From 285e7aa179a6081531be3274772b89e8989fd107 Mon Sep 17 00:00:00 2001 From: Alexander Barkov Date: Thu, 26 Sep 2013 18:02:17 +0400 Subject: MDEV-4425 REGEXP enhancements --- pcre/AUTHORS | 45 + pcre/CMakeLists.txt | 973 ++ pcre/ChangeLog | 5149 ++++++++ pcre/HACKING | 473 + pcre/INSTALL | 370 + pcre/LICENCE | 92 + pcre/NEWS | 635 + pcre/NON-AUTOTOOLS-BUILD | 764 ++ pcre/NON-UNIX-USE | 7 + pcre/README | 974 ++ pcre/RunGrepTest | 567 + pcre/RunTest | 1063 ++ pcre/RunTest.bat | 616 + pcre/cmake/COPYING-CMAKE-SCRIPTS | 22 + pcre/cmake/FindEditline.cmake | 17 + pcre/cmake/FindPackageHandleStandardArgs.cmake | 58 + pcre/cmake/FindReadline.cmake | 29 + pcre/config-cmake.h.in | 54 + pcre/config.h | 54 + pcre/configure.ac | 1093 ++ pcre/doc/html/NON-AUTOTOOLS-BUILD.txt | 764 ++ pcre/doc/html/README.txt | 974 ++ pcre/doc/html/index.html | 185 + pcre/doc/html/pcre-config.html | 109 + pcre/doc/html/pcre.html | 210 + pcre/doc/html/pcre16.html | 384 + pcre/doc/html/pcre32.html | 383 + pcre/doc/html/pcre_assign_jit_stack.html | 76 + pcre/doc/html/pcre_compile.html | 108 + pcre/doc/html/pcre_compile2.html | 112 + pcre/doc/html/pcre_config.html | 91 + pcre/doc/html/pcre_copy_named_substring.html | 65 + pcre/doc/html/pcre_copy_substring.html | 61 + pcre/doc/html/pcre_dfa_exec.html | 129 + pcre/doc/html/pcre_exec.html | 111 + pcre/doc/html/pcre_free_study.html | 46 + pcre/doc/html/pcre_free_substring.html | 46 + pcre/doc/html/pcre_free_substring_list.html | 46 + pcre/doc/html/pcre_fullinfo.html | 108 + pcre/doc/html/pcre_get_named_substring.html | 68 + pcre/doc/html/pcre_get_stringnumber.html | 57 + pcre/doc/html/pcre_get_stringtable_entries.html | 60 + pcre/doc/html/pcre_get_substring.html | 64 + pcre/doc/html/pcre_get_substring_list.html | 61 + pcre/doc/html/pcre_jit_exec.html | 108 + pcre/doc/html/pcre_jit_stack_alloc.html | 55 + pcre/doc/html/pcre_jit_stack_free.html | 48 + pcre/doc/html/pcre_maketables.html | 48 + pcre/doc/html/pcre_pattern_to_host_byte_order.html | 58 + pcre/doc/html/pcre_refcount.html | 51 + pcre/doc/html/pcre_study.html | 68 + pcre/doc/html/pcre_utf16_to_host_byte_order.html | 57 + pcre/doc/html/pcre_utf32_to_host_byte_order.html | 57 + pcre/doc/html/pcre_version.html | 46 + pcre/doc/html/pcreapi.html | 2859 +++++ pcre/doc/html/pcrebuild.html | 534 + pcre/doc/html/pcrecallout.html | 251 + pcre/doc/html/pcrecompat.html | 229 + pcre/doc/html/pcrecpp.html | 368 + pcre/doc/html/pcredemo.html | 426 + pcre/doc/html/pcregrep.html | 757 ++ pcre/doc/html/pcrejit.html | 452 + pcre/doc/html/pcrelimits.html | 86 + pcre/doc/html/pcrematching.html | 233 + pcre/doc/html/pcrepartial.html | 499 + pcre/doc/html/pcrepattern.html | 3118 +++++ pcre/doc/html/pcreperform.html | 195 + pcre/doc/html/pcreposix.html | 292 + pcre/doc/html/pcreprecompile.html | 158 + pcre/doc/html/pcresample.html | 110 + pcre/doc/html/pcrestack.html | 225 + pcre/doc/html/pcresyntax.html | 525 + pcre/doc/html/pcretest.html | 1100 ++ pcre/doc/html/pcreunicode.html | 262 + pcre/doc/index.html.src | 185 + pcre/doc/pcre-config.1 | 92 + pcre/doc/pcre-config.txt | 86 + pcre/doc/pcre.3 | 215 + pcre/doc/pcre.txt | 10181 +++++++++++++++ pcre/doc/pcre16.3 | 391 + pcre/doc/pcre32.3 | 390 + pcre/doc/pcre_assign_jit_stack.3 | 61 + pcre/doc/pcre_compile.3 | 98 + pcre/doc/pcre_compile2.3 | 106 + pcre/doc/pcre_config.3 | 76 + pcre/doc/pcre_copy_named_substring.3 | 59 + pcre/doc/pcre_copy_substring.3 | 52 + pcre/doc/pcre_dfa_exec.3 | 126 + pcre/doc/pcre_exec.3 | 104 + pcre/doc/pcre_free_study.3 | 31 + pcre/doc/pcre_free_substring.3 | 31 + pcre/doc/pcre_free_substring_list.3 | 31 + pcre/doc/pcre_fullinfo.3 | 95 + pcre/doc/pcre_get_named_substring.3 | 62 + pcre/doc/pcre_get_stringnumber.3 | 45 + pcre/doc/pcre_get_stringtable_entries.3 | 48 + pcre/doc/pcre_get_substring.3 | 55 + pcre/doc/pcre_get_substring_list.3 | 49 + pcre/doc/pcre_jit_exec.3 | 104 + pcre/doc/pcre_jit_stack_alloc.3 | 45 + pcre/doc/pcre_jit_stack_free.3 | 35 + pcre/doc/pcre_maketables.3 | 33 + pcre/doc/pcre_pattern_to_host_byte_order.3 | 46 + pcre/doc/pcre_refcount.3 | 36 + pcre/doc/pcre_study.3 | 56 + pcre/doc/pcre_utf16_to_host_byte_order.3 | 46 + pcre/doc/pcre_utf32_to_host_byte_order.3 | 46 + pcre/doc/pcre_version.3 | 31 + pcre/doc/pcreapi.3 | 2891 +++++ pcre/doc/pcrebuild.3 | 550 + pcre/doc/pcrecallout.3 | 222 + pcre/doc/pcrecompat.3 | 195 + pcre/doc/pcrecpp.3 | 348 + pcre/doc/pcredemo.3 | 424 + pcre/doc/pcregrep.1 | 679 + pcre/doc/pcregrep.txt | 740 ++ pcre/doc/pcrejit.3 | 431 + pcre/doc/pcrelimits.3 | 67 + pcre/doc/pcrematching.3 | 206 + pcre/doc/pcrepartial.3 | 467 + pcre/doc/pcrepattern.3 | 3150 +++++ pcre/doc/pcreperform.3 | 177 + pcre/doc/pcreposix.3 | 270 + pcre/doc/pcreprecompile.3 | 151 + pcre/doc/pcresample.3 | 99 + pcre/doc/pcrestack.3 | 215 + pcre/doc/pcresyntax.3 | 500 + pcre/doc/pcretest.1 | 1099 ++ pcre/doc/pcretest.txt | 1034 ++ pcre/doc/pcreunicode.3 | 249 + pcre/doc/perltest.txt | 42 + pcre/pcre.h | 663 + pcre/pcre.h.in | 663 + pcre/pcre16_byte_order.c | 45 + pcre/pcre16_chartables.c | 45 + pcre/pcre16_compile.c | 45 + pcre/pcre16_config.c | 45 + pcre/pcre16_dfa_exec.c | 45 + pcre/pcre16_exec.c | 45 + pcre/pcre16_fullinfo.c | 45 + pcre/pcre16_get.c | 45 + pcre/pcre16_globals.c | 45 + pcre/pcre16_jit_compile.c | 45 + pcre/pcre16_maketables.c | 45 + pcre/pcre16_newline.c | 45 + pcre/pcre16_ord2utf16.c | 90 + pcre/pcre16_printint.c | 45 + pcre/pcre16_refcount.c | 45 + pcre/pcre16_string_utils.c | 45 + pcre/pcre16_study.c | 45 + pcre/pcre16_tables.c | 45 + pcre/pcre16_ucd.c | 45 + pcre/pcre16_utf16_utils.c | 130 + pcre/pcre16_valid_utf16.c | 137 + pcre/pcre16_version.c | 45 + pcre/pcre16_xclass.c | 45 + pcre/pcre32_byte_order.c | 45 + pcre/pcre32_chartables.c | 45 + pcre/pcre32_compile.c | 45 + pcre/pcre32_config.c | 45 + pcre/pcre32_dfa_exec.c | 45 + pcre/pcre32_exec.c | 45 + pcre/pcre32_fullinfo.c | 45 + pcre/pcre32_get.c | 45 + pcre/pcre32_globals.c | 45 + pcre/pcre32_jit_compile.c | 45 + pcre/pcre32_maketables.c | 45 + pcre/pcre32_newline.c | 45 + pcre/pcre32_ord2utf32.c | 82 + pcre/pcre32_printint.c | 45 + pcre/pcre32_refcount.c | 45 + pcre/pcre32_string_utils.c | 45 + pcre/pcre32_study.c | 45 + pcre/pcre32_tables.c | 45 + pcre/pcre32_ucd.c | 45 + pcre/pcre32_utf32_utils.c | 141 + pcre/pcre32_valid_utf32.c | 124 + pcre/pcre32_version.c | 45 + pcre/pcre32_xclass.c | 45 + pcre/pcre_byte_order.c | 319 + pcre/pcre_chartables.c | 198 + pcre/pcre_chartables.c.dist | 198 + pcre/pcre_compile.c | 8432 ++++++++++++ pcre/pcre_config.c | 186 + pcre/pcre_dfa_exec.c | 3608 ++++++ pcre/pcre_exec.c | 7213 +++++++++++ pcre/pcre_fullinfo.c | 241 + pcre/pcre_get.c | 662 + pcre/pcre_globals.c | 84 + pcre/pcre_internal.h | 2771 ++++ pcre/pcre_jit_compile.c | 9752 ++++++++++++++ pcre/pcre_jit_test.c | 1691 +++ pcre/pcre_maketables.c | 151 + pcre/pcre_newline.c | 210 + pcre/pcre_ord2utf8.c | 94 + pcre/pcre_printint.c | 766 ++ pcre/pcre_refcount.c | 92 + pcre/pcre_scanner.cc | 199 + pcre/pcre_scanner.h | 172 + pcre/pcre_scanner_unittest.cc | 159 + pcre/pcre_string_utils.c | 211 + pcre/pcre_stringpiece.cc | 43 + pcre/pcre_stringpiece.h | 179 + pcre/pcre_stringpiece.h.in | 179 + pcre/pcre_stringpiece_unittest.cc | 150 + pcre/pcre_study.c | 1562 +++ pcre/pcre_tables.c | 658 + pcre/pcre_ucd.c | 3298 +++++ pcre/pcre_valid_utf8.c | 301 + pcre/pcre_version.c | 98 + pcre/pcre_xclass.c | 212 + pcre/pcrecpp.cc | 922 ++ pcre/pcrecpp.h | 710 ++ pcre/pcrecpp_internal.h | 71 + pcre/pcrecpp_unittest.cc | 1291 ++ pcre/pcrecpparg.h | 174 + pcre/pcrecpparg.h.in | 174 + pcre/pcredemo.c | 406 + pcre/pcregrep.c | 3186 +++++ pcre/pcreposix.c | 420 + pcre/pcreposix.h | 150 + pcre/pcretest.c | 5524 ++++++++ pcre/testdata/grepbinary | Bin 0 -> 45 bytes pcre/testdata/grepfilelist | 3 + pcre/testdata/grepinput | 611 + pcre/testdata/grepinput3 | 15 + pcre/testdata/grepinput8 | 11 + pcre/testdata/grepinputv | 4 + pcre/testdata/grepinputx | 43 + pcre/testdata/greplist | 7 + pcre/testdata/grepoutput | 742 ++ pcre/testdata/grepoutput8 | 12 + pcre/testdata/grepoutputN | 16 + pcre/testdata/greppatN4 | 2 + pcre/testdata/saved16 | Bin 0 -> 86 bytes pcre/testdata/saved16BE-1 | Bin 0 -> 410 bytes pcre/testdata/saved16BE-2 | Bin 0 -> 344 bytes pcre/testdata/saved16LE-1 | Bin 0 -> 410 bytes pcre/testdata/saved16LE-2 | Bin 0 -> 344 bytes pcre/testdata/saved32 | Bin 0 -> 108 bytes pcre/testdata/saved32BE-1 | Bin 0 -> 552 bytes pcre/testdata/saved32BE-2 | Bin 0 -> 456 bytes pcre/testdata/saved32LE-1 | Bin 0 -> 552 bytes pcre/testdata/saved32LE-2 | Bin 0 -> 456 bytes pcre/testdata/saved8 | Bin 0 -> 77 bytes pcre/testdata/testinput1 | 5611 ++++++++ pcre/testdata/testinput10 | 1401 ++ pcre/testdata/testinput11 | 135 + pcre/testdata/testinput12 | 90 + pcre/testdata/testinput13 | 9 + pcre/testdata/testinput14 | 329 + pcre/testdata/testinput15 | 364 + pcre/testdata/testinput16 | 35 + pcre/testdata/testinput17 | 296 + pcre/testdata/testinput18 | 298 + pcre/testdata/testinput19 | 22 + pcre/testdata/testinput2 | 3840 ++++++ pcre/testdata/testinput20 | 19 + pcre/testdata/testinput21 | 22 + pcre/testdata/testinput22 | 18 + pcre/testdata/testinput23 | 16 + pcre/testdata/testinput24 | 9 + pcre/testdata/testinput25 | 32 + pcre/testdata/testinput26 | 10 + pcre/testdata/testinput3 | 95 + pcre/testdata/testinput4 | 714 ++ pcre/testdata/testinput5 | 772 ++ pcre/testdata/testinput6 | 1327 ++ pcre/testdata/testinput7 | 738 ++ pcre/testdata/testinput8 | 4804 +++++++ pcre/testdata/testinput9 | 717 ++ pcre/testdata/testinputEBC | 121 + pcre/testdata/testoutput1 | 9203 ++++++++++++++ pcre/testdata/testoutput10 | 2851 +++++ pcre/testdata/testoutput11-16 | 713 ++ pcre/testdata/testoutput11-32 | 713 ++ pcre/testdata/testoutput11-8 | 713 ++ pcre/testdata/testoutput12 | 176 + pcre/testdata/testoutput13 | 21 + pcre/testdata/testoutput14 | 476 + pcre/testdata/testoutput15 | 1137 ++ pcre/testdata/testoutput16 | 121 + pcre/testdata/testoutput17 | 505 + pcre/testdata/testoutput18-16 | 1024 ++ pcre/testdata/testoutput18-32 | 1021 ++ pcre/testdata/testoutput19 | 88 + pcre/testdata/testoutput2 | 12668 +++++++++++++++++++ pcre/testdata/testoutput20 | 27 + pcre/testdata/testoutput21-16 | 96 + pcre/testdata/testoutput21-32 | 96 + pcre/testdata/testoutput22-16 | 76 + pcre/testdata/testoutput22-32 | 76 + pcre/testdata/testoutput23 | 42 + pcre/testdata/testoutput24 | 11 + pcre/testdata/testoutput25 | 79 + pcre/testdata/testoutput26 | 12 + pcre/testdata/testoutput3 | 169 + pcre/testdata/testoutput4 | 1260 ++ pcre/testdata/testoutput5 | 1849 +++ pcre/testdata/testoutput6 | 2149 ++++ pcre/testdata/testoutput7 | 1581 +++ pcre/testdata/testoutput8 | 8023 ++++++++++++ pcre/testdata/testoutput9 | 1371 ++ pcre/testdata/testoutputEBC | 182 + pcre/testdata/wintestinput3 | 91 + pcre/testdata/wintestoutput3 | 166 + pcre/ucp.h | 197 + 307 files changed, 191241 insertions(+) create mode 100644 pcre/AUTHORS create mode 100644 pcre/CMakeLists.txt create mode 100644 pcre/ChangeLog create mode 100644 pcre/HACKING create mode 100644 pcre/INSTALL create mode 100644 pcre/LICENCE create mode 100644 pcre/NEWS create mode 100644 pcre/NON-AUTOTOOLS-BUILD create mode 100644 pcre/NON-UNIX-USE create mode 100644 pcre/README create mode 100755 pcre/RunGrepTest create mode 100755 pcre/RunTest create mode 100644 pcre/RunTest.bat create mode 100644 pcre/cmake/COPYING-CMAKE-SCRIPTS create mode 100644 pcre/cmake/FindEditline.cmake create mode 100644 pcre/cmake/FindPackageHandleStandardArgs.cmake create mode 100644 pcre/cmake/FindReadline.cmake create mode 100644 pcre/config-cmake.h.in create mode 100644 pcre/config.h create mode 100644 pcre/configure.ac create mode 100644 pcre/doc/html/NON-AUTOTOOLS-BUILD.txt create mode 100644 pcre/doc/html/README.txt create mode 100644 pcre/doc/html/index.html create mode 100644 pcre/doc/html/pcre-config.html create mode 100644 pcre/doc/html/pcre.html create mode 100644 pcre/doc/html/pcre16.html create mode 100644 pcre/doc/html/pcre32.html create mode 100644 pcre/doc/html/pcre_assign_jit_stack.html create mode 100644 pcre/doc/html/pcre_compile.html create mode 100644 pcre/doc/html/pcre_compile2.html create mode 100644 pcre/doc/html/pcre_config.html create mode 100644 pcre/doc/html/pcre_copy_named_substring.html create mode 100644 pcre/doc/html/pcre_copy_substring.html create mode 100644 pcre/doc/html/pcre_dfa_exec.html create mode 100644 pcre/doc/html/pcre_exec.html create mode 100644 pcre/doc/html/pcre_free_study.html create mode 100644 pcre/doc/html/pcre_free_substring.html create mode 100644 pcre/doc/html/pcre_free_substring_list.html create mode 100644 pcre/doc/html/pcre_fullinfo.html create mode 100644 pcre/doc/html/pcre_get_named_substring.html create mode 100644 pcre/doc/html/pcre_get_stringnumber.html create mode 100644 pcre/doc/html/pcre_get_stringtable_entries.html create mode 100644 pcre/doc/html/pcre_get_substring.html create mode 100644 pcre/doc/html/pcre_get_substring_list.html create mode 100644 pcre/doc/html/pcre_jit_exec.html create mode 100644 pcre/doc/html/pcre_jit_stack_alloc.html create mode 100644 pcre/doc/html/pcre_jit_stack_free.html create mode 100644 pcre/doc/html/pcre_maketables.html create mode 100644 pcre/doc/html/pcre_pattern_to_host_byte_order.html create mode 100644 pcre/doc/html/pcre_refcount.html create mode 100644 pcre/doc/html/pcre_study.html create mode 100644 pcre/doc/html/pcre_utf16_to_host_byte_order.html create mode 100644 pcre/doc/html/pcre_utf32_to_host_byte_order.html create mode 100644 pcre/doc/html/pcre_version.html create mode 100644 pcre/doc/html/pcreapi.html create mode 100644 pcre/doc/html/pcrebuild.html create mode 100644 pcre/doc/html/pcrecallout.html create mode 100644 pcre/doc/html/pcrecompat.html create mode 100644 pcre/doc/html/pcrecpp.html create mode 100644 pcre/doc/html/pcredemo.html create mode 100644 pcre/doc/html/pcregrep.html create mode 100644 pcre/doc/html/pcrejit.html create mode 100644 pcre/doc/html/pcrelimits.html create mode 100644 pcre/doc/html/pcrematching.html create mode 100644 pcre/doc/html/pcrepartial.html create mode 100644 pcre/doc/html/pcrepattern.html create mode 100644 pcre/doc/html/pcreperform.html create mode 100644 pcre/doc/html/pcreposix.html create mode 100644 pcre/doc/html/pcreprecompile.html create mode 100644 pcre/doc/html/pcresample.html create mode 100644 pcre/doc/html/pcrestack.html create mode 100644 pcre/doc/html/pcresyntax.html create mode 100644 pcre/doc/html/pcretest.html create mode 100644 pcre/doc/html/pcreunicode.html create mode 100644 pcre/doc/index.html.src create mode 100644 pcre/doc/pcre-config.1 create mode 100644 pcre/doc/pcre-config.txt create mode 100644 pcre/doc/pcre.3 create mode 100644 pcre/doc/pcre.txt create mode 100644 pcre/doc/pcre16.3 create mode 100644 pcre/doc/pcre32.3 create mode 100644 pcre/doc/pcre_assign_jit_stack.3 create mode 100644 pcre/doc/pcre_compile.3 create mode 100644 pcre/doc/pcre_compile2.3 create mode 100644 pcre/doc/pcre_config.3 create mode 100644 pcre/doc/pcre_copy_named_substring.3 create mode 100644 pcre/doc/pcre_copy_substring.3 create mode 100644 pcre/doc/pcre_dfa_exec.3 create mode 100644 pcre/doc/pcre_exec.3 create mode 100644 pcre/doc/pcre_free_study.3 create mode 100644 pcre/doc/pcre_free_substring.3 create mode 100644 pcre/doc/pcre_free_substring_list.3 create mode 100644 pcre/doc/pcre_fullinfo.3 create mode 100644 pcre/doc/pcre_get_named_substring.3 create mode 100644 pcre/doc/pcre_get_stringnumber.3 create mode 100644 pcre/doc/pcre_get_stringtable_entries.3 create mode 100644 pcre/doc/pcre_get_substring.3 create mode 100644 pcre/doc/pcre_get_substring_list.3 create mode 100644 pcre/doc/pcre_jit_exec.3 create mode 100644 pcre/doc/pcre_jit_stack_alloc.3 create mode 100644 pcre/doc/pcre_jit_stack_free.3 create mode 100644 pcre/doc/pcre_maketables.3 create mode 100644 pcre/doc/pcre_pattern_to_host_byte_order.3 create mode 100644 pcre/doc/pcre_refcount.3 create mode 100644 pcre/doc/pcre_study.3 create mode 100644 pcre/doc/pcre_utf16_to_host_byte_order.3 create mode 100644 pcre/doc/pcre_utf32_to_host_byte_order.3 create mode 100644 pcre/doc/pcre_version.3 create mode 100644 pcre/doc/pcreapi.3 create mode 100644 pcre/doc/pcrebuild.3 create mode 100644 pcre/doc/pcrecallout.3 create mode 100644 pcre/doc/pcrecompat.3 create mode 100644 pcre/doc/pcrecpp.3 create mode 100644 pcre/doc/pcredemo.3 create mode 100644 pcre/doc/pcregrep.1 create mode 100644 pcre/doc/pcregrep.txt create mode 100644 pcre/doc/pcrejit.3 create mode 100644 pcre/doc/pcrelimits.3 create mode 100644 pcre/doc/pcrematching.3 create mode 100644 pcre/doc/pcrepartial.3 create mode 100644 pcre/doc/pcrepattern.3 create mode 100644 pcre/doc/pcreperform.3 create mode 100644 pcre/doc/pcreposix.3 create mode 100644 pcre/doc/pcreprecompile.3 create mode 100644 pcre/doc/pcresample.3 create mode 100644 pcre/doc/pcrestack.3 create mode 100644 pcre/doc/pcresyntax.3 create mode 100644 pcre/doc/pcretest.1 create mode 100644 pcre/doc/pcretest.txt create mode 100644 pcre/doc/pcreunicode.3 create mode 100644 pcre/doc/perltest.txt create mode 100644 pcre/pcre.h create mode 100644 pcre/pcre.h.in create mode 100644 pcre/pcre16_byte_order.c create mode 100644 pcre/pcre16_chartables.c create mode 100644 pcre/pcre16_compile.c create mode 100644 pcre/pcre16_config.c create mode 100644 pcre/pcre16_dfa_exec.c create mode 100644 pcre/pcre16_exec.c create mode 100644 pcre/pcre16_fullinfo.c create mode 100644 pcre/pcre16_get.c create mode 100644 pcre/pcre16_globals.c create mode 100644 pcre/pcre16_jit_compile.c create mode 100644 pcre/pcre16_maketables.c create mode 100644 pcre/pcre16_newline.c create mode 100644 pcre/pcre16_ord2utf16.c create mode 100644 pcre/pcre16_printint.c create mode 100644 pcre/pcre16_refcount.c create mode 100644 pcre/pcre16_string_utils.c create mode 100644 pcre/pcre16_study.c create mode 100644 pcre/pcre16_tables.c create mode 100644 pcre/pcre16_ucd.c create mode 100644 pcre/pcre16_utf16_utils.c create mode 100644 pcre/pcre16_valid_utf16.c create mode 100644 pcre/pcre16_version.c create mode 100644 pcre/pcre16_xclass.c create mode 100644 pcre/pcre32_byte_order.c create mode 100644 pcre/pcre32_chartables.c create mode 100644 pcre/pcre32_compile.c create mode 100644 pcre/pcre32_config.c create mode 100644 pcre/pcre32_dfa_exec.c create mode 100644 pcre/pcre32_exec.c create mode 100644 pcre/pcre32_fullinfo.c create mode 100644 pcre/pcre32_get.c create mode 100644 pcre/pcre32_globals.c create mode 100644 pcre/pcre32_jit_compile.c create mode 100644 pcre/pcre32_maketables.c create mode 100644 pcre/pcre32_newline.c create mode 100644 pcre/pcre32_ord2utf32.c create mode 100644 pcre/pcre32_printint.c create mode 100644 pcre/pcre32_refcount.c create mode 100644 pcre/pcre32_string_utils.c create mode 100644 pcre/pcre32_study.c create mode 100644 pcre/pcre32_tables.c create mode 100644 pcre/pcre32_ucd.c create mode 100644 pcre/pcre32_utf32_utils.c create mode 100644 pcre/pcre32_valid_utf32.c create mode 100644 pcre/pcre32_version.c create mode 100644 pcre/pcre32_xclass.c create mode 100644 pcre/pcre_byte_order.c create mode 100644 pcre/pcre_chartables.c create mode 100644 pcre/pcre_chartables.c.dist create mode 100644 pcre/pcre_compile.c create mode 100644 pcre/pcre_config.c create mode 100644 pcre/pcre_dfa_exec.c create mode 100644 pcre/pcre_exec.c create mode 100644 pcre/pcre_fullinfo.c create mode 100644 pcre/pcre_get.c create mode 100644 pcre/pcre_globals.c create mode 100644 pcre/pcre_internal.h create mode 100644 pcre/pcre_jit_compile.c create mode 100644 pcre/pcre_jit_test.c create mode 100644 pcre/pcre_maketables.c create mode 100644 pcre/pcre_newline.c create mode 100644 pcre/pcre_ord2utf8.c create mode 100644 pcre/pcre_printint.c create mode 100644 pcre/pcre_refcount.c create mode 100644 pcre/pcre_scanner.cc create mode 100644 pcre/pcre_scanner.h create mode 100644 pcre/pcre_scanner_unittest.cc create mode 100644 pcre/pcre_string_utils.c create mode 100644 pcre/pcre_stringpiece.cc create mode 100644 pcre/pcre_stringpiece.h create mode 100644 pcre/pcre_stringpiece.h.in create mode 100644 pcre/pcre_stringpiece_unittest.cc create mode 100644 pcre/pcre_study.c create mode 100644 pcre/pcre_tables.c create mode 100644 pcre/pcre_ucd.c create mode 100644 pcre/pcre_valid_utf8.c create mode 100644 pcre/pcre_version.c create mode 100644 pcre/pcre_xclass.c create mode 100644 pcre/pcrecpp.cc create mode 100644 pcre/pcrecpp.h create mode 100644 pcre/pcrecpp_internal.h create mode 100644 pcre/pcrecpp_unittest.cc create mode 100644 pcre/pcrecpparg.h create mode 100644 pcre/pcrecpparg.h.in create mode 100644 pcre/pcredemo.c create mode 100644 pcre/pcregrep.c create mode 100644 pcre/pcreposix.c create mode 100644 pcre/pcreposix.h create mode 100644 pcre/pcretest.c create mode 100644 pcre/testdata/grepbinary create mode 100644 pcre/testdata/grepfilelist create mode 100644 pcre/testdata/grepinput create mode 100644 pcre/testdata/grepinput3 create mode 100644 pcre/testdata/grepinput8 create mode 100644 pcre/testdata/grepinputv create mode 100644 pcre/testdata/grepinputx create mode 100644 pcre/testdata/greplist create mode 100644 pcre/testdata/grepoutput create mode 100644 pcre/testdata/grepoutput8 create mode 100644 pcre/testdata/grepoutputN create mode 100644 pcre/testdata/greppatN4 create mode 100644 pcre/testdata/saved16 create mode 100644 pcre/testdata/saved16BE-1 create mode 100644 pcre/testdata/saved16BE-2 create mode 100644 pcre/testdata/saved16LE-1 create mode 100644 pcre/testdata/saved16LE-2 create mode 100644 pcre/testdata/saved32 create mode 100644 pcre/testdata/saved32BE-1 create mode 100644 pcre/testdata/saved32BE-2 create mode 100644 pcre/testdata/saved32LE-1 create mode 100644 pcre/testdata/saved32LE-2 create mode 100644 pcre/testdata/saved8 create mode 100644 pcre/testdata/testinput1 create mode 100644 pcre/testdata/testinput10 create mode 100644 pcre/testdata/testinput11 create mode 100644 pcre/testdata/testinput12 create mode 100644 pcre/testdata/testinput13 create mode 100644 pcre/testdata/testinput14 create mode 100644 pcre/testdata/testinput15 create mode 100644 pcre/testdata/testinput16 create mode 100644 pcre/testdata/testinput17 create mode 100644 pcre/testdata/testinput18 create mode 100644 pcre/testdata/testinput19 create mode 100644 pcre/testdata/testinput2 create mode 100644 pcre/testdata/testinput20 create mode 100644 pcre/testdata/testinput21 create mode 100644 pcre/testdata/testinput22 create mode 100644 pcre/testdata/testinput23 create mode 100644 pcre/testdata/testinput24 create mode 100644 pcre/testdata/testinput25 create mode 100644 pcre/testdata/testinput26 create mode 100644 pcre/testdata/testinput3 create mode 100644 pcre/testdata/testinput4 create mode 100644 pcre/testdata/testinput5 create mode 100644 pcre/testdata/testinput6 create mode 100644 pcre/testdata/testinput7 create mode 100644 pcre/testdata/testinput8 create mode 100644 pcre/testdata/testinput9 create mode 100644 pcre/testdata/testinputEBC create mode 100644 pcre/testdata/testoutput1 create mode 100644 pcre/testdata/testoutput10 create mode 100644 pcre/testdata/testoutput11-16 create mode 100644 pcre/testdata/testoutput11-32 create mode 100644 pcre/testdata/testoutput11-8 create mode 100644 pcre/testdata/testoutput12 create mode 100644 pcre/testdata/testoutput13 create mode 100644 pcre/testdata/testoutput14 create mode 100644 pcre/testdata/testoutput15 create mode 100644 pcre/testdata/testoutput16 create mode 100644 pcre/testdata/testoutput17 create mode 100644 pcre/testdata/testoutput18-16 create mode 100644 pcre/testdata/testoutput18-32 create mode 100644 pcre/testdata/testoutput19 create mode 100644 pcre/testdata/testoutput2 create mode 100644 pcre/testdata/testoutput20 create mode 100644 pcre/testdata/testoutput21-16 create mode 100644 pcre/testdata/testoutput21-32 create mode 100644 pcre/testdata/testoutput22-16 create mode 100644 pcre/testdata/testoutput22-32 create mode 100644 pcre/testdata/testoutput23 create mode 100644 pcre/testdata/testoutput24 create mode 100644 pcre/testdata/testoutput25 create mode 100644 pcre/testdata/testoutput26 create mode 100644 pcre/testdata/testoutput3 create mode 100644 pcre/testdata/testoutput4 create mode 100644 pcre/testdata/testoutput5 create mode 100644 pcre/testdata/testoutput6 create mode 100644 pcre/testdata/testoutput7 create mode 100644 pcre/testdata/testoutput8 create mode 100644 pcre/testdata/testoutput9 create mode 100644 pcre/testdata/testoutputEBC create mode 100644 pcre/testdata/wintestinput3 create mode 100644 pcre/testdata/wintestoutput3 create mode 100644 pcre/ucp.h (limited to 'pcre') diff --git a/pcre/AUTHORS b/pcre/AUTHORS new file mode 100644 index 00000000000..97d8c71dd67 --- /dev/null +++ b/pcre/AUTHORS @@ -0,0 +1,45 @@ +THE MAIN PCRE LIBRARY +--------------------- + +Written by: Philip Hazel +Email local part: ph10 +Email domain: cam.ac.uk + +University of Cambridge Computing Service, +Cambridge, England. + +Copyright (c) 1997-2013 University of Cambridge +All rights reserved + + +PCRE JUST-IN-TIME COMPILATION SUPPORT +------------------------------------- + +Written by: Zoltan Herczeg +Email local part: hzmester +Emain domain: freemail.hu + +Copyright(c) 2010-2013 Zoltan Herczeg +All rights reserved. + + +STACK-LESS JUST-IN-TIME COMPILER +-------------------------------- + +Written by: Zoltan Herczeg +Email local part: hzmester +Emain domain: freemail.hu + +Copyright(c) 2009-2013 Zoltan Herczeg +All rights reserved. + + +THE C++ WRAPPER LIBRARY +----------------------- + +Written by: Google Inc. + +Copyright (c) 2007-2012 Google Inc +All rights reserved + +#### diff --git a/pcre/CMakeLists.txt b/pcre/CMakeLists.txt new file mode 100644 index 00000000000..4767f9c7e9c --- /dev/null +++ b/pcre/CMakeLists.txt @@ -0,0 +1,973 @@ +# CMakeLists.txt +# +# +# This file allows building PCRE with the CMake configuration and build +# tool. Download CMake in source or binary form from http://www.cmake.org/ +# +# Original listfile by Christian Ehrlicher +# Refined and expanded by Daniel Richard G. +# 2007-09-14 mod by Sheri so 7.4 supported configuration options can be entered +# 2007-09-19 Adjusted by PH to retain previous default settings +# 2007-12-26 (a) On UNIX, use names libpcre instead of just pcre +# (b) Ensure pcretest and pcregrep link with the local library, +# not a previously-installed one. +# (c) Add PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, and +# PCRE_SUPPORT_LIBBZ2. +# 2008-01-20 Brought up to date to include several new features by Christian +# Ehrlicher. +# 2008-01-22 Sheri added options for backward compatibility of library names +# when building with minGW: +# if "ON", NON_STANDARD_LIB_PREFIX causes shared libraries to +# be built without "lib" as prefix. (The libraries will be named +# pcre.dll, pcreposix.dll and pcrecpp.dll). +# if "ON", NON_STANDARD_LIB_SUFFIX causes shared libraries to +# be built with suffix of "-0.dll". (The libraries will be named +# libpcre-0.dll, libpcreposix-0.dll and libpcrecpp-0.dll - same names +# built by default with Configure and Make. +# 2008-01-23 PH removed the automatic build of pcredemo. +# 2008-04-22 PH modified READLINE support so it finds NCURSES when needed. +# 2008-07-03 PH updated for revised UCP property support (change of files) +# 2009-03-23 PH applied Steven Van Ingelgem's patch to change the name +# CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE +# is included within another project. +# 2009-03-23 PH applied a modified version of Steven Van Ingelgem's patches to +# add options to stop the building of pcregrep and the tests, and +# to disable the final configuration report. +# 2009-04-11 PH applied Christian Ehrlicher's patch to show compiler flags that +# are set by specifying a release type. +# 2010-01-02 PH added test for stdint.h +# 2010-03-02 PH added test for inttypes.h +# 2011-08-01 PH added PCREGREP_BUFSIZE +# 2011-08-22 PH added PCRE_SUPPORT_JIT +# 2011-09-06 PH modified WIN32 ADD_TEST line as suggested by Sergey Cherepanov +# 2011-09-06 PH added PCRE_SUPPORT_PCREGREP_JIT +# 2011-10-04 Sheri added support for including coff data in windows shared libraries +# compiled with MINGW if pcre.rc and/or pcreposix.rc are placed in +# the source dir by the user prior to building +# 2011-10-04 Sheri changed various add_test's to use exes' location built instead +# of DEBUG location only (likely only matters in MSVC) +# 2011-10-04 Sheri added scripts to provide needed variables to RunTest and +# RunGrepTest (used for UNIX and Msys) +# 2011-10-04 Sheri added scripts to provide needed variables and to execute +# RunTest.bat in Win32 (for effortless testing with "make test") +# 2011-10-04 Sheri Increased minimum required cmake version +# 2012-01-06 PH removed pcre_info.c and added pcre_string_utils.c +# 2012-01-10 Zoltan Herczeg added libpcre16 support +# 2012-01-13 Stephen Kelly added out of source build support +# 2012-01-17 PH applied Stephen Kelly's patch to parse the version data out +# of the configure.ac file +# 2012-02-26 PH added support for libedit +# 2012-09-06 PH added support for PCRE_EBCDIC_NL25 +# 2012-09-08 ChPe added PCRE32 support +# 2012-10-23 PH added support for VALGRIND and GCOV +# 2012-12-08 PH added patch from Daniel Richard G to quash some MSVC warnings + +PROJECT(PCRE C CXX) + +# Increased minimum to 2.8.0 to support newer add_test features + +CMAKE_MINIMUM_REQUIRED(VERSION 2.8.0) + +SET(CMAKE_MODULE_PATH ${PROJECT_SOURCE_DIR}/cmake) # for FindReadline.cmake + +# Enable UTF support +OPTION(PCRE_SUPPORT_UTF "uft support" ON) + +# Enable Unicode properties +OPTION(PCRE_SUPPORT_UNICODE_PROPERTIES "Unicode properties" ON) + +INCLUDE_DIRECTORIES(${CMAKE_SOURCE_DIR}/include) + +#INSTALL(FILES "my_regex.h" DESTINATION ${INSTALL_INCLUDEDIR}/private COMPONENT Development) + +# external packages +FIND_PACKAGE( BZip2 ) +FIND_PACKAGE( ZLIB ) +FIND_PACKAGE( Readline ) +FIND_PACKAGE( Editline ) + +# Configuration checks + +INCLUDE(CheckIncludeFile) +INCLUDE(CheckIncludeFileCXX) +INCLUDE(CheckFunctionExists) +INCLUDE(CheckTypeSize) + +CHECK_INCLUDE_FILE(dirent.h HAVE_DIRENT_H) +CHECK_INCLUDE_FILE(stdint.h HAVE_STDINT_H) +CHECK_INCLUDE_FILE(inttypes.h HAVE_INTTYPES_H) +CHECK_INCLUDE_FILE(sys/stat.h HAVE_SYS_STAT_H) +CHECK_INCLUDE_FILE(sys/types.h HAVE_SYS_TYPES_H) +CHECK_INCLUDE_FILE(unistd.h HAVE_UNISTD_H) +CHECK_INCLUDE_FILE(windows.h HAVE_WINDOWS_H) + +CHECK_INCLUDE_FILE_CXX(type_traits.h HAVE_TYPE_TRAITS_H) +CHECK_INCLUDE_FILE_CXX(bits/type_traits.h HAVE_BITS_TYPE_TRAITS_H) + +CHECK_FUNCTION_EXISTS(bcopy HAVE_BCOPY) +CHECK_FUNCTION_EXISTS(memmove HAVE_MEMMOVE) +CHECK_FUNCTION_EXISTS(strerror HAVE_STRERROR) +CHECK_FUNCTION_EXISTS(strtoll HAVE_STRTOLL) +CHECK_FUNCTION_EXISTS(strtoq HAVE_STRTOQ) +CHECK_FUNCTION_EXISTS(_strtoi64 HAVE__STRTOI64) + +CHECK_TYPE_SIZE("long long" LONG_LONG) +CHECK_TYPE_SIZE("unsigned long long" UNSIGNED_LONG_LONG) + +# User-configurable options +# +# (Note: CMakeSetup displays these in alphabetical order, regardless of +# the order we use here) + +SET(BUILD_SHARED_LIBS OFF CACHE BOOL + "Build shared libraries instead of static ones.") + +OPTION(PCRE_BUILD_PCRE8 "Build 8 bit PCRE library" ON) + +OPTION(PCRE_BUILD_PCRE16 "Build 16 bit PCRE library" OFF) + +OPTION(PCRE_BUILD_PCRE32 "Build 32 bit PCRE library" OFF) + +OPTION(PCRE_BUILD_PCRECPP "Build the PCRE C++ library (pcrecpp)." ON) + +SET(PCRE_EBCDIC OFF CACHE BOOL + "Use EBCDIC coding instead of ASCII. (This is rarely used outside of mainframe systems.)") + +SET(PCRE_EBCDIC_NL25 OFF CACHE BOOL + "Use 0x25 as EBCDIC NL character instead of 0x15; implies EBCDIC.") + +SET(PCRE_LINK_SIZE "2" CACHE STRING + "Internal link size (2, 3 or 4 allowed). See LINK_SIZE in config.h.in for details.") + +SET(PCRE_MATCH_LIMIT "10000000" CACHE STRING + "Default limit on internal looping. See MATCH_LIMIT in config.h.in for details.") + +SET(PCRE_MATCH_LIMIT_RECURSION "MATCH_LIMIT" CACHE STRING + "Default limit on internal recursion. See MATCH_LIMIT_RECURSION in config.h.in for details.") + +SET(PCREGREP_BUFSIZE "20480" CACHE STRING + "Buffer size parameter for pcregrep. See PCREGREP_BUFSIZE in config.h.in for details.") + +SET(PCRE_NEWLINE "LF" CACHE STRING + "What to recognize as a newline (one of CR, LF, CRLF, ANY, ANYCRLF).") + +SET(PCRE_NO_RECURSE OFF CACHE BOOL + "If ON, then don't use stack recursion when matching. See NO_RECURSE in config.h.in for details.") + +SET(PCRE_POSIX_MALLOC_THRESHOLD "10" CACHE STRING + "Threshold for malloc() usage. See POSIX_MALLOC_THRESHOLD in config.h.in for details.") + +SET(PCRE_SUPPORT_JIT OFF CACHE BOOL + "Enable support for Just-in-time compiling.") + +SET(PCRE_SUPPORT_PCREGREP_JIT ON CACHE BOOL + "Enable use of Just-in-time compiling in pcregrep.") + +SET(PCRE_SUPPORT_UTF OFF CACHE BOOL + "Enable support for Unicode Transformation Format (UTF-8/UTF-16/UTF-32) encoding.") + +SET(PCRE_SUPPORT_UNICODE_PROPERTIES OFF CACHE BOOL + "Enable support for Unicode properties (if set, UTF support will be enabled as well).") + +SET(PCRE_SUPPORT_BSR_ANYCRLF OFF CACHE BOOL + "ON=Backslash-R matches only LF CR and CRLF, OFF=Backslash-R matches all Unicode Linebreaks") + +SET(PCRE_SUPPORT_VALGRIND OFF CACHE BOOL + "Enable Valgrind support.") + +SET(PCRE_SUPPORT_COVERAGE OFF CACHE BOOL + "Enable code coverage support using gcov.") + +OPTION(PCRE_SHOW_REPORT "Show the final configuration report" ON) +OPTION(PCRE_BUILD_PCREGREP "Build pcregrep" ON) +OPTION(PCRE_BUILD_TESTS "Build the tests" ON) + +IF (MINGW) + OPTION(NON_STANDARD_LIB_PREFIX + "ON=Shared libraries built in mingw will be named pcre.dll, etc., instead of libpcre.dll, etc." + OFF) + + OPTION(NON_STANDARD_LIB_SUFFIX + "ON=Shared libraries built in mingw will be named libpcre-0.dll, etc., instead of libpcre.dll, etc." + OFF) +ENDIF(MINGW) + +# bzip2 lib +IF(BZIP2_FOUND) + OPTION (PCRE_SUPPORT_LIBBZ2 "Enable support for linking pcregrep with libbz2." ON) +ENDIF(BZIP2_FOUND) +IF(PCRE_SUPPORT_LIBBZ2) + INCLUDE_DIRECTORIES(${BZIP2_INCLUDE_DIR}) +ENDIF(PCRE_SUPPORT_LIBBZ2) + +# zlib +IF(ZLIB_FOUND) + OPTION (PCRE_SUPPORT_LIBZ "Enable support for linking pcregrep with libz." ON) +ENDIF(ZLIB_FOUND) +IF(PCRE_SUPPORT_LIBZ) + INCLUDE_DIRECTORIES(${ZLIB_INCLUDE_DIR}) +ENDIF(PCRE_SUPPORT_LIBZ) + +# editline lib +IF(EDITLINE_FOUND) + OPTION (PCRE_SUPPORT_LIBEDIT "Enable support for linking pcretest with libedit." OFF) +ENDIF(EDITLINE_FOUND) +IF(PCRE_SUPPORT_LIBEDIT) + INCLUDE_DIRECTORIES(${EDITLINE_INCLUDE_DIR}) +ENDIF(PCRE_SUPPORT_LIBEDIT) + +# readline lib +IF(READLINE_FOUND) + OPTION (PCRE_SUPPORT_LIBREADLINE "Enable support for linking pcretest with libreadline." ON) +ENDIF(READLINE_FOUND) +IF(PCRE_SUPPORT_LIBREADLINE) + INCLUDE_DIRECTORIES(${READLINE_INCLUDE_DIR}) +ENDIF(PCRE_SUPPORT_LIBREADLINE) + +# Prepare build configuration + +SET(pcre_have_type_traits 0) +SET(pcre_have_bits_type_traits 0) + +IF(HAVE_TYPE_TRAITS_H) + SET(pcre_have_type_traits 1) +ENDIF(HAVE_TYPE_TRAITS_H) + +IF(HAVE_BITS_TYPE_TRAITS_H) + SET(pcre_have_bits_type_traits 1) +ENDIF(HAVE_BITS_TYPE_TRAITS_H) + +SET(pcre_have_long_long 0) +SET(pcre_have_ulong_long 0) + +IF(HAVE_LONG_LONG) + SET(pcre_have_long_long 1) +ENDIF(HAVE_LONG_LONG) + +IF(HAVE_UNSIGNED_LONG_LONG) + SET(pcre_have_ulong_long 1) +ENDIF(HAVE_UNSIGNED_LONG_LONG) + +IF(NOT BUILD_SHARED_LIBS) + SET(PCRE_STATIC 1) +ENDIF(NOT BUILD_SHARED_LIBS) + +IF(NOT PCRE_BUILD_PCRE8 AND NOT PCRE_BUILD_PCRE16 AND NOT PCRE_BUILD_PCRE32) + MESSAGE(FATAL_ERROR "At least one of PCRE_BUILD_PCRE8, PCRE_BUILD_PCRE16 or PCRE_BUILD_PCRE32 must be enabled") +ENDIF(NOT PCRE_BUILD_PCRE8 AND NOT PCRE_BUILD_PCRE16 AND NOT PCRE_BUILD_PCRE32) + +IF(PCRE_BUILD_PCRE8) + SET(SUPPORT_PCRE8 1) +ENDIF(PCRE_BUILD_PCRE8) + +IF(PCRE_BUILD_PCRE16) + SET(SUPPORT_PCRE16 1) +ENDIF(PCRE_BUILD_PCRE16) + +IF(PCRE_BUILD_PCRE32) + SET(SUPPORT_PCRE32 1) +ENDIF(PCRE_BUILD_PCRE32) + +IF(PCRE_BUILD_PCRECPP AND NOT PCRE_BUILD_PCRE8) + MESSAGE(STATUS "** PCRE_BUILD_PCRE8 must be enabled for the C++ library support") + SET(PCRE_BUILD_PCRECPP OFF) +ENDIF(PCRE_BUILD_PCRECPP AND NOT PCRE_BUILD_PCRE8) + +IF(PCRE_BUILD_PCREGREP AND NOT PCRE_BUILD_PCRE8) + MESSAGE(STATUS "** PCRE_BUILD_PCRE8 must be enabled for the pcregrep program") + SET(PCRE_BUILD_PCREGREP OFF) +ENDIF(PCRE_BUILD_PCREGREP AND NOT PCRE_BUILD_PCRE8) + +IF(PCRE_SUPPORT_LIBREADLINE AND PCRE_SUPPORT_LIBEDIT) + MESSAGE(FATAL_ERROR "Only one of libreadline or libeditline can be specified") +ENDIF(PCRE_SUPPORT_LIBREADLINE AND PCRE_SUPPORT_LIBEDIT) + +IF(PCRE_SUPPORT_BSR_ANYCRLF) + SET(BSR_ANYCRLF 1) +ENDIF(PCRE_SUPPORT_BSR_ANYCRLF) + +IF(PCRE_SUPPORT_UTF OR PCRE_SUPPORT_UNICODE_PROPERTIES) + SET(SUPPORT_UTF 1) + SET(PCRE_SUPPORT_UTF ON) +ENDIF(PCRE_SUPPORT_UTF OR PCRE_SUPPORT_UNICODE_PROPERTIES) + +IF(PCRE_SUPPORT_UNICODE_PROPERTIES) + SET(SUPPORT_UCP 1) +ENDIF(PCRE_SUPPORT_UNICODE_PROPERTIES) + +IF(PCRE_SUPPORT_JIT) + SET(SUPPORT_JIT 1) +ENDIF(PCRE_SUPPORT_JIT) + +IF(PCRE_SUPPORT_PCREGREP_JIT) + SET(SUPPORT_PCREGREP_JIT 1) +ENDIF(PCRE_SUPPORT_PCREGREP_JIT) + +IF(PCRE_SUPPORT_VALGRIND) + SET(SUPPORT_VALGRIND 1) +ENDIF(PCRE_SUPPORT_VALGRIND) + +IF(PCRE_SUPPORT_COVERAGE) + SET(SUPPORT_GCOV 1) + IF(NOT CMAKE_COMPILER_IS_GNUCC) + MESSAGE(FATAL_ERROR "Code coverage reports can only be generated when using GCC") + ENDIF(NOT CMAKE_COMPILER_IS_GNUCC) +ENDIF(PCRE_SUPPORT_COVERAGE) + +# This next one used to contain +# SET(PCRETEST_LIBS ${READLINE_LIBRARY}) +# but I was advised to add the NCURSES test as well, along with +# some modifications to cmake/FindReadline.cmake which should +# make it possible to override the default if necessary. PH + +IF(PCRE_SUPPORT_LIBREADLINE) + SET(SUPPORT_LIBREADLINE 1) + SET(PCRETEST_LIBS ${READLINE_LIBRARY} ${NCURSES_LIBRARY}) +ENDIF(PCRE_SUPPORT_LIBREADLINE) + +# libedit is a plug-compatible alternative to libreadline + +IF(PCRE_SUPPORT_LIBEDIT) + SET(SUPPORT_LIBEDIT 1) + SET(PCRETEST_LIBS ${EDITLINE_LIBRARY} ${NCURSES_LIBRARY}) +ENDIF(PCRE_SUPPORT_LIBEDIT) + +IF(PCRE_SUPPORT_LIBZ) + SET(SUPPORT_LIBZ 1) + SET(PCREGREP_LIBS ${PCREGREP_LIBS} ${ZLIB_LIBRARIES}) +ENDIF(PCRE_SUPPORT_LIBZ) + +IF(PCRE_SUPPORT_LIBBZ2) + SET(SUPPORT_LIBBZ2 1) + SET(PCREGREP_LIBS ${PCREGREP_LIBS} ${BZIP2_LIBRARIES}) +ENDIF(PCRE_SUPPORT_LIBBZ2) + +SET(NEWLINE "") + +IF(PCRE_NEWLINE STREQUAL "LF") + SET(NEWLINE "10") +ENDIF(PCRE_NEWLINE STREQUAL "LF") +IF(PCRE_NEWLINE STREQUAL "CR") + SET(NEWLINE "13") +ENDIF(PCRE_NEWLINE STREQUAL "CR") +IF(PCRE_NEWLINE STREQUAL "CRLF") + SET(NEWLINE "3338") +ENDIF(PCRE_NEWLINE STREQUAL "CRLF") +IF(PCRE_NEWLINE STREQUAL "ANY") + SET(NEWLINE "-1") +ENDIF(PCRE_NEWLINE STREQUAL "ANY") +IF(PCRE_NEWLINE STREQUAL "ANYCRLF") + SET(NEWLINE "-2") +ENDIF(PCRE_NEWLINE STREQUAL "ANYCRLF") + +IF(NEWLINE STREQUAL "") + MESSAGE(FATAL_ERROR "The PCRE_NEWLINE variable must be set to one of the following values: \"LF\", \"CR\", \"CRLF\", \"ANY\", \"ANYCRLF\".") +ENDIF(NEWLINE STREQUAL "") + +IF(PCRE_EBCDIC) + SET(EBCDIC 1) +IF(PCRE_NEWLINE STREQUAL "LF") + SET(NEWLINE "21") +ENDIF(PCRE_NEWLINE STREQUAL "LF") +IF(PCRE_NEWLINE STREQUAL "CRLF") + SET(NEWLINE "3349") +ENDIF(PCRE_NEWLINE STREQUAL "CRLF") +ENDIF(PCRE_EBCDIC) + +IF(PCRE_EBCDIC_NL25) + SET(EBCDIC 1) + SET(EBCDIC_NL25 1) +IF(PCRE_NEWLINE STREQUAL "LF") + SET(NEWLINE "37") +ENDIF(PCRE_NEWLINE STREQUAL "LF") +IF(PCRE_NEWLINE STREQUAL "CRLF") + SET(NEWLINE "3365") +ENDIF(PCRE_NEWLINE STREQUAL "CRLF") +ENDIF(PCRE_EBCDIC_NL25) + +IF(PCRE_NO_RECURSE) + SET(NO_RECURSE 1) +ENDIF(PCRE_NO_RECURSE) + +# Output files +CONFIGURE_FILE(config-cmake.h.in + ${PROJECT_BINARY_DIR}/config.h + @ONLY) + +# Parse version numbers and date out of configure.ac + +file(STRINGS ${PROJECT_SOURCE_DIR}/configure.ac + configure_lines + LIMIT_COUNT 50 # Read only the first 50 lines of the file +) + +set(SEARCHED_VARIABLES "pcre_major" "pcre_minor" "pcre_prerelease" "pcre_date") +foreach(configure_line ${configure_lines}) + foreach(_substitution_variable ${SEARCHED_VARIABLES}) + string(TOUPPER ${_substitution_variable} _substitution_variable_upper) + if (NOT ${_substitution_variable_upper}) + string(REGEX MATCH "m4_define\\(${_substitution_variable}, \\[(.*)\\]" MACTHED_STRING ${configure_line}) + if (CMAKE_MATCH_1) + set(${_substitution_variable_upper} ${CMAKE_MATCH_1}) + endif() + endif() + endforeach() +endforeach() + +CONFIGURE_FILE(pcre.h.in + ${PROJECT_BINARY_DIR}/pcre.h + @ONLY) + +# What about pcre-config and libpcre.pc? + +IF(PCRE_BUILD_PCRECPP) + CONFIGURE_FILE(pcre_stringpiece.h.in + ${PROJECT_BINARY_DIR}/pcre_stringpiece.h + @ONLY) + + CONFIGURE_FILE(pcrecpparg.h.in + ${PROJECT_BINARY_DIR}/pcrecpparg.h + @ONLY) +ENDIF(PCRE_BUILD_PCRECPP) + +# Character table generation + +OPTION(PCRE_REBUILD_CHARTABLES "Rebuild char tables" OFF) +IF(PCRE_REBUILD_CHARTABLES) + ADD_EXECUTABLE(dftables dftables.c) + + GET_TARGET_PROPERTY(DFTABLES_EXE dftables LOCATION) + + ADD_CUSTOM_COMMAND( + COMMENT "Generating character tables (pcre_chartables.c) for current locale" + DEPENDS dftables + COMMAND ${DFTABLES_EXE} + ARGS ${PROJECT_BINARY_DIR}/pcre_chartables.c + OUTPUT ${PROJECT_BINARY_DIR}/pcre_chartables.c + ) +ELSE(PCRE_REBUILD_CHARTABLES) + CONFIGURE_FILE(${PROJECT_SOURCE_DIR}/pcre_chartables.c.dist + ${PROJECT_BINARY_DIR}/pcre_chartables.c + COPYONLY) +ENDIF(PCRE_REBUILD_CHARTABLES) + +# Source code + +SET(PCRE_HEADERS ${PROJECT_BINARY_DIR}/pcre.h) + +IF(PCRE_BUILD_PCRE8) +SET(PCRE_SOURCES + pcre_byte_order.c + pcre_chartables.c + pcre_compile.c + pcre_config.c + pcre_dfa_exec.c + pcre_exec.c + pcre_fullinfo.c + pcre_get.c + pcre_globals.c + pcre_jit_compile.c + pcre_maketables.c + pcre_newline.c + pcre_ord2utf8.c + pcre_refcount.c + pcre_string_utils.c + pcre_study.c + pcre_tables.c + pcre_ucd.c + pcre_valid_utf8.c + pcre_version.c + pcre_xclass.c +) + +SET(PCREPOSIX_HEADERS pcreposix.h) + +SET(PCREPOSIX_SOURCES pcreposix.c) + +ENDIF(PCRE_BUILD_PCRE8) + +IF(PCRE_BUILD_PCRE16) +SET(PCRE16_SOURCES + pcre16_byte_order.c + pcre16_chartables.c + pcre16_compile.c + pcre16_config.c + pcre16_dfa_exec.c + pcre16_exec.c + pcre16_fullinfo.c + pcre16_get.c + pcre16_globals.c + pcre16_jit_compile.c + pcre16_maketables.c + pcre16_newline.c + pcre16_ord2utf16.c + pcre16_refcount.c + pcre16_string_utils.c + pcre16_study.c + pcre16_tables.c + pcre16_ucd.c + pcre16_utf16_utils.c + pcre16_valid_utf16.c + pcre16_version.c + pcre16_xclass.c +) +ENDIF(PCRE_BUILD_PCRE16) + +IF(PCRE_BUILD_PCRE32) +SET(PCRE32_SOURCES + pcre32_byte_order.c + pcre32_chartables.c + pcre32_compile.c + pcre32_config.c + pcre32_dfa_exec.c + pcre32_exec.c + pcre32_fullinfo.c + pcre32_get.c + pcre32_globals.c + pcre32_jit_compile.c + pcre32_maketables.c + pcre32_newline.c + pcre32_ord2utf32.c + pcre32_refcount.c + pcre32_string_utils.c + pcre32_study.c + pcre32_tables.c + pcre32_ucd.c + pcre32_utf32_utils.c + pcre32_valid_utf32.c + pcre32_version.c + pcre32_xclass.c +) +ENDIF(PCRE_BUILD_PCRE32) + +IF(MINGW AND NOT PCRE_STATIC) +IF (EXISTS ${PROJECT_SOURCE_DIR}/pcre.rc) +ADD_CUSTOM_COMMAND(OUTPUT ${PROJECT_SOURCE_DIR}/pcre.o +PRE-LINK +COMMAND windres ARGS pcre.rc pcre.o +WORKING_DIRECTORY ${PROJECT_SOURCE_DIR} +COMMENT Using pcre coff info in mingw build) +SET(PCRE_SOURCES + ${PCRE_SOURCES} ${PROJECT_SOURCE_DIR}/pcre.o +) +ENDIF(EXISTS ${PROJECT_SOURCE_DIR}/pcre.rc) +IF (EXISTS ${PROJECT_SOURCE_DIR}/pcreposix.rc) +ADD_CUSTOM_COMMAND(OUTPUT ${PROJECT_SOURCE_DIR}/pcreposix.o +PRE-LINK +COMMAND windres ARGS pcreposix.rc pcreposix.o +WORKING_DIRECTORY ${PROJECT_SOURCE_DIR} +COMMENT Using pcreposix coff info in mingw build) +SET(PCREPOSIX_SOURCES + ${PCREPOSIX_SOURCES} ${PROJECT_SOURCE_DIR}/pcreposix.o +) +ENDIF(EXISTS ${PROJECT_SOURCE_DIR}/pcreposix.rc) +ENDIF(MINGW AND NOT PCRE_STATIC) + +SET(PCRECPP_HEADERS + pcrecpp.h + pcre_scanner.h + ${PROJECT_BINARY_DIR}/pcrecpparg.h + ${PROJECT_BINARY_DIR}/pcre_stringpiece.h +) + +SET(PCRECPP_SOURCES + pcrecpp.cc + pcre_scanner.cc + pcre_stringpiece.cc +) + +# Build setup + +ADD_DEFINITIONS(-DHAVE_CONFIG_H) + +IF(MSVC) + ADD_DEFINITIONS(-D_CRT_SECURE_NO_DEPRECATE -D_CRT_SECURE_NO_WARNINGS) +ENDIF(MSVC) + +SET(CMAKE_INCLUDE_CURRENT_DIR 1) +# needed to make sure to not link debug libs +# against release libs and vice versa +IF(WIN32) + SET(CMAKE_DEBUG_POSTFIX "d") +ENDIF(WIN32) + +SET(targets) + +# Libraries +# pcre +IF(PCRE_BUILD_PCRE8) +ADD_LIBRARY(pcre ${PCRE_HEADERS} ${PCRE_SOURCES} ${PROJECT_BINARY_DIR}/config.h) +SET(targets ${targets} pcre) +ADD_LIBRARY(pcreposix ${PCREPOSIX_HEADERS} ${PCREPOSIX_SOURCES}) +SET(targets ${targets} pcreposix) +TARGET_LINK_LIBRARIES(pcreposix pcre) + +IF(MINGW AND NOT PCRE_STATIC) + IF(NON_STANDARD_LIB_PREFIX) + SET_TARGET_PROPERTIES(pcre pcreposix + PROPERTIES PREFIX "" + ) + ENDIF(NON_STANDARD_LIB_PREFIX) + + IF(NON_STANDARD_LIB_SUFFIX) + SET_TARGET_PROPERTIES(pcre pcreposix + PROPERTIES SUFFIX "-0.dll" + ) + ENDIF(NON_STANDARD_LIB_SUFFIX) +ENDIF(MINGW AND NOT PCRE_STATIC) + +ENDIF(PCRE_BUILD_PCRE8) + +IF(PCRE_BUILD_PCRE16) +ADD_LIBRARY(pcre16 ${PCRE_HEADERS} ${PCRE16_SOURCES} ${PROJECT_BINARY_DIR}/config.h) +SET(targets ${targets} pcre16) + +IF(MINGW AND NOT PCRE_STATIC) + IF(NON_STANDARD_LIB_PREFIX) + SET_TARGET_PROPERTIES(pcre16 + PROPERTIES PREFIX "" + ) + ENDIF(NON_STANDARD_LIB_PREFIX) + + IF(NON_STANDARD_LIB_SUFFIX) + SET_TARGET_PROPERTIES(pcre16 + PROPERTIES SUFFIX "-0.dll" + ) + ENDIF(NON_STANDARD_LIB_SUFFIX) +ENDIF(MINGW AND NOT PCRE_STATIC) + +ENDIF(PCRE_BUILD_PCRE16) + +IF(PCRE_BUILD_PCRE32) +ADD_LIBRARY(pcre32 ${PCRE_HEADERS} ${PCRE32_SOURCES} ${PROJECT_BINARY_DIR}/config.h) +SET(targets ${targets} pcre32) + +IF(MINGW AND NOT PCRE_STATIC) + IF(NON_STANDARD_LIB_PREFIX) + SET_TARGET_PROPERTIES(pcre32 + PROPERTIES PREFIX "" + ) + ENDIF(NON_STANDARD_LIB_PREFIX) + + IF(NON_STANDARD_LIB_SUFFIX) + SET_TARGET_PROPERTIES(pcre32 + PROPERTIES SUFFIX "-0.dll" + ) + ENDIF(NON_STANDARD_LIB_SUFFIX) +ENDIF(MINGW AND NOT PCRE_STATIC) + +ENDIF(PCRE_BUILD_PCRE32) + +# pcrecpp +IF(PCRE_BUILD_PCRECPP) +ADD_LIBRARY(pcrecpp ${PCRECPP_HEADERS} ${PCRECPP_SOURCES}) +SET(targets ${targets} pcrecpp) +TARGET_LINK_LIBRARIES(pcrecpp pcre) + + IF(MINGW AND NOT PCRE_STATIC) + IF(NON_STANDARD_LIB_PREFIX) + SET_TARGET_PROPERTIES(pcrecpp + PROPERTIES PREFIX "" + ) + ENDIF(NON_STANDARD_LIB_PREFIX) + + IF(NON_STANDARD_LIB_SUFFIX) + SET_TARGET_PROPERTIES(pcrecpp + PROPERTIES SUFFIX "-0.dll" + ) + ENDIF(NON_STANDARD_LIB_SUFFIX) + ENDIF(MINGW AND NOT PCRE_STATIC) +ENDIF(PCRE_BUILD_PCRECPP) + + +# Executables + +# Removed by PH (2008-01-23) because pcredemo shouldn't really be built +# automatically, and it gave trouble in some environments anyway. +# ADD_EXECUTABLE(pcredemo pcredemo.c) +# TARGET_LINK_LIBRARIES(pcredemo pcreposix) +# IF(NOT BUILD_SHARED_LIBS) +# # make sure to not use declspec(dllimport) in static mode on windows +# SET_TARGET_PROPERTIES(pcredemo PROPERTIES COMPILE_FLAGS "-DPCRE_STATIC") +# ENDIF(NOT BUILD_SHARED_LIBS) + +IF(PCRE_BUILD_PCREGREP) + ADD_EXECUTABLE(pcregrep pcregrep.c) + SET(targets ${targets} pcregrep) + TARGET_LINK_LIBRARIES(pcregrep pcreposix ${PCREGREP_LIBS}) +ENDIF(PCRE_BUILD_PCREGREP) + +# Testing +IF(PCRE_BUILD_TESTS) + ENABLE_TESTING() + + SET(PCRETEST_SOURCES pcretest.c) + IF(PCRE_BUILD_PCRE8) + LIST(APPEND PCRETEST_SOURCES pcre_printint.c) + ENDIF(PCRE_BUILD_PCRE8) + IF(PCRE_BUILD_PCRE16) + LIST(APPEND PCRETEST_SOURCES pcre16_printint.c) + ENDIF(PCRE_BUILD_PCRE16) + IF(PCRE_BUILD_PCRE32) + LIST(APPEND PCRETEST_SOURCES pcre32_printint.c) + ENDIF(PCRE_BUILD_PCRE32) + + ADD_EXECUTABLE(pcretest ${PCRETEST_SOURCES}) + SET(targets ${targets} pcretest) + IF(PCRE_BUILD_PCRE8) + LIST(APPEND PCRETEST_LIBS pcreposix pcre) + ENDIF(PCRE_BUILD_PCRE8) + IF(PCRE_BUILD_PCRE16) + LIST(APPEND PCRETEST_LIBS pcre16) + ENDIF(PCRE_BUILD_PCRE16) + IF(PCRE_BUILD_PCRE32) + LIST(APPEND PCRETEST_LIBS pcre32) + ENDIF(PCRE_BUILD_PCRE32) + TARGET_LINK_LIBRARIES(pcretest ${PCRETEST_LIBS}) + + IF(PCRE_SUPPORT_JIT) + ADD_EXECUTABLE(pcre_jit_test pcre_jit_test.c) + SET(targets ${targets} pcre_jit_test) + SET(PCRE_JIT_TEST_LIBS ) + IF(PCRE_BUILD_PCRE8) + LIST(APPEND PCRE_JIT_TEST_LIBS pcre) + ENDIF(PCRE_BUILD_PCRE8) + IF(PCRE_BUILD_PCRE16) + LIST(APPEND PCRE_JIT_TEST_LIBS pcre16) + ENDIF(PCRE_BUILD_PCRE16) + IF(PCRE_BUILD_PCRE32) + LIST(APPEND PCRE_JIT_TEST_LIBS pcre32) + ENDIF(PCRE_BUILD_PCRE32) + TARGET_LINK_LIBRARIES(pcre_jit_test ${PCRE_JIT_TEST_LIBS}) + ENDIF(PCRE_SUPPORT_JIT) + + IF(PCRE_BUILD_PCRECPP) + ADD_EXECUTABLE(pcrecpp_unittest pcrecpp_unittest.cc) + SET(targets ${targets} pcrecpp_unittest) + TARGET_LINK_LIBRARIES(pcrecpp_unittest pcrecpp) + IF(MINGW AND NON_STANDARD_LIB_NAMES AND NOT PCRE_STATIC) + SET_TARGET_PROPERTIES(pcrecpp + PROPERTIES PREFIX "" + ) + ENDIF(MINGW AND NON_STANDARD_LIB_NAMES AND NOT PCRE_STATIC) + + ADD_EXECUTABLE(pcre_scanner_unittest pcre_scanner_unittest.cc) + SET(targets ${targets} pcre_scanner_unittest) + TARGET_LINK_LIBRARIES(pcre_scanner_unittest pcrecpp) + + ADD_EXECUTABLE(pcre_stringpiece_unittest pcre_stringpiece_unittest.cc) + SET(targets ${targets} pcre_stringpiece_unittest) + TARGET_LINK_LIBRARIES(pcre_stringpiece_unittest pcrecpp) + ENDIF(PCRE_BUILD_PCRECPP) + + # exes in Debug location tested by the RunTest shell script + # via "make test" + IF(PCRE_BUILD_PCREGREP) + GET_TARGET_PROPERTY(PCREGREP_EXE pcregrep DEBUG_LOCATION) + ENDIF(PCRE_BUILD_PCREGREP) + + GET_TARGET_PROPERTY(PCRETEST_EXE pcretest DEBUG_LOCATION) + STRING(REGEX REPLACE "[^/]+$" "" PCRETEST_DIR ${PCRETEST_EXE}) + +# ================================================= + # Write out a CTest configuration file + # + FILE(WRITE ${PROJECT_BINARY_DIR}/CTestCustom.ctest + "# This is a generated file. +MESSAGE(\"When testing is complete, review test output in the +\\\"${PROJECT_BINARY_DIR}/Testing/Temporary\\\" folder.\") +MESSAGE(\" \") +") + + FILE(WRITE ${PROJECT_BINARY_DIR}/pcre_test.sh + "#! /bin/sh +# This is a generated file. +srcdir=${PROJECT_SOURCE_DIR} +pcretest=${PCRETEST_EXE} +cd ${PCRETEST_DIR} +source ${PROJECT_SOURCE_DIR}/RunTest +if test \"$?\" != \"0\"; then exit 1; fi +# End +") + + IF(UNIX) + ADD_TEST(pcre_test sh ${PROJECT_BINARY_DIR}/pcre_test.sh) + ENDIF(UNIX) + + IF(PCRE_BUILD_PCREGREP) + FILE(WRITE ${PROJECT_BINARY_DIR}/pcre_grep_test.sh + "#! /bin/sh +# This is a generated file. +srcdir=${PROJECT_SOURCE_DIR} +pcregrep=${PCREGREP_EXE} +pcretest=${PCRETEST_EXE} +cd ${PCRETEST_DIR} +source ${PROJECT_SOURCE_DIR}/RunGrepTest +if test \"$?\" != \"0\"; then exit 1; fi +# End +") + + IF(UNIX) + ADD_TEST(pcre_grep_test sh ${PROJECT_BINARY_DIR}/pcre_grep_test.sh) + ENDIF(UNIX) + ENDIF(PCRE_BUILD_PCREGREP) + + IF(WIN32) + # Provide environment for executing the bat file version of RunTest + FILE(TO_NATIVE_PATH ${PROJECT_SOURCE_DIR} winsrc) + FILE(TO_NATIVE_PATH ${PROJECT_BINARY_DIR} winbin) + FILE(TO_NATIVE_PATH ${PCRETEST_EXE} winexe) + + FILE(WRITE ${PROJECT_BINARY_DIR}/pcre_test.bat + "\@REM This is a generated file. +\@echo off +setlocal +SET srcdir=\"${winsrc}\" +SET pcretest=\"${winexe}\" +if not [%CMAKE_CONFIG_TYPE%]==[] SET pcretest=\"${winbin}\\%CMAKE_CONFIG_TYPE%\\pcretest.exe\" +call %srcdir%\\RunTest.Bat +if errorlevel 1 exit /b 1 +echo RunTest.bat tests successfully completed +") + + ADD_TEST(NAME pcre_test_bat + COMMAND pcre_test.bat) + SET_TESTS_PROPERTIES(pcre_test_bat PROPERTIES + PASS_REGULAR_EXPRESSION "RunTest\\.bat tests successfully completed") + + IF("$ENV{OSTYPE}" STREQUAL "msys") + # Both the sh and bat file versions of RunTest are run if make test is used + # in msys + ADD_TEST(pcre_test_sh sh.exe ${PROJECT_BINARY_DIR}/pcre_test.sh) + IF(PCRE_BUILD_PCREGREP) + ADD_TEST(pcre_grep_test sh.exe ${PROJECT_BINARY_DIR}/pcre_grep_test.sh) + ENDIF(PCRE_BUILD_PCREGREP) + ENDIF("$ENV{OSTYPE}" STREQUAL "msys") + + ENDIF(WIN32) + + # Changed to accommodate testing whichever location was just built + + IF(PCRE_SUPPORT_JIT) + ADD_TEST(pcre_jit_test pcre_jit_test) + ENDIF(PCRE_SUPPORT_JIT) + + IF(PCRE_BUILD_PCRECPP) + ADD_TEST(pcrecpp_test pcrecpp_unittest) + ADD_TEST(pcre_scanner_test pcre_scanner_unittest) + ADD_TEST(pcre_stringpiece_test pcre_stringpiece_unittest) + ENDIF(PCRE_BUILD_PCRECPP) + +ENDIF(PCRE_BUILD_TESTS) + +# Installation +SET(CMAKE_INSTALL_ALWAYS 1) + +INSTALL(TARGETS ${targets} + RUNTIME DESTINATION bin + LIBRARY DESTINATION lib + ARCHIVE DESTINATION lib) + +INSTALL(FILES ${PCRE_HEADERS} ${PCREPOSIX_HEADERS} DESTINATION include) + +FILE(GLOB html ${PROJECT_SOURCE_DIR}/doc/html/*.html) +FILE(GLOB man1 ${PROJECT_SOURCE_DIR}/doc/*.1) +FILE(GLOB man3 ${PROJECT_SOURCE_DIR}/doc/*.3) + +IF(PCRE_BUILD_PCRECPP) + INSTALL(FILES ${PCRECPP_HEADERS} DESTINATION include) +ELSE(PCRE_BUILD_PCRECPP) + # Remove pcrecpp.3 + FOREACH(man ${man3}) + GET_FILENAME_COMPONENT(man_tmp ${man} NAME) + IF(NOT man_tmp STREQUAL "pcrecpp.3") + SET(man3_new ${man3} ${man}) + ENDIF(NOT man_tmp STREQUAL "pcrecpp.3") + ENDFOREACH(man ${man3}) + SET(man3 ${man3_new}) +ENDIF(PCRE_BUILD_PCRECPP) + +INSTALL(FILES ${man1} DESTINATION man/man1) +INSTALL(FILES ${man3} DESTINATION man/man3) +INSTALL(FILES ${html} DESTINATION share/doc/pcre/html) + +# help, only for nice output +IF(BUILD_SHARED_LIBS) + SET(BUILD_STATIC_LIBS OFF) +ELSE(BUILD_SHARED_LIBS) + SET(BUILD_STATIC_LIBS ON) +ENDIF(BUILD_SHARED_LIBS) + +IF(PCRE_SHOW_REPORT) + STRING(TOUPPER "${CMAKE_BUILD_TYPE}" buildtype) + IF (CMAKE_C_FLAGS) + SET(cfsp " ") + ENDIF(CMAKE_C_FLAGS) + IF (CMAKE_CXX_FLAGS) + SET(cxxfsp " ") + ENDIF(CMAKE_CXX_FLAGS) + MESSAGE(STATUS "") + MESSAGE(STATUS "") + MESSAGE(STATUS "PCRE configuration summary:") + MESSAGE(STATUS "") + MESSAGE(STATUS " Install prefix .................. : ${CMAKE_INSTALL_PREFIX}") + MESSAGE(STATUS " C compiler ...................... : ${CMAKE_C_COMPILER}") + MESSAGE(STATUS " C++ compiler .................... : ${CMAKE_CXX_COMPILER}") + MESSAGE(STATUS " C compiler flags ................ : ${CMAKE_C_FLAGS}${cfsp}${CMAKE_C_FLAGS_${buildtype}}") + MESSAGE(STATUS " C++ compiler flags .............. : ${CMAKE_CXX_FLAGS}${cxxfsp}${CMAKE_CXX_FLAGS_${buildtype}}") + MESSAGE(STATUS "") + MESSAGE(STATUS " Build 8 bit PCRE library ........ : ${PCRE_BUILD_PCRE8}") + MESSAGE(STATUS " Build 16 bit PCRE library ....... : ${PCRE_BUILD_PCRE16}") + MESSAGE(STATUS " Build 32 bit PCRE library ....... : ${PCRE_BUILD_PCRE32}") + MESSAGE(STATUS " Build C++ library ............... : ${PCRE_BUILD_PCRECPP}") + MESSAGE(STATUS " Enable JIT compiling support .... : ${PCRE_SUPPORT_JIT}") + MESSAGE(STATUS " Enable UTF support .............. : ${PCRE_SUPPORT_UTF}") + MESSAGE(STATUS " Unicode properties .............. : ${PCRE_SUPPORT_UNICODE_PROPERTIES}") + MESSAGE(STATUS " Newline char/sequence ........... : ${PCRE_NEWLINE}") + MESSAGE(STATUS " \\R matches only ANYCRLF ......... : ${PCRE_SUPPORT_BSR_ANYCRLF}") + MESSAGE(STATUS " EBCDIC coding ................... : ${PCRE_EBCDIC}") + MESSAGE(STATUS " EBCDIC coding with NL=0x25 ...... : ${PCRE_EBCDIC_NL25}") + MESSAGE(STATUS " Rebuild char tables ............. : ${PCRE_REBUILD_CHARTABLES}") + MESSAGE(STATUS " No stack recursion .............. : ${PCRE_NO_RECURSE}") + MESSAGE(STATUS " POSIX mem threshold ............. : ${PCRE_POSIX_MALLOC_THRESHOLD}") + MESSAGE(STATUS " Internal link size .............. : ${PCRE_LINK_SIZE}") + MESSAGE(STATUS " Match limit ..................... : ${PCRE_MATCH_LIMIT}") + MESSAGE(STATUS " Match limit recursion ........... : ${PCRE_MATCH_LIMIT_RECURSION}") + MESSAGE(STATUS " Build shared libs ............... : ${BUILD_SHARED_LIBS}") + MESSAGE(STATUS " Build static libs ............... : ${BUILD_STATIC_LIBS}") + MESSAGE(STATUS " Build pcregrep .................. : ${PCRE_BUILD_PCREGREP}") + MESSAGE(STATUS " Enable JIT in pcregrep .......... : ${PCRE_SUPPORT_PCREGREP_JIT}") + MESSAGE(STATUS " Buffer size for pcregrep ........ : ${PCREGREP_BUFSIZE}") + MESSAGE(STATUS " Build tests (implies pcretest .. : ${PCRE_BUILD_TESTS}") + MESSAGE(STATUS " and pcregrep)") + IF(ZLIB_FOUND) + MESSAGE(STATUS " Link pcregrep with libz ......... : ${PCRE_SUPPORT_LIBZ}") + ELSE(ZLIB_FOUND) + MESSAGE(STATUS " Link pcregrep with libz ......... : Library not found" ) + ENDIF(ZLIB_FOUND) + IF(BZIP2_FOUND) + MESSAGE(STATUS " Link pcregrep with libbz2 ....... : ${PCRE_SUPPORT_LIBBZ2}") + ELSE(BZIP2_FOUND) + MESSAGE(STATUS " Link pcregrep with libbz2 ....... : Library not found" ) + ENDIF(BZIP2_FOUND) + IF(EDITLINE_FOUND) + MESSAGE(STATUS " Link pcretest with libeditline .. : ${PCRE_SUPPORT_LIBEDIT}") + ELSE(EDITLINE_FOUND) + MESSAGE(STATUS " Link pcretest with libeditline .. : Library not found" ) + ENDIF(EDITLINE_FOUND) + IF(READLINE_FOUND) + MESSAGE(STATUS " Link pcretest with libreadline .. : ${PCRE_SUPPORT_LIBREADLINE}") + ELSE(READLINE_FOUND) + MESSAGE(STATUS " Link pcretest with libreadline .. : Library not found" ) + ENDIF(READLINE_FOUND) + MESSAGE(STATUS " Support Valgrind .................: ${PCRE_SUPPORT_VALGRIND}") + MESSAGE(STATUS " Support coverage .................: ${PCRE_SUPPORT_COVERAGE}") + + IF(MINGW AND NOT PCRE_STATIC) + MESSAGE(STATUS " Non-standard dll names (prefix) . : ${NON_STANDARD_LIB_PREFIX}") + MESSAGE(STATUS " Non-standard dll names (suffix) . : ${NON_STANDARD_LIB_SUFFIX}") + ENDIF(MINGW AND NOT PCRE_STATIC) + MESSAGE(STATUS "") +ENDIF(PCRE_SHOW_REPORT) + +# end CMakeLists.txt diff --git a/pcre/ChangeLog b/pcre/ChangeLog new file mode 100644 index 00000000000..21cada49c9e --- /dev/null +++ b/pcre/ChangeLog @@ -0,0 +1,5149 @@ +ChangeLog for PCRE +------------------ + +Version 8.33 28-May-2013 +-------------------------- + +1. Added 'U' to some constants that are compared to unsigned integers, to + avoid compiler signed/unsigned warnings. Added (int) casts to unsigned + variables that are added to signed variables, to ensure the result is + signed and can be negated. + +2. Applied patch by Daniel Richard G for quashing MSVC warnings to the + CMake config files. + +3. Revise the creation of config.h.generic so that all boolean macros are + #undefined, whereas non-boolean macros are #ifndef/#endif-ed. This makes + overriding via -D on the command line possible. + +4. Changing the definition of the variable "op" in pcre_exec.c from pcre_uchar + to unsigned int is reported to make a quite noticeable speed difference in + a specific Windows environment. Testing on Linux did also appear to show + some benefit (and it is clearly not harmful). Also fixed the definition of + Xop which should be unsigned. + +5. Related to (4), changing the definition of the intermediate variable cc + in repeated character loops from pcre_uchar to pcre_uint32 also gave speed + improvements. + +6. Fix forward search in JIT when link size is 3 or greater. Also removed some + unnecessary spaces. + +7. Adjust autogen.sh and configure.ac to lose warnings given by automake 1.12 + and later. + +8. Fix two buffer over read issues in 16 and 32 bit modes. Affects JIT only. + +9. Optimizing fast_forward_start_bits in JIT. + +10. Adding support for callouts in JIT, and fixing some issues revealed + during this work. Namely: + + (a) Unoptimized capturing brackets incorrectly reset on backtrack. + + (b) Minimum length was not checked before the matching is started. + +11. The value of capture_last that is passed to callouts was incorrect in some + cases when there was a capture on one path that was subsequently abandoned + after a backtrack. Also, the capture_last value is now reset after a + recursion, since all captures are also reset in this case. + +12. The interpreter no longer returns the "too many substrings" error in the + case when an overflowing capture is in a branch that is subsequently + abandoned after a backtrack. + +13. In the pathological case when an offset vector of size 2 is used, pcretest + now prints out the matched string after a yield of 0 or 1. + +14. Inlining subpatterns in recursions, when certain conditions are fulfilled. + Only supported by the JIT compiler at the moment. + +15. JIT compiler now supports 32 bit Macs thanks to Lawrence Velazquez. + +16. Partial matches now set offsets[2] to the "bumpalong" value, that is, the + offset of the starting point of the matching process, provided the offsets + vector is large enough. + +17. The \A escape now records a lookbehind value of 1, though its execution + does not actually inspect the previous character. This is to ensure that, + in partial multi-segment matching, at least one character from the old + segment is retained when a new segment is processed. Otherwise, if there + are no lookbehinds in the pattern, \A might match incorrectly at the start + of a new segment. + +18. Added some #ifdef __VMS code into pcretest.c to help VMS implementations. + +19. Redefined some pcre_uchar variables in pcre_exec.c as pcre_uint32; this + gives some modest performance improvement in 8-bit mode. + +20. Added the PCRE-specific property \p{Xuc} for matching characters that can + be expressed in certain programming languages using Universal Character + Names. + +21. Unicode validation has been updated in the light of Unicode Corrigendum #9, + which points out that "non characters" are not "characters that may not + appear in Unicode strings" but rather "characters that are reserved for + internal use and have only local meaning". + +22. When a pattern was compiled with automatic callouts (PCRE_AUTO_CALLOUT) and + there was a conditional group that depended on an assertion, if the + assertion was false, the callout that immediately followed the alternation + in the condition was skipped when pcre_exec() was used for matching. + +23. Allow an explicit callout to be inserted before an assertion that is the + condition for a conditional group, for compatibility with automatic + callouts, which always insert a callout at this point. + +24. In 8.31, (*COMMIT) was confined to within a recursive subpattern. Perl also + confines (*SKIP) and (*PRUNE) in the same way, and this has now been done. + +25. (*PRUNE) is now supported by the JIT compiler. + +26. Fix infinite loop when /(?<=(*SKIP)ac)a/ is matched against aa. + +27. Fix the case where there are two or more SKIPs with arguments that may be + ignored. + +28. (*SKIP) is now supported by the JIT compiler. + +29. (*THEN) is now supported by the JIT compiler. + +30. Update RunTest with additional test selector options. + +31. The way PCRE handles backtracking verbs has been changed in two ways. + + (1) Previously, in something like (*COMMIT)(*SKIP), COMMIT would override + SKIP. Now, PCRE acts on whichever backtracking verb is reached first by + backtracking. In some cases this makes it more Perl-compatible, but Perl's + rather obscure rules do not always do the same thing. + + (2) Previously, backtracking verbs were confined within assertions. This is + no longer the case for positive assertions, except for (*ACCEPT). Again, + this sometimes improves Perl compatibility, and sometimes does not. + +32. A number of tests that were in test 2 because Perl did things differently + have been moved to test 1, because either Perl or PCRE has changed, and + these tests are now compatible. + +32. Backtracking control verbs are now handled in the same way in JIT and + interpreter. + +33. An opening parenthesis in a MARK/PRUNE/SKIP/THEN name in a pattern that + contained a forward subroutine reference caused a compile error. + +34. Auto-detect and optimize limited repetitions in JIT. + +35. Implement PCRE_NEVER_UTF to lock out the use of UTF, in particular, + blocking (*UTF) etc. + +36. In the interpreter, maximizing pattern repetitions for characters and + character types now use tail recursion, which reduces stack usage. + +37. The value of the max lookbehind was not correctly preserved if a compiled + and saved regex was reloaded on a host of different endianness. + +38. Implemented (*LIMIT_MATCH) and (*LIMIT_RECURSION). As part of the extension + of the compiled pattern block, expand the flags field from 16 to 32 bits + because it was almost full. + +39. Try madvise first before posix_madvise. + +40. Change 7 for PCRE 7.9 made it impossible for pcregrep to find empty lines + with a pattern such as ^$. It has taken 4 years for anybody to notice! The + original change locked out all matches of empty strings. This has been + changed so that one match of an empty string per line is recognized. + Subsequent searches on the same line (for colouring or for --only-matching, + for example) do not recognize empty strings. + +41. Applied a user patch to fix a number of spelling mistakes in comments. + +42. Data lines longer than 65536 caused pcretest to crash. + +43. Clarified the data type for length and startoffset arguments for pcre_exec + and pcre_dfa_exec in the function-specific man pages, where they were + explicitly stated to be in bytes, never having been updated. I also added + some clarification to the pcreapi man page. + +44. A call to pcre_dfa_exec() with an output vector size less than 2 caused + a segmentation fault. + + +Version 8.32 30-November-2012 +----------------------------- + +1. Improved JIT compiler optimizations for first character search and single + character iterators. + +2. Supporting IBM XL C compilers for PPC architectures in the JIT compiler. + Patch by Daniel Richard G. + +3. Single character iterator optimizations in the JIT compiler. + +4. Improved JIT compiler optimizations for character ranges. + +5. Rename the "leave" variable names to "quit" to improve WinCE compatibility. + Reported by Giuseppe D'Angelo. + +6. The PCRE_STARTLINE bit, indicating that a match can occur only at the start + of a line, was being set incorrectly in cases where .* appeared inside + atomic brackets at the start of a pattern, or where there was a subsequent + *PRUNE or *SKIP. + +7. Improved instruction cache flush for POWER/PowerPC. + Patch by Daniel Richard G. + +8. Fixed a number of issues in pcregrep, making it more compatible with GNU + grep: + + (a) There is now no limit to the number of patterns to be matched. + + (b) An error is given if a pattern is too long. + + (c) Multiple uses of --exclude, --exclude-dir, --include, and --include-dir + are now supported. + + (d) --exclude-from and --include-from (multiple use) have been added. + + (e) Exclusions and inclusions now apply to all files and directories, not + just to those obtained from scanning a directory recursively. + + (f) Multiple uses of -f and --file-list are now supported. + + (g) In a Windows environment, the default for -d has been changed from + "read" (the GNU grep default) to "skip", because otherwise the presence + of a directory in the file list provokes an error. + + (h) The documentation has been revised and clarified in places. + +9. Improve the matching speed of capturing brackets. + +10. Changed the meaning of \X so that it now matches a Unicode extended + grapheme cluster. + +11. Patch by Daniel Richard G to the autoconf files to add a macro for sorting + out POSIX threads when JIT support is configured. + +12. Added support for PCRE_STUDY_EXTRA_NEEDED. + +13. In the POSIX wrapper regcomp() function, setting re_nsub field in the preg + structure could go wrong in environments where size_t is not the same size + as int. + +14. Applied user-supplied patch to pcrecpp.cc to allow PCRE_NO_UTF8_CHECK to be + set. + +15. The EBCDIC support had decayed; later updates to the code had included + explicit references to (e.g.) \x0a instead of CHAR_LF. There has been a + general tidy up of EBCDIC-related issues, and the documentation was also + not quite right. There is now a test that can be run on ASCII systems to + check some of the EBCDIC-related things (but is it not a full test). + +16. The new PCRE_STUDY_EXTRA_NEEDED option is now used by pcregrep, resulting + in a small tidy to the code. + +17. Fix JIT tests when UTF is disabled and both 8 and 16 bit mode are enabled. + +18. If the --only-matching (-o) option in pcregrep is specified multiple + times, each one causes appropriate output. For example, -o1 -o2 outputs the + substrings matched by the 1st and 2nd capturing parentheses. A separating + string can be specified by --om-separator (default empty). + +19. Improving the first n character searches. + +20. Turn case lists for horizontal and vertical white space into macros so that + they are defined only once. + +21. This set of changes together give more compatible Unicode case-folding + behaviour for characters that have more than one other case when UCP + support is available. + + (a) The Unicode property table now has offsets into a new table of sets of + three or more characters that are case-equivalent. The MultiStage2.py + script that generates these tables (the pcre_ucd.c file) now scans + CaseFolding.txt instead of UnicodeData.txt for character case + information. + + (b) The code for adding characters or ranges of characters to a character + class has been abstracted into a generalized function that also handles + case-independence. In UTF-mode with UCP support, this uses the new data + to handle characters with more than one other case. + + (c) A bug that is fixed as a result of (b) is that codepoints less than 256 + whose other case is greater than 256 are now correctly matched + caselessly. Previously, the high codepoint matched the low one, but not + vice versa. + + (d) The processing of \h, \H, \v, and \ in character classes now makes use + of the new class addition function, using character lists defined as + macros alongside the case definitions of 20 above. + + (e) Caseless back references now work with characters that have more than + one other case. + + (f) General caseless matching of characters with more than one other case + is supported. + +22. Unicode character properties were updated from Unicode 6.2.0 + +23. Improved CMake support under Windows. Patch by Daniel Richard G. + +24. Add support for 32-bit character strings, and UTF-32 + +25. Major JIT compiler update (code refactoring and bugfixing). + Experimental Sparc 32 support is added. + +26. Applied a modified version of Daniel Richard G's patch to create + pcre.h.generic and config.h.generic by "make" instead of in the + PrepareRelease script. + +27. Added a definition for CHAR_NULL (helpful for the z/OS port), and use it in + pcre_compile.c when checking for a zero character. + +28. Introducing a native interface for JIT. Through this interface, the compiled + machine code can be directly executed. The purpose of this interface is to + provide fast pattern matching, so several sanity checks are not performed. + However, feature tests are still performed. The new interface provides + 1.4x speedup compared to the old one. + +29. If pcre_exec() or pcre_dfa_exec() was called with a negative value for + the subject string length, the error given was PCRE_ERROR_BADOFFSET, which + was confusing. There is now a new error PCRE_ERROR_BADLENGTH for this case. + +30. In 8-bit UTF-8 mode, pcretest failed to give an error for data codepoints + greater than 0x7fffffff (which cannot be represented in UTF-8, even under + the "old" RFC 2279). Instead, it ended up passing a negative length to + pcre_exec(). + +31. Add support for GCC's visibility feature to hide internal functions. + +32. Running "pcretest -C pcre8" or "pcretest -C pcre16" gave a spurious error + "unknown -C option" after outputting 0 or 1. + +33. There is now support for generating a code coverage report for the test + suite in environments where gcc is the compiler and lcov is installed. This + is mainly for the benefit of the developers. + +34. If PCRE is built with --enable-valgrind, certain memory regions are marked + unaddressable using valgrind annotations, allowing valgrind to detect + invalid memory accesses. This is mainly for the benefit of the developers. + +25. (*UTF) can now be used to start a pattern in any of the three libraries. + +26. Give configure error if --enable-cpp but no C++ compiler found. + + +Version 8.31 06-July-2012 +------------------------- + +1. Fixing a wrong JIT test case and some compiler warnings. + +2. Removed a bashism from the RunTest script. + +3. Add a cast to pcre_exec.c to fix the warning "unary minus operator applied + to unsigned type, result still unsigned" that was given by an MS compiler + on encountering the code "-sizeof(xxx)". + +4. Partial matching support is added to the JIT compiler. + +5. Fixed several bugs concerned with partial matching of items that consist + of more than one character: + + (a) /^(..)\1/ did not partially match "aba" because checking references was + done on an "all or nothing" basis. This also applied to repeated + references. + + (b) \R did not give a hard partial match if \r was found at the end of the + subject. + + (c) \X did not give a hard partial match after matching one or more + characters at the end of the subject. + + (d) When newline was set to CRLF, a pattern such as /a$/ did not recognize + a partial match for the string "\r". + + (e) When newline was set to CRLF, the metacharacter "." did not recognize + a partial match for a CR character at the end of the subject string. + +6. If JIT is requested using /S++ or -s++ (instead of just /S+ or -s+) when + running pcretest, the text "(JIT)" added to the output whenever JIT is + actually used to run the match. + +7. Individual JIT compile options can be set in pcretest by following -s+[+] + or /S+[+] with a digit between 1 and 7. + +8. OP_NOT now supports any UTF character not just single-byte ones. + +9. (*MARK) control verb is now supported by the JIT compiler. + +10. The command "./RunTest list" lists the available tests without actually + running any of them. (Because I keep forgetting what they all are.) + +11. Add PCRE_INFO_MAXLOOKBEHIND. + +12. Applied a (slightly modified) user-supplied patch that improves performance + when the heap is used for recursion (compiled with --disable-stack-for- + recursion). Instead of malloc and free for each heap frame each time a + logical recursion happens, frames are retained on a chain and re-used where + possible. This sometimes gives as much as 30% improvement. + +13. As documented, (*COMMIT) is now confined to within a recursive subpattern + call. + +14. As documented, (*COMMIT) is now confined to within a positive assertion. + +15. It is now possible to link pcretest with libedit as an alternative to + libreadline. + +16. (*COMMIT) control verb is now supported by the JIT compiler. + +17. The Unicode data tables have been updated to Unicode 6.1.0. + +18. Added --file-list option to pcregrep. + +19. Added binary file support to pcregrep, including the -a, --binary-files, + -I, and --text options. + +20. The madvise function is renamed for posix_madvise for QNX compatibility + reasons. Fixed by Giuseppe D'Angelo. + +21. Fixed a bug for backward assertions with REVERSE 0 in the JIT compiler. + +22. Changed the option for creating symbolic links for 16-bit man pages from + -s to -sf so that re-installing does not cause issues. + +23. Support PCRE_NO_START_OPTIMIZE in JIT as (*MARK) support requires it. + +24. Fixed a very old bug in pcretest that caused errors with restarted DFA + matches in certain environments (the workspace was not being correctly + retained). Also added to pcre_dfa_exec() a simple plausibility check on + some of the workspace data at the beginning of a restart. + +25. \s*\R was auto-possessifying the \s* when it should not, whereas \S*\R + was not doing so when it should - probably a typo introduced by SVN 528 + (change 8.10/14). + +26. When PCRE_UCP was not set, \w+\x{c4} was incorrectly auto-possessifying the + \w+ when the character tables indicated that \x{c4} was a word character. + There were several related cases, all because the tests for doing a table + lookup were testing for characters less than 127 instead of 255. + +27. If a pattern contains capturing parentheses that are not used in a match, + their slots in the ovector are set to -1. For those that are higher than + any matched groups, this happens at the end of processing. In the case when + there were back references that the ovector was too small to contain + (causing temporary malloc'd memory to be used during matching), and the + highest capturing number was not used, memory off the end of the ovector + was incorrectly being set to -1. (It was using the size of the temporary + memory instead of the true size.) + +28. To catch bugs like 27 using valgrind, when pcretest is asked to specify an + ovector size, it uses memory at the end of the block that it has got. + +29. Check for an overlong MARK name and give an error at compile time. The + limit is 255 for the 8-bit library and 65535 for the 16-bit library. + +30. JIT compiler update. + +31. JIT is now supported on jailbroken iOS devices. Thanks for Ruiger + Rill for the patch. + +32. Put spaces around SLJIT_PRINT_D in the JIT compiler. Required by CXX11. + +33. Variable renamings in the PCRE-JIT compiler. No functionality change. + +34. Fixed typos in pcregrep: in two places there was SUPPORT_LIBZ2 instead of + SUPPORT_LIBBZ2. This caused a build problem when bzip2 but not gzip (zlib) + was enabled. + +35. Improve JIT code generation for greedy plus quantifier. + +36. When /((?:a?)*)*c/ or /((?>a?)*)*c/ was matched against "aac", it set group + 1 to "aa" instead of to an empty string. The bug affected repeated groups + that could potentially match an empty string. + +37. Optimizing single character iterators in JIT. + +38. Wide characters specified with \uxxxx in JavaScript mode are now subject to + the same checks as \x{...} characters in non-JavaScript mode. Specifically, + codepoints that are too big for the mode are faulted, and in a UTF mode, + disallowed codepoints are also faulted. + +39. If PCRE was compiled with UTF support, in three places in the DFA + matcher there was code that should only have been obeyed in UTF mode, but + was being obeyed unconditionally. In 8-bit mode this could cause incorrect + processing when bytes with values greater than 127 were present. In 16-bit + mode the bug would be provoked by values in the range 0xfc00 to 0xdc00. In + both cases the values are those that cannot be the first data item in a UTF + character. The three items that might have provoked this were recursions, + possessively repeated groups, and atomic groups. + +40. Ensure that libpcre is explicitly listed in the link commands for pcretest + and pcregrep, because some OS require shared objects to be explicitly + passed to ld, causing the link step to fail if they are not. + +41. There were two incorrect #ifdefs in pcre_study.c, meaning that, in 16-bit + mode, patterns that started with \h* or \R* might be incorrectly matched. + + +Version 8.30 04-February-2012 +----------------------------- + +1. Renamed "isnumber" as "is_a_number" because in some Mac environments this + name is defined in ctype.h. + +2. Fixed a bug in fixed-length calculation for lookbehinds that would show up + only in quite long subpatterns. + +3. Removed the function pcre_info(), which has been obsolete and deprecated + since it was replaced by pcre_fullinfo() in February 2000. + +4. For a non-anchored pattern, if (*SKIP) was given with a name that did not + match a (*MARK), and the match failed at the start of the subject, a + reference to memory before the start of the subject could occur. This bug + was introduced by fix 17 of release 8.21. + +5. A reference to an unset group with zero minimum repetition was giving + totally wrong answers (in non-JavaScript-compatibility mode). For example, + /(another)?(\1?)test/ matched against "hello world test". This bug was + introduced in release 8.13. + +6. Add support for 16-bit character strings (a large amount of work involving + many changes and refactorings). + +7. RunGrepTest failed on msys because \r\n was replaced by whitespace when the + command "pattern=`printf 'xxx\r\njkl'`" was run. The pattern is now taken + from a file. + +8. Ovector size of 2 is also supported by JIT based pcre_exec (the ovector size + rounding is not applied in this particular case). + +9. The invalid Unicode surrogate codepoints U+D800 to U+DFFF are now rejected + if they appear, or are escaped, in patterns. + +10. Get rid of a number of -Wunused-but-set-variable warnings. + +11. The pattern /(?=(*:x))(q|)/ matches an empty string, and returns the mark + "x". The similar pattern /(?=(*:x))((*:y)q|)/ did not return a mark at all. + Oddly, Perl behaves the same way. PCRE has been fixed so that this pattern + also returns the mark "x". This bug applied to capturing parentheses, + non-capturing parentheses, and atomic parentheses. It also applied to some + assertions. + +12. Stephen Kelly's patch to CMakeLists.txt allows it to parse the version + information out of configure.ac instead of relying on pcre.h.generic, which + is not stored in the repository. + +13. Applied Dmitry V. Levin's patch for a more portable method for linking with + -lreadline. + +14. ZH added PCRE_CONFIG_JITTARGET; added its output to pcretest -C. + +15. Applied Graycode's patch to put the top-level frame on the stack rather + than the heap when not using the stack for recursion. This gives a + performance improvement in many cases when recursion is not deep. + +16. Experimental code added to "pcretest -C" to output the stack frame size. + + +Version 8.21 12-Dec-2011 +------------------------ + +1. Updating the JIT compiler. + +2. JIT compiler now supports OP_NCREF, OP_RREF and OP_NRREF. New test cases + are added as well. + +3. Fix cache-flush issue on PowerPC (It is still an experimental JIT port). + PCRE_EXTRA_TABLES is not suported by JIT, and should be checked before + calling _pcre_jit_exec. Some extra comments are added. + +4. (*MARK) settings inside atomic groups that do not contain any capturing + parentheses, for example, (?>a(*:m)), were not being passed out. This bug + was introduced by change 18 for 8.20. + +5. Supporting of \x, \U and \u in JavaScript compatibility mode based on the + ECMA-262 standard. + +6. Lookbehinds such as (?<=a{2}b) that contained a fixed repetition were + erroneously being rejected as "not fixed length" if PCRE_CASELESS was set. + This bug was probably introduced by change 9 of 8.13. + +7. While fixing 6 above, I noticed that a number of other items were being + incorrectly rejected as "not fixed length". This arose partly because newer + opcodes had not been added to the fixed-length checking code. I have (a) + corrected the bug and added tests for these items, and (b) arranged for an + error to occur if an unknown opcode is encountered while checking for fixed + length instead of just assuming "not fixed length". The items that were + rejected were: (*ACCEPT), (*COMMIT), (*FAIL), (*MARK), (*PRUNE), (*SKIP), + (*THEN), \h, \H, \v, \V, and single character negative classes with fixed + repetitions, e.g. [^a]{3}, with and without PCRE_CASELESS. + +8. A possessively repeated conditional subpattern such as (?(?=c)c|d)++ was + being incorrectly compiled and would have given unpredicatble results. + +9. A possessively repeated subpattern with minimum repeat count greater than + one behaved incorrectly. For example, (A){2,}+ behaved as if it was + (A)(A)++ which meant that, after a subsequent mismatch, backtracking into + the first (A) could occur when it should not. + +10. Add a cast and remove a redundant test from the code. + +11. JIT should use pcre_malloc/pcre_free for allocation. + +12. Updated pcre-config so that it no longer shows -L/usr/lib, which seems + best practice nowadays, and helps with cross-compiling. (If the exec_prefix + is anything other than /usr, -L is still shown). + +13. In non-UTF-8 mode, \C is now supported in lookbehinds and DFA matching. + +14. Perl does not support \N without a following name in a [] class; PCRE now + also gives an error. + +15. If a forward reference was repeated with an upper limit of around 2000, + it caused the error "internal error: overran compiling workspace". The + maximum number of forward references (including repeats) was limited by the + internal workspace, and dependent on the LINK_SIZE. The code has been + rewritten so that the workspace expands (via pcre_malloc) if necessary, and + the default depends on LINK_SIZE. There is a new upper limit (for safety) + of around 200,000 forward references. While doing this, I also speeded up + the filling in of repeated forward references. + +16. A repeated forward reference in a pattern such as (a)(?2){2}(.) was + incorrectly expecting the subject to contain another "a" after the start. + +17. When (*SKIP:name) is activated without a corresponding (*MARK:name) earlier + in the match, the SKIP should be ignored. This was not happening; instead + the SKIP was being treated as NOMATCH. For patterns such as + /A(*MARK:A)A+(*SKIP:B)Z|AAC/ this meant that the AAC branch was never + tested. + +18. The behaviour of (*MARK), (*PRUNE), and (*THEN) has been reworked and is + now much more compatible with Perl, in particular in cases where the result + is a non-match for a non-anchored pattern. For example, if + /b(*:m)f|a(*:n)w/ is matched against "abc", the non-match returns the name + "m", where previously it did not return a name. A side effect of this + change is that for partial matches, the last encountered mark name is + returned, as for non matches. A number of tests that were previously not + Perl-compatible have been moved into the Perl-compatible test files. The + refactoring has had the pleasing side effect of removing one argument from + the match() function, thus reducing its stack requirements. + +19. If the /S+ option was used in pcretest to study a pattern using JIT, + subsequent uses of /S (without +) incorrectly behaved like /S+. + +21. Retrieve executable code size support for the JIT compiler and fixing + some warnings. + +22. A caseless match of a UTF-8 character whose other case uses fewer bytes did + not work when the shorter character appeared right at the end of the + subject string. + +23. Added some (int) casts to non-JIT modules to reduce warnings on 64-bit + systems. + +24. Added PCRE_INFO_JITSIZE to pass on the value from (21) above, and also + output it when the /M option is used in pcretest. + +25. The CheckMan script was not being included in the distribution. Also, added + an explicit "perl" to run Perl scripts from the PrepareRelease script + because this is reportedly needed in Windows. + +26. If study data was being save in a file and studying had not found a set of + "starts with" bytes for the pattern, the data written to the file (though + never used) was taken from uninitialized memory and so caused valgrind to + complain. + +27. Updated RunTest.bat as provided by Sheri Pierce. + +28. Fixed a possible uninitialized memory bug in pcre_jit_compile.c. + +29. Computation of memory usage for the table of capturing group names was + giving an unnecessarily large value. + + +Version 8.20 21-Oct-2011 +------------------------ + +1. Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had + a POSIX class. After further experiments with Perl, which convinced me that + Perl has bugs and confusions, a closing square bracket is no longer allowed + in a POSIX name. This bug also affected patterns with classes that started + with full stops. + +2. If a pattern such as /(a)b|ac/ is matched against "ac", there is no + captured substring, but while checking the failing first alternative, + substring 1 is temporarily captured. If the output vector supplied to + pcre_exec() was not big enough for this capture, the yield of the function + was still zero ("insufficient space for captured substrings"). This cannot + be totally fixed without adding another stack variable, which seems a lot + of expense for a edge case. However, I have improved the situation in cases + such as /(a)(b)x|abc/ matched against "abc", where the return code + indicates that fewer than the maximum number of slots in the ovector have + been set. + +3. Related to (2) above: when there are more back references in a pattern than + slots in the output vector, pcre_exec() uses temporary memory during + matching, and copies in the captures as far as possible afterwards. It was + using the entire output vector, but this conflicts with the specification + that only 2/3 is used for passing back captured substrings. Now it uses + only the first 2/3, for compatibility. This is, of course, another edge + case. + +4. Zoltan Herczeg's just-in-time compiler support has been integrated into the + main code base, and can be used by building with --enable-jit. When this is + done, pcregrep automatically uses it unless --disable-pcregrep-jit or the + runtime --no-jit option is given. + +5. When the number of matches in a pcre_dfa_exec() run exactly filled the + ovector, the return from the function was zero, implying that there were + other matches that did not fit. The correct "exactly full" value is now + returned. + +6. If a subpattern that was called recursively or as a subroutine contained + (*PRUNE) or any other control that caused it to give a non-standard return, + invalid errors such as "Error -26 (nested recursion at the same subject + position)" or even infinite loops could occur. + +7. If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped + computing the minimum length on reaching *ACCEPT, and so ended up with the + wrong value of 1 rather than 0. Further investigation indicates that + computing a minimum subject length in the presence of *ACCEPT is difficult + (think back references, subroutine calls), and so I have changed the code + so that no minimum is registered for a pattern that contains *ACCEPT. + +8. If (*THEN) was present in the first (true) branch of a conditional group, + it was not handled as intended. [But see 16 below.] + +9. Replaced RunTest.bat and CMakeLists.txt with improved versions provided by + Sheri Pierce. + +10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that + the first byte in a match must be "a". + +11. Change 17 for 8.13 increased the recursion depth for patterns like + /a(?:.)*?a/ drastically. I've improved things by remembering whether a + pattern contains any instances of (*THEN). If it does not, the old + optimizations are restored. It would be nice to do this on a per-group + basis, but at the moment that is not feasible. + +12. In some environments, the output of pcretest -C is CRLF terminated. This + broke RunTest's code that checks for the link size. A single white space + character after the value is now allowed for. + +13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french". + For "fr", it uses the Windows-specific input and output files. + +14. If (*THEN) appeared in a group that was called recursively or as a + subroutine, it did not work as intended. [But see next item.] + +15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex + pattern fragments (but not containing any | characters). If A and B are + matched, but there is a failure in C so that it backtracks to (*THEN), PCRE + was behaving differently to Perl. PCRE backtracked into A, but Perl goes to + D. In other words, Perl considers parentheses that do not contain any | + characters to be part of a surrounding alternative, whereas PCRE was + treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles + differently. PCRE now behaves in the same way as Perl, except in the case + of subroutine/recursion calls such as (?1) which have in any case always + been different (but PCRE had them first :-). + +16. Related to 15 above: Perl does not treat the | in a conditional group as + creating alternatives. Such a group is treated in the same way as an + ordinary group without any | characters when processing (*THEN). PCRE has + been changed to match Perl's behaviour. + +17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the + RunGrepTest script failed. + +18. Change 22 for version 13 caused atomic groups to use more stack. This is + inevitable for groups that contain captures, but it can lead to a lot of + stack use in large patterns. The old behaviour has been restored for atomic + groups that do not contain any capturing parentheses. + +19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not + suppress the check for a minimum subject length at run time. (If it was + given to pcre_exec() or pcre_dfa_exec() it did work.) + +20. Fixed an ASCII-dependent infelicity in pcretest that would have made it + fail to work when decoding hex characters in data strings in EBCDIC + environments. + +21. It appears that in at least one Mac OS environment, the isxdigit() function + is implemented as a macro that evaluates to its argument more than once, + contravening the C 90 Standard (I haven't checked a later standard). There + was an instance in pcretest which caused it to go wrong when processing + \x{...} escapes in subject strings. The has been rewritten to avoid using + things like p++ in the argument of isxdigit(). + + +Version 8.13 16-Aug-2011 +------------------------ + +1. The Unicode data tables have been updated to Unicode 6.0.0. + +2. Two minor typos in pcre_internal.h have been fixed. + +3. Added #include to pcre_scanner_unittest.cc, pcrecpp.cc, and + pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr() + in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2). + +4. There were a number of related bugs in the code for matching backrefences + caselessly in UTF-8 mode when codes for the characters concerned were + different numbers of bytes. For example, U+023A and U+2C65 are an upper + and lower case pair, using 2 and 3 bytes, respectively. The main bugs were: + (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte + code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a + 2-byte code at the end of the subject (it thought there wasn't enough data + left). + +5. Comprehensive information about what went wrong is now returned by + pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long + as the output vector has at least 2 elements. The offset of the start of + the failing character and a reason code are placed in the vector. + +6. When the UTF-8 string check fails for pcre_compile(), the offset that is + now returned is for the first byte of the failing character, instead of the + last byte inspected. This is an incompatible change, but I hope it is small + enough not to be a problem. It makes the returned offset consistent with + pcre_exec() and pcre_dfa_exec(). + +7. pcretest now gives a text phrase as well as the error number when + pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check + failure, the offset and reason code are output. + +8. When \R was used with a maximizing quantifier it failed to skip backwards + over a \r\n pair if the subsequent match failed. Instead, it just skipped + back over a single character (\n). This seems wrong (because it treated the + two characters as a single entity when going forwards), conflicts with the + documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the + behaviour of \R* different to (\R)*, which also seems wrong. The behaviour + has been changed. + +9. Some internal refactoring has changed the processing so that the handling + of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile + time (the PCRE_DOTALL option was changed this way some time ago: version + 7.7 change 16). This has made it possible to abolish the OP_OPT op code, + which was always a bit of a fudge. It also means that there is one less + argument for the match() function, which reduces its stack requirements + slightly. This change also fixes an incompatibility with Perl: the pattern + (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match. + +10. More internal refactoring has drastically reduced the number of recursive + calls to match() for possessively repeated groups such as (abc)++ when + using pcre_exec(). + +11. While implementing 10, a number of bugs in the handling of groups were + discovered and fixed: + + (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind). + (a|)*(?1) gave a compile-time internal error. + ((a|)+)+ did not notice that the outer group could match an empty string. + (^a|^)+ was not marked as anchored. + (.*a|.*)+ was not marked as matching at start or after a newline. + +12. Yet more internal refactoring has removed another argument from the match() + function. Special calls to this function are now indicated by setting a + value in a variable in the "match data" data block. + +13. Be more explicit in pcre_study() instead of relying on "default" for + opcodes that mean there is no starting character; this means that when new + ones are added and accidentally left out of pcre_study(), testing should + pick them up. + +14. The -s option of pcretest has been documented for ages as being an old + synonym of -m (show memory usage). I have changed it to mean "force study + for every regex", that is, assume /S for every regex. This is similar to -i + and -d etc. It's slightly incompatible, but I'm hoping nobody is still + using it. It makes it easier to run collections of tests with and without + study enabled, and thereby test pcre_study() more easily. All the standard + tests are now run with and without -s (but some patterns can be marked as + "never study" - see 20 below). + +15. When (*ACCEPT) was used in a subpattern that was called recursively, the + restoration of the capturing data to the outer values was not happening + correctly. + +16. If a recursively called subpattern ended with (*ACCEPT) and matched an + empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole + pattern had matched an empty string, and so incorrectly returned a no + match. + +17. There was optimizing code for the last branch of non-capturing parentheses, + and also for the obeyed branch of a conditional subexpression, which used + tail recursion to cut down on stack usage. Unfortunately, now that there is + the possibility of (*THEN) occurring in these branches, tail recursion is + no longer possible because the return has to be checked for (*THEN). These + two optimizations have therefore been removed. [But see 8.20/11 above.] + +18. If a pattern containing \R was studied, it was assumed that \R always + matched two bytes, thus causing the minimum subject length to be + incorrectly computed because \R can also match just one byte. + +19. If a pattern containing (*ACCEPT) was studied, the minimum subject length + was incorrectly computed. + +20. If /S is present twice on a test pattern in pcretest input, it now + *disables* studying, thereby overriding the use of -s on the command line + (see 14 above). This is necessary for one or two tests to keep the output + identical in both cases. + +21. When (*ACCEPT) was used in an assertion that matched an empty string and + PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion. + +22. When an atomic group that contained a capturing parenthesis was + successfully matched, but the branch in which it appeared failed, the + capturing was not being forgotten if a higher numbered group was later + captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing + group 1 to "a", when in fact it should be unset. This applied to multi- + branched capturing and non-capturing groups, repeated or not, and also to + positive assertions (capturing in negative assertions does not happen + in PCRE) and also to nested atomic groups. + +23. Add the ++ qualifier feature to pcretest, to show the remainder of the + subject after a captured substring, to make it easier to tell which of a + number of identical substrings has been captured. + +24. The way atomic groups are processed by pcre_exec() has been changed so that + if they are repeated, backtracking one repetition now resets captured + values correctly. For example, if ((?>(a+)b)+aabab) is matched against + "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as + "aaa". Previously, it would have been "a". As part of this code + refactoring, the way recursive calls are handled has also been changed. + +25. If an assertion condition captured any substrings, they were not passed + back unless some other capturing happened later. For example, if + (?(?=(a))a) was matched against "a", no capturing was returned. + +26. When studying a pattern that contained subroutine calls or assertions, + the code for finding the minimum length of a possible match was handling + direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where + group 1 called group 2 while simultaneously a separate group 2 called group + 1). A stack overflow occurred in this case. I have fixed this by limiting + the recursion depth to 10. + +27. Updated RunTest.bat in the distribution to the version supplied by Tom + Fortmann. This supports explicit test numbers on the command line, and has + argument validation and error reporting. + +28. An instance of \X with an unlimited repeat could fail if at any point the + first character it looked at was a mark character. + +29. Some minor code refactoring concerning Unicode properties and scripts + should reduce the stack requirement of match() slightly. + +30. Added the '=' option to pcretest to check the setting of unused capturing + slots at the end of the pattern, which are documented as being -1, but are + not included in the return count. + +31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE + compiled something random. Now it gives a compile-time error (as does + Perl). + +32. A *MARK encountered during the processing of a positive assertion is now + recorded and passed back (compatible with Perl). + +33. If --only-matching or --colour was set on a pcregrep call whose pattern + had alternative anchored branches, the search for a second match in a line + was done as if at the line start. Thus, for example, /^01|^02/ incorrectly + matched the line "0102" twice. The same bug affected patterns that started + with a backwards assertion. For example /\b01|\b02/ also matched "0102" + twice. + +34. Previously, PCRE did not allow quantification of assertions. However, Perl + does, and because of capturing effects, quantifying parenthesized + assertions may at times be useful. Quantifiers are now allowed for + parenthesized assertions. + +35. A minor code tidy in pcre_compile() when checking options for \R usage. + +36. \g was being checked for fancy things in a character class, when it should + just be a literal "g". + +37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the + appearance of a nested POSIX class supersedes an apparent external class. + For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also, + unescaped square brackets may also appear as part of class names. For + example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves + more like Perl. (But see 8.20/1 above.) + +38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this + was because it thought it was \N{name}, which is not supported). + +39. Add minix to OS list not supporting the -S option in pcretest. + +40. PCRE tries to detect cases of infinite recursion at compile time, but it + cannot analyze patterns in sufficient detail to catch mutual recursions + such as ((?1))((?2)). There is now a runtime test that gives an error if a + subgroup is called recursively as a subpattern for a second time at the + same position in the subject string. In previous releases this might have + been caught by the recursion limit, or it might have run out of stack. + +41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can + happen only once. PCRE was, however incorrectly giving a compile time error + "recursive call could loop indefinitely" because it cannot analyze the + pattern in sufficient detail. The compile time test no longer happens when + PCRE is compiling a conditional subpattern, but actual runaway loops are + now caught at runtime (see 40 above). + +42. It seems that Perl allows any characters other than a closing parenthesis + to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE + has been changed to be the same. + +43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so + as not to get warnings when autogen.sh is called. Also changed + AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro). + +44. To help people who use pcregrep to scan files containing exceedingly long + lines, the following changes have been made: + + (a) The default value of the buffer size parameter has been increased from + 8K to 20K. (The actual buffer used is three times this size.) + + (b) The default can be changed by ./configure --with-pcregrep-bufsize when + PCRE is built. + + (c) A --buffer-size=n option has been added to pcregrep, to allow the size + to be set at run time. + + (d) Numerical values in pcregrep options can be followed by K or M, for + example --buffer-size=50K. + + (e) If a line being scanned overflows pcregrep's buffer, an error is now + given and the return code is set to 2. + +45. Add a pointer to the latest mark to the callout data block. + +46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a + partial match of an empty string instead of no match. This was specific to + the use of ".". + +47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a + complete match instead of a partial match. This bug was dependent on both + the PCRE_UTF8 and PCRE_DOTALL options being set. + +48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the + starting byte set, because \b was not being ignored. + + +Version 8.12 15-Jan-2011 +------------------------ + +1. Fixed some typos in the markup of the man pages, and wrote a script that + checks for such things as part of the documentation building process. + +2. On a big-endian 64-bit system, pcregrep did not correctly process the + --match-limit and --recursion-limit options (added for 8.11). In + particular, this made one of the standard tests fail. (The integer value + went into the wrong half of a long int.) + +3. If the --colour option was given to pcregrep with -v (invert match), it + did strange things, either producing crazy output, or crashing. It should, + of course, ignore a request for colour when reporting lines that do not + match. + +4. Another pcregrep bug caused similar problems if --colour was specified with + -M (multiline) and the pattern match finished with a line ending. + +5. In pcregrep, when a pattern that ended with a literal newline sequence was + matched in multiline mode, the following line was shown as part of the + match. This seems wrong, so I have changed it. + +6. Another pcregrep bug in multiline mode, when --colour was specified, caused + the check for further matches in the same line (so they could be coloured) + to overrun the end of the current line. If another match was found, it was + incorrectly shown (and then shown again when found in the next line). + +7. If pcregrep was compiled under Windows, there was a reference to the + function pcregrep_exit() before it was defined. I am assuming this was + the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was + reported by a user. I've moved the definition above the reference. + + +Version 8.11 10-Dec-2010 +------------------------ + +1. (*THEN) was not working properly if there were untried alternatives prior + to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it + backtracked to try for "b" instead of moving to the next alternative branch + at the same level (in this case, to look for "c"). The Perl documentation + is clear that when (*THEN) is backtracked onto, it goes to the "next + alternative in the innermost enclosing group". + +2. (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern + such as (A(*COMMIT)B(*THEN)C|D) any failure after matching A should + result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and + (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides + (*THEN). + +3. If \s appeared in a character class, it removed the VT character from + the class, even if it had been included by some previous item, for example + in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part + of \s, but is part of the POSIX "space" class.) + +4. A partial match never returns an empty string (because you can always + match an empty string at the end of the subject); however the checking for + an empty string was starting at the "start of match" point. This has been + changed to the "earliest inspected character" point, because the returned + data for a partial match starts at this character. This means that, for + example, /(?<=abc)def/ gives a partial match for the subject "abc" + (previously it gave "no match"). + +5. Changes have been made to the way PCRE_PARTIAL_HARD affects the matching + of $, \z, \Z, \b, and \B. If the match point is at the end of the string, + previously a full match would be given. However, setting PCRE_PARTIAL_HARD + has an implication that the given string is incomplete (because a partial + match is preferred over a full match). For this reason, these items now + give a partial match in this situation. [Aside: previously, the one case + /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial + match rather than a full match, which was wrong by the old rules, but is + now correct.] + +6. There was a bug in the handling of #-introduced comments, recognized when + PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set. + If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose + UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when + scanning for the end of the comment. (*Character* 0x85 is an "any" newline, + but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several + places in pcre_compile(). + +7. Related to (6) above, when pcre_compile() was skipping #-introduced + comments when looking ahead for named forward references to subpatterns, + the only newline sequence it recognized was NL. It now handles newlines + according to the set newline convention. + +8. SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the + former, but used strtoul(), whereas pcretest avoided strtoul() but did not + cater for a lack of strerror(). These oversights have been fixed. + +9. Added --match-limit and --recursion-limit to pcregrep. + +10. Added two casts needed to build with Visual Studio when NO_RECURSE is set. + +11. When the -o option was used, pcregrep was setting a return code of 1, even + when matches were found, and --line-buffered was not being honoured. + +12. Added an optional parentheses number to the -o and --only-matching options + of pcregrep. + +13. Imitating Perl's /g action for multiple matches is tricky when the pattern + can match an empty string. The code to do it in pcretest and pcredemo + needed fixing: + + (a) When the newline convention was "crlf", pcretest got it wrong, skipping + only one byte after an empty string match just before CRLF (this case + just got forgotten; "any" and "anycrlf" were OK). + + (b) The pcretest code also had a bug, causing it to loop forever in UTF-8 + mode when an empty string match preceded an ASCII character followed by + a non-ASCII character. (The code for advancing by one character rather + than one byte was nonsense.) + + (c) The pcredemo.c sample program did not have any code at all to handle + the cases when CRLF is a valid newline sequence. + +14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given + as a starting offset was within the subject string. There is now a new + error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is + negative or greater than the length of the string. In order to test this, + pcretest is extended to allow the setting of negative starting offsets. + +15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the + starting offset points to the beginning of a UTF-8 character was + unnecessarily clumsy. I tidied it up. + +16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a + bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD. + +17. Nobody had reported that the --include_dir option, which was added in + release 7.7 should have been called --include-dir (hyphen, not underscore) + for compatibility with GNU grep. I have changed it to --include-dir, but + left --include_dir as an undocumented synonym, and the same for + --exclude-dir, though that is not available in GNU grep, at least as of + release 2.5.4. + +18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8 + characters from a string of bytes) have been redefined so as not to use + loops, in order to improve performance in some environments. At the same + time, I abstracted some of the common code into auxiliary macros to save + repetition (this should not affect the compiled code). + +19. If \c was followed by a multibyte UTF-8 character, bad things happened. A + compile-time error is now given if \c is not followed by an ASCII + character, that is, a byte less than 128. (In EBCDIC mode, the code is + different, and any byte value is allowed.) + +20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_ + START_OPTIMIZE option, which is now allowed at compile time - but just + passed through to pcre_exec() or pcre_dfa_exec(). This makes it available + to pcregrep and other applications that have no direct access to PCRE + options. The new /Y option in pcretest sets this option when calling + pcre_compile(). + +21. Change 18 of release 8.01 broke the use of named subpatterns for recursive + back references. Groups containing recursive back references were forced to + be atomic by that change, but in the case of named groups, the amount of + memory required was incorrectly computed, leading to "Failed: internal + error: code overflow". This has been fixed. + +22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and + pcretest.c, to avoid build problems in some Borland environments. + + +Version 8.10 25-Jun-2010 +------------------------ + +1. Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and + THEN. + +2. (*ACCEPT) was not working when inside an atomic group. + +3. Inside a character class, \B is treated as a literal by default, but + faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option + causes the error). The code is unchanged, but I tidied the documentation. + +4. Inside a character class, PCRE always treated \R and \X as literals, + whereas Perl faults them if its -w option is set. I have changed PCRE so + that it faults them when PCRE_EXTRA is set. + +5. Added support for \N, which always matches any character other than + newline. (It is the same as "." when PCRE_DOTALL is not set.) + +6. When compiling pcregrep with newer versions of gcc which may have + FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite', + declared with attribute warn_unused_result" were given. Just casting the + result to (void) does not stop the warnings; a more elaborate fudge is + needed. I've used a macro to implement this. + +7. Minor change to pcretest.c to avoid a compiler warning. + +8. Added four artifical Unicode properties to help with an option to make + \s etc use properties (see next item). The new properties are: Xan + (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word). + +9. Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes + use Unicode properties. (*UCP) at the start of a pattern can be used to set + this option. Modified pcretest to add /W to test this facility. Added + REG_UCP to make it available via the POSIX interface. + +10. Added --line-buffered to pcregrep. + +11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was + studied, and the match started with a letter with a code point greater than + 127 whose first byte was different to the first byte of the other case of + the letter, the other case of this starting letter was not recognized + (#976). + +12. If a pattern that was studied started with a repeated Unicode property + test, for example, \p{Nd}+, there was the theoretical possibility of + setting up an incorrect bitmap of starting bytes, but fortunately it could + not have actually happened in practice until change 8 above was made (it + added property types that matched character-matching opcodes). + +13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of + possible starting bytes for non-anchored patterns. + +14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes + \R, and also a number of cases that involve Unicode properties, both + explicit and implicit when PCRE_UCP is set. + +15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8 + input, it could crash or give wrong results if characters with values + greater than 0xc0 were present in the subject string. (Detail: it assumed + UTF-8 input when processing these items.) + +16. Added a lot of (int) casts to avoid compiler warnings in systems where + size_t is 64-bit (#991). + +17. Added a check for running out of memory when PCRE is compiled with + --disable-stack-for-recursion (#990). + +18. If the last data line in a file for pcretest does not have a newline on + the end, a newline was missing in the output. + +19. The default pcre_chartables.c file recognizes only ASCII characters (values + less than 128) in its various bitmaps. However, there is a facility for + generating tables according to the current locale when PCRE is compiled. It + turns out that in some environments, 0x85 and 0xa0, which are Unicode space + characters, are recognized by isspace() and therefore were getting set in + these tables, and indeed these tables seem to approximate to ISO 8859. This + caused a problem in UTF-8 mode when pcre_study() was used to create a list + of bytes that can start a match. For \s, it was including 0x85 and 0xa0, + which of course cannot start UTF-8 characters. I have changed the code so + that only real ASCII characters (less than 128) and the correct starting + bytes for UTF-8 encodings are set for characters greater than 127 when in + UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different + altogether.) + +20. Added the /T option to pcretest so as to be able to run tests with non- + standard character tables, thus making it possible to include the tests + used for 19 above in the standard set of tests. + +21. A pattern such as (?&t)(?#()(?(DEFINE)(?a)) which has a forward + reference to a subpattern the other side of a comment that contains an + opening parenthesis caused either an internal compiling error, or a + reference to the wrong subpattern. + + +Version 8.02 19-Mar-2010 +------------------------ + +1. The Unicode data tables have been updated to Unicode 5.2.0. + +2. Added the option --libs-cpp to pcre-config, but only when C++ support is + configured. + +3. Updated the licensing terms in the pcregexp.pas file, as agreed with the + original author of that file, following a query about its status. + +4. On systems that do not have stdint.h (e.g. Solaris), check for and include + inttypes.h instead. This fixes a bug that was introduced by change 8.01/8. + +5. A pattern such as (?&t)*+(?(DEFINE)(?.)) which has a possessive + quantifier applied to a forward-referencing subroutine call, could compile + incorrect code or give the error "internal error: previously-checked + referenced subpattern not found". + +6. Both MS Visual Studio and Symbian OS have problems with initializing + variables to point to external functions. For these systems, therefore, + pcre_malloc etc. are now initialized to local functions that call the + relevant global functions. + +7. There were two entries missing in the vectors called coptable and poptable + in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors. + I've fixed the data, and added a kludgy way of testing at compile time that + the lengths are correct (equal to the number of opcodes). + +8. Following on from 7, I added a similar kludge to check the length of the + eint vector in pcreposix.c. + +9. Error texts for pcre_compile() are held as one long string to avoid too + much relocation at load time. To find a text, the string is searched, + counting zeros. There was no check for running off the end of the string, + which could happen if a new error number was added without updating the + string. + +10. \K gave a compile-time error if it appeared in a lookbehind assersion. + +11. \K was not working if it appeared in an atomic group or in a group that + was called as a "subroutine", or in an assertion. Perl 5.11 documents that + \K is "not well defined" if used in an assertion. PCRE now accepts it if + the assertion is positive, but not if it is negative. + +12. Change 11 fortuitously reduced the size of the stack frame used in the + "match()" function of pcre_exec.c by one pointer. Forthcoming + implementation of support for (*MARK) will need an extra pointer on the + stack; I have reserved it now, so that the stack frame size does not + decrease. + +13. A pattern such as (?P(?P0)|(?P>L2)(?P>L1)) in which the only other + item in branch that calls a recursion is a subroutine call - as in the + second branch in the above example - was incorrectly given the compile- + time error "recursive call could loop indefinitely" because pcre_compile() + was not correctly checking the subroutine for matching a non-empty string. + +14. The checks for overrunning compiling workspace could trigger after an + overrun had occurred. This is a "should never occur" error, but it can be + triggered by pathological patterns such as hundreds of nested parentheses. + The checks now trigger 100 bytes before the end of the workspace. + +15. Fix typo in configure.ac: "srtoq" should be "strtoq". + + +Version 8.01 19-Jan-2010 +------------------------ + +1. If a pattern contained a conditional subpattern with only one branch (in + particular, this includes all (*DEFINE) patterns), a call to pcre_study() + computed the wrong minimum data length (which is of course zero for such + subpatterns). This could cause incorrect "no match" results. + +2. For patterns such as (?i)a(?-i)b|c where an option setting at the start of + the pattern is reset in the first branch, pcre_compile() failed with + "internal error: code overflow at offset...". This happened only when + the reset was to the original external option setting. (An optimization + abstracts leading options settings into an external setting, which was the + cause of this.) + +3. A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one + of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the + assertion pattern did not match (meaning that the assertion was true), it + was incorrectly treated as false if the SKIP had been reached during the + matching. This also applied to assertions used as conditions. + +4. If an item that is not supported by pcre_dfa_exec() was encountered in an + assertion subpattern, including such a pattern used as a condition, + unpredictable results occurred, instead of the error return + PCRE_ERROR_DFA_UITEM. + +5. The C++ GlobalReplace function was not working like Perl for the special + situation when an empty string is matched. It now does the fancy magic + stuff that is necessary. + +6. In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been + removed. (These were left over from very, very early versions of PCRE.) + +7. Some cosmetic changes to the code to make life easier when compiling it + as part of something else: + + (a) Change DEBUG to PCRE_DEBUG. + + (b) In pcre_compile(), rename the member of the "branch_chain" structure + called "current" as "current_branch", to prevent a collision with the + Linux macro when compiled as a kernel module. + + (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to + prevent a collision with the Linux macro when compiled as a kernel + module. + +8. In pcre_compile() there are some checks for integer overflows that used to + cast potentially large values to (double). This has been changed to that + when building, a check for int64_t is made, and if it is found, it is used + instead, thus avoiding the use of floating point arithmetic. (There is no + other use of FP in PCRE.) If int64_t is not found, the fallback is to + double. + +9. Added two casts to avoid signed/unsigned warnings from VS Studio Express + 2005 (difference between two addresses compared to an unsigned value). + +10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a + custom one, because of the following reported problem in Windows: + + - libbz2 uses the Pascal calling convention (WINAPI) for the functions + under Win32. + - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h", + therefore missing the function definition. + - The compiler thus generates a "C" signature for the test function. + - The linker fails to find the "C" function. + - PCRE fails to configure if asked to do so against libbz2. + +11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these + messages were output: + + Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and + rerunning libtoolize, to keep the correct libtool macros in-tree. + Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am. + + I have done both of these things. + +12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec() + most of the time, it *can* run out if it is given a pattern that contains a + runaway infinite recursion. I updated the discussion in the pcrestack man + page. + +13. Now that we have gone to the x.xx style of version numbers, the minor + version may start with zero. Using 08 or 09 is a bad idea because users + might check the value of PCRE_MINOR in their code, and 08 or 09 may be + interpreted as invalid octal numbers. I've updated the previous comment in + configure.ac, and also added a check that gives an error if 08 or 09 are + used. + +14. Change 8.00/11 was not quite complete: code had been accidentally omitted, + causing partial matching to fail when the end of the subject matched \W + in a UTF-8 pattern where \W was quantified with a minimum of 3. + +15. There were some discrepancies between the declarations in pcre_internal.h + of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and + their definitions. The declarations used "const uschar *" and the + definitions used USPTR. Even though USPTR is normally defined as "const + unsigned char *" (and uschar is typedeffed as "unsigned char"), it was + reported that: "This difference in casting confuses some C++ compilers, for + example, SunCC recognizes above declarations as different functions and + generates broken code for hbpcre." I have changed the declarations to use + USPTR. + +16. GNU libtool is named differently on some systems. The autogen.sh script now + tries several variants such as glibtoolize (MacOSX) and libtoolize1x + (FreeBSD). + +17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00 + (strtoXX undefined when compiling pcrecpp.cc). The patch contains this + comment: "Figure out how to create a longlong from a string: strtoll and + equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for + instance, but it only takes 2 args instead of 3!" + +18. A subtle bug concerned with back references has been fixed by a change of + specification, with a corresponding code fix. A pattern such as + ^(xa|=?\1a)+$ which contains a back reference inside the group to which it + refers, was giving matches when it shouldn't. For example, xa=xaaa would + match that pattern. Interestingly, Perl (at least up to 5.11.3) has the + same bug. Such groups have to be quantified to be useful, or contained + inside another quantified group. (If there's no repetition, the reference + can never match.) The problem arises because, having left the group and + moved on to the rest of the pattern, a later failure that backtracks into + the group uses the captured value from the final iteration of the group + rather than the correct earlier one. I have fixed this in PCRE by forcing + any group that contains a reference to itself to be an atomic group; that + is, there cannot be any backtracking into it once it has completed. This is + similar to recursive and subroutine calls. + + +Version 8.00 19-Oct-09 +---------------------- + +1. The table for translating pcre_compile() error codes into POSIX error codes + was out-of-date, and there was no check on the pcre_compile() error code + being within the table. This could lead to an OK return being given in + error. + +2. Changed the call to open a subject file in pcregrep from fopen(pathname, + "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests + in a Windows environment. + +3. The pcregrep --count option prints the count for each file even when it is + zero, as does GNU grep. However, pcregrep was also printing all files when + --files-with-matches was added. Now, when both options are given, it prints + counts only for those files that have at least one match. (GNU grep just + prints the file name in this circumstance, but including the count seems + more useful - otherwise, why use --count?) Also ensured that the + combination -clh just lists non-zero counts, with no names. + +4. The long form of the pcregrep -F option was incorrectly implemented as + --fixed_strings instead of --fixed-strings. This is an incompatible change, + but it seems right to fix it, and I didn't think it was worth preserving + the old behaviour. + +5. The command line items --regex=pattern and --regexp=pattern were not + recognized by pcregrep, which required --regex pattern or --regexp pattern + (with a space rather than an '='). The man page documented the '=' forms, + which are compatible with GNU grep; these now work. + +6. No libpcreposix.pc file was created for pkg-config; there was just + libpcre.pc and libpcrecpp.pc. The omission has been rectified. + +7. Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size + when UCP support is not needed, by modifying the Python script that + generates it from Unicode data files. This should not matter if the module + is correctly used as a library, but I received one complaint about 50K of + unwanted data. My guess is that the person linked everything into his + program rather than using a library. Anyway, it does no harm. + +8. A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger + was a minimum greater than 1 for a wide character in a possessive + repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8 + which had an unlimited repeat of a nested, fixed maximum repeat of a wide + character. Chaos in the form of incorrect output or a compiling loop could + result. + +9. The restrictions on what a pattern can contain when partial matching is + requested for pcre_exec() have been removed. All patterns can now be + partially matched by this function. In addition, if there are at least two + slots in the offset vector, the offset of the earliest inspected character + for the match and the offset of the end of the subject are set in them when + PCRE_ERROR_PARTIAL is returned. + +10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is + synonymous with PCRE_PARTIAL, for backwards compatibility, and + PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match, + and may be more useful for multi-segment matching. + +11. Partial matching with pcre_exec() is now more intuitive. A partial match + used to be given if ever the end of the subject was reached; now it is + given only if matching could not proceed because another character was + needed. This makes a difference in some odd cases such as Z(*FAIL) with the + string "Z", which now yields "no match" instead of "partial match". In the + case of pcre_dfa_exec(), "no match" is given if every matching path for the + final character ended with (*FAIL). + +12. Restarting a match using pcre_dfa_exec() after a partial match did not work + if the pattern had a "must contain" character that was already found in the + earlier partial match, unless partial matching was again requested. For + example, with the pattern /dog.(body)?/, the "must contain" character is + "g". If the first part-match was for the string "dog", restarting with + "sbody" failed. This bug has been fixed. + +13. The string returned by pcre_dfa_exec() after a partial match has been + changed so that it starts at the first inspected character rather than the + first character of the match. This makes a difference only if the pattern + starts with a lookbehind assertion or \b or \B (\K is not supported by + pcre_dfa_exec()). It's an incompatible change, but it makes the two + matching functions compatible, and I think it's the right thing to do. + +14. Added a pcredemo man page, created automatically from the pcredemo.c file, + so that the demonstration program is easily available in environments where + PCRE has not been installed from source. + +15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp, + libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared + library. + +16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user. + It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it + is not the first non-POSIX option to be added. Clearly some people find + these options useful. + +17. If a caller to the POSIX matching function regexec() passes a non-zero + value for nmatch with a NULL value for pmatch, the value of + nmatch is forced to zero. + +18. RunGrepTest did not have a test for the availability of the -u option of + the diff command, as RunTest does. It now checks in the same way as + RunTest, and also checks for the -b option. + +19. If an odd number of negated classes containing just a single character + interposed, within parentheses, between a forward reference to a named + subpattern and the definition of the subpattern, compilation crashed with + an internal error, complaining that it could not find the referenced + subpattern. An example of a crashing pattern is /(?&A)(([^m])(?))/. + [The bug was that it was starting one character too far in when skipping + over the character class, thus treating the ] as data rather than + terminating the class. This meant it could skip too much.] + +20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the + /g option in pcretest when the pattern contains \K, which makes it possible + to have an empty string match not at the start, even when the pattern is + anchored. Updated pcretest and pcredemo to use this option. + +21. If the maximum number of capturing subpatterns in a recursion was greater + than the maximum at the outer level, the higher number was returned, but + with unset values at the outer level. The correct (outer level) value is + now given. + +22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of + PCRE did not set those parentheses (unlike Perl). I have now found a way to + make it do so. The string so far is captured, making this feature + compatible with Perl. + +23. The tests have been re-organized, adding tests 11 and 12, to make it + possible to check the Perl 5.10 features against Perl 5.10. + +24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine + pattern matches a fixed length string. PCRE did not allow this; now it + does. Neither allows recursion. + +25. I finally figured out how to implement a request to provide the minimum + length of subject string that was needed in order to match a given pattern. + (It was back references and recursion that I had previously got hung up + on.) This code has now been added to pcre_study(); it finds a lower bound + to the length of subject needed. It is not necessarily the greatest lower + bound, but using it to avoid searching strings that are too short does give + some useful speed-ups. The value is available to calling programs via + pcre_fullinfo(). + +26. While implementing 25, I discovered to my embarrassment that pcretest had + not been passing the result of pcre_study() to pcre_dfa_exec(), so the + study optimizations had never been tested with that matching function. + Oops. What is worse, even when it was passed study data, there was a bug in + pcre_dfa_exec() that meant it never actually used it. Double oops. There + were also very few tests of studied patterns with pcre_dfa_exec(). + +27. If (?| is used to create subpatterns with duplicate numbers, they are now + allowed to have the same name, even if PCRE_DUPNAMES is not set. However, + on the other side of the coin, they are no longer allowed to have different + names, because these cannot be distinguished in PCRE, and this has caused + confusion. (This is a difference from Perl.) + +28. When duplicate subpattern names are present (necessarily with different + numbers, as required by 27 above), and a test is made by name in a + conditional pattern, either for a subpattern having been matched, or for + recursion in such a pattern, all the associated numbered subpatterns are + tested, and the overall condition is true if the condition is true for any + one of them. This is the way Perl works, and is also more like the way + testing by number works. + + +Version 7.9 11-Apr-09 +--------------------- + +1. When building with support for bzlib/zlib (pcregrep) and/or readline + (pcretest), all targets were linked against these libraries. This included + libpcre, libpcreposix, and libpcrecpp, even though they do not use these + libraries. This caused unwanted dependencies to be created. This problem + has been fixed, and now only pcregrep is linked with bzlib/zlib and only + pcretest is linked with readline. + +2. The "typedef int BOOL" in pcre_internal.h that was included inside the + "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been + moved outside it again, because FALSE and TRUE are already defined in AIX, + but BOOL is not. + +3. The pcre_config() function was treating the PCRE_MATCH_LIMIT and + PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints. + +4. The pcregrep documentation said spaces were inserted as well as colons (or + hyphens) following file names and line numbers when outputting matching + lines. This is not true; no spaces are inserted. I have also clarified the + wording for the --colour (or --color) option. + +5. In pcregrep, when --colour was used with -o, the list of matching strings + was not coloured; this is different to GNU grep, so I have changed it to be + the same. + +6. When --colo(u)r was used in pcregrep, only the first matching substring in + each matching line was coloured. Now it goes on to look for further matches + of any of the test patterns, which is the same behaviour as GNU grep. + +7. A pattern that could match an empty string could cause pcregrep to loop; it + doesn't make sense to accept an empty string match in pcregrep, so I have + locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this + seems to be how GNU grep behaves. [But see later change 40 for release + 8.33.] + +8. The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at + start or after a newline", because the conditional assertion was not being + correctly handled. The rule now is that both the assertion and what follows + in the first alternative must satisfy the test. + +9. If auto-callout was enabled in a pattern with a conditional group whose + condition was an assertion, PCRE could crash during matching, both with + pcre_exec() and pcre_dfa_exec(). + +10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was + used for matching. + +11. Unicode property support in character classes was not working for + characters (bytes) greater than 127 when not in UTF-8 mode. + +12. Added the -M command line option to pcretest. + +14. Added the non-standard REG_NOTEMPTY option to the POSIX interface. + +15. Added the PCRE_NO_START_OPTIMIZE match-time option. + +16. Added comments and documentation about mis-use of no_arg in the C++ + wrapper. + +17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch + from Martin Jerabek that uses macro names for all relevant character and + string constants. + +18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and + SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without + SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of + these, but not everybody uses configure. + +19. A conditional group that had only one branch was not being correctly + recognized as an item that could match an empty string. This meant that an + enclosing group might also not be so recognized, causing infinite looping + (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$ + with the subject "ab", where knowledge that the repeated group can match + nothing is needed in order to break the loop. + +20. If a pattern that was compiled with callouts was matched using pcre_dfa_ + exec(), but without supplying a callout function, matching went wrong. + +21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory + leak if the size of the offset vector was greater than 30. When the vector + is smaller, the saved offsets during recursion go onto a local stack + vector, but for larger vectors malloc() is used. It was failing to free + when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal" + error, in fact). + +22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the + heapframe that is used only when UTF-8 support is enabled. This caused no + problem, but was untidy. + +23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name + CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is + included within another project. + +24. Steven Van Ingelgem's patches to add more options to the CMake support, + slightly modified by me: + + (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including + not building pcregrep. + + (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only + if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep. + +25. Forward references, both numeric and by name, in patterns that made use of + duplicate group numbers, could behave incorrectly or give incorrect errors, + because when scanning forward to find the reference group, PCRE was not + taking into account the duplicate group numbers. A pattern such as + ^X(?3)(a)(?|(b)|(q))(Y) is an example. + +26. Changed a few more instances of "const unsigned char *" to USPTR, making + the feature of a custom pointer more persuasive (as requested by a user). + +27. Wrapped the definitions of fileno and isatty for Windows, which appear in + pcretest.c, inside #ifndefs, because it seems they are sometimes already + pre-defined. + +28. Added support for (*UTF8) at the start of a pattern. + +29. Arrange for flags added by the "release type" setting in CMake to be shown + in the configuration summary. + + +Version 7.8 05-Sep-08 +--------------------- + +1. Replaced UCP searching code with optimized version as implemented for Ad + Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two- + stage table and inline lookup instead of a function, giving speed ups of 2 + to 5 times on some simple patterns that I tested. Permission was given to + distribute the MultiStage2.py script that generates the tables (it's not in + the tarball, but is in the Subversion repository). + +2. Updated the Unicode datatables to Unicode 5.1.0. This adds yet more + scripts. + +3. Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained + a group with a zero qualifier. The result of the study could be incorrect, + or the function might crash, depending on the pattern. + +4. Caseless matching was not working for non-ASCII characters in back + references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}. + It now works when Unicode Property Support is available. + +5. In pcretest, an escape such as \x{de} in the data was always generating + a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in + non-UTF-8 mode. If the value is greater than 255, it gives a warning about + truncation. + +6. Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...). + +7. Added two (int) casts to pcregrep when printing the difference of two + pointers, in case they are 64-bit values. + +8. Added comments about Mac OS X stack usage to the pcrestack man page and to + test 2 if it fails. + +9. Added PCRE_CALL_CONVENTION just before the names of all exported functions, + and a #define of that name to empty if it is not externally set. This is to + allow users of MSVC to set it if necessary. + +10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from + the convenience functions in the pcre_get.c source file. + +11. An option change at the start of a pattern that had top-level alternatives + could cause overwriting and/or a crash. This command provoked a crash in + some environments: + + printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest + + This potential security problem was recorded as CVE-2008-2371. + +12. For a pattern where the match had to start at the beginning or immediately + after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and + pcre_dfa_exec() could read past the end of the passed subject if there was + no match. To help with detecting such bugs (e.g. with valgrind), I modified + pcretest so that it places the subject at the end of its malloc-ed buffer. + +13. The change to pcretest in 12 above threw up a couple more cases when pcre_ + exec() might read past the end of the data buffer in UTF-8 mode. + +14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and + the data contained the byte 0x85 as part of a UTF-8 character within its + first line. This applied both to normal and DFA matching. + +15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example, + /^[^d]*?$/8 failed to match "abc". + +16. Added a missing copyright notice to pcrecpp_internal.h. + +17. Make it more clear in the documentation that values returned from + pcre_exec() in ovector are byte offsets, not character counts. + +18. Tidied a few places to stop certain compilers from issuing warnings. + +19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as + supplied by Stefan Weber. I made a further small update for 7.8 because + there is a change of source arrangements: the pcre_searchfuncs.c module is + replaced by pcre_ucd.c. + + +Version 7.7 07-May-08 +--------------------- + +1. Applied Craig's patch to sort out a long long problem: "If we can't convert + a string to a long long, pretend we don't even have a long long." This is + done by checking for the strtoq, strtoll, and _strtoi64 functions. + +2. Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with + pre-7.6 versions, which defined a global no_arg variable instead of putting + it in the RE class. (See also #8 below.) + +3. Remove a line of dead code, identified by coverity and reported by Nuno + Lopes. + +4. Fixed two related pcregrep bugs involving -r with --include or --exclude: + + (1) The include/exclude patterns were being applied to the whole pathnames + of files, instead of just to the final components. + + (2) If there was more than one level of directory, the subdirectories were + skipped unless they satisfied the include/exclude conditions. This is + inconsistent with GNU grep (and could even be seen as contrary to the + pcregrep specification - which I improved to make it absolutely clear). + The action now is always to scan all levels of directory, and just + apply the include/exclude patterns to regular files. + +5. Added the --include_dir and --exclude_dir patterns to pcregrep, and used + --exclude_dir in the tests to avoid scanning .svn directories. + +6. Applied Craig's patch to the QuoteMeta function so that it escapes the + NUL character as backslash + 0 rather than backslash + NUL, because PCRE + doesn't support NULs in patterns. + +7. Added some missing "const"s to declarations of static tables in + pcre_compile.c and pcre_dfa_exec.c. + +8. Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was + caused by fix #2 above. (Subsequently also a second patch to fix the + first patch. And a third patch - this was a messy problem.) + +9. Applied Craig's patch to remove the use of push_back(). + +10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX + matching function regexec(). + +11. Added support for the Oniguruma syntax \g, \g, \g'name', \g'n', + which, however, unlike Perl's \g{...}, are subroutine calls, not back + references. PCRE supports relative numbers with this syntax (I don't think + Oniguruma does). + +12. Previously, a group with a zero repeat such as (...){0} was completely + omitted from the compiled regex. However, this means that if the group + was called as a subroutine from elsewhere in the pattern, things went wrong + (an internal error was given). Such groups are now left in the compiled + pattern, with a new opcode that causes them to be skipped at execution + time. + +13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes + to the way PCRE behaves: + + (a) A lone ] character is dis-allowed (Perl treats it as data). + + (b) A back reference to an unmatched subpattern matches an empty string + (Perl fails the current match path). + + (c) A data ] in a character class must be notated as \] because if the + first data character in a class is ], it defines an empty class. (In + Perl it is not possible to have an empty class.) The empty class [] + never matches; it forces failure and is equivalent to (*FAIL) or (?!). + The negative empty class [^] matches any one character, independently + of the DOTALL setting. + +14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a + non-existent subpattern following a character class starting with ']' and + containing () gave an internal compiling error instead of "reference to + non-existent subpattern". Fortunately, when the pattern did exist, the + compiled code was correct. (When scanning forwards to check for the + existence of the subpattern, it was treating the data ']' as terminating + the class, so got the count wrong. When actually compiling, the reference + was subsequently set up correctly.) + +15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile; + it was being rejected as not supported by pcre_dfa_exec(), even though + other assertions are supported. I have made pcre_dfa_exec() support + (*FAIL). + +16. The implementation of 13c above involved the invention of a new opcode, + OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s + cannot be changed at match time, I realized I could make a small + improvement to matching performance by compiling OP_ALLANY instead of + OP_ANY for "." when DOTALL was set, and then removing the runtime tests + on the OP_ANY path. + +17. Compiling pcretest on Windows with readline support failed without the + following two fixes: (1) Make the unistd.h include conditional on + HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno. + +18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the + ncurses library to be included for pcretest when ReadLine support is + requested, but also to allow for it to be overridden. This patch came from + Daniel Bergstrm. + +19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined + as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused + any errors with the current Unicode tables. Thanks to Peter Kankowski for + spotting this. + + +Version 7.6 28-Jan-08 +--------------------- + +1. A character class containing a very large number of characters with + codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer + overflow. + +2. Patch to cut out the "long long" test in pcrecpp_unittest when + HAVE_LONG_LONG is not defined. + +3. Applied Christian Ehrlicher's patch to update the CMake build files to + bring them up to date and include new features. This patch includes: + + - Fixed PH's badly added libz and libbz2 support. + - Fixed a problem with static linking. + - Added pcredemo. [But later removed - see 7 below.] + - Fixed dftables problem and added an option. + - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and + HAVE_LONG_LONG. + - Added readline support for pcretest. + - Added an listing of the option settings after cmake has run. + +4. A user submitted a patch to Makefile that makes it easy to create + "pcre.dll" under mingw when using Configure/Make. I added stuff to + Makefile.am that cause it to include this special target, without + affecting anything else. Note that the same mingw target plus all + the other distribution libraries and programs are now supported + when configuring with CMake (see 6 below) instead of with + Configure/Make. + +5. Applied Craig's patch that moves no_arg into the RE class in the C++ code. + This is an attempt to solve the reported problem "pcrecpp::no_arg is not + exported in the Windows port". It has not yet been confirmed that the patch + solves the problem, but it does no harm. + +6. Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and + NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured + with CMake, and also correct the comment about stack recursion. + +7. Remove the automatic building of pcredemo from the ./configure system and + from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example + of a program that users should build themselves after PCRE is installed, so + building it automatically is not really right. What is more, it gave + trouble in some build environments. + +8. Further tidies to CMakeLists.txt from Sheri and Christian. + + +Version 7.5 10-Jan-08 +--------------------- + +1. Applied a patch from Craig: "This patch makes it possible to 'ignore' + values in parens when parsing an RE using the C++ wrapper." + +2. Negative specials like \S did not work in character classes in UTF-8 mode. + Characters greater than 255 were excluded from the class instead of being + included. + +3. The same bug as (2) above applied to negated POSIX classes such as + [:^space:]. + +4. PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it + defined or documented. It seems to have been a typo for PCRE_STATIC, so + I have changed it. + +5. The construct (?&) was not diagnosed as a syntax error (it referenced the + first named subpattern) and a construct such as (?&a) would reference the + first named subpattern whose name started with "a" (in other words, the + length check was missing). Both these problems are fixed. "Subpattern name + expected" is now given for (?&) (a zero-length name), and this patch also + makes it give the same error for \k'' (previously it complained that that + was a reference to a non-existent subpattern). + +6. The erroneous patterns (?+-a) and (?-+a) give different error messages; + this is right because (?- can be followed by option settings as well as by + digits. I have, however, made the messages clearer. + +7. Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns + than the number used in the conditional) now cause a compile-time error. + This is actually not compatible with Perl, which accepts such patterns, but + treats the conditional as always being FALSE (as PCRE used to), but it + seems to me that giving a diagnostic is better. + +8. Change "alphameric" to the more common word "alphanumeric" in comments + and messages. + +9. Fix two occurrences of "backslash" in comments that should have been + "backspace". + +10. Remove two redundant lines of code that can never be obeyed (their function + was moved elsewhere). + +11. The program that makes PCRE's Unicode character property table had a bug + which caused it to generate incorrect table entries for sequences of + characters that have the same character type, but are in different scripts. + It amalgamated them into a single range, with the script of the first of + them. In other words, some characters were in the wrong script. There were + thirteen such cases, affecting characters in the following ranges: + + U+002b0 - U+002c1 + U+0060c - U+0060d + U+0061e - U+00612 + U+0064b - U+0065e + U+0074d - U+0076d + U+01800 - U+01805 + U+01d00 - U+01d77 + U+01d9b - U+01dbf + U+0200b - U+0200f + U+030fc - U+030fe + U+03260 - U+0327f + U+0fb46 - U+0fbb1 + U+10450 - U+1049d + +12. The -o option (show only the matching part of a line) for pcregrep was not + compatible with GNU grep in that, if there was more than one match in a + line, it showed only the first of them. It now behaves in the same way as + GNU grep. + +13. If the -o and -v options were combined for pcregrep, it printed a blank + line for every non-matching line. GNU grep prints nothing, and pcregrep now + does the same. The return code can be used to tell if there were any + non-matching lines. + +14. Added --file-offsets and --line-offsets to pcregrep. + +15. The pattern (?=something)(?R) was not being diagnosed as a potentially + infinitely looping recursion. The bug was that positive lookaheads were not + being skipped when checking for a possible empty match (negative lookaheads + and both kinds of lookbehind were skipped). + +16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the + inclusion of to before rather than after the definition of + INVALID_FILE_ATTRIBUTES (patch from David Byron). + +17. Specifying a possessive quantifier with a specific limit for a Unicode + character property caused pcre_compile() to compile bad code, which led at + runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this + are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that + caused the error; without that there was no problem. + +18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2. + +19. Added --enable-pcretest-libreadline. + +20. In pcrecpp.cc, the variable 'count' was incremented twice in + RE::GlobalReplace(). As a result, the number of replacements returned was + double what it should be. I removed one of the increments, but Craig sent a + later patch that removed the other one (the right fix) and added unit tests + that check the return values (which was not done before). + +21. Several CMake things: + + (1) Arranged that, when cmake is used on Unix, the libraries end up with + the names libpcre and libpcreposix, not just pcre and pcreposix. + + (2) The above change means that pcretest and pcregrep are now correctly + linked with the newly-built libraries, not previously installed ones. + + (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2. + +22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.* + crashed when matching a string such as a\x{2029}b (note that \x{2029} is a + UTF-8 newline character). The key issue is that the pattern starts .*; + this means that the match must be either at the beginning, or after a + newline. The bug was in the code for advancing after a failed match and + checking that the new position followed a newline. It was not taking + account of UTF-8 characters correctly. + +23. PCRE was behaving differently from Perl in the way it recognized POSIX + character classes. PCRE was not treating the sequence [:...:] as a + character class unless the ... were all letters. Perl, however, seems to + allow any characters between [: and :], though of course it rejects as + unknown any "names" that contain non-letters, because all the known class + names consist only of letters. Thus, Perl gives an error for [[:1234:]], + for example, whereas PCRE did not - it did not recognize a POSIX character + class. This seemed a bit dangerous, so the code has been changed to be + closer to Perl. The behaviour is not identical to Perl, because PCRE will + diagnose an unknown class for, for example, [[:l\ower:]] where Perl will + treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where + Perl does, and where it didn't before. + +24. Rewrite so as to remove the single use of %n from pcregrep because in some + Windows environments %n is disabled by default. + + +Version 7.4 21-Sep-07 +--------------------- + +1. Change 7.3/28 was implemented for classes by looking at the bitmap. This + means that a class such as [\s] counted as "explicit reference to CR or + LF". That isn't really right - the whole point of the change was to try to + help when there was an actual mention of one of the two characters. So now + the change happens only if \r or \n (or a literal CR or LF) character is + encountered. + +2. The 32-bit options word was also used for 6 internal flags, but the numbers + of both had grown to the point where there were only 3 bits left. + Fortunately, there was spare space in the data structure, and so I have + moved the internal flags into a new 16-bit field to free up more option + bits. + +3. The appearance of (?J) at the start of a pattern set the DUPNAMES option, + but did not set the internal JCHANGED flag - either of these is enough to + control the way the "get" function works - but the PCRE_INFO_JCHANGED + facility is supposed to tell if (?J) was ever used, so now (?J) at the + start sets both bits. + +4. Added options (at build time, compile time, exec time) to change \R from + matching any Unicode line ending sequence to just matching CR, LF, or CRLF. + +5. doc/pcresyntax.html was missing from the distribution. + +6. Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward + compatibility, even though it is no longer used. + +7. Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and + strtoull to pcrecpp.cc to select the available functions in WIN32 when the + windows.h file is present (where different names are used). [This was + reversed later after testing - see 16 below.] + +8. Changed all #include to #include "config.h". There were also + some further cases that I changed to "pcre.h". + +9. When pcregrep was used with the --colour option, it missed the line ending + sequence off the lines that it output. + +10. It was pointed out to me that arrays of string pointers cause lots of + relocations when a shared library is dynamically loaded. A technique of + using a single long string with a table of offsets can drastically reduce + these. I have refactored PCRE in four places to do this. The result is + dramatic: + + Originally: 290 + After changing UCP table: 187 + After changing error message table: 43 + After changing table of "verbs" 36 + After changing table of Posix names 22 + + Thanks to the folks working on Gregex for glib for this insight. + +11. --disable-stack-for-recursion caused compiling to fail unless -enable- + unicode-properties was also set. + +12. Updated the tests so that they work when \R is defaulted to ANYCRLF. + +13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously + checked only for CRLF. + +14. Added casts to pcretest.c to avoid compiler warnings. + +15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings. + +16. Added Craig's patch to remove the WINDOWS_H tests, that were not working, + and instead check for _strtoi64 explicitly, and avoid the use of snprintf() + entirely. This removes changes made in 7 above. + +17. The CMake files have been updated, and there is now more information about + building with CMake in the NON-UNIX-USE document. + + +Version 7.3 28-Aug-07 +--------------------- + + 1. In the rejigging of the build system that eventually resulted in 7.1, the + line "#include " was included in pcre_internal.h. The use of angle + brackets there is not right, since it causes compilers to look for an + installed pcre.h, not the version that is in the source that is being + compiled (which of course may be different). I have changed it back to: + + #include "pcre.h" + + I have a vague recollection that the change was concerned with compiling in + different directories, but in the new build system, that is taken care of + by the VPATH setting the Makefile. + + 2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed + when the subject happened to end in the byte 0x85 (e.g. if the last + character was \x{1ec5}). *Character* 0x85 is one of the "any" newline + characters but of course it shouldn't be taken as a newline when it is part + of another character. The bug was that, for an unlimited repeat of . in + not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by + characters when looking for a newline. + + 3. A small performance improvement in the DOTALL UTF-8 mode .* case. + + 4. Debugging: adjusted the names of opcodes for different kinds of parentheses + in debug output. + + 5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for + long printing in the pcrecpp unittest when running under MinGW. + + 6. ESC_K was left out of the EBCDIC table. + + 7. Change 7.0/38 introduced a new limit on the number of nested non-capturing + parentheses; I made it 1000, which seemed large enough. Unfortunately, the + limit also applies to "virtual nesting" when a pattern is recursive, and in + this case 1000 isn't so big. I have been able to remove this limit at the + expense of backing off one optimization in certain circumstances. Normally, + when pcre_exec() would call its internal match() function recursively and + immediately return the result unconditionally, it uses a "tail recursion" + feature to save stack. However, when a subpattern that can match an empty + string has an unlimited repetition quantifier, it no longer makes this + optimization. That gives it a stack frame in which to save the data for + checking that an empty string has been matched. Previously this was taken + from the 1000-entry workspace that had been reserved. So now there is no + explicit limit, but more stack is used. + + 8. Applied Daniel's patches to solve problems with the import/export magic + syntax that is required for Windows, and which was going wrong for the + pcreposix and pcrecpp parts of the library. These were overlooked when this + problem was solved for the main library. + + 9. There were some crude static tests to avoid integer overflow when computing + the size of patterns that contain repeated groups with explicit upper + limits. As the maximum quantifier is 65535, the maximum group length was + set at 30,000 so that the product of these two numbers did not overflow a + 32-bit integer. However, it turns out that people want to use groups that + are longer than 30,000 bytes (though not repeat them that many times). + Change 7.0/17 (the refactoring of the way the pattern size is computed) has + made it possible to implement the integer overflow checks in a much more + dynamic way, which I have now done. The artificial limitation on group + length has been removed - we now have only the limit on the total length of + the compiled pattern, which depends on the LINK_SIZE setting. + +10. Fixed a bug in the documentation for get/copy named substring when + duplicate names are permitted. If none of the named substrings are set, the + functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an + empty string. + +11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E + instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error, + because the ] is interpreted as the first data character and the + terminating ] is not found. PCRE has been made compatible with Perl in this + regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could + cause memory overwriting. + +10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty + string has been matched (to stop an infinite loop). It was not recognizing + a conditional subpattern that could match an empty string if that + subpattern was within another subpattern. For example, it looped when + trying to match (((?(1)X|))*) but it was OK with ((?(1)X|)*) where the + condition was not nested. This bug has been fixed. + +12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack + past the start of the subject in the presence of bytes with the top bit + set, for example "\x8aBCD". + +13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE), + (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT). + +14. Optimized (?!) to (*FAIL). + +15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629. + This restricts code points to be within the range 0 to 0x10FFFF, excluding + the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the + full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still + does: it's just the validity check that is more restrictive. + +16. Inserted checks for integer overflows during escape sequence (backslash) + processing, and also fixed erroneous offset values for syntax errors during + backslash processing. + +17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above) + for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80". + +18. An unterminated class in a pattern like (?1)\c[ with a "forward reference" + caused an overrun. + +19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with + something other than just ASCII characters) inside a group that had an + unlimited repeat caused a loop at compile time (while checking to see + whether the group could match an empty string). + +20. Debugging a pattern containing \p or \P could cause a crash. For example, + [\P{Any}] did so. (Error in the code for printing property names.) + +21. An orphan \E inside a character class could cause a crash. + +22. A repeated capturing bracket such as (A)? could cause a wild memory + reference during compilation. + +23. There are several functions in pcre_compile() that scan along a compiled + expression for various reasons (e.g. to see if it's fixed length for look + behind). There were bugs in these functions when a repeated \p or \P was + present in the pattern. These operators have additional parameters compared + with \d, etc, and these were not being taken into account when moving along + the compiled data. Specifically: + + (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed + length. + + (b) An item such as \pL+ within a repeated group could cause crashes or + loops. + + (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect + "reference to non-existent subpattern" error. + + (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time. + +24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte + characters were involved (for example /\S{2}/8g with "A\x{a3}BC"). + +25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop. + +26. Patterns such as [\P{Yi}A] which include \p or \P and just one other + character were causing crashes (broken optimization). + +27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing + \p or \P) caused a compile-time loop. + +28. More problems have arisen in unanchored patterns when CRLF is a valid line + break. For example, the unstudied pattern [\r\n]A does not match the string + "\r\nA" because change 7.0/46 below moves the current point on by two + characters after failing to match at the start. However, the pattern \nA + *does* match, because it doesn't start till \n, and if [\r\n]A is studied, + the same is true. There doesn't seem any very clean way out of this, but + what I have chosen to do makes the common cases work: PCRE now takes note + of whether there can be an explicit match for \r or \n anywhere in the + pattern, and if so, 7.0/46 no longer applies. As part of this change, + there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled + pattern has explicit CR or LF references. + +29. Added (*CR) etc for changing newline setting at start of pattern. + + +Version 7.2 19-Jun-07 +--------------------- + + 1. If the fr_FR locale cannot be found for test 3, try the "french" locale, + which is apparently normally available under Windows. + + 2. Re-jig the pcregrep tests with different newline settings in an attempt + to make them independent of the local environment's newline setting. + + 3. Add code to configure.ac to remove -g from the CFLAGS default settings. + + 4. Some of the "internals" tests were previously cut out when the link size + was not 2, because the output contained actual offsets. The recent new + "Z" feature of pcretest means that these can be cut out, making the tests + usable with all link sizes. + + 5. Implemented Stan Switzer's goto replacement for longjmp() when not using + stack recursion. This gives a massive performance boost under BSD, but just + a small improvement under Linux. However, it saves one field in the frame + in all cases. + + 6. Added more features from the forthcoming Perl 5.10: + + (a) (?-n) (where n is a string of digits) is a relative subroutine or + recursion call. It refers to the nth most recently opened parentheses. + + (b) (?+n) is also a relative subroutine call; it refers to the nth next + to be opened parentheses. + + (c) Conditions that refer to capturing parentheses can be specified + relatively, for example, (?(-2)... or (?(+3)... + + (d) \K resets the start of the current match so that everything before + is not part of it. + + (e) \k{name} is synonymous with \k and \k'name' (.NET compatible). + + (f) \g{name} is another synonym - part of Perl 5.10's unification of + reference syntax. + + (g) (?| introduces a group in which the numbering of parentheses in each + alternative starts with the same number. + + (h) \h, \H, \v, and \V match horizontal and vertical whitespace. + + 7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and + PCRE_INFO_JCHANGED. + + 8. A pattern such as (.*(.)?)* caused pcre_exec() to fail by either not + terminating or by crashing. Diagnosed by Viktor Griph; it was in the code + for detecting groups that can match an empty string. + + 9. A pattern with a very large number of alternatives (more than several + hundred) was running out of internal workspace during the pre-compile + phase, where pcre_compile() figures out how much memory will be needed. A + bit of new cunning has reduced the workspace needed for groups with + alternatives. The 1000-alternative test pattern now uses 12 bytes of + workspace instead of running out of the 4096 that are available. + +10. Inserted some missing (unsigned int) casts to get rid of compiler warnings. + +11. Applied patch from Google to remove an optimization that didn't quite work. + The report of the bug said: + + pcrecpp::RE("a*").FullMatch("aaa") matches, while + pcrecpp::RE("a*?").FullMatch("aaa") does not, and + pcrecpp::RE("a*?\\z").FullMatch("aaa") does again. + +12. If \p or \P was used in non-UTF-8 mode on a character greater than 127 + it matched the wrong number of bytes. + + +Version 7.1 24-Apr-07 +--------------------- + + 1. Applied Bob Rossi and Daniel G's patches to convert the build system to one + that is more "standard", making use of automake and other Autotools. There + is some re-arrangement of the files and adjustment of comments consequent + on this. + + 2. Part of the patch fixed a problem with the pcregrep tests. The test of -r + for recursive directory scanning broke on some systems because the files + are not scanned in any specific order and on different systems the order + was different. A call to "sort" has been inserted into RunGrepTest for the + approprate test as a short-term fix. In the longer term there may be an + alternative. + + 3. I had an email from Eric Raymond about problems translating some of PCRE's + man pages to HTML (despite the fact that I distribute HTML pages, some + people do their own conversions for various reasons). The problems + concerned the use of low-level troff macros .br and .in. I have therefore + removed all such uses from the man pages (some were redundant, some could + be replaced by .nf/.fi pairs). The 132html script that I use to generate + HTML has been updated to handle .nf/.fi and to complain if it encounters + .br or .in. + + 4. Updated comments in configure.ac that get placed in config.h.in and also + arranged for config.h to be included in the distribution, with the name + config.h.generic, for the benefit of those who have to compile without + Autotools (compare pcre.h, which is now distributed as pcre.h.generic). + + 5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan + Weber: (1) pcre_internal.h was missing some function renames; (2) updated + makevp.bat for the current PCRE, using the additional files + makevp_c.txt, makevp_l.txt, and pcregexp.pas. + + 6. A Windows user reported a minor discrepancy with test 2, which turned out + to be caused by a trailing space on an input line that had got lost in his + copy. The trailing space was an accident, so I've just removed it. + + 7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told + that is needed. + + 8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c) + as "const" (a) because they are and (b) because it helps the PHP + maintainers who have recently made a script to detect big data structures + in the php code that should be moved to the .rodata section. I remembered + to update Builducptable as well, so it won't revert if ucptable.h is ever + re-created. + + 9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c, + pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in + order to be able to cut out the UTF-8 tables in the latter when UTF-8 + support is not required. This saves 1.5-2K of code, which is important in + some applications. + + Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c + so as not to refer to the tables, even though these functions will never be + called when UTF-8 support is disabled. Otherwise there are problems with a + shared library. + +10. Fixed two bugs in the emulated memmove() function in pcre_internal.h: + + (a) It was defining its arguments as char * instead of void *. + + (b) It was assuming that all moves were upwards in memory; this was true + a long time ago when I wrote it, but is no longer the case. + + The emulated memove() is provided for those environments that have neither + memmove() nor bcopy(). I didn't think anyone used it these days, but that + is clearly not the case, as these two bugs were recently reported. + +11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt, + and Detrail to create the HTML documentation, the .txt form of the man + pages, and it removes trailing spaces from listed files. It also creates + pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter + case, it wraps all the #defines with #ifndefs. This script should be run + before "make dist". + +12. Fixed two fairly obscure bugs concerned with quantified caseless matching + with Unicode property support. + + (a) For a maximizing quantifier, if the two different cases of the + character were of different lengths in their UTF-8 codings (there are + some cases like this - I found 11), and the matching function had to + back up over a mixture of the two cases, it incorrectly assumed they + were both the same length. + + (b) When PCRE was configured to use the heap rather than the stack for + recursion during matching, it was not correctly preserving the data for + the other case of a UTF-8 character when checking ahead for a match + while processing a minimizing repeat. If the check also involved + matching a wide character, but failed, corruption could cause an + erroneous result when trying to check for a repeat of the original + character. + +13. Some tidying changes to the testing mechanism: + + (a) The RunTest script now detects the internal link size and whether there + is UTF-8 and UCP support by running ./pcretest -C instead of relying on + values substituted by "configure". (The RunGrepTest script already did + this for UTF-8.) The configure.ac script no longer substitutes the + relevant variables. + + (b) The debugging options /B and /D in pcretest show the compiled bytecode + with length and offset values. This means that the output is different + for different internal link sizes. Test 2 is skipped for link sizes + other than 2 because of this, bypassing the problem. Unfortunately, + there was also a test in test 3 (the locale tests) that used /B and + failed for link sizes other than 2. Rather than cut the whole test out, + I have added a new /Z option to pcretest that replaces the length and + offset values with spaces. This is now used to make test 3 independent + of link size. (Test 2 will be tidied up later.) + +14. If erroroffset was passed as NULL to pcre_compile, it provoked a + segmentation fault instead of returning the appropriate error message. + +15. In multiline mode when the newline sequence was set to "any", the pattern + ^$ would give a match between the \r and \n of a subject such as "A\r\nB". + This doesn't seem right; it now treats the CRLF combination as the line + ending, and so does not match in that case. It's only a pattern such as ^$ + that would hit this one: something like ^ABC$ would have failed after \r + and then tried again after \r\n. + +16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub" + in an attempt to make files that differ only in their line terminators + compare equal. This works on Linux. + +17. Under certain error circumstances pcregrep might try to free random memory + as it exited. This is now fixed, thanks to valgrind. + +19. In pcretest, if the pattern /(?m)^$/g was matched against the string + "abc\r\n\r\n", it found an unwanted second match after the second \r. This + was because its rules for how to advance for /g after matching an empty + string at the end of a line did not allow for this case. They now check for + it specially. + +20. pcretest is supposed to handle patterns and data of any length, by + extending its buffers when necessary. It was getting this wrong when the + buffer for a data line had to be extended. + +21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or + CRLF as a newline sequence. + +22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut + out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but + I have nevertheless tidied it up. + +23. Added some casts to kill warnings from HP-UX ia64 compiler. + +24. Added a man page for pcre-config. + + +Version 7.0 19-Dec-06 +--------------------- + + 1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by + moving to gcc 4.1.1. + + 2. The -S option for pcretest uses setrlimit(); I had omitted to #include + sys/time.h, which is documented as needed for this function. It doesn't + seem to matter on Linux, but it showed up on some releases of OS X. + + 3. It seems that there are systems where bytes whose values are greater than + 127 match isprint() in the "C" locale. The "C" locale should be the + default when a C program starts up. In most systems, only ASCII printing + characters match isprint(). This difference caused the output from pcretest + to vary, making some of the tests fail. I have changed pcretest so that: + + (a) When it is outputting text in the compiled version of a pattern, bytes + other than 32-126 are always shown as hex escapes. + + (b) When it is outputting text that is a matched part of a subject string, + it does the same, unless a different locale has been set for the match + (using the /L modifier). In this case, it uses isprint() to decide. + + 4. Fixed a major bug that caused incorrect computation of the amount of memory + required for a compiled pattern when options that changed within the + pattern affected the logic of the preliminary scan that determines the + length. The relevant options are -x, and -i in UTF-8 mode. The result was + that the computed length was too small. The symptoms of this bug were + either the PCRE error "internal error: code overflow" from pcre_compile(), + or a glibc crash with a message such as "pcretest: free(): invalid next + size (fast)". Examples of patterns that provoked this bug (shown in + pcretest format) are: + + /(?-x: )/x + /(?x)(?-x: \s*#\s*)/ + /((?i)[\x{c0}])/8 + /(?i:[\x{c0}])/8 + + HOWEVER: Change 17 below makes this fix obsolete as the memory computation + is now done differently. + + 5. Applied patches from Google to: (a) add a QuoteMeta function to the C++ + wrapper classes; (b) implement a new function in the C++ scanner that is + more efficient than the old way of doing things because it avoids levels of + recursion in the regex matching; (c) add a paragraph to the documentation + for the FullMatch() function. + + 6. The escape sequence \n was being treated as whatever was defined as + "newline". Not only was this contrary to the documentation, which states + that \n is character 10 (hex 0A), but it also went horribly wrong when + "newline" was defined as CRLF. This has been fixed. + + 7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c) + was being set to -1 for the "end of line" case (supposedly a value that no + character can have). Though this value is never used (the check for end of + line is "zero bytes in current character"), it caused compiler complaints. + I've changed it to 0xffffffff. + + 8. In pcre_version.c, the version string was being built by a sequence of + C macros that, in the event of PCRE_PRERELEASE being defined as an empty + string (as it is for production releases) called a macro with an empty + argument. The C standard says the result of this is undefined. The gcc + compiler treats it as an empty string (which was what was wanted) but it is + reported that Visual C gives an error. The source has been hacked around to + avoid this problem. + + 9. On the advice of a Windows user, included and in Windows + builds of pcretest, and changed the call to _setmode() to use _O_BINARY + instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all + of them did). + +10. Originally, pcretest opened its input and output without "b"; then I was + told that "b" was needed in some environments, so it was added for release + 5.0 to both the input and output. (It makes no difference on Unix-like + systems.) Later I was told that it is wrong for the input on Windows. I've + now abstracted the modes into two macros, to make it easier to fiddle with + them, and removed "b" from the input mode under Windows. + +11. Added pkgconfig support for the C++ wrapper library, libpcrecpp. + +12. Added -help and --help to pcretest as an official way of being reminded + of the options. + +13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in + and pcrecpp.cc because they annoy compilers at high warning levels. + +14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop. + +15. Fixed an occurrence of == in configure.ac that should have been = (shell + scripts are not C programs :-) and which was not noticed because it works + on Linux. + +16. pcretest is supposed to handle any length of pattern and data line (as one + line or as a continued sequence of lines) by extending its input buffer if + necessary. This feature was broken for very long pattern lines, leading to + a string of junk being passed to pcre_compile() if the pattern was longer + than about 50K. + +17. I have done a major re-factoring of the way pcre_compile() computes the + amount of memory needed for a compiled pattern. Previously, there was code + that made a preliminary scan of the pattern in order to do this. That was + OK when PCRE was new, but as the facilities have expanded, it has become + harder and harder to keep it in step with the real compile phase, and there + have been a number of bugs (see for example, 4 above). I have now found a + cunning way of running the real compile function in a "fake" mode that + enables it to compute how much memory it would need, while actually only + ever using a few hundred bytes of working memory and without too many + tests of the mode. This should make future maintenance and development + easier. A side effect of this work is that the limit of 200 on the nesting + depth of parentheses has been removed (though this was never a serious + limitation, I suspect). However, there is a downside: pcre_compile() now + runs more slowly than before (30% or more, depending on the pattern). I + hope this isn't a big issue. There is no effect on runtime performance. + +18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a + newline (only possible for the last line of a file) and it was a + pattern that set a locale (followed by /Lsomething), pcretest crashed. + +19. Added additional timing features to pcretest. (1) The -tm option now times + matching only, not compiling. (2) Both -t and -tm can be followed, as a + separate command line item, by a number that specifies the number of + repeats to use when timing. The default is 50000; this gives better + precision, but takes uncomfortably long for very large patterns. + +20. Extended pcre_study() to be more clever in cases where a branch of a + subpattern has no definite first character. For example, (a*|b*)[cd] would + previously give no result from pcre_study(). Now it recognizes that the + first character must be a, b, c, or d. + +21. There was an incorrect error "recursive call could loop indefinitely" if + a subpattern (or the entire pattern) that was being tested for matching an + empty string contained only one non-empty item after a nested subpattern. + For example, the pattern (?>\x{100}*)\d(?R) provoked this error + incorrectly, because the \d was being skipped in the check. + +22. The pcretest program now has a new pattern option /B and a command line + option -b, which is equivalent to adding /B to every pattern. This causes + it to show the compiled bytecode, without the additional information that + -d shows. The effect of -d is now the same as -b with -i (and similarly, /D + is the same as /B/I). + +23. A new optimization is now able automatically to treat some sequences such + as a*b as a*+b. More specifically, if something simple (such as a character + or a simple class like \d) has an unlimited quantifier, and is followed by + something that cannot possibly match the quantified thing, the quantifier + is automatically "possessified". + +24. A recursive reference to a subpattern whose number was greater than 39 + went wrong under certain circumstances in UTF-8 mode. This bug could also + have affected the operation of pcre_study(). + +25. Realized that a little bit of performance could be had by replacing + (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters. + +26. Timing data from pcretest is now shown to 4 decimal places instead of 3. + +27. Possessive quantifiers such as a++ were previously implemented by turning + them into atomic groups such as ($>a+). Now they have their own opcodes, + which improves performance. This includes the automatically created ones + from 23 above. + +28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a + lookahead was broken if it was not anchored. PCRE was mistakenly expecting + the first matched character to be a colon. This applied both to named and + numbered groups. + +29. The ucpinternal.h header file was missing its idempotency #ifdef. + +30. I was sent a "project" file called libpcre.a.dev which I understand makes + building PCRE on Windows easier, so I have included it in the distribution. + +31. There is now a check in pcretest against a ridiculously large number being + returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G + loop, the loop is abandoned. + +32. Forward references to subpatterns in conditions such as (?(2)...) where + subpattern 2 is defined later cause pcre_compile() to search forwards in + the pattern for the relevant set of parentheses. This search went wrong + when there were unescaped parentheses in a character class, parentheses + escaped with \Q...\E, or parentheses in a #-comment in /x mode. + +33. "Subroutine" calls and backreferences were previously restricted to + referencing subpatterns earlier in the regex. This restriction has now + been removed. + +34. Added a number of extra features that are going to be in Perl 5.10. On the + whole, these are just syntactic alternatives for features that PCRE had + previously implemented using the Python syntax or my own invention. The + other formats are all retained for compatibility. + + (a) Named groups can now be defined as (?...) or (?'name'...) as well + as (?P...). The new forms, as well as being in Perl 5.10, are + also .NET compatible. + + (b) A recursion or subroutine call to a named group can now be defined as + (?&name) as well as (?P>name). + + (c) A backreference to a named group can now be defined as \k or + \k'name' as well as (?P=name). The new forms, as well as being in Perl + 5.10, are also .NET compatible. + + (d) A conditional reference to a named group can now use the syntax + (?() or (?('name') as well as (?(name). + + (e) A "conditional group" of the form (?(DEFINE)...) can be used to define + groups (named and numbered) that are never evaluated inline, but can be + called as "subroutines" from elsewhere. In effect, the DEFINE condition + is always false. There may be only one alternative in such a group. + + (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well + as the simple (?(R). The condition is true only if the most recent + recursion is that of the given number or name. It does not search out + through the entire recursion stack. + + (g) The escape \gN or \g{N} has been added, where N is a positive or + negative number, specifying an absolute or relative reference. + +35. Tidied to get rid of some further signed/unsigned compiler warnings and + some "unreachable code" warnings. + +36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other + things, this adds five new scripts. + +37. Perl ignores orphaned \E escapes completely. PCRE now does the same. + There were also incompatibilities regarding the handling of \Q..\E inside + character classes, for example with patterns like [\Qa\E-\Qz\E] where the + hyphen was adjacent to \Q or \E. I hope I've cleared all this up now. + +38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group + matches an empty string, and forcibly breaks the loop. There were bugs in + this code in non-simple cases. For a pattern such as ^(a()*)* matched + against aaaa the result was just "a" rather than "aaaa", for example. Two + separate and independent bugs (that affected different cases) have been + fixed. + +39. Refactored the code to abolish the use of different opcodes for small + capturing bracket numbers. This is a tidy that I avoided doing when I + removed the limit on the number of capturing brackets for 3.5 back in 2001. + The new approach is not only tidier, it makes it possible to reduce the + memory needed to fix the previous bug (38). + +40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline + sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when + processing dot, circumflex, or dollar metacharacters, or #-comments in /x + mode. + +41. Add \R to match any Unicode newline sequence, as suggested in the Unicode + report. + +42. Applied patch, originally from Ari Pollak, modified by Google, to allow + copy construction and assignment in the C++ wrapper. + +43. Updated pcregrep to support "--newline=any". In the process, I fixed a + couple of bugs that could have given wrong results in the "--newline=crlf" + case. + +44. Added a number of casts and did some reorganization of signed/unsigned int + variables following suggestions from Dair Grant. Also renamed the variable + "this" as "item" because it is a C++ keyword. + +45. Arranged for dftables to add + + #include "pcre_internal.h" + + to pcre_chartables.c because without it, gcc 4.x may remove the array + definition from the final binary if PCRE is built into a static library and + dead code stripping is activated. + +46. For an unanchored pattern, if a match attempt fails at the start of a + newline sequence, and the newline setting is CRLF or ANY, and the next two + characters are CRLF, advance by two characters instead of one. + + +Version 6.7 04-Jul-06 +--------------------- + + 1. In order to handle tests when input lines are enormously long, pcretest has + been re-factored so that it automatically extends its buffers when + necessary. The code is crude, but this _is_ just a test program. The + default size has been increased from 32K to 50K. + + 2. The code in pcre_study() was using the value of the re argument before + testing it for NULL. (Of course, in any sensible call of the function, it + won't be NULL.) + + 3. The memmove() emulation function in pcre_internal.h, which is used on + systems that lack both memmove() and bcopy() - that is, hardly ever - + was missing a "static" storage class specifier. + + 4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns + containing an extended class (one that cannot be represented by a bitmap + because it contains high-valued characters or Unicode property items, e.g. + [\pZ]). Almost always one would set UTF-8 mode when processing such a + pattern, but PCRE should not loop if you do not (it no longer does). + [Detail: two cases were found: (a) a repeated subpattern containing an + extended class; (b) a recursive reference to a subpattern that followed a + previous extended class. It wasn't skipping over the extended class + correctly when UTF-8 mode was not set.] + + 5. A negated single-character class was not being recognized as fixed-length + in lookbehind assertions such as (?<=[^f]), leading to an incorrect + compile error "lookbehind assertion is not fixed length". + + 6. The RunPerlTest auxiliary script was showing an unexpected difference + between PCRE and Perl for UTF-8 tests. It turns out that it is hard to + write a Perl script that can interpret lines of an input file either as + byte characters or as UTF-8, which is what "perltest" was being required to + do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you + can't do is switch easily at run time between having the "use utf8;" pragma + or not. In the end, I fudged it by using the RunPerlTest script to insert + "use utf8;" explicitly for the UTF-8 tests. + + 7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at + the end of the subject string, contrary to the documentation and to what + Perl does. This was true of both matching functions. Now it matches only at + the start of the subject and immediately after *internal* newlines. + + 8. A call of pcre_fullinfo() from pcretest to get the option bits was passing + a pointer to an int instead of a pointer to an unsigned long int. This + caused problems on 64-bit systems. + + 9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another + instance of the 'standard' template library not being so standard". + +10. There was no check on the number of named subpatterns nor the maximum + length of a subpattern name. The product of these values is used to compute + the size of the memory block for a compiled pattern. By supplying a very + long subpattern name and a large number of named subpatterns, the size + computation could be caused to overflow. This is now prevented by limiting + the length of names to 32 characters, and the number of named subpatterns + to 10,000. + +11. Subpatterns that are repeated with specific counts have to be replicated in + the compiled pattern. The size of memory for this was computed from the + length of the subpattern and the repeat count. The latter is limited to + 65535, but there was no limit on the former, meaning that integer overflow + could in principle occur. The compiled length of a repeated subpattern is + now limited to 30,000 bytes in order to prevent this. + +12. Added the optional facility to have named substrings with the same name. + +13. Added the ability to use a named substring as a condition, using the + Python syntax: (?(name)yes|no). This overloads (?(R)... and names that + are numbers (not recommended). Forward references are permitted. + +14. Added forward references in named backreferences (if you see what I mean). + +15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the + pattern could run off the end of the subject. For example, the pattern + "(?s)(.{1,5})"8 did this with the subject "ab". + +16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if + PCRE_CASELESS was set when matching characters that were quantified with ? + or *. + +17. A character class other than a single negated character that had a minimum + but no maximum quantifier - for example [ab]{6,} - was not handled + correctly by pce_dfa_exec(). It would match only one character. + +18. A valid (though odd) pattern that looked like a POSIX character + class but used an invalid character after [ (for example [[,abc,]]) caused + pcre_compile() to give the error "Failed: internal error: code overflow" or + in some cases to crash with a glibc free() error. This could even happen if + the pattern terminated after [[ but there just happened to be a sequence of + letters, a binary zero, and a closing ] in the memory that followed. + +19. Perl's treatment of octal escapes in the range \400 to \777 has changed + over the years. Originally (before any Unicode support), just the bottom 8 + bits were taken. Thus, for example, \500 really meant \100. Nowadays the + output from "man perlunicode" includes this: + + The regular expression compiler produces polymorphic opcodes. That + is, the pattern adapts to the data and automatically switches to + the Unicode character scheme when presented with Unicode data--or + instead uses a traditional byte scheme when presented with byte + data. + + Sadly, a wide octal escape does not cause a switch, and in a string with + no other multibyte characters, these octal escapes are treated as before. + Thus, in Perl, the pattern /\500/ actually matches \100 but the pattern + /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a + Unicode string. + + I have not perpetrated such confusion in PCRE. Up till now, it took just + the bottom 8 bits, as in old Perl. I have now made octal escapes with + values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they + translate to the appropriate multibyte character. + +29. Applied some refactoring to reduce the number of warnings from Microsoft + and Borland compilers. This has included removing the fudge introduced + seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused + a warning about an unused variable. + +21. PCRE has not included VT (character 0x0b) in the set of whitespace + characters since release 4.0, because Perl (from release 5.004) does not. + [Or at least, is documented not to: some releases seem to be in conflict + with the documentation.] However, when a pattern was studied with + pcre_study() and all its branches started with \s, PCRE still included VT + as a possible starting character. Of course, this did no harm; it just + caused an unnecessary match attempt. + +22. Removed a now-redundant internal flag bit that recorded the fact that case + dependency changed within the pattern. This was once needed for "required + byte" processing, but is no longer used. This recovers a now-scarce options + bit. Also moved the least significant internal flag bit to the most- + significant bit of the word, which was not previously used (hangover from + the days when it was an int rather than a uint) to free up another bit for + the future. + +23. Added support for CRLF line endings as well as CR and LF. As well as the + default being selectable at build time, it can now be changed at runtime + via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to + specify that it is scanning data with non-default line endings. + +24. Changed the definition of CXXLINK to make it agree with the definition of + LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS. + +25. Applied Ian Taylor's patches to avoid using another stack frame for tail + recursions. This makes a big different to stack usage for some patterns. + +26. If a subpattern containing a named recursion or subroutine reference such + as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of + the space required for the compiled pattern went wrong and gave too small a + value. Depending on the environment, this could lead to "Failed: internal + error: code overflow at offset 49" or "glibc detected double free or + corruption" errors. + +27. Applied patches from Google (a) to support the new newline modes and (b) to + advance over multibyte UTF-8 characters in GlobalReplace. + +28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a + difference for some implementation of PCRE in some Windows version. + +29. Added some extra testing facilities to pcretest: + + \q in a data line sets the "match limit" value + \Q in a data line sets the "match recursion limt" value + -S sets the stack size, where is in megabytes + + The -S option isn't available for Windows. + + +Version 6.6 06-Feb-06 +--------------------- + + 1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined + in pcreposix.h. I have copied the definition from pcre.h. + + 2. Change 25 for 6.5 broke compilation in a build directory out-of-tree + because pcre.h is no longer a built file. + + 3. Added Jeff Friedl's additional debugging patches to pcregrep. These are + not normally included in the compiled code. + + +Version 6.5 01-Feb-06 +--------------------- + + 1. When using the partial match feature with pcre_dfa_exec(), it was not + anchoring the second and subsequent partial matches at the new starting + point. This could lead to incorrect results. For example, with the pattern + /1234/, partially matching against "123" and then "a4" gave a match. + + 2. Changes to pcregrep: + + (a) All non-match returns from pcre_exec() were being treated as failures + to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an + error message is output. Some extra information is given for the + PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are + probably the only errors that are likely to be caused by users (by + specifying a regex that has nested indefinite repeats, for instance). + If there are more than 20 of these errors, pcregrep is abandoned. + + (b) A binary zero was treated as data while matching, but terminated the + output line if it was written out. This has been fixed: binary zeroes + are now no different to any other data bytes. + + (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is + used to set a locale for matching. The --locale=xxxx long option has + been added (no short equivalent) to specify a locale explicitly on the + pcregrep command, overriding the environment variables. + + (d) When -B was used with -n, some line numbers in the output were one less + than they should have been. + + (e) Added the -o (--only-matching) option. + + (f) If -A or -C was used with -c (count only), some lines of context were + accidentally printed for the final match. + + (g) Added the -H (--with-filename) option. + + (h) The combination of options -rh failed to suppress file names for files + that were found from directory arguments. + + (i) Added the -D (--devices) and -d (--directories) options. + + (j) Added the -F (--fixed-strings) option. + + (k) Allow "-" to be used as a file name for -f as well as for a data file. + + (l) Added the --colo(u)r option. + + (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it + is not present by default. + + 3. A nasty bug was discovered in the handling of recursive patterns, that is, + items such as (?R) or (?1), when the recursion could match a number of + alternatives. If it matched one of the alternatives, but subsequently, + outside the recursion, there was a failure, the code tried to back up into + the recursion. However, because of the way PCRE is implemented, this is not + possible, and the result was an incorrect result from the match. + + In order to prevent this happening, the specification of recursion has + been changed so that all such subpatterns are automatically treated as + atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)). + + 4. I had overlooked the fact that, in some locales, there are characters for + which isalpha() is true but neither isupper() nor islower() are true. In + the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine + and ordfeminine) are like this. This affected the treatment of \w and \W + when they appeared in character classes, but not when they appeared outside + a character class. The bit map for "word" characters is now created + separately from the results of isalnum() instead of just taking it from the + upper, lower, and digit maps. (Plus the underscore character, of course.) + + 5. The above bug also affected the handling of POSIX character classes such as + [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's + permanent tables. Instead, the bit maps for such a class were previously + created as the appropriate unions of the upper, lower, and digit bitmaps. + Now they are created by subtraction from the [[:word:]] class, which has + its own bitmap. + + 6. The [[:blank:]] character class matches horizontal, but not vertical space. + It is created by subtracting the vertical space characters (\x09, \x0a, + \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the + subtraction was done in the overall bitmap for a character class, meaning + that a class such as [\x0c[:blank:]] was incorrect because \x0c would not + be recognized. This bug has been fixed. + + 7. Patches from the folks at Google: + + (a) pcrecpp.cc: "to handle a corner case that may or may not happen in + real life, but is still worth protecting against". + + (b) pcrecpp.cc: "corrects a bug when negative radixes are used with + regular expressions". + + (c) pcre_scanner.cc: avoid use of std::count() because not all systems + have it. + + (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by + "configure" and the latter not, in order to fix a problem somebody had + with compiling the Arg class on HP-UX. + + (e) Improve the error-handling of the C++ wrapper a little bit. + + (f) New tests for checking recursion limiting. + + 8. The pcre_memmove() function, which is used only if the environment does not + have a standard memmove() function (and is therefore rarely compiled), + contained two bugs: (a) use of int instead of size_t, and (b) it was not + returning a result (though PCRE never actually uses the result). + + 9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously + large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is + returned instead of calling malloc() with an overflowing number that would + most likely cause subsequent chaos. + +10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag. + +11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled + with this option is matched, the nmatch and pmatch options of regexec() are + ignored. + +12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is + provided in case anyone wants to the the POSIX interface with UTF-8 + strings. + +13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the + C++ linking (needed for some HP-UX environments). + +14. Avoid compiler warnings in get_ucpname() when compiled without UCP support + (unused parameter) and in the pcre_printint() function (omitted "default" + switch label when the default is to do nothing). + +15. Added some code to make it possible, when PCRE is compiled as a C++ + library, to replace subject pointers for pcre_exec() with a smart pointer + class, thus making it possible to process discontinuous strings. + +16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform + much the same function. They were added by different people who were trying + to make PCRE easy to compile on non-Unix systems. It has been suggested + that PCRE_EXPORT be abolished now that there is more automatic apparatus + for compiling on Windows systems. I have therefore replaced it with + PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it + defaults to "extern" for C or "extern C" for C++, which works fine on + Unix-like systems. It is now possible to override the value of PCRE_DATA_ + SCOPE with something explicit in config.h. In addition: + + (a) pcreposix.h still had just "extern" instead of either of these macros; + I have replaced it with PCRE_DATA_SCOPE. + + (b) Functions such as _pcre_xclass(), which are internal to the library, + but external in the C sense, all had PCRE_EXPORT in their definitions. + This is apparently wrong for the Windows case, so I have removed it. + (It makes no difference on Unix-like systems.) + +17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting + of recursive calls to match(). This is different to MATCH_LIMIT because + that limits the total number of calls to match(), not all of which increase + the depth of recursion. Limiting the recursion depth limits the amount of + stack (or heap if NO_RECURSE is set) that is used. The default can be set + when PCRE is compiled, and changed at run time. A patch from Google adds + this functionality to the C++ interface. + +18. Changes to the handling of Unicode character properties: + + (a) Updated the table to Unicode 4.1.0. + + (b) Recognize characters that are not in the table as "Cn" (undefined). + + (c) I revised the way the table is implemented to a much improved format + which includes recognition of ranges. It now supports the ranges that + are defined in UnicodeData.txt, and it also amalgamates other + characters into ranges. This has reduced the number of entries in the + table from around 16,000 to around 3,000, thus reducing its size + considerably. I realized I did not need to use a tree structure after + all - a binary chop search is just as efficient. Having reduced the + number of entries, I extended their size from 6 bytes to 8 bytes to + allow for more data. + + (d) Added support for Unicode script names via properties such as \p{Han}. + +19. In UTF-8 mode, a backslash followed by a non-Ascii character was not + matching that character. + +20. When matching a repeated Unicode property with a minimum greater than zero, + (for example \pL{2,}), PCRE could look past the end of the subject if it + reached it while seeking the minimum number of characters. This could + happen only if some of the characters were more than one byte long, because + there is a check for at least the minimum number of bytes. + +21. Refactored the implementation of \p and \P so as to be more general, to + allow for more different types of property in future. This has changed the + compiled form incompatibly. Anybody with saved compiled patterns that use + \p or \P will have to recompile them. + +22. Added "Any" and "L&" to the supported property types. + +23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode, + but give a compile time error if the value is greater than 0xff. + +24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were + accidentally not being installed or uninstalled. + +25. The pcre.h file was built from pcre.h.in, but the only changes that were + made were to insert the current release number. This seemed silly, because + it made things harder for people building PCRE on systems that don't run + "configure". I have turned pcre.h into a distributed file, no longer built + by "configure", with the version identification directly included. There is + no longer a pcre.h.in file. + + However, this change necessitated a change to the pcre-config script as + well. It is built from pcre-config.in, and one of the substitutions was the + release number. I have updated configure.ac so that ./configure now finds + the release number by grepping pcre.h. + +26. Added the ability to run the tests under valgrind. + + +Version 6.4 05-Sep-05 +--------------------- + + 1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines + "--" to be printed when multiple files were scanned, even when none of the + -A, -B, or -C options were used. This is not compatible with Gnu grep, so I + consider it to be a bug, and have restored the previous behaviour. + + 2. A couple of code tidies to get rid of compiler warnings. + + 3. The pcretest program used to cheat by referring to symbols in the library + whose names begin with _pcre_. These are internal symbols that are not + really supposed to be visible externally, and in some environments it is + possible to suppress them. The cheating is now confined to including + certain files from the library's source, which is a bit cleaner. + + 4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the + file's purpose clearer. + + 5. Reorganized pcre_ucp_findchar(). + + +Version 6.3 15-Aug-05 +--------------------- + + 1. The file libpcre.pc.in did not have general read permission in the tarball. + + 2. There were some problems when building without C++ support: + + (a) If C++ support was not built, "make install" and "make test" still + tried to test it. + + (b) There were problems when the value of CXX was explicitly set. Some + changes have been made to try to fix these, and ... + + (c) --disable-cpp can now be used to explicitly disable C++ support. + + (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a + backslash in a target when C++ was disabled. This confuses some + versions of "make", apparently. Using an intermediate variable solves + this. (Same for CPP_LOBJ.) + + 3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK) + (non-Windows) now includes $(CFLAGS) because these flags are sometimes + necessary on certain architectures. + + 4. Added a setting of -export-symbols-regex to the link command to remove + those symbols that are exported in the C sense, but actually are local + within the library, and not documented. Their names all begin with + "_pcre_". This is not a perfect job, because (a) we have to except some + symbols that pcretest ("illegally") uses, and (b) the facility isn't always + available (and never for static libraries). I have made a note to try to + find a way round (a) in the future. + + +Version 6.2 01-Aug-05 +--------------------- + + 1. There was no test for integer overflow of quantifier values. A construction + such as {1111111111111111} would give undefined results. What is worse, if + a minimum quantifier for a parenthesized subpattern overflowed and became + negative, the calculation of the memory size went wrong. This could have + led to memory overwriting. + + 2. Building PCRE using VPATH was broken. Hopefully it is now fixed. + + 3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like + operating environments where this matters. + + 4. Applied Giuseppe Maxia's patch to add additional features for controlling + PCRE options from within the C++ wrapper. + + 5. Named capturing subpatterns were not being correctly counted when a pattern + was compiled. This caused two problems: (a) If there were more than 100 + such subpatterns, the calculation of the memory needed for the whole + compiled pattern went wrong, leading to an overflow error. (b) Numerical + back references of the form \12, where the number was greater than 9, were + not recognized as back references, even though there were sufficient + previous subpatterns. + + 6. Two minor patches to pcrecpp.cc in order to allow it to compile on older + versions of gcc, e.g. 2.95.4. + + +Version 6.1 21-Jun-05 +--------------------- + + 1. There was one reference to the variable "posix" in pcretest.c that was not + surrounded by "#if !defined NOPOSIX". + + 2. Make it possible to compile pcretest without DFA support, UTF8 support, or + the cross-check on the old pcre_info() function, for the benefit of the + cut-down version of PCRE that is currently imported into Exim. + + 3. A (silly) pattern starting with (?i)(?-i) caused an internal space + allocation error. I've done the easy fix, which wastes 2 bytes for sensible + patterns that start (?i) but I don't think that matters. The use of (?i) is + just an example; this all applies to the other options as well. + + 4. Since libtool seems to echo the compile commands it is issuing, the output + from "make" can be reduced a bit by putting "@" in front of each libtool + compile command. + + 5. Patch from the folks at Google for configure.in to be a bit more thorough + in checking for a suitable C++ installation before trying to compile the + C++ stuff. This should fix a reported problem when a compiler was present, + but no suitable headers. + + 6. The man pages all had just "PCRE" as their title. I have changed them to + be the relevant file name. I have also arranged that these names are + retained in the file doc/pcre.txt, which is a concatenation in text format + of all the man pages except the little individual ones for each function. + + 7. The NON-UNIX-USE file had not been updated for the different set of source + files that come with release 6. I also added a few comments about the C++ + wrapper. + + +Version 6.0 07-Jun-05 +--------------------- + + 1. Some minor internal re-organization to help with my DFA experiments. + + 2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that + didn't matter for the library itself when fully configured, but did matter + when compiling without UCP support, or within Exim, where the ucp files are + not imported. + + 3. Refactoring of the library code to split up the various functions into + different source modules. The addition of the new DFA matching code (see + below) to a single monolithic source would have made it really too + unwieldy, quite apart from causing all the code to be include in a + statically linked application, when only some functions are used. This is + relevant even without the DFA addition now that patterns can be compiled in + one application and matched in another. + + The downside of splitting up is that there have to be some external + functions and data tables that are used internally in different modules of + the library but which are not part of the API. These have all had their + names changed to start with "_pcre_" so that they are unlikely to clash + with other external names. + + 4. Added an alternate matching function, pcre_dfa_exec(), which matches using + a different (DFA) algorithm. Although it is slower than the original + function, it does have some advantages for certain types of matching + problem. + + 5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(), + including restarting after a partial match. + + 6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not + defined when compiling for Windows was sent to me. I have put it into the + code, though I have no means of testing or verifying it. + + 7. Added the pcre_refcount() auxiliary function. + + 8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to + match before or at the first newline in the subject string. In pcretest, + the /f option on a pattern can be used to set this. + + 9. A repeated \w when used in UTF-8 mode with characters greater than 256 + would behave wrongly. This has been present in PCRE since release 4.0. + +10. A number of changes to the pcregrep command: + + (a) Refactored how -x works; insert ^(...)$ instead of setting + PCRE_ANCHORED and checking the length, in preparation for adding + something similar for -w. + + (b) Added the -w (match as a word) option. + + (c) Refactored the way lines are read and buffered so as to have more + than one at a time available. + + (d) Implemented a pcregrep test script. + + (e) Added the -M (multiline match) option. This allows patterns to match + over several lines of the subject. The buffering ensures that at least + 8K, or the rest of the document (whichever is the shorter) is available + for matching (and similarly the previous 8K for lookbehind assertions). + + (f) Changed the --help output so that it now says + + -w, --word-regex(p) + + instead of two lines, one with "regex" and the other with "regexp" + because that confused at least one person since the short forms are the + same. (This required a bit of code, as the output is generated + automatically from a table. It wasn't just a text change.) + + (g) -- can be used to terminate pcregrep options if the next thing isn't an + option but starts with a hyphen. Could be a pattern or a path name + starting with a hyphen, for instance. + + (h) "-" can be given as a file name to represent stdin. + + (i) When file names are being printed, "(standard input)" is used for + the standard input, for compatibility with GNU grep. Previously + "" was used. + + (j) The option --label=xxx can be used to supply a name to be used for + stdin when file names are being printed. There is no short form. + + (k) Re-factored the options decoding logic because we are going to add + two more options that take data. Such options can now be given in four + different ways, e.g. "-fname", "-f name", "--file=name", "--file name". + + (l) Added the -A, -B, and -C options for requesting that lines of context + around matches be printed. + + (m) Added the -L option to print the names of files that do not contain + any matching lines, that is, the complement of -l. + + (n) The return code is 2 if any file cannot be opened, but pcregrep does + continue to scan other files. + + (o) The -s option was incorrectly implemented. For compatibility with other + greps, it now suppresses the error message for a non-existent or non- + accessible file (but not the return code). There is a new option called + -q that suppresses the output of matching lines, which was what -s was + previously doing. + + (p) Added --include and --exclude options to specify files for inclusion + and exclusion when recursing. + +11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly. + Hopefully, it now does. + +12. Missing cast in pcre_study(). + +13. Added an "uninstall" target to the makefile. + +14. Replaced "extern" in the function prototypes in Makefile.in with + "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix + world, but is set differently for Windows. + +15. Added a second compiling function called pcre_compile2(). The only + difference is that it has an extra argument, which is a pointer to an + integer error code. When there is a compile-time failure, this is set + non-zero, in addition to the error test pointer being set to point to an + error message. The new argument may be NULL if no error number is required + (but then you may as well call pcre_compile(), which is now just a + wrapper). This facility is provided because some applications need a + numeric error indication, but it has also enabled me to tidy up the way + compile-time errors are handled in the POSIX wrapper. + +16. Added VPATH=.libs to the makefile; this should help when building with one + prefix path and installing with another. (Or so I'm told by someone who + knows more about this stuff than I do.) + +17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This + passes PCRE_DOTALL to the pcre_compile() function, making the "." character + match everything, including newlines. This is not POSIX-compatible, but + somebody wanted the feature. From pcretest it can be activated by using + both the P and the s flags. + +18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one. + +19. libpcre.pc was being incorrectly installed as executable. + +20. A couple of places in pcretest check for end-of-line by looking for '\n'; + it now also looks for '\r' so that it will work unmodified on Windows. + +21. Added Google's contributed C++ wrapper to the distribution. + +22. Added some untidy missing memory free() calls in pcretest, to keep + Electric Fence happy when testing. + + + +Version 5.0 13-Sep-04 +--------------------- + + 1. Internal change: literal characters are no longer packed up into items + containing multiple characters in a single byte-string. Each character + is now matched using a separate opcode. However, there may be more than one + byte in the character in UTF-8 mode. + + 2. The pcre_callout_block structure has two new fields: pattern_position and + next_item_length. These contain the offset in the pattern to the next match + item, and its length, respectively. + + 3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic + insertion of callouts before each pattern item. Added the /C option to + pcretest to make use of this. + + 4. On the advice of a Windows user, the lines + + #if defined(_WIN32) || defined(WIN32) + _setmode( _fileno( stdout ), 0x8000 ); + #endif /* defined(_WIN32) || defined(WIN32) */ + + have been added to the source of pcretest. This apparently does useful + magic in relation to line terminators. + + 5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb" + for the benefit of those environments where the "b" makes a difference. + + 6. The icc compiler has the same options as gcc, but "configure" doesn't seem + to know about it. I have put a hack into configure.in that adds in code + to set GCC=yes if CC=icc. This seems to end up at a point in the + generated configure script that is early enough to affect the setting of + compiler options, which is what is needed, but I have no means of testing + whether it really works. (The user who reported this had patched the + generated configure script, which of course I cannot do.) + + LATER: After change 22 below (new libtool files), the configure script + seems to know about icc (and also ecc). Therefore, I have commented out + this hack in configure.in. + + 7. Added support for pkg-config (2 patches were sent in). + + 8. Negated POSIX character classes that used a combination of internal tables + were completely broken. These were [[:^alpha:]], [[:^alnum:]], and + [[:^ascii]]. Typically, they would match almost any characters. The other + POSIX classes were not broken in this way. + + 9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed + to find the match, as PCRE was deluded into thinking that the match had to + start at the start point or following a newline. The same bug applied to + patterns with negative forward assertions or any backward assertions + preceding ".*" at the start, unless the pattern required a fixed first + character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed. + +10. In UTF-8 mode, when moving forwards in the subject after a failed match + starting at the last subject character, bytes beyond the end of the subject + string were read. + +11. Renamed the variable "class" as "classbits" to make life easier for C++ + users. (Previously there was a macro definition, but it apparently wasn't + enough.) + +12. Added the new field "tables" to the extra data so that tables can be passed + in at exec time, or the internal tables can be re-selected. This allows + a compiled regex to be saved and re-used at a later time by a different + program that might have everything at different addresses. + +13. Modified the pcre-config script so that, when run on Solaris, it shows a + -R library as well as a -L library. + +14. The debugging options of pcretest (-d on the command line or D on a + pattern) showed incorrect output for anything following an extended class + that contained multibyte characters and which was followed by a quantifier. + +15. Added optional support for general category Unicode character properties + via the \p, \P, and \X escapes. Unicode property support implies UTF-8 + support. It adds about 90K to the size of the library. The meanings of the + inbuilt class escapes such as \d and \s have NOT been changed. + +16. Updated pcredemo.c to include calls to free() to release the memory for the + compiled pattern. + +17. The generated file chartables.c was being created in the source directory + instead of in the building directory. This caused the build to fail if the + source directory was different from the building directory, and was + read-only. + +18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE + file. No doubt somebody will tell me if they don't make sense... Also added + Dan Mooney's comments about building on OpenVMS. + +19. Added support for partial matching via the PCRE_PARTIAL option for + pcre_exec() and the \P data escape in pcretest. + +20. Extended pcretest with 3 new pattern features: + + (i) A pattern option of the form ">rest-of-line" causes pcretest to + write the compiled pattern to the file whose name is "rest-of-line". + This is a straight binary dump of the data, with the saved pointer to + the character tables forced to be NULL. The study data, if any, is + written too. After writing, pcretest reads a new pattern. + + (ii) If, instead of a pattern, ": new target + : new target + : use native compiler + : use native linker + : handle Windows platform correctly + : ditto + : ditto + copy DLL to top builddir before testing + + As part of these changes, -no-undefined was removed again. This was reported + to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea + in any case. + +3. Some tidies to get rid of compiler warnings: + + . In the match_data structure, match_limit was an unsigned long int, whereas + match_call_count was an int. I've made them both unsigned long ints. + + . In pcretest the fact that a const uschar * doesn't automatically cast to + a void * provoked a warning. + + . Turning on some more compiler warnings threw up some "shadow" variables + and a few more missing casts. + +4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8 + option, a class that contained a single character with a value between 128 + and 255 (e.g. /[\xFF]/) caused PCRE to crash. + +5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8 + option, a class that contained several characters, but with at least one + whose value was between 128 and 255 caused PCRE to crash. + + +Version 4.1 12-Mar-03 +--------------------- + +1. Compiling with gcc -pedantic found a couple of places where casts were +needed, and a string in dftables.c that was longer than standard compilers are +required to support. + +2. Compiling with Sun's compiler found a few more places where the code could +be tidied up in order to avoid warnings. + +3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the +first of these names is deprecated in the latest Autoconf in favour of the name +CC_FOR_BUILD, because "host" is typically used to mean the system on which the +compiled code will be run. I can't find a reference for HOST_CFLAGS, but by +analogy I have changed it to CFLAGS_FOR_BUILD. + +4. Added -no-undefined to the linking command in the Makefile, because this is +apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the +linking step for the pcreposix library. + +5. PCRE was failing to diagnose the case of two named groups with the same +name. + +6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a +literal character that is needed in the subject for a match, and scans along to +ensure that it is present before embarking on the full matching process. This +saves time in cases of nested unlimited repeats that are never going to match. +Problem: the scan can take a lot of time if the subject is very long (e.g. +megabytes), thus penalizing straightforward matches. It is now done only if the +amount of subject to be scanned is less than 1000 bytes. + +7. A lesser problem with the same optimization is that it was recording the +first character of an anchored pattern as "needed", thus provoking a search +right along the subject, even when the first match of the pattern was going to +fail. The "needed" character is now not set for anchored patterns, unless it +follows something in the pattern that is of non-fixed length. Thus, it still +fulfils its original purpose of finding quick non-matches in cases of nested +unlimited repeats, but isn't used for simple anchored patterns such as /^abc/. + + +Version 4.0 17-Feb-03 +--------------------- + +1. If a comment in an extended regex that started immediately after a meta-item +extended to the end of string, PCRE compiled incorrect data. This could lead to +all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not. + +2. Moved to autoconf 2.53 and libtool 1.4.2. + +3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently, +the special perltest8 script is no longer needed - all the tests can be run +from a single perltest script. + +4. From 5.004, Perl has not included the VT character (0x0b) in the set defined +by \s. It has now been removed in PCRE. This means it isn't recognized as +whitespace in /x regexes too, which is the same as Perl. Note that the POSIX +class [:space:] *does* include VT, thereby creating a mess. + +5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only +space and tab. + +6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use +its new features into the main test script, reducing the number of scripts. + +7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions +were backward compatible, and made the (?i) apply to the whole pattern, as if +/i were given. Now it behaves more logically, and applies the option setting +only to what follows. PCRE has been changed to follow suit. However, if it +finds options settings right at the start of the pattern, it extracts them into +the global options, as before. Thus, they show up in the info data. + +8. Added support for the \Q...\E escape sequence. Characters in between are +treated as literals. This is slightly different from Perl in that $ and @ are +also handled as literals inside the quotes. In Perl, they will cause variable +interpolation. Note the following examples: + + Pattern PCRE matches Perl matches + + \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz + \Qabc\$xyz\E abc\$xyz abc\$xyz + \Qabc\E\$\Qxyz\E abc$xyz abc$xyz + +For compatibility with Perl, \Q...\E sequences are recognized inside character +classes as well as outside them. + +9. Re-organized 3 code statements in pcretest to avoid "overflow in +floating-point constant arithmetic" warnings from a Microsoft compiler. Added a +(size_t) cast to one statement in pcretest and one in pcreposix to avoid +signed/unsigned warnings. + +10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o +option for pcretest, so I've replaced it by a simple function that does just +that job. + +11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or +"pcregrep -". + +12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's +Java package. This provides some syntactic sugar for simple cases of what my +documentation calls "once-only subpatterns". A pattern such as x*+ is the same +as (?>x*). In other words, if what is inside (?>...) is just a single repeated +item, you can use this simplified notation. Note that only makes sense with +greedy quantifiers. Consequently, the use of the possessive quantifier forces +greediness, whatever the setting of the PCRE_UNGREEDY option. + +13. A change of greediness default within a pattern was not taking effect at +the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized +subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option +was abstracted outside. + +14. PCRE now supports the \G assertion. It is true when the current matching +position is at the start point of the match. This differs from \A when the +starting offset is non-zero. Used with the /g option of pcretest (or similar +code), it works in the same way as it does for Perl's /g option. If all +alternatives of a regex begin with \G, the expression is anchored to the start +match position, and the "anchored" flag is set in the compiled expression. + +15. Some bugs concerning the handling of certain option changes within patterns +have been fixed. These applied to options other than (?ims). For example, +"a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have +been the other way round. Some of this was related to change 7 above. + +16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX +features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/ +and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports +POSIX classes only within a class (e.g. /[[:alpha:]]/). + +17. Added support for Perl's \C escape. This matches one byte, even in UTF8 +mode. Unlike ".", it always matches newline, whatever the setting of +PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind +assertions. Perl allows it, but it doesn't (in general) work because it can't +calculate the length of the lookbehind. At least, that's the case for Perl +5.8.0 - I've been told they are going to document that it doesn't work in +future. + +18. Added an error diagnosis for escapes that PCRE does not support: these are +\L, \l, \N, \P, \p, \U, \u, and \X. + +19. Although correctly diagnosing a missing ']' in a character class, PCRE was +reading past the end of the pattern in cases such as /[abcd/. + +20. PCRE was getting more memory than necessary for patterns with classes that +contained both POSIX named classes and other characters, e.g. /[[:space:]abc/. + +21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for +compiling PCRE for use with Virtual Pascal. + +22. Small fix to the Makefile to make it work properly if the build is done +outside the source tree. + +23. Added a new extension: a condition to go with recursion. If a conditional +subpattern starts with (?(R) the "true" branch is used if recursion has +happened, whereas the "false" branch is used only at the top level. + +24. When there was a very long string of literal characters (over 255 bytes +without UTF support, over 250 bytes with UTF support), the computation of how +much memory was required could be incorrect, leading to segfaults or other +strange effects. + +25. PCRE was incorrectly assuming anchoring (either to start of subject or to +start of line for a non-DOTALL pattern) when a pattern started with (.*) and +there was a subsequent back reference to those brackets. This meant that, for +example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't +possible to check for precisely this case. All we can do is abandon the +optimization if .* occurs inside capturing brackets when there are any back +references whatsoever. (See below for a better fix that came later.) + +26. The handling of the optimization for finding the first character of a +non-anchored pattern, and for finding a character that is required later in the +match were failing in some cases. This didn't break the matching; it just +failed to optimize when it could. The way this is done has been re-implemented. + +27. Fixed typo in error message for invalid (?R item (it said "(?p"). + +28. Added a new feature that provides some of the functionality that Perl +provides with (?{...}). The facility is termed a "callout". The way it is done +in PCRE is for the caller to provide an optional function, by setting +pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a +global variable. By default it is unset, which disables all calling out. To get +the function called, the regex must include (?C) at appropriate points. This +is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C). +This provides a means of identifying different callout points. When PCRE +reaches such a point in the regex, if pcre_callout has been set, the external +function is called. It is provided with data in a structure called +pcre_callout_block, which is defined in pcre.h. If the function returns 0, +matching continues; if it returns a non-zero value, the match at the current +point fails. However, backtracking will occur if possible. [This was changed +later and other features added - see item 49 below.] + +29. pcretest is upgraded to test the callout functionality. It provides a +callout function that displays information. By default, it shows the start of +the match and the current position in the text. There are some new data escapes +to vary what happens: + + \C+ in addition, show current contents of captured substrings + \C- do not supply a callout function + \C!n return 1 when callout number n is reached + \C!n!m return 1 when callout number n is reached for the mth time + +30. If pcregrep was called with the -l option and just a single file name, it +output "" if a match was found, instead of the file name. + +31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing +slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to +pcre_exec(). This saves a malloc/free per call. The default value of +POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold +when configuring. + +32. The default maximum size of a compiled pattern is 64K. There have been a +few cases of people hitting this limit. The code now uses macros to handle the +storing of links as offsets within the compiled pattern. It defaults to 2-byte +links, but this can be changed to 3 or 4 bytes by --with-link-size when +configuring. Tests 2 and 5 work only with 2-byte links because they output +debugging information about compiled patterns. + +33. Internal code re-arrangements: + +(a) Moved the debugging function for printing out a compiled regex into + its own source file (printint.c) and used #include to pull it into + pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two + separate copies. + +(b) Defined the list of op-code names for debugging as a macro in + internal.h so that it is next to the definition of the opcodes. + +(c) Defined a table of op-code lengths for simpler skipping along compiled + code. This is again a macro in internal.h so that it is next to the + definition of the opcodes. + +34. Added support for recursive calls to individual subpatterns, along the +lines of Robin Houston's patch (but implemented somewhat differently). + +35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to +allow it to read and process whole directories in Win32. This code was +contributed by Lionel Fourquaux; it has not been tested by me. + +36. Added support for named subpatterns. The Python syntax (?P...) is +used to name a group. Names consist of alphanumerics and underscores, and must +be unique. Back references use the syntax (?P=name) and recursive calls use +(?P>name) which is a PCRE extension to the Python extension. Groups still have +numbers. The function pcre_fullinfo() can be used after compilation to extract +a name/number map. There are three relevant calls: + + PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map + PCRE_INFO_NAMECOUNT yields the number of entries + PCRE_INFO_NAMETABLE yields a pointer to the map. + +The map is a vector of fixed-size entries. The size of each entry depends on +the length of the longest name used. The first two bytes of each entry are the +group number, most significant byte first. There follows the corresponding +name, zero terminated. The names are in alphabetical order. + +37. Make the maximum literal string in the compiled code 250 for the non-UTF-8 +case instead of 255. Making it the same both with and without UTF-8 support +means that the same test output works with both. + +38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid +calling malloc() with a zero argument. + +39. Change 25 above had to resort to a heavy-handed test for the .* anchoring +optimization. I've improved things by keeping a bitmap of backreferences with +numbers 1-31 so that if .* occurs inside capturing brackets that are not in +fact referenced, the optimization can be applied. It is unlikely that a +relevant occurrence of .* (i.e. one which might indicate anchoring or forcing +the match to follow \n) will appear inside brackets with a number greater than +31, but if it does, any back reference > 31 suppresses the optimization. + +40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect +of disabling numbered capturing parentheses. Any opening parenthesis that is +not followed by ? behaves as if it were followed by ?: but named parentheses +can still be used for capturing (and they will acquire numbers in the usual +way). + +41. Redesigned the return codes from the match() function into yes/no/error so +that errors can be passed back from deep inside the nested calls. A malloc +failure while inside a recursive subpattern call now causes the +PCRE_ERROR_NOMEMORY return instead of quietly going wrong. + +42. It is now possible to set a limit on the number of times the match() +function is called in a call to pcre_exec(). This facility makes it possible to +limit the amount of recursion and backtracking, though not in a directly +obvious way, because the match() function is used in a number of different +circumstances. The count starts from zero for each position in the subject +string (for non-anchored patterns). The default limit is, for compatibility, a +large number, namely 10 000 000. You can change this in two ways: + +(a) When configuring PCRE before making, you can use --with-match-limit=n + to set a default value for the compiled library. + +(b) For each call to pcre_exec(), you can pass a pcre_extra block in which + a different value is set. See 45 below. + +If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT. + +43. Added a new function pcre_config(int, void *) to enable run-time extraction +of things that can be changed at compile time. The first argument specifies +what is wanted and the second points to where the information is to be placed. +The current list of available information is: + + PCRE_CONFIG_UTF8 + +The output is an integer that is set to one if UTF-8 support is available; +otherwise it is set to zero. + + PCRE_CONFIG_NEWLINE + +The output is an integer that it set to the value of the code that is used for +newline. It is either LF (10) or CR (13). + + PCRE_CONFIG_LINK_SIZE + +The output is an integer that contains the number of bytes used for internal +linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above. + + PCRE_CONFIG_POSIX_MALLOC_THRESHOLD + +The output is an integer that contains the threshold above which the POSIX +interface uses malloc() for output vectors. See item 31 above. + + PCRE_CONFIG_MATCH_LIMIT + +The output is an unsigned integer that contains the default limit of the number +of match() calls in a pcre_exec() execution. See 42 above. + +44. pcretest has been upgraded by the addition of the -C option. This causes it +to extract all the available output from the new pcre_config() function, and to +output it. The program then exits immediately. + +45. A need has arisen to pass over additional data with calls to pcre_exec() in +order to support additional features. One way would have been to define +pcre_exec2() (for example) with extra arguments, but this would not have been +extensible, and would also have required all calls to the original function to +be mapped to the new one. Instead, I have chosen to extend the mechanism that +is used for passing in "extra" data from pcre_study(). + +The pcre_extra structure is now exposed and defined in pcre.h. It currently +contains the following fields: + + flags a bitmap indicating which of the following fields are set + study_data opaque data from pcre_study() + match_limit a way of specifying a limit on match() calls for a specific + call to pcre_exec() + callout_data data for callouts (see 49 below) + +The flag bits are also defined in pcre.h, and are + + PCRE_EXTRA_STUDY_DATA + PCRE_EXTRA_MATCH_LIMIT + PCRE_EXTRA_CALLOUT_DATA + +The pcre_study() function now returns one of these new pcre_extra blocks, with +the actual study data pointed to by the study_data field, and the +PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as +before. That is, this change is entirely upwards-compatible and requires no +change to existing code. + +If you want to pass in additional data to pcre_exec(), you can either place it +in a pcre_extra block provided by pcre_study(), or create your own pcre_extra +block. + +46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a +data string contains the escape sequence \M, pcretest calls pcre_exec() several +times with different match limits, until it finds the minimum value needed for +pcre_exec() to complete. The value is then output. This can be instructive; for +most simple matches the number is quite small, but for pathological cases it +gets very large very quickly. + +47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It +returns the size of the data block pointed to by the study_data field in a +pcre_extra block, that is, the value that was passed as the argument to +pcre_malloc() when PCRE was getting memory in which to place the information +created by pcre_study(). The fourth argument should point to a size_t variable. +pcretest has been extended so that this information is shown after a successful +pcre_study() call when information about the compiled regex is being displayed. + +48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR) +because what follows is always an absolute path. (Later: it turns out that this +is more than cosmetic for MinGW, because it doesn't like empty path +components.) + +49. Some changes have been made to the callout feature (see 28 above): + +(i) A callout function now has three choices for what it returns: + + 0 => success, carry on matching + > 0 => failure at this point, but backtrack if possible + < 0 => serious error, return this value from pcre_exec() + + Negative values should normally be chosen from the set of PCRE_ERROR_xxx + values. In particular, returning PCRE_ERROR_NOMATCH forces a standard + "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for + use by callout functions. It will never be used by PCRE itself. + +(ii) The pcre_extra structure (see 45 above) has a void * field called + callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The + pcre_callout_block structure has a field of the same name. The contents of + the field passed in the pcre_extra structure are passed to the callout + function in the corresponding field in the callout block. This makes it + easier to use the same callout-containing regex from multiple threads. For + testing, the pcretest program has a new data escape + + \C*n pass the number n (may be negative) as callout_data + + If the callout function in pcretest receives a non-zero value as + callout_data, it returns that value. + +50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also, +there were some redundant $(CFLAGS) in commands that are now specified as +$(LINK), which already includes $(CFLAGS). + +51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE +has been compiled with UTF-8 support *and* pcre_compile() has been compiled +with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume +one-byte characters throughout. Note that case-insensitive matching applies +only to characters whose values are less than 256. PCRE doesn't support the +notion of cases for higher-valued characters. + +(i) A character class whose characters are all within 0-255 is handled as + a bit map, and the map is inverted for negative classes. Previously, a + character > 255 always failed to match such a class; however it should + match if the class was a negative one (e.g. [^ab]). This has been fixed. + +(ii) A negated character class with a single character < 255 is coded as + "not this character" (OP_NOT). This wasn't working properly when the test + character was multibyte, either singly or repeated. + +(iii) Repeats of multibyte characters are now handled correctly in UTF-8 + mode, for example: \x{100}{2,3}. + +(iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either + singly or repeated) now correctly test multibyte characters. However, + PCRE doesn't recognize any characters with values greater than 255 as + digits, spaces, or word characters. Such characters always match \D, \S, + and \W, and never match \d, \s, or \w. + +(v) Classes may now contain characters and character ranges with values + greater than 255. For example: [ab\x{100}-\x{400}]. + +(vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call + PCRE in UTF-8 mode. + +52. The info request value PCRE_INFO_FIRSTCHAR has been renamed +PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is +retained for backwards compatibility. (Note that LASTLITERAL is also a byte +value.) + +53. The single man page has become too large. I have therefore split it up into +a number of separate man pages. These also give rise to individual HTML pages; +these are now put in a separate directory, and there is an index.html page that +lists them all. Some hyperlinking between the pages has been installed. + +54. Added convenience functions for handling named capturing parentheses. + +55. Unknown escapes inside character classes (e.g. [\M]) and escapes that +aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also +true in PCRE, except when the PCRE_EXTENDED option is set, in which case they +are faulted. + +56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when +calling configure. These values are used when compiling the dftables.c program +which is run to generate the source of the default character tables. They +default to the values of CC and CFLAGS. If you are cross-compiling PCRE, +you will need to set these values. + +57. Updated the building process for Windows DLL, as provided by Fred Cox. + + +Version 3.9 02-Jan-02 +--------------------- + +1. A bit of extraneous text had somehow crept into the pcregrep documentation. + +2. If --disable-static was given, the building process failed when trying to +build pcretest and pcregrep. (For some reason it was using libtool to compile +them, which is not right, as they aren't part of the library.) + + +Version 3.8 18-Dec-01 +--------------------- + +1. The experimental UTF-8 code was completely screwed up. It was packing the +bytes in the wrong order. How dumb can you get? + + +Version 3.7 29-Oct-01 +--------------------- + +1. In updating pcretest to check change 1 of version 3.6, I screwed up. +This caused pcretest, when used on the test data, to segfault. Unfortunately, +this didn't happen under Solaris 8, where I normally test things. + +2. The Makefile had to be changed to make it work on BSD systems, where 'make' +doesn't seem to recognize that ./xxx and xxx are the same file. (This entry +isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made +this fix an hour or so after the initial 3.7 release.) + + +Version 3.6 23-Oct-01 +--------------------- + +1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if +offsets passed as NULL with zero offset count. + +2. The config.guess and config.sub files had not been updated when I moved to +the latest autoconf. + + +Version 3.5 15-Aug-01 +--------------------- + +1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that +had been forgotten. + +2. By using declared but undefined structures, we can avoid using "void" +definitions in pcre.h while keeping the internal definitions of the structures +private. + +3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a +user point of view, this means that both static and shared libraries are built +by default, but this can be individually controlled. More of the work of +handling this static/shared cases is now inside libtool instead of PCRE's make +file. + +4. The pcretest utility is now installed along with pcregrep because it is +useful for users (to test regexs) and by doing this, it automatically gets +relinked by libtool. The documentation has been turned into a man page, so +there are now .1, .txt, and .html versions in /doc. + +5. Upgrades to pcregrep: + (i) Added long-form option names like gnu grep. + (ii) Added --help to list all options with an explanatory phrase. + (iii) Added -r, --recursive to recurse into sub-directories. + (iv) Added -f, --file to read patterns from a file. + +6. pcre_exec() was referring to its "code" argument before testing that +argument for NULL (and giving an error if it was NULL). + +7. Upgraded Makefile.in to allow for compiling in a different directory from +the source directory. + +8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the +options bits, the pointer it was passed was to an int instead of to an unsigned +long int. This mattered only on 64-bit systems. + +9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is +generated) instead of pcre.in, which it its source. Also made the same change +in several of the .c files. + +10. A new release of gcc defines printf() as a macro, which broke pcretest +because it had an ifdef in the middle of a string argument for printf(). Fixed +by using separate calls to printf(). + +11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure +script, to force use of CR or LF instead of \n in the source. On non-Unix +systems, the value can be set in config.h. + +12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an +absolute limit. Changed the text of the error message to make this clear, and +likewise updated the man page. + +13. The limit of 99 on the number of capturing subpatterns has been removed. +The new limit is 65535, which I hope will not be a "real" limit. + + +Version 3.4 22-Aug-00 +--------------------- + +1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *. + +2. Diagnose condition (?(0) as an error instead of crashing on matching. + + +Version 3.3 01-Aug-00 +--------------------- + +1. If an octal character was given, but the value was greater than \377, it +was not getting masked to the least significant bits, as documented. This could +lead to crashes in some systems. + +2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats +the hyphen as a literal. PCRE used to give an error; it now behaves like Perl. + +3. Added the functions pcre_free_substring() and pcre_free_substring_list(). +These just pass their arguments on to (pcre_free)(), but they are provided +because some uses of PCRE bind it to non-C systems that can call its functions, +but cannot call free() or pcre_free() directly. + +4. Add "make test" as a synonym for "make check". Corrected some comments in +the Makefile. + +5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the +Makefile. + +6. Changed the name of pgrep to pcregrep, because Solaris has introduced a +command called pgrep for grepping around the active processes. + +7. Added the beginnings of support for UTF-8 character strings. + +8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and +RANLIB to ./ltconfig so that they are used by libtool. I think these are all +the relevant ones. (AR is not passed because ./ltconfig does its own figuring +out for the ar command.) + + +Version 3.2 12-May-00 +--------------------- + +This is purely a bug fixing release. + +1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead +of ZA. This was just one example of several cases that could provoke this bug, +which was introduced by change 9 of version 2.00. The code for breaking +infinite loops after an iteration that matches an empty string was't working +correctly. + +2. The pcretest program was not imitating Perl correctly for the pattern /a*/g +when matched against abbab (for example). After matching an empty string, it +wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this +caused it to match further down the string than it should. + +3. The code contained an inclusion of sys/types.h. It isn't clear why this +was there because it doesn't seem to be needed, and it causes trouble on some +systems, as it is not a Standard C header. It has been removed. + +4. Made 4 silly changes to the source to avoid stupid compiler warnings that +were reported on the Macintosh. The changes were from + + while ((c = *(++ptr)) != 0 && c != '\n'); +to + while ((c = *(++ptr)) != 0 && c != '\n') ; + +Totally extraordinary, but if that's what it takes... + +5. PCRE is being used in one environment where neither memmove() nor bcopy() is +available. Added HAVE_BCOPY and an autoconf test for it; if neither +HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which +assumes the way PCRE uses memmove() (always moving upwards). + +6. PCRE is being used in one environment where strchr() is not available. There +was only one use in pcre.c, and writing it out to avoid strchr() probably gives +faster code anyway. + + +Version 3.1 09-Feb-00 +--------------------- + +The only change in this release is the fixing of some bugs in Makefile.in for +the "install" target: + +(1) It was failing to install pcreposix.h. + +(2) It was overwriting the pcre.3 man page with the pcreposix.3 man page. + + +Version 3.0 01-Feb-00 +--------------------- + +1. Add support for the /+ modifier to perltest (to output $` like it does in +pcretest). + +2. Add support for the /g modifier to perltest. + +3. Fix pcretest so that it behaves even more like Perl for /g when the pattern +matches null strings. + +4. Fix perltest so that it doesn't do unwanted things when fed an empty +pattern. Perl treats empty patterns specially - it reuses the most recent +pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this +effect. + +5. The POSIX interface was broken in that it was just handing over the POSIX +captured string vector to pcre_exec(), but (since release 2.00) PCRE has +required a bigger vector, with some working space on the end. This means that +the POSIX wrapper now has to get and free some memory, and copy the results. + +6. Added some simple autoconf support, placing the test data and the +documentation in separate directories, re-organizing some of the +information files, and making it build pcre-config (a GNU standard). Also added +libtool support for building PCRE as a shared library, which is now the +default. + +7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and +09 are not valid octal constants. Single digits will be used for minor values +less than 10. + +8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that +existing programs that set these in the POSIX interface can use PCRE without +modification. + +9. Added a new function, pcre_fullinfo() with an extensible interface. It can +return all that pcre_info() returns, plus additional data. The pcre_info() +function is retained for compatibility, but is considered to be obsolete. + +10. Added experimental recursion feature (?R) to handle one common case that +Perl 5.6 will be able to do with (?p{...}). + +11. Added support for POSIX character classes like [:alpha:], which Perl is +adopting. + + +Version 2.08 31-Aug-99 +---------------------- + +1. When startoffset was not zero and the pattern began with ".*", PCRE was not +trying to match at the startoffset position, but instead was moving forward to +the next newline as if a previous match had failed. + +2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G, +and could get into a loop if a null string was matched other than at the start +of the subject. + +3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can +be distinguished at compile time, and for completeness also added PCRE_DATE. + +5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL +in GnuWin32 environments. + + +Version 2.07 29-Jul-99 +---------------------- + +1. The documentation is now supplied in plain text form and HTML as well as in +the form of man page sources. + +2. C++ compilers don't like assigning (void *) values to other pointer types. +In particular this affects malloc(). Although there is no problem in Standard +C, I've put in casts to keep C++ compilers happy. + +3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call +should be (const char *). + +4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may +be useful for non-Unix systems who don't want to bother with the POSIX stuff. +However, I haven't made this a standard facility. The documentation doesn't +mention it, and the Makefile doesn't support it. + +5. The Makefile now contains an "install" target, with editable destinations at +the top of the file. The pcretest program is not installed. + +6. pgrep -V now gives the PCRE version number and date. + +7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was +causing the entire string to be ignored, instead of just the last character. + +8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a +non-matching string, it can take a very, very long time, even for strings of +quite modest length, because of the nested recursion. PCRE now does better in +some of these cases. It does this by remembering the last required literal +character in the pattern, and pre-searching the subject to ensure it is present +before running the real match. In other words, it applies a heuristic to detect +some types of certain failure quickly, and in the above example, if presented +with a string that has no trailing " it gives "no match" very quickly. + +9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored; +other alternatives are tried instead. + + +Version 2.06 09-Jun-99 +---------------------- + +1. Change pcretest's output for amount of store used to show just the code +space, because the remainder (the data block) varies in size between 32-bit and +64-bit systems. + +2. Added an extra argument to pcre_exec() to supply an offset in the subject to +start matching at. This allows lookbehinds to work when searching for multiple +occurrences in a string. + +3. Added additional options to pcretest for testing multiple occurrences: + + /+ outputs the rest of the string that follows a match + /g loops for multiple occurrences, using the new startoffset argument + /G loops for multiple occurrences by passing an incremented pointer + +4. PCRE wasn't doing the "first character" optimization for patterns starting +with \b or \B, though it was doing it for other lookbehind assertions. That is, +it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with +the letter 'x'. On long subject strings, this gives a significant speed-up. + + +Version 2.05 21-Apr-99 +---------------------- + +1. Changed the type of magic_number from int to long int so that it works +properly on 16-bit systems. + +2. Fixed a bug which caused patterns starting with .* not to work correctly +when the subject string contained newline characters. PCRE was assuming +anchoring for such patterns in all cases, which is not correct because .* will +not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if +DOTALL is set at top level; otherwise it knows that patterns starting with .* +must be retried after every newline in the subject. + + +Version 2.04 18-Feb-99 +---------------------- + +1. For parenthesized subpatterns with repeats whose minimum was zero, the +computation of the store needed to hold the pattern was incorrect (too large). +If such patterns were nested a few deep, this could multiply and become a real +problem. + +2. Added /M option to pcretest to show the memory requirement of a specific +pattern. Made -m a synonym of -s (which does this globally) for compatibility. + +3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being +compiled in such a way that the backtracking after subsequent failure was +pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of +((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size. + + +Version 2.03 02-Feb-99 +---------------------- + +1. Fixed typo and small mistake in man page. + +2. Added 4th condition (GPL supersedes if conflict) and created separate +LICENCE file containing the conditions. + +3. Updated pcretest so that patterns such as /abc\/def/ work like they do in +Perl, that is the internal \ allows the delimiter to be included in the +pattern. Locked out the use of \ as a delimiter. If \ immediately follows +the final delimiter, add \ to the end of the pattern (to test the error). + +4. Added the convenience functions for extracting substrings after a successful +match. Updated pcretest to make it able to test these functions. + + +Version 2.02 14-Jan-99 +---------------------- + +1. Initialized the working variables associated with each extraction so that +their saving and restoring doesn't refer to uninitialized store. + +2. Put dummy code into study.c in order to trick the optimizer of the IBM C +compiler for OS/2 into generating correct code. Apparently IBM isn't going to +fix the problem. + +3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution +calls, and wasn't printing the correct value for compiling calls. Increased the +default value of LOOPREPEAT, and the number of significant figures in the +times. + +4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT. + +5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid +a building problem on Windows NT with a FAT file system. + + +Version 2.01 21-Oct-98 +---------------------- + +1. Changed the API for pcre_compile() to allow for the provision of a pointer +to character tables built by pcre_maketables() in the current locale. If NULL +is passed, the default tables are used. + + +Version 2.00 24-Sep-98 +---------------------- + +1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable +it any more. + +2. Allow quantification of (?>) groups, and make it work correctly. + +3. The first character computation wasn't working for (?>) groups. + +4. Correct the implementation of \Z (it is permitted to match on the \n at the +end of the subject) and add 5.005's \z, which really does match only at the +very end of the subject. + +5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater. + +6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and +DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005 +localized options. All options to pcre_study() were also removed. + +7. Add other new features from 5.005: + + $(?<= positive lookbehind + $(?a*))*/ (a PCRE_EXTRA facility). + + +Version 1.00 18-Nov-97 +---------------------- + +1. Added compile-time macros to support systems such as SunOS4 which don't have +memmove() or strerror() but have other things that can be used instead. + +2. Arranged that "make clean" removes the executables. + + +Version 0.99 27-Oct-97 +---------------------- + +1. Fixed bug in code for optimizing classes with only one character. It was +initializing a 32-byte map regardless, which could cause it to run off the end +of the memory it had got. + +2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction. + + +Version 0.98 22-Oct-97 +---------------------- + +1. Fixed bug in code for handling temporary memory usage when there are more +back references than supplied space in the ovector. This could cause segfaults. + + +Version 0.97 21-Oct-97 +---------------------- + +1. Added the \X "cut" facility, conditional on PCRE_EXTRA. + +2. Optimized negated single characters not to use a bit map. + +3. Brought error texts together as macro definitions; clarified some of them; +fixed one that was wrong - it said "range out of order" when it meant "invalid +escape sequence". + +4. Changed some char * arguments to const char *. + +5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX). + +6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in +pcretest. + + +Version 0.96 16-Oct-97 +---------------------- + +1. Added a simple "pgrep" utility to the distribution. + +2. Fixed an incompatibility with Perl: "{" is now treated as a normal character +unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}" +where "ddd" means "one or more decimal digits". + +3. Fixed serious bug. If a pattern had a back reference, but the call to +pcre_exec() didn't supply a large enough ovector to record the related +identifying subpattern, the match always failed. PCRE now remembers the number +of the largest back reference, and gets some temporary memory in which to save +the offsets during matching if necessary, in order to ensure that +backreferences always work. + +4. Increased the compatibility with Perl in a number of ways: + + (a) . no longer matches \n by default; an option PCRE_DOTALL is provided + to request this handling. The option can be set at compile or exec time. + + (b) $ matches before a terminating newline by default; an option + PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline + mode). The option can be set at compile or exec time. + + (c) The handling of \ followed by a digit other than 0 is now supposed to be + the same as Perl's. If the decimal number it represents is less than 10 + or there aren't that many previous left capturing parentheses, an octal + escape is read. Inside a character class, it's always an octal escape, + even if it is a single digit. + + (d) An escaped but undefined alphabetic character is taken as a literal, + unless PCRE_EXTRA is set. Currently this just reserves the remaining + escapes. + + (e) {0} is now permitted. (The previous item is removed from the compiled + pattern). + +5. Changed all the names of code files so that the basic parts are no longer +than 10 characters, and abolished the teeny "globals.c" file. + +6. Changed the handling of character classes; they are now done with a 32-byte +bit map always. + +7. Added the -d and /D options to pcretest to make it possible to look at the +internals of compilation without having to recompile pcre. + + +Version 0.95 23-Sep-97 +---------------------- + +1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or +\x20 at the start of a run of normal characters. These were being treated as +real characters, instead of the source characters being re-checked. + + +Version 0.94 18-Sep-97 +---------------------- + +1. The functions are now thread-safe, with the caveat that the global variables +containing pointers to malloc() and free() or alternative functions are the +same for all threads. + +2. Get pcre_study() to generate a bitmap of initial characters for non- +anchored patterns when this is possible, and use it if passed to pcre_exec(). + + +Version 0.93 15-Sep-97 +---------------------- + +1. /(b)|(:+)/ was computing an incorrect first character. + +2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(), +but not actually doing anything yet. + +3. Treat "-" characters in classes that cannot be part of ranges as literals, +as Perl does (e.g. [-az] or [az-]). + +4. Set the anchored flag if a branch starts with .* or .*? because that tests +all possible positions. + +5. Split up into different modules to avoid including unneeded functions in a +compiled binary. However, compile and exec are still in one module. The "study" +function is split off. + +6. The character tables are now in a separate module whose source is generated +by an auxiliary program - but can then be edited by hand if required. There are +now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or +toupper() in the code. + +7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and +make them global. Abolish the function for setting them, as the caller can now +set them directly. + + +Version 0.92 11-Sep-97 +---------------------- + +1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character +(e.g. /a{1,3}/) was broken (I mis-optimized it). + +2. Caseless matching was not working in character classes if the characters in +the pattern were in upper case. + +3. Make ranges like [W-c] work in the same way as Perl for caseless matching. + +4. Make PCRE_ANCHORED public and accept as a compile option. + +5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and +PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to +pass them. + +6. Give an error if bad option bits passed at compile or run time. + +7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to +pcretest to cause it to pass that flag. + +8. Add pcre_info(), to get the number of identifying subpatterns, the stored +options, and the first character, if set. + +9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character. + + +Version 0.91 10-Sep-97 +---------------------- + +1. PCRE was failing to diagnose unlimited repeats of subpatterns that could +match the empty string as in /(a*)*/. It was looping and ultimately crashing. + +2. PCRE was looping on encountering an indefinitely repeated back reference to +a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what +Perl does - treats the match as successful. + +**** diff --git a/pcre/HACKING b/pcre/HACKING new file mode 100644 index 00000000000..a90ddf87975 --- /dev/null +++ b/pcre/HACKING @@ -0,0 +1,473 @@ +Technical Notes about PCRE +-------------------------- + +These are very rough technical notes that record potentially useful information +about PCRE internals. For information about testing PCRE, see the pcretest +documentation and the comment at the head of the RunTest file. + + +Historical note 1 +----------------- + +Many years ago I implemented some regular expression functions to an algorithm +suggested by Martin Richards. These were not Unix-like in form, and were quite +restricted in what they could do by comparison with Perl. The interesting part +about the algorithm was that the amount of space required to hold the compiled +form of an expression was known in advance. The code to apply an expression did +not operate by backtracking, as the original Henry Spencer code and current +Perl code does, but instead checked all possibilities simultaneously by keeping +a list of current states and checking all of them as it advanced through the +subject string. In the terminology of Jeffrey Friedl's book, it was a "DFA +algorithm", though it was not a traditional Finite State Machine (FSM). When +the pattern was all used up, all remaining states were possible matches, and +the one matching the longest subset of the subject string was chosen. This did +not necessarily maximize the individual wild portions of the pattern, as is +expected in Unix and Perl-style regular expressions. + + +Historical note 2 +----------------- + +By contrast, the code originally written by Henry Spencer (which was +subsequently heavily modified for Perl) compiles the expression twice: once in +a dummy mode in order to find out how much store will be needed, and then for +real. (The Perl version probably doesn't do this any more; I'm talking about +the original library.) The execution function operates by backtracking and +maximizing (or, optionally, minimizing in Perl) the amount of the subject that +matches individual wild portions of the pattern. This is an "NFA algorithm" in +Friedl's terminology. + + +OK, here's the real stuff +------------------------- + +For the set of functions that form the "basic" PCRE library (which are +unrelated to those mentioned above), I tried at first to invent an algorithm +that used an amount of store bounded by a multiple of the number of characters +in the pattern, to save on compiling time. However, because of the greater +complexity in Perl regular expressions, I couldn't do this. In any case, a +first pass through the pattern is helpful for other reasons. + + +Support for 16-bit and 32-bit data strings +------------------------------------------- + +From release 8.30, PCRE supports 16-bit as well as 8-bit data strings; and from +release 8.32, PCRE supports 32-bit data strings. The library can be compiled +in any combination of 8-bit, 16-bit or 32-bit modes, creating different +libraries. In the description that follows, the word "short" is +used for a 16-bit data quantity, and the word "unit" is used for a quantity +that is a byte in 8-bit mode, a short in 16-bit mode and a 32-bit unsigned +integer in 32-bit mode. However, so as not to over-complicate the text, the +names of PCRE functions are given in 8-bit form only. + + +Computing the memory requirement: how it was +-------------------------------------------- + +Up to and including release 6.7, PCRE worked by running a very degenerate first +pass to calculate a maximum store size, and then a second pass to do the real +compile - which might use a bit less than the predicted amount of memory. The +idea was that this would turn out faster than the Henry Spencer code because +the first pass is degenerate and the second pass can just store stuff straight +into the vector, which it knows is big enough. + + +Computing the memory requirement: how it is +------------------------------------------- + +By the time I was working on a potential 6.8 release, the degenerate first pass +had become very complicated and hard to maintain. Indeed one of the early +things I did for 6.8 was to fix Yet Another Bug in the memory computation. Then +I had a flash of inspiration as to how I could run the real compile function in +a "fake" mode that enables it to compute how much memory it would need, while +actually only ever using a few hundred bytes of working memory, and without too +many tests of the mode that might slow it down. So I refactored the compiling +functions to work this way. This got rid of about 600 lines of source. It +should make future maintenance and development easier. As this was such a major +change, I never released 6.8, instead upping the number to 7.0 (other quite +major changes were also present in the 7.0 release). + +A side effect of this work was that the previous limit of 200 on the nesting +depth of parentheses was removed. However, there is a downside: pcre_compile() +runs more slowly than before (30% or more, depending on the pattern) because it +is doing a full analysis of the pattern. My hope was that this would not be a +big issue, and in the event, nobody has commented on it. + + +Traditional matching function +----------------------------- + +The "traditional", and original, matching function is called pcre_exec(), and +it implements an NFA algorithm, similar to the original Henry Spencer algorithm +and the way that Perl works. This is not surprising, since it is intended to be +as compatible with Perl as possible. This is the function most users of PCRE +will use most of the time. From release 8.20, if PCRE is compiled with +just-in-time (JIT) support, and studying a compiled pattern with JIT is +successful, the JIT code is run instead of the normal pcre_exec() code, but the +result is the same. + + +Supplementary matching function +------------------------------- + +From PCRE 6.0, there is also a supplementary matching function called +pcre_dfa_exec(). This implements a DFA matching algorithm that searches +simultaneously for all possible matches that start at one point in the subject +string. (Going back to my roots: see Historical Note 1 above.) This function +intreprets the same compiled pattern data as pcre_exec(); however, not all the +facilities are available, and those that are do not always work in quite the +same way. See the user documentation for details. + +The algorithm that is used for pcre_dfa_exec() is not a traditional FSM, +because it may have a number of states active at one time. More work would be +needed at compile time to produce a traditional FSM where only one state is +ever active at once. I believe some other regex matchers work this way. + + +Changeable options +------------------ + +The /i, /m, or /s options (PCRE_CASELESS, PCRE_MULTILINE, PCRE_DOTALL) may +change in the middle of patterns. From PCRE 8.13, their processing is handled +entirely at compile time by generating different opcodes for the different +settings. The runtime functions do not need to keep track of an options state +any more. + + +Format of compiled patterns +--------------------------- + +The compiled form of a pattern is a vector of units (bytes in 8-bit mode, or +shorts in 16-bit mode, 32-bit unsigned integers in 32-bit mode), containing +items of variable length. The first unit in an item contains an opcode, and +the length of the item is either implicit in the opcode or contained in the +data that follows it. + +In many cases listed below, LINK_SIZE data values are specified for offsets +within the compiled pattern. LINK_SIZE always specifies a number of bytes. The +default value for LINK_SIZE is 2, but PCRE can be compiled to use 3-byte or +4-byte values for these offsets, although this impairs the performance. (3-byte +LINK_SIZE values are available only in 8-bit mode.) Specifing a LINK_SIZE +larger than 2 is necessary only when patterns whose compiled length is greater +than 64K are going to be processed. In this description, we assume the "normal" +compilation options. Data values that are counts (e.g. for quantifiers) are +always just two bytes long (one short in 16-bit mode). + +Opcodes with no following data +------------------------------ + +These items are all just one unit long + + OP_END end of pattern + OP_ANY match any one character other than newline + OP_ALLANY match any one character, including newline + OP_ANYBYTE match any single byte, even in UTF-8 mode + OP_SOD match start of data: \A + OP_SOM, start of match (subject + offset): \G + OP_SET_SOM, set start of match (\K) + OP_CIRC ^ (start of data) + OP_CIRCM ^ multiline mode (start of data or after newline) + OP_NOT_WORD_BOUNDARY \W + OP_WORD_BOUNDARY \w + OP_NOT_DIGIT \D + OP_DIGIT \d + OP_NOT_HSPACE \H + OP_HSPACE \h + OP_NOT_WHITESPACE \S + OP_WHITESPACE \s + OP_NOT_VSPACE \V + OP_VSPACE \v + OP_NOT_WORDCHAR \W + OP_WORDCHAR \w + OP_EODN match end of data or \n at end: \Z + OP_EOD match end of data: \z + OP_DOLL $ (end of data, or before final newline) + OP_DOLLM $ multiline mode (end of data or before newline) + OP_EXTUNI match an extended Unicode character + OP_ANYNL match any Unicode newline sequence + + OP_ACCEPT ) These are Perl 5.10's "backtracking control + OP_COMMIT ) verbs". If OP_ACCEPT is inside capturing + OP_FAIL ) parentheses, it may be preceded by one or more + OP_PRUNE ) OP_CLOSE, followed by a 2-byte number, + OP_SKIP ) indicating which parentheses must be closed. + + +Backtracking control verbs with (optional) data +----------------------------------------------- + +(*THEN) without an argument generates the opcode OP_THEN and no following data. +OP_MARK is followed by the mark name, preceded by a one-unit length, and +followed by a binary zero. For (*PRUNE), (*SKIP), and (*THEN) with arguments, +the opcodes OP_PRUNE_ARG, OP_SKIP_ARG, and OP_THEN_ARG are used, with the name +following in the same format. + + +Matching literal characters +--------------------------- + +The OP_CHAR opcode is followed by a single character that is to be matched +casefully. For caseless matching, OP_CHARI is used. In UTF-8 or UTF-16 modes, +the character may be more than one unit long. In UTF-32 mode, characters +are always exactly one unit long. + + +Repeating single characters +--------------------------- + +The common repeats (*, +, ?), when applied to a single character, use the +following opcodes, which come in caseful and caseless versions: + + Caseful Caseless + OP_STAR OP_STARI + OP_MINSTAR OP_MINSTARI + OP_POSSTAR OP_POSSTARI + OP_PLUS OP_PLUSI + OP_MINPLUS OP_MINPLUSI + OP_POSPLUS OP_POSPLUSI + OP_QUERY OP_QUERYI + OP_MINQUERY OP_MINQUERYI + OP_POSQUERY OP_POSQUERYI + +Each opcode is followed by the character that is to be repeated. In ASCII mode, +these are two-unit items; in UTF-8 or UTF-16 modes, the length is variable; in +UTF-32 mode these are one-unit items. +Those with "MIN" in their names are the minimizing versions. Those with "POS" +in their names are possessive versions. Other repeats make use of these +opcodes: + + Caseful Caseless + OP_UPTO OP_UPTOI + OP_MINUPTO OP_MINUPTOI + OP_POSUPTO OP_POSUPTOI + OP_EXACT OP_EXACTI + +Each of these is followed by a two-byte (one short) count (most significant +byte first in 8-bit mode) and then the repeated character. OP_UPTO matches from +0 to the given number. A repeat with a non-zero minimum and a fixed maximum is +coded as an OP_EXACT followed by an OP_UPTO (or OP_MINUPTO or OPT_POSUPTO). + + +Repeating character types +------------------------- + +Repeats of things like \d are done exactly as for single characters, except +that instead of a character, the opcode for the type is stored in the data +unit. The opcodes are: + + OP_TYPESTAR + OP_TYPEMINSTAR + OP_TYPEPOSSTAR + OP_TYPEPLUS + OP_TYPEMINPLUS + OP_TYPEPOSPLUS + OP_TYPEQUERY + OP_TYPEMINQUERY + OP_TYPEPOSQUERY + OP_TYPEUPTO + OP_TYPEMINUPTO + OP_TYPEPOSUPTO + OP_TYPEEXACT + + +Match by Unicode property +------------------------- + +OP_PROP and OP_NOTPROP are used for positive and negative matches of a +character by testing its Unicode property (the \p and \P escape sequences). +Each is followed by two units that encode the desired property as a type and a +value. + +Repeats of these items use the OP_TYPESTAR etc. set of opcodes, followed by +three units: OP_PROP or OP_NOTPROP, and then the desired property type and +value. + + +Character classes +----------------- + +If there is only one character in the class, OP_CHAR or OP_CHARI is used for a +positive class, and OP_NOT or OP_NOTI for a negative one (that is, for +something like [^a]). + +Another set of 13 repeating opcodes (called OP_NOTSTAR etc.) are used for +repeated, negated, single-character classes. The normal single-character +opcodes (OP_STAR, etc.) are used for repeated positive single-character +classes. + +When there is more than one character in a class and all the characters are +less than 256, OP_CLASS is used for a positive class, and OP_NCLASS for a +negative one. In either case, the opcode is followed by a 32-byte (16-short) +bit map containing a 1 bit for every character that is acceptable. The bits are +counted from the least significant end of each unit. In caseless mode, bits for +both cases are set. + +The reason for having both OP_CLASS and OP_NCLASS is so that, in UTF-8/16/32 mode, +subject characters with values greater than 255 can be handled correctly. For +OP_CLASS they do not match, whereas for OP_NCLASS they do. + +For classes containing characters with values greater than 255, OP_XCLASS is +used. It optionally uses a bit map (if any characters lie within it), followed +by a list of pairs (for a range) and single characters. In caseless mode, both +cases are explicitly listed. There is a flag character than indicates whether +it is a positive or a negative class. + + +Back references +--------------- + +OP_REF (caseful) or OP_REFI (caseless) is followed by two bytes (one short) +containing the reference number. + + +Repeating character classes and back references +----------------------------------------------- + +Single-character classes are handled specially (see above). This section +applies to OP_CLASS and OP_REF[I]. In both cases, the repeat information +follows the base item. The matching code looks at the following opcode to see +if it is one of + + OP_CRSTAR + OP_CRMINSTAR + OP_CRPLUS + OP_CRMINPLUS + OP_CRQUERY + OP_CRMINQUERY + OP_CRRANGE + OP_CRMINRANGE + +All but the last two are just single-unit items. The others are followed by +four bytes (two shorts) of data, comprising the minimum and maximum repeat +counts. There are no special possessive opcodes for these repeats; a possessive +repeat is compiled into an atomic group. + + +Brackets and alternation +------------------------ + +A pair of non-capturing (round) brackets is wrapped round each expression at +compile time, so alternation always happens in the context of brackets. + +[Note for North Americans: "bracket" to some English speakers, including +myself, can be round, square, curly, or pointy. Hence this usage rather than +"parentheses".] + +Non-capturing brackets use the opcode OP_BRA. Originally PCRE was limited to 99 +capturing brackets and it used a different opcode for each one. From release +3.5, the limit was removed by putting the bracket number into the data for +higher-numbered brackets. From release 7.0 all capturing brackets are handled +this way, using the single opcode OP_CBRA. + +A bracket opcode is followed by LINK_SIZE bytes which give the offset to the +next alternative OP_ALT or, if there aren't any branches, to the matching +OP_KET opcode. Each OP_ALT is followed by LINK_SIZE bytes giving the offset to +the next one, or to the OP_KET opcode. For capturing brackets, the bracket +number immediately follows the offset, always as a 2-byte (one short) item. + +OP_KET is used for subpatterns that do not repeat indefinitely, and +OP_KETRMIN and OP_KETRMAX are used for indefinite repetitions, minimally or +maximally respectively (see below for possessive repetitions). All three are +followed by LINK_SIZE bytes giving (as a positive number) the offset back to +the matching bracket opcode. + +If a subpattern is quantified such that it is permitted to match zero times, it +is preceded by one of OP_BRAZERO, OP_BRAMINZERO, or OP_SKIPZERO. These are +single-unit opcodes that tell the matcher that skipping the following +subpattern entirely is a valid branch. In the case of the first two, not +skipping the pattern is also valid (greedy and non-greedy). The third is used +when a pattern has the quantifier {0,0}. It cannot be entirely discarded, +because it may be called as a subroutine from elsewhere in the regex. + +A subpattern with an indefinite maximum repetition is replicated in the +compiled data its minimum number of times (or once with OP_BRAZERO if the +minimum is zero), with the final copy terminating with OP_KETRMIN or OP_KETRMAX +as appropriate. + +A subpattern with a bounded maximum repetition is replicated in a nested +fashion up to the maximum number of times, with OP_BRAZERO or OP_BRAMINZERO +before each replication after the minimum, so that, for example, (abc){2,5} is +compiled as (abc)(abc)((abc)((abc)(abc)?)?)?, except that each bracketed group +has the same number. + +When a repeated subpattern has an unbounded upper limit, it is checked to see +whether it could match an empty string. If this is the case, the opcode in the +final replication is changed to OP_SBRA or OP_SCBRA. This tells the matcher +that it needs to check for matching an empty string when it hits OP_KETRMIN or +OP_KETRMAX, and if so, to break the loop. + +Possessive brackets +------------------- + +When a repeated group (capturing or non-capturing) is marked as possessive by +the "+" notation, e.g. (abc)++, different opcodes are used. Their names all +have POS on the end, e.g. OP_BRAPOS instead of OP_BRA and OP_SCPBRPOS instead +of OP_SCBRA. The end of such a group is marked by OP_KETRPOS. If the minimum +repetition is zero, the group is preceded by OP_BRAPOSZERO. + + +Assertions +---------- + +Forward assertions are just like other subpatterns, but starting with one of +the opcodes OP_ASSERT or OP_ASSERT_NOT. Backward assertions use the opcodes +OP_ASSERTBACK and OP_ASSERTBACK_NOT, and the first opcode inside the assertion +is OP_REVERSE, followed by a two byte (one short) count of the number of +characters to move back the pointer in the subject string. In ASCII mode, the +count is a number of units, but in UTF-8/16 mode each character may occupy more +than one unit; in UTF-32 mode each character occupies exactly one unit. +A separate count is present in each alternative of a lookbehind +assertion, allowing them to have different fixed lengths. + + +Once-only (atomic) subpatterns +------------------------------ + +These are also just like other subpatterns, but they start with the opcode +OP_ONCE. The check for matching an empty string in an unbounded repeat is +handled entirely at runtime, so there is just this one opcode. + + +Conditional subpatterns +----------------------- + +These are like other subpatterns, but they start with the opcode OP_COND, or +OP_SCOND for one that might match an empty string in an unbounded repeat. If +the condition is a back reference, this is stored at the start of the +subpattern using the opcode OP_CREF followed by two bytes (one short) +containing the reference number. OP_NCREF is used instead if the reference was +generated by name (so that the runtime code knows to check for duplicate +names). + +If the condition is "in recursion" (coded as "(?(R)"), or "in recursion of +group x" (coded as "(?(Rx)"), the group number is stored at the start of the +subpattern using the opcode OP_RREF or OP_NRREF (cf OP_NCREF), and a value of +zero for "the whole pattern". For a DEFINE condition, just the single unit +OP_DEF is used (it has no associated data). Otherwise, a conditional subpattern +always starts with one of the assertions. + + +Recursion +--------- + +Recursion either matches the current regex, or some subexpression. The opcode +OP_RECURSE is followed by an value which is the offset to the starting bracket +from the start of the whole pattern. From release 6.5, OP_RECURSE is +automatically wrapped inside OP_ONCE brackets (because otherwise some patterns +broke it). OP_RECURSE is also used for "subroutine" calls, even though they +are not strictly a recursion. + + +Callout +------- + +OP_CALLOUT is followed by one unit of data that holds a callout number in the +range 0 to 254 for manual callouts, or 255 for an automatic callout. In both +cases there follows a two-byte (one short) value giving the offset in the +pattern to the start of the following item, and another two-byte (one short) +item giving the length of the next item. + + +Philip Hazel +February 2012 diff --git a/pcre/INSTALL b/pcre/INSTALL new file mode 100644 index 00000000000..007e9396d0a --- /dev/null +++ b/pcre/INSTALL @@ -0,0 +1,370 @@ +Installation Instructions +************************* + +Copyright (C) 1994-1996, 1999-2002, 2004-2013 Free Software Foundation, +Inc. + + Copying and distribution of this file, with or without modification, +are permitted in any medium without royalty provided the copyright +notice and this notice are preserved. This file is offered as-is, +without warranty of any kind. + +Basic Installation +================== + + Briefly, the shell commands `./configure; make; make install' should +configure, build, and install this package. The following +more-detailed instructions are generic; see the `README' file for +instructions specific to this package. Some packages provide this +`INSTALL' file but do not implement all of the features documented +below. The lack of an optional feature in a given package is not +necessarily a bug. More recommendations for GNU packages can be found +in *note Makefile Conventions: (standards)Makefile Conventions. + + The `configure' shell script attempts to guess correct values for +various system-dependent variables used during compilation. It uses +those values to create a `Makefile' in each directory of the package. +It may also create one or more `.h' files containing system-dependent +definitions. Finally, it creates a shell script `config.status' that +you can run in the future to recreate the current configuration, and a +file `config.log' containing compiler output (useful mainly for +debugging `configure'). + + It can also use an optional file (typically called `config.cache' +and enabled with `--cache-file=config.cache' or simply `-C') that saves +the results of its tests to speed up reconfiguring. Caching is +disabled by default to prevent problems with accidental use of stale +cache files. + + If you need to do unusual things to compile the package, please try +to figure out how `configure' could check whether to do them, and mail +diffs or instructions to the address given in the `README' so they can +be considered for the next release. If you are using the cache, and at +some point `config.cache' contains results you don't want to keep, you +may remove or edit it. + + The file `configure.ac' (or `configure.in') is used to create +`configure' by a program called `autoconf'. You need `configure.ac' if +you want to change it or regenerate `configure' using a newer version +of `autoconf'. + + The simplest way to compile this package is: + + 1. `cd' to the directory containing the package's source code and type + `./configure' to configure the package for your system. + + Running `configure' might take a while. While running, it prints + some messages telling which features it is checking for. + + 2. Type `make' to compile the package. + + 3. Optionally, type `make check' to run any self-tests that come with + the package, generally using the just-built uninstalled binaries. + + 4. Type `make install' to install the programs and any data files and + documentation. When installing into a prefix owned by root, it is + recommended that the package be configured and built as a regular + user, and only the `make install' phase executed with root + privileges. + + 5. Optionally, type `make installcheck' to repeat any self-tests, but + this time using the binaries in their final installed location. + This target does not install anything. Running this target as a + regular user, particularly if the prior `make install' required + root privileges, verifies that the installation completed + correctly. + + 6. You can remove the program binaries and object files from the + source code directory by typing `make clean'. To also remove the + files that `configure' created (so you can compile the package for + a different kind of computer), type `make distclean'. There is + also a `make maintainer-clean' target, but that is intended mainly + for the package's developers. If you use it, you may have to get + all sorts of other programs in order to regenerate files that came + with the distribution. + + 7. Often, you can also type `make uninstall' to remove the installed + files again. In practice, not all packages have tested that + uninstallation works correctly, even though it is required by the + GNU Coding Standards. + + 8. Some packages, particularly those that use Automake, provide `make + distcheck', which can by used by developers to test that all other + targets like `make install' and `make uninstall' work correctly. + This target is generally not run by end users. + +Compilers and Options +===================== + + Some systems require unusual options for compilation or linking that +the `configure' script does not know about. Run `./configure --help' +for details on some of the pertinent environment variables. + + You can give `configure' initial values for configuration parameters +by setting variables in the command line or in the environment. Here +is an example: + + ./configure CC=c99 CFLAGS=-g LIBS=-lposix + + *Note Defining Variables::, for more details. + +Compiling For Multiple Architectures +==================================== + + You can compile the package for more than one kind of computer at the +same time, by placing the object files for each architecture in their +own directory. To do this, you can use GNU `make'. `cd' to the +directory where you want the object files and executables to go and run +the `configure' script. `configure' automatically checks for the +source code in the directory that `configure' is in and in `..'. This +is known as a "VPATH" build. + + With a non-GNU `make', it is safer to compile the package for one +architecture at a time in the source code directory. After you have +installed the package for one architecture, use `make distclean' before +reconfiguring for another architecture. + + On MacOS X 10.5 and later systems, you can create libraries and +executables that work on multiple system types--known as "fat" or +"universal" binaries--by specifying multiple `-arch' options to the +compiler but only a single `-arch' option to the preprocessor. Like +this: + + ./configure CC="gcc -arch i386 -arch x86_64 -arch ppc -arch ppc64" \ + CXX="g++ -arch i386 -arch x86_64 -arch ppc -arch ppc64" \ + CPP="gcc -E" CXXCPP="g++ -E" + + This is not guaranteed to produce working output in all cases, you +may have to build one architecture at a time and combine the results +using the `lipo' tool if you have problems. + +Installation Names +================== + + By default, `make install' installs the package's commands under +`/usr/local/bin', include files under `/usr/local/include', etc. You +can specify an installation prefix other than `/usr/local' by giving +`configure' the option `--prefix=PREFIX', where PREFIX must be an +absolute file name. + + You can specify separate installation prefixes for +architecture-specific files and architecture-independent files. If you +pass the option `--exec-prefix=PREFIX' to `configure', the package uses +PREFIX as the prefix for installing programs and libraries. +Documentation and other data files still use the regular prefix. + + In addition, if you use an unusual directory layout you can give +options like `--bindir=DIR' to specify different values for particular +kinds of files. Run `configure --help' for a list of the directories +you can set and what kinds of files go in them. In general, the +default for these options is expressed in terms of `${prefix}', so that +specifying just `--prefix' will affect all of the other directory +specifications that were not explicitly provided. + + The most portable way to affect installation locations is to pass the +correct locations to `configure'; however, many packages provide one or +both of the following shortcuts of passing variable assignments to the +`make install' command line to change installation locations without +having to reconfigure or recompile. + + The first method involves providing an override variable for each +affected directory. For example, `make install +prefix=/alternate/directory' will choose an alternate location for all +directory configuration variables that were expressed in terms of +`${prefix}'. Any directories that were specified during `configure', +but not in terms of `${prefix}', must each be overridden at install +time for the entire installation to be relocated. The approach of +makefile variable overrides for each directory variable is required by +the GNU Coding Standards, and ideally causes no recompilation. +However, some platforms have known limitations with the semantics of +shared libraries that end up requiring recompilation when using this +method, particularly noticeable in packages that use GNU Libtool. + + The second method involves providing the `DESTDIR' variable. For +example, `make install DESTDIR=/alternate/directory' will prepend +`/alternate/directory' before all installation names. The approach of +`DESTDIR' overrides is not required by the GNU Coding Standards, and +does not work on platforms that have drive letters. On the other hand, +it does better at avoiding recompilation issues, and works well even +when some directory options were not specified in terms of `${prefix}' +at `configure' time. + +Optional Features +================= + + If the package supports it, you can cause programs to be installed +with an extra prefix or suffix on their names by giving `configure' the +option `--program-prefix=PREFIX' or `--program-suffix=SUFFIX'. + + Some packages pay attention to `--enable-FEATURE' options to +`configure', where FEATURE indicates an optional part of the package. +They may also pay attention to `--with-PACKAGE' options, where PACKAGE +is something like `gnu-as' or `x' (for the X Window System). The +`README' should mention any `--enable-' and `--with-' options that the +package recognizes. + + For packages that use the X Window System, `configure' can usually +find the X include and library files automatically, but if it doesn't, +you can use the `configure' options `--x-includes=DIR' and +`--x-libraries=DIR' to specify their locations. + + Some packages offer the ability to configure how verbose the +execution of `make' will be. For these packages, running `./configure +--enable-silent-rules' sets the default to minimal output, which can be +overridden with `make V=1'; while running `./configure +--disable-silent-rules' sets the default to verbose, which can be +overridden with `make V=0'. + +Particular systems +================== + + On HP-UX, the default C compiler is not ANSI C compatible. If GNU +CC is not installed, it is recommended to use the following options in +order to use an ANSI C compiler: + + ./configure CC="cc -Ae -D_XOPEN_SOURCE=500" + +and if that doesn't work, install pre-built binaries of GCC for HP-UX. + + HP-UX `make' updates targets which have the same time stamps as +their prerequisites, which makes it generally unusable when shipped +generated files such as `configure' are involved. Use GNU `make' +instead. + + On OSF/1 a.k.a. Tru64, some versions of the default C compiler cannot +parse its `' header file. The option `-nodtk' can be used as +a workaround. If GNU CC is not installed, it is therefore recommended +to try + + ./configure CC="cc" + +and if that doesn't work, try + + ./configure CC="cc -nodtk" + + On Solaris, don't put `/usr/ucb' early in your `PATH'. This +directory contains several dysfunctional programs; working variants of +these programs are available in `/usr/bin'. So, if you need `/usr/ucb' +in your `PATH', put it _after_ `/usr/bin'. + + On Haiku, software installed for all users goes in `/boot/common', +not `/usr/local'. It is recommended to use the following options: + + ./configure --prefix=/boot/common + +Specifying the System Type +========================== + + There may be some features `configure' cannot figure out +automatically, but needs to determine by the type of machine the package +will run on. Usually, assuming the package is built to be run on the +_same_ architectures, `configure' can figure that out, but if it prints +a message saying it cannot guess the machine type, give it the +`--build=TYPE' option. TYPE can either be a short name for the system +type, such as `sun4', or a canonical name which has the form: + + CPU-COMPANY-SYSTEM + +where SYSTEM can have one of these forms: + + OS + KERNEL-OS + + See the file `config.sub' for the possible values of each field. If +`config.sub' isn't included in this package, then this package doesn't +need to know the machine type. + + If you are _building_ compiler tools for cross-compiling, you should +use the option `--target=TYPE' to select the type of system they will +produce code for. + + If you want to _use_ a cross compiler, that generates code for a +platform different from the build platform, you should specify the +"host" platform (i.e., that on which the generated programs will +eventually be run) with `--host=TYPE'. + +Sharing Defaults +================ + + If you want to set default values for `configure' scripts to share, +you can create a site shell script called `config.site' that gives +default values for variables like `CC', `cache_file', and `prefix'. +`configure' looks for `PREFIX/share/config.site' if it exists, then +`PREFIX/etc/config.site' if it exists. Or, you can set the +`CONFIG_SITE' environment variable to the location of the site script. +A warning: not all `configure' scripts look for a site script. + +Defining Variables +================== + + Variables not defined in a site shell script can be set in the +environment passed to `configure'. However, some packages may run +configure again during the build, and the customized values of these +variables may be lost. In order to avoid this problem, you should set +them in the `configure' command line, using `VAR=value'. For example: + + ./configure CC=/usr/local2/bin/gcc + +causes the specified `gcc' to be used as the C compiler (unless it is +overridden in the site shell script). + +Unfortunately, this technique does not work for `CONFIG_SHELL' due to +an Autoconf limitation. Until the limitation is lifted, you can use +this workaround: + + CONFIG_SHELL=/bin/bash ./configure CONFIG_SHELL=/bin/bash + +`configure' Invocation +====================== + + `configure' recognizes the following options to control how it +operates. + +`--help' +`-h' + Print a summary of all of the options to `configure', and exit. + +`--help=short' +`--help=recursive' + Print a summary of the options unique to this package's + `configure', and exit. The `short' variant lists options used + only in the top level, while the `recursive' variant lists options + also present in any nested packages. + +`--version' +`-V' + Print the version of Autoconf used to generate the `configure' + script, and exit. + +`--cache-file=FILE' + Enable the cache: use and save the results of the tests in FILE, + traditionally `config.cache'. FILE defaults to `/dev/null' to + disable caching. + +`--config-cache' +`-C' + Alias for `--cache-file=config.cache'. + +`--quiet' +`--silent' +`-q' + Do not print messages saying which checks are being made. To + suppress all normal output, redirect it to `/dev/null' (any error + messages will still be shown). + +`--srcdir=DIR' + Look for the package's source code in directory DIR. Usually + `configure' can determine that directory automatically. + +`--prefix=DIR' + Use DIR as the installation prefix. *note Installation Names:: + for more details, including other options available for fine-tuning + the installation locations. + +`--no-create' +`-n' + Run the configure checks, but stop before creating any output + files. + +`configure' also accepts some other, not widely useful, options. Run +`configure --help' for more details. diff --git a/pcre/LICENCE b/pcre/LICENCE new file mode 100644 index 00000000000..3aff6a62c00 --- /dev/null +++ b/pcre/LICENCE @@ -0,0 +1,92 @@ +PCRE LICENCE +------------ + +PCRE is a library of functions to support regular expressions whose syntax +and semantics are as close as possible to those of the Perl 5 language. + +Release 8 of PCRE is distributed under the terms of the "BSD" licence, as +specified below. The documentation for PCRE, supplied in the "doc" +directory, is distributed under the same terms as the software itself. + +The basic library functions are written in C and are freestanding. Also +included in the distribution is a set of C++ wrapper functions, and a +just-in-time compiler that can be used to optimize pattern matching. These +are both optional features that can be omitted when the library is built. + + +THE BASIC LIBRARY FUNCTIONS +--------------------------- + +Written by: Philip Hazel +Email local part: ph10 +Email domain: cam.ac.uk + +University of Cambridge Computing Service, +Cambridge, England. + +Copyright (c) 1997-2013 University of Cambridge +All rights reserved. + + +PCRE JUST-IN-TIME COMPILATION SUPPORT +------------------------------------- + +Written by: Zoltan Herczeg +Email local part: hzmester +Emain domain: freemail.hu + +Copyright(c) 2010-2013 Zoltan Herczeg +All rights reserved. + + +STACK-LESS JUST-IN-TIME COMPILER +-------------------------------- + +Written by: Zoltan Herczeg +Email local part: hzmester +Emain domain: freemail.hu + +Copyright(c) 2009-2013 Zoltan Herczeg +All rights reserved. + + +THE C++ WRAPPER FUNCTIONS +------------------------- + +Contributed by: Google Inc. + +Copyright (c) 2007-2012, Google Inc. +All rights reserved. + + +THE "BSD" LICENCE +----------------- + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + + * Redistributions of source code must retain the above copyright notice, + this list of conditions and the following disclaimer. + + * Redistributions in binary form must reproduce the above copyright + notice, this list of conditions and the following disclaimer in the + documentation and/or other materials provided with the distribution. + + * Neither the name of the University of Cambridge nor the name of Google + Inc. nor the names of their contributors may be used to endorse or + promote products derived from this software without specific prior + written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE +ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE +LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR +CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF +SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS +INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN +CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) +ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE +POSSIBILITY OF SUCH DAMAGE. + +End diff --git a/pcre/NEWS b/pcre/NEWS new file mode 100644 index 00000000000..2ff69cbdd78 --- /dev/null +++ b/pcre/NEWS @@ -0,0 +1,635 @@ +News about PCRE releases +------------------------ + +Release 8.33 28-May-2013 +-------------------------- + +A number of bugs are fixed, and some performance improvements have been made. +There are also some new features, of which these are the most important: + +. The behaviour of the backtracking verbs has been rationalized and + documented in more detail. + +. JIT now supports callouts and all of the backtracking verbs. + +. Unicode validation has been updated in the light of Unicode Corrigendum #9, + which points out that "non characters" are not "characters that may not + appear in Unicode strings" but rather "characters that are reserved for + internal use and have only local meaning". + +. (*LIMIT_MATCH=d) and (*LIMIT_RECURSION=d) have been added so that the + creator of a pattern can specify lower (but not higher) limits for the + matching process. + +. The PCRE_NEVER_UTF option is available to prevent pattern-writers from using + the (*UTF) feature, as this could be a security issue. + + +Release 8.32 30-November-2012 +----------------------------- + +This release fixes a number of bugs, but also has some new features. These are +the highlights: + +. There is now support for 32-bit character strings and UTF-32. Like the + 16-bit support, this is done by compiling a separate 32-bit library. + +. \X now matches a Unicode extended grapheme cluster. + +. Case-independent matching of Unicode characters that have more than one + "other case" now makes all three (or more) characters equivalent. This + applies, for example, to Greek Sigma, which has two lowercase versions. + +. Unicode character properties are updated to Unicode 6.2.0. + +. The EBCDIC support, which had decayed, has had a spring clean. + +. A number of JIT optimizations have been added, which give faster JIT + execution speed. In addition, a new direct interface to JIT execution is + available. This bypasses some of the sanity checks of pcre_exec() to give a + noticeable speed-up. + +. A number of issues in pcregrep have been fixed, making it more compatible + with GNU grep. In particular, --exclude and --include (and variants) apply + to all files now, not just those obtained from scanning a directory + recursively. In Windows environments, the default action for directories is + now "skip" instead of "read" (which provokes an error). + +. If the --only-matching (-o) option in pcregrep is specified multiple + times, each one causes appropriate output. For example, -o1 -o2 outputs the + substrings matched by the 1st and 2nd capturing parentheses. A separating + string can be specified by --om-separator (default empty). + +. When PCRE is built via Autotools using a version of gcc that has the + "visibility" feature, it is used to hide internal library functions that are + not part of the public API. + + +Release 8.31 06-July-2012 +------------------------- + +This is mainly a bug-fixing release, with a small number of developments: + +. The JIT compiler now supports partial matching and the (*MARK) and + (*COMMIT) verbs. + +. PCRE_INFO_MAXLOOKBEHIND can be used to find the longest lookbehind in a + pattern. + +. There should be a performance improvement when using the heap instead of the + stack for recursion. + +. pcregrep can now be linked with libedit as an alternative to libreadline. + +. pcregrep now has a --file-list option where the list of files to scan is + given as a file. + +. pcregrep now recognizes binary files and there are related options. + +. The Unicode tables have been updated to 6.1.0. + +As always, the full list of changes is in the ChangeLog file. + + +Release 8.30 04-February-2012 +----------------------------- + +Release 8.30 introduces a major new feature: support for 16-bit character +strings, compiled as a separate library. There are a few changes to the +8-bit library, in addition to some bug fixes. + +. The pcre_info() function, which has been obsolete for over 10 years, has + been removed. + +. When a compiled pattern was saved to a file and later reloaded on a host + with different endianness, PCRE used automatically to swap the bytes in some + of the data fields. With the advent of the 16-bit library, where more of this + swapping is needed, it is no longer done automatically. Instead, the bad + endianness is detected and a specific error is given. The user can then call + a new function called pcre_pattern_to_host_byte_order() (or an equivalent + 16-bit function) to do the swap. + +. In UTF-8 mode, the values 0xd800 to 0xdfff are not legal Unicode + code points and are now faulted. (They are the so-called "surrogates" + that are reserved for coding high values in UTF-16.) + + +Release 8.21 12-Dec-2011 +------------------------ + +This is almost entirely a bug-fix release. The only new feature is the ability +to obtain the size of the memory used by the JIT compiler. + + +Release 8.20 21-Oct-2011 +------------------------ + +The main change in this release is the inclusion of Zoltan Herczeg's +just-in-time compiler support, which can be accessed by building PCRE with +--enable-jit. Large performance benefits can be had in many situations. 8.20 +also fixes an unfortunate bug that was introduced in 8.13 as well as tidying up +a number of infelicities and differences from Perl. + + +Release 8.13 16-Aug-2011 +------------------------ + +This is mainly a bug-fix release. There has been a lot of internal refactoring. +The Unicode tables have been updated. The only new feature in the library is +the passing of *MARK information to callouts. Some additions have been made to +pcretest to make testing easier and more comprehensive. There is a new option +for pcregrep to adjust its internal buffer size. + + +Release 8.12 15-Jan-2011 +------------------------ + +This release fixes some bugs in pcregrep, one of which caused the tests to fail +on 64-bit big-endian systems. There are no changes to the code of the library. + + +Release 8.11 10-Dec-2010 +------------------------ + +A number of bugs in the library and in pcregrep have been fixed. As always, see +ChangeLog for details. The following are the non-bug-fix changes: + +. Added --match-limit and --recursion-limit to pcregrep. + +. Added an optional parentheses number to the -o and --only-matching options + of pcregrep. + +. Changed the way PCRE_PARTIAL_HARD affects the matching of $, \z, \Z, \b, and + \B. + +. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a + bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD. + +. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_ + START_OPTIMIZE option, which is now allowed at compile time + + +Release 8.10 25-Jun-2010 +------------------------ + +There are two major additions: support for (*MARK) and friends, and the option +PCRE_UCP, which changes the behaviour of \b, \d, \s, and \w (and their +opposites) so that they make use of Unicode properties. There are also a number +of lesser new features, and several bugs have been fixed. A new option, +--line-buffered, has been added to pcregrep, for use when it is connected to +pipes. + + +Release 8.02 19-Mar-2010 +------------------------ + +Another bug-fix release. + + +Release 8.01 19-Jan-2010 +------------------------ + +This is a bug-fix release. Several bugs in the code itself and some bugs and +infelicities in the build system have been fixed. + + +Release 8.00 19-Oct-09 +---------------------- + +Bugs have been fixed in the library and in pcregrep. There are also some +enhancements. Restrictions on patterns used for partial matching have been +removed, extra information is given for partial matches, the partial matching +process has been improved, and an option to make a partial match override a +full match is available. The "study" process has been enhanced by finding a +lower bound matching length. Groups with duplicate numbers may now have +duplicated names without the use of PCRE_DUPNAMES. However, they may not have +different names. The documentation has been revised to reflect these changes. +The version number has been expanded to 3 digits as it is clear that the rate +of change is not slowing down. + + +Release 7.9 11-Apr-09 +--------------------- + +Mostly bugfixes and tidies with just a couple of minor functional additions. + + +Release 7.8 05-Sep-08 +--------------------- + +More bug fixes, plus a performance improvement in Unicode character property +lookup. + + +Release 7.7 07-May-08 +--------------------- + +This is once again mainly a bug-fix release, but there are a couple of new +features. + + +Release 7.6 28-Jan-08 +--------------------- + +The main reason for having this release so soon after 7.5 is because it fixes a +potential buffer overflow problem in pcre_compile() when run in UTF-8 mode. In +addition, the CMake configuration files have been brought up to date. + + +Release 7.5 10-Jan-08 +--------------------- + +This is mainly a bug-fix release. However the ability to link pcregrep with +libz or libbz2 and the ability to link pcretest with libreadline have been +added. Also the --line-offsets and --file-offsets options were added to +pcregrep. + + +Release 7.4 21-Sep-07 +--------------------- + +The only change of specification is the addition of options to control whether +\R matches any Unicode line ending (the default) or just CR, LF, and CRLF. +Otherwise, the changes are bug fixes and a refactoring to reduce the number of +relocations needed in a shared library. There have also been some documentation +updates, in particular, some more information about using CMake to build PCRE +has been added to the NON-UNIX-USE file. + + +Release 7.3 28-Aug-07 +--------------------- + +Most changes are bug fixes. Some that are not: + +1. There is some support for Perl 5.10's experimental "backtracking control + verbs" such as (*PRUNE). + +2. UTF-8 checking is now as per RFC 3629 instead of RFC 2279; this is more + restrictive in the strings it accepts. + +3. Checking for potential integer overflow has been made more dynamic, and as a + consequence there is no longer a hard limit on the size of a subpattern that + has a limited repeat count. + +4. When CRLF is a valid line-ending sequence, pcre_exec() and pcre_dfa_exec() + no longer advance by two characters instead of one when an unanchored match + fails at CRLF if there are explicit CR or LF matches within the pattern. + This gets rid of some anomalous effects that previously occurred. + +5. Some PCRE-specific settings for varying the newline options at the start of + a pattern have been added. + + +Release 7.2 19-Jun-07 +--------------------- + +WARNING: saved patterns that were compiled by earlier versions of PCRE must be +recompiled for use with 7.2 (necessitated by the addition of \K, \h, \H, \v, +and \V). + +Correction to the notes for 7.1: the note about shared libraries for Windows is +wrong. Previously, three libraries were built, but each could function +independently. For example, the pcreposix library also included all the +functions from the basic pcre library. The change is that the three libraries +are no longer independent. They are like the Unix libraries. To use the +pcreposix functions, for example, you need to link with both the pcreposix and +the basic pcre library. + +Some more features from Perl 5.10 have been added: + + (?-n) and (?+n) relative references for recursion and subroutines. + + (?(-n) and (?(+n) relative references as conditions. + + \k{name} and \g{name} are synonyms for \k. + + \K to reset the start of the matched string; for example, (foo)\Kbar + matches bar preceded by foo, but only sets bar as the matched string. + + (?| introduces a group where the capturing parentheses in each alternative + start from the same number; for example, (?|(abc)|(xyz)) sets capturing + parentheses number 1 in both cases. + + \h, \H, \v, \V match horizontal and vertical whitespace, respectively. + + +Release 7.1 24-Apr-07 +--------------------- + +There is only one new feature in this release: a linebreak setting of +PCRE_NEWLINE_ANYCRLF. It is a cut-down version of PCRE_NEWLINE_ANY, which +recognizes only CRLF, CR, and LF as linebreaks. + +A few bugs are fixed (see ChangeLog for details), but the major change is a +complete re-implementation of the build system. This now has full Autotools +support and so is now "standard" in some sense. It should help with compiling +PCRE in a wide variety of environments. + +NOTE: when building shared libraries for Windows, three dlls are now built, +called libpcre, libpcreposix, and libpcrecpp. Previously, everything was +included in a single dll. + +Another important change is that the dftables auxiliary program is no longer +compiled and run at "make" time by default. Instead, a default set of character +tables (assuming ASCII coding) is used. If you want to use dftables to generate +the character tables as previously, add --enable-rebuild-chartables to the +"configure" command. You must do this if you are compiling PCRE to run on a +system that uses EBCDIC code. + +There is a discussion about character tables in the README file. The default is +not to use dftables so that that there is no problem when cross-compiling. + + +Release 7.0 19-Dec-06 +--------------------- + +This release has a new major number because there have been some internal +upheavals to facilitate the addition of new optimizations and other facilities, +and to make subsequent maintenance and extension easier. Compilation is likely +to be a bit slower, but there should be no major effect on runtime performance. +Previously compiled patterns are NOT upwards compatible with this release. If +you have saved compiled patterns from a previous release, you will have to +re-compile them. Important changes that are visible to users are: + +1. The Unicode property tables have been updated to Unicode 5.0.0, which adds + some more scripts. + +2. The option PCRE_NEWLINE_ANY causes PCRE to recognize any Unicode newline + sequence as a newline. + +3. The \R escape matches a single Unicode newline sequence as a single unit. + +4. New features that will appear in Perl 5.10 are now in PCRE. These include + alternative Perl syntax for named parentheses, and Perl syntax for + recursion. + +5. The C++ wrapper interface has been extended by the addition of a + QuoteMeta function and the ability to allow copy construction and + assignment. + +For a complete list of changes, see the ChangeLog file. + + +Release 6.7 04-Jul-06 +--------------------- + +The main additions to this release are the ability to use the same name for +multiple sets of parentheses, and support for CRLF line endings in both the +library and pcregrep (and in pcretest for testing). + +Thanks to Ian Taylor, the stack usage for many kinds of pattern has been +significantly reduced for certain subject strings. + + +Release 6.5 01-Feb-06 +--------------------- + +Important changes in this release: + +1. A number of new features have been added to pcregrep. + +2. The Unicode property tables have been updated to Unicode 4.1.0, and the + supported properties have been extended with script names such as "Arabic", + and the derived properties "Any" and "L&". This has necessitated a change to + the interal format of compiled patterns. Any saved compiled patterns that + use \p or \P must be recompiled. + +3. The specification of recursion in patterns has been changed so that all + recursive subpatterns are automatically treated as atomic groups. Thus, for + example, (?R) is treated as if it were (?>(?R)). This is necessary because + otherwise there are situations where recursion does not work. + +See the ChangeLog for a complete list of changes, which include a number of bug +fixes and tidies. + + +Release 6.0 07-Jun-05 +--------------------- + +The release number has been increased to 6.0 because of the addition of several +major new pieces of functionality. + +A new function, pcre_dfa_exec(), which implements pattern matching using a DFA +algorithm, has been added. This has a number of advantages for certain cases, +though it does run more slowly, and lacks the ability to capture substrings. On +the other hand, it does find all matches, not just the first, and it works +better for partial matching. The pcrematching man page discusses the +differences. + +The pcretest program has been enhanced so that it can make use of the new +pcre_dfa_exec() matching function and the extra features it provides. + +The distribution now includes a C++ wrapper library. This is built +automatically if a C++ compiler is found. The pcrecpp man page discusses this +interface. + +The code itself has been re-organized into many more files, one for each +function, so it no longer requires everything to be linked in when static +linkage is used. As a consequence, some internal functions have had to have +their names exposed. These functions all have names starting with _pcre_. They +are undocumented, and are not intended for use by outside callers. + +The pcregrep program has been enhanced with new functionality such as +multiline-matching and options for output more matching context. See the +ChangeLog for a complete list of changes to the library and the utility +programs. + + +Release 5.0 13-Sep-04 +--------------------- + +The licence under which PCRE is released has been changed to the more +conventional "BSD" licence. + +In the code, some bugs have been fixed, and there are also some major changes +in this release (which is why I've increased the number to 5.0). Some changes +are internal rearrangements, and some provide a number of new facilities. The +new features are: + +1. There's an "automatic callout" feature that inserts callouts before every + item in the regex, and there's a new callout field that gives the position + in the pattern - useful for debugging and tracing. + +2. The extra_data structure can now be used to pass in a set of character + tables at exec time. This is useful if compiled regex are saved and re-used + at a later time when the tables may not be at the same address. If the + default internal tables are used, the pointer saved with the compiled + pattern is now set to NULL, which means that you don't need to do anything + special unless you are using custom tables. + +3. It is possible, with some restrictions on the content of the regex, to + request "partial" matching. A special return code is given if all of the + subject string matched part of the regex. This could be useful for testing + an input field as it is being typed. + +4. There is now some optional support for Unicode character properties, which + means that the patterns items such as \p{Lu} and \X can now be used. Only + the general category properties are supported. If PCRE is compiled with this + support, an additional 90K data structure is include, which increases the + size of the library dramatically. + +5. There is support for saving compiled patterns and re-using them later. + +6. There is support for running regular expressions that were compiled on a + different host with the opposite endianness. + +7. The pcretest program has been extended to accommodate the new features. + +The main internal rearrangement is that sequences of literal characters are no +longer handled as strings. Instead, each character is handled on its own. This +makes some UTF-8 handling easier, and makes the support of partial matching +possible. Compiled patterns containing long literal strings will be larger as a +result of this change; I hope that performance will not be much affected. + + +Release 4.5 01-Dec-03 +--------------------- + +Again mainly a bug-fix and tidying release, with only a couple of new features: + +1. It's possible now to compile PCRE so that it does not use recursive +function calls when matching. Instead it gets memory from the heap. This slows +things down, but may be necessary on systems with limited stacks. + +2. UTF-8 string checking has been tightened to reject overlong sequences and to +check that a starting offset points to the start of a character. Failure of the +latter returns a new error code: PCRE_ERROR_BADUTF8_OFFSET. + +3. PCRE can now be compiled for systems that use EBCDIC code. + + +Release 4.4 21-Aug-03 +--------------------- + +This is mainly a bug-fix and tidying release. The only new feature is that PCRE +checks UTF-8 strings for validity by default. There is an option to suppress +this, just in case anybody wants that teeny extra bit of performance. + + +Releases 4.1 - 4.3 +------------------ + +Sorry, I forgot about updating the NEWS file for these releases. Please take a +look at ChangeLog. + + +Release 4.0 17-Feb-03 +--------------------- + +There have been a lot of changes for the 4.0 release, adding additional +functionality and mending bugs. Below is a list of the highlights of the new +functionality. For full details of these features, please consult the +documentation. For a complete list of changes, see the ChangeLog file. + +1. Support for Perl's \Q...\E escapes. + +2. "Possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's Java +package. They provide some syntactic sugar for simple cases of "atomic +grouping". + +3. Support for the \G assertion. It is true when the current matching position +is at the start point of the match. + +4. A new feature that provides some of the functionality that Perl provides +with (?{...}). The facility is termed a "callout". The way it is done in PCRE +is for the caller to provide an optional function, by setting pcre_callout to +its entry point. To get the function called, the regex must include (?C) at +appropriate points. + +5. Support for recursive calls to individual subpatterns. This makes it really +easy to get totally confused. + +6. Support for named subpatterns. The Python syntax (?P...) is used to +name a group. + +7. Several extensions to UTF-8 support; it is now fairly complete. There is an +option for pcregrep to make it operate in UTF-8 mode. + +8. The single man page has been split into a number of separate man pages. +These also give rise to individual HTML pages which are put in a separate +directory. There is an index.html page that lists them all. Some hyperlinking +between the pages has been installed. + + +Release 3.5 15-Aug-01 +--------------------- + +1. The configuring system has been upgraded to use later versions of autoconf +and libtool. By default it builds both a shared and a static library if the OS +supports it. You can use --disable-shared or --disable-static on the configure +command if you want only one of them. + +2. The pcretest utility is now installed along with pcregrep because it is +useful for users (to test regexs) and by doing this, it automatically gets +relinked by libtool. The documentation has been turned into a man page, so +there are now .1, .txt, and .html versions in /doc. + +3. Upgrades to pcregrep: + (i) Added long-form option names like gnu grep. + (ii) Added --help to list all options with an explanatory phrase. + (iii) Added -r, --recursive to recurse into sub-directories. + (iv) Added -f, --file to read patterns from a file. + +4. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure +script, to force use of CR or LF instead of \n in the source. On non-Unix +systems, the value can be set in config.h. + +5. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an +absolute limit. Changed the text of the error message to make this clear, and +likewise updated the man page. + +6. The limit of 99 on the number of capturing subpatterns has been removed. +The new limit is 65535, which I hope will not be a "real" limit. + + +Release 3.3 01-Aug-00 +--------------------- + +There is some support for UTF-8 character strings. This is incomplete and +experimental. The documentation describes what is and what is not implemented. +Otherwise, this is just a bug-fixing release. + + +Release 3.0 01-Feb-00 +--------------------- + +1. A "configure" script is now used to configure PCRE for Unix systems. It +builds a Makefile, a config.h file, and the pcre-config script. + +2. PCRE is built as a shared library by default. + +3. There is support for POSIX classes such as [:alpha:]. + +5. There is an experimental recursion feature. + +---------------------------------------------------------------------------- + IMPORTANT FOR THOSE UPGRADING FROM VERSIONS BEFORE 2.00 + +Please note that there has been a change in the API such that a larger +ovector is required at matching time, to provide some additional workspace. +The new man page has details. This change was necessary in order to support +some of the new functionality in Perl 5.005. + + IMPORTANT FOR THOSE UPGRADING FROM VERSION 2.00 + +Another (I hope this is the last!) change has been made to the API for the +pcre_compile() function. An additional argument has been added to make it +possible to pass over a pointer to character tables built in the current +locale by pcre_maketables(). To use the default tables, this new argument +should be passed as NULL. + + IMPORTANT FOR THOSE UPGRADING FROM VERSION 2.05 + +Yet another (and again I hope this really is the last) change has been made +to the API for the pcre_exec() function. An additional argument has been +added to make it possible to start the match other than at the start of the +subject string. This is important if there are lookbehinds. The new man +page has the details, but you just want to convert existing programs, all +you need to do is to stick in a new fifth argument to pcre_exec(), with a +value of zero. For example, change + + pcre_exec(pattern, extra, subject, length, options, ovec, ovecsize) +to + pcre_exec(pattern, extra, subject, length, 0, options, ovec, ovecsize) + +**** diff --git a/pcre/NON-AUTOTOOLS-BUILD b/pcre/NON-AUTOTOOLS-BUILD new file mode 100644 index 00000000000..74075f0160a --- /dev/null +++ b/pcre/NON-AUTOTOOLS-BUILD @@ -0,0 +1,764 @@ +Building PCRE without using autotools +------------------------------------- + +This document contains the following sections: + + General + Generic instructions for the PCRE C library + The C++ wrapper functions + Building for virtual Pascal + Stack size in Windows environments + Linking programs in Windows environments + Calling conventions in Windows environments + Comments about Win32 builds + Building PCRE on Windows with CMake + Use of relative paths with CMake on Windows + Testing with RunTest.bat + Building under Windows CE with Visual Studio 200x + Building under Windows with BCC5.5 + Building using Borland C++ Builder 2007 (CB2007) and higher + Building PCRE on OpenVMS + Building PCRE on Stratus OpenVOS + Building PCRE on native z/OS and z/VM + + +GENERAL + +I (Philip Hazel) have no experience of Windows or VMS sytems and how their +libraries work. The items in the PCRE distribution and Makefile that relate to +anything other than Linux systems are untested by me. + +There are some other comments and files (including some documentation in CHM +format) in the Contrib directory on the FTP site: + + ftp://ftp.csx.cam.ac.uk/pub/software/programming/pcre/Contrib + +The basic PCRE library consists entirely of code written in Standard C, and so +should compile successfully on any system that has a Standard C compiler and +library. The C++ wrapper functions are a separate issue (see below). + +The PCRE distribution includes a "configure" file for use by the configure/make +(autotools) build system, as found in many Unix-like environments. The README +file contains information about the options for "configure". + +There is also support for CMake, which some users prefer, especially in Windows +environments, though it can also be run in Unix-like environments. See the +section entitled "Building PCRE on Windows with CMake" below. + +Versions of config.h and pcre.h are distributed in the PCRE tarballs under the +names config.h.generic and pcre.h.generic. These are provided for those who +build PCRE without using "configure" or CMake. If you use "configure" or CMake, +the .generic versions are not used. + + +GENERIC INSTRUCTIONS FOR THE PCRE C LIBRARY + +The following are generic instructions for building the PCRE C library "by +hand". If you are going to use CMake, this section does not apply to you; you +can skip ahead to the CMake section. + + (1) Copy or rename the file config.h.generic as config.h, and edit the macro + settings that it contains to whatever is appropriate for your environment. + + In particular, you can alter the definition of the NEWLINE macro to + specify what character(s) you want to be interpreted as line terminators. + In an EBCDIC environment, you MUST change NEWLINE, because its default + value is 10, an ASCII LF. The usual EBCDIC newline character is 21 (0x15, + NL), though in some cases it may be 37 (0x25). + + When you compile any of the PCRE modules, you must specify -DHAVE_CONFIG_H + to your compiler so that config.h is included in the sources. + + An alternative approach is not to edit config.h, but to use -D on the + compiler command line to make any changes that you need to the + configuration options. In this case -DHAVE_CONFIG_H must not be set. + + NOTE: There have been occasions when the way in which certain parameters + in config.h are used has changed between releases. (In the configure/make + world, this is handled automatically.) When upgrading to a new release, + you are strongly advised to review config.h.generic before re-using what + you had previously. + + (2) Copy or rename the file pcre.h.generic as pcre.h. + + (3) EITHER: + Copy or rename file pcre_chartables.c.dist as pcre_chartables.c. + + OR: + Compile dftables.c as a stand-alone program (using -DHAVE_CONFIG_H if + you have set up config.h), and then run it with the single argument + "pcre_chartables.c". This generates a set of standard character tables + and writes them to that file. The tables are generated using the default + C locale for your system. If you want to use a locale that is specified + by LC_xxx environment variables, add the -L option to the dftables + command. You must use this method if you are building on a system that + uses EBCDIC code. + + The tables in pcre_chartables.c are defaults. The caller of PCRE can + specify alternative tables at run time. + + (4) Ensure that you have the following header files: + + pcre_internal.h + ucp.h + + (5) For an 8-bit library, compile the following source files, setting + -DHAVE_CONFIG_H as a compiler option if you have set up config.h with your + configuration, or else use other -D settings to change the configuration + as required. + + pcre_byte_order.c + pcre_chartables.c + pcre_compile.c + pcre_config.c + pcre_dfa_exec.c + pcre_exec.c + pcre_fullinfo.c + pcre_get.c + pcre_globals.c + pcre_jit_compile.c + pcre_maketables.c + pcre_newline.c + pcre_ord2utf8.c + pcre_refcount.c + pcre_string_utils.c + pcre_study.c + pcre_tables.c + pcre_ucd.c + pcre_valid_utf8.c + pcre_version.c + pcre_xclass.c + + Make sure that you include -I. in the compiler command (or equivalent for + an unusual compiler) so that all included PCRE header files are first + sought in the current directory. Otherwise you run the risk of picking up + a previously-installed file from somewhere else. + + Note that you must still compile pcre_jit_compile.c, even if you have not + defined SUPPORT_JIT in config.h, because when JIT support is not + configured, dummy functions are compiled. When JIT support IS configured, + pcre_jit_compile.c #includes sources from the sljit subdirectory, where + there should be 16 files, all of whose names begin with "sljit". + + (6) Now link all the compiled code into an object library in whichever form + your system keeps such libraries. This is the basic PCRE C 8-bit library. + If your system has static and shared libraries, you may have to do this + once for each type. + + (7) If you want to build a 16-bit library (as well as, or instead of the 8-bit + or 32-bit libraries) repeat steps 5-6 with the following files: + + pcre16_byte_order.c + pcre16_chartables.c + pcre16_compile.c + pcre16_config.c + pcre16_dfa_exec.c + pcre16_exec.c + pcre16_fullinfo.c + pcre16_get.c + pcre16_globals.c + pcre16_jit_compile.c + pcre16_maketables.c + pcre16_newline.c + pcre16_ord2utf16.c + pcre16_refcount.c + pcre16_string_utils.c + pcre16_study.c + pcre16_tables.c + pcre16_ucd.c + pcre16_utf16_utils.c + pcre16_valid_utf16.c + pcre16_version.c + pcre16_xclass.c + + (8) If you want to build a 16-bit library (as well as, or instead of the 8-bit + or 32-bit libraries) repeat steps 5-6 with the following files: + + pcre32_byte_order.c + pcre32_chartables.c + pcre32_compile.c + pcre32_config.c + pcre32_dfa_exec.c + pcre32_exec.c + pcre32_fullinfo.c + pcre32_get.c + pcre32_globals.c + pcre32_jit_compile.c + pcre32_maketables.c + pcre32_newline.c + pcre32_ord2utf32.c + pcre32_refcount.c + pcre32_string_utils.c + pcre32_study.c + pcre32_tables.c + pcre32_ucd.c + pcre32_utf32_utils.c + pcre32_valid_utf32.c + pcre32_version.c + pcre32_xclass.c + + (9) If you want to build the POSIX wrapper functions (which apply only to the + 8-bit library), ensure that you have the pcreposix.h file and then compile + pcreposix.c (remembering -DHAVE_CONFIG_H if necessary). Link the result + (on its own) as the pcreposix library. + +(10) The pcretest program can be linked with any combination of the 8-bit, + 16-bit and 32-bit libraries (depending on what you selected in config.h). + Compile pcretest.c and pcre_printint.c (again, don't forget + -DHAVE_CONFIG_H) and link them together with the appropriate library/ies. + If you compiled an 8-bit library, pcretest also needs the pcreposix + wrapper library unless you compiled it with -DNOPOSIX. + +(11) Run pcretest on the testinput files in the testdata directory, and check + that the output matches the corresponding testoutput files. There are + comments about what each test does in the section entitled "Testing PCRE" + in the README file. If you compiled more than one of the 8-bit, 16-bit and + 32-bit libraries, you need to run pcretest with the -16 option to do + 16-bit tests and with the -32 option to do 32-bit tests. + + Some tests are relevant only when certain build-time options are selected. + For example, test 4 is for UTF-8/UTF-16/UTF-32 support, and will not run + if you have built PCRE without it. See the comments at the start of each + testinput file. If you have a suitable Unix-like shell, the RunTest script + will run the appropriate tests for you. The command "RunTest list" will + output a list of all the tests. + + Note that the supplied files are in Unix format, with just LF characters + as line terminators. You may need to edit them to change this if your + system uses a different convention. If you are using Windows, you probably + should use the wintestinput3 file instead of testinput3 (and the + corresponding output file). This is a locale test; wintestinput3 sets the + locale to "french" rather than "fr_FR", and there some minor output + differences. + +(12) If you have built PCRE with SUPPORT_JIT, the JIT features will be tested + by the testdata files. However, you might also like to build and run + the freestanding JIT test program, pcre_jit_test.c. + +(13) If you want to use the pcregrep command, compile and link pcregrep.c; it + uses only the basic 8-bit PCRE library (it does not need the pcreposix + library). + + +THE C++ WRAPPER FUNCTIONS + +The PCRE distribution also contains some C++ wrapper functions and tests, +applicable to the 8-bit library, which were contributed by Google Inc. On a +system that can use "configure" and "make", the functions are automatically +built into a library called pcrecpp. It should be straightforward to compile +the .cc files manually on other systems. The files called xxx_unittest.cc are +test programs for each of the corresponding xxx.cc files. + + +BUILDING FOR VIRTUAL PASCAL + +A script for building PCRE using Borland's C++ compiler for use with VPASCAL +was contributed by Alexander Tokarev. Stefan Weber updated the script and added +additional files. The following files in the distribution are for building PCRE +for use with VP/Borland: makevp_c.txt, makevp_l.txt, makevp.bat, pcregexp.pas. + + +STACK SIZE IN WINDOWS ENVIRONMENTS + +The default processor stack size of 1Mb in some Windows environments is too +small for matching patterns that need much recursion. In particular, test 2 may +fail because of this. Normally, running out of stack causes a crash, but there +have been cases where the test program has just died silently. See your linker +documentation for how to increase stack size if you experience problems. The +Linux default of 8Mb is a reasonable choice for the stack, though even that can +be too small for some pattern/subject combinations. + +PCRE has a compile configuration option to disable the use of stack for +recursion so that heap is used instead. However, pattern matching is +significantly slower when this is done. There is more about stack usage in the +"pcrestack" documentation. + + +LINKING PROGRAMS IN WINDOWS ENVIRONMENTS + +If you want to statically link a program against a PCRE library in the form of +a non-dll .a file, you must define PCRE_STATIC before including pcre.h or +pcrecpp.h, otherwise the pcre_malloc() and pcre_free() exported functions will +be declared __declspec(dllimport), with unwanted results. + + +CALLING CONVENTIONS IN WINDOWS ENVIRONMENTS + +It is possible to compile programs to use different calling conventions using +MSVC. Search the web for "calling conventions" for more information. To make it +easier to change the calling convention for the exported functions in the +PCRE library, the macro PCRE_CALL_CONVENTION is present in all the external +definitions. It can be set externally when compiling (e.g. in CFLAGS). If it is +not set, it defaults to empty; the default calling convention is then used +(which is what is wanted most of the time). + + +COMMENTS ABOUT WIN32 BUILDS (see also "BUILDING PCRE ON WINDOWS WITH CMAKE") + +There are two ways of building PCRE using the "configure, make, make install" +paradigm on Windows systems: using MinGW or using Cygwin. These are not at all +the same thing; they are completely different from each other. There is also +support for building using CMake, which some users find a more straightforward +way of building PCRE under Windows. + +The MinGW home page (http://www.mingw.org/) says this: + + MinGW: A collection of freely available and freely distributable Windows + specific header files and import libraries combined with GNU toolsets that + allow one to produce native Windows programs that do not rely on any + 3rd-party C runtime DLLs. + +The Cygwin home page (http://www.cygwin.com/) says this: + + Cygwin is a Linux-like environment for Windows. It consists of two parts: + + . A DLL (cygwin1.dll) which acts as a Linux API emulation layer providing + substantial Linux API functionality + + . A collection of tools which provide Linux look and feel. + + The Cygwin DLL currently works with all recent, commercially released x86 32 + bit and 64 bit versions of Windows, with the exception of Windows CE. + +On both MinGW and Cygwin, PCRE should build correctly using: + + ./configure && make && make install + +This should create two libraries called libpcre and libpcreposix, and, if you +have enabled building the C++ wrapper, a third one called libpcrecpp. These are +independent libraries: when you link with libpcreposix or libpcrecpp you must +also link with libpcre, which contains the basic functions. (Some earlier +releases of PCRE included the basic libpcre functions in libpcreposix. This no +longer happens.) + +A user submitted a special-purpose patch that makes it easy to create +"pcre.dll" under mingw32 using the "msys" environment. It provides "pcre.dll" +as a special target. If you use this target, no other files are built, and in +particular, the pcretest and pcregrep programs are not built. An example of how +this might be used is: + + ./configure --enable-utf --disable-cpp CFLAGS="-03 -s"; make pcre.dll + +Using Cygwin's compiler generates libraries and executables that depend on +cygwin1.dll. If a library that is generated this way is distributed, +cygwin1.dll has to be distributed as well. Since cygwin1.dll is under the GPL +licence, this forces not only PCRE to be under the GPL, but also the entire +application. A distributor who wants to keep their own code proprietary must +purchase an appropriate Cygwin licence. + +MinGW has no such restrictions. The MinGW compiler generates a library or +executable that can run standalone on Windows without any third party dll or +licensing issues. + +But there is more complication: + +If a Cygwin user uses the -mno-cygwin Cygwin gcc flag, what that really does is +to tell Cygwin's gcc to use the MinGW gcc. Cygwin's gcc is only acting as a +front end to MinGW's gcc (if you install Cygwin's gcc, you get both Cygwin's +gcc and MinGW's gcc). So, a user can: + +. Build native binaries by using MinGW or by getting Cygwin and using + -mno-cygwin. + +. Build binaries that depend on cygwin1.dll by using Cygwin with the normal + compiler flags. + +The test files that are supplied with PCRE are in UNIX format, with LF +characters as line terminators. Unless your PCRE library uses a default newline +option that includes LF as a valid newline, it may be necessary to change the +line terminators in the test files to get some of the tests to work. + + +BUILDING PCRE ON WINDOWS WITH CMAKE + +CMake is an alternative configuration facility that can be used instead of +"configure". CMake creates project files (make files, solution files, etc.) +tailored to numerous development environments, including Visual Studio, +Borland, Msys, MinGW, NMake, and Unix. If possible, use short paths with no +spaces in the names for your CMake installation and your PCRE source and build +directories. + +The following instructions were contributed by a PCRE user. If they are not +followed exactly, errors may occur. In the event that errors do occur, it is +recommended that you delete the CMake cache before attempting to repeat the +CMake build process. In the CMake GUI, the cache can be deleted by selecting +"File > Delete Cache". + +1. Install the latest CMake version available from http://www.cmake.org/, and + ensure that cmake\bin is on your path. + +2. Unzip (retaining folder structure) the PCRE source tree into a source + directory such as C:\pcre. You should ensure your local date and time + is not earlier than the file dates in your source dir if the release is + very new. + +3. Create a new, empty build directory, preferably a subdirectory of the + source dir. For example, C:\pcre\pcre-xx\build. + +4. Run cmake-gui from the Shell envirornment of your build tool, for example, + Msys for Msys/MinGW or Visual Studio Command Prompt for VC/VC++. Do not try + to start Cmake from the Windows Start menu, as this can lead to errors. + +5. Enter C:\pcre\pcre-xx and C:\pcre\pcre-xx\build for the source and build + directories, respectively. + +6. Hit the "Configure" button. + +7. Select the particular IDE / build tool that you are using (Visual + Studio, MSYS makefiles, MinGW makefiles, etc.) + +8. The GUI will then list several configuration options. This is where + you can enable UTF-8 support or other PCRE optional features. + +9. Hit "Configure" again. The adjacent "Generate" button should now be + active. + +10. Hit "Generate". + +11. The build directory should now contain a usable build system, be it a + solution file for Visual Studio, makefiles for MinGW, etc. Exit from + cmake-gui and use the generated build system with your compiler or IDE. + E.g., for MinGW you can run "make", or for Visual Studio, open the PCRE + solution, select the desired configuration (Debug, or Release, etc.) and + build the ALL_BUILD project. + +12. If during configuration with cmake-gui you've elected to build the test + programs, you can execute them by building the test project. E.g., for + MinGW: "make test"; for Visual Studio build the RUN_TESTS project. The + most recent build configuration is targeted by the tests. A summary of + test results is presented. Complete test output is subsequently + available for review in Testing\Temporary under your build dir. + + +USE OF RELATIVE PATHS WITH CMAKE ON WINDOWS + +A PCRE user comments as follows: I thought that others may want to know the +current state of CMAKE_USE_RELATIVE_PATHS support on Windows. Here it is: + +-- AdditionalIncludeDirectories is only partially modified (only the + first path - see below) +-- Only some of the contained file paths are modified - shown below for + pcre.vcproj +-- It properly modifies + +I am sure CMake people can fix that if they want to. Until then one will +need to replace existing absolute paths in project files with relative +paths manually (e.g. from VS) - relative to project file location. I did +just that before being told to try CMAKE_USE_RELATIVE_PATHS. Not a big +deal. + +AdditionalIncludeDirectories="E:\builds\pcre\build;E:\builds\pcre\pcre-7.5;" +AdditionalIncludeDirectories=".;E:\builds\pcre\pcre-7.5;" + +RelativePath="pcre.h" +RelativePath="pcre_chartables.c" +RelativePath="pcre_chartables.c.rule" + + +TESTING WITH RUNTEST.BAT + +If configured with CMake, building the test project ("make test" or building +ALL_TESTS in Visual Studio) creates (and runs) pcre_test.bat (and depending +on your configuration options, possibly other test programs) in the build +directory. Pcre_test.bat runs RunTest.Bat with correct source and exe paths. + +For manual testing with RunTest.bat, provided the build dir is a subdirectory +of the source directory: Open command shell window. Chdir to the location +of your pcretest.exe and pcregrep.exe programs. Call RunTest.bat with +"..\RunTest.Bat" or "..\..\RunTest.bat" as appropriate. + +To run only a particular test with RunTest.Bat provide a test number argument. + +Otherwise: + +1. Copy RunTest.bat into the directory where pcretest.exe and pcregrep.exe + have been created. + +2. Edit RunTest.bat to indentify the full or relative location of + the pcre source (wherein which the testdata folder resides), e.g.: + + set srcdir=C:\pcre\pcre-8.20 + +3. In a Windows command environment, chdir to the location of your bat and + exe programs. + +4. Run RunTest.bat. Test outputs will automatically be compared to expected + results, and discrepancies will be identified in the console output. + +To independently test the just-in-time compiler, run pcre_jit_test.exe. +To test pcrecpp, run pcrecpp_unittest.exe, pcre_stringpiece_unittest.exe and +pcre_scanner_unittest.exe. + + +BUILDING UNDER WINDOWS CE WITH VISUAL STUDIO 200x + +Vincent Richomme sent a zip archive of files to help with this process. They +can be found in the file "pcre-vsbuild.zip" in the Contrib directory of the FTP +site. + + +BUILDING UNDER WINDOWS WITH BCC5.5 + +Michael Roy sent these comments about building PCRE under Windows with BCC5.5: + +Some of the core BCC libraries have a version of PCRE from 1998 built in, which +can lead to pcre_exec() giving an erroneous PCRE_ERROR_NULL from a version +mismatch. I'm including an easy workaround below, if you'd like to include it +in the non-unix instructions: + +When linking a project with BCC5.5, pcre.lib must be included before any of the +libraries cw32.lib, cw32i.lib, cw32mt.lib, and cw32mti.lib on the command line. + + +BUILDING USING BORLAND C++ BUILDER 2007 (CB2007) AND HIGHER + +A PCRE user sent these comments about this environment (see also the comment +from another user that follows them): + +The XE versions of C++ Builder come with a RegularExpressionsCore class which +contain a version of TPerlRegEx. However, direct use of the C PCRE library may +be desirable. + +The default makevp.bat, however, supplied with PCRE builds a version of PCRE +that is not usable with any version of C++ Builder because the compiler ships +with an embedded version of PCRE, version 2.01 from 1998! [See also the note +about BCC5.5 above.] If you want to use PCRE you'll need to rename the +functions (pcre_compile to pcre_compile_bcc, etc) or do as I have done and just +use the 16 bit versions. I'm using std::wstring everywhere anyway. Since the +embedded version of PCRE does not have the 16 bit function names, there is no +conflict. + +Building PCRE using a C++ Builder static library project file (recommended): + +1. Rename or remove pcre.h, pcreposi.h, and pcreposix.h from your C++ Builder +original include path. + +2. Download PCRE from pcre.org and extract to a directory. + +3. Rename pcre_chartables.c.dist to pcre_chartables.c, pcre.h.generic to +pcre.h, and config.h.generic to config.h. + +4. Edit pcre.h and pcre_config.c so that they include config.h. + +5. Edit config.h like so: + +Comment out the following lines: +#define PACKAGE "pcre" +#define PACKAGE_BUGREPORT "" +#define PACKAGE_NAME "PCRE" +#define PACKAGE_STRING "PCRE 8.32" +#define PACKAGE_TARNAME "pcre" +#define PACKAGE_URL "" +#define PACKAGE_VERSION "8.32" + +Add the following lines: +#ifndef SUPPORT_UTF +#define SUPPORT_UTF 100 // any value is fine +#endif + +#ifndef SUPPORT_UCP +#define SUPPORT_UCP 101 // any value is fine +#endif + +#ifndef SUPPORT_UCP +#define SUPPORT_PCRE16 102 // any value is fine +#endif + +#ifndef SUPPORT_UTF8 +#define SUPPORT_UTF8 103 // any value is fine +#endif + +6. Build a C++ Builder project using the IDE. Go to File / New / Other and +choose Static Library. You can name it pcre.cbproj or whatever. Now set your +paths by going to Project / Options. Set the Include path. Do this from the +"Base" option to apply to both Release and Debug builds. Now add the following +files to the project: + +pcre.h +pcre16_byte_order.c +pcre16_chartables.c +pcre16_compile.c +pcre16_config.c +pcre16_dfa_exec.c +pcre16_exec.c +pcre16_fullinfo.c +pcre16_get.c +pcre16_globals.c +pcre16_maketables.c +pcre16_newline.c +pcre16_ord2utf16.c +pcre16_printint.c +pcre16_refcount.c +pcre16_string_utils.c +pcre16_study.c +pcre16_tables.c +pcre16_ucd.c +pcre16_utf16_utils.c +pcre16_valid_utf16.c +pcre16_version.c +pcre16_xclass.c + +//Optional +pcre_version.c + +7. After compiling the .lib file, copy the .lib and header files to a project +you want to use PCRE with. Enjoy. + +Optional ... Building PCRE using the makevp.bat file: + +1. Edit makevp_c.txt and makevp_l.txt and change all the names to the 16 bit +versions. + +2. Edit makevp.bat and set the path to C++ Builder. Run makevp.bat. + +Another PCRE user added this comment: + +Another approach I successfully used for some years with BCB 5 and 6 was to +make sure that include and library paths of PCRE are configured before the +default paths of the IDE in the dialogs where one can manage those paths. +Afterwards one can open the project files using a text editor and manually add +the self created library for pcre itself, pcrecpp doesn't ship with the IDE, in +the library nodes where the IDE manages its own libraries to link against in +front of the IDE-own libraries. This way one can use the default PCRE function +names without getting access violations on runtime. + + + + +BUILDING PCRE ON OPENVMS + +Stephen Hoffman sent the following, in December 2012: + +"Here is a very short write-up on the +OpenVMS port and here + + + +is a zip with the OpenVMS files, and with one modified testing-related PCRE +file." This is a port of PCRE 8.32. + +Earlier, Dan Mooney sent the following comments about building PCRE on OpenVMS. +They relate to an older version of PCRE that used fewer source files, so the +exact commands will need changing. See the current list of source files above. + +"It was quite easy to compile and link the library. I don't have a formal +make file but the attached file [reproduced below] contains the OpenVMS DCL +commands I used to build the library. I had to add #define +POSIX_MALLOC_THRESHOLD 10 to pcre.h since it was not defined anywhere. + +The library was built on: +O/S: HP OpenVMS v7.3-1 +Compiler: Compaq C v6.5-001-48BCD +Linker: vA13-01 + +The test results did not match 100% due to the issues you mention in your +documentation regarding isprint(), iscntrl(), isgraph() and ispunct(). I +modified some of the character tables temporarily and was able to get the +results to match. Tests using the fr locale did not match since I don't have +that locale loaded. The study size was always reported to be 3 less than the +value in the standard test output files." + +========================= +$! This DCL procedure builds PCRE on OpenVMS +$! +$! I followed the instructions in the non-unix-use file in the distribution. +$! +$ COMPILE == "CC/LIST/NOMEMBER_ALIGNMENT/PREFIX_LIBRARY_ENTRIES=ALL_ENTRIES +$ COMPILE DFTABLES.C +$ LINK/EXE=DFTABLES.EXE DFTABLES.OBJ +$ RUN DFTABLES.EXE/OUTPUT=CHARTABLES.C +$ COMPILE MAKETABLES.C +$ COMPILE GET.C +$ COMPILE STUDY.C +$! I had to set POSIX_MALLOC_THRESHOLD to 10 in PCRE.H since the symbol +$! did not seem to be defined anywhere. +$! I edited pcre.h and added #DEFINE SUPPORT_UTF8 to enable UTF8 support. +$ COMPILE PCRE.C +$ LIB/CREATE PCRE MAKETABLES.OBJ, GET.OBJ, STUDY.OBJ, PCRE.OBJ +$! I had to set POSIX_MALLOC_THRESHOLD to 10 in PCRE.H since the symbol +$! did not seem to be defined anywhere. +$ COMPILE PCREPOSIX.C +$ LIB/CREATE PCREPOSIX PCREPOSIX.OBJ +$ COMPILE PCRETEST.C +$ LINK/EXE=PCRETEST.EXE PCRETEST.OBJ, PCRE/LIB, PCREPOSIX/LIB +$! C programs that want access to command line arguments must be +$! defined as a symbol +$ PCRETEST :== "$ SYS$ROADSUSERS:[DMOONEY.REGEXP]PCRETEST.EXE" +$! Arguments must be enclosed in quotes. +$ PCRETEST "-C" +$! Test results: +$! +$! The test results did not match 100%. The functions isprint(), iscntrl(), +$! isgraph() and ispunct() on OpenVMS must not produce the same results +$! as the system that built the test output files provided with the +$! distribution. +$! +$! The study size did not match and was always 3 less on OpenVMS. +$! +$! Locale could not be set to fr +$! +========================= + + +BUILDING PCRE ON STRATUS OPENVOS + +These notes on the port of PCRE to VOS (lightly edited) were supplied by +Ashutosh Warikoo, whose email address has the local part awarikoo and the +domain nse.co.in. The port was for version 7.9 in August 2009. + +1. Building PCRE + +I built pcre on OpenVOS Release 17.0.1at using GNU Tools 3.4a without any +problems. I used the following packages to build PCRE: + + ftp://ftp.stratus.com/pub/vos/posix/ga/posix.save.evf.gz + +Please read and follow the instructions that come with these packages. To start +the build of pcre, from the root of the package type: + + ./build.sh + +2. Installing PCRE + +Once you have successfully built PCRE, login to the SysAdmin group, switch to +the root user, and type + + [ !create_dir (master_disk)>usr --if needed ] + [ !create_dir (master_disk)>usr>local --if needed ] + !gmake install + +This installs PCRE and its man pages into /usr/local. You can add +(master_disk)>usr>local>bin to your command search paths, or if you are in +BASH, add /usr/local/bin to the PATH environment variable. + +4. Restrictions + +This port requires readline library optionally. However during the build I +faced some yet unexplored errors while linking with readline. As it was an +optional component I chose to disable it. + +5. Known Problems + +I ran the test suite, but you will have to be your own judge of whether this +command, and this port, suits your purposes. If you find any problems that +appear to be related to the port itself, please let me know. Please see the +build.log file in the root of the package also. + + +BUILDING PCRE ON NATIVE Z/OS AND Z/VM + +z/OS and z/VM are operating systems for mainframe computers, produced by IBM. +The character code used is EBCDIC, not ASCII or Unicode. In z/OS, UNIX APIs and +applications can be supported through UNIX System Services, and in such an +environment PCRE can be built in the same way as in other systems. However, in +native z/OS (without UNIX System Services) and in z/VM, special ports are +required. For details, please see this web site: + + http://www.zaconsultants.net + +There is also a mirror here: + + http://www.vsoft-software.com/downloads.html + +========================== +Last Updated: 14 May 2013 diff --git a/pcre/NON-UNIX-USE b/pcre/NON-UNIX-USE new file mode 100644 index 00000000000..a25546b6ff5 --- /dev/null +++ b/pcre/NON-UNIX-USE @@ -0,0 +1,7 @@ +Compiling PCRE on non-Unix systems +---------------------------------- + +This has been renamed to better reflect its contents. Please see the file +NON-AUTOTOOLS-BUILD for details of how to build PCRE without using autotools. + +#### diff --git a/pcre/README b/pcre/README new file mode 100644 index 00000000000..e14935efeba --- /dev/null +++ b/pcre/README @@ -0,0 +1,974 @@ +README file for PCRE (Perl-compatible regular expression library) +----------------------------------------------------------------- + +The latest release of PCRE is always available in three alternative formats +from: + + ftp://ftp.csx.cam.ac.uk/pub/software/programming/pcre/pcre-xxx.tar.gz + ftp://ftp.csx.cam.ac.uk/pub/software/programming/pcre/pcre-xxx.tar.bz2 + ftp://ftp.csx.cam.ac.uk/pub/software/programming/pcre/pcre-xxx.zip + +There is a mailing list for discussion about the development of PCRE at + + pcre-dev@exim.org + +Please read the NEWS file if you are upgrading from a previous release. +The contents of this README file are: + + The PCRE APIs + Documentation for PCRE + Contributions by users of PCRE + Building PCRE on non-Unix-like systems + Building PCRE without using autotools + Building PCRE using autotools + Retrieving configuration information + Shared libraries + Cross-compiling using autotools + Using HP's ANSI C++ compiler (aCC) + Compiling in Tru64 using native compilers + Using Sun's compilers for Solaris + Using PCRE from MySQL + Making new tarballs + Testing PCRE + Character tables + File manifest + + +The PCRE APIs +------------- + +PCRE is written in C, and it has its own API. There are three sets of +functions, one for the 8-bit library, which processes strings of bytes, one for +the 16-bit library, which processes strings of 16-bit values, and one for the +32-bit library, which processes strings of 32-bit values. The distribution also +includes a set of C++ wrapper functions (see the pcrecpp man page for details), +courtesy of Google Inc., which can be used to call the 8-bit PCRE library from +C++. + +In addition, there is a set of C wrapper functions (again, just for the 8-bit +library) that are based on the POSIX regular expression API (see the pcreposix +man page). These end up in the library called libpcreposix. Note that this just +provides a POSIX calling interface to PCRE; the regular expressions themselves +still follow Perl syntax and semantics. The POSIX API is restricted, and does +not give full access to all of PCRE's facilities. + +The header file for the POSIX-style functions is called pcreposix.h. The +official POSIX name is regex.h, but I did not want to risk possible problems +with existing files of that name by distributing it that way. To use PCRE with +an existing program that uses the POSIX API, pcreposix.h will have to be +renamed or pointed at by a link. + +If you are using the POSIX interface to PCRE and there is already a POSIX regex +library installed on your system, as well as worrying about the regex.h header +file (as mentioned above), you must also take care when linking programs to +ensure that they link with PCRE's libpcreposix library. Otherwise they may pick +up the POSIX functions of the same name from the other library. + +One way of avoiding this confusion is to compile PCRE with the addition of +-Dregcomp=PCREregcomp (and similarly for the other POSIX functions) to the +compiler flags (CFLAGS if you are using "configure" -- see below). This has the +effect of renaming the functions so that the names no longer clash. Of course, +you have to do the same thing for your applications, or write them using the +new names. + + +Documentation for PCRE +---------------------- + +If you install PCRE in the normal way on a Unix-like system, you will end up +with a set of man pages whose names all start with "pcre". The one that is just +called "pcre" lists all the others. In addition to these man pages, the PCRE +documentation is supplied in two other forms: + + 1. There are files called doc/pcre.txt, doc/pcregrep.txt, and + doc/pcretest.txt in the source distribution. The first of these is a + concatenation of the text forms of all the section 3 man pages except + those that summarize individual functions. The other two are the text + forms of the section 1 man pages for the pcregrep and pcretest commands. + These text forms are provided for ease of scanning with text editors or + similar tools. They are installed in /share/doc/pcre, where + is the installation prefix (defaulting to /usr/local). + + 2. A set of files containing all the documentation in HTML form, hyperlinked + in various ways, and rooted in a file called index.html, is distributed in + doc/html and installed in /share/doc/pcre/html. + +Users of PCRE have contributed files containing the documentation for various +releases in CHM format. These can be found in the Contrib directory of the FTP +site (see next section). + + +Contributions by users of PCRE +------------------------------ + +You can find contributions from PCRE users in the directory + + ftp://ftp.csx.cam.ac.uk/pub/software/programming/pcre/Contrib + +There is a README file giving brief descriptions of what they are. Some are +complete in themselves; others are pointers to URLs containing relevant files. +Some of this material is likely to be well out-of-date. Several of the earlier +contributions provided support for compiling PCRE on various flavours of +Windows (I myself do not use Windows). Nowadays there is more Windows support +in the standard distribution, so these contibutions have been archived. + + +Building PCRE on non-Unix-like systems +-------------------------------------- + +For a non-Unix-like system, please read the comments in the file +NON-AUTOTOOLS-BUILD, though if your system supports the use of "configure" and +"make" you may be able to build PCRE using autotools in the same way as for +many Unix-like systems. + +PCRE can also be configured using the GUI facility provided by CMake's +cmake-gui command. This creates Makefiles, solution files, etc. The file +NON-AUTOTOOLS-BUILD has information about CMake. + +PCRE has been compiled on many different operating systems. It should be +straightforward to build PCRE on any system that has a Standard C compiler and +library, because it uses only Standard C functions. + + +Building PCRE without using autotools +------------------------------------- + +The use of autotools (in particular, libtool) is problematic in some +environments, even some that are Unix or Unix-like. See the NON-AUTOTOOLS-BUILD +file for ways of building PCRE without using autotools. + + +Building PCRE using autotools +----------------------------- + +If you are using HP's ANSI C++ compiler (aCC), please see the special note +in the section entitled "Using HP's ANSI C++ compiler (aCC)" below. + +The following instructions assume the use of the widely used "configure; make; +make install" (autotools) process. + +To build PCRE on system that supports autotools, first run the "configure" +command from the PCRE distribution directory, with your current directory set +to the directory where you want the files to be created. This command is a +standard GNU "autoconf" configuration script, for which generic instructions +are supplied in the file INSTALL. + +Most commonly, people build PCRE within its own distribution directory, and in +this case, on many systems, just running "./configure" is sufficient. However, +the usual methods of changing standard defaults are available. For example: + +CFLAGS='-O2 -Wall' ./configure --prefix=/opt/local + +This command specifies that the C compiler should be run with the flags '-O2 +-Wall' instead of the default, and that "make install" should install PCRE +under /opt/local instead of the default /usr/local. + +If you want to build in a different directory, just run "configure" with that +directory as current. For example, suppose you have unpacked the PCRE source +into /source/pcre/pcre-xxx, but you want to build it in /build/pcre/pcre-xxx: + +cd /build/pcre/pcre-xxx +/source/pcre/pcre-xxx/configure + +PCRE is written in C and is normally compiled as a C library. However, it is +possible to build it as a C++ library, though the provided building apparatus +does not have any features to support this. + +There are some optional features that can be included or omitted from the PCRE +library. They are also documented in the pcrebuild man page. + +. By default, both shared and static libraries are built. You can change this + by adding one of these options to the "configure" command: + + --disable-shared + --disable-static + + (See also "Shared libraries on Unix-like systems" below.) + +. By default, only the 8-bit library is built. If you add --enable-pcre16 to + the "configure" command, the 16-bit library is also built. If you add + --enable-pcre32 to the "configure" command, the 32-bit library is also built. + If you want only the 16-bit or 32-bit library, use --disable-pcre8 to disable + building the 8-bit library. + +. If you are building the 8-bit library and want to suppress the building of + the C++ wrapper library, you can add --disable-cpp to the "configure" + command. Otherwise, when "configure" is run without --disable-pcre8, it will + try to find a C++ compiler and C++ header files, and if it succeeds, it will + try to build the C++ wrapper. + +. If you want to include support for just-in-time compiling, which can give + large performance improvements on certain platforms, add --enable-jit to the + "configure" command. This support is available only for certain hardware + architectures. If you try to enable it on an unsupported architecture, there + will be a compile time error. + +. When JIT support is enabled, pcregrep automatically makes use of it, unless + you add --disable-pcregrep-jit to the "configure" command. + +. If you want to make use of the support for UTF-8 Unicode character strings in + the 8-bit library, or UTF-16 Unicode character strings in the 16-bit library, + or UTF-32 Unicode character strings in the 32-bit library, you must add + --enable-utf to the "configure" command. Without it, the code for handling + UTF-8, UTF-16 and UTF-8 is not included in the relevant library. Even + when --enable-utf is included, the use of a UTF encoding still has to be + enabled by an option at run time. When PCRE is compiled with this option, its + input can only either be ASCII or UTF-8/16/32, even when running on EBCDIC + platforms. It is not possible to use both --enable-utf and --enable-ebcdic at + the same time. + +. There are no separate options for enabling UTF-8, UTF-16 and UTF-32 + independently because that would allow ridiculous settings such as requesting + UTF-16 support while building only the 8-bit library. However, the option + --enable-utf8 is retained for backwards compatibility with earlier releases + that did not support 16-bit or 32-bit character strings. It is synonymous with + --enable-utf. It is not possible to configure one library with UTF support + and the other without in the same configuration. + +. If, in addition to support for UTF-8/16/32 character strings, you want to + include support for the \P, \p, and \X sequences that recognize Unicode + character properties, you must add --enable-unicode-properties to the + "configure" command. This adds about 30K to the size of the library (in the + form of a property table); only the basic two-letter properties such as Lu + are supported. + +. You can build PCRE to recognize either CR or LF or the sequence CRLF or any + of the preceding, or any of the Unicode newline sequences as indicating the + end of a line. Whatever you specify at build time is the default; the caller + of PCRE can change the selection at run time. The default newline indicator + is a single LF character (the Unix standard). You can specify the default + newline indicator by adding --enable-newline-is-cr or --enable-newline-is-lf + or --enable-newline-is-crlf or --enable-newline-is-anycrlf or + --enable-newline-is-any to the "configure" command, respectively. + + If you specify --enable-newline-is-cr or --enable-newline-is-crlf, some of + the standard tests will fail, because the lines in the test files end with + LF. Even if the files are edited to change the line endings, there are likely + to be some failures. With --enable-newline-is-anycrlf or + --enable-newline-is-any, many tests should succeed, but there may be some + failures. + +. By default, the sequence \R in a pattern matches any Unicode line ending + sequence. This is independent of the option specifying what PCRE considers to + be the end of a line (see above). However, the caller of PCRE can restrict \R + to match only CR, LF, or CRLF. You can make this the default by adding + --enable-bsr-anycrlf to the "configure" command (bsr = "backslash R"). + +. When called via the POSIX interface, PCRE uses malloc() to get additional + storage for processing capturing parentheses if there are more than 10 of + them in a pattern. You can increase this threshold by setting, for example, + + --with-posix-malloc-threshold=20 + + on the "configure" command. + +. PCRE has a counter that can be set to limit the amount of resources it uses. + If the limit is exceeded during a match, the match fails. The default is ten + million. You can change the default by setting, for example, + + --with-match-limit=500000 + + on the "configure" command. This is just the default; individual calls to + pcre_exec() can supply their own value. There is more discussion on the + pcreapi man page. + +. There is a separate counter that limits the depth of recursive function calls + during a matching process. This also has a default of ten million, which is + essentially "unlimited". You can change the default by setting, for example, + + --with-match-limit-recursion=500000 + + Recursive function calls use up the runtime stack; running out of stack can + cause programs to crash in strange ways. There is a discussion about stack + sizes in the pcrestack man page. + +. The default maximum compiled pattern size is around 64K. You can increase + this by adding --with-link-size=3 to the "configure" command. In the 8-bit + library, PCRE then uses three bytes instead of two for offsets to different + parts of the compiled pattern. In the 16-bit library, --with-link-size=3 is + the same as --with-link-size=4, which (in both libraries) uses four-byte + offsets. Increasing the internal link size reduces performance. In the 32-bit + library, the only supported link size is 4. + +. You can build PCRE so that its internal match() function that is called from + pcre_exec() does not call itself recursively. Instead, it uses memory blocks + obtained from the heap via the special functions pcre_stack_malloc() and + pcre_stack_free() to save data that would otherwise be saved on the stack. To + build PCRE like this, use + + --disable-stack-for-recursion + + on the "configure" command. PCRE runs more slowly in this mode, but it may be + necessary in environments with limited stack sizes. This applies only to the + normal execution of the pcre_exec() function; if JIT support is being + successfully used, it is not relevant. Equally, it does not apply to + pcre_dfa_exec(), which does not use deeply nested recursion. There is a + discussion about stack sizes in the pcrestack man page. + +. For speed, PCRE uses four tables for manipulating and identifying characters + whose code point values are less than 256. By default, it uses a set of + tables for ASCII encoding that is part of the distribution. If you specify + + --enable-rebuild-chartables + + a program called dftables is compiled and run in the default C locale when + you obey "make". It builds a source file called pcre_chartables.c. If you do + not specify this option, pcre_chartables.c is created as a copy of + pcre_chartables.c.dist. See "Character tables" below for further information. + +. It is possible to compile PCRE for use on systems that use EBCDIC as their + character code (as opposed to ASCII/Unicode) by specifying + + --enable-ebcdic + + This automatically implies --enable-rebuild-chartables (see above). However, + when PCRE is built this way, it always operates in EBCDIC. It cannot support + both EBCDIC and UTF-8/16/32. There is a second option, --enable-ebcdic-nl25, + which specifies that the code value for the EBCDIC NL character is 0x25 + instead of the default 0x15. + +. In environments where valgrind is installed, if you specify + + --enable-valgrind + + PCRE will use valgrind annotations to mark certain memory regions as + unaddressable. This allows it to detect invalid memory accesses, and is + mostly useful for debugging PCRE itself. + +. In environments where the gcc compiler is used and lcov version 1.6 or above + is installed, if you specify + + --enable-coverage + + the build process implements a code coverage report for the test suite. The + report is generated by running "make coverage". If ccache is installed on + your system, it must be disabled when building PCRE for coverage reporting. + You can do this by setting the environment variable CCACHE_DISABLE=1 before + running "make" to build PCRE. + +. The pcregrep program currently supports only 8-bit data files, and so + requires the 8-bit PCRE library. It is possible to compile pcregrep to use + libz and/or libbz2, in order to read .gz and .bz2 files (respectively), by + specifying one or both of + + --enable-pcregrep-libz + --enable-pcregrep-libbz2 + + Of course, the relevant libraries must be installed on your system. + +. The default size of internal buffer used by pcregrep can be set by, for + example: + + --with-pcregrep-bufsize=50K + + The default value is 20K. + +. It is possible to compile pcretest so that it links with the libreadline + or libedit libraries, by specifying, respectively, + + --enable-pcretest-libreadline or --enable-pcretest-libedit + + If this is done, when pcretest's input is from a terminal, it reads it using + the readline() function. This provides line-editing and history facilities. + Note that libreadline is GPL-licenced, so if you distribute a binary of + pcretest linked in this way, there may be licensing issues. These can be + avoided by linking with libedit (which has a BSD licence) instead. + + Enabling libreadline causes the -lreadline option to be added to the pcretest + build. In many operating environments with a sytem-installed readline + library this is sufficient. However, in some environments (e.g. if an + unmodified distribution version of readline is in use), it may be necessary + to specify something like LIBS="-lncurses" as well. This is because, to quote + the readline INSTALL, "Readline uses the termcap functions, but does not link + with the termcap or curses library itself, allowing applications which link + with readline the to choose an appropriate library." If you get error + messages about missing functions tgetstr, tgetent, tputs, tgetflag, or tgoto, + this is the problem, and linking with the ncurses library should fix it. + +The "configure" script builds the following files for the basic C library: + +. Makefile the makefile that builds the library +. config.h build-time configuration options for the library +. pcre.h the public PCRE header file +. pcre-config script that shows the building settings such as CFLAGS + that were set for "configure" +. libpcre.pc ) data for the pkg-config command +. libpcre16.pc ) +. libpcre32.pc ) +. libpcreposix.pc ) +. libtool script that builds shared and/or static libraries + +Versions of config.h and pcre.h are distributed in the PCRE tarballs under the +names config.h.generic and pcre.h.generic. These are provided for those who +have to built PCRE without using "configure" or CMake. If you use "configure" +or CMake, the .generic versions are not used. + +When building the 8-bit library, if a C++ compiler is found, the following +files are also built: + +. libpcrecpp.pc data for the pkg-config command +. pcrecpparg.h header file for calling PCRE via the C++ wrapper +. pcre_stringpiece.h header for the C++ "stringpiece" functions + +The "configure" script also creates config.status, which is an executable +script that can be run to recreate the configuration, and config.log, which +contains compiler output from tests that "configure" runs. + +Once "configure" has run, you can run "make". This builds the the libraries +libpcre, libpcre16 and/or libpcre32, and a test program called pcretest. If you +enabled JIT support with --enable-jit, a test program called pcre_jit_test is +built as well. + +If the 8-bit library is built, libpcreposix and the pcregrep command are also +built, and if a C++ compiler was found on your system, and you did not disable +it with --disable-cpp, "make" builds the C++ wrapper library, which is called +libpcrecpp, as well as some test programs called pcrecpp_unittest, +pcre_scanner_unittest, and pcre_stringpiece_unittest. + +The command "make check" runs all the appropriate tests. Details of the PCRE +tests are given below in a separate section of this document. + +You can use "make install" to install PCRE into live directories on your +system. The following are installed (file names are all relative to the + that is set when "configure" is run): + + Commands (bin): + pcretest + pcregrep (if 8-bit support is enabled) + pcre-config + + Libraries (lib): + libpcre16 (if 16-bit support is enabled) + libpcre32 (if 32-bit support is enabled) + libpcre (if 8-bit support is enabled) + libpcreposix (if 8-bit support is enabled) + libpcrecpp (if 8-bit and C++ support is enabled) + + Configuration information (lib/pkgconfig): + libpcre16.pc + libpcre32.pc + libpcre.pc + libpcreposix.pc + libpcrecpp.pc (if C++ support is enabled) + + Header files (include): + pcre.h + pcreposix.h + pcre_scanner.h ) + pcre_stringpiece.h ) if C++ support is enabled + pcrecpp.h ) + pcrecpparg.h ) + + Man pages (share/man/man{1,3}): + pcregrep.1 + pcretest.1 + pcre-config.1 + pcre.3 + pcre*.3 (lots more pages, all starting "pcre") + + HTML documentation (share/doc/pcre/html): + index.html + *.html (lots more pages, hyperlinked from index.html) + + Text file documentation (share/doc/pcre): + AUTHORS + COPYING + ChangeLog + LICENCE + NEWS + README + pcre.txt (a concatenation of the man(3) pages) + pcretest.txt the pcretest man page + pcregrep.txt the pcregrep man page + pcre-config.txt the pcre-config man page + +If you want to remove PCRE from your system, you can run "make uninstall". +This removes all the files that "make install" installed. However, it does not +remove any directories, because these are often shared with other programs. + + +Retrieving configuration information +------------------------------------ + +Running "make install" installs the command pcre-config, which can be used to +recall information about the PCRE configuration and installation. For example: + + pcre-config --version + +prints the version number, and + + pcre-config --libs + +outputs information about where the library is installed. This command can be +included in makefiles for programs that use PCRE, saving the programmer from +having to remember too many details. + +The pkg-config command is another system for saving and retrieving information +about installed libraries. Instead of separate commands for each library, a +single command is used. For example: + + pkg-config --cflags pcre + +The data is held in *.pc files that are installed in a directory called +/lib/pkgconfig. + + +Shared libraries +---------------- + +The default distribution builds PCRE as shared libraries and static libraries, +as long as the operating system supports shared libraries. Shared library +support relies on the "libtool" script which is built as part of the +"configure" process. + +The libtool script is used to compile and link both shared and static +libraries. They are placed in a subdirectory called .libs when they are newly +built. The programs pcretest and pcregrep are built to use these uninstalled +libraries (by means of wrapper scripts in the case of shared libraries). When +you use "make install" to install shared libraries, pcregrep and pcretest are +automatically re-built to use the newly installed shared libraries before being +installed themselves. However, the versions left in the build directory still +use the uninstalled libraries. + +To build PCRE using static libraries only you must use --disable-shared when +configuring it. For example: + +./configure --prefix=/usr/gnu --disable-shared + +Then run "make" in the usual way. Similarly, you can use --disable-static to +build only shared libraries. + + +Cross-compiling using autotools +------------------------------- + +You can specify CC and CFLAGS in the normal way to the "configure" command, in +order to cross-compile PCRE for some other host. However, you should NOT +specify --enable-rebuild-chartables, because if you do, the dftables.c source +file is compiled and run on the local host, in order to generate the inbuilt +character tables (the pcre_chartables.c file). This will probably not work, +because dftables.c needs to be compiled with the local compiler, not the cross +compiler. + +When --enable-rebuild-chartables is not specified, pcre_chartables.c is created +by making a copy of pcre_chartables.c.dist, which is a default set of tables +that assumes ASCII code. Cross-compiling with the default tables should not be +a problem. + +If you need to modify the character tables when cross-compiling, you should +move pcre_chartables.c.dist out of the way, then compile dftables.c by hand and +run it on the local host to make a new version of pcre_chartables.c.dist. +Then when you cross-compile PCRE this new version of the tables will be used. + + +Using HP's ANSI C++ compiler (aCC) +---------------------------------- + +Unless C++ support is disabled by specifying the "--disable-cpp" option of the +"configure" script, you must include the "-AA" option in the CXXFLAGS +environment variable in order for the C++ components to compile correctly. + +Also, note that the aCC compiler on PA-RISC platforms may have a defect whereby +needed libraries fail to get included when specifying the "-AA" compiler +option. If you experience unresolved symbols when linking the C++ programs, +use the workaround of specifying the following environment variable prior to +running the "configure" script: + + CXXLDFLAGS="-lstd_v2 -lCsup_v2" + + +Compiling in Tru64 using native compilers +----------------------------------------- + +The following error may occur when compiling with native compilers in the Tru64 +operating system: + + CXX libpcrecpp_la-pcrecpp.lo +cxx: Error: /usr/lib/cmplrs/cxx/V7.1-006/include/cxx/iosfwd, line 58: #error + directive: "cannot include iosfwd -- define __USE_STD_IOSTREAM to + override default - see section 7.1.2 of the C++ Using Guide" +#error "cannot include iosfwd -- define __USE_STD_IOSTREAM to override default +- see section 7.1.2 of the C++ Using Guide" + +This may be followed by other errors, complaining that 'namespace "std" has no +member'. The solution to this is to add the line + +#define __USE_STD_IOSTREAM 1 + +to the config.h file. + + +Using Sun's compilers for Solaris +--------------------------------- + +A user reports that the following configurations work on Solaris 9 sparcv9 and +Solaris 9 x86 (32-bit): + + Solaris 9 sparcv9: ./configure --disable-cpp CC=/bin/cc CFLAGS="-m64 -g" + Solaris 9 x86: ./configure --disable-cpp CC=/bin/cc CFLAGS="-g" + + +Using PCRE from MySQL +--------------------- + +On systems where both PCRE and MySQL are installed, it is possible to make use +of PCRE from within MySQL, as an alternative to the built-in pattern matching. +There is a web page that tells you how to do this: + + http://www.mysqludf.org/lib_mysqludf_preg/index.php + + +Making new tarballs +------------------- + +The command "make dist" creates three PCRE tarballs, in tar.gz, tar.bz2, and +zip formats. The command "make distcheck" does the same, but then does a trial +build of the new distribution to ensure that it works. + +If you have modified any of the man page sources in the doc directory, you +should first run the PrepareRelease script before making a distribution. This +script creates the .txt and HTML forms of the documentation from the man pages. + + +Testing PCRE +------------ + +To test the basic PCRE library on a Unix-like system, run the RunTest script. +There is another script called RunGrepTest that tests the options of the +pcregrep command. If the C++ wrapper library is built, three test programs +called pcrecpp_unittest, pcre_scanner_unittest, and pcre_stringpiece_unittest +are also built. When JIT support is enabled, another test program called +pcre_jit_test is built. + +Both the scripts and all the program tests are run if you obey "make check" or +"make test". For other environments, see the instructions in +NON-AUTOTOOLS-BUILD. + +The RunTest script runs the pcretest test program (which is documented in its +own man page) on each of the relevant testinput files in the testdata +directory, and compares the output with the contents of the corresponding +testoutput files. RunTest uses a file called testtry to hold the main output +from pcretest. Other files whose names begin with "test" are used as working +files in some tests. + +Some tests are relevant only when certain build-time options were selected. For +example, the tests for UTF-8/16/32 support are run only if --enable-utf was +used. RunTest outputs a comment when it skips a test. + +Many of the tests that are not skipped are run up to three times. The second +run forces pcre_study() to be called for all patterns except for a few in some +tests that are marked "never study" (see the pcretest program for how this is +done). If JIT support is available, the non-DFA tests are run a third time, +this time with a forced pcre_study() with the PCRE_STUDY_JIT_COMPILE option. +This testing can be suppressed by putting "nojit" on the RunTest command line. + +The entire set of tests is run once for each of the 8-bit, 16-bit and 32-bit +libraries that are enabled. If you want to run just one set of tests, call +RunTest with either the -8, -16 or -32 option. + +If valgrind is installed, you can run the tests under it by putting "valgrind" +on the RunTest command line. To run pcretest on just one or more specific test +files, give their numbers as arguments to RunTest, for example: + + RunTest 2 7 11 + +You can also specify ranges of tests such as 3-6 or 3- (meaning 3 to the +end), or a number preceded by ~ to exclude a test. For example: + + Runtest 3-15 ~10 + +This runs tests 3 to 15, excluding test 10, and just ~13 runs all the tests +except test 13. Whatever order the arguments are in, the tests are always run +in numerical order. + +You can also call RunTest with the single argument "list" to cause it to output +a list of tests. + +The first test file can be fed directly into the perltest.pl script to check +that Perl gives the same results. The only difference you should see is in the +first few lines, where the Perl version is given instead of the PCRE version. + +The second set of tests check pcre_fullinfo(), pcre_study(), +pcre_copy_substring(), pcre_get_substring(), pcre_get_substring_list(), error +detection, and run-time flags that are specific to PCRE, as well as the POSIX +wrapper API. It also uses the debugging flags to check some of the internals of +pcre_compile(). + +If you build PCRE with a locale setting that is not the standard C locale, the +character tables may be different (see next paragraph). In some cases, this may +cause failures in the second set of tests. For example, in a locale where the +isprint() function yields TRUE for characters in the range 128-255, the use of +[:isascii:] inside a character class defines a different set of characters, and +this shows up in this test as a difference in the compiled code, which is being +listed for checking. Where the comparison test output contains [\x00-\x7f] the +test will contain [\x00-\xff], and similarly in some other cases. This is not a +bug in PCRE. + +The third set of tests checks pcre_maketables(), the facility for building a +set of character tables for a specific locale and using them instead of the +default tables. The tests make use of the "fr_FR" (French) locale. Before +running the test, the script checks for the presence of this locale by running +the "locale" command. If that command fails, or if it doesn't include "fr_FR" +in the list of available locales, the third test cannot be run, and a comment +is output to say why. If running this test produces instances of the error + + ** Failed to set locale "fr_FR" + +in the comparison output, it means that locale is not available on your system, +despite being listed by "locale". This does not mean that PCRE is broken. + +[If you are trying to run this test on Windows, you may be able to get it to +work by changing "fr_FR" to "french" everywhere it occurs. Alternatively, use +RunTest.bat. The version of RunTest.bat included with PCRE 7.4 and above uses +Windows versions of test 2. More info on using RunTest.bat is included in the +document entitled NON-UNIX-USE.] + +The fourth and fifth tests check the UTF-8/16/32 support and error handling and +internal UTF features of PCRE that are not relevant to Perl, respectively. The +sixth and seventh tests do the same for Unicode character properties support. + +The eighth, ninth, and tenth tests check the pcre_dfa_exec() alternative +matching function, in non-UTF-8/16/32 mode, UTF-8/16/32 mode, and UTF-8/16/32 +mode with Unicode property support, respectively. + +The eleventh test checks some internal offsets and code size features; it is +run only when the default "link size" of 2 is set (in other cases the sizes +change) and when Unicode property support is enabled. + +The twelfth test is run only when JIT support is available, and the thirteenth +test is run only when JIT support is not available. They test some JIT-specific +features such as information output from pcretest about JIT compilation. + +The fourteenth, fifteenth, and sixteenth tests are run only in 8-bit mode, and +the seventeenth, eighteenth, and nineteenth tests are run only in 16/32-bit +mode. These are tests that generate different output in the two modes. They are +for general cases, UTF-8/16/32 support, and Unicode property support, +respectively. + +The twentieth test is run only in 16/32-bit mode. It tests some specific +16/32-bit features of the DFA matching engine. + +The twenty-first and twenty-second tests are run only in 16/32-bit mode, when +the link size is set to 2 for the 16-bit library. They test reloading +pre-compiled patterns. + +The twenty-third and twenty-fourth tests are run only in 16-bit mode. They are +for general cases, and UTF-16 support, respectively. + +The twenty-fifth and twenty-sixth tests are run only in 32-bit mode. They are +for general cases, and UTF-32 support, respectively. + + +Character tables +---------------- + +For speed, PCRE uses four tables for manipulating and identifying characters +whose code point values are less than 256. The final argument of the +pcre_compile() function is a pointer to a block of memory containing the +concatenated tables. A call to pcre_maketables() can be used to generate a set +of tables in the current locale. If the final argument for pcre_compile() is +passed as NULL, a set of default tables that is built into the binary is used. + +The source file called pcre_chartables.c contains the default set of tables. By +default, this is created as a copy of pcre_chartables.c.dist, which contains +tables for ASCII coding. However, if --enable-rebuild-chartables is specified +for ./configure, a different version of pcre_chartables.c is built by the +program dftables (compiled from dftables.c), which uses the ANSI C character +handling functions such as isalnum(), isalpha(), isupper(), islower(), etc. to +build the table sources. This means that the default C locale which is set for +your system will control the contents of these default tables. You can change +the default tables by editing pcre_chartables.c and then re-building PCRE. If +you do this, you should take care to ensure that the file does not get +automatically re-generated. The best way to do this is to move +pcre_chartables.c.dist out of the way and replace it with your customized +tables. + +When the dftables program is run as a result of --enable-rebuild-chartables, +it uses the default C locale that is set on your system. It does not pay +attention to the LC_xxx environment variables. In other words, it uses the +system's default locale rather than whatever the compiling user happens to have +set. If you really do want to build a source set of character tables in a +locale that is specified by the LC_xxx variables, you can run the dftables +program by hand with the -L option. For example: + + ./dftables -L pcre_chartables.c.special + +The first two 256-byte tables provide lower casing and case flipping functions, +respectively. The next table consists of three 32-byte bit maps which identify +digits, "word" characters, and white space, respectively. These are used when +building 32-byte bit maps that represent character classes for code points less +than 256. + +The final 256-byte table has bits indicating various character types, as +follows: + + 1 white space character + 2 letter + 4 decimal digit + 8 hexadecimal digit + 16 alphanumeric or '_' + 128 regular expression metacharacter or binary zero + +You should not alter the set of characters that contain the 128 bit, as that +will cause PCRE to malfunction. + + +File manifest +------------- + +The distribution should contain the files listed below. Where a file name is +given as pcre[16|32]_xxx it means that there are three files, one with the name +pcre_xxx, one with the name pcre16_xx, and a third with the name pcre32_xxx. + +(A) Source files of the PCRE library functions and their headers: + + dftables.c auxiliary program for building pcre_chartables.c + when --enable-rebuild-chartables is specified + + pcre_chartables.c.dist a default set of character tables that assume ASCII + coding; used, unless --enable-rebuild-chartables is + specified, by copying to pcre[16]_chartables.c + + pcreposix.c ) + pcre[16|32]_byte_order.c ) + pcre[16|32]_compile.c ) + pcre[16|32]_config.c ) + pcre[16|32]_dfa_exec.c ) + pcre[16|32]_exec.c ) + pcre[16|32]_fullinfo.c ) + pcre[16|32]_get.c ) sources for the functions in the library, + pcre[16|32]_globals.c ) and some internal functions that they use + pcre[16|32]_jit_compile.c ) + pcre[16|32]_maketables.c ) + pcre[16|32]_newline.c ) + pcre[16|32]_refcount.c ) + pcre[16|32]_string_utils.c ) + pcre[16|32]_study.c ) + pcre[16|32]_tables.c ) + pcre[16|32]_ucd.c ) + pcre[16|32]_version.c ) + pcre[16|32]_xclass.c ) + pcre_ord2utf8.c ) + pcre_valid_utf8.c ) + pcre16_ord2utf16.c ) + pcre16_utf16_utils.c ) + pcre16_valid_utf16.c ) + pcre32_utf32_utils.c ) + pcre32_valid_utf32.c ) + + pcre[16|32]_printint.c ) debugging function that is used by pcretest, + ) and can also be #included in pcre_compile() + + pcre.h.in template for pcre.h when built by "configure" + pcreposix.h header for the external POSIX wrapper API + pcre_internal.h header for internal use + sljit/* 16 files that make up the JIT compiler + ucp.h header for Unicode property handling + + config.h.in template for config.h, which is built by "configure" + + pcrecpp.h public header file for the C++ wrapper + pcrecpparg.h.in template for another C++ header file + pcre_scanner.h public header file for C++ scanner functions + pcrecpp.cc ) + pcre_scanner.cc ) source for the C++ wrapper library + + pcre_stringpiece.h.in template for pcre_stringpiece.h, the header for the + C++ stringpiece functions + pcre_stringpiece.cc source for the C++ stringpiece functions + +(B) Source files for programs that use PCRE: + + pcredemo.c simple demonstration of coding calls to PCRE + pcregrep.c source of a grep utility that uses PCRE + pcretest.c comprehensive test program + +(C) Auxiliary files: + + 132html script to turn "man" pages into HTML + AUTHORS information about the author of PCRE + ChangeLog log of changes to the code + CleanTxt script to clean nroff output for txt man pages + Detrail script to remove trailing spaces + HACKING some notes about the internals of PCRE + INSTALL generic installation instructions + LICENCE conditions for the use of PCRE + COPYING the same, using GNU's standard name + Makefile.in ) template for Unix Makefile, which is built by + ) "configure" + Makefile.am ) the automake input that was used to create + ) Makefile.in + NEWS important changes in this release + NON-UNIX-USE the previous name for NON-AUTOTOOLS-BUILD + NON-AUTOTOOLS-BUILD notes on building PCRE without using autotools + PrepareRelease script to make preparations for "make dist" + README this file + RunTest a Unix shell script for running tests + RunGrepTest a Unix shell script for pcregrep tests + aclocal.m4 m4 macros (generated by "aclocal") + config.guess ) files used by libtool, + config.sub ) used only when building a shared library + configure a configuring shell script (built by autoconf) + configure.ac ) the autoconf input that was used to build + ) "configure" and config.h + depcomp ) script to find program dependencies, generated by + ) automake + doc/*.3 man page sources for PCRE + doc/*.1 man page sources for pcregrep and pcretest + doc/index.html.src the base HTML page + doc/html/* HTML documentation + doc/pcre.txt plain text version of the man pages + doc/pcretest.txt plain text documentation of test program + doc/perltest.txt plain text documentation of Perl test program + install-sh a shell script for installing files + libpcre16.pc.in template for libpcre16.pc for pkg-config + libpcre32.pc.in template for libpcre32.pc for pkg-config + libpcre.pc.in template for libpcre.pc for pkg-config + libpcreposix.pc.in template for libpcreposix.pc for pkg-config + libpcrecpp.pc.in template for libpcrecpp.pc for pkg-config + ltmain.sh file used to build a libtool script + missing ) common stub for a few missing GNU programs while + ) installing, generated by automake + mkinstalldirs script for making install directories + perltest.pl Perl test program + pcre-config.in source of script which retains PCRE information + pcre_jit_test.c test program for the JIT compiler + pcrecpp_unittest.cc ) + pcre_scanner_unittest.cc ) test programs for the C++ wrapper + pcre_stringpiece_unittest.cc ) + testdata/testinput* test data for main library tests + testdata/testoutput* expected test results + testdata/grep* input and output for pcregrep tests + testdata/* other supporting test files + +(D) Auxiliary files for cmake support + + cmake/COPYING-CMAKE-SCRIPTS + cmake/FindPackageHandleStandardArgs.cmake + cmake/FindEditline.cmake + cmake/FindReadline.cmake + CMakeLists.txt + config-cmake.h.in + +(E) Auxiliary files for VPASCAL + + makevp.bat + makevp_c.txt + makevp_l.txt + pcregexp.pas + +(F) Auxiliary files for building PCRE "by hand" + + pcre.h.generic ) a version of the public PCRE header file + ) for use in non-"configure" environments + config.h.generic ) a version of config.h for use in non-"configure" + ) environments + +(F) Miscellaneous + + RunTest.bat a script for running tests under Windows + +Philip Hazel +Email local part: ph10 +Email domain: cam.ac.uk +Last updated: 28 April 2013 diff --git a/pcre/RunGrepTest b/pcre/RunGrepTest new file mode 100755 index 00000000000..daaf8af47ac --- /dev/null +++ b/pcre/RunGrepTest @@ -0,0 +1,567 @@ +#! /bin/sh + +# Run pcregrep tests. The assumption is that the PCRE tests check the library +# itself. What we are checking here is the file handling and options that are +# supported by pcregrep. This script must be run in the build directory. + +# Set the C locale, so that sort(1) behaves predictably. + +LC_ALL=C +export LC_ALL + +# Remove any non-default colouring and aliases that the caller may have set. + +unset PCREGREP_COLOUR PCREGREP_COLOR +unset cp ls mv rm + +# Remember the current (build) directory, set the program to be tested, and +# valgrind settings when requested. + +builddir=`pwd` +pcregrep=$builddir/pcregrep + +valgrind= +while [ $# -gt 0 ] ; do + case $1 in + valgrind) valgrind="valgrind -q --leak-check=no --smc-check=all";; + *) echo "RunGrepTest: Unknown argument $1"; exit 1;; + esac + shift +done + +echo " " +pcregrep_version=`$pcregrep -V` +if [ "$valgrind" = "" ] ; then + echo "Testing $pcregrep_version" +else + echo "Testing $pcregrep_version using valgrind" +fi + +# Set up a suitable "diff" command for comparison. Some systems have a diff +# that lacks a -u option. Try to deal with this; better do the test for the -b +# option as well. + +cf="diff" +diff -b /dev/null /dev/null 2>/dev/null && cf="diff -b" +diff -u /dev/null /dev/null 2>/dev/null && cf="diff -u" +diff -ub /dev/null /dev/null 2>/dev/null && cf="diff -ub" + +# If this test is being run from "make check", $srcdir will be set. If not, set +# it to the current or parent directory, whichever one contains the test data. +# Subsequently, we run most of the pcregrep tests in the source directory so +# that the file names in the output are always the same. + +if [ -z "$srcdir" -o ! -d "$srcdir/testdata" ] ; then + if [ -d "./testdata" ] ; then + srcdir=. + elif [ -d "../testdata" ] ; then + srcdir=.. + else + echo "Cannot find the testdata directory" + exit 1 + fi +fi + +# Check for the availability of UTF-8 support + +./pcretest -C utf >/dev/null +utf8=$? + +echo "Testing pcregrep main features" + +echo "---------------------------- Test 1 ------------------------------" >testtry +(cd $srcdir; $valgrind $pcregrep PATTERN ./testdata/grepinput) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 2 ------------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep '^PATTERN' ./testdata/grepinput) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 3 ------------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -in PATTERN ./testdata/grepinput) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 4 ------------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -ic PATTERN ./testdata/grepinput) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 5 ------------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -in PATTERN ./testdata/grepinput ./testdata/grepinputx) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 6 ------------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -inh PATTERN ./testdata/grepinput ./testdata/grepinputx) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 7 ------------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -il PATTERN ./testdata/grepinput ./testdata/grepinputx) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 8 ------------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -l PATTERN ./testdata/grepinput ./testdata/grepinputx) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 9 ------------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -q PATTERN ./testdata/grepinput ./testdata/grepinputx) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 10 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -q NEVER-PATTERN ./testdata/grepinput ./testdata/grepinputx) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 11 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -vn pattern ./testdata/grepinputx) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 12 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -ix pattern ./testdata/grepinputx) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 13 -----------------------------" >>testtry +echo seventeen >testtemp1 +(cd $srcdir; $valgrind $pcregrep -f./testdata/greplist -f $builddir/testtemp1 ./testdata/grepinputx) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 14 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -w pat ./testdata/grepinput ./testdata/grepinputx) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 15 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep 'abc^*' ./testdata/grepinput) 2>>testtry >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 16 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep abc ./testdata/grepinput ./testdata/nonexistfile) 2>>testtry >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 17 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -M 'the\noutput' ./testdata/grepinput) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 18 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -Mn '(the\noutput|dog\.\n--)' ./testdata/grepinput) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 19 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -Mix 'Pattern' ./testdata/grepinputx) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 20 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -Mixn 'complete pair\nof lines' ./testdata/grepinputx) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 21 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -nA3 'four' ./testdata/grepinputx) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 22 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -nB3 'four' ./testdata/grepinputx) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 23 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -C3 'four' ./testdata/grepinputx) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 24 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -A9 'four' ./testdata/grepinputx) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 25 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -nB9 'four' ./testdata/grepinputx) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 26 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -A9 -B9 'four' ./testdata/grepinputx) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 27 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -A10 'four' ./testdata/grepinputx) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 28 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -nB10 'four' ./testdata/grepinputx) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 29 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -C12 -B10 'four' ./testdata/grepinputx) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 30 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -inB3 'pattern' ./testdata/grepinput ./testdata/grepinputx) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 31 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -inA3 'pattern' ./testdata/grepinput ./testdata/grepinputx) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 32 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -L 'fox' ./testdata/grepinput ./testdata/grepinputx) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 33 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep 'fox' ./testdata/grepnonexist) >>testtry 2>&1 +echo "RC=$?" >>testtry + +echo "---------------------------- Test 34 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -s 'fox' ./testdata/grepnonexist) >>testtry 2>&1 +echo "RC=$?" >>testtry + +echo "---------------------------- Test 35 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -L -r --include=grepinputx --include grepinput8 --exclude-dir='^\.' 'fox' ./testdata | sort) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 36 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -L -r --include=grepinput --exclude 'grepinput$' --exclude=grepinput8 --exclude-dir='^\.' 'fox' ./testdata | sort) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 37 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep '^(a+)*\d' ./testdata/grepinput) >>testtry 2>teststderr +echo "RC=$?" >>testtry +echo "======== STDERR ========" >>testtry +cat teststderr >>testtry + +echo "---------------------------- Test 38 ------------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep '>\x00<' ./testdata/grepinput) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 39 ------------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -A1 'before the binary zero' ./testdata/grepinput) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 40 ------------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -B1 'after the binary zero' ./testdata/grepinput) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 41 ------------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -B1 -o '\w+ the binary zero' ./testdata/grepinput) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 42 ------------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -B1 -onH '\w+ the binary zero' ./testdata/grepinput) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 43 ------------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -on 'before|zero|after' ./testdata/grepinput) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 44 ------------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -on -e before -ezero -e after ./testdata/grepinput) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 45 ------------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -on -f ./testdata/greplist -e binary ./testdata/grepinput) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 46 ------------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -eabc -e '(unclosed' ./testdata/grepinput) 2>>testtry >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 47 ------------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -Fx "AB.VE +elephant" ./testdata/grepinput) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 48 ------------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -F "AB.VE +elephant" ./testdata/grepinput) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 49 ------------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -F -e DATA -e "AB.VE +elephant" ./testdata/grepinput) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 50 ------------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep "^(abc|def|ghi|jkl)" ./testdata/grepinputx) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 51 ------------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -Mv "brown\sfox" ./testdata/grepinputv) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 52 ------------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep --colour=always jumps ./testdata/grepinputv) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 53 ------------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep --file-offsets 'before|zero|after' ./testdata/grepinput) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 54 ------------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep --line-offsets 'before|zero|after' ./testdata/grepinput) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 55 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -f./testdata/greplist --color=always ./testdata/grepinputx) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 56 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -c lazy ./testdata/grepinput*) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 57 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -c -l lazy ./testdata/grepinput*) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 58 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep --regex=PATTERN ./testdata/grepinput) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 59 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep --regexp=PATTERN ./testdata/grepinput) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 60 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep --regex PATTERN ./testdata/grepinput) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 61 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep --regexp PATTERN ./testdata/grepinput) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 62 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep --match-limit=1000 --no-jit -M 'This is a file(.|\R)*file.' ./testdata/grepinput) >>testtry 2>&1 +echo "RC=$?" >>testtry + +echo "---------------------------- Test 63 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep --recursion-limit=1000 --no-jit -M 'This is a file(.|\R)*file.' ./testdata/grepinput) >>testtry 2>&1 +echo "RC=$?" >>testtry + +echo "---------------------------- Test 64 ------------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -o1 '(?<=PAT)TERN (ap(pear)s)' ./testdata/grepinput) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 65 ------------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -o2 '(?<=PAT)TERN (ap(pear)s)' ./testdata/grepinput) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 66 ------------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -o3 '(?<=PAT)TERN (ap(pear)s)' ./testdata/grepinput) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 67 ------------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -o12 '(?<=PAT)TERN (ap(pear)s)' ./testdata/grepinput) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 68 ------------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep --only-matching=2 '(?<=PAT)TERN (ap(pear)s)' ./testdata/grepinput) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 69 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -vn --colour=always pattern ./testdata/grepinputx) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 70 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep --color=always -M "triple:\t.*\n\n" ./testdata/grepinput3) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 71 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -o "^01|^02|^03" ./testdata/grepinput) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 72 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep --color=always "^01|^02|^03" ./testdata/grepinput) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 73 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -o --colour=always "^01|^02|^03" ./testdata/grepinput) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 74 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -o "^01|02|^03" ./testdata/grepinput) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 75 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep --color=always "^01|02|^03" ./testdata/grepinput) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 76 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -o --colour=always "^01|02|^03" ./testdata/grepinput) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 77 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -o "^01|^02|03" ./testdata/grepinput) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 78 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep --color=always "^01|^02|03" ./testdata/grepinput) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 79 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -o --colour=always "^01|^02|03" ./testdata/grepinput) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 80 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -o "\b01|\b02" ./testdata/grepinput) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 81 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep --color=always "\\b01|\\b02" ./testdata/grepinput) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 82 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -o --colour=always "\\b01|\\b02" ./testdata/grepinput) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 83 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep --buffer-size=100 "^a" ./testdata/grepinput3) >>testtry 2>&1 +echo "RC=$?" >>testtry + +echo "---------------------------- Test 84 -----------------------------" >>testtry +echo testdata/grepinput3 >testtemp1 +(cd $srcdir; $valgrind $pcregrep --file-list ./testdata/grepfilelist --file-list $builddir/testtemp1 "fox|complete|t7") >>testtry 2>&1 +echo "RC=$?" >>testtry + +echo "---------------------------- Test 85 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep --file-list=./testdata/grepfilelist "dolor" ./testdata/grepinput3) >>testtry 2>&1 +echo "RC=$?" >>testtry + +echo "---------------------------- Test 86 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep "dog" ./testdata/grepbinary) >>testtry 2>&1 +echo "RC=$?" >>testtry + +echo "---------------------------- Test 87 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep "cat" ./testdata/grepbinary) >>testtry 2>&1 +echo "RC=$?" >>testtry + +echo "---------------------------- Test 88 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -v "cat" ./testdata/grepbinary) >>testtry 2>&1 +echo "RC=$?" >>testtry + +echo "---------------------------- Test 89 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -I "dog" ./testdata/grepbinary) >>testtry 2>&1 +echo "RC=$?" >>testtry + +echo "---------------------------- Test 90 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep --binary-files=without-match "dog" ./testdata/grepbinary) >>testtry 2>&1 +echo "RC=$?" >>testtry + +echo "---------------------------- Test 91 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -a "dog" ./testdata/grepbinary) >>testtry 2>&1 +echo "RC=$?" >>testtry + +echo "---------------------------- Test 92 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep --binary-files=text "dog" ./testdata/grepbinary) >>testtry 2>&1 +echo "RC=$?" >>testtry + +echo "---------------------------- Test 93 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep --text "dog" ./testdata/grepbinary) >>testtry 2>&1 +echo "RC=$?" >>testtry + +echo "---------------------------- Test 94 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -L -r --include=grepinputx --include grepinput8 'fox' ./testdata/grepinput* | sort) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 95 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep --file-list ./testdata/grepfilelist --exclude grepinputv "fox|complete") >>testtry 2>&1 +echo "RC=$?" >>testtry + +echo "---------------------------- Test 96 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -L -r --include-dir=testdata --exclude '^(?!grepinput)' 'fox' ./test* | sort) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 97 -----------------------------" >>testtry +echo "grepinput$" >testtemp1 +echo "grepinput8" >>testtemp1 +(cd $srcdir; $valgrind $pcregrep -L -r --include=grepinput --exclude-from $builddir/testtemp1 --exclude-dir='^\.' 'fox' ./testdata | sort) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 98 -----------------------------" >>testtry +echo "grepinput$" >testtemp1 +echo "grepinput8" >>testtemp1 +(cd $srcdir; $valgrind $pcregrep -L -r --exclude=grepinput3 --include=grepinput --exclude-from $builddir/testtemp1 --exclude-dir='^\.' 'fox' ./testdata | sort) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 99 -----------------------------" >>testtry +echo "grepinput$" >testtemp1 +echo "grepinput8" >testtemp2 +(cd $srcdir; $valgrind $pcregrep -L -r --include grepinput --exclude-from $builddir/testtemp1 --exclude-from=$builddir/testtemp2 --exclude-dir='^\.' 'fox' ./testdata | sort) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 100 ------------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -Ho2 --only-matching=1 -o3 '(\w+) binary (\w+)(\.)?' ./testdata/grepinput) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 101 ------------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -o3 -Ho2 -o12 --only-matching=1 -o3 --colour=always --om-separator='|' '(\w+) binary (\w+)(\.)?' ./testdata/grepinput) >>testtry +echo "RC=$?" >>testtry + +echo "---------------------------- Test 102 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -n "^$" ./testdata/grepinput3) >>testtry 2>&1 +echo "RC=$?" >>testtry + +echo "---------------------------- Test 103 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep --only-matching "^$" ./testdata/grepinput3) >>testtry 2>&1 +echo "RC=$?" >>testtry + +echo "---------------------------- Test 104 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep -n --only-matching "^$" ./testdata/grepinput3) >>testtry 2>&1 +echo "RC=$?" >>testtry + +echo "---------------------------- Test 105 -----------------------------" >>testtry +(cd $srcdir; $valgrind $pcregrep --colour=always "ipsum|" ./testdata/grepinput3) >>testtry 2>&1 +echo "RC=$?" >>testtry + + +# Now compare the results. + +$cf $srcdir/testdata/grepoutput testtry +if [ $? != 0 ] ; then exit 1; fi + + +# These tests require UTF-8 support + +if [ $utf8 -ne 0 ] ; then + echo "Testing pcregrep UTF-8 features" + + echo "---------------------------- Test U1 ------------------------------" >testtry + (cd $srcdir; $valgrind $pcregrep -n -u --newline=any "^X" ./testdata/grepinput8) >>testtry + echo "RC=$?" >>testtry + + echo "---------------------------- Test U2 ------------------------------" >>testtry + (cd $srcdir; $valgrind $pcregrep -n -u -C 3 --newline=any "Match" ./testdata/grepinput8) >>testtry + echo "RC=$?" >>testtry + + $cf $srcdir/testdata/grepoutput8 testtry + if [ $? != 0 ] ; then exit 1; fi + +else + echo "Skipping pcregrep UTF-8 tests: no UTF-8 support in PCRE library" +fi + + +# We go to some contortions to try to ensure that the tests for the various +# newline settings will work in environments where the normal newline sequence +# is not \n. Do not use exported files, whose line endings might be changed. +# Instead, create an input file using printf so that its contents are exactly +# what we want. Note the messy fudge to get printf to write a string that +# starts with a hyphen. These tests are run in the build directory. + +echo "Testing pcregrep newline settings" +printf "abc\rdef\r\nghi\njkl" >testNinput + +printf "%c--------------------------- Test N1 ------------------------------\r\n" - >testtry +$valgrind $pcregrep -n -N CR "^(abc|def|ghi|jkl)" testNinput >>testtry + +printf "%c--------------------------- Test N2 ------------------------------\r\n" - >>testtry +$valgrind $pcregrep -n --newline=crlf "^(abc|def|ghi|jkl)" testNinput >>testtry + +printf "%c--------------------------- Test N3 ------------------------------\r\n" - >>testtry +pattern=`printf 'def\rjkl'` +$valgrind $pcregrep -n --newline=cr -F "$pattern" testNinput >>testtry + +printf "%c--------------------------- Test N4 ------------------------------\r\n" - >>testtry +$valgrind $pcregrep -n --newline=crlf -F -f $srcdir/testdata/greppatN4 testNinput >>testtry + +printf "%c--------------------------- Test N5 ------------------------------\r\n" - >>testtry +$valgrind $pcregrep -n --newline=any "^(abc|def|ghi|jkl)" testNinput >>testtry + +printf "%c--------------------------- Test N6 ------------------------------\r\n" - >>testtry +$valgrind $pcregrep -n --newline=anycrlf "^(abc|def|ghi|jkl)" testNinput >>testtry + +$cf $srcdir/testdata/grepoutputN testtry +if [ $? != 0 ] ; then exit 1; fi + +exit 0 + +# End diff --git a/pcre/RunTest b/pcre/RunTest new file mode 100755 index 00000000000..505a0bf14b5 --- /dev/null +++ b/pcre/RunTest @@ -0,0 +1,1063 @@ +#! /bin/sh + +############################################################################### +# Run the PCRE tests using the pcretest program. The appropriate tests are +# selected, depending on which build-time options were used. +# +# All tests are now run both with and without -s, to ensure that everything is +# tested with and without studying. However, there are some tests that produce +# different output after studying, typically when we are tracing the actual +# matching process (for example, using auto-callouts). In these few cases, the +# tests are duplicated in the files, one with /S to force studying always, and +# one with /SS to force *not* studying always. The use of -s doesn't then make +# any difference to their output. There is also one test which compiles invalid +# UTF-8 with the UTF-8 check turned off; for this, studying must also be +# disabled with /SS. +# +# When JIT support is available, all the tests are also run with -s+ to test +# (again, almost) everything with studying and the JIT option, unless "nojit" +# is given on the command line. There are also two tests for JIT-specific +# features, one to be run when JIT support is available (unless "nojit" is +# specified), and one when it is not. +# +# Whichever of the 8-, 16- and 32-bit libraries exist are tested. It is also +# possible to select which to test by giving "-8", "-16" or "-32" on the +# command line. +# +# As well as "nojit", "-8", "-16", and "-32", arguments for this script are +# individual test numbers, ranges of tests such as 3-6 or 3- (meaning 3 to the +# end), or a number preceded by ~ to exclude a test. For example, "3-15 ~10" +# runs tests 3 to 15, excluding test 10, and just "~10" runs all the tests +# except test 10. Whatever order the arguments are in, the tests are always run +# in numerical order. + +# Other arguments can be one of the words "valgrind", "valgrind-log", or "sim" +# followed by an argument to run cross- compiled executables under a simulator, +# for example: +# +# RunTest 3 sim "qemu-arm -s 8388608" +# +# There are two special cases where only one argument is allowed: +# +# If the first and only argument is "ebcdic", the script runs the special +# EBCDIC test that can be useful for checking certain EBCDIC features, even +# when run in an ASCII environment. +# +# If the script is obeyed as "RunTest list", a list of available tests is +# output, but none of them are run. +############################################################################### + +# Define test titles in variables so that they can be output as a list. Some +# of them are modified (e.g. with -8 or -16) when used in the actual tests. + +title1="Test 1: Main functionality (Compatible with Perl >= 5.10)" +title2="Test 2: API, errors, internals, and non-Perl stuff" +title3="Test 3: Locale-specific features" +title4A="Test 4: UTF" +title4B=" support (Compatible with Perl >= 5.10)" +title5="Test 5: API, internals, and non-Perl stuff for UTF" +title6="Test 6: Unicode property support (Compatible with Perl >= 5.10)" +title7="Test 7: API, internals, and non-Perl stuff for Unicode property support" +title8="Test 8: DFA matching main functionality" +title9="Test 9: DFA matching with UTF" +title10="Test 10: DFA matching with Unicode properties" +title11="Test 11: Internal offsets and code size tests" +title12="Test 12: JIT-specific features (JIT available)" +title13="Test 13: JIT-specific features (JIT not available)" +title14="Test 14: Specials for the basic 8-bit library" +title15="Test 15: Specials for the 8-bit library with UTF-8 support" +title16="Test 16: Specials for the 8-bit library with Unicode propery support" +title17="Test 17: Specials for the basic 16/32-bit library" +title18="Test 18: Specials for the 16/32-bit library with UTF-16/32 support" +title19="Test 19: Specials for the 16/32-bit library with Unicode property support" +title20="Test 20: DFA specials for the basic 16/32-bit library" +title21="Test 21: Reloads for the basic 16/32-bit library" +title22="Test 22: Reloads for the 16/32-bit library with UTF-16/32 support" +title23="Test 23: Specials for the 16-bit library" +title24="Test 24: Specials for the 16-bit library with UTF-16 support" +title25="Test 25: Specials for the 32-bit library" +title26="Test 26: Specials for the 32-bit library with UTF-32 support" + +maxtest=26 + +if [ $# -eq 1 -a "$1" = "list" ]; then + echo $title1 + echo $title2 "(not UTF)" + echo $title3 + echo $title4A $title4B + echo $title5 support + echo $title6 + echo $title7 + echo $title8 + echo $title9 + echo $title10 + echo $title11 + echo $title12 + echo $title13 + echo $title14 + echo $title15 + echo $title16 + echo $title17 + echo $title18 + echo $title19 + echo $title20 + echo $title21 + echo $title22 + echo $title23 + echo $title24 + echo $title25 + echo $title26 + exit 0 +fi + +# Set up a suitable "diff" command for comparison. Some systems +# have a diff that lacks a -u option. Try to deal with this. + +cf="diff" +diff -u /dev/null /dev/null 2>/dev/null && cf="diff -u" + +# Find the test data + +if [ -n "$srcdir" -a -d "$srcdir" ] ; then + testdata="$srcdir/testdata" +elif [ -d "./testdata" ] ; then + testdata=./testdata +elif [ -d "../testdata" ] ; then + testdata=../testdata +else + echo "Cannot find the testdata directory" + exit 1 +fi + + +# ------ Special EBCDIC Test ------- + +if [ $# -eq 1 -a "$1" = "ebcdic" ]; then + ./pcretest -C ebcdic >/dev/null + ebcdic=$? + if [ $ebcdic -ne 1 ] ; then + echo "Cannot run EBCDIC tests: EBCDIC support not compiled" + exit 1 + fi + + for opt in "" "-s" "-dfa" "-s -dfa"; do + ./pcretest -q $opt $testdata/testinputEBC >testtry + if [ $? = 0 ] ; then + $cf $testdata/testoutputEBC testtry + if [ $? != 0 ] ; then exit 1; fi + else exit 1 + fi + if [ "$opt" = "-s" ] ; then echo " OK with study" + elif [ "$opt" = "-dfa" ] ; then echo " OK using DFA" + elif [ "$opt" = "-s -dfa" ] ; then echo " OK using DFA with study" + else echo " OK" + fi + done + +exit 0 +fi + + +# ------ Normal Tests ------ + +# Default values + +arg8= +arg16= +arg32= +nojit= +sim= +skip= +valgrind= + +# This is in case the caller has set aliases (as I do - PH) +unset cp ls mv rm + +# Process options and select which tests to run; for those that are explicitly +# requested, check that the necessary optional facilities are available. + +do1=no +do2=no +do3=no +do4=no +do5=no +do6=no +do7=no +do8=no +do9=no +do10=no +do11=no +do12=no +do13=no +do14=no +do15=no +do16=no +do17=no +do18=no +do19=no +do20=no +do21=no +do22=no +do23=no +do24=no +do25=no +do26=no + +while [ $# -gt 0 ] ; do + case $1 in + 1) do1=yes;; + 2) do2=yes;; + 3) do3=yes;; + 4) do4=yes;; + 5) do5=yes;; + 6) do6=yes;; + 7) do7=yes;; + 8) do8=yes;; + 9) do9=yes;; + 10) do10=yes;; + 11) do11=yes;; + 12) do12=yes;; + 13) do13=yes;; + 14) do14=yes;; + 15) do15=yes;; + 16) do16=yes;; + 17) do17=yes;; + 18) do18=yes;; + 19) do19=yes;; + 20) do20=yes;; + 21) do21=yes;; + 22) do22=yes;; + 23) do23=yes;; + 24) do24=yes;; + 25) do25=yes;; + 26) do26=yes;; + -8) arg8=yes;; + -16) arg16=yes;; + -32) arg32=yes;; + nojit) nojit=yes;; + sim) shift; sim=$1;; + valgrind) valgrind="valgrind --tool=memcheck -q --smc-check=all";; + valgrind-log) valgrind="valgrind --tool=memcheck --num-callers=30 --leak-check=no --error-limit=no --smc-check=all --log-file=report.%p ";; + ~*) + if expr "$1" : '~[0-9][0-9]*$' >/dev/null; then + skip="$skip `expr "$1" : '~\([0-9]*\)*$'`" + else + echo "Unknown option or test selector '$1'"; exit 1 + fi + ;; + *-*) + if expr "$1" : '[0-9][0-9]*-[0-9]*$' >/dev/null; then + tf=`expr "$1" : '\([0-9]*\)'` + tt=`expr "$1" : '.*-\([0-9]*\)'` + if [ "$tt" = "" ] ; then tt=$maxtest; fi + if expr \( "$tf" "<" 1 \) \| \( "$tt" ">" "$maxtest" \) >/dev/null; then + echo "Invalid test range '$1'"; exit 1 + fi + while expr "$tf" "<=" "$tt" >/dev/null; do + eval do${tf}=yes + tf=`expr $tf + 1` + done + else + echo "Invalid test range '$1'"; exit 1 + fi + ;; + *) echo "Unknown option or test selector '$1'"; exit 1;; + esac + shift +done + +# Find which optional facilities are available. + +$sim ./pcretest -C linksize >/dev/null +link_size=$? +if [ $link_size -lt 2 ] ; then + echo "Failed to find internal link size" + exit 1 +fi +if [ $link_size -gt 4 ] ; then + echo "Failed to find internal link size" + exit 1 +fi + +# All of 8-bit, 16-bit, and 32-bit character strings may be supported, but only +# one need be. + +$sim ./pcretest -C pcre8 >/dev/null +support8=$? +$sim ./pcretest -C pcre16 >/dev/null +support16=$? +$sim ./pcretest -C pcre32 >/dev/null +support32=$? + +# Initialize all bitsizes skipped + +test8=skip +test16=skip +test32=skip + +# If no bitsize arguments, select all that are available + +if [ "$arg8$arg16$arg32" = "" ] ; then + if [ $support8 -ne 0 ] ; then + test8= + fi + if [ $support16 -ne 0 ] ; then + test16=-16 + fi + if [ $support32 -ne 0 ] ; then + test32=-32 + fi + +# Select requested bit sizes + +else + if [ "$arg8" = yes ] ; then + if [ $support8 -eq 0 ] ; then + echo "Cannot run 8-bit library tests: 8-bit library not compiled" + exit 1 + fi + test8= + fi + if [ "$arg16" = yes ] ; then + if [ $support16 -eq 0 ] ; then + echo "Cannot run 16-bit library tests: 16-bit library not compiled" + exit 1 + fi + test16=-16 + fi + if [ "$arg32" = yes ] ; then + if [ $support32 -eq 0 ] ; then + echo "Cannot run 32-bit library tests: 32-bit library not compiled" + exit 1 + fi + test32=-32 + fi +fi + +# UTF support always applies to all bit sizes if both are supported; we can't +# have UTF-8 support without UTF-16 support (for example). + +$sim ./pcretest -C utf >/dev/null +utf=$? + +$sim ./pcretest -C ucp >/dev/null +ucp=$? + +jitopt= +$sim ./pcretest -C jit >/dev/null +jit=$? +if [ $jit -ne 0 -a "$nojit" != "yes" ] ; then + jitopt=-s+ +fi + +# Handle any explicit skips + +for i in $skip; do eval do$i=no; done + +# If any unsuitable tests were explicitly requested, grumble. + +if [ $utf -eq 0 ] ; then + if [ $do4 = yes ] ; then + echo "Can't run test 4 because UTF support is not configured" + exit 1 + fi + if [ $do5 = yes ] ; then + echo "Can't run test 5 because UTF support is not configured" + exit 1 + fi + if [ $do9 = yes ] ; then + echo "Can't run test 8 because UTF support is not configured" + exit 1 + fi + if [ $do15 = yes ] ; then + echo "Can't run test 15 because UTF support is not configured" + exit 1 + fi + if [ $do18 = yes ] ; then + echo "Can't run test 18 because UTF support is not configured" + fi + if [ $do22 = yes ] ; then + echo "Can't run test 22 because UTF support is not configured" + fi +fi + +if [ $ucp -eq 0 ] ; then + if [ $do6 = yes ] ; then + echo "Can't run test 6 because Unicode property support is not configured" + exit 1 + fi + if [ $do7 = yes ] ; then + echo "Can't run test 7 because Unicode property support is not configured" + exit 1 + fi + if [ $do10 = yes ] ; then + echo "Can't run test 10 because Unicode property support is not configured" + exit 1 + fi + if [ $do16 = yes ] ; then + echo "Can't run test 16 because Unicode property support is not configured" + exit 1 + fi + if [ $do19 = yes ] ; then + echo "Can't run test 19 because Unicode property support is not configured" + exit 1 + fi +fi + +if [ $link_size -ne 2 ] ; then + if [ $do11 = yes ] ; then + echo "Can't run test 11 because the link size ($link_size) is not 2" + exit 1 + fi +fi + +if [ $jit -eq 0 ] ; then + if [ $do12 = "yes" ] ; then + echo "Can't run test 12 because JIT support is not configured" + exit 1 + fi +else + if [ $do13 = "yes" ] ; then + echo "Can't run test 13 because JIT support is configured" + exit 1 + fi +fi + +# If no specific tests were requested, select all. Those that are not +# relevant will be automatically skipped. + +if [ $do1 = no -a $do2 = no -a $do3 = no -a $do4 = no -a \ + $do5 = no -a $do6 = no -a $do7 = no -a $do8 = no -a \ + $do9 = no -a $do10 = no -a $do11 = no -a $do12 = no -a \ + $do13 = no -a $do14 = no -a $do15 = no -a $do16 = no -a \ + $do17 = no -a $do18 = no -a $do19 = no -a $do20 = no -a \ + $do21 = no -a $do22 = no -a $do23 = no -a $do24 = no -a \ + $do25 = no -a $do26 = no ] ; then + do1=yes + do2=yes + do3=yes + do4=yes + do5=yes + do6=yes + do7=yes + do8=yes + do9=yes + do10=yes + do11=yes + do12=yes + do13=yes + do14=yes + do15=yes + do16=yes + do17=yes + do18=yes + do19=yes + do20=yes + do21=yes + do22=yes + do23=yes + do24=yes + do25=yes + do26=yes +fi + +# Handle any explicit skips (again, so that an argument list may consist only +# of explicit skips). + +for i in $skip; do eval do$i=no; done + +# Show which release and which test data + +echo "" +echo PCRE C library tests using test data from $testdata +$sim ./pcretest /dev/null + +for bmode in "$test8" "$test16" "$test32"; do + case "$bmode" in + skip) continue;; + -16) if [ "$test8$test32" != "skipskip" ] ; then echo ""; fi + bits=16; echo "---- Testing 16-bit library ----"; echo "";; + -32) if [ "$test8$test16" != "skipskip" ] ; then echo ""; fi + bits=32; echo "---- Testing 32-bit library ----"; echo "";; + *) bits=8; echo "---- Testing 8-bit library ----"; echo "";; + esac + +# Primary test, compatible with JIT and all versions of Perl >= 5.8 + +if [ $do1 = yes ] ; then + echo $title1 + for opt in "" "-s" $jitopt; do + $sim $valgrind ./pcretest -q $bmode $opt $testdata/testinput1 testtry + if [ $? = 0 ] ; then + $cf $testdata/testoutput1 testtry + if [ $? != 0 ] ; then exit 1; fi + else exit 1 + fi + if [ "$opt" = "-s" ] ; then echo " OK with study" + elif [ "$opt" = "-s+" ] ; then echo " OK with JIT study" + else echo " OK" + fi + done +fi + +# PCRE tests that are not JIT or Perl-compatible: API, errors, internals + +if [ $do2 = yes ] ; then + echo $title2 "(not UTF-$bits)" + for opt in "" "-s" $jitopt; do + $sim $valgrind ./pcretest -q $bmode $opt $testdata/testinput2 testtry + if [ $? = 0 ] ; then + $cf $testdata/testoutput2 testtry + if [ $? != 0 ] ; then exit 1; fi + else + echo " " + echo "** Test 2 requires a lot of stack. If it has crashed with a" + echo "** segmentation fault, it may be that you do not have enough" + echo "** stack available by default. Please see the 'pcrestack' man" + echo "** page for a discussion of PCRE's stack usage." + echo " " + exit 1 + fi + if [ "$opt" = "-s" ] ; then echo " OK with study" + elif [ "$opt" = "-s+" ] ; then echo " OK with JIT study" + else echo " OK" + fi + done +fi + +# Locale-specific tests, provided that either the "fr_FR" or the "french" +# locale is available. The former is the Unix-like standard; the latter is +# for Windows. Another possibility is "fr", which needs to be run against +# the Windows-specific input and output files. + +if [ $do3 = yes ] ; then + locale -a | grep '^fr_FR$' >/dev/null + if [ $? -eq 0 ] ; then + locale=fr_FR + infile=$testdata/testinput3 + outfile=$testdata/testoutput3 + else + infile=test3input + outfile=test3output + locale -a | grep '^french$' >/dev/null + if [ $? -eq 0 ] ; then + locale=french + sed 's/fr_FR/french/' $testdata/testinput3 >test3input + sed 's/fr_FR/french/' $testdata/testoutput3 >test3output + else + locale -a | grep '^fr$' >/dev/null + if [ $? -eq 0 ] ; then + locale=fr + sed 's/fr_FR/fr/' $testdata/wintestinput3 >test3input + sed 's/fr_FR/fr/' $testdata/wintestoutput3 >test3output + else + locale= + fi + fi + fi + + if [ "$locale" != "" ] ; then + echo $title3 "(using '$locale' locale)" + for opt in "" "-s" $jitopt; do + $sim $valgrind ./pcretest -q $bmode $opt $infile testtry + if [ $? = 0 ] ; then + $cf $outfile testtry + if [ $? != 0 ] ; then + echo " " + echo "Locale test did not run entirely successfully." + echo "This usually means that there is a problem with the locale" + echo "settings rather than a bug in PCRE." + break; + else + if [ "$opt" = "-s" ] ; then echo " OK with study" + elif [ "$opt" = "-s+" ] ; then echo " OK with JIT study" + else echo " OK" + fi + fi + else exit 1 + fi + done + else + echo "Cannot test locale-specific features - none of the 'fr_FR', 'fr' or" + echo "'french' locales exist, or the \"locale\" command is not available" + echo "to check for them." + echo " " + fi +fi + +# Additional tests for UTF support + +if [ $do4 = yes ] ; then + echo ${title4A}-${bits}${title4B} + if [ $utf -eq 0 ] ; then + echo " Skipped because UTF-$bits support is not available" + else + for opt in "" "-s" $jitopt; do + $sim $valgrind ./pcretest -q $bmode $opt $testdata/testinput4 testtry + if [ $? = 0 ] ; then + $cf $testdata/testoutput4 testtry + if [ $? != 0 ] ; then exit 1; fi + else exit 1 + fi + if [ "$opt" = "-s" ] ; then echo " OK with study" + elif [ "$opt" = "-s+" ] ; then echo " OK with JIT study" + else echo " OK" + fi + done + fi +fi + +if [ $do5 = yes ] ; then + echo ${title5}-${bits} support + if [ $utf -eq 0 ] ; then + echo " Skipped because UTF-$bits support is not available" + else + for opt in "" "-s" $jitopt; do + $sim $valgrind ./pcretest -q $bmode $opt $testdata/testinput5 testtry + if [ $? = 0 ] ; then + $cf $testdata/testoutput5 testtry + if [ $? != 0 ] ; then exit 1; fi + else exit 1 + fi + if [ "$opt" = "-s" ] ; then echo " OK with study" + elif [ "$opt" = "-s+" ] ; then echo " OK with JIT study" + else echo " OK" + fi + done + fi +fi + +if [ $do6 = yes ] ; then + echo $title6 + if [ $utf -eq 0 -o $ucp -eq 0 ] ; then + echo " Skipped because Unicode property support is not available" + else + for opt in "" "-s" $jitopt; do + $sim $valgrind ./pcretest -q $bmode $opt $testdata/testinput6 testtry + if [ $? = 0 ] ; then + $cf $testdata/testoutput6 testtry + if [ $? != 0 ] ; then exit 1; fi + else exit 1 + fi + if [ "$opt" = "-s" ] ; then echo " OK with study" + elif [ "$opt" = "-s+" ] ; then echo " OK with JIT study" + else echo " OK" + fi + done + fi +fi + +# Test non-Perl-compatible Unicode property support + +if [ $do7 = yes ] ; then + echo $title7 + if [ $utf -eq 0 -o $ucp -eq 0 ] ; then + echo " Skipped because Unicode property support is not available" + else + for opt in "" "-s" $jitopt; do + $sim $valgrind ./pcretest -q $bmode $opt $testdata/testinput7 testtry + if [ $? = 0 ] ; then + $cf $testdata/testoutput7 testtry + if [ $? != 0 ] ; then exit 1; fi + else exit 1 + fi + if [ "$opt" = "-s" ] ; then echo " OK with study" + elif [ "$opt" = "-s+" ] ; then echo " OK with JIT study" + else echo " OK" + fi + done + fi +fi + +# Tests for DFA matching support + +if [ $do8 = yes ] ; then + echo $title8 + for opt in "" "-s"; do + $sim $valgrind ./pcretest -q $bmode $opt -dfa $testdata/testinput8 testtry + if [ $? = 0 ] ; then + $cf $testdata/testoutput8 testtry + if [ $? != 0 ] ; then exit 1; fi + else exit 1 + fi + if [ "$opt" = "-s" ] ; then echo " OK with study" ; else echo " OK"; fi + done +fi + +if [ $do9 = yes ] ; then + echo ${title9}-${bits} + if [ $utf -eq 0 ] ; then + echo " Skipped because UTF-$bits support is not available" + else + for opt in "" "-s"; do + $sim $valgrind ./pcretest -q $bmode $opt -dfa $testdata/testinput9 testtry + if [ $? = 0 ] ; then + $cf $testdata/testoutput9 testtry + if [ $? != 0 ] ; then exit 1; fi + else exit 1 + fi + if [ "$opt" = "-s" ] ; then echo " OK with study" ; else echo " OK"; fi + done + fi +fi + +if [ $do10 = yes ] ; then + echo $title10 + if [ $utf -eq 0 -o $ucp -eq 0 ] ; then + echo " Skipped because Unicode property support is not available" + else + for opt in "" "-s"; do + $sim $valgrind ./pcretest -q $bmode $opt -dfa $testdata/testinput10 testtry + if [ $? = 0 ] ; then + $cf $testdata/testoutput10 testtry + if [ $? != 0 ] ; then exit 1; fi + else exit 1 + fi + if [ "$opt" = "-s" ] ; then echo " OK with study" ; else echo " OK"; fi + done + fi +fi + +# Test of internal offsets and code sizes. This test is run only when there +# is Unicode property support and the link size is 2. The actual tests are +# mostly the same as in some of the above, but in this test we inspect some +# offsets and sizes that require a known link size. This is a doublecheck for +# the maintainer, just in case something changes unexpectely. The output from +# this test is not the same in 8-bit and 16-bit modes. + +if [ $do11 = yes ] ; then + echo $title11 + if [ $link_size -ne 2 ] ; then + echo " Skipped because link size is not 2" + elif [ $ucp -eq 0 ] ; then + echo " Skipped because Unicode property support is not available" + else + for opt in "" "-s"; do + $sim $valgrind ./pcretest -q $bmode $opt $testdata/testinput11 testtry + if [ $? = 0 ] ; then + $cf $testdata/testoutput11-$bits testtry + if [ $? != 0 ] ; then exit 1; fi + else exit 1 + fi + if [ "$opt" = "-s" ] ; then echo " OK with study" ; else echo " OK"; fi + done + fi +fi + +# Test JIT-specific features when JIT is available + +if [ $do12 = yes ] ; then + echo $title12 + if [ $jit -eq 0 -o "$nojit" = "yes" ] ; then + echo " Skipped because JIT is not available or not usable" + else + $sim $valgrind ./pcretest -q $bmode $testdata/testinput12 testtry + if [ $? = 0 ] ; then + $cf $testdata/testoutput12 testtry + if [ $? != 0 ] ; then exit 1; fi + else exit 1 + fi + echo " OK" + fi +fi + +# Test JIT-specific features when JIT is not available + +if [ $do13 = yes ] ; then + echo $title13 + if [ $jit -ne 0 ] ; then + echo " Skipped because JIT is available" + else + $sim $valgrind ./pcretest -q $bmode $testdata/testinput13 testtry + if [ $? = 0 ] ; then + $cf $testdata/testoutput13 testtry + if [ $? != 0 ] ; then exit 1; fi + else exit 1 + fi + echo " OK" + fi +fi + +# Tests for 8-bit-specific features + +if [ "$do14" = yes ] ; then + echo $title14 + if [ "$bits" = "16" -o "$bits" = "32" ] ; then + echo " Skipped when running 16/32-bit tests" + else + cp -f $testdata/saved16 testsaved16 + cp -f $testdata/saved32 testsaved32 + for opt in "" "-s" $jitopt; do + $sim $valgrind ./pcretest -q $bmode $opt $testdata/testinput14 testtry + if [ $? = 0 ] ; then + $cf $testdata/testoutput14 testtry + if [ $? != 0 ] ; then exit 1; fi + else exit 1 + fi + if [ "$opt" = "-s" ] ; then echo " OK with study" + elif [ "$opt" = "-s+" ] ; then echo " OK with JIT study" + else echo " OK" + fi + done + fi +fi + +# Tests for 8-bit-specific features (needs UTF-8 support) + +if [ "$do15" = yes ] ; then + echo $title15 + if [ "$bits" = "16" -o "$bits" = "32" ] ; then + echo " Skipped when running 16/32-bit tests" + elif [ $utf -eq 0 ] ; then + echo " Skipped because UTF-$bits support is not available" + else + for opt in "" "-s" $jitopt; do + $sim $valgrind ./pcretest -q $bmode $opt $testdata/testinput15 testtry + if [ $? = 0 ] ; then + $cf $testdata/testoutput15 testtry + if [ $? != 0 ] ; then exit 1; fi + else exit 1 + fi + if [ "$opt" = "-s" ] ; then echo " OK with study" + elif [ "$opt" = "-s+" ] ; then echo " OK with JIT study" + else echo " OK" + fi + done + fi +fi + +# Tests for 8-bit-specific features (Unicode property support) + +if [ $do16 = yes ] ; then + echo $title16 + if [ "$bits" = "16" -o "$bits" = "32" ] ; then + echo " Skipped when running 16/32-bit tests" + elif [ $ucp -eq 0 ] ; then + echo " Skipped because Unicode property support is not available" + else + for opt in "" "-s" $jitopt; do + $sim $valgrind ./pcretest -q $bmode $opt $testdata/testinput16 testtry + if [ $? = 0 ] ; then + $cf $testdata/testoutput16 testtry + if [ $? != 0 ] ; then exit 1; fi + else exit 1 + fi + if [ "$opt" = "-s" ] ; then echo " OK with study" + elif [ "$opt" = "-s+" ] ; then echo " OK with JIT study" + else echo " OK" + fi + done + fi +fi + +# Tests for 16/32-bit-specific features + +if [ $do17 = yes ] ; then + echo $title17 + if [ "$bits" = "8" ] ; then + echo " Skipped when running 8-bit tests" + else + for opt in "" "-s" $jitopt; do + $sim $valgrind ./pcretest -q $bmode $opt $testdata/testinput17 testtry + if [ $? = 0 ] ; then + $cf $testdata/testoutput17 testtry + if [ $? != 0 ] ; then exit 1; fi + else exit 1 + fi + if [ "$opt" = "-s" ] ; then echo " OK with study" + elif [ "$opt" = "-s+" ] ; then echo " OK with JIT study" + else echo " OK" + fi + done + fi +fi + +# Tests for 16/32-bit-specific features (UTF-16/32 support) + +if [ $do18 = yes ] ; then + echo $title18 + if [ "$bits" = "8" ] ; then + echo " Skipped when running 8-bit tests" + elif [ $utf -eq 0 ] ; then + echo " Skipped because UTF-$bits support is not available" + else + for opt in "" "-s" $jitopt; do + $sim $valgrind ./pcretest -q $bmode $opt $testdata/testinput18 testtry + if [ $? = 0 ] ; then + $cf $testdata/testoutput18-$bits testtry + if [ $? != 0 ] ; then exit 1; fi + else exit 1 + fi + if [ "$opt" = "-s" ] ; then echo " OK with study" + elif [ "$opt" = "-s+" ] ; then echo " OK with JIT study" + else echo " OK" + fi + done + fi +fi + +# Tests for 16/32-bit-specific features (Unicode property support) + +if [ $do19 = yes ] ; then + echo $title19 + if [ "$bits" = "8" ] ; then + echo " Skipped when running 8-bit tests" + elif [ $ucp -eq 0 ] ; then + echo " Skipped because Unicode property support is not available" + else + for opt in "" "-s" $jitopt; do + $sim $valgrind ./pcretest -q $bmode $opt $testdata/testinput19 testtry + if [ $? = 0 ] ; then + $cf $testdata/testoutput19 testtry + if [ $? != 0 ] ; then exit 1; fi + else exit 1 + fi + if [ "$opt" = "-s" ] ; then echo " OK with study" + elif [ "$opt" = "-s+" ] ; then echo " OK with JIT study" + else echo " OK" + fi + done + fi +fi + +# Tests for 16/32-bit-specific features in DFA non-UTF-16/32 mode + +if [ $do20 = yes ] ; then + echo $title20 + if [ "$bits" = "8" ] ; then + echo " Skipped when running 8-bit tests" + else + for opt in "" "-s"; do + $sim $valgrind ./pcretest -q $bmode $opt -dfa $testdata/testinput20 testtry + if [ $? = 0 ] ; then + $cf $testdata/testoutput20 testtry + if [ $? != 0 ] ; then exit 1; fi + else exit 1 + fi + if [ "$opt" = "-s" ] ; then echo " OK with study" + else echo " OK" + fi + done + fi +fi + +# Tests for reloads with 16/32-bit library + +if [ $do21 = yes ] ; then + echo $title21 + if [ "$bits" = "8" ] ; then + echo " Skipped when running 8-bit tests" + elif [ $link_size -ne 2 ] ; then + echo " Skipped because link size is not 2" + else + cp -f $testdata/saved8 testsaved8 + cp -f $testdata/saved16LE-1 testsaved16LE-1 + cp -f $testdata/saved16BE-1 testsaved16BE-1 + cp -f $testdata/saved32LE-1 testsaved32LE-1 + cp -f $testdata/saved32BE-1 testsaved32BE-1 + $sim $valgrind ./pcretest -q $bmode $testdata/testinput21 testtry + if [ $? = 0 ] ; then + $cf $testdata/testoutput21-$bits testtry + if [ $? != 0 ] ; then exit 1; fi + else exit 1 + fi + echo " OK" + fi +fi + +# Tests for reloads with 16/32-bit library (UTF-16 support) + +if [ $do22 = yes ] ; then + echo $title22 + if [ "$bits" = "8" ] ; then + echo " Skipped when running 8-bit tests" + elif [ $utf -eq 0 ] ; then + echo " Skipped because UTF-$bits support is not available" + elif [ $link_size -ne 2 ] ; then + echo " Skipped because link size is not 2" + else + cp -f $testdata/saved16LE-2 testsaved16LE-2 + cp -f $testdata/saved16BE-2 testsaved16BE-2 + cp -f $testdata/saved32LE-2 testsaved32LE-2 + cp -f $testdata/saved32BE-2 testsaved32BE-2 + $sim $valgrind ./pcretest -q $bmode $testdata/testinput22 testtry + if [ $? = 0 ] ; then + $cf $testdata/testoutput22-$bits testtry + if [ $? != 0 ] ; then exit 1; fi + else exit 1 + fi + echo " OK" + fi +fi + +if [ $do23 = yes ] ; then + echo $title23 + if [ "$bits" = "8" -o "$bits" = "32" ] ; then + echo " Skipped when running 8/32-bit tests" + else + $sim $valgrind ./pcretest -q $bmode $testdata/testinput23 testtry + if [ $? = 0 ] ; then + $cf $testdata/testoutput23 testtry + if [ $? != 0 ] ; then exit 1; fi + else exit 1 + fi + echo " OK" + fi +fi + +if [ $do24 = yes ] ; then + echo $title24 + if [ "$bits" = "8" -o "$bits" = "32" ] ; then + echo " Skipped when running 8/32-bit tests" + elif [ $utf -eq 0 ] ; then + echo " Skipped because UTF-$bits support is not available" + else + $sim $valgrind ./pcretest -q $bmode $testdata/testinput24 testtry + if [ $? = 0 ] ; then + $cf $testdata/testoutput24 testtry + if [ $? != 0 ] ; then exit 1; fi + else exit 1 + fi + echo " OK" + fi +fi + +if [ $do25 = yes ] ; then + echo $title25 + if [ "$bits" = "8" -o "$bits" = "16" ] ; then + echo " Skipped when running 8/16-bit tests" + else + $sim $valgrind ./pcretest -q $bmode $testdata/testinput25 testtry + if [ $? = 0 ] ; then + $cf $testdata/testoutput25 testtry + if [ $? != 0 ] ; then exit 1; fi + else exit 1 + fi + echo " OK" + fi +fi + +if [ $do26 = yes ] ; then + echo $title26 + if [ "$bits" = "8" -o "$bits" = "16" ] ; then + echo " Skipped when running 8/16-bit tests" + elif [ $utf -eq 0 ] ; then + echo " Skipped because UTF-$bits support is not available" + else + $sim $valgrind ./pcretest -q $bmode $testdata/testinput26 testtry + if [ $? = 0 ] ; then + $cf $testdata/testoutput26 testtry + if [ $? != 0 ] ; then exit 1; fi + else exit 1 + fi + echo " OK" + fi +fi + +# End of loop for 8/16/32-bit tests +done + +# Clean up local working files +rm -f test3input test3output testNinput testsaved* teststderr teststdout testtry + +# End diff --git a/pcre/RunTest.bat b/pcre/RunTest.bat new file mode 100644 index 00000000000..35d7f71f9e8 --- /dev/null +++ b/pcre/RunTest.bat @@ -0,0 +1,616 @@ +@echo off +@rem This file must use CRLF linebreaks to function properly +@rem and requires both pcretest and pcregrep +@rem This file was originally contributed by Ralf Junker, and touched up by +@rem Daniel Richard G. Tests 10-12 added by Philip H. +@rem Philip H also changed test 3 to use "wintest" files. +@rem +@rem Updated by Tom Fortmann to support explicit test numbers on the command line. +@rem Added argument validation and added error reporting. +@rem +@rem MS Windows batch file to run pcretest on testfiles with the correct +@rem options. +@rem +@rem Sheri Pierce added logic to skip feature dependent tests +@rem tests 4 5 9 15 and 18 require utf support +@rem tests 6 7 10 16 and 19 require ucp support +@rem 11 requires ucp and link size 2 +@rem 12 requires presence of jit support +@rem 13 requires absence of jit support +@rem Sheri P also added override tests for study and jit testing +@rem Zoltan Herczeg added libpcre16 support +@rem Zoltan Herczeg added libpcre32 support + +setlocal enabledelayedexpansion +if [%srcdir%]==[] ( +if exist testdata\ set srcdir=.) +if [%srcdir%]==[] ( +if exist ..\testdata\ set srcdir=..) +if [%srcdir%]==[] ( +if exist ..\..\testdata\ set srcdir=..\..) +if NOT exist %srcdir%\testdata\ ( +Error: echo distribution testdata folder not found! +call :conferror +exit /b 1 +goto :eof +) + +if [%pcretest%]==[] set pcretest=.\pcretest.exe + +echo source dir is %srcdir% +echo pcretest=%pcretest% + +if NOT exist %pcretest% ( +echo Error: %pcretest% not found! +echo. +call :conferror +exit /b 1 +) + +%pcretest% -C linksize >NUL +set link_size=%ERRORLEVEL% +%pcretest% -C pcre8 >NUL +set support8=%ERRORLEVEL% +%pcretest% -C pcre16 >NUL +set support16=%ERRORLEVEL% +%pcretest% -C pcre32 >NUL +set support32=%ERRORLEVEL% +%pcretest% -C utf >NUL +set utf=%ERRORLEVEL% +%pcretest% -C ucp >NUL +set ucp=%ERRORLEVEL% +%pcretest% -C jit >NUL +set jit=%ERRORLEVEL% + +if %support8% EQU 1 ( +if not exist testout8 md testout8 +if not exist testoutstudy8 md testoutstudy8 +if not exist testoutjit8 md testoutjit8 +) + +if %support16% EQU 1 ( +if not exist testout16 md testout16 +if not exist testoutstudy16 md testoutstudy16 +if not exist testoutjit16 md testoutjit16 +) + +if %support16% EQU 1 ( +if not exist testout32 md testout32 +if not exist testoutstudy32 md testoutstudy32 +if not exist testoutjit32 md testoutjit32 +) + +set do1=no +set do2=no +set do3=no +set do4=no +set do5=no +set do6=no +set do7=no +set do8=no +set do9=no +set do10=no +set do11=no +set do12=no +set do13=no +set do14=no +set do15=no +set do16=no +set do17=no +set do18=no +set do19=no +set do20=no +set do21=no +set do22=no +set do23=no +set do24=no +set do25=no +set do26=no +set all=yes + +for %%a in (%*) do ( + set valid=no + for %%v in (1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26) do if %%v == %%a set valid=yes + if "!valid!" == "yes" ( + set do%%a=yes + set all=no +) else ( + echo Invalid test number - %%a! + echo Usage %0 [ test_number ] ... + echo Where test_number is one or more optional test numbers 1 through 26, default is all tests. + exit /b 1 +) +) +set failed="no" + +if "%all%" == "yes" ( + set do1=yes + set do2=yes + set do3=yes + set do4=yes + set do5=yes + set do6=yes + set do7=yes + set do8=yes + set do9=yes + set do10=yes + set do11=yes + set do12=yes + set do13=yes + set do14=yes + set do15=yes + set do16=yes + set do17=yes + set do18=yes + set do19=yes + set do20=yes + set do21=yes + set do22=yes + set do23=yes + set do24=yes + set do25=yes + set do26=yes +) + +@echo RunTest.bat's pcretest output is written to newly created subfolders named +@echo testout, testoutstudy and testoutjit. +@echo. + +set mode= +set bits=8 + +:nextMode +if "%mode%" == "" ( + if %support8% EQU 0 goto modeSkip + echo. + echo ---- Testing 8-bit library ---- + echo. +) +if "%mode%" == "-16" ( + if %support16% EQU 0 goto modeSkip + echo. + echo ---- Testing 16-bit library ---- + echo. +) +if "%mode%" == "-32" ( + if %support32% EQU 0 goto modeSkip + echo. + echo ---- Testing 32-bit library ---- + echo. +) +if "%do1%" == "yes" call :do1 +if "%do2%" == "yes" call :do2 +if "%do3%" == "yes" call :do3 +if "%do4%" == "yes" call :do4 +if "%do5%" == "yes" call :do5 +if "%do6%" == "yes" call :do6 +if "%do7%" == "yes" call :do7 +if "%do8%" == "yes" call :do8 +if "%do9%" == "yes" call :do9 +if "%do10%" == "yes" call :do10 +if "%do11%" == "yes" call :do11 +if "%do12%" == "yes" call :do12 +if "%do13%" == "yes" call :do13 +if "%do14%" == "yes" call :do14 +if "%do15%" == "yes" call :do15 +if "%do16%" == "yes" call :do16 +if "%do17%" == "yes" call :do17 +if "%do18%" == "yes" call :do18 +if "%do19%" == "yes" call :do19 +if "%do20%" == "yes" call :do20 +if "%do21%" == "yes" call :do21 +if "%do22%" == "yes" call :do22 +if "%do23%" == "yes" call :do23 +if "%do24%" == "yes" call :do24 +if "%do25%" == "yes" call :do25 +if "%do26%" == "yes" call :do26 +:modeSkip +if "%mode%" == "" ( + set mode=-16 + set bits=16 + goto nextMode +) +if "%mode%" == "-16" ( + set mode=-32 + set bits=32 + goto nextMode +) + +@rem If mode is -32, testing is finished +if %failed% == "yes" ( +echo In above output, one or more of the various tests failed! +exit /b 1 +) +echo All OK +goto :eof + +:runsub +@rem Function to execute pcretest and compare the output +@rem Arguments are as follows: +@rem +@rem 1 = test number +@rem 2 = outputdir +@rem 3 = test name use double quotes +@rem 4 - 9 = pcretest options + +if [%1] == [] ( + echo Missing test number argument! + exit /b 1 +) + +if [%2] == [] ( + echo Missing outputdir! + exit /b 1 +) + +if [%3] == [] ( + echo Missing test name argument! + exit /b 1 +) + +set testinput=testinput%1 +set testoutput=testoutput%1 +if exist %srcdir%\testdata\win%testinput% ( + set testinput=wintestinput%1 + set testoutput=wintestoutput%1 +) + +echo Test %1: %3 +%pcretest% %mode% %4 %5 %6 %7 %8 %9 %srcdir%\testdata\%testinput% >%2%bits%\%testoutput% +if errorlevel 1 ( + echo. failed executing command-line: + echo. %pcretest% %mode% %4 %5 %6 %7 %8 %9 %srcdir%\testdata\%testinput% ^>%2%bits%\%testoutput% + set failed="yes" + goto :eof +) + +set type= +if [%1]==[11] ( + set type=-%bits% +) +if [%1]==[18] ( + set type=-%bits% +) +if [%1]==[21] ( + set type=-%bits% +) +if [%1]==[22] ( + set type=-%bits% +) + +fc /n %srcdir%\testdata\%testoutput%%type% %2%bits%\%testoutput% >NUL + +if errorlevel 1 ( + echo. failed comparison: fc /n %srcdir%\testdata\%testoutput% %2%bits%\%testoutput% + if [%1]==[2] ( + echo. + echo ** Test 2 requires a lot of stack. PCRE can be configured to + echo ** use heap for recursion. Otherwise, to pass Test 2 + echo ** you generally need to allocate 8 mb stack to PCRE. + echo ** See the 'pcrestack' page for a discussion of PCRE's + echo ** stack usage. + echo. +) + if [%1]==[3] ( + echo. + echo ** Test 3 failure usually means french locale is not + echo ** available on the system, rather than a bug or problem with PCRE. + echo. + goto :eof +) + + set failed="yes" + goto :eof +) + +echo. Passed. +goto :eof + +:do1 +call :runsub 1 testout "Main functionality (Compatible with Perl >= 5.10)" -q +call :runsub 1 testoutstudy "Test with Study Override" -q -s +if %jit% EQU 1 call :runsub 1 testoutjit "Test with JIT Override" -q -s+ +goto :eof + +:do2 + call :runsub 2 testout "API, errors, internals, and non-Perl stuff" -q + call :runsub 2 testoutstudy "Test with Study Override" -q -s + if %jit% EQU 1 call :runsub 2 testoutjit "Test with JIT Override" -q -s+ +goto :eof + +:do3 + call :runsub 3 testout "Locale-specific features" -q + call :runsub 3 testoutstudy "Test with Study Override" -q -s + if %jit% EQU 1 call :runsub 3 testoutjit "Test with JIT Override" -q -s+ +goto :eof + +:do4 +if %utf% EQU 0 ( + echo Test 4 Skipped due to absence of UTF-%bits% support. + goto :eof +) + call :runsub 4 testout "UTF-%bits% support - (Compatible with Perl >= 5.10)" -q + call :runsub 4 testoutstudy "Test with Study Override" -q -s + if %jit% EQU 1 call :runsub 4 testoutjit "Test with JIT Override" -q -s+ +goto :eof + +:do5 +if %utf% EQU 0 ( + echo Test 5 Skipped due to absence of UTF-%bits% support. + goto :eof +) + call :runsub 5 testout "API, internals, and non-Perl stuff for UTF-%bits%" -q + call :runsub 5 testoutstudy "Test with Study Override" -q -s + if %jit% EQU 1 call :runsub 5 testoutjit "Test with JIT Override" -q -s+ +goto :eof + +:do6 +if %ucp% EQU 0 ( + echo Test 6 Skipped due to absence of Unicode property support. + goto :eof +) + call :runsub 6 testout "Unicode property support (Compatible with Perl >= 5.10)" -q + call :runsub 6 testoutstudy "Test with Study Override" -q -s + if %jit% EQU 1 call :runsub 6 testoutjit "Test with JIT Override" -q -s+ +goto :eof + +:do7 +if %ucp% EQU 0 ( + echo Test 7 Skipped due to absence of Unicode property support. + goto :eof +) + call :runsub 7 testout "API, internals, and non-Perl stuff for Unicode property support" -q + call :runsub 7 testoutstudy "Test with Study Override" -q -s + if %jit% EQU 1 call :runsub 7 testoutjit "Test with JIT Override" -q -s+ +goto :eof + +:do8 + call :runsub 8 testout "DFA matching main functionality" -q -dfa + call :runsub 8 testoutstudy "Test with Study Override" -q -dfa -s +goto :eof + +:do9 +if %utf% EQU 0 ( + echo Test 9 Skipped due to absence of UTF-%bits% support. + goto :eof +) + call :runsub 9 testout "DFA matching with UTF-%bits%" -q -dfa + call :runsub 9 testoutstudy "Test with Study Override" -q -dfa -s + goto :eof + +:do10 +if %ucp% EQU 0 ( + echo Test 10 Skipped due to absence of Unicode property support. + goto :eof +) + call :runsub 10 testout "DFA matching with Unicode properties" -q -dfa + call :runsub 10 testoutstudy "Test with Study Override" -q -dfa -s +goto :eof + +:do11 +if NOT %link_size% EQU 2 ( + echo Test 11 Skipped because link size is not 2. + goto :eof +) +if %ucp% EQU 0 ( + echo Test 11 Skipped due to absence of Unicode property support. + goto :eof +) + call :runsub 11 testout "Internal offsets and code size tests" -q + call :runsub 11 testoutstudy "Test with Study Override" -q -s +goto :eof + +:do12 +if %jit% EQU 0 ( + echo Test 12 Skipped due to absence of JIT support. + goto :eof +) + call :runsub 12 testout "JIT-specific features (JIT available)" -q +goto :eof + +:do13 +if %jit% EQU 1 ( + echo Test 13 Skipped due to presence of JIT support. + goto :eof +) + call :runsub 13 testout "JIT-specific features (JIT not available)" -q +goto :eof + +:do14 +if NOT %bits% EQU 8 ( + echo Test 14 Skipped when running 16/32-bit tests. + goto :eof +) + copy /Y %srcdir%\testdata\saved16 testsaved16 + copy /Y %srcdir%\testdata\saved32 testsaved32 + call :runsub 14 testout "Specials for the basic 8-bit library" -q + call :runsub 14 testoutstudy "Test with Study Override" -q -s + if %jit% EQU 1 call :runsub 14 testoutjit "Test with JIT Override" -q -s+ +goto :eof + +:do15 +if NOT %bits% EQU 8 ( + echo Test 15 Skipped when running 16/32-bit tests. + goto :eof +) +if %utf% EQU 0 ( + echo Test 15 Skipped due to absence of UTF-%bits% support. + goto :eof +) + call :runsub 15 testout "Specials for the 8-bit library with UTF-%bits% support" -q + call :runsub 15 testoutstudy "Test with Study Override" -q -s + if %jit% EQU 1 call :runsub 15 testoutjit "Test with JIT Override" -q -s+ +goto :eof + +:do16 +if NOT %bits% EQU 8 ( + echo Test 16 Skipped when running 16/32-bit tests. + goto :eof +) +if %ucp% EQU 0 ( + echo Test 16 Skipped due to absence of Unicode property support. + goto :eof +) + call :runsub 16 testout "Specials for the 8-bit library with Unicode propery support" -q + call :runsub 16 testoutstudy "Test with Study Override" -q -s + if %jit% EQU 1 call :runsub 16 testoutjit "Test with JIT Override" -q -s+ +goto :eof + +:do17 +if %bits% EQU 8 ( + echo Test 17 Skipped when running 8-bit tests. + goto :eof +) + call :runsub 17 testout "Specials for the basic 16/32-bit library" -q + call :runsub 17 testoutstudy "Test with Study Override" -q -s + if %jit% EQU 1 call :runsub 17 testoutjit "Test with JIT Override" -q -s+ +goto :eof + +:do18 +if %bits% EQU 8 ( + echo Test 18 Skipped when running 8-bit tests. + goto :eof +) +if %utf% EQU 0 ( + echo Test 18 Skipped due to absence of UTF-%bits% support. + goto :eof +) + call :runsub 18 testout "Specials for the 16/32-bit library with UTF-%bits% support" -q + call :runsub 18 testoutstudy "Test with Study Override" -q -s + if %jit% EQU 1 call :runsub 18 testoutjit "Test with JIT Override" -q -s+ +goto :eof + +:do19 +if %bits% EQU 8 ( + echo Test 19 Skipped when running 8-bit tests. + goto :eof +) +if %ucp% EQU 0 ( + echo Test 19 Skipped due to absence of Unicode property support. + goto :eof +) + call :runsub 19 testout "Specials for the 16/32-bit library with Unicode property support" -q + call :runsub 19 testoutstudy "Test with Study Override" -q -s + if %jit% EQU 1 call :runsub 19 testoutjit "Test with JIT Override" -q -s+ +goto :eof + +:do20 +if %bits% EQU 8 ( + echo Test 20 Skipped when running 8-bit tests. + goto :eof +) + call :runsub 20 testout "DFA specials for the basic 16/32-bit library" -q -dfa + call :runsub 20 testoutstudy "Test with Study Override" -q -dfa -s +goto :eof + +:do21 +if %bits% EQU 8 ( + echo Test 21 Skipped when running 8-bit tests. + goto :eof +) +if NOT %link_size% EQU 2 ( + echo Test 21 Skipped because link size is not 2. + goto :eof +) +copy /Y %srcdir%\testdata\saved8 testsaved8 +copy /Y %srcdir%\testdata\saved16LE-1 testsaved16LE-1 +copy /Y %srcdir%\testdata\saved16BE-1 testsaved16BE-1 +copy /Y %srcdir%\testdata\saved32LE-1 testsaved32LE-1 +copy /Y %srcdir%\testdata\saved32BE-1 testsaved32BE-1 +call :runsub 21 testout "Reloads for the basic 16/32-bit library" -q +call :runsub 21 testoutstudy "Test with Study Override" -q -s +if %jit% EQU 1 call :runsub 21 testoutjit "Test with JIT Override" -q -s+ +goto :eof + +:do22 +if %bits% EQU 8 ( + echo Test 22 Skipped when running 8-bit tests. + goto :eof +) +if %utf% EQU 0 ( + echo Test 22 Skipped due to absence of UTF-%bits% support. + goto :eof +) +if NOT %link_size% EQU 2 ( + echo Test 22 Skipped because link size is not 2. + goto :eof +) +copy /Y %srcdir%\testdata\saved16LE-2 testsaved16LE-2 +copy /Y %srcdir%\testdata\saved16BE-2 testsaved16BE-2 +copy /Y %srcdir%\testdata\saved32LE-2 testsaved32LE-2 +copy /Y %srcdir%\testdata\saved32BE-2 testsaved32BE-2 +call :runsub 22 testout "Reloads for the 16/32-bit library with UTF-16/32 support" -q +call :runsub 22 testoutstudy "Test with Study Override" -q -s +if %jit% EQU 1 call :runsub 22 testoutjit "Test with JIT Override" -q -s+ +goto :eof + +:do23 +if NOT %bits% EQU 16 ( + echo Test 23 Skipped when running 8/32-bit tests. + goto :eof +) +call :runsub 23 testout "Specials for the 16-bit library" -q +call :runsub 23 testoutstudy "Test with Study Override" -q -s +if %jit% EQU 1 call :runsub 23 testoutjit "Test with JIT Override" -q -s+ +goto :eof + +:do24 +if NOT %bits% EQU 16 ( + echo Test 24 Skipped when running 8/32-bit tests. + goto :eof +) +if %utf% EQU 0 ( + echo Test 24 Skipped due to absence of UTF-%bits% support. + goto :eof +) +call :runsub 24 testout "Specials for the 16-bit library with UTF-16 support" -q +call :runsub 24 testoutstudy "Test with Study Override" -q -s +if %jit% EQU 1 call :runsub 24 testoutjit "Test with JIT Override" -q -s+ +goto :eof + +:do25 +if NOT %bits% EQU 32 ( + echo Test 25 Skipped when running 8/16-bit tests. + goto :eof +) +call :runsub 25 testout "Specials for the 32-bit library" -q +call :runsub 25 testoutstudy "Test with Study Override" -q -s +if %jit% EQU 1 call :runsub 25 testoutjit "Test with JIT Override" -q -s+ +goto :eof + +:do26 +if NOT %bits% EQU 32 ( + echo Test 26 Skipped when running 8/16-bit tests. + goto :eof +) +if %utf% EQU 0 ( + echo Test 26 Skipped due to absence of UTF-%bits% support. + goto :eof +) +call :runsub 26 testout "Specials for the 32-bit library with UTF-32 support" -q +call :runsub 26 testoutstudy "Test with Study Override" -q -s +if %jit% EQU 1 call :runsub 26 testoutjit "Test with JIT Override" -q -s+ +goto :eof + +:conferror +@echo. +@echo Either your build is incomplete or you have a configuration error. +@echo. +@echo If configured with cmake and executed via "make test" or the MSVC "RUN_TESTS" +@echo project, pcre_test.bat defines variables and automatically calls RunTest.bat. +@echo For manual testing of all available features, after configuring with cmake +@echo and building, you can run the built pcre_test.bat. For best results with +@echo cmake builds and tests avoid directories with full path names that include +@echo spaces for source or build. +@echo. +@echo Otherwise, if the build dir is in a subdir of the source dir, testdata needed +@echo for input and verification should be found automatically when (from the +@echo location of the the built exes) you call RunTest.bat. By default RunTest.bat +@echo runs all tests compatible with the linked pcre library but it can be given +@echo a test number as an argument. +@echo. +@echo If the build dir is not under the source dir you can either copy your exes +@echo to the source folder or copy RunTest.bat and the testdata folder to the +@echo location of your built exes and then run RunTest.bat. +@echo. +goto :eof diff --git a/pcre/cmake/COPYING-CMAKE-SCRIPTS b/pcre/cmake/COPYING-CMAKE-SCRIPTS new file mode 100644 index 00000000000..4b417765f3a --- /dev/null +++ b/pcre/cmake/COPYING-CMAKE-SCRIPTS @@ -0,0 +1,22 @@ +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions +are met: + +1. Redistributions of source code must retain the copyright + notice, this list of conditions and the following disclaimer. +2. Redistributions in binary form must reproduce the copyright + notice, this list of conditions and the following disclaimer in the + documentation and/or other materials provided with the distribution. +3. The name of the author may not be used to endorse or promote products + derived from this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR +IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES +OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. +IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, +INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT +NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, +DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY +THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT +(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF +THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. diff --git a/pcre/cmake/FindEditline.cmake b/pcre/cmake/FindEditline.cmake new file mode 100644 index 00000000000..2d0b7cc543c --- /dev/null +++ b/pcre/cmake/FindEditline.cmake @@ -0,0 +1,17 @@ +# Modified from FindReadline.cmake (PH Feb 2012) + +if(EDITLINE_INCLUDE_DIR AND EDITLINE_LIBRARY AND NCURSES_LIBRARY) + set(EDITLINE_FOUND TRUE) +else(EDITLINE_INCLUDE_DIR AND EDITLINE_LIBRARY AND NCURSES_LIBRARY) + FIND_PATH(EDITLINE_INCLUDE_DIR readline.h + /usr/include/editline + /usr/include/edit/readline + /usr/include/readline + ) + + FIND_LIBRARY(EDITLINE_LIBRARY NAMES edit) + include(FindPackageHandleStandardArgs) + FIND_PACKAGE_HANDLE_STANDARD_ARGS(Editline DEFAULT_MSG EDITLINE_INCLUDE_DIR EDITLINE_LIBRARY ) + + MARK_AS_ADVANCED(EDITLINE_INCLUDE_DIR EDITLINE_LIBRARY) +endif(EDITLINE_INCLUDE_DIR AND EDITLINE_LIBRARY AND NCURSES_LIBRARY) diff --git a/pcre/cmake/FindPackageHandleStandardArgs.cmake b/pcre/cmake/FindPackageHandleStandardArgs.cmake new file mode 100644 index 00000000000..151d8125031 --- /dev/null +++ b/pcre/cmake/FindPackageHandleStandardArgs.cmake @@ -0,0 +1,58 @@ +# FIND_PACKAGE_HANDLE_STANDARD_ARGS(NAME (DEFAULT_MSG|"Custom failure message") VAR1 ... ) +# This macro is intended to be used in FindXXX.cmake modules files. +# It handles the REQUIRED and QUIET argument to FIND_PACKAGE() and +# it also sets the _FOUND variable. +# The package is found if all variables listed are TRUE. +# Example: +# +# FIND_PACKAGE_HANDLE_STANDARD_ARGS(LibXml2 DEFAULT_MSG LIBXML2_LIBRARIES LIBXML2_INCLUDE_DIR) +# +# LibXml2 is considered to be found, if both LIBXML2_LIBRARIES and +# LIBXML2_INCLUDE_DIR are valid. Then also LIBXML2_FOUND is set to TRUE. +# If it is not found and REQUIRED was used, it fails with FATAL_ERROR, +# independent whether QUIET was used or not. +# If it is found, the location is reported using the VAR1 argument, so +# here a message "Found LibXml2: /usr/lib/libxml2.so" will be printed out. +# If the second argument is DEFAULT_MSG, the message in the failure case will +# be "Could NOT find LibXml2", if you don't like this message you can specify +# your own custom failure message there. + +MACRO(FIND_PACKAGE_HANDLE_STANDARD_ARGS _NAME _FAIL_MSG _VAR1 ) + + IF("${_FAIL_MSG}" STREQUAL "DEFAULT_MSG") + IF (${_NAME}_FIND_REQUIRED) + SET(_FAIL_MESSAGE "Could not find REQUIRED package ${_NAME}") + ELSE (${_NAME}_FIND_REQUIRED) + SET(_FAIL_MESSAGE "Could not find OPTIONAL package ${_NAME}") + ENDIF (${_NAME}_FIND_REQUIRED) + ELSE("${_FAIL_MSG}" STREQUAL "DEFAULT_MSG") + SET(_FAIL_MESSAGE "${_FAIL_MSG}") + ENDIF("${_FAIL_MSG}" STREQUAL "DEFAULT_MSG") + + STRING(TOUPPER ${_NAME} _NAME_UPPER) + + SET(${_NAME_UPPER}_FOUND TRUE) + IF(NOT ${_VAR1}) + SET(${_NAME_UPPER}_FOUND FALSE) + ENDIF(NOT ${_VAR1}) + + FOREACH(_CURRENT_VAR ${ARGN}) + IF(NOT ${_CURRENT_VAR}) + SET(${_NAME_UPPER}_FOUND FALSE) + ENDIF(NOT ${_CURRENT_VAR}) + ENDFOREACH(_CURRENT_VAR) + + IF (${_NAME_UPPER}_FOUND) + IF (NOT ${_NAME}_FIND_QUIETLY) + MESSAGE(STATUS "Found ${_NAME}: ${${_VAR1}}") + ENDIF (NOT ${_NAME}_FIND_QUIETLY) + ELSE (${_NAME_UPPER}_FOUND) + IF (${_NAME}_FIND_REQUIRED) + MESSAGE(FATAL_ERROR "${_FAIL_MESSAGE}") + ELSE (${_NAME}_FIND_REQUIRED) + IF (NOT ${_NAME}_FIND_QUIETLY) + MESSAGE(STATUS "${_FAIL_MESSAGE}") + ENDIF (NOT ${_NAME}_FIND_QUIETLY) + ENDIF (${_NAME}_FIND_REQUIRED) + ENDIF (${_NAME_UPPER}_FOUND) +ENDMACRO(FIND_PACKAGE_HANDLE_STANDARD_ARGS) diff --git a/pcre/cmake/FindReadline.cmake b/pcre/cmake/FindReadline.cmake new file mode 100644 index 00000000000..1d4cc558431 --- /dev/null +++ b/pcre/cmake/FindReadline.cmake @@ -0,0 +1,29 @@ +# from http://websvn.kde.org/trunk/KDE/kdeedu/cmake/modules/FindReadline.cmake +# http://websvn.kde.org/trunk/KDE/kdeedu/cmake/modules/COPYING-CMAKE-SCRIPTS +# --> BSD licensed +# +# GNU Readline library finder +if(READLINE_INCLUDE_DIR AND READLINE_LIBRARY AND NCURSES_LIBRARY) + set(READLINE_FOUND TRUE) +else(READLINE_INCLUDE_DIR AND READLINE_LIBRARY AND NCURSES_LIBRARY) + FIND_PATH(READLINE_INCLUDE_DIR readline/readline.h + /usr/include/readline + ) + +# 2008-04-22 The next clause used to read like this: +# +# FIND_LIBRARY(READLINE_LIBRARY NAMES readline) +# FIND_LIBRARY(NCURSES_LIBRARY NAMES ncurses ) +# include(FindPackageHandleStandardArgs) +# FIND_PACKAGE_HANDLE_STANDARD_ARGS(Readline DEFAULT_MSG NCURSES_LIBRARY READLINE_INCLUDE_DIR READLINE_LIBRARY ) +# +# I was advised to modify it such that it will find an ncurses library if +# required, but not if one was explicitly given, that is, it allows the +# default to be overridden. PH + + FIND_LIBRARY(READLINE_LIBRARY NAMES readline) + include(FindPackageHandleStandardArgs) + FIND_PACKAGE_HANDLE_STANDARD_ARGS(Readline DEFAULT_MSG READLINE_INCLUDE_DIR READLINE_LIBRARY ) + + MARK_AS_ADVANCED(READLINE_INCLUDE_DIR READLINE_LIBRARY) +endif(READLINE_INCLUDE_DIR AND READLINE_LIBRARY AND NCURSES_LIBRARY) diff --git a/pcre/config-cmake.h.in b/pcre/config-cmake.h.in new file mode 100644 index 00000000000..a93263fa8cc --- /dev/null +++ b/pcre/config-cmake.h.in @@ -0,0 +1,54 @@ +/* config.h for CMake builds */ + +#cmakedefine HAVE_DIRENT_H 1 +#cmakedefine HAVE_SYS_STAT_H 1 +#cmakedefine HAVE_SYS_TYPES_H 1 +#cmakedefine HAVE_UNISTD_H 1 +#cmakedefine HAVE_WINDOWS_H 1 + +#cmakedefine HAVE_TYPE_TRAITS_H 1 +#cmakedefine HAVE_BITS_TYPE_TRAITS_H 1 + +#cmakedefine HAVE_BCOPY 1 +#cmakedefine HAVE_MEMMOVE 1 +#cmakedefine HAVE_STRERROR 1 +#cmakedefine HAVE_STRTOLL 1 +#cmakedefine HAVE_STRTOQ 1 +#cmakedefine HAVE__STRTOI64 1 + +#cmakedefine PCRE_STATIC 1 + +#cmakedefine SUPPORT_PCRE8 1 +#cmakedefine SUPPORT_PCRE16 1 +#cmakedefine SUPPORT_PCRE32 1 +#cmakedefine SUPPORT_JIT 1 +#cmakedefine SUPPORT_PCREGREP_JIT 1 +#cmakedefine SUPPORT_UTF 1 +#cmakedefine SUPPORT_UCP 1 +#cmakedefine EBCDIC 1 +#cmakedefine EBCDIC_NL25 1 +#cmakedefine BSR_ANYCRLF 1 +#cmakedefine NO_RECURSE 1 + +#cmakedefine HAVE_LONG_LONG 1 +#cmakedefine HAVE_UNSIGNED_LONG_LONG 1 + +#cmakedefine SUPPORT_LIBBZ2 1 +#cmakedefine SUPPORT_LIBZ 1 +#cmakedefine SUPPORT_LIBEDIT 1 +#cmakedefine SUPPORT_LIBREADLINE 1 + +#cmakedefine SUPPORT_VALGRIND 1 +#cmakedefine SUPPORT_GCOV 1 + +#define NEWLINE @NEWLINE@ +#define POSIX_MALLOC_THRESHOLD @PCRE_POSIX_MALLOC_THRESHOLD@ +#define LINK_SIZE @PCRE_LINK_SIZE@ +#define MATCH_LIMIT @PCRE_MATCH_LIMIT@ +#define MATCH_LIMIT_RECURSION @PCRE_MATCH_LIMIT_RECURSION@ +#define PCREGREP_BUFSIZE @PCREGREP_BUFSIZE@ + +#define MAX_NAME_SIZE 32 +#define MAX_NAME_COUNT 10000 + +/* end config.h for CMake builds */ diff --git a/pcre/config.h b/pcre/config.h new file mode 100644 index 00000000000..c6357434770 --- /dev/null +++ b/pcre/config.h @@ -0,0 +1,54 @@ +/* config.h for CMake builds */ + +#define HAVE_DIRENT_H 1 +#define HAVE_SYS_STAT_H 1 +#define HAVE_SYS_TYPES_H 1 +#define HAVE_UNISTD_H 1 +/* #undef HAVE_WINDOWS_H */ + +/* #undef HAVE_TYPE_TRAITS_H */ +/* #undef HAVE_BITS_TYPE_TRAITS_H */ + +#define HAVE_BCOPY 1 +#define HAVE_MEMMOVE 1 +#define HAVE_STRERROR 1 +#define HAVE_STRTOLL 1 +#define HAVE_STRTOQ 1 +/* #undef HAVE__STRTOI64 */ + +#define PCRE_STATIC 1 + +#define SUPPORT_PCRE8 1 +/* #undef SUPPORT_PCRE16 */ +/* #undef SUPPORT_PCRE32 */ +/* #undef SUPPORT_JIT */ +#define SUPPORT_PCREGREP_JIT 1 +#define SUPPORT_UTF 1 +#define SUPPORT_UCP 1 +/* #undef EBCDIC */ +/* #undef EBCDIC_NL25 */ +/* #undef BSR_ANYCRLF */ +/* #undef NO_RECURSE */ + +#define HAVE_LONG_LONG 1 +#define HAVE_UNSIGNED_LONG_LONG 1 + +#define SUPPORT_LIBBZ2 1 +#define SUPPORT_LIBZ 1 +/* #undef SUPPORT_LIBEDIT */ +#define SUPPORT_LIBREADLINE 1 + +/* #undef SUPPORT_VALGRIND */ +/* #undef SUPPORT_GCOV */ + +#define NEWLINE 10 +#define POSIX_MALLOC_THRESHOLD 10 +#define LINK_SIZE 2 +#define MATCH_LIMIT 10000000 +#define MATCH_LIMIT_RECURSION MATCH_LIMIT +#define PCREGREP_BUFSIZE 20480 + +#define MAX_NAME_SIZE 32 +#define MAX_NAME_COUNT 10000 + +/* end config.h for CMake builds */ diff --git a/pcre/configure.ac b/pcre/configure.ac new file mode 100644 index 00000000000..e70ceae2336 --- /dev/null +++ b/pcre/configure.ac @@ -0,0 +1,1093 @@ +dnl Process this file with autoconf to produce a configure script. + +dnl NOTE FOR MAINTAINERS: Do not use minor version numbers 08 or 09 because +dnl the leading zeros may cause them to be treated as invalid octal constants +dnl if a PCRE user writes code that uses PCRE_MINOR as a number. There is now +dnl a check further down that throws an error if 08 or 09 are used. + +dnl The PCRE_PRERELEASE feature is for identifying release candidates. It might +dnl be defined as -RC2, for example. For real releases, it should be empty. + +m4_define(pcre_major, [8]) +m4_define(pcre_minor, [33]) +m4_define(pcre_prerelease, []) +m4_define(pcre_date, [2013-05-28]) + +# NOTE: The CMakeLists.txt file searches for the above variables in the first +# 50 lines of this file. Please update that if the variables above are moved. + +# Libtool shared library interface versions (current:revision:age) +m4_define(libpcre_version, [3:1:2]) +m4_define(libpcre16_version, [2:1:2]) +m4_define(libpcre32_version, [0:1:0]) +m4_define(libpcreposix_version, [0:2:0]) +m4_define(libpcrecpp_version, [0:0:0]) + +AC_PREREQ(2.57) +AC_INIT(PCRE, pcre_major.pcre_minor[]pcre_prerelease, , pcre) +AC_CONFIG_SRCDIR([pcre.h.in]) +AM_INIT_AUTOMAKE([dist-bzip2 dist-zip]) +m4_ifdef([AM_SILENT_RULES], [AM_SILENT_RULES([yes])]) +AC_CONFIG_HEADERS(config.h) + +# This is a new thing required to stop a warning from automake 1.12 +m4_ifdef([AM_PROG_AR], [AM_PROG_AR]) + +# This was added at the suggestion of libtoolize (03-Jan-10) +AC_CONFIG_MACRO_DIR([m4]) + +# The default CFLAGS and CXXFLAGS in Autoconf are "-g -O2" for gcc and just +# "-g" for any other compiler. There doesn't seem to be a standard way of +# getting rid of the -g (which I don't think is needed for a production +# library). This fudge seems to achieve the necessary. First, we remember the +# externally set values of CFLAGS and CXXFLAGS. Then call the AC_PROG_CC and +# AC_PROG_CXX macros to find the compilers - if CFLAGS and CXXFLAGS are not +# set, they will be set to Autoconf's defaults. Afterwards, if the original +# values were not set, remove the -g from the Autoconf defaults. +# (PH 02-May-07) + +remember_set_CFLAGS="$CFLAGS" +remember_set_CXXFLAGS="$CXXFLAGS" + +AC_PROG_CC +AC_PROG_CXX +AM_PROG_CC_C_O + +if test "x$remember_set_CFLAGS" = "x" +then + if test "$CFLAGS" = "-g -O2" + then + CFLAGS="-O2" + elif test "$CFLAGS" = "-g" + then + CFLAGS="" + fi +fi + +if test "x$remember_set_CXXFLAGS" = "x" +then + if test "$CXXFLAGS" = "-g -O2" + then + CXXFLAGS="-O2" + elif test "$CXXFLAGS" = "-g" + then + CXXFLAGS="" + fi +fi + +# AC_PROG_CXX will return "g++" even if no c++ compiler is installed. +# Check for that case, and just disable c++ code if g++ doesn't run. +AC_LANG_PUSH(C++) +AC_COMPILE_IFELSE([AC_LANG_PROGRAM([],[])],, CXX=""; CXXCP=""; CXXFLAGS="") +AC_LANG_POP + +# Check for a 64-bit integer type +AC_TYPE_INT64_T + +AC_PROG_INSTALL +AC_LIBTOOL_WIN32_DLL +LT_INIT +AC_PROG_LN_S + +# Check for GCC visibility feature + +PCRE_VISIBILITY + +# Versioning + +PCRE_MAJOR="pcre_major" +PCRE_MINOR="pcre_minor" +PCRE_PRERELEASE="pcre_prerelease" +PCRE_DATE="pcre_date" + +if test "$PCRE_MINOR" = "08" -o "$PCRE_MINOR" = "09" +then + echo "***" + echo "*** Minor version number $PCRE_MINOR must not be used. ***" + echo "*** Use only 01 to 07 or 10 onwards, to avoid octal issues. ***" + echo "***" + exit 1 +fi + +AC_SUBST(PCRE_MAJOR) +AC_SUBST(PCRE_MINOR) +AC_SUBST(PCRE_PRERELEASE) +AC_SUBST(PCRE_DATE) + +# Set a more sensible default value for $(htmldir). +if test "x$htmldir" = 'x${docdir}' +then + htmldir='${docdir}/html' +fi + +# Handle --disable-pcre8 (enabled by default) +AC_ARG_ENABLE(pcre8, + AS_HELP_STRING([--disable-pcre8], + [disable 8 bit character support]), + , enable_pcre8=unset) +AC_SUBST(enable_pcre8) + +# Handle --enable-pcre16 (disabled by default) +AC_ARG_ENABLE(pcre16, + AS_HELP_STRING([--enable-pcre16], + [enable 16 bit character support]), + , enable_pcre16=unset) +AC_SUBST(enable_pcre16) + +# Handle --enable-pcre32 (disabled by default) +AC_ARG_ENABLE(pcre32, + AS_HELP_STRING([--enable-pcre32], + [enable 32 bit character support]), + , enable_pcre32=unset) +AC_SUBST(enable_pcre32) + +# Handle --disable-cpp. The substitution of enable_cpp is needed for use in +# pcre-config. +AC_ARG_ENABLE(cpp, + AS_HELP_STRING([--disable-cpp], + [disable C++ support]), + , enable_cpp=unset) +AC_SUBST(enable_cpp) + +# Handle --enable-jit (disabled by default) +AC_ARG_ENABLE(jit, + AS_HELP_STRING([--enable-jit], + [enable Just-In-Time compiling support]), + , enable_jit=no) + +# Handle --disable-pcregrep-jit (enabled by default) +AC_ARG_ENABLE(pcregrep-jit, + AS_HELP_STRING([--disable-pcregrep-jit], + [disable JIT support in pcregrep]), + , enable_pcregrep_jit=yes) + +# Handle --enable-rebuild-chartables +AC_ARG_ENABLE(rebuild-chartables, + AS_HELP_STRING([--enable-rebuild-chartables], + [rebuild character tables in current locale]), + , enable_rebuild_chartables=no) + +# Handle --enable-utf8 (disabled by default) +AC_ARG_ENABLE(utf8, + AS_HELP_STRING([--enable-utf8], + [another name for --enable-utf. Kept only for compatibility reasons]), + , enable_utf8=unset) + +# Handle --enable-utf (disabled by default) +AC_ARG_ENABLE(utf, + AS_HELP_STRING([--enable-utf], + [enable UTF-8/16/32 support (incompatible with --enable-ebcdic)]), + , enable_utf=unset) + +# Handle --enable-unicode-properties +AC_ARG_ENABLE(unicode-properties, + AS_HELP_STRING([--enable-unicode-properties], + [enable Unicode properties support (implies --enable-utf)]), + , enable_unicode_properties=no) + +# Handle newline options +ac_pcre_newline=lf +AC_ARG_ENABLE(newline-is-cr, + AS_HELP_STRING([--enable-newline-is-cr], + [use CR as newline character]), + ac_pcre_newline=cr) +AC_ARG_ENABLE(newline-is-lf, + AS_HELP_STRING([--enable-newline-is-lf], + [use LF as newline character (default)]), + ac_pcre_newline=lf) +AC_ARG_ENABLE(newline-is-crlf, + AS_HELP_STRING([--enable-newline-is-crlf], + [use CRLF as newline sequence]), + ac_pcre_newline=crlf) +AC_ARG_ENABLE(newline-is-anycrlf, + AS_HELP_STRING([--enable-newline-is-anycrlf], + [use CR, LF, or CRLF as newline sequence]), + ac_pcre_newline=anycrlf) +AC_ARG_ENABLE(newline-is-any, + AS_HELP_STRING([--enable-newline-is-any], + [use any valid Unicode newline sequence]), + ac_pcre_newline=any) +enable_newline="$ac_pcre_newline" + +# Handle --enable-bsr-anycrlf +AC_ARG_ENABLE(bsr-anycrlf, + AS_HELP_STRING([--enable-bsr-anycrlf], + [\R matches only CR, LF, CRLF by default]), + , enable_bsr_anycrlf=no) + +# Handle --enable-ebcdic +AC_ARG_ENABLE(ebcdic, + AS_HELP_STRING([--enable-ebcdic], + [assume EBCDIC coding rather than ASCII; incompatible with --enable-utf; use only in (uncommon) EBCDIC environments; it implies --enable-rebuild-chartables]), + , enable_ebcdic=no) + +# Handle --enable-ebcdic-nl25 +AC_ARG_ENABLE(ebcdic-nl25, + AS_HELP_STRING([--enable-ebcdic-nl25], + [set EBCDIC code for NL to 0x25 instead of 0x15; it implies --enable-ebcdic]), + , enable_ebcdic_nl25=no) + +# Handle --disable-stack-for-recursion +AC_ARG_ENABLE(stack-for-recursion, + AS_HELP_STRING([--disable-stack-for-recursion], + [don't use stack recursion when matching]), + , enable_stack_for_recursion=yes) + +# Handle --enable-pcregrep-libz +AC_ARG_ENABLE(pcregrep-libz, + AS_HELP_STRING([--enable-pcregrep-libz], + [link pcregrep with libz to handle .gz files]), + , enable_pcregrep_libz=no) + +# Handle --enable-pcregrep-libbz2 +AC_ARG_ENABLE(pcregrep-libbz2, + AS_HELP_STRING([--enable-pcregrep-libbz2], + [link pcregrep with libbz2 to handle .bz2 files]), + , enable_pcregrep_libbz2=no) + +# Handle --with-pcregrep-bufsize=N +AC_ARG_WITH(pcregrep-bufsize, + AS_HELP_STRING([--with-pcregrep-bufsize=N], + [pcregrep buffer size (default=20480)]), + , with_pcregrep_bufsize=20480) + +# Handle --enable-pcretest-libedit +AC_ARG_ENABLE(pcretest-libedit, + AS_HELP_STRING([--enable-pcretest-libedit], + [link pcretest with libedit]), + , enable_pcretest_libedit=no) + +# Handle --enable-pcretest-libreadline +AC_ARG_ENABLE(pcretest-libreadline, + AS_HELP_STRING([--enable-pcretest-libreadline], + [link pcretest with libreadline]), + , enable_pcretest_libreadline=no) + +# Handle --with-posix-malloc-threshold=NBYTES +AC_ARG_WITH(posix-malloc-threshold, + AS_HELP_STRING([--with-posix-malloc-threshold=NBYTES], + [threshold for POSIX malloc usage (default=10)]), + , with_posix_malloc_threshold=10) + +# Handle --with-link-size=N +AC_ARG_WITH(link-size, + AS_HELP_STRING([--with-link-size=N], + [internal link size (2, 3, or 4 allowed; default=2)]), + , with_link_size=2) + +# Handle --with-match-limit=N +AC_ARG_WITH(match-limit, + AS_HELP_STRING([--with-match-limit=N], + [default limit on internal looping (default=10000000)]), + , with_match_limit=10000000) + +# Handle --with-match-limit_recursion=N +# +# Note: In config.h, the default is to define MATCH_LIMIT_RECURSION +# symbolically as MATCH_LIMIT, which in turn is defined to be some numeric +# value (e.g. 10000000). MATCH_LIMIT_RECURSION can otherwise be set to some +# different numeric value (or even the same numeric value as MATCH_LIMIT, +# though no longer defined in terms of the latter). +# +AC_ARG_WITH(match-limit-recursion, + AS_HELP_STRING([--with-match-limit-recursion=N], + [default limit on internal recursion (default=MATCH_LIMIT)]), + , with_match_limit_recursion=MATCH_LIMIT) + +# Handle --enable-valgrind +AC_ARG_ENABLE(valgrind, + AS_HELP_STRING([--enable-valgrind], + [valgrind support]), + , enable_valgrind=no) + +# Enable code coverage reports using gcov +AC_ARG_ENABLE(coverage, + AS_HELP_STRING([--enable-coverage], + [enable code coverage reports using gcov]), + , enable_coverage=no) + +# Copy enable_utf8 value to enable_utf for compatibility reasons +if test "x$enable_utf8" != "xunset" +then + if test "x$enable_utf" != "xunset" + then + AC_MSG_ERROR([--enable/disable-utf8 is kept only for compatibility reasons and its value is copied to --enable/disable-utf. Newer code must use --enable/disable-utf alone.]) + fi + enable_utf=$enable_utf8 +fi + +# Set the default value for pcre8 +if test "x$enable_pcre8" = "xunset" +then + enable_pcre8=yes +fi + +# Set the default value for pcre16 +if test "x$enable_pcre16" = "xunset" +then + enable_pcre16=no +fi + +# Set the default value for pcre32 +if test "x$enable_pcre32" = "xunset" +then + enable_pcre32=no +fi + +# Make sure enable_pcre8 or enable_pcre16 was set +if test "x$enable_pcre8$enable_pcre16$enable_pcre32" = "xnonono" +then + AC_MSG_ERROR([At least one of 8, 16 or 32 bit pcre library must be enabled]) +fi + +# Make sure that if enable_unicode_properties was set, that UTF support is enabled. +if test "x$enable_unicode_properties" = "xyes" +then + if test "x$enable_utf" = "xno" + then + AC_MSG_ERROR([support for Unicode properties requires UTF-8/16/32 support]) + fi + enable_utf=yes +fi + +# enable_utf is disabled by default. +if test "x$enable_utf" = "xunset" +then + enable_utf=no +fi + +# enable_cpp copies the value of enable_pcre8 by default +if test "x$enable_cpp" = "xunset" +then + enable_cpp=$enable_pcre8 +fi + +# Make sure that if enable_cpp was set, that enable_pcre8 support is enabled +if test "x$enable_cpp" = "xyes" +then + if test "x$enable_pcre8" = "xno" + then + AC_MSG_ERROR([C++ library requires pcre library with 8 bit characters]) + fi +fi + +# Convert the newline identifier into the appropriate integer value. The first +# three are ASCII values 0x0a, 0x0d, and 0x0d0a, but if EBCDIC is enabled, they +# are changed below. + +case "$enable_newline" in + lf) ac_pcre_newline_value=10 ;; + cr) ac_pcre_newline_value=13 ;; + crlf) ac_pcre_newline_value=3338 ;; + anycrlf) ac_pcre_newline_value=-2 ;; + any) ac_pcre_newline_value=-1 ;; + *) + AC_MSG_ERROR([invalid argument \"$enable_newline\" to --enable-newline option]) + ;; +esac + +# --enable-ebcdic-nl25 implies --enable-ebcdic +if test "x$enable_ebcdic_nl25" = "xyes"; then + enable_ebcdic=yes +fi + +# Make sure that if enable_ebcdic is set, rebuild_chartables is also enabled, +# and the newline value is adjusted appropriately (CR is still 13, but LF is +# 21 or 37). Also check that UTF support is not requested, because PCRE cannot +# handle EBCDIC and UTF in the same build. To do so it would need to use +# different character constants depending on the mode. +# +if test "x$enable_ebcdic" = "xyes"; then + enable_rebuild_chartables=yes + + if test "x$enable_utf" = "xyes"; then + AC_MSG_ERROR([support for EBCDIC and UTF-8/16/32 cannot be enabled at the same time]) + fi + + if test "x$enable_ebcdic_nl25" = "xno"; then + case "$ac_pcre_newline_value" in + 10) ac_pcre_newline_value=21 ;; + 3338) ac_pcre_newline_value=3349 ;; + esac + else + case "$ac_pcre_newline_value" in + 10) ac_pcre_newline_value=37 ;; + 3338) ac_pcre_newline_value=3365 ;; + esac + fi +fi + +# Check argument to --with-link-size +case "$with_link_size" in + 2|3|4) ;; + *) + AC_MSG_ERROR([invalid argument \"$with_link_size\" to --with-link-size option]) + ;; +esac + +AH_TOP([ +/* PCRE is written in Standard C, but there are a few non-standard things it +can cope with, allowing it to run on SunOS4 and other "close to standard" +systems. + +In environments that support the GNU autotools, config.h.in is converted into +config.h by the "configure" script. In environments that use CMake, +config-cmake.in is converted into config.h. If you are going to build PCRE "by +hand" without using "configure" or CMake, you should copy the distributed +config.h.generic to config.h, and edit the macro definitions to be the way you +need them. You must then add -DHAVE_CONFIG_H to all of your compile commands, +so that config.h is included at the start of every source. + +Alternatively, you can avoid editing by using -D on the compiler command line +to set the macro values. In this case, you do not have to set -DHAVE_CONFIG_H, +but if you do, default values will be taken from config.h for non-boolean +macros that are not defined on the command line. + +Boolean macros such as HAVE_STDLIB_H and SUPPORT_PCRE8 should either be defined +(conventionally to 1) for TRUE, and not defined at all for FALSE. All such +macros are listed as a commented #undef in config.h.generic. Macros such as +MATCH_LIMIT, whose actual value is relevant, have defaults defined, but are +surrounded by #ifndef/#endif lines so that the value can be overridden by -D. + +PCRE uses memmove() if HAVE_MEMMOVE is defined; otherwise it uses bcopy() if +HAVE_BCOPY is defined. If your system has neither bcopy() nor memmove(), make +sure both macros are undefined; an emulation function will then be used. */]) + +# Checks for header files. +AC_HEADER_STDC +AC_CHECK_HEADERS(limits.h sys/types.h sys/stat.h dirent.h windows.h) + +# The files below are C++ header files. +pcre_have_type_traits="0" +pcre_have_bits_type_traits="0" + +if test "x$enable_cpp" = "xyes" -a -z "$CXX"; then + AC_MSG_ERROR([You need a C++ compiler for C++ support.]) +fi + +if test "x$enable_cpp" = "xyes" -a -n "$CXX" +then +AC_LANG_PUSH(C++) + +# Older versions of pcre defined pcrecpp::no_arg, but in new versions +# it's called pcrecpp::RE::no_arg. For backwards ABI compatibility, +# we want to make one an alias for the other. Different systems do +# this in different ways. Some systems, for instance, can do it via +# a linker flag: -alias (for os x 10.5) or -i (for os x <=10.4). +OLD_LDFLAGS="$LDFLAGS" +for flag in "-alias,__ZN7pcrecpp2RE6no_argE,__ZN7pcrecpp6no_argE" \ + "-i__ZN7pcrecpp6no_argE:__ZN7pcrecpp2RE6no_argE"; do + AC_MSG_CHECKING([for alias support in the linker]) + LDFLAGS="$OLD_LDFLAGS -Wl,$flag" + # We try to run the linker with this new ld flag. If the link fails, + # we give up and remove the new flag from LDFLAGS. + AC_LINK_IFELSE([AC_LANG_PROGRAM([namespace pcrecpp { + class RE { static int no_arg; }; + int RE::no_arg; + }], + [])], + [AC_MSG_RESULT([yes]); + EXTRA_LIBPCRECPP_LDFLAGS="$EXTRA_LIBPCRECPP_LDFLAGS -Wl,$flag"; + break;], + AC_MSG_RESULT([no])) +done +LDFLAGS="$OLD_LDFLAGS" + +# We could be more clever here, given we're doing AC_SUBST with this +# (eg set a var to be the name of the include file we want). But we're not +# so it's easy to change back to 'regular' autoconf vars if we needed to. +AC_CHECK_HEADERS(string, [pcre_have_cpp_headers="1"], + [pcre_have_cpp_headers="0"]) +AC_CHECK_HEADERS(bits/type_traits.h, [pcre_have_bits_type_traits="1"], + [pcre_have_bits_type_traits="0"]) +AC_CHECK_HEADERS(type_traits.h, [pcre_have_type_traits="1"], + [pcre_have_type_traits="0"]) + +# (This isn't c++-specific, but is only used in pcrecpp.cc, so try this +# in a c++ context. This matters becuase strtoimax is C99 and may not +# be supported by the C++ compiler.) +# Figure out how to create a longlong from a string: strtoll and +# equiv. It's not enough to call AC_CHECK_FUNCS: hpux has a +# strtoll, for instance, but it only takes 2 args instead of 3! +# We have to call AH_TEMPLATE since AC_DEFINE_UNQUOTED below is complex. +AH_TEMPLATE(HAVE_STRTOQ, [Define to 1 if you have `strtoq'.]) +AH_TEMPLATE(HAVE_STRTOLL, [Define to 1 if you have `strtoll'.]) +AH_TEMPLATE(HAVE__STRTOI64, [Define to 1 if you have `_strtoi64'.]) +AH_TEMPLATE(HAVE_STRTOIMAX, [Define to 1 if you have `strtoimax'.]) +have_strto_fn=0 +for fn in strtoq strtoll _strtoi64 strtoimax; do + AC_MSG_CHECKING([for $fn]) + if test "$fn" = strtoimax; then + include=stdint.h + else + include=stdlib.h + fi + AC_COMPILE_IFELSE([AC_LANG_PROGRAM([#include <$include>], + [char* e; return $fn("100", &e, 10)])], + [AC_MSG_RESULT(yes) + AC_DEFINE_UNQUOTED(HAVE_`echo $fn | tr abcdefghijklmnopqrstuvwxyz ABCDEFGHIJKLMNOPQRSTUVWXYZ`, 1, + [Define to 1 if you have `$fn'.]) + have_strto_fn=1 + break], + [AC_MSG_RESULT(no)]) +done + +if test "$have_strto_fn" = 1; then + AC_CHECK_TYPES([long long], + [pcre_have_long_long="1"], + [pcre_have_long_long="0"]) + AC_CHECK_TYPES([unsigned long long], + [pcre_have_ulong_long="1"], + [pcre_have_ulong_long="0"]) +else + pcre_have_long_long="0" + pcre_have_ulong_long="0" +fi +AC_SUBST(pcre_have_long_long) +AC_SUBST(pcre_have_ulong_long) + +AC_LANG_POP +fi +# Using AC_SUBST eliminates the need to include config.h in a public .h file +AC_SUBST(pcre_have_type_traits) +AC_SUBST(pcre_have_bits_type_traits) + +# Conditional compilation +AM_CONDITIONAL(WITH_PCRE8, test "x$enable_pcre8" = "xyes") +AM_CONDITIONAL(WITH_PCRE16, test "x$enable_pcre16" = "xyes") +AM_CONDITIONAL(WITH_PCRE32, test "x$enable_pcre32" = "xyes") +AM_CONDITIONAL(WITH_PCRE_CPP, test "x$enable_cpp" = "xyes") +AM_CONDITIONAL(WITH_REBUILD_CHARTABLES, test "x$enable_rebuild_chartables" = "xyes") +AM_CONDITIONAL(WITH_JIT, test "x$enable_jit" = "xyes") +AM_CONDITIONAL(WITH_UTF, test "x$enable_utf" = "xyes") +AM_CONDITIONAL(WITH_VALGRIND, test "x$enable_valgrind" = "xyes") + +# Checks for typedefs, structures, and compiler characteristics. + +AC_C_CONST +AC_TYPE_SIZE_T + +# Checks for library functions. + +AC_CHECK_FUNCS(bcopy memmove strerror) + +# Check for the availability of libz (aka zlib) + +AC_CHECK_HEADERS([zlib.h], [HAVE_ZLIB_H=1]) +AC_CHECK_LIB([z], [gzopen], [HAVE_LIBZ=1]) + +# Check for the availability of libbz2. Originally we just used AC_CHECK_LIB, +# as for libz. However, this had the following problem, diagnosed and fixed by +# a user: +# +# - libbz2 uses the Pascal calling convention (WINAPI) for the functions +# under Win32. +# - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h", +# therefore missing the function definition. +# - The compiler thus generates a "C" signature for the test function. +# - The linker fails to find the "C" function. +# - PCRE fails to configure if asked to do so against libbz2. +# +# Solution: +# +# - Replace the AC_CHECK_LIB test with a custom test. + +AC_CHECK_HEADERS([bzlib.h], [HAVE_BZLIB_H=1]) +# Original test +# AC_CHECK_LIB([bz2], [BZ2_bzopen], [HAVE_LIBBZ2=1]) +# +# Custom test follows + +AC_MSG_CHECKING([for libbz2]) +OLD_LIBS="$LIBS" +LIBS="$LIBS -lbz2" +AC_LINK_IFELSE([AC_LANG_PROGRAM([[ +#ifdef HAVE_BZLIB_H +#include +#endif]], +[[return (int)BZ2_bzopen("conftest", "rb");]])], +[AC_MSG_RESULT([yes]);HAVE_LIBBZ2=1; break;], +AC_MSG_RESULT([no])) +LIBS="$OLD_LIBS" + +# Check for the availabiity of libreadline + +if test "$enable_pcretest_libreadline" = "yes"; then + AC_CHECK_HEADERS([readline/readline.h], [HAVE_READLINE_H=1]) + AC_CHECK_HEADERS([readline/history.h], [HAVE_HISTORY_H=1]) + AC_CHECK_LIB([readline], [readline], [LIBREADLINE="-lreadline"], + [unset ac_cv_lib_readline_readline; + AC_CHECK_LIB([readline], [readline], [LIBREADLINE="-ltinfo"], + [unset ac_cv_lib_readline_readline; + AC_CHECK_LIB([readline], [readline], [LIBREADLINE="-lcurses"], + [unset ac_cv_lib_readline_readline; + AC_CHECK_LIB([readline], [readline], [LIBREADLINE="-lncurses"], + [unset ac_cv_lib_readline_readline; + AC_CHECK_LIB([readline], [readline], [LIBREADLINE="-lncursesw"], + [unset ac_cv_lib_readline_readline; + AC_CHECK_LIB([readline], [readline], [LIBREADLINE="-ltermcap"], + [LIBREADLINE=""], + [-ltermcap])], + [-lncursesw])], + [-lncurses])], + [-lcurses])], + [-ltinfo])]) + AC_SUBST(LIBREADLINE) + if test -n "$LIBREADLINE"; then + if test "$LIBREADLINE" != "-lreadline"; then + echo "-lreadline needs $LIBREADLINE" + LIBREADLINE="-lreadline $LIBREADLINE" + fi + fi +fi + + +# Check for the availability of libedit. Different distributions put its +# headers in different places. Try to cover the most common ones. + +if test "$enable_pcretest_libedit" = "yes"; then + AC_CHECK_HEADERS([editline/readline.h], [HAVE_EDITLINE_READLINE_H=1], + [AC_CHECK_HEADERS([edit/readline/readline.h], [HAVE_READLINE_READLINE_H=1], + [AC_CHECK_HEADERS([readline/readline.h], [HAVE_READLINE_READLINE_H=1])])]) + AC_CHECK_LIB([edit], [readline], [LIBEDIT="-ledit"]) +fi + +# This facilitates -ansi builds under Linux +dnl AC_DEFINE([_GNU_SOURCE], [], [Enable GNU extensions in glibc]) + +PCRE_STATIC_CFLAG="" +if test "x$enable_shared" = "xno" ; then + AC_DEFINE([PCRE_STATIC], [1], [ + Define to any value if linking statically (TODO: make nice with Libtool)]) + PCRE_STATIC_CFLAG="-DPCRE_STATIC" +fi +AC_SUBST(PCRE_STATIC_CFLAG) + +# Here is where pcre specific defines are handled + +if test "$enable_pcre8" = "yes"; then + AC_DEFINE([SUPPORT_PCRE8], [], [ + Define to any value to enable the 8 bit PCRE library.]) +fi + +if test "$enable_pcre16" = "yes"; then + AC_DEFINE([SUPPORT_PCRE16], [], [ + Define to any value to enable the 16 bit PCRE library.]) +fi + +if test "$enable_pcre32" = "yes"; then + AC_DEFINE([SUPPORT_PCRE32], [], [ + Define to any value to enable the 32 bit PCRE library.]) +fi + +if test "$enable_jit" = "yes"; then + AX_PTHREAD([], [AC_MSG_ERROR([JIT support requires pthreads])]) + CC="$PTHREAD_CC" + CFLAGS="$PTHREAD_CFLAGS $CFLAGS" + LIBS="$PTHREAD_LIBS $LIBS" + AC_DEFINE([SUPPORT_JIT], [], [ + Define to any value to enable support for Just-In-Time compiling.]) +else + enable_pcregrep_jit="no" +fi + +if test "$enable_pcregrep_jit" = "yes"; then + AC_DEFINE([SUPPORT_PCREGREP_JIT], [], [ + Define to any value to enable JIT support in pcregrep.]) +fi + +if test "$enable_utf" = "yes"; then + AC_DEFINE([SUPPORT_UTF], [], [ + Define to any value to enable support for the UTF-8/16/32 Unicode encoding. + This will work even in an EBCDIC environment, but it is incompatible + with the EBCDIC macro. That is, PCRE can support *either* EBCDIC + code *or* ASCII/UTF-8/16/32, but not both at once.]) +fi + +if test "$enable_unicode_properties" = "yes"; then + AC_DEFINE([SUPPORT_UCP], [], [ + Define to any value to enable support for Unicode properties.]) +fi + +if test "$enable_stack_for_recursion" = "no"; then + AC_DEFINE([NO_RECURSE], [], [ + PCRE uses recursive function calls to handle backtracking while + matching. This can sometimes be a problem on systems that have + stacks of limited size. Define NO_RECURSE to any value to get a + version that doesn't use recursion in the match() function; instead + it creates its own stack by steam using pcre_recurse_malloc() to obtain + memory from the heap. For more detail, see the comments and other stuff + just above the match() function.]) +fi + +if test "$enable_pcregrep_libz" = "yes"; then + AC_DEFINE([SUPPORT_LIBZ], [], [ + Define to any value to allow pcregrep to be linked with libz, so that it is + able to handle .gz files.]) +fi + +if test "$enable_pcregrep_libbz2" = "yes"; then + AC_DEFINE([SUPPORT_LIBBZ2], [], [ + Define to any value to allow pcregrep to be linked with libbz2, so that it + is able to handle .bz2 files.]) +fi + +if test $with_pcregrep_bufsize -lt 8192 ; then + with_pcregrep_bufsize="8192" +fi + +AC_DEFINE_UNQUOTED([PCREGREP_BUFSIZE], [$with_pcregrep_bufsize], [ + The value of PCREGREP_BUFSIZE determines the size of buffer used by pcregrep + to hold parts of the file it is searching. This is also the minimum value. + The actual amount of memory used by pcregrep is three times this number, + because it allows for the buffering of "before" and "after" lines.]) + +if test "$enable_pcretest_libedit" = "yes"; then + AC_DEFINE([SUPPORT_LIBEDIT], [], [ + Define to any value to allow pcretest to be linked with libedit.]) + LIBREADLINE="$LIBEDIT" +elif test "$enable_pcretest_libreadline" = "yes"; then + AC_DEFINE([SUPPORT_LIBREADLINE], [], [ + Define to any value to allow pcretest to be linked with libreadline.]) +fi + +AC_DEFINE_UNQUOTED([NEWLINE], [$ac_pcre_newline_value], [ + The value of NEWLINE determines the default newline character sequence. PCRE + client programs can override this by selecting other values at run time. In + ASCII environments, the value can be 10 (LF), 13 (CR), or 3338 (CRLF); in + EBCDIC environments the value can be 21 or 37 (LF), 13 (CR), or 3349 or 3365 + (CRLF) because there are two alternative codepoints (0x15 and 0x25) that are + used as the NL line terminator that is equivalent to ASCII LF. In both ASCII + and EBCDIC environments the value can also be -1 (ANY), or -2 (ANYCRLF).]) + +if test "$enable_bsr_anycrlf" = "yes"; then + AC_DEFINE([BSR_ANYCRLF], [], [ + By default, the \R escape sequence matches any Unicode line ending + character or sequence of characters. If BSR_ANYCRLF is defined (to any + value), this is changed so that backslash-R matches only CR, LF, or CRLF. + The build-time default can be overridden by the user of PCRE at runtime.]) +fi + +AC_DEFINE_UNQUOTED([LINK_SIZE], [$with_link_size], [ + The value of LINK_SIZE determines the number of bytes used to store + links as offsets within the compiled regex. The default is 2, which + allows for compiled patterns up to 64K long. This covers the vast + majority of cases. However, PCRE can also be compiled to use 3 or 4 + bytes instead. This allows for longer patterns in extreme cases.]) + +AC_DEFINE_UNQUOTED([POSIX_MALLOC_THRESHOLD], [$with_posix_malloc_threshold], [ + When calling PCRE via the POSIX interface, additional working storage + is required for holding the pointers to capturing substrings because + PCRE requires three integers per substring, whereas the POSIX + interface provides only two. If the number of expected substrings is + small, the wrapper function uses space on the stack, because this is + faster than using malloc() for each call. The threshold above which + the stack is no longer used is defined by POSIX_MALLOC_THRESHOLD.]) + +AC_DEFINE_UNQUOTED([MATCH_LIMIT], [$with_match_limit], [ + The value of MATCH_LIMIT determines the default number of times the + internal match() function can be called during a single execution of + pcre_exec(). There is a runtime interface for setting a different + limit. The limit exists in order to catch runaway regular + expressions that take for ever to determine that they do not match. + The default is set very large so that it does not accidentally catch + legitimate cases.]) + +AC_DEFINE_UNQUOTED([MATCH_LIMIT_RECURSION], [$with_match_limit_recursion], [ + The above limit applies to all calls of match(), whether or not they + increase the recursion depth. In some environments it is desirable + to limit the depth of recursive calls of match() more strictly, in + order to restrict the maximum amount of stack (or heap, if + NO_RECURSE is defined) that is used. The value of + MATCH_LIMIT_RECURSION applies only to recursive calls of match(). To + have any useful effect, it must be less than the value of + MATCH_LIMIT. The default is to use the same value as MATCH_LIMIT. + There is a runtime method for setting a different limit.]) + +AC_DEFINE([MAX_NAME_SIZE], [32], [ + This limit is parameterized just in case anybody ever wants to + change it. Care must be taken if it is increased, because it guards + against integer overflow caused by enormously large patterns.]) + +AC_DEFINE([MAX_NAME_COUNT], [10000], [ + This limit is parameterized just in case anybody ever wants to + change it. Care must be taken if it is increased, because it guards + against integer overflow caused by enormously large patterns.]) + +AH_VERBATIM([PCRE_EXP_DEFN], [ +/* If you are compiling for a system other than a Unix-like system or + Win32, and it needs some magic to be inserted before the definition + of a function that is exported by the library, define this macro to + contain the relevant magic. If you do not define this macro, a suitable + __declspec value is used for Windows systems; in other environments + "extern" is used for a C compiler and "extern C" for a C++ compiler. + This macro apears at the start of every exported function that is part + of the external API. It does not appear on functions that are "external" + in the C sense, but which are internal to the library. */ +#undef PCRE_EXP_DEFN]) + +if test "$enable_ebcdic" = "yes"; then + AC_DEFINE_UNQUOTED([EBCDIC], [], [ + If you are compiling for a system that uses EBCDIC instead of ASCII + character codes, define this macro to any value. You must also edit the + NEWLINE macro below to set a suitable EBCDIC newline, commonly 21 (0x15). + On systems that can use "configure" or CMake to set EBCDIC, NEWLINE is + automatically adjusted. When EBCDIC is set, PCRE assumes that all input + strings are in EBCDIC. If you do not define this macro, PCRE will assume + input strings are ASCII or UTF-8/16/32 Unicode. It is not possible to build + a version of PCRE that supports both EBCDIC and UTF-8/16/32.]) +fi + +if test "$enable_ebcdic_nl25" = "yes"; then + AC_DEFINE_UNQUOTED([EBCDIC_NL25], [], [ + In an EBCDIC environment, define this macro to any value to arrange for + the NL character to be 0x25 instead of the default 0x15. NL plays the role + that LF does in an ASCII/Unicode environment. The value must also be set in + the NEWLINE macro below. On systems that can use "configure" or CMake to + set EBCDIC_NL25, the adjustment of NEWLINE is automatic.]) +fi + +if test "$enable_valgrind" = "yes"; then + AC_DEFINE_UNQUOTED([SUPPORT_VALGRIND], [], [ + Define to any value for valgrind support to find invalid memory reads.]) +fi + +# Platform specific issues +NO_UNDEFINED= +EXPORT_ALL_SYMBOLS= +case $host_os in + cygwin* | mingw* ) + if test X"$enable_shared" = Xyes; then + NO_UNDEFINED="-no-undefined" + EXPORT_ALL_SYMBOLS="-Wl,--export-all-symbols" + fi + ;; +esac + +# The extra LDFLAGS for each particular library +# (Note: The libpcre*_version bits are m4 variables, assigned above) + +EXTRA_LIBPCRE_LDFLAGS="$EXTRA_LIBPCRE_LDFLAGS \ + $NO_UNDEFINED -version-info libpcre_version" + +EXTRA_LIBPCRE16_LDFLAGS="$EXTRA_LIBPCRE16_LDFLAGS \ + $NO_UNDEFINED -version-info libpcre16_version" + +EXTRA_LIBPCRE32_LDFLAGS="$EXTRA_LIBPCRE32_LDFLAGS \ + $NO_UNDEFINED -version-info libpcre32_version" + +EXTRA_LIBPCREPOSIX_LDFLAGS="$EXTRA_LIBPCREPOSIX_LDFLAGS \ + $NO_UNDEFINED -version-info libpcreposix_version" + +EXTRA_LIBPCRECPP_LDFLAGS="$EXTRA_LIBPCRECPP_LDFLAGS \ + $NO_UNDEFINED -version-info libpcrecpp_version \ + $EXPORT_ALL_SYMBOLS" + +AC_SUBST(EXTRA_LIBPCRE_LDFLAGS) +AC_SUBST(EXTRA_LIBPCRE16_LDFLAGS) +AC_SUBST(EXTRA_LIBPCRE32_LDFLAGS) +AC_SUBST(EXTRA_LIBPCREPOSIX_LDFLAGS) +AC_SUBST(EXTRA_LIBPCRECPP_LDFLAGS) + +# When we run 'make distcheck', use these arguments. Turning off compiler +# optimization makes it run faster. +DISTCHECK_CONFIGURE_FLAGS="CFLAGS='' CXXFLAGS='' --enable-pcre16 --enable-pcre32 --enable-jit --enable-cpp --enable-unicode-properties" +AC_SUBST(DISTCHECK_CONFIGURE_FLAGS) + +# Check that, if --enable-pcregrep-libz or --enable-pcregrep-libbz2 is +# specified, the relevant library is available. + +if test "$enable_pcregrep_libz" = "yes"; then + if test "$HAVE_ZLIB_H" != "1"; then + echo "** Cannot --enable-pcregrep-libz because zlib.h was not found" + exit 1 + fi + if test "$HAVE_LIBZ" != "1"; then + echo "** Cannot --enable-pcregrep-libz because libz was not found" + exit 1 + fi + LIBZ="-lz" +fi +AC_SUBST(LIBZ) + +if test "$enable_pcregrep_libbz2" = "yes"; then + if test "$HAVE_BZLIB_H" != "1"; then + echo "** Cannot --enable-pcregrep-libbz2 because bzlib.h was not found" + exit 1 + fi + if test "$HAVE_LIBBZ2" != "1"; then + echo "** Cannot --enable-pcregrep-libbz2 because libbz2 was not found" + exit 1 + fi + LIBBZ2="-lbz2" +fi +AC_SUBST(LIBBZ2) + +# Similarly for --enable-pcretest-readline + +if test "$enable_pcretest_libedit" = "yes"; then + if test "$enable_pcretest_libreadline" = "yes"; then + echo "** Cannot use both --enable-pcretest-libedit and --enable-pcretest-readline" + exit 1 + fi + if test "$HAVE_EDITLINE_READLINE_H" != "1" -a \ + "$HAVE_READLINE_READLINE_H" != "1"; then + echo "** Cannot --enable-pcretest-libedit because neither editline/readline.h" + echo "** nor readline/readline.h was found." + exit 1 + fi + if test -z "$LIBEDIT"; then + echo "** Cannot --enable-pcretest-libedit because libedit library was not found." + exit 1 + fi +fi + +if test "$enable_pcretest_libreadline" = "yes"; then + if test "$HAVE_READLINE_H" != "1"; then + echo "** Cannot --enable-pcretest-readline because readline/readline.h was not found." + exit 1 + fi + if test "$HAVE_HISTORY_H" != "1"; then + echo "** Cannot --enable-pcretest-readline because readline/history.h was not found." + exit 1 + fi + if test -z "$LIBREADLINE"; then + echo "** Cannot --enable-pcretest-readline because readline library was not found." + exit 1 + fi +fi + +# Check for valgrind + +if test "$enable_valgrind" = "yes"; then + m4_ifdef([PKG_CHECK_MODULES], + [PKG_CHECK_MODULES([VALGRIND],[valgrind])], + [AC_MSG_ERROR([pkg-config not supported])]) +fi + +# test code coverage reporting +if test "$enable_coverage" = "yes"; then + if test "x$GCC" != "xyes"; then + AC_MSG_ERROR([Code coverage reports can only be generated when using GCC]) + fi + + # ccache is incompatible with gcov + AC_PATH_PROG([SHTOOL],[shtool],[false]) + case `$SHTOOL path $CC` in + *ccache*) cc_ccache=yes;; + *) cc_ccache=no;; + esac + + if test "$cc_ccache" = "yes"; then + if test -z "$CCACHE_DISABLE" -o "$CCACHE_DISABLE" != "1"; then + AC_MSG_ERROR([must export CCACHE_DISABLE=1 to disable ccache for code coverage]) + fi + fi + + AC_ARG_VAR([LCOV],[the ltp lcov program]) + AC_PATH_PROG([LCOV],[lcov],[false]) + if test "x$LCOV" = "xfalse"; then + AC_MSG_ERROR([lcov not found]) + fi + + AC_ARG_VAR([GENHTML],[the ltp genhtml program]) + AC_PATH_PROG([GENHTML],[genhtml],[false]) + if test "x$GENHTML" = "xfalse"; then + AC_MSG_ERROR([genhtml not found]) + fi + + AC_DEFINE([SUPPORT_GCOV],[1], [ + Define to allow pcretest and pcregrep to be linked with gcov, so that they + are able to generate code coverage reports.]) + + # And add flags needed for gcov + GCOV_CFLAGS="-O0 -ggdb3 -fprofile-arcs -ftest-coverage" + GCOV_CXXFLAGS="-O0 -ggdb3 -fprofile-arcs -ftest-coverage" + GCOV_LIBS="-lgcov" + AC_SUBST([GCOV_CFLAGS]) + AC_SUBST([GCOV_CXXFLAGS]) + AC_SUBST([GCOV_LIBS]) +fi # enable_coverage + +AM_CONDITIONAL([WITH_GCOV],[test "x$enable_coverage" = "xyes"]) + +# Produce these files, in addition to config.h. +AC_CONFIG_FILES( + Makefile + libpcre.pc + libpcre16.pc + libpcre32.pc + libpcreposix.pc + libpcrecpp.pc + pcre-config + pcre.h + pcre_stringpiece.h + pcrecpparg.h +) + +# Make the generated script files executable. +AC_CONFIG_COMMANDS([script-chmod], [chmod a+x pcre-config]) + +# Make sure that pcre_chartables.c is removed in case the method for +# creating it was changed by reconfiguration. +AC_CONFIG_COMMANDS([delete-old-chartables], [rm -f pcre_chartables.c]) + +AC_OUTPUT + +# Print out a nice little message after configure is run displaying the +# chosen options. + +ebcdic_nl_code=n/a +if test "$enable_ebcdic_nl25" = "yes"; then + ebcdic_nl_code=0x25 +elif test "$enable_ebcdic" = "yes"; then + ebcdic_nl_code=0x15 +fi + +cat < Delete Cache". + +1. Install the latest CMake version available from http://www.cmake.org/, and + ensure that cmake\bin is on your path. + +2. Unzip (retaining folder structure) the PCRE source tree into a source + directory such as C:\pcre. You should ensure your local date and time + is not earlier than the file dates in your source dir if the release is + very new. + +3. Create a new, empty build directory, preferably a subdirectory of the + source dir. For example, C:\pcre\pcre-xx\build. + +4. Run cmake-gui from the Shell envirornment of your build tool, for example, + Msys for Msys/MinGW or Visual Studio Command Prompt for VC/VC++. Do not try + to start Cmake from the Windows Start menu, as this can lead to errors. + +5. Enter C:\pcre\pcre-xx and C:\pcre\pcre-xx\build for the source and build + directories, respectively. + +6. Hit the "Configure" button. + +7. Select the particular IDE / build tool that you are using (Visual + Studio, MSYS makefiles, MinGW makefiles, etc.) + +8. The GUI will then list several configuration options. This is where + you can enable UTF-8 support or other PCRE optional features. + +9. Hit "Configure" again. The adjacent "Generate" button should now be + active. + +10. Hit "Generate". + +11. The build directory should now contain a usable build system, be it a + solution file for Visual Studio, makefiles for MinGW, etc. Exit from + cmake-gui and use the generated build system with your compiler or IDE. + E.g., for MinGW you can run "make", or for Visual Studio, open the PCRE + solution, select the desired configuration (Debug, or Release, etc.) and + build the ALL_BUILD project. + +12. If during configuration with cmake-gui you've elected to build the test + programs, you can execute them by building the test project. E.g., for + MinGW: "make test"; for Visual Studio build the RUN_TESTS project. The + most recent build configuration is targeted by the tests. A summary of + test results is presented. Complete test output is subsequently + available for review in Testing\Temporary under your build dir. + + +USE OF RELATIVE PATHS WITH CMAKE ON WINDOWS + +A PCRE user comments as follows: I thought that others may want to know the +current state of CMAKE_USE_RELATIVE_PATHS support on Windows. Here it is: + +-- AdditionalIncludeDirectories is only partially modified (only the + first path - see below) +-- Only some of the contained file paths are modified - shown below for + pcre.vcproj +-- It properly modifies + +I am sure CMake people can fix that if they want to. Until then one will +need to replace existing absolute paths in project files with relative +paths manually (e.g. from VS) - relative to project file location. I did +just that before being told to try CMAKE_USE_RELATIVE_PATHS. Not a big +deal. + +AdditionalIncludeDirectories="E:\builds\pcre\build;E:\builds\pcre\pcre-7.5;" +AdditionalIncludeDirectories=".;E:\builds\pcre\pcre-7.5;" + +RelativePath="pcre.h" +RelativePath="pcre_chartables.c" +RelativePath="pcre_chartables.c.rule" + + +TESTING WITH RUNTEST.BAT + +If configured with CMake, building the test project ("make test" or building +ALL_TESTS in Visual Studio) creates (and runs) pcre_test.bat (and depending +on your configuration options, possibly other test programs) in the build +directory. Pcre_test.bat runs RunTest.Bat with correct source and exe paths. + +For manual testing with RunTest.bat, provided the build dir is a subdirectory +of the source directory: Open command shell window. Chdir to the location +of your pcretest.exe and pcregrep.exe programs. Call RunTest.bat with +"..\RunTest.Bat" or "..\..\RunTest.bat" as appropriate. + +To run only a particular test with RunTest.Bat provide a test number argument. + +Otherwise: + +1. Copy RunTest.bat into the directory where pcretest.exe and pcregrep.exe + have been created. + +2. Edit RunTest.bat to indentify the full or relative location of + the pcre source (wherein which the testdata folder resides), e.g.: + + set srcdir=C:\pcre\pcre-8.20 + +3. In a Windows command environment, chdir to the location of your bat and + exe programs. + +4. Run RunTest.bat. Test outputs will automatically be compared to expected + results, and discrepancies will be identified in the console output. + +To independently test the just-in-time compiler, run pcre_jit_test.exe. +To test pcrecpp, run pcrecpp_unittest.exe, pcre_stringpiece_unittest.exe and +pcre_scanner_unittest.exe. + + +BUILDING UNDER WINDOWS CE WITH VISUAL STUDIO 200x + +Vincent Richomme sent a zip archive of files to help with this process. They +can be found in the file "pcre-vsbuild.zip" in the Contrib directory of the FTP +site. + + +BUILDING UNDER WINDOWS WITH BCC5.5 + +Michael Roy sent these comments about building PCRE under Windows with BCC5.5: + +Some of the core BCC libraries have a version of PCRE from 1998 built in, which +can lead to pcre_exec() giving an erroneous PCRE_ERROR_NULL from a version +mismatch. I'm including an easy workaround below, if you'd like to include it +in the non-unix instructions: + +When linking a project with BCC5.5, pcre.lib must be included before any of the +libraries cw32.lib, cw32i.lib, cw32mt.lib, and cw32mti.lib on the command line. + + +BUILDING USING BORLAND C++ BUILDER 2007 (CB2007) AND HIGHER + +A PCRE user sent these comments about this environment (see also the comment +from another user that follows them): + +The XE versions of C++ Builder come with a RegularExpressionsCore class which +contain a version of TPerlRegEx. However, direct use of the C PCRE library may +be desirable. + +The default makevp.bat, however, supplied with PCRE builds a version of PCRE +that is not usable with any version of C++ Builder because the compiler ships +with an embedded version of PCRE, version 2.01 from 1998! [See also the note +about BCC5.5 above.] If you want to use PCRE you'll need to rename the +functions (pcre_compile to pcre_compile_bcc, etc) or do as I have done and just +use the 16 bit versions. I'm using std::wstring everywhere anyway. Since the +embedded version of PCRE does not have the 16 bit function names, there is no +conflict. + +Building PCRE using a C++ Builder static library project file (recommended): + +1. Rename or remove pcre.h, pcreposi.h, and pcreposix.h from your C++ Builder +original include path. + +2. Download PCRE from pcre.org and extract to a directory. + +3. Rename pcre_chartables.c.dist to pcre_chartables.c, pcre.h.generic to +pcre.h, and config.h.generic to config.h. + +4. Edit pcre.h and pcre_config.c so that they include config.h. + +5. Edit config.h like so: + +Comment out the following lines: +#define PACKAGE "pcre" +#define PACKAGE_BUGREPORT "" +#define PACKAGE_NAME "PCRE" +#define PACKAGE_STRING "PCRE 8.32" +#define PACKAGE_TARNAME "pcre" +#define PACKAGE_URL "" +#define PACKAGE_VERSION "8.32" + +Add the following lines: +#ifndef SUPPORT_UTF +#define SUPPORT_UTF 100 // any value is fine +#endif + +#ifndef SUPPORT_UCP +#define SUPPORT_UCP 101 // any value is fine +#endif + +#ifndef SUPPORT_UCP +#define SUPPORT_PCRE16 102 // any value is fine +#endif + +#ifndef SUPPORT_UTF8 +#define SUPPORT_UTF8 103 // any value is fine +#endif + +6. Build a C++ Builder project using the IDE. Go to File / New / Other and +choose Static Library. You can name it pcre.cbproj or whatever. Now set your +paths by going to Project / Options. Set the Include path. Do this from the +"Base" option to apply to both Release and Debug builds. Now add the following +files to the project: + +pcre.h +pcre16_byte_order.c +pcre16_chartables.c +pcre16_compile.c +pcre16_config.c +pcre16_dfa_exec.c +pcre16_exec.c +pcre16_fullinfo.c +pcre16_get.c +pcre16_globals.c +pcre16_maketables.c +pcre16_newline.c +pcre16_ord2utf16.c +pcre16_printint.c +pcre16_refcount.c +pcre16_string_utils.c +pcre16_study.c +pcre16_tables.c +pcre16_ucd.c +pcre16_utf16_utils.c +pcre16_valid_utf16.c +pcre16_version.c +pcre16_xclass.c + +//Optional +pcre_version.c + +7. After compiling the .lib file, copy the .lib and header files to a project +you want to use PCRE with. Enjoy. + +Optional ... Building PCRE using the makevp.bat file: + +1. Edit makevp_c.txt and makevp_l.txt and change all the names to the 16 bit +versions. + +2. Edit makevp.bat and set the path to C++ Builder. Run makevp.bat. + +Another PCRE user added this comment: + +Another approach I successfully used for some years with BCB 5 and 6 was to +make sure that include and library paths of PCRE are configured before the +default paths of the IDE in the dialogs where one can manage those paths. +Afterwards one can open the project files using a text editor and manually add +the self created library for pcre itself, pcrecpp doesn't ship with the IDE, in +the library nodes where the IDE manages its own libraries to link against in +front of the IDE-own libraries. This way one can use the default PCRE function +names without getting access violations on runtime. + + + + +BUILDING PCRE ON OPENVMS + +Stephen Hoffman sent the following, in December 2012: + +"Here is a very short write-up on the +OpenVMS port and here + + + +is a zip with the OpenVMS files, and with one modified testing-related PCRE +file." This is a port of PCRE 8.32. + +Earlier, Dan Mooney sent the following comments about building PCRE on OpenVMS. +They relate to an older version of PCRE that used fewer source files, so the +exact commands will need changing. See the current list of source files above. + +"It was quite easy to compile and link the library. I don't have a formal +make file but the attached file [reproduced below] contains the OpenVMS DCL +commands I used to build the library. I had to add #define +POSIX_MALLOC_THRESHOLD 10 to pcre.h since it was not defined anywhere. + +The library was built on: +O/S: HP OpenVMS v7.3-1 +Compiler: Compaq C v6.5-001-48BCD +Linker: vA13-01 + +The test results did not match 100% due to the issues you mention in your +documentation regarding isprint(), iscntrl(), isgraph() and ispunct(). I +modified some of the character tables temporarily and was able to get the +results to match. Tests using the fr locale did not match since I don't have +that locale loaded. The study size was always reported to be 3 less than the +value in the standard test output files." + +========================= +$! This DCL procedure builds PCRE on OpenVMS +$! +$! I followed the instructions in the non-unix-use file in the distribution. +$! +$ COMPILE == "CC/LIST/NOMEMBER_ALIGNMENT/PREFIX_LIBRARY_ENTRIES=ALL_ENTRIES +$ COMPILE DFTABLES.C +$ LINK/EXE=DFTABLES.EXE DFTABLES.OBJ +$ RUN DFTABLES.EXE/OUTPUT=CHARTABLES.C +$ COMPILE MAKETABLES.C +$ COMPILE GET.C +$ COMPILE STUDY.C +$! I had to set POSIX_MALLOC_THRESHOLD to 10 in PCRE.H since the symbol +$! did not seem to be defined anywhere. +$! I edited pcre.h and added #DEFINE SUPPORT_UTF8 to enable UTF8 support. +$ COMPILE PCRE.C +$ LIB/CREATE PCRE MAKETABLES.OBJ, GET.OBJ, STUDY.OBJ, PCRE.OBJ +$! I had to set POSIX_MALLOC_THRESHOLD to 10 in PCRE.H since the symbol +$! did not seem to be defined anywhere. +$ COMPILE PCREPOSIX.C +$ LIB/CREATE PCREPOSIX PCREPOSIX.OBJ +$ COMPILE PCRETEST.C +$ LINK/EXE=PCRETEST.EXE PCRETEST.OBJ, PCRE/LIB, PCREPOSIX/LIB +$! C programs that want access to command line arguments must be +$! defined as a symbol +$ PCRETEST :== "$ SYS$ROADSUSERS:[DMOONEY.REGEXP]PCRETEST.EXE" +$! Arguments must be enclosed in quotes. +$ PCRETEST "-C" +$! Test results: +$! +$! The test results did not match 100%. The functions isprint(), iscntrl(), +$! isgraph() and ispunct() on OpenVMS must not produce the same results +$! as the system that built the test output files provided with the +$! distribution. +$! +$! The study size did not match and was always 3 less on OpenVMS. +$! +$! Locale could not be set to fr +$! +========================= + + +BUILDING PCRE ON STRATUS OPENVOS + +These notes on the port of PCRE to VOS (lightly edited) were supplied by +Ashutosh Warikoo, whose email address has the local part awarikoo and the +domain nse.co.in. The port was for version 7.9 in August 2009. + +1. Building PCRE + +I built pcre on OpenVOS Release 17.0.1at using GNU Tools 3.4a without any +problems. I used the following packages to build PCRE: + + ftp://ftp.stratus.com/pub/vos/posix/ga/posix.save.evf.gz + +Please read and follow the instructions that come with these packages. To start +the build of pcre, from the root of the package type: + + ./build.sh + +2. Installing PCRE + +Once you have successfully built PCRE, login to the SysAdmin group, switch to +the root user, and type + + [ !create_dir (master_disk)>usr --if needed ] + [ !create_dir (master_disk)>usr>local --if needed ] + !gmake install + +This installs PCRE and its man pages into /usr/local. You can add +(master_disk)>usr>local>bin to your command search paths, or if you are in +BASH, add /usr/local/bin to the PATH environment variable. + +4. Restrictions + +This port requires readline library optionally. However during the build I +faced some yet unexplored errors while linking with readline. As it was an +optional component I chose to disable it. + +5. Known Problems + +I ran the test suite, but you will have to be your own judge of whether this +command, and this port, suits your purposes. If you find any problems that +appear to be related to the port itself, please let me know. Please see the +build.log file in the root of the package also. + + +BUILDING PCRE ON NATIVE Z/OS AND Z/VM + +z/OS and z/VM are operating systems for mainframe computers, produced by IBM. +The character code used is EBCDIC, not ASCII or Unicode. In z/OS, UNIX APIs and +applications can be supported through UNIX System Services, and in such an +environment PCRE can be built in the same way as in other systems. However, in +native z/OS (without UNIX System Services) and in z/VM, special ports are +required. For details, please see this web site: + + http://www.zaconsultants.net + +There is also a mirror here: + + http://www.vsoft-software.com/downloads.html + +========================== +Last Updated: 14 May 2013 diff --git a/pcre/doc/html/README.txt b/pcre/doc/html/README.txt new file mode 100644 index 00000000000..e14935efeba --- /dev/null +++ b/pcre/doc/html/README.txt @@ -0,0 +1,974 @@ +README file for PCRE (Perl-compatible regular expression library) +----------------------------------------------------------------- + +The latest release of PCRE is always available in three alternative formats +from: + + ftp://ftp.csx.cam.ac.uk/pub/software/programming/pcre/pcre-xxx.tar.gz + ftp://ftp.csx.cam.ac.uk/pub/software/programming/pcre/pcre-xxx.tar.bz2 + ftp://ftp.csx.cam.ac.uk/pub/software/programming/pcre/pcre-xxx.zip + +There is a mailing list for discussion about the development of PCRE at + + pcre-dev@exim.org + +Please read the NEWS file if you are upgrading from a previous release. +The contents of this README file are: + + The PCRE APIs + Documentation for PCRE + Contributions by users of PCRE + Building PCRE on non-Unix-like systems + Building PCRE without using autotools + Building PCRE using autotools + Retrieving configuration information + Shared libraries + Cross-compiling using autotools + Using HP's ANSI C++ compiler (aCC) + Compiling in Tru64 using native compilers + Using Sun's compilers for Solaris + Using PCRE from MySQL + Making new tarballs + Testing PCRE + Character tables + File manifest + + +The PCRE APIs +------------- + +PCRE is written in C, and it has its own API. There are three sets of +functions, one for the 8-bit library, which processes strings of bytes, one for +the 16-bit library, which processes strings of 16-bit values, and one for the +32-bit library, which processes strings of 32-bit values. The distribution also +includes a set of C++ wrapper functions (see the pcrecpp man page for details), +courtesy of Google Inc., which can be used to call the 8-bit PCRE library from +C++. + +In addition, there is a set of C wrapper functions (again, just for the 8-bit +library) that are based on the POSIX regular expression API (see the pcreposix +man page). These end up in the library called libpcreposix. Note that this just +provides a POSIX calling interface to PCRE; the regular expressions themselves +still follow Perl syntax and semantics. The POSIX API is restricted, and does +not give full access to all of PCRE's facilities. + +The header file for the POSIX-style functions is called pcreposix.h. The +official POSIX name is regex.h, but I did not want to risk possible problems +with existing files of that name by distributing it that way. To use PCRE with +an existing program that uses the POSIX API, pcreposix.h will have to be +renamed or pointed at by a link. + +If you are using the POSIX interface to PCRE and there is already a POSIX regex +library installed on your system, as well as worrying about the regex.h header +file (as mentioned above), you must also take care when linking programs to +ensure that they link with PCRE's libpcreposix library. Otherwise they may pick +up the POSIX functions of the same name from the other library. + +One way of avoiding this confusion is to compile PCRE with the addition of +-Dregcomp=PCREregcomp (and similarly for the other POSIX functions) to the +compiler flags (CFLAGS if you are using "configure" -- see below). This has the +effect of renaming the functions so that the names no longer clash. Of course, +you have to do the same thing for your applications, or write them using the +new names. + + +Documentation for PCRE +---------------------- + +If you install PCRE in the normal way on a Unix-like system, you will end up +with a set of man pages whose names all start with "pcre". The one that is just +called "pcre" lists all the others. In addition to these man pages, the PCRE +documentation is supplied in two other forms: + + 1. There are files called doc/pcre.txt, doc/pcregrep.txt, and + doc/pcretest.txt in the source distribution. The first of these is a + concatenation of the text forms of all the section 3 man pages except + those that summarize individual functions. The other two are the text + forms of the section 1 man pages for the pcregrep and pcretest commands. + These text forms are provided for ease of scanning with text editors or + similar tools. They are installed in /share/doc/pcre, where + is the installation prefix (defaulting to /usr/local). + + 2. A set of files containing all the documentation in HTML form, hyperlinked + in various ways, and rooted in a file called index.html, is distributed in + doc/html and installed in /share/doc/pcre/html. + +Users of PCRE have contributed files containing the documentation for various +releases in CHM format. These can be found in the Contrib directory of the FTP +site (see next section). + + +Contributions by users of PCRE +------------------------------ + +You can find contributions from PCRE users in the directory + + ftp://ftp.csx.cam.ac.uk/pub/software/programming/pcre/Contrib + +There is a README file giving brief descriptions of what they are. Some are +complete in themselves; others are pointers to URLs containing relevant files. +Some of this material is likely to be well out-of-date. Several of the earlier +contributions provided support for compiling PCRE on various flavours of +Windows (I myself do not use Windows). Nowadays there is more Windows support +in the standard distribution, so these contibutions have been archived. + + +Building PCRE on non-Unix-like systems +-------------------------------------- + +For a non-Unix-like system, please read the comments in the file +NON-AUTOTOOLS-BUILD, though if your system supports the use of "configure" and +"make" you may be able to build PCRE using autotools in the same way as for +many Unix-like systems. + +PCRE can also be configured using the GUI facility provided by CMake's +cmake-gui command. This creates Makefiles, solution files, etc. The file +NON-AUTOTOOLS-BUILD has information about CMake. + +PCRE has been compiled on many different operating systems. It should be +straightforward to build PCRE on any system that has a Standard C compiler and +library, because it uses only Standard C functions. + + +Building PCRE without using autotools +------------------------------------- + +The use of autotools (in particular, libtool) is problematic in some +environments, even some that are Unix or Unix-like. See the NON-AUTOTOOLS-BUILD +file for ways of building PCRE without using autotools. + + +Building PCRE using autotools +----------------------------- + +If you are using HP's ANSI C++ compiler (aCC), please see the special note +in the section entitled "Using HP's ANSI C++ compiler (aCC)" below. + +The following instructions assume the use of the widely used "configure; make; +make install" (autotools) process. + +To build PCRE on system that supports autotools, first run the "configure" +command from the PCRE distribution directory, with your current directory set +to the directory where you want the files to be created. This command is a +standard GNU "autoconf" configuration script, for which generic instructions +are supplied in the file INSTALL. + +Most commonly, people build PCRE within its own distribution directory, and in +this case, on many systems, just running "./configure" is sufficient. However, +the usual methods of changing standard defaults are available. For example: + +CFLAGS='-O2 -Wall' ./configure --prefix=/opt/local + +This command specifies that the C compiler should be run with the flags '-O2 +-Wall' instead of the default, and that "make install" should install PCRE +under /opt/local instead of the default /usr/local. + +If you want to build in a different directory, just run "configure" with that +directory as current. For example, suppose you have unpacked the PCRE source +into /source/pcre/pcre-xxx, but you want to build it in /build/pcre/pcre-xxx: + +cd /build/pcre/pcre-xxx +/source/pcre/pcre-xxx/configure + +PCRE is written in C and is normally compiled as a C library. However, it is +possible to build it as a C++ library, though the provided building apparatus +does not have any features to support this. + +There are some optional features that can be included or omitted from the PCRE +library. They are also documented in the pcrebuild man page. + +. By default, both shared and static libraries are built. You can change this + by adding one of these options to the "configure" command: + + --disable-shared + --disable-static + + (See also "Shared libraries on Unix-like systems" below.) + +. By default, only the 8-bit library is built. If you add --enable-pcre16 to + the "configure" command, the 16-bit library is also built. If you add + --enable-pcre32 to the "configure" command, the 32-bit library is also built. + If you want only the 16-bit or 32-bit library, use --disable-pcre8 to disable + building the 8-bit library. + +. If you are building the 8-bit library and want to suppress the building of + the C++ wrapper library, you can add --disable-cpp to the "configure" + command. Otherwise, when "configure" is run without --disable-pcre8, it will + try to find a C++ compiler and C++ header files, and if it succeeds, it will + try to build the C++ wrapper. + +. If you want to include support for just-in-time compiling, which can give + large performance improvements on certain platforms, add --enable-jit to the + "configure" command. This support is available only for certain hardware + architectures. If you try to enable it on an unsupported architecture, there + will be a compile time error. + +. When JIT support is enabled, pcregrep automatically makes use of it, unless + you add --disable-pcregrep-jit to the "configure" command. + +. If you want to make use of the support for UTF-8 Unicode character strings in + the 8-bit library, or UTF-16 Unicode character strings in the 16-bit library, + or UTF-32 Unicode character strings in the 32-bit library, you must add + --enable-utf to the "configure" command. Without it, the code for handling + UTF-8, UTF-16 and UTF-8 is not included in the relevant library. Even + when --enable-utf is included, the use of a UTF encoding still has to be + enabled by an option at run time. When PCRE is compiled with this option, its + input can only either be ASCII or UTF-8/16/32, even when running on EBCDIC + platforms. It is not possible to use both --enable-utf and --enable-ebcdic at + the same time. + +. There are no separate options for enabling UTF-8, UTF-16 and UTF-32 + independently because that would allow ridiculous settings such as requesting + UTF-16 support while building only the 8-bit library. However, the option + --enable-utf8 is retained for backwards compatibility with earlier releases + that did not support 16-bit or 32-bit character strings. It is synonymous with + --enable-utf. It is not possible to configure one library with UTF support + and the other without in the same configuration. + +. If, in addition to support for UTF-8/16/32 character strings, you want to + include support for the \P, \p, and \X sequences that recognize Unicode + character properties, you must add --enable-unicode-properties to the + "configure" command. This adds about 30K to the size of the library (in the + form of a property table); only the basic two-letter properties such as Lu + are supported. + +. You can build PCRE to recognize either CR or LF or the sequence CRLF or any + of the preceding, or any of the Unicode newline sequences as indicating the + end of a line. Whatever you specify at build time is the default; the caller + of PCRE can change the selection at run time. The default newline indicator + is a single LF character (the Unix standard). You can specify the default + newline indicator by adding --enable-newline-is-cr or --enable-newline-is-lf + or --enable-newline-is-crlf or --enable-newline-is-anycrlf or + --enable-newline-is-any to the "configure" command, respectively. + + If you specify --enable-newline-is-cr or --enable-newline-is-crlf, some of + the standard tests will fail, because the lines in the test files end with + LF. Even if the files are edited to change the line endings, there are likely + to be some failures. With --enable-newline-is-anycrlf or + --enable-newline-is-any, many tests should succeed, but there may be some + failures. + +. By default, the sequence \R in a pattern matches any Unicode line ending + sequence. This is independent of the option specifying what PCRE considers to + be the end of a line (see above). However, the caller of PCRE can restrict \R + to match only CR, LF, or CRLF. You can make this the default by adding + --enable-bsr-anycrlf to the "configure" command (bsr = "backslash R"). + +. When called via the POSIX interface, PCRE uses malloc() to get additional + storage for processing capturing parentheses if there are more than 10 of + them in a pattern. You can increase this threshold by setting, for example, + + --with-posix-malloc-threshold=20 + + on the "configure" command. + +. PCRE has a counter that can be set to limit the amount of resources it uses. + If the limit is exceeded during a match, the match fails. The default is ten + million. You can change the default by setting, for example, + + --with-match-limit=500000 + + on the "configure" command. This is just the default; individual calls to + pcre_exec() can supply their own value. There is more discussion on the + pcreapi man page. + +. There is a separate counter that limits the depth of recursive function calls + during a matching process. This also has a default of ten million, which is + essentially "unlimited". You can change the default by setting, for example, + + --with-match-limit-recursion=500000 + + Recursive function calls use up the runtime stack; running out of stack can + cause programs to crash in strange ways. There is a discussion about stack + sizes in the pcrestack man page. + +. The default maximum compiled pattern size is around 64K. You can increase + this by adding --with-link-size=3 to the "configure" command. In the 8-bit + library, PCRE then uses three bytes instead of two for offsets to different + parts of the compiled pattern. In the 16-bit library, --with-link-size=3 is + the same as --with-link-size=4, which (in both libraries) uses four-byte + offsets. Increasing the internal link size reduces performance. In the 32-bit + library, the only supported link size is 4. + +. You can build PCRE so that its internal match() function that is called from + pcre_exec() does not call itself recursively. Instead, it uses memory blocks + obtained from the heap via the special functions pcre_stack_malloc() and + pcre_stack_free() to save data that would otherwise be saved on the stack. To + build PCRE like this, use + + --disable-stack-for-recursion + + on the "configure" command. PCRE runs more slowly in this mode, but it may be + necessary in environments with limited stack sizes. This applies only to the + normal execution of the pcre_exec() function; if JIT support is being + successfully used, it is not relevant. Equally, it does not apply to + pcre_dfa_exec(), which does not use deeply nested recursion. There is a + discussion about stack sizes in the pcrestack man page. + +. For speed, PCRE uses four tables for manipulating and identifying characters + whose code point values are less than 256. By default, it uses a set of + tables for ASCII encoding that is part of the distribution. If you specify + + --enable-rebuild-chartables + + a program called dftables is compiled and run in the default C locale when + you obey "make". It builds a source file called pcre_chartables.c. If you do + not specify this option, pcre_chartables.c is created as a copy of + pcre_chartables.c.dist. See "Character tables" below for further information. + +. It is possible to compile PCRE for use on systems that use EBCDIC as their + character code (as opposed to ASCII/Unicode) by specifying + + --enable-ebcdic + + This automatically implies --enable-rebuild-chartables (see above). However, + when PCRE is built this way, it always operates in EBCDIC. It cannot support + both EBCDIC and UTF-8/16/32. There is a second option, --enable-ebcdic-nl25, + which specifies that the code value for the EBCDIC NL character is 0x25 + instead of the default 0x15. + +. In environments where valgrind is installed, if you specify + + --enable-valgrind + + PCRE will use valgrind annotations to mark certain memory regions as + unaddressable. This allows it to detect invalid memory accesses, and is + mostly useful for debugging PCRE itself. + +. In environments where the gcc compiler is used and lcov version 1.6 or above + is installed, if you specify + + --enable-coverage + + the build process implements a code coverage report for the test suite. The + report is generated by running "make coverage". If ccache is installed on + your system, it must be disabled when building PCRE for coverage reporting. + You can do this by setting the environment variable CCACHE_DISABLE=1 before + running "make" to build PCRE. + +. The pcregrep program currently supports only 8-bit data files, and so + requires the 8-bit PCRE library. It is possible to compile pcregrep to use + libz and/or libbz2, in order to read .gz and .bz2 files (respectively), by + specifying one or both of + + --enable-pcregrep-libz + --enable-pcregrep-libbz2 + + Of course, the relevant libraries must be installed on your system. + +. The default size of internal buffer used by pcregrep can be set by, for + example: + + --with-pcregrep-bufsize=50K + + The default value is 20K. + +. It is possible to compile pcretest so that it links with the libreadline + or libedit libraries, by specifying, respectively, + + --enable-pcretest-libreadline or --enable-pcretest-libedit + + If this is done, when pcretest's input is from a terminal, it reads it using + the readline() function. This provides line-editing and history facilities. + Note that libreadline is GPL-licenced, so if you distribute a binary of + pcretest linked in this way, there may be licensing issues. These can be + avoided by linking with libedit (which has a BSD licence) instead. + + Enabling libreadline causes the -lreadline option to be added to the pcretest + build. In many operating environments with a sytem-installed readline + library this is sufficient. However, in some environments (e.g. if an + unmodified distribution version of readline is in use), it may be necessary + to specify something like LIBS="-lncurses" as well. This is because, to quote + the readline INSTALL, "Readline uses the termcap functions, but does not link + with the termcap or curses library itself, allowing applications which link + with readline the to choose an appropriate library." If you get error + messages about missing functions tgetstr, tgetent, tputs, tgetflag, or tgoto, + this is the problem, and linking with the ncurses library should fix it. + +The "configure" script builds the following files for the basic C library: + +. Makefile the makefile that builds the library +. config.h build-time configuration options for the library +. pcre.h the public PCRE header file +. pcre-config script that shows the building settings such as CFLAGS + that were set for "configure" +. libpcre.pc ) data for the pkg-config command +. libpcre16.pc ) +. libpcre32.pc ) +. libpcreposix.pc ) +. libtool script that builds shared and/or static libraries + +Versions of config.h and pcre.h are distributed in the PCRE tarballs under the +names config.h.generic and pcre.h.generic. These are provided for those who +have to built PCRE without using "configure" or CMake. If you use "configure" +or CMake, the .generic versions are not used. + +When building the 8-bit library, if a C++ compiler is found, the following +files are also built: + +. libpcrecpp.pc data for the pkg-config command +. pcrecpparg.h header file for calling PCRE via the C++ wrapper +. pcre_stringpiece.h header for the C++ "stringpiece" functions + +The "configure" script also creates config.status, which is an executable +script that can be run to recreate the configuration, and config.log, which +contains compiler output from tests that "configure" runs. + +Once "configure" has run, you can run "make". This builds the the libraries +libpcre, libpcre16 and/or libpcre32, and a test program called pcretest. If you +enabled JIT support with --enable-jit, a test program called pcre_jit_test is +built as well. + +If the 8-bit library is built, libpcreposix and the pcregrep command are also +built, and if a C++ compiler was found on your system, and you did not disable +it with --disable-cpp, "make" builds the C++ wrapper library, which is called +libpcrecpp, as well as some test programs called pcrecpp_unittest, +pcre_scanner_unittest, and pcre_stringpiece_unittest. + +The command "make check" runs all the appropriate tests. Details of the PCRE +tests are given below in a separate section of this document. + +You can use "make install" to install PCRE into live directories on your +system. The following are installed (file names are all relative to the + that is set when "configure" is run): + + Commands (bin): + pcretest + pcregrep (if 8-bit support is enabled) + pcre-config + + Libraries (lib): + libpcre16 (if 16-bit support is enabled) + libpcre32 (if 32-bit support is enabled) + libpcre (if 8-bit support is enabled) + libpcreposix (if 8-bit support is enabled) + libpcrecpp (if 8-bit and C++ support is enabled) + + Configuration information (lib/pkgconfig): + libpcre16.pc + libpcre32.pc + libpcre.pc + libpcreposix.pc + libpcrecpp.pc (if C++ support is enabled) + + Header files (include): + pcre.h + pcreposix.h + pcre_scanner.h ) + pcre_stringpiece.h ) if C++ support is enabled + pcrecpp.h ) + pcrecpparg.h ) + + Man pages (share/man/man{1,3}): + pcregrep.1 + pcretest.1 + pcre-config.1 + pcre.3 + pcre*.3 (lots more pages, all starting "pcre") + + HTML documentation (share/doc/pcre/html): + index.html + *.html (lots more pages, hyperlinked from index.html) + + Text file documentation (share/doc/pcre): + AUTHORS + COPYING + ChangeLog + LICENCE + NEWS + README + pcre.txt (a concatenation of the man(3) pages) + pcretest.txt the pcretest man page + pcregrep.txt the pcregrep man page + pcre-config.txt the pcre-config man page + +If you want to remove PCRE from your system, you can run "make uninstall". +This removes all the files that "make install" installed. However, it does not +remove any directories, because these are often shared with other programs. + + +Retrieving configuration information +------------------------------------ + +Running "make install" installs the command pcre-config, which can be used to +recall information about the PCRE configuration and installation. For example: + + pcre-config --version + +prints the version number, and + + pcre-config --libs + +outputs information about where the library is installed. This command can be +included in makefiles for programs that use PCRE, saving the programmer from +having to remember too many details. + +The pkg-config command is another system for saving and retrieving information +about installed libraries. Instead of separate commands for each library, a +single command is used. For example: + + pkg-config --cflags pcre + +The data is held in *.pc files that are installed in a directory called +/lib/pkgconfig. + + +Shared libraries +---------------- + +The default distribution builds PCRE as shared libraries and static libraries, +as long as the operating system supports shared libraries. Shared library +support relies on the "libtool" script which is built as part of the +"configure" process. + +The libtool script is used to compile and link both shared and static +libraries. They are placed in a subdirectory called .libs when they are newly +built. The programs pcretest and pcregrep are built to use these uninstalled +libraries (by means of wrapper scripts in the case of shared libraries). When +you use "make install" to install shared libraries, pcregrep and pcretest are +automatically re-built to use the newly installed shared libraries before being +installed themselves. However, the versions left in the build directory still +use the uninstalled libraries. + +To build PCRE using static libraries only you must use --disable-shared when +configuring it. For example: + +./configure --prefix=/usr/gnu --disable-shared + +Then run "make" in the usual way. Similarly, you can use --disable-static to +build only shared libraries. + + +Cross-compiling using autotools +------------------------------- + +You can specify CC and CFLAGS in the normal way to the "configure" command, in +order to cross-compile PCRE for some other host. However, you should NOT +specify --enable-rebuild-chartables, because if you do, the dftables.c source +file is compiled and run on the local host, in order to generate the inbuilt +character tables (the pcre_chartables.c file). This will probably not work, +because dftables.c needs to be compiled with the local compiler, not the cross +compiler. + +When --enable-rebuild-chartables is not specified, pcre_chartables.c is created +by making a copy of pcre_chartables.c.dist, which is a default set of tables +that assumes ASCII code. Cross-compiling with the default tables should not be +a problem. + +If you need to modify the character tables when cross-compiling, you should +move pcre_chartables.c.dist out of the way, then compile dftables.c by hand and +run it on the local host to make a new version of pcre_chartables.c.dist. +Then when you cross-compile PCRE this new version of the tables will be used. + + +Using HP's ANSI C++ compiler (aCC) +---------------------------------- + +Unless C++ support is disabled by specifying the "--disable-cpp" option of the +"configure" script, you must include the "-AA" option in the CXXFLAGS +environment variable in order for the C++ components to compile correctly. + +Also, note that the aCC compiler on PA-RISC platforms may have a defect whereby +needed libraries fail to get included when specifying the "-AA" compiler +option. If you experience unresolved symbols when linking the C++ programs, +use the workaround of specifying the following environment variable prior to +running the "configure" script: + + CXXLDFLAGS="-lstd_v2 -lCsup_v2" + + +Compiling in Tru64 using native compilers +----------------------------------------- + +The following error may occur when compiling with native compilers in the Tru64 +operating system: + + CXX libpcrecpp_la-pcrecpp.lo +cxx: Error: /usr/lib/cmplrs/cxx/V7.1-006/include/cxx/iosfwd, line 58: #error + directive: "cannot include iosfwd -- define __USE_STD_IOSTREAM to + override default - see section 7.1.2 of the C++ Using Guide" +#error "cannot include iosfwd -- define __USE_STD_IOSTREAM to override default +- see section 7.1.2 of the C++ Using Guide" + +This may be followed by other errors, complaining that 'namespace "std" has no +member'. The solution to this is to add the line + +#define __USE_STD_IOSTREAM 1 + +to the config.h file. + + +Using Sun's compilers for Solaris +--------------------------------- + +A user reports that the following configurations work on Solaris 9 sparcv9 and +Solaris 9 x86 (32-bit): + + Solaris 9 sparcv9: ./configure --disable-cpp CC=/bin/cc CFLAGS="-m64 -g" + Solaris 9 x86: ./configure --disable-cpp CC=/bin/cc CFLAGS="-g" + + +Using PCRE from MySQL +--------------------- + +On systems where both PCRE and MySQL are installed, it is possible to make use +of PCRE from within MySQL, as an alternative to the built-in pattern matching. +There is a web page that tells you how to do this: + + http://www.mysqludf.org/lib_mysqludf_preg/index.php + + +Making new tarballs +------------------- + +The command "make dist" creates three PCRE tarballs, in tar.gz, tar.bz2, and +zip formats. The command "make distcheck" does the same, but then does a trial +build of the new distribution to ensure that it works. + +If you have modified any of the man page sources in the doc directory, you +should first run the PrepareRelease script before making a distribution. This +script creates the .txt and HTML forms of the documentation from the man pages. + + +Testing PCRE +------------ + +To test the basic PCRE library on a Unix-like system, run the RunTest script. +There is another script called RunGrepTest that tests the options of the +pcregrep command. If the C++ wrapper library is built, three test programs +called pcrecpp_unittest, pcre_scanner_unittest, and pcre_stringpiece_unittest +are also built. When JIT support is enabled, another test program called +pcre_jit_test is built. + +Both the scripts and all the program tests are run if you obey "make check" or +"make test". For other environments, see the instructions in +NON-AUTOTOOLS-BUILD. + +The RunTest script runs the pcretest test program (which is documented in its +own man page) on each of the relevant testinput files in the testdata +directory, and compares the output with the contents of the corresponding +testoutput files. RunTest uses a file called testtry to hold the main output +from pcretest. Other files whose names begin with "test" are used as working +files in some tests. + +Some tests are relevant only when certain build-time options were selected. For +example, the tests for UTF-8/16/32 support are run only if --enable-utf was +used. RunTest outputs a comment when it skips a test. + +Many of the tests that are not skipped are run up to three times. The second +run forces pcre_study() to be called for all patterns except for a few in some +tests that are marked "never study" (see the pcretest program for how this is +done). If JIT support is available, the non-DFA tests are run a third time, +this time with a forced pcre_study() with the PCRE_STUDY_JIT_COMPILE option. +This testing can be suppressed by putting "nojit" on the RunTest command line. + +The entire set of tests is run once for each of the 8-bit, 16-bit and 32-bit +libraries that are enabled. If you want to run just one set of tests, call +RunTest with either the -8, -16 or -32 option. + +If valgrind is installed, you can run the tests under it by putting "valgrind" +on the RunTest command line. To run pcretest on just one or more specific test +files, give their numbers as arguments to RunTest, for example: + + RunTest 2 7 11 + +You can also specify ranges of tests such as 3-6 or 3- (meaning 3 to the +end), or a number preceded by ~ to exclude a test. For example: + + Runtest 3-15 ~10 + +This runs tests 3 to 15, excluding test 10, and just ~13 runs all the tests +except test 13. Whatever order the arguments are in, the tests are always run +in numerical order. + +You can also call RunTest with the single argument "list" to cause it to output +a list of tests. + +The first test file can be fed directly into the perltest.pl script to check +that Perl gives the same results. The only difference you should see is in the +first few lines, where the Perl version is given instead of the PCRE version. + +The second set of tests check pcre_fullinfo(), pcre_study(), +pcre_copy_substring(), pcre_get_substring(), pcre_get_substring_list(), error +detection, and run-time flags that are specific to PCRE, as well as the POSIX +wrapper API. It also uses the debugging flags to check some of the internals of +pcre_compile(). + +If you build PCRE with a locale setting that is not the standard C locale, the +character tables may be different (see next paragraph). In some cases, this may +cause failures in the second set of tests. For example, in a locale where the +isprint() function yields TRUE for characters in the range 128-255, the use of +[:isascii:] inside a character class defines a different set of characters, and +this shows up in this test as a difference in the compiled code, which is being +listed for checking. Where the comparison test output contains [\x00-\x7f] the +test will contain [\x00-\xff], and similarly in some other cases. This is not a +bug in PCRE. + +The third set of tests checks pcre_maketables(), the facility for building a +set of character tables for a specific locale and using them instead of the +default tables. The tests make use of the "fr_FR" (French) locale. Before +running the test, the script checks for the presence of this locale by running +the "locale" command. If that command fails, or if it doesn't include "fr_FR" +in the list of available locales, the third test cannot be run, and a comment +is output to say why. If running this test produces instances of the error + + ** Failed to set locale "fr_FR" + +in the comparison output, it means that locale is not available on your system, +despite being listed by "locale". This does not mean that PCRE is broken. + +[If you are trying to run this test on Windows, you may be able to get it to +work by changing "fr_FR" to "french" everywhere it occurs. Alternatively, use +RunTest.bat. The version of RunTest.bat included with PCRE 7.4 and above uses +Windows versions of test 2. More info on using RunTest.bat is included in the +document entitled NON-UNIX-USE.] + +The fourth and fifth tests check the UTF-8/16/32 support and error handling and +internal UTF features of PCRE that are not relevant to Perl, respectively. The +sixth and seventh tests do the same for Unicode character properties support. + +The eighth, ninth, and tenth tests check the pcre_dfa_exec() alternative +matching function, in non-UTF-8/16/32 mode, UTF-8/16/32 mode, and UTF-8/16/32 +mode with Unicode property support, respectively. + +The eleventh test checks some internal offsets and code size features; it is +run only when the default "link size" of 2 is set (in other cases the sizes +change) and when Unicode property support is enabled. + +The twelfth test is run only when JIT support is available, and the thirteenth +test is run only when JIT support is not available. They test some JIT-specific +features such as information output from pcretest about JIT compilation. + +The fourteenth, fifteenth, and sixteenth tests are run only in 8-bit mode, and +the seventeenth, eighteenth, and nineteenth tests are run only in 16/32-bit +mode. These are tests that generate different output in the two modes. They are +for general cases, UTF-8/16/32 support, and Unicode property support, +respectively. + +The twentieth test is run only in 16/32-bit mode. It tests some specific +16/32-bit features of the DFA matching engine. + +The twenty-first and twenty-second tests are run only in 16/32-bit mode, when +the link size is set to 2 for the 16-bit library. They test reloading +pre-compiled patterns. + +The twenty-third and twenty-fourth tests are run only in 16-bit mode. They are +for general cases, and UTF-16 support, respectively. + +The twenty-fifth and twenty-sixth tests are run only in 32-bit mode. They are +for general cases, and UTF-32 support, respectively. + + +Character tables +---------------- + +For speed, PCRE uses four tables for manipulating and identifying characters +whose code point values are less than 256. The final argument of the +pcre_compile() function is a pointer to a block of memory containing the +concatenated tables. A call to pcre_maketables() can be used to generate a set +of tables in the current locale. If the final argument for pcre_compile() is +passed as NULL, a set of default tables that is built into the binary is used. + +The source file called pcre_chartables.c contains the default set of tables. By +default, this is created as a copy of pcre_chartables.c.dist, which contains +tables for ASCII coding. However, if --enable-rebuild-chartables is specified +for ./configure, a different version of pcre_chartables.c is built by the +program dftables (compiled from dftables.c), which uses the ANSI C character +handling functions such as isalnum(), isalpha(), isupper(), islower(), etc. to +build the table sources. This means that the default C locale which is set for +your system will control the contents of these default tables. You can change +the default tables by editing pcre_chartables.c and then re-building PCRE. If +you do this, you should take care to ensure that the file does not get +automatically re-generated. The best way to do this is to move +pcre_chartables.c.dist out of the way and replace it with your customized +tables. + +When the dftables program is run as a result of --enable-rebuild-chartables, +it uses the default C locale that is set on your system. It does not pay +attention to the LC_xxx environment variables. In other words, it uses the +system's default locale rather than whatever the compiling user happens to have +set. If you really do want to build a source set of character tables in a +locale that is specified by the LC_xxx variables, you can run the dftables +program by hand with the -L option. For example: + + ./dftables -L pcre_chartables.c.special + +The first two 256-byte tables provide lower casing and case flipping functions, +respectively. The next table consists of three 32-byte bit maps which identify +digits, "word" characters, and white space, respectively. These are used when +building 32-byte bit maps that represent character classes for code points less +than 256. + +The final 256-byte table has bits indicating various character types, as +follows: + + 1 white space character + 2 letter + 4 decimal digit + 8 hexadecimal digit + 16 alphanumeric or '_' + 128 regular expression metacharacter or binary zero + +You should not alter the set of characters that contain the 128 bit, as that +will cause PCRE to malfunction. + + +File manifest +------------- + +The distribution should contain the files listed below. Where a file name is +given as pcre[16|32]_xxx it means that there are three files, one with the name +pcre_xxx, one with the name pcre16_xx, and a third with the name pcre32_xxx. + +(A) Source files of the PCRE library functions and their headers: + + dftables.c auxiliary program for building pcre_chartables.c + when --enable-rebuild-chartables is specified + + pcre_chartables.c.dist a default set of character tables that assume ASCII + coding; used, unless --enable-rebuild-chartables is + specified, by copying to pcre[16]_chartables.c + + pcreposix.c ) + pcre[16|32]_byte_order.c ) + pcre[16|32]_compile.c ) + pcre[16|32]_config.c ) + pcre[16|32]_dfa_exec.c ) + pcre[16|32]_exec.c ) + pcre[16|32]_fullinfo.c ) + pcre[16|32]_get.c ) sources for the functions in the library, + pcre[16|32]_globals.c ) and some internal functions that they use + pcre[16|32]_jit_compile.c ) + pcre[16|32]_maketables.c ) + pcre[16|32]_newline.c ) + pcre[16|32]_refcount.c ) + pcre[16|32]_string_utils.c ) + pcre[16|32]_study.c ) + pcre[16|32]_tables.c ) + pcre[16|32]_ucd.c ) + pcre[16|32]_version.c ) + pcre[16|32]_xclass.c ) + pcre_ord2utf8.c ) + pcre_valid_utf8.c ) + pcre16_ord2utf16.c ) + pcre16_utf16_utils.c ) + pcre16_valid_utf16.c ) + pcre32_utf32_utils.c ) + pcre32_valid_utf32.c ) + + pcre[16|32]_printint.c ) debugging function that is used by pcretest, + ) and can also be #included in pcre_compile() + + pcre.h.in template for pcre.h when built by "configure" + pcreposix.h header for the external POSIX wrapper API + pcre_internal.h header for internal use + sljit/* 16 files that make up the JIT compiler + ucp.h header for Unicode property handling + + config.h.in template for config.h, which is built by "configure" + + pcrecpp.h public header file for the C++ wrapper + pcrecpparg.h.in template for another C++ header file + pcre_scanner.h public header file for C++ scanner functions + pcrecpp.cc ) + pcre_scanner.cc ) source for the C++ wrapper library + + pcre_stringpiece.h.in template for pcre_stringpiece.h, the header for the + C++ stringpiece functions + pcre_stringpiece.cc source for the C++ stringpiece functions + +(B) Source files for programs that use PCRE: + + pcredemo.c simple demonstration of coding calls to PCRE + pcregrep.c source of a grep utility that uses PCRE + pcretest.c comprehensive test program + +(C) Auxiliary files: + + 132html script to turn "man" pages into HTML + AUTHORS information about the author of PCRE + ChangeLog log of changes to the code + CleanTxt script to clean nroff output for txt man pages + Detrail script to remove trailing spaces + HACKING some notes about the internals of PCRE + INSTALL generic installation instructions + LICENCE conditions for the use of PCRE + COPYING the same, using GNU's standard name + Makefile.in ) template for Unix Makefile, which is built by + ) "configure" + Makefile.am ) the automake input that was used to create + ) Makefile.in + NEWS important changes in this release + NON-UNIX-USE the previous name for NON-AUTOTOOLS-BUILD + NON-AUTOTOOLS-BUILD notes on building PCRE without using autotools + PrepareRelease script to make preparations for "make dist" + README this file + RunTest a Unix shell script for running tests + RunGrepTest a Unix shell script for pcregrep tests + aclocal.m4 m4 macros (generated by "aclocal") + config.guess ) files used by libtool, + config.sub ) used only when building a shared library + configure a configuring shell script (built by autoconf) + configure.ac ) the autoconf input that was used to build + ) "configure" and config.h + depcomp ) script to find program dependencies, generated by + ) automake + doc/*.3 man page sources for PCRE + doc/*.1 man page sources for pcregrep and pcretest + doc/index.html.src the base HTML page + doc/html/* HTML documentation + doc/pcre.txt plain text version of the man pages + doc/pcretest.txt plain text documentation of test program + doc/perltest.txt plain text documentation of Perl test program + install-sh a shell script for installing files + libpcre16.pc.in template for libpcre16.pc for pkg-config + libpcre32.pc.in template for libpcre32.pc for pkg-config + libpcre.pc.in template for libpcre.pc for pkg-config + libpcreposix.pc.in template for libpcreposix.pc for pkg-config + libpcrecpp.pc.in template for libpcrecpp.pc for pkg-config + ltmain.sh file used to build a libtool script + missing ) common stub for a few missing GNU programs while + ) installing, generated by automake + mkinstalldirs script for making install directories + perltest.pl Perl test program + pcre-config.in source of script which retains PCRE information + pcre_jit_test.c test program for the JIT compiler + pcrecpp_unittest.cc ) + pcre_scanner_unittest.cc ) test programs for the C++ wrapper + pcre_stringpiece_unittest.cc ) + testdata/testinput* test data for main library tests + testdata/testoutput* expected test results + testdata/grep* input and output for pcregrep tests + testdata/* other supporting test files + +(D) Auxiliary files for cmake support + + cmake/COPYING-CMAKE-SCRIPTS + cmake/FindPackageHandleStandardArgs.cmake + cmake/FindEditline.cmake + cmake/FindReadline.cmake + CMakeLists.txt + config-cmake.h.in + +(E) Auxiliary files for VPASCAL + + makevp.bat + makevp_c.txt + makevp_l.txt + pcregexp.pas + +(F) Auxiliary files for building PCRE "by hand" + + pcre.h.generic ) a version of the public PCRE header file + ) for use in non-"configure" environments + config.h.generic ) a version of config.h for use in non-"configure" + ) environments + +(F) Miscellaneous + + RunTest.bat a script for running tests under Windows + +Philip Hazel +Email local part: ph10 +Email domain: cam.ac.uk +Last updated: 28 April 2013 diff --git a/pcre/doc/html/index.html b/pcre/doc/html/index.html new file mode 100644 index 00000000000..352c55df2f1 --- /dev/null +++ b/pcre/doc/html/index.html @@ -0,0 +1,185 @@ + + + +PCRE specification + + +

Perl-compatible Regular Expressions (PCRE)

+

+The HTML documentation for PCRE consists of a number of pages that are listed +below in alphabetical order. If you are new to PCRE, please read the first one +first. +

+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
pcre  Introductory page
pcre-config  Information about the installation configuration
pcre16  Discussion of the 16-bit PCRE library
pcre32  Discussion of the 32-bit PCRE library
pcreapi  PCRE's native API
pcrebuild  Building PCRE
pcrecallout  The callout facility
pcrecompat  Compability with Perl
pcrecpp  The C++ wrapper for the PCRE library
pcredemo  A demonstration C program that uses the PCRE library
pcregrep  The pcregrep command
pcrejit  Discussion of the just-in-time optimization support
pcrelimits  Details of size and other limits
pcrematching  Discussion of the two matching algorithms
pcrepartial  Using PCRE for partial matching
pcrepattern  Specification of the regular expressions supported by PCRE
pcreperform  Some comments on performance
pcreposix  The POSIX API to the PCRE 8-bit library
pcreprecompile  How to save and re-use compiled patterns
pcresample  Discussion of the pcredemo program
pcrestack  Discussion of PCRE's stack usage
pcresyntax  Syntax quick-reference summary
pcretest  The pcretest command for testing PCRE
pcreunicode  Discussion of Unicode and UTF-8/UTF-16/UTF-32 support
+ +

+There are also individual pages that summarize the interface for each function +in the library. There is a single page for each triple of 8-bit/16-bit/32-bit +functions. +

+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
pcre_assign_jit_stack  Assign stack for JIT matching
pcre_compile  Compile a regular expression
pcre_compile2  Compile a regular expression (alternate interface)
pcre_config  Show build-time configuration options
pcre_copy_named_substring  Extract named substring into given buffer
pcre_copy_substring  Extract numbered substring into given buffer
pcre_dfa_exec  Match a compiled pattern to a subject string + (DFA algorithm; not Perl compatible)
pcre_exec  Match a compiled pattern to a subject string + (Perl compatible)
pcre_free_study  Free study data
pcre_free_substring  Free extracted substring
pcre_free_substring_list  Free list of extracted substrings
pcre_fullinfo  Extract information about a pattern
pcre_get_named_substring  Extract named substring into new memory
pcre_get_stringnumber  Convert captured string name to number
pcre_get_stringtable_entries  Find table entries for given string name
pcre_get_substring  Extract numbered substring into new memory
pcre_get_substring_list  Extract all substrings into new memory
pcre_jit_exec  Fast path interface to JIT matching
pcre_jit_stack_alloc  Create a stack for JIT matching
pcre_jit_stack_free  Free a JIT matching stack
pcre_maketables  Build character tables in current locale
pcre_pattern_to_host_byte_order  Convert compiled pattern to host byte order if necessary
pcre_refcount  Maintain reference count in compiled pattern
pcre_study  Study a compiled pattern
pcre_utf16_to_host_byte_order  Convert UTF-16 string to host byte order if necessary
pcre_utf32_to_host_byte_order  Convert UTF-32 string to host byte order if necessary
pcre_version  Return PCRE version and release date
+ + diff --git a/pcre/doc/html/pcre-config.html b/pcre/doc/html/pcre-config.html new file mode 100644 index 00000000000..c2e3ba28ab6 --- /dev/null +++ b/pcre/doc/html/pcre-config.html @@ -0,0 +1,109 @@ + + +pcre-config specification + + +

pcre-config man page

+

+Return to the PCRE index page. +

+

+This page is part of the PCRE HTML documentation. It was generated automatically +from the original man page. If there is any nonsense in it, please consult the +man page, in case the conversion went wrong. +
+

+
SYNOPSIS
+

+pcre-config [--prefix] [--exec-prefix] [--version] [--libs] +[--libs16] [--libs32] [--libs-cpp] [--libs-posix] +[--cflags] [--cflags-posix] +

+
DESCRIPTION
+

+pcre-config returns the configuration of the installed PCRE +libraries and the options required to compile a program to use them. Some of +the options apply only to the 8-bit, or 16-bit, or 32-bit libraries, +respectively, and are +not available if only one of those libraries has been built. If an unavailable +option is encountered, the "usage" information is output. +

+
OPTIONS
+

+--prefix +Writes the directory prefix used in the PCRE installation for architecture +independent files (/usr on many systems, /usr/local on some +systems) to the standard output. +

+

+--exec-prefix +Writes the directory prefix used in the PCRE installation for architecture +dependent files (normally the same as --prefix) to the standard output. +

+

+--version +Writes the version number of the installed PCRE libraries to the standard +output. +

+

+--libs +Writes to the standard output the command line options required to link +with the 8-bit PCRE library (-lpcre on many systems). +

+

+--libs16 +Writes to the standard output the command line options required to link +with the 16-bit PCRE library (-lpcre16 on many systems). +

+

+--libs32 +Writes to the standard output the command line options required to link +with the 32-bit PCRE library (-lpcre32 on many systems). +

+

+--libs-cpp +Writes to the standard output the command line options required to link with +PCRE's C++ wrapper library (-lpcrecpp -lpcre on many +systems). +

+

+--libs-posix +Writes to the standard output the command line options required to link with +PCRE's POSIX API wrapper library (-lpcreposix -lpcre on many +systems). +

+

+--cflags +Writes to the standard output the command line options required to compile +files that use PCRE (this may include some -I options, but is blank on +many systems). +

+

+--cflags-posix +Writes to the standard output the command line options required to compile +files that use PCRE's POSIX API wrapper library (this may include some -I +options, but is blank on many systems). +

+
SEE ALSO
+

+pcre(3) +

+
AUTHOR
+

+This manual page was originally written by Mark Baker for the Debian GNU/Linux +system. It has been subsequently revised as a generic PCRE man page. +

+
REVISION
+

+Last updated: 24 June 2012 +
+

+Return to the PCRE index page. +

diff --git a/pcre/doc/html/pcre.html b/pcre/doc/html/pcre.html new file mode 100644 index 00000000000..692f651b257 --- /dev/null +++ b/pcre/doc/html/pcre.html @@ -0,0 +1,210 @@ + + +pcre specification + + +

pcre man page

+

+Return to the PCRE index page. +

+

+This page is part of the PCRE HTML documentation. It was generated automatically +from the original man page. If there is any nonsense in it, please consult the +man page, in case the conversion went wrong. +
+

+
INTRODUCTION
+

+The PCRE library is a set of functions that implement regular expression +pattern matching using the same syntax and semantics as Perl, with just a few +differences. Some features that appeared in Python and PCRE before they +appeared in Perl are also available using the Python syntax, there is some +support for one or two .NET and Oniguruma syntax items, and there is an option +for requesting some minor changes that give better JavaScript compatibility. +

+

+Starting with release 8.30, it is possible to compile two separate PCRE +libraries: the original, which supports 8-bit character strings (including +UTF-8 strings), and a second library that supports 16-bit character strings +(including UTF-16 strings). The build process allows either one or both to be +built. The majority of the work to make this possible was done by Zoltan +Herczeg. +

+

+Starting with release 8.32 it is possible to compile a third separate PCRE +library that supports 32-bit character strings (including UTF-32 strings). The +build process allows any combination of the 8-, 16- and 32-bit libraries. The +work to make this possible was done by Christian Persch. +

+

+The three libraries contain identical sets of functions, except that the names +in the 16-bit library start with pcre16_ instead of pcre_, and the +names in the 32-bit library start with pcre32_ instead of pcre_. To +avoid over-complication and reduce the documentation maintenance load, most of +the documentation describes the 8-bit library, with the differences for the +16-bit and 32-bit libraries described separately in the +pcre16 +and +pcre32 +pages. References to functions or structures of the form pcre[16|32]_xxx +should be read as meaning "pcre_xxx when using the 8-bit library, +pcre16_xxx when using the 16-bit library, or pcre32_xxx when using +the 32-bit library". +

+

+The current implementation of PCRE corresponds approximately with Perl 5.12, +including support for UTF-8/16/32 encoded strings and Unicode general category +properties. However, UTF-8/16/32 and Unicode support has to be explicitly +enabled; it is not the default. The Unicode tables correspond to Unicode +release 6.2.0. +

+

+In addition to the Perl-compatible matching function, PCRE contains an +alternative function that matches the same compiled patterns in a different +way. In certain circumstances, the alternative function has some advantages. +For a discussion of the two matching algorithms, see the +pcrematching +page. +

+

+PCRE is written in C and released as a C library. A number of people have +written wrappers and interfaces of various kinds. In particular, Google Inc. +have provided a comprehensive C++ wrapper for the 8-bit library. This is now +included as part of the PCRE distribution. The +pcrecpp +page has details of this interface. Other people's contributions can be found +in the Contrib directory at the primary FTP site, which is: +ftp://ftp.csx.cam.ac.uk/pub/software/programming/pcre +

+

+Details of exactly which Perl regular expression features are and are not +supported by PCRE are given in separate documents. See the +pcrepattern +and +pcrecompat +pages. There is a syntax summary in the +pcresyntax +page. +

+

+Some features of PCRE can be included, excluded, or changed when the library is +built. The +pcre_config() +function makes it possible for a client to discover which features are +available. The features themselves are described in the +pcrebuild +page. Documentation about building PCRE for various operating systems can be +found in the +README +and +NON-AUTOTOOLS_BUILD +files in the source distribution. +

+

+The libraries contains a number of undocumented internal functions and data +tables that are used by more than one of the exported external functions, but +which are not intended for use by external callers. Their names all begin with +"_pcre_" or "_pcre16_" or "_pcre32_", which hopefully will not provoke any name +clashes. In some environments, it is possible to control which external symbols +are exported when a shared library is built, and in these cases the +undocumented symbols are not exported. +

+
SECURITY CONSIDERATIONS
+

+If you are using PCRE in a non-UTF application that permits users to supply +arbitrary patterns for compilation, you should be aware of a feature that +allows users to turn on UTF support from within a pattern, provided that PCRE +was built with UTF support. For example, an 8-bit pattern that begins with +"(*UTF8)" or "(*UTF)" turns on UTF-8 mode, which interprets patterns and +subjects as strings of UTF-8 characters instead of individual 8-bit characters. +This causes both the pattern and any data against which it is matched to be +checked for UTF-8 validity. If the data string is very long, such a check might +use sufficiently many resources as to cause your application to lose +performance. +

+

+One way of guarding against this possibility is to use the +pcre_fullinfo() function to check the compiled pattern's options for UTF. +Alternatively, from release 8.33, you can set the PCRE_NEVER_UTF option at +compile time. This causes an compile time error if a pattern contains a +UTF-setting sequence. +

+

+If your application is one that supports UTF, be aware that validity checking +can take time. If the same data string is to be matched many times, you can use +the PCRE_NO_UTF[8|16|32]_CHECK option for the second and subsequent matches to +save redundant checks. +

+

+Another way that performance can be hit is by running a pattern that has a very +large search tree against a string that will never match. Nested unlimited +repeats in a pattern are a common example. PCRE provides some protection +against this: see the PCRE_EXTRA_MATCH_LIMIT feature in the +pcreapi +page. +

+
USER DOCUMENTATION
+

+The user documentation for PCRE comprises a number of different sections. In +the "man" format, each of these is a separate "man page". In the HTML format, +each is a separate page, linked from the index page. In the plain text format, +all the sections, except the pcredemo section, are concatenated, for ease +of searching. The sections are as follows: +

+  pcre              this document
+  pcre-config       show PCRE installation configuration information
+  pcre16            details of the 16-bit library
+  pcre32            details of the 32-bit library
+  pcreapi           details of PCRE's native C API
+  pcrebuild         building PCRE
+  pcrecallout       details of the callout feature
+  pcrecompat        discussion of Perl compatibility
+  pcrecpp           details of the C++ wrapper for the 8-bit library
+  pcredemo          a demonstration C program that uses PCRE
+  pcregrep          description of the pcregrep command (8-bit only)
+  pcrejit           discussion of the just-in-time optimization support
+  pcrelimits        details of size and other limits
+  pcrematching      discussion of the two matching algorithms
+  pcrepartial       details of the partial matching facility
+  pcrepattern       syntax and semantics of supported regular expressions
+  pcreperform       discussion of performance issues
+  pcreposix         the POSIX-compatible C API for the 8-bit library
+  pcreprecompile    details of saving and re-using precompiled patterns
+  pcresample        discussion of the pcredemo program
+  pcrestack         discussion of stack usage
+  pcresyntax        quick syntax reference
+  pcretest          description of the pcretest testing command
+  pcreunicode       discussion of Unicode and UTF-8/16/32 support
+
+In addition, in the "man" and HTML formats, there is a short page for each +C library function, listing its arguments and results. +

+
AUTHOR
+

+Philip Hazel +
+University Computing Service +
+Cambridge CB2 3QH, England. +
+

+

+Putting an actual email address here seems to have been a spam magnet, so I've +taken it away. If you want to email me, use my two initials, followed by the +two digits 10, at the domain cam.ac.uk. +

+
REVISION
+

+Last updated: 13 May 2013 +
+Copyright © 1997-2013 University of Cambridge. +
+

+Return to the PCRE index page. +

diff --git a/pcre/doc/html/pcre16.html b/pcre/doc/html/pcre16.html new file mode 100644 index 00000000000..3ade219092c --- /dev/null +++ b/pcre/doc/html/pcre16.html @@ -0,0 +1,384 @@ + + +pcre16 specification + + +

pcre16 man page

+

+Return to the PCRE index page. +

+

+This page is part of the PCRE HTML documentation. It was generated automatically +from the original man page. If there is any nonsense in it, please consult the +man page, in case the conversion went wrong. +
+

+

+#include <pcre.h> +

+
PCRE 16-BIT API BASIC FUNCTIONS
+

+pcre16 *pcre16_compile(PCRE_SPTR16 pattern, int options, +const char **errptr, int *erroffset, +const unsigned char *tableptr); +

+

+pcre16 *pcre16_compile2(PCRE_SPTR16 pattern, int options, +int *errorcodeptr, +const char **errptr, int *erroffset, +const unsigned char *tableptr); +

+

+pcre16_extra *pcre16_study(const pcre16 *code, int options, +const char **errptr); +

+

+void pcre16_free_study(pcre16_extra *extra); +

+

+int pcre16_exec(const pcre16 *code, const pcre16_extra *extra, +PCRE_SPTR16 subject, int length, int startoffset, +int options, int *ovector, int ovecsize); +

+

+int pcre16_dfa_exec(const pcre16 *code, const pcre16_extra *extra, +PCRE_SPTR16 subject, int length, int startoffset, +int options, int *ovector, int ovecsize, +int *workspace, int wscount); +

+
PCRE 16-BIT API STRING EXTRACTION FUNCTIONS
+

+int pcre16_copy_named_substring(const pcre16 *code, +PCRE_SPTR16 subject, int *ovector, +int stringcount, PCRE_SPTR16 stringname, +PCRE_UCHAR16 *buffer, int buffersize); +

+

+int pcre16_copy_substring(PCRE_SPTR16 subject, int *ovector, +int stringcount, int stringnumber, PCRE_UCHAR16 *buffer, +int buffersize); +

+

+int pcre16_get_named_substring(const pcre16 *code, +PCRE_SPTR16 subject, int *ovector, +int stringcount, PCRE_SPTR16 stringname, +PCRE_SPTR16 *stringptr); +

+

+int pcre16_get_stringnumber(const pcre16 *code, +PCRE_SPTR16 name); +

+

+int pcre16_get_stringtable_entries(const pcre16 *code, +PCRE_SPTR16 name, PCRE_UCHAR16 **first, PCRE_UCHAR16 **last); +

+

+int pcre16_get_substring(PCRE_SPTR16 subject, int *ovector, +int stringcount, int stringnumber, +PCRE_SPTR16 *stringptr); +

+

+int pcre16_get_substring_list(PCRE_SPTR16 subject, +int *ovector, int stringcount, PCRE_SPTR16 **listptr); +

+

+void pcre16_free_substring(PCRE_SPTR16 stringptr); +

+

+void pcre16_free_substring_list(PCRE_SPTR16 *stringptr); +

+
PCRE 16-BIT API AUXILIARY FUNCTIONS
+

+pcre16_jit_stack *pcre16_jit_stack_alloc(int startsize, int maxsize); +

+

+void pcre16_jit_stack_free(pcre16_jit_stack *stack); +

+

+void pcre16_assign_jit_stack(pcre16_extra *extra, +pcre16_jit_callback callback, void *data); +

+

+const unsigned char *pcre16_maketables(void); +

+

+int pcre16_fullinfo(const pcre16 *code, const pcre16_extra *extra, +int what, void *where); +

+

+int pcre16_refcount(pcre16 *code, int adjust); +

+

+int pcre16_config(int what, void *where); +

+

+const char *pcre16_version(void); +

+

+int pcre16_pattern_to_host_byte_order(pcre16 *code, +pcre16_extra *extra, const unsigned char *tables); +

+
PCRE 16-BIT API INDIRECTED FUNCTIONS
+

+void *(*pcre16_malloc)(size_t); +

+

+void (*pcre16_free)(void *); +

+

+void *(*pcre16_stack_malloc)(size_t); +

+

+void (*pcre16_stack_free)(void *); +

+

+int (*pcre16_callout)(pcre16_callout_block *); +

+
PCRE 16-BIT API 16-BIT-ONLY FUNCTION
+

+int pcre16_utf16_to_host_byte_order(PCRE_UCHAR16 *output, +PCRE_SPTR16 input, int length, int *byte_order, +int keep_boms); +

+
THE PCRE 16-BIT LIBRARY
+

+Starting with release 8.30, it is possible to compile a PCRE library that +supports 16-bit character strings, including UTF-16 strings, as well as or +instead of the original 8-bit library. The majority of the work to make this +possible was done by Zoltan Herczeg. The two libraries contain identical sets +of functions, used in exactly the same way. Only the names of the functions and +the data types of their arguments and results are different. To avoid +over-complication and reduce the documentation maintenance load, most of the +PCRE documentation describes the 8-bit library, with only occasional references +to the 16-bit library. This page describes what is different when you use the +16-bit library. +

+

+WARNING: A single application can be linked with both libraries, but you must +take care when processing any particular pattern to use functions from just one +library. For example, if you want to study a pattern that was compiled with +pcre16_compile(), you must do so with pcre16_study(), not +pcre_study(), and you must free the study data with +pcre16_free_study(). +

+
THE HEADER FILE
+

+There is only one header file, pcre.h. It contains prototypes for all the +functions in all libraries, as well as definitions of flags, structures, error +codes, etc. +

+
THE LIBRARY NAME
+

+In Unix-like systems, the 16-bit library is called libpcre16, and can +normally be accesss by adding -lpcre16 to the command for linking an +application that uses PCRE. +

+
STRING TYPES
+

+In the 8-bit library, strings are passed to PCRE library functions as vectors +of bytes with the C type "char *". In the 16-bit library, strings are passed as +vectors of unsigned 16-bit quantities. The macro PCRE_UCHAR16 specifies an +appropriate data type, and PCRE_SPTR16 is defined as "const PCRE_UCHAR16 *". In +very many environments, "short int" is a 16-bit data type. When PCRE is built, +it defines PCRE_UCHAR16 as "unsigned short int", but checks that it really is a +16-bit data type. If it is not, the build fails with an error message telling +the maintainer to modify the definition appropriately. +

+
STRUCTURE TYPES
+

+The types of the opaque structures that are used for compiled 16-bit patterns +and JIT stacks are pcre16 and pcre16_jit_stack respectively. The +type of the user-accessible structure that is returned by pcre16_study() +is pcre16_extra, and the type of the structure that is used for passing +data to a callout function is pcre16_callout_block. These structures +contain the same fields, with the same names, as their 8-bit counterparts. The +only difference is that pointers to character strings are 16-bit instead of +8-bit types. +

+
16-BIT FUNCTIONS
+

+For every function in the 8-bit library there is a corresponding function in +the 16-bit library with a name that starts with pcre16_ instead of +pcre_. The prototypes are listed above. In addition, there is one extra +function, pcre16_utf16_to_host_byte_order(). This is a utility function +that converts a UTF-16 character string to host byte order if necessary. The +other 16-bit functions expect the strings they are passed to be in host byte +order. +

+

+The input and output arguments of +pcre16_utf16_to_host_byte_order() may point to the same address, that is, +conversion in place is supported. The output buffer must be at least as long as +the input. +

+

+The length argument specifies the number of 16-bit data units in the +input string; a negative value specifies a zero-terminated string. +

+

+If byte_order is NULL, it is assumed that the string starts off in host +byte order. This may be changed by byte-order marks (BOMs) anywhere in the +string (commonly as the first character). +

+

+If byte_order is not NULL, a non-zero value of the integer to which it +points means that the input starts off in host byte order, otherwise the +opposite order is assumed. Again, BOMs in the string can change this. The final +byte order is passed back at the end of processing. +

+

+If keep_boms is not zero, byte-order mark characters (0xfeff) are copied +into the output string. Otherwise they are discarded. +

+

+The result of the function is the number of 16-bit units placed into the output +buffer, including the zero terminator if the string was zero-terminated. +

+
SUBJECT STRING OFFSETS
+

+The lengths and starting offsets of subject strings must be specified in 16-bit +data units, and the offsets within subject strings that are returned by the +matching functions are in also 16-bit units rather than bytes. +

+
NAMED SUBPATTERNS
+

+The name-to-number translation table that is maintained for named subpatterns +uses 16-bit characters. The pcre16_get_stringtable_entries() function +returns the length of each entry in the table as the number of 16-bit data +units. +

+
OPTION NAMES
+

+There are two new general option names, PCRE_UTF16 and PCRE_NO_UTF16_CHECK, +which correspond to PCRE_UTF8 and PCRE_NO_UTF8_CHECK in the 8-bit library. In +fact, these new options define the same bits in the options word. There is a +discussion about the +validity of UTF-16 strings +in the +pcreunicode +page. +

+

+For the pcre16_config() function there is an option PCRE_CONFIG_UTF16 +that returns 1 if UTF-16 support is configured, otherwise 0. If this option is +given to pcre_config() or pcre32_config(), or if the +PCRE_CONFIG_UTF8 or PCRE_CONFIG_UTF32 option is given to pcre16_config(), +the result is the PCRE_ERROR_BADOPTION error. +

+
CHARACTER CODES
+

+In 16-bit mode, when PCRE_UTF16 is not set, character values are treated in the +same way as in 8-bit, non UTF-8 mode, except, of course, that they can range +from 0 to 0xffff instead of 0 to 0xff. Character types for characters less than +0xff can therefore be influenced by the locale in the same way as before. +Characters greater than 0xff have only one case, and no "type" (such as letter +or digit). +

+

+In UTF-16 mode, the character code is Unicode, in the range 0 to 0x10ffff, with +the exception of values in the range 0xd800 to 0xdfff because those are +"surrogate" values that are used in pairs to encode values greater than 0xffff. +

+

+A UTF-16 string can indicate its endianness by special code knows as a +byte-order mark (BOM). The PCRE functions do not handle this, expecting strings +to be in host byte order. A utility function called +pcre16_utf16_to_host_byte_order() is provided to help with this (see +above). +

+
ERROR NAMES
+

+The errors PCRE_ERROR_BADUTF16_OFFSET and PCRE_ERROR_SHORTUTF16 correspond to +their 8-bit counterparts. The error PCRE_ERROR_BADMODE is given when a compiled +pattern is passed to a function that processes patterns in the other +mode, for example, if a pattern compiled with pcre_compile() is passed to +pcre16_exec(). +

+

+There are new error codes whose names begin with PCRE_UTF16_ERR for invalid +UTF-16 strings, corresponding to the PCRE_UTF8_ERR codes for UTF-8 strings that +are described in the section entitled +"Reason codes for invalid UTF-8 strings" +in the main +pcreapi +page. The UTF-16 errors are: +

+  PCRE_UTF16_ERR1  Missing low surrogate at end of string
+  PCRE_UTF16_ERR2  Invalid low surrogate follows high surrogate
+  PCRE_UTF16_ERR3  Isolated low surrogate
+  PCRE_UTF16_ERR4  Non-character
+
+

+
ERROR TEXTS
+

+If there is an error while compiling a pattern, the error text that is passed +back by pcre16_compile() or pcre16_compile2() is still an 8-bit +character string, zero-terminated. +

+
CALLOUTS
+

+The subject and mark fields in the callout block that is passed to +a callout function point to 16-bit vectors. +

+
TESTING
+

+The pcretest program continues to operate with 8-bit input and output +files, but it can be used for testing the 16-bit library. If it is run with the +command line option -16, patterns and subject strings are converted from +8-bit to 16-bit before being passed to PCRE, and the 16-bit library functions +are used instead of the 8-bit ones. Returned 16-bit strings are converted to +8-bit for output. If both the 8-bit and the 32-bit libraries were not compiled, +pcretest defaults to 16-bit and the -16 option is ignored. +

+

+When PCRE is being built, the RunTest script that is called by "make +check" uses the pcretest -C option to discover which of the 8-bit, +16-bit and 32-bit libraries has been built, and runs the tests appropriately. +

+
NOT SUPPORTED IN 16-BIT MODE
+

+Not all the features of the 8-bit library are available with the 16-bit +library. The C++ and POSIX wrapper functions support only the 8-bit library, +and the pcregrep program is at present 8-bit only. +

+
AUTHOR
+

+Philip Hazel +
+University Computing Service +
+Cambridge CB2 3QH, England. +
+

+
REVISION
+

+Last updated: 12 May 2013 +
+Copyright © 1997-2013 University of Cambridge. +
+

+Return to the PCRE index page. +

diff --git a/pcre/doc/html/pcre32.html b/pcre/doc/html/pcre32.html new file mode 100644 index 00000000000..2155ee88173 --- /dev/null +++ b/pcre/doc/html/pcre32.html @@ -0,0 +1,383 @@ + + +pcre32 specification + + +

pcre32 man page

+

+Return to the PCRE index page. +

+

+This page is part of the PCRE HTML documentation. It was generated automatically +from the original man page. If there is any nonsense in it, please consult the +man page, in case the conversion went wrong. +
+

+

+#include <pcre.h> +

+
PCRE 32-BIT API BASIC FUNCTIONS
+

+pcre32 *pcre32_compile(PCRE_SPTR32 pattern, int options, +const char **errptr, int *erroffset, +const unsigned char *tableptr); +

+

+pcre32 *pcre32_compile2(PCRE_SPTR32 pattern, int options, +int *errorcodeptr, +const char **errptr, int *erroffset, +const unsigned char *tableptr); +

+

+pcre32_extra *pcre32_study(const pcre32 *code, int options, +const char **errptr); +

+

+void pcre32_free_study(pcre32_extra *extra); +

+

+int pcre32_exec(const pcre32 *code, const pcre32_extra *extra, +PCRE_SPTR32 subject, int length, int startoffset, +int options, int *ovector, int ovecsize); +

+

+int pcre32_dfa_exec(const pcre32 *code, const pcre32_extra *extra, +PCRE_SPTR32 subject, int length, int startoffset, +int options, int *ovector, int ovecsize, +int *workspace, int wscount); +

+
PCRE 32-BIT API STRING EXTRACTION FUNCTIONS
+

+int pcre32_copy_named_substring(const pcre32 *code, +PCRE_SPTR32 subject, int *ovector, +int stringcount, PCRE_SPTR32 stringname, +PCRE_UCHAR32 *buffer, int buffersize); +

+

+int pcre32_copy_substring(PCRE_SPTR32 subject, int *ovector, +int stringcount, int stringnumber, PCRE_UCHAR32 *buffer, +int buffersize); +

+

+int pcre32_get_named_substring(const pcre32 *code, +PCRE_SPTR32 subject, int *ovector, +int stringcount, PCRE_SPTR32 stringname, +PCRE_SPTR32 *stringptr); +

+

+int pcre32_get_stringnumber(const pcre32 *code, +PCRE_SPTR32 name); +

+

+int pcre32_get_stringtable_entries(const pcre32 *code, +PCRE_SPTR32 name, PCRE_UCHAR32 **first, PCRE_UCHAR32 **last); +

+

+int pcre32_get_substring(PCRE_SPTR32 subject, int *ovector, +int stringcount, int stringnumber, +PCRE_SPTR32 *stringptr); +

+

+int pcre32_get_substring_list(PCRE_SPTR32 subject, +int *ovector, int stringcount, PCRE_SPTR32 **listptr); +

+

+void pcre32_free_substring(PCRE_SPTR32 stringptr); +

+

+void pcre32_free_substring_list(PCRE_SPTR32 *stringptr); +

+
PCRE 32-BIT API AUXILIARY FUNCTIONS
+

+pcre32_jit_stack *pcre32_jit_stack_alloc(int startsize, int maxsize); +

+

+void pcre32_jit_stack_free(pcre32_jit_stack *stack); +

+

+void pcre32_assign_jit_stack(pcre32_extra *extra, +pcre32_jit_callback callback, void *data); +

+

+const unsigned char *pcre32_maketables(void); +

+

+int pcre32_fullinfo(const pcre32 *code, const pcre32_extra *extra, +int what, void *where); +

+

+int pcre32_refcount(pcre32 *code, int adjust); +

+

+int pcre32_config(int what, void *where); +

+

+const char *pcre32_version(void); +

+

+int pcre32_pattern_to_host_byte_order(pcre32 *code, +pcre32_extra *extra, const unsigned char *tables); +

+
PCRE 32-BIT API INDIRECTED FUNCTIONS
+

+void *(*pcre32_malloc)(size_t); +

+

+void (*pcre32_free)(void *); +

+

+void *(*pcre32_stack_malloc)(size_t); +

+

+void (*pcre32_stack_free)(void *); +

+

+int (*pcre32_callout)(pcre32_callout_block *); +

+
PCRE 32-BIT API 32-BIT-ONLY FUNCTION
+

+int pcre32_utf32_to_host_byte_order(PCRE_UCHAR32 *output, +PCRE_SPTR32 input, int length, int *byte_order, +int keep_boms); +

+
THE PCRE 32-BIT LIBRARY
+

+Starting with release 8.32, it is possible to compile a PCRE library that +supports 32-bit character strings, including UTF-32 strings, as well as or +instead of the original 8-bit library. This work was done by Christian Persch, +based on the work done by Zoltan Herczeg for the 16-bit library. All three +libraries contain identical sets of functions, used in exactly the same way. +Only the names of the functions and the data types of their arguments and +results are different. To avoid over-complication and reduce the documentation +maintenance load, most of the PCRE documentation describes the 8-bit library, +with only occasional references to the 16-bit and 32-bit libraries. This page +describes what is different when you use the 32-bit library. +

+

+WARNING: A single application can be linked with all or any of the three +libraries, but you must take care when processing any particular pattern +to use functions from just one library. For example, if you want to study +a pattern that was compiled with pcre32_compile(), you must do so +with pcre32_study(), not pcre_study(), and you must free the +study data with pcre32_free_study(). +

+
THE HEADER FILE
+

+There is only one header file, pcre.h. It contains prototypes for all the +functions in all libraries, as well as definitions of flags, structures, error +codes, etc. +

+
THE LIBRARY NAME
+

+In Unix-like systems, the 32-bit library is called libpcre32, and can +normally be accesss by adding -lpcre32 to the command for linking an +application that uses PCRE. +

+
STRING TYPES
+

+In the 8-bit library, strings are passed to PCRE library functions as vectors +of bytes with the C type "char *". In the 32-bit library, strings are passed as +vectors of unsigned 32-bit quantities. The macro PCRE_UCHAR32 specifies an +appropriate data type, and PCRE_SPTR32 is defined as "const PCRE_UCHAR32 *". In +very many environments, "unsigned int" is a 32-bit data type. When PCRE is +built, it defines PCRE_UCHAR32 as "unsigned int", but checks that it really is +a 32-bit data type. If it is not, the build fails with an error message telling +the maintainer to modify the definition appropriately. +

+
STRUCTURE TYPES
+

+The types of the opaque structures that are used for compiled 32-bit patterns +and JIT stacks are pcre32 and pcre32_jit_stack respectively. The +type of the user-accessible structure that is returned by pcre32_study() +is pcre32_extra, and the type of the structure that is used for passing +data to a callout function is pcre32_callout_block. These structures +contain the same fields, with the same names, as their 8-bit counterparts. The +only difference is that pointers to character strings are 32-bit instead of +8-bit types. +

+
32-BIT FUNCTIONS
+

+For every function in the 8-bit library there is a corresponding function in +the 32-bit library with a name that starts with pcre32_ instead of +pcre_. The prototypes are listed above. In addition, there is one extra +function, pcre32_utf32_to_host_byte_order(). This is a utility function +that converts a UTF-32 character string to host byte order if necessary. The +other 32-bit functions expect the strings they are passed to be in host byte +order. +

+

+The input and output arguments of +pcre32_utf32_to_host_byte_order() may point to the same address, that is, +conversion in place is supported. The output buffer must be at least as long as +the input. +

+

+The length argument specifies the number of 32-bit data units in the +input string; a negative value specifies a zero-terminated string. +

+

+If byte_order is NULL, it is assumed that the string starts off in host +byte order. This may be changed by byte-order marks (BOMs) anywhere in the +string (commonly as the first character). +

+

+If byte_order is not NULL, a non-zero value of the integer to which it +points means that the input starts off in host byte order, otherwise the +opposite order is assumed. Again, BOMs in the string can change this. The final +byte order is passed back at the end of processing. +

+

+If keep_boms is not zero, byte-order mark characters (0xfeff) are copied +into the output string. Otherwise they are discarded. +

+

+The result of the function is the number of 32-bit units placed into the output +buffer, including the zero terminator if the string was zero-terminated. +

+
SUBJECT STRING OFFSETS
+

+The lengths and starting offsets of subject strings must be specified in 32-bit +data units, and the offsets within subject strings that are returned by the +matching functions are in also 32-bit units rather than bytes. +

+
NAMED SUBPATTERNS
+

+The name-to-number translation table that is maintained for named subpatterns +uses 32-bit characters. The pcre32_get_stringtable_entries() function +returns the length of each entry in the table as the number of 32-bit data +units. +

+
OPTION NAMES
+

+There are two new general option names, PCRE_UTF32 and PCRE_NO_UTF32_CHECK, +which correspond to PCRE_UTF8 and PCRE_NO_UTF8_CHECK in the 8-bit library. In +fact, these new options define the same bits in the options word. There is a +discussion about the +validity of UTF-32 strings +in the +pcreunicode +page. +

+

+For the pcre32_config() function there is an option PCRE_CONFIG_UTF32 +that returns 1 if UTF-32 support is configured, otherwise 0. If this option is +given to pcre_config() or pcre16_config(), or if the +PCRE_CONFIG_UTF8 or PCRE_CONFIG_UTF16 option is given to pcre32_config(), +the result is the PCRE_ERROR_BADOPTION error. +

+
CHARACTER CODES
+

+In 32-bit mode, when PCRE_UTF32 is not set, character values are treated in the +same way as in 8-bit, non UTF-8 mode, except, of course, that they can range +from 0 to 0x7fffffff instead of 0 to 0xff. Character types for characters less +than 0xff can therefore be influenced by the locale in the same way as before. +Characters greater than 0xff have only one case, and no "type" (such as letter +or digit). +

+

+In UTF-32 mode, the character code is Unicode, in the range 0 to 0x10ffff, with +the exception of values in the range 0xd800 to 0xdfff because those are +"surrogate" values that are ill-formed in UTF-32. +

+

+A UTF-32 string can indicate its endianness by special code knows as a +byte-order mark (BOM). The PCRE functions do not handle this, expecting strings +to be in host byte order. A utility function called +pcre32_utf32_to_host_byte_order() is provided to help with this (see +above). +

+
ERROR NAMES
+

+The error PCRE_ERROR_BADUTF32 corresponds to its 8-bit counterpart. +The error PCRE_ERROR_BADMODE is given when a compiled +pattern is passed to a function that processes patterns in the other +mode, for example, if a pattern compiled with pcre_compile() is passed to +pcre32_exec(). +

+

+There are new error codes whose names begin with PCRE_UTF32_ERR for invalid +UTF-32 strings, corresponding to the PCRE_UTF8_ERR codes for UTF-8 strings that +are described in the section entitled +"Reason codes for invalid UTF-8 strings" +in the main +pcreapi +page. The UTF-32 errors are: +

+  PCRE_UTF32_ERR1  Surrogate character (range from 0xd800 to 0xdfff)
+  PCRE_UTF32_ERR2  Non-character
+  PCRE_UTF32_ERR3  Character > 0x10ffff
+
+

+
ERROR TEXTS
+

+If there is an error while compiling a pattern, the error text that is passed +back by pcre32_compile() or pcre32_compile2() is still an 8-bit +character string, zero-terminated. +

+
CALLOUTS
+

+The subject and mark fields in the callout block that is passed to +a callout function point to 32-bit vectors. +

+
TESTING
+

+The pcretest program continues to operate with 8-bit input and output +files, but it can be used for testing the 32-bit library. If it is run with the +command line option -32, patterns and subject strings are converted from +8-bit to 32-bit before being passed to PCRE, and the 32-bit library functions +are used instead of the 8-bit ones. Returned 32-bit strings are converted to +8-bit for output. If both the 8-bit and the 16-bit libraries were not compiled, +pcretest defaults to 32-bit and the -32 option is ignored. +

+

+When PCRE is being built, the RunTest script that is called by "make +check" uses the pcretest -C option to discover which of the 8-bit, +16-bit and 32-bit libraries has been built, and runs the tests appropriately. +

+
NOT SUPPORTED IN 32-BIT MODE
+

+Not all the features of the 8-bit library are available with the 32-bit +library. The C++ and POSIX wrapper functions support only the 8-bit library, +and the pcregrep program is at present 8-bit only. +

+
AUTHOR
+

+Philip Hazel +
+University Computing Service +
+Cambridge CB2 3QH, England. +
+

+
REVISION
+

+Last updated: 12 May 2013 +
+Copyright © 1997-2013 University of Cambridge. +
+

+Return to the PCRE index page. +

diff --git a/pcre/doc/html/pcre_assign_jit_stack.html b/pcre/doc/html/pcre_assign_jit_stack.html new file mode 100644 index 00000000000..d77d4e1c9b4 --- /dev/null +++ b/pcre/doc/html/pcre_assign_jit_stack.html @@ -0,0 +1,76 @@ + + +pcre_assign_jit_stack specification + + +

pcre_assign_jit_stack man page

+

+Return to the PCRE index page. +

+

+This page is part of the PCRE HTML documentation. It was generated automatically +from the original man page. If there is any nonsense in it, please consult the +man page, in case the conversion went wrong. +
+
+SYNOPSIS +
+

+#include <pcre.h> +

+

+void pcre_assign_jit_stack(pcre_extra *extra, +pcre_jit_callback callback, void *data); +

+

+void pcre16_assign_jit_stack(pcre16_extra *extra, +pcre16_jit_callback callback, void *data); +

+

+void pcre32_assign_jit_stack(pcre32_extra *extra, +pcre32_jit_callback callback, void *data); +

+
+DESCRIPTION +
+

+This function provides control over the memory used as a stack at run-time by a +call to pcre[16|32]_exec() with a pattern that has been successfully +compiled with JIT optimization. The arguments are: +

+  extra     the data pointer returned by pcre[16|32]_study()
+  callback  a callback function
+  data      a JIT stack or a value to be passed to the callback
+              function
+
+

+

+If callback is NULL and data is NULL, an internal 32K block on +the machine stack is used. +

+

+If callback is NULL and data is not NULL, data must +be a valid JIT stack, the result of calling pcre[16|32]_jit_stack_alloc(). +

+

+If callback not NULL, it is called with data as an argument at +the start of matching, in order to set up a JIT stack. If the result is NULL, +the internal 32K stack is used; otherwise the return value must be a valid JIT +stack, the result of calling pcre[16|32]_jit_stack_alloc(). +

+

+You may safely assign the same JIT stack to multiple patterns, as long as they +are all matched in the same thread. In a multithread application, each thread +must use its own JIT stack. For more details, see the +pcrejit +page. +

+

+There is a complete description of the PCRE native API in the +pcreapi +page and a description of the POSIX API in the +pcreposix +page. +

+Return to the PCRE index page. +

diff --git a/pcre/doc/html/pcre_compile.html b/pcre/doc/html/pcre_compile.html new file mode 100644 index 00000000000..012147516cf --- /dev/null +++ b/pcre/doc/html/pcre_compile.html @@ -0,0 +1,108 @@ + + +pcre_compile specification + + +

pcre_compile man page

+

+Return to the PCRE index page. +

+

+This page is part of the PCRE HTML documentation. It was generated automatically +from the original man page. If there is any nonsense in it, please consult the +man page, in case the conversion went wrong. +
+
+SYNOPSIS +
+

+#include <pcre.h> +

+

+pcre *pcre_compile(const char *pattern, int options, +const char **errptr, int *erroffset, +const unsigned char *tableptr); +

+

+pcre16 *pcre16_compile(PCRE_SPTR16 pattern, int options, +const char **errptr, int *erroffset, +const unsigned char *tableptr); +

+

+pcre32 *pcre32_compile(PCRE_SPTR32 pattern, int options, +const char **errptr, int *erroffset, +const unsigned char *tableptr); +

+
+DESCRIPTION +
+

+This function compiles a regular expression into an internal form. It is the +same as pcre[16|32]_compile2(), except for the absence of the +errorcodeptr argument. Its arguments are: +

+  pattern       A zero-terminated string containing the
+                  regular expression to be compiled
+  options       Zero or more option bits
+  errptr        Where to put an error message
+  erroffset     Offset in pattern where error was found
+  tableptr      Pointer to character tables, or NULL to
+                  use the built-in default
+
+The option bits are: +
+  PCRE_ANCHORED           Force pattern anchoring
+  PCRE_AUTO_CALLOUT       Compile automatic callouts
+  PCRE_BSR_ANYCRLF        \R matches only CR, LF, or CRLF
+  PCRE_BSR_UNICODE        \R matches all Unicode line endings
+  PCRE_CASELESS           Do caseless matching
+  PCRE_DOLLAR_ENDONLY     $ not to match newline at end
+  PCRE_DOTALL             . matches anything including NL
+  PCRE_DUPNAMES           Allow duplicate names for subpatterns
+  PCRE_EXTENDED           Ignore white space and # comments
+  PCRE_EXTRA              PCRE extra features
+                            (not much use currently)
+  PCRE_FIRSTLINE          Force matching to be before newline
+  PCRE_JAVASCRIPT_COMPAT  JavaScript compatibility
+  PCRE_MULTILINE          ^ and $ match newlines within data
+  PCRE_NEWLINE_ANY        Recognize any Unicode newline sequence
+  PCRE_NEWLINE_ANYCRLF    Recognize CR, LF, and CRLF as newline
+                            sequences
+  PCRE_NEWLINE_CR         Set CR as the newline sequence
+  PCRE_NEWLINE_CRLF       Set CRLF as the newline sequence
+  PCRE_NEWLINE_LF         Set LF as the newline sequence
+  PCRE_NO_AUTO_CAPTURE    Disable numbered capturing paren-
+                            theses (named ones available)
+  PCRE_NO_UTF16_CHECK     Do not check the pattern for UTF-16
+                            validity (only relevant if
+                            PCRE_UTF16 is set)
+  PCRE_NO_UTF32_CHECK     Do not check the pattern for UTF-32
+                            validity (only relevant if
+                            PCRE_UTF32 is set)
+  PCRE_NO_UTF8_CHECK      Do not check the pattern for UTF-8
+                            validity (only relevant if
+                            PCRE_UTF8 is set)
+  PCRE_UCP                Use Unicode properties for \d, \w, etc.
+  PCRE_UNGREEDY           Invert greediness of quantifiers
+  PCRE_UTF16              Run in pcre16_compile() UTF-16 mode
+  PCRE_UTF32              Run in pcre32_compile() UTF-32 mode
+  PCRE_UTF8               Run in pcre_compile() UTF-8 mode
+
+PCRE must be built with UTF support in order to use PCRE_UTF8/16/32 and +PCRE_NO_UTF8/16/32_CHECK, and with UCP support if PCRE_UCP is used. +

+

+The yield of the function is a pointer to a private data structure that +contains the compiled pattern, or NULL if an error was detected. Note that +compiling regular expressions with one version of PCRE for use with a different +version is not guaranteed to work and may cause crashes. +

+

+There is a complete description of the PCRE native API in the +pcreapi +page and a description of the POSIX API in the +pcreposix +page. +

+Return to the PCRE index page. +

diff --git a/pcre/doc/html/pcre_compile2.html b/pcre/doc/html/pcre_compile2.html new file mode 100644 index 00000000000..7d76bd9cae0 --- /dev/null +++ b/pcre/doc/html/pcre_compile2.html @@ -0,0 +1,112 @@ + + +pcre_compile2 specification + + +

pcre_compile2 man page

+

+Return to the PCRE index page. +

+

+This page is part of the PCRE HTML documentation. It was generated automatically +from the original man page. If there is any nonsense in it, please consult the +man page, in case the conversion went wrong. +
+
+SYNOPSIS +
+

+#include <pcre.h> +

+

+pcre *pcre_compile2(const char *pattern, int options, +int *errorcodeptr, +const char **errptr, int *erroffset, +const unsigned char *tableptr); +

+

+pcre16 *pcre16_compile2(PCRE_SPTR16 pattern, int options, +int *errorcodeptr, +const char **errptr, int *erroffset, +const unsigned char *tableptr); +

+

+pcre32 *pcre32_compile2(PCRE_SPTR32 pattern, int options, +int *errorcodeptr, +const char **errptr, int *erroffset, +const unsigned char *tableptr); +

+
+DESCRIPTION +
+

+This function compiles a regular expression into an internal form. It is the +same as pcre[16|32]_compile(), except for the addition of the +errorcodeptr argument. The arguments are: +

+  pattern       A zero-terminated string containing the
+                  regular expression to be compiled
+  options       Zero or more option bits
+  errorcodeptr  Where to put an error code
+  errptr        Where to put an error message
+  erroffset     Offset in pattern where error was found
+  tableptr      Pointer to character tables, or NULL to
+                  use the built-in default
+
+The option bits are: +
+  PCRE_ANCHORED           Force pattern anchoring
+  PCRE_AUTO_CALLOUT       Compile automatic callouts
+  PCRE_BSR_ANYCRLF        \R matches only CR, LF, or CRLF
+  PCRE_BSR_UNICODE        \R matches all Unicode line endings
+  PCRE_CASELESS           Do caseless matching
+  PCRE_DOLLAR_ENDONLY     $ not to match newline at end
+  PCRE_DOTALL             . matches anything including NL
+  PCRE_DUPNAMES           Allow duplicate names for subpatterns
+  PCRE_EXTENDED           Ignore white space and # comments
+  PCRE_EXTRA              PCRE extra features
+                            (not much use currently)
+  PCRE_FIRSTLINE          Force matching to be before newline
+  PCRE_JAVASCRIPT_COMPAT  JavaScript compatibility
+  PCRE_MULTILINE          ^ and $ match newlines within data
+  PCRE_NEWLINE_ANY        Recognize any Unicode newline sequence
+  PCRE_NEWLINE_ANYCRLF    Recognize CR, LF, and CRLF as newline
+                            sequences
+  PCRE_NEWLINE_CR         Set CR as the newline sequence
+  PCRE_NEWLINE_CRLF       Set CRLF as the newline sequence
+  PCRE_NEWLINE_LF         Set LF as the newline sequence
+  PCRE_NO_AUTO_CAPTURE    Disable numbered capturing paren-
+                            theses (named ones available)
+  PCRE_NO_UTF16_CHECK     Do not check the pattern for UTF-16
+                            validity (only relevant if
+                            PCRE_UTF16 is set)
+  PCRE_NO_UTF32_CHECK     Do not check the pattern for UTF-32
+                            validity (only relevant if
+                            PCRE_UTF32 is set)
+  PCRE_NO_UTF8_CHECK      Do not check the pattern for UTF-8
+                            validity (only relevant if
+                            PCRE_UTF8 is set)
+  PCRE_UCP                Use Unicode properties for \d, \w, etc.
+  PCRE_UNGREEDY           Invert greediness of quantifiers
+  PCRE_UTF16              Run pcre16_compile() in UTF-16 mode
+  PCRE_UTF32              Run pcre32_compile() in UTF-32 mode
+  PCRE_UTF8               Run pcre_compile() in UTF-8 mode
+
+PCRE must be built with UTF support in order to use PCRE_UTF8/16/32 and +PCRE_NO_UTF8/16/32_CHECK, and with UCP support if PCRE_UCP is used. +

+

+The yield of the function is a pointer to a private data structure that +contains the compiled pattern, or NULL if an error was detected. Note that +compiling regular expressions with one version of PCRE for use with a different +version is not guaranteed to work and may cause crashes. +

+

+There is a complete description of the PCRE native API in the +pcreapi +page and a description of the POSIX API in the +pcreposix +page. +

+Return to the PCRE index page. +

diff --git a/pcre/doc/html/pcre_config.html b/pcre/doc/html/pcre_config.html new file mode 100644 index 00000000000..fc10d183933 --- /dev/null +++ b/pcre/doc/html/pcre_config.html @@ -0,0 +1,91 @@ + + +pcre_config specification + + +

pcre_config man page

+

+Return to the PCRE index page. +

+

+This page is part of the PCRE HTML documentation. It was generated automatically +from the original man page. If there is any nonsense in it, please consult the +man page, in case the conversion went wrong. +
+
+SYNOPSIS +
+

+#include <pcre.h> +

+

+int pcre_config(int what, void *where); +

+

+int pcre16_config(int what, void *where); +

+

+int pcre32_config(int what, void *where); +

+
+DESCRIPTION +
+

+This function makes it possible for a client program to find out which optional +features are available in the version of the PCRE library it is using. The +arguments are as follows: +

+  what     A code specifying what information is required
+  where    Points to where to put the data
+
+The where argument must point to an integer variable, except for +PCRE_CONFIG_MATCH_LIMIT and PCRE_CONFIG_MATCH_LIMIT_RECURSION, when it must +point to an unsigned long integer. The available codes are: +
+  PCRE_CONFIG_JIT           Availability of just-in-time compiler
+                              support (1=yes 0=no)
+  PCRE_CONFIG_JITTARGET     String containing information about the
+                              target architecture for the JIT compiler,
+                              or NULL if there is no JIT support
+  PCRE_CONFIG_LINK_SIZE     Internal link size: 2, 3, or 4
+  PCRE_CONFIG_MATCH_LIMIT   Internal resource limit
+  PCRE_CONFIG_MATCH_LIMIT_RECURSION
+                            Internal recursion depth limit
+  PCRE_CONFIG_NEWLINE       Value of the default newline sequence:
+                                13 (0x000d)    for CR
+                                10 (0x000a)    for LF
+                              3338 (0x0d0a)    for CRLF
+                                -2             for ANYCRLF
+                                -1             for ANY
+  PCRE_CONFIG_BSR           Indicates what \R matches by default:
+                                 0             all Unicode line endings
+                                 1             CR, LF, or CRLF only
+  PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
+                            Threshold of return slots, above which
+                              malloc() is used by the POSIX API
+  PCRE_CONFIG_STACKRECURSE  Recursion implementation (1=stack 0=heap)
+  PCRE_CONFIG_UTF16         Availability of UTF-16 support (1=yes
+                               0=no); option for pcre16_config()
+  PCRE_CONFIG_UTF32         Availability of UTF-32 support (1=yes
+                               0=no); option for pcre32_config()
+  PCRE_CONFIG_UTF8          Availability of UTF-8 support (1=yes 0=no);
+                              option for pcre_config()
+  PCRE_CONFIG_UNICODE_PROPERTIES
+                            Availability of Unicode property support
+                              (1=yes 0=no)
+
+The function yields 0 on success or PCRE_ERROR_BADOPTION otherwise. That error +is also given if PCRE_CONFIG_UTF16 or PCRE_CONFIG_UTF32 is passed to +pcre_config(), if PCRE_CONFIG_UTF8 or PCRE_CONFIG_UTF32 is passed to +pcre16_config(), or if PCRE_CONFIG_UTF8 or PCRE_CONFIG_UTF16 is passed to +pcre32_config(). +

+

+There is a complete description of the PCRE native API in the +pcreapi +page and a description of the POSIX API in the +pcreposix +page. +

+Return to the PCRE index page. +

diff --git a/pcre/doc/html/pcre_copy_named_substring.html b/pcre/doc/html/pcre_copy_named_substring.html new file mode 100644 index 00000000000..ae4f6905ccc --- /dev/null +++ b/pcre/doc/html/pcre_copy_named_substring.html @@ -0,0 +1,65 @@ + + +pcre_copy_named_substring specification + + +

pcre_copy_named_substring man page

+

+Return to the PCRE index page. +

+

+This page is part of the PCRE HTML documentation. It was generated automatically +from the original man page. If there is any nonsense in it, please consult the +man page, in case the conversion went wrong. +
+
+SYNOPSIS +
+

+#include <pcre.h> +

+

+int pcre_copy_named_substring(const pcre *code, +const char *subject, int *ovector, +int stringcount, const char *stringname, +char *buffer, int buffersize); +

+

+int pcre16_copy_named_substring(const pcre16 *code, +PCRE_SPTR16 subject, int *ovector, +int stringcount, PCRE_SPTR16 stringname, +PCRE_UCHAR16 *buffer, int buffersize); +

+

+int pcre32_copy_named_substring(const pcre32 *code, +PCRE_SPTR32 subject, int *ovector, +int stringcount, PCRE_SPTR32 stringname, +PCRE_UCHAR32 *buffer, int buffersize); +

+
+DESCRIPTION +
+

+This is a convenience function for extracting a captured substring, identified +by name, into a given buffer. The arguments are: +

+  code          Pattern that was successfully matched
+  subject       Subject that has been successfully matched
+  ovector       Offset vector that pcre[16|32]_exec() used
+  stringcount   Value returned by pcre[16|32]_exec()
+  stringname    Name of the required substring
+  buffer        Buffer to receive the string
+  buffersize    Size of buffer
+
+The yield is the length of the substring, PCRE_ERROR_NOMEMORY if the buffer was +too small, or PCRE_ERROR_NOSUBSTRING if the string name is invalid. +

+

+There is a complete description of the PCRE native API in the +pcreapi +page and a description of the POSIX API in the +pcreposix +page. +

+Return to the PCRE index page. +

diff --git a/pcre/doc/html/pcre_copy_substring.html b/pcre/doc/html/pcre_copy_substring.html new file mode 100644 index 00000000000..12bfb636a77 --- /dev/null +++ b/pcre/doc/html/pcre_copy_substring.html @@ -0,0 +1,61 @@ + + +pcre_copy_substring specification + + +

pcre_copy_substring man page

+

+Return to the PCRE index page. +

+

+This page is part of the PCRE HTML documentation. It was generated automatically +from the original man page. If there is any nonsense in it, please consult the +man page, in case the conversion went wrong. +
+
+SYNOPSIS +
+

+#include <pcre.h> +

+

+int pcre_copy_substring(const char *subject, int *ovector, +int stringcount, int stringnumber, char *buffer, +int buffersize); +

+

+int pcre16_copy_substring(PCRE_SPTR16 subject, int *ovector, +int stringcount, int stringnumber, PCRE_UCHAR16 *buffer, +int buffersize); +

+

+int pcre32_copy_substring(PCRE_SPTR32 subject, int *ovector, +int stringcount, int stringnumber, PCRE_UCHAR32 *buffer, +int buffersize); +

+
+DESCRIPTION +
+

+This is a convenience function for extracting a captured substring into a given +buffer. The arguments are: +

+  subject       Subject that has been successfully matched
+  ovector       Offset vector that pcre[16|32]_exec() used
+  stringcount   Value returned by pcre[16|32]_exec()
+  stringnumber  Number of the required substring
+  buffer        Buffer to receive the string
+  buffersize    Size of buffer
+
+The yield is the length of the string, PCRE_ERROR_NOMEMORY if the buffer was +too small, or PCRE_ERROR_NOSUBSTRING if the string number is invalid. +

+

+There is a complete description of the PCRE native API in the +pcreapi +page and a description of the POSIX API in the +pcreposix +page. +

+Return to the PCRE index page. +

diff --git a/pcre/doc/html/pcre_dfa_exec.html b/pcre/doc/html/pcre_dfa_exec.html new file mode 100644 index 00000000000..e91b670eac6 --- /dev/null +++ b/pcre/doc/html/pcre_dfa_exec.html @@ -0,0 +1,129 @@ + + +pcre_dfa_exec specification + + +

pcre_dfa_exec man page

+

+Return to the PCRE index page. +

+

+This page is part of the PCRE HTML documentation. It was generated automatically +from the original man page. If there is any nonsense in it, please consult the +man page, in case the conversion went wrong. +
+
+SYNOPSIS +
+

+#include <pcre.h> +

+

+int pcre_dfa_exec(const pcre *code, const pcre_extra *extra, +const char *subject, int length, int startoffset, +int options, int *ovector, int ovecsize, +int *workspace, int wscount); +

+

+int pcre16_dfa_exec(const pcre16 *code, const pcre16_extra *extra, +PCRE_SPTR16 subject, int length, int startoffset, +int options, int *ovector, int ovecsize, +int *workspace, int wscount); +

+

+int pcre32_dfa_exec(const pcre32 *code, const pcre32_extra *extra, +PCRE_SPTR32 subject, int length, int startoffset, +int options, int *ovector, int ovecsize, +int *workspace, int wscount); +

+
+DESCRIPTION +
+

+This function matches a compiled regular expression against a given subject +string, using an alternative matching algorithm that scans the subject string +just once (not Perl-compatible). Note that the main, Perl-compatible, +matching function is pcre[16|32]_exec(). The arguments for this function +are: +

+  code         Points to the compiled pattern
+  extra        Points to an associated pcre[16|32]_extra structure,
+                 or is NULL
+  subject      Points to the subject string
+  length       Length of the subject string
+  startoffset  Offset in the subject at which to start matching
+  options      Option bits
+  ovector      Points to a vector of ints for result offsets
+  ovecsize     Number of elements in the vector
+  workspace    Points to a vector of ints used as working space
+  wscount      Number of elements in the vector
+
+The units for length and startoffset are bytes for +pcre_exec(), 16-bit data items for pcre16_exec(), and 32-bit items +for pcre32_exec(). The options are: +
+  PCRE_ANCHORED          Match only at the first position
+  PCRE_BSR_ANYCRLF       \R matches only CR, LF, or CRLF
+  PCRE_BSR_UNICODE       \R matches all Unicode line endings
+  PCRE_NEWLINE_ANY       Recognize any Unicode newline sequence
+  PCRE_NEWLINE_ANYCRLF   Recognize CR, LF, & CRLF as newline sequences
+  PCRE_NEWLINE_CR        Recognize CR as the only newline sequence
+  PCRE_NEWLINE_CRLF      Recognize CRLF as the only newline sequence
+  PCRE_NEWLINE_LF        Recognize LF as the only newline sequence
+  PCRE_NOTBOL            Subject is not the beginning of a line
+  PCRE_NOTEOL            Subject is not the end of a line
+  PCRE_NOTEMPTY          An empty string is not a valid match
+  PCRE_NOTEMPTY_ATSTART  An empty string at the start of the subject
+                           is not a valid match
+  PCRE_NO_START_OPTIMIZE Do not do "start-match" optimizations
+  PCRE_NO_UTF16_CHECK    Do not check the subject for UTF-16
+                           validity (only relevant if PCRE_UTF16
+                           was set at compile time)
+  PCRE_NO_UTF32_CHECK    Do not check the subject for UTF-32
+                           validity (only relevant if PCRE_UTF32
+                           was set at compile time)
+  PCRE_NO_UTF8_CHECK     Do not check the subject for UTF-8
+                           validity (only relevant if PCRE_UTF8
+                           was set at compile time)
+  PCRE_PARTIAL           ) Return PCRE_ERROR_PARTIAL for a partial
+  PCRE_PARTIAL_SOFT      )   match if no full matches are found
+  PCRE_PARTIAL_HARD      Return PCRE_ERROR_PARTIAL for a partial match
+                           even if there is a full match as well
+  PCRE_DFA_SHORTEST      Return only the shortest match
+  PCRE_DFA_RESTART       Restart after a partial match
+
+There are restrictions on what may appear in a pattern when using this matching +function. Details are given in the +pcrematching +documentation. For details of partial matching, see the +pcrepartial +page. +

+

+A pcre[16|32]_extra structure contains the following fields: +

+  flags            Bits indicating which fields are set
+  study_data       Opaque data from pcre[16|32]_study()
+  match_limit      Limit on internal resource use
+  match_limit_recursion  Limit on internal recursion depth
+  callout_data     Opaque data passed back to callouts
+  tables           Points to character tables or is NULL
+  mark             For passing back a *MARK pointer
+  executable_jit   Opaque data from JIT compilation
+
+The flag bits are PCRE_EXTRA_STUDY_DATA, PCRE_EXTRA_MATCH_LIMIT, +PCRE_EXTRA_MATCH_LIMIT_RECURSION, PCRE_EXTRA_CALLOUT_DATA, +PCRE_EXTRA_TABLES, PCRE_EXTRA_MARK and PCRE_EXTRA_EXECUTABLE_JIT. For this +matching function, the match_limit and match_limit_recursion fields +are not used, and must not be set. The PCRE_EXTRA_EXECUTABLE_JIT flag and +the corresponding variable are ignored. +

+

+There is a complete description of the PCRE native API in the +pcreapi +page and a description of the POSIX API in the +pcreposix +page. +

+Return to the PCRE index page. +

diff --git a/pcre/doc/html/pcre_exec.html b/pcre/doc/html/pcre_exec.html new file mode 100644 index 00000000000..0cc3bb7bb8b --- /dev/null +++ b/pcre/doc/html/pcre_exec.html @@ -0,0 +1,111 @@ + + +pcre_exec specification + + +

pcre_exec man page

+

+Return to the PCRE index page. +

+

+This page is part of the PCRE HTML documentation. It was generated automatically +from the original man page. If there is any nonsense in it, please consult the +man page, in case the conversion went wrong. +
+
+SYNOPSIS +
+

+#include <pcre.h> +

+

+int pcre_exec(const pcre *code, const pcre_extra *extra, +const char *subject, int length, int startoffset, +int options, int *ovector, int ovecsize); +

+

+int pcre16_exec(const pcre16 *code, const pcre16_extra *extra, +PCRE_SPTR16 subject, int length, int startoffset, +int options, int *ovector, int ovecsize); +

+

+int pcre32_exec(const pcre32 *code, const pcre32_extra *extra, +PCRE_SPTR32 subject, int length, int startoffset, +int options, int *ovector, int ovecsize); +

+
+DESCRIPTION +
+

+This function matches a compiled regular expression against a given subject +string, using a matching algorithm that is similar to Perl's. It returns +offsets to captured substrings. Its arguments are: +

+  code         Points to the compiled pattern
+  extra        Points to an associated pcre[16|32]_extra structure,
+                 or is NULL
+  subject      Points to the subject string
+  length       Length of the subject string
+  startoffset  Offset in the subject at which to start matching
+  options      Option bits
+  ovector      Points to a vector of ints for result offsets
+  ovecsize     Number of elements in the vector (a multiple of 3)
+
+The units for length and startoffset are bytes for +pcre_exec(), 16-bit data items for pcre16_exec(), and 32-bit items +for pcre32_exec(). The options are: +
+  PCRE_ANCHORED          Match only at the first position
+  PCRE_BSR_ANYCRLF       \R matches only CR, LF, or CRLF
+  PCRE_BSR_UNICODE       \R matches all Unicode line endings
+  PCRE_NEWLINE_ANY       Recognize any Unicode newline sequence
+  PCRE_NEWLINE_ANYCRLF   Recognize CR, LF, & CRLF as newline sequences
+  PCRE_NEWLINE_CR        Recognize CR as the only newline sequence
+  PCRE_NEWLINE_CRLF      Recognize CRLF as the only newline sequence
+  PCRE_NEWLINE_LF        Recognize LF as the only newline sequence
+  PCRE_NOTBOL            Subject string is not the beginning of a line
+  PCRE_NOTEOL            Subject string is not the end of a line
+  PCRE_NOTEMPTY          An empty string is not a valid match
+  PCRE_NOTEMPTY_ATSTART  An empty string at the start of the subject
+                           is not a valid match
+  PCRE_NO_START_OPTIMIZE Do not do "start-match" optimizations
+  PCRE_NO_UTF16_CHECK    Do not check the subject for UTF-16
+                           validity (only relevant if PCRE_UTF16
+                           was set at compile time)
+  PCRE_NO_UTF32_CHECK    Do not check the subject for UTF-32
+                           validity (only relevant if PCRE_UTF32
+                           was set at compile time)
+  PCRE_NO_UTF8_CHECK     Do not check the subject for UTF-8
+                           validity (only relevant if PCRE_UTF8
+                           was set at compile time)
+  PCRE_PARTIAL           ) Return PCRE_ERROR_PARTIAL for a partial
+  PCRE_PARTIAL_SOFT      )   match if no full matches are found
+  PCRE_PARTIAL_HARD      Return PCRE_ERROR_PARTIAL for a partial match
+                           if that is found before a full match
+
+For details of partial matching, see the +pcrepartial +page. A pcre_extra structure contains the following fields: +
+  flags            Bits indicating which fields are set
+  study_data       Opaque data from pcre[16|32]_study()
+  match_limit      Limit on internal resource use
+  match_limit_recursion  Limit on internal recursion depth
+  callout_data     Opaque data passed back to callouts
+  tables           Points to character tables or is NULL
+  mark             For passing back a *MARK pointer
+  executable_jit   Opaque data from JIT compilation
+
+The flag bits are PCRE_EXTRA_STUDY_DATA, PCRE_EXTRA_MATCH_LIMIT, +PCRE_EXTRA_MATCH_LIMIT_RECURSION, PCRE_EXTRA_CALLOUT_DATA, +PCRE_EXTRA_TABLES, PCRE_EXTRA_MARK and PCRE_EXTRA_EXECUTABLE_JIT. +

+

+There is a complete description of the PCRE native API in the +pcreapi +page and a description of the POSIX API in the +pcreposix +page. +

+Return to the PCRE index page. +

diff --git a/pcre/doc/html/pcre_free_study.html b/pcre/doc/html/pcre_free_study.html new file mode 100644 index 00000000000..7f9e10e8632 --- /dev/null +++ b/pcre/doc/html/pcre_free_study.html @@ -0,0 +1,46 @@ + + +pcre_free_study specification + + +

pcre_free_study man page

+

+Return to the PCRE index page. +

+

+This page is part of the PCRE HTML documentation. It was generated automatically +from the original man page. If there is any nonsense in it, please consult the +man page, in case the conversion went wrong. +
+
+SYNOPSIS +
+

+#include <pcre.h> +

+

+void pcre_free_study(pcre_extra *extra); +

+

+void pcre16_free_study(pcre16_extra *extra); +

+

+void pcre32_free_study(pcre32_extra *extra); +

+
+DESCRIPTION +
+

+This function is used to free the memory used for the data generated by a call +to pcre[16|32]_study() when it is no longer needed. The argument must be the +result of such a call. +

+

+There is a complete description of the PCRE native API in the +pcreapi +page and a description of the POSIX API in the +pcreposix +page. +

+Return to the PCRE index page. +

diff --git a/pcre/doc/html/pcre_free_substring.html b/pcre/doc/html/pcre_free_substring.html new file mode 100644 index 00000000000..1fe66107463 --- /dev/null +++ b/pcre/doc/html/pcre_free_substring.html @@ -0,0 +1,46 @@ + + +pcre_free_substring specification + + +

pcre_free_substring man page

+

+Return to the PCRE index page. +

+

+This page is part of the PCRE HTML documentation. It was generated automatically +from the original man page. If there is any nonsense in it, please consult the +man page, in case the conversion went wrong. +
+
+SYNOPSIS +
+

+#include <pcre.h> +

+

+void pcre_free_substring(const char *stringptr); +

+

+void pcre16_free_substring(PCRE_SPTR16 stringptr); +

+

+void pcre32_free_substring(PCRE_SPTR32 stringptr); +

+
+DESCRIPTION +
+

+This is a convenience function for freeing the store obtained by a previous +call to pcre[16|32]_get_substring() or pcre[16|32]_get_named_substring(). +Its only argument is a pointer to the string. +

+

+There is a complete description of the PCRE native API in the +pcreapi +page and a description of the POSIX API in the +pcreposix +page. +

+Return to the PCRE index page. +

diff --git a/pcre/doc/html/pcre_free_substring_list.html b/pcre/doc/html/pcre_free_substring_list.html new file mode 100644 index 00000000000..c0861780b42 --- /dev/null +++ b/pcre/doc/html/pcre_free_substring_list.html @@ -0,0 +1,46 @@ + + +pcre_free_substring_list specification + + +

pcre_free_substring_list man page

+

+Return to the PCRE index page. +

+

+This page is part of the PCRE HTML documentation. It was generated automatically +from the original man page. If there is any nonsense in it, please consult the +man page, in case the conversion went wrong. +
+
+SYNOPSIS +
+

+#include <pcre.h> +

+

+void pcre_free_substring_list(const char **stringptr); +

+

+void pcre16_free_substring_list(PCRE_SPTR16 *stringptr); +

+

+void pcre32_free_substring_list(PCRE_SPTR32 *stringptr); +

+
+DESCRIPTION +
+

+This is a convenience function for freeing the store obtained by a previous +call to pcre[16|32]_get_substring_list(). Its only argument is a pointer to +the list of string pointers. +

+

+There is a complete description of the PCRE native API in the +pcreapi +page and a description of the POSIX API in the +pcreposix +page. +

+Return to the PCRE index page. +

diff --git a/pcre/doc/html/pcre_fullinfo.html b/pcre/doc/html/pcre_fullinfo.html new file mode 100644 index 00000000000..d353432bc52 --- /dev/null +++ b/pcre/doc/html/pcre_fullinfo.html @@ -0,0 +1,108 @@ + + +pcre_fullinfo specification + + +

pcre_fullinfo man page

+

+Return to the PCRE index page. +

+

+This page is part of the PCRE HTML documentation. It was generated automatically +from the original man page. If there is any nonsense in it, please consult the +man page, in case the conversion went wrong. +
+
+SYNOPSIS +
+

+#include <pcre.h> +

+

+int pcre_fullinfo(const pcre *code, const pcre_extra *extra, +int what, void *where); +

+

+int pcre16_fullinfo(const pcre16 *code, const pcre16_extra *extra, +int what, void *where); +

+

+int pcre32_fullinfo(const pcre32 *code, const pcre32_extra *extra, +int what, void *where); +

+
+DESCRIPTION +
+

+This function returns information about a compiled pattern. Its arguments are: +

+  code                      Compiled regular expression
+  extra                     Result of pcre[16|32]_study() or NULL
+  what                      What information is required
+  where                     Where to put the information
+
+The following information is available: +
+  PCRE_INFO_BACKREFMAX      Number of highest back reference
+  PCRE_INFO_CAPTURECOUNT    Number of capturing subpatterns
+  PCRE_INFO_DEFAULT_TABLES  Pointer to default tables
+  PCRE_INFO_FIRSTBYTE       Fixed first data unit for a match, or
+                              -1 for start of string
+                                 or after newline, or
+                              -2 otherwise
+  PCRE_INFO_FIRSTTABLE      Table of first data units (after studying)
+  PCRE_INFO_HASCRORLF       Return 1 if explicit CR or LF matches exist
+  PCRE_INFO_JCHANGED        Return 1 if (?J) or (?-J) was used
+  PCRE_INFO_JIT             Return 1 after successful JIT compilation
+  PCRE_INFO_JITSIZE         Size of JIT compiled code
+  PCRE_INFO_LASTLITERAL     Literal last data unit required
+  PCRE_INFO_MINLENGTH       Lower bound length of matching strings
+  PCRE_INFO_NAMECOUNT       Number of named subpatterns
+  PCRE_INFO_NAMEENTRYSIZE   Size of name table entry
+  PCRE_INFO_NAMETABLE       Pointer to name table
+  PCRE_INFO_OKPARTIAL       Return 1 if partial matching can be tried
+                              (always returns 1 after release 8.00)
+  PCRE_INFO_OPTIONS         Option bits used for compilation
+  PCRE_INFO_SIZE            Size of compiled pattern
+  PCRE_INFO_STUDYSIZE       Size of study data
+  PCRE_INFO_FIRSTCHARACTER      Fixed first data unit for a match
+  PCRE_INFO_FIRSTCHARACTERFLAGS Returns
+                                  1 if there is a first data character set, which can
+                                    then be retrieved using PCRE_INFO_FIRSTCHARACTER,
+                                  2 if the first character is at the start of the data
+                                    string or after a newline, and
+                                  0 otherwise
+  PCRE_INFO_REQUIREDCHAR      Literal last data unit required
+  PCRE_INFO_REQUIREDCHARFLAGS Returns 1 if the last data character is set (which can then
+                              be retrieved using PCRE_INFO_REQUIREDCHAR); 0 otherwise
+
+The where argument must point to an integer variable, except for the +following what values: +
+  PCRE_INFO_DEFAULT_TABLES  const unsigned char *
+  PCRE_INFO_FIRSTTABLE      const unsigned char *
+  PCRE_INFO_NAMETABLE       PCRE_SPTR16           (16-bit library)
+  PCRE_INFO_NAMETABLE       PCRE_SPTR32           (32-bit library)
+  PCRE_INFO_NAMETABLE       const unsigned char * (8-bit library)
+  PCRE_INFO_OPTIONS         unsigned long int
+  PCRE_INFO_SIZE            size_t
+  PCRE_INFO_FIRSTCHARACTER  uint32_t
+  PCRE_INFO_REQUIREDCHAR    uint32_t
+
+The yield of the function is zero on success or: +
+  PCRE_ERROR_NULL           the argument code was NULL
+                            the argument where was NULL
+  PCRE_ERROR_BADMAGIC       the "magic number" was not found
+  PCRE_ERROR_BADOPTION      the value of what was invalid
+
+

+

+There is a complete description of the PCRE native API in the +pcreapi +page and a description of the POSIX API in the +pcreposix +page. +

+Return to the PCRE index page. +

diff --git a/pcre/doc/html/pcre_get_named_substring.html b/pcre/doc/html/pcre_get_named_substring.html new file mode 100644 index 00000000000..6150ad71424 --- /dev/null +++ b/pcre/doc/html/pcre_get_named_substring.html @@ -0,0 +1,68 @@ + + +pcre_get_named_substring specification + + +

pcre_get_named_substring man page

+

+Return to the PCRE index page. +

+

+This page is part of the PCRE HTML documentation. It was generated automatically +from the original man page. If there is any nonsense in it, please consult the +man page, in case the conversion went wrong. +
+
+SYNOPSIS +
+

+#include <pcre.h> +

+

+int pcre_get_named_substring(const pcre *code, +const char *subject, int *ovector, +int stringcount, const char *stringname, +const char **stringptr); +

+

+int pcre16_get_named_substring(const pcre16 *code, +PCRE_SPTR16 subject, int *ovector, +int stringcount, PCRE_SPTR16 stringname, +PCRE_SPTR16 *stringptr); +

+

+int pcre32_get_named_substring(const pcre32 *code, +PCRE_SPTR32 subject, int *ovector, +int stringcount, PCRE_SPTR32 stringname, +PCRE_SPTR32 *stringptr); +

+
+DESCRIPTION +
+

+This is a convenience function for extracting a captured substring by name. The +arguments are: +

+  code          Compiled pattern
+  subject       Subject that has been successfully matched
+  ovector       Offset vector that pcre[16|32]_exec() used
+  stringcount   Value returned by pcre[16|32]_exec()
+  stringname    Name of the required substring
+  stringptr     Where to put the string pointer
+
+The memory in which the substring is placed is obtained by calling +pcre[16|32]_malloc(). The convenience function +pcre[16|32]_free_substring() can be used to free it when it is no longer +needed. The yield of the function is the length of the extracted substring, +PCRE_ERROR_NOMEMORY if sufficient memory could not be obtained, or +PCRE_ERROR_NOSUBSTRING if the string name is invalid. +

+

+There is a complete description of the PCRE native API in the +pcreapi +page and a description of the POSIX API in the +pcreposix +page. +

+Return to the PCRE index page. +

diff --git a/pcre/doc/html/pcre_get_stringnumber.html b/pcre/doc/html/pcre_get_stringnumber.html new file mode 100644 index 00000000000..08967de3f7d --- /dev/null +++ b/pcre/doc/html/pcre_get_stringnumber.html @@ -0,0 +1,57 @@ + + +pcre_get_stringnumber specification + + +

pcre_get_stringnumber man page

+

+Return to the PCRE index page. +

+

+This page is part of the PCRE HTML documentation. It was generated automatically +from the original man page. If there is any nonsense in it, please consult the +man page, in case the conversion went wrong. +
+
+SYNOPSIS +
+

+#include <pcre.h> +

+

+int pcre_get_stringnumber(const pcre *code, +const char *name); +

+

+int pcre16_get_stringnumber(const pcre16 *code, +PCRE_SPTR16 name); +

+

+int pcre32_get_stringnumber(const pcre32 *code, +PCRE_SPTR32 name); +

+
+DESCRIPTION +
+

+This convenience function finds the number of a named substring capturing +parenthesis in a compiled pattern. Its arguments are: +

+  code    Compiled regular expression
+  name    Name whose number is required
+
+The yield of the function is the number of the parenthesis if the name is +found, or PCRE_ERROR_NOSUBSTRING otherwise. When duplicate names are allowed +(PCRE_DUPNAMES is set), it is not defined which of the numbers is returned by +pcre[16|32]_get_stringnumber(). You can obtain the complete list by calling +pcre[16|32]_get_stringtable_entries(). +

+

+There is a complete description of the PCRE native API in the +pcreapi +page and a description of the POSIX API in the +pcreposix +page. +

+Return to the PCRE index page. +

diff --git a/pcre/doc/html/pcre_get_stringtable_entries.html b/pcre/doc/html/pcre_get_stringtable_entries.html new file mode 100644 index 00000000000..38f9c0c902f --- /dev/null +++ b/pcre/doc/html/pcre_get_stringtable_entries.html @@ -0,0 +1,60 @@ + + +pcre_get_stringtable_entries specification + + +

pcre_get_stringtable_entries man page

+

+Return to the PCRE index page. +

+

+This page is part of the PCRE HTML documentation. It was generated automatically +from the original man page. If there is any nonsense in it, please consult the +man page, in case the conversion went wrong. +
+
+SYNOPSIS +
+

+#include <pcre.h> +

+

+int pcre_get_stringtable_entries(const pcre *code, +const char *name, char **first, char **last); +

+

+int pcre16_get_stringtable_entries(const pcre16 *code, +PCRE_SPTR16 name, PCRE_UCHAR16 **first, PCRE_UCHAR16 **last); +

+

+int pcre32_get_stringtable_entries(const pcre32 *code, +PCRE_SPTR32 name, PCRE_UCHAR32 **first, PCRE_UCHAR32 **last); +

+
+DESCRIPTION +
+

+This convenience function finds, for a compiled pattern, the first and last +entries for a given name in the table that translates capturing parenthesis +names into numbers. When names are required to be unique (PCRE_DUPNAMES is +not set), it is usually easier to use pcre[16|32]_get_stringnumber() +instead. +

+  code    Compiled regular expression
+  name    Name whose entries required
+  first   Where to return a pointer to the first entry
+  last    Where to return a pointer to the last entry
+
+The yield of the function is the length of each entry, or +PCRE_ERROR_NOSUBSTRING if none are found. +

+

+There is a complete description of the PCRE native API, including the format of +the table entries, in the +pcreapi +page, and a description of the POSIX API in the +pcreposix +page. +

+Return to the PCRE index page. +

diff --git a/pcre/doc/html/pcre_get_substring.html b/pcre/doc/html/pcre_get_substring.html new file mode 100644 index 00000000000..2a5a610f971 --- /dev/null +++ b/pcre/doc/html/pcre_get_substring.html @@ -0,0 +1,64 @@ + + +pcre_get_substring specification + + +

pcre_get_substring man page

+

+Return to the PCRE index page. +

+

+This page is part of the PCRE HTML documentation. It was generated automatically +from the original man page. If there is any nonsense in it, please consult the +man page, in case the conversion went wrong. +
+
+SYNOPSIS +
+

+#include <pcre.h> +

+

+int pcre_get_substring(const char *subject, int *ovector, +int stringcount, int stringnumber, +const char **stringptr); +

+

+int pcre16_get_substring(PCRE_SPTR16 subject, int *ovector, +int stringcount, int stringnumber, +PCRE_SPTR16 *stringptr); +

+

+int pcre32_get_substring(PCRE_SPTR32 subject, int *ovector, +int stringcount, int stringnumber, +PCRE_SPTR32 *stringptr); +

+
+DESCRIPTION +
+

+This is a convenience function for extracting a captured substring. The +arguments are: +

+  subject       Subject that has been successfully matched
+  ovector       Offset vector that pcre[16|32]_exec() used
+  stringcount   Value returned by pcre[16|32]_exec()
+  stringnumber  Number of the required substring
+  stringptr     Where to put the string pointer
+
+The memory in which the substring is placed is obtained by calling +pcre[16|32]_malloc(). The convenience function +pcre[16|32]_free_substring() can be used to free it when it is no longer +needed. The yield of the function is the length of the substring, +PCRE_ERROR_NOMEMORY if sufficient memory could not be obtained, or +PCRE_ERROR_NOSUBSTRING if the string number is invalid. +

+

+There is a complete description of the PCRE native API in the +pcreapi +page and a description of the POSIX API in the +pcreposix +page. +

+Return to the PCRE index page. +

diff --git a/pcre/doc/html/pcre_get_substring_list.html b/pcre/doc/html/pcre_get_substring_list.html new file mode 100644 index 00000000000..85edef4b9f4 --- /dev/null +++ b/pcre/doc/html/pcre_get_substring_list.html @@ -0,0 +1,61 @@ + + +pcre_get_substring_list specification + + +

pcre_get_substring_list man page

+

+Return to the PCRE index page. +

+

+This page is part of the PCRE HTML documentation. It was generated automatically +from the original man page. If there is any nonsense in it, please consult the +man page, in case the conversion went wrong. +
+
+SYNOPSIS +
+

+#include <pcre.h> +

+

+int pcre_get_substring_list(const char *subject, +int *ovector, int stringcount, const char ***listptr); +

+

+int pcre16_get_substring_list(PCRE_SPTR16 subject, +int *ovector, int stringcount, PCRE_SPTR16 **listptr); +

+

+int pcre32_get_substring_list(PCRE_SPTR32 subject, +int *ovector, int stringcount, PCRE_SPTR32 **listptr); +

+
+DESCRIPTION +
+

+This is a convenience function for extracting a list of all the captured +substrings. The arguments are: +

+  subject       Subject that has been successfully matched
+  ovector       Offset vector that pcre[16|32]_exec used
+  stringcount   Value returned by pcre[16|32]_exec
+  listptr       Where to put a pointer to the list
+
+The memory in which the substrings and the list are placed is obtained by +calling pcre[16|32]_malloc(). The convenience function +pcre[16|32]_free_substring_list() can be used to free it when it is no +longer needed. A pointer to a list of pointers is put in the variable whose +address is in listptr. The list is terminated by a NULL pointer. The +yield of the function is zero on success or PCRE_ERROR_NOMEMORY if sufficient +memory could not be obtained. +

+

+There is a complete description of the PCRE native API in the +pcreapi +page and a description of the POSIX API in the +pcreposix +page. +

+Return to the PCRE index page. +

diff --git a/pcre/doc/html/pcre_jit_exec.html b/pcre/doc/html/pcre_jit_exec.html new file mode 100644 index 00000000000..0c63503a997 --- /dev/null +++ b/pcre/doc/html/pcre_jit_exec.html @@ -0,0 +1,108 @@ + + +pcre_jit_exec specification + + +

pcre_jit_exec man page

+

+Return to the PCRE index page. +

+

+This page is part of the PCRE HTML documentation. It was generated automatically +from the original man page. If there is any nonsense in it, please consult the +man page, in case the conversion went wrong. +
+
+SYNOPSIS +
+

+#include <pcre.h> +

+

+int pcre_jit_exec(const pcre *code, const pcre_extra *extra, +const char *subject, int length, int startoffset, +int options, int *ovector, int ovecsize, +pcre_jit_stack *jstack); +

+

+int pcre16_jit_exec(const pcre16 *code, const pcre16_extra *extra, +PCRE_SPTR16 subject, int length, int startoffset, +int options, int *ovector, int ovecsize, +pcre_jit_stack *jstack); +

+

+int pcre32_jit_exec(const pcre32 *code, const pcre32_extra *extra, +PCRE_SPTR32 subject, int length, int startoffset, +int options, int *ovector, int ovecsize, +pcre_jit_stack *jstack); +

+
+DESCRIPTION +
+

+This function matches a compiled regular expression that has been successfully +studied with one of the JIT options against a given subject string, using a +matching algorithm that is similar to Perl's. It is a "fast path" interface to +JIT, and it bypasses some of the sanity checks that pcre_exec() applies. +It returns offsets to captured substrings. Its arguments are: +

+  code         Points to the compiled pattern
+  extra        Points to an associated pcre[16|32]_extra structure,
+                 or is NULL
+  subject      Points to the subject string
+  length       Length of the subject string, in bytes
+  startoffset  Offset in bytes in the subject at which to
+                 start matching
+  options      Option bits
+  ovector      Points to a vector of ints for result offsets
+  ovecsize     Number of elements in the vector (a multiple of 3)
+  jstack       Pointer to a JIT stack
+
+The allowed options are: +
+  PCRE_NOTBOL            Subject string is not the beginning of a line
+  PCRE_NOTEOL            Subject string is not the end of a line
+  PCRE_NOTEMPTY          An empty string is not a valid match
+  PCRE_NOTEMPTY_ATSTART  An empty string at the start of the subject
+                           is not a valid match
+  PCRE_NO_UTF16_CHECK    Do not check the subject for UTF-16
+                           validity (only relevant if PCRE_UTF16
+                           was set at compile time)
+  PCRE_NO_UTF32_CHECK    Do not check the subject for UTF-32
+                           validity (only relevant if PCRE_UTF32
+                           was set at compile time)
+  PCRE_NO_UTF8_CHECK     Do not check the subject for UTF-8
+                           validity (only relevant if PCRE_UTF8
+                           was set at compile time)
+  PCRE_PARTIAL           ) Return PCRE_ERROR_PARTIAL for a partial
+  PCRE_PARTIAL_SOFT      )   match if no full matches are found
+  PCRE_PARTIAL_HARD      Return PCRE_ERROR_PARTIAL for a partial match
+                           if that is found before a full match
+
+However, the PCRE_NO_UTF[8|16|32]_CHECK options have no effect, as this check +is never applied. For details of partial matching, see the +pcrepartial +page. A pcre_extra structure contains the following fields: +
+  flags            Bits indicating which fields are set
+  study_data       Opaque data from pcre[16|32]_study()
+  match_limit      Limit on internal resource use
+  match_limit_recursion  Limit on internal recursion depth
+  callout_data     Opaque data passed back to callouts
+  tables           Points to character tables or is NULL
+  mark             For passing back a *MARK pointer
+  executable_jit   Opaque data from JIT compilation
+
+The flag bits are PCRE_EXTRA_STUDY_DATA, PCRE_EXTRA_MATCH_LIMIT, +PCRE_EXTRA_MATCH_LIMIT_RECURSION, PCRE_EXTRA_CALLOUT_DATA, +PCRE_EXTRA_TABLES, PCRE_EXTRA_MARK and PCRE_EXTRA_EXECUTABLE_JIT. +

+

+There is a complete description of the PCRE native API in the +pcreapi +page and a description of the JIT API in the +pcrejit +page. +

+Return to the PCRE index page. +

diff --git a/pcre/doc/html/pcre_jit_stack_alloc.html b/pcre/doc/html/pcre_jit_stack_alloc.html new file mode 100644 index 00000000000..4153ee594ff --- /dev/null +++ b/pcre/doc/html/pcre_jit_stack_alloc.html @@ -0,0 +1,55 @@ + + +pcre_jit_stack_alloc specification + + +

pcre_jit_stack_alloc man page

+

+Return to the PCRE index page. +

+

+This page is part of the PCRE HTML documentation. It was generated automatically +from the original man page. If there is any nonsense in it, please consult the +man page, in case the conversion went wrong. +
+
+SYNOPSIS +
+

+#include <pcre.h> +

+

+pcre_jit_stack *pcre_jit_stack_alloc(int startsize, +int maxsize); +

+

+pcre16_jit_stack *pcre16_jit_stack_alloc(int startsize, +int maxsize); +

+

+pcre32_jit_stack *pcre32_jit_stack_alloc(int startsize, +int maxsize); +

+
+DESCRIPTION +
+

+This function is used to create a stack for use by the code compiled by the JIT +optimization of pcre[16|32]_study(). The arguments are a starting size for +the stack, and a maximum size to which it is allowed to grow. The result can be +passed to the JIT run-time code by pcre[16|32]_assign_jit_stack(), or that +function can set up a callback for obtaining a stack. A maximum stack size of +512K to 1M should be more than enough for any pattern. For more details, see +the +pcrejit +page. +

+

+There is a complete description of the PCRE native API in the +pcreapi +page and a description of the POSIX API in the +pcreposix +page. +

+Return to the PCRE index page. +

diff --git a/pcre/doc/html/pcre_jit_stack_free.html b/pcre/doc/html/pcre_jit_stack_free.html new file mode 100644 index 00000000000..8bd06e4655a --- /dev/null +++ b/pcre/doc/html/pcre_jit_stack_free.html @@ -0,0 +1,48 @@ + + +pcre_jit_stack_free specification + + +

pcre_jit_stack_free man page

+

+Return to the PCRE index page. +

+

+This page is part of the PCRE HTML documentation. It was generated automatically +from the original man page. If there is any nonsense in it, please consult the +man page, in case the conversion went wrong. +
+
+SYNOPSIS +
+

+#include <pcre.h> +

+

+void pcre_jit_stack_free(pcre_jit_stack *stack); +

+

+void pcre16_jit_stack_free(pcre16_jit_stack *stack); +

+

+void pcre32_jit_stack_free(pcre32_jit_stack *stack); +

+
+DESCRIPTION +
+

+This function is used to free a JIT stack that was created by +pcre[16|32]_jit_stack_alloc() when it is no longer needed. For more details, +see the +pcrejit +page. +

+

+There is a complete description of the PCRE native API in the +pcreapi +page and a description of the POSIX API in the +pcreposix +page. +

+Return to the PCRE index page. +

diff --git a/pcre/doc/html/pcre_maketables.html b/pcre/doc/html/pcre_maketables.html new file mode 100644 index 00000000000..3a7b5ebc4a9 --- /dev/null +++ b/pcre/doc/html/pcre_maketables.html @@ -0,0 +1,48 @@ + + +pcre_maketables specification + + +

pcre_maketables man page

+

+Return to the PCRE index page. +

+

+This page is part of the PCRE HTML documentation. It was generated automatically +from the original man page. If there is any nonsense in it, please consult the +man page, in case the conversion went wrong. +
+
+SYNOPSIS +
+

+#include <pcre.h> +

+

+const unsigned char *pcre_maketables(void); +

+

+const unsigned char *pcre16_maketables(void); +

+

+const unsigned char *pcre32_maketables(void); +

+
+DESCRIPTION +
+

+This function builds a set of character tables for character values less than +256. These can be passed to pcre[16|32]_compile() to override PCRE's +internal, built-in tables (which were made by pcre[16|32]_maketables() when +PCRE was compiled). You might want to do this if you are using a non-standard +locale. The function yields a pointer to the tables. +

+

+There is a complete description of the PCRE native API in the +pcreapi +page and a description of the POSIX API in the +pcreposix +page. +

+Return to the PCRE index page. +

diff --git a/pcre/doc/html/pcre_pattern_to_host_byte_order.html b/pcre/doc/html/pcre_pattern_to_host_byte_order.html new file mode 100644 index 00000000000..68d6f5a10a3 --- /dev/null +++ b/pcre/doc/html/pcre_pattern_to_host_byte_order.html @@ -0,0 +1,58 @@ + + +pcre_pattern_to_host_byte_order specification + + +

pcre_pattern_to_host_byte_order man page

+

+Return to the PCRE index page. +

+

+This page is part of the PCRE HTML documentation. It was generated automatically +from the original man page. If there is any nonsense in it, please consult the +man page, in case the conversion went wrong. +
+
+SYNOPSIS +
+

+#include <pcre.h> +

+

+int pcre_pattern_to_host_byte_order(pcre *code, +pcre_extra *extra, const unsigned char *tables); +

+

+int pcre16_pattern_to_host_byte_order(pcre16 *code, +pcre16_extra *extra, const unsigned char *tables); +

+

+int pcre32_pattern_to_host_byte_order(pcre32 *code, +pcre32_extra *extra, const unsigned char *tables); +

+
+DESCRIPTION +
+

+This function ensures that the bytes in 2-byte and 4-byte values in a compiled +pattern are in the correct order for the current host. It is useful when a +pattern that has been compiled on one host is transferred to another that might +have different endianness. The arguments are: +

+  code         A compiled regular expression
+  extra        Points to an associated pcre[16|32]_extra structure,
+                 or is NULL
+  tables       Pointer to character tables, or NULL to
+                 set the built-in default
+
+The result is 0 for success, a negative PCRE_ERROR_xxx value otherwise. +

+

+There is a complete description of the PCRE native API in the +pcreapi +page and a description of the POSIX API in the +pcreposix +page. +

+Return to the PCRE index page. +

diff --git a/pcre/doc/html/pcre_refcount.html b/pcre/doc/html/pcre_refcount.html new file mode 100644 index 00000000000..bfb92e6d8a9 --- /dev/null +++ b/pcre/doc/html/pcre_refcount.html @@ -0,0 +1,51 @@ + + +pcre_refcount specification + + +

pcre_refcount man page

+

+Return to the PCRE index page. +

+

+This page is part of the PCRE HTML documentation. It was generated automatically +from the original man page. If there is any nonsense in it, please consult the +man page, in case the conversion went wrong. +
+
+SYNOPSIS +
+

+#include <pcre.h> +

+

+int pcre_refcount(pcre *code, int adjust); +

+

+int pcre16_refcount(pcre16 *code, int adjust); +

+

+int pcre32_refcount(pcre32 *code, int adjust); +

+
+DESCRIPTION +
+

+This function is used to maintain a reference count inside a data block that +contains a compiled pattern. Its arguments are: +

+  code                      Compiled regular expression
+  adjust                    Adjustment to reference value
+
+The yield of the function is the adjusted reference value, which is constrained +to lie between 0 and 65535. +

+

+There is a complete description of the PCRE native API in the +pcreapi +page and a description of the POSIX API in the +pcreposix +page. +

+Return to the PCRE index page. +

diff --git a/pcre/doc/html/pcre_study.html b/pcre/doc/html/pcre_study.html new file mode 100644 index 00000000000..2baf54c4689 --- /dev/null +++ b/pcre/doc/html/pcre_study.html @@ -0,0 +1,68 @@ + + +pcre_study specification + + +

pcre_study man page

+

+Return to the PCRE index page. +

+

+This page is part of the PCRE HTML documentation. It was generated automatically +from the original man page. If there is any nonsense in it, please consult the +man page, in case the conversion went wrong. +
+
+SYNOPSIS +
+

+#include <pcre.h> +

+

+pcre_extra *pcre_study(const pcre *code, int options, +const char **errptr); +

+

+pcre16_extra *pcre16_study(const pcre16 *code, int options, +const char **errptr); +

+

+pcre32_extra *pcre32_study(const pcre32 *code, int options, +const char **errptr); +

+
+DESCRIPTION +
+

+This function studies a compiled pattern, to see if additional information can +be extracted that might speed up matching. Its arguments are: +

+  code       A compiled regular expression
+  options    Options for pcre[16|32]_study()
+  errptr     Where to put an error message
+
+If the function succeeds, it returns a value that can be passed to +pcre[16|32]_exec() or pcre[16|32]_dfa_exec() via their extra +arguments. +

+

+If the function returns NULL, either it could not find any additional +information, or there was an error. You can tell the difference by looking at +the error value. It is NULL in first case. +

+

+The only option is PCRE_STUDY_JIT_COMPILE. It requests just-in-time compilation +if possible. If PCRE has been compiled without JIT support, this option is +ignored. See the +pcrejit +page for further details. +

+

+There is a complete description of the PCRE native API in the +pcreapi +page and a description of the POSIX API in the +pcreposix +page. +

+Return to the PCRE index page. +

diff --git a/pcre/doc/html/pcre_utf16_to_host_byte_order.html b/pcre/doc/html/pcre_utf16_to_host_byte_order.html new file mode 100644 index 00000000000..164e2365a15 --- /dev/null +++ b/pcre/doc/html/pcre_utf16_to_host_byte_order.html @@ -0,0 +1,57 @@ + + +pcre_utf16_to_host_byte_order specification + + +

pcre_utf16_to_host_byte_order man page

+

+Return to the PCRE index page. +

+

+This page is part of the PCRE HTML documentation. It was generated automatically +from the original man page. If there is any nonsense in it, please consult the +man page, in case the conversion went wrong. +
+
+SYNOPSIS +
+

+#include <pcre.h> +

+

+int pcre16_utf16_to_host_byte_order(PCRE_UCHAR16 *output, +PCRE_SPTR16 input, int length, int *host_byte_order, +int keep_boms); +

+
+DESCRIPTION +
+

+This function, which exists only in the 16-bit library, converts a UTF-16 +string to the correct order for the current host, taking account of any byte +order marks (BOMs) within the string. Its arguments are: +

+  output           pointer to output buffer, may be the same as input
+  input            pointer to input buffer
+  length           number of 16-bit units in the input, or negative for
+                     a zero-terminated string
+  host_byte_order  a NULL value or a non-zero value pointed to means
+                     start in host byte order
+  keep_boms        if non-zero, BOMs are copied to the output string
+
+The result of the function is the number of 16-bit units placed into the output +buffer, including the zero terminator if the string was zero-terminated. +

+

+If host_byte_order is not NULL, it is set to indicate the byte order that +is current at the end of the string. +

+

+There is a complete description of the PCRE native API in the +pcreapi +page and a description of the POSIX API in the +pcreposix +page. +

+Return to the PCRE index page. +

diff --git a/pcre/doc/html/pcre_utf32_to_host_byte_order.html b/pcre/doc/html/pcre_utf32_to_host_byte_order.html new file mode 100644 index 00000000000..967c00db802 --- /dev/null +++ b/pcre/doc/html/pcre_utf32_to_host_byte_order.html @@ -0,0 +1,57 @@ + + +pcre_utf32_to_host_byte_order specification + + +

pcre_utf32_to_host_byte_order man page

+

+Return to the PCRE index page. +

+

+This page is part of the PCRE HTML documentation. It was generated automatically +from the original man page. If there is any nonsense in it, please consult the +man page, in case the conversion went wrong. +
+
+SYNOPSIS +
+

+#include <pcre.h> +

+

+int pcre32_utf32_to_host_byte_order(PCRE_UCHAR32 *output, +PCRE_SPTR32 input, int length, int *host_byte_order, +int keep_boms); +

+
+DESCRIPTION +
+

+This function, which exists only in the 32-bit library, converts a UTF-32 +string to the correct order for the current host, taking account of any byte +order marks (BOMs) within the string. Its arguments are: +

+  output           pointer to output buffer, may be the same as input
+  input            pointer to input buffer
+  length           number of 32-bit units in the input, or negative for
+                     a zero-terminated string
+  host_byte_order  a NULL value or a non-zero value pointed to means
+                     start in host byte order
+  keep_boms        if non-zero, BOMs are copied to the output string
+
+The result of the function is the number of 32-bit units placed into the output +buffer, including the zero terminator if the string was zero-terminated. +

+

+If host_byte_order is not NULL, it is set to indicate the byte order that +is current at the end of the string. +

+

+There is a complete description of the PCRE native API in the +pcreapi +page and a description of the POSIX API in the +pcreposix +page. +

+Return to the PCRE index page. +

diff --git a/pcre/doc/html/pcre_version.html b/pcre/doc/html/pcre_version.html new file mode 100644 index 00000000000..d33e7189559 --- /dev/null +++ b/pcre/doc/html/pcre_version.html @@ -0,0 +1,46 @@ + + +pcre_version specification + + +

pcre_version man page

+

+Return to the PCRE index page. +

+

+This page is part of the PCRE HTML documentation. It was generated automatically +from the original man page. If there is any nonsense in it, please consult the +man page, in case the conversion went wrong. +
+
+SYNOPSIS +
+

+#include <pcre.h> +

+

+const char *pcre_version(void); +

+

+const char *pcre16_version(void); +

+

+const char *pcre32_version(void); +

+
+DESCRIPTION +
+

+This function (even in the 16-bit and 32-bit libraries) returns a +zero-terminated, 8-bit character string that gives the version number of the +PCRE library and the date of its release. +

+

+There is a complete description of the PCRE native API in the +pcreapi +page and a description of the POSIX API in the +pcreposix +page. +

+Return to the PCRE index page. +

diff --git a/pcre/doc/html/pcreapi.html b/pcre/doc/html/pcreapi.html new file mode 100644 index 00000000000..2a00d9ba041 --- /dev/null +++ b/pcre/doc/html/pcreapi.html @@ -0,0 +1,2859 @@ + + +pcreapi specification + + +

pcreapi man page

+

+Return to the PCRE index page. +

+

+This page is part of the PCRE HTML documentation. It was generated automatically +from the original man page. If there is any nonsense in it, please consult the +man page, in case the conversion went wrong. +
+

+

+#include <pcre.h> +

+
PCRE NATIVE API BASIC FUNCTIONS
+

+pcre *pcre_compile(const char *pattern, int options, +const char **errptr, int *erroffset, +const unsigned char *tableptr); +

+

+pcre *pcre_compile2(const char *pattern, int options, +int *errorcodeptr, +const char **errptr, int *erroffset, +const unsigned char *tableptr); +

+

+pcre_extra *pcre_study(const pcre *code, int options, +const char **errptr); +

+

+void pcre_free_study(pcre_extra *extra); +

+

+int pcre_exec(const pcre *code, const pcre_extra *extra, +const char *subject, int length, int startoffset, +int options, int *ovector, int ovecsize); +

+

+int pcre_dfa_exec(const pcre *code, const pcre_extra *extra, +const char *subject, int length, int startoffset, +int options, int *ovector, int ovecsize, +int *workspace, int wscount); +

+
PCRE NATIVE API STRING EXTRACTION FUNCTIONS
+

+int pcre_copy_named_substring(const pcre *code, +const char *subject, int *ovector, +int stringcount, const char *stringname, +char *buffer, int buffersize); +

+

+int pcre_copy_substring(const char *subject, int *ovector, +int stringcount, int stringnumber, char *buffer, +int buffersize); +

+

+int pcre_get_named_substring(const pcre *code, +const char *subject, int *ovector, +int stringcount, const char *stringname, +const char **stringptr); +

+

+int pcre_get_stringnumber(const pcre *code, +const char *name); +

+

+int pcre_get_stringtable_entries(const pcre *code, +const char *name, char **first, char **last); +

+

+int pcre_get_substring(const char *subject, int *ovector, +int stringcount, int stringnumber, +const char **stringptr); +

+

+int pcre_get_substring_list(const char *subject, +int *ovector, int stringcount, const char ***listptr); +

+

+void pcre_free_substring(const char *stringptr); +

+

+void pcre_free_substring_list(const char **stringptr); +

+
PCRE NATIVE API AUXILIARY FUNCTIONS
+

+int pcre_jit_exec(const pcre *code, const pcre_extra *extra, +const char *subject, int length, int startoffset, +int options, int *ovector, int ovecsize, +pcre_jit_stack *jstack); +

+

+pcre_jit_stack *pcre_jit_stack_alloc(int startsize, int maxsize); +

+

+void pcre_jit_stack_free(pcre_jit_stack *stack); +

+

+void pcre_assign_jit_stack(pcre_extra *extra, +pcre_jit_callback callback, void *data); +

+

+const unsigned char *pcre_maketables(void); +

+

+int pcre_fullinfo(const pcre *code, const pcre_extra *extra, +int what, void *where); +

+

+int pcre_refcount(pcre *code, int adjust); +

+

+int pcre_config(int what, void *where); +

+

+const char *pcre_version(void); +

+

+int pcre_pattern_to_host_byte_order(pcre *code, +pcre_extra *extra, const unsigned char *tables); +

+
PCRE NATIVE API INDIRECTED FUNCTIONS
+

+void *(*pcre_malloc)(size_t); +

+

+void (*pcre_free)(void *); +

+

+void *(*pcre_stack_malloc)(size_t); +

+

+void (*pcre_stack_free)(void *); +

+

+int (*pcre_callout)(pcre_callout_block *); +

+
PCRE 8-BIT, 16-BIT, AND 32-BIT LIBRARIES
+

+As well as support for 8-bit character strings, PCRE also supports 16-bit +strings (from release 8.30) and 32-bit strings (from release 8.32), by means of +two additional libraries. They can be built as well as, or instead of, the +8-bit library. To avoid too much complication, this document describes the +8-bit versions of the functions, with only occasional references to the 16-bit +and 32-bit libraries. +

+

+The 16-bit and 32-bit functions operate in the same way as their 8-bit +counterparts; they just use different data types for their arguments and +results, and their names start with pcre16_ or pcre32_ instead of +pcre_. For every option that has UTF8 in its name (for example, +PCRE_UTF8), there are corresponding 16-bit and 32-bit names with UTF8 replaced +by UTF16 or UTF32, respectively. This facility is in fact just cosmetic; the +16-bit and 32-bit option names define the same bit values. +

+

+References to bytes and UTF-8 in this document should be read as references to +16-bit data units and UTF-16 when using the 16-bit library, or 32-bit data +units and UTF-32 when using the 32-bit library, unless specified otherwise. +More details of the specific differences for the 16-bit and 32-bit libraries +are given in the +pcre16 +and +pcre32 +pages. +

+
PCRE API OVERVIEW
+

+PCRE has its own native API, which is described in this document. There are +also some wrapper functions (for the 8-bit library only) that correspond to the +POSIX regular expression API, but they do not give access to all the +functionality. They are described in the +pcreposix +documentation. Both of these APIs define a set of C function calls. A C++ +wrapper (again for the 8-bit library only) is also distributed with PCRE. It is +documented in the +pcrecpp +page. +

+

+The native API C function prototypes are defined in the header file +pcre.h, and on Unix-like systems the (8-bit) library itself is called +libpcre. It can normally be accessed by adding -lpcre to the +command for linking an application that uses PCRE. The header file defines the +macros PCRE_MAJOR and PCRE_MINOR to contain the major and minor release numbers +for the library. Applications can use these to include support for different +releases of PCRE. +

+

+In a Windows environment, if you want to statically link an application program +against a non-dll pcre.a file, you must define PCRE_STATIC before +including pcre.h or pcrecpp.h, because otherwise the +pcre_malloc() and pcre_free() exported functions will be declared +__declspec(dllimport), with unwanted results. +

+

+The functions pcre_compile(), pcre_compile2(), pcre_study(), +and pcre_exec() are used for compiling and matching regular expressions +in a Perl-compatible manner. A sample program that demonstrates the simplest +way of using them is provided in the file called pcredemo.c in the PCRE +source distribution. A listing of this program is given in the +pcredemo +documentation, and the +pcresample +documentation describes how to compile and run it. +

+

+Just-in-time compiler support is an optional feature of PCRE that can be built +in appropriate hardware environments. It greatly speeds up the matching +performance of many patterns. Simple programs can easily request that it be +used if available, by setting an option that is ignored when it is not +relevant. More complicated programs might need to make use of the functions +pcre_jit_stack_alloc(), pcre_jit_stack_free(), and +pcre_assign_jit_stack() in order to control the JIT code's memory usage. +

+

+From release 8.32 there is also a direct interface for JIT execution, which +gives improved performance. The JIT-specific functions are discussed in the +pcrejit +documentation. +

+

+A second matching function, pcre_dfa_exec(), which is not +Perl-compatible, is also provided. This uses a different algorithm for the +matching. The alternative algorithm finds all possible matches (at a given +point in the subject), and scans the subject just once (unless there are +lookbehind assertions). However, this algorithm does not return captured +substrings. A description of the two matching algorithms and their advantages +and disadvantages is given in the +pcrematching +documentation. +

+

+In addition to the main compiling and matching functions, there are convenience +functions for extracting captured substrings from a subject string that is +matched by pcre_exec(). They are: +

+  pcre_copy_substring()
+  pcre_copy_named_substring()
+  pcre_get_substring()
+  pcre_get_named_substring()
+  pcre_get_substring_list()
+  pcre_get_stringnumber()
+  pcre_get_stringtable_entries()
+
+pcre_free_substring() and pcre_free_substring_list() are also +provided, to free the memory used for extracted strings. +

+

+The function pcre_maketables() is used to build a set of character tables +in the current locale for passing to pcre_compile(), pcre_exec(), +or pcre_dfa_exec(). This is an optional facility that is provided for +specialist use. Most commonly, no special tables are passed, in which case +internal tables that are generated when PCRE is built are used. +

+

+The function pcre_fullinfo() is used to find out information about a +compiled pattern. The function pcre_version() returns a pointer to a +string containing the version of PCRE and its date of release. +

+

+The function pcre_refcount() maintains a reference count in a data block +containing a compiled pattern. This is provided for the benefit of +object-oriented applications. +

+

+The global variables pcre_malloc and pcre_free initially contain +the entry points of the standard malloc() and free() functions, +respectively. PCRE calls the memory management functions via these variables, +so a calling program can replace them if it wishes to intercept the calls. This +should be done before calling any PCRE functions. +

+

+The global variables pcre_stack_malloc and pcre_stack_free are also +indirections to memory management functions. These special functions are used +only when PCRE is compiled to use the heap for remembering data, instead of +recursive function calls, when running the pcre_exec() function. See the +pcrebuild +documentation for details of how to do this. It is a non-standard way of +building PCRE, for use in environments that have limited stacks. Because of the +greater use of memory management, it runs more slowly. Separate functions are +provided so that special-purpose external code can be used for this case. When +used, these functions are always called in a stack-like manner (last obtained, +first freed), and always for memory blocks of the same size. There is a +discussion about PCRE's stack usage in the +pcrestack +documentation. +

+

+The global variable pcre_callout initially contains NULL. It can be set +by the caller to a "callout" function, which PCRE will then call at specified +points during a matching operation. Details are given in the +pcrecallout +documentation. +

+
NEWLINES
+

+PCRE supports five different conventions for indicating line breaks in +strings: a single CR (carriage return) character, a single LF (linefeed) +character, the two-character sequence CRLF, any of the three preceding, or any +Unicode newline sequence. The Unicode newline sequences are the three just +mentioned, plus the single characters VT (vertical tab, U+000B), FF (form feed, +U+000C), NEL (next line, U+0085), LS (line separator, U+2028), and PS +(paragraph separator, U+2029). +

+

+Each of the first three conventions is used by at least one operating system as +its standard newline sequence. When PCRE is built, a default can be specified. +The default default is LF, which is the Unix standard. When PCRE is run, the +default can be overridden, either when a pattern is compiled, or when it is +matched. +

+

+At compile time, the newline convention can be specified by the options +argument of pcre_compile(), or it can be specified by special text at the +start of the pattern itself; this overrides any other settings. See the +pcrepattern +page for details of the special character sequences. +

+

+In the PCRE documentation the word "newline" is used to mean "the character or +pair of characters that indicate a line break". The choice of newline +convention affects the handling of the dot, circumflex, and dollar +metacharacters, the handling of #-comments in /x mode, and, when CRLF is a +recognized line ending sequence, the match position advancement for a +non-anchored pattern. There is more detail about this in the +section on pcre_exec() options +below. +

+

+The choice of newline convention does not affect the interpretation of +the \n or \r escape sequences, nor does it affect what \R matches, which is +controlled in a similar way, but by separate options. +

+
MULTITHREADING
+

+The PCRE functions can be used in multi-threading applications, with the +proviso that the memory management functions pointed to by pcre_malloc, +pcre_free, pcre_stack_malloc, and pcre_stack_free, and the +callout function pointed to by pcre_callout, are shared by all threads. +

+

+The compiled form of a regular expression is not altered during matching, so +the same compiled pattern can safely be used by several threads at once. +

+

+If the just-in-time optimization feature is being used, it needs separate +memory stack areas for each thread. See the +pcrejit +documentation for more details. +

+
SAVING PRECOMPILED PATTERNS FOR LATER USE
+

+The compiled form of a regular expression can be saved and re-used at a later +time, possibly by a different program, and even on a host other than the one on +which it was compiled. Details are given in the +pcreprecompile +documentation, which includes a description of the +pcre_pattern_to_host_byte_order() function. However, compiling a regular +expression with one version of PCRE for use with a different version is not +guaranteed to work and may cause crashes. +

+
CHECKING BUILD-TIME OPTIONS
+

+int pcre_config(int what, void *where); +

+

+The function pcre_config() makes it possible for a PCRE client to +discover which optional features have been compiled into the PCRE library. The +pcrebuild +documentation has more details about these optional features. +

+

+The first argument for pcre_config() is an integer, specifying which +information is required; the second argument is a pointer to a variable into +which the information is placed. The returned value is zero on success, or the +negative error code PCRE_ERROR_BADOPTION if the value in the first argument is +not recognized. The following information is available: +

+  PCRE_CONFIG_UTF8
+
+The output is an integer that is set to one if UTF-8 support is available; +otherwise it is set to zero. This value should normally be given to the 8-bit +version of this function, pcre_config(). If it is given to the 16-bit +or 32-bit version of this function, the result is PCRE_ERROR_BADOPTION. +
+  PCRE_CONFIG_UTF16
+
+The output is an integer that is set to one if UTF-16 support is available; +otherwise it is set to zero. This value should normally be given to the 16-bit +version of this function, pcre16_config(). If it is given to the 8-bit +or 32-bit version of this function, the result is PCRE_ERROR_BADOPTION. +
+  PCRE_CONFIG_UTF32
+
+The output is an integer that is set to one if UTF-32 support is available; +otherwise it is set to zero. This value should normally be given to the 32-bit +version of this function, pcre32_config(). If it is given to the 8-bit +or 16-bit version of this function, the result is PCRE_ERROR_BADOPTION. +
+  PCRE_CONFIG_UNICODE_PROPERTIES
+
+The output is an integer that is set to one if support for Unicode character +properties is available; otherwise it is set to zero. +
+  PCRE_CONFIG_JIT
+
+The output is an integer that is set to one if support for just-in-time +compiling is available; otherwise it is set to zero. +
+  PCRE_CONFIG_JITTARGET
+
+The output is a pointer to a zero-terminated "const char *" string. If JIT +support is available, the string contains the name of the architecture for +which the JIT compiler is configured, for example "x86 32bit (little endian + +unaligned)". If JIT support is not available, the result is NULL. +
+  PCRE_CONFIG_NEWLINE
+
+The output is an integer whose value specifies the default character sequence +that is recognized as meaning "newline". The values that are supported in +ASCII/Unicode environments are: 10 for LF, 13 for CR, 3338 for CRLF, -2 for +ANYCRLF, and -1 for ANY. In EBCDIC environments, CR, ANYCRLF, and ANY yield the +same values. However, the value for LF is normally 21, though some EBCDIC +environments use 37. The corresponding values for CRLF are 3349 and 3365. The +default should normally correspond to the standard sequence for your operating +system. +
+  PCRE_CONFIG_BSR
+
+The output is an integer whose value indicates what character sequences the \R +escape sequence matches by default. A value of 0 means that \R matches any +Unicode line ending sequence; a value of 1 means that \R matches only CR, LF, +or CRLF. The default can be overridden when a pattern is compiled or matched. +
+  PCRE_CONFIG_LINK_SIZE
+
+The output is an integer that contains the number of bytes used for internal +linkage in compiled regular expressions. For the 8-bit library, the value can +be 2, 3, or 4. For the 16-bit library, the value is either 2 or 4 and is still +a number of bytes. For the 32-bit library, the value is either 2 or 4 and is +still a number of bytes. The default value of 2 is sufficient for all but the +most massive patterns, since it allows the compiled pattern to be up to 64K in +size. Larger values allow larger regular expressions to be compiled, at the +expense of slower matching. +
+  PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
+
+The output is an integer that contains the threshold above which the POSIX +interface uses malloc() for output vectors. Further details are given in +the +pcreposix +documentation. +
+  PCRE_CONFIG_MATCH_LIMIT
+
+The output is a long integer that gives the default limit for the number of +internal matching function calls in a pcre_exec() execution. Further +details are given with pcre_exec() below. +
+  PCRE_CONFIG_MATCH_LIMIT_RECURSION
+
+The output is a long integer that gives the default limit for the depth of +recursion when calling the internal matching function in a pcre_exec() +execution. Further details are given with pcre_exec() below. +
+  PCRE_CONFIG_STACKRECURSE
+
+The output is an integer that is set to one if internal recursion when running +pcre_exec() is implemented by recursive function calls that use the stack +to remember their state. This is the usual way that PCRE is compiled. The +output is zero if PCRE was compiled to use blocks of data on the heap instead +of recursive function calls. In this case, pcre_stack_malloc and +pcre_stack_free are called to manage memory blocks on the heap, thus +avoiding the use of the stack. +

+
COMPILING A PATTERN
+

+pcre *pcre_compile(const char *pattern, int options, +const char **errptr, int *erroffset, +const unsigned char *tableptr); +pcre *pcre_compile2(const char *pattern, int options, +int *errorcodeptr, +const char **errptr, int *erroffset, +const unsigned char *tableptr); +

+

+Either of the functions pcre_compile() or pcre_compile2() can be +called to compile a pattern into an internal form. The only difference between +the two interfaces is that pcre_compile2() has an additional argument, +errorcodeptr, via which a numerical error code can be returned. To avoid +too much repetition, we refer just to pcre_compile() below, but the +information applies equally to pcre_compile2(). +

+

+The pattern is a C string terminated by a binary zero, and is passed in the +pattern argument. A pointer to a single block of memory that is obtained +via pcre_malloc is returned. This contains the compiled code and related +data. The pcre type is defined for the returned block; this is a typedef +for a structure whose contents are not externally defined. It is up to the +caller to free the memory (via pcre_free) when it is no longer required. +

+

+Although the compiled code of a PCRE regex is relocatable, that is, it does not +depend on memory location, the complete pcre data block is not +fully relocatable, because it may contain a copy of the tableptr +argument, which is an address (see below). +

+

+The options argument contains various bit settings that affect the +compilation. It should be zero if no options are required. The available +options are described below. Some of them (in particular, those that are +compatible with Perl, but some others as well) can also be set and unset from +within the pattern (see the detailed description in the +pcrepattern +documentation). For those options that can be different in different parts of +the pattern, the contents of the options argument specifies their +settings at the start of compilation and execution. The PCRE_ANCHORED, +PCRE_BSR_xxx, PCRE_NEWLINE_xxx, PCRE_NO_UTF8_CHECK, and +PCRE_NO_START_OPTIMIZE options can be set at the time of matching as well as at +compile time. +

+

+If errptr is NULL, pcre_compile() returns NULL immediately. +Otherwise, if compilation of a pattern fails, pcre_compile() returns +NULL, and sets the variable pointed to by errptr to point to a textual +error message. This is a static string that is part of the library. You must +not try to free it. Normally, the offset from the start of the pattern to the +data unit that was being processed when the error was discovered is placed in +the variable pointed to by erroffset, which must not be NULL (if it is, +an immediate error is given). However, for an invalid UTF-8 or UTF-16 string, +the offset is that of the first data unit of the failing character. +

+

+Some errors are not detected until the whole pattern has been scanned; in these +cases, the offset passed back is the length of the pattern. Note that the +offset is in data units, not characters, even in a UTF mode. It may sometimes +point into the middle of a UTF-8 or UTF-16 character. +

+

+If pcre_compile2() is used instead of pcre_compile(), and the +errorcodeptr argument is not NULL, a non-zero error code number is +returned via this argument in the event of an error. This is in addition to the +textual error message. Error codes and messages are listed below. +

+

+If the final argument, tableptr, is NULL, PCRE uses a default set of +character tables that are built when PCRE is compiled, using the default C +locale. Otherwise, tableptr must be an address that is the result of a +call to pcre_maketables(). This value is stored with the compiled +pattern, and used again by pcre_exec(), unless another table pointer is +passed to it. For more discussion, see the section on locale support below. +

+

+This code fragment shows a typical straightforward call to pcre_compile(): +

+  pcre *re;
+  const char *error;
+  int erroffset;
+  re = pcre_compile(
+    "^A.*Z",          /* the pattern */
+    0,                /* default options */
+    &error,           /* for error message */
+    &erroffset,       /* for error offset */
+    NULL);            /* use default character tables */
+
+The following names for option bits are defined in the pcre.h header +file: +
+  PCRE_ANCHORED
+
+If this bit is set, the pattern is forced to be "anchored", that is, it is +constrained to match only at the first matching point in the string that is +being searched (the "subject string"). This effect can also be achieved by +appropriate constructs in the pattern itself, which is the only way to do it in +Perl. +
+  PCRE_AUTO_CALLOUT
+
+If this bit is set, pcre_compile() automatically inserts callout items, +all with number 255, before each pattern item. For discussion of the callout +facility, see the +pcrecallout +documentation. +
+  PCRE_BSR_ANYCRLF
+  PCRE_BSR_UNICODE
+
+These options (which are mutually exclusive) control what the \R escape +sequence matches. The choice is either to match only CR, LF, or CRLF, or to +match any Unicode newline sequence. The default is specified when PCRE is +built. It can be overridden from within the pattern, or by setting an option +when a compiled pattern is matched. +
+  PCRE_CASELESS
+
+If this bit is set, letters in the pattern match both upper and lower case +letters. It is equivalent to Perl's /i option, and it can be changed within a +pattern by a (?i) option setting. In UTF-8 mode, PCRE always understands the +concept of case for characters whose values are less than 128, so caseless +matching is always possible. For characters with higher values, the concept of +case is supported if PCRE is compiled with Unicode property support, but not +otherwise. If you want to use caseless matching for characters 128 and above, +you must ensure that PCRE is compiled with Unicode property support as well as +with UTF-8 support. +
+  PCRE_DOLLAR_ENDONLY
+
+If this bit is set, a dollar metacharacter in the pattern matches only at the +end of the subject string. Without this option, a dollar also matches +immediately before a newline at the end of the string (but not before any other +newlines). The PCRE_DOLLAR_ENDONLY option is ignored if PCRE_MULTILINE is set. +There is no equivalent to this option in Perl, and no way to set it within a +pattern. +
+  PCRE_DOTALL
+
+If this bit is set, a dot metacharacter in the pattern matches a character of +any value, including one that indicates a newline. However, it only ever +matches one character, even if newlines are coded as CRLF. Without this option, +a dot does not match when the current position is at a newline. This option is +equivalent to Perl's /s option, and it can be changed within a pattern by a +(?s) option setting. A negative class such as [^a] always matches newline +characters, independent of the setting of this option. +
+  PCRE_DUPNAMES
+
+If this bit is set, names used to identify capturing subpatterns need not be +unique. This can be helpful for certain types of pattern when it is known that +only one instance of the named subpattern can ever be matched. There are more +details of named subpatterns below; see also the +pcrepattern +documentation. +
+  PCRE_EXTENDED
+
+If this bit is set, white space data characters in the pattern are totally +ignored except when escaped or inside a character class. White space does not +include the VT character (code 11). In addition, characters between an +unescaped # outside a character class and the next newline, inclusive, are also +ignored. This is equivalent to Perl's /x option, and it can be changed within a +pattern by a (?x) option setting. +

+

+Which characters are interpreted as newlines is controlled by the options +passed to pcre_compile() or by a special sequence at the start of the +pattern, as described in the section entitled +"Newline conventions" +in the pcrepattern documentation. Note that the end of this type of +comment is a literal newline sequence in the pattern; escape sequences that +happen to represent a newline do not count. +

+

+This option makes it possible to include comments inside complicated patterns. +Note, however, that this applies only to data characters. White space characters +may never appear within special character sequences in a pattern, for example +within the sequence (?( that introduces a conditional subpattern. +

+  PCRE_EXTRA
+
+This option was invented in order to turn on additional functionality of PCRE +that is incompatible with Perl, but it is currently of very little use. When +set, any backslash in a pattern that is followed by a letter that has no +special meaning causes an error, thus reserving these combinations for future +expansion. By default, as in Perl, a backslash followed by a letter with no +special meaning is treated as a literal. (Perl can, however, be persuaded to +give an error for this, by running it with the -w option.) There are at present +no other features controlled by this option. It can also be set by a (?X) +option setting within a pattern. +
+  PCRE_FIRSTLINE
+
+If this option is set, an unanchored pattern is required to match before or at +the first newline in the subject string, though the matched text may continue +over the newline. +
+  PCRE_JAVASCRIPT_COMPAT
+
+If this option is set, PCRE's behaviour is changed in some ways so that it is +compatible with JavaScript rather than Perl. The changes are as follows: +

+

+(1) A lone closing square bracket in a pattern causes a compile-time error, +because this is illegal in JavaScript (by default it is treated as a data +character). Thus, the pattern AB]CD becomes illegal when this option is set. +

+

+(2) At run time, a back reference to an unset subpattern group matches an empty +string (by default this causes the current matching alternative to fail). A +pattern such as (\1)(a) succeeds when this option is set (assuming it can find +an "a" in the subject), whereas it fails by default, for Perl compatibility. +

+

+(3) \U matches an upper case "U" character; by default \U causes a compile +time error (Perl uses \U to upper case subsequent characters). +

+

+(4) \u matches a lower case "u" character unless it is followed by four +hexadecimal digits, in which case the hexadecimal number defines the code point +to match. By default, \u causes a compile time error (Perl uses it to upper +case the following character). +

+

+(5) \x matches a lower case "x" character unless it is followed by two +hexadecimal digits, in which case the hexadecimal number defines the code point +to match. By default, as in Perl, a hexadecimal number is always expected after +\x, but it may have zero, one, or two digits (so, for example, \xz matches a +binary zero character followed by z). +

+  PCRE_MULTILINE
+
+By default, for the purposes of matching "start of line" and "end of line", +PCRE treats the subject string as consisting of a single line of characters, +even if it actually contains newlines. The "start of line" metacharacter (^) +matches only at the start of the string, and the "end of line" metacharacter +($) matches only at the end of the string, or before a terminating newline +(except when PCRE_DOLLAR_ENDONLY is set). Note, however, that unless +PCRE_DOTALL is set, the "any character" metacharacter (.) does not match at a +newline. This behaviour (for ^, $, and dot) is the same as Perl. +

+

+When PCRE_MULTILINE it is set, the "start of line" and "end of line" constructs +match immediately following or immediately before internal newlines in the +subject string, respectively, as well as at the very start and end. This is +equivalent to Perl's /m option, and it can be changed within a pattern by a +(?m) option setting. If there are no newlines in a subject string, or no +occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect. +

+  PCRE_NEVER_UTF
+
+This option locks out interpretation of the pattern as UTF-8 (or UTF-16 or +UTF-32 in the 16-bit and 32-bit libraries). In particular, it prevents the +creator of the pattern from switching to UTF interpretation by starting the +pattern with (*UTF). This may be useful in applications that process patterns +from external sources. The combination of PCRE_UTF8 and PCRE_NEVER_UTF also +causes an error. +
+  PCRE_NEWLINE_CR
+  PCRE_NEWLINE_LF
+  PCRE_NEWLINE_CRLF
+  PCRE_NEWLINE_ANYCRLF
+  PCRE_NEWLINE_ANY
+
+These options override the default newline definition that was chosen when PCRE +was built. Setting the first or the second specifies that a newline is +indicated by a single character (CR or LF, respectively). Setting +PCRE_NEWLINE_CRLF specifies that a newline is indicated by the two-character +CRLF sequence. Setting PCRE_NEWLINE_ANYCRLF specifies that any of the three +preceding sequences should be recognized. Setting PCRE_NEWLINE_ANY specifies +that any Unicode newline sequence should be recognized. +

+

+In an ASCII/Unicode environment, the Unicode newline sequences are the three +just mentioned, plus the single characters VT (vertical tab, U+000B), FF (form +feed, U+000C), NEL (next line, U+0085), LS (line separator, U+2028), and PS +(paragraph separator, U+2029). For the 8-bit library, the last two are +recognized only in UTF-8 mode. +

+

+When PCRE is compiled to run in an EBCDIC (mainframe) environment, the code for +CR is 0x0d, the same as ASCII. However, the character code for LF is normally +0x15, though in some EBCDIC environments 0x25 is used. Whichever of these is +not LF is made to correspond to Unicode's NEL character. EBCDIC codes are all +less than 256. For more details, see the +pcrebuild +documentation. +

+

+The newline setting in the options word uses three bits that are treated +as a number, giving eight possibilities. Currently only six are used (default +plus the five values above). This means that if you set more than one newline +option, the combination may or may not be sensible. For example, +PCRE_NEWLINE_CR with PCRE_NEWLINE_LF is equivalent to PCRE_NEWLINE_CRLF, but +other combinations may yield unused numbers and cause an error. +

+

+The only time that a line break in a pattern is specially recognized when +compiling is when PCRE_EXTENDED is set. CR and LF are white space characters, +and so are ignored in this mode. Also, an unescaped # outside a character class +indicates a comment that lasts until after the next line break sequence. In +other circumstances, line break sequences in patterns are treated as literal +data. +

+

+The newline option that is set at compile time becomes the default that is used +for pcre_exec() and pcre_dfa_exec(), but it can be overridden. +

+  PCRE_NO_AUTO_CAPTURE
+
+If this option is set, it disables the use of numbered capturing parentheses in +the pattern. Any opening parenthesis that is not followed by ? behaves as if it +were followed by ?: but named parentheses can still be used for capturing (and +they acquire numbers in the usual way). There is no equivalent of this option +in Perl. +
+  PCRE_NO_START_OPTIMIZE
+
+This is an option that acts at matching time; that is, it is really an option +for pcre_exec() or pcre_dfa_exec(). If it is set at compile time, +it is remembered with the compiled pattern and assumed at matching time. This +is necessary if you want to use JIT execution, because the JIT compiler needs +to know whether or not this option is set. For details see the discussion of +PCRE_NO_START_OPTIMIZE +below. +
+  PCRE_UCP
+
+This option changes the way PCRE processes \B, \b, \D, \d, \S, \s, \W, +\w, and some of the POSIX character classes. By default, only ASCII characters +are recognized, but if PCRE_UCP is set, Unicode properties are used instead to +classify characters. More details are given in the section on +generic character types +in the +pcrepattern +page. If you set PCRE_UCP, matching one of the items it affects takes much +longer. The option is available only if PCRE has been compiled with Unicode +property support. +
+  PCRE_UNGREEDY
+
+This option inverts the "greediness" of the quantifiers so that they are not +greedy by default, but become greedy if followed by "?". It is not compatible +with Perl. It can also be set by a (?U) option setting within the pattern. +
+  PCRE_UTF8
+
+This option causes PCRE to regard both the pattern and the subject as strings +of UTF-8 characters instead of single-byte strings. However, it is available +only when PCRE is built to include UTF support. If not, the use of this option +provokes an error. Details of how this option changes the behaviour of PCRE are +given in the +pcreunicode +page. +
+  PCRE_NO_UTF8_CHECK
+
+When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is +automatically checked. There is a discussion about the +validity of UTF-8 strings +in the +pcreunicode +page. If an invalid UTF-8 sequence is found, pcre_compile() returns an +error. If you already know that your pattern is valid, and you want to skip +this check for performance reasons, you can set the PCRE_NO_UTF8_CHECK option. +When it is set, the effect of passing an invalid UTF-8 string as a pattern is +undefined. It may cause your program to crash. Note that this option can also +be passed to pcre_exec() and pcre_dfa_exec(), to suppress the +validity checking of subject strings only. If the same string is being matched +many times, the option can be safely set for the second and subsequent +matchings to improve performance. +

+
COMPILATION ERROR CODES
+

+The following table lists the error codes than may be returned by +pcre_compile2(), along with the error messages that may be returned by +both compiling functions. Note that error messages are always 8-bit ASCII +strings, even in 16-bit or 32-bit mode. As PCRE has developed, some error codes +have fallen out of use. To avoid confusion, they have not been re-used. +

+   0  no error
+   1  \ at end of pattern
+   2  \c at end of pattern
+   3  unrecognized character follows \
+   4  numbers out of order in {} quantifier
+   5  number too big in {} quantifier
+   6  missing terminating ] for character class
+   7  invalid escape sequence in character class
+   8  range out of order in character class
+   9  nothing to repeat
+  10  [this code is not in use]
+  11  internal error: unexpected repeat
+  12  unrecognized character after (? or (?-
+  13  POSIX named classes are supported only within a class
+  14  missing )
+  15  reference to non-existent subpattern
+  16  erroffset passed as NULL
+  17  unknown option bit(s) set
+  18  missing ) after comment
+  19  [this code is not in use]
+  20  regular expression is too large
+  21  failed to get memory
+  22  unmatched parentheses
+  23  internal error: code overflow
+  24  unrecognized character after (?<
+  25  lookbehind assertion is not fixed length
+  26  malformed number or name after (?(
+  27  conditional group contains more than two branches
+  28  assertion expected after (?(
+  29  (?R or (?[+-]digits must be followed by )
+  30  unknown POSIX class name
+  31  POSIX collating elements are not supported
+  32  this version of PCRE is compiled without UTF support
+  33  [this code is not in use]
+  34  character value in \x{...} sequence is too large
+  35  invalid condition (?(0)
+  36  \C not allowed in lookbehind assertion
+  37  PCRE does not support \L, \l, \N{name}, \U, or \u
+  38  number after (?C is > 255
+  39  closing ) for (?C expected
+  40  recursive call could loop indefinitely
+  41  unrecognized character after (?P
+  42  syntax error in subpattern name (missing terminator)
+  43  two named subpatterns have the same name
+  44  invalid UTF-8 string (specifically UTF-8)
+  45  support for \P, \p, and \X has not been compiled
+  46  malformed \P or \p sequence
+  47  unknown property name after \P or \p
+  48  subpattern name is too long (maximum 32 characters)
+  49  too many named subpatterns (maximum 10000)
+  50  [this code is not in use]
+  51  octal value is greater than \377 in 8-bit non-UTF-8 mode
+  52  internal error: overran compiling workspace
+  53  internal error: previously-checked referenced subpattern
+        not found
+  54  DEFINE group contains more than one branch
+  55  repeating a DEFINE group is not allowed
+  56  inconsistent NEWLINE options
+  57  \g is not followed by a braced, angle-bracketed, or quoted
+        name/number or by a plain number
+  58  a numbered reference must not be zero
+  59  an argument is not allowed for (*ACCEPT), (*FAIL), or (*COMMIT)
+  60  (*VERB) not recognized or malformed
+  61  number is too big
+  62  subpattern name expected
+  63  digit expected after (?+
+  64  ] is an invalid data character in JavaScript compatibility mode
+  65  different names for subpatterns of the same number are
+        not allowed
+  66  (*MARK) must have an argument
+  67  this version of PCRE is not compiled with Unicode property
+        support
+  68  \c must be followed by an ASCII character
+  69  \k is not followed by a braced, angle-bracketed, or quoted name
+  70  internal error: unknown opcode in find_fixedlength()
+  71  \N is not supported in a class
+  72  too many forward references
+  73  disallowed Unicode code point (>= 0xd800 && <= 0xdfff)
+  74  invalid UTF-16 string (specifically UTF-16)
+  75  name is too long in (*MARK), (*PRUNE), (*SKIP), or (*THEN)
+  76  character value in \u.... sequence is too large
+  77  invalid UTF-32 string (specifically UTF-32)
+
+The numbers 32 and 10000 in errors 48 and 49 are defaults; different values may +be used if the limits were changed when PCRE was built. +

+
STUDYING A PATTERN
+

+pcre_extra *pcre_study(const pcre *code, int options +const char **errptr); +

+

+If a compiled pattern is going to be used several times, it is worth spending +more time analyzing it in order to speed up the time taken for matching. The +function pcre_study() takes a pointer to a compiled pattern as its first +argument. If studying the pattern produces additional information that will +help speed up matching, pcre_study() returns a pointer to a +pcre_extra block, in which the study_data field points to the +results of the study. +

+

+The returned value from pcre_study() can be passed directly to +pcre_exec() or pcre_dfa_exec(). However, a pcre_extra block +also contains other fields that can be set by the caller before the block is +passed; these are described +below +in the section on matching a pattern. +

+

+If studying the pattern does not produce any useful information, +pcre_study() returns NULL by default. In that circumstance, if the +calling program wants to pass any of the other fields to pcre_exec() or +pcre_dfa_exec(), it must set up its own pcre_extra block. However, +if pcre_study() is called with the PCRE_STUDY_EXTRA_NEEDED option, it +returns a pcre_extra block even if studying did not find any additional +information. It may still return NULL, however, if an error occurs in +pcre_study(). +

+

+The second argument of pcre_study() contains option bits. There are three +further options in addition to PCRE_STUDY_EXTRA_NEEDED: +

+  PCRE_STUDY_JIT_COMPILE
+  PCRE_STUDY_JIT_PARTIAL_HARD_COMPILE
+  PCRE_STUDY_JIT_PARTIAL_SOFT_COMPILE
+
+If any of these are set, and the just-in-time compiler is available, the +pattern is further compiled into machine code that executes much faster than +the pcre_exec() interpretive matching function. If the just-in-time +compiler is not available, these options are ignored. All undefined bits in the +options argument must be zero. +

+

+JIT compilation is a heavyweight optimization. It can take some time for +patterns to be analyzed, and for one-off matches and simple patterns the +benefit of faster execution might be offset by a much slower study time. +Not all patterns can be optimized by the JIT compiler. For those that cannot be +handled, matching automatically falls back to the pcre_exec() +interpreter. For more details, see the +pcrejit +documentation. +

+

+The third argument for pcre_study() is a pointer for an error message. If +studying succeeds (even if no data is returned), the variable it points to is +set to NULL. Otherwise it is set to point to a textual error message. This is a +static string that is part of the library. You must not try to free it. You +should test the error pointer for NULL after calling pcre_study(), to be +sure that it has run successfully. +

+

+When you are finished with a pattern, you can free the memory used for the +study data by calling pcre_free_study(). This function was added to the +API for release 8.20. For earlier versions, the memory could be freed with +pcre_free(), just like the pattern itself. This will still work in cases +where JIT optimization is not used, but it is advisable to change to the new +function when convenient. +

+

+This is a typical way in which pcre_study() is used (except that in a +real application there should be tests for errors): +

+  int rc;
+  pcre *re;
+  pcre_extra *sd;
+  re = pcre_compile("pattern", 0, &error, &erroroffset, NULL);
+  sd = pcre_study(
+    re,             /* result of pcre_compile() */
+    0,              /* no options */
+    &error);        /* set to NULL or points to a message */
+  rc = pcre_exec(   /* see below for details of pcre_exec() options */
+    re, sd, "subject", 7, 0, 0, ovector, 30);
+  ...
+  pcre_free_study(sd);
+  pcre_free(re);
+
+Studying a pattern does two things: first, a lower bound for the length of +subject string that is needed to match the pattern is computed. This does not +mean that there are any strings of that length that match, but it does +guarantee that no shorter strings match. The value is used to avoid wasting +time by trying to match strings that are shorter than the lower bound. You can +find out the value in a calling program via the pcre_fullinfo() function. +

+

+Studying a pattern is also useful for non-anchored patterns that do not have a +single fixed starting character. A bitmap of possible starting bytes is +created. This speeds up finding a position in the subject at which to start +matching. (In 16-bit mode, the bitmap is used for 16-bit values less than 256. +In 32-bit mode, the bitmap is used for 32-bit values less than 256.) +

+

+These two optimizations apply to both pcre_exec() and +pcre_dfa_exec(), and the information is also used by the JIT compiler. +The optimizations can be disabled by setting the PCRE_NO_START_OPTIMIZE option. +You might want to do this if your pattern contains callouts or (*MARK) and you +want to make use of these facilities in cases where matching fails. +

+

+PCRE_NO_START_OPTIMIZE can be specified at either compile time or execution +time. However, if PCRE_NO_START_OPTIMIZE is passed to pcre_exec(), (that +is, after any JIT compilation has happened) JIT execution is disabled. For JIT +execution to work with PCRE_NO_START_OPTIMIZE, the option must be set at +compile time. +

+

+There is a longer discussion of PCRE_NO_START_OPTIMIZE +below. +

+
LOCALE SUPPORT
+

+PCRE handles caseless matching, and determines whether characters are letters, +digits, or whatever, by reference to a set of tables, indexed by character +value. When running in UTF-8 mode, this applies only to characters +with codes less than 128. By default, higher-valued codes never match escapes +such as \w or \d, but they can be tested with \p if PCRE is built with +Unicode character property support. Alternatively, the PCRE_UCP option can be +set at compile time; this causes \w and friends to use Unicode property +support instead of built-in tables. The use of locales with Unicode is +discouraged. If you are handling characters with codes greater than 128, you +should either use UTF-8 and Unicode, or use locales, but not try to mix the +two. +

+

+PCRE contains an internal set of tables that are used when the final argument +of pcre_compile() is NULL. These are sufficient for many applications. +Normally, the internal tables recognize only ASCII characters. However, when +PCRE is built, it is possible to cause the internal tables to be rebuilt in the +default "C" locale of the local system, which may cause them to be different. +

+

+The internal tables can always be overridden by tables supplied by the +application that calls PCRE. These may be created in a different locale from +the default. As more and more applications change to using Unicode, the need +for this locale support is expected to die away. +

+

+External tables are built by calling the pcre_maketables() function, +which has no arguments, in the relevant locale. The result can then be passed +to pcre_compile() or pcre_exec() as often as necessary. For +example, to build and use tables that are appropriate for the French locale +(where accented characters with values greater than 128 are treated as letters), +the following code could be used: +

+  setlocale(LC_CTYPE, "fr_FR");
+  tables = pcre_maketables();
+  re = pcre_compile(..., tables);
+
+The locale name "fr_FR" is used on Linux and other Unix-like systems; if you +are using Windows, the name for the French locale is "french". +

+

+When pcre_maketables() runs, the tables are built in memory that is +obtained via pcre_malloc. It is the caller's responsibility to ensure +that the memory containing the tables remains available for as long as it is +needed. +

+

+The pointer that is passed to pcre_compile() is saved with the compiled +pattern, and the same tables are used via this pointer by pcre_study() +and normally also by pcre_exec(). Thus, by default, for any single +pattern, compilation, studying and matching all happen in the same locale, but +different patterns can be compiled in different locales. +

+

+It is possible to pass a table pointer or NULL (indicating the use of the +internal tables) to pcre_exec(). Although not intended for this purpose, +this facility could be used to match a pattern in a different locale from the +one in which it was compiled. Passing table pointers at run time is discussed +below in the section on matching a pattern. +

+
INFORMATION ABOUT A PATTERN
+

+int pcre_fullinfo(const pcre *code, const pcre_extra *extra, +int what, void *where); +

+

+The pcre_fullinfo() function returns information about a compiled +pattern. It replaces the pcre_info() function, which was removed from the +library at version 8.30, after more than 10 years of obsolescence. +

+

+The first argument for pcre_fullinfo() is a pointer to the compiled +pattern. The second argument is the result of pcre_study(), or NULL if +the pattern was not studied. The third argument specifies which piece of +information is required, and the fourth argument is a pointer to a variable +to receive the data. The yield of the function is zero for success, or one of +the following negative numbers: +

+  PCRE_ERROR_NULL           the argument code was NULL
+                            the argument where was NULL
+  PCRE_ERROR_BADMAGIC       the "magic number" was not found
+  PCRE_ERROR_BADENDIANNESS  the pattern was compiled with different
+                            endianness
+  PCRE_ERROR_BADOPTION      the value of what was invalid
+  PCRE_ERROR_UNSET          the requested field is not set
+
+The "magic number" is placed at the start of each compiled pattern as an simple +check against passing an arbitrary memory pointer. The endianness error can +occur if a compiled pattern is saved and reloaded on a different host. Here is +a typical call of pcre_fullinfo(), to obtain the length of the compiled +pattern: +
+  int rc;
+  size_t length;
+  rc = pcre_fullinfo(
+    re,               /* result of pcre_compile() */
+    sd,               /* result of pcre_study(), or NULL */
+    PCRE_INFO_SIZE,   /* what is required */
+    &length);         /* where to put the data */
+
+The possible values for the third argument are defined in pcre.h, and are +as follows: +
+  PCRE_INFO_BACKREFMAX
+
+Return the number of the highest back reference in the pattern. The fourth +argument should point to an int variable. Zero is returned if there are +no back references. +
+  PCRE_INFO_CAPTURECOUNT
+
+Return the number of capturing subpatterns in the pattern. The fourth argument +should point to an int variable. +
+  PCRE_INFO_DEFAULT_TABLES
+
+Return a pointer to the internal default character tables within PCRE. The +fourth argument should point to an unsigned char * variable. This +information call is provided for internal use by the pcre_study() +function. External callers can cause PCRE to use its internal tables by passing +a NULL table pointer. +
+  PCRE_INFO_FIRSTBYTE
+
+Return information about the first data unit of any matched string, for a +non-anchored pattern. (The name of this option refers to the 8-bit library, +where data units are bytes.) The fourth argument should point to an int +variable. +

+

+If there is a fixed first value, for example, the letter "c" from a pattern +such as (cat|cow|coyote), its value is returned. In the 8-bit library, the +value is always less than 256. In the 16-bit library the value can be up to +0xffff. In the 32-bit library the value can be up to 0x10ffff. +

+

+If there is no fixed first value, and if either +
+
+(a) the pattern was compiled with the PCRE_MULTILINE option, and every branch +starts with "^", or +
+
+(b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not set +(if it were set, the pattern would be anchored), +
+
+-1 is returned, indicating that the pattern matches only at the start of a +subject string or after any newline within the string. Otherwise -2 is +returned. For anchored patterns, -2 is returned. +

+

+Since for the 32-bit library using the non-UTF-32 mode, this function is unable +to return the full 32-bit range of the character, this value is deprecated; +instead the PCRE_INFO_FIRSTCHARACTERFLAGS and PCRE_INFO_FIRSTCHARACTER values +should be used. +

+  PCRE_INFO_FIRSTTABLE
+
+If the pattern was studied, and this resulted in the construction of a 256-bit +table indicating a fixed set of values for the first data unit in any matching +string, a pointer to the table is returned. Otherwise NULL is returned. The +fourth argument should point to an unsigned char * variable. +
+  PCRE_INFO_HASCRORLF
+
+Return 1 if the pattern contains any explicit matches for CR or LF characters, +otherwise 0. The fourth argument should point to an int variable. An +explicit match is either a literal CR or LF character, or \r or \n. +
+  PCRE_INFO_JCHANGED
+
+Return 1 if the (?J) or (?-J) option setting is used in the pattern, otherwise +0. The fourth argument should point to an int variable. (?J) and +(?-J) set and unset the local PCRE_DUPNAMES option, respectively. +
+  PCRE_INFO_JIT
+
+Return 1 if the pattern was studied with one of the JIT options, and +just-in-time compiling was successful. The fourth argument should point to an +int variable. A return value of 0 means that JIT support is not available +in this version of PCRE, or that the pattern was not studied with a JIT option, +or that the JIT compiler could not handle this particular pattern. See the +pcrejit +documentation for details of what can and cannot be handled. +
+  PCRE_INFO_JITSIZE
+
+If the pattern was successfully studied with a JIT option, return the size of +the JIT compiled code, otherwise return zero. The fourth argument should point +to a size_t variable. +
+  PCRE_INFO_LASTLITERAL
+
+Return the value of the rightmost literal data unit that must exist in any +matched string, other than at its start, if such a value has been recorded. The +fourth argument should point to an int variable. If there is no such +value, -1 is returned. For anchored patterns, a last literal value is recorded +only if it follows something of variable length. For example, for the pattern +/^a\d+z\d+/ the returned value is "z", but for /^a\dz\d/ the returned value +is -1. +

+

+Since for the 32-bit library using the non-UTF-32 mode, this function is unable +to return the full 32-bit range of the character, this value is deprecated; +instead the PCRE_INFO_REQUIREDCHARFLAGS and PCRE_INFO_REQUIREDCHAR values should +be used. +

+  PCRE_INFO_MATCHLIMIT
+
+If the pattern set a match limit by including an item of the form +(*LIMIT_MATCH=nnnn) at the start, the value is returned. The fourth argument +should point to an unsigned 32-bit integer. If no such value has been set, the +call to pcre_fullinfo() returns the error PCRE_ERROR_UNSET. +
+  PCRE_INFO_MAXLOOKBEHIND
+
+Return the number of characters (NB not data units) in the longest lookbehind +assertion in the pattern. This information is useful when doing multi-segment +matching using the partial matching facilities. Note that the simple assertions +\b and \B require a one-character lookbehind. \A also registers a +one-character lookbehind, though it does not actually inspect the previous +character. This is to ensure that at least one character from the old segment +is retained when a new segment is processed. Otherwise, if there are no +lookbehinds in the pattern, \A might match incorrectly at the start of a new +segment. +
+  PCRE_INFO_MINLENGTH
+
+If the pattern was studied and a minimum length for matching subject strings +was computed, its value is returned. Otherwise the returned value is -1. The +value is a number of characters, which in UTF mode may be different from the +number of data units. The fourth argument should point to an int +variable. A non-negative value is a lower bound to the length of any matching +string. There may not be any strings of that length that do actually match, but +every string that does match is at least that long. +
+  PCRE_INFO_NAMECOUNT
+  PCRE_INFO_NAMEENTRYSIZE
+  PCRE_INFO_NAMETABLE
+
+PCRE supports the use of named as well as numbered capturing parentheses. The +names are just an additional way of identifying the parentheses, which still +acquire numbers. Several convenience functions such as +pcre_get_named_substring() are provided for extracting captured +substrings by name. It is also possible to extract the data directly, by first +converting the name to a number in order to access the correct pointers in the +output vector (described with pcre_exec() below). To do the conversion, +you need to use the name-to-number map, which is described by these three +values. +

+

+The map consists of a number of fixed-size entries. PCRE_INFO_NAMECOUNT gives +the number of entries, and PCRE_INFO_NAMEENTRYSIZE gives the size of each +entry; both of these return an int value. The entry size depends on the +length of the longest name. PCRE_INFO_NAMETABLE returns a pointer to the first +entry of the table. This is a pointer to char in the 8-bit library, where +the first two bytes of each entry are the number of the capturing parenthesis, +most significant byte first. In the 16-bit library, the pointer points to +16-bit data units, the first of which contains the parenthesis number. In the +32-bit library, the pointer points to 32-bit data units, the first of which +contains the parenthesis number. The rest of the entry is the corresponding +name, zero terminated. +

+

+The names are in alphabetical order. Duplicate names may appear if (?| is used +to create multiple groups with the same number, as described in the +section on duplicate subpattern numbers +in the +pcrepattern +page. Duplicate names for subpatterns with different numbers are permitted only +if PCRE_DUPNAMES is set. In all cases of duplicate names, they appear in the +table in the order in which they were found in the pattern. In the absence of +(?| this is the order of increasing number; when (?| is used this is not +necessarily the case because later subpatterns may have lower numbers. +

+

+As a simple example of the name/number table, consider the following pattern +after compilation by the 8-bit library (assume PCRE_EXTENDED is set, so white +space - including newlines - is ignored): +

+  (?<date> (?<year>(\d\d)?\d\d) - (?<month>\d\d) - (?<day>\d\d) )
+
+There are four named subpatterns, so the table has four entries, and each entry +in the table is eight bytes long. The table is as follows, with non-printing +bytes shows in hexadecimal, and undefined bytes shown as ??: +
+  00 01 d  a  t  e  00 ??
+  00 05 d  a  y  00 ?? ??
+  00 04 m  o  n  t  h  00
+  00 02 y  e  a  r  00 ??
+
+When writing code to extract data from named subpatterns using the +name-to-number map, remember that the length of the entries is likely to be +different for each compiled pattern. +
+  PCRE_INFO_OKPARTIAL
+
+Return 1 if the pattern can be used for partial matching with +pcre_exec(), otherwise 0. The fourth argument should point to an +int variable. From release 8.00, this always returns 1, because the +restrictions that previously applied to partial matching have been lifted. The +pcrepartial +documentation gives details of partial matching. +
+  PCRE_INFO_OPTIONS
+
+Return a copy of the options with which the pattern was compiled. The fourth +argument should point to an unsigned long int variable. These option bits +are those specified in the call to pcre_compile(), modified by any +top-level option settings at the start of the pattern itself. In other words, +they are the options that will be in force when matching starts. For example, +if the pattern /(?im)abc(?-i)d/ is compiled with the PCRE_EXTENDED option, the +result is PCRE_CASELESS, PCRE_MULTILINE, and PCRE_EXTENDED. +

+

+A pattern is automatically anchored by PCRE if all of its top-level +alternatives begin with one of the following: +

+  ^     unless PCRE_MULTILINE is set
+  \A    always
+  \G    always
+  .*    if PCRE_DOTALL is set and there are no back references to the subpattern in which .* appears
+
+For such patterns, the PCRE_ANCHORED bit is set in the options returned by +pcre_fullinfo(). +
+  PCRE_INFO_RECURSIONLIMIT
+
+If the pattern set a recursion limit by including an item of the form +(*LIMIT_RECURSION=nnnn) at the start, the value is returned. The fourth +argument should point to an unsigned 32-bit integer. If no such value has been +set, the call to pcre_fullinfo() returns the error PCRE_ERROR_UNSET. +
+  PCRE_INFO_SIZE
+
+Return the size of the compiled pattern in bytes (for all three libraries). The +fourth argument should point to a size_t variable. This value does not +include the size of the pcre structure that is returned by +pcre_compile(). The value that is passed as the argument to +pcre_malloc() when pcre_compile() is getting memory in which to +place the compiled data is the value returned by this option plus the size of +the pcre structure. Studying a compiled pattern, with or without JIT, +does not alter the value returned by this option. +
+  PCRE_INFO_STUDYSIZE
+
+Return the size in bytes (for all three libraries) of the data block pointed to +by the study_data field in a pcre_extra block. If pcre_extra +is NULL, or there is no study data, zero is returned. The fourth argument +should point to a size_t variable. The study_data field is set by +pcre_study() to record information that will speed up matching (see the +section entitled +"Studying a pattern" +above). The format of the study_data block is private, but its length +is made available via this option so that it can be saved and restored (see the +pcreprecompile +documentation for details). +
+  PCRE_INFO_FIRSTCHARACTERFLAGS
+
+Return information about the first data unit of any matched string, for a +non-anchored pattern. The fourth argument should point to an int +variable. +

+

+If there is a fixed first value, for example, the letter "c" from a pattern +such as (cat|cow|coyote), 1 is returned, and the character value can be +retrieved using PCRE_INFO_FIRSTCHARACTER. +

+

+If there is no fixed first value, and if either +
+
+(a) the pattern was compiled with the PCRE_MULTILINE option, and every branch +starts with "^", or +
+
+(b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not set +(if it were set, the pattern would be anchored), +
+
+2 is returned, indicating that the pattern matches only at the start of a +subject string or after any newline within the string. Otherwise 0 is +returned. For anchored patterns, 0 is returned. +

+  PCRE_INFO_FIRSTCHARACTER
+
+Return the fixed first character value, if PCRE_INFO_FIRSTCHARACTERFLAGS +returned 1; otherwise returns 0. The fourth argument should point to an +uint_t variable. +

+

+In the 8-bit library, the value is always less than 256. In the 16-bit library +the value can be up to 0xffff. In the 32-bit library in UTF-32 mode the value +can be up to 0x10ffff, and up to 0xffffffff when not using UTF-32 mode. +

+

+If there is no fixed first value, and if either +
+
+(a) the pattern was compiled with the PCRE_MULTILINE option, and every branch +starts with "^", or +
+
+(b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not set +(if it were set, the pattern would be anchored), +
+
+-1 is returned, indicating that the pattern matches only at the start of a +subject string or after any newline within the string. Otherwise -2 is +returned. For anchored patterns, -2 is returned. +

+  PCRE_INFO_REQUIREDCHARFLAGS
+
+Returns 1 if there is a rightmost literal data unit that must exist in any +matched string, other than at its start. The fourth argument should point to +an int variable. If there is no such value, 0 is returned. If returning +1, the character value itself can be retrieved using PCRE_INFO_REQUIREDCHAR. +

+

+For anchored patterns, a last literal value is recorded only if it follows +something of variable length. For example, for the pattern /^a\d+z\d+/ the +returned value 1 (with "z" returned from PCRE_INFO_REQUIREDCHAR), but for +/^a\dz\d/ the returned value is 0. +

+  PCRE_INFO_REQUIREDCHAR
+
+Return the value of the rightmost literal data unit that must exist in any +matched string, other than at its start, if such a value has been recorded. The +fourth argument should point to an uint32_t variable. If there is no such +value, 0 is returned. +

+
REFERENCE COUNTS
+

+int pcre_refcount(pcre *code, int adjust); +

+

+The pcre_refcount() function is used to maintain a reference count in the +data block that contains a compiled pattern. It is provided for the benefit of +applications that operate in an object-oriented manner, where different parts +of the application may be using the same compiled pattern, but you want to free +the block when they are all done. +

+

+When a pattern is compiled, the reference count field is initialized to zero. +It is changed only by calling this function, whose action is to add the +adjust value (which may be positive or negative) to it. The yield of the +function is the new value. However, the value of the count is constrained to +lie between 0 and 65535, inclusive. If the new value is outside these limits, +it is forced to the appropriate limit value. +

+

+Except when it is zero, the reference count is not correctly preserved if a +pattern is compiled on one host and then transferred to a host whose byte-order +is different. (This seems a highly unlikely scenario.) +

+
MATCHING A PATTERN: THE TRADITIONAL FUNCTION
+

+int pcre_exec(const pcre *code, const pcre_extra *extra, +const char *subject, int length, int startoffset, +int options, int *ovector, int ovecsize); +

+

+The function pcre_exec() is called to match a subject string against a +compiled pattern, which is passed in the code argument. If the +pattern was studied, the result of the study should be passed in the +extra argument. You can call pcre_exec() with the same code +and extra arguments as many times as you like, in order to match +different subject strings with the same pattern. +

+

+This function is the main matching facility of the library, and it operates in +a Perl-like manner. For specialist use there is also an alternative matching +function, which is described +below +in the section about the pcre_dfa_exec() function. +

+

+In most applications, the pattern will have been compiled (and optionally +studied) in the same process that calls pcre_exec(). However, it is +possible to save compiled patterns and study data, and then use them later +in different processes, possibly even on different hosts. For a discussion +about this, see the +pcreprecompile +documentation. +

+

+Here is an example of a simple call to pcre_exec(): +

+  int rc;
+  int ovector[30];
+  rc = pcre_exec(
+    re,             /* result of pcre_compile() */
+    NULL,           /* we didn't study the pattern */
+    "some string",  /* the subject string */
+    11,             /* the length of the subject string */
+    0,              /* start at offset 0 in the subject */
+    0,              /* default options */
+    ovector,        /* vector of integers for substring information */
+    30);            /* number of elements (NOT size in bytes) */
+
+

+
+Extra data for pcre_exec() +
+

+If the extra argument is not NULL, it must point to a pcre_extra +data block. The pcre_study() function returns such a block (when it +doesn't return NULL), but you can also create one for yourself, and pass +additional information in it. The pcre_extra block contains the following +fields (not necessarily in this order): +

+  unsigned long int flags;
+  void *study_data;
+  void *executable_jit;
+  unsigned long int match_limit;
+  unsigned long int match_limit_recursion;
+  void *callout_data;
+  const unsigned char *tables;
+  unsigned char **mark;
+
+In the 16-bit version of this structure, the mark field has type +"PCRE_UCHAR16 **". +
+
+In the 32-bit version of this structure, the mark field has type +"PCRE_UCHAR32 **". +

+

+The flags field is used to specify which of the other fields are set. The +flag bits are: +

+  PCRE_EXTRA_CALLOUT_DATA
+  PCRE_EXTRA_EXECUTABLE_JIT
+  PCRE_EXTRA_MARK
+  PCRE_EXTRA_MATCH_LIMIT
+  PCRE_EXTRA_MATCH_LIMIT_RECURSION
+  PCRE_EXTRA_STUDY_DATA
+  PCRE_EXTRA_TABLES
+
+Other flag bits should be set to zero. The study_data field and sometimes +the executable_jit field are set in the pcre_extra block that is +returned by pcre_study(), together with the appropriate flag bits. You +should not set these yourself, but you may add to the block by setting other +fields and their corresponding flag bits. +

+

+The match_limit field provides a means of preventing PCRE from using up a +vast amount of resources when running patterns that are not going to match, +but which have a very large number of possibilities in their search trees. The +classic example is a pattern that uses nested unlimited repeats. +

+

+Internally, pcre_exec() uses a function called match(), which it +calls repeatedly (sometimes recursively). The limit set by match_limit is +imposed on the number of times this function is called during a match, which +has the effect of limiting the amount of backtracking that can take place. For +patterns that are not anchored, the count restarts from zero for each position +in the subject string. +

+

+When pcre_exec() is called with a pattern that was successfully studied +with a JIT option, the way that the matching is executed is entirely different. +However, there is still the possibility of runaway matching that goes on for a +very long time, and so the match_limit value is also used in this case +(but in a different way) to limit how long the matching can continue. +

+

+The default value for the limit can be set when PCRE is built; the default +default is 10 million, which handles all but the most extreme cases. You can +override the default by suppling pcre_exec() with a pcre_extra +block in which match_limit is set, and PCRE_EXTRA_MATCH_LIMIT is set in +the flags field. If the limit is exceeded, pcre_exec() returns +PCRE_ERROR_MATCHLIMIT. +

+

+A value for the match limit may also be supplied by an item at the start of a +pattern of the form +

+  (*LIMIT_MATCH=d)
+
+where d is a decimal number. However, such a setting is ignored unless d is +less than the limit set by the caller of pcre_exec() or, if no such limit +is set, less than the default. +

+

+The match_limit_recursion field is similar to match_limit, but +instead of limiting the total number of times that match() is called, it +limits the depth of recursion. The recursion depth is a smaller number than the +total number of calls, because not all calls to match() are recursive. +This limit is of use only if it is set smaller than match_limit. +

+

+Limiting the recursion depth limits the amount of machine stack that can be +used, or, when PCRE has been compiled to use memory on the heap instead of the +stack, the amount of heap memory that can be used. This limit is not relevant, +and is ignored, when matching is done using JIT compiled code. +

+

+The default value for match_limit_recursion can be set when PCRE is +built; the default default is the same value as the default for +match_limit. You can override the default by suppling pcre_exec() +with a pcre_extra block in which match_limit_recursion is set, and +PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the flags field. If the limit +is exceeded, pcre_exec() returns PCRE_ERROR_RECURSIONLIMIT. +

+

+A value for the recursion limit may also be supplied by an item at the start of +a pattern of the form +

+  (*LIMIT_RECURSION=d)
+
+where d is a decimal number. However, such a setting is ignored unless d is +less than the limit set by the caller of pcre_exec() or, if no such limit +is set, less than the default. +

+

+The callout_data field is used in conjunction with the "callout" feature, +and is described in the +pcrecallout +documentation. +

+

+The tables field is used to pass a character tables pointer to +pcre_exec(); this overrides the value that is stored with the compiled +pattern. A non-NULL value is stored with the compiled pattern only if custom +tables were supplied to pcre_compile() via its tableptr argument. +If NULL is passed to pcre_exec() using this mechanism, it forces PCRE's +internal tables to be used. This facility is helpful when re-using patterns +that have been saved after compiling with an external set of tables, because +the external tables might be at a different address when pcre_exec() is +called. See the +pcreprecompile +documentation for a discussion of saving compiled patterns for later use. +

+

+If PCRE_EXTRA_MARK is set in the flags field, the mark field must +be set to point to a suitable variable. If the pattern contains any +backtracking control verbs such as (*MARK:NAME), and the execution ends up with +a name to pass back, a pointer to the name string (zero terminated) is placed +in the variable pointed to by the mark field. The names are within the +compiled pattern; if you wish to retain such a name you must copy it before +freeing the memory of a compiled pattern. If there is no name to pass back, the +variable pointed to by the mark field is set to NULL. For details of the +backtracking control verbs, see the section entitled +"Backtracking control" +in the +pcrepattern +documentation. +

+
+Option bits for pcre_exec() +
+

+The unused bits of the options argument for pcre_exec() must be +zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_xxx, +PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART, +PCRE_NO_START_OPTIMIZE, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL_HARD, and +PCRE_PARTIAL_SOFT. +

+

+If the pattern was successfully studied with one of the just-in-time (JIT) +compile options, the only supported options for JIT execution are +PCRE_NO_UTF8_CHECK, PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, +PCRE_NOTEMPTY_ATSTART, PCRE_PARTIAL_HARD, and PCRE_PARTIAL_SOFT. If an +unsupported option is used, JIT execution is disabled and the normal +interpretive code in pcre_exec() is run. +

+  PCRE_ANCHORED
+
+The PCRE_ANCHORED option limits pcre_exec() to matching at the first +matching position. If a pattern was compiled with PCRE_ANCHORED, or turned out +to be anchored by virtue of its contents, it cannot be made unachored at +matching time. +
+  PCRE_BSR_ANYCRLF
+  PCRE_BSR_UNICODE
+
+These options (which are mutually exclusive) control what the \R escape +sequence matches. The choice is either to match only CR, LF, or CRLF, or to +match any Unicode newline sequence. These options override the choice that was +made or defaulted when the pattern was compiled. +
+  PCRE_NEWLINE_CR
+  PCRE_NEWLINE_LF
+  PCRE_NEWLINE_CRLF
+  PCRE_NEWLINE_ANYCRLF
+  PCRE_NEWLINE_ANY
+
+These options override the newline definition that was chosen or defaulted when +the pattern was compiled. For details, see the description of +pcre_compile() above. During matching, the newline choice affects the +behaviour of the dot, circumflex, and dollar metacharacters. It may also alter +the way the match position is advanced after a match failure for an unanchored +pattern. +

+

+When PCRE_NEWLINE_CRLF, PCRE_NEWLINE_ANYCRLF, or PCRE_NEWLINE_ANY is set, and a +match attempt for an unanchored pattern fails when the current position is at a +CRLF sequence, and the pattern contains no explicit matches for CR or LF +characters, the match position is advanced by two characters instead of one, in +other words, to after the CRLF. +

+

+The above rule is a compromise that makes the most common cases work as +expected. For example, if the pattern is .+A (and the PCRE_DOTALL option is not +set), it does not match the string "\r\nA" because, after failing at the +start, it skips both the CR and the LF before retrying. However, the pattern +[\r\n]A does match that string, because it contains an explicit CR or LF +reference, and so advances only by one character after the first failure. +

+

+An explicit match for CR of LF is either a literal appearance of one of those +characters, or one of the \r or \n escape sequences. Implicit matches such as +[^X] do not count, nor does \s (which includes CR and LF in the characters +that it matches). +

+

+Notwithstanding the above, anomalous effects may still occur when CRLF is a +valid newline sequence and explicit \r or \n escapes appear in the pattern. +

+  PCRE_NOTBOL
+
+This option specifies that first character of the subject string is not the +beginning of a line, so the circumflex metacharacter should not match before +it. Setting this without PCRE_MULTILINE (at compile time) causes circumflex +never to match. This option affects only the behaviour of the circumflex +metacharacter. It does not affect \A. +
+  PCRE_NOTEOL
+
+This option specifies that the end of the subject string is not the end of a +line, so the dollar metacharacter should not match it nor (except in multiline +mode) a newline immediately before it. Setting this without PCRE_MULTILINE (at +compile time) causes dollar never to match. This option affects only the +behaviour of the dollar metacharacter. It does not affect \Z or \z. +
+  PCRE_NOTEMPTY
+
+An empty string is not considered to be a valid match if this option is set. If +there are alternatives in the pattern, they are tried. If all the alternatives +match the empty string, the entire match fails. For example, if the pattern +
+  a?b?
+
+is applied to a string not beginning with "a" or "b", it matches an empty +string at the start of the subject. With PCRE_NOTEMPTY set, this match is not +valid, so PCRE searches further into the string for occurrences of "a" or "b". +
+  PCRE_NOTEMPTY_ATSTART
+
+This is like PCRE_NOTEMPTY, except that an empty string match that is not at +the start of the subject is permitted. If the pattern is anchored, such a match +can occur only if the pattern contains \K. +

+

+Perl has no direct equivalent of PCRE_NOTEMPTY or PCRE_NOTEMPTY_ATSTART, but it +does make a special case of a pattern match of the empty string within its +split() function, and when using the /g modifier. It is possible to +emulate Perl's behaviour after matching a null string by first trying the match +again at the same offset with PCRE_NOTEMPTY_ATSTART and PCRE_ANCHORED, and then +if that fails, by advancing the starting offset (see below) and trying an +ordinary match again. There is some code that demonstrates how to do this in +the +pcredemo +sample program. In the most general case, you have to check to see if the +newline convention recognizes CRLF as a newline, and if so, and the current +character is CR followed by LF, advance the starting offset by two characters +instead of one. +

+  PCRE_NO_START_OPTIMIZE
+
+There are a number of optimizations that pcre_exec() uses at the start of +a match, in order to speed up the process. For example, if it is known that an +unanchored match must start with a specific character, it searches the subject +for that character, and fails immediately if it cannot find it, without +actually running the main matching function. This means that a special item +such as (*COMMIT) at the start of a pattern is not considered until after a +suitable starting point for the match has been found. Also, when callouts or +(*MARK) items are in use, these "start-up" optimizations can cause them to be +skipped if the pattern is never actually used. The start-up optimizations are +in effect a pre-scan of the subject that takes place before the pattern is run. +

+

+The PCRE_NO_START_OPTIMIZE option disables the start-up optimizations, possibly +causing performance to suffer, but ensuring that in cases where the result is +"no match", the callouts do occur, and that items such as (*COMMIT) and (*MARK) +are considered at every possible starting position in the subject string. If +PCRE_NO_START_OPTIMIZE is set at compile time, it cannot be unset at matching +time. The use of PCRE_NO_START_OPTIMIZE at matching time (that is, passing it +to pcre_exec()) disables JIT execution; in this situation, matching is +always done using interpretively. +

+

+Setting PCRE_NO_START_OPTIMIZE can change the outcome of a matching operation. +Consider the pattern +

+  (*COMMIT)ABC
+
+When this is compiled, PCRE records the fact that a match must start with the +character "A". Suppose the subject string is "DEFABC". The start-up +optimization scans along the subject, finds "A" and runs the first match +attempt from there. The (*COMMIT) item means that the pattern must match the +current starting position, which in this case, it does. However, if the same +match is run with PCRE_NO_START_OPTIMIZE set, the initial scan along the +subject string does not happen. The first match attempt is run starting from +"D" and when this fails, (*COMMIT) prevents any further matches being tried, so +the overall result is "no match". If the pattern is studied, more start-up +optimizations may be used. For example, a minimum length for the subject may be +recorded. Consider the pattern +
+  (*MARK:A)(X|Y)
+
+The minimum length for a match is one character. If the subject is "ABC", there +will be attempts to match "ABC", "BC", "C", and then finally an empty string. +If the pattern is studied, the final attempt does not take place, because PCRE +knows that the subject is too short, and so the (*MARK) is never encountered. +In this case, studying the pattern does not affect the overall match result, +which is still "no match", but it does affect the auxiliary information that is +returned. +
+  PCRE_NO_UTF8_CHECK
+
+When PCRE_UTF8 is set at compile time, the validity of the subject as a UTF-8 +string is automatically checked when pcre_exec() is subsequently called. +The entire string is checked before any other processing takes place. The value +of startoffset is also checked to ensure that it points to the start of a +UTF-8 character. There is a discussion about the +validity of UTF-8 strings +in the +pcreunicode +page. If an invalid sequence of bytes is found, pcre_exec() returns the +error PCRE_ERROR_BADUTF8 or, if PCRE_PARTIAL_HARD is set and the problem is a +truncated character at the end of the subject, PCRE_ERROR_SHORTUTF8. In both +cases, information about the precise nature of the error may also be returned +(see the descriptions of these errors in the section entitled \fIError return +values from\fP pcre_exec() +below). +If startoffset contains a value that does not point to the start of a +UTF-8 character (or to the end of the subject), PCRE_ERROR_BADUTF8_OFFSET is +returned. +

+

+If you already know that your subject is valid, and you want to skip these +checks for performance reasons, you can set the PCRE_NO_UTF8_CHECK option when +calling pcre_exec(). You might want to do this for the second and +subsequent calls to pcre_exec() if you are making repeated calls to find +all the matches in a single subject string. However, you should be sure that +the value of startoffset points to the start of a character (or the end +of the subject). When PCRE_NO_UTF8_CHECK is set, the effect of passing an +invalid string as a subject or an invalid value of startoffset is +undefined. Your program may crash. +

+  PCRE_PARTIAL_HARD
+  PCRE_PARTIAL_SOFT
+
+These options turn on the partial matching feature. For backwards +compatibility, PCRE_PARTIAL is a synonym for PCRE_PARTIAL_SOFT. A partial match +occurs if the end of the subject string is reached successfully, but there are +not enough subject characters to complete the match. If this happens when +PCRE_PARTIAL_SOFT (but not PCRE_PARTIAL_HARD) is set, matching continues by +testing any remaining alternatives. Only if no complete match can be found is +PCRE_ERROR_PARTIAL returned instead of PCRE_ERROR_NOMATCH. In other words, +PCRE_PARTIAL_SOFT says that the caller is prepared to handle a partial match, +but only if no complete match can be found. +

+

+If PCRE_PARTIAL_HARD is set, it overrides PCRE_PARTIAL_SOFT. In this case, if a +partial match is found, pcre_exec() immediately returns +PCRE_ERROR_PARTIAL, without considering any other alternatives. In other words, +when PCRE_PARTIAL_HARD is set, a partial match is considered to be more +important that an alternative complete match. +

+

+In both cases, the portion of the string that was inspected when the partial +match was found is set as the first matching string. There is a more detailed +discussion of partial and multi-segment matching, with examples, in the +pcrepartial +documentation. +

+
+The string to be matched by pcre_exec() +
+

+The subject string is passed to pcre_exec() as a pointer in +subject, a length in length, and a starting offset in +startoffset. The units for length and startoffset are bytes +for the 8-bit library, 16-bit data items for the 16-bit library, and 32-bit +data items for the 32-bit library. +

+

+If startoffset is negative or greater than the length of the subject, +pcre_exec() returns PCRE_ERROR_BADOFFSET. When the starting offset is +zero, the search for a match starts at the beginning of the subject, and this +is by far the most common case. In UTF-8 or UTF-16 mode, the offset must point +to the start of a character, or the end of the subject (in UTF-32 mode, one +data unit equals one character, so all offsets are valid). Unlike the pattern +string, the subject may contain binary zeroes. +

+

+A non-zero starting offset is useful when searching for another match in the +same subject by calling pcre_exec() again after a previous success. +Setting startoffset differs from just passing over a shortened string and +setting PCRE_NOTBOL in the case of a pattern that begins with any kind of +lookbehind. For example, consider the pattern +

+  \Biss\B
+
+which finds occurrences of "iss" in the middle of words. (\B matches only if +the current position in the subject is not a word boundary.) When applied to +the string "Mississipi" the first call to pcre_exec() finds the first +occurrence. If pcre_exec() is called again with just the remainder of the +subject, namely "issipi", it does not match, because \B is always false at the +start of the subject, which is deemed to be a word boundary. However, if +pcre_exec() is passed the entire string again, but with startoffset +set to 4, it finds the second occurrence of "iss" because it is able to look +behind the starting point to discover that it is preceded by a letter. +

+

+Finding all the matches in a subject is tricky when the pattern can match an +empty string. It is possible to emulate Perl's /g behaviour by first trying the +match again at the same offset, with the PCRE_NOTEMPTY_ATSTART and +PCRE_ANCHORED options, and then if that fails, advancing the starting offset +and trying an ordinary match again. There is some code that demonstrates how to +do this in the +pcredemo +sample program. In the most general case, you have to check to see if the +newline convention recognizes CRLF as a newline, and if so, and the current +character is CR followed by LF, advance the starting offset by two characters +instead of one. +

+

+If a non-zero starting offset is passed when the pattern is anchored, one +attempt to match at the given offset is made. This can only succeed if the +pattern does not require the match to be at the start of the subject. +

+
+How pcre_exec() returns captured substrings +
+

+In general, a pattern matches a certain portion of the subject, and in +addition, further substrings from the subject may be picked out by parts of the +pattern. Following the usage in Jeffrey Friedl's book, this is called +"capturing" in what follows, and the phrase "capturing subpattern" is used for +a fragment of a pattern that picks out a substring. PCRE supports several other +kinds of parenthesized subpattern that do not cause substrings to be captured. +

+

+Captured substrings are returned to the caller via a vector of integers whose +address is passed in ovector. The number of elements in the vector is +passed in ovecsize, which must be a non-negative number. Note: this +argument is NOT the size of ovector in bytes. +

+

+The first two-thirds of the vector is used to pass back captured substrings, +each substring using a pair of integers. The remaining third of the vector is +used as workspace by pcre_exec() while matching capturing subpatterns, +and is not available for passing back information. The number passed in +ovecsize should always be a multiple of three. If it is not, it is +rounded down. +

+

+When a match is successful, information about captured substrings is returned +in pairs of integers, starting at the beginning of ovector, and +continuing up to two-thirds of its length at the most. The first element of +each pair is set to the offset of the first character in a substring, and the +second is set to the offset of the first character after the end of a +substring. These values are always data unit offsets, even in UTF mode. They +are byte offsets in the 8-bit library, 16-bit data item offsets in the 16-bit +library, and 32-bit data item offsets in the 32-bit library. Note: they +are not character counts. +

+

+The first pair of integers, ovector[0] and ovector[1], identify the +portion of the subject string matched by the entire pattern. The next pair is +used for the first capturing subpattern, and so on. The value returned by +pcre_exec() is one more than the highest numbered pair that has been set. +For example, if two substrings have been captured, the returned value is 3. If +there are no capturing subpatterns, the return value from a successful match is +1, indicating that just the first pair of offsets has been set. +

+

+If a capturing subpattern is matched repeatedly, it is the last portion of the +string that it matched that is returned. +

+

+If the vector is too small to hold all the captured substring offsets, it is +used as far as possible (up to two-thirds of its length), and the function +returns a value of zero. If neither the actual string matched nor any captured +substrings are of interest, pcre_exec() may be called with ovector +passed as NULL and ovecsize as zero. However, if the pattern contains +back references and the ovector is not big enough to remember the related +substrings, PCRE has to get additional memory for use during matching. Thus it +is usually advisable to supply an ovector of reasonable size. +

+

+There are some cases where zero is returned (indicating vector overflow) when +in fact the vector is exactly the right size for the final match. For example, +consider the pattern +

+  (a)(?:(b)c|bd)
+
+If a vector of 6 elements (allowing for only 1 captured substring) is given +with subject string "abd", pcre_exec() will try to set the second +captured string, thereby recording a vector overflow, before failing to match +"c" and backing up to try the second alternative. The zero return, however, +does correctly indicate that the maximum number of slots (namely 2) have been +filled. In similar cases where there is temporary overflow, but the final +number of used slots is actually less than the maximum, a non-zero value is +returned. +

+

+The pcre_fullinfo() function can be used to find out how many capturing +subpatterns there are in a compiled pattern. The smallest size for +ovector that will allow for n captured substrings, in addition to +the offsets of the substring matched by the whole pattern, is (n+1)*3. +

+

+It is possible for capturing subpattern number n+1 to match some part of +the subject when subpattern n has not been used at all. For example, if +the string "abc" is matched against the pattern (a|(z))(bc) the return from the +function is 4, and subpatterns 1 and 3 are matched, but 2 is not. When this +happens, both values in the offset pairs corresponding to unused subpatterns +are set to -1. +

+

+Offset values that correspond to unused subpatterns at the end of the +expression are also set to -1. For example, if the string "abc" is matched +against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The +return from the function is 2, because the highest used capturing subpattern +number is 1, and the offsets for for the second and third capturing subpatterns +(assuming the vector is large enough, of course) are set to -1. +

+

+Note: Elements in the first two-thirds of ovector that do not +correspond to capturing parentheses in the pattern are never changed. That is, +if a pattern contains n capturing parentheses, no more than +ovector[0] to ovector[2n+1] are set by pcre_exec(). The other +elements (in the first two-thirds) retain whatever values they previously had. +

+

+Some convenience functions are provided for extracting the captured substrings +as separate strings. These are described below. +

+
+Error return values from pcre_exec() +
+

+If pcre_exec() fails, it returns a negative number. The following are +defined in the header file: +

+  PCRE_ERROR_NOMATCH        (-1)
+
+The subject string did not match the pattern. +
+  PCRE_ERROR_NULL           (-2)
+
+Either code or subject was passed as NULL, or ovector was +NULL and ovecsize was not zero. +
+  PCRE_ERROR_BADOPTION      (-3)
+
+An unrecognized bit was set in the options argument. +
+  PCRE_ERROR_BADMAGIC       (-4)
+
+PCRE stores a 4-byte "magic number" at the start of the compiled code, to catch +the case when it is passed a junk pointer and to detect when a pattern that was +compiled in an environment of one endianness is run in an environment with the +other endianness. This is the error that PCRE gives when the magic number is +not present. +
+  PCRE_ERROR_UNKNOWN_OPCODE (-5)
+
+While running the pattern match, an unknown item was encountered in the +compiled pattern. This error could be caused by a bug in PCRE or by overwriting +of the compiled pattern. +
+  PCRE_ERROR_NOMEMORY       (-6)
+
+If a pattern contains back references, but the ovector that is passed to +pcre_exec() is not big enough to remember the referenced substrings, PCRE +gets a block of memory at the start of matching to use for this purpose. If the +call via pcre_malloc() fails, this error is given. The memory is +automatically freed at the end of matching. +

+

+This error is also given if pcre_stack_malloc() fails in +pcre_exec(). This can happen only when PCRE has been compiled with +--disable-stack-for-recursion. +

+  PCRE_ERROR_NOSUBSTRING    (-7)
+
+This error is used by the pcre_copy_substring(), +pcre_get_substring(), and pcre_get_substring_list() functions (see +below). It is never returned by pcre_exec(). +
+  PCRE_ERROR_MATCHLIMIT     (-8)
+
+The backtracking limit, as specified by the match_limit field in a +pcre_extra structure (or defaulted) was reached. See the description +above. +
+  PCRE_ERROR_CALLOUT        (-9)
+
+This error is never generated by pcre_exec() itself. It is provided for +use by callout functions that want to yield a distinctive error code. See the +pcrecallout +documentation for details. +
+  PCRE_ERROR_BADUTF8        (-10)
+
+A string that contains an invalid UTF-8 byte sequence was passed as a subject, +and the PCRE_NO_UTF8_CHECK option was not set. If the size of the output vector +(ovecsize) is at least 2, the byte offset to the start of the the invalid +UTF-8 character is placed in the first element, and a reason code is placed in +the second element. The reason codes are listed in the +following section. +For backward compatibility, if PCRE_PARTIAL_HARD is set and the problem is a +truncated UTF-8 character at the end of the subject (reason codes 1 to 5), +PCRE_ERROR_SHORTUTF8 is returned instead of PCRE_ERROR_BADUTF8. +
+  PCRE_ERROR_BADUTF8_OFFSET (-11)
+
+The UTF-8 byte sequence that was passed as a subject was checked and found to +be valid (the PCRE_NO_UTF8_CHECK option was not set), but the value of +startoffset did not point to the beginning of a UTF-8 character or the +end of the subject. +
+  PCRE_ERROR_PARTIAL        (-12)
+
+The subject string did not match, but it did match partially. See the +pcrepartial +documentation for details of partial matching. +
+  PCRE_ERROR_BADPARTIAL     (-13)
+
+This code is no longer in use. It was formerly returned when the PCRE_PARTIAL +option was used with a compiled pattern containing items that were not +supported for partial matching. From release 8.00 onwards, there are no +restrictions on partial matching. +
+  PCRE_ERROR_INTERNAL       (-14)
+
+An unexpected internal error has occurred. This error could be caused by a bug +in PCRE or by overwriting of the compiled pattern. +
+  PCRE_ERROR_BADCOUNT       (-15)
+
+This error is given if the value of the ovecsize argument is negative. +
+  PCRE_ERROR_RECURSIONLIMIT (-21)
+
+The internal recursion limit, as specified by the match_limit_recursion +field in a pcre_extra structure (or defaulted) was reached. See the +description above. +
+  PCRE_ERROR_BADNEWLINE     (-23)
+
+An invalid combination of PCRE_NEWLINE_xxx options was given. +
+  PCRE_ERROR_BADOFFSET      (-24)
+
+The value of startoffset was negative or greater than the length of the +subject, that is, the value in length. +
+  PCRE_ERROR_SHORTUTF8      (-25)
+
+This error is returned instead of PCRE_ERROR_BADUTF8 when the subject string +ends with a truncated UTF-8 character and the PCRE_PARTIAL_HARD option is set. +Information about the failure is returned as for PCRE_ERROR_BADUTF8. It is in +fact sufficient to detect this case, but this special error code for +PCRE_PARTIAL_HARD precedes the implementation of returned information; it is +retained for backwards compatibility. +
+  PCRE_ERROR_RECURSELOOP    (-26)
+
+This error is returned when pcre_exec() detects a recursion loop within +the pattern. Specifically, it means that either the whole pattern or a +subpattern has been called recursively for the second time at the same position +in the subject string. Some simple patterns that might do this are detected and +faulted at compile time, but more complicated cases, in particular mutual +recursions between two different subpatterns, cannot be detected until run +time. +
+  PCRE_ERROR_JIT_STACKLIMIT (-27)
+
+This error is returned when a pattern that was successfully studied using a +JIT compile option is being matched, but the memory available for the +just-in-time processing stack is not large enough. See the +pcrejit +documentation for more details. +
+  PCRE_ERROR_BADMODE        (-28)
+
+This error is given if a pattern that was compiled by the 8-bit library is +passed to a 16-bit or 32-bit library function, or vice versa. +
+  PCRE_ERROR_BADENDIANNESS  (-29)
+
+This error is given if a pattern that was compiled and saved is reloaded on a +host with different endianness. The utility function +pcre_pattern_to_host_byte_order() can be used to convert such a pattern +so that it runs on the new host. +
+  PCRE_ERROR_JIT_BADOPTION
+
+This error is returned when a pattern that was successfully studied using a JIT +compile option is being matched, but the matching mode (partial or complete +match) does not correspond to any JIT compilation mode. When the JIT fast path +function is used, this error may be also given for invalid options. See the +pcrejit +documentation for more details. +
+  PCRE_ERROR_BADLENGTH      (-32)
+
+This error is given if pcre_exec() is called with a negative value for +the length argument. +

+

+Error numbers -16 to -20, -22, and 30 are not used by pcre_exec(). +

+
+Reason codes for invalid UTF-8 strings +
+

+This section applies only to the 8-bit library. The corresponding information +for the 16-bit and 32-bit libraries is given in the +pcre16 +and +pcre32 +pages. +

+

+When pcre_exec() returns either PCRE_ERROR_BADUTF8 or +PCRE_ERROR_SHORTUTF8, and the size of the output vector (ovecsize) is at +least 2, the offset of the start of the invalid UTF-8 character is placed in +the first output vector element (ovector[0]) and a reason code is placed +in the second element (ovector[1]). The reason codes are given names in +the pcre.h header file: +

+  PCRE_UTF8_ERR1
+  PCRE_UTF8_ERR2
+  PCRE_UTF8_ERR3
+  PCRE_UTF8_ERR4
+  PCRE_UTF8_ERR5
+
+The string ends with a truncated UTF-8 character; the code specifies how many +bytes are missing (1 to 5). Although RFC 3629 restricts UTF-8 characters to be +no longer than 4 bytes, the encoding scheme (originally defined by RFC 2279) +allows for up to 6 bytes, and this is checked first; hence the possibility of +4 or 5 missing bytes. +
+  PCRE_UTF8_ERR6
+  PCRE_UTF8_ERR7
+  PCRE_UTF8_ERR8
+  PCRE_UTF8_ERR9
+  PCRE_UTF8_ERR10
+
+The two most significant bits of the 2nd, 3rd, 4th, 5th, or 6th byte of the +character do not have the binary value 0b10 (that is, either the most +significant bit is 0, or the next bit is 1). +
+  PCRE_UTF8_ERR11
+  PCRE_UTF8_ERR12
+
+A character that is valid by the RFC 2279 rules is either 5 or 6 bytes long; +these code points are excluded by RFC 3629. +
+  PCRE_UTF8_ERR13
+
+A 4-byte character has a value greater than 0x10fff; these code points are +excluded by RFC 3629. +
+  PCRE_UTF8_ERR14
+
+A 3-byte character has a value in the range 0xd800 to 0xdfff; this range of +code points are reserved by RFC 3629 for use with UTF-16, and so are excluded +from UTF-8. +
+  PCRE_UTF8_ERR15
+  PCRE_UTF8_ERR16
+  PCRE_UTF8_ERR17
+  PCRE_UTF8_ERR18
+  PCRE_UTF8_ERR19
+
+A 2-, 3-, 4-, 5-, or 6-byte character is "overlong", that is, it codes for a +value that can be represented by fewer bytes, which is invalid. For example, +the two bytes 0xc0, 0xae give the value 0x2e, whose correct coding uses just +one byte. +
+  PCRE_UTF8_ERR20
+
+The two most significant bits of the first byte of a character have the binary +value 0b10 (that is, the most significant bit is 1 and the second is 0). Such a +byte can only validly occur as the second or subsequent byte of a multi-byte +character. +
+  PCRE_UTF8_ERR21
+
+The first byte of a character has the value 0xfe or 0xff. These values can +never occur in a valid UTF-8 string. +
+  PCRE_UTF8_ERR22
+
+This error code was formerly used when the presence of a so-called +"non-character" caused an error. Unicode corrigendum #9 makes it clear that +such characters should not cause a string to be rejected, and so this code is +no longer in use and is never returned. +

+
EXTRACTING CAPTURED SUBSTRINGS BY NUMBER
+

+int pcre_copy_substring(const char *subject, int *ovector, +int stringcount, int stringnumber, char *buffer, +int buffersize); +

+

+int pcre_get_substring(const char *subject, int *ovector, +int stringcount, int stringnumber, +const char **stringptr); +

+

+int pcre_get_substring_list(const char *subject, +int *ovector, int stringcount, const char ***listptr); +

+

+Captured substrings can be accessed directly by using the offsets returned by +pcre_exec() in ovector. For convenience, the functions +pcre_copy_substring(), pcre_get_substring(), and +pcre_get_substring_list() are provided for extracting captured substrings +as new, separate, zero-terminated strings. These functions identify substrings +by number. The next section describes functions for extracting named +substrings. +

+

+A substring that contains a binary zero is correctly extracted and has a +further zero added on the end, but the result is not, of course, a C string. +However, you can process such a string by referring to the length that is +returned by pcre_copy_substring() and pcre_get_substring(). +Unfortunately, the interface to pcre_get_substring_list() is not adequate +for handling strings containing binary zeros, because the end of the final +string is not independently indicated. +

+

+The first three arguments are the same for all three of these functions: +subject is the subject string that has just been successfully matched, +ovector is a pointer to the vector of integer offsets that was passed to +pcre_exec(), and stringcount is the number of substrings that were +captured by the match, including the substring that matched the entire regular +expression. This is the value returned by pcre_exec() if it is greater +than zero. If pcre_exec() returned zero, indicating that it ran out of +space in ovector, the value passed as stringcount should be the +number of elements in the vector divided by three. +

+

+The functions pcre_copy_substring() and pcre_get_substring() +extract a single substring, whose number is given as stringnumber. A +value of zero extracts the substring that matched the entire pattern, whereas +higher values extract the captured substrings. For pcre_copy_substring(), +the string is placed in buffer, whose length is given by +buffersize, while for pcre_get_substring() a new block of memory is +obtained via pcre_malloc, and its address is returned via +stringptr. The yield of the function is the length of the string, not +including the terminating zero, or one of these error codes: +

+  PCRE_ERROR_NOMEMORY       (-6)
+
+The buffer was too small for pcre_copy_substring(), or the attempt to get +memory failed for pcre_get_substring(). +
+  PCRE_ERROR_NOSUBSTRING    (-7)
+
+There is no substring whose number is stringnumber. +

+

+The pcre_get_substring_list() function extracts all available substrings +and builds a list of pointers to them. All this is done in a single block of +memory that is obtained via pcre_malloc. The address of the memory block +is returned via listptr, which is also the start of the list of string +pointers. The end of the list is marked by a NULL pointer. The yield of the +function is zero if all went well, or the error code +

+  PCRE_ERROR_NOMEMORY       (-6)
+
+if the attempt to get the memory block failed. +

+

+When any of these functions encounter a substring that is unset, which can +happen when capturing subpattern number n+1 matches some part of the +subject, but subpattern n has not been used at all, they return an empty +string. This can be distinguished from a genuine zero-length substring by +inspecting the appropriate offset in ovector, which is negative for unset +substrings. +

+

+The two convenience functions pcre_free_substring() and +pcre_free_substring_list() can be used to free the memory returned by +a previous call of pcre_get_substring() or +pcre_get_substring_list(), respectively. They do nothing more than call +the function pointed to by pcre_free, which of course could be called +directly from a C program. However, PCRE is used in some situations where it is +linked via a special interface to another programming language that cannot use +pcre_free directly; it is for these cases that the functions are +provided. +

+
EXTRACTING CAPTURED SUBSTRINGS BY NAME
+

+int pcre_get_stringnumber(const pcre *code, +const char *name); +

+

+int pcre_copy_named_substring(const pcre *code, +const char *subject, int *ovector, +int stringcount, const char *stringname, +char *buffer, int buffersize); +

+

+int pcre_get_named_substring(const pcre *code, +const char *subject, int *ovector, +int stringcount, const char *stringname, +const char **stringptr); +

+

+To extract a substring by name, you first have to find associated number. +For example, for this pattern +

+  (a+)b(?<xxx>\d+)...
+
+the number of the subpattern called "xxx" is 2. If the name is known to be +unique (PCRE_DUPNAMES was not set), you can find the number from the name by +calling pcre_get_stringnumber(). The first argument is the compiled +pattern, and the second is the name. The yield of the function is the +subpattern number, or PCRE_ERROR_NOSUBSTRING (-7) if there is no subpattern of +that name. +

+

+Given the number, you can extract the substring directly, or use one of the +functions described in the previous section. For convenience, there are also +two functions that do the whole job. +

+

+Most of the arguments of pcre_copy_named_substring() and +pcre_get_named_substring() are the same as those for the similarly named +functions that extract by number. As these are described in the previous +section, they are not re-described here. There are just two differences: +

+

+First, instead of a substring number, a substring name is given. Second, there +is an extra argument, given at the start, which is a pointer to the compiled +pattern. This is needed in order to gain access to the name-to-number +translation table. +

+

+These functions call pcre_get_stringnumber(), and if it succeeds, they +then call pcre_copy_substring() or pcre_get_substring(), as +appropriate. NOTE: If PCRE_DUPNAMES is set and there are duplicate names, +the behaviour may not be what you want (see the next section). +

+

+Warning: If the pattern uses the (?| feature to set up multiple +subpatterns with the same number, as described in the +section on duplicate subpattern numbers +in the +pcrepattern +page, you cannot use names to distinguish the different subpatterns, because +names are not included in the compiled code. The matching process uses only +numbers. For this reason, the use of different names for subpatterns of the +same number causes an error at compile time. +

+
DUPLICATE SUBPATTERN NAMES
+

+int pcre_get_stringtable_entries(const pcre *code, +const char *name, char **first, char **last); +

+

+When a pattern is compiled with the PCRE_DUPNAMES option, names for subpatterns +are not required to be unique. (Duplicate names are always allowed for +subpatterns with the same number, created by using the (?| feature. Indeed, if +such subpatterns are named, they are required to use the same names.) +

+

+Normally, patterns with duplicate names are such that in any one match, only +one of the named subpatterns participates. An example is shown in the +pcrepattern +documentation. +

+

+When duplicates are present, pcre_copy_named_substring() and +pcre_get_named_substring() return the first substring corresponding to +the given name that is set. If none are set, PCRE_ERROR_NOSUBSTRING (-7) is +returned; no data is returned. The pcre_get_stringnumber() function +returns one of the numbers that are associated with the name, but it is not +defined which it is. +

+

+If you want to get full details of all captured substrings for a given name, +you must use the pcre_get_stringtable_entries() function. The first +argument is the compiled pattern, and the second is the name. The third and +fourth are pointers to variables which are updated by the function. After it +has run, they point to the first and last entries in the name-to-number table +for the given name. The function itself returns the length of each entry, or +PCRE_ERROR_NOSUBSTRING (-7) if there are none. The format of the table is +described above in the section entitled Information about a pattern +above. +Given all the relevant entries for the name, you can extract each of their +numbers, and hence the captured data, if any. +

+
FINDING ALL POSSIBLE MATCHES
+

+The traditional matching function uses a similar algorithm to Perl, which stops +when it finds the first match, starting at a given point in the subject. If you +want to find all possible matches, or the longest possible match, consider +using the alternative matching function (see below) instead. If you cannot use +the alternative function, but still need to find all possible matches, you +can kludge it up by making use of the callout facility, which is described in +the +pcrecallout +documentation. +

+

+What you have to do is to insert a callout right at the end of the pattern. +When your callout function is called, extract and save the current matched +substring. Then return 1, which forces pcre_exec() to backtrack and try +other alternatives. Ultimately, when it runs out of matches, pcre_exec() +will yield PCRE_ERROR_NOMATCH. +

+
OBTAINING AN ESTIMATE OF STACK USAGE
+

+Matching certain patterns using pcre_exec() can use a lot of process +stack, which in certain environments can be rather limited in size. Some users +find it helpful to have an estimate of the amount of stack that is used by +pcre_exec(), to help them set recursion limits, as described in the +pcrestack +documentation. The estimate that is output by pcretest when called with +the -m and -C options is obtained by calling pcre_exec with +the values NULL, NULL, NULL, -999, and -999 for its first five arguments. +

+

+Normally, if its first argument is NULL, pcre_exec() immediately returns +the negative error code PCRE_ERROR_NULL, but with this special combination of +arguments, it returns instead a negative number whose absolute value is the +approximate stack frame size in bytes. (A negative number is used so that it is +clear that no match has happened.) The value is approximate because in some +cases, recursive calls to pcre_exec() occur when there are one or two +additional variables on the stack. +

+

+If PCRE has been compiled to use the heap instead of the stack for recursion, +the value returned is the size of each block that is obtained from the heap. +

+
MATCHING A PATTERN: THE ALTERNATIVE FUNCTION
+

+int pcre_dfa_exec(const pcre *code, const pcre_extra *extra, +const char *subject, int length, int startoffset, +int options, int *ovector, int ovecsize, +int *workspace, int wscount); +

+

+The function pcre_dfa_exec() is called to match a subject string against +a compiled pattern, using a matching algorithm that scans the subject string +just once, and does not backtrack. This has different characteristics to the +normal algorithm, and is not compatible with Perl. Some of the features of PCRE +patterns are not supported. Nevertheless, there are times when this kind of +matching can be useful. For a discussion of the two matching algorithms, and a +list of features that pcre_dfa_exec() does not support, see the +pcrematching +documentation. +

+

+The arguments for the pcre_dfa_exec() function are the same as for +pcre_exec(), plus two extras. The ovector argument is used in a +different way, and this is described below. The other common arguments are used +in the same way as for pcre_exec(), so their description is not repeated +here. +

+

+The two additional arguments provide workspace for the function. The workspace +vector should contain at least 20 elements. It is used for keeping track of +multiple paths through the pattern tree. More workspace will be needed for +patterns and subjects where there are a lot of potential matches. +

+

+Here is an example of a simple call to pcre_dfa_exec(): +

+  int rc;
+  int ovector[10];
+  int wspace[20];
+  rc = pcre_dfa_exec(
+    re,             /* result of pcre_compile() */
+    NULL,           /* we didn't study the pattern */
+    "some string",  /* the subject string */
+    11,             /* the length of the subject string */
+    0,              /* start at offset 0 in the subject */
+    0,              /* default options */
+    ovector,        /* vector of integers for substring information */
+    10,             /* number of elements (NOT size in bytes) */
+    wspace,         /* working space vector */
+    20);            /* number of elements (NOT size in bytes) */
+
+

+
+Option bits for pcre_dfa_exec() +
+

+The unused bits of the options argument for pcre_dfa_exec() must be +zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_xxx, +PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART, +PCRE_NO_UTF8_CHECK, PCRE_BSR_ANYCRLF, PCRE_BSR_UNICODE, PCRE_NO_START_OPTIMIZE, +PCRE_PARTIAL_HARD, PCRE_PARTIAL_SOFT, PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART. +All but the last four of these are exactly the same as for pcre_exec(), +so their description is not repeated here. +

+  PCRE_PARTIAL_HARD
+  PCRE_PARTIAL_SOFT
+
+These have the same general effect as they do for pcre_exec(), but the +details are slightly different. When PCRE_PARTIAL_HARD is set for +pcre_dfa_exec(), it returns PCRE_ERROR_PARTIAL if the end of the subject +is reached and there is still at least one matching possibility that requires +additional characters. This happens even if some complete matches have also +been found. When PCRE_PARTIAL_SOFT is set, the return code PCRE_ERROR_NOMATCH +is converted into PCRE_ERROR_PARTIAL if the end of the subject is reached, +there have been no complete matches, but there is still at least one matching +possibility. The portion of the string that was inspected when the longest +partial match was found is set as the first matching string in both cases. +There is a more detailed discussion of partial and multi-segment matching, with +examples, in the +pcrepartial +documentation. +
+  PCRE_DFA_SHORTEST
+
+Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as +soon as it has found one match. Because of the way the alternative algorithm +works, this is necessarily the shortest possible match at the first possible +matching point in the subject string. +
+  PCRE_DFA_RESTART
+
+When pcre_dfa_exec() returns a partial match, it is possible to call it +again, with additional subject characters, and have it continue with the same +match. The PCRE_DFA_RESTART option requests this action; when it is set, the +workspace and wscount options must reference the same vector as +before because data about the match so far is left in them after a partial +match. There is more discussion of this facility in the +pcrepartial +documentation. +

+
+Successful returns from pcre_dfa_exec() +
+

+When pcre_dfa_exec() succeeds, it may have matched more than one +substring in the subject. Note, however, that all the matches from one run of +the function start at the same point in the subject. The shorter matches are +all initial substrings of the longer matches. For example, if the pattern +

+  <.*>
+
+is matched against the string +
+  This is <something> <something else> <something further> no more
+
+the three matched strings are +
+  <something>
+  <something> <something else>
+  <something> <something else> <something further>
+
+On success, the yield of the function is a number greater than zero, which is +the number of matched substrings. The substrings themselves are returned in +ovector. Each string uses two elements; the first is the offset to the +start, and the second is the offset to the end. In fact, all the strings have +the same start offset. (Space could have been saved by giving this only once, +but it was decided to retain some compatibility with the way pcre_exec() +returns data, even though the meaning of the strings is different.) +

+

+The strings are returned in reverse order of length; that is, the longest +matching string is given first. If there were too many matches to fit into +ovector, the yield of the function is zero, and the vector is filled with +the longest matches. Unlike pcre_exec(), pcre_dfa_exec() can use +the entire ovector for returning matched strings. +

+
+Error returns from pcre_dfa_exec() +
+

+The pcre_dfa_exec() function returns a negative number when it fails. +Many of the errors are the same as for pcre_exec(), and these are +described +above. +There are in addition the following errors that are specific to +pcre_dfa_exec(): +

+  PCRE_ERROR_DFA_UITEM      (-16)
+
+This return is given if pcre_dfa_exec() encounters an item in the pattern +that it does not support, for instance, the use of \C or a back reference. +
+  PCRE_ERROR_DFA_UCOND      (-17)
+
+This return is given if pcre_dfa_exec() encounters a condition item that +uses a back reference for the condition, or a test for recursion in a specific +group. These are not supported. +
+  PCRE_ERROR_DFA_UMLIMIT    (-18)
+
+This return is given if pcre_dfa_exec() is called with an extra +block that contains a setting of the match_limit or +match_limit_recursion fields. This is not supported (these fields are +meaningless for DFA matching). +
+  PCRE_ERROR_DFA_WSSIZE     (-19)
+
+This return is given if pcre_dfa_exec() runs out of space in the +workspace vector. +
+  PCRE_ERROR_DFA_RECURSE    (-20)
+
+When a recursive subpattern is processed, the matching function calls itself +recursively, using private vectors for ovector and workspace. This +error is given if the output vector is not large enough. This should be +extremely rare, as a vector of size 1000 is used. +
+  PCRE_ERROR_DFA_BADRESTART (-30)
+
+When pcre_dfa_exec() is called with the PCRE_DFA_RESTART option, +some plausibility checks are made on the contents of the workspace, which +should contain data about the previous partial match. If any of these checks +fail, this error is given. +

+
SEE ALSO
+

+pcre16(3), pcre32(3), pcrebuild(3), pcrecallout(3), +pcrecpp(3)(3), pcrematching(3), pcrepartial(3), +pcreposix(3), pcreprecompile(3), pcresample(3), +pcrestack(3). +

+
AUTHOR
+

+Philip Hazel +
+University Computing Service +
+Cambridge CB2 3QH, England. +
+

+
REVISION
+

+Last updated: 12 May 2013 +
+Copyright © 1997-2013 University of Cambridge. +
+

+Return to the PCRE index page. +

diff --git a/pcre/doc/html/pcrebuild.html b/pcre/doc/html/pcrebuild.html new file mode 100644 index 00000000000..03c8cbe0b21 --- /dev/null +++ b/pcre/doc/html/pcrebuild.html @@ -0,0 +1,534 @@ + + +pcrebuild specification + + +

pcrebuild man page

+

+Return to the PCRE index page. +

+

+This page is part of the PCRE HTML documentation. It was generated automatically +from the original man page. If there is any nonsense in it, please consult the +man page, in case the conversion went wrong. +
+

+
BUILDING PCRE
+

+PCRE is distributed with a configure script that can be used to build the +library in Unix-like environments using the applications known as Autotools. +Also in the distribution are files to support building using CMake +instead of configure. The text file +README +contains general information about building with Autotools (some of which is +repeated below), and also has some comments about building on various operating +systems. There is a lot more information about building PCRE without using +Autotools (including information about using CMake and building "by +hand") in the text file called +NON-AUTOTOOLS-BUILD. +You should consult this file as well as the +README +file if you are building in a non-Unix-like environment. +

+
PCRE BUILD-TIME OPTIONS
+

+The rest of this document describes the optional features of PCRE that can be +selected when the library is compiled. It assumes use of the configure +script, where the optional features are selected or deselected by providing +options to configure before running the make command. However, the +same options can be selected in both Unix-like and non-Unix-like environments +using the GUI facility of cmake-gui if you are using CMake instead +of configure to build PCRE. +

+

+If you are not using Autotools or CMake, option selection can be done by +editing the config.h file, or by passing parameter settings to the +compiler, as described in +NON-AUTOTOOLS-BUILD. +

+

+The complete list of options for configure (which includes the standard +ones such as the selection of the installation directory) can be obtained by +running +

+  ./configure --help
+
+The following sections include descriptions of options whose names begin with +--enable or --disable. These settings specify changes to the defaults for the +configure command. Because of the way that configure works, +--enable and --disable always come in pairs, so the complementary option always +exists as well, but as it specifies the default, it is not described. +

+
BUILDING 8-BIT, 16-BIT AND 32-BIT LIBRARIES
+

+By default, a library called libpcre is built, containing functions that +take string arguments contained in vectors of bytes, either as single-byte +characters, or interpreted as UTF-8 strings. You can also build a separate +library, called libpcre16, in which strings are contained in vectors of +16-bit data units and interpreted either as single-unit characters or UTF-16 +strings, by adding +

+  --enable-pcre16
+
+to the configure command. You can also build yet another separate +library, called libpcre32, in which strings are contained in vectors of +32-bit data units and interpreted either as single-unit characters or UTF-32 +strings, by adding +
+  --enable-pcre32
+
+to the configure command. If you do not want the 8-bit library, add +
+  --disable-pcre8
+
+as well. At least one of the three libraries must be built. Note that the C++ +and POSIX wrappers are for the 8-bit library only, and that pcregrep is +an 8-bit program. None of these are built if you select only the 16-bit or +32-bit libraries. +

+
BUILDING SHARED AND STATIC LIBRARIES
+

+The Autotools PCRE building process uses libtool to build both shared and +static libraries by default. You can suppress one of these by adding one of +

+  --disable-shared
+  --disable-static
+
+to the configure command, as required. +

+
C++ SUPPORT
+

+By default, if the 8-bit library is being built, the configure script +will search for a C++ compiler and C++ header files. If it finds them, it +automatically builds the C++ wrapper library (which supports only 8-bit +strings). You can disable this by adding +

+  --disable-cpp
+
+to the configure command. +

+
UTF-8, UTF-16 AND UTF-32 SUPPORT
+

+To build PCRE with support for UTF Unicode character strings, add +

+  --enable-utf
+
+to the configure command. This setting applies to all three libraries, +adding support for UTF-8 to the 8-bit library, support for UTF-16 to the 16-bit +library, and support for UTF-32 to the to the 32-bit library. There are no +separate options for enabling UTF-8, UTF-16 and UTF-32 independently because +that would allow ridiculous settings such as requesting UTF-16 support while +building only the 8-bit library. It is not possible to build one library with +UTF support and another without in the same configuration. (For backwards +compatibility, --enable-utf8 is a synonym of --enable-utf.) +

+

+Of itself, this setting does not make PCRE treat strings as UTF-8, UTF-16 or +UTF-32. As well as compiling PCRE with this option, you also have have to set +the PCRE_UTF8, PCRE_UTF16 or PCRE_UTF32 option (as appropriate) when you call +one of the pattern compiling functions. +

+

+If you set --enable-utf when compiling in an EBCDIC environment, PCRE expects +its input to be either ASCII or UTF-8 (depending on the run-time option). It is +not possible to support both EBCDIC and UTF-8 codes in the same version of the +library. Consequently, --enable-utf and --enable-ebcdic are mutually +exclusive. +

+
UNICODE CHARACTER PROPERTY SUPPORT
+

+UTF support allows the libraries to process character codepoints up to 0x10ffff +in the strings that they handle. On its own, however, it does not provide any +facilities for accessing the properties of such characters. If you want to be +able to use the pattern escapes \P, \p, and \X, which refer to Unicode +character properties, you must add +

+  --enable-unicode-properties
+
+to the configure command. This implies UTF support, even if you have +not explicitly requested it. +

+

+Including Unicode property support adds around 30K of tables to the PCRE +library. Only the general category properties such as Lu and Nd are +supported. Details are given in the +pcrepattern +documentation. +

+
JUST-IN-TIME COMPILER SUPPORT
+

+Just-in-time compiler support is included in the build by specifying +

+  --enable-jit
+
+This support is available only for certain hardware architectures. If this +option is set for an unsupported architecture, a compile time error occurs. +See the +pcrejit +documentation for a discussion of JIT usage. When JIT support is enabled, +pcregrep automatically makes use of it, unless you add +
+  --disable-pcregrep-jit
+
+to the "configure" command. +

+
CODE VALUE OF NEWLINE
+

+By default, PCRE interprets the linefeed (LF) character as indicating the end +of a line. This is the normal newline character on Unix-like systems. You can +compile PCRE to use carriage return (CR) instead, by adding +

+  --enable-newline-is-cr
+
+to the configure command. There is also a --enable-newline-is-lf option, +which explicitly specifies linefeed as the newline character. +
+
+Alternatively, you can specify that line endings are to be indicated by the two +character sequence CRLF. If you want this, add +
+  --enable-newline-is-crlf
+
+to the configure command. There is a fourth option, specified by +
+  --enable-newline-is-anycrlf
+
+which causes PCRE to recognize any of the three sequences CR, LF, or CRLF as +indicating a line ending. Finally, a fifth option, specified by +
+  --enable-newline-is-any
+
+causes PCRE to recognize any Unicode newline sequence. +

+

+Whatever line ending convention is selected when PCRE is built can be +overridden when the library functions are called. At build time it is +conventional to use the standard for your operating system. +

+
WHAT \R MATCHES
+

+By default, the sequence \R in a pattern matches any Unicode newline sequence, +whatever has been selected as the line ending sequence. If you specify +

+  --enable-bsr-anycrlf
+
+the default is changed so that \R matches only CR, LF, or CRLF. Whatever is +selected when PCRE is built can be overridden when the library functions are +called. +

+
POSIX MALLOC USAGE
+

+When the 8-bit library is called through the POSIX interface (see the +pcreposix +documentation), additional working storage is required for holding the pointers +to capturing substrings, because PCRE requires three integers per substring, +whereas the POSIX interface provides only two. If the number of expected +substrings is small, the wrapper function uses space on the stack, because this +is faster than using malloc() for each call. The default threshold above +which the stack is no longer used is 10; it can be changed by adding a setting +such as +

+  --with-posix-malloc-threshold=20
+
+to the configure command. +

+
HANDLING VERY LARGE PATTERNS
+

+Within a compiled pattern, offset values are used to point from one part to +another (for example, from an opening parenthesis to an alternation +metacharacter). By default, in the 8-bit and 16-bit libraries, two-byte values +are used for these offsets, leading to a maximum size for a compiled pattern of +around 64K. This is sufficient to handle all but the most gigantic patterns. +Nevertheless, some people do want to process truly enormous patterns, so it is +possible to compile PCRE to use three-byte or four-byte offsets by adding a +setting such as +

+  --with-link-size=3
+
+to the configure command. The value given must be 2, 3, or 4. For the +16-bit library, a value of 3 is rounded up to 4. In these libraries, using +longer offsets slows down the operation of PCRE because it has to load +additional data when handling them. For the 32-bit library the value is always +4 and cannot be overridden; the value of --with-link-size is ignored. +

+
AVOIDING EXCESSIVE STACK USAGE
+

+When matching with the pcre_exec() function, PCRE implements backtracking +by making recursive calls to an internal function called match(). In +environments where the size of the stack is limited, this can severely limit +PCRE's operation. (The Unix environment does not usually suffer from this +problem, but it may sometimes be necessary to increase the maximum stack size. +There is a discussion in the +pcrestack +documentation.) An alternative approach to recursion that uses memory from the +heap to remember data, instead of using recursive function calls, has been +implemented to work round the problem of limited stack size. If you want to +build a version of PCRE that works this way, add +

+  --disable-stack-for-recursion
+
+to the configure command. With this configuration, PCRE will use the +pcre_stack_malloc and pcre_stack_free variables to call memory +management functions. By default these point to malloc() and +free(), but you can replace the pointers so that your own functions are +used instead. +

+

+Separate functions are provided rather than using pcre_malloc and +pcre_free because the usage is very predictable: the block sizes +requested are always the same, and the blocks are always freed in reverse +order. A calling program might be able to implement optimized functions that +perform better than malloc() and free(). PCRE runs noticeably more +slowly when built in this way. This option affects only the pcre_exec() +function; it is not relevant for pcre_dfa_exec(). +

+
LIMITING PCRE RESOURCE USAGE
+

+Internally, PCRE has a function called match(), which it calls repeatedly +(sometimes recursively) when matching a pattern with the pcre_exec() +function. By controlling the maximum number of times this function may be +called during a single matching operation, a limit can be placed on the +resources used by a single call to pcre_exec(). The limit can be changed +at run time, as described in the +pcreapi +documentation. The default is 10 million, but this can be changed by adding a +setting such as +

+  --with-match-limit=500000
+
+to the configure command. This setting has no effect on the +pcre_dfa_exec() matching function. +

+

+In some environments it is desirable to limit the depth of recursive calls of +match() more strictly than the total number of calls, in order to +restrict the maximum amount of stack (or heap, if --disable-stack-for-recursion +is specified) that is used. A second limit controls this; it defaults to the +value that is set for --with-match-limit, which imposes no additional +constraints. However, you can set a lower limit by adding, for example, +

+  --with-match-limit-recursion=10000
+
+to the configure command. This value can also be overridden at run time. +

+
CREATING CHARACTER TABLES AT BUILD TIME
+

+PCRE uses fixed tables for processing characters whose code values are less +than 256. By default, PCRE is built with a set of tables that are distributed +in the file pcre_chartables.c.dist. These tables are for ASCII codes +only. If you add +

+  --enable-rebuild-chartables
+
+to the configure command, the distributed tables are no longer used. +Instead, a program called dftables is compiled and run. This outputs the +source for new set of tables, created in the default locale of your C run-time +system. (This method of replacing the tables does not work if you are cross +compiling, because dftables is run on the local host. If you need to +create alternative tables when cross compiling, you will have to do so "by +hand".) +

+
USING EBCDIC CODE
+

+PCRE assumes by default that it will run in an environment where the character +code is ASCII (or Unicode, which is a superset of ASCII). This is the case for +most computer operating systems. PCRE can, however, be compiled to run in an +EBCDIC environment by adding +

+  --enable-ebcdic
+
+to the configure command. This setting implies +--enable-rebuild-chartables. You should only use it if you know that you are in +an EBCDIC environment (for example, an IBM mainframe operating system). The +--enable-ebcdic option is incompatible with --enable-utf. +

+

+The EBCDIC character that corresponds to an ASCII LF is assumed to have the +value 0x15 by default. However, in some EBCDIC environments, 0x25 is used. In +such an environment you should use +

+  --enable-ebcdic-nl25
+
+as well as, or instead of, --enable-ebcdic. The EBCDIC character for CR has the +same value as in ASCII, namely, 0x0d. Whichever of 0x15 and 0x25 is not +chosen as LF is made to correspond to the Unicode NEL character (which, in +Unicode, is 0x85). +

+

+The options that select newline behaviour, such as --enable-newline-is-cr, +and equivalent run-time options, refer to these character values in an EBCDIC +environment. +

+
PCREGREP OPTIONS FOR COMPRESSED FILE SUPPORT
+

+By default, pcregrep reads all files as plain text. You can build it so +that it recognizes files whose names end in .gz or .bz2, and reads +them with libz or libbz2, respectively, by adding one or both of +

+  --enable-pcregrep-libz
+  --enable-pcregrep-libbz2
+
+to the configure command. These options naturally require that the +relevant libraries are installed on your system. Configuration will fail if +they are not. +

+
PCREGREP BUFFER SIZE
+

+pcregrep uses an internal buffer to hold a "window" on the file it is +scanning, in order to be able to output "before" and "after" lines when it +finds a match. The size of the buffer is controlled by a parameter whose +default value is 20K. The buffer itself is three times this size, but because +of the way it is used for holding "before" lines, the longest line that is +guaranteed to be processable is the parameter size. You can change the default +parameter value by adding, for example, +

+  --with-pcregrep-bufsize=50K
+
+to the configure command. The caller of \fPpcregrep\fP can, however, +override this value by specifying a run-time option. +

+
PCRETEST OPTION FOR LIBREADLINE SUPPORT
+

+If you add +

+  --enable-pcretest-libreadline
+
+to the configure command, pcretest is linked with the +libreadline library, and when its input is from a terminal, it reads it +using the readline() function. This provides line-editing and history +facilities. Note that libreadline is GPL-licensed, so if you distribute a +binary of pcretest linked in this way, there may be licensing issues. +

+

+Setting this option causes the -lreadline option to be added to the +pcretest build. In many operating environments with a sytem-installed +libreadline this is sufficient. However, in some environments (e.g. +if an unmodified distribution version of readline is in use), some extra +configuration may be necessary. The INSTALL file for libreadline says +this: +

+  "Readline uses the termcap functions, but does not link with the
+  termcap or curses library itself, allowing applications which link
+  with readline the to choose an appropriate library."
+
+If your environment has not been set up so that an appropriate library is +automatically included, you may need to add something like +
+  LIBS="-ncurses"
+
+immediately before the configure command. +

+
DEBUGGING WITH VALGRIND SUPPORT
+

+By adding the +

+  --enable-valgrind
+
+option to to the configure command, PCRE will use valgrind annotations +to mark certain memory regions as unaddressable. This allows it to detect +invalid memory accesses, and is mostly useful for debugging PCRE itself. +

+
CODE COVERAGE REPORTING
+

+If your C compiler is gcc, you can build a version of PCRE that can generate a +code coverage report for its test suite. To enable this, you must install +lcov version 1.6 or above. Then specify +

+  --enable-coverage
+
+to the configure command and build PCRE in the usual way. +

+

+Note that using ccache (a caching C compiler) is incompatible with code +coverage reporting. If you have configured ccache to run automatically +on your system, you must set the environment variable +

+  CCACHE_DISABLE=1
+
+before running make to build PCRE, so that ccache is not used. +

+

+When --enable-coverage is used, the following addition targets are added to the +Makefile: +

+  make coverage
+
+This creates a fresh coverage report for the PCRE test suite. It is equivalent +to running "make coverage-reset", "make coverage-baseline", "make check", and +then "make coverage-report". +
+  make coverage-reset
+
+This zeroes the coverage counters, but does nothing else. +
+  make coverage-baseline
+
+This captures baseline coverage information. +
+  make coverage-report
+
+This creates the coverage report. +
+  make coverage-clean-report
+
+This removes the generated coverage report without cleaning the coverage data +itself. +
+  make coverage-clean-data
+
+This removes the captured coverage data without removing the coverage files +created at compile time (*.gcno). +
+  make coverage-clean
+
+This cleans all coverage data including the generated coverage report. For more +information about code coverage, see the gcov and lcov +documentation. +

+
SEE ALSO
+

+pcreapi(3), pcre16, pcre32, pcre_config(3). +

+
AUTHOR
+

+Philip Hazel +
+University Computing Service +
+Cambridge CB2 3QH, England. +
+

+
REVISION
+

+Last updated: 12 May 2013 +
+Copyright © 1997-2013 University of Cambridge. +
+

+Return to the PCRE index page. +

diff --git a/pcre/doc/html/pcrecallout.html b/pcre/doc/html/pcrecallout.html new file mode 100644 index 00000000000..7233bb671e7 --- /dev/null +++ b/pcre/doc/html/pcrecallout.html @@ -0,0 +1,251 @@ + + +pcrecallout specification + + +

pcrecallout man page

+

+Return to the PCRE index page. +

+

+This page is part of the PCRE HTML documentation. It was generated automatically +from the original man page. If there is any nonsense in it, please consult the +man page, in case the conversion went wrong. +
+

+
SYNOPSIS
+

+#include <pcre.h> +

+

+int (*pcre_callout)(pcre_callout_block *); +

+

+int (*pcre16_callout)(pcre16_callout_block *); +

+

+int (*pcre32_callout)(pcre32_callout_block *); +

+
DESCRIPTION
+

+PCRE provides a feature called "callout", which is a means of temporarily +passing control to the caller of PCRE in the middle of pattern matching. The +caller of PCRE provides an external function by putting its entry point in the +global variable pcre_callout (pcre16_callout for the 16-bit +library, pcre32_callout for the 32-bit library). By default, this +variable contains NULL, which disables all calling out. +

+

+Within a regular expression, (?C) indicates the points at which the external +function is to be called. Different callout points can be identified by putting +a number less than 256 after the letter C. The default value is zero. +For example, this pattern has two callout points: +

+  (?C1)abc(?C2)def
+
+If the PCRE_AUTO_CALLOUT option bit is set when a pattern is compiled, PCRE +automatically inserts callouts, all with number 255, before each item in the +pattern. For example, if PCRE_AUTO_CALLOUT is used with the pattern +
+  A(\d{2}|--)
+
+it is processed as if it were +
+
+(?C255)A(?C255)((?C255)\d{2}(?C255)|(?C255)-(?C255)-(?C255))(?C255) +
+
+Notice that there is a callout before and after each parenthesis and +alternation bar. If the pattern contains a conditional group whose condition is +an assertion, an automatic callout is inserted immediately before the +condition. Such a callout may also be inserted explicitly, for example: +
+  (?(?C9)(?=a)ab|de)
+
+This applies only to assertion conditions (because they are themselves +independent groups). +

+

+Automatic callouts can be used for tracking the progress of pattern matching. +The +pcretest +command has an option that sets automatic callouts; when it is used, the output +indicates how the pattern is matched. This is useful information when you are +trying to optimize the performance of a particular pattern. +

+
MISSING CALLOUTS
+

+You should be aware that, because of optimizations in the way PCRE matches +patterns by default, callouts sometimes do not happen. For example, if the +pattern is +

+  ab(?C4)cd
+
+PCRE knows that any matching string must contain the letter "d". If the subject +string is "abyz", the lack of "d" means that matching doesn't ever start, and +the callout is never reached. However, with "abyd", though the result is still +no match, the callout is obeyed. +

+

+If the pattern is studied, PCRE knows the minimum length of a matching string, +and will immediately give a "no match" return without actually running a match +if the subject is not long enough, or, for unanchored patterns, if it has +been scanned far enough. +

+

+You can disable these optimizations by passing the PCRE_NO_START_OPTIMIZE +option to the matching function, or by starting the pattern with +(*NO_START_OPT). This slows down the matching process, but does ensure that +callouts such as the example above are obeyed. +

+
THE CALLOUT INTERFACE
+

+During matching, when PCRE reaches a callout point, the external function +defined by pcre_callout or pcre[16|32]_callout is called +(if it is set). This applies to both normal and DFA matching. The only +argument to the callout function is a pointer to a pcre_callout +or pcre[16|32]_callout block. +These structures contains the following fields: +

+  int           version;
+  int           callout_number;
+  int          *offset_vector;
+  const char   *subject;           (8-bit version)
+  PCRE_SPTR16   subject;           (16-bit version)
+  PCRE_SPTR32   subject;           (32-bit version)
+  int           subject_length;
+  int           start_match;
+  int           current_position;
+  int           capture_top;
+  int           capture_last;
+  void         *callout_data;
+  int           pattern_position;
+  int           next_item_length;
+  const unsigned char *mark;       (8-bit version)
+  const PCRE_UCHAR16  *mark;       (16-bit version)
+  const PCRE_UCHAR32  *mark;       (32-bit version)
+
+The version field is an integer containing the version number of the +block format. The initial version was 0; the current version is 2. The version +number will change again in future if additional fields are added, but the +intention is never to remove any of the existing fields. +

+

+The callout_number field contains the number of the callout, as compiled +into the pattern (that is, the number after ?C for manual callouts, and 255 for +automatically generated callouts). +

+

+The offset_vector field is a pointer to the vector of offsets that was +passed by the caller to the matching function. When pcre_exec() or +pcre[16|32]_exec() is used, the contents can be inspected, in order to +extract substrings that have been matched so far, in the same way as for +extracting substrings after a match has completed. For the DFA matching +functions, this field is not useful. +

+

+The subject and subject_length fields contain copies of the values +that were passed to the matching function. +

+

+The start_match field normally contains the offset within the subject at +which the current match attempt started. However, if the escape sequence \K +has been encountered, this value is changed to reflect the modified starting +point. If the pattern is not anchored, the callout function may be called +several times from the same point in the pattern for different starting points +in the subject. +

+

+The current_position field contains the offset within the subject of the +current match pointer. +

+

+When the pcre_exec() or pcre[16|32]_exec() is used, the +capture_top field contains one more than the number of the highest +numbered captured substring so far. If no substrings have been captured, the +value of capture_top is one. This is always the case when the DFA +functions are used, because they do not support captured substrings. +

+

+The capture_last field contains the number of the most recently captured +substring. However, when a recursion exits, the value reverts to what it was +outside the recursion, as do the values of all captured substrings. If no +substrings have been captured, the value of capture_last is -1. This is +always the case for the DFA matching functions. +

+

+The callout_data field contains a value that is passed to a matching +function specifically so that it can be passed back in callouts. It is passed +in the callout_data field of a pcre_extra or pcre[16|32]_extra +data structure. If no such data was passed, the value of callout_data in +a callout block is NULL. There is a description of the pcre_extra +structure in the +pcreapi +documentation. +

+

+The pattern_position field is present from version 1 of the callout +structure. It contains the offset to the next item to be matched in the pattern +string. +

+

+The next_item_length field is present from version 1 of the callout +structure. It contains the length of the next item to be matched in the pattern +string. When the callout immediately precedes an alternation bar, a closing +parenthesis, or the end of the pattern, the length is zero. When the callout +precedes an opening parenthesis, the length is that of the entire subpattern. +

+

+The pattern_position and next_item_length fields are intended to +help in distinguishing between different automatic callouts, which all have the +same callout number. However, they are set for all callouts. +

+

+The mark field is present from version 2 of the callout structure. In +callouts from pcre_exec() or pcre[16|32]_exec() it contains a +pointer to the zero-terminated name of the most recently passed (*MARK), +(*PRUNE), or (*THEN) item in the match, or NULL if no such items have been +passed. Instances of (*PRUNE) or (*THEN) without a name do not obliterate a +previous (*MARK). In callouts from the DFA matching functions this field always +contains NULL. +

+
RETURN VALUES
+

+The external callout function returns an integer to PCRE. If the value is zero, +matching proceeds as normal. If the value is greater than zero, matching fails +at the current point, but the testing of other matching possibilities goes +ahead, just as if a lookahead assertion had failed. If the value is less than +zero, the match is abandoned, the matching function returns the negative value. +

+

+Negative values should normally be chosen from the set of PCRE_ERROR_xxx +values. In particular, PCRE_ERROR_NOMATCH forces a standard "no match" failure. +The error number PCRE_ERROR_CALLOUT is reserved for use by callout functions; +it will never be used by PCRE itself. +

+
AUTHOR
+

+Philip Hazel +
+University Computing Service +
+Cambridge CB2 3QH, England. +
+

+
REVISION
+

+Last updated: 03 March 2013 +
+Copyright © 1997-2013 University of Cambridge. +
+

+Return to the PCRE index page. +

diff --git a/pcre/doc/html/pcrecompat.html b/pcre/doc/html/pcrecompat.html new file mode 100644 index 00000000000..14e20c5f4b4 --- /dev/null +++ b/pcre/doc/html/pcrecompat.html @@ -0,0 +1,229 @@ + + +pcrecompat specification + + +

pcrecompat man page

+

+Return to the PCRE index page. +

+

+This page is part of the PCRE HTML documentation. It was generated automatically +from the original man page. If there is any nonsense in it, please consult the +man page, in case the conversion went wrong. +
+
+DIFFERENCES BETWEEN PCRE AND PERL +
+

+This document describes the differences in the ways that PCRE and Perl handle +regular expressions. The differences described here are with respect to Perl +versions 5.10 and above. +

+

+1. PCRE has only a subset of Perl's Unicode support. Details of what it does +have are given in the +pcreunicode +page. +

+

+2. PCRE allows repeat quantifiers only on parenthesized assertions, but they do +not mean what you might think. For example, (?!a){3} does not assert that the +next three characters are not "a". It just asserts that the next character is +not "a" three times (in principle: PCRE optimizes this to run the assertion +just once). Perl allows repeat quantifiers on other assertions such as \b, but +these do not seem to have any use. +

+

+3. Capturing subpatterns that occur inside negative lookahead assertions are +counted, but their entries in the offsets vector are never set. Perl sometimes +(but not always) sets its numerical variables from inside negative assertions. +

+

+4. Though binary zero characters are supported in the subject string, they are +not allowed in a pattern string because it is passed as a normal C string, +terminated by zero. The escape sequence \0 can be used in the pattern to +represent a binary zero. +

+

+5. The following Perl escape sequences are not supported: \l, \u, \L, +\U, and \N when followed by a character name or Unicode value. (\N on its +own, matching a non-newline character, is supported.) In fact these are +implemented by Perl's general string-handling and are not part of its pattern +matching engine. If any of these are encountered by PCRE, an error is +generated by default. However, if the PCRE_JAVASCRIPT_COMPAT option is set, +\U and \u are interpreted as JavaScript interprets them. +

+

+6. The Perl escape sequences \p, \P, and \X are supported only if PCRE is +built with Unicode character property support. The properties that can be +tested with \p and \P are limited to the general category properties such as +Lu and Nd, script names such as Greek or Han, and the derived properties Any +and L&. PCRE does support the Cs (surrogate) property, which Perl does not; the +Perl documentation says "Because Perl hides the need for the user to understand +the internal representation of Unicode characters, there is no need to +implement the somewhat messy concept of surrogates." +

+

+7. PCRE does support the \Q...\E escape for quoting substrings. Characters in +between are treated as literals. This is slightly different from Perl in that $ +and @ are also handled as literals inside the quotes. In Perl, they cause +variable interpolation (but of course PCRE does not have variables). Note the +following examples: +

+    Pattern            PCRE matches      Perl matches
+
+    \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
+    \Qabc\$xyz\E       abc\$xyz          abc\$xyz
+    \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
+
+The \Q...\E sequence is recognized both inside and outside character classes. +

+

+8. Fairly obviously, PCRE does not support the (?{code}) and (??{code}) +constructions. However, there is support for recursive patterns. This is not +available in Perl 5.8, but it is in Perl 5.10. Also, the PCRE "callout" +feature allows an external function to be called during pattern matching. See +the +pcrecallout +documentation for details. +

+

+9. Subpatterns that are called as subroutines (whether or not recursively) are +always treated as atomic groups in PCRE. This is like Python, but unlike Perl. +Captured values that are set outside a subroutine call can be reference from +inside in PCRE, but not in Perl. There is a discussion that explains these +differences in more detail in the +section on recursion differences from Perl +in the +pcrepattern +page. +

+

+10. If any of the backtracking control verbs are used in a subpattern that is +called as a subroutine (whether or not recursively), their effect is confined +to that subpattern; it does not extend to the surrounding pattern. This is not +always the case in Perl. In particular, if (*THEN) is present in a group that +is called as a subroutine, its action is limited to that group, even if the +group does not contain any | characters. Note that such subpatterns are +processed as anchored at the point where they are tested. +

+

+11. If a pattern contains more than one backtracking control verb, the first +one that is backtracked onto acts. For example, in the pattern +A(*COMMIT)B(*PRUNE)C a failure in B triggers (*COMMIT), but a failure in C +triggers (*PRUNE). Perl's behaviour is more complex; in many cases it is the +same as PCRE, but there are examples where it differs. +

+

+12. Most backtracking verbs in assertions have their normal actions. They are +not confined to the assertion. +

+

+13. There are some differences that are concerned with the settings of captured +strings when part of a pattern is repeated. For example, matching "aba" against +the pattern /^(a(b)?)+$/ in Perl leaves $2 unset, but in PCRE it is set to "b". +

+

+14. PCRE's handling of duplicate subpattern numbers and duplicate subpattern +names is not as general as Perl's. This is a consequence of the fact the PCRE +works internally just with numbers, using an external table to translate +between numbers and names. In particular, a pattern such as (?|(?<a>A)|(?<b)B), +where the two capturing parentheses have the same number but different names, +is not supported, and causes an error at compile time. If it were allowed, it +would not be possible to distinguish which parentheses matched, because both +names map to capturing subpattern number 1. To avoid this confusing situation, +an error is given at compile time. +

+

+15. Perl recognizes comments in some places that PCRE does not, for example, +between the ( and ? at the start of a subpattern. If the /x modifier is set, +Perl allows white space between ( and ? but PCRE never does, even if the +PCRE_EXTENDED option is set. +

+

+16. In PCRE, the upper/lower case character properties Lu and Ll are not +affected when case-independent matching is specified. For example, \p{Lu} +always matches an upper case letter. I think Perl has changed in this respect; +in the release at the time of writing (5.16), \p{Lu} and \p{Ll} match all +letters, regardless of case, when case independence is specified. +

+

+17. PCRE provides some extensions to the Perl regular expression facilities. +Perl 5.10 includes new features that are not in earlier versions of Perl, some +of which (such as named parentheses) have been in PCRE for some time. This list +is with respect to Perl 5.10: +
+
+(a) Although lookbehind assertions in PCRE must match fixed length strings, +each alternative branch of a lookbehind assertion can match a different length +of string. Perl requires them all to have the same length. +
+
+(b) If PCRE_DOLLAR_ENDONLY is set and PCRE_MULTILINE is not set, the $ +meta-character matches only at the very end of the string. +
+
+(c) If PCRE_EXTRA is set, a backslash followed by a letter with no special +meaning is faulted. Otherwise, like Perl, the backslash is quietly ignored. +(Perl can be made to issue a warning.) +
+
+(d) If PCRE_UNGREEDY is set, the greediness of the repetition quantifiers is +inverted, that is, by default they are not greedy, but if followed by a +question mark they are. +
+
+(e) PCRE_ANCHORED can be used at matching time to force a pattern to be tried +only at the first matching position in the subject string. +
+
+(f) The PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART, and +PCRE_NO_AUTO_CAPTURE options for pcre_exec() have no Perl equivalents. +
+
+(g) The \R escape sequence can be restricted to match only CR, LF, or CRLF +by the PCRE_BSR_ANYCRLF option. +
+
+(h) The callout facility is PCRE-specific. +
+
+(i) The partial matching facility is PCRE-specific. +
+
+(j) Patterns compiled by PCRE can be saved and re-used at a later time, even on +different hosts that have the other endianness. However, this does not apply to +optimized data created by the just-in-time compiler. +
+
+(k) The alternative matching functions (pcre_dfa_exec(), +pcre16_dfa_exec() and pcre32_dfa_exec(),) match in a different way +and are not Perl-compatible. +
+
+(l) PCRE recognizes some special sequences such as (*CR) at the start of +a pattern that set overall options that cannot be changed within the pattern. +

+
+AUTHOR +
+

+Philip Hazel +
+University Computing Service +
+Cambridge CB2 3QH, England. +
+

+
+REVISION +
+

+Last updated: 19 March 2013 +
+Copyright © 1997-2013 University of Cambridge. +
+

+Return to the PCRE index page. +

diff --git a/pcre/doc/html/pcrecpp.html b/pcre/doc/html/pcrecpp.html new file mode 100644 index 00000000000..b7eac3a3d7a --- /dev/null +++ b/pcre/doc/html/pcrecpp.html @@ -0,0 +1,368 @@ + + +pcrecpp specification + + +

pcrecpp man page

+

+Return to the PCRE index page. +

+

+This page is part of the PCRE HTML documentation. It was generated automatically +from the original man page. If there is any nonsense in it, please consult the +man page, in case the conversion went wrong. +
+

+
SYNOPSIS OF C++ WRAPPER
+

+#include <pcrecpp.h> +

+
DESCRIPTION
+

+The C++ wrapper for PCRE was provided by Google Inc. Some additional +functionality was added by Giuseppe Maxia. This brief man page was constructed +from the notes in the pcrecpp.h file, which should be consulted for +further details. Note that the C++ wrapper supports only the original 8-bit +PCRE library. There is no 16-bit or 32-bit support at present. +

+
MATCHING INTERFACE
+

+The "FullMatch" operation checks that supplied text matches a supplied pattern +exactly. If pointer arguments are supplied, it copies matched sub-strings that +match sub-patterns into them. +

+  Example: successful match
+     pcrecpp::RE re("h.*o");
+     re.FullMatch("hello");
+
+  Example: unsuccessful match (requires full match):
+     pcrecpp::RE re("e");
+     !re.FullMatch("hello");
+
+  Example: creating a temporary RE object:
+     pcrecpp::RE("h.*o").FullMatch("hello");
+
+You can pass in a "const char*" or a "string" for "text". The examples below +tend to use a const char*. You can, as in the different examples above, store +the RE object explicitly in a variable or use a temporary RE object. The +examples below use one mode or the other arbitrarily. Either could correctly be +used for any of these examples. +

+

+You must supply extra pointer arguments to extract matched subpieces. +

+  Example: extracts "ruby" into "s" and 1234 into "i"
+     int i;
+     string s;
+     pcrecpp::RE re("(\\w+):(\\d+)");
+     re.FullMatch("ruby:1234", &s, &i);
+
+  Example: does not try to extract any extra sub-patterns
+     re.FullMatch("ruby:1234", &s);
+
+  Example: does not try to extract into NULL
+     re.FullMatch("ruby:1234", NULL, &i);
+
+  Example: integer overflow causes failure
+     !re.FullMatch("ruby:1234567891234", NULL, &i);
+
+  Example: fails because there aren't enough sub-patterns:
+     !pcrecpp::RE("\\w+:\\d+").FullMatch("ruby:1234", &s);
+
+  Example: fails because string cannot be stored in integer
+     !pcrecpp::RE("(.*)").FullMatch("ruby", &i);
+
+The provided pointer arguments can be pointers to any scalar numeric +type, or one of: +
+   string        (matched piece is copied to string)
+   StringPiece   (StringPiece is mutated to point to matched piece)
+   T             (where "bool T::ParseFrom(const char*, int)" exists)
+   NULL          (the corresponding matched sub-pattern is not copied)
+
+The function returns true iff all of the following conditions are satisfied: +
+  a. "text" matches "pattern" exactly;
+
+  b. The number of matched sub-patterns is >= number of supplied
+     pointers;
+
+  c. The "i"th argument has a suitable type for holding the
+     string captured as the "i"th sub-pattern. If you pass in
+     void * NULL for the "i"th argument, or a non-void * NULL
+     of the correct type, or pass fewer arguments than the
+     number of sub-patterns, "i"th captured sub-pattern is
+     ignored.
+
+CAVEAT: An optional sub-pattern that does not exist in the matched +string is assigned the empty string. Therefore, the following will +return false (because the empty string is not a valid number): +
+   int number;
+   pcrecpp::RE::FullMatch("abc", "[a-z]+(\\d+)?", &number);
+
+The matching interface supports at most 16 arguments per call. +If you need more, consider using the more general interface +pcrecpp::RE::DoMatch. See pcrecpp.h for the signature for +DoMatch. +

+

+NOTE: Do not use no_arg, which is used internally to mark the end of a +list of optional arguments, as a placeholder for missing arguments, as this can +lead to segfaults. +

+
QUOTING METACHARACTERS
+

+You can use the "QuoteMeta" operation to insert backslashes before all +potentially meaningful characters in a string. The returned string, used as a +regular expression, will exactly match the original string. +

+  Example:
+     string quoted = RE::QuoteMeta(unquoted);
+
+Note that it's legal to escape a character even if it has no special meaning in +a regular expression -- so this function does that. (This also makes it +identical to the perl function of the same name; see "perldoc -f quotemeta".) +For example, "1.5-2.0?" becomes "1\.5\-2\.0\?". +

+
PARTIAL MATCHES
+

+You can use the "PartialMatch" operation when you want the pattern +to match any substring of the text. +

+  Example: simple search for a string:
+     pcrecpp::RE("ell").PartialMatch("hello");
+
+  Example: find first number in a string:
+     int number;
+     pcrecpp::RE re("(\\d+)");
+     re.PartialMatch("x*100 + 20", &number);
+     assert(number == 100);
+
+

+
UTF-8 AND THE MATCHING INTERFACE
+

+By default, pattern and text are plain text, one byte per character. The UTF8 +flag, passed to the constructor, causes both pattern and string to be treated +as UTF-8 text, still a byte stream but potentially multiple bytes per +character. In practice, the text is likelier to be UTF-8 than the pattern, but +the match returned may depend on the UTF8 flag, so always use it when matching +UTF8 text. For example, "." will match one byte normally but with UTF8 set may +match up to three bytes of a multi-byte character. +

+  Example:
+     pcrecpp::RE_Options options;
+     options.set_utf8();
+     pcrecpp::RE re(utf8_pattern, options);
+     re.FullMatch(utf8_string);
+
+  Example: using the convenience function UTF8():
+     pcrecpp::RE re(utf8_pattern, pcrecpp::UTF8());
+     re.FullMatch(utf8_string);
+
+NOTE: The UTF8 flag is ignored if pcre was not configured with the +
+      --enable-utf8 flag.
+
+

+
PASSING MODIFIERS TO THE REGULAR EXPRESSION ENGINE
+

+PCRE defines some modifiers to change the behavior of the regular expression +engine. The C++ wrapper defines an auxiliary class, RE_Options, as a vehicle to +pass such modifiers to a RE class. Currently, the following modifiers are +supported: +

+   modifier              description               Perl corresponding
+
+   PCRE_CASELESS         case insensitive match      /i
+   PCRE_MULTILINE        multiple lines match        /m
+   PCRE_DOTALL           dot matches newlines        /s
+   PCRE_DOLLAR_ENDONLY   $ matches only at end       N/A
+   PCRE_EXTRA            strict escape parsing       N/A
+   PCRE_EXTENDED         ignore white spaces         /x
+   PCRE_UTF8             handles UTF8 chars          built-in
+   PCRE_UNGREEDY         reverses * and *?           N/A
+   PCRE_NO_AUTO_CAPTURE  disables capturing parens   N/A (*)
+
+(*) Both Perl and PCRE allow non capturing parentheses by means of the +"?:" modifier within the pattern itself. e.g. (?:ab|cd) does not +capture, while (ab|cd) does. +

+

+For a full account on how each modifier works, please check the +PCRE API reference page. +

+

+For each modifier, there are two member functions whose name is made +out of the modifier in lowercase, without the "PCRE_" prefix. For +instance, PCRE_CASELESS is handled by +

+  bool caseless()
+
+which returns true if the modifier is set, and +
+  RE_Options & set_caseless(bool)
+
+which sets or unsets the modifier. Moreover, PCRE_EXTRA_MATCH_LIMIT can be +accessed through the set_match_limit() and match_limit() member +functions. Setting match_limit to a non-zero value will limit the +execution of pcre to keep it from doing bad things like blowing the stack or +taking an eternity to return a result. A value of 5000 is good enough to stop +stack blowup in a 2MB thread stack. Setting match_limit to zero disables +match limiting. Alternatively, you can call match_limit_recursion() +which uses PCRE_EXTRA_MATCH_LIMIT_RECURSION to limit how much PCRE +recurses. match_limit() limits the number of matches PCRE does; +match_limit_recursion() limits the depth of internal recursion, and +therefore the amount of stack that is used. +

+

+Normally, to pass one or more modifiers to a RE class, you declare +a RE_Options object, set the appropriate options, and pass this +object to a RE constructor. Example: +

+   RE_Options opt;
+   opt.set_caseless(true);
+   if (RE("HELLO", opt).PartialMatch("hello world")) ...
+
+RE_options has two constructors. The default constructor takes no arguments and +creates a set of flags that are off by default. The optional parameter +option_flags is to facilitate transfer of legacy code from C programs. +This lets you do +
+   RE(pattern,
+     RE_Options(PCRE_CASELESS|PCRE_MULTILINE)).PartialMatch(str);
+
+However, new code is better off doing +
+   RE(pattern,
+     RE_Options().set_caseless(true).set_multiline(true))
+       .PartialMatch(str);
+
+If you are going to pass one of the most used modifiers, there are some +convenience functions that return a RE_Options class with the +appropriate modifier already set: CASELESS(), UTF8(), +MULTILINE(), DOTALL(), and EXTENDED(). +

+

+If you need to set several options at once, and you don't want to go through +the pains of declaring a RE_Options object and setting several options, there +is a parallel method that give you such ability on the fly. You can concatenate +several set_xxxxx() member functions, since each of them returns a +reference to its class object. For example, to pass PCRE_CASELESS, +PCRE_EXTENDED, and PCRE_MULTILINE to a RE with one statement, you may write: +

+   RE(" ^ xyz \\s+ .* blah$",
+     RE_Options()
+       .set_caseless(true)
+       .set_extended(true)
+       .set_multiline(true)).PartialMatch(sometext);
+
+
+

+
SCANNING TEXT INCREMENTALLY
+

+The "Consume" operation may be useful if you want to repeatedly +match regular expressions at the front of a string and skip over +them as they match. This requires use of the "StringPiece" type, +which represents a sub-range of a real string. Like RE, StringPiece +is defined in the pcrecpp namespace. +

+  Example: read lines of the form "var = value" from a string.
+     string contents = ...;                 // Fill string somehow
+     pcrecpp::StringPiece input(contents);  // Wrap in a StringPiece
+
+     string var;
+     int value;
+     pcrecpp::RE re("(\\w+) = (\\d+)\n");
+     while (re.Consume(&input, &var, &value)) {
+       ...;
+     }
+
+Each successful call to "Consume" will set "var/value", and also +advance "input" so it points past the matched text. +

+

+The "FindAndConsume" operation is similar to "Consume" but does not +anchor your match at the beginning of the string. For example, you +could extract all words from a string by repeatedly calling +

+  pcrecpp::RE("(\\w+)").FindAndConsume(&input, &word)
+
+

+
PARSING HEX/OCTAL/C-RADIX NUMBERS
+

+By default, if you pass a pointer to a numeric value, the +corresponding text is interpreted as a base-10 number. You can +instead wrap the pointer with a call to one of the operators Hex(), +Octal(), or CRadix() to interpret the text in another base. The +CRadix operator interprets C-style "0" (base-8) and "0x" (base-16) +prefixes, but defaults to base-10. +

+  Example:
+    int a, b, c, d;
+    pcrecpp::RE re("(.*) (.*) (.*) (.*)");
+    re.FullMatch("100 40 0100 0x40",
+                 pcrecpp::Octal(&a), pcrecpp::Hex(&b),
+                 pcrecpp::CRadix(&c), pcrecpp::CRadix(&d));
+
+will leave 64 in a, b, c, and d. +

+
REPLACING PARTS OF STRINGS
+

+You can replace the first match of "pattern" in "str" with "rewrite". +Within "rewrite", backslash-escaped digits (\1 to \9) can be +used to insert text matching corresponding parenthesized group +from the pattern. \0 in "rewrite" refers to the entire matching +text. For example: +

+  string s = "yabba dabba doo";
+  pcrecpp::RE("b+").Replace("d", &s);
+
+will leave "s" containing "yada dabba doo". The result is true if the pattern +matches and a replacement occurs, false otherwise. +

+

+GlobalReplace is like Replace except that it replaces all +occurrences of the pattern in the string with the rewrite. Replacements are +not subject to re-matching. For example: +

+  string s = "yabba dabba doo";
+  pcrecpp::RE("b+").GlobalReplace("d", &s);
+
+will leave "s" containing "yada dada doo". It returns the number of +replacements made. +

+

+Extract is like Replace, except that if the pattern matches, +"rewrite" is copied into "out" (an additional argument) with substitutions. +The non-matching portions of "text" are ignored. Returns true iff a match +occurred and the extraction happened successfully; if no match occurs, the +string is left unaffected. +

+
AUTHOR
+

+The C++ wrapper was contributed by Google Inc. +
+Copyright © 2007 Google Inc. +
+

+
REVISION
+

+Last updated: 08 January 2012 +
+

+Return to the PCRE index page. +

diff --git a/pcre/doc/html/pcredemo.html b/pcre/doc/html/pcredemo.html new file mode 100644 index 00000000000..894a9308261 --- /dev/null +++ b/pcre/doc/html/pcredemo.html @@ -0,0 +1,426 @@ + + +pcredemo specification + + +

pcredemo man page

+

+Return to the PCRE index page. +

+

+This page is part of the PCRE HTML documentation. It was generated automatically +from the original man page. If there is any nonsense in it, please consult the +man page, in case the conversion went wrong. +
+

    +
+
+/*************************************************
+*           PCRE DEMONSTRATION PROGRAM           *
+*************************************************/
+
+/* This is a demonstration program to illustrate the most straightforward ways
+of calling the PCRE regular expression library from a C program. See the
+pcresample documentation for a short discussion ("man pcresample" if you have
+the PCRE man pages installed).
+
+In Unix-like environments, if PCRE is installed in your standard system
+libraries, you should be able to compile this program using this command:
+
+gcc -Wall pcredemo.c -lpcre -o pcredemo
+
+If PCRE is not installed in a standard place, it is likely to be installed with
+support for the pkg-config mechanism. If you have pkg-config, you can compile
+this program using this command:
+
+gcc -Wall pcredemo.c `pkg-config --cflags --libs libpcre` -o pcredemo
+
+If you do not have pkg-config, you may have to use this:
+
+gcc -Wall pcredemo.c -I/usr/local/include -L/usr/local/lib \
+  -R/usr/local/lib -lpcre -o pcredemo
+
+Replace "/usr/local/include" and "/usr/local/lib" with wherever the include and
+library files for PCRE are installed on your system. Only some operating
+systems (e.g. Solaris) use the -R option.
+
+Building under Windows:
+
+If you want to statically link this program against a non-dll .a file, you must
+define PCRE_STATIC before including pcre.h, otherwise the pcre_malloc() and
+pcre_free() exported functions will be declared __declspec(dllimport), with
+unwanted results. So in this environment, uncomment the following line. */
+
+/* #define PCRE_STATIC */
+
+#include <stdio.h>
+#include <string.h>
+#include <pcre.h>
+
+#define OVECCOUNT 30    /* should be a multiple of 3 */
+
+
+int main(int argc, char **argv)
+{
+pcre *re;
+const char *error;
+char *pattern;
+char *subject;
+unsigned char *name_table;
+unsigned int option_bits;
+int erroffset;
+int find_all;
+int crlf_is_newline;
+int namecount;
+int name_entry_size;
+int ovector[OVECCOUNT];
+int subject_length;
+int rc, i;
+int utf8;
+
+
+/**************************************************************************
+* First, sort out the command line. There is only one possible option at  *
+* the moment, "-g" to request repeated matching to find all occurrences,  *
+* like Perl's /g option. We set the variable find_all to a non-zero value *
+* if the -g option is present. Apart from that, there must be exactly two *
+* arguments.                                                              *
+**************************************************************************/
+
+find_all = 0;
+for (i = 1; i < argc; i++)
+  {
+  if (strcmp(argv[i], "-g") == 0) find_all = 1;
+    else break;
+  }
+
+/* After the options, we require exactly two arguments, which are the pattern,
+and the subject string. */
+
+if (argc - i != 2)
+  {
+  printf("Two arguments required: a regex and a subject string\n");
+  return 1;
+  }
+
+pattern = argv[i];
+subject = argv[i+1];
+subject_length = (int)strlen(subject);
+
+
+/*************************************************************************
+* Now we are going to compile the regular expression pattern, and handle *
+* and errors that are detected.                                          *
+*************************************************************************/
+
+re = pcre_compile(
+  pattern,              /* the pattern */
+  0,                    /* default options */
+  &error,               /* for error message */
+  &erroffset,           /* for error offset */
+  NULL);                /* use default character tables */
+
+/* Compilation failed: print the error message and exit */
+
+if (re == NULL)
+  {
+  printf("PCRE compilation failed at offset %d: %s\n", erroffset, error);
+  return 1;
+  }
+
+
+/*************************************************************************
+* If the compilation succeeded, we call PCRE again, in order to do a     *
+* pattern match against the subject string. This does just ONE match. If *
+* further matching is needed, it will be done below.                     *
+*************************************************************************/
+
+rc = pcre_exec(
+  re,                   /* the compiled pattern */
+  NULL,                 /* no extra data - we didn't study the pattern */
+  subject,              /* the subject string */
+  subject_length,       /* the length of the subject */
+  0,                    /* start at offset 0 in the subject */
+  0,                    /* default options */
+  ovector,              /* output vector for substring information */
+  OVECCOUNT);           /* number of elements in the output vector */
+
+/* Matching failed: handle error cases */
+
+if (rc < 0)
+  {
+  switch(rc)
+    {
+    case PCRE_ERROR_NOMATCH: printf("No match\n"); break;
+    /*
+    Handle other special cases if you like
+    */
+    default: printf("Matching error %d\n", rc); break;
+    }
+  pcre_free(re);     /* Release memory used for the compiled pattern */
+  return 1;
+  }
+
+/* Match succeded */
+
+printf("\nMatch succeeded at offset %d\n", ovector[0]);
+
+
+/*************************************************************************
+* We have found the first match within the subject string. If the output *
+* vector wasn't big enough, say so. Then output any substrings that were *
+* captured.                                                              *
+*************************************************************************/
+
+/* The output vector wasn't big enough */
+
+if (rc == 0)
+  {
+  rc = OVECCOUNT/3;
+  printf("ovector only has room for %d captured substrings\n", rc - 1);
+  }
+
+/* Show substrings stored in the output vector by number. Obviously, in a real
+application you might want to do things other than print them. */
+
+for (i = 0; i < rc; i++)
+  {
+  char *substring_start = subject + ovector[2*i];
+  int substring_length = ovector[2*i+1] - ovector[2*i];
+  printf("%2d: %.*s\n", i, substring_length, substring_start);
+  }
+
+
+/**************************************************************************
+* That concludes the basic part of this demonstration program. We have    *
+* compiled a pattern, and performed a single match. The code that follows *
+* shows first how to access named substrings, and then how to code for    *
+* repeated matches on the same subject.                                   *
+**************************************************************************/
+
+/* See if there are any named substrings, and if so, show them by name. First
+we have to extract the count of named parentheses from the pattern. */
+
+(void)pcre_fullinfo(
+  re,                   /* the compiled pattern */
+  NULL,                 /* no extra data - we didn't study the pattern */
+  PCRE_INFO_NAMECOUNT,  /* number of named substrings */
+  &namecount);          /* where to put the answer */
+
+if (namecount <= 0) printf("No named substrings\n"); else
+  {
+  unsigned char *tabptr;
+  printf("Named substrings\n");
+
+  /* Before we can access the substrings, we must extract the table for
+  translating names to numbers, and the size of each entry in the table. */
+
+  (void)pcre_fullinfo(
+    re,                       /* the compiled pattern */
+    NULL,                     /* no extra data - we didn't study the pattern */
+    PCRE_INFO_NAMETABLE,      /* address of the table */
+    &name_table);             /* where to put the answer */
+
+  (void)pcre_fullinfo(
+    re,                       /* the compiled pattern */
+    NULL,                     /* no extra data - we didn't study the pattern */
+    PCRE_INFO_NAMEENTRYSIZE,  /* size of each entry in the table */
+    &name_entry_size);        /* where to put the answer */
+
+  /* Now we can scan the table and, for each entry, print the number, the name,
+  and the substring itself. */
+
+  tabptr = name_table;
+  for (i = 0; i < namecount; i++)
+    {
+    int n = (tabptr[0] << 8) | tabptr[1];
+    printf("(%d) %*s: %.*s\n", n, name_entry_size - 3, tabptr + 2,
+      ovector[2*n+1] - ovector[2*n], subject + ovector[2*n]);
+    tabptr += name_entry_size;
+    }
+  }
+
+
+/*************************************************************************
+* If the "-g" option was given on the command line, we want to continue  *
+* to search for additional matches in the subject string, in a similar   *
+* way to the /g option in Perl. This turns out to be trickier than you   *
+* might think because of the possibility of matching an empty string.    *
+* What happens is as follows:                                            *
+*                                                                        *
+* If the previous match was NOT for an empty string, we can just start   *
+* the next match at the end of the previous one.                         *
+*                                                                        *
+* If the previous match WAS for an empty string, we can't do that, as it *
+* would lead to an infinite loop. Instead, a special call of pcre_exec() *
+* is made with the PCRE_NOTEMPTY_ATSTART and PCRE_ANCHORED flags set.    *
+* The first of these tells PCRE that an empty string at the start of the *
+* subject is not a valid match; other possibilities must be tried. The   *
+* second flag restricts PCRE to one match attempt at the initial string  *
+* position. If this match succeeds, an alternative to the empty string   *
+* match has been found, and we can print it and proceed round the loop,  *
+* advancing by the length of whatever was found. If this match does not  *
+* succeed, we still stay in the loop, advancing by just one character.   *
+* In UTF-8 mode, which can be set by (*UTF8) in the pattern, this may be *
+* more than one byte.                                                    *
+*                                                                        *
+* However, there is a complication concerned with newlines. When the     *
+* newline convention is such that CRLF is a valid newline, we must       *
+* advance by two characters rather than one. The newline convention can  *
+* be set in the regex by (*CR), etc.; if not, we must find the default.  *
+*************************************************************************/
+
+if (!find_all)     /* Check for -g */
+  {
+  pcre_free(re);   /* Release the memory used for the compiled pattern */
+  return 0;        /* Finish unless -g was given */
+  }
+
+/* Before running the loop, check for UTF-8 and whether CRLF is a valid newline
+sequence. First, find the options with which the regex was compiled; extract
+the UTF-8 state, and mask off all but the newline options. */
+
+(void)pcre_fullinfo(re, NULL, PCRE_INFO_OPTIONS, &option_bits);
+utf8 = option_bits & PCRE_UTF8;
+option_bits &= PCRE_NEWLINE_CR|PCRE_NEWLINE_LF|PCRE_NEWLINE_CRLF|
+               PCRE_NEWLINE_ANY|PCRE_NEWLINE_ANYCRLF;
+
+/* If no newline options were set, find the default newline convention from the
+build configuration. */
+
+if (option_bits == 0)
+  {
+  int d;
+  (void)pcre_config(PCRE_CONFIG_NEWLINE, &d);
+  /* Note that these values are always the ASCII ones, even in
+  EBCDIC environments. CR = 13, NL = 10. */
+  option_bits = (d == 13)? PCRE_NEWLINE_CR :
+          (d == 10)? PCRE_NEWLINE_LF :
+          (d == (13<<8 | 10))? PCRE_NEWLINE_CRLF :
+          (d == -2)? PCRE_NEWLINE_ANYCRLF :
+          (d == -1)? PCRE_NEWLINE_ANY : 0;
+  }
+
+/* See if CRLF is a valid newline sequence. */
+
+crlf_is_newline =
+     option_bits == PCRE_NEWLINE_ANY ||
+     option_bits == PCRE_NEWLINE_CRLF ||
+     option_bits == PCRE_NEWLINE_ANYCRLF;
+
+/* Loop for second and subsequent matches */
+
+for (;;)
+  {
+  int options = 0;                 /* Normally no options */
+  int start_offset = ovector[1];   /* Start at end of previous match */
+
+  /* If the previous match was for an empty string, we are finished if we are
+  at the end of the subject. Otherwise, arrange to run another match at the
+  same point to see if a non-empty match can be found. */
+
+  if (ovector[0] == ovector[1])
+    {
+    if (ovector[0] == subject_length) break;
+    options = PCRE_NOTEMPTY_ATSTART | PCRE_ANCHORED;
+    }
+
+  /* Run the next matching operation */
+
+  rc = pcre_exec(
+    re,                   /* the compiled pattern */
+    NULL,                 /* no extra data - we didn't study the pattern */
+    subject,              /* the subject string */
+    subject_length,       /* the length of the subject */
+    start_offset,         /* starting offset in the subject */
+    options,              /* options */
+    ovector,              /* output vector for substring information */
+    OVECCOUNT);           /* number of elements in the output vector */
+
+  /* This time, a result of NOMATCH isn't an error. If the value in "options"
+  is zero, it just means we have found all possible matches, so the loop ends.
+  Otherwise, it means we have failed to find a non-empty-string match at a
+  point where there was a previous empty-string match. In this case, we do what
+  Perl does: advance the matching position by one character, and continue. We
+  do this by setting the "end of previous match" offset, because that is picked
+  up at the top of the loop as the point at which to start again.
+
+  There are two complications: (a) When CRLF is a valid newline sequence, and
+  the current position is just before it, advance by an extra byte. (b)
+  Otherwise we must ensure that we skip an entire UTF-8 character if we are in
+  UTF-8 mode. */
+
+  if (rc == PCRE_ERROR_NOMATCH)
+    {
+    if (options == 0) break;                    /* All matches found */
+    ovector[1] = start_offset + 1;              /* Advance one byte */
+    if (crlf_is_newline &&                      /* If CRLF is newline & */
+        start_offset < subject_length - 1 &&    /* we are at CRLF, */
+        subject[start_offset] == '\r' &&
+        subject[start_offset + 1] == '\n')
+      ovector[1] += 1;                          /* Advance by one more. */
+    else if (utf8)                              /* Otherwise, ensure we */
+      {                                         /* advance a whole UTF-8 */
+      while (ovector[1] < subject_length)       /* character. */
+        {
+        if ((subject[ovector[1]] & 0xc0) != 0x80) break;
+        ovector[1] += 1;
+        }
+      }
+    continue;    /* Go round the loop again */
+    }
+
+  /* Other matching errors are not recoverable. */
+
+  if (rc < 0)
+    {
+    printf("Matching error %d\n", rc);
+    pcre_free(re);    /* Release memory used for the compiled pattern */
+    return 1;
+    }
+
+  /* Match succeded */
+
+  printf("\nMatch succeeded again at offset %d\n", ovector[0]);
+
+  /* The match succeeded, but the output vector wasn't big enough. */
+
+  if (rc == 0)
+    {
+    rc = OVECCOUNT/3;
+    printf("ovector only has room for %d captured substrings\n", rc - 1);
+    }
+
+  /* As before, show substrings stored in the output vector by number, and then
+  also any named substrings. */
+
+  for (i = 0; i < rc; i++)
+    {
+    char *substring_start = subject + ovector[2*i];
+    int substring_length = ovector[2*i+1] - ovector[2*i];
+    printf("%2d: %.*s\n", i, substring_length, substring_start);
+    }
+
+  if (namecount <= 0) printf("No named substrings\n"); else
+    {
+    unsigned char *tabptr = name_table;
+    printf("Named substrings\n");
+    for (i = 0; i < namecount; i++)
+      {
+      int n = (tabptr[0] << 8) | tabptr[1];
+      printf("(%d) %*s: %.*s\n", n, name_entry_size - 3, tabptr + 2,
+        ovector[2*n+1] - ovector[2*n], subject + ovector[2*n]);
+      tabptr += name_entry_size;
+      }
+    }
+  }      /* End of loop to find second and subsequent matches */
+
+printf("\n");
+pcre_free(re);       /* Release memory used for the compiled pattern */
+return 0;
+}
+
+/* End of pcredemo.c */
+

+Return to the PCRE index page.
+

diff --git a/pcre/doc/html/pcregrep.html b/pcre/doc/html/pcregrep.html
new file mode 100644
index 00000000000..bac8f9a43fd
--- /dev/null
+++ b/pcre/doc/html/pcregrep.html
@@ -0,0 +1,757 @@
+
+
+pcregrep specification
+
+
+
pcregrep man page

+

+Return to the PCRE index page.
+

+

+This page is part of the PCRE HTML documentation. It was generated automatically
+from the original man page. If there is any nonsense in it, please consult the
+man page, in case the conversion went wrong.
+

+

+
SYNOPSIS

+

+pcregrep [options] [long options] [pattern] [path1 path2 ...]
+

+
DESCRIPTION

+

+pcregrep searches files for character patterns, in the same way as other
+grep commands do, but it uses the PCRE regular expression library to support
+patterns that are compatible with the regular expressions of Perl 5. See
+pcrepattern(3)
+for a full description of syntax and semantics of the regular expressions
+that PCRE supports.
+

+

+Patterns, whether supplied on the command line or in a separate file, are given
+without delimiters. For example:
+
+  pcregrep Thursday /etc/motd
+

+If you attempt to use delimiters (for example, by surrounding a pattern with
+slashes, as is common in Perl scripts), they are interpreted as part of the
+pattern. Quotes can of course be used to delimit patterns on the command line
+because they are interpreted by the shell, and indeed quotes are required if a
+pattern contains white space or shell metacharacters.
+

+

+The first argument that follows any option settings is treated as the single
+pattern to be matched when neither -e nor -f is present.
+Conversely, when one or both of these options are used to specify patterns, all
+arguments are treated as path names. At least one of -e, -f, or an
+argument pattern must be provided.
+

+

+If no files are specified, pcregrep reads the standard input. The
+standard input can also be referenced by a name consisting of a single hyphen.
+For example:
+
+  pcregrep some-pattern /file1 - /file3
+

+By default, each line that matches a pattern is copied to the standard
+output, and if there is more than one file, the file name is output at the
+start of each line, followed by a colon. However, there are options that can
+change how pcregrep behaves. In particular, the -M option makes it
+possible to search for patterns that span line boundaries. What defines a line
+boundary is controlled by the -N (--newline) option.
+

+

+The amount of memory used for buffering files that are being scanned is
+controlled by a parameter that can be set by the --buffer-size option.
+The default value for this parameter is specified when pcregrep is built,
+with the default default being 20K. A block of memory three times this size is
+used (to allow for buffering "before" and "after" lines). An error occurs if a
+line overflows the buffer.
+

+

+Patterns can be no longer than 8K or BUFSIZ bytes, whichever is the greater.
+BUFSIZ is defined in <stdio.h>. When there is more than one pattern
+(specified by the use of -e and/or -f), each pattern is applied to
+each line in the order in which they are defined, except that all the -e
+patterns are tried before the -f patterns.
+

+

+By default, as soon as one pattern matches a line, no further patterns are
+considered. However, if --colour (or --color) is used to colour the
+matching substrings, or if --only-matching, --file-offsets, or
+--line-offsets is used to output only the part of the line that matched
+(either shown literally, or as an offset), scanning resumes immediately
+following the match, so that further matches on the same line can be found. If
+there are multiple patterns, they are all tried on the remainder of the line,
+but patterns that follow the one that matched are not tried on the earlier part
+of the line.
+

+

+This behaviour means that the order in which multiple patterns are specified
+can affect the output when one of the above options is used. This is no longer
+the same behaviour as GNU grep, which now manages to display earlier matches
+for later patterns (as long as there is no overlap).
+

+

+Patterns that can match an empty string are accepted, but empty string
+matches are never recognized. An example is the pattern "(super)?(man)?", in
+which all components are optional. This pattern finds all occurrences of both
+"super" and "man"; the output differs from matching with "super|man" when only
+the matching substrings are being shown.
+

+

+If the LC_ALL or LC_CTYPE environment variable is set,
+pcregrep uses the value to set a locale when calling the PCRE library.
+The --locale option can be used to override this.
+

+
SUPPORT FOR COMPRESSED FILES

+

+It is possible to compile pcregrep so that it uses libz or
+libbz2 to read files whose names end in .gz or .bz2,
+respectively. You can find out whether your binary has support for one or both
+of these file types by running it with the --help option. If the
+appropriate support is not present, files are treated as plain text. The
+standard input is always so treated.
+

+
BINARY FILES

+

+By default, a file that contains a binary zero byte within the first 1024 bytes
+is identified as a binary file, and is processed specially. (GNU grep also
+identifies binary files in this manner.) See the --binary-files option
+for a means of changing the way binary files are handled.
+

+
OPTIONS

+

+The order in which some of the options appear can affect the output. For
+example, both the -h and -l options affect the printing of file
+names. Whichever comes later in the command line will be the one that takes
+effect. Similarly, except where noted below, if an option is given twice, the
+later setting is used. Numerical values for options may be followed by K or M,
+to signify multiplication by 1024 or 1024*1024 respectively.
+

+

+--
+This terminates the list of options. It is useful if the next item on the
+command line starts with a hyphen but is not an option. This allows for the
+processing of patterns and filenames that start with hyphens.
+

+

+-A number, --after-context=number
+Output number lines of context after each matching line. If filenames
+and/or line numbers are being output, a hyphen separator is used instead of a
+colon for the context lines. A line containing "--" is output between each
+group of lines, unless they are in fact contiguous in the input file. The value
+of number is expected to be relatively small. However, pcregrep
+guarantees to have up to 8K of following text available for context output.
+

+

+-a, --text
+Treat binary files as text. This is equivalent to
+--binary-files=text.
+

+

+-B number, --before-context=number
+Output number lines of context before each matching line. If filenames
+and/or line numbers are being output, a hyphen separator is used instead of a
+colon for the context lines. A line containing "--" is output between each
+group of lines, unless they are in fact contiguous in the input file. The value
+of number is expected to be relatively small. However, pcregrep
+guarantees to have up to 8K of preceding text available for context output.
+

+

+--binary-files=word
+Specify how binary files are to be processed. If the word is "binary" (the
+default), pattern matching is performed on binary files, but the only output is
+"Binary file <name> matches" when a match succeeds. If the word is "text",
+which is equivalent to the -a or --text option, binary files are
+processed in the same way as any other file. In this case, when a match
+succeeds, the output may be binary garbage, which can have nasty effects if
+sent to a terminal. If the word is "without-match", which is equivalent to the
+-I option, binary files are not processed at all; they are assumed not to
+be of interest.
+

+

+--buffer-size=number
+Set the parameter that controls how much memory is used for buffering files
+that are being scanned.
+

+

+-C number, --context=number
+Output number lines of context both before and after each matching line.
+This is equivalent to setting both -A and -B to the same value.
+

+

+-c, --count
+Do not output individual lines from the files that are being scanned; instead
+output the number of lines that would otherwise have been shown. If no lines
+are selected, the number zero is output. If several files are are being
+scanned, a count is output for each of them. However, if the
+--files-with-matches option is also used, only those files whose counts
+are greater than zero are listed. When -c is used, the -A,
+-B, and -C options are ignored.
+

+

+--colour, --color
+If this option is given without any data, it is equivalent to "--colour=auto".
+If data is required, it must be given in the same shell item, separated by an
+equals sign.
+

+

+--colour=value, --color=value
+This option specifies under what circumstances the parts of a line that matched
+a pattern should be coloured in the output. By default, the output is not
+coloured. The value (which is optional, see above) may be "never", "always", or
+"auto". In the latter case, colouring happens only if the standard output is
+connected to a terminal. More resources are used when colouring is enabled,
+because pcregrep has to search for all possible matches in a line, not
+just one, in order to colour them all.
+

+

+The colour that is used can be specified by setting the environment variable
+PCREGREP_COLOUR or PCREGREP_COLOR. The value of this variable should be a
+string of two numbers, separated by a semicolon. They are copied directly into
+the control string for setting colour on a terminal, so it is your
+responsibility to ensure that they make sense. If neither of the environment
+variables is set, the default is "1;31", which gives red.
+

+

+-D action, --devices=action
+If an input path is not a regular file or a directory, "action" specifies how
+it is to be processed. Valid values are "read" (the default) or "skip"
+(silently skip the path).
+

+

+-d action, --directories=action
+If an input path is a directory, "action" specifies how it is to be processed.
+Valid values are "read" (the default in non-Windows environments, for
+compatibility with GNU grep), "recurse" (equivalent to the -r option), or
+"skip" (silently skip the path, the default in Windows environments). In the
+"read" case, directories are read as if they were ordinary files. In some
+operating systems the effect of reading a directory like this is an immediate
+end-of-file; in others it may provoke an error.
+

+

+-e pattern, --regex=pattern, --regexp=pattern
+Specify a pattern to be matched. This option can be used multiple times in
+order to specify several patterns. It can also be used as a way of specifying a
+single pattern that starts with a hyphen. When -e is used, no argument
+pattern is taken from the command line; all arguments are treated as file
+names. There is no limit to the number of patterns. They are applied to each
+line in the order in which they are defined until one matches.
+

+

+If -f is used with -e, the command line patterns are matched first,
+followed by the patterns from the file(s), independent of the order in which
+these options are specified. Note that multiple use of -e is not the same
+as a single pattern with alternatives. For example, X|Y finds the first
+character in a line that is X or Y, whereas if the two patterns are given
+separately, with X first, pcregrep finds X if it is present, even if it
+follows Y in the line. It finds Y only if there is no X in the line. This
+matters only if you are using -o or --colo(u)r to show the part(s)
+of the line that matched.
+

+

+--exclude=pattern
+Files (but not directories) whose names match the pattern are skipped without
+being processed. This applies to all files, whether listed on the command line,
+obtained from --file-list, or by scanning a directory. The pattern is a
+PCRE regular expression, and is matched against the final component of the file
+name, not the entire path. The -F, -w, and -x options do not
+apply to this pattern. The option may be given any number of times in order to
+specify multiple patterns. If a file name matches both an --include
+and an --exclude pattern, it is excluded. There is no short form for this
+option.
+

+

+--exclude-from=filename
+Treat each non-empty line of the file as the data for an --exclude
+option. What constitutes a newline when reading the file is the operating
+system's default. The --newline option has no effect on this option. This
+option may be given more than once in order to specify a number of files to
+read.
+

+

+--exclude-dir=pattern
+Directories whose names match the pattern are skipped without being processed,
+whatever the setting of the --recursive option. This applies to all
+directories, whether listed on the command line, obtained from
+--file-list, or by scanning a parent directory. The pattern is a PCRE
+regular expression, and is matched against the final component of the directory
+name, not the entire path. The -F, -w, and -x options do not
+apply to this pattern. The option may be given any number of times in order to
+specify more than one pattern. If a directory matches both --include-dir
+and --exclude-dir, it is excluded. There is no short form for this
+option.
+

+

+-F, --fixed-strings
+Interpret each data-matching pattern as a list of fixed strings, separated by
+newlines, instead of as a regular expression. What constitutes a newline for
+this purpose is controlled by the --newline option. The -w (match
+as a word) and -x (match whole line) options can be used with -F.
+They apply to each of the fixed strings. A line is selected if any of the fixed
+strings are found in it (subject to -w or -x, if present). This
+option applies only to the patterns that are matched against the contents of
+files; it does not apply to patterns specified by any of the --include or
+--exclude options.
+

+

+-f filename, --file=filename
+Read patterns from the file, one per line, and match them against
+each line of input. What constitutes a newline when reading the file is the
+operating system's default. The --newline option has no effect on this
+option. Trailing white space is removed from each line, and blank lines are
+ignored. An empty file contains no patterns and therefore matches nothing. See
+also the comments about multiple patterns versus a single pattern with
+alternatives in the description of -e above.
+

+

+If this option is given more than once, all the specified files are
+read. A data line is output if any of the patterns match it. A filename can
+be given as "-" to refer to the standard input. When -f is used, patterns
+specified on the command line using -e may also be present; they are
+tested before the file's patterns. However, no other pattern is taken from the
+command line; all arguments are treated as the names of paths to be searched.
+

+

+--file-list=filename
+Read a list of files and/or directories that are to be scanned from the given
+file, one per line. Trailing white space is removed from each line, and blank
+lines are ignored. These paths are processed before any that are listed on the
+command line. The filename can be given as "-" to refer to the standard input.
+If --file and --file-list are both specified as "-", patterns are
+read first. This is useful only when the standard input is a terminal, from
+which further lines (the list of files) can be read after an end-of-file
+indication. If this option is given more than once, all the specified files are
+read.
+

+

+--file-offsets
+Instead of showing lines or parts of lines that match, show each match as an
+offset from the start of the file and a length, separated by a comma. In this
+mode, no context is shown. That is, the -A, -B, and -C
+options are ignored. If there is more than one match in a line, each of them is
+shown separately. This option is mutually exclusive with --line-offsets
+and --only-matching.
+

+

+-H, --with-filename
+Force the inclusion of the filename at the start of output lines when searching
+a single file. By default, the filename is not shown in this case. For matching
+lines, the filename is followed by a colon; for context lines, a hyphen
+separator is used. If a line number is also being output, it follows the file
+name.
+

+

+-h, --no-filename
+Suppress the output filenames when searching multiple files. By default,
+filenames are shown when multiple files are searched. For matching lines, the
+filename is followed by a colon; for context lines, a hyphen separator is used.
+If a line number is also being output, it follows the file name.
+

+

+--help
+Output a help message, giving brief details of the command options and file
+type support, and then exit. Anything else on the command line is
+ignored.
+

+

+-I
+Treat binary files as never matching. This is equivalent to
+--binary-files=without-match.
+

+

+-i, --ignore-case
+Ignore upper/lower case distinctions during comparisons.
+

+

+--include=pattern
+If any --include patterns are specified, the only files that are
+processed are those that match one of the patterns (and do not match an
+--exclude pattern). This option does not affect directories, but it
+applies to all files, whether listed on the command line, obtained from
+--file-list, or by scanning a directory. The pattern is a PCRE regular
+expression, and is matched against the final component of the file name, not
+the entire path. The -F, -w, and -x options do not apply to
+this pattern. The option may be given any number of times. If a file name
+matches both an --include and an --exclude pattern, it is excluded.
+There is no short form for this option.
+

+

+--include-from=filename
+Treat each non-empty line of the file as the data for an --include
+option. What constitutes a newline for this purpose is the operating system's
+default. The --newline option has no effect on this option. This option
+may be given any number of times; all the files are read.
+

+

+--include-dir=pattern
+If any --include-dir patterns are specified, the only directories that
+are processed are those that match one of the patterns (and do not match an
+--exclude-dir pattern). This applies to all directories, whether listed
+on the command line, obtained from --file-list, or by scanning a parent
+directory. The pattern is a PCRE regular expression, and is matched against the
+final component of the directory name, not the entire path. The -F,
+-w, and -x options do not apply to this pattern. The option may be
+given any number of times. If a directory matches both --include-dir and
+--exclude-dir, it is excluded. There is no short form for this option.
+

+

+-L, --files-without-match
+Instead of outputting lines from the files, just output the names of the files
+that do not contain any lines that would have been output. Each file name is
+output once, on a separate line.
+

+

+-l, --files-with-matches
+Instead of outputting lines from the files, just output the names of the files
+containing lines that would have been output. Each file name is output
+once, on a separate line. Searching normally stops as soon as a matching line
+is found in a file. However, if the -c (count) option is also used,
+matching continues in order to obtain the correct count, and those files that
+have at least one match are listed along with their counts. Using this option
+with -c is a way of suppressing the listing of files with no matches.
+

+

+--label=name
+This option supplies a name to be used for the standard input when file names
+are being output. If not supplied, "(standard input)" is used. There is no
+short form for this option.
+

+

+--line-buffered
+When this option is given, input is read and processed line by line, and the
+output is flushed after each write. By default, input is read in large chunks,
+unless pcregrep can determine that it is reading from a terminal (which
+is currently possible only in Unix-like environments). Output to terminal is
+normally automatically flushed by the operating system. This option can be
+useful when the input or output is attached to a pipe and you do not want
+pcregrep to buffer up large amounts of data. However, its use will affect
+performance, and the -M (multiline) option ceases to work.
+

+

+--line-offsets
+Instead of showing lines or parts of lines that match, show each match as a
+line number, the offset from the start of the line, and a length. The line
+number is terminated by a colon (as usual; see the -n option), and the
+offset and length are separated by a comma. In this mode, no context is shown.
+That is, the -A, -B, and -C options are ignored. If there is
+more than one match in a line, each of them is shown separately. This option is
+mutually exclusive with --file-offsets and --only-matching.
+

+

+--locale=locale-name
+This option specifies a locale to be used for pattern matching. It overrides
+the value in the LC_ALL or LC_CTYPE environment variables. If no
+locale is specified, the PCRE library's default (usually the "C" locale) is
+used. There is no short form for this option.
+

+

+--match-limit=number
+Processing some regular expression patterns can require a very large amount of
+memory, leading in some cases to a program crash if not enough is available.
+Other patterns may take a very long time to search for all possible matching
+strings. The pcre_exec() function that is called by pcregrep to do
+the matching has two parameters that can limit the resources that it uses.
+

+

+The --match-limit option provides a means of limiting resource usage
+when processing patterns that are not going to match, but which have a very
+large number of possibilities in their search trees. The classic example is a
+pattern that uses nested unlimited repeats. Internally, PCRE uses a function
+called match() which it calls repeatedly (sometimes recursively). The
+limit set by --match-limit is imposed on the number of times this
+function is called during a match, which has the effect of limiting the amount
+of backtracking that can take place.
+

+

+The --recursion-limit option is similar to --match-limit, but
+instead of limiting the total number of times that match() is called, it
+limits the depth of recursive calls, which in turn limits the amount of memory
+that can be used. The recursion depth is a smaller number than the total number
+of calls, because not all calls to match() are recursive. This limit is
+of use only if it is set smaller than --match-limit.
+

+

+There are no short forms for these options. The default settings are specified
+when the PCRE library is compiled, with the default default being 10 million.
+

+

+-M, --multiline
+Allow patterns to match more than one line. When this option is given, patterns
+may usefully contain literal newline characters and internal occurrences of ^
+and $ characters. The output for a successful match may consist of more than
+one line, the last of which is the one in which the match ended. If the matched
+string ends with a newline sequence the output ends at the end of that line.
+

+

+When this option is set, the PCRE library is called in "multiline" mode.
+There is a limit to the number of lines that can be matched, imposed by the way
+that pcregrep buffers the input file as it scans it. However,
+pcregrep ensures that at least 8K characters or the rest of the document
+(whichever is the shorter) are available for forward matching, and similarly
+the previous 8K characters (or all the previous characters, if fewer than 8K)
+are guaranteed to be available for lookbehind assertions. This option does not
+work when input is read line by line (see \fP--line-buffered\fP.)
+

+

+-N newline-type, --newline=newline-type
+The PCRE library supports five different conventions for indicating
+the ends of lines. They are the single-character sequences CR (carriage return)
+and LF (linefeed), the two-character sequence CRLF, an "anycrlf" convention,
+which recognizes any of the preceding three types, and an "any" convention, in
+which any Unicode line ending sequence is assumed to end a line. The Unicode
+sequences are the three just mentioned, plus VT (vertical tab, U+000B), FF
+(form feed, U+000C), NEL (next line, U+0085), LS (line separator, U+2028), and
+PS (paragraph separator, U+2029).
+

+

+When the PCRE library is built, a default line-ending sequence is specified.
+This is normally the standard sequence for the operating system. Unless
+otherwise specified by this option, pcregrep uses the library's default.
+The possible values for this option are CR, LF, CRLF, ANYCRLF, or ANY. This
+makes it possible to use pcregrep to scan files that have come from other
+environments without having to modify their line endings. If the data that is
+being scanned does not agree with the convention set by this option,
+pcregrep may behave in strange ways. Note that this option does not
+apply to files specified by the -f, --exclude-from, or
+--include-from options, which are expected to use the operating system's
+standard newline sequence.
+

+

+-n, --line-number
+Precede each output line by its line number in the file, followed by a colon
+for matching lines or a hyphen for context lines. If the filename is also being
+output, it precedes the line number. This option is forced if
+--line-offsets is used.
+

+

+--no-jit
+If the PCRE library is built with support for just-in-time compiling (which
+speeds up matching), pcregrep automatically makes use of this, unless it
+was explicitly disabled at build time. This option can be used to disable the
+use of JIT at run time. It is provided for testing and working round problems.
+It should never be needed in normal use.
+

+

+-o, --only-matching
+Show only the part of the line that matched a pattern instead of the whole
+line. In this mode, no context is shown. That is, the -A, -B, and
+-C options are ignored. If there is more than one match in a line, each
+of them is shown separately. If -o is combined with -v (invert the
+sense of the match to find non-matching lines), no output is generated, but the
+return code is set appropriately. If the matched portion of the line is empty,
+nothing is output unless the file name or line number are being printed, in
+which case they are shown on an otherwise empty line. This option is mutually
+exclusive with --file-offsets and --line-offsets.
+

+

+-onumber, --only-matching=number
+Show only the part of the line that matched the capturing parentheses of the
+given number. Up to 32 capturing parentheses are supported, and -o0 is
+equivalent to -o without a number. Because these options can be given
+without an argument (see above), if an argument is present, it must be given in
+the same shell item, for example, -o3 or --only-matching=2. The comments given
+for the non-argument case above also apply to this case. If the specified
+capturing parentheses do not exist in the pattern, or were not set in the
+match, nothing is output unless the file name or line number are being printed.
+

+

+If this option is given multiple times, multiple substrings are output, in the
+order the options are given. For example, -o3 -o1 -o3 causes the substrings
+matched by capturing parentheses 3 and 1 and then 3 again to be output. By
+default, there is no separator (but see the next option).
+

+

+--om-separator=text
+Specify a separating string for multiple occurrences of -o. The default
+is an empty string. Separating strings are never coloured.
+

+

+-q, --quiet
+Work quietly, that is, display nothing except error messages. The exit
+status indicates whether or not any matches were found.
+

+

+-r, --recursive
+If any given path is a directory, recursively scan the files it contains,
+taking note of any --include and --exclude settings. By default, a
+directory is read as a normal file; in some operating systems this gives an
+immediate end-of-file. This option is a shorthand for setting the -d
+option to "recurse".
+

+

+--recursion-limit=number
+See --match-limit above.
+

+

+-s, --no-messages
+Suppress error messages about non-existent or unreadable files. Such files are
+quietly skipped. However, the return code is still 2, even if matches were
+found in other files.
+

+

+-u, --utf-8
+Operate in UTF-8 mode. This option is available only if PCRE has been compiled
+with UTF-8 support. All patterns (including those for any --exclude and
+--include options) and all subject lines that are scanned must be valid
+strings of UTF-8 characters.
+

+

+-V, --version
+Write the version numbers of pcregrep and the PCRE library to the
+standard output and then exit. Anything else on the command line is
+ignored.
+

+

+-v, --invert-match
+Invert the sense of the match, so that lines which do not match any of
+the patterns are the ones that are found.
+

+

+-w, --word-regex, --word-regexp
+Force the patterns to match only whole words. This is equivalent to having \b
+at the start and end of the pattern. This option applies only to the patterns
+that are matched against the contents of files; it does not apply to patterns
+specified by any of the --include or --exclude options.
+

+

+-x, --line-regex, --line-regexp
+Force the patterns to be anchored (each must start matching at the beginning of
+a line) and in addition, require them to match entire lines. This is equivalent
+to having ^ and $ characters at the start and end of each alternative branch in
+every pattern. This option applies only to the patterns that are matched
+against the contents of files; it does not apply to patterns specified by any
+of the --include or --exclude options.
+

+
ENVIRONMENT VARIABLES

+

+The environment variables LC_ALL and LC_CTYPE are examined, in that
+order, for a locale. The first one that is set is used. This can be overridden
+by the --locale option. If no locale is set, the PCRE library's default
+(usually the "C" locale) is used.
+

+
NEWLINES

+

+The -N (--newline) option allows pcregrep to scan files with
+different newline conventions from the default. Any parts of the input files
+that are written to the standard output are copied identically, with whatever
+newline sequences they have in the input. However, the setting of this option
+does not affect the interpretation of files specified by the -f,
+--exclude-from, or --include-from options, which are assumed to use
+the operating system's standard newline sequence, nor does it affect the way in
+which pcregrep writes informational messages to the standard error and
+output streams. For these it uses the string "\n" to indicate newlines,
+relying on the C I/O library to convert this to an appropriate sequence.
+

+
OPTIONS COMPATIBILITY

+

+Many of the short and long forms of pcregrep's options are the same
+as in the GNU grep program. Any long option of the form
+--xxx-regexp (GNU terminology) is also available as --xxx-regex
+(PCRE terminology). However, the --file-list, --file-offsets,
+--include-dir, --line-offsets, --locale, --match-limit,
+-M, --multiline, -N, --newline, --om-separator,
+--recursion-limit, -u, and --utf-8 options are specific to
+pcregrep, as is the use of the --only-matching option with a
+capturing parentheses number.
+

+

+Although most of the common options work the same way, a few are different in
+pcregrep. For example, the --include option's argument is a glob
+for GNU grep, but a regular expression for pcregrep. If both the
+-c and -l options are given, GNU grep lists only file names,
+without counts, but pcregrep gives the counts.
+

+
OPTIONS WITH DATA

+

+There are four different ways in which an option with data can be specified.
+If a short form option is used, the data may follow immediately, or (with one
+exception) in the next command line item. For example:
+
+  -f/some/file
+  -f /some/file
+

+The exception is the -o option, which may appear with or without data.
+Because of this, if data is present, it must follow immediately in the same
+item, for example -o3.
+

+

+If a long form option is used, the data may appear in the same command line
+item, separated by an equals character, or (with two exceptions) it may appear
+in the next command line item. For example:
+
+  --file=/some/file
+  --file /some/file
+

+Note, however, that if you want to supply a file name beginning with ~ as data
+in a shell command, and have the shell expand ~ to a home directory, you must
+separate the file name from the option, because the shell does not treat ~
+specially unless it is at the start of an item.
+

+

+The exceptions to the above are the --colour (or --color) and
+--only-matching options, for which the data is optional. If one of these
+options does have data, it must be given in the first form, using an equals
+character. Otherwise pcregrep will assume that it has no data.
+

+
MATCHING ERRORS

+

+It is possible to supply a regular expression that takes a very long time to
+fail to match certain lines. Such patterns normally involve nested indefinite
+repeats, for example: (a+)*\d when matched against a line of a's with no final
+digit. The PCRE matching function has a resource limit that causes it to abort
+in these circumstances. If this happens, pcregrep outputs an error
+message and the line that caused the problem to the standard error stream. If
+there are more than 20 such errors, pcregrep gives up.
+

+

+The --match-limit option of pcregrep can be used to set the overall
+resource limit; there is a second option called --recursion-limit that
+sets a limit on the amount of memory (usually stack) that is used (see the
+discussion of these options above).
+

+
DIAGNOSTICS

+

+Exit status is 0 if any matches were found, 1 if no matches were found, and 2
+for syntax errors, overlong lines, non-existent or inaccessible files (even if
+matches were found in other files) or too many matching errors. Using the
+-s option to suppress error messages about inaccessible files does not
+affect the return code.
+

+
SEE ALSO

+

+pcrepattern(3), pcresyntax(3), pcretest(1).
+

+
AUTHOR

+

+Philip Hazel
+

+University Computing Service
+

+Cambridge CB2 3QH, England.
+

+

+
REVISION

+

+Last updated: 13 September 2012
+

+Copyright © 1997-2012 University of Cambridge.
+

+

+Return to the PCRE index page.
+

diff --git a/pcre/doc/html/pcrejit.html b/pcre/doc/html/pcrejit.html
new file mode 100644
index 00000000000..210f1da0262
--- /dev/null
+++ b/pcre/doc/html/pcrejit.html
@@ -0,0 +1,452 @@
+
+
+pcrejit specification
+
+
+
pcrejit man page

+

+Return to the PCRE index page.
+

+

+This page is part of the PCRE HTML documentation. It was generated automatically
+from the original man page. If there is any nonsense in it, please consult the
+man page, in case the conversion went wrong.
+

+

+
PCRE JUST-IN-TIME COMPILER SUPPORT

+

+Just-in-time compiling is a heavyweight optimization that can greatly speed up
+pattern matching. However, it comes at the cost of extra processing before the
+match is performed. Therefore, it is of most benefit when the same pattern is
+going to be matched many times. This does not necessarily mean many calls of a
+matching function; if the pattern is not anchored, matching attempts may take
+place many times at various positions in the subject, even for a single call.
+Therefore, if the subject string is very long, it may still pay to use JIT for
+one-off matches.
+

+

+JIT support applies only to the traditional Perl-compatible matching function.
+It does not apply when the DFA matching function is being used. The code for
+this support was written by Zoltan Herczeg.
+

+
8-BIT, 16-BIT AND 32-BIT SUPPORT

+

+JIT support is available for all of the 8-bit, 16-bit and 32-bit PCRE
+libraries. To keep this documentation simple, only the 8-bit interface is
+described in what follows. If you are using the 16-bit library, substitute the
+16-bit functions and 16-bit structures (for example, pcre16_jit_stack
+instead of pcre_jit_stack). If you are using the 32-bit library,
+substitute the 32-bit functions and 32-bit structures (for example,
+pcre32_jit_stack instead of pcre_jit_stack).
+

+
AVAILABILITY OF JIT SUPPORT

+

+JIT support is an optional feature of PCRE. The "configure" option --enable-jit
+(or equivalent CMake option) must be set when PCRE is built if you want to use
+JIT. The support is limited to the following hardware platforms:
+
+  ARM v5, v7, and Thumb2
+  Intel x86 32-bit and 64-bit
+  MIPS 32-bit
+  Power PC 32-bit and 64-bit
+  SPARC 32-bit (experimental)
+

+If --enable-jit is set on an unsupported platform, compilation fails.
+

+

+A program that is linked with PCRE 8.20 or later can tell if JIT support is
+available by calling pcre_config() with the PCRE_CONFIG_JIT option. The
+result is 1 when JIT is available, and 0 otherwise. However, a simple program
+does not need to check this in order to use JIT. The normal API is implemented
+in a way that falls back to the interpretive code if JIT is not available. For
+programs that need the best possible performance, there is also a "fast path"
+API that is JIT-specific.
+

+

+If your program may sometimes be linked with versions of PCRE that are older
+than 8.20, but you want to use JIT when it is available, you can test
+the values of PCRE_MAJOR and PCRE_MINOR, or the existence of a JIT macro such
+as PCRE_CONFIG_JIT, for compile-time control of your code.
+

+
SIMPLE USE OF JIT

+

+You have to do two things to make use of the JIT support in the simplest way:
+
+  (1) Call pcre_study() with the PCRE_STUDY_JIT_COMPILE option for
+      each compiled pattern, and pass the resulting pcre_extra block to
+      pcre_exec().
+
+  (2) Use pcre_free_study() to free the pcre_extra block when it is
+      no longer needed, instead of just freeing it yourself. This ensures that
+      any JIT data is also freed.
+

+For a program that may be linked with pre-8.20 versions of PCRE, you can insert
+
+  #ifndef PCRE_STUDY_JIT_COMPILE
+  #define PCRE_STUDY_JIT_COMPILE 0
+  #endif
+

+so that no option is passed to pcre_study(), and then use something like
+this to free the study data:
+
+  #ifdef PCRE_CONFIG_JIT
+      pcre_free_study(study_ptr);
+  #else
+      pcre_free(study_ptr);
+  #endif
+

+PCRE_STUDY_JIT_COMPILE requests the JIT compiler to generate code for complete
+matches. If you want to run partial matches using the PCRE_PARTIAL_HARD or
+PCRE_PARTIAL_SOFT options of pcre_exec(), you should set one or both of
+the following options in addition to, or instead of, PCRE_STUDY_JIT_COMPILE
+when you call pcre_study():
+
+  PCRE_STUDY_JIT_PARTIAL_HARD_COMPILE
+  PCRE_STUDY_JIT_PARTIAL_SOFT_COMPILE
+

+The JIT compiler generates different optimized code for each of the three
+modes (normal, soft partial, hard partial). When pcre_exec() is called,
+the appropriate code is run if it is available. Otherwise, the pattern is
+matched using interpretive code.
+

+

+In some circumstances you may need to call additional functions. These are
+described in the section entitled
+"Controlling the JIT stack"
+below.
+

+

+If JIT support is not available, PCRE_STUDY_JIT_COMPILE etc. are ignored, and
+no JIT data is created. Otherwise, the compiled pattern is passed to the JIT
+compiler, which turns it into machine code that executes much faster than the
+normal interpretive code. When pcre_exec() is passed a pcre_extra
+block containing a pointer to JIT code of the appropriate mode (normal or
+hard/soft partial), it obeys that code instead of running the interpreter. The
+result is identical, but the compiled JIT code runs much faster.
+

+

+There are some pcre_exec() options that are not supported for JIT
+execution. There are also some pattern items that JIT cannot handle. Details
+are given below. In both cases, execution automatically falls back to the
+interpretive code. If you want to know whether JIT was actually used for a
+particular match, you should arrange for a JIT callback function to be set up
+as described in the section entitled
+"Controlling the JIT stack"
+below, even if you do not need to supply a non-default JIT stack. Such a
+callback function is called whenever JIT code is about to be obeyed. If the
+execution options are not right for JIT execution, the callback function is not
+obeyed.
+

+

+If the JIT compiler finds an unsupported item, no JIT data is generated. You
+can find out if JIT execution is available after studying a pattern by calling
+pcre_fullinfo() with the PCRE_INFO_JIT option. A result of 1 means that
+JIT compilation was successful. A result of 0 means that JIT support is not
+available, or the pattern was not studied with PCRE_STUDY_JIT_COMPILE etc., or
+the JIT compiler was not able to handle the pattern.
+

+

+Once a pattern has been studied, with or without JIT, it can be used as many
+times as you like for matching different subject strings.
+

+
UNSUPPORTED OPTIONS AND PATTERN ITEMS

+

+The only pcre_exec() options that are supported for JIT execution are
+PCRE_NO_UTF8_CHECK, PCRE_NO_UTF16_CHECK, PCRE_NO_UTF32_CHECK, PCRE_NOTBOL,
+PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART, PCRE_PARTIAL_HARD, and
+PCRE_PARTIAL_SOFT.
+

+

+The only unsupported pattern items are \C (match a single data unit) when
+running in a UTF mode, and a callout immediately before an assertion condition
+in a conditional group.
+

+
RETURN VALUES FROM JIT EXECUTION

+

+When a pattern is matched using JIT execution, the return values are the same
+as those given by the interpretive pcre_exec() code, with the addition of
+one new error code: PCRE_ERROR_JIT_STACKLIMIT. This means that the memory used
+for the JIT stack was insufficient. See
+"Controlling the JIT stack"
+below for a discussion of JIT stack usage. For compatibility with the
+interpretive pcre_exec() code, no more than two-thirds of the
+ovector argument is used for passing back captured substrings.
+

+

+The error code PCRE_ERROR_MATCHLIMIT is returned by the JIT code if searching a
+very large pattern tree goes on for too long, as it is in the same circumstance
+when JIT is not used, but the details of exactly what is counted are not the
+same. The PCRE_ERROR_RECURSIONLIMIT error code is never returned by JIT
+execution.
+

+
SAVING AND RESTORING COMPILED PATTERNS

+

+The code that is generated by the JIT compiler is architecture-specific, and is
+also position dependent. For those reasons it cannot be saved (in a file or
+database) and restored later like the bytecode and other data of a compiled
+pattern. Saving and restoring compiled patterns is not something many people
+do. More detail about this facility is given in the
+pcreprecompile
+documentation. It should be possible to run pcre_study() on a saved and
+restored pattern, and thereby recreate the JIT data, but because JIT
+compilation uses significant resources, it is probably not worth doing this;
+you might as well recompile the original pattern.
+

+
CONTROLLING THE JIT STACK

+

+When the compiled JIT code runs, it needs a block of memory to use as a stack.
+By default, it uses 32K on the machine stack. However, some large or
+complicated patterns need more than this. The error PCRE_ERROR_JIT_STACKLIMIT
+is given when there is not enough stack. Three functions are provided for
+managing blocks of memory for use as JIT stacks. There is further discussion
+about the use of JIT stacks in the section entitled
+"JIT stack FAQ"
+below.
+

+

+The pcre_jit_stack_alloc() function creates a JIT stack. Its arguments
+are a starting size and a maximum size, and it returns a pointer to an opaque
+structure of type pcre_jit_stack, or NULL if there is an error. The
+pcre_jit_stack_free() function can be used to free a stack that is no
+longer needed. (For the technically minded: the address space is allocated by
+mmap or VirtualAlloc.)
+

+

+JIT uses far less memory for recursion than the interpretive code,
+and a maximum stack size of 512K to 1M should be more than enough for any
+pattern.
+

+

+The pcre_assign_jit_stack() function specifies which stack JIT code
+should use. Its arguments are as follows:
+
+  pcre_extra         *extra
+  pcre_jit_callback  callback
+  void               *data
+

+The extra argument must be the result of studying a pattern with
+PCRE_STUDY_JIT_COMPILE etc. There are three cases for the values of the other
+two options:
+
+  (1) If callback is NULL and data is NULL, an internal 32K block
+      on the machine stack is used.
+
+  (2) If callback is NULL and data is not NULL, data must be
+      a valid JIT stack, the result of calling pcre_jit_stack_alloc().
+
+  (3) If callback is not NULL, it must point to a function that is
+      called with data as an argument at the start of matching, in
+      order to set up a JIT stack. If the return from the callback
+      function is NULL, the internal 32K stack is used; otherwise the
+      return value must be a valid JIT stack, the result of calling
+      pcre_jit_stack_alloc().
+

+A callback function is obeyed whenever JIT code is about to be run; it is not
+obeyed when pcre_exec() is called with options that are incompatible for
+JIT execution. A callback function can therefore be used to determine whether a
+match operation was executed by JIT or by the interpreter.
+

+

+You may safely use the same JIT stack for more than one pattern (either by
+assigning directly or by callback), as long as the patterns are all matched
+sequentially in the same thread. In a multithread application, if you do not
+specify a JIT stack, or if you assign or pass back NULL from a callback, that
+is thread-safe, because each thread has its own machine stack. However, if you
+assign or pass back a non-NULL JIT stack, this must be a different stack for
+each thread so that the application is thread-safe.
+

+

+Strictly speaking, even more is allowed. You can assign the same non-NULL stack
+to any number of patterns as long as they are not used for matching by multiple
+threads at the same time. For example, you can assign the same stack to all
+compiled patterns, and use a global mutex in the callback to wait until the
+stack is available for use. However, this is an inefficient solution, and not
+recommended.
+

+

+This is a suggestion for how a multithreaded program that needs to set up
+non-default JIT stacks might operate:
+
+  During thread initalization
+    thread_local_var = pcre_jit_stack_alloc(...)
+
+  During thread exit
+    pcre_jit_stack_free(thread_local_var)
+
+  Use a one-line callback function
+    return thread_local_var
+

+All the functions described in this section do nothing if JIT is not available,
+and pcre_assign_jit_stack() does nothing unless the extra argument
+is non-NULL and points to a pcre_extra block that is the result of a
+successful study with PCRE_STUDY_JIT_COMPILE etc.
+

+
JIT STACK FAQ

+

+(1) Why do we need JIT stacks?
+

+

+PCRE (and JIT) is a recursive, depth-first engine, so it needs a stack where
+the local data of the current node is pushed before checking its child nodes.
+Allocating real machine stack on some platforms is difficult. For example, the
+stack chain needs to be updated every time if we extend the stack on PowerPC.
+Although it is possible, its updating time overhead decreases performance. So
+we do the recursion in memory.
+

+

+(2) Why don't we simply allocate blocks of memory with malloc()?
+

+

+Modern operating systems have a nice feature: they can reserve an address space
+instead of allocating memory. We can safely allocate memory pages inside this
+address space, so the stack could grow without moving memory data (this is
+important because of pointers). Thus we can allocate 1M address space, and use
+only a single memory page (usually 4K) if that is enough. However, we can still
+grow up to 1M anytime if needed.
+

+

+(3) Who "owns" a JIT stack?
+

+

+The owner of the stack is the user program, not the JIT studied pattern or
+anything else. The user program must ensure that if a stack is used by
+pcre_exec(), (that is, it is assigned to the pattern currently running),
+that stack must not be used by any other threads (to avoid overwriting the same
+memory area). The best practice for multithreaded programs is to allocate a
+stack for each thread, and return this stack through the JIT callback function.
+

+

+(4) When should a JIT stack be freed?
+

+

+You can free a JIT stack at any time, as long as it will not be used by
+pcre_exec() again. When you assign the stack to a pattern, only a pointer
+is set. There is no reference counting or any other magic. You can free the
+patterns and stacks in any order, anytime. Just do not call
+pcre_exec() with a pattern pointing to an already freed stack, as that
+will cause SEGFAULT. (Also, do not free a stack currently used by
+pcre_exec() in another thread). You can also replace the stack for a
+pattern at any time. You can even free the previous stack before assigning a
+replacement.
+

+

+(5) Should I allocate/free a stack every time before/after calling
+pcre_exec()?
+

+

+No, because this is too costly in terms of resources. However, you could
+implement some clever idea which release the stack if it is not used in let's
+say two minutes. The JIT callback can help to achieve this without keeping a
+list of the currently JIT studied patterns.
+

+

+(6) OK, the stack is for long term memory allocation. But what happens if a
+pattern causes stack overflow with a stack of 1M? Is that 1M kept until the
+stack is freed?
+

+

+Especially on embedded sytems, it might be a good idea to release memory
+sometimes without freeing the stack. There is no API for this at the moment.
+Probably a function call which returns with the currently allocated memory for
+any stack and another which allows releasing memory (shrinking the stack) would
+be a good idea if someone needs this.
+

+

+(7) This is too much of a headache. Isn't there any better solution for JIT
+stack handling?
+

+

+No, thanks to Windows. If POSIX threads were used everywhere, we could throw
+out this complicated API.
+

+
EXAMPLE CODE

+

+This is a single-threaded example that specifies a JIT stack without using a
+callback.
+
+  int rc;
+  int ovector[30];
+  pcre *re;
+  pcre_extra *extra;
+  pcre_jit_stack *jit_stack;
+
+  re = pcre_compile(pattern, 0, &error, &erroffset, NULL);
+  /* Check for errors */
+  extra = pcre_study(re, PCRE_STUDY_JIT_COMPILE, &error);
+  jit_stack = pcre_jit_stack_alloc(32*1024, 512*1024);
+  /* Check for error (NULL) */
+  pcre_assign_jit_stack(extra, NULL, jit_stack);
+  rc = pcre_exec(re, extra, subject, length, 0, 0, ovector, 30);
+  /* Check results */
+  pcre_free(re);
+  pcre_free_study(extra);
+  pcre_jit_stack_free(jit_stack);
+
+

+

+
JIT FAST PATH API

+

+Because the API described above falls back to interpreted execution when JIT is
+not available, it is convenient for programs that are written for general use
+in many environments. However, calling JIT via pcre_exec() does have a
+performance impact. Programs that are written for use where JIT is known to be
+available, and which need the best possible performance, can instead use a
+"fast path" API to call JIT execution directly instead of calling
+pcre_exec() (obviously only for patterns that have been successfully
+studied by JIT).
+

+

+The fast path function is called pcre_jit_exec(), and it takes exactly
+the same arguments as pcre_exec(), plus one additional argument that
+must point to a JIT stack. The JIT stack arrangements described above do not
+apply. The return values are the same as for pcre_exec().
+

+

+When you call pcre_exec(), as well as testing for invalid options, a
+number of other sanity checks are performed on the arguments. For example, if
+the subject pointer is NULL, or its length is negative, an immediate error is
+given. Also, unless PCRE_NO_UTF[8|16|32] is set, a UTF subject string is tested
+for validity. In the interests of speed, these checks do not happen on the JIT
+fast path, and if invalid data is passed, the result is undefined.
+

+

+Bypassing the sanity checks and the pcre_exec() wrapping can give
+speedups of more than 10%.
+

+
SEE ALSO

+

+pcreapi(3)
+

+
AUTHOR

+

+Philip Hazel (FAQ by Zoltan Herczeg)
+

+University Computing Service
+

+Cambridge CB2 3QH, England.
+

+

+
REVISION

+

+Last updated: 17 March 2013
+

+Copyright © 1997-2013 University of Cambridge.
+

+

+Return to the PCRE index page.
+

diff --git a/pcre/doc/html/pcrelimits.html b/pcre/doc/html/pcrelimits.html
new file mode 100644
index 00000000000..b83a801040a
--- /dev/null
+++ b/pcre/doc/html/pcrelimits.html
@@ -0,0 +1,86 @@
+
+
+pcrelimits specification
+
+
+
pcrelimits man page

+

+Return to the PCRE index page.
+

+

+This page is part of the PCRE HTML documentation. It was generated automatically
+from the original man page. If there is any nonsense in it, please consult the
+man page, in case the conversion went wrong.
+

+

+SIZE AND OTHER LIMITATIONS
+

+

+There are some size limitations in PCRE but it is hoped that they will never in
+practice be relevant.
+

+

+The maximum length of a compiled pattern is approximately 64K data units (bytes
+for the 8-bit library, 32-bit units for the 32-bit library, and 32-bit units for
+the 32-bit library) if PCRE is compiled with the default internal linkage size
+of 2 bytes. If you want to process regular expressions that are truly enormous,
+you can compile PCRE with an internal linkage size of 3 or 4 (when building the
+16-bit or 32-bit library, 3 is rounded up to 4). See the README file in
+the source distribution and the
+pcrebuild
+documentation for details. In these cases the limit is substantially larger.
+However, the speed of execution is slower.
+

+

+All values in repeating quantifiers must be less than 65536.
+

+

+There is no limit to the number of parenthesized subpatterns, but there can be
+no more than 65535 capturing subpatterns.
+

+

+There is a limit to the number of forward references to subsequent subpatterns
+of around 200,000. Repeated forward references with fixed upper limits, for
+example, (?2){0,100} when subpattern number 2 is to the right, are included in
+the count. There is no limit to the number of backward references.
+

+

+The maximum length of name for a named subpattern is 32 characters, and the
+maximum number of named subpatterns is 10000.
+

+

+The maximum length of a name in a (*MARK), (*PRUNE), (*SKIP), or (*THEN) verb
+is 255 for the 8-bit library and 65535 for the 16-bit and 32-bit library.
+

+

+The maximum length of a subject string is the largest positive number that an
+integer variable can hold. However, when using the traditional matching
+function, PCRE uses recursion to handle subpatterns and indefinite repetition.
+This means that the available stack space may limit the size of a subject
+string that can be processed by certain patterns. For a discussion of stack
+issues, see the
+pcrestack
+documentation.
+

+

+AUTHOR
+

+

+Philip Hazel
+

+University Computing Service
+

+Cambridge CB2 3QH, England.
+

+

+

+REVISION
+

+

+Last updated: 04 May 2012
+

+Copyright © 1997-2012 University of Cambridge.
+

+

+Return to the PCRE index page.
+

diff --git a/pcre/doc/html/pcrematching.html b/pcre/doc/html/pcrematching.html
new file mode 100644
index 00000000000..f1854314594
--- /dev/null
+++ b/pcre/doc/html/pcrematching.html
@@ -0,0 +1,233 @@
+
+
+pcrematching specification
+
+
+
pcrematching man page

+

+Return to the PCRE index page.
+

+

+This page is part of the PCRE HTML documentation. It was generated automatically
+from the original man page. If there is any nonsense in it, please consult the
+man page, in case the conversion went wrong.
+

+

+
PCRE MATCHING ALGORITHMS

+

+This document describes the two different algorithms that are available in PCRE
+for matching a compiled regular expression against a given subject string. The
+"standard" algorithm is the one provided by the pcre_exec(),
+pcre16_exec() and pcre32_exec() functions. These work in the same
+as as Perl's matching function, and provide a Perl-compatible matching operation.
+The just-in-time (JIT) optimization that is described in the
+pcrejit
+documentation is compatible with these functions.
+

+

+An alternative algorithm is provided by the pcre_dfa_exec(),
+pcre16_dfa_exec() and pcre32_dfa_exec() functions; they operate in
+a different way, and are not Perl-compatible. This alternative has advantages
+and disadvantages compared with the standard algorithm, and these are described
+below.
+

+

+When there is only one possible way in which a given subject string can match a
+pattern, the two algorithms give the same answer. A difference arises, however,
+when there are multiple possibilities. For example, if the pattern
+
+  ^<.*>
+

+is matched against the string
+
+  <something> <something else> <something further>
+

+there are three possible answers. The standard algorithm finds only one of
+them, whereas the alternative algorithm finds all three.
+

+
REGULAR EXPRESSIONS AS TREES

+

+The set of strings that are matched by a regular expression can be represented
+as a tree structure. An unlimited repetition in the pattern makes the tree of
+infinite size, but it is still a tree. Matching the pattern to a given subject
+string (from a given starting point) can be thought of as a search of the tree.
+There are two ways to search a tree: depth-first and breadth-first, and these
+correspond to the two matching algorithms provided by PCRE.
+

+
THE STANDARD MATCHING ALGORITHM

+

+In the terminology of Jeffrey Friedl's book "Mastering Regular
+Expressions", the standard algorithm is an "NFA algorithm". It conducts a
+depth-first search of the pattern tree. That is, it proceeds along a single
+path through the tree, checking that the subject matches what is required. When
+there is a mismatch, the algorithm tries any alternatives at the current point,
+and if they all fail, it backs up to the previous branch point in the tree, and
+tries the next alternative branch at that level. This often involves backing up
+(moving to the left) in the subject string as well. The order in which
+repetition branches are tried is controlled by the greedy or ungreedy nature of
+the quantifier.
+

+

+If a leaf node is reached, a matching string has been found, and at that point
+the algorithm stops. Thus, if there is more than one possible match, this
+algorithm returns the first one that it finds. Whether this is the shortest,
+the longest, or some intermediate length depends on the way the greedy and
+ungreedy repetition quantifiers are specified in the pattern.
+

+

+Because it ends up with a single path through the tree, it is relatively
+straightforward for this algorithm to keep track of the substrings that are
+matched by portions of the pattern in parentheses. This provides support for
+capturing parentheses and back references.
+

+
THE ALTERNATIVE MATCHING ALGORITHM

+

+This algorithm conducts a breadth-first search of the tree. Starting from the
+first matching point in the subject, it scans the subject string from left to
+right, once, character by character, and as it does this, it remembers all the
+paths through the tree that represent valid matches. In Friedl's terminology,
+this is a kind of "DFA algorithm", though it is not implemented as a
+traditional finite state machine (it keeps multiple states active
+simultaneously).
+

+

+Although the general principle of this matching algorithm is that it scans the
+subject string only once, without backtracking, there is one exception: when a
+lookaround assertion is encountered, the characters following or preceding the
+current point have to be independently inspected.
+

+

+The scan continues until either the end of the subject is reached, or there are
+no more unterminated paths. At this point, terminated paths represent the
+different matching possibilities (if there are none, the match has failed).
+Thus, if there is more than one possible match, this algorithm finds all of
+them, and in particular, it finds the longest. The matches are returned in
+decreasing order of length. There is an option to stop the algorithm after the
+first match (which is necessarily the shortest) is found.
+

+

+Note that all the matches that are found start at the same point in the
+subject. If the pattern
+
+  cat(er(pillar)?)?
+

+is matched against the string "the caterpillar catchment", the result will be
+the three strings "caterpillar", "cater", and "cat" that start at the fifth
+character of the subject. The algorithm does not automatically move on to find
+matches that start at later positions.
+

+

+There are a number of features of PCRE regular expressions that are not
+supported by the alternative matching algorithm. They are as follows:
+

+

+1. Because the algorithm finds all possible matches, the greedy or ungreedy
+nature of repetition quantifiers is not relevant. Greedy and ungreedy
+quantifiers are treated in exactly the same way. However, possessive
+quantifiers can make a difference when what follows could also match what is
+quantified, for example in a pattern like this:
+
+  ^a++\w!
+

+This pattern matches "aaab!" but not "aaa!", which would be matched by a
+non-possessive quantifier. Similarly, if an atomic group is present, it is
+matched as if it were a standalone pattern at the current point, and the
+longest match is then "locked in" for the rest of the overall pattern.
+

+

+2. When dealing with multiple paths through the tree simultaneously, it is not
+straightforward to keep track of captured substrings for the different matching
+possibilities, and PCRE's implementation of this algorithm does not attempt to
+do this. This means that no captured substrings are available.
+

+

+3. Because no substrings are captured, back references within the pattern are
+not supported, and cause errors if encountered.
+

+

+4. For the same reason, conditional expressions that use a backreference as the
+condition or test for a specific group recursion are not supported.
+

+

+5. Because many paths through the tree may be active, the \K escape sequence,
+which resets the start of the match when encountered (but may be on some paths
+and not on others), is not supported. It causes an error if encountered.
+

+

+6. Callouts are supported, but the value of the capture_top field is
+always 1, and the value of the capture_last field is always -1.
+

+

+7. The \C escape sequence, which (in the standard algorithm) always matches a
+single data unit, even in UTF-8, UTF-16 or UTF-32 modes, is not supported in
+these modes, because the alternative algorithm moves through the subject string
+one character (not data unit) at a time, for all active paths through the tree.
+

+

+8. Except for (*FAIL), the backtracking control verbs such as (*PRUNE) are not
+supported. (*FAIL) is supported, and behaves like a failing negative assertion.
+

+
ADVANTAGES OF THE ALTERNATIVE ALGORITHM

+

+Using the alternative matching algorithm provides the following advantages:
+

+

+1. All possible matches (at a single point in the subject) are automatically
+found, and in particular, the longest match is found. To find more than one
+match using the standard algorithm, you have to do kludgy things with
+callouts.
+

+

+2. Because the alternative algorithm scans the subject string just once, and
+never needs to backtrack (except for lookbehinds), it is possible to pass very
+long subject strings to the matching function in several pieces, checking for
+partial matching each time. Although it is possible to do multi-segment
+matching using the standard algorithm by retaining partially matched
+substrings, it is more complicated. The
+pcrepartial
+documentation gives details of partial matching and discusses multi-segment
+matching.
+

+
DISADVANTAGES OF THE ALTERNATIVE ALGORITHM

+

+The alternative algorithm suffers from a number of disadvantages:
+

+

+1. It is substantially slower than the standard algorithm. This is partly
+because it has to search for all possible matches, but is also because it is
+less susceptible to optimization.
+

+

+2. Capturing parentheses and back references are not supported.
+

+

+3. Although atomic groups are supported, their use does not provide the
+performance advantage that it does for the standard algorithm.
+

+
AUTHOR

+

+Philip Hazel
+

+University Computing Service
+

+Cambridge CB2 3QH, England.
+

+

+
REVISION

+

+Last updated: 08 January 2012
+

+Copyright © 1997-2012 University of Cambridge.
+

+

+Return to the PCRE index page.
+

diff --git a/pcre/doc/html/pcrepartial.html b/pcre/doc/html/pcrepartial.html
new file mode 100644
index 00000000000..98d34f0bcc2
--- /dev/null
+++ b/pcre/doc/html/pcrepartial.html
@@ -0,0 +1,499 @@
+
+
+pcrepartial specification
+
+
+
pcrepartial man page

+

+Return to the PCRE index page.
+

+

+This page is part of the PCRE HTML documentation. It was generated automatically
+from the original man page. If there is any nonsense in it, please consult the
+man page, in case the conversion went wrong.
+

+

+
PARTIAL MATCHING IN PCRE

+

+In normal use of PCRE, if the subject string that is passed to a matching
+function matches as far as it goes, but is too short to match the entire
+pattern, PCRE_ERROR_NOMATCH is returned. There are circumstances where it might
+be helpful to distinguish this case from other cases in which there is no
+match.
+

+

+Consider, for example, an application where a human is required to type in data
+for a field with specific formatting requirements. An example might be a date
+in the form ddmmmyy, defined by this pattern:
+
+  ^\d?\d(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\d\d$
+

+If the application sees the user's keystrokes one by one, and can check that
+what has been typed so far is potentially valid, it is able to raise an error
+as soon as a mistake is made, by beeping and not reflecting the character that
+has been typed, for example. This immediate feedback is likely to be a better
+user interface than a check that is delayed until the entire string has been
+entered. Partial matching can also be useful when the subject string is very
+long and is not all available at once.
+

+

+PCRE supports partial matching by means of the PCRE_PARTIAL_SOFT and
+PCRE_PARTIAL_HARD options, which can be set when calling any of the matching
+functions. For backwards compatibility, PCRE_PARTIAL is a synonym for
+PCRE_PARTIAL_SOFT. The essential difference between the two options is whether
+or not a partial match is preferred to an alternative complete match, though
+the details differ between the two types of matching function. If both options
+are set, PCRE_PARTIAL_HARD takes precedence.
+

+

+If you want to use partial matching with just-in-time optimized code, you must
+call pcre_study(), pcre16_study() or  pcre32_study() with one
+or both of these options:
+
+  PCRE_STUDY_JIT_PARTIAL_SOFT_COMPILE
+  PCRE_STUDY_JIT_PARTIAL_HARD_COMPILE
+

+PCRE_STUDY_JIT_COMPILE should also be set if you are going to run non-partial
+matches on the same pattern. If the appropriate JIT study mode has not been set
+for a match, the interpretive matching code is used.
+

+

+Setting a partial matching option disables two of PCRE's standard
+optimizations. PCRE remembers the last literal data unit in a pattern, and
+abandons matching immediately if it is not present in the subject string. This
+optimization cannot be used for a subject string that might match only
+partially. If the pattern was studied, PCRE knows the minimum length of a
+matching string, and does not bother to run the matching function on shorter
+strings. This optimization is also disabled for partial matching.
+

+
PARTIAL MATCHING USING pcre_exec() OR pcre[16|32]_exec()

+

+A partial match occurs during a call to pcre_exec() or
+pcre[16|32]_exec() when the end of the subject string is reached
+successfully, but matching cannot continue because more characters are needed.
+However, at least one character in the subject must have been inspected. This
+character need not form part of the final matched string; lookbehind assertions
+and the \K escape sequence provide ways of inspecting characters before the
+start of a matched substring. The requirement for inspecting at least one
+character exists because an empty string can always be matched; without such a
+restriction there would always be a partial match of an empty string at the end
+of the subject.
+

+

+If there are at least two slots in the offsets vector when a partial match is
+returned, the first slot is set to the offset of the earliest character that
+was inspected. For convenience, the second offset points to the end of the
+subject so that a substring can easily be identified. If there are at least
+three slots in the offsets vector, the third slot is set to the offset of the
+character where matching started.
+

+

+For the majority of patterns, the contents of the first and third slots will be
+the same. However, for patterns that contain lookbehind assertions, or begin
+with \b or \B, characters before the one where matching started may have been
+inspected while carrying out the match. For example, consider this pattern:
+
+  /(?<=abc)123/
+

+This pattern matches "123", but only if it is preceded by "abc". If the subject
+string is "xyzabc12", the first two offsets after a partial match are for the
+substring "abc12", because all these characters were inspected. However, the
+third offset is set to 6, because that is the offset where matching began.
+

+

+What happens when a partial match is identified depends on which of the two
+partial matching options are set.
+

+

+PCRE_PARTIAL_SOFT WITH pcre_exec() OR pcre[16|32]_exec()
+

+

+If PCRE_PARTIAL_SOFT is set when pcre_exec() or pcre[16|32]_exec()
+identifies a partial match, the partial match is remembered, but matching
+continues as normal, and other alternatives in the pattern are tried. If no
+complete match can be found, PCRE_ERROR_PARTIAL is returned instead of
+PCRE_ERROR_NOMATCH.
+

+

+This option is "soft" because it prefers a complete match over a partial match.
+All the various matching items in a pattern behave as if the subject string is
+potentially complete. For example, \z, \Z, and $ match at the end of the
+subject, as normal, and for \b and \B the end of the subject is treated as a
+non-alphanumeric.
+

+

+If there is more than one partial match, the first one that was found provides
+the data that is returned. Consider this pattern:
+
+  /123\w+X|dogY/
+

+If this is matched against the subject string "abc123dog", both
+alternatives fail to match, but the end of the subject is reached during
+matching, so PCRE_ERROR_PARTIAL is returned. The offsets are set to 3 and 9,
+identifying "123dog" as the first partial match that was found. (In this
+example, there are two partial matches, because "dog" on its own partially
+matches the second alternative.)
+

+

+PCRE_PARTIAL_HARD WITH pcre_exec() OR pcre[16|32]_exec()
+

+

+If PCRE_PARTIAL_HARD is set for pcre_exec() or pcre[16|32]_exec(),
+PCRE_ERROR_PARTIAL is returned as soon as a partial match is found, without
+continuing to search for possible complete matches. This option is "hard"
+because it prefers an earlier partial match over a later complete match. For
+this reason, the assumption is made that the end of the supplied subject string
+may not be the true end of the available data, and so, if \z, \Z, \b, \B,
+or $ are encountered at the end of the subject, the result is
+PCRE_ERROR_PARTIAL, provided that at least one character in the subject has
+been inspected.
+

+

+Setting PCRE_PARTIAL_HARD also affects the way UTF-8 and UTF-16
+subject strings are checked for validity. Normally, an invalid sequence
+causes the error PCRE_ERROR_BADUTF8 or PCRE_ERROR_BADUTF16. However, in the
+special case of a truncated character at the end of the subject,
+PCRE_ERROR_SHORTUTF8 or PCRE_ERROR_SHORTUTF16 is returned when
+PCRE_PARTIAL_HARD is set.
+

+

+Comparing hard and soft partial matching
+

+

+The difference between the two partial matching options can be illustrated by a
+pattern such as:
+
+  /dog(sbody)?/
+

+This matches either "dog" or "dogsbody", greedily (that is, it prefers the
+longer string if possible). If it is matched against the string "dog" with
+PCRE_PARTIAL_SOFT, it yields a complete match for "dog". However, if
+PCRE_PARTIAL_HARD is set, the result is PCRE_ERROR_PARTIAL. On the other hand,
+if the pattern is made ungreedy the result is different:
+
+  /dog(sbody)??/
+

+In this case the result is always a complete match because that is found first,
+and matching never continues after finding a complete match. It might be easier
+to follow this explanation by thinking of the two patterns like this:
+
+  /dog(sbody)?/    is the same as  /dogsbody|dog/
+  /dog(sbody)??/   is the same as  /dog|dogsbody/
+

+The second pattern will never match "dogsbody", because it will always find the
+shorter match first.
+

+
PARTIAL MATCHING USING pcre_dfa_exec() OR pcre[16|32]_dfa_exec()

+

+The DFA functions move along the subject string character by character, without
+backtracking, searching for all possible matches simultaneously. If the end of
+the subject is reached before the end of the pattern, there is the possibility
+of a partial match, again provided that at least one character has been
+inspected.
+

+

+When PCRE_PARTIAL_SOFT is set, PCRE_ERROR_PARTIAL is returned only if there
+have been no complete matches. Otherwise, the complete matches are returned.
+However, if PCRE_PARTIAL_HARD is set, a partial match takes precedence over any
+complete matches. The portion of the string that was inspected when the longest
+partial match was found is set as the first matching string, provided there are
+at least two slots in the offsets vector.
+

+

+Because the DFA functions always search for all possible matches, and there is
+no difference between greedy and ungreedy repetition, their behaviour is
+different from the standard functions when PCRE_PARTIAL_HARD is set. Consider
+the string "dog" matched against the ungreedy pattern shown above:
+
+  /dog(sbody)??/
+

+Whereas the standard functions stop as soon as they find the complete match for
+"dog", the DFA functions also find the partial match for "dogsbody", and so
+return that when PCRE_PARTIAL_HARD is set.
+

+
PARTIAL MATCHING AND WORD BOUNDARIES

+

+If a pattern ends with one of sequences \b or \B, which test for word
+boundaries, partial matching with PCRE_PARTIAL_SOFT can give counter-intuitive
+results. Consider this pattern:
+
+  /\bcat\b/
+

+This matches "cat", provided there is a word boundary at either end. If the
+subject string is "the cat", the comparison of the final "t" with a following
+character cannot take place, so a partial match is found. However, normal
+matching carries on, and \b matches at the end of the subject when the last
+character is a letter, so a complete match is found. The result, therefore, is
+not PCRE_ERROR_PARTIAL. Using PCRE_PARTIAL_HARD in this case does yield
+PCRE_ERROR_PARTIAL, because then the partial match takes precedence.
+

+
FORMERLY RESTRICTED PATTERNS

+

+For releases of PCRE prior to 8.00, because of the way certain internal
+optimizations were implemented in the pcre_exec() function, the
+PCRE_PARTIAL option (predecessor of PCRE_PARTIAL_SOFT) could not be used with
+all patterns. From release 8.00 onwards, the restrictions no longer apply, and
+partial matching with can be requested for any pattern.
+

+

+Items that were formerly restricted were repeated single characters and
+repeated metasequences. If PCRE_PARTIAL was set for a pattern that did not
+conform to the restrictions, pcre_exec() returned the error code
+PCRE_ERROR_BADPARTIAL (-13). This error code is no longer in use. The
+PCRE_INFO_OKPARTIAL call to pcre_fullinfo() to find out if a compiled
+pattern can be used for partial matching now always returns 1.
+

+
EXAMPLE OF PARTIAL MATCHING USING PCRETEST

+

+If the escape sequence \P is present in a pcretest data line, the
+PCRE_PARTIAL_SOFT option is used for the match. Here is a run of pcretest
+that uses the date example quoted above:
+
+    re> /^\d?\d(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\d\d$/
+  data> 25jun04\P
+   0: 25jun04
+   1: jun
+  data> 25dec3\P
+  Partial match: 23dec3
+  data> 3ju\P
+  Partial match: 3ju
+  data> 3juj\P
+  No match
+  data> j\P
+  No match
+

+The first data string is matched completely, so pcretest shows the
+matched substrings. The remaining four strings do not match the complete
+pattern, but the first two are partial matches. Similar output is obtained
+if DFA matching is used.
+

+

+If the escape sequence \P is present more than once in a pcretest data
+line, the PCRE_PARTIAL_HARD option is set for the match.
+

+
MULTI-SEGMENT MATCHING WITH pcre_dfa_exec() OR pcre[16|32]_dfa_exec()

+

+When a partial match has been found using a DFA matching function, it is
+possible to continue the match by providing additional subject data and calling
+the function again with the same compiled regular expression, this time setting
+the PCRE_DFA_RESTART option. You must pass the same working space as before,
+because this is where details of the previous partial match are stored. Here is
+an example using pcretest, using the \R escape sequence to set the
+PCRE_DFA_RESTART option (\D specifies the use of the DFA matching function):
+
+    re> /^\d?\d(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\d\d$/
+  data> 23ja\P\D
+  Partial match: 23ja
+  data> n05\R\D
+   0: n05
+

+The first call has "23ja" as the subject, and requests partial matching; the
+second call has "n05" as the subject for the continued (restarted) match.
+Notice that when the match is complete, only the last part is shown; PCRE does
+not retain the previously partially-matched string. It is up to the calling
+program to do that if it needs to.
+

+

+You can set the PCRE_PARTIAL_SOFT or PCRE_PARTIAL_HARD options with
+PCRE_DFA_RESTART to continue partial matching over multiple segments. This
+facility can be used to pass very long subject strings to the DFA matching
+functions.
+

+
MULTI-SEGMENT MATCHING WITH pcre_exec() OR pcre[16|32]_exec()

+

+From release 8.00, the standard matching functions can also be used to do
+multi-segment matching. Unlike the DFA functions, it is not possible to
+restart the previous match with a new segment of data. Instead, new data must
+be added to the previous subject string, and the entire match re-run, starting
+from the point where the partial match occurred. Earlier data can be discarded.
+

+

+It is best to use PCRE_PARTIAL_HARD in this situation, because it does not
+treat the end of a segment as the end of the subject when matching \z, \Z,
+\b, \B, and $. Consider an unanchored pattern that matches dates:
+
+    re> /\d?\d(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\d\d/
+  data> The date is 23ja\P\P
+  Partial match: 23ja
+

+At this stage, an application could discard the text preceding "23ja", add on
+text from the next segment, and call the matching function again. Unlike the
+DFA matching functions, the entire matching string must always be available,
+and the complete matching process occurs for each call, so more memory and more
+processing time is needed.
+

+

+Note: If the pattern contains lookbehind assertions, or \K, or starts
+with \b or \B, the string that is returned for a partial match includes
+characters that precede the start of what would be returned for a complete
+match, because it contains all the characters that were inspected during the
+partial match.
+

+
ISSUES WITH MULTI-SEGMENT MATCHING

+

+Certain types of pattern may give problems with multi-segment matching,
+whichever matching function is used.
+

+

+1. If the pattern contains a test for the beginning of a line, you need to pass
+the PCRE_NOTBOL option when the subject string for any call does start at the
+beginning of a line. There is also a PCRE_NOTEOL option, but in practice when
+doing multi-segment matching you should be using PCRE_PARTIAL_HARD, which
+includes the effect of PCRE_NOTEOL.
+

+

+2. Lookbehind assertions that have already been obeyed are catered for in the
+offsets that are returned for a partial match. However a lookbehind assertion
+later in the pattern could require even earlier characters to be inspected. You
+can handle this case by using the PCRE_INFO_MAXLOOKBEHIND option of the
+pcre_fullinfo() or pcre[16|32]_fullinfo() functions to obtain the
+length of the longest lookbehind in the pattern. This length is given in
+characters, not bytes. If you always retain at least that many characters
+before the partially matched string, all should be well. (Of course, near the
+start of the subject, fewer characters may be present; in that case all
+characters should be retained.)
+

+

+From release 8.33, there is a more accurate way of deciding which characters to
+retain. Instead of subtracting the length of the longest lookbehind from the
+earliest inspected character (offsets[0]), the match start position
+(offsets[2]) should be used, and the next match attempt started at the
+offsets[2] character by setting the startoffset argument of
+pcre_exec() or pcre_dfa_exec().
+

+

+For example, if the pattern "(?<=123)abc" is partially
+matched against the string "xx123a", the three offset values returned are 2, 6,
+and 5. This indicates that the matching process that gave a partial match
+started at offset 5, but the characters "123a" were all inspected. The maximum
+lookbehind for that pattern is 3, so taking that away from 5 shows that we need
+only keep "123a", and the next match attempt can be started at offset 3 (that
+is, at "a") when further characters have been added. When the match start is
+not the earliest inspected character, pcretest shows it explicitly:
+
+    re> "(?<=123)abc"
+  data> xx123a\P\P
+  Partial match at offset 5: 123a
+

+

+

+3. Because a partial match must always contain at least one character, what
+might be considered a partial match of an empty string actually gives a "no
+match" result. For example:
+
+    re> /c(?<=abc)x/
+  data> ab\P
+  No match
+

+If the next segment begins "cx", a match should be found, but this will only
+happen if characters from the previous segment are retained. For this reason, a
+"no match" result should be interpreted as "partial match of an empty string"
+when the pattern contains lookbehinds.
+

+

+4. Matching a subject string that is split into multiple segments may not
+always produce exactly the same result as matching over one single long string,
+especially when PCRE_PARTIAL_SOFT is used. The section "Partial Matching and
+Word Boundaries" above describes an issue that arises if the pattern ends with
+\b or \B. Another kind of difference may occur when there are multiple
+matching possibilities, because (for PCRE_PARTIAL_SOFT) a partial match result
+is given only when there are no completed matches. This means that as soon as
+the shortest match has been found, continuation to a new subject segment is no
+longer possible. Consider again this pcretest example:
+
+    re> /dog(sbody)?/
+  data> dogsb\P
+   0: dog
+  data> do\P\D
+  Partial match: do
+  data> gsb\R\P\D
+   0: g
+  data> dogsbody\D
+   0: dogsbody
+   1: dog
+

+The first data line passes the string "dogsb" to a standard matching function,
+setting the PCRE_PARTIAL_SOFT option. Although the string is a partial match
+for "dogsbody", the result is not PCRE_ERROR_PARTIAL, because the shorter
+string "dog" is a complete match. Similarly, when the subject is presented to
+a DFA matching function in several parts ("do" and "gsb" being the first two)
+the match stops when "dog" has been found, and it is not possible to continue.
+On the other hand, if "dogsbody" is presented as a single string, a DFA
+matching function finds both matches.
+

+

+Because of these problems, it is best to use PCRE_PARTIAL_HARD when matching
+multi-segment data. The example above then behaves differently:
+
+    re> /dog(sbody)?/
+  data> dogsb\P\P
+  Partial match: dogsb
+  data> do\P\D
+  Partial match: do
+  data> gsb\R\P\P\D
+  Partial match: gsb
+

+5. Patterns that contain alternatives at the top level which do not all start
+with the same pattern item may not work as expected when PCRE_DFA_RESTART is
+used. For example, consider this pattern:
+
+  1234|3789
+

+If the first part of the subject is "ABC123", a partial match of the first
+alternative is found at offset 3. There is no partial match for the second
+alternative, because such a match does not start at the same point in the
+subject string. Attempting to continue with the string "7890" does not yield a
+match because only those alternatives that match at one point in the subject
+are remembered. The problem arises because the start of the second alternative
+matches within the first alternative. There is no problem with anchored
+patterns or patterns such as:
+
+  1234|ABCD
+

+where no string can be a partial match for both alternatives. This is not a
+problem if a standard matching function is used, because the entire match has
+to be rerun each time:
+
+    re> /1234|3789/
+  data> ABC123\P\P
+  Partial match: 123
+  data> 1237890
+   0: 3789
+

+Of course, instead of using PCRE_DFA_RESTART, the same technique of re-running
+the entire match can also be used with the DFA matching functions. Another
+possibility is to work with two buffers. If a partial match at offset n
+in the first buffer is followed by "no match" when PCRE_DFA_RESTART is used on
+the second buffer, you can then try a new match starting at offset n+1 in
+the first buffer.
+

+
AUTHOR

+

+Philip Hazel
+

+University Computing Service
+

+Cambridge CB2 3QH, England.
+

+

+
REVISION

+

+Last updated: 20 February 2013
+

+Copyright © 1997-2013 University of Cambridge.
+

+

+Return to the PCRE index page.
+

diff --git a/pcre/doc/html/pcrepattern.html b/pcre/doc/html/pcrepattern.html
new file mode 100644
index 00000000000..7e837e5a87e
--- /dev/null
+++ b/pcre/doc/html/pcrepattern.html
@@ -0,0 +1,3118 @@
+
+
+pcrepattern specification
+
+
+
pcrepattern man page

+

+Return to the PCRE index page.
+

+

+This page is part of the PCRE HTML documentation. It was generated automatically
+from the original man page. If there is any nonsense in it, please consult the
+man page, in case the conversion went wrong.
+

+

+
PCRE REGULAR EXPRESSION DETAILS

+

+The syntax and semantics of the regular expressions that are supported by PCRE
+are described in detail below. There is a quick-reference syntax summary in the
+pcresyntax
+page. PCRE tries to match Perl syntax and semantics as closely as it can. PCRE
+also supports some alternative regular expression syntax (which does not
+conflict with the Perl syntax) in order to provide some compatibility with
+regular expressions in Python, .NET, and Oniguruma.
+

+

+Perl's regular expressions are described in its own documentation, and
+regular expressions in general are covered in a number of books, some of which
+have copious examples. Jeffrey Friedl's "Mastering Regular Expressions",
+published by O'Reilly, covers regular expressions in great detail. This
+description of PCRE's regular expressions is intended as reference material.
+

+

+This document discusses the patterns that are supported by PCRE when one its
+main matching functions, pcre_exec() (8-bit) or pcre[16|32]_exec()
+(16- or 32-bit), is used. PCRE also has alternative matching functions,
+pcre_dfa_exec() and pcre[16|32_dfa_exec(), which match using a
+different algorithm that is not Perl-compatible. Some of the features discussed
+below are not available when DFA matching is used. The advantages and
+disadvantages of the alternative functions, and how they differ from the normal
+functions, are discussed in the
+pcrematching
+page.
+

+
SPECIAL START-OF-PATTERN ITEMS

+

+A number of options that can be passed to pcre_compile() can also be set
+by special items at the start of a pattern. These are not Perl-compatible, but
+are provided to make these options accessible to pattern writers who are not
+able to change the program that processes the pattern. Any number of these
+items may appear, but they must all be together right at the start of the
+pattern string, and the letters must be in upper case.
+

+

+UTF support
+

+

+The original operation of PCRE was on strings of one-byte characters. However,
+there is now also support for UTF-8 strings in the original library, an
+extra library that supports 16-bit and UTF-16 character strings, and a
+third library that supports 32-bit and UTF-32 character strings. To use these
+features, PCRE must be built to include appropriate support. When using UTF
+strings you must either call the compiling function with the PCRE_UTF8,
+PCRE_UTF16, or PCRE_UTF32 option, or the pattern must start with one of
+these special sequences:
+
+  (*UTF8)
+  (*UTF16)
+  (*UTF32)
+  (*UTF)
+

+(*UTF) is a generic sequence that can be used with any of the libraries.
+Starting a pattern with such a sequence is equivalent to setting the relevant
+option. How setting a UTF mode affects pattern matching is mentioned in several
+places below. There is also a summary of features in the
+pcreunicode
+page.
+

+

+Some applications that allow their users to supply patterns may wish to
+restrict them to non-UTF data for security reasons. If the PCRE_NEVER_UTF
+option is set at compile time, (*UTF) etc. are not allowed, and their
+appearance causes an error.
+

+

+Unicode property support
+

+

+Another special sequence that may appear at the start of a pattern is
+
+  (*UCP)
+

+This has the same effect as setting the PCRE_UCP option: it causes sequences
+such as \d and \w to use Unicode properties to determine character types,
+instead of recognizing only characters with codes less than 128 via a lookup
+table.
+

+

+Disabling start-up optimizations
+

+

+If a pattern starts with (*NO_START_OPT), it has the same effect as setting the
+PCRE_NO_START_OPTIMIZE option either at compile or matching time.
+

+

+Newline conventions
+

+

+PCRE supports five different conventions for indicating line breaks in
+strings: a single CR (carriage return) character, a single LF (linefeed)
+character, the two-character sequence CRLF, any of the three preceding, or any
+Unicode newline sequence. The
+pcreapi
+page has
+further discussion
+about newlines, and shows how to set the newline convention in the
+options arguments for the compiling and matching functions.
+

+

+It is also possible to specify a newline convention by starting a pattern
+string with one of the following five sequences:
+
+  (*CR)        carriage return
+  (*LF)        linefeed
+  (*CRLF)      carriage return, followed by linefeed
+  (*ANYCRLF)   any of the three above
+  (*ANY)       all Unicode newline sequences
+

+These override the default and the options given to the compiling function. For
+example, on a Unix system where LF is the default newline sequence, the pattern
+
+  (*CR)a.b
+

+changes the convention to CR. That pattern matches "a\nb" because LF is no
+longer a newline. If more than one of these settings is present, the last one
+is used.
+

+

+The newline convention affects where the circumflex and dollar assertions are
+true. It also affects the interpretation of the dot metacharacter when
+PCRE_DOTALL is not set, and the behaviour of \N. However, it does not affect
+what the \R escape sequence matches. By default, this is any Unicode newline
+sequence, for Perl compatibility. However, this can be changed; see the
+description of \R in the section entitled
+"Newline sequences"
+below. A change of \R setting can be combined with a change of newline
+convention.
+

+

+Setting match and recursion limits
+

+

+The caller of pcre_exec() can set a limit on the number of times the
+internal match() function is called and on the maximum depth of
+recursive calls. These facilities are provided to catch runaway matches that
+are provoked by patterns with huge matching trees (a typical example is a
+pattern with nested unlimited repeats) and to avoid running out of system stack
+by too much recursion. When one of these limits is reached, pcre_exec()
+gives an error return. The limits can also be set by items at the start of the
+pattern of the form
+
+  (*LIMIT_MATCH=d)
+  (*LIMIT_RECURSION=d)
+

+where d is any number of decimal digits. However, the value of the setting must
+be less than the value set by the caller of pcre_exec() for it to have
+any effect. In other words, the pattern writer can lower the limit set by the
+programmer, but not raise it. If there is more than one setting of one of these
+limits, the lower value is used.
+

+
EBCDIC CHARACTER CODES

+

+PCRE can be compiled to run in an environment that uses EBCDIC as its character
+code rather than ASCII or Unicode (typically a mainframe system). In the
+sections below, character code values are ASCII or Unicode; in an EBCDIC
+environment these characters may have different code values, and there are no
+code points greater than 255.
+

+
CHARACTERS AND METACHARACTERS

+

+A regular expression is a pattern that is matched against a subject string from
+left to right. Most characters stand for themselves in a pattern, and match the
+corresponding characters in the subject. As a trivial example, the pattern
+
+  The quick brown fox
+

+matches a portion of a subject string that is identical to itself. When
+caseless matching is specified (the PCRE_CASELESS option), letters are matched
+independently of case. In a UTF mode, PCRE always understands the concept of
+case for characters whose values are less than 128, so caseless matching is
+always possible. For characters with higher values, the concept of case is
+supported if PCRE is compiled with Unicode property support, but not otherwise.
+If you want to use caseless matching for characters 128 and above, you must
+ensure that PCRE is compiled with Unicode property support as well as with
+UTF support.
+

+

+The power of regular expressions comes from the ability to include alternatives
+and repetitions in the pattern. These are encoded in the pattern by the use of
+metacharacters, which do not stand for themselves but instead are
+interpreted in some special way.
+

+

+There are two different sets of metacharacters: those that are recognized
+anywhere in the pattern except within square brackets, and those that are
+recognized within square brackets. Outside square brackets, the metacharacters
+are as follows:
+
+  \      general escape character with several uses
+  ^      assert start of string (or line, in multiline mode)
+  $      assert end of string (or line, in multiline mode)
+  .      match any character except newline (by default)
+  [      start character class definition
+  |      start of alternative branch
+  (      start subpattern
+  )      end subpattern
+  ?      extends the meaning of (
+         also 0 or 1 quantifier
+         also quantifier minimizer
+  *      0 or more quantifier
+  +      1 or more quantifier
+         also "possessive quantifier"
+  {      start min/max quantifier
+

+Part of a pattern that is in square brackets is called a "character class". In
+a character class the only metacharacters are:
+
+  \      general escape character
+  ^      negate the class, but only if the first character
+  -      indicates character range
+  [      POSIX character class (only if followed by POSIX syntax)
+  ]      terminates the character class
+

+The following sections describe the use of each of the metacharacters.
+

+
BACKSLASH

+

+The backslash character has several uses. Firstly, if it is followed by a
+character that is not a number or a letter, it takes away any special meaning
+that character may have. This use of backslash as an escape character applies
+both inside and outside character classes.
+

+

+For example, if you want to match a * character, you write \* in the pattern.
+This escaping action applies whether or not the following character would
+otherwise be interpreted as a metacharacter, so it is always safe to precede a
+non-alphanumeric with backslash to specify that it stands for itself. In
+particular, if you want to match a backslash, you write \\.
+

+

+In a UTF mode, only ASCII numbers and letters have any special meaning after a
+backslash. All other characters (in particular, those whose codepoints are
+greater than 127) are treated as literals.
+

+

+If a pattern is compiled with the PCRE_EXTENDED option, white space in the
+pattern (other than in a character class) and characters between a # outside
+a character class and the next newline are ignored. An escaping backslash can
+be used to include a white space or # character as part of the pattern.
+

+

+If you want to remove the special meaning from a sequence of characters, you
+can do so by putting them between \Q and \E. This is different from Perl in
+that $ and @ are handled as literals in \Q...\E sequences in PCRE, whereas in
+Perl, $ and @ cause variable interpolation. Note the following examples:
+
+  Pattern            PCRE matches   Perl matches
+
+  \Qabc$xyz\E        abc$xyz        abc followed by the contents of $xyz
+  \Qabc\$xyz\E       abc\$xyz       abc\$xyz
+  \Qabc\E\$\Qxyz\E   abc$xyz        abc$xyz
+

+The \Q...\E sequence is recognized both inside and outside character classes.
+An isolated \E that is not preceded by \Q is ignored. If \Q is not followed
+by \E later in the pattern, the literal interpretation continues to the end of
+the pattern (that is, \E is assumed at the end). If the isolated \Q is inside
+a character class, this causes an error, because the character class is not
+terminated.
+

+

+Non-printing characters
+

+

+A second use of backslash provides a way of encoding non-printing characters
+in patterns in a visible manner. There is no restriction on the appearance of
+non-printing characters, apart from the binary zero that terminates a pattern,
+but when a pattern is being prepared by text editing, it is often easier to use
+one of the following escape sequences than the binary character it represents:
+
+  \a        alarm, that is, the BEL character (hex 07)
+  \cx       "control-x", where x is any ASCII character
+  \e        escape (hex 1B)
+  \f        form feed (hex 0C)
+  \n        linefeed (hex 0A)
+  \r        carriage return (hex 0D)
+  \t        tab (hex 09)
+  \ddd      character with octal code ddd, or back reference
+  \xhh      character with hex code hh
+  \x{hhh..} character with hex code hhh.. (non-JavaScript mode)
+  \uhhhh    character with hex code hhhh (JavaScript mode only)
+

+The precise effect of \cx on ASCII characters is as follows: if x is a lower
+case letter, it is converted to upper case. Then bit 6 of the character (hex
+40) is inverted. Thus \cA to \cZ become hex 01 to hex 1A (A is 41, Z is 5A),
+but \c{ becomes hex 3B ({ is 7B), and \c; becomes hex 7B (; is 3B). If the
+data item (byte or 16-bit value) following \c has a value greater than 127, a
+compile-time error occurs. This locks out non-ASCII characters in all modes.
+

+

+The \c facility was designed for use with ASCII characters, but with the
+extension to Unicode it is even less useful than it once was. It is, however,
+recognized when PCRE is compiled in EBCDIC mode, where data items are always
+bytes. In this mode, all values are valid after \c. If the next character is a
+lower case letter, it is converted to upper case. Then the 0xc0 bits of the
+byte are inverted. Thus \cA becomes hex 01, as in ASCII (A is C1), but because
+the EBCDIC letters are disjoint, \cZ becomes hex 29 (Z is E9), and other
+characters also generate different values.
+

+

+By default, after \x, from zero to two hexadecimal digits are read (letters
+can be in upper or lower case). Any number of hexadecimal digits may appear
+between \x{ and }, but the character code is constrained as follows:
+
+  8-bit non-UTF mode    less than 0x100
+  8-bit UTF-8 mode      less than 0x10ffff and a valid codepoint
+  16-bit non-UTF mode   less than 0x10000
+  16-bit UTF-16 mode    less than 0x10ffff and a valid codepoint
+  32-bit non-UTF mode   less than 0x80000000
+  32-bit UTF-32 mode    less than 0x10ffff and a valid codepoint
+

+Invalid Unicode codepoints are the range 0xd800 to 0xdfff (the so-called
+"surrogate" codepoints), and 0xffef.
+

+

+If characters other than hexadecimal digits appear between \x{ and }, or if
+there is no terminating }, this form of escape is not recognized. Instead, the
+initial \x will be interpreted as a basic hexadecimal escape, with no
+following digits, giving a character whose value is zero.
+

+

+If the PCRE_JAVASCRIPT_COMPAT option is set, the interpretation of \x is
+as just described only when it is followed by two hexadecimal digits.
+Otherwise, it matches a literal "x" character. In JavaScript mode, support for
+code points greater than 256 is provided by \u, which must be followed by
+four hexadecimal digits; otherwise it matches a literal "u" character.
+Character codes specified by \u in JavaScript mode are constrained in the same
+was as those specified by \x in non-JavaScript mode.
+

+

+Characters whose value is less than 256 can be defined by either of the two
+syntaxes for \x (or by \u in JavaScript mode). There is no difference in the
+way they are handled. For example, \xdc is exactly the same as \x{dc} (or
+\u00dc in JavaScript mode).
+

+

+After \0 up to two further octal digits are read. If there are fewer than two
+digits, just those that are present are used. Thus the sequence \0\x\07
+specifies two binary zeros followed by a BEL character (code value 7). Make
+sure you supply two digits after the initial zero if the pattern character that
+follows is itself an octal digit.
+

+

+The handling of a backslash followed by a digit other than 0 is complicated.
+Outside a character class, PCRE reads it and any following digits as a decimal
+number. If the number is less than 10, or if there have been at least that many
+previous capturing left parentheses in the expression, the entire sequence is
+taken as a back reference. A description of how this works is given
+later,
+following the discussion of
+parenthesized subpatterns.
+

+

+Inside a character class, or if the decimal number is greater than 9 and there
+have not been that many capturing subpatterns, PCRE re-reads up to three octal
+digits following the backslash, and uses them to generate a data character. Any
+subsequent digits stand for themselves. The value of the character is
+constrained in the same way as characters specified in hexadecimal.
+For example:
+
+  \040   is another way of writing an ASCII space
+  \40    is the same, provided there are fewer than 40 previous capturing subpatterns
+  \7     is always a back reference
+  \11    might be a back reference, or another way of writing a tab
+  \011   is always a tab
+  \0113  is a tab followed by the character "3"
+  \113   might be a back reference, otherwise the character with octal code 113
+  \377   might be a back reference, otherwise the value 255 (decimal)
+  \81    is either a back reference, or a binary zero followed by the two characters "8" and "1"
+

+Note that octal values of 100 or greater must not be introduced by a leading
+zero, because no more than three octal digits are ever read.
+

+

+All the sequences that define a single character value can be used both inside
+and outside character classes. In addition, inside a character class, \b is
+interpreted as the backspace character (hex 08).
+

+

+\N is not allowed in a character class. \B, \R, and \X are not special
+inside a character class. Like other unrecognized escape sequences, they are
+treated as the literal characters "B", "R", and "X" by default, but cause an
+error if the PCRE_EXTRA option is set. Outside a character class, these
+sequences have different meanings.
+

+

+Unsupported escape sequences
+

+

+In Perl, the sequences \l, \L, \u, and \U are recognized by its string
+handler and used to modify the case of following characters. By default, PCRE
+does not support these escape sequences. However, if the PCRE_JAVASCRIPT_COMPAT
+option is set, \U matches a "U" character, and \u can be used to define a
+character by code point, as described in the previous section.
+

+

+Absolute and relative back references
+

+

+The sequence \g followed by an unsigned or a negative number, optionally
+enclosed in braces, is an absolute or relative back reference. A named back
+reference can be coded as \g{name}. Back references are discussed
+later,
+following the discussion of
+parenthesized subpatterns.
+

+

+Absolute and relative subroutine calls
+

+

+For compatibility with Oniguruma, the non-Perl syntax \g followed by a name or
+a number enclosed either in angle brackets or single quotes, is an alternative
+syntax for referencing a subpattern as a "subroutine". Details are discussed
+later.
+Note that \g{...} (Perl syntax) and \g<...> (Oniguruma syntax) are not
+synonymous. The former is a back reference; the latter is a
+subroutine
+call.
+

+

+Generic character types
+

+

+Another use of backslash is for specifying generic character types:
+
+  \d     any decimal digit
+  \D     any character that is not a decimal digit
+  \h     any horizontal white space character
+  \H     any character that is not a horizontal white space character
+  \s     any white space character
+  \S     any character that is not a white space character
+  \v     any vertical white space character
+  \V     any character that is not a vertical white space character
+  \w     any "word" character
+  \W     any "non-word" character
+

+There is also the single sequence \N, which matches a non-newline character.
+This is the same as
+the "." metacharacter
+when PCRE_DOTALL is not set. Perl also uses \N to match characters by name;
+PCRE does not support this.
+

+

+Each pair of lower and upper case escape sequences partitions the complete set
+of characters into two disjoint sets. Any given character matches one, and only
+one, of each pair. The sequences can appear both inside and outside character
+classes. They each match one character of the appropriate type. If the current
+matching point is at the end of the subject string, all of them fail, because
+there is no character to match.
+

+

+For compatibility with Perl, \s does not match the VT character (code 11).
+This makes it different from the the POSIX "space" class. The \s characters
+are HT (9), LF (10), FF (12), CR (13), and space (32). If "use locale;" is
+included in a Perl script, \s may match the VT character. In PCRE, it never
+does.
+

+

+A "word" character is an underscore or any character that is a letter or digit.
+By default, the definition of letters and digits is controlled by PCRE's
+low-valued character tables, and may vary if locale-specific matching is taking
+place (see
+"Locale support"
+in the
+pcreapi
+page). For example, in a French locale such as "fr_FR" in Unix-like systems,
+or "french" in Windows, some character codes greater than 128 are used for
+accented letters, and these are then matched by \w. The use of locales with
+Unicode is discouraged.
+

+

+By default, in a UTF mode, characters with values greater than 128 never match
+\d, \s, or \w, and always match \D, \S, and \W. These sequences retain
+their original meanings from before UTF support was available, mainly for
+efficiency reasons. However, if PCRE is compiled with Unicode property support,
+and the PCRE_UCP option is set, the behaviour is changed so that Unicode
+properties are used to determine character types, as follows:
+
+  \d  any character that \p{Nd} matches (decimal digit)
+  \s  any character that \p{Z} matches, plus HT, LF, FF, CR
+  \w  any character that \p{L} or \p{N} matches, plus underscore
+

+The upper case escapes match the inverse sets of characters. Note that \d
+matches only decimal digits, whereas \w matches any Unicode digit, as well as
+any Unicode letter, and underscore. Note also that PCRE_UCP affects \b, and
+\B because they are defined in terms of \w and \W. Matching these sequences
+is noticeably slower when PCRE_UCP is set.
+

+

+The sequences \h, \H, \v, and \V are features that were added to Perl at
+release 5.10. In contrast to the other sequences, which match only ASCII
+characters by default, these always match certain high-valued codepoints,
+whether or not PCRE_UCP is set. The horizontal space characters are:
+
+  U+0009     Horizontal tab (HT)
+  U+0020     Space
+  U+00A0     Non-break space
+  U+1680     Ogham space mark
+  U+180E     Mongolian vowel separator
+  U+2000     En quad
+  U+2001     Em quad
+  U+2002     En space
+  U+2003     Em space
+  U+2004     Three-per-em space
+  U+2005     Four-per-em space
+  U+2006     Six-per-em space
+  U+2007     Figure space
+  U+2008     Punctuation space
+  U+2009     Thin space
+  U+200A     Hair space
+  U+202F     Narrow no-break space
+  U+205F     Medium mathematical space
+  U+3000     Ideographic space
+

+The vertical space characters are:
+
+  U+000A     Linefeed (LF)
+  U+000B     Vertical tab (VT)
+  U+000C     Form feed (FF)
+  U+000D     Carriage return (CR)
+  U+0085     Next line (NEL)
+  U+2028     Line separator
+  U+2029     Paragraph separator
+

+In 8-bit, non-UTF-8 mode, only the characters with codepoints less than 256 are
+relevant.
+

+

+Newline sequences
+

+

+Outside a character class, by default, the escape sequence \R matches any
+Unicode newline sequence. In 8-bit non-UTF-8 mode \R is equivalent to the
+following:
+
+  (?>\r\n|\n|\x0b|\f|\r|\x85)
+

+This is an example of an "atomic group", details of which are given
+below.
+This particular group matches either the two-character sequence CR followed by
+LF, or one of the single characters LF (linefeed, U+000A), VT (vertical tab,
+U+000B), FF (form feed, U+000C), CR (carriage return, U+000D), or NEL (next
+line, U+0085). The two-character sequence is treated as a single unit that
+cannot be split.
+

+

+In other modes, two additional characters whose codepoints are greater than 255
+are added: LS (line separator, U+2028) and PS (paragraph separator, U+2029).
+Unicode character property support is not needed for these characters to be
+recognized.
+

+

+It is possible to restrict \R to match only CR, LF, or CRLF (instead of the
+complete set of Unicode line endings) by setting the option PCRE_BSR_ANYCRLF
+either at compile time or when the pattern is matched. (BSR is an abbrevation
+for "backslash R".) This can be made the default when PCRE is built; if this is
+the case, the other behaviour can be requested via the PCRE_BSR_UNICODE option.
+It is also possible to specify these settings by starting a pattern string with
+one of the following sequences:
+
+  (*BSR_ANYCRLF)   CR, LF, or CRLF only
+  (*BSR_UNICODE)   any Unicode newline sequence
+

+These override the default and the options given to the compiling function, but
+they can themselves be overridden by options given to a matching function. Note
+that these special settings, which are not Perl-compatible, are recognized only
+at the very start of a pattern, and that they must be in upper case. If more
+than one of them is present, the last one is used. They can be combined with a
+change of newline convention; for example, a pattern can start with:
+
+  (*ANY)(*BSR_ANYCRLF)
+

+They can also be combined with the (*UTF8), (*UTF16), (*UTF32), (*UTF) or
+(*UCP) special sequences. Inside a character class, \R is treated as an
+unrecognized escape sequence, and so matches the letter "R" by default, but
+causes an error if PCRE_EXTRA is set.
+

+

+Unicode character properties
+

+

+When PCRE is built with Unicode character property support, three additional
+escape sequences that match characters with specific properties are available.
+When in 8-bit non-UTF-8 mode, these sequences are of course limited to testing
+characters whose codepoints are less than 256, but they do work in this mode.
+The extra escape sequences are:
+
+  \p{xx}   a character with the xx property
+  \P{xx}   a character without the xx property
+  \X       a Unicode extended grapheme cluster
+

+The property names represented by xx above are limited to the Unicode
+script names, the general category properties, "Any", which matches any
+character (including newline), and some special PCRE properties (described
+in the
+next section).
+Other Perl properties such as "InMusicalSymbols" are not currently supported by
+PCRE. Note that \P{Any} does not match any characters, so always causes a
+match failure.
+

+

+Sets of Unicode characters are defined as belonging to certain scripts. A
+character from one of these sets can be matched using a script name. For
+example:
+
+  \p{Greek}
+  \P{Han}
+

+Those that are not part of an identified script are lumped together as
+"Common". The current list of scripts is:
+

+

+Arabic,
+Armenian,
+Avestan,
+Balinese,
+Bamum,
+Batak,
+Bengali,
+Bopomofo,
+Brahmi,
+Braille,
+Buginese,
+Buhid,
+Canadian_Aboriginal,
+Carian,
+Chakma,
+Cham,
+Cherokee,
+Common,
+Coptic,
+Cuneiform,
+Cypriot,
+Cyrillic,
+Deseret,
+Devanagari,
+Egyptian_Hieroglyphs,
+Ethiopic,
+Georgian,
+Glagolitic,
+Gothic,
+Greek,
+Gujarati,
+Gurmukhi,
+Han,
+Hangul,
+Hanunoo,
+Hebrew,
+Hiragana,
+Imperial_Aramaic,
+Inherited,
+Inscriptional_Pahlavi,
+Inscriptional_Parthian,
+Javanese,
+Kaithi,
+Kannada,
+Katakana,
+Kayah_Li,
+Kharoshthi,
+Khmer,
+Lao,
+Latin,
+Lepcha,
+Limbu,
+Linear_B,
+Lisu,
+Lycian,
+Lydian,
+Malayalam,
+Mandaic,
+Meetei_Mayek,
+Meroitic_Cursive,
+Meroitic_Hieroglyphs,
+Miao,
+Mongolian,
+Myanmar,
+New_Tai_Lue,
+Nko,
+Ogham,
+Old_Italic,
+Old_Persian,
+Old_South_Arabian,
+Old_Turkic,
+Ol_Chiki,
+Oriya,
+Osmanya,
+Phags_Pa,
+Phoenician,
+Rejang,
+Runic,
+Samaritan,
+Saurashtra,
+Sharada,
+Shavian,
+Sinhala,
+Sora_Sompeng,
+Sundanese,
+Syloti_Nagri,
+Syriac,
+Tagalog,
+Tagbanwa,
+Tai_Le,
+Tai_Tham,
+Tai_Viet,
+Takri,
+Tamil,
+Telugu,
+Thaana,
+Thai,
+Tibetan,
+Tifinagh,
+Ugaritic,
+Vai,
+Yi.
+

+

+Each character has exactly one Unicode general category property, specified by
+a two-letter abbreviation. For compatibility with Perl, negation can be
+specified by including a circumflex between the opening brace and the property
+name. For example, \p{^Lu} is the same as \P{Lu}.
+

+

+If only one letter is specified with \p or \P, it includes all the general
+category properties that start with that letter. In this case, in the absence
+of negation, the curly brackets in the escape sequence are optional; these two
+examples have the same effect:
+
+  \p{L}
+  \pL
+

+The following general category property codes are supported:
+
+  C     Other
+  Cc    Control
+  Cf    Format
+  Cn    Unassigned
+  Co    Private use
+  Cs    Surrogate
+
+  L     Letter
+  Ll    Lower case letter
+  Lm    Modifier letter
+  Lo    Other letter
+  Lt    Title case letter
+  Lu    Upper case letter
+
+  M     Mark
+  Mc    Spacing mark
+  Me    Enclosing mark
+  Mn    Non-spacing mark
+
+  N     Number
+  Nd    Decimal number
+  Nl    Letter number
+  No    Other number
+
+  P     Punctuation
+  Pc    Connector punctuation
+  Pd    Dash punctuation
+  Pe    Close punctuation
+  Pf    Final punctuation
+  Pi    Initial punctuation
+  Po    Other punctuation
+  Ps    Open punctuation
+
+  S     Symbol
+  Sc    Currency symbol
+  Sk    Modifier symbol
+  Sm    Mathematical symbol
+  So    Other symbol
+
+  Z     Separator
+  Zl    Line separator
+  Zp    Paragraph separator
+  Zs    Space separator
+

+The special property L& is also supported: it matches a character that has
+the Lu, Ll, or Lt property, in other words, a letter that is not classified as
+a modifier or "other".
+

+

+The Cs (Surrogate) property applies only to characters in the range U+D800 to
+U+DFFF. Such characters are not valid in Unicode strings and so
+cannot be tested by PCRE, unless UTF validity checking has been turned off
+(see the discussion of PCRE_NO_UTF8_CHECK, PCRE_NO_UTF16_CHECK and
+PCRE_NO_UTF32_CHECK in the
+pcreapi
+page). Perl does not support the Cs property.
+

+

+The long synonyms for property names that Perl supports (such as \p{Letter})
+are not supported by PCRE, nor is it permitted to prefix any of these
+properties with "Is".
+

+

+No character that is in the Unicode table has the Cn (unassigned) property.
+Instead, this property is assumed for any code point that is not in the
+Unicode table.
+

+

+Specifying caseless matching does not affect these escape sequences. For
+example, \p{Lu} always matches only upper case letters. This is different from
+the behaviour of current versions of Perl.
+

+

+Matching characters by Unicode property is not fast, because PCRE has to do a
+multistage table lookup in order to find a character's property. That is why
+the traditional escape sequences such as \d and \w do not use Unicode
+properties in PCRE by default, though you can make them do so by setting the
+PCRE_UCP option or by starting the pattern with (*UCP).
+

+

+Extended grapheme clusters
+

+

+The \X escape matches any number of Unicode characters that form an "extended
+grapheme cluster", and treats the sequence as an atomic group
+(see below).
+Up to and including release 8.31, PCRE matched an earlier, simpler definition
+that was equivalent to
+
+  (?>\PM\pM*)
+

+That is, it matched a character without the "mark" property, followed by zero
+or more characters with the "mark" property. Characters with the "mark"
+property are typically non-spacing accents that affect the preceding character.
+

+

+This simple definition was extended in Unicode to include more complicated
+kinds of composite character by giving each character a grapheme breaking
+property, and creating rules that use these properties to define the boundaries
+of extended grapheme clusters. In releases of PCRE later than 8.31, \X matches
+one of these clusters.
+

+

+\X always matches at least one character. Then it decides whether to add
+additional characters according to the following rules for ending a cluster:
+

+

+1. End at the end of the subject string.
+

+

+2. Do not end between CR and LF; otherwise end after any control character.
+

+

+3. Do not break Hangul (a Korean script) syllable sequences. Hangul characters
+are of five types: L, V, T, LV, and LVT. An L character may be followed by an
+L, V, LV, or LVT character; an LV or V character may be followed by a V or T
+character; an LVT or T character may be follwed only by a T character.
+

+

+4. Do not end before extending characters or spacing marks. Characters with
+the "mark" property always have the "extend" grapheme breaking property.
+

+

+5. Do not end after prepend characters.
+

+

+6. Otherwise, end the cluster.
+

+

+PCRE's additional properties
+

+

+As well as the standard Unicode properties described above, PCRE supports four
+more that make it possible to convert traditional escape sequences such as \w
+and \s and POSIX character classes to use Unicode properties. PCRE uses these
+non-standard, non-Perl properties internally when PCRE_UCP is set. However,
+they may also be used explicitly. These properties are:
+
+  Xan   Any alphanumeric character
+  Xps   Any POSIX space character
+  Xsp   Any Perl space character
+  Xwd   Any Perl "word" character
+

+Xan matches characters that have either the L (letter) or the N (number)
+property. Xps matches the characters tab, linefeed, vertical tab, form feed, or
+carriage return, and any other character that has the Z (separator) property.
+Xsp is the same as Xps, except that vertical tab is excluded. Xwd matches the
+same characters as Xan, plus underscore.
+

+

+There is another non-standard property, Xuc, which matches any character that
+can be represented by a Universal Character Name in C++ and other programming
+languages. These are the characters $, @, ` (grave accent), and all characters
+with Unicode code points greater than or equal to U+00A0, except for the
+surrogates U+D800 to U+DFFF. Note that most base (ASCII) characters are
+excluded. (Universal Character Names are of the form \uHHHH or \UHHHHHHHH
+where H is a hexadecimal digit. Note that the Xuc property does not match these
+sequences but the characters that they represent.)
+

+

+Resetting the match start
+

+

+The escape sequence \K causes any previously matched characters not to be
+included in the final matched sequence. For example, the pattern:
+
+  foo\Kbar
+

+matches "foobar", but reports that it has matched "bar". This feature is
+similar to a lookbehind assertion
+(described below).
+However, in this case, the part of the subject before the real match does not
+have to be of fixed length, as lookbehind assertions do. The use of \K does
+not interfere with the setting of
+captured substrings.
+For example, when the pattern
+
+  (foo)\Kbar
+

+matches "foobar", the first substring is still set to "foo".
+

+

+Perl documents that the use of \K within assertions is "not well defined". In
+PCRE, \K is acted upon when it occurs inside positive assertions, but is
+ignored in negative assertions.
+

+

+Simple assertions
+

+

+The final use of backslash is for certain simple assertions. An assertion
+specifies a condition that has to be met at a particular point in a match,
+without consuming any characters from the subject string. The use of
+subpatterns for more complicated assertions is described
+below.
+The backslashed assertions are:
+
+  \b     matches at a word boundary
+  \B     matches when not at a word boundary
+  \A     matches at the start of the subject
+  \Z     matches at the end of the subject
+          also matches before a newline at the end of the subject
+  \z     matches only at the end of the subject
+  \G     matches at the first matching position in the subject
+

+Inside a character class, \b has a different meaning; it matches the backspace
+character. If any other of these assertions appears in a character class, by
+default it matches the corresponding literal character (for example, \B
+matches the letter B). However, if the PCRE_EXTRA option is set, an "invalid
+escape sequence" error is generated instead.
+

+

+A word boundary is a position in the subject string where the current character
+and the previous character do not both match \w or \W (i.e. one matches
+\w and the other matches \W), or the start or end of the string if the
+first or last character matches \w, respectively. In a UTF mode, the meanings
+of \w and \W can be changed by setting the PCRE_UCP option. When this is
+done, it also affects \b and \B. Neither PCRE nor Perl has a separate "start
+of word" or "end of word" metasequence. However, whatever follows \b normally
+determines which it is. For example, the fragment \ba matches "a" at the start
+of a word.
+

+

+The \A, \Z, and \z assertions differ from the traditional circumflex and
+dollar (described in the next section) in that they only ever match at the very
+start and end of the subject string, whatever options are set. Thus, they are
+independent of multiline mode. These three assertions are not affected by the
+PCRE_NOTBOL or PCRE_NOTEOL options, which affect only the behaviour of the
+circumflex and dollar metacharacters. However, if the startoffset
+argument of pcre_exec() is non-zero, indicating that matching is to start
+at a point other than the beginning of the subject, \A can never match. The
+difference between \Z and \z is that \Z matches before a newline at the end
+of the string as well as at the very end, whereas \z matches only at the end.
+

+

+The \G assertion is true only when the current matching position is at the
+start point of the match, as specified by the startoffset argument of
+pcre_exec(). It differs from \A when the value of startoffset is
+non-zero. By calling pcre_exec() multiple times with appropriate
+arguments, you can mimic Perl's /g option, and it is in this kind of
+implementation where \G can be useful.
+

+

+Note, however, that PCRE's interpretation of \G, as the start of the current
+match, is subtly different from Perl's, which defines it as the end of the
+previous match. In Perl, these can be different when the previously matched
+string was empty. Because PCRE does just one match at a time, it cannot
+reproduce this behaviour.
+

+

+If all the alternatives of a pattern begin with \G, the expression is anchored
+to the starting match position, and the "anchored" flag is set in the compiled
+regular expression.
+

+
CIRCUMFLEX AND DOLLAR

+

+The circumflex and dollar metacharacters are zero-width assertions. That is,
+they test for a particular condition being true without consuming any
+characters from the subject string.
+

+

+Outside a character class, in the default matching mode, the circumflex
+character is an assertion that is true only if the current matching point is at
+the start of the subject string. If the startoffset argument of
+pcre_exec() is non-zero, circumflex can never match if the PCRE_MULTILINE
+option is unset. Inside a character class, circumflex has an entirely different
+meaning
+(see below).
+

+

+Circumflex need not be the first character of the pattern if a number of
+alternatives are involved, but it should be the first thing in each alternative
+in which it appears if the pattern is ever to match that branch. If all
+possible alternatives start with a circumflex, that is, if the pattern is
+constrained to match only at the start of the subject, it is said to be an
+"anchored" pattern. (There are also other constructs that can cause a pattern
+to be anchored.)
+

+

+The dollar character is an assertion that is true only if the current matching
+point is at the end of the subject string, or immediately before a newline at
+the end of the string (by default). Note, however, that it does not actually
+match the newline. Dollar need not be the last character of the pattern if a
+number of alternatives are involved, but it should be the last item in any
+branch in which it appears. Dollar has no special meaning in a character class.
+

+

+The meaning of dollar can be changed so that it matches only at the very end of
+the string, by setting the PCRE_DOLLAR_ENDONLY option at compile time. This
+does not affect the \Z assertion.
+

+

+The meanings of the circumflex and dollar characters are changed if the
+PCRE_MULTILINE option is set. When this is the case, a circumflex matches
+immediately after internal newlines as well as at the start of the subject
+string. It does not match after a newline that ends the string. A dollar
+matches before any newlines in the string, as well as at the very end, when
+PCRE_MULTILINE is set. When newline is specified as the two-character
+sequence CRLF, isolated CR and LF characters do not indicate newlines.
+

+

+For example, the pattern /^abc$/ matches the subject string "def\nabc" (where
+\n represents a newline) in multiline mode, but not otherwise. Consequently,
+patterns that are anchored in single line mode because all branches start with
+^ are not anchored in multiline mode, and a match for circumflex is possible
+when the startoffset argument of pcre_exec() is non-zero. The
+PCRE_DOLLAR_ENDONLY option is ignored if PCRE_MULTILINE is set.
+

+

+Note that the sequences \A, \Z, and \z can be used to match the start and
+end of the subject in both modes, and if all branches of a pattern start with
+\A it is always anchored, whether or not PCRE_MULTILINE is set.
+

+
FULL STOP (PERIOD, DOT) AND \N

+

+Outside a character class, a dot in the pattern matches any one character in
+the subject string except (by default) a character that signifies the end of a
+line.
+

+

+When a line ending is defined as a single character, dot never matches that
+character; when the two-character sequence CRLF is used, dot does not match CR
+if it is immediately followed by LF, but otherwise it matches all characters
+(including isolated CRs and LFs). When any Unicode line endings are being
+recognized, dot does not match CR or LF or any of the other line ending
+characters.
+

+

+The behaviour of dot with regard to newlines can be changed. If the PCRE_DOTALL
+option is set, a dot matches any one character, without exception. If the
+two-character sequence CRLF is present in the subject string, it takes two dots
+to match it.
+

+

+The handling of dot is entirely independent of the handling of circumflex and
+dollar, the only relationship being that they both involve newlines. Dot has no
+special meaning in a character class.
+

+

+The escape sequence \N behaves like a dot, except that it is not affected by
+the PCRE_DOTALL option. In other words, it matches any character except one
+that signifies the end of a line. Perl also uses \N to match characters by
+name; PCRE does not support this.
+

+
MATCHING A SINGLE DATA UNIT

+

+Outside a character class, the escape sequence \C matches any one data unit,
+whether or not a UTF mode is set. In the 8-bit library, one data unit is one
+byte; in the 16-bit library it is a 16-bit unit; in the 32-bit library it is
+a 32-bit unit. Unlike a dot, \C always
+matches line-ending characters. The feature is provided in Perl in order to
+match individual bytes in UTF-8 mode, but it is unclear how it can usefully be
+used. Because \C breaks up characters into individual data units, matching one
+unit with \C in a UTF mode means that the rest of the string may start with a
+malformed UTF character. This has undefined results, because PCRE assumes that
+it is dealing with valid UTF strings (and by default it checks this at the
+start of processing unless the PCRE_NO_UTF8_CHECK, PCRE_NO_UTF16_CHECK or
+PCRE_NO_UTF32_CHECK option is used).
+

+

+PCRE does not allow \C to appear in lookbehind assertions
+(described below)
+in a UTF mode, because this would make it impossible to calculate the length of
+the lookbehind.
+

+

+In general, the \C escape sequence is best avoided. However, one
+way of using it that avoids the problem of malformed UTF characters is to use a
+lookahead to check the length of the next character, as in this pattern, which
+could be used with a UTF-8 string (ignore white space and line breaks):
+
+  (?| (?=[\x00-\x7f])(\C) |
+      (?=[\x80-\x{7ff}])(\C)(\C) |
+      (?=[\x{800}-\x{ffff}])(\C)(\C)(\C) |
+      (?=[\x{10000}-\x{1fffff}])(\C)(\C)(\C)(\C))
+

+A group that starts with (?| resets the capturing parentheses numbers in each
+alternative (see
+"Duplicate Subpattern Numbers"
+below). The assertions at the start of each branch check the next UTF-8
+character for values whose encoding uses 1, 2, 3, or 4 bytes, respectively. The
+character's individual bytes are then captured by the appropriate number of
+groups.
+

+
SQUARE BRACKETS AND CHARACTER CLASSES

+

+An opening square bracket introduces a character class, terminated by a closing
+square bracket. A closing square bracket on its own is not special by default.
+However, if the PCRE_JAVASCRIPT_COMPAT option is set, a lone closing square
+bracket causes a compile-time error. If a closing square bracket is required as
+a member of the class, it should be the first data character in the class
+(after an initial circumflex, if present) or escaped with a backslash.
+

+

+A character class matches a single character in the subject. In a UTF mode, the
+character may be more than one data unit long. A matched character must be in
+the set of characters defined by the class, unless the first character in the
+class definition is a circumflex, in which case the subject character must not
+be in the set defined by the class. If a circumflex is actually required as a
+member of the class, ensure it is not the first character, or escape it with a
+backslash.
+

+

+For example, the character class [aeiou] matches any lower case vowel, while
+[^aeiou] matches any character that is not a lower case vowel. Note that a
+circumflex is just a convenient notation for specifying the characters that
+are in the class by enumerating those that are not. A class that starts with a
+circumflex is not an assertion; it still consumes a character from the subject
+string, and therefore it fails if the current pointer is at the end of the
+string.
+

+

+In UTF-8 (UTF-16, UTF-32) mode, characters with values greater than 255 (0xffff)
+can be included in a class as a literal string of data units, or by using the
+\x{ escaping mechanism.
+

+

+When caseless matching is set, any letters in a class represent both their
+upper case and lower case versions, so for example, a caseless [aeiou] matches
+"A" as well as "a", and a caseless [^aeiou] does not match "A", whereas a
+caseful version would. In a UTF mode, PCRE always understands the concept of
+case for characters whose values are less than 128, so caseless matching is
+always possible. For characters with higher values, the concept of case is
+supported if PCRE is compiled with Unicode property support, but not otherwise.
+If you want to use caseless matching in a UTF mode for characters 128 and
+above, you must ensure that PCRE is compiled with Unicode property support as
+well as with UTF support.
+

+

+Characters that might indicate line breaks are never treated in any special way
+when matching character classes, whatever line-ending sequence is in use, and
+whatever setting of the PCRE_DOTALL and PCRE_MULTILINE options is used. A class
+such as [^a] always matches one of these characters.
+

+

+The minus (hyphen) character can be used to specify a range of characters in a
+character class. For example, [d-m] matches any letter between d and m,
+inclusive. If a minus character is required in a class, it must be escaped with
+a backslash or appear in a position where it cannot be interpreted as
+indicating a range, typically as the first or last character in the class.
+

+

+It is not possible to have the literal character "]" as the end character of a
+range. A pattern such as [W-]46] is interpreted as a class of two characters
+("W" and "-") followed by a literal string "46]", so it would match "W46]" or
+"-46]". However, if the "]" is escaped with a backslash it is interpreted as
+the end of range, so [W-\]46] is interpreted as a class containing a range
+followed by two other characters. The octal or hexadecimal representation of
+"]" can also be used to end a range.
+

+

+Ranges operate in the collating sequence of character values. They can also be
+used for characters specified numerically, for example [\000-\037]. Ranges
+can include any characters that are valid for the current mode.
+

+

+If a range that includes letters is used when caseless matching is set, it
+matches the letters in either case. For example, [W-c] is equivalent to
+[][\\^_`wxyzabc], matched caselessly, and in a non-UTF mode, if character
+tables for a French locale are in use, [\xc8-\xcb] matches accented E
+characters in both cases. In UTF modes, PCRE supports the concept of case for
+characters with values greater than 128 only when it is compiled with Unicode
+property support.
+

+

+The character escape sequences \d, \D, \h, \H, \p, \P, \s, \S, \v,
+\V, \w, and \W may appear in a character class, and add the characters that
+they match to the class. For example, [\dABCDEF] matches any hexadecimal
+digit. In UTF modes, the PCRE_UCP option affects the meanings of \d, \s, \w
+and their upper case partners, just as it does when they appear outside a
+character class, as described in the section entitled
+"Generic character types"
+above. The escape sequence \b has a different meaning inside a character
+class; it matches the backspace character. The sequences \B, \N, \R, and \X
+are not special inside a character class. Like any other unrecognized escape
+sequences, they are treated as the literal characters "B", "N", "R", and "X" by
+default, but cause an error if the PCRE_EXTRA option is set.
+

+

+A circumflex can conveniently be used with the upper case character types to
+specify a more restricted set of characters than the matching lower case type.
+For example, the class [^\W_] matches any letter or digit, but not underscore,
+whereas [\w] includes underscore. A positive character class should be read as
+"something OR something OR ..." and a negative class as "NOT something AND NOT
+something AND NOT ...".
+

+

+The only metacharacters that are recognized in character classes are backslash,
+hyphen (only where it can be interpreted as specifying a range), circumflex
+(only at the start), opening square bracket (only when it can be interpreted as
+introducing a POSIX class name - see the next section), and the terminating
+closing square bracket. However, escaping other non-alphanumeric characters
+does no harm.
+

+
POSIX CHARACTER CLASSES

+

+Perl supports the POSIX notation for character classes. This uses names
+enclosed by [: and :] within the enclosing square brackets. PCRE also supports
+this notation. For example,
+
+  [01[:alpha:]%]
+

+matches "0", "1", any alphabetic character, or "%". The supported class names
+are:
+
+  alnum    letters and digits
+  alpha    letters
+  ascii    character codes 0 - 127
+  blank    space or tab only
+  cntrl    control characters
+  digit    decimal digits (same as \d)
+  graph    printing characters, excluding space
+  lower    lower case letters
+  print    printing characters, including space
+  punct    printing characters, excluding letters and digits and space
+  space    white space (not quite the same as \s)
+  upper    upper case letters
+  word     "word" characters (same as \w)
+  xdigit   hexadecimal digits
+

+The "space" characters are HT (9), LF (10), VT (11), FF (12), CR (13), and
+space (32). Notice that this list includes the VT character (code 11). This
+makes "space" different to \s, which does not include VT (for Perl
+compatibility).
+

+

+The name "word" is a Perl extension, and "blank" is a GNU extension from Perl
+5.8. Another Perl extension is negation, which is indicated by a ^ character
+after the colon. For example,
+
+  [12[:^digit:]]
+

+matches "1", "2", or any non-digit. PCRE (and Perl) also recognize the POSIX
+syntax [.ch.] and [=ch=] where "ch" is a "collating element", but these are not
+supported, and an error is given if they are encountered.
+

+

+By default, in UTF modes, characters with values greater than 128 do not match
+any of the POSIX character classes. However, if the PCRE_UCP option is passed
+to pcre_compile(), some of the classes are changed so that Unicode
+character properties are used. This is achieved by replacing the POSIX classes
+by other sequences, as follows:
+
+  [:alnum:]  becomes  \p{Xan}
+  [:alpha:]  becomes  \p{L}
+  [:blank:]  becomes  \h
+  [:digit:]  becomes  \p{Nd}
+  [:lower:]  becomes  \p{Ll}
+  [:space:]  becomes  \p{Xps}
+  [:upper:]  becomes  \p{Lu}
+  [:word:]   becomes  \p{Xwd}
+

+Negated versions, such as [:^alpha:] use \P instead of \p. The other POSIX
+classes are unchanged, and match only characters with code points less than
+128.
+

+
VERTICAL BAR

+

+Vertical bar characters are used to separate alternative patterns. For example,
+the pattern
+
+  gilbert|sullivan
+

+matches either "gilbert" or "sullivan". Any number of alternatives may appear,
+and an empty alternative is permitted (matching the empty string). The matching
+process tries each alternative in turn, from left to right, and the first one
+that succeeds is used. If the alternatives are within a subpattern
+(defined below),
+"succeeds" means matching the rest of the main pattern as well as the
+alternative in the subpattern.
+

+
INTERNAL OPTION SETTING

+

+The settings of the PCRE_CASELESS, PCRE_MULTILINE, PCRE_DOTALL, and
+PCRE_EXTENDED options (which are Perl-compatible) can be changed from within
+the pattern by a sequence of Perl option letters enclosed between "(?" and ")".
+The option letters are
+
+  i  for PCRE_CASELESS
+  m  for PCRE_MULTILINE
+  s  for PCRE_DOTALL
+  x  for PCRE_EXTENDED
+

+For example, (?im) sets caseless, multiline matching. It is also possible to
+unset these options by preceding the letter with a hyphen, and a combined
+setting and unsetting such as (?im-sx), which sets PCRE_CASELESS and
+PCRE_MULTILINE while unsetting PCRE_DOTALL and PCRE_EXTENDED, is also
+permitted. If a letter appears both before and after the hyphen, the option is
+unset.
+

+

+The PCRE-specific options PCRE_DUPNAMES, PCRE_UNGREEDY, and PCRE_EXTRA can be
+changed in the same way as the Perl-compatible options by using the characters
+J, U and X respectively.
+

+

+When one of these option changes occurs at top level (that is, not inside
+subpattern parentheses), the change applies to the remainder of the pattern
+that follows. If the change is placed right at the start of a pattern, PCRE
+extracts it into the global options (and it will therefore show up in data
+extracted by the pcre_fullinfo() function).
+

+

+An option change within a subpattern (see below for a description of
+subpatterns) affects only that part of the subpattern that follows it, so
+
+  (a(?i)b)c
+

+matches abc and aBc and no other strings (assuming PCRE_CASELESS is not used).
+By this means, options can be made to have different settings in different
+parts of the pattern. Any changes made in one alternative do carry on
+into subsequent branches within the same subpattern. For example,
+
+  (a(?i)b|c)
+

+matches "ab", "aB", "c", and "C", even though when matching "C" the first
+branch is abandoned before the option setting. This is because the effects of
+option settings happen at compile time. There would be some very weird
+behaviour otherwise.
+

+

+Note: There are other PCRE-specific options that can be set by the
+application when the compiling or matching functions are called. In some cases
+the pattern can contain special leading sequences such as (*CRLF) to override
+what the application has set or what has been defaulted. Details are given in
+the section entitled
+"Newline sequences"
+above. There are also the (*UTF8), (*UTF16),(*UTF32), and (*UCP) leading
+sequences that can be used to set UTF and Unicode property modes; they are
+equivalent to setting the PCRE_UTF8, PCRE_UTF16, PCRE_UTF32 and the PCRE_UCP
+options, respectively. The (*UTF) sequence is a generic version that can be
+used with any of the libraries. However, the application can set the
+PCRE_NEVER_UTF option, which locks out the use of the (*UTF) sequences.
+

+
SUBPATTERNS

+

+Subpatterns are delimited by parentheses (round brackets), which can be nested.
+Turning part of a pattern into a subpattern does two things:
+

+

+1. It localizes a set of alternatives. For example, the pattern
+
+  cat(aract|erpillar|)
+

+matches "cataract", "caterpillar", or "cat". Without the parentheses, it would
+match "cataract", "erpillar" or an empty string.
+

+

+2. It sets up the subpattern as a capturing subpattern. This means that, when
+the whole pattern matches, that portion of the subject string that matched the
+subpattern is passed back to the caller via the ovector argument of the
+matching function. (This applies only to the traditional matching functions;
+the DFA matching functions do not support capturing.)
+

+

+Opening parentheses are counted from left to right (starting from 1) to obtain
+numbers for the capturing subpatterns. For example, if the string "the red
+king" is matched against the pattern
+
+  the ((red|white) (king|queen))
+

+the captured substrings are "red king", "red", and "king", and are numbered 1,
+2, and 3, respectively.
+

+

+The fact that plain parentheses fulfil two functions is not always helpful.
+There are often times when a grouping subpattern is required without a
+capturing requirement. If an opening parenthesis is followed by a question mark
+and a colon, the subpattern does not do any capturing, and is not counted when
+computing the number of any subsequent capturing subpatterns. For example, if
+the string "the white queen" is matched against the pattern
+
+  the ((?:red|white) (king|queen))
+

+the captured substrings are "white queen" and "queen", and are numbered 1 and
+2. The maximum number of capturing subpatterns is 65535.
+

+

+As a convenient shorthand, if any option settings are required at the start of
+a non-capturing subpattern, the option letters may appear between the "?" and
+the ":". Thus the two patterns
+
+  (?i:saturday|sunday)
+  (?:(?i)saturday|sunday)
+

+match exactly the same set of strings. Because alternative branches are tried
+from left to right, and options are not reset until the end of the subpattern
+is reached, an option setting in one branch does affect subsequent branches, so
+the above patterns match "SUNDAY" as well as "Saturday".
+

+
DUPLICATE SUBPATTERN NUMBERS

+

+Perl 5.10 introduced a feature whereby each alternative in a subpattern uses
+the same numbers for its capturing parentheses. Such a subpattern starts with
+(?| and is itself a non-capturing subpattern. For example, consider this
+pattern:
+
+  (?|(Sat)ur|(Sun))day
+

+Because the two alternatives are inside a (?| group, both sets of capturing
+parentheses are numbered one. Thus, when the pattern matches, you can look
+at captured substring number one, whichever alternative matched. This construct
+is useful when you want to capture part, but not all, of one of a number of
+alternatives. Inside a (?| group, parentheses are numbered as usual, but the
+number is reset at the start of each branch. The numbers of any capturing
+parentheses that follow the subpattern start after the highest number used in
+any branch. The following example is taken from the Perl documentation. The
+numbers underneath show in which buffer the captured content will be stored.
+
+  # before  ---------------branch-reset----------- after
+  / ( a )  (?| x ( y ) z | (p (q) r) | (t) u (v) ) ( z ) /x
+  # 1            2         2  3        2     3     4
+

+A back reference to a numbered subpattern uses the most recent value that is
+set for that number by any subpattern. The following pattern matches "abcabc"
+or "defdef":
+
+  /(?|(abc)|(def))\1/
+

+In contrast, a subroutine call to a numbered subpattern always refers to the
+first one in the pattern with the given number. The following pattern matches
+"abcabc" or "defabc":
+
+  /(?|(abc)|(def))(?1)/
+

+If a
+condition test
+for a subpattern's having matched refers to a non-unique number, the test is
+true if any of the subpatterns of that number have matched.
+

+

+An alternative approach to using this "branch reset" feature is to use
+duplicate named subpatterns, as described in the next section.
+

+
NAMED SUBPATTERNS

+

+Identifying capturing parentheses by number is simple, but it can be very hard
+to keep track of the numbers in complicated regular expressions. Furthermore,
+if an expression is modified, the numbers may change. To help with this
+difficulty, PCRE supports the naming of subpatterns. This feature was not
+added to Perl until release 5.10. Python had the feature earlier, and PCRE
+introduced it at release 4.0, using the Python syntax. PCRE now supports both
+the Perl and the Python syntax. Perl allows identically numbered subpatterns to
+have different names, but PCRE does not.
+

+

+In PCRE, a subpattern can be named in one of three ways: (?<name>...) or
+(?'name'...) as in Perl, or (?P<name>...) as in Python. References to capturing
+parentheses from other parts of the pattern, such as
+back references,
+recursion,
+and
+conditions,
+can be made by name as well as by number.
+

+

+Names consist of up to 32 alphanumeric characters and underscores. Named
+capturing parentheses are still allocated numbers as well as names, exactly as
+if the names were not present. The PCRE API provides function calls for
+extracting the name-to-number translation table from a compiled pattern. There
+is also a convenience function for extracting a captured substring by name.
+

+

+By default, a name must be unique within a pattern, but it is possible to relax
+this constraint by setting the PCRE_DUPNAMES option at compile time. (Duplicate
+names are also always permitted for subpatterns with the same number, set up as
+described in the previous section.) Duplicate names can be useful for patterns
+where only one instance of the named parentheses can match. Suppose you want to
+match the name of a weekday, either as a 3-letter abbreviation or as the full
+name, and in both cases you want to extract the abbreviation. This pattern
+(ignoring the line breaks) does the job:
+
+  (?<DN>Mon|Fri|Sun)(?:day)?|
+  (?<DN>Tue)(?:sday)?|
+  (?<DN>Wed)(?:nesday)?|
+  (?<DN>Thu)(?:rsday)?|
+  (?<DN>Sat)(?:urday)?
+

+There are five capturing substrings, but only one is ever set after a match.
+(An alternative way of solving this problem is to use a "branch reset"
+subpattern, as described in the previous section.)
+

+

+The convenience function for extracting the data by name returns the substring
+for the first (and in this example, the only) subpattern of that name that
+matched. This saves searching to find which numbered subpattern it was.
+

+

+If you make a back reference to a non-unique named subpattern from elsewhere in
+the pattern, the one that corresponds to the first occurrence of the name is
+used. In the absence of duplicate numbers (see the previous section) this is
+the one with the lowest number. If you use a named reference in a condition
+test (see the
+section about conditions
+below), either to check whether a subpattern has matched, or to check for
+recursion, all subpatterns with the same name are tested. If the condition is
+true for any one of them, the overall condition is true. This is the same
+behaviour as testing by number. For further details of the interfaces for
+handling named subpatterns, see the
+pcreapi
+documentation.
+

+

+Warning: You cannot use different names to distinguish between two
+subpatterns with the same number because PCRE uses only the numbers when
+matching. For this reason, an error is given at compile time if different names
+are given to subpatterns with the same number. However, you can give the same
+name to subpatterns with the same number, even when PCRE_DUPNAMES is not set.
+

+
REPETITION

+

+Repetition is specified by quantifiers, which can follow any of the following
+items:
+
+  a literal data character
+  the dot metacharacter
+  the \C escape sequence
+  the \X escape sequence
+  the \R escape sequence
+  an escape such as \d or \pL that matches a single character
+  a character class
+  a back reference (see next section)
+  a parenthesized subpattern (including assertions)
+  a subroutine call to a subpattern (recursive or otherwise)
+

+The general repetition quantifier specifies a minimum and maximum number of
+permitted matches, by giving the two numbers in curly brackets (braces),
+separated by a comma. The numbers must be less than 65536, and the first must
+be less than or equal to the second. For example:
+
+  z{2,4}
+

+matches "zz", "zzz", or "zzzz". A closing brace on its own is not a special
+character. If the second number is omitted, but the comma is present, there is
+no upper limit; if the second number and the comma are both omitted, the
+quantifier specifies an exact number of required matches. Thus
+
+  [aeiou]{3,}
+

+matches at least 3 successive vowels, but may match many more, while
+
+  \d{8}
+

+matches exactly 8 digits. An opening curly bracket that appears in a position
+where a quantifier is not allowed, or one that does not match the syntax of a
+quantifier, is taken as a literal character. For example, {,6} is not a
+quantifier, but a literal string of four characters.
+

+

+In UTF modes, quantifiers apply to characters rather than to individual data
+units. Thus, for example, \x{100}{2} matches two characters, each of
+which is represented by a two-byte sequence in a UTF-8 string. Similarly,
+\X{3} matches three Unicode extended grapheme clusters, each of which may be
+several data units long (and they may be of different lengths).
+

+

+The quantifier {0} is permitted, causing the expression to behave as if the
+previous item and the quantifier were not present. This may be useful for
+subpatterns that are referenced as
+subroutines
+from elsewhere in the pattern (but see also the section entitled
+"Defining subpatterns for use by reference only"
+below). Items other than subpatterns that have a {0} quantifier are omitted
+from the compiled pattern.
+

+

+For convenience, the three most common quantifiers have single-character
+abbreviations:
+
+  *    is equivalent to {0,}
+  +    is equivalent to {1,}
+  ?    is equivalent to {0,1}
+

+It is possible to construct infinite loops by following a subpattern that can
+match no characters with a quantifier that has no upper limit, for example:
+
+  (a?)*
+

+Earlier versions of Perl and PCRE used to give an error at compile time for
+such patterns. However, because there are cases where this can be useful, such
+patterns are now accepted, but if any repetition of the subpattern does in fact
+match no characters, the loop is forcibly broken.
+

+

+By default, the quantifiers are "greedy", that is, they match as much as
+possible (up to the maximum number of permitted times), without causing the
+rest of the pattern to fail. The classic example of where this gives problems
+is in trying to match comments in C programs. These appear between /* and */
+and within the comment, individual * and / characters may appear. An attempt to
+match C comments by applying the pattern
+
+  /\*.*\*/
+

+to the string
+
+  /* first comment */  not comment  /* second comment */
+

+fails, because it matches the entire string owing to the greediness of the .*
+item.
+

+

+However, if a quantifier is followed by a question mark, it ceases to be
+greedy, and instead matches the minimum number of times possible, so the
+pattern
+
+  /\*.*?\*/
+

+does the right thing with the C comments. The meaning of the various
+quantifiers is not otherwise changed, just the preferred number of matches.
+Do not confuse this use of question mark with its use as a quantifier in its
+own right. Because it has two uses, it can sometimes appear doubled, as in
+
+  \d??\d
+

+which matches one digit by preference, but can match two if that is the only
+way the rest of the pattern matches.
+

+

+If the PCRE_UNGREEDY option is set (an option that is not available in Perl),
+the quantifiers are not greedy by default, but individual ones can be made
+greedy by following them with a question mark. In other words, it inverts the
+default behaviour.
+

+

+When a parenthesized subpattern is quantified with a minimum repeat count that
+is greater than 1 or with a limited maximum, more memory is required for the
+compiled pattern, in proportion to the size of the minimum or maximum.
+

+

+If a pattern starts with .* or .{0,} and the PCRE_DOTALL option (equivalent
+to Perl's /s) is set, thus allowing the dot to match newlines, the pattern is
+implicitly anchored, because whatever follows will be tried against every
+character position in the subject string, so there is no point in retrying the
+overall match at any position after the first. PCRE normally treats such a
+pattern as though it were preceded by \A.
+

+

+In cases where it is known that the subject string contains no newlines, it is
+worth setting PCRE_DOTALL in order to obtain this optimization, or
+alternatively using ^ to indicate anchoring explicitly.
+

+

+However, there are some cases where the optimization cannot be used. When .*
+is inside capturing parentheses that are the subject of a back reference
+elsewhere in the pattern, a match at the start may fail where a later one
+succeeds. Consider, for example:
+
+  (.*)abc\1
+

+If the subject is "xyz123abc123" the match point is the fourth character. For
+this reason, such a pattern is not implicitly anchored.
+

+

+Another case where implicit anchoring is not applied is when the leading .* is
+inside an atomic group. Once again, a match at the start may fail where a later
+one succeeds. Consider this pattern:
+
+  (?>.*?a)b
+

+It matches "ab" in the subject "aab". The use of the backtracking control verbs
+(*PRUNE) and (*SKIP) also disable this optimization.
+

+

+When a capturing subpattern is repeated, the value captured is the substring
+that matched the final iteration. For example, after
+
+  (tweedle[dume]{3}\s*)+
+

+has matched "tweedledum tweedledee" the value of the captured substring is
+"tweedledee". However, if there are nested capturing subpatterns, the
+corresponding captured values may have been set in previous iterations. For
+example, after
+
+  /(a|(b))+/
+

+matches "aba" the value of the second captured substring is "b".
+

+
ATOMIC GROUPING AND POSSESSIVE QUANTIFIERS

+

+With both maximizing ("greedy") and minimizing ("ungreedy" or "lazy")
+repetition, failure of what follows normally causes the repeated item to be
+re-evaluated to see if a different number of repeats allows the rest of the
+pattern to match. Sometimes it is useful to prevent this, either to change the
+nature of the match, or to cause it fail earlier than it otherwise might, when
+the author of the pattern knows there is no point in carrying on.
+

+

+Consider, for example, the pattern \d+foo when applied to the subject line
+
+  123456bar
+

+After matching all 6 digits and then failing to match "foo", the normal
+action of the matcher is to try again with only 5 digits matching the \d+
+item, and then with 4, and so on, before ultimately failing. "Atomic grouping"
+(a term taken from Jeffrey Friedl's book) provides the means for specifying
+that once a subpattern has matched, it is not to be re-evaluated in this way.
+

+

+If we use atomic grouping for the previous example, the matcher gives up
+immediately on failing to match "foo" the first time. The notation is a kind of
+special parenthesis, starting with (?> as in this example:
+
+  (?>\d+)foo
+

+This kind of parenthesis "locks up" the  part of the pattern it contains once
+it has matched, and a failure further into the pattern is prevented from
+backtracking into it. Backtracking past it to previous items, however, works as
+normal.
+

+

+An alternative description is that a subpattern of this type matches the string
+of characters that an identical standalone pattern would match, if anchored at
+the current point in the subject string.
+

+

+Atomic grouping subpatterns are not capturing subpatterns. Simple cases such as
+the above example can be thought of as a maximizing repeat that must swallow
+everything it can. So, while both \d+ and \d+? are prepared to adjust the
+number of digits they match in order to make the rest of the pattern match,
+(?>\d+) can only match an entire sequence of digits.
+

+

+Atomic groups in general can of course contain arbitrarily complicated
+subpatterns, and can be nested. However, when the subpattern for an atomic
+group is just a single repeated item, as in the example above, a simpler
+notation, called a "possessive quantifier" can be used. This consists of an
+additional + character following a quantifier. Using this notation, the
+previous example can be rewritten as
+
+  \d++foo
+

+Note that a possessive quantifier can be used with an entire group, for
+example:
+
+  (abc|xyz){2,3}+
+

+Possessive quantifiers are always greedy; the setting of the PCRE_UNGREEDY
+option is ignored. They are a convenient notation for the simpler forms of
+atomic group. However, there is no difference in the meaning of a possessive
+quantifier and the equivalent atomic group, though there may be a performance
+difference; possessive quantifiers should be slightly faster.
+

+

+The possessive quantifier syntax is an extension to the Perl 5.8 syntax.
+Jeffrey Friedl originated the idea (and the name) in the first edition of his
+book. Mike McCloskey liked it, so implemented it when he built Sun's Java
+package, and PCRE copied it from there. It ultimately found its way into Perl
+at release 5.10.
+

+

+PCRE has an optimization that automatically "possessifies" certain simple
+pattern constructs. For example, the sequence A+B is treated as A++B because
+there is no point in backtracking into a sequence of A's when B must follow.
+

+

+When a pattern contains an unlimited repeat inside a subpattern that can itself
+be repeated an unlimited number of times, the use of an atomic group is the
+only way to avoid some failing matches taking a very long time indeed. The
+pattern
+
+  (\D+|<\d+>)*[!?]
+

+matches an unlimited number of substrings that either consist of non-digits, or
+digits enclosed in <>, followed by either ! or ?. When it matches, it runs
+quickly. However, if it is applied to
+
+  aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+

+it takes a long time before reporting failure. This is because the string can
+be divided between the internal \D+ repeat and the external * repeat in a
+large number of ways, and all have to be tried. (The example uses [!?] rather
+than a single character at the end, because both PCRE and Perl have an
+optimization that allows for fast failure when a single character is used. They
+remember the last single character that is required for a match, and fail early
+if it is not present in the string.) If the pattern is changed so that it uses
+an atomic group, like this:
+
+  ((?>\D+)|<\d+>)*[!?]
+

+sequences of non-digits cannot be broken, and failure happens quickly.
+

+
BACK REFERENCES

+

+Outside a character class, a backslash followed by a digit greater than 0 (and
+possibly further digits) is a back reference to a capturing subpattern earlier
+(that is, to its left) in the pattern, provided there have been that many
+previous capturing left parentheses.
+

+

+However, if the decimal number following the backslash is less than 10, it is
+always taken as a back reference, and causes an error only if there are not
+that many capturing left parentheses in the entire pattern. In other words, the
+parentheses that are referenced need not be to the left of the reference for
+numbers less than 10. A "forward back reference" of this type can make sense
+when a repetition is involved and the subpattern to the right has participated
+in an earlier iteration.
+

+

+It is not possible to have a numerical "forward back reference" to a subpattern
+whose number is 10 or more using this syntax because a sequence such as \50 is
+interpreted as a character defined in octal. See the subsection entitled
+"Non-printing characters"
+above
+for further details of the handling of digits following a backslash. There is
+no such problem when named parentheses are used. A back reference to any
+subpattern is possible using named parentheses (see below).
+

+

+Another way of avoiding the ambiguity inherent in the use of digits following a
+backslash is to use the \g escape sequence. This escape must be followed by an
+unsigned number or a negative number, optionally enclosed in braces. These
+examples are all identical:
+
+  (ring), \1
+  (ring), \g1
+  (ring), \g{1}
+

+An unsigned number specifies an absolute reference without the ambiguity that
+is present in the older syntax. It is also useful when literal digits follow
+the reference. A negative number is a relative reference. Consider this
+example:
+
+  (abc(def)ghi)\g{-1}
+

+The sequence \g{-1} is a reference to the most recently started capturing
+subpattern before \g, that is, is it equivalent to \2 in this example.
+Similarly, \g{-2} would be equivalent to \1. The use of relative references
+can be helpful in long patterns, and also in patterns that are created by
+joining together fragments that contain references within themselves.
+

+

+A back reference matches whatever actually matched the capturing subpattern in
+the current subject string, rather than anything matching the subpattern
+itself (see
+"Subpatterns as subroutines"
+below for a way of doing that). So the pattern
+
+  (sens|respons)e and \1ibility
+

+matches "sense and sensibility" and "response and responsibility", but not
+"sense and responsibility". If caseful matching is in force at the time of the
+back reference, the case of letters is relevant. For example,
+
+  ((?i)rah)\s+\1
+

+matches "rah rah" and "RAH RAH", but not "RAH rah", even though the original
+capturing subpattern is matched caselessly.
+

+

+There are several different ways of writing back references to named
+subpatterns. The .NET syntax \k{name} and the Perl syntax \k<name> or
+\k'name' are supported, as is the Python syntax (?P=name). Perl 5.10's unified
+back reference syntax, in which \g can be used for both numeric and named
+references, is also supported. We could rewrite the above example in any of
+the following ways:
+
+  (?<p1>(?i)rah)\s+\k<p1>
+  (?'p1'(?i)rah)\s+\k{p1}
+  (?P<p1>(?i)rah)\s+(?P=p1)
+  (?<p1>(?i)rah)\s+\g{p1}
+

+A subpattern that is referenced by name may appear in the pattern before or
+after the reference.
+

+

+There may be more than one back reference to the same subpattern. If a
+subpattern has not actually been used in a particular match, any back
+references to it always fail by default. For example, the pattern
+
+  (a|(bc))\2
+

+always fails if it starts to match "a" rather than "bc". However, if the
+PCRE_JAVASCRIPT_COMPAT option is set at compile time, a back reference to an
+unset value matches an empty string.
+

+

+Because there may be many capturing parentheses in a pattern, all digits
+following a backslash are taken as part of a potential back reference number.
+If the pattern continues with a digit character, some delimiter must be used to
+terminate the back reference. If the PCRE_EXTENDED option is set, this can be
+white space. Otherwise, the \g{ syntax or an empty comment (see
+"Comments"
+below) can be used.
+

+

+Recursive back references
+

+

+A back reference that occurs inside the parentheses to which it refers fails
+when the subpattern is first used, so, for example, (a\1) never matches.
+However, such references can be useful inside repeated subpatterns. For
+example, the pattern
+
+  (a|b\1)+
+

+matches any number of "a"s and also "aba", "ababbaa" etc. At each iteration of
+the subpattern, the back reference matches the character string corresponding
+to the previous iteration. In order for this to work, the pattern must be such
+that the first iteration does not need to match the back reference. This can be
+done using alternation, as in the example above, or by a quantifier with a
+minimum of zero.
+

+

+Back references of this type cause the group that they reference to be treated
+as an
+atomic group.
+Once the whole group has been matched, a subsequent matching failure cannot
+cause backtracking into the middle of the group.
+

+
ASSERTIONS

+

+An assertion is a test on the characters following or preceding the current
+matching point that does not actually consume any characters. The simple
+assertions coded as \b, \B, \A, \G, \Z, \z, ^ and $ are described
+above.
+

+

+More complicated assertions are coded as subpatterns. There are two kinds:
+those that look ahead of the current position in the subject string, and those
+that look behind it. An assertion subpattern is matched in the normal way,
+except that it does not cause the current matching position to be changed.
+

+

+Assertion subpatterns are not capturing subpatterns. If such an assertion
+contains capturing subpatterns within it, these are counted for the purposes of
+numbering the capturing subpatterns in the whole pattern. However, substring
+capturing is carried out only for positive assertions. (Perl sometimes, but not
+always, does do capturing in negative assertions.)
+

+

+For compatibility with Perl, assertion subpatterns may be repeated; though
+it makes no sense to assert the same thing several times, the side effect of
+capturing parentheses may occasionally be useful. In practice, there only three
+cases:
+

+

+(1) If the quantifier is {0}, the assertion is never obeyed during matching.
+However, it may contain internal capturing parenthesized groups that are called
+from elsewhere via the
+subroutine mechanism.
+

+

+(2) If quantifier is {0,n} where n is greater than zero, it is treated as if it
+were {0,1}. At run time, the rest of the pattern match is tried with and
+without the assertion, the order depending on the greediness of the quantifier.
+

+

+(3) If the minimum repetition is greater than zero, the quantifier is ignored.
+The assertion is obeyed just once when encountered during matching.
+

+

+Lookahead assertions
+

+

+Lookahead assertions start with (?= for positive assertions and (?! for
+negative assertions. For example,
+
+  \w+(?=;)
+

+matches a word followed by a semicolon, but does not include the semicolon in
+the match, and
+
+  foo(?!bar)
+

+matches any occurrence of "foo" that is not followed by "bar". Note that the
+apparently similar pattern
+
+  (?!foo)bar
+

+does not find an occurrence of "bar" that is preceded by something other than
+"foo"; it finds any occurrence of "bar" whatsoever, because the assertion
+(?!foo) is always true when the next three characters are "bar". A
+lookbehind assertion is needed to achieve the other effect.
+

+

+If you want to force a matching failure at some point in a pattern, the most
+convenient way to do it is with (?!) because an empty string always matches, so
+an assertion that requires there not to be an empty string must always fail.
+The backtracking control verb (*FAIL) or (*F) is a synonym for (?!).
+

+

+Lookbehind assertions
+

+

+Lookbehind assertions start with (?<= for positive assertions and (?<! for
+negative assertions. For example,
+
+  (?<!foo)bar
+

+does find an occurrence of "bar" that is not preceded by "foo". The contents of
+a lookbehind assertion are restricted such that all the strings it matches must
+have a fixed length. However, if there are several top-level alternatives, they
+do not all have to have the same fixed length. Thus
+
+  (?<=bullock|donkey)
+

+is permitted, but
+
+  (?<!dogs?|cats?)
+

+causes an error at compile time. Branches that match different length strings
+are permitted only at the top level of a lookbehind assertion. This is an
+extension compared with Perl, which requires all branches to match the same
+length of string. An assertion such as
+
+  (?<=ab(c|de))
+

+is not permitted, because its single top-level branch can match two different
+lengths, but it is acceptable to PCRE if rewritten to use two top-level
+branches:
+
+  (?<=abc|abde)
+

+In some cases, the escape sequence \K
+(see above)
+can be used instead of a lookbehind assertion to get round the fixed-length
+restriction.
+

+

+The implementation of lookbehind assertions is, for each alternative, to
+temporarily move the current position back by the fixed length and then try to
+match. If there are insufficient characters before the current position, the
+assertion fails.
+

+

+In a UTF mode, PCRE does not allow the \C escape (which matches a single data
+unit even in a UTF mode) to appear in lookbehind assertions, because it makes
+it impossible to calculate the length of the lookbehind. The \X and \R
+escapes, which can match different numbers of data units, are also not
+permitted.
+

+

+"Subroutine"
+calls (see below) such as (?2) or (?&X) are permitted in lookbehinds, as long
+as the subpattern matches a fixed-length string.
+Recursion,
+however, is not supported.
+

+

+Possessive quantifiers can be used in conjunction with lookbehind assertions to
+specify efficient matching of fixed-length strings at the end of subject
+strings. Consider a simple pattern such as
+
+  abcd$
+

+when applied to a long string that does not match. Because matching proceeds
+from left to right, PCRE will look for each "a" in the subject and then see if
+what follows matches the rest of the pattern. If the pattern is specified as
+
+  ^.*abcd$
+

+the initial .* matches the entire string at first, but when this fails (because
+there is no following "a"), it backtracks to match all but the last character,
+then all but the last two characters, and so on. Once again the search for "a"
+covers the entire string, from right to left, so we are no better off. However,
+if the pattern is written as
+
+  ^.*+(?<=abcd)
+

+there can be no backtracking for the .*+ item; it can match only the entire
+string. The subsequent lookbehind assertion does a single test on the last four
+characters. If it fails, the match fails immediately. For long strings, this
+approach makes a significant difference to the processing time.
+

+

+Using multiple assertions
+

+

+Several assertions (of any sort) may occur in succession. For example,
+
+  (?<=\d{3})(?<!999)foo
+

+matches "foo" preceded by three digits that are not "999". Notice that each of
+the assertions is applied independently at the same point in the subject
+string. First there is a check that the previous three characters are all
+digits, and then there is a check that the same three characters are not "999".
+This pattern does not match "foo" preceded by six characters, the first
+of which are digits and the last three of which are not "999". For example, it
+doesn't match "123abcfoo". A pattern to do that is
+
+  (?<=\d{3}...)(?<!999)foo
+

+This time the first assertion looks at the preceding six characters, checking
+that the first three are digits, and then the second assertion checks that the
+preceding three characters are not "999".
+

+

+Assertions can be nested in any combination. For example,
+
+  (?<=(?<!foo)bar)baz
+

+matches an occurrence of "baz" that is preceded by "bar" which in turn is not
+preceded by "foo", while
+
+  (?<=\d{3}(?!999)...)foo
+

+is another pattern that matches "foo" preceded by three digits and any three
+characters that are not "999".
+

+
CONDITIONAL SUBPATTERNS

+

+It is possible to cause the matching process to obey a subpattern
+conditionally or to choose between two alternative subpatterns, depending on
+the result of an assertion, or whether a specific capturing subpattern has
+already been matched. The two possible forms of conditional subpattern are:
+
+  (?(condition)yes-pattern)
+  (?(condition)yes-pattern|no-pattern)
+

+If the condition is satisfied, the yes-pattern is used; otherwise the
+no-pattern (if present) is used. If there are more than two alternatives in the
+subpattern, a compile-time error occurs. Each of the two alternatives may
+itself contain nested subpatterns of any form, including conditional
+subpatterns; the restriction to two alternatives applies only at the level of
+the condition. This pattern fragment is an example where the alternatives are
+complex:
+
+  (?(1) (A|B|C) | (D | (?(2)E|F) | E) )
+
+

+

+

+There are four kinds of condition: references to subpatterns, references to
+recursion, a pseudo-condition called DEFINE, and assertions.
+

+

+Checking for a used subpattern by number
+

+

+If the text between the parentheses consists of a sequence of digits, the
+condition is true if a capturing subpattern of that number has previously
+matched. If there is more than one capturing subpattern with the same number
+(see the earlier
+section about duplicate subpattern numbers),
+the condition is true if any of them have matched. An alternative notation is
+to precede the digits with a plus or minus sign. In this case, the subpattern
+number is relative rather than absolute. The most recently opened parentheses
+can be referenced by (?(-1), the next most recent by (?(-2), and so on. Inside
+loops it can also make sense to refer to subsequent groups. The next
+parentheses to be opened can be referenced as (?(+1), and so on. (The value
+zero in any of these forms is not used; it provokes a compile-time error.)
+

+

+Consider the following pattern, which contains non-significant white space to
+make it more readable (assume the PCRE_EXTENDED option) and to divide it into
+three parts for ease of discussion:
+
+  ( \( )?    [^()]+    (?(1) \) )
+

+The first part matches an optional opening parenthesis, and if that
+character is present, sets it as the first captured substring. The second part
+matches one or more characters that are not parentheses. The third part is a
+conditional subpattern that tests whether or not the first set of parentheses
+matched. If they did, that is, if subject started with an opening parenthesis,
+the condition is true, and so the yes-pattern is executed and a closing
+parenthesis is required. Otherwise, since no-pattern is not present, the
+subpattern matches nothing. In other words, this pattern matches a sequence of
+non-parentheses, optionally enclosed in parentheses.
+

+

+If you were embedding this pattern in a larger one, you could use a relative
+reference:
+
+  ...other stuff... ( \( )?    [^()]+    (?(-1) \) ) ...
+

+This makes the fragment independent of the parentheses in the larger pattern.
+

+

+Checking for a used subpattern by name
+

+

+Perl uses the syntax (?(<name>)...) or (?('name')...) to test for a used
+subpattern by name. For compatibility with earlier versions of PCRE, which had
+this facility before Perl, the syntax (?(name)...) is also recognized. However,
+there is a possible ambiguity with this syntax, because subpattern names may
+consist entirely of digits. PCRE looks first for a named subpattern; if it
+cannot find one and the name consists entirely of digits, PCRE looks for a
+subpattern of that number, which must be greater than zero. Using subpattern
+names that consist entirely of digits is not recommended.
+

+

+Rewriting the above example to use a named subpattern gives this:
+
+  (?<OPEN> \( )?    [^()]+    (?(<OPEN>) \) )
+

+If the name used in a condition of this kind is a duplicate, the test is
+applied to all subpatterns of the same name, and is true if any one of them has
+matched.
+

+

+Checking for pattern recursion
+

+

+If the condition is the string (R), and there is no subpattern with the name R,
+the condition is true if a recursive call to the whole pattern or any
+subpattern has been made. If digits or a name preceded by ampersand follow the
+letter R, for example:
+
+  (?(R3)...) or (?(R&name)...)
+

+the condition is true if the most recent recursion is into a subpattern whose
+number or name is given. This condition does not check the entire recursion
+stack. If the name used in a condition of this kind is a duplicate, the test is
+applied to all subpatterns of the same name, and is true if any one of them is
+the most recent recursion.
+

+

+At "top level", all these recursion test conditions are false.
+The syntax for recursive patterns
+is described below.
+

+

+Defining subpatterns for use by reference only
+

+

+If the condition is the string (DEFINE), and there is no subpattern with the
+name DEFINE, the condition is always false. In this case, there may be only one
+alternative in the subpattern. It is always skipped if control reaches this
+point in the pattern; the idea of DEFINE is that it can be used to define
+subroutines that can be referenced from elsewhere. (The use of
+subroutines
+is described below.) For example, a pattern to match an IPv4 address such as
+"192.168.23.245" could be written like this (ignore white space and line
+breaks):
+
+  (?(DEFINE) (?<byte> 2[0-4]\d | 25[0-5] | 1\d\d | [1-9]?\d) )
+  \b (?&byte) (\.(?&byte)){3} \b
+

+The first part of the pattern is a DEFINE group inside which a another group
+named "byte" is defined. This matches an individual component of an IPv4
+address (a number less than 256). When matching takes place, this part of the
+pattern is skipped because DEFINE acts like a false condition. The rest of the
+pattern uses references to the named group to match the four dot-separated
+components of an IPv4 address, insisting on a word boundary at each end.
+

+

+Assertion conditions
+

+

+If the condition is not in any of the above formats, it must be an assertion.
+This may be a positive or negative lookahead or lookbehind assertion. Consider
+this pattern, again containing non-significant white space, and with the two
+alternatives on the second line:
+
+  (?(?=[^a-z]*[a-z])
+  \d{2}-[a-z]{3}-\d{2}  |  \d{2}-\d{2}-\d{2} )
+

+The condition is a positive lookahead assertion that matches an optional
+sequence of non-letters followed by a letter. In other words, it tests for the
+presence of at least one letter in the subject. If a letter is found, the
+subject is matched against the first alternative; otherwise it is matched
+against the second. This pattern matches strings in one of the two forms
+dd-aaa-dd or dd-dd-dd, where aaa are letters and dd are digits.
+

+
COMMENTS

+

+There are two ways of including comments in patterns that are processed by
+PCRE. In both cases, the start of the comment must not be in a character class,
+nor in the middle of any other sequence of related characters such as (?: or a
+subpattern name or number. The characters that make up a comment play no part
+in the pattern matching.
+

+

+The sequence (?# marks the start of a comment that continues up to the next
+closing parenthesis. Nested parentheses are not permitted. If the PCRE_EXTENDED
+option is set, an unescaped # character also introduces a comment, which in
+this case continues to immediately after the next newline character or
+character sequence in the pattern. Which characters are interpreted as newlines
+is controlled by the options passed to a compiling function or by a special
+sequence at the start of the pattern, as described in the section entitled
+"Newline conventions"
+above. Note that the end of this type of comment is a literal newline sequence
+in the pattern; escape sequences that happen to represent a newline do not
+count. For example, consider this pattern when PCRE_EXTENDED is set, and the
+default newline convention is in force:
+
+  abc #comment \n still comment
+

+On encountering the # character, pcre_compile() skips along, looking for
+a newline in the pattern. The sequence \n is still literal at this stage, so
+it does not terminate the comment. Only an actual character with the code value
+0x0a (the default newline) does so.
+

+
RECURSIVE PATTERNS

+

+Consider the problem of matching a string in parentheses, allowing for
+unlimited nested parentheses. Without the use of recursion, the best that can
+be done is to use a pattern that matches up to some fixed depth of nesting. It
+is not possible to handle an arbitrary nesting depth.
+

+

+For some time, Perl has provided a facility that allows regular expressions to
+recurse (amongst other things). It does this by interpolating Perl code in the
+expression at run time, and the code can refer to the expression itself. A Perl
+pattern using code interpolation to solve the parentheses problem can be
+created like this:
+
+  $re = qr{\( (?: (?>[^()]+) | (?p{$re}) )* \)}x;
+

+The (?p{...}) item interpolates Perl code at run time, and in this case refers
+recursively to the pattern in which it appears.
+

+

+Obviously, PCRE cannot support the interpolation of Perl code. Instead, it
+supports special syntax for recursion of the entire pattern, and also for
+individual subpattern recursion. After its introduction in PCRE and Python,
+this kind of recursion was subsequently introduced into Perl at release 5.10.
+

+

+A special item that consists of (? followed by a number greater than zero and a
+closing parenthesis is a recursive subroutine call of the subpattern of the
+given number, provided that it occurs inside that subpattern. (If not, it is a
+non-recursive subroutine
+call, which is described in the next section.) The special item (?R) or (?0) is
+a recursive call of the entire regular expression.
+

+

+This PCRE pattern solves the nested parentheses problem (assume the
+PCRE_EXTENDED option is set so that white space is ignored):
+
+  \( ( [^()]++ | (?R) )* \)
+

+First it matches an opening parenthesis. Then it matches any number of
+substrings which can either be a sequence of non-parentheses, or a recursive
+match of the pattern itself (that is, a correctly parenthesized substring).
+Finally there is a closing parenthesis. Note the use of a possessive quantifier
+to avoid backtracking into sequences of non-parentheses.
+

+

+If this were part of a larger pattern, you would not want to recurse the entire
+pattern, so instead you could use this:
+
+  ( \( ( [^()]++ | (?1) )* \) )
+

+We have put the pattern into parentheses, and caused the recursion to refer to
+them instead of the whole pattern.
+

+

+In a larger pattern, keeping track of parenthesis numbers can be tricky. This
+is made easier by the use of relative references. Instead of (?1) in the
+pattern above you can write (?-2) to refer to the second most recently opened
+parentheses preceding the recursion. In other words, a negative number counts
+capturing parentheses leftwards from the point at which it is encountered.
+

+

+It is also possible to refer to subsequently opened parentheses, by writing
+references such as (?+2). However, these cannot be recursive because the
+reference is not inside the parentheses that are referenced. They are always
+non-recursive subroutine
+calls, as described in the next section.
+

+

+An alternative approach is to use named parentheses instead. The Perl syntax
+for this is (?&name); PCRE's earlier syntax (?P>name) is also supported. We
+could rewrite the above example as follows:
+
+  (?<pn> \( ( [^()]++ | (?&pn) )* \) )
+

+If there is more than one subpattern with the same name, the earliest one is
+used.
+

+

+This particular example pattern that we have been looking at contains nested
+unlimited repeats, and so the use of a possessive quantifier for matching
+strings of non-parentheses is important when applying the pattern to strings
+that do not match. For example, when this pattern is applied to
+
+  (aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa()
+

+it yields "no match" quickly. However, if a possessive quantifier is not used,
+the match runs for a very long time indeed because there are so many different
+ways the + and * repeats can carve up the subject, and all have to be tested
+before failure can be reported.
+

+

+At the end of a match, the values of capturing parentheses are those from
+the outermost level. If you want to obtain intermediate values, a callout
+function can be used (see below and the
+pcrecallout
+documentation). If the pattern above is matched against
+
+  (ab(cd)ef)
+

+the value for the inner capturing parentheses (numbered 2) is "ef", which is
+the last value taken on at the top level. If a capturing subpattern is not
+matched at the top level, its final captured value is unset, even if it was
+(temporarily) set at a deeper level during the matching process.
+

+

+If there are more than 15 capturing parentheses in a pattern, PCRE has to
+obtain extra memory to store data during a recursion, which it does by using
+pcre_malloc, freeing it via pcre_free afterwards. If no memory can
+be obtained, the match fails with the PCRE_ERROR_NOMEMORY error.
+

+

+Do not confuse the (?R) item with the condition (R), which tests for recursion.
+Consider this pattern, which matches text in angle brackets, allowing for
+arbitrary nesting. Only digits are allowed in nested brackets (that is, when
+recursing), whereas any characters are permitted at the outer level.
+
+  < (?: (?(R) \d++  | [^<>]*+) | (?R)) * >
+

+In this pattern, (?(R) is the start of a conditional subpattern, with two
+different alternatives for the recursive and non-recursive cases. The (?R) item
+is the actual recursive call.
+

+

+Differences in recursion processing between PCRE and Perl
+

+

+Recursion processing in PCRE differs from Perl in two important ways. In PCRE
+(like Python, but unlike Perl), a recursive subpattern call is always treated
+as an atomic group. That is, once it has matched some of the subject string, it
+is never re-entered, even if it contains untried alternatives and there is a
+subsequent matching failure. This can be illustrated by the following pattern,
+which purports to match a palindromic string that contains an odd number of
+characters (for example, "a", "aba", "abcba", "abcdcba"):
+
+  ^(.|(.)(?1)\2)$
+

+The idea is that it either matches a single character, or two identical
+characters surrounding a sub-palindrome. In Perl, this pattern works; in PCRE
+it does not if the pattern is longer than three characters. Consider the
+subject string "abcba":
+

+

+At the top level, the first character is matched, but as it is not at the end
+of the string, the first alternative fails; the second alternative is taken
+and the recursion kicks in. The recursive call to subpattern 1 successfully
+matches the next character ("b"). (Note that the beginning and end of line
+tests are not part of the recursion).
+

+

+Back at the top level, the next character ("c") is compared with what
+subpattern 2 matched, which was "a". This fails. Because the recursion is
+treated as an atomic group, there are now no backtracking points, and so the
+entire match fails. (Perl is able, at this point, to re-enter the recursion and
+try the second alternative.) However, if the pattern is written with the
+alternatives in the other order, things are different:
+
+  ^((.)(?1)\2|.)$
+

+This time, the recursing alternative is tried first, and continues to recurse
+until it runs out of characters, at which point the recursion fails. But this
+time we do have another alternative to try at the higher level. That is the big
+difference: in the previous case the remaining alternative is at a deeper
+recursion level, which PCRE cannot use.
+

+

+To change the pattern so that it matches all palindromic strings, not just
+those with an odd number of characters, it is tempting to change the pattern to
+this:
+
+  ^((.)(?1)\2|.?)$
+

+Again, this works in Perl, but not in PCRE, and for the same reason. When a
+deeper recursion has matched a single character, it cannot be entered again in
+order to match an empty string. The solution is to separate the two cases, and
+write out the odd and even cases as alternatives at the higher level:
+
+  ^(?:((.)(?1)\2|)|((.)(?3)\4|.))
+

+If you want to match typical palindromic phrases, the pattern has to ignore all
+non-word characters, which can be done like this:
+
+  ^\W*+(?:((.)\W*+(?1)\W*+\2|)|((.)\W*+(?3)\W*+\4|\W*+.\W*+))\W*+$
+

+If run with the PCRE_CASELESS option, this pattern matches phrases such as "A
+man, a plan, a canal: Panama!" and it works well in both PCRE and Perl. Note
+the use of the possessive quantifier *+ to avoid backtracking into sequences of
+non-word characters. Without this, PCRE takes a great deal longer (ten times or
+more) to match typical phrases, and Perl takes so long that you think it has
+gone into a loop.
+

+

+WARNING: The palindrome-matching patterns above work only if the subject
+string does not start with a palindrome that is shorter than the entire string.
+For example, although "abcba" is correctly matched, if the subject is "ababa",
+PCRE finds the palindrome "aba" at the start, then fails at top level because
+the end of the string does not follow. Once again, it cannot jump back into the
+recursion to try other alternatives, so the entire match fails.
+

+

+The second way in which PCRE and Perl differ in their recursion processing is
+in the handling of captured values. In Perl, when a subpattern is called
+recursively or as a subpattern (see the next section), it has no access to any
+values that were captured outside the recursion, whereas in PCRE these values
+can be referenced. Consider this pattern:
+
+  ^(.)(\1|a(?2))
+

+In PCRE, this pattern matches "bab". The first capturing parentheses match "b",
+then in the second group, when the back reference \1 fails to match "b", the
+second alternative matches "a" and then recurses. In the recursion, \1 does
+now match "b" and so the whole match succeeds. In Perl, the pattern fails to
+match because inside the recursive call \1 cannot access the externally set
+value.
+

+
SUBPATTERNS AS SUBROUTINES

+

+If the syntax for a recursive subpattern call (either by number or by
+name) is used outside the parentheses to which it refers, it operates like a
+subroutine in a programming language. The called subpattern may be defined
+before or after the reference. A numbered reference can be absolute or
+relative, as in these examples:
+
+  (...(absolute)...)...(?2)...
+  (...(relative)...)...(?-1)...
+  (...(?+1)...(relative)...
+

+An earlier example pointed out that the pattern
+
+  (sens|respons)e and \1ibility
+

+matches "sense and sensibility" and "response and responsibility", but not
+"sense and responsibility". If instead the pattern
+
+  (sens|respons)e and (?1)ibility
+

+is used, it does match "sense and responsibility" as well as the other two
+strings. Another example is given in the discussion of DEFINE above.
+

+

+All subroutine calls, whether recursive or not, are always treated as atomic
+groups. That is, once a subroutine has matched some of the subject string, it
+is never re-entered, even if it contains untried alternatives and there is a
+subsequent matching failure. Any capturing parentheses that are set during the
+subroutine call revert to their previous values afterwards.
+

+

+Processing options such as case-independence are fixed when a subpattern is
+defined, so if it is used as a subroutine, such options cannot be changed for
+different calls. For example, consider this pattern:
+
+  (abc)(?i:(?-1))
+

+It matches "abcabc". It does not match "abcABC" because the change of
+processing option does not affect the called subpattern.
+

+
ONIGURUMA SUBROUTINE SYNTAX

+

+For compatibility with Oniguruma, the non-Perl syntax \g followed by a name or
+a number enclosed either in angle brackets or single quotes, is an alternative
+syntax for referencing a subpattern as a subroutine, possibly recursively. Here
+are two of the examples used above, rewritten using this syntax:
+
+  (?<pn> \( ( (?>[^()]+) | \g<pn> )* \) )
+  (sens|respons)e and \g'1'ibility
+

+PCRE supports an extension to Oniguruma: if a number is preceded by a
+plus or a minus sign it is taken as a relative reference. For example:
+
+  (abc)(?i:\g<-1>)
+

+Note that \g{...} (Perl syntax) and \g<...> (Oniguruma syntax) are not
+synonymous. The former is a back reference; the latter is a subroutine call.
+

+
CALLOUTS

+

+Perl has a feature whereby using the sequence (?{...}) causes arbitrary Perl
+code to be obeyed in the middle of matching a regular expression. This makes it
+possible, amongst other things, to extract different substrings that match the
+same pair of parentheses when there is a repetition.
+

+

+PCRE provides a similar feature, but of course it cannot obey arbitrary Perl
+code. The feature is called "callout". The caller of PCRE provides an external
+function by putting its entry point in the global variable pcre_callout
+(8-bit library) or pcre[16|32]_callout (16-bit or 32-bit library).
+By default, this variable contains NULL, which disables all calling out.
+

+

+Within a regular expression, (?C) indicates the points at which the external
+function is to be called. If you want to identify different callout points, you
+can put a number less than 256 after the letter C. The default value is zero.
+For example, this pattern has two callout points:
+
+  (?C1)abc(?C2)def
+

+If the PCRE_AUTO_CALLOUT flag is passed to a compiling function, callouts are
+automatically installed before each item in the pattern. They are all numbered
+255. If there is a conditional group in the pattern whose condition is an
+assertion, an additional callout is inserted just before the condition. An
+explicit callout may also be set at this position, as in this example:
+
+  (?(?C9)(?=a)abc|def)
+

+Note that this applies only to assertion conditions, not to other types of
+condition.
+

+

+During matching, when PCRE reaches a callout point, the external function is
+called. It is provided with the number of the callout, the position in the
+pattern, and, optionally, one item of data originally supplied by the caller of
+the matching function. The callout function may cause matching to proceed, to
+backtrack, or to fail altogether. A complete description of the interface to
+the callout function is given in the
+pcrecallout
+documentation.
+

+
BACKTRACKING CONTROL

+

+Perl 5.10 introduced a number of "Special Backtracking Control Verbs", which
+are still described in the Perl documentation as "experimental and subject to
+change or removal in a future version of Perl". It goes on to say: "Their usage
+in production code should be noted to avoid problems during upgrades." The same
+remarks apply to the PCRE features described in this section.
+

+

+The new verbs make use of what was previously invalid syntax: an opening
+parenthesis followed by an asterisk. They are generally of the form
+(*VERB) or (*VERB:NAME). Some may take either form, possibly behaving
+differently depending on whether or not a name is present. A name is any
+sequence of characters that does not include a closing parenthesis. The maximum
+length of name is 255 in the 8-bit library and 65535 in the 16-bit and 32-bit
+libraries. If the name is empty, that is, if the closing parenthesis
+immediately follows the colon, the effect is as if the colon were not there.
+Any number of these verbs may occur in a pattern.
+

+

+Since these verbs are specifically related to backtracking, most of them can be
+used only when the pattern is to be matched using one of the traditional
+matching functions, because these use a backtracking algorithm. With the
+exception of (*FAIL), which behaves like a failing negative assertion, the
+backtracking control verbs cause an error if encountered by a DFA matching
+function.
+

+

+The behaviour of these verbs in
+repeated groups,
+assertions,
+and in
+subpatterns called as subroutines
+(whether or not recursively) is documented below.
+

+

+Optimizations that affect backtracking verbs
+

+

+PCRE contains some optimizations that are used to speed up matching by running
+some checks at the start of each match attempt. For example, it may know the
+minimum length of matching subject, or that a particular character must be
+present. When one of these optimizations bypasses the running of a match, any
+included backtracking verbs will not, of course, be processed. You can suppress
+the start-of-match optimizations by setting the PCRE_NO_START_OPTIMIZE option
+when calling pcre_compile() or pcre_exec(), or by starting the
+pattern with (*NO_START_OPT). There is more discussion of this option in the
+section entitled
+"Option bits for pcre_exec()"
+in the
+pcreapi
+documentation.
+

+

+Experiments with Perl suggest that it too has similar optimizations, sometimes
+leading to anomalous results.
+

+

+Verbs that act immediately
+

+

+The following verbs act as soon as they are encountered. They may not be
+followed by a name.
+
+   (*ACCEPT)
+

+This verb causes the match to end successfully, skipping the remainder of the
+pattern. However, when it is inside a subpattern that is called as a
+subroutine, only that subpattern is ended successfully. Matching then continues
+at the outer level. If (*ACCEPT) in triggered in a positive assertion, the
+assertion succeeds; in a negative assertion, the assertion fails.
+

+

+If (*ACCEPT) is inside capturing parentheses, the data so far is captured. For
+example:
+
+  A((?:A|B(*ACCEPT)|C)D)
+

+This matches "AB", "AAD", or "ACD"; when it matches "AB", "B" is captured by
+the outer parentheses.
+
+  (*FAIL) or (*F)
+

+This verb causes a matching failure, forcing backtracking to occur. It is
+equivalent to (?!) but easier to read. The Perl documentation notes that it is
+probably useful only when combined with (?{}) or (??{}). Those are, of course,
+Perl features that are not present in PCRE. The nearest equivalent is the
+callout feature, as for example in this pattern:
+
+  a+(?C)(*FAIL)
+

+A match with the string "aaaa" always fails, but the callout is taken before
+each backtrack happens (in this example, 10 times).
+

+

+Recording which path was taken
+

+

+There is one verb whose main purpose is to track how a match was arrived at,
+though it also has a secondary use in conjunction with advancing the match
+starting point (see (*SKIP) below).
+
+  (*MARK:NAME) or (*:NAME)
+

+A name is always required with this verb. There may be as many instances of
+(*MARK) as you like in a pattern, and their names do not have to be unique.
+

+

+When a match succeeds, the name of the last-encountered (*MARK:NAME),
+(*PRUNE:NAME), or (*THEN:NAME) on the matching path is passed back to the
+caller as described in the section entitled
+"Extra data for pcre_exec()"
+in the
+pcreapi
+documentation. Here is an example of pcretest output, where the /K
+modifier requests the retrieval and outputting of (*MARK) data:
+
+    re> /X(*MARK:A)Y|X(*MARK:B)Z/K
+  data> XY
+   0: XY
+  MK: A
+  XZ
+   0: XZ
+  MK: B
+

+The (*MARK) name is tagged with "MK:" in this output, and in this example it
+indicates which of the two alternatives matched. This is a more efficient way
+of obtaining this information than putting each alternative in its own
+capturing parentheses.
+

+

+If a verb with a name is encountered in a positive assertion that is true, the
+name is recorded and passed back if it is the last-encountered. This does not
+happen for negative assertions or failing positive assertions.
+

+

+After a partial match or a failed match, the last encountered name in the
+entire match process is returned. For example:
+
+    re> /X(*MARK:A)Y|X(*MARK:B)Z/K
+  data> XP
+  No match, mark = B
+

+Note that in this unanchored example the mark is retained from the match
+attempt that started at the letter "X" in the subject. Subsequent match
+attempts starting at "P" and then with an empty string do not get as far as the
+(*MARK) item, but nevertheless do not reset it.
+

+

+If you are interested in (*MARK) values after failed matches, you should
+probably set the PCRE_NO_START_OPTIMIZE option
+(see above)
+to ensure that the match is always attempted.
+

+

+Verbs that act after backtracking
+

+

+The following verbs do nothing when they are encountered. Matching continues
+with what follows, but if there is no subsequent match, causing a backtrack to
+the verb, a failure is forced. That is, backtracking cannot pass to the left of
+the verb. However, when one of these verbs appears inside an atomic group or an
+assertion that is true, its effect is confined to that group, because once the
+group has been matched, there is never any backtracking into it. In this
+situation, backtracking can "jump back" to the left of the entire atomic group
+or assertion. (Remember also, as stated above, that this localization also
+applies in subroutine calls.)
+

+

+These verbs differ in exactly what kind of failure occurs when backtracking
+reaches them. The behaviour described below is what happens when the verb is
+not in a subroutine or an assertion. Subsequent sections cover these special
+cases.
+
+  (*COMMIT)
+

+This verb, which may not be followed by a name, causes the whole match to fail
+outright if there is a later matching failure that causes backtracking to reach
+it. Even if the pattern is unanchored, no further attempts to find a match by
+advancing the starting point take place. If (*COMMIT) is the only backtracking
+verb that is encountered, once it has been passed pcre_exec() is
+committed to finding a match at the current starting point, or not at all. For
+example:
+
+  a+(*COMMIT)b
+

+This matches "xxaab" but not "aacaab". It can be thought of as a kind of
+dynamic anchor, or "I've started, so I must finish." The name of the most
+recently passed (*MARK) in the path is passed back when (*COMMIT) forces a
+match failure.
+

+

+If there is more than one backtracking verb in a pattern, a different one that
+follows (*COMMIT) may be triggered first, so merely passing (*COMMIT) during a
+match does not always guarantee that a match must be at this starting point.
+

+

+Note that (*COMMIT) at the start of a pattern is not the same as an anchor,
+unless PCRE's start-of-match optimizations are turned off, as shown in this
+pcretest example:
+
+    re> /(*COMMIT)abc/
+  data> xyzabc
+   0: abc
+  xyzabc\Y
+  No match
+

+PCRE knows that any match must start with "a", so the optimization skips along
+the subject to "a" before running the first match attempt, which succeeds. When
+the optimization is disabled by the \Y escape in the second subject, the match
+starts at "x" and so the (*COMMIT) causes it to fail without trying any other
+starting points.
+
+  (*PRUNE) or (*PRUNE:NAME)
+

+This verb causes the match to fail at the current starting position in the
+subject if there is a later matching failure that causes backtracking to reach
+it. If the pattern is unanchored, the normal "bumpalong" advance to the next
+starting character then happens. Backtracking can occur as usual to the left of
+(*PRUNE), before it is reached, or when matching to the right of (*PRUNE), but
+if there is no match to the right, backtracking cannot cross (*PRUNE). In
+simple cases, the use of (*PRUNE) is just an alternative to an atomic group or
+possessive quantifier, but there are some uses of (*PRUNE) that cannot be
+expressed in any other way. In an anchored pattern (*PRUNE) has the same effect
+as (*COMMIT).
+

+

+The behaviour of (*PRUNE:NAME) is the not the same as (*MARK:NAME)(*PRUNE).
+It is like (*MARK:NAME) in that the name is remembered for passing back to the
+caller. However, (*SKIP:NAME) searches only for names set with (*MARK).
+
+  (*SKIP)
+

+This verb, when given without a name, is like (*PRUNE), except that if the
+pattern is unanchored, the "bumpalong" advance is not to the next character,
+but to the position in the subject where (*SKIP) was encountered. (*SKIP)
+signifies that whatever text was matched leading up to it cannot be part of a
+successful match. Consider:
+
+  a+(*SKIP)b
+

+If the subject is "aaaac...", after the first match attempt fails (starting at
+the first character in the string), the starting point skips on to start the
+next attempt at "c". Note that a possessive quantifer does not have the same
+effect as this example; although it would suppress backtracking during the
+first match attempt, the second attempt would start at the second character
+instead of skipping on to "c".
+
+  (*SKIP:NAME)
+

+When (*SKIP) has an associated name, its behaviour is modified. When it is
+triggered, the previous path through the pattern is searched for the most
+recent (*MARK) that has the same name. If one is found, the "bumpalong" advance
+is to the subject position that corresponds to that (*MARK) instead of to where
+(*SKIP) was encountered. If no (*MARK) with a matching name is found, the
+(*SKIP) is ignored.
+

+

+Note that (*SKIP:NAME) searches only for names set by (*MARK:NAME). It ignores
+names that are set by (*PRUNE:NAME) or (*THEN:NAME).
+
+  (*THEN) or (*THEN:NAME)
+

+This verb causes a skip to the next innermost alternative when backtracking
+reaches it. That is, it cancels any further backtracking within the current
+alternative. Its name comes from the observation that it can be used for a
+pattern-based if-then-else block:
+
+  ( COND1 (*THEN) FOO | COND2 (*THEN) BAR | COND3 (*THEN) BAZ ) ...
+

+If the COND1 pattern matches, FOO is tried (and possibly further items after
+the end of the group if FOO succeeds); on failure, the matcher skips to the
+second alternative and tries COND2, without backtracking into COND1. If that
+succeeds and BAR fails, COND3 is tried. If subsequently BAZ fails, there are no
+more alternatives, so there is a backtrack to whatever came before the entire
+group. If (*THEN) is not inside an alternation, it acts like (*PRUNE).
+

+

+The behaviour of (*THEN:NAME) is the not the same as (*MARK:NAME)(*THEN).
+It is like (*MARK:NAME) in that the name is remembered for passing back to the
+caller. However, (*SKIP:NAME) searches only for names set with (*MARK).
+

+

+A subpattern that does not contain a | character is just a part of the
+enclosing alternative; it is not a nested alternation with only one
+alternative. The effect of (*THEN) extends beyond such a subpattern to the
+enclosing alternative. Consider this pattern, where A, B, etc. are complex
+pattern fragments that do not contain any | characters at this level:
+
+  A (B(*THEN)C) | D
+

+If A and B are matched, but there is a failure in C, matching does not
+backtrack into A; instead it moves to the next alternative, that is, D.
+However, if the subpattern containing (*THEN) is given an alternative, it
+behaves differently:
+
+  A (B(*THEN)C | (*FAIL)) | D
+

+The effect of (*THEN) is now confined to the inner subpattern. After a failure
+in C, matching moves to (*FAIL), which causes the whole subpattern to fail
+because there are no more alternatives to try. In this case, matching does now
+backtrack into A.
+

+

+Note that a conditional subpattern is not considered as having two
+alternatives, because only one is ever used. In other words, the | character in
+a conditional subpattern has a different meaning. Ignoring white space,
+consider:
+
+  ^.*? (?(?=a) a | b(*THEN)c )
+

+If the subject is "ba", this pattern does not match. Because .*? is ungreedy,
+it initially matches zero characters. The condition (?=a) then fails, the
+character "b" is matched, but "c" is not. At this point, matching does not
+backtrack to .*? as might perhaps be expected from the presence of the |
+character. The conditional subpattern is part of the single alternative that
+comprises the whole pattern, and so the match fails. (If there was a backtrack
+into .*?, allowing it to match "b", the match would succeed.)
+

+

+The verbs just described provide four different "strengths" of control when
+subsequent matching fails. (*THEN) is the weakest, carrying on the match at the
+next alternative. (*PRUNE) comes next, failing the match at the current
+starting position, but allowing an advance to the next character (for an
+unanchored pattern). (*SKIP) is similar, except that the advance may be more
+than one character. (*COMMIT) is the strongest, causing the entire match to
+fail.
+

+

+More than one backtracking verb
+

+

+If more than one backtracking verb is present in a pattern, the one that is
+backtracked onto first acts. For example, consider this pattern, where A, B,
+etc. are complex pattern fragments:
+
+  (A(*COMMIT)B(*THEN)C|ABD)
+

+If A matches but B fails, the backtrack to (*COMMIT) causes the entire match to
+fail. However, if A and B match, but C fails, the backtrack to (*THEN) causes
+the next alternative (ABD) to be tried. This behaviour is consistent, but is
+not always the same as Perl's. It means that if two or more backtracking verbs
+appear in succession, all the the last of them has no effect. Consider this
+example:
+
+  ...(*COMMIT)(*PRUNE)...
+

+If there is a matching failure to the right, backtracking onto (*PRUNE) cases
+it to be triggered, and its action is taken. There can never be a backtrack
+onto (*COMMIT).
+

+

+Backtracking verbs in repeated groups
+

+

+PCRE differs from Perl in its handling of backtracking verbs in repeated
+groups. For example, consider:
+
+  /(a(*COMMIT)b)+ac/
+

+If the subject is "abac", Perl matches, but PCRE fails because the (*COMMIT) in
+the second repeat of the group acts.
+

+

+Backtracking verbs in assertions
+

+

+(*FAIL) in an assertion has its normal effect: it forces an immediate backtrack.
+

+

+(*ACCEPT) in a positive assertion causes the assertion to succeed without any
+further processing. In a negative assertion, (*ACCEPT) causes the assertion to
+fail without any further processing.
+

+

+The other backtracking verbs are not treated specially if they appear in a
+positive assertion. In particular, (*THEN) skips to the next alternative in the
+innermost enclosing group that has alternations, whether or not this is within
+the assertion.
+

+

+Negative assertions are, however, different, in order to ensure that changing a
+positive assertion into a negative assertion changes its result. Backtracking
+into (*COMMIT), (*SKIP), or (*PRUNE) causes a negative assertion to be true,
+without considering any further alternative branches in the assertion.
+Backtracking into (*THEN) causes it to skip to the next enclosing alternative
+within the assertion (the normal behaviour), but if the assertion does not have
+such an alternative, (*THEN) behaves like (*PRUNE).
+

+

+Backtracking verbs in subroutines
+

+

+These behaviours occur whether or not the subpattern is called recursively.
+Perl's treatment of subroutines is different in some cases.
+

+

+(*FAIL) in a subpattern called as a subroutine has its normal effect: it forces
+an immediate backtrack.
+

+

+(*ACCEPT) in a subpattern called as a subroutine causes the subroutine match to
+succeed without any further processing. Matching then continues after the
+subroutine call.
+

+

+(*COMMIT), (*SKIP), and (*PRUNE) in a subpattern called as a subroutine cause
+the subroutine match to fail.
+

+

+(*THEN) skips to the next alternative in the innermost enclosing group within
+the subpattern that has alternatives. If there is no such group within the
+subpattern, (*THEN) causes the subroutine match to fail.
+

+
SEE ALSO

+

+pcreapi(3), pcrecallout(3), pcrematching(3),
+pcresyntax(3), pcre(3), pcre16(3), pcre32(3).
+

+
AUTHOR

+

+Philip Hazel
+

+University Computing Service
+

+Cambridge CB2 3QH, England.
+

+

+
REVISION

+

+Last updated: 26 April 2013
+

+Copyright © 1997-2013 University of Cambridge.
+

+

+Return to the PCRE index page.
+

diff --git a/pcre/doc/html/pcreperform.html b/pcre/doc/html/pcreperform.html
new file mode 100644
index 00000000000..dda207f9018
--- /dev/null
+++ b/pcre/doc/html/pcreperform.html
@@ -0,0 +1,195 @@
+
+
+pcreperform specification
+
+
+
pcreperform man page

+

+Return to the PCRE index page.
+

+

+This page is part of the PCRE HTML documentation. It was generated automatically
+from the original man page. If there is any nonsense in it, please consult the
+man page, in case the conversion went wrong.
+

+

+PCRE PERFORMANCE
+

+

+Two aspects of performance are discussed below: memory usage and processing
+time. The way you express your pattern as a regular expression can affect both
+of them.
+

+

+COMPILED PATTERN MEMORY USAGE
+

+

+Patterns are compiled by PCRE into a reasonably efficient interpretive code, so
+that most simple patterns do not use much memory. However, there is one case
+where the memory usage of a compiled pattern can be unexpectedly large. If a
+parenthesized subpattern has a quantifier with a minimum greater than 1 and/or
+a limited maximum, the whole subpattern is repeated in the compiled code. For
+example, the pattern
+
+  (abc|def){2,4}
+

+is compiled as if it were
+
+  (abc|def)(abc|def)((abc|def)(abc|def)?)?
+

+(Technical aside: It is done this way so that backtrack points within each of
+the repetitions can be independently maintained.)
+

+

+For regular expressions whose quantifiers use only small numbers, this is not
+usually a problem. However, if the numbers are large, and particularly if such
+repetitions are nested, the memory usage can become an embarrassment. For
+example, the very simple pattern
+
+  ((ab){1,1000}c){1,3}
+

+uses 51K bytes when compiled using the 8-bit library. When PCRE is compiled
+with its default internal pointer size of two bytes, the size limit on a
+compiled pattern is 64K data units, and this is reached with the above pattern
+if the outer repetition is increased from 3 to 4. PCRE can be compiled to use
+larger internal pointers and thus handle larger compiled patterns, but it is
+better to try to rewrite your pattern to use less memory if you can.
+

+

+One way of reducing the memory usage for such patterns is to make use of PCRE's
+"subroutine"
+facility. Re-writing the above pattern as
+
+  ((ab)(?2){0,999}c)(?1){0,2}
+

+reduces the memory requirements to 18K, and indeed it remains under 20K even
+with the outer repetition increased to 100. However, this pattern is not
+exactly equivalent, because the "subroutine" calls are treated as
+atomic groups
+into which there can be no backtracking if there is a subsequent matching
+failure. Therefore, PCRE cannot do this kind of rewriting automatically.
+Furthermore, there is a noticeable loss of speed when executing the modified
+pattern. Nevertheless, if the atomic grouping is not a problem and the loss of
+speed is acceptable, this kind of rewriting will allow you to process patterns
+that PCRE cannot otherwise handle.
+

+

+STACK USAGE AT RUN TIME
+

+

+When pcre_exec() or pcre[16|32]_exec() is used for matching, certain
+kinds of pattern can cause it to use large amounts of the process stack. In
+some environments the default process stack is quite small, and if it runs out
+the result is often SIGSEGV. This issue is probably the most frequently raised
+problem with PCRE. Rewriting your pattern can often help. The
+pcrestack
+documentation discusses this issue in detail.
+

+

+PROCESSING TIME
+

+

+Certain items in regular expression patterns are processed more efficiently
+than others. It is more efficient to use a character class like [aeiou] than a
+set of single-character alternatives such as (a|e|i|o|u). In general, the
+simplest construction that provides the required behaviour is usually the most
+efficient. Jeffrey Friedl's book contains a lot of useful general discussion
+about optimizing regular expressions for efficient performance. This document
+contains a few observations about PCRE.
+

+

+Using Unicode character properties (the \p, \P, and \X escapes) is slow,
+because PCRE has to use a multi-stage table lookup whenever it needs a
+character's property. If you can find an alternative pattern that does not use
+character properties, it will probably be faster.
+

+

+By default, the escape sequences \b, \d, \s, and \w, and the POSIX
+character classes such as [:alpha:] do not use Unicode properties, partly for
+backwards compatibility, and partly for performance reasons. However, you can
+set PCRE_UCP if you want Unicode character properties to be used. This can
+double the matching time for items such as \d, when matched with
+a traditional matching function; the performance loss is less with
+a DFA matching function, and in both cases there is not much difference for
+\b.
+

+

+When a pattern begins with .* not in parentheses, or in parentheses that are
+not the subject of a backreference, and the PCRE_DOTALL option is set, the
+pattern is implicitly anchored by PCRE, since it can match only at the start of
+a subject string. However, if PCRE_DOTALL is not set, PCRE cannot make this
+optimization, because the . metacharacter does not then match a newline, and if
+the subject string contains newlines, the pattern may match from the character
+immediately following one of them instead of from the very start. For example,
+the pattern
+
+  .*second
+

+matches the subject "first\nand second" (where \n stands for a newline
+character), with the match starting at the seventh character. In order to do
+this, PCRE has to retry the match starting after every newline in the subject.
+

+

+If you are using such a pattern with subject strings that do not contain
+newlines, the best performance is obtained by setting PCRE_DOTALL, or starting
+the pattern with ^.* or ^.*? to indicate explicit anchoring. That saves PCRE
+from having to scan along the subject looking for a newline to restart at.
+

+

+Beware of patterns that contain nested indefinite repeats. These can take a
+long time to run when applied to a string that does not match. Consider the
+pattern fragment
+
+  ^(a+)*
+

+This can match "aaaa" in 16 different ways, and this number increases very
+rapidly as the string gets longer. (The * repeat can match 0, 1, 2, 3, or 4
+times, and for each of those cases other than 0 or 4, the + repeats can match
+different numbers of times.) When the remainder of the pattern is such that the
+entire match is going to fail, PCRE has in principle to try every possible
+variation, and this can take an extremely long time, even for relatively short
+strings.
+

+

+An optimization catches some of the more simple cases such as
+
+  (a+)*b
+

+where a literal character follows. Before embarking on the standard matching
+procedure, PCRE checks that there is a "b" later in the subject string, and if
+there is not, it fails the match immediately. However, when there is no
+following literal this optimization cannot be used. You can see the difference
+by comparing the behaviour of
+
+  (a+)*\d
+

+with the pattern above. The former gives a failure almost instantly when
+applied to a whole line of "a" characters, whereas the latter takes an
+appreciable time with strings longer than about 20 characters.
+

+

+In many cases, the solution to this kind of performance issue is to use an
+atomic group or a possessive quantifier.
+

+

+AUTHOR
+

+

+Philip Hazel
+

+University Computing Service
+

+Cambridge CB2 3QH, England.
+

+

+

+REVISION
+

+

+Last updated: 25 August 2012
+

+Copyright © 1997-2012 University of Cambridge.
+

+

+Return to the PCRE index page.
+

diff --git a/pcre/doc/html/pcreposix.html b/pcre/doc/html/pcreposix.html
new file mode 100644
index 00000000000..0e5b296a086
--- /dev/null
+++ b/pcre/doc/html/pcreposix.html
@@ -0,0 +1,292 @@
+
+
+pcreposix specification
+
+
+
pcreposix man page

+

+Return to the PCRE index page.
+

+

+This page is part of the PCRE HTML documentation. It was generated automatically
+from the original man page. If there is any nonsense in it, please consult the
+man page, in case the conversion went wrong.
+

+

+
SYNOPSIS OF POSIX API

+

+#include <pcreposix.h>
+

+

+int regcomp(regex_t *preg, const char *pattern,
+int cflags);
+

+

+int regexec(regex_t *preg, const char *string,
+size_t nmatch, regmatch_t pmatch[], int eflags);
+

+

+size_t regerror(int errcode, const regex_t *preg,
+char *errbuf, size_t errbuf_size);
+

+

+void regfree(regex_t *preg);
+

+
DESCRIPTION

+

+This set of functions provides a POSIX-style API for the PCRE regular
+expression 8-bit library. See the
+pcreapi
+documentation for a description of PCRE's native API, which contains much
+additional functionality. There is no POSIX-style wrapper for PCRE's 16-bit
+and 32-bit library.
+

+

+The functions described here are just wrapper functions that ultimately call
+the PCRE native API. Their prototypes are defined in the pcreposix.h
+header file, and on Unix systems the library itself is called
+pcreposix.a, so can be accessed by adding -lpcreposix to the
+command for linking an application that uses them. Because the POSIX functions
+call the native ones, it is also necessary to add -lpcre.
+

+

+I have implemented only those POSIX option bits that can be reasonably mapped
+to PCRE native options. In addition, the option REG_EXTENDED is defined with
+the value zero. This has no effect, but since programs that are written to the
+POSIX interface often use it, this makes it easier to slot in PCRE as a
+replacement library. Other POSIX options are not even defined.
+

+

+There are also some other options that are not defined by POSIX. These have
+been added at the request of users who want to make use of certain
+PCRE-specific features via the POSIX calling interface.
+

+

+When PCRE is called via these functions, it is only the API that is POSIX-like
+in style. The syntax and semantics of the regular expressions themselves are
+still those of Perl, subject to the setting of various PCRE options, as
+described below. "POSIX-like in style" means that the API approximates to the
+POSIX definition; it is not fully POSIX-compatible, and in multi-byte encoding
+domains it is probably even less compatible.
+

+

+The header for these functions is supplied as pcreposix.h to avoid any
+potential clash with other POSIX libraries. It can, of course, be renamed or
+aliased as regex.h, which is the "correct" name. It provides two
+structure types, regex_t for compiled internal forms, and
+regmatch_t for returning captured substrings. It also defines some
+constants whose names start with "REG_"; these are used for setting options and
+identifying error codes.
+

+
COMPILING A PATTERN

+

+The function regcomp() is called to compile a pattern into an
+internal form. The pattern is a C string terminated by a binary zero, and
+is passed in the argument pattern. The preg argument is a pointer
+to a regex_t structure that is used as a base for storing information
+about the compiled regular expression.
+

+

+The argument cflags is either zero, or contains one or more of the bits
+defined by the following macros:
+
+  REG_DOTALL
+

+The PCRE_DOTALL option is set when the regular expression is passed for
+compilation to the native function. Note that REG_DOTALL is not part of the
+POSIX standard.
+
+  REG_ICASE
+

+The PCRE_CASELESS option is set when the regular expression is passed for
+compilation to the native function.
+
+  REG_NEWLINE
+

+The PCRE_MULTILINE option is set when the regular expression is passed for
+compilation to the native function. Note that this does not mimic the
+defined POSIX behaviour for REG_NEWLINE (see the following section).
+
+  REG_NOSUB
+

+The PCRE_NO_AUTO_CAPTURE option is set when the regular expression is passed
+for compilation to the native function. In addition, when a pattern that is
+compiled with this flag is passed to regexec() for matching, the
+nmatch and pmatch arguments are ignored, and no captured strings
+are returned.
+
+  REG_UCP
+

+The PCRE_UCP option is set when the regular expression is passed for
+compilation to the native function. This causes PCRE to use Unicode properties
+when matchine \d, \w, etc., instead of just recognizing ASCII values. Note
+that REG_UTF8 is not part of the POSIX standard.
+
+  REG_UNGREEDY
+

+The PCRE_UNGREEDY option is set when the regular expression is passed for
+compilation to the native function. Note that REG_UNGREEDY is not part of the
+POSIX standard.
+
+  REG_UTF8
+

+The PCRE_UTF8 option is set when the regular expression is passed for
+compilation to the native function. This causes the pattern itself and all data
+strings used for matching it to be treated as UTF-8 strings. Note that REG_UTF8
+is not part of the POSIX standard.
+

+

+In the absence of these flags, no options are passed to the native function.
+This means the the regex is compiled with PCRE default semantics. In
+particular, the way it handles newline characters in the subject string is the
+Perl way, not the POSIX way. Note that setting PCRE_MULTILINE has only
+some of the effects specified for REG_NEWLINE. It does not affect the way
+newlines are matched by . (they are not) or by a negative class such as [^a]
+(they are).
+

+

+The yield of regcomp() is zero on success, and non-zero otherwise. The
+preg structure is filled in on success, and one member of the structure
+is public: re_nsub contains the number of capturing subpatterns in
+the regular expression. Various error codes are defined in the header file.
+

+

+NOTE: If the yield of regcomp() is non-zero, you must not attempt to
+use the contents of the preg structure. If, for example, you pass it to
+regexec(), the result is undefined and your program is likely to crash.
+

+
MATCHING NEWLINE CHARACTERS

+

+This area is not simple, because POSIX and Perl take different views of things.
+It is not possible to get PCRE to obey POSIX semantics, but then PCRE was never
+intended to be a POSIX engine. The following table lists the different
+possibilities for matching newline characters in PCRE:
+
+                          Default   Change with
+
+  . matches newline          no     PCRE_DOTALL
+  newline matches [^a]       yes    not changeable
+  $ matches \n at end        yes    PCRE_DOLLARENDONLY
+  $ matches \n in middle     no     PCRE_MULTILINE
+  ^ matches \n in middle     no     PCRE_MULTILINE
+

+This is the equivalent table for POSIX:
+
+                          Default   Change with
+
+  . matches newline          yes    REG_NEWLINE
+  newline matches [^a]       yes    REG_NEWLINE
+  $ matches \n at end        no     REG_NEWLINE
+  $ matches \n in middle     no     REG_NEWLINE
+  ^ matches \n in middle     no     REG_NEWLINE
+

+PCRE's behaviour is the same as Perl's, except that there is no equivalent for
+PCRE_DOLLAR_ENDONLY in Perl. In both PCRE and Perl, there is no way to stop
+newline from matching [^a].
+

+

+The default POSIX newline handling can be obtained by setting PCRE_DOTALL and
+PCRE_DOLLAR_ENDONLY, but there is no way to make PCRE behave exactly as for the
+REG_NEWLINE action.
+

+
MATCHING A PATTERN

+

+The function regexec() is called to match a compiled pattern preg
+against a given string, which is by default terminated by a zero byte
+(but see REG_STARTEND below), subject to the options in eflags. These can
+be:
+
+  REG_NOTBOL
+

+The PCRE_NOTBOL option is set when calling the underlying PCRE matching
+function.
+
+  REG_NOTEMPTY
+

+The PCRE_NOTEMPTY option is set when calling the underlying PCRE matching
+function. Note that REG_NOTEMPTY is not part of the POSIX standard. However,
+setting this option can give more POSIX-like behaviour in some situations.
+
+  REG_NOTEOL
+

+The PCRE_NOTEOL option is set when calling the underlying PCRE matching
+function.
+
+  REG_STARTEND
+

+The string is considered to start at string + pmatch[0].rm_so and
+to have a terminating NUL located at string + pmatch[0].rm_eo
+(there need not actually be a NUL at that location), regardless of the value of
+nmatch. This is a BSD extension, compatible with but not specified by
+IEEE Standard 1003.2 (POSIX.2), and should be used with caution in software
+intended to be portable to other systems. Note that a non-zero rm_so does
+not imply REG_NOTBOL; REG_STARTEND affects only the location of the string, not
+how it is matched.
+

+

+If the pattern was compiled with the REG_NOSUB flag, no data about any matched
+strings is returned. The nmatch and pmatch arguments of
+regexec() are ignored.
+

+

+If the value of nmatch is zero, or if the value pmatch is NULL,
+no data about any matched strings is returned.
+

+

+Otherwise,the portion of the string that was matched, and also any captured
+substrings, are returned via the pmatch argument, which points to an
+array of nmatch structures of type regmatch_t, containing the
+members rm_so and rm_eo. These contain the offset to the first
+character of each substring and the offset to the first character after the end
+of each substring, respectively. The 0th element of the vector relates to the
+entire portion of string that was matched; subsequent elements relate to
+the capturing subpatterns of the regular expression. Unused entries in the
+array have both structure members set to -1.
+

+

+A successful match yields a zero return; various error codes are defined in the
+header file, of which REG_NOMATCH is the "expected" failure code.
+

+
ERROR MESSAGES

+

+The regerror() function maps a non-zero errorcode from either
+regcomp() or regexec() to a printable message. If preg is not
+NULL, the error should have arisen from the use of that structure. A message
+terminated by a binary zero is placed in errbuf. The length of the
+message, including the zero, is limited to errbuf_size. The yield of the
+function is the size of buffer needed to hold the whole message.
+

+
MEMORY USAGE

+

+Compiling a regular expression causes memory to be allocated and associated
+with the preg structure. The function regfree() frees all such
+memory, after which preg may no longer be used as a compiled expression.
+

+
AUTHOR

+

+Philip Hazel
+

+University Computing Service
+

+Cambridge CB2 3QH, England.
+

+

+
REVISION

+

+Last updated: 09 January 2012
+

+Copyright © 1997-2012 University of Cambridge.
+

+

+Return to the PCRE index page.
+

diff --git a/pcre/doc/html/pcreprecompile.html b/pcre/doc/html/pcreprecompile.html
new file mode 100644
index 00000000000..beb9e242451
--- /dev/null
+++ b/pcre/doc/html/pcreprecompile.html
@@ -0,0 +1,158 @@
+
+
+pcreprecompile specification
+
+
+
pcreprecompile man page

+

+Return to the PCRE index page.
+

+

+This page is part of the PCRE HTML documentation. It was generated automatically
+from the original man page. If there is any nonsense in it, please consult the
+man page, in case the conversion went wrong.
+

+

+
SAVING AND RE-USING PRECOMPILED PCRE PATTERNS

+

+If you are running an application that uses a large number of regular
+expression patterns, it may be useful to store them in a precompiled form
+instead of having to compile them every time the application is run.
+If you are not using any private character tables (see the
+pcre_maketables()
+documentation), this is relatively straightforward. If you are using private
+tables, it is a little bit more complicated. However, if you are using the
+just-in-time optimization feature, it is not possible to save and reload the
+JIT data.
+

+

+If you save compiled patterns to a file, you can copy them to a different host
+and run them there. If the two hosts have different endianness (byte order),
+you should run the pcre[16|32]_pattern_to_host_byte_order() function on the
+new host before trying to match the pattern. The matching functions return
+PCRE_ERROR_BADENDIANNESS if they detect a pattern with the wrong endianness.
+

+

+Compiling regular expressions with one version of PCRE for use with a different
+version is not guaranteed to work and may cause crashes, and saving and
+restoring a compiled pattern loses any JIT optimization data.
+

+
SAVING A COMPILED PATTERN

+

+The value returned by pcre[16|32]_compile() points to a single block of
+memory that holds the compiled pattern and associated data. You can find the
+length of this block in bytes by calling pcre[16|32]_fullinfo() with an
+argument of PCRE_INFO_SIZE. You can then save the data in any appropriate
+manner. Here is sample code for the 8-bit library that compiles a pattern and
+writes it to a file. It assumes that the variable fd refers to a file
+that is open for output:
+
+  int erroroffset, rc, size;
+  char *error;
+  pcre *re;
+
+  re = pcre_compile("my pattern", 0, &error, &erroroffset, NULL);
+  if (re == NULL) { ... handle errors ... }
+  rc = pcre_fullinfo(re, NULL, PCRE_INFO_SIZE, &size);
+  if (rc < 0) { ... handle errors ... }
+  rc = fwrite(re, 1, size, fd);
+  if (rc != size) { ... handle errors ... }
+

+In this example, the bytes that comprise the compiled pattern are copied
+exactly. Note that this is binary data that may contain any of the 256 possible
+byte values. On systems that make a distinction between binary and non-binary
+data, be sure that the file is opened for binary output.
+

+

+If you want to write more than one pattern to a file, you will have to devise a
+way of separating them. For binary data, preceding each pattern with its length
+is probably the most straightforward approach. Another possibility is to write
+out the data in hexadecimal instead of binary, one pattern to a line.
+

+

+Saving compiled patterns in a file is only one possible way of storing them for
+later use. They could equally well be saved in a database, or in the memory of
+some daemon process that passes them via sockets to the processes that want
+them.
+

+

+If the pattern has been studied, it is also possible to save the normal study
+data in a similar way to the compiled pattern itself. However, if the
+PCRE_STUDY_JIT_COMPILE was used, the just-in-time data that is created cannot
+be saved because it is too dependent on the current environment. When studying
+generates additional information, pcre[16|32]_study() returns a pointer to a
+pcre[16|32]_extra data block. Its format is defined in the
+section on matching a pattern
+in the
+pcreapi
+documentation. The study_data field points to the binary study data, and
+this is what you must save (not the pcre[16|32]_extra block itself). The
+length of the study data can be obtained by calling pcre[16|32]_fullinfo()
+with an argument of PCRE_INFO_STUDYSIZE. Remember to check that
+pcre[16|32]_study() did return a non-NULL value before trying to save the
+study data.
+

+
RE-USING A PRECOMPILED PATTERN

+

+Re-using a precompiled pattern is straightforward. Having reloaded it into main
+memory, called pcre[16|32]_pattern_to_host_byte_order() if necessary,
+you pass its pointer to pcre[16|32]_exec() or pcre[16|32]_dfa_exec() in
+the usual way.
+

+

+However, if you passed a pointer to custom character tables when the pattern
+was compiled (the tableptr argument of pcre[16|32]_compile()), you
+must now pass a similar pointer to pcre[16|32]_exec() or
+pcre[16|32]_dfa_exec(), because the value saved with the compiled pattern
+will obviously be nonsense. A field in a pcre[16|32]_extra() block is used
+to pass this data, as described in the
+section on matching a pattern
+in the
+pcreapi
+documentation.
+

+

+If you did not provide custom character tables when the pattern was compiled,
+the pointer in the compiled pattern is NULL, which causes the matching
+functions to use PCRE's internal tables. Thus, you do not need to take any
+special action at run time in this case.
+

+

+If you saved study data with the compiled pattern, you need to create your own
+pcre[16|32]_extra data block and set the study_data field to point to the
+reloaded study data. You must also set the PCRE_EXTRA_STUDY_DATA bit in the
+flags field to indicate that study data is present. Then pass the
+pcre[16|32]_extra block to the matching function in the usual way. If the
+pattern was studied for just-in-time optimization, that data cannot be saved,
+and so is lost by a save/restore cycle.
+

+
COMPATIBILITY WITH DIFFERENT PCRE RELEASES

+

+In general, it is safest to recompile all saved patterns when you update to a
+new PCRE release, though not all updates actually require this.
+

+
AUTHOR

+

+Philip Hazel
+

+University Computing Service
+

+Cambridge CB2 3QH, England.
+

+

+
REVISION

+

+Last updated: 24 June 2012
+

+Copyright © 1997-2012 University of Cambridge.
+

+

+Return to the PCRE index page.
+

diff --git a/pcre/doc/html/pcresample.html b/pcre/doc/html/pcresample.html
new file mode 100644
index 00000000000..aca9184e00e
--- /dev/null
+++ b/pcre/doc/html/pcresample.html
@@ -0,0 +1,110 @@
+
+
+pcresample specification
+
+
+
pcresample man page

+

+Return to the PCRE index page.
+

+

+This page is part of the PCRE HTML documentation. It was generated automatically
+from the original man page. If there is any nonsense in it, please consult the
+man page, in case the conversion went wrong.
+

+

+PCRE SAMPLE PROGRAM
+

+

+A simple, complete demonstration program, to get you started with using PCRE,
+is supplied in the file pcredemo.c in the PCRE distribution. A listing of
+this program is given in the
+pcredemo
+documentation. If you do not have a copy of the PCRE distribution, you can save
+this listing to re-create pcredemo.c.
+

+

+The demonstration program, which uses the original PCRE 8-bit library, compiles
+the regular expression that is its first argument, and matches it against the
+subject string in its second argument. No PCRE options are set, and default
+character tables are used. If matching succeeds, the program outputs the
+portion of the subject that matched, together with the contents of any captured
+substrings.
+

+

+If the -g option is given on the command line, the program then goes on to
+check for further matches of the same regular expression in the same subject
+string. The logic is a little bit tricky because of the possibility of matching
+an empty string. Comments in the code explain what is going on.
+

+

+If PCRE is installed in the standard include and library directories for your
+operating system, you should be able to compile the demonstration program using
+this command:
+
+  gcc -o pcredemo pcredemo.c -lpcre
+

+If PCRE is installed elsewhere, you may need to add additional options to the
+command line. For example, on a Unix-like system that has PCRE installed in
+/usr/local, you can compile the demonstration program using a command
+like this:
+
+  gcc -o pcredemo -I/usr/local/include pcredemo.c -L/usr/local/lib -lpcre
+

+In a Windows environment, if you want to statically link the program against a
+non-dll pcre.a file, you must uncomment the line that defines PCRE_STATIC
+before including pcre.h, because otherwise the pcre_malloc() and
+pcre_free() exported functions will be declared
+__declspec(dllimport), with unwanted results.
+

+

+Once you have compiled and linked the demonstration program, you can run simple
+tests like this:
+
+  ./pcredemo 'cat|dog' 'the cat sat on the mat'
+  ./pcredemo -g 'cat|dog' 'the dog sat on the cat'
+

+Note that there is a much more comprehensive test program, called
+pcretest,
+which supports many more facilities for testing regular expressions and both
+PCRE libraries. The
+pcredemo
+program is provided as a simple coding example.
+

+

+If you try to run
+pcredemo
+when PCRE is not installed in the standard library directory, you may get an
+error like this on some operating systems (e.g. Solaris):
+
+  ld.so.1: a.out: fatal: libpcre.so.0: open failed: No such file or directory
+

+This is caused by the way shared library support works on those systems. You
+need to add
+
+  -R/usr/local/lib
+

+(for example) to the compile command to get round this problem.
+

+

+AUTHOR
+

+

+Philip Hazel
+

+University Computing Service
+

+Cambridge CB2 3QH, England.
+

+

+

+REVISION
+

+

+Last updated: 10 January 2012
+

+Copyright © 1997-2012 University of Cambridge.
+

+

+Return to the PCRE index page.
+

diff --git a/pcre/doc/html/pcrestack.html b/pcre/doc/html/pcrestack.html
new file mode 100644
index 00000000000..af6406d0708
--- /dev/null
+++ b/pcre/doc/html/pcrestack.html
@@ -0,0 +1,225 @@
+
+
+pcrestack specification
+
+
+
pcrestack man page

+

+Return to the PCRE index page.
+

+

+This page is part of the PCRE HTML documentation. It was generated automatically
+from the original man page. If there is any nonsense in it, please consult the
+man page, in case the conversion went wrong.
+

+

+PCRE DISCUSSION OF STACK USAGE
+

+

+When you call pcre[16|32]_exec(), it makes use of an internal function
+called match(). This calls itself recursively at branch points in the
+pattern, in order to remember the state of the match so that it can back up and
+try a different alternative if the first one fails. As matching proceeds deeper
+and deeper into the tree of possibilities, the recursion depth increases. The
+match() function is also called in other circumstances, for example,
+whenever a parenthesized sub-pattern is entered, and in certain cases of
+repetition.
+

+

+Not all calls of match() increase the recursion depth; for an item such
+as a* it may be called several times at the same level, after matching
+different numbers of a's. Furthermore, in a number of cases where the result of
+the recursive call would immediately be passed back as the result of the
+current call (a "tail recursion"), the function is just restarted instead.
+

+

+The above comments apply when pcre[16|32]_exec() is run in its normal
+interpretive manner. If the pattern was studied with the
+PCRE_STUDY_JIT_COMPILE option, and just-in-time compiling was successful, and
+the options passed to pcre[16|32]_exec() were not incompatible, the matching
+process uses the JIT-compiled code instead of the match() function. In
+this case, the memory requirements are handled entirely differently. See the
+pcrejit
+documentation for details.
+

+

+The pcre[16|32]_dfa_exec() function operates in an entirely different way,
+and uses recursion only when there is a regular expression recursion or
+subroutine call in the pattern. This includes the processing of assertion and
+"once-only" subpatterns, which are handled like subroutine calls. Normally,
+these are never very deep, and the limit on the complexity of
+pcre[16|32]_dfa_exec() is controlled by the amount of workspace it is given.
+However, it is possible to write patterns with runaway infinite recursions;
+such patterns will cause pcre[16|32]_dfa_exec() to run out of stack. At
+present, there is no protection against this.
+

+

+The comments that follow do NOT apply to pcre[16|32]_dfa_exec(); they are
+relevant only for pcre[16|32]_exec() without the JIT optimization.
+

+

+Reducing pcre[16|32]_exec()'s stack usage
+

+

+Each time that match() is actually called recursively, it uses memory
+from the process stack. For certain kinds of pattern and data, very large
+amounts of stack may be needed, despite the recognition of "tail recursion".
+You can often reduce the amount of recursion, and therefore the amount of stack
+used, by modifying the pattern that is being matched. Consider, for example,
+this pattern:
+
+  ([^<]|<(?!inet))+
+

+It matches from wherever it starts until it encounters "<inet" or the end of
+the data, and is the kind of pattern that might be used when processing an XML
+file. Each iteration of the outer parentheses matches either one character that
+is not "<" or a "<" that is not followed by "inet". However, each time a
+parenthesis is processed, a recursion occurs, so this formulation uses a stack
+frame for each matched character. For a long string, a lot of stack is
+required. Consider now this rewritten pattern, which matches exactly the same
+strings:
+
+  ([^<]++|<(?!inet))+
+

+This uses very much less stack, because runs of characters that do not contain
+"<" are "swallowed" in one item inside the parentheses. Recursion happens only
+when a "<" character that is not followed by "inet" is encountered (and we
+assume this is relatively rare). A possessive quantifier is used to stop any
+backtracking into the runs of non-"<" characters, but that is not related to
+stack usage.
+

+

+This example shows that one way of avoiding stack problems when matching long
+subject strings is to write repeated parenthesized subpatterns to match more
+than one character whenever possible.
+

+

+Compiling PCRE to use heap instead of stack for pcre[16|32]_exec()
+

+

+In environments where stack memory is constrained, you might want to compile
+PCRE to use heap memory instead of stack for remembering back-up points when
+pcre[16|32]_exec() is running. This makes it run a lot more slowly, however.
+Details of how to do this are given in the
+pcrebuild
+documentation. When built in this way, instead of using the stack, PCRE obtains
+and frees memory by calling the functions that are pointed to by the
+pcre[16|32]_stack_malloc and pcre[16|32]_stack_free variables. By
+default, these point to malloc() and free(), but you can replace
+the pointers to cause PCRE to use your own functions. Since the block sizes are
+always the same, and are always freed in reverse order, it may be possible to
+implement customized memory handlers that are more efficient than the standard
+functions.
+

+

+Limiting pcre[16|32]_exec()'s stack usage
+

+

+You can set limits on the number of times that match() is called, both in
+total and recursively. If a limit is exceeded, pcre[16|32]_exec() returns an
+error code. Setting suitable limits should prevent it from running out of
+stack. The default values of the limits are very large, and unlikely ever to
+operate. They can be changed when PCRE is built, and they can also be set when
+pcre[16|32]_exec() is called. For details of these interfaces, see the
+pcrebuild
+documentation and the
+section on extra data for pcre[16|32]_exec()
+in the
+pcreapi
+documentation.
+

+

+As a very rough rule of thumb, you should reckon on about 500 bytes per
+recursion. Thus, if you want to limit your stack usage to 8Mb, you should set
+the limit at 16000 recursions. A 64Mb stack, on the other hand, can support
+around 128000 recursions.
+

+

+In Unix-like environments, the pcretest test program has a command line
+option (-S) that can be used to increase the size of its stack. As long
+as the stack is large enough, another option (-M) can be used to find the
+smallest limits that allow a particular pattern to match a given subject
+string. This is done by calling pcre[16|32]_exec() repeatedly with different
+limits.
+

+

+Obtaining an estimate of stack usage
+

+

+The actual amount of stack used per recursion can vary quite a lot, depending
+on the compiler that was used to build PCRE and the optimization or debugging
+options that were set for it. The rule of thumb value of 500 bytes mentioned
+above may be larger or smaller than what is actually needed. A better
+approximation can be obtained by running this command:
+
+  pcretest -m -C
+

+The -C option causes pcretest to output information about the
+options with which PCRE was compiled. When -m is also given (before
+-C), information about stack use is given in a line like this:
+
+  Match recursion uses stack: approximate frame size = 640 bytes
+

+The value is approximate because some recursions need a bit more (up to perhaps
+16 more bytes).
+

+

+If the above command is given when PCRE is compiled to use the heap instead of
+the stack for recursion, the value that is output is the size of each block
+that is obtained from the heap.
+

+

+Changing stack size in Unix-like systems
+

+

+In Unix-like environments, there is not often a problem with the stack unless
+very long strings are involved, though the default limit on stack size varies
+from system to system. Values from 8Mb to 64Mb are common. You can find your
+default limit by running the command:
+
+  ulimit -s
+

+Unfortunately, the effect of running out of stack is often SIGSEGV, though
+sometimes a more explicit error message is given. You can normally increase the
+limit on stack size by code such as this:
+
+  struct rlimit rlim;
+  getrlimit(RLIMIT_STACK, &rlim);
+  rlim.rlim_cur = 100*1024*1024;
+  setrlimit(RLIMIT_STACK, &rlim);
+

+This reads the current limits (soft and hard) using getrlimit(), then
+attempts to increase the soft limit to 100Mb using setrlimit(). You must
+do this before calling pcre[16|32]_exec().
+

+

+Changing stack size in Mac OS X
+

+

+Using setrlimit(), as described above, should also work on Mac OS X. It
+is also possible to set a stack size when linking a program. There is a
+discussion about stack sizes in Mac OS X at this web site:
+http://developer.apple.com/qa/qa2005/qa1419.html.
+

+

+AUTHOR
+

+

+Philip Hazel
+

+University Computing Service
+

+Cambridge CB2 3QH, England.
+

+

+

+REVISION
+

+

+Last updated: 24 June 2012
+

+Copyright © 1997-2012 University of Cambridge.
+

+

+Return to the PCRE index page.
+

diff --git a/pcre/doc/html/pcresyntax.html b/pcre/doc/html/pcresyntax.html
new file mode 100644
index 00000000000..b32e8b1689f
--- /dev/null
+++ b/pcre/doc/html/pcresyntax.html
@@ -0,0 +1,525 @@
+
+
+pcresyntax specification
+
+
+
pcresyntax man page

+

+Return to the PCRE index page.
+

+

+This page is part of the PCRE HTML documentation. It was generated automatically
+from the original man page. If there is any nonsense in it, please consult the
+man page, in case the conversion went wrong.
+

+

+
PCRE REGULAR EXPRESSION SYNTAX SUMMARY

+

+The full syntax and semantics of the regular expressions that are supported by
+PCRE are described in the
+pcrepattern
+documentation. This document contains a quick-reference summary of the syntax.
+

+
QUOTING

+

+
+  \x         where x is non-alphanumeric is a literal x
+  \Q...\E    treat enclosed characters as literal
+

+

+
CHARACTERS

+

+
+  \a         alarm, that is, the BEL character (hex 07)
+  \cx        "control-x", where x is any ASCII character
+  \e         escape (hex 1B)
+  \f         form feed (hex 0C)
+  \n         newline (hex 0A)
+  \r         carriage return (hex 0D)
+  \t         tab (hex 09)
+  \ddd       character with octal code ddd, or backreference
+  \xhh       character with hex code hh
+  \x{hhh..}  character with hex code hhh..
+

+

+
CHARACTER TYPES

+

+
+  .          any character except newline;
+               in dotall mode, any character whatsoever
+  \C         one data unit, even in UTF mode (best avoided)
+  \d         a decimal digit
+  \D         a character that is not a decimal digit
+  \h         a horizontal white space character
+  \H         a character that is not a horizontal white space character
+  \N         a character that is not a newline
+  \p{xx}     a character with the xx property
+  \P{xx}     a character without the xx property
+  \R         a newline sequence
+  \s         a white space character
+  \S         a character that is not a white space character
+  \v         a vertical white space character
+  \V         a character that is not a vertical white space character
+  \w         a "word" character
+  \W         a "non-word" character
+  \X         a Unicode extended grapheme cluster
+

+In PCRE, by default, \d, \D, \s, \S, \w, and \W recognize only ASCII
+characters, even in a UTF mode. However, this can be changed by setting the
+PCRE_UCP option.
+

+
GENERAL CATEGORY PROPERTIES FOR \p and \P

+

+
+  C          Other
+  Cc         Control
+  Cf         Format
+  Cn         Unassigned
+  Co         Private use
+  Cs         Surrogate
+
+  L          Letter
+  Ll         Lower case letter
+  Lm         Modifier letter
+  Lo         Other letter
+  Lt         Title case letter
+  Lu         Upper case letter
+  L&         Ll, Lu, or Lt
+
+  M          Mark
+  Mc         Spacing mark
+  Me         Enclosing mark
+  Mn         Non-spacing mark
+
+  N          Number
+  Nd         Decimal number
+  Nl         Letter number
+  No         Other number
+
+  P          Punctuation
+  Pc         Connector punctuation
+  Pd         Dash punctuation
+  Pe         Close punctuation
+  Pf         Final punctuation
+  Pi         Initial punctuation
+  Po         Other punctuation
+  Ps         Open punctuation
+
+  S          Symbol
+  Sc         Currency symbol
+  Sk         Modifier symbol
+  Sm         Mathematical symbol
+  So         Other symbol
+
+  Z          Separator
+  Zl         Line separator
+  Zp         Paragraph separator
+  Zs         Space separator
+

+

+
PCRE SPECIAL CATEGORY PROPERTIES FOR \p and \P

+

+
+  Xan        Alphanumeric: union of properties L and N
+  Xps        POSIX space: property Z or tab, NL, VT, FF, CR
+  Xsp        Perl space: property Z or tab, NL, FF, CR
+  Xuc        Univerally-named character: one that can be
+               represented by a Universal Character Name
+  Xwd        Perl word: property Xan or underscore
+

+

+
SCRIPT NAMES FOR \p AND \P

+

+Arabic,
+Armenian,
+Avestan,
+Balinese,
+Bamum,
+Batak,
+Bengali,
+Bopomofo,
+Brahmi,
+Braille,
+Buginese,
+Buhid,
+Canadian_Aboriginal,
+Carian,
+Chakma,
+Cham,
+Cherokee,
+Common,
+Coptic,
+Cuneiform,
+Cypriot,
+Cyrillic,
+Deseret,
+Devanagari,
+Egyptian_Hieroglyphs,
+Ethiopic,
+Georgian,
+Glagolitic,
+Gothic,
+Greek,
+Gujarati,
+Gurmukhi,
+Han,
+Hangul,
+Hanunoo,
+Hebrew,
+Hiragana,
+Imperial_Aramaic,
+Inherited,
+Inscriptional_Pahlavi,
+Inscriptional_Parthian,
+Javanese,
+Kaithi,
+Kannada,
+Katakana,
+Kayah_Li,
+Kharoshthi,
+Khmer,
+Lao,
+Latin,
+Lepcha,
+Limbu,
+Linear_B,
+Lisu,
+Lycian,
+Lydian,
+Malayalam,
+Mandaic,
+Meetei_Mayek,
+Meroitic_Cursive,
+Meroitic_Hieroglyphs,
+Miao,
+Mongolian,
+Myanmar,
+New_Tai_Lue,
+Nko,
+Ogham,
+Old_Italic,
+Old_Persian,
+Old_South_Arabian,
+Old_Turkic,
+Ol_Chiki,
+Oriya,
+Osmanya,
+Phags_Pa,
+Phoenician,
+Rejang,
+Runic,
+Samaritan,
+Saurashtra,
+Sharada,
+Shavian,
+Sinhala,
+Sora_Sompeng,
+Sundanese,
+Syloti_Nagri,
+Syriac,
+Tagalog,
+Tagbanwa,
+Tai_Le,
+Tai_Tham,
+Tai_Viet,
+Takri,
+Tamil,
+Telugu,
+Thaana,
+Thai,
+Tibetan,
+Tifinagh,
+Ugaritic,
+Vai,
+Yi.
+

+
CHARACTER CLASSES

+

+
+  [...]       positive character class
+  [^...]      negative character class
+  [x-y]       range (can be used for hex characters)
+  [[:xxx:]]   positive POSIX named set
+  [[:^xxx:]]  negative POSIX named set
+
+  alnum       alphanumeric
+  alpha       alphabetic
+  ascii       0-127
+  blank       space or tab
+  cntrl       control character
+  digit       decimal digit
+  graph       printing, excluding space
+  lower       lower case letter
+  print       printing, including space
+  punct       printing, excluding alphanumeric
+  space       white space
+  upper       upper case letter
+  word        same as \w
+  xdigit      hexadecimal digit
+

+In PCRE, POSIX character set names recognize only ASCII characters by default,
+but some of them use Unicode properties if PCRE_UCP is set. You can use
+\Q...\E inside a character class.
+

+
QUANTIFIERS

+

+
+  ?           0 or 1, greedy
+  ?+          0 or 1, possessive
+  ??          0 or 1, lazy
+  *           0 or more, greedy
+  *+          0 or more, possessive
+  *?          0 or more, lazy
+  +           1 or more, greedy
+  ++          1 or more, possessive
+  +?          1 or more, lazy
+  {n}         exactly n
+  {n,m}       at least n, no more than m, greedy
+  {n,m}+      at least n, no more than m, possessive
+  {n,m}?      at least n, no more than m, lazy
+  {n,}        n or more, greedy
+  {n,}+       n or more, possessive
+  {n,}?       n or more, lazy
+

+

+
ANCHORS AND SIMPLE ASSERTIONS

+

+
+  \b          word boundary
+  \B          not a word boundary
+  ^           start of subject
+               also after internal newline in multiline mode
+  \A          start of subject
+  $           end of subject
+               also before newline at end of subject
+               also before internal newline in multiline mode
+  \Z          end of subject
+               also before newline at end of subject
+  \z          end of subject
+  \G          first matching position in subject
+

+

+
MATCH POINT RESET

+

+
+  \K          reset start of match
+

+

+
ALTERNATION

+

+
+  expr|expr|expr...
+

+

+
CAPTURING

+

+
+  (...)           capturing group
+  (?<name>...)    named capturing group (Perl)
+  (?'name'...)    named capturing group (Perl)
+  (?P<name>...)   named capturing group (Python)
+  (?:...)         non-capturing group
+  (?|...)         non-capturing group; reset group numbers for
+                   capturing groups in each alternative
+

+

+
ATOMIC GROUPS

+

+
+  (?>...)         atomic, non-capturing group
+

+

+
COMMENT

+

+
+  (?#....)        comment (not nestable)
+

+

+
OPTION SETTING

+

+
+  (?i)            caseless
+  (?J)            allow duplicate names
+  (?m)            multiline
+  (?s)            single line (dotall)
+  (?U)            default ungreedy (lazy)
+  (?x)            extended (ignore white space)
+  (?-...)         unset option(s)
+

+The following are recognized only at the start of a pattern or after one of the
+newline-setting options with similar syntax:
+
+  (*LIMIT_MATCH=d) set the match limit to d (decimal number)
+  (*LIMIT_RECURSION=d) set the recursion limit to d (decimal number)
+  (*NO_START_OPT) no start-match optimization (PCRE_NO_START_OPTIMIZE)
+  (*UTF8)         set UTF-8 mode: 8-bit library (PCRE_UTF8)
+  (*UTF16)        set UTF-16 mode: 16-bit library (PCRE_UTF16)
+  (*UTF32)        set UTF-32 mode: 32-bit library (PCRE_UTF32)
+  (*UTF)          set appropriate UTF mode for the library in use
+  (*UCP)          set PCRE_UCP (use Unicode properties for \d etc)
+

+

+
LOOKAHEAD AND LOOKBEHIND ASSERTIONS

+

+
+  (?=...)         positive look ahead
+  (?!...)         negative look ahead
+  (?<=...)        positive look behind
+  (?<!...)        negative look behind
+

+Each top-level branch of a look behind must be of a fixed length.
+

+
BACKREFERENCES

+

+
+  \n              reference by number (can be ambiguous)
+  \gn             reference by number
+  \g{n}           reference by number
+  \g{-n}          relative reference by number
+  \k<name>        reference by name (Perl)
+  \k'name'        reference by name (Perl)
+  \g{name}        reference by name (Perl)
+  \k{name}        reference by name (.NET)
+  (?P=name)       reference by name (Python)
+

+

+
SUBROUTINE REFERENCES (POSSIBLY RECURSIVE)

+

+
+  (?R)            recurse whole pattern
+  (?n)            call subpattern by absolute number
+  (?+n)           call subpattern by relative number
+  (?-n)           call subpattern by relative number
+  (?&name)        call subpattern by name (Perl)
+  (?P>name)       call subpattern by name (Python)
+  \g<name>        call subpattern by name (Oniguruma)
+  \g'name'        call subpattern by name (Oniguruma)
+  \g<n>           call subpattern by absolute number (Oniguruma)
+  \g'n'           call subpattern by absolute number (Oniguruma)
+  \g<+n>          call subpattern by relative number (PCRE extension)
+  \g'+n'          call subpattern by relative number (PCRE extension)
+  \g<-n>          call subpattern by relative number (PCRE extension)
+  \g'-n'          call subpattern by relative number (PCRE extension)
+

+

+
CONDITIONAL PATTERNS

+

+
+  (?(condition)yes-pattern)
+  (?(condition)yes-pattern|no-pattern)
+
+  (?(n)...        absolute reference condition
+  (?(+n)...       relative reference condition
+  (?(-n)...       relative reference condition
+  (?(<name>)...   named reference condition (Perl)
+  (?('name')...   named reference condition (Perl)
+  (?(name)...     named reference condition (PCRE)
+  (?(R)...        overall recursion condition
+  (?(Rn)...       specific group recursion condition
+  (?(R&name)...   specific recursion condition
+  (?(DEFINE)...   define subpattern for reference
+  (?(assert)...   assertion condition
+

+

+
BACKTRACKING CONTROL

+

+The following act immediately they are reached:
+
+  (*ACCEPT)       force successful match
+  (*FAIL)         force backtrack; synonym (*F)
+  (*MARK:NAME)    set name to be passed back; synonym (*:NAME)
+

+The following act only when a subsequent match failure causes a backtrack to
+reach them. They all force a match failure, but they differ in what happens
+afterwards. Those that advance the start-of-match point do so only if the
+pattern is not anchored.
+
+  (*COMMIT)       overall failure, no advance of starting point
+  (*PRUNE)        advance to next starting character
+  (*PRUNE:NAME)   equivalent to (*MARK:NAME)(*PRUNE)
+  (*SKIP)         advance to current matching position
+  (*SKIP:NAME)    advance to position corresponding to an earlier
+                  (*MARK:NAME); if not found, the (*SKIP) is ignored
+  (*THEN)         local failure, backtrack to next alternation
+  (*THEN:NAME)    equivalent to (*MARK:NAME)(*THEN)
+

+

+
NEWLINE CONVENTIONS

+

+These are recognized only at the very start of the pattern or after a
+(*BSR_...), (*UTF8), (*UTF16), (*UTF32) or (*UCP) option.
+
+  (*CR)           carriage return only
+  (*LF)           linefeed only
+  (*CRLF)         carriage return followed by linefeed
+  (*ANYCRLF)      all three of the above
+  (*ANY)          any Unicode newline sequence
+

+

+
WHAT \R MATCHES

+

+These are recognized only at the very start of the pattern or after a
+(*...) option that sets the newline convention or a UTF or UCP mode.
+
+  (*BSR_ANYCRLF)  CR, LF, or CRLF
+  (*BSR_UNICODE)  any Unicode newline sequence
+

+

+
CALLOUTS

+

+
+  (?C)      callout
+  (?Cn)     callout with data n
+

+

+
SEE ALSO

+

+pcrepattern(3), pcreapi(3), pcrecallout(3),
+pcrematching(3), pcre(3).
+

+
AUTHOR

+

+Philip Hazel
+

+University Computing Service
+

+Cambridge CB2 3QH, England.
+

+

+
REVISION

+

+Last updated: 26 April 2013
+

+Copyright © 1997-2013 University of Cambridge.
+

+

+Return to the PCRE index page.
+

diff --git a/pcre/doc/html/pcretest.html b/pcre/doc/html/pcretest.html
new file mode 100644
index 00000000000..17d1ade6246
--- /dev/null
+++ b/pcre/doc/html/pcretest.html
@@ -0,0 +1,1100 @@
+
+
+pcretest specification
+
+
+
pcretest man page

+

+Return to the PCRE index page.
+

+

+This page is part of the PCRE HTML documentation. It was generated automatically
+from the original man page. If there is any nonsense in it, please consult the
+man page, in case the conversion went wrong.
+

+

+
SYNOPSIS

+

+pcretest [options] [input file [output file]]
+

+

+pcretest was written as a test program for the PCRE regular expression
+library itself, but it can also be used for experimenting with regular
+expressions. This document describes the features of the test program; for
+details of the regular expressions themselves, see the
+pcrepattern
+documentation. For details of the PCRE library function calls and their
+options, see the
+pcreapi
+,
+pcre16
+and
+pcre32
+documentation.
+

+

+The input for pcretest is a sequence of regular expression patterns and
+strings to be matched, as described below. The output shows the result of each
+match. Options on the command line and the patterns control PCRE options and
+exactly what is output.
+

+

+As PCRE has evolved, it has acquired many different features, and as a result,
+pcretest now has rather a lot of obscure options for testing every
+possible feature. Some of these options are specifically designed for use in
+conjunction with the test script and data files that are distributed as part of
+PCRE, and are unlikely to be of use otherwise. They are all documented here,
+but without much justification.
+

+
INPUT DATA FORMAT

+

+Input to pcretest is processed line by line, either by calling the C
+library's fgets() function, or via the libreadline library (see
+below). In Unix-like environments, fgets() treats any bytes other than
+newline as data characters. However, in some Windows environments character 26
+(hex 1A) causes an immediate end of file, and no further data is read. For
+maximum portability, therefore, it is safest to use only ASCII characters in
+pcretest input files.
+

+
PCRE's 8-BIT, 16-BIT AND 32-BIT LIBRARIES

+

+From release 8.30, two separate PCRE libraries can be built. The original one
+supports 8-bit character strings, whereas the newer 16-bit library supports
+character strings encoded in 16-bit units. From release 8.32, a third library
+can be built, supporting character strings encoded in 32-bit units. The
+pcretest program can be used to test all three libraries. However, it is
+itself still an 8-bit program, reading 8-bit input and writing 8-bit output.
+When testing the 16-bit or 32-bit library, the patterns and data strings are
+converted to 16- or 32-bit format before being passed to the PCRE library
+functions. Results are converted to 8-bit for output.
+

+

+References to functions and structures of the form pcre[16|32]_xx below
+mean "pcre_xx when using the 8-bit library, pcre16_xx when using
+the 16-bit library, or pcre32_xx when using the 32-bit library".
+

+
COMMAND LINE OPTIONS

+

+-8
+If both the 8-bit library has been built, this option causes the 8-bit library
+to be used (which is the default); if the 8-bit library has not been built,
+this option causes an error.
+

+

+-16
+If both the 8-bit or the 32-bit, and the 16-bit libraries have been built, this
+option causes the 16-bit library to be used. If only the 16-bit library has been
+built, this is the default (so has no effect). If only the 8-bit or the 32-bit
+library has been built, this option causes an error.
+

+

+-32
+If both the 8-bit or the 16-bit, and the 32-bit libraries have been built, this
+option causes the 32-bit library to be used. If only the 32-bit library has been
+built, this is the default (so has no effect). If only the 8-bit or the 16-bit
+library has been built, this option causes an error.
+

+

+-b
+Behave as if each pattern has the /B (show byte code) modifier; the
+internal form is output after compilation.
+

+

+-C
+Output the version number of the PCRE library, and all available information
+about the optional features that are included, and then exit with zero exit
+code. All other options are ignored.
+

+

+-C option
+Output information about a specific build-time option, then exit. This
+functionality is intended for use in scripts such as RunTest. The
+following options output the value and set the exit code as indicated:
+
+  ebcdic-nl  the code for LF (= NL) in an EBCDIC environment:
+               0x15 or 0x25
+               0 if used in an ASCII environment
+               exit code is always 0
+  linksize   the configured internal link size (2, 3, or 4)
+               exit code is set to the link size
+  newline    the default newline setting:
+               CR, LF, CRLF, ANYCRLF, or ANY
+               exit code is always 0
+

+The following options output 1 for true or 0 for false, and set the exit code
+to the same value:
+
+  ebcdic     compiled for an EBCDIC environment
+  jit        just-in-time support is available
+  pcre16     the 16-bit library was built
+  pcre32     the 32-bit library was built
+  pcre8      the 8-bit library was built
+  ucp        Unicode property support is available
+  utf        UTF-8 and/or UTF-16 and/or UTF-32 support
+               is available
+

+If an unknown option is given, an error message is output; the exit code is 0.
+

+

+-d
+Behave as if each pattern has the /D (debug) modifier; the internal
+form and information about the compiled pattern is output after compilation;
+-d is equivalent to -b -i.
+

+

+-dfa
+Behave as if each data line contains the \D escape sequence; this causes the
+alternative matching function, pcre[16|32]_dfa_exec(), to be used instead
+of the standard pcre[16|32]_exec() function (more detail is given below).
+

+

+-help
+Output a brief summary these options and then exit.
+

+

+-i
+Behave as if each pattern has the /I modifier; information about the
+compiled pattern is given after compilation.
+

+

+-M
+Behave as if each data line contains the \M escape sequence; this causes
+PCRE to discover the minimum MATCH_LIMIT and MATCH_LIMIT_RECURSION settings by
+calling pcre[16|32]_exec() repeatedly with different limits.
+

+

+-m
+Output the size of each compiled pattern after it has been compiled. This is
+equivalent to adding /M to each regular expression. The size is given in
+bytes for both libraries.
+

+

+-o osize
+Set the number of elements in the output vector that is used when calling
+pcre[16|32]_exec() or pcre[16|32]_dfa_exec() to be osize. The
+default value is 45, which is enough for 14 capturing subexpressions for
+pcre[16|32]_exec() or 22 different matches for
+pcre[16|32]_dfa_exec().
+The vector size can be changed for individual matching calls by including \O
+in the data line (see below).
+

+

+-p
+Behave as if each pattern has the /P modifier; the POSIX wrapper API is
+used to call PCRE. None of the other options has any effect when -p is
+set. This option can be used only with the 8-bit library.
+

+

+-q
+Do not output the version number of pcretest at the start of execution.
+

+

+-S size
+On Unix-like systems, set the size of the run-time stack to size
+megabytes.
+

+

+-s or -s+
+Behave as if each pattern has the /S modifier; in other words, force each
+pattern to be studied. If -s+ is used, all the JIT compile options are
+passed to pcre[16|32]_study(), causing just-in-time optimization to be set
+up if it is available, for both full and partial matching. Specific JIT compile
+options can be selected by following -s+ with a digit in the range 1 to
+7, which selects the JIT compile modes as follows:
+
+  1  normal match only
+  2  soft partial match only
+  3  normal match and soft partial match
+  4  hard partial match only
+  6  soft and hard partial match
+  7  all three modes (default)
+

+If -s++ is used instead of -s+ (with or without a following digit),
+the text "(JIT)" is added to the first output line after a match or no match
+when JIT-compiled code was actually used.
+

+

+Note that there are pattern options that can override -s, either
+specifying no studying at all, or suppressing JIT compilation.
+

+

+If the /I or /D option is present on a pattern (requesting output
+about the compiled pattern), information about the result of studying is not
+included when studying is caused only by -s and neither -i nor
+-d is present on the command line. This behaviour means that the output
+from tests that are run with and without -s should be identical, except
+when options that output information about the actual running of a match are
+set.
+

+

+The -M, -t, and -tm options, which give information about
+resources used, are likely to produce different output with and without
+-s. Output may also differ if the /C option is present on an
+individual pattern. This uses callouts to trace the the matching process, and
+this may be different between studied and non-studied patterns. If the pattern
+contains (*MARK) items there may also be differences, for the same reason. The
+-s command line option can be overridden for specific patterns that
+should never be studied (see the /S pattern modifier below).
+

+

+-t
+Run each compile, study, and match many times with a timer, and output
+resulting time per compile or match (in milliseconds). Do not set -m with
+-t, because you will then get the size output a zillion times, and the
+timing will be distorted. You can control the number of iterations that are
+used for timing by following -t with a number (as a separate item on the
+command line). For example, "-t 1000" would iterate 1000 times. The default is
+to iterate 500000 times.
+

+

+-tm
+This is like -t except that it times only the matching phase, not the
+compile or study phases.
+

+
DESCRIPTION

+

+If pcretest is given two filename arguments, it reads from the first and
+writes to the second. If it is given only one filename argument, it reads from
+that file and writes to stdout. Otherwise, it reads from stdin and writes to
+stdout, and prompts for each line of input, using "re>" to prompt for regular
+expressions, and "data>" to prompt for data lines.
+

+

+When pcretest is built, a configuration option can specify that it should
+be linked with the libreadline library. When this is done, if the input
+is from a terminal, it is read using the readline() function. This
+provides line-editing and history facilities. The output from the -help
+option states whether or not readline() will be used.
+

+

+The program handles any number of sets of input on a single input file. Each
+set starts with a regular expression, and continues with any number of data
+lines to be matched against the pattern.
+

+

+Each data line is matched separately and independently. If you want to do
+multi-line matches, you have to use the \n escape sequence (or \r or \r\n,
+etc., depending on the newline setting) in a single line of input to encode the
+newline sequences. There is no limit on the length of data lines; the input
+buffer is automatically extended if it is too small.
+

+

+An empty line signals the end of the data lines, at which point a new regular
+expression is read. The regular expressions are given enclosed in any
+non-alphanumeric delimiters other than backslash, for example:
+
+  /(a|bc)x+yz/
+

+White space before the initial delimiter is ignored. A regular expression may
+be continued over several input lines, in which case the newline characters are
+included within it. It is possible to include the delimiter within the pattern
+by escaping it, for example
+
+  /abc\/def/
+

+If you do so, the escape and the delimiter form part of the pattern, but since
+delimiters are always non-alphanumeric, this does not affect its interpretation.
+If the terminating delimiter is immediately followed by a backslash, for
+example,
+
+  /abc/\
+

+then a backslash is added to the end of the pattern. This is done to provide a
+way of testing the error condition that arises if a pattern finishes with a
+backslash, because
+
+  /abc\/
+

+is interpreted as the first line of a pattern that starts with "abc/", causing
+pcretest to read the next line as a continuation of the regular expression.
+

+
PATTERN MODIFIERS

+

+A pattern may be followed by any number of modifiers, which are mostly single
+characters, though some of these can be qualified by further characters.
+Following Perl usage, these are referred to below as, for example, "the
+/i modifier", even though the delimiter of the pattern need not always be
+a slash, and no slash is used when writing modifiers. White space may appear
+between the final pattern delimiter and the first modifier, and between the
+modifiers themselves. For reference, here is a complete list of modifiers. They
+fall into several groups that are described in detail in the following
+sections.
+
+  /8              set UTF mode
+  /9              set PCRE_NEVER_UTF (locks out UTF mode)
+  /?              disable UTF validity check
+  /+              show remainder of subject after match
+  /=              show all captures (not just those that are set)
+
+  /A              set PCRE_ANCHORED
+  /B              show compiled code
+  /C              set PCRE_AUTO_CALLOUT
+  /D              same as /B plus /I
+  /E              set PCRE_DOLLAR_ENDONLY
+  /F              flip byte order in compiled pattern
+  /f              set PCRE_FIRSTLINE
+  /G              find all matches (shorten string)
+  /g              find all matches (use startoffset)
+  /I              show information about pattern
+  /i              set PCRE_CASELESS
+  /J              set PCRE_DUPNAMES
+  /K              show backtracking control names
+  /L              set locale
+  /M              show compiled memory size
+  /m              set PCRE_MULTILINE
+  /N              set PCRE_NO_AUTO_CAPTURE
+  /P              use the POSIX wrapper
+  /S              study the pattern after compilation
+  /s              set PCRE_DOTALL
+  /T              select character tables
+  /U              set PCRE_UNGREEDY
+  /W              set PCRE_UCP
+  /X              set PCRE_EXTRA
+  /x              set PCRE_EXTENDED
+  /Y              set PCRE_NO_START_OPTIMIZE
+  /Z              don't show lengths in /B output
+
+  /<any>          set PCRE_NEWLINE_ANY
+  /<anycrlf>      set PCRE_NEWLINE_ANYCRLF
+  /<cr>           set PCRE_NEWLINE_CR
+  /<crlf>         set PCRE_NEWLINE_CRLF
+  /<lf>           set PCRE_NEWLINE_LF
+  /<bsr_anycrlf>  set PCRE_BSR_ANYCRLF
+  /<bsr_unicode>  set PCRE_BSR_UNICODE
+  /<JS>           set PCRE_JAVASCRIPT_COMPAT
+
+

+

+

+Perl-compatible modifiers
+

+

+The /i, /m, /s, and /x modifiers set the PCRE_CASELESS,
+PCRE_MULTILINE, PCRE_DOTALL, or PCRE_EXTENDED options, respectively, when
+pcre[16|32]_compile() is called. These four modifier letters have the same
+effect as they do in Perl. For example:
+
+  /caseless/i
+
+

+

+

+Modifiers for other PCRE options
+

+

+The following table shows additional modifiers for setting PCRE compile-time
+options that do not correspond to anything in Perl:
+
+  /8              PCRE_UTF8           ) when using the 8-bit
+  /?              PCRE_NO_UTF8_CHECK  )   library
+
+  /8              PCRE_UTF16          ) when using the 16-bit
+  /?              PCRE_NO_UTF16_CHECK )   library
+
+  /8              PCRE_UTF32          ) when using the 32-bit
+  /?              PCRE_NO_UTF32_CHECK )   library
+
+  /9              PCRE_NEVER_UTF
+  /A              PCRE_ANCHORED
+  /C              PCRE_AUTO_CALLOUT
+  /E              PCRE_DOLLAR_ENDONLY
+  /f              PCRE_FIRSTLINE
+  /J              PCRE_DUPNAMES
+  /N              PCRE_NO_AUTO_CAPTURE
+  /U              PCRE_UNGREEDY
+  /W              PCRE_UCP
+  /X              PCRE_EXTRA
+  /Y              PCRE_NO_START_OPTIMIZE
+  /<any>          PCRE_NEWLINE_ANY
+  /<anycrlf>      PCRE_NEWLINE_ANYCRLF
+  /<cr>           PCRE_NEWLINE_CR
+  /<crlf>         PCRE_NEWLINE_CRLF
+  /<lf>           PCRE_NEWLINE_LF
+  /<bsr_anycrlf>  PCRE_BSR_ANYCRLF
+  /<bsr_unicode>  PCRE_BSR_UNICODE
+  /<JS>           PCRE_JAVASCRIPT_COMPAT
+

+The modifiers that are enclosed in angle brackets are literal strings as shown,
+including the angle brackets, but the letters within can be in either case.
+This example sets multiline matching with CRLF as the line ending sequence:
+
+  /^abc/m<CRLF>
+

+As well as turning on the PCRE_UTF8/16/32 option, the /8 modifier causes
+all non-printing characters in output strings to be printed using the
+\x{hh...} notation. Otherwise, those less than 0x100 are output in hex without
+the curly brackets.
+

+

+Full details of the PCRE options are given in the
+pcreapi
+documentation.
+

+

+Finding all matches in a string
+

+

+Searching for all possible matches within each subject string can be requested
+by the /g or /G modifier. After finding a match, PCRE is called
+again to search the remainder of the subject string. The difference between
+/g and /G is that the former uses the startoffset argument to
+pcre[16|32]_exec() to start searching at a new point within the entire
+string (which is in effect what Perl does), whereas the latter passes over a
+shortened substring. This makes a difference to the matching process if the
+pattern begins with a lookbehind assertion (including \b or \B).
+

+

+If any call to pcre[16|32]_exec() in a /g or /G sequence matches
+an empty string, the next call is done with the PCRE_NOTEMPTY_ATSTART and
+PCRE_ANCHORED flags set in order to search for another, non-empty, match at the
+same point. If this second match fails, the start offset is advanced, and the
+normal match is retried. This imitates the way Perl handles such cases when
+using the /g modifier or the split() function. Normally, the start
+offset is advanced by one character, but if the newline convention recognizes
+CRLF as a newline, and the current character is CR followed by LF, an advance
+of two is used.
+

+

+Other modifiers
+

+

+There are yet more modifiers for controlling the way pcretest
+operates.
+

+

+The /+ modifier requests that as well as outputting the substring that
+matched the entire pattern, pcretest should in addition output the
+remainder of the subject string. This is useful for tests where the subject
+contains multiple copies of the same substring. If the + modifier appears
+twice, the same action is taken for captured substrings. In each case the
+remainder is output on the following line with a plus character following the
+capture number. Note that this modifier must not immediately follow the /S
+modifier because /S+ and /S++ have other meanings.
+

+

+The /= modifier requests that the values of all potential captured
+parentheses be output after a match. By default, only those up to the highest
+one actually used in the match are output (corresponding to the return code
+from pcre[16|32]_exec()). Values in the offsets vector corresponding to
+higher numbers should be set to -1, and these are output as "<unset>". This
+modifier gives a way of checking that this is happening.
+

+

+The /B modifier is a debugging feature. It requests that pcretest
+output a representation of the compiled code after compilation. Normally this
+information contains length and offset values; however, if /Z is also
+present, this data is replaced by spaces. This is a special feature for use in
+the automatic test scripts; it ensures that the same output is generated for
+different internal link sizes.
+

+

+The /D modifier is a PCRE debugging feature, and is equivalent to
+/BI, that is, both the /B and the /I modifiers.
+

+

+The /F modifier causes pcretest to flip the byte order of the
+2-byte and 4-byte fields in the compiled pattern. This facility is for testing
+the feature in PCRE that allows it to execute patterns that were compiled on a
+host with a different endianness. This feature is not available when the POSIX
+interface to PCRE is being used, that is, when the /P pattern modifier is
+specified. See also the section about saving and reloading compiled patterns
+below.
+

+

+The /I modifier requests that pcretest output information about the
+compiled pattern (whether it is anchored, has a fixed first character, and
+so on). It does this by calling pcre[16|32]_fullinfo() after compiling a
+pattern. If the pattern is studied, the results of that are also output.
+

+

+The /K modifier requests pcretest to show names from backtracking
+control verbs that are returned from calls to pcre[16|32]_exec(). It causes
+pcretest to create a pcre[16|32]_extra block if one has not already
+been created by a call to pcre[16|32]_study(), and to set the
+PCRE_EXTRA_MARK flag and the mark field within it, every time that
+pcre[16|32]_exec() is called. If the variable that the mark field
+points to is non-NULL for a match, non-match, or partial match, pcretest
+prints the string to which it points. For a match, this is shown on a line by
+itself, tagged with "MK:". For a non-match it is added to the message.
+

+

+The /L modifier must be followed directly by the name of a locale, for
+example,
+
+  /pattern/Lfr_FR
+

+For this reason, it must be the last modifier. The given locale is set,
+pcre[16|32]_maketables() is called to build a set of character tables for
+the locale, and this is then passed to pcre[16|32]_compile() when compiling
+the regular expression. Without an /L (or /T) modifier, NULL is
+passed as the tables pointer; that is, /L applies only to the expression
+on which it appears.
+

+

+The /M modifier causes the size in bytes of the memory block used to hold
+the compiled pattern to be output. This does not include the size of the
+pcre[16|32] block; it is just the actual compiled data. If the pattern is
+successfully studied with the PCRE_STUDY_JIT_COMPILE option, the size of the
+JIT compiled code is also output.
+

+

+The /S modifier causes pcre[16|32]_study() to be called after the
+expression has been compiled, and the results used when the expression is
+matched. There are a number of qualifying characters that may follow /S.
+They may appear in any order.
+

+

+If S is followed by an exclamation mark, pcre[16|32]_study() is called
+with the PCRE_STUDY_EXTRA_NEEDED option, causing it always to return a
+pcre_extra block, even when studying discovers no useful information.
+

+

+If /S is followed by a second S character, it suppresses studying, even
+if it was requested externally by the -s command line option. This makes
+it possible to specify that certain patterns are always studied, and others are
+never studied, independently of -s. This feature is used in the test
+files in a few cases where the output is different when the pattern is studied.
+

+

+If the /S modifier is followed by a + character, the call to
+pcre[16|32]_study() is made with all the JIT study options, requesting
+just-in-time optimization support if it is available, for both normal and
+partial matching. If you want to restrict the JIT compiling modes, you can
+follow /S+ with a digit in the range 1 to 7:
+
+  1  normal match only
+  2  soft partial match only
+  3  normal match and soft partial match
+  4  hard partial match only
+  6  soft and hard partial match
+  7  all three modes (default)
+

+If /S++ is used instead of /S+ (with or without a following digit),
+the text "(JIT)" is added to the first output line after a match or no match
+when JIT-compiled code was actually used.
+

+

+Note that there is also an independent /+ modifier; it must not be given
+immediately after /S or /S+ because this will be misinterpreted.
+

+

+If JIT studying is successful, the compiled JIT code will automatically be used
+when pcre[16|32]_exec() is run, except when incompatible run-time options
+are specified. For more details, see the
+pcrejit
+documentation. See also the \J escape sequence below for a way of
+setting the size of the JIT stack.
+

+

+Finally, if /S is followed by a minus character, JIT compilation is
+suppressed, even if it was requested externally by the -s command line
+option. This makes it possible to specify that JIT is never to be used for
+certain patterns.
+

+

+The /T modifier must be followed by a single digit. It causes a specific
+set of built-in character tables to be passed to pcre[16|32]_compile(). It
+is used in the standard PCRE tests to check behaviour with different character
+tables. The digit specifies the tables as follows:
+
+  0   the default ASCII tables, as distributed in
+        pcre_chartables.c.dist
+  1   a set of tables defining ISO 8859 characters
+

+In table 1, some characters whose codes are greater than 128 are identified as
+letters, digits, spaces, etc.
+

+

+Using the POSIX wrapper API
+

+

+The /P modifier causes pcretest to call PCRE via the POSIX wrapper
+API rather than its native API. This supports only the 8-bit library. When
+/P is set, the following modifiers set options for the regcomp()
+function:
+
+  /i    REG_ICASE
+  /m    REG_NEWLINE
+  /N    REG_NOSUB
+  /s    REG_DOTALL     )
+  /U    REG_UNGREEDY   ) These options are not part of
+  /W    REG_UCP        )   the POSIX standard
+  /8    REG_UTF8       )
+

+The /+ modifier works as described above. All other modifiers are
+ignored.
+

+
DATA LINES

+

+Before each data line is passed to pcre[16|32]_exec(), leading and trailing
+white space is removed, and it is then scanned for \ escapes. Some of these
+are pretty esoteric features, intended for checking out some of the more
+complicated features of PCRE. If you are just testing "ordinary" regular
+expressions, you probably don't need any of these. The following escapes are
+recognized:
+
+  \a         alarm (BEL, \x07)
+  \b         backspace (\x08)
+  \e         escape (\x27)
+  \f         form feed (\x0c)
+  \n         newline (\x0a)
+  \qdd       set the PCRE_MATCH_LIMIT limit to dd (any number of digits)
+  \r         carriage return (\x0d)
+  \t         tab (\x09)
+  \v         vertical tab (\x0b)
+  \nnn       octal character (up to 3 octal digits); always
+               a byte unless > 255 in UTF-8 or 16-bit or 32-bit mode
+  \xhh       hexadecimal byte (up to 2 hex digits)
+  \x{hh...}  hexadecimal character (any number of hex digits)
+  \A         pass the PCRE_ANCHORED option to pcre[16|32]_exec() or pcre[16|32]_dfa_exec()
+  \B         pass the PCRE_NOTBOL option to pcre[16|32]_exec() or pcre[16|32]_dfa_exec()
+  \Cdd       call pcre[16|32]_copy_substring() for substring dd after a successful match (number less than 32)
+  \Cname     call pcre[16|32]_copy_named_substring() for substring "name" after a successful match (name termin-
+               ated by next non alphanumeric character)
+  \C+        show the current captured substrings at callout time
+  \C-        do not supply a callout function
+  \C!n       return 1 instead of 0 when callout number n is reached
+  \C!n!m     return 1 instead of 0 when callout number n is reached for the nth time
+  \C*n       pass the number n (may be negative) as callout data; this is used as the callout return value
+  \D         use the pcre[16|32]_dfa_exec() match function
+  \F         only shortest match for pcre[16|32]_dfa_exec()
+  \Gdd       call pcre[16|32]_get_substring() for substring dd after a successful match (number less than 32)
+  \Gname     call pcre[16|32]_get_named_substring() for substring "name" after a successful match (name termin-
+               ated by next non-alphanumeric character)
+  \Jdd       set up a JIT stack of dd kilobytes maximum (any number of digits)
+  \L         call pcre[16|32]_get_substringlist() after a successful match
+  \M         discover the minimum MATCH_LIMIT and MATCH_LIMIT_RECURSION settings
+  \N         pass the PCRE_NOTEMPTY option to pcre[16|32]_exec() or pcre[16|32]_dfa_exec(); if used twice, pass the
+               PCRE_NOTEMPTY_ATSTART option
+  \Odd       set the size of the output vector passed to pcre[16|32]_exec() to dd (any number of digits)
+  \P         pass the PCRE_PARTIAL_SOFT option to pcre[16|32]_exec() or pcre[16|32]_dfa_exec(); if used twice, pass the
+               PCRE_PARTIAL_HARD option
+  \Qdd       set the PCRE_MATCH_LIMIT_RECURSION limit to dd (any number of digits)
+  \R         pass the PCRE_DFA_RESTART option to pcre[16|32]_dfa_exec()
+  \S         output details of memory get/free calls during matching
+  \Y         pass the PCRE_NO_START_OPTIMIZE option to pcre[16|32]_exec() or pcre[16|32]_dfa_exec()
+  \Z         pass the PCRE_NOTEOL option to pcre[16|32]_exec() or pcre[16|32]_dfa_exec()
+  \?         pass the PCRE_NO_UTF[8|16|32]_CHECK option to pcre[16|32]_exec() or pcre[16|32]_dfa_exec()
+  \>dd       start the match at offset dd (optional "-"; then any number of digits); this sets the startoffset
+               argument for pcre[16|32]_exec() or pcre[16|32]_dfa_exec()
+  \<cr>      pass the PCRE_NEWLINE_CR option to pcre[16|32]_exec() or pcre[16|32]_dfa_exec()
+  \<lf>      pass the PCRE_NEWLINE_LF option to pcre[16|32]_exec() or pcre[16|32]_dfa_exec()
+  \<crlf>    pass the PCRE_NEWLINE_CRLF option to pcre[16|32]_exec() or pcre[16|32]_dfa_exec()
+  \<anycrlf> pass the PCRE_NEWLINE_ANYCRLF option to pcre[16|32]_exec() or pcre[16|32]_dfa_exec()
+  \<any>     pass the PCRE_NEWLINE_ANY option to pcre[16|32]_exec() or pcre[16|32]_dfa_exec()
+

+The use of \x{hh...} is not dependent on the use of the /8 modifier on
+the pattern. It is recognized always. There may be any number of hexadecimal
+digits inside the braces; invalid values provoke error messages.
+

+

+Note that \xhh specifies one byte rather than one character in UTF-8 mode;
+this makes it possible to construct invalid UTF-8 sequences for testing
+purposes. On the other hand, \x{hh} is interpreted as a UTF-8 character in
+UTF-8 mode, generating more than one byte if the value is greater than 127.
+When testing the 8-bit library not in UTF-8 mode, \x{hh} generates one byte
+for values less than 256, and causes an error for greater values.
+

+

+In UTF-16 mode, all 4-digit \x{hhhh} values are accepted. This makes it
+possible to construct invalid UTF-16 sequences for testing purposes.
+

+

+In UTF-32 mode, all 4- to 8-digit \x{...} values are accepted. This makes it
+possible to construct invalid UTF-32 sequences for testing purposes.
+

+

+The escapes that specify line ending sequences are literal strings, exactly as
+shown. No more than one newline setting should be present in any data line.
+

+

+A backslash followed by anything else just escapes the anything else. If
+the very last character is a backslash, it is ignored. This gives a way of
+passing an empty line as data, since a real empty line terminates the data
+input.
+

+

+The \J escape provides a way of setting the maximum stack size that is
+used by the just-in-time optimization code. It is ignored if JIT optimization
+is not being used. Providing a stack that is larger than the default 32K is
+necessary only for very complicated patterns.
+

+

+If \M is present, pcretest calls pcre[16|32]_exec() several times,
+with different values in the match_limit and match_limit_recursion
+fields of the pcre[16|32]_extra data structure, until it finds the minimum
+numbers for each parameter that allow pcre[16|32]_exec() to complete without
+error. Because this is testing a specific feature of the normal interpretive
+pcre[16|32]_exec() execution, the use of any JIT optimization that might
+have been set up by the /S+ qualifier of -s+ option is disabled.
+

+

+The match_limit number is a measure of the amount of backtracking
+that takes place, and checking it out can be instructive. For most simple
+matches, the number is quite small, but for patterns with very large numbers of
+matching possibilities, it can become large very quickly with increasing length
+of subject string. The match_limit_recursion number is a measure of how
+much stack (or, if PCRE is compiled with NO_RECURSE, how much heap) memory is
+needed to complete the match attempt.
+

+

+When \O is used, the value specified may be higher or lower than the size set
+by the -O command line option (or defaulted to 45); \O applies only to
+the call of pcre[16|32]_exec() for the line in which it appears.
+

+

+If the /P modifier was present on the pattern, causing the POSIX wrapper
+API to be used, the only option-setting sequences that have any effect are \B,
+\N, and \Z, causing REG_NOTBOL, REG_NOTEMPTY, and REG_NOTEOL, respectively,
+to be passed to regexec().
+

+
THE ALTERNATIVE MATCHING FUNCTION

+

+By default, pcretest uses the standard PCRE matching function,
+pcre[16|32]_exec() to match each data line. PCRE also supports an
+alternative matching function, pcre[16|32]_dfa_test(), which operates in a
+different way, and has some restrictions. The differences between the two
+functions are described in the
+pcrematching
+documentation.
+

+

+If a data line contains the \D escape sequence, or if the command line
+contains the -dfa option, the alternative matching function is used.
+This function finds all possible matches at a given point. If, however, the \F
+escape sequence is present in the data line, it stops after the first match is
+found. This is always the shortest possible match.
+

+
DEFAULT OUTPUT FROM PCRETEST

+

+This section describes the output when the normal matching function,
+pcre[16|32]_exec(), is being used.
+

+

+When a match succeeds, pcretest outputs the list of captured substrings
+that pcre[16|32]_exec() returns, starting with number 0 for the string that
+matched the whole pattern. Otherwise, it outputs "No match" when the return is
+PCRE_ERROR_NOMATCH, and "Partial match:" followed by the partially matching
+substring when pcre[16|32]_exec() returns PCRE_ERROR_PARTIAL. (Note that
+this is the entire substring that was inspected during the partial match; it
+may include characters before the actual match start if a lookbehind assertion,
+\K, \b, or \B was involved.) For any other return, pcretest outputs
+the PCRE negative error number and a short descriptive phrase. If the error is
+a failed UTF string check, the offset of the start of the failing character and
+the reason code are also output, provided that the size of the output vector is
+at least two. Here is an example of an interactive pcretest run.
+
+  $ pcretest
+  PCRE version 8.13 2011-04-30
+
+    re> /^abc(\d+)/
+  data> abc123
+   0: abc123
+   1: 123
+  data> xyz
+  No match
+

+Unset capturing substrings that are not followed by one that is set are not
+returned by pcre[16|32]_exec(), and are not shown by pcretest. In the
+following example, there are two capturing substrings, but when the first data
+line is matched, the second, unset substring is not shown. An "internal" unset
+substring is shown as "<unset>", as for the second data line.
+
+    re> /(a)|(b)/
+  data> a
+   0: a
+   1: a
+  data> b
+   0: b
+   1: <unset>
+   2: b
+

+If the strings contain any non-printing characters, they are output as \xhh
+escapes if the value is less than 256 and UTF mode is not set. Otherwise they
+are output as \x{hh...} escapes. See below for the definition of non-printing
+characters. If the pattern has the /+ modifier, the output for substring
+0 is followed by the the rest of the subject string, identified by "0+" like
+this:
+
+    re> /cat/+
+  data> cataract
+   0: cat
+   0+ aract
+

+If the pattern has the /g or /G modifier, the results of successive
+matching attempts are output in sequence, like this:
+
+    re> /\Bi(\w\w)/g
+  data> Mississippi
+   0: iss
+   1: ss
+   0: iss
+   1: ss
+   0: ipp
+   1: pp
+

+"No match" is output only if the first match attempt fails. Here is an example
+of a failure message (the offset 4 that is specified by \>4 is past the end of
+the subject string):
+
+    re> /xyz/
+  data> xyz\>4
+  Error -24 (bad offset value)
+

+

+

+If any of the sequences \C, \G, or \L are present in a
+data line that is successfully matched, the substrings extracted by the
+convenience functions are output with C, G, or L after the string number
+instead of a colon. This is in addition to the normal full list. The string
+length (that is, the return from the extraction function) is given in
+parentheses after each string for \C and \G.
+

+

+Note that whereas patterns can be continued over several lines (a plain ">"
+prompt is used for continuations), data lines may not. However newlines can be
+included in data by means of the \n escape (or \r, \r\n, etc., depending on
+the newline sequence setting).
+

+
OUTPUT FROM THE ALTERNATIVE MATCHING FUNCTION

+

+When the alternative matching function, pcre[16|32]_dfa_exec(), is used (by
+means of the \D escape sequence or the -dfa command line option), the
+output consists of a list of all the matches that start at the first point in
+the subject where there is at least one match. For example:
+
+    re> /(tang|tangerine|tan)/
+  data> yellow tangerine\D
+   0: tangerine
+   1: tang
+   2: tan
+

+(Using the normal matching function on this data finds only "tang".) The
+longest matching string is always given first (and numbered zero). After a
+PCRE_ERROR_PARTIAL return, the output is "Partial match:", followed by the
+partially matching substring. (Note that this is the entire substring that was
+inspected during the partial match; it may include characters before the actual
+match start if a lookbehind assertion, \K, \b, or \B was involved.)
+

+

+If /g is present on the pattern, the search for further matches resumes
+at the end of the longest match. For example:
+
+    re> /(tang|tangerine|tan)/g
+  data> yellow tangerine and tangy sultana\D
+   0: tangerine
+   1: tang
+   2: tan
+   0: tang
+   1: tan
+   0: tan
+

+Since the matching function does not support substring capture, the escape
+sequences that are concerned with captured substrings are not relevant.
+

+
RESTARTING AFTER A PARTIAL MATCH

+

+When the alternative matching function has given the PCRE_ERROR_PARTIAL return,
+indicating that the subject partially matched the pattern, you can restart the
+match with additional subject data by means of the \R escape sequence. For
+example:
+
+    re> /^\d?\d(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\d\d$/
+  data> 23ja\P\D
+  Partial match: 23ja
+  data> n05\R\D
+   0: n05
+

+For further information about partial matching, see the
+pcrepartial
+documentation.
+

+
CALLOUTS

+

+If the pattern contains any callout requests, pcretest's callout function
+is called during matching. This works with both matching functions. By default,
+the called function displays the callout number, the start and current
+positions in the text at the callout time, and the next pattern item to be
+tested. For example:
+
+  --->pqrabcdef
+    0    ^  ^     \d
+

+This output indicates that callout number 0 occurred for a match attempt
+starting at the fourth character of the subject string, when the pointer was at
+the seventh character of the data, and when the next pattern item was \d. Just
+one circumflex is output if the start and current positions are the same.
+

+

+Callouts numbered 255 are assumed to be automatic callouts, inserted as a
+result of the /C pattern modifier. In this case, instead of showing the
+callout number, the offset in the pattern, preceded by a plus, is output. For
+example:
+
+    re> /\d?[A-E]\*/C
+  data> E*
+  --->E*
+   +0 ^      \d?
+   +3 ^      [A-E]
+   +8 ^^     \*
+  +10 ^ ^
+   0: E*
+

+If a pattern contains (*MARK) items, an additional line is output whenever
+a change of latest mark is passed to the callout function. For example:
+
+    re> /a(*MARK:X)bc/C
+  data> abc
+  --->abc
+   +0 ^       a
+   +1 ^^      (*MARK:X)
+  +10 ^^      b
+  Latest Mark: X
+  +11 ^ ^     c
+  +12 ^  ^
+   0: abc
+

+The mark changes between matching "a" and "b", but stays the same for the rest
+of the match, so nothing more is output. If, as a result of backtracking, the
+mark reverts to being unset, the text "<unset>" is output.
+

+

+The callout function in pcretest returns zero (carry on matching) by
+default, but you can use a \C item in a data line (as described above) to
+change this and other parameters of the callout.
+

+

+Inserting callouts can be helpful when using pcretest to check
+complicated regular expressions. For further information about callouts, see
+the
+pcrecallout
+documentation.
+

+
NON-PRINTING CHARACTERS

+

+When pcretest is outputting text in the compiled version of a pattern,
+bytes other than 32-126 are always treated as non-printing characters are are
+therefore shown as hex escapes.
+

+

+When pcretest is outputting text that is a matched part of a subject
+string, it behaves in the same way, unless a different locale has been set for
+the pattern (using the /L modifier). In this case, the isprint()
+function to distinguish printing and non-printing characters.
+

+
SAVING AND RELOADING COMPILED PATTERNS

+

+The facilities described in this section are not available when the POSIX
+interface to PCRE is being used, that is, when the /P pattern modifier is
+specified.
+

+

+When the POSIX interface is not in use, you can cause pcretest to write a
+compiled pattern to a file, by following the modifiers with > and a file name.
+For example:
+
+  /pattern/im >/some/file
+

+See the
+pcreprecompile
+documentation for a discussion about saving and re-using compiled patterns.
+Note that if the pattern was successfully studied with JIT optimization, the
+JIT data cannot be saved.
+

+

+The data that is written is binary. The first eight bytes are the length of the
+compiled pattern data followed by the length of the optional study data, each
+written as four bytes in big-endian order (most significant byte first). If
+there is no study data (either the pattern was not studied, or studying did not
+return any data), the second length is zero. The lengths are followed by an
+exact copy of the compiled pattern. If there is additional study data, this
+(excluding any JIT data) follows immediately after the compiled pattern. After
+writing the file, pcretest expects to read a new pattern.
+

+

+A saved pattern can be reloaded into pcretest by specifying < and a file
+name instead of a pattern. The name of the file must not contain a < character,
+as otherwise pcretest will interpret the line as a pattern delimited by <
+characters.
+For example:
+
+   re> </some/file
+  Compiled pattern loaded from /some/file
+  No study data
+

+If the pattern was previously studied with the JIT optimization, the JIT
+information cannot be saved and restored, and so is lost. When the pattern has
+been loaded, pcretest proceeds to read data lines in the usual way.
+

+

+You can copy a file written by pcretest to a different host and reload it
+there, even if the new host has opposite endianness to the one on which the
+pattern was compiled. For example, you can compile on an i86 machine and run on
+a SPARC machine. When a pattern is reloaded on a host with different
+endianness, the confirmation message is changed to:
+
+  Compiled pattern (byte-inverted) loaded from /some/file
+

+The test suite contains some saved pre-compiled patterns with different
+endianness. These are reloaded using "<!" instead of just "<". This suppresses
+the "(byte-inverted)" text so that the output is the same on all hosts. It also
+forces debugging output once the pattern has been reloaded.
+

+

+File names for saving and reloading can be absolute or relative, but note that
+the shell facility of expanding a file name that starts with a tilde (~) is not
+available.
+

+

+The ability to save and reload files in pcretest is intended for testing
+and experimentation. It is not intended for production use because only a
+single pattern can be written to a file. Furthermore, there is no facility for
+supplying custom character tables for use with a reloaded pattern. If the
+original pattern was compiled with custom tables, an attempt to match a subject
+string using a reloaded pattern is likely to cause pcretest to crash.
+Finally, if you attempt to load a file that is not in the correct format, the
+result is undefined.
+

+
SEE ALSO

+

+pcre(3), pcre16(3), pcre32(3), pcreapi(3),
+pcrecallout(3),
+pcrejit, pcrematching(3), pcrepartial(d),
+pcrepattern(3), pcreprecompile(3).
+

+
AUTHOR

+

+Philip Hazel
+

+University Computing Service
+

+Cambridge CB2 3QH, England.
+

+

+
REVISION

+

+Last updated: 26 April 2013
+

+Copyright © 1997-2013 University of Cambridge.
+

+

+Return to the PCRE index page.
+

diff --git a/pcre/doc/html/pcreunicode.html b/pcre/doc/html/pcreunicode.html
new file mode 100644
index 00000000000..ab36bc61e37
--- /dev/null
+++ b/pcre/doc/html/pcreunicode.html
@@ -0,0 +1,262 @@
+
+
+pcreunicode specification
+
+
+
pcreunicode man page

+

+Return to the PCRE index page.
+

+

+This page is part of the PCRE HTML documentation. It was generated automatically
+from the original man page. If there is any nonsense in it, please consult the
+man page, in case the conversion went wrong.
+

+

+UTF-8, UTF-16, UTF-32, AND UNICODE PROPERTY SUPPORT
+

+

+As well as UTF-8 support, PCRE also supports UTF-16 (from release 8.30) and
+UTF-32 (from release 8.32), by means of two additional libraries. They can be
+built as well as, or instead of, the 8-bit library.
+

+

+UTF-8 SUPPORT
+

+

+In order process UTF-8 strings, you must build PCRE's 8-bit library with UTF
+support, and, in addition, you must call
+pcre_compile()
+with the PCRE_UTF8 option flag, or the pattern must start with the sequence
+(*UTF8) or (*UTF). When either of these is the case, both the pattern and any
+subject strings that are matched against it are treated as UTF-8 strings
+instead of strings of individual 1-byte characters.
+

+

+UTF-16 AND UTF-32 SUPPORT
+

+

+In order process UTF-16 or UTF-32 strings, you must build PCRE's 16-bit or
+32-bit library with UTF support, and, in addition, you must call
+pcre16_compile()
+or
+pcre32_compile()
+with the PCRE_UTF16 or PCRE_UTF32 option flag, as appropriate. Alternatively,
+the pattern must start with the sequence (*UTF16), (*UTF32), as appropriate, or
+(*UTF), which can be used with either library. When UTF mode is set, both the
+pattern and any subject strings that are matched against it are treated as
+UTF-16 or UTF-32 strings instead of strings of individual 16-bit or 32-bit
+characters.
+

+

+UTF SUPPORT OVERHEAD
+

+

+If you compile PCRE with UTF support, but do not use it at run time, the
+library will be a bit bigger, but the additional run time overhead is limited
+to testing the PCRE_UTF[8|16|32] flag occasionally, so should not be very big.
+

+

+UNICODE PROPERTY SUPPORT
+

+

+If PCRE is built with Unicode character property support (which implies UTF
+support), the escape sequences \p{..}, \P{..}, and \X can be used.
+The available properties that can be tested are limited to the general
+category properties such as Lu for an upper case letter or Nd for a decimal
+number, the Unicode script names such as Arabic or Han, and the derived
+properties Any and L&. Full lists is given in the
+pcrepattern
+and
+pcresyntax
+documentation. Only the short names for properties are supported. For example,
+\p{L} matches a letter. Its Perl synonym, \p{Letter}, is not supported.
+Furthermore, in Perl, many properties may optionally be prefixed by "Is", for
+compatibility with Perl 5.6. PCRE does not support this.
+

+

+Validity of UTF-8 strings
+

+

+When you set the PCRE_UTF8 flag, the byte strings passed as patterns and
+subjects are (by default) checked for validity on entry to the relevant
+functions. The entire string is checked before any other processing takes
+place. From release 7.3 of PCRE, the check is according the rules of RFC 3629,
+which are themselves derived from the Unicode specification. Earlier releases
+of PCRE followed the rules of RFC 2279, which allows the full range of 31-bit
+values (0 to 0x7FFFFFFF). The current check allows only values in the range U+0
+to U+10FFFF, excluding the surrogate area. (From release 8.33 the so-called
+"non-character" code points are no longer excluded because Unicode corrigendum
+#9 makes it clear that they should not be.)
+

+

+Characters in the "Surrogate Area" of Unicode are reserved for use by UTF-16,
+where they are used in pairs to encode codepoints with values greater than
+0xFFFF. The code points that are encoded by UTF-16 pairs are available
+independently in the UTF-8 and UTF-32 encodings. (In other words, the whole
+surrogate thing is a fudge for UTF-16 which unfortunately messes up UTF-8 and
+UTF-32.)
+

+

+If an invalid UTF-8 string is passed to PCRE, an error return is given. At
+compile time, the only additional information is the offset to the first byte
+of the failing character. The run-time functions pcre_exec() and
+pcre_dfa_exec() also pass back this information, as well as a more
+detailed reason code if the caller has provided memory in which to do this.
+

+

+In some situations, you may already know that your strings are valid, and
+therefore want to skip these checks in order to improve performance, for
+example in the case of a long subject string that is being scanned repeatedly.
+If you set the PCRE_NO_UTF8_CHECK flag at compile time or at run time, PCRE
+assumes that the pattern or subject it is given (respectively) contains only
+valid UTF-8 codes. In this case, it does not diagnose an invalid UTF-8 string.
+

+

+Note that passing PCRE_NO_UTF8_CHECK to pcre_compile() just disables the
+check for the pattern; it does not also apply to subject strings. If you want
+to disable the check for a subject string you must pass this option to
+pcre_exec() or pcre_dfa_exec().
+

+

+If you pass an invalid UTF-8 string when PCRE_NO_UTF8_CHECK is set, the result
+is undefined and your program may crash.
+

+

+Validity of UTF-16 strings
+

+

+When you set the PCRE_UTF16 flag, the strings of 16-bit data units that are
+passed as patterns and subjects are (by default) checked for validity on entry
+to the relevant functions. Values other than those in the surrogate range
+U+D800 to U+DFFF are independent code points. Values in the surrogate range
+must be used in pairs in the correct manner.
+

+

+If an invalid UTF-16 string is passed to PCRE, an error return is given. At
+compile time, the only additional information is the offset to the first data
+unit of the failing character. The run-time functions pcre16_exec() and
+pcre16_dfa_exec() also pass back this information, as well as a more
+detailed reason code if the caller has provided memory in which to do this.
+

+

+In some situations, you may already know that your strings are valid, and
+therefore want to skip these checks in order to improve performance. If you set
+the PCRE_NO_UTF16_CHECK flag at compile time or at run time, PCRE assumes that
+the pattern or subject it is given (respectively) contains only valid UTF-16
+sequences. In this case, it does not diagnose an invalid UTF-16 string.
+However, if an invalid string is passed, the result is undefined.
+

+

+Validity of UTF-32 strings
+

+

+When you set the PCRE_UTF32 flag, the strings of 32-bit data units that are
+passed as patterns and subjects are (by default) checked for validity on entry
+to the relevant functions.  This check allows only values in the range U+0
+to U+10FFFF, excluding the surrogate area U+D800 to U+DFFF.
+

+

+If an invalid UTF-32 string is passed to PCRE, an error return is given. At
+compile time, the only additional information is the offset to the first data
+unit of the failing character. The run-time functions pcre32_exec() and
+pcre32_dfa_exec() also pass back this information, as well as a more
+detailed reason code if the caller has provided memory in which to do this.
+

+

+In some situations, you may already know that your strings are valid, and
+therefore want to skip these checks in order to improve performance. If you set
+the PCRE_NO_UTF32_CHECK flag at compile time or at run time, PCRE assumes that
+the pattern or subject it is given (respectively) contains only valid UTF-32
+sequences. In this case, it does not diagnose an invalid UTF-32 string.
+However, if an invalid string is passed, the result is undefined.
+

+

+General comments about UTF modes
+

+

+1. Codepoints less than 256 can be specified in patterns by either braced or
+unbraced hexadecimal escape sequences (for example, \x{b3} or \xb3). Larger
+values have to use braced sequences.
+

+

+2. Octal numbers up to \777 are recognized, and in UTF-8 mode they match
+two-byte characters for values greater than \177.
+

+

+3. Repeat quantifiers apply to complete UTF characters, not to individual
+data units, for example: \x{100}{3}.
+

+

+4. The dot metacharacter matches one UTF character instead of a single data
+unit.
+

+

+5. The escape sequence \C can be used to match a single byte in UTF-8 mode, or
+a single 16-bit data unit in UTF-16 mode, or a single 32-bit data unit in
+UTF-32 mode, but its use can lead to some strange effects because it breaks up
+multi-unit characters (see the description of \C in the
+pcrepattern
+documentation). The use of \C is not supported in the alternative matching
+function pcre[16|32]_dfa_exec(), nor is it supported in UTF mode by the
+JIT optimization of pcre[16|32]_exec(). If JIT optimization is requested
+for a UTF pattern that contains \C, it will not succeed, and so the matching
+will be carried out by the normal interpretive function.
+

+

+6. The character escapes \b, \B, \d, \D, \s, \S, \w, and \W correctly
+test characters of any code value, but, by default, the characters that PCRE
+recognizes as digits, spaces, or word characters remain the same set as in
+non-UTF mode, all with values less than 256. This remains true even when PCRE
+is built to include Unicode property support, because to do otherwise would
+slow down PCRE in many common cases. Note in particular that this applies to
+\b and \B, because they are defined in terms of \w and \W. If you really
+want to test for a wider sense of, say, "digit", you can use explicit Unicode
+property tests such as \p{Nd}. Alternatively, if you set the PCRE_UCP option,
+the way that the character escapes work is changed so that Unicode properties
+are used to determine which characters match. There are more details in the
+section on
+generic character types
+in the
+pcrepattern
+documentation.
+

+

+7. Similarly, characters that match the POSIX named character classes are all
+low-valued characters, unless the PCRE_UCP option is set.
+

+

+8. However, the horizontal and vertical white space matching escapes (\h, \H,
+\v, and \V) do match all the appropriate Unicode characters, whether or not
+PCRE_UCP is set.
+

+

+9. Case-insensitive matching applies only to characters whose values are less
+than 128, unless PCRE is built with Unicode property support. A few Unicode
+characters such as Greek sigma have more than two codepoints that are
+case-equivalent. Up to and including PCRE release 8.31, only one-to-one case
+mappings were supported, but later releases (with Unicode property support) do
+treat as case-equivalent all versions of characters such as Greek sigma.
+

+

+AUTHOR
+

+

+Philip Hazel
+

+University Computing Service
+

+Cambridge CB2 3QH, England.
+

+

+

+REVISION
+

+

+Last updated: 27 February 2013
+

+Copyright © 1997-2013 University of Cambridge.
+

+

+Return to the PCRE index page.
+

diff --git a/pcre/doc/index.html.src b/pcre/doc/index.html.src
new file mode 100644
index 00000000000..887f4d75071
--- /dev/null
+++ b/pcre/doc/index.html.src
@@ -0,0 +1,185 @@
+
+      
+
+PCRE specification
+
+
+
Perl-compatible Regular Expressions (PCRE)

+

+The HTML documentation for PCRE consists of a number of pages that are listed
+below in alphabetical order. If you are new to PCRE, please read the first one
+first.
+

+
+
+
+    
+
+
+    
+
+
+    
+
+
+    
+
+
+    
+
+
+    
+
+
+    
+
+
+    
+
+
+    
+
+
+    
+
+
+    
+
+
+    
+
+
+    
+
+
+    
+
+
+    
+
+
+    
+
+
+    
+
+
+    
+
+
+    
+
+
+    
+
+
+    
+
+
+    
+
+
+    
+
+
+    
+
pcre  Introductory page
pcre-config  Information about the installation configuration
pcre16  Discussion of the 16-bit PCRE library
pcre32  Discussion of the 32-bit PCRE library
pcreapi  PCRE's native API
pcrebuild  Building PCRE
pcrecallout  The callout facility
pcrecompat  Compability with Perl
pcrecpp  The C++ wrapper for the PCRE library
pcredemo  A demonstration C program that uses the PCRE library
pcregrep  The pcregrep command
pcrejit  Discussion of the just-in-time optimization support
pcrelimits  Details of size and other limits
pcrematching  Discussion of the two matching algorithms
pcrepartial  Using PCRE for partial matching
pcrepattern  Specification of the regular expressions supported by PCRE
pcreperform  Some comments on performance
pcreposix  The POSIX API to the PCRE 8-bit library
pcreprecompile  How to save and re-use compiled patterns
pcresample  Discussion of the pcredemo program
pcrestack  Discussion of PCRE's stack usage
pcresyntax  Syntax quick-reference summary
pcretest  The pcretest command for testing PCRE
pcreunicode  Discussion of Unicode and UTF-8/UTF-16/UTF-32 support

+
+

+There are also individual pages that summarize the interface for each function
+in the library. There is a single page for each triple of 8-bit/16-bit/32-bit
+functions.
+

+
+    
+
+
+    
+
+
+    
+
+
+    
+
+
+    
+
+
+    
+
+
+    
+
+
+    
+
+
+    
+
+
+    
+
+
+    
+
+
+    
+
+
+    
+
+
+    
+
+
+    
+
+
+    
+
+
+    
+
+
+    
+
+
+    
+
+
+    
+
+
+    
+
+
+    
+    
+
+    
+
+
+    
+
+
+    
+
+
+    
+
+
+    
+
+
+    
+
pcre_assign_jit_stack  Assign stack for JIT matching
pcre_compile  Compile a regular expression
pcre_compile2  Compile a regular expression (alternate interface)
pcre_config  Show build-time configuration options
pcre_copy_named_substring  Extract named substring into given buffer
pcre_copy_substring  Extract numbered substring into given buffer
pcre_dfa_exec  Match a compiled pattern to a subject string
+    (DFA algorithm; not Perl compatible)
pcre_exec  Match a compiled pattern to a subject string
+    (Perl compatible)
pcre_free_study  Free study data
pcre_free_substring  Free extracted substring
pcre_free_substring_list  Free list of extracted substrings
pcre_fullinfo  Extract information about a pattern
pcre_get_named_substring  Extract named substring into new memory
pcre_get_stringnumber  Convert captured string name to number
pcre_get_stringtable_entries  Find table entries for given string name
pcre_get_substring  Extract numbered substring into new memory
pcre_get_substring_list  Extract all substrings into new memory
pcre_jit_exec  Fast path interface to JIT matching
pcre_jit_stack_alloc  Create a stack for JIT matching
pcre_jit_stack_free  Free a JIT matching stack
pcre_maketables  Build character tables in current locale
pcre_pattern_to_host_byte_order  Convert compiled pattern to host byte order if necessary
pcre_refcount  Maintain reference count in compiled pattern
pcre_study  Study a compiled pattern
pcre_utf16_to_host_byte_order  Convert UTF-16 string to host byte order if necessary
pcre_utf32_to_host_byte_order  Convert UTF-32 string to host byte order if necessary
pcre_version  Return PCRE version and release date

+
+
diff --git a/pcre/doc/pcre-config.1 b/pcre/doc/pcre-config.1
new file mode 100644
index 00000000000..92a4b58bc4a
--- /dev/null
+++ b/pcre/doc/pcre-config.1
@@ -0,0 +1,92 @@
+.TH PCRE-CONFIG 1 "01 January 2012" "PCRE 8.30"
+.SH NAME
+pcre-config - program to return PCRE configuration
+.SH SYNOPSIS
+.rs
+.sp
+.B pcre-config  [--prefix] [--exec-prefix] [--version] [--libs]
+.ti +5n
+.B              [--libs16] [--libs32] [--libs-cpp] [--libs-posix]
+.ti +5n
+.B              [--cflags] [--cflags-posix]
+.
+.
+.SH DESCRIPTION
+.rs
+.sp
+\fBpcre-config\fP returns the configuration of the installed PCRE
+libraries and the options required to compile a program to use them. Some of
+the options apply only to the 8-bit, or 16-bit, or 32-bit libraries,
+respectively, and are
+not available if only one of those libraries has been built. If an unavailable
+option is encountered, the "usage" information is output.
+.
+.
+.SH OPTIONS
+.rs
+.TP 10
+\fB--prefix\fP
+Writes the directory prefix used in the PCRE installation for architecture
+independent files (\fI/usr\fP on many systems, \fI/usr/local\fP on some
+systems) to the standard output.
+.TP 10
+\fB--exec-prefix\fP
+Writes the directory prefix used in the PCRE installation for architecture
+dependent files (normally the same as \fB--prefix\fP) to the standard output.
+.TP 10
+\fB--version\fP
+Writes the version number of the installed PCRE libraries to the standard
+output.
+.TP 10
+\fB--libs\fP
+Writes to the standard output the command line options required to link
+with the 8-bit PCRE library (\fB-lpcre\fP on many systems).
+.TP 10
+\fB--libs16\fP
+Writes to the standard output the command line options required to link
+with the 16-bit PCRE library (\fB-lpcre16\fP on many systems).
+.TP 10
+\fB--libs32\fP
+Writes to the standard output the command line options required to link
+with the 32-bit PCRE library (\fB-lpcre32\fP on many systems).
+.TP 10
+\fB--libs-cpp\fP
+Writes to the standard output the command line options required to link with
+PCRE's C++ wrapper library (\fB-lpcrecpp\fP \fB-lpcre\fP on many
+systems).
+.TP 10
+\fB--libs-posix\fP
+Writes to the standard output the command line options required to link with
+PCRE's POSIX API wrapper library (\fB-lpcreposix\fP \fB-lpcre\fP on many
+systems).
+.TP 10
+\fB--cflags\fP
+Writes to the standard output the command line options required to compile
+files that use PCRE (this may include some \fB-I\fP options, but is blank on
+many systems).
+.TP 10
+\fB--cflags-posix\fP
+Writes to the standard output the command line options required to compile
+files that use PCRE's POSIX API wrapper library (this may include some \fB-I\fP
+options, but is blank on many systems).
+.
+.
+.SH "SEE ALSO"
+.rs
+.sp
+\fBpcre(3)\fP
+.
+.
+.SH AUTHOR
+.rs
+.sp
+This manual page was originally written by Mark Baker for the Debian GNU/Linux
+system. It has been subsequently revised as a generic PCRE man page.
+.
+.
+.SH REVISION
+.rs
+.sp
+.nf
+Last updated: 24 June 2012
+.fi
diff --git a/pcre/doc/pcre-config.txt b/pcre/doc/pcre-config.txt
new file mode 100644
index 00000000000..97a6abed62c
--- /dev/null
+++ b/pcre/doc/pcre-config.txt
@@ -0,0 +1,86 @@
+PCRE-CONFIG(1)              General Commands Manual             PCRE-CONFIG(1)
+
+
+
+NAME
+       pcre-config - program to return PCRE configuration
+
+SYNOPSIS
+
+       pcre-config [--prefix] [--exec-prefix] [--version] [--libs]
+            [--libs16] [--libs32] [--libs-cpp] [--libs-posix]
+            [--cflags] [--cflags-posix]
+
+
+DESCRIPTION
+
+       pcre-config  returns  the configuration of the installed PCRE libraries
+       and the options required to compile a program to use them. Some of  the
+       options  apply  only  to  the  8-bit,  or  16-bit, or 32-bit libraries,
+       respectively, and are not available if only one of those libraries  has
+       been built. If an unavailable option is encountered, the "usage" infor-
+       mation is output.
+
+
+OPTIONS
+
+       --prefix  Writes the directory prefix used in the PCRE installation for
+                 architecture   independent   files  (/usr  on  many  systems,
+                 /usr/local on some systems) to the standard output.
+
+       --exec-prefix
+                 Writes the directory prefix used in the PCRE installation for
+                 architecture  dependent files (normally the same as --prefix)
+                 to the standard output.
+
+       --version Writes the version number of the installed PCRE libraries  to
+                 the standard output.
+
+       --libs    Writes  to  the  standard  output  the  command  line options
+                 required to link with the 8-bit PCRE library (-lpcre on  many
+                 systems).
+
+       --libs16  Writes  to  the  standard  output  the  command  line options
+                 required to link with the 16-bit PCRE  library  (-lpcre16  on
+                 many systems).
+
+       --libs32  Writes  to  the  standard  output  the  command  line options
+                 required to link with the 32-bit PCRE  library  (-lpcre32  on
+                 many systems).
+
+       --libs-cpp
+                 Writes  to  the  standard  output  the  command  line options
+                 required to link with PCRE's C++ wrapper  library  (-lpcrecpp
+                 -lpcre on many systems).
+
+       --libs-posix
+                 Writes  to  the  standard  output  the  command  line options
+                 required to  link  with  PCRE's  POSIX  API  wrapper  library
+                 (-lpcreposix -lpcre on many systems).
+
+       --cflags  Writes  to  the  standard  output  the  command  line options
+                 required to compile files that use  PCRE  (this  may  include
+                 some -I options, but is blank on many systems).
+
+       --cflags-posix
+                 Writes  to  the  standard  output  the  command  line options
+                 required to compile files that use PCRE's POSIX  API  wrapper
+                 library  (this  may  include some -I options, but is blank on
+                 many systems).
+
+
+SEE ALSO
+
+       pcre(3)
+
+
+AUTHOR
+
+       This manual page was originally written by Mark Baker  for  the  Debian
+       GNU/Linux  system.  It  has been subsequently revised as a generic PCRE
+       man page.
+
+
+REVISION
+
+       Last updated: 24 June 2012
diff --git a/pcre/doc/pcre.3 b/pcre/doc/pcre.3
new file mode 100644
index 00000000000..c2f9164052c
--- /dev/null
+++ b/pcre/doc/pcre.3
@@ -0,0 +1,215 @@
+.TH PCRE 3 "13 May 2013" "PCRE 8.33"
+.SH NAME
+PCRE - Perl-compatible regular expressions
+.SH INTRODUCTION
+.rs
+.sp
+The PCRE library is a set of functions that implement regular expression
+pattern matching using the same syntax and semantics as Perl, with just a few
+differences. Some features that appeared in Python and PCRE before they
+appeared in Perl are also available using the Python syntax, there is some
+support for one or two .NET and Oniguruma syntax items, and there is an option
+for requesting some minor changes that give better JavaScript compatibility.
+.P
+Starting with release 8.30, it is possible to compile two separate PCRE
+libraries: the original, which supports 8-bit character strings (including
+UTF-8 strings), and a second library that supports 16-bit character strings
+(including UTF-16 strings). The build process allows either one or both to be
+built. The majority of the work to make this possible was done by Zoltan
+Herczeg.
+.P
+Starting with release 8.32 it is possible to compile a third separate PCRE
+library that supports 32-bit character strings (including UTF-32 strings). The
+build process allows any combination of the 8-, 16- and 32-bit libraries. The
+work to make this possible was done by Christian Persch.
+.P
+The three libraries contain identical sets of functions, except that the names
+in the 16-bit library start with \fBpcre16_\fP instead of \fBpcre_\fP, and the
+names in the 32-bit library start with \fBpcre32_\fP instead of \fBpcre_\fP. To
+avoid over-complication and reduce the documentation maintenance load, most of
+the documentation describes the 8-bit library, with the differences for the
+16-bit and 32-bit libraries described separately in the
+.\" HREF
+\fBpcre16\fP
+and
+.\" HREF
+\fBpcre32\fP
+.\"
+pages. References to functions or structures of the form \fIpcre[16|32]_xxx\fP
+should be read as meaning "\fIpcre_xxx\fP when using the 8-bit library,
+\fIpcre16_xxx\fP when using the 16-bit library, or \fIpcre32_xxx\fP when using
+the 32-bit library".
+.P
+The current implementation of PCRE corresponds approximately with Perl 5.12,
+including support for UTF-8/16/32 encoded strings and Unicode general category
+properties. However, UTF-8/16/32 and Unicode support has to be explicitly
+enabled; it is not the default. The Unicode tables correspond to Unicode
+release 6.2.0.
+.P
+In addition to the Perl-compatible matching function, PCRE contains an
+alternative function that matches the same compiled patterns in a different
+way. In certain circumstances, the alternative function has some advantages.
+For a discussion of the two matching algorithms, see the
+.\" HREF
+\fBpcrematching\fP
+.\"
+page.
+.P
+PCRE is written in C and released as a C library. A number of people have
+written wrappers and interfaces of various kinds. In particular, Google Inc.
+have provided a comprehensive C++ wrapper for the 8-bit library. This is now
+included as part of the PCRE distribution. The
+.\" HREF
+\fBpcrecpp\fP
+.\"
+page has details of this interface. Other people's contributions can be found
+in the \fIContrib\fP directory at the primary FTP site, which is:
+.sp
+.\" HTML 
+.\" 
+ftp://ftp.csx.cam.ac.uk/pub/software/programming/pcre
+.\"
+.P
+Details of exactly which Perl regular expression features are and are not
+supported by PCRE are given in separate documents. See the
+.\" HREF
+\fBpcrepattern\fP
+.\"
+and
+.\" HREF
+\fBpcrecompat\fP
+.\"
+pages. There is a syntax summary in the
+.\" HREF
+\fBpcresyntax\fP
+.\"
+page.
+.P
+Some features of PCRE can be included, excluded, or changed when the library is
+built. The
+.\" HREF
+\fBpcre_config()\fP
+.\"
+function makes it possible for a client to discover which features are
+available. The features themselves are described in the
+.\" HREF
+\fBpcrebuild\fP
+.\"
+page. Documentation about building PCRE for various operating systems can be
+found in the
+.\" HTML 
+.\" 
+\fBREADME\fP
+.\"
+and
+.\" HTML 
+.\" 
+\fBNON-AUTOTOOLS_BUILD\fP
+.\"
+files in the source distribution.
+.P
+The libraries contains a number of undocumented internal functions and data
+tables that are used by more than one of the exported external functions, but
+which are not intended for use by external callers. Their names all begin with
+"_pcre_" or "_pcre16_" or "_pcre32_", which hopefully will not provoke any name
+clashes. In some environments, it is possible to control which external symbols
+are exported when a shared library is built, and in these cases the
+undocumented symbols are not exported.
+.
+.
+.SH "SECURITY CONSIDERATIONS"
+.rs
+.sp
+If you are using PCRE in a non-UTF application that permits users to supply
+arbitrary patterns for compilation, you should be aware of a feature that
+allows users to turn on UTF support from within a pattern, provided that PCRE
+was built with UTF support. For example, an 8-bit pattern that begins with
+"(*UTF8)" or "(*UTF)" turns on UTF-8 mode, which interprets patterns and
+subjects as strings of UTF-8 characters instead of individual 8-bit characters.
+This causes both the pattern and any data against which it is matched to be
+checked for UTF-8 validity. If the data string is very long, such a check might
+use sufficiently many resources as to cause your application to lose
+performance.
+.P
+One way of guarding against this possibility is to use the
+\fBpcre_fullinfo()\fP function to check the compiled pattern's options for UTF.
+Alternatively, from release 8.33, you can set the PCRE_NEVER_UTF option at
+compile time. This causes an compile time error if a pattern contains a
+UTF-setting sequence.
+.P
+If your application is one that supports UTF, be aware that validity checking
+can take time. If the same data string is to be matched many times, you can use
+the PCRE_NO_UTF[8|16|32]_CHECK option for the second and subsequent matches to
+save redundant checks.
+.P
+Another way that performance can be hit is by running a pattern that has a very
+large search tree against a string that will never match. Nested unlimited
+repeats in a pattern are a common example. PCRE provides some protection
+against this: see the PCRE_EXTRA_MATCH_LIMIT feature in the
+.\" HREF
+\fBpcreapi\fP
+.\"
+page.
+.
+.
+.SH "USER DOCUMENTATION"
+.rs
+.sp
+The user documentation for PCRE comprises a number of different sections. In
+the "man" format, each of these is a separate "man page". In the HTML format,
+each is a separate page, linked from the index page. In the plain text format,
+all the sections, except the \fBpcredemo\fP section, are concatenated, for ease
+of searching. The sections are as follows:
+.sp
+  pcre              this document
+  pcre-config       show PCRE installation configuration information
+  pcre16            details of the 16-bit library
+  pcre32            details of the 32-bit library
+  pcreapi           details of PCRE's native C API
+  pcrebuild         building PCRE
+  pcrecallout       details of the callout feature
+  pcrecompat        discussion of Perl compatibility
+  pcrecpp           details of the C++ wrapper for the 8-bit library
+  pcredemo          a demonstration C program that uses PCRE
+  pcregrep          description of the \fBpcregrep\fP command (8-bit only)
+  pcrejit           discussion of the just-in-time optimization support
+  pcrelimits        details of size and other limits
+  pcrematching      discussion of the two matching algorithms
+  pcrepartial       details of the partial matching facility
+.\" JOIN
+  pcrepattern       syntax and semantics of supported
+                      regular expressions
+  pcreperform       discussion of performance issues
+  pcreposix         the POSIX-compatible C API for the 8-bit library
+  pcreprecompile    details of saving and re-using precompiled patterns
+  pcresample        discussion of the pcredemo program
+  pcrestack         discussion of stack usage
+  pcresyntax        quick syntax reference
+  pcretest          description of the \fBpcretest\fP testing command
+  pcreunicode       discussion of Unicode and UTF-8/16/32 support
+.sp
+In addition, in the "man" and HTML formats, there is a short page for each
+C library function, listing its arguments and results.
+.
+.
+.SH AUTHOR
+.rs
+.sp
+.nf
+Philip Hazel
+University Computing Service
+Cambridge CB2 3QH, England.
+.fi
+.P
+Putting an actual email address here seems to have been a spam magnet, so I've
+taken it away. If you want to email me, use my two initials, followed by the
+two digits 10, at the domain cam.ac.uk.
+.
+.
+.SH REVISION
+.rs
+.sp
+.nf
+Last updated: 13 May 2013
+Copyright (c) 1997-2013 University of Cambridge.
+.fi
diff --git a/pcre/doc/pcre.txt b/pcre/doc/pcre.txt
new file mode 100644
index 00000000000..193203e8a32
--- /dev/null
+++ b/pcre/doc/pcre.txt
@@ -0,0 +1,10181 @@
+-----------------------------------------------------------------------------
+This file contains a concatenation of the PCRE man pages, converted to plain
+text format for ease of searching with a text editor, or for use on systems
+that do not have a man page processor. The small individual files that give
+synopses of each function in the library have not been included. Neither has
+the pcredemo program. There are separate text files for the pcregrep and
+pcretest commands.
+-----------------------------------------------------------------------------
+
+
+PCRE(3)                    Library Functions Manual                    PCRE(3)
+
+
+
+NAME
+       PCRE - Perl-compatible regular expressions
+
+INTRODUCTION
+
+       The  PCRE  library is a set of functions that implement regular expres-
+       sion pattern matching using the same syntax and semantics as Perl, with
+       just  a few differences. Some features that appeared in Python and PCRE
+       before they appeared in Perl are also available using the  Python  syn-
+       tax,  there  is  some  support for one or two .NET and Oniguruma syntax
+       items, and there is an option for requesting some  minor  changes  that
+       give better JavaScript compatibility.
+
+       Starting with release 8.30, it is possible to compile two separate PCRE
+       libraries:  the  original,  which  supports  8-bit  character   strings
+       (including  UTF-8  strings),  and a second library that supports 16-bit
+       character strings (including UTF-16 strings). The build process  allows
+       either  one  or both to be built. The majority of the work to make this
+       possible was done by Zoltan Herczeg.
+
+       Starting with release 8.32 it is possible to compile a  third  separate
+       PCRE  library  that supports 32-bit character strings (including UTF-32
+       strings). The build process allows any combination of the 8-,  16-  and
+       32-bit  libraries. The work to make this possible was done by Christian
+       Persch.
+
+       The three libraries contain identical sets of  functions,  except  that
+       the  names  in  the 16-bit library start with pcre16_ instead of pcre_,
+       and the names in the 32-bit  library  start  with  pcre32_  instead  of
+       pcre_.  To avoid over-complication and reduce the documentation mainte-
+       nance load, most of the documentation describes the 8-bit library, with
+       the  differences  for  the  16-bit and 32-bit libraries described sepa-
+       rately in the pcre16 and  pcre32  pages.  References  to  functions  or
+       structures  of  the  form  pcre[16|32]_xxx  should  be  read as meaning
+       "pcre_xxx when using the  8-bit  library,  pcre16_xxx  when  using  the
+       16-bit library, or pcre32_xxx when using the 32-bit library".
+
+       The  current implementation of PCRE corresponds approximately with Perl
+       5.12, including support for UTF-8/16/32  encoded  strings  and  Unicode
+       general  category  properties. However, UTF-8/16/32 and Unicode support
+       has to be explicitly enabled; it is not the default. The Unicode tables
+       correspond to Unicode release 6.2.0.
+
+       In  addition to the Perl-compatible matching function, PCRE contains an
+       alternative function that matches the same compiled patterns in a  dif-
+       ferent way. In certain circumstances, the alternative function has some
+       advantages.  For a discussion of the two matching algorithms,  see  the
+       pcrematching page.
+
+       PCRE  is  written  in C and released as a C library. A number of people
+       have written wrappers and interfaces of various kinds.  In  particular,
+       Google  Inc.   have  provided a comprehensive C++ wrapper for the 8-bit
+       library. This is now included as part of  the  PCRE  distribution.  The
+       pcrecpp  page  has  details of this interface. Other people's contribu-
+       tions can be found in the Contrib directory at the  primary  FTP  site,
+       which is:
+
+       ftp://ftp.csx.cam.ac.uk/pub/software/programming/pcre
+
+       Details  of  exactly which Perl regular expression features are and are
+       not supported by PCRE are given in separate documents. See the pcrepat-
+       tern  and pcrecompat pages. There is a syntax summary in the pcresyntax
+       page.
+
+       Some features of PCRE can be included, excluded, or  changed  when  the
+       library  is  built.  The pcre_config() function makes it possible for a
+       client to discover which features are  available.  The  features  them-
+       selves  are described in the pcrebuild page. Documentation about build-
+       ing PCRE for various operating systems can be found in the  README  and
+       NON-AUTOTOOLS_BUILD files in the source distribution.
+
+       The  libraries contains a number of undocumented internal functions and
+       data tables that are used by more than one  of  the  exported  external
+       functions,  but  which  are  not  intended for use by external callers.
+       Their names all begin with "_pcre_" or "_pcre16_" or "_pcre32_",  which
+       hopefully  will  not provoke any name clashes. In some environments, it
+       is possible to control which  external  symbols  are  exported  when  a
+       shared  library  is  built, and in these cases the undocumented symbols
+       are not exported.
+
+
+SECURITY CONSIDERATIONS
+
+       If you are using PCRE in a non-UTF application that  permits  users  to
+       supply  arbitrary  patterns  for  compilation, you should be aware of a
+       feature that allows users to turn on UTF support from within a pattern,
+       provided  that  PCRE  was built with UTF support. For example, an 8-bit
+       pattern that begins with "(*UTF8)" or "(*UTF)"  turns  on  UTF-8  mode,
+       which  interprets  patterns and subjects as strings of UTF-8 characters
+       instead of individual 8-bit characters.  This causes both  the  pattern
+       and any data against which it is matched to be checked for UTF-8 valid-
+       ity. If the data string is very long, such a  check  might  use  suffi-
+       ciently  many  resources  as  to cause your application to lose perfor-
+       mance.
+
+       One  way  of  guarding  against  this  possibility  is   to   use   the
+       pcre_fullinfo()  function  to  check the compiled pattern's options for
+       UTF.  Alternatively, from release 8.33, you can set the  PCRE_NEVER_UTF
+       option  at compile time. This causes an compile time error if a pattern
+       contains a UTF-setting sequence.
+
+       If your application is one that supports UTF, be  aware  that  validity
+       checking  can  take time. If the same data string is to be matched many
+       times, you can use the PCRE_NO_UTF[8|16|32]_CHECK option for the second
+       and subsequent matches to save redundant checks.
+
+       Another  way  that  performance can be hit is by running a pattern that
+       has a very large search tree against a string that  will  never  match.
+       Nested  unlimited  repeats in a pattern are a common example. PCRE pro-
+       vides some protection against this: see the PCRE_EXTRA_MATCH_LIMIT fea-
+       ture in the pcreapi page.
+
+
+USER DOCUMENTATION
+
+       The  user  documentation  for PCRE comprises a number of different sec-
+       tions. In the "man" format, each of these is a separate "man page".  In
+       the  HTML  format, each is a separate page, linked from the index page.
+       In the plain text format, all the sections, except  the  pcredemo  sec-
+       tion, are concatenated, for ease of searching. The sections are as fol-
+       lows:
+
+         pcre              this document
+         pcre-config       show PCRE installation configuration information
+         pcre16            details of the 16-bit library
+         pcre32            details of the 32-bit library
+         pcreapi           details of PCRE's native C API
+         pcrebuild         building PCRE
+         pcrecallout       details of the callout feature
+         pcrecompat        discussion of Perl compatibility
+         pcrecpp           details of the C++ wrapper for the 8-bit library
+         pcredemo          a demonstration C program that uses PCRE
+         pcregrep          description of the pcregrep command (8-bit only)
+         pcrejit           discussion of the just-in-time optimization support
+         pcrelimits        details of size and other limits
+         pcrematching      discussion of the two matching algorithms
+         pcrepartial       details of the partial matching facility
+         pcrepattern       syntax and semantics of supported
+                             regular expressions
+         pcreperform       discussion of performance issues
+         pcreposix         the POSIX-compatible C API for the 8-bit library
+         pcreprecompile    details of saving and re-using precompiled patterns
+         pcresample        discussion of the pcredemo program
+         pcrestack         discussion of stack usage
+         pcresyntax        quick syntax reference
+         pcretest          description of the pcretest testing command
+         pcreunicode       discussion of Unicode and UTF-8/16/32 support
+
+       In addition, in the "man" and HTML formats, there is a short  page  for
+       each C library function, listing its arguments and results.
+
+
+AUTHOR
+
+       Philip Hazel
+       University Computing Service
+       Cambridge CB2 3QH, England.
+
+       Putting  an actual email address here seems to have been a spam magnet,
+       so I've taken it away. If you want to email me, use  my  two  initials,
+       followed by the two digits 10, at the domain cam.ac.uk.
+
+
+REVISION
+
+       Last updated: 13 May 2013
+       Copyright (c) 1997-2013 University of Cambridge.
+------------------------------------------------------------------------------
+
+
+PCRE(3)                    Library Functions Manual                    PCRE(3)
+
+
+
+NAME
+       PCRE - Perl-compatible regular expressions
+
+       #include 
+
+
+PCRE 16-BIT API BASIC FUNCTIONS
+
+       pcre16 *pcre16_compile(PCRE_SPTR16 pattern, int options,
+            const char **errptr, int *erroffset,
+            const unsigned char *tableptr);
+
+       pcre16 *pcre16_compile2(PCRE_SPTR16 pattern, int options,
+            int *errorcodeptr,
+            const char **errptr, int *erroffset,
+            const unsigned char *tableptr);
+
+       pcre16_extra *pcre16_study(const pcre16 *code, int options,
+            const char **errptr);
+
+       void pcre16_free_study(pcre16_extra *extra);
+
+       int pcre16_exec(const pcre16 *code, const pcre16_extra *extra,
+            PCRE_SPTR16 subject, int length, int startoffset,
+            int options, int *ovector, int ovecsize);
+
+       int pcre16_dfa_exec(const pcre16 *code, const pcre16_extra *extra,
+            PCRE_SPTR16 subject, int length, int startoffset,
+            int options, int *ovector, int ovecsize,
+            int *workspace, int wscount);
+
+
+PCRE 16-BIT API STRING EXTRACTION FUNCTIONS
+
+       int pcre16_copy_named_substring(const pcre16 *code,
+            PCRE_SPTR16 subject, int *ovector,
+            int stringcount, PCRE_SPTR16 stringname,
+            PCRE_UCHAR16 *buffer, int buffersize);
+
+       int pcre16_copy_substring(PCRE_SPTR16 subject, int *ovector,
+            int stringcount, int stringnumber, PCRE_UCHAR16 *buffer,
+            int buffersize);
+
+       int pcre16_get_named_substring(const pcre16 *code,
+            PCRE_SPTR16 subject, int *ovector,
+            int stringcount, PCRE_SPTR16 stringname,
+            PCRE_SPTR16 *stringptr);
+
+       int pcre16_get_stringnumber(const pcre16 *code,
+            PCRE_SPTR16 name);
+
+       int pcre16_get_stringtable_entries(const pcre16 *code,
+            PCRE_SPTR16 name, PCRE_UCHAR16 **first, PCRE_UCHAR16 **last);
+
+       int pcre16_get_substring(PCRE_SPTR16 subject, int *ovector,
+            int stringcount, int stringnumber,
+            PCRE_SPTR16 *stringptr);
+
+       int pcre16_get_substring_list(PCRE_SPTR16 subject,
+            int *ovector, int stringcount, PCRE_SPTR16 **listptr);
+
+       void pcre16_free_substring(PCRE_SPTR16 stringptr);
+
+       void pcre16_free_substring_list(PCRE_SPTR16 *stringptr);
+
+
+PCRE 16-BIT API AUXILIARY FUNCTIONS
+
+       pcre16_jit_stack *pcre16_jit_stack_alloc(int startsize, int maxsize);
+
+       void pcre16_jit_stack_free(pcre16_jit_stack *stack);
+
+       void pcre16_assign_jit_stack(pcre16_extra *extra,
+            pcre16_jit_callback callback, void *data);
+
+       const unsigned char *pcre16_maketables(void);
+
+       int pcre16_fullinfo(const pcre16 *code, const pcre16_extra *extra,
+            int what, void *where);
+
+       int pcre16_refcount(pcre16 *code, int adjust);
+
+       int pcre16_config(int what, void *where);
+
+       const char *pcre16_version(void);
+
+       int pcre16_pattern_to_host_byte_order(pcre16 *code,
+            pcre16_extra *extra, const unsigned char *tables);
+
+
+PCRE 16-BIT API INDIRECTED FUNCTIONS
+
+       void *(*pcre16_malloc)(size_t);
+
+       void (*pcre16_free)(void *);
+
+       void *(*pcre16_stack_malloc)(size_t);
+
+       void (*pcre16_stack_free)(void *);
+
+       int (*pcre16_callout)(pcre16_callout_block *);
+
+
+PCRE 16-BIT API 16-BIT-ONLY FUNCTION
+
+       int pcre16_utf16_to_host_byte_order(PCRE_UCHAR16 *output,
+            PCRE_SPTR16 input, int length, int *byte_order,
+            int keep_boms);
+
+
+THE PCRE 16-BIT LIBRARY
+
+       Starting  with  release  8.30, it is possible to compile a PCRE library
+       that supports 16-bit character strings, including  UTF-16  strings,  as
+       well  as  or instead of the original 8-bit library. The majority of the
+       work to make  this  possible  was  done  by  Zoltan  Herczeg.  The  two
+       libraries contain identical sets of functions, used in exactly the same
+       way. Only the names of the functions and the data types of their  argu-
+       ments  and results are different. To avoid over-complication and reduce
+       the documentation maintenance load,  most  of  the  PCRE  documentation
+       describes  the  8-bit  library,  with only occasional references to the
+       16-bit library. This page describes what is different when you use  the
+       16-bit library.
+
+       WARNING:  A  single  application can be linked with both libraries, but
+       you must take care when processing any particular pattern to use  func-
+       tions  from  just one library. For example, if you want to study a pat-
+       tern that was compiled with  pcre16_compile(),  you  must  do  so  with
+       pcre16_study(), not pcre_study(), and you must free the study data with
+       pcre16_free_study().
+
+
+THE HEADER FILE
+
+       There is only one header file, pcre.h. It contains prototypes  for  all
+       the functions in all libraries, as well as definitions of flags, struc-
+       tures, error codes, etc.
+
+
+THE LIBRARY NAME
+
+       In Unix-like systems, the 16-bit library is called libpcre16,  and  can
+       normally  be  accesss  by adding -lpcre16 to the command for linking an
+       application that uses PCRE.
+
+
+STRING TYPES
+
+       In the 8-bit library, strings are passed to PCRE library  functions  as
+       vectors  of  bytes  with  the  C  type "char *". In the 16-bit library,
+       strings are passed as vectors of unsigned 16-bit quantities. The  macro
+       PCRE_UCHAR16  specifies  an  appropriate  data type, and PCRE_SPTR16 is
+       defined as "const PCRE_UCHAR16 *". In very  many  environments,  "short
+       int" is a 16-bit data type. When PCRE is built, it defines PCRE_UCHAR16
+       as "unsigned short int", but checks that it really  is  a  16-bit  data
+       type.  If  it is not, the build fails with an error message telling the
+       maintainer to modify the definition appropriately.
+
+
+STRUCTURE TYPES
+
+       The types of the opaque structures that are used  for  compiled  16-bit
+       patterns  and  JIT stacks are pcre16 and pcre16_jit_stack respectively.
+       The  type  of  the  user-accessible  structure  that  is  returned   by
+       pcre16_study()  is  pcre16_extra, and the type of the structure that is
+       used for passing data to a callout  function  is  pcre16_callout_block.
+       These structures contain the same fields, with the same names, as their
+       8-bit counterparts. The only difference is that pointers  to  character
+       strings are 16-bit instead of 8-bit types.
+
+
+16-BIT FUNCTIONS
+
+       For  every function in the 8-bit library there is a corresponding func-
+       tion in the 16-bit library with a name that starts with pcre16_ instead
+       of  pcre_.  The  prototypes are listed above. In addition, there is one
+       extra function, pcre16_utf16_to_host_byte_order(). This  is  a  utility
+       function  that converts a UTF-16 character string to host byte order if
+       necessary. The other 16-bit  functions  expect  the  strings  they  are
+       passed to be in host byte order.
+
+       The input and output arguments of pcre16_utf16_to_host_byte_order() may
+       point to the same address, that is, conversion in place  is  supported.
+       The output buffer must be at least as long as the input.
+
+       The  length  argument  specifies the number of 16-bit data units in the
+       input string; a negative value specifies a zero-terminated string.
+
+       If byte_order is NULL, it is assumed that the string starts off in host
+       byte  order. This may be changed by byte-order marks (BOMs) anywhere in
+       the string (commonly as the first character).
+
+       If byte_order is not NULL, a non-zero value of the integer to which  it
+       points  means  that  the input starts off in host byte order, otherwise
+       the opposite order is assumed. Again, BOMs in  the  string  can  change
+       this. The final byte order is passed back at the end of processing.
+
+       If  keep_boms  is  not  zero,  byte-order  mark characters (0xfeff) are
+       copied into the output string. Otherwise they are discarded.
+
+       The result of the function is the number of 16-bit  units  placed  into
+       the  output  buffer,  including  the  zero terminator if the string was
+       zero-terminated.
+
+
+SUBJECT STRING OFFSETS
+
+       The lengths and starting offsets of subject strings must  be  specified
+       in  16-bit  data units, and the offsets within subject strings that are
+       returned by the matching functions are in also 16-bit units rather than
+       bytes.
+
+
+NAMED SUBPATTERNS
+
+       The  name-to-number translation table that is maintained for named sub-
+       patterns uses 16-bit characters.  The  pcre16_get_stringtable_entries()
+       function returns the length of each entry in the table as the number of
+       16-bit data units.
+
+
+OPTION NAMES
+
+       There   are   two   new   general   option   names,   PCRE_UTF16    and
+       PCRE_NO_UTF16_CHECK,     which     correspond    to    PCRE_UTF8    and
+       PCRE_NO_UTF8_CHECK in the 8-bit library. In  fact,  these  new  options
+       define  the  same bits in the options word. There is a discussion about
+       the validity of UTF-16 strings in the pcreunicode page.
+
+       For the pcre16_config() function there is an  option  PCRE_CONFIG_UTF16
+       that  returns  1  if UTF-16 support is configured, otherwise 0. If this
+       option  is  given  to  pcre_config()  or  pcre32_config(),  or  if  the
+       PCRE_CONFIG_UTF8  or  PCRE_CONFIG_UTF32  option is given to pcre16_con-
+       fig(), the result is the PCRE_ERROR_BADOPTION error.
+
+
+CHARACTER CODES
+
+       In 16-bit mode, when  PCRE_UTF16  is  not  set,  character  values  are
+       treated in the same way as in 8-bit, non UTF-8 mode, except, of course,
+       that they can range from 0 to 0xffff instead of 0  to  0xff.  Character
+       types  for characters less than 0xff can therefore be influenced by the
+       locale in the same way as before.  Characters greater  than  0xff  have
+       only one case, and no "type" (such as letter or digit).
+
+       In  UTF-16  mode,  the  character  code  is  Unicode, in the range 0 to
+       0x10ffff, with the exception of values in the range  0xd800  to  0xdfff
+       because  those  are "surrogate" values that are used in pairs to encode
+       values greater than 0xffff.
+
+       A UTF-16 string can indicate its endianness by special code knows as  a
+       byte-order mark (BOM). The PCRE functions do not handle this, expecting
+       strings  to  be  in  host  byte  order.  A  utility   function   called
+       pcre16_utf16_to_host_byte_order()  is  provided  to help with this (see
+       above).
+
+
+ERROR NAMES
+
+       The errors PCRE_ERROR_BADUTF16_OFFSET and PCRE_ERROR_SHORTUTF16  corre-
+       spond  to  their  8-bit  counterparts.  The error PCRE_ERROR_BADMODE is
+       given when a compiled pattern is passed to a  function  that  processes
+       patterns  in  the  other  mode, for example, if a pattern compiled with
+       pcre_compile() is passed to pcre16_exec().
+
+       There are new error codes whose names  begin  with  PCRE_UTF16_ERR  for
+       invalid  UTF-16  strings,  corresponding to the PCRE_UTF8_ERR codes for
+       UTF-8 strings that are described in the section entitled "Reason  codes
+       for  invalid UTF-8 strings" in the main pcreapi page. The UTF-16 errors
+       are:
+
+         PCRE_UTF16_ERR1  Missing low surrogate at end of string
+         PCRE_UTF16_ERR2  Invalid low surrogate follows high surrogate
+         PCRE_UTF16_ERR3  Isolated low surrogate
+         PCRE_UTF16_ERR4  Non-character
+
+
+ERROR TEXTS
+
+       If there is an error while compiling a pattern, the error text that  is
+       passed  back by pcre16_compile() or pcre16_compile2() is still an 8-bit
+       character string, zero-terminated.
+
+
+CALLOUTS
+
+       The subject and mark fields in the callout block that is  passed  to  a
+       callout function point to 16-bit vectors.
+
+
+TESTING
+
+       The  pcretest  program continues to operate with 8-bit input and output
+       files, but it can be used for testing the 16-bit library. If it is  run
+       with the command line option -16, patterns and subject strings are con-
+       verted from 8-bit to 16-bit before being passed to PCRE, and the 16-bit
+       library  functions  are used instead of the 8-bit ones. Returned 16-bit
+       strings are converted to 8-bit for output. If both the  8-bit  and  the
+       32-bit libraries were not compiled, pcretest defaults to 16-bit and the
+       -16 option is ignored.
+
+       When PCRE is being built, the RunTest script that is  called  by  "make
+       check"  uses  the  pcretest  -C  option to discover which of the 8-bit,
+       16-bit and 32-bit libraries has been built, and runs the  tests  appro-
+       priately.
+
+
+NOT SUPPORTED IN 16-BIT MODE
+
+       Not all the features of the 8-bit library are available with the 16-bit
+       library. The C++ and POSIX wrapper functions  support  only  the  8-bit
+       library, and the pcregrep program is at present 8-bit only.
+
+
+AUTHOR
+
+       Philip Hazel
+       University Computing Service
+       Cambridge CB2 3QH, England.
+
+
+REVISION
+
+       Last updated: 12 May 2013
+       Copyright (c) 1997-2013 University of Cambridge.
+------------------------------------------------------------------------------
+
+
+PCRE(3)                    Library Functions Manual                    PCRE(3)
+
+
+
+NAME
+       PCRE - Perl-compatible regular expressions
+
+       #include 
+
+
+PCRE 32-BIT API BASIC FUNCTIONS
+
+       pcre32 *pcre32_compile(PCRE_SPTR32 pattern, int options,
+            const char **errptr, int *erroffset,
+            const unsigned char *tableptr);
+
+       pcre32 *pcre32_compile2(PCRE_SPTR32 pattern, int options,
+            int *errorcodeptr,
+            const char **errptr, int *erroffset,
+            const unsigned char *tableptr);
+
+       pcre32_extra *pcre32_study(const pcre32 *code, int options,
+            const char **errptr);
+
+       void pcre32_free_study(pcre32_extra *extra);
+
+       int pcre32_exec(const pcre32 *code, const pcre32_extra *extra,
+            PCRE_SPTR32 subject, int length, int startoffset,
+            int options, int *ovector, int ovecsize);
+
+       int pcre32_dfa_exec(const pcre32 *code, const pcre32_extra *extra,
+            PCRE_SPTR32 subject, int length, int startoffset,
+            int options, int *ovector, int ovecsize,
+            int *workspace, int wscount);
+
+
+PCRE 32-BIT API STRING EXTRACTION FUNCTIONS
+
+       int pcre32_copy_named_substring(const pcre32 *code,
+            PCRE_SPTR32 subject, int *ovector,
+            int stringcount, PCRE_SPTR32 stringname,
+            PCRE_UCHAR32 *buffer, int buffersize);
+
+       int pcre32_copy_substring(PCRE_SPTR32 subject, int *ovector,
+            int stringcount, int stringnumber, PCRE_UCHAR32 *buffer,
+            int buffersize);
+
+       int pcre32_get_named_substring(const pcre32 *code,
+            PCRE_SPTR32 subject, int *ovector,
+            int stringcount, PCRE_SPTR32 stringname,
+            PCRE_SPTR32 *stringptr);
+
+       int pcre32_get_stringnumber(const pcre32 *code,
+            PCRE_SPTR32 name);
+
+       int pcre32_get_stringtable_entries(const pcre32 *code,
+            PCRE_SPTR32 name, PCRE_UCHAR32 **first, PCRE_UCHAR32 **last);
+
+       int pcre32_get_substring(PCRE_SPTR32 subject, int *ovector,
+            int stringcount, int stringnumber,
+            PCRE_SPTR32 *stringptr);
+
+       int pcre32_get_substring_list(PCRE_SPTR32 subject,
+            int *ovector, int stringcount, PCRE_SPTR32 **listptr);
+
+       void pcre32_free_substring(PCRE_SPTR32 stringptr);
+
+       void pcre32_free_substring_list(PCRE_SPTR32 *stringptr);
+
+
+PCRE 32-BIT API AUXILIARY FUNCTIONS
+
+       pcre32_jit_stack *pcre32_jit_stack_alloc(int startsize, int maxsize);
+
+       void pcre32_jit_stack_free(pcre32_jit_stack *stack);
+
+       void pcre32_assign_jit_stack(pcre32_extra *extra,
+            pcre32_jit_callback callback, void *data);
+
+       const unsigned char *pcre32_maketables(void);
+
+       int pcre32_fullinfo(const pcre32 *code, const pcre32_extra *extra,
+            int what, void *where);
+
+       int pcre32_refcount(pcre32 *code, int adjust);
+
+       int pcre32_config(int what, void *where);
+
+       const char *pcre32_version(void);
+
+       int pcre32_pattern_to_host_byte_order(pcre32 *code,
+            pcre32_extra *extra, const unsigned char *tables);
+
+
+PCRE 32-BIT API INDIRECTED FUNCTIONS
+
+       void *(*pcre32_malloc)(size_t);
+
+       void (*pcre32_free)(void *);
+
+       void *(*pcre32_stack_malloc)(size_t);
+
+       void (*pcre32_stack_free)(void *);
+
+       int (*pcre32_callout)(pcre32_callout_block *);
+
+
+PCRE 32-BIT API 32-BIT-ONLY FUNCTION
+
+       int pcre32_utf32_to_host_byte_order(PCRE_UCHAR32 *output,
+            PCRE_SPTR32 input, int length, int *byte_order,
+            int keep_boms);
+
+
+THE PCRE 32-BIT LIBRARY
+
+       Starting  with  release  8.32, it is possible to compile a PCRE library
+       that supports 32-bit character strings, including  UTF-32  strings,  as
+       well as or instead of the original 8-bit library. This work was done by
+       Christian Persch, based on the work done  by  Zoltan  Herczeg  for  the
+       16-bit  library.  All  three  libraries contain identical sets of func-
+       tions, used in exactly the same way.  Only the names of  the  functions
+       and  the  data  types  of their arguments and results are different. To
+       avoid over-complication and reduce the documentation maintenance  load,
+       most  of  the PCRE documentation describes the 8-bit library, with only
+       occasional references to the 16-bit and  32-bit  libraries.  This  page
+       describes what is different when you use the 32-bit library.
+
+       WARNING:  A  single  application  can  be linked with all or any of the
+       three libraries, but you must take care when processing any  particular
+       pattern  to  use  functions  from just one library. For example, if you
+       want to study a pattern that was compiled  with  pcre32_compile(),  you
+       must do so with pcre32_study(), not pcre_study(), and you must free the
+       study data with pcre32_free_study().
+
+
+THE HEADER FILE
+
+       There is only one header file, pcre.h. It contains prototypes  for  all
+       the functions in all libraries, as well as definitions of flags, struc-
+       tures, error codes, etc.
+
+
+THE LIBRARY NAME
+
+       In Unix-like systems, the 32-bit library is called libpcre32,  and  can
+       normally  be  accesss  by adding -lpcre32 to the command for linking an
+       application that uses PCRE.
+
+
+STRING TYPES
+
+       In the 8-bit library, strings are passed to PCRE library  functions  as
+       vectors  of  bytes  with  the  C  type "char *". In the 32-bit library,
+       strings are passed as vectors of unsigned 32-bit quantities. The  macro
+       PCRE_UCHAR32  specifies  an  appropriate  data type, and PCRE_SPTR32 is
+       defined as "const PCRE_UCHAR32 *". In very many environments, "unsigned
+       int" is a 32-bit data type. When PCRE is built, it defines PCRE_UCHAR32
+       as "unsigned int", but checks that it really is a 32-bit data type.  If
+       it is not, the build fails with an error message telling the maintainer
+       to modify the definition appropriately.
+
+
+STRUCTURE TYPES
+
+       The types of the opaque structures that are used  for  compiled  32-bit
+       patterns  and  JIT stacks are pcre32 and pcre32_jit_stack respectively.
+       The  type  of  the  user-accessible  structure  that  is  returned   by
+       pcre32_study()  is  pcre32_extra, and the type of the structure that is
+       used for passing data to a callout  function  is  pcre32_callout_block.
+       These structures contain the same fields, with the same names, as their
+       8-bit counterparts. The only difference is that pointers  to  character
+       strings are 32-bit instead of 8-bit types.
+
+
+32-BIT FUNCTIONS
+
+       For  every function in the 8-bit library there is a corresponding func-
+       tion in the 32-bit library with a name that starts with pcre32_ instead
+       of  pcre_.  The  prototypes are listed above. In addition, there is one
+       extra function, pcre32_utf32_to_host_byte_order(). This  is  a  utility
+       function  that converts a UTF-32 character string to host byte order if
+       necessary. The other 32-bit  functions  expect  the  strings  they  are
+       passed to be in host byte order.
+
+       The input and output arguments of pcre32_utf32_to_host_byte_order() may
+       point to the same address, that is, conversion in place  is  supported.
+       The output buffer must be at least as long as the input.
+
+       The  length  argument  specifies the number of 32-bit data units in the
+       input string; a negative value specifies a zero-terminated string.
+
+       If byte_order is NULL, it is assumed that the string starts off in host
+       byte  order. This may be changed by byte-order marks (BOMs) anywhere in
+       the string (commonly as the first character).
+
+       If byte_order is not NULL, a non-zero value of the integer to which  it
+       points  means  that  the input starts off in host byte order, otherwise
+       the opposite order is assumed. Again, BOMs in  the  string  can  change
+       this. The final byte order is passed back at the end of processing.
+
+       If  keep_boms  is  not  zero,  byte-order  mark characters (0xfeff) are
+       copied into the output string. Otherwise they are discarded.
+
+       The result of the function is the number of 32-bit  units  placed  into
+       the  output  buffer,  including  the  zero terminator if the string was
+       zero-terminated.
+
+
+SUBJECT STRING OFFSETS
+
+       The lengths and starting offsets of subject strings must  be  specified
+       in  32-bit  data units, and the offsets within subject strings that are
+       returned by the matching functions are in also 32-bit units rather than
+       bytes.
+
+
+NAMED SUBPATTERNS
+
+       The  name-to-number translation table that is maintained for named sub-
+       patterns uses 32-bit characters.  The  pcre32_get_stringtable_entries()
+       function returns the length of each entry in the table as the number of
+       32-bit data units.
+
+
+OPTION NAMES
+
+       There   are   two   new   general   option   names,   PCRE_UTF32    and
+       PCRE_NO_UTF32_CHECK,     which     correspond    to    PCRE_UTF8    and
+       PCRE_NO_UTF8_CHECK in the 8-bit library. In  fact,  these  new  options
+       define  the  same bits in the options word. There is a discussion about
+       the validity of UTF-32 strings in the pcreunicode page.
+
+       For the pcre32_config() function there is an  option  PCRE_CONFIG_UTF32
+       that  returns  1  if UTF-32 support is configured, otherwise 0. If this
+       option  is  given  to  pcre_config()  or  pcre16_config(),  or  if  the
+       PCRE_CONFIG_UTF8  or  PCRE_CONFIG_UTF16  option is given to pcre32_con-
+       fig(), the result is the PCRE_ERROR_BADOPTION error.
+
+
+CHARACTER CODES
+
+       In 32-bit mode, when  PCRE_UTF32  is  not  set,  character  values  are
+       treated in the same way as in 8-bit, non UTF-8 mode, except, of course,
+       that they can range from 0 to 0x7fffffff instead of 0 to 0xff.  Charac-
+       ter  types for characters less than 0xff can therefore be influenced by
+       the locale in the same way as before.   Characters  greater  than  0xff
+       have only one case, and no "type" (such as letter or digit).
+
+       In  UTF-32  mode,  the  character  code  is  Unicode, in the range 0 to
+       0x10ffff, with the exception of values in the range  0xd800  to  0xdfff
+       because those are "surrogate" values that are ill-formed in UTF-32.
+
+       A  UTF-32 string can indicate its endianness by special code knows as a
+       byte-order mark (BOM). The PCRE functions do not handle this, expecting
+       strings   to   be  in  host  byte  order.  A  utility  function  called
+       pcre32_utf32_to_host_byte_order() is provided to help  with  this  (see
+       above).
+
+
+ERROR NAMES
+
+       The  error  PCRE_ERROR_BADUTF32  corresponds  to its 8-bit counterpart.
+       The error PCRE_ERROR_BADMODE is given when a compiled pattern is passed
+       to  a  function that processes patterns in the other mode, for example,
+       if a pattern compiled with pcre_compile() is passed to pcre32_exec().
+
+       There are new error codes whose names  begin  with  PCRE_UTF32_ERR  for
+       invalid  UTF-32  strings,  corresponding to the PCRE_UTF8_ERR codes for
+       UTF-8 strings that are described in the section entitled "Reason  codes
+       for  invalid UTF-8 strings" in the main pcreapi page. The UTF-32 errors
+       are:
+
+         PCRE_UTF32_ERR1  Surrogate character (range from 0xd800 to 0xdfff)
+         PCRE_UTF32_ERR2  Non-character
+         PCRE_UTF32_ERR3  Character > 0x10ffff
+
+
+ERROR TEXTS
+
+       If there is an error while compiling a pattern, the error text that  is
+       passed  back by pcre32_compile() or pcre32_compile2() is still an 8-bit
+       character string, zero-terminated.
+
+
+CALLOUTS
+
+       The subject and mark fields in the callout block that is  passed  to  a
+       callout function point to 32-bit vectors.
+
+
+TESTING
+
+       The  pcretest  program continues to operate with 8-bit input and output
+       files, but it can be used for testing the 32-bit library. If it is  run
+       with the command line option -32, patterns and subject strings are con-
+       verted from 8-bit to 32-bit before being passed to PCRE, and the 32-bit
+       library  functions  are used instead of the 8-bit ones. Returned 32-bit
+       strings are converted to 8-bit for output. If both the  8-bit  and  the
+       16-bit libraries were not compiled, pcretest defaults to 32-bit and the
+       -32 option is ignored.
+
+       When PCRE is being built, the RunTest script that is  called  by  "make
+       check"  uses  the  pcretest  -C  option to discover which of the 8-bit,
+       16-bit and 32-bit libraries has been built, and runs the  tests  appro-
+       priately.
+
+
+NOT SUPPORTED IN 32-BIT MODE
+
+       Not all the features of the 8-bit library are available with the 32-bit
+       library. The C++ and POSIX wrapper functions  support  only  the  8-bit
+       library, and the pcregrep program is at present 8-bit only.
+
+
+AUTHOR
+
+       Philip Hazel
+       University Computing Service
+       Cambridge CB2 3QH, England.
+
+
+REVISION
+
+       Last updated: 12 May 2013
+       Copyright (c) 1997-2013 University of Cambridge.
+------------------------------------------------------------------------------
+
+
+PCREBUILD(3)               Library Functions Manual               PCREBUILD(3)
+
+
+
+NAME
+       PCRE - Perl-compatible regular expressions
+
+BUILDING PCRE
+
+       PCRE  is  distributed with a configure script that can be used to build
+       the library in Unix-like environments using the applications  known  as
+       Autotools.   Also  in  the  distribution  are files to support building
+       using CMake instead of configure. The text file README contains general
+       information  about  building  with Autotools (some of which is repeated
+       below), and also has some comments about building on various  operating
+       systems.  There  is  a lot more information about building PCRE without
+       using Autotools (including information about using CMake  and  building
+       "by  hand")  in  the  text file called NON-AUTOTOOLS-BUILD.  You should
+       consult this file as well as the README file if you are building  in  a
+       non-Unix-like environment.
+
+
+PCRE BUILD-TIME OPTIONS
+
+       The  rest of this document describes the optional features of PCRE that
+       can be selected when the library is compiled. It  assumes  use  of  the
+       configure  script,  where  the  optional features are selected or dese-
+       lected by providing options to configure before running the  make  com-
+       mand.  However,  the same options can be selected in both Unix-like and
+       non-Unix-like environments using the GUI facility of cmake-gui  if  you
+       are using CMake instead of configure to build PCRE.
+
+       If  you  are not using Autotools or CMake, option selection can be done
+       by editing the config.h file, or by passing parameter settings  to  the
+       compiler, as described in NON-AUTOTOOLS-BUILD.
+
+       The complete list of options for configure (which includes the standard
+       ones such as the  selection  of  the  installation  directory)  can  be
+       obtained by running
+
+         ./configure --help
+
+       The  following  sections  include  descriptions  of options whose names
+       begin with --enable or --disable. These settings specify changes to the
+       defaults  for  the configure command. Because of the way that configure
+       works, --enable and --disable always come in pairs, so  the  complemen-
+       tary  option always exists as well, but as it specifies the default, it
+       is not described.
+
+
+BUILDING 8-BIT, 16-BIT AND 32-BIT LIBRARIES
+
+       By default, a library called libpcre  is  built,  containing  functions
+       that  take  string  arguments  contained in vectors of bytes, either as
+       single-byte characters, or interpreted as UTF-8 strings. You  can  also
+       build  a  separate library, called libpcre16, in which strings are con-
+       tained in vectors of 16-bit data units and interpreted either  as  sin-
+       gle-unit characters or UTF-16 strings, by adding
+
+         --enable-pcre16
+
+       to  the  configure  command.  You  can  also build yet another separate
+       library, called libpcre32, in which strings are contained in vectors of
+       32-bit  data  units and interpreted either as single-unit characters or
+       UTF-32 strings, by adding
+
+         --enable-pcre32
+
+       to the configure command. If you do not want the 8-bit library, add
+
+         --disable-pcre8
+
+       as well. At least one of the three libraries must be built.  Note  that
+       the  C++  and  POSIX  wrappers are for the 8-bit library only, and that
+       pcregrep is an 8-bit program. None of these are  built  if  you  select
+       only the 16-bit or 32-bit libraries.
+
+
+BUILDING SHARED AND STATIC LIBRARIES
+
+       The  Autotools  PCRE building process uses libtool to build both shared
+       and static libraries by default. You  can  suppress  one  of  these  by
+       adding one of
+
+         --disable-shared
+         --disable-static
+
+       to the configure command, as required.
+
+
+C++ SUPPORT
+
+       By  default,  if the 8-bit library is being built, the configure script
+       will search for a C++ compiler and C++ header files. If it finds  them,
+       it  automatically  builds  the C++ wrapper library (which supports only
+       8-bit strings). You can disable this by adding
+
+         --disable-cpp
+
+       to the configure command.
+
+
+UTF-8, UTF-16 AND UTF-32 SUPPORT
+
+       To build PCRE with support for UTF Unicode character strings, add
+
+         --enable-utf
+
+       to the configure command. This setting applies to all three  libraries,
+       adding  support  for  UTF-8 to the 8-bit library, support for UTF-16 to
+       the 16-bit library, and  support  for  UTF-32  to  the  to  the  32-bit
+       library.  There  are no separate options for enabling UTF-8, UTF-16 and
+       UTF-32 independently because that would allow ridiculous settings  such
+       as  requesting UTF-16 support while building only the 8-bit library. It
+       is not possible to build one library with UTF support and another with-
+       out  in the same configuration. (For backwards compatibility, --enable-
+       utf8 is a synonym of --enable-utf.)
+
+       Of itself, this setting does not make  PCRE  treat  strings  as  UTF-8,
+       UTF-16  or UTF-32. As well as compiling PCRE with this option, you also
+       have have to set the PCRE_UTF8, PCRE_UTF16  or  PCRE_UTF32  option  (as
+       appropriate) when you call one of the pattern compiling functions.
+
+       If  you  set --enable-utf when compiling in an EBCDIC environment, PCRE
+       expects its input to be either ASCII or UTF-8 (depending  on  the  run-
+       time option). It is not possible to support both EBCDIC and UTF-8 codes
+       in the same version of  the  library.  Consequently,  --enable-utf  and
+       --enable-ebcdic are mutually exclusive.
+
+
+UNICODE CHARACTER PROPERTY SUPPORT
+
+       UTF  support allows the libraries to process character codepoints up to
+       0x10ffff in the strings that they handle. On its own, however, it  does
+       not provide any facilities for accessing the properties of such charac-
+       ters. If you want to be able to use the pattern escapes \P, \p, and \X,
+       which refer to Unicode character properties, you must add
+
+         --enable-unicode-properties
+
+       to  the  configure  command. This implies UTF support, even if you have
+       not explicitly requested it.
+
+       Including Unicode property support adds around 30K  of  tables  to  the
+       PCRE  library.  Only  the general category properties such as Lu and Nd
+       are supported. Details are given in the pcrepattern documentation.
+
+
+JUST-IN-TIME COMPILER SUPPORT
+
+       Just-in-time compiler support is included in the build by specifying
+
+         --enable-jit
+
+       This support is available only for certain hardware  architectures.  If
+       this  option  is  set  for  an unsupported architecture, a compile time
+       error occurs.  See the pcrejit documentation for a  discussion  of  JIT
+       usage. When JIT support is enabled, pcregrep automatically makes use of
+       it, unless you add
+
+         --disable-pcregrep-jit
+
+       to the "configure" command.
+
+
+CODE VALUE OF NEWLINE
+
+       By default, PCRE interprets the linefeed (LF) character  as  indicating
+       the  end  of  a line. This is the normal newline character on Unix-like
+       systems. You can compile PCRE to use carriage return (CR)  instead,  by
+       adding
+
+         --enable-newline-is-cr
+
+       to  the  configure  command.  There  is  also  a --enable-newline-is-lf
+       option, which explicitly specifies linefeed as the newline character.
+
+       Alternatively, you can specify that line endings are to be indicated by
+       the two character sequence CRLF. If you want this, add
+
+         --enable-newline-is-crlf
+
+       to the configure command. There is a fourth option, specified by
+
+         --enable-newline-is-anycrlf
+
+       which  causes  PCRE  to recognize any of the three sequences CR, LF, or
+       CRLF as indicating a line ending. Finally, a fifth option, specified by
+
+         --enable-newline-is-any
+
+       causes PCRE to recognize any Unicode newline sequence.
+
+       Whatever line ending convention is selected when PCRE is built  can  be
+       overridden  when  the library functions are called. At build time it is
+       conventional to use the standard for your operating system.
+
+
+WHAT \R MATCHES
+
+       By default, the sequence \R in a pattern matches  any  Unicode  newline
+       sequence,  whatever  has  been selected as the line ending sequence. If
+       you specify
+
+         --enable-bsr-anycrlf
+
+       the default is changed so that \R matches only CR, LF, or  CRLF.  What-
+       ever  is selected when PCRE is built can be overridden when the library
+       functions are called.
+
+
+POSIX MALLOC USAGE
+
+       When the 8-bit library is called through the POSIX interface  (see  the
+       pcreposix  documentation),  additional  working storage is required for
+       holding the pointers to capturing  substrings,  because  PCRE  requires
+       three integers per substring, whereas the POSIX interface provides only
+       two. If the number of expected substrings is small, the  wrapper  func-
+       tion  uses  space  on the stack, because this is faster than using mal-
+       loc() for each call. The default threshold above which the stack is  no
+       longer used is 10; it can be changed by adding a setting such as
+
+         --with-posix-malloc-threshold=20
+
+       to the configure command.
+
+
+HANDLING VERY LARGE PATTERNS
+
+       Within  a  compiled  pattern,  offset values are used to point from one
+       part to another (for example, from an opening parenthesis to an  alter-
+       nation  metacharacter).  By default, in the 8-bit and 16-bit libraries,
+       two-byte values are used for these offsets, leading to a  maximum  size
+       for  a compiled pattern of around 64K. This is sufficient to handle all
+       but the most gigantic patterns.  Nevertheless, some people do  want  to
+       process  truly  enormous patterns, so it is possible to compile PCRE to
+       use three-byte or four-byte offsets by adding a setting such as
+
+         --with-link-size=3
+
+       to the configure command. The value given must be 2, 3, or 4.  For  the
+       16-bit  library,  a  value of 3 is rounded up to 4. In these libraries,
+       using longer offsets slows down the operation of PCRE because it has to
+       load  additional  data  when  handling them. For the 32-bit library the
+       value is always 4 and cannot be overridden; the value  of  --with-link-
+       size is ignored.
+
+
+AVOIDING EXCESSIVE STACK USAGE
+
+       When matching with the pcre_exec() function, PCRE implements backtrack-
+       ing by making recursive calls to an internal function  called  match().
+       In  environments  where  the size of the stack is limited, this can se-
+       verely limit PCRE's operation. (The Unix environment does  not  usually
+       suffer from this problem, but it may sometimes be necessary to increase
+       the maximum stack size.  There is a discussion in the  pcrestack  docu-
+       mentation.)  An alternative approach to recursion that uses memory from
+       the heap to remember data, instead of using recursive  function  calls,
+       has  been  implemented to work round the problem of limited stack size.
+       If you want to build a version of PCRE that works this way, add
+
+         --disable-stack-for-recursion
+
+       to the configure command. With this configuration, PCRE  will  use  the
+       pcre_stack_malloc  and pcre_stack_free variables to call memory manage-
+       ment functions. By default these point to malloc() and free(), but  you
+       can replace the pointers so that your own functions are used instead.
+
+       Separate  functions  are  provided  rather  than  using pcre_malloc and
+       pcre_free because the  usage  is  very  predictable:  the  block  sizes
+       requested  are  always  the  same,  and  the blocks are always freed in
+       reverse order. A calling program might be able to  implement  optimized
+       functions  that  perform  better  than  malloc()  and free(). PCRE runs
+       noticeably more slowly when built in this way. This option affects only
+       the pcre_exec() function; it is not relevant for pcre_dfa_exec().
+
+
+LIMITING PCRE RESOURCE USAGE
+
+       Internally,  PCRE has a function called match(), which it calls repeat-
+       edly  (sometimes  recursively)  when  matching  a  pattern   with   the
+       pcre_exec()  function.  By controlling the maximum number of times this
+       function may be called during a single matching operation, a limit  can
+       be  placed  on  the resources used by a single call to pcre_exec(). The
+       limit can be changed at run time, as described in the pcreapi  documen-
+       tation.  The default is 10 million, but this can be changed by adding a
+       setting such as
+
+         --with-match-limit=500000
+
+       to  the  configure  command.  This  setting  has  no  effect   on   the
+       pcre_dfa_exec() matching function.
+
+       In  some  environments  it is desirable to limit the depth of recursive
+       calls of match() more strictly than the total number of calls, in order
+       to  restrict  the maximum amount of stack (or heap, if --disable-stack-
+       for-recursion is specified) that is used. A second limit controls this;
+       it  defaults  to  the  value  that is set for --with-match-limit, which
+       imposes no additional constraints. However, you can set a  lower  limit
+       by adding, for example,
+
+         --with-match-limit-recursion=10000
+
+       to  the  configure  command.  This  value can also be overridden at run
+       time.
+
+
+CREATING CHARACTER TABLES AT BUILD TIME
+
+       PCRE uses fixed tables for processing characters whose code values  are
+       less  than 256. By default, PCRE is built with a set of tables that are
+       distributed in the file pcre_chartables.c.dist. These  tables  are  for
+       ASCII codes only. If you add
+
+         --enable-rebuild-chartables
+
+       to  the  configure  command, the distributed tables are no longer used.
+       Instead, a program called dftables is compiled and  run.  This  outputs
+       the source for new set of tables, created in the default locale of your
+       C run-time system. (This method of replacing the tables does  not  work
+       if  you are cross compiling, because dftables is run on the local host.
+       If you need to create alternative tables when cross compiling, you will
+       have to do so "by hand".)
+
+
+USING EBCDIC CODE
+
+       PCRE  assumes  by  default that it will run in an environment where the
+       character code is ASCII (or Unicode, which is  a  superset  of  ASCII).
+       This  is  the  case for most computer operating systems. PCRE can, how-
+       ever, be compiled to run in an EBCDIC environment by adding
+
+         --enable-ebcdic
+
+       to the configure command. This setting implies --enable-rebuild-charta-
+       bles.  You  should  only  use  it if you know that you are in an EBCDIC
+       environment (for example,  an  IBM  mainframe  operating  system).  The
+       --enable-ebcdic option is incompatible with --enable-utf.
+
+       The EBCDIC character that corresponds to an ASCII LF is assumed to have
+       the value 0x15 by default. However, in some EBCDIC  environments,  0x25
+       is used. In such an environment you should use
+
+         --enable-ebcdic-nl25
+
+       as well as, or instead of, --enable-ebcdic. The EBCDIC character for CR
+       has the same value as in ASCII, namely, 0x0d.  Whichever  of  0x15  and
+       0x25 is not chosen as LF is made to correspond to the Unicode NEL char-
+       acter (which, in Unicode, is 0x85).
+
+       The options that select newline behaviour, such as --enable-newline-is-
+       cr, and equivalent run-time options, refer to these character values in
+       an EBCDIC environment.
+
+
+PCREGREP OPTIONS FOR COMPRESSED FILE SUPPORT
+
+       By default, pcregrep reads all files as plain text. You can build it so
+       that it recognizes files whose names end in .gz or .bz2, and reads them
+       with libz or libbz2, respectively, by adding one or both of
+
+         --enable-pcregrep-libz
+         --enable-pcregrep-libbz2
+
+       to the configure command. These options naturally require that the rel-
+       evant  libraries  are installed on your system. Configuration will fail
+       if they are not.
+
+
+PCREGREP BUFFER SIZE
+
+       pcregrep uses an internal buffer to hold a "window" on the file  it  is
+       scanning, in order to be able to output "before" and "after" lines when
+       it finds a match. The size of the buffer is controlled by  a  parameter
+       whose default value is 20K. The buffer itself is three times this size,
+       but because of the way it is used for holding "before" lines, the long-
+       est  line  that  is guaranteed to be processable is the parameter size.
+       You can change the default parameter value by adding, for example,
+
+         --with-pcregrep-bufsize=50K
+
+       to the configure command. The caller of pcregrep can, however, override
+       this value by specifying a run-time option.
+
+
+PCRETEST OPTION FOR LIBREADLINE SUPPORT
+
+       If you add
+
+         --enable-pcretest-libreadline
+
+       to  the  configure  command,  pcretest  is  linked with the libreadline
+       library, and when its input is from a terminal, it reads it  using  the
+       readline() function. This provides line-editing and history facilities.
+       Note that libreadline is GPL-licensed, so if you distribute a binary of
+       pcretest linked in this way, there may be licensing issues.
+
+       Setting  this  option  causes  the -lreadline option to be added to the
+       pcretest build. In many operating environments with  a  sytem-installed
+       libreadline this is sufficient. However, in some environments (e.g.  if
+       an unmodified distribution version of readline is in use),  some  extra
+       configuration  may  be necessary. The INSTALL file for libreadline says
+       this:
+
+         "Readline uses the termcap functions, but does not link with the
+         termcap or curses library itself, allowing applications which link
+         with readline the to choose an appropriate library."
+
+       If your environment has not been set up so that an appropriate  library
+       is automatically included, you may need to add something like
+
+         LIBS="-ncurses"
+
+       immediately before the configure command.
+
+
+DEBUGGING WITH VALGRIND SUPPORT
+
+       By adding the
+
+         --enable-valgrind
+
+       option  to to the configure command, PCRE will use valgrind annotations
+       to mark certain memory regions as  unaddressable.  This  allows  it  to
+       detect invalid memory accesses, and is mostly useful for debugging PCRE
+       itself.
+
+
+CODE COVERAGE REPORTING
+
+       If your C compiler is gcc, you can build a version  of  PCRE  that  can
+       generate a code coverage report for its test suite. To enable this, you
+       must install lcov version 1.6 or above. Then specify
+
+         --enable-coverage
+
+       to the configure command and build PCRE in the usual way.
+
+       Note that using ccache (a caching C compiler) is incompatible with code
+       coverage  reporting. If you have configured ccache to run automatically
+       on your system, you must set the environment variable
+
+         CCACHE_DISABLE=1
+
+       before running make to build PCRE, so that ccache is not used.
+
+       When --enable-coverage is used,  the  following  addition  targets  are
+       added to the Makefile:
+
+         make coverage
+
+       This  creates  a  fresh  coverage report for the PCRE test suite. It is
+       equivalent to running "make coverage-reset", "make  coverage-baseline",
+       "make check", and then "make coverage-report".
+
+         make coverage-reset
+
+       This zeroes the coverage counters, but does nothing else.
+
+         make coverage-baseline
+
+       This captures baseline coverage information.
+
+         make coverage-report
+
+       This creates the coverage report.
+
+         make coverage-clean-report
+
+       This  removes the generated coverage report without cleaning the cover-
+       age data itself.
+
+         make coverage-clean-data
+
+       This removes the captured coverage data without removing  the  coverage
+       files created at compile time (*.gcno).
+
+         make coverage-clean
+
+       This  cleans all coverage data including the generated coverage report.
+       For more information about code coverage, see the gcov and  lcov  docu-
+       mentation.
+
+
+SEE ALSO
+
+       pcreapi(3), pcre16, pcre32, pcre_config(3).
+
+
+AUTHOR
+
+       Philip Hazel
+       University Computing Service
+       Cambridge CB2 3QH, England.
+
+
+REVISION
+
+       Last updated: 12 May 2013
+       Copyright (c) 1997-2013 University of Cambridge.
+------------------------------------------------------------------------------
+
+
+PCREMATCHING(3)            Library Functions Manual            PCREMATCHING(3)
+
+
+
+NAME
+       PCRE - Perl-compatible regular expressions
+
+PCRE MATCHING ALGORITHMS
+
+       This document describes the two different algorithms that are available
+       in PCRE for matching a compiled regular expression against a given sub-
+       ject  string.  The  "standard"  algorithm  is  the  one provided by the
+       pcre_exec(), pcre16_exec() and pcre32_exec() functions. These  work  in
+       the  same as as Perl's matching function, and provide a Perl-compatible
+       matching  operation.   The  just-in-time  (JIT)  optimization  that  is
+       described  in  the pcrejit documentation is compatible with these func-
+       tions.
+
+       An  alternative  algorithm  is   provided   by   the   pcre_dfa_exec(),
+       pcre16_dfa_exec()  and  pcre32_dfa_exec()  functions; they operate in a
+       different way, and are not Perl-compatible. This alternative has advan-
+       tages and disadvantages compared with the standard algorithm, and these
+       are described below.
+
+       When there is only one possible way in which a given subject string can
+       match  a pattern, the two algorithms give the same answer. A difference
+       arises, however, when there are multiple possibilities. For example, if
+       the pattern
+
+         ^<.*>
+
+       is matched against the string
+
+           
+
+       there are three possible answers. The standard algorithm finds only one
+       of them, whereas the alternative algorithm finds all three.
+
+
+REGULAR EXPRESSIONS AS TREES
+
+       The set of strings that are matched by a regular expression can be rep-
+       resented  as  a  tree structure. An unlimited repetition in the pattern
+       makes the tree of infinite size, but it is still a tree.  Matching  the
+       pattern  to a given subject string (from a given starting point) can be
+       thought of as a search of the tree.  There are two  ways  to  search  a
+       tree:  depth-first  and  breadth-first, and these correspond to the two
+       matching algorithms provided by PCRE.
+
+
+THE STANDARD MATCHING ALGORITHM
+
+       In the terminology of Jeffrey Friedl's book "Mastering Regular  Expres-
+       sions",  the  standard  algorithm  is an "NFA algorithm". It conducts a
+       depth-first search of the pattern tree. That is, it  proceeds  along  a
+       single path through the tree, checking that the subject matches what is
+       required. When there is a mismatch, the algorithm  tries  any  alterna-
+       tives  at  the  current point, and if they all fail, it backs up to the
+       previous branch point in the  tree,  and  tries  the  next  alternative
+       branch  at  that  level.  This often involves backing up (moving to the
+       left) in the subject string as well.  The  order  in  which  repetition
+       branches  are  tried  is controlled by the greedy or ungreedy nature of
+       the quantifier.
+
+       If a leaf node is reached, a matching string has  been  found,  and  at
+       that  point the algorithm stops. Thus, if there is more than one possi-
+       ble match, this algorithm returns the first one that it finds.  Whether
+       this  is the shortest, the longest, or some intermediate length depends
+       on the way the greedy and ungreedy repetition quantifiers are specified
+       in the pattern.
+
+       Because  it  ends  up  with a single path through the tree, it is rela-
+       tively straightforward for this algorithm to keep  track  of  the  sub-
+       strings  that  are  matched  by portions of the pattern in parentheses.
+       This provides support for capturing parentheses and back references.
+
+
+THE ALTERNATIVE MATCHING ALGORITHM
+
+       This algorithm conducts a breadth-first search of  the  tree.  Starting
+       from  the  first  matching  point  in the subject, it scans the subject
+       string from left to right, once, character by character, and as it does
+       this,  it remembers all the paths through the tree that represent valid
+       matches. In Friedl's terminology, this is a kind  of  "DFA  algorithm",
+       though  it is not implemented as a traditional finite state machine (it
+       keeps multiple states active simultaneously).
+
+       Although the general principle of this matching algorithm  is  that  it
+       scans  the subject string only once, without backtracking, there is one
+       exception: when a lookaround assertion is encountered,  the  characters
+       following  or  preceding  the  current  point  have to be independently
+       inspected.
+
+       The scan continues until either the end of the subject is  reached,  or
+       there  are  no more unterminated paths. At this point, terminated paths
+       represent the different matching possibilities (if there are none,  the
+       match  has  failed).   Thus,  if there is more than one possible match,
+       this algorithm finds all of them, and in particular, it finds the long-
+       est.  The  matches are returned in decreasing order of length. There is
+       an option to stop the algorithm after the first match (which is  neces-
+       sarily the shortest) is found.
+
+       Note that all the matches that are found start at the same point in the
+       subject. If the pattern
+
+         cat(er(pillar)?)?
+
+       is matched against the string "the caterpillar catchment",  the  result
+       will  be the three strings "caterpillar", "cater", and "cat" that start
+       at the fifth character of the subject. The algorithm does not automati-
+       cally move on to find matches that start at later positions.
+
+       There are a number of features of PCRE regular expressions that are not
+       supported by the alternative matching algorithm. They are as follows:
+
+       1. Because the algorithm finds all  possible  matches,  the  greedy  or
+       ungreedy  nature  of repetition quantifiers is not relevant. Greedy and
+       ungreedy quantifiers are treated in exactly the same way. However, pos-
+       sessive  quantifiers can make a difference when what follows could also
+       match what is quantified, for example in a pattern like this:
+
+         ^a++\w!
+
+       This pattern matches "aaab!" but not "aaa!", which would be matched  by
+       a  non-possessive quantifier. Similarly, if an atomic group is present,
+       it is matched as if it were a standalone pattern at the current  point,
+       and  the  longest match is then "locked in" for the rest of the overall
+       pattern.
+
+       2. When dealing with multiple paths through the tree simultaneously, it
+       is  not  straightforward  to  keep track of captured substrings for the
+       different matching possibilities, and  PCRE's  implementation  of  this
+       algorithm does not attempt to do this. This means that no captured sub-
+       strings are available.
+
+       3. Because no substrings are captured, back references within the  pat-
+       tern are not supported, and cause errors if encountered.
+
+       4.  For  the same reason, conditional expressions that use a backrefer-
+       ence as the condition or test for a specific group  recursion  are  not
+       supported.
+
+       5.  Because  many  paths  through the tree may be active, the \K escape
+       sequence, which resets the start of the match when encountered (but may
+       be  on  some  paths  and not on others), is not supported. It causes an
+       error if encountered.
+
+       6. Callouts are supported, but the value of the  capture_top  field  is
+       always 1, and the value of the capture_last field is always -1.
+
+       7.  The  \C  escape  sequence, which (in the standard algorithm) always
+       matches a single data unit, even in UTF-8, UTF-16 or UTF-32  modes,  is
+       not  supported  in these modes, because the alternative algorithm moves
+       through the subject string one character (not data unit) at a time, for
+       all active paths through the tree.
+
+       8.  Except for (*FAIL), the backtracking control verbs such as (*PRUNE)
+       are not supported. (*FAIL) is supported, and  behaves  like  a  failing
+       negative assertion.
+
+
+ADVANTAGES OF THE ALTERNATIVE ALGORITHM
+
+       Using  the alternative matching algorithm provides the following advan-
+       tages:
+
+       1. All possible matches (at a single point in the subject) are automat-
+       ically  found,  and  in particular, the longest match is found. To find
+       more than one match using the standard algorithm, you have to do kludgy
+       things with callouts.
+
+       2.  Because  the  alternative  algorithm  scans the subject string just
+       once, and never needs to backtrack (except for lookbehinds), it is pos-
+       sible  to  pass  very  long subject strings to the matching function in
+       several pieces, checking for partial matching each time. Although it is
+       possible  to  do multi-segment matching using the standard algorithm by
+       retaining partially matched substrings, it  is  more  complicated.  The
+       pcrepartial  documentation  gives  details of partial matching and dis-
+       cusses multi-segment matching.
+
+
+DISADVANTAGES OF THE ALTERNATIVE ALGORITHM
+
+       The alternative algorithm suffers from a number of disadvantages:
+
+       1. It is substantially slower than  the  standard  algorithm.  This  is
+       partly  because  it has to search for all possible matches, but is also
+       because it is less susceptible to optimization.
+
+       2. Capturing parentheses and back references are not supported.
+
+       3. Although atomic groups are supported, their use does not provide the
+       performance advantage that it does for the standard algorithm.
+
+
+AUTHOR
+
+       Philip Hazel
+       University Computing Service
+       Cambridge CB2 3QH, England.
+
+
+REVISION
+
+       Last updated: 08 January 2012
+       Copyright (c) 1997-2012 University of Cambridge.
+------------------------------------------------------------------------------
+
+
+PCREAPI(3)                 Library Functions Manual                 PCREAPI(3)
+
+
+
+NAME
+       PCRE - Perl-compatible regular expressions
+
+       #include 
+
+
+PCRE NATIVE API BASIC FUNCTIONS
+
+       pcre *pcre_compile(const char *pattern, int options,
+            const char **errptr, int *erroffset,
+            const unsigned char *tableptr);
+
+       pcre *pcre_compile2(const char *pattern, int options,
+            int *errorcodeptr,
+            const char **errptr, int *erroffset,
+            const unsigned char *tableptr);
+
+       pcre_extra *pcre_study(const pcre *code, int options,
+            const char **errptr);
+
+       void pcre_free_study(pcre_extra *extra);
+
+       int pcre_exec(const pcre *code, const pcre_extra *extra,
+            const char *subject, int length, int startoffset,
+            int options, int *ovector, int ovecsize);
+
+       int pcre_dfa_exec(const pcre *code, const pcre_extra *extra,
+            const char *subject, int length, int startoffset,
+            int options, int *ovector, int ovecsize,
+            int *workspace, int wscount);
+
+
+PCRE NATIVE API STRING EXTRACTION FUNCTIONS
+
+       int pcre_copy_named_substring(const pcre *code,
+            const char *subject, int *ovector,
+            int stringcount, const char *stringname,
+            char *buffer, int buffersize);
+
+       int pcre_copy_substring(const char *subject, int *ovector,
+            int stringcount, int stringnumber, char *buffer,
+            int buffersize);
+
+       int pcre_get_named_substring(const pcre *code,
+            const char *subject, int *ovector,
+            int stringcount, const char *stringname,
+            const char **stringptr);
+
+       int pcre_get_stringnumber(const pcre *code,
+            const char *name);
+
+       int pcre_get_stringtable_entries(const pcre *code,
+            const char *name, char **first, char **last);
+
+       int pcre_get_substring(const char *subject, int *ovector,
+            int stringcount, int stringnumber,
+            const char **stringptr);
+
+       int pcre_get_substring_list(const char *subject,
+            int *ovector, int stringcount, const char ***listptr);
+
+       void pcre_free_substring(const char *stringptr);
+
+       void pcre_free_substring_list(const char **stringptr);
+
+
+PCRE NATIVE API AUXILIARY FUNCTIONS
+
+       int pcre_jit_exec(const pcre *code, const pcre_extra *extra,
+            const char *subject, int length, int startoffset,
+            int options, int *ovector, int ovecsize,
+            pcre_jit_stack *jstack);
+
+       pcre_jit_stack *pcre_jit_stack_alloc(int startsize, int maxsize);
+
+       void pcre_jit_stack_free(pcre_jit_stack *stack);
+
+       void pcre_assign_jit_stack(pcre_extra *extra,
+            pcre_jit_callback callback, void *data);
+
+       const unsigned char *pcre_maketables(void);
+
+       int pcre_fullinfo(const pcre *code, const pcre_extra *extra,
+            int what, void *where);
+
+       int pcre_refcount(pcre *code, int adjust);
+
+       int pcre_config(int what, void *where);
+
+       const char *pcre_version(void);
+
+       int pcre_pattern_to_host_byte_order(pcre *code,
+            pcre_extra *extra, const unsigned char *tables);
+
+
+PCRE NATIVE API INDIRECTED FUNCTIONS
+
+       void *(*pcre_malloc)(size_t);
+
+       void (*pcre_free)(void *);
+
+       void *(*pcre_stack_malloc)(size_t);
+
+       void (*pcre_stack_free)(void *);
+
+       int (*pcre_callout)(pcre_callout_block *);
+
+
+PCRE 8-BIT, 16-BIT, AND 32-BIT LIBRARIES
+
+       As  well  as  support  for  8-bit character strings, PCRE also supports
+       16-bit strings (from release 8.30) and  32-bit  strings  (from  release
+       8.32),  by means of two additional libraries. They can be built as well
+       as, or instead of, the 8-bit library. To avoid too  much  complication,
+       this  document describes the 8-bit versions of the functions, with only
+       occasional references to the 16-bit and 32-bit libraries.
+
+       The 16-bit and 32-bit functions operate in the same way as their  8-bit
+       counterparts;  they  just  use different data types for their arguments
+       and results, and their names start with pcre16_ or pcre32_  instead  of
+       pcre_.  For  every  option  that  has  UTF8  in  its name (for example,
+       PCRE_UTF8), there are corresponding 16-bit and 32-bit names  with  UTF8
+       replaced by UTF16 or UTF32, respectively. This facility is in fact just
+       cosmetic; the 16-bit and 32-bit option names define the same  bit  val-
+       ues.
+
+       References to bytes and UTF-8 in this document should be read as refer-
+       ences to 16-bit data units and UTF-16 when using the 16-bit library, or
+       32-bit  data  units  and  UTF-32  when using the 32-bit library, unless
+       specified otherwise.  More details of the specific differences for  the
+       16-bit and 32-bit libraries are given in the pcre16 and pcre32 pages.
+
+
+PCRE API OVERVIEW
+
+       PCRE has its own native API, which is described in this document. There
+       are also some wrapper functions (for the 8-bit library only) that  cor-
+       respond  to  the  POSIX  regular  expression  API, but they do not give
+       access to all the functionality. They are described  in  the  pcreposix
+       documentation.  Both  of these APIs define a set of C function calls. A
+       C++ wrapper (again for the 8-bit library only) is also distributed with
+       PCRE. It is documented in the pcrecpp page.
+
+       The  native  API  C  function prototypes are defined in the header file
+       pcre.h, and on Unix-like systems the (8-bit) library itself  is  called
+       libpcre.  It  can  normally be accessed by adding -lpcre to the command
+       for linking an application that uses PCRE. The header file defines  the
+       macros PCRE_MAJOR and PCRE_MINOR to contain the major and minor release
+       numbers for the library. Applications can use these to include  support
+       for different releases of PCRE.
+
+       In a Windows environment, if you want to statically link an application
+       program against a non-dll pcre.a  file,  you  must  define  PCRE_STATIC
+       before  including  pcre.h or pcrecpp.h, because otherwise the pcre_mal-
+       loc()   and   pcre_free()   exported   functions   will   be   declared
+       __declspec(dllimport), with unwanted results.
+
+       The   functions   pcre_compile(),  pcre_compile2(),  pcre_study(),  and
+       pcre_exec() are used for compiling and matching regular expressions  in
+       a  Perl-compatible  manner. A sample program that demonstrates the sim-
+       plest way of using them is provided in the file  called  pcredemo.c  in
+       the PCRE source distribution. A listing of this program is given in the
+       pcredemo documentation, and the pcresample documentation describes  how
+       to compile and run it.
+
+       Just-in-time  compiler  support is an optional feature of PCRE that can
+       be built in appropriate hardware environments. It greatly speeds up the
+       matching  performance  of  many  patterns.  Simple  programs can easily
+       request that it be used if available, by  setting  an  option  that  is
+       ignored  when  it is not relevant. More complicated programs might need
+       to    make    use    of    the    functions     pcre_jit_stack_alloc(),
+       pcre_jit_stack_free(),  and pcre_assign_jit_stack() in order to control
+       the JIT code's memory usage.
+
+       From release 8.32 there is also a direct interface for  JIT  execution,
+       which  gives  improved performance. The JIT-specific functions are dis-
+       cussed in the pcrejit documentation.
+
+       A second matching function, pcre_dfa_exec(), which is not Perl-compati-
+       ble,  is  also provided. This uses a different algorithm for the match-
+       ing. The alternative algorithm finds all possible matches (at  a  given
+       point  in  the  subject), and scans the subject just once (unless there
+       are lookbehind assertions). However, this  algorithm  does  not  return
+       captured  substrings.  A description of the two matching algorithms and
+       their advantages and disadvantages is given in the  pcrematching  docu-
+       mentation.
+
+       In  addition  to  the  main compiling and matching functions, there are
+       convenience functions for extracting captured substrings from a subject
+       string that is matched by pcre_exec(). They are:
+
+         pcre_copy_substring()
+         pcre_copy_named_substring()
+         pcre_get_substring()
+         pcre_get_named_substring()
+         pcre_get_substring_list()
+         pcre_get_stringnumber()
+         pcre_get_stringtable_entries()
+
+       pcre_free_substring() and pcre_free_substring_list() are also provided,
+       to free the memory used for extracted strings.
+
+       The function pcre_maketables() is used to  build  a  set  of  character
+       tables   in   the   current   locale  for  passing  to  pcre_compile(),
+       pcre_exec(), or pcre_dfa_exec(). This is an optional facility  that  is
+       provided  for  specialist  use.  Most  commonly,  no special tables are
+       passed, in which case internal tables that are generated when  PCRE  is
+       built are used.
+
+       The  function  pcre_fullinfo()  is used to find out information about a
+       compiled pattern. The function pcre_version() returns a  pointer  to  a
+       string containing the version of PCRE and its date of release.
+
+       The  function  pcre_refcount()  maintains  a  reference count in a data
+       block containing a compiled pattern. This is provided for  the  benefit
+       of object-oriented applications.
+
+       The  global  variables  pcre_malloc and pcre_free initially contain the
+       entry points of the standard malloc()  and  free()  functions,  respec-
+       tively. PCRE calls the memory management functions via these variables,
+       so a calling program can replace them if it  wishes  to  intercept  the
+       calls. This should be done before calling any PCRE functions.
+
+       The  global  variables  pcre_stack_malloc  and pcre_stack_free are also
+       indirections to memory management functions.  These  special  functions
+       are  used  only  when  PCRE is compiled to use the heap for remembering
+       data, instead of recursive function calls, when running the pcre_exec()
+       function.  See  the  pcrebuild  documentation  for details of how to do
+       this. It is a non-standard way of building PCRE, for  use  in  environ-
+       ments  that  have  limited stacks. Because of the greater use of memory
+       management, it runs more slowly. Separate  functions  are  provided  so
+       that  special-purpose  external  code  can  be used for this case. When
+       used, these functions are always called in a  stack-like  manner  (last
+       obtained,  first freed), and always for memory blocks of the same size.
+       There is a discussion about PCRE's stack usage in the  pcrestack  docu-
+       mentation.
+
+       The global variable pcre_callout initially contains NULL. It can be set
+       by the caller to a "callout" function, which PCRE  will  then  call  at
+       specified  points during a matching operation. Details are given in the
+       pcrecallout documentation.
+
+
+NEWLINES
+
+       PCRE supports five different conventions for indicating line breaks  in
+       strings:  a  single  CR (carriage return) character, a single LF (line-
+       feed) character, the two-character sequence CRLF, any of the three pre-
+       ceding,  or any Unicode newline sequence. The Unicode newline sequences
+       are the three just mentioned, plus the single characters  VT  (vertical
+       tab, U+000B), FF (form feed, U+000C), NEL (next line, U+0085), LS (line
+       separator, U+2028), and PS (paragraph separator, U+2029).
+
+       Each of the first three conventions is used by at least  one  operating
+       system  as its standard newline sequence. When PCRE is built, a default
+       can be specified.  The default default is LF, which is the  Unix  stan-
+       dard.  When  PCRE  is run, the default can be overridden, either when a
+       pattern is compiled, or when it is matched.
+
+       At compile time, the newline convention can be specified by the options
+       argument  of  pcre_compile(), or it can be specified by special text at
+       the start of the pattern itself; this overrides any other settings. See
+       the pcrepattern page for details of the special character sequences.
+
+       In the PCRE documentation the word "newline" is used to mean "the char-
+       acter or pair of characters that indicate a line break". The choice  of
+       newline  convention  affects  the  handling of the dot, circumflex, and
+       dollar metacharacters, the handling of #-comments in /x mode, and, when
+       CRLF  is a recognized line ending sequence, the match position advance-
+       ment for a non-anchored pattern. There is more detail about this in the
+       section on pcre_exec() options below.
+
+       The  choice of newline convention does not affect the interpretation of
+       the \n or \r escape sequences, nor does  it  affect  what  \R  matches,
+       which is controlled in a similar way, but by separate options.
+
+
+MULTITHREADING
+
+       The  PCRE  functions  can be used in multi-threading applications, with
+       the  proviso  that  the  memory  management  functions  pointed  to  by
+       pcre_malloc, pcre_free, pcre_stack_malloc, and pcre_stack_free, and the
+       callout function pointed to by pcre_callout, are shared by all threads.
+
+       The compiled form of a regular expression is not altered during  match-
+       ing, so the same compiled pattern can safely be used by several threads
+       at once.
+
+       If the just-in-time optimization feature is being used, it needs  sepa-
+       rate  memory stack areas for each thread. See the pcrejit documentation
+       for more details.
+
+
+SAVING PRECOMPILED PATTERNS FOR LATER USE
+
+       The compiled form of a regular expression can be saved and re-used at a
+       later  time,  possibly by a different program, and even on a host other
+       than the one on which  it  was  compiled.  Details  are  given  in  the
+       pcreprecompile  documentation,  which  includes  a  description  of the
+       pcre_pattern_to_host_byte_order() function. However, compiling a  regu-
+       lar  expression  with one version of PCRE for use with a different ver-
+       sion is not guaranteed to work and may cause crashes.
+
+
+CHECKING BUILD-TIME OPTIONS
+
+       int pcre_config(int what, void *where);
+
+       The function pcre_config() makes it possible for a PCRE client to  dis-
+       cover which optional features have been compiled into the PCRE library.
+       The pcrebuild documentation has more details about these optional  fea-
+       tures.
+
+       The  first  argument  for pcre_config() is an integer, specifying which
+       information is required; the second argument is a pointer to a variable
+       into  which  the  information  is placed. The returned value is zero on
+       success, or the negative error code PCRE_ERROR_BADOPTION if  the  value
+       in  the  first argument is not recognized. The following information is
+       available:
+
+         PCRE_CONFIG_UTF8
+
+       The output is an integer that is set to one if UTF-8 support is  avail-
+       able;  otherwise it is set to zero. This value should normally be given
+       to the 8-bit version of this function, pcre_config(). If it is given to
+       the   16-bit  or  32-bit  version  of  this  function,  the  result  is
+       PCRE_ERROR_BADOPTION.
+
+         PCRE_CONFIG_UTF16
+
+       The output is an integer that is set to one if UTF-16 support is avail-
+       able;  otherwise it is set to zero. This value should normally be given
+       to the 16-bit version of this function, pcre16_config(). If it is given
+       to  the  8-bit  or  32-bit  version  of  this  function,  the result is
+       PCRE_ERROR_BADOPTION.
+
+         PCRE_CONFIG_UTF32
+
+       The output is an integer that is set to one if UTF-32 support is avail-
+       able;  otherwise it is set to zero. This value should normally be given
+       to the 32-bit version of this function, pcre32_config(). If it is given
+       to  the  8-bit  or  16-bit  version  of  this  function,  the result is
+       PCRE_ERROR_BADOPTION.
+
+         PCRE_CONFIG_UNICODE_PROPERTIES
+
+       The output is an integer that is set to  one  if  support  for  Unicode
+       character properties is available; otherwise it is set to zero.
+
+         PCRE_CONFIG_JIT
+
+       The output is an integer that is set to one if support for just-in-time
+       compiling is available; otherwise it is set to zero.
+
+         PCRE_CONFIG_JITTARGET
+
+       The output is a pointer to a zero-terminated "const char *" string.  If
+       JIT support is available, the string contains the name of the architec-
+       ture for which the JIT compiler is configured, for example  "x86  32bit
+       (little  endian  +  unaligned)".  If  JIT support is not available, the
+       result is NULL.
+
+         PCRE_CONFIG_NEWLINE
+
+       The output is an integer whose value specifies  the  default  character
+       sequence  that  is recognized as meaning "newline". The values that are
+       supported in ASCII/Unicode environments are: 10 for LF, 13 for CR, 3338
+       for  CRLF,  -2 for ANYCRLF, and -1 for ANY. In EBCDIC environments, CR,
+       ANYCRLF, and ANY yield the same values. However, the value  for  LF  is
+       normally  21, though some EBCDIC environments use 37. The corresponding
+       values for CRLF are 3349 and 3365. The default should  normally  corre-
+       spond to the standard sequence for your operating system.
+
+         PCRE_CONFIG_BSR
+
+       The output is an integer whose value indicates what character sequences
+       the \R escape sequence matches by default. A value of 0 means  that  \R
+       matches  any  Unicode  line ending sequence; a value of 1 means that \R
+       matches only CR, LF, or CRLF. The default can be overridden when a pat-
+       tern is compiled or matched.
+
+         PCRE_CONFIG_LINK_SIZE
+
+       The  output  is  an  integer that contains the number of bytes used for
+       internal  linkage  in  compiled  regular  expressions.  For  the  8-bit
+       library, the value can be 2, 3, or 4. For the 16-bit library, the value
+       is either 2 or 4 and is  still  a  number  of  bytes.  For  the  32-bit
+       library, the value is either 2 or 4 and is still a number of bytes. The
+       default value of 2 is sufficient for all but the most massive patterns,
+       since  it  allows  the compiled pattern to be up to 64K in size. Larger
+       values allow larger regular expressions to be compiled, at the  expense
+       of slower matching.
+
+         PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
+
+       The  output  is  an integer that contains the threshold above which the
+       POSIX interface uses malloc() for output vectors. Further  details  are
+       given in the pcreposix documentation.
+
+         PCRE_CONFIG_MATCH_LIMIT
+
+       The  output is a long integer that gives the default limit for the num-
+       ber of internal matching function calls  in  a  pcre_exec()  execution.
+       Further details are given with pcre_exec() below.
+
+         PCRE_CONFIG_MATCH_LIMIT_RECURSION
+
+       The output is a long integer that gives the default limit for the depth
+       of  recursion  when  calling  the  internal  matching  function  in   a
+       pcre_exec()  execution.  Further  details  are  given  with pcre_exec()
+       below.
+
+         PCRE_CONFIG_STACKRECURSE
+
+       The output is an integer that is set to one if internal recursion  when
+       running pcre_exec() is implemented by recursive function calls that use
+       the stack to remember their state. This is the usual way that  PCRE  is
+       compiled. The output is zero if PCRE was compiled to use blocks of data
+       on the  heap  instead  of  recursive  function  calls.  In  this  case,
+       pcre_stack_malloc  and  pcre_stack_free  are  called  to  manage memory
+       blocks on the heap, thus avoiding the use of the stack.
+
+
+COMPILING A PATTERN
+
+       pcre *pcre_compile(const char *pattern, int options,
+            const char **errptr, int *erroffset,
+            const unsigned char *tableptr);
+
+       pcre *pcre_compile2(const char *pattern, int options,
+            int *errorcodeptr,
+            const char **errptr, int *erroffset,
+            const unsigned char *tableptr);
+
+       Either of the functions pcre_compile() or pcre_compile2() can be called
+       to compile a pattern into an internal form. The only difference between
+       the two interfaces is that pcre_compile2() has an additional  argument,
+       errorcodeptr,  via  which  a  numerical  error code can be returned. To
+       avoid too much repetition, we refer just to pcre_compile()  below,  but
+       the information applies equally to pcre_compile2().
+
+       The pattern is a C string terminated by a binary zero, and is passed in
+       the pattern argument. A pointer to a single block  of  memory  that  is
+       obtained  via  pcre_malloc is returned. This contains the compiled code
+       and related data. The pcre type is defined for the returned block; this
+       is a typedef for a structure whose contents are not externally defined.
+       It is up to the caller to free the memory (via pcre_free) when it is no
+       longer required.
+
+       Although  the compiled code of a PCRE regex is relocatable, that is, it
+       does not depend on memory location, the complete pcre data block is not
+       fully  relocatable, because it may contain a copy of the tableptr argu-
+       ment, which is an address (see below).
+
+       The options argument contains various bit settings that affect the com-
+       pilation.  It  should be zero if no options are required. The available
+       options are described below. Some of them (in  particular,  those  that
+       are  compatible with Perl, but some others as well) can also be set and
+       unset from within the pattern (see  the  detailed  description  in  the
+       pcrepattern  documentation). For those options that can be different in
+       different parts of the pattern, the contents of  the  options  argument
+       specifies their settings at the start of compilation and execution. The
+       PCRE_ANCHORED, PCRE_BSR_xxx, PCRE_NEWLINE_xxx, PCRE_NO_UTF8_CHECK,  and
+       PCRE_NO_START_OPTIMIZE  options  can  be set at the time of matching as
+       well as at compile time.
+
+       If errptr is NULL, pcre_compile() returns NULL immediately.  Otherwise,
+       if  compilation  of  a  pattern fails, pcre_compile() returns NULL, and
+       sets the variable pointed to by errptr to point to a textual error mes-
+       sage. This is a static string that is part of the library. You must not
+       try to free it. Normally, the offset from the start of the  pattern  to
+       the data unit that was being processed when the error was discovered is
+       placed in the variable pointed to by erroffset, which must not be  NULL
+       (if  it is, an immediate error is given). However, for an invalid UTF-8
+       or UTF-16 string, the offset is that of the  first  data  unit  of  the
+       failing character.
+
+       Some  errors are not detected until the whole pattern has been scanned;
+       in these cases, the offset passed back is the length  of  the  pattern.
+       Note  that  the  offset is in data units, not characters, even in a UTF
+       mode. It may sometimes point into the middle of a UTF-8 or UTF-16 char-
+       acter.
+
+       If  pcre_compile2()  is  used instead of pcre_compile(), and the error-
+       codeptr argument is not NULL, a non-zero error code number is  returned
+       via  this argument in the event of an error. This is in addition to the
+       textual error message. Error codes and messages are listed below.
+
+       If the final argument, tableptr, is NULL, PCRE uses a  default  set  of
+       character  tables  that  are  built  when  PCRE  is compiled, using the
+       default C locale. Otherwise, tableptr must be an address  that  is  the
+       result  of  a  call to pcre_maketables(). This value is stored with the
+       compiled pattern, and used again by pcre_exec(), unless  another  table
+       pointer is passed to it. For more discussion, see the section on locale
+       support below.
+
+       This code fragment shows a typical straightforward  call  to  pcre_com-
+       pile():
+
+         pcre *re;
+         const char *error;
+         int erroffset;
+         re = pcre_compile(
+           "^A.*Z",          /* the pattern */
+           0,                /* default options */
+           &error,           /* for error message */
+           &erroffset,       /* for error offset */
+           NULL);            /* use default character tables */
+
+       The  following  names  for option bits are defined in the pcre.h header
+       file:
+
+         PCRE_ANCHORED
+
+       If this bit is set, the pattern is forced to be "anchored", that is, it
+       is  constrained to match only at the first matching point in the string
+       that is being searched (the "subject string"). This effect can also  be
+       achieved  by appropriate constructs in the pattern itself, which is the
+       only way to do it in Perl.
+
+         PCRE_AUTO_CALLOUT
+
+       If this bit is set, pcre_compile() automatically inserts callout items,
+       all  with  number  255, before each pattern item. For discussion of the
+       callout facility, see the pcrecallout documentation.
+
+         PCRE_BSR_ANYCRLF
+         PCRE_BSR_UNICODE
+
+       These options (which are mutually exclusive) control what the \R escape
+       sequence  matches.  The choice is either to match only CR, LF, or CRLF,
+       or to match any Unicode newline sequence. The default is specified when
+       PCRE is built. It can be overridden from within the pattern, or by set-
+       ting an option when a compiled pattern is matched.
+
+         PCRE_CASELESS
+
+       If this bit is set, letters in the pattern match both upper  and  lower
+       case  letters.  It  is  equivalent  to  Perl's /i option, and it can be
+       changed within a pattern by a (?i) option setting. In UTF-8 mode,  PCRE
+       always  understands the concept of case for characters whose values are
+       less than 128, so caseless matching is always possible. For  characters
+       with  higher  values,  the concept of case is supported if PCRE is com-
+       piled with Unicode property support, but not otherwise. If you want  to
+       use  caseless  matching  for  characters 128 and above, you must ensure
+       that PCRE is compiled with Unicode property support  as  well  as  with
+       UTF-8 support.
+
+         PCRE_DOLLAR_ENDONLY
+
+       If  this bit is set, a dollar metacharacter in the pattern matches only
+       at the end of the subject string. Without this option,  a  dollar  also
+       matches  immediately before a newline at the end of the string (but not
+       before any other newlines). The PCRE_DOLLAR_ENDONLY option  is  ignored
+       if  PCRE_MULTILINE  is  set.   There is no equivalent to this option in
+       Perl, and no way to set it within a pattern.
+
+         PCRE_DOTALL
+
+       If this bit is set, a dot metacharacter in the pattern matches a  char-
+       acter of any value, including one that indicates a newline. However, it
+       only ever matches one character, even if newlines are  coded  as  CRLF.
+       Without  this option, a dot does not match when the current position is
+       at a newline. This option is equivalent to Perl's /s option, and it can
+       be  changed within a pattern by a (?s) option setting. A negative class
+       such as [^a] always matches newline characters, independent of the set-
+       ting of this option.
+
+         PCRE_DUPNAMES
+
+       If  this  bit is set, names used to identify capturing subpatterns need
+       not be unique. This can be helpful for certain types of pattern when it
+       is  known  that  only  one instance of the named subpattern can ever be
+       matched. There are more details of named subpatterns  below;  see  also
+       the pcrepattern documentation.
+
+         PCRE_EXTENDED
+
+       If  this  bit  is  set,  white space data characters in the pattern are
+       totally ignored except when escaped or inside a character class.  White
+       space does not include the VT character (code 11). In addition, charac-
+       ters between an unescaped # outside a character class and the next new-
+       line,  inclusive,  are  also  ignored.  This is equivalent to Perl's /x
+       option, and it can be changed within a pattern by a  (?x)  option  set-
+       ting.
+
+       Which  characters  are  interpreted  as  newlines  is controlled by the
+       options passed to pcre_compile() or by a special sequence at the  start
+       of  the  pattern, as described in the section entitled "Newline conven-
+       tions" in the pcrepattern documentation. Note that the end of this type
+       of  comment  is  a  literal  newline  sequence  in  the pattern; escape
+       sequences that happen to represent a newline do not count.
+
+       This option makes it possible to include  comments  inside  complicated
+       patterns.   Note,  however,  that this applies only to data characters.
+       White space  characters  may  never  appear  within  special  character
+       sequences in a pattern, for example within the sequence (?( that intro-
+       duces a conditional subpattern.
+
+         PCRE_EXTRA
+
+       This option was invented in order to turn on  additional  functionality
+       of  PCRE  that  is  incompatible with Perl, but it is currently of very
+       little use. When set, any backslash in a pattern that is followed by  a
+       letter  that  has  no  special  meaning causes an error, thus reserving
+       these combinations for future expansion. By  default,  as  in  Perl,  a
+       backslash  followed by a letter with no special meaning is treated as a
+       literal. (Perl can, however, be persuaded to give an error for this, by
+       running  it with the -w option.) There are at present no other features
+       controlled by this option. It can also be set by a (?X) option  setting
+       within a pattern.
+
+         PCRE_FIRSTLINE
+
+       If  this  option  is  set,  an  unanchored pattern is required to match
+       before or at the first  newline  in  the  subject  string,  though  the
+       matched text may continue over the newline.
+
+         PCRE_JAVASCRIPT_COMPAT
+
+       If this option is set, PCRE's behaviour is changed in some ways so that
+       it is compatible with JavaScript rather than Perl. The changes  are  as
+       follows:
+
+       (1)  A  lone  closing square bracket in a pattern causes a compile-time
+       error, because this is illegal in JavaScript (by default it is  treated
+       as a data character). Thus, the pattern AB]CD becomes illegal when this
+       option is set.
+
+       (2) At run time, a back reference to an unset subpattern group  matches
+       an  empty  string (by default this causes the current matching alterna-
+       tive to fail). A pattern such as (\1)(a) succeeds when this  option  is
+       set  (assuming  it can find an "a" in the subject), whereas it fails by
+       default, for Perl compatibility.
+
+       (3) \U matches an upper case "U" character; by default \U causes a com-
+       pile time error (Perl uses \U to upper case subsequent characters).
+
+       (4) \u matches a lower case "u" character unless it is followed by four
+       hexadecimal digits, in which case the hexadecimal  number  defines  the
+       code  point  to match. By default, \u causes a compile time error (Perl
+       uses it to upper case the following character).
+
+       (5) \x matches a lower case "x" character unless it is followed by  two
+       hexadecimal  digits,  in  which case the hexadecimal number defines the
+       code point to match. By default, as in Perl, a  hexadecimal  number  is
+       always expected after \x, but it may have zero, one, or two digits (so,
+       for example, \xz matches a binary zero character followed by z).
+
+         PCRE_MULTILINE
+
+       By default, for the purposes of matching "start of line"  and  "end  of
+       line", PCRE treats the subject string as consisting of a single line of
+       characters, even if it actually contains newlines. The "start of  line"
+       metacharacter (^) matches only at the start of the string, and the "end
+       of line" metacharacter ($) matches only at the end of  the  string,  or
+       before  a terminating newline (except when PCRE_DOLLAR_ENDONLY is set).
+       Note, however, that unless PCRE_DOTALL  is  set,  the  "any  character"
+       metacharacter  (.)  does not match at a newline. This behaviour (for ^,
+       $, and dot) is the same as Perl.
+
+       When PCRE_MULTILINE it is set, the "start of line" and  "end  of  line"
+       constructs  match  immediately following or immediately before internal
+       newlines in the subject string, respectively, as well as  at  the  very
+       start  and  end.  This is equivalent to Perl's /m option, and it can be
+       changed within a pattern by a (?m) option setting. If there are no new-
+       lines  in  a  subject string, or no occurrences of ^ or $ in a pattern,
+       setting PCRE_MULTILINE has no effect.
+
+         PCRE_NEVER_UTF
+
+       This option locks out interpretation of the pattern as UTF-8 (or UTF-16
+       or  UTF-32  in the 16-bit and 32-bit libraries). In particular, it pre-
+       vents the creator of the pattern from switching to  UTF  interpretation
+       by starting the pattern with (*UTF). This may be useful in applications
+       that  process  patterns  from  external  sources.  The  combination  of
+       PCRE_UTF8 and PCRE_NEVER_UTF also causes an error.
+
+         PCRE_NEWLINE_CR
+         PCRE_NEWLINE_LF
+         PCRE_NEWLINE_CRLF
+         PCRE_NEWLINE_ANYCRLF
+         PCRE_NEWLINE_ANY
+
+       These  options  override the default newline definition that was chosen
+       when PCRE was built. Setting the first or the second specifies  that  a
+       newline  is  indicated  by a single character (CR or LF, respectively).
+       Setting PCRE_NEWLINE_CRLF specifies that a newline is indicated by  the
+       two-character  CRLF  sequence.  Setting  PCRE_NEWLINE_ANYCRLF specifies
+       that any of the three preceding sequences should be recognized. Setting
+       PCRE_NEWLINE_ANY  specifies that any Unicode newline sequence should be
+       recognized.
+
+       In an ASCII/Unicode environment, the Unicode newline sequences are  the
+       three  just  mentioned,  plus  the  single characters VT (vertical tab,
+       U+000B), FF (form feed, U+000C), NEL (next line, U+0085), LS (line sep-
+       arator,  U+2028),  and  PS (paragraph separator, U+2029). For the 8-bit
+       library, the last two are recognized only in UTF-8 mode.
+
+       When PCRE is compiled to run in an EBCDIC (mainframe) environment,  the
+       code for CR is 0x0d, the same as ASCII. However, the character code for
+       LF is normally 0x15, though in some EBCDIC environments 0x25  is  used.
+       Whichever  of  these  is  not LF is made to correspond to Unicode's NEL
+       character. EBCDIC codes are all less than 256. For  more  details,  see
+       the pcrebuild documentation.
+
+       The  newline  setting  in  the  options  word  uses three bits that are
+       treated as a number, giving eight possibilities. Currently only six are
+       used  (default  plus the five values above). This means that if you set
+       more than one newline option, the combination may or may not be  sensi-
+       ble. For example, PCRE_NEWLINE_CR with PCRE_NEWLINE_LF is equivalent to
+       PCRE_NEWLINE_CRLF, but other combinations may yield unused numbers  and
+       cause an error.
+
+       The  only  time  that a line break in a pattern is specially recognized
+       when compiling is when PCRE_EXTENDED is set. CR and LF are white  space
+       characters,  and so are ignored in this mode. Also, an unescaped # out-
+       side a character class indicates a comment that lasts until  after  the
+       next  line break sequence. In other circumstances, line break sequences
+       in patterns are treated as literal data.
+
+       The newline option that is set at compile time becomes the default that
+       is used for pcre_exec() and pcre_dfa_exec(), but it can be overridden.
+
+         PCRE_NO_AUTO_CAPTURE
+
+       If this option is set, it disables the use of numbered capturing paren-
+       theses in the pattern. Any opening parenthesis that is not followed  by
+       ?  behaves as if it were followed by ?: but named parentheses can still
+       be used for capturing (and they acquire  numbers  in  the  usual  way).
+       There is no equivalent of this option in Perl.
+
+         PCRE_NO_START_OPTIMIZE
+
+       This  is an option that acts at matching time; that is, it is really an
+       option for pcre_exec() or pcre_dfa_exec(). If  it  is  set  at  compile
+       time,  it is remembered with the compiled pattern and assumed at match-
+       ing time. This is necessary if you want to use JIT  execution,  because
+       the  JIT  compiler needs to know whether or not this option is set. For
+       details see the discussion of PCRE_NO_START_OPTIMIZE below.
+
+         PCRE_UCP
+
+       This option changes the way PCRE processes \B, \b, \D, \d, \S, \s,  \W,
+       \w,  and  some  of  the POSIX character classes. By default, only ASCII
+       characters are recognized, but if PCRE_UCP is set,  Unicode  properties
+       are  used instead to classify characters. More details are given in the
+       section on generic character types in the pcrepattern page. If you  set
+       PCRE_UCP,  matching  one of the items it affects takes much longer. The
+       option is available only if PCRE has been compiled with  Unicode  prop-
+       erty support.
+
+         PCRE_UNGREEDY
+
+       This  option  inverts  the "greediness" of the quantifiers so that they
+       are not greedy by default, but become greedy if followed by "?". It  is
+       not  compatible  with Perl. It can also be set by a (?U) option setting
+       within the pattern.
+
+         PCRE_UTF8
+
+       This option causes PCRE to regard both the pattern and the  subject  as
+       strings of UTF-8 characters instead of single-byte strings. However, it
+       is available only when PCRE is built to include UTF  support.  If  not,
+       the  use  of  this option provokes an error. Details of how this option
+       changes the behaviour of PCRE are given in the pcreunicode page.
+
+         PCRE_NO_UTF8_CHECK
+
+       When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is
+       automatically  checked.  There  is  a  discussion about the validity of
+       UTF-8 strings in the pcreunicode page. If an invalid UTF-8 sequence  is
+       found,  pcre_compile()  returns an error. If you already know that your
+       pattern is valid, and you want to skip this check for performance  rea-
+       sons,  you  can set the PCRE_NO_UTF8_CHECK option.  When it is set, the
+       effect of passing an invalid UTF-8 string as a pattern is undefined. It
+       may  cause  your  program  to  crash. Note that this option can also be
+       passed to pcre_exec() and pcre_dfa_exec(),  to  suppress  the  validity
+       checking  of  subject strings only. If the same string is being matched
+       many times, the option can be safely set for the second and  subsequent
+       matchings to improve performance.
+
+
+COMPILATION ERROR CODES
+
+       The  following  table  lists  the  error  codes than may be returned by
+       pcre_compile2(), along with the error messages that may be returned  by
+       both  compiling  functions.  Note  that error messages are always 8-bit
+       ASCII strings, even in 16-bit or 32-bit mode. As  PCRE  has  developed,
+       some  error codes have fallen out of use. To avoid confusion, they have
+       not been re-used.
+
+          0  no error
+          1  \ at end of pattern
+          2  \c at end of pattern
+          3  unrecognized character follows \
+          4  numbers out of order in {} quantifier
+          5  number too big in {} quantifier
+          6  missing terminating ] for character class
+          7  invalid escape sequence in character class
+          8  range out of order in character class
+          9  nothing to repeat
+         10  [this code is not in use]
+         11  internal error: unexpected repeat
+         12  unrecognized character after (? or (?-
+         13  POSIX named classes are supported only within a class
+         14  missing )
+         15  reference to non-existent subpattern
+         16  erroffset passed as NULL
+         17  unknown option bit(s) set
+         18  missing ) after comment
+         19  [this code is not in use]
+         20  regular expression is too large
+         21  failed to get memory
+         22  unmatched parentheses
+         23  internal error: code overflow
+         24  unrecognized character after (?<
+         25  lookbehind assertion is not fixed length
+         26  malformed number or name after (?(
+         27  conditional group contains more than two branches
+         28  assertion expected after (?(
+         29  (?R or (?[+-]digits must be followed by )
+         30  unknown POSIX class name
+         31  POSIX collating elements are not supported
+         32  this version of PCRE is compiled without UTF support
+         33  [this code is not in use]
+         34  character value in \x{...} sequence is too large
+         35  invalid condition (?(0)
+         36  \C not allowed in lookbehind assertion
+         37  PCRE does not support \L, \l, \N{name}, \U, or \u
+         38  number after (?C is > 255
+         39  closing ) for (?C expected
+         40  recursive call could loop indefinitely
+         41  unrecognized character after (?P
+         42  syntax error in subpattern name (missing terminator)
+         43  two named subpatterns have the same name
+         44  invalid UTF-8 string (specifically UTF-8)
+         45  support for \P, \p, and \X has not been compiled
+         46  malformed \P or \p sequence
+         47  unknown property name after \P or \p
+         48  subpattern name is too long (maximum 32 characters)
+         49  too many named subpatterns (maximum 10000)
+         50  [this code is not in use]
+         51  octal value is greater than \377 in 8-bit non-UTF-8 mode
+         52  internal error: overran compiling workspace
+         53  internal error: previously-checked referenced subpattern
+               not found
+         54  DEFINE group contains more than one branch
+         55  repeating a DEFINE group is not allowed
+         56  inconsistent NEWLINE options
+         57  \g is not followed by a braced, angle-bracketed, or quoted
+               name/number or by a plain number
+         58  a numbered reference must not be zero
+         59  an argument is not allowed for (*ACCEPT), (*FAIL), or (*COMMIT)
+         60  (*VERB) not recognized or malformed
+         61  number is too big
+         62  subpattern name expected
+         63  digit expected after (?+
+         64  ] is an invalid data character in JavaScript compatibility mode
+         65  different names for subpatterns of the same number are
+               not allowed
+         66  (*MARK) must have an argument
+         67  this version of PCRE is not compiled with Unicode property
+               support
+         68  \c must be followed by an ASCII character
+         69  \k is not followed by a braced, angle-bracketed, or quoted name
+         70  internal error: unknown opcode in find_fixedlength()
+         71  \N is not supported in a class
+         72  too many forward references
+         73  disallowed Unicode code point (>= 0xd800 && <= 0xdfff)
+         74  invalid UTF-16 string (specifically UTF-16)
+         75  name is too long in (*MARK), (*PRUNE), (*SKIP), or (*THEN)
+         76  character value in \u.... sequence is too large
+         77  invalid UTF-32 string (specifically UTF-32)
+
+       The numbers 32 and 10000 in errors 48 and 49  are  defaults;  different
+       values may be used if the limits were changed when PCRE was built.
+
+
+STUDYING A PATTERN
+
+       pcre_extra *pcre_study(const pcre *code, int options
+            const char **errptr);
+
+       If  a  compiled  pattern is going to be used several times, it is worth
+       spending more time analyzing it in order to speed up the time taken for
+       matching.  The function pcre_study() takes a pointer to a compiled pat-
+       tern as its first argument. If studying the pattern produces additional
+       information  that  will  help speed up matching, pcre_study() returns a
+       pointer to a pcre_extra block, in which the study_data field points  to
+       the results of the study.
+
+       The  returned  value  from  pcre_study()  can  be  passed  directly  to
+       pcre_exec() or pcre_dfa_exec(). However, a pcre_extra block  also  con-
+       tains  other  fields  that can be set by the caller before the block is
+       passed; these are described below in the section on matching a pattern.
+
+       If studying the  pattern  does  not  produce  any  useful  information,
+       pcre_study()  returns  NULL  by  default.  In that circumstance, if the
+       calling program wants to pass any of the other fields to pcre_exec() or
+       pcre_dfa_exec(),  it  must set up its own pcre_extra block. However, if
+       pcre_study() is called  with  the  PCRE_STUDY_EXTRA_NEEDED  option,  it
+       returns a pcre_extra block even if studying did not find any additional
+       information. It may still return NULL, however, if an error  occurs  in
+       pcre_study().
+
+       The  second  argument  of  pcre_study() contains option bits. There are
+       three further options in addition to PCRE_STUDY_EXTRA_NEEDED:
+
+         PCRE_STUDY_JIT_COMPILE
+         PCRE_STUDY_JIT_PARTIAL_HARD_COMPILE
+         PCRE_STUDY_JIT_PARTIAL_SOFT_COMPILE
+
+       If any of these are set, and the just-in-time  compiler  is  available,
+       the  pattern  is  further compiled into machine code that executes much
+       faster than the pcre_exec()  interpretive  matching  function.  If  the
+       just-in-time  compiler is not available, these options are ignored. All
+       undefined bits in the options argument must be zero.
+
+       JIT compilation is a heavyweight optimization. It can  take  some  time
+       for  patterns  to  be analyzed, and for one-off matches and simple pat-
+       terns the benefit of faster execution might be offset by a much  slower
+       study time.  Not all patterns can be optimized by the JIT compiler. For
+       those that cannot be handled, matching automatically falls back to  the
+       pcre_exec()  interpreter.  For more details, see the pcrejit documenta-
+       tion.
+
+       The third argument for pcre_study() is a pointer for an error  message.
+       If  studying  succeeds  (even  if no data is returned), the variable it
+       points to is set to NULL. Otherwise it is set to  point  to  a  textual
+       error message. This is a static string that is part of the library. You
+       must not try to free it. You should test the  error  pointer  for  NULL
+       after calling pcre_study(), to be sure that it has run successfully.
+
+       When  you are finished with a pattern, you can free the memory used for
+       the study data by calling pcre_free_study(). This function was added to
+       the  API  for  release  8.20. For earlier versions, the memory could be
+       freed with pcre_free(), just like the pattern itself. This  will  still
+       work  in  cases where JIT optimization is not used, but it is advisable
+       to change to the new function when convenient.
+
+       This is a typical way in which pcre_study() is used (except that  in  a
+       real application there should be tests for errors):
+
+         int rc;
+         pcre *re;
+         pcre_extra *sd;
+         re = pcre_compile("pattern", 0, &error, &erroroffset, NULL);
+         sd = pcre_study(
+           re,             /* result of pcre_compile() */
+           0,              /* no options */
+           &error);        /* set to NULL or points to a message */
+         rc = pcre_exec(   /* see below for details of pcre_exec() options */
+           re, sd, "subject", 7, 0, 0, ovector, 30);
+         ...
+         pcre_free_study(sd);
+         pcre_free(re);
+
+       Studying a pattern does two things: first, a lower bound for the length
+       of subject string that is needed to match the pattern is computed. This
+       does not mean that there are any strings of that length that match, but
+       it does guarantee that no shorter strings match. The value is  used  to
+       avoid wasting time by trying to match strings that are shorter than the
+       lower bound. You can find out the value in a calling  program  via  the
+       pcre_fullinfo() function.
+
+       Studying a pattern is also useful for non-anchored patterns that do not
+       have a single fixed starting character. A bitmap of  possible  starting
+       bytes  is  created. This speeds up finding a position in the subject at
+       which to start matching. (In 16-bit mode, the bitmap is used for 16-bit
+       values  less  than  256.  In 32-bit mode, the bitmap is used for 32-bit
+       values less than 256.)
+
+       These two optimizations apply to both pcre_exec() and  pcre_dfa_exec(),
+       and  the  information  is also used by the JIT compiler.  The optimiza-
+       tions can be disabled by  setting  the  PCRE_NO_START_OPTIMIZE  option.
+       You  might want to do this if your pattern contains callouts or (*MARK)
+       and you want to make use of these facilities in  cases  where  matching
+       fails.
+
+       PCRE_NO_START_OPTIMIZE  can be specified at either compile time or exe-
+       cution  time.  However,  if   PCRE_NO_START_OPTIMIZE   is   passed   to
+       pcre_exec(), (that is, after any JIT compilation has happened) JIT exe-
+       cution is disabled. For JIT execution to work with  PCRE_NO_START_OPTI-
+       MIZE, the option must be set at compile time.
+
+       There is a longer discussion of PCRE_NO_START_OPTIMIZE below.
+
+
+LOCALE SUPPORT
+
+       PCRE  handles  caseless matching, and determines whether characters are
+       letters, digits, or whatever, by reference to a set of tables,  indexed
+       by  character  value.  When running in UTF-8 mode, this applies only to
+       characters with codes less than 128. By  default,  higher-valued  codes
+       never match escapes such as \w or \d, but they can be tested with \p if
+       PCRE is built with Unicode character property  support.  Alternatively,
+       the  PCRE_UCP  option  can  be  set at compile time; this causes \w and
+       friends to use Unicode property support instead of built-in tables. The
+       use of locales with Unicode is discouraged. If you are handling charac-
+       ters with codes greater than 128, you should either use UTF-8 and  Uni-
+       code, or use locales, but not try to mix the two.
+
+       PCRE  contains  an  internal set of tables that are used when the final
+       argument of pcre_compile() is  NULL.  These  are  sufficient  for  many
+       applications.  Normally, the internal tables recognize only ASCII char-
+       acters. However, when PCRE is built, it is possible to cause the inter-
+       nal tables to be rebuilt in the default "C" locale of the local system,
+       which may cause them to be different.
+
+       The internal tables can always be overridden by tables supplied by  the
+       application that calls PCRE. These may be created in a different locale
+       from the default. As more and more applications change  to  using  Uni-
+       code, the need for this locale support is expected to die away.
+
+       External  tables  are  built by calling the pcre_maketables() function,
+       which has no arguments, in the relevant locale. The result can then  be
+       passed  to  pcre_compile()  or  pcre_exec()  as often as necessary. For
+       example, to build and use tables that are appropriate  for  the  French
+       locale  (where  accented  characters  with  values greater than 128 are
+       treated as letters), the following code could be used:
+
+         setlocale(LC_CTYPE, "fr_FR");
+         tables = pcre_maketables();
+         re = pcre_compile(..., tables);
+
+       The locale name "fr_FR" is used on Linux and other  Unix-like  systems;
+       if you are using Windows, the name for the French locale is "french".
+
+       When  pcre_maketables()  runs,  the  tables are built in memory that is
+       obtained via pcre_malloc. It is the caller's responsibility  to  ensure
+       that  the memory containing the tables remains available for as long as
+       it is needed.
+
+       The pointer that is passed to pcre_compile() is saved with the compiled
+       pattern,  and the same tables are used via this pointer by pcre_study()
+       and normally also by pcre_exec(). Thus, by default, for any single pat-
+       tern, compilation, studying and matching all happen in the same locale,
+       but different patterns can be compiled in different locales.
+
+       It is possible to pass a table pointer or NULL (indicating the  use  of
+       the  internal  tables)  to  pcre_exec(). Although not intended for this
+       purpose, this facility could be used to match a pattern in a  different
+       locale from the one in which it was compiled. Passing table pointers at
+       run time is discussed below in the section on matching a pattern.
+
+
+INFORMATION ABOUT A PATTERN
+
+       int pcre_fullinfo(const pcre *code, const pcre_extra *extra,
+            int what, void *where);
+
+       The pcre_fullinfo() function returns information about a compiled  pat-
+       tern.  It replaces the pcre_info() function, which was removed from the
+       library at version 8.30, after more than 10 years of obsolescence.
+
+       The first argument for pcre_fullinfo() is a  pointer  to  the  compiled
+       pattern.  The second argument is the result of pcre_study(), or NULL if
+       the pattern was not studied. The third argument specifies  which  piece
+       of  information  is required, and the fourth argument is a pointer to a
+       variable to receive the data. The yield of the  function  is  zero  for
+       success, or one of the following negative numbers:
+
+         PCRE_ERROR_NULL           the argument code was NULL
+                                   the argument where was NULL
+         PCRE_ERROR_BADMAGIC       the "magic number" was not found
+         PCRE_ERROR_BADENDIANNESS  the pattern was compiled with different
+                                   endianness
+         PCRE_ERROR_BADOPTION      the value of what was invalid
+         PCRE_ERROR_UNSET          the requested field is not set
+
+       The  "magic  number" is placed at the start of each compiled pattern as
+       an simple check against passing an arbitrary memory pointer. The  endi-
+       anness error can occur if a compiled pattern is saved and reloaded on a
+       different host. Here is a typical call of  pcre_fullinfo(),  to  obtain
+       the length of the compiled pattern:
+
+         int rc;
+         size_t length;
+         rc = pcre_fullinfo(
+           re,               /* result of pcre_compile() */
+           sd,               /* result of pcre_study(), or NULL */
+           PCRE_INFO_SIZE,   /* what is required */
+           &length);         /* where to put the data */
+
+       The  possible  values for the third argument are defined in pcre.h, and
+       are as follows:
+
+         PCRE_INFO_BACKREFMAX
+
+       Return the number of the highest back reference  in  the  pattern.  The
+       fourth  argument  should  point to an int variable. Zero is returned if
+       there are no back references.
+
+         PCRE_INFO_CAPTURECOUNT
+
+       Return the number of capturing subpatterns in the pattern.  The  fourth
+       argument should point to an int variable.
+
+         PCRE_INFO_DEFAULT_TABLES
+
+       Return  a pointer to the internal default character tables within PCRE.
+       The fourth argument should point to an unsigned char *  variable.  This
+       information call is provided for internal use by the pcre_study() func-
+       tion. External callers can cause PCRE to use  its  internal  tables  by
+       passing a NULL table pointer.
+
+         PCRE_INFO_FIRSTBYTE
+
+       Return information about the first data unit of any matched string, for
+       a non-anchored pattern. (The name of this option refers  to  the  8-bit
+       library,  where data units are bytes.) The fourth argument should point
+       to an int variable.
+
+       If there is a fixed first value, for example, the  letter  "c"  from  a
+       pattern  such  as (cat|cow|coyote), its value is returned. In the 8-bit
+       library, the value is always less than 256. In the 16-bit  library  the
+       value can be up to 0xffff. In the 32-bit library the value can be up to
+       0x10ffff.
+
+       If there is no fixed first value, and if either
+
+       (a) the pattern was compiled with the PCRE_MULTILINE option, and  every
+       branch starts with "^", or
+
+       (b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not
+       set (if it were set, the pattern would be anchored),
+
+       -1 is returned, indicating that the pattern matches only at  the  start
+       of  a  subject string or after any newline within the string. Otherwise
+       -2 is returned. For anchored patterns, -2 is returned.
+
+       Since for the 32-bit library using the non-UTF-32 mode,  this  function
+       is  unable to return the full 32-bit range of the character, this value
+       is   deprecated;   instead   the   PCRE_INFO_FIRSTCHARACTERFLAGS    and
+       PCRE_INFO_FIRSTCHARACTER values should be used.
+
+         PCRE_INFO_FIRSTTABLE
+
+       If  the pattern was studied, and this resulted in the construction of a
+       256-bit table indicating a fixed set of values for the first data  unit
+       in  any  matching string, a pointer to the table is returned. Otherwise
+       NULL is returned. The fourth argument should point to an unsigned  char
+       * variable.
+
+         PCRE_INFO_HASCRORLF
+
+       Return  1  if  the  pattern  contains any explicit matches for CR or LF
+       characters, otherwise 0. The fourth argument should  point  to  an  int
+       variable.  An explicit match is either a literal CR or LF character, or
+       \r or \n.
+
+         PCRE_INFO_JCHANGED
+
+       Return 1 if the (?J) or (?-J) option setting is used  in  the  pattern,
+       otherwise  0. The fourth argument should point to an int variable. (?J)
+       and (?-J) set and unset the local PCRE_DUPNAMES option, respectively.
+
+         PCRE_INFO_JIT
+
+       Return 1 if the pattern was studied with one of the  JIT  options,  and
+       just-in-time compiling was successful. The fourth argument should point
+       to an int variable. A return value of 0 means that JIT support  is  not
+       available  in this version of PCRE, or that the pattern was not studied
+       with a JIT option, or that the JIT compiler could not handle this  par-
+       ticular  pattern. See the pcrejit documentation for details of what can
+       and cannot be handled.
+
+         PCRE_INFO_JITSIZE
+
+       If the pattern was successfully studied with a JIT option,  return  the
+       size  of the JIT compiled code, otherwise return zero. The fourth argu-
+       ment should point to a size_t variable.
+
+         PCRE_INFO_LASTLITERAL
+
+       Return the value of the rightmost literal data unit that must exist  in
+       any  matched  string, other than at its start, if such a value has been
+       recorded. The fourth argument should point to an int variable. If there
+       is no such value, -1 is returned. For anchored patterns, a last literal
+       value is recorded only if it follows something of variable length.  For
+       example, for the pattern /^a\d+z\d+/ the returned value is "z", but for
+       /^a\dz\d/ the returned value is -1.
+
+       Since for the 32-bit library using the non-UTF-32 mode,  this  function
+       is  unable to return the full 32-bit range of the character, this value
+       is   deprecated;   instead    the    PCRE_INFO_REQUIREDCHARFLAGS    and
+       PCRE_INFO_REQUIREDCHAR values should be used.
+
+         PCRE_INFO_MATCHLIMIT
+
+       If  the  pattern  set  a  match  limit by including an item of the form
+       (*LIMIT_MATCH=nnnn) at the start, the value  is  returned.  The  fourth
+       argument  should  point to an unsigned 32-bit integer. If no such value
+       has  been  set,  the  call  to  pcre_fullinfo()   returns   the   error
+       PCRE_ERROR_UNSET.
+
+         PCRE_INFO_MAXLOOKBEHIND
+
+       Return  the  number  of  characters  (NB not data units) in the longest
+       lookbehind assertion in the pattern. This information  is  useful  when
+       doing  multi-segment  matching  using  the partial matching facilities.
+       Note that the simple assertions \b and \B require a one-character look-
+       behind.  \A  also  registers a one-character lookbehind, though it does
+       not actually inspect the previous character. This is to ensure that  at
+       least one character from the old segment is retained when a new segment
+       is processed. Otherwise, if there are no lookbehinds in the pattern, \A
+       might match incorrectly at the start of a new segment.
+
+         PCRE_INFO_MINLENGTH
+
+       If  the  pattern  was studied and a minimum length for matching subject
+       strings was computed, its value is  returned.  Otherwise  the  returned
+       value is -1. The value is a number of characters, which in UTF mode may
+       be different from the number of data units. The fourth argument  should
+       point  to an int variable. A non-negative value is a lower bound to the
+       length of any matching string. There may not be  any  strings  of  that
+       length  that  do actually match, but every string that does match is at
+       least that long.
+
+         PCRE_INFO_NAMECOUNT
+         PCRE_INFO_NAMEENTRYSIZE
+         PCRE_INFO_NAMETABLE
+
+       PCRE supports the use of named as well as numbered capturing  parenthe-
+       ses.  The names are just an additional way of identifying the parenthe-
+       ses, which still acquire numbers. Several convenience functions such as
+       pcre_get_named_substring()  are  provided  for extracting captured sub-
+       strings by name. It is also possible to extract the data  directly,  by
+       first  converting  the  name to a number in order to access the correct
+       pointers in the output vector (described with pcre_exec() below). To do
+       the  conversion,  you  need  to  use  the  name-to-number map, which is
+       described by these three values.
+
+       The map consists of a number of fixed-size entries. PCRE_INFO_NAMECOUNT
+       gives the number of entries, and PCRE_INFO_NAMEENTRYSIZE gives the size
+       of each entry; both of these  return  an  int  value.  The  entry  size
+       depends  on the length of the longest name. PCRE_INFO_NAMETABLE returns
+       a pointer to the first entry of the table. This is a pointer to char in
+       the 8-bit library, where the first two bytes of each entry are the num-
+       ber of the capturing parenthesis, most significant byte first.  In  the
+       16-bit  library,  the pointer points to 16-bit data units, the first of
+       which contains the parenthesis  number.  In  the  32-bit  library,  the
+       pointer  points  to  32-bit data units, the first of which contains the
+       parenthesis number. The rest of the entry is  the  corresponding  name,
+       zero terminated.
+
+       The  names are in alphabetical order. Duplicate names may appear if (?|
+       is used to create multiple groups with the same number, as described in
+       the  section  on  duplicate subpattern numbers in the pcrepattern page.
+       Duplicate names for subpatterns with different  numbers  are  permitted
+       only  if  PCRE_DUPNAMES  is  set. In all cases of duplicate names, they
+       appear in the table in the order in which they were found in  the  pat-
+       tern.  In  the  absence  of (?| this is the order of increasing number;
+       when (?| is used this is not necessarily the case because later subpat-
+       terns may have lower numbers.
+
+       As  a  simple  example of the name/number table, consider the following
+       pattern after compilation by the 8-bit library (assume PCRE_EXTENDED is
+       set, so white space - including newlines - is ignored):
+
+         (? (?(\d\d)?\d\d) -
+         (?\d\d) - (?\d\d) )
+
+       There  are  four  named subpatterns, so the table has four entries, and
+       each entry in the table is eight bytes long. The table is  as  follows,
+       with non-printing bytes shows in hexadecimal, and undefined bytes shown
+       as ??:
+
+         00 01 d  a  t  e  00 ??
+         00 05 d  a  y  00 ?? ??
+         00 04 m  o  n  t  h  00
+         00 02 y  e  a  r  00 ??
+
+       When writing code to extract data  from  named  subpatterns  using  the
+       name-to-number  map,  remember that the length of the entries is likely
+       to be different for each compiled pattern.
+
+         PCRE_INFO_OKPARTIAL
+
+       Return 1  if  the  pattern  can  be  used  for  partial  matching  with
+       pcre_exec(),  otherwise  0.  The fourth argument should point to an int
+       variable. From  release  8.00,  this  always  returns  1,  because  the
+       restrictions  that  previously  applied  to  partial matching have been
+       lifted. The pcrepartial documentation gives details of  partial  match-
+       ing.
+
+         PCRE_INFO_OPTIONS
+
+       Return  a  copy of the options with which the pattern was compiled. The
+       fourth argument should point to an unsigned long  int  variable.  These
+       option bits are those specified in the call to pcre_compile(), modified
+       by any top-level option settings at the start of the pattern itself. In
+       other  words,  they are the options that will be in force when matching
+       starts. For example, if the pattern /(?im)abc(?-i)d/ is  compiled  with
+       the  PCRE_EXTENDED option, the result is PCRE_CASELESS, PCRE_MULTILINE,
+       and PCRE_EXTENDED.
+
+       A pattern is automatically anchored by PCRE if  all  of  its  top-level
+       alternatives begin with one of the following:
+
+         ^     unless PCRE_MULTILINE is set
+         \A    always
+         \G    always
+         .*    if PCRE_DOTALL is set and there are no back
+                 references to the subpattern in which .* appears
+
+       For such patterns, the PCRE_ANCHORED bit is set in the options returned
+       by pcre_fullinfo().
+
+         PCRE_INFO_RECURSIONLIMIT
+
+       If the pattern set a recursion limit by including an item of  the  form
+       (*LIMIT_RECURSION=nnnn) at the start, the value is returned. The fourth
+       argument should point to an unsigned 32-bit integer. If no  such  value
+       has   been   set,   the  call  to  pcre_fullinfo()  returns  the  error
+       PCRE_ERROR_UNSET.
+
+         PCRE_INFO_SIZE
+
+       Return the size of  the  compiled  pattern  in  bytes  (for  all  three
+       libraries). The fourth argument should point to a size_t variable. This
+       value does not include the size of the pcre structure that is  returned
+       by  pcre_compile().  The  value  that  is  passed  as  the  argument to
+       pcre_malloc() when pcre_compile() is getting memory in which  to  place
+       the compiled data is the value returned by this option plus the size of
+       the pcre structure. Studying a compiled pattern, with or  without  JIT,
+       does not alter the value returned by this option.
+
+         PCRE_INFO_STUDYSIZE
+
+       Return  the  size  in bytes (for all three libraries) of the data block
+       pointed to by the study_data field in a pcre_extra block. If pcre_extra
+       is  NULL, or there is no study data, zero is returned. The fourth argu-
+       ment should point to a size_t variable. The study_data field is set  by
+       pcre_study() to record information that will speed up matching (see the
+       section entitled  "Studying  a  pattern"  above).  The  format  of  the
+       study_data  block is private, but its length is made available via this
+       option so that it can be saved and  restored  (see  the  pcreprecompile
+       documentation for details).
+
+         PCRE_INFO_FIRSTCHARACTERFLAGS
+
+       Return information about the first data unit of any matched string, for
+       a non-anchored pattern. The fourth argument  should  point  to  an  int
+       variable.
+
+       If  there  is  a  fixed first value, for example, the letter "c" from a
+       pattern such as (cat|cow|coyote), 1  is  returned,  and  the  character
+       value can be retrieved using PCRE_INFO_FIRSTCHARACTER.
+
+       If there is no fixed first value, and if either
+
+       (a)  the pattern was compiled with the PCRE_MULTILINE option, and every
+       branch starts with "^", or
+
+       (b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not
+       set (if it were set, the pattern would be anchored),
+
+       2 is returned, indicating that the pattern matches only at the start of
+       a subject string or after any newline within the string. Otherwise 0 is
+       returned. For anchored patterns, 0 is returned.
+
+         PCRE_INFO_FIRSTCHARACTER
+
+       Return  the  fixed  first character value, if PCRE_INFO_FIRSTCHARACTER-
+       FLAGS returned 1; otherwise returns 0. The fourth argument should point
+       to an uint_t variable.
+
+       In  the 8-bit library, the value is always less than 256. In the 16-bit
+       library the value can be up to 0xffff. In the 32-bit library in  UTF-32
+       mode  the  value  can  be up to 0x10ffff, and up to 0xffffffff when not
+       using UTF-32 mode.
+
+       If there is no fixed first value, and if either
+
+       (a) the pattern was compiled with the PCRE_MULTILINE option, and  every
+       branch starts with "^", or
+
+       (b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not
+       set (if it were set, the pattern would be anchored),
+
+       -1 is returned, indicating that the pattern matches only at  the  start
+       of  a  subject string or after any newline within the string. Otherwise
+       -2 is returned. For anchored patterns, -2 is returned.
+
+         PCRE_INFO_REQUIREDCHARFLAGS
+
+       Returns 1 if there is a rightmost literal data unit that must exist  in
+       any matched string, other than at its start. The fourth argument should
+       point to an int variable. If there is no such value, 0 is returned.  If
+       returning  1,  the  character  value  itself  can  be  retrieved  using
+       PCRE_INFO_REQUIREDCHAR.
+
+       For anchored patterns, a last literal value is recorded only if it fol-
+       lows  something  of  variable  length.  For  example,  for  the pattern
+       /^a\d+z\d+/  the   returned   value   1   (with   "z"   returned   from
+       PCRE_INFO_REQUIREDCHAR), but for /^a\dz\d/ the returned value is 0.
+
+         PCRE_INFO_REQUIREDCHAR
+
+       Return  the value of the rightmost literal data unit that must exist in
+       any matched string, other than at its start, if such a value  has  been
+       recorded.  The fourth argument should point to an uint32_t variable. If
+       there is no such value, 0 is returned.
+
+
+REFERENCE COUNTS
+
+       int pcre_refcount(pcre *code, int adjust);
+
+       The pcre_refcount() function is used to maintain a reference  count  in
+       the data block that contains a compiled pattern. It is provided for the
+       benefit of applications that  operate  in  an  object-oriented  manner,
+       where different parts of the application may be using the same compiled
+       pattern, but you want to free the block when they are all done.
+
+       When a pattern is compiled, the reference count field is initialized to
+       zero.   It is changed only by calling this function, whose action is to
+       add the adjust value (which may be positive or  negative)  to  it.  The
+       yield of the function is the new value. However, the value of the count
+       is constrained to lie between 0 and 65535, inclusive. If the new  value
+       is outside these limits, it is forced to the appropriate limit value.
+
+       Except  when it is zero, the reference count is not correctly preserved
+       if a pattern is compiled on one host and then  transferred  to  a  host
+       whose byte-order is different. (This seems a highly unlikely scenario.)
+
+
+MATCHING A PATTERN: THE TRADITIONAL FUNCTION
+
+       int pcre_exec(const pcre *code, const pcre_extra *extra,
+            const char *subject, int length, int startoffset,
+            int options, int *ovector, int ovecsize);
+
+       The  function pcre_exec() is called to match a subject string against a
+       compiled pattern, which is passed in the code argument. If the  pattern
+       was  studied,  the  result  of  the study should be passed in the extra
+       argument. You can call pcre_exec() with the same code and  extra  argu-
+       ments  as  many  times as you like, in order to match different subject
+       strings with the same pattern.
+
+       This function is the main matching facility  of  the  library,  and  it
+       operates  in  a  Perl-like  manner. For specialist use there is also an
+       alternative matching function, which is described below in the  section
+       about the pcre_dfa_exec() function.
+
+       In  most applications, the pattern will have been compiled (and option-
+       ally studied) in the same process that calls pcre_exec().  However,  it
+       is possible to save compiled patterns and study data, and then use them
+       later in different processes, possibly even on different hosts.  For  a
+       discussion about this, see the pcreprecompile documentation.
+
+       Here is an example of a simple call to pcre_exec():
+
+         int rc;
+         int ovector[30];
+         rc = pcre_exec(
+           re,             /* result of pcre_compile() */
+           NULL,           /* we didn't study the pattern */
+           "some string",  /* the subject string */
+           11,             /* the length of the subject string */
+           0,              /* start at offset 0 in the subject */
+           0,              /* default options */
+           ovector,        /* vector of integers for substring information */
+           30);            /* number of elements (NOT size in bytes) */
+
+   Extra data for pcre_exec()
+
+       If  the  extra argument is not NULL, it must point to a pcre_extra data
+       block. The pcre_study() function returns such a block (when it  doesn't
+       return  NULL), but you can also create one for yourself, and pass addi-
+       tional information in it. The pcre_extra block contains  the  following
+       fields (not necessarily in this order):
+
+         unsigned long int flags;
+         void *study_data;
+         void *executable_jit;
+         unsigned long int match_limit;
+         unsigned long int match_limit_recursion;
+         void *callout_data;
+         const unsigned char *tables;
+         unsigned char **mark;
+
+       In  the  16-bit  version  of  this  structure,  the mark field has type
+       "PCRE_UCHAR16 **".
+
+       In the 32-bit version of  this  structure,  the  mark  field  has  type
+       "PCRE_UCHAR32 **".
+
+       The  flags  field is used to specify which of the other fields are set.
+       The flag bits are:
+
+         PCRE_EXTRA_CALLOUT_DATA
+         PCRE_EXTRA_EXECUTABLE_JIT
+         PCRE_EXTRA_MARK
+         PCRE_EXTRA_MATCH_LIMIT
+         PCRE_EXTRA_MATCH_LIMIT_RECURSION
+         PCRE_EXTRA_STUDY_DATA
+         PCRE_EXTRA_TABLES
+
+       Other flag bits should be set to zero. The study_data field  and  some-
+       times  the executable_jit field are set in the pcre_extra block that is
+       returned by pcre_study(), together with the appropriate flag bits.  You
+       should  not set these yourself, but you may add to the block by setting
+       other fields and their corresponding flag bits.
+
+       The match_limit field provides a means of preventing PCRE from using up
+       a  vast amount of resources when running patterns that are not going to
+       match, but which have a very large number  of  possibilities  in  their
+       search  trees. The classic example is a pattern that uses nested unlim-
+       ited repeats.
+
+       Internally, pcre_exec() uses a function called match(), which it  calls
+       repeatedly  (sometimes  recursively).  The  limit set by match_limit is
+       imposed on the number of times this function is called during a  match,
+       which  has  the  effect of limiting the amount of backtracking that can
+       take place. For patterns that are not anchored, the count restarts from
+       zero for each position in the subject string.
+
+       When pcre_exec() is called with a pattern that was successfully studied
+       with a JIT option, the way that the matching is  executed  is  entirely
+       different.  However, there is still the possibility of runaway matching
+       that goes on for a very long time, and so the match_limit value is also
+       used in this case (but in a different way) to limit how long the match-
+       ing can continue.
+
+       The default value for the limit can be set  when  PCRE  is  built;  the
+       default  default  is 10 million, which handles all but the most extreme
+       cases. You can override the default  by  suppling  pcre_exec()  with  a
+       pcre_extra     block    in    which    match_limit    is    set,    and
+       PCRE_EXTRA_MATCH_LIMIT is set in the  flags  field.  If  the  limit  is
+       exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
+
+       A  value  for  the  match  limit may also be supplied by an item at the
+       start of a pattern of the form
+
+         (*LIMIT_MATCH=d)
+
+       where d is a decimal number. However, such a setting is ignored  unless
+       d  is  less  than  the limit set by the caller of pcre_exec() or, if no
+       such limit is set, less than the default.
+
+       The match_limit_recursion field is similar to match_limit, but  instead
+       of limiting the total number of times that match() is called, it limits
+       the depth of recursion. The recursion depth is a  smaller  number  than
+       the  total number of calls, because not all calls to match() are recur-
+       sive.  This limit is of use only if it is set smaller than match_limit.
+
+       Limiting the recursion depth limits the amount of  machine  stack  that
+       can  be used, or, when PCRE has been compiled to use memory on the heap
+       instead of the stack, the amount of heap memory that can be used.  This
+       limit  is not relevant, and is ignored, when matching is done using JIT
+       compiled code.
+
+       The default value for match_limit_recursion can be  set  when  PCRE  is
+       built;  the  default  default  is  the  same  value  as the default for
+       match_limit. You can override the default by suppling pcre_exec()  with
+       a   pcre_extra   block  in  which  match_limit_recursion  is  set,  and
+       PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in  the  flags  field.  If  the
+       limit is exceeded, pcre_exec() returns PCRE_ERROR_RECURSIONLIMIT.
+
+       A  value for the recursion limit may also be supplied by an item at the
+       start of a pattern of the form
+
+         (*LIMIT_RECURSION=d)
+
+       where d is a decimal number. However, such a setting is ignored  unless
+       d  is  less  than  the limit set by the caller of pcre_exec() or, if no
+       such limit is set, less than the default.
+
+       The callout_data field is used in conjunction with the  "callout"  fea-
+       ture, and is described in the pcrecallout documentation.
+
+       The  tables  field  is  used  to  pass  a  character  tables pointer to
+       pcre_exec(); this overrides the value that is stored with the  compiled
+       pattern.  A  non-NULL value is stored with the compiled pattern only if
+       custom tables were supplied to pcre_compile() via  its  tableptr  argu-
+       ment.  If NULL is passed to pcre_exec() using this mechanism, it forces
+       PCRE's internal tables to be used. This facility is  helpful  when  re-
+       using  patterns  that  have been saved after compiling with an external
+       set of tables, because the external tables  might  be  at  a  different
+       address  when  pcre_exec() is called. See the pcreprecompile documenta-
+       tion for a discussion of saving compiled patterns for later use.
+
+       If PCRE_EXTRA_MARK is set in the flags field, the mark  field  must  be
+       set  to point to a suitable variable. If the pattern contains any back-
+       tracking control verbs such as (*MARK:NAME), and the execution ends  up
+       with  a  name  to  pass back, a pointer to the name string (zero termi-
+       nated) is placed in the variable pointed to  by  the  mark  field.  The
+       names  are  within  the  compiled pattern; if you wish to retain such a
+       name you must copy it before freeing the memory of a compiled  pattern.
+       If  there  is no name to pass back, the variable pointed to by the mark
+       field is set to NULL. For details of the  backtracking  control  verbs,
+       see the section entitled "Backtracking control" in the pcrepattern doc-
+       umentation.
+
+   Option bits for pcre_exec()
+
+       The unused bits of the options argument for pcre_exec() must  be  zero.
+       The  only  bits  that  may  be set are PCRE_ANCHORED, PCRE_NEWLINE_xxx,
+       PCRE_NOTBOL,   PCRE_NOTEOL,    PCRE_NOTEMPTY,    PCRE_NOTEMPTY_ATSTART,
+       PCRE_NO_START_OPTIMIZE,   PCRE_NO_UTF8_CHECK,   PCRE_PARTIAL_HARD,  and
+       PCRE_PARTIAL_SOFT.
+
+       If the pattern was successfully studied with one  of  the  just-in-time
+       (JIT) compile options, the only supported options for JIT execution are
+       PCRE_NO_UTF8_CHECK,    PCRE_NOTBOL,     PCRE_NOTEOL,     PCRE_NOTEMPTY,
+       PCRE_NOTEMPTY_ATSTART,  PCRE_PARTIAL_HARD, and PCRE_PARTIAL_SOFT. If an
+       unsupported option is used, JIT execution is disabled  and  the  normal
+       interpretive code in pcre_exec() is run.
+
+         PCRE_ANCHORED
+
+       The  PCRE_ANCHORED  option  limits pcre_exec() to matching at the first
+       matching position. If a pattern was  compiled  with  PCRE_ANCHORED,  or
+       turned  out to be anchored by virtue of its contents, it cannot be made
+       unachored at matching time.
+
+         PCRE_BSR_ANYCRLF
+         PCRE_BSR_UNICODE
+
+       These options (which are mutually exclusive) control what the \R escape
+       sequence  matches.  The choice is either to match only CR, LF, or CRLF,
+       or to match any Unicode newline sequence. These  options  override  the
+       choice that was made or defaulted when the pattern was compiled.
+
+         PCRE_NEWLINE_CR
+         PCRE_NEWLINE_LF
+         PCRE_NEWLINE_CRLF
+         PCRE_NEWLINE_ANYCRLF
+         PCRE_NEWLINE_ANY
+
+       These  options  override  the  newline  definition  that  was chosen or
+       defaulted when the pattern was compiled. For details, see the  descrip-
+       tion  of  pcre_compile()  above.  During  matching,  the newline choice
+       affects the behaviour of the dot, circumflex,  and  dollar  metacharac-
+       ters.  It may also alter the way the match position is advanced after a
+       match failure for an unanchored pattern.
+
+       When PCRE_NEWLINE_CRLF, PCRE_NEWLINE_ANYCRLF,  or  PCRE_NEWLINE_ANY  is
+       set,  and a match attempt for an unanchored pattern fails when the cur-
+       rent position is at a  CRLF  sequence,  and  the  pattern  contains  no
+       explicit  matches  for  CR  or  LF  characters,  the  match position is
+       advanced by two characters instead of one, in other words, to after the
+       CRLF.
+
+       The above rule is a compromise that makes the most common cases work as
+       expected. For example, if the  pattern  is  .+A  (and  the  PCRE_DOTALL
+       option is not set), it does not match the string "\r\nA" because, after
+       failing at the start, it skips both the CR and the LF before  retrying.
+       However,  the  pattern  [\r\n]A does match that string, because it con-
+       tains an explicit CR or LF reference, and so advances only by one char-
+       acter after the first failure.
+
+       An explicit match for CR of LF is either a literal appearance of one of
+       those characters, or one of the \r or  \n  escape  sequences.  Implicit
+       matches  such  as [^X] do not count, nor does \s (which includes CR and
+       LF in the characters that it matches).
+
+       Notwithstanding the above, anomalous effects may still occur when  CRLF
+       is a valid newline sequence and explicit \r or \n escapes appear in the
+       pattern.
+
+         PCRE_NOTBOL
+
+       This option specifies that first character of the subject string is not
+       the  beginning  of  a  line, so the circumflex metacharacter should not
+       match before it. Setting this without PCRE_MULTILINE (at compile  time)
+       causes  circumflex  never to match. This option affects only the behav-
+       iour of the circumflex metacharacter. It does not affect \A.
+
+         PCRE_NOTEOL
+
+       This option specifies that the end of the subject string is not the end
+       of  a line, so the dollar metacharacter should not match it nor (except
+       in multiline mode) a newline immediately before it. Setting this  with-
+       out PCRE_MULTILINE (at compile time) causes dollar never to match. This
+       option affects only the behaviour of the dollar metacharacter. It  does
+       not affect \Z or \z.
+
+         PCRE_NOTEMPTY
+
+       An empty string is not considered to be a valid match if this option is
+       set. If there are alternatives in the pattern, they are tried.  If  all
+       the  alternatives  match  the empty string, the entire match fails. For
+       example, if the pattern
+
+         a?b?
+
+       is applied to a string not beginning with "a" or  "b",  it  matches  an
+       empty  string at the start of the subject. With PCRE_NOTEMPTY set, this
+       match is not valid, so PCRE searches further into the string for occur-
+       rences of "a" or "b".
+
+         PCRE_NOTEMPTY_ATSTART
+
+       This  is  like PCRE_NOTEMPTY, except that an empty string match that is
+       not at the start of  the  subject  is  permitted.  If  the  pattern  is
+       anchored, such a match can occur only if the pattern contains \K.
+
+       Perl     has    no    direct    equivalent    of    PCRE_NOTEMPTY    or
+       PCRE_NOTEMPTY_ATSTART, but it does make a special  case  of  a  pattern
+       match  of  the empty string within its split() function, and when using
+       the /g modifier. It is  possible  to  emulate  Perl's  behaviour  after
+       matching a null string by first trying the match again at the same off-
+       set with PCRE_NOTEMPTY_ATSTART and  PCRE_ANCHORED,  and  then  if  that
+       fails, by advancing the starting offset (see below) and trying an ordi-
+       nary match again. There is some code that demonstrates how to  do  this
+       in  the  pcredemo sample program. In the most general case, you have to
+       check to see if the newline convention recognizes CRLF  as  a  newline,
+       and  if so, and the current character is CR followed by LF, advance the
+       starting offset by two characters instead of one.
+
+         PCRE_NO_START_OPTIMIZE
+
+       There are a number of optimizations that pcre_exec() uses at the  start
+       of  a  match,  in  order to speed up the process. For example, if it is
+       known that an unanchored match must start with a specific character, it
+       searches  the  subject  for that character, and fails immediately if it
+       cannot find it, without actually running the  main  matching  function.
+       This means that a special item such as (*COMMIT) at the start of a pat-
+       tern is not considered until after a suitable starting  point  for  the
+       match  has been found. Also, when callouts or (*MARK) items are in use,
+       these "start-up" optimizations can cause them to be skipped if the pat-
+       tern is never actually used. The start-up optimizations are in effect a
+       pre-scan of the subject that takes place before the pattern is run.
+
+       The PCRE_NO_START_OPTIMIZE option disables the start-up  optimizations,
+       possibly  causing  performance  to  suffer,  but ensuring that in cases
+       where the result is "no match", the callouts do occur, and  that  items
+       such as (*COMMIT) and (*MARK) are considered at every possible starting
+       position in the subject string. If  PCRE_NO_START_OPTIMIZE  is  set  at
+       compile  time,  it  cannot  be  unset  at  matching  time.  The  use of
+       PCRE_NO_START_OPTIMIZE  at  matching  time  (that  is,  passing  it  to
+       pcre_exec())  disables  JIT  execution;  in this situation, matching is
+       always done using interpretively.
+
+       Setting PCRE_NO_START_OPTIMIZE can change the  outcome  of  a  matching
+       operation.  Consider the pattern
+
+         (*COMMIT)ABC
+
+       When  this  is  compiled, PCRE records the fact that a match must start
+       with the character "A". Suppose the subject  string  is  "DEFABC".  The
+       start-up  optimization  scans along the subject, finds "A" and runs the
+       first match attempt from there. The (*COMMIT) item means that the  pat-
+       tern  must  match the current starting position, which in this case, it
+       does. However, if the same match  is  run  with  PCRE_NO_START_OPTIMIZE
+       set,  the  initial  scan  along the subject string does not happen. The
+       first match attempt is run starting  from  "D"  and  when  this  fails,
+       (*COMMIT)  prevents  any  further  matches  being tried, so the overall
+       result is "no match". If the pattern is studied,  more  start-up  opti-
+       mizations  may  be  used. For example, a minimum length for the subject
+       may be recorded. Consider the pattern
+
+         (*MARK:A)(X|Y)
+
+       The minimum length for a match is one  character.  If  the  subject  is
+       "ABC",  there  will  be  attempts  to  match "ABC", "BC", "C", and then
+       finally an empty string.  If the pattern is studied, the final  attempt
+       does  not take place, because PCRE knows that the subject is too short,
+       and so the (*MARK) is never encountered.  In this  case,  studying  the
+       pattern  does  not  affect the overall match result, which is still "no
+       match", but it does affect the auxiliary information that is returned.
+
+         PCRE_NO_UTF8_CHECK
+
+       When PCRE_UTF8 is set at compile time, the validity of the subject as a
+       UTF-8  string is automatically checked when pcre_exec() is subsequently
+       called.  The entire string is checked before any other processing takes
+       place.  The  value  of  startoffset  is  also checked to ensure that it
+       points to the start of a UTF-8 character. There is a  discussion  about
+       the  validity  of  UTF-8 strings in the pcreunicode page. If an invalid
+       sequence  of  bytes   is   found,   pcre_exec()   returns   the   error
+       PCRE_ERROR_BADUTF8 or, if PCRE_PARTIAL_HARD is set and the problem is a
+       truncated character at the end of the subject, PCRE_ERROR_SHORTUTF8. In
+       both  cases, information about the precise nature of the error may also
+       be returned (see the descriptions of these errors in the section  enti-
+       tled  Error return values from pcre_exec() below).  If startoffset con-
+       tains a value that does not point to the start of a UTF-8 character (or
+       to the end of the subject), PCRE_ERROR_BADUTF8_OFFSET is returned.
+
+       If  you  already  know that your subject is valid, and you want to skip
+       these   checks   for   performance   reasons,   you   can    set    the
+       PCRE_NO_UTF8_CHECK  option  when calling pcre_exec(). You might want to
+       do this for the second and subsequent calls to pcre_exec() if  you  are
+       making  repeated  calls  to  find  all  the matches in a single subject
+       string. However, you should be  sure  that  the  value  of  startoffset
+       points  to  the  start of a character (or the end of the subject). When
+       PCRE_NO_UTF8_CHECK is set, the effect of passing an invalid string as a
+       subject  or  an invalid value of startoffset is undefined. Your program
+       may crash.
+
+         PCRE_PARTIAL_HARD
+         PCRE_PARTIAL_SOFT
+
+       These options turn on the partial matching feature. For backwards  com-
+       patibility,  PCRE_PARTIAL is a synonym for PCRE_PARTIAL_SOFT. A partial
+       match occurs if the end of the subject string is reached  successfully,
+       but  there  are not enough subject characters to complete the match. If
+       this happens when PCRE_PARTIAL_SOFT (but not PCRE_PARTIAL_HARD) is set,
+       matching  continues  by  testing any remaining alternatives. Only if no
+       complete match can be found is PCRE_ERROR_PARTIAL returned  instead  of
+       PCRE_ERROR_NOMATCH.  In  other  words,  PCRE_PARTIAL_SOFT says that the
+       caller is prepared to handle a partial match, but only if  no  complete
+       match can be found.
+
+       If  PCRE_PARTIAL_HARD  is  set, it overrides PCRE_PARTIAL_SOFT. In this
+       case, if a partial match  is  found,  pcre_exec()  immediately  returns
+       PCRE_ERROR_PARTIAL,  without  considering  any  other  alternatives. In
+       other words, when PCRE_PARTIAL_HARD is set, a partial match is  consid-
+       ered to be more important that an alternative complete match.
+
+       In  both  cases,  the portion of the string that was inspected when the
+       partial match was found is set as the first matching string. There is a
+       more  detailed  discussion  of partial and multi-segment matching, with
+       examples, in the pcrepartial documentation.
+
+   The string to be matched by pcre_exec()
+
+       The subject string is passed to pcre_exec() as a pointer in subject,  a
+       length  in  length, and a starting offset in startoffset. The units for
+       length and startoffset are bytes for the  8-bit  library,  16-bit  data
+       items  for  the  16-bit  library,  and 32-bit data items for the 32-bit
+       library.
+
+       If startoffset is negative or greater than the length of  the  subject,
+       pcre_exec()  returns  PCRE_ERROR_BADOFFSET. When the starting offset is
+       zero, the search for a match starts at the beginning  of  the  subject,
+       and  this  is by far the most common case. In UTF-8 or UTF-16 mode, the
+       offset must point to the start of a character, or the end of  the  sub-
+       ject  (in  UTF-32 mode, one data unit equals one character, so all off-
+       sets are valid). Unlike the pattern string,  the  subject  may  contain
+       binary zeroes.
+
+       A  non-zero  starting offset is useful when searching for another match
+       in the same subject by calling pcre_exec() again after a previous  suc-
+       cess.   Setting  startoffset differs from just passing over a shortened
+       string and setting PCRE_NOTBOL in the case of  a  pattern  that  begins
+       with any kind of lookbehind. For example, consider the pattern
+
+         \Biss\B
+
+       which  finds  occurrences  of "iss" in the middle of words. (\B matches
+       only if the current position in the subject is not  a  word  boundary.)
+       When  applied  to the string "Mississipi" the first call to pcre_exec()
+       finds the first occurrence. If pcre_exec() is called  again  with  just
+       the  remainder  of  the  subject,  namely  "issipi", it does not match,
+       because \B is always false at the start of the subject, which is deemed
+       to  be  a  word  boundary. However, if pcre_exec() is passed the entire
+       string again, but with startoffset set to 4, it finds the second occur-
+       rence  of "iss" because it is able to look behind the starting point to
+       discover that it is preceded by a letter.
+
+       Finding all the matches in a subject is tricky  when  the  pattern  can
+       match an empty string. It is possible to emulate Perl's /g behaviour by
+       first  trying  the  match  again  at  the   same   offset,   with   the
+       PCRE_NOTEMPTY_ATSTART  and  PCRE_ANCHORED  options,  and  then  if that
+       fails, advancing the starting  offset  and  trying  an  ordinary  match
+       again. There is some code that demonstrates how to do this in the pcre-
+       demo sample program. In the most general case, you have to check to see
+       if  the newline convention recognizes CRLF as a newline, and if so, and
+       the current character is CR followed by LF, advance the starting offset
+       by two characters instead of one.
+
+       If  a  non-zero starting offset is passed when the pattern is anchored,
+       one attempt to match at the given offset is made. This can only succeed
+       if  the  pattern  does  not require the match to be at the start of the
+       subject.
+
+   How pcre_exec() returns captured substrings
+
+       In general, a pattern matches a certain portion of the subject, and  in
+       addition,  further  substrings  from  the  subject may be picked out by
+       parts of the pattern. Following the usage  in  Jeffrey  Friedl's  book,
+       this  is  called "capturing" in what follows, and the phrase "capturing
+       subpattern" is used for a fragment of a pattern that picks out  a  sub-
+       string.  PCRE  supports several other kinds of parenthesized subpattern
+       that do not cause substrings to be captured.
+
+       Captured substrings are returned to the caller via a vector of integers
+       whose  address is passed in ovector. The number of elements in the vec-
+       tor is passed in ovecsize, which must be a non-negative  number.  Note:
+       this argument is NOT the size of ovector in bytes.
+
+       The  first  two-thirds of the vector is used to pass back captured sub-
+       strings, each substring using a pair of integers. The  remaining  third
+       of  the  vector is used as workspace by pcre_exec() while matching cap-
+       turing subpatterns, and is not available for passing back  information.
+       The  number passed in ovecsize should always be a multiple of three. If
+       it is not, it is rounded down.
+
+       When a match is successful, information about  captured  substrings  is
+       returned  in  pairs  of integers, starting at the beginning of ovector,
+       and continuing up to two-thirds of its length at the  most.  The  first
+       element  of  each pair is set to the offset of the first character in a
+       substring, and the second is set to the offset of the  first  character
+       after  the  end  of a substring. These values are always data unit off-
+       sets, even in UTF mode. They are byte offsets  in  the  8-bit  library,
+       16-bit  data  item  offsets in the 16-bit library, and 32-bit data item
+       offsets in the 32-bit library. Note: they are not character counts.
+
+       The first pair of integers, ovector[0]  and  ovector[1],  identify  the
+       portion  of  the subject string matched by the entire pattern. The next
+       pair is used for the first capturing subpattern, and so on.  The  value
+       returned by pcre_exec() is one more than the highest numbered pair that
+       has been set.  For example, if two substrings have been  captured,  the
+       returned  value is 3. If there are no capturing subpatterns, the return
+       value from a successful match is 1, indicating that just the first pair
+       of offsets has been set.
+
+       If a capturing subpattern is matched repeatedly, it is the last portion
+       of the string that it matched that is returned.
+
+       If the vector is too small to hold all the captured substring  offsets,
+       it is used as far as possible (up to two-thirds of its length), and the
+       function returns a value of zero. If neither the actual string  matched
+       nor  any captured substrings are of interest, pcre_exec() may be called
+       with ovector passed as NULL and ovecsize as zero. However, if the  pat-
+       tern  contains  back  references  and  the ovector is not big enough to
+       remember the related substrings, PCRE has to get additional memory  for
+       use  during matching. Thus it is usually advisable to supply an ovector
+       of reasonable size.
+
+       There are some cases where zero is returned  (indicating  vector  over-
+       flow)  when  in fact the vector is exactly the right size for the final
+       match. For example, consider the pattern
+
+         (a)(?:(b)c|bd)
+
+       If a vector of 6 elements (allowing for only 1 captured  substring)  is
+       given with subject string "abd", pcre_exec() will try to set the second
+       captured string, thereby recording a vector overflow, before failing to
+       match  "c"  and  backing  up  to  try  the second alternative. The zero
+       return, however, does correctly indicate that  the  maximum  number  of
+       slots (namely 2) have been filled. In similar cases where there is tem-
+       porary overflow, but the final number of used slots  is  actually  less
+       than the maximum, a non-zero value is returned.
+
+       The pcre_fullinfo() function can be used to find out how many capturing
+       subpatterns there are in a compiled  pattern.  The  smallest  size  for
+       ovector  that  will allow for n captured substrings, in addition to the
+       offsets of the substring matched by the whole pattern, is (n+1)*3.
+
+       It is possible for capturing subpattern number n+1 to match  some  part
+       of the subject when subpattern n has not been used at all. For example,
+       if the string "abc" is matched  against  the  pattern  (a|(z))(bc)  the
+       return from the function is 4, and subpatterns 1 and 3 are matched, but
+       2 is not. When this happens, both values in  the  offset  pairs  corre-
+       sponding to unused subpatterns are set to -1.
+
+       Offset  values  that correspond to unused subpatterns at the end of the
+       expression are also set to -1. For example,  if  the  string  "abc"  is
+       matched  against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not
+       matched. The return from the function is 2, because  the  highest  used
+       capturing  subpattern  number  is 1, and the offsets for for the second
+       and third capturing subpatterns (assuming the vector is  large  enough,
+       of course) are set to -1.
+
+       Note:  Elements  in  the first two-thirds of ovector that do not corre-
+       spond to capturing parentheses in the pattern are never  changed.  That
+       is,  if  a pattern contains n capturing parentheses, no more than ovec-
+       tor[0] to ovector[2n+1] are set by pcre_exec(). The other elements  (in
+       the first two-thirds) retain whatever values they previously had.
+
+       Some  convenience  functions  are  provided for extracting the captured
+       substrings as separate strings. These are described below.
+
+   Error return values from pcre_exec()
+
+       If pcre_exec() fails, it returns a negative number. The  following  are
+       defined in the header file:
+
+         PCRE_ERROR_NOMATCH        (-1)
+
+       The subject string did not match the pattern.
+
+         PCRE_ERROR_NULL           (-2)
+
+       Either  code  or  subject  was  passed as NULL, or ovector was NULL and
+       ovecsize was not zero.
+
+         PCRE_ERROR_BADOPTION      (-3)
+
+       An unrecognized bit was set in the options argument.
+
+         PCRE_ERROR_BADMAGIC       (-4)
+
+       PCRE stores a 4-byte "magic number" at the start of the compiled  code,
+       to catch the case when it is passed a junk pointer and to detect when a
+       pattern that was compiled in an environment of one endianness is run in
+       an  environment  with the other endianness. This is the error that PCRE
+       gives when the magic number is not present.
+
+         PCRE_ERROR_UNKNOWN_OPCODE (-5)
+
+       While running the pattern match, an unknown item was encountered in the
+       compiled  pattern.  This  error  could be caused by a bug in PCRE or by
+       overwriting of the compiled pattern.
+
+         PCRE_ERROR_NOMEMORY       (-6)
+
+       If a pattern contains back references, but the ovector that  is  passed
+       to pcre_exec() is not big enough to remember the referenced substrings,
+       PCRE gets a block of memory at the start of matching to  use  for  this
+       purpose.  If the call via pcre_malloc() fails, this error is given. The
+       memory is automatically freed at the end of matching.
+
+       This error is also given if pcre_stack_malloc() fails  in  pcre_exec().
+       This  can happen only when PCRE has been compiled with --disable-stack-
+       for-recursion.
+
+         PCRE_ERROR_NOSUBSTRING    (-7)
+
+       This error is used by the pcre_copy_substring(),  pcre_get_substring(),
+       and  pcre_get_substring_list()  functions  (see  below).  It  is  never
+       returned by pcre_exec().
+
+         PCRE_ERROR_MATCHLIMIT     (-8)
+
+       The backtracking limit, as specified by  the  match_limit  field  in  a
+       pcre_extra  structure  (or  defaulted) was reached. See the description
+       above.
+
+         PCRE_ERROR_CALLOUT        (-9)
+
+       This error is never generated by pcre_exec() itself. It is provided for
+       use  by  callout functions that want to yield a distinctive error code.
+       See the pcrecallout documentation for details.
+
+         PCRE_ERROR_BADUTF8        (-10)
+
+       A string that contains an invalid UTF-8 byte sequence was passed  as  a
+       subject,  and the PCRE_NO_UTF8_CHECK option was not set. If the size of
+       the output vector (ovecsize) is at least 2,  the  byte  offset  to  the
+       start  of  the  the invalid UTF-8 character is placed in the first ele-
+       ment, and a reason code is placed in the  second  element.  The  reason
+       codes are listed in the following section.  For backward compatibility,
+       if PCRE_PARTIAL_HARD is set and the problem is a truncated UTF-8  char-
+       acter   at   the   end   of   the   subject  (reason  codes  1  to  5),
+       PCRE_ERROR_SHORTUTF8 is returned instead of PCRE_ERROR_BADUTF8.
+
+         PCRE_ERROR_BADUTF8_OFFSET (-11)
+
+       The UTF-8 byte sequence that was passed as a subject  was  checked  and
+       found  to be valid (the PCRE_NO_UTF8_CHECK option was not set), but the
+       value of startoffset did not point to the beginning of a UTF-8  charac-
+       ter or the end of the subject.
+
+         PCRE_ERROR_PARTIAL        (-12)
+
+       The  subject  string did not match, but it did match partially. See the
+       pcrepartial documentation for details of partial matching.
+
+         PCRE_ERROR_BADPARTIAL     (-13)
+
+       This code is no longer in  use.  It  was  formerly  returned  when  the
+       PCRE_PARTIAL  option  was used with a compiled pattern containing items
+       that were  not  supported  for  partial  matching.  From  release  8.00
+       onwards, there are no restrictions on partial matching.
+
+         PCRE_ERROR_INTERNAL       (-14)
+
+       An  unexpected  internal error has occurred. This error could be caused
+       by a bug in PCRE or by overwriting of the compiled pattern.
+
+         PCRE_ERROR_BADCOUNT       (-15)
+
+       This error is given if the value of the ovecsize argument is negative.
+
+         PCRE_ERROR_RECURSIONLIMIT (-21)
+
+       The internal recursion limit, as specified by the match_limit_recursion
+       field  in  a  pcre_extra  structure (or defaulted) was reached. See the
+       description above.
+
+         PCRE_ERROR_BADNEWLINE     (-23)
+
+       An invalid combination of PCRE_NEWLINE_xxx options was given.
+
+         PCRE_ERROR_BADOFFSET      (-24)
+
+       The value of startoffset was negative or greater than the length of the
+       subject, that is, the value in length.
+
+         PCRE_ERROR_SHORTUTF8      (-25)
+
+       This  error  is returned instead of PCRE_ERROR_BADUTF8 when the subject
+       string ends with a truncated UTF-8 character and the  PCRE_PARTIAL_HARD
+       option  is  set.   Information  about  the  failure  is returned as for
+       PCRE_ERROR_BADUTF8. It is in fact sufficient to detect this  case,  but
+       this  special error code for PCRE_PARTIAL_HARD precedes the implementa-
+       tion of returned information; it is retained for backwards  compatibil-
+       ity.
+
+         PCRE_ERROR_RECURSELOOP    (-26)
+
+       This error is returned when pcre_exec() detects a recursion loop within
+       the pattern. Specifically, it means that either the whole pattern or  a
+       subpattern  has been called recursively for the second time at the same
+       position in the subject string. Some simple patterns that might do this
+       are  detected  and faulted at compile time, but more complicated cases,
+       in particular mutual recursions between two different subpatterns, can-
+       not be detected until run time.
+
+         PCRE_ERROR_JIT_STACKLIMIT (-27)
+
+       This  error  is  returned  when a pattern that was successfully studied
+       using a JIT compile option is being matched, but the  memory  available
+       for  the  just-in-time  processing  stack  is not large enough. See the
+       pcrejit documentation for more details.
+
+         PCRE_ERROR_BADMODE        (-28)
+
+       This error is given if a pattern that was compiled by the 8-bit library
+       is passed to a 16-bit or 32-bit library function, or vice versa.
+
+         PCRE_ERROR_BADENDIANNESS  (-29)
+
+       This  error  is  given  if  a  pattern  that  was compiled and saved is
+       reloaded on a host with  different  endianness.  The  utility  function
+       pcre_pattern_to_host_byte_order() can be used to convert such a pattern
+       so that it runs on the new host.
+
+         PCRE_ERROR_JIT_BADOPTION
+
+       This error is returned when a pattern  that  was  successfully  studied
+       using  a  JIT  compile  option  is being matched, but the matching mode
+       (partial or complete match) does not correspond to any JIT  compilation
+       mode.  When  the JIT fast path function is used, this error may be also
+       given for invalid options.  See  the  pcrejit  documentation  for  more
+       details.
+
+         PCRE_ERROR_BADLENGTH      (-32)
+
+       This  error is given if pcre_exec() is called with a negative value for
+       the length argument.
+
+       Error numbers -16 to -20, -22, and 30 are not used by pcre_exec().
+
+   Reason codes for invalid UTF-8 strings
+
+       This section applies only  to  the  8-bit  library.  The  corresponding
+       information  for the 16-bit and 32-bit libraries is given in the pcre16
+       and pcre32 pages.
+
+       When pcre_exec() returns either PCRE_ERROR_BADUTF8 or PCRE_ERROR_SHORT-
+       UTF8,  and  the size of the output vector (ovecsize) is at least 2, the
+       offset of the start of the invalid UTF-8 character  is  placed  in  the
+       first output vector element (ovector[0]) and a reason code is placed in
+       the second element (ovector[1]). The reason codes are  given  names  in
+       the pcre.h header file:
+
+         PCRE_UTF8_ERR1
+         PCRE_UTF8_ERR2
+         PCRE_UTF8_ERR3
+         PCRE_UTF8_ERR4
+         PCRE_UTF8_ERR5
+
+       The  string  ends  with a truncated UTF-8 character; the code specifies
+       how many bytes are missing (1 to 5). Although RFC 3629 restricts  UTF-8
+       characters  to  be  no longer than 4 bytes, the encoding scheme (origi-
+       nally defined by RFC 2279) allows for  up  to  6  bytes,  and  this  is
+       checked first; hence the possibility of 4 or 5 missing bytes.
+
+         PCRE_UTF8_ERR6
+         PCRE_UTF8_ERR7
+         PCRE_UTF8_ERR8
+         PCRE_UTF8_ERR9
+         PCRE_UTF8_ERR10
+
+       The two most significant bits of the 2nd, 3rd, 4th, 5th, or 6th byte of
+       the character do not have the binary value 0b10 (that  is,  either  the
+       most significant bit is 0, or the next bit is 1).
+
+         PCRE_UTF8_ERR11
+         PCRE_UTF8_ERR12
+
+       A  character that is valid by the RFC 2279 rules is either 5 or 6 bytes
+       long; these code points are excluded by RFC 3629.
+
+         PCRE_UTF8_ERR13
+
+       A 4-byte character has a value greater than 0x10fff; these code  points
+       are excluded by RFC 3629.
+
+         PCRE_UTF8_ERR14
+
+       A  3-byte  character  has  a  value in the range 0xd800 to 0xdfff; this
+       range of code points are reserved by RFC 3629 for use with UTF-16,  and
+       so are excluded from UTF-8.
+
+         PCRE_UTF8_ERR15
+         PCRE_UTF8_ERR16
+         PCRE_UTF8_ERR17
+         PCRE_UTF8_ERR18
+         PCRE_UTF8_ERR19
+
+       A  2-, 3-, 4-, 5-, or 6-byte character is "overlong", that is, it codes
+       for a value that can be represented by fewer bytes, which  is  invalid.
+       For  example,  the two bytes 0xc0, 0xae give the value 0x2e, whose cor-
+       rect coding uses just one byte.
+
+         PCRE_UTF8_ERR20
+
+       The two most significant bits of the first byte of a character have the
+       binary  value 0b10 (that is, the most significant bit is 1 and the sec-
+       ond is 0). Such a byte can only validly occur as the second  or  subse-
+       quent byte of a multi-byte character.
+
+         PCRE_UTF8_ERR21
+
+       The  first byte of a character has the value 0xfe or 0xff. These values
+       can never occur in a valid UTF-8 string.
+
+         PCRE_UTF8_ERR22
+
+       This error code was formerly used when  the  presence  of  a  so-called
+       "non-character"  caused an error. Unicode corrigendum #9 makes it clear
+       that such characters should not cause a string to be rejected,  and  so
+       this code is no longer in use and is never returned.
+
+
+EXTRACTING CAPTURED SUBSTRINGS BY NUMBER
+
+       int pcre_copy_substring(const char *subject, int *ovector,
+            int stringcount, int stringnumber, char *buffer,
+            int buffersize);
+
+       int pcre_get_substring(const char *subject, int *ovector,
+            int stringcount, int stringnumber,
+            const char **stringptr);
+
+       int pcre_get_substring_list(const char *subject,
+            int *ovector, int stringcount, const char ***listptr);
+
+       Captured  substrings  can  be  accessed  directly  by using the offsets
+       returned by pcre_exec() in  ovector.  For  convenience,  the  functions
+       pcre_copy_substring(),    pcre_get_substring(),    and    pcre_get_sub-
+       string_list() are provided for extracting captured substrings  as  new,
+       separate,  zero-terminated strings. These functions identify substrings
+       by number. The next section describes functions  for  extracting  named
+       substrings.
+
+       A  substring that contains a binary zero is correctly extracted and has
+       a further zero added on the end, but the result is not, of course, a  C
+       string.   However,  you  can  process such a string by referring to the
+       length that is  returned  by  pcre_copy_substring()  and  pcre_get_sub-
+       string().  Unfortunately, the interface to pcre_get_substring_list() is
+       not adequate for handling strings containing binary zeros, because  the
+       end of the final string is not independently indicated.
+
+       The  first  three  arguments  are the same for all three of these func-
+       tions: subject is the subject string that has  just  been  successfully
+       matched, ovector is a pointer to the vector of integer offsets that was
+       passed to pcre_exec(), and stringcount is the number of substrings that
+       were  captured  by  the match, including the substring that matched the
+       entire regular expression. This is the value returned by pcre_exec() if
+       it  is greater than zero. If pcre_exec() returned zero, indicating that
+       it ran out of space in ovector, the value passed as stringcount  should
+       be the number of elements in the vector divided by three.
+
+       The  functions pcre_copy_substring() and pcre_get_substring() extract a
+       single substring, whose number is given as  stringnumber.  A  value  of
+       zero  extracts  the  substring that matched the entire pattern, whereas
+       higher values  extract  the  captured  substrings.  For  pcre_copy_sub-
+       string(),  the  string  is  placed  in buffer, whose length is given by
+       buffersize, while for pcre_get_substring() a new  block  of  memory  is
+       obtained  via  pcre_malloc,  and its address is returned via stringptr.
+       The yield of the function is the length of the  string,  not  including
+       the terminating zero, or one of these error codes:
+
+         PCRE_ERROR_NOMEMORY       (-6)
+
+       The  buffer  was too small for pcre_copy_substring(), or the attempt to
+       get memory failed for pcre_get_substring().
+
+         PCRE_ERROR_NOSUBSTRING    (-7)
+
+       There is no substring whose number is stringnumber.
+
+       The pcre_get_substring_list()  function  extracts  all  available  sub-
+       strings  and  builds  a list of pointers to them. All this is done in a
+       single block of memory that is obtained via pcre_malloc. The address of
+       the  memory  block  is returned via listptr, which is also the start of
+       the list of string pointers. The end of the list is marked  by  a  NULL
+       pointer.  The  yield  of  the function is zero if all went well, or the
+       error code
+
+         PCRE_ERROR_NOMEMORY       (-6)
+
+       if the attempt to get the memory block failed.
+
+       When any of these functions encounter a substring that is unset,  which
+       can  happen  when  capturing subpattern number n+1 matches some part of
+       the subject, but subpattern n has not been used at all, they return  an
+       empty string. This can be distinguished from a genuine zero-length sub-
+       string by inspecting the appropriate offset in ovector, which is  nega-
+       tive for unset substrings.
+
+       The  two convenience functions pcre_free_substring() and pcre_free_sub-
+       string_list() can be used to free the memory  returned  by  a  previous
+       call  of  pcre_get_substring()  or  pcre_get_substring_list(),  respec-
+       tively. They do nothing more than  call  the  function  pointed  to  by
+       pcre_free,  which  of course could be called directly from a C program.
+       However, PCRE is used in some situations where it is linked via a  spe-
+       cial   interface  to  another  programming  language  that  cannot  use
+       pcre_free directly; it is for these cases that the functions  are  pro-
+       vided.
+
+
+EXTRACTING CAPTURED SUBSTRINGS BY NAME
+
+       int pcre_get_stringnumber(const pcre *code,
+            const char *name);
+
+       int pcre_copy_named_substring(const pcre *code,
+            const char *subject, int *ovector,
+            int stringcount, const char *stringname,
+            char *buffer, int buffersize);
+
+       int pcre_get_named_substring(const pcre *code,
+            const char *subject, int *ovector,
+            int stringcount, const char *stringname,
+            const char **stringptr);
+
+       To  extract a substring by name, you first have to find associated num-
+       ber.  For example, for this pattern
+
+         (a+)b(?\d+)...
+
+       the number of the subpattern called "xxx" is 2. If the name is known to
+       be unique (PCRE_DUPNAMES was not set), you can find the number from the
+       name by calling pcre_get_stringnumber(). The first argument is the com-
+       piled pattern, and the second is the name. The yield of the function is
+       the subpattern number, or PCRE_ERROR_NOSUBSTRING (-7) if  there  is  no
+       subpattern of that name.
+
+       Given the number, you can extract the substring directly, or use one of
+       the functions described in the previous section. For convenience, there
+       are also two functions that do the whole job.
+
+       Most    of    the    arguments   of   pcre_copy_named_substring()   and
+       pcre_get_named_substring() are the same  as  those  for  the  similarly
+       named  functions  that extract by number. As these are described in the
+       previous section, they are not re-described here. There  are  just  two
+       differences:
+
+       First,  instead  of a substring number, a substring name is given. Sec-
+       ond, there is an extra argument, given at the start, which is a pointer
+       to  the compiled pattern. This is needed in order to gain access to the
+       name-to-number translation table.
+
+       These functions call pcre_get_stringnumber(), and if it succeeds,  they
+       then  call  pcre_copy_substring() or pcre_get_substring(), as appropri-
+       ate. NOTE: If PCRE_DUPNAMES is set and there are duplicate  names,  the
+       behaviour may not be what you want (see the next section).
+
+       Warning: If the pattern uses the (?| feature to set up multiple subpat-
+       terns with the same number, as described in the  section  on  duplicate
+       subpattern  numbers  in  the  pcrepattern page, you cannot use names to
+       distinguish the different subpatterns, because names are  not  included
+       in  the compiled code. The matching process uses only numbers. For this
+       reason, the use of different names for subpatterns of the  same  number
+       causes an error at compile time.
+
+
+DUPLICATE SUBPATTERN NAMES
+
+       int pcre_get_stringtable_entries(const pcre *code,
+            const char *name, char **first, char **last);
+
+       When  a  pattern  is  compiled with the PCRE_DUPNAMES option, names for
+       subpatterns are not required to be unique. (Duplicate names are  always
+       allowed  for subpatterns with the same number, created by using the (?|
+       feature. Indeed, if such subpatterns are named, they  are  required  to
+       use the same names.)
+
+       Normally, patterns with duplicate names are such that in any one match,
+       only one of the named subpatterns participates. An example is shown  in
+       the pcrepattern documentation.
+
+       When    duplicates   are   present,   pcre_copy_named_substring()   and
+       pcre_get_named_substring() return the first substring corresponding  to
+       the  given  name  that  is set. If none are set, PCRE_ERROR_NOSUBSTRING
+       (-7) is returned; no  data  is  returned.  The  pcre_get_stringnumber()
+       function  returns one of the numbers that are associated with the name,
+       but it is not defined which it is.
+
+       If you want to get full details of all captured substrings for a  given
+       name,  you  must  use  the pcre_get_stringtable_entries() function. The
+       first argument is the compiled pattern, and the second is the name. The
+       third  and  fourth  are  pointers to variables which are updated by the
+       function. After it has run, they point to the first and last entries in
+       the  name-to-number  table  for  the  given  name.  The function itself
+       returns the length of each entry,  or  PCRE_ERROR_NOSUBSTRING  (-7)  if
+       there  are none. The format of the table is described above in the sec-
+       tion entitled Information about a pattern above.  Given all  the  rele-
+       vant  entries  for the name, you can extract each of their numbers, and
+       hence the captured data, if any.
+
+
+FINDING ALL POSSIBLE MATCHES
+
+       The traditional matching function uses a  similar  algorithm  to  Perl,
+       which stops when it finds the first match, starting at a given point in
+       the subject. If you want to find all possible matches, or  the  longest
+       possible  match,  consider using the alternative matching function (see
+       below) instead. If you cannot use the alternative function,  but  still
+       need  to  find all possible matches, you can kludge it up by making use
+       of the callout facility, which is described in the pcrecallout documen-
+       tation.
+
+       What you have to do is to insert a callout right at the end of the pat-
+       tern.  When your callout function is called, extract and save the  cur-
+       rent  matched  substring.  Then  return  1, which forces pcre_exec() to
+       backtrack and try other alternatives. Ultimately, when it runs  out  of
+       matches, pcre_exec() will yield PCRE_ERROR_NOMATCH.
+
+
+OBTAINING AN ESTIMATE OF STACK USAGE
+
+       Matching  certain  patterns  using pcre_exec() can use a lot of process
+       stack, which in certain environments can be  rather  limited  in  size.
+       Some  users  find it helpful to have an estimate of the amount of stack
+       that is used by pcre_exec(), to help  them  set  recursion  limits,  as
+       described  in  the pcrestack documentation. The estimate that is output
+       by pcretest when called with the -m and -C options is obtained by call-
+       ing  pcre_exec with the values NULL, NULL, NULL, -999, and -999 for its
+       first five arguments.
+
+       Normally, if  its  first  argument  is  NULL,  pcre_exec()  immediately
+       returns  the negative error code PCRE_ERROR_NULL, but with this special
+       combination of arguments, it returns instead a  negative  number  whose
+       absolute  value  is the approximate stack frame size in bytes. (A nega-
+       tive number is used so that it is clear that no  match  has  happened.)
+       The  value  is  approximate  because  in some cases, recursive calls to
+       pcre_exec() occur when there are one or two additional variables on the
+       stack.
+
+       If  PCRE  has  been  compiled  to use the heap instead of the stack for
+       recursion, the value returned  is  the  size  of  each  block  that  is
+       obtained from the heap.
+
+
+MATCHING A PATTERN: THE ALTERNATIVE FUNCTION
+
+       int pcre_dfa_exec(const pcre *code, const pcre_extra *extra,
+            const char *subject, int length, int startoffset,
+            int options, int *ovector, int ovecsize,
+            int *workspace, int wscount);
+
+       The  function  pcre_dfa_exec()  is  called  to  match  a subject string
+       against a compiled pattern, using a matching algorithm that  scans  the
+       subject  string  just  once, and does not backtrack. This has different
+       characteristics to the normal algorithm, and  is  not  compatible  with
+       Perl.  Some  of the features of PCRE patterns are not supported. Never-
+       theless, there are times when this kind of matching can be useful.  For
+       a  discussion  of  the  two matching algorithms, and a list of features
+       that pcre_dfa_exec() does not support, see the pcrematching  documenta-
+       tion.
+
+       The  arguments  for  the  pcre_dfa_exec()  function are the same as for
+       pcre_exec(), plus two extras. The ovector argument is used in a differ-
+       ent  way,  and  this is described below. The other common arguments are
+       used in the same way as for pcre_exec(), so their  description  is  not
+       repeated here.
+
+       The  two  additional  arguments provide workspace for the function. The
+       workspace vector should contain at least 20 elements. It  is  used  for
+       keeping  track  of  multiple  paths  through  the  pattern  tree.  More
+       workspace will be needed for patterns and subjects where  there  are  a
+       lot of potential matches.
+
+       Here is an example of a simple call to pcre_dfa_exec():
+
+         int rc;
+         int ovector[10];
+         int wspace[20];
+         rc = pcre_dfa_exec(
+           re,             /* result of pcre_compile() */
+           NULL,           /* we didn't study the pattern */
+           "some string",  /* the subject string */
+           11,             /* the length of the subject string */
+           0,              /* start at offset 0 in the subject */
+           0,              /* default options */
+           ovector,        /* vector of integers for substring information */
+           10,             /* number of elements (NOT size in bytes) */
+           wspace,         /* working space vector */
+           20);            /* number of elements (NOT size in bytes) */
+
+   Option bits for pcre_dfa_exec()
+
+       The  unused  bits  of  the options argument for pcre_dfa_exec() must be
+       zero. The only bits  that  may  be  set  are  PCRE_ANCHORED,  PCRE_NEW-
+       LINE_xxx,        PCRE_NOTBOL,        PCRE_NOTEOL,        PCRE_NOTEMPTY,
+       PCRE_NOTEMPTY_ATSTART,      PCRE_NO_UTF8_CHECK,       PCRE_BSR_ANYCRLF,
+       PCRE_BSR_UNICODE,  PCRE_NO_START_OPTIMIZE, PCRE_PARTIAL_HARD, PCRE_PAR-
+       TIAL_SOFT, PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART.  All but  the  last
+       four  of  these  are  exactly  the  same  as  for pcre_exec(), so their
+       description is not repeated here.
+
+         PCRE_PARTIAL_HARD
+         PCRE_PARTIAL_SOFT
+
+       These have the same general effect as they do for pcre_exec(), but  the
+       details  are  slightly  different.  When  PCRE_PARTIAL_HARD  is set for
+       pcre_dfa_exec(), it returns PCRE_ERROR_PARTIAL if the end of  the  sub-
+       ject  is  reached  and there is still at least one matching possibility
+       that requires additional characters. This happens even if some complete
+       matches have also been found. When PCRE_PARTIAL_SOFT is set, the return
+       code PCRE_ERROR_NOMATCH is converted into PCRE_ERROR_PARTIAL if the end
+       of  the  subject  is  reached, there have been no complete matches, but
+       there is still at least one matching possibility. The  portion  of  the
+       string  that  was inspected when the longest partial match was found is
+       set as the first matching string  in  both  cases.   There  is  a  more
+       detailed  discussion  of partial and multi-segment matching, with exam-
+       ples, in the pcrepartial documentation.
+
+         PCRE_DFA_SHORTEST
+
+       Setting the PCRE_DFA_SHORTEST option causes the matching  algorithm  to
+       stop as soon as it has found one match. Because of the way the alterna-
+       tive algorithm works, this is necessarily the shortest  possible  match
+       at the first possible matching point in the subject string.
+
+         PCRE_DFA_RESTART
+
+       When pcre_dfa_exec() returns a partial match, it is possible to call it
+       again, with additional subject characters, and have  it  continue  with
+       the  same match. The PCRE_DFA_RESTART option requests this action; when
+       it is set, the workspace and wscount options must  reference  the  same
+       vector  as  before  because data about the match so far is left in them
+       after a partial match. There is more discussion of this facility in the
+       pcrepartial documentation.
+
+   Successful returns from pcre_dfa_exec()
+
+       When  pcre_dfa_exec()  succeeds, it may have matched more than one sub-
+       string in the subject. Note, however, that all the matches from one run
+       of  the  function  start  at the same point in the subject. The shorter
+       matches are all initial substrings of the longer matches. For  example,
+       if the pattern
+
+         <.*>
+
+       is matched against the string
+
+         This is    no more
+
+       the three matched strings are
+
+         
+          
+           
+
+       On  success,  the  yield of the function is a number greater than zero,
+       which is the number of matched substrings.  The  substrings  themselves
+       are  returned  in  ovector. Each string uses two elements; the first is
+       the offset to the start, and the second is the offset to  the  end.  In
+       fact,  all  the  strings  have the same start offset. (Space could have
+       been saved by giving this only once, but it was decided to retain  some
+       compatibility  with  the  way pcre_exec() returns data, even though the
+       meaning of the strings is different.)
+
+       The strings are returned in reverse order of length; that is, the long-
+       est  matching  string is given first. If there were too many matches to
+       fit into ovector, the yield of the function is zero, and the vector  is
+       filled  with  the  longest matches. Unlike pcre_exec(), pcre_dfa_exec()
+       can use the entire ovector for returning matched strings.
+
+   Error returns from pcre_dfa_exec()
+
+       The pcre_dfa_exec() function returns a negative number when  it  fails.
+       Many  of  the  errors  are  the  same as for pcre_exec(), and these are
+       described above.  There are in addition the following errors  that  are
+       specific to pcre_dfa_exec():
+
+         PCRE_ERROR_DFA_UITEM      (-16)
+
+       This  return is given if pcre_dfa_exec() encounters an item in the pat-
+       tern that it does not support, for instance, the use of \C  or  a  back
+       reference.
+
+         PCRE_ERROR_DFA_UCOND      (-17)
+
+       This  return  is  given  if pcre_dfa_exec() encounters a condition item
+       that uses a back reference for the condition, or a test  for  recursion
+       in a specific group. These are not supported.
+
+         PCRE_ERROR_DFA_UMLIMIT    (-18)
+
+       This  return  is given if pcre_dfa_exec() is called with an extra block
+       that contains a setting of  the  match_limit  or  match_limit_recursion
+       fields.  This  is  not  supported (these fields are meaningless for DFA
+       matching).
+
+         PCRE_ERROR_DFA_WSSIZE     (-19)
+
+       This return is given if  pcre_dfa_exec()  runs  out  of  space  in  the
+       workspace vector.
+
+         PCRE_ERROR_DFA_RECURSE    (-20)
+
+       When  a  recursive subpattern is processed, the matching function calls
+       itself recursively, using private vectors for  ovector  and  workspace.
+       This  error  is  given  if  the output vector is not large enough. This
+       should be extremely rare, as a vector of size 1000 is used.
+
+         PCRE_ERROR_DFA_BADRESTART (-30)
+
+       When pcre_dfa_exec() is called with the PCRE_DFA_RESTART  option,  some
+       plausibility  checks  are  made on the contents of the workspace, which
+       should contain data about the previous partial match. If any  of  these
+       checks fail, this error is given.
+
+
+SEE ALSO
+
+       pcre16(3),   pcre32(3),  pcrebuild(3),  pcrecallout(3),  pcrecpp(3)(3),
+       pcrematching(3), pcrepartial(3), pcreposix(3), pcreprecompile(3), pcre-
+       sample(3), pcrestack(3).
+
+
+AUTHOR
+
+       Philip Hazel
+       University Computing Service
+       Cambridge CB2 3QH, England.
+
+
+REVISION
+
+       Last updated: 12 May 2013
+       Copyright (c) 1997-2013 University of Cambridge.
+------------------------------------------------------------------------------
+
+
+PCRECALLOUT(3)             Library Functions Manual             PCRECALLOUT(3)
+
+
+
+NAME
+       PCRE - Perl-compatible regular expressions
+
+SYNOPSIS
+
+       #include 
+
+       int (*pcre_callout)(pcre_callout_block *);
+
+       int (*pcre16_callout)(pcre16_callout_block *);
+
+       int (*pcre32_callout)(pcre32_callout_block *);
+
+
+DESCRIPTION
+
+       PCRE provides a feature called "callout", which is a means of temporar-
+       ily passing control to the caller of PCRE  in  the  middle  of  pattern
+       matching.  The  caller of PCRE provides an external function by putting
+       its entry point in the global variable pcre_callout (pcre16_callout for
+       the 16-bit library, pcre32_callout for the 32-bit library). By default,
+       this variable contains NULL, which disables all calling out.
+
+       Within a regular expression, (?C) indicates the  points  at  which  the
+       external  function  is  to  be  called. Different callout points can be
+       identified by putting a number less than 256 after the  letter  C.  The
+       default  value  is  zero.   For  example,  this pattern has two callout
+       points:
+
+         (?C1)abc(?C2)def
+
+       If the PCRE_AUTO_CALLOUT option bit is set when a pattern is  compiled,
+       PCRE  automatically  inserts callouts, all with number 255, before each
+       item in the pattern. For example, if PCRE_AUTO_CALLOUT is used with the
+       pattern
+
+         A(\d{2}|--)
+
+       it is processed as if it were
+
+       (?C255)A(?C255)((?C255)\d{2}(?C255)|(?C255)-(?C255)-(?C255))(?C255)
+
+       Notice  that  there  is a callout before and after each parenthesis and
+       alternation bar. If the pattern contains a conditional group whose con-
+       dition  is  an  assertion, an automatic callout is inserted immediately
+       before the condition. Such a callout may also be  inserted  explicitly,
+       for example:
+
+         (?(?C9)(?=a)ab|de)
+
+       This  applies only to assertion conditions (because they are themselves
+       independent groups).
+
+       Automatic callouts can be used for tracking  the  progress  of  pattern
+       matching.  The pcretest command has an option that sets automatic call-
+       outs; when it is used, the output indicates how the pattern is matched.
+       This  is useful information when you are trying to optimize the perfor-
+       mance of a particular pattern.
+
+
+MISSING CALLOUTS
+
+       You should be aware that, because of  optimizations  in  the  way  PCRE
+       matches  patterns  by  default,  callouts  sometimes do not happen. For
+       example, if the pattern is
+
+         ab(?C4)cd
+
+       PCRE knows that any matching string must contain the letter "d". If the
+       subject  string  is "abyz", the lack of "d" means that matching doesn't
+       ever start, and the callout is never  reached.  However,  with  "abyd",
+       though the result is still no match, the callout is obeyed.
+
+       If  the pattern is studied, PCRE knows the minimum length of a matching
+       string, and will immediately give a "no match" return without  actually
+       running  a  match if the subject is not long enough, or, for unanchored
+       patterns, if it has been scanned far enough.
+
+       You can disable these optimizations by passing the  PCRE_NO_START_OPTI-
+       MIZE  option  to the matching function, or by starting the pattern with
+       (*NO_START_OPT). This slows down the matching process, but does  ensure
+       that callouts such as the example above are obeyed.
+
+
+THE CALLOUT INTERFACE
+
+       During  matching, when PCRE reaches a callout point, the external func-
+       tion defined by pcre_callout or pcre[16|32]_callout is called (if it is
+       set).  This  applies to both normal and DFA matching. The only argument
+       to  the  callout  function  is  a  pointer   to   a   pcre_callout   or
+       pcre[16|32]_callout  block.   These  structures  contains the following
+       fields:
+
+         int           version;
+         int           callout_number;
+         int          *offset_vector;
+         const char   *subject;           (8-bit version)
+         PCRE_SPTR16   subject;           (16-bit version)
+         PCRE_SPTR32   subject;           (32-bit version)
+         int           subject_length;
+         int           start_match;
+         int           current_position;
+         int           capture_top;
+         int           capture_last;
+         void         *callout_data;
+         int           pattern_position;
+         int           next_item_length;
+         const unsigned char *mark;       (8-bit version)
+         const PCRE_UCHAR16  *mark;       (16-bit version)
+         const PCRE_UCHAR32  *mark;       (32-bit version)
+
+       The version field is an integer containing the version  number  of  the
+       block  format. The initial version was 0; the current version is 2. The
+       version number will change again in future  if  additional  fields  are
+       added, but the intention is never to remove any of the existing fields.
+
+       The  callout_number  field  contains the number of the callout, as com-
+       piled into the pattern (that is, the number after ?C for  manual  call-
+       outs, and 255 for automatically generated callouts).
+
+       The  offset_vector field is a pointer to the vector of offsets that was
+       passed by the caller to the  matching  function.  When  pcre_exec()  or
+       pcre[16|32]_exec()  is used, the contents can be inspected, in order to
+       extract substrings that have been matched so far, in the  same  way  as
+       for  extracting  substrings  after  a  match has completed. For the DFA
+       matching functions, this field is not useful.
+
+       The subject and subject_length fields contain copies of the values that
+       were passed to the matching function.
+
+       The  start_match  field normally contains the offset within the subject
+       at which the current match attempt  started.  However,  if  the  escape
+       sequence  \K has been encountered, this value is changed to reflect the
+       modified starting point. If the pattern is not  anchored,  the  callout
+       function may be called several times from the same point in the pattern
+       for different starting points in the subject.
+
+       The current_position field contains the offset within  the  subject  of
+       the current match pointer.
+
+       When  the  pcre_exec()  or  pcre[16|32]_exec() is used, the capture_top
+       field contains one more than the number of the  highest  numbered  cap-
+       tured  substring so far. If no substrings have been captured, the value
+       of capture_top is one. This is always the case when the  DFA  functions
+       are used, because they do not support captured substrings.
+
+       The  capture_last  field  contains the number of the most recently cap-
+       tured substring. However, when a recursion exits, the value reverts  to
+       what  it  was  outside  the recursion, as do the values of all captured
+       substrings. If no substrings have been  captured,  the  value  of  cap-
+       ture_last  is  -1.  This  is always the case for the DFA matching func-
+       tions.
+
+       The callout_data field contains a value that is passed  to  a  matching
+       function  specifically so that it can be passed back in callouts. It is
+       passed in the callout_data field of a pcre_extra  or  pcre[16|32]_extra
+       data  structure.  If no such data was passed, the value of callout_data
+       in a callout block is NULL. There is a description  of  the  pcre_extra
+       structure in the pcreapi documentation.
+
+       The  pattern_position  field  is  present from version 1 of the callout
+       structure. It contains the offset to the next item to be matched in the
+       pattern string.
+
+       The  next_item_length  field  is  present from version 1 of the callout
+       structure. It contains the length of the next item to be matched in the
+       pattern  string.  When  the callout immediately precedes an alternation
+       bar, a closing parenthesis, or the end of the pattern,  the  length  is
+       zero.  When  the callout precedes an opening parenthesis, the length is
+       that of the entire subpattern.
+
+       The pattern_position and next_item_length fields are intended  to  help
+       in  distinguishing between different automatic callouts, which all have
+       the same callout number. However, they are set for all callouts.
+
+       The mark field is present from version 2 of the callout  structure.  In
+       callouts  from  pcre_exec() or pcre[16|32]_exec() it contains a pointer
+       to the zero-terminated  name  of  the  most  recently  passed  (*MARK),
+       (*PRUNE),  or  (*THEN) item in the match, or NULL if no such items have
+       been passed. Instances of (*PRUNE) or (*THEN) without  a  name  do  not
+       obliterate  a previous (*MARK). In callouts from the DFA matching func-
+       tions this field always contains NULL.
+
+
+RETURN VALUES
+
+       The external callout function returns an integer to PCRE. If the  value
+       is  zero,  matching  proceeds  as  normal. If the value is greater than
+       zero, matching fails at the current point, but  the  testing  of  other
+       matching possibilities goes ahead, just as if a lookahead assertion had
+       failed. If the value is less than zero, the  match  is  abandoned,  the
+       matching function returns the negative value.
+
+       Negative   values   should   normally   be   chosen  from  the  set  of
+       PCRE_ERROR_xxx values. In particular, PCRE_ERROR_NOMATCH forces a stan-
+       dard  "no  match"  failure.   The  error  number  PCRE_ERROR_CALLOUT is
+       reserved for use by callout functions; it will never be  used  by  PCRE
+       itself.
+
+
+AUTHOR
+
+       Philip Hazel
+       University Computing Service
+       Cambridge CB2 3QH, England.
+
+
+REVISION
+
+       Last updated: 03 March 2013
+       Copyright (c) 1997-2013 University of Cambridge.
+------------------------------------------------------------------------------
+
+
+PCRECOMPAT(3)              Library Functions Manual              PCRECOMPAT(3)
+
+
+
+NAME
+       PCRE - Perl-compatible regular expressions
+
+DIFFERENCES BETWEEN PCRE AND PERL
+
+       This  document describes the differences in the ways that PCRE and Perl
+       handle regular expressions. The differences  described  here  are  with
+       respect to Perl versions 5.10 and above.
+
+       1. PCRE has only a subset of Perl's Unicode support. Details of what it
+       does have are given in the pcreunicode page.
+
+       2. PCRE allows repeat quantifiers only on parenthesized assertions, but
+       they  do  not mean what you might think. For example, (?!a){3} does not
+       assert that the next three characters are not "a". It just asserts that
+       the next character is not "a" three times (in principle: PCRE optimizes
+       this to run the assertion just once). Perl allows repeat quantifiers on
+       other assertions such as \b, but these do not seem to have any use.
+
+       3.  Capturing  subpatterns  that occur inside negative lookahead asser-
+       tions are counted, but their entries in the offsets  vector  are  never
+       set.  Perl sometimes (but not always) sets its numerical variables from
+       inside negative assertions.
+
+       4. Though binary zero characters are supported in the  subject  string,
+       they are not allowed in a pattern string because it is passed as a nor-
+       mal C string, terminated by zero. The escape sequence \0 can be used in
+       the pattern to represent a binary zero.
+
+       5.  The  following Perl escape sequences are not supported: \l, \u, \L,
+       \U, and \N when followed by a character name or Unicode value.  (\N  on
+       its own, matching a non-newline character, is supported.) In fact these
+       are implemented by Perl's general string-handling and are not  part  of
+       its  pattern  matching engine. If any of these are encountered by PCRE,
+       an error is generated by default. However, if the  PCRE_JAVASCRIPT_COM-
+       PAT  option  is set, \U and \u are interpreted as JavaScript interprets
+       them.
+
+       6. The Perl escape sequences \p, \P, and \X are supported only if  PCRE
+       is  built  with Unicode character property support. The properties that
+       can be tested with \p and \P are limited to the general category  prop-
+       erties  such  as  Lu and Nd, script names such as Greek or Han, and the
+       derived properties Any and L&. PCRE does  support  the  Cs  (surrogate)
+       property,  which  Perl  does  not; the Perl documentation says "Because
+       Perl hides the need for the user to understand the internal representa-
+       tion  of Unicode characters, there is no need to implement the somewhat
+       messy concept of surrogates."
+
+       7. PCRE does support the \Q...\E escape for quoting substrings. Charac-
+       ters  in  between  are  treated as literals. This is slightly different
+       from Perl in that $ and @ are  also  handled  as  literals  inside  the
+       quotes.  In Perl, they cause variable interpolation (but of course PCRE
+       does not have variables). Note the following examples:
+
+           Pattern            PCRE matches      Perl matches
+
+           \Qabc$xyz\E        abc$xyz           abc followed by the
+                                                  contents of $xyz
+           \Qabc\$xyz\E       abc\$xyz          abc\$xyz
+           \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
+
+       The \Q...\E sequence is recognized both inside  and  outside  character
+       classes.
+
+       8. Fairly obviously, PCRE does not support the (?{code}) and (??{code})
+       constructions. However, there is support for recursive  patterns.  This
+       is  not  available  in Perl 5.8, but it is in Perl 5.10. Also, the PCRE
+       "callout" feature allows an external function to be called during  pat-
+       tern matching. See the pcrecallout documentation for details.
+
+       9.  Subpatterns  that  are called as subroutines (whether or not recur-
+       sively) are always treated as atomic  groups  in  PCRE.  This  is  like
+       Python,  but  unlike Perl.  Captured values that are set outside a sub-
+       routine call can be reference from inside in PCRE,  but  not  in  Perl.
+       There is a discussion that explains these differences in more detail in
+       the section on recursion differences from Perl in the pcrepattern page.
+
+       10. If any of the backtracking control verbs are used in  a  subpattern
+       that  is  called  as  a  subroutine (whether or not recursively), their
+       effect is confined to that subpattern; it does not extend to  the  sur-
+       rounding  pattern.  This is not always the case in Perl. In particular,
+       if (*THEN) is present in a group that is called as  a  subroutine,  its
+       action is limited to that group, even if the group does not contain any
+       | characters. Note that such subpatterns are processed as  anchored  at
+       the point where they are tested.
+
+       11.  If a pattern contains more than one backtracking control verb, the
+       first one that is backtracked onto acts. For example,  in  the  pattern
+       A(*COMMIT)B(*PRUNE)C  a  failure in B triggers (*COMMIT), but a failure
+       in C triggers (*PRUNE). Perl's behaviour is more complex; in many cases
+       it is the same as PCRE, but there are examples where it differs.
+
+       12.  Most  backtracking  verbs in assertions have their normal actions.
+       They are not confined to the assertion.
+
+       13. There are some differences that are concerned with the settings  of
+       captured  strings  when  part  of  a  pattern is repeated. For example,
+       matching "aba" against the  pattern  /^(a(b)?)+$/  in  Perl  leaves  $2
+       unset, but in PCRE it is set to "b".
+
+       14.  PCRE's handling of duplicate subpattern numbers and duplicate sub-
+       pattern names is not as general as Perl's. This is a consequence of the
+       fact the PCRE works internally just with numbers, using an external ta-
+       ble to translate between numbers and names. In  particular,  a  pattern
+       such  as  (?|(?A)|(?  (Oniguruma  syntax)  are  not synonymous. The former is a back
+       reference; the latter is a subroutine call.
+
+   Generic character types
+
+       Another use of backslash is for specifying generic character types:
+
+         \d     any decimal digit
+         \D     any character that is not a decimal digit
+         \h     any horizontal white space character
+         \H     any character that is not a horizontal white space character
+         \s     any white space character
+         \S     any character that is not a white space character
+         \v     any vertical white space character
+         \V     any character that is not a vertical white space character
+         \w     any "word" character
+         \W     any "non-word" character
+
+       There is also the single sequence \N, which matches a non-newline char-
+       acter.   This  is the same as the "." metacharacter when PCRE_DOTALL is
+       not set. Perl also uses \N to match characters by name; PCRE  does  not
+       support this.
+
+       Each  pair of lower and upper case escape sequences partitions the com-
+       plete set of characters into two disjoint  sets.  Any  given  character
+       matches  one, and only one, of each pair. The sequences can appear both
+       inside and outside character classes. They each match one character  of
+       the  appropriate  type.  If the current matching point is at the end of
+       the subject string, all of them fail, because there is no character  to
+       match.
+
+       For  compatibility  with Perl, \s does not match the VT character (code
+       11).  This makes it different from the the POSIX "space" class. The  \s
+       characters  are  HT  (9), LF (10), FF (12), CR (13), and space (32). If
+       "use locale;" is included in a Perl script, \s may match the VT charac-
+       ter. In PCRE, it never does.
+
+       A  "word"  character is an underscore or any character that is a letter
+       or digit.  By default, the definition of letters  and  digits  is  con-
+       trolled  by PCRE's low-valued character tables, and may vary if locale-
+       specific matching is taking place (see "Locale support" in the  pcreapi
+       page).  For  example,  in  a French locale such as "fr_FR" in Unix-like
+       systems, or "french" in Windows, some character codes greater than  128
+       are  used  for  accented letters, and these are then matched by \w. The
+       use of locales with Unicode is discouraged.
+
+       By default, in a UTF mode, characters  with  values  greater  than  128
+       never  match  \d,  \s,  or  \w,  and always match \D, \S, and \W. These
+       sequences retain their original meanings from before  UTF  support  was
+       available,  mainly for efficiency reasons. However, if PCRE is compiled
+       with Unicode property support, and the PCRE_UCP option is set, the  be-
+       haviour  is  changed  so  that Unicode properties are used to determine
+       character types, as follows:
+
+         \d  any character that \p{Nd} matches (decimal digit)
+         \s  any character that \p{Z} matches, plus HT, LF, FF, CR
+         \w  any character that \p{L} or \p{N} matches, plus underscore
+
+       The upper case escapes match the inverse sets of characters. Note  that
+       \d  matches  only decimal digits, whereas \w matches any Unicode digit,
+       as well as any Unicode letter, and underscore. Note also that  PCRE_UCP
+       affects  \b,  and  \B  because  they are defined in terms of \w and \W.
+       Matching these sequences is noticeably slower when PCRE_UCP is set.
+
+       The sequences \h, \H, \v, and \V are features that were added  to  Perl
+       at  release  5.10. In contrast to the other sequences, which match only
+       ASCII characters by default, these  always  match  certain  high-valued
+       codepoints,  whether or not PCRE_UCP is set. The horizontal space char-
+       acters are:
+
+         U+0009     Horizontal tab (HT)
+         U+0020     Space
+         U+00A0     Non-break space
+         U+1680     Ogham space mark
+         U+180E     Mongolian vowel separator
+         U+2000     En quad
+         U+2001     Em quad
+         U+2002     En space
+         U+2003     Em space
+         U+2004     Three-per-em space
+         U+2005     Four-per-em space
+         U+2006     Six-per-em space
+         U+2007     Figure space
+         U+2008     Punctuation space
+         U+2009     Thin space
+         U+200A     Hair space
+         U+202F     Narrow no-break space
+         U+205F     Medium mathematical space
+         U+3000     Ideographic space
+
+       The vertical space characters are:
+
+         U+000A     Linefeed (LF)
+         U+000B     Vertical tab (VT)
+         U+000C     Form feed (FF)
+         U+000D     Carriage return (CR)
+         U+0085     Next line (NEL)
+         U+2028     Line separator
+         U+2029     Paragraph separator
+
+       In 8-bit, non-UTF-8 mode, only the characters with codepoints less than
+       256 are relevant.
+
+   Newline sequences
+
+       Outside  a  character class, by default, the escape sequence \R matches
+       any Unicode newline sequence. In 8-bit non-UTF-8 mode \R is  equivalent
+       to the following:
+
+         (?>\r\n|\n|\x0b|\f|\r|\x85)
+
+       This  is  an  example  of an "atomic group", details of which are given
+       below.  This particular group matches either the two-character sequence
+       CR  followed  by  LF,  or  one  of  the single characters LF (linefeed,
+       U+000A), VT (vertical tab, U+000B), FF (form feed,  U+000C),  CR  (car-
+       riage  return,  U+000D),  or NEL (next line, U+0085). The two-character
+       sequence is treated as a single unit that cannot be split.
+
+       In other modes, two additional characters whose codepoints are  greater
+       than 255 are added: LS (line separator, U+2028) and PS (paragraph sepa-
+       rator, U+2029).  Unicode character property support is not  needed  for
+       these characters to be recognized.
+
+       It is possible to restrict \R to match only CR, LF, or CRLF (instead of
+       the complete set  of  Unicode  line  endings)  by  setting  the  option
+       PCRE_BSR_ANYCRLF either at compile time or when the pattern is matched.
+       (BSR is an abbrevation for "backslash R".) This can be made the default
+       when  PCRE  is  built;  if this is the case, the other behaviour can be
+       requested via the PCRE_BSR_UNICODE option.   It  is  also  possible  to
+       specify  these  settings  by  starting a pattern string with one of the
+       following sequences:
+
+         (*BSR_ANYCRLF)   CR, LF, or CRLF only
+         (*BSR_UNICODE)   any Unicode newline sequence
+
+       These override the default and the options given to the compiling func-
+       tion,  but  they  can  themselves  be  overridden by options given to a
+       matching function. Note that these  special  settings,  which  are  not
+       Perl-compatible,  are  recognized  only at the very start of a pattern,
+       and that they must be in upper case.  If  more  than  one  of  them  is
+       present,  the  last  one is used. They can be combined with a change of
+       newline convention; for example, a pattern can start with:
+
+         (*ANY)(*BSR_ANYCRLF)
+
+       They can also be combined with the (*UTF8), (*UTF16), (*UTF32),  (*UTF)
+       or (*UCP) special sequences. Inside a character class, \R is treated as
+       an unrecognized escape sequence, and  so  matches  the  letter  "R"  by
+       default, but causes an error if PCRE_EXTRA is set.
+
+   Unicode character properties
+
+       When PCRE is built with Unicode character property support, three addi-
+       tional escape sequences that match characters with specific  properties
+       are  available.   When  in 8-bit non-UTF-8 mode, these sequences are of
+       course limited to testing characters whose  codepoints  are  less  than
+       256, but they do work in this mode.  The extra escape sequences are:
+
+         \p{xx}   a character with the xx property
+         \P{xx}   a character without the xx property
+         \X       a Unicode extended grapheme cluster
+
+       The  property  names represented by xx above are limited to the Unicode
+       script names, the general category properties, "Any", which matches any
+       character   (including  newline),  and  some  special  PCRE  properties
+       (described in the next section).  Other Perl properties such as  "InMu-
+       sicalSymbols"  are  not  currently supported by PCRE. Note that \P{Any}
+       does not match any characters, so always causes a match failure.
+
+       Sets of Unicode characters are defined as belonging to certain scripts.
+       A  character from one of these sets can be matched using a script name.
+       For example:
+
+         \p{Greek}
+         \P{Han}
+
+       Those that are not part of an identified script are lumped together  as
+       "Common". The current list of scripts is:
+
+       Arabic,  Armenian,  Avestan, Balinese, Bamum, Batak, Bengali, Bopomofo,
+       Brahmi, Braille, Buginese, Buhid, Canadian_Aboriginal, Carian,  Chakma,
+       Cham,  Cherokee, Common, Coptic, Cuneiform, Cypriot, Cyrillic, Deseret,
+       Devanagari,  Egyptian_Hieroglyphs,  Ethiopic,   Georgian,   Glagolitic,
+       Gothic,  Greek, Gujarati, Gurmukhi, Han, Hangul, Hanunoo, Hebrew, Hira-
+       gana,  Imperial_Aramaic,  Inherited,  Inscriptional_Pahlavi,   Inscrip-
+       tional_Parthian,   Javanese,   Kaithi,   Kannada,  Katakana,  Kayah_Li,
+       Kharoshthi, Khmer, Lao, Latin, Lepcha, Limbu, Linear_B,  Lisu,  Lycian,
+       Lydian,    Malayalam,    Mandaic,    Meetei_Mayek,    Meroitic_Cursive,
+       Meroitic_Hieroglyphs,  Miao,  Mongolian,  Myanmar,  New_Tai_Lue,   Nko,
+       Ogham,    Old_Italic,   Old_Persian,   Old_South_Arabian,   Old_Turkic,
+       Ol_Chiki, Oriya, Osmanya, Phags_Pa, Phoenician, Rejang, Runic,  Samari-
+       tan,  Saurashtra,  Sharada,  Shavian, Sinhala, Sora_Sompeng, Sundanese,
+       Syloti_Nagri, Syriac, Tagalog, Tagbanwa,  Tai_Le,  Tai_Tham,  Tai_Viet,
+       Takri,  Tamil,  Telugu, Thaana, Thai, Tibetan, Tifinagh, Ugaritic, Vai,
+       Yi.
+
+       Each character has exactly one Unicode general category property, spec-
+       ified  by a two-letter abbreviation. For compatibility with Perl, nega-
+       tion can be specified by including a  circumflex  between  the  opening
+       brace  and  the  property  name.  For  example,  \p{^Lu} is the same as
+       \P{Lu}.
+
+       If only one letter is specified with \p or \P, it includes all the gen-
+       eral  category properties that start with that letter. In this case, in
+       the absence of negation, the curly brackets in the escape sequence  are
+       optional; these two examples have the same effect:
+
+         \p{L}
+         \pL
+
+       The following general category property codes are supported:
+
+         C     Other
+         Cc    Control
+         Cf    Format
+         Cn    Unassigned
+         Co    Private use
+         Cs    Surrogate
+
+         L     Letter
+         Ll    Lower case letter
+         Lm    Modifier letter
+         Lo    Other letter
+         Lt    Title case letter
+         Lu    Upper case letter
+
+         M     Mark
+         Mc    Spacing mark
+         Me    Enclosing mark
+         Mn    Non-spacing mark
+
+         N     Number
+         Nd    Decimal number
+         Nl    Letter number
+         No    Other number
+
+         P     Punctuation
+         Pc    Connector punctuation
+         Pd    Dash punctuation
+         Pe    Close punctuation
+         Pf    Final punctuation
+         Pi    Initial punctuation
+         Po    Other punctuation
+         Ps    Open punctuation
+
+         S     Symbol
+         Sc    Currency symbol
+         Sk    Modifier symbol
+         Sm    Mathematical symbol
+         So    Other symbol
+
+         Z     Separator
+         Zl    Line separator
+         Zp    Paragraph separator
+         Zs    Space separator
+
+       The  special property L& is also supported: it matches a character that
+       has the Lu, Ll, or Lt property, in other words, a letter  that  is  not
+       classified as a modifier or "other".
+
+       The  Cs  (Surrogate)  property  applies only to characters in the range
+       U+D800 to U+DFFF. Such characters are not valid in Unicode strings  and
+       so  cannot  be  tested  by  PCRE, unless UTF validity checking has been
+       turned    off    (see    the    discussion    of    PCRE_NO_UTF8_CHECK,
+       PCRE_NO_UTF16_CHECK  and PCRE_NO_UTF32_CHECK in the pcreapi page). Perl
+       does not support the Cs property.
+
+       The long synonyms for  property  names  that  Perl  supports  (such  as
+       \p{Letter})  are  not  supported by PCRE, nor is it permitted to prefix
+       any of these properties with "Is".
+
+       No character that is in the Unicode table has the Cn (unassigned) prop-
+       erty.  Instead, this property is assumed for any code point that is not
+       in the Unicode table.
+
+       Specifying caseless matching does not affect  these  escape  sequences.
+       For  example,  \p{Lu}  always  matches only upper case letters. This is
+       different from the behaviour of current versions of Perl.
+
+       Matching characters by Unicode property is not fast, because  PCRE  has
+       to  do  a  multistage table lookup in order to find a character's prop-
+       erty. That is why the traditional escape sequences such as \d and \w do
+       not use Unicode properties in PCRE by default, though you can make them
+       do so by setting the PCRE_UCP option or by starting  the  pattern  with
+       (*UCP).
+
+   Extended grapheme clusters
+
+       The  \X  escape  matches  any number of Unicode characters that form an
+       "extended grapheme cluster", and treats the sequence as an atomic group
+       (see  below).   Up  to and including release 8.31, PCRE matched an ear-
+       lier, simpler definition that was equivalent to
+
+         (?>\PM\pM*)
+
+       That is, it matched a character without the "mark"  property,  followed
+       by  zero  or  more characters with the "mark" property. Characters with
+       the "mark" property are typically non-spacing accents that  affect  the
+       preceding character.
+
+       This  simple definition was extended in Unicode to include more compli-
+       cated kinds of composite character by giving each character a  grapheme
+       breaking  property,  and  creating  rules  that use these properties to
+       define the boundaries of extended grapheme  clusters.  In  releases  of
+       PCRE later than 8.31, \X matches one of these clusters.
+
+       \X  always  matches  at least one character. Then it decides whether to
+       add additional characters according to the following rules for ending a
+       cluster:
+
+       1. End at the end of the subject string.
+
+       2.  Do not end between CR and LF; otherwise end after any control char-
+       acter.
+
+       3. Do not break Hangul (a Korean  script)  syllable  sequences.  Hangul
+       characters  are of five types: L, V, T, LV, and LVT. An L character may
+       be followed by an L, V, LV, or LVT character; an LV or V character  may
+       be followed by a V or T character; an LVT or T character may be follwed
+       only by a T character.
+
+       4. Do not end before extending characters or spacing marks.  Characters
+       with  the  "mark"  property  always have the "extend" grapheme breaking
+       property.
+
+       5. Do not end after prepend characters.
+
+       6. Otherwise, end the cluster.
+
+   PCRE's additional properties
+
+       As well as the standard Unicode properties described above,  PCRE  sup-
+       ports  four  more  that  make it possible to convert traditional escape
+       sequences such as \w and \s and POSIX character classes to use  Unicode
+       properties.  PCRE  uses  these non-standard, non-Perl properties inter-
+       nally when PCRE_UCP is set. However, they may also be used  explicitly.
+       These properties are:
+
+         Xan   Any alphanumeric character
+         Xps   Any POSIX space character
+         Xsp   Any Perl space character
+         Xwd   Any Perl "word" character
+
+       Xan  matches  characters that have either the L (letter) or the N (num-
+       ber) property. Xps matches the characters tab, linefeed, vertical  tab,
+       form  feed,  or carriage return, and any other character that has the Z
+       (separator) property.  Xsp is the same as Xps, except that vertical tab
+       is excluded. Xwd matches the same characters as Xan, plus underscore.
+
+       There  is another non-standard property, Xuc, which matches any charac-
+       ter that can be represented by a Universal Character Name  in  C++  and
+       other  programming  languages.  These are the characters $, @, ` (grave
+       accent), and all characters with Unicode code points  greater  than  or
+       equal  to U+00A0, except for the surrogates U+D800 to U+DFFF. Note that
+       most base (ASCII) characters are excluded. (Universal  Character  Names
+       are  of  the  form \uHHHH or \UHHHHHHHH where H is a hexadecimal digit.
+       Note that the Xuc property does not match these sequences but the char-
+       acters that they represent.)
+
+   Resetting the match start
+
+       The  escape sequence \K causes any previously matched characters not to
+       be included in the final matched sequence. For example, the pattern:
+
+         foo\Kbar
+
+       matches "foobar", but reports that it has matched "bar".  This  feature
+       is  similar  to  a lookbehind assertion (described below).  However, in
+       this case, the part of the subject before the real match does not  have
+       to  be of fixed length, as lookbehind assertions do. The use of \K does
+       not interfere with the setting of captured  substrings.   For  example,
+       when the pattern
+
+         (foo)\Kbar
+
+       matches "foobar", the first substring is still set to "foo".
+
+       Perl  documents  that  the  use  of  \K  within assertions is "not well
+       defined". In PCRE, \K is acted upon  when  it  occurs  inside  positive
+       assertions, but is ignored in negative assertions.
+
+   Simple assertions
+
+       The  final use of backslash is for certain simple assertions. An asser-
+       tion specifies a condition that has to be met at a particular point  in
+       a  match, without consuming any characters from the subject string. The
+       use of subpatterns for more complicated assertions is described  below.
+       The backslashed assertions are:
+
+         \b     matches at a word boundary
+         \B     matches when not at a word boundary
+         \A     matches at the start of the subject
+         \Z     matches at the end of the subject
+                 also matches before a newline at the end of the subject
+         \z     matches only at the end of the subject
+         \G     matches at the first matching position in the subject
+
+       Inside  a  character  class, \b has a different meaning; it matches the
+       backspace character. If any other of  these  assertions  appears  in  a
+       character  class, by default it matches the corresponding literal char-
+       acter  (for  example,  \B  matches  the  letter  B).  However,  if  the
+       PCRE_EXTRA  option is set, an "invalid escape sequence" error is gener-
+       ated instead.
+
+       A word boundary is a position in the subject string where  the  current
+       character  and  the previous character do not both match \w or \W (i.e.
+       one matches \w and the other matches \W), or the start or  end  of  the
+       string  if  the  first or last character matches \w, respectively. In a
+       UTF mode, the meanings of \w and \W  can  be  changed  by  setting  the
+       PCRE_UCP  option. When this is done, it also affects \b and \B. Neither
+       PCRE nor Perl has a separate "start of word" or "end of  word"  metase-
+       quence.  However,  whatever follows \b normally determines which it is.
+       For example, the fragment \ba matches "a" at the start of a word.
+
+       The \A, \Z, and \z assertions differ from  the  traditional  circumflex
+       and dollar (described in the next section) in that they only ever match
+       at the very start and end of the subject string, whatever  options  are
+       set.  Thus,  they are independent of multiline mode. These three asser-
+       tions are not affected by the PCRE_NOTBOL or PCRE_NOTEOL options, which
+       affect  only the behaviour of the circumflex and dollar metacharacters.
+       However, if the startoffset argument of pcre_exec() is non-zero,  indi-
+       cating that matching is to start at a point other than the beginning of
+       the subject, \A can never match. The difference between \Z  and  \z  is
+       that \Z matches before a newline at the end of the string as well as at
+       the very end, whereas \z matches only at the end.
+
+       The \G assertion is true only when the current matching position is  at
+       the  start point of the match, as specified by the startoffset argument
+       of pcre_exec(). It differs from \A when the  value  of  startoffset  is
+       non-zero.  By calling pcre_exec() multiple times with appropriate argu-
+       ments, you can mimic Perl's /g option, and it is in this kind of imple-
+       mentation where \G can be useful.
+
+       Note,  however,  that  PCRE's interpretation of \G, as the start of the
+       current match, is subtly different from Perl's, which defines it as the
+       end  of  the  previous  match. In Perl, these can be different when the
+       previously matched string was empty. Because PCRE does just  one  match
+       at a time, it cannot reproduce this behaviour.
+
+       If  all  the alternatives of a pattern begin with \G, the expression is
+       anchored to the starting match position, and the "anchored" flag is set
+       in the compiled regular expression.
+
+
+CIRCUMFLEX AND DOLLAR
+
+       The  circumflex  and  dollar  metacharacters are zero-width assertions.
+       That is, they test for a particular condition being true  without  con-
+       suming any characters from the subject string.
+
+       Outside a character class, in the default matching mode, the circumflex
+       character is an assertion that is true only  if  the  current  matching
+       point  is  at the start of the subject string. If the startoffset argu-
+       ment of pcre_exec() is non-zero, circumflex  can  never  match  if  the
+       PCRE_MULTILINE  option  is  unset. Inside a character class, circumflex
+       has an entirely different meaning (see below).
+
+       Circumflex need not be the first character of the pattern if  a  number
+       of  alternatives are involved, but it should be the first thing in each
+       alternative in which it appears if the pattern is ever  to  match  that
+       branch.  If all possible alternatives start with a circumflex, that is,
+       if the pattern is constrained to match only at the start  of  the  sub-
+       ject,  it  is  said  to be an "anchored" pattern. (There are also other
+       constructs that can cause a pattern to be anchored.)
+
+       The dollar character is an assertion that is true only if  the  current
+       matching  point  is  at  the  end of the subject string, or immediately
+       before a newline at the end of the string (by default). Note,  however,
+       that  it  does  not  actually match the newline. Dollar need not be the
+       last character of the pattern if a number of alternatives are involved,
+       but  it should be the last item in any branch in which it appears. Dol-
+       lar has no special meaning in a character class.
+
+       The meaning of dollar can be changed so that it  matches  only  at  the
+       very  end  of  the string, by setting the PCRE_DOLLAR_ENDONLY option at
+       compile time. This does not affect the \Z assertion.
+
+       The meanings of the circumflex and dollar characters are changed if the
+       PCRE_MULTILINE  option  is  set.  When  this  is the case, a circumflex
+       matches immediately after internal newlines as well as at the start  of
+       the  subject  string.  It  does not match after a newline that ends the
+       string. A dollar matches before any newlines in the string, as well  as
+       at  the very end, when PCRE_MULTILINE is set. When newline is specified
+       as the two-character sequence CRLF, isolated CR and  LF  characters  do
+       not indicate newlines.
+
+       For  example, the pattern /^abc$/ matches the subject string "def\nabc"
+       (where \n represents a newline) in multiline mode, but  not  otherwise.
+       Consequently,  patterns  that  are anchored in single line mode because
+       all branches start with ^ are not anchored in  multiline  mode,  and  a
+       match  for  circumflex  is  possible  when  the startoffset argument of
+       pcre_exec() is non-zero. The PCRE_DOLLAR_ENDONLY option is  ignored  if
+       PCRE_MULTILINE is set.
+
+       Note  that  the sequences \A, \Z, and \z can be used to match the start
+       and end of the subject in both modes, and if all branches of a  pattern
+       start  with  \A it is always anchored, whether or not PCRE_MULTILINE is
+       set.
+
+
+FULL STOP (PERIOD, DOT) AND \N
+
+       Outside a character class, a dot in the pattern matches any one charac-
+       ter  in  the subject string except (by default) a character that signi-
+       fies the end of a line.
+
+       When a line ending is defined as a single character, dot never  matches
+       that  character; when the two-character sequence CRLF is used, dot does
+       not match CR if it is immediately followed  by  LF,  but  otherwise  it
+       matches  all characters (including isolated CRs and LFs). When any Uni-
+       code line endings are being recognized, dot does not match CR or LF  or
+       any of the other line ending characters.
+
+       The  behaviour  of  dot  with regard to newlines can be changed. If the
+       PCRE_DOTALL option is set, a dot matches  any  one  character,  without
+       exception. If the two-character sequence CRLF is present in the subject
+       string, it takes two dots to match it.
+
+       The handling of dot is entirely independent of the handling of  circum-
+       flex  and  dollar,  the  only relationship being that they both involve
+       newlines. Dot has no special meaning in a character class.
+
+       The escape sequence \N behaves like  a  dot,  except  that  it  is  not
+       affected  by  the  PCRE_DOTALL  option.  In other words, it matches any
+       character except one that signifies the end of a line. Perl  also  uses
+       \N to match characters by name; PCRE does not support this.
+
+
+MATCHING A SINGLE DATA UNIT
+
+       Outside  a character class, the escape sequence \C matches any one data
+       unit, whether or not a UTF mode is set. In the 8-bit library, one  data
+       unit  is  one  byte;  in the 16-bit library it is a 16-bit unit; in the
+       32-bit library it is a 32-bit unit. Unlike a  dot,  \C  always  matches
+       line-ending  characters.  The  feature  is provided in Perl in order to
+       match individual bytes in UTF-8 mode, but it is unclear how it can use-
+       fully  be  used.  Because  \C breaks up characters into individual data
+       units, matching one unit with \C in a UTF mode means that the  rest  of
+       the string may start with a malformed UTF character. This has undefined
+       results, because PCRE assumes that it is dealing with valid UTF strings
+       (and  by  default  it checks this at the start of processing unless the
+       PCRE_NO_UTF8_CHECK, PCRE_NO_UTF16_CHECK or  PCRE_NO_UTF32_CHECK  option
+       is used).
+
+       PCRE  does  not  allow \C to appear in lookbehind assertions (described
+       below) in a UTF mode, because this would make it impossible  to  calcu-
+       late the length of the lookbehind.
+
+       In general, the \C escape sequence is best avoided. However, one way of
+       using it that avoids the problem of malformed UTF characters is to  use
+       a  lookahead to check the length of the next character, as in this pat-
+       tern, which could be used with a UTF-8 string (ignore white  space  and
+       line breaks):
+
+         (?| (?=[\x00-\x7f])(\C) |
+             (?=[\x80-\x{7ff}])(\C)(\C) |
+             (?=[\x{800}-\x{ffff}])(\C)(\C)(\C) |
+             (?=[\x{10000}-\x{1fffff}])(\C)(\C)(\C)(\C))
+
+       A  group  that starts with (?| resets the capturing parentheses numbers
+       in each alternative (see "Duplicate  Subpattern  Numbers"  below).  The
+       assertions  at  the start of each branch check the next UTF-8 character
+       for values whose encoding uses 1, 2, 3, or 4 bytes,  respectively.  The
+       character's  individual bytes are then captured by the appropriate num-
+       ber of groups.
+
+
+SQUARE BRACKETS AND CHARACTER CLASSES
+
+       An opening square bracket introduces a character class, terminated by a
+       closing square bracket. A closing square bracket on its own is not spe-
+       cial by default.  However, if the PCRE_JAVASCRIPT_COMPAT option is set,
+       a lone closing square bracket causes a compile-time error. If a closing
+       square bracket is required as a member of the class, it should  be  the
+       first  data  character  in  the  class (after an initial circumflex, if
+       present) or escaped with a backslash.
+
+       A character class matches a single character in the subject. In  a  UTF
+       mode,  the  character  may  be  more than one data unit long. A matched
+       character must be in the set of characters defined by the class, unless
+       the  first  character in the class definition is a circumflex, in which
+       case the subject character must not be in the set defined by the class.
+       If  a  circumflex is actually required as a member of the class, ensure
+       it is not the first character, or escape it with a backslash.
+
+       For example, the character class [aeiou] matches any lower case  vowel,
+       while  [^aeiou]  matches  any character that is not a lower case vowel.
+       Note that a circumflex is just a convenient notation for specifying the
+       characters  that  are in the class by enumerating those that are not. A
+       class that starts with a circumflex is not an assertion; it still  con-
+       sumes  a  character  from the subject string, and therefore it fails if
+       the current pointer is at the end of the string.
+
+       In UTF-8 (UTF-16, UTF-32) mode, characters with values greater than 255
+       (0xffff)  can be included in a class as a literal string of data units,
+       or by using the \x{ escaping mechanism.
+
+       When caseless matching is set, any letters in a  class  represent  both
+       their  upper  case  and lower case versions, so for example, a caseless
+       [aeiou] matches "A" as well as "a", and a caseless  [^aeiou]  does  not
+       match  "A", whereas a caseful version would. In a UTF mode, PCRE always
+       understands the concept of case for characters whose  values  are  less
+       than  128, so caseless matching is always possible. For characters with
+       higher values, the concept of case is supported  if  PCRE  is  compiled
+       with  Unicode  property support, but not otherwise.  If you want to use
+       caseless matching in a UTF mode for characters 128 and above, you  must
+       ensure  that  PCRE is compiled with Unicode property support as well as
+       with UTF support.
+
+       Characters that might indicate line breaks are  never  treated  in  any
+       special  way  when  matching  character  classes,  whatever line-ending
+       sequence is in  use,  and  whatever  setting  of  the  PCRE_DOTALL  and
+       PCRE_MULTILINE options is used. A class such as [^a] always matches one
+       of these characters.
+
+       The minus (hyphen) character can be used to specify a range of  charac-
+       ters  in  a  character  class.  For  example,  [d-m] matches any letter
+       between d and m, inclusive. If a  minus  character  is  required  in  a
+       class,  it  must  be  escaped  with a backslash or appear in a position
+       where it cannot be interpreted as indicating a range, typically as  the
+       first or last character in the class.
+
+       It is not possible to have the literal character "]" as the end charac-
+       ter of a range. A pattern such as [W-]46] is interpreted as a class  of
+       two  characters ("W" and "-") followed by a literal string "46]", so it
+       would match "W46]" or "-46]". However, if the "]"  is  escaped  with  a
+       backslash  it is interpreted as the end of range, so [W-\]46] is inter-
+       preted as a class containing a range followed by two other  characters.
+       The  octal or hexadecimal representation of "]" can also be used to end
+       a range.
+
+       Ranges operate in the collating sequence of character values. They  can
+       also   be  used  for  characters  specified  numerically,  for  example
+       [\000-\037]. Ranges can include any characters that are valid  for  the
+       current mode.
+
+       If a range that includes letters is used when caseless matching is set,
+       it matches the letters in either case. For example, [W-c] is equivalent
+       to  [][\\^_`wxyzabc],  matched  caselessly,  and  in a non-UTF mode, if
+       character tables for a French locale are in  use,  [\xc8-\xcb]  matches
+       accented  E  characters  in both cases. In UTF modes, PCRE supports the
+       concept of case for characters with values greater than 128  only  when
+       it is compiled with Unicode property support.
+
+       The  character escape sequences \d, \D, \h, \H, \p, \P, \s, \S, \v, \V,
+       \w, and \W may appear in a character class, and add the characters that
+       they  match to the class. For example, [\dABCDEF] matches any hexadeci-
+       mal digit. In UTF modes, the PCRE_UCP option affects  the  meanings  of
+       \d,  \s,  \w  and  their upper case partners, just as it does when they
+       appear outside a character class, as described in the section  entitled
+       "Generic character types" above. The escape sequence \b has a different
+       meaning inside a character class; it matches the  backspace  character.
+       The  sequences  \B,  \N,  \R, and \X are not special inside a character
+       class. Like any other unrecognized escape sequences, they  are  treated
+       as  the literal characters "B", "N", "R", and "X" by default, but cause
+       an error if the PCRE_EXTRA option is set.
+
+       A circumflex can conveniently be used with  the  upper  case  character
+       types  to specify a more restricted set of characters than the matching
+       lower case type.  For example, the class [^\W_] matches any  letter  or
+       digit, but not underscore, whereas [\w] includes underscore. A positive
+       character class should be read as "something OR something OR ..." and a
+       negative class as "NOT something AND NOT something AND NOT ...".
+
+       The  only  metacharacters  that are recognized in character classes are
+       backslash, hyphen (only where it can be  interpreted  as  specifying  a
+       range),  circumflex  (only  at the start), opening square bracket (only
+       when it can be interpreted as introducing a POSIX class name - see  the
+       next  section),  and  the  terminating closing square bracket. However,
+       escaping other non-alphanumeric characters does no harm.
+
+
+POSIX CHARACTER CLASSES
+
+       Perl supports the POSIX notation for character classes. This uses names
+       enclosed  by  [: and :] within the enclosing square brackets. PCRE also
+       supports this notation. For example,
+
+         [01[:alpha:]%]
+
+       matches "0", "1", any alphabetic character, or "%". The supported class
+       names are:
+
+         alnum    letters and digits
+         alpha    letters
+         ascii    character codes 0 - 127
+         blank    space or tab only
+         cntrl    control characters
+         digit    decimal digits (same as \d)
+         graph    printing characters, excluding space
+         lower    lower case letters
+         print    printing characters, including space
+         punct    printing characters, excluding letters and digits and space
+         space    white space (not quite the same as \s)
+         upper    upper case letters
+         word     "word" characters (same as \w)
+         xdigit   hexadecimal digits
+
+       The  "space" characters are HT (9), LF (10), VT (11), FF (12), CR (13),
+       and space (32). Notice that this list includes the VT  character  (code
+       11). This makes "space" different to \s, which does not include VT (for
+       Perl compatibility).
+
+       The name "word" is a Perl extension, and "blank"  is  a  GNU  extension
+       from  Perl  5.8. Another Perl extension is negation, which is indicated
+       by a ^ character after the colon. For example,
+
+         [12[:^digit:]]
+
+       matches "1", "2", or any non-digit. PCRE (and Perl) also recognize  the
+       POSIX syntax [.ch.] and [=ch=] where "ch" is a "collating element", but
+       these are not supported, and an error is given if they are encountered.
+
+       By default, in UTF modes, characters with values greater  than  128  do
+       not  match any of the POSIX character classes. However, if the PCRE_UCP
+       option is passed to pcre_compile(), some of the classes are changed  so
+       that Unicode character properties are used. This is achieved by replac-
+       ing the POSIX classes by other sequences, as follows:
+
+         [:alnum:]  becomes  \p{Xan}
+         [:alpha:]  becomes  \p{L}
+         [:blank:]  becomes  \h
+         [:digit:]  becomes  \p{Nd}
+         [:lower:]  becomes  \p{Ll}
+         [:space:]  becomes  \p{Xps}
+         [:upper:]  becomes  \p{Lu}
+         [:word:]   becomes  \p{Xwd}
+
+       Negated versions, such as [:^alpha:] use \P instead of  \p.  The  other
+       POSIX classes are unchanged, and match only characters with code points
+       less than 128.
+
+
+VERTICAL BAR
+
+       Vertical bar characters are used to separate alternative patterns.  For
+       example, the pattern
+
+         gilbert|sullivan
+
+       matches  either "gilbert" or "sullivan". Any number of alternatives may
+       appear, and an empty  alternative  is  permitted  (matching  the  empty
+       string). The matching process tries each alternative in turn, from left
+       to right, and the first one that succeeds is used. If the  alternatives
+       are  within a subpattern (defined below), "succeeds" means matching the
+       rest of the main pattern as well as the alternative in the subpattern.
+
+
+INTERNAL OPTION SETTING
+
+       The settings of the  PCRE_CASELESS,  PCRE_MULTILINE,  PCRE_DOTALL,  and
+       PCRE_EXTENDED  options  (which are Perl-compatible) can be changed from
+       within the pattern by  a  sequence  of  Perl  option  letters  enclosed
+       between "(?" and ")".  The option letters are
+
+         i  for PCRE_CASELESS
+         m  for PCRE_MULTILINE
+         s  for PCRE_DOTALL
+         x  for PCRE_EXTENDED
+
+       For example, (?im) sets caseless, multiline matching. It is also possi-
+       ble to unset these options by preceding the letter with a hyphen, and a
+       combined  setting and unsetting such as (?im-sx), which sets PCRE_CASE-
+       LESS and PCRE_MULTILINE while unsetting PCRE_DOTALL and  PCRE_EXTENDED,
+       is  also  permitted.  If  a  letter  appears  both before and after the
+       hyphen, the option is unset.
+
+       The PCRE-specific options PCRE_DUPNAMES, PCRE_UNGREEDY, and  PCRE_EXTRA
+       can  be changed in the same way as the Perl-compatible options by using
+       the characters J, U and X respectively.
+
+       When one of these option changes occurs at  top  level  (that  is,  not
+       inside  subpattern parentheses), the change applies to the remainder of
+       the pattern that follows. If the change is placed right at the start of
+       a pattern, PCRE extracts it into the global options (and it will there-
+       fore show up in data extracted by the pcre_fullinfo() function).
+
+       An option change within a subpattern (see below for  a  description  of
+       subpatterns)  affects only that part of the subpattern that follows it,
+       so
+
+         (a(?i)b)c
+
+       matches abc and aBc and no other strings (assuming PCRE_CASELESS is not
+       used).   By  this means, options can be made to have different settings
+       in different parts of the pattern. Any changes made in one  alternative
+       do  carry  on  into subsequent branches within the same subpattern. For
+       example,
+
+         (a(?i)b|c)
+
+       matches "ab", "aB", "c", and "C", even though  when  matching  "C"  the
+       first  branch  is  abandoned before the option setting. This is because
+       the effects of option settings happen at compile time. There  would  be
+       some very weird behaviour otherwise.
+
+       Note:  There  are  other  PCRE-specific  options that can be set by the
+       application when the compiling or matching  functions  are  called.  In
+       some  cases  the  pattern can contain special leading sequences such as
+       (*CRLF) to override what the application  has  set  or  what  has  been
+       defaulted.   Details   are  given  in  the  section  entitled  "Newline
+       sequences" above. There are also the  (*UTF8),  (*UTF16),(*UTF32),  and
+       (*UCP)  leading sequences that can be used to set UTF and Unicode prop-
+       erty modes; they are equivalent to setting the  PCRE_UTF8,  PCRE_UTF16,
+       PCRE_UTF32  and the PCRE_UCP options, respectively. The (*UTF) sequence
+       is a generic version that can be used with any of the  libraries.  How-
+       ever,  the  application  can set the PCRE_NEVER_UTF option, which locks
+       out the use of the (*UTF) sequences.
+
+
+SUBPATTERNS
+
+       Subpatterns are delimited by parentheses (round brackets), which can be
+       nested.  Turning part of a pattern into a subpattern does two things:
+
+       1. It localizes a set of alternatives. For example, the pattern
+
+         cat(aract|erpillar|)
+
+       matches  "cataract",  "caterpillar", or "cat". Without the parentheses,
+       it would match "cataract", "erpillar" or an empty string.
+
+       2. It sets up the subpattern as  a  capturing  subpattern.  This  means
+       that,  when  the  whole  pattern  matches,  that portion of the subject
+       string that matched the subpattern is passed back to the caller via the
+       ovector  argument  of  the matching function. (This applies only to the
+       traditional matching functions; the DFA matching functions do not  sup-
+       port capturing.)
+
+       Opening parentheses are counted from left to right (starting from 1) to
+       obtain numbers for the  capturing  subpatterns.  For  example,  if  the
+       string "the red king" is matched against the pattern
+
+         the ((red|white) (king|queen))
+
+       the captured substrings are "red king", "red", and "king", and are num-
+       bered 1, 2, and 3, respectively.
+
+       The fact that plain parentheses fulfil  two  functions  is  not  always
+       helpful.   There are often times when a grouping subpattern is required
+       without a capturing requirement. If an opening parenthesis is  followed
+       by  a question mark and a colon, the subpattern does not do any captur-
+       ing, and is not counted when computing the  number  of  any  subsequent
+       capturing  subpatterns. For example, if the string "the white queen" is
+       matched against the pattern
+
+         the ((?:red|white) (king|queen))
+
+       the captured substrings are "white queen" and "queen", and are numbered
+       1 and 2. The maximum number of capturing subpatterns is 65535.
+
+       As  a  convenient shorthand, if any option settings are required at the
+       start of a non-capturing subpattern,  the  option  letters  may  appear
+       between the "?" and the ":". Thus the two patterns
+
+         (?i:saturday|sunday)
+         (?:(?i)saturday|sunday)
+
+       match exactly the same set of strings. Because alternative branches are
+       tried from left to right, and options are not reset until  the  end  of
+       the  subpattern is reached, an option setting in one branch does affect
+       subsequent branches, so the above patterns match "SUNDAY"  as  well  as
+       "Saturday".
+
+
+DUPLICATE SUBPATTERN NUMBERS
+
+       Perl 5.10 introduced a feature whereby each alternative in a subpattern
+       uses the same numbers for its capturing parentheses. Such a  subpattern
+       starts  with (?| and is itself a non-capturing subpattern. For example,
+       consider this pattern:
+
+         (?|(Sat)ur|(Sun))day
+
+       Because the two alternatives are inside a (?| group, both sets of  cap-
+       turing  parentheses  are  numbered one. Thus, when the pattern matches,
+       you can look at captured substring number  one,  whichever  alternative
+       matched.  This  construct  is useful when you want to capture part, but
+       not all, of one of a number of alternatives. Inside a (?| group, paren-
+       theses  are  numbered as usual, but the number is reset at the start of
+       each branch. The numbers of any capturing parentheses that  follow  the
+       subpattern  start after the highest number used in any branch. The fol-
+       lowing example is taken from the Perl documentation. The numbers under-
+       neath show in which buffer the captured content will be stored.
+
+         # before  ---------------branch-reset----------- after
+         / ( a )  (?| x ( y ) z | (p (q) r) | (t) u (v) ) ( z ) /x
+         # 1            2         2  3        2     3     4
+
+       A  back  reference  to a numbered subpattern uses the most recent value
+       that is set for that number by any subpattern.  The  following  pattern
+       matches "abcabc" or "defdef":
+
+         /(?|(abc)|(def))\1/
+
+       In  contrast,  a subroutine call to a numbered subpattern always refers
+       to the first one in the pattern with the given  number.  The  following
+       pattern matches "abcabc" or "defabc":
+
+         /(?|(abc)|(def))(?1)/
+
+       If  a condition test for a subpattern's having matched refers to a non-
+       unique number, the test is true if any of the subpatterns of that  num-
+       ber have matched.
+
+       An  alternative approach to using this "branch reset" feature is to use
+       duplicate named subpatterns, as described in the next section.
+
+
+NAMED SUBPATTERNS
+
+       Identifying capturing parentheses by number is simple, but  it  can  be
+       very  hard  to keep track of the numbers in complicated regular expres-
+       sions. Furthermore, if an  expression  is  modified,  the  numbers  may
+       change.  To help with this difficulty, PCRE supports the naming of sub-
+       patterns. This feature was not added to Perl until release 5.10. Python
+       had  the  feature earlier, and PCRE introduced it at release 4.0, using
+       the Python syntax. PCRE now supports both the Perl and the Python  syn-
+       tax.  Perl  allows  identically  numbered subpatterns to have different
+       names, but PCRE does not.
+
+       In PCRE, a subpattern can be named in one of three  ways:  (?...)
+       or  (?'name'...)  as in Perl, or (?P...) as in Python. References
+       to capturing parentheses from other parts of the pattern, such as  back
+       references,  recursion,  and conditions, can be made by name as well as
+       by number.
+
+       Names consist of up to  32  alphanumeric  characters  and  underscores.
+       Named  capturing  parentheses  are  still  allocated numbers as well as
+       names, exactly as if the names were not present. The PCRE API  provides
+       function calls for extracting the name-to-number translation table from
+       a compiled pattern. There is also a convenience function for extracting
+       a captured substring by name.
+
+       By  default, a name must be unique within a pattern, but it is possible
+       to relax this constraint by setting the PCRE_DUPNAMES option at compile
+       time.  (Duplicate  names are also always permitted for subpatterns with
+       the same number, set up as described in the previous  section.)  Dupli-
+       cate  names  can  be useful for patterns where only one instance of the
+       named parentheses can match. Suppose you want to match the  name  of  a
+       weekday,  either as a 3-letter abbreviation or as the full name, and in
+       both cases you want to extract the abbreviation. This pattern (ignoring
+       the line breaks) does the job:
+
+         (?Mon|Fri|Sun)(?:day)?|
+         (?Tue)(?:sday)?|
+         (?Wed)(?:nesday)?|
+         (?Thu)(?:rsday)?|
+         (?Sat)(?:urday)?
+
+       There  are  five capturing substrings, but only one is ever set after a
+       match.  (An alternative way of solving this problem is to use a "branch
+       reset" subpattern, as described in the previous section.)
+
+       The  convenience  function  for extracting the data by name returns the
+       substring for the first (and in this example, the only)  subpattern  of
+       that  name  that  matched.  This saves searching to find which numbered
+       subpattern it was.
+
+       If you make a back reference to  a  non-unique  named  subpattern  from
+       elsewhere  in the pattern, the one that corresponds to the first occur-
+       rence of the name is used. In the absence of duplicate numbers (see the
+       previous  section) this is the one with the lowest number. If you use a
+       named reference in a condition test (see the section  about  conditions
+       below),  either  to check whether a subpattern has matched, or to check
+       for recursion, all subpatterns with the same name are  tested.  If  the
+       condition  is  true for any one of them, the overall condition is true.
+       This is the same behaviour as testing by number. For further details of
+       the interfaces for handling named subpatterns, see the pcreapi documen-
+       tation.
+
+       Warning: You cannot use different names to distinguish between two sub-
+       patterns  with  the same number because PCRE uses only the numbers when
+       matching. For this reason, an error is given at compile time if differ-
+       ent  names  are given to subpatterns with the same number. However, you
+       can give the same name to subpatterns with the same number,  even  when
+       PCRE_DUPNAMES is not set.
+
+
+REPETITION
+
+       Repetition  is  specified  by  quantifiers, which can follow any of the
+       following items:
+
+         a literal data character
+         the dot metacharacter
+         the \C escape sequence
+         the \X escape sequence
+         the \R escape sequence
+         an escape such as \d or \pL that matches a single character
+         a character class
+         a back reference (see next section)
+         a parenthesized subpattern (including assertions)
+         a subroutine call to a subpattern (recursive or otherwise)
+
+       The general repetition quantifier specifies a minimum and maximum  num-
+       ber  of  permitted matches, by giving the two numbers in curly brackets
+       (braces), separated by a comma. The numbers must be  less  than  65536,
+       and the first must be less than or equal to the second. For example:
+
+         z{2,4}
+
+       matches  "zz",  "zzz",  or  "zzzz". A closing brace on its own is not a
+       special character. If the second number is omitted, but  the  comma  is
+       present,  there  is  no upper limit; if the second number and the comma
+       are both omitted, the quantifier specifies an exact number of  required
+       matches. Thus
+
+         [aeiou]{3,}
+
+       matches at least 3 successive vowels, but may match many more, while
+
+         \d{8}
+
+       matches  exactly  8  digits. An opening curly bracket that appears in a
+       position where a quantifier is not allowed, or one that does not  match
+       the  syntax of a quantifier, is taken as a literal character. For exam-
+       ple, {,6} is not a quantifier, but a literal string of four characters.
+
+       In UTF modes, quantifiers apply to characters rather than to individual
+       data  units. Thus, for example, \x{100}{2} matches two characters, each
+       of which is represented by a two-byte sequence in a UTF-8 string. Simi-
+       larly,  \X{3} matches three Unicode extended grapheme clusters, each of
+       which may be several data units long (and  they  may  be  of  different
+       lengths).
+
+       The quantifier {0} is permitted, causing the expression to behave as if
+       the previous item and the quantifier were not present. This may be use-
+       ful  for  subpatterns that are referenced as subroutines from elsewhere
+       in the pattern (but see also the section entitled "Defining subpatterns
+       for  use  by  reference only" below). Items other than subpatterns that
+       have a {0} quantifier are omitted from the compiled pattern.
+
+       For convenience, the three most common quantifiers have  single-charac-
+       ter abbreviations:
+
+         *    is equivalent to {0,}
+         +    is equivalent to {1,}
+         ?    is equivalent to {0,1}
+
+       It  is  possible  to construct infinite loops by following a subpattern
+       that can match no characters with a quantifier that has no upper limit,
+       for example:
+
+         (a?)*
+
+       Earlier versions of Perl and PCRE used to give an error at compile time
+       for such patterns. However, because there are cases where this  can  be
+       useful,  such  patterns  are now accepted, but if any repetition of the
+       subpattern does in fact match no characters, the loop is forcibly  bro-
+       ken.
+
+       By  default,  the quantifiers are "greedy", that is, they match as much
+       as possible (up to the maximum  number  of  permitted  times),  without
+       causing  the  rest of the pattern to fail. The classic example of where
+       this gives problems is in trying to match comments in C programs. These
+       appear  between  /*  and  */ and within the comment, individual * and /
+       characters may appear. An attempt to match C comments by  applying  the
+       pattern
+
+         /\*.*\*/
+
+       to the string
+
+         /* first comment */  not comment  /* second comment */
+
+       fails,  because it matches the entire string owing to the greediness of
+       the .*  item.
+
+       However, if a quantifier is followed by a question mark, it  ceases  to
+       be greedy, and instead matches the minimum number of times possible, so
+       the pattern
+
+         /\*.*?\*/
+
+       does the right thing with the C comments. The meaning  of  the  various
+       quantifiers  is  not  otherwise  changed,  just the preferred number of
+       matches.  Do not confuse this use of question mark with its  use  as  a
+       quantifier  in its own right. Because it has two uses, it can sometimes
+       appear doubled, as in
+
+         \d??\d
+
+       which matches one digit by preference, but can match two if that is the
+       only way the rest of the pattern matches.
+
+       If  the PCRE_UNGREEDY option is set (an option that is not available in
+       Perl), the quantifiers are not greedy by default, but  individual  ones
+       can  be  made  greedy  by following them with a question mark. In other
+       words, it inverts the default behaviour.
+
+       When a parenthesized subpattern is quantified  with  a  minimum  repeat
+       count  that is greater than 1 or with a limited maximum, more memory is
+       required for the compiled pattern, in proportion to  the  size  of  the
+       minimum or maximum.
+
+       If a pattern starts with .* or .{0,} and the PCRE_DOTALL option (equiv-
+       alent to Perl's /s) is set, thus allowing the dot  to  match  newlines,
+       the  pattern  is  implicitly anchored, because whatever follows will be
+       tried against every character position in the subject string, so  there
+       is  no  point  in  retrying the overall match at any position after the
+       first. PCRE normally treats such a pattern as though it  were  preceded
+       by \A.
+
+       In  cases  where  it  is known that the subject string contains no new-
+       lines, it is worth setting PCRE_DOTALL in order to  obtain  this  opti-
+       mization, or alternatively using ^ to indicate anchoring explicitly.
+
+       However,  there  are  some cases where the optimization cannot be used.
+       When .*  is inside capturing parentheses that are the subject of a back
+       reference elsewhere in the pattern, a match at the start may fail where
+       a later one succeeds. Consider, for example:
+
+         (.*)abc\1
+
+       If the subject is "xyz123abc123" the match point is the fourth  charac-
+       ter. For this reason, such a pattern is not implicitly anchored.
+
+       Another  case where implicit anchoring is not applied is when the lead-
+       ing .* is inside an atomic group. Once again, a match at the start  may
+       fail where a later one succeeds. Consider this pattern:
+
+         (?>.*?a)b
+
+       It  matches "ab" in the subject "aab". The use of the backtracking con-
+       trol verbs (*PRUNE) and (*SKIP) also disable this optimization.
+
+       When a capturing subpattern is repeated, the value captured is the sub-
+       string that matched the final iteration. For example, after
+
+         (tweedle[dume]{3}\s*)+
+
+       has matched "tweedledum tweedledee" the value of the captured substring
+       is "tweedledee". However, if there are  nested  capturing  subpatterns,
+       the  corresponding captured values may have been set in previous itera-
+       tions. For example, after
+
+         /(a|(b))+/
+
+       matches "aba" the value of the second captured substring is "b".
+
+
+ATOMIC GROUPING AND POSSESSIVE QUANTIFIERS
+
+       With both maximizing ("greedy") and minimizing ("ungreedy"  or  "lazy")
+       repetition,  failure  of what follows normally causes the repeated item
+       to be re-evaluated to see if a different number of repeats  allows  the
+       rest  of  the pattern to match. Sometimes it is useful to prevent this,
+       either to change the nature of the match, or to cause it  fail  earlier
+       than  it otherwise might, when the author of the pattern knows there is
+       no point in carrying on.
+
+       Consider, for example, the pattern \d+foo when applied to  the  subject
+       line
+
+         123456bar
+
+       After matching all 6 digits and then failing to match "foo", the normal
+       action of the matcher is to try again with only 5 digits  matching  the
+       \d+  item,  and  then  with  4,  and  so on, before ultimately failing.
+       "Atomic grouping" (a term taken from Jeffrey  Friedl's  book)  provides
+       the  means for specifying that once a subpattern has matched, it is not
+       to be re-evaluated in this way.
+
+       If we use atomic grouping for the previous example, the  matcher  gives
+       up  immediately  on failing to match "foo" the first time. The notation
+       is a kind of special parenthesis, starting with (?> as in this example:
+
+         (?>\d+)foo
+
+       This kind of parenthesis "locks up" the  part of the  pattern  it  con-
+       tains  once  it  has matched, and a failure further into the pattern is
+       prevented from backtracking into it. Backtracking past it  to  previous
+       items, however, works as normal.
+
+       An  alternative  description  is that a subpattern of this type matches
+       the string of characters that an  identical  standalone  pattern  would
+       match, if anchored at the current point in the subject string.
+
+       Atomic grouping subpatterns are not capturing subpatterns. Simple cases
+       such as the above example can be thought of as a maximizing repeat that
+       must  swallow  everything  it can. So, while both \d+ and \d+? are pre-
+       pared to adjust the number of digits they match in order  to  make  the
+       rest of the pattern match, (?>\d+) can only match an entire sequence of
+       digits.
+
+       Atomic groups in general can of course contain arbitrarily  complicated
+       subpatterns,  and  can  be  nested. However, when the subpattern for an
+       atomic group is just a single repeated item, as in the example above, a
+       simpler  notation,  called  a "possessive quantifier" can be used. This
+       consists of an additional + character  following  a  quantifier.  Using
+       this notation, the previous example can be rewritten as
+
+         \d++foo
+
+       Note that a possessive quantifier can be used with an entire group, for
+       example:
+
+         (abc|xyz){2,3}+
+
+       Possessive  quantifiers  are  always  greedy;  the   setting   of   the
+       PCRE_UNGREEDY option is ignored. They are a convenient notation for the
+       simpler forms of atomic group. However, there is no difference  in  the
+       meaning  of  a  possessive  quantifier and the equivalent atomic group,
+       though there may be a performance  difference;  possessive  quantifiers
+       should be slightly faster.
+
+       The  possessive  quantifier syntax is an extension to the Perl 5.8 syn-
+       tax.  Jeffrey Friedl originated the idea (and the name)  in  the  first
+       edition of his book. Mike McCloskey liked it, so implemented it when he
+       built Sun's Java package, and PCRE copied it from there. It  ultimately
+       found its way into Perl at release 5.10.
+
+       PCRE has an optimization that automatically "possessifies" certain sim-
+       ple pattern constructs. For example, the sequence  A+B  is  treated  as
+       A++B  because  there is no point in backtracking into a sequence of A's
+       when B must follow.
+
+       When a pattern contains an unlimited repeat inside  a  subpattern  that
+       can  itself  be  repeated  an  unlimited number of times, the use of an
+       atomic group is the only way to avoid some  failing  matches  taking  a
+       very long time indeed. The pattern
+
+         (\D+|<\d+>)*[!?]
+
+       matches  an  unlimited number of substrings that either consist of non-
+       digits, or digits enclosed in <>, followed by either ! or  ?.  When  it
+       matches, it runs quickly. However, if it is applied to
+
+         aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+
+       it  takes  a  long  time  before reporting failure. This is because the
+       string can be divided between the internal \D+ repeat and the  external
+       *  repeat  in  a  large  number of ways, and all have to be tried. (The
+       example uses [!?] rather than a single character at  the  end,  because
+       both  PCRE  and  Perl have an optimization that allows for fast failure
+       when a single character is used. They remember the last single  charac-
+       ter  that  is required for a match, and fail early if it is not present
+       in the string.) If the pattern is changed so that  it  uses  an  atomic
+       group, like this:
+
+         ((?>\D+)|<\d+>)*[!?]
+
+       sequences of non-digits cannot be broken, and failure happens quickly.
+
+
+BACK REFERENCES
+
+       Outside a character class, a backslash followed by a digit greater than
+       0 (and possibly further digits) is a back reference to a capturing sub-
+       pattern  earlier  (that is, to its left) in the pattern, provided there
+       have been that many previous capturing left parentheses.
+
+       However, if the decimal number following the backslash is less than 10,
+       it  is  always  taken  as a back reference, and causes an error only if
+       there are not that many capturing left parentheses in the  entire  pat-
+       tern.  In  other words, the parentheses that are referenced need not be
+       to the left of the reference for numbers less than 10. A "forward  back
+       reference"  of  this  type can make sense when a repetition is involved
+       and the subpattern to the right has participated in an  earlier  itera-
+       tion.
+
+       It  is  not  possible to have a numerical "forward back reference" to a
+       subpattern whose number is 10 or  more  using  this  syntax  because  a
+       sequence  such  as  \50 is interpreted as a character defined in octal.
+       See the subsection entitled "Non-printing characters" above for further
+       details  of  the  handling of digits following a backslash. There is no
+       such problem when named parentheses are used. A back reference  to  any
+       subpattern is possible using named parentheses (see below).
+
+       Another  way  of  avoiding  the ambiguity inherent in the use of digits
+       following a backslash is to use the \g  escape  sequence.  This  escape
+       must be followed by an unsigned number or a negative number, optionally
+       enclosed in braces. These examples are all identical:
+
+         (ring), \1
+         (ring), \g1
+         (ring), \g{1}
+
+       An unsigned number specifies an absolute reference without the  ambigu-
+       ity that is present in the older syntax. It is also useful when literal
+       digits follow the reference. A negative number is a relative reference.
+       Consider this example:
+
+         (abc(def)ghi)\g{-1}
+
+       The sequence \g{-1} is a reference to the most recently started captur-
+       ing subpattern before \g, that is, is it equivalent to \2 in this exam-
+       ple.   Similarly, \g{-2} would be equivalent to \1. The use of relative
+       references can be helpful in long patterns, and also in  patterns  that
+       are  created  by  joining  together  fragments  that contain references
+       within themselves.
+
+       A back reference matches whatever actually matched the  capturing  sub-
+       pattern  in  the  current subject string, rather than anything matching
+       the subpattern itself (see "Subpatterns as subroutines" below for a way
+       of doing that). So the pattern
+
+         (sens|respons)e and \1ibility
+
+       matches  "sense and sensibility" and "response and responsibility", but
+       not "sense and responsibility". If caseful matching is in force at  the
+       time  of the back reference, the case of letters is relevant. For exam-
+       ple,
+
+         ((?i)rah)\s+\1
+
+       matches "rah rah" and "RAH RAH", but not "RAH  rah",  even  though  the
+       original capturing subpattern is matched caselessly.
+
+       There  are  several  different ways of writing back references to named
+       subpatterns. The .NET syntax \k{name} and the Perl syntax  \k  or
+       \k'name'  are supported, as is the Python syntax (?P=name). Perl 5.10's
+       unified back reference syntax, in which \g can be used for both numeric
+       and  named  references,  is  also supported. We could rewrite the above
+       example in any of the following ways:
+
+         (?(?i)rah)\s+\k
+         (?'p1'(?i)rah)\s+\k{p1}
+         (?P(?i)rah)\s+(?P=p1)
+         (?(?i)rah)\s+\g{p1}
+
+       A subpattern that is referenced by  name  may  appear  in  the  pattern
+       before or after the reference.
+
+       There  may be more than one back reference to the same subpattern. If a
+       subpattern has not actually been used in a particular match,  any  back
+       references to it always fail by default. For example, the pattern
+
+         (a|(bc))\2
+
+       always  fails  if  it starts to match "a" rather than "bc". However, if
+       the PCRE_JAVASCRIPT_COMPAT option is set at compile time, a back refer-
+       ence to an unset value matches an empty string.
+
+       Because  there may be many capturing parentheses in a pattern, all dig-
+       its following a backslash are taken as part of a potential back  refer-
+       ence  number.   If  the  pattern continues with a digit character, some
+       delimiter must  be  used  to  terminate  the  back  reference.  If  the
+       PCRE_EXTENDED  option  is  set, this can be white space. Otherwise, the
+       \g{ syntax or an empty comment (see "Comments" below) can be used.
+
+   Recursive back references
+
+       A back reference that occurs inside the parentheses to which it  refers
+       fails  when  the subpattern is first used, so, for example, (a\1) never
+       matches.  However, such references can be useful inside  repeated  sub-
+       patterns. For example, the pattern
+
+         (a|b\1)+
+
+       matches any number of "a"s and also "aba", "ababbaa" etc. At each iter-
+       ation of the subpattern,  the  back  reference  matches  the  character
+       string  corresponding  to  the previous iteration. In order for this to
+       work, the pattern must be such that the first iteration does  not  need
+       to  match the back reference. This can be done using alternation, as in
+       the example above, or by a quantifier with a minimum of zero.
+
+       Back references of this type cause the group that they reference to  be
+       treated  as  an atomic group.  Once the whole group has been matched, a
+       subsequent matching failure cannot cause backtracking into  the  middle
+       of the group.
+
+
+ASSERTIONS
+
+       An  assertion  is  a  test on the characters following or preceding the
+       current matching point that does not actually consume  any  characters.
+       The  simple  assertions  coded  as  \b, \B, \A, \G, \Z, \z, ^ and $ are
+       described above.
+
+       More complicated assertions are coded as  subpatterns.  There  are  two
+       kinds:  those  that  look  ahead of the current position in the subject
+       string, and those that look  behind  it.  An  assertion  subpattern  is
+       matched  in  the  normal way, except that it does not cause the current
+       matching position to be changed.
+
+       Assertion subpatterns are not capturing subpatterns. If such an  asser-
+       tion  contains  capturing  subpatterns within it, these are counted for
+       the purposes of numbering the capturing subpatterns in the  whole  pat-
+       tern.  However,  substring  capturing  is carried out only for positive
+       assertions. (Perl sometimes, but not always, does do capturing in nega-
+       tive assertions.)
+
+       For  compatibility  with  Perl,  assertion subpatterns may be repeated;
+       though it makes no sense to assert the same thing  several  times,  the
+       side  effect  of  capturing  parentheses may occasionally be useful. In
+       practice, there only three cases:
+
+       (1) If the quantifier is {0}, the  assertion  is  never  obeyed  during
+       matching.   However,  it  may  contain internal capturing parenthesized
+       groups that are called from elsewhere via the subroutine mechanism.
+
+       (2) If quantifier is {0,n} where n is greater than zero, it is  treated
+       as  if  it  were  {0,1}.  At run time, the rest of the pattern match is
+       tried with and without the assertion, the order depending on the greed-
+       iness of the quantifier.
+
+       (3)  If  the minimum repetition is greater than zero, the quantifier is
+       ignored.  The assertion is obeyed just  once  when  encountered  during
+       matching.
+
+   Lookahead assertions
+
+       Lookahead assertions start with (?= for positive assertions and (?! for
+       negative assertions. For example,
+
+         \w+(?=;)
+
+       matches a word followed by a semicolon, but does not include the  semi-
+       colon in the match, and
+
+         foo(?!bar)
+
+       matches  any  occurrence  of  "foo" that is not followed by "bar". Note
+       that the apparently similar pattern
+
+         (?!foo)bar
+
+       does not find an occurrence of "bar"  that  is  preceded  by  something
+       other  than "foo"; it finds any occurrence of "bar" whatsoever, because
+       the assertion (?!foo) is always true when the next three characters are
+       "bar". A lookbehind assertion is needed to achieve the other effect.
+
+       If you want to force a matching failure at some point in a pattern, the
+       most convenient way to do it is  with  (?!)  because  an  empty  string
+       always  matches, so an assertion that requires there not to be an empty
+       string must always fail.  The backtracking control verb (*FAIL) or (*F)
+       is a synonym for (?!).
+
+   Lookbehind assertions
+
+       Lookbehind  assertions start with (?<= for positive assertions and (?)...) or (?('name')...)  to  test  for  a
+       used  subpattern  by  name.  For compatibility with earlier versions of
+       PCRE, which had this facility before Perl, the syntax  (?(name)...)  is
+       also  recognized. However, there is a possible ambiguity with this syn-
+       tax, because subpattern names may  consist  entirely  of  digits.  PCRE
+       looks  first for a named subpattern; if it cannot find one and the name
+       consists entirely of digits, PCRE looks for a subpattern of  that  num-
+       ber,  which must be greater than zero. Using subpattern names that con-
+       sist entirely of digits is not recommended.
+
+       Rewriting the above example to use a named subpattern gives this:
+
+         (? \( )?    [^()]+    (?() \) )
+
+       If the name used in a condition of this kind is a duplicate,  the  test
+       is  applied to all subpatterns of the same name, and is true if any one
+       of them has matched.
+
+   Checking for pattern recursion
+
+       If the condition is the string (R), and there is no subpattern with the
+       name  R, the condition is true if a recursive call to the whole pattern
+       or any subpattern has been made. If digits or a name preceded by amper-
+       sand follow the letter R, for example:
+
+         (?(R3)...) or (?(R&name)...)
+
+       the condition is true if the most recent recursion is into a subpattern
+       whose number or name is given. This condition does not check the entire
+       recursion  stack.  If  the  name  used in a condition of this kind is a
+       duplicate, the test is applied to all subpatterns of the same name, and
+       is true if any one of them is the most recent recursion.
+
+       At  "top  level",  all  these recursion test conditions are false.  The
+       syntax for recursive patterns is described below.
+
+   Defining subpatterns for use by reference only
+
+       If the condition is the string (DEFINE), and  there  is  no  subpattern
+       with  the  name  DEFINE,  the  condition is always false. In this case,
+       there may be only one alternative  in  the  subpattern.  It  is  always
+       skipped  if  control  reaches  this  point  in the pattern; the idea of
+       DEFINE is that it can be used to define subroutines that can be  refer-
+       enced  from elsewhere. (The use of subroutines is described below.) For
+       example, a pattern to match an IPv4 address  such  as  "192.168.23.245"
+       could be written like this (ignore white space and line breaks):
+
+         (?(DEFINE) (? 2[0-4]\d | 25[0-5] | 1\d\d | [1-9]?\d) )
+         \b (?&byte) (\.(?&byte)){3} \b
+
+       The  first part of the pattern is a DEFINE group inside which a another
+       group named "byte" is defined. This matches an individual component  of
+       an  IPv4  address  (a number less than 256). When matching takes place,
+       this part of the pattern is skipped because DEFINE acts  like  a  false
+       condition.  The  rest of the pattern uses references to the named group
+       to match the four dot-separated components of an IPv4 address,  insist-
+       ing on a word boundary at each end.
+
+   Assertion conditions
+
+       If  the  condition  is  not  in any of the above formats, it must be an
+       assertion.  This may be a positive or negative lookahead or  lookbehind
+       assertion.  Consider  this  pattern,  again  containing non-significant
+       white space, and with the two alternatives on the second line:
+
+         (?(?=[^a-z]*[a-z])
+         \d{2}-[a-z]{3}-\d{2}  |  \d{2}-\d{2}-\d{2} )
+
+       The condition  is  a  positive  lookahead  assertion  that  matches  an
+       optional  sequence of non-letters followed by a letter. In other words,
+       it tests for the presence of at least one letter in the subject.  If  a
+       letter  is found, the subject is matched against the first alternative;
+       otherwise it is  matched  against  the  second.  This  pattern  matches
+       strings  in  one  of the two forms dd-aaa-dd or dd-dd-dd, where aaa are
+       letters and dd are digits.
+
+
+COMMENTS
+
+       There are two ways of including comments in patterns that are processed
+       by PCRE. In both cases, the start of the comment must not be in a char-
+       acter class, nor in the middle of any other sequence of related charac-
+       ters  such  as  (?: or a subpattern name or number. The characters that
+       make up a comment play no part in the pattern matching.
+
+       The sequence (?# marks the start of a comment that continues up to  the
+       next  closing parenthesis. Nested parentheses are not permitted. If the
+       PCRE_EXTENDED option is set, an unescaped # character also introduces a
+       comment,  which  in  this  case continues to immediately after the next
+       newline character or character sequence in the pattern.  Which  charac-
+       ters are interpreted as newlines is controlled by the options passed to
+       a compiling function or by a special sequence at the start of the  pat-
+       tern, as described in the section entitled "Newline conventions" above.
+       Note that the end of this type of comment is a literal newline sequence
+       in  the pattern; escape sequences that happen to represent a newline do
+       not count. For example, consider this  pattern  when  PCRE_EXTENDED  is
+       set, and the default newline convention is in force:
+
+         abc #comment \n still comment
+
+       On  encountering  the  # character, pcre_compile() skips along, looking
+       for a newline in the pattern. The sequence \n is still literal at  this
+       stage,  so  it does not terminate the comment. Only an actual character
+       with the code value 0x0a (the default newline) does so.
+
+
+RECURSIVE PATTERNS
+
+       Consider the problem of matching a string in parentheses, allowing  for
+       unlimited  nested  parentheses.  Without the use of recursion, the best
+       that can be done is to use a pattern that  matches  up  to  some  fixed
+       depth  of  nesting.  It  is not possible to handle an arbitrary nesting
+       depth.
+
+       For some time, Perl has provided a facility that allows regular expres-
+       sions  to recurse (amongst other things). It does this by interpolating
+       Perl code in the expression at run time, and the code can refer to  the
+       expression itself. A Perl pattern using code interpolation to solve the
+       parentheses problem can be created like this:
+
+         $re = qr{\( (?: (?>[^()]+) | (?p{$re}) )* \)}x;
+
+       The (?p{...}) item interpolates Perl code at run time, and in this case
+       refers recursively to the pattern in which it appears.
+
+       Obviously, PCRE cannot support the interpolation of Perl code. Instead,
+       it supports special syntax for recursion of  the  entire  pattern,  and
+       also  for  individual  subpattern  recursion. After its introduction in
+       PCRE and Python, this kind of  recursion  was  subsequently  introduced
+       into Perl at release 5.10.
+
+       A  special  item  that consists of (? followed by a number greater than
+       zero and a closing parenthesis is a recursive subroutine  call  of  the
+       subpattern  of  the  given  number, provided that it occurs inside that
+       subpattern. (If not, it is a non-recursive subroutine  call,  which  is
+       described  in  the  next  section.)  The special item (?R) or (?0) is a
+       recursive call of the entire regular expression.
+
+       This PCRE pattern solves the nested  parentheses  problem  (assume  the
+       PCRE_EXTENDED option is set so that white space is ignored):
+
+         \( ( [^()]++ | (?R) )* \)
+
+       First  it matches an opening parenthesis. Then it matches any number of
+       substrings which can either be a  sequence  of  non-parentheses,  or  a
+       recursive  match  of the pattern itself (that is, a correctly parenthe-
+       sized substring).  Finally there is a closing parenthesis. Note the use
+       of a possessive quantifier to avoid backtracking into sequences of non-
+       parentheses.
+
+       If this were part of a larger pattern, you would not  want  to  recurse
+       the entire pattern, so instead you could use this:
+
+         ( \( ( [^()]++ | (?1) )* \) )
+
+       We  have  put the pattern into parentheses, and caused the recursion to
+       refer to them instead of the whole pattern.
+
+       In a larger pattern,  keeping  track  of  parenthesis  numbers  can  be
+       tricky.  This is made easier by the use of relative references. Instead
+       of (?1) in the pattern above you can write (?-2) to refer to the second
+       most  recently  opened  parentheses  preceding  the recursion. In other
+       words, a negative number counts capturing  parentheses  leftwards  from
+       the point at which it is encountered.
+
+       It  is  also  possible  to refer to subsequently opened parentheses, by
+       writing references such as (?+2). However, these  cannot  be  recursive
+       because  the  reference  is  not inside the parentheses that are refer-
+       enced. They are always non-recursive subroutine calls, as described  in
+       the next section.
+
+       An  alternative  approach is to use named parentheses instead. The Perl
+       syntax for this is (?&name); PCRE's earlier syntax  (?P>name)  is  also
+       supported. We could rewrite the above example as follows:
+
+         (? \( ( [^()]++ | (?&pn) )* \) )
+
+       If  there  is more than one subpattern with the same name, the earliest
+       one is used.
+
+       This particular example pattern that we have been looking  at  contains
+       nested unlimited repeats, and so the use of a possessive quantifier for
+       matching strings of non-parentheses is important when applying the pat-
+       tern  to  strings  that do not match. For example, when this pattern is
+       applied to
+
+         (aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa()
+
+       it yields "no match" quickly. However, if a  possessive  quantifier  is
+       not  used, the match runs for a very long time indeed because there are
+       so many different ways the + and * repeats can carve  up  the  subject,
+       and all have to be tested before failure can be reported.
+
+       At  the  end  of a match, the values of capturing parentheses are those
+       from the outermost level. If you want to obtain intermediate values,  a
+       callout  function can be used (see below and the pcrecallout documenta-
+       tion). If the pattern above is matched against
+
+         (ab(cd)ef)
+
+       the value for the inner capturing parentheses  (numbered  2)  is  "ef",
+       which  is the last value taken on at the top level. If a capturing sub-
+       pattern is not matched at the top level, its final  captured  value  is
+       unset,  even  if  it was (temporarily) set at a deeper level during the
+       matching process.
+
+       If there are more than 15 capturing parentheses in a pattern, PCRE  has
+       to  obtain extra memory to store data during a recursion, which it does
+       by using pcre_malloc, freeing it via pcre_free afterwards. If no memory
+       can be obtained, the match fails with the PCRE_ERROR_NOMEMORY error.
+
+       Do  not  confuse  the (?R) item with the condition (R), which tests for
+       recursion.  Consider this pattern, which matches text in  angle  brack-
+       ets,  allowing for arbitrary nesting. Only digits are allowed in nested
+       brackets (that is, when recursing), whereas any characters are  permit-
+       ted at the outer level.
+
+         < (?: (?(R) \d++  | [^<>]*+) | (?R)) * >
+
+       In  this  pattern, (?(R) is the start of a conditional subpattern, with
+       two different alternatives for the recursive and  non-recursive  cases.
+       The (?R) item is the actual recursive call.
+
+   Differences in recursion processing between PCRE and Perl
+
+       Recursion  processing  in PCRE differs from Perl in two important ways.
+       In PCRE (like Python, but unlike Perl), a recursive subpattern call  is
+       always treated as an atomic group. That is, once it has matched some of
+       the subject string, it is never re-entered, even if it contains untried
+       alternatives  and  there  is a subsequent matching failure. This can be
+       illustrated by the following pattern, which purports to match a  palin-
+       dromic  string  that contains an odd number of characters (for example,
+       "a", "aba", "abcba", "abcdcba"):
+
+         ^(.|(.)(?1)\2)$
+
+       The idea is that it either matches a single character, or two identical
+       characters  surrounding  a sub-palindrome. In Perl, this pattern works;
+       in PCRE it does not if the pattern is  longer  than  three  characters.
+       Consider the subject string "abcba":
+
+       At  the  top level, the first character is matched, but as it is not at
+       the end of the string, the first alternative fails; the second alterna-
+       tive is taken and the recursion kicks in. The recursive call to subpat-
+       tern 1 successfully matches the next character ("b").  (Note  that  the
+       beginning and end of line tests are not part of the recursion).
+
+       Back  at  the top level, the next character ("c") is compared with what
+       subpattern 2 matched, which was "a". This fails. Because the  recursion
+       is  treated  as  an atomic group, there are now no backtracking points,
+       and so the entire match fails. (Perl is able, at  this  point,  to  re-
+       enter  the  recursion  and try the second alternative.) However, if the
+       pattern is written with the alternatives in the other order, things are
+       different:
+
+         ^((.)(?1)\2|.)$
+
+       This  time,  the recursing alternative is tried first, and continues to
+       recurse until it runs out of characters, at which point  the  recursion
+       fails.  But  this  time  we  do  have another alternative to try at the
+       higher level. That is the big difference:  in  the  previous  case  the
+       remaining alternative is at a deeper recursion level, which PCRE cannot
+       use.
+
+       To change the pattern so that it matches all palindromic  strings,  not
+       just  those  with an odd number of characters, it is tempting to change
+       the pattern to this:
+
+         ^((.)(?1)\2|.?)$
+
+       Again, this works in Perl, but not in PCRE, and for  the  same  reason.
+       When  a  deeper  recursion has matched a single character, it cannot be
+       entered again in order to match an empty string.  The  solution  is  to
+       separate  the two cases, and write out the odd and even cases as alter-
+       natives at the higher level:
+
+         ^(?:((.)(?1)\2|)|((.)(?3)\4|.))
+
+       If you want to match typical palindromic phrases, the  pattern  has  to
+       ignore all non-word characters, which can be done like this:
+
+         ^\W*+(?:((.)\W*+(?1)\W*+\2|)|((.)\W*+(?3)\W*+\4|\W*+.\W*+))\W*+$
+
+       If run with the PCRE_CASELESS option, this pattern matches phrases such
+       as "A man, a plan, a canal: Panama!" and it works well in both PCRE and
+       Perl.  Note the use of the possessive quantifier *+ to avoid backtrack-
+       ing into sequences of non-word characters. Without this, PCRE  takes  a
+       great  deal  longer  (ten  times or more) to match typical phrases, and
+       Perl takes so long that you think it has gone into a loop.
+
+       WARNING: The palindrome-matching patterns above work only if  the  sub-
+       ject  string  does not start with a palindrome that is shorter than the
+       entire string.  For example, although "abcba" is correctly matched,  if
+       the  subject  is "ababa", PCRE finds the palindrome "aba" at the start,
+       then fails at top level because the end of the string does not  follow.
+       Once  again, it cannot jump back into the recursion to try other alter-
+       natives, so the entire match fails.
+
+       The second way in which PCRE and Perl differ in  their  recursion  pro-
+       cessing  is in the handling of captured values. In Perl, when a subpat-
+       tern is called recursively or as a subpattern (see the  next  section),
+       it  has  no  access to any values that were captured outside the recur-
+       sion, whereas in PCRE these values can  be  referenced.  Consider  this
+       pattern:
+
+         ^(.)(\1|a(?2))
+
+       In  PCRE,  this  pattern matches "bab". The first capturing parentheses
+       match "b", then in the second group, when the back reference  \1  fails
+       to  match "b", the second alternative matches "a" and then recurses. In
+       the recursion, \1 does now match "b" and so the whole  match  succeeds.
+       In  Perl,  the pattern fails to match because inside the recursive call
+       \1 cannot access the externally set value.
+
+
+SUBPATTERNS AS SUBROUTINES
+
+       If the syntax for a recursive subpattern call (either by number  or  by
+       name)  is  used outside the parentheses to which it refers, it operates
+       like a subroutine in a programming language. The called subpattern  may
+       be  defined  before or after the reference. A numbered reference can be
+       absolute or relative, as in these examples:
+
+         (...(absolute)...)...(?2)...
+         (...(relative)...)...(?-1)...
+         (...(?+1)...(relative)...
+
+       An earlier example pointed out that the pattern
+
+         (sens|respons)e and \1ibility
+
+       matches "sense and sensibility" and "response and responsibility",  but
+       not "sense and responsibility". If instead the pattern
+
+         (sens|respons)e and (?1)ibility
+
+       is  used, it does match "sense and responsibility" as well as the other
+       two strings. Another example is  given  in  the  discussion  of  DEFINE
+       above.
+
+       All  subroutine  calls, whether recursive or not, are always treated as
+       atomic groups. That is, once a subroutine has matched some of the  sub-
+       ject string, it is never re-entered, even if it contains untried alter-
+       natives and there is  a  subsequent  matching  failure.  Any  capturing
+       parentheses  that  are  set  during the subroutine call revert to their
+       previous values afterwards.
+
+       Processing options such as case-independence are fixed when  a  subpat-
+       tern  is defined, so if it is used as a subroutine, such options cannot
+       be changed for different calls. For example, consider this pattern:
+
+         (abc)(?i:(?-1))
+
+       It matches "abcabc". It does not match "abcABC" because the  change  of
+       processing option does not affect the called subpattern.
+
+
+ONIGURUMA SUBROUTINE SYNTAX
+
+       For  compatibility with Oniguruma, the non-Perl syntax \g followed by a
+       name or a number enclosed either in angle brackets or single quotes, is
+       an  alternative  syntax  for  referencing a subpattern as a subroutine,
+       possibly recursively. Here are two of the examples used above,  rewrit-
+       ten using this syntax:
+
+         (? \( ( (?>[^()]+) | \g )* \) )
+         (sens|respons)e and \g'1'ibility
+
+       PCRE  supports  an extension to Oniguruma: if a number is preceded by a
+       plus or a minus sign it is taken as a relative reference. For example:
+
+         (abc)(?i:\g<-1>)
+
+       Note that \g{...} (Perl syntax) and \g<...> (Oniguruma syntax) are  not
+       synonymous.  The former is a back reference; the latter is a subroutine
+       call.
+
+
+CALLOUTS
+
+       Perl has a feature whereby using the sequence (?{...}) causes arbitrary
+       Perl  code to be obeyed in the middle of matching a regular expression.
+       This makes it possible, amongst other things, to extract different sub-
+       strings that match the same pair of parentheses when there is a repeti-
+       tion.
+
+       PCRE provides a similar feature, but of course it cannot obey arbitrary
+       Perl code. The feature is called "callout". The caller of PCRE provides
+       an external function by putting its entry point in the global  variable
+       pcre_callout  (8-bit  library) or pcre[16|32]_callout (16-bit or 32-bit
+       library).  By default, this variable contains NULL, which disables  all
+       calling out.
+
+       Within  a  regular  expression,  (?C) indicates the points at which the
+       external function is to be called. If you want  to  identify  different
+       callout  points, you can put a number less than 256 after the letter C.
+       The default value is zero.  For example, this pattern has  two  callout
+       points:
+
+         (?C1)abc(?C2)def
+
+       If  the PCRE_AUTO_CALLOUT flag is passed to a compiling function, call-
+       outs are automatically installed before each item in the pattern.  They
+       are  all  numbered  255. If there is a conditional group in the pattern
+       whose condition is an assertion, an additional callout is inserted just
+       before the condition. An explicit callout may also be set at this posi-
+       tion, as in this example:
+
+         (?(?C9)(?=a)abc|def)
+
+       Note that this applies only to assertion conditions, not to other types
+       of condition.
+
+       During  matching, when PCRE reaches a callout point, the external func-
+       tion is called. It is provided with the  number  of  the  callout,  the
+       position  in  the pattern, and, optionally, one item of data originally
+       supplied by the caller of the matching function. The  callout  function
+       may  cause  matching to proceed, to backtrack, or to fail altogether. A
+       complete description of the interface to the callout function is  given
+       in the pcrecallout documentation.
+
+
+BACKTRACKING CONTROL
+
+       Perl  5.10 introduced a number of "Special Backtracking Control Verbs",
+       which are still described in the Perl  documentation  as  "experimental
+       and  subject to change or removal in a future version of Perl". It goes
+       on to say: "Their usage in production code should  be  noted  to  avoid
+       problems  during upgrades." The same remarks apply to the PCRE features
+       described in this section.
+
+       The new verbs make use of what was previously invalid syntax: an  open-
+       ing parenthesis followed by an asterisk. They are generally of the form
+       (*VERB) or (*VERB:NAME). Some may take either form,  possibly  behaving
+       differently  depending  on  whether or not a name is present. A name is
+       any sequence of characters that does not include a closing parenthesis.
+       The maximum length of name is 255 in the 8-bit library and 65535 in the
+       16-bit and 32-bit libraries. If the name is  empty,  that  is,  if  the
+       closing  parenthesis immediately follows the colon, the effect is as if
+       the colon were not there.  Any number of these verbs  may  occur  in  a
+       pattern.
+
+       Since  these  verbs  are  specifically related to backtracking, most of
+       them can be used only when the pattern is to be matched  using  one  of
+       the  traditional  matching  functions, because these use a backtracking
+       algorithm. With the exception of (*FAIL), which behaves like a  failing
+       negative  assertion,  the  backtracking control verbs cause an error if
+       encountered by a DFA matching function.
+
+       The behaviour of these verbs in repeated  groups,  assertions,  and  in
+       subpatterns called as subroutines (whether or not recursively) is docu-
+       mented below.
+
+   Optimizations that affect backtracking verbs
+
+       PCRE contains some optimizations that are used to speed up matching  by
+       running some checks at the start of each match attempt. For example, it
+       may know the minimum length of matching subject, or that  a  particular
+       character must be present. When one of these optimizations bypasses the
+       running of a match,  any  included  backtracking  verbs  will  not,  of
+       course, be processed. You can suppress the start-of-match optimizations
+       by setting the PCRE_NO_START_OPTIMIZE  option  when  calling  pcre_com-
+       pile() or pcre_exec(), or by starting the pattern with (*NO_START_OPT).
+       There is more discussion of this option in the section entitled "Option
+       bits for pcre_exec()" in the pcreapi documentation.
+
+       Experiments  with  Perl  suggest that it too has similar optimizations,
+       sometimes leading to anomalous results.
+
+   Verbs that act immediately
+
+       The following verbs act as soon as they are encountered. They  may  not
+       be followed by a name.
+
+          (*ACCEPT)
+
+       This  verb causes the match to end successfully, skipping the remainder
+       of the pattern. However, when it is inside a subpattern that is  called
+       as  a  subroutine, only that subpattern is ended successfully. Matching
+       then continues at the outer level. If (*ACCEPT) in triggered in a posi-
+       tive  assertion,  the  assertion succeeds; in a negative assertion, the
+       assertion fails.
+
+       If (*ACCEPT) is inside capturing parentheses, the data so far  is  cap-
+       tured. For example:
+
+         A((?:A|B(*ACCEPT)|C)D)
+
+       This  matches  "AB", "AAD", or "ACD"; when it matches "AB", "B" is cap-
+       tured by the outer parentheses.
+
+         (*FAIL) or (*F)
+
+       This verb causes a matching failure, forcing backtracking to occur.  It
+       is  equivalent to (?!) but easier to read. The Perl documentation notes
+       that it is probably useful only when combined  with  (?{})  or  (??{}).
+       Those  are,  of course, Perl features that are not present in PCRE. The
+       nearest equivalent is the callout feature, as for example in this  pat-
+       tern:
+
+         a+(?C)(*FAIL)
+
+       A  match  with the string "aaaa" always fails, but the callout is taken
+       before each backtrack happens (in this example, 10 times).
+
+   Recording which path was taken
+
+       There is one verb whose main purpose  is  to  track  how  a  match  was
+       arrived  at,  though  it  also  has a secondary use in conjunction with
+       advancing the match starting point (see (*SKIP) below).
+
+         (*MARK:NAME) or (*:NAME)
+
+       A name is always  required  with  this  verb.  There  may  be  as  many
+       instances  of  (*MARK) as you like in a pattern, and their names do not
+       have to be unique.
+
+       When a match succeeds, the name of the  last-encountered  (*MARK:NAME),
+       (*PRUNE:NAME),  or  (*THEN:NAME) on the matching path is passed back to
+       the caller as  described  in  the  section  entitled  "Extra  data  for
+       pcre_exec()"  in  the  pcreapi  documentation.  Here  is  an example of
+       pcretest output, where the /K modifier requests the retrieval and  out-
+       putting of (*MARK) data:
+
+           re> /X(*MARK:A)Y|X(*MARK:B)Z/K
+         data> XY
+          0: XY
+         MK: A
+         XZ
+          0: XZ
+         MK: B
+
+       The (*MARK) name is tagged with "MK:" in this output, and in this exam-
+       ple it indicates which of the two alternatives matched. This is a  more
+       efficient  way of obtaining this information than putting each alterna-
+       tive in its own capturing parentheses.
+
+       If a verb with a name is encountered in a positive  assertion  that  is
+       true,  the  name  is recorded and passed back if it is the last-encoun-
+       tered. This does not happen for negative assertions or failing positive
+       assertions.
+
+       After  a  partial match or a failed match, the last encountered name in
+       the entire match process is returned. For example:
+
+           re> /X(*MARK:A)Y|X(*MARK:B)Z/K
+         data> XP
+         No match, mark = B
+
+       Note that in this unanchored example the  mark  is  retained  from  the
+       match attempt that started at the letter "X" in the subject. Subsequent
+       match attempts starting at "P" and then with an empty string do not get
+       as far as the (*MARK) item, but nevertheless do not reset it.
+
+       If  you  are  interested  in  (*MARK)  values after failed matches, you
+       should probably set the PCRE_NO_START_OPTIMIZE option  (see  above)  to
+       ensure that the match is always attempted.
+
+   Verbs that act after backtracking
+
+       The following verbs do nothing when they are encountered. Matching con-
+       tinues with what follows, but if there is no subsequent match,  causing
+       a  backtrack  to  the  verb, a failure is forced. That is, backtracking
+       cannot pass to the left of the verb. However, when one of  these  verbs
+       appears inside an atomic group or an assertion that is true, its effect
+       is confined to that group, because once the  group  has  been  matched,
+       there  is never any backtracking into it. In this situation, backtrack-
+       ing can "jump back" to the left of the entire atomic  group  or  asser-
+       tion.  (Remember  also,  as  stated  above, that this localization also
+       applies in subroutine calls.)
+
+       These verbs differ in exactly what kind of failure  occurs  when  back-
+       tracking  reaches  them.  The behaviour described below is what happens
+       when the verb is not in a subroutine or an assertion.  Subsequent  sec-
+       tions cover these special cases.
+
+         (*COMMIT)
+
+       This  verb, which may not be followed by a name, causes the whole match
+       to fail outright if there is a later matching failure that causes back-
+       tracking  to  reach  it.  Even if the pattern is unanchored, no further
+       attempts to find a match by advancing the starting point take place. If
+       (*COMMIT)  is  the  only backtracking verb that is encountered, once it
+       has been passed pcre_exec() is committed to finding a match at the cur-
+       rent starting point, or not at all. For example:
+
+         a+(*COMMIT)b
+
+       This  matches  "xxaab" but not "aacaab". It can be thought of as a kind
+       of dynamic anchor, or "I've started, so I must finish." The name of the
+       most  recently passed (*MARK) in the path is passed back when (*COMMIT)
+       forces a match failure.
+
+       If there is more than one backtracking verb in a pattern,  a  different
+       one  that  follows  (*COMMIT) may be triggered first, so merely passing
+       (*COMMIT) during a match does not always guarantee that a match must be
+       at this starting point.
+
+       Note  that  (*COMMIT)  at  the start of a pattern is not the same as an
+       anchor, unless PCRE's start-of-match optimizations are turned  off,  as
+       shown in this pcretest example:
+
+           re> /(*COMMIT)abc/
+         data> xyzabc
+          0: abc
+         xyzabc\Y
+         No match
+
+       PCRE  knows  that  any  match  must start with "a", so the optimization
+       skips along the subject to "a" before running the first match  attempt,
+       which  succeeds.  When the optimization is disabled by the \Y escape in
+       the second subject, the match starts at "x" and so the (*COMMIT) causes
+       it to fail without trying any other starting points.
+
+         (*PRUNE) or (*PRUNE:NAME)
+
+       This  verb causes the match to fail at the current starting position in
+       the subject if there is a later matching failure that causes backtrack-
+       ing  to  reach it. If the pattern is unanchored, the normal "bumpalong"
+       advance to the next starting character then happens.  Backtracking  can
+       occur  as  usual to the left of (*PRUNE), before it is reached, or when
+       matching to the right of (*PRUNE), but if there  is  no  match  to  the
+       right,  backtracking cannot cross (*PRUNE). In simple cases, the use of
+       (*PRUNE) is just an alternative to an atomic group or possessive  quan-
+       tifier, but there are some uses of (*PRUNE) that cannot be expressed in
+       any other way. In an anchored pattern (*PRUNE) has the same  effect  as
+       (*COMMIT).
+
+       The   behaviour   of   (*PRUNE:NAME)   is   the   not   the   same   as
+       (*MARK:NAME)(*PRUNE).  It is like (*MARK:NAME)  in  that  the  name  is
+       remembered  for  passing  back  to  the  caller.  However, (*SKIP:NAME)
+       searches only for names set with (*MARK).
+
+         (*SKIP)
+
+       This verb, when given without a name, is like (*PRUNE), except that  if
+       the  pattern  is unanchored, the "bumpalong" advance is not to the next
+       character, but to the position in the subject where (*SKIP) was encoun-
+       tered.  (*SKIP)  signifies that whatever text was matched leading up to
+       it cannot be part of a successful match. Consider:
+
+         a+(*SKIP)b
+
+       If the subject is "aaaac...",  after  the  first  match  attempt  fails
+       (starting  at  the  first  character in the string), the starting point
+       skips on to start the next attempt at "c". Note that a possessive quan-
+       tifer  does not have the same effect as this example; although it would
+       suppress backtracking  during  the  first  match  attempt,  the  second
+       attempt  would  start at the second character instead of skipping on to
+       "c".
+
+         (*SKIP:NAME)
+
+       When (*SKIP) has an associated name, its behaviour is modified. When it
+       is triggered, the previous path through the pattern is searched for the
+       most recent (*MARK) that has the  same  name.  If  one  is  found,  the
+       "bumpalong" advance is to the subject position that corresponds to that
+       (*MARK) instead of to where (*SKIP) was encountered. If no (*MARK) with
+       a matching name is found, the (*SKIP) is ignored.
+
+       Note  that (*SKIP:NAME) searches only for names set by (*MARK:NAME). It
+       ignores names that are set by (*PRUNE:NAME) or (*THEN:NAME).
+
+         (*THEN) or (*THEN:NAME)
+
+       This verb causes a skip to the next innermost  alternative  when  back-
+       tracking  reaches  it.  That  is,  it  cancels any further backtracking
+       within the current alternative. Its name  comes  from  the  observation
+       that it can be used for a pattern-based if-then-else block:
+
+         ( COND1 (*THEN) FOO | COND2 (*THEN) BAR | COND3 (*THEN) BAZ ) ...
+
+       If  the COND1 pattern matches, FOO is tried (and possibly further items
+       after the end of the group if FOO succeeds); on  failure,  the  matcher
+       skips  to  the second alternative and tries COND2, without backtracking
+       into COND1. If that succeeds and BAR fails, COND3 is tried.  If  subse-
+       quently  BAZ fails, there are no more alternatives, so there is a back-
+       track to whatever came before the  entire  group.  If  (*THEN)  is  not
+       inside an alternation, it acts like (*PRUNE).
+
+       The    behaviour   of   (*THEN:NAME)   is   the   not   the   same   as
+       (*MARK:NAME)(*THEN).  It is like  (*MARK:NAME)  in  that  the  name  is
+       remembered  for  passing  back  to  the  caller.  However, (*SKIP:NAME)
+       searches only for names set with (*MARK).
+
+       A subpattern that does not contain a | character is just a part of  the
+       enclosing  alternative;  it  is  not a nested alternation with only one
+       alternative. The effect of (*THEN) extends beyond such a subpattern  to
+       the  enclosing alternative. Consider this pattern, where A, B, etc. are
+       complex pattern fragments that do not contain any | characters at  this
+       level:
+
+         A (B(*THEN)C) | D
+
+       If  A and B are matched, but there is a failure in C, matching does not
+       backtrack into A; instead it moves to the next alternative, that is, D.
+       However,  if the subpattern containing (*THEN) is given an alternative,
+       it behaves differently:
+
+         A (B(*THEN)C | (*FAIL)) | D
+
+       The effect of (*THEN) is now confined to the inner subpattern. After  a
+       failure in C, matching moves to (*FAIL), which causes the whole subpat-
+       tern to fail because there are no more alternatives  to  try.  In  this
+       case, matching does now backtrack into A.
+
+       Note  that  a  conditional  subpattern  is not considered as having two
+       alternatives, because only one is ever used.  In  other  words,  the  |
+       character in a conditional subpattern has a different meaning. Ignoring
+       white space, consider:
+
+         ^.*? (?(?=a) a | b(*THEN)c )
+
+       If the subject is "ba", this pattern does not  match.  Because  .*?  is
+       ungreedy,  it  initially  matches  zero characters. The condition (?=a)
+       then fails, the character "b" is matched,  but  "c"  is  not.  At  this
+       point,  matching does not backtrack to .*? as might perhaps be expected
+       from the presence of the | character.  The  conditional  subpattern  is
+       part of the single alternative that comprises the whole pattern, and so
+       the match fails. (If there was a backtrack into  .*?,  allowing  it  to
+       match "b", the match would succeed.)
+
+       The  verbs just described provide four different "strengths" of control
+       when subsequent matching fails. (*THEN) is the weakest, carrying on the
+       match  at  the next alternative. (*PRUNE) comes next, failing the match
+       at the current starting position, but allowing an advance to  the  next
+       character  (for an unanchored pattern). (*SKIP) is similar, except that
+       the advance may be more than one character. (*COMMIT) is the strongest,
+       causing the entire match to fail.
+
+   More than one backtracking verb
+
+       If  more  than  one  backtracking verb is present in a pattern, the one
+       that is backtracked onto first acts. For example,  consider  this  pat-
+       tern, where A, B, etc. are complex pattern fragments:
+
+         (A(*COMMIT)B(*THEN)C|ABD)
+
+       If  A matches but B fails, the backtrack to (*COMMIT) causes the entire
+       match to fail. However, if A and B match, but C fails, the backtrack to
+       (*THEN)  causes  the next alternative (ABD) to be tried. This behaviour
+       is consistent, but is not always the same as Perl's. It means  that  if
+       two  or  more backtracking verbs appear in succession, all the the last
+       of them has no effect. Consider this example:
+
+         ...(*COMMIT)(*PRUNE)...
+
+       If there is a matching failure to the right, backtracking onto (*PRUNE)
+       cases it to be triggered, and its action is taken. There can never be a
+       backtrack onto (*COMMIT).
+
+   Backtracking verbs in repeated groups
+
+       PCRE differs from  Perl  in  its  handling  of  backtracking  verbs  in
+       repeated groups. For example, consider:
+
+         /(a(*COMMIT)b)+ac/
+
+       If  the  subject  is  "abac",  Perl matches, but PCRE fails because the
+       (*COMMIT) in the second repeat of the group acts.
+
+   Backtracking verbs in assertions
+
+       (*FAIL) in an assertion has its normal effect: it forces  an  immediate
+       backtrack.
+
+       (*ACCEPT) in a positive assertion causes the assertion to succeed with-
+       out any further processing. In a negative assertion,  (*ACCEPT)  causes
+       the assertion to fail without any further processing.
+
+       The  other  backtracking verbs are not treated specially if they appear
+       in a positive assertion. In  particular,  (*THEN)  skips  to  the  next
+       alternative  in  the  innermost  enclosing group that has alternations,
+       whether or not this is within the assertion.
+
+       Negative assertions are, however, different, in order  to  ensure  that
+       changing  a  positive  assertion  into a negative assertion changes its
+       result. Backtracking into (*COMMIT), (*SKIP), or (*PRUNE) causes a neg-
+       ative assertion to be true, without considering any further alternative
+       branches in the assertion.  Backtracking into (*THEN) causes it to skip
+       to  the next enclosing alternative within the assertion (the normal be-
+       haviour), but if the assertion  does  not  have  such  an  alternative,
+       (*THEN) behaves like (*PRUNE).
+
+   Backtracking verbs in subroutines
+
+       These  behaviours  occur whether or not the subpattern is called recur-
+       sively.  Perl's treatment of subroutines is different in some cases.
+
+       (*FAIL) in a subpattern called as a subroutine has its  normal  effect:
+       it forces an immediate backtrack.
+
+       (*ACCEPT)  in a subpattern called as a subroutine causes the subroutine
+       match to succeed without any further processing. Matching then  contin-
+       ues after the subroutine call.
+
+       (*COMMIT), (*SKIP), and (*PRUNE) in a subpattern called as a subroutine
+       cause the subroutine match to fail.
+
+       (*THEN) skips to the next alternative in the innermost enclosing  group
+       within  the subpattern that has alternatives. If there is no such group
+       within the subpattern, (*THEN) causes the subroutine match to fail.
+
+
+SEE ALSO
+
+       pcreapi(3), pcrecallout(3),  pcrematching(3),  pcresyntax(3),  pcre(3),
+       pcre16(3), pcre32(3).
+
+
+AUTHOR
+
+       Philip Hazel
+       University Computing Service
+       Cambridge CB2 3QH, England.
+
+
+REVISION
+
+       Last updated: 26 April 2013
+       Copyright (c) 1997-2013 University of Cambridge.
+------------------------------------------------------------------------------
+
+
+PCRESYNTAX(3)              Library Functions Manual              PCRESYNTAX(3)
+
+
+
+NAME
+       PCRE - Perl-compatible regular expressions
+
+PCRE REGULAR EXPRESSION SYNTAX SUMMARY
+
+       The  full syntax and semantics of the regular expressions that are sup-
+       ported by PCRE are described in  the  pcrepattern  documentation.  This
+       document contains a quick-reference summary of the syntax.
+
+
+QUOTING
+
+         \x         where x is non-alphanumeric is a literal x
+         \Q...\E    treat enclosed characters as literal
+
+
+CHARACTERS
+
+         \a         alarm, that is, the BEL character (hex 07)
+         \cx        "control-x", where x is any ASCII character
+         \e         escape (hex 1B)
+         \f         form feed (hex 0C)
+         \n         newline (hex 0A)
+         \r         carriage return (hex 0D)
+         \t         tab (hex 09)
+         \ddd       character with octal code ddd, or backreference
+         \xhh       character with hex code hh
+         \x{hhh..}  character with hex code hhh..
+
+
+CHARACTER TYPES
+
+         .          any character except newline;
+                      in dotall mode, any character whatsoever
+         \C         one data unit, even in UTF mode (best avoided)
+         \d         a decimal digit
+         \D         a character that is not a decimal digit
+         \h         a horizontal white space character
+         \H         a character that is not a horizontal white space character
+         \N         a character that is not a newline
+         \p{xx}     a character with the xx property
+         \P{xx}     a character without the xx property
+         \R         a newline sequence
+         \s         a white space character
+         \S         a character that is not a white space character
+         \v         a vertical white space character
+         \V         a character that is not a vertical white space character
+         \w         a "word" character
+         \W         a "non-word" character
+         \X         a Unicode extended grapheme cluster
+
+       In  PCRE,  by  default, \d, \D, \s, \S, \w, and \W recognize only ASCII
+       characters, even in a UTF mode. However, this can be changed by setting
+       the PCRE_UCP option.
+
+
+GENERAL CATEGORY PROPERTIES FOR \p and \P
+
+         C          Other
+         Cc         Control
+         Cf         Format
+         Cn         Unassigned
+         Co         Private use
+         Cs         Surrogate
+
+         L          Letter
+         Ll         Lower case letter
+         Lm         Modifier letter
+         Lo         Other letter
+         Lt         Title case letter
+         Lu         Upper case letter
+         L&         Ll, Lu, or Lt
+
+         M          Mark
+         Mc         Spacing mark
+         Me         Enclosing mark
+         Mn         Non-spacing mark
+
+         N          Number
+         Nd         Decimal number
+         Nl         Letter number
+         No         Other number
+
+         P          Punctuation
+         Pc         Connector punctuation
+         Pd         Dash punctuation
+         Pe         Close punctuation
+         Pf         Final punctuation
+         Pi         Initial punctuation
+         Po         Other punctuation
+         Ps         Open punctuation
+
+         S          Symbol
+         Sc         Currency symbol
+         Sk         Modifier symbol
+         Sm         Mathematical symbol
+         So         Other symbol
+
+         Z          Separator
+         Zl         Line separator
+         Zp         Paragraph separator
+         Zs         Space separator
+
+
+PCRE SPECIAL CATEGORY PROPERTIES FOR \p and \P
+
+         Xan        Alphanumeric: union of properties L and N
+         Xps        POSIX space: property Z or tab, NL, VT, FF, CR
+         Xsp        Perl space: property Z or tab, NL, FF, CR
+         Xuc        Univerally-named character: one that can be
+                      represented by a Universal Character Name
+         Xwd        Perl word: property Xan or underscore
+
+
+SCRIPT NAMES FOR \p AND \P
+
+       Arabic,  Armenian,  Avestan, Balinese, Bamum, Batak, Bengali, Bopomofo,
+       Brahmi, Braille, Buginese, Buhid, Canadian_Aboriginal, Carian,  Chakma,
+       Cham,  Cherokee, Common, Coptic, Cuneiform, Cypriot, Cyrillic, Deseret,
+       Devanagari,  Egyptian_Hieroglyphs,  Ethiopic,   Georgian,   Glagolitic,
+       Gothic,  Greek, Gujarati, Gurmukhi, Han, Hangul, Hanunoo, Hebrew, Hira-
+       gana,  Imperial_Aramaic,  Inherited,  Inscriptional_Pahlavi,   Inscrip-
+       tional_Parthian,   Javanese,   Kaithi,   Kannada,  Katakana,  Kayah_Li,
+       Kharoshthi, Khmer, Lao, Latin, Lepcha, Limbu, Linear_B,  Lisu,  Lycian,
+       Lydian,    Malayalam,    Mandaic,    Meetei_Mayek,    Meroitic_Cursive,
+       Meroitic_Hieroglyphs,  Miao,  Mongolian,  Myanmar,  New_Tai_Lue,   Nko,
+       Ogham,    Old_Italic,   Old_Persian,   Old_South_Arabian,   Old_Turkic,
+       Ol_Chiki, Oriya, Osmanya, Phags_Pa, Phoenician, Rejang, Runic,  Samari-
+       tan,  Saurashtra,  Sharada,  Shavian, Sinhala, Sora_Sompeng, Sundanese,
+       Syloti_Nagri, Syriac, Tagalog, Tagbanwa,  Tai_Le,  Tai_Tham,  Tai_Viet,
+       Takri,  Tamil,  Telugu, Thaana, Thai, Tibetan, Tifinagh, Ugaritic, Vai,
+       Yi.
+
+
+CHARACTER CLASSES
+
+         [...]       positive character class
+         [^...]      negative character class
+         [x-y]       range (can be used for hex characters)
+         [[:xxx:]]   positive POSIX named set
+         [[:^xxx:]]  negative POSIX named set
+
+         alnum       alphanumeric
+         alpha       alphabetic
+         ascii       0-127
+         blank       space or tab
+         cntrl       control character
+         digit       decimal digit
+         graph       printing, excluding space
+         lower       lower case letter
+         print       printing, including space
+         punct       printing, excluding alphanumeric
+         space       white space
+         upper       upper case letter
+         word        same as \w
+         xdigit      hexadecimal digit
+
+       In PCRE, POSIX character set names recognize only ASCII  characters  by
+       default,  but  some  of them use Unicode properties if PCRE_UCP is set.
+       You can use \Q...\E inside a character class.
+
+
+QUANTIFIERS
+
+         ?           0 or 1, greedy
+         ?+          0 or 1, possessive
+         ??          0 or 1, lazy
+         *           0 or more, greedy
+         *+          0 or more, possessive
+         *?          0 or more, lazy
+         +           1 or more, greedy
+         ++          1 or more, possessive
+         +?          1 or more, lazy
+         {n}         exactly n
+         {n,m}       at least n, no more than m, greedy
+         {n,m}+      at least n, no more than m, possessive
+         {n,m}?      at least n, no more than m, lazy
+         {n,}        n or more, greedy
+         {n,}+       n or more, possessive
+         {n,}?       n or more, lazy
+
+
+ANCHORS AND SIMPLE ASSERTIONS
+
+         \b          word boundary
+         \B          not a word boundary
+         ^           start of subject
+                      also after internal newline in multiline mode
+         \A          start of subject
+         $           end of subject
+                      also before newline at end of subject
+                      also before internal newline in multiline mode
+         \Z          end of subject
+                      also before newline at end of subject
+         \z          end of subject
+         \G          first matching position in subject
+
+
+MATCH POINT RESET
+
+         \K          reset start of match
+
+
+ALTERNATION
+
+         expr|expr|expr...
+
+
+CAPTURING
+
+         (...)           capturing group
+         (?...)    named capturing group (Perl)
+         (?'name'...)    named capturing group (Perl)
+         (?P...)   named capturing group (Python)
+         (?:...)         non-capturing group
+         (?|...)         non-capturing group; reset group numbers for
+                          capturing groups in each alternative
+
+
+ATOMIC GROUPS
+
+         (?>...)         atomic, non-capturing group
+
+
+COMMENT
+
+         (?#....)        comment (not nestable)
+
+
+OPTION SETTING
+
+         (?i)            caseless
+         (?J)            allow duplicate names
+         (?m)            multiline
+         (?s)            single line (dotall)
+         (?U)            default ungreedy (lazy)
+         (?x)            extended (ignore white space)
+         (?-...)         unset option(s)
+
+       The following are recognized only at the start of a  pattern  or  after
+       one of the newline-setting options with similar syntax:
+
+         (*LIMIT_MATCH=d) set the match limit to d (decimal number)
+         (*LIMIT_RECURSION=d) set the recursion limit to d (decimal number)
+         (*NO_START_OPT) no start-match optimization (PCRE_NO_START_OPTIMIZE)
+         (*UTF8)         set UTF-8 mode: 8-bit library (PCRE_UTF8)
+         (*UTF16)        set UTF-16 mode: 16-bit library (PCRE_UTF16)
+         (*UTF32)        set UTF-32 mode: 32-bit library (PCRE_UTF32)
+         (*UTF)          set appropriate UTF mode for the library in use
+         (*UCP)          set PCRE_UCP (use Unicode properties for \d etc)
+
+
+LOOKAHEAD AND LOOKBEHIND ASSERTIONS
+
+         (?=...)         positive look ahead
+         (?!...)         negative look ahead
+         (?<=...)        positive look behind
+         (?        reference by name (Perl)
+         \k'name'        reference by name (Perl)
+         \g{name}        reference by name (Perl)
+         \k{name}        reference by name (.NET)
+         (?P=name)       reference by name (Python)
+
+
+SUBROUTINE REFERENCES (POSSIBLY RECURSIVE)
+
+         (?R)            recurse whole pattern
+         (?n)            call subpattern by absolute number
+         (?+n)           call subpattern by relative number
+         (?-n)           call subpattern by relative number
+         (?&name)        call subpattern by name (Perl)
+         (?P>name)       call subpattern by name (Python)
+         \g        call subpattern by name (Oniguruma)
+         \g'name'        call subpattern by name (Oniguruma)
+         \g           call subpattern by absolute number (Oniguruma)
+         \g'n'           call subpattern by absolute number (Oniguruma)
+         \g<+n>          call subpattern by relative number (PCRE extension)
+         \g'+n'          call subpattern by relative number (PCRE extension)
+         \g<-n>          call subpattern by relative number (PCRE extension)
+         \g'-n'          call subpattern by relative number (PCRE extension)
+
+
+CONDITIONAL PATTERNS
+
+         (?(condition)yes-pattern)
+         (?(condition)yes-pattern|no-pattern)
+
+         (?(n)...        absolute reference condition
+         (?(+n)...       relative reference condition
+         (?(-n)...       relative reference condition
+         (?()...   named reference condition (Perl)
+         (?('name')...   named reference condition (Perl)
+         (?(name)...     named reference condition (PCRE)
+         (?(R)...        overall recursion condition
+         (?(Rn)...       specific group recursion condition
+         (?(R&name)...   specific recursion condition
+         (?(DEFINE)...   define subpattern for reference
+         (?(assert)...   assertion condition
+
+
+BACKTRACKING CONTROL
+
+       The following act immediately they are reached:
+
+         (*ACCEPT)       force successful match
+         (*FAIL)         force backtrack; synonym (*F)
+         (*MARK:NAME)    set name to be passed back; synonym (*:NAME)
+
+       The  following  act only when a subsequent match failure causes a back-
+       track to reach them. They all force a match failure, but they differ in
+       what happens afterwards. Those that advance the start-of-match point do
+       so only if the pattern is not anchored.
+
+         (*COMMIT)       overall failure, no advance of starting point
+         (*PRUNE)        advance to next starting character
+         (*PRUNE:NAME)   equivalent to (*MARK:NAME)(*PRUNE)
+         (*SKIP)         advance to current matching position
+         (*SKIP:NAME)    advance to position corresponding to an earlier
+                         (*MARK:NAME); if not found, the (*SKIP) is ignored
+         (*THEN)         local failure, backtrack to next alternation
+         (*THEN:NAME)    equivalent to (*MARK:NAME)(*THEN)
+
+
+NEWLINE CONVENTIONS
+
+       These are recognized only at the very start of the pattern or  after  a
+       (*BSR_...), (*UTF8), (*UTF16), (*UTF32) or (*UCP) option.
+
+         (*CR)           carriage return only
+         (*LF)           linefeed only
+         (*CRLF)         carriage return followed by linefeed
+         (*ANYCRLF)      all three of the above
+         (*ANY)          any Unicode newline sequence
+
+
+WHAT \R MATCHES
+
+       These  are  recognized only at the very start of the pattern or after a
+       (*...) option that sets the newline convention or a UTF or UCP mode.
+
+         (*BSR_ANYCRLF)  CR, LF, or CRLF
+         (*BSR_UNICODE)  any Unicode newline sequence
+
+
+CALLOUTS
+
+         (?C)      callout
+         (?Cn)     callout with data n
+
+
+SEE ALSO
+
+       pcrepattern(3), pcreapi(3), pcrecallout(3), pcrematching(3), pcre(3).
+
+
+AUTHOR
+
+       Philip Hazel
+       University Computing Service
+       Cambridge CB2 3QH, England.
+
+
+REVISION
+
+       Last updated: 26 April 2013
+       Copyright (c) 1997-2013 University of Cambridge.
+------------------------------------------------------------------------------
+
+
+PCREUNICODE(3)             Library Functions Manual             PCREUNICODE(3)
+
+
+
+NAME
+       PCRE - Perl-compatible regular expressions
+
+UTF-8, UTF-16, UTF-32, AND UNICODE PROPERTY SUPPORT
+
+       As well as UTF-8 support, PCRE also supports UTF-16 (from release 8.30)
+       and UTF-32 (from release 8.32), by means of two  additional  libraries.
+       They can be built as well as, or instead of, the 8-bit library.
+
+
+UTF-8 SUPPORT
+
+       In  order  process  UTF-8  strings, you must build PCRE's 8-bit library
+       with UTF support, and, in addition, you must call  pcre_compile()  with
+       the  PCRE_UTF8 option flag, or the pattern must start with the sequence
+       (*UTF8) or (*UTF). When either of these is the case, both  the  pattern
+       and  any  subject  strings  that  are matched against it are treated as
+       UTF-8 strings instead of strings of individual 1-byte characters.
+
+
+UTF-16 AND UTF-32 SUPPORT
+
+       In order process UTF-16 or UTF-32 strings, you must build PCRE's 16-bit
+       or  32-bit  library  with  UTF support, and, in addition, you must call
+       pcre16_compile() or pcre32_compile() with the PCRE_UTF16 or  PCRE_UTF32
+       option flag, as appropriate. Alternatively, the pattern must start with
+       the sequence (*UTF16), (*UTF32), as appropriate, or (*UTF),  which  can
+       be used with either library. When UTF mode is set, both the pattern and
+       any subject strings that are matched against it are treated  as  UTF-16
+       or  UTF-32  strings  instead  of strings of individual 16-bit or 32-bit
+       characters.
+
+
+UTF SUPPORT OVERHEAD
+
+       If you compile PCRE with UTF support, but do not use it  at  run  time,
+       the  library will be a bit bigger, but the additional run time overhead
+       is limited to  testing  the  PCRE_UTF[8|16|32]  flag  occasionally,  so
+       should not be very big.
+
+
+UNICODE PROPERTY SUPPORT
+
+       If PCRE is built with Unicode character property support (which implies
+       UTF support), the escape sequences \p{..}, \P{..}, and \X can be  used.
+       The  available properties that can be tested are limited to the general
+       category properties such as Lu for an upper case letter  or  Nd  for  a
+       decimal number, the Unicode script names such as Arabic or Han, and the
+       derived properties Any and L&. Full lists is given in  the  pcrepattern
+       and  pcresyntax  documentation. Only the short names for properties are
+       supported. For example, \p{L}  matches  a  letter.  Its  Perl  synonym,
+       \p{Letter},  is  not  supported.  Furthermore, in Perl, many properties
+       may optionally be prefixed by "Is", for compatibility  with  Perl  5.6.
+       PCRE does not support this.
+
+   Validity of UTF-8 strings
+
+       When  you  set  the PCRE_UTF8 flag, the byte strings passed as patterns
+       and subjects are (by default) checked for validity on entry to the rel-
+       evant functions. The entire string is checked before any other process-
+       ing takes place. From release 7.3 of PCRE, the check is  according  the
+       rules of RFC 3629, which are themselves derived from the Unicode speci-
+       fication. Earlier releases of PCRE followed  the  rules  of  RFC  2279,
+       which  allows  the  full  range of 31-bit values (0 to 0x7FFFFFFF). The
+       current check allows only values in the range U+0 to U+10FFFF,  exclud-
+       ing  the  surrogate area. (From release 8.33 the so-called "non-charac-
+       ter" code points are no longer excluded because Unicode corrigendum  #9
+       makes it clear that they should not be.)
+
+       Characters  in  the "Surrogate Area" of Unicode are reserved for use by
+       UTF-16, where they are used in pairs to encode codepoints  with  values
+       greater  than  0xFFFF. The code points that are encoded by UTF-16 pairs
+       are available independently in the  UTF-8  and  UTF-32  encodings.  (In
+       other  words,  the  whole  surrogate  thing is a fudge for UTF-16 which
+       unfortunately messes up UTF-8 and UTF-32.)
+
+       If an invalid UTF-8 string is passed to PCRE, an error return is given.
+       At  compile  time, the only additional information is the offset to the
+       first byte of the failing character. The run-time functions pcre_exec()
+       and  pcre_dfa_exec() also pass back this information, as well as a more
+       detailed reason code if the caller has provided memory in which  to  do
+       this.
+
+       In  some  situations, you may already know that your strings are valid,
+       and therefore want to skip these checks in  order  to  improve  perfor-
+       mance,  for  example in the case of a long subject string that is being
+       scanned repeatedly.  If you set the PCRE_NO_UTF8_CHECK flag at  compile
+       time  or  at  run  time, PCRE assumes that the pattern or subject it is
+       given (respectively) contains only valid UTF-8 codes. In this case,  it
+       does not diagnose an invalid UTF-8 string.
+
+       Note  that  passing  PCRE_NO_UTF8_CHECK to pcre_compile() just disables
+       the check for the pattern; it does not also apply to  subject  strings.
+       If  you  want  to  disable the check for a subject string you must pass
+       this option to pcre_exec() or pcre_dfa_exec().
+
+       If you pass an invalid UTF-8 string when PCRE_NO_UTF8_CHECK is set, the
+       result is undefined and your program may crash.
+
+   Validity of UTF-16 strings
+
+       When you set the PCRE_UTF16 flag, the strings of 16-bit data units that
+       are passed as patterns and subjects are (by default) checked for valid-
+       ity  on entry to the relevant functions. Values other than those in the
+       surrogate range U+D800 to U+DFFF are independent code points. Values in
+       the surrogate range must be used in pairs in the correct manner.
+
+       If  an  invalid  UTF-16  string  is  passed to PCRE, an error return is
+       given. At compile time, the only additional information is  the  offset
+       to the first data unit of the failing character. The run-time functions
+       pcre16_exec() and pcre16_dfa_exec() also pass back this information, as
+       well  as  a more detailed reason code if the caller has provided memory
+       in which to do this.
+
+       In some situations, you may already know that your strings  are  valid,
+       and  therefore  want  to  skip these checks in order to improve perfor-
+       mance. If you set the PCRE_NO_UTF16_CHECK flag at compile  time  or  at
+       run time, PCRE assumes that the pattern or subject it is given (respec-
+       tively) contains only valid UTF-16 sequences. In this case, it does not
+       diagnose  an  invalid  UTF-16 string.  However, if an invalid string is
+       passed, the result is undefined.
+
+   Validity of UTF-32 strings
+
+       When you set the PCRE_UTF32 flag, the strings of 32-bit data units that
+       are passed as patterns and subjects are (by default) checked for valid-
+       ity on entry to the relevant functions.  This check allows only  values
+       in  the  range  U+0 to U+10FFFF, excluding the surrogate area U+D800 to
+       U+DFFF.
+
+       If an invalid UTF-32 string is passed  to  PCRE,  an  error  return  is
+       given.  At  compile time, the only additional information is the offset
+       to the first data unit of the failing character. The run-time functions
+       pcre32_exec() and pcre32_dfa_exec() also pass back this information, as
+       well as a more detailed reason code if the caller has  provided  memory
+       in which to do this.
+
+       In  some  situations, you may already know that your strings are valid,
+       and therefore want to skip these checks in  order  to  improve  perfor-
+       mance.  If  you  set the PCRE_NO_UTF32_CHECK flag at compile time or at
+       run time, PCRE assumes that the pattern or subject it is given (respec-
+       tively) contains only valid UTF-32 sequences. In this case, it does not
+       diagnose an invalid UTF-32 string.  However, if an  invalid  string  is
+       passed, the result is undefined.
+
+   General comments about UTF modes
+
+       1.  Codepoints  less  than  256  can be specified in patterns by either
+       braced or unbraced hexadecimal escape sequences (for example, \x{b3} or
+       \xb3). Larger values have to use braced sequences.
+
+       2.  Octal  numbers  up  to  \777 are recognized, and in UTF-8 mode they
+       match two-byte characters for values greater than \177.
+
+       3. Repeat quantifiers apply to complete UTF characters, not to individ-
+       ual data units, for example: \x{100}{3}.
+
+       4.  The dot metacharacter matches one UTF character instead of a single
+       data unit.
+
+       5. The escape sequence \C can be used to match a single byte  in  UTF-8
+       mode,  or  a single 16-bit data unit in UTF-16 mode, or a single 32-bit
+       data unit in UTF-32 mode, but its use can lead to some strange  effects
+       because  it  breaks up multi-unit characters (see the description of \C
+       in the pcrepattern documentation). The use of \C is  not  supported  in
+       the  alternative  matching  function  pcre[16|32]_dfa_exec(), nor is it
+       supported in UTF mode by the JIT optimization of pcre[16|32]_exec(). If
+       JIT  optimization  is  requested for a UTF pattern that contains \C, it
+       will not succeed, and so the matching will be carried out by the normal
+       interpretive function.
+
+       6.  The  character escapes \b, \B, \d, \D, \s, \S, \w, and \W correctly
+       test characters of any code value, but, by default, the characters that
+       PCRE  recognizes  as digits, spaces, or word characters remain the same
+       set as in non-UTF mode, all with values less  than  256.  This  remains
+       true  even  when  PCRE  is  built  to include Unicode property support,
+       because to do otherwise would slow down PCRE in many common cases. Note
+       in  particular that this applies to \b and \B, because they are defined
+       in terms of \w and \W. If you really want to test for a wider sense of,
+       say,  "digit",  you  can  use  explicit  Unicode property tests such as
+       \p{Nd}. Alternatively, if you set the PCRE_UCP option, the way that the
+       character  escapes  work is changed so that Unicode properties are used
+       to determine which characters match. There are more details in the sec-
+       tion on generic character types in the pcrepattern documentation.
+
+       7.  Similarly,  characters that match the POSIX named character classes
+       are all low-valued characters, unless the PCRE_UCP option is set.
+
+       8. However, the horizontal and vertical white  space  matching  escapes
+       (\h,  \H,  \v, and \V) do match all the appropriate Unicode characters,
+       whether or not PCRE_UCP is set.
+
+       9. Case-insensitive matching applies only to  characters  whose  values
+       are  less than 128, unless PCRE is built with Unicode property support.
+       A few Unicode characters such as Greek sigma have more than  two  code-
+       points that are case-equivalent. Up to and including PCRE release 8.31,
+       only one-to-one case mappings were supported, but later releases  (with
+       Unicode  property  support) do treat as case-equivalent all versions of
+       characters such as Greek sigma.
+
+
+AUTHOR
+
+       Philip Hazel
+       University Computing Service
+       Cambridge CB2 3QH, England.
+
+
+REVISION
+
+       Last updated: 27 February 2013
+       Copyright (c) 1997-2013 University of Cambridge.
+------------------------------------------------------------------------------
+
+
+PCREJIT(3)                 Library Functions Manual                 PCREJIT(3)
+
+
+
+NAME
+       PCRE - Perl-compatible regular expressions
+
+PCRE JUST-IN-TIME COMPILER SUPPORT
+
+       Just-in-time  compiling  is a heavyweight optimization that can greatly
+       speed up pattern matching. However, it comes at the cost of extra  pro-
+       cessing before the match is performed. Therefore, it is of most benefit
+       when the same pattern is going to be matched many times. This does  not
+       necessarily  mean  many calls of a matching function; if the pattern is
+       not anchored, matching attempts may take place many  times  at  various
+       positions  in  the  subject, even for a single call.  Therefore, if the
+       subject string is very long, it may still pay to use  JIT  for  one-off
+       matches.
+
+       JIT  support  applies  only to the traditional Perl-compatible matching
+       function.  It does not apply when the DFA matching  function  is  being
+       used. The code for this support was written by Zoltan Herczeg.
+
+
+8-BIT, 16-BIT AND 32-BIT SUPPORT
+
+       JIT  support  is available for all of the 8-bit, 16-bit and 32-bit PCRE
+       libraries. To keep this documentation simple, only the 8-bit  interface
+       is described in what follows. If you are using the 16-bit library, sub-
+       stitute the  16-bit  functions  and  16-bit  structures  (for  example,
+       pcre16_jit_stack  instead  of  pcre_jit_stack).  If  you  are using the
+       32-bit library, substitute the 32-bit functions and  32-bit  structures
+       (for example, pcre32_jit_stack instead of pcre_jit_stack).
+
+
+AVAILABILITY OF JIT SUPPORT
+
+       JIT  support  is  an  optional  feature of PCRE. The "configure" option
+       --enable-jit (or equivalent CMake option) must  be  set  when  PCRE  is
+       built  if  you want to use JIT. The support is limited to the following
+       hardware platforms:
+
+         ARM v5, v7, and Thumb2
+         Intel x86 32-bit and 64-bit
+         MIPS 32-bit
+         Power PC 32-bit and 64-bit
+         SPARC 32-bit (experimental)
+
+       If --enable-jit is set on an unsupported platform, compilation fails.
+
+       A program that is linked with PCRE 8.20 or later can tell if  JIT  sup-
+       port  is  available  by  calling pcre_config() with the PCRE_CONFIG_JIT
+       option. The result is 1 when JIT is available, and  0  otherwise.  How-
+       ever, a simple program does not need to check this in order to use JIT.
+       The normal API is implemented in a way that falls back to the interpre-
+       tive code if JIT is not available. For programs that need the best pos-
+       sible performance, there is also a "fast path"  API  that  is  JIT-spe-
+       cific.
+
+       If  your program may sometimes be linked with versions of PCRE that are
+       older than 8.20, but you want to use JIT when it is available, you  can
+       test the values of PCRE_MAJOR and PCRE_MINOR, or the existence of a JIT
+       macro such as PCRE_CONFIG_JIT, for compile-time control of your code.
+
+
+SIMPLE USE OF JIT
+
+       You have to do two things to make use of the JIT support  in  the  sim-
+       plest way:
+
+         (1) Call pcre_study() with the PCRE_STUDY_JIT_COMPILE option for
+             each compiled pattern, and pass the resulting pcre_extra block to
+             pcre_exec().
+
+         (2) Use pcre_free_study() to free the pcre_extra block when it is
+             no  longer  needed,  instead  of  just  freeing it yourself. This
+       ensures that
+             any JIT data is also freed.
+
+       For a program that may be linked with pre-8.20 versions  of  PCRE,  you
+       can insert
+
+         #ifndef PCRE_STUDY_JIT_COMPILE
+         #define PCRE_STUDY_JIT_COMPILE 0
+         #endif
+
+       so  that  no  option  is passed to pcre_study(), and then use something
+       like this to free the study data:
+
+         #ifdef PCRE_CONFIG_JIT
+             pcre_free_study(study_ptr);
+         #else
+             pcre_free(study_ptr);
+         #endif
+
+       PCRE_STUDY_JIT_COMPILE requests the JIT compiler to generate  code  for
+       complete  matches.  If  you  want  to  run  partial  matches  using the
+       PCRE_PARTIAL_HARD or  PCRE_PARTIAL_SOFT  options  of  pcre_exec(),  you
+       should  set  one  or  both  of the following options in addition to, or
+       instead of, PCRE_STUDY_JIT_COMPILE when you call pcre_study():
+
+         PCRE_STUDY_JIT_PARTIAL_HARD_COMPILE
+         PCRE_STUDY_JIT_PARTIAL_SOFT_COMPILE
+
+       The JIT compiler generates different optimized code  for  each  of  the
+       three  modes  (normal, soft partial, hard partial). When pcre_exec() is
+       called, the appropriate code is run if it is available. Otherwise,  the
+       pattern is matched using interpretive code.
+
+       In  some circumstances you may need to call additional functions. These
+       are described in the  section  entitled  "Controlling  the  JIT  stack"
+       below.
+
+       If  JIT  support  is  not  available,  PCRE_STUDY_JIT_COMPILE  etc. are
+       ignored, and no JIT data is created. Otherwise, the compiled pattern is
+       passed  to the JIT compiler, which turns it into machine code that exe-
+       cutes much faster than the normal interpretive code.  When  pcre_exec()
+       is  passed  a  pcre_extra block containing a pointer to JIT code of the
+       appropriate mode (normal or hard/soft  partial),  it  obeys  that  code
+       instead  of  running  the interpreter. The result is identical, but the
+       compiled JIT code runs much faster.
+
+       There are some pcre_exec() options that are not supported for JIT  exe-
+       cution.  There  are  also  some  pattern  items that JIT cannot handle.
+       Details are given below. In both cases, execution  automatically  falls
+       back  to  the  interpretive  code.  If you want to know whether JIT was
+       actually used for a particular match, you  should  arrange  for  a  JIT
+       callback  function  to  be  set up as described in the section entitled
+       "Controlling the JIT stack" below, even if you do not need to supply  a
+       non-default  JIT stack. Such a callback function is called whenever JIT
+       code is about to be obeyed. If the execution options are not right  for
+       JIT execution, the callback function is not obeyed.
+
+       If  the  JIT  compiler finds an unsupported item, no JIT data is gener-
+       ated. You can find out if JIT execution is available after  studying  a
+       pattern  by  calling  pcre_fullinfo()  with the PCRE_INFO_JIT option. A
+       result of 1 means that JIT compilation was successful. A  result  of  0
+       means that JIT support is not available, or the pattern was not studied
+       with PCRE_STUDY_JIT_COMPILE etc., or the JIT compiler was not  able  to
+       handle the pattern.
+
+       Once a pattern has been studied, with or without JIT, it can be used as
+       many times as you like for matching different subject strings.
+
+
+UNSUPPORTED OPTIONS AND PATTERN ITEMS
+
+       The only pcre_exec() options that are supported for JIT  execution  are
+       PCRE_NO_UTF8_CHECK, PCRE_NO_UTF16_CHECK, PCRE_NO_UTF32_CHECK, PCRE_NOT-
+       BOL,  PCRE_NOTEOL,  PCRE_NOTEMPTY,   PCRE_NOTEMPTY_ATSTART,   PCRE_PAR-
+       TIAL_HARD, and PCRE_PARTIAL_SOFT.
+
+       The  only  unsupported  pattern items are \C (match a single data unit)
+       when running in a UTF mode, and a callout immediately before an  asser-
+       tion condition in a conditional group.
+
+
+RETURN VALUES FROM JIT EXECUTION
+
+       When  a  pattern  is matched using JIT execution, the return values are
+       the same as those given by the interpretive pcre_exec() code, with  the
+       addition  of  one new error code: PCRE_ERROR_JIT_STACKLIMIT. This means
+       that the memory used for the JIT stack was insufficient. See  "Control-
+       ling the JIT stack" below for a discussion of JIT stack usage. For com-
+       patibility with the interpretive pcre_exec() code, no  more  than  two-
+       thirds  of  the ovector argument is used for passing back captured sub-
+       strings.
+
+       The error code PCRE_ERROR_MATCHLIMIT is returned by  the  JIT  code  if
+       searching  a  very large pattern tree goes on for too long, as it is in
+       the same circumstance when JIT is not used, but the details of  exactly
+       what  is  counted are not the same. The PCRE_ERROR_RECURSIONLIMIT error
+       code is never returned by JIT execution.
+
+
+SAVING AND RESTORING COMPILED PATTERNS
+
+       The code that is generated by the  JIT  compiler  is  architecture-spe-
+       cific,  and  is also position dependent. For those reasons it cannot be
+       saved (in a file or database) and restored later like the bytecode  and
+       other  data  of  a compiled pattern. Saving and restoring compiled pat-
+       terns is not something many people do. More detail about this  facility
+       is  given in the pcreprecompile documentation. It should be possible to
+       run pcre_study() on a saved and restored pattern, and thereby  recreate
+       the  JIT  data, but because JIT compilation uses significant resources,
+       it is probably not worth doing this; you might as  well  recompile  the
+       original pattern.
+
+
+CONTROLLING THE JIT STACK
+
+       When the compiled JIT code runs, it needs a block of memory to use as a
+       stack.  By default, it uses 32K on the  machine  stack.  However,  some
+       large   or   complicated  patterns  need  more  than  this.  The  error
+       PCRE_ERROR_JIT_STACKLIMIT is given when  there  is  not  enough  stack.
+       Three  functions  are provided for managing blocks of memory for use as
+       JIT stacks. There is further discussion about the use of JIT stacks  in
+       the section entitled "JIT stack FAQ" below.
+
+       The  pcre_jit_stack_alloc() function creates a JIT stack. Its arguments
+       are a starting size and a maximum size, and it returns a pointer to  an
+       opaque  structure of type pcre_jit_stack, or NULL if there is an error.
+       The pcre_jit_stack_free() function can be used to free a stack that  is
+       no  longer  needed.  (For  the technically minded: the address space is
+       allocated by mmap or VirtualAlloc.)
+
+       JIT uses far less memory for recursion than the interpretive code,  and
+       a  maximum  stack size of 512K to 1M should be more than enough for any
+       pattern.
+
+       The pcre_assign_jit_stack() function specifies  which  stack  JIT  code
+       should use. Its arguments are as follows:
+
+         pcre_extra         *extra
+         pcre_jit_callback  callback
+         void               *data
+
+       The  extra  argument  must  be  the  result  of studying a pattern with
+       PCRE_STUDY_JIT_COMPILE etc. There are three cases for the values of the
+       other two options:
+
+         (1) If callback is NULL and data is NULL, an internal 32K block
+             on the machine stack is used.
+
+         (2) If callback is NULL and data is not NULL, data must be
+             a valid JIT stack, the result of calling pcre_jit_stack_alloc().
+
+         (3) If callback is not NULL, it must point to a function that is
+             called with data as an argument at the start of matching, in
+             order to set up a JIT stack. If the return from the callback
+             function is NULL, the internal 32K stack is used; otherwise the
+             return value must be a valid JIT stack, the result of calling
+             pcre_jit_stack_alloc().
+
+       A  callback function is obeyed whenever JIT code is about to be run; it
+       is not obeyed when pcre_exec() is called with options that  are  incom-
+       patible for JIT execution. A callback function can therefore be used to
+       determine whether a match operation was  executed  by  JIT  or  by  the
+       interpreter.
+
+       You may safely use the same JIT stack for more than one pattern (either
+       by assigning directly or by callback), as long as the patterns are  all
+       matched  sequentially in the same thread. In a multithread application,
+       if you do not specify a JIT stack, or if you assign or pass  back  NULL
+       from  a  callback, that is thread-safe, because each thread has its own
+       machine stack. However, if you assign  or  pass  back  a  non-NULL  JIT
+       stack,  this  must  be  a  different  stack for each thread so that the
+       application is thread-safe.
+
+       Strictly speaking, even more is allowed. You can assign the  same  non-
+       NULL  stack  to any number of patterns as long as they are not used for
+       matching by multiple threads at the same time.  For  example,  you  can
+       assign  the same stack to all compiled patterns, and use a global mutex
+       in the callback to wait until the stack is available for use.  However,
+       this is an inefficient solution, and not recommended.
+
+       This  is a suggestion for how a multithreaded program that needs to set
+       up non-default JIT stacks might operate:
+
+         During thread initalization
+           thread_local_var = pcre_jit_stack_alloc(...)
+
+         During thread exit
+           pcre_jit_stack_free(thread_local_var)
+
+         Use a one-line callback function
+           return thread_local_var
+
+       All the functions described in this section do nothing if  JIT  is  not
+       available,  and  pcre_assign_jit_stack()  does nothing unless the extra
+       argument is non-NULL and points to  a  pcre_extra  block  that  is  the
+       result of a successful study with PCRE_STUDY_JIT_COMPILE etc.
+
+
+JIT STACK FAQ
+
+       (1) Why do we need JIT stacks?
+
+       PCRE  (and JIT) is a recursive, depth-first engine, so it needs a stack
+       where the local data of the current node is pushed before checking  its
+       child nodes.  Allocating real machine stack on some platforms is diffi-
+       cult. For example, the stack chain needs to be updated every time if we
+       extend  the  stack  on  PowerPC.  Although it is possible, its updating
+       time overhead decreases performance. So we do the recursion in memory.
+
+       (2) Why don't we simply allocate blocks of memory with malloc()?
+
+       Modern operating systems have a  nice  feature:  they  can  reserve  an
+       address space instead of allocating memory. We can safely allocate mem-
+       ory pages inside this address space, so the stack  could  grow  without
+       moving memory data (this is important because of pointers). Thus we can
+       allocate 1M address space, and use only a single memory  page  (usually
+       4K)  if  that is enough. However, we can still grow up to 1M anytime if
+       needed.
+
+       (3) Who "owns" a JIT stack?
+
+       The owner of the stack is the user program, not the JIT studied pattern
+       or  anything else. The user program must ensure that if a stack is used
+       by pcre_exec(), (that is, it is assigned to the pattern currently  run-
+       ning), that stack must not be used by any other threads (to avoid over-
+       writing the same memory area). The best practice for multithreaded pro-
+       grams  is  to  allocate  a stack for each thread, and return this stack
+       through the JIT callback function.
+
+       (4) When should a JIT stack be freed?
+
+       You can free a JIT stack at any time, as long as it will not be used by
+       pcre_exec()  again.  When  you  assign  the  stack to a pattern, only a
+       pointer is set. There is no reference counting or any other magic.  You
+       can  free  the  patterns  and stacks in any order, anytime. Just do not
+       call pcre_exec() with a pattern pointing to an already freed stack,  as
+       that  will cause SEGFAULT. (Also, do not free a stack currently used by
+       pcre_exec() in another thread). You can also replace the  stack  for  a
+       pattern  at  any  time.  You  can  even  free the previous stack before
+       assigning a replacement.
+
+       (5) Should I allocate/free a  stack  every  time  before/after  calling
+       pcre_exec()?
+
+       No,  because  this  is  too  costly in terms of resources. However, you
+       could implement some clever idea which release the stack if it  is  not
+       used  in  let's  say  two minutes. The JIT callback can help to achieve
+       this without keeping a list of the currently JIT studied patterns.
+
+       (6) OK, the stack is for long term memory allocation. But what  happens
+       if  a pattern causes stack overflow with a stack of 1M? Is that 1M kept
+       until the stack is freed?
+
+       Especially on embedded sytems, it might be a good idea to release  mem-
+       ory  sometimes  without  freeing the stack. There is no API for this at
+       the moment.  Probably a function call which returns with the  currently
+       allocated  memory for any stack and another which allows releasing mem-
+       ory (shrinking the stack) would be a good idea if someone needs this.
+
+       (7) This is too much of a headache. Isn't there any better solution for
+       JIT stack handling?
+
+       No,  thanks to Windows. If POSIX threads were used everywhere, we could
+       throw out this complicated API.
+
+
+EXAMPLE CODE
+
+       This is a single-threaded example that specifies a  JIT  stack  without
+       using a callback.
+
+         int rc;
+         int ovector[30];
+         pcre *re;
+         pcre_extra *extra;
+         pcre_jit_stack *jit_stack;
+
+         re = pcre_compile(pattern, 0, &error, &erroffset, NULL);
+         /* Check for errors */
+         extra = pcre_study(re, PCRE_STUDY_JIT_COMPILE, &error);
+         jit_stack = pcre_jit_stack_alloc(32*1024, 512*1024);
+         /* Check for error (NULL) */
+         pcre_assign_jit_stack(extra, NULL, jit_stack);
+         rc = pcre_exec(re, extra, subject, length, 0, 0, ovector, 30);
+         /* Check results */
+         pcre_free(re);
+         pcre_free_study(extra);
+         pcre_jit_stack_free(jit_stack);
+
+
+JIT FAST PATH API
+
+       Because  the  API  described  above falls back to interpreted execution
+       when JIT is not available, it is convenient for programs that are writ-
+       ten  for  general  use  in  many environments. However, calling JIT via
+       pcre_exec() does have a performance impact. Programs that  are  written
+       for  use  where  JIT  is known to be available, and which need the best
+       possible performance, can instead use a "fast path"  API  to  call  JIT
+       execution  directly  instead of calling pcre_exec() (obviously only for
+       patterns that have been successfully studied by JIT).
+
+       The fast path function is called pcre_jit_exec(), and it takes  exactly
+       the  same  arguments  as pcre_exec(), plus one additional argument that
+       must point to a JIT stack. The JIT stack arrangements  described  above
+       do not apply. The return values are the same as for pcre_exec().
+
+       When  you  call  pcre_exec(), as well as testing for invalid options, a
+       number of other sanity checks are performed on the arguments. For exam-
+       ple,  if  the  subject  pointer  is NULL, or its length is negative, an
+       immediate error is given. Also, unless PCRE_NO_UTF[8|16|32] is  set,  a
+       UTF  subject  string is tested for validity. In the interests of speed,
+       these checks do not happen on the JIT fast path, and if invalid data is
+       passed, the result is undefined.
+
+       Bypassing  the  sanity  checks  and  the  pcre_exec() wrapping can give
+       speedups of more than 10%.
+
+
+SEE ALSO
+
+       pcreapi(3)
+
+
+AUTHOR
+
+       Philip Hazel (FAQ by Zoltan Herczeg)
+       University Computing Service
+       Cambridge CB2 3QH, England.
+
+
+REVISION
+
+       Last updated: 17 March 2013
+       Copyright (c) 1997-2013 University of Cambridge.
+------------------------------------------------------------------------------
+
+
+PCREPARTIAL(3)             Library Functions Manual             PCREPARTIAL(3)
+
+
+
+NAME
+       PCRE - Perl-compatible regular expressions
+
+PARTIAL MATCHING IN PCRE
+
+       In normal use of PCRE, if the subject string that is passed to a match-
+       ing function matches as far as it goes, but is too short to  match  the
+       entire pattern, PCRE_ERROR_NOMATCH is returned. There are circumstances
+       where it might be helpful to distinguish this case from other cases  in
+       which there is no match.
+
+       Consider, for example, an application where a human is required to type
+       in data for a field with specific formatting requirements.  An  example
+       might be a date in the form ddmmmyy, defined by this pattern:
+
+         ^\d?\d(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\d\d$
+
+       If the application sees the user's keystrokes one by one, and can check
+       that what has been typed so far is potentially valid,  it  is  able  to
+       raise  an  error  as  soon  as  a  mistake  is made, by beeping and not
+       reflecting the character that has been typed, for example. This immedi-
+       ate  feedback is likely to be a better user interface than a check that
+       is delayed until the entire string has been entered.  Partial  matching
+       can  also be useful when the subject string is very long and is not all
+       available at once.
+
+       PCRE supports partial matching by means of  the  PCRE_PARTIAL_SOFT  and
+       PCRE_PARTIAL_HARD  options,  which  can  be set when calling any of the
+       matching functions. For backwards compatibility, PCRE_PARTIAL is a syn-
+       onym  for  PCRE_PARTIAL_SOFT.  The essential difference between the two
+       options is whether or not a partial match is preferred to  an  alterna-
+       tive complete match, though the details differ between the two types of
+       matching function. If both options  are  set,  PCRE_PARTIAL_HARD  takes
+       precedence.
+
+       If  you  want to use partial matching with just-in-time optimized code,
+       you must call pcre_study(), pcre16_study() or  pcre32_study() with  one
+       or both of these options:
+
+         PCRE_STUDY_JIT_PARTIAL_SOFT_COMPILE
+         PCRE_STUDY_JIT_PARTIAL_HARD_COMPILE
+
+       PCRE_STUDY_JIT_COMPILE  should also be set if you are going to run non-
+       partial matches on the same pattern. If the appropriate JIT study  mode
+       has not been set for a match, the interpretive matching code is used.
+
+       Setting a partial matching option disables two of PCRE's standard opti-
+       mizations. PCRE remembers the last literal data unit in a pattern,  and
+       abandons  matching  immediately  if  it  is  not present in the subject
+       string. This optimization cannot be used  for  a  subject  string  that
+       might  match only partially. If the pattern was studied, PCRE knows the
+       minimum length of a matching string, and does not  bother  to  run  the
+       matching  function  on  shorter strings. This optimization is also dis-
+       abled for partial matching.
+
+
+PARTIAL MATCHING USING pcre_exec() OR pcre[16|32]_exec()
+
+       A  partial   match   occurs   during   a   call   to   pcre_exec()   or
+       pcre[16|32]_exec()  when  the end of the subject string is reached suc-
+       cessfully, but matching cannot continue  because  more  characters  are
+       needed.   However, at least one character in the subject must have been
+       inspected. This character need not  form  part  of  the  final  matched
+       string;  lookbehind  assertions and the \K escape sequence provide ways
+       of inspecting characters before the start of a matched  substring.  The
+       requirement  for  inspecting  at  least one character exists because an
+       empty string can always be matched; without such  a  restriction  there
+       would  always  be  a partial match of an empty string at the end of the
+       subject.
+
+       If there are at least two slots in the offsets vector  when  a  partial
+       match  is returned, the first slot is set to the offset of the earliest
+       character that was inspected. For convenience, the second offset points
+       to the end of the subject so that a substring can easily be identified.
+       If there are at least three slots in the offsets vector, the third slot
+       is set to the offset of the character where matching started.
+
+       For the majority of patterns, the contents of the first and third slots
+       will be the same. However, for patterns that contain lookbehind  asser-
+       tions, or begin with \b or \B, characters before the one where matching
+       started may have been inspected while carrying out the match. For exam-
+       ple, consider this pattern:
+
+         /(?<=abc)123/
+
+       This pattern matches "123", but only if it is preceded by "abc". If the
+       subject string is "xyzabc12", the first two  offsets  after  a  partial
+       match  are for the substring "abc12", because all these characters were
+       inspected. However, the third offset is set to 6, because that  is  the
+       offset where matching began.
+
+       What happens when a partial match is identified depends on which of the
+       two partial matching options are set.
+
+   PCRE_PARTIAL_SOFT WITH pcre_exec() OR pcre[16|32]_exec()
+
+       If PCRE_PARTIAL_SOFT is  set  when  pcre_exec()  or  pcre[16|32]_exec()
+       identifies a partial match, the partial match is remembered, but match-
+       ing continues as normal, and other  alternatives  in  the  pattern  are
+       tried.  If  no  complete  match  can  be  found,  PCRE_ERROR_PARTIAL is
+       returned instead of PCRE_ERROR_NOMATCH.
+
+       This option is "soft" because it prefers a complete match over  a  par-
+       tial  match.   All the various matching items in a pattern behave as if
+       the subject string is potentially complete. For example, \z, \Z, and  $
+       match  at  the end of the subject, as normal, and for \b and \B the end
+       of the subject is treated as a non-alphanumeric.
+
+       If there is more than one partial match, the first one that  was  found
+       provides the data that is returned. Consider this pattern:
+
+         /123\w+X|dogY/
+
+       If  this is matched against the subject string "abc123dog", both alter-
+       natives fail to match, but the end of the  subject  is  reached  during
+       matching,  so  PCRE_ERROR_PARTIAL is returned. The offsets are set to 3
+       and 9, identifying "123dog" as the first partial match that was  found.
+       (In  this  example, there are two partial matches, because "dog" on its
+       own partially matches the second alternative.)
+
+   PCRE_PARTIAL_HARD WITH pcre_exec() OR pcre[16|32]_exec()
+
+       If PCRE_PARTIAL_HARD is  set  for  pcre_exec()  or  pcre[16|32]_exec(),
+       PCRE_ERROR_PARTIAL  is  returned  as  soon as a partial match is found,
+       without continuing to search for possible complete matches. This option
+       is "hard" because it prefers an earlier partial match over a later com-
+       plete match. For this reason, the assumption is made that  the  end  of
+       the  supplied  subject  string may not be the true end of the available
+       data, and so, if \z, \Z, \b, \B, or $ are encountered at the end of the
+       subject,  the  result is PCRE_ERROR_PARTIAL, provided that at least one
+       character in the subject has been inspected.
+
+       Setting PCRE_PARTIAL_HARD also affects the way UTF-8 and UTF-16 subject
+       strings  are checked for validity. Normally, an invalid sequence causes
+       the error PCRE_ERROR_BADUTF8 or PCRE_ERROR_BADUTF16.  However,  in  the
+       special  case  of  a  truncated  character  at  the end of the subject,
+       PCRE_ERROR_SHORTUTF8  or   PCRE_ERROR_SHORTUTF16   is   returned   when
+       PCRE_PARTIAL_HARD is set.
+
+   Comparing hard and soft partial matching
+
+       The  difference  between the two partial matching options can be illus-
+       trated by a pattern such as:
+
+         /dog(sbody)?/
+
+       This matches either "dog" or "dogsbody", greedily (that is, it  prefers
+       the  longer  string  if  possible). If it is matched against the string
+       "dog" with PCRE_PARTIAL_SOFT, it yields a  complete  match  for  "dog".
+       However, if PCRE_PARTIAL_HARD is set, the result is PCRE_ERROR_PARTIAL.
+       On the other hand, if the pattern is made ungreedy the result  is  dif-
+       ferent:
+
+         /dog(sbody)??/
+
+       In  this  case  the  result  is always a complete match because that is
+       found first, and matching never  continues  after  finding  a  complete
+       match. It might be easier to follow this explanation by thinking of the
+       two patterns like this:
+
+         /dog(sbody)?/    is the same as  /dogsbody|dog/
+         /dog(sbody)??/   is the same as  /dog|dogsbody/
+
+       The second pattern will never match "dogsbody", because it will  always
+       find the shorter match first.
+
+
+PARTIAL MATCHING USING pcre_dfa_exec() OR pcre[16|32]_dfa_exec()
+
+       The DFA functions move along the subject string character by character,
+       without backtracking, searching for  all  possible  matches  simultane-
+       ously.  If the end of the subject is reached before the end of the pat-
+       tern, there is the possibility of a partial match, again provided  that
+       at least one character has been inspected.
+
+       When  PCRE_PARTIAL_SOFT  is set, PCRE_ERROR_PARTIAL is returned only if
+       there have been no complete matches. Otherwise,  the  complete  matches
+       are  returned.   However,  if PCRE_PARTIAL_HARD is set, a partial match
+       takes precedence over any complete matches. The portion of  the  string
+       that  was  inspected when the longest partial match was found is set as
+       the first matching string, provided there are at least two slots in the
+       offsets vector.
+
+       Because  the  DFA functions always search for all possible matches, and
+       there is no difference between greedy and  ungreedy  repetition,  their
+       behaviour  is  different  from  the  standard  functions when PCRE_PAR-
+       TIAL_HARD is  set.  Consider  the  string  "dog"  matched  against  the
+       ungreedy pattern shown above:
+
+         /dog(sbody)??/
+
+       Whereas  the  standard functions stop as soon as they find the complete
+       match for "dog", the DFA functions also  find  the  partial  match  for
+       "dogsbody", and so return that when PCRE_PARTIAL_HARD is set.
+
+
+PARTIAL MATCHING AND WORD BOUNDARIES
+
+       If  a  pattern ends with one of sequences \b or \B, which test for word
+       boundaries, partial matching with PCRE_PARTIAL_SOFT can  give  counter-
+       intuitive results. Consider this pattern:
+
+         /\bcat\b/
+
+       This matches "cat", provided there is a word boundary at either end. If
+       the subject string is "the cat", the comparison of the final "t" with a
+       following  character  cannot  take  place, so a partial match is found.
+       However, normal matching carries on, and \b matches at the end  of  the
+       subject  when  the  last  character is a letter, so a complete match is
+       found.  The  result,  therefore,  is  not   PCRE_ERROR_PARTIAL.   Using
+       PCRE_PARTIAL_HARD  in  this case does yield PCRE_ERROR_PARTIAL, because
+       then the partial match takes precedence.
+
+
+FORMERLY RESTRICTED PATTERNS
+
+       For releases of PCRE prior to 8.00, because of the way certain internal
+       optimizations   were  implemented  in  the  pcre_exec()  function,  the
+       PCRE_PARTIAL option (predecessor of  PCRE_PARTIAL_SOFT)  could  not  be
+       used  with all patterns. From release 8.00 onwards, the restrictions no
+       longer apply, and partial matching with can be requested for  any  pat-
+       tern.
+
+       Items that were formerly restricted were repeated single characters and
+       repeated metasequences. If PCRE_PARTIAL was set for a pattern that  did
+       not  conform  to  the restrictions, pcre_exec() returned the error code
+       PCRE_ERROR_BADPARTIAL (-13). This error code is no longer in  use.  The
+       PCRE_INFO_OKPARTIAL  call  to pcre_fullinfo() to find out if a compiled
+       pattern can be used for partial matching now always returns 1.
+
+
+EXAMPLE OF PARTIAL MATCHING USING PCRETEST
+
+       If the escape sequence \P is present  in  a  pcretest  data  line,  the
+       PCRE_PARTIAL_SOFT  option  is  used  for  the  match.  Here is a run of
+       pcretest that uses the date example quoted above:
+
+           re> /^\d?\d(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\d\d$/
+         data> 25jun04\P
+          0: 25jun04
+          1: jun
+         data> 25dec3\P
+         Partial match: 23dec3
+         data> 3ju\P
+         Partial match: 3ju
+         data> 3juj\P
+         No match
+         data> j\P
+         No match
+
+       The first data string is matched  completely,  so  pcretest  shows  the
+       matched  substrings.  The  remaining four strings do not match the com-
+       plete pattern, but the first two are partial matches. Similar output is
+       obtained if DFA matching is used.
+
+       If  the escape sequence \P is present more than once in a pcretest data
+       line, the PCRE_PARTIAL_HARD option is set for the match.
+
+
+MULTI-SEGMENT MATCHING WITH pcre_dfa_exec() OR pcre[16|32]_dfa_exec()
+
+       When a partial match has been found using a DFA matching  function,  it
+       is  possible to continue the match by providing additional subject data
+       and calling the function again with the same compiled  regular  expres-
+       sion,  this time setting the PCRE_DFA_RESTART option. You must pass the
+       same working space as before, because this is where details of the pre-
+       vious  partial  match  are  stored.  Here is an example using pcretest,
+       using the \R escape sequence to set  the  PCRE_DFA_RESTART  option  (\D
+       specifies the use of the DFA matching function):
+
+           re> /^\d?\d(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\d\d$/
+         data> 23ja\P\D
+         Partial match: 23ja
+         data> n05\R\D
+          0: n05
+
+       The  first  call has "23ja" as the subject, and requests partial match-
+       ing; the second call  has  "n05"  as  the  subject  for  the  continued
+       (restarted)  match.   Notice  that when the match is complete, only the
+       last part is shown; PCRE does  not  retain  the  previously  partially-
+       matched  string. It is up to the calling program to do that if it needs
+       to.
+
+       You can set the PCRE_PARTIAL_SOFT  or  PCRE_PARTIAL_HARD  options  with
+       PCRE_DFA_RESTART  to  continue partial matching over multiple segments.
+       This facility can be used to pass very long subject strings to the  DFA
+       matching functions.
+
+
+MULTI-SEGMENT MATCHING WITH pcre_exec() OR pcre[16|32]_exec()
+
+       From  release 8.00, the standard matching functions can also be used to
+       do multi-segment matching. Unlike the DFA functions, it is not possible
+       to  restart the previous match with a new segment of data. Instead, new
+       data must be added to the previous subject string, and the entire match
+       re-run,  starting from the point where the partial match occurred. Ear-
+       lier data can be discarded.
+
+       It is best to use PCRE_PARTIAL_HARD in this situation, because it  does
+       not  treat the end of a segment as the end of the subject when matching
+       \z, \Z, \b, \B, and $. Consider  an  unanchored  pattern  that  matches
+       dates:
+
+           re> /\d?\d(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\d\d/
+         data> The date is 23ja\P\P
+         Partial match: 23ja
+
+       At  this stage, an application could discard the text preceding "23ja",
+       add on text from the next  segment,  and  call  the  matching  function
+       again.  Unlike  the  DFA matching functions, the entire matching string
+       must always be available, and the complete matching process occurs  for
+       each call, so more memory and more processing time is needed.
+
+       Note:  If  the pattern contains lookbehind assertions, or \K, or starts
+       with \b or \B, the string that is returned for a partial match includes
+       characters  that precede the start of what would be returned for a com-
+       plete match, because it contains all the characters that were inspected
+       during the partial match.
+
+
+ISSUES WITH MULTI-SEGMENT MATCHING
+
+       Certain types of pattern may give problems with multi-segment matching,
+       whichever matching function is used.
+
+       1. If the pattern contains a test for the beginning of a line, you need
+       to  pass  the  PCRE_NOTBOL  option when the subject string for any call
+       does start at the beginning of a line.  There  is  also  a  PCRE_NOTEOL
+       option, but in practice when doing multi-segment matching you should be
+       using PCRE_PARTIAL_HARD, which includes the effect of PCRE_NOTEOL.
+
+       2. Lookbehind assertions that have already been obeyed are catered  for
+       in the offsets that are returned for a partial match. However a lookbe-
+       hind assertion later in the pattern could require even earlier  charac-
+       ters   to  be  inspected.  You  can  handle  this  case  by  using  the
+       PCRE_INFO_MAXLOOKBEHIND    option    of    the    pcre_fullinfo()    or
+       pcre[16|32]_fullinfo()  functions  to  obtain the length of the longest
+       lookbehind in the pattern. This length  is  given  in  characters,  not
+       bytes.  If  you  always retain at least that many characters before the
+       partially matched string, all should be  well.  (Of  course,  near  the
+       start of the subject, fewer characters may be present; in that case all
+       characters should be retained.)
+
+       From release 8.33, there is a more accurate way of deciding which char-
+       acters  to  retain.  Instead  of  subtracting the length of the longest
+       lookbehind from the  earliest  inspected  character  (offsets[0]),  the
+       match  start  position  (offsets[2]) should be used, and the next match
+       attempt started at the offsets[2] character by setting the  startoffset
+       argument of pcre_exec() or pcre_dfa_exec().
+
+       For  example, if the pattern "(?<=123)abc" is partially matched against
+       the string "xx123a", the three offset values returned are 2, 6, and  5.
+       This  indicates  that  the  matching  process that gave a partial match
+       started at offset 5, but the characters "123a" were all inspected.  The
+       maximum  lookbehind  for  that pattern is 3, so taking that away from 5
+       shows that we need only keep "123a", and the next match attempt can  be
+       started at offset 3 (that is, at "a") when further characters have been
+       added. When the match start is not the  earliest  inspected  character,
+       pcretest shows it explicitly:
+
+           re> "(?<=123)abc"
+         data> xx123a\P\P
+         Partial match at offset 5: 123a
+
+       3.  Because a partial match must always contain at least one character,
+       what might be considered a partial match of an  empty  string  actually
+       gives a "no match" result. For example:
+
+           re> /c(?<=abc)x/
+         data> ab\P
+         No match
+
+       If the next segment begins "cx", a match should be found, but this will
+       only happen if characters from the previous segment are  retained.  For
+       this  reason,  a  "no  match"  result should be interpreted as "partial
+       match of an empty string" when the pattern contains lookbehinds.
+
+       4. Matching a subject string that is split into multiple  segments  may
+       not  always produce exactly the same result as matching over one single
+       long string, especially when PCRE_PARTIAL_SOFT  is  used.  The  section
+       "Partial  Matching  and  Word Boundaries" above describes an issue that
+       arises if the pattern ends with \b or \B. Another  kind  of  difference
+       may  occur when there are multiple matching possibilities, because (for
+       PCRE_PARTIAL_SOFT) a partial match result is given only when there  are
+       no completed matches. This means that as soon as the shortest match has
+       been found, continuation to a new subject segment is no  longer  possi-
+       ble. Consider again this pcretest example:
+
+           re> /dog(sbody)?/
+         data> dogsb\P
+          0: dog
+         data> do\P\D
+         Partial match: do
+         data> gsb\R\P\D
+          0: g
+         data> dogsbody\D
+          0: dogsbody
+          1: dog
+
+       The  first  data  line passes the string "dogsb" to a standard matching
+       function, setting the PCRE_PARTIAL_SOFT option. Although the string  is
+       a  partial  match for "dogsbody", the result is not PCRE_ERROR_PARTIAL,
+       because the shorter string "dog" is a complete match.  Similarly,  when
+       the  subject  is  presented to a DFA matching function in several parts
+       ("do" and "gsb" being the first two) the match  stops  when  "dog"  has
+       been  found, and it is not possible to continue.  On the other hand, if
+       "dogsbody" is presented as a single string,  a  DFA  matching  function
+       finds both matches.
+
+       Because  of  these  problems,  it is best to use PCRE_PARTIAL_HARD when
+       matching multi-segment data. The example  above  then  behaves  differ-
+       ently:
+
+           re> /dog(sbody)?/
+         data> dogsb\P\P
+         Partial match: dogsb
+         data> do\P\D
+         Partial match: do
+         data> gsb\R\P\P\D
+         Partial match: gsb
+
+       5. Patterns that contain alternatives at the top level which do not all
+       start with the  same  pattern  item  may  not  work  as  expected  when
+       PCRE_DFA_RESTART is used. For example, consider this pattern:
+
+         1234|3789
+
+       If  the  first  part of the subject is "ABC123", a partial match of the
+       first alternative is found at offset 3. There is no partial  match  for
+       the second alternative, because such a match does not start at the same
+       point in the subject string. Attempting to  continue  with  the  string
+       "7890"  does  not  yield  a  match because only those alternatives that
+       match at one point in the subject are remembered.  The  problem  arises
+       because  the  start  of the second alternative matches within the first
+       alternative. There is no problem with  anchored  patterns  or  patterns
+       such as:
+
+         1234|ABCD
+
+       where  no  string can be a partial match for both alternatives. This is
+       not a problem if a standard matching  function  is  used,  because  the
+       entire match has to be rerun each time:
+
+           re> /1234|3789/
+         data> ABC123\P\P
+         Partial match: 123
+         data> 1237890
+          0: 3789
+
+       Of course, instead of using PCRE_DFA_RESTART, the same technique of re-
+       running the entire match can also be used with the DFA  matching  func-
+       tions.  Another  possibility  is to work with two buffers. If a partial
+       match at offset n in the first buffer is followed by  "no  match"  when
+       PCRE_DFA_RESTART  is  used on the second buffer, you can then try a new
+       match starting at offset n+1 in the first buffer.
+
+
+AUTHOR
+
+       Philip Hazel
+       University Computing Service
+       Cambridge CB2 3QH, England.
+
+
+REVISION
+
+       Last updated: 20 February 2013
+       Copyright (c) 1997-2013 University of Cambridge.
+------------------------------------------------------------------------------
+
+
+PCREPRECOMPILE(3)          Library Functions Manual          PCREPRECOMPILE(3)
+
+
+
+NAME
+       PCRE - Perl-compatible regular expressions
+
+SAVING AND RE-USING PRECOMPILED PCRE PATTERNS
+
+       If  you  are running an application that uses a large number of regular
+       expression patterns, it may be useful to store them  in  a  precompiled
+       form  instead  of  having to compile them every time the application is
+       run.  If you are not  using  any  private  character  tables  (see  the
+       pcre_maketables()  documentation),  this is relatively straightforward.
+       If you are using private tables, it is a little bit  more  complicated.
+       However,  if you are using the just-in-time optimization feature, it is
+       not possible to save and reload the JIT data.
+
+       If you save compiled patterns to a file, you can copy them to a differ-
+       ent host and run them there. If the two hosts have different endianness
+       (byte    order),    you     should     run     the     pcre[16|32]_pat-
+       tern_to_host_byte_order()  function  on  the  new host before trying to
+       match the pattern. The matching functions return  PCRE_ERROR_BADENDIAN-
+       NESS if they detect a pattern with the wrong endianness.
+
+       Compiling  regular  expressions with one version of PCRE for use with a
+       different version is not guaranteed to work and may cause crashes,  and
+       saving  and  restoring  a  compiled  pattern loses any JIT optimization
+       data.
+
+
+SAVING A COMPILED PATTERN
+
+       The value returned by pcre[16|32]_compile() points to a single block of
+       memory  that  holds  the  compiled pattern and associated data. You can
+       find   the   length   of   this   block    in    bytes    by    calling
+       pcre[16|32]_fullinfo() with an argument of PCRE_INFO_SIZE. You can then
+       save the data in any appropriate manner. Here is sample  code  for  the
+       8-bit  library  that  compiles  a  pattern  and writes it to a file. It
+       assumes that the variable fd refers to a file that is open for output:
+
+         int erroroffset, rc, size;
+         char *error;
+         pcre *re;
+
+         re = pcre_compile("my pattern", 0, &error, &erroroffset, NULL);
+         if (re == NULL) { ... handle errors ... }
+         rc = pcre_fullinfo(re, NULL, PCRE_INFO_SIZE, &size);
+         if (rc < 0) { ... handle errors ... }
+         rc = fwrite(re, 1, size, fd);
+         if (rc != size) { ... handle errors ... }
+
+       In this example, the bytes  that  comprise  the  compiled  pattern  are
+       copied  exactly.  Note that this is binary data that may contain any of
+       the 256 possible byte  values.  On  systems  that  make  a  distinction
+       between binary and non-binary data, be sure that the file is opened for
+       binary output.
+
+       If you want to write more than one pattern to a file, you will have  to
+       devise  a  way of separating them. For binary data, preceding each pat-
+       tern with its length is probably  the  most  straightforward  approach.
+       Another  possibility is to write out the data in hexadecimal instead of
+       binary, one pattern to a line.
+
+       Saving compiled patterns in a file is only one possible way of  storing
+       them  for later use. They could equally well be saved in a database, or
+       in the memory of some daemon process that passes them  via  sockets  to
+       the processes that want them.
+
+       If the pattern has been studied, it is also possible to save the normal
+       study data in a similar way to the compiled pattern itself. However, if
+       the PCRE_STUDY_JIT_COMPILE was used, the just-in-time data that is cre-
+       ated cannot be saved because it is too dependent on the  current  envi-
+       ronment.    When    studying    generates    additional    information,
+       pcre[16|32]_study() returns  a  pointer  to  a  pcre[16|32]_extra  data
+       block.  Its  format  is defined in the section on matching a pattern in
+       the pcreapi documentation. The study_data field points  to  the  binary
+       study  data,  and this is what you must save (not the pcre[16|32]_extra
+       block itself). The length of the study data can be obtained by  calling
+       pcre[16|32]_fullinfo()  with an argument of PCRE_INFO_STUDYSIZE. Remem-
+       ber to check that  pcre[16|32]_study()  did  return  a  non-NULL  value
+       before trying to save the study data.
+
+
+RE-USING A PRECOMPILED PATTERN
+
+       Re-using  a  precompiled pattern is straightforward. Having reloaded it
+       into main memory,  called  pcre[16|32]_pattern_to_host_byte_order()  if
+       necessary,    you   pass   its   pointer   to   pcre[16|32]_exec()   or
+       pcre[16|32]_dfa_exec() in the usual way.
+
+       However, if you passed a pointer to custom character  tables  when  the
+       pattern  was compiled (the tableptr argument of pcre[16|32]_compile()),
+       you  must  now  pass  a  similar  pointer  to   pcre[16|32]_exec()   or
+       pcre[16|32]_dfa_exec(),  because the value saved with the compiled pat-
+       tern will obviously be nonsense. A field in a pcre[16|32]_extra() block
+       is  used  to  pass this data, as described in the section on matching a
+       pattern in the pcreapi documentation.
+
+       If you did not provide custom character tables  when  the  pattern  was
+       compiled, the pointer in the compiled pattern is NULL, which causes the
+       matching functions to use PCRE's internal tables. Thus, you do not need
+       to take any special action at run time in this case.
+
+       If  you  saved study data with the compiled pattern, you need to create
+       your own pcre[16|32]_extra data block and set the study_data  field  to
+       point   to   the   reloaded   study   data.   You  must  also  set  the
+       PCRE_EXTRA_STUDY_DATA bit in the flags field  to  indicate  that  study
+       data  is present. Then pass the pcre[16|32]_extra block to the matching
+       function in the usual way. If the pattern was studied for  just-in-time
+       optimization,  that  data  cannot  be  saved,  and  so  is  lost  by  a
+       save/restore cycle.
+
+
+COMPATIBILITY WITH DIFFERENT PCRE RELEASES
+
+       In general, it is safest to  recompile  all  saved  patterns  when  you
+       update  to  a new PCRE release, though not all updates actually require
+       this.
+
+
+AUTHOR
+
+       Philip Hazel
+       University Computing Service
+       Cambridge CB2 3QH, England.
+
+
+REVISION
+
+       Last updated: 24 June 2012
+       Copyright (c) 1997-2012 University of Cambridge.
+------------------------------------------------------------------------------
+
+
+PCREPERFORM(3)             Library Functions Manual             PCREPERFORM(3)
+
+
+
+NAME
+       PCRE - Perl-compatible regular expressions
+
+PCRE PERFORMANCE
+
+       Two  aspects  of performance are discussed below: memory usage and pro-
+       cessing time. The way you express your pattern as a regular  expression
+       can affect both of them.
+
+
+COMPILED PATTERN MEMORY USAGE
+
+       Patterns  are compiled by PCRE into a reasonably efficient interpretive
+       code, so that most simple patterns do not  use  much  memory.  However,
+       there  is  one case where the memory usage of a compiled pattern can be
+       unexpectedly large. If a parenthesized subpattern has a quantifier with
+       a minimum greater than 1 and/or a limited maximum, the whole subpattern
+       is repeated in the compiled code. For example, the pattern
+
+         (abc|def){2,4}
+
+       is compiled as if it were
+
+         (abc|def)(abc|def)((abc|def)(abc|def)?)?
+
+       (Technical aside: It is done this way so that backtrack  points  within
+       each of the repetitions can be independently maintained.)
+
+       For  regular expressions whose quantifiers use only small numbers, this
+       is not usually a problem. However, if the numbers are large,  and  par-
+       ticularly  if  such repetitions are nested, the memory usage can become
+       an embarrassment. For example, the very simple pattern
+
+         ((ab){1,1000}c){1,3}
+
+       uses 51K bytes when compiled using the 8-bit library. When PCRE is com-
+       piled  with  its  default  internal pointer size of two bytes, the size
+       limit on a compiled pattern is 64K data units, and this is reached with
+       the  above  pattern  if  the outer repetition is increased from 3 to 4.
+       PCRE can be compiled to use larger internal pointers  and  thus  handle
+       larger  compiled patterns, but it is better to try to rewrite your pat-
+       tern to use less memory if you can.
+
+       One way of reducing the memory usage for such patterns is to  make  use
+       of PCRE's "subroutine" facility. Re-writing the above pattern as
+
+         ((ab)(?2){0,999}c)(?1){0,2}
+
+       reduces the memory requirements to 18K, and indeed it remains under 20K
+       even with the outer repetition increased to 100. However, this  pattern
+       is  not  exactly equivalent, because the "subroutine" calls are treated
+       as atomic groups into which there can be no backtracking if there is  a
+       subsequent  matching  failure.  Therefore,  PCRE cannot do this kind of
+       rewriting automatically.  Furthermore, there is a  noticeable  loss  of
+       speed  when executing the modified pattern. Nevertheless, if the atomic
+       grouping is not a problem and the loss of  speed  is  acceptable,  this
+       kind  of  rewriting will allow you to process patterns that PCRE cannot
+       otherwise handle.
+
+
+STACK USAGE AT RUN TIME
+
+       When pcre_exec() or pcre[16|32]_exec() is used  for  matching,  certain
+       kinds  of  pattern  can  cause  it  to use large amounts of the process
+       stack. In some environments the default process stack is  quite  small,
+       and  if it runs out the result is often SIGSEGV. This issue is probably
+       the most frequently raised problem with PCRE.  Rewriting  your  pattern
+       can  often  help.  The  pcrestack documentation discusses this issue in
+       detail.
+
+
+PROCESSING TIME
+
+       Certain items in regular expression patterns are processed  more  effi-
+       ciently than others. It is more efficient to use a character class like
+       [aeiou]  than  a  set  of   single-character   alternatives   such   as
+       (a|e|i|o|u).  In  general,  the simplest construction that provides the
+       required behaviour is usually the most efficient. Jeffrey Friedl's book
+       contains  a  lot  of useful general discussion about optimizing regular
+       expressions for efficient performance. This  document  contains  a  few
+       observations about PCRE.
+
+       Using  Unicode  character  properties  (the  \p, \P, and \X escapes) is
+       slow, because PCRE has to use a multi-stage table  lookup  whenever  it
+       needs  a  character's  property. If you can find an alternative pattern
+       that does not use character properties, it will probably be faster.
+
+       By default, the escape sequences \b, \d, \s,  and  \w,  and  the  POSIX
+       character  classes  such  as  [:alpha:]  do not use Unicode properties,
+       partly for backwards compatibility, and partly for performance reasons.
+       However,  you can set PCRE_UCP if you want Unicode character properties
+       to be used. This can double the matching time for  items  such  as  \d,
+       when matched with a traditional matching function; the performance loss
+       is less with a DFA matching function, and in both cases  there  is  not
+       much difference for \b.
+
+       When  a  pattern  begins  with .* not in parentheses, or in parentheses
+       that are not the subject of a backreference, and the PCRE_DOTALL option
+       is  set, the pattern is implicitly anchored by PCRE, since it can match
+       only at the start of a subject string. However, if PCRE_DOTALL  is  not
+       set,  PCRE  cannot  make this optimization, because the . metacharacter
+       does not then match a newline, and if the subject string contains  new-
+       lines,  the  pattern may match from the character immediately following
+       one of them instead of from the very start. For example, the pattern
+
+         .*second
+
+       matches the subject "first\nand second" (where \n stands for a  newline
+       character),  with the match starting at the seventh character. In order
+       to do this, PCRE has to retry the match starting after every newline in
+       the subject.
+
+       If  you  are using such a pattern with subject strings that do not con-
+       tain newlines, the best performance is obtained by setting PCRE_DOTALL,
+       or  starting  the pattern with ^.* or ^.*? to indicate explicit anchor-
+       ing. That saves PCRE from having to scan along the subject looking  for
+       a newline to restart at.
+
+       Beware  of  patterns  that contain nested indefinite repeats. These can
+       take a long time to run when applied to a string that does  not  match.
+       Consider the pattern fragment
+
+         ^(a+)*
+
+       This  can  match "aaaa" in 16 different ways, and this number increases
+       very rapidly as the string gets longer. (The * repeat can match  0,  1,
+       2,  3, or 4 times, and for each of those cases other than 0 or 4, the +
+       repeats can match different numbers of times.) When  the  remainder  of
+       the pattern is such that the entire match is going to fail, PCRE has in
+       principle to try  every  possible  variation,  and  this  can  take  an
+       extremely long time, even for relatively short strings.
+
+       An optimization catches some of the more simple cases such as
+
+         (a+)*b
+
+       where  a  literal  character  follows. Before embarking on the standard
+       matching procedure, PCRE checks that there is a "b" later in  the  sub-
+       ject  string, and if there is not, it fails the match immediately. How-
+       ever, when there is no following literal this  optimization  cannot  be
+       used. You can see the difference by comparing the behaviour of
+
+         (a+)*\d
+
+       with  the  pattern  above.  The former gives a failure almost instantly
+       when applied to a whole line of  "a"  characters,  whereas  the  latter
+       takes an appreciable time with strings longer than about 20 characters.
+
+       In many cases, the solution to this kind of performance issue is to use
+       an atomic group or a possessive quantifier.
+
+
+AUTHOR
+
+       Philip Hazel
+       University Computing Service
+       Cambridge CB2 3QH, England.
+
+
+REVISION
+
+       Last updated: 25 August 2012
+       Copyright (c) 1997-2012 University of Cambridge.
+------------------------------------------------------------------------------
+
+
+PCREPOSIX(3)               Library Functions Manual               PCREPOSIX(3)
+
+
+
+NAME
+       PCRE - Perl-compatible regular expressions.
+
+SYNOPSIS OF POSIX API
+
+       #include 
+
+       int regcomp(regex_t *preg, const char *pattern,
+            int cflags);
+
+       int regexec(regex_t *preg, const char *string,
+            size_t nmatch, regmatch_t pmatch[], int eflags);
+
+       size_t regerror(int errcode, const regex_t *preg,
+            char *errbuf, size_t errbuf_size);
+
+       void regfree(regex_t *preg);
+
+
+DESCRIPTION
+
+       This  set  of functions provides a POSIX-style API for the PCRE regular
+       expression 8-bit library. See the pcreapi documentation for a  descrip-
+       tion  of  PCRE's native API, which contains much additional functional-
+       ity. There is no POSIX-style  wrapper  for  PCRE's  16-bit  and  32-bit
+       library.
+
+       The functions described here are just wrapper functions that ultimately
+       call  the  PCRE  native  API.  Their  prototypes  are  defined  in  the
+       pcreposix.h  header  file,  and  on  Unix systems the library itself is
+       called pcreposix.a, so can be accessed by  adding  -lpcreposix  to  the
+       command  for  linking  an application that uses them. Because the POSIX
+       functions call the native ones, it is also necessary to add -lpcre.
+
+       I have implemented only those POSIX option bits that can be  reasonably
+       mapped  to PCRE native options. In addition, the option REG_EXTENDED is
+       defined with the value zero. This has no  effect,  but  since  programs
+       that  are  written  to  the POSIX interface often use it, this makes it
+       easier to slot in PCRE as a replacement library.  Other  POSIX  options
+       are not even defined.
+
+       There  are also some other options that are not defined by POSIX. These
+       have been added at the request of users who want to make use of certain
+       PCRE-specific features via the POSIX calling interface.
+
+       When  PCRE  is  called  via these functions, it is only the API that is
+       POSIX-like in style. The syntax and semantics of  the  regular  expres-
+       sions  themselves  are  still  those of Perl, subject to the setting of
+       various PCRE options, as described below. "POSIX-like in  style"  means
+       that  the  API  approximates  to  the POSIX definition; it is not fully
+       POSIX-compatible, and in multi-byte encoding  domains  it  is  probably
+       even less compatible.
+
+       The  header for these functions is supplied as pcreposix.h to avoid any
+       potential clash with other POSIX  libraries.  It  can,  of  course,  be
+       renamed or aliased as regex.h, which is the "correct" name. It provides
+       two structure types, regex_t for  compiled  internal  forms,  and  reg-
+       match_t  for  returning  captured substrings. It also defines some con-
+       stants whose names start  with  "REG_";  these  are  used  for  setting
+       options and identifying error codes.
+
+
+COMPILING A PATTERN
+
+       The  function regcomp() is called to compile a pattern into an internal
+       form. The pattern is a C string terminated by a  binary  zero,  and  is
+       passed  in  the  argument  pattern. The preg argument is a pointer to a
+       regex_t structure that is used as a base for storing information  about
+       the compiled regular expression.
+
+       The argument cflags is either zero, or contains one or more of the bits
+       defined by the following macros:
+
+         REG_DOTALL
+
+       The PCRE_DOTALL option is set when the regular expression is passed for
+       compilation to the native function. Note that REG_DOTALL is not part of
+       the POSIX standard.
+
+         REG_ICASE
+
+       The PCRE_CASELESS option is set when the regular expression  is  passed
+       for compilation to the native function.
+
+         REG_NEWLINE
+
+       The  PCRE_MULTILINE option is set when the regular expression is passed
+       for compilation to the native function. Note that this does  not  mimic
+       the  defined  POSIX  behaviour  for REG_NEWLINE (see the following sec-
+       tion).
+
+         REG_NOSUB
+
+       The PCRE_NO_AUTO_CAPTURE option is set when the regular  expression  is
+       passed for compilation to the native function. In addition, when a pat-
+       tern that is compiled with this flag is passed to regexec() for  match-
+       ing,  the  nmatch  and  pmatch  arguments  are ignored, and no captured
+       strings are returned.
+
+         REG_UCP
+
+       The PCRE_UCP option is set when the regular expression  is  passed  for
+       compilation  to  the  native  function. This causes PCRE to use Unicode
+       properties when matchine \d, \w,  etc.,  instead  of  just  recognizing
+       ASCII values. Note that REG_UTF8 is not part of the POSIX standard.
+
+         REG_UNGREEDY
+
+       The  PCRE_UNGREEDY  option is set when the regular expression is passed
+       for compilation to the native function. Note that REG_UNGREEDY  is  not
+       part of the POSIX standard.
+
+         REG_UTF8
+
+       The  PCRE_UTF8  option is set when the regular expression is passed for
+       compilation to the native function. This causes the pattern itself  and
+       all  data  strings used for matching it to be treated as UTF-8 strings.
+       Note that REG_UTF8 is not part of the POSIX standard.
+
+       In the absence of these flags, no options  are  passed  to  the  native
+       function.   This  means  the  the  regex  is compiled with PCRE default
+       semantics. In particular, the way it handles newline characters in  the
+       subject  string  is  the Perl way, not the POSIX way. Note that setting
+       PCRE_MULTILINE has only some of the effects specified for  REG_NEWLINE.
+       It  does not affect the way newlines are matched by . (they are not) or
+       by a negative class such as [^a] (they are).
+
+       The yield of regcomp() is zero on success, and non-zero otherwise.  The
+       preg structure is filled in on success, and one member of the structure
+       is public: re_nsub contains the number of capturing subpatterns in  the
+       regular expression. Various error codes are defined in the header file.
+
+       NOTE:  If  the  yield of regcomp() is non-zero, you must not attempt to
+       use the contents of the preg structure. If, for example, you pass it to
+       regexec(), the result is undefined and your program is likely to crash.
+
+
+MATCHING NEWLINE CHARACTERS
+
+       This area is not simple, because POSIX and Perl take different views of
+       things.  It is not possible to get PCRE to obey  POSIX  semantics,  but
+       then  PCRE was never intended to be a POSIX engine. The following table
+       lists the different possibilities for matching  newline  characters  in
+       PCRE:
+
+                                 Default   Change with
+
+         . matches newline          no     PCRE_DOTALL
+         newline matches [^a]       yes    not changeable
+         $ matches \n at end        yes    PCRE_DOLLARENDONLY
+         $ matches \n in middle     no     PCRE_MULTILINE
+         ^ matches \n in middle     no     PCRE_MULTILINE
+
+       This is the equivalent table for POSIX:
+
+                                 Default   Change with
+
+         . matches newline          yes    REG_NEWLINE
+         newline matches [^a]       yes    REG_NEWLINE
+         $ matches \n at end        no     REG_NEWLINE
+         $ matches \n in middle     no     REG_NEWLINE
+         ^ matches \n in middle     no     REG_NEWLINE
+
+       PCRE's behaviour is the same as Perl's, except that there is no equiva-
+       lent for PCRE_DOLLAR_ENDONLY in Perl. In both PCRE and Perl,  there  is
+       no way to stop newline from matching [^a].
+
+       The   default  POSIX  newline  handling  can  be  obtained  by  setting
+       PCRE_DOTALL and PCRE_DOLLAR_ENDONLY, but there is no way to  make  PCRE
+       behave exactly as for the REG_NEWLINE action.
+
+
+MATCHING A PATTERN
+
+       The  function  regexec()  is  called  to  match a compiled pattern preg
+       against a given string, which is by default terminated by a  zero  byte
+       (but  see  REG_STARTEND below), subject to the options in eflags. These
+       can be:
+
+         REG_NOTBOL
+
+       The PCRE_NOTBOL option is set when calling the underlying PCRE matching
+       function.
+
+         REG_NOTEMPTY
+
+       The PCRE_NOTEMPTY option is set when calling the underlying PCRE match-
+       ing function. Note that REG_NOTEMPTY is not part of the POSIX standard.
+       However, setting this option can give more POSIX-like behaviour in some
+       situations.
+
+         REG_NOTEOL
+
+       The PCRE_NOTEOL option is set when calling the underlying PCRE matching
+       function.
+
+         REG_STARTEND
+
+       The  string  is  considered to start at string + pmatch[0].rm_so and to
+       have a terminating NUL located at string + pmatch[0].rm_eo (there  need
+       not  actually  be  a  NUL at that location), regardless of the value of
+       nmatch. This is a BSD extension, compatible with but not  specified  by
+       IEEE  Standard  1003.2  (POSIX.2),  and  should be used with caution in
+       software intended to be portable to other systems. Note that a non-zero
+       rm_so does not imply REG_NOTBOL; REG_STARTEND affects only the location
+       of the string, not how it is matched.
+
+       If the pattern was compiled with the REG_NOSUB flag, no data about  any
+       matched  strings  is  returned.  The  nmatch  and  pmatch  arguments of
+       regexec() are ignored.
+
+       If the value of nmatch is zero, or if the value pmatch is NULL, no data
+       about any matched strings is returned.
+
+       Otherwise,the portion of the string that was matched, and also any cap-
+       tured substrings, are returned via the pmatch argument, which points to
+       an  array  of nmatch structures of type regmatch_t, containing the mem-
+       bers rm_so and rm_eo. These contain the offset to the  first  character
+       of  each  substring and the offset to the first character after the end
+       of each substring, respectively. The 0th element of the vector  relates
+       to  the  entire portion of string that was matched; subsequent elements
+       relate to the capturing subpatterns of the regular  expression.  Unused
+       entries in the array have both structure members set to -1.
+
+       A  successful  match  yields  a  zero  return;  various error codes are
+       defined in the header file, of  which  REG_NOMATCH  is  the  "expected"
+       failure code.
+
+
+ERROR MESSAGES
+
+       The regerror() function maps a non-zero errorcode from either regcomp()
+       or regexec() to a printable message. If preg is  not  NULL,  the  error
+       should have arisen from the use of that structure. A message terminated
+       by a binary zero is placed  in  errbuf.  The  length  of  the  message,
+       including  the  zero, is limited to errbuf_size. The yield of the func-
+       tion is the size of buffer needed to hold the whole message.
+
+
+MEMORY USAGE
+
+       Compiling a regular expression causes memory to be allocated and  asso-
+       ciated  with  the preg structure. The function regfree() frees all such
+       memory, after which preg may no longer be used as  a  compiled  expres-
+       sion.
+
+
+AUTHOR
+
+       Philip Hazel
+       University Computing Service
+       Cambridge CB2 3QH, England.
+
+
+REVISION
+
+       Last updated: 09 January 2012
+       Copyright (c) 1997-2012 University of Cambridge.
+------------------------------------------------------------------------------
+
+
+PCRECPP(3)                 Library Functions Manual                 PCRECPP(3)
+
+
+
+NAME
+       PCRE - Perl-compatible regular expressions.
+
+SYNOPSIS OF C++ WRAPPER
+
+       #include 
+
+
+DESCRIPTION
+
+       The  C++  wrapper  for PCRE was provided by Google Inc. Some additional
+       functionality was added by Giuseppe Maxia. This brief man page was con-
+       structed  from  the  notes  in the pcrecpp.h file, which should be con-
+       sulted for further details. Note that the C++ wrapper supports only the
+       original  8-bit  PCRE  library. There is no 16-bit or 32-bit support at
+       present.
+
+
+MATCHING INTERFACE
+
+       The "FullMatch" operation checks that supplied text matches a  supplied
+       pattern  exactly.  If pointer arguments are supplied, it copies matched
+       sub-strings that match sub-patterns into them.
+
+         Example: successful match
+            pcrecpp::RE re("h.*o");
+            re.FullMatch("hello");
+
+         Example: unsuccessful match (requires full match):
+            pcrecpp::RE re("e");
+            !re.FullMatch("hello");
+
+         Example: creating a temporary RE object:
+            pcrecpp::RE("h.*o").FullMatch("hello");
+
+       You can pass in a "const char*" or a "string" for "text". The  examples
+       below  tend to use a const char*. You can, as in the different examples
+       above, store the RE object explicitly in a variable or use a  temporary
+       RE  object.  The  examples below use one mode or the other arbitrarily.
+       Either could correctly be used for any of these examples.
+
+       You must supply extra pointer arguments to extract matched subpieces.
+
+         Example: extracts "ruby" into "s" and 1234 into "i"
+            int i;
+            string s;
+            pcrecpp::RE re("(\\w+):(\\d+)");
+            re.FullMatch("ruby:1234", &s, &i);
+
+         Example: does not try to extract any extra sub-patterns
+            re.FullMatch("ruby:1234", &s);
+
+         Example: does not try to extract into NULL
+            re.FullMatch("ruby:1234", NULL, &i);
+
+         Example: integer overflow causes failure
+            !re.FullMatch("ruby:1234567891234", NULL, &i);
+
+         Example: fails because there aren't enough sub-patterns:
+            !pcrecpp::RE("\\w+:\\d+").FullMatch("ruby:1234", &s);
+
+         Example: fails because string cannot be stored in integer
+            !pcrecpp::RE("(.*)").FullMatch("ruby", &i);
+
+       The provided pointer arguments can be pointers to  any  scalar  numeric
+       type, or one of:
+
+          string        (matched piece is copied to string)
+          StringPiece   (StringPiece is mutated to point to matched piece)
+          T             (where "bool T::ParseFrom(const char*, int)" exists)
+          NULL          (the corresponding matched sub-pattern is not copied)
+
+       The  function returns true iff all of the following conditions are sat-
+       isfied:
+
+         a. "text" matches "pattern" exactly;
+
+         b. The number of matched sub-patterns is >= number of supplied
+            pointers;
+
+         c. The "i"th argument has a suitable type for holding the
+            string captured as the "i"th sub-pattern. If you pass in
+            void * NULL for the "i"th argument, or a non-void * NULL
+            of the correct type, or pass fewer arguments than the
+            number of sub-patterns, "i"th captured sub-pattern is
+            ignored.
+
+       CAVEAT: An optional sub-pattern that does  not  exist  in  the  matched
+       string  is  assigned  the  empty  string. Therefore, the following will
+       return false (because the empty string is not a valid number):
+
+          int number;
+          pcrecpp::RE::FullMatch("abc", "[a-z]+(\\d+)?", &number);
+
+       The matching interface supports at most 16 arguments per call.  If  you
+       need    more,    consider    using    the    more   general   interface
+       pcrecpp::RE::DoMatch. See pcrecpp.h for the signature for DoMatch.
+
+       NOTE: Do not use no_arg, which is used internally to mark the end of  a
+       list  of optional arguments, as a placeholder for missing arguments, as
+       this can lead to segfaults.
+
+
+QUOTING METACHARACTERS
+
+       You can use the "QuoteMeta" operation to insert backslashes before  all
+       potentially  meaningful  characters  in  a string. The returned string,
+       used as a regular expression, will exactly match the original string.
+
+         Example:
+            string quoted = RE::QuoteMeta(unquoted);
+
+       Note that it's legal to escape a character even if it  has  no  special
+       meaning  in  a  regular expression -- so this function does that. (This
+       also makes it identical to the perl function  of  the  same  name;  see
+       "perldoc    -f    quotemeta".)    For   example,   "1.5-2.0?"   becomes
+       "1\.5\-2\.0\?".
+
+
+PARTIAL MATCHES
+
+       You can use the "PartialMatch" operation when you want the  pattern  to
+       match any substring of the text.
+
+         Example: simple search for a string:
+            pcrecpp::RE("ell").PartialMatch("hello");
+
+         Example: find first number in a string:
+            int number;
+            pcrecpp::RE re("(\\d+)");
+            re.PartialMatch("x*100 + 20", &number);
+            assert(number == 100);
+
+
+UTF-8 AND THE MATCHING INTERFACE
+
+       By  default,  pattern  and text are plain text, one byte per character.
+       The UTF8 flag, passed to  the  constructor,  causes  both  pattern  and
+       string to be treated as UTF-8 text, still a byte stream but potentially
+       multiple bytes per character. In practice, the text is likelier  to  be
+       UTF-8  than  the pattern, but the match returned may depend on the UTF8
+       flag, so always use it when matching UTF8 text. For example,  "."  will
+       match  one  byte normally but with UTF8 set may match up to three bytes
+       of a multi-byte character.
+
+         Example:
+            pcrecpp::RE_Options options;
+            options.set_utf8();
+            pcrecpp::RE re(utf8_pattern, options);
+            re.FullMatch(utf8_string);
+
+         Example: using the convenience function UTF8():
+            pcrecpp::RE re(utf8_pattern, pcrecpp::UTF8());
+            re.FullMatch(utf8_string);
+
+       NOTE: The UTF8 flag is ignored if pcre was not configured with the
+             --enable-utf8 flag.
+
+
+PASSING MODIFIERS TO THE REGULAR EXPRESSION ENGINE
+
+       PCRE defines some modifiers to  change  the  behavior  of  the  regular
+       expression   engine.  The  C++  wrapper  defines  an  auxiliary  class,
+       RE_Options, as a vehicle to pass such modifiers to  a  RE  class.  Cur-
+       rently, the following modifiers are supported:
+
+          modifier              description               Perl corresponding
+
+          PCRE_CASELESS         case insensitive match      /i
+          PCRE_MULTILINE        multiple lines match        /m
+          PCRE_DOTALL           dot matches newlines        /s
+          PCRE_DOLLAR_ENDONLY   $ matches only at end       N/A
+          PCRE_EXTRA            strict escape parsing       N/A
+          PCRE_EXTENDED         ignore white spaces         /x
+          PCRE_UTF8             handles UTF8 chars          built-in
+          PCRE_UNGREEDY         reverses * and *?           N/A
+          PCRE_NO_AUTO_CAPTURE  disables capturing parens   N/A (*)
+
+       (*)  Both Perl and PCRE allow non capturing parentheses by means of the
+       "?:" modifier within the pattern itself. e.g. (?:ab|cd) does  not  cap-
+       ture, while (ab|cd) does.
+
+       For  a  full  account on how each modifier works, please check the PCRE
+       API reference page.
+
+       For each modifier, there are two member functions whose  name  is  made
+       out  of  the  modifier  in  lowercase,  without the "PCRE_" prefix. For
+       instance, PCRE_CASELESS is handled by
+
+         bool caseless()
+
+       which returns true if the modifier is set, and
+
+         RE_Options & set_caseless(bool)
+
+       which sets or unsets the modifier. Moreover, PCRE_EXTRA_MATCH_LIMIT can
+       be  accessed  through  the  set_match_limit()  and match_limit() member
+       functions. Setting match_limit to a non-zero value will limit the  exe-
+       cution  of pcre to keep it from doing bad things like blowing the stack
+       or taking an eternity to return a result.  A  value  of  5000  is  good
+       enough  to stop stack blowup in a 2MB thread stack. Setting match_limit
+       to  zero  disables  match  limiting.  Alternatively,   you   can   call
+       match_limit_recursion()  which uses PCRE_EXTRA_MATCH_LIMIT_RECURSION to
+       limit how much  PCRE  recurses.  match_limit()  limits  the  number  of
+       matches PCRE does; match_limit_recursion() limits the depth of internal
+       recursion, and therefore the amount of stack that is used.
+
+       Normally, to pass one or more modifiers to a RE class,  you  declare  a
+       RE_Options object, set the appropriate options, and pass this object to
+       a RE constructor. Example:
+
+          RE_Options opt;
+          opt.set_caseless(true);
+          if (RE("HELLO", opt).PartialMatch("hello world")) ...
+
+       RE_options has two constructors. The default constructor takes no argu-
+       ments  and creates a set of flags that are off by default. The optional
+       parameter option_flags is to facilitate transfer of legacy code from  C
+       programs.  This lets you do
+
+          RE(pattern,
+            RE_Options(PCRE_CASELESS|PCRE_MULTILINE)).PartialMatch(str);
+
+       However, new code is better off doing
+
+          RE(pattern,
+            RE_Options().set_caseless(true).set_multiline(true))
+              .PartialMatch(str);
+
+       If you are going to pass one of the most used modifiers, there are some
+       convenience functions that return a RE_Options class with the appropri-
+       ate  modifier  already  set: CASELESS(), UTF8(), MULTILINE(), DOTALL(),
+       and EXTENDED().
+
+       If you need to set several options at once, and you don't  want  to  go
+       through  the pains of declaring a RE_Options object and setting several
+       options, there is a parallel method that give you such ability  on  the
+       fly.  You  can  concatenate several set_xxxxx() member functions, since
+       each of them returns a reference to its class object. For  example,  to
+       pass  PCRE_CASELESS, PCRE_EXTENDED, and PCRE_MULTILINE to a RE with one
+       statement, you may write:
+
+          RE(" ^ xyz \\s+ .* blah$",
+            RE_Options()
+              .set_caseless(true)
+              .set_extended(true)
+              .set_multiline(true)).PartialMatch(sometext);
+
+
+SCANNING TEXT INCREMENTALLY
+
+       The "Consume" operation may be useful if you want to  repeatedly  match
+       regular expressions at the front of a string and skip over them as they
+       match. This requires use of the "StringPiece" type, which represents  a
+       sub-range  of  a  real  string.  Like RE, StringPiece is defined in the
+       pcrecpp namespace.
+
+         Example: read lines of the form "var = value" from a string.
+            string contents = ...;                 // Fill string somehow
+            pcrecpp::StringPiece input(contents);  // Wrap in a StringPiece
+
+            string var;
+            int value;
+            pcrecpp::RE re("(\\w+) = (\\d+)\n");
+            while (re.Consume(&input, &var, &value)) {
+              ...;
+            }
+
+       Each successful call  to  "Consume"  will  set  "var/value",  and  also
+       advance "input" so it points past the matched text.
+
+       The  "FindAndConsume"  operation  is  similar to "Consume" but does not
+       anchor your match at the beginning of  the  string.  For  example,  you
+       could extract all words from a string by repeatedly calling
+
+         pcrecpp::RE("(\\w+)").FindAndConsume(&input, &word)
+
+
+PARSING HEX/OCTAL/C-RADIX NUMBERS
+
+       By default, if you pass a pointer to a numeric value, the corresponding
+       text is interpreted as a base-10  number.  You  can  instead  wrap  the
+       pointer with a call to one of the operators Hex(), Octal(), or CRadix()
+       to interpret the text in another base. The CRadix  operator  interprets
+       C-style  "0"  (base-8)  and  "0x"  (base-16)  prefixes, but defaults to
+       base-10.
+
+         Example:
+           int a, b, c, d;
+           pcrecpp::RE re("(.*) (.*) (.*) (.*)");
+           re.FullMatch("100 40 0100 0x40",
+                        pcrecpp::Octal(&a), pcrecpp::Hex(&b),
+                        pcrecpp::CRadix(&c), pcrecpp::CRadix(&d));
+
+       will leave 64 in a, b, c, and d.
+
+
+REPLACING PARTS OF STRINGS
+
+       You can replace the first match of "pattern" in "str"  with  "rewrite".
+       Within  "rewrite",  backslash-escaped  digits (\1 to \9) can be used to
+       insert text matching corresponding parenthesized group  from  the  pat-
+       tern. \0 in "rewrite" refers to the entire matching text. For example:
+
+         string s = "yabba dabba doo";
+         pcrecpp::RE("b+").Replace("d", &s);
+
+       will  leave  "s" containing "yada dabba doo". The result is true if the
+       pattern matches and a replacement occurs, false otherwise.
+
+       GlobalReplace is like Replace except that it replaces  all  occurrences
+       of  the  pattern  in  the string with the rewrite. Replacements are not
+       subject to re-matching. For example:
+
+         string s = "yabba dabba doo";
+         pcrecpp::RE("b+").GlobalReplace("d", &s);
+
+       will leave "s" containing "yada dada doo". It  returns  the  number  of
+       replacements made.
+
+       Extract  is like Replace, except that if the pattern matches, "rewrite"
+       is copied into "out" (an additional argument) with substitutions.   The
+       non-matching  portions  of "text" are ignored. Returns true iff a match
+       occurred and the extraction happened successfully;  if no match occurs,
+       the string is left unaffected.
+
+
+AUTHOR
+
+       The C++ wrapper was contributed by Google Inc.
+       Copyright (c) 2007 Google Inc.
+
+
+REVISION
+
+       Last updated: 08 January 2012
+------------------------------------------------------------------------------
+
+
+PCRESAMPLE(3)              Library Functions Manual              PCRESAMPLE(3)
+
+
+
+NAME
+       PCRE - Perl-compatible regular expressions
+
+PCRE SAMPLE PROGRAM
+
+       A simple, complete demonstration program, to get you started with using
+       PCRE, is supplied in the file pcredemo.c in the  PCRE  distribution.  A
+       listing  of this program is given in the pcredemo documentation. If you
+       do not have a copy of the PCRE distribution, you can save this  listing
+       to re-create pcredemo.c.
+
+       The  demonstration program, which uses the original PCRE 8-bit library,
+       compiles the regular expression that is its first argument, and matches
+       it  against  the subject string in its second argument. No PCRE options
+       are set, and default character tables are used. If  matching  succeeds,
+       the  program  outputs the portion of the subject that matched, together
+       with the contents of any captured substrings.
+
+       If the -g option is given on the command line, the program then goes on
+       to check for further matches of the same regular expression in the same
+       subject string. The logic is a little bit tricky because of the  possi-
+       bility  of  matching an empty string. Comments in the code explain what
+       is going on.
+
+       If PCRE is installed in the standard include  and  library  directories
+       for your operating system, you should be able to compile the demonstra-
+       tion program using this command:
+
+         gcc -o pcredemo pcredemo.c -lpcre
+
+       If PCRE is installed elsewhere, you may need to add additional  options
+       to  the  command line. For example, on a Unix-like system that has PCRE
+       installed in /usr/local, you  can  compile  the  demonstration  program
+       using a command like this:
+
+         gcc -o pcredemo -I/usr/local/include pcredemo.c \
+             -L/usr/local/lib -lpcre
+
+       In  a  Windows  environment, if you want to statically link the program
+       against a non-dll pcre.a file, you must uncomment the line that defines
+       PCRE_STATIC  before  including  pcre.h, because otherwise the pcre_mal-
+       loc()   and   pcre_free()   exported   functions   will   be   declared
+       __declspec(dllimport), with unwanted results.
+
+       Once  you  have  compiled and linked the demonstration program, you can
+       run simple tests like this:
+
+         ./pcredemo 'cat|dog' 'the cat sat on the mat'
+         ./pcredemo -g 'cat|dog' 'the dog sat on the cat'
+
+       Note that there is a  much  more  comprehensive  test  program,  called
+       pcretest,  which  supports  many  more  facilities  for testing regular
+       expressions and both PCRE libraries. The pcredemo program  is  provided
+       as a simple coding example.
+
+       If  you  try to run pcredemo when PCRE is not installed in the standard
+       library directory, you may get an error like  this  on  some  operating
+       systems (e.g. Solaris):
+
+         ld.so.1:  a.out:  fatal:  libpcre.so.0:  open failed: No such file or
+       directory
+
+       This is caused by the way shared library support works  on  those  sys-
+       tems. You need to add
+
+         -R/usr/local/lib
+
+       (for example) to the compile command to get round this problem.
+
+
+AUTHOR
+
+       Philip Hazel
+       University Computing Service
+       Cambridge CB2 3QH, England.
+
+
+REVISION
+
+       Last updated: 10 January 2012
+       Copyright (c) 1997-2012 University of Cambridge.
+------------------------------------------------------------------------------
+PCRELIMITS(3)              Library Functions Manual              PCRELIMITS(3)
+
+
+
+NAME
+       PCRE - Perl-compatible regular expressions
+
+SIZE AND OTHER LIMITATIONS
+
+       There  are some size limitations in PCRE but it is hoped that they will
+       never in practice be relevant.
+
+       The maximum length of a compiled  pattern  is  approximately  64K  data
+       units  (bytes  for  the  8-bit  library,  32-bit  units  for the 32-bit
+       library, and 32-bit units for the 32-bit library) if PCRE  is  compiled
+       with  the  default  internal  linkage  size  of 2 bytes. If you want to
+       process regular expressions that are truly enormous,  you  can  compile
+       PCRE  with an internal linkage size of 3 or 4 (when building the 16-bit
+       or 32-bit library, 3 is rounded up to 4). See the README  file  in  the
+       source  distribution  and  the  pcrebuild documentation for details. In
+       these cases the limit is substantially larger.  However, the  speed  of
+       execution is slower.
+
+       All values in repeating quantifiers must be less than 65536.
+
+       There is no limit to the number of parenthesized subpatterns, but there
+       can be no more than 65535 capturing subpatterns.
+
+       There is a limit to the number of forward references to subsequent sub-
+       patterns  of  around  200,000.  Repeated  forward references with fixed
+       upper limits, for example, (?2){0,100} when subpattern number 2  is  to
+       the  right,  are included in the count. There is no limit to the number
+       of backward references.
+
+       The maximum length of name for a named subpattern is 32 characters, and
+       the maximum number of named subpatterns is 10000.
+
+       The  maximum  length  of  a  name  in  a (*MARK), (*PRUNE), (*SKIP), or
+       (*THEN) verb is 255 for the 8-bit library and 65535 for the 16-bit  and
+       32-bit library.
+
+       The  maximum  length of a subject string is the largest positive number
+       that an integer variable can hold. However, when using the  traditional
+       matching function, PCRE uses recursion to handle subpatterns and indef-
+       inite repetition.  This means that the available stack space may  limit
+       the size of a subject string that can be processed by certain patterns.
+       For a discussion of stack issues, see the pcrestack documentation.
+
+
+AUTHOR
+
+       Philip Hazel
+       University Computing Service
+       Cambridge CB2 3QH, England.
+
+
+REVISION
+
+       Last updated: 04 May 2012
+       Copyright (c) 1997-2012 University of Cambridge.
+------------------------------------------------------------------------------
+
+
+PCRESTACK(3)               Library Functions Manual               PCRESTACK(3)
+
+
+
+NAME
+       PCRE - Perl-compatible regular expressions
+
+PCRE DISCUSSION OF STACK USAGE
+
+       When  you call pcre[16|32]_exec(), it makes use of an internal function
+       called match(). This calls itself recursively at branch points  in  the
+       pattern,  in  order  to  remember the state of the match so that it can
+       back up and try a different alternative if  the  first  one  fails.  As
+       matching proceeds deeper and deeper into the tree of possibilities, the
+       recursion depth increases. The match() function is also called in other
+       circumstances,  for  example,  whenever  a parenthesized sub-pattern is
+       entered, and in certain cases of repetition.
+
+       Not all calls of match() increase the recursion depth; for an item such
+       as  a* it may be called several times at the same level, after matching
+       different numbers of a's. Furthermore, in a number of cases  where  the
+       result  of  the  recursive call would immediately be passed back as the
+       result of the current call (a "tail recursion"), the function  is  just
+       restarted instead.
+
+       The  above  comments apply when pcre[16|32]_exec() is run in its normal
+       interpretive  manner.   If   the   pattern   was   studied   with   the
+       PCRE_STUDY_JIT_COMPILE  option, and just-in-time compiling was success-
+       ful, and the options passed to pcre[16|32]_exec() were  not  incompati-
+       ble,  the  matching  process  uses the JIT-compiled code instead of the
+       match() function. In this case, the  memory  requirements  are  handled
+       entirely differently. See the pcrejit documentation for details.
+
+       The  pcre[16|32]_dfa_exec()  function operates in an entirely different
+       way, and uses recursion only when there is a regular expression  recur-
+       sion or subroutine call in the pattern. This includes the processing of
+       assertion and "once-only" subpatterns, which are handled  like  subrou-
+       tine  calls.  Normally, these are never very deep, and the limit on the
+       complexity of pcre[16|32]_dfa_exec() is controlled  by  the  amount  of
+       workspace  it is given.  However, it is possible to write patterns with
+       runaway    infinite    recursions;    such    patterns    will    cause
+       pcre[16|32]_dfa_exec()  to  run  out  of stack. At present, there is no
+       protection against this.
+
+       The comments that follow do NOT apply to  pcre[16|32]_dfa_exec();  they
+       are relevant only for pcre[16|32]_exec() without the JIT optimization.
+
+   Reducing pcre[16|32]_exec()'s stack usage
+
+       Each  time  that match() is actually called recursively, it uses memory
+       from the process stack. For certain kinds of  pattern  and  data,  very
+       large  amounts of stack may be needed, despite the recognition of "tail
+       recursion".  You can often reduce the amount of recursion,  and  there-
+       fore  the  amount of stack used, by modifying the pattern that is being
+       matched. Consider, for example, this pattern:
+
+         ([^<]|<(?!inet))+
+
+       It matches from wherever it starts until it encounters "
+.
+.
+.SH "PCRE 16-BIT API BASIC FUNCTIONS"
+.rs
+.sp
+.SM
+.B pcre16 *pcre16_compile(PCRE_SPTR16 \fIpattern\fP, int \fIoptions\fP,
+.ti +5n
+.B const char **\fIerrptr\fP, int *\fIerroffset\fP,
+.ti +5n
+.B const unsigned char *\fItableptr\fP);
+.PP
+.B pcre16 *pcre16_compile2(PCRE_SPTR16 \fIpattern\fP, int \fIoptions\fP,
+.ti +5n
+.B int *\fIerrorcodeptr\fP,
+.ti +5n
+.B const char **\fIerrptr\fP, int *\fIerroffset\fP,
+.ti +5n
+.B const unsigned char *\fItableptr\fP);
+.PP
+.B pcre16_extra *pcre16_study(const pcre16 *\fIcode\fP, int \fIoptions\fP,
+.ti +5n
+.B const char **\fIerrptr\fP);
+.PP
+.B void pcre16_free_study(pcre16_extra *\fIextra\fP);
+.PP
+.B int pcre16_exec(const pcre16 *\fIcode\fP, "const pcre16_extra *\fIextra\fP,"
+.ti +5n
+.B "PCRE_SPTR16 \fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
+.ti +5n
+.B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);
+.PP
+.B int pcre16_dfa_exec(const pcre16 *\fIcode\fP, "const pcre16_extra *\fIextra\fP,"
+.ti +5n
+.B "PCRE_SPTR16 \fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
+.ti +5n
+.B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
+.ti +5n
+.B int *\fIworkspace\fP, int \fIwscount\fP);
+.
+.
+.SH "PCRE 16-BIT API STRING EXTRACTION FUNCTIONS"
+.rs
+.sp
+.B int pcre16_copy_named_substring(const pcre16 *\fIcode\fP,
+.ti +5n
+.B PCRE_SPTR16 \fIsubject\fP, int *\fIovector\fP,
+.ti +5n
+.B int \fIstringcount\fP, PCRE_SPTR16 \fIstringname\fP,
+.ti +5n
+.B PCRE_UCHAR16 *\fIbuffer\fP, int \fIbuffersize\fP);
+.PP
+.B int pcre16_copy_substring(PCRE_SPTR16 \fIsubject\fP, int *\fIovector\fP,
+.ti +5n
+.B int \fIstringcount\fP, int \fIstringnumber\fP, PCRE_UCHAR16 *\fIbuffer\fP,
+.ti +5n
+.B int \fIbuffersize\fP);
+.PP
+.B int pcre16_get_named_substring(const pcre16 *\fIcode\fP,
+.ti +5n
+.B PCRE_SPTR16 \fIsubject\fP, int *\fIovector\fP,
+.ti +5n
+.B int \fIstringcount\fP, PCRE_SPTR16 \fIstringname\fP,
+.ti +5n
+.B PCRE_SPTR16 *\fIstringptr\fP);
+.PP
+.B int pcre16_get_stringnumber(const pcre16 *\fIcode\fP,
+.ti +5n
+.B PCRE_SPTR16 \fIname\fP);
+.PP
+.B int pcre16_get_stringtable_entries(const pcre16 *\fIcode\fP,
+.ti +5n
+.B PCRE_SPTR16 \fIname\fP, PCRE_UCHAR16 **\fIfirst\fP, PCRE_UCHAR16 **\fIlast\fP);
+.PP
+.B int pcre16_get_substring(PCRE_SPTR16 \fIsubject\fP, int *\fIovector\fP,
+.ti +5n
+.B int \fIstringcount\fP, int \fIstringnumber\fP,
+.ti +5n
+.B PCRE_SPTR16 *\fIstringptr\fP);
+.PP
+.B int pcre16_get_substring_list(PCRE_SPTR16 \fIsubject\fP,
+.ti +5n
+.B int *\fIovector\fP, int \fIstringcount\fP, "PCRE_SPTR16 **\fIlistptr\fP);"
+.PP
+.B void pcre16_free_substring(PCRE_SPTR16 \fIstringptr\fP);
+.PP
+.B void pcre16_free_substring_list(PCRE_SPTR16 *\fIstringptr\fP);
+.
+.
+.SH "PCRE 16-BIT API AUXILIARY FUNCTIONS"
+.rs
+.sp
+.B pcre16_jit_stack *pcre16_jit_stack_alloc(int \fIstartsize\fP, int \fImaxsize\fP);
+.PP
+.B void pcre16_jit_stack_free(pcre16_jit_stack *\fIstack\fP);
+.PP
+.B void pcre16_assign_jit_stack(pcre16_extra *\fIextra\fP,
+.ti +5n
+.B pcre16_jit_callback \fIcallback\fP, void *\fIdata\fP);
+.PP
+.B const unsigned char *pcre16_maketables(void);
+.PP
+.B int pcre16_fullinfo(const pcre16 *\fIcode\fP, "const pcre16_extra *\fIextra\fP,"
+.ti +5n
+.B int \fIwhat\fP, void *\fIwhere\fP);
+.PP
+.B int pcre16_refcount(pcre16 *\fIcode\fP, int \fIadjust\fP);
+.PP
+.B int pcre16_config(int \fIwhat\fP, void *\fIwhere\fP);
+.PP
+.B const char *pcre16_version(void);
+.PP
+.B int pcre16_pattern_to_host_byte_order(pcre16 *\fIcode\fP,
+.ti +5n
+.B pcre16_extra *\fIextra\fP, const unsigned char *\fItables\fP);
+.
+.
+.SH "PCRE 16-BIT API INDIRECTED FUNCTIONS"
+.rs
+.sp
+.B void *(*pcre16_malloc)(size_t);
+.PP
+.B void (*pcre16_free)(void *);
+.PP
+.B void *(*pcre16_stack_malloc)(size_t);
+.PP
+.B void (*pcre16_stack_free)(void *);
+.PP
+.B int (*pcre16_callout)(pcre16_callout_block *);
+.
+.
+.SH "PCRE 16-BIT API 16-BIT-ONLY FUNCTION"
+.rs
+.sp
+.B int pcre16_utf16_to_host_byte_order(PCRE_UCHAR16 *\fIoutput\fP,
+.ti +5n
+.B PCRE_SPTR16 \fIinput\fP, int \fIlength\fP, int *\fIbyte_order\fP,
+.ti +5n
+.B int \fIkeep_boms\fP);
+.
+.
+.SH "THE PCRE 16-BIT LIBRARY"
+.rs
+.sp
+Starting with release 8.30, it is possible to compile a PCRE library that
+supports 16-bit character strings, including UTF-16 strings, as well as or
+instead of the original 8-bit library. The majority of the work to make this
+possible was done by Zoltan Herczeg. The two libraries contain identical sets
+of functions, used in exactly the same way. Only the names of the functions and
+the data types of their arguments and results are different. To avoid
+over-complication and reduce the documentation maintenance load, most of the
+PCRE documentation describes the 8-bit library, with only occasional references
+to the 16-bit library. This page describes what is different when you use the
+16-bit library.
+.P
+WARNING: A single application can be linked with both libraries, but you must
+take care when processing any particular pattern to use functions from just one
+library. For example, if you want to study a pattern that was compiled with
+\fBpcre16_compile()\fP, you must do so with \fBpcre16_study()\fP, not
+\fBpcre_study()\fP, and you must free the study data with
+\fBpcre16_free_study()\fP.
+.
+.
+.SH "THE HEADER FILE"
+.rs
+.sp
+There is only one header file, \fBpcre.h\fP. It contains prototypes for all the
+functions in all libraries, as well as definitions of flags, structures, error
+codes, etc.
+.
+.
+.SH "THE LIBRARY NAME"
+.rs
+.sp
+In Unix-like systems, the 16-bit library is called \fBlibpcre16\fP, and can
+normally be accesss by adding \fB-lpcre16\fP to the command for linking an
+application that uses PCRE.
+.
+.
+.SH "STRING TYPES"
+.rs
+.sp
+In the 8-bit library, strings are passed to PCRE library functions as vectors
+of bytes with the C type "char *". In the 16-bit library, strings are passed as
+vectors of unsigned 16-bit quantities. The macro PCRE_UCHAR16 specifies an
+appropriate data type, and PCRE_SPTR16 is defined as "const PCRE_UCHAR16 *". In
+very many environments, "short int" is a 16-bit data type. When PCRE is built,
+it defines PCRE_UCHAR16 as "unsigned short int", but checks that it really is a
+16-bit data type. If it is not, the build fails with an error message telling
+the maintainer to modify the definition appropriately.
+.
+.
+.SH "STRUCTURE TYPES"
+.rs
+.sp
+The types of the opaque structures that are used for compiled 16-bit patterns
+and JIT stacks are \fBpcre16\fP and \fBpcre16_jit_stack\fP respectively. The
+type of the user-accessible structure that is returned by \fBpcre16_study()\fP
+is \fBpcre16_extra\fP, and the type of the structure that is used for passing
+data to a callout function is \fBpcre16_callout_block\fP. These structures
+contain the same fields, with the same names, as their 8-bit counterparts. The
+only difference is that pointers to character strings are 16-bit instead of
+8-bit types.
+.
+.
+.SH "16-BIT FUNCTIONS"
+.rs
+.sp
+For every function in the 8-bit library there is a corresponding function in
+the 16-bit library with a name that starts with \fBpcre16_\fP instead of
+\fBpcre_\fP. The prototypes are listed above. In addition, there is one extra
+function, \fBpcre16_utf16_to_host_byte_order()\fP. This is a utility function
+that converts a UTF-16 character string to host byte order if necessary. The
+other 16-bit functions expect the strings they are passed to be in host byte
+order.
+.P
+The \fIinput\fP and \fIoutput\fP arguments of
+\fBpcre16_utf16_to_host_byte_order()\fP may point to the same address, that is,
+conversion in place is supported. The output buffer must be at least as long as
+the input.
+.P
+The \fIlength\fP argument specifies the number of 16-bit data units in the
+input string; a negative value specifies a zero-terminated string.
+.P
+If \fIbyte_order\fP is NULL, it is assumed that the string starts off in host
+byte order. This may be changed by byte-order marks (BOMs) anywhere in the
+string (commonly as the first character).
+.P
+If \fIbyte_order\fP is not NULL, a non-zero value of the integer to which it
+points means that the input starts off in host byte order, otherwise the
+opposite order is assumed. Again, BOMs in the string can change this. The final
+byte order is passed back at the end of processing.
+.P
+If \fIkeep_boms\fP is not zero, byte-order mark characters (0xfeff) are copied
+into the output string. Otherwise they are discarded.
+.P
+The result of the function is the number of 16-bit units placed into the output
+buffer, including the zero terminator if the string was zero-terminated.
+.
+.
+.SH "SUBJECT STRING OFFSETS"
+.rs
+.sp
+The lengths and starting offsets of subject strings must be specified in 16-bit
+data units, and the offsets within subject strings that are returned by the
+matching functions are in also 16-bit units rather than bytes.
+.
+.
+.SH "NAMED SUBPATTERNS"
+.rs
+.sp
+The name-to-number translation table that is maintained for named subpatterns
+uses 16-bit characters. The \fBpcre16_get_stringtable_entries()\fP function
+returns the length of each entry in the table as the number of 16-bit data
+units.
+.
+.
+.SH "OPTION NAMES"
+.rs
+.sp
+There are two new general option names, PCRE_UTF16 and PCRE_NO_UTF16_CHECK,
+which correspond to PCRE_UTF8 and PCRE_NO_UTF8_CHECK in the 8-bit library. In
+fact, these new options define the same bits in the options word. There is a
+discussion about the
+.\" HTML 
+.\" 
+validity of UTF-16 strings
+.\"
+in the
+.\" HREF
+\fBpcreunicode\fP
+.\"
+page.
+.P
+For the \fBpcre16_config()\fP function there is an option PCRE_CONFIG_UTF16
+that returns 1 if UTF-16 support is configured, otherwise 0. If this option is
+given to \fBpcre_config()\fP or \fBpcre32_config()\fP, or if the
+PCRE_CONFIG_UTF8 or PCRE_CONFIG_UTF32 option is given to \fBpcre16_config()\fP,
+the result is the PCRE_ERROR_BADOPTION error.
+.
+.
+.SH "CHARACTER CODES"
+.rs
+.sp
+In 16-bit mode, when PCRE_UTF16 is not set, character values are treated in the
+same way as in 8-bit, non UTF-8 mode, except, of course, that they can range
+from 0 to 0xffff instead of 0 to 0xff. Character types for characters less than
+0xff can therefore be influenced by the locale in the same way as before.
+Characters greater than 0xff have only one case, and no "type" (such as letter
+or digit).
+.P
+In UTF-16 mode, the character code is Unicode, in the range 0 to 0x10ffff, with
+the exception of values in the range 0xd800 to 0xdfff because those are
+"surrogate" values that are used in pairs to encode values greater than 0xffff.
+.P
+A UTF-16 string can indicate its endianness by special code knows as a
+byte-order mark (BOM). The PCRE functions do not handle this, expecting strings
+to be in host byte order. A utility function called
+\fBpcre16_utf16_to_host_byte_order()\fP is provided to help with this (see
+above).
+.
+.
+.SH "ERROR NAMES"
+.rs
+.sp
+The errors PCRE_ERROR_BADUTF16_OFFSET and PCRE_ERROR_SHORTUTF16 correspond to
+their 8-bit counterparts. The error PCRE_ERROR_BADMODE is given when a compiled
+pattern is passed to a function that processes patterns in the other
+mode, for example, if a pattern compiled with \fBpcre_compile()\fP is passed to
+\fBpcre16_exec()\fP.
+.P
+There are new error codes whose names begin with PCRE_UTF16_ERR for invalid
+UTF-16 strings, corresponding to the PCRE_UTF8_ERR codes for UTF-8 strings that
+are described in the section entitled
+.\" HTML 
+.\" 
+"Reason codes for invalid UTF-8 strings"
+.\"
+in the main
+.\" HREF
+\fBpcreapi\fP
+.\"
+page. The UTF-16 errors are:
+.sp
+  PCRE_UTF16_ERR1  Missing low surrogate at end of string
+  PCRE_UTF16_ERR2  Invalid low surrogate follows high surrogate
+  PCRE_UTF16_ERR3  Isolated low surrogate
+  PCRE_UTF16_ERR4  Non-character
+.
+.
+.SH "ERROR TEXTS"
+.rs
+.sp
+If there is an error while compiling a pattern, the error text that is passed
+back by \fBpcre16_compile()\fP or \fBpcre16_compile2()\fP is still an 8-bit
+character string, zero-terminated.
+.
+.
+.SH "CALLOUTS"
+.rs
+.sp
+The \fIsubject\fP and \fImark\fP fields in the callout block that is passed to
+a callout function point to 16-bit vectors.
+.
+.
+.SH "TESTING"
+.rs
+.sp
+The \fBpcretest\fP program continues to operate with 8-bit input and output
+files, but it can be used for testing the 16-bit library. If it is run with the
+command line option \fB-16\fP, patterns and subject strings are converted from
+8-bit to 16-bit before being passed to PCRE, and the 16-bit library functions
+are used instead of the 8-bit ones. Returned 16-bit strings are converted to
+8-bit for output. If both the 8-bit and the 32-bit libraries were not compiled,
+\fBpcretest\fP defaults to 16-bit and the \fB-16\fP option is ignored.
+.P
+When PCRE is being built, the \fBRunTest\fP script that is called by "make
+check" uses the \fBpcretest\fP \fB-C\fP option to discover which of the 8-bit,
+16-bit and 32-bit libraries has been built, and runs the tests appropriately.
+.
+.
+.SH "NOT SUPPORTED IN 16-BIT MODE"
+.rs
+.sp
+Not all the features of the 8-bit library are available with the 16-bit
+library. The C++ and POSIX wrapper functions support only the 8-bit library,
+and the \fBpcregrep\fP program is at present 8-bit only.
+.
+.
+.SH AUTHOR
+.rs
+.sp
+.nf
+Philip Hazel
+University Computing Service
+Cambridge CB2 3QH, England.
+.fi
+.
+.
+.SH REVISION
+.rs
+.sp
+.nf
+Last updated: 12 May 2013
+Copyright (c) 1997-2013 University of Cambridge.
+.fi
diff --git a/pcre/doc/pcre32.3 b/pcre/doc/pcre32.3
new file mode 100644
index 00000000000..516c8ee570c
--- /dev/null
+++ b/pcre/doc/pcre32.3
@@ -0,0 +1,390 @@
+.TH PCRE 3 "12 May 2013" "PCRE 8.33"
+.SH NAME
+PCRE - Perl-compatible regular expressions
+.sp
+.B #include 
+.
+.
+.SH "PCRE 32-BIT API BASIC FUNCTIONS"
+.rs
+.sp
+.SM
+.B pcre32 *pcre32_compile(PCRE_SPTR32 \fIpattern\fP, int \fIoptions\fP,
+.ti +5n
+.B const char **\fIerrptr\fP, int *\fIerroffset\fP,
+.ti +5n
+.B const unsigned char *\fItableptr\fP);
+.PP
+.B pcre32 *pcre32_compile2(PCRE_SPTR32 \fIpattern\fP, int \fIoptions\fP,
+.ti +5n
+.B int *\fIerrorcodeptr\fP,
+.ti +5n
+.B const char **\fIerrptr\fP, int *\fIerroffset\fP,
+.ti +5n
+.B const unsigned char *\fItableptr\fP);
+.PP
+.B pcre32_extra *pcre32_study(const pcre32 *\fIcode\fP, int \fIoptions\fP,
+.ti +5n
+.B const char **\fIerrptr\fP);
+.PP
+.B void pcre32_free_study(pcre32_extra *\fIextra\fP);
+.PP
+.B int pcre32_exec(const pcre32 *\fIcode\fP, "const pcre32_extra *\fIextra\fP,"
+.ti +5n
+.B "PCRE_SPTR32 \fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
+.ti +5n
+.B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);
+.PP
+.B int pcre32_dfa_exec(const pcre32 *\fIcode\fP, "const pcre32_extra *\fIextra\fP,"
+.ti +5n
+.B "PCRE_SPTR32 \fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
+.ti +5n
+.B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
+.ti +5n
+.B int *\fIworkspace\fP, int \fIwscount\fP);
+.
+.
+.SH "PCRE 32-BIT API STRING EXTRACTION FUNCTIONS"
+.rs
+.sp
+.B int pcre32_copy_named_substring(const pcre32 *\fIcode\fP,
+.ti +5n
+.B PCRE_SPTR32 \fIsubject\fP, int *\fIovector\fP,
+.ti +5n
+.B int \fIstringcount\fP, PCRE_SPTR32 \fIstringname\fP,
+.ti +5n
+.B PCRE_UCHAR32 *\fIbuffer\fP, int \fIbuffersize\fP);
+.PP
+.B int pcre32_copy_substring(PCRE_SPTR32 \fIsubject\fP, int *\fIovector\fP,
+.ti +5n
+.B int \fIstringcount\fP, int \fIstringnumber\fP, PCRE_UCHAR32 *\fIbuffer\fP,
+.ti +5n
+.B int \fIbuffersize\fP);
+.PP
+.B int pcre32_get_named_substring(const pcre32 *\fIcode\fP,
+.ti +5n
+.B PCRE_SPTR32 \fIsubject\fP, int *\fIovector\fP,
+.ti +5n
+.B int \fIstringcount\fP, PCRE_SPTR32 \fIstringname\fP,
+.ti +5n
+.B PCRE_SPTR32 *\fIstringptr\fP);
+.PP
+.B int pcre32_get_stringnumber(const pcre32 *\fIcode\fP,
+.ti +5n
+.B PCRE_SPTR32 \fIname\fP);
+.PP
+.B int pcre32_get_stringtable_entries(const pcre32 *\fIcode\fP,
+.ti +5n
+.B PCRE_SPTR32 \fIname\fP, PCRE_UCHAR32 **\fIfirst\fP, PCRE_UCHAR32 **\fIlast\fP);
+.PP
+.B int pcre32_get_substring(PCRE_SPTR32 \fIsubject\fP, int *\fIovector\fP,
+.ti +5n
+.B int \fIstringcount\fP, int \fIstringnumber\fP,
+.ti +5n
+.B PCRE_SPTR32 *\fIstringptr\fP);
+.PP
+.B int pcre32_get_substring_list(PCRE_SPTR32 \fIsubject\fP,
+.ti +5n
+.B int *\fIovector\fP, int \fIstringcount\fP, "PCRE_SPTR32 **\fIlistptr\fP);"
+.PP
+.B void pcre32_free_substring(PCRE_SPTR32 \fIstringptr\fP);
+.PP
+.B void pcre32_free_substring_list(PCRE_SPTR32 *\fIstringptr\fP);
+.
+.
+.SH "PCRE 32-BIT API AUXILIARY FUNCTIONS"
+.rs
+.sp
+.B pcre32_jit_stack *pcre32_jit_stack_alloc(int \fIstartsize\fP, int \fImaxsize\fP);
+.PP
+.B void pcre32_jit_stack_free(pcre32_jit_stack *\fIstack\fP);
+.PP
+.B void pcre32_assign_jit_stack(pcre32_extra *\fIextra\fP,
+.ti +5n
+.B pcre32_jit_callback \fIcallback\fP, void *\fIdata\fP);
+.PP
+.B const unsigned char *pcre32_maketables(void);
+.PP
+.B int pcre32_fullinfo(const pcre32 *\fIcode\fP, "const pcre32_extra *\fIextra\fP,"
+.ti +5n
+.B int \fIwhat\fP, void *\fIwhere\fP);
+.PP
+.B int pcre32_refcount(pcre32 *\fIcode\fP, int \fIadjust\fP);
+.PP
+.B int pcre32_config(int \fIwhat\fP, void *\fIwhere\fP);
+.PP
+.B const char *pcre32_version(void);
+.PP
+.B int pcre32_pattern_to_host_byte_order(pcre32 *\fIcode\fP,
+.ti +5n
+.B pcre32_extra *\fIextra\fP, const unsigned char *\fItables\fP);
+.
+.
+.SH "PCRE 32-BIT API INDIRECTED FUNCTIONS"
+.rs
+.sp
+.B void *(*pcre32_malloc)(size_t);
+.PP
+.B void (*pcre32_free)(void *);
+.PP
+.B void *(*pcre32_stack_malloc)(size_t);
+.PP
+.B void (*pcre32_stack_free)(void *);
+.PP
+.B int (*pcre32_callout)(pcre32_callout_block *);
+.
+.
+.SH "PCRE 32-BIT API 32-BIT-ONLY FUNCTION"
+.rs
+.sp
+.B int pcre32_utf32_to_host_byte_order(PCRE_UCHAR32 *\fIoutput\fP,
+.ti +5n
+.B PCRE_SPTR32 \fIinput\fP, int \fIlength\fP, int *\fIbyte_order\fP,
+.ti +5n
+.B int \fIkeep_boms\fP);
+.
+.
+.SH "THE PCRE 32-BIT LIBRARY"
+.rs
+.sp
+Starting with release 8.32, it is possible to compile a PCRE library that
+supports 32-bit character strings, including UTF-32 strings, as well as or
+instead of the original 8-bit library. This work was done by Christian Persch,
+based on the work done by Zoltan Herczeg for the 16-bit library. All three
+libraries contain identical sets of functions, used in exactly the same way.
+Only the names of the functions and the data types of their arguments and
+results are different. To avoid over-complication and reduce the documentation
+maintenance load, most of the PCRE documentation describes the 8-bit library,
+with only occasional references to the 16-bit and 32-bit libraries. This page
+describes what is different when you use the 32-bit library.
+.P
+WARNING: A single application can be linked with all or any of the three
+libraries, but you must take care when processing any particular pattern
+to use functions from just one library. For example, if you want to study
+a pattern that was compiled with \fBpcre32_compile()\fP, you must do so
+with \fBpcre32_study()\fP, not \fBpcre_study()\fP, and you must free the
+study data with \fBpcre32_free_study()\fP.
+.
+.
+.SH "THE HEADER FILE"
+.rs
+.sp
+There is only one header file, \fBpcre.h\fP. It contains prototypes for all the
+functions in all libraries, as well as definitions of flags, structures, error
+codes, etc.
+.
+.
+.SH "THE LIBRARY NAME"
+.rs
+.sp
+In Unix-like systems, the 32-bit library is called \fBlibpcre32\fP, and can
+normally be accesss by adding \fB-lpcre32\fP to the command for linking an
+application that uses PCRE.
+.
+.
+.SH "STRING TYPES"
+.rs
+.sp
+In the 8-bit library, strings are passed to PCRE library functions as vectors
+of bytes with the C type "char *". In the 32-bit library, strings are passed as
+vectors of unsigned 32-bit quantities. The macro PCRE_UCHAR32 specifies an
+appropriate data type, and PCRE_SPTR32 is defined as "const PCRE_UCHAR32 *". In
+very many environments, "unsigned int" is a 32-bit data type. When PCRE is
+built, it defines PCRE_UCHAR32 as "unsigned int", but checks that it really is
+a 32-bit data type. If it is not, the build fails with an error message telling
+the maintainer to modify the definition appropriately.
+.
+.
+.SH "STRUCTURE TYPES"
+.rs
+.sp
+The types of the opaque structures that are used for compiled 32-bit patterns
+and JIT stacks are \fBpcre32\fP and \fBpcre32_jit_stack\fP respectively. The
+type of the user-accessible structure that is returned by \fBpcre32_study()\fP
+is \fBpcre32_extra\fP, and the type of the structure that is used for passing
+data to a callout function is \fBpcre32_callout_block\fP. These structures
+contain the same fields, with the same names, as their 8-bit counterparts. The
+only difference is that pointers to character strings are 32-bit instead of
+8-bit types.
+.
+.
+.SH "32-BIT FUNCTIONS"
+.rs
+.sp
+For every function in the 8-bit library there is a corresponding function in
+the 32-bit library with a name that starts with \fBpcre32_\fP instead of
+\fBpcre_\fP. The prototypes are listed above. In addition, there is one extra
+function, \fBpcre32_utf32_to_host_byte_order()\fP. This is a utility function
+that converts a UTF-32 character string to host byte order if necessary. The
+other 32-bit functions expect the strings they are passed to be in host byte
+order.
+.P
+The \fIinput\fP and \fIoutput\fP arguments of
+\fBpcre32_utf32_to_host_byte_order()\fP may point to the same address, that is,
+conversion in place is supported. The output buffer must be at least as long as
+the input.
+.P
+The \fIlength\fP argument specifies the number of 32-bit data units in the
+input string; a negative value specifies a zero-terminated string.
+.P
+If \fIbyte_order\fP is NULL, it is assumed that the string starts off in host
+byte order. This may be changed by byte-order marks (BOMs) anywhere in the
+string (commonly as the first character).
+.P
+If \fIbyte_order\fP is not NULL, a non-zero value of the integer to which it
+points means that the input starts off in host byte order, otherwise the
+opposite order is assumed. Again, BOMs in the string can change this. The final
+byte order is passed back at the end of processing.
+.P
+If \fIkeep_boms\fP is not zero, byte-order mark characters (0xfeff) are copied
+into the output string. Otherwise they are discarded.
+.P
+The result of the function is the number of 32-bit units placed into the output
+buffer, including the zero terminator if the string was zero-terminated.
+.
+.
+.SH "SUBJECT STRING OFFSETS"
+.rs
+.sp
+The lengths and starting offsets of subject strings must be specified in 32-bit
+data units, and the offsets within subject strings that are returned by the
+matching functions are in also 32-bit units rather than bytes.
+.
+.
+.SH "NAMED SUBPATTERNS"
+.rs
+.sp
+The name-to-number translation table that is maintained for named subpatterns
+uses 32-bit characters. The \fBpcre32_get_stringtable_entries()\fP function
+returns the length of each entry in the table as the number of 32-bit data
+units.
+.
+.
+.SH "OPTION NAMES"
+.rs
+.sp
+There are two new general option names, PCRE_UTF32 and PCRE_NO_UTF32_CHECK,
+which correspond to PCRE_UTF8 and PCRE_NO_UTF8_CHECK in the 8-bit library. In
+fact, these new options define the same bits in the options word. There is a
+discussion about the
+.\" HTML 
+.\" 
+validity of UTF-32 strings
+.\"
+in the
+.\" HREF
+\fBpcreunicode\fP
+.\"
+page.
+.P
+For the \fBpcre32_config()\fP function there is an option PCRE_CONFIG_UTF32
+that returns 1 if UTF-32 support is configured, otherwise 0. If this option is
+given to \fBpcre_config()\fP or \fBpcre16_config()\fP, or if the
+PCRE_CONFIG_UTF8 or PCRE_CONFIG_UTF16 option is given to \fBpcre32_config()\fP,
+the result is the PCRE_ERROR_BADOPTION error.
+.
+.
+.SH "CHARACTER CODES"
+.rs
+.sp
+In 32-bit mode, when PCRE_UTF32 is not set, character values are treated in the
+same way as in 8-bit, non UTF-8 mode, except, of course, that they can range
+from 0 to 0x7fffffff instead of 0 to 0xff. Character types for characters less
+than 0xff can therefore be influenced by the locale in the same way as before.
+Characters greater than 0xff have only one case, and no "type" (such as letter
+or digit).
+.P
+In UTF-32 mode, the character code is Unicode, in the range 0 to 0x10ffff, with
+the exception of values in the range 0xd800 to 0xdfff because those are
+"surrogate" values that are ill-formed in UTF-32.
+.P
+A UTF-32 string can indicate its endianness by special code knows as a
+byte-order mark (BOM). The PCRE functions do not handle this, expecting strings
+to be in host byte order. A utility function called
+\fBpcre32_utf32_to_host_byte_order()\fP is provided to help with this (see
+above).
+.
+.
+.SH "ERROR NAMES"
+.rs
+.sp
+The error PCRE_ERROR_BADUTF32 corresponds to its 8-bit counterpart.
+The error PCRE_ERROR_BADMODE is given when a compiled
+pattern is passed to a function that processes patterns in the other
+mode, for example, if a pattern compiled with \fBpcre_compile()\fP is passed to
+\fBpcre32_exec()\fP.
+.P
+There are new error codes whose names begin with PCRE_UTF32_ERR for invalid
+UTF-32 strings, corresponding to the PCRE_UTF8_ERR codes for UTF-8 strings that
+are described in the section entitled
+.\" HTML 
+.\" 
+"Reason codes for invalid UTF-8 strings"
+.\"
+in the main
+.\" HREF
+\fBpcreapi\fP
+.\"
+page. The UTF-32 errors are:
+.sp
+  PCRE_UTF32_ERR1  Surrogate character (range from 0xd800 to 0xdfff)
+  PCRE_UTF32_ERR2  Non-character
+  PCRE_UTF32_ERR3  Character > 0x10ffff
+.
+.
+.SH "ERROR TEXTS"
+.rs
+.sp
+If there is an error while compiling a pattern, the error text that is passed
+back by \fBpcre32_compile()\fP or \fBpcre32_compile2()\fP is still an 8-bit
+character string, zero-terminated.
+.
+.
+.SH "CALLOUTS"
+.rs
+.sp
+The \fIsubject\fP and \fImark\fP fields in the callout block that is passed to
+a callout function point to 32-bit vectors.
+.
+.
+.SH "TESTING"
+.rs
+.sp
+The \fBpcretest\fP program continues to operate with 8-bit input and output
+files, but it can be used for testing the 32-bit library. If it is run with the
+command line option \fB-32\fP, patterns and subject strings are converted from
+8-bit to 32-bit before being passed to PCRE, and the 32-bit library functions
+are used instead of the 8-bit ones. Returned 32-bit strings are converted to
+8-bit for output. If both the 8-bit and the 16-bit libraries were not compiled,
+\fBpcretest\fP defaults to 32-bit and the \fB-32\fP option is ignored.
+.P
+When PCRE is being built, the \fBRunTest\fP script that is called by "make
+check" uses the \fBpcretest\fP \fB-C\fP option to discover which of the 8-bit,
+16-bit and 32-bit libraries has been built, and runs the tests appropriately.
+.
+.
+.SH "NOT SUPPORTED IN 32-BIT MODE"
+.rs
+.sp
+Not all the features of the 8-bit library are available with the 32-bit
+library. The C++ and POSIX wrapper functions support only the 8-bit library,
+and the \fBpcregrep\fP program is at present 8-bit only.
+.
+.
+.SH AUTHOR
+.rs
+.sp
+.nf
+Philip Hazel
+University Computing Service
+Cambridge CB2 3QH, England.
+.fi
+.
+.
+.SH REVISION
+.rs
+.sp
+.nf
+Last updated: 12 May 2013
+Copyright (c) 1997-2013 University of Cambridge.
+.fi
diff --git a/pcre/doc/pcre_assign_jit_stack.3 b/pcre/doc/pcre_assign_jit_stack.3
new file mode 100644
index 00000000000..e1563b6e06d
--- /dev/null
+++ b/pcre/doc/pcre_assign_jit_stack.3
@@ -0,0 +1,61 @@
+.TH PCRE_ASSIGN_JIT_STACK 3 "24 June 2012" "PCRE 8.30"
+.SH NAME
+PCRE - Perl-compatible regular expressions
+.SH SYNOPSIS
+.rs
+.sp
+.B #include 
+.PP
+.SM
+.B void pcre_assign_jit_stack(pcre_extra *\fIextra\fP,
+.ti +5n
+.B pcre_jit_callback \fIcallback\fP, void *\fIdata\fP);
+.PP
+.B void pcre16_assign_jit_stack(pcre16_extra *\fIextra\fP,
+.ti +5n
+.B pcre16_jit_callback \fIcallback\fP, void *\fIdata\fP);
+.PP
+.B void pcre32_assign_jit_stack(pcre32_extra *\fIextra\fP,
+.ti +5n
+.B pcre32_jit_callback \fIcallback\fP, void *\fIdata\fP);
+.
+.SH DESCRIPTION
+.rs
+.sp
+This function provides control over the memory used as a stack at run-time by a
+call to \fBpcre[16|32]_exec()\fP with a pattern that has been successfully
+compiled with JIT optimization. The arguments are:
+.sp
+  extra     the data pointer returned by \fBpcre[16|32]_study()\fP
+  callback  a callback function
+  data      a JIT stack or a value to be passed to the callback
+              function
+.P
+If \fIcallback\fP is NULL and \fIdata\fP is NULL, an internal 32K block on
+the machine stack is used.
+.P
+If \fIcallback\fP is NULL and \fIdata\fP is not NULL, \fIdata\fP must
+be a valid JIT stack, the result of calling \fBpcre[16|32]_jit_stack_alloc()\fP.
+.P
+If \fIcallback\fP not NULL, it is called with \fIdata\fP as an argument at
+the start of matching, in order to set up a JIT stack. If the result is NULL,
+the internal 32K stack is used; otherwise the return value must be a valid JIT
+stack, the result of calling \fBpcre[16|32]_jit_stack_alloc()\fP.
+.P
+You may safely assign the same JIT stack to multiple patterns, as long as they
+are all matched in the same thread. In a multithread application, each thread
+must use its own JIT stack. For more details, see the
+.\" HREF
+\fBpcrejit\fP
+.\"
+page.
+.P
+There is a complete description of the PCRE native API in the
+.\" HREF
+\fBpcreapi\fP
+.\"
+page and a description of the POSIX API in the
+.\" HREF
+\fBpcreposix\fP
+.\"
+page.
diff --git a/pcre/doc/pcre_compile.3 b/pcre/doc/pcre_compile.3
new file mode 100644
index 00000000000..d09768d8021
--- /dev/null
+++ b/pcre/doc/pcre_compile.3
@@ -0,0 +1,98 @@
+.TH PCRE_COMPILE 3 "24 June 2012" "PCRE 8.30"
+.SH NAME
+PCRE - Perl-compatible regular expressions
+.SH SYNOPSIS
+.rs
+.sp
+.B #include 
+.PP
+.SM
+.B pcre *pcre_compile(const char *\fIpattern\fP, int \fIoptions\fP,
+.ti +5n
+.B const char **\fIerrptr\fP, int *\fIerroffset\fP,
+.ti +5n
+.B const unsigned char *\fItableptr\fP);
+.PP
+.B pcre16 *pcre16_compile(PCRE_SPTR16 \fIpattern\fP, int \fIoptions\fP,
+.ti +5n
+.B const char **\fIerrptr\fP, int *\fIerroffset\fP,
+.ti +5n
+.B const unsigned char *\fItableptr\fP);
+.PP
+.B pcre32 *pcre32_compile(PCRE_SPTR32 \fIpattern\fP, int \fIoptions\fP,
+.ti +5n
+.B const char **\fIerrptr\fP, int *\fIerroffset\fP,
+.ti +5n
+.B const unsigned char *\fItableptr\fP);
+.
+.SH DESCRIPTION
+.rs
+.sp
+This function compiles a regular expression into an internal form. It is the
+same as \fBpcre[16|32]_compile2()\fP, except for the absence of the
+\fIerrorcodeptr\fP argument. Its arguments are:
+.sp
+  \fIpattern\fP       A zero-terminated string containing the
+                  regular expression to be compiled
+  \fIoptions\fP       Zero or more option bits
+  \fIerrptr\fP        Where to put an error message
+  \fIerroffset\fP     Offset in pattern where error was found
+  \fItableptr\fP      Pointer to character tables, or NULL to
+                  use the built-in default
+.sp
+The option bits are:
+.sp
+  PCRE_ANCHORED           Force pattern anchoring
+  PCRE_AUTO_CALLOUT       Compile automatic callouts
+  PCRE_BSR_ANYCRLF        \eR matches only CR, LF, or CRLF
+  PCRE_BSR_UNICODE        \eR matches all Unicode line endings
+  PCRE_CASELESS           Do caseless matching
+  PCRE_DOLLAR_ENDONLY     $ not to match newline at end
+  PCRE_DOTALL             . matches anything including NL
+  PCRE_DUPNAMES           Allow duplicate names for subpatterns
+  PCRE_EXTENDED           Ignore white space and # comments
+  PCRE_EXTRA              PCRE extra features
+                            (not much use currently)
+  PCRE_FIRSTLINE          Force matching to be before newline
+  PCRE_JAVASCRIPT_COMPAT  JavaScript compatibility
+  PCRE_MULTILINE          ^ and $ match newlines within data
+  PCRE_NEWLINE_ANY        Recognize any Unicode newline sequence
+  PCRE_NEWLINE_ANYCRLF    Recognize CR, LF, and CRLF as newline
+                            sequences
+  PCRE_NEWLINE_CR         Set CR as the newline sequence
+  PCRE_NEWLINE_CRLF       Set CRLF as the newline sequence
+  PCRE_NEWLINE_LF         Set LF as the newline sequence
+  PCRE_NO_AUTO_CAPTURE    Disable numbered capturing paren-
+                            theses (named ones available)
+  PCRE_NO_UTF16_CHECK     Do not check the pattern for UTF-16
+                            validity (only relevant if
+                            PCRE_UTF16 is set)
+  PCRE_NO_UTF32_CHECK     Do not check the pattern for UTF-32
+                            validity (only relevant if
+                            PCRE_UTF32 is set)
+  PCRE_NO_UTF8_CHECK      Do not check the pattern for UTF-8
+                            validity (only relevant if
+                            PCRE_UTF8 is set)
+  PCRE_UCP                Use Unicode properties for \ed, \ew, etc.
+  PCRE_UNGREEDY           Invert greediness of quantifiers
+  PCRE_UTF16              Run in \fBpcre16_compile()\fP UTF-16 mode
+  PCRE_UTF32              Run in \fBpcre32_compile()\fP UTF-32 mode
+  PCRE_UTF8               Run in \fBpcre_compile()\fP UTF-8 mode
+.sp
+PCRE must be built with UTF support in order to use PCRE_UTF8/16/32 and
+PCRE_NO_UTF8/16/32_CHECK, and with UCP support if PCRE_UCP is used.
+.P
+The yield of the function is a pointer to a private data structure that
+contains the compiled pattern, or NULL if an error was detected. Note that
+compiling regular expressions with one version of PCRE for use with a different
+version is not guaranteed to work and may cause crashes.
+.P
+There is a complete description of the PCRE native API in the
+.\" HREF
+\fBpcreapi\fP
+.\"
+page and a description of the POSIX API in the
+.\" HREF
+\fBpcreposix\fP
+.\"
+page.
diff --git a/pcre/doc/pcre_compile2.3 b/pcre/doc/pcre_compile2.3
new file mode 100644
index 00000000000..1fcae433b43
--- /dev/null
+++ b/pcre/doc/pcre_compile2.3
@@ -0,0 +1,106 @@
+.TH PCRE_COMPILE2 3 "24 June 2012" "PCRE 8.30"
+.SH NAME
+PCRE - Perl-compatible regular expressions
+.SH SYNOPSIS
+.rs
+.sp
+.B #include 
+.PP
+.SM
+.B pcre *pcre_compile2(const char *\fIpattern\fP, int \fIoptions\fP,
+.ti +5n
+.B int *\fIerrorcodeptr\fP,
+.ti +5n
+.B const char **\fIerrptr\fP, int *\fIerroffset\fP,
+.ti +5n
+.B const unsigned char *\fItableptr\fP);
+.PP
+.B pcre16 *pcre16_compile2(PCRE_SPTR16 \fIpattern\fP, int \fIoptions\fP,
+.ti +5n
+.B int *\fIerrorcodeptr\fP,
+.ti +5n
+.B const char **\fIerrptr\fP, int *\fIerroffset\fP,
+.ti +5n
+.B const unsigned char *\fItableptr\fP);
+.PP
+.B pcre32 *pcre32_compile2(PCRE_SPTR32 \fIpattern\fP, int \fIoptions\fP,
+.ti +5n
+.B int *\fIerrorcodeptr\fP,
+.ti +5n
+.B const char **\fIerrptr\fP, int *\fIerroffset\fP,
+.ti +5n
+.B const unsigned char *\fItableptr\fP);
+.
+.SH DESCRIPTION
+.rs
+.sp
+This function compiles a regular expression into an internal form. It is the
+same as \fBpcre[16|32]_compile()\fP, except for the addition of the
+\fIerrorcodeptr\fP argument. The arguments are:
+.
+.sp
+  \fIpattern\fP       A zero-terminated string containing the
+                  regular expression to be compiled
+  \fIoptions\fP       Zero or more option bits
+  \fIerrorcodeptr\fP  Where to put an error code
+  \fIerrptr\fP        Where to put an error message
+  \fIerroffset\fP     Offset in pattern where error was found
+  \fItableptr\fP      Pointer to character tables, or NULL to
+                  use the built-in default
+.sp
+The option bits are:
+.sp
+  PCRE_ANCHORED           Force pattern anchoring
+  PCRE_AUTO_CALLOUT       Compile automatic callouts
+  PCRE_BSR_ANYCRLF        \eR matches only CR, LF, or CRLF
+  PCRE_BSR_UNICODE        \eR matches all Unicode line endings
+  PCRE_CASELESS           Do caseless matching
+  PCRE_DOLLAR_ENDONLY     $ not to match newline at end
+  PCRE_DOTALL             . matches anything including NL
+  PCRE_DUPNAMES           Allow duplicate names for subpatterns
+  PCRE_EXTENDED           Ignore white space and # comments
+  PCRE_EXTRA              PCRE extra features
+                            (not much use currently)
+  PCRE_FIRSTLINE          Force matching to be before newline
+  PCRE_JAVASCRIPT_COMPAT  JavaScript compatibility
+  PCRE_MULTILINE          ^ and $ match newlines within data
+  PCRE_NEWLINE_ANY        Recognize any Unicode newline sequence
+  PCRE_NEWLINE_ANYCRLF    Recognize CR, LF, and CRLF as newline
+                            sequences
+  PCRE_NEWLINE_CR         Set CR as the newline sequence
+  PCRE_NEWLINE_CRLF       Set CRLF as the newline sequence
+  PCRE_NEWLINE_LF         Set LF as the newline sequence
+  PCRE_NO_AUTO_CAPTURE    Disable numbered capturing paren-
+                            theses (named ones available)
+  PCRE_NO_UTF16_CHECK     Do not check the pattern for UTF-16
+                            validity (only relevant if
+                            PCRE_UTF16 is set)
+  PCRE_NO_UTF32_CHECK     Do not check the pattern for UTF-32
+                            validity (only relevant if
+                            PCRE_UTF32 is set)
+  PCRE_NO_UTF8_CHECK      Do not check the pattern for UTF-8
+                            validity (only relevant if
+                            PCRE_UTF8 is set)
+  PCRE_UCP                Use Unicode properties for \ed, \ew, etc.
+  PCRE_UNGREEDY           Invert greediness of quantifiers
+  PCRE_UTF16              Run \fBpcre16_compile()\fP in UTF-16 mode
+  PCRE_UTF32              Run \fBpcre32_compile()\fP in UTF-32 mode
+  PCRE_UTF8               Run \fBpcre_compile()\fP in UTF-8 mode
+.sp
+PCRE must be built with UTF support in order to use PCRE_UTF8/16/32 and
+PCRE_NO_UTF8/16/32_CHECK, and with UCP support if PCRE_UCP is used.
+.P
+The yield of the function is a pointer to a private data structure that
+contains the compiled pattern, or NULL if an error was detected. Note that
+compiling regular expressions with one version of PCRE for use with a different
+version is not guaranteed to work and may cause crashes.
+.P
+There is a complete description of the PCRE native API in the
+.\" HREF
+\fBpcreapi\fP
+.\"
+page and a description of the POSIX API in the
+.\" HREF
+\fBpcreposix\fP
+.\"
+page.
diff --git a/pcre/doc/pcre_config.3 b/pcre/doc/pcre_config.3
new file mode 100644
index 00000000000..5a6e6be7a47
--- /dev/null
+++ b/pcre/doc/pcre_config.3
@@ -0,0 +1,76 @@
+.TH PCRE_CONFIG 3 "24 June 2012" "PCRE 8.30"
+.SH NAME
+PCRE - Perl-compatible regular expressions
+.SH SYNOPSIS
+.rs
+.sp
+.B #include 
+.PP
+.SM
+.B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);
+.PP
+.B int pcre16_config(int \fIwhat\fP, void *\fIwhere\fP);
+.PP
+.B int pcre32_config(int \fIwhat\fP, void *\fIwhere\fP);
+.
+.SH DESCRIPTION
+.rs
+.sp
+This function makes it possible for a client program to find out which optional
+features are available in the version of the PCRE library it is using. The
+arguments are as follows:
+.sp
+  \fIwhat\fP     A code specifying what information is required
+  \fIwhere\fP    Points to where to put the data
+.sp
+The \fIwhere\fP argument must point to an integer variable, except for
+PCRE_CONFIG_MATCH_LIMIT and PCRE_CONFIG_MATCH_LIMIT_RECURSION, when it must
+point to an unsigned long integer. The available codes are:
+.sp
+  PCRE_CONFIG_JIT           Availability of just-in-time compiler
+                              support (1=yes 0=no)
+  PCRE_CONFIG_JITTARGET     String containing information about the
+                              target architecture for the JIT compiler,
+                              or NULL if there is no JIT support
+  PCRE_CONFIG_LINK_SIZE     Internal link size: 2, 3, or 4
+  PCRE_CONFIG_MATCH_LIMIT   Internal resource limit
+  PCRE_CONFIG_MATCH_LIMIT_RECURSION
+                            Internal recursion depth limit
+  PCRE_CONFIG_NEWLINE       Value of the default newline sequence:
+                                13 (0x000d)    for CR
+                                10 (0x000a)    for LF
+                              3338 (0x0d0a)    for CRLF
+                                -2             for ANYCRLF
+                                -1             for ANY
+  PCRE_CONFIG_BSR           Indicates what \eR matches by default:
+                                 0             all Unicode line endings
+                                 1             CR, LF, or CRLF only
+  PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
+                            Threshold of return slots, above which
+                              \fBmalloc()\fP is used by the POSIX API
+  PCRE_CONFIG_STACKRECURSE  Recursion implementation (1=stack 0=heap)
+  PCRE_CONFIG_UTF16         Availability of UTF-16 support (1=yes
+                               0=no); option for \fBpcre16_config()\fP
+  PCRE_CONFIG_UTF32         Availability of UTF-32 support (1=yes
+                               0=no); option for \fBpcre32_config()\fP
+  PCRE_CONFIG_UTF8          Availability of UTF-8 support (1=yes 0=no);
+                              option for \fBpcre_config()\fP
+  PCRE_CONFIG_UNICODE_PROPERTIES
+                            Availability of Unicode property support
+                              (1=yes 0=no)
+.sp
+The function yields 0 on success or PCRE_ERROR_BADOPTION otherwise. That error
+is also given if PCRE_CONFIG_UTF16 or PCRE_CONFIG_UTF32 is passed to
+\fBpcre_config()\fP, if PCRE_CONFIG_UTF8 or PCRE_CONFIG_UTF32 is passed to
+\fBpcre16_config()\fP, or if PCRE_CONFIG_UTF8 or PCRE_CONFIG_UTF16 is passed to
+\fBpcre32_config()\fP.
+.P
+There is a complete description of the PCRE native API in the
+.\" HREF
+\fBpcreapi\fP
+.\"
+page and a description of the POSIX API in the
+.\" HREF
+\fBpcreposix\fP
+.\"
+page.
diff --git a/pcre/doc/pcre_copy_named_substring.3 b/pcre/doc/pcre_copy_named_substring.3
new file mode 100644
index 00000000000..e3281d8feae
--- /dev/null
+++ b/pcre/doc/pcre_copy_named_substring.3
@@ -0,0 +1,59 @@
+.TH PCRE_COPY_NAMED_SUBSTRING 3 "24 June 2012" "PCRE 8.30"
+.SH NAME
+PCRE - Perl-compatible regular expressions
+.SH SYNOPSIS
+.rs
+.sp
+.B #include 
+.PP
+.SM
+.B int pcre_copy_named_substring(const pcre *\fIcode\fP,
+.ti +5n
+.B const char *\fIsubject\fP, int *\fIovector\fP,
+.ti +5n
+.B int \fIstringcount\fP, const char *\fIstringname\fP,
+.ti +5n
+.B char *\fIbuffer\fP, int \fIbuffersize\fP);
+.PP
+.B int pcre16_copy_named_substring(const pcre16 *\fIcode\fP,
+.ti +5n
+.B PCRE_SPTR16 \fIsubject\fP, int *\fIovector\fP,
+.ti +5n
+.B int \fIstringcount\fP, PCRE_SPTR16 \fIstringname\fP,
+.ti +5n
+.B PCRE_UCHAR16 *\fIbuffer\fP, int \fIbuffersize\fP);
+.PP
+.B int pcre32_copy_named_substring(const pcre32 *\fIcode\fP,
+.ti +5n
+.B PCRE_SPTR32 \fIsubject\fP, int *\fIovector\fP,
+.ti +5n
+.B int \fIstringcount\fP, PCRE_SPTR32 \fIstringname\fP,
+.ti +5n
+.B PCRE_UCHAR32 *\fIbuffer\fP, int \fIbuffersize\fP);
+.
+.SH DESCRIPTION
+.rs
+.sp
+This is a convenience function for extracting a captured substring, identified
+by name, into a given buffer. The arguments are:
+.sp
+  \fIcode\fP          Pattern that was successfully matched
+  \fIsubject\fP       Subject that has been successfully matched
+  \fIovector\fP       Offset vector that \fBpcre[16|32]_exec()\fP used
+  \fIstringcount\fP   Value returned by \fBpcre[16|32]_exec()\fP
+  \fIstringname\fP    Name of the required substring
+  \fIbuffer\fP        Buffer to receive the string
+  \fIbuffersize\fP    Size of buffer
+.sp
+The yield is the length of the substring, PCRE_ERROR_NOMEMORY if the buffer was
+too small, or PCRE_ERROR_NOSUBSTRING if the string name is invalid.
+.P
+There is a complete description of the PCRE native API in the
+.\" HREF
+\fBpcreapi\fP
+.\"
+page and a description of the POSIX API in the
+.\" HREF
+\fBpcreposix\fP
+.\"
+page.
diff --git a/pcre/doc/pcre_copy_substring.3 b/pcre/doc/pcre_copy_substring.3
new file mode 100644
index 00000000000..96bff3a2c66
--- /dev/null
+++ b/pcre/doc/pcre_copy_substring.3
@@ -0,0 +1,52 @@
+.TH PCRE_COPY_SUBSTRING 3 "24 June 2012" "PCRE 8.30"
+.SH NAME
+PCRE - Perl-compatible regular expressions
+.SH SYNOPSIS
+.rs
+.sp
+.B #include 
+.PP
+.SM
+.B int pcre_copy_substring(const char *\fIsubject\fP, int *\fIovector\fP,
+.ti +5n
+.B int \fIstringcount\fP, int \fIstringnumber\fP, char *\fIbuffer\fP,
+.ti +5n
+.B int \fIbuffersize\fP);
+.PP
+.B int pcre16_copy_substring(PCRE_SPTR16 \fIsubject\fP, int *\fIovector\fP,
+.ti +5n
+.B int \fIstringcount\fP, int \fIstringnumber\fP, PCRE_UCHAR16 *\fIbuffer\fP,
+.ti +5n
+.B int \fIbuffersize\fP);
+.PP
+.B int pcre32_copy_substring(PCRE_SPTR32 \fIsubject\fP, int *\fIovector\fP,
+.ti +5n
+.B int \fIstringcount\fP, int \fIstringnumber\fP, PCRE_UCHAR32 *\fIbuffer\fP,
+.ti +5n
+.B int \fIbuffersize\fP);
+.
+.SH DESCRIPTION
+.rs
+.sp
+This is a convenience function for extracting a captured substring into a given
+buffer. The arguments are:
+.sp
+  \fIsubject\fP       Subject that has been successfully matched
+  \fIovector\fP       Offset vector that \fBpcre[16|32]_exec()\fP used
+  \fIstringcount\fP   Value returned by \fBpcre[16|32]_exec()\fP
+  \fIstringnumber\fP  Number of the required substring
+  \fIbuffer\fP        Buffer to receive the string
+  \fIbuffersize\fP    Size of buffer
+.sp
+The yield is the length of the string, PCRE_ERROR_NOMEMORY if the buffer was
+too small, or PCRE_ERROR_NOSUBSTRING if the string number is invalid.
+.P
+There is a complete description of the PCRE native API in the
+.\" HREF
+\fBpcreapi\fP
+.\"
+page and a description of the POSIX API in the
+.\" HREF
+\fBpcreposix\fP
+.\"
+page.
diff --git a/pcre/doc/pcre_dfa_exec.3 b/pcre/doc/pcre_dfa_exec.3
new file mode 100644
index 00000000000..9bc7448b6a3
--- /dev/null
+++ b/pcre/doc/pcre_dfa_exec.3
@@ -0,0 +1,126 @@
+.TH PCRE_DFA_EXEC 3 "12 May 2013" "PCRE 8.33"
+.SH NAME
+PCRE - Perl-compatible regular expressions
+.SH SYNOPSIS
+.rs
+.sp
+.B #include 
+.PP
+.SM
+.B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
+.ti +5n
+.B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
+.ti +5n
+.B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
+.ti +5n
+.B int *\fIworkspace\fP, int \fIwscount\fP);
+.PP
+.B int pcre16_dfa_exec(const pcre16 *\fIcode\fP, "const pcre16_extra *\fIextra\fP,"
+.ti +5n
+.B "PCRE_SPTR16 \fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
+.ti +5n
+.B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
+.ti +5n
+.B int *\fIworkspace\fP, int \fIwscount\fP);
+.PP
+.B int pcre32_dfa_exec(const pcre32 *\fIcode\fP, "const pcre32_extra *\fIextra\fP,"
+.ti +5n
+.B "PCRE_SPTR32 \fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
+.ti +5n
+.B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
+.ti +5n
+.B int *\fIworkspace\fP, int \fIwscount\fP);
+.
+.SH DESCRIPTION
+.rs
+.sp
+This function matches a compiled regular expression against a given subject
+string, using an alternative matching algorithm that scans the subject string
+just once (\fInot\fP Perl-compatible). Note that the main, Perl-compatible,
+matching function is \fBpcre[16|32]_exec()\fP. The arguments for this function
+are:
+.sp
+  \fIcode\fP         Points to the compiled pattern
+  \fIextra\fP        Points to an associated \fBpcre[16|32]_extra\fP structure,
+                 or is NULL
+  \fIsubject\fP      Points to the subject string
+  \fIlength\fP       Length of the subject string
+  \fIstartoffset\fP  Offset in the subject at which to start matching
+  \fIoptions\fP      Option bits
+  \fIovector\fP      Points to a vector of ints for result offsets
+  \fIovecsize\fP     Number of elements in the vector
+  \fIworkspace\fP    Points to a vector of ints used as working space
+  \fIwscount\fP      Number of elements in the vector
+.sp
+The units for \fIlength\fP and \fIstartoffset\fP are bytes for
+\fBpcre_exec()\fP, 16-bit data items for \fBpcre16_exec()\fP, and 32-bit items
+for \fBpcre32_exec()\fP. The options are:
+.sp
+  PCRE_ANCHORED          Match only at the first position
+  PCRE_BSR_ANYCRLF       \eR matches only CR, LF, or CRLF
+  PCRE_BSR_UNICODE       \eR matches all Unicode line endings
+  PCRE_NEWLINE_ANY       Recognize any Unicode newline sequence
+  PCRE_NEWLINE_ANYCRLF   Recognize CR, LF, & CRLF as newline sequences
+  PCRE_NEWLINE_CR        Recognize CR as the only newline sequence
+  PCRE_NEWLINE_CRLF      Recognize CRLF as the only newline sequence
+  PCRE_NEWLINE_LF        Recognize LF as the only newline sequence
+  PCRE_NOTBOL            Subject is not the beginning of a line
+  PCRE_NOTEOL            Subject is not the end of a line
+  PCRE_NOTEMPTY          An empty string is not a valid match
+  PCRE_NOTEMPTY_ATSTART  An empty string at the start of the subject
+                           is not a valid match
+  PCRE_NO_START_OPTIMIZE Do not do "start-match" optimizations
+  PCRE_NO_UTF16_CHECK    Do not check the subject for UTF-16
+                           validity (only relevant if PCRE_UTF16
+                           was set at compile time)
+  PCRE_NO_UTF32_CHECK    Do not check the subject for UTF-32
+                           validity (only relevant if PCRE_UTF32
+                           was set at compile time)
+  PCRE_NO_UTF8_CHECK     Do not check the subject for UTF-8
+                           validity (only relevant if PCRE_UTF8
+                           was set at compile time)
+  PCRE_PARTIAL           ) Return PCRE_ERROR_PARTIAL for a partial
+  PCRE_PARTIAL_SOFT      )   match if no full matches are found
+  PCRE_PARTIAL_HARD      Return PCRE_ERROR_PARTIAL for a partial match
+                           even if there is a full match as well
+  PCRE_DFA_SHORTEST      Return only the shortest match
+  PCRE_DFA_RESTART       Restart after a partial match
+.sp
+There are restrictions on what may appear in a pattern when using this matching
+function. Details are given in the
+.\" HREF
+\fBpcrematching\fP
+.\"
+documentation. For details of partial matching, see the
+.\" HREF
+\fBpcrepartial\fP
+.\"
+page.
+.P
+A \fBpcre[16|32]_extra\fP structure contains the following fields:
+.sp
+  \fIflags\fP            Bits indicating which fields are set
+  \fIstudy_data\fP       Opaque data from \fBpcre[16|32]_study()\fP
+  \fImatch_limit\fP      Limit on internal resource use
+  \fImatch_limit_recursion\fP  Limit on internal recursion depth
+  \fIcallout_data\fP     Opaque data passed back to callouts
+  \fItables\fP           Points to character tables or is NULL
+  \fImark\fP             For passing back a *MARK pointer
+  \fIexecutable_jit\fP   Opaque data from JIT compilation
+.sp
+The flag bits are PCRE_EXTRA_STUDY_DATA, PCRE_EXTRA_MATCH_LIMIT,
+PCRE_EXTRA_MATCH_LIMIT_RECURSION, PCRE_EXTRA_CALLOUT_DATA,
+PCRE_EXTRA_TABLES, PCRE_EXTRA_MARK and PCRE_EXTRA_EXECUTABLE_JIT. For this
+matching function, the \fImatch_limit\fP and \fImatch_limit_recursion\fP fields
+are not used, and must not be set. The PCRE_EXTRA_EXECUTABLE_JIT flag and
+the corresponding variable are ignored.
+.P
+There is a complete description of the PCRE native API in the
+.\" HREF
+\fBpcreapi\fP
+.\"
+page and a description of the POSIX API in the
+.\" HREF
+\fBpcreposix\fP
+.\"
+page.
diff --git a/pcre/doc/pcre_exec.3 b/pcre/doc/pcre_exec.3
new file mode 100644
index 00000000000..c92c2a57a69
--- /dev/null
+++ b/pcre/doc/pcre_exec.3
@@ -0,0 +1,104 @@
+.TH PCRE_EXEC 3 "12 May 2013" "PCRE 8.33"
+.SH NAME
+PCRE - Perl-compatible regular expressions
+.SH SYNOPSIS
+.rs
+.sp
+.B #include 
+.PP
+.SM
+.B int pcre_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
+.ti +5n
+.B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
+.ti +5n
+.B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);
+.PP
+.B int pcre16_exec(const pcre16 *\fIcode\fP, "const pcre16_extra *\fIextra\fP,"
+.ti +5n
+.B "PCRE_SPTR16 \fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
+.ti +5n
+.B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);
+.PP
+.B int pcre32_exec(const pcre32 *\fIcode\fP, "const pcre32_extra *\fIextra\fP,"
+.ti +5n
+.B "PCRE_SPTR32 \fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
+.ti +5n
+.B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);
+.
+.SH DESCRIPTION
+.rs
+.sp
+This function matches a compiled regular expression against a given subject
+string, using a matching algorithm that is similar to Perl's. It returns
+offsets to captured substrings. Its arguments are:
+.sp
+  \fIcode\fP         Points to the compiled pattern
+  \fIextra\fP        Points to an associated \fBpcre[16|32]_extra\fP structure,
+                 or is NULL
+  \fIsubject\fP      Points to the subject string
+  \fIlength\fP       Length of the subject string
+  \fIstartoffset\fP  Offset in the subject at which to start matching
+  \fIoptions\fP      Option bits
+  \fIovector\fP      Points to a vector of ints for result offsets
+  \fIovecsize\fP     Number of elements in the vector (a multiple of 3)
+.sp
+The units for \fIlength\fP and \fIstartoffset\fP are bytes for
+\fBpcre_exec()\fP, 16-bit data items for \fBpcre16_exec()\fP, and 32-bit items
+for \fBpcre32_exec()\fP. The options are:
+.sp
+  PCRE_ANCHORED          Match only at the first position
+  PCRE_BSR_ANYCRLF       \eR matches only CR, LF, or CRLF
+  PCRE_BSR_UNICODE       \eR matches all Unicode line endings
+  PCRE_NEWLINE_ANY       Recognize any Unicode newline sequence
+  PCRE_NEWLINE_ANYCRLF   Recognize CR, LF, & CRLF as newline sequences
+  PCRE_NEWLINE_CR        Recognize CR as the only newline sequence
+  PCRE_NEWLINE_CRLF      Recognize CRLF as the only newline sequence
+  PCRE_NEWLINE_LF        Recognize LF as the only newline sequence
+  PCRE_NOTBOL            Subject string is not the beginning of a line
+  PCRE_NOTEOL            Subject string is not the end of a line
+  PCRE_NOTEMPTY          An empty string is not a valid match
+  PCRE_NOTEMPTY_ATSTART  An empty string at the start of the subject
+                           is not a valid match
+  PCRE_NO_START_OPTIMIZE Do not do "start-match" optimizations
+  PCRE_NO_UTF16_CHECK    Do not check the subject for UTF-16
+                           validity (only relevant if PCRE_UTF16
+                           was set at compile time)
+  PCRE_NO_UTF32_CHECK    Do not check the subject for UTF-32
+                           validity (only relevant if PCRE_UTF32
+                           was set at compile time)
+  PCRE_NO_UTF8_CHECK     Do not check the subject for UTF-8
+                           validity (only relevant if PCRE_UTF8
+                           was set at compile time)
+  PCRE_PARTIAL           ) Return PCRE_ERROR_PARTIAL for a partial
+  PCRE_PARTIAL_SOFT      )   match if no full matches are found
+  PCRE_PARTIAL_HARD      Return PCRE_ERROR_PARTIAL for a partial match
+                           if that is found before a full match
+.sp
+For details of partial matching, see the
+.\" HREF
+\fBpcrepartial\fP
+.\"
+page. A \fBpcre_extra\fP structure contains the following fields:
+.sp
+  \fIflags\fP            Bits indicating which fields are set
+  \fIstudy_data\fP       Opaque data from \fBpcre[16|32]_study()\fP
+  \fImatch_limit\fP      Limit on internal resource use
+  \fImatch_limit_recursion\fP  Limit on internal recursion depth
+  \fIcallout_data\fP     Opaque data passed back to callouts
+  \fItables\fP           Points to character tables or is NULL
+  \fImark\fP             For passing back a *MARK pointer
+  \fIexecutable_jit\fP   Opaque data from JIT compilation
+.sp
+The flag bits are PCRE_EXTRA_STUDY_DATA, PCRE_EXTRA_MATCH_LIMIT,
+PCRE_EXTRA_MATCH_LIMIT_RECURSION, PCRE_EXTRA_CALLOUT_DATA,
+PCRE_EXTRA_TABLES, PCRE_EXTRA_MARK and PCRE_EXTRA_EXECUTABLE_JIT.
+.P
+There is a complete description of the PCRE native API in the
+.\" HREF
+\fBpcreapi\fP
+.\"
+page and a description of the POSIX API in the
+.\" HREF
+\fBpcreposix\fP
+.\"
+page.
diff --git a/pcre/doc/pcre_free_study.3 b/pcre/doc/pcre_free_study.3
new file mode 100644
index 00000000000..8826b73597b
--- /dev/null
+++ b/pcre/doc/pcre_free_study.3
@@ -0,0 +1,31 @@
+.TH PCRE_FREE_STUDY 3 "24 June 2012" "PCRE 8.30"
+.SH NAME
+PCRE - Perl-compatible regular expressions
+.SH SYNOPSIS
+.rs
+.sp
+.B #include 
+.PP
+.SM
+.B void pcre_free_study(pcre_extra *\fIextra\fP);
+.PP
+.B void pcre16_free_study(pcre16_extra *\fIextra\fP);
+.PP
+.B void pcre32_free_study(pcre32_extra *\fIextra\fP);
+.
+.SH DESCRIPTION
+.rs
+.sp
+This function is used to free the memory used for the data generated by a call
+to \fBpcre[16|32]_study()\fP when it is no longer needed. The argument must be the
+result of such a call.
+.P
+There is a complete description of the PCRE native API in the
+.\" HREF
+\fBpcreapi\fP
+.\"
+page and a description of the POSIX API in the
+.\" HREF
+\fBpcreposix\fP
+.\"
+page.
diff --git a/pcre/doc/pcre_free_substring.3 b/pcre/doc/pcre_free_substring.3
new file mode 100644
index 00000000000..88c04019f40
--- /dev/null
+++ b/pcre/doc/pcre_free_substring.3
@@ -0,0 +1,31 @@
+.TH PCRE_FREE_SUBSTRING 3 "24 June 2012" "PCRE 8.30"
+.SH NAME
+PCRE - Perl-compatible regular expressions
+.SH SYNOPSIS
+.rs
+.sp
+.B #include 
+.PP
+.SM
+.B void pcre_free_substring(const char *\fIstringptr\fP);
+.PP
+.B void pcre16_free_substring(PCRE_SPTR16 \fIstringptr\fP);
+.PP
+.B void pcre32_free_substring(PCRE_SPTR32 \fIstringptr\fP);
+.
+.SH DESCRIPTION
+.rs
+.sp
+This is a convenience function for freeing the store obtained by a previous
+call to \fBpcre[16|32]_get_substring()\fP or \fBpcre[16|32]_get_named_substring()\fP.
+Its only argument is a pointer to the string.
+.P
+There is a complete description of the PCRE native API in the
+.\" HREF
+\fBpcreapi\fP
+.\"
+page and a description of the POSIX API in the
+.\" HREF
+\fBpcreposix\fP
+.\"
+page.
diff --git a/pcre/doc/pcre_free_substring_list.3 b/pcre/doc/pcre_free_substring_list.3
new file mode 100644
index 00000000000..248b4bd01b9
--- /dev/null
+++ b/pcre/doc/pcre_free_substring_list.3
@@ -0,0 +1,31 @@
+.TH PCRE_FREE_SUBSTRING_LIST 3 "24 June 2012" "PCRE 8.30"
+.SH NAME
+PCRE - Perl-compatible regular expressions
+.SH SYNOPSIS
+.rs
+.sp
+.B #include 
+.PP
+.SM
+.B void pcre_free_substring_list(const char **\fIstringptr\fP);
+.PP
+.B void pcre16_free_substring_list(PCRE_SPTR16 *\fIstringptr\fP);
+.PP
+.B void pcre32_free_substring_list(PCRE_SPTR32 *\fIstringptr\fP);
+.
+.SH DESCRIPTION
+.rs
+.sp
+This is a convenience function for freeing the store obtained by a previous
+call to \fBpcre[16|32]_get_substring_list()\fP. Its only argument is a pointer to
+the list of string pointers.
+.P
+There is a complete description of the PCRE native API in the
+.\" HREF
+\fBpcreapi\fP
+.\"
+page and a description of the POSIX API in the
+.\" HREF
+\fBpcreposix\fP
+.\"
+page.
diff --git a/pcre/doc/pcre_fullinfo.3 b/pcre/doc/pcre_fullinfo.3
new file mode 100644
index 00000000000..ad640fc66fc
--- /dev/null
+++ b/pcre/doc/pcre_fullinfo.3
@@ -0,0 +1,95 @@
+.TH PCRE_FULLINFO 3 "24 June 2012" "PCRE 8.30"
+.SH NAME
+PCRE - Perl-compatible regular expressions
+.SH SYNOPSIS
+.rs
+.sp
+.B #include 
+.PP
+.SM
+.B int pcre_fullinfo(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
+.ti +5n
+.B int \fIwhat\fP, void *\fIwhere\fP);
+.PP
+.B int pcre16_fullinfo(const pcre16 *\fIcode\fP, "const pcre16_extra *\fIextra\fP,"
+.ti +5n
+.B int \fIwhat\fP, void *\fIwhere\fP);
+.PP
+.B int pcre32_fullinfo(const pcre32 *\fIcode\fP, "const pcre32_extra *\fIextra\fP,"
+.ti +5n
+.B int \fIwhat\fP, void *\fIwhere\fP);
+.
+.SH DESCRIPTION
+.rs
+.sp
+This function returns information about a compiled pattern. Its arguments are:
+.sp
+  \fIcode\fP                      Compiled regular expression
+  \fIextra\fP                     Result of \fBpcre[16|32]_study()\fP or NULL
+  \fIwhat\fP                      What information is required
+  \fIwhere\fP                     Where to put the information
+.sp
+The following information is available:
+.sp
+  PCRE_INFO_BACKREFMAX      Number of highest back reference
+  PCRE_INFO_CAPTURECOUNT    Number of capturing subpatterns
+  PCRE_INFO_DEFAULT_TABLES  Pointer to default tables
+  PCRE_INFO_FIRSTBYTE       Fixed first data unit for a match, or
+                              -1 for start of string
+                                 or after newline, or
+                              -2 otherwise
+  PCRE_INFO_FIRSTTABLE      Table of first data units (after studying)
+  PCRE_INFO_HASCRORLF       Return 1 if explicit CR or LF matches exist
+  PCRE_INFO_JCHANGED        Return 1 if (?J) or (?-J) was used
+  PCRE_INFO_JIT             Return 1 after successful JIT compilation
+  PCRE_INFO_JITSIZE         Size of JIT compiled code
+  PCRE_INFO_LASTLITERAL     Literal last data unit required
+  PCRE_INFO_MINLENGTH       Lower bound length of matching strings
+  PCRE_INFO_NAMECOUNT       Number of named subpatterns
+  PCRE_INFO_NAMEENTRYSIZE   Size of name table entry
+  PCRE_INFO_NAMETABLE       Pointer to name table
+  PCRE_INFO_OKPARTIAL       Return 1 if partial matching can be tried
+                              (always returns 1 after release 8.00)
+  PCRE_INFO_OPTIONS         Option bits used for compilation
+  PCRE_INFO_SIZE            Size of compiled pattern
+  PCRE_INFO_STUDYSIZE       Size of study data
+  PCRE_INFO_FIRSTCHARACTER      Fixed first data unit for a match
+  PCRE_INFO_FIRSTCHARACTERFLAGS Returns
+                                  1 if there is a first data character set, which can
+                                    then be retrieved using PCRE_INFO_FIRSTCHARACTER,
+                                  2 if the first character is at the start of the data
+                                    string or after a newline, and
+                                  0 otherwise
+  PCRE_INFO_REQUIREDCHAR      Literal last data unit required
+  PCRE_INFO_REQUIREDCHARFLAGS Returns 1 if the last data character is set (which can then
+                              be retrieved using PCRE_INFO_REQUIREDCHAR); 0 otherwise
+.sp
+The \fIwhere\fP argument must point to an integer variable, except for the
+following \fIwhat\fP values:
+.sp
+  PCRE_INFO_DEFAULT_TABLES  const unsigned char *
+  PCRE_INFO_FIRSTTABLE      const unsigned char *
+  PCRE_INFO_NAMETABLE       PCRE_SPTR16           (16-bit library)
+  PCRE_INFO_NAMETABLE       PCRE_SPTR32           (32-bit library)
+  PCRE_INFO_NAMETABLE       const unsigned char * (8-bit library)
+  PCRE_INFO_OPTIONS         unsigned long int
+  PCRE_INFO_SIZE            size_t
+  PCRE_INFO_FIRSTCHARACTER  uint32_t
+  PCRE_INFO_REQUIREDCHAR    uint32_t
+.sp
+The yield of the function is zero on success or:
+.sp
+  PCRE_ERROR_NULL           the argument \fIcode\fP was NULL
+                            the argument \fIwhere\fP was NULL
+  PCRE_ERROR_BADMAGIC       the "magic number" was not found
+  PCRE_ERROR_BADOPTION      the value of \fIwhat\fP was invalid
+.P
+There is a complete description of the PCRE native API in the
+.\" HREF
+\fBpcreapi\fP
+.\"
+page and a description of the POSIX API in the
+.\" HREF
+\fBpcreposix\fP
+.\"
+page.
diff --git a/pcre/doc/pcre_get_named_substring.3 b/pcre/doc/pcre_get_named_substring.3
new file mode 100644
index 00000000000..f81a243c6bf
--- /dev/null
+++ b/pcre/doc/pcre_get_named_substring.3
@@ -0,0 +1,62 @@
+.TH PCRE_GET_NAMED_SUBSTRING 3 "24 June 2012" "PCRE 8.30"
+.SH NAME
+PCRE - Perl-compatible regular expressions
+.SH SYNOPSIS
+.rs
+.sp
+.B #include 
+.PP
+.SM
+.B int pcre_get_named_substring(const pcre *\fIcode\fP,
+.ti +5n
+.B const char *\fIsubject\fP, int *\fIovector\fP,
+.ti +5n
+.B int \fIstringcount\fP, const char *\fIstringname\fP,
+.ti +5n
+.B const char **\fIstringptr\fP);
+.PP
+.B int pcre16_get_named_substring(const pcre16 *\fIcode\fP,
+.ti +5n
+.B PCRE_SPTR16 \fIsubject\fP, int *\fIovector\fP,
+.ti +5n
+.B int \fIstringcount\fP, PCRE_SPTR16 \fIstringname\fP,
+.ti +5n
+.B PCRE_SPTR16 *\fIstringptr\fP);
+.PP
+.B int pcre32_get_named_substring(const pcre32 *\fIcode\fP,
+.ti +5n
+.B PCRE_SPTR32 \fIsubject\fP, int *\fIovector\fP,
+.ti +5n
+.B int \fIstringcount\fP, PCRE_SPTR32 \fIstringname\fP,
+.ti +5n
+.B PCRE_SPTR32 *\fIstringptr\fP);
+.
+.SH DESCRIPTION
+.rs
+.sp
+This is a convenience function for extracting a captured substring by name. The
+arguments are:
+.sp
+  \fIcode\fP          Compiled pattern
+  \fIsubject\fP       Subject that has been successfully matched
+  \fIovector\fP       Offset vector that \fBpcre[16|32]_exec()\fP used
+  \fIstringcount\fP   Value returned by \fBpcre[16|32]_exec()\fP
+  \fIstringname\fP    Name of the required substring
+  \fIstringptr\fP     Where to put the string pointer
+.sp
+The memory in which the substring is placed is obtained by calling
+\fBpcre[16|32]_malloc()\fP. The convenience function
+\fBpcre[16|32]_free_substring()\fP can be used to free it when it is no longer
+needed. The yield of the function is the length of the extracted substring,
+PCRE_ERROR_NOMEMORY if sufficient memory could not be obtained, or
+PCRE_ERROR_NOSUBSTRING if the string name is invalid.
+.P
+There is a complete description of the PCRE native API in the
+.\" HREF
+\fBpcreapi\fP
+.\"
+page and a description of the POSIX API in the
+.\" HREF
+\fBpcreposix\fP
+.\"
+page.
diff --git a/pcre/doc/pcre_get_stringnumber.3 b/pcre/doc/pcre_get_stringnumber.3
new file mode 100644
index 00000000000..7def00bafd1
--- /dev/null
+++ b/pcre/doc/pcre_get_stringnumber.3
@@ -0,0 +1,45 @@
+.TH PCRE_GET_STRINGNUMBER 3 "24 June 2012" "PCRE 8.30"
+.SH NAME
+PCRE - Perl-compatible regular expressions
+.SH SYNOPSIS
+.rs
+.sp
+.B #include 
+.PP
+.SM
+.B int pcre_get_stringnumber(const pcre *\fIcode\fP,
+.ti +5n
+.B const char *\fIname\fP);
+.PP
+.B int pcre16_get_stringnumber(const pcre16 *\fIcode\fP,
+.ti +5n
+.B PCRE_SPTR16 \fIname\fP);
+.PP
+.B int pcre32_get_stringnumber(const pcre32 *\fIcode\fP,
+.ti +5n
+.B PCRE_SPTR32 \fIname\fP);
+.
+.SH DESCRIPTION
+.rs
+.sp
+This convenience function finds the number of a named substring capturing
+parenthesis in a compiled pattern. Its arguments are:
+.sp
+  \fIcode\fP    Compiled regular expression
+  \fIname\fP    Name whose number is required
+.sp
+The yield of the function is the number of the parenthesis if the name is
+found, or PCRE_ERROR_NOSUBSTRING otherwise. When duplicate names are allowed
+(PCRE_DUPNAMES is set), it is not defined which of the numbers is returned by
+\fBpcre[16|32]_get_stringnumber()\fP. You can obtain the complete list by calling
+\fBpcre[16|32]_get_stringtable_entries()\fP.
+.P
+There is a complete description of the PCRE native API in the
+.\" HREF
+\fBpcreapi\fP
+.\"
+page and a description of the POSIX API in the
+.\" HREF
+\fBpcreposix\fP
+.\"
+page.
diff --git a/pcre/doc/pcre_get_stringtable_entries.3 b/pcre/doc/pcre_get_stringtable_entries.3
new file mode 100644
index 00000000000..39178165c59
--- /dev/null
+++ b/pcre/doc/pcre_get_stringtable_entries.3
@@ -0,0 +1,48 @@
+.TH PCRE_GET_STRINGTABLE_ENTRIES 3 "24 June 2012" "PCRE 8.30"
+.SH NAME
+PCRE - Perl-compatible regular expressions
+.SH SYNOPSIS
+.rs
+.sp
+.B #include 
+.PP
+.SM
+.B int pcre_get_stringtable_entries(const pcre *\fIcode\fP,
+.ti +5n
+.B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
+.PP
+.B int pcre16_get_stringtable_entries(const pcre16 *\fIcode\fP,
+.ti +5n
+.B PCRE_SPTR16 \fIname\fP, PCRE_UCHAR16 **\fIfirst\fP, PCRE_UCHAR16 **\fIlast\fP);
+.PP
+.B int pcre32_get_stringtable_entries(const pcre32 *\fIcode\fP,
+.ti +5n
+.B PCRE_SPTR32 \fIname\fP, PCRE_UCHAR32 **\fIfirst\fP, PCRE_UCHAR32 **\fIlast\fP);
+.
+.SH DESCRIPTION
+.rs
+.sp
+This convenience function finds, for a compiled pattern, the first and last
+entries for a given name in the table that translates capturing parenthesis
+names into numbers. When names are required to be unique (PCRE_DUPNAMES is
+\fInot\fP set), it is usually easier to use \fBpcre[16|32]_get_stringnumber()\fP
+instead.
+.sp
+  \fIcode\fP    Compiled regular expression
+  \fIname\fP    Name whose entries required
+  \fIfirst\fP   Where to return a pointer to the first entry
+  \fIlast\fP    Where to return a pointer to the last entry
+.sp
+The yield of the function is the length of each entry, or
+PCRE_ERROR_NOSUBSTRING if none are found.
+.P
+There is a complete description of the PCRE native API, including the format of
+the table entries, in the
+.\" HREF
+\fBpcreapi\fP
+.\"
+page, and a description of the POSIX API in the
+.\" HREF
+\fBpcreposix\fP
+.\"
+page.
diff --git a/pcre/doc/pcre_get_substring.3 b/pcre/doc/pcre_get_substring.3
new file mode 100644
index 00000000000..d5bc60c45d1
--- /dev/null
+++ b/pcre/doc/pcre_get_substring.3
@@ -0,0 +1,55 @@
+.TH PCRE_GET_SUBSTRING 3 "24 June 2012" "PCRE 8.30"
+.SH NAME
+PCRE - Perl-compatible regular expressions
+.SH SYNOPSIS
+.rs
+.sp
+.B #include 
+.PP
+.SM
+.B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,
+.ti +5n
+.B int \fIstringcount\fP, int \fIstringnumber\fP,
+.ti +5n
+.B const char **\fIstringptr\fP);
+.PP
+.B int pcre16_get_substring(PCRE_SPTR16 \fIsubject\fP, int *\fIovector\fP,
+.ti +5n
+.B int \fIstringcount\fP, int \fIstringnumber\fP,
+.ti +5n
+.B PCRE_SPTR16 *\fIstringptr\fP);
+.PP
+.B int pcre32_get_substring(PCRE_SPTR32 \fIsubject\fP, int *\fIovector\fP,
+.ti +5n
+.B int \fIstringcount\fP, int \fIstringnumber\fP,
+.ti +5n
+.B PCRE_SPTR32 *\fIstringptr\fP);
+.
+.SH DESCRIPTION
+.rs
+.sp
+This is a convenience function for extracting a captured substring. The
+arguments are:
+.sp
+  \fIsubject\fP       Subject that has been successfully matched
+  \fIovector\fP       Offset vector that \fBpcre[16|32]_exec()\fP used
+  \fIstringcount\fP   Value returned by \fBpcre[16|32]_exec()\fP
+  \fIstringnumber\fP  Number of the required substring
+  \fIstringptr\fP     Where to put the string pointer
+.sp
+The memory in which the substring is placed is obtained by calling
+\fBpcre[16|32]_malloc()\fP. The convenience function
+\fBpcre[16|32]_free_substring()\fP can be used to free it when it is no longer
+needed. The yield of the function is the length of the substring,
+PCRE_ERROR_NOMEMORY if sufficient memory could not be obtained, or
+PCRE_ERROR_NOSUBSTRING if the string number is invalid.
+.P
+There is a complete description of the PCRE native API in the
+.\" HREF
+\fBpcreapi\fP
+.\"
+page and a description of the POSIX API in the
+.\" HREF
+\fBpcreposix\fP
+.\"
+page.
diff --git a/pcre/doc/pcre_get_substring_list.3 b/pcre/doc/pcre_get_substring_list.3
new file mode 100644
index 00000000000..a1a57496b19
--- /dev/null
+++ b/pcre/doc/pcre_get_substring_list.3
@@ -0,0 +1,49 @@
+.TH PCRE_GET_SUBSTRING_LIST 3 "24 June 2012" "PCRE 8.30"
+.SH NAME
+PCRE - Perl-compatible regular expressions
+.SH SYNOPSIS
+.rs
+.sp
+.B #include 
+.PP
+.SM
+.B int pcre_get_substring_list(const char *\fIsubject\fP,
+.ti +5n
+.B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"
+.PP
+.B int pcre16_get_substring_list(PCRE_SPTR16 \fIsubject\fP,
+.ti +5n
+.B int *\fIovector\fP, int \fIstringcount\fP, "PCRE_SPTR16 **\fIlistptr\fP);"
+.PP
+.B int pcre32_get_substring_list(PCRE_SPTR32 \fIsubject\fP,
+.ti +5n
+.B int *\fIovector\fP, int \fIstringcount\fP, "PCRE_SPTR32 **\fIlistptr\fP);"
+.
+.SH DESCRIPTION
+.rs
+.sp
+This is a convenience function for extracting a list of all the captured
+substrings. The arguments are:
+.sp
+  \fIsubject\fP       Subject that has been successfully matched
+  \fIovector\fP       Offset vector that \fBpcre[16|32]_exec\fP used
+  \fIstringcount\fP   Value returned by \fBpcre[16|32]_exec\fP
+  \fIlistptr\fP       Where to put a pointer to the list
+.sp
+The memory in which the substrings and the list are placed is obtained by
+calling \fBpcre[16|32]_malloc()\fP. The convenience function
+\fBpcre[16|32]_free_substring_list()\fP can be used to free it when it is no
+longer needed. A pointer to a list of pointers is put in the variable whose
+address is in \fIlistptr\fP. The list is terminated by a NULL pointer. The
+yield of the function is zero on success or PCRE_ERROR_NOMEMORY if sufficient
+memory could not be obtained.
+.P
+There is a complete description of the PCRE native API in the
+.\" HREF
+\fBpcreapi\fP
+.\"
+page and a description of the POSIX API in the
+.\" HREF
+\fBpcreposix\fP
+.\"
+page.
diff --git a/pcre/doc/pcre_jit_exec.3 b/pcre/doc/pcre_jit_exec.3
new file mode 100644
index 00000000000..8b9d05bf1ee
--- /dev/null
+++ b/pcre/doc/pcre_jit_exec.3
@@ -0,0 +1,104 @@
+.TH PCRE_EXEC 3 "31 October 2012" "PCRE 8.30"
+.SH NAME
+PCRE - Perl-compatible regular expressions
+.SH SYNOPSIS
+.rs
+.sp
+.B #include 
+.PP
+.SM
+.B int pcre_jit_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
+.ti +5n
+.B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
+.ti +5n
+.B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
+.ti +5n
+.B pcre_jit_stack *\fIjstack\fP);
+.PP
+.B int pcre16_jit_exec(const pcre16 *\fIcode\fP, "const pcre16_extra *\fIextra\fP,"
+.ti +5n
+.B "PCRE_SPTR16 \fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
+.ti +5n
+.B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
+.ti +5n
+.B pcre_jit_stack *\fIjstack\fP);
+.PP
+.B int pcre32_jit_exec(const pcre32 *\fIcode\fP, "const pcre32_extra *\fIextra\fP,"
+.ti +5n
+.B "PCRE_SPTR32 \fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
+.ti +5n
+.B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
+.ti +5n
+.B pcre_jit_stack *\fIjstack\fP);
+.
+.SH DESCRIPTION
+.rs
+.sp
+This function matches a compiled regular expression that has been successfully
+studied with one of the JIT options against a given subject string, using a
+matching algorithm that is similar to Perl's. It is a "fast path" interface to
+JIT, and it bypasses some of the sanity checks that \fBpcre_exec()\fP applies.
+It returns offsets to captured substrings. Its arguments are:
+.sp
+  \fIcode\fP         Points to the compiled pattern
+  \fIextra\fP        Points to an associated \fBpcre[16|32]_extra\fP structure,
+                 or is NULL
+  \fIsubject\fP      Points to the subject string
+  \fIlength\fP       Length of the subject string, in bytes
+  \fIstartoffset\fP  Offset in bytes in the subject at which to
+                 start matching
+  \fIoptions\fP      Option bits
+  \fIovector\fP      Points to a vector of ints for result offsets
+  \fIovecsize\fP     Number of elements in the vector (a multiple of 3)
+  \fIjstack\fP       Pointer to a JIT stack
+.sp
+The allowed options are:
+.sp
+  PCRE_NOTBOL            Subject string is not the beginning of a line
+  PCRE_NOTEOL            Subject string is not the end of a line
+  PCRE_NOTEMPTY          An empty string is not a valid match
+  PCRE_NOTEMPTY_ATSTART  An empty string at the start of the subject
+                           is not a valid match
+  PCRE_NO_UTF16_CHECK    Do not check the subject for UTF-16
+                           validity (only relevant if PCRE_UTF16
+                           was set at compile time)
+  PCRE_NO_UTF32_CHECK    Do not check the subject for UTF-32
+                           validity (only relevant if PCRE_UTF32
+                           was set at compile time)
+  PCRE_NO_UTF8_CHECK     Do not check the subject for UTF-8
+                           validity (only relevant if PCRE_UTF8
+                           was set at compile time)
+  PCRE_PARTIAL           ) Return PCRE_ERROR_PARTIAL for a partial
+  PCRE_PARTIAL_SOFT      )   match if no full matches are found
+  PCRE_PARTIAL_HARD      Return PCRE_ERROR_PARTIAL for a partial match
+                           if that is found before a full match
+.sp
+However, the PCRE_NO_UTF[8|16|32]_CHECK options have no effect, as this check
+is never applied. For details of partial matching, see the
+.\" HREF
+\fBpcrepartial\fP
+.\"
+page. A \fBpcre_extra\fP structure contains the following fields:
+.sp
+  \fIflags\fP            Bits indicating which fields are set
+  \fIstudy_data\fP       Opaque data from \fBpcre[16|32]_study()\fP
+  \fImatch_limit\fP      Limit on internal resource use
+  \fImatch_limit_recursion\fP  Limit on internal recursion depth
+  \fIcallout_data\fP     Opaque data passed back to callouts
+  \fItables\fP           Points to character tables or is NULL
+  \fImark\fP             For passing back a *MARK pointer
+  \fIexecutable_jit\fP   Opaque data from JIT compilation
+.sp
+The flag bits are PCRE_EXTRA_STUDY_DATA, PCRE_EXTRA_MATCH_LIMIT,
+PCRE_EXTRA_MATCH_LIMIT_RECURSION, PCRE_EXTRA_CALLOUT_DATA,
+PCRE_EXTRA_TABLES, PCRE_EXTRA_MARK and PCRE_EXTRA_EXECUTABLE_JIT.
+.P
+There is a complete description of the PCRE native API in the
+.\" HREF
+\fBpcreapi\fP
+.\"
+page and a description of the JIT API in the
+.\" HREF
+\fBpcrejit\fP
+.\"
+page.
diff --git a/pcre/doc/pcre_jit_stack_alloc.3 b/pcre/doc/pcre_jit_stack_alloc.3
new file mode 100644
index 00000000000..5d2a117882a
--- /dev/null
+++ b/pcre/doc/pcre_jit_stack_alloc.3
@@ -0,0 +1,45 @@
+.TH PCRE_JIT_STACK_ALLOC 3 "24 June 2012" "PCRE 8.30"
+.SH NAME
+PCRE - Perl-compatible regular expressions
+.SH SYNOPSIS
+.rs
+.sp
+.B #include 
+.PP
+.SM
+.B pcre_jit_stack *pcre_jit_stack_alloc(int \fIstartsize\fP,
+.ti +5n
+.B int \fImaxsize\fP);
+.PP
+.B pcre16_jit_stack *pcre16_jit_stack_alloc(int \fIstartsize\fP,
+.ti +5n
+.B int \fImaxsize\fP);
+.PP
+.B pcre32_jit_stack *pcre32_jit_stack_alloc(int \fIstartsize\fP,
+.ti +5n
+.B int \fImaxsize\fP);
+.
+.SH DESCRIPTION
+.rs
+.sp
+This function is used to create a stack for use by the code compiled by the JIT
+optimization of \fBpcre[16|32]_study()\fP. The arguments are a starting size for
+the stack, and a maximum size to which it is allowed to grow. The result can be
+passed to the JIT run-time code by \fBpcre[16|32]_assign_jit_stack()\fP, or that
+function can set up a callback for obtaining a stack. A maximum stack size of
+512K to 1M should be more than enough for any pattern. For more details, see
+the
+.\" HREF
+\fBpcrejit\fP
+.\"
+page.
+.P
+There is a complete description of the PCRE native API in the
+.\" HREF
+\fBpcreapi\fP
+.\"
+page and a description of the POSIX API in the
+.\" HREF
+\fBpcreposix\fP
+.\"
+page.
diff --git a/pcre/doc/pcre_jit_stack_free.3 b/pcre/doc/pcre_jit_stack_free.3
new file mode 100644
index 00000000000..494724e844f
--- /dev/null
+++ b/pcre/doc/pcre_jit_stack_free.3
@@ -0,0 +1,35 @@
+.TH PCRE_JIT_STACK_FREE 3 "24 June 2012" "PCRE 8.30"
+.SH NAME
+PCRE - Perl-compatible regular expressions
+.SH SYNOPSIS
+.rs
+.sp
+.B #include 
+.PP
+.SM
+.B void pcre_jit_stack_free(pcre_jit_stack *\fIstack\fP);
+.PP
+.B void pcre16_jit_stack_free(pcre16_jit_stack *\fIstack\fP);
+.PP
+.B void pcre32_jit_stack_free(pcre32_jit_stack *\fIstack\fP);
+.
+.SH DESCRIPTION
+.rs
+.sp
+This function is used to free a JIT stack that was created by
+\fBpcre[16|32]_jit_stack_alloc()\fP when it is no longer needed. For more details,
+see the
+.\" HREF
+\fBpcrejit\fP
+.\"
+page.
+.P
+There is a complete description of the PCRE native API in the
+.\" HREF
+\fBpcreapi\fP
+.\"
+page and a description of the POSIX API in the
+.\" HREF
+\fBpcreposix\fP
+.\"
+page.
diff --git a/pcre/doc/pcre_maketables.3 b/pcre/doc/pcre_maketables.3
new file mode 100644
index 00000000000..b2c3d23aa68
--- /dev/null
+++ b/pcre/doc/pcre_maketables.3
@@ -0,0 +1,33 @@
+.TH PCRE_MAKETABLES 3 "24 June 2012" "PCRE 8.30"
+.SH NAME
+PCRE - Perl-compatible regular expressions
+.SH SYNOPSIS
+.rs
+.sp
+.B #include 
+.PP
+.SM
+.B const unsigned char *pcre_maketables(void);
+.PP
+.B const unsigned char *pcre16_maketables(void);
+.PP
+.B const unsigned char *pcre32_maketables(void);
+.
+.SH DESCRIPTION
+.rs
+.sp
+This function builds a set of character tables for character values less than
+256. These can be passed to \fBpcre[16|32]_compile()\fP to override PCRE's
+internal, built-in tables (which were made by \fBpcre[16|32]_maketables()\fP when
+PCRE was compiled). You might want to do this if you are using a non-standard
+locale. The function yields a pointer to the tables.
+.P
+There is a complete description of the PCRE native API in the
+.\" HREF
+\fBpcreapi\fP
+.\"
+page and a description of the POSIX API in the
+.\" HREF
+\fBpcreposix\fP
+.\"
+page.
diff --git a/pcre/doc/pcre_pattern_to_host_byte_order.3 b/pcre/doc/pcre_pattern_to_host_byte_order.3
new file mode 100644
index 00000000000..4c306eef178
--- /dev/null
+++ b/pcre/doc/pcre_pattern_to_host_byte_order.3
@@ -0,0 +1,46 @@
+.TH PCRE_PATTERN_TO_HOST_BYTE_ORDER 3 "24 June 2012" "PCRE 8.30"
+.SH NAME
+PCRE - Perl-compatible regular expressions
+.SH SYNOPSIS
+.rs
+.sp
+.B #include 
+.PP
+.SM
+.B int pcre_pattern_to_host_byte_order(pcre *\fIcode\fP,
+.ti +5n
+.B pcre_extra *\fIextra\fP, const unsigned char *\fItables\fP);
+.PP
+.B int pcre16_pattern_to_host_byte_order(pcre16 *\fIcode\fP,
+.ti +5n
+.B pcre16_extra *\fIextra\fP, const unsigned char *\fItables\fP);
+.PP
+.B int pcre32_pattern_to_host_byte_order(pcre32 *\fIcode\fP,
+.ti +5n
+.B pcre32_extra *\fIextra\fP, const unsigned char *\fItables\fP);
+.
+.SH DESCRIPTION
+.rs
+.sp
+This function ensures that the bytes in 2-byte and 4-byte values in a compiled
+pattern are in the correct order for the current host. It is useful when a
+pattern that has been compiled on one host is transferred to another that might
+have different endianness. The arguments are:
+.sp
+  \fIcode\fP         A compiled regular expression
+  \fIextra\fP        Points to an associated \fBpcre[16|32]_extra\fP structure,
+                 or is NULL
+  \fItables\fP       Pointer to character tables, or NULL to
+                 set the built-in default
+.sp
+The result is 0 for success, a negative PCRE_ERROR_xxx value otherwise.
+.P
+There is a complete description of the PCRE native API in the
+.\" HREF
+\fBpcreapi\fP
+.\"
+page and a description of the POSIX API in the
+.\" HREF
+\fBpcreposix\fP
+.\"
+page.
diff --git a/pcre/doc/pcre_refcount.3 b/pcre/doc/pcre_refcount.3
new file mode 100644
index 00000000000..45a41fef6a0
--- /dev/null
+++ b/pcre/doc/pcre_refcount.3
@@ -0,0 +1,36 @@
+.TH PCRE_REFCOUNT 3 "24 June 2012" "PCRE 8.30"
+.SH NAME
+PCRE - Perl-compatible regular expressions
+.SH SYNOPSIS
+.rs
+.sp
+.B #include 
+.PP
+.SM
+.B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
+.PP
+.B int pcre16_refcount(pcre16 *\fIcode\fP, int \fIadjust\fP);
+.PP
+.B int pcre32_refcount(pcre32 *\fIcode\fP, int \fIadjust\fP);
+.
+.SH DESCRIPTION
+.rs
+.sp
+This function is used to maintain a reference count inside a data block that
+contains a compiled pattern. Its arguments are:
+.sp
+  \fIcode\fP                      Compiled regular expression
+  \fIadjust\fP                    Adjustment to reference value
+.sp
+The yield of the function is the adjusted reference value, which is constrained
+to lie between 0 and 65535.
+.P
+There is a complete description of the PCRE native API in the
+.\" HREF
+\fBpcreapi\fP
+.\"
+page and a description of the POSIX API in the
+.\" HREF
+\fBpcreposix\fP
+.\"
+page.
diff --git a/pcre/doc/pcre_study.3 b/pcre/doc/pcre_study.3
new file mode 100644
index 00000000000..1f2b465afff
--- /dev/null
+++ b/pcre/doc/pcre_study.3
@@ -0,0 +1,56 @@
+.TH PCRE_STUDY 3 " 24 June 2012" "PCRE 8.30"
+.SH NAME
+PCRE - Perl-compatible regular expressions
+.SH SYNOPSIS
+.rs
+.sp
+.B #include 
+.PP
+.SM
+.B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP,
+.ti +5n
+.B const char **\fIerrptr\fP);
+.PP
+.B pcre16_extra *pcre16_study(const pcre16 *\fIcode\fP, int \fIoptions\fP,
+.ti +5n
+.B const char **\fIerrptr\fP);
+.PP
+.B pcre32_extra *pcre32_study(const pcre32 *\fIcode\fP, int \fIoptions\fP,
+.ti +5n
+.B const char **\fIerrptr\fP);
+.
+.SH DESCRIPTION
+.rs
+.sp
+This function studies a compiled pattern, to see if additional information can
+be extracted that might speed up matching. Its arguments are:
+.sp
+  \fIcode\fP       A compiled regular expression
+  \fIoptions\fP    Options for \fBpcre[16|32]_study()\fP
+  \fIerrptr\fP     Where to put an error message
+.sp
+If the function succeeds, it returns a value that can be passed to
+\fBpcre[16|32]_exec()\fP or \fBpcre[16|32]_dfa_exec()\fP via their \fIextra\fP
+arguments.
+.P
+If the function returns NULL, either it could not find any additional
+information, or there was an error. You can tell the difference by looking at
+the error value. It is NULL in first case.
+.P
+The only option is PCRE_STUDY_JIT_COMPILE. It requests just-in-time compilation
+if possible. If PCRE has been compiled without JIT support, this option is
+ignored. See the
+.\" HREF
+\fBpcrejit\fP
+.\"
+page for further details.
+.P
+There is a complete description of the PCRE native API in the
+.\" HREF
+\fBpcreapi\fP
+.\"
+page and a description of the POSIX API in the
+.\" HREF
+\fBpcreposix\fP
+.\"
+page.
diff --git a/pcre/doc/pcre_utf16_to_host_byte_order.3 b/pcre/doc/pcre_utf16_to_host_byte_order.3
new file mode 100644
index 00000000000..8f0d2d4e8a9
--- /dev/null
+++ b/pcre/doc/pcre_utf16_to_host_byte_order.3
@@ -0,0 +1,46 @@
+.TH PCRE_UTF16_TO_HOST_BYTE_ORDER 3 "21 January 2012" "PCRE 8.30"
+.SH NAME
+PCRE - Perl-compatible regular expressions
+.SH SYNOPSIS
+.rs
+.sp
+.B #include 
+.PP
+.SM
+.B int pcre16_utf16_to_host_byte_order(PCRE_UCHAR16 *\fIoutput\fP,
+.ti +5n
+.B PCRE_SPTR16 \fIinput\fP, int \fIlength\fP, int *\fIhost_byte_order\fP,
+.ti +5n
+.B int \fIkeep_boms\fP);
+.
+.
+.SH DESCRIPTION
+.rs
+.sp
+This function, which exists only in the 16-bit library, converts a UTF-16
+string to the correct order for the current host, taking account of any byte
+order marks (BOMs) within the string. Its arguments are:
+.sp
+  \fIoutput\fP           pointer to output buffer, may be the same as \fIinput\fP
+  \fIinput\fP            pointer to input buffer
+  \fIlength\fP           number of 16-bit units in the input, or negative for
+                     a zero-terminated string
+  \fIhost_byte_order\fP  a NULL value or a non-zero value pointed to means
+                     start in host byte order
+  \fIkeep_boms\fP        if non-zero, BOMs are copied to the output string
+.sp
+The result of the function is the number of 16-bit units placed into the output
+buffer, including the zero terminator if the string was zero-terminated.
+.P
+If \fIhost_byte_order\fP is not NULL, it is set to indicate the byte order that
+is current at the end of the string.
+.P
+There is a complete description of the PCRE native API in the
+.\" HREF
+\fBpcreapi\fP
+.\"
+page and a description of the POSIX API in the
+.\" HREF
+\fBpcreposix\fP
+.\"
+page.
diff --git a/pcre/doc/pcre_utf32_to_host_byte_order.3 b/pcre/doc/pcre_utf32_to_host_byte_order.3
new file mode 100644
index 00000000000..fa4c1796b81
--- /dev/null
+++ b/pcre/doc/pcre_utf32_to_host_byte_order.3
@@ -0,0 +1,46 @@
+.TH PCRE_UTF32_TO_HOST_BYTE_ORDER 3 "24 June 2012" "PCRE 8.30"
+.SH NAME
+PCRE - Perl-compatible regular expressions
+.SH SYNOPSIS
+.rs
+.sp
+.B #include 
+.PP
+.SM
+.B int pcre32_utf32_to_host_byte_order(PCRE_UCHAR32 *\fIoutput\fP,
+.ti +5n
+.B PCRE_SPTR32 \fIinput\fP, int \fIlength\fP, int *\fIhost_byte_order\fP,
+.ti +5n
+.B int \fIkeep_boms\fP);
+.
+.
+.SH DESCRIPTION
+.rs
+.sp
+This function, which exists only in the 32-bit library, converts a UTF-32
+string to the correct order for the current host, taking account of any byte
+order marks (BOMs) within the string. Its arguments are:
+.sp
+  \fIoutput\fP           pointer to output buffer, may be the same as \fIinput\fP
+  \fIinput\fP            pointer to input buffer
+  \fIlength\fP           number of 32-bit units in the input, or negative for
+                     a zero-terminated string
+  \fIhost_byte_order\fP  a NULL value or a non-zero value pointed to means
+                     start in host byte order
+  \fIkeep_boms\fP        if non-zero, BOMs are copied to the output string
+.sp
+The result of the function is the number of 32-bit units placed into the output
+buffer, including the zero terminator if the string was zero-terminated.
+.P
+If \fIhost_byte_order\fP is not NULL, it is set to indicate the byte order that
+is current at the end of the string.
+.P
+There is a complete description of the PCRE native API in the
+.\" HREF
+\fBpcreapi\fP
+.\"
+page and a description of the POSIX API in the
+.\" HREF
+\fBpcreposix\fP
+.\"
+page.
diff --git a/pcre/doc/pcre_version.3 b/pcre/doc/pcre_version.3
new file mode 100644
index 00000000000..0f4973f9c7d
--- /dev/null
+++ b/pcre/doc/pcre_version.3
@@ -0,0 +1,31 @@
+.TH PCRE_VERSION 3 "24 June 2012" "PCRE 8.30"
+.SH NAME
+PCRE - Perl-compatible regular expressions
+.SH SYNOPSIS
+.rs
+.sp
+.B #include 
+.PP
+.SM
+.B const char *pcre_version(void);
+.PP
+.B const char *pcre16_version(void);
+.PP
+.B const char *pcre32_version(void);
+.
+.SH DESCRIPTION
+.rs
+.sp
+This function (even in the 16-bit and 32-bit libraries) returns a
+zero-terminated, 8-bit character string that gives the version number of the
+PCRE library and the date of its release.
+.P
+There is a complete description of the PCRE native API in the
+.\" HREF
+\fBpcreapi\fP
+.\"
+page and a description of the POSIX API in the
+.\" HREF
+\fBpcreposix\fP
+.\"
+page.
diff --git a/pcre/doc/pcreapi.3 b/pcre/doc/pcreapi.3
new file mode 100644
index 00000000000..4718c7fd71b
--- /dev/null
+++ b/pcre/doc/pcreapi.3
@@ -0,0 +1,2891 @@
+.TH PCREAPI 3 "12 May 2013" "PCRE 8.33"
+.SH NAME
+PCRE - Perl-compatible regular expressions
+.sp
+.B #include 
+.
+.
+.SH "PCRE NATIVE API BASIC FUNCTIONS"
+.rs
+.sp
+.SM
+.B pcre *pcre_compile(const char *\fIpattern\fP, int \fIoptions\fP,
+.ti +5n
+.B const char **\fIerrptr\fP, int *\fIerroffset\fP,
+.ti +5n
+.B const unsigned char *\fItableptr\fP);
+.PP
+.B pcre *pcre_compile2(const char *\fIpattern\fP, int \fIoptions\fP,
+.ti +5n
+.B int *\fIerrorcodeptr\fP,
+.ti +5n
+.B const char **\fIerrptr\fP, int *\fIerroffset\fP,
+.ti +5n
+.B const unsigned char *\fItableptr\fP);
+.PP
+.B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP,
+.ti +5n
+.B const char **\fIerrptr\fP);
+.PP
+.B void pcre_free_study(pcre_extra *\fIextra\fP);
+.PP
+.B int pcre_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
+.ti +5n
+.B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
+.ti +5n
+.B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);
+.PP
+.B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
+.ti +5n
+.B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
+.ti +5n
+.B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
+.ti +5n
+.B int *\fIworkspace\fP, int \fIwscount\fP);
+.
+.
+.SH "PCRE NATIVE API STRING EXTRACTION FUNCTIONS"
+.rs
+.sp
+.B int pcre_copy_named_substring(const pcre *\fIcode\fP,
+.ti +5n
+.B const char *\fIsubject\fP, int *\fIovector\fP,
+.ti +5n
+.B int \fIstringcount\fP, const char *\fIstringname\fP,
+.ti +5n
+.B char *\fIbuffer\fP, int \fIbuffersize\fP);
+.PP
+.B int pcre_copy_substring(const char *\fIsubject\fP, int *\fIovector\fP,
+.ti +5n
+.B int \fIstringcount\fP, int \fIstringnumber\fP, char *\fIbuffer\fP,
+.ti +5n
+.B int \fIbuffersize\fP);
+.PP
+.B int pcre_get_named_substring(const pcre *\fIcode\fP,
+.ti +5n
+.B const char *\fIsubject\fP, int *\fIovector\fP,
+.ti +5n
+.B int \fIstringcount\fP, const char *\fIstringname\fP,
+.ti +5n
+.B const char **\fIstringptr\fP);
+.PP
+.B int pcre_get_stringnumber(const pcre *\fIcode\fP,
+.ti +5n
+.B const char *\fIname\fP);
+.PP
+.B int pcre_get_stringtable_entries(const pcre *\fIcode\fP,
+.ti +5n
+.B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
+.PP
+.B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,
+.ti +5n
+.B int \fIstringcount\fP, int \fIstringnumber\fP,
+.ti +5n
+.B const char **\fIstringptr\fP);
+.PP
+.B int pcre_get_substring_list(const char *\fIsubject\fP,
+.ti +5n
+.B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"
+.PP
+.B void pcre_free_substring(const char *\fIstringptr\fP);
+.PP
+.B void pcre_free_substring_list(const char **\fIstringptr\fP);
+.
+.
+.SH "PCRE NATIVE API AUXILIARY FUNCTIONS"
+.rs
+.sp
+.B int pcre_jit_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
+.ti +5n
+.B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
+.ti +5n
+.B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
+.ti +5n
+.B pcre_jit_stack *\fIjstack\fP);
+.PP
+.B pcre_jit_stack *pcre_jit_stack_alloc(int \fIstartsize\fP, int \fImaxsize\fP);
+.PP
+.B void pcre_jit_stack_free(pcre_jit_stack *\fIstack\fP);
+.PP
+.B void pcre_assign_jit_stack(pcre_extra *\fIextra\fP,
+.ti +5n
+.B pcre_jit_callback \fIcallback\fP, void *\fIdata\fP);
+.PP
+.B const unsigned char *pcre_maketables(void);
+.PP
+.B int pcre_fullinfo(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
+.ti +5n
+.B int \fIwhat\fP, void *\fIwhere\fP);
+.PP
+.B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
+.PP
+.B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);
+.PP
+.B const char *pcre_version(void);
+.PP
+.B int pcre_pattern_to_host_byte_order(pcre *\fIcode\fP,
+.ti +5n
+.B pcre_extra *\fIextra\fP, const unsigned char *\fItables\fP);
+.
+.
+.SH "PCRE NATIVE API INDIRECTED FUNCTIONS"
+.rs
+.sp
+.B void *(*pcre_malloc)(size_t);
+.PP
+.B void (*pcre_free)(void *);
+.PP
+.B void *(*pcre_stack_malloc)(size_t);
+.PP
+.B void (*pcre_stack_free)(void *);
+.PP
+.B int (*pcre_callout)(pcre_callout_block *);
+.
+.
+.SH "PCRE 8-BIT, 16-BIT, AND 32-BIT LIBRARIES"
+.rs
+.sp
+As well as support for 8-bit character strings, PCRE also supports 16-bit
+strings (from release 8.30) and 32-bit strings (from release 8.32), by means of
+two additional libraries. They can be built as well as, or instead of, the
+8-bit library. To avoid too much complication, this document describes the
+8-bit versions of the functions, with only occasional references to the 16-bit
+and 32-bit libraries.
+.P
+The 16-bit and 32-bit functions operate in the same way as their 8-bit
+counterparts; they just use different data types for their arguments and
+results, and their names start with \fBpcre16_\fP or \fBpcre32_\fP instead of
+\fBpcre_\fP. For every option that has UTF8 in its name (for example,
+PCRE_UTF8), there are corresponding 16-bit and 32-bit names with UTF8 replaced
+by UTF16 or UTF32, respectively. This facility is in fact just cosmetic; the
+16-bit and 32-bit option names define the same bit values.
+.P
+References to bytes and UTF-8 in this document should be read as references to
+16-bit data units and UTF-16 when using the 16-bit library, or 32-bit data
+units and UTF-32 when using the 32-bit library, unless specified otherwise.
+More details of the specific differences for the 16-bit and 32-bit libraries
+are given in the
+.\" HREF
+\fBpcre16\fP
+.\"
+and
+.\" HREF
+\fBpcre32\fP
+.\"
+pages.
+.
+.
+.SH "PCRE API OVERVIEW"
+.rs
+.sp
+PCRE has its own native API, which is described in this document. There are
+also some wrapper functions (for the 8-bit library only) that correspond to the
+POSIX regular expression API, but they do not give access to all the
+functionality. They are described in the
+.\" HREF
+\fBpcreposix\fP
+.\"
+documentation. Both of these APIs define a set of C function calls. A C++
+wrapper (again for the 8-bit library only) is also distributed with PCRE. It is
+documented in the
+.\" HREF
+\fBpcrecpp\fP
+.\"
+page.
+.P
+The native API C function prototypes are defined in the header file
+\fBpcre.h\fP, and on Unix-like systems the (8-bit) library itself is called
+\fBlibpcre\fP. It can normally be accessed by adding \fB-lpcre\fP to the
+command for linking an application that uses PCRE. The header file defines the
+macros PCRE_MAJOR and PCRE_MINOR to contain the major and minor release numbers
+for the library. Applications can use these to include support for different
+releases of PCRE.
+.P
+In a Windows environment, if you want to statically link an application program
+against a non-dll \fBpcre.a\fP file, you must define PCRE_STATIC before
+including \fBpcre.h\fP or \fBpcrecpp.h\fP, because otherwise the
+\fBpcre_malloc()\fP and \fBpcre_free()\fP exported functions will be declared
+\fB__declspec(dllimport)\fP, with unwanted results.
+.P
+The functions \fBpcre_compile()\fP, \fBpcre_compile2()\fP, \fBpcre_study()\fP,
+and \fBpcre_exec()\fP are used for compiling and matching regular expressions
+in a Perl-compatible manner. A sample program that demonstrates the simplest
+way of using them is provided in the file called \fIpcredemo.c\fP in the PCRE
+source distribution. A listing of this program is given in the
+.\" HREF
+\fBpcredemo\fP
+.\"
+documentation, and the
+.\" HREF
+\fBpcresample\fP
+.\"
+documentation describes how to compile and run it.
+.P
+Just-in-time compiler support is an optional feature of PCRE that can be built
+in appropriate hardware environments. It greatly speeds up the matching
+performance of many patterns. Simple programs can easily request that it be
+used if available, by setting an option that is ignored when it is not
+relevant. More complicated programs might need to make use of the functions
+\fBpcre_jit_stack_alloc()\fP, \fBpcre_jit_stack_free()\fP, and
+\fBpcre_assign_jit_stack()\fP in order to control the JIT code's memory usage.
+.P
+From release 8.32 there is also a direct interface for JIT execution, which
+gives improved performance. The JIT-specific functions are discussed in the
+.\" HREF
+\fBpcrejit\fP
+.\"
+documentation.
+.P
+A second matching function, \fBpcre_dfa_exec()\fP, which is not
+Perl-compatible, is also provided. This uses a different algorithm for the
+matching. The alternative algorithm finds all possible matches (at a given
+point in the subject), and scans the subject just once (unless there are
+lookbehind assertions). However, this algorithm does not return captured
+substrings. A description of the two matching algorithms and their advantages
+and disadvantages is given in the
+.\" HREF
+\fBpcrematching\fP
+.\"
+documentation.
+.P
+In addition to the main compiling and matching functions, there are convenience
+functions for extracting captured substrings from a subject string that is
+matched by \fBpcre_exec()\fP. They are:
+.sp
+  \fBpcre_copy_substring()\fP
+  \fBpcre_copy_named_substring()\fP
+  \fBpcre_get_substring()\fP
+  \fBpcre_get_named_substring()\fP
+  \fBpcre_get_substring_list()\fP
+  \fBpcre_get_stringnumber()\fP
+  \fBpcre_get_stringtable_entries()\fP
+.sp
+\fBpcre_free_substring()\fP and \fBpcre_free_substring_list()\fP are also
+provided, to free the memory used for extracted strings.
+.P
+The function \fBpcre_maketables()\fP is used to build a set of character tables
+in the current locale for passing to \fBpcre_compile()\fP, \fBpcre_exec()\fP,
+or \fBpcre_dfa_exec()\fP. This is an optional facility that is provided for
+specialist use. Most commonly, no special tables are passed, in which case
+internal tables that are generated when PCRE is built are used.
+.P
+The function \fBpcre_fullinfo()\fP is used to find out information about a
+compiled pattern. The function \fBpcre_version()\fP returns a pointer to a
+string containing the version of PCRE and its date of release.
+.P
+The function \fBpcre_refcount()\fP maintains a reference count in a data block
+containing a compiled pattern. This is provided for the benefit of
+object-oriented applications.
+.P
+The global variables \fBpcre_malloc\fP and \fBpcre_free\fP initially contain
+the entry points of the standard \fBmalloc()\fP and \fBfree()\fP functions,
+respectively. PCRE calls the memory management functions via these variables,
+so a calling program can replace them if it wishes to intercept the calls. This
+should be done before calling any PCRE functions.
+.P
+The global variables \fBpcre_stack_malloc\fP and \fBpcre_stack_free\fP are also
+indirections to memory management functions. These special functions are used
+only when PCRE is compiled to use the heap for remembering data, instead of
+recursive function calls, when running the \fBpcre_exec()\fP function. See the
+.\" HREF
+\fBpcrebuild\fP
+.\"
+documentation for details of how to do this. It is a non-standard way of
+building PCRE, for use in environments that have limited stacks. Because of the
+greater use of memory management, it runs more slowly. Separate functions are
+provided so that special-purpose external code can be used for this case. When
+used, these functions are always called in a stack-like manner (last obtained,
+first freed), and always for memory blocks of the same size. There is a
+discussion about PCRE's stack usage in the
+.\" HREF
+\fBpcrestack\fP
+.\"
+documentation.
+.P
+The global variable \fBpcre_callout\fP initially contains NULL. It can be set
+by the caller to a "callout" function, which PCRE will then call at specified
+points during a matching operation. Details are given in the
+.\" HREF
+\fBpcrecallout\fP
+.\"
+documentation.
+.
+.
+.\" HTML 
+.SH NEWLINES
+.rs
+.sp
+PCRE supports five different conventions for indicating line breaks in
+strings: a single CR (carriage return) character, a single LF (linefeed)
+character, the two-character sequence CRLF, any of the three preceding, or any
+Unicode newline sequence. The Unicode newline sequences are the three just
+mentioned, plus the single characters VT (vertical tab, U+000B), FF (form feed,
+U+000C), NEL (next line, U+0085), LS (line separator, U+2028), and PS
+(paragraph separator, U+2029).
+.P
+Each of the first three conventions is used by at least one operating system as
+its standard newline sequence. When PCRE is built, a default can be specified.
+The default default is LF, which is the Unix standard. When PCRE is run, the
+default can be overridden, either when a pattern is compiled, or when it is
+matched.
+.P
+At compile time, the newline convention can be specified by the \fIoptions\fP
+argument of \fBpcre_compile()\fP, or it can be specified by special text at the
+start of the pattern itself; this overrides any other settings. See the
+.\" HREF
+\fBpcrepattern\fP
+.\"
+page for details of the special character sequences.
+.P
+In the PCRE documentation the word "newline" is used to mean "the character or
+pair of characters that indicate a line break". The choice of newline
+convention affects the handling of the dot, circumflex, and dollar
+metacharacters, the handling of #-comments in /x mode, and, when CRLF is a
+recognized line ending sequence, the match position advancement for a
+non-anchored pattern. There is more detail about this in the
+.\" HTML 
+.\" 
+section on \fBpcre_exec()\fP options
+.\"
+below.
+.P
+The choice of newline convention does not affect the interpretation of
+the \en or \er escape sequences, nor does it affect what \eR matches, which is
+controlled in a similar way, but by separate options.
+.
+.
+.SH MULTITHREADING
+.rs
+.sp
+The PCRE functions can be used in multi-threading applications, with the
+proviso that the memory management functions pointed to by \fBpcre_malloc\fP,
+\fBpcre_free\fP, \fBpcre_stack_malloc\fP, and \fBpcre_stack_free\fP, and the
+callout function pointed to by \fBpcre_callout\fP, are shared by all threads.
+.P
+The compiled form of a regular expression is not altered during matching, so
+the same compiled pattern can safely be used by several threads at once.
+.P
+If the just-in-time optimization feature is being used, it needs separate
+memory stack areas for each thread. See the
+.\" HREF
+\fBpcrejit\fP
+.\"
+documentation for more details.
+.
+.
+.SH "SAVING PRECOMPILED PATTERNS FOR LATER USE"
+.rs
+.sp
+The compiled form of a regular expression can be saved and re-used at a later
+time, possibly by a different program, and even on a host other than the one on
+which it was compiled. Details are given in the
+.\" HREF
+\fBpcreprecompile\fP
+.\"
+documentation, which includes a description of the
+\fBpcre_pattern_to_host_byte_order()\fP function. However, compiling a regular
+expression with one version of PCRE for use with a different version is not
+guaranteed to work and may cause crashes.
+.
+.
+.SH "CHECKING BUILD-TIME OPTIONS"
+.rs
+.sp
+.B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);
+.PP
+The function \fBpcre_config()\fP makes it possible for a PCRE client to
+discover which optional features have been compiled into the PCRE library. The
+.\" HREF
+\fBpcrebuild\fP
+.\"
+documentation has more details about these optional features.
+.P
+The first argument for \fBpcre_config()\fP is an integer, specifying which
+information is required; the second argument is a pointer to a variable into
+which the information is placed. The returned value is zero on success, or the
+negative error code PCRE_ERROR_BADOPTION if the value in the first argument is
+not recognized. The following information is available:
+.sp
+  PCRE_CONFIG_UTF8
+.sp
+The output is an integer that is set to one if UTF-8 support is available;
+otherwise it is set to zero. This value should normally be given to the 8-bit
+version of this function, \fBpcre_config()\fP. If it is given to the 16-bit
+or 32-bit version of this function, the result is PCRE_ERROR_BADOPTION.
+.sp
+  PCRE_CONFIG_UTF16
+.sp
+The output is an integer that is set to one if UTF-16 support is available;
+otherwise it is set to zero. This value should normally be given to the 16-bit
+version of this function, \fBpcre16_config()\fP. If it is given to the 8-bit
+or 32-bit version of this function, the result is PCRE_ERROR_BADOPTION.
+.sp
+  PCRE_CONFIG_UTF32
+.sp
+The output is an integer that is set to one if UTF-32 support is available;
+otherwise it is set to zero. This value should normally be given to the 32-bit
+version of this function, \fBpcre32_config()\fP. If it is given to the 8-bit
+or 16-bit version of this function, the result is PCRE_ERROR_BADOPTION.
+.sp
+  PCRE_CONFIG_UNICODE_PROPERTIES
+.sp
+The output is an integer that is set to one if support for Unicode character
+properties is available; otherwise it is set to zero.
+.sp
+  PCRE_CONFIG_JIT
+.sp
+The output is an integer that is set to one if support for just-in-time
+compiling is available; otherwise it is set to zero.
+.sp
+  PCRE_CONFIG_JITTARGET
+.sp
+The output is a pointer to a zero-terminated "const char *" string. If JIT
+support is available, the string contains the name of the architecture for
+which the JIT compiler is configured, for example "x86 32bit (little endian +
+unaligned)". If JIT support is not available, the result is NULL.
+.sp
+  PCRE_CONFIG_NEWLINE
+.sp
+The output is an integer whose value specifies the default character sequence
+that is recognized as meaning "newline". The values that are supported in
+ASCII/Unicode environments are: 10 for LF, 13 for CR, 3338 for CRLF, -2 for
+ANYCRLF, and -1 for ANY. In EBCDIC environments, CR, ANYCRLF, and ANY yield the
+same values. However, the value for LF is normally 21, though some EBCDIC
+environments use 37. The corresponding values for CRLF are 3349 and 3365. The
+default should normally correspond to the standard sequence for your operating
+system.
+.sp
+  PCRE_CONFIG_BSR
+.sp
+The output is an integer whose value indicates what character sequences the \eR
+escape sequence matches by default. A value of 0 means that \eR matches any
+Unicode line ending sequence; a value of 1 means that \eR matches only CR, LF,
+or CRLF. The default can be overridden when a pattern is compiled or matched.
+.sp
+  PCRE_CONFIG_LINK_SIZE
+.sp
+The output is an integer that contains the number of bytes used for internal
+linkage in compiled regular expressions. For the 8-bit library, the value can
+be 2, 3, or 4. For the 16-bit library, the value is either 2 or 4 and is still
+a number of bytes. For the 32-bit library, the value is either 2 or 4 and is
+still a number of bytes. The default value of 2 is sufficient for all but the
+most massive patterns, since it allows the compiled pattern to be up to 64K in
+size. Larger values allow larger regular expressions to be compiled, at the
+expense of slower matching.
+.sp
+  PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
+.sp
+The output is an integer that contains the threshold above which the POSIX
+interface uses \fBmalloc()\fP for output vectors. Further details are given in
+the
+.\" HREF
+\fBpcreposix\fP
+.\"
+documentation.
+.sp
+  PCRE_CONFIG_MATCH_LIMIT
+.sp
+The output is a long integer that gives the default limit for the number of
+internal matching function calls in a \fBpcre_exec()\fP execution. Further
+details are given with \fBpcre_exec()\fP below.
+.sp
+  PCRE_CONFIG_MATCH_LIMIT_RECURSION
+.sp
+The output is a long integer that gives the default limit for the depth of
+recursion when calling the internal matching function in a \fBpcre_exec()\fP
+execution. Further details are given with \fBpcre_exec()\fP below.
+.sp
+  PCRE_CONFIG_STACKRECURSE
+.sp
+The output is an integer that is set to one if internal recursion when running
+\fBpcre_exec()\fP is implemented by recursive function calls that use the stack
+to remember their state. This is the usual way that PCRE is compiled. The
+output is zero if PCRE was compiled to use blocks of data on the heap instead
+of recursive function calls. In this case, \fBpcre_stack_malloc\fP and
+\fBpcre_stack_free\fP are called to manage memory blocks on the heap, thus
+avoiding the use of the stack.
+.
+.
+.SH "COMPILING A PATTERN"
+.rs
+.sp
+.B pcre *pcre_compile(const char *\fIpattern\fP, int \fIoptions\fP,
+.ti +5n
+.B const char **\fIerrptr\fP, int *\fIerroffset\fP,
+.ti +5n
+.B const unsigned char *\fItableptr\fP);
+.sp
+.B pcre *pcre_compile2(const char *\fIpattern\fP, int \fIoptions\fP,
+.ti +5n
+.B int *\fIerrorcodeptr\fP,
+.ti +5n
+.B const char **\fIerrptr\fP, int *\fIerroffset\fP,
+.ti +5n
+.B const unsigned char *\fItableptr\fP);
+.P
+Either of the functions \fBpcre_compile()\fP or \fBpcre_compile2()\fP can be
+called to compile a pattern into an internal form. The only difference between
+the two interfaces is that \fBpcre_compile2()\fP has an additional argument,
+\fIerrorcodeptr\fP, via which a numerical error code can be returned. To avoid
+too much repetition, we refer just to \fBpcre_compile()\fP below, but the
+information applies equally to \fBpcre_compile2()\fP.
+.P
+The pattern is a C string terminated by a binary zero, and is passed in the
+\fIpattern\fP argument. A pointer to a single block of memory that is obtained
+via \fBpcre_malloc\fP is returned. This contains the compiled code and related
+data. The \fBpcre\fP type is defined for the returned block; this is a typedef
+for a structure whose contents are not externally defined. It is up to the
+caller to free the memory (via \fBpcre_free\fP) when it is no longer required.
+.P
+Although the compiled code of a PCRE regex is relocatable, that is, it does not
+depend on memory location, the complete \fBpcre\fP data block is not
+fully relocatable, because it may contain a copy of the \fItableptr\fP
+argument, which is an address (see below).
+.P
+The \fIoptions\fP argument contains various bit settings that affect the
+compilation. It should be zero if no options are required. The available
+options are described below. Some of them (in particular, those that are
+compatible with Perl, but some others as well) can also be set and unset from
+within the pattern (see the detailed description in the
+.\" HREF
+\fBpcrepattern\fP
+.\"
+documentation). For those options that can be different in different parts of
+the pattern, the contents of the \fIoptions\fP argument specifies their
+settings at the start of compilation and execution. The PCRE_ANCHORED,
+PCRE_BSR_\fIxxx\fP, PCRE_NEWLINE_\fIxxx\fP, PCRE_NO_UTF8_CHECK, and
+PCRE_NO_START_OPTIMIZE options can be set at the time of matching as well as at
+compile time.
+.P
+If \fIerrptr\fP is NULL, \fBpcre_compile()\fP returns NULL immediately.
+Otherwise, if compilation of a pattern fails, \fBpcre_compile()\fP returns
+NULL, and sets the variable pointed to by \fIerrptr\fP to point to a textual
+error message. This is a static string that is part of the library. You must
+not try to free it. Normally, the offset from the start of the pattern to the
+data unit that was being processed when the error was discovered is placed in
+the variable pointed to by \fIerroffset\fP, which must not be NULL (if it is,
+an immediate error is given). However, for an invalid UTF-8 or UTF-16 string,
+the offset is that of the first data unit of the failing character.
+.P
+Some errors are not detected until the whole pattern has been scanned; in these
+cases, the offset passed back is the length of the pattern. Note that the
+offset is in data units, not characters, even in a UTF mode. It may sometimes
+point into the middle of a UTF-8 or UTF-16 character.
+.P
+If \fBpcre_compile2()\fP is used instead of \fBpcre_compile()\fP, and the
+\fIerrorcodeptr\fP argument is not NULL, a non-zero error code number is
+returned via this argument in the event of an error. This is in addition to the
+textual error message. Error codes and messages are listed below.
+.P
+If the final argument, \fItableptr\fP, is NULL, PCRE uses a default set of
+character tables that are built when PCRE is compiled, using the default C
+locale. Otherwise, \fItableptr\fP must be an address that is the result of a
+call to \fBpcre_maketables()\fP. This value is stored with the compiled
+pattern, and used again by \fBpcre_exec()\fP, unless another table pointer is
+passed to it. For more discussion, see the section on locale support below.
+.P
+This code fragment shows a typical straightforward call to \fBpcre_compile()\fP:
+.sp
+  pcre *re;
+  const char *error;
+  int erroffset;
+  re = pcre_compile(
+    "^A.*Z",          /* the pattern */
+    0,                /* default options */
+    &error,           /* for error message */
+    &erroffset,       /* for error offset */
+    NULL);            /* use default character tables */
+.sp
+The following names for option bits are defined in the \fBpcre.h\fP header
+file:
+.sp
+  PCRE_ANCHORED
+.sp
+If this bit is set, the pattern is forced to be "anchored", that is, it is
+constrained to match only at the first matching point in the string that is
+being searched (the "subject string"). This effect can also be achieved by
+appropriate constructs in the pattern itself, which is the only way to do it in
+Perl.
+.sp
+  PCRE_AUTO_CALLOUT
+.sp
+If this bit is set, \fBpcre_compile()\fP automatically inserts callout items,
+all with number 255, before each pattern item. For discussion of the callout
+facility, see the
+.\" HREF
+\fBpcrecallout\fP
+.\"
+documentation.
+.sp
+  PCRE_BSR_ANYCRLF
+  PCRE_BSR_UNICODE
+.sp
+These options (which are mutually exclusive) control what the \eR escape
+sequence matches. The choice is either to match only CR, LF, or CRLF, or to
+match any Unicode newline sequence. The default is specified when PCRE is
+built. It can be overridden from within the pattern, or by setting an option
+when a compiled pattern is matched.
+.sp
+  PCRE_CASELESS
+.sp
+If this bit is set, letters in the pattern match both upper and lower case
+letters. It is equivalent to Perl's /i option, and it can be changed within a
+pattern by a (?i) option setting. In UTF-8 mode, PCRE always understands the
+concept of case for characters whose values are less than 128, so caseless
+matching is always possible. For characters with higher values, the concept of
+case is supported if PCRE is compiled with Unicode property support, but not
+otherwise. If you want to use caseless matching for characters 128 and above,
+you must ensure that PCRE is compiled with Unicode property support as well as
+with UTF-8 support.
+.sp
+  PCRE_DOLLAR_ENDONLY
+.sp
+If this bit is set, a dollar metacharacter in the pattern matches only at the
+end of the subject string. Without this option, a dollar also matches
+immediately before a newline at the end of the string (but not before any other
+newlines). The PCRE_DOLLAR_ENDONLY option is ignored if PCRE_MULTILINE is set.
+There is no equivalent to this option in Perl, and no way to set it within a
+pattern.
+.sp
+  PCRE_DOTALL
+.sp
+If this bit is set, a dot metacharacter in the pattern matches a character of
+any value, including one that indicates a newline. However, it only ever
+matches one character, even if newlines are coded as CRLF. Without this option,
+a dot does not match when the current position is at a newline. This option is
+equivalent to Perl's /s option, and it can be changed within a pattern by a
+(?s) option setting. A negative class such as [^a] always matches newline
+characters, independent of the setting of this option.
+.sp
+  PCRE_DUPNAMES
+.sp
+If this bit is set, names used to identify capturing subpatterns need not be
+unique. This can be helpful for certain types of pattern when it is known that
+only one instance of the named subpattern can ever be matched. There are more
+details of named subpatterns below; see also the
+.\" HREF
+\fBpcrepattern\fP
+.\"
+documentation.
+.sp
+  PCRE_EXTENDED
+.sp
+If this bit is set, white space data characters in the pattern are totally
+ignored except when escaped or inside a character class. White space does not
+include the VT character (code 11). In addition, characters between an
+unescaped # outside a character class and the next newline, inclusive, are also
+ignored. This is equivalent to Perl's /x option, and it can be changed within a
+pattern by a (?x) option setting.
+.P
+Which characters are interpreted as newlines is controlled by the options
+passed to \fBpcre_compile()\fP or by a special sequence at the start of the
+pattern, as described in the section entitled
+.\" HTML 
+.\" 
+"Newline conventions"
+.\"
+in the \fBpcrepattern\fP documentation. Note that the end of this type of
+comment is a literal newline sequence in the pattern; escape sequences that
+happen to represent a newline do not count.
+.P
+This option makes it possible to include comments inside complicated patterns.
+Note, however, that this applies only to data characters. White space characters
+may never appear within special character sequences in a pattern, for example
+within the sequence (?( that introduces a conditional subpattern.
+.sp
+  PCRE_EXTRA
+.sp
+This option was invented in order to turn on additional functionality of PCRE
+that is incompatible with Perl, but it is currently of very little use. When
+set, any backslash in a pattern that is followed by a letter that has no
+special meaning causes an error, thus reserving these combinations for future
+expansion. By default, as in Perl, a backslash followed by a letter with no
+special meaning is treated as a literal. (Perl can, however, be persuaded to
+give an error for this, by running it with the -w option.) There are at present
+no other features controlled by this option. It can also be set by a (?X)
+option setting within a pattern.
+.sp
+  PCRE_FIRSTLINE
+.sp
+If this option is set, an unanchored pattern is required to match before or at
+the first newline in the subject string, though the matched text may continue
+over the newline.
+.sp
+  PCRE_JAVASCRIPT_COMPAT
+.sp
+If this option is set, PCRE's behaviour is changed in some ways so that it is
+compatible with JavaScript rather than Perl. The changes are as follows:
+.P
+(1) A lone closing square bracket in a pattern causes a compile-time error,
+because this is illegal in JavaScript (by default it is treated as a data
+character). Thus, the pattern AB]CD becomes illegal when this option is set.
+.P
+(2) At run time, a back reference to an unset subpattern group matches an empty
+string (by default this causes the current matching alternative to fail). A
+pattern such as (\e1)(a) succeeds when this option is set (assuming it can find
+an "a" in the subject), whereas it fails by default, for Perl compatibility.
+.P
+(3) \eU matches an upper case "U" character; by default \eU causes a compile
+time error (Perl uses \eU to upper case subsequent characters).
+.P
+(4) \eu matches a lower case "u" character unless it is followed by four
+hexadecimal digits, in which case the hexadecimal number defines the code point
+to match. By default, \eu causes a compile time error (Perl uses it to upper
+case the following character).
+.P
+(5) \ex matches a lower case "x" character unless it is followed by two
+hexadecimal digits, in which case the hexadecimal number defines the code point
+to match. By default, as in Perl, a hexadecimal number is always expected after
+\ex, but it may have zero, one, or two digits (so, for example, \exz matches a
+binary zero character followed by z).
+.sp
+  PCRE_MULTILINE
+.sp
+By default, for the purposes of matching "start of line" and "end of line",
+PCRE treats the subject string as consisting of a single line of characters,
+even if it actually contains newlines. The "start of line" metacharacter (^)
+matches only at the start of the string, and the "end of line" metacharacter
+($) matches only at the end of the string, or before a terminating newline
+(except when PCRE_DOLLAR_ENDONLY is set). Note, however, that unless
+PCRE_DOTALL is set, the "any character" metacharacter (.) does not match at a
+newline. This behaviour (for ^, $, and dot) is the same as Perl.
+.P
+When PCRE_MULTILINE it is set, the "start of line" and "end of line" constructs
+match immediately following or immediately before internal newlines in the
+subject string, respectively, as well as at the very start and end. This is
+equivalent to Perl's /m option, and it can be changed within a pattern by a
+(?m) option setting. If there are no newlines in a subject string, or no
+occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.
+.sp
+  PCRE_NEVER_UTF
+.sp
+This option locks out interpretation of the pattern as UTF-8 (or UTF-16 or
+UTF-32 in the 16-bit and 32-bit libraries). In particular, it prevents the
+creator of the pattern from switching to UTF interpretation by starting the
+pattern with (*UTF). This may be useful in applications that process patterns
+from external sources. The combination of PCRE_UTF8 and PCRE_NEVER_UTF also
+causes an error.
+.sp
+  PCRE_NEWLINE_CR
+  PCRE_NEWLINE_LF
+  PCRE_NEWLINE_CRLF
+  PCRE_NEWLINE_ANYCRLF
+  PCRE_NEWLINE_ANY
+.sp
+These options override the default newline definition that was chosen when PCRE
+was built. Setting the first or the second specifies that a newline is
+indicated by a single character (CR or LF, respectively). Setting
+PCRE_NEWLINE_CRLF specifies that a newline is indicated by the two-character
+CRLF sequence. Setting PCRE_NEWLINE_ANYCRLF specifies that any of the three
+preceding sequences should be recognized. Setting PCRE_NEWLINE_ANY specifies
+that any Unicode newline sequence should be recognized.
+.P
+In an ASCII/Unicode environment, the Unicode newline sequences are the three
+just mentioned, plus the single characters VT (vertical tab, U+000B), FF (form
+feed, U+000C), NEL (next line, U+0085), LS (line separator, U+2028), and PS
+(paragraph separator, U+2029). For the 8-bit library, the last two are
+recognized only in UTF-8 mode.
+.P
+When PCRE is compiled to run in an EBCDIC (mainframe) environment, the code for
+CR is 0x0d, the same as ASCII. However, the character code for LF is normally
+0x15, though in some EBCDIC environments 0x25 is used. Whichever of these is
+not LF is made to correspond to Unicode's NEL character. EBCDIC codes are all
+less than 256. For more details, see the
+.\" HREF
+\fBpcrebuild\fP
+.\"
+documentation.
+.P
+The newline setting in the options word uses three bits that are treated
+as a number, giving eight possibilities. Currently only six are used (default
+plus the five values above). This means that if you set more than one newline
+option, the combination may or may not be sensible. For example,
+PCRE_NEWLINE_CR with PCRE_NEWLINE_LF is equivalent to PCRE_NEWLINE_CRLF, but
+other combinations may yield unused numbers and cause an error.
+.P
+The only time that a line break in a pattern is specially recognized when
+compiling is when PCRE_EXTENDED is set. CR and LF are white space characters,
+and so are ignored in this mode. Also, an unescaped # outside a character class
+indicates a comment that lasts until after the next line break sequence. In
+other circumstances, line break sequences in patterns are treated as literal
+data.
+.P
+The newline option that is set at compile time becomes the default that is used
+for \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, but it can be overridden.
+.sp
+  PCRE_NO_AUTO_CAPTURE
+.sp
+If this option is set, it disables the use of numbered capturing parentheses in
+the pattern. Any opening parenthesis that is not followed by ? behaves as if it
+were followed by ?: but named parentheses can still be used for capturing (and
+they acquire numbers in the usual way). There is no equivalent of this option
+in Perl.
+.sp
+  PCRE_NO_START_OPTIMIZE
+.sp
+This is an option that acts at matching time; that is, it is really an option
+for \fBpcre_exec()\fP or \fBpcre_dfa_exec()\fP. If it is set at compile time,
+it is remembered with the compiled pattern and assumed at matching time. This
+is necessary if you want to use JIT execution, because the JIT compiler needs
+to know whether or not this option is set. For details see the discussion of
+PCRE_NO_START_OPTIMIZE
+.\" HTML 
+.\" 
+below.
+.\"
+.sp
+  PCRE_UCP
+.sp
+This option changes the way PCRE processes \eB, \eb, \eD, \ed, \eS, \es, \eW,
+\ew, and some of the POSIX character classes. By default, only ASCII characters
+are recognized, but if PCRE_UCP is set, Unicode properties are used instead to
+classify characters. More details are given in the section on
+.\" HTML 
+.\" 
+generic character types
+.\"
+in the
+.\" HREF
+\fBpcrepattern\fP
+.\"
+page. If you set PCRE_UCP, matching one of the items it affects takes much
+longer. The option is available only if PCRE has been compiled with Unicode
+property support.
+.sp
+  PCRE_UNGREEDY
+.sp
+This option inverts the "greediness" of the quantifiers so that they are not
+greedy by default, but become greedy if followed by "?". It is not compatible
+with Perl. It can also be set by a (?U) option setting within the pattern.
+.sp
+  PCRE_UTF8
+.sp
+This option causes PCRE to regard both the pattern and the subject as strings
+of UTF-8 characters instead of single-byte strings. However, it is available
+only when PCRE is built to include UTF support. If not, the use of this option
+provokes an error. Details of how this option changes the behaviour of PCRE are
+given in the
+.\" HREF
+\fBpcreunicode\fP
+.\"
+page.
+.sp
+  PCRE_NO_UTF8_CHECK
+.sp
+When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is
+automatically checked. There is a discussion about the
+.\" HTML 
+.\" 
+validity of UTF-8 strings
+.\"
+in the
+.\" HREF
+\fBpcreunicode\fP
+.\"
+page. If an invalid UTF-8 sequence is found, \fBpcre_compile()\fP returns an
+error. If you already know that your pattern is valid, and you want to skip
+this check for performance reasons, you can set the PCRE_NO_UTF8_CHECK option.
+When it is set, the effect of passing an invalid UTF-8 string as a pattern is
+undefined. It may cause your program to crash. Note that this option can also
+be passed to \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, to suppress the
+validity checking of subject strings only. If the same string is being matched
+many times, the option can be safely set for the second and subsequent
+matchings to improve performance.
+.
+.
+.SH "COMPILATION ERROR CODES"
+.rs
+.sp
+The following table lists the error codes than may be returned by
+\fBpcre_compile2()\fP, along with the error messages that may be returned by
+both compiling functions. Note that error messages are always 8-bit ASCII
+strings, even in 16-bit or 32-bit mode. As PCRE has developed, some error codes
+have fallen out of use. To avoid confusion, they have not been re-used.
+.sp
+   0  no error
+   1  \e at end of pattern
+   2  \ec at end of pattern
+   3  unrecognized character follows \e
+   4  numbers out of order in {} quantifier
+   5  number too big in {} quantifier
+   6  missing terminating ] for character class
+   7  invalid escape sequence in character class
+   8  range out of order in character class
+   9  nothing to repeat
+  10  [this code is not in use]
+  11  internal error: unexpected repeat
+  12  unrecognized character after (? or (?-
+  13  POSIX named classes are supported only within a class
+  14  missing )
+  15  reference to non-existent subpattern
+  16  erroffset passed as NULL
+  17  unknown option bit(s) set
+  18  missing ) after comment
+  19  [this code is not in use]
+  20  regular expression is too large
+  21  failed to get memory
+  22  unmatched parentheses
+  23  internal error: code overflow
+  24  unrecognized character after (?<
+  25  lookbehind assertion is not fixed length
+  26  malformed number or name after (?(
+  27  conditional group contains more than two branches
+  28  assertion expected after (?(
+  29  (?R or (?[+-]digits must be followed by )
+  30  unknown POSIX class name
+  31  POSIX collating elements are not supported
+  32  this version of PCRE is compiled without UTF support
+  33  [this code is not in use]
+  34  character value in \ex{...} sequence is too large
+  35  invalid condition (?(0)
+  36  \eC not allowed in lookbehind assertion
+  37  PCRE does not support \eL, \el, \eN{name}, \eU, or \eu
+  38  number after (?C is > 255
+  39  closing ) for (?C expected
+  40  recursive call could loop indefinitely
+  41  unrecognized character after (?P
+  42  syntax error in subpattern name (missing terminator)
+  43  two named subpatterns have the same name
+  44  invalid UTF-8 string (specifically UTF-8)
+  45  support for \eP, \ep, and \eX has not been compiled
+  46  malformed \eP or \ep sequence
+  47  unknown property name after \eP or \ep
+  48  subpattern name is too long (maximum 32 characters)
+  49  too many named subpatterns (maximum 10000)
+  50  [this code is not in use]
+  51  octal value is greater than \e377 in 8-bit non-UTF-8 mode
+  52  internal error: overran compiling workspace
+  53  internal error: previously-checked referenced subpattern
+        not found
+  54  DEFINE group contains more than one branch
+  55  repeating a DEFINE group is not allowed
+  56  inconsistent NEWLINE options
+  57  \eg is not followed by a braced, angle-bracketed, or quoted
+        name/number or by a plain number
+  58  a numbered reference must not be zero
+  59  an argument is not allowed for (*ACCEPT), (*FAIL), or (*COMMIT)
+  60  (*VERB) not recognized or malformed
+  61  number is too big
+  62  subpattern name expected
+  63  digit expected after (?+
+  64  ] is an invalid data character in JavaScript compatibility mode
+  65  different names for subpatterns of the same number are
+        not allowed
+  66  (*MARK) must have an argument
+  67  this version of PCRE is not compiled with Unicode property
+        support
+  68  \ec must be followed by an ASCII character
+  69  \ek is not followed by a braced, angle-bracketed, or quoted name
+  70  internal error: unknown opcode in find_fixedlength()
+  71  \eN is not supported in a class
+  72  too many forward references
+  73  disallowed Unicode code point (>= 0xd800 && <= 0xdfff)
+  74  invalid UTF-16 string (specifically UTF-16)
+  75  name is too long in (*MARK), (*PRUNE), (*SKIP), or (*THEN)
+  76  character value in \eu.... sequence is too large
+  77  invalid UTF-32 string (specifically UTF-32)
+.sp
+The numbers 32 and 10000 in errors 48 and 49 are defaults; different values may
+be used if the limits were changed when PCRE was built.
+.
+.
+.\" HTML 
+.SH "STUDYING A PATTERN"
+.rs
+.sp
+.B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP
+.ti +5n
+.B const char **\fIerrptr\fP);
+.PP
+If a compiled pattern is going to be used several times, it is worth spending
+more time analyzing it in order to speed up the time taken for matching. The
+function \fBpcre_study()\fP takes a pointer to a compiled pattern as its first
+argument. If studying the pattern produces additional information that will
+help speed up matching, \fBpcre_study()\fP returns a pointer to a
+\fBpcre_extra\fP block, in which the \fIstudy_data\fP field points to the
+results of the study.
+.P
+The returned value from \fBpcre_study()\fP can be passed directly to
+\fBpcre_exec()\fP or \fBpcre_dfa_exec()\fP. However, a \fBpcre_extra\fP block
+also contains other fields that can be set by the caller before the block is
+passed; these are described
+.\" HTML 
+.\" 
+below
+.\"
+in the section on matching a pattern.
+.P
+If studying the pattern does not produce any useful information,
+\fBpcre_study()\fP returns NULL by default. In that circumstance, if the
+calling program wants to pass any of the other fields to \fBpcre_exec()\fP or
+\fBpcre_dfa_exec()\fP, it must set up its own \fBpcre_extra\fP block. However,
+if \fBpcre_study()\fP is called with the PCRE_STUDY_EXTRA_NEEDED option, it
+returns a \fBpcre_extra\fP block even if studying did not find any additional
+information. It may still return NULL, however, if an error occurs in
+\fBpcre_study()\fP.
+.P
+The second argument of \fBpcre_study()\fP contains option bits. There are three
+further options in addition to PCRE_STUDY_EXTRA_NEEDED:
+.sp
+  PCRE_STUDY_JIT_COMPILE
+  PCRE_STUDY_JIT_PARTIAL_HARD_COMPILE
+  PCRE_STUDY_JIT_PARTIAL_SOFT_COMPILE
+.sp
+If any of these are set, and the just-in-time compiler is available, the
+pattern is further compiled into machine code that executes much faster than
+the \fBpcre_exec()\fP interpretive matching function. If the just-in-time
+compiler is not available, these options are ignored. All undefined bits in the
+\fIoptions\fP argument must be zero.
+.P
+JIT compilation is a heavyweight optimization. It can take some time for
+patterns to be analyzed, and for one-off matches and simple patterns the
+benefit of faster execution might be offset by a much slower study time.
+Not all patterns can be optimized by the JIT compiler. For those that cannot be
+handled, matching automatically falls back to the \fBpcre_exec()\fP
+interpreter. For more details, see the
+.\" HREF
+\fBpcrejit\fP
+.\"
+documentation.
+.P
+The third argument for \fBpcre_study()\fP is a pointer for an error message. If
+studying succeeds (even if no data is returned), the variable it points to is
+set to NULL. Otherwise it is set to point to a textual error message. This is a
+static string that is part of the library. You must not try to free it. You
+should test the error pointer for NULL after calling \fBpcre_study()\fP, to be
+sure that it has run successfully.
+.P
+When you are finished with a pattern, you can free the memory used for the
+study data by calling \fBpcre_free_study()\fP. This function was added to the
+API for release 8.20. For earlier versions, the memory could be freed with
+\fBpcre_free()\fP, just like the pattern itself. This will still work in cases
+where JIT optimization is not used, but it is advisable to change to the new
+function when convenient.
+.P
+This is a typical way in which \fBpcre_study\fP() is used (except that in a
+real application there should be tests for errors):
+.sp
+  int rc;
+  pcre *re;
+  pcre_extra *sd;
+  re = pcre_compile("pattern", 0, &error, &erroroffset, NULL);
+  sd = pcre_study(
+    re,             /* result of pcre_compile() */
+    0,              /* no options */
+    &error);        /* set to NULL or points to a message */
+  rc = pcre_exec(   /* see below for details of pcre_exec() options */
+    re, sd, "subject", 7, 0, 0, ovector, 30);
+  ...
+  pcre_free_study(sd);
+  pcre_free(re);
+.sp
+Studying a pattern does two things: first, a lower bound for the length of
+subject string that is needed to match the pattern is computed. This does not
+mean that there are any strings of that length that match, but it does
+guarantee that no shorter strings match. The value is used to avoid wasting
+time by trying to match strings that are shorter than the lower bound. You can
+find out the value in a calling program via the \fBpcre_fullinfo()\fP function.
+.P
+Studying a pattern is also useful for non-anchored patterns that do not have a
+single fixed starting character. A bitmap of possible starting bytes is
+created. This speeds up finding a position in the subject at which to start
+matching. (In 16-bit mode, the bitmap is used for 16-bit values less than 256.
+In 32-bit mode, the bitmap is used for 32-bit values less than 256.)
+.P
+These two optimizations apply to both \fBpcre_exec()\fP and
+\fBpcre_dfa_exec()\fP, and the information is also used by the JIT compiler.
+The optimizations can be disabled by setting the PCRE_NO_START_OPTIMIZE option.
+You might want to do this if your pattern contains callouts or (*MARK) and you
+want to make use of these facilities in cases where matching fails.
+.P
+PCRE_NO_START_OPTIMIZE can be specified at either compile time or execution
+time. However, if PCRE_NO_START_OPTIMIZE is passed to \fBpcre_exec()\fP, (that
+is, after any JIT compilation has happened) JIT execution is disabled. For JIT
+execution to work with PCRE_NO_START_OPTIMIZE, the option must be set at
+compile time.
+.P
+There is a longer discussion of PCRE_NO_START_OPTIMIZE
+.\" HTML 
+.\" 
+below.
+.\"
+.
+.
+.\" HTML 
+.SH "LOCALE SUPPORT"
+.rs
+.sp
+PCRE handles caseless matching, and determines whether characters are letters,
+digits, or whatever, by reference to a set of tables, indexed by character
+value. When running in UTF-8 mode, this applies only to characters
+with codes less than 128. By default, higher-valued codes never match escapes
+such as \ew or \ed, but they can be tested with \ep if PCRE is built with
+Unicode character property support. Alternatively, the PCRE_UCP option can be
+set at compile time; this causes \ew and friends to use Unicode property
+support instead of built-in tables. The use of locales with Unicode is
+discouraged. If you are handling characters with codes greater than 128, you
+should either use UTF-8 and Unicode, or use locales, but not try to mix the
+two.
+.P
+PCRE contains an internal set of tables that are used when the final argument
+of \fBpcre_compile()\fP is NULL. These are sufficient for many applications.
+Normally, the internal tables recognize only ASCII characters. However, when
+PCRE is built, it is possible to cause the internal tables to be rebuilt in the
+default "C" locale of the local system, which may cause them to be different.
+.P
+The internal tables can always be overridden by tables supplied by the
+application that calls PCRE. These may be created in a different locale from
+the default. As more and more applications change to using Unicode, the need
+for this locale support is expected to die away.
+.P
+External tables are built by calling the \fBpcre_maketables()\fP function,
+which has no arguments, in the relevant locale. The result can then be passed
+to \fBpcre_compile()\fP or \fBpcre_exec()\fP as often as necessary. For
+example, to build and use tables that are appropriate for the French locale
+(where accented characters with values greater than 128 are treated as letters),
+the following code could be used:
+.sp
+  setlocale(LC_CTYPE, "fr_FR");
+  tables = pcre_maketables();
+  re = pcre_compile(..., tables);
+.sp
+The locale name "fr_FR" is used on Linux and other Unix-like systems; if you
+are using Windows, the name for the French locale is "french".
+.P
+When \fBpcre_maketables()\fP runs, the tables are built in memory that is
+obtained via \fBpcre_malloc\fP. It is the caller's responsibility to ensure
+that the memory containing the tables remains available for as long as it is
+needed.
+.P
+The pointer that is passed to \fBpcre_compile()\fP is saved with the compiled
+pattern, and the same tables are used via this pointer by \fBpcre_study()\fP
+and normally also by \fBpcre_exec()\fP. Thus, by default, for any single
+pattern, compilation, studying and matching all happen in the same locale, but
+different patterns can be compiled in different locales.
+.P
+It is possible to pass a table pointer or NULL (indicating the use of the
+internal tables) to \fBpcre_exec()\fP. Although not intended for this purpose,
+this facility could be used to match a pattern in a different locale from the
+one in which it was compiled. Passing table pointers at run time is discussed
+below in the section on matching a pattern.
+.
+.
+.\" HTML 
+.SH "INFORMATION ABOUT A PATTERN"
+.rs
+.sp
+.B int pcre_fullinfo(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
+.ti +5n
+.B int \fIwhat\fP, void *\fIwhere\fP);
+.PP
+The \fBpcre_fullinfo()\fP function returns information about a compiled
+pattern. It replaces the \fBpcre_info()\fP function, which was removed from the
+library at version 8.30, after more than 10 years of obsolescence.
+.P
+The first argument for \fBpcre_fullinfo()\fP is a pointer to the compiled
+pattern. The second argument is the result of \fBpcre_study()\fP, or NULL if
+the pattern was not studied. The third argument specifies which piece of
+information is required, and the fourth argument is a pointer to a variable
+to receive the data. The yield of the function is zero for success, or one of
+the following negative numbers:
+.sp
+  PCRE_ERROR_NULL           the argument \fIcode\fP was NULL
+                            the argument \fIwhere\fP was NULL
+  PCRE_ERROR_BADMAGIC       the "magic number" was not found
+  PCRE_ERROR_BADENDIANNESS  the pattern was compiled with different
+                            endianness
+  PCRE_ERROR_BADOPTION      the value of \fIwhat\fP was invalid
+  PCRE_ERROR_UNSET          the requested field is not set
+.sp
+The "magic number" is placed at the start of each compiled pattern as an simple
+check against passing an arbitrary memory pointer. The endianness error can
+occur if a compiled pattern is saved and reloaded on a different host. Here is
+a typical call of \fBpcre_fullinfo()\fP, to obtain the length of the compiled
+pattern:
+.sp
+  int rc;
+  size_t length;
+  rc = pcre_fullinfo(
+    re,               /* result of pcre_compile() */
+    sd,               /* result of pcre_study(), or NULL */
+    PCRE_INFO_SIZE,   /* what is required */
+    &length);         /* where to put the data */
+.sp
+The possible values for the third argument are defined in \fBpcre.h\fP, and are
+as follows:
+.sp
+  PCRE_INFO_BACKREFMAX
+.sp
+Return the number of the highest back reference in the pattern. The fourth
+argument should point to an \fBint\fP variable. Zero is returned if there are
+no back references.
+.sp
+  PCRE_INFO_CAPTURECOUNT
+.sp
+Return the number of capturing subpatterns in the pattern. The fourth argument
+should point to an \fBint\fP variable.
+.sp
+  PCRE_INFO_DEFAULT_TABLES
+.sp
+Return a pointer to the internal default character tables within PCRE. The
+fourth argument should point to an \fBunsigned char *\fP variable. This
+information call is provided for internal use by the \fBpcre_study()\fP
+function. External callers can cause PCRE to use its internal tables by passing
+a NULL table pointer.
+.sp
+  PCRE_INFO_FIRSTBYTE
+.sp
+Return information about the first data unit of any matched string, for a
+non-anchored pattern. (The name of this option refers to the 8-bit library,
+where data units are bytes.) The fourth argument should point to an \fBint\fP
+variable.
+.P
+If there is a fixed first value, for example, the letter "c" from a pattern
+such as (cat|cow|coyote), its value is returned. In the 8-bit library, the
+value is always less than 256. In the 16-bit library the value can be up to
+0xffff. In the 32-bit library the value can be up to 0x10ffff.
+.P
+If there is no fixed first value, and if either
+.sp
+(a) the pattern was compiled with the PCRE_MULTILINE option, and every branch
+starts with "^", or
+.sp
+(b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not set
+(if it were set, the pattern would be anchored),
+.sp
+-1 is returned, indicating that the pattern matches only at the start of a
+subject string or after any newline within the string. Otherwise -2 is
+returned. For anchored patterns, -2 is returned.
+.P
+Since for the 32-bit library using the non-UTF-32 mode, this function is unable
+to return the full 32-bit range of the character, this value is deprecated;
+instead the PCRE_INFO_FIRSTCHARACTERFLAGS and PCRE_INFO_FIRSTCHARACTER values
+should be used.
+.sp
+  PCRE_INFO_FIRSTTABLE
+.sp
+If the pattern was studied, and this resulted in the construction of a 256-bit
+table indicating a fixed set of values for the first data unit in any matching
+string, a pointer to the table is returned. Otherwise NULL is returned. The
+fourth argument should point to an \fBunsigned char *\fP variable.
+.sp
+  PCRE_INFO_HASCRORLF
+.sp
+Return 1 if the pattern contains any explicit matches for CR or LF characters,
+otherwise 0. The fourth argument should point to an \fBint\fP variable. An
+explicit match is either a literal CR or LF character, or \er or \en.
+.sp
+  PCRE_INFO_JCHANGED
+.sp
+Return 1 if the (?J) or (?-J) option setting is used in the pattern, otherwise
+0. The fourth argument should point to an \fBint\fP variable. (?J) and
+(?-J) set and unset the local PCRE_DUPNAMES option, respectively.
+.sp
+  PCRE_INFO_JIT
+.sp
+Return 1 if the pattern was studied with one of the JIT options, and
+just-in-time compiling was successful. The fourth argument should point to an
+\fBint\fP variable. A return value of 0 means that JIT support is not available
+in this version of PCRE, or that the pattern was not studied with a JIT option,
+or that the JIT compiler could not handle this particular pattern. See the
+.\" HREF
+\fBpcrejit\fP
+.\"
+documentation for details of what can and cannot be handled.
+.sp
+  PCRE_INFO_JITSIZE
+.sp
+If the pattern was successfully studied with a JIT option, return the size of
+the JIT compiled code, otherwise return zero. The fourth argument should point
+to a \fBsize_t\fP variable.
+.sp
+  PCRE_INFO_LASTLITERAL
+.sp
+Return the value of the rightmost literal data unit that must exist in any
+matched string, other than at its start, if such a value has been recorded. The
+fourth argument should point to an \fBint\fP variable. If there is no such
+value, -1 is returned. For anchored patterns, a last literal value is recorded
+only if it follows something of variable length. For example, for the pattern
+/^a\ed+z\ed+/ the returned value is "z", but for /^a\edz\ed/ the returned value
+is -1.
+.P
+Since for the 32-bit library using the non-UTF-32 mode, this function is unable
+to return the full 32-bit range of the character, this value is deprecated;
+instead the PCRE_INFO_REQUIREDCHARFLAGS and PCRE_INFO_REQUIREDCHAR values should
+be used.
+.sp
+  PCRE_INFO_MATCHLIMIT
+.sp
+If the pattern set a match limit by including an item of the form
+(*LIMIT_MATCH=nnnn) at the start, the value is returned. The fourth argument
+should point to an unsigned 32-bit integer. If no such value has been set, the
+call to \fBpcre_fullinfo()\fP returns the error PCRE_ERROR_UNSET.
+.sp
+  PCRE_INFO_MAXLOOKBEHIND
+.sp
+Return the number of characters (NB not data units) in the longest lookbehind
+assertion in the pattern. This information is useful when doing multi-segment
+matching using the partial matching facilities. Note that the simple assertions
+\eb and \eB require a one-character lookbehind. \eA also registers a
+one-character lookbehind, though it does not actually inspect the previous
+character. This is to ensure that at least one character from the old segment
+is retained when a new segment is processed. Otherwise, if there are no
+lookbehinds in the pattern, \eA might match incorrectly at the start of a new
+segment.
+.sp
+  PCRE_INFO_MINLENGTH
+.sp
+If the pattern was studied and a minimum length for matching subject strings
+was computed, its value is returned. Otherwise the returned value is -1. The
+value is a number of characters, which in UTF mode may be different from the
+number of data units. The fourth argument should point to an \fBint\fP
+variable. A non-negative value is a lower bound to the length of any matching
+string. There may not be any strings of that length that do actually match, but
+every string that does match is at least that long.
+.sp
+  PCRE_INFO_NAMECOUNT
+  PCRE_INFO_NAMEENTRYSIZE
+  PCRE_INFO_NAMETABLE
+.sp
+PCRE supports the use of named as well as numbered capturing parentheses. The
+names are just an additional way of identifying the parentheses, which still
+acquire numbers. Several convenience functions such as
+\fBpcre_get_named_substring()\fP are provided for extracting captured
+substrings by name. It is also possible to extract the data directly, by first
+converting the name to a number in order to access the correct pointers in the
+output vector (described with \fBpcre_exec()\fP below). To do the conversion,
+you need to use the name-to-number map, which is described by these three
+values.
+.P
+The map consists of a number of fixed-size entries. PCRE_INFO_NAMECOUNT gives
+the number of entries, and PCRE_INFO_NAMEENTRYSIZE gives the size of each
+entry; both of these return an \fBint\fP value. The entry size depends on the
+length of the longest name. PCRE_INFO_NAMETABLE returns a pointer to the first
+entry of the table. This is a pointer to \fBchar\fP in the 8-bit library, where
+the first two bytes of each entry are the number of the capturing parenthesis,
+most significant byte first. In the 16-bit library, the pointer points to
+16-bit data units, the first of which contains the parenthesis number. In the
+32-bit library, the pointer points to 32-bit data units, the first of which
+contains the parenthesis number. The rest of the entry is the corresponding
+name, zero terminated.
+.P
+The names are in alphabetical order. Duplicate names may appear if (?| is used
+to create multiple groups with the same number, as described in the
+.\" HTML 
+.\" 
+section on duplicate subpattern numbers
+.\"
+in the
+.\" HREF
+\fBpcrepattern\fP
+.\"
+page. Duplicate names for subpatterns with different numbers are permitted only
+if PCRE_DUPNAMES is set. In all cases of duplicate names, they appear in the
+table in the order in which they were found in the pattern. In the absence of
+(?| this is the order of increasing number; when (?| is used this is not
+necessarily the case because later subpatterns may have lower numbers.
+.P
+As a simple example of the name/number table, consider the following pattern
+after compilation by the 8-bit library (assume PCRE_EXTENDED is set, so white
+space - including newlines - is ignored):
+.sp
+.\" JOIN
+  (? (?(\ed\ed)?\ed\ed) -
+  (?\ed\ed) - (?\ed\ed) )
+.sp
+There are four named subpatterns, so the table has four entries, and each entry
+in the table is eight bytes long. The table is as follows, with non-printing
+bytes shows in hexadecimal, and undefined bytes shown as ??:
+.sp
+  00 01 d  a  t  e  00 ??
+  00 05 d  a  y  00 ?? ??
+  00 04 m  o  n  t  h  00
+  00 02 y  e  a  r  00 ??
+.sp
+When writing code to extract data from named subpatterns using the
+name-to-number map, remember that the length of the entries is likely to be
+different for each compiled pattern.
+.sp
+  PCRE_INFO_OKPARTIAL
+.sp
+Return 1 if the pattern can be used for partial matching with
+\fBpcre_exec()\fP, otherwise 0. The fourth argument should point to an
+\fBint\fP variable. From release 8.00, this always returns 1, because the
+restrictions that previously applied to partial matching have been lifted. The
+.\" HREF
+\fBpcrepartial\fP
+.\"
+documentation gives details of partial matching.
+.sp
+  PCRE_INFO_OPTIONS
+.sp
+Return a copy of the options with which the pattern was compiled. The fourth
+argument should point to an \fBunsigned long int\fP variable. These option bits
+are those specified in the call to \fBpcre_compile()\fP, modified by any
+top-level option settings at the start of the pattern itself. In other words,
+they are the options that will be in force when matching starts. For example,
+if the pattern /(?im)abc(?-i)d/ is compiled with the PCRE_EXTENDED option, the
+result is PCRE_CASELESS, PCRE_MULTILINE, and PCRE_EXTENDED.
+.P
+A pattern is automatically anchored by PCRE if all of its top-level
+alternatives begin with one of the following:
+.sp
+  ^     unless PCRE_MULTILINE is set
+  \eA    always
+  \eG    always
+.\" JOIN
+  .*    if PCRE_DOTALL is set and there are no back
+          references to the subpattern in which .* appears
+.sp
+For such patterns, the PCRE_ANCHORED bit is set in the options returned by
+\fBpcre_fullinfo()\fP.
+.sp
+  PCRE_INFO_RECURSIONLIMIT
+.sp
+If the pattern set a recursion limit by including an item of the form
+(*LIMIT_RECURSION=nnnn) at the start, the value is returned. The fourth
+argument should point to an unsigned 32-bit integer. If no such value has been
+set, the call to \fBpcre_fullinfo()\fP returns the error PCRE_ERROR_UNSET.
+.sp
+  PCRE_INFO_SIZE
+.sp
+Return the size of the compiled pattern in bytes (for all three libraries). The
+fourth argument should point to a \fBsize_t\fP variable. This value does not
+include the size of the \fBpcre\fP structure that is returned by
+\fBpcre_compile()\fP. The value that is passed as the argument to
+\fBpcre_malloc()\fP when \fBpcre_compile()\fP is getting memory in which to
+place the compiled data is the value returned by this option plus the size of
+the \fBpcre\fP structure. Studying a compiled pattern, with or without JIT,
+does not alter the value returned by this option.
+.sp
+  PCRE_INFO_STUDYSIZE
+.sp
+Return the size in bytes (for all three libraries) of the data block pointed to
+by the \fIstudy_data\fP field in a \fBpcre_extra\fP block. If \fBpcre_extra\fP
+is NULL, or there is no study data, zero is returned. The fourth argument
+should point to a \fBsize_t\fP variable. The \fIstudy_data\fP field is set by
+\fBpcre_study()\fP to record information that will speed up matching (see the
+section entitled
+.\" HTML 
+.\" 
+"Studying a pattern"
+.\"
+above). The format of the \fIstudy_data\fP block is private, but its length
+is made available via this option so that it can be saved and restored (see the
+.\" HREF
+\fBpcreprecompile\fP
+.\"
+documentation for details).
+.sp
+  PCRE_INFO_FIRSTCHARACTERFLAGS
+.sp
+Return information about the first data unit of any matched string, for a
+non-anchored pattern. The fourth argument should point to an \fBint\fP
+variable.
+.P
+If there is a fixed first value, for example, the letter "c" from a pattern
+such as (cat|cow|coyote), 1 is returned, and the character value can be
+retrieved using PCRE_INFO_FIRSTCHARACTER.
+.P
+If there is no fixed first value, and if either
+.sp
+(a) the pattern was compiled with the PCRE_MULTILINE option, and every branch
+starts with "^", or
+.sp
+(b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not set
+(if it were set, the pattern would be anchored),
+.sp
+2 is returned, indicating that the pattern matches only at the start of a
+subject string or after any newline within the string. Otherwise 0 is
+returned. For anchored patterns, 0 is returned.
+.sp
+  PCRE_INFO_FIRSTCHARACTER
+.sp
+Return the fixed first character value, if PCRE_INFO_FIRSTCHARACTERFLAGS
+returned 1; otherwise returns 0. The fourth argument should point to an
+\fBuint_t\fP variable.
+.P
+In the 8-bit library, the value is always less than 256. In the 16-bit library
+the value can be up to 0xffff. In the 32-bit library in UTF-32 mode the value
+can be up to 0x10ffff, and up to 0xffffffff when not using UTF-32 mode.
+.P
+If there is no fixed first value, and if either
+.sp
+(a) the pattern was compiled with the PCRE_MULTILINE option, and every branch
+starts with "^", or
+.sp
+(b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not set
+(if it were set, the pattern would be anchored),
+.sp
+-1 is returned, indicating that the pattern matches only at the start of a
+subject string or after any newline within the string. Otherwise -2 is
+returned. For anchored patterns, -2 is returned.
+.sp
+  PCRE_INFO_REQUIREDCHARFLAGS
+.sp
+Returns 1 if there is a rightmost literal data unit that must exist in any
+matched string, other than at its start. The fourth argument should  point to
+an \fBint\fP variable. If there is no such value, 0 is returned. If returning
+1, the character value itself can be retrieved using PCRE_INFO_REQUIREDCHAR.
+.P
+For anchored patterns, a last literal value is recorded only if it follows
+something of variable length. For example, for the pattern /^a\ed+z\ed+/ the
+returned value 1 (with "z" returned from PCRE_INFO_REQUIREDCHAR), but for
+/^a\edz\ed/ the returned value is 0.
+.sp
+  PCRE_INFO_REQUIREDCHAR
+.sp
+Return the value of the rightmost literal data unit that must exist in any
+matched string, other than at its start, if such a value has been recorded. The
+fourth argument should point to an \fBuint32_t\fP variable. If there is no such
+value, 0 is returned.
+.
+.
+.SH "REFERENCE COUNTS"
+.rs
+.sp
+.B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
+.PP
+The \fBpcre_refcount()\fP function is used to maintain a reference count in the
+data block that contains a compiled pattern. It is provided for the benefit of
+applications that operate in an object-oriented manner, where different parts
+of the application may be using the same compiled pattern, but you want to free
+the block when they are all done.
+.P
+When a pattern is compiled, the reference count field is initialized to zero.
+It is changed only by calling this function, whose action is to add the
+\fIadjust\fP value (which may be positive or negative) to it. The yield of the
+function is the new value. However, the value of the count is constrained to
+lie between 0 and 65535, inclusive. If the new value is outside these limits,
+it is forced to the appropriate limit value.
+.P
+Except when it is zero, the reference count is not correctly preserved if a
+pattern is compiled on one host and then transferred to a host whose byte-order
+is different. (This seems a highly unlikely scenario.)
+.
+.
+.SH "MATCHING A PATTERN: THE TRADITIONAL FUNCTION"
+.rs
+.sp
+.B int pcre_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
+.ti +5n
+.B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
+.ti +5n
+.B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);
+.P
+The function \fBpcre_exec()\fP is called to match a subject string against a
+compiled pattern, which is passed in the \fIcode\fP argument. If the
+pattern was studied, the result of the study should be passed in the
+\fIextra\fP argument. You can call \fBpcre_exec()\fP with the same \fIcode\fP
+and \fIextra\fP arguments as many times as you like, in order to match
+different subject strings with the same pattern.
+.P
+This function is the main matching facility of the library, and it operates in
+a Perl-like manner. For specialist use there is also an alternative matching
+function, which is described
+.\" HTML 
+.\" 
+below
+.\"
+in the section about the \fBpcre_dfa_exec()\fP function.
+.P
+In most applications, the pattern will have been compiled (and optionally
+studied) in the same process that calls \fBpcre_exec()\fP. However, it is
+possible to save compiled patterns and study data, and then use them later
+in different processes, possibly even on different hosts. For a discussion
+about this, see the
+.\" HREF
+\fBpcreprecompile\fP
+.\"
+documentation.
+.P
+Here is an example of a simple call to \fBpcre_exec()\fP:
+.sp
+  int rc;
+  int ovector[30];
+  rc = pcre_exec(
+    re,             /* result of pcre_compile() */
+    NULL,           /* we didn't study the pattern */
+    "some string",  /* the subject string */
+    11,             /* the length of the subject string */
+    0,              /* start at offset 0 in the subject */
+    0,              /* default options */
+    ovector,        /* vector of integers for substring information */
+    30);            /* number of elements (NOT size in bytes) */
+.
+.
+.\" HTML 
+.SS "Extra data for \fBpcre_exec()\fR"
+.rs
+.sp
+If the \fIextra\fP argument is not NULL, it must point to a \fBpcre_extra\fP
+data block. The \fBpcre_study()\fP function returns such a block (when it
+doesn't return NULL), but you can also create one for yourself, and pass
+additional information in it. The \fBpcre_extra\fP block contains the following
+fields (not necessarily in this order):
+.sp
+  unsigned long int \fIflags\fP;
+  void *\fIstudy_data\fP;
+  void *\fIexecutable_jit\fP;
+  unsigned long int \fImatch_limit\fP;
+  unsigned long int \fImatch_limit_recursion\fP;
+  void *\fIcallout_data\fP;
+  const unsigned char *\fItables\fP;
+  unsigned char **\fImark\fP;
+.sp
+In the 16-bit version of this structure, the \fImark\fP field has type
+"PCRE_UCHAR16 **".
+.sp
+In the 32-bit version of this structure, the \fImark\fP field has type
+"PCRE_UCHAR32 **".
+.P
+The \fIflags\fP field is used to specify which of the other fields are set. The
+flag bits are:
+.sp
+  PCRE_EXTRA_CALLOUT_DATA
+  PCRE_EXTRA_EXECUTABLE_JIT
+  PCRE_EXTRA_MARK
+  PCRE_EXTRA_MATCH_LIMIT
+  PCRE_EXTRA_MATCH_LIMIT_RECURSION
+  PCRE_EXTRA_STUDY_DATA
+  PCRE_EXTRA_TABLES
+.sp
+Other flag bits should be set to zero. The \fIstudy_data\fP field and sometimes
+the \fIexecutable_jit\fP field are set in the \fBpcre_extra\fP block that is
+returned by \fBpcre_study()\fP, together with the appropriate flag bits. You
+should not set these yourself, but you may add to the block by setting other
+fields and their corresponding flag bits.
+.P
+The \fImatch_limit\fP field provides a means of preventing PCRE from using up a
+vast amount of resources when running patterns that are not going to match,
+but which have a very large number of possibilities in their search trees. The
+classic example is a pattern that uses nested unlimited repeats.
+.P
+Internally, \fBpcre_exec()\fP uses a function called \fBmatch()\fP, which it
+calls repeatedly (sometimes recursively). The limit set by \fImatch_limit\fP is
+imposed on the number of times this function is called during a match, which
+has the effect of limiting the amount of backtracking that can take place. For
+patterns that are not anchored, the count restarts from zero for each position
+in the subject string.
+.P
+When \fBpcre_exec()\fP is called with a pattern that was successfully studied
+with a JIT option, the way that the matching is executed is entirely different.
+However, there is still the possibility of runaway matching that goes on for a
+very long time, and so the \fImatch_limit\fP value is also used in this case
+(but in a different way) to limit how long the matching can continue.
+.P
+The default value for the limit can be set when PCRE is built; the default
+default is 10 million, which handles all but the most extreme cases. You can
+override the default by suppling \fBpcre_exec()\fP with a \fBpcre_extra\fP
+block in which \fImatch_limit\fP is set, and PCRE_EXTRA_MATCH_LIMIT is set in
+the \fIflags\fP field. If the limit is exceeded, \fBpcre_exec()\fP returns
+PCRE_ERROR_MATCHLIMIT.
+.P
+A value for the match limit may also be supplied by an item at the start of a
+pattern of the form
+.sp
+  (*LIMIT_MATCH=d)
+.sp
+where d is a decimal number. However, such a setting is ignored unless d is
+less than the limit set by the caller of \fBpcre_exec()\fP or, if no such limit
+is set, less than the default.
+.P
+The \fImatch_limit_recursion\fP field is similar to \fImatch_limit\fP, but
+instead of limiting the total number of times that \fBmatch()\fP is called, it
+limits the depth of recursion. The recursion depth is a smaller number than the
+total number of calls, because not all calls to \fBmatch()\fP are recursive.
+This limit is of use only if it is set smaller than \fImatch_limit\fP.
+.P
+Limiting the recursion depth limits the amount of machine stack that can be
+used, or, when PCRE has been compiled to use memory on the heap instead of the
+stack, the amount of heap memory that can be used. This limit is not relevant,
+and is ignored, when matching is done using JIT compiled code.
+.P
+The default value for \fImatch_limit_recursion\fP can be set when PCRE is
+built; the default default is the same value as the default for
+\fImatch_limit\fP. You can override the default by suppling \fBpcre_exec()\fP
+with a \fBpcre_extra\fP block in which \fImatch_limit_recursion\fP is set, and
+PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the \fIflags\fP field. If the limit
+is exceeded, \fBpcre_exec()\fP returns PCRE_ERROR_RECURSIONLIMIT.
+.P
+A value for the recursion limit may also be supplied by an item at the start of
+a pattern of the form
+.sp
+  (*LIMIT_RECURSION=d)
+.sp
+where d is a decimal number. However, such a setting is ignored unless d is
+less than the limit set by the caller of \fBpcre_exec()\fP or, if no such limit
+is set, less than the default.
+.P
+The \fIcallout_data\fP field is used in conjunction with the "callout" feature,
+and is described in the
+.\" HREF
+\fBpcrecallout\fP
+.\"
+documentation.
+.P
+The \fItables\fP field is used to pass a character tables pointer to
+\fBpcre_exec()\fP; this overrides the value that is stored with the compiled
+pattern. A non-NULL value is stored with the compiled pattern only if custom
+tables were supplied to \fBpcre_compile()\fP via its \fItableptr\fP argument.
+If NULL is passed to \fBpcre_exec()\fP using this mechanism, it forces PCRE's
+internal tables to be used. This facility is helpful when re-using patterns
+that have been saved after compiling with an external set of tables, because
+the external tables might be at a different address when \fBpcre_exec()\fP is
+called. See the
+.\" HREF
+\fBpcreprecompile\fP
+.\"
+documentation for a discussion of saving compiled patterns for later use.
+.P
+If PCRE_EXTRA_MARK is set in the \fIflags\fP field, the \fImark\fP field must
+be set to point to a suitable variable. If the pattern contains any
+backtracking control verbs such as (*MARK:NAME), and the execution ends up with
+a name to pass back, a pointer to the name string (zero terminated) is placed
+in the variable pointed to by the \fImark\fP field. The names are within the
+compiled pattern; if you wish to retain such a name you must copy it before
+freeing the memory of a compiled pattern. If there is no name to pass back, the
+variable pointed to by the \fImark\fP field is set to NULL. For details of the
+backtracking control verbs, see the section entitled
+.\" HTML 
+.\" 
+"Backtracking control"
+.\"
+in the
+.\" HREF
+\fBpcrepattern\fP
+.\"
+documentation.
+.
+.
+.\" HTML 
+.SS "Option bits for \fBpcre_exec()\fP"
+.rs
+.sp
+The unused bits of the \fIoptions\fP argument for \fBpcre_exec()\fP must be
+zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
+PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART,
+PCRE_NO_START_OPTIMIZE, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL_HARD, and
+PCRE_PARTIAL_SOFT.
+.P
+If the pattern was successfully studied with one of the just-in-time (JIT)
+compile options, the only supported options for JIT execution are
+PCRE_NO_UTF8_CHECK, PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY,
+PCRE_NOTEMPTY_ATSTART, PCRE_PARTIAL_HARD, and PCRE_PARTIAL_SOFT. If an
+unsupported option is used, JIT execution is disabled and the normal
+interpretive code in \fBpcre_exec()\fP is run.
+.sp
+  PCRE_ANCHORED
+.sp
+The PCRE_ANCHORED option limits \fBpcre_exec()\fP to matching at the first
+matching position. If a pattern was compiled with PCRE_ANCHORED, or turned out
+to be anchored by virtue of its contents, it cannot be made unachored at
+matching time.
+.sp
+  PCRE_BSR_ANYCRLF
+  PCRE_BSR_UNICODE
+.sp
+These options (which are mutually exclusive) control what the \eR escape
+sequence matches. The choice is either to match only CR, LF, or CRLF, or to
+match any Unicode newline sequence. These options override the choice that was
+made or defaulted when the pattern was compiled.
+.sp
+  PCRE_NEWLINE_CR
+  PCRE_NEWLINE_LF
+  PCRE_NEWLINE_CRLF
+  PCRE_NEWLINE_ANYCRLF
+  PCRE_NEWLINE_ANY
+.sp
+These options override the newline definition that was chosen or defaulted when
+the pattern was compiled. For details, see the description of
+\fBpcre_compile()\fP above. During matching, the newline choice affects the
+behaviour of the dot, circumflex, and dollar metacharacters. It may also alter
+the way the match position is advanced after a match failure for an unanchored
+pattern.
+.P
+When PCRE_NEWLINE_CRLF, PCRE_NEWLINE_ANYCRLF, or PCRE_NEWLINE_ANY is set, and a
+match attempt for an unanchored pattern fails when the current position is at a
+CRLF sequence, and the pattern contains no explicit matches for CR or LF
+characters, the match position is advanced by two characters instead of one, in
+other words, to after the CRLF.
+.P
+The above rule is a compromise that makes the most common cases work as
+expected. For example, if the pattern is .+A (and the PCRE_DOTALL option is not
+set), it does not match the string "\er\enA" because, after failing at the
+start, it skips both the CR and the LF before retrying. However, the pattern
+[\er\en]A does match that string, because it contains an explicit CR or LF
+reference, and so advances only by one character after the first failure.
+.P
+An explicit match for CR of LF is either a literal appearance of one of those
+characters, or one of the \er or \en escape sequences. Implicit matches such as
+[^X] do not count, nor does \es (which includes CR and LF in the characters
+that it matches).
+.P
+Notwithstanding the above, anomalous effects may still occur when CRLF is a
+valid newline sequence and explicit \er or \en escapes appear in the pattern.
+.sp
+  PCRE_NOTBOL
+.sp
+This option specifies that first character of the subject string is not the
+beginning of a line, so the circumflex metacharacter should not match before
+it. Setting this without PCRE_MULTILINE (at compile time) causes circumflex
+never to match. This option affects only the behaviour of the circumflex
+metacharacter. It does not affect \eA.
+.sp
+  PCRE_NOTEOL
+.sp
+This option specifies that the end of the subject string is not the end of a
+line, so the dollar metacharacter should not match it nor (except in multiline
+mode) a newline immediately before it. Setting this without PCRE_MULTILINE (at
+compile time) causes dollar never to match. This option affects only the
+behaviour of the dollar metacharacter. It does not affect \eZ or \ez.
+.sp
+  PCRE_NOTEMPTY
+.sp
+An empty string is not considered to be a valid match if this option is set. If
+there are alternatives in the pattern, they are tried. If all the alternatives
+match the empty string, the entire match fails. For example, if the pattern
+.sp
+  a?b?
+.sp
+is applied to a string not beginning with "a" or "b", it matches an empty
+string at the start of the subject. With PCRE_NOTEMPTY set, this match is not
+valid, so PCRE searches further into the string for occurrences of "a" or "b".
+.sp
+  PCRE_NOTEMPTY_ATSTART
+.sp
+This is like PCRE_NOTEMPTY, except that an empty string match that is not at
+the start of the subject is permitted. If the pattern is anchored, such a match
+can occur only if the pattern contains \eK.
+.P
+Perl has no direct equivalent of PCRE_NOTEMPTY or PCRE_NOTEMPTY_ATSTART, but it
+does make a special case of a pattern match of the empty string within its
+\fBsplit()\fP function, and when using the /g modifier. It is possible to
+emulate Perl's behaviour after matching a null string by first trying the match
+again at the same offset with PCRE_NOTEMPTY_ATSTART and PCRE_ANCHORED, and then
+if that fails, by advancing the starting offset (see below) and trying an
+ordinary match again. There is some code that demonstrates how to do this in
+the
+.\" HREF
+\fBpcredemo\fP
+.\"
+sample program. In the most general case, you have to check to see if the
+newline convention recognizes CRLF as a newline, and if so, and the current
+character is CR followed by LF, advance the starting offset by two characters
+instead of one.
+.sp
+  PCRE_NO_START_OPTIMIZE
+.sp
+There are a number of optimizations that \fBpcre_exec()\fP uses at the start of
+a match, in order to speed up the process. For example, if it is known that an
+unanchored match must start with a specific character, it searches the subject
+for that character, and fails immediately if it cannot find it, without
+actually running the main matching function. This means that a special item
+such as (*COMMIT) at the start of a pattern is not considered until after a
+suitable starting point for the match has been found. Also, when callouts or
+(*MARK) items are in use, these "start-up" optimizations can cause them to be
+skipped if the pattern is never actually used. The start-up optimizations are
+in effect a pre-scan of the subject that takes place before the pattern is run.
+.P
+The PCRE_NO_START_OPTIMIZE option disables the start-up optimizations, possibly
+causing performance to suffer, but ensuring that in cases where the result is
+"no match", the callouts do occur, and that items such as (*COMMIT) and (*MARK)
+are considered at every possible starting position in the subject string. If
+PCRE_NO_START_OPTIMIZE is set at compile time, it cannot be unset at matching
+time. The use of PCRE_NO_START_OPTIMIZE at matching time (that is, passing it
+to \fBpcre_exec()\fP) disables JIT execution; in this situation, matching is
+always done using interpretively.
+.P
+Setting PCRE_NO_START_OPTIMIZE can change the outcome of a matching operation.
+Consider the pattern
+.sp
+  (*COMMIT)ABC
+.sp
+When this is compiled, PCRE records the fact that a match must start with the
+character "A". Suppose the subject string is "DEFABC". The start-up
+optimization scans along the subject, finds "A" and runs the first match
+attempt from there. The (*COMMIT) item means that the pattern must match the
+current starting position, which in this case, it does. However, if the same
+match is run with PCRE_NO_START_OPTIMIZE set, the initial scan along the
+subject string does not happen. The first match attempt is run starting from
+"D" and when this fails, (*COMMIT) prevents any further matches being tried, so
+the overall result is "no match". If the pattern is studied, more start-up
+optimizations may be used. For example, a minimum length for the subject may be
+recorded. Consider the pattern
+.sp
+  (*MARK:A)(X|Y)
+.sp
+The minimum length for a match is one character. If the subject is "ABC", there
+will be attempts to match "ABC", "BC", "C", and then finally an empty string.
+If the pattern is studied, the final attempt does not take place, because PCRE
+knows that the subject is too short, and so the (*MARK) is never encountered.
+In this case, studying the pattern does not affect the overall match result,
+which is still "no match", but it does affect the auxiliary information that is
+returned.
+.sp
+  PCRE_NO_UTF8_CHECK
+.sp
+When PCRE_UTF8 is set at compile time, the validity of the subject as a UTF-8
+string is automatically checked when \fBpcre_exec()\fP is subsequently called.
+The entire string is checked before any other processing takes place. The value
+of \fIstartoffset\fP is also checked to ensure that it points to the start of a
+UTF-8 character. There is a discussion about the
+.\" HTML 
+.\" 
+validity of UTF-8 strings
+.\"
+in the
+.\" HREF
+\fBpcreunicode\fP
+.\"
+page. If an invalid sequence of bytes is found, \fBpcre_exec()\fP returns the
+error PCRE_ERROR_BADUTF8 or, if PCRE_PARTIAL_HARD is set and the problem is a
+truncated character at the end of the subject, PCRE_ERROR_SHORTUTF8. In both
+cases, information about the precise nature of the error may also be returned
+(see the descriptions of these errors in the section entitled \fIError return
+values from\fP \fBpcre_exec()\fP
+.\" HTML 
+.\" 
+below).
+.\"
+If \fIstartoffset\fP contains a value that does not point to the start of a
+UTF-8 character (or to the end of the subject), PCRE_ERROR_BADUTF8_OFFSET is
+returned.
+.P
+If you already know that your subject is valid, and you want to skip these
+checks for performance reasons, you can set the PCRE_NO_UTF8_CHECK option when
+calling \fBpcre_exec()\fP. You might want to do this for the second and
+subsequent calls to \fBpcre_exec()\fP if you are making repeated calls to find
+all the matches in a single subject string. However, you should be sure that
+the value of \fIstartoffset\fP points to the start of a character (or the end
+of the subject). When PCRE_NO_UTF8_CHECK is set, the effect of passing an
+invalid string as a subject or an invalid value of \fIstartoffset\fP is
+undefined. Your program may crash.
+.sp
+  PCRE_PARTIAL_HARD
+  PCRE_PARTIAL_SOFT
+.sp
+These options turn on the partial matching feature. For backwards
+compatibility, PCRE_PARTIAL is a synonym for PCRE_PARTIAL_SOFT. A partial match
+occurs if the end of the subject string is reached successfully, but there are
+not enough subject characters to complete the match. If this happens when
+PCRE_PARTIAL_SOFT (but not PCRE_PARTIAL_HARD) is set, matching continues by
+testing any remaining alternatives. Only if no complete match can be found is
+PCRE_ERROR_PARTIAL returned instead of PCRE_ERROR_NOMATCH. In other words,
+PCRE_PARTIAL_SOFT says that the caller is prepared to handle a partial match,
+but only if no complete match can be found.
+.P
+If PCRE_PARTIAL_HARD is set, it overrides PCRE_PARTIAL_SOFT. In this case, if a
+partial match is found, \fBpcre_exec()\fP immediately returns
+PCRE_ERROR_PARTIAL, without considering any other alternatives. In other words,
+when PCRE_PARTIAL_HARD is set, a partial match is considered to be more
+important that an alternative complete match.
+.P
+In both cases, the portion of the string that was inspected when the partial
+match was found is set as the first matching string. There is a more detailed
+discussion of partial and multi-segment matching, with examples, in the
+.\" HREF
+\fBpcrepartial\fP
+.\"
+documentation.
+.
+.
+.SS "The string to be matched by \fBpcre_exec()\fP"
+.rs
+.sp
+The subject string is passed to \fBpcre_exec()\fP as a pointer in
+\fIsubject\fP, a length in \fIlength\fP, and a starting offset in
+\fIstartoffset\fP. The units for \fIlength\fP and \fIstartoffset\fP are bytes
+for the 8-bit library, 16-bit data items for the 16-bit library, and 32-bit
+data items for the 32-bit library.
+.P
+If \fIstartoffset\fP is negative or greater than the length of the subject,
+\fBpcre_exec()\fP returns PCRE_ERROR_BADOFFSET. When the starting offset is
+zero, the search for a match starts at the beginning of the subject, and this
+is by far the most common case. In UTF-8 or UTF-16 mode, the offset must point
+to the start of a character, or the end of the subject (in UTF-32 mode, one
+data unit equals one character, so all offsets are valid). Unlike the pattern
+string, the subject may contain binary zeroes.
+.P
+A non-zero starting offset is useful when searching for another match in the
+same subject by calling \fBpcre_exec()\fP again after a previous success.
+Setting \fIstartoffset\fP differs from just passing over a shortened string and
+setting PCRE_NOTBOL in the case of a pattern that begins with any kind of
+lookbehind. For example, consider the pattern
+.sp
+  \eBiss\eB
+.sp
+which finds occurrences of "iss" in the middle of words. (\eB matches only if
+the current position in the subject is not a word boundary.) When applied to
+the string "Mississipi" the first call to \fBpcre_exec()\fP finds the first
+occurrence. If \fBpcre_exec()\fP is called again with just the remainder of the
+subject, namely "issipi", it does not match, because \eB is always false at the
+start of the subject, which is deemed to be a word boundary. However, if
+\fBpcre_exec()\fP is passed the entire string again, but with \fIstartoffset\fP
+set to 4, it finds the second occurrence of "iss" because it is able to look
+behind the starting point to discover that it is preceded by a letter.
+.P
+Finding all the matches in a subject is tricky when the pattern can match an
+empty string. It is possible to emulate Perl's /g behaviour by first trying the
+match again at the same offset, with the PCRE_NOTEMPTY_ATSTART and
+PCRE_ANCHORED options, and then if that fails, advancing the starting offset
+and trying an ordinary match again. There is some code that demonstrates how to
+do this in the
+.\" HREF
+\fBpcredemo\fP
+.\"
+sample program. In the most general case, you have to check to see if the
+newline convention recognizes CRLF as a newline, and if so, and the current
+character is CR followed by LF, advance the starting offset by two characters
+instead of one.
+.P
+If a non-zero starting offset is passed when the pattern is anchored, one
+attempt to match at the given offset is made. This can only succeed if the
+pattern does not require the match to be at the start of the subject.
+.
+.
+.SS "How \fBpcre_exec()\fP returns captured substrings"
+.rs
+.sp
+In general, a pattern matches a certain portion of the subject, and in
+addition, further substrings from the subject may be picked out by parts of the
+pattern. Following the usage in Jeffrey Friedl's book, this is called
+"capturing" in what follows, and the phrase "capturing subpattern" is used for
+a fragment of a pattern that picks out a substring. PCRE supports several other
+kinds of parenthesized subpattern that do not cause substrings to be captured.
+.P
+Captured substrings are returned to the caller via a vector of integers whose
+address is passed in \fIovector\fP. The number of elements in the vector is
+passed in \fIovecsize\fP, which must be a non-negative number. \fBNote\fP: this
+argument is NOT the size of \fIovector\fP in bytes.
+.P
+The first two-thirds of the vector is used to pass back captured substrings,
+each substring using a pair of integers. The remaining third of the vector is
+used as workspace by \fBpcre_exec()\fP while matching capturing subpatterns,
+and is not available for passing back information. The number passed in
+\fIovecsize\fP should always be a multiple of three. If it is not, it is
+rounded down.
+.P
+When a match is successful, information about captured substrings is returned
+in pairs of integers, starting at the beginning of \fIovector\fP, and
+continuing up to two-thirds of its length at the most. The first element of
+each pair is set to the offset of the first character in a substring, and the
+second is set to the offset of the first character after the end of a
+substring. These values are always data unit offsets, even in UTF mode. They
+are byte offsets in the 8-bit library, 16-bit data item offsets in the 16-bit
+library, and 32-bit data item offsets in the 32-bit library. \fBNote\fP: they
+are not character counts.
+.P
+The first pair of integers, \fIovector[0]\fP and \fIovector[1]\fP, identify the
+portion of the subject string matched by the entire pattern. The next pair is
+used for the first capturing subpattern, and so on. The value returned by
+\fBpcre_exec()\fP is one more than the highest numbered pair that has been set.
+For example, if two substrings have been captured, the returned value is 3. If
+there are no capturing subpatterns, the return value from a successful match is
+1, indicating that just the first pair of offsets has been set.
+.P
+If a capturing subpattern is matched repeatedly, it is the last portion of the
+string that it matched that is returned.
+.P
+If the vector is too small to hold all the captured substring offsets, it is
+used as far as possible (up to two-thirds of its length), and the function
+returns a value of zero. If neither the actual string matched nor any captured
+substrings are of interest, \fBpcre_exec()\fP may be called with \fIovector\fP
+passed as NULL and \fIovecsize\fP as zero. However, if the pattern contains
+back references and the \fIovector\fP is not big enough to remember the related
+substrings, PCRE has to get additional memory for use during matching. Thus it
+is usually advisable to supply an \fIovector\fP of reasonable size.
+.P
+There are some cases where zero is returned (indicating vector overflow) when
+in fact the vector is exactly the right size for the final match. For example,
+consider the pattern
+.sp
+  (a)(?:(b)c|bd)
+.sp
+If a vector of 6 elements (allowing for only 1 captured substring) is given
+with subject string "abd", \fBpcre_exec()\fP will try to set the second
+captured string, thereby recording a vector overflow, before failing to match
+"c" and backing up to try the second alternative. The zero return, however,
+does correctly indicate that the maximum number of slots (namely 2) have been
+filled. In similar cases where there is temporary overflow, but the final
+number of used slots is actually less than the maximum, a non-zero value is
+returned.
+.P
+The \fBpcre_fullinfo()\fP function can be used to find out how many capturing
+subpatterns there are in a compiled pattern. The smallest size for
+\fIovector\fP that will allow for \fIn\fP captured substrings, in addition to
+the offsets of the substring matched by the whole pattern, is (\fIn\fP+1)*3.
+.P
+It is possible for capturing subpattern number \fIn+1\fP to match some part of
+the subject when subpattern \fIn\fP has not been used at all. For example, if
+the string "abc" is matched against the pattern (a|(z))(bc) the return from the
+function is 4, and subpatterns 1 and 3 are matched, but 2 is not. When this
+happens, both values in the offset pairs corresponding to unused subpatterns
+are set to -1.
+.P
+Offset values that correspond to unused subpatterns at the end of the
+expression are also set to -1. For example, if the string "abc" is matched
+against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The
+return from the function is 2, because the highest used capturing subpattern
+number is 1, and the offsets for for the second and third capturing subpatterns
+(assuming the vector is large enough, of course) are set to -1.
+.P
+\fBNote\fP: Elements in the first two-thirds of \fIovector\fP that do not
+correspond to capturing parentheses in the pattern are never changed. That is,
+if a pattern contains \fIn\fP capturing parentheses, no more than
+\fIovector[0]\fP to \fIovector[2n+1]\fP are set by \fBpcre_exec()\fP. The other
+elements (in the first two-thirds) retain whatever values they previously had.
+.P
+Some convenience functions are provided for extracting the captured substrings
+as separate strings. These are described below.
+.
+.
+.\" HTML 
+.SS "Error return values from \fBpcre_exec()\fP"
+.rs
+.sp
+If \fBpcre_exec()\fP fails, it returns a negative number. The following are
+defined in the header file:
+.sp
+  PCRE_ERROR_NOMATCH        (-1)
+.sp
+The subject string did not match the pattern.
+.sp
+  PCRE_ERROR_NULL           (-2)
+.sp
+Either \fIcode\fP or \fIsubject\fP was passed as NULL, or \fIovector\fP was
+NULL and \fIovecsize\fP was not zero.
+.sp
+  PCRE_ERROR_BADOPTION      (-3)
+.sp
+An unrecognized bit was set in the \fIoptions\fP argument.
+.sp
+  PCRE_ERROR_BADMAGIC       (-4)
+.sp
+PCRE stores a 4-byte "magic number" at the start of the compiled code, to catch
+the case when it is passed a junk pointer and to detect when a pattern that was
+compiled in an environment of one endianness is run in an environment with the
+other endianness. This is the error that PCRE gives when the magic number is
+not present.
+.sp
+  PCRE_ERROR_UNKNOWN_OPCODE (-5)
+.sp
+While running the pattern match, an unknown item was encountered in the
+compiled pattern. This error could be caused by a bug in PCRE or by overwriting
+of the compiled pattern.
+.sp
+  PCRE_ERROR_NOMEMORY       (-6)
+.sp
+If a pattern contains back references, but the \fIovector\fP that is passed to
+\fBpcre_exec()\fP is not big enough to remember the referenced substrings, PCRE
+gets a block of memory at the start of matching to use for this purpose. If the
+call via \fBpcre_malloc()\fP fails, this error is given. The memory is
+automatically freed at the end of matching.
+.P
+This error is also given if \fBpcre_stack_malloc()\fP fails in
+\fBpcre_exec()\fP. This can happen only when PCRE has been compiled with
+\fB--disable-stack-for-recursion\fP.
+.sp
+  PCRE_ERROR_NOSUBSTRING    (-7)
+.sp
+This error is used by the \fBpcre_copy_substring()\fP,
+\fBpcre_get_substring()\fP, and \fBpcre_get_substring_list()\fP functions (see
+below). It is never returned by \fBpcre_exec()\fP.
+.sp
+  PCRE_ERROR_MATCHLIMIT     (-8)
+.sp
+The backtracking limit, as specified by the \fImatch_limit\fP field in a
+\fBpcre_extra\fP structure (or defaulted) was reached. See the description
+above.
+.sp
+  PCRE_ERROR_CALLOUT        (-9)
+.sp
+This error is never generated by \fBpcre_exec()\fP itself. It is provided for
+use by callout functions that want to yield a distinctive error code. See the
+.\" HREF
+\fBpcrecallout\fP
+.\"
+documentation for details.
+.sp
+  PCRE_ERROR_BADUTF8        (-10)
+.sp
+A string that contains an invalid UTF-8 byte sequence was passed as a subject,
+and the PCRE_NO_UTF8_CHECK option was not set. If the size of the output vector
+(\fIovecsize\fP) is at least 2, the byte offset to the start of the the invalid
+UTF-8 character is placed in the first element, and a reason code is placed in
+the second element. The reason codes are listed in the
+.\" HTML 
+.\" 
+following section.
+.\"
+For backward compatibility, if PCRE_PARTIAL_HARD is set and the problem is a
+truncated UTF-8 character at the end of the subject (reason codes 1 to 5),
+PCRE_ERROR_SHORTUTF8 is returned instead of PCRE_ERROR_BADUTF8.
+.sp
+  PCRE_ERROR_BADUTF8_OFFSET (-11)
+.sp
+The UTF-8 byte sequence that was passed as a subject was checked and found to
+be valid (the PCRE_NO_UTF8_CHECK option was not set), but the value of
+\fIstartoffset\fP did not point to the beginning of a UTF-8 character or the
+end of the subject.
+.sp
+  PCRE_ERROR_PARTIAL        (-12)
+.sp
+The subject string did not match, but it did match partially. See the
+.\" HREF
+\fBpcrepartial\fP
+.\"
+documentation for details of partial matching.
+.sp
+  PCRE_ERROR_BADPARTIAL     (-13)
+.sp
+This code is no longer in use. It was formerly returned when the PCRE_PARTIAL
+option was used with a compiled pattern containing items that were not
+supported for partial matching. From release 8.00 onwards, there are no
+restrictions on partial matching.
+.sp
+  PCRE_ERROR_INTERNAL       (-14)
+.sp
+An unexpected internal error has occurred. This error could be caused by a bug
+in PCRE or by overwriting of the compiled pattern.
+.sp
+  PCRE_ERROR_BADCOUNT       (-15)
+.sp
+This error is given if the value of the \fIovecsize\fP argument is negative.
+.sp
+  PCRE_ERROR_RECURSIONLIMIT (-21)
+.sp
+The internal recursion limit, as specified by the \fImatch_limit_recursion\fP
+field in a \fBpcre_extra\fP structure (or defaulted) was reached. See the
+description above.
+.sp
+  PCRE_ERROR_BADNEWLINE     (-23)
+.sp
+An invalid combination of PCRE_NEWLINE_\fIxxx\fP options was given.
+.sp
+  PCRE_ERROR_BADOFFSET      (-24)
+.sp
+The value of \fIstartoffset\fP was negative or greater than the length of the
+subject, that is, the value in \fIlength\fP.
+.sp
+  PCRE_ERROR_SHORTUTF8      (-25)
+.sp
+This error is returned instead of PCRE_ERROR_BADUTF8 when the subject string
+ends with a truncated UTF-8 character and the PCRE_PARTIAL_HARD option is set.
+Information about the failure is returned as for PCRE_ERROR_BADUTF8. It is in
+fact sufficient to detect this case, but this special error code for
+PCRE_PARTIAL_HARD precedes the implementation of returned information; it is
+retained for backwards compatibility.
+.sp
+  PCRE_ERROR_RECURSELOOP    (-26)
+.sp
+This error is returned when \fBpcre_exec()\fP detects a recursion loop within
+the pattern. Specifically, it means that either the whole pattern or a
+subpattern has been called recursively for the second time at the same position
+in the subject string. Some simple patterns that might do this are detected and
+faulted at compile time, but more complicated cases, in particular mutual
+recursions between two different subpatterns, cannot be detected until run
+time.
+.sp
+  PCRE_ERROR_JIT_STACKLIMIT (-27)
+.sp
+This error is returned when a pattern that was successfully studied using a
+JIT compile option is being matched, but the memory available for the
+just-in-time processing stack is not large enough. See the
+.\" HREF
+\fBpcrejit\fP
+.\"
+documentation for more details.
+.sp
+  PCRE_ERROR_BADMODE        (-28)
+.sp
+This error is given if a pattern that was compiled by the 8-bit library is
+passed to a 16-bit or 32-bit library function, or vice versa.
+.sp
+  PCRE_ERROR_BADENDIANNESS  (-29)
+.sp
+This error is given if a pattern that was compiled and saved is reloaded on a
+host with different endianness. The utility function
+\fBpcre_pattern_to_host_byte_order()\fP can be used to convert such a pattern
+so that it runs on the new host.
+.sp
+  PCRE_ERROR_JIT_BADOPTION
+.sp
+This error is returned when a pattern that was successfully studied using a JIT
+compile option is being matched, but the matching mode (partial or complete
+match) does not correspond to any JIT compilation mode. When the JIT fast path
+function is used, this error may be also given for invalid options. See the
+.\" HREF
+\fBpcrejit\fP
+.\"
+documentation for more details.
+.sp
+  PCRE_ERROR_BADLENGTH      (-32)
+.sp
+This error is given if \fBpcre_exec()\fP is called with a negative value for
+the \fIlength\fP argument.
+.P
+Error numbers -16 to -20, -22, and 30 are not used by \fBpcre_exec()\fP.
+.
+.
+.\" HTML 
+.SS "Reason codes for invalid UTF-8 strings"
+.rs
+.sp
+This section applies only to the 8-bit library. The corresponding information
+for the 16-bit and 32-bit libraries is given in the
+.\" HREF
+\fBpcre16\fP
+.\"
+and
+.\" HREF
+\fBpcre32\fP
+.\"
+pages.
+.P
+When \fBpcre_exec()\fP returns either PCRE_ERROR_BADUTF8 or
+PCRE_ERROR_SHORTUTF8, and the size of the output vector (\fIovecsize\fP) is at
+least 2, the offset of the start of the invalid UTF-8 character is placed in
+the first output vector element (\fIovector[0]\fP) and a reason code is placed
+in the second element (\fIovector[1]\fP). The reason codes are given names in
+the \fBpcre.h\fP header file:
+.sp
+  PCRE_UTF8_ERR1
+  PCRE_UTF8_ERR2
+  PCRE_UTF8_ERR3
+  PCRE_UTF8_ERR4
+  PCRE_UTF8_ERR5
+.sp
+The string ends with a truncated UTF-8 character; the code specifies how many
+bytes are missing (1 to 5). Although RFC 3629 restricts UTF-8 characters to be
+no longer than 4 bytes, the encoding scheme (originally defined by RFC 2279)
+allows for up to 6 bytes, and this is checked first; hence the possibility of
+4 or 5 missing bytes.
+.sp
+  PCRE_UTF8_ERR6
+  PCRE_UTF8_ERR7
+  PCRE_UTF8_ERR8
+  PCRE_UTF8_ERR9
+  PCRE_UTF8_ERR10
+.sp
+The two most significant bits of the 2nd, 3rd, 4th, 5th, or 6th byte of the
+character do not have the binary value 0b10 (that is, either the most
+significant bit is 0, or the next bit is 1).
+.sp
+  PCRE_UTF8_ERR11
+  PCRE_UTF8_ERR12
+.sp
+A character that is valid by the RFC 2279 rules is either 5 or 6 bytes long;
+these code points are excluded by RFC 3629.
+.sp
+  PCRE_UTF8_ERR13
+.sp
+A 4-byte character has a value greater than 0x10fff; these code points are
+excluded by RFC 3629.
+.sp
+  PCRE_UTF8_ERR14
+.sp
+A 3-byte character has a value in the range 0xd800 to 0xdfff; this range of
+code points are reserved by RFC 3629 for use with UTF-16, and so are excluded
+from UTF-8.
+.sp
+  PCRE_UTF8_ERR15
+  PCRE_UTF8_ERR16
+  PCRE_UTF8_ERR17
+  PCRE_UTF8_ERR18
+  PCRE_UTF8_ERR19
+.sp
+A 2-, 3-, 4-, 5-, or 6-byte character is "overlong", that is, it codes for a
+value that can be represented by fewer bytes, which is invalid. For example,
+the two bytes 0xc0, 0xae give the value 0x2e, whose correct coding uses just
+one byte.
+.sp
+  PCRE_UTF8_ERR20
+.sp
+The two most significant bits of the first byte of a character have the binary
+value 0b10 (that is, the most significant bit is 1 and the second is 0). Such a
+byte can only validly occur as the second or subsequent byte of a multi-byte
+character.
+.sp
+  PCRE_UTF8_ERR21
+.sp
+The first byte of a character has the value 0xfe or 0xff. These values can
+never occur in a valid UTF-8 string.
+.sp
+  PCRE_UTF8_ERR22
+.sp
+This error code was formerly used when the presence of a so-called
+"non-character" caused an error. Unicode corrigendum #9 makes it clear that
+such characters should not cause a string to be rejected, and so this code is
+no longer in use and is never returned.
+.
+.
+.SH "EXTRACTING CAPTURED SUBSTRINGS BY NUMBER"
+.rs
+.sp
+.B int pcre_copy_substring(const char *\fIsubject\fP, int *\fIovector\fP,
+.ti +5n
+.B int \fIstringcount\fP, int \fIstringnumber\fP, char *\fIbuffer\fP,
+.ti +5n
+.B int \fIbuffersize\fP);
+.PP
+.B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,
+.ti +5n
+.B int \fIstringcount\fP, int \fIstringnumber\fP,
+.ti +5n
+.B const char **\fIstringptr\fP);
+.PP
+.B int pcre_get_substring_list(const char *\fIsubject\fP,
+.ti +5n
+.B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"
+.PP
+Captured substrings can be accessed directly by using the offsets returned by
+\fBpcre_exec()\fP in \fIovector\fP. For convenience, the functions
+\fBpcre_copy_substring()\fP, \fBpcre_get_substring()\fP, and
+\fBpcre_get_substring_list()\fP are provided for extracting captured substrings
+as new, separate, zero-terminated strings. These functions identify substrings
+by number. The next section describes functions for extracting named
+substrings.
+.P
+A substring that contains a binary zero is correctly extracted and has a
+further zero added on the end, but the result is not, of course, a C string.
+However, you can process such a string by referring to the length that is
+returned by \fBpcre_copy_substring()\fP and \fBpcre_get_substring()\fP.
+Unfortunately, the interface to \fBpcre_get_substring_list()\fP is not adequate
+for handling strings containing binary zeros, because the end of the final
+string is not independently indicated.
+.P
+The first three arguments are the same for all three of these functions:
+\fIsubject\fP is the subject string that has just been successfully matched,
+\fIovector\fP is a pointer to the vector of integer offsets that was passed to
+\fBpcre_exec()\fP, and \fIstringcount\fP is the number of substrings that were
+captured by the match, including the substring that matched the entire regular
+expression. This is the value returned by \fBpcre_exec()\fP if it is greater
+than zero. If \fBpcre_exec()\fP returned zero, indicating that it ran out of
+space in \fIovector\fP, the value passed as \fIstringcount\fP should be the
+number of elements in the vector divided by three.
+.P
+The functions \fBpcre_copy_substring()\fP and \fBpcre_get_substring()\fP
+extract a single substring, whose number is given as \fIstringnumber\fP. A
+value of zero extracts the substring that matched the entire pattern, whereas
+higher values extract the captured substrings. For \fBpcre_copy_substring()\fP,
+the string is placed in \fIbuffer\fP, whose length is given by
+\fIbuffersize\fP, while for \fBpcre_get_substring()\fP a new block of memory is
+obtained via \fBpcre_malloc\fP, and its address is returned via
+\fIstringptr\fP. The yield of the function is the length of the string, not
+including the terminating zero, or one of these error codes:
+.sp
+  PCRE_ERROR_NOMEMORY       (-6)
+.sp
+The buffer was too small for \fBpcre_copy_substring()\fP, or the attempt to get
+memory failed for \fBpcre_get_substring()\fP.
+.sp
+  PCRE_ERROR_NOSUBSTRING    (-7)
+.sp
+There is no substring whose number is \fIstringnumber\fP.
+.P
+The \fBpcre_get_substring_list()\fP function extracts all available substrings
+and builds a list of pointers to them. All this is done in a single block of
+memory that is obtained via \fBpcre_malloc\fP. The address of the memory block
+is returned via \fIlistptr\fP, which is also the start of the list of string
+pointers. The end of the list is marked by a NULL pointer. The yield of the
+function is zero if all went well, or the error code
+.sp
+  PCRE_ERROR_NOMEMORY       (-6)
+.sp
+if the attempt to get the memory block failed.
+.P
+When any of these functions encounter a substring that is unset, which can
+happen when capturing subpattern number \fIn+1\fP matches some part of the
+subject, but subpattern \fIn\fP has not been used at all, they return an empty
+string. This can be distinguished from a genuine zero-length substring by
+inspecting the appropriate offset in \fIovector\fP, which is negative for unset
+substrings.
+.P
+The two convenience functions \fBpcre_free_substring()\fP and
+\fBpcre_free_substring_list()\fP can be used to free the memory returned by
+a previous call of \fBpcre_get_substring()\fP or
+\fBpcre_get_substring_list()\fP, respectively. They do nothing more than call
+the function pointed to by \fBpcre_free\fP, which of course could be called
+directly from a C program. However, PCRE is used in some situations where it is
+linked via a special interface to another programming language that cannot use
+\fBpcre_free\fP directly; it is for these cases that the functions are
+provided.
+.
+.
+.SH "EXTRACTING CAPTURED SUBSTRINGS BY NAME"
+.rs
+.sp
+.B int pcre_get_stringnumber(const pcre *\fIcode\fP,
+.ti +5n
+.B const char *\fIname\fP);
+.PP
+.B int pcre_copy_named_substring(const pcre *\fIcode\fP,
+.ti +5n
+.B const char *\fIsubject\fP, int *\fIovector\fP,
+.ti +5n
+.B int \fIstringcount\fP, const char *\fIstringname\fP,
+.ti +5n
+.B char *\fIbuffer\fP, int \fIbuffersize\fP);
+.PP
+.B int pcre_get_named_substring(const pcre *\fIcode\fP,
+.ti +5n
+.B const char *\fIsubject\fP, int *\fIovector\fP,
+.ti +5n
+.B int \fIstringcount\fP, const char *\fIstringname\fP,
+.ti +5n
+.B const char **\fIstringptr\fP);
+.PP
+To extract a substring by name, you first have to find associated number.
+For example, for this pattern
+.sp
+  (a+)b(?\ed+)...
+.sp
+the number of the subpattern called "xxx" is 2. If the name is known to be
+unique (PCRE_DUPNAMES was not set), you can find the number from the name by
+calling \fBpcre_get_stringnumber()\fP. The first argument is the compiled
+pattern, and the second is the name. The yield of the function is the
+subpattern number, or PCRE_ERROR_NOSUBSTRING (-7) if there is no subpattern of
+that name.
+.P
+Given the number, you can extract the substring directly, or use one of the
+functions described in the previous section. For convenience, there are also
+two functions that do the whole job.
+.P
+Most of the arguments of \fBpcre_copy_named_substring()\fP and
+\fBpcre_get_named_substring()\fP are the same as those for the similarly named
+functions that extract by number. As these are described in the previous
+section, they are not re-described here. There are just two differences:
+.P
+First, instead of a substring number, a substring name is given. Second, there
+is an extra argument, given at the start, which is a pointer to the compiled
+pattern. This is needed in order to gain access to the name-to-number
+translation table.
+.P
+These functions call \fBpcre_get_stringnumber()\fP, and if it succeeds, they
+then call \fBpcre_copy_substring()\fP or \fBpcre_get_substring()\fP, as
+appropriate. \fBNOTE:\fP If PCRE_DUPNAMES is set and there are duplicate names,
+the behaviour may not be what you want (see the next section).
+.P
+\fBWarning:\fP If the pattern uses the (?| feature to set up multiple
+subpatterns with the same number, as described in the
+.\" HTML 
+.\" 
+section on duplicate subpattern numbers
+.\"
+in the
+.\" HREF
+\fBpcrepattern\fP
+.\"
+page, you cannot use names to distinguish the different subpatterns, because
+names are not included in the compiled code. The matching process uses only
+numbers. For this reason, the use of different names for subpatterns of the
+same number causes an error at compile time.
+.
+.
+.SH "DUPLICATE SUBPATTERN NAMES"
+.rs
+.sp
+.B int pcre_get_stringtable_entries(const pcre *\fIcode\fP,
+.ti +5n
+.B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
+.PP
+When a pattern is compiled with the PCRE_DUPNAMES option, names for subpatterns
+are not required to be unique. (Duplicate names are always allowed for
+subpatterns with the same number, created by using the (?| feature. Indeed, if
+such subpatterns are named, they are required to use the same names.)
+.P
+Normally, patterns with duplicate names are such that in any one match, only
+one of the named subpatterns participates. An example is shown in the
+.\" HREF
+\fBpcrepattern\fP
+.\"
+documentation.
+.P
+When duplicates are present, \fBpcre_copy_named_substring()\fP and
+\fBpcre_get_named_substring()\fP return the first substring corresponding to
+the given name that is set. If none are set, PCRE_ERROR_NOSUBSTRING (-7) is
+returned; no data is returned. The \fBpcre_get_stringnumber()\fP function
+returns one of the numbers that are associated with the name, but it is not
+defined which it is.
+.P
+If you want to get full details of all captured substrings for a given name,
+you must use the \fBpcre_get_stringtable_entries()\fP function. The first
+argument is the compiled pattern, and the second is the name. The third and
+fourth are pointers to variables which are updated by the function. After it
+has run, they point to the first and last entries in the name-to-number table
+for the given name. The function itself returns the length of each entry, or
+PCRE_ERROR_NOSUBSTRING (-7) if there are none. The format of the table is
+described above in the section entitled \fIInformation about a pattern\fP
+.\" HTML 
+.\" 
+above.
+.\"
+Given all the relevant entries for the name, you can extract each of their
+numbers, and hence the captured data, if any.
+.
+.
+.SH "FINDING ALL POSSIBLE MATCHES"
+.rs
+.sp
+The traditional matching function uses a similar algorithm to Perl, which stops
+when it finds the first match, starting at a given point in the subject. If you
+want to find all possible matches, or the longest possible match, consider
+using the alternative matching function (see below) instead. If you cannot use
+the alternative function, but still need to find all possible matches, you
+can kludge it up by making use of the callout facility, which is described in
+the
+.\" HREF
+\fBpcrecallout\fP
+.\"
+documentation.
+.P
+What you have to do is to insert a callout right at the end of the pattern.
+When your callout function is called, extract and save the current matched
+substring. Then return 1, which forces \fBpcre_exec()\fP to backtrack and try
+other alternatives. Ultimately, when it runs out of matches, \fBpcre_exec()\fP
+will yield PCRE_ERROR_NOMATCH.
+.
+.
+.SH "OBTAINING AN ESTIMATE OF STACK USAGE"
+.rs
+.sp
+Matching certain patterns using \fBpcre_exec()\fP can use a lot of process
+stack, which in certain environments can be rather limited in size. Some users
+find it helpful to have an estimate of the amount of stack that is used by
+\fBpcre_exec()\fP, to help them set recursion limits, as described in the
+.\" HREF
+\fBpcrestack\fP
+.\"
+documentation. The estimate that is output by \fBpcretest\fP when called with
+the \fB-m\fP and \fB-C\fP options is obtained by calling \fBpcre_exec\fP with
+the values NULL, NULL, NULL, -999, and -999 for its first five arguments.
+.P
+Normally, if its first argument is NULL, \fBpcre_exec()\fP immediately returns
+the negative error code PCRE_ERROR_NULL, but with this special combination of
+arguments, it returns instead a negative number whose absolute value is the
+approximate stack frame size in bytes. (A negative number is used so that it is
+clear that no match has happened.) The value is approximate because in some
+cases, recursive calls to \fBpcre_exec()\fP occur when there are one or two
+additional variables on the stack.
+.P
+If PCRE has been compiled to use the heap instead of the stack for recursion,
+the value returned is the size of each block that is obtained from the heap.
+.
+.
+.\" HTML 
+.SH "MATCHING A PATTERN: THE ALTERNATIVE FUNCTION"
+.rs
+.sp
+.B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
+.ti +5n
+.B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
+.ti +5n
+.B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
+.ti +5n
+.B int *\fIworkspace\fP, int \fIwscount\fP);
+.P
+The function \fBpcre_dfa_exec()\fP is called to match a subject string against
+a compiled pattern, using a matching algorithm that scans the subject string
+just once, and does not backtrack. This has different characteristics to the
+normal algorithm, and is not compatible with Perl. Some of the features of PCRE
+patterns are not supported. Nevertheless, there are times when this kind of
+matching can be useful. For a discussion of the two matching algorithms, and a
+list of features that \fBpcre_dfa_exec()\fP does not support, see the
+.\" HREF
+\fBpcrematching\fP
+.\"
+documentation.
+.P
+The arguments for the \fBpcre_dfa_exec()\fP function are the same as for
+\fBpcre_exec()\fP, plus two extras. The \fIovector\fP argument is used in a
+different way, and this is described below. The other common arguments are used
+in the same way as for \fBpcre_exec()\fP, so their description is not repeated
+here.
+.P
+The two additional arguments provide workspace for the function. The workspace
+vector should contain at least 20 elements. It is used for keeping track of
+multiple paths through the pattern tree. More workspace will be needed for
+patterns and subjects where there are a lot of potential matches.
+.P
+Here is an example of a simple call to \fBpcre_dfa_exec()\fP:
+.sp
+  int rc;
+  int ovector[10];
+  int wspace[20];
+  rc = pcre_dfa_exec(
+    re,             /* result of pcre_compile() */
+    NULL,           /* we didn't study the pattern */
+    "some string",  /* the subject string */
+    11,             /* the length of the subject string */
+    0,              /* start at offset 0 in the subject */
+    0,              /* default options */
+    ovector,        /* vector of integers for substring information */
+    10,             /* number of elements (NOT size in bytes) */
+    wspace,         /* working space vector */
+    20);            /* number of elements (NOT size in bytes) */
+.
+.SS "Option bits for \fBpcre_dfa_exec()\fP"
+.rs
+.sp
+The unused bits of the \fIoptions\fP argument for \fBpcre_dfa_exec()\fP must be
+zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
+PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART,
+PCRE_NO_UTF8_CHECK, PCRE_BSR_ANYCRLF, PCRE_BSR_UNICODE, PCRE_NO_START_OPTIMIZE,
+PCRE_PARTIAL_HARD, PCRE_PARTIAL_SOFT, PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART.
+All but the last four of these are exactly the same as for \fBpcre_exec()\fP,
+so their description is not repeated here.
+.sp
+  PCRE_PARTIAL_HARD
+  PCRE_PARTIAL_SOFT
+.sp
+These have the same general effect as they do for \fBpcre_exec()\fP, but the
+details are slightly different. When PCRE_PARTIAL_HARD is set for
+\fBpcre_dfa_exec()\fP, it returns PCRE_ERROR_PARTIAL if the end of the subject
+is reached and there is still at least one matching possibility that requires
+additional characters. This happens even if some complete matches have also
+been found. When PCRE_PARTIAL_SOFT is set, the return code PCRE_ERROR_NOMATCH
+is converted into PCRE_ERROR_PARTIAL if the end of the subject is reached,
+there have been no complete matches, but there is still at least one matching
+possibility. The portion of the string that was inspected when the longest
+partial match was found is set as the first matching string in both cases.
+There is a more detailed discussion of partial and multi-segment matching, with
+examples, in the
+.\" HREF
+\fBpcrepartial\fP
+.\"
+documentation.
+.sp
+  PCRE_DFA_SHORTEST
+.sp
+Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as
+soon as it has found one match. Because of the way the alternative algorithm
+works, this is necessarily the shortest possible match at the first possible
+matching point in the subject string.
+.sp
+  PCRE_DFA_RESTART
+.sp
+When \fBpcre_dfa_exec()\fP returns a partial match, it is possible to call it
+again, with additional subject characters, and have it continue with the same
+match. The PCRE_DFA_RESTART option requests this action; when it is set, the
+\fIworkspace\fP and \fIwscount\fP options must reference the same vector as
+before because data about the match so far is left in them after a partial
+match. There is more discussion of this facility in the
+.\" HREF
+\fBpcrepartial\fP
+.\"
+documentation.
+.
+.
+.SS "Successful returns from \fBpcre_dfa_exec()\fP"
+.rs
+.sp
+When \fBpcre_dfa_exec()\fP succeeds, it may have matched more than one
+substring in the subject. Note, however, that all the matches from one run of
+the function start at the same point in the subject. The shorter matches are
+all initial substrings of the longer matches. For example, if the pattern
+.sp
+  <.*>
+.sp
+is matched against the string
+.sp
+  This is    no more
+.sp
+the three matched strings are
+.sp
+  
+   
+    
+.sp
+On success, the yield of the function is a number greater than zero, which is
+the number of matched substrings. The substrings themselves are returned in
+\fIovector\fP. Each string uses two elements; the first is the offset to the
+start, and the second is the offset to the end. In fact, all the strings have
+the same start offset. (Space could have been saved by giving this only once,
+but it was decided to retain some compatibility with the way \fBpcre_exec()\fP
+returns data, even though the meaning of the strings is different.)
+.P
+The strings are returned in reverse order of length; that is, the longest
+matching string is given first. If there were too many matches to fit into
+\fIovector\fP, the yield of the function is zero, and the vector is filled with
+the longest matches. Unlike \fBpcre_exec()\fP, \fBpcre_dfa_exec()\fP can use
+the entire \fIovector\fP for returning matched strings.
+.
+.
+.SS "Error returns from \fBpcre_dfa_exec()\fP"
+.rs
+.sp
+The \fBpcre_dfa_exec()\fP function returns a negative number when it fails.
+Many of the errors are the same as for \fBpcre_exec()\fP, and these are
+described
+.\" HTML 
+.\" 
+above.
+.\"
+There are in addition the following errors that are specific to
+\fBpcre_dfa_exec()\fP:
+.sp
+  PCRE_ERROR_DFA_UITEM      (-16)
+.sp
+This return is given if \fBpcre_dfa_exec()\fP encounters an item in the pattern
+that it does not support, for instance, the use of \eC or a back reference.
+.sp
+  PCRE_ERROR_DFA_UCOND      (-17)
+.sp
+This return is given if \fBpcre_dfa_exec()\fP encounters a condition item that
+uses a back reference for the condition, or a test for recursion in a specific
+group. These are not supported.
+.sp
+  PCRE_ERROR_DFA_UMLIMIT    (-18)
+.sp
+This return is given if \fBpcre_dfa_exec()\fP is called with an \fIextra\fP
+block that contains a setting of the \fImatch_limit\fP or
+\fImatch_limit_recursion\fP fields. This is not supported (these fields are
+meaningless for DFA matching).
+.sp
+  PCRE_ERROR_DFA_WSSIZE     (-19)
+.sp
+This return is given if \fBpcre_dfa_exec()\fP runs out of space in the
+\fIworkspace\fP vector.
+.sp
+  PCRE_ERROR_DFA_RECURSE    (-20)
+.sp
+When a recursive subpattern is processed, the matching function calls itself
+recursively, using private vectors for \fIovector\fP and \fIworkspace\fP. This
+error is given if the output vector is not large enough. This should be
+extremely rare, as a vector of size 1000 is used.
+.sp
+  PCRE_ERROR_DFA_BADRESTART (-30)
+.sp
+When \fBpcre_dfa_exec()\fP is called with the \fBPCRE_DFA_RESTART\fP option,
+some plausibility checks are made on the contents of the workspace, which
+should contain data about the previous partial match. If any of these checks
+fail, this error is given.
+.
+.
+.SH "SEE ALSO"
+.rs
+.sp
+\fBpcre16\fP(3), \fBpcre32\fP(3), \fBpcrebuild\fP(3), \fBpcrecallout\fP(3),
+\fBpcrecpp(3)\fP(3), \fBpcrematching\fP(3), \fBpcrepartial\fP(3),
+\fBpcreposix\fP(3), \fBpcreprecompile\fP(3), \fBpcresample\fP(3),
+\fBpcrestack\fP(3).
+.
+.
+.SH AUTHOR
+.rs
+.sp
+.nf
+Philip Hazel
+University Computing Service
+Cambridge CB2 3QH, England.
+.fi
+.
+.
+.SH REVISION
+.rs
+.sp
+.nf
+Last updated: 12 May 2013
+Copyright (c) 1997-2013 University of Cambridge.
+.fi
diff --git a/pcre/doc/pcrebuild.3 b/pcre/doc/pcrebuild.3
new file mode 100644
index 00000000000..403f2ae32f1
--- /dev/null
+++ b/pcre/doc/pcrebuild.3
@@ -0,0 +1,550 @@
+.TH PCREBUILD 3 "12 May 2013" "PCRE 8.33"
+.SH NAME
+PCRE - Perl-compatible regular expressions
+.
+.
+.SH "BUILDING PCRE"
+.rs
+.sp
+PCRE is distributed with a \fBconfigure\fP script that can be used to build the
+library in Unix-like environments using the applications known as Autotools.
+Also in the distribution are files to support building using \fBCMake\fP
+instead of \fBconfigure\fP. The text file
+.\" HTML 
+.\" 
+\fBREADME\fP
+.\"
+contains general information about building with Autotools (some of which is
+repeated below), and also has some comments about building on various operating
+systems. There is a lot more information about building PCRE without using
+Autotools (including information about using \fBCMake\fP and building "by
+hand") in the text file called
+.\" HTML 
+.\" 
+\fBNON-AUTOTOOLS-BUILD\fP.
+.\"
+You should consult this file as well as the
+.\" HTML 
+.\" 
+\fBREADME\fP
+.\"
+file if you are building in a non-Unix-like environment.
+.
+.
+.SH "PCRE BUILD-TIME OPTIONS"
+.rs
+.sp
+The rest of this document describes the optional features of PCRE that can be
+selected when the library is compiled. It assumes use of the \fBconfigure\fP
+script, where the optional features are selected or deselected by providing
+options to \fBconfigure\fP before running the \fBmake\fP command. However, the
+same options can be selected in both Unix-like and non-Unix-like environments
+using the GUI facility of \fBcmake-gui\fP if you are using \fBCMake\fP instead
+of \fBconfigure\fP to build PCRE.
+.P
+If you are not using Autotools or \fBCMake\fP, option selection can be done by
+editing the \fBconfig.h\fP file, or by passing parameter settings to the
+compiler, as described in
+.\" HTML 
+.\" 
+\fBNON-AUTOTOOLS-BUILD\fP.
+.\"
+.P
+The complete list of options for \fBconfigure\fP (which includes the standard
+ones such as the selection of the installation directory) can be obtained by
+running
+.sp
+  ./configure --help
+.sp
+The following sections include descriptions of options whose names begin with
+--enable or --disable. These settings specify changes to the defaults for the
+\fBconfigure\fP command. Because of the way that \fBconfigure\fP works,
+--enable and --disable always come in pairs, so the complementary option always
+exists as well, but as it specifies the default, it is not described.
+.
+.
+.SH "BUILDING 8-BIT, 16-BIT AND 32-BIT LIBRARIES"
+.rs
+.sp
+By default, a library called \fBlibpcre\fP is built, containing functions that
+take string arguments contained in vectors of bytes, either as single-byte
+characters, or interpreted as UTF-8 strings. You can also build a separate
+library, called \fBlibpcre16\fP, in which strings are contained in vectors of
+16-bit data units and interpreted either as single-unit characters or UTF-16
+strings, by adding
+.sp
+  --enable-pcre16
+.sp
+to the \fBconfigure\fP command. You can also build yet another separate
+library, called \fBlibpcre32\fP, in which strings are contained in vectors of
+32-bit data units and interpreted either as single-unit characters or UTF-32
+strings, by adding
+.sp
+  --enable-pcre32
+.sp
+to the \fBconfigure\fP command. If you do not want the 8-bit library, add
+.sp
+  --disable-pcre8
+.sp
+as well. At least one of the three libraries must be built. Note that the C++
+and POSIX wrappers are for the 8-bit library only, and that \fBpcregrep\fP is
+an 8-bit program. None of these are built if you select only the 16-bit or
+32-bit libraries.
+.
+.
+.SH "BUILDING SHARED AND STATIC LIBRARIES"
+.rs
+.sp
+The Autotools PCRE building process uses \fBlibtool\fP to build both shared and
+static libraries by default. You can suppress one of these by adding one of
+.sp
+  --disable-shared
+  --disable-static
+.sp
+to the \fBconfigure\fP command, as required.
+.
+.
+.SH "C++ SUPPORT"
+.rs
+.sp
+By default, if the 8-bit library is being built, the \fBconfigure\fP script
+will search for a C++ compiler and C++ header files. If it finds them, it
+automatically builds the C++ wrapper library (which supports only 8-bit
+strings). You can disable this by adding
+.sp
+  --disable-cpp
+.sp
+to the \fBconfigure\fP command.
+.
+.
+.SH "UTF-8, UTF-16 AND UTF-32 SUPPORT"
+.rs
+.sp
+To build PCRE with support for UTF Unicode character strings, add
+.sp
+  --enable-utf
+.sp
+to the \fBconfigure\fP command. This setting applies to all three libraries,
+adding support for UTF-8 to the 8-bit library, support for UTF-16 to the 16-bit
+library, and support for UTF-32 to the to the 32-bit library. There are no
+separate options for enabling UTF-8, UTF-16 and UTF-32 independently because
+that would allow ridiculous settings such as requesting UTF-16 support while
+building only the 8-bit library. It is not possible to build one library with
+UTF support and another without in the same configuration. (For backwards
+compatibility, --enable-utf8 is a synonym of --enable-utf.)
+.P
+Of itself, this setting does not make PCRE treat strings as UTF-8, UTF-16 or
+UTF-32. As well as compiling PCRE with this option, you also have have to set
+the PCRE_UTF8, PCRE_UTF16 or PCRE_UTF32 option (as appropriate) when you call
+one of the pattern compiling functions.
+.P
+If you set --enable-utf when compiling in an EBCDIC environment, PCRE expects
+its input to be either ASCII or UTF-8 (depending on the run-time option). It is
+not possible to support both EBCDIC and UTF-8 codes in the same version of the
+library. Consequently, --enable-utf and --enable-ebcdic are mutually
+exclusive.
+.
+.
+.SH "UNICODE CHARACTER PROPERTY SUPPORT"
+.rs
+.sp
+UTF support allows the libraries to process character codepoints up to 0x10ffff
+in the strings that they handle. On its own, however, it does not provide any
+facilities for accessing the properties of such characters. If you want to be
+able to use the pattern escapes \eP, \ep, and \eX, which refer to Unicode
+character properties, you must add
+.sp
+  --enable-unicode-properties
+.sp
+to the \fBconfigure\fP command. This implies UTF support, even if you have
+not explicitly requested it.
+.P
+Including Unicode property support adds around 30K of tables to the PCRE
+library. Only the general category properties such as \fILu\fP and \fINd\fP are
+supported. Details are given in the
+.\" HREF
+\fBpcrepattern\fP
+.\"
+documentation.
+.
+.
+.SH "JUST-IN-TIME COMPILER SUPPORT"
+.rs
+.sp
+Just-in-time compiler support is included in the build by specifying
+.sp
+  --enable-jit
+.sp
+This support is available only for certain hardware architectures. If this
+option is set for an unsupported architecture, a compile time error occurs.
+See the
+.\" HREF
+\fBpcrejit\fP
+.\"
+documentation for a discussion of JIT usage. When JIT support is enabled,
+pcregrep automatically makes use of it, unless you add
+.sp
+  --disable-pcregrep-jit
+.sp
+to the "configure" command.
+.
+.
+.SH "CODE VALUE OF NEWLINE"
+.rs
+.sp
+By default, PCRE interprets the linefeed (LF) character as indicating the end
+of a line. This is the normal newline character on Unix-like systems. You can
+compile PCRE to use carriage return (CR) instead, by adding
+.sp
+  --enable-newline-is-cr
+.sp
+to the \fBconfigure\fP command. There is also a --enable-newline-is-lf option,
+which explicitly specifies linefeed as the newline character.
+.sp
+Alternatively, you can specify that line endings are to be indicated by the two
+character sequence CRLF. If you want this, add
+.sp
+  --enable-newline-is-crlf
+.sp
+to the \fBconfigure\fP command. There is a fourth option, specified by
+.sp
+  --enable-newline-is-anycrlf
+.sp
+which causes PCRE to recognize any of the three sequences CR, LF, or CRLF as
+indicating a line ending. Finally, a fifth option, specified by
+.sp
+  --enable-newline-is-any
+.sp
+causes PCRE to recognize any Unicode newline sequence.
+.P
+Whatever line ending convention is selected when PCRE is built can be
+overridden when the library functions are called. At build time it is
+conventional to use the standard for your operating system.
+.
+.
+.SH "WHAT \eR MATCHES"
+.rs
+.sp
+By default, the sequence \eR in a pattern matches any Unicode newline sequence,
+whatever has been selected as the line ending sequence. If you specify
+.sp
+  --enable-bsr-anycrlf
+.sp
+the default is changed so that \eR matches only CR, LF, or CRLF. Whatever is
+selected when PCRE is built can be overridden when the library functions are
+called.
+.
+.
+.SH "POSIX MALLOC USAGE"
+.rs
+.sp
+When the 8-bit library is called through the POSIX interface (see the
+.\" HREF
+\fBpcreposix\fP
+.\"
+documentation), additional working storage is required for holding the pointers
+to capturing substrings, because PCRE requires three integers per substring,
+whereas the POSIX interface provides only two. If the number of expected
+substrings is small, the wrapper function uses space on the stack, because this
+is faster than using \fBmalloc()\fP for each call. The default threshold above
+which the stack is no longer used is 10; it can be changed by adding a setting
+such as
+.sp
+  --with-posix-malloc-threshold=20
+.sp
+to the \fBconfigure\fP command.
+.
+.
+.SH "HANDLING VERY LARGE PATTERNS"
+.rs
+.sp
+Within a compiled pattern, offset values are used to point from one part to
+another (for example, from an opening parenthesis to an alternation
+metacharacter). By default, in the 8-bit and 16-bit libraries, two-byte values
+are used for these offsets, leading to a maximum size for a compiled pattern of
+around 64K. This is sufficient to handle all but the most gigantic patterns.
+Nevertheless, some people do want to process truly enormous patterns, so it is
+possible to compile PCRE to use three-byte or four-byte offsets by adding a
+setting such as
+.sp
+  --with-link-size=3
+.sp
+to the \fBconfigure\fP command. The value given must be 2, 3, or 4. For the
+16-bit library, a value of 3 is rounded up to 4. In these libraries, using
+longer offsets slows down the operation of PCRE because it has to load
+additional data when handling them. For the 32-bit library the value is always
+4 and cannot be overridden; the value of --with-link-size is ignored.
+.
+.
+.SH "AVOIDING EXCESSIVE STACK USAGE"
+.rs
+.sp
+When matching with the \fBpcre_exec()\fP function, PCRE implements backtracking
+by making recursive calls to an internal function called \fBmatch()\fP. In
+environments where the size of the stack is limited, this can severely limit
+PCRE's operation. (The Unix environment does not usually suffer from this
+problem, but it may sometimes be necessary to increase the maximum stack size.
+There is a discussion in the
+.\" HREF
+\fBpcrestack\fP
+.\"
+documentation.) An alternative approach to recursion that uses memory from the
+heap to remember data, instead of using recursive function calls, has been
+implemented to work round the problem of limited stack size. If you want to
+build a version of PCRE that works this way, add
+.sp
+  --disable-stack-for-recursion
+.sp
+to the \fBconfigure\fP command. With this configuration, PCRE will use the
+\fBpcre_stack_malloc\fP and \fBpcre_stack_free\fP variables to call memory
+management functions. By default these point to \fBmalloc()\fP and
+\fBfree()\fP, but you can replace the pointers so that your own functions are
+used instead.
+.P
+Separate functions are provided rather than using \fBpcre_malloc\fP and
+\fBpcre_free\fP because the usage is very predictable: the block sizes
+requested are always the same, and the blocks are always freed in reverse
+order. A calling program might be able to implement optimized functions that
+perform better than \fBmalloc()\fP and \fBfree()\fP. PCRE runs noticeably more
+slowly when built in this way. This option affects only the \fBpcre_exec()\fP
+function; it is not relevant for \fBpcre_dfa_exec()\fP.
+.
+.
+.SH "LIMITING PCRE RESOURCE USAGE"
+.rs
+.sp
+Internally, PCRE has a function called \fBmatch()\fP, which it calls repeatedly
+(sometimes recursively) when matching a pattern with the \fBpcre_exec()\fP
+function. By controlling the maximum number of times this function may be
+called during a single matching operation, a limit can be placed on the
+resources used by a single call to \fBpcre_exec()\fP. The limit can be changed
+at run time, as described in the
+.\" HREF
+\fBpcreapi\fP
+.\"
+documentation. The default is 10 million, but this can be changed by adding a
+setting such as
+.sp
+  --with-match-limit=500000
+.sp
+to the \fBconfigure\fP command. This setting has no effect on the
+\fBpcre_dfa_exec()\fP matching function.
+.P
+In some environments it is desirable to limit the depth of recursive calls of
+\fBmatch()\fP more strictly than the total number of calls, in order to
+restrict the maximum amount of stack (or heap, if --disable-stack-for-recursion
+is specified) that is used. A second limit controls this; it defaults to the
+value that is set for --with-match-limit, which imposes no additional
+constraints. However, you can set a lower limit by adding, for example,
+.sp
+  --with-match-limit-recursion=10000
+.sp
+to the \fBconfigure\fP command. This value can also be overridden at run time.
+.
+.
+.SH "CREATING CHARACTER TABLES AT BUILD TIME"
+.rs
+.sp
+PCRE uses fixed tables for processing characters whose code values are less
+than 256. By default, PCRE is built with a set of tables that are distributed
+in the file \fIpcre_chartables.c.dist\fP. These tables are for ASCII codes
+only. If you add
+.sp
+  --enable-rebuild-chartables
+.sp
+to the \fBconfigure\fP command, the distributed tables are no longer used.
+Instead, a program called \fBdftables\fP is compiled and run. This outputs the
+source for new set of tables, created in the default locale of your C run-time
+system. (This method of replacing the tables does not work if you are cross
+compiling, because \fBdftables\fP is run on the local host. If you need to
+create alternative tables when cross compiling, you will have to do so "by
+hand".)
+.
+.
+.SH "USING EBCDIC CODE"
+.rs
+.sp
+PCRE assumes by default that it will run in an environment where the character
+code is ASCII (or Unicode, which is a superset of ASCII). This is the case for
+most computer operating systems. PCRE can, however, be compiled to run in an
+EBCDIC environment by adding
+.sp
+  --enable-ebcdic
+.sp
+to the \fBconfigure\fP command. This setting implies
+--enable-rebuild-chartables. You should only use it if you know that you are in
+an EBCDIC environment (for example, an IBM mainframe operating system). The
+--enable-ebcdic option is incompatible with --enable-utf.
+.P
+The EBCDIC character that corresponds to an ASCII LF is assumed to have the
+value 0x15 by default. However, in some EBCDIC environments, 0x25 is used. In
+such an environment you should use
+.sp
+  --enable-ebcdic-nl25
+.sp
+as well as, or instead of, --enable-ebcdic. The EBCDIC character for CR has the
+same value as in ASCII, namely, 0x0d. Whichever of 0x15 and 0x25 is \fInot\fP
+chosen as LF is made to correspond to the Unicode NEL character (which, in
+Unicode, is 0x85).
+.P
+The options that select newline behaviour, such as --enable-newline-is-cr,
+and equivalent run-time options, refer to these character values in an EBCDIC
+environment.
+.
+.
+.SH "PCREGREP OPTIONS FOR COMPRESSED FILE SUPPORT"
+.rs
+.sp
+By default, \fBpcregrep\fP reads all files as plain text. You can build it so
+that it recognizes files whose names end in \fB.gz\fP or \fB.bz2\fP, and reads
+them with \fBlibz\fP or \fBlibbz2\fP, respectively, by adding one or both of
+.sp
+  --enable-pcregrep-libz
+  --enable-pcregrep-libbz2
+.sp
+to the \fBconfigure\fP command. These options naturally require that the
+relevant libraries are installed on your system. Configuration will fail if
+they are not.
+.
+.
+.SH "PCREGREP BUFFER SIZE"
+.rs
+.sp
+\fBpcregrep\fP uses an internal buffer to hold a "window" on the file it is
+scanning, in order to be able to output "before" and "after" lines when it
+finds a match. The size of the buffer is controlled by a parameter whose
+default value is 20K. The buffer itself is three times this size, but because
+of the way it is used for holding "before" lines, the longest line that is
+guaranteed to be processable is the parameter size. You can change the default
+parameter value by adding, for example,
+.sp
+  --with-pcregrep-bufsize=50K
+.sp
+to the \fBconfigure\fP command. The caller of \fPpcregrep\fP can, however,
+override this value by specifying a run-time option.
+.
+.
+.SH "PCRETEST OPTION FOR LIBREADLINE SUPPORT"
+.rs
+.sp
+If you add
+.sp
+  --enable-pcretest-libreadline
+.sp
+to the \fBconfigure\fP command, \fBpcretest\fP is linked with the
+\fBlibreadline\fP library, and when its input is from a terminal, it reads it
+using the \fBreadline()\fP function. This provides line-editing and history
+facilities. Note that \fBlibreadline\fP is GPL-licensed, so if you distribute a
+binary of \fBpcretest\fP linked in this way, there may be licensing issues.
+.P
+Setting this option causes the \fB-lreadline\fP option to be added to the
+\fBpcretest\fP build. In many operating environments with a sytem-installed
+\fBlibreadline\fP this is sufficient. However, in some environments (e.g.
+if an unmodified distribution version of readline is in use), some extra
+configuration may be necessary. The INSTALL file for \fBlibreadline\fP says
+this:
+.sp
+  "Readline uses the termcap functions, but does not link with the
+  termcap or curses library itself, allowing applications which link
+  with readline the to choose an appropriate library."
+.sp
+If your environment has not been set up so that an appropriate library is
+automatically included, you may need to add something like
+.sp
+  LIBS="-ncurses"
+.sp
+immediately before the \fBconfigure\fP command.
+.
+.
+.SH "DEBUGGING WITH VALGRIND SUPPORT"
+.rs
+.sp
+By adding the
+.sp
+  --enable-valgrind
+.sp
+option to to the \fBconfigure\fP command, PCRE will use valgrind annotations
+to mark certain memory regions as unaddressable. This allows it to detect
+invalid memory accesses, and is mostly useful for debugging PCRE itself.
+.
+.
+.SH "CODE COVERAGE REPORTING"
+.rs
+.sp
+If your C compiler is gcc, you can build a version of PCRE that can generate a
+code coverage report for its test suite. To enable this, you must install
+\fBlcov\fP version 1.6 or above. Then specify
+.sp
+  --enable-coverage
+.sp
+to the \fBconfigure\fP command and build PCRE in the usual way.
+.P
+Note that using \fBccache\fP (a caching C compiler) is incompatible with code
+coverage reporting. If you have configured \fBccache\fP to run automatically
+on your system, you must set the environment variable
+.sp
+  CCACHE_DISABLE=1
+.sp
+before running \fBmake\fP to build PCRE, so that \fBccache\fP is not used.
+.P
+When --enable-coverage is used, the following addition targets are added to the
+\fIMakefile\fP:
+.sp
+  make coverage
+.sp
+This creates a fresh coverage report for the PCRE test suite. It is equivalent
+to running "make coverage-reset", "make coverage-baseline", "make check", and
+then "make coverage-report".
+.sp
+  make coverage-reset
+.sp
+This zeroes the coverage counters, but does nothing else.
+.sp
+  make coverage-baseline
+.sp
+This captures baseline coverage information.
+.sp
+  make coverage-report
+.sp
+This creates the coverage report.
+.sp
+  make coverage-clean-report
+.sp
+This removes the generated coverage report without cleaning the coverage data
+itself.
+.sp
+  make coverage-clean-data
+.sp
+This removes the captured coverage data without removing the coverage files
+created at compile time (*.gcno).
+.sp
+  make coverage-clean
+.sp
+This cleans all coverage data including the generated coverage report. For more
+information about code coverage, see the \fBgcov\fP and \fBlcov\fP
+documentation.
+.
+.
+.SH "SEE ALSO"
+.rs
+.sp
+\fBpcreapi\fP(3), \fBpcre16\fP, \fBpcre32\fP, \fBpcre_config\fP(3).
+.
+.
+.SH AUTHOR
+.rs
+.sp
+.nf
+Philip Hazel
+University Computing Service
+Cambridge CB2 3QH, England.
+.fi
+.
+.
+.SH REVISION
+.rs
+.sp
+.nf
+Last updated: 12 May 2013
+Copyright (c) 1997-2013 University of Cambridge.
+.fi
diff --git a/pcre/doc/pcrecallout.3 b/pcre/doc/pcrecallout.3
new file mode 100644
index 00000000000..79e2bb931aa
--- /dev/null
+++ b/pcre/doc/pcrecallout.3
@@ -0,0 +1,222 @@
+.TH PCRECALLOUT 3 "03 March 2013" "PCRE 8.33"
+.SH NAME
+PCRE - Perl-compatible regular expressions
+.SH SYNOPSIS
+.rs
+.sp
+.B #include 
+.PP
+.SM
+.B int (*pcre_callout)(pcre_callout_block *);
+.PP
+.B int (*pcre16_callout)(pcre16_callout_block *);
+.PP
+.B int (*pcre32_callout)(pcre32_callout_block *);
+.
+.SH DESCRIPTION
+.rs
+.sp
+PCRE provides a feature called "callout", which is a means of temporarily
+passing control to the caller of PCRE in the middle of pattern matching. The
+caller of PCRE provides an external function by putting its entry point in the
+global variable \fIpcre_callout\fP (\fIpcre16_callout\fP for the 16-bit
+library, \fIpcre32_callout\fP for the 32-bit library). By default, this
+variable contains NULL, which disables all calling out.
+.P
+Within a regular expression, (?C) indicates the points at which the external
+function is to be called. Different callout points can be identified by putting
+a number less than 256 after the letter C. The default value is zero.
+For example, this pattern has two callout points:
+.sp
+  (?C1)abc(?C2)def
+.sp
+If the PCRE_AUTO_CALLOUT option bit is set when a pattern is compiled, PCRE
+automatically inserts callouts, all with number 255, before each item in the
+pattern. For example, if PCRE_AUTO_CALLOUT is used with the pattern
+.sp
+  A(\ed{2}|--)
+.sp
+it is processed as if it were
+.sp
+(?C255)A(?C255)((?C255)\ed{2}(?C255)|(?C255)-(?C255)-(?C255))(?C255)
+.sp
+Notice that there is a callout before and after each parenthesis and
+alternation bar. If the pattern contains a conditional group whose condition is
+an assertion, an automatic callout is inserted immediately before the
+condition. Such a callout may also be inserted explicitly, for example:
+.sp
+  (?(?C9)(?=a)ab|de)
+.sp
+This applies only to assertion conditions (because they are themselves
+independent groups).
+.P
+Automatic callouts can be used for tracking the progress of pattern matching.
+The
+.\" HREF
+\fBpcretest\fP
+.\"
+command has an option that sets automatic callouts; when it is used, the output
+indicates how the pattern is matched. This is useful information when you are
+trying to optimize the performance of a particular pattern.
+.
+.
+.SH "MISSING CALLOUTS"
+.rs
+.sp
+You should be aware that, because of optimizations in the way PCRE matches
+patterns by default, callouts sometimes do not happen. For example, if the
+pattern is
+.sp
+  ab(?C4)cd
+.sp
+PCRE knows that any matching string must contain the letter "d". If the subject
+string is "abyz", the lack of "d" means that matching doesn't ever start, and
+the callout is never reached. However, with "abyd", though the result is still
+no match, the callout is obeyed.
+.P
+If the pattern is studied, PCRE knows the minimum length of a matching string,
+and will immediately give a "no match" return without actually running a match
+if the subject is not long enough, or, for unanchored patterns, if it has
+been scanned far enough.
+.P
+You can disable these optimizations by passing the PCRE_NO_START_OPTIMIZE
+option to the matching function, or by starting the pattern with
+(*NO_START_OPT). This slows down the matching process, but does ensure that
+callouts such as the example above are obeyed.
+.
+.
+.SH "THE CALLOUT INTERFACE"
+.rs
+.sp
+During matching, when PCRE reaches a callout point, the external function
+defined by \fIpcre_callout\fP or \fIpcre[16|32]_callout\fP is called
+(if it is set). This applies to both normal and DFA matching. The only
+argument to the callout function is a pointer to a \fBpcre_callout\fP
+or \fBpcre[16|32]_callout\fP block.
+These structures contains the following fields:
+.sp
+  int           \fIversion\fP;
+  int           \fIcallout_number\fP;
+  int          *\fIoffset_vector\fP;
+  const char   *\fIsubject\fP;           (8-bit version)
+  PCRE_SPTR16   \fIsubject\fP;           (16-bit version)
+  PCRE_SPTR32   \fIsubject\fP;           (32-bit version)
+  int           \fIsubject_length\fP;
+  int           \fIstart_match\fP;
+  int           \fIcurrent_position\fP;
+  int           \fIcapture_top\fP;
+  int           \fIcapture_last\fP;
+  void         *\fIcallout_data\fP;
+  int           \fIpattern_position\fP;
+  int           \fInext_item_length\fP;
+  const unsigned char *\fImark\fP;       (8-bit version)
+  const PCRE_UCHAR16  *\fImark\fP;       (16-bit version)
+  const PCRE_UCHAR32  *\fImark\fP;       (32-bit version)
+.sp
+The \fIversion\fP field is an integer containing the version number of the
+block format. The initial version was 0; the current version is 2. The version
+number will change again in future if additional fields are added, but the
+intention is never to remove any of the existing fields.
+.P
+The \fIcallout_number\fP field contains the number of the callout, as compiled
+into the pattern (that is, the number after ?C for manual callouts, and 255 for
+automatically generated callouts).
+.P
+The \fIoffset_vector\fP field is a pointer to the vector of offsets that was
+passed by the caller to the matching function. When \fBpcre_exec()\fP or
+\fBpcre[16|32]_exec()\fP is used, the contents can be inspected, in order to
+extract substrings that have been matched so far, in the same way as for
+extracting substrings after a match has completed. For the DFA matching
+functions, this field is not useful.
+.P
+The \fIsubject\fP and \fIsubject_length\fP fields contain copies of the values
+that were passed to the matching function.
+.P
+The \fIstart_match\fP field normally contains the offset within the subject at
+which the current match attempt started. However, if the escape sequence \eK
+has been encountered, this value is changed to reflect the modified starting
+point. If the pattern is not anchored, the callout function may be called
+several times from the same point in the pattern for different starting points
+in the subject.
+.P
+The \fIcurrent_position\fP field contains the offset within the subject of the
+current match pointer.
+.P
+When the \fBpcre_exec()\fP or \fBpcre[16|32]_exec()\fP is used, the
+\fIcapture_top\fP field contains one more than the number of the highest
+numbered captured substring so far. If no substrings have been captured, the
+value of \fIcapture_top\fP is one. This is always the case when the DFA
+functions are used, because they do not support captured substrings.
+.P
+The \fIcapture_last\fP field contains the number of the most recently captured
+substring. However, when a recursion exits, the value reverts to what it was
+outside the recursion, as do the values of all captured substrings. If no
+substrings have been captured, the value of \fIcapture_last\fP is -1. This is
+always the case for the DFA matching functions.
+.P
+The \fIcallout_data\fP field contains a value that is passed to a matching
+function specifically so that it can be passed back in callouts. It is passed
+in the \fIcallout_data\fP field of a \fBpcre_extra\fP or \fBpcre[16|32]_extra\fP
+data structure. If no such data was passed, the value of \fIcallout_data\fP in
+a callout block is NULL. There is a description of the \fBpcre_extra\fP
+structure in the
+.\" HREF
+\fBpcreapi\fP
+.\"
+documentation.
+.P
+The \fIpattern_position\fP field is present from version 1 of the callout
+structure. It contains the offset to the next item to be matched in the pattern
+string.
+.P
+The \fInext_item_length\fP field is present from version 1 of the callout
+structure. It contains the length of the next item to be matched in the pattern
+string. When the callout immediately precedes an alternation bar, a closing
+parenthesis, or the end of the pattern, the length is zero. When the callout
+precedes an opening parenthesis, the length is that of the entire subpattern.
+.P
+The \fIpattern_position\fP and \fInext_item_length\fP fields are intended to
+help in distinguishing between different automatic callouts, which all have the
+same callout number. However, they are set for all callouts.
+.P
+The \fImark\fP field is present from version 2 of the callout structure. In
+callouts from \fBpcre_exec()\fP or \fBpcre[16|32]_exec()\fP it contains a
+pointer to the zero-terminated name of the most recently passed (*MARK),
+(*PRUNE), or (*THEN) item in the match, or NULL if no such items have been
+passed. Instances of (*PRUNE) or (*THEN) without a name do not obliterate a
+previous (*MARK). In callouts from the DFA matching functions this field always
+contains NULL.
+.
+.
+.SH "RETURN VALUES"
+.rs
+.sp
+The external callout function returns an integer to PCRE. If the value is zero,
+matching proceeds as normal. If the value is greater than zero, matching fails
+at the current point, but the testing of other matching possibilities goes
+ahead, just as if a lookahead assertion had failed. If the value is less than
+zero, the match is abandoned, the matching function returns the negative value.
+.P
+Negative values should normally be chosen from the set of PCRE_ERROR_xxx
+values. In particular, PCRE_ERROR_NOMATCH forces a standard "no match" failure.
+The error number PCRE_ERROR_CALLOUT is reserved for use by callout functions;
+it will never be used by PCRE itself.
+.
+.
+.SH AUTHOR
+.rs
+.sp
+.nf
+Philip Hazel
+University Computing Service
+Cambridge CB2 3QH, England.
+.fi
+.
+.
+.SH REVISION
+.rs
+.sp
+.nf
+Last updated: 03 March 2013
+Copyright (c) 1997-2013 University of Cambridge.
+.fi
diff --git a/pcre/doc/pcrecompat.3 b/pcre/doc/pcrecompat.3
new file mode 100644
index 00000000000..2c109ea9999
--- /dev/null
+++ b/pcre/doc/pcrecompat.3
@@ -0,0 +1,195 @@
+.TH PCRECOMPAT 3 "19 March 2013" "PCRE 8.33"
+.SH NAME
+PCRE - Perl-compatible regular expressions
+.SH "DIFFERENCES BETWEEN PCRE AND PERL"
+.rs
+.sp
+This document describes the differences in the ways that PCRE and Perl handle
+regular expressions. The differences described here are with respect to Perl
+versions 5.10 and above.
+.P
+1. PCRE has only a subset of Perl's Unicode support. Details of what it does
+have are given in the
+.\" HREF
+\fBpcreunicode\fP
+.\"
+page.
+.P
+2. PCRE allows repeat quantifiers only on parenthesized assertions, but they do
+not mean what you might think. For example, (?!a){3} does not assert that the
+next three characters are not "a". It just asserts that the next character is
+not "a" three times (in principle: PCRE optimizes this to run the assertion
+just once). Perl allows repeat quantifiers on other assertions such as \eb, but
+these do not seem to have any use.
+.P
+3. Capturing subpatterns that occur inside negative lookahead assertions are
+counted, but their entries in the offsets vector are never set. Perl sometimes
+(but not always) sets its numerical variables from inside negative assertions.
+.P
+4. Though binary zero characters are supported in the subject string, they are
+not allowed in a pattern string because it is passed as a normal C string,
+terminated by zero. The escape sequence \e0 can be used in the pattern to
+represent a binary zero.
+.P
+5. The following Perl escape sequences are not supported: \el, \eu, \eL,
+\eU, and \eN when followed by a character name or Unicode value. (\eN on its
+own, matching a non-newline character, is supported.) In fact these are
+implemented by Perl's general string-handling and are not part of its pattern
+matching engine. If any of these are encountered by PCRE, an error is
+generated by default. However, if the PCRE_JAVASCRIPT_COMPAT option is set,
+\eU and \eu are interpreted as JavaScript interprets them.
+.P
+6. The Perl escape sequences \ep, \eP, and \eX are supported only if PCRE is
+built with Unicode character property support. The properties that can be
+tested with \ep and \eP are limited to the general category properties such as
+Lu and Nd, script names such as Greek or Han, and the derived properties Any
+and L&. PCRE does support the Cs (surrogate) property, which Perl does not; the
+Perl documentation says "Because Perl hides the need for the user to understand
+the internal representation of Unicode characters, there is no need to
+implement the somewhat messy concept of surrogates."
+.P
+7. PCRE does support the \eQ...\eE escape for quoting substrings. Characters in
+between are treated as literals. This is slightly different from Perl in that $
+and @ are also handled as literals inside the quotes. In Perl, they cause
+variable interpolation (but of course PCRE does not have variables). Note the
+following examples:
+.sp
+    Pattern            PCRE matches      Perl matches
+.sp
+.\" JOIN
+    \eQabc$xyz\eE        abc$xyz           abc followed by the
+                                           contents of $xyz
+    \eQabc\e$xyz\eE       abc\e$xyz          abc\e$xyz
+    \eQabc\eE\e$\eQxyz\eE   abc$xyz           abc$xyz
+.sp
+The \eQ...\eE sequence is recognized both inside and outside character classes.
+.P
+8. Fairly obviously, PCRE does not support the (?{code}) and (??{code})
+constructions. However, there is support for recursive patterns. This is not
+available in Perl 5.8, but it is in Perl 5.10. Also, the PCRE "callout"
+feature allows an external function to be called during pattern matching. See
+the
+.\" HREF
+\fBpcrecallout\fP
+.\"
+documentation for details.
+.P
+9. Subpatterns that are called as subroutines (whether or not recursively) are
+always treated as atomic groups in PCRE. This is like Python, but unlike Perl.
+Captured values that are set outside a subroutine call can be reference from
+inside in PCRE, but not in Perl. There is a discussion that explains these
+differences in more detail in the
+.\" HTML 
+.\" 
+section on recursion differences from Perl
+.\"
+in the
+.\" HREF
+\fBpcrepattern\fP
+.\"
+page.
+.P
+10. If any of the backtracking control verbs are used in a subpattern that is
+called as a subroutine (whether or not recursively), their effect is confined
+to that subpattern; it does not extend to the surrounding pattern. This is not
+always the case in Perl. In particular, if (*THEN) is present in a group that
+is called as a subroutine, its action is limited to that group, even if the
+group does not contain any | characters. Note that such subpatterns are
+processed as anchored at the point where they are tested.
+.P
+11. If a pattern contains more than one backtracking control verb, the first
+one that is backtracked onto acts. For example, in the pattern
+A(*COMMIT)B(*PRUNE)C a failure in B triggers (*COMMIT), but a failure in C
+triggers (*PRUNE). Perl's behaviour is more complex; in many cases it is the
+same as PCRE, but there are examples where it differs.
+.P
+12. Most backtracking verbs in assertions have their normal actions. They are
+not confined to the assertion.
+.P
+13. There are some differences that are concerned with the settings of captured
+strings when part of a pattern is repeated. For example, matching "aba" against
+the pattern /^(a(b)?)+$/ in Perl leaves $2 unset, but in PCRE it is set to "b".
+.P
+14. PCRE's handling of duplicate subpattern numbers and duplicate subpattern
+names is not as general as Perl's. This is a consequence of the fact the PCRE
+works internally just with numbers, using an external table to translate
+between numbers and names. In particular, a pattern such as (?|(?A)|(?
+.
+.SH DESCRIPTION
+.rs
+.sp
+The C++ wrapper for PCRE was provided by Google Inc. Some additional
+functionality was added by Giuseppe Maxia. This brief man page was constructed
+from the notes in the \fIpcrecpp.h\fP file, which should be consulted for
+further details. Note that the C++ wrapper supports only the original 8-bit
+PCRE library. There is no 16-bit or 32-bit support at present.
+.
+.
+.SH "MATCHING INTERFACE"
+.rs
+.sp
+The "FullMatch" operation checks that supplied text matches a supplied pattern
+exactly. If pointer arguments are supplied, it copies matched sub-strings that
+match sub-patterns into them.
+.sp
+  Example: successful match
+     pcrecpp::RE re("h.*o");
+     re.FullMatch("hello");
+.sp
+  Example: unsuccessful match (requires full match):
+     pcrecpp::RE re("e");
+     !re.FullMatch("hello");
+.sp
+  Example: creating a temporary RE object:
+     pcrecpp::RE("h.*o").FullMatch("hello");
+.sp
+You can pass in a "const char*" or a "string" for "text". The examples below
+tend to use a const char*. You can, as in the different examples above, store
+the RE object explicitly in a variable or use a temporary RE object. The
+examples below use one mode or the other arbitrarily. Either could correctly be
+used for any of these examples.
+.P
+You must supply extra pointer arguments to extract matched subpieces.
+.sp
+  Example: extracts "ruby" into "s" and 1234 into "i"
+     int i;
+     string s;
+     pcrecpp::RE re("(\e\ew+):(\e\ed+)");
+     re.FullMatch("ruby:1234", &s, &i);
+.sp
+  Example: does not try to extract any extra sub-patterns
+     re.FullMatch("ruby:1234", &s);
+.sp
+  Example: does not try to extract into NULL
+     re.FullMatch("ruby:1234", NULL, &i);
+.sp
+  Example: integer overflow causes failure
+     !re.FullMatch("ruby:1234567891234", NULL, &i);
+.sp
+  Example: fails because there aren't enough sub-patterns:
+     !pcrecpp::RE("\e\ew+:\e\ed+").FullMatch("ruby:1234", &s);
+.sp
+  Example: fails because string cannot be stored in integer
+     !pcrecpp::RE("(.*)").FullMatch("ruby", &i);
+.sp
+The provided pointer arguments can be pointers to any scalar numeric
+type, or one of:
+.sp
+   string        (matched piece is copied to string)
+   StringPiece   (StringPiece is mutated to point to matched piece)
+   T             (where "bool T::ParseFrom(const char*, int)" exists)
+   NULL          (the corresponding matched sub-pattern is not copied)
+.sp
+The function returns true iff all of the following conditions are satisfied:
+.sp
+  a. "text" matches "pattern" exactly;
+.sp
+  b. The number of matched sub-patterns is >= number of supplied
+     pointers;
+.sp
+  c. The "i"th argument has a suitable type for holding the
+     string captured as the "i"th sub-pattern. If you pass in
+     void * NULL for the "i"th argument, or a non-void * NULL
+     of the correct type, or pass fewer arguments than the
+     number of sub-patterns, "i"th captured sub-pattern is
+     ignored.
+.sp
+CAVEAT: An optional sub-pattern that does not exist in the matched
+string is assigned the empty string. Therefore, the following will
+return false (because the empty string is not a valid number):
+.sp
+   int number;
+   pcrecpp::RE::FullMatch("abc", "[a-z]+(\e\ed+)?", &number);
+.sp
+The matching interface supports at most 16 arguments per call.
+If you need more, consider using the more general interface
+\fBpcrecpp::RE::DoMatch\fP. See \fBpcrecpp.h\fP for the signature for
+\fBDoMatch\fP.
+.P
+NOTE: Do not use \fBno_arg\fP, which is used internally to mark the end of a
+list of optional arguments, as a placeholder for missing arguments, as this can
+lead to segfaults.
+.
+.
+.SH "QUOTING METACHARACTERS"
+.rs
+.sp
+You can use the "QuoteMeta" operation to insert backslashes before all
+potentially meaningful characters in a string. The returned string, used as a
+regular expression, will exactly match the original string.
+.sp
+  Example:
+     string quoted = RE::QuoteMeta(unquoted);
+.sp
+Note that it's legal to escape a character even if it has no special meaning in
+a regular expression -- so this function does that. (This also makes it
+identical to the perl function of the same name; see "perldoc -f quotemeta".)
+For example, "1.5-2.0?" becomes "1\e.5\e-2\e.0\e?".
+.
+.SH "PARTIAL MATCHES"
+.rs
+.sp
+You can use the "PartialMatch" operation when you want the pattern
+to match any substring of the text.
+.sp
+  Example: simple search for a string:
+     pcrecpp::RE("ell").PartialMatch("hello");
+.sp
+  Example: find first number in a string:
+     int number;
+     pcrecpp::RE re("(\e\ed+)");
+     re.PartialMatch("x*100 + 20", &number);
+     assert(number == 100);
+.
+.
+.SH "UTF-8 AND THE MATCHING INTERFACE"
+.rs
+.sp
+By default, pattern and text are plain text, one byte per character. The UTF8
+flag, passed to the constructor, causes both pattern and string to be treated
+as UTF-8 text, still a byte stream but potentially multiple bytes per
+character. In practice, the text is likelier to be UTF-8 than the pattern, but
+the match returned may depend on the UTF8 flag, so always use it when matching
+UTF8 text. For example, "." will match one byte normally but with UTF8 set may
+match up to three bytes of a multi-byte character.
+.sp
+  Example:
+     pcrecpp::RE_Options options;
+     options.set_utf8();
+     pcrecpp::RE re(utf8_pattern, options);
+     re.FullMatch(utf8_string);
+.sp
+  Example: using the convenience function UTF8():
+     pcrecpp::RE re(utf8_pattern, pcrecpp::UTF8());
+     re.FullMatch(utf8_string);
+.sp
+NOTE: The UTF8 flag is ignored if pcre was not configured with the
+      --enable-utf8 flag.
+.
+.
+.SH "PASSING MODIFIERS TO THE REGULAR EXPRESSION ENGINE"
+.rs
+.sp
+PCRE defines some modifiers to change the behavior of the regular expression
+engine. The C++ wrapper defines an auxiliary class, RE_Options, as a vehicle to
+pass such modifiers to a RE class. Currently, the following modifiers are
+supported:
+.sp
+   modifier              description               Perl corresponding
+.sp
+   PCRE_CASELESS         case insensitive match      /i
+   PCRE_MULTILINE        multiple lines match        /m
+   PCRE_DOTALL           dot matches newlines        /s
+   PCRE_DOLLAR_ENDONLY   $ matches only at end       N/A
+   PCRE_EXTRA            strict escape parsing       N/A
+   PCRE_EXTENDED         ignore white spaces         /x
+   PCRE_UTF8             handles UTF8 chars          built-in
+   PCRE_UNGREEDY         reverses * and *?           N/A
+   PCRE_NO_AUTO_CAPTURE  disables capturing parens   N/A (*)
+.sp
+(*) Both Perl and PCRE allow non capturing parentheses by means of the
+"?:" modifier within the pattern itself. e.g. (?:ab|cd) does not
+capture, while (ab|cd) does.
+.P
+For a full account on how each modifier works, please check the
+PCRE API reference page.
+.P
+For each modifier, there are two member functions whose name is made
+out of the modifier in lowercase, without the "PCRE_" prefix. For
+instance, PCRE_CASELESS is handled by
+.sp
+  bool caseless()
+.sp
+which returns true if the modifier is set, and
+.sp
+  RE_Options & set_caseless(bool)
+.sp
+which sets or unsets the modifier. Moreover, PCRE_EXTRA_MATCH_LIMIT can be
+accessed through the \fBset_match_limit()\fP and \fBmatch_limit()\fP member
+functions. Setting \fImatch_limit\fP to a non-zero value will limit the
+execution of pcre to keep it from doing bad things like blowing the stack or
+taking an eternity to return a result. A value of 5000 is good enough to stop
+stack blowup in a 2MB thread stack. Setting \fImatch_limit\fP to zero disables
+match limiting. Alternatively, you can call \fBmatch_limit_recursion()\fP
+which uses PCRE_EXTRA_MATCH_LIMIT_RECURSION to limit how much PCRE
+recurses. \fBmatch_limit()\fP limits the number of matches PCRE does;
+\fBmatch_limit_recursion()\fP limits the depth of internal recursion, and
+therefore the amount of stack that is used.
+.P
+Normally, to pass one or more modifiers to a RE class, you declare
+a \fIRE_Options\fP object, set the appropriate options, and pass this
+object to a RE constructor. Example:
+.sp
+   RE_Options opt;
+   opt.set_caseless(true);
+   if (RE("HELLO", opt).PartialMatch("hello world")) ...
+.sp
+RE_options has two constructors. The default constructor takes no arguments and
+creates a set of flags that are off by default. The optional parameter
+\fIoption_flags\fP is to facilitate transfer of legacy code from C programs.
+This lets you do
+.sp
+   RE(pattern,
+     RE_Options(PCRE_CASELESS|PCRE_MULTILINE)).PartialMatch(str);
+.sp
+However, new code is better off doing
+.sp
+   RE(pattern,
+     RE_Options().set_caseless(true).set_multiline(true))
+       .PartialMatch(str);
+.sp
+If you are going to pass one of the most used modifiers, there are some
+convenience functions that return a RE_Options class with the
+appropriate modifier already set: \fBCASELESS()\fP, \fBUTF8()\fP,
+\fBMULTILINE()\fP, \fBDOTALL\fP(), and \fBEXTENDED()\fP.
+.P
+If you need to set several options at once, and you don't want to go through
+the pains of declaring a RE_Options object and setting several options, there
+is a parallel method that give you such ability on the fly. You can concatenate
+several \fBset_xxxxx()\fP member functions, since each of them returns a
+reference to its class object. For example, to pass PCRE_CASELESS,
+PCRE_EXTENDED, and PCRE_MULTILINE to a RE with one statement, you may write:
+.sp
+   RE(" ^ xyz \e\es+ .* blah$",
+     RE_Options()
+       .set_caseless(true)
+       .set_extended(true)
+       .set_multiline(true)).PartialMatch(sometext);
+.sp
+.
+.
+.SH "SCANNING TEXT INCREMENTALLY"
+.rs
+.sp
+The "Consume" operation may be useful if you want to repeatedly
+match regular expressions at the front of a string and skip over
+them as they match. This requires use of the "StringPiece" type,
+which represents a sub-range of a real string. Like RE, StringPiece
+is defined in the pcrecpp namespace.
+.sp
+  Example: read lines of the form "var = value" from a string.
+     string contents = ...;                 // Fill string somehow
+     pcrecpp::StringPiece input(contents);  // Wrap in a StringPiece
+.sp
+     string var;
+     int value;
+     pcrecpp::RE re("(\e\ew+) = (\e\ed+)\en");
+     while (re.Consume(&input, &var, &value)) {
+       ...;
+     }
+.sp
+Each successful call to "Consume" will set "var/value", and also
+advance "input" so it points past the matched text.
+.P
+The "FindAndConsume" operation is similar to "Consume" but does not
+anchor your match at the beginning of the string. For example, you
+could extract all words from a string by repeatedly calling
+.sp
+  pcrecpp::RE("(\e\ew+)").FindAndConsume(&input, &word)
+.
+.
+.SH "PARSING HEX/OCTAL/C-RADIX NUMBERS"
+.rs
+.sp
+By default, if you pass a pointer to a numeric value, the
+corresponding text is interpreted as a base-10 number. You can
+instead wrap the pointer with a call to one of the operators Hex(),
+Octal(), or CRadix() to interpret the text in another base. The
+CRadix operator interprets C-style "0" (base-8) and "0x" (base-16)
+prefixes, but defaults to base-10.
+.sp
+  Example:
+    int a, b, c, d;
+    pcrecpp::RE re("(.*) (.*) (.*) (.*)");
+    re.FullMatch("100 40 0100 0x40",
+                 pcrecpp::Octal(&a), pcrecpp::Hex(&b),
+                 pcrecpp::CRadix(&c), pcrecpp::CRadix(&d));
+.sp
+will leave 64 in a, b, c, and d.
+.
+.
+.SH "REPLACING PARTS OF STRINGS"
+.rs
+.sp
+You can replace the first match of "pattern" in "str" with "rewrite".
+Within "rewrite", backslash-escaped digits (\e1 to \e9) can be
+used to insert text matching corresponding parenthesized group
+from the pattern. \e0 in "rewrite" refers to the entire matching
+text. For example:
+.sp
+  string s = "yabba dabba doo";
+  pcrecpp::RE("b+").Replace("d", &s);
+.sp
+will leave "s" containing "yada dabba doo". The result is true if the pattern
+matches and a replacement occurs, false otherwise.
+.P
+\fBGlobalReplace\fP is like \fBReplace\fP except that it replaces all
+occurrences of the pattern in the string with the rewrite. Replacements are
+not subject to re-matching. For example:
+.sp
+  string s = "yabba dabba doo";
+  pcrecpp::RE("b+").GlobalReplace("d", &s);
+.sp
+will leave "s" containing "yada dada doo". It returns the number of
+replacements made.
+.P
+\fBExtract\fP is like \fBReplace\fP, except that if the pattern matches,
+"rewrite" is copied into "out" (an additional argument) with substitutions.
+The non-matching portions of "text" are ignored. Returns true iff a match
+occurred and the extraction happened successfully;  if no match occurs, the
+string is left unaffected.
+.
+.
+.SH AUTHOR
+.rs
+.sp
+.nf
+The C++ wrapper was contributed by Google Inc.
+Copyright (c) 2007 Google Inc.
+.fi
+.
+.
+.SH REVISION
+.rs
+.sp
+.nf
+Last updated: 08 January 2012
+.fi
diff --git a/pcre/doc/pcredemo.3 b/pcre/doc/pcredemo.3
new file mode 100644
index 00000000000..194629b1fac
--- /dev/null
+++ b/pcre/doc/pcredemo.3
@@ -0,0 +1,424 @@
+.\" Start example.
+.de EX
+.  nr mE \\n(.f
+.  nf
+.  nh
+.  ft CW
+..
+.
+.
+.\" End example.
+.de EE
+.  ft \\n(mE
+.  fi
+.  hy \\n(HY
+..
+.
+.EX
+/*************************************************
+*           PCRE DEMONSTRATION PROGRAM           *
+*************************************************/
+
+/* This is a demonstration program to illustrate the most straightforward ways
+of calling the PCRE regular expression library from a C program. See the
+pcresample documentation for a short discussion ("man pcresample" if you have
+the PCRE man pages installed).
+
+In Unix-like environments, if PCRE is installed in your standard system
+libraries, you should be able to compile this program using this command:
+
+gcc -Wall pcredemo.c -lpcre -o pcredemo
+
+If PCRE is not installed in a standard place, it is likely to be installed with
+support for the pkg-config mechanism. If you have pkg-config, you can compile
+this program using this command:
+
+gcc -Wall pcredemo.c `pkg-config --cflags --libs libpcre` -o pcredemo
+
+If you do not have pkg-config, you may have to use this:
+
+gcc -Wall pcredemo.c -I/usr/local/include -L/usr/local/lib \e
+  -R/usr/local/lib -lpcre -o pcredemo
+
+Replace "/usr/local/include" and "/usr/local/lib" with wherever the include and
+library files for PCRE are installed on your system. Only some operating
+systems (e.g. Solaris) use the -R option.
+
+Building under Windows:
+
+If you want to statically link this program against a non-dll .a file, you must
+define PCRE_STATIC before including pcre.h, otherwise the pcre_malloc() and
+pcre_free() exported functions will be declared __declspec(dllimport), with
+unwanted results. So in this environment, uncomment the following line. */
+
+/* #define PCRE_STATIC */
+
+#include 
+#include 
+#include 
+
+#define OVECCOUNT 30    /* should be a multiple of 3 */
+
+
+int main(int argc, char **argv)
+{
+pcre *re;
+const char *error;
+char *pattern;
+char *subject;
+unsigned char *name_table;
+unsigned int option_bits;
+int erroffset;
+int find_all;
+int crlf_is_newline;
+int namecount;
+int name_entry_size;
+int ovector[OVECCOUNT];
+int subject_length;
+int rc, i;
+int utf8;
+
+
+/**************************************************************************
+* First, sort out the command line. There is only one possible option at  *
+* the moment, "-g" to request repeated matching to find all occurrences,  *
+* like Perl's /g option. We set the variable find_all to a non-zero value *
+* if the -g option is present. Apart from that, there must be exactly two *
+* arguments.                                                              *
+**************************************************************************/
+
+find_all = 0;
+for (i = 1; i < argc; i++)
+  {
+  if (strcmp(argv[i], "-g") == 0) find_all = 1;
+    else break;
+  }
+
+/* After the options, we require exactly two arguments, which are the pattern,
+and the subject string. */
+
+if (argc - i != 2)
+  {
+  printf("Two arguments required: a regex and a subject string\en");
+  return 1;
+  }
+
+pattern = argv[i];
+subject = argv[i+1];
+subject_length = (int)strlen(subject);
+
+
+/*************************************************************************
+* Now we are going to compile the regular expression pattern, and handle *
+* and errors that are detected.                                          *
+*************************************************************************/
+
+re = pcre_compile(
+  pattern,              /* the pattern */
+  0,                    /* default options */
+  &error,               /* for error message */
+  &erroffset,           /* for error offset */
+  NULL);                /* use default character tables */
+
+/* Compilation failed: print the error message and exit */
+
+if (re == NULL)
+  {
+  printf("PCRE compilation failed at offset %d: %s\en", erroffset, error);
+  return 1;
+  }
+
+
+/*************************************************************************
+* If the compilation succeeded, we call PCRE again, in order to do a     *
+* pattern match against the subject string. This does just ONE match. If *
+* further matching is needed, it will be done below.                     *
+*************************************************************************/
+
+rc = pcre_exec(
+  re,                   /* the compiled pattern */
+  NULL,                 /* no extra data - we didn't study the pattern */
+  subject,              /* the subject string */
+  subject_length,       /* the length of the subject */
+  0,                    /* start at offset 0 in the subject */
+  0,                    /* default options */
+  ovector,              /* output vector for substring information */
+  OVECCOUNT);           /* number of elements in the output vector */
+
+/* Matching failed: handle error cases */
+
+if (rc < 0)
+  {
+  switch(rc)
+    {
+    case PCRE_ERROR_NOMATCH: printf("No match\en"); break;
+    /*
+    Handle other special cases if you like
+    */
+    default: printf("Matching error %d\en", rc); break;
+    }
+  pcre_free(re);     /* Release memory used for the compiled pattern */
+  return 1;
+  }
+
+/* Match succeded */
+
+printf("\enMatch succeeded at offset %d\en", ovector[0]);
+
+
+/*************************************************************************
+* We have found the first match within the subject string. If the output *
+* vector wasn't big enough, say so. Then output any substrings that were *
+* captured.                                                              *
+*************************************************************************/
+
+/* The output vector wasn't big enough */
+
+if (rc == 0)
+  {
+  rc = OVECCOUNT/3;
+  printf("ovector only has room for %d captured substrings\en", rc - 1);
+  }
+
+/* Show substrings stored in the output vector by number. Obviously, in a real
+application you might want to do things other than print them. */
+
+for (i = 0; i < rc; i++)
+  {
+  char *substring_start = subject + ovector[2*i];
+  int substring_length = ovector[2*i+1] - ovector[2*i];
+  printf("%2d: %.*s\en", i, substring_length, substring_start);
+  }
+
+
+/**************************************************************************
+* That concludes the basic part of this demonstration program. We have    *
+* compiled a pattern, and performed a single match. The code that follows *
+* shows first how to access named substrings, and then how to code for    *
+* repeated matches on the same subject.                                   *
+**************************************************************************/
+
+/* See if there are any named substrings, and if so, show them by name. First
+we have to extract the count of named parentheses from the pattern. */
+
+(void)pcre_fullinfo(
+  re,                   /* the compiled pattern */
+  NULL,                 /* no extra data - we didn't study the pattern */
+  PCRE_INFO_NAMECOUNT,  /* number of named substrings */
+  &namecount);          /* where to put the answer */
+
+if (namecount <= 0) printf("No named substrings\en"); else
+  {
+  unsigned char *tabptr;
+  printf("Named substrings\en");
+
+  /* Before we can access the substrings, we must extract the table for
+  translating names to numbers, and the size of each entry in the table. */
+
+  (void)pcre_fullinfo(
+    re,                       /* the compiled pattern */
+    NULL,                     /* no extra data - we didn't study the pattern */
+    PCRE_INFO_NAMETABLE,      /* address of the table */
+    &name_table);             /* where to put the answer */
+
+  (void)pcre_fullinfo(
+    re,                       /* the compiled pattern */
+    NULL,                     /* no extra data - we didn't study the pattern */
+    PCRE_INFO_NAMEENTRYSIZE,  /* size of each entry in the table */
+    &name_entry_size);        /* where to put the answer */
+
+  /* Now we can scan the table and, for each entry, print the number, the name,
+  and the substring itself. */
+
+  tabptr = name_table;
+  for (i = 0; i < namecount; i++)
+    {
+    int n = (tabptr[0] << 8) | tabptr[1];
+    printf("(%d) %*s: %.*s\en", n, name_entry_size - 3, tabptr + 2,
+      ovector[2*n+1] - ovector[2*n], subject + ovector[2*n]);
+    tabptr += name_entry_size;
+    }
+  }
+
+
+/*************************************************************************
+* If the "-g" option was given on the command line, we want to continue  *
+* to search for additional matches in the subject string, in a similar   *
+* way to the /g option in Perl. This turns out to be trickier than you   *
+* might think because of the possibility of matching an empty string.    *
+* What happens is as follows:                                            *
+*                                                                        *
+* If the previous match was NOT for an empty string, we can just start   *
+* the next match at the end of the previous one.                         *
+*                                                                        *
+* If the previous match WAS for an empty string, we can't do that, as it *
+* would lead to an infinite loop. Instead, a special call of pcre_exec() *
+* is made with the PCRE_NOTEMPTY_ATSTART and PCRE_ANCHORED flags set.    *
+* The first of these tells PCRE that an empty string at the start of the *
+* subject is not a valid match; other possibilities must be tried. The   *
+* second flag restricts PCRE to one match attempt at the initial string  *
+* position. If this match succeeds, an alternative to the empty string   *
+* match has been found, and we can print it and proceed round the loop,  *
+* advancing by the length of whatever was found. If this match does not  *
+* succeed, we still stay in the loop, advancing by just one character.   *
+* In UTF-8 mode, which can be set by (*UTF8) in the pattern, this may be *
+* more than one byte.                                                    *
+*                                                                        *
+* However, there is a complication concerned with newlines. When the     *
+* newline convention is such that CRLF is a valid newline, we must       *
+* advance by two characters rather than one. The newline convention can  *
+* be set in the regex by (*CR), etc.; if not, we must find the default.  *
+*************************************************************************/
+
+if (!find_all)     /* Check for -g */
+  {
+  pcre_free(re);   /* Release the memory used for the compiled pattern */
+  return 0;        /* Finish unless -g was given */
+  }
+
+/* Before running the loop, check for UTF-8 and whether CRLF is a valid newline
+sequence. First, find the options with which the regex was compiled; extract
+the UTF-8 state, and mask off all but the newline options. */
+
+(void)pcre_fullinfo(re, NULL, PCRE_INFO_OPTIONS, &option_bits);
+utf8 = option_bits & PCRE_UTF8;
+option_bits &= PCRE_NEWLINE_CR|PCRE_NEWLINE_LF|PCRE_NEWLINE_CRLF|
+               PCRE_NEWLINE_ANY|PCRE_NEWLINE_ANYCRLF;
+
+/* If no newline options were set, find the default newline convention from the
+build configuration. */
+
+if (option_bits == 0)
+  {
+  int d;
+  (void)pcre_config(PCRE_CONFIG_NEWLINE, &d);
+  /* Note that these values are always the ASCII ones, even in
+  EBCDIC environments. CR = 13, NL = 10. */
+  option_bits = (d == 13)? PCRE_NEWLINE_CR :
+          (d == 10)? PCRE_NEWLINE_LF :
+          (d == (13<<8 | 10))? PCRE_NEWLINE_CRLF :
+          (d == -2)? PCRE_NEWLINE_ANYCRLF :
+          (d == -1)? PCRE_NEWLINE_ANY : 0;
+  }
+
+/* See if CRLF is a valid newline sequence. */
+
+crlf_is_newline =
+     option_bits == PCRE_NEWLINE_ANY ||
+     option_bits == PCRE_NEWLINE_CRLF ||
+     option_bits == PCRE_NEWLINE_ANYCRLF;
+
+/* Loop for second and subsequent matches */
+
+for (;;)
+  {
+  int options = 0;                 /* Normally no options */
+  int start_offset = ovector[1];   /* Start at end of previous match */
+
+  /* If the previous match was for an empty string, we are finished if we are
+  at the end of the subject. Otherwise, arrange to run another match at the
+  same point to see if a non-empty match can be found. */
+
+  if (ovector[0] == ovector[1])
+    {
+    if (ovector[0] == subject_length) break;
+    options = PCRE_NOTEMPTY_ATSTART | PCRE_ANCHORED;
+    }
+
+  /* Run the next matching operation */
+
+  rc = pcre_exec(
+    re,                   /* the compiled pattern */
+    NULL,                 /* no extra data - we didn't study the pattern */
+    subject,              /* the subject string */
+    subject_length,       /* the length of the subject */
+    start_offset,         /* starting offset in the subject */
+    options,              /* options */
+    ovector,              /* output vector for substring information */
+    OVECCOUNT);           /* number of elements in the output vector */
+
+  /* This time, a result of NOMATCH isn't an error. If the value in "options"
+  is zero, it just means we have found all possible matches, so the loop ends.
+  Otherwise, it means we have failed to find a non-empty-string match at a
+  point where there was a previous empty-string match. In this case, we do what
+  Perl does: advance the matching position by one character, and continue. We
+  do this by setting the "end of previous match" offset, because that is picked
+  up at the top of the loop as the point at which to start again.
+
+  There are two complications: (a) When CRLF is a valid newline sequence, and
+  the current position is just before it, advance by an extra byte. (b)
+  Otherwise we must ensure that we skip an entire UTF-8 character if we are in
+  UTF-8 mode. */
+
+  if (rc == PCRE_ERROR_NOMATCH)
+    {
+    if (options == 0) break;                    /* All matches found */
+    ovector[1] = start_offset + 1;              /* Advance one byte */
+    if (crlf_is_newline &&                      /* If CRLF is newline & */
+        start_offset < subject_length - 1 &&    /* we are at CRLF, */
+        subject[start_offset] == '\er' &&
+        subject[start_offset + 1] == '\en')
+      ovector[1] += 1;                          /* Advance by one more. */
+    else if (utf8)                              /* Otherwise, ensure we */
+      {                                         /* advance a whole UTF-8 */
+      while (ovector[1] < subject_length)       /* character. */
+        {
+        if ((subject[ovector[1]] & 0xc0) != 0x80) break;
+        ovector[1] += 1;
+        }
+      }
+    continue;    /* Go round the loop again */
+    }
+
+  /* Other matching errors are not recoverable. */
+
+  if (rc < 0)
+    {
+    printf("Matching error %d\en", rc);
+    pcre_free(re);    /* Release memory used for the compiled pattern */
+    return 1;
+    }
+
+  /* Match succeded */
+
+  printf("\enMatch succeeded again at offset %d\en", ovector[0]);
+
+  /* The match succeeded, but the output vector wasn't big enough. */
+
+  if (rc == 0)
+    {
+    rc = OVECCOUNT/3;
+    printf("ovector only has room for %d captured substrings\en", rc - 1);
+    }
+
+  /* As before, show substrings stored in the output vector by number, and then
+  also any named substrings. */
+
+  for (i = 0; i < rc; i++)
+    {
+    char *substring_start = subject + ovector[2*i];
+    int substring_length = ovector[2*i+1] - ovector[2*i];
+    printf("%2d: %.*s\en", i, substring_length, substring_start);
+    }
+
+  if (namecount <= 0) printf("No named substrings\en"); else
+    {
+    unsigned char *tabptr = name_table;
+    printf("Named substrings\en");
+    for (i = 0; i < namecount; i++)
+      {
+      int n = (tabptr[0] << 8) | tabptr[1];
+      printf("(%d) %*s: %.*s\en", n, name_entry_size - 3, tabptr + 2,
+        ovector[2*n+1] - ovector[2*n], subject + ovector[2*n]);
+      tabptr += name_entry_size;
+      }
+    }
+  }      /* End of loop to find second and subsequent matches */
+
+printf("\en");
+pcre_free(re);       /* Release memory used for the compiled pattern */
+return 0;
+}
+
+/* End of pcredemo.c */
+.EE
diff --git a/pcre/doc/pcregrep.1 b/pcre/doc/pcregrep.1
new file mode 100644
index 00000000000..7fa5b65e030
--- /dev/null
+++ b/pcre/doc/pcregrep.1
@@ -0,0 +1,679 @@
+.TH PCREGREP 1 "13 September 2012" "PCRE 8.32"
+.SH NAME
+pcregrep - a grep with Perl-compatible regular expressions.
+.SH SYNOPSIS
+.B pcregrep [options] [long options] [pattern] [path1 path2 ...]
+.
+.SH DESCRIPTION
+.rs
+.sp
+\fBpcregrep\fP searches files for character patterns, in the same way as other
+grep commands do, but it uses the PCRE regular expression library to support
+patterns that are compatible with the regular expressions of Perl 5. See
+.\" HREF
+\fBpcrepattern\fP(3)
+.\"
+for a full description of syntax and semantics of the regular expressions
+that PCRE supports.
+.P
+Patterns, whether supplied on the command line or in a separate file, are given
+without delimiters. For example:
+.sp
+  pcregrep Thursday /etc/motd
+.sp
+If you attempt to use delimiters (for example, by surrounding a pattern with
+slashes, as is common in Perl scripts), they are interpreted as part of the
+pattern. Quotes can of course be used to delimit patterns on the command line
+because they are interpreted by the shell, and indeed quotes are required if a
+pattern contains white space or shell metacharacters.
+.P
+The first argument that follows any option settings is treated as the single
+pattern to be matched when neither \fB-e\fP nor \fB-f\fP is present.
+Conversely, when one or both of these options are used to specify patterns, all
+arguments are treated as path names. At least one of \fB-e\fP, \fB-f\fP, or an
+argument pattern must be provided.
+.P
+If no files are specified, \fBpcregrep\fP reads the standard input. The
+standard input can also be referenced by a name consisting of a single hyphen.
+For example:
+.sp
+  pcregrep some-pattern /file1 - /file3
+.sp
+By default, each line that matches a pattern is copied to the standard
+output, and if there is more than one file, the file name is output at the
+start of each line, followed by a colon. However, there are options that can
+change how \fBpcregrep\fP behaves. In particular, the \fB-M\fP option makes it
+possible to search for patterns that span line boundaries. What defines a line
+boundary is controlled by the \fB-N\fP (\fB--newline\fP) option.
+.P
+The amount of memory used for buffering files that are being scanned is
+controlled by a parameter that can be set by the \fB--buffer-size\fP option.
+The default value for this parameter is specified when \fBpcregrep\fP is built,
+with the default default being 20K. A block of memory three times this size is
+used (to allow for buffering "before" and "after" lines). An error occurs if a
+line overflows the buffer.
+.P
+Patterns can be no longer than 8K or BUFSIZ bytes, whichever is the greater.
+BUFSIZ is defined in \fB\fP. When there is more than one pattern
+(specified by the use of \fB-e\fP and/or \fB-f\fP), each pattern is applied to
+each line in the order in which they are defined, except that all the \fB-e\fP
+patterns are tried before the \fB-f\fP patterns.
+.P
+By default, as soon as one pattern matches a line, no further patterns are
+considered. However, if \fB--colour\fP (or \fB--color\fP) is used to colour the
+matching substrings, or if \fB--only-matching\fP, \fB--file-offsets\fP, or
+\fB--line-offsets\fP is used to output only the part of the line that matched
+(either shown literally, or as an offset), scanning resumes immediately
+following the match, so that further matches on the same line can be found. If
+there are multiple patterns, they are all tried on the remainder of the line,
+but patterns that follow the one that matched are not tried on the earlier part
+of the line.
+.P
+This behaviour means that the order in which multiple patterns are specified
+can affect the output when one of the above options is used. This is no longer
+the same behaviour as GNU grep, which now manages to display earlier matches
+for later patterns (as long as there is no overlap).
+.P
+Patterns that can match an empty string are accepted, but empty string
+matches are never recognized. An example is the pattern "(super)?(man)?", in
+which all components are optional. This pattern finds all occurrences of both
+"super" and "man"; the output differs from matching with "super|man" when only
+the matching substrings are being shown.
+.P
+If the \fBLC_ALL\fP or \fBLC_CTYPE\fP environment variable is set,
+\fBpcregrep\fP uses the value to set a locale when calling the PCRE library.
+The \fB--locale\fP option can be used to override this.
+.
+.
+.SH "SUPPORT FOR COMPRESSED FILES"
+.rs
+.sp
+It is possible to compile \fBpcregrep\fP so that it uses \fBlibz\fP or
+\fBlibbz2\fP to read files whose names end in \fB.gz\fP or \fB.bz2\fP,
+respectively. You can find out whether your binary has support for one or both
+of these file types by running it with the \fB--help\fP option. If the
+appropriate support is not present, files are treated as plain text. The
+standard input is always so treated.
+.
+.
+.SH "BINARY FILES"
+.rs
+.sp
+By default, a file that contains a binary zero byte within the first 1024 bytes
+is identified as a binary file, and is processed specially. (GNU grep also
+identifies binary files in this manner.) See the \fB--binary-files\fP option
+for a means of changing the way binary files are handled.
+.
+.
+.SH OPTIONS
+.rs
+.sp
+The order in which some of the options appear can affect the output. For
+example, both the \fB-h\fP and \fB-l\fP options affect the printing of file
+names. Whichever comes later in the command line will be the one that takes
+effect. Similarly, except where noted below, if an option is given twice, the
+later setting is used. Numerical values for options may be followed by K or M,
+to signify multiplication by 1024 or 1024*1024 respectively.
+.TP 10
+\fB--\fP
+This terminates the list of options. It is useful if the next item on the
+command line starts with a hyphen but is not an option. This allows for the
+processing of patterns and filenames that start with hyphens.
+.TP
+\fB-A\fP \fInumber\fP, \fB--after-context=\fP\fInumber\fP
+Output \fInumber\fP lines of context after each matching line. If filenames
+and/or line numbers are being output, a hyphen separator is used instead of a
+colon for the context lines. A line containing "--" is output between each
+group of lines, unless they are in fact contiguous in the input file. The value
+of \fInumber\fP is expected to be relatively small. However, \fBpcregrep\fP
+guarantees to have up to 8K of following text available for context output.
+.TP
+\fB-a\fP, \fB--text\fP
+Treat binary files as text. This is equivalent to
+\fB--binary-files\fP=\fItext\fP.
+.TP
+\fB-B\fP \fInumber\fP, \fB--before-context=\fP\fInumber\fP
+Output \fInumber\fP lines of context before each matching line. If filenames
+and/or line numbers are being output, a hyphen separator is used instead of a
+colon for the context lines. A line containing "--" is output between each
+group of lines, unless they are in fact contiguous in the input file. The value
+of \fInumber\fP is expected to be relatively small. However, \fBpcregrep\fP
+guarantees to have up to 8K of preceding text available for context output.
+.TP
+\fB--binary-files=\fP\fIword\fP
+Specify how binary files are to be processed. If the word is "binary" (the
+default), pattern matching is performed on binary files, but the only output is
+"Binary file  matches" when a match succeeds. If the word is "text",
+which is equivalent to the \fB-a\fP or \fB--text\fP option, binary files are
+processed in the same way as any other file. In this case, when a match
+succeeds, the output may be binary garbage, which can have nasty effects if
+sent to a terminal. If the word is "without-match", which is equivalent to the
+\fB-I\fP option, binary files are not processed at all; they are assumed not to
+be of interest.
+.TP
+\fB--buffer-size=\fP\fInumber\fP
+Set the parameter that controls how much memory is used for buffering files
+that are being scanned.
+.TP
+\fB-C\fP \fInumber\fP, \fB--context=\fP\fInumber\fP
+Output \fInumber\fP lines of context both before and after each matching line.
+This is equivalent to setting both \fB-A\fP and \fB-B\fP to the same value.
+.TP
+\fB-c\fP, \fB--count\fP
+Do not output individual lines from the files that are being scanned; instead
+output the number of lines that would otherwise have been shown. If no lines
+are selected, the number zero is output. If several files are are being
+scanned, a count is output for each of them. However, if the
+\fB--files-with-matches\fP option is also used, only those files whose counts
+are greater than zero are listed. When \fB-c\fP is used, the \fB-A\fP,
+\fB-B\fP, and \fB-C\fP options are ignored.
+.TP
+\fB--colour\fP, \fB--color\fP
+If this option is given without any data, it is equivalent to "--colour=auto".
+If data is required, it must be given in the same shell item, separated by an
+equals sign.
+.TP
+\fB--colour=\fP\fIvalue\fP, \fB--color=\fP\fIvalue\fP
+This option specifies under what circumstances the parts of a line that matched
+a pattern should be coloured in the output. By default, the output is not
+coloured. The value (which is optional, see above) may be "never", "always", or
+"auto". In the latter case, colouring happens only if the standard output is
+connected to a terminal. More resources are used when colouring is enabled,
+because \fBpcregrep\fP has to search for all possible matches in a line, not
+just one, in order to colour them all.
+.sp
+The colour that is used can be specified by setting the environment variable
+PCREGREP_COLOUR or PCREGREP_COLOR. The value of this variable should be a
+string of two numbers, separated by a semicolon. They are copied directly into
+the control string for setting colour on a terminal, so it is your
+responsibility to ensure that they make sense. If neither of the environment
+variables is set, the default is "1;31", which gives red.
+.TP
+\fB-D\fP \fIaction\fP, \fB--devices=\fP\fIaction\fP
+If an input path is not a regular file or a directory, "action" specifies how
+it is to be processed. Valid values are "read" (the default) or "skip"
+(silently skip the path).
+.TP
+\fB-d\fP \fIaction\fP, \fB--directories=\fP\fIaction\fP
+If an input path is a directory, "action" specifies how it is to be processed.
+Valid values are "read" (the default in non-Windows environments, for
+compatibility with GNU grep), "recurse" (equivalent to the \fB-r\fP option), or
+"skip" (silently skip the path, the default in Windows environments). In the
+"read" case, directories are read as if they were ordinary files. In some
+operating systems the effect of reading a directory like this is an immediate
+end-of-file; in others it may provoke an error.
+.TP
+\fB-e\fP \fIpattern\fP, \fB--regex=\fP\fIpattern\fP, \fB--regexp=\fP\fIpattern\fP
+Specify a pattern to be matched. This option can be used multiple times in
+order to specify several patterns. It can also be used as a way of specifying a
+single pattern that starts with a hyphen. When \fB-e\fP is used, no argument
+pattern is taken from the command line; all arguments are treated as file
+names. There is no limit to the number of patterns. They are applied to each
+line in the order in which they are defined until one matches.
+.sp
+If \fB-f\fP is used with \fB-e\fP, the command line patterns are matched first,
+followed by the patterns from the file(s), independent of the order in which
+these options are specified. Note that multiple use of \fB-e\fP is not the same
+as a single pattern with alternatives. For example, X|Y finds the first
+character in a line that is X or Y, whereas if the two patterns are given
+separately, with X first, \fBpcregrep\fP finds X if it is present, even if it
+follows Y in the line. It finds Y only if there is no X in the line. This
+matters only if you are using \fB-o\fP or \fB--colo(u)r\fP to show the part(s)
+of the line that matched.
+.TP
+\fB--exclude\fP=\fIpattern\fP
+Files (but not directories) whose names match the pattern are skipped without
+being processed. This applies to all files, whether listed on the command line,
+obtained from \fB--file-list\fP, or by scanning a directory. The pattern is a
+PCRE regular expression, and is matched against the final component of the file
+name, not the entire path. The \fB-F\fP, \fB-w\fP, and \fB-x\fP options do not
+apply to this pattern. The option may be given any number of times in order to
+specify multiple patterns. If a file name matches both an \fB--include\fP
+and an \fB--exclude\fP pattern, it is excluded. There is no short form for this
+option.
+.TP
+\fB--exclude-from=\fP\fIfilename\fP
+Treat each non-empty line of the file as the data for an \fB--exclude\fP
+option. What constitutes a newline when reading the file is the operating
+system's default. The \fB--newline\fP option has no effect on this option. This
+option may be given more than once in order to specify a number of files to
+read.
+.TP
+\fB--exclude-dir\fP=\fIpattern\fP
+Directories whose names match the pattern are skipped without being processed,
+whatever the setting of the \fB--recursive\fP option. This applies to all
+directories, whether listed on the command line, obtained from
+\fB--file-list\fP, or by scanning a parent directory. The pattern is a PCRE
+regular expression, and is matched against the final component of the directory
+name, not the entire path. The \fB-F\fP, \fB-w\fP, and \fB-x\fP options do not
+apply to this pattern. The option may be given any number of times in order to
+specify more than one pattern. If a directory matches both \fB--include-dir\fP
+and \fB--exclude-dir\fP, it is excluded. There is no short form for this
+option.
+.TP
+\fB-F\fP, \fB--fixed-strings\fP
+Interpret each data-matching pattern as a list of fixed strings, separated by
+newlines, instead of as a regular expression. What constitutes a newline for
+this purpose is controlled by the \fB--newline\fP option. The \fB-w\fP (match
+as a word) and \fB-x\fP (match whole line) options can be used with \fB-F\fP.
+They apply to each of the fixed strings. A line is selected if any of the fixed
+strings are found in it (subject to \fB-w\fP or \fB-x\fP, if present). This
+option applies only to the patterns that are matched against the contents of
+files; it does not apply to patterns specified by any of the \fB--include\fP or
+\fB--exclude\fP options.
+.TP
+\fB-f\fP \fIfilename\fP, \fB--file=\fP\fIfilename\fP
+Read patterns from the file, one per line, and match them against
+each line of input. What constitutes a newline when reading the file is the
+operating system's default. The \fB--newline\fP option has no effect on this
+option. Trailing white space is removed from each line, and blank lines are
+ignored. An empty file contains no patterns and therefore matches nothing. See
+also the comments about multiple patterns versus a single pattern with
+alternatives in the description of \fB-e\fP above.
+.sp
+If this option is given more than once, all the specified files are
+read. A data line is output if any of the patterns match it. A filename can
+be given as "-" to refer to the standard input. When \fB-f\fP is used, patterns
+specified on the command line using \fB-e\fP may also be present; they are
+tested before the file's patterns. However, no other pattern is taken from the
+command line; all arguments are treated as the names of paths to be searched.
+.TP
+\fB--file-list\fP=\fIfilename\fP
+Read a list of files and/or directories that are to be scanned from the given
+file, one per line. Trailing white space is removed from each line, and blank
+lines are ignored. These paths are processed before any that are listed on the
+command line. The filename can be given as "-" to refer to the standard input.
+If \fB--file\fP and \fB--file-list\fP are both specified as "-", patterns are
+read first. This is useful only when the standard input is a terminal, from
+which further lines (the list of files) can be read after an end-of-file
+indication. If this option is given more than once, all the specified files are
+read.
+.TP
+\fB--file-offsets\fP
+Instead of showing lines or parts of lines that match, show each match as an
+offset from the start of the file and a length, separated by a comma. In this
+mode, no context is shown. That is, the \fB-A\fP, \fB-B\fP, and \fB-C\fP
+options are ignored. If there is more than one match in a line, each of them is
+shown separately. This option is mutually exclusive with \fB--line-offsets\fP
+and \fB--only-matching\fP.
+.TP
+\fB-H\fP, \fB--with-filename\fP
+Force the inclusion of the filename at the start of output lines when searching
+a single file. By default, the filename is not shown in this case. For matching
+lines, the filename is followed by a colon; for context lines, a hyphen
+separator is used. If a line number is also being output, it follows the file
+name.
+.TP
+\fB-h\fP, \fB--no-filename\fP
+Suppress the output filenames when searching multiple files. By default,
+filenames are shown when multiple files are searched. For matching lines, the
+filename is followed by a colon; for context lines, a hyphen separator is used.
+If a line number is also being output, it follows the file name.
+.TP
+\fB--help\fP
+Output a help message, giving brief details of the command options and file
+type support, and then exit. Anything else on the command line is
+ignored.
+.TP
+\fB-I\fP
+Treat binary files as never matching. This is equivalent to
+\fB--binary-files\fP=\fIwithout-match\fP.
+.TP
+\fB-i\fP, \fB--ignore-case\fP
+Ignore upper/lower case distinctions during comparisons.
+.TP
+\fB--include\fP=\fIpattern\fP
+If any \fB--include\fP patterns are specified, the only files that are
+processed are those that match one of the patterns (and do not match an
+\fB--exclude\fP pattern). This option does not affect directories, but it
+applies to all files, whether listed on the command line, obtained from
+\fB--file-list\fP, or by scanning a directory. The pattern is a PCRE regular
+expression, and is matched against the final component of the file name, not
+the entire path. The \fB-F\fP, \fB-w\fP, and \fB-x\fP options do not apply to
+this pattern. The option may be given any number of times. If a file name
+matches both an \fB--include\fP and an \fB--exclude\fP pattern, it is excluded.
+There is no short form for this option.
+.TP
+\fB--include-from=\fP\fIfilename\fP
+Treat each non-empty line of the file as the data for an \fB--include\fP
+option. What constitutes a newline for this purpose is the operating system's
+default. The \fB--newline\fP option has no effect on this option. This option
+may be given any number of times; all the files are read.
+.TP
+\fB--include-dir\fP=\fIpattern\fP
+If any \fB--include-dir\fP patterns are specified, the only directories that
+are processed are those that match one of the patterns (and do not match an
+\fB--exclude-dir\fP pattern). This applies to all directories, whether listed
+on the command line, obtained from \fB--file-list\fP, or by scanning a parent
+directory. The pattern is a PCRE regular expression, and is matched against the
+final component of the directory name, not the entire path. The \fB-F\fP,
+\fB-w\fP, and \fB-x\fP options do not apply to this pattern. The option may be
+given any number of times. If a directory matches both \fB--include-dir\fP and
+\fB--exclude-dir\fP, it is excluded. There is no short form for this option.
+.TP
+\fB-L\fP, \fB--files-without-match\fP
+Instead of outputting lines from the files, just output the names of the files
+that do not contain any lines that would have been output. Each file name is
+output once, on a separate line.
+.TP
+\fB-l\fP, \fB--files-with-matches\fP
+Instead of outputting lines from the files, just output the names of the files
+containing lines that would have been output. Each file name is output
+once, on a separate line. Searching normally stops as soon as a matching line
+is found in a file. However, if the \fB-c\fP (count) option is also used,
+matching continues in order to obtain the correct count, and those files that
+have at least one match are listed along with their counts. Using this option
+with \fB-c\fP is a way of suppressing the listing of files with no matches.
+.TP
+\fB--label\fP=\fIname\fP
+This option supplies a name to be used for the standard input when file names
+are being output. If not supplied, "(standard input)" is used. There is no
+short form for this option.
+.TP
+\fB--line-buffered\fP
+When this option is given, input is read and processed line by line, and the
+output is flushed after each write. By default, input is read in large chunks,
+unless \fBpcregrep\fP can determine that it is reading from a terminal (which
+is currently possible only in Unix-like environments). Output to terminal is
+normally automatically flushed by the operating system. This option can be
+useful when the input or output is attached to a pipe and you do not want
+\fBpcregrep\fP to buffer up large amounts of data. However, its use will affect
+performance, and the \fB-M\fP (multiline) option ceases to work.
+.TP
+\fB--line-offsets\fP
+Instead of showing lines or parts of lines that match, show each match as a
+line number, the offset from the start of the line, and a length. The line
+number is terminated by a colon (as usual; see the \fB-n\fP option), and the
+offset and length are separated by a comma. In this mode, no context is shown.
+That is, the \fB-A\fP, \fB-B\fP, and \fB-C\fP options are ignored. If there is
+more than one match in a line, each of them is shown separately. This option is
+mutually exclusive with \fB--file-offsets\fP and \fB--only-matching\fP.
+.TP
+\fB--locale\fP=\fIlocale-name\fP
+This option specifies a locale to be used for pattern matching. It overrides
+the value in the \fBLC_ALL\fP or \fBLC_CTYPE\fP environment variables. If no
+locale is specified, the PCRE library's default (usually the "C" locale) is
+used. There is no short form for this option.
+.TP
+\fB--match-limit\fP=\fInumber\fP
+Processing some regular expression patterns can require a very large amount of
+memory, leading in some cases to a program crash if not enough is available.
+Other patterns may take a very long time to search for all possible matching
+strings. The \fBpcre_exec()\fP function that is called by \fBpcregrep\fP to do
+the matching has two parameters that can limit the resources that it uses.
+.sp
+The \fB--match-limit\fP option provides a means of limiting resource usage
+when processing patterns that are not going to match, but which have a very
+large number of possibilities in their search trees. The classic example is a
+pattern that uses nested unlimited repeats. Internally, PCRE uses a function
+called \fBmatch()\fP which it calls repeatedly (sometimes recursively). The
+limit set by \fB--match-limit\fP is imposed on the number of times this
+function is called during a match, which has the effect of limiting the amount
+of backtracking that can take place.
+.sp
+The \fB--recursion-limit\fP option is similar to \fB--match-limit\fP, but
+instead of limiting the total number of times that \fBmatch()\fP is called, it
+limits the depth of recursive calls, which in turn limits the amount of memory
+that can be used. The recursion depth is a smaller number than the total number
+of calls, because not all calls to \fBmatch()\fP are recursive. This limit is
+of use only if it is set smaller than \fB--match-limit\fP.
+.sp
+There are no short forms for these options. The default settings are specified
+when the PCRE library is compiled, with the default default being 10 million.
+.TP
+\fB-M\fP, \fB--multiline\fP
+Allow patterns to match more than one line. When this option is given, patterns
+may usefully contain literal newline characters and internal occurrences of ^
+and $ characters. The output for a successful match may consist of more than
+one line, the last of which is the one in which the match ended. If the matched
+string ends with a newline sequence the output ends at the end of that line.
+.sp
+When this option is set, the PCRE library is called in "multiline" mode.
+There is a limit to the number of lines that can be matched, imposed by the way
+that \fBpcregrep\fP buffers the input file as it scans it. However,
+\fBpcregrep\fP ensures that at least 8K characters or the rest of the document
+(whichever is the shorter) are available for forward matching, and similarly
+the previous 8K characters (or all the previous characters, if fewer than 8K)
+are guaranteed to be available for lookbehind assertions. This option does not
+work when input is read line by line (see \fP--line-buffered\fP.)
+.TP
+\fB-N\fP \fInewline-type\fP, \fB--newline\fP=\fInewline-type\fP
+The PCRE library supports five different conventions for indicating
+the ends of lines. They are the single-character sequences CR (carriage return)
+and LF (linefeed), the two-character sequence CRLF, an "anycrlf" convention,
+which recognizes any of the preceding three types, and an "any" convention, in
+which any Unicode line ending sequence is assumed to end a line. The Unicode
+sequences are the three just mentioned, plus VT (vertical tab, U+000B), FF
+(form feed, U+000C), NEL (next line, U+0085), LS (line separator, U+2028), and
+PS (paragraph separator, U+2029).
+.sp
+When the PCRE library is built, a default line-ending sequence is specified.
+This is normally the standard sequence for the operating system. Unless
+otherwise specified by this option, \fBpcregrep\fP uses the library's default.
+The possible values for this option are CR, LF, CRLF, ANYCRLF, or ANY. This
+makes it possible to use \fBpcregrep\fP to scan files that have come from other
+environments without having to modify their line endings. If the data that is
+being scanned does not agree with the convention set by this option,
+\fBpcregrep\fP may behave in strange ways. Note that this option does not
+apply to files specified by the \fB-f\fP, \fB--exclude-from\fP, or
+\fB--include-from\fP options, which are expected to use the operating system's
+standard newline sequence.
+.TP
+\fB-n\fP, \fB--line-number\fP
+Precede each output line by its line number in the file, followed by a colon
+for matching lines or a hyphen for context lines. If the filename is also being
+output, it precedes the line number. This option is forced if
+\fB--line-offsets\fP is used.
+.TP
+\fB--no-jit\fP
+If the PCRE library is built with support for just-in-time compiling (which
+speeds up matching), \fBpcregrep\fP automatically makes use of this, unless it
+was explicitly disabled at build time. This option can be used to disable the
+use of JIT at run time. It is provided for testing and working round problems.
+It should never be needed in normal use.
+.TP
+\fB-o\fP, \fB--only-matching\fP
+Show only the part of the line that matched a pattern instead of the whole
+line. In this mode, no context is shown. That is, the \fB-A\fP, \fB-B\fP, and
+\fB-C\fP options are ignored. If there is more than one match in a line, each
+of them is shown separately. If \fB-o\fP is combined with \fB-v\fP (invert the
+sense of the match to find non-matching lines), no output is generated, but the
+return code is set appropriately. If the matched portion of the line is empty,
+nothing is output unless the file name or line number are being printed, in
+which case they are shown on an otherwise empty line. This option is mutually
+exclusive with \fB--file-offsets\fP and \fB--line-offsets\fP.
+.TP
+\fB-o\fP\fInumber\fP, \fB--only-matching\fP=\fInumber\fP
+Show only the part of the line that matched the capturing parentheses of the
+given number. Up to 32 capturing parentheses are supported, and -o0 is
+equivalent to \fB-o\fP without a number. Because these options can be given
+without an argument (see above), if an argument is present, it must be given in
+the same shell item, for example, -o3 or --only-matching=2. The comments given
+for the non-argument case above also apply to this case. If the specified
+capturing parentheses do not exist in the pattern, or were not set in the
+match, nothing is output unless the file name or line number are being printed.
+.sp
+If this option is given multiple times, multiple substrings are output, in the
+order the options are given. For example, -o3 -o1 -o3 causes the substrings
+matched by capturing parentheses 3 and 1 and then 3 again to be output. By
+default, there is no separator (but see the next option).
+.TP
+\fB--om-separator\fP=\fItext\fP
+Specify a separating string for multiple occurrences of \fB-o\fP. The default
+is an empty string. Separating strings are never coloured.
+.TP
+\fB-q\fP, \fB--quiet\fP
+Work quietly, that is, display nothing except error messages. The exit
+status indicates whether or not any matches were found.
+.TP
+\fB-r\fP, \fB--recursive\fP
+If any given path is a directory, recursively scan the files it contains,
+taking note of any \fB--include\fP and \fB--exclude\fP settings. By default, a
+directory is read as a normal file; in some operating systems this gives an
+immediate end-of-file. This option is a shorthand for setting the \fB-d\fP
+option to "recurse".
+.TP
+\fB--recursion-limit\fP=\fInumber\fP
+See \fB--match-limit\fP above.
+.TP
+\fB-s\fP, \fB--no-messages\fP
+Suppress error messages about non-existent or unreadable files. Such files are
+quietly skipped. However, the return code is still 2, even if matches were
+found in other files.
+.TP
+\fB-u\fP, \fB--utf-8\fP
+Operate in UTF-8 mode. This option is available only if PCRE has been compiled
+with UTF-8 support. All patterns (including those for any \fB--exclude\fP and
+\fB--include\fP options) and all subject lines that are scanned must be valid
+strings of UTF-8 characters.
+.TP
+\fB-V\fP, \fB--version\fP
+Write the version numbers of \fBpcregrep\fP and the PCRE library to the
+standard output and then exit. Anything else on the command line is
+ignored.
+.TP
+\fB-v\fP, \fB--invert-match\fP
+Invert the sense of the match, so that lines which do \fInot\fP match any of
+the patterns are the ones that are found.
+.TP
+\fB-w\fP, \fB--word-regex\fP, \fB--word-regexp\fP
+Force the patterns to match only whole words. This is equivalent to having \eb
+at the start and end of the pattern. This option applies only to the patterns
+that are matched against the contents of files; it does not apply to patterns
+specified by any of the \fB--include\fP or \fB--exclude\fP options.
+.TP
+\fB-x\fP, \fB--line-regex\fP, \fB--line-regexp\fP
+Force the patterns to be anchored (each must start matching at the beginning of
+a line) and in addition, require them to match entire lines. This is equivalent
+to having ^ and $ characters at the start and end of each alternative branch in
+every pattern. This option applies only to the patterns that are matched
+against the contents of files; it does not apply to patterns specified by any
+of the \fB--include\fP or \fB--exclude\fP options.
+.
+.
+.SH "ENVIRONMENT VARIABLES"
+.rs
+.sp
+The environment variables \fBLC_ALL\fP and \fBLC_CTYPE\fP are examined, in that
+order, for a locale. The first one that is set is used. This can be overridden
+by the \fB--locale\fP option. If no locale is set, the PCRE library's default
+(usually the "C" locale) is used.
+.
+.
+.SH "NEWLINES"
+.rs
+.sp
+The \fB-N\fP (\fB--newline\fP) option allows \fBpcregrep\fP to scan files with
+different newline conventions from the default. Any parts of the input files
+that are written to the standard output are copied identically, with whatever
+newline sequences they have in the input. However, the setting of this option
+does not affect the interpretation of files specified by the \fB-f\fP,
+\fB--exclude-from\fP, or \fB--include-from\fP options, which are assumed to use
+the operating system's standard newline sequence, nor does it affect the way in
+which \fBpcregrep\fP writes informational messages to the standard error and
+output streams. For these it uses the string "\en" to indicate newlines,
+relying on the C I/O library to convert this to an appropriate sequence.
+.
+.
+.SH "OPTIONS COMPATIBILITY"
+.rs
+.sp
+Many of the short and long forms of \fBpcregrep\fP's options are the same
+as in the GNU \fBgrep\fP program. Any long option of the form
+\fB--xxx-regexp\fP (GNU terminology) is also available as \fB--xxx-regex\fP
+(PCRE terminology). However, the \fB--file-list\fP, \fB--file-offsets\fP,
+\fB--include-dir\fP, \fB--line-offsets\fP, \fB--locale\fP, \fB--match-limit\fP,
+\fB-M\fP, \fB--multiline\fP, \fB-N\fP, \fB--newline\fP, \fB--om-separator\fP,
+\fB--recursion-limit\fP, \fB-u\fP, and \fB--utf-8\fP options are specific to
+\fBpcregrep\fP, as is the use of the \fB--only-matching\fP option with a
+capturing parentheses number.
+.P
+Although most of the common options work the same way, a few are different in
+\fBpcregrep\fP. For example, the \fB--include\fP option's argument is a glob
+for GNU \fBgrep\fP, but a regular expression for \fBpcregrep\fP. If both the
+\fB-c\fP and \fB-l\fP options are given, GNU grep lists only file names,
+without counts, but \fBpcregrep\fP gives the counts.
+.
+.
+.SH "OPTIONS WITH DATA"
+.rs
+.sp
+There are four different ways in which an option with data can be specified.
+If a short form option is used, the data may follow immediately, or (with one
+exception) in the next command line item. For example:
+.sp
+  -f/some/file
+  -f /some/file
+.sp
+The exception is the \fB-o\fP option, which may appear with or without data.
+Because of this, if data is present, it must follow immediately in the same
+item, for example -o3.
+.P
+If a long form option is used, the data may appear in the same command line
+item, separated by an equals character, or (with two exceptions) it may appear
+in the next command line item. For example:
+.sp
+  --file=/some/file
+  --file /some/file
+.sp
+Note, however, that if you want to supply a file name beginning with ~ as data
+in a shell command, and have the shell expand ~ to a home directory, you must
+separate the file name from the option, because the shell does not treat ~
+specially unless it is at the start of an item.
+.P
+The exceptions to the above are the \fB--colour\fP (or \fB--color\fP) and
+\fB--only-matching\fP options, for which the data is optional. If one of these
+options does have data, it must be given in the first form, using an equals
+character. Otherwise \fBpcregrep\fP will assume that it has no data.
+.
+.
+.SH "MATCHING ERRORS"
+.rs
+.sp
+It is possible to supply a regular expression that takes a very long time to
+fail to match certain lines. Such patterns normally involve nested indefinite
+repeats, for example: (a+)*\ed when matched against a line of a's with no final
+digit. The PCRE matching function has a resource limit that causes it to abort
+in these circumstances. If this happens, \fBpcregrep\fP outputs an error
+message and the line that caused the problem to the standard error stream. If
+there are more than 20 such errors, \fBpcregrep\fP gives up.
+.P
+The \fB--match-limit\fP option of \fBpcregrep\fP can be used to set the overall
+resource limit; there is a second option called \fB--recursion-limit\fP that
+sets a limit on the amount of memory (usually stack) that is used (see the
+discussion of these options above).
+.
+.
+.SH DIAGNOSTICS
+.rs
+.sp
+Exit status is 0 if any matches were found, 1 if no matches were found, and 2
+for syntax errors, overlong lines, non-existent or inaccessible files (even if
+matches were found in other files) or too many matching errors. Using the
+\fB-s\fP option to suppress error messages about inaccessible files does not
+affect the return code.
+.
+.
+.SH "SEE ALSO"
+.rs
+.sp
+\fBpcrepattern\fP(3), \fBpcresyntax\fP(3), \fBpcretest\fP(1).
+.
+.
+.SH AUTHOR
+.rs
+.sp
+.nf
+Philip Hazel
+University Computing Service
+Cambridge CB2 3QH, England.
+.fi
+.
+.
+.SH REVISION
+.rs
+.sp
+.nf
+Last updated: 13 September 2012
+Copyright (c) 1997-2012 University of Cambridge.
+.fi
diff --git a/pcre/doc/pcregrep.txt b/pcre/doc/pcregrep.txt
new file mode 100644
index 00000000000..b31eb77c304
--- /dev/null
+++ b/pcre/doc/pcregrep.txt
@@ -0,0 +1,740 @@
+PCREGREP(1)                 General Commands Manual                PCREGREP(1)
+
+
+
+NAME
+       pcregrep - a grep with Perl-compatible regular expressions.
+
+SYNOPSIS
+       pcregrep [options] [long options] [pattern] [path1 path2 ...]
+
+
+DESCRIPTION
+
+       pcregrep  searches  files  for  character  patterns, in the same way as
+       other grep commands do, but it uses the PCRE regular expression library
+       to support patterns that are compatible with the regular expressions of
+       Perl 5. See pcrepattern(3) for a full description of syntax and  seman-
+       tics of the regular expressions that PCRE supports.
+
+       Patterns,  whether  supplied on the command line or in a separate file,
+       are given without delimiters. For example:
+
+         pcregrep Thursday /etc/motd
+
+       If you attempt to use delimiters (for example, by surrounding a pattern
+       with  slashes,  as  is common in Perl scripts), they are interpreted as
+       part of the pattern. Quotes can of course be used to  delimit  patterns
+       on  the  command  line  because  they are interpreted by the shell, and
+       indeed quotes are required if a pattern contains white space  or  shell
+       metacharacters.
+
+       The  first  argument that follows any option settings is treated as the
+       single pattern to be matched when neither -e nor -f is  present.   Con-
+       versely,  when  one  or  both of these options are used to specify pat-
+       terns, all arguments are treated as path names. At least one of -e, -f,
+       or an argument pattern must be provided.
+
+       If no files are specified, pcregrep reads the standard input. The stan-
+       dard input can also be referenced by a  name  consisting  of  a  single
+       hyphen.  For example:
+
+         pcregrep some-pattern /file1 - /file3
+
+       By  default, each line that matches a pattern is copied to the standard
+       output, and if there is more than one file, the file name is output  at
+       the start of each line, followed by a colon. However, there are options
+       that can change how pcregrep behaves.  In  particular,  the  -M  option
+       makes  it  possible  to  search for patterns that span line boundaries.
+       What defines a line  boundary  is  controlled  by  the  -N  (--newline)
+       option.
+
+       The amount of memory used for buffering files that are being scanned is
+       controlled by a parameter that can be set by the --buffer-size  option.
+       The  default  value  for  this  parameter is specified when pcregrep is
+       built, with the default default being 20K.  A  block  of  memory  three
+       times  this  size  is used (to allow for buffering "before" and "after"
+       lines). An error occurs if a line overflows the buffer.
+
+       Patterns can be no longer than 8K or BUFSIZ  bytes,  whichever  is  the
+       greater.   BUFSIZ  is defined in . When there is more than one
+       pattern (specified by the use of -e and/or -f), each pattern is applied
+       to  each  line  in the order in which they are defined, except that all
+       the -e patterns are tried before the -f patterns.
+
+       By default, as soon as one pattern matches a line, no further  patterns
+       are considered. However, if --colour (or --color) is used to colour the
+       matching substrings, or if --only-matching, --file-offsets, or  --line-
+       offsets  is  used  to  output  only  the  part of the line that matched
+       (either shown literally, or as an offset), scanning resumes immediately
+       following  the  match,  so that further matches on the same line can be
+       found. If there are multiple  patterns,  they  are  all  tried  on  the
+       remainder  of  the  line, but patterns that follow the one that matched
+       are not tried on the earlier part of the line.
+
+       This behaviour means that the order  in  which  multiple  patterns  are
+       specified  can affect the output when one of the above options is used.
+       This is no longer the same behaviour as GNU grep, which now manages  to
+       display  earlier  matches  for  later  patterns (as long as there is no
+       overlap).
+
+       Patterns that can match an empty string are accepted, but empty  string
+       matches   are   never   recognized.   An   example   is   the   pattern
+       "(super)?(man)?", in which all components are  optional.  This  pattern
+       finds  all  occurrences  of  both "super" and "man"; the output differs
+       from matching with "super|man" when only the  matching  substrings  are
+       being shown.
+
+       If  the  LC_ALL  or LC_CTYPE environment variable is set, pcregrep uses
+       the value to set a locale when calling the PCRE library.  The  --locale
+       option can be used to override this.
+
+
+SUPPORT FOR COMPRESSED FILES
+
+       It  is  possible  to compile pcregrep so that it uses libz or libbz2 to
+       read files whose names end in .gz or .bz2, respectively. You  can  find
+       out whether your binary has support for one or both of these file types
+       by running it with the --help option. If the appropriate support is not
+       present,  files are treated as plain text. The standard input is always
+       so treated.
+
+
+BINARY FILES
+
+       By default, a file that contains a binary zero byte  within  the  first
+       1024  bytes is identified as a binary file, and is processed specially.
+       (GNU grep also  identifies  binary  files  in  this  manner.)  See  the
+       --binary-files  option for a means of changing the way binary files are
+       handled.
+
+
+OPTIONS
+
+       The order in which some of the options appear can  affect  the  output.
+       For  example,  both  the  -h and -l options affect the printing of file
+       names. Whichever comes later in the command line will be the  one  that
+       takes  effect.  Similarly,  except  where  noted below, if an option is
+       given twice, the later setting is used. Numerical  values  for  options
+       may  be  followed  by  K  or  M,  to  signify multiplication by 1024 or
+       1024*1024 respectively.
+
+       --        This terminates the list of options. It is useful if the next
+                 item  on  the command line starts with a hyphen but is not an
+                 option. This allows for the processing of patterns and  file-
+                 names that start with hyphens.
+
+       -A number, --after-context=number
+                 Output  number  lines of context after each matching line. If
+                 filenames and/or line numbers are being output, a hyphen sep-
+                 arator  is  used  instead of a colon for the context lines. A
+                 line containing "--" is output between each group  of  lines,
+                 unless  they  are  in  fact contiguous in the input file. The
+                 value of number is expected to be relatively small.  However,
+                 pcregrep guarantees to have up to 8K of following text avail-
+                 able for context output.
+
+       -a, --text
+                 Treat binary files as text. This is equivalent  to  --binary-
+                 files=text.
+
+       -B number, --before-context=number
+                 Output  number lines of context before each matching line. If
+                 filenames and/or line numbers are being output, a hyphen sep-
+                 arator  is  used  instead of a colon for the context lines. A
+                 line containing "--" is output between each group  of  lines,
+                 unless  they  are  in  fact contiguous in the input file. The
+                 value of number is expected to be relatively small.  However,
+                 pcregrep guarantees to have up to 8K of preceding text avail-
+                 able for context output.
+
+       --binary-files=word
+                 Specify how binary files are to be processed. If the word  is
+                 "binary"  (the  default),  pattern  matching  is performed on
+                 binary files, but the only  output  is  "Binary  file  
+                 matches"  when a match succeeds. If the word is "text", which
+                 is equivalent to the -a or --text option,  binary  files  are
+                 processed  in  the  same way as any other file. In this case,
+                 when a match succeeds, the  output  may  be  binary  garbage,
+                 which  can  have  nasty effects if sent to a terminal. If the
+                 word is  "without-match",  which  is  equivalent  to  the  -I
+                 option,  binary  files  are  not  processed  at all; they are
+                 assumed not to be of interest.
+
+       --buffer-size=number
+                 Set the parameter that controls how much memory is  used  for
+                 buffering files that are being scanned.
+
+       -C number, --context=number
+                 Output  number  lines  of  context both before and after each
+                 matching line.  This is equivalent to setting both -A and  -B
+                 to the same value.
+
+       -c, --count
+                 Do  not output individual lines from the files that are being
+                 scanned; instead output the number of lines that would other-
+                 wise  have  been  shown. If no lines are selected, the number
+                 zero is output. If several files are  are  being  scanned,  a
+                 count  is  output  for each of them. However, if the --files-
+                 with-matches option is also  used,  only  those  files  whose
+                 counts are greater than zero are listed. When -c is used, the
+                 -A, -B, and -C options are ignored.
+
+       --colour, --color
+                 If this option is given without any data, it is equivalent to
+                 "--colour=auto".   If  data  is required, it must be given in
+                 the same shell item, separated by an equals sign.
+
+       --colour=value, --color=value
+                 This option specifies under what circumstances the parts of a
+                 line that matched a pattern should be coloured in the output.
+                 By default, the output is not coloured. The value  (which  is
+                 optional,  see above) may be "never", "always", or "auto". In
+                 the latter case, colouring happens only if the standard  out-
+                 put  is connected to a terminal. More resources are used when
+                 colouring is enabled, because pcregrep has to search for  all
+                 possible  matches in a line, not just one, in order to colour
+                 them all.
+
+                 The colour that is used can be specified by setting the envi-
+                 ronment variable PCREGREP_COLOUR or PCREGREP_COLOR. The value
+                 of this variable should be a string of two numbers, separated
+                 by  a  semicolon.  They  are copied directly into the control
+                 string for setting colour  on  a  terminal,  so  it  is  your
+                 responsibility  to ensure that they make sense. If neither of
+                 the environment variables is  set,  the  default  is  "1;31",
+                 which gives red.
+
+       -D action, --devices=action
+                 If  an  input  path  is  not  a  regular file or a directory,
+                 "action" specifies how it is to be  processed.  Valid  values
+                 are "read" (the default) or "skip" (silently skip the path).
+
+       -d action, --directories=action
+                 If an input path is a directory, "action" specifies how it is
+                 to be processed.  Valid values are  "read"  (the  default  in
+                 non-Windows  environments,  for compatibility with GNU grep),
+                 "recurse" (equivalent to the -r option), or "skip"  (silently
+                 skip  the  path, the default in Windows environments). In the
+                 "read" case, directories are read as if  they  were  ordinary
+                 files.  In  some  operating  systems  the effect of reading a
+                 directory like this is an immediate end-of-file; in others it
+                 may provoke an error.
+
+       -e pattern, --regex=pattern, --regexp=pattern
+                 Specify a pattern to be matched. This option can be used mul-
+                 tiple times in order to specify several patterns. It can also
+                 be  used  as a way of specifying a single pattern that starts
+                 with a hyphen. When -e is used, no argument pattern is  taken
+                 from  the  command  line;  all  arguments are treated as file
+                 names. There is no limit to the number of patterns. They  are
+                 applied  to  each line in the order in which they are defined
+                 until one matches.
+
+                 If -f is used with -e, the command line patterns are  matched
+                 first, followed by the patterns from the file(s), independent
+                 of the order in which these options are specified. Note  that
+                 multiple  use  of -e is not the same as a single pattern with
+                 alternatives. For example, X|Y finds the first character in a
+                 line  that  is  X or Y, whereas if the two patterns are given
+                 separately, with X first, pcregrep finds X if it is  present,
+                 even if it follows Y in the line. It finds Y only if there is
+                 no X in the line. This matters only if you are  using  -o  or
+                 --colo(u)r to show the part(s) of the line that matched.
+
+       --exclude=pattern
+                 Files (but not directories) whose names match the pattern are
+                 skipped without being processed. This applies to  all  files,
+                 whether  listed  on  the  command line, obtained from --file-
+                 list, or by scanning a directory. The pattern is a PCRE regu-
+                 lar expression, and is matched against the final component of
+                 the file name, not the  entire  path.  The  -F,  -w,  and  -x
+                 options do not apply to this pattern. The option may be given
+                 any number of times in order to specify multiple patterns. If
+                 a  file  name matches both an --include and an --exclude pat-
+                 tern, it is excluded. There is no short form for this option.
+
+       --exclude-from=filename
+                 Treat each non-empty line of the file  as  the  data  for  an
+                 --exclude option. What constitutes a newline when reading the
+                 file is the operating system's default. The --newline  option
+                 has  no  effect on this option. This option may be given more
+                 than once in order to specify a number of files to read.
+
+       --exclude-dir=pattern
+                 Directories whose names match the pattern are skipped without
+                 being  processed,  whatever  the  setting  of the --recursive
+                 option. This applies to all directories,  whether  listed  on
+                 the command line, obtained from --file-list, or by scanning a
+                 parent directory. The pattern is a PCRE  regular  expression,
+                 and  is  matched against the final component of the directory
+                 name, not the entire path. The -F, -w, and -x options do  not
+                 apply  to this pattern. The option may be given any number of
+                 times in order to specify more than one pattern. If a  direc-
+                 tory  matches  both  --include-dir  and  --exclude-dir, it is
+                 excluded. There is no short form for this option.
+
+       -F, --fixed-strings
+                 Interpret each data-matching  pattern  as  a  list  of  fixed
+                 strings,  separated  by  newlines,  instead  of  as a regular
+                 expression. What constitutes a newline for  this  purpose  is
+                 controlled  by the --newline option. The -w (match as a word)
+                 and -x (match whole line) options can be used with -F.   They
+                 apply to each of the fixed strings. A line is selected if any
+                 of the fixed strings are found in it (subject to -w or -x, if
+                 present).  This  option applies only to the patterns that are
+                 matched against the contents of files; it does not  apply  to
+                 patterns  specified  by  any  of  the  --include or --exclude
+                 options.
+
+       -f filename, --file=filename
+                 Read patterns from the file, one per  line,  and  match  them
+                 against  each  line of input. What constitutes a newline when
+                 reading the file  is  the  operating  system's  default.  The
+                 --newline option has no effect on this option. Trailing white
+                 space is removed from each line, and blank lines are ignored.
+                 An  empty  file  contains  no  patterns and therefore matches
+                 nothing. See also the comments about multiple patterns versus
+                 a  single  pattern with alternatives in the description of -e
+                 above.
+
+                 If this option is given more than  once,  all  the  specified
+                 files  are read. A data line is output if any of the patterns
+                 match it. A filename can be given as  "-"  to  refer  to  the
+                 standard  input.  When  -f is used, patterns specified on the
+                 command line using -e may also be present;  they  are  tested
+                 before  the  file's  patterns.  However,  no other pattern is
+                 taken from the command line; all arguments are treated as the
+                 names of paths to be searched.
+
+       --file-list=filename
+                 Read  a  list  of  files  and/or  directories  that are to be
+                 scanned from the given file, one  per  line.  Trailing  white
+                 space is removed from each line, and blank lines are ignored.
+                 These paths are processed before any that are listed  on  the
+                 command  line.  The  filename can be given as "-" to refer to
+                 the standard input.  If --file and --file-list are both spec-
+                 ified  as  "-",  patterns are read first. This is useful only
+                 when the standard input is a  terminal,  from  which  further
+                 lines  (the  list  of files) can be read after an end-of-file
+                 indication. If this option is given more than once,  all  the
+                 specified files are read.
+
+       --file-offsets
+                 Instead  of  showing lines or parts of lines that match, show
+                 each match as an offset from the start  of  the  file  and  a
+                 length,  separated  by  a  comma. In this mode, no context is
+                 shown. That is, the -A, -B, and -C options  are  ignored.  If
+                 there is more than one match in a line, each of them is shown
+                 separately. This option is mutually  exclusive  with  --line-
+                 offsets and --only-matching.
+
+       -H, --with-filename
+                 Force  the  inclusion  of the filename at the start of output
+                 lines when searching a single file. By default, the  filename
+                 is  not  shown in this case. For matching lines, the filename
+                 is followed by a colon; for context lines, a hyphen separator
+                 is  used.  If  a line number is also being output, it follows
+                 the file name.
+
+       -h, --no-filename
+                 Suppress the output filenames when searching multiple  files.
+                 By  default,  filenames  are  shown  when  multiple files are
+                 searched. For matching lines, the filename is followed  by  a
+                 colon;  for  context lines, a hyphen separator is used.  If a
+                 line number is also being output, it follows the file name.
+
+       --help    Output a help message, giving brief details  of  the  command
+                 options  and  file type support, and then exit. Anything else
+                 on the command line is ignored.
+
+       -I        Treat binary files as never matching. This is  equivalent  to
+                 --binary-files=without-match.
+
+       -i, --ignore-case
+                 Ignore upper/lower case distinctions during comparisons.
+
+       --include=pattern
+                 If  any --include patterns are specified, the only files that
+                 are processed are those that match one of the  patterns  (and
+                 do  not  match  an  --exclude  pattern). This option does not
+                 affect directories, but it  applies  to  all  files,  whether
+                 listed  on the command line, obtained from --file-list, or by
+                 scanning a directory. The pattern is a PCRE  regular  expres-
+                 sion,  and is matched against the final component of the file
+                 name, not the entire path. The -F, -w, and -x options do  not
+                 apply  to this pattern. The option may be given any number of
+                 times. If a file  name  matches  both  an  --include  and  an
+                 --exclude  pattern,  it  is excluded.  There is no short form
+                 for this option.
+
+       --include-from=filename
+                 Treat each non-empty line of the file  as  the  data  for  an
+                 --include option. What constitutes a newline for this purpose
+                 is the operating system's default. The --newline  option  has
+                 no effect on this option. This option may be given any number
+                 of times; all the files are read.
+
+       --include-dir=pattern
+                 If any --include-dir patterns are specified, the only  direc-
+                 tories  that  are  processed  are those that match one of the
+                 patterns (and do not match an  --exclude-dir  pattern).  This
+                 applies  to  all  directories,  whether listed on the command
+                 line, obtained from --file-list,  or  by  scanning  a  parent
+                 directory.  The  pattern is a PCRE regular expression, and is
+                 matched against the final component of  the  directory  name,
+                 not  the entire path. The -F, -w, and -x options do not apply
+                 to this pattern. The option may be given any number of times.
+                 If  a directory matches both --include-dir and --exclude-dir,
+                 it is excluded. There is no short form for this option.
+
+       -L, --files-without-match
+                 Instead of outputting lines from the files, just  output  the
+                 names  of  the files that do not contain any lines that would
+                 have been output. Each file name is output once, on  a  sepa-
+                 rate line.
+
+       -l, --files-with-matches
+                 Instead  of  outputting lines from the files, just output the
+                 names of the files containing lines that would have been out-
+                 put.  Each  file  name  is  output  once, on a separate line.
+                 Searching normally stops as soon as a matching line is  found
+                 in  a  file.  However, if the -c (count) option is also used,
+                 matching continues in order to obtain the correct count,  and
+                 those  files  that  have  at least one match are listed along
+                 with their counts. Using this option with -c is a way of sup-
+                 pressing the listing of files with no matches.
+
+       --label=name
+                 This option supplies a name to be used for the standard input
+                 when file names are being output. If not supplied, "(standard
+                 input)" is used. There is no short form for this option.
+
+       --line-buffered
+                 When  this  option is given, input is read and processed line
+                 by line, and the output  is  flushed  after  each  write.  By
+                 default,  input  is read in large chunks, unless pcregrep can
+                 determine that it is reading from a terminal (which  is  cur-
+                 rently  possible  only  in Unix-like environments). Output to
+                 terminal is normally automatically flushed by  the  operating
+                 system. This option can be useful when the input or output is
+                 attached to a pipe and you do not want pcregrep to buffer  up
+                 large  amounts  of data. However, its use will affect perfor-
+                 mance, and the -M (multiline) option ceases to work.
+
+       --line-offsets
+                 Instead of showing lines or parts of lines that  match,  show
+                 each match as a line number, the offset from the start of the
+                 line, and a length. The line number is terminated by a  colon
+                 (as  usual; see the -n option), and the offset and length are
+                 separated by a comma. In this  mode,  no  context  is  shown.
+                 That  is, the -A, -B, and -C options are ignored. If there is
+                 more than one match in a line, each of them  is  shown  sepa-
+                 rately. This option is mutually exclusive with --file-offsets
+                 and --only-matching.
+
+       --locale=locale-name
+                 This option specifies a locale to be used for pattern  match-
+                 ing.  It  overrides the value in the LC_ALL or LC_CTYPE envi-
+                 ronment variables.  If  no  locale  is  specified,  the  PCRE
+                 library's  default (usually the "C" locale) is used. There is
+                 no short form for this option.
+
+       --match-limit=number
+                 Processing some regular expression  patterns  can  require  a
+                 very  large amount of memory, leading in some cases to a pro-
+                 gram crash if not enough is available.   Other  patterns  may
+                 take  a  very  long  time to search for all possible matching
+                 strings. The pcre_exec() function that is called by  pcregrep
+                 to  do  the  matching  has  two parameters that can limit the
+                 resources that it uses.
+
+                 The  --match-limit  option  provides  a  means  of   limiting
+                 resource usage when processing patterns that are not going to
+                 match, but which have a very large number of possibilities in
+                 their  search  trees.  The  classic example is a pattern that
+                 uses nested unlimited repeats. Internally, PCRE uses a  func-
+                 tion  called  match()  which  it  calls repeatedly (sometimes
+                 recursively). The limit set by --match-limit  is  imposed  on
+                 the  number  of times this function is called during a match,
+                 which has the effect of limiting the amount  of  backtracking
+                 that can take place.
+
+                 The --recursion-limit option is similar to --match-limit, but
+                 instead of limiting the total number of times that match() is
+                 called, it limits the depth of recursive calls, which in turn
+                 limits the amount of memory that can be used.  The  recursion
+                 depth  is  a  smaller  number than the total number of calls,
+                 because not all calls to match() are recursive. This limit is
+                 of use only if it is set smaller than --match-limit.
+
+                 There  are no short forms for these options. The default set-
+                 tings are specified when the PCRE library is  compiled,  with
+                 the default default being 10 million.
+
+       -M, --multiline
+                 Allow  patterns to match more than one line. When this option
+                 is given, patterns may usefully contain literal newline char-
+                 acters  and  internal  occurrences of ^ and $ characters. The
+                 output for a successful match may consist of  more  than  one
+                 line,  the last of which is the one in which the match ended.
+                 If the matched string ends with a newline sequence the output
+                 ends at the end of that line.
+
+                 When  this option is set, the PCRE library is called in "mul-
+                 tiline" mode.  There is a limit to the number of  lines  that
+                 can  be matched, imposed by the way that pcregrep buffers the
+                 input file as it scans it. However, pcregrep ensures that  at
+                 least 8K characters or the rest of the document (whichever is
+                 the shorter) are available for forward  matching,  and  simi-
+                 larly the previous 8K characters (or all the previous charac-
+                 ters, if fewer than 8K) are guaranteed to  be  available  for
+                 lookbehind  assertions.  This option does not work when input
+                 is read line by line (see --line-buffered.)
+
+       -N newline-type, --newline=newline-type
+                 The PCRE library  supports  five  different  conventions  for
+                 indicating  the  ends of lines. They are the single-character
+                 sequences CR (carriage return) and LF  (linefeed),  the  two-
+                 character  sequence CRLF, an "anycrlf" convention, which rec-
+                 ognizes any of the preceding three types, and an  "any"  con-
+                 vention, in which any Unicode line ending sequence is assumed
+                 to end a line. The Unicode sequences are the three just  men-
+                 tioned,  plus  VT  (vertical  tab,  U+000B),  FF  (form feed,
+                 U+000C),  NEL  (next  line,  U+0085),  LS  (line   separator,
+                 U+2028), and PS (paragraph separator, U+2029).
+
+                 When  the  PCRE  library  is  built,  a  default  line-ending
+                 sequence  is  specified.   This  is  normally  the   standard
+                 sequence for the operating system. Unless otherwise specified
+                 by this option, pcregrep uses  the  library's  default.   The
+                 possible values for this option are CR, LF, CRLF, ANYCRLF, or
+                 ANY. This makes it possible to use  pcregrep  to  scan  files
+                 that have come from other environments without having to mod-
+                 ify their line endings. If the data  that  is  being  scanned
+                 does  not agree with the convention set by this option, pcre-
+                 grep may behave in strange ways. Note that this  option  does
+                 not  apply  to  files specified by the -f, --exclude-from, or
+                 --include-from options, which are expected to use the operat-
+                 ing system's standard newline sequence.
+
+       -n, --line-number
+                 Precede each output line by its line number in the file, fol-
+                 lowed by a colon for matching lines or a hyphen  for  context
+                 lines.  If the filename is also being output, it precedes the
+                 line number. This option is forced if --line-offsets is used.
+
+       --no-jit  If the PCRE library is built with  support  for  just-in-time
+                 compiling  (which speeds up matching), pcregrep automatically
+                 makes use of this, unless it was explicitly disabled at build
+                 time.  This  option  can be used to disable the use of JIT at
+                 run time. It is provided for testing and working round  prob-
+                 lems.  It should never be needed in normal use.
+
+       -o, --only-matching
+                 Show only the part of the line that matched a pattern instead
+                 of the whole line. In this mode, no context  is  shown.  That
+                 is,  the -A, -B, and -C options are ignored. If there is more
+                 than one match in a line, each of them is  shown  separately.
+                 If  -o  is combined with -v (invert the sense of the match to
+                 find non-matching lines), no output  is  generated,  but  the
+                 return  code  is set appropriately. If the matched portion of
+                 the line is empty, nothing is output unless the file name  or
+                 line  number  are being printed, in which case they are shown
+                 on an otherwise empty line. This option is mutually exclusive
+                 with --file-offsets and --line-offsets.
+
+       -onumber, --only-matching=number
+                 Show  only  the  part  of the line that matched the capturing
+                 parentheses of the given number. Up to 32 capturing parenthe-
+                 ses are supported, and -o0 is equivalent to -o without a num-
+                 ber. Because these options can be given without  an  argument
+                 (see  above),  if an argument is present, it must be given in
+                 the same shell item, for example, -o3  or  --only-matching=2.
+                 The comments given for the non-argument case above also apply
+                 to this case. If the specified capturing parentheses  do  not
+                 exist  in  the pattern, or were not set in the match, nothing
+                 is output unless the file  name  or  line  number  are  being
+                 printed.
+
+                 If  this  option is given multiple times, multiple substrings
+                 are output, in the order the options are given. For  example,
+                 -o3 -o1 -o3 causes the substrings matched by capturing paren-
+                 theses 3 and 1 and then 3 again to  be  output.  By  default,
+                 there is no separator (but see the next option).
+
+       --om-separator=text
+                 Specify  a  separating string for multiple occurrences of -o.
+                 The default is an empty string. Separating strings are  never
+                 coloured.
+
+       -q, --quiet
+                 Work quietly, that is, display nothing except error messages.
+                 The exit status indicates whether or  not  any  matches  were
+                 found.
+
+       -r, --recursive
+                 If  any given path is a directory, recursively scan the files
+                 it contains, taking note of any --include and --exclude  set-
+                 tings.  By  default, a directory is read as a normal file; in
+                 some operating systems this gives an  immediate  end-of-file.
+                 This  option  is  a  shorthand  for  setting the -d option to
+                 "recurse".
+
+       --recursion-limit=number
+                 See --match-limit above.
+
+       -s, --no-messages
+                 Suppress error  messages  about  non-existent  or  unreadable
+                 files.  Such  files  are quietly skipped. However, the return
+                 code is still 2, even if matches were found in other files.
+
+       -u, --utf-8
+                 Operate in UTF-8 mode. This option is available only if  PCRE
+                 has been compiled with UTF-8 support. All patterns (including
+                 those for any --exclude and --include options) and  all  sub-
+                 ject  lines  that  are scanned must be valid strings of UTF-8
+                 characters.
+
+       -V, --version
+                 Write the version numbers of pcregrep and the PCRE library to
+                 the  standard output and then exit. Anything else on the com-
+                 mand line is ignored.
+
+       -v, --invert-match
+                 Invert the sense of the match, so that  lines  which  do  not
+                 match any of the patterns are the ones that are found.
+
+       -w, --word-regex, --word-regexp
+                 Force the patterns to match only whole words. This is equiva-
+                 lent to having \b at the start and end of the  pattern.  This
+                 option  applies only to the patterns that are matched against
+                 the contents of files; it does not apply to  patterns  speci-
+                 fied by any of the --include or --exclude options.
+
+       -x, --line-regex, --line-regexp
+                 Force  the  patterns to be anchored (each must start matching
+                 at the beginning of a line) and in addition, require them  to
+                 match  entire  lines.  This  is  equivalent to having ^ and $
+                 characters at the start and end of each alternative branch in
+                 every  pattern. This option applies only to the patterns that
+                 are matched against the contents of files; it does not  apply
+                 to  patterns  specified  by any of the --include or --exclude
+                 options.
+
+
+ENVIRONMENT VARIABLES
+
+       The environment variables LC_ALL and LC_CTYPE  are  examined,  in  that
+       order,  for  a  locale.  The first one that is set is used. This can be
+       overridden by the --locale option.  If  no  locale  is  set,  the  PCRE
+       library's default (usually the "C" locale) is used.
+
+
+NEWLINES
+
+       The  -N (--newline) option allows pcregrep to scan files with different
+       newline conventions from the default. Any parts of the input files that
+       are  written  to the standard output are copied identically, with what-
+       ever newline sequences they have in the input. However, the setting  of
+       this  option  does  not affect the interpretation of files specified by
+       the -f, --exclude-from, or --include-from options, which are assumed to
+       use  the  operating  system's  standard  newline  sequence, nor does it
+       affect the way in which pcregrep writes informational messages  to  the
+       standard error and output streams. For these it uses the string "\n" to
+       indicate newlines, relying on the C I/O library to convert this  to  an
+       appropriate sequence.
+
+
+OPTIONS COMPATIBILITY
+
+       Many  of the short and long forms of pcregrep's options are the same as
+       in the GNU grep program. Any long option of the form --xxx-regexp  (GNU
+       terminology)  is also available as --xxx-regex (PCRE terminology). How-
+       ever, the --file-list, --file-offsets,  --include-dir,  --line-offsets,
+       --locale,  --match-limit,  -M, --multiline, -N, --newline, --om-separa-
+       tor, --recursion-limit, -u, and --utf-8 options are specific  to  pcre-
+       grep,  as  is  the  use  of the --only-matching option with a capturing
+       parentheses number.
+
+       Although most of the common options work the same way, a few  are  dif-
+       ferent  in  pcregrep. For example, the --include option's argument is a
+       glob for GNU grep, but a regular expression for pcregrep. If  both  the
+       -c  and  -l  options are given, GNU grep lists only file names, without
+       counts, but pcregrep gives the counts.
+
+
+OPTIONS WITH DATA
+
+       There are four different ways in which an option with data can be spec-
+       ified.   If  a  short  form option is used, the data may follow immedi-
+       ately, or (with one exception) in the next command line item. For exam-
+       ple:
+
+         -f/some/file
+         -f /some/file
+
+       The  exception is the -o option, which may appear with or without data.
+       Because of this, if data is present, it must follow immediately in  the
+       same item, for example -o3.
+
+       If  a long form option is used, the data may appear in the same command
+       line item, separated by an equals character, or (with  two  exceptions)
+       it may appear in the next command line item. For example:
+
+         --file=/some/file
+         --file /some/file
+
+       Note,  however, that if you want to supply a file name beginning with ~
+       as data in a shell command, and have the  shell  expand  ~  to  a  home
+       directory, you must separate the file name from the option, because the
+       shell does not treat ~ specially unless it is at the start of an item.
+
+       The exceptions to the above are the --colour (or --color)  and  --only-
+       matching  options,  for  which  the  data  is optional. If one of these
+       options does have data, it must be given in the first  form,  using  an
+       equals character. Otherwise pcregrep will assume that it has no data.
+
+
+MATCHING ERRORS
+
+       It  is  possible  to supply a regular expression that takes a very long
+       time to fail to match certain lines.  Such  patterns  normally  involve
+       nested  indefinite repeats, for example: (a+)*\d when matched against a
+       line of a's with no final digit.  The  PCRE  matching  function  has  a
+       resource  limit that causes it to abort in these circumstances. If this
+       happens, pcregrep outputs an error message and the line that caused the
+       problem  to  the  standard error stream. If there are more than 20 such
+       errors, pcregrep gives up.
+
+       The --match-limit option of pcregrep can be used  to  set  the  overall
+       resource  limit; there is a second option called --recursion-limit that
+       sets a limit on the amount of memory (usually stack) that is used  (see
+       the discussion of these options above).
+
+
+DIAGNOSTICS
+
+       Exit status is 0 if any matches were found, 1 if no matches were found,
+       and 2 for syntax errors, overlong lines, non-existent  or  inaccessible
+       files  (even if matches were found in other files) or too many matching
+       errors. Using the -s option to suppress error messages about inaccessi-
+       ble files does not affect the return code.
+
+
+SEE ALSO
+
+       pcrepattern(3), pcresyntax(3), pcretest(1).
+
+
+AUTHOR
+
+       Philip Hazel
+       University Computing Service
+       Cambridge CB2 3QH, England.
+
+
+REVISION
+
+       Last updated: 13 September 2012
+       Copyright (c) 1997-2012 University of Cambridge.
diff --git a/pcre/doc/pcrejit.3 b/pcre/doc/pcrejit.3
new file mode 100644
index 00000000000..341403f7c84
--- /dev/null
+++ b/pcre/doc/pcrejit.3
@@ -0,0 +1,431 @@
+.TH PCREJIT 3 "17 March 2013" "PCRE 8.33"
+.SH NAME
+PCRE - Perl-compatible regular expressions
+.SH "PCRE JUST-IN-TIME COMPILER SUPPORT"
+.rs
+.sp
+Just-in-time compiling is a heavyweight optimization that can greatly speed up
+pattern matching. However, it comes at the cost of extra processing before the
+match is performed. Therefore, it is of most benefit when the same pattern is
+going to be matched many times. This does not necessarily mean many calls of a
+matching function; if the pattern is not anchored, matching attempts may take
+place many times at various positions in the subject, even for a single call.
+Therefore, if the subject string is very long, it may still pay to use JIT for
+one-off matches.
+.P
+JIT support applies only to the traditional Perl-compatible matching function.
+It does not apply when the DFA matching function is being used. The code for
+this support was written by Zoltan Herczeg.
+.
+.
+.SH "8-BIT, 16-BIT AND 32-BIT SUPPORT"
+.rs
+.sp
+JIT support is available for all of the 8-bit, 16-bit and 32-bit PCRE
+libraries. To keep this documentation simple, only the 8-bit interface is
+described in what follows. If you are using the 16-bit library, substitute the
+16-bit functions and 16-bit structures (for example, \fIpcre16_jit_stack\fP
+instead of \fIpcre_jit_stack\fP). If you are using the 32-bit library,
+substitute the 32-bit functions and 32-bit structures (for example,
+\fIpcre32_jit_stack\fP instead of \fIpcre_jit_stack\fP).
+.
+.
+.SH "AVAILABILITY OF JIT SUPPORT"
+.rs
+.sp
+JIT support is an optional feature of PCRE. The "configure" option --enable-jit
+(or equivalent CMake option) must be set when PCRE is built if you want to use
+JIT. The support is limited to the following hardware platforms:
+.sp
+  ARM v5, v7, and Thumb2
+  Intel x86 32-bit and 64-bit
+  MIPS 32-bit
+  Power PC 32-bit and 64-bit
+  SPARC 32-bit (experimental)
+.sp
+If --enable-jit is set on an unsupported platform, compilation fails.
+.P
+A program that is linked with PCRE 8.20 or later can tell if JIT support is
+available by calling \fBpcre_config()\fP with the PCRE_CONFIG_JIT option. The
+result is 1 when JIT is available, and 0 otherwise. However, a simple program
+does not need to check this in order to use JIT. The normal API is implemented
+in a way that falls back to the interpretive code if JIT is not available. For
+programs that need the best possible performance, there is also a "fast path"
+API that is JIT-specific.
+.P
+If your program may sometimes be linked with versions of PCRE that are older
+than 8.20, but you want to use JIT when it is available, you can test
+the values of PCRE_MAJOR and PCRE_MINOR, or the existence of a JIT macro such
+as PCRE_CONFIG_JIT, for compile-time control of your code.
+.
+.
+.SH "SIMPLE USE OF JIT"
+.rs
+.sp
+You have to do two things to make use of the JIT support in the simplest way:
+.sp
+  (1) Call \fBpcre_study()\fP with the PCRE_STUDY_JIT_COMPILE option for
+      each compiled pattern, and pass the resulting \fBpcre_extra\fP block to
+      \fBpcre_exec()\fP.
+.sp
+  (2) Use \fBpcre_free_study()\fP to free the \fBpcre_extra\fP block when it is
+      no longer needed, instead of just freeing it yourself. This ensures that
+      any JIT data is also freed.
+.sp
+For a program that may be linked with pre-8.20 versions of PCRE, you can insert
+.sp
+  #ifndef PCRE_STUDY_JIT_COMPILE
+  #define PCRE_STUDY_JIT_COMPILE 0
+  #endif
+.sp
+so that no option is passed to \fBpcre_study()\fP, and then use something like
+this to free the study data:
+.sp
+  #ifdef PCRE_CONFIG_JIT
+      pcre_free_study(study_ptr);
+  #else
+      pcre_free(study_ptr);
+  #endif
+.sp
+PCRE_STUDY_JIT_COMPILE requests the JIT compiler to generate code for complete
+matches. If you want to run partial matches using the PCRE_PARTIAL_HARD or
+PCRE_PARTIAL_SOFT options of \fBpcre_exec()\fP, you should set one or both of
+the following options in addition to, or instead of, PCRE_STUDY_JIT_COMPILE
+when you call \fBpcre_study()\fP:
+.sp
+  PCRE_STUDY_JIT_PARTIAL_HARD_COMPILE
+  PCRE_STUDY_JIT_PARTIAL_SOFT_COMPILE
+.sp
+The JIT compiler generates different optimized code for each of the three
+modes (normal, soft partial, hard partial). When \fBpcre_exec()\fP is called,
+the appropriate code is run if it is available. Otherwise, the pattern is
+matched using interpretive code.
+.P
+In some circumstances you may need to call additional functions. These are
+described in the section entitled
+.\" HTML 
+.\" 
+"Controlling the JIT stack"
+.\"
+below.
+.P
+If JIT support is not available, PCRE_STUDY_JIT_COMPILE etc. are ignored, and
+no JIT data is created. Otherwise, the compiled pattern is passed to the JIT
+compiler, which turns it into machine code that executes much faster than the
+normal interpretive code. When \fBpcre_exec()\fP is passed a \fBpcre_extra\fP
+block containing a pointer to JIT code of the appropriate mode (normal or
+hard/soft partial), it obeys that code instead of running the interpreter. The
+result is identical, but the compiled JIT code runs much faster.
+.P
+There are some \fBpcre_exec()\fP options that are not supported for JIT
+execution. There are also some pattern items that JIT cannot handle. Details
+are given below. In both cases, execution automatically falls back to the
+interpretive code. If you want to know whether JIT was actually used for a
+particular match, you should arrange for a JIT callback function to be set up
+as described in the section entitled
+.\" HTML 
+.\" 
+"Controlling the JIT stack"
+.\"
+below, even if you do not need to supply a non-default JIT stack. Such a
+callback function is called whenever JIT code is about to be obeyed. If the
+execution options are not right for JIT execution, the callback function is not
+obeyed.
+.P
+If the JIT compiler finds an unsupported item, no JIT data is generated. You
+can find out if JIT execution is available after studying a pattern by calling
+\fBpcre_fullinfo()\fP with the PCRE_INFO_JIT option. A result of 1 means that
+JIT compilation was successful. A result of 0 means that JIT support is not
+available, or the pattern was not studied with PCRE_STUDY_JIT_COMPILE etc., or
+the JIT compiler was not able to handle the pattern.
+.P
+Once a pattern has been studied, with or without JIT, it can be used as many
+times as you like for matching different subject strings.
+.
+.
+.SH "UNSUPPORTED OPTIONS AND PATTERN ITEMS"
+.rs
+.sp
+The only \fBpcre_exec()\fP options that are supported for JIT execution are
+PCRE_NO_UTF8_CHECK, PCRE_NO_UTF16_CHECK, PCRE_NO_UTF32_CHECK, PCRE_NOTBOL,
+PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART, PCRE_PARTIAL_HARD, and
+PCRE_PARTIAL_SOFT.
+.P
+The only unsupported pattern items are \eC (match a single data unit) when
+running in a UTF mode, and a callout immediately before an assertion condition
+in a conditional group.
+.
+.
+.SH "RETURN VALUES FROM JIT EXECUTION"
+.rs
+.sp
+When a pattern is matched using JIT execution, the return values are the same
+as those given by the interpretive \fBpcre_exec()\fP code, with the addition of
+one new error code: PCRE_ERROR_JIT_STACKLIMIT. This means that the memory used
+for the JIT stack was insufficient. See
+.\" HTML 
+.\" 
+"Controlling the JIT stack"
+.\"
+below for a discussion of JIT stack usage. For compatibility with the
+interpretive \fBpcre_exec()\fP code, no more than two-thirds of the
+\fIovector\fP argument is used for passing back captured substrings.
+.P
+The error code PCRE_ERROR_MATCHLIMIT is returned by the JIT code if searching a
+very large pattern tree goes on for too long, as it is in the same circumstance
+when JIT is not used, but the details of exactly what is counted are not the
+same. The PCRE_ERROR_RECURSIONLIMIT error code is never returned by JIT
+execution.
+.
+.
+.SH "SAVING AND RESTORING COMPILED PATTERNS"
+.rs
+.sp
+The code that is generated by the JIT compiler is architecture-specific, and is
+also position dependent. For those reasons it cannot be saved (in a file or
+database) and restored later like the bytecode and other data of a compiled
+pattern. Saving and restoring compiled patterns is not something many people
+do. More detail about this facility is given in the
+.\" HREF
+\fBpcreprecompile\fP
+.\"
+documentation. It should be possible to run \fBpcre_study()\fP on a saved and
+restored pattern, and thereby recreate the JIT data, but because JIT
+compilation uses significant resources, it is probably not worth doing this;
+you might as well recompile the original pattern.
+.
+.
+.\" HTML 
+.SH "CONTROLLING THE JIT STACK"
+.rs
+.sp
+When the compiled JIT code runs, it needs a block of memory to use as a stack.
+By default, it uses 32K on the machine stack. However, some large or
+complicated patterns need more than this. The error PCRE_ERROR_JIT_STACKLIMIT
+is given when there is not enough stack. Three functions are provided for
+managing blocks of memory for use as JIT stacks. There is further discussion
+about the use of JIT stacks in the section entitled
+.\" HTML 
+.\" 
+"JIT stack FAQ"
+.\"
+below.
+.P
+The \fBpcre_jit_stack_alloc()\fP function creates a JIT stack. Its arguments
+are a starting size and a maximum size, and it returns a pointer to an opaque
+structure of type \fBpcre_jit_stack\fP, or NULL if there is an error. The
+\fBpcre_jit_stack_free()\fP function can be used to free a stack that is no
+longer needed. (For the technically minded: the address space is allocated by
+mmap or VirtualAlloc.)
+.P
+JIT uses far less memory for recursion than the interpretive code,
+and a maximum stack size of 512K to 1M should be more than enough for any
+pattern.
+.P
+The \fBpcre_assign_jit_stack()\fP function specifies which stack JIT code
+should use. Its arguments are as follows:
+.sp
+  pcre_extra         *extra
+  pcre_jit_callback  callback
+  void               *data
+.sp
+The \fIextra\fP argument must be the result of studying a pattern with
+PCRE_STUDY_JIT_COMPILE etc. There are three cases for the values of the other
+two options:
+.sp
+  (1) If \fIcallback\fP is NULL and \fIdata\fP is NULL, an internal 32K block
+      on the machine stack is used.
+.sp
+  (2) If \fIcallback\fP is NULL and \fIdata\fP is not NULL, \fIdata\fP must be
+      a valid JIT stack, the result of calling \fBpcre_jit_stack_alloc()\fP.
+.sp
+  (3) If \fIcallback\fP is not NULL, it must point to a function that is
+      called with \fIdata\fP as an argument at the start of matching, in
+      order to set up a JIT stack. If the return from the callback
+      function is NULL, the internal 32K stack is used; otherwise the
+      return value must be a valid JIT stack, the result of calling
+      \fBpcre_jit_stack_alloc()\fP.
+.sp
+A callback function is obeyed whenever JIT code is about to be run; it is not
+obeyed when \fBpcre_exec()\fP is called with options that are incompatible for
+JIT execution. A callback function can therefore be used to determine whether a
+match operation was executed by JIT or by the interpreter.
+.P
+You may safely use the same JIT stack for more than one pattern (either by
+assigning directly or by callback), as long as the patterns are all matched
+sequentially in the same thread. In a multithread application, if you do not
+specify a JIT stack, or if you assign or pass back NULL from a callback, that
+is thread-safe, because each thread has its own machine stack. However, if you
+assign or pass back a non-NULL JIT stack, this must be a different stack for
+each thread so that the application is thread-safe.
+.P
+Strictly speaking, even more is allowed. You can assign the same non-NULL stack
+to any number of patterns as long as they are not used for matching by multiple
+threads at the same time. For example, you can assign the same stack to all
+compiled patterns, and use a global mutex in the callback to wait until the
+stack is available for use. However, this is an inefficient solution, and not
+recommended.
+.P
+This is a suggestion for how a multithreaded program that needs to set up
+non-default JIT stacks might operate:
+.sp
+  During thread initalization
+    thread_local_var = pcre_jit_stack_alloc(...)
+.sp
+  During thread exit
+    pcre_jit_stack_free(thread_local_var)
+.sp
+  Use a one-line callback function
+    return thread_local_var
+.sp
+All the functions described in this section do nothing if JIT is not available,
+and \fBpcre_assign_jit_stack()\fP does nothing unless the \fBextra\fP argument
+is non-NULL and points to a \fBpcre_extra\fP block that is the result of a
+successful study with PCRE_STUDY_JIT_COMPILE etc.
+.
+.
+.\" HTML 
+.SH "JIT STACK FAQ"
+.rs
+.sp
+(1) Why do we need JIT stacks?
+.sp
+PCRE (and JIT) is a recursive, depth-first engine, so it needs a stack where
+the local data of the current node is pushed before checking its child nodes.
+Allocating real machine stack on some platforms is difficult. For example, the
+stack chain needs to be updated every time if we extend the stack on PowerPC.
+Although it is possible, its updating time overhead decreases performance. So
+we do the recursion in memory.
+.P
+(2) Why don't we simply allocate blocks of memory with \fBmalloc()\fP?
+.sp
+Modern operating systems have a nice feature: they can reserve an address space
+instead of allocating memory. We can safely allocate memory pages inside this
+address space, so the stack could grow without moving memory data (this is
+important because of pointers). Thus we can allocate 1M address space, and use
+only a single memory page (usually 4K) if that is enough. However, we can still
+grow up to 1M anytime if needed.
+.P
+(3) Who "owns" a JIT stack?
+.sp
+The owner of the stack is the user program, not the JIT studied pattern or
+anything else. The user program must ensure that if a stack is used by
+\fBpcre_exec()\fP, (that is, it is assigned to the pattern currently running),
+that stack must not be used by any other threads (to avoid overwriting the same
+memory area). The best practice for multithreaded programs is to allocate a
+stack for each thread, and return this stack through the JIT callback function.
+.P
+(4) When should a JIT stack be freed?
+.sp
+You can free a JIT stack at any time, as long as it will not be used by
+\fBpcre_exec()\fP again. When you assign the stack to a pattern, only a pointer
+is set. There is no reference counting or any other magic. You can free the
+patterns and stacks in any order, anytime. Just \fIdo not\fP call
+\fBpcre_exec()\fP with a pattern pointing to an already freed stack, as that
+will cause SEGFAULT. (Also, do not free a stack currently used by
+\fBpcre_exec()\fP in another thread). You can also replace the stack for a
+pattern at any time. You can even free the previous stack before assigning a
+replacement.
+.P
+(5) Should I allocate/free a stack every time before/after calling
+\fBpcre_exec()\fP?
+.sp
+No, because this is too costly in terms of resources. However, you could
+implement some clever idea which release the stack if it is not used in let's
+say two minutes. The JIT callback can help to achieve this without keeping a
+list of the currently JIT studied patterns.
+.P
+(6) OK, the stack is for long term memory allocation. But what happens if a
+pattern causes stack overflow with a stack of 1M? Is that 1M kept until the
+stack is freed?
+.sp
+Especially on embedded sytems, it might be a good idea to release memory
+sometimes without freeing the stack. There is no API for this at the moment.
+Probably a function call which returns with the currently allocated memory for
+any stack and another which allows releasing memory (shrinking the stack) would
+be a good idea if someone needs this.
+.P
+(7) This is too much of a headache. Isn't there any better solution for JIT
+stack handling?
+.sp
+No, thanks to Windows. If POSIX threads were used everywhere, we could throw
+out this complicated API.
+.
+.
+.SH "EXAMPLE CODE"
+.rs
+.sp
+This is a single-threaded example that specifies a JIT stack without using a
+callback.
+.sp
+  int rc;
+  int ovector[30];
+  pcre *re;
+  pcre_extra *extra;
+  pcre_jit_stack *jit_stack;
+.sp
+  re = pcre_compile(pattern, 0, &error, &erroffset, NULL);
+  /* Check for errors */
+  extra = pcre_study(re, PCRE_STUDY_JIT_COMPILE, &error);
+  jit_stack = pcre_jit_stack_alloc(32*1024, 512*1024);
+  /* Check for error (NULL) */
+  pcre_assign_jit_stack(extra, NULL, jit_stack);
+  rc = pcre_exec(re, extra, subject, length, 0, 0, ovector, 30);
+  /* Check results */
+  pcre_free(re);
+  pcre_free_study(extra);
+  pcre_jit_stack_free(jit_stack);
+.sp
+.
+.
+.SH "JIT FAST PATH API"
+.rs
+.sp
+Because the API described above falls back to interpreted execution when JIT is
+not available, it is convenient for programs that are written for general use
+in many environments. However, calling JIT via \fBpcre_exec()\fP does have a
+performance impact. Programs that are written for use where JIT is known to be
+available, and which need the best possible performance, can instead use a
+"fast path" API to call JIT execution directly instead of calling
+\fBpcre_exec()\fP (obviously only for patterns that have been successfully
+studied by JIT).
+.P
+The fast path function is called \fBpcre_jit_exec()\fP, and it takes exactly
+the same arguments as \fBpcre_exec()\fP, plus one additional argument that
+must point to a JIT stack. The JIT stack arrangements described above do not
+apply. The return values are the same as for \fBpcre_exec()\fP.
+.P
+When you call \fBpcre_exec()\fP, as well as testing for invalid options, a
+number of other sanity checks are performed on the arguments. For example, if
+the subject pointer is NULL, or its length is negative, an immediate error is
+given. Also, unless PCRE_NO_UTF[8|16|32] is set, a UTF subject string is tested
+for validity. In the interests of speed, these checks do not happen on the JIT
+fast path, and if invalid data is passed, the result is undefined.
+.P
+Bypassing the sanity checks and the \fBpcre_exec()\fP wrapping can give
+speedups of more than 10%.
+.
+.
+.SH "SEE ALSO"
+.rs
+.sp
+\fBpcreapi\fP(3)
+.
+.
+.SH AUTHOR
+.rs
+.sp
+.nf
+Philip Hazel (FAQ by Zoltan Herczeg)
+University Computing Service
+Cambridge CB2 3QH, England.
+.fi
+.
+.
+.SH REVISION
+.rs
+.sp
+.nf
+Last updated: 17 March 2013
+Copyright (c) 1997-2013 University of Cambridge.
+.fi
diff --git a/pcre/doc/pcrelimits.3 b/pcre/doc/pcrelimits.3
new file mode 100644
index 00000000000..14ffbc466ed
--- /dev/null
+++ b/pcre/doc/pcrelimits.3
@@ -0,0 +1,67 @@
+.TH PCRELIMITS 3 "24 June 2012" "PCRE 8.30"
+.SH NAME
+PCRE - Perl-compatible regular expressions
+.SH "SIZE AND OTHER LIMITATIONS"
+.rs
+.sp
+There are some size limitations in PCRE but it is hoped that they will never in
+practice be relevant.
+.P
+The maximum length of a compiled pattern is approximately 64K data units (bytes
+for the 8-bit library, 32-bit units for the 32-bit library, and 32-bit units for
+the 32-bit library) if PCRE is compiled with the default internal linkage size
+of 2 bytes. If you want to process regular expressions that are truly enormous,
+you can compile PCRE with an internal linkage size of 3 or 4 (when building the
+16-bit or 32-bit library, 3 is rounded up to 4). See the \fBREADME\fP file in
+the source distribution and the
+.\" HREF
+\fBpcrebuild\fP
+.\"
+documentation for details. In these cases the limit is substantially larger.
+However, the speed of execution is slower.
+.P
+All values in repeating quantifiers must be less than 65536.
+.P
+There is no limit to the number of parenthesized subpatterns, but there can be
+no more than 65535 capturing subpatterns.
+.P
+There is a limit to the number of forward references to subsequent subpatterns
+of around 200,000. Repeated forward references with fixed upper limits, for
+example, (?2){0,100} when subpattern number 2 is to the right, are included in
+the count. There is no limit to the number of backward references.
+.P
+The maximum length of name for a named subpattern is 32 characters, and the
+maximum number of named subpatterns is 10000.
+.P
+The maximum length of a name in a (*MARK), (*PRUNE), (*SKIP), or (*THEN) verb
+is 255 for the 8-bit library and 65535 for the 16-bit and 32-bit library.
+.P
+The maximum length of a subject string is the largest positive number that an
+integer variable can hold. However, when using the traditional matching
+function, PCRE uses recursion to handle subpatterns and indefinite repetition.
+This means that the available stack space may limit the size of a subject
+string that can be processed by certain patterns. For a discussion of stack
+issues, see the
+.\" HREF
+\fBpcrestack\fP
+.\"
+documentation.
+.
+.
+.SH AUTHOR
+.rs
+.sp
+.nf
+Philip Hazel
+University Computing Service
+Cambridge CB2 3QH, England.
+.fi
+.
+.
+.SH REVISION
+.rs
+.sp
+.nf
+Last updated: 04 May 2012
+Copyright (c) 1997-2012 University of Cambridge.
+.fi
diff --git a/pcre/doc/pcrematching.3 b/pcre/doc/pcrematching.3
new file mode 100644
index 00000000000..a9977d5e05a
--- /dev/null
+++ b/pcre/doc/pcrematching.3
@@ -0,0 +1,206 @@
+.TH PCREMATCHING 3 "08 January 2012" "PCRE 8.30"
+.SH NAME
+PCRE - Perl-compatible regular expressions
+.SH "PCRE MATCHING ALGORITHMS"
+.rs
+.sp
+This document describes the two different algorithms that are available in PCRE
+for matching a compiled regular expression against a given subject string. The
+"standard" algorithm is the one provided by the \fBpcre_exec()\fP,
+\fBpcre16_exec()\fP and \fBpcre32_exec()\fP functions. These work in the same
+as as Perl's matching function, and provide a Perl-compatible matching operation.
+The just-in-time (JIT) optimization that is described in the
+.\" HREF
+\fBpcrejit\fP
+.\"
+documentation is compatible with these functions.
+.P
+An alternative algorithm is provided by the \fBpcre_dfa_exec()\fP,
+\fBpcre16_dfa_exec()\fP and \fBpcre32_dfa_exec()\fP functions; they operate in
+a different way, and are not Perl-compatible. This alternative has advantages
+and disadvantages compared with the standard algorithm, and these are described
+below.
+.P
+When there is only one possible way in which a given subject string can match a
+pattern, the two algorithms give the same answer. A difference arises, however,
+when there are multiple possibilities. For example, if the pattern
+.sp
+  ^<.*>
+.sp
+is matched against the string
+.sp
+    
+.sp
+there are three possible answers. The standard algorithm finds only one of
+them, whereas the alternative algorithm finds all three.
+.
+.
+.SH "REGULAR EXPRESSIONS AS TREES"
+.rs
+.sp
+The set of strings that are matched by a regular expression can be represented
+as a tree structure. An unlimited repetition in the pattern makes the tree of
+infinite size, but it is still a tree. Matching the pattern to a given subject
+string (from a given starting point) can be thought of as a search of the tree.
+There are two ways to search a tree: depth-first and breadth-first, and these
+correspond to the two matching algorithms provided by PCRE.
+.
+.
+.SH "THE STANDARD MATCHING ALGORITHM"
+.rs
+.sp
+In the terminology of Jeffrey Friedl's book "Mastering Regular
+Expressions", the standard algorithm is an "NFA algorithm". It conducts a
+depth-first search of the pattern tree. That is, it proceeds along a single
+path through the tree, checking that the subject matches what is required. When
+there is a mismatch, the algorithm tries any alternatives at the current point,
+and if they all fail, it backs up to the previous branch point in the tree, and
+tries the next alternative branch at that level. This often involves backing up
+(moving to the left) in the subject string as well. The order in which
+repetition branches are tried is controlled by the greedy or ungreedy nature of
+the quantifier.
+.P
+If a leaf node is reached, a matching string has been found, and at that point
+the algorithm stops. Thus, if there is more than one possible match, this
+algorithm returns the first one that it finds. Whether this is the shortest,
+the longest, or some intermediate length depends on the way the greedy and
+ungreedy repetition quantifiers are specified in the pattern.
+.P
+Because it ends up with a single path through the tree, it is relatively
+straightforward for this algorithm to keep track of the substrings that are
+matched by portions of the pattern in parentheses. This provides support for
+capturing parentheses and back references.
+.
+.
+.SH "THE ALTERNATIVE MATCHING ALGORITHM"
+.rs
+.sp
+This algorithm conducts a breadth-first search of the tree. Starting from the
+first matching point in the subject, it scans the subject string from left to
+right, once, character by character, and as it does this, it remembers all the
+paths through the tree that represent valid matches. In Friedl's terminology,
+this is a kind of "DFA algorithm", though it is not implemented as a
+traditional finite state machine (it keeps multiple states active
+simultaneously).
+.P
+Although the general principle of this matching algorithm is that it scans the
+subject string only once, without backtracking, there is one exception: when a
+lookaround assertion is encountered, the characters following or preceding the
+current point have to be independently inspected.
+.P
+The scan continues until either the end of the subject is reached, or there are
+no more unterminated paths. At this point, terminated paths represent the
+different matching possibilities (if there are none, the match has failed).
+Thus, if there is more than one possible match, this algorithm finds all of
+them, and in particular, it finds the longest. The matches are returned in
+decreasing order of length. There is an option to stop the algorithm after the
+first match (which is necessarily the shortest) is found.
+.P
+Note that all the matches that are found start at the same point in the
+subject. If the pattern
+.sp
+  cat(er(pillar)?)?
+.sp
+is matched against the string "the caterpillar catchment", the result will be
+the three strings "caterpillar", "cater", and "cat" that start at the fifth
+character of the subject. The algorithm does not automatically move on to find
+matches that start at later positions.
+.P
+There are a number of features of PCRE regular expressions that are not
+supported by the alternative matching algorithm. They are as follows:
+.P
+1. Because the algorithm finds all possible matches, the greedy or ungreedy
+nature of repetition quantifiers is not relevant. Greedy and ungreedy
+quantifiers are treated in exactly the same way. However, possessive
+quantifiers can make a difference when what follows could also match what is
+quantified, for example in a pattern like this:
+.sp
+  ^a++\ew!
+.sp
+This pattern matches "aaab!" but not "aaa!", which would be matched by a
+non-possessive quantifier. Similarly, if an atomic group is present, it is
+matched as if it were a standalone pattern at the current point, and the
+longest match is then "locked in" for the rest of the overall pattern.
+.P
+2. When dealing with multiple paths through the tree simultaneously, it is not
+straightforward to keep track of captured substrings for the different matching
+possibilities, and PCRE's implementation of this algorithm does not attempt to
+do this. This means that no captured substrings are available.
+.P
+3. Because no substrings are captured, back references within the pattern are
+not supported, and cause errors if encountered.
+.P
+4. For the same reason, conditional expressions that use a backreference as the
+condition or test for a specific group recursion are not supported.
+.P
+5. Because many paths through the tree may be active, the \eK escape sequence,
+which resets the start of the match when encountered (but may be on some paths
+and not on others), is not supported. It causes an error if encountered.
+.P
+6. Callouts are supported, but the value of the \fIcapture_top\fP field is
+always 1, and the value of the \fIcapture_last\fP field is always -1.
+.P
+7. The \eC escape sequence, which (in the standard algorithm) always matches a
+single data unit, even in UTF-8, UTF-16 or UTF-32 modes, is not supported in
+these modes, because the alternative algorithm moves through the subject string
+one character (not data unit) at a time, for all active paths through the tree.
+.P
+8. Except for (*FAIL), the backtracking control verbs such as (*PRUNE) are not
+supported. (*FAIL) is supported, and behaves like a failing negative assertion.
+.
+.
+.SH "ADVANTAGES OF THE ALTERNATIVE ALGORITHM"
+.rs
+.sp
+Using the alternative matching algorithm provides the following advantages:
+.P
+1. All possible matches (at a single point in the subject) are automatically
+found, and in particular, the longest match is found. To find more than one
+match using the standard algorithm, you have to do kludgy things with
+callouts.
+.P
+2. Because the alternative algorithm scans the subject string just once, and
+never needs to backtrack (except for lookbehinds), it is possible to pass very
+long subject strings to the matching function in several pieces, checking for
+partial matching each time. Although it is possible to do multi-segment
+matching using the standard algorithm by retaining partially matched
+substrings, it is more complicated. The
+.\" HREF
+\fBpcrepartial\fP
+.\"
+documentation gives details of partial matching and discusses multi-segment
+matching.
+.
+.
+.SH "DISADVANTAGES OF THE ALTERNATIVE ALGORITHM"
+.rs
+.sp
+The alternative algorithm suffers from a number of disadvantages:
+.P
+1. It is substantially slower than the standard algorithm. This is partly
+because it has to search for all possible matches, but is also because it is
+less susceptible to optimization.
+.P
+2. Capturing parentheses and back references are not supported.
+.P
+3. Although atomic groups are supported, their use does not provide the
+performance advantage that it does for the standard algorithm.
+.
+.
+.SH AUTHOR
+.rs
+.sp
+.nf
+Philip Hazel
+University Computing Service
+Cambridge CB2 3QH, England.
+.fi
+.
+.
+.SH REVISION
+.rs
+.sp
+.nf
+Last updated: 08 January 2012
+Copyright (c) 1997-2012 University of Cambridge.
+.fi
diff --git a/pcre/doc/pcrepartial.3 b/pcre/doc/pcrepartial.3
new file mode 100644
index 00000000000..4041838c3e9
--- /dev/null
+++ b/pcre/doc/pcrepartial.3
@@ -0,0 +1,467 @@
+.TH PCREPARTIAL 3 "20 February 2013" "PCRE 8.33"
+.SH NAME
+PCRE - Perl-compatible regular expressions
+.SH "PARTIAL MATCHING IN PCRE"
+.rs
+.sp
+In normal use of PCRE, if the subject string that is passed to a matching
+function matches as far as it goes, but is too short to match the entire
+pattern, PCRE_ERROR_NOMATCH is returned. There are circumstances where it might
+be helpful to distinguish this case from other cases in which there is no
+match.
+.P
+Consider, for example, an application where a human is required to type in data
+for a field with specific formatting requirements. An example might be a date
+in the form \fIddmmmyy\fP, defined by this pattern:
+.sp
+  ^\ed?\ed(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\ed\ed$
+.sp
+If the application sees the user's keystrokes one by one, and can check that
+what has been typed so far is potentially valid, it is able to raise an error
+as soon as a mistake is made, by beeping and not reflecting the character that
+has been typed, for example. This immediate feedback is likely to be a better
+user interface than a check that is delayed until the entire string has been
+entered. Partial matching can also be useful when the subject string is very
+long and is not all available at once.
+.P
+PCRE supports partial matching by means of the PCRE_PARTIAL_SOFT and
+PCRE_PARTIAL_HARD options, which can be set when calling any of the matching
+functions. For backwards compatibility, PCRE_PARTIAL is a synonym for
+PCRE_PARTIAL_SOFT. The essential difference between the two options is whether
+or not a partial match is preferred to an alternative complete match, though
+the details differ between the two types of matching function. If both options
+are set, PCRE_PARTIAL_HARD takes precedence.
+.P
+If you want to use partial matching with just-in-time optimized code, you must
+call \fBpcre_study()\fP, \fBpcre16_study()\fP or  \fBpcre32_study()\fP with one
+or both of these options:
+.sp
+  PCRE_STUDY_JIT_PARTIAL_SOFT_COMPILE
+  PCRE_STUDY_JIT_PARTIAL_HARD_COMPILE
+.sp
+PCRE_STUDY_JIT_COMPILE should also be set if you are going to run non-partial
+matches on the same pattern. If the appropriate JIT study mode has not been set
+for a match, the interpretive matching code is used.
+.P
+Setting a partial matching option disables two of PCRE's standard
+optimizations. PCRE remembers the last literal data unit in a pattern, and
+abandons matching immediately if it is not present in the subject string. This
+optimization cannot be used for a subject string that might match only
+partially. If the pattern was studied, PCRE knows the minimum length of a
+matching string, and does not bother to run the matching function on shorter
+strings. This optimization is also disabled for partial matching.
+.
+.
+.SH "PARTIAL MATCHING USING pcre_exec() OR pcre[16|32]_exec()"
+.rs
+.sp
+A partial match occurs during a call to \fBpcre_exec()\fP or
+\fBpcre[16|32]_exec()\fP when the end of the subject string is reached
+successfully, but matching cannot continue because more characters are needed.
+However, at least one character in the subject must have been inspected. This
+character need not form part of the final matched string; lookbehind assertions
+and the \eK escape sequence provide ways of inspecting characters before the
+start of a matched substring. The requirement for inspecting at least one
+character exists because an empty string can always be matched; without such a
+restriction there would always be a partial match of an empty string at the end
+of the subject.
+.P
+If there are at least two slots in the offsets vector when a partial match is
+returned, the first slot is set to the offset of the earliest character that
+was inspected. For convenience, the second offset points to the end of the
+subject so that a substring can easily be identified. If there are at least
+three slots in the offsets vector, the third slot is set to the offset of the
+character where matching started.
+.P
+For the majority of patterns, the contents of the first and third slots will be
+the same. However, for patterns that contain lookbehind assertions, or begin
+with \eb or \eB, characters before the one where matching started may have been
+inspected while carrying out the match. For example, consider this pattern:
+.sp
+  /(?<=abc)123/
+.sp
+This pattern matches "123", but only if it is preceded by "abc". If the subject
+string is "xyzabc12", the first two offsets after a partial match are for the
+substring "abc12", because all these characters were inspected. However, the
+third offset is set to 6, because that is the offset where matching began.
+.P
+What happens when a partial match is identified depends on which of the two
+partial matching options are set.
+.
+.
+.SS "PCRE_PARTIAL_SOFT WITH pcre_exec() OR pcre[16|32]_exec()"
+.rs
+.sp
+If PCRE_PARTIAL_SOFT is set when \fBpcre_exec()\fP or \fBpcre[16|32]_exec()\fP
+identifies a partial match, the partial match is remembered, but matching
+continues as normal, and other alternatives in the pattern are tried. If no
+complete match can be found, PCRE_ERROR_PARTIAL is returned instead of
+PCRE_ERROR_NOMATCH.
+.P
+This option is "soft" because it prefers a complete match over a partial match.
+All the various matching items in a pattern behave as if the subject string is
+potentially complete. For example, \ez, \eZ, and $ match at the end of the
+subject, as normal, and for \eb and \eB the end of the subject is treated as a
+non-alphanumeric.
+.P
+If there is more than one partial match, the first one that was found provides
+the data that is returned. Consider this pattern:
+.sp
+  /123\ew+X|dogY/
+.sp
+If this is matched against the subject string "abc123dog", both
+alternatives fail to match, but the end of the subject is reached during
+matching, so PCRE_ERROR_PARTIAL is returned. The offsets are set to 3 and 9,
+identifying "123dog" as the first partial match that was found. (In this
+example, there are two partial matches, because "dog" on its own partially
+matches the second alternative.)
+.
+.
+.SS "PCRE_PARTIAL_HARD WITH pcre_exec() OR pcre[16|32]_exec()"
+.rs
+.sp
+If PCRE_PARTIAL_HARD is set for \fBpcre_exec()\fP or \fBpcre[16|32]_exec()\fP,
+PCRE_ERROR_PARTIAL is returned as soon as a partial match is found, without
+continuing to search for possible complete matches. This option is "hard"
+because it prefers an earlier partial match over a later complete match. For
+this reason, the assumption is made that the end of the supplied subject string
+may not be the true end of the available data, and so, if \ez, \eZ, \eb, \eB,
+or $ are encountered at the end of the subject, the result is
+PCRE_ERROR_PARTIAL, provided that at least one character in the subject has
+been inspected.
+.P
+Setting PCRE_PARTIAL_HARD also affects the way UTF-8 and UTF-16
+subject strings are checked for validity. Normally, an invalid sequence
+causes the error PCRE_ERROR_BADUTF8 or PCRE_ERROR_BADUTF16. However, in the
+special case of a truncated character at the end of the subject,
+PCRE_ERROR_SHORTUTF8 or PCRE_ERROR_SHORTUTF16 is returned when
+PCRE_PARTIAL_HARD is set.
+.
+.
+.SS "Comparing hard and soft partial matching"
+.rs
+.sp
+The difference between the two partial matching options can be illustrated by a
+pattern such as:
+.sp
+  /dog(sbody)?/
+.sp
+This matches either "dog" or "dogsbody", greedily (that is, it prefers the
+longer string if possible). If it is matched against the string "dog" with
+PCRE_PARTIAL_SOFT, it yields a complete match for "dog". However, if
+PCRE_PARTIAL_HARD is set, the result is PCRE_ERROR_PARTIAL. On the other hand,
+if the pattern is made ungreedy the result is different:
+.sp
+  /dog(sbody)??/
+.sp
+In this case the result is always a complete match because that is found first,
+and matching never continues after finding a complete match. It might be easier
+to follow this explanation by thinking of the two patterns like this:
+.sp
+  /dog(sbody)?/    is the same as  /dogsbody|dog/
+  /dog(sbody)??/   is the same as  /dog|dogsbody/
+.sp
+The second pattern will never match "dogsbody", because it will always find the
+shorter match first.
+.
+.
+.SH "PARTIAL MATCHING USING pcre_dfa_exec() OR pcre[16|32]_dfa_exec()"
+.rs
+.sp
+The DFA functions move along the subject string character by character, without
+backtracking, searching for all possible matches simultaneously. If the end of
+the subject is reached before the end of the pattern, there is the possibility
+of a partial match, again provided that at least one character has been
+inspected.
+.P
+When PCRE_PARTIAL_SOFT is set, PCRE_ERROR_PARTIAL is returned only if there
+have been no complete matches. Otherwise, the complete matches are returned.
+However, if PCRE_PARTIAL_HARD is set, a partial match takes precedence over any
+complete matches. The portion of the string that was inspected when the longest
+partial match was found is set as the first matching string, provided there are
+at least two slots in the offsets vector.
+.P
+Because the DFA functions always search for all possible matches, and there is
+no difference between greedy and ungreedy repetition, their behaviour is
+different from the standard functions when PCRE_PARTIAL_HARD is set. Consider
+the string "dog" matched against the ungreedy pattern shown above:
+.sp
+  /dog(sbody)??/
+.sp
+Whereas the standard functions stop as soon as they find the complete match for
+"dog", the DFA functions also find the partial match for "dogsbody", and so
+return that when PCRE_PARTIAL_HARD is set.
+.
+.
+.SH "PARTIAL MATCHING AND WORD BOUNDARIES"
+.rs
+.sp
+If a pattern ends with one of sequences \eb or \eB, which test for word
+boundaries, partial matching with PCRE_PARTIAL_SOFT can give counter-intuitive
+results. Consider this pattern:
+.sp
+  /\ebcat\eb/
+.sp
+This matches "cat", provided there is a word boundary at either end. If the
+subject string is "the cat", the comparison of the final "t" with a following
+character cannot take place, so a partial match is found. However, normal
+matching carries on, and \eb matches at the end of the subject when the last
+character is a letter, so a complete match is found. The result, therefore, is
+\fInot\fP PCRE_ERROR_PARTIAL. Using PCRE_PARTIAL_HARD in this case does yield
+PCRE_ERROR_PARTIAL, because then the partial match takes precedence.
+.
+.
+.SH "FORMERLY RESTRICTED PATTERNS"
+.rs
+.sp
+For releases of PCRE prior to 8.00, because of the way certain internal
+optimizations were implemented in the \fBpcre_exec()\fP function, the
+PCRE_PARTIAL option (predecessor of PCRE_PARTIAL_SOFT) could not be used with
+all patterns. From release 8.00 onwards, the restrictions no longer apply, and
+partial matching with can be requested for any pattern.
+.P
+Items that were formerly restricted were repeated single characters and
+repeated metasequences. If PCRE_PARTIAL was set for a pattern that did not
+conform to the restrictions, \fBpcre_exec()\fP returned the error code
+PCRE_ERROR_BADPARTIAL (-13). This error code is no longer in use. The
+PCRE_INFO_OKPARTIAL call to \fBpcre_fullinfo()\fP to find out if a compiled
+pattern can be used for partial matching now always returns 1.
+.
+.
+.SH "EXAMPLE OF PARTIAL MATCHING USING PCRETEST"
+.rs
+.sp
+If the escape sequence \eP is present in a \fBpcretest\fP data line, the
+PCRE_PARTIAL_SOFT option is used for the match. Here is a run of \fBpcretest\fP
+that uses the date example quoted above:
+.sp
+    re> /^\ed?\ed(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\ed\ed$/
+  data> 25jun04\eP
+   0: 25jun04
+   1: jun
+  data> 25dec3\eP
+  Partial match: 23dec3
+  data> 3ju\eP
+  Partial match: 3ju
+  data> 3juj\eP
+  No match
+  data> j\eP
+  No match
+.sp
+The first data string is matched completely, so \fBpcretest\fP shows the
+matched substrings. The remaining four strings do not match the complete
+pattern, but the first two are partial matches. Similar output is obtained
+if DFA matching is used.
+.P
+If the escape sequence \eP is present more than once in a \fBpcretest\fP data
+line, the PCRE_PARTIAL_HARD option is set for the match.
+.
+.
+.SH "MULTI-SEGMENT MATCHING WITH pcre_dfa_exec() OR pcre[16|32]_dfa_exec()"
+.rs
+.sp
+When a partial match has been found using a DFA matching function, it is
+possible to continue the match by providing additional subject data and calling
+the function again with the same compiled regular expression, this time setting
+the PCRE_DFA_RESTART option. You must pass the same working space as before,
+because this is where details of the previous partial match are stored. Here is
+an example using \fBpcretest\fP, using the \eR escape sequence to set the
+PCRE_DFA_RESTART option (\eD specifies the use of the DFA matching function):
+.sp
+    re> /^\ed?\ed(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\ed\ed$/
+  data> 23ja\eP\eD
+  Partial match: 23ja
+  data> n05\eR\eD
+   0: n05
+.sp
+The first call has "23ja" as the subject, and requests partial matching; the
+second call has "n05" as the subject for the continued (restarted) match.
+Notice that when the match is complete, only the last part is shown; PCRE does
+not retain the previously partially-matched string. It is up to the calling
+program to do that if it needs to.
+.P
+You can set the PCRE_PARTIAL_SOFT or PCRE_PARTIAL_HARD options with
+PCRE_DFA_RESTART to continue partial matching over multiple segments. This
+facility can be used to pass very long subject strings to the DFA matching
+functions.
+.
+.
+.SH "MULTI-SEGMENT MATCHING WITH pcre_exec() OR pcre[16|32]_exec()"
+.rs
+.sp
+From release 8.00, the standard matching functions can also be used to do
+multi-segment matching. Unlike the DFA functions, it is not possible to
+restart the previous match with a new segment of data. Instead, new data must
+be added to the previous subject string, and the entire match re-run, starting
+from the point where the partial match occurred. Earlier data can be discarded.
+.P
+It is best to use PCRE_PARTIAL_HARD in this situation, because it does not
+treat the end of a segment as the end of the subject when matching \ez, \eZ,
+\eb, \eB, and $. Consider an unanchored pattern that matches dates:
+.sp
+    re> /\ed?\ed(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\ed\ed/
+  data> The date is 23ja\eP\eP
+  Partial match: 23ja
+.sp
+At this stage, an application could discard the text preceding "23ja", add on
+text from the next segment, and call the matching function again. Unlike the
+DFA matching functions, the entire matching string must always be available,
+and the complete matching process occurs for each call, so more memory and more
+processing time is needed.
+.P
+\fBNote:\fP If the pattern contains lookbehind assertions, or \eK, or starts
+with \eb or \eB, the string that is returned for a partial match includes
+characters that precede the start of what would be returned for a complete
+match, because it contains all the characters that were inspected during the
+partial match.
+.
+.
+.SH "ISSUES WITH MULTI-SEGMENT MATCHING"
+.rs
+.sp
+Certain types of pattern may give problems with multi-segment matching,
+whichever matching function is used.
+.P
+1. If the pattern contains a test for the beginning of a line, you need to pass
+the PCRE_NOTBOL option when the subject string for any call does start at the
+beginning of a line. There is also a PCRE_NOTEOL option, but in practice when
+doing multi-segment matching you should be using PCRE_PARTIAL_HARD, which
+includes the effect of PCRE_NOTEOL.
+.P
+2. Lookbehind assertions that have already been obeyed are catered for in the
+offsets that are returned for a partial match. However a lookbehind assertion
+later in the pattern could require even earlier characters to be inspected. You
+can handle this case by using the PCRE_INFO_MAXLOOKBEHIND option of the
+\fBpcre_fullinfo()\fP or \fBpcre[16|32]_fullinfo()\fP functions to obtain the
+length of the longest lookbehind in the pattern. This length is given in
+characters, not bytes. If you always retain at least that many characters
+before the partially matched string, all should be well. (Of course, near the
+start of the subject, fewer characters may be present; in that case all
+characters should be retained.)
+.P
+From release 8.33, there is a more accurate way of deciding which characters to
+retain. Instead of subtracting the length of the longest lookbehind from the
+earliest inspected character (\fIoffsets[0]\fP), the match start position
+(\fIoffsets[2]\fP) should be used, and the next match attempt started at the
+\fIoffsets[2]\fP character by setting the \fIstartoffset\fP argument of
+\fBpcre_exec()\fP or \fBpcre_dfa_exec()\fP.
+.P
+For example, if the pattern "(?<=123)abc" is partially
+matched against the string "xx123a", the three offset values returned are 2, 6,
+and 5. This indicates that the matching process that gave a partial match
+started at offset 5, but the characters "123a" were all inspected. The maximum
+lookbehind for that pattern is 3, so taking that away from 5 shows that we need
+only keep "123a", and the next match attempt can be started at offset 3 (that
+is, at "a") when further characters have been added. When the match start is
+not the earliest inspected character, \fBpcretest\fP shows it explicitly:
+.sp
+    re> "(?<=123)abc"
+  data> xx123a\eP\eP
+  Partial match at offset 5: 123a
+.P
+3. Because a partial match must always contain at least one character, what
+might be considered a partial match of an empty string actually gives a "no
+match" result. For example:
+.sp
+    re> /c(?<=abc)x/
+  data> ab\eP
+  No match
+.sp
+If the next segment begins "cx", a match should be found, but this will only
+happen if characters from the previous segment are retained. For this reason, a
+"no match" result should be interpreted as "partial match of an empty string"
+when the pattern contains lookbehinds.
+.P
+4. Matching a subject string that is split into multiple segments may not
+always produce exactly the same result as matching over one single long string,
+especially when PCRE_PARTIAL_SOFT is used. The section "Partial Matching and
+Word Boundaries" above describes an issue that arises if the pattern ends with
+\eb or \eB. Another kind of difference may occur when there are multiple
+matching possibilities, because (for PCRE_PARTIAL_SOFT) a partial match result
+is given only when there are no completed matches. This means that as soon as
+the shortest match has been found, continuation to a new subject segment is no
+longer possible. Consider again this \fBpcretest\fP example:
+.sp
+    re> /dog(sbody)?/
+  data> dogsb\eP
+   0: dog
+  data> do\eP\eD
+  Partial match: do
+  data> gsb\eR\eP\eD
+   0: g
+  data> dogsbody\eD
+   0: dogsbody
+   1: dog
+.sp
+The first data line passes the string "dogsb" to a standard matching function,
+setting the PCRE_PARTIAL_SOFT option. Although the string is a partial match
+for "dogsbody", the result is not PCRE_ERROR_PARTIAL, because the shorter
+string "dog" is a complete match. Similarly, when the subject is presented to
+a DFA matching function in several parts ("do" and "gsb" being the first two)
+the match stops when "dog" has been found, and it is not possible to continue.
+On the other hand, if "dogsbody" is presented as a single string, a DFA
+matching function finds both matches.
+.P
+Because of these problems, it is best to use PCRE_PARTIAL_HARD when matching
+multi-segment data. The example above then behaves differently:
+.sp
+    re> /dog(sbody)?/
+  data> dogsb\eP\eP
+  Partial match: dogsb
+  data> do\eP\eD
+  Partial match: do
+  data> gsb\eR\eP\eP\eD
+  Partial match: gsb
+.sp
+5. Patterns that contain alternatives at the top level which do not all start
+with the same pattern item may not work as expected when PCRE_DFA_RESTART is
+used. For example, consider this pattern:
+.sp
+  1234|3789
+.sp
+If the first part of the subject is "ABC123", a partial match of the first
+alternative is found at offset 3. There is no partial match for the second
+alternative, because such a match does not start at the same point in the
+subject string. Attempting to continue with the string "7890" does not yield a
+match because only those alternatives that match at one point in the subject
+are remembered. The problem arises because the start of the second alternative
+matches within the first alternative. There is no problem with anchored
+patterns or patterns such as:
+.sp
+  1234|ABCD
+.sp
+where no string can be a partial match for both alternatives. This is not a
+problem if a standard matching function is used, because the entire match has
+to be rerun each time:
+.sp
+    re> /1234|3789/
+  data> ABC123\eP\eP
+  Partial match: 123
+  data> 1237890
+   0: 3789
+.sp
+Of course, instead of using PCRE_DFA_RESTART, the same technique of re-running
+the entire match can also be used with the DFA matching functions. Another
+possibility is to work with two buffers. If a partial match at offset \fIn\fP
+in the first buffer is followed by "no match" when PCRE_DFA_RESTART is used on
+the second buffer, you can then try a new match starting at offset \fIn+1\fP in
+the first buffer.
+.
+.
+.SH AUTHOR
+.rs
+.sp
+.nf
+Philip Hazel
+University Computing Service
+Cambridge CB2 3QH, England.
+.fi
+.
+.
+.SH REVISION
+.rs
+.sp
+.nf
+Last updated: 20 February 2013
+Copyright (c) 1997-2013 University of Cambridge.
+.fi
diff --git a/pcre/doc/pcrepattern.3 b/pcre/doc/pcrepattern.3
new file mode 100644
index 00000000000..39711014996
--- /dev/null
+++ b/pcre/doc/pcrepattern.3
@@ -0,0 +1,3150 @@
+.TH PCREPATTERN 3 "26 April 2013" "PCRE 8.33"
+.SH NAME
+PCRE - Perl-compatible regular expressions
+.SH "PCRE REGULAR EXPRESSION DETAILS"
+.rs
+.sp
+The syntax and semantics of the regular expressions that are supported by PCRE
+are described in detail below. There is a quick-reference syntax summary in the
+.\" HREF
+\fBpcresyntax\fP
+.\"
+page. PCRE tries to match Perl syntax and semantics as closely as it can. PCRE
+also supports some alternative regular expression syntax (which does not
+conflict with the Perl syntax) in order to provide some compatibility with
+regular expressions in Python, .NET, and Oniguruma.
+.P
+Perl's regular expressions are described in its own documentation, and
+regular expressions in general are covered in a number of books, some of which
+have copious examples. Jeffrey Friedl's "Mastering Regular Expressions",
+published by O'Reilly, covers regular expressions in great detail. This
+description of PCRE's regular expressions is intended as reference material.
+.P
+This document discusses the patterns that are supported by PCRE when one its
+main matching functions, \fBpcre_exec()\fP (8-bit) or \fBpcre[16|32]_exec()\fP
+(16- or 32-bit), is used. PCRE also has alternative matching functions,
+\fBpcre_dfa_exec()\fP and \fBpcre[16|32_dfa_exec()\fP, which match using a
+different algorithm that is not Perl-compatible. Some of the features discussed
+below are not available when DFA matching is used. The advantages and
+disadvantages of the alternative functions, and how they differ from the normal
+functions, are discussed in the
+.\" HREF
+\fBpcrematching\fP
+.\"
+page.
+.
+.
+.SH "SPECIAL START-OF-PATTERN ITEMS"
+.rs
+.sp
+A number of options that can be passed to \fBpcre_compile()\fP can also be set
+by special items at the start of a pattern. These are not Perl-compatible, but
+are provided to make these options accessible to pattern writers who are not
+able to change the program that processes the pattern. Any number of these
+items may appear, but they must all be together right at the start of the
+pattern string, and the letters must be in upper case.
+.
+.
+.SS "UTF support"
+.rs
+.sp
+The original operation of PCRE was on strings of one-byte characters. However,
+there is now also support for UTF-8 strings in the original library, an
+extra library that supports 16-bit and UTF-16 character strings, and a
+third library that supports 32-bit and UTF-32 character strings. To use these
+features, PCRE must be built to include appropriate support. When using UTF
+strings you must either call the compiling function with the PCRE_UTF8,
+PCRE_UTF16, or PCRE_UTF32 option, or the pattern must start with one of
+these special sequences:
+.sp
+  (*UTF8)
+  (*UTF16)
+  (*UTF32)
+  (*UTF)
+.sp
+(*UTF) is a generic sequence that can be used with any of the libraries.
+Starting a pattern with such a sequence is equivalent to setting the relevant
+option. How setting a UTF mode affects pattern matching is mentioned in several
+places below. There is also a summary of features in the
+.\" HREF
+\fBpcreunicode\fP
+.\"
+page.
+.P
+Some applications that allow their users to supply patterns may wish to
+restrict them to non-UTF data for security reasons. If the PCRE_NEVER_UTF
+option is set at compile time, (*UTF) etc. are not allowed, and their
+appearance causes an error.
+.
+.
+.SS "Unicode property support"
+.rs
+.sp
+Another special sequence that may appear at the start of a pattern is
+.sp
+  (*UCP)
+.sp
+This has the same effect as setting the PCRE_UCP option: it causes sequences
+such as \ed and \ew to use Unicode properties to determine character types,
+instead of recognizing only characters with codes less than 128 via a lookup
+table.
+.
+.
+.SS "Disabling start-up optimizations"
+.rs
+.sp
+If a pattern starts with (*NO_START_OPT), it has the same effect as setting the
+PCRE_NO_START_OPTIMIZE option either at compile or matching time.
+.
+.
+.\" HTML 
+.SS "Newline conventions"
+.rs
+.sp
+PCRE supports five different conventions for indicating line breaks in
+strings: a single CR (carriage return) character, a single LF (linefeed)
+character, the two-character sequence CRLF, any of the three preceding, or any
+Unicode newline sequence. The
+.\" HREF
+\fBpcreapi\fP
+.\"
+page has
+.\" HTML 
+.\" 
+further discussion
+.\"
+about newlines, and shows how to set the newline convention in the
+\fIoptions\fP arguments for the compiling and matching functions.
+.P
+It is also possible to specify a newline convention by starting a pattern
+string with one of the following five sequences:
+.sp
+  (*CR)        carriage return
+  (*LF)        linefeed
+  (*CRLF)      carriage return, followed by linefeed
+  (*ANYCRLF)   any of the three above
+  (*ANY)       all Unicode newline sequences
+.sp
+These override the default and the options given to the compiling function. For
+example, on a Unix system where LF is the default newline sequence, the pattern
+.sp
+  (*CR)a.b
+.sp
+changes the convention to CR. That pattern matches "a\enb" because LF is no
+longer a newline. If more than one of these settings is present, the last one
+is used.
+.P
+The newline convention affects where the circumflex and dollar assertions are
+true. It also affects the interpretation of the dot metacharacter when
+PCRE_DOTALL is not set, and the behaviour of \eN. However, it does not affect
+what the \eR escape sequence matches. By default, this is any Unicode newline
+sequence, for Perl compatibility. However, this can be changed; see the
+description of \eR in the section entitled
+.\" HTML 
+.\" 
+"Newline sequences"
+.\"
+below. A change of \eR setting can be combined with a change of newline
+convention.
+.
+.
+.SS "Setting match and recursion limits"
+.rs
+.sp
+The caller of \fBpcre_exec()\fP can set a limit on the number of times the
+internal \fBmatch()\fP function is called and on the maximum depth of
+recursive calls. These facilities are provided to catch runaway matches that
+are provoked by patterns with huge matching trees (a typical example is a
+pattern with nested unlimited repeats) and to avoid running out of system stack
+by too much recursion. When one of these limits is reached, \fBpcre_exec()\fP
+gives an error return. The limits can also be set by items at the start of the
+pattern of the form
+.sp
+  (*LIMIT_MATCH=d)
+  (*LIMIT_RECURSION=d)
+.sp
+where d is any number of decimal digits. However, the value of the setting must
+be less than the value set by the caller of \fBpcre_exec()\fP for it to have
+any effect. In other words, the pattern writer can lower the limit set by the
+programmer, but not raise it. If there is more than one setting of one of these
+limits, the lower value is used.
+.
+.
+.SH "EBCDIC CHARACTER CODES"
+.rs
+.sp
+PCRE can be compiled to run in an environment that uses EBCDIC as its character
+code rather than ASCII or Unicode (typically a mainframe system). In the
+sections below, character code values are ASCII or Unicode; in an EBCDIC
+environment these characters may have different code values, and there are no
+code points greater than 255.
+.
+.
+.SH "CHARACTERS AND METACHARACTERS"
+.rs
+.sp
+A regular expression is a pattern that is matched against a subject string from
+left to right. Most characters stand for themselves in a pattern, and match the
+corresponding characters in the subject. As a trivial example, the pattern
+.sp
+  The quick brown fox
+.sp
+matches a portion of a subject string that is identical to itself. When
+caseless matching is specified (the PCRE_CASELESS option), letters are matched
+independently of case. In a UTF mode, PCRE always understands the concept of
+case for characters whose values are less than 128, so caseless matching is
+always possible. For characters with higher values, the concept of case is
+supported if PCRE is compiled with Unicode property support, but not otherwise.
+If you want to use caseless matching for characters 128 and above, you must
+ensure that PCRE is compiled with Unicode property support as well as with
+UTF support.
+.P
+The power of regular expressions comes from the ability to include alternatives
+and repetitions in the pattern. These are encoded in the pattern by the use of
+\fImetacharacters\fP, which do not stand for themselves but instead are
+interpreted in some special way.
+.P
+There are two different sets of metacharacters: those that are recognized
+anywhere in the pattern except within square brackets, and those that are
+recognized within square brackets. Outside square brackets, the metacharacters
+are as follows:
+.sp
+  \e      general escape character with several uses
+  ^      assert start of string (or line, in multiline mode)
+  $      assert end of string (or line, in multiline mode)
+  .      match any character except newline (by default)
+  [      start character class definition
+  |      start of alternative branch
+  (      start subpattern
+  )      end subpattern
+  ?      extends the meaning of (
+         also 0 or 1 quantifier
+         also quantifier minimizer
+  *      0 or more quantifier
+  +      1 or more quantifier
+         also "possessive quantifier"
+  {      start min/max quantifier
+.sp
+Part of a pattern that is in square brackets is called a "character class". In
+a character class the only metacharacters are:
+.sp
+  \e      general escape character
+  ^      negate the class, but only if the first character
+  -      indicates character range
+.\" JOIN
+  [      POSIX character class (only if followed by POSIX
+           syntax)
+  ]      terminates the character class
+.sp
+The following sections describe the use of each of the metacharacters.
+.
+.
+.SH BACKSLASH
+.rs
+.sp
+The backslash character has several uses. Firstly, if it is followed by a
+character that is not a number or a letter, it takes away any special meaning
+that character may have. This use of backslash as an escape character applies
+both inside and outside character classes.
+.P
+For example, if you want to match a * character, you write \e* in the pattern.
+This escaping action applies whether or not the following character would
+otherwise be interpreted as a metacharacter, so it is always safe to precede a
+non-alphanumeric with backslash to specify that it stands for itself. In
+particular, if you want to match a backslash, you write \e\e.
+.P
+In a UTF mode, only ASCII numbers and letters have any special meaning after a
+backslash. All other characters (in particular, those whose codepoints are
+greater than 127) are treated as literals.
+.P
+If a pattern is compiled with the PCRE_EXTENDED option, white space in the
+pattern (other than in a character class) and characters between a # outside
+a character class and the next newline are ignored. An escaping backslash can
+be used to include a white space or # character as part of the pattern.
+.P
+If you want to remove the special meaning from a sequence of characters, you
+can do so by putting them between \eQ and \eE. This is different from Perl in
+that $ and @ are handled as literals in \eQ...\eE sequences in PCRE, whereas in
+Perl, $ and @ cause variable interpolation. Note the following examples:
+.sp
+  Pattern            PCRE matches   Perl matches
+.sp
+.\" JOIN
+  \eQabc$xyz\eE        abc$xyz        abc followed by the
+                                      contents of $xyz
+  \eQabc\e$xyz\eE       abc\e$xyz       abc\e$xyz
+  \eQabc\eE\e$\eQxyz\eE   abc$xyz        abc$xyz
+.sp
+The \eQ...\eE sequence is recognized both inside and outside character classes.
+An isolated \eE that is not preceded by \eQ is ignored. If \eQ is not followed
+by \eE later in the pattern, the literal interpretation continues to the end of
+the pattern (that is, \eE is assumed at the end). If the isolated \eQ is inside
+a character class, this causes an error, because the character class is not
+terminated.
+.
+.
+.\" HTML 
+.SS "Non-printing characters"
+.rs
+.sp
+A second use of backslash provides a way of encoding non-printing characters
+in patterns in a visible manner. There is no restriction on the appearance of
+non-printing characters, apart from the binary zero that terminates a pattern,
+but when a pattern is being prepared by text editing, it is often easier to use
+one of the following escape sequences than the binary character it represents:
+.sp
+  \ea        alarm, that is, the BEL character (hex 07)
+  \ecx       "control-x", where x is any ASCII character
+  \ee        escape (hex 1B)
+  \ef        form feed (hex 0C)
+  \en        linefeed (hex 0A)
+  \er        carriage return (hex 0D)
+  \et        tab (hex 09)
+  \eddd      character with octal code ddd, or back reference
+  \exhh      character with hex code hh
+  \ex{hhh..} character with hex code hhh.. (non-JavaScript mode)
+  \euhhhh    character with hex code hhhh (JavaScript mode only)
+.sp
+The precise effect of \ecx on ASCII characters is as follows: if x is a lower
+case letter, it is converted to upper case. Then bit 6 of the character (hex
+40) is inverted. Thus \ecA to \ecZ become hex 01 to hex 1A (A is 41, Z is 5A),
+but \ec{ becomes hex 3B ({ is 7B), and \ec; becomes hex 7B (; is 3B). If the
+data item (byte or 16-bit value) following \ec has a value greater than 127, a
+compile-time error occurs. This locks out non-ASCII characters in all modes.
+.P
+The \ec facility was designed for use with ASCII characters, but with the
+extension to Unicode it is even less useful than it once was. It is, however,
+recognized when PCRE is compiled in EBCDIC mode, where data items are always
+bytes. In this mode, all values are valid after \ec. If the next character is a
+lower case letter, it is converted to upper case. Then the 0xc0 bits of the
+byte are inverted. Thus \ecA becomes hex 01, as in ASCII (A is C1), but because
+the EBCDIC letters are disjoint, \ecZ becomes hex 29 (Z is E9), and other
+characters also generate different values.
+.P
+By default, after \ex, from zero to two hexadecimal digits are read (letters
+can be in upper or lower case). Any number of hexadecimal digits may appear
+between \ex{ and }, but the character code is constrained as follows:
+.sp
+  8-bit non-UTF mode    less than 0x100
+  8-bit UTF-8 mode      less than 0x10ffff and a valid codepoint
+  16-bit non-UTF mode   less than 0x10000
+  16-bit UTF-16 mode    less than 0x10ffff and a valid codepoint
+  32-bit non-UTF mode   less than 0x80000000
+  32-bit UTF-32 mode    less than 0x10ffff and a valid codepoint
+.sp
+Invalid Unicode codepoints are the range 0xd800 to 0xdfff (the so-called
+"surrogate" codepoints), and 0xffef.
+.P
+If characters other than hexadecimal digits appear between \ex{ and }, or if
+there is no terminating }, this form of escape is not recognized. Instead, the
+initial \ex will be interpreted as a basic hexadecimal escape, with no
+following digits, giving a character whose value is zero.
+.P
+If the PCRE_JAVASCRIPT_COMPAT option is set, the interpretation of \ex is
+as just described only when it is followed by two hexadecimal digits.
+Otherwise, it matches a literal "x" character. In JavaScript mode, support for
+code points greater than 256 is provided by \eu, which must be followed by
+four hexadecimal digits; otherwise it matches a literal "u" character.
+Character codes specified by \eu in JavaScript mode are constrained in the same
+was as those specified by \ex in non-JavaScript mode.
+.P
+Characters whose value is less than 256 can be defined by either of the two
+syntaxes for \ex (or by \eu in JavaScript mode). There is no difference in the
+way they are handled. For example, \exdc is exactly the same as \ex{dc} (or
+\eu00dc in JavaScript mode).
+.P
+After \e0 up to two further octal digits are read. If there are fewer than two
+digits, just those that are present are used. Thus the sequence \e0\ex\e07
+specifies two binary zeros followed by a BEL character (code value 7). Make
+sure you supply two digits after the initial zero if the pattern character that
+follows is itself an octal digit.
+.P
+The handling of a backslash followed by a digit other than 0 is complicated.
+Outside a character class, PCRE reads it and any following digits as a decimal
+number. If the number is less than 10, or if there have been at least that many
+previous capturing left parentheses in the expression, the entire sequence is
+taken as a \fIback reference\fP. A description of how this works is given
+.\" HTML 
+.\" 
+later,
+.\"
+following the discussion of
+.\" HTML 
+.\" 
+parenthesized subpatterns.
+.\"
+.P
+Inside a character class, or if the decimal number is greater than 9 and there
+have not been that many capturing subpatterns, PCRE re-reads up to three octal
+digits following the backslash, and uses them to generate a data character. Any
+subsequent digits stand for themselves. The value of the character is
+constrained in the same way as characters specified in hexadecimal.
+For example:
+.sp
+  \e040   is another way of writing an ASCII space
+.\" JOIN
+  \e40    is the same, provided there are fewer than 40
+            previous capturing subpatterns
+  \e7     is always a back reference
+.\" JOIN
+  \e11    might be a back reference, or another way of
+            writing a tab
+  \e011   is always a tab
+  \e0113  is a tab followed by the character "3"
+.\" JOIN
+  \e113   might be a back reference, otherwise the
+            character with octal code 113
+.\" JOIN
+  \e377   might be a back reference, otherwise
+            the value 255 (decimal)
+.\" JOIN
+  \e81    is either a back reference, or a binary zero
+            followed by the two characters "8" and "1"
+.sp
+Note that octal values of 100 or greater must not be introduced by a leading
+zero, because no more than three octal digits are ever read.
+.P
+All the sequences that define a single character value can be used both inside
+and outside character classes. In addition, inside a character class, \eb is
+interpreted as the backspace character (hex 08).
+.P
+\eN is not allowed in a character class. \eB, \eR, and \eX are not special
+inside a character class. Like other unrecognized escape sequences, they are
+treated as the literal characters "B", "R", and "X" by default, but cause an
+error if the PCRE_EXTRA option is set. Outside a character class, these
+sequences have different meanings.
+.
+.
+.SS "Unsupported escape sequences"
+.rs
+.sp
+In Perl, the sequences \el, \eL, \eu, and \eU are recognized by its string
+handler and used to modify the case of following characters. By default, PCRE
+does not support these escape sequences. However, if the PCRE_JAVASCRIPT_COMPAT
+option is set, \eU matches a "U" character, and \eu can be used to define a
+character by code point, as described in the previous section.
+.
+.
+.SS "Absolute and relative back references"
+.rs
+.sp
+The sequence \eg followed by an unsigned or a negative number, optionally
+enclosed in braces, is an absolute or relative back reference. A named back
+reference can be coded as \eg{name}. Back references are discussed
+.\" HTML 
+.\" 
+later,
+.\"
+following the discussion of
+.\" HTML 
+.\" 
+parenthesized subpatterns.
+.\"
+.
+.
+.SS "Absolute and relative subroutine calls"
+.rs
+.sp
+For compatibility with Oniguruma, the non-Perl syntax \eg followed by a name or
+a number enclosed either in angle brackets or single quotes, is an alternative
+syntax for referencing a subpattern as a "subroutine". Details are discussed
+.\" HTML 
+.\" 
+later.
+.\"
+Note that \eg{...} (Perl syntax) and \eg<...> (Oniguruma syntax) are \fInot\fP
+synonymous. The former is a back reference; the latter is a
+.\" HTML 
+.\" 
+subroutine
+.\"
+call.
+.
+.
+.\" HTML 
+.SS "Generic character types"
+.rs
+.sp
+Another use of backslash is for specifying generic character types:
+.sp
+  \ed     any decimal digit
+  \eD     any character that is not a decimal digit
+  \eh     any horizontal white space character
+  \eH     any character that is not a horizontal white space character
+  \es     any white space character
+  \eS     any character that is not a white space character
+  \ev     any vertical white space character
+  \eV     any character that is not a vertical white space character
+  \ew     any "word" character
+  \eW     any "non-word" character
+.sp
+There is also the single sequence \eN, which matches a non-newline character.
+This is the same as
+.\" HTML 
+.\" 
+the "." metacharacter
+.\"
+when PCRE_DOTALL is not set. Perl also uses \eN to match characters by name;
+PCRE does not support this.
+.P
+Each pair of lower and upper case escape sequences partitions the complete set
+of characters into two disjoint sets. Any given character matches one, and only
+one, of each pair. The sequences can appear both inside and outside character
+classes. They each match one character of the appropriate type. If the current
+matching point is at the end of the subject string, all of them fail, because
+there is no character to match.
+.P
+For compatibility with Perl, \es does not match the VT character (code 11).
+This makes it different from the the POSIX "space" class. The \es characters
+are HT (9), LF (10), FF (12), CR (13), and space (32). If "use locale;" is
+included in a Perl script, \es may match the VT character. In PCRE, it never
+does.
+.P
+A "word" character is an underscore or any character that is a letter or digit.
+By default, the definition of letters and digits is controlled by PCRE's
+low-valued character tables, and may vary if locale-specific matching is taking
+place (see
+.\" HTML 
+.\" 
+"Locale support"
+.\"
+in the
+.\" HREF
+\fBpcreapi\fP
+.\"
+page). For example, in a French locale such as "fr_FR" in Unix-like systems,
+or "french" in Windows, some character codes greater than 128 are used for
+accented letters, and these are then matched by \ew. The use of locales with
+Unicode is discouraged.
+.P
+By default, in a UTF mode, characters with values greater than 128 never match
+\ed, \es, or \ew, and always match \eD, \eS, and \eW. These sequences retain
+their original meanings from before UTF support was available, mainly for
+efficiency reasons. However, if PCRE is compiled with Unicode property support,
+and the PCRE_UCP option is set, the behaviour is changed so that Unicode
+properties are used to determine character types, as follows:
+.sp
+  \ed  any character that \ep{Nd} matches (decimal digit)
+  \es  any character that \ep{Z} matches, plus HT, LF, FF, CR
+  \ew  any character that \ep{L} or \ep{N} matches, plus underscore
+.sp
+The upper case escapes match the inverse sets of characters. Note that \ed
+matches only decimal digits, whereas \ew matches any Unicode digit, as well as
+any Unicode letter, and underscore. Note also that PCRE_UCP affects \eb, and
+\eB because they are defined in terms of \ew and \eW. Matching these sequences
+is noticeably slower when PCRE_UCP is set.
+.P
+The sequences \eh, \eH, \ev, and \eV are features that were added to Perl at
+release 5.10. In contrast to the other sequences, which match only ASCII
+characters by default, these always match certain high-valued codepoints,
+whether or not PCRE_UCP is set. The horizontal space characters are:
+.sp
+  U+0009     Horizontal tab (HT)
+  U+0020     Space
+  U+00A0     Non-break space
+  U+1680     Ogham space mark
+  U+180E     Mongolian vowel separator
+  U+2000     En quad
+  U+2001     Em quad
+  U+2002     En space
+  U+2003     Em space
+  U+2004     Three-per-em space
+  U+2005     Four-per-em space
+  U+2006     Six-per-em space
+  U+2007     Figure space
+  U+2008     Punctuation space
+  U+2009     Thin space
+  U+200A     Hair space
+  U+202F     Narrow no-break space
+  U+205F     Medium mathematical space
+  U+3000     Ideographic space
+.sp
+The vertical space characters are:
+.sp
+  U+000A     Linefeed (LF)
+  U+000B     Vertical tab (VT)
+  U+000C     Form feed (FF)
+  U+000D     Carriage return (CR)
+  U+0085     Next line (NEL)
+  U+2028     Line separator
+  U+2029     Paragraph separator
+.sp
+In 8-bit, non-UTF-8 mode, only the characters with codepoints less than 256 are
+relevant.
+.
+.
+.\" HTML 
+.SS "Newline sequences"
+.rs
+.sp
+Outside a character class, by default, the escape sequence \eR matches any
+Unicode newline sequence. In 8-bit non-UTF-8 mode \eR is equivalent to the
+following:
+.sp
+  (?>\er\en|\en|\ex0b|\ef|\er|\ex85)
+.sp
+This is an example of an "atomic group", details of which are given
+.\" HTML 
+.\" 
+below.
+.\"
+This particular group matches either the two-character sequence CR followed by
+LF, or one of the single characters LF (linefeed, U+000A), VT (vertical tab,
+U+000B), FF (form feed, U+000C), CR (carriage return, U+000D), or NEL (next
+line, U+0085). The two-character sequence is treated as a single unit that
+cannot be split.
+.P
+In other modes, two additional characters whose codepoints are greater than 255
+are added: LS (line separator, U+2028) and PS (paragraph separator, U+2029).
+Unicode character property support is not needed for these characters to be
+recognized.
+.P
+It is possible to restrict \eR to match only CR, LF, or CRLF (instead of the
+complete set of Unicode line endings) by setting the option PCRE_BSR_ANYCRLF
+either at compile time or when the pattern is matched. (BSR is an abbrevation
+for "backslash R".) This can be made the default when PCRE is built; if this is
+the case, the other behaviour can be requested via the PCRE_BSR_UNICODE option.
+It is also possible to specify these settings by starting a pattern string with
+one of the following sequences:
+.sp
+  (*BSR_ANYCRLF)   CR, LF, or CRLF only
+  (*BSR_UNICODE)   any Unicode newline sequence
+.sp
+These override the default and the options given to the compiling function, but
+they can themselves be overridden by options given to a matching function. Note
+that these special settings, which are not Perl-compatible, are recognized only
+at the very start of a pattern, and that they must be in upper case. If more
+than one of them is present, the last one is used. They can be combined with a
+change of newline convention; for example, a pattern can start with:
+.sp
+  (*ANY)(*BSR_ANYCRLF)
+.sp
+They can also be combined with the (*UTF8), (*UTF16), (*UTF32), (*UTF) or
+(*UCP) special sequences. Inside a character class, \eR is treated as an
+unrecognized escape sequence, and so matches the letter "R" by default, but
+causes an error if PCRE_EXTRA is set.
+.
+.
+.\" HTML 
+.SS Unicode character properties
+.rs
+.sp
+When PCRE is built with Unicode character property support, three additional
+escape sequences that match characters with specific properties are available.
+When in 8-bit non-UTF-8 mode, these sequences are of course limited to testing
+characters whose codepoints are less than 256, but they do work in this mode.
+The extra escape sequences are:
+.sp
+  \ep{\fIxx\fP}   a character with the \fIxx\fP property
+  \eP{\fIxx\fP}   a character without the \fIxx\fP property
+  \eX       a Unicode extended grapheme cluster
+.sp
+The property names represented by \fIxx\fP above are limited to the Unicode
+script names, the general category properties, "Any", which matches any
+character (including newline), and some special PCRE properties (described
+in the
+.\" HTML 
+.\" 
+next section).
+.\"
+Other Perl properties such as "InMusicalSymbols" are not currently supported by
+PCRE. Note that \eP{Any} does not match any characters, so always causes a
+match failure.
+.P
+Sets of Unicode characters are defined as belonging to certain scripts. A
+character from one of these sets can be matched using a script name. For
+example:
+.sp
+  \ep{Greek}
+  \eP{Han}
+.sp
+Those that are not part of an identified script are lumped together as
+"Common". The current list of scripts is:
+.P
+Arabic,
+Armenian,
+Avestan,
+Balinese,
+Bamum,
+Batak,
+Bengali,
+Bopomofo,
+Brahmi,
+Braille,
+Buginese,
+Buhid,
+Canadian_Aboriginal,
+Carian,
+Chakma,
+Cham,
+Cherokee,
+Common,
+Coptic,
+Cuneiform,
+Cypriot,
+Cyrillic,
+Deseret,
+Devanagari,
+Egyptian_Hieroglyphs,
+Ethiopic,
+Georgian,
+Glagolitic,
+Gothic,
+Greek,
+Gujarati,
+Gurmukhi,
+Han,
+Hangul,
+Hanunoo,
+Hebrew,
+Hiragana,
+Imperial_Aramaic,
+Inherited,
+Inscriptional_Pahlavi,
+Inscriptional_Parthian,
+Javanese,
+Kaithi,
+Kannada,
+Katakana,
+Kayah_Li,
+Kharoshthi,
+Khmer,
+Lao,
+Latin,
+Lepcha,
+Limbu,
+Linear_B,
+Lisu,
+Lycian,
+Lydian,
+Malayalam,
+Mandaic,
+Meetei_Mayek,
+Meroitic_Cursive,
+Meroitic_Hieroglyphs,
+Miao,
+Mongolian,
+Myanmar,
+New_Tai_Lue,
+Nko,
+Ogham,
+Old_Italic,
+Old_Persian,
+Old_South_Arabian,
+Old_Turkic,
+Ol_Chiki,
+Oriya,
+Osmanya,
+Phags_Pa,
+Phoenician,
+Rejang,
+Runic,
+Samaritan,
+Saurashtra,
+Sharada,
+Shavian,
+Sinhala,
+Sora_Sompeng,
+Sundanese,
+Syloti_Nagri,
+Syriac,
+Tagalog,
+Tagbanwa,
+Tai_Le,
+Tai_Tham,
+Tai_Viet,
+Takri,
+Tamil,
+Telugu,
+Thaana,
+Thai,
+Tibetan,
+Tifinagh,
+Ugaritic,
+Vai,
+Yi.
+.P
+Each character has exactly one Unicode general category property, specified by
+a two-letter abbreviation. For compatibility with Perl, negation can be
+specified by including a circumflex between the opening brace and the property
+name. For example, \ep{^Lu} is the same as \eP{Lu}.
+.P
+If only one letter is specified with \ep or \eP, it includes all the general
+category properties that start with that letter. In this case, in the absence
+of negation, the curly brackets in the escape sequence are optional; these two
+examples have the same effect:
+.sp
+  \ep{L}
+  \epL
+.sp
+The following general category property codes are supported:
+.sp
+  C     Other
+  Cc    Control
+  Cf    Format
+  Cn    Unassigned
+  Co    Private use
+  Cs    Surrogate
+.sp
+  L     Letter
+  Ll    Lower case letter
+  Lm    Modifier letter
+  Lo    Other letter
+  Lt    Title case letter
+  Lu    Upper case letter
+.sp
+  M     Mark
+  Mc    Spacing mark
+  Me    Enclosing mark
+  Mn    Non-spacing mark
+.sp
+  N     Number
+  Nd    Decimal number
+  Nl    Letter number
+  No    Other number
+.sp
+  P     Punctuation
+  Pc    Connector punctuation
+  Pd    Dash punctuation
+  Pe    Close punctuation
+  Pf    Final punctuation
+  Pi    Initial punctuation
+  Po    Other punctuation
+  Ps    Open punctuation
+.sp
+  S     Symbol
+  Sc    Currency symbol
+  Sk    Modifier symbol
+  Sm    Mathematical symbol
+  So    Other symbol
+.sp
+  Z     Separator
+  Zl    Line separator
+  Zp    Paragraph separator
+  Zs    Space separator
+.sp
+The special property L& is also supported: it matches a character that has
+the Lu, Ll, or Lt property, in other words, a letter that is not classified as
+a modifier or "other".
+.P
+The Cs (Surrogate) property applies only to characters in the range U+D800 to
+U+DFFF. Such characters are not valid in Unicode strings and so
+cannot be tested by PCRE, unless UTF validity checking has been turned off
+(see the discussion of PCRE_NO_UTF8_CHECK, PCRE_NO_UTF16_CHECK and
+PCRE_NO_UTF32_CHECK in the
+.\" HREF
+\fBpcreapi\fP
+.\"
+page). Perl does not support the Cs property.
+.P
+The long synonyms for property names that Perl supports (such as \ep{Letter})
+are not supported by PCRE, nor is it permitted to prefix any of these
+properties with "Is".
+.P
+No character that is in the Unicode table has the Cn (unassigned) property.
+Instead, this property is assumed for any code point that is not in the
+Unicode table.
+.P
+Specifying caseless matching does not affect these escape sequences. For
+example, \ep{Lu} always matches only upper case letters. This is different from
+the behaviour of current versions of Perl.
+.P
+Matching characters by Unicode property is not fast, because PCRE has to do a
+multistage table lookup in order to find a character's property. That is why
+the traditional escape sequences such as \ed and \ew do not use Unicode
+properties in PCRE by default, though you can make them do so by setting the
+PCRE_UCP option or by starting the pattern with (*UCP).
+.
+.
+.SS Extended grapheme clusters
+.rs
+.sp
+The \eX escape matches any number of Unicode characters that form an "extended
+grapheme cluster", and treats the sequence as an atomic group
+.\" HTML 
+.\" 
+(see below).
+.\"
+Up to and including release 8.31, PCRE matched an earlier, simpler definition
+that was equivalent to
+.sp
+  (?>\ePM\epM*)
+.sp
+That is, it matched a character without the "mark" property, followed by zero
+or more characters with the "mark" property. Characters with the "mark"
+property are typically non-spacing accents that affect the preceding character.
+.P
+This simple definition was extended in Unicode to include more complicated
+kinds of composite character by giving each character a grapheme breaking
+property, and creating rules that use these properties to define the boundaries
+of extended grapheme clusters. In releases of PCRE later than 8.31, \eX matches
+one of these clusters.
+.P
+\eX always matches at least one character. Then it decides whether to add
+additional characters according to the following rules for ending a cluster:
+.P
+1. End at the end of the subject string.
+.P
+2. Do not end between CR and LF; otherwise end after any control character.
+.P
+3. Do not break Hangul (a Korean script) syllable sequences. Hangul characters
+are of five types: L, V, T, LV, and LVT. An L character may be followed by an
+L, V, LV, or LVT character; an LV or V character may be followed by a V or T
+character; an LVT or T character may be follwed only by a T character.
+.P
+4. Do not end before extending characters or spacing marks. Characters with
+the "mark" property always have the "extend" grapheme breaking property.
+.P
+5. Do not end after prepend characters.
+.P
+6. Otherwise, end the cluster.
+.
+.
+.\" HTML 
+.SS PCRE's additional properties
+.rs
+.sp
+As well as the standard Unicode properties described above, PCRE supports four
+more that make it possible to convert traditional escape sequences such as \ew
+and \es and POSIX character classes to use Unicode properties. PCRE uses these
+non-standard, non-Perl properties internally when PCRE_UCP is set. However,
+they may also be used explicitly. These properties are:
+.sp
+  Xan   Any alphanumeric character
+  Xps   Any POSIX space character
+  Xsp   Any Perl space character
+  Xwd   Any Perl "word" character
+.sp
+Xan matches characters that have either the L (letter) or the N (number)
+property. Xps matches the characters tab, linefeed, vertical tab, form feed, or
+carriage return, and any other character that has the Z (separator) property.
+Xsp is the same as Xps, except that vertical tab is excluded. Xwd matches the
+same characters as Xan, plus underscore.
+.P
+There is another non-standard property, Xuc, which matches any character that
+can be represented by a Universal Character Name in C++ and other programming
+languages. These are the characters $, @, ` (grave accent), and all characters
+with Unicode code points greater than or equal to U+00A0, except for the
+surrogates U+D800 to U+DFFF. Note that most base (ASCII) characters are
+excluded. (Universal Character Names are of the form \euHHHH or \eUHHHHHHHH
+where H is a hexadecimal digit. Note that the Xuc property does not match these
+sequences but the characters that they represent.)
+.
+.
+.\" HTML 
+.SS "Resetting the match start"
+.rs
+.sp
+The escape sequence \eK causes any previously matched characters not to be
+included in the final matched sequence. For example, the pattern:
+.sp
+  foo\eKbar
+.sp
+matches "foobar", but reports that it has matched "bar". This feature is
+similar to a lookbehind assertion
+.\" HTML 
+.\" 
+(described below).
+.\"
+However, in this case, the part of the subject before the real match does not
+have to be of fixed length, as lookbehind assertions do. The use of \eK does
+not interfere with the setting of
+.\" HTML 
+.\" 
+captured substrings.
+.\"
+For example, when the pattern
+.sp
+  (foo)\eKbar
+.sp
+matches "foobar", the first substring is still set to "foo".
+.P
+Perl documents that the use of \eK within assertions is "not well defined". In
+PCRE, \eK is acted upon when it occurs inside positive assertions, but is
+ignored in negative assertions.
+.
+.
+.\" HTML 
+.SS "Simple assertions"
+.rs
+.sp
+The final use of backslash is for certain simple assertions. An assertion
+specifies a condition that has to be met at a particular point in a match,
+without consuming any characters from the subject string. The use of
+subpatterns for more complicated assertions is described
+.\" HTML 
+.\" 
+below.
+.\"
+The backslashed assertions are:
+.sp
+  \eb     matches at a word boundary
+  \eB     matches when not at a word boundary
+  \eA     matches at the start of the subject
+  \eZ     matches at the end of the subject
+          also matches before a newline at the end of the subject
+  \ez     matches only at the end of the subject
+  \eG     matches at the first matching position in the subject
+.sp
+Inside a character class, \eb has a different meaning; it matches the backspace
+character. If any other of these assertions appears in a character class, by
+default it matches the corresponding literal character (for example, \eB
+matches the letter B). However, if the PCRE_EXTRA option is set, an "invalid
+escape sequence" error is generated instead.
+.P
+A word boundary is a position in the subject string where the current character
+and the previous character do not both match \ew or \eW (i.e. one matches
+\ew and the other matches \eW), or the start or end of the string if the
+first or last character matches \ew, respectively. In a UTF mode, the meanings
+of \ew and \eW can be changed by setting the PCRE_UCP option. When this is
+done, it also affects \eb and \eB. Neither PCRE nor Perl has a separate "start
+of word" or "end of word" metasequence. However, whatever follows \eb normally
+determines which it is. For example, the fragment \eba matches "a" at the start
+of a word.
+.P
+The \eA, \eZ, and \ez assertions differ from the traditional circumflex and
+dollar (described in the next section) in that they only ever match at the very
+start and end of the subject string, whatever options are set. Thus, they are
+independent of multiline mode. These three assertions are not affected by the
+PCRE_NOTBOL or PCRE_NOTEOL options, which affect only the behaviour of the
+circumflex and dollar metacharacters. However, if the \fIstartoffset\fP
+argument of \fBpcre_exec()\fP is non-zero, indicating that matching is to start
+at a point other than the beginning of the subject, \eA can never match. The
+difference between \eZ and \ez is that \eZ matches before a newline at the end
+of the string as well as at the very end, whereas \ez matches only at the end.
+.P
+The \eG assertion is true only when the current matching position is at the
+start point of the match, as specified by the \fIstartoffset\fP argument of
+\fBpcre_exec()\fP. It differs from \eA when the value of \fIstartoffset\fP is
+non-zero. By calling \fBpcre_exec()\fP multiple times with appropriate
+arguments, you can mimic Perl's /g option, and it is in this kind of
+implementation where \eG can be useful.
+.P
+Note, however, that PCRE's interpretation of \eG, as the start of the current
+match, is subtly different from Perl's, which defines it as the end of the
+previous match. In Perl, these can be different when the previously matched
+string was empty. Because PCRE does just one match at a time, it cannot
+reproduce this behaviour.
+.P
+If all the alternatives of a pattern begin with \eG, the expression is anchored
+to the starting match position, and the "anchored" flag is set in the compiled
+regular expression.
+.
+.
+.SH "CIRCUMFLEX AND DOLLAR"
+.rs
+.sp
+The circumflex and dollar metacharacters are zero-width assertions. That is,
+they test for a particular condition being true without consuming any
+characters from the subject string.
+.P
+Outside a character class, in the default matching mode, the circumflex
+character is an assertion that is true only if the current matching point is at
+the start of the subject string. If the \fIstartoffset\fP argument of
+\fBpcre_exec()\fP is non-zero, circumflex can never match if the PCRE_MULTILINE
+option is unset. Inside a character class, circumflex has an entirely different
+meaning
+.\" HTML 
+.\" 
+(see below).
+.\"
+.P
+Circumflex need not be the first character of the pattern if a number of
+alternatives are involved, but it should be the first thing in each alternative
+in which it appears if the pattern is ever to match that branch. If all
+possible alternatives start with a circumflex, that is, if the pattern is
+constrained to match only at the start of the subject, it is said to be an
+"anchored" pattern. (There are also other constructs that can cause a pattern
+to be anchored.)
+.P
+The dollar character is an assertion that is true only if the current matching
+point is at the end of the subject string, or immediately before a newline at
+the end of the string (by default). Note, however, that it does not actually
+match the newline. Dollar need not be the last character of the pattern if a
+number of alternatives are involved, but it should be the last item in any
+branch in which it appears. Dollar has no special meaning in a character class.
+.P
+The meaning of dollar can be changed so that it matches only at the very end of
+the string, by setting the PCRE_DOLLAR_ENDONLY option at compile time. This
+does not affect the \eZ assertion.
+.P
+The meanings of the circumflex and dollar characters are changed if the
+PCRE_MULTILINE option is set. When this is the case, a circumflex matches
+immediately after internal newlines as well as at the start of the subject
+string. It does not match after a newline that ends the string. A dollar
+matches before any newlines in the string, as well as at the very end, when
+PCRE_MULTILINE is set. When newline is specified as the two-character
+sequence CRLF, isolated CR and LF characters do not indicate newlines.
+.P
+For example, the pattern /^abc$/ matches the subject string "def\enabc" (where
+\en represents a newline) in multiline mode, but not otherwise. Consequently,
+patterns that are anchored in single line mode because all branches start with
+^ are not anchored in multiline mode, and a match for circumflex is possible
+when the \fIstartoffset\fP argument of \fBpcre_exec()\fP is non-zero. The
+PCRE_DOLLAR_ENDONLY option is ignored if PCRE_MULTILINE is set.
+.P
+Note that the sequences \eA, \eZ, and \ez can be used to match the start and
+end of the subject in both modes, and if all branches of a pattern start with
+\eA it is always anchored, whether or not PCRE_MULTILINE is set.
+.
+.
+.\" HTML 
+.SH "FULL STOP (PERIOD, DOT) AND \eN"
+.rs
+.sp
+Outside a character class, a dot in the pattern matches any one character in
+the subject string except (by default) a character that signifies the end of a
+line.
+.P
+When a line ending is defined as a single character, dot never matches that
+character; when the two-character sequence CRLF is used, dot does not match CR
+if it is immediately followed by LF, but otherwise it matches all characters
+(including isolated CRs and LFs). When any Unicode line endings are being
+recognized, dot does not match CR or LF or any of the other line ending
+characters.
+.P
+The behaviour of dot with regard to newlines can be changed. If the PCRE_DOTALL
+option is set, a dot matches any one character, without exception. If the
+two-character sequence CRLF is present in the subject string, it takes two dots
+to match it.
+.P
+The handling of dot is entirely independent of the handling of circumflex and
+dollar, the only relationship being that they both involve newlines. Dot has no
+special meaning in a character class.
+.P
+The escape sequence \eN behaves like a dot, except that it is not affected by
+the PCRE_DOTALL option. In other words, it matches any character except one
+that signifies the end of a line. Perl also uses \eN to match characters by
+name; PCRE does not support this.
+.
+.
+.SH "MATCHING A SINGLE DATA UNIT"
+.rs
+.sp
+Outside a character class, the escape sequence \eC matches any one data unit,
+whether or not a UTF mode is set. In the 8-bit library, one data unit is one
+byte; in the 16-bit library it is a 16-bit unit; in the 32-bit library it is
+a 32-bit unit. Unlike a dot, \eC always
+matches line-ending characters. The feature is provided in Perl in order to
+match individual bytes in UTF-8 mode, but it is unclear how it can usefully be
+used. Because \eC breaks up characters into individual data units, matching one
+unit with \eC in a UTF mode means that the rest of the string may start with a
+malformed UTF character. This has undefined results, because PCRE assumes that
+it is dealing with valid UTF strings (and by default it checks this at the
+start of processing unless the PCRE_NO_UTF8_CHECK, PCRE_NO_UTF16_CHECK or
+PCRE_NO_UTF32_CHECK option is used).
+.P
+PCRE does not allow \eC to appear in lookbehind assertions
+.\" HTML 
+.\" 
+(described below)
+.\"
+in a UTF mode, because this would make it impossible to calculate the length of
+the lookbehind.
+.P
+In general, the \eC escape sequence is best avoided. However, one
+way of using it that avoids the problem of malformed UTF characters is to use a
+lookahead to check the length of the next character, as in this pattern, which
+could be used with a UTF-8 string (ignore white space and line breaks):
+.sp
+  (?| (?=[\ex00-\ex7f])(\eC) |
+      (?=[\ex80-\ex{7ff}])(\eC)(\eC) |
+      (?=[\ex{800}-\ex{ffff}])(\eC)(\eC)(\eC) |
+      (?=[\ex{10000}-\ex{1fffff}])(\eC)(\eC)(\eC)(\eC))
+.sp
+A group that starts with (?| resets the capturing parentheses numbers in each
+alternative (see
+.\" HTML 
+.\" 
+"Duplicate Subpattern Numbers"
+.\"
+below). The assertions at the start of each branch check the next UTF-8
+character for values whose encoding uses 1, 2, 3, or 4 bytes, respectively. The
+character's individual bytes are then captured by the appropriate number of
+groups.
+.
+.
+.\" HTML 
+.SH "SQUARE BRACKETS AND CHARACTER CLASSES"
+.rs
+.sp
+An opening square bracket introduces a character class, terminated by a closing
+square bracket. A closing square bracket on its own is not special by default.
+However, if the PCRE_JAVASCRIPT_COMPAT option is set, a lone closing square
+bracket causes a compile-time error. If a closing square bracket is required as
+a member of the class, it should be the first data character in the class
+(after an initial circumflex, if present) or escaped with a backslash.
+.P
+A character class matches a single character in the subject. In a UTF mode, the
+character may be more than one data unit long. A matched character must be in
+the set of characters defined by the class, unless the first character in the
+class definition is a circumflex, in which case the subject character must not
+be in the set defined by the class. If a circumflex is actually required as a
+member of the class, ensure it is not the first character, or escape it with a
+backslash.
+.P
+For example, the character class [aeiou] matches any lower case vowel, while
+[^aeiou] matches any character that is not a lower case vowel. Note that a
+circumflex is just a convenient notation for specifying the characters that
+are in the class by enumerating those that are not. A class that starts with a
+circumflex is not an assertion; it still consumes a character from the subject
+string, and therefore it fails if the current pointer is at the end of the
+string.
+.P
+In UTF-8 (UTF-16, UTF-32) mode, characters with values greater than 255 (0xffff)
+can be included in a class as a literal string of data units, or by using the
+\ex{ escaping mechanism.
+.P
+When caseless matching is set, any letters in a class represent both their
+upper case and lower case versions, so for example, a caseless [aeiou] matches
+"A" as well as "a", and a caseless [^aeiou] does not match "A", whereas a
+caseful version would. In a UTF mode, PCRE always understands the concept of
+case for characters whose values are less than 128, so caseless matching is
+always possible. For characters with higher values, the concept of case is
+supported if PCRE is compiled with Unicode property support, but not otherwise.
+If you want to use caseless matching in a UTF mode for characters 128 and
+above, you must ensure that PCRE is compiled with Unicode property support as
+well as with UTF support.
+.P
+Characters that might indicate line breaks are never treated in any special way
+when matching character classes, whatever line-ending sequence is in use, and
+whatever setting of the PCRE_DOTALL and PCRE_MULTILINE options is used. A class
+such as [^a] always matches one of these characters.
+.P
+The minus (hyphen) character can be used to specify a range of characters in a
+character class. For example, [d-m] matches any letter between d and m,
+inclusive. If a minus character is required in a class, it must be escaped with
+a backslash or appear in a position where it cannot be interpreted as
+indicating a range, typically as the first or last character in the class.
+.P
+It is not possible to have the literal character "]" as the end character of a
+range. A pattern such as [W-]46] is interpreted as a class of two characters
+("W" and "-") followed by a literal string "46]", so it would match "W46]" or
+"-46]". However, if the "]" is escaped with a backslash it is interpreted as
+the end of range, so [W-\e]46] is interpreted as a class containing a range
+followed by two other characters. The octal or hexadecimal representation of
+"]" can also be used to end a range.
+.P
+Ranges operate in the collating sequence of character values. They can also be
+used for characters specified numerically, for example [\e000-\e037]. Ranges
+can include any characters that are valid for the current mode.
+.P
+If a range that includes letters is used when caseless matching is set, it
+matches the letters in either case. For example, [W-c] is equivalent to
+[][\e\e^_`wxyzabc], matched caselessly, and in a non-UTF mode, if character
+tables for a French locale are in use, [\exc8-\excb] matches accented E
+characters in both cases. In UTF modes, PCRE supports the concept of case for
+characters with values greater than 128 only when it is compiled with Unicode
+property support.
+.P
+The character escape sequences \ed, \eD, \eh, \eH, \ep, \eP, \es, \eS, \ev,
+\eV, \ew, and \eW may appear in a character class, and add the characters that
+they match to the class. For example, [\edABCDEF] matches any hexadecimal
+digit. In UTF modes, the PCRE_UCP option affects the meanings of \ed, \es, \ew
+and their upper case partners, just as it does when they appear outside a
+character class, as described in the section entitled
+.\" HTML 
+.\" 
+"Generic character types"
+.\"
+above. The escape sequence \eb has a different meaning inside a character
+class; it matches the backspace character. The sequences \eB, \eN, \eR, and \eX
+are not special inside a character class. Like any other unrecognized escape
+sequences, they are treated as the literal characters "B", "N", "R", and "X" by
+default, but cause an error if the PCRE_EXTRA option is set.
+.P
+A circumflex can conveniently be used with the upper case character types to
+specify a more restricted set of characters than the matching lower case type.
+For example, the class [^\eW_] matches any letter or digit, but not underscore,
+whereas [\ew] includes underscore. A positive character class should be read as
+"something OR something OR ..." and a negative class as "NOT something AND NOT
+something AND NOT ...".
+.P
+The only metacharacters that are recognized in character classes are backslash,
+hyphen (only where it can be interpreted as specifying a range), circumflex
+(only at the start), opening square bracket (only when it can be interpreted as
+introducing a POSIX class name - see the next section), and the terminating
+closing square bracket. However, escaping other non-alphanumeric characters
+does no harm.
+.
+.
+.SH "POSIX CHARACTER CLASSES"
+.rs
+.sp
+Perl supports the POSIX notation for character classes. This uses names
+enclosed by [: and :] within the enclosing square brackets. PCRE also supports
+this notation. For example,
+.sp
+  [01[:alpha:]%]
+.sp
+matches "0", "1", any alphabetic character, or "%". The supported class names
+are:
+.sp
+  alnum    letters and digits
+  alpha    letters
+  ascii    character codes 0 - 127
+  blank    space or tab only
+  cntrl    control characters
+  digit    decimal digits (same as \ed)
+  graph    printing characters, excluding space
+  lower    lower case letters
+  print    printing characters, including space
+  punct    printing characters, excluding letters and digits and space
+  space    white space (not quite the same as \es)
+  upper    upper case letters
+  word     "word" characters (same as \ew)
+  xdigit   hexadecimal digits
+.sp
+The "space" characters are HT (9), LF (10), VT (11), FF (12), CR (13), and
+space (32). Notice that this list includes the VT character (code 11). This
+makes "space" different to \es, which does not include VT (for Perl
+compatibility).
+.P
+The name "word" is a Perl extension, and "blank" is a GNU extension from Perl
+5.8. Another Perl extension is negation, which is indicated by a ^ character
+after the colon. For example,
+.sp
+  [12[:^digit:]]
+.sp
+matches "1", "2", or any non-digit. PCRE (and Perl) also recognize the POSIX
+syntax [.ch.] and [=ch=] where "ch" is a "collating element", but these are not
+supported, and an error is given if they are encountered.
+.P
+By default, in UTF modes, characters with values greater than 128 do not match
+any of the POSIX character classes. However, if the PCRE_UCP option is passed
+to \fBpcre_compile()\fP, some of the classes are changed so that Unicode
+character properties are used. This is achieved by replacing the POSIX classes
+by other sequences, as follows:
+.sp
+  [:alnum:]  becomes  \ep{Xan}
+  [:alpha:]  becomes  \ep{L}
+  [:blank:]  becomes  \eh
+  [:digit:]  becomes  \ep{Nd}
+  [:lower:]  becomes  \ep{Ll}
+  [:space:]  becomes  \ep{Xps}
+  [:upper:]  becomes  \ep{Lu}
+  [:word:]   becomes  \ep{Xwd}
+.sp
+Negated versions, such as [:^alpha:] use \eP instead of \ep. The other POSIX
+classes are unchanged, and match only characters with code points less than
+128.
+.
+.
+.SH "VERTICAL BAR"
+.rs
+.sp
+Vertical bar characters are used to separate alternative patterns. For example,
+the pattern
+.sp
+  gilbert|sullivan
+.sp
+matches either "gilbert" or "sullivan". Any number of alternatives may appear,
+and an empty alternative is permitted (matching the empty string). The matching
+process tries each alternative in turn, from left to right, and the first one
+that succeeds is used. If the alternatives are within a subpattern
+.\" HTML 
+.\" 
+(defined below),
+.\"
+"succeeds" means matching the rest of the main pattern as well as the
+alternative in the subpattern.
+.
+.
+.SH "INTERNAL OPTION SETTING"
+.rs
+.sp
+The settings of the PCRE_CASELESS, PCRE_MULTILINE, PCRE_DOTALL, and
+PCRE_EXTENDED options (which are Perl-compatible) can be changed from within
+the pattern by a sequence of Perl option letters enclosed between "(?" and ")".
+The option letters are
+.sp
+  i  for PCRE_CASELESS
+  m  for PCRE_MULTILINE
+  s  for PCRE_DOTALL
+  x  for PCRE_EXTENDED
+.sp
+For example, (?im) sets caseless, multiline matching. It is also possible to
+unset these options by preceding the letter with a hyphen, and a combined
+setting and unsetting such as (?im-sx), which sets PCRE_CASELESS and
+PCRE_MULTILINE while unsetting PCRE_DOTALL and PCRE_EXTENDED, is also
+permitted. If a letter appears both before and after the hyphen, the option is
+unset.
+.P
+The PCRE-specific options PCRE_DUPNAMES, PCRE_UNGREEDY, and PCRE_EXTRA can be
+changed in the same way as the Perl-compatible options by using the characters
+J, U and X respectively.
+.P
+When one of these option changes occurs at top level (that is, not inside
+subpattern parentheses), the change applies to the remainder of the pattern
+that follows. If the change is placed right at the start of a pattern, PCRE
+extracts it into the global options (and it will therefore show up in data
+extracted by the \fBpcre_fullinfo()\fP function).
+.P
+An option change within a subpattern (see below for a description of
+subpatterns) affects only that part of the subpattern that follows it, so
+.sp
+  (a(?i)b)c
+.sp
+matches abc and aBc and no other strings (assuming PCRE_CASELESS is not used).
+By this means, options can be made to have different settings in different
+parts of the pattern. Any changes made in one alternative do carry on
+into subsequent branches within the same subpattern. For example,
+.sp
+  (a(?i)b|c)
+.sp
+matches "ab", "aB", "c", and "C", even though when matching "C" the first
+branch is abandoned before the option setting. This is because the effects of
+option settings happen at compile time. There would be some very weird
+behaviour otherwise.
+.P
+\fBNote:\fP There are other PCRE-specific options that can be set by the
+application when the compiling or matching functions are called. In some cases
+the pattern can contain special leading sequences such as (*CRLF) to override
+what the application has set or what has been defaulted. Details are given in
+the section entitled
+.\" HTML 
+.\" 
+"Newline sequences"
+.\"
+above. There are also the (*UTF8), (*UTF16),(*UTF32), and (*UCP) leading
+sequences that can be used to set UTF and Unicode property modes; they are
+equivalent to setting the PCRE_UTF8, PCRE_UTF16, PCRE_UTF32 and the PCRE_UCP
+options, respectively. The (*UTF) sequence is a generic version that can be
+used with any of the libraries. However, the application can set the
+PCRE_NEVER_UTF option, which locks out the use of the (*UTF) sequences.
+.
+.
+.\" HTML 
+.SH SUBPATTERNS
+.rs
+.sp
+Subpatterns are delimited by parentheses (round brackets), which can be nested.
+Turning part of a pattern into a subpattern does two things:
+.sp
+1. It localizes a set of alternatives. For example, the pattern
+.sp
+  cat(aract|erpillar|)
+.sp
+matches "cataract", "caterpillar", or "cat". Without the parentheses, it would
+match "cataract", "erpillar" or an empty string.
+.sp
+2. It sets up the subpattern as a capturing subpattern. This means that, when
+the whole pattern matches, that portion of the subject string that matched the
+subpattern is passed back to the caller via the \fIovector\fP argument of the
+matching function. (This applies only to the traditional matching functions;
+the DFA matching functions do not support capturing.)
+.P
+Opening parentheses are counted from left to right (starting from 1) to obtain
+numbers for the capturing subpatterns. For example, if the string "the red
+king" is matched against the pattern
+.sp
+  the ((red|white) (king|queen))
+.sp
+the captured substrings are "red king", "red", and "king", and are numbered 1,
+2, and 3, respectively.
+.P
+The fact that plain parentheses fulfil two functions is not always helpful.
+There are often times when a grouping subpattern is required without a
+capturing requirement. If an opening parenthesis is followed by a question mark
+and a colon, the subpattern does not do any capturing, and is not counted when
+computing the number of any subsequent capturing subpatterns. For example, if
+the string "the white queen" is matched against the pattern
+.sp
+  the ((?:red|white) (king|queen))
+.sp
+the captured substrings are "white queen" and "queen", and are numbered 1 and
+2. The maximum number of capturing subpatterns is 65535.
+.P
+As a convenient shorthand, if any option settings are required at the start of
+a non-capturing subpattern, the option letters may appear between the "?" and
+the ":". Thus the two patterns
+.sp
+  (?i:saturday|sunday)
+  (?:(?i)saturday|sunday)
+.sp
+match exactly the same set of strings. Because alternative branches are tried
+from left to right, and options are not reset until the end of the subpattern
+is reached, an option setting in one branch does affect subsequent branches, so
+the above patterns match "SUNDAY" as well as "Saturday".
+.
+.
+.\" HTML 
+.SH "DUPLICATE SUBPATTERN NUMBERS"
+.rs
+.sp
+Perl 5.10 introduced a feature whereby each alternative in a subpattern uses
+the same numbers for its capturing parentheses. Such a subpattern starts with
+(?| and is itself a non-capturing subpattern. For example, consider this
+pattern:
+.sp
+  (?|(Sat)ur|(Sun))day
+.sp
+Because the two alternatives are inside a (?| group, both sets of capturing
+parentheses are numbered one. Thus, when the pattern matches, you can look
+at captured substring number one, whichever alternative matched. This construct
+is useful when you want to capture part, but not all, of one of a number of
+alternatives. Inside a (?| group, parentheses are numbered as usual, but the
+number is reset at the start of each branch. The numbers of any capturing
+parentheses that follow the subpattern start after the highest number used in
+any branch. The following example is taken from the Perl documentation. The
+numbers underneath show in which buffer the captured content will be stored.
+.sp
+  # before  ---------------branch-reset----------- after
+  / ( a )  (?| x ( y ) z | (p (q) r) | (t) u (v) ) ( z ) /x
+  # 1            2         2  3        2     3     4
+.sp
+A back reference to a numbered subpattern uses the most recent value that is
+set for that number by any subpattern. The following pattern matches "abcabc"
+or "defdef":
+.sp
+  /(?|(abc)|(def))\e1/
+.sp
+In contrast, a subroutine call to a numbered subpattern always refers to the
+first one in the pattern with the given number. The following pattern matches
+"abcabc" or "defabc":
+.sp
+  /(?|(abc)|(def))(?1)/
+.sp
+If a
+.\" HTML 
+.\" 
+condition test
+.\"
+for a subpattern's having matched refers to a non-unique number, the test is
+true if any of the subpatterns of that number have matched.
+.P
+An alternative approach to using this "branch reset" feature is to use
+duplicate named subpatterns, as described in the next section.
+.
+.
+.SH "NAMED SUBPATTERNS"
+.rs
+.sp
+Identifying capturing parentheses by number is simple, but it can be very hard
+to keep track of the numbers in complicated regular expressions. Furthermore,
+if an expression is modified, the numbers may change. To help with this
+difficulty, PCRE supports the naming of subpatterns. This feature was not
+added to Perl until release 5.10. Python had the feature earlier, and PCRE
+introduced it at release 4.0, using the Python syntax. PCRE now supports both
+the Perl and the Python syntax. Perl allows identically numbered subpatterns to
+have different names, but PCRE does not.
+.P
+In PCRE, a subpattern can be named in one of three ways: (?...) or
+(?'name'...) as in Perl, or (?P...) as in Python. References to capturing
+parentheses from other parts of the pattern, such as
+.\" HTML 
+.\" 
+back references,
+.\"
+.\" HTML 
+.\" 
+recursion,
+.\"
+and
+.\" HTML 
+.\" 
+conditions,
+.\"
+can be made by name as well as by number.
+.P
+Names consist of up to 32 alphanumeric characters and underscores. Named
+capturing parentheses are still allocated numbers as well as names, exactly as
+if the names were not present. The PCRE API provides function calls for
+extracting the name-to-number translation table from a compiled pattern. There
+is also a convenience function for extracting a captured substring by name.
+.P
+By default, a name must be unique within a pattern, but it is possible to relax
+this constraint by setting the PCRE_DUPNAMES option at compile time. (Duplicate
+names are also always permitted for subpatterns with the same number, set up as
+described in the previous section.) Duplicate names can be useful for patterns
+where only one instance of the named parentheses can match. Suppose you want to
+match the name of a weekday, either as a 3-letter abbreviation or as the full
+name, and in both cases you want to extract the abbreviation. This pattern
+(ignoring the line breaks) does the job:
+.sp
+  (?Mon|Fri|Sun)(?:day)?|
+  (?Tue)(?:sday)?|
+  (?Wed)(?:nesday)?|
+  (?Thu)(?:rsday)?|
+  (?Sat)(?:urday)?
+.sp
+There are five capturing substrings, but only one is ever set after a match.
+(An alternative way of solving this problem is to use a "branch reset"
+subpattern, as described in the previous section.)
+.P
+The convenience function for extracting the data by name returns the substring
+for the first (and in this example, the only) subpattern of that name that
+matched. This saves searching to find which numbered subpattern it was.
+.P
+If you make a back reference to a non-unique named subpattern from elsewhere in
+the pattern, the one that corresponds to the first occurrence of the name is
+used. In the absence of duplicate numbers (see the previous section) this is
+the one with the lowest number. If you use a named reference in a condition
+test (see the
+.\"
+.\" HTML 
+.\" 
+section about conditions
+.\"
+below), either to check whether a subpattern has matched, or to check for
+recursion, all subpatterns with the same name are tested. If the condition is
+true for any one of them, the overall condition is true. This is the same
+behaviour as testing by number. For further details of the interfaces for
+handling named subpatterns, see the
+.\" HREF
+\fBpcreapi\fP
+.\"
+documentation.
+.P
+\fBWarning:\fP You cannot use different names to distinguish between two
+subpatterns with the same number because PCRE uses only the numbers when
+matching. For this reason, an error is given at compile time if different names
+are given to subpatterns with the same number. However, you can give the same
+name to subpatterns with the same number, even when PCRE_DUPNAMES is not set.
+.
+.
+.SH REPETITION
+.rs
+.sp
+Repetition is specified by quantifiers, which can follow any of the following
+items:
+.sp
+  a literal data character
+  the dot metacharacter
+  the \eC escape sequence
+  the \eX escape sequence
+  the \eR escape sequence
+  an escape such as \ed or \epL that matches a single character
+  a character class
+  a back reference (see next section)
+  a parenthesized subpattern (including assertions)
+  a subroutine call to a subpattern (recursive or otherwise)
+.sp
+The general repetition quantifier specifies a minimum and maximum number of
+permitted matches, by giving the two numbers in curly brackets (braces),
+separated by a comma. The numbers must be less than 65536, and the first must
+be less than or equal to the second. For example:
+.sp
+  z{2,4}
+.sp
+matches "zz", "zzz", or "zzzz". A closing brace on its own is not a special
+character. If the second number is omitted, but the comma is present, there is
+no upper limit; if the second number and the comma are both omitted, the
+quantifier specifies an exact number of required matches. Thus
+.sp
+  [aeiou]{3,}
+.sp
+matches at least 3 successive vowels, but may match many more, while
+.sp
+  \ed{8}
+.sp
+matches exactly 8 digits. An opening curly bracket that appears in a position
+where a quantifier is not allowed, or one that does not match the syntax of a
+quantifier, is taken as a literal character. For example, {,6} is not a
+quantifier, but a literal string of four characters.
+.P
+In UTF modes, quantifiers apply to characters rather than to individual data
+units. Thus, for example, \ex{100}{2} matches two characters, each of
+which is represented by a two-byte sequence in a UTF-8 string. Similarly,
+\eX{3} matches three Unicode extended grapheme clusters, each of which may be
+several data units long (and they may be of different lengths).
+.P
+The quantifier {0} is permitted, causing the expression to behave as if the
+previous item and the quantifier were not present. This may be useful for
+subpatterns that are referenced as
+.\" HTML 
+.\" 
+subroutines
+.\"
+from elsewhere in the pattern (but see also the section entitled
+.\" HTML 
+.\" 
+"Defining subpatterns for use by reference only"
+.\"
+below). Items other than subpatterns that have a {0} quantifier are omitted
+from the compiled pattern.
+.P
+For convenience, the three most common quantifiers have single-character
+abbreviations:
+.sp
+  *    is equivalent to {0,}
+  +    is equivalent to {1,}
+  ?    is equivalent to {0,1}
+.sp
+It is possible to construct infinite loops by following a subpattern that can
+match no characters with a quantifier that has no upper limit, for example:
+.sp
+  (a?)*
+.sp
+Earlier versions of Perl and PCRE used to give an error at compile time for
+such patterns. However, because there are cases where this can be useful, such
+patterns are now accepted, but if any repetition of the subpattern does in fact
+match no characters, the loop is forcibly broken.
+.P
+By default, the quantifiers are "greedy", that is, they match as much as
+possible (up to the maximum number of permitted times), without causing the
+rest of the pattern to fail. The classic example of where this gives problems
+is in trying to match comments in C programs. These appear between /* and */
+and within the comment, individual * and / characters may appear. An attempt to
+match C comments by applying the pattern
+.sp
+  /\e*.*\e*/
+.sp
+to the string
+.sp
+  /* first comment */  not comment  /* second comment */
+.sp
+fails, because it matches the entire string owing to the greediness of the .*
+item.
+.P
+However, if a quantifier is followed by a question mark, it ceases to be
+greedy, and instead matches the minimum number of times possible, so the
+pattern
+.sp
+  /\e*.*?\e*/
+.sp
+does the right thing with the C comments. The meaning of the various
+quantifiers is not otherwise changed, just the preferred number of matches.
+Do not confuse this use of question mark with its use as a quantifier in its
+own right. Because it has two uses, it can sometimes appear doubled, as in
+.sp
+  \ed??\ed
+.sp
+which matches one digit by preference, but can match two if that is the only
+way the rest of the pattern matches.
+.P
+If the PCRE_UNGREEDY option is set (an option that is not available in Perl),
+the quantifiers are not greedy by default, but individual ones can be made
+greedy by following them with a question mark. In other words, it inverts the
+default behaviour.
+.P
+When a parenthesized subpattern is quantified with a minimum repeat count that
+is greater than 1 or with a limited maximum, more memory is required for the
+compiled pattern, in proportion to the size of the minimum or maximum.
+.P
+If a pattern starts with .* or .{0,} and the PCRE_DOTALL option (equivalent
+to Perl's /s) is set, thus allowing the dot to match newlines, the pattern is
+implicitly anchored, because whatever follows will be tried against every
+character position in the subject string, so there is no point in retrying the
+overall match at any position after the first. PCRE normally treats such a
+pattern as though it were preceded by \eA.
+.P
+In cases where it is known that the subject string contains no newlines, it is
+worth setting PCRE_DOTALL in order to obtain this optimization, or
+alternatively using ^ to indicate anchoring explicitly.
+.P
+However, there are some cases where the optimization cannot be used. When .*
+is inside capturing parentheses that are the subject of a back reference
+elsewhere in the pattern, a match at the start may fail where a later one
+succeeds. Consider, for example:
+.sp
+  (.*)abc\e1
+.sp
+If the subject is "xyz123abc123" the match point is the fourth character. For
+this reason, such a pattern is not implicitly anchored.
+.P
+Another case where implicit anchoring is not applied is when the leading .* is
+inside an atomic group. Once again, a match at the start may fail where a later
+one succeeds. Consider this pattern:
+.sp
+  (?>.*?a)b
+.sp
+It matches "ab" in the subject "aab". The use of the backtracking control verbs
+(*PRUNE) and (*SKIP) also disable this optimization.
+.P
+When a capturing subpattern is repeated, the value captured is the substring
+that matched the final iteration. For example, after
+.sp
+  (tweedle[dume]{3}\es*)+
+.sp
+has matched "tweedledum tweedledee" the value of the captured substring is
+"tweedledee". However, if there are nested capturing subpatterns, the
+corresponding captured values may have been set in previous iterations. For
+example, after
+.sp
+  /(a|(b))+/
+.sp
+matches "aba" the value of the second captured substring is "b".
+.
+.
+.\" HTML 
+.SH "ATOMIC GROUPING AND POSSESSIVE QUANTIFIERS"
+.rs
+.sp
+With both maximizing ("greedy") and minimizing ("ungreedy" or "lazy")
+repetition, failure of what follows normally causes the repeated item to be
+re-evaluated to see if a different number of repeats allows the rest of the
+pattern to match. Sometimes it is useful to prevent this, either to change the
+nature of the match, or to cause it fail earlier than it otherwise might, when
+the author of the pattern knows there is no point in carrying on.
+.P
+Consider, for example, the pattern \ed+foo when applied to the subject line
+.sp
+  123456bar
+.sp
+After matching all 6 digits and then failing to match "foo", the normal
+action of the matcher is to try again with only 5 digits matching the \ed+
+item, and then with 4, and so on, before ultimately failing. "Atomic grouping"
+(a term taken from Jeffrey Friedl's book) provides the means for specifying
+that once a subpattern has matched, it is not to be re-evaluated in this way.
+.P
+If we use atomic grouping for the previous example, the matcher gives up
+immediately on failing to match "foo" the first time. The notation is a kind of
+special parenthesis, starting with (?> as in this example:
+.sp
+  (?>\ed+)foo
+.sp
+This kind of parenthesis "locks up" the  part of the pattern it contains once
+it has matched, and a failure further into the pattern is prevented from
+backtracking into it. Backtracking past it to previous items, however, works as
+normal.
+.P
+An alternative description is that a subpattern of this type matches the string
+of characters that an identical standalone pattern would match, if anchored at
+the current point in the subject string.
+.P
+Atomic grouping subpatterns are not capturing subpatterns. Simple cases such as
+the above example can be thought of as a maximizing repeat that must swallow
+everything it can. So, while both \ed+ and \ed+? are prepared to adjust the
+number of digits they match in order to make the rest of the pattern match,
+(?>\ed+) can only match an entire sequence of digits.
+.P
+Atomic groups in general can of course contain arbitrarily complicated
+subpatterns, and can be nested. However, when the subpattern for an atomic
+group is just a single repeated item, as in the example above, a simpler
+notation, called a "possessive quantifier" can be used. This consists of an
+additional + character following a quantifier. Using this notation, the
+previous example can be rewritten as
+.sp
+  \ed++foo
+.sp
+Note that a possessive quantifier can be used with an entire group, for
+example:
+.sp
+  (abc|xyz){2,3}+
+.sp
+Possessive quantifiers are always greedy; the setting of the PCRE_UNGREEDY
+option is ignored. They are a convenient notation for the simpler forms of
+atomic group. However, there is no difference in the meaning of a possessive
+quantifier and the equivalent atomic group, though there may be a performance
+difference; possessive quantifiers should be slightly faster.
+.P
+The possessive quantifier syntax is an extension to the Perl 5.8 syntax.
+Jeffrey Friedl originated the idea (and the name) in the first edition of his
+book. Mike McCloskey liked it, so implemented it when he built Sun's Java
+package, and PCRE copied it from there. It ultimately found its way into Perl
+at release 5.10.
+.P
+PCRE has an optimization that automatically "possessifies" certain simple
+pattern constructs. For example, the sequence A+B is treated as A++B because
+there is no point in backtracking into a sequence of A's when B must follow.
+.P
+When a pattern contains an unlimited repeat inside a subpattern that can itself
+be repeated an unlimited number of times, the use of an atomic group is the
+only way to avoid some failing matches taking a very long time indeed. The
+pattern
+.sp
+  (\eD+|<\ed+>)*[!?]
+.sp
+matches an unlimited number of substrings that either consist of non-digits, or
+digits enclosed in <>, followed by either ! or ?. When it matches, it runs
+quickly. However, if it is applied to
+.sp
+  aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+.sp
+it takes a long time before reporting failure. This is because the string can
+be divided between the internal \eD+ repeat and the external * repeat in a
+large number of ways, and all have to be tried. (The example uses [!?] rather
+than a single character at the end, because both PCRE and Perl have an
+optimization that allows for fast failure when a single character is used. They
+remember the last single character that is required for a match, and fail early
+if it is not present in the string.) If the pattern is changed so that it uses
+an atomic group, like this:
+.sp
+  ((?>\eD+)|<\ed+>)*[!?]
+.sp
+sequences of non-digits cannot be broken, and failure happens quickly.
+.
+.
+.\" HTML 
+.SH "BACK REFERENCES"
+.rs
+.sp
+Outside a character class, a backslash followed by a digit greater than 0 (and
+possibly further digits) is a back reference to a capturing subpattern earlier
+(that is, to its left) in the pattern, provided there have been that many
+previous capturing left parentheses.
+.P
+However, if the decimal number following the backslash is less than 10, it is
+always taken as a back reference, and causes an error only if there are not
+that many capturing left parentheses in the entire pattern. In other words, the
+parentheses that are referenced need not be to the left of the reference for
+numbers less than 10. A "forward back reference" of this type can make sense
+when a repetition is involved and the subpattern to the right has participated
+in an earlier iteration.
+.P
+It is not possible to have a numerical "forward back reference" to a subpattern
+whose number is 10 or more using this syntax because a sequence such as \e50 is
+interpreted as a character defined in octal. See the subsection entitled
+"Non-printing characters"
+.\" HTML 
+.\" 
+above
+.\"
+for further details of the handling of digits following a backslash. There is
+no such problem when named parentheses are used. A back reference to any
+subpattern is possible using named parentheses (see below).
+.P
+Another way of avoiding the ambiguity inherent in the use of digits following a
+backslash is to use the \eg escape sequence. This escape must be followed by an
+unsigned number or a negative number, optionally enclosed in braces. These
+examples are all identical:
+.sp
+  (ring), \e1
+  (ring), \eg1
+  (ring), \eg{1}
+.sp
+An unsigned number specifies an absolute reference without the ambiguity that
+is present in the older syntax. It is also useful when literal digits follow
+the reference. A negative number is a relative reference. Consider this
+example:
+.sp
+  (abc(def)ghi)\eg{-1}
+.sp
+The sequence \eg{-1} is a reference to the most recently started capturing
+subpattern before \eg, that is, is it equivalent to \e2 in this example.
+Similarly, \eg{-2} would be equivalent to \e1. The use of relative references
+can be helpful in long patterns, and also in patterns that are created by
+joining together fragments that contain references within themselves.
+.P
+A back reference matches whatever actually matched the capturing subpattern in
+the current subject string, rather than anything matching the subpattern
+itself (see
+.\" HTML 
+.\" 
+"Subpatterns as subroutines"
+.\"
+below for a way of doing that). So the pattern
+.sp
+  (sens|respons)e and \e1ibility
+.sp
+matches "sense and sensibility" and "response and responsibility", but not
+"sense and responsibility". If caseful matching is in force at the time of the
+back reference, the case of letters is relevant. For example,
+.sp
+  ((?i)rah)\es+\e1
+.sp
+matches "rah rah" and "RAH RAH", but not "RAH rah", even though the original
+capturing subpattern is matched caselessly.
+.P
+There are several different ways of writing back references to named
+subpatterns. The .NET syntax \ek{name} and the Perl syntax \ek or
+\ek'name' are supported, as is the Python syntax (?P=name). Perl 5.10's unified
+back reference syntax, in which \eg can be used for both numeric and named
+references, is also supported. We could rewrite the above example in any of
+the following ways:
+.sp
+  (?(?i)rah)\es+\ek
+  (?'p1'(?i)rah)\es+\ek{p1}
+  (?P(?i)rah)\es+(?P=p1)
+  (?(?i)rah)\es+\eg{p1}
+.sp
+A subpattern that is referenced by name may appear in the pattern before or
+after the reference.
+.P
+There may be more than one back reference to the same subpattern. If a
+subpattern has not actually been used in a particular match, any back
+references to it always fail by default. For example, the pattern
+.sp
+  (a|(bc))\e2
+.sp
+always fails if it starts to match "a" rather than "bc". However, if the
+PCRE_JAVASCRIPT_COMPAT option is set at compile time, a back reference to an
+unset value matches an empty string.
+.P
+Because there may be many capturing parentheses in a pattern, all digits
+following a backslash are taken as part of a potential back reference number.
+If the pattern continues with a digit character, some delimiter must be used to
+terminate the back reference. If the PCRE_EXTENDED option is set, this can be
+white space. Otherwise, the \eg{ syntax or an empty comment (see
+.\" HTML 
+.\" 
+"Comments"
+.\"
+below) can be used.
+.
+.SS "Recursive back references"
+.rs
+.sp
+A back reference that occurs inside the parentheses to which it refers fails
+when the subpattern is first used, so, for example, (a\e1) never matches.
+However, such references can be useful inside repeated subpatterns. For
+example, the pattern
+.sp
+  (a|b\e1)+
+.sp
+matches any number of "a"s and also "aba", "ababbaa" etc. At each iteration of
+the subpattern, the back reference matches the character string corresponding
+to the previous iteration. In order for this to work, the pattern must be such
+that the first iteration does not need to match the back reference. This can be
+done using alternation, as in the example above, or by a quantifier with a
+minimum of zero.
+.P
+Back references of this type cause the group that they reference to be treated
+as an
+.\" HTML 
+.\" 
+atomic group.
+.\"
+Once the whole group has been matched, a subsequent matching failure cannot
+cause backtracking into the middle of the group.
+.
+.
+.\" HTML 
+.SH ASSERTIONS
+.rs
+.sp
+An assertion is a test on the characters following or preceding the current
+matching point that does not actually consume any characters. The simple
+assertions coded as \eb, \eB, \eA, \eG, \eZ, \ez, ^ and $ are described
+.\" HTML 
+.\" 
+above.
+.\"
+.P
+More complicated assertions are coded as subpatterns. There are two kinds:
+those that look ahead of the current position in the subject string, and those
+that look behind it. An assertion subpattern is matched in the normal way,
+except that it does not cause the current matching position to be changed.
+.P
+Assertion subpatterns are not capturing subpatterns. If such an assertion
+contains capturing subpatterns within it, these are counted for the purposes of
+numbering the capturing subpatterns in the whole pattern. However, substring
+capturing is carried out only for positive assertions. (Perl sometimes, but not
+always, does do capturing in negative assertions.)
+.P
+For compatibility with Perl, assertion subpatterns may be repeated; though
+it makes no sense to assert the same thing several times, the side effect of
+capturing parentheses may occasionally be useful. In practice, there only three
+cases:
+.sp
+(1) If the quantifier is {0}, the assertion is never obeyed during matching.
+However, it may contain internal capturing parenthesized groups that are called
+from elsewhere via the
+.\" HTML 
+.\" 
+subroutine mechanism.
+.\"
+.sp
+(2) If quantifier is {0,n} where n is greater than zero, it is treated as if it
+were {0,1}. At run time, the rest of the pattern match is tried with and
+without the assertion, the order depending on the greediness of the quantifier.
+.sp
+(3) If the minimum repetition is greater than zero, the quantifier is ignored.
+The assertion is obeyed just once when encountered during matching.
+.
+.
+.SS "Lookahead assertions"
+.rs
+.sp
+Lookahead assertions start with (?= for positive assertions and (?! for
+negative assertions. For example,
+.sp
+  \ew+(?=;)
+.sp
+matches a word followed by a semicolon, but does not include the semicolon in
+the match, and
+.sp
+  foo(?!bar)
+.sp
+matches any occurrence of "foo" that is not followed by "bar". Note that the
+apparently similar pattern
+.sp
+  (?!foo)bar
+.sp
+does not find an occurrence of "bar" that is preceded by something other than
+"foo"; it finds any occurrence of "bar" whatsoever, because the assertion
+(?!foo) is always true when the next three characters are "bar". A
+lookbehind assertion is needed to achieve the other effect.
+.P
+If you want to force a matching failure at some point in a pattern, the most
+convenient way to do it is with (?!) because an empty string always matches, so
+an assertion that requires there not to be an empty string must always fail.
+The backtracking control verb (*FAIL) or (*F) is a synonym for (?!).
+.
+.
+.\" HTML 
+.SS "Lookbehind assertions"
+.rs
+.sp
+Lookbehind assertions start with (?<= for positive assertions and (?
+.\" 
+(see above)
+.\"
+can be used instead of a lookbehind assertion to get round the fixed-length
+restriction.
+.P
+The implementation of lookbehind assertions is, for each alternative, to
+temporarily move the current position back by the fixed length and then try to
+match. If there are insufficient characters before the current position, the
+assertion fails.
+.P
+In a UTF mode, PCRE does not allow the \eC escape (which matches a single data
+unit even in a UTF mode) to appear in lookbehind assertions, because it makes
+it impossible to calculate the length of the lookbehind. The \eX and \eR
+escapes, which can match different numbers of data units, are also not
+permitted.
+.P
+.\" HTML 
+.\" 
+"Subroutine"
+.\"
+calls (see below) such as (?2) or (?&X) are permitted in lookbehinds, as long
+as the subpattern matches a fixed-length string.
+.\" HTML 
+.\" 
+Recursion,
+.\"
+however, is not supported.
+.P
+Possessive quantifiers can be used in conjunction with lookbehind assertions to
+specify efficient matching of fixed-length strings at the end of subject
+strings. Consider a simple pattern such as
+.sp
+  abcd$
+.sp
+when applied to a long string that does not match. Because matching proceeds
+from left to right, PCRE will look for each "a" in the subject and then see if
+what follows matches the rest of the pattern. If the pattern is specified as
+.sp
+  ^.*abcd$
+.sp
+the initial .* matches the entire string at first, but when this fails (because
+there is no following "a"), it backtracks to match all but the last character,
+then all but the last two characters, and so on. Once again the search for "a"
+covers the entire string, from right to left, so we are no better off. However,
+if the pattern is written as
+.sp
+  ^.*+(?<=abcd)
+.sp
+there can be no backtracking for the .*+ item; it can match only the entire
+string. The subsequent lookbehind assertion does a single test on the last four
+characters. If it fails, the match fails immediately. For long strings, this
+approach makes a significant difference to the processing time.
+.
+.
+.SS "Using multiple assertions"
+.rs
+.sp
+Several assertions (of any sort) may occur in succession. For example,
+.sp
+  (?<=\ed{3})(?
+.SH "CONDITIONAL SUBPATTERNS"
+.rs
+.sp
+It is possible to cause the matching process to obey a subpattern
+conditionally or to choose between two alternative subpatterns, depending on
+the result of an assertion, or whether a specific capturing subpattern has
+already been matched. The two possible forms of conditional subpattern are:
+.sp
+  (?(condition)yes-pattern)
+  (?(condition)yes-pattern|no-pattern)
+.sp
+If the condition is satisfied, the yes-pattern is used; otherwise the
+no-pattern (if present) is used. If there are more than two alternatives in the
+subpattern, a compile-time error occurs. Each of the two alternatives may
+itself contain nested subpatterns of any form, including conditional
+subpatterns; the restriction to two alternatives applies only at the level of
+the condition. This pattern fragment is an example where the alternatives are
+complex:
+.sp
+  (?(1) (A|B|C) | (D | (?(2)E|F) | E) )
+.sp
+.P
+There are four kinds of condition: references to subpatterns, references to
+recursion, a pseudo-condition called DEFINE, and assertions.
+.
+.SS "Checking for a used subpattern by number"
+.rs
+.sp
+If the text between the parentheses consists of a sequence of digits, the
+condition is true if a capturing subpattern of that number has previously
+matched. If there is more than one capturing subpattern with the same number
+(see the earlier
+.\"
+.\" HTML 
+.\" 
+section about duplicate subpattern numbers),
+.\"
+the condition is true if any of them have matched. An alternative notation is
+to precede the digits with a plus or minus sign. In this case, the subpattern
+number is relative rather than absolute. The most recently opened parentheses
+can be referenced by (?(-1), the next most recent by (?(-2), and so on. Inside
+loops it can also make sense to refer to subsequent groups. The next
+parentheses to be opened can be referenced as (?(+1), and so on. (The value
+zero in any of these forms is not used; it provokes a compile-time error.)
+.P
+Consider the following pattern, which contains non-significant white space to
+make it more readable (assume the PCRE_EXTENDED option) and to divide it into
+three parts for ease of discussion:
+.sp
+  ( \e( )?    [^()]+    (?(1) \e) )
+.sp
+The first part matches an optional opening parenthesis, and if that
+character is present, sets it as the first captured substring. The second part
+matches one or more characters that are not parentheses. The third part is a
+conditional subpattern that tests whether or not the first set of parentheses
+matched. If they did, that is, if subject started with an opening parenthesis,
+the condition is true, and so the yes-pattern is executed and a closing
+parenthesis is required. Otherwise, since no-pattern is not present, the
+subpattern matches nothing. In other words, this pattern matches a sequence of
+non-parentheses, optionally enclosed in parentheses.
+.P
+If you were embedding this pattern in a larger one, you could use a relative
+reference:
+.sp
+  ...other stuff... ( \e( )?    [^()]+    (?(-1) \e) ) ...
+.sp
+This makes the fragment independent of the parentheses in the larger pattern.
+.
+.SS "Checking for a used subpattern by name"
+.rs
+.sp
+Perl uses the syntax (?()...) or (?('name')...) to test for a used
+subpattern by name. For compatibility with earlier versions of PCRE, which had
+this facility before Perl, the syntax (?(name)...) is also recognized. However,
+there is a possible ambiguity with this syntax, because subpattern names may
+consist entirely of digits. PCRE looks first for a named subpattern; if it
+cannot find one and the name consists entirely of digits, PCRE looks for a
+subpattern of that number, which must be greater than zero. Using subpattern
+names that consist entirely of digits is not recommended.
+.P
+Rewriting the above example to use a named subpattern gives this:
+.sp
+  (? \e( )?    [^()]+    (?() \e) )
+.sp
+If the name used in a condition of this kind is a duplicate, the test is
+applied to all subpatterns of the same name, and is true if any one of them has
+matched.
+.
+.SS "Checking for pattern recursion"
+.rs
+.sp
+If the condition is the string (R), and there is no subpattern with the name R,
+the condition is true if a recursive call to the whole pattern or any
+subpattern has been made. If digits or a name preceded by ampersand follow the
+letter R, for example:
+.sp
+  (?(R3)...) or (?(R&name)...)
+.sp
+the condition is true if the most recent recursion is into a subpattern whose
+number or name is given. This condition does not check the entire recursion
+stack. If the name used in a condition of this kind is a duplicate, the test is
+applied to all subpatterns of the same name, and is true if any one of them is
+the most recent recursion.
+.P
+At "top level", all these recursion test conditions are false.
+.\" HTML 
+.\" 
+The syntax for recursive patterns
+.\"
+is described below.
+.
+.\" HTML 
+.SS "Defining subpatterns for use by reference only"
+.rs
+.sp
+If the condition is the string (DEFINE), and there is no subpattern with the
+name DEFINE, the condition is always false. In this case, there may be only one
+alternative in the subpattern. It is always skipped if control reaches this
+point in the pattern; the idea of DEFINE is that it can be used to define
+subroutines that can be referenced from elsewhere. (The use of
+.\" HTML 
+.\" 
+subroutines
+.\"
+is described below.) For example, a pattern to match an IPv4 address such as
+"192.168.23.245" could be written like this (ignore white space and line
+breaks):
+.sp
+  (?(DEFINE) (? 2[0-4]\ed | 25[0-5] | 1\ed\ed | [1-9]?\ed) )
+  \eb (?&byte) (\e.(?&byte)){3} \eb
+.sp
+The first part of the pattern is a DEFINE group inside which a another group
+named "byte" is defined. This matches an individual component of an IPv4
+address (a number less than 256). When matching takes place, this part of the
+pattern is skipped because DEFINE acts like a false condition. The rest of the
+pattern uses references to the named group to match the four dot-separated
+components of an IPv4 address, insisting on a word boundary at each end.
+.
+.SS "Assertion conditions"
+.rs
+.sp
+If the condition is not in any of the above formats, it must be an assertion.
+This may be a positive or negative lookahead or lookbehind assertion. Consider
+this pattern, again containing non-significant white space, and with the two
+alternatives on the second line:
+.sp
+  (?(?=[^a-z]*[a-z])
+  \ed{2}-[a-z]{3}-\ed{2}  |  \ed{2}-\ed{2}-\ed{2} )
+.sp
+The condition is a positive lookahead assertion that matches an optional
+sequence of non-letters followed by a letter. In other words, it tests for the
+presence of at least one letter in the subject. If a letter is found, the
+subject is matched against the first alternative; otherwise it is matched
+against the second. This pattern matches strings in one of the two forms
+dd-aaa-dd or dd-dd-dd, where aaa are letters and dd are digits.
+.
+.
+.\" HTML 
+.SH COMMENTS
+.rs
+.sp
+There are two ways of including comments in patterns that are processed by
+PCRE. In both cases, the start of the comment must not be in a character class,
+nor in the middle of any other sequence of related characters such as (?: or a
+subpattern name or number. The characters that make up a comment play no part
+in the pattern matching.
+.P
+The sequence (?# marks the start of a comment that continues up to the next
+closing parenthesis. Nested parentheses are not permitted. If the PCRE_EXTENDED
+option is set, an unescaped # character also introduces a comment, which in
+this case continues to immediately after the next newline character or
+character sequence in the pattern. Which characters are interpreted as newlines
+is controlled by the options passed to a compiling function or by a special
+sequence at the start of the pattern, as described in the section entitled
+.\" HTML 
+.\" 
+"Newline conventions"
+.\"
+above. Note that the end of this type of comment is a literal newline sequence
+in the pattern; escape sequences that happen to represent a newline do not
+count. For example, consider this pattern when PCRE_EXTENDED is set, and the
+default newline convention is in force:
+.sp
+  abc #comment \en still comment
+.sp
+On encountering the # character, \fBpcre_compile()\fP skips along, looking for
+a newline in the pattern. The sequence \en is still literal at this stage, so
+it does not terminate the comment. Only an actual character with the code value
+0x0a (the default newline) does so.
+.
+.
+.\" HTML 
+.SH "RECURSIVE PATTERNS"
+.rs
+.sp
+Consider the problem of matching a string in parentheses, allowing for
+unlimited nested parentheses. Without the use of recursion, the best that can
+be done is to use a pattern that matches up to some fixed depth of nesting. It
+is not possible to handle an arbitrary nesting depth.
+.P
+For some time, Perl has provided a facility that allows regular expressions to
+recurse (amongst other things). It does this by interpolating Perl code in the
+expression at run time, and the code can refer to the expression itself. A Perl
+pattern using code interpolation to solve the parentheses problem can be
+created like this:
+.sp
+  $re = qr{\e( (?: (?>[^()]+) | (?p{$re}) )* \e)}x;
+.sp
+The (?p{...}) item interpolates Perl code at run time, and in this case refers
+recursively to the pattern in which it appears.
+.P
+Obviously, PCRE cannot support the interpolation of Perl code. Instead, it
+supports special syntax for recursion of the entire pattern, and also for
+individual subpattern recursion. After its introduction in PCRE and Python,
+this kind of recursion was subsequently introduced into Perl at release 5.10.
+.P
+A special item that consists of (? followed by a number greater than zero and a
+closing parenthesis is a recursive subroutine call of the subpattern of the
+given number, provided that it occurs inside that subpattern. (If not, it is a
+.\" HTML 
+.\" 
+non-recursive subroutine
+.\"
+call, which is described in the next section.) The special item (?R) or (?0) is
+a recursive call of the entire regular expression.
+.P
+This PCRE pattern solves the nested parentheses problem (assume the
+PCRE_EXTENDED option is set so that white space is ignored):
+.sp
+  \e( ( [^()]++ | (?R) )* \e)
+.sp
+First it matches an opening parenthesis. Then it matches any number of
+substrings which can either be a sequence of non-parentheses, or a recursive
+match of the pattern itself (that is, a correctly parenthesized substring).
+Finally there is a closing parenthesis. Note the use of a possessive quantifier
+to avoid backtracking into sequences of non-parentheses.
+.P
+If this were part of a larger pattern, you would not want to recurse the entire
+pattern, so instead you could use this:
+.sp
+  ( \e( ( [^()]++ | (?1) )* \e) )
+.sp
+We have put the pattern into parentheses, and caused the recursion to refer to
+them instead of the whole pattern.
+.P
+In a larger pattern, keeping track of parenthesis numbers can be tricky. This
+is made easier by the use of relative references. Instead of (?1) in the
+pattern above you can write (?-2) to refer to the second most recently opened
+parentheses preceding the recursion. In other words, a negative number counts
+capturing parentheses leftwards from the point at which it is encountered.
+.P
+It is also possible to refer to subsequently opened parentheses, by writing
+references such as (?+2). However, these cannot be recursive because the
+reference is not inside the parentheses that are referenced. They are always
+.\" HTML 
+.\" 
+non-recursive subroutine
+.\"
+calls, as described in the next section.
+.P
+An alternative approach is to use named parentheses instead. The Perl syntax
+for this is (?&name); PCRE's earlier syntax (?P>name) is also supported. We
+could rewrite the above example as follows:
+.sp
+  (? \e( ( [^()]++ | (?&pn) )* \e) )
+.sp
+If there is more than one subpattern with the same name, the earliest one is
+used.
+.P
+This particular example pattern that we have been looking at contains nested
+unlimited repeats, and so the use of a possessive quantifier for matching
+strings of non-parentheses is important when applying the pattern to strings
+that do not match. For example, when this pattern is applied to
+.sp
+  (aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa()
+.sp
+it yields "no match" quickly. However, if a possessive quantifier is not used,
+the match runs for a very long time indeed because there are so many different
+ways the + and * repeats can carve up the subject, and all have to be tested
+before failure can be reported.
+.P
+At the end of a match, the values of capturing parentheses are those from
+the outermost level. If you want to obtain intermediate values, a callout
+function can be used (see below and the
+.\" HREF
+\fBpcrecallout\fP
+.\"
+documentation). If the pattern above is matched against
+.sp
+  (ab(cd)ef)
+.sp
+the value for the inner capturing parentheses (numbered 2) is "ef", which is
+the last value taken on at the top level. If a capturing subpattern is not
+matched at the top level, its final captured value is unset, even if it was
+(temporarily) set at a deeper level during the matching process.
+.P
+If there are more than 15 capturing parentheses in a pattern, PCRE has to
+obtain extra memory to store data during a recursion, which it does by using
+\fBpcre_malloc\fP, freeing it via \fBpcre_free\fP afterwards. If no memory can
+be obtained, the match fails with the PCRE_ERROR_NOMEMORY error.
+.P
+Do not confuse the (?R) item with the condition (R), which tests for recursion.
+Consider this pattern, which matches text in angle brackets, allowing for
+arbitrary nesting. Only digits are allowed in nested brackets (that is, when
+recursing), whereas any characters are permitted at the outer level.
+.sp
+  < (?: (?(R) \ed++  | [^<>]*+) | (?R)) * >
+.sp
+In this pattern, (?(R) is the start of a conditional subpattern, with two
+different alternatives for the recursive and non-recursive cases. The (?R) item
+is the actual recursive call.
+.
+.
+.\" HTML 
+.SS "Differences in recursion processing between PCRE and Perl"
+.rs
+.sp
+Recursion processing in PCRE differs from Perl in two important ways. In PCRE
+(like Python, but unlike Perl), a recursive subpattern call is always treated
+as an atomic group. That is, once it has matched some of the subject string, it
+is never re-entered, even if it contains untried alternatives and there is a
+subsequent matching failure. This can be illustrated by the following pattern,
+which purports to match a palindromic string that contains an odd number of
+characters (for example, "a", "aba", "abcba", "abcdcba"):
+.sp
+  ^(.|(.)(?1)\e2)$
+.sp
+The idea is that it either matches a single character, or two identical
+characters surrounding a sub-palindrome. In Perl, this pattern works; in PCRE
+it does not if the pattern is longer than three characters. Consider the
+subject string "abcba":
+.P
+At the top level, the first character is matched, but as it is not at the end
+of the string, the first alternative fails; the second alternative is taken
+and the recursion kicks in. The recursive call to subpattern 1 successfully
+matches the next character ("b"). (Note that the beginning and end of line
+tests are not part of the recursion).
+.P
+Back at the top level, the next character ("c") is compared with what
+subpattern 2 matched, which was "a". This fails. Because the recursion is
+treated as an atomic group, there are now no backtracking points, and so the
+entire match fails. (Perl is able, at this point, to re-enter the recursion and
+try the second alternative.) However, if the pattern is written with the
+alternatives in the other order, things are different:
+.sp
+  ^((.)(?1)\e2|.)$
+.sp
+This time, the recursing alternative is tried first, and continues to recurse
+until it runs out of characters, at which point the recursion fails. But this
+time we do have another alternative to try at the higher level. That is the big
+difference: in the previous case the remaining alternative is at a deeper
+recursion level, which PCRE cannot use.
+.P
+To change the pattern so that it matches all palindromic strings, not just
+those with an odd number of characters, it is tempting to change the pattern to
+this:
+.sp
+  ^((.)(?1)\e2|.?)$
+.sp
+Again, this works in Perl, but not in PCRE, and for the same reason. When a
+deeper recursion has matched a single character, it cannot be entered again in
+order to match an empty string. The solution is to separate the two cases, and
+write out the odd and even cases as alternatives at the higher level:
+.sp
+  ^(?:((.)(?1)\e2|)|((.)(?3)\e4|.))
+.sp
+If you want to match typical palindromic phrases, the pattern has to ignore all
+non-word characters, which can be done like this:
+.sp
+  ^\eW*+(?:((.)\eW*+(?1)\eW*+\e2|)|((.)\eW*+(?3)\eW*+\e4|\eW*+.\eW*+))\eW*+$
+.sp
+If run with the PCRE_CASELESS option, this pattern matches phrases such as "A
+man, a plan, a canal: Panama!" and it works well in both PCRE and Perl. Note
+the use of the possessive quantifier *+ to avoid backtracking into sequences of
+non-word characters. Without this, PCRE takes a great deal longer (ten times or
+more) to match typical phrases, and Perl takes so long that you think it has
+gone into a loop.
+.P
+\fBWARNING\fP: The palindrome-matching patterns above work only if the subject
+string does not start with a palindrome that is shorter than the entire string.
+For example, although "abcba" is correctly matched, if the subject is "ababa",
+PCRE finds the palindrome "aba" at the start, then fails at top level because
+the end of the string does not follow. Once again, it cannot jump back into the
+recursion to try other alternatives, so the entire match fails.
+.P
+The second way in which PCRE and Perl differ in their recursion processing is
+in the handling of captured values. In Perl, when a subpattern is called
+recursively or as a subpattern (see the next section), it has no access to any
+values that were captured outside the recursion, whereas in PCRE these values
+can be referenced. Consider this pattern:
+.sp
+  ^(.)(\e1|a(?2))
+.sp
+In PCRE, this pattern matches "bab". The first capturing parentheses match "b",
+then in the second group, when the back reference \e1 fails to match "b", the
+second alternative matches "a" and then recurses. In the recursion, \e1 does
+now match "b" and so the whole match succeeds. In Perl, the pattern fails to
+match because inside the recursive call \e1 cannot access the externally set
+value.
+.
+.
+.\" HTML 
+.SH "SUBPATTERNS AS SUBROUTINES"
+.rs
+.sp
+If the syntax for a recursive subpattern call (either by number or by
+name) is used outside the parentheses to which it refers, it operates like a
+subroutine in a programming language. The called subpattern may be defined
+before or after the reference. A numbered reference can be absolute or
+relative, as in these examples:
+.sp
+  (...(absolute)...)...(?2)...
+  (...(relative)...)...(?-1)...
+  (...(?+1)...(relative)...
+.sp
+An earlier example pointed out that the pattern
+.sp
+  (sens|respons)e and \e1ibility
+.sp
+matches "sense and sensibility" and "response and responsibility", but not
+"sense and responsibility". If instead the pattern
+.sp
+  (sens|respons)e and (?1)ibility
+.sp
+is used, it does match "sense and responsibility" as well as the other two
+strings. Another example is given in the discussion of DEFINE above.
+.P
+All subroutine calls, whether recursive or not, are always treated as atomic
+groups. That is, once a subroutine has matched some of the subject string, it
+is never re-entered, even if it contains untried alternatives and there is a
+subsequent matching failure. Any capturing parentheses that are set during the
+subroutine call revert to their previous values afterwards.
+.P
+Processing options such as case-independence are fixed when a subpattern is
+defined, so if it is used as a subroutine, such options cannot be changed for
+different calls. For example, consider this pattern:
+.sp
+  (abc)(?i:(?-1))
+.sp
+It matches "abcabc". It does not match "abcABC" because the change of
+processing option does not affect the called subpattern.
+.
+.
+.\" HTML 
+.SH "ONIGURUMA SUBROUTINE SYNTAX"
+.rs
+.sp
+For compatibility with Oniguruma, the non-Perl syntax \eg followed by a name or
+a number enclosed either in angle brackets or single quotes, is an alternative
+syntax for referencing a subpattern as a subroutine, possibly recursively. Here
+are two of the examples used above, rewritten using this syntax:
+.sp
+  (? \e( ( (?>[^()]+) | \eg )* \e) )
+  (sens|respons)e and \eg'1'ibility
+.sp
+PCRE supports an extension to Oniguruma: if a number is preceded by a
+plus or a minus sign it is taken as a relative reference. For example:
+.sp
+  (abc)(?i:\eg<-1>)
+.sp
+Note that \eg{...} (Perl syntax) and \eg<...> (Oniguruma syntax) are \fInot\fP
+synonymous. The former is a back reference; the latter is a subroutine call.
+.
+.
+.SH CALLOUTS
+.rs
+.sp
+Perl has a feature whereby using the sequence (?{...}) causes arbitrary Perl
+code to be obeyed in the middle of matching a regular expression. This makes it
+possible, amongst other things, to extract different substrings that match the
+same pair of parentheses when there is a repetition.
+.P
+PCRE provides a similar feature, but of course it cannot obey arbitrary Perl
+code. The feature is called "callout". The caller of PCRE provides an external
+function by putting its entry point in the global variable \fIpcre_callout\fP
+(8-bit library) or \fIpcre[16|32]_callout\fP (16-bit or 32-bit library).
+By default, this variable contains NULL, which disables all calling out.
+.P
+Within a regular expression, (?C) indicates the points at which the external
+function is to be called. If you want to identify different callout points, you
+can put a number less than 256 after the letter C. The default value is zero.
+For example, this pattern has two callout points:
+.sp
+  (?C1)abc(?C2)def
+.sp
+If the PCRE_AUTO_CALLOUT flag is passed to a compiling function, callouts are
+automatically installed before each item in the pattern. They are all numbered
+255. If there is a conditional group in the pattern whose condition is an
+assertion, an additional callout is inserted just before the condition. An
+explicit callout may also be set at this position, as in this example:
+.sp
+  (?(?C9)(?=a)abc|def)
+.sp
+Note that this applies only to assertion conditions, not to other types of
+condition.
+.P
+During matching, when PCRE reaches a callout point, the external function is
+called. It is provided with the number of the callout, the position in the
+pattern, and, optionally, one item of data originally supplied by the caller of
+the matching function. The callout function may cause matching to proceed, to
+backtrack, or to fail altogether. A complete description of the interface to
+the callout function is given in the
+.\" HREF
+\fBpcrecallout\fP
+.\"
+documentation.
+.
+.
+.\" HTML 
+.SH "BACKTRACKING CONTROL"
+.rs
+.sp
+Perl 5.10 introduced a number of "Special Backtracking Control Verbs", which
+are still described in the Perl documentation as "experimental and subject to
+change or removal in a future version of Perl". It goes on to say: "Their usage
+in production code should be noted to avoid problems during upgrades." The same
+remarks apply to the PCRE features described in this section.
+.P
+The new verbs make use of what was previously invalid syntax: an opening
+parenthesis followed by an asterisk. They are generally of the form
+(*VERB) or (*VERB:NAME). Some may take either form, possibly behaving
+differently depending on whether or not a name is present. A name is any
+sequence of characters that does not include a closing parenthesis. The maximum
+length of name is 255 in the 8-bit library and 65535 in the 16-bit and 32-bit
+libraries. If the name is empty, that is, if the closing parenthesis
+immediately follows the colon, the effect is as if the colon were not there.
+Any number of these verbs may occur in a pattern.
+.P
+Since these verbs are specifically related to backtracking, most of them can be
+used only when the pattern is to be matched using one of the traditional
+matching functions, because these use a backtracking algorithm. With the
+exception of (*FAIL), which behaves like a failing negative assertion, the
+backtracking control verbs cause an error if encountered by a DFA matching
+function.
+.P
+The behaviour of these verbs in
+.\" HTML 
+.\" 
+repeated groups,
+.\"
+.\" HTML 
+.\" 
+assertions,
+.\"
+and in
+.\" HTML 
+.\" 
+subpatterns called as subroutines
+.\"
+(whether or not recursively) is documented below.
+.
+.
+.\" HTML 
+.SS "Optimizations that affect backtracking verbs"
+.rs
+.sp
+PCRE contains some optimizations that are used to speed up matching by running
+some checks at the start of each match attempt. For example, it may know the
+minimum length of matching subject, or that a particular character must be
+present. When one of these optimizations bypasses the running of a match, any
+included backtracking verbs will not, of course, be processed. You can suppress
+the start-of-match optimizations by setting the PCRE_NO_START_OPTIMIZE option
+when calling \fBpcre_compile()\fP or \fBpcre_exec()\fP, or by starting the
+pattern with (*NO_START_OPT). There is more discussion of this option in the
+section entitled
+.\" HTML 
+.\" 
+"Option bits for \fBpcre_exec()\fP"
+.\"
+in the
+.\" HREF
+\fBpcreapi\fP
+.\"
+documentation.
+.P
+Experiments with Perl suggest that it too has similar optimizations, sometimes
+leading to anomalous results.
+.
+.
+.SS "Verbs that act immediately"
+.rs
+.sp
+The following verbs act as soon as they are encountered. They may not be
+followed by a name.
+.sp
+   (*ACCEPT)
+.sp
+This verb causes the match to end successfully, skipping the remainder of the
+pattern. However, when it is inside a subpattern that is called as a
+subroutine, only that subpattern is ended successfully. Matching then continues
+at the outer level. If (*ACCEPT) in triggered in a positive assertion, the
+assertion succeeds; in a negative assertion, the assertion fails.
+.P
+If (*ACCEPT) is inside capturing parentheses, the data so far is captured. For
+example:
+.sp
+  A((?:A|B(*ACCEPT)|C)D)
+.sp
+This matches "AB", "AAD", or "ACD"; when it matches "AB", "B" is captured by
+the outer parentheses.
+.sp
+  (*FAIL) or (*F)
+.sp
+This verb causes a matching failure, forcing backtracking to occur. It is
+equivalent to (?!) but easier to read. The Perl documentation notes that it is
+probably useful only when combined with (?{}) or (??{}). Those are, of course,
+Perl features that are not present in PCRE. The nearest equivalent is the
+callout feature, as for example in this pattern:
+.sp
+  a+(?C)(*FAIL)
+.sp
+A match with the string "aaaa" always fails, but the callout is taken before
+each backtrack happens (in this example, 10 times).
+.
+.
+.SS "Recording which path was taken"
+.rs
+.sp
+There is one verb whose main purpose is to track how a match was arrived at,
+though it also has a secondary use in conjunction with advancing the match
+starting point (see (*SKIP) below).
+.sp
+  (*MARK:NAME) or (*:NAME)
+.sp
+A name is always required with this verb. There may be as many instances of
+(*MARK) as you like in a pattern, and their names do not have to be unique.
+.P
+When a match succeeds, the name of the last-encountered (*MARK:NAME),
+(*PRUNE:NAME), or (*THEN:NAME) on the matching path is passed back to the
+caller as described in the section entitled
+.\" HTML 
+.\" 
+"Extra data for \fBpcre_exec()\fP"
+.\"
+in the
+.\" HREF
+\fBpcreapi\fP
+.\"
+documentation. Here is an example of \fBpcretest\fP output, where the /K
+modifier requests the retrieval and outputting of (*MARK) data:
+.sp
+    re> /X(*MARK:A)Y|X(*MARK:B)Z/K
+  data> XY
+   0: XY
+  MK: A
+  XZ
+   0: XZ
+  MK: B
+.sp
+The (*MARK) name is tagged with "MK:" in this output, and in this example it
+indicates which of the two alternatives matched. This is a more efficient way
+of obtaining this information than putting each alternative in its own
+capturing parentheses.
+.P
+If a verb with a name is encountered in a positive assertion that is true, the
+name is recorded and passed back if it is the last-encountered. This does not
+happen for negative assertions or failing positive assertions.
+.P
+After a partial match or a failed match, the last encountered name in the
+entire match process is returned. For example:
+.sp
+    re> /X(*MARK:A)Y|X(*MARK:B)Z/K
+  data> XP
+  No match, mark = B
+.sp
+Note that in this unanchored example the mark is retained from the match
+attempt that started at the letter "X" in the subject. Subsequent match
+attempts starting at "P" and then with an empty string do not get as far as the
+(*MARK) item, but nevertheless do not reset it.
+.P
+If you are interested in (*MARK) values after failed matches, you should
+probably set the PCRE_NO_START_OPTIMIZE option
+.\" HTML 
+.\" 
+(see above)
+.\"
+to ensure that the match is always attempted.
+.
+.
+.SS "Verbs that act after backtracking"
+.rs
+.sp
+The following verbs do nothing when they are encountered. Matching continues
+with what follows, but if there is no subsequent match, causing a backtrack to
+the verb, a failure is forced. That is, backtracking cannot pass to the left of
+the verb. However, when one of these verbs appears inside an atomic group or an
+assertion that is true, its effect is confined to that group, because once the
+group has been matched, there is never any backtracking into it. In this
+situation, backtracking can "jump back" to the left of the entire atomic group
+or assertion. (Remember also, as stated above, that this localization also
+applies in subroutine calls.)
+.P
+These verbs differ in exactly what kind of failure occurs when backtracking
+reaches them. The behaviour described below is what happens when the verb is
+not in a subroutine or an assertion. Subsequent sections cover these special
+cases.
+.sp
+  (*COMMIT)
+.sp
+This verb, which may not be followed by a name, causes the whole match to fail
+outright if there is a later matching failure that causes backtracking to reach
+it. Even if the pattern is unanchored, no further attempts to find a match by
+advancing the starting point take place. If (*COMMIT) is the only backtracking
+verb that is encountered, once it has been passed \fBpcre_exec()\fP is
+committed to finding a match at the current starting point, or not at all. For
+example:
+.sp
+  a+(*COMMIT)b
+.sp
+This matches "xxaab" but not "aacaab". It can be thought of as a kind of
+dynamic anchor, or "I've started, so I must finish." The name of the most
+recently passed (*MARK) in the path is passed back when (*COMMIT) forces a
+match failure.
+.P
+If there is more than one backtracking verb in a pattern, a different one that
+follows (*COMMIT) may be triggered first, so merely passing (*COMMIT) during a
+match does not always guarantee that a match must be at this starting point.
+.P
+Note that (*COMMIT) at the start of a pattern is not the same as an anchor,
+unless PCRE's start-of-match optimizations are turned off, as shown in this
+\fBpcretest\fP example:
+.sp
+    re> /(*COMMIT)abc/
+  data> xyzabc
+   0: abc
+  xyzabc\eY
+  No match
+.sp
+PCRE knows that any match must start with "a", so the optimization skips along
+the subject to "a" before running the first match attempt, which succeeds. When
+the optimization is disabled by the \eY escape in the second subject, the match
+starts at "x" and so the (*COMMIT) causes it to fail without trying any other
+starting points.
+.sp
+  (*PRUNE) or (*PRUNE:NAME)
+.sp
+This verb causes the match to fail at the current starting position in the
+subject if there is a later matching failure that causes backtracking to reach
+it. If the pattern is unanchored, the normal "bumpalong" advance to the next
+starting character then happens. Backtracking can occur as usual to the left of
+(*PRUNE), before it is reached, or when matching to the right of (*PRUNE), but
+if there is no match to the right, backtracking cannot cross (*PRUNE). In
+simple cases, the use of (*PRUNE) is just an alternative to an atomic group or
+possessive quantifier, but there are some uses of (*PRUNE) that cannot be
+expressed in any other way. In an anchored pattern (*PRUNE) has the same effect
+as (*COMMIT).
+.P
+The behaviour of (*PRUNE:NAME) is the not the same as (*MARK:NAME)(*PRUNE).
+It is like (*MARK:NAME) in that the name is remembered for passing back to the
+caller. However, (*SKIP:NAME) searches only for names set with (*MARK).
+.sp
+  (*SKIP)
+.sp
+This verb, when given without a name, is like (*PRUNE), except that if the
+pattern is unanchored, the "bumpalong" advance is not to the next character,
+but to the position in the subject where (*SKIP) was encountered. (*SKIP)
+signifies that whatever text was matched leading up to it cannot be part of a
+successful match. Consider:
+.sp
+  a+(*SKIP)b
+.sp
+If the subject is "aaaac...", after the first match attempt fails (starting at
+the first character in the string), the starting point skips on to start the
+next attempt at "c". Note that a possessive quantifer does not have the same
+effect as this example; although it would suppress backtracking during the
+first match attempt, the second attempt would start at the second character
+instead of skipping on to "c".
+.sp
+  (*SKIP:NAME)
+.sp
+When (*SKIP) has an associated name, its behaviour is modified. When it is
+triggered, the previous path through the pattern is searched for the most
+recent (*MARK) that has the same name. If one is found, the "bumpalong" advance
+is to the subject position that corresponds to that (*MARK) instead of to where
+(*SKIP) was encountered. If no (*MARK) with a matching name is found, the
+(*SKIP) is ignored.
+.P
+Note that (*SKIP:NAME) searches only for names set by (*MARK:NAME). It ignores
+names that are set by (*PRUNE:NAME) or (*THEN:NAME).
+.sp
+  (*THEN) or (*THEN:NAME)
+.sp
+This verb causes a skip to the next innermost alternative when backtracking
+reaches it. That is, it cancels any further backtracking within the current
+alternative. Its name comes from the observation that it can be used for a
+pattern-based if-then-else block:
+.sp
+  ( COND1 (*THEN) FOO | COND2 (*THEN) BAR | COND3 (*THEN) BAZ ) ...
+.sp
+If the COND1 pattern matches, FOO is tried (and possibly further items after
+the end of the group if FOO succeeds); on failure, the matcher skips to the
+second alternative and tries COND2, without backtracking into COND1. If that
+succeeds and BAR fails, COND3 is tried. If subsequently BAZ fails, there are no
+more alternatives, so there is a backtrack to whatever came before the entire
+group. If (*THEN) is not inside an alternation, it acts like (*PRUNE).
+.P
+The behaviour of (*THEN:NAME) is the not the same as (*MARK:NAME)(*THEN).
+It is like (*MARK:NAME) in that the name is remembered for passing back to the
+caller. However, (*SKIP:NAME) searches only for names set with (*MARK).
+.P
+A subpattern that does not contain a | character is just a part of the
+enclosing alternative; it is not a nested alternation with only one
+alternative. The effect of (*THEN) extends beyond such a subpattern to the
+enclosing alternative. Consider this pattern, where A, B, etc. are complex
+pattern fragments that do not contain any | characters at this level:
+.sp
+  A (B(*THEN)C) | D
+.sp
+If A and B are matched, but there is a failure in C, matching does not
+backtrack into A; instead it moves to the next alternative, that is, D.
+However, if the subpattern containing (*THEN) is given an alternative, it
+behaves differently:
+.sp
+  A (B(*THEN)C | (*FAIL)) | D
+.sp
+The effect of (*THEN) is now confined to the inner subpattern. After a failure
+in C, matching moves to (*FAIL), which causes the whole subpattern to fail
+because there are no more alternatives to try. In this case, matching does now
+backtrack into A.
+.P
+Note that a conditional subpattern is not considered as having two
+alternatives, because only one is ever used. In other words, the | character in
+a conditional subpattern has a different meaning. Ignoring white space,
+consider:
+.sp
+  ^.*? (?(?=a) a | b(*THEN)c )
+.sp
+If the subject is "ba", this pattern does not match. Because .*? is ungreedy,
+it initially matches zero characters. The condition (?=a) then fails, the
+character "b" is matched, but "c" is not. At this point, matching does not
+backtrack to .*? as might perhaps be expected from the presence of the |
+character. The conditional subpattern is part of the single alternative that
+comprises the whole pattern, and so the match fails. (If there was a backtrack
+into .*?, allowing it to match "b", the match would succeed.)
+.P
+The verbs just described provide four different "strengths" of control when
+subsequent matching fails. (*THEN) is the weakest, carrying on the match at the
+next alternative. (*PRUNE) comes next, failing the match at the current
+starting position, but allowing an advance to the next character (for an
+unanchored pattern). (*SKIP) is similar, except that the advance may be more
+than one character. (*COMMIT) is the strongest, causing the entire match to
+fail.
+.
+.
+.SS "More than one backtracking verb"
+.rs
+.sp
+If more than one backtracking verb is present in a pattern, the one that is
+backtracked onto first acts. For example, consider this pattern, where A, B,
+etc. are complex pattern fragments:
+.sp
+  (A(*COMMIT)B(*THEN)C|ABD)
+.sp
+If A matches but B fails, the backtrack to (*COMMIT) causes the entire match to
+fail. However, if A and B match, but C fails, the backtrack to (*THEN) causes
+the next alternative (ABD) to be tried. This behaviour is consistent, but is
+not always the same as Perl's. It means that if two or more backtracking verbs
+appear in succession, all the the last of them has no effect. Consider this
+example:
+.sp
+  ...(*COMMIT)(*PRUNE)...
+.sp
+If there is a matching failure to the right, backtracking onto (*PRUNE) cases
+it to be triggered, and its action is taken. There can never be a backtrack
+onto (*COMMIT).
+.
+.
+.\" HTML 
+.SS "Backtracking verbs in repeated groups"
+.rs
+.sp
+PCRE differs from Perl in its handling of backtracking verbs in repeated
+groups. For example, consider:
+.sp
+  /(a(*COMMIT)b)+ac/
+.sp
+If the subject is "abac", Perl matches, but PCRE fails because the (*COMMIT) in
+the second repeat of the group acts.
+.
+.
+.\" HTML 
+.SS "Backtracking verbs in assertions"
+.rs
+.sp
+(*FAIL) in an assertion has its normal effect: it forces an immediate backtrack.
+.P
+(*ACCEPT) in a positive assertion causes the assertion to succeed without any
+further processing. In a negative assertion, (*ACCEPT) causes the assertion to
+fail without any further processing.
+.P
+The other backtracking verbs are not treated specially if they appear in a
+positive assertion. In particular, (*THEN) skips to the next alternative in the
+innermost enclosing group that has alternations, whether or not this is within
+the assertion.
+.P
+Negative assertions are, however, different, in order to ensure that changing a
+positive assertion into a negative assertion changes its result. Backtracking
+into (*COMMIT), (*SKIP), or (*PRUNE) causes a negative assertion to be true,
+without considering any further alternative branches in the assertion.
+Backtracking into (*THEN) causes it to skip to the next enclosing alternative
+within the assertion (the normal behaviour), but if the assertion does not have
+such an alternative, (*THEN) behaves like (*PRUNE).
+.
+.
+.\" HTML 
+.SS "Backtracking verbs in subroutines"
+.rs
+.sp
+These behaviours occur whether or not the subpattern is called recursively.
+Perl's treatment of subroutines is different in some cases.
+.P
+(*FAIL) in a subpattern called as a subroutine has its normal effect: it forces
+an immediate backtrack.
+.P
+(*ACCEPT) in a subpattern called as a subroutine causes the subroutine match to
+succeed without any further processing. Matching then continues after the
+subroutine call.
+.P
+(*COMMIT), (*SKIP), and (*PRUNE) in a subpattern called as a subroutine cause
+the subroutine match to fail.
+.P
+(*THEN) skips to the next alternative in the innermost enclosing group within
+the subpattern that has alternatives. If there is no such group within the
+subpattern, (*THEN) causes the subroutine match to fail.
+.
+.
+.SH "SEE ALSO"
+.rs
+.sp
+\fBpcreapi\fP(3), \fBpcrecallout\fP(3), \fBpcrematching\fP(3),
+\fBpcresyntax\fP(3), \fBpcre\fP(3), \fBpcre16(3)\fP, \fBpcre32(3)\fP.
+.
+.
+.SH AUTHOR
+.rs
+.sp
+.nf
+Philip Hazel
+University Computing Service
+Cambridge CB2 3QH, England.
+.fi
+.
+.
+.SH REVISION
+.rs
+.sp
+.nf
+Last updated: 26 April 2013
+Copyright (c) 1997-2013 University of Cambridge.
+.fi
diff --git a/pcre/doc/pcreperform.3 b/pcre/doc/pcreperform.3
new file mode 100644
index 00000000000..fb2aa95926a
--- /dev/null
+++ b/pcre/doc/pcreperform.3
@@ -0,0 +1,177 @@
+.TH PCREPERFORM 3 "09 January 2012" "PCRE 8.30"
+.SH NAME
+PCRE - Perl-compatible regular expressions
+.SH "PCRE PERFORMANCE"
+.rs
+.sp
+Two aspects of performance are discussed below: memory usage and processing
+time. The way you express your pattern as a regular expression can affect both
+of them.
+.
+.SH "COMPILED PATTERN MEMORY USAGE"
+.rs
+.sp
+Patterns are compiled by PCRE into a reasonably efficient interpretive code, so
+that most simple patterns do not use much memory. However, there is one case
+where the memory usage of a compiled pattern can be unexpectedly large. If a
+parenthesized subpattern has a quantifier with a minimum greater than 1 and/or
+a limited maximum, the whole subpattern is repeated in the compiled code. For
+example, the pattern
+.sp
+  (abc|def){2,4}
+.sp
+is compiled as if it were
+.sp
+  (abc|def)(abc|def)((abc|def)(abc|def)?)?
+.sp
+(Technical aside: It is done this way so that backtrack points within each of
+the repetitions can be independently maintained.)
+.P
+For regular expressions whose quantifiers use only small numbers, this is not
+usually a problem. However, if the numbers are large, and particularly if such
+repetitions are nested, the memory usage can become an embarrassment. For
+example, the very simple pattern
+.sp
+  ((ab){1,1000}c){1,3}
+.sp
+uses 51K bytes when compiled using the 8-bit library. When PCRE is compiled
+with its default internal pointer size of two bytes, the size limit on a
+compiled pattern is 64K data units, and this is reached with the above pattern
+if the outer repetition is increased from 3 to 4. PCRE can be compiled to use
+larger internal pointers and thus handle larger compiled patterns, but it is
+better to try to rewrite your pattern to use less memory if you can.
+.P
+One way of reducing the memory usage for such patterns is to make use of PCRE's
+.\" HTML 
+.\" 
+"subroutine"
+.\"
+facility. Re-writing the above pattern as
+.sp
+  ((ab)(?2){0,999}c)(?1){0,2}
+.sp
+reduces the memory requirements to 18K, and indeed it remains under 20K even
+with the outer repetition increased to 100. However, this pattern is not
+exactly equivalent, because the "subroutine" calls are treated as
+.\" HTML 
+.\" 
+atomic groups
+.\"
+into which there can be no backtracking if there is a subsequent matching
+failure. Therefore, PCRE cannot do this kind of rewriting automatically.
+Furthermore, there is a noticeable loss of speed when executing the modified
+pattern. Nevertheless, if the atomic grouping is not a problem and the loss of
+speed is acceptable, this kind of rewriting will allow you to process patterns
+that PCRE cannot otherwise handle.
+.
+.
+.SH "STACK USAGE AT RUN TIME"
+.rs
+.sp
+When \fBpcre_exec()\fP or \fBpcre[16|32]_exec()\fP is used for matching, certain
+kinds of pattern can cause it to use large amounts of the process stack. In
+some environments the default process stack is quite small, and if it runs out
+the result is often SIGSEGV. This issue is probably the most frequently raised
+problem with PCRE. Rewriting your pattern can often help. The
+.\" HREF
+\fBpcrestack\fP
+.\"
+documentation discusses this issue in detail.
+.
+.
+.SH "PROCESSING TIME"
+.rs
+.sp
+Certain items in regular expression patterns are processed more efficiently
+than others. It is more efficient to use a character class like [aeiou] than a
+set of single-character alternatives such as (a|e|i|o|u). In general, the
+simplest construction that provides the required behaviour is usually the most
+efficient. Jeffrey Friedl's book contains a lot of useful general discussion
+about optimizing regular expressions for efficient performance. This document
+contains a few observations about PCRE.
+.P
+Using Unicode character properties (the \ep, \eP, and \eX escapes) is slow,
+because PCRE has to use a multi-stage table lookup whenever it needs a
+character's property. If you can find an alternative pattern that does not use
+character properties, it will probably be faster.
+.P
+By default, the escape sequences \eb, \ed, \es, and \ew, and the POSIX
+character classes such as [:alpha:] do not use Unicode properties, partly for
+backwards compatibility, and partly for performance reasons. However, you can
+set PCRE_UCP if you want Unicode character properties to be used. This can
+double the matching time for items such as \ed, when matched with
+a traditional matching function; the performance loss is less with
+a DFA matching function, and in both cases there is not much difference for
+\eb.
+.P
+When a pattern begins with .* not in parentheses, or in parentheses that are
+not the subject of a backreference, and the PCRE_DOTALL option is set, the
+pattern is implicitly anchored by PCRE, since it can match only at the start of
+a subject string. However, if PCRE_DOTALL is not set, PCRE cannot make this
+optimization, because the . metacharacter does not then match a newline, and if
+the subject string contains newlines, the pattern may match from the character
+immediately following one of them instead of from the very start. For example,
+the pattern
+.sp
+  .*second
+.sp
+matches the subject "first\enand second" (where \en stands for a newline
+character), with the match starting at the seventh character. In order to do
+this, PCRE has to retry the match starting after every newline in the subject.
+.P
+If you are using such a pattern with subject strings that do not contain
+newlines, the best performance is obtained by setting PCRE_DOTALL, or starting
+the pattern with ^.* or ^.*? to indicate explicit anchoring. That saves PCRE
+from having to scan along the subject looking for a newline to restart at.
+.P
+Beware of patterns that contain nested indefinite repeats. These can take a
+long time to run when applied to a string that does not match. Consider the
+pattern fragment
+.sp
+  ^(a+)*
+.sp
+This can match "aaaa" in 16 different ways, and this number increases very
+rapidly as the string gets longer. (The * repeat can match 0, 1, 2, 3, or 4
+times, and for each of those cases other than 0 or 4, the + repeats can match
+different numbers of times.) When the remainder of the pattern is such that the
+entire match is going to fail, PCRE has in principle to try every possible
+variation, and this can take an extremely long time, even for relatively short
+strings.
+.P
+An optimization catches some of the more simple cases such as
+.sp
+  (a+)*b
+.sp
+where a literal character follows. Before embarking on the standard matching
+procedure, PCRE checks that there is a "b" later in the subject string, and if
+there is not, it fails the match immediately. However, when there is no
+following literal this optimization cannot be used. You can see the difference
+by comparing the behaviour of
+.sp
+  (a+)*\ed
+.sp
+with the pattern above. The former gives a failure almost instantly when
+applied to a whole line of "a" characters, whereas the latter takes an
+appreciable time with strings longer than about 20 characters.
+.P
+In many cases, the solution to this kind of performance issue is to use an
+atomic group or a possessive quantifier.
+.
+.
+.SH AUTHOR
+.rs
+.sp
+.nf
+Philip Hazel
+University Computing Service
+Cambridge CB2 3QH, England.
+.fi
+.
+.
+.SH REVISION
+.rs
+.sp
+.nf
+Last updated: 25 August 2012
+Copyright (c) 1997-2012 University of Cambridge.
+.fi
diff --git a/pcre/doc/pcreposix.3 b/pcre/doc/pcreposix.3
new file mode 100644
index 00000000000..b25a89193c9
--- /dev/null
+++ b/pcre/doc/pcreposix.3
@@ -0,0 +1,270 @@
+.TH PCREPOSIX 3 "09 January 2012" "PCRE 8.30"
+.SH NAME
+PCRE - Perl-compatible regular expressions.
+.SH "SYNOPSIS OF POSIX API"
+.rs
+.sp
+.B #include 
+.PP
+.SM
+.B int regcomp(regex_t *\fIpreg\fP, const char *\fIpattern\fP,
+.ti +5n
+.B int \fIcflags\fP);
+.PP
+.B int regexec(regex_t *\fIpreg\fP, const char *\fIstring\fP,
+.ti +5n
+.B size_t \fInmatch\fP, regmatch_t \fIpmatch\fP[], int \fIeflags\fP);
+.PP
+.B size_t regerror(int \fIerrcode\fP, const regex_t *\fIpreg\fP,
+.ti +5n
+.B char *\fIerrbuf\fP, size_t \fIerrbuf_size\fP);
+.PP
+.B void regfree(regex_t *\fIpreg\fP);
+.
+.SH DESCRIPTION
+.rs
+.sp
+This set of functions provides a POSIX-style API for the PCRE regular
+expression 8-bit library. See the
+.\" HREF
+\fBpcreapi\fP
+.\"
+documentation for a description of PCRE's native API, which contains much
+additional functionality. There is no POSIX-style wrapper for PCRE's 16-bit
+and 32-bit library.
+.P
+The functions described here are just wrapper functions that ultimately call
+the PCRE native API. Their prototypes are defined in the \fBpcreposix.h\fP
+header file, and on Unix systems the library itself is called
+\fBpcreposix.a\fP, so can be accessed by adding \fB-lpcreposix\fP to the
+command for linking an application that uses them. Because the POSIX functions
+call the native ones, it is also necessary to add \fB-lpcre\fP.
+.P
+I have implemented only those POSIX option bits that can be reasonably mapped
+to PCRE native options. In addition, the option REG_EXTENDED is defined with
+the value zero. This has no effect, but since programs that are written to the
+POSIX interface often use it, this makes it easier to slot in PCRE as a
+replacement library. Other POSIX options are not even defined.
+.P
+There are also some other options that are not defined by POSIX. These have
+been added at the request of users who want to make use of certain
+PCRE-specific features via the POSIX calling interface.
+.P
+When PCRE is called via these functions, it is only the API that is POSIX-like
+in style. The syntax and semantics of the regular expressions themselves are
+still those of Perl, subject to the setting of various PCRE options, as
+described below. "POSIX-like in style" means that the API approximates to the
+POSIX definition; it is not fully POSIX-compatible, and in multi-byte encoding
+domains it is probably even less compatible.
+.P
+The header for these functions is supplied as \fBpcreposix.h\fP to avoid any
+potential clash with other POSIX libraries. It can, of course, be renamed or
+aliased as \fBregex.h\fP, which is the "correct" name. It provides two
+structure types, \fIregex_t\fP for compiled internal forms, and
+\fIregmatch_t\fP for returning captured substrings. It also defines some
+constants whose names start with "REG_"; these are used for setting options and
+identifying error codes.
+.
+.
+.SH "COMPILING A PATTERN"
+.rs
+.sp
+The function \fBregcomp()\fP is called to compile a pattern into an
+internal form. The pattern is a C string terminated by a binary zero, and
+is passed in the argument \fIpattern\fP. The \fIpreg\fP argument is a pointer
+to a \fBregex_t\fP structure that is used as a base for storing information
+about the compiled regular expression.
+.P
+The argument \fIcflags\fP is either zero, or contains one or more of the bits
+defined by the following macros:
+.sp
+  REG_DOTALL
+.sp
+The PCRE_DOTALL option is set when the regular expression is passed for
+compilation to the native function. Note that REG_DOTALL is not part of the
+POSIX standard.
+.sp
+  REG_ICASE
+.sp
+The PCRE_CASELESS option is set when the regular expression is passed for
+compilation to the native function.
+.sp
+  REG_NEWLINE
+.sp
+The PCRE_MULTILINE option is set when the regular expression is passed for
+compilation to the native function. Note that this does \fInot\fP mimic the
+defined POSIX behaviour for REG_NEWLINE (see the following section).
+.sp
+  REG_NOSUB
+.sp
+The PCRE_NO_AUTO_CAPTURE option is set when the regular expression is passed
+for compilation to the native function. In addition, when a pattern that is
+compiled with this flag is passed to \fBregexec()\fP for matching, the
+\fInmatch\fP and \fIpmatch\fP arguments are ignored, and no captured strings
+are returned.
+.sp
+  REG_UCP
+.sp
+The PCRE_UCP option is set when the regular expression is passed for
+compilation to the native function. This causes PCRE to use Unicode properties
+when matchine \ed, \ew, etc., instead of just recognizing ASCII values. Note
+that REG_UTF8 is not part of the POSIX standard.
+.sp
+  REG_UNGREEDY
+.sp
+The PCRE_UNGREEDY option is set when the regular expression is passed for
+compilation to the native function. Note that REG_UNGREEDY is not part of the
+POSIX standard.
+.sp
+  REG_UTF8
+.sp
+The PCRE_UTF8 option is set when the regular expression is passed for
+compilation to the native function. This causes the pattern itself and all data
+strings used for matching it to be treated as UTF-8 strings. Note that REG_UTF8
+is not part of the POSIX standard.
+.P
+In the absence of these flags, no options are passed to the native function.
+This means the the regex is compiled with PCRE default semantics. In
+particular, the way it handles newline characters in the subject string is the
+Perl way, not the POSIX way. Note that setting PCRE_MULTILINE has only
+\fIsome\fP of the effects specified for REG_NEWLINE. It does not affect the way
+newlines are matched by . (they are not) or by a negative class such as [^a]
+(they are).
+.P
+The yield of \fBregcomp()\fP is zero on success, and non-zero otherwise. The
+\fIpreg\fP structure is filled in on success, and one member of the structure
+is public: \fIre_nsub\fP contains the number of capturing subpatterns in
+the regular expression. Various error codes are defined in the header file.
+.P
+NOTE: If the yield of \fBregcomp()\fP is non-zero, you must not attempt to
+use the contents of the \fIpreg\fP structure. If, for example, you pass it to
+\fBregexec()\fP, the result is undefined and your program is likely to crash.
+.
+.
+.SH "MATCHING NEWLINE CHARACTERS"
+.rs
+.sp
+This area is not simple, because POSIX and Perl take different views of things.
+It is not possible to get PCRE to obey POSIX semantics, but then PCRE was never
+intended to be a POSIX engine. The following table lists the different
+possibilities for matching newline characters in PCRE:
+.sp
+                          Default   Change with
+.sp
+  . matches newline          no     PCRE_DOTALL
+  newline matches [^a]       yes    not changeable
+  $ matches \en at end        yes    PCRE_DOLLARENDONLY
+  $ matches \en in middle     no     PCRE_MULTILINE
+  ^ matches \en in middle     no     PCRE_MULTILINE
+.sp
+This is the equivalent table for POSIX:
+.sp
+                          Default   Change with
+.sp
+  . matches newline          yes    REG_NEWLINE
+  newline matches [^a]       yes    REG_NEWLINE
+  $ matches \en at end        no     REG_NEWLINE
+  $ matches \en in middle     no     REG_NEWLINE
+  ^ matches \en in middle     no     REG_NEWLINE
+.sp
+PCRE's behaviour is the same as Perl's, except that there is no equivalent for
+PCRE_DOLLAR_ENDONLY in Perl. In both PCRE and Perl, there is no way to stop
+newline from matching [^a].
+.P
+The default POSIX newline handling can be obtained by setting PCRE_DOTALL and
+PCRE_DOLLAR_ENDONLY, but there is no way to make PCRE behave exactly as for the
+REG_NEWLINE action.
+.
+.
+.SH "MATCHING A PATTERN"
+.rs
+.sp
+The function \fBregexec()\fP is called to match a compiled pattern \fIpreg\fP
+against a given \fIstring\fP, which is by default terminated by a zero byte
+(but see REG_STARTEND below), subject to the options in \fIeflags\fP. These can
+be:
+.sp
+  REG_NOTBOL
+.sp
+The PCRE_NOTBOL option is set when calling the underlying PCRE matching
+function.
+.sp
+  REG_NOTEMPTY
+.sp
+The PCRE_NOTEMPTY option is set when calling the underlying PCRE matching
+function. Note that REG_NOTEMPTY is not part of the POSIX standard. However,
+setting this option can give more POSIX-like behaviour in some situations.
+.sp
+  REG_NOTEOL
+.sp
+The PCRE_NOTEOL option is set when calling the underlying PCRE matching
+function.
+.sp
+  REG_STARTEND
+.sp
+The string is considered to start at \fIstring\fP + \fIpmatch[0].rm_so\fP and
+to have a terminating NUL located at \fIstring\fP + \fIpmatch[0].rm_eo\fP
+(there need not actually be a NUL at that location), regardless of the value of
+\fInmatch\fP. This is a BSD extension, compatible with but not specified by
+IEEE Standard 1003.2 (POSIX.2), and should be used with caution in software
+intended to be portable to other systems. Note that a non-zero \fIrm_so\fP does
+not imply REG_NOTBOL; REG_STARTEND affects only the location of the string, not
+how it is matched.
+.P
+If the pattern was compiled with the REG_NOSUB flag, no data about any matched
+strings is returned. The \fInmatch\fP and \fIpmatch\fP arguments of
+\fBregexec()\fP are ignored.
+.P
+If the value of \fInmatch\fP is zero, or if the value \fIpmatch\fP is NULL,
+no data about any matched strings is returned.
+.P
+Otherwise,the portion of the string that was matched, and also any captured
+substrings, are returned via the \fIpmatch\fP argument, which points to an
+array of \fInmatch\fP structures of type \fIregmatch_t\fP, containing the
+members \fIrm_so\fP and \fIrm_eo\fP. These contain the offset to the first
+character of each substring and the offset to the first character after the end
+of each substring, respectively. The 0th element of the vector relates to the
+entire portion of \fIstring\fP that was matched; subsequent elements relate to
+the capturing subpatterns of the regular expression. Unused entries in the
+array have both structure members set to -1.
+.P
+A successful match yields a zero return; various error codes are defined in the
+header file, of which REG_NOMATCH is the "expected" failure code.
+.
+.
+.SH "ERROR MESSAGES"
+.rs
+.sp
+The \fBregerror()\fP function maps a non-zero errorcode from either
+\fBregcomp()\fP or \fBregexec()\fP to a printable message. If \fIpreg\fP is not
+NULL, the error should have arisen from the use of that structure. A message
+terminated by a binary zero is placed in \fIerrbuf\fP. The length of the
+message, including the zero, is limited to \fIerrbuf_size\fP. The yield of the
+function is the size of buffer needed to hold the whole message.
+.
+.
+.SH MEMORY USAGE
+.rs
+.sp
+Compiling a regular expression causes memory to be allocated and associated
+with the \fIpreg\fP structure. The function \fBregfree()\fP frees all such
+memory, after which \fIpreg\fP may no longer be used as a compiled expression.
+.
+.
+.SH AUTHOR
+.rs
+.sp
+.nf
+Philip Hazel
+University Computing Service
+Cambridge CB2 3QH, England.
+.fi
+.
+.
+.SH REVISION
+.rs
+.sp
+.nf
+Last updated: 09 January 2012
+Copyright (c) 1997-2012 University of Cambridge.
+.fi
diff --git a/pcre/doc/pcreprecompile.3 b/pcre/doc/pcreprecompile.3
new file mode 100644
index 00000000000..39eb82b02cc
--- /dev/null
+++ b/pcre/doc/pcreprecompile.3
@@ -0,0 +1,151 @@
+.TH PCREPRECOMPILE 3 "24 June 2012" "PCRE 8.30"
+.SH NAME
+PCRE - Perl-compatible regular expressions
+.SH "SAVING AND RE-USING PRECOMPILED PCRE PATTERNS"
+.rs
+.sp
+If you are running an application that uses a large number of regular
+expression patterns, it may be useful to store them in a precompiled form
+instead of having to compile them every time the application is run.
+If you are not using any private character tables (see the
+.\" HREF
+\fBpcre_maketables()\fP
+.\"
+documentation), this is relatively straightforward. If you are using private
+tables, it is a little bit more complicated. However, if you are using the
+just-in-time optimization feature, it is not possible to save and reload the
+JIT data.
+.P
+If you save compiled patterns to a file, you can copy them to a different host
+and run them there. If the two hosts have different endianness (byte order),
+you should run the \fBpcre[16|32]_pattern_to_host_byte_order()\fP function on the
+new host before trying to match the pattern. The matching functions return
+PCRE_ERROR_BADENDIANNESS if they detect a pattern with the wrong endianness.
+.P
+Compiling regular expressions with one version of PCRE for use with a different
+version is not guaranteed to work and may cause crashes, and saving and
+restoring a compiled pattern loses any JIT optimization data.
+.
+.
+.SH "SAVING A COMPILED PATTERN"
+.rs
+.sp
+The value returned by \fBpcre[16|32]_compile()\fP points to a single block of
+memory that holds the compiled pattern and associated data. You can find the
+length of this block in bytes by calling \fBpcre[16|32]_fullinfo()\fP with an
+argument of PCRE_INFO_SIZE. You can then save the data in any appropriate
+manner. Here is sample code for the 8-bit library that compiles a pattern and
+writes it to a file. It assumes that the variable \fIfd\fP refers to a file
+that is open for output:
+.sp
+  int erroroffset, rc, size;
+  char *error;
+  pcre *re;
+.sp
+  re = pcre_compile("my pattern", 0, &error, &erroroffset, NULL);
+  if (re == NULL) { ... handle errors ... }
+  rc = pcre_fullinfo(re, NULL, PCRE_INFO_SIZE, &size);
+  if (rc < 0) { ... handle errors ... }
+  rc = fwrite(re, 1, size, fd);
+  if (rc != size) { ... handle errors ... }
+.sp
+In this example, the bytes that comprise the compiled pattern are copied
+exactly. Note that this is binary data that may contain any of the 256 possible
+byte values. On systems that make a distinction between binary and non-binary
+data, be sure that the file is opened for binary output.
+.P
+If you want to write more than one pattern to a file, you will have to devise a
+way of separating them. For binary data, preceding each pattern with its length
+is probably the most straightforward approach. Another possibility is to write
+out the data in hexadecimal instead of binary, one pattern to a line.
+.P
+Saving compiled patterns in a file is only one possible way of storing them for
+later use. They could equally well be saved in a database, or in the memory of
+some daemon process that passes them via sockets to the processes that want
+them.
+.P
+If the pattern has been studied, it is also possible to save the normal study
+data in a similar way to the compiled pattern itself. However, if the
+PCRE_STUDY_JIT_COMPILE was used, the just-in-time data that is created cannot
+be saved because it is too dependent on the current environment. When studying
+generates additional information, \fBpcre[16|32]_study()\fP returns a pointer to a
+\fBpcre[16|32]_extra\fP data block. Its format is defined in the
+.\" HTML 
+.\" 
+section on matching a pattern
+.\"
+in the
+.\" HREF
+\fBpcreapi\fP
+.\"
+documentation. The \fIstudy_data\fP field points to the binary study data, and
+this is what you must save (not the \fBpcre[16|32]_extra\fP block itself). The
+length of the study data can be obtained by calling \fBpcre[16|32]_fullinfo()\fP
+with an argument of PCRE_INFO_STUDYSIZE. Remember to check that
+\fBpcre[16|32]_study()\fP did return a non-NULL value before trying to save the
+study data.
+.
+.
+.SH "RE-USING A PRECOMPILED PATTERN"
+.rs
+.sp
+Re-using a precompiled pattern is straightforward. Having reloaded it into main
+memory, called \fBpcre[16|32]_pattern_to_host_byte_order()\fP if necessary,
+you pass its pointer to \fBpcre[16|32]_exec()\fP or \fBpcre[16|32]_dfa_exec()\fP in
+the usual way.
+.P
+However, if you passed a pointer to custom character tables when the pattern
+was compiled (the \fItableptr\fP argument of \fBpcre[16|32]_compile()\fP), you
+must now pass a similar pointer to \fBpcre[16|32]_exec()\fP or
+\fBpcre[16|32]_dfa_exec()\fP, because the value saved with the compiled pattern
+will obviously be nonsense. A field in a \fBpcre[16|32]_extra()\fP block is used
+to pass this data, as described in the
+.\" HTML 
+.\" 
+section on matching a pattern
+.\"
+in the
+.\" HREF
+\fBpcreapi\fP
+.\"
+documentation.
+.P
+If you did not provide custom character tables when the pattern was compiled,
+the pointer in the compiled pattern is NULL, which causes the matching
+functions to use PCRE's internal tables. Thus, you do not need to take any
+special action at run time in this case.
+.P
+If you saved study data with the compiled pattern, you need to create your own
+\fBpcre[16|32]_extra\fP data block and set the \fIstudy_data\fP field to point to the
+reloaded study data. You must also set the PCRE_EXTRA_STUDY_DATA bit in the
+\fIflags\fP field to indicate that study data is present. Then pass the
+\fBpcre[16|32]_extra\fP block to the matching function in the usual way. If the
+pattern was studied for just-in-time optimization, that data cannot be saved,
+and so is lost by a save/restore cycle.
+.
+.
+.SH "COMPATIBILITY WITH DIFFERENT PCRE RELEASES"
+.rs
+.sp
+In general, it is safest to recompile all saved patterns when you update to a
+new PCRE release, though not all updates actually require this.
+.
+.
+.
+.SH AUTHOR
+.rs
+.sp
+.nf
+Philip Hazel
+University Computing Service
+Cambridge CB2 3QH, England.
+.fi
+.
+.
+.SH REVISION
+.rs
+.sp
+.nf
+Last updated: 24 June 2012
+Copyright (c) 1997-2012 University of Cambridge.
+.fi
diff --git a/pcre/doc/pcresample.3 b/pcre/doc/pcresample.3
new file mode 100644
index 00000000000..d7fe7ec546b
--- /dev/null
+++ b/pcre/doc/pcresample.3
@@ -0,0 +1,99 @@
+.TH PCRESAMPLE 3 "10 January 2012" "PCRE 8.30"
+.SH NAME
+PCRE - Perl-compatible regular expressions
+.SH "PCRE SAMPLE PROGRAM"
+.rs
+.sp
+A simple, complete demonstration program, to get you started with using PCRE,
+is supplied in the file \fIpcredemo.c\fP in the PCRE distribution. A listing of
+this program is given in the
+.\" HREF
+\fBpcredemo\fP
+.\"
+documentation. If you do not have a copy of the PCRE distribution, you can save
+this listing to re-create \fIpcredemo.c\fP.
+.P
+The demonstration program, which uses the original PCRE 8-bit library, compiles
+the regular expression that is its first argument, and matches it against the
+subject string in its second argument. No PCRE options are set, and default
+character tables are used. If matching succeeds, the program outputs the
+portion of the subject that matched, together with the contents of any captured
+substrings.
+.P
+If the -g option is given on the command line, the program then goes on to
+check for further matches of the same regular expression in the same subject
+string. The logic is a little bit tricky because of the possibility of matching
+an empty string. Comments in the code explain what is going on.
+.P
+If PCRE is installed in the standard include and library directories for your
+operating system, you should be able to compile the demonstration program using
+this command:
+.sp
+  gcc -o pcredemo pcredemo.c -lpcre
+.sp
+If PCRE is installed elsewhere, you may need to add additional options to the
+command line. For example, on a Unix-like system that has PCRE installed in
+\fI/usr/local\fP, you can compile the demonstration program using a command
+like this:
+.sp
+.\" JOINSH
+  gcc -o pcredemo -I/usr/local/include pcredemo.c \e
+      -L/usr/local/lib -lpcre
+.sp
+In a Windows environment, if you want to statically link the program against a
+non-dll \fBpcre.a\fP file, you must uncomment the line that defines PCRE_STATIC
+before including \fBpcre.h\fP, because otherwise the \fBpcre_malloc()\fP and
+\fBpcre_free()\fP exported functions will be declared
+\fB__declspec(dllimport)\fP, with unwanted results.
+.P
+Once you have compiled and linked the demonstration program, you can run simple
+tests like this:
+.sp
+  ./pcredemo 'cat|dog' 'the cat sat on the mat'
+  ./pcredemo -g 'cat|dog' 'the dog sat on the cat'
+.sp
+Note that there is a much more comprehensive test program, called
+.\" HREF
+\fBpcretest\fP,
+.\"
+which supports many more facilities for testing regular expressions and both
+PCRE libraries. The
+.\" HREF
+\fBpcredemo\fP
+.\"
+program is provided as a simple coding example.
+.P
+If you try to run
+.\" HREF
+\fBpcredemo\fP
+.\"
+when PCRE is not installed in the standard library directory, you may get an
+error like this on some operating systems (e.g. Solaris):
+.sp
+  ld.so.1: a.out: fatal: libpcre.so.0: open failed: No such file or directory
+.sp
+This is caused by the way shared library support works on those systems. You
+need to add
+.sp
+  -R/usr/local/lib
+.sp
+(for example) to the compile command to get round this problem.
+.
+.
+.SH AUTHOR
+.rs
+.sp
+.nf
+Philip Hazel
+University Computing Service
+Cambridge CB2 3QH, England.
+.fi
+.
+.
+.SH REVISION
+.rs
+.sp
+.nf
+Last updated: 10 January 2012
+Copyright (c) 1997-2012 University of Cambridge.
+.fi
diff --git a/pcre/doc/pcrestack.3 b/pcre/doc/pcrestack.3
new file mode 100644
index 00000000000..798f0bca63e
--- /dev/null
+++ b/pcre/doc/pcrestack.3
@@ -0,0 +1,215 @@
+.TH PCRESTACK 3 "24 June 2012" "PCRE 8.30"
+.SH NAME
+PCRE - Perl-compatible regular expressions
+.SH "PCRE DISCUSSION OF STACK USAGE"
+.rs
+.sp
+When you call \fBpcre[16|32]_exec()\fP, it makes use of an internal function
+called \fBmatch()\fP. This calls itself recursively at branch points in the
+pattern, in order to remember the state of the match so that it can back up and
+try a different alternative if the first one fails. As matching proceeds deeper
+and deeper into the tree of possibilities, the recursion depth increases. The
+\fBmatch()\fP function is also called in other circumstances, for example,
+whenever a parenthesized sub-pattern is entered, and in certain cases of
+repetition.
+.P
+Not all calls of \fBmatch()\fP increase the recursion depth; for an item such
+as a* it may be called several times at the same level, after matching
+different numbers of a's. Furthermore, in a number of cases where the result of
+the recursive call would immediately be passed back as the result of the
+current call (a "tail recursion"), the function is just restarted instead.
+.P
+The above comments apply when \fBpcre[16|32]_exec()\fP is run in its normal
+interpretive manner. If the pattern was studied with the
+PCRE_STUDY_JIT_COMPILE option, and just-in-time compiling was successful, and
+the options passed to \fBpcre[16|32]_exec()\fP were not incompatible, the matching
+process uses the JIT-compiled code instead of the \fBmatch()\fP function. In
+this case, the memory requirements are handled entirely differently. See the
+.\" HREF
+\fBpcrejit\fP
+.\"
+documentation for details.
+.P
+The \fBpcre[16|32]_dfa_exec()\fP function operates in an entirely different way,
+and uses recursion only when there is a regular expression recursion or
+subroutine call in the pattern. This includes the processing of assertion and
+"once-only" subpatterns, which are handled like subroutine calls. Normally,
+these are never very deep, and the limit on the complexity of
+\fBpcre[16|32]_dfa_exec()\fP is controlled by the amount of workspace it is given.
+However, it is possible to write patterns with runaway infinite recursions;
+such patterns will cause \fBpcre[16|32]_dfa_exec()\fP to run out of stack. At
+present, there is no protection against this.
+.P
+The comments that follow do NOT apply to \fBpcre[16|32]_dfa_exec()\fP; they are
+relevant only for \fBpcre[16|32]_exec()\fP without the JIT optimization.
+.
+.
+.SS "Reducing \fBpcre[16|32]_exec()\fP's stack usage"
+.rs
+.sp
+Each time that \fBmatch()\fP is actually called recursively, it uses memory
+from the process stack. For certain kinds of pattern and data, very large
+amounts of stack may be needed, despite the recognition of "tail recursion".
+You can often reduce the amount of recursion, and therefore the amount of stack
+used, by modifying the pattern that is being matched. Consider, for example,
+this pattern:
+.sp
+  ([^<]|<(?!inet))+
+.sp
+It matches from wherever it starts until it encounters "
+.\" 
+section on extra data for \fBpcre[16|32]_exec()\fP
+.\"
+in the
+.\" HREF
+\fBpcreapi\fP
+.\"
+documentation.
+.P
+As a very rough rule of thumb, you should reckon on about 500 bytes per
+recursion. Thus, if you want to limit your stack usage to 8Mb, you should set
+the limit at 16000 recursions. A 64Mb stack, on the other hand, can support
+around 128000 recursions.
+.P
+In Unix-like environments, the \fBpcretest\fP test program has a command line
+option (\fB-S\fP) that can be used to increase the size of its stack. As long
+as the stack is large enough, another option (\fB-M\fP) can be used to find the
+smallest limits that allow a particular pattern to match a given subject
+string. This is done by calling \fBpcre[16|32]_exec()\fP repeatedly with different
+limits.
+.
+.
+.SS "Obtaining an estimate of stack usage"
+.rs
+.sp
+The actual amount of stack used per recursion can vary quite a lot, depending
+on the compiler that was used to build PCRE and the optimization or debugging
+options that were set for it. The rule of thumb value of 500 bytes mentioned
+above may be larger or smaller than what is actually needed. A better
+approximation can be obtained by running this command:
+.sp
+  pcretest -m -C
+.sp
+The \fB-C\fP option causes \fBpcretest\fP to output information about the
+options with which PCRE was compiled. When \fB-m\fP is also given (before
+\fB-C\fP), information about stack use is given in a line like this:
+.sp
+  Match recursion uses stack: approximate frame size = 640 bytes
+.sp
+The value is approximate because some recursions need a bit more (up to perhaps
+16 more bytes).
+.P
+If the above command is given when PCRE is compiled to use the heap instead of
+the stack for recursion, the value that is output is the size of each block
+that is obtained from the heap.
+.
+.
+.SS "Changing stack size in Unix-like systems"
+.rs
+.sp
+In Unix-like environments, there is not often a problem with the stack unless
+very long strings are involved, though the default limit on stack size varies
+from system to system. Values from 8Mb to 64Mb are common. You can find your
+default limit by running the command:
+.sp
+  ulimit -s
+.sp
+Unfortunately, the effect of running out of stack is often SIGSEGV, though
+sometimes a more explicit error message is given. You can normally increase the
+limit on stack size by code such as this:
+.sp
+  struct rlimit rlim;
+  getrlimit(RLIMIT_STACK, &rlim);
+  rlim.rlim_cur = 100*1024*1024;
+  setrlimit(RLIMIT_STACK, &rlim);
+.sp
+This reads the current limits (soft and hard) using \fBgetrlimit()\fP, then
+attempts to increase the soft limit to 100Mb using \fBsetrlimit()\fP. You must
+do this before calling \fBpcre[16|32]_exec()\fP.
+.
+.
+.SS "Changing stack size in Mac OS X"
+.rs
+.sp
+Using \fBsetrlimit()\fP, as described above, should also work on Mac OS X. It
+is also possible to set a stack size when linking a program. There is a
+discussion about stack sizes in Mac OS X at this web site:
+.\" HTML 
+.\" 
+http://developer.apple.com/qa/qa2005/qa1419.html.
+.\"
+.
+.
+.SH AUTHOR
+.rs
+.sp
+.nf
+Philip Hazel
+University Computing Service
+Cambridge CB2 3QH, England.
+.fi
+.
+.
+.SH REVISION
+.rs
+.sp
+.nf
+Last updated: 24 June 2012
+Copyright (c) 1997-2012 University of Cambridge.
+.fi
diff --git a/pcre/doc/pcresyntax.3 b/pcre/doc/pcresyntax.3
new file mode 100644
index 00000000000..399bbe2535a
--- /dev/null
+++ b/pcre/doc/pcresyntax.3
@@ -0,0 +1,500 @@
+.TH PCRESYNTAX 3 "26 April 2013" "PCRE 8.33"
+.SH NAME
+PCRE - Perl-compatible regular expressions
+.SH "PCRE REGULAR EXPRESSION SYNTAX SUMMARY"
+.rs
+.sp
+The full syntax and semantics of the regular expressions that are supported by
+PCRE are described in the
+.\" HREF
+\fBpcrepattern\fP
+.\"
+documentation. This document contains a quick-reference summary of the syntax.
+.
+.
+.SH "QUOTING"
+.rs
+.sp
+  \ex         where x is non-alphanumeric is a literal x
+  \eQ...\eE    treat enclosed characters as literal
+.
+.
+.SH "CHARACTERS"
+.rs
+.sp
+  \ea         alarm, that is, the BEL character (hex 07)
+  \ecx        "control-x", where x is any ASCII character
+  \ee         escape (hex 1B)
+  \ef         form feed (hex 0C)
+  \en         newline (hex 0A)
+  \er         carriage return (hex 0D)
+  \et         tab (hex 09)
+  \eddd       character with octal code ddd, or backreference
+  \exhh       character with hex code hh
+  \ex{hhh..}  character with hex code hhh..
+.
+.
+.SH "CHARACTER TYPES"
+.rs
+.sp
+  .          any character except newline;
+               in dotall mode, any character whatsoever
+  \eC         one data unit, even in UTF mode (best avoided)
+  \ed         a decimal digit
+  \eD         a character that is not a decimal digit
+  \eh         a horizontal white space character
+  \eH         a character that is not a horizontal white space character
+  \eN         a character that is not a newline
+  \ep{\fIxx\fP}     a character with the \fIxx\fP property
+  \eP{\fIxx\fP}     a character without the \fIxx\fP property
+  \eR         a newline sequence
+  \es         a white space character
+  \eS         a character that is not a white space character
+  \ev         a vertical white space character
+  \eV         a character that is not a vertical white space character
+  \ew         a "word" character
+  \eW         a "non-word" character
+  \eX         a Unicode extended grapheme cluster
+.sp
+In PCRE, by default, \ed, \eD, \es, \eS, \ew, and \eW recognize only ASCII
+characters, even in a UTF mode. However, this can be changed by setting the
+PCRE_UCP option.
+.
+.
+.SH "GENERAL CATEGORY PROPERTIES FOR \ep and \eP"
+.rs
+.sp
+  C          Other
+  Cc         Control
+  Cf         Format
+  Cn         Unassigned
+  Co         Private use
+  Cs         Surrogate
+.sp
+  L          Letter
+  Ll         Lower case letter
+  Lm         Modifier letter
+  Lo         Other letter
+  Lt         Title case letter
+  Lu         Upper case letter
+  L&         Ll, Lu, or Lt
+.sp
+  M          Mark
+  Mc         Spacing mark
+  Me         Enclosing mark
+  Mn         Non-spacing mark
+.sp
+  N          Number
+  Nd         Decimal number
+  Nl         Letter number
+  No         Other number
+.sp
+  P          Punctuation
+  Pc         Connector punctuation
+  Pd         Dash punctuation
+  Pe         Close punctuation
+  Pf         Final punctuation
+  Pi         Initial punctuation
+  Po         Other punctuation
+  Ps         Open punctuation
+.sp
+  S          Symbol
+  Sc         Currency symbol
+  Sk         Modifier symbol
+  Sm         Mathematical symbol
+  So         Other symbol
+.sp
+  Z          Separator
+  Zl         Line separator
+  Zp         Paragraph separator
+  Zs         Space separator
+.
+.
+.SH "PCRE SPECIAL CATEGORY PROPERTIES FOR \ep and \eP"
+.rs
+.sp
+  Xan        Alphanumeric: union of properties L and N
+  Xps        POSIX space: property Z or tab, NL, VT, FF, CR
+  Xsp        Perl space: property Z or tab, NL, FF, CR
+  Xuc        Univerally-named character: one that can be
+               represented by a Universal Character Name
+  Xwd        Perl word: property Xan or underscore
+.
+.
+.SH "SCRIPT NAMES FOR \ep AND \eP"
+.rs
+.sp
+Arabic,
+Armenian,
+Avestan,
+Balinese,
+Bamum,
+Batak,
+Bengali,
+Bopomofo,
+Brahmi,
+Braille,
+Buginese,
+Buhid,
+Canadian_Aboriginal,
+Carian,
+Chakma,
+Cham,
+Cherokee,
+Common,
+Coptic,
+Cuneiform,
+Cypriot,
+Cyrillic,
+Deseret,
+Devanagari,
+Egyptian_Hieroglyphs,
+Ethiopic,
+Georgian,
+Glagolitic,
+Gothic,
+Greek,
+Gujarati,
+Gurmukhi,
+Han,
+Hangul,
+Hanunoo,
+Hebrew,
+Hiragana,
+Imperial_Aramaic,
+Inherited,
+Inscriptional_Pahlavi,
+Inscriptional_Parthian,
+Javanese,
+Kaithi,
+Kannada,
+Katakana,
+Kayah_Li,
+Kharoshthi,
+Khmer,
+Lao,
+Latin,
+Lepcha,
+Limbu,
+Linear_B,
+Lisu,
+Lycian,
+Lydian,
+Malayalam,
+Mandaic,
+Meetei_Mayek,
+Meroitic_Cursive,
+Meroitic_Hieroglyphs,
+Miao,
+Mongolian,
+Myanmar,
+New_Tai_Lue,
+Nko,
+Ogham,
+Old_Italic,
+Old_Persian,
+Old_South_Arabian,
+Old_Turkic,
+Ol_Chiki,
+Oriya,
+Osmanya,
+Phags_Pa,
+Phoenician,
+Rejang,
+Runic,
+Samaritan,
+Saurashtra,
+Sharada,
+Shavian,
+Sinhala,
+Sora_Sompeng,
+Sundanese,
+Syloti_Nagri,
+Syriac,
+Tagalog,
+Tagbanwa,
+Tai_Le,
+Tai_Tham,
+Tai_Viet,
+Takri,
+Tamil,
+Telugu,
+Thaana,
+Thai,
+Tibetan,
+Tifinagh,
+Ugaritic,
+Vai,
+Yi.
+.
+.
+.SH "CHARACTER CLASSES"
+.rs
+.sp
+  [...]       positive character class
+  [^...]      negative character class
+  [x-y]       range (can be used for hex characters)
+  [[:xxx:]]   positive POSIX named set
+  [[:^xxx:]]  negative POSIX named set
+.sp
+  alnum       alphanumeric
+  alpha       alphabetic
+  ascii       0-127
+  blank       space or tab
+  cntrl       control character
+  digit       decimal digit
+  graph       printing, excluding space
+  lower       lower case letter
+  print       printing, including space
+  punct       printing, excluding alphanumeric
+  space       white space
+  upper       upper case letter
+  word        same as \ew
+  xdigit      hexadecimal digit
+.sp
+In PCRE, POSIX character set names recognize only ASCII characters by default,
+but some of them use Unicode properties if PCRE_UCP is set. You can use
+\eQ...\eE inside a character class.
+.
+.
+.SH "QUANTIFIERS"
+.rs
+.sp
+  ?           0 or 1, greedy
+  ?+          0 or 1, possessive
+  ??          0 or 1, lazy
+  *           0 or more, greedy
+  *+          0 or more, possessive
+  *?          0 or more, lazy
+  +           1 or more, greedy
+  ++          1 or more, possessive
+  +?          1 or more, lazy
+  {n}         exactly n
+  {n,m}       at least n, no more than m, greedy
+  {n,m}+      at least n, no more than m, possessive
+  {n,m}?      at least n, no more than m, lazy
+  {n,}        n or more, greedy
+  {n,}+       n or more, possessive
+  {n,}?       n or more, lazy
+.
+.
+.SH "ANCHORS AND SIMPLE ASSERTIONS"
+.rs
+.sp
+  \eb          word boundary
+  \eB          not a word boundary
+  ^           start of subject
+               also after internal newline in multiline mode
+  \eA          start of subject
+  $           end of subject
+               also before newline at end of subject
+               also before internal newline in multiline mode
+  \eZ          end of subject
+               also before newline at end of subject
+  \ez          end of subject
+  \eG          first matching position in subject
+.
+.
+.SH "MATCH POINT RESET"
+.rs
+.sp
+  \eK          reset start of match
+.
+.
+.SH "ALTERNATION"
+.rs
+.sp
+  expr|expr|expr...
+.
+.
+.SH "CAPTURING"
+.rs
+.sp
+  (...)           capturing group
+  (?...)    named capturing group (Perl)
+  (?'name'...)    named capturing group (Perl)
+  (?P...)   named capturing group (Python)
+  (?:...)         non-capturing group
+  (?|...)         non-capturing group; reset group numbers for
+                   capturing groups in each alternative
+.
+.
+.SH "ATOMIC GROUPS"
+.rs
+.sp
+  (?>...)         atomic, non-capturing group
+.
+.
+.
+.
+.SH "COMMENT"
+.rs
+.sp
+  (?#....)        comment (not nestable)
+.
+.
+.SH "OPTION SETTING"
+.rs
+.sp
+  (?i)            caseless
+  (?J)            allow duplicate names
+  (?m)            multiline
+  (?s)            single line (dotall)
+  (?U)            default ungreedy (lazy)
+  (?x)            extended (ignore white space)
+  (?-...)         unset option(s)
+.sp
+The following are recognized only at the start of a pattern or after one of the
+newline-setting options with similar syntax:
+.sp
+  (*LIMIT_MATCH=d) set the match limit to d (decimal number)
+  (*LIMIT_RECURSION=d) set the recursion limit to d (decimal number)
+  (*NO_START_OPT) no start-match optimization (PCRE_NO_START_OPTIMIZE)
+  (*UTF8)         set UTF-8 mode: 8-bit library (PCRE_UTF8)
+  (*UTF16)        set UTF-16 mode: 16-bit library (PCRE_UTF16)
+  (*UTF32)        set UTF-32 mode: 32-bit library (PCRE_UTF32)
+  (*UTF)          set appropriate UTF mode for the library in use
+  (*UCP)          set PCRE_UCP (use Unicode properties for \ed etc)
+.
+.
+.SH "LOOKAHEAD AND LOOKBEHIND ASSERTIONS"
+.rs
+.sp
+  (?=...)         positive look ahead
+  (?!...)         negative look ahead
+  (?<=...)        positive look behind
+  (?        reference by name (Perl)
+  \ek'name'        reference by name (Perl)
+  \eg{name}        reference by name (Perl)
+  \ek{name}        reference by name (.NET)
+  (?P=name)       reference by name (Python)
+.
+.
+.SH "SUBROUTINE REFERENCES (POSSIBLY RECURSIVE)"
+.rs
+.sp
+  (?R)            recurse whole pattern
+  (?n)            call subpattern by absolute number
+  (?+n)           call subpattern by relative number
+  (?-n)           call subpattern by relative number
+  (?&name)        call subpattern by name (Perl)
+  (?P>name)       call subpattern by name (Python)
+  \eg        call subpattern by name (Oniguruma)
+  \eg'name'        call subpattern by name (Oniguruma)
+  \eg           call subpattern by absolute number (Oniguruma)
+  \eg'n'           call subpattern by absolute number (Oniguruma)
+  \eg<+n>          call subpattern by relative number (PCRE extension)
+  \eg'+n'          call subpattern by relative number (PCRE extension)
+  \eg<-n>          call subpattern by relative number (PCRE extension)
+  \eg'-n'          call subpattern by relative number (PCRE extension)
+.
+.
+.SH "CONDITIONAL PATTERNS"
+.rs
+.sp
+  (?(condition)yes-pattern)
+  (?(condition)yes-pattern|no-pattern)
+.sp
+  (?(n)...        absolute reference condition
+  (?(+n)...       relative reference condition
+  (?(-n)...       relative reference condition
+  (?()...   named reference condition (Perl)
+  (?('name')...   named reference condition (Perl)
+  (?(name)...     named reference condition (PCRE)
+  (?(R)...        overall recursion condition
+  (?(Rn)...       specific group recursion condition
+  (?(R&name)...   specific recursion condition
+  (?(DEFINE)...   define subpattern for reference
+  (?(assert)...   assertion condition
+.
+.
+.SH "BACKTRACKING CONTROL"
+.rs
+.sp
+The following act immediately they are reached:
+.sp
+  (*ACCEPT)       force successful match
+  (*FAIL)         force backtrack; synonym (*F)
+  (*MARK:NAME)    set name to be passed back; synonym (*:NAME)
+.sp
+The following act only when a subsequent match failure causes a backtrack to
+reach them. They all force a match failure, but they differ in what happens
+afterwards. Those that advance the start-of-match point do so only if the
+pattern is not anchored.
+.sp
+  (*COMMIT)       overall failure, no advance of starting point
+  (*PRUNE)        advance to next starting character
+  (*PRUNE:NAME)   equivalent to (*MARK:NAME)(*PRUNE)
+  (*SKIP)         advance to current matching position
+  (*SKIP:NAME)    advance to position corresponding to an earlier
+                  (*MARK:NAME); if not found, the (*SKIP) is ignored
+  (*THEN)         local failure, backtrack to next alternation
+  (*THEN:NAME)    equivalent to (*MARK:NAME)(*THEN)
+.
+.
+.SH "NEWLINE CONVENTIONS"
+.rs
+.sp
+These are recognized only at the very start of the pattern or after a
+(*BSR_...), (*UTF8), (*UTF16), (*UTF32) or (*UCP) option.
+.sp
+  (*CR)           carriage return only
+  (*LF)           linefeed only
+  (*CRLF)         carriage return followed by linefeed
+  (*ANYCRLF)      all three of the above
+  (*ANY)          any Unicode newline sequence
+.
+.
+.SH "WHAT \eR MATCHES"
+.rs
+.sp
+These are recognized only at the very start of the pattern or after a
+(*...) option that sets the newline convention or a UTF or UCP mode.
+.sp
+  (*BSR_ANYCRLF)  CR, LF, or CRLF
+  (*BSR_UNICODE)  any Unicode newline sequence
+.
+.
+.SH "CALLOUTS"
+.rs
+.sp
+  (?C)      callout
+  (?Cn)     callout with data n
+.
+.
+.SH "SEE ALSO"
+.rs
+.sp
+\fBpcrepattern\fP(3), \fBpcreapi\fP(3), \fBpcrecallout\fP(3),
+\fBpcrematching\fP(3), \fBpcre\fP(3).
+.
+.
+.SH AUTHOR
+.rs
+.sp
+.nf
+Philip Hazel
+University Computing Service
+Cambridge CB2 3QH, England.
+.fi
+.
+.
+.SH REVISION
+.rs
+.sp
+.nf
+Last updated: 26 April 2013
+Copyright (c) 1997-2013 University of Cambridge.
+.fi
diff --git a/pcre/doc/pcretest.1 b/pcre/doc/pcretest.1
new file mode 100644
index 00000000000..b71c897f2bb
--- /dev/null
+++ b/pcre/doc/pcretest.1
@@ -0,0 +1,1099 @@
+.TH PCRETEST 1 "26 April 2013" "PCRE 8.33"
+.SH NAME
+pcretest - a program for testing Perl-compatible regular expressions.
+.SH SYNOPSIS
+.rs
+.sp
+.B pcretest "[options] [input file [output file]]"
+.sp
+\fBpcretest\fP was written as a test program for the PCRE regular expression
+library itself, but it can also be used for experimenting with regular
+expressions. This document describes the features of the test program; for
+details of the regular expressions themselves, see the
+.\" HREF
+\fBpcrepattern\fP
+.\"
+documentation. For details of the PCRE library function calls and their
+options, see the
+.\" HREF
+\fBpcreapi\fP
+.\"
+,
+.\" HREF
+\fBpcre16\fP
+and
+.\" HREF
+\fBpcre32\fP
+.\"
+documentation.
+.P
+The input for \fBpcretest\fP is a sequence of regular expression patterns and
+strings to be matched, as described below. The output shows the result of each
+match. Options on the command line and the patterns control PCRE options and
+exactly what is output.
+.P
+As PCRE has evolved, it has acquired many different features, and as a result,
+\fBpcretest\fP now has rather a lot of obscure options for testing every
+possible feature. Some of these options are specifically designed for use in
+conjunction with the test script and data files that are distributed as part of
+PCRE, and are unlikely to be of use otherwise. They are all documented here,
+but without much justification.
+.
+.
+.SH "INPUT DATA FORMAT"
+.rs
+.sp
+Input to \fBpcretest\fP is processed line by line, either by calling the C
+library's \fBfgets()\fP function, or via the \fBlibreadline\fP library (see
+below). In Unix-like environments, \fBfgets()\fP treats any bytes other than
+newline as data characters. However, in some Windows environments character 26
+(hex 1A) causes an immediate end of file, and no further data is read. For
+maximum portability, therefore, it is safest to use only ASCII characters in
+\fBpcretest\fP input files.
+.
+.
+.SH "PCRE's 8-BIT, 16-BIT AND 32-BIT LIBRARIES"
+.rs
+.sp
+From release 8.30, two separate PCRE libraries can be built. The original one
+supports 8-bit character strings, whereas the newer 16-bit library supports
+character strings encoded in 16-bit units. From release 8.32, a third library
+can be built, supporting character strings encoded in 32-bit units. The
+\fBpcretest\fP program can be used to test all three libraries. However, it is
+itself still an 8-bit program, reading 8-bit input and writing 8-bit output.
+When testing the 16-bit or 32-bit library, the patterns and data strings are
+converted to 16- or 32-bit format before being passed to the PCRE library
+functions. Results are converted to 8-bit for output.
+.P
+References to functions and structures of the form \fBpcre[16|32]_xx\fP below
+mean "\fBpcre_xx\fP when using the 8-bit library, \fBpcre16_xx\fP when using
+the 16-bit library, or \fBpcre32_xx\fP when using the 32-bit library".
+.
+.
+.SH "COMMAND LINE OPTIONS"
+.rs
+.TP 10
+\fB-8\fP
+If both the 8-bit library has been built, this option causes the 8-bit library
+to be used (which is the default); if the 8-bit library has not been built,
+this option causes an error.
+.TP 10
+\fB-16\fP
+If both the 8-bit or the 32-bit, and the 16-bit libraries have been built, this
+option causes the 16-bit library to be used. If only the 16-bit library has been
+built, this is the default (so has no effect). If only the 8-bit or the 32-bit
+library has been built, this option causes an error.
+.TP 10
+\fB-32\fP
+If both the 8-bit or the 16-bit, and the 32-bit libraries have been built, this
+option causes the 32-bit library to be used. If only the 32-bit library has been
+built, this is the default (so has no effect). If only the 8-bit or the 16-bit
+library has been built, this option causes an error.
+.TP 10
+\fB-b\fP
+Behave as if each pattern has the \fB/B\fP (show byte code) modifier; the
+internal form is output after compilation.
+.TP 10
+\fB-C\fP
+Output the version number of the PCRE library, and all available information
+about the optional features that are included, and then exit with zero exit
+code. All other options are ignored.
+.TP 10
+\fB-C\fP \fIoption\fP
+Output information about a specific build-time option, then exit. This
+functionality is intended for use in scripts such as \fBRunTest\fP. The
+following options output the value and set the exit code as indicated:
+.sp
+  ebcdic-nl  the code for LF (= NL) in an EBCDIC environment:
+               0x15 or 0x25
+               0 if used in an ASCII environment
+               exit code is always 0
+  linksize   the configured internal link size (2, 3, or 4)
+               exit code is set to the link size
+  newline    the default newline setting:
+               CR, LF, CRLF, ANYCRLF, or ANY
+               exit code is always 0
+.sp
+The following options output 1 for true or 0 for false, and set the exit code
+to the same value:
+.sp
+  ebcdic     compiled for an EBCDIC environment
+  jit        just-in-time support is available
+  pcre16     the 16-bit library was built
+  pcre32     the 32-bit library was built
+  pcre8      the 8-bit library was built
+  ucp        Unicode property support is available
+  utf        UTF-8 and/or UTF-16 and/or UTF-32 support
+               is available
+.sp
+If an unknown option is given, an error message is output; the exit code is 0.
+.TP 10
+\fB-d\fP
+Behave as if each pattern has the \fB/D\fP (debug) modifier; the internal
+form and information about the compiled pattern is output after compilation;
+\fB-d\fP is equivalent to \fB-b -i\fP.
+.TP 10
+\fB-dfa\fP
+Behave as if each data line contains the \eD escape sequence; this causes the
+alternative matching function, \fBpcre[16|32]_dfa_exec()\fP, to be used instead
+of the standard \fBpcre[16|32]_exec()\fP function (more detail is given below).
+.TP 10
+\fB-help\fP
+Output a brief summary these options and then exit.
+.TP 10
+\fB-i\fP
+Behave as if each pattern has the \fB/I\fP modifier; information about the
+compiled pattern is given after compilation.
+.TP 10
+\fB-M\fP
+Behave as if each data line contains the \eM escape sequence; this causes
+PCRE to discover the minimum MATCH_LIMIT and MATCH_LIMIT_RECURSION settings by
+calling \fBpcre[16|32]_exec()\fP repeatedly with different limits.
+.TP 10
+\fB-m\fP
+Output the size of each compiled pattern after it has been compiled. This is
+equivalent to adding \fB/M\fP to each regular expression. The size is given in
+bytes for both libraries.
+.TP 10
+\fB-o\fP \fIosize\fP
+Set the number of elements in the output vector that is used when calling
+\fBpcre[16|32]_exec()\fP or \fBpcre[16|32]_dfa_exec()\fP to be \fIosize\fP. The
+default value is 45, which is enough for 14 capturing subexpressions for
+\fBpcre[16|32]_exec()\fP or 22 different matches for
+\fBpcre[16|32]_dfa_exec()\fP.
+The vector size can be changed for individual matching calls by including \eO
+in the data line (see below).
+.TP 10
+\fB-p\fP
+Behave as if each pattern has the \fB/P\fP modifier; the POSIX wrapper API is
+used to call PCRE. None of the other options has any effect when \fB-p\fP is
+set. This option can be used only with the 8-bit library.
+.TP 10
+\fB-q\fP
+Do not output the version number of \fBpcretest\fP at the start of execution.
+.TP 10
+\fB-S\fP \fIsize\fP
+On Unix-like systems, set the size of the run-time stack to \fIsize\fP
+megabytes.
+.TP 10
+\fB-s\fP or \fB-s+\fP
+Behave as if each pattern has the \fB/S\fP modifier; in other words, force each
+pattern to be studied. If \fB-s+\fP is used, all the JIT compile options are
+passed to \fBpcre[16|32]_study()\fP, causing just-in-time optimization to be set
+up if it is available, for both full and partial matching. Specific JIT compile
+options can be selected by following \fB-s+\fP with a digit in the range 1 to
+7, which selects the JIT compile modes as follows:
+.sp
+  1  normal match only
+  2  soft partial match only
+  3  normal match and soft partial match
+  4  hard partial match only
+  6  soft and hard partial match
+  7  all three modes (default)
+.sp
+If \fB-s++\fP is used instead of \fB-s+\fP (with or without a following digit),
+the text "(JIT)" is added to the first output line after a match or no match
+when JIT-compiled code was actually used.
+.sp
+Note that there are pattern options that can override \fB-s\fP, either
+specifying no studying at all, or suppressing JIT compilation.
+.sp
+If the \fB/I\fP or \fB/D\fP option is present on a pattern (requesting output
+about the compiled pattern), information about the result of studying is not
+included when studying is caused only by \fB-s\fP and neither \fB-i\fP nor
+\fB-d\fP is present on the command line. This behaviour means that the output
+from tests that are run with and without \fB-s\fP should be identical, except
+when options that output information about the actual running of a match are
+set.
+.sp
+The \fB-M\fP, \fB-t\fP, and \fB-tm\fP options, which give information about
+resources used, are likely to produce different output with and without
+\fB-s\fP. Output may also differ if the \fB/C\fP option is present on an
+individual pattern. This uses callouts to trace the the matching process, and
+this may be different between studied and non-studied patterns. If the pattern
+contains (*MARK) items there may also be differences, for the same reason. The
+\fB-s\fP command line option can be overridden for specific patterns that
+should never be studied (see the \fB/S\fP pattern modifier below).
+.TP 10
+\fB-t\fP
+Run each compile, study, and match many times with a timer, and output
+resulting time per compile or match (in milliseconds). Do not set \fB-m\fP with
+\fB-t\fP, because you will then get the size output a zillion times, and the
+timing will be distorted. You can control the number of iterations that are
+used for timing by following \fB-t\fP with a number (as a separate item on the
+command line). For example, "-t 1000" would iterate 1000 times. The default is
+to iterate 500000 times.
+.TP 10
+\fB-tm\fP
+This is like \fB-t\fP except that it times only the matching phase, not the
+compile or study phases.
+.
+.
+.SH DESCRIPTION
+.rs
+.sp
+If \fBpcretest\fP is given two filename arguments, it reads from the first and
+writes to the second. If it is given only one filename argument, it reads from
+that file and writes to stdout. Otherwise, it reads from stdin and writes to
+stdout, and prompts for each line of input, using "re>" to prompt for regular
+expressions, and "data>" to prompt for data lines.
+.P
+When \fBpcretest\fP is built, a configuration option can specify that it should
+be linked with the \fBlibreadline\fP library. When this is done, if the input
+is from a terminal, it is read using the \fBreadline()\fP function. This
+provides line-editing and history facilities. The output from the \fB-help\fP
+option states whether or not \fBreadline()\fP will be used.
+.P
+The program handles any number of sets of input on a single input file. Each
+set starts with a regular expression, and continues with any number of data
+lines to be matched against the pattern.
+.P
+Each data line is matched separately and independently. If you want to do
+multi-line matches, you have to use the \en escape sequence (or \er or \er\en,
+etc., depending on the newline setting) in a single line of input to encode the
+newline sequences. There is no limit on the length of data lines; the input
+buffer is automatically extended if it is too small.
+.P
+An empty line signals the end of the data lines, at which point a new regular
+expression is read. The regular expressions are given enclosed in any
+non-alphanumeric delimiters other than backslash, for example:
+.sp
+  /(a|bc)x+yz/
+.sp
+White space before the initial delimiter is ignored. A regular expression may
+be continued over several input lines, in which case the newline characters are
+included within it. It is possible to include the delimiter within the pattern
+by escaping it, for example
+.sp
+  /abc\e/def/
+.sp
+If you do so, the escape and the delimiter form part of the pattern, but since
+delimiters are always non-alphanumeric, this does not affect its interpretation.
+If the terminating delimiter is immediately followed by a backslash, for
+example,
+.sp
+  /abc/\e
+.sp
+then a backslash is added to the end of the pattern. This is done to provide a
+way of testing the error condition that arises if a pattern finishes with a
+backslash, because
+.sp
+  /abc\e/
+.sp
+is interpreted as the first line of a pattern that starts with "abc/", causing
+pcretest to read the next line as a continuation of the regular expression.
+.
+.
+.SH "PATTERN MODIFIERS"
+.rs
+.sp
+A pattern may be followed by any number of modifiers, which are mostly single
+characters, though some of these can be qualified by further characters.
+Following Perl usage, these are referred to below as, for example, "the
+\fB/i\fP modifier", even though the delimiter of the pattern need not always be
+a slash, and no slash is used when writing modifiers. White space may appear
+between the final pattern delimiter and the first modifier, and between the
+modifiers themselves. For reference, here is a complete list of modifiers. They
+fall into several groups that are described in detail in the following
+sections.
+.sp
+  \fB/8\fP              set UTF mode
+  \fB/9\fP              set PCRE_NEVER_UTF (locks out UTF mode)
+  \fB/?\fP              disable UTF validity check
+  \fB/+\fP              show remainder of subject after match
+  \fB/=\fP              show all captures (not just those that are set)
+.sp
+  \fB/A\fP              set PCRE_ANCHORED
+  \fB/B\fP              show compiled code
+  \fB/C\fP              set PCRE_AUTO_CALLOUT
+  \fB/D\fP              same as \fB/B\fP plus \fB/I\fP
+  \fB/E\fP              set PCRE_DOLLAR_ENDONLY
+  \fB/F\fP              flip byte order in compiled pattern
+  \fB/f\fP              set PCRE_FIRSTLINE
+  \fB/G\fP              find all matches (shorten string)
+  \fB/g\fP              find all matches (use startoffset)
+  \fB/I\fP              show information about pattern
+  \fB/i\fP              set PCRE_CASELESS
+  \fB/J\fP              set PCRE_DUPNAMES
+  \fB/K\fP              show backtracking control names
+  \fB/L\fP              set locale
+  \fB/M\fP              show compiled memory size
+  \fB/m\fP              set PCRE_MULTILINE
+  \fB/N\fP              set PCRE_NO_AUTO_CAPTURE
+  \fB/P\fP              use the POSIX wrapper
+  \fB/S\fP              study the pattern after compilation
+  \fB/s\fP              set PCRE_DOTALL
+  \fB/T\fP              select character tables
+  \fB/U\fP              set PCRE_UNGREEDY
+  \fB/W\fP              set PCRE_UCP
+  \fB/X\fP              set PCRE_EXTRA
+  \fB/x\fP              set PCRE_EXTENDED
+  \fB/Y\fP              set PCRE_NO_START_OPTIMIZE
+  \fB/Z\fP              don't show lengths in \fB/B\fP output
+.sp
+  \fB/\fP          set PCRE_NEWLINE_ANY
+  \fB/\fP      set PCRE_NEWLINE_ANYCRLF
+  \fB/\fP           set PCRE_NEWLINE_CR
+  \fB/\fP         set PCRE_NEWLINE_CRLF
+  \fB/\fP           set PCRE_NEWLINE_LF
+  \fB/\fP  set PCRE_BSR_ANYCRLF
+  \fB/\fP  set PCRE_BSR_UNICODE
+  \fB/\fP           set PCRE_JAVASCRIPT_COMPAT
+.sp
+.
+.
+.SS "Perl-compatible modifiers"
+.rs
+.sp
+The \fB/i\fP, \fB/m\fP, \fB/s\fP, and \fB/x\fP modifiers set the PCRE_CASELESS,
+PCRE_MULTILINE, PCRE_DOTALL, or PCRE_EXTENDED options, respectively, when
+\fBpcre[16|32]_compile()\fP is called. These four modifier letters have the same
+effect as they do in Perl. For example:
+.sp
+  /caseless/i
+.sp
+.
+.
+.SS "Modifiers for other PCRE options"
+.rs
+.sp
+The following table shows additional modifiers for setting PCRE compile-time
+options that do not correspond to anything in Perl:
+.sp
+  \fB/8\fP              PCRE_UTF8           ) when using the 8-bit
+  \fB/?\fP              PCRE_NO_UTF8_CHECK  )   library
+.sp
+  \fB/8\fP              PCRE_UTF16          ) when using the 16-bit
+  \fB/?\fP              PCRE_NO_UTF16_CHECK )   library
+.sp
+  \fB/8\fP              PCRE_UTF32          ) when using the 32-bit
+  \fB/?\fP              PCRE_NO_UTF32_CHECK )   library
+.sp
+  \fB/9\fP              PCRE_NEVER_UTF
+  \fB/A\fP              PCRE_ANCHORED
+  \fB/C\fP              PCRE_AUTO_CALLOUT
+  \fB/E\fP              PCRE_DOLLAR_ENDONLY
+  \fB/f\fP              PCRE_FIRSTLINE
+  \fB/J\fP              PCRE_DUPNAMES
+  \fB/N\fP              PCRE_NO_AUTO_CAPTURE
+  \fB/U\fP              PCRE_UNGREEDY
+  \fB/W\fP              PCRE_UCP
+  \fB/X\fP              PCRE_EXTRA
+  \fB/Y\fP              PCRE_NO_START_OPTIMIZE
+  \fB/\fP          PCRE_NEWLINE_ANY
+  \fB/\fP      PCRE_NEWLINE_ANYCRLF
+  \fB/\fP           PCRE_NEWLINE_CR
+  \fB/\fP         PCRE_NEWLINE_CRLF
+  \fB/\fP           PCRE_NEWLINE_LF
+  \fB/\fP  PCRE_BSR_ANYCRLF
+  \fB/\fP  PCRE_BSR_UNICODE
+  \fB/\fP           PCRE_JAVASCRIPT_COMPAT
+.sp
+The modifiers that are enclosed in angle brackets are literal strings as shown,
+including the angle brackets, but the letters within can be in either case.
+This example sets multiline matching with CRLF as the line ending sequence:
+.sp
+  /^abc/m
+.sp
+As well as turning on the PCRE_UTF8/16/32 option, the \fB/8\fP modifier causes
+all non-printing characters in output strings to be printed using the
+\ex{hh...} notation. Otherwise, those less than 0x100 are output in hex without
+the curly brackets.
+.P
+Full details of the PCRE options are given in the
+.\" HREF
+\fBpcreapi\fP
+.\"
+documentation.
+.
+.
+.SS "Finding all matches in a string"
+.rs
+.sp
+Searching for all possible matches within each subject string can be requested
+by the \fB/g\fP or \fB/G\fP modifier. After finding a match, PCRE is called
+again to search the remainder of the subject string. The difference between
+\fB/g\fP and \fB/G\fP is that the former uses the \fIstartoffset\fP argument to
+\fBpcre[16|32]_exec()\fP to start searching at a new point within the entire
+string (which is in effect what Perl does), whereas the latter passes over a
+shortened substring. This makes a difference to the matching process if the
+pattern begins with a lookbehind assertion (including \eb or \eB).
+.P
+If any call to \fBpcre[16|32]_exec()\fP in a \fB/g\fP or \fB/G\fP sequence matches
+an empty string, the next call is done with the PCRE_NOTEMPTY_ATSTART and
+PCRE_ANCHORED flags set in order to search for another, non-empty, match at the
+same point. If this second match fails, the start offset is advanced, and the
+normal match is retried. This imitates the way Perl handles such cases when
+using the \fB/g\fP modifier or the \fBsplit()\fP function. Normally, the start
+offset is advanced by one character, but if the newline convention recognizes
+CRLF as a newline, and the current character is CR followed by LF, an advance
+of two is used.
+.
+.
+.SS "Other modifiers"
+.rs
+.sp
+There are yet more modifiers for controlling the way \fBpcretest\fP
+operates.
+.P
+The \fB/+\fP modifier requests that as well as outputting the substring that
+matched the entire pattern, \fBpcretest\fP should in addition output the
+remainder of the subject string. This is useful for tests where the subject
+contains multiple copies of the same substring. If the \fB+\fP modifier appears
+twice, the same action is taken for captured substrings. In each case the
+remainder is output on the following line with a plus character following the
+capture number. Note that this modifier must not immediately follow the /S
+modifier because /S+ and /S++ have other meanings.
+.P
+The \fB/=\fP modifier requests that the values of all potential captured
+parentheses be output after a match. By default, only those up to the highest
+one actually used in the match are output (corresponding to the return code
+from \fBpcre[16|32]_exec()\fP). Values in the offsets vector corresponding to
+higher numbers should be set to -1, and these are output as "". This
+modifier gives a way of checking that this is happening.
+.P
+The \fB/B\fP modifier is a debugging feature. It requests that \fBpcretest\fP
+output a representation of the compiled code after compilation. Normally this
+information contains length and offset values; however, if \fB/Z\fP is also
+present, this data is replaced by spaces. This is a special feature for use in
+the automatic test scripts; it ensures that the same output is generated for
+different internal link sizes.
+.P
+The \fB/D\fP modifier is a PCRE debugging feature, and is equivalent to
+\fB/BI\fP, that is, both the \fB/B\fP and the \fB/I\fP modifiers.
+.P
+The \fB/F\fP modifier causes \fBpcretest\fP to flip the byte order of the
+2-byte and 4-byte fields in the compiled pattern. This facility is for testing
+the feature in PCRE that allows it to execute patterns that were compiled on a
+host with a different endianness. This feature is not available when the POSIX
+interface to PCRE is being used, that is, when the \fB/P\fP pattern modifier is
+specified. See also the section about saving and reloading compiled patterns
+below.
+.P
+The \fB/I\fP modifier requests that \fBpcretest\fP output information about the
+compiled pattern (whether it is anchored, has a fixed first character, and
+so on). It does this by calling \fBpcre[16|32]_fullinfo()\fP after compiling a
+pattern. If the pattern is studied, the results of that are also output.
+.P
+The \fB/K\fP modifier requests \fBpcretest\fP to show names from backtracking
+control verbs that are returned from calls to \fBpcre[16|32]_exec()\fP. It causes
+\fBpcretest\fP to create a \fBpcre[16|32]_extra\fP block if one has not already
+been created by a call to \fBpcre[16|32]_study()\fP, and to set the
+PCRE_EXTRA_MARK flag and the \fBmark\fP field within it, every time that
+\fBpcre[16|32]_exec()\fP is called. If the variable that the \fBmark\fP field
+points to is non-NULL for a match, non-match, or partial match, \fBpcretest\fP
+prints the string to which it points. For a match, this is shown on a line by
+itself, tagged with "MK:". For a non-match it is added to the message.
+.P
+The \fB/L\fP modifier must be followed directly by the name of a locale, for
+example,
+.sp
+  /pattern/Lfr_FR
+.sp
+For this reason, it must be the last modifier. The given locale is set,
+\fBpcre[16|32]_maketables()\fP is called to build a set of character tables for
+the locale, and this is then passed to \fBpcre[16|32]_compile()\fP when compiling
+the regular expression. Without an \fB/L\fP (or \fB/T\fP) modifier, NULL is
+passed as the tables pointer; that is, \fB/L\fP applies only to the expression
+on which it appears.
+.P
+The \fB/M\fP modifier causes the size in bytes of the memory block used to hold
+the compiled pattern to be output. This does not include the size of the
+\fBpcre[16|32]\fP block; it is just the actual compiled data. If the pattern is
+successfully studied with the PCRE_STUDY_JIT_COMPILE option, the size of the
+JIT compiled code is also output.
+.P
+The \fB/S\fP modifier causes \fBpcre[16|32]_study()\fP to be called after the
+expression has been compiled, and the results used when the expression is
+matched. There are a number of qualifying characters that may follow \fB/S\fP.
+They may appear in any order.
+.P
+If \fBS\fP is followed by an exclamation mark, \fBpcre[16|32]_study()\fP is called
+with the PCRE_STUDY_EXTRA_NEEDED option, causing it always to return a
+\fBpcre_extra\fP block, even when studying discovers no useful information.
+.P
+If \fB/S\fP is followed by a second S character, it suppresses studying, even
+if it was requested externally by the \fB-s\fP command line option. This makes
+it possible to specify that certain patterns are always studied, and others are
+never studied, independently of \fB-s\fP. This feature is used in the test
+files in a few cases where the output is different when the pattern is studied.
+.P
+If the \fB/S\fP modifier is followed by a + character, the call to
+\fBpcre[16|32]_study()\fP is made with all the JIT study options, requesting
+just-in-time optimization support if it is available, for both normal and
+partial matching. If you want to restrict the JIT compiling modes, you can
+follow \fB/S+\fP with a digit in the range 1 to 7:
+.sp
+  1  normal match only
+  2  soft partial match only
+  3  normal match and soft partial match
+  4  hard partial match only
+  6  soft and hard partial match
+  7  all three modes (default)
+.sp
+If \fB/S++\fP is used instead of \fB/S+\fP (with or without a following digit),
+the text "(JIT)" is added to the first output line after a match or no match
+when JIT-compiled code was actually used.
+.P
+Note that there is also an independent \fB/+\fP modifier; it must not be given
+immediately after \fB/S\fP or \fB/S+\fP because this will be misinterpreted.
+.P
+If JIT studying is successful, the compiled JIT code will automatically be used
+when \fBpcre[16|32]_exec()\fP is run, except when incompatible run-time options
+are specified. For more details, see the
+.\" HREF
+\fBpcrejit\fP
+.\"
+documentation. See also the \fB\eJ\fP escape sequence below for a way of
+setting the size of the JIT stack.
+.P
+Finally, if \fB/S\fP is followed by a minus character, JIT compilation is
+suppressed, even if it was requested externally by the \fB-s\fP command line
+option. This makes it possible to specify that JIT is never to be used for
+certain patterns.
+.P
+The \fB/T\fP modifier must be followed by a single digit. It causes a specific
+set of built-in character tables to be passed to \fBpcre[16|32]_compile()\fP. It
+is used in the standard PCRE tests to check behaviour with different character
+tables. The digit specifies the tables as follows:
+.sp
+  0   the default ASCII tables, as distributed in
+        pcre_chartables.c.dist
+  1   a set of tables defining ISO 8859 characters
+.sp
+In table 1, some characters whose codes are greater than 128 are identified as
+letters, digits, spaces, etc.
+.
+.
+.SS "Using the POSIX wrapper API"
+.rs
+.sp
+The \fB/P\fP modifier causes \fBpcretest\fP to call PCRE via the POSIX wrapper
+API rather than its native API. This supports only the 8-bit library. When
+\fB/P\fP is set, the following modifiers set options for the \fBregcomp()\fP
+function:
+.sp
+  /i    REG_ICASE
+  /m    REG_NEWLINE
+  /N    REG_NOSUB
+  /s    REG_DOTALL     )
+  /U    REG_UNGREEDY   ) These options are not part of
+  /W    REG_UCP        )   the POSIX standard
+  /8    REG_UTF8       )
+.sp
+The \fB/+\fP modifier works as described above. All other modifiers are
+ignored.
+.
+.
+.SH "DATA LINES"
+.rs
+.sp
+Before each data line is passed to \fBpcre[16|32]_exec()\fP, leading and trailing
+white space is removed, and it is then scanned for \e escapes. Some of these
+are pretty esoteric features, intended for checking out some of the more
+complicated features of PCRE. If you are just testing "ordinary" regular
+expressions, you probably don't need any of these. The following escapes are
+recognized:
+.sp
+  \ea         alarm (BEL, \ex07)
+  \eb         backspace (\ex08)
+  \ee         escape (\ex27)
+  \ef         form feed (\ex0c)
+  \en         newline (\ex0a)
+.\" JOIN
+  \eqdd       set the PCRE_MATCH_LIMIT limit to dd
+               (any number of digits)
+  \er         carriage return (\ex0d)
+  \et         tab (\ex09)
+  \ev         vertical tab (\ex0b)
+  \ennn       octal character (up to 3 octal digits); always
+               a byte unless > 255 in UTF-8 or 16-bit or 32-bit mode
+  \exhh       hexadecimal byte (up to 2 hex digits)
+  \ex{hh...}  hexadecimal character (any number of hex digits)
+.\" JOIN
+  \eA         pass the PCRE_ANCHORED option to \fBpcre[16|32]_exec()\fP
+               or \fBpcre[16|32]_dfa_exec()\fP
+.\" JOIN
+  \eB         pass the PCRE_NOTBOL option to \fBpcre[16|32]_exec()\fP
+               or \fBpcre[16|32]_dfa_exec()\fP
+.\" JOIN
+  \eCdd       call pcre[16|32]_copy_substring() for substring dd
+               after a successful match (number less than 32)
+.\" JOIN
+  \eCname     call pcre[16|32]_copy_named_substring() for substring
+               "name" after a successful match (name termin-
+               ated by next non alphanumeric character)
+.\" JOIN
+  \eC+        show the current captured substrings at callout
+               time
+  \eC-        do not supply a callout function
+.\" JOIN
+  \eC!n       return 1 instead of 0 when callout number n is
+               reached
+.\" JOIN
+  \eC!n!m     return 1 instead of 0 when callout number n is
+               reached for the nth time
+.\" JOIN
+  \eC*n       pass the number n (may be negative) as callout
+               data; this is used as the callout return value
+  \eD         use the \fBpcre[16|32]_dfa_exec()\fP match function
+  \eF         only shortest match for \fBpcre[16|32]_dfa_exec()\fP
+.\" JOIN
+  \eGdd       call pcre[16|32]_get_substring() for substring dd
+               after a successful match (number less than 32)
+.\" JOIN
+  \eGname     call pcre[16|32]_get_named_substring() for substring
+               "name" after a successful match (name termin-
+               ated by next non-alphanumeric character)
+.\" JOIN
+  \eJdd       set up a JIT stack of dd kilobytes maximum (any
+               number of digits)
+.\" JOIN
+  \eL         call pcre[16|32]_get_substringlist() after a
+               successful match
+.\" JOIN
+  \eM         discover the minimum MATCH_LIMIT and
+               MATCH_LIMIT_RECURSION settings
+.\" JOIN
+  \eN         pass the PCRE_NOTEMPTY option to \fBpcre[16|32]_exec()\fP
+               or \fBpcre[16|32]_dfa_exec()\fP; if used twice, pass the
+               PCRE_NOTEMPTY_ATSTART option
+.\" JOIN
+  \eOdd       set the size of the output vector passed to
+               \fBpcre[16|32]_exec()\fP to dd (any number of digits)
+.\" JOIN
+  \eP         pass the PCRE_PARTIAL_SOFT option to \fBpcre[16|32]_exec()\fP
+               or \fBpcre[16|32]_dfa_exec()\fP; if used twice, pass the
+               PCRE_PARTIAL_HARD option
+.\" JOIN
+  \eQdd       set the PCRE_MATCH_LIMIT_RECURSION limit to dd
+               (any number of digits)
+  \eR         pass the PCRE_DFA_RESTART option to \fBpcre[16|32]_dfa_exec()\fP
+  \eS         output details of memory get/free calls during matching
+.\" JOIN
+  \eY         pass the PCRE_NO_START_OPTIMIZE option to \fBpcre[16|32]_exec()\fP
+               or \fBpcre[16|32]_dfa_exec()\fP
+.\" JOIN
+  \eZ         pass the PCRE_NOTEOL option to \fBpcre[16|32]_exec()\fP
+               or \fBpcre[16|32]_dfa_exec()\fP
+.\" JOIN
+  \e?         pass the PCRE_NO_UTF[8|16|32]_CHECK option to
+               \fBpcre[16|32]_exec()\fP or \fBpcre[16|32]_dfa_exec()\fP
+.\" JOIN
+  \e>dd       start the match at offset dd (optional "-"; then
+               any number of digits); this sets the \fIstartoffset\fP
+               argument for \fBpcre[16|32]_exec()\fP or \fBpcre[16|32]_dfa_exec()\fP
+.\" JOIN
+  \e      pass the PCRE_NEWLINE_CR option to \fBpcre[16|32]_exec()\fP
+               or \fBpcre[16|32]_dfa_exec()\fP
+.\" JOIN
+  \e      pass the PCRE_NEWLINE_LF option to \fBpcre[16|32]_exec()\fP
+               or \fBpcre[16|32]_dfa_exec()\fP
+.\" JOIN
+  \e    pass the PCRE_NEWLINE_CRLF option to \fBpcre[16|32]_exec()\fP
+               or \fBpcre[16|32]_dfa_exec()\fP
+.\" JOIN
+  \e pass the PCRE_NEWLINE_ANYCRLF option to \fBpcre[16|32]_exec()\fP
+               or \fBpcre[16|32]_dfa_exec()\fP
+.\" JOIN
+  \e     pass the PCRE_NEWLINE_ANY option to \fBpcre[16|32]_exec()\fP
+               or \fBpcre[16|32]_dfa_exec()\fP
+.sp
+The use of \ex{hh...} is not dependent on the use of the \fB/8\fP modifier on
+the pattern. It is recognized always. There may be any number of hexadecimal
+digits inside the braces; invalid values provoke error messages.
+.P
+Note that \exhh specifies one byte rather than one character in UTF-8 mode;
+this makes it possible to construct invalid UTF-8 sequences for testing
+purposes. On the other hand, \ex{hh} is interpreted as a UTF-8 character in
+UTF-8 mode, generating more than one byte if the value is greater than 127.
+When testing the 8-bit library not in UTF-8 mode, \ex{hh} generates one byte
+for values less than 256, and causes an error for greater values.
+.P
+In UTF-16 mode, all 4-digit \ex{hhhh} values are accepted. This makes it
+possible to construct invalid UTF-16 sequences for testing purposes.
+.P
+In UTF-32 mode, all 4- to 8-digit \ex{...} values are accepted. This makes it
+possible to construct invalid UTF-32 sequences for testing purposes.
+.P
+The escapes that specify line ending sequences are literal strings, exactly as
+shown. No more than one newline setting should be present in any data line.
+.P
+A backslash followed by anything else just escapes the anything else. If
+the very last character is a backslash, it is ignored. This gives a way of
+passing an empty line as data, since a real empty line terminates the data
+input.
+.P
+The \fB\eJ\fP escape provides a way of setting the maximum stack size that is
+used by the just-in-time optimization code. It is ignored if JIT optimization
+is not being used. Providing a stack that is larger than the default 32K is
+necessary only for very complicated patterns.
+.P
+If \eM is present, \fBpcretest\fP calls \fBpcre[16|32]_exec()\fP several times,
+with different values in the \fImatch_limit\fP and \fImatch_limit_recursion\fP
+fields of the \fBpcre[16|32]_extra\fP data structure, until it finds the minimum
+numbers for each parameter that allow \fBpcre[16|32]_exec()\fP to complete without
+error. Because this is testing a specific feature of the normal interpretive
+\fBpcre[16|32]_exec()\fP execution, the use of any JIT optimization that might
+have been set up by the \fB/S+\fP qualifier of \fB-s+\fP option is disabled.
+.P
+The \fImatch_limit\fP number is a measure of the amount of backtracking
+that takes place, and checking it out can be instructive. For most simple
+matches, the number is quite small, but for patterns with very large numbers of
+matching possibilities, it can become large very quickly with increasing length
+of subject string. The \fImatch_limit_recursion\fP number is a measure of how
+much stack (or, if PCRE is compiled with NO_RECURSE, how much heap) memory is
+needed to complete the match attempt.
+.P
+When \eO is used, the value specified may be higher or lower than the size set
+by the \fB-O\fP command line option (or defaulted to 45); \eO applies only to
+the call of \fBpcre[16|32]_exec()\fP for the line in which it appears.
+.P
+If the \fB/P\fP modifier was present on the pattern, causing the POSIX wrapper
+API to be used, the only option-setting sequences that have any effect are \eB,
+\eN, and \eZ, causing REG_NOTBOL, REG_NOTEMPTY, and REG_NOTEOL, respectively,
+to be passed to \fBregexec()\fP.
+.
+.
+.SH "THE ALTERNATIVE MATCHING FUNCTION"
+.rs
+.sp
+By default, \fBpcretest\fP uses the standard PCRE matching function,
+\fBpcre[16|32]_exec()\fP to match each data line. PCRE also supports an
+alternative matching function, \fBpcre[16|32]_dfa_test()\fP, which operates in a
+different way, and has some restrictions. The differences between the two
+functions are described in the
+.\" HREF
+\fBpcrematching\fP
+.\"
+documentation.
+.P
+If a data line contains the \eD escape sequence, or if the command line
+contains the \fB-dfa\fP option, the alternative matching function is used.
+This function finds all possible matches at a given point. If, however, the \eF
+escape sequence is present in the data line, it stops after the first match is
+found. This is always the shortest possible match.
+.
+.
+.SH "DEFAULT OUTPUT FROM PCRETEST"
+.rs
+.sp
+This section describes the output when the normal matching function,
+\fBpcre[16|32]_exec()\fP, is being used.
+.P
+When a match succeeds, \fBpcretest\fP outputs the list of captured substrings
+that \fBpcre[16|32]_exec()\fP returns, starting with number 0 for the string that
+matched the whole pattern. Otherwise, it outputs "No match" when the return is
+PCRE_ERROR_NOMATCH, and "Partial match:" followed by the partially matching
+substring when \fBpcre[16|32]_exec()\fP returns PCRE_ERROR_PARTIAL. (Note that
+this is the entire substring that was inspected during the partial match; it
+may include characters before the actual match start if a lookbehind assertion,
+\eK, \eb, or \eB was involved.) For any other return, \fBpcretest\fP outputs
+the PCRE negative error number and a short descriptive phrase. If the error is
+a failed UTF string check, the offset of the start of the failing character and
+the reason code are also output, provided that the size of the output vector is
+at least two. Here is an example of an interactive \fBpcretest\fP run.
+.sp
+  $ pcretest
+  PCRE version 8.13 2011-04-30
+.sp
+    re> /^abc(\ed+)/
+  data> abc123
+   0: abc123
+   1: 123
+  data> xyz
+  No match
+.sp
+Unset capturing substrings that are not followed by one that is set are not
+returned by \fBpcre[16|32]_exec()\fP, and are not shown by \fBpcretest\fP. In the
+following example, there are two capturing substrings, but when the first data
+line is matched, the second, unset substring is not shown. An "internal" unset
+substring is shown as "", as for the second data line.
+.sp
+    re> /(a)|(b)/
+  data> a
+   0: a
+   1: a
+  data> b
+   0: b
+   1: 
+   2: b
+.sp
+If the strings contain any non-printing characters, they are output as \exhh
+escapes if the value is less than 256 and UTF mode is not set. Otherwise they
+are output as \ex{hh...} escapes. See below for the definition of non-printing
+characters. If the pattern has the \fB/+\fP modifier, the output for substring
+0 is followed by the the rest of the subject string, identified by "0+" like
+this:
+.sp
+    re> /cat/+
+  data> cataract
+   0: cat
+   0+ aract
+.sp
+If the pattern has the \fB/g\fP or \fB/G\fP modifier, the results of successive
+matching attempts are output in sequence, like this:
+.sp
+    re> /\eBi(\ew\ew)/g
+  data> Mississippi
+   0: iss
+   1: ss
+   0: iss
+   1: ss
+   0: ipp
+   1: pp
+.sp
+"No match" is output only if the first match attempt fails. Here is an example
+of a failure message (the offset 4 that is specified by \e>4 is past the end of
+the subject string):
+.sp
+    re> /xyz/
+  data> xyz\e>4
+  Error -24 (bad offset value)
+.P
+If any of the sequences \fB\eC\fP, \fB\eG\fP, or \fB\eL\fP are present in a
+data line that is successfully matched, the substrings extracted by the
+convenience functions are output with C, G, or L after the string number
+instead of a colon. This is in addition to the normal full list. The string
+length (that is, the return from the extraction function) is given in
+parentheses after each string for \fB\eC\fP and \fB\eG\fP.
+.P
+Note that whereas patterns can be continued over several lines (a plain ">"
+prompt is used for continuations), data lines may not. However newlines can be
+included in data by means of the \en escape (or \er, \er\en, etc., depending on
+the newline sequence setting).
+.
+.
+.
+.SH "OUTPUT FROM THE ALTERNATIVE MATCHING FUNCTION"
+.rs
+.sp
+When the alternative matching function, \fBpcre[16|32]_dfa_exec()\fP, is used (by
+means of the \eD escape sequence or the \fB-dfa\fP command line option), the
+output consists of a list of all the matches that start at the first point in
+the subject where there is at least one match. For example:
+.sp
+    re> /(tang|tangerine|tan)/
+  data> yellow tangerine\eD
+   0: tangerine
+   1: tang
+   2: tan
+.sp
+(Using the normal matching function on this data finds only "tang".) The
+longest matching string is always given first (and numbered zero). After a
+PCRE_ERROR_PARTIAL return, the output is "Partial match:", followed by the
+partially matching substring. (Note that this is the entire substring that was
+inspected during the partial match; it may include characters before the actual
+match start if a lookbehind assertion, \eK, \eb, or \eB was involved.)
+.P
+If \fB/g\fP is present on the pattern, the search for further matches resumes
+at the end of the longest match. For example:
+.sp
+    re> /(tang|tangerine|tan)/g
+  data> yellow tangerine and tangy sultana\eD
+   0: tangerine
+   1: tang
+   2: tan
+   0: tang
+   1: tan
+   0: tan
+.sp
+Since the matching function does not support substring capture, the escape
+sequences that are concerned with captured substrings are not relevant.
+.
+.
+.SH "RESTARTING AFTER A PARTIAL MATCH"
+.rs
+.sp
+When the alternative matching function has given the PCRE_ERROR_PARTIAL return,
+indicating that the subject partially matched the pattern, you can restart the
+match with additional subject data by means of the \eR escape sequence. For
+example:
+.sp
+    re> /^\ed?\ed(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\ed\ed$/
+  data> 23ja\eP\eD
+  Partial match: 23ja
+  data> n05\eR\eD
+   0: n05
+.sp
+For further information about partial matching, see the
+.\" HREF
+\fBpcrepartial\fP
+.\"
+documentation.
+.
+.
+.SH CALLOUTS
+.rs
+.sp
+If the pattern contains any callout requests, \fBpcretest\fP's callout function
+is called during matching. This works with both matching functions. By default,
+the called function displays the callout number, the start and current
+positions in the text at the callout time, and the next pattern item to be
+tested. For example:
+.sp
+  --->pqrabcdef
+    0    ^  ^     \ed
+.sp
+This output indicates that callout number 0 occurred for a match attempt
+starting at the fourth character of the subject string, when the pointer was at
+the seventh character of the data, and when the next pattern item was \ed. Just
+one circumflex is output if the start and current positions are the same.
+.P
+Callouts numbered 255 are assumed to be automatic callouts, inserted as a
+result of the \fB/C\fP pattern modifier. In this case, instead of showing the
+callout number, the offset in the pattern, preceded by a plus, is output. For
+example:
+.sp
+    re> /\ed?[A-E]\e*/C
+  data> E*
+  --->E*
+   +0 ^      \ed?
+   +3 ^      [A-E]
+   +8 ^^     \e*
+  +10 ^ ^
+   0: E*
+.sp
+If a pattern contains (*MARK) items, an additional line is output whenever
+a change of latest mark is passed to the callout function. For example:
+.sp
+    re> /a(*MARK:X)bc/C
+  data> abc
+  --->abc
+   +0 ^       a
+   +1 ^^      (*MARK:X)
+  +10 ^^      b
+  Latest Mark: X
+  +11 ^ ^     c
+  +12 ^  ^
+   0: abc
+.sp
+The mark changes between matching "a" and "b", but stays the same for the rest
+of the match, so nothing more is output. If, as a result of backtracking, the
+mark reverts to being unset, the text "" is output.
+.P
+The callout function in \fBpcretest\fP returns zero (carry on matching) by
+default, but you can use a \eC item in a data line (as described above) to
+change this and other parameters of the callout.
+.P
+Inserting callouts can be helpful when using \fBpcretest\fP to check
+complicated regular expressions. For further information about callouts, see
+the
+.\" HREF
+\fBpcrecallout\fP
+.\"
+documentation.
+.
+.
+.
+.SH "NON-PRINTING CHARACTERS"
+.rs
+.sp
+When \fBpcretest\fP is outputting text in the compiled version of a pattern,
+bytes other than 32-126 are always treated as non-printing characters are are
+therefore shown as hex escapes.
+.P
+When \fBpcretest\fP is outputting text that is a matched part of a subject
+string, it behaves in the same way, unless a different locale has been set for
+the pattern (using the \fB/L\fP modifier). In this case, the \fBisprint()\fP
+function to distinguish printing and non-printing characters.
+.
+.
+.
+.SH "SAVING AND RELOADING COMPILED PATTERNS"
+.rs
+.sp
+The facilities described in this section are not available when the POSIX
+interface to PCRE is being used, that is, when the \fB/P\fP pattern modifier is
+specified.
+.P
+When the POSIX interface is not in use, you can cause \fBpcretest\fP to write a
+compiled pattern to a file, by following the modifiers with > and a file name.
+For example:
+.sp
+  /pattern/im >/some/file
+.sp
+See the
+.\" HREF
+\fBpcreprecompile\fP
+.\"
+documentation for a discussion about saving and re-using compiled patterns.
+Note that if the pattern was successfully studied with JIT optimization, the
+JIT data cannot be saved.
+.P
+The data that is written is binary. The first eight bytes are the length of the
+compiled pattern data followed by the length of the optional study data, each
+written as four bytes in big-endian order (most significant byte first). If
+there is no study data (either the pattern was not studied, or studying did not
+return any data), the second length is zero. The lengths are followed by an
+exact copy of the compiled pattern. If there is additional study data, this
+(excluding any JIT data) follows immediately after the compiled pattern. After
+writing the file, \fBpcretest\fP expects to read a new pattern.
+.P
+A saved pattern can be reloaded into \fBpcretest\fP by specifying < and a file
+name instead of a pattern. The name of the file must not contain a < character,
+as otherwise \fBpcretest\fP will interpret the line as a pattern delimited by <
+characters.
+For example:
+.sp
+   re> " to prompt for regular expressions, and "data>" to prompt for data
+       lines.
+
+       When pcretest is built, a configuration  option  can  specify  that  it
+       should  be  linked  with the libreadline library. When this is done, if
+       the input is from a terminal, it is read using the readline() function.
+       This  provides line-editing and history facilities. The output from the
+       -help option states whether or not readline() will be used.
+
+       The program handles any number of sets of input on a single input file.
+       Each  set starts with a regular expression, and continues with any num-
+       ber of data lines to be matched against the pattern.
+
+       Each data line is matched separately and independently. If you want  to
+       do multi-line matches, you have to use the \n escape sequence (or \r or
+       \r\n, etc., depending on the newline setting) in a single line of input
+       to  encode  the  newline  sequences. There is no limit on the length of
+       data lines; the input buffer is automatically extended  if  it  is  too
+       small.
+
+       An  empty  line signals the end of the data lines, at which point a new
+       regular expression is read. The regular expressions are given  enclosed
+       in any non-alphanumeric delimiters other than backslash, for example:
+
+         /(a|bc)x+yz/
+
+       White  space before the initial delimiter is ignored. A regular expres-
+       sion may be continued over several input lines, in which case the  new-
+       line  characters  are included within it. It is possible to include the
+       delimiter within the pattern by escaping it, for example
+
+         /abc\/def/
+
+       If you do so, the escape and the delimiter form part  of  the  pattern,
+       but  since delimiters are always non-alphanumeric, this does not affect
+       its interpretation.  If the terminating delimiter is  immediately  fol-
+       lowed by a backslash, for example,
+
+         /abc/\
+
+       then  a  backslash  is added to the end of the pattern. This is done to
+       provide a way of testing the error condition that arises if  a  pattern
+       finishes with a backslash, because
+
+         /abc\/
+
+       is  interpreted as the first line of a pattern that starts with "abc/",
+       causing pcretest to read the next line as a continuation of the regular
+       expression.
+
+
+PATTERN MODIFIERS
+
+       A  pattern may be followed by any number of modifiers, which are mostly
+       single characters, though some of these can  be  qualified  by  further
+       characters.   Following Perl usage, these are referred to below as, for
+       example, "the /i modifier", even though the delimiter  of  the  pattern
+       need  not  always  be  a slash, and no slash is used when writing modi-
+       fiers. White space may appear between the final pattern  delimiter  and
+       the  first  modifier,  and between the modifiers themselves. For refer-
+       ence, here is a complete list of  modifiers.  They  fall  into  several
+       groups that are described in detail in the following sections.
+
+         /8              set UTF mode
+         /9              set PCRE_NEVER_UTF (locks out UTF mode)
+         /?              disable UTF validity check
+         /+              show remainder of subject after match
+         /=              show all captures (not just those that are set)
+
+         /A              set PCRE_ANCHORED
+         /B              show compiled code
+         /C              set PCRE_AUTO_CALLOUT
+         /D              same as /B plus /I
+         /E              set PCRE_DOLLAR_ENDONLY
+         /F              flip byte order in compiled pattern
+         /f              set PCRE_FIRSTLINE
+         /G              find all matches (shorten string)
+         /g              find all matches (use startoffset)
+         /I              show information about pattern
+         /i              set PCRE_CASELESS
+         /J              set PCRE_DUPNAMES
+         /K              show backtracking control names
+         /L              set locale
+         /M              show compiled memory size
+         /m              set PCRE_MULTILINE
+         /N              set PCRE_NO_AUTO_CAPTURE
+         /P              use the POSIX wrapper
+         /S              study the pattern after compilation
+         /s              set PCRE_DOTALL
+         /T              select character tables
+         /U              set PCRE_UNGREEDY
+         /W              set PCRE_UCP
+         /X              set PCRE_EXTRA
+         /x              set PCRE_EXTENDED
+         /Y              set PCRE_NO_START_OPTIMIZE
+         /Z              don't show lengths in /B output
+
+         /          set PCRE_NEWLINE_ANY
+         /      set PCRE_NEWLINE_ANYCRLF
+         /           set PCRE_NEWLINE_CR
+         /         set PCRE_NEWLINE_CRLF
+         /           set PCRE_NEWLINE_LF
+         /  set PCRE_BSR_ANYCRLF
+         /  set PCRE_BSR_UNICODE
+         /           set PCRE_JAVASCRIPT_COMPAT
+
+
+   Perl-compatible modifiers
+
+       The /i, /m, /s, and /x modifiers set the PCRE_CASELESS, PCRE_MULTILINE,
+       PCRE_DOTALL,   or    PCRE_EXTENDED    options,    respectively,    when
+       pcre[16|32]_compile()  is  called. These four modifier letters have the
+       same effect as they do in Perl. For example:
+
+         /caseless/i
+
+
+   Modifiers for other PCRE options
+
+       The following table shows additional modifiers for  setting  PCRE  com-
+       pile-time options that do not correspond to anything in Perl:
+
+         /8              PCRE_UTF8           ) when using the 8-bit
+         /?              PCRE_NO_UTF8_CHECK  )   library
+
+         /8              PCRE_UTF16          ) when using the 16-bit
+         /?              PCRE_NO_UTF16_CHECK )   library
+
+         /8              PCRE_UTF32          ) when using the 32-bit
+         /?              PCRE_NO_UTF32_CHECK )   library
+
+         /9              PCRE_NEVER_UTF
+         /A              PCRE_ANCHORED
+         /C              PCRE_AUTO_CALLOUT
+         /E              PCRE_DOLLAR_ENDONLY
+         /f              PCRE_FIRSTLINE
+         /J              PCRE_DUPNAMES
+         /N              PCRE_NO_AUTO_CAPTURE
+         /U              PCRE_UNGREEDY
+         /W              PCRE_UCP
+         /X              PCRE_EXTRA
+         /Y              PCRE_NO_START_OPTIMIZE
+         /          PCRE_NEWLINE_ANY
+         /      PCRE_NEWLINE_ANYCRLF
+         /           PCRE_NEWLINE_CR
+         /         PCRE_NEWLINE_CRLF
+         /           PCRE_NEWLINE_LF
+         /  PCRE_BSR_ANYCRLF
+         /  PCRE_BSR_UNICODE
+         /           PCRE_JAVASCRIPT_COMPAT
+
+       The  modifiers  that are enclosed in angle brackets are literal strings
+       as shown, including the angle brackets, but the letters within  can  be
+       in  either case.  This example sets multiline matching with CRLF as the
+       line ending sequence:
+
+         /^abc/m
+
+       As well as turning on  the  PCRE_UTF8/16/32  option,  the  /8  modifier
+       causes  all  non-printing  characters  in  output strings to be printed
+       using the \x{hh...} notation. Otherwise, those less than 0x100 are out-
+       put in hex without the curly brackets.
+
+       Full  details  of  the PCRE options are given in the pcreapi documenta-
+       tion.
+
+   Finding all matches in a string
+
+       Searching for all possible matches within each subject  string  can  be
+       requested  by  the  /g  or  /G modifier. After finding a match, PCRE is
+       called again to search the remainder of the subject string. The differ-
+       ence between /g and /G is that the former uses the startoffset argument
+       to pcre[16|32]_exec() to start searching at  a  new  point  within  the
+       entire  string  (which is in effect what Perl does), whereas the latter
+       passes over a shortened substring.  This  makes  a  difference  to  the
+       matching  process  if  the  pattern  begins with a lookbehind assertion
+       (including \b or \B).
+
+       If any call to pcre[16|32]_exec() in a /g or  /G  sequence  matches  an
+       empty  string, the next call is done with the PCRE_NOTEMPTY_ATSTART and
+       PCRE_ANCHORED flags set in order  to  search  for  another,  non-empty,
+       match  at  the same point. If this second match fails, the start offset
+       is advanced, and the normal match is retried.  This  imitates  the  way
+       Perl handles such cases when using the /g modifier or the split() func-
+       tion. Normally, the start offset is advanced by one character,  but  if
+       the  newline  convention  recognizes CRLF as a newline, and the current
+       character is CR followed by LF, an advance of two is used.
+
+   Other modifiers
+
+       There are yet more modifiers for controlling the way pcretest operates.
+
+       The /+ modifier requests that as well as outputting the substring  that
+       matched  the  entire  pattern,  pcretest  should in addition output the
+       remainder of the subject string. This is useful  for  tests  where  the
+       subject  contains multiple copies of the same substring. If the + modi-
+       fier appears twice, the same action is taken for  captured  substrings.
+       In  each case the remainder is output on the following line with a plus
+       character following the capture number. Note that  this  modifier  must
+       not  immediately follow the /S modifier because /S+ and /S++ have other
+       meanings.
+
+       The /= modifier requests that the  values  of  all  potential  captured
+       parentheses  be  output after a match. By default, only those up to the
+       highest one actually used in the match are output (corresponding to the
+       return code from pcre[16|32]_exec()). Values in the offsets vector cor-
+       responding to higher numbers should be set to -1, and these are  output
+       as  "".  This modifier gives a way of checking that this is hap-
+       pening.
+
+       The /B modifier is a debugging feature. It requests that pcretest  out-
+       put  a  representation of the compiled code after compilation. Normally
+       this information contains length and offset values; however, if  /Z  is
+       also  present,  this data is replaced by spaces. This is a special fea-
+       ture for use in the automatic test scripts; it ensures  that  the  same
+       output is generated for different internal link sizes.
+
+       The  /D modifier is a PCRE debugging feature, and is equivalent to /BI,
+       that is, both the /B and the /I modifiers.
+
+       The /F modifier causes pcretest to flip the byte order  of  the  2-byte
+       and 4-byte fields in the compiled pattern. This facility is for testing
+       the feature in PCRE that allows it to execute patterns that  were  com-
+       piled on a host with a different endianness. This feature is not avail-
+       able when the POSIX interface to PCRE is being used, that is, when  the
+       /P pattern modifier is specified. See also the section about saving and
+       reloading compiled patterns below.
+
+       The /I modifier requests that pcretest  output  information  about  the
+       compiled  pattern (whether it is anchored, has a fixed first character,
+       and so on). It does this by calling pcre[16|32]_fullinfo()  after  com-
+       piling  a  pattern.  If the pattern is studied, the results of that are
+       also output.
+
+       The /K modifier requests pcretest to show names from backtracking  con-
+       trol  verbs  that  are  returned  from  calls to pcre[16|32]_exec(). It
+       causes pcretest to create a pcre[16|32]_extra  block  if  one  has  not
+       already  been  created by a call to pcre[16|32]_study(), and to set the
+       PCRE_EXTRA_MARK flag and the mark field  within  it,  every  time  that
+       pcre[16|32]_exec()  is  called.  If  the  variable  that the mark field
+       points to is  non-NULL  for  a  match,  non-match,  or  partial  match,
+       pcretest  prints  the  string  to which it points. For a match, this is
+       shown on a line by itself, tagged with "MK:". For  a  non-match  it  is
+       added to the message.
+
+       The  /L modifier must be followed directly by the name of a locale, for
+       example,
+
+         /pattern/Lfr_FR
+
+       For this reason, it must be the last modifier. The given locale is set,
+       pcre[16|32]_maketables()  is  called to build a set of character tables
+       for the locale, and this is then passed to  pcre[16|32]_compile()  when
+       compiling  the regular expression. Without an /L (or /T) modifier, NULL
+       is passed as the tables pointer;  that  is,  /L  applies  only  to  the
+       expression on which it appears.
+
+       The  /M  modifier  causes the size in bytes of the memory block used to
+       hold the compiled pattern to be output. This does not include the  size
+       of  the  pcre[16|32] block; it is just the actual compiled data. If the
+       pattern is successfully studied with the PCRE_STUDY_JIT_COMPILE option,
+       the size of the JIT compiled code is also output.
+
+       The  /S  modifier  causes  pcre[16|32]_study()  to  be called after the
+       expression has been compiled, and the results used when the  expression
+       is matched. There are a number of qualifying characters that may follow
+       /S.  They may appear in any order.
+
+       If S is followed by an exclamation mark, pcre[16|32]_study() is  called
+       with  the PCRE_STUDY_EXTRA_NEEDED option, causing it always to return a
+       pcre_extra block, even when studying discovers no useful information.
+
+       If /S is followed by a second S character, it suppresses studying, even
+       if  it  was  requested  externally  by the -s command line option. This
+       makes it possible to specify that certain patterns are always  studied,
+       and others are never studied, independently of -s. This feature is used
+       in the test files in a few cases where the output is different when the
+       pattern is studied.
+
+       If  the  /S  modifier  is  followed  by  a  +  character,  the  call to
+       pcre[16|32]_study() is made with all the JIT study options,  requesting
+       just-in-time  optimization  support if it is available, for both normal
+       and partial matching. If you want to restrict the JIT compiling  modes,
+       you can follow /S+ with a digit in the range 1 to 7:
+
+         1  normal match only
+         2  soft partial match only
+         3  normal match and soft partial match
+         4  hard partial match only
+         6  soft and hard partial match
+         7  all three modes (default)
+
+       If /S++ is used instead of /S+ (with or without a following digit), the
+       text "(JIT)" is added to the first output line  after  a  match  or  no
+       match when JIT-compiled code was actually used.
+
+       Note  that  there  is  also  an independent /+ modifier; it must not be
+       given immediately after /S or /S+ because this will be misinterpreted.
+
+       If JIT studying is successful, the compiled JIT code will automatically
+       be  used  when pcre[16|32]_exec() is run, except when incompatible run-
+       time options are specified. For more details, see the pcrejit  documen-
+       tation.  See also the \J escape sequence below for a way of setting the
+       size of the JIT stack.
+
+       Finally, if /S is followed by a minus  character,  JIT  compilation  is
+       suppressed,  even if it was requested externally by the -s command line
+       option. This makes it possible to specify that JIT is never to be  used
+       for certain patterns.
+
+       The  /T  modifier  must be followed by a single digit. It causes a spe-
+       cific set of built-in character tables to be passed to pcre[16|32]_com-
+       pile().  It  is used in the standard PCRE tests to check behaviour with
+       different character tables. The digit specifies the tables as follows:
+
+         0   the default ASCII tables, as distributed in
+               pcre_chartables.c.dist
+         1   a set of tables defining ISO 8859 characters
+
+       In table 1, some characters whose codes are greater than 128 are  iden-
+       tified as letters, digits, spaces, etc.
+
+   Using the POSIX wrapper API
+
+       The  /P modifier causes pcretest to call PCRE via the POSIX wrapper API
+       rather than its native API. This supports only the 8-bit library.  When
+       /P  is set, the following modifiers set options for the regcomp() func-
+       tion:
+
+         /i    REG_ICASE
+         /m    REG_NEWLINE
+         /N    REG_NOSUB
+         /s    REG_DOTALL     )
+         /U    REG_UNGREEDY   ) These options are not part of
+         /W    REG_UCP        )   the POSIX standard
+         /8    REG_UTF8       )
+
+       The /+ modifier works as  described  above.  All  other  modifiers  are
+       ignored.
+
+
+DATA LINES
+
+       Before  each  data  line  is  passed to pcre[16|32]_exec(), leading and
+       trailing white space is removed, and it is then scanned for \  escapes.
+       Some  of  these are pretty esoteric features, intended for checking out
+       some of the more complicated features of PCRE. If you are just  testing
+       "ordinary"  regular  expressions, you probably don't need any of these.
+       The following escapes are recognized:
+
+         \a         alarm (BEL, \x07)
+         \b         backspace (\x08)
+         \e         escape (\x27)
+         \f         form feed (\x0c)
+         \n         newline (\x0a)
+         \qdd       set the PCRE_MATCH_LIMIT limit to dd
+                      (any number of digits)
+         \r         carriage return (\x0d)
+         \t         tab (\x09)
+         \v         vertical tab (\x0b)
+         \nnn       octal character (up to 3 octal digits); always
+                      a byte unless > 255 in UTF-8 or 16-bit or 32-bit mode
+         \xhh       hexadecimal byte (up to 2 hex digits)
+         \x{hh...}  hexadecimal character (any number of hex digits)
+         \A         pass the PCRE_ANCHORED option to pcre[16|32]_exec()
+                      or pcre[16|32]_dfa_exec()
+         \B         pass the PCRE_NOTBOL option to pcre[16|32]_exec()
+                      or pcre[16|32]_dfa_exec()
+         \Cdd       call pcre[16|32]_copy_substring() for substring dd
+                      after a successful match (number less than 32)
+         \Cname     call pcre[16|32]_copy_named_substring() for substring
+                      "name" after a successful match (name termin-
+                      ated by next non alphanumeric character)
+         \C+        show the current captured substrings at callout
+                      time
+         \C-        do not supply a callout function
+         \C!n       return 1 instead of 0 when callout number n is
+                      reached
+         \C!n!m     return 1 instead of 0 when callout number n is
+                      reached for the nth time
+         \C*n       pass the number n (may be negative) as callout
+                      data; this is used as the callout return value
+         \D         use the pcre[16|32]_dfa_exec() match function
+         \F         only shortest match for pcre[16|32]_dfa_exec()
+         \Gdd       call pcre[16|32]_get_substring() for substring dd
+                      after a successful match (number less than 32)
+         \Gname     call pcre[16|32]_get_named_substring() for substring
+                      "name" after a successful match (name termin-
+                      ated by next non-alphanumeric character)
+         \Jdd       set up a JIT stack of dd kilobytes maximum (any
+                      number of digits)
+         \L         call pcre[16|32]_get_substringlist() after a
+                      successful match
+         \M         discover the minimum MATCH_LIMIT and
+                      MATCH_LIMIT_RECURSION settings
+         \N         pass the PCRE_NOTEMPTY option to pcre[16|32]_exec()
+                      or pcre[16|32]_dfa_exec(); if used twice, pass the
+                      PCRE_NOTEMPTY_ATSTART option
+         \Odd       set the size of the output vector passed to
+                      pcre[16|32]_exec() to dd (any number of digits)
+         \P         pass the PCRE_PARTIAL_SOFT option to pcre[16|32]_exec()
+                      or pcre[16|32]_dfa_exec(); if used twice, pass the
+                      PCRE_PARTIAL_HARD option
+         \Qdd       set the PCRE_MATCH_LIMIT_RECURSION limit to dd
+                      (any number of digits)
+         \R         pass the PCRE_DFA_RESTART option to pcre[16|32]_dfa_exec()
+         \S         output details of memory get/free calls during matching
+         \Y            pass    the    PCRE_NO_START_OPTIMIZE     option     to
+       pcre[16|32]_exec()
+                      or pcre[16|32]_dfa_exec()
+         \Z         pass the PCRE_NOTEOL option to pcre[16|32]_exec()
+                      or pcre[16|32]_dfa_exec()
+         \?         pass the PCRE_NO_UTF[8|16|32]_CHECK option to
+                      pcre[16|32]_exec() or pcre[16|32]_dfa_exec()
+         \>dd       start the match at offset dd (optional "-"; then
+                      any number of digits); this sets the startoffset
+                      argument         for        pcre[16|32]_exec()        or
+       pcre[16|32]_dfa_exec()
+         \      pass the PCRE_NEWLINE_CR option to pcre[16|32]_exec()
+                      or pcre[16|32]_dfa_exec()
+         \      pass the PCRE_NEWLINE_LF option to pcre[16|32]_exec()
+                      or pcre[16|32]_dfa_exec()
+         \    pass the PCRE_NEWLINE_CRLF option to pcre[16|32]_exec()
+                      or pcre[16|32]_dfa_exec()
+         \ pass the PCRE_NEWLINE_ANYCRLF option to pcre[16|32]_exec()
+                      or pcre[16|32]_dfa_exec()
+         \     pass the PCRE_NEWLINE_ANY option to pcre[16|32]_exec()
+                      or pcre[16|32]_dfa_exec()
+
+       The use of \x{hh...} is not dependent on the use of the /8 modifier  on
+       the  pattern. It is recognized always. There may be any number of hexa-
+       decimal digits inside the braces; invalid  values  provoke  error  mes-
+       sages.
+
+       Note  that  \xhh  specifies one byte rather than one character in UTF-8
+       mode; this makes it possible to construct invalid UTF-8  sequences  for
+       testing  purposes.  On the other hand, \x{hh} is interpreted as a UTF-8
+       character in UTF-8 mode, generating more than one byte if the value  is
+       greater  than  127.   When testing the 8-bit library not in UTF-8 mode,
+       \x{hh} generates one byte for values less than 256, and causes an error
+       for greater values.
+
+       In UTF-16 mode, all 4-digit \x{hhhh} values are accepted. This makes it
+       possible to construct invalid UTF-16 sequences for testing purposes.
+
+       In UTF-32 mode, all 4- to 8-digit \x{...}  values  are  accepted.  This
+       makes  it  possible  to  construct invalid UTF-32 sequences for testing
+       purposes.
+
+       The escapes that specify line ending  sequences  are  literal  strings,
+       exactly as shown. No more than one newline setting should be present in
+       any data line.
+
+       A backslash followed by anything else just escapes the  anything  else.
+       If  the very last character is a backslash, it is ignored. This gives a
+       way of passing an empty line as data, since a real  empty  line  termi-
+       nates the data input.
+
+       The  \J escape provides a way of setting the maximum stack size that is
+       used by the just-in-time optimization code. It is ignored if JIT  opti-
+       mization  is  not being used. Providing a stack that is larger than the
+       default 32K is necessary only for very complicated patterns.
+
+       If \M is present, pcretest calls pcre[16|32]_exec() several times, with
+       different values in the match_limit and match_limit_recursion fields of
+       the pcre[16|32]_extra data structure, until it finds the  minimum  num-
+       bers for each parameter that allow pcre[16|32]_exec() to complete with-
+       out error. Because this is testing a specific  feature  of  the  normal
+       interpretive pcre[16|32]_exec() execution, the use of any JIT optimiza-
+       tion that might have been set up by the /S+ qualifier of -s+ option  is
+       disabled.
+
+       The  match_limit number is a measure of the amount of backtracking that
+       takes place, and checking it out can be instructive.  For  most  simple
+       matches,  the  number  is quite small, but for patterns with very large
+       numbers of matching possibilities, it can  become  large  very  quickly
+       with  increasing  length  of  subject string. The match_limit_recursion
+       number is a measure of how much stack (or, if  PCRE  is  compiled  with
+       NO_RECURSE,  how  much  heap)  memory  is  needed to complete the match
+       attempt.
+
+       When \O is used, the value specified may be higher or  lower  than  the
+       size set by the -O command line option (or defaulted to 45); \O applies
+       only to the call  of  pcre[16|32]_exec()  for  the  line  in  which  it
+       appears.
+
+       If  the /P modifier was present on the pattern, causing the POSIX wrap-
+       per API to be used, the only option-setting  sequences  that  have  any
+       effect  are  \B,  \N,  and  \Z,  causing  REG_NOTBOL, REG_NOTEMPTY, and
+       REG_NOTEOL, respectively, to be passed to regexec().
+
+
+THE ALTERNATIVE MATCHING FUNCTION
+
+       By  default,  pcretest  uses  the  standard  PCRE  matching   function,
+       pcre[16|32]_exec()  to  match  each  data  line.  PCRE also supports an
+       alternative matching function, pcre[16|32]_dfa_test(),  which  operates
+       in  a different way, and has some restrictions. The differences between
+       the two functions are described in the pcrematching documentation.
+
+       If a data line contains the \D escape sequence, or if the command  line
+       contains  the  -dfa  option, the alternative matching function is used.
+       This function finds all possible matches at a given point. If, however,
+       the  \F escape sequence is present in the data line, it stops after the
+       first match is found. This is always the shortest possible match.
+
+
+DEFAULT OUTPUT FROM PCRETEST
+
+       This section describes the output when the  normal  matching  function,
+       pcre[16|32]_exec(), is being used.
+
+       When a match succeeds, pcretest outputs the list of captured substrings
+       that pcre[16|32]_exec() returns, starting with number 0 for the  string
+       that  matched  the whole pattern. Otherwise, it outputs "No match" when
+       the return is PCRE_ERROR_NOMATCH, and "Partial match:" followed by  the
+       partially    matching   substring   when   pcre[16|32]_exec()   returns
+       PCRE_ERROR_PARTIAL. (Note that this is the entire  substring  that  was
+       inspected  during  the  partial match; it may include characters before
+       the actual match start if a lookbehind assertion, \K,  \b,  or  \B  was
+       involved.)  For  any  other  return, pcretest outputs the PCRE negative
+       error number and a short descriptive phrase. If the error is  a  failed
+       UTF  string check, the offset of the start of the failing character and
+       the reason code are also output, provided that the size of  the  output
+       vector  is  at least two. Here is an example of an interactive pcretest
+       run.
+
+         $ pcretest
+         PCRE version 8.13 2011-04-30
+
+           re> /^abc(\d+)/
+         data> abc123
+          0: abc123
+          1: 123
+         data> xyz
+         No match
+
+       Unset capturing substrings that are not followed by one that is set are
+       not  returned  by pcre[16|32]_exec(), and are not shown by pcretest. In
+       the following example, there are two capturing substrings, but when the
+       first  data  line is matched, the second, unset substring is not shown.
+       An "internal" unset substring is shown as "", as for the  second
+       data line.
+
+           re> /(a)|(b)/
+         data> a
+          0: a
+          1: a
+         data> b
+          0: b
+          1: 
+          2: b
+
+       If  the strings contain any non-printing characters, they are output as
+       \xhh escapes if the value is less than 256 and UTF  mode  is  not  set.
+       Otherwise they are output as \x{hh...} escapes. See below for the defi-
+       nition of non-printing characters. If the pattern has the /+  modifier,
+       the  output  for substring 0 is followed by the the rest of the subject
+       string, identified by "0+" like this:
+
+           re> /cat/+
+         data> cataract
+          0: cat
+          0+ aract
+
+       If the pattern has the /g or /G modifier,  the  results  of  successive
+       matching attempts are output in sequence, like this:
+
+           re> /\Bi(\w\w)/g
+         data> Mississippi
+          0: iss
+          1: ss
+          0: iss
+          1: ss
+          0: ipp
+          1: pp
+
+       "No  match" is output only if the first match attempt fails. Here is an
+       example of a failure message (the offset 4 that is specified by \>4  is
+       past the end of the subject string):
+
+           re> /xyz/
+         data> xyz\>4
+         Error -24 (bad offset value)
+
+       If  any  of the sequences \C, \G, or \L are present in a data line that
+       is successfully matched, the substrings extracted  by  the  convenience
+       functions are output with C, G, or L after the string number instead of
+       a colon. This is in addition to the normal full list. The string length
+       (that  is,  the return from the extraction function) is given in paren-
+       theses after each string for \C and \G.
+
+       Note that whereas patterns can be continued over several lines (a plain
+       ">" prompt is used for continuations), data lines may not. However new-
+       lines can be included in data by means of the \n escape (or  \r,  \r\n,
+       etc., depending on the newline sequence setting).
+
+
+OUTPUT FROM THE ALTERNATIVE MATCHING FUNCTION
+
+       When the alternative matching function, pcre[16|32]_dfa_exec(), is used
+       (by means of the \D escape sequence or the -dfa command  line  option),
+       the  output  consists  of  a  list of all the matches that start at the
+       first point in the subject where there is at least one match. For exam-
+       ple:
+
+           re> /(tang|tangerine|tan)/
+         data> yellow tangerine\D
+          0: tangerine
+          1: tang
+          2: tan
+
+       (Using  the  normal  matching function on this data finds only "tang".)
+       The longest matching string is always given first (and numbered  zero).
+       After a PCRE_ERROR_PARTIAL return, the output is "Partial match:", fol-
+       lowed by the partially matching  substring.  (Note  that  this  is  the
+       entire  substring  that  was inspected during the partial match; it may
+       include characters before the actual match start if a lookbehind asser-
+       tion, \K, \b, or \B was involved.)
+
+       If /g is present on the pattern, the search for further matches resumes
+       at the end of the longest match. For example:
+
+           re> /(tang|tangerine|tan)/g
+         data> yellow tangerine and tangy sultana\D
+          0: tangerine
+          1: tang
+          2: tan
+          0: tang
+          1: tan
+          0: tan
+
+       Since the matching function does not  support  substring  capture,  the
+       escape  sequences  that  are concerned with captured substrings are not
+       relevant.
+
+
+RESTARTING AFTER A PARTIAL MATCH
+
+       When the alternative matching function has given the PCRE_ERROR_PARTIAL
+       return,  indicating that the subject partially matched the pattern, you
+       can restart the match with additional subject data by means of  the  \R
+       escape sequence. For example:
+
+           re> /^\d?\d(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\d\d$/
+         data> 23ja\P\D
+         Partial match: 23ja
+         data> n05\R\D
+          0: n05
+
+       For  further  information  about  partial matching, see the pcrepartial
+       documentation.
+
+
+CALLOUTS
+
+       If the pattern contains any callout requests, pcretest's callout  func-
+       tion  is  called  during  matching. This works with both matching func-
+       tions. By default, the called function displays the callout number, the
+       start  and  current  positions in the text at the callout time, and the
+       next pattern item to be tested. For example:
+
+         --->pqrabcdef
+           0    ^  ^     \d
+
+       This output indicates that  callout  number  0  occurred  for  a  match
+       attempt  starting  at  the fourth character of the subject string, when
+       the pointer was at the seventh character of the data, and when the next
+       pattern  item  was  \d.  Just one circumflex is output if the start and
+       current positions are the same.
+
+       Callouts numbered 255 are assumed to be automatic callouts, inserted as
+       a  result  of the /C pattern modifier. In this case, instead of showing
+       the callout number, the offset in the pattern, preceded by a  plus,  is
+       output. For example:
+
+           re> /\d?[A-E]\*/C
+         data> E*
+         --->E*
+          +0 ^      \d?
+          +3 ^      [A-E]
+          +8 ^^     \*
+         +10 ^ ^
+          0: E*
+
+       If a pattern contains (*MARK) items, an additional line is output when-
+       ever a change of latest mark is passed to  the  callout  function.  For
+       example:
+
+           re> /a(*MARK:X)bc/C
+         data> abc
+         --->abc
+          +0 ^       a
+          +1 ^^      (*MARK:X)
+         +10 ^^      b
+         Latest Mark: X
+         +11 ^ ^     c
+         +12 ^  ^
+          0: abc
+
+       The  mark  changes between matching "a" and "b", but stays the same for
+       the rest of the match, so nothing more is output. If, as  a  result  of
+       backtracking,  the  mark  reverts to being unset, the text "" is
+       output.
+
+       The callout function in pcretest returns zero (carry  on  matching)  by
+       default,  but you can use a \C item in a data line (as described above)
+       to change this and other parameters of the callout.
+
+       Inserting callouts can be helpful when using pcretest to check  compli-
+       cated  regular expressions. For further information about callouts, see
+       the pcrecallout documentation.
+
+
+NON-PRINTING CHARACTERS
+
+       When pcretest is outputting text in the compiled version of a  pattern,
+       bytes  other  than 32-126 are always treated as non-printing characters
+       are are therefore shown as hex escapes.
+
+       When pcretest is outputting text that is a matched part  of  a  subject
+       string,  it behaves in the same way, unless a different locale has been
+       set for the  pattern  (using  the  /L  modifier).  In  this  case,  the
+       isprint() function to distinguish printing and non-printing characters.
+
+
+SAVING AND RELOADING COMPILED PATTERNS
+
+       The  facilities  described  in  this section are not available when the
+       POSIX interface to PCRE is being used, that is,  when  the  /P  pattern
+       modifier is specified.
+
+       When the POSIX interface is not in use, you can cause pcretest to write
+       a compiled pattern to a file, by following the modifiers with >  and  a
+       file name.  For example:
+
+         /pattern/im >/some/file
+
+       See  the pcreprecompile documentation for a discussion about saving and
+       re-using compiled patterns.  Note that if the pattern was  successfully
+       studied with JIT optimization, the JIT data cannot be saved.
+
+       The  data  that  is  written  is  binary. The first eight bytes are the
+       length of the compiled pattern data  followed  by  the  length  of  the
+       optional  study  data,  each  written as four bytes in big-endian order
+       (most significant byte first). If there is no study  data  (either  the
+       pattern was not studied, or studying did not return any data), the sec-
+       ond length is zero. The lengths are followed by an exact  copy  of  the
+       compiled  pattern.  If  there is additional study data, this (excluding
+       any JIT data) follows immediately after  the  compiled  pattern.  After
+       writing the file, pcretest expects to read a new pattern.
+
+       A  saved  pattern  can  be reloaded into pcretest by specifying < and a
+       file name instead of a pattern. The name of the file must not contain a
+       < character, as otherwise pcretest will interpret the line as a pattern
+       delimited by < characters.  For example:
+
+          re> 
+.SS "Validity of UTF-8 strings"
+.rs
+.sp
+When you set the PCRE_UTF8 flag, the byte strings passed as patterns and
+subjects are (by default) checked for validity on entry to the relevant
+functions. The entire string is checked before any other processing takes
+place. From release 7.3 of PCRE, the check is according the rules of RFC 3629,
+which are themselves derived from the Unicode specification. Earlier releases
+of PCRE followed the rules of RFC 2279, which allows the full range of 31-bit
+values (0 to 0x7FFFFFFF). The current check allows only values in the range U+0
+to U+10FFFF, excluding the surrogate area. (From release 8.33 the so-called
+"non-character" code points are no longer excluded because Unicode corrigendum
+#9 makes it clear that they should not be.)
+.P
+Characters in the "Surrogate Area" of Unicode are reserved for use by UTF-16,
+where they are used in pairs to encode codepoints with values greater than
+0xFFFF. The code points that are encoded by UTF-16 pairs are available
+independently in the UTF-8 and UTF-32 encodings. (In other words, the whole
+surrogate thing is a fudge for UTF-16 which unfortunately messes up UTF-8 and
+UTF-32.)
+.P
+If an invalid UTF-8 string is passed to PCRE, an error return is given. At
+compile time, the only additional information is the offset to the first byte
+of the failing character. The run-time functions \fBpcre_exec()\fP and
+\fBpcre_dfa_exec()\fP also pass back this information, as well as a more
+detailed reason code if the caller has provided memory in which to do this.
+.P
+In some situations, you may already know that your strings are valid, and
+therefore want to skip these checks in order to improve performance, for
+example in the case of a long subject string that is being scanned repeatedly.
+If you set the PCRE_NO_UTF8_CHECK flag at compile time or at run time, PCRE
+assumes that the pattern or subject it is given (respectively) contains only
+valid UTF-8 codes. In this case, it does not diagnose an invalid UTF-8 string.
+.P
+Note that passing PCRE_NO_UTF8_CHECK to \fBpcre_compile()\fP just disables the
+check for the pattern; it does not also apply to subject strings. If you want
+to disable the check for a subject string you must pass this option to
+\fBpcre_exec()\fP or \fBpcre_dfa_exec()\fP.
+.P
+If you pass an invalid UTF-8 string when PCRE_NO_UTF8_CHECK is set, the result
+is undefined and your program may crash.
+.
+.
+.\" HTML 
+.SS "Validity of UTF-16 strings"
+.rs
+.sp
+When you set the PCRE_UTF16 flag, the strings of 16-bit data units that are
+passed as patterns and subjects are (by default) checked for validity on entry
+to the relevant functions. Values other than those in the surrogate range
+U+D800 to U+DFFF are independent code points. Values in the surrogate range
+must be used in pairs in the correct manner.
+.P
+If an invalid UTF-16 string is passed to PCRE, an error return is given. At
+compile time, the only additional information is the offset to the first data
+unit of the failing character. The run-time functions \fBpcre16_exec()\fP and
+\fBpcre16_dfa_exec()\fP also pass back this information, as well as a more
+detailed reason code if the caller has provided memory in which to do this.
+.P
+In some situations, you may already know that your strings are valid, and
+therefore want to skip these checks in order to improve performance. If you set
+the PCRE_NO_UTF16_CHECK flag at compile time or at run time, PCRE assumes that
+the pattern or subject it is given (respectively) contains only valid UTF-16
+sequences. In this case, it does not diagnose an invalid UTF-16 string.
+However, if an invalid string is passed, the result is undefined.
+.
+.
+.\" HTML 
+.SS "Validity of UTF-32 strings"
+.rs
+.sp
+When you set the PCRE_UTF32 flag, the strings of 32-bit data units that are
+passed as patterns and subjects are (by default) checked for validity on entry
+to the relevant functions.  This check allows only values in the range U+0
+to U+10FFFF, excluding the surrogate area U+D800 to U+DFFF.
+.P
+If an invalid UTF-32 string is passed to PCRE, an error return is given. At
+compile time, the only additional information is the offset to the first data
+unit of the failing character. The run-time functions \fBpcre32_exec()\fP and
+\fBpcre32_dfa_exec()\fP also pass back this information, as well as a more
+detailed reason code if the caller has provided memory in which to do this.
+.P
+In some situations, you may already know that your strings are valid, and
+therefore want to skip these checks in order to improve performance. If you set
+the PCRE_NO_UTF32_CHECK flag at compile time or at run time, PCRE assumes that
+the pattern or subject it is given (respectively) contains only valid UTF-32
+sequences. In this case, it does not diagnose an invalid UTF-32 string.
+However, if an invalid string is passed, the result is undefined.
+.
+.
+.SS "General comments about UTF modes"
+.rs
+.sp
+1. Codepoints less than 256 can be specified in patterns by either braced or
+unbraced hexadecimal escape sequences (for example, \ex{b3} or \exb3). Larger
+values have to use braced sequences.
+.P
+2. Octal numbers up to \e777 are recognized, and in UTF-8 mode they match
+two-byte characters for values greater than \e177.
+.P
+3. Repeat quantifiers apply to complete UTF characters, not to individual
+data units, for example: \ex{100}{3}.
+.P
+4. The dot metacharacter matches one UTF character instead of a single data
+unit.
+.P
+5. The escape sequence \eC can be used to match a single byte in UTF-8 mode, or
+a single 16-bit data unit in UTF-16 mode, or a single 32-bit data unit in
+UTF-32 mode, but its use can lead to some strange effects because it breaks up
+multi-unit characters (see the description of \eC in the
+.\" HREF
+\fBpcrepattern\fP
+.\"
+documentation). The use of \eC is not supported in the alternative matching
+function \fBpcre[16|32]_dfa_exec()\fP, nor is it supported in UTF mode by the
+JIT optimization of \fBpcre[16|32]_exec()\fP. If JIT optimization is requested
+for a UTF pattern that contains \eC, it will not succeed, and so the matching
+will be carried out by the normal interpretive function.
+.P
+6. The character escapes \eb, \eB, \ed, \eD, \es, \eS, \ew, and \eW correctly
+test characters of any code value, but, by default, the characters that PCRE
+recognizes as digits, spaces, or word characters remain the same set as in
+non-UTF mode, all with values less than 256. This remains true even when PCRE
+is built to include Unicode property support, because to do otherwise would
+slow down PCRE in many common cases. Note in particular that this applies to
+\eb and \eB, because they are defined in terms of \ew and \eW. If you really
+want to test for a wider sense of, say, "digit", you can use explicit Unicode
+property tests such as \ep{Nd}. Alternatively, if you set the PCRE_UCP option,
+the way that the character escapes work is changed so that Unicode properties
+are used to determine which characters match. There are more details in the
+section on
+.\" HTML 
+.\" 
+generic character types
+.\"
+in the
+.\" HREF
+\fBpcrepattern\fP
+.\"
+documentation.
+.P
+7. Similarly, characters that match the POSIX named character classes are all
+low-valued characters, unless the PCRE_UCP option is set.
+.P
+8. However, the horizontal and vertical white space matching escapes (\eh, \eH,
+\ev, and \eV) do match all the appropriate Unicode characters, whether or not
+PCRE_UCP is set.
+.P
+9. Case-insensitive matching applies only to characters whose values are less
+than 128, unless PCRE is built with Unicode property support. A few Unicode
+characters such as Greek sigma have more than two codepoints that are
+case-equivalent. Up to and including PCRE release 8.31, only one-to-one case
+mappings were supported, but later releases (with Unicode property support) do
+treat as case-equivalent all versions of characters such as Greek sigma.
+.
+.
+.SH AUTHOR
+.rs
+.sp
+.nf
+Philip Hazel
+University Computing Service
+Cambridge CB2 3QH, England.
+.fi
+.
+.
+.SH REVISION
+.rs
+.sp
+.nf
+Last updated: 27 February 2013
+Copyright (c) 1997-2013 University of Cambridge.
+.fi
diff --git a/pcre/doc/perltest.txt b/pcre/doc/perltest.txt
new file mode 100644
index 00000000000..bb1a52a48e1
--- /dev/null
+++ b/pcre/doc/perltest.txt
@@ -0,0 +1,42 @@
+The perltest program
+--------------------
+
+The perltest.pl script tests Perl's regular expressions; it has the same
+specification as pcretest, and so can be given identical input, except that
+input patterns can be followed only by Perl's lower case modifiers and certain
+other pcretest modifiers that are either handled or ignored:
+
+  /+   recognized and handled by perltest
+  /++  the second + is ignored
+  /8   recognized and handled by perltest
+  /J   ignored
+  /K   ignored
+  /W   ignored
+  /S   ignored
+  /SS  ignored
+  /Y   ignored
+
+The pcretest \Y escape in data lines is removed before matching. The data lines
+are processed as Perl double-quoted strings, so if they contain " $ or @
+characters, these have to be escaped. For this reason, all such characters in
+the Perl-compatible testinput1 file are escaped so that they can be used for
+perltest as well as for pcretest. The special upper case pattern modifiers such
+as /A that pcretest recognizes, and its special data line escapes, are not used
+in the Perl-compatible test file. The output should be identical, apart from
+the initial identifying banner.
+
+The perltest.pl script can also test UTF-8 features. It recognizes the special
+modifier /8 that pcretest uses to invoke UTF-8 functionality. The testinput4
+and testinput6 files can be fed to perltest to run compatible UTF-8 tests.
+However, it is necessary to add "use utf8; require Encode" to the script to
+make this work correctly. I have not managed to find a way to handle this
+automatically.
+
+The other testinput files are not suitable for feeding to perltest.pl, since
+they make use of the special upper case modifiers and escapes that pcretest
+uses to test certain features of PCRE. Some of these files also contain
+malformed regular expressions, in order to check that PCRE diagnoses them
+correctly.
+
+Philip Hazel
+January 2012
diff --git a/pcre/pcre.h b/pcre/pcre.h
new file mode 100644
index 00000000000..0e479667672
--- /dev/null
+++ b/pcre/pcre.h
@@ -0,0 +1,663 @@
+/*************************************************
+*       Perl-Compatible Regular Expressions      *
+*************************************************/
+
+/* This is the public header file for the PCRE library, to be #included by
+applications that call the PCRE functions.
+
+           Copyright (c) 1997-2013 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+#ifndef _PCRE_H
+#define _PCRE_H
+
+/* The current PCRE version information. */
+
+#define PCRE_MAJOR          8
+#define PCRE_MINOR          33
+#define PCRE_PRERELEASE     
+#define PCRE_DATE           2013-05-28
+
+/* When an application links to a PCRE DLL in Windows, the symbols that are
+imported have to be identified as such. When building PCRE, the appropriate
+export setting is defined in pcre_internal.h, which includes this file. So we
+don't change existing definitions of PCRE_EXP_DECL and PCRECPP_EXP_DECL. */
+
+#if defined(_WIN32) && !defined(PCRE_STATIC)
+#  ifndef PCRE_EXP_DECL
+#    define PCRE_EXP_DECL  extern __declspec(dllimport)
+#  endif
+#  ifdef __cplusplus
+#    ifndef PCRECPP_EXP_DECL
+#      define PCRECPP_EXP_DECL  extern __declspec(dllimport)
+#    endif
+#    ifndef PCRECPP_EXP_DEFN
+#      define PCRECPP_EXP_DEFN  __declspec(dllimport)
+#    endif
+#  endif
+#endif
+
+/* By default, we use the standard "extern" declarations. */
+
+#ifndef PCRE_EXP_DECL
+#  ifdef __cplusplus
+#    define PCRE_EXP_DECL  extern "C"
+#  else
+#    define PCRE_EXP_DECL  extern
+#  endif
+#endif
+
+#ifdef __cplusplus
+#  ifndef PCRECPP_EXP_DECL
+#    define PCRECPP_EXP_DECL  extern
+#  endif
+#  ifndef PCRECPP_EXP_DEFN
+#    define PCRECPP_EXP_DEFN
+#  endif
+#endif
+
+/* Have to include stdlib.h in order to ensure that size_t is defined;
+it is needed here for malloc. */
+
+#include 
+
+/* Allow for C++ users */
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+/* Public options. Some are compile-time only, some are run-time only, and some
+are both. Most of the compile-time options are saved with the compiled regex so
+that they can be inspected during studying (and therefore JIT compiling). Note
+that pcre_study() has its own set of options. Originally, all the options
+defined here used distinct bits. However, almost all the bits in a 32-bit word
+are now used, so in order to conserve them, option bits that were previously
+only recognized at matching time (i.e. by pcre_exec() or pcre_dfa_exec()) may
+also be used for compile-time options that affect only compiling and are not
+relevant for studying or JIT compiling.
+
+Some options for pcre_compile() change its behaviour but do not affect the
+behaviour of the execution functions. Other options are passed through to the
+execution functions and affect their behaviour, with or without affecting the
+behaviour of pcre_compile().
+
+Options that can be passed to pcre_compile() are tagged Cx below, with these
+variants:
+
+C1   Affects compile only
+C2   Does not affect compile; affects exec, dfa_exec
+C3   Affects compile, exec, dfa_exec
+C4   Affects compile, exec, dfa_exec, study
+C5   Affects compile, exec, study
+
+Options that can be set for pcre_exec() and/or pcre_dfa_exec() are flagged with
+E and D, respectively. They take precedence over C3, C4, and C5 settings passed
+from pcre_compile(). Those that are compatible with JIT execution are flagged
+with J. */
+
+#define PCRE_CASELESS           0x00000001  /* C1       */
+#define PCRE_MULTILINE          0x00000002  /* C1       */
+#define PCRE_DOTALL             0x00000004  /* C1       */
+#define PCRE_EXTENDED           0x00000008  /* C1       */
+#define PCRE_ANCHORED           0x00000010  /* C4 E D   */
+#define PCRE_DOLLAR_ENDONLY     0x00000020  /* C2       */
+#define PCRE_EXTRA              0x00000040  /* C1       */
+#define PCRE_NOTBOL             0x00000080  /*    E D J */
+#define PCRE_NOTEOL             0x00000100  /*    E D J */
+#define PCRE_UNGREEDY           0x00000200  /* C1       */
+#define PCRE_NOTEMPTY           0x00000400  /*    E D J */
+#define PCRE_UTF8               0x00000800  /* C4        )          */
+#define PCRE_UTF16              0x00000800  /* C4        ) Synonyms */
+#define PCRE_UTF32              0x00000800  /* C4        )          */
+#define PCRE_NO_AUTO_CAPTURE    0x00001000  /* C1       */
+#define PCRE_NO_UTF8_CHECK      0x00002000  /* C1 E D J  )          */
+#define PCRE_NO_UTF16_CHECK     0x00002000  /* C1 E D J  ) Synonyms */
+#define PCRE_NO_UTF32_CHECK     0x00002000  /* C1 E D J  )          */
+#define PCRE_AUTO_CALLOUT       0x00004000  /* C1       */
+#define PCRE_PARTIAL_SOFT       0x00008000  /*    E D J  ) Synonyms */
+#define PCRE_PARTIAL            0x00008000  /*    E D J  )          */
+
+/* This pair use the same bit. */
+#define PCRE_NEVER_UTF          0x00010000  /* C1        ) Overlaid */
+#define PCRE_DFA_SHORTEST       0x00010000  /*      D    ) Overlaid */
+
+#define PCRE_DFA_RESTART        0x00020000  /*      D   */
+#define PCRE_FIRSTLINE          0x00040000  /* C3       */
+#define PCRE_DUPNAMES           0x00080000  /* C1       */
+#define PCRE_NEWLINE_CR         0x00100000  /* C3 E D   */
+#define PCRE_NEWLINE_LF         0x00200000  /* C3 E D   */
+#define PCRE_NEWLINE_CRLF       0x00300000  /* C3 E D   */
+#define PCRE_NEWLINE_ANY        0x00400000  /* C3 E D   */
+#define PCRE_NEWLINE_ANYCRLF    0x00500000  /* C3 E D   */
+#define PCRE_BSR_ANYCRLF        0x00800000  /* C3 E D   */
+#define PCRE_BSR_UNICODE        0x01000000  /* C3 E D   */
+#define PCRE_JAVASCRIPT_COMPAT  0x02000000  /* C5       */
+#define PCRE_NO_START_OPTIMIZE  0x04000000  /* C2 E D    ) Synonyms */
+#define PCRE_NO_START_OPTIMISE  0x04000000  /* C2 E D    )          */
+#define PCRE_PARTIAL_HARD       0x08000000  /*    E D J */
+#define PCRE_NOTEMPTY_ATSTART   0x10000000  /*    E D J */
+#define PCRE_UCP                0x20000000  /* C3       */
+
+/* Exec-time and get/set-time error codes */
+
+#define PCRE_ERROR_NOMATCH          (-1)
+#define PCRE_ERROR_NULL             (-2)
+#define PCRE_ERROR_BADOPTION        (-3)
+#define PCRE_ERROR_BADMAGIC         (-4)
+#define PCRE_ERROR_UNKNOWN_OPCODE   (-5)
+#define PCRE_ERROR_UNKNOWN_NODE     (-5)  /* For backward compatibility */
+#define PCRE_ERROR_NOMEMORY         (-6)
+#define PCRE_ERROR_NOSUBSTRING      (-7)
+#define PCRE_ERROR_MATCHLIMIT       (-8)
+#define PCRE_ERROR_CALLOUT          (-9)  /* Never used by PCRE itself */
+#define PCRE_ERROR_BADUTF8         (-10)  /* Same for 8/16/32 */
+#define PCRE_ERROR_BADUTF16        (-10)  /* Same for 8/16/32 */
+#define PCRE_ERROR_BADUTF32        (-10)  /* Same for 8/16/32 */
+#define PCRE_ERROR_BADUTF8_OFFSET  (-11)  /* Same for 8/16 */
+#define PCRE_ERROR_BADUTF16_OFFSET (-11)  /* Same for 8/16 */
+#define PCRE_ERROR_PARTIAL         (-12)
+#define PCRE_ERROR_BADPARTIAL      (-13)
+#define PCRE_ERROR_INTERNAL        (-14)
+#define PCRE_ERROR_BADCOUNT        (-15)
+#define PCRE_ERROR_DFA_UITEM       (-16)
+#define PCRE_ERROR_DFA_UCOND       (-17)
+#define PCRE_ERROR_DFA_UMLIMIT     (-18)
+#define PCRE_ERROR_DFA_WSSIZE      (-19)
+#define PCRE_ERROR_DFA_RECURSE     (-20)
+#define PCRE_ERROR_RECURSIONLIMIT  (-21)
+#define PCRE_ERROR_NULLWSLIMIT     (-22)  /* No longer actually used */
+#define PCRE_ERROR_BADNEWLINE      (-23)
+#define PCRE_ERROR_BADOFFSET       (-24)
+#define PCRE_ERROR_SHORTUTF8       (-25)
+#define PCRE_ERROR_SHORTUTF16      (-25)  /* Same for 8/16 */
+#define PCRE_ERROR_RECURSELOOP     (-26)
+#define PCRE_ERROR_JIT_STACKLIMIT  (-27)
+#define PCRE_ERROR_BADMODE         (-28)
+#define PCRE_ERROR_BADENDIANNESS   (-29)
+#define PCRE_ERROR_DFA_BADRESTART  (-30)
+#define PCRE_ERROR_JIT_BADOPTION   (-31)
+#define PCRE_ERROR_BADLENGTH       (-32)
+#define PCRE_ERROR_UNSET           (-33)
+
+/* Specific error codes for UTF-8 validity checks */
+
+#define PCRE_UTF8_ERR0               0
+#define PCRE_UTF8_ERR1               1
+#define PCRE_UTF8_ERR2               2
+#define PCRE_UTF8_ERR3               3
+#define PCRE_UTF8_ERR4               4
+#define PCRE_UTF8_ERR5               5
+#define PCRE_UTF8_ERR6               6
+#define PCRE_UTF8_ERR7               7
+#define PCRE_UTF8_ERR8               8
+#define PCRE_UTF8_ERR9               9
+#define PCRE_UTF8_ERR10             10
+#define PCRE_UTF8_ERR11             11
+#define PCRE_UTF8_ERR12             12
+#define PCRE_UTF8_ERR13             13
+#define PCRE_UTF8_ERR14             14
+#define PCRE_UTF8_ERR15             15
+#define PCRE_UTF8_ERR16             16
+#define PCRE_UTF8_ERR17             17
+#define PCRE_UTF8_ERR18             18
+#define PCRE_UTF8_ERR19             19
+#define PCRE_UTF8_ERR20             20
+#define PCRE_UTF8_ERR21             21
+#define PCRE_UTF8_ERR22             22  /* Unused (was non-character) */
+
+/* Specific error codes for UTF-16 validity checks */
+
+#define PCRE_UTF16_ERR0              0
+#define PCRE_UTF16_ERR1              1
+#define PCRE_UTF16_ERR2              2
+#define PCRE_UTF16_ERR3              3
+#define PCRE_UTF16_ERR4              4  /* Unused (was non-character) */
+
+/* Specific error codes for UTF-32 validity checks */
+
+#define PCRE_UTF32_ERR0              0
+#define PCRE_UTF32_ERR1              1
+#define PCRE_UTF32_ERR2              2  /* Unused (was non-character) */
+#define PCRE_UTF32_ERR3              3
+
+/* Request types for pcre_fullinfo() */
+
+#define PCRE_INFO_OPTIONS            0
+#define PCRE_INFO_SIZE               1
+#define PCRE_INFO_CAPTURECOUNT       2
+#define PCRE_INFO_BACKREFMAX         3
+#define PCRE_INFO_FIRSTBYTE          4
+#define PCRE_INFO_FIRSTCHAR          4  /* For backwards compatibility */
+#define PCRE_INFO_FIRSTTABLE         5
+#define PCRE_INFO_LASTLITERAL        6
+#define PCRE_INFO_NAMEENTRYSIZE      7
+#define PCRE_INFO_NAMECOUNT          8
+#define PCRE_INFO_NAMETABLE          9
+#define PCRE_INFO_STUDYSIZE         10
+#define PCRE_INFO_DEFAULT_TABLES    11
+#define PCRE_INFO_OKPARTIAL         12
+#define PCRE_INFO_JCHANGED          13
+#define PCRE_INFO_HASCRORLF         14
+#define PCRE_INFO_MINLENGTH         15
+#define PCRE_INFO_JIT               16
+#define PCRE_INFO_JITSIZE           17
+#define PCRE_INFO_MAXLOOKBEHIND     18
+#define PCRE_INFO_FIRSTCHARACTER    19
+#define PCRE_INFO_FIRSTCHARACTERFLAGS 20
+#define PCRE_INFO_REQUIREDCHAR      21
+#define PCRE_INFO_REQUIREDCHARFLAGS 22
+#define PCRE_INFO_MATCHLIMIT        23
+#define PCRE_INFO_RECURSIONLIMIT    24
+
+/* Request types for pcre_config(). Do not re-arrange, in order to remain
+compatible. */
+
+#define PCRE_CONFIG_UTF8                    0
+#define PCRE_CONFIG_NEWLINE                 1
+#define PCRE_CONFIG_LINK_SIZE               2
+#define PCRE_CONFIG_POSIX_MALLOC_THRESHOLD  3
+#define PCRE_CONFIG_MATCH_LIMIT             4
+#define PCRE_CONFIG_STACKRECURSE            5
+#define PCRE_CONFIG_UNICODE_PROPERTIES      6
+#define PCRE_CONFIG_MATCH_LIMIT_RECURSION   7
+#define PCRE_CONFIG_BSR                     8
+#define PCRE_CONFIG_JIT                     9
+#define PCRE_CONFIG_UTF16                  10
+#define PCRE_CONFIG_JITTARGET              11
+#define PCRE_CONFIG_UTF32                  12
+
+/* Request types for pcre_study(). Do not re-arrange, in order to remain
+compatible. */
+
+#define PCRE_STUDY_JIT_COMPILE                0x0001
+#define PCRE_STUDY_JIT_PARTIAL_SOFT_COMPILE   0x0002
+#define PCRE_STUDY_JIT_PARTIAL_HARD_COMPILE   0x0004
+#define PCRE_STUDY_EXTRA_NEEDED               0x0008
+
+/* Bit flags for the pcre[16|32]_extra structure. Do not re-arrange or redefine
+these bits, just add new ones on the end, in order to remain compatible. */
+
+#define PCRE_EXTRA_STUDY_DATA             0x0001
+#define PCRE_EXTRA_MATCH_LIMIT            0x0002
+#define PCRE_EXTRA_CALLOUT_DATA           0x0004
+#define PCRE_EXTRA_TABLES                 0x0008
+#define PCRE_EXTRA_MATCH_LIMIT_RECURSION  0x0010
+#define PCRE_EXTRA_MARK                   0x0020
+#define PCRE_EXTRA_EXECUTABLE_JIT         0x0040
+
+/* Types */
+
+struct real_pcre;                 /* declaration; the definition is private  */
+typedef struct real_pcre pcre;
+
+struct real_pcre16;               /* declaration; the definition is private  */
+typedef struct real_pcre16 pcre16;
+
+struct real_pcre32;               /* declaration; the definition is private  */
+typedef struct real_pcre32 pcre32;
+
+struct real_pcre_jit_stack;       /* declaration; the definition is private  */
+typedef struct real_pcre_jit_stack pcre_jit_stack;
+
+struct real_pcre16_jit_stack;     /* declaration; the definition is private  */
+typedef struct real_pcre16_jit_stack pcre16_jit_stack;
+
+struct real_pcre32_jit_stack;     /* declaration; the definition is private  */
+typedef struct real_pcre32_jit_stack pcre32_jit_stack;
+
+/* If PCRE is compiled with 16 bit character support, PCRE_UCHAR16 must contain
+a 16 bit wide signed data type. Otherwise it can be a dummy data type since
+pcre16 functions are not implemented. There is a check for this in pcre_internal.h. */
+#ifndef PCRE_UCHAR16
+#define PCRE_UCHAR16 unsigned short
+#endif
+
+#ifndef PCRE_SPTR16
+#define PCRE_SPTR16 const PCRE_UCHAR16 *
+#endif
+
+/* If PCRE is compiled with 32 bit character support, PCRE_UCHAR32 must contain
+a 32 bit wide signed data type. Otherwise it can be a dummy data type since
+pcre32 functions are not implemented. There is a check for this in pcre_internal.h. */
+#ifndef PCRE_UCHAR32
+#define PCRE_UCHAR32 unsigned int
+#endif
+
+#ifndef PCRE_SPTR32
+#define PCRE_SPTR32 const PCRE_UCHAR32 *
+#endif
+
+/* When PCRE is compiled as a C++ library, the subject pointer type can be
+replaced with a custom type. For conventional use, the public interface is a
+const char *. */
+
+#ifndef PCRE_SPTR
+#define PCRE_SPTR const char *
+#endif
+
+/* The structure for passing additional data to pcre_exec(). This is defined in
+such as way as to be extensible. Always add new fields at the end, in order to
+remain compatible. */
+
+typedef struct pcre_extra {
+  unsigned long int flags;        /* Bits for which fields are set */
+  void *study_data;               /* Opaque data from pcre_study() */
+  unsigned long int match_limit;  /* Maximum number of calls to match() */
+  void *callout_data;             /* Data passed back in callouts */
+  const unsigned char *tables;    /* Pointer to character tables */
+  unsigned long int match_limit_recursion; /* Max recursive calls to match() */
+  unsigned char **mark;           /* For passing back a mark pointer */
+  void *executable_jit;           /* Contains a pointer to a compiled jit code */
+} pcre_extra;
+
+/* Same structure as above, but with 16 bit char pointers. */
+
+typedef struct pcre16_extra {
+  unsigned long int flags;        /* Bits for which fields are set */
+  void *study_data;               /* Opaque data from pcre_study() */
+  unsigned long int match_limit;  /* Maximum number of calls to match() */
+  void *callout_data;             /* Data passed back in callouts */
+  const unsigned char *tables;    /* Pointer to character tables */
+  unsigned long int match_limit_recursion; /* Max recursive calls to match() */
+  PCRE_UCHAR16 **mark;            /* For passing back a mark pointer */
+  void *executable_jit;           /* Contains a pointer to a compiled jit code */
+} pcre16_extra;
+
+/* Same structure as above, but with 32 bit char pointers. */
+
+typedef struct pcre32_extra {
+  unsigned long int flags;        /* Bits for which fields are set */
+  void *study_data;               /* Opaque data from pcre_study() */
+  unsigned long int match_limit;  /* Maximum number of calls to match() */
+  void *callout_data;             /* Data passed back in callouts */
+  const unsigned char *tables;    /* Pointer to character tables */
+  unsigned long int match_limit_recursion; /* Max recursive calls to match() */
+  PCRE_UCHAR32 **mark;            /* For passing back a mark pointer */
+  void *executable_jit;           /* Contains a pointer to a compiled jit code */
+} pcre32_extra;
+
+/* The structure for passing out data via the pcre_callout_function. We use a
+structure so that new fields can be added on the end in future versions,
+without changing the API of the function, thereby allowing old clients to work
+without modification. */
+
+typedef struct pcre_callout_block {
+  int          version;           /* Identifies version of block */
+  /* ------------------------ Version 0 ------------------------------- */
+  int          callout_number;    /* Number compiled into pattern */
+  int         *offset_vector;     /* The offset vector */
+  PCRE_SPTR    subject;           /* The subject being matched */
+  int          subject_length;    /* The length of the subject */
+  int          start_match;       /* Offset to start of this match attempt */
+  int          current_position;  /* Where we currently are in the subject */
+  int          capture_top;       /* Max current capture */
+  int          capture_last;      /* Most recently closed capture */
+  void        *callout_data;      /* Data passed in with the call */
+  /* ------------------- Added for Version 1 -------------------------- */
+  int          pattern_position;  /* Offset to next item in the pattern */
+  int          next_item_length;  /* Length of next item in the pattern */
+  /* ------------------- Added for Version 2 -------------------------- */
+  const unsigned char *mark;      /* Pointer to current mark or NULL    */
+  /* ------------------------------------------------------------------ */
+} pcre_callout_block;
+
+/* Same structure as above, but with 16 bit char pointers. */
+
+typedef struct pcre16_callout_block {
+  int          version;           /* Identifies version of block */
+  /* ------------------------ Version 0 ------------------------------- */
+  int          callout_number;    /* Number compiled into pattern */
+  int         *offset_vector;     /* The offset vector */
+  PCRE_SPTR16  subject;           /* The subject being matched */
+  int          subject_length;    /* The length of the subject */
+  int          start_match;       /* Offset to start of this match attempt */
+  int          current_position;  /* Where we currently are in the subject */
+  int          capture_top;       /* Max current capture */
+  int          capture_last;      /* Most recently closed capture */
+  void        *callout_data;      /* Data passed in with the call */
+  /* ------------------- Added for Version 1 -------------------------- */
+  int          pattern_position;  /* Offset to next item in the pattern */
+  int          next_item_length;  /* Length of next item in the pattern */
+  /* ------------------- Added for Version 2 -------------------------- */
+  const PCRE_UCHAR16 *mark;       /* Pointer to current mark or NULL    */
+  /* ------------------------------------------------------------------ */
+} pcre16_callout_block;
+
+/* Same structure as above, but with 32 bit char pointers. */
+
+typedef struct pcre32_callout_block {
+  int          version;           /* Identifies version of block */
+  /* ------------------------ Version 0 ------------------------------- */
+  int          callout_number;    /* Number compiled into pattern */
+  int         *offset_vector;     /* The offset vector */
+  PCRE_SPTR32  subject;           /* The subject being matched */
+  int          subject_length;    /* The length of the subject */
+  int          start_match;       /* Offset to start of this match attempt */
+  int          current_position;  /* Where we currently are in the subject */
+  int          capture_top;       /* Max current capture */
+  int          capture_last;      /* Most recently closed capture */
+  void        *callout_data;      /* Data passed in with the call */
+  /* ------------------- Added for Version 1 -------------------------- */
+  int          pattern_position;  /* Offset to next item in the pattern */
+  int          next_item_length;  /* Length of next item in the pattern */
+  /* ------------------- Added for Version 2 -------------------------- */
+  const PCRE_UCHAR32 *mark;       /* Pointer to current mark or NULL    */
+  /* ------------------------------------------------------------------ */
+} pcre32_callout_block;
+
+/* Indirection for store get and free functions. These can be set to
+alternative malloc/free functions if required. Special ones are used in the
+non-recursive case for "frames". There is also an optional callout function
+that is triggered by the (?) regex item. For Virtual Pascal, these definitions
+have to take another form. */
+
+#ifndef VPCOMPAT
+PCRE_EXP_DECL void *(*pcre_malloc)(size_t);
+PCRE_EXP_DECL void  (*pcre_free)(void *);
+PCRE_EXP_DECL void *(*pcre_stack_malloc)(size_t);
+PCRE_EXP_DECL void  (*pcre_stack_free)(void *);
+PCRE_EXP_DECL int   (*pcre_callout)(pcre_callout_block *);
+
+PCRE_EXP_DECL void *(*pcre16_malloc)(size_t);
+PCRE_EXP_DECL void  (*pcre16_free)(void *);
+PCRE_EXP_DECL void *(*pcre16_stack_malloc)(size_t);
+PCRE_EXP_DECL void  (*pcre16_stack_free)(void *);
+PCRE_EXP_DECL int   (*pcre16_callout)(pcre16_callout_block *);
+
+PCRE_EXP_DECL void *(*pcre32_malloc)(size_t);
+PCRE_EXP_DECL void  (*pcre32_free)(void *);
+PCRE_EXP_DECL void *(*pcre32_stack_malloc)(size_t);
+PCRE_EXP_DECL void  (*pcre32_stack_free)(void *);
+PCRE_EXP_DECL int   (*pcre32_callout)(pcre32_callout_block *);
+#else   /* VPCOMPAT */
+PCRE_EXP_DECL void *pcre_malloc(size_t);
+PCRE_EXP_DECL void  pcre_free(void *);
+PCRE_EXP_DECL void *pcre_stack_malloc(size_t);
+PCRE_EXP_DECL void  pcre_stack_free(void *);
+PCRE_EXP_DECL int   pcre_callout(pcre_callout_block *);
+
+PCRE_EXP_DECL void *pcre16_malloc(size_t);
+PCRE_EXP_DECL void  pcre16_free(void *);
+PCRE_EXP_DECL void *pcre16_stack_malloc(size_t);
+PCRE_EXP_DECL void  pcre16_stack_free(void *);
+PCRE_EXP_DECL int   pcre16_callout(pcre16_callout_block *);
+
+PCRE_EXP_DECL void *pcre32_malloc(size_t);
+PCRE_EXP_DECL void  pcre32_free(void *);
+PCRE_EXP_DECL void *pcre32_stack_malloc(size_t);
+PCRE_EXP_DECL void  pcre32_stack_free(void *);
+PCRE_EXP_DECL int   pcre32_callout(pcre32_callout_block *);
+#endif  /* VPCOMPAT */
+
+/* User defined callback which provides a stack just before the match starts. */
+
+typedef pcre_jit_stack *(*pcre_jit_callback)(void *);
+typedef pcre16_jit_stack *(*pcre16_jit_callback)(void *);
+typedef pcre32_jit_stack *(*pcre32_jit_callback)(void *);
+
+/* Exported PCRE functions */
+
+PCRE_EXP_DECL pcre *pcre_compile(const char *, int, const char **, int *,
+                  const unsigned char *);
+PCRE_EXP_DECL pcre16 *pcre16_compile(PCRE_SPTR16, int, const char **, int *,
+                  const unsigned char *);
+PCRE_EXP_DECL pcre32 *pcre32_compile(PCRE_SPTR32, int, const char **, int *,
+                  const unsigned char *);
+PCRE_EXP_DECL pcre *pcre_compile2(const char *, int, int *, const char **,
+                  int *, const unsigned char *);
+PCRE_EXP_DECL pcre16 *pcre16_compile2(PCRE_SPTR16, int, int *, const char **,
+                  int *, const unsigned char *);
+PCRE_EXP_DECL pcre32 *pcre32_compile2(PCRE_SPTR32, int, int *, const char **,
+                  int *, const unsigned char *);
+PCRE_EXP_DECL int  pcre_config(int, void *);
+PCRE_EXP_DECL int  pcre16_config(int, void *);
+PCRE_EXP_DECL int  pcre32_config(int, void *);
+PCRE_EXP_DECL int  pcre_copy_named_substring(const pcre *, const char *,
+                  int *, int, const char *, char *, int);
+PCRE_EXP_DECL int  pcre16_copy_named_substring(const pcre16 *, PCRE_SPTR16,
+                  int *, int, PCRE_SPTR16, PCRE_UCHAR16 *, int);
+PCRE_EXP_DECL int  pcre32_copy_named_substring(const pcre32 *, PCRE_SPTR32,
+                  int *, int, PCRE_SPTR32, PCRE_UCHAR32 *, int);
+PCRE_EXP_DECL int  pcre_copy_substring(const char *, int *, int, int,
+                  char *, int);
+PCRE_EXP_DECL int  pcre16_copy_substring(PCRE_SPTR16, int *, int, int,
+                  PCRE_UCHAR16 *, int);
+PCRE_EXP_DECL int  pcre32_copy_substring(PCRE_SPTR32, int *, int, int,
+                  PCRE_UCHAR32 *, int);
+PCRE_EXP_DECL int  pcre_dfa_exec(const pcre *, const pcre_extra *,
+                  const char *, int, int, int, int *, int , int *, int);
+PCRE_EXP_DECL int  pcre16_dfa_exec(const pcre16 *, const pcre16_extra *,
+                  PCRE_SPTR16, int, int, int, int *, int , int *, int);
+PCRE_EXP_DECL int  pcre32_dfa_exec(const pcre32 *, const pcre32_extra *,
+                  PCRE_SPTR32, int, int, int, int *, int , int *, int);
+PCRE_EXP_DECL int  pcre_exec(const pcre *, const pcre_extra *, PCRE_SPTR,
+                   int, int, int, int *, int);
+PCRE_EXP_DECL int  pcre16_exec(const pcre16 *, const pcre16_extra *,
+                   PCRE_SPTR16, int, int, int, int *, int);
+PCRE_EXP_DECL int  pcre32_exec(const pcre32 *, const pcre32_extra *,
+                   PCRE_SPTR32, int, int, int, int *, int);
+PCRE_EXP_DECL int  pcre_jit_exec(const pcre *, const pcre_extra *,
+                   PCRE_SPTR, int, int, int, int *, int,
+                   pcre_jit_stack *);
+PCRE_EXP_DECL int  pcre16_jit_exec(const pcre16 *, const pcre16_extra *,
+                   PCRE_SPTR16, int, int, int, int *, int,
+                   pcre16_jit_stack *);
+PCRE_EXP_DECL int  pcre32_jit_exec(const pcre32 *, const pcre32_extra *,
+                   PCRE_SPTR32, int, int, int, int *, int,
+                   pcre32_jit_stack *);
+PCRE_EXP_DECL void pcre_free_substring(const char *);
+PCRE_EXP_DECL void pcre16_free_substring(PCRE_SPTR16);
+PCRE_EXP_DECL void pcre32_free_substring(PCRE_SPTR32);
+PCRE_EXP_DECL void pcre_free_substring_list(const char **);
+PCRE_EXP_DECL void pcre16_free_substring_list(PCRE_SPTR16 *);
+PCRE_EXP_DECL void pcre32_free_substring_list(PCRE_SPTR32 *);
+PCRE_EXP_DECL int  pcre_fullinfo(const pcre *, const pcre_extra *, int,
+                  void *);
+PCRE_EXP_DECL int  pcre16_fullinfo(const pcre16 *, const pcre16_extra *, int,
+                  void *);
+PCRE_EXP_DECL int  pcre32_fullinfo(const pcre32 *, const pcre32_extra *, int,
+                  void *);
+PCRE_EXP_DECL int  pcre_get_named_substring(const pcre *, const char *,
+                  int *, int, const char *, const char **);
+PCRE_EXP_DECL int  pcre16_get_named_substring(const pcre16 *, PCRE_SPTR16,
+                  int *, int, PCRE_SPTR16, PCRE_SPTR16 *);
+PCRE_EXP_DECL int  pcre32_get_named_substring(const pcre32 *, PCRE_SPTR32,
+                  int *, int, PCRE_SPTR32, PCRE_SPTR32 *);
+PCRE_EXP_DECL int  pcre_get_stringnumber(const pcre *, const char *);
+PCRE_EXP_DECL int  pcre16_get_stringnumber(const pcre16 *, PCRE_SPTR16);
+PCRE_EXP_DECL int  pcre32_get_stringnumber(const pcre32 *, PCRE_SPTR32);
+PCRE_EXP_DECL int  pcre_get_stringtable_entries(const pcre *, const char *,
+                  char **, char **);
+PCRE_EXP_DECL int  pcre16_get_stringtable_entries(const pcre16 *, PCRE_SPTR16,
+                  PCRE_UCHAR16 **, PCRE_UCHAR16 **);
+PCRE_EXP_DECL int  pcre32_get_stringtable_entries(const pcre32 *, PCRE_SPTR32,
+                  PCRE_UCHAR32 **, PCRE_UCHAR32 **);
+PCRE_EXP_DECL int  pcre_get_substring(const char *, int *, int, int,
+                  const char **);
+PCRE_EXP_DECL int  pcre16_get_substring(PCRE_SPTR16, int *, int, int,
+                  PCRE_SPTR16 *);
+PCRE_EXP_DECL int  pcre32_get_substring(PCRE_SPTR32, int *, int, int,
+                  PCRE_SPTR32 *);
+PCRE_EXP_DECL int  pcre_get_substring_list(const char *, int *, int,
+                  const char ***);
+PCRE_EXP_DECL int  pcre16_get_substring_list(PCRE_SPTR16, int *, int,
+                  PCRE_SPTR16 **);
+PCRE_EXP_DECL int  pcre32_get_substring_list(PCRE_SPTR32, int *, int,
+                  PCRE_SPTR32 **);
+PCRE_EXP_DECL const unsigned char *pcre_maketables(void);
+PCRE_EXP_DECL const unsigned char *pcre16_maketables(void);
+PCRE_EXP_DECL const unsigned char *pcre32_maketables(void);
+PCRE_EXP_DECL int  pcre_refcount(pcre *, int);
+PCRE_EXP_DECL int  pcre16_refcount(pcre16 *, int);
+PCRE_EXP_DECL int  pcre32_refcount(pcre32 *, int);
+PCRE_EXP_DECL pcre_extra *pcre_study(const pcre *, int, const char **);
+PCRE_EXP_DECL pcre16_extra *pcre16_study(const pcre16 *, int, const char **);
+PCRE_EXP_DECL pcre32_extra *pcre32_study(const pcre32 *, int, const char **);
+PCRE_EXP_DECL void pcre_free_study(pcre_extra *);
+PCRE_EXP_DECL void pcre16_free_study(pcre16_extra *);
+PCRE_EXP_DECL void pcre32_free_study(pcre32_extra *);
+PCRE_EXP_DECL const char *pcre_version(void);
+PCRE_EXP_DECL const char *pcre16_version(void);
+PCRE_EXP_DECL const char *pcre32_version(void);
+
+/* Utility functions for byte order swaps. */
+PCRE_EXP_DECL int  pcre_pattern_to_host_byte_order(pcre *, pcre_extra *,
+                  const unsigned char *);
+PCRE_EXP_DECL int  pcre16_pattern_to_host_byte_order(pcre16 *, pcre16_extra *,
+                  const unsigned char *);
+PCRE_EXP_DECL int  pcre32_pattern_to_host_byte_order(pcre32 *, pcre32_extra *,
+                  const unsigned char *);
+PCRE_EXP_DECL int  pcre16_utf16_to_host_byte_order(PCRE_UCHAR16 *,
+                  PCRE_SPTR16, int, int *, int);
+PCRE_EXP_DECL int  pcre32_utf32_to_host_byte_order(PCRE_UCHAR32 *,
+                  PCRE_SPTR32, int, int *, int);
+
+/* JIT compiler related functions. */
+
+PCRE_EXP_DECL pcre_jit_stack *pcre_jit_stack_alloc(int, int);
+PCRE_EXP_DECL pcre16_jit_stack *pcre16_jit_stack_alloc(int, int);
+PCRE_EXP_DECL pcre32_jit_stack *pcre32_jit_stack_alloc(int, int);
+PCRE_EXP_DECL void pcre_jit_stack_free(pcre_jit_stack *);
+PCRE_EXP_DECL void pcre16_jit_stack_free(pcre16_jit_stack *);
+PCRE_EXP_DECL void pcre32_jit_stack_free(pcre32_jit_stack *);
+PCRE_EXP_DECL void pcre_assign_jit_stack(pcre_extra *,
+                  pcre_jit_callback, void *);
+PCRE_EXP_DECL void pcre16_assign_jit_stack(pcre16_extra *,
+                  pcre16_jit_callback, void *);
+PCRE_EXP_DECL void pcre32_assign_jit_stack(pcre32_extra *,
+                  pcre32_jit_callback, void *);
+
+#ifdef __cplusplus
+}  /* extern "C" */
+#endif
+
+#endif /* End of pcre.h */
diff --git a/pcre/pcre.h.in b/pcre/pcre.h.in
new file mode 100644
index 00000000000..d6766339090
--- /dev/null
+++ b/pcre/pcre.h.in
@@ -0,0 +1,663 @@
+/*************************************************
+*       Perl-Compatible Regular Expressions      *
+*************************************************/
+
+/* This is the public header file for the PCRE library, to be #included by
+applications that call the PCRE functions.
+
+           Copyright (c) 1997-2013 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+#ifndef _PCRE_H
+#define _PCRE_H
+
+/* The current PCRE version information. */
+
+#define PCRE_MAJOR          @PCRE_MAJOR@
+#define PCRE_MINOR          @PCRE_MINOR@
+#define PCRE_PRERELEASE     @PCRE_PRERELEASE@
+#define PCRE_DATE           @PCRE_DATE@
+
+/* When an application links to a PCRE DLL in Windows, the symbols that are
+imported have to be identified as such. When building PCRE, the appropriate
+export setting is defined in pcre_internal.h, which includes this file. So we
+don't change existing definitions of PCRE_EXP_DECL and PCRECPP_EXP_DECL. */
+
+#if defined(_WIN32) && !defined(PCRE_STATIC)
+#  ifndef PCRE_EXP_DECL
+#    define PCRE_EXP_DECL  extern __declspec(dllimport)
+#  endif
+#  ifdef __cplusplus
+#    ifndef PCRECPP_EXP_DECL
+#      define PCRECPP_EXP_DECL  extern __declspec(dllimport)
+#    endif
+#    ifndef PCRECPP_EXP_DEFN
+#      define PCRECPP_EXP_DEFN  __declspec(dllimport)
+#    endif
+#  endif
+#endif
+
+/* By default, we use the standard "extern" declarations. */
+
+#ifndef PCRE_EXP_DECL
+#  ifdef __cplusplus
+#    define PCRE_EXP_DECL  extern "C"
+#  else
+#    define PCRE_EXP_DECL  extern
+#  endif
+#endif
+
+#ifdef __cplusplus
+#  ifndef PCRECPP_EXP_DECL
+#    define PCRECPP_EXP_DECL  extern
+#  endif
+#  ifndef PCRECPP_EXP_DEFN
+#    define PCRECPP_EXP_DEFN
+#  endif
+#endif
+
+/* Have to include stdlib.h in order to ensure that size_t is defined;
+it is needed here for malloc. */
+
+#include 
+
+/* Allow for C++ users */
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+/* Public options. Some are compile-time only, some are run-time only, and some
+are both. Most of the compile-time options are saved with the compiled regex so
+that they can be inspected during studying (and therefore JIT compiling). Note
+that pcre_study() has its own set of options. Originally, all the options
+defined here used distinct bits. However, almost all the bits in a 32-bit word
+are now used, so in order to conserve them, option bits that were previously
+only recognized at matching time (i.e. by pcre_exec() or pcre_dfa_exec()) may
+also be used for compile-time options that affect only compiling and are not
+relevant for studying or JIT compiling.
+
+Some options for pcre_compile() change its behaviour but do not affect the
+behaviour of the execution functions. Other options are passed through to the
+execution functions and affect their behaviour, with or without affecting the
+behaviour of pcre_compile().
+
+Options that can be passed to pcre_compile() are tagged Cx below, with these
+variants:
+
+C1   Affects compile only
+C2   Does not affect compile; affects exec, dfa_exec
+C3   Affects compile, exec, dfa_exec
+C4   Affects compile, exec, dfa_exec, study
+C5   Affects compile, exec, study
+
+Options that can be set for pcre_exec() and/or pcre_dfa_exec() are flagged with
+E and D, respectively. They take precedence over C3, C4, and C5 settings passed
+from pcre_compile(). Those that are compatible with JIT execution are flagged
+with J. */
+
+#define PCRE_CASELESS           0x00000001  /* C1       */
+#define PCRE_MULTILINE          0x00000002  /* C1       */
+#define PCRE_DOTALL             0x00000004  /* C1       */
+#define PCRE_EXTENDED           0x00000008  /* C1       */
+#define PCRE_ANCHORED           0x00000010  /* C4 E D   */
+#define PCRE_DOLLAR_ENDONLY     0x00000020  /* C2       */
+#define PCRE_EXTRA              0x00000040  /* C1       */
+#define PCRE_NOTBOL             0x00000080  /*    E D J */
+#define PCRE_NOTEOL             0x00000100  /*    E D J */
+#define PCRE_UNGREEDY           0x00000200  /* C1       */
+#define PCRE_NOTEMPTY           0x00000400  /*    E D J */
+#define PCRE_UTF8               0x00000800  /* C4        )          */
+#define PCRE_UTF16              0x00000800  /* C4        ) Synonyms */
+#define PCRE_UTF32              0x00000800  /* C4        )          */
+#define PCRE_NO_AUTO_CAPTURE    0x00001000  /* C1       */
+#define PCRE_NO_UTF8_CHECK      0x00002000  /* C1 E D J  )          */
+#define PCRE_NO_UTF16_CHECK     0x00002000  /* C1 E D J  ) Synonyms */
+#define PCRE_NO_UTF32_CHECK     0x00002000  /* C1 E D J  )          */
+#define PCRE_AUTO_CALLOUT       0x00004000  /* C1       */
+#define PCRE_PARTIAL_SOFT       0x00008000  /*    E D J  ) Synonyms */
+#define PCRE_PARTIAL            0x00008000  /*    E D J  )          */
+
+/* This pair use the same bit. */
+#define PCRE_NEVER_UTF          0x00010000  /* C1        ) Overlaid */
+#define PCRE_DFA_SHORTEST       0x00010000  /*      D    ) Overlaid */
+
+#define PCRE_DFA_RESTART        0x00020000  /*      D   */
+#define PCRE_FIRSTLINE          0x00040000  /* C3       */
+#define PCRE_DUPNAMES           0x00080000  /* C1       */
+#define PCRE_NEWLINE_CR         0x00100000  /* C3 E D   */
+#define PCRE_NEWLINE_LF         0x00200000  /* C3 E D   */
+#define PCRE_NEWLINE_CRLF       0x00300000  /* C3 E D   */
+#define PCRE_NEWLINE_ANY        0x00400000  /* C3 E D   */
+#define PCRE_NEWLINE_ANYCRLF    0x00500000  /* C3 E D   */
+#define PCRE_BSR_ANYCRLF        0x00800000  /* C3 E D   */
+#define PCRE_BSR_UNICODE        0x01000000  /* C3 E D   */
+#define PCRE_JAVASCRIPT_COMPAT  0x02000000  /* C5       */
+#define PCRE_NO_START_OPTIMIZE  0x04000000  /* C2 E D    ) Synonyms */
+#define PCRE_NO_START_OPTIMISE  0x04000000  /* C2 E D    )          */
+#define PCRE_PARTIAL_HARD       0x08000000  /*    E D J */
+#define PCRE_NOTEMPTY_ATSTART   0x10000000  /*    E D J */
+#define PCRE_UCP                0x20000000  /* C3       */
+
+/* Exec-time and get/set-time error codes */
+
+#define PCRE_ERROR_NOMATCH          (-1)
+#define PCRE_ERROR_NULL             (-2)
+#define PCRE_ERROR_BADOPTION        (-3)
+#define PCRE_ERROR_BADMAGIC         (-4)
+#define PCRE_ERROR_UNKNOWN_OPCODE   (-5)
+#define PCRE_ERROR_UNKNOWN_NODE     (-5)  /* For backward compatibility */
+#define PCRE_ERROR_NOMEMORY         (-6)
+#define PCRE_ERROR_NOSUBSTRING      (-7)
+#define PCRE_ERROR_MATCHLIMIT       (-8)
+#define PCRE_ERROR_CALLOUT          (-9)  /* Never used by PCRE itself */
+#define PCRE_ERROR_BADUTF8         (-10)  /* Same for 8/16/32 */
+#define PCRE_ERROR_BADUTF16        (-10)  /* Same for 8/16/32 */
+#define PCRE_ERROR_BADUTF32        (-10)  /* Same for 8/16/32 */
+#define PCRE_ERROR_BADUTF8_OFFSET  (-11)  /* Same for 8/16 */
+#define PCRE_ERROR_BADUTF16_OFFSET (-11)  /* Same for 8/16 */
+#define PCRE_ERROR_PARTIAL         (-12)
+#define PCRE_ERROR_BADPARTIAL      (-13)
+#define PCRE_ERROR_INTERNAL        (-14)
+#define PCRE_ERROR_BADCOUNT        (-15)
+#define PCRE_ERROR_DFA_UITEM       (-16)
+#define PCRE_ERROR_DFA_UCOND       (-17)
+#define PCRE_ERROR_DFA_UMLIMIT     (-18)
+#define PCRE_ERROR_DFA_WSSIZE      (-19)
+#define PCRE_ERROR_DFA_RECURSE     (-20)
+#define PCRE_ERROR_RECURSIONLIMIT  (-21)
+#define PCRE_ERROR_NULLWSLIMIT     (-22)  /* No longer actually used */
+#define PCRE_ERROR_BADNEWLINE      (-23)
+#define PCRE_ERROR_BADOFFSET       (-24)
+#define PCRE_ERROR_SHORTUTF8       (-25)
+#define PCRE_ERROR_SHORTUTF16      (-25)  /* Same for 8/16 */
+#define PCRE_ERROR_RECURSELOOP     (-26)
+#define PCRE_ERROR_JIT_STACKLIMIT  (-27)
+#define PCRE_ERROR_BADMODE         (-28)
+#define PCRE_ERROR_BADENDIANNESS   (-29)
+#define PCRE_ERROR_DFA_BADRESTART  (-30)
+#define PCRE_ERROR_JIT_BADOPTION   (-31)
+#define PCRE_ERROR_BADLENGTH       (-32)
+#define PCRE_ERROR_UNSET           (-33)
+
+/* Specific error codes for UTF-8 validity checks */
+
+#define PCRE_UTF8_ERR0               0
+#define PCRE_UTF8_ERR1               1
+#define PCRE_UTF8_ERR2               2
+#define PCRE_UTF8_ERR3               3
+#define PCRE_UTF8_ERR4               4
+#define PCRE_UTF8_ERR5               5
+#define PCRE_UTF8_ERR6               6
+#define PCRE_UTF8_ERR7               7
+#define PCRE_UTF8_ERR8               8
+#define PCRE_UTF8_ERR9               9
+#define PCRE_UTF8_ERR10             10
+#define PCRE_UTF8_ERR11             11
+#define PCRE_UTF8_ERR12             12
+#define PCRE_UTF8_ERR13             13
+#define PCRE_UTF8_ERR14             14
+#define PCRE_UTF8_ERR15             15
+#define PCRE_UTF8_ERR16             16
+#define PCRE_UTF8_ERR17             17
+#define PCRE_UTF8_ERR18             18
+#define PCRE_UTF8_ERR19             19
+#define PCRE_UTF8_ERR20             20
+#define PCRE_UTF8_ERR21             21
+#define PCRE_UTF8_ERR22             22  /* Unused (was non-character) */
+
+/* Specific error codes for UTF-16 validity checks */
+
+#define PCRE_UTF16_ERR0              0
+#define PCRE_UTF16_ERR1              1
+#define PCRE_UTF16_ERR2              2
+#define PCRE_UTF16_ERR3              3
+#define PCRE_UTF16_ERR4              4  /* Unused (was non-character) */
+
+/* Specific error codes for UTF-32 validity checks */
+
+#define PCRE_UTF32_ERR0              0
+#define PCRE_UTF32_ERR1              1
+#define PCRE_UTF32_ERR2              2  /* Unused (was non-character) */
+#define PCRE_UTF32_ERR3              3
+
+/* Request types for pcre_fullinfo() */
+
+#define PCRE_INFO_OPTIONS            0
+#define PCRE_INFO_SIZE               1
+#define PCRE_INFO_CAPTURECOUNT       2
+#define PCRE_INFO_BACKREFMAX         3
+#define PCRE_INFO_FIRSTBYTE          4
+#define PCRE_INFO_FIRSTCHAR          4  /* For backwards compatibility */
+#define PCRE_INFO_FIRSTTABLE         5
+#define PCRE_INFO_LASTLITERAL        6
+#define PCRE_INFO_NAMEENTRYSIZE      7
+#define PCRE_INFO_NAMECOUNT          8
+#define PCRE_INFO_NAMETABLE          9
+#define PCRE_INFO_STUDYSIZE         10
+#define PCRE_INFO_DEFAULT_TABLES    11
+#define PCRE_INFO_OKPARTIAL         12
+#define PCRE_INFO_JCHANGED          13
+#define PCRE_INFO_HASCRORLF         14
+#define PCRE_INFO_MINLENGTH         15
+#define PCRE_INFO_JIT               16
+#define PCRE_INFO_JITSIZE           17
+#define PCRE_INFO_MAXLOOKBEHIND     18
+#define PCRE_INFO_FIRSTCHARACTER    19
+#define PCRE_INFO_FIRSTCHARACTERFLAGS 20
+#define PCRE_INFO_REQUIREDCHAR      21
+#define PCRE_INFO_REQUIREDCHARFLAGS 22
+#define PCRE_INFO_MATCHLIMIT        23
+#define PCRE_INFO_RECURSIONLIMIT    24
+
+/* Request types for pcre_config(). Do not re-arrange, in order to remain
+compatible. */
+
+#define PCRE_CONFIG_UTF8                    0
+#define PCRE_CONFIG_NEWLINE                 1
+#define PCRE_CONFIG_LINK_SIZE               2
+#define PCRE_CONFIG_POSIX_MALLOC_THRESHOLD  3
+#define PCRE_CONFIG_MATCH_LIMIT             4
+#define PCRE_CONFIG_STACKRECURSE            5
+#define PCRE_CONFIG_UNICODE_PROPERTIES      6
+#define PCRE_CONFIG_MATCH_LIMIT_RECURSION   7
+#define PCRE_CONFIG_BSR                     8
+#define PCRE_CONFIG_JIT                     9
+#define PCRE_CONFIG_UTF16                  10
+#define PCRE_CONFIG_JITTARGET              11
+#define PCRE_CONFIG_UTF32                  12
+
+/* Request types for pcre_study(). Do not re-arrange, in order to remain
+compatible. */
+
+#define PCRE_STUDY_JIT_COMPILE                0x0001
+#define PCRE_STUDY_JIT_PARTIAL_SOFT_COMPILE   0x0002
+#define PCRE_STUDY_JIT_PARTIAL_HARD_COMPILE   0x0004
+#define PCRE_STUDY_EXTRA_NEEDED               0x0008
+
+/* Bit flags for the pcre[16|32]_extra structure. Do not re-arrange or redefine
+these bits, just add new ones on the end, in order to remain compatible. */
+
+#define PCRE_EXTRA_STUDY_DATA             0x0001
+#define PCRE_EXTRA_MATCH_LIMIT            0x0002
+#define PCRE_EXTRA_CALLOUT_DATA           0x0004
+#define PCRE_EXTRA_TABLES                 0x0008
+#define PCRE_EXTRA_MATCH_LIMIT_RECURSION  0x0010
+#define PCRE_EXTRA_MARK                   0x0020
+#define PCRE_EXTRA_EXECUTABLE_JIT         0x0040
+
+/* Types */
+
+struct real_pcre;                 /* declaration; the definition is private  */
+typedef struct real_pcre pcre;
+
+struct real_pcre16;               /* declaration; the definition is private  */
+typedef struct real_pcre16 pcre16;
+
+struct real_pcre32;               /* declaration; the definition is private  */
+typedef struct real_pcre32 pcre32;
+
+struct real_pcre_jit_stack;       /* declaration; the definition is private  */
+typedef struct real_pcre_jit_stack pcre_jit_stack;
+
+struct real_pcre16_jit_stack;     /* declaration; the definition is private  */
+typedef struct real_pcre16_jit_stack pcre16_jit_stack;
+
+struct real_pcre32_jit_stack;     /* declaration; the definition is private  */
+typedef struct real_pcre32_jit_stack pcre32_jit_stack;
+
+/* If PCRE is compiled with 16 bit character support, PCRE_UCHAR16 must contain
+a 16 bit wide signed data type. Otherwise it can be a dummy data type since
+pcre16 functions are not implemented. There is a check for this in pcre_internal.h. */
+#ifndef PCRE_UCHAR16
+#define PCRE_UCHAR16 unsigned short
+#endif
+
+#ifndef PCRE_SPTR16
+#define PCRE_SPTR16 const PCRE_UCHAR16 *
+#endif
+
+/* If PCRE is compiled with 32 bit character support, PCRE_UCHAR32 must contain
+a 32 bit wide signed data type. Otherwise it can be a dummy data type since
+pcre32 functions are not implemented. There is a check for this in pcre_internal.h. */
+#ifndef PCRE_UCHAR32
+#define PCRE_UCHAR32 unsigned int
+#endif
+
+#ifndef PCRE_SPTR32
+#define PCRE_SPTR32 const PCRE_UCHAR32 *
+#endif
+
+/* When PCRE is compiled as a C++ library, the subject pointer type can be
+replaced with a custom type. For conventional use, the public interface is a
+const char *. */
+
+#ifndef PCRE_SPTR
+#define PCRE_SPTR const char *
+#endif
+
+/* The structure for passing additional data to pcre_exec(). This is defined in
+such as way as to be extensible. Always add new fields at the end, in order to
+remain compatible. */
+
+typedef struct pcre_extra {
+  unsigned long int flags;        /* Bits for which fields are set */
+  void *study_data;               /* Opaque data from pcre_study() */
+  unsigned long int match_limit;  /* Maximum number of calls to match() */
+  void *callout_data;             /* Data passed back in callouts */
+  const unsigned char *tables;    /* Pointer to character tables */
+  unsigned long int match_limit_recursion; /* Max recursive calls to match() */
+  unsigned char **mark;           /* For passing back a mark pointer */
+  void *executable_jit;           /* Contains a pointer to a compiled jit code */
+} pcre_extra;
+
+/* Same structure as above, but with 16 bit char pointers. */
+
+typedef struct pcre16_extra {
+  unsigned long int flags;        /* Bits for which fields are set */
+  void *study_data;               /* Opaque data from pcre_study() */
+  unsigned long int match_limit;  /* Maximum number of calls to match() */
+  void *callout_data;             /* Data passed back in callouts */
+  const unsigned char *tables;    /* Pointer to character tables */
+  unsigned long int match_limit_recursion; /* Max recursive calls to match() */
+  PCRE_UCHAR16 **mark;            /* For passing back a mark pointer */
+  void *executable_jit;           /* Contains a pointer to a compiled jit code */
+} pcre16_extra;
+
+/* Same structure as above, but with 32 bit char pointers. */
+
+typedef struct pcre32_extra {
+  unsigned long int flags;        /* Bits for which fields are set */
+  void *study_data;               /* Opaque data from pcre_study() */
+  unsigned long int match_limit;  /* Maximum number of calls to match() */
+  void *callout_data;             /* Data passed back in callouts */
+  const unsigned char *tables;    /* Pointer to character tables */
+  unsigned long int match_limit_recursion; /* Max recursive calls to match() */
+  PCRE_UCHAR32 **mark;            /* For passing back a mark pointer */
+  void *executable_jit;           /* Contains a pointer to a compiled jit code */
+} pcre32_extra;
+
+/* The structure for passing out data via the pcre_callout_function. We use a
+structure so that new fields can be added on the end in future versions,
+without changing the API of the function, thereby allowing old clients to work
+without modification. */
+
+typedef struct pcre_callout_block {
+  int          version;           /* Identifies version of block */
+  /* ------------------------ Version 0 ------------------------------- */
+  int          callout_number;    /* Number compiled into pattern */
+  int         *offset_vector;     /* The offset vector */
+  PCRE_SPTR    subject;           /* The subject being matched */
+  int          subject_length;    /* The length of the subject */
+  int          start_match;       /* Offset to start of this match attempt */
+  int          current_position;  /* Where we currently are in the subject */
+  int          capture_top;       /* Max current capture */
+  int          capture_last;      /* Most recently closed capture */
+  void        *callout_data;      /* Data passed in with the call */
+  /* ------------------- Added for Version 1 -------------------------- */
+  int          pattern_position;  /* Offset to next item in the pattern */
+  int          next_item_length;  /* Length of next item in the pattern */
+  /* ------------------- Added for Version 2 -------------------------- */
+  const unsigned char *mark;      /* Pointer to current mark or NULL    */
+  /* ------------------------------------------------------------------ */
+} pcre_callout_block;
+
+/* Same structure as above, but with 16 bit char pointers. */
+
+typedef struct pcre16_callout_block {
+  int          version;           /* Identifies version of block */
+  /* ------------------------ Version 0 ------------------------------- */
+  int          callout_number;    /* Number compiled into pattern */
+  int         *offset_vector;     /* The offset vector */
+  PCRE_SPTR16  subject;           /* The subject being matched */
+  int          subject_length;    /* The length of the subject */
+  int          start_match;       /* Offset to start of this match attempt */
+  int          current_position;  /* Where we currently are in the subject */
+  int          capture_top;       /* Max current capture */
+  int          capture_last;      /* Most recently closed capture */
+  void        *callout_data;      /* Data passed in with the call */
+  /* ------------------- Added for Version 1 -------------------------- */
+  int          pattern_position;  /* Offset to next item in the pattern */
+  int          next_item_length;  /* Length of next item in the pattern */
+  /* ------------------- Added for Version 2 -------------------------- */
+  const PCRE_UCHAR16 *mark;       /* Pointer to current mark or NULL    */
+  /* ------------------------------------------------------------------ */
+} pcre16_callout_block;
+
+/* Same structure as above, but with 32 bit char pointers. */
+
+typedef struct pcre32_callout_block {
+  int          version;           /* Identifies version of block */
+  /* ------------------------ Version 0 ------------------------------- */
+  int          callout_number;    /* Number compiled into pattern */
+  int         *offset_vector;     /* The offset vector */
+  PCRE_SPTR32  subject;           /* The subject being matched */
+  int          subject_length;    /* The length of the subject */
+  int          start_match;       /* Offset to start of this match attempt */
+  int          current_position;  /* Where we currently are in the subject */
+  int          capture_top;       /* Max current capture */
+  int          capture_last;      /* Most recently closed capture */
+  void        *callout_data;      /* Data passed in with the call */
+  /* ------------------- Added for Version 1 -------------------------- */
+  int          pattern_position;  /* Offset to next item in the pattern */
+  int          next_item_length;  /* Length of next item in the pattern */
+  /* ------------------- Added for Version 2 -------------------------- */
+  const PCRE_UCHAR32 *mark;       /* Pointer to current mark or NULL    */
+  /* ------------------------------------------------------------------ */
+} pcre32_callout_block;
+
+/* Indirection for store get and free functions. These can be set to
+alternative malloc/free functions if required. Special ones are used in the
+non-recursive case for "frames". There is also an optional callout function
+that is triggered by the (?) regex item. For Virtual Pascal, these definitions
+have to take another form. */
+
+#ifndef VPCOMPAT
+PCRE_EXP_DECL void *(*pcre_malloc)(size_t);
+PCRE_EXP_DECL void  (*pcre_free)(void *);
+PCRE_EXP_DECL void *(*pcre_stack_malloc)(size_t);
+PCRE_EXP_DECL void  (*pcre_stack_free)(void *);
+PCRE_EXP_DECL int   (*pcre_callout)(pcre_callout_block *);
+
+PCRE_EXP_DECL void *(*pcre16_malloc)(size_t);
+PCRE_EXP_DECL void  (*pcre16_free)(void *);
+PCRE_EXP_DECL void *(*pcre16_stack_malloc)(size_t);
+PCRE_EXP_DECL void  (*pcre16_stack_free)(void *);
+PCRE_EXP_DECL int   (*pcre16_callout)(pcre16_callout_block *);
+
+PCRE_EXP_DECL void *(*pcre32_malloc)(size_t);
+PCRE_EXP_DECL void  (*pcre32_free)(void *);
+PCRE_EXP_DECL void *(*pcre32_stack_malloc)(size_t);
+PCRE_EXP_DECL void  (*pcre32_stack_free)(void *);
+PCRE_EXP_DECL int   (*pcre32_callout)(pcre32_callout_block *);
+#else   /* VPCOMPAT */
+PCRE_EXP_DECL void *pcre_malloc(size_t);
+PCRE_EXP_DECL void  pcre_free(void *);
+PCRE_EXP_DECL void *pcre_stack_malloc(size_t);
+PCRE_EXP_DECL void  pcre_stack_free(void *);
+PCRE_EXP_DECL int   pcre_callout(pcre_callout_block *);
+
+PCRE_EXP_DECL void *pcre16_malloc(size_t);
+PCRE_EXP_DECL void  pcre16_free(void *);
+PCRE_EXP_DECL void *pcre16_stack_malloc(size_t);
+PCRE_EXP_DECL void  pcre16_stack_free(void *);
+PCRE_EXP_DECL int   pcre16_callout(pcre16_callout_block *);
+
+PCRE_EXP_DECL void *pcre32_malloc(size_t);
+PCRE_EXP_DECL void  pcre32_free(void *);
+PCRE_EXP_DECL void *pcre32_stack_malloc(size_t);
+PCRE_EXP_DECL void  pcre32_stack_free(void *);
+PCRE_EXP_DECL int   pcre32_callout(pcre32_callout_block *);
+#endif  /* VPCOMPAT */
+
+/* User defined callback which provides a stack just before the match starts. */
+
+typedef pcre_jit_stack *(*pcre_jit_callback)(void *);
+typedef pcre16_jit_stack *(*pcre16_jit_callback)(void *);
+typedef pcre32_jit_stack *(*pcre32_jit_callback)(void *);
+
+/* Exported PCRE functions */
+
+PCRE_EXP_DECL pcre *pcre_compile(const char *, int, const char **, int *,
+                  const unsigned char *);
+PCRE_EXP_DECL pcre16 *pcre16_compile(PCRE_SPTR16, int, const char **, int *,
+                  const unsigned char *);
+PCRE_EXP_DECL pcre32 *pcre32_compile(PCRE_SPTR32, int, const char **, int *,
+                  const unsigned char *);
+PCRE_EXP_DECL pcre *pcre_compile2(const char *, int, int *, const char **,
+                  int *, const unsigned char *);
+PCRE_EXP_DECL pcre16 *pcre16_compile2(PCRE_SPTR16, int, int *, const char **,
+                  int *, const unsigned char *);
+PCRE_EXP_DECL pcre32 *pcre32_compile2(PCRE_SPTR32, int, int *, const char **,
+                  int *, const unsigned char *);
+PCRE_EXP_DECL int  pcre_config(int, void *);
+PCRE_EXP_DECL int  pcre16_config(int, void *);
+PCRE_EXP_DECL int  pcre32_config(int, void *);
+PCRE_EXP_DECL int  pcre_copy_named_substring(const pcre *, const char *,
+                  int *, int, const char *, char *, int);
+PCRE_EXP_DECL int  pcre16_copy_named_substring(const pcre16 *, PCRE_SPTR16,
+                  int *, int, PCRE_SPTR16, PCRE_UCHAR16 *, int);
+PCRE_EXP_DECL int  pcre32_copy_named_substring(const pcre32 *, PCRE_SPTR32,
+                  int *, int, PCRE_SPTR32, PCRE_UCHAR32 *, int);
+PCRE_EXP_DECL int  pcre_copy_substring(const char *, int *, int, int,
+                  char *, int);
+PCRE_EXP_DECL int  pcre16_copy_substring(PCRE_SPTR16, int *, int, int,
+                  PCRE_UCHAR16 *, int);
+PCRE_EXP_DECL int  pcre32_copy_substring(PCRE_SPTR32, int *, int, int,
+                  PCRE_UCHAR32 *, int);
+PCRE_EXP_DECL int  pcre_dfa_exec(const pcre *, const pcre_extra *,
+                  const char *, int, int, int, int *, int , int *, int);
+PCRE_EXP_DECL int  pcre16_dfa_exec(const pcre16 *, const pcre16_extra *,
+                  PCRE_SPTR16, int, int, int, int *, int , int *, int);
+PCRE_EXP_DECL int  pcre32_dfa_exec(const pcre32 *, const pcre32_extra *,
+                  PCRE_SPTR32, int, int, int, int *, int , int *, int);
+PCRE_EXP_DECL int  pcre_exec(const pcre *, const pcre_extra *, PCRE_SPTR,
+                   int, int, int, int *, int);
+PCRE_EXP_DECL int  pcre16_exec(const pcre16 *, const pcre16_extra *,
+                   PCRE_SPTR16, int, int, int, int *, int);
+PCRE_EXP_DECL int  pcre32_exec(const pcre32 *, const pcre32_extra *,
+                   PCRE_SPTR32, int, int, int, int *, int);
+PCRE_EXP_DECL int  pcre_jit_exec(const pcre *, const pcre_extra *,
+                   PCRE_SPTR, int, int, int, int *, int,
+                   pcre_jit_stack *);
+PCRE_EXP_DECL int  pcre16_jit_exec(const pcre16 *, const pcre16_extra *,
+                   PCRE_SPTR16, int, int, int, int *, int,
+                   pcre16_jit_stack *);
+PCRE_EXP_DECL int  pcre32_jit_exec(const pcre32 *, const pcre32_extra *,
+                   PCRE_SPTR32, int, int, int, int *, int,
+                   pcre32_jit_stack *);
+PCRE_EXP_DECL void pcre_free_substring(const char *);
+PCRE_EXP_DECL void pcre16_free_substring(PCRE_SPTR16);
+PCRE_EXP_DECL void pcre32_free_substring(PCRE_SPTR32);
+PCRE_EXP_DECL void pcre_free_substring_list(const char **);
+PCRE_EXP_DECL void pcre16_free_substring_list(PCRE_SPTR16 *);
+PCRE_EXP_DECL void pcre32_free_substring_list(PCRE_SPTR32 *);
+PCRE_EXP_DECL int  pcre_fullinfo(const pcre *, const pcre_extra *, int,
+                  void *);
+PCRE_EXP_DECL int  pcre16_fullinfo(const pcre16 *, const pcre16_extra *, int,
+                  void *);
+PCRE_EXP_DECL int  pcre32_fullinfo(const pcre32 *, const pcre32_extra *, int,
+                  void *);
+PCRE_EXP_DECL int  pcre_get_named_substring(const pcre *, const char *,
+                  int *, int, const char *, const char **);
+PCRE_EXP_DECL int  pcre16_get_named_substring(const pcre16 *, PCRE_SPTR16,
+                  int *, int, PCRE_SPTR16, PCRE_SPTR16 *);
+PCRE_EXP_DECL int  pcre32_get_named_substring(const pcre32 *, PCRE_SPTR32,
+                  int *, int, PCRE_SPTR32, PCRE_SPTR32 *);
+PCRE_EXP_DECL int  pcre_get_stringnumber(const pcre *, const char *);
+PCRE_EXP_DECL int  pcre16_get_stringnumber(const pcre16 *, PCRE_SPTR16);
+PCRE_EXP_DECL int  pcre32_get_stringnumber(const pcre32 *, PCRE_SPTR32);
+PCRE_EXP_DECL int  pcre_get_stringtable_entries(const pcre *, const char *,
+                  char **, char **);
+PCRE_EXP_DECL int  pcre16_get_stringtable_entries(const pcre16 *, PCRE_SPTR16,
+                  PCRE_UCHAR16 **, PCRE_UCHAR16 **);
+PCRE_EXP_DECL int  pcre32_get_stringtable_entries(const pcre32 *, PCRE_SPTR32,
+                  PCRE_UCHAR32 **, PCRE_UCHAR32 **);
+PCRE_EXP_DECL int  pcre_get_substring(const char *, int *, int, int,
+                  const char **);
+PCRE_EXP_DECL int  pcre16_get_substring(PCRE_SPTR16, int *, int, int,
+                  PCRE_SPTR16 *);
+PCRE_EXP_DECL int  pcre32_get_substring(PCRE_SPTR32, int *, int, int,
+                  PCRE_SPTR32 *);
+PCRE_EXP_DECL int  pcre_get_substring_list(const char *, int *, int,
+                  const char ***);
+PCRE_EXP_DECL int  pcre16_get_substring_list(PCRE_SPTR16, int *, int,
+                  PCRE_SPTR16 **);
+PCRE_EXP_DECL int  pcre32_get_substring_list(PCRE_SPTR32, int *, int,
+                  PCRE_SPTR32 **);
+PCRE_EXP_DECL const unsigned char *pcre_maketables(void);
+PCRE_EXP_DECL const unsigned char *pcre16_maketables(void);
+PCRE_EXP_DECL const unsigned char *pcre32_maketables(void);
+PCRE_EXP_DECL int  pcre_refcount(pcre *, int);
+PCRE_EXP_DECL int  pcre16_refcount(pcre16 *, int);
+PCRE_EXP_DECL int  pcre32_refcount(pcre32 *, int);
+PCRE_EXP_DECL pcre_extra *pcre_study(const pcre *, int, const char **);
+PCRE_EXP_DECL pcre16_extra *pcre16_study(const pcre16 *, int, const char **);
+PCRE_EXP_DECL pcre32_extra *pcre32_study(const pcre32 *, int, const char **);
+PCRE_EXP_DECL void pcre_free_study(pcre_extra *);
+PCRE_EXP_DECL void pcre16_free_study(pcre16_extra *);
+PCRE_EXP_DECL void pcre32_free_study(pcre32_extra *);
+PCRE_EXP_DECL const char *pcre_version(void);
+PCRE_EXP_DECL const char *pcre16_version(void);
+PCRE_EXP_DECL const char *pcre32_version(void);
+
+/* Utility functions for byte order swaps. */
+PCRE_EXP_DECL int  pcre_pattern_to_host_byte_order(pcre *, pcre_extra *,
+                  const unsigned char *);
+PCRE_EXP_DECL int  pcre16_pattern_to_host_byte_order(pcre16 *, pcre16_extra *,
+                  const unsigned char *);
+PCRE_EXP_DECL int  pcre32_pattern_to_host_byte_order(pcre32 *, pcre32_extra *,
+                  const unsigned char *);
+PCRE_EXP_DECL int  pcre16_utf16_to_host_byte_order(PCRE_UCHAR16 *,
+                  PCRE_SPTR16, int, int *, int);
+PCRE_EXP_DECL int  pcre32_utf32_to_host_byte_order(PCRE_UCHAR32 *,
+                  PCRE_SPTR32, int, int *, int);
+
+/* JIT compiler related functions. */
+
+PCRE_EXP_DECL pcre_jit_stack *pcre_jit_stack_alloc(int, int);
+PCRE_EXP_DECL pcre16_jit_stack *pcre16_jit_stack_alloc(int, int);
+PCRE_EXP_DECL pcre32_jit_stack *pcre32_jit_stack_alloc(int, int);
+PCRE_EXP_DECL void pcre_jit_stack_free(pcre_jit_stack *);
+PCRE_EXP_DECL void pcre16_jit_stack_free(pcre16_jit_stack *);
+PCRE_EXP_DECL void pcre32_jit_stack_free(pcre32_jit_stack *);
+PCRE_EXP_DECL void pcre_assign_jit_stack(pcre_extra *,
+                  pcre_jit_callback, void *);
+PCRE_EXP_DECL void pcre16_assign_jit_stack(pcre16_extra *,
+                  pcre16_jit_callback, void *);
+PCRE_EXP_DECL void pcre32_assign_jit_stack(pcre32_extra *,
+                  pcre32_jit_callback, void *);
+
+#ifdef __cplusplus
+}  /* extern "C" */
+#endif
+
+#endif /* End of pcre.h */
diff --git a/pcre/pcre16_byte_order.c b/pcre/pcre16_byte_order.c
new file mode 100644
index 00000000000..11d2973a3db
--- /dev/null
+++ b/pcre/pcre16_byte_order.c
@@ -0,0 +1,45 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+/* Generate code with 16 bit character support. */
+#define COMPILE_PCRE16
+
+#include "pcre_byte_order.c"
+
+/* End of pcre16_byte_order.c */
diff --git a/pcre/pcre16_chartables.c b/pcre/pcre16_chartables.c
new file mode 100644
index 00000000000..7c0ff35f5e0
--- /dev/null
+++ b/pcre/pcre16_chartables.c
@@ -0,0 +1,45 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+/* Generate code with 16 bit character support. */
+#define COMPILE_PCRE16
+
+#include "pcre_chartables.c"
+
+/* End of pcre16_chartables.c */
diff --git a/pcre/pcre16_compile.c b/pcre/pcre16_compile.c
new file mode 100644
index 00000000000..e499b670877
--- /dev/null
+++ b/pcre/pcre16_compile.c
@@ -0,0 +1,45 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+/* Generate code with 16 bit character support. */
+#define COMPILE_PCRE16
+
+#include "pcre_compile.c"
+
+/* End of pcre16_compile.c */
diff --git a/pcre/pcre16_config.c b/pcre/pcre16_config.c
new file mode 100644
index 00000000000..b52138764f6
--- /dev/null
+++ b/pcre/pcre16_config.c
@@ -0,0 +1,45 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+/* Generate code with 16 bit character support. */
+#define COMPILE_PCRE16
+
+#include "pcre_config.c"
+
+/* End of pcre16_config.c */
diff --git a/pcre/pcre16_dfa_exec.c b/pcre/pcre16_dfa_exec.c
new file mode 100644
index 00000000000..2ba740e972b
--- /dev/null
+++ b/pcre/pcre16_dfa_exec.c
@@ -0,0 +1,45 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+/* Generate code with 16 bit character support. */
+#define COMPILE_PCRE16
+
+#include "pcre_dfa_exec.c"
+
+/* End of pcre16_dfa_exec.c */
diff --git a/pcre/pcre16_exec.c b/pcre/pcre16_exec.c
new file mode 100644
index 00000000000..7417b1770c6
--- /dev/null
+++ b/pcre/pcre16_exec.c
@@ -0,0 +1,45 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+/* Generate code with 16 bit character support. */
+#define COMPILE_PCRE16
+
+#include "pcre_exec.c"
+
+/* End of pcre16_exec.c */
diff --git a/pcre/pcre16_fullinfo.c b/pcre/pcre16_fullinfo.c
new file mode 100644
index 00000000000..544dca6ed5c
--- /dev/null
+++ b/pcre/pcre16_fullinfo.c
@@ -0,0 +1,45 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+/* Generate code with 16 bit character support. */
+#define COMPILE_PCRE16
+
+#include "pcre_fullinfo.c"
+
+/* End of pcre16_fullinfo.c */
diff --git a/pcre/pcre16_get.c b/pcre/pcre16_get.c
new file mode 100644
index 00000000000..3ded08c622c
--- /dev/null
+++ b/pcre/pcre16_get.c
@@ -0,0 +1,45 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+/* Generate code with 16 bit character support. */
+#define COMPILE_PCRE16
+
+#include "pcre_get.c"
+
+/* End of pcre16_get.c */
diff --git a/pcre/pcre16_globals.c b/pcre/pcre16_globals.c
new file mode 100644
index 00000000000..a136b3d8c22
--- /dev/null
+++ b/pcre/pcre16_globals.c
@@ -0,0 +1,45 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+/* Generate code with 16 bit character support. */
+#define COMPILE_PCRE16
+
+#include "pcre_globals.c"
+
+/* End of pcre16_globals.c */
diff --git a/pcre/pcre16_jit_compile.c b/pcre/pcre16_jit_compile.c
new file mode 100644
index 00000000000..ab0cacd7646
--- /dev/null
+++ b/pcre/pcre16_jit_compile.c
@@ -0,0 +1,45 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+/* Generate code with 16 bit character support. */
+#define COMPILE_PCRE16
+
+#include "pcre_jit_compile.c"
+
+/* End of pcre16_jit_compile.c */
diff --git a/pcre/pcre16_maketables.c b/pcre/pcre16_maketables.c
new file mode 100644
index 00000000000..b1cd1c579d6
--- /dev/null
+++ b/pcre/pcre16_maketables.c
@@ -0,0 +1,45 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+/* Generate code with 16 bit character support. */
+#define COMPILE_PCRE16
+
+#include "pcre_maketables.c"
+
+/* End of pcre16_maketables.c */
diff --git a/pcre/pcre16_newline.c b/pcre/pcre16_newline.c
new file mode 100644
index 00000000000..7fe201400f5
--- /dev/null
+++ b/pcre/pcre16_newline.c
@@ -0,0 +1,45 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+/* Generate code with 16 bit character support. */
+#define COMPILE_PCRE16
+
+#include "pcre_newline.c"
+
+/* End of pcre16_newline.c */
diff --git a/pcre/pcre16_ord2utf16.c b/pcre/pcre16_ord2utf16.c
new file mode 100644
index 00000000000..8e2ce5ea6c5
--- /dev/null
+++ b/pcre/pcre16_ord2utf16.c
@@ -0,0 +1,90 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+
+/* This file contains a private PCRE function that converts an ordinal
+character value into a UTF16 string. */
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+/* Generate code with 16 bit character support. */
+#define COMPILE_PCRE16
+
+#include "pcre_internal.h"
+
+/*************************************************
+*       Convert character value to UTF-16         *
+*************************************************/
+
+/* This function takes an integer value in the range 0 - 0x10ffff
+and encodes it as a UTF-16 character in 1 to 2 pcre_uchars.
+
+Arguments:
+  cvalue     the character value
+  buffer     pointer to buffer for result - at least 2 pcre_uchars long
+
+Returns:     number of characters placed in the buffer
+*/
+
+unsigned int
+PRIV(ord2utf)(pcre_uint32 cvalue, pcre_uchar *buffer)
+{
+#ifdef SUPPORT_UTF
+
+if (cvalue <= 0xffff)
+  {
+  *buffer = (pcre_uchar)cvalue;
+  return 1;
+  }
+
+cvalue -= 0x10000;
+*buffer++ = 0xd800 | (cvalue >> 10);
+*buffer = 0xdc00 | (cvalue & 0x3ff);
+return 2;
+
+#else /* SUPPORT_UTF */
+(void)(cvalue);  /* Keep compiler happy; this function won't ever be */
+(void)(buffer);  /* called when SUPPORT_UTF is not defined. */
+return 0;
+#endif /* SUPPORT_UTF */
+}
+
+/* End of pcre16_ord2utf16.c */
diff --git a/pcre/pcre16_printint.c b/pcre/pcre16_printint.c
new file mode 100644
index 00000000000..33d8c340200
--- /dev/null
+++ b/pcre/pcre16_printint.c
@@ -0,0 +1,45 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+/* Generate code with 16 bit character support. */
+#define COMPILE_PCRE16
+
+#include "pcre_printint.c"
+
+/* End of pcre16_printint.c */
diff --git a/pcre/pcre16_refcount.c b/pcre/pcre16_refcount.c
new file mode 100644
index 00000000000..d3d15439737
--- /dev/null
+++ b/pcre/pcre16_refcount.c
@@ -0,0 +1,45 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+/* Generate code with 16 bit character support. */
+#define COMPILE_PCRE16
+
+#include "pcre_refcount.c"
+
+/* End of pcre16_refcount.c */
diff --git a/pcre/pcre16_string_utils.c b/pcre/pcre16_string_utils.c
new file mode 100644
index 00000000000..382c40799fb
--- /dev/null
+++ b/pcre/pcre16_string_utils.c
@@ -0,0 +1,45 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+/* Generate code with 16 bit character support. */
+#define COMPILE_PCRE16
+
+#include "pcre_string_utils.c"
+
+/* End of pcre16_string_utils.c */
diff --git a/pcre/pcre16_study.c b/pcre/pcre16_study.c
new file mode 100644
index 00000000000..f87de081fc4
--- /dev/null
+++ b/pcre/pcre16_study.c
@@ -0,0 +1,45 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+/* Generate code with 16 bit character support. */
+#define COMPILE_PCRE16
+
+#include "pcre_study.c"
+
+/* End of pcre16_study.c */
diff --git a/pcre/pcre16_tables.c b/pcre/pcre16_tables.c
new file mode 100644
index 00000000000..d84297093a4
--- /dev/null
+++ b/pcre/pcre16_tables.c
@@ -0,0 +1,45 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+/* Generate code with 16 bit character support. */
+#define COMPILE_PCRE16
+
+#include "pcre_tables.c"
+
+/* End of pcre16_tables.c */
diff --git a/pcre/pcre16_ucd.c b/pcre/pcre16_ucd.c
new file mode 100644
index 00000000000..ee23439a013
--- /dev/null
+++ b/pcre/pcre16_ucd.c
@@ -0,0 +1,45 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+/* Generate code with 16 bit character support. */
+#define COMPILE_PCRE16
+
+#include "pcre_ucd.c"
+
+/* End of pcre16_ucd.c */
diff --git a/pcre/pcre16_utf16_utils.c b/pcre/pcre16_utf16_utils.c
new file mode 100644
index 00000000000..49ced0c0b1c
--- /dev/null
+++ b/pcre/pcre16_utf16_utils.c
@@ -0,0 +1,130 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+
+/* This module contains a function for converting any UTF-16 character
+strings to host byte order. */
+
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+/* Generate code with 16 bit character support. */
+#define COMPILE_PCRE16
+
+#include "pcre_internal.h"
+
+/*************************************************
+*  Convert any UTF-16 string to host byte order  *
+*************************************************/
+
+/* This function takes an UTF-16 string and converts
+it to host byte order. The length can be explicitly set,
+or automatically detected for zero terminated strings.
+BOMs can be kept or discarded during the conversion.
+Conversion can be done in place (output == input).
+
+Arguments:
+  output     the output buffer, its size must be greater
+             or equal than the input string
+  input      any UTF-16 string
+  length     the number of 16-bit units in the input string
+             can be less than zero for zero terminated strings
+  host_byte_order
+             A non-zero value means the input is in host byte
+             order, which can be dynamically changed by BOMs later.
+             Initially it contains the starting byte order and returns
+             with the last byte order so it can be used for stream
+             processing. It can be NULL, which set the host byte
+             order mode by default.
+  keep_boms  for a non-zero value, the BOM (0xfeff) characters
+             are copied as well
+
+Returns:     the number of 16-bit units placed into the output buffer,
+             including the zero-terminator
+*/
+
+int
+pcre16_utf16_to_host_byte_order(PCRE_UCHAR16 *output, PCRE_SPTR16 input,
+  int length, int *host_byte_order, int keep_boms)
+{
+#ifdef SUPPORT_UTF
+/* This function converts any UTF-16 string to host byte order and optionally
+removes any Byte Order Marks (BOMS). Returns with the remainig length. */
+int host_bo = host_byte_order != NULL ? *host_byte_order : 1;
+pcre_uchar *optr = (pcre_uchar *)output;
+const pcre_uchar *iptr = (const pcre_uchar *)input;
+const pcre_uchar *end;
+/* The c variable must be unsigned. */
+register pcre_uchar c;
+
+if (length < 0)
+  length = STRLEN_UC(iptr) + 1;
+end = iptr + length;
+
+while (iptr < end)
+  {
+  c = *iptr++;
+  if (c == 0xfeff || c == 0xfffe)
+    {
+    /* Detecting the byte order of the machine is unnecessary, it is
+    enough to know that the UTF-16 string has the same byte order or not. */
+    host_bo = c == 0xfeff;
+    if (keep_boms != 0)
+      *optr++ = 0xfeff;
+    else
+      length--;
+    }
+  else
+    *optr++ = host_bo ? c : ((c >> 8) | (c << 8)); /* Flip bytes if needed. */
+  }
+if (host_byte_order != NULL)
+  *host_byte_order = host_bo;
+
+#else /* Not SUPPORT_UTF */
+(void)(output);  /* Keep picky compilers happy */
+(void)(input);
+(void)(keep_boms);
+(void)(host_byte_order);
+#endif /* SUPPORT_UTF */
+return length;
+}
+
+/* End of pcre16_utf16_utils.c */
diff --git a/pcre/pcre16_valid_utf16.c b/pcre/pcre16_valid_utf16.c
new file mode 100644
index 00000000000..09076539d09
--- /dev/null
+++ b/pcre/pcre16_valid_utf16.c
@@ -0,0 +1,137 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2013 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+
+/* This module contains an internal function for validating UTF-16 character
+strings. */
+
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+/* Generate code with 16 bit character support. */
+#define COMPILE_PCRE16
+
+#include "pcre_internal.h"
+
+
+/*************************************************
+*         Validate a UTF-16 string                *
+*************************************************/
+
+/* This function is called (optionally) at the start of compile or match, to
+check that a supposed UTF-16 string is actually valid. The early check means
+that subsequent code can assume it is dealing with a valid string. The check
+can be turned off for maximum performance, but the consequences of supplying an
+invalid string are then undefined.
+
+From release 8.21 more information about the details of the error are passed
+back in the returned value:
+
+PCRE_UTF16_ERR0  No error
+PCRE_UTF16_ERR1  Missing low surrogate at the end of the string
+PCRE_UTF16_ERR2  Invalid low surrogate
+PCRE_UTF16_ERR3  Isolated low surrogate
+PCRE_UTF16_ERR4  Unused (was non-character)
+
+Arguments:
+  string       points to the string
+  length       length of string, or -1 if the string is zero-terminated
+  errp         pointer to an error position offset variable
+
+Returns:       = 0    if the string is a valid UTF-16 string
+               > 0    otherwise, setting the offset of the bad character
+*/
+
+int
+PRIV(valid_utf)(PCRE_PUCHAR string, int length, int *erroroffset)
+{
+#ifdef SUPPORT_UTF
+register PCRE_PUCHAR p;
+register pcre_uint32 c;
+
+if (length < 0)
+  {
+  for (p = string; *p != 0; p++);
+  length = p - string;
+  }
+
+for (p = string; length-- > 0; p++)
+  {
+  c = *p;
+
+  if ((c & 0xf800) != 0xd800)
+    {
+    /* Normal UTF-16 code point. Neither high nor low surrogate. */
+    }
+  else if ((c & 0x0400) == 0)
+    {
+    /* High surrogate. Must be a followed by a low surrogate. */
+    if (length == 0)
+      {
+      *erroroffset = p - string;
+      return PCRE_UTF16_ERR1;
+      }
+    p++;
+    length--;
+    if ((*p & 0xfc00) != 0xdc00)
+      {
+      *erroroffset = p - string;
+      return PCRE_UTF16_ERR2;
+      }
+    }
+  else
+    {
+    /* Isolated low surrogate. Always an error. */
+    *erroroffset = p - string;
+    return PCRE_UTF16_ERR3;
+    }
+  }
+
+#else  /* SUPPORT_UTF */
+(void)(string);  /* Keep picky compilers happy */
+(void)(length);
+(void)(erroroffset);
+#endif /* SUPPORT_UTF */
+
+return PCRE_UTF16_ERR0;   /* This indicates success */
+}
+
+/* End of pcre16_valid_utf16.c */
diff --git a/pcre/pcre16_version.c b/pcre/pcre16_version.c
new file mode 100644
index 00000000000..e991b1a8cfd
--- /dev/null
+++ b/pcre/pcre16_version.c
@@ -0,0 +1,45 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+/* Generate code with 16 bit character support. */
+#define COMPILE_PCRE16
+
+#include "pcre_version.c"
+
+/* End of pcre16_version.c */
diff --git a/pcre/pcre16_xclass.c b/pcre/pcre16_xclass.c
new file mode 100644
index 00000000000..5aac2a36c68
--- /dev/null
+++ b/pcre/pcre16_xclass.c
@@ -0,0 +1,45 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+/* Generate code with 16 bit character support. */
+#define COMPILE_PCRE16
+
+#include "pcre_xclass.c"
+
+/* End of pcre16_xclass.c */
diff --git a/pcre/pcre32_byte_order.c b/pcre/pcre32_byte_order.c
new file mode 100644
index 00000000000..9cf5362730a
--- /dev/null
+++ b/pcre/pcre32_byte_order.c
@@ -0,0 +1,45 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+/* Generate code with 32 bit character support. */
+#define COMPILE_PCRE32
+
+#include "pcre_byte_order.c"
+
+/* End of pcre32_byte_order.c */
diff --git a/pcre/pcre32_chartables.c b/pcre/pcre32_chartables.c
new file mode 100644
index 00000000000..b5d8c23dbf1
--- /dev/null
+++ b/pcre/pcre32_chartables.c
@@ -0,0 +1,45 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+/* Generate code with 32 bit character support. */
+#define COMPILE_PCRE32
+
+#include "pcre_chartables.c"
+
+/* End of pcre32_chartables.c */
diff --git a/pcre/pcre32_compile.c b/pcre/pcre32_compile.c
new file mode 100644
index 00000000000..d781eb377e0
--- /dev/null
+++ b/pcre/pcre32_compile.c
@@ -0,0 +1,45 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+/* Generate code with 32 bit character support. */
+#define COMPILE_PCRE32
+
+#include "pcre_compile.c"
+
+/* End of pcre32_compile.c */
diff --git a/pcre/pcre32_config.c b/pcre/pcre32_config.c
new file mode 100644
index 00000000000..d63f3e9ea23
--- /dev/null
+++ b/pcre/pcre32_config.c
@@ -0,0 +1,45 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+/* Generate code with 32 bit character support. */
+#define COMPILE_PCRE32
+
+#include "pcre_config.c"
+
+/* End of pcre32_config.c */
diff --git a/pcre/pcre32_dfa_exec.c b/pcre/pcre32_dfa_exec.c
new file mode 100644
index 00000000000..b0bfd34f04d
--- /dev/null
+++ b/pcre/pcre32_dfa_exec.c
@@ -0,0 +1,45 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+/* Generate code with 32 bit character support. */
+#define COMPILE_PCRE32
+
+#include "pcre_dfa_exec.c"
+
+/* End of pcre32_dfa_exec.c */
diff --git a/pcre/pcre32_exec.c b/pcre/pcre32_exec.c
new file mode 100644
index 00000000000..8170ed77d35
--- /dev/null
+++ b/pcre/pcre32_exec.c
@@ -0,0 +1,45 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+/* Generate code with 32 bit character support. */
+#define COMPILE_PCRE32
+
+#include "pcre_exec.c"
+
+/* End of pcre32_exec.c */
diff --git a/pcre/pcre32_fullinfo.c b/pcre/pcre32_fullinfo.c
new file mode 100644
index 00000000000..6ecc5209a08
--- /dev/null
+++ b/pcre/pcre32_fullinfo.c
@@ -0,0 +1,45 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+/* Generate code with 32 bit character support. */
+#define COMPILE_PCRE32
+
+#include "pcre_fullinfo.c"
+
+/* End of pcre32_fullinfo.c */
diff --git a/pcre/pcre32_get.c b/pcre/pcre32_get.c
new file mode 100644
index 00000000000..d35deee0cd4
--- /dev/null
+++ b/pcre/pcre32_get.c
@@ -0,0 +1,45 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+/* Generate code with 32 bit character support. */
+#define COMPILE_PCRE32
+
+#include "pcre_get.c"
+
+/* End of pcre32_get.c */
diff --git a/pcre/pcre32_globals.c b/pcre/pcre32_globals.c
new file mode 100644
index 00000000000..32e0914ca6d
--- /dev/null
+++ b/pcre/pcre32_globals.c
@@ -0,0 +1,45 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+/* Generate code with 32 bit character support. */
+#define COMPILE_PCRE32
+
+#include "pcre_globals.c"
+
+/* End of pcre32_globals.c */
diff --git a/pcre/pcre32_jit_compile.c b/pcre/pcre32_jit_compile.c
new file mode 100644
index 00000000000..2e7c6f97c96
--- /dev/null
+++ b/pcre/pcre32_jit_compile.c
@@ -0,0 +1,45 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+/* Generate code with 32 bit character support. */
+#define COMPILE_PCRE32
+
+#include "pcre_jit_compile.c"
+
+/* End of pcre32_jit_compile.c */
diff --git a/pcre/pcre32_maketables.c b/pcre/pcre32_maketables.c
new file mode 100644
index 00000000000..5d1b1c64c96
--- /dev/null
+++ b/pcre/pcre32_maketables.c
@@ -0,0 +1,45 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+/* Generate code with 32 bit character support. */
+#define COMPILE_PCRE32
+
+#include "pcre_maketables.c"
+
+/* End of pcre32_maketables.c */
diff --git a/pcre/pcre32_newline.c b/pcre/pcre32_newline.c
new file mode 100644
index 00000000000..7f8d5360cdc
--- /dev/null
+++ b/pcre/pcre32_newline.c
@@ -0,0 +1,45 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+/* Generate code with 32 bit character support. */
+#define COMPILE_PCRE32
+
+#include "pcre_newline.c"
+
+/* End of pcre32_newline.c */
diff --git a/pcre/pcre32_ord2utf32.c b/pcre/pcre32_ord2utf32.c
new file mode 100644
index 00000000000..606bcb3d7ba
--- /dev/null
+++ b/pcre/pcre32_ord2utf32.c
@@ -0,0 +1,82 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+
+/* This file contains a private PCRE function that converts an ordinal
+character value into a UTF32 string. */
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+/* Generate code with 32 bit character support. */
+#define COMPILE_PCRE32
+
+#include "pcre_internal.h"
+
+/*************************************************
+*       Convert character value to UTF-32         *
+*************************************************/
+
+/* This function takes an integer value in the range 0 - 0x10ffff
+and encodes it as a UTF-32 character in 1 pcre_uchars.
+
+Arguments:
+  cvalue     the character value
+  buffer     pointer to buffer for result - at least 1 pcre_uchars long
+
+Returns:     number of characters placed in the buffer
+*/
+
+unsigned int
+PRIV(ord2utf)(pcre_uint32 cvalue, pcre_uchar *buffer)
+{
+#ifdef SUPPORT_UTF
+
+*buffer = (pcre_uchar)cvalue;
+return 1;
+
+#else /* SUPPORT_UTF */
+(void)(cvalue);  /* Keep compiler happy; this function won't ever be */
+(void)(buffer);  /* called when SUPPORT_UTF is not defined. */
+return 0;
+#endif /* SUPPORT_UTF */
+}
+
+/* End of pcre32_ord2utf32.c */
diff --git a/pcre/pcre32_printint.c b/pcre/pcre32_printint.c
new file mode 100644
index 00000000000..f3fd7b25e2c
--- /dev/null
+++ b/pcre/pcre32_printint.c
@@ -0,0 +1,45 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+/* Generate code with 32 bit character support. */
+#define COMPILE_PCRE32
+
+#include "pcre_printint.c"
+
+/* End of pcre32_printint.c */
diff --git a/pcre/pcre32_refcount.c b/pcre/pcre32_refcount.c
new file mode 100644
index 00000000000..dbdf432d82a
--- /dev/null
+++ b/pcre/pcre32_refcount.c
@@ -0,0 +1,45 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+/* Generate code with 32 bit character support. */
+#define COMPILE_PCRE32
+
+#include "pcre_refcount.c"
+
+/* End of pcre32_refcount.c */
diff --git a/pcre/pcre32_string_utils.c b/pcre/pcre32_string_utils.c
new file mode 100644
index 00000000000..e37b3d4805f
--- /dev/null
+++ b/pcre/pcre32_string_utils.c
@@ -0,0 +1,45 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+/* Generate code with 32 bit character support. */
+#define COMPILE_PCRE32
+
+#include "pcre_string_utils.c"
+
+/* End of pcre32_string_utils.c */
diff --git a/pcre/pcre32_study.c b/pcre/pcre32_study.c
new file mode 100644
index 00000000000..d3a3afed791
--- /dev/null
+++ b/pcre/pcre32_study.c
@@ -0,0 +1,45 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+/* Generate code with 32 bit character support. */
+#define COMPILE_PCRE32
+
+#include "pcre_study.c"
+
+/* End of pcre32_study.c */
diff --git a/pcre/pcre32_tables.c b/pcre/pcre32_tables.c
new file mode 100644
index 00000000000..3d94cca33a1
--- /dev/null
+++ b/pcre/pcre32_tables.c
@@ -0,0 +1,45 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+/* Generate code with 32 bit character support. */
+#define COMPILE_PCRE32
+
+#include "pcre_tables.c"
+
+/* End of pcre32_tables.c */
diff --git a/pcre/pcre32_ucd.c b/pcre/pcre32_ucd.c
new file mode 100644
index 00000000000..befe22d3435
--- /dev/null
+++ b/pcre/pcre32_ucd.c
@@ -0,0 +1,45 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+/* Generate code with 32 bit character support. */
+#define COMPILE_PCRE32
+
+#include "pcre_ucd.c"
+
+/* End of pcre32_ucd.c */
diff --git a/pcre/pcre32_utf32_utils.c b/pcre/pcre32_utf32_utils.c
new file mode 100644
index 00000000000..f844e237165
--- /dev/null
+++ b/pcre/pcre32_utf32_utils.c
@@ -0,0 +1,141 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+
+/* This module contains a function for converting any UTF-32 character
+strings to host byte order. */
+
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+/* Generate code with 32 bit character support. */
+#define COMPILE_PCRE32
+
+#include "pcre_internal.h"
+
+#ifdef SUPPORT_UTF
+static pcre_uint32
+swap_uint32(pcre_uint32 value)
+{
+return ((value & 0x000000ff) << 24) |
+       ((value & 0x0000ff00) <<  8) |
+       ((value & 0x00ff0000) >>  8) |
+       (value >> 24);
+}
+#endif
+
+
+/*************************************************
+*  Convert any UTF-32 string to host byte order  *
+*************************************************/
+
+/* This function takes an UTF-32 string and converts
+it to host byte order. The length can be explicitly set,
+or automatically detected for zero terminated strings.
+BOMs can be kept or discarded during the conversion.
+Conversion can be done in place (output == input).
+
+Arguments:
+  output     the output buffer, its size must be greater
+             or equal than the input string
+  input      any UTF-32 string
+  length     the number of 32-bit units in the input string
+             can be less than zero for zero terminated strings
+  host_byte_order
+             A non-zero value means the input is in host byte
+             order, which can be dynamically changed by BOMs later.
+             Initially it contains the starting byte order and returns
+             with the last byte order so it can be used for stream
+             processing. It can be NULL, which set the host byte
+             order mode by default.
+  keep_boms  for a non-zero value, the BOM (0xfeff) characters
+             are copied as well
+
+Returns:     the number of 32-bit units placed into the output buffer,
+             including the zero-terminator
+*/
+
+int
+pcre32_utf32_to_host_byte_order(PCRE_UCHAR32 *output, PCRE_SPTR32 input,
+  int length, int *host_byte_order, int keep_boms)
+{
+#ifdef SUPPORT_UTF
+/* This function converts any UTF-32 string to host byte order and optionally
+removes any Byte Order Marks (BOMS). Returns with the remainig length. */
+int host_bo = host_byte_order != NULL ? *host_byte_order : 1;
+pcre_uchar *optr = (pcre_uchar *)output;
+const pcre_uchar *iptr = (const pcre_uchar *)input;
+const pcre_uchar *end;
+/* The c variable must be unsigned. */
+register pcre_uchar c;
+
+if (length < 0)
+  end = iptr + STRLEN_UC(iptr) + 1;
+else
+  end = iptr + length;
+
+while (iptr < end)
+  {
+  c = *iptr++;
+  if (c == 0x0000feffu || c == 0xfffe0000u)
+    {
+    /* Detecting the byte order of the machine is unnecessary, it is
+    enough to know that the UTF-32 string has the same byte order or not. */
+    host_bo = c == 0x0000feffu;
+    if (keep_boms != 0)
+      *optr++ = 0x0000feffu;
+    }
+  else
+    *optr++ = host_bo ? c : swap_uint32(c);
+  }
+if (host_byte_order != NULL)
+  *host_byte_order = host_bo;
+
+#else /* SUPPORT_UTF */
+(void)(output);  /* Keep picky compilers happy */
+(void)(input);
+(void)(keep_boms);
+(void)(host_byte_order);
+#endif /* SUPPORT_UTF */
+return length;
+}
+
+/* End of pcre32_utf32_utils.c */
diff --git a/pcre/pcre32_valid_utf32.c b/pcre/pcre32_valid_utf32.c
new file mode 100644
index 00000000000..94cda1a2c4c
--- /dev/null
+++ b/pcre/pcre32_valid_utf32.c
@@ -0,0 +1,124 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2013 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+
+/* This module contains an internal function for validating UTF-32 character
+strings. */
+
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+/* Generate code with 32 bit character support. */
+#define COMPILE_PCRE32
+
+#include "pcre_internal.h"
+
+/*************************************************
+*         Validate a UTF-32 string                *
+*************************************************/
+
+/* This function is called (optionally) at the start of compile or match, to
+check that a supposed UTF-32 string is actually valid. The early check means
+that subsequent code can assume it is dealing with a valid string. The check
+can be turned off for maximum performance, but the consequences of supplying an
+invalid string are then undefined.
+
+More information about the details of the error are passed
+back in the returned value:
+
+PCRE_UTF32_ERR0  No error
+PCRE_UTF32_ERR1  Surrogate character
+PCRE_UTF32_ERR2  Unused (was non-character)
+PCRE_UTF32_ERR3  Character > 0x10ffff
+
+Arguments:
+  string       points to the string
+  length       length of string, or -1 if the string is zero-terminated
+  errp         pointer to an error position offset variable
+
+Returns:       = 0    if the string is a valid UTF-32 string
+               > 0    otherwise, setting the offset of the bad character
+*/
+
+int
+PRIV(valid_utf)(PCRE_PUCHAR string, int length, int *erroroffset)
+{
+#ifdef SUPPORT_UTF
+register PCRE_PUCHAR p;
+register pcre_uchar c;
+
+if (length < 0)
+  {
+  for (p = string; *p != 0; p++);
+  length = p - string;
+  }
+
+for (p = string; length-- > 0; p++)
+  {
+  c = *p;
+
+  if ((c & 0xfffff800u) != 0xd800u)
+    {
+    /* Normal UTF-32 code point. Neither high nor low surrogate. */
+    if (c > 0x10ffffu)
+      {
+      *erroroffset = p - string;
+      return PCRE_UTF32_ERR3;
+      }
+    }
+  else
+    {
+    /* A surrogate */
+    *erroroffset = p - string;
+    return PCRE_UTF32_ERR1;
+    }
+  }
+
+#else  /* SUPPORT_UTF */
+(void)(string);  /* Keep picky compilers happy */
+(void)(length);
+(void)(erroroffset);
+#endif /* SUPPORT_UTF */
+
+return PCRE_UTF32_ERR0;   /* This indicates success */
+}
+
+/* End of pcre32_valid_utf32.c */
diff --git a/pcre/pcre32_version.c b/pcre/pcre32_version.c
new file mode 100644
index 00000000000..fdaad9b0859
--- /dev/null
+++ b/pcre/pcre32_version.c
@@ -0,0 +1,45 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+/* Generate code with 32 bit character support. */
+#define COMPILE_PCRE32
+
+#include "pcre_version.c"
+
+/* End of pcre32_version.c */
diff --git a/pcre/pcre32_xclass.c b/pcre/pcre32_xclass.c
new file mode 100644
index 00000000000..5662408ad5f
--- /dev/null
+++ b/pcre/pcre32_xclass.c
@@ -0,0 +1,45 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+/* Generate code with 32 bit character support. */
+#define COMPILE_PCRE32
+
+#include "pcre_xclass.c"
+
+/* End of pcre32_xclass.c */
diff --git a/pcre/pcre_byte_order.c b/pcre/pcre_byte_order.c
new file mode 100644
index 00000000000..01cbca36ffe
--- /dev/null
+++ b/pcre/pcre_byte_order.c
@@ -0,0 +1,319 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2013 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+
+/* This module contains an internal function that tests a compiled pattern to
+see if it was compiled with the opposite endianness. If so, it uses an
+auxiliary local function to flip the appropriate bytes. */
+
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include "pcre_internal.h"
+
+
+/*************************************************
+*             Swap byte functions                *
+*************************************************/
+
+/* The following functions swap the bytes of a pcre_uint16
+and pcre_uint32 value.
+
+Arguments:
+  value        any number
+
+Returns:       the byte swapped value
+*/
+
+static pcre_uint32
+swap_uint32(pcre_uint32 value)
+{
+return ((value & 0x000000ff) << 24) |
+       ((value & 0x0000ff00) <<  8) |
+       ((value & 0x00ff0000) >>  8) |
+       (value >> 24);
+}
+
+static pcre_uint16
+swap_uint16(pcre_uint16 value)
+{
+return (value >> 8) | (value << 8);
+}
+
+
+/*************************************************
+*       Test for a byte-flipped compiled regex   *
+*************************************************/
+
+/* This function swaps the bytes of a compiled pattern usually
+loaded form the disk. It also sets the tables pointer, which
+is likely an invalid pointer after reload.
+
+Arguments:
+  argument_re     points to the compiled expression
+  extra_data      points to extra data or is NULL
+  tables          points to the character tables or NULL
+
+Returns:          0 if the swap is successful, negative on error
+*/
+
+#if defined COMPILE_PCRE8
+PCRE_EXP_DECL int pcre_pattern_to_host_byte_order(pcre *argument_re,
+  pcre_extra *extra_data, const unsigned char *tables)
+#elif defined COMPILE_PCRE16
+PCRE_EXP_DECL int pcre16_pattern_to_host_byte_order(pcre16 *argument_re,
+  pcre16_extra *extra_data, const unsigned char *tables)
+#elif defined COMPILE_PCRE32
+PCRE_EXP_DECL int pcre32_pattern_to_host_byte_order(pcre32 *argument_re,
+  pcre32_extra *extra_data, const unsigned char *tables)
+#endif
+{
+REAL_PCRE *re = (REAL_PCRE *)argument_re;
+pcre_study_data *study;
+#ifndef COMPILE_PCRE8
+pcre_uchar *ptr;
+int length;
+#if defined SUPPORT_UTF && defined COMPILE_PCRE16
+BOOL utf;
+BOOL utf16_char;
+#endif /* SUPPORT_UTF && COMPILE_PCRE16 */
+#endif /* !COMPILE_PCRE8 */
+
+if (re == NULL) return PCRE_ERROR_NULL;
+if (re->magic_number == MAGIC_NUMBER)
+  {
+  if ((re->flags & PCRE_MODE) == 0) return PCRE_ERROR_BADMODE;
+  re->tables = tables;
+  return 0;
+  }
+
+if (re->magic_number != REVERSED_MAGIC_NUMBER) return PCRE_ERROR_BADMAGIC;
+if ((swap_uint32(re->flags) & PCRE_MODE) == 0) return PCRE_ERROR_BADMODE;
+
+re->magic_number = MAGIC_NUMBER;
+re->size = swap_uint32(re->size);
+re->options = swap_uint32(re->options);
+re->flags = swap_uint32(re->flags);
+re->limit_match = swap_uint32(re->limit_match);
+re->limit_recursion = swap_uint32(re->limit_recursion);
+
+#if defined COMPILE_PCRE8 || defined COMPILE_PCRE16
+re->first_char = swap_uint16(re->first_char);
+re->req_char = swap_uint16(re->req_char);
+#elif defined COMPILE_PCRE32
+re->first_char = swap_uint32(re->first_char);
+re->req_char = swap_uint32(re->req_char);
+#endif
+
+re->max_lookbehind = swap_uint16(re->max_lookbehind);
+re->top_bracket = swap_uint16(re->top_bracket);
+re->top_backref = swap_uint16(re->top_backref);
+re->name_table_offset = swap_uint16(re->name_table_offset);
+re->name_entry_size = swap_uint16(re->name_entry_size);
+re->name_count = swap_uint16(re->name_count);
+re->ref_count = swap_uint16(re->ref_count);
+re->tables = tables;
+
+if (extra_data != NULL && (extra_data->flags & PCRE_EXTRA_STUDY_DATA) != 0)
+  {
+  study = (pcre_study_data *)extra_data->study_data;
+  study->size = swap_uint32(study->size);
+  study->flags = swap_uint32(study->flags);
+  study->minlength = swap_uint32(study->minlength);
+  }
+
+#ifndef COMPILE_PCRE8
+ptr = (pcre_uchar *)re + re->name_table_offset;
+length = re->name_count * re->name_entry_size;
+#if defined SUPPORT_UTF && defined COMPILE_PCRE16
+utf = (re->options & PCRE_UTF16) != 0;
+utf16_char = FALSE;
+#endif /* SUPPORT_UTF && COMPILE_PCRE16 */
+
+while(TRUE)
+  {
+  /* Swap previous characters. */
+  while (length-- > 0)
+    {
+#if defined COMPILE_PCRE16
+    *ptr = swap_uint16(*ptr);
+#elif defined COMPILE_PCRE32
+    *ptr = swap_uint32(*ptr);
+#endif
+    ptr++;
+    }
+#if defined SUPPORT_UTF && defined COMPILE_PCRE16
+  if (utf16_char)
+    {
+    if (HAS_EXTRALEN(ptr[-1]))
+      {
+      /* We know that there is only one extra character in UTF-16. */
+      *ptr = swap_uint16(*ptr);
+      ptr++;
+      }
+    }
+  utf16_char = FALSE;
+#endif /* SUPPORT_UTF */
+
+  /* Get next opcode. */
+  length = 0;
+#if defined COMPILE_PCRE16
+  *ptr = swap_uint16(*ptr);
+#elif defined COMPILE_PCRE32
+  *ptr = swap_uint32(*ptr);
+#endif
+  switch (*ptr)
+    {
+    case OP_END:
+    return 0;
+
+#if defined SUPPORT_UTF && defined COMPILE_PCRE16
+    case OP_CHAR:
+    case OP_CHARI:
+    case OP_NOT:
+    case OP_NOTI:
+    case OP_STAR:
+    case OP_MINSTAR:
+    case OP_PLUS:
+    case OP_MINPLUS:
+    case OP_QUERY:
+    case OP_MINQUERY:
+    case OP_UPTO:
+    case OP_MINUPTO:
+    case OP_EXACT:
+    case OP_POSSTAR:
+    case OP_POSPLUS:
+    case OP_POSQUERY:
+    case OP_POSUPTO:
+    case OP_STARI:
+    case OP_MINSTARI:
+    case OP_PLUSI:
+    case OP_MINPLUSI:
+    case OP_QUERYI:
+    case OP_MINQUERYI:
+    case OP_UPTOI:
+    case OP_MINUPTOI:
+    case OP_EXACTI:
+    case OP_POSSTARI:
+    case OP_POSPLUSI:
+    case OP_POSQUERYI:
+    case OP_POSUPTOI:
+    case OP_NOTSTAR:
+    case OP_NOTMINSTAR:
+    case OP_NOTPLUS:
+    case OP_NOTMINPLUS:
+    case OP_NOTQUERY:
+    case OP_NOTMINQUERY:
+    case OP_NOTUPTO:
+    case OP_NOTMINUPTO:
+    case OP_NOTEXACT:
+    case OP_NOTPOSSTAR:
+    case OP_NOTPOSPLUS:
+    case OP_NOTPOSQUERY:
+    case OP_NOTPOSUPTO:
+    case OP_NOTSTARI:
+    case OP_NOTMINSTARI:
+    case OP_NOTPLUSI:
+    case OP_NOTMINPLUSI:
+    case OP_NOTQUERYI:
+    case OP_NOTMINQUERYI:
+    case OP_NOTUPTOI:
+    case OP_NOTMINUPTOI:
+    case OP_NOTEXACTI:
+    case OP_NOTPOSSTARI:
+    case OP_NOTPOSPLUSI:
+    case OP_NOTPOSQUERYI:
+    case OP_NOTPOSUPTOI:
+    if (utf) utf16_char = TRUE;
+#endif
+    /* Fall through. */
+
+    default:
+    length = PRIV(OP_lengths)[*ptr] - 1;
+    break;
+
+    case OP_CLASS:
+    case OP_NCLASS:
+    /* Skip the character bit map. */
+    ptr += 32/sizeof(pcre_uchar);
+    length = 0;
+    break;
+
+    case OP_XCLASS:
+    /* Reverse the size of the XCLASS instance. */
+    ptr++;
+#if defined COMPILE_PCRE16
+    *ptr = swap_uint16(*ptr);
+#elif defined COMPILE_PCRE32
+    *ptr = swap_uint32(*ptr);
+#endif
+#ifndef COMPILE_PCRE32
+    if (LINK_SIZE > 1)
+      {
+      /* LINK_SIZE can be 1 or 2 in 16 bit mode. */
+      ptr++;
+      *ptr = swap_uint16(*ptr);
+      }
+#endif
+    ptr++;
+    length = (GET(ptr, -LINK_SIZE)) - (1 + LINK_SIZE + 1);
+#if defined COMPILE_PCRE16
+    *ptr = swap_uint16(*ptr);
+#elif defined COMPILE_PCRE32
+    *ptr = swap_uint32(*ptr);
+#endif
+    if ((*ptr & XCL_MAP) != 0)
+      {
+      /* Skip the character bit map. */
+      ptr += 32/sizeof(pcre_uchar);
+      length -= 32/sizeof(pcre_uchar);
+      }
+    break;
+    }
+  ptr++;
+  }
+/* Control should never reach here in 16/32 bit mode. */
+#endif /* !COMPILE_PCRE8 */
+
+return 0;
+}
+
+/* End of pcre_byte_order.c */
diff --git a/pcre/pcre_chartables.c b/pcre/pcre_chartables.c
new file mode 100644
index 00000000000..2a39e9ff33a
--- /dev/null
+++ b/pcre/pcre_chartables.c
@@ -0,0 +1,198 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* This file contains character tables that are used when no external tables
+are passed to PCRE by the application that calls it. The tables are used only
+for characters whose code values are less than 256.
+
+This is a default version of the tables that assumes ASCII encoding. A program
+called dftables (which is distributed with PCRE) can be used to build
+alternative versions of this file. This is necessary if you are running in an
+EBCDIC environment, or if you want to default to a different encoding, for
+example ISO-8859-1. When dftables is run, it creates these tables in the
+current locale. If PCRE is configured with --enable-rebuild-chartables, this
+happens automatically.
+
+The following #includes are present because without them gcc 4.x may remove the
+array definition from the final binary if PCRE is built into a static library
+and dead code stripping is activated. This leads to link errors. Pulling in the
+header ensures that the array gets flagged as "someone outside this compilation
+unit might reference this" and so it will always be supplied to the linker. */
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include "pcre_internal.h"
+
+const pcre_uint8 PRIV(default_tables)[] = {
+
+/* This table is a lower casing table. */
+
+    0,  1,  2,  3,  4,  5,  6,  7,
+    8,  9, 10, 11, 12, 13, 14, 15,
+   16, 17, 18, 19, 20, 21, 22, 23,
+   24, 25, 26, 27, 28, 29, 30, 31,
+   32, 33, 34, 35, 36, 37, 38, 39,
+   40, 41, 42, 43, 44, 45, 46, 47,
+   48, 49, 50, 51, 52, 53, 54, 55,
+   56, 57, 58, 59, 60, 61, 62, 63,
+   64, 97, 98, 99,100,101,102,103,
+  104,105,106,107,108,109,110,111,
+  112,113,114,115,116,117,118,119,
+  120,121,122, 91, 92, 93, 94, 95,
+   96, 97, 98, 99,100,101,102,103,
+  104,105,106,107,108,109,110,111,
+  112,113,114,115,116,117,118,119,
+  120,121,122,123,124,125,126,127,
+  128,129,130,131,132,133,134,135,
+  136,137,138,139,140,141,142,143,
+  144,145,146,147,148,149,150,151,
+  152,153,154,155,156,157,158,159,
+  160,161,162,163,164,165,166,167,
+  168,169,170,171,172,173,174,175,
+  176,177,178,179,180,181,182,183,
+  184,185,186,187,188,189,190,191,
+  192,193,194,195,196,197,198,199,
+  200,201,202,203,204,205,206,207,
+  208,209,210,211,212,213,214,215,
+  216,217,218,219,220,221,222,223,
+  224,225,226,227,228,229,230,231,
+  232,233,234,235,236,237,238,239,
+  240,241,242,243,244,245,246,247,
+  248,249,250,251,252,253,254,255,
+
+/* This table is a case flipping table. */
+
+    0,  1,  2,  3,  4,  5,  6,  7,
+    8,  9, 10, 11, 12, 13, 14, 15,
+   16, 17, 18, 19, 20, 21, 22, 23,
+   24, 25, 26, 27, 28, 29, 30, 31,
+   32, 33, 34, 35, 36, 37, 38, 39,
+   40, 41, 42, 43, 44, 45, 46, 47,
+   48, 49, 50, 51, 52, 53, 54, 55,
+   56, 57, 58, 59, 60, 61, 62, 63,
+   64, 97, 98, 99,100,101,102,103,
+  104,105,106,107,108,109,110,111,
+  112,113,114,115,116,117,118,119,
+  120,121,122, 91, 92, 93, 94, 95,
+   96, 65, 66, 67, 68, 69, 70, 71,
+   72, 73, 74, 75, 76, 77, 78, 79,
+   80, 81, 82, 83, 84, 85, 86, 87,
+   88, 89, 90,123,124,125,126,127,
+  128,129,130,131,132,133,134,135,
+  136,137,138,139,140,141,142,143,
+  144,145,146,147,148,149,150,151,
+  152,153,154,155,156,157,158,159,
+  160,161,162,163,164,165,166,167,
+  168,169,170,171,172,173,174,175,
+  176,177,178,179,180,181,182,183,
+  184,185,186,187,188,189,190,191,
+  192,193,194,195,196,197,198,199,
+  200,201,202,203,204,205,206,207,
+  208,209,210,211,212,213,214,215,
+  216,217,218,219,220,221,222,223,
+  224,225,226,227,228,229,230,231,
+  232,233,234,235,236,237,238,239,
+  240,241,242,243,244,245,246,247,
+  248,249,250,251,252,253,254,255,
+
+/* This table contains bit maps for various character classes. Each map is 32
+bytes long and the bits run from the least significant end of each byte. The
+classes that have their own maps are: space, xdigit, digit, upper, lower, word,
+graph, print, punct, and cntrl. Other classes are built from combinations. */
+
+  0x00,0x3e,0x00,0x00,0x01,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+
+  0x00,0x00,0x00,0x00,0x00,0x00,0xff,0x03,
+  0x7e,0x00,0x00,0x00,0x7e,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+
+  0x00,0x00,0x00,0x00,0x00,0x00,0xff,0x03,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+  0xfe,0xff,0xff,0x07,0x00,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0xfe,0xff,0xff,0x07,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+
+  0x00,0x00,0x00,0x00,0x00,0x00,0xff,0x03,
+  0xfe,0xff,0xff,0x87,0xfe,0xff,0xff,0x07,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+
+  0x00,0x00,0x00,0x00,0xfe,0xff,0xff,0xff,
+  0xff,0xff,0xff,0xff,0xff,0xff,0xff,0x7f,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+
+  0x00,0x00,0x00,0x00,0xff,0xff,0xff,0xff,
+  0xff,0xff,0xff,0xff,0xff,0xff,0xff,0x7f,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+
+  0x00,0x00,0x00,0x00,0xfe,0xff,0x00,0xfc,
+  0x01,0x00,0x00,0xf8,0x01,0x00,0x00,0x78,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+
+  0xff,0xff,0xff,0xff,0x00,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x80,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+
+/* This table identifies various classes of character by individual bits:
+  0x01   white space character
+  0x02   letter
+  0x04   decimal digit
+  0x08   hexadecimal digit
+  0x10   alphanumeric or '_'
+  0x80   regular expression metacharacter or binary zero
+*/
+
+  0x80,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*   0-  7 */
+  0x00,0x01,0x01,0x00,0x01,0x01,0x00,0x00, /*   8- 15 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*  16- 23 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*  24- 31 */
+  0x01,0x00,0x00,0x00,0x80,0x00,0x00,0x00, /*    - '  */
+  0x80,0x80,0x80,0x80,0x00,0x00,0x80,0x00, /*  ( - /  */
+  0x1c,0x1c,0x1c,0x1c,0x1c,0x1c,0x1c,0x1c, /*  0 - 7  */
+  0x1c,0x1c,0x00,0x00,0x00,0x00,0x00,0x80, /*  8 - ?  */
+  0x00,0x1a,0x1a,0x1a,0x1a,0x1a,0x1a,0x12, /*  @ - G  */
+  0x12,0x12,0x12,0x12,0x12,0x12,0x12,0x12, /*  H - O  */
+  0x12,0x12,0x12,0x12,0x12,0x12,0x12,0x12, /*  P - W  */
+  0x12,0x12,0x12,0x80,0x80,0x00,0x80,0x10, /*  X - _  */
+  0x00,0x1a,0x1a,0x1a,0x1a,0x1a,0x1a,0x12, /*  ` - g  */
+  0x12,0x12,0x12,0x12,0x12,0x12,0x12,0x12, /*  h - o  */
+  0x12,0x12,0x12,0x12,0x12,0x12,0x12,0x12, /*  p - w  */
+  0x12,0x12,0x12,0x80,0x80,0x00,0x00,0x00, /*  x -127 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 128-135 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 136-143 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 144-151 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 152-159 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 160-167 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 168-175 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 176-183 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 184-191 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 192-199 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 200-207 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 208-215 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 216-223 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 224-231 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 232-239 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 240-247 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00};/* 248-255 */
+
+/* End of pcre_chartables.c */
diff --git a/pcre/pcre_chartables.c.dist b/pcre/pcre_chartables.c.dist
new file mode 100644
index 00000000000..2a39e9ff33a
--- /dev/null
+++ b/pcre/pcre_chartables.c.dist
@@ -0,0 +1,198 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* This file contains character tables that are used when no external tables
+are passed to PCRE by the application that calls it. The tables are used only
+for characters whose code values are less than 256.
+
+This is a default version of the tables that assumes ASCII encoding. A program
+called dftables (which is distributed with PCRE) can be used to build
+alternative versions of this file. This is necessary if you are running in an
+EBCDIC environment, or if you want to default to a different encoding, for
+example ISO-8859-1. When dftables is run, it creates these tables in the
+current locale. If PCRE is configured with --enable-rebuild-chartables, this
+happens automatically.
+
+The following #includes are present because without them gcc 4.x may remove the
+array definition from the final binary if PCRE is built into a static library
+and dead code stripping is activated. This leads to link errors. Pulling in the
+header ensures that the array gets flagged as "someone outside this compilation
+unit might reference this" and so it will always be supplied to the linker. */
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include "pcre_internal.h"
+
+const pcre_uint8 PRIV(default_tables)[] = {
+
+/* This table is a lower casing table. */
+
+    0,  1,  2,  3,  4,  5,  6,  7,
+    8,  9, 10, 11, 12, 13, 14, 15,
+   16, 17, 18, 19, 20, 21, 22, 23,
+   24, 25, 26, 27, 28, 29, 30, 31,
+   32, 33, 34, 35, 36, 37, 38, 39,
+   40, 41, 42, 43, 44, 45, 46, 47,
+   48, 49, 50, 51, 52, 53, 54, 55,
+   56, 57, 58, 59, 60, 61, 62, 63,
+   64, 97, 98, 99,100,101,102,103,
+  104,105,106,107,108,109,110,111,
+  112,113,114,115,116,117,118,119,
+  120,121,122, 91, 92, 93, 94, 95,
+   96, 97, 98, 99,100,101,102,103,
+  104,105,106,107,108,109,110,111,
+  112,113,114,115,116,117,118,119,
+  120,121,122,123,124,125,126,127,
+  128,129,130,131,132,133,134,135,
+  136,137,138,139,140,141,142,143,
+  144,145,146,147,148,149,150,151,
+  152,153,154,155,156,157,158,159,
+  160,161,162,163,164,165,166,167,
+  168,169,170,171,172,173,174,175,
+  176,177,178,179,180,181,182,183,
+  184,185,186,187,188,189,190,191,
+  192,193,194,195,196,197,198,199,
+  200,201,202,203,204,205,206,207,
+  208,209,210,211,212,213,214,215,
+  216,217,218,219,220,221,222,223,
+  224,225,226,227,228,229,230,231,
+  232,233,234,235,236,237,238,239,
+  240,241,242,243,244,245,246,247,
+  248,249,250,251,252,253,254,255,
+
+/* This table is a case flipping table. */
+
+    0,  1,  2,  3,  4,  5,  6,  7,
+    8,  9, 10, 11, 12, 13, 14, 15,
+   16, 17, 18, 19, 20, 21, 22, 23,
+   24, 25, 26, 27, 28, 29, 30, 31,
+   32, 33, 34, 35, 36, 37, 38, 39,
+   40, 41, 42, 43, 44, 45, 46, 47,
+   48, 49, 50, 51, 52, 53, 54, 55,
+   56, 57, 58, 59, 60, 61, 62, 63,
+   64, 97, 98, 99,100,101,102,103,
+  104,105,106,107,108,109,110,111,
+  112,113,114,115,116,117,118,119,
+  120,121,122, 91, 92, 93, 94, 95,
+   96, 65, 66, 67, 68, 69, 70, 71,
+   72, 73, 74, 75, 76, 77, 78, 79,
+   80, 81, 82, 83, 84, 85, 86, 87,
+   88, 89, 90,123,124,125,126,127,
+  128,129,130,131,132,133,134,135,
+  136,137,138,139,140,141,142,143,
+  144,145,146,147,148,149,150,151,
+  152,153,154,155,156,157,158,159,
+  160,161,162,163,164,165,166,167,
+  168,169,170,171,172,173,174,175,
+  176,177,178,179,180,181,182,183,
+  184,185,186,187,188,189,190,191,
+  192,193,194,195,196,197,198,199,
+  200,201,202,203,204,205,206,207,
+  208,209,210,211,212,213,214,215,
+  216,217,218,219,220,221,222,223,
+  224,225,226,227,228,229,230,231,
+  232,233,234,235,236,237,238,239,
+  240,241,242,243,244,245,246,247,
+  248,249,250,251,252,253,254,255,
+
+/* This table contains bit maps for various character classes. Each map is 32
+bytes long and the bits run from the least significant end of each byte. The
+classes that have their own maps are: space, xdigit, digit, upper, lower, word,
+graph, print, punct, and cntrl. Other classes are built from combinations. */
+
+  0x00,0x3e,0x00,0x00,0x01,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+
+  0x00,0x00,0x00,0x00,0x00,0x00,0xff,0x03,
+  0x7e,0x00,0x00,0x00,0x7e,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+
+  0x00,0x00,0x00,0x00,0x00,0x00,0xff,0x03,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+  0xfe,0xff,0xff,0x07,0x00,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0xfe,0xff,0xff,0x07,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+
+  0x00,0x00,0x00,0x00,0x00,0x00,0xff,0x03,
+  0xfe,0xff,0xff,0x87,0xfe,0xff,0xff,0x07,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+
+  0x00,0x00,0x00,0x00,0xfe,0xff,0xff,0xff,
+  0xff,0xff,0xff,0xff,0xff,0xff,0xff,0x7f,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+
+  0x00,0x00,0x00,0x00,0xff,0xff,0xff,0xff,
+  0xff,0xff,0xff,0xff,0xff,0xff,0xff,0x7f,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+
+  0x00,0x00,0x00,0x00,0xfe,0xff,0x00,0xfc,
+  0x01,0x00,0x00,0xf8,0x01,0x00,0x00,0x78,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+
+  0xff,0xff,0xff,0xff,0x00,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x80,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+
+/* This table identifies various classes of character by individual bits:
+  0x01   white space character
+  0x02   letter
+  0x04   decimal digit
+  0x08   hexadecimal digit
+  0x10   alphanumeric or '_'
+  0x80   regular expression metacharacter or binary zero
+*/
+
+  0x80,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*   0-  7 */
+  0x00,0x01,0x01,0x00,0x01,0x01,0x00,0x00, /*   8- 15 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*  16- 23 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*  24- 31 */
+  0x01,0x00,0x00,0x00,0x80,0x00,0x00,0x00, /*    - '  */
+  0x80,0x80,0x80,0x80,0x00,0x00,0x80,0x00, /*  ( - /  */
+  0x1c,0x1c,0x1c,0x1c,0x1c,0x1c,0x1c,0x1c, /*  0 - 7  */
+  0x1c,0x1c,0x00,0x00,0x00,0x00,0x00,0x80, /*  8 - ?  */
+  0x00,0x1a,0x1a,0x1a,0x1a,0x1a,0x1a,0x12, /*  @ - G  */
+  0x12,0x12,0x12,0x12,0x12,0x12,0x12,0x12, /*  H - O  */
+  0x12,0x12,0x12,0x12,0x12,0x12,0x12,0x12, /*  P - W  */
+  0x12,0x12,0x12,0x80,0x80,0x00,0x80,0x10, /*  X - _  */
+  0x00,0x1a,0x1a,0x1a,0x1a,0x1a,0x1a,0x12, /*  ` - g  */
+  0x12,0x12,0x12,0x12,0x12,0x12,0x12,0x12, /*  h - o  */
+  0x12,0x12,0x12,0x12,0x12,0x12,0x12,0x12, /*  p - w  */
+  0x12,0x12,0x12,0x80,0x80,0x00,0x00,0x00, /*  x -127 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 128-135 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 136-143 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 144-151 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 152-159 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 160-167 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 168-175 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 176-183 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 184-191 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 192-199 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 200-207 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 208-215 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 216-223 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 224-231 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 232-239 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 240-247 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00};/* 248-255 */
+
+/* End of pcre_chartables.c */
diff --git a/pcre/pcre_compile.c b/pcre/pcre_compile.c
new file mode 100644
index 00000000000..1f7bdb40b19
--- /dev/null
+++ b/pcre/pcre_compile.c
@@ -0,0 +1,8432 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2013 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+
+/* This module contains the external function pcre_compile(), along with
+supporting internal functions that are not used by other modules. */
+
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#define NLBLOCK cd             /* Block containing newline information */
+#define PSSTART start_pattern  /* Field containing processed string start */
+#define PSEND   end_pattern    /* Field containing processed string end */
+
+#include "pcre_internal.h"
+
+
+/* When PCRE_DEBUG is defined, we need the pcre(16|32)_printint() function, which
+is also used by pcretest. PCRE_DEBUG is not defined when building a production
+library. We do not need to select pcre16_printint.c specially, because the
+COMPILE_PCREx macro will already be appropriately set. */
+
+#ifdef PCRE_DEBUG
+/* pcre_printint.c should not include any headers */
+#define PCRE_INCLUDED
+#include "pcre_printint.c"
+#undef PCRE_INCLUDED
+#endif
+
+
+/* Macro for setting individual bits in class bitmaps. */
+
+#define SETBIT(a,b) a[(b)/8] |= (1 << ((b)&7))
+
+/* Maximum length value to check against when making sure that the integer that
+holds the compiled pattern length does not overflow. We make it a bit less than
+INT_MAX to allow for adding in group terminating bytes, so that we don't have
+to check them every time. */
+
+#define OFLOW_MAX (INT_MAX - 20)
+
+/* Definitions to allow mutual recursion */
+
+static int
+  add_list_to_class(pcre_uint8 *, pcre_uchar **, int, compile_data *,
+    const pcre_uint32 *, unsigned int);
+
+static BOOL
+  compile_regex(int, pcre_uchar **, const pcre_uchar **, int *, BOOL, BOOL, int, int,
+    pcre_uint32 *, pcre_int32 *, pcre_uint32 *, pcre_int32 *, branch_chain *,
+    compile_data *, int *);
+
+
+
+/*************************************************
+*      Code parameters and static tables         *
+*************************************************/
+
+/* This value specifies the size of stack workspace that is used during the
+first pre-compile phase that determines how much memory is required. The regex
+is partly compiled into this space, but the compiled parts are discarded as
+soon as they can be, so that hopefully there will never be an overrun. The code
+does, however, check for an overrun. The largest amount I've seen used is 218,
+so this number is very generous.
+
+The same workspace is used during the second, actual compile phase for
+remembering forward references to groups so that they can be filled in at the
+end. Each entry in this list occupies LINK_SIZE bytes, so even when LINK_SIZE
+is 4 there is plenty of room for most patterns. However, the memory can get
+filled up by repetitions of forward references, for example patterns like
+/(?1){0,1999}(b)/, and one user did hit the limit. The code has been changed so
+that the workspace is expanded using malloc() in this situation. The value
+below is therefore a minimum, and we put a maximum on it for safety. The
+minimum is now also defined in terms of LINK_SIZE so that the use of malloc()
+kicks in at the same number of forward references in all cases. */
+
+#define COMPILE_WORK_SIZE (2048*LINK_SIZE)
+#define COMPILE_WORK_SIZE_MAX (100*COMPILE_WORK_SIZE)
+
+/* The overrun tests check for a slightly smaller size so that they detect the
+overrun before it actually does run off the end of the data block. */
+
+#define WORK_SIZE_SAFETY_MARGIN (100)
+
+/* Private flags added to firstchar and reqchar. */
+
+#define REQ_CASELESS    (1 << 0)        /* Indicates caselessness */
+#define REQ_VARY        (1 << 1)        /* Reqchar followed non-literal item */
+/* Negative values for the firstchar and reqchar flags */
+#define REQ_UNSET       (-2)
+#define REQ_NONE        (-1)
+
+/* Repeated character flags. */
+
+#define UTF_LENGTH     0x10000000l      /* The char contains its length. */
+
+/* Table for handling escaped characters in the range '0'-'z'. Positive returns
+are simple data values; negative values are for special things like \d and so
+on. Zero means further processing is needed (for things like \x), or the escape
+is invalid. */
+
+#ifndef EBCDIC
+
+/* This is the "normal" table for ASCII systems or for EBCDIC systems running
+in UTF-8 mode. */
+
+static const short int escapes[] = {
+     0,                       0,
+     0,                       0,
+     0,                       0,
+     0,                       0,
+     0,                       0,
+     CHAR_COLON,              CHAR_SEMICOLON,
+     CHAR_LESS_THAN_SIGN,     CHAR_EQUALS_SIGN,
+     CHAR_GREATER_THAN_SIGN,  CHAR_QUESTION_MARK,
+     CHAR_COMMERCIAL_AT,      -ESC_A,
+     -ESC_B,                  -ESC_C,
+     -ESC_D,                  -ESC_E,
+     0,                       -ESC_G,
+     -ESC_H,                  0,
+     0,                       -ESC_K,
+     0,                       0,
+     -ESC_N,                  0,
+     -ESC_P,                  -ESC_Q,
+     -ESC_R,                  -ESC_S,
+     0,                       0,
+     -ESC_V,                  -ESC_W,
+     -ESC_X,                  0,
+     -ESC_Z,                  CHAR_LEFT_SQUARE_BRACKET,
+     CHAR_BACKSLASH,          CHAR_RIGHT_SQUARE_BRACKET,
+     CHAR_CIRCUMFLEX_ACCENT,  CHAR_UNDERSCORE,
+     CHAR_GRAVE_ACCENT,       7,
+     -ESC_b,                  0,
+     -ESC_d,                  ESC_e,
+     ESC_f,                   0,
+     -ESC_h,                  0,
+     0,                       -ESC_k,
+     0,                       0,
+     ESC_n,                   0,
+     -ESC_p,                  0,
+     ESC_r,                   -ESC_s,
+     ESC_tee,                 0,
+     -ESC_v,                  -ESC_w,
+     0,                       0,
+     -ESC_z
+};
+
+#else
+
+/* This is the "abnormal" table for EBCDIC systems without UTF-8 support. */
+
+static const short int escapes[] = {
+/*  48 */     0,     0,      0,     '.',    '<',   '(',    '+',    '|',
+/*  50 */   '&',     0,      0,       0,      0,     0,      0,      0,
+/*  58 */     0,     0,    '!',     '$',    '*',   ')',    ';',    '~',
+/*  60 */   '-',   '/',      0,       0,      0,     0,      0,      0,
+/*  68 */     0,     0,    '|',     ',',    '%',   '_',    '>',    '?',
+/*  70 */     0,     0,      0,       0,      0,     0,      0,      0,
+/*  78 */     0,   '`',    ':',     '#',    '@',  '\'',    '=',    '"',
+/*  80 */     0,     7, -ESC_b,       0, -ESC_d, ESC_e,  ESC_f,      0,
+/*  88 */-ESC_h,     0,      0,     '{',      0,     0,      0,      0,
+/*  90 */     0,     0, -ESC_k,     'l',      0, ESC_n,      0, -ESC_p,
+/*  98 */     0, ESC_r,      0,     '}',      0,     0,      0,      0,
+/*  A0 */     0,   '~', -ESC_s, ESC_tee,      0,-ESC_v, -ESC_w,      0,
+/*  A8 */     0,-ESC_z,      0,       0,      0,   '[',      0,      0,
+/*  B0 */     0,     0,      0,       0,      0,     0,      0,      0,
+/*  B8 */     0,     0,      0,       0,      0,   ']',    '=',    '-',
+/*  C0 */   '{',-ESC_A, -ESC_B,  -ESC_C, -ESC_D,-ESC_E,      0, -ESC_G,
+/*  C8 */-ESC_H,     0,      0,       0,      0,     0,      0,      0,
+/*  D0 */   '}',     0, -ESC_K,       0,      0,-ESC_N,      0, -ESC_P,
+/*  D8 */-ESC_Q,-ESC_R,      0,       0,      0,     0,      0,      0,
+/*  E0 */  '\\',     0, -ESC_S,       0,      0,-ESC_V, -ESC_W, -ESC_X,
+/*  E8 */     0,-ESC_Z,      0,       0,      0,     0,      0,      0,
+/*  F0 */     0,     0,      0,       0,      0,     0,      0,      0,
+/*  F8 */     0,     0,      0,       0,      0,     0,      0,      0
+};
+#endif
+
+
+/* Table of special "verbs" like (*PRUNE). This is a short table, so it is
+searched linearly. Put all the names into a single string, in order to reduce
+the number of relocations when a shared library is dynamically linked. The
+string is built from string macros so that it works in UTF-8 mode on EBCDIC
+platforms. */
+
+typedef struct verbitem {
+  int   len;                 /* Length of verb name */
+  int   op;                  /* Op when no arg, or -1 if arg mandatory */
+  int   op_arg;              /* Op when arg present, or -1 if not allowed */
+} verbitem;
+
+static const char verbnames[] =
+  "\0"                       /* Empty name is a shorthand for MARK */
+  STRING_MARK0
+  STRING_ACCEPT0
+  STRING_COMMIT0
+  STRING_F0
+  STRING_FAIL0
+  STRING_PRUNE0
+  STRING_SKIP0
+  STRING_THEN;
+
+static const verbitem verbs[] = {
+  { 0, -1,        OP_MARK },
+  { 4, -1,        OP_MARK },
+  { 6, OP_ACCEPT, -1 },
+  { 6, OP_COMMIT, -1 },
+  { 1, OP_FAIL,   -1 },
+  { 4, OP_FAIL,   -1 },
+  { 5, OP_PRUNE,  OP_PRUNE_ARG },
+  { 4, OP_SKIP,   OP_SKIP_ARG  },
+  { 4, OP_THEN,   OP_THEN_ARG  }
+};
+
+static const int verbcount = sizeof(verbs)/sizeof(verbitem);
+
+
+/* Tables of names of POSIX character classes and their lengths. The names are
+now all in a single string, to reduce the number of relocations when a shared
+library is dynamically loaded. The list of lengths is terminated by a zero
+length entry. The first three must be alpha, lower, upper, as this is assumed
+for handling case independence. */
+
+static const char posix_names[] =
+  STRING_alpha0 STRING_lower0 STRING_upper0 STRING_alnum0
+  STRING_ascii0 STRING_blank0 STRING_cntrl0 STRING_digit0
+  STRING_graph0 STRING_print0 STRING_punct0 STRING_space0
+  STRING_word0  STRING_xdigit;
+
+static const pcre_uint8 posix_name_lengths[] = {
+  5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 4, 6, 0 };
+
+/* Table of class bit maps for each POSIX class. Each class is formed from a
+base map, with an optional addition or removal of another map. Then, for some
+classes, there is some additional tweaking: for [:blank:] the vertical space
+characters are removed, and for [:alpha:] and [:alnum:] the underscore
+character is removed. The triples in the table consist of the base map offset,
+second map offset or -1 if no second map, and a non-negative value for map
+addition or a negative value for map subtraction (if there are two maps). The
+absolute value of the third field has these meanings: 0 => no tweaking, 1 =>
+remove vertical space characters, 2 => remove underscore. */
+
+static const int posix_class_maps[] = {
+  cbit_word,  cbit_digit, -2,             /* alpha */
+  cbit_lower, -1,          0,             /* lower */
+  cbit_upper, -1,          0,             /* upper */
+  cbit_word,  -1,          2,             /* alnum - word without underscore */
+  cbit_print, cbit_cntrl,  0,             /* ascii */
+  cbit_space, -1,          1,             /* blank - a GNU extension */
+  cbit_cntrl, -1,          0,             /* cntrl */
+  cbit_digit, -1,          0,             /* digit */
+  cbit_graph, -1,          0,             /* graph */
+  cbit_print, -1,          0,             /* print */
+  cbit_punct, -1,          0,             /* punct */
+  cbit_space, -1,          0,             /* space */
+  cbit_word,  -1,          0,             /* word - a Perl extension */
+  cbit_xdigit,-1,          0              /* xdigit */
+};
+
+/* Table of substitutes for \d etc when PCRE_UCP is set. The POSIX class
+substitutes must be in the order of the names, defined above, and there are
+both positive and negative cases. NULL means no substitute. */
+
+#ifdef SUPPORT_UCP
+static const pcre_uchar string_PNd[]  = {
+  CHAR_BACKSLASH, CHAR_P, CHAR_LEFT_CURLY_BRACKET,
+  CHAR_N, CHAR_d, CHAR_RIGHT_CURLY_BRACKET, '\0' };
+static const pcre_uchar string_pNd[]  = {
+  CHAR_BACKSLASH, CHAR_p, CHAR_LEFT_CURLY_BRACKET,
+  CHAR_N, CHAR_d, CHAR_RIGHT_CURLY_BRACKET, '\0' };
+static const pcre_uchar string_PXsp[] = {
+  CHAR_BACKSLASH, CHAR_P, CHAR_LEFT_CURLY_BRACKET,
+  CHAR_X, CHAR_s, CHAR_p, CHAR_RIGHT_CURLY_BRACKET, '\0' };
+static const pcre_uchar string_pXsp[] = {
+  CHAR_BACKSLASH, CHAR_p, CHAR_LEFT_CURLY_BRACKET,
+  CHAR_X, CHAR_s, CHAR_p, CHAR_RIGHT_CURLY_BRACKET, '\0' };
+static const pcre_uchar string_PXwd[] = {
+  CHAR_BACKSLASH, CHAR_P, CHAR_LEFT_CURLY_BRACKET,
+  CHAR_X, CHAR_w, CHAR_d, CHAR_RIGHT_CURLY_BRACKET, '\0' };
+static const pcre_uchar string_pXwd[] = {
+  CHAR_BACKSLASH, CHAR_p, CHAR_LEFT_CURLY_BRACKET,
+  CHAR_X, CHAR_w, CHAR_d, CHAR_RIGHT_CURLY_BRACKET, '\0' };
+
+static const pcre_uchar *substitutes[] = {
+  string_PNd,           /* \D */
+  string_pNd,           /* \d */
+  string_PXsp,          /* \S */       /* NOTE: Xsp is Perl space */
+  string_pXsp,          /* \s */
+  string_PXwd,          /* \W */
+  string_pXwd           /* \w */
+};
+
+static const pcre_uchar string_pL[] =   {
+  CHAR_BACKSLASH, CHAR_p, CHAR_LEFT_CURLY_BRACKET,
+  CHAR_L, CHAR_RIGHT_CURLY_BRACKET, '\0' };
+static const pcre_uchar string_pLl[] =  {
+  CHAR_BACKSLASH, CHAR_p, CHAR_LEFT_CURLY_BRACKET,
+  CHAR_L, CHAR_l, CHAR_RIGHT_CURLY_BRACKET, '\0' };
+static const pcre_uchar string_pLu[] =  {
+  CHAR_BACKSLASH, CHAR_p, CHAR_LEFT_CURLY_BRACKET,
+  CHAR_L, CHAR_u, CHAR_RIGHT_CURLY_BRACKET, '\0' };
+static const pcre_uchar string_pXan[] = {
+  CHAR_BACKSLASH, CHAR_p, CHAR_LEFT_CURLY_BRACKET,
+  CHAR_X, CHAR_a, CHAR_n, CHAR_RIGHT_CURLY_BRACKET, '\0' };
+static const pcre_uchar string_h[] =    {
+  CHAR_BACKSLASH, CHAR_h, '\0' };
+static const pcre_uchar string_pXps[] = {
+  CHAR_BACKSLASH, CHAR_p, CHAR_LEFT_CURLY_BRACKET,
+  CHAR_X, CHAR_p, CHAR_s, CHAR_RIGHT_CURLY_BRACKET, '\0' };
+static const pcre_uchar string_PL[] =   {
+  CHAR_BACKSLASH, CHAR_P, CHAR_LEFT_CURLY_BRACKET,
+  CHAR_L, CHAR_RIGHT_CURLY_BRACKET, '\0' };
+static const pcre_uchar string_PLl[] =  {
+  CHAR_BACKSLASH, CHAR_P, CHAR_LEFT_CURLY_BRACKET,
+  CHAR_L, CHAR_l, CHAR_RIGHT_CURLY_BRACKET, '\0' };
+static const pcre_uchar string_PLu[] =  {
+  CHAR_BACKSLASH, CHAR_P, CHAR_LEFT_CURLY_BRACKET,
+  CHAR_L, CHAR_u, CHAR_RIGHT_CURLY_BRACKET, '\0' };
+static const pcre_uchar string_PXan[] = {
+  CHAR_BACKSLASH, CHAR_P, CHAR_LEFT_CURLY_BRACKET,
+  CHAR_X, CHAR_a, CHAR_n, CHAR_RIGHT_CURLY_BRACKET, '\0' };
+static const pcre_uchar string_H[] =    {
+  CHAR_BACKSLASH, CHAR_H, '\0' };
+static const pcre_uchar string_PXps[] = {
+  CHAR_BACKSLASH, CHAR_P, CHAR_LEFT_CURLY_BRACKET,
+  CHAR_X, CHAR_p, CHAR_s, CHAR_RIGHT_CURLY_BRACKET, '\0' };
+
+static const pcre_uchar *posix_substitutes[] = {
+  string_pL,            /* alpha */
+  string_pLl,           /* lower */
+  string_pLu,           /* upper */
+  string_pXan,          /* alnum */
+  NULL,                 /* ascii */
+  string_h,             /* blank */
+  NULL,                 /* cntrl */
+  string_pNd,           /* digit */
+  NULL,                 /* graph */
+  NULL,                 /* print */
+  NULL,                 /* punct */
+  string_pXps,          /* space */    /* NOTE: Xps is POSIX space */
+  string_pXwd,          /* word */
+  NULL,                 /* xdigit */
+  /* Negated cases */
+  string_PL,            /* ^alpha */
+  string_PLl,           /* ^lower */
+  string_PLu,           /* ^upper */
+  string_PXan,          /* ^alnum */
+  NULL,                 /* ^ascii */
+  string_H,             /* ^blank */
+  NULL,                 /* ^cntrl */
+  string_PNd,           /* ^digit */
+  NULL,                 /* ^graph */
+  NULL,                 /* ^print */
+  NULL,                 /* ^punct */
+  string_PXps,          /* ^space */   /* NOTE: Xps is POSIX space */
+  string_PXwd,          /* ^word */
+  NULL                  /* ^xdigit */
+};
+#define POSIX_SUBSIZE (sizeof(posix_substitutes) / sizeof(pcre_uchar *))
+#endif
+
+#define STRING(a)  # a
+#define XSTRING(s) STRING(s)
+
+/* The texts of compile-time error messages. These are "char *" because they
+are passed to the outside world. Do not ever re-use any error number, because
+they are documented. Always add a new error instead. Messages marked DEAD below
+are no longer used. This used to be a table of strings, but in order to reduce
+the number of relocations needed when a shared library is loaded dynamically,
+it is now one long string. We cannot use a table of offsets, because the
+lengths of inserts such as XSTRING(MAX_NAME_SIZE) are not known. Instead, we
+simply count through to the one we want - this isn't a performance issue
+because these strings are used only when there is a compilation error.
+
+Each substring ends with \0 to insert a null character. This includes the final
+substring, so that the whole string ends with \0\0, which can be detected when
+counting through. */
+
+static const char error_texts[] =
+  "no error\0"
+  "\\ at end of pattern\0"
+  "\\c at end of pattern\0"
+  "unrecognized character follows \\\0"
+  "numbers out of order in {} quantifier\0"
+  /* 5 */
+  "number too big in {} quantifier\0"
+  "missing terminating ] for character class\0"
+  "invalid escape sequence in character class\0"
+  "range out of order in character class\0"
+  "nothing to repeat\0"
+  /* 10 */
+  "operand of unlimited repeat could match the empty string\0"  /** DEAD **/
+  "internal error: unexpected repeat\0"
+  "unrecognized character after (? or (?-\0"
+  "POSIX named classes are supported only within a class\0"
+  "missing )\0"
+  /* 15 */
+  "reference to non-existent subpattern\0"
+  "erroffset passed as NULL\0"
+  "unknown option bit(s) set\0"
+  "missing ) after comment\0"
+  "parentheses nested too deeply\0"  /** DEAD **/
+  /* 20 */
+  "regular expression is too large\0"
+  "failed to get memory\0"
+  "unmatched parentheses\0"
+  "internal error: code overflow\0"
+  "unrecognized character after (?<\0"
+  /* 25 */
+  "lookbehind assertion is not fixed length\0"
+  "malformed number or name after (?(\0"
+  "conditional group contains more than two branches\0"
+  "assertion expected after (?(\0"
+  "(?R or (?[+-]digits must be followed by )\0"
+  /* 30 */
+  "unknown POSIX class name\0"
+  "POSIX collating elements are not supported\0"
+  "this version of PCRE is compiled without UTF support\0"
+  "spare error\0"  /** DEAD **/
+  "character value in \\x{...} sequence is too large\0"
+  /* 35 */
+  "invalid condition (?(0)\0"
+  "\\C not allowed in lookbehind assertion\0"
+  "PCRE does not support \\L, \\l, \\N{name}, \\U, or \\u\0"
+  "number after (?C is > 255\0"
+  "closing ) for (?C expected\0"
+  /* 40 */
+  "recursive call could loop indefinitely\0"
+  "unrecognized character after (?P\0"
+  "syntax error in subpattern name (missing terminator)\0"
+  "two named subpatterns have the same name\0"
+  "invalid UTF-8 string\0"
+  /* 45 */
+  "support for \\P, \\p, and \\X has not been compiled\0"
+  "malformed \\P or \\p sequence\0"
+  "unknown property name after \\P or \\p\0"
+  "subpattern name is too long (maximum " XSTRING(MAX_NAME_SIZE) " characters)\0"
+  "too many named subpatterns (maximum " XSTRING(MAX_NAME_COUNT) ")\0"
+  /* 50 */
+  "repeated subpattern is too long\0"    /** DEAD **/
+  "octal value is greater than \\377 in 8-bit non-UTF-8 mode\0"
+  "internal error: overran compiling workspace\0"
+  "internal error: previously-checked referenced subpattern not found\0"
+  "DEFINE group contains more than one branch\0"
+  /* 55 */
+  "repeating a DEFINE group is not allowed\0"  /** DEAD **/
+  "inconsistent NEWLINE options\0"
+  "\\g is not followed by a braced, angle-bracketed, or quoted name/number or by a plain number\0"
+  "a numbered reference must not be zero\0"
+  "an argument is not allowed for (*ACCEPT), (*FAIL), or (*COMMIT)\0"
+  /* 60 */
+  "(*VERB) not recognized or malformed\0"
+  "number is too big\0"
+  "subpattern name expected\0"
+  "digit expected after (?+\0"
+  "] is an invalid data character in JavaScript compatibility mode\0"
+  /* 65 */
+  "different names for subpatterns of the same number are not allowed\0"
+  "(*MARK) must have an argument\0"
+  "this version of PCRE is not compiled with Unicode property support\0"
+  "\\c must be followed by an ASCII character\0"
+  "\\k is not followed by a braced, angle-bracketed, or quoted name\0"
+  /* 70 */
+  "internal error: unknown opcode in find_fixedlength()\0"
+  "\\N is not supported in a class\0"
+  "too many forward references\0"
+  "disallowed Unicode code point (>= 0xd800 && <= 0xdfff)\0"
+  "invalid UTF-16 string\0"
+  /* 75 */
+  "name is too long in (*MARK), (*PRUNE), (*SKIP), or (*THEN)\0"
+  "character value in \\u.... sequence is too large\0"
+  "invalid UTF-32 string\0"
+  "setting UTF is disabled by the application\0"
+  ;
+
+/* Table to identify digits and hex digits. This is used when compiling
+patterns. Note that the tables in chartables are dependent on the locale, and
+may mark arbitrary characters as digits - but the PCRE compiling code expects
+to handle only 0-9, a-z, and A-Z as digits when compiling. That is why we have
+a private table here. It costs 256 bytes, but it is a lot faster than doing
+character value tests (at least in some simple cases I timed), and in some
+applications one wants PCRE to compile efficiently as well as match
+efficiently.
+
+For convenience, we use the same bit definitions as in chartables:
+
+  0x04   decimal digit
+  0x08   hexadecimal digit
+
+Then we can use ctype_digit and ctype_xdigit in the code. */
+
+/* Using a simple comparison for decimal numbers rather than a memory read
+is much faster, and the resulting code is simpler (the compiler turns it
+into a subtraction and unsigned comparison). */
+
+#define IS_DIGIT(x) ((x) >= CHAR_0 && (x) <= CHAR_9)
+
+#ifndef EBCDIC
+
+/* This is the "normal" case, for ASCII systems, and EBCDIC systems running in
+UTF-8 mode. */
+
+static const pcre_uint8 digitab[] =
+  {
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*   0-  7 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*   8- 15 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*  16- 23 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*  24- 31 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*    - '  */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*  ( - /  */
+  0x0c,0x0c,0x0c,0x0c,0x0c,0x0c,0x0c,0x0c, /*  0 - 7  */
+  0x0c,0x0c,0x00,0x00,0x00,0x00,0x00,0x00, /*  8 - ?  */
+  0x00,0x08,0x08,0x08,0x08,0x08,0x08,0x00, /*  @ - G  */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*  H - O  */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*  P - W  */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*  X - _  */
+  0x00,0x08,0x08,0x08,0x08,0x08,0x08,0x00, /*  ` - g  */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*  h - o  */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*  p - w  */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*  x -127 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 128-135 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 136-143 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 144-151 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 152-159 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 160-167 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 168-175 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 176-183 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 184-191 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 192-199 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 200-207 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 208-215 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 216-223 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 224-231 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 232-239 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 240-247 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00};/* 248-255 */
+
+#else
+
+/* This is the "abnormal" case, for EBCDIC systems not running in UTF-8 mode. */
+
+static const pcre_uint8 digitab[] =
+  {
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*   0-  7  0 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*   8- 15    */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*  16- 23 10 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*  24- 31    */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*  32- 39 20 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*  40- 47    */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*  48- 55 30 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*  56- 63    */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*    - 71 40 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*  72- |     */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*  & - 87 50 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*  88- 95    */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*  - -103 60 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 104- ?     */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 112-119 70 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 120- "     */
+  0x00,0x08,0x08,0x08,0x08,0x08,0x08,0x00, /* 128- g  80 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*  h -143    */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 144- p  90 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*  q -159    */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 160- x  A0 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*  y -175    */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*  ^ -183 B0 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 184-191    */
+  0x00,0x08,0x08,0x08,0x08,0x08,0x08,0x00, /*  { - G  C0 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*  H -207    */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*  } - P  D0 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*  Q -223    */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*  \ - X  E0 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*  Y -239    */
+  0x0c,0x0c,0x0c,0x0c,0x0c,0x0c,0x0c,0x0c, /*  0 - 7  F0 */
+  0x0c,0x0c,0x00,0x00,0x00,0x00,0x00,0x00};/*  8 -255    */
+
+static const pcre_uint8 ebcdic_chartab[] = { /* chartable partial dup */
+  0x80,0x00,0x00,0x00,0x00,0x01,0x00,0x00, /*   0-  7 */
+  0x00,0x00,0x00,0x00,0x01,0x01,0x00,0x00, /*   8- 15 */
+  0x00,0x00,0x00,0x00,0x00,0x01,0x00,0x00, /*  16- 23 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*  24- 31 */
+  0x00,0x00,0x00,0x00,0x00,0x01,0x00,0x00, /*  32- 39 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*  40- 47 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*  48- 55 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*  56- 63 */
+  0x01,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*    - 71 */
+  0x00,0x00,0x00,0x80,0x00,0x80,0x80,0x80, /*  72- |  */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*  & - 87 */
+  0x00,0x00,0x00,0x80,0x80,0x80,0x00,0x00, /*  88- 95 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*  - -103 */
+  0x00,0x00,0x00,0x00,0x00,0x10,0x00,0x80, /* 104- ?  */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 112-119 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 120- "  */
+  0x00,0x1a,0x1a,0x1a,0x1a,0x1a,0x1a,0x12, /* 128- g  */
+  0x12,0x12,0x00,0x00,0x00,0x00,0x00,0x00, /*  h -143 */
+  0x00,0x12,0x12,0x12,0x12,0x12,0x12,0x12, /* 144- p  */
+  0x12,0x12,0x00,0x00,0x00,0x00,0x00,0x00, /*  q -159 */
+  0x00,0x00,0x12,0x12,0x12,0x12,0x12,0x12, /* 160- x  */
+  0x12,0x12,0x00,0x00,0x00,0x00,0x00,0x00, /*  y -175 */
+  0x80,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*  ^ -183 */
+  0x00,0x00,0x80,0x00,0x00,0x00,0x00,0x00, /* 184-191 */
+  0x80,0x1a,0x1a,0x1a,0x1a,0x1a,0x1a,0x12, /*  { - G  */
+  0x12,0x12,0x00,0x00,0x00,0x00,0x00,0x00, /*  H -207 */
+  0x00,0x12,0x12,0x12,0x12,0x12,0x12,0x12, /*  } - P  */
+  0x12,0x12,0x00,0x00,0x00,0x00,0x00,0x00, /*  Q -223 */
+  0x00,0x00,0x12,0x12,0x12,0x12,0x12,0x12, /*  \ - X  */
+  0x12,0x12,0x00,0x00,0x00,0x00,0x00,0x00, /*  Y -239 */
+  0x1c,0x1c,0x1c,0x1c,0x1c,0x1c,0x1c,0x1c, /*  0 - 7  */
+  0x1c,0x1c,0x00,0x00,0x00,0x00,0x00,0x00};/*  8 -255 */
+#endif
+
+
+
+/*************************************************
+*            Find an error text                  *
+*************************************************/
+
+/* The error texts are now all in one long string, to save on relocations. As
+some of the text is of unknown length, we can't use a table of offsets.
+Instead, just count through the strings. This is not a performance issue
+because it happens only when there has been a compilation error.
+
+Argument:   the error number
+Returns:    pointer to the error string
+*/
+
+static const char *
+find_error_text(int n)
+{
+const char *s = error_texts;
+for (; n > 0; n--)
+  {
+  while (*s++ != CHAR_NULL) {};
+  if (*s == CHAR_NULL) return "Error text not found (please report)";
+  }
+return s;
+}
+
+
+/*************************************************
+*           Expand the workspace                 *
+*************************************************/
+
+/* This function is called during the second compiling phase, if the number of
+forward references fills the existing workspace, which is originally a block on
+the stack. A larger block is obtained from malloc() unless the ultimate limit
+has been reached or the increase will be rather small.
+
+Argument: pointer to the compile data block
+Returns:  0 if all went well, else an error number
+*/
+
+static int
+expand_workspace(compile_data *cd)
+{
+pcre_uchar *newspace;
+int newsize = cd->workspace_size * 2;
+
+if (newsize > COMPILE_WORK_SIZE_MAX) newsize = COMPILE_WORK_SIZE_MAX;
+if (cd->workspace_size >= COMPILE_WORK_SIZE_MAX ||
+    newsize - cd->workspace_size < WORK_SIZE_SAFETY_MARGIN)
+ return ERR72;
+
+newspace = (PUBL(malloc))(IN_UCHARS(newsize));
+if (newspace == NULL) return ERR21;
+memcpy(newspace, cd->start_workspace, cd->workspace_size * sizeof(pcre_uchar));
+cd->hwm = (pcre_uchar *)newspace + (cd->hwm - cd->start_workspace);
+if (cd->workspace_size > COMPILE_WORK_SIZE)
+  (PUBL(free))((void *)cd->start_workspace);
+cd->start_workspace = newspace;
+cd->workspace_size = newsize;
+return 0;
+}
+
+
+
+/*************************************************
+*            Check for counted repeat            *
+*************************************************/
+
+/* This function is called when a '{' is encountered in a place where it might
+start a quantifier. It looks ahead to see if it really is a quantifier or not.
+It is only a quantifier if it is one of the forms {ddd} {ddd,} or {ddd,ddd}
+where the ddds are digits.
+
+Arguments:
+  p         pointer to the first char after '{'
+
+Returns:    TRUE or FALSE
+*/
+
+static BOOL
+is_counted_repeat(const pcre_uchar *p)
+{
+if (!IS_DIGIT(*p)) return FALSE;
+p++;
+while (IS_DIGIT(*p)) p++;
+if (*p == CHAR_RIGHT_CURLY_BRACKET) return TRUE;
+
+if (*p++ != CHAR_COMMA) return FALSE;
+if (*p == CHAR_RIGHT_CURLY_BRACKET) return TRUE;
+
+if (!IS_DIGIT(*p)) return FALSE;
+p++;
+while (IS_DIGIT(*p)) p++;
+
+return (*p == CHAR_RIGHT_CURLY_BRACKET);
+}
+
+
+
+/*************************************************
+*            Handle escapes                      *
+*************************************************/
+
+/* This function is called when a \ has been encountered. It either returns a
+positive value for a simple escape such as \n, or 0 for a data character
+which will be placed in chptr. A backreference to group n is returned as
+negative n. When UTF-8 is enabled, a positive value greater than 255 may
+be returned in chptr.
+On entry,ptr is pointing at the \. On exit, it is on the final character of the
+escape sequence.
+
+Arguments:
+  ptrptr         points to the pattern position pointer
+  chptr          points to the data character
+  errorcodeptr   points to the errorcode variable
+  bracount       number of previous extracting brackets
+  options        the options bits
+  isclass        TRUE if inside a character class
+
+Returns:         zero => a data character
+                 positive => a special escape sequence
+                 negative => a back reference
+                 on error, errorcodeptr is set
+*/
+
+static int
+check_escape(const pcre_uchar **ptrptr, pcre_uint32 *chptr, int *errorcodeptr,
+  int bracount, int options, BOOL isclass)
+{
+/* PCRE_UTF16 has the same value as PCRE_UTF8. */
+BOOL utf = (options & PCRE_UTF8) != 0;
+const pcre_uchar *ptr = *ptrptr + 1;
+pcre_uint32 c;
+int escape = 0;
+int i;
+
+GETCHARINCTEST(c, ptr);           /* Get character value, increment pointer */
+ptr--;                            /* Set pointer back to the last byte */
+
+/* If backslash is at the end of the pattern, it's an error. */
+
+if (c == CHAR_NULL) *errorcodeptr = ERR1;
+
+/* Non-alphanumerics are literals. For digits or letters, do an initial lookup
+in a table. A non-zero result is something that can be returned immediately.
+Otherwise further processing may be required. */
+
+#ifndef EBCDIC  /* ASCII/UTF-8 coding */
+/* Not alphanumeric */
+else if (c < CHAR_0 || c > CHAR_z) {}
+else if ((i = escapes[c - CHAR_0]) != 0)
+  { if (i > 0) c = (pcre_uint32)i; else escape = -i; }
+
+#else           /* EBCDIC coding */
+/* Not alphanumeric */
+else if (c < CHAR_a || (!MAX_255(c) || (ebcdic_chartab[c] & 0x0E) == 0)) {}
+else if ((i = escapes[c - 0x48]) != 0)  { if (i > 0) c = (pcre_uint32)i; else escape = -i; }
+#endif
+
+/* Escapes that need further processing, or are illegal. */
+
+else
+  {
+  const pcre_uchar *oldptr;
+  BOOL braced, negated, overflow;
+  int s;
+
+  switch (c)
+    {
+    /* A number of Perl escapes are not handled by PCRE. We give an explicit
+    error. */
+
+    case CHAR_l:
+    case CHAR_L:
+    *errorcodeptr = ERR37;
+    break;
+
+    case CHAR_u:
+    if ((options & PCRE_JAVASCRIPT_COMPAT) != 0)
+      {
+      /* In JavaScript, \u must be followed by four hexadecimal numbers.
+      Otherwise it is a lowercase u letter. */
+      if (MAX_255(ptr[1]) && (digitab[ptr[1]] & ctype_xdigit) != 0
+        && MAX_255(ptr[2]) && (digitab[ptr[2]] & ctype_xdigit) != 0
+        && MAX_255(ptr[3]) && (digitab[ptr[3]] & ctype_xdigit) != 0
+        && MAX_255(ptr[4]) && (digitab[ptr[4]] & ctype_xdigit) != 0)
+        {
+        c = 0;
+        for (i = 0; i < 4; ++i)
+          {
+          register pcre_uint32 cc = *(++ptr);
+#ifndef EBCDIC  /* ASCII/UTF-8 coding */
+          if (cc >= CHAR_a) cc -= 32;               /* Convert to upper case */
+          c = (c << 4) + cc - ((cc < CHAR_A)? CHAR_0 : (CHAR_A - 10));
+#else           /* EBCDIC coding */
+          if (cc >= CHAR_a && cc <= CHAR_z) cc += 64;  /* Convert to upper case */
+          c = (c << 4) + cc - ((cc >= CHAR_0)? CHAR_0 : (CHAR_A - 10));
+#endif
+          }
+
+#if defined COMPILE_PCRE8
+        if (c > (utf ? 0x10ffffU : 0xffU))
+#elif defined COMPILE_PCRE16
+        if (c > (utf ? 0x10ffffU : 0xffffU))
+#elif defined COMPILE_PCRE32
+        if (utf && c > 0x10ffffU)
+#endif
+          {
+          *errorcodeptr = ERR76;
+          }
+        else if (utf && c >= 0xd800 && c <= 0xdfff) *errorcodeptr = ERR73;
+        }
+      }
+    else
+      *errorcodeptr = ERR37;
+    break;
+
+    case CHAR_U:
+    /* In JavaScript, \U is an uppercase U letter. */
+    if ((options & PCRE_JAVASCRIPT_COMPAT) == 0) *errorcodeptr = ERR37;
+    break;
+
+    /* In a character class, \g is just a literal "g". Outside a character
+    class, \g must be followed by one of a number of specific things:
+
+    (1) A number, either plain or braced. If positive, it is an absolute
+    backreference. If negative, it is a relative backreference. This is a Perl
+    5.10 feature.
+
+    (2) Perl 5.10 also supports \g{name} as a reference to a named group. This
+    is part of Perl's movement towards a unified syntax for back references. As
+    this is synonymous with \k{name}, we fudge it up by pretending it really
+    was \k.
+
+    (3) For Oniguruma compatibility we also support \g followed by a name or a
+    number either in angle brackets or in single quotes. However, these are
+    (possibly recursive) subroutine calls, _not_ backreferences. Just return
+    the ESC_g code (cf \k). */
+
+    case CHAR_g:
+    if (isclass) break;
+    if (ptr[1] == CHAR_LESS_THAN_SIGN || ptr[1] == CHAR_APOSTROPHE)
+      {
+      escape = ESC_g;
+      break;
+      }
+
+    /* Handle the Perl-compatible cases */
+
+    if (ptr[1] == CHAR_LEFT_CURLY_BRACKET)
+      {
+      const pcre_uchar *p;
+      for (p = ptr+2; *p != CHAR_NULL && *p != CHAR_RIGHT_CURLY_BRACKET; p++)
+        if (*p != CHAR_MINUS && !IS_DIGIT(*p)) break;
+      if (*p != CHAR_NULL && *p != CHAR_RIGHT_CURLY_BRACKET)
+        {
+        escape = ESC_k;
+        break;
+        }
+      braced = TRUE;
+      ptr++;
+      }
+    else braced = FALSE;
+
+    if (ptr[1] == CHAR_MINUS)
+      {
+      negated = TRUE;
+      ptr++;
+      }
+    else negated = FALSE;
+
+    /* The integer range is limited by the machine's int representation. */
+    s = 0;
+    overflow = FALSE;
+    while (IS_DIGIT(ptr[1]))
+      {
+      if (s > INT_MAX / 10 - 1) /* Integer overflow */
+        {
+        overflow = TRUE;
+        break;
+        }
+      s = s * 10 + (int)(*(++ptr) - CHAR_0);
+      }
+    if (overflow) /* Integer overflow */
+      {
+      while (IS_DIGIT(ptr[1]))
+        ptr++;
+      *errorcodeptr = ERR61;
+      break;
+      }
+
+    if (braced && *(++ptr) != CHAR_RIGHT_CURLY_BRACKET)
+      {
+      *errorcodeptr = ERR57;
+      break;
+      }
+
+    if (s == 0)
+      {
+      *errorcodeptr = ERR58;
+      break;
+      }
+
+    if (negated)
+      {
+      if (s > bracount)
+        {
+        *errorcodeptr = ERR15;
+        break;
+        }
+      s = bracount - (s - 1);
+      }
+
+    escape = -s;
+    break;
+
+    /* The handling of escape sequences consisting of a string of digits
+    starting with one that is not zero is not straightforward. By experiment,
+    the way Perl works seems to be as follows:
+
+    Outside a character class, the digits are read as a decimal number. If the
+    number is less than 10, or if there are that many previous extracting
+    left brackets, then it is a back reference. Otherwise, up to three octal
+    digits are read to form an escaped byte. Thus \123 is likely to be octal
+    123 (cf \0123, which is octal 012 followed by the literal 3). If the octal
+    value is greater than 377, the least significant 8 bits are taken. Inside a
+    character class, \ followed by a digit is always an octal number. */
+
+    case CHAR_1: case CHAR_2: case CHAR_3: case CHAR_4: case CHAR_5:
+    case CHAR_6: case CHAR_7: case CHAR_8: case CHAR_9:
+
+    if (!isclass)
+      {
+      oldptr = ptr;
+      /* The integer range is limited by the machine's int representation. */
+      s = (int)(c -CHAR_0);
+      overflow = FALSE;
+      while (IS_DIGIT(ptr[1]))
+        {
+        if (s > INT_MAX / 10 - 1) /* Integer overflow */
+          {
+          overflow = TRUE;
+          break;
+          }
+        s = s * 10 + (int)(*(++ptr) - CHAR_0);
+        }
+      if (overflow) /* Integer overflow */
+        {
+        while (IS_DIGIT(ptr[1]))
+          ptr++;
+        *errorcodeptr = ERR61;
+        break;
+        }
+      if (s < 10 || s <= bracount)
+        {
+        escape = -s;
+        break;
+        }
+      ptr = oldptr;      /* Put the pointer back and fall through */
+      }
+
+    /* Handle an octal number following \. If the first digit is 8 or 9, Perl
+    generates a binary zero byte and treats the digit as a following literal.
+    Thus we have to pull back the pointer by one. */
+
+    if ((c = *ptr) >= CHAR_8)
+      {
+      ptr--;
+      c = 0;
+      break;
+      }
+
+    /* \0 always starts an octal number, but we may drop through to here with a
+    larger first octal digit. The original code used just to take the least
+    significant 8 bits of octal numbers (I think this is what early Perls used
+    to do). Nowadays we allow for larger numbers in UTF-8 mode and 16-bit mode,
+    but no more than 3 octal digits. */
+
+    case CHAR_0:
+    c -= CHAR_0;
+    while(i++ < 2 && ptr[1] >= CHAR_0 && ptr[1] <= CHAR_7)
+        c = c * 8 + *(++ptr) - CHAR_0;
+#ifdef COMPILE_PCRE8
+    if (!utf && c > 0xff) *errorcodeptr = ERR51;
+#endif
+    break;
+
+    /* \x is complicated. \x{ddd} is a character number which can be greater
+    than 0xff in utf or non-8bit mode, but only if the ddd are hex digits.
+    If not, { is treated as a data character. */
+
+    case CHAR_x:
+    if ((options & PCRE_JAVASCRIPT_COMPAT) != 0)
+      {
+      /* In JavaScript, \x must be followed by two hexadecimal numbers.
+      Otherwise it is a lowercase x letter. */
+      if (MAX_255(ptr[1]) && (digitab[ptr[1]] & ctype_xdigit) != 0
+        && MAX_255(ptr[2]) && (digitab[ptr[2]] & ctype_xdigit) != 0)
+        {
+        c = 0;
+        for (i = 0; i < 2; ++i)
+          {
+          register pcre_uint32 cc = *(++ptr);
+#ifndef EBCDIC  /* ASCII/UTF-8 coding */
+          if (cc >= CHAR_a) cc -= 32;               /* Convert to upper case */
+          c = (c << 4) + cc - ((cc < CHAR_A)? CHAR_0 : (CHAR_A - 10));
+#else           /* EBCDIC coding */
+          if (cc >= CHAR_a && cc <= CHAR_z) cc += 64;  /* Convert to upper case */
+          c = (c << 4) + cc - ((cc >= CHAR_0)? CHAR_0 : (CHAR_A - 10));
+#endif
+          }
+        }
+      break;
+      }
+
+    if (ptr[1] == CHAR_LEFT_CURLY_BRACKET)
+      {
+      const pcre_uchar *pt = ptr + 2;
+
+      c = 0;
+      overflow = FALSE;
+      while (MAX_255(*pt) && (digitab[*pt] & ctype_xdigit) != 0)
+        {
+        register pcre_uint32 cc = *pt++;
+        if (c == 0 && cc == CHAR_0) continue;     /* Leading zeroes */
+
+#ifdef COMPILE_PCRE32
+        if (c >= 0x10000000l) { overflow = TRUE; break; }
+#endif
+
+#ifndef EBCDIC  /* ASCII/UTF-8 coding */
+        if (cc >= CHAR_a) cc -= 32;               /* Convert to upper case */
+        c = (c << 4) + cc - ((cc < CHAR_A)? CHAR_0 : (CHAR_A - 10));
+#else           /* EBCDIC coding */
+        if (cc >= CHAR_a && cc <= CHAR_z) cc += 64;  /* Convert to upper case */
+        c = (c << 4) + cc - ((cc >= CHAR_0)? CHAR_0 : (CHAR_A - 10));
+#endif
+
+#if defined COMPILE_PCRE8
+        if (c > (utf ? 0x10ffffU : 0xffU)) { overflow = TRUE; break; }
+#elif defined COMPILE_PCRE16
+        if (c > (utf ? 0x10ffffU : 0xffffU)) { overflow = TRUE; break; }
+#elif defined COMPILE_PCRE32
+        if (utf && c > 0x10ffffU) { overflow = TRUE; break; }
+#endif
+        }
+
+      if (overflow)
+        {
+        while (MAX_255(*pt) && (digitab[*pt] & ctype_xdigit) != 0) pt++;
+        *errorcodeptr = ERR34;
+        }
+
+      if (*pt == CHAR_RIGHT_CURLY_BRACKET)
+        {
+        if (utf && c >= 0xd800 && c <= 0xdfff) *errorcodeptr = ERR73;
+        ptr = pt;
+        break;
+        }
+
+      /* If the sequence of hex digits does not end with '}', then we don't
+      recognize this construct; fall through to the normal \x handling. */
+      }
+
+    /* Read just a single-byte hex-defined char */
+
+    c = 0;
+    while (i++ < 2 && MAX_255(ptr[1]) && (digitab[ptr[1]] & ctype_xdigit) != 0)
+      {
+      pcre_uint32 cc;                          /* Some compilers don't like */
+      cc = *(++ptr);                           /* ++ in initializers */
+#ifndef EBCDIC  /* ASCII/UTF-8 coding */
+      if (cc >= CHAR_a) cc -= 32;              /* Convert to upper case */
+      c = c * 16 + cc - ((cc < CHAR_A)? CHAR_0 : (CHAR_A - 10));
+#else           /* EBCDIC coding */
+      if (cc <= CHAR_z) cc += 64;              /* Convert to upper case */
+      c = c * 16 + cc - ((cc >= CHAR_0)? CHAR_0 : (CHAR_A - 10));
+#endif
+      }
+    break;
+
+    /* For \c, a following letter is upper-cased; then the 0x40 bit is flipped.
+    An error is given if the byte following \c is not an ASCII character. This
+    coding is ASCII-specific, but then the whole concept of \cx is
+    ASCII-specific. (However, an EBCDIC equivalent has now been added.) */
+
+    case CHAR_c:
+    c = *(++ptr);
+    if (c == CHAR_NULL)
+      {
+      *errorcodeptr = ERR2;
+      break;
+      }
+#ifndef EBCDIC    /* ASCII/UTF-8 coding */
+    if (c > 127)  /* Excludes all non-ASCII in either mode */
+      {
+      *errorcodeptr = ERR68;
+      break;
+      }
+    if (c >= CHAR_a && c <= CHAR_z) c -= 32;
+    c ^= 0x40;
+#else             /* EBCDIC coding */
+    if (c >= CHAR_a && c <= CHAR_z) c += 64;
+    c ^= 0xC0;
+#endif
+    break;
+
+    /* PCRE_EXTRA enables extensions to Perl in the matter of escapes. Any
+    other alphanumeric following \ is an error if PCRE_EXTRA was set;
+    otherwise, for Perl compatibility, it is a literal. This code looks a bit
+    odd, but there used to be some cases other than the default, and there may
+    be again in future, so I haven't "optimized" it. */
+
+    default:
+    if ((options & PCRE_EXTRA) != 0) switch(c)
+      {
+      default:
+      *errorcodeptr = ERR3;
+      break;
+      }
+    break;
+    }
+  }
+
+/* Perl supports \N{name} for character names, as well as plain \N for "not
+newline". PCRE does not support \N{name}. However, it does support
+quantification such as \N{2,3}. */
+
+if (escape == ESC_N && ptr[1] == CHAR_LEFT_CURLY_BRACKET &&
+     !is_counted_repeat(ptr+2))
+  *errorcodeptr = ERR37;
+
+/* If PCRE_UCP is set, we change the values for \d etc. */
+
+if ((options & PCRE_UCP) != 0 && escape >= ESC_D && escape <= ESC_w)
+  escape += (ESC_DU - ESC_D);
+
+/* Set the pointer to the final character before returning. */
+
+*ptrptr = ptr;
+*chptr = c;
+return escape;
+}
+
+#ifdef SUPPORT_UCP
+/*************************************************
+*               Handle \P and \p                 *
+*************************************************/
+
+/* This function is called after \P or \p has been encountered, provided that
+PCRE is compiled with support for Unicode properties. On entry, ptrptr is
+pointing at the P or p. On exit, it is pointing at the final character of the
+escape sequence.
+
+Argument:
+  ptrptr         points to the pattern position pointer
+  negptr         points to a boolean that is set TRUE for negation else FALSE
+  ptypeptr       points to an unsigned int that is set to the type value
+  pdataptr       points to an unsigned int that is set to the detailed property value
+  errorcodeptr   points to the error code variable
+
+Returns:         TRUE if the type value was found, or FALSE for an invalid type
+*/
+
+static BOOL
+get_ucp(const pcre_uchar **ptrptr, BOOL *negptr, unsigned int *ptypeptr,
+  unsigned int *pdataptr, int *errorcodeptr)
+{
+pcre_uchar c;
+int i, bot, top;
+const pcre_uchar *ptr = *ptrptr;
+pcre_uchar name[32];
+
+c = *(++ptr);
+if (c == CHAR_NULL) goto ERROR_RETURN;
+
+*negptr = FALSE;
+
+/* \P or \p can be followed by a name in {}, optionally preceded by ^ for
+negation. */
+
+if (c == CHAR_LEFT_CURLY_BRACKET)
+  {
+  if (ptr[1] == CHAR_CIRCUMFLEX_ACCENT)
+    {
+    *negptr = TRUE;
+    ptr++;
+    }
+  for (i = 0; i < (int)(sizeof(name) / sizeof(pcre_uchar)) - 1; i++)
+    {
+    c = *(++ptr);
+    if (c == CHAR_NULL) goto ERROR_RETURN;
+    if (c == CHAR_RIGHT_CURLY_BRACKET) break;
+    name[i] = c;
+    }
+  if (c != CHAR_RIGHT_CURLY_BRACKET) goto ERROR_RETURN;
+  name[i] = 0;
+  }
+
+/* Otherwise there is just one following character */
+
+else
+  {
+  name[0] = c;
+  name[1] = 0;
+  }
+
+*ptrptr = ptr;
+
+/* Search for a recognized property name using binary chop */
+
+bot = 0;
+top = PRIV(utt_size);
+
+while (bot < top)
+  {
+  int r;
+  i = (bot + top) >> 1;
+  r = STRCMP_UC_C8(name, PRIV(utt_names) + PRIV(utt)[i].name_offset);
+  if (r == 0)
+    {
+    *ptypeptr = PRIV(utt)[i].type;
+    *pdataptr = PRIV(utt)[i].value;
+    return TRUE;
+    }
+  if (r > 0) bot = i + 1; else top = i;
+  }
+
+*errorcodeptr = ERR47;
+*ptrptr = ptr;
+return FALSE;
+
+ERROR_RETURN:
+*errorcodeptr = ERR46;
+*ptrptr = ptr;
+return FALSE;
+}
+#endif
+
+
+
+
+/*************************************************
+*         Read repeat counts                     *
+*************************************************/
+
+/* Read an item of the form {n,m} and return the values. This is called only
+after is_counted_repeat() has confirmed that a repeat-count quantifier exists,
+so the syntax is guaranteed to be correct, but we need to check the values.
+
+Arguments:
+  p              pointer to first char after '{'
+  minp           pointer to int for min
+  maxp           pointer to int for max
+                 returned as -1 if no max
+  errorcodeptr   points to error code variable
+
+Returns:         pointer to '}' on success;
+                 current ptr on error, with errorcodeptr set non-zero
+*/
+
+static const pcre_uchar *
+read_repeat_counts(const pcre_uchar *p, int *minp, int *maxp, int *errorcodeptr)
+{
+int min = 0;
+int max = -1;
+
+/* Read the minimum value and do a paranoid check: a negative value indicates
+an integer overflow. */
+
+while (IS_DIGIT(*p)) min = min * 10 + (int)(*p++ - CHAR_0);
+if (min < 0 || min > 65535)
+  {
+  *errorcodeptr = ERR5;
+  return p;
+  }
+
+/* Read the maximum value if there is one, and again do a paranoid on its size.
+Also, max must not be less than min. */
+
+if (*p == CHAR_RIGHT_CURLY_BRACKET) max = min; else
+  {
+  if (*(++p) != CHAR_RIGHT_CURLY_BRACKET)
+    {
+    max = 0;
+    while(IS_DIGIT(*p)) max = max * 10 + (int)(*p++ - CHAR_0);
+    if (max < 0 || max > 65535)
+      {
+      *errorcodeptr = ERR5;
+      return p;
+      }
+    if (max < min)
+      {
+      *errorcodeptr = ERR4;
+      return p;
+      }
+    }
+  }
+
+/* Fill in the required variables, and pass back the pointer to the terminating
+'}'. */
+
+*minp = min;
+*maxp = max;
+return p;
+}
+
+
+
+/*************************************************
+*  Subroutine for finding forward reference      *
+*************************************************/
+
+/* This recursive function is called only from find_parens() below. The
+top-level call starts at the beginning of the pattern. All other calls must
+start at a parenthesis. It scans along a pattern's text looking for capturing
+subpatterns, and counting them. If it finds a named pattern that matches the
+name it is given, it returns its number. Alternatively, if the name is NULL, it
+returns when it reaches a given numbered subpattern. Recursion is used to keep
+track of subpatterns that reset the capturing group numbers - the (?| feature.
+
+This function was originally called only from the second pass, in which we know
+that if (?< or (?' or (?P< is encountered, the name will be correctly
+terminated because that is checked in the first pass. There is now one call to
+this function in the first pass, to check for a recursive back reference by
+name (so that we can make the whole group atomic). In this case, we need check
+only up to the current position in the pattern, and that is still OK because
+and previous occurrences will have been checked. To make this work, the test
+for "end of pattern" is a check against cd->end_pattern in the main loop,
+instead of looking for a binary zero. This means that the special first-pass
+call can adjust cd->end_pattern temporarily. (Checks for binary zero while
+processing items within the loop are OK, because afterwards the main loop will
+terminate.)
+
+Arguments:
+  ptrptr       address of the current character pointer (updated)
+  cd           compile background data
+  name         name to seek, or NULL if seeking a numbered subpattern
+  lorn         name length, or subpattern number if name is NULL
+  xmode        TRUE if we are in /x mode
+  utf          TRUE if we are in UTF-8 / UTF-16 / UTF-32 mode
+  count        pointer to the current capturing subpattern number (updated)
+
+Returns:       the number of the named subpattern, or -1 if not found
+*/
+
+static int
+find_parens_sub(pcre_uchar **ptrptr, compile_data *cd, const pcre_uchar *name, int lorn,
+  BOOL xmode, BOOL utf, int *count)
+{
+pcre_uchar *ptr = *ptrptr;
+int start_count = *count;
+int hwm_count = start_count;
+BOOL dup_parens = FALSE;
+
+/* If the first character is a parenthesis, check on the type of group we are
+dealing with. The very first call may not start with a parenthesis. */
+
+if (ptr[0] == CHAR_LEFT_PARENTHESIS)
+  {
+  /* Handle specials such as (*SKIP) or (*UTF8) etc. */
+
+  if (ptr[1] == CHAR_ASTERISK)
+    {
+    ptr += 2;
+    while (ptr < cd->end_pattern && *ptr != CHAR_RIGHT_PARENTHESIS) ptr++;
+    }
+
+  /* Handle a normal, unnamed capturing parenthesis. */
+
+  else if (ptr[1] != CHAR_QUESTION_MARK)
+    {
+    *count += 1;
+    if (name == NULL && *count == lorn) return *count;
+    ptr++;
+    }
+
+  /* All cases now have (? at the start. Remember when we are in a group
+  where the parenthesis numbers are duplicated. */
+
+  else if (ptr[2] == CHAR_VERTICAL_LINE)
+    {
+    ptr += 3;
+    dup_parens = TRUE;
+    }
+
+  /* Handle comments; all characters are allowed until a ket is reached. */
+
+  else if (ptr[2] == CHAR_NUMBER_SIGN)
+    {
+    for (ptr += 3; *ptr != CHAR_NULL; ptr++)
+      if (*ptr == CHAR_RIGHT_PARENTHESIS) break;
+    goto FAIL_EXIT;
+    }
+
+  /* Handle a condition. If it is an assertion, just carry on so that it
+  is processed as normal. If not, skip to the closing parenthesis of the
+  condition (there can't be any nested parens). */
+
+  else if (ptr[2] == CHAR_LEFT_PARENTHESIS)
+    {
+    ptr += 2;
+    if (ptr[1] != CHAR_QUESTION_MARK)
+      {
+      while (*ptr != CHAR_NULL && *ptr != CHAR_RIGHT_PARENTHESIS) ptr++;
+      if (*ptr != CHAR_NULL) ptr++;
+      }
+    }
+
+  /* Start with (? but not a condition. */
+
+  else
+    {
+    ptr += 2;
+    if (*ptr == CHAR_P) ptr++;                      /* Allow optional P */
+
+    /* We have to disambiguate (? for named groups */
+
+    if ((*ptr == CHAR_LESS_THAN_SIGN && ptr[1] != CHAR_EXCLAMATION_MARK &&
+        ptr[1] != CHAR_EQUALS_SIGN) || *ptr == CHAR_APOSTROPHE)
+      {
+      pcre_uchar term;
+      const pcre_uchar *thisname;
+      *count += 1;
+      if (name == NULL && *count == lorn) return *count;
+      term = *ptr++;
+      if (term == CHAR_LESS_THAN_SIGN) term = CHAR_GREATER_THAN_SIGN;
+      thisname = ptr;
+      while (*ptr != term) ptr++;
+      if (name != NULL && lorn == (int)(ptr - thisname) &&
+          STRNCMP_UC_UC(name, thisname, (unsigned int)lorn) == 0)
+        return *count;
+      term++;
+      }
+    }
+  }
+
+/* Past any initial parenthesis handling, scan for parentheses or vertical
+bars. Stop if we get to cd->end_pattern. Note that this is important for the
+first-pass call when this value is temporarily adjusted to stop at the current
+position. So DO NOT change this to a test for binary zero. */
+
+for (; ptr < cd->end_pattern; ptr++)
+  {
+  /* Skip over backslashed characters and also entire \Q...\E */
+
+  if (*ptr == CHAR_BACKSLASH)
+    {
+    if (*(++ptr) == CHAR_NULL) goto FAIL_EXIT;
+    if (*ptr == CHAR_Q) for (;;)
+      {
+      while (*(++ptr) != CHAR_NULL && *ptr != CHAR_BACKSLASH) {};
+      if (*ptr == CHAR_NULL) goto FAIL_EXIT;
+      if (*(++ptr) == CHAR_E) break;
+      }
+    continue;
+    }
+
+  /* Skip over character classes; this logic must be similar to the way they
+  are handled for real. If the first character is '^', skip it. Also, if the
+  first few characters (either before or after ^) are \Q\E or \E we skip them
+  too. This makes for compatibility with Perl. Note the use of STR macros to
+  encode "Q\\E" so that it works in UTF-8 on EBCDIC platforms. */
+
+  if (*ptr == CHAR_LEFT_SQUARE_BRACKET)
+    {
+    BOOL negate_class = FALSE;
+    for (;;)
+      {
+      if (ptr[1] == CHAR_BACKSLASH)
+        {
+        if (ptr[2] == CHAR_E)
+          ptr+= 2;
+        else if (STRNCMP_UC_C8(ptr + 2,
+                 STR_Q STR_BACKSLASH STR_E, 3) == 0)
+          ptr += 4;
+        else
+          break;
+        }
+      else if (!negate_class && ptr[1] == CHAR_CIRCUMFLEX_ACCENT)
+        {
+        negate_class = TRUE;
+        ptr++;
+        }
+      else break;
+      }
+
+    /* If the next character is ']', it is a data character that must be
+    skipped, except in JavaScript compatibility mode. */
+
+    if (ptr[1] == CHAR_RIGHT_SQUARE_BRACKET &&
+        (cd->external_options & PCRE_JAVASCRIPT_COMPAT) == 0)
+      ptr++;
+
+    while (*(++ptr) != CHAR_RIGHT_SQUARE_BRACKET)
+      {
+      if (*ptr == CHAR_NULL) return -1;
+      if (*ptr == CHAR_BACKSLASH)
+        {
+        if (*(++ptr) == CHAR_NULL) goto FAIL_EXIT;
+        if (*ptr == CHAR_Q) for (;;)
+          {
+          while (*(++ptr) != CHAR_NULL && *ptr != CHAR_BACKSLASH) {};
+          if (*ptr == CHAR_NULL) goto FAIL_EXIT;
+          if (*(++ptr) == CHAR_E) break;
+          }
+        continue;
+        }
+      }
+    continue;
+    }
+
+  /* Skip comments in /x mode */
+
+  if (xmode && *ptr == CHAR_NUMBER_SIGN)
+    {
+    ptr++;
+    while (*ptr != CHAR_NULL)
+      {
+      if (IS_NEWLINE(ptr)) { ptr += cd->nllen - 1; break; }
+      ptr++;
+#ifdef SUPPORT_UTF
+      if (utf) FORWARDCHAR(ptr);
+#endif
+      }
+    if (*ptr == CHAR_NULL) goto FAIL_EXIT;
+    continue;
+    }
+
+  /* Check for the special metacharacters */
+
+  if (*ptr == CHAR_LEFT_PARENTHESIS)
+    {
+    int rc = find_parens_sub(&ptr, cd, name, lorn, xmode, utf, count);
+    if (rc > 0) return rc;
+    if (*ptr == CHAR_NULL) goto FAIL_EXIT;
+    }
+
+  else if (*ptr == CHAR_RIGHT_PARENTHESIS)
+    {
+    if (dup_parens && *count < hwm_count) *count = hwm_count;
+    goto FAIL_EXIT;
+    }
+
+  else if (*ptr == CHAR_VERTICAL_LINE && dup_parens)
+    {
+    if (*count > hwm_count) hwm_count = *count;
+    *count = start_count;
+    }
+  }
+
+FAIL_EXIT:
+*ptrptr = ptr;
+return -1;
+}
+
+
+
+
+/*************************************************
+*       Find forward referenced subpattern       *
+*************************************************/
+
+/* This function scans along a pattern's text looking for capturing
+subpatterns, and counting them. If it finds a named pattern that matches the
+name it is given, it returns its number. Alternatively, if the name is NULL, it
+returns when it reaches a given numbered subpattern. This is used for forward
+references to subpatterns. We used to be able to start this scan from the
+current compiling point, using the current count value from cd->bracount, and
+do it all in a single loop, but the addition of the possibility of duplicate
+subpattern numbers means that we have to scan from the very start, in order to
+take account of such duplicates, and to use a recursive function to keep track
+of the different types of group.
+
+Arguments:
+  cd           compile background data
+  name         name to seek, or NULL if seeking a numbered subpattern
+  lorn         name length, or subpattern number if name is NULL
+  xmode        TRUE if we are in /x mode
+  utf          TRUE if we are in UTF-8 / UTF-16 / UTF-32 mode
+
+Returns:       the number of the found subpattern, or -1 if not found
+*/
+
+static int
+find_parens(compile_data *cd, const pcre_uchar *name, int lorn, BOOL xmode,
+  BOOL utf)
+{
+pcre_uchar *ptr = (pcre_uchar *)cd->start_pattern;
+int count = 0;
+int rc;
+
+/* If the pattern does not start with an opening parenthesis, the first call
+to find_parens_sub() will scan right to the end (if necessary). However, if it
+does start with a parenthesis, find_parens_sub() will return when it hits the
+matching closing parens. That is why we have to have a loop. */
+
+for (;;)
+  {
+  rc = find_parens_sub(&ptr, cd, name, lorn, xmode, utf, &count);
+  if (rc > 0 || *ptr++ == CHAR_NULL) break;
+  }
+
+return rc;
+}
+
+
+
+
+/*************************************************
+*      Find first significant op code            *
+*************************************************/
+
+/* This is called by several functions that scan a compiled expression looking
+for a fixed first character, or an anchoring op code etc. It skips over things
+that do not influence this. For some calls, it makes sense to skip negative
+forward and all backward assertions, and also the \b assertion; for others it
+does not.
+
+Arguments:
+  code         pointer to the start of the group
+  skipassert   TRUE if certain assertions are to be skipped
+
+Returns:       pointer to the first significant opcode
+*/
+
+static const pcre_uchar*
+first_significant_code(const pcre_uchar *code, BOOL skipassert)
+{
+for (;;)
+  {
+  switch ((int)*code)
+    {
+    case OP_ASSERT_NOT:
+    case OP_ASSERTBACK:
+    case OP_ASSERTBACK_NOT:
+    if (!skipassert) return code;
+    do code += GET(code, 1); while (*code == OP_ALT);
+    code += PRIV(OP_lengths)[*code];
+    break;
+
+    case OP_WORD_BOUNDARY:
+    case OP_NOT_WORD_BOUNDARY:
+    if (!skipassert) return code;
+    /* Fall through */
+
+    case OP_CALLOUT:
+    case OP_CREF:
+    case OP_NCREF:
+    case OP_RREF:
+    case OP_NRREF:
+    case OP_DEF:
+    code += PRIV(OP_lengths)[*code];
+    break;
+
+    default:
+    return code;
+    }
+  }
+/* Control never reaches here */
+}
+
+
+
+
+/*************************************************
+*        Find the fixed length of a branch       *
+*************************************************/
+
+/* Scan a branch and compute the fixed length of subject that will match it,
+if the length is fixed. This is needed for dealing with backward assertions.
+In UTF8 mode, the result is in characters rather than bytes. The branch is
+temporarily terminated with OP_END when this function is called.
+
+This function is called when a backward assertion is encountered, so that if it
+fails, the error message can point to the correct place in the pattern.
+However, we cannot do this when the assertion contains subroutine calls,
+because they can be forward references. We solve this by remembering this case
+and doing the check at the end; a flag specifies which mode we are running in.
+
+Arguments:
+  code     points to the start of the pattern (the bracket)
+  utf      TRUE in UTF-8 / UTF-16 / UTF-32 mode
+  atend    TRUE if called when the pattern is complete
+  cd       the "compile data" structure
+
+Returns:   the fixed length,
+             or -1 if there is no fixed length,
+             or -2 if \C was encountered (in UTF-8 mode only)
+             or -3 if an OP_RECURSE item was encountered and atend is FALSE
+             or -4 if an unknown opcode was encountered (internal error)
+*/
+
+static int
+find_fixedlength(pcre_uchar *code, BOOL utf, BOOL atend, compile_data *cd)
+{
+int length = -1;
+
+register int branchlength = 0;
+register pcre_uchar *cc = code + 1 + LINK_SIZE;
+
+/* Scan along the opcodes for this branch. If we get to the end of the
+branch, check the length against that of the other branches. */
+
+for (;;)
+  {
+  int d;
+  pcre_uchar *ce, *cs;
+  register pcre_uchar op = *cc;
+
+  switch (op)
+    {
+    /* We only need to continue for OP_CBRA (normal capturing bracket) and
+    OP_BRA (normal non-capturing bracket) because the other variants of these
+    opcodes are all concerned with unlimited repeated groups, which of course
+    are not of fixed length. */
+
+    case OP_CBRA:
+    case OP_BRA:
+    case OP_ONCE:
+    case OP_ONCE_NC:
+    case OP_COND:
+    d = find_fixedlength(cc + ((op == OP_CBRA)? IMM2_SIZE : 0), utf, atend, cd);
+    if (d < 0) return d;
+    branchlength += d;
+    do cc += GET(cc, 1); while (*cc == OP_ALT);
+    cc += 1 + LINK_SIZE;
+    break;
+
+    /* Reached end of a branch; if it's a ket it is the end of a nested call.
+    If it's ALT it is an alternation in a nested call. An ACCEPT is effectively
+    an ALT. If it is END it's the end of the outer call. All can be handled by
+    the same code. Note that we must not include the OP_KETRxxx opcodes here,
+    because they all imply an unlimited repeat. */
+
+    case OP_ALT:
+    case OP_KET:
+    case OP_END:
+    case OP_ACCEPT:
+    case OP_ASSERT_ACCEPT:
+    if (length < 0) length = branchlength;
+      else if (length != branchlength) return -1;
+    if (*cc != OP_ALT) return length;
+    cc += 1 + LINK_SIZE;
+    branchlength = 0;
+    break;
+
+    /* A true recursion implies not fixed length, but a subroutine call may
+    be OK. If the subroutine is a forward reference, we can't deal with
+    it until the end of the pattern, so return -3. */
+
+    case OP_RECURSE:
+    if (!atend) return -3;
+    cs = ce = (pcre_uchar *)cd->start_code + GET(cc, 1);  /* Start subpattern */
+    do ce += GET(ce, 1); while (*ce == OP_ALT);           /* End subpattern */
+    if (cc > cs && cc < ce) return -1;                    /* Recursion */
+    d = find_fixedlength(cs + IMM2_SIZE, utf, atend, cd);
+    if (d < 0) return d;
+    branchlength += d;
+    cc += 1 + LINK_SIZE;
+    break;
+
+    /* Skip over assertive subpatterns */
+
+    case OP_ASSERT:
+    case OP_ASSERT_NOT:
+    case OP_ASSERTBACK:
+    case OP_ASSERTBACK_NOT:
+    do cc += GET(cc, 1); while (*cc == OP_ALT);
+    cc += PRIV(OP_lengths)[*cc];
+    break;
+
+    /* Skip over things that don't match chars */
+
+    case OP_MARK:
+    case OP_PRUNE_ARG:
+    case OP_SKIP_ARG:
+    case OP_THEN_ARG:
+    cc += cc[1] + PRIV(OP_lengths)[*cc];
+    break;
+
+    case OP_CALLOUT:
+    case OP_CIRC:
+    case OP_CIRCM:
+    case OP_CLOSE:
+    case OP_COMMIT:
+    case OP_CREF:
+    case OP_DEF:
+    case OP_DOLL:
+    case OP_DOLLM:
+    case OP_EOD:
+    case OP_EODN:
+    case OP_FAIL:
+    case OP_NCREF:
+    case OP_NRREF:
+    case OP_NOT_WORD_BOUNDARY:
+    case OP_PRUNE:
+    case OP_REVERSE:
+    case OP_RREF:
+    case OP_SET_SOM:
+    case OP_SKIP:
+    case OP_SOD:
+    case OP_SOM:
+    case OP_THEN:
+    case OP_WORD_BOUNDARY:
+    cc += PRIV(OP_lengths)[*cc];
+    break;
+
+    /* Handle literal characters */
+
+    case OP_CHAR:
+    case OP_CHARI:
+    case OP_NOT:
+    case OP_NOTI:
+    branchlength++;
+    cc += 2;
+#ifdef SUPPORT_UTF
+    if (utf && HAS_EXTRALEN(cc[-1])) cc += GET_EXTRALEN(cc[-1]);
+#endif
+    break;
+
+    /* Handle exact repetitions. The count is already in characters, but we
+    need to skip over a multibyte character in UTF8 mode.  */
+
+    case OP_EXACT:
+    case OP_EXACTI:
+    case OP_NOTEXACT:
+    case OP_NOTEXACTI:
+    branchlength += (int)GET2(cc,1);
+    cc += 2 + IMM2_SIZE;
+#ifdef SUPPORT_UTF
+    if (utf && HAS_EXTRALEN(cc[-1])) cc += GET_EXTRALEN(cc[-1]);
+#endif
+    break;
+
+    case OP_TYPEEXACT:
+    branchlength += GET2(cc,1);
+    if (cc[1 + IMM2_SIZE] == OP_PROP || cc[1 + IMM2_SIZE] == OP_NOTPROP)
+      cc += 2;
+    cc += 1 + IMM2_SIZE + 1;
+    break;
+
+    /* Handle single-char matchers */
+
+    case OP_PROP:
+    case OP_NOTPROP:
+    cc += 2;
+    /* Fall through */
+
+    case OP_HSPACE:
+    case OP_VSPACE:
+    case OP_NOT_HSPACE:
+    case OP_NOT_VSPACE:
+    case OP_NOT_DIGIT:
+    case OP_DIGIT:
+    case OP_NOT_WHITESPACE:
+    case OP_WHITESPACE:
+    case OP_NOT_WORDCHAR:
+    case OP_WORDCHAR:
+    case OP_ANY:
+    case OP_ALLANY:
+    branchlength++;
+    cc++;
+    break;
+
+    /* The single-byte matcher isn't allowed. This only happens in UTF-8 mode;
+    otherwise \C is coded as OP_ALLANY. */
+
+    case OP_ANYBYTE:
+    return -2;
+
+    /* Check a class for variable quantification */
+
+    case OP_CLASS:
+    case OP_NCLASS:
+#if defined SUPPORT_UTF || defined COMPILE_PCRE16 || defined COMPILE_PCRE32
+    case OP_XCLASS:
+    /* The original code caused an unsigned overflow in 64 bit systems,
+    so now we use a conditional statement. */
+    if (op == OP_XCLASS)
+      cc += GET(cc, 1);
+    else
+      cc += PRIV(OP_lengths)[OP_CLASS];
+#else
+    cc += PRIV(OP_lengths)[OP_CLASS];
+#endif
+
+    switch (*cc)
+      {
+      case OP_CRPLUS:
+      case OP_CRMINPLUS:
+      case OP_CRSTAR:
+      case OP_CRMINSTAR:
+      case OP_CRQUERY:
+      case OP_CRMINQUERY:
+      return -1;
+
+      case OP_CRRANGE:
+      case OP_CRMINRANGE:
+      if (GET2(cc,1) != GET2(cc,1+IMM2_SIZE)) return -1;
+      branchlength += (int)GET2(cc,1);
+      cc += 1 + 2 * IMM2_SIZE;
+      break;
+
+      default:
+      branchlength++;
+      }
+    break;
+
+    /* Anything else is variable length */
+
+    case OP_ANYNL:
+    case OP_BRAMINZERO:
+    case OP_BRAPOS:
+    case OP_BRAPOSZERO:
+    case OP_BRAZERO:
+    case OP_CBRAPOS:
+    case OP_EXTUNI:
+    case OP_KETRMAX:
+    case OP_KETRMIN:
+    case OP_KETRPOS:
+    case OP_MINPLUS:
+    case OP_MINPLUSI:
+    case OP_MINQUERY:
+    case OP_MINQUERYI:
+    case OP_MINSTAR:
+    case OP_MINSTARI:
+    case OP_MINUPTO:
+    case OP_MINUPTOI:
+    case OP_NOTMINPLUS:
+    case OP_NOTMINPLUSI:
+    case OP_NOTMINQUERY:
+    case OP_NOTMINQUERYI:
+    case OP_NOTMINSTAR:
+    case OP_NOTMINSTARI:
+    case OP_NOTMINUPTO:
+    case OP_NOTMINUPTOI:
+    case OP_NOTPLUS:
+    case OP_NOTPLUSI:
+    case OP_NOTPOSPLUS:
+    case OP_NOTPOSPLUSI:
+    case OP_NOTPOSQUERY:
+    case OP_NOTPOSQUERYI:
+    case OP_NOTPOSSTAR:
+    case OP_NOTPOSSTARI:
+    case OP_NOTPOSUPTO:
+    case OP_NOTPOSUPTOI:
+    case OP_NOTQUERY:
+    case OP_NOTQUERYI:
+    case OP_NOTSTAR:
+    case OP_NOTSTARI:
+    case OP_NOTUPTO:
+    case OP_NOTUPTOI:
+    case OP_PLUS:
+    case OP_PLUSI:
+    case OP_POSPLUS:
+    case OP_POSPLUSI:
+    case OP_POSQUERY:
+    case OP_POSQUERYI:
+    case OP_POSSTAR:
+    case OP_POSSTARI:
+    case OP_POSUPTO:
+    case OP_POSUPTOI:
+    case OP_QUERY:
+    case OP_QUERYI:
+    case OP_REF:
+    case OP_REFI:
+    case OP_SBRA:
+    case OP_SBRAPOS:
+    case OP_SCBRA:
+    case OP_SCBRAPOS:
+    case OP_SCOND:
+    case OP_SKIPZERO:
+    case OP_STAR:
+    case OP_STARI:
+    case OP_TYPEMINPLUS:
+    case OP_TYPEMINQUERY:
+    case OP_TYPEMINSTAR:
+    case OP_TYPEMINUPTO:
+    case OP_TYPEPLUS:
+    case OP_TYPEPOSPLUS:
+    case OP_TYPEPOSQUERY:
+    case OP_TYPEPOSSTAR:
+    case OP_TYPEPOSUPTO:
+    case OP_TYPEQUERY:
+    case OP_TYPESTAR:
+    case OP_TYPEUPTO:
+    case OP_UPTO:
+    case OP_UPTOI:
+    return -1;
+
+    /* Catch unrecognized opcodes so that when new ones are added they
+    are not forgotten, as has happened in the past. */
+
+    default:
+    return -4;
+    }
+  }
+/* Control never gets here */
+}
+
+
+
+
+/*************************************************
+*    Scan compiled regex for specific bracket    *
+*************************************************/
+
+/* This little function scans through a compiled pattern until it finds a
+capturing bracket with the given number, or, if the number is negative, an
+instance of OP_REVERSE for a lookbehind. The function is global in the C sense
+so that it can be called from pcre_study() when finding the minimum matching
+length.
+
+Arguments:
+  code        points to start of expression
+  utf         TRUE in UTF-8 / UTF-16 / UTF-32 mode
+  number      the required bracket number or negative to find a lookbehind
+
+Returns:      pointer to the opcode for the bracket, or NULL if not found
+*/
+
+const pcre_uchar *
+PRIV(find_bracket)(const pcre_uchar *code, BOOL utf, int number)
+{
+for (;;)
+  {
+  register pcre_uchar c = *code;
+
+  if (c == OP_END) return NULL;
+
+  /* XCLASS is used for classes that cannot be represented just by a bit
+  map. This includes negated single high-valued characters. The length in
+  the table is zero; the actual length is stored in the compiled code. */
+
+  if (c == OP_XCLASS) code += GET(code, 1);
+
+  /* Handle recursion */
+
+  else if (c == OP_REVERSE)
+    {
+    if (number < 0) return (pcre_uchar *)code;
+    code += PRIV(OP_lengths)[c];
+    }
+
+  /* Handle capturing bracket */
+
+  else if (c == OP_CBRA || c == OP_SCBRA ||
+           c == OP_CBRAPOS || c == OP_SCBRAPOS)
+    {
+    int n = (int)GET2(code, 1+LINK_SIZE);
+    if (n == number) return (pcre_uchar *)code;
+    code += PRIV(OP_lengths)[c];
+    }
+
+  /* Otherwise, we can get the item's length from the table, except that for
+  repeated character types, we have to test for \p and \P, which have an extra
+  two bytes of parameters, and for MARK/PRUNE/SKIP/THEN with an argument, we
+  must add in its length. */
+
+  else
+    {
+    switch(c)
+      {
+      case OP_TYPESTAR:
+      case OP_TYPEMINSTAR:
+      case OP_TYPEPLUS:
+      case OP_TYPEMINPLUS:
+      case OP_TYPEQUERY:
+      case OP_TYPEMINQUERY:
+      case OP_TYPEPOSSTAR:
+      case OP_TYPEPOSPLUS:
+      case OP_TYPEPOSQUERY:
+      if (code[1] == OP_PROP || code[1] == OP_NOTPROP) code += 2;
+      break;
+
+      case OP_TYPEUPTO:
+      case OP_TYPEMINUPTO:
+      case OP_TYPEEXACT:
+      case OP_TYPEPOSUPTO:
+      if (code[1 + IMM2_SIZE] == OP_PROP || code[1 + IMM2_SIZE] == OP_NOTPROP)
+        code += 2;
+      break;
+
+      case OP_MARK:
+      case OP_PRUNE_ARG:
+      case OP_SKIP_ARG:
+      case OP_THEN_ARG:
+      code += code[1];
+      break;
+      }
+
+    /* Add in the fixed length from the table */
+
+    code += PRIV(OP_lengths)[c];
+
+  /* In UTF-8 mode, opcodes that are followed by a character may be followed by
+  a multi-byte character. The length in the table is a minimum, so we have to
+  arrange to skip the extra bytes. */
+
+#if defined SUPPORT_UTF && !defined COMPILE_PCRE32
+    if (utf) switch(c)
+      {
+      case OP_CHAR:
+      case OP_CHARI:
+      case OP_EXACT:
+      case OP_EXACTI:
+      case OP_UPTO:
+      case OP_UPTOI:
+      case OP_MINUPTO:
+      case OP_MINUPTOI:
+      case OP_POSUPTO:
+      case OP_POSUPTOI:
+      case OP_STAR:
+      case OP_STARI:
+      case OP_MINSTAR:
+      case OP_MINSTARI:
+      case OP_POSSTAR:
+      case OP_POSSTARI:
+      case OP_PLUS:
+      case OP_PLUSI:
+      case OP_MINPLUS:
+      case OP_MINPLUSI:
+      case OP_POSPLUS:
+      case OP_POSPLUSI:
+      case OP_QUERY:
+      case OP_QUERYI:
+      case OP_MINQUERY:
+      case OP_MINQUERYI:
+      case OP_POSQUERY:
+      case OP_POSQUERYI:
+      if (HAS_EXTRALEN(code[-1])) code += GET_EXTRALEN(code[-1]);
+      break;
+      }
+#else
+    (void)(utf);  /* Keep compiler happy by referencing function argument */
+#endif
+    }
+  }
+}
+
+
+
+/*************************************************
+*   Scan compiled regex for recursion reference  *
+*************************************************/
+
+/* This little function scans through a compiled pattern until it finds an
+instance of OP_RECURSE.
+
+Arguments:
+  code        points to start of expression
+  utf         TRUE in UTF-8 / UTF-16 / UTF-32 mode
+
+Returns:      pointer to the opcode for OP_RECURSE, or NULL if not found
+*/
+
+static const pcre_uchar *
+find_recurse(const pcre_uchar *code, BOOL utf)
+{
+for (;;)
+  {
+  register pcre_uchar c = *code;
+  if (c == OP_END) return NULL;
+  if (c == OP_RECURSE) return code;
+
+  /* XCLASS is used for classes that cannot be represented just by a bit
+  map. This includes negated single high-valued characters. The length in
+  the table is zero; the actual length is stored in the compiled code. */
+
+  if (c == OP_XCLASS) code += GET(code, 1);
+
+  /* Otherwise, we can get the item's length from the table, except that for
+  repeated character types, we have to test for \p and \P, which have an extra
+  two bytes of parameters, and for MARK/PRUNE/SKIP/THEN with an argument, we
+  must add in its length. */
+
+  else
+    {
+    switch(c)
+      {
+      case OP_TYPESTAR:
+      case OP_TYPEMINSTAR:
+      case OP_TYPEPLUS:
+      case OP_TYPEMINPLUS:
+      case OP_TYPEQUERY:
+      case OP_TYPEMINQUERY:
+      case OP_TYPEPOSSTAR:
+      case OP_TYPEPOSPLUS:
+      case OP_TYPEPOSQUERY:
+      if (code[1] == OP_PROP || code[1] == OP_NOTPROP) code += 2;
+      break;
+
+      case OP_TYPEPOSUPTO:
+      case OP_TYPEUPTO:
+      case OP_TYPEMINUPTO:
+      case OP_TYPEEXACT:
+      if (code[1 + IMM2_SIZE] == OP_PROP || code[1 + IMM2_SIZE] == OP_NOTPROP)
+        code += 2;
+      break;
+
+      case OP_MARK:
+      case OP_PRUNE_ARG:
+      case OP_SKIP_ARG:
+      case OP_THEN_ARG:
+      code += code[1];
+      break;
+      }
+
+    /* Add in the fixed length from the table */
+
+    code += PRIV(OP_lengths)[c];
+
+    /* In UTF-8 mode, opcodes that are followed by a character may be followed
+    by a multi-byte character. The length in the table is a minimum, so we have
+    to arrange to skip the extra bytes. */
+
+#if defined SUPPORT_UTF && !defined COMPILE_PCRE32
+    if (utf) switch(c)
+      {
+      case OP_CHAR:
+      case OP_CHARI:
+      case OP_NOT:
+      case OP_NOTI:
+      case OP_EXACT:
+      case OP_EXACTI:
+      case OP_NOTEXACT:
+      case OP_NOTEXACTI:
+      case OP_UPTO:
+      case OP_UPTOI:
+      case OP_NOTUPTO:
+      case OP_NOTUPTOI:
+      case OP_MINUPTO:
+      case OP_MINUPTOI:
+      case OP_NOTMINUPTO:
+      case OP_NOTMINUPTOI:
+      case OP_POSUPTO:
+      case OP_POSUPTOI:
+      case OP_NOTPOSUPTO:
+      case OP_NOTPOSUPTOI:
+      case OP_STAR:
+      case OP_STARI:
+      case OP_NOTSTAR:
+      case OP_NOTSTARI:
+      case OP_MINSTAR:
+      case OP_MINSTARI:
+      case OP_NOTMINSTAR:
+      case OP_NOTMINSTARI:
+      case OP_POSSTAR:
+      case OP_POSSTARI:
+      case OP_NOTPOSSTAR:
+      case OP_NOTPOSSTARI:
+      case OP_PLUS:
+      case OP_PLUSI:
+      case OP_NOTPLUS:
+      case OP_NOTPLUSI:
+      case OP_MINPLUS:
+      case OP_MINPLUSI:
+      case OP_NOTMINPLUS:
+      case OP_NOTMINPLUSI:
+      case OP_POSPLUS:
+      case OP_POSPLUSI:
+      case OP_NOTPOSPLUS:
+      case OP_NOTPOSPLUSI:
+      case OP_QUERY:
+      case OP_QUERYI:
+      case OP_NOTQUERY:
+      case OP_NOTQUERYI:
+      case OP_MINQUERY:
+      case OP_MINQUERYI:
+      case OP_NOTMINQUERY:
+      case OP_NOTMINQUERYI:
+      case OP_POSQUERY:
+      case OP_POSQUERYI:
+      case OP_NOTPOSQUERY:
+      case OP_NOTPOSQUERYI:
+      if (HAS_EXTRALEN(code[-1])) code += GET_EXTRALEN(code[-1]);
+      break;
+      }
+#else
+    (void)(utf);  /* Keep compiler happy by referencing function argument */
+#endif
+    }
+  }
+}
+
+
+
+/*************************************************
+*    Scan compiled branch for non-emptiness      *
+*************************************************/
+
+/* This function scans through a branch of a compiled pattern to see whether it
+can match the empty string or not. It is called from could_be_empty()
+below and from compile_branch() when checking for an unlimited repeat of a
+group that can match nothing. Note that first_significant_code() skips over
+backward and negative forward assertions when its final argument is TRUE. If we
+hit an unclosed bracket, we return "empty" - this means we've struck an inner
+bracket whose current branch will already have been scanned.
+
+Arguments:
+  code        points to start of search
+  endcode     points to where to stop
+  utf         TRUE if in UTF-8 / UTF-16 / UTF-32 mode
+  cd          contains pointers to tables etc.
+
+Returns:      TRUE if what is matched could be empty
+*/
+
+static BOOL
+could_be_empty_branch(const pcre_uchar *code, const pcre_uchar *endcode,
+  BOOL utf, compile_data *cd)
+{
+register pcre_uchar c;
+for (code = first_significant_code(code + PRIV(OP_lengths)[*code], TRUE);
+     code < endcode;
+     code = first_significant_code(code + PRIV(OP_lengths)[c], TRUE))
+  {
+  const pcre_uchar *ccode;
+
+  c = *code;
+
+  /* Skip over forward assertions; the other assertions are skipped by
+  first_significant_code() with a TRUE final argument. */
+
+  if (c == OP_ASSERT)
+    {
+    do code += GET(code, 1); while (*code == OP_ALT);
+    c = *code;
+    continue;
+    }
+
+  /* For a recursion/subroutine call, if its end has been reached, which
+  implies a backward reference subroutine call, we can scan it. If it's a
+  forward reference subroutine call, we can't. To detect forward reference
+  we have to scan up the list that is kept in the workspace. This function is
+  called only when doing the real compile, not during the pre-compile that
+  measures the size of the compiled pattern. */
+
+  if (c == OP_RECURSE)
+    {
+    const pcre_uchar *scode;
+    BOOL empty_branch;
+
+    /* Test for forward reference */
+
+    for (scode = cd->start_workspace; scode < cd->hwm; scode += LINK_SIZE)
+      if ((int)GET(scode, 0) == (int)(code + 1 - cd->start_code)) return TRUE;
+
+    /* Not a forward reference, test for completed backward reference */
+
+    empty_branch = FALSE;
+    scode = cd->start_code + GET(code, 1);
+    if (GET(scode, 1) == 0) return TRUE;    /* Unclosed */
+
+    /* Completed backwards reference */
+
+    do
+      {
+      if (could_be_empty_branch(scode, endcode, utf, cd))
+        {
+        empty_branch = TRUE;
+        break;
+        }
+      scode += GET(scode, 1);
+      }
+    while (*scode == OP_ALT);
+
+    if (!empty_branch) return FALSE;  /* All branches are non-empty */
+    continue;
+    }
+
+  /* Groups with zero repeats can of course be empty; skip them. */
+
+  if (c == OP_BRAZERO || c == OP_BRAMINZERO || c == OP_SKIPZERO ||
+      c == OP_BRAPOSZERO)
+    {
+    code += PRIV(OP_lengths)[c];
+    do code += GET(code, 1); while (*code == OP_ALT);
+    c = *code;
+    continue;
+    }
+
+  /* A nested group that is already marked as "could be empty" can just be
+  skipped. */
+
+  if (c == OP_SBRA  || c == OP_SBRAPOS ||
+      c == OP_SCBRA || c == OP_SCBRAPOS)
+    {
+    do code += GET(code, 1); while (*code == OP_ALT);
+    c = *code;
+    continue;
+    }
+
+  /* For other groups, scan the branches. */
+
+  if (c == OP_BRA  || c == OP_BRAPOS ||
+      c == OP_CBRA || c == OP_CBRAPOS ||
+      c == OP_ONCE || c == OP_ONCE_NC ||
+      c == OP_COND)
+    {
+    BOOL empty_branch;
+    if (GET(code, 1) == 0) return TRUE;    /* Hit unclosed bracket */
+
+    /* If a conditional group has only one branch, there is a second, implied,
+    empty branch, so just skip over the conditional, because it could be empty.
+    Otherwise, scan the individual branches of the group. */
+
+    if (c == OP_COND && code[GET(code, 1)] != OP_ALT)
+      code += GET(code, 1);
+    else
+      {
+      empty_branch = FALSE;
+      do
+        {
+        if (!empty_branch && could_be_empty_branch(code, endcode, utf, cd))
+          empty_branch = TRUE;
+        code += GET(code, 1);
+        }
+      while (*code == OP_ALT);
+      if (!empty_branch) return FALSE;   /* All branches are non-empty */
+      }
+
+    c = *code;
+    continue;
+    }
+
+  /* Handle the other opcodes */
+
+  switch (c)
+    {
+    /* Check for quantifiers after a class. XCLASS is used for classes that
+    cannot be represented just by a bit map. This includes negated single
+    high-valued characters. The length in PRIV(OP_lengths)[] is zero; the
+    actual length is stored in the compiled code, so we must update "code"
+    here. */
+
+#if defined SUPPORT_UTF || !defined COMPILE_PCRE8
+    case OP_XCLASS:
+    ccode = code += GET(code, 1);
+    goto CHECK_CLASS_REPEAT;
+#endif
+
+    case OP_CLASS:
+    case OP_NCLASS:
+    ccode = code + PRIV(OP_lengths)[OP_CLASS];
+
+#if defined SUPPORT_UTF || !defined COMPILE_PCRE8
+    CHECK_CLASS_REPEAT:
+#endif
+
+    switch (*ccode)
+      {
+      case OP_CRSTAR:            /* These could be empty; continue */
+      case OP_CRMINSTAR:
+      case OP_CRQUERY:
+      case OP_CRMINQUERY:
+      break;
+
+      default:                   /* Non-repeat => class must match */
+      case OP_CRPLUS:            /* These repeats aren't empty */
+      case OP_CRMINPLUS:
+      return FALSE;
+
+      case OP_CRRANGE:
+      case OP_CRMINRANGE:
+      if (GET2(ccode, 1) > 0) return FALSE;  /* Minimum > 0 */
+      break;
+      }
+    break;
+
+    /* Opcodes that must match a character */
+
+    case OP_PROP:
+    case OP_NOTPROP:
+    case OP_EXTUNI:
+    case OP_NOT_DIGIT:
+    case OP_DIGIT:
+    case OP_NOT_WHITESPACE:
+    case OP_WHITESPACE:
+    case OP_NOT_WORDCHAR:
+    case OP_WORDCHAR:
+    case OP_ANY:
+    case OP_ALLANY:
+    case OP_ANYBYTE:
+    case OP_CHAR:
+    case OP_CHARI:
+    case OP_NOT:
+    case OP_NOTI:
+    case OP_PLUS:
+    case OP_MINPLUS:
+    case OP_POSPLUS:
+    case OP_EXACT:
+    case OP_NOTPLUS:
+    case OP_NOTMINPLUS:
+    case OP_NOTPOSPLUS:
+    case OP_NOTEXACT:
+    case OP_TYPEPLUS:
+    case OP_TYPEMINPLUS:
+    case OP_TYPEPOSPLUS:
+    case OP_TYPEEXACT:
+    return FALSE;
+
+    /* These are going to continue, as they may be empty, but we have to
+    fudge the length for the \p and \P cases. */
+
+    case OP_TYPESTAR:
+    case OP_TYPEMINSTAR:
+    case OP_TYPEPOSSTAR:
+    case OP_TYPEQUERY:
+    case OP_TYPEMINQUERY:
+    case OP_TYPEPOSQUERY:
+    if (code[1] == OP_PROP || code[1] == OP_NOTPROP) code += 2;
+    break;
+
+    /* Same for these */
+
+    case OP_TYPEUPTO:
+    case OP_TYPEMINUPTO:
+    case OP_TYPEPOSUPTO:
+    if (code[1 + IMM2_SIZE] == OP_PROP || code[1 + IMM2_SIZE] == OP_NOTPROP)
+      code += 2;
+    break;
+
+    /* End of branch */
+
+    case OP_KET:
+    case OP_KETRMAX:
+    case OP_KETRMIN:
+    case OP_KETRPOS:
+    case OP_ALT:
+    return TRUE;
+
+    /* In UTF-8 mode, STAR, MINSTAR, POSSTAR, QUERY, MINQUERY, POSQUERY, UPTO,
+    MINUPTO, and POSUPTO may be followed by a multibyte character */
+
+#if defined SUPPORT_UTF && !defined COMPILE_PCRE32
+    case OP_STAR:
+    case OP_STARI:
+    case OP_MINSTAR:
+    case OP_MINSTARI:
+    case OP_POSSTAR:
+    case OP_POSSTARI:
+    case OP_QUERY:
+    case OP_QUERYI:
+    case OP_MINQUERY:
+    case OP_MINQUERYI:
+    case OP_POSQUERY:
+    case OP_POSQUERYI:
+    if (utf && HAS_EXTRALEN(code[1])) code += GET_EXTRALEN(code[1]);
+    break;
+
+    case OP_UPTO:
+    case OP_UPTOI:
+    case OP_MINUPTO:
+    case OP_MINUPTOI:
+    case OP_POSUPTO:
+    case OP_POSUPTOI:
+    if (utf && HAS_EXTRALEN(code[1 + IMM2_SIZE])) code += GET_EXTRALEN(code[1 + IMM2_SIZE]);
+    break;
+#endif
+
+    /* MARK, and PRUNE/SKIP/THEN with an argument must skip over the argument
+    string. */
+
+    case OP_MARK:
+    case OP_PRUNE_ARG:
+    case OP_SKIP_ARG:
+    case OP_THEN_ARG:
+    code += code[1];
+    break;
+
+    /* None of the remaining opcodes are required to match a character. */
+
+    default:
+    break;
+    }
+  }
+
+return TRUE;
+}
+
+
+
+/*************************************************
+*    Scan compiled regex for non-emptiness       *
+*************************************************/
+
+/* This function is called to check for left recursive calls. We want to check
+the current branch of the current pattern to see if it could match the empty
+string. If it could, we must look outwards for branches at other levels,
+stopping when we pass beyond the bracket which is the subject of the recursion.
+This function is called only during the real compile, not during the
+pre-compile.
+
+Arguments:
+  code        points to start of the recursion
+  endcode     points to where to stop (current RECURSE item)
+  bcptr       points to the chain of current (unclosed) branch starts
+  utf         TRUE if in UTF-8 / UTF-16 / UTF-32 mode
+  cd          pointers to tables etc
+
+Returns:      TRUE if what is matched could be empty
+*/
+
+static BOOL
+could_be_empty(const pcre_uchar *code, const pcre_uchar *endcode,
+  branch_chain *bcptr, BOOL utf, compile_data *cd)
+{
+while (bcptr != NULL && bcptr->current_branch >= code)
+  {
+  if (!could_be_empty_branch(bcptr->current_branch, endcode, utf, cd))
+    return FALSE;
+  bcptr = bcptr->outer;
+  }
+return TRUE;
+}
+
+
+
+/*************************************************
+*           Check for POSIX class syntax         *
+*************************************************/
+
+/* This function is called when the sequence "[:" or "[." or "[=" is
+encountered in a character class. It checks whether this is followed by a
+sequence of characters terminated by a matching ":]" or ".]" or "=]". If we
+reach an unescaped ']' without the special preceding character, return FALSE.
+
+Originally, this function only recognized a sequence of letters between the
+terminators, but it seems that Perl recognizes any sequence of characters,
+though of course unknown POSIX names are subsequently rejected. Perl gives an
+"Unknown POSIX class" error for [:f\oo:] for example, where previously PCRE
+didn't consider this to be a POSIX class. Likewise for [:1234:].
+
+The problem in trying to be exactly like Perl is in the handling of escapes. We
+have to be sure that [abc[:x\]pqr] is *not* treated as containing a POSIX
+class, but [abc[:x\]pqr:]] is (so that an error can be generated). The code
+below handles the special case of \], but does not try to do any other escape
+processing. This makes it different from Perl for cases such as [:l\ower:]
+where Perl recognizes it as the POSIX class "lower" but PCRE does not recognize
+"l\ower". This is a lesser evil that not diagnosing bad classes when Perl does,
+I think.
+
+A user pointed out that PCRE was rejecting [:a[:digit:]] whereas Perl was not.
+It seems that the appearance of a nested POSIX class supersedes an apparent
+external class. For example, [:a[:digit:]b:] matches "a", "b", ":", or
+a digit.
+
+In Perl, unescaped square brackets may also appear as part of class names. For
+example, [:a[:abc]b:] gives unknown POSIX class "[:abc]b:]". However, for
+[:a[:abc]b][b:] it gives unknown POSIX class "[:abc]b][b:]", which does not
+seem right at all. PCRE does not allow closing square brackets in POSIX class
+names.
+
+Arguments:
+  ptr      pointer to the initial [
+  endptr   where to return the end pointer
+
+Returns:   TRUE or FALSE
+*/
+
+static BOOL
+check_posix_syntax(const pcre_uchar *ptr, const pcre_uchar **endptr)
+{
+pcre_uchar terminator;          /* Don't combine these lines; the Solaris cc */
+terminator = *(++ptr);   /* compiler warns about "non-constant" initializer. */
+for (++ptr; *ptr != CHAR_NULL; ptr++)
+  {
+  if (*ptr == CHAR_BACKSLASH && ptr[1] == CHAR_RIGHT_SQUARE_BRACKET)
+    ptr++;
+  else if (*ptr == CHAR_RIGHT_SQUARE_BRACKET) return FALSE;
+  else
+    {
+    if (*ptr == terminator && ptr[1] == CHAR_RIGHT_SQUARE_BRACKET)
+      {
+      *endptr = ptr;
+      return TRUE;
+      }
+    if (*ptr == CHAR_LEFT_SQUARE_BRACKET &&
+         (ptr[1] == CHAR_COLON || ptr[1] == CHAR_DOT ||
+          ptr[1] == CHAR_EQUALS_SIGN) &&
+        check_posix_syntax(ptr, endptr))
+      return FALSE;
+    }
+  }
+return FALSE;
+}
+
+
+
+
+/*************************************************
+*          Check POSIX class name                *
+*************************************************/
+
+/* This function is called to check the name given in a POSIX-style class entry
+such as [:alnum:].
+
+Arguments:
+  ptr        points to the first letter
+  len        the length of the name
+
+Returns:     a value representing the name, or -1 if unknown
+*/
+
+static int
+check_posix_name(const pcre_uchar *ptr, int len)
+{
+const char *pn = posix_names;
+register int yield = 0;
+while (posix_name_lengths[yield] != 0)
+  {
+  if (len == posix_name_lengths[yield] &&
+    STRNCMP_UC_C8(ptr, pn, (unsigned int)len) == 0) return yield;
+  pn += posix_name_lengths[yield] + 1;
+  yield++;
+  }
+return -1;
+}
+
+
+/*************************************************
+*    Adjust OP_RECURSE items in repeated group   *
+*************************************************/
+
+/* OP_RECURSE items contain an offset from the start of the regex to the group
+that is referenced. This means that groups can be replicated for fixed
+repetition simply by copying (because the recursion is allowed to refer to
+earlier groups that are outside the current group). However, when a group is
+optional (i.e. the minimum quantifier is zero), OP_BRAZERO or OP_SKIPZERO is
+inserted before it, after it has been compiled. This means that any OP_RECURSE
+items within it that refer to the group itself or any contained groups have to
+have their offsets adjusted. That one of the jobs of this function. Before it
+is called, the partially compiled regex must be temporarily terminated with
+OP_END.
+
+This function has been extended with the possibility of forward references for
+recursions and subroutine calls. It must also check the list of such references
+for the group we are dealing with. If it finds that one of the recursions in
+the current group is on this list, it adjusts the offset in the list, not the
+value in the reference (which is a group number).
+
+Arguments:
+  group      points to the start of the group
+  adjust     the amount by which the group is to be moved
+  utf        TRUE in UTF-8 / UTF-16 / UTF-32 mode
+  cd         contains pointers to tables etc.
+  save_hwm   the hwm forward reference pointer at the start of the group
+
+Returns:     nothing
+*/
+
+static void
+adjust_recurse(pcre_uchar *group, int adjust, BOOL utf, compile_data *cd,
+  pcre_uchar *save_hwm)
+{
+pcre_uchar *ptr = group;
+
+while ((ptr = (pcre_uchar *)find_recurse(ptr, utf)) != NULL)
+  {
+  int offset;
+  pcre_uchar *hc;
+
+  /* See if this recursion is on the forward reference list. If so, adjust the
+  reference. */
+
+  for (hc = save_hwm; hc < cd->hwm; hc += LINK_SIZE)
+    {
+    offset = (int)GET(hc, 0);
+    if (cd->start_code + offset == ptr + 1)
+      {
+      PUT(hc, 0, offset + adjust);
+      break;
+      }
+    }
+
+  /* Otherwise, adjust the recursion offset if it's after the start of this
+  group. */
+
+  if (hc >= cd->hwm)
+    {
+    offset = (int)GET(ptr, 1);
+    if (cd->start_code + offset >= group) PUT(ptr, 1, offset + adjust);
+    }
+
+  ptr += 1 + LINK_SIZE;
+  }
+}
+
+
+
+/*************************************************
+*        Insert an automatic callout point       *
+*************************************************/
+
+/* This function is called when the PCRE_AUTO_CALLOUT option is set, to insert
+callout points before each pattern item.
+
+Arguments:
+  code           current code pointer
+  ptr            current pattern pointer
+  cd             pointers to tables etc
+
+Returns:         new code pointer
+*/
+
+static pcre_uchar *
+auto_callout(pcre_uchar *code, const pcre_uchar *ptr, compile_data *cd)
+{
+*code++ = OP_CALLOUT;
+*code++ = 255;
+PUT(code, 0, (int)(ptr - cd->start_pattern));  /* Pattern offset */
+PUT(code, LINK_SIZE, 0);                       /* Default length */
+return code + 2 * LINK_SIZE;
+}
+
+
+
+/*************************************************
+*         Complete a callout item                *
+*************************************************/
+
+/* A callout item contains the length of the next item in the pattern, which
+we can't fill in till after we have reached the relevant point. This is used
+for both automatic and manual callouts.
+
+Arguments:
+  previous_callout   points to previous callout item
+  ptr                current pattern pointer
+  cd                 pointers to tables etc
+
+Returns:             nothing
+*/
+
+static void
+complete_callout(pcre_uchar *previous_callout, const pcre_uchar *ptr, compile_data *cd)
+{
+int length = (int)(ptr - cd->start_pattern - GET(previous_callout, 2));
+PUT(previous_callout, 2 + LINK_SIZE, length);
+}
+
+
+
+#ifdef SUPPORT_UCP
+/*************************************************
+*           Get othercase range                  *
+*************************************************/
+
+/* This function is passed the start and end of a class range, in UTF-8 mode
+with UCP support. It searches up the characters, looking for ranges of
+characters in the "other" case. Each call returns the next one, updating the
+start address. A character with multiple other cases is returned on its own
+with a special return value.
+
+Arguments:
+  cptr        points to starting character value; updated
+  d           end value
+  ocptr       where to put start of othercase range
+  odptr       where to put end of othercase range
+
+Yield:        -1 when no more
+               0 when a range is returned
+              >0 the CASESET offset for char with multiple other cases
+                in this case, ocptr contains the original
+*/
+
+static int
+get_othercase_range(pcre_uint32 *cptr, pcre_uint32 d, pcre_uint32 *ocptr,
+  pcre_uint32 *odptr)
+{
+pcre_uint32 c, othercase, next;
+unsigned int co;
+
+/* Find the first character that has an other case. If it has multiple other
+cases, return its case offset value. */
+
+for (c = *cptr; c <= d; c++)
+  {
+  if ((co = UCD_CASESET(c)) != 0)
+    {
+    *ocptr = c++;   /* Character that has the set */
+    *cptr = c;      /* Rest of input range */
+    return (int)co;
+    }
+  if ((othercase = UCD_OTHERCASE(c)) != c) break;
+  }
+
+if (c > d) return -1;  /* Reached end of range */
+
+*ocptr = othercase;
+next = othercase + 1;
+
+for (++c; c <= d; c++)
+  {
+  if (UCD_OTHERCASE(c) != next) break;
+  next++;
+  }
+
+*odptr = next - 1;     /* End of othercase range */
+*cptr = c;             /* Rest of input range */
+return 0;
+}
+
+
+
+/*************************************************
+*        Check a character and a property        *
+*************************************************/
+
+/* This function is called by check_auto_possessive() when a property item
+is adjacent to a fixed character.
+
+Arguments:
+  c            the character
+  ptype        the property type
+  pdata        the data for the type
+  negated      TRUE if it's a negated property (\P or \p{^)
+
+Returns:       TRUE if auto-possessifying is OK
+*/
+
+static BOOL
+check_char_prop(pcre_uint32 c, unsigned int ptype, unsigned int pdata, BOOL negated)
+{
+#ifdef SUPPORT_UCP
+const pcre_uint32 *p;
+#endif
+
+const ucd_record *prop = GET_UCD(c);
+
+switch(ptype)
+  {
+  case PT_LAMP:
+  return (prop->chartype == ucp_Lu ||
+          prop->chartype == ucp_Ll ||
+          prop->chartype == ucp_Lt) == negated;
+
+  case PT_GC:
+  return (pdata == PRIV(ucp_gentype)[prop->chartype]) == negated;
+
+  case PT_PC:
+  return (pdata == prop->chartype) == negated;
+
+  case PT_SC:
+  return (pdata == prop->script) == negated;
+
+  /* These are specials */
+
+  case PT_ALNUM:
+  return (PRIV(ucp_gentype)[prop->chartype] == ucp_L ||
+          PRIV(ucp_gentype)[prop->chartype] == ucp_N) == negated;
+
+  case PT_SPACE:    /* Perl space */
+  return (PRIV(ucp_gentype)[prop->chartype] == ucp_Z ||
+          c == CHAR_HT || c == CHAR_NL || c == CHAR_FF || c == CHAR_CR)
+          == negated;
+
+  case PT_PXSPACE:  /* POSIX space */
+  return (PRIV(ucp_gentype)[prop->chartype] == ucp_Z ||
+          c == CHAR_HT || c == CHAR_NL || c == CHAR_VT ||
+          c == CHAR_FF || c == CHAR_CR)
+          == negated;
+
+  case PT_WORD:
+  return (PRIV(ucp_gentype)[prop->chartype] == ucp_L ||
+          PRIV(ucp_gentype)[prop->chartype] == ucp_N ||
+          c == CHAR_UNDERSCORE) == negated;
+
+#ifdef SUPPORT_UCP
+  case PT_CLIST:
+  p = PRIV(ucd_caseless_sets) + prop->caseset;
+  for (;;)
+    {
+    if (c < *p) return !negated;
+    if (c == *p++) return negated;
+    }
+  break;  /* Control never reaches here */
+#endif
+  }
+
+return FALSE;
+}
+#endif  /* SUPPORT_UCP */
+
+
+
+/*************************************************
+*     Check if auto-possessifying is possible    *
+*************************************************/
+
+/* This function is called for unlimited repeats of certain items, to see
+whether the next thing could possibly match the repeated item. If not, it makes
+sense to automatically possessify the repeated item.
+
+Arguments:
+  previous      pointer to the repeated opcode
+  utf           TRUE in UTF-8 / UTF-16 / UTF-32 mode
+  ptr           next character in pattern
+  options       options bits
+  cd            contains pointers to tables etc.
+
+Returns:        TRUE if possessifying is wanted
+*/
+
+static BOOL
+check_auto_possessive(const pcre_uchar *previous, BOOL utf,
+  const pcre_uchar *ptr, int options, compile_data *cd)
+{
+pcre_uint32 c = NOTACHAR;
+pcre_uint32 next;
+int escape;
+pcre_uchar op_code = *previous++;
+
+/* Skip whitespace and comments in extended mode */
+
+if ((options & PCRE_EXTENDED) != 0)
+  {
+  for (;;)
+    {
+    while (MAX_255(*ptr) && (cd->ctypes[*ptr] & ctype_space) != 0) ptr++;
+    if (*ptr == CHAR_NUMBER_SIGN)
+      {
+      ptr++;
+      while (*ptr != CHAR_NULL)
+        {
+        if (IS_NEWLINE(ptr)) { ptr += cd->nllen; break; }
+        ptr++;
+#ifdef SUPPORT_UTF
+        if (utf) FORWARDCHAR(ptr);
+#endif
+        }
+      }
+    else break;
+    }
+  }
+
+/* If the next item is one that we can handle, get its value. A non-negative
+value is a character, a negative value is an escape value. */
+
+if (*ptr == CHAR_BACKSLASH)
+  {
+  int temperrorcode = 0;
+  escape = check_escape(&ptr, &next, &temperrorcode, cd->bracount, options,
+    FALSE);
+  if (temperrorcode != 0) return FALSE;
+  ptr++;    /* Point after the escape sequence */
+  }
+else if (!MAX_255(*ptr) || (cd->ctypes[*ptr] & ctype_meta) == 0)
+  {
+  escape = 0;
+#ifdef SUPPORT_UTF
+  if (utf) { GETCHARINC(next, ptr); } else
+#endif
+  next = *ptr++;
+  }
+else return FALSE;
+
+/* Skip whitespace and comments in extended mode */
+
+if ((options & PCRE_EXTENDED) != 0)
+  {
+  for (;;)
+    {
+    while (MAX_255(*ptr) && (cd->ctypes[*ptr] & ctype_space) != 0) ptr++;
+    if (*ptr == CHAR_NUMBER_SIGN)
+      {
+      ptr++;
+      while (*ptr != CHAR_NULL)
+        {
+        if (IS_NEWLINE(ptr)) { ptr += cd->nllen; break; }
+        ptr++;
+#ifdef SUPPORT_UTF
+        if (utf) FORWARDCHAR(ptr);
+#endif
+        }
+      }
+    else break;
+    }
+  }
+
+/* If the next thing is itself optional, we have to give up. */
+
+if (*ptr == CHAR_ASTERISK || *ptr == CHAR_QUESTION_MARK ||
+  STRNCMP_UC_C8(ptr, STR_LEFT_CURLY_BRACKET STR_0 STR_COMMA, 3) == 0)
+    return FALSE;
+
+/* If the previous item is a character, get its value. */
+
+if (op_code == OP_CHAR || op_code == OP_CHARI ||
+    op_code == OP_NOT || op_code == OP_NOTI)
+  {
+#ifdef SUPPORT_UTF
+  GETCHARTEST(c, previous);
+#else
+  c = *previous;
+#endif
+  }
+
+/* Now compare the next item with the previous opcode. First, handle cases when
+the next item is a character. */
+
+if (escape == 0)
+  {
+  /* For a caseless UTF match, the next character may have more than one other
+  case, which maps to the special PT_CLIST property. Check this first. */
+
+#ifdef SUPPORT_UCP
+  if (utf && c != NOTACHAR && (options & PCRE_CASELESS) != 0)
+    {
+    unsigned int ocs = UCD_CASESET(next);
+    if (ocs > 0) return check_char_prop(c, PT_CLIST, ocs, op_code >= OP_NOT);
+    }
+#endif
+
+  switch(op_code)
+    {
+    case OP_CHAR:
+    return c != next;
+
+    /* For CHARI (caseless character) we must check the other case. If we have
+    Unicode property support, we can use it to test the other case of
+    high-valued characters. We know that next can have only one other case,
+    because multi-other-case characters are dealt with above. */
+
+    case OP_CHARI:
+    if (c == next) return FALSE;
+#ifdef SUPPORT_UTF
+    if (utf)
+      {
+      pcre_uint32 othercase;
+      if (next < 128) othercase = cd->fcc[next]; else
+#ifdef SUPPORT_UCP
+      othercase = UCD_OTHERCASE(next);
+#else
+      othercase = NOTACHAR;
+#endif
+      return c != othercase;
+      }
+    else
+#endif  /* SUPPORT_UTF */
+    return (c != TABLE_GET(next, cd->fcc, next));  /* Not UTF */
+
+    case OP_NOT:
+    return c == next;
+
+    case OP_NOTI:
+    if (c == next) return TRUE;
+#ifdef SUPPORT_UTF
+    if (utf)
+      {
+      pcre_uint32 othercase;
+      if (next < 128) othercase = cd->fcc[next]; else
+#ifdef SUPPORT_UCP
+      othercase = UCD_OTHERCASE(next);
+#else
+      othercase = NOTACHAR;
+#endif
+      return c == othercase;
+      }
+    else
+#endif  /* SUPPORT_UTF */
+    return (c == TABLE_GET(next, cd->fcc, next));  /* Not UTF */
+
+    /* Note that OP_DIGIT etc. are generated only when PCRE_UCP is *not* set.
+    When it is set, \d etc. are converted into OP_(NOT_)PROP codes. */
+
+    case OP_DIGIT:
+    return next > 255 || (cd->ctypes[next] & ctype_digit) == 0;
+
+    case OP_NOT_DIGIT:
+    return next <= 255 && (cd->ctypes[next] & ctype_digit) != 0;
+
+    case OP_WHITESPACE:
+    return next > 255 || (cd->ctypes[next] & ctype_space) == 0;
+
+    case OP_NOT_WHITESPACE:
+    return next <= 255 && (cd->ctypes[next] & ctype_space) != 0;
+
+    case OP_WORDCHAR:
+    return next > 255 || (cd->ctypes[next] & ctype_word) == 0;
+
+    case OP_NOT_WORDCHAR:
+    return next <= 255 && (cd->ctypes[next] & ctype_word) != 0;
+
+    case OP_HSPACE:
+    case OP_NOT_HSPACE:
+    switch(next)
+      {
+      HSPACE_CASES:
+      return op_code == OP_NOT_HSPACE;
+
+      default:
+      return op_code != OP_NOT_HSPACE;
+      }
+
+    case OP_ANYNL:
+    case OP_VSPACE:
+    case OP_NOT_VSPACE:
+    switch(next)
+      {
+      VSPACE_CASES:
+      return op_code == OP_NOT_VSPACE;
+
+      default:
+      return op_code != OP_NOT_VSPACE;
+      }
+
+#ifdef SUPPORT_UCP
+    case OP_PROP:
+    return check_char_prop(next, previous[0], previous[1], FALSE);
+
+    case OP_NOTPROP:
+    return check_char_prop(next, previous[0], previous[1], TRUE);
+#endif
+
+    default:
+    return FALSE;
+    }
+  }
+
+/* Handle the case when the next item is \d, \s, etc. Note that when PCRE_UCP
+is set, \d turns into ESC_du rather than ESC_d, etc., so ESC_d etc. are
+generated only when PCRE_UCP is *not* set, that is, when only ASCII
+characteristics are recognized. Similarly, the opcodes OP_DIGIT etc. are
+replaced by OP_PROP codes when PCRE_UCP is set. */
+
+switch(op_code)
+  {
+  case OP_CHAR:
+  case OP_CHARI:
+  switch(escape)
+    {
+    case ESC_d:
+    return c > 255 || (cd->ctypes[c] & ctype_digit) == 0;
+
+    case ESC_D:
+    return c <= 255 && (cd->ctypes[c] & ctype_digit) != 0;
+
+    case ESC_s:
+    return c > 255 || (cd->ctypes[c] & ctype_space) == 0;
+
+    case ESC_S:
+    return c <= 255 && (cd->ctypes[c] & ctype_space) != 0;
+
+    case ESC_w:
+    return c > 255 || (cd->ctypes[c] & ctype_word) == 0;
+
+    case ESC_W:
+    return c <= 255 && (cd->ctypes[c] & ctype_word) != 0;
+
+    case ESC_h:
+    case ESC_H:
+    switch(c)
+      {
+      HSPACE_CASES:
+      return escape != ESC_h;
+
+      default:
+      return escape == ESC_h;
+      }
+
+    case ESC_v:
+    case ESC_V:
+    switch(c)
+      {
+      VSPACE_CASES:
+      return escape != ESC_v;
+
+      default:
+      return escape == ESC_v;
+      }
+
+    /* When PCRE_UCP is set, these values get generated for \d etc. Find
+    their substitutions and process them. The result will always be either
+    ESC_p or ESC_P. Then fall through to process those values. */
+
+#ifdef SUPPORT_UCP
+    case ESC_du:
+    case ESC_DU:
+    case ESC_wu:
+    case ESC_WU:
+    case ESC_su:
+    case ESC_SU:
+      {
+      int temperrorcode = 0;
+      ptr = substitutes[escape - ESC_DU];
+      escape = check_escape(&ptr, &next, &temperrorcode, 0, options, FALSE);
+      if (temperrorcode != 0) return FALSE;
+      ptr++;    /* For compatibility */
+      }
+    /* Fall through */
+
+    case ESC_p:
+    case ESC_P:
+      {
+      unsigned int ptype = 0, pdata = 0;
+      int errorcodeptr;
+      BOOL negated;
+
+      ptr--;      /* Make ptr point at the p or P */
+      if (!get_ucp(&ptr, &negated, &ptype, &pdata, &errorcodeptr))
+        return FALSE;
+      ptr++;      /* Point past the final curly ket */
+
+      /* If the property item is optional, we have to give up. (When generated
+      from \d etc by PCRE_UCP, this test will have been applied much earlier,
+      to the original \d etc. At this point, ptr will point to a zero byte. */
+
+      if (*ptr == CHAR_ASTERISK || *ptr == CHAR_QUESTION_MARK ||
+        STRNCMP_UC_C8(ptr, STR_LEFT_CURLY_BRACKET STR_0 STR_COMMA, 3) == 0)
+          return FALSE;
+
+      /* Do the property check. */
+
+      return check_char_prop(c, ptype, pdata, (escape == ESC_P) != negated);
+      }
+#endif
+
+    default:
+    return FALSE;
+    }
+
+  /* In principle, support for Unicode properties should be integrated here as
+  well. It means re-organizing the above code so as to get hold of the property
+  values before switching on the op-code. However, I wonder how many patterns
+  combine ASCII \d etc with Unicode properties? (Note that if PCRE_UCP is set,
+  these op-codes are never generated.) */
+
+  case OP_DIGIT:
+  return escape == ESC_D || escape == ESC_s || escape == ESC_W ||
+         escape == ESC_h || escape == ESC_v || escape == ESC_R;
+
+  case OP_NOT_DIGIT:
+  return escape == ESC_d;
+
+  case OP_WHITESPACE:
+  return escape == ESC_S || escape == ESC_d || escape == ESC_w;
+
+  case OP_NOT_WHITESPACE:
+  return escape == ESC_s || escape == ESC_h || escape == ESC_v || escape == ESC_R;
+
+  case OP_HSPACE:
+  return escape == ESC_S || escape == ESC_H || escape == ESC_d ||
+         escape == ESC_w || escape == ESC_v || escape == ESC_R;
+
+  case OP_NOT_HSPACE:
+  return escape == ESC_h;
+
+  /* Can't have \S in here because VT matches \S (Perl anomaly) */
+  case OP_ANYNL:
+  case OP_VSPACE:
+  return escape == ESC_V || escape == ESC_d || escape == ESC_w;
+
+  case OP_NOT_VSPACE:
+  return escape == ESC_v || escape == ESC_R;
+
+  case OP_WORDCHAR:
+  return escape == ESC_W || escape == ESC_s || escape == ESC_h ||
+         escape == ESC_v || escape == ESC_R;
+
+  case OP_NOT_WORDCHAR:
+  return escape == ESC_w || escape == ESC_d;
+
+  default:
+  return FALSE;
+  }
+
+/* Control does not reach here */
+}
+
+
+
+/*************************************************
+*        Add a character or range to a class     *
+*************************************************/
+
+/* This function packages up the logic of adding a character or range of
+characters to a class. The character values in the arguments will be within the
+valid values for the current mode (8-bit, 16-bit, UTF, etc). This function is
+mutually recursive with the function immediately below.
+
+Arguments:
+  classbits     the bit map for characters < 256
+  uchardptr     points to the pointer for extra data
+  options       the options word
+  cd            contains pointers to tables etc.
+  start         start of range character
+  end           end of range character
+
+Returns:        the number of < 256 characters added
+                the pointer to extra data is updated
+*/
+
+static int
+add_to_class(pcre_uint8 *classbits, pcre_uchar **uchardptr, int options,
+  compile_data *cd, pcre_uint32 start, pcre_uint32 end)
+{
+pcre_uint32 c;
+int n8 = 0;
+
+/* If caseless matching is required, scan the range and process alternate
+cases. In Unicode, there are 8-bit characters that have alternate cases that
+are greater than 255 and vice-versa. Sometimes we can just extend the original
+range. */
+
+if ((options & PCRE_CASELESS) != 0)
+  {
+#ifdef SUPPORT_UCP
+  if ((options & PCRE_UTF8) != 0)
+    {
+    int rc;
+    pcre_uint32 oc, od;
+
+    options &= ~PCRE_CASELESS;   /* Remove for recursive calls */
+    c = start;
+
+    while ((rc = get_othercase_range(&c, end, &oc, &od)) >= 0)
+      {
+      /* Handle a single character that has more than one other case. */
+
+      if (rc > 0) n8 += add_list_to_class(classbits, uchardptr, options, cd,
+        PRIV(ucd_caseless_sets) + rc, oc);
+
+      /* Do nothing if the other case range is within the original range. */
+
+      else if (oc >= start && od <= end) continue;
+
+      /* Extend the original range if there is overlap, noting that if oc < c, we
+      can't have od > end because a subrange is always shorter than the basic
+      range. Otherwise, use a recursive call to add the additional range. */
+
+      else if (oc < start && od >= start - 1) start = oc; /* Extend downwards */
+      else if (od > end && oc <= end + 1) end = od;       /* Extend upwards */
+      else n8 += add_to_class(classbits, uchardptr, options, cd, oc, od);
+      }
+    }
+  else
+#endif  /* SUPPORT_UCP */
+
+  /* Not UTF-mode, or no UCP */
+
+  for (c = start; c <= end && c < 256; c++)
+    {
+    SETBIT(classbits, cd->fcc[c]);
+    n8++;
+    }
+  }
+
+/* Now handle the original range. Adjust the final value according to the bit
+length - this means that the same lists of (e.g.) horizontal spaces can be used
+in all cases. */
+
+#if defined COMPILE_PCRE8
+#ifdef SUPPORT_UTF
+  if ((options & PCRE_UTF8) == 0)
+#endif
+  if (end > 0xff) end = 0xff;
+
+#elif defined COMPILE_PCRE16
+#ifdef SUPPORT_UTF
+  if ((options & PCRE_UTF16) == 0)
+#endif
+  if (end > 0xffff) end = 0xffff;
+
+#endif /* COMPILE_PCRE[8|16] */
+
+/* If all characters are less than 256, use the bit map. Otherwise use extra
+data. */
+
+if (end < 0x100)
+  {
+  for (c = start; c <= end; c++)
+    {
+    n8++;
+    SETBIT(classbits, c);
+    }
+  }
+
+else
+  {
+  pcre_uchar *uchardata = *uchardptr;
+
+#ifdef SUPPORT_UTF
+  if ((options & PCRE_UTF8) != 0)  /* All UTFs use the same flag bit */
+    {
+    if (start < end)
+      {
+      *uchardata++ = XCL_RANGE;
+      uchardata += PRIV(ord2utf)(start, uchardata);
+      uchardata += PRIV(ord2utf)(end, uchardata);
+      }
+    else if (start == end)
+      {
+      *uchardata++ = XCL_SINGLE;
+      uchardata += PRIV(ord2utf)(start, uchardata);
+      }
+    }
+  else
+#endif  /* SUPPORT_UTF */
+
+  /* Without UTF support, character values are constrained by the bit length,
+  and can only be > 256 for 16-bit and 32-bit libraries. */
+
+#ifdef COMPILE_PCRE8
+    {}
+#else
+  if (start < end)
+    {
+    *uchardata++ = XCL_RANGE;
+    *uchardata++ = start;
+    *uchardata++ = end;
+    }
+  else if (start == end)
+    {
+    *uchardata++ = XCL_SINGLE;
+    *uchardata++ = start;
+    }
+#endif
+
+  *uchardptr = uchardata;   /* Updata extra data pointer */
+  }
+
+return n8;    /* Number of 8-bit characters */
+}
+
+
+
+
+/*************************************************
+*        Add a list of characters to a class     *
+*************************************************/
+
+/* This function is used for adding a list of case-equivalent characters to a
+class, and also for adding a list of horizontal or vertical whitespace. If the
+list is in order (which it should be), ranges of characters are detected and
+handled appropriately. This function is mutually recursive with the function
+above.
+
+Arguments:
+  classbits     the bit map for characters < 256
+  uchardptr     points to the pointer for extra data
+  options       the options word
+  cd            contains pointers to tables etc.
+  p             points to row of 32-bit values, terminated by NOTACHAR
+  except        character to omit; this is used when adding lists of
+                  case-equivalent characters to avoid including the one we
+                  already know about
+
+Returns:        the number of < 256 characters added
+                the pointer to extra data is updated
+*/
+
+static int
+add_list_to_class(pcre_uint8 *classbits, pcre_uchar **uchardptr, int options,
+  compile_data *cd, const pcre_uint32 *p, unsigned int except)
+{
+int n8 = 0;
+while (p[0] < NOTACHAR)
+  {
+  int n = 0;
+  if (p[0] != except)
+    {
+    while(p[n+1] == p[0] + n + 1) n++;
+    n8 += add_to_class(classbits, uchardptr, options, cd, p[0], p[n]);
+    }
+  p += n + 1;
+  }
+return n8;
+}
+
+
+
+/*************************************************
+*    Add characters not in a list to a class     *
+*************************************************/
+
+/* This function is used for adding the complement of a list of horizontal or
+vertical whitespace to a class. The list must be in order.
+
+Arguments:
+  classbits     the bit map for characters < 256
+  uchardptr     points to the pointer for extra data
+  options       the options word
+  cd            contains pointers to tables etc.
+  p             points to row of 32-bit values, terminated by NOTACHAR
+
+Returns:        the number of < 256 characters added
+                the pointer to extra data is updated
+*/
+
+static int
+add_not_list_to_class(pcre_uint8 *classbits, pcre_uchar **uchardptr,
+  int options, compile_data *cd, const pcre_uint32 *p)
+{
+BOOL utf = (options & PCRE_UTF8) != 0;
+int n8 = 0;
+if (p[0] > 0)
+  n8 += add_to_class(classbits, uchardptr, options, cd, 0, p[0] - 1);
+while (p[0] < NOTACHAR)
+  {
+  while (p[1] == p[0] + 1) p++;
+  n8 += add_to_class(classbits, uchardptr, options, cd, p[0] + 1,
+    (p[1] == NOTACHAR) ? (utf ? 0x10ffffu : 0xffffffffu) : p[1] - 1);
+  p++;
+  }
+return n8;
+}
+
+
+
+/*************************************************
+*           Compile one branch                   *
+*************************************************/
+
+/* Scan the pattern, compiling it into the a vector. If the options are
+changed during the branch, the pointer is used to change the external options
+bits. This function is used during the pre-compile phase when we are trying
+to find out the amount of memory needed, as well as during the real compile
+phase. The value of lengthptr distinguishes the two phases.
+
+Arguments:
+  optionsptr     pointer to the option bits
+  codeptr        points to the pointer to the current code point
+  ptrptr         points to the current pattern pointer
+  errorcodeptr   points to error code variable
+  firstcharptr    place to put the first required character
+  firstcharflagsptr place to put the first character flags, or a negative number
+  reqcharptr     place to put the last required character
+  reqcharflagsptr place to put the last required character flags, or a negative number
+  bcptr          points to current branch chain
+  cond_depth     conditional nesting depth
+  cd             contains pointers to tables etc.
+  lengthptr      NULL during the real compile phase
+                 points to length accumulator during pre-compile phase
+
+Returns:         TRUE on success
+                 FALSE, with *errorcodeptr set non-zero on error
+*/
+
+static BOOL
+compile_branch(int *optionsptr, pcre_uchar **codeptr,
+  const pcre_uchar **ptrptr, int *errorcodeptr,
+  pcre_uint32 *firstcharptr, pcre_int32 *firstcharflagsptr,
+  pcre_uint32 *reqcharptr, pcre_int32 *reqcharflagsptr,
+  branch_chain *bcptr, int cond_depth,
+  compile_data *cd, int *lengthptr)
+{
+int repeat_type, op_type;
+int repeat_min = 0, repeat_max = 0;      /* To please picky compilers */
+int bravalue = 0;
+int greedy_default, greedy_non_default;
+pcre_uint32 firstchar, reqchar;
+pcre_int32 firstcharflags, reqcharflags;
+pcre_uint32 zeroreqchar, zerofirstchar;
+pcre_int32 zeroreqcharflags, zerofirstcharflags;
+pcre_int32 req_caseopt, reqvary, tempreqvary;
+int options = *optionsptr;               /* May change dynamically */
+int after_manual_callout = 0;
+int length_prevgroup = 0;
+register pcre_uint32 c;
+int escape;
+register pcre_uchar *code = *codeptr;
+pcre_uchar *last_code = code;
+pcre_uchar *orig_code = code;
+pcre_uchar *tempcode;
+BOOL inescq = FALSE;
+BOOL groupsetfirstchar = FALSE;
+const pcre_uchar *ptr = *ptrptr;
+const pcre_uchar *tempptr;
+const pcre_uchar *nestptr = NULL;
+pcre_uchar *previous = NULL;
+pcre_uchar *previous_callout = NULL;
+pcre_uchar *save_hwm = NULL;
+pcre_uint8 classbits[32];
+
+/* We can fish out the UTF-8 setting once and for all into a BOOL, but we
+must not do this for other options (e.g. PCRE_EXTENDED) because they may change
+dynamically as we process the pattern. */
+
+#ifdef SUPPORT_UTF
+/* PCRE_UTF[16|32] have the same value as PCRE_UTF8. */
+BOOL utf = (options & PCRE_UTF8) != 0;
+#ifndef COMPILE_PCRE32
+pcre_uchar utf_chars[6];
+#endif
+#else
+BOOL utf = FALSE;
+#endif
+
+/* Helper variables for OP_XCLASS opcode (for characters > 255). We define
+class_uchardata always so that it can be passed to add_to_class() always,
+though it will not be used in non-UTF 8-bit cases. This avoids having to supply
+alternative calls for the different cases. */
+
+pcre_uchar *class_uchardata;
+#if defined SUPPORT_UTF || !defined COMPILE_PCRE8
+BOOL xclass;
+pcre_uchar *class_uchardata_base;
+#endif
+
+#ifdef PCRE_DEBUG
+if (lengthptr != NULL) DPRINTF((">> start branch\n"));
+#endif
+
+/* Set up the default and non-default settings for greediness */
+
+greedy_default = ((options & PCRE_UNGREEDY) != 0);
+greedy_non_default = greedy_default ^ 1;
+
+/* Initialize no first byte, no required byte. REQ_UNSET means "no char
+matching encountered yet". It gets changed to REQ_NONE if we hit something that
+matches a non-fixed char first char; reqchar just remains unset if we never
+find one.
+
+When we hit a repeat whose minimum is zero, we may have to adjust these values
+to take the zero repeat into account. This is implemented by setting them to
+zerofirstbyte and zeroreqchar when such a repeat is encountered. The individual
+item types that can be repeated set these backoff variables appropriately. */
+
+firstchar = reqchar = zerofirstchar = zeroreqchar = 0;
+firstcharflags = reqcharflags = zerofirstcharflags = zeroreqcharflags = REQ_UNSET;
+
+/* The variable req_caseopt contains either the REQ_CASELESS value
+or zero, according to the current setting of the caseless flag. The
+REQ_CASELESS leaves the lower 28 bit empty. It is added into the
+firstchar or reqchar variables to record the case status of the
+value. This is used only for ASCII characters. */
+
+req_caseopt = ((options & PCRE_CASELESS) != 0)? REQ_CASELESS:0;
+
+/* Switch on next character until the end of the branch */
+
+for (;; ptr++)
+  {
+  BOOL negate_class;
+  BOOL should_flip_negation;
+  BOOL possessive_quantifier;
+  BOOL is_quantifier;
+  BOOL is_recurse;
+  BOOL reset_bracount;
+  int class_has_8bitchar;
+  int class_one_char;
+  int newoptions;
+  int recno;
+  int refsign;
+  int skipbytes;
+  pcre_uint32 subreqchar, subfirstchar;
+  pcre_int32 subreqcharflags, subfirstcharflags;
+  int terminator;
+  unsigned int mclength;
+  unsigned int tempbracount;
+  pcre_uint32 ec;
+  pcre_uchar mcbuffer[8];
+
+  /* Get next character in the pattern */
+
+  c = *ptr;
+
+  /* If we are at the end of a nested substitution, revert to the outer level
+  string. Nesting only happens one level deep. */
+
+  if (c == CHAR_NULL && nestptr != NULL)
+    {
+    ptr = nestptr;
+    nestptr = NULL;
+    c = *ptr;
+    }
+
+  /* If we are in the pre-compile phase, accumulate the length used for the
+  previous cycle of this loop. */
+
+  if (lengthptr != NULL)
+    {
+#ifdef PCRE_DEBUG
+    if (code > cd->hwm) cd->hwm = code;                 /* High water info */
+#endif
+    if (code > cd->start_workspace + cd->workspace_size -
+        WORK_SIZE_SAFETY_MARGIN)                       /* Check for overrun */
+      {
+      *errorcodeptr = ERR52;
+      goto FAILED;
+      }
+
+    /* There is at least one situation where code goes backwards: this is the
+    case of a zero quantifier after a class (e.g. [ab]{0}). At compile time,
+    the class is simply eliminated. However, it is created first, so we have to
+    allow memory for it. Therefore, don't ever reduce the length at this point.
+    */
+
+    if (code < last_code) code = last_code;
+
+    /* Paranoid check for integer overflow */
+
+    if (OFLOW_MAX - *lengthptr < code - last_code)
+      {
+      *errorcodeptr = ERR20;
+      goto FAILED;
+      }
+
+    *lengthptr += (int)(code - last_code);
+    DPRINTF(("length=%d added %d c=%c (0x%x)\n", *lengthptr,
+      (int)(code - last_code), c, c));
+
+    /* If "previous" is set and it is not at the start of the work space, move
+    it back to there, in order to avoid filling up the work space. Otherwise,
+    if "previous" is NULL, reset the current code pointer to the start. */
+
+    if (previous != NULL)
+      {
+      if (previous > orig_code)
+        {
+        memmove(orig_code, previous, IN_UCHARS(code - previous));
+        code -= previous - orig_code;
+        previous = orig_code;
+        }
+      }
+    else code = orig_code;
+
+    /* Remember where this code item starts so we can pick up the length
+    next time round. */
+
+    last_code = code;
+    }
+
+  /* In the real compile phase, just check the workspace used by the forward
+  reference list. */
+
+  else if (cd->hwm > cd->start_workspace + cd->workspace_size -
+           WORK_SIZE_SAFETY_MARGIN)
+    {
+    *errorcodeptr = ERR52;
+    goto FAILED;
+    }
+
+  /* If in \Q...\E, check for the end; if not, we have a literal */
+
+  if (inescq && c != CHAR_NULL)
+    {
+    if (c == CHAR_BACKSLASH && ptr[1] == CHAR_E)
+      {
+      inescq = FALSE;
+      ptr++;
+      continue;
+      }
+    else
+      {
+      if (previous_callout != NULL)
+        {
+        if (lengthptr == NULL)  /* Don't attempt in pre-compile phase */
+          complete_callout(previous_callout, ptr, cd);
+        previous_callout = NULL;
+        }
+      if ((options & PCRE_AUTO_CALLOUT) != 0)
+        {
+        previous_callout = code;
+        code = auto_callout(code, ptr, cd);
+        }
+      goto NORMAL_CHAR;
+      }
+    }
+
+  /* Fill in length of a previous callout, except when the next thing is
+  a quantifier. */
+
+  is_quantifier =
+    c == CHAR_ASTERISK || c == CHAR_PLUS || c == CHAR_QUESTION_MARK ||
+    (c == CHAR_LEFT_CURLY_BRACKET && is_counted_repeat(ptr+1));
+
+  if (!is_quantifier && previous_callout != NULL &&
+       after_manual_callout-- <= 0)
+    {
+    if (lengthptr == NULL)      /* Don't attempt in pre-compile phase */
+      complete_callout(previous_callout, ptr, cd);
+    previous_callout = NULL;
+    }
+
+  /* In extended mode, skip white space and comments. */
+
+  if ((options & PCRE_EXTENDED) != 0)
+    {
+    if (MAX_255(*ptr) && (cd->ctypes[c] & ctype_space) != 0) continue;
+    if (c == CHAR_NUMBER_SIGN)
+      {
+      ptr++;
+      while (*ptr != CHAR_NULL)
+        {
+        if (IS_NEWLINE(ptr)) { ptr += cd->nllen - 1; break; }
+        ptr++;
+#ifdef SUPPORT_UTF
+        if (utf) FORWARDCHAR(ptr);
+#endif
+        }
+      if (*ptr != CHAR_NULL) continue;
+
+      /* Else fall through to handle end of string */
+      c = 0;
+      }
+    }
+
+  /* No auto callout for quantifiers. */
+
+  if ((options & PCRE_AUTO_CALLOUT) != 0 && !is_quantifier)
+    {
+    previous_callout = code;
+    code = auto_callout(code, ptr, cd);
+    }
+
+  switch(c)
+    {
+    /* ===================================================================*/
+    case 0:                        /* The branch terminates at string end */
+    case CHAR_VERTICAL_LINE:       /* or | or ) */
+    case CHAR_RIGHT_PARENTHESIS:
+    *firstcharptr = firstchar;
+    *firstcharflagsptr = firstcharflags;
+    *reqcharptr = reqchar;
+    *reqcharflagsptr = reqcharflags;
+    *codeptr = code;
+    *ptrptr = ptr;
+    if (lengthptr != NULL)
+      {
+      if (OFLOW_MAX - *lengthptr < code - last_code)
+        {
+        *errorcodeptr = ERR20;
+        goto FAILED;
+        }
+      *lengthptr += (int)(code - last_code);   /* To include callout length */
+      DPRINTF((">> end branch\n"));
+      }
+    return TRUE;
+
+
+    /* ===================================================================*/
+    /* Handle single-character metacharacters. In multiline mode, ^ disables
+    the setting of any following char as a first character. */
+
+    case CHAR_CIRCUMFLEX_ACCENT:
+    previous = NULL;
+    if ((options & PCRE_MULTILINE) != 0)
+      {
+      if (firstcharflags == REQ_UNSET) firstcharflags = REQ_NONE;
+      *code++ = OP_CIRCM;
+      }
+    else *code++ = OP_CIRC;
+    break;
+
+    case CHAR_DOLLAR_SIGN:
+    previous = NULL;
+    *code++ = ((options & PCRE_MULTILINE) != 0)? OP_DOLLM : OP_DOLL;
+    break;
+
+    /* There can never be a first char if '.' is first, whatever happens about
+    repeats. The value of reqchar doesn't change either. */
+
+    case CHAR_DOT:
+    if (firstcharflags == REQ_UNSET) firstcharflags = REQ_NONE;
+    zerofirstchar = firstchar;
+    zerofirstcharflags = firstcharflags;
+    zeroreqchar = reqchar;
+    zeroreqcharflags = reqcharflags;
+    previous = code;
+    *code++ = ((options & PCRE_DOTALL) != 0)? OP_ALLANY: OP_ANY;
+    break;
+
+
+    /* ===================================================================*/
+    /* Character classes. If the included characters are all < 256, we build a
+    32-byte bitmap of the permitted characters, except in the special case
+    where there is only one such character. For negated classes, we build the
+    map as usual, then invert it at the end. However, we use a different opcode
+    so that data characters > 255 can be handled correctly.
+
+    If the class contains characters outside the 0-255 range, a different
+    opcode is compiled. It may optionally have a bit map for characters < 256,
+    but those above are are explicitly listed afterwards. A flag byte tells
+    whether the bitmap is present, and whether this is a negated class or not.
+
+    In JavaScript compatibility mode, an isolated ']' causes an error. In
+    default (Perl) mode, it is treated as a data character. */
+
+    case CHAR_RIGHT_SQUARE_BRACKET:
+    if ((cd->external_options & PCRE_JAVASCRIPT_COMPAT) != 0)
+      {
+      *errorcodeptr = ERR64;
+      goto FAILED;
+      }
+    goto NORMAL_CHAR;
+
+    case CHAR_LEFT_SQUARE_BRACKET:
+    previous = code;
+
+    /* PCRE supports POSIX class stuff inside a class. Perl gives an error if
+    they are encountered at the top level, so we'll do that too. */
+
+    if ((ptr[1] == CHAR_COLON || ptr[1] == CHAR_DOT ||
+         ptr[1] == CHAR_EQUALS_SIGN) &&
+        check_posix_syntax(ptr, &tempptr))
+      {
+      *errorcodeptr = (ptr[1] == CHAR_COLON)? ERR13 : ERR31;
+      goto FAILED;
+      }
+
+    /* If the first character is '^', set the negation flag and skip it. Also,
+    if the first few characters (either before or after ^) are \Q\E or \E we
+    skip them too. This makes for compatibility with Perl. */
+
+    negate_class = FALSE;
+    for (;;)
+      {
+      c = *(++ptr);
+      if (c == CHAR_BACKSLASH)
+        {
+        if (ptr[1] == CHAR_E)
+          ptr++;
+        else if (STRNCMP_UC_C8(ptr + 1, STR_Q STR_BACKSLASH STR_E, 3) == 0)
+          ptr += 3;
+        else
+          break;
+        }
+      else if (!negate_class && c == CHAR_CIRCUMFLEX_ACCENT)
+        negate_class = TRUE;
+      else break;
+      }
+
+    /* Empty classes are allowed in JavaScript compatibility mode. Otherwise,
+    an initial ']' is taken as a data character -- the code below handles
+    that. In JS mode, [] must always fail, so generate OP_FAIL, whereas
+    [^] must match any character, so generate OP_ALLANY. */
+
+    if (c == CHAR_RIGHT_SQUARE_BRACKET &&
+        (cd->external_options & PCRE_JAVASCRIPT_COMPAT) != 0)
+      {
+      *code++ = negate_class? OP_ALLANY : OP_FAIL;
+      if (firstcharflags == REQ_UNSET) firstcharflags = REQ_NONE;
+      zerofirstchar = firstchar;
+      zerofirstcharflags = firstcharflags;
+      break;
+      }
+
+    /* If a class contains a negative special such as \S, we need to flip the
+    negation flag at the end, so that support for characters > 255 works
+    correctly (they are all included in the class). */
+
+    should_flip_negation = FALSE;
+
+    /* For optimization purposes, we track some properties of the class:
+    class_has_8bitchar will be non-zero if the class contains at least one <
+    256 character; class_one_char will be 1 if the class contains just one
+    character. */
+
+    class_has_8bitchar = 0;
+    class_one_char = 0;
+
+    /* Initialize the 32-char bit map to all zeros. We build the map in a
+    temporary bit of memory, in case the class contains fewer than two
+    8-bit characters because in that case the compiled code doesn't use the bit
+    map. */
+
+    memset(classbits, 0, 32 * sizeof(pcre_uint8));
+
+#if defined SUPPORT_UTF || !defined COMPILE_PCRE8
+    xclass = FALSE;
+    class_uchardata = code + LINK_SIZE + 2;   /* For XCLASS items */
+    class_uchardata_base = class_uchardata;   /* Save the start */
+#endif
+
+    /* Process characters until ] is reached. By writing this as a "do" it
+    means that an initial ] is taken as a data character. At the start of the
+    loop, c contains the first byte of the character. */
+
+    if (c != CHAR_NULL) do
+      {
+      const pcre_uchar *oldptr;
+
+#ifdef SUPPORT_UTF
+      if (utf && HAS_EXTRALEN(c))
+        {                           /* Braces are required because the */
+        GETCHARLEN(c, ptr, ptr);    /* macro generates multiple statements */
+        }
+#endif
+
+#if defined SUPPORT_UTF || !defined COMPILE_PCRE8
+      /* In the pre-compile phase, accumulate the length of any extra
+      data and reset the pointer. This is so that very large classes that
+      contain a zillion > 255 characters no longer overwrite the work space
+      (which is on the stack). We have to remember that there was XCLASS data,
+      however. */
+
+      if (lengthptr != NULL && class_uchardata > class_uchardata_base)
+        {
+        xclass = TRUE;
+        *lengthptr += class_uchardata - class_uchardata_base;
+        class_uchardata = class_uchardata_base;
+        }
+#endif
+
+      /* Inside \Q...\E everything is literal except \E */
+
+      if (inescq)
+        {
+        if (c == CHAR_BACKSLASH && ptr[1] == CHAR_E)  /* If we are at \E */
+          {
+          inescq = FALSE;                   /* Reset literal state */
+          ptr++;                            /* Skip the 'E' */
+          continue;                         /* Carry on with next */
+          }
+        goto CHECK_RANGE;                   /* Could be range if \E follows */
+        }
+
+      /* Handle POSIX class names. Perl allows a negation extension of the
+      form [:^name:]. A square bracket that doesn't match the syntax is
+      treated as a literal. We also recognize the POSIX constructions
+      [.ch.] and [=ch=] ("collating elements") and fault them, as Perl
+      5.6 and 5.8 do. */
+
+      if (c == CHAR_LEFT_SQUARE_BRACKET &&
+          (ptr[1] == CHAR_COLON || ptr[1] == CHAR_DOT ||
+           ptr[1] == CHAR_EQUALS_SIGN) && check_posix_syntax(ptr, &tempptr))
+        {
+        BOOL local_negate = FALSE;
+        int posix_class, taboffset, tabopt;
+        register const pcre_uint8 *cbits = cd->cbits;
+        pcre_uint8 pbits[32];
+
+        if (ptr[1] != CHAR_COLON)
+          {
+          *errorcodeptr = ERR31;
+          goto FAILED;
+          }
+
+        ptr += 2;
+        if (*ptr == CHAR_CIRCUMFLEX_ACCENT)
+          {
+          local_negate = TRUE;
+          should_flip_negation = TRUE;  /* Note negative special */
+          ptr++;
+          }
+
+        posix_class = check_posix_name(ptr, (int)(tempptr - ptr));
+        if (posix_class < 0)
+          {
+          *errorcodeptr = ERR30;
+          goto FAILED;
+          }
+
+        /* If matching is caseless, upper and lower are converted to
+        alpha. This relies on the fact that the class table starts with
+        alpha, lower, upper as the first 3 entries. */
+
+        if ((options & PCRE_CASELESS) != 0 && posix_class <= 2)
+          posix_class = 0;
+
+        /* When PCRE_UCP is set, some of the POSIX classes are converted to
+        different escape sequences that use Unicode properties. */
+
+#ifdef SUPPORT_UCP
+        if ((options & PCRE_UCP) != 0)
+          {
+          int pc = posix_class + ((local_negate)? POSIX_SUBSIZE/2 : 0);
+          if (posix_substitutes[pc] != NULL)
+            {
+            nestptr = tempptr + 1;
+            ptr = posix_substitutes[pc] - 1;
+            continue;
+            }
+          }
+#endif
+        /* In the non-UCP case, we build the bit map for the POSIX class in a
+        chunk of local store because we may be adding and subtracting from it,
+        and we don't want to subtract bits that may be in the main map already.
+        At the end we or the result into the bit map that is being built. */
+
+        posix_class *= 3;
+
+        /* Copy in the first table (always present) */
+
+        memcpy(pbits, cbits + posix_class_maps[posix_class],
+          32 * sizeof(pcre_uint8));
+
+        /* If there is a second table, add or remove it as required. */
+
+        taboffset = posix_class_maps[posix_class + 1];
+        tabopt = posix_class_maps[posix_class + 2];
+
+        if (taboffset >= 0)
+          {
+          if (tabopt >= 0)
+            for (c = 0; c < 32; c++) pbits[c] |= cbits[c + taboffset];
+          else
+            for (c = 0; c < 32; c++) pbits[c] &= ~cbits[c + taboffset];
+          }
+
+        /* Now see if we need to remove any special characters. An option
+        value of 1 removes vertical space and 2 removes underscore. */
+
+        if (tabopt < 0) tabopt = -tabopt;
+        if (tabopt == 1) pbits[1] &= ~0x3c;
+          else if (tabopt == 2) pbits[11] &= 0x7f;
+
+        /* Add the POSIX table or its complement into the main table that is
+        being built and we are done. */
+
+        if (local_negate)
+          for (c = 0; c < 32; c++) classbits[c] |= ~pbits[c];
+        else
+          for (c = 0; c < 32; c++) classbits[c] |= pbits[c];
+
+        ptr = tempptr + 1;
+        /* Every class contains at least one < 256 character. */
+        class_has_8bitchar = 1;
+        /* Every class contains at least two characters. */
+        class_one_char = 2;
+        continue;    /* End of POSIX syntax handling */
+        }
+
+      /* Backslash may introduce a single character, or it may introduce one
+      of the specials, which just set a flag. The sequence \b is a special
+      case. Inside a class (and only there) it is treated as backspace. We
+      assume that other escapes have more than one character in them, so
+      speculatively set both class_has_8bitchar and class_one_char bigger
+      than one. Unrecognized escapes fall through and are either treated
+      as literal characters (by default), or are faulted if
+      PCRE_EXTRA is set. */
+
+      if (c == CHAR_BACKSLASH)
+        {
+        escape = check_escape(&ptr, &ec, errorcodeptr, cd->bracount, options,
+          TRUE);
+        if (*errorcodeptr != 0) goto FAILED;
+        if (escape == 0) c = ec;
+        else if (escape == ESC_b) c = CHAR_BS; /* \b is backspace in a class */
+        else if (escape == ESC_N)          /* \N is not supported in a class */
+          {
+          *errorcodeptr = ERR71;
+          goto FAILED;
+          }
+        else if (escape == ESC_Q)            /* Handle start of quoted string */
+          {
+          if (ptr[1] == CHAR_BACKSLASH && ptr[2] == CHAR_E)
+            {
+            ptr += 2; /* avoid empty string */
+            }
+          else inescq = TRUE;
+          continue;
+          }
+        else if (escape == ESC_E) continue;  /* Ignore orphan \E */
+
+        else
+          {
+          register const pcre_uint8 *cbits = cd->cbits;
+          /* Every class contains at least two < 256 characters. */
+          class_has_8bitchar++;
+          /* Every class contains at least two characters. */
+          class_one_char += 2;
+
+          switch (escape)
+            {
+#ifdef SUPPORT_UCP
+            case ESC_du:     /* These are the values given for \d etc */
+            case ESC_DU:     /* when PCRE_UCP is set. We replace the */
+            case ESC_wu:     /* escape sequence with an appropriate \p */
+            case ESC_WU:     /* or \P to test Unicode properties instead */
+            case ESC_su:     /* of the default ASCII testing. */
+            case ESC_SU:
+            nestptr = ptr;
+            ptr = substitutes[escape - ESC_DU] - 1;  /* Just before substitute */
+            class_has_8bitchar--;                /* Undo! */
+            continue;
+#endif
+            case ESC_d:
+            for (c = 0; c < 32; c++) classbits[c] |= cbits[c+cbit_digit];
+            continue;
+
+            case ESC_D:
+            should_flip_negation = TRUE;
+            for (c = 0; c < 32; c++) classbits[c] |= ~cbits[c+cbit_digit];
+            continue;
+
+            case ESC_w:
+            for (c = 0; c < 32; c++) classbits[c] |= cbits[c+cbit_word];
+            continue;
+
+            case ESC_W:
+            should_flip_negation = TRUE;
+            for (c = 0; c < 32; c++) classbits[c] |= ~cbits[c+cbit_word];
+            continue;
+
+            /* Perl 5.004 onwards omits VT from \s, but we must preserve it
+            if it was previously set by something earlier in the character
+            class. Luckily, the value of CHAR_VT is 0x0b in both ASCII and
+            EBCDIC, so we lazily just adjust the appropriate bit. */
+
+            case ESC_s:
+            classbits[0] |= cbits[cbit_space];
+            classbits[1] |= cbits[cbit_space+1] & ~0x08;
+            for (c = 2; c < 32; c++) classbits[c] |= cbits[c+cbit_space];
+            continue;
+
+            case ESC_S:
+            should_flip_negation = TRUE;
+            for (c = 0; c < 32; c++) classbits[c] |= ~cbits[c+cbit_space];
+            classbits[1] |= 0x08;    /* Perl 5.004 onwards omits VT from \s */
+            continue;
+
+            /* The rest apply in both UCP and non-UCP cases. */
+
+            case ESC_h:
+            (void)add_list_to_class(classbits, &class_uchardata, options, cd,
+              PRIV(hspace_list), NOTACHAR);
+            continue;
+
+            case ESC_H:
+            (void)add_not_list_to_class(classbits, &class_uchardata, options,
+              cd, PRIV(hspace_list));
+            continue;
+
+            case ESC_v:
+            (void)add_list_to_class(classbits, &class_uchardata, options, cd,
+              PRIV(vspace_list), NOTACHAR);
+            continue;
+
+            case ESC_V:
+            (void)add_not_list_to_class(classbits, &class_uchardata, options,
+              cd, PRIV(vspace_list));
+            continue;
+
+#ifdef SUPPORT_UCP
+            case ESC_p:
+            case ESC_P:
+              {
+              BOOL negated;
+              unsigned int ptype = 0, pdata = 0;
+              if (!get_ucp(&ptr, &negated, &ptype, &pdata, errorcodeptr))
+                goto FAILED;
+              *class_uchardata++ = ((escape == ESC_p) != negated)?
+                XCL_PROP : XCL_NOTPROP;
+              *class_uchardata++ = ptype;
+              *class_uchardata++ = pdata;
+              class_has_8bitchar--;                /* Undo! */
+              continue;
+              }
+#endif
+            /* Unrecognized escapes are faulted if PCRE is running in its
+            strict mode. By default, for compatibility with Perl, they are
+            treated as literals. */
+
+            default:
+            if ((options & PCRE_EXTRA) != 0)
+              {
+              *errorcodeptr = ERR7;
+              goto FAILED;
+              }
+            class_has_8bitchar--;    /* Undo the speculative increase. */
+            class_one_char -= 2;     /* Undo the speculative increase. */
+            c = *ptr;                /* Get the final character and fall through */
+            break;
+            }
+          }
+
+        /* Fall through if the escape just defined a single character (c >= 0).
+        This may be greater than 256. */
+
+        escape = 0;
+
+        }   /* End of backslash handling */
+
+      /* A character may be followed by '-' to form a range. However, Perl does
+      not permit ']' to be the end of the range. A '-' character at the end is
+      treated as a literal. Perl ignores orphaned \E sequences entirely. The
+      code for handling \Q and \E is messy. */
+
+      CHECK_RANGE:
+      while (ptr[1] == CHAR_BACKSLASH && ptr[2] == CHAR_E)
+        {
+        inescq = FALSE;
+        ptr += 2;
+        }
+      oldptr = ptr;
+
+      /* Remember if \r or \n were explicitly used */
+
+      if (c == CHAR_CR || c == CHAR_NL) cd->external_flags |= PCRE_HASCRORLF;
+
+      /* Check for range */
+
+      if (!inescq && ptr[1] == CHAR_MINUS)
+        {
+        pcre_uint32 d;
+        ptr += 2;
+        while (*ptr == CHAR_BACKSLASH && ptr[1] == CHAR_E) ptr += 2;
+
+        /* If we hit \Q (not followed by \E) at this point, go into escaped
+        mode. */
+
+        while (*ptr == CHAR_BACKSLASH && ptr[1] == CHAR_Q)
+          {
+          ptr += 2;
+          if (*ptr == CHAR_BACKSLASH && ptr[1] == CHAR_E)
+            { ptr += 2; continue; }
+          inescq = TRUE;
+          break;
+          }
+
+        /* Minus (hyphen) at the end of a class is treated as a literal, so put
+        back the pointer and jump to handle the character that preceded it. */
+
+        if (*ptr == CHAR_NULL || (!inescq && *ptr == CHAR_RIGHT_SQUARE_BRACKET))
+          {
+          ptr = oldptr;
+          goto CLASS_SINGLE_CHARACTER;
+          }
+
+        /* Otherwise, we have a potential range; pick up the next character */
+
+#ifdef SUPPORT_UTF
+        if (utf)
+          {                           /* Braces are required because the */
+          GETCHARLEN(d, ptr, ptr);    /* macro generates multiple statements */
+          }
+        else
+#endif
+        d = *ptr;  /* Not UTF-8 mode */
+
+        /* The second part of a range can be a single-character escape, but
+        not any of the other escapes. Perl 5.6 treats a hyphen as a literal
+        in such circumstances. */
+
+        if (!inescq && d == CHAR_BACKSLASH)
+          {
+          int descape;
+          descape = check_escape(&ptr, &d, errorcodeptr, cd->bracount, options, TRUE);
+          if (*errorcodeptr != 0) goto FAILED;
+
+          /* \b is backspace; any other special means the '-' was literal. */
+
+          if (descape != 0)
+            {
+            if (descape == ESC_b) d = CHAR_BS; else
+              {
+              ptr = oldptr;
+              goto CLASS_SINGLE_CHARACTER;  /* A few lines below */
+              }
+            }
+          }
+
+        /* Check that the two values are in the correct order. Optimize
+        one-character ranges. */
+
+        if (d < c)
+          {
+          *errorcodeptr = ERR8;
+          goto FAILED;
+          }
+        if (d == c) goto CLASS_SINGLE_CHARACTER;  /* A few lines below */
+
+        /* We have found a character range, so single character optimizations
+        cannot be done anymore. Any value greater than 1 indicates that there
+        is more than one character. */
+
+        class_one_char = 2;
+
+        /* Remember an explicit \r or \n, and add the range to the class. */
+
+        if (d == CHAR_CR || d == CHAR_NL) cd->external_flags |= PCRE_HASCRORLF;
+
+        class_has_8bitchar +=
+          add_to_class(classbits, &class_uchardata, options, cd, c, d);
+
+        continue;   /* Go get the next char in the class */
+        }
+
+      /* Handle a single character - we can get here for a normal non-escape
+      char, or after \ that introduces a single character or for an apparent
+      range that isn't. Only the value 1 matters for class_one_char, so don't
+      increase it if it is already 2 or more ... just in case there's a class
+      with a zillion characters in it. */
+
+      CLASS_SINGLE_CHARACTER:
+      if (class_one_char < 2) class_one_char++;
+
+      /* If class_one_char is 1, we have the first single character in the
+      class, and there have been no prior ranges, or XCLASS items generated by
+      escapes. If this is the final character in the class, we can optimize by
+      turning the item into a 1-character OP_CHAR[I] if it's positive, or
+      OP_NOT[I] if it's negative. In the positive case, it can cause firstchar
+      to be set. Otherwise, there can be no first char if this item is first,
+      whatever repeat count may follow. In the case of reqchar, save the
+      previous value for reinstating. */
+
+      if (class_one_char == 1 && ptr[1] == CHAR_RIGHT_SQUARE_BRACKET)
+        {
+        ptr++;
+        zeroreqchar = reqchar;
+        zeroreqcharflags = reqcharflags;
+
+        if (negate_class)
+          {
+#ifdef SUPPORT_UCP
+          int d;
+#endif
+          if (firstcharflags == REQ_UNSET) firstcharflags = REQ_NONE;
+          zerofirstchar = firstchar;
+          zerofirstcharflags = firstcharflags;
+
+          /* For caseless UTF-8 mode when UCP support is available, check
+          whether this character has more than one other case. If so, generate
+          a special OP_NOTPROP item instead of OP_NOTI. */
+
+#ifdef SUPPORT_UCP
+          if (utf && (options & PCRE_CASELESS) != 0 &&
+              (d = UCD_CASESET(c)) != 0)
+            {
+            *code++ = OP_NOTPROP;
+            *code++ = PT_CLIST;
+            *code++ = d;
+            }
+          else
+#endif
+          /* Char has only one other case, or UCP not available */
+
+            {
+            *code++ = ((options & PCRE_CASELESS) != 0)? OP_NOTI: OP_NOT;
+#if defined SUPPORT_UTF && !defined COMPILE_PCRE32
+            if (utf && c > MAX_VALUE_FOR_SINGLE_CHAR)
+              code += PRIV(ord2utf)(c, code);
+            else
+#endif
+              *code++ = c;
+            }
+
+          /* We are finished with this character class */
+
+          goto END_CLASS;
+          }
+
+        /* For a single, positive character, get the value into mcbuffer, and
+        then we can handle this with the normal one-character code. */
+
+#if defined SUPPORT_UTF && !defined COMPILE_PCRE32
+        if (utf && c > MAX_VALUE_FOR_SINGLE_CHAR)
+          mclength = PRIV(ord2utf)(c, mcbuffer);
+        else
+#endif
+          {
+          mcbuffer[0] = c;
+          mclength = 1;
+          }
+        goto ONE_CHAR;
+        }       /* End of 1-char optimization */
+
+      /* There is more than one character in the class, or an XCLASS item
+      has been generated. Add this character to the class. */
+
+      class_has_8bitchar +=
+        add_to_class(classbits, &class_uchardata, options, cd, c, c);
+      }
+
+    /* Loop until ']' reached. This "while" is the end of the "do" far above.
+    If we are at the end of an internal nested string, revert to the outer
+    string. */
+
+    while (((c = *(++ptr)) != CHAR_NULL ||
+           (nestptr != NULL &&
+             (ptr = nestptr, nestptr = NULL, c = *(++ptr)) != CHAR_NULL)) &&
+           (c != CHAR_RIGHT_SQUARE_BRACKET || inescq));
+
+    /* Check for missing terminating ']' */
+
+    if (c == CHAR_NULL)
+      {
+      *errorcodeptr = ERR6;
+      goto FAILED;
+      }
+
+    /* We will need an XCLASS if data has been placed in class_uchardata. In
+    the second phase this is a sufficient test. However, in the pre-compile
+    phase, class_uchardata gets emptied to prevent workspace overflow, so it
+    only if the very last character in the class needs XCLASS will it contain
+    anything at this point. For this reason, xclass gets set TRUE above when
+    uchar_classdata is emptied, and that's why this code is the way it is here
+    instead of just doing a test on class_uchardata below. */
+
+#if defined SUPPORT_UTF || !defined COMPILE_PCRE8
+    if (class_uchardata > class_uchardata_base) xclass = TRUE;
+#endif
+
+    /* If this is the first thing in the branch, there can be no first char
+    setting, whatever the repeat count. Any reqchar setting must remain
+    unchanged after any kind of repeat. */
+
+    if (firstcharflags == REQ_UNSET) firstcharflags = REQ_NONE;
+    zerofirstchar = firstchar;
+    zerofirstcharflags = firstcharflags;
+    zeroreqchar = reqchar;
+    zeroreqcharflags = reqcharflags;
+
+    /* If there are characters with values > 255, we have to compile an
+    extended class, with its own opcode, unless there was a negated special
+    such as \S in the class, and PCRE_UCP is not set, because in that case all
+    characters > 255 are in the class, so any that were explicitly given as
+    well can be ignored. If (when there are explicit characters > 255 that must
+    be listed) there are no characters < 256, we can omit the bitmap in the
+    actual compiled code. */
+
+#ifdef SUPPORT_UTF
+    if (xclass && (!should_flip_negation || (options & PCRE_UCP) != 0))
+#elif !defined COMPILE_PCRE8
+    if (xclass && !should_flip_negation)
+#endif
+#if defined SUPPORT_UTF || !defined COMPILE_PCRE8
+      {
+      *class_uchardata++ = XCL_END;    /* Marks the end of extra data */
+      *code++ = OP_XCLASS;
+      code += LINK_SIZE;
+      *code = negate_class? XCL_NOT:0;
+
+      /* If the map is required, move up the extra data to make room for it;
+      otherwise just move the code pointer to the end of the extra data. */
+
+      if (class_has_8bitchar > 0)
+        {
+        *code++ |= XCL_MAP;
+        memmove(code + (32 / sizeof(pcre_uchar)), code,
+          IN_UCHARS(class_uchardata - code));
+        memcpy(code, classbits, 32);
+        code = class_uchardata + (32 / sizeof(pcre_uchar));
+        }
+      else code = class_uchardata;
+
+      /* Now fill in the complete length of the item */
+
+      PUT(previous, 1, (int)(code - previous));
+      break;   /* End of class handling */
+      }
+#endif
+
+    /* If there are no characters > 255, or they are all to be included or
+    excluded, set the opcode to OP_CLASS or OP_NCLASS, depending on whether the
+    whole class was negated and whether there were negative specials such as \S
+    (non-UCP) in the class. Then copy the 32-byte map into the code vector,
+    negating it if necessary. */
+
+    *code++ = (negate_class == should_flip_negation) ? OP_CLASS : OP_NCLASS;
+    if (lengthptr == NULL)    /* Save time in the pre-compile phase */
+      {
+      if (negate_class)
+        for (c = 0; c < 32; c++) classbits[c] = ~classbits[c];
+      memcpy(code, classbits, 32);
+      }
+    code += 32 / sizeof(pcre_uchar);
+
+    END_CLASS:
+    break;
+
+
+    /* ===================================================================*/
+    /* Various kinds of repeat; '{' is not necessarily a quantifier, but this
+    has been tested above. */
+
+    case CHAR_LEFT_CURLY_BRACKET:
+    if (!is_quantifier) goto NORMAL_CHAR;
+    ptr = read_repeat_counts(ptr+1, &repeat_min, &repeat_max, errorcodeptr);
+    if (*errorcodeptr != 0) goto FAILED;
+    goto REPEAT;
+
+    case CHAR_ASTERISK:
+    repeat_min = 0;
+    repeat_max = -1;
+    goto REPEAT;
+
+    case CHAR_PLUS:
+    repeat_min = 1;
+    repeat_max = -1;
+    goto REPEAT;
+
+    case CHAR_QUESTION_MARK:
+    repeat_min = 0;
+    repeat_max = 1;
+
+    REPEAT:
+    if (previous == NULL)
+      {
+      *errorcodeptr = ERR9;
+      goto FAILED;
+      }
+
+    if (repeat_min == 0)
+      {
+      firstchar = zerofirstchar;    /* Adjust for zero repeat */
+      firstcharflags = zerofirstcharflags;
+      reqchar = zeroreqchar;        /* Ditto */
+      reqcharflags = zeroreqcharflags;
+      }
+
+    /* Remember whether this is a variable length repeat */
+
+    reqvary = (repeat_min == repeat_max)? 0 : REQ_VARY;
+
+    op_type = 0;                    /* Default single-char op codes */
+    possessive_quantifier = FALSE;  /* Default not possessive quantifier */
+
+    /* Save start of previous item, in case we have to move it up in order to
+    insert something before it. */
+
+    tempcode = previous;
+
+    /* If the next character is '+', we have a possessive quantifier. This
+    implies greediness, whatever the setting of the PCRE_UNGREEDY option.
+    If the next character is '?' this is a minimizing repeat, by default,
+    but if PCRE_UNGREEDY is set, it works the other way round. We change the
+    repeat type to the non-default. */
+
+    if (ptr[1] == CHAR_PLUS)
+      {
+      repeat_type = 0;                  /* Force greedy */
+      possessive_quantifier = TRUE;
+      ptr++;
+      }
+    else if (ptr[1] == CHAR_QUESTION_MARK)
+      {
+      repeat_type = greedy_non_default;
+      ptr++;
+      }
+    else repeat_type = greedy_default;
+
+    /* If previous was a recursion call, wrap it in atomic brackets so that
+    previous becomes the atomic group. All recursions were so wrapped in the
+    past, but it no longer happens for non-repeated recursions. In fact, the
+    repeated ones could be re-implemented independently so as not to need this,
+    but for the moment we rely on the code for repeating groups. */
+
+    if (*previous == OP_RECURSE)
+      {
+      memmove(previous + 1 + LINK_SIZE, previous, IN_UCHARS(1 + LINK_SIZE));
+      *previous = OP_ONCE;
+      PUT(previous, 1, 2 + 2*LINK_SIZE);
+      previous[2 + 2*LINK_SIZE] = OP_KET;
+      PUT(previous, 3 + 2*LINK_SIZE, 2 + 2*LINK_SIZE);
+      code += 2 + 2 * LINK_SIZE;
+      length_prevgroup = 3 + 3*LINK_SIZE;
+
+      /* When actually compiling, we need to check whether this was a forward
+      reference, and if so, adjust the offset. */
+
+      if (lengthptr == NULL && cd->hwm >= cd->start_workspace + LINK_SIZE)
+        {
+        int offset = GET(cd->hwm, -LINK_SIZE);
+        if (offset == previous + 1 - cd->start_code)
+          PUT(cd->hwm, -LINK_SIZE, offset + 1 + LINK_SIZE);
+        }
+      }
+
+    /* Now handle repetition for the different types of item. */
+
+    /* If previous was a character or negated character match, abolish the item
+    and generate a repeat item instead. If a char item has a minimum of more
+    than one, ensure that it is set in reqchar - it might not be if a sequence
+    such as x{3} is the first thing in a branch because the x will have gone
+    into firstchar instead.  */
+
+    if (*previous == OP_CHAR || *previous == OP_CHARI
+        || *previous == OP_NOT || *previous == OP_NOTI)
+      {
+      switch (*previous)
+        {
+        default: /* Make compiler happy. */
+        case OP_CHAR:  op_type = OP_STAR - OP_STAR; break;
+        case OP_CHARI: op_type = OP_STARI - OP_STAR; break;
+        case OP_NOT:   op_type = OP_NOTSTAR - OP_STAR; break;
+        case OP_NOTI:  op_type = OP_NOTSTARI - OP_STAR; break;
+        }
+
+      /* Deal with UTF characters that take up more than one character. It's
+      easier to write this out separately than try to macrify it. Use c to
+      hold the length of the character in bytes, plus UTF_LENGTH to flag that
+      it's a length rather than a small character. */
+
+#if defined SUPPORT_UTF && !defined COMPILE_PCRE32
+      if (utf && NOT_FIRSTCHAR(code[-1]))
+        {
+        pcre_uchar *lastchar = code - 1;
+        BACKCHAR(lastchar);
+        c = (int)(code - lastchar);     /* Length of UTF-8 character */
+        memcpy(utf_chars, lastchar, IN_UCHARS(c)); /* Save the char */
+        c |= UTF_LENGTH;                /* Flag c as a length */
+        }
+      else
+#endif /* SUPPORT_UTF */
+
+      /* Handle the case of a single charater - either with no UTF support, or
+      with UTF disabled, or for a single character UTF character. */
+        {
+        c = code[-1];
+        if (*previous <= OP_CHARI && repeat_min > 1)
+          {
+          reqchar = c;
+          reqcharflags = req_caseopt | cd->req_varyopt;
+          }
+        }
+
+      /* If the repetition is unlimited, it pays to see if the next thing on
+      the line is something that cannot possibly match this character. If so,
+      automatically possessifying this item gains some performance in the case
+      where the match fails. */
+
+      if (!possessive_quantifier &&
+          repeat_max < 0 &&
+          check_auto_possessive(previous, utf, ptr + 1, options, cd))
+        {
+        repeat_type = 0;    /* Force greedy */
+        possessive_quantifier = TRUE;
+        }
+
+      goto OUTPUT_SINGLE_REPEAT;   /* Code shared with single character types */
+      }
+
+    /* If previous was a character type match (\d or similar), abolish it and
+    create a suitable repeat item. The code is shared with single-character
+    repeats by setting op_type to add a suitable offset into repeat_type. Note
+    the the Unicode property types will be present only when SUPPORT_UCP is
+    defined, but we don't wrap the little bits of code here because it just
+    makes it horribly messy. */
+
+    else if (*previous < OP_EODN)
+      {
+      pcre_uchar *oldcode;
+      int prop_type, prop_value;
+      op_type = OP_TYPESTAR - OP_STAR;  /* Use type opcodes */
+      c = *previous;
+
+      if (!possessive_quantifier &&
+          repeat_max < 0 &&
+          check_auto_possessive(previous, utf, ptr + 1, options, cd))
+        {
+        repeat_type = 0;    /* Force greedy */
+        possessive_quantifier = TRUE;
+        }
+
+      OUTPUT_SINGLE_REPEAT:
+      if (*previous == OP_PROP || *previous == OP_NOTPROP)
+        {
+        prop_type = previous[1];
+        prop_value = previous[2];
+        }
+      else prop_type = prop_value = -1;
+
+      oldcode = code;
+      code = previous;                  /* Usually overwrite previous item */
+
+      /* If the maximum is zero then the minimum must also be zero; Perl allows
+      this case, so we do too - by simply omitting the item altogether. */
+
+      if (repeat_max == 0) goto END_REPEAT;
+
+      /* Combine the op_type with the repeat_type */
+
+      repeat_type += op_type;
+
+      /* A minimum of zero is handled either as the special case * or ?, or as
+      an UPTO, with the maximum given. */
+
+      if (repeat_min == 0)
+        {
+        if (repeat_max == -1) *code++ = OP_STAR + repeat_type;
+          else if (repeat_max == 1) *code++ = OP_QUERY + repeat_type;
+        else
+          {
+          *code++ = OP_UPTO + repeat_type;
+          PUT2INC(code, 0, repeat_max);
+          }
+        }
+
+      /* A repeat minimum of 1 is optimized into some special cases. If the
+      maximum is unlimited, we use OP_PLUS. Otherwise, the original item is
+      left in place and, if the maximum is greater than 1, we use OP_UPTO with
+      one less than the maximum. */
+
+      else if (repeat_min == 1)
+        {
+        if (repeat_max == -1)
+          *code++ = OP_PLUS + repeat_type;
+        else
+          {
+          code = oldcode;                 /* leave previous item in place */
+          if (repeat_max == 1) goto END_REPEAT;
+          *code++ = OP_UPTO + repeat_type;
+          PUT2INC(code, 0, repeat_max - 1);
+          }
+        }
+
+      /* The case {n,n} is just an EXACT, while the general case {n,m} is
+      handled as an EXACT followed by an UPTO. */
+
+      else
+        {
+        *code++ = OP_EXACT + op_type;  /* NB EXACT doesn't have repeat_type */
+        PUT2INC(code, 0, repeat_min);
+
+        /* If the maximum is unlimited, insert an OP_STAR. Before doing so,
+        we have to insert the character for the previous code. For a repeated
+        Unicode property match, there are two extra bytes that define the
+        required property. In UTF-8 mode, long characters have their length in
+        c, with the UTF_LENGTH bit as a flag. */
+
+        if (repeat_max < 0)
+          {
+#if defined SUPPORT_UTF && !defined COMPILE_PCRE32
+          if (utf && (c & UTF_LENGTH) != 0)
+            {
+            memcpy(code, utf_chars, IN_UCHARS(c & 7));
+            code += c & 7;
+            }
+          else
+#endif
+            {
+            *code++ = c;
+            if (prop_type >= 0)
+              {
+              *code++ = prop_type;
+              *code++ = prop_value;
+              }
+            }
+          *code++ = OP_STAR + repeat_type;
+          }
+
+        /* Else insert an UPTO if the max is greater than the min, again
+        preceded by the character, for the previously inserted code. If the
+        UPTO is just for 1 instance, we can use QUERY instead. */
+
+        else if (repeat_max != repeat_min)
+          {
+#if defined SUPPORT_UTF && !defined COMPILE_PCRE32
+          if (utf && (c & UTF_LENGTH) != 0)
+            {
+            memcpy(code, utf_chars, IN_UCHARS(c & 7));
+            code += c & 7;
+            }
+          else
+#endif
+          *code++ = c;
+          if (prop_type >= 0)
+            {
+            *code++ = prop_type;
+            *code++ = prop_value;
+            }
+          repeat_max -= repeat_min;
+
+          if (repeat_max == 1)
+            {
+            *code++ = OP_QUERY + repeat_type;
+            }
+          else
+            {
+            *code++ = OP_UPTO + repeat_type;
+            PUT2INC(code, 0, repeat_max);
+            }
+          }
+        }
+
+      /* The character or character type itself comes last in all cases. */
+
+#if defined SUPPORT_UTF && !defined COMPILE_PCRE32
+      if (utf && (c & UTF_LENGTH) != 0)
+        {
+        memcpy(code, utf_chars, IN_UCHARS(c & 7));
+        code += c & 7;
+        }
+      else
+#endif
+      *code++ = c;
+
+      /* For a repeated Unicode property match, there are two extra bytes that
+      define the required property. */
+
+#ifdef SUPPORT_UCP
+      if (prop_type >= 0)
+        {
+        *code++ = prop_type;
+        *code++ = prop_value;
+        }
+#endif
+      }
+
+    /* If previous was a character class or a back reference, we put the repeat
+    stuff after it, but just skip the item if the repeat was {0,0}. */
+
+    else if (*previous == OP_CLASS ||
+             *previous == OP_NCLASS ||
+#if defined SUPPORT_UTF || !defined COMPILE_PCRE8
+             *previous == OP_XCLASS ||
+#endif
+             *previous == OP_REF ||
+             *previous == OP_REFI)
+      {
+      if (repeat_max == 0)
+        {
+        code = previous;
+        goto END_REPEAT;
+        }
+
+      if (repeat_min == 0 && repeat_max == -1)
+        *code++ = OP_CRSTAR + repeat_type;
+      else if (repeat_min == 1 && repeat_max == -1)
+        *code++ = OP_CRPLUS + repeat_type;
+      else if (repeat_min == 0 && repeat_max == 1)
+        *code++ = OP_CRQUERY + repeat_type;
+      else
+        {
+        *code++ = OP_CRRANGE + repeat_type;
+        PUT2INC(code, 0, repeat_min);
+        if (repeat_max == -1) repeat_max = 0;  /* 2-byte encoding for max */
+        PUT2INC(code, 0, repeat_max);
+        }
+      }
+
+    /* If previous was a bracket group, we may have to replicate it in certain
+    cases. Note that at this point we can encounter only the "basic" bracket
+    opcodes such as BRA and CBRA, as this is the place where they get converted
+    into the more special varieties such as BRAPOS and SBRA. A test for >=
+    OP_ASSERT and <= OP_COND includes ASSERT, ASSERT_NOT, ASSERTBACK,
+    ASSERTBACK_NOT, ONCE, BRA, CBRA, and COND. Originally, PCRE did not allow
+    repetition of assertions, but now it does, for Perl compatibility. */
+
+    else if (*previous >= OP_ASSERT && *previous <= OP_COND)
+      {
+      register int i;
+      int len = (int)(code - previous);
+      pcre_uchar *bralink = NULL;
+      pcre_uchar *brazeroptr = NULL;
+
+      /* Repeating a DEFINE group is pointless, but Perl allows the syntax, so
+      we just ignore the repeat. */
+
+      if (*previous == OP_COND && previous[LINK_SIZE+1] == OP_DEF)
+        goto END_REPEAT;
+
+      /* There is no sense in actually repeating assertions. The only potential
+      use of repetition is in cases when the assertion is optional. Therefore,
+      if the minimum is greater than zero, just ignore the repeat. If the
+      maximum is not not zero or one, set it to 1. */
+
+      if (*previous < OP_ONCE)    /* Assertion */
+        {
+        if (repeat_min > 0) goto END_REPEAT;
+        if (repeat_max < 0 || repeat_max > 1) repeat_max = 1;
+        }
+
+      /* The case of a zero minimum is special because of the need to stick
+      OP_BRAZERO in front of it, and because the group appears once in the
+      data, whereas in other cases it appears the minimum number of times. For
+      this reason, it is simplest to treat this case separately, as otherwise
+      the code gets far too messy. There are several special subcases when the
+      minimum is zero. */
+
+      if (repeat_min == 0)
+        {
+        /* If the maximum is also zero, we used to just omit the group from the
+        output altogether, like this:
+
+        ** if (repeat_max == 0)
+        **   {
+        **   code = previous;
+        **   goto END_REPEAT;
+        **   }
+
+        However, that fails when a group or a subgroup within it is referenced
+        as a subroutine from elsewhere in the pattern, so now we stick in
+        OP_SKIPZERO in front of it so that it is skipped on execution. As we
+        don't have a list of which groups are referenced, we cannot do this
+        selectively.
+
+        If the maximum is 1 or unlimited, we just have to stick in the BRAZERO
+        and do no more at this point. However, we do need to adjust any
+        OP_RECURSE calls inside the group that refer to the group itself or any
+        internal or forward referenced group, because the offset is from the
+        start of the whole regex. Temporarily terminate the pattern while doing
+        this. */
+
+        if (repeat_max <= 1)    /* Covers 0, 1, and unlimited */
+          {
+          *code = OP_END;
+          adjust_recurse(previous, 1, utf, cd, save_hwm);
+          memmove(previous + 1, previous, IN_UCHARS(len));
+          code++;
+          if (repeat_max == 0)
+            {
+            *previous++ = OP_SKIPZERO;
+            goto END_REPEAT;
+            }
+          brazeroptr = previous;    /* Save for possessive optimizing */
+          *previous++ = OP_BRAZERO + repeat_type;
+          }
+
+        /* If the maximum is greater than 1 and limited, we have to replicate
+        in a nested fashion, sticking OP_BRAZERO before each set of brackets.
+        The first one has to be handled carefully because it's the original
+        copy, which has to be moved up. The remainder can be handled by code
+        that is common with the non-zero minimum case below. We have to
+        adjust the value or repeat_max, since one less copy is required. Once
+        again, we may have to adjust any OP_RECURSE calls inside the group. */
+
+        else
+          {
+          int offset;
+          *code = OP_END;
+          adjust_recurse(previous, 2 + LINK_SIZE, utf, cd, save_hwm);
+          memmove(previous + 2 + LINK_SIZE, previous, IN_UCHARS(len));
+          code += 2 + LINK_SIZE;
+          *previous++ = OP_BRAZERO + repeat_type;
+          *previous++ = OP_BRA;
+
+          /* We chain together the bracket offset fields that have to be
+          filled in later when the ends of the brackets are reached. */
+
+          offset = (bralink == NULL)? 0 : (int)(previous - bralink);
+          bralink = previous;
+          PUTINC(previous, 0, offset);
+          }
+
+        repeat_max--;
+        }
+
+      /* If the minimum is greater than zero, replicate the group as many
+      times as necessary, and adjust the maximum to the number of subsequent
+      copies that we need. If we set a first char from the group, and didn't
+      set a required char, copy the latter from the former. If there are any
+      forward reference subroutine calls in the group, there will be entries on
+      the workspace list; replicate these with an appropriate increment. */
+
+      else
+        {
+        if (repeat_min > 1)
+          {
+          /* In the pre-compile phase, we don't actually do the replication. We
+          just adjust the length as if we had. Do some paranoid checks for
+          potential integer overflow. The INT64_OR_DOUBLE type is a 64-bit
+          integer type when available, otherwise double. */
+
+          if (lengthptr != NULL)
+            {
+            int delta = (repeat_min - 1)*length_prevgroup;
+            if ((INT64_OR_DOUBLE)(repeat_min - 1)*
+                  (INT64_OR_DOUBLE)length_prevgroup >
+                    (INT64_OR_DOUBLE)INT_MAX ||
+                OFLOW_MAX - *lengthptr < delta)
+              {
+              *errorcodeptr = ERR20;
+              goto FAILED;
+              }
+            *lengthptr += delta;
+            }
+
+          /* This is compiling for real. If there is a set first byte for
+          the group, and we have not yet set a "required byte", set it. Make
+          sure there is enough workspace for copying forward references before
+          doing the copy. */
+
+          else
+            {
+            if (groupsetfirstchar && reqcharflags < 0)
+              {
+              reqchar = firstchar;
+              reqcharflags = firstcharflags;
+              }
+
+            for (i = 1; i < repeat_min; i++)
+              {
+              pcre_uchar *hc;
+              pcre_uchar *this_hwm = cd->hwm;
+              memcpy(code, previous, IN_UCHARS(len));
+
+              while (cd->hwm > cd->start_workspace + cd->workspace_size -
+                     WORK_SIZE_SAFETY_MARGIN - (this_hwm - save_hwm))
+                {
+                int save_offset = save_hwm - cd->start_workspace;
+                int this_offset = this_hwm - cd->start_workspace;
+                *errorcodeptr = expand_workspace(cd);
+                if (*errorcodeptr != 0) goto FAILED;
+                save_hwm = (pcre_uchar *)cd->start_workspace + save_offset;
+                this_hwm = (pcre_uchar *)cd->start_workspace + this_offset;
+                }
+
+              for (hc = save_hwm; hc < this_hwm; hc += LINK_SIZE)
+                {
+                PUT(cd->hwm, 0, GET(hc, 0) + len);
+                cd->hwm += LINK_SIZE;
+                }
+              save_hwm = this_hwm;
+              code += len;
+              }
+            }
+          }
+
+        if (repeat_max > 0) repeat_max -= repeat_min;
+        }
+
+      /* This code is common to both the zero and non-zero minimum cases. If
+      the maximum is limited, it replicates the group in a nested fashion,
+      remembering the bracket starts on a stack. In the case of a zero minimum,
+      the first one was set up above. In all cases the repeat_max now specifies
+      the number of additional copies needed. Again, we must remember to
+      replicate entries on the forward reference list. */
+
+      if (repeat_max >= 0)
+        {
+        /* In the pre-compile phase, we don't actually do the replication. We
+        just adjust the length as if we had. For each repetition we must add 1
+        to the length for BRAZERO and for all but the last repetition we must
+        add 2 + 2*LINKSIZE to allow for the nesting that occurs. Do some
+        paranoid checks to avoid integer overflow. The INT64_OR_DOUBLE type is
+        a 64-bit integer type when available, otherwise double. */
+
+        if (lengthptr != NULL && repeat_max > 0)
+          {
+          int delta = repeat_max * (length_prevgroup + 1 + 2 + 2*LINK_SIZE) -
+                      2 - 2*LINK_SIZE;   /* Last one doesn't nest */
+          if ((INT64_OR_DOUBLE)repeat_max *
+                (INT64_OR_DOUBLE)(length_prevgroup + 1 + 2 + 2*LINK_SIZE)
+                  > (INT64_OR_DOUBLE)INT_MAX ||
+              OFLOW_MAX - *lengthptr < delta)
+            {
+            *errorcodeptr = ERR20;
+            goto FAILED;
+            }
+          *lengthptr += delta;
+          }
+
+        /* This is compiling for real */
+
+        else for (i = repeat_max - 1; i >= 0; i--)
+          {
+          pcre_uchar *hc;
+          pcre_uchar *this_hwm = cd->hwm;
+
+          *code++ = OP_BRAZERO + repeat_type;
+
+          /* All but the final copy start a new nesting, maintaining the
+          chain of brackets outstanding. */
+
+          if (i != 0)
+            {
+            int offset;
+            *code++ = OP_BRA;
+            offset = (bralink == NULL)? 0 : (int)(code - bralink);
+            bralink = code;
+            PUTINC(code, 0, offset);
+            }
+
+          memcpy(code, previous, IN_UCHARS(len));
+
+          /* Ensure there is enough workspace for forward references before
+          copying them. */
+
+          while (cd->hwm > cd->start_workspace + cd->workspace_size -
+                 WORK_SIZE_SAFETY_MARGIN - (this_hwm - save_hwm))
+            {
+            int save_offset = save_hwm - cd->start_workspace;
+            int this_offset = this_hwm - cd->start_workspace;
+            *errorcodeptr = expand_workspace(cd);
+            if (*errorcodeptr != 0) goto FAILED;
+            save_hwm = (pcre_uchar *)cd->start_workspace + save_offset;
+            this_hwm = (pcre_uchar *)cd->start_workspace + this_offset;
+            }
+
+          for (hc = save_hwm; hc < this_hwm; hc += LINK_SIZE)
+            {
+            PUT(cd->hwm, 0, GET(hc, 0) + len + ((i != 0)? 2+LINK_SIZE : 1));
+            cd->hwm += LINK_SIZE;
+            }
+          save_hwm = this_hwm;
+          code += len;
+          }
+
+        /* Now chain through the pending brackets, and fill in their length
+        fields (which are holding the chain links pro tem). */
+
+        while (bralink != NULL)
+          {
+          int oldlinkoffset;
+          int offset = (int)(code - bralink + 1);
+          pcre_uchar *bra = code - offset;
+          oldlinkoffset = GET(bra, 1);
+          bralink = (oldlinkoffset == 0)? NULL : bralink - oldlinkoffset;
+          *code++ = OP_KET;
+          PUTINC(code, 0, offset);
+          PUT(bra, 1, offset);
+          }
+        }
+
+      /* If the maximum is unlimited, set a repeater in the final copy. For
+      ONCE brackets, that's all we need to do. However, possessively repeated
+      ONCE brackets can be converted into non-capturing brackets, as the
+      behaviour of (?:xx)++ is the same as (?>xx)++ and this saves having to
+      deal with possessive ONCEs specially.
+
+      Otherwise, when we are doing the actual compile phase, check to see
+      whether this group is one that could match an empty string. If so,
+      convert the initial operator to the S form (e.g. OP_BRA -> OP_SBRA) so
+      that runtime checking can be done. [This check is also applied to ONCE
+      groups at runtime, but in a different way.]
+
+      Then, if the quantifier was possessive and the bracket is not a
+      conditional, we convert the BRA code to the POS form, and the KET code to
+      KETRPOS. (It turns out to be convenient at runtime to detect this kind of
+      subpattern at both the start and at the end.) The use of special opcodes
+      makes it possible to reduce greatly the stack usage in pcre_exec(). If
+      the group is preceded by OP_BRAZERO, convert this to OP_BRAPOSZERO.
+
+      Then, if the minimum number of matches is 1 or 0, cancel the possessive
+      flag so that the default action below, of wrapping everything inside
+      atomic brackets, does not happen. When the minimum is greater than 1,
+      there will be earlier copies of the group, and so we still have to wrap
+      the whole thing. */
+
+      else
+        {
+        pcre_uchar *ketcode = code - 1 - LINK_SIZE;
+        pcre_uchar *bracode = ketcode - GET(ketcode, 1);
+
+        /* Convert possessive ONCE brackets to non-capturing */
+
+        if ((*bracode == OP_ONCE || *bracode == OP_ONCE_NC) &&
+            possessive_quantifier) *bracode = OP_BRA;
+
+        /* For non-possessive ONCE brackets, all we need to do is to
+        set the KET. */
+
+        if (*bracode == OP_ONCE || *bracode == OP_ONCE_NC)
+          *ketcode = OP_KETRMAX + repeat_type;
+
+        /* Handle non-ONCE brackets and possessive ONCEs (which have been
+        converted to non-capturing above). */
+
+        else
+          {
+          /* In the compile phase, check for empty string matching. */
+
+          if (lengthptr == NULL)
+            {
+            pcre_uchar *scode = bracode;
+            do
+              {
+              if (could_be_empty_branch(scode, ketcode, utf, cd))
+                {
+                *bracode += OP_SBRA - OP_BRA;
+                break;
+                }
+              scode += GET(scode, 1);
+              }
+            while (*scode == OP_ALT);
+            }
+
+          /* Handle possessive quantifiers. */
+
+          if (possessive_quantifier)
+            {
+            /* For COND brackets, we wrap the whole thing in a possessively
+            repeated non-capturing bracket, because we have not invented POS
+            versions of the COND opcodes. Because we are moving code along, we
+            must ensure that any pending recursive references are updated. */
+
+            if (*bracode == OP_COND || *bracode == OP_SCOND)
+              {
+              int nlen = (int)(code - bracode);
+              *code = OP_END;
+              adjust_recurse(bracode, 1 + LINK_SIZE, utf, cd, save_hwm);
+              memmove(bracode + 1 + LINK_SIZE, bracode, IN_UCHARS(nlen));
+              code += 1 + LINK_SIZE;
+              nlen += 1 + LINK_SIZE;
+              *bracode = OP_BRAPOS;
+              *code++ = OP_KETRPOS;
+              PUTINC(code, 0, nlen);
+              PUT(bracode, 1, nlen);
+              }
+
+            /* For non-COND brackets, we modify the BRA code and use KETRPOS. */
+
+            else
+              {
+              *bracode += 1;              /* Switch to xxxPOS opcodes */
+              *ketcode = OP_KETRPOS;
+              }
+
+            /* If the minimum is zero, mark it as possessive, then unset the
+            possessive flag when the minimum is 0 or 1. */
+
+            if (brazeroptr != NULL) *brazeroptr = OP_BRAPOSZERO;
+            if (repeat_min < 2) possessive_quantifier = FALSE;
+            }
+
+          /* Non-possessive quantifier */
+
+          else *ketcode = OP_KETRMAX + repeat_type;
+          }
+        }
+      }
+
+    /* If previous is OP_FAIL, it was generated by an empty class [] in
+    JavaScript mode. The other ways in which OP_FAIL can be generated, that is
+    by (*FAIL) or (?!) set previous to NULL, which gives a "nothing to repeat"
+    error above. We can just ignore the repeat in JS case. */
+
+    else if (*previous == OP_FAIL) goto END_REPEAT;
+
+    /* Else there's some kind of shambles */
+
+    else
+      {
+      *errorcodeptr = ERR11;
+      goto FAILED;
+      }
+
+    /* If the character following a repeat is '+', or if certain optimization
+    tests above succeeded, possessive_quantifier is TRUE. For some opcodes,
+    there are special alternative opcodes for this case. For anything else, we
+    wrap the entire repeated item inside OP_ONCE brackets. Logically, the '+'
+    notation is just syntactic sugar, taken from Sun's Java package, but the
+    special opcodes can optimize it.
+
+    Some (but not all) possessively repeated subpatterns have already been
+    completely handled in the code just above. For them, possessive_quantifier
+    is always FALSE at this stage.
+
+    Note that the repeated item starts at tempcode, not at previous, which
+    might be the first part of a string whose (former) last char we repeated.
+
+    Possessifying an 'exact' quantifier has no effect, so we can ignore it. But
+    an 'upto' may follow. We skip over an 'exact' item, and then test the
+    length of what remains before proceeding. */
+
+    if (possessive_quantifier)
+      {
+      int len;
+
+      if (*tempcode == OP_TYPEEXACT)
+        tempcode += PRIV(OP_lengths)[*tempcode] +
+          ((tempcode[1 + IMM2_SIZE] == OP_PROP
+          || tempcode[1 + IMM2_SIZE] == OP_NOTPROP)? 2 : 0);
+
+      else if (*tempcode == OP_EXACT || *tempcode == OP_NOTEXACT)
+        {
+        tempcode += PRIV(OP_lengths)[*tempcode];
+#ifdef SUPPORT_UTF
+        if (utf && HAS_EXTRALEN(tempcode[-1]))
+          tempcode += GET_EXTRALEN(tempcode[-1]);
+#endif
+        }
+
+      len = (int)(code - tempcode);
+      if (len > 0) switch (*tempcode)
+        {
+        case OP_STAR:  *tempcode = OP_POSSTAR; break;
+        case OP_PLUS:  *tempcode = OP_POSPLUS; break;
+        case OP_QUERY: *tempcode = OP_POSQUERY; break;
+        case OP_UPTO:  *tempcode = OP_POSUPTO; break;
+
+        case OP_STARI:  *tempcode = OP_POSSTARI; break;
+        case OP_PLUSI:  *tempcode = OP_POSPLUSI; break;
+        case OP_QUERYI: *tempcode = OP_POSQUERYI; break;
+        case OP_UPTOI:  *tempcode = OP_POSUPTOI; break;
+
+        case OP_NOTSTAR:  *tempcode = OP_NOTPOSSTAR; break;
+        case OP_NOTPLUS:  *tempcode = OP_NOTPOSPLUS; break;
+        case OP_NOTQUERY: *tempcode = OP_NOTPOSQUERY; break;
+        case OP_NOTUPTO:  *tempcode = OP_NOTPOSUPTO; break;
+
+        case OP_NOTSTARI:  *tempcode = OP_NOTPOSSTARI; break;
+        case OP_NOTPLUSI:  *tempcode = OP_NOTPOSPLUSI; break;
+        case OP_NOTQUERYI: *tempcode = OP_NOTPOSQUERYI; break;
+        case OP_NOTUPTOI:  *tempcode = OP_NOTPOSUPTOI; break;
+
+        case OP_TYPESTAR:  *tempcode = OP_TYPEPOSSTAR; break;
+        case OP_TYPEPLUS:  *tempcode = OP_TYPEPOSPLUS; break;
+        case OP_TYPEQUERY: *tempcode = OP_TYPEPOSQUERY; break;
+        case OP_TYPEUPTO:  *tempcode = OP_TYPEPOSUPTO; break;
+
+        /* Because we are moving code along, we must ensure that any
+        pending recursive references are updated. */
+
+        default:
+        *code = OP_END;
+        adjust_recurse(tempcode, 1 + LINK_SIZE, utf, cd, save_hwm);
+        memmove(tempcode + 1 + LINK_SIZE, tempcode, IN_UCHARS(len));
+        code += 1 + LINK_SIZE;
+        len += 1 + LINK_SIZE;
+        tempcode[0] = OP_ONCE;
+        *code++ = OP_KET;
+        PUTINC(code, 0, len);
+        PUT(tempcode, 1, len);
+        break;
+        }
+      }
+
+    /* In all case we no longer have a previous item. We also set the
+    "follows varying string" flag for subsequently encountered reqchars if
+    it isn't already set and we have just passed a varying length item. */
+
+    END_REPEAT:
+    previous = NULL;
+    cd->req_varyopt |= reqvary;
+    break;
+
+
+    /* ===================================================================*/
+    /* Start of nested parenthesized sub-expression, or comment or lookahead or
+    lookbehind or option setting or condition or all the other extended
+    parenthesis forms.  */
+
+    case CHAR_LEFT_PARENTHESIS:
+    newoptions = options;
+    skipbytes = 0;
+    bravalue = OP_CBRA;
+    save_hwm = cd->hwm;
+    reset_bracount = FALSE;
+
+    /* First deal with various "verbs" that can be introduced by '*'. */
+
+    ptr++;
+    if (ptr[0] == CHAR_ASTERISK && (ptr[1] == ':'
+         || (MAX_255(ptr[1]) && ((cd->ctypes[ptr[1]] & ctype_letter) != 0))))
+      {
+      int i, namelen;
+      int arglen = 0;
+      const char *vn = verbnames;
+      const pcre_uchar *name = ptr + 1;
+      const pcre_uchar *arg = NULL;
+      previous = NULL;
+      ptr++;
+      while (MAX_255(*ptr) && (cd->ctypes[*ptr] & ctype_letter) != 0) ptr++;
+      namelen = (int)(ptr - name);
+
+      /* It appears that Perl allows any characters whatsoever, other than
+      a closing parenthesis, to appear in arguments, so we no longer insist on
+      letters, digits, and underscores. */
+
+      if (*ptr == CHAR_COLON)
+        {
+        arg = ++ptr;
+        while (*ptr != CHAR_NULL && *ptr != CHAR_RIGHT_PARENTHESIS) ptr++;
+        arglen = (int)(ptr - arg);
+        if ((unsigned int)arglen > MAX_MARK)
+          {
+          *errorcodeptr = ERR75;
+          goto FAILED;
+          }
+        }
+
+      if (*ptr != CHAR_RIGHT_PARENTHESIS)
+        {
+        *errorcodeptr = ERR60;
+        goto FAILED;
+        }
+
+      /* Scan the table of verb names */
+
+      for (i = 0; i < verbcount; i++)
+        {
+        if (namelen == verbs[i].len &&
+            STRNCMP_UC_C8(name, vn, namelen) == 0)
+          {
+          int setverb;
+
+          /* Check for open captures before ACCEPT and convert it to
+          ASSERT_ACCEPT if in an assertion. */
+
+          if (verbs[i].op == OP_ACCEPT)
+            {
+            open_capitem *oc;
+            if (arglen != 0)
+              {
+              *errorcodeptr = ERR59;
+              goto FAILED;
+              }
+            cd->had_accept = TRUE;
+            for (oc = cd->open_caps; oc != NULL; oc = oc->next)
+              {
+              *code++ = OP_CLOSE;
+              PUT2INC(code, 0, oc->number);
+              }
+            setverb = *code++ =
+              (cd->assert_depth > 0)? OP_ASSERT_ACCEPT : OP_ACCEPT;
+
+            /* Do not set firstchar after *ACCEPT */
+            if (firstcharflags == REQ_UNSET) firstcharflags = REQ_NONE;
+            }
+
+          /* Handle other cases with/without an argument */
+
+          else if (arglen == 0)
+            {
+            if (verbs[i].op < 0)   /* Argument is mandatory */
+              {
+              *errorcodeptr = ERR66;
+              goto FAILED;
+              }
+            setverb = *code++ = verbs[i].op;
+            }
+
+          else
+            {
+            if (verbs[i].op_arg < 0)   /* Argument is forbidden */
+              {
+              *errorcodeptr = ERR59;
+              goto FAILED;
+              }
+            setverb = *code++ = verbs[i].op_arg;
+            *code++ = arglen;
+            memcpy(code, arg, IN_UCHARS(arglen));
+            code += arglen;
+            *code++ = 0;
+            }
+
+          switch (setverb)
+            {
+            case OP_THEN:
+            case OP_THEN_ARG:
+            cd->external_flags |= PCRE_HASTHEN;
+            break;
+
+            case OP_PRUNE:
+            case OP_PRUNE_ARG:
+            case OP_SKIP:
+            case OP_SKIP_ARG:
+            cd->had_pruneorskip = TRUE;
+            break;
+            }
+
+          break;  /* Found verb, exit loop */
+          }
+
+        vn += verbs[i].len + 1;
+        }
+
+      if (i < verbcount) continue;    /* Successfully handled a verb */
+      *errorcodeptr = ERR60;          /* Verb not recognized */
+      goto FAILED;
+      }
+
+    /* Deal with the extended parentheses; all are introduced by '?', and the
+    appearance of any of them means that this is not a capturing group. */
+
+    else if (*ptr == CHAR_QUESTION_MARK)
+      {
+      int i, set, unset, namelen;
+      int *optset;
+      const pcre_uchar *name;
+      pcre_uchar *slot;
+
+      switch (*(++ptr))
+        {
+        case CHAR_NUMBER_SIGN:                 /* Comment; skip to ket */
+        ptr++;
+        while (*ptr != CHAR_NULL && *ptr != CHAR_RIGHT_PARENTHESIS) ptr++;
+        if (*ptr == CHAR_NULL)
+          {
+          *errorcodeptr = ERR18;
+          goto FAILED;
+          }
+        continue;
+
+
+        /* ------------------------------------------------------------ */
+        case CHAR_VERTICAL_LINE:  /* Reset capture count for each branch */
+        reset_bracount = TRUE;
+        /* Fall through */
+
+        /* ------------------------------------------------------------ */
+        case CHAR_COLON:          /* Non-capturing bracket */
+        bravalue = OP_BRA;
+        ptr++;
+        break;
+
+
+        /* ------------------------------------------------------------ */
+        case CHAR_LEFT_PARENTHESIS:
+        bravalue = OP_COND;       /* Conditional group */
+        tempptr = ptr;
+
+        /* A condition can be an assertion, a number (referring to a numbered
+        group), a name (referring to a named group), or 'R', referring to
+        recursion. R and R&name are also permitted for recursion tests.
+
+        There are several syntaxes for testing a named group: (?(name)) is used
+        by Python; Perl 5.10 onwards uses (?() or (?('name')).
+
+        There are two unfortunate ambiguities, caused by history. (a) 'R' can
+        be the recursive thing or the name 'R' (and similarly for 'R' followed
+        by digits), and (b) a number could be a name that consists of digits.
+        In both cases, we look for a name first; if not found, we try the other
+        cases.
+
+        For compatibility with auto-callouts, we allow a callout to be
+        specified before a condition that is an assertion. First, check for the
+        syntax of a callout; if found, adjust the temporary pointer that is
+        used to check for an assertion condition. That's all that is needed! */
+
+        if (ptr[1] == CHAR_QUESTION_MARK && ptr[2] == CHAR_C)
+          {
+          for (i = 3;; i++) if (!IS_DIGIT(ptr[i])) break;
+          if (ptr[i] == CHAR_RIGHT_PARENTHESIS)
+            tempptr += i + 1;
+          }
+
+        /* For conditions that are assertions, check the syntax, and then exit
+        the switch. This will take control down to where bracketed groups,
+        including assertions, are processed. */
+
+        if (tempptr[1] == CHAR_QUESTION_MARK &&
+              (tempptr[2] == CHAR_EQUALS_SIGN ||
+               tempptr[2] == CHAR_EXCLAMATION_MARK ||
+               tempptr[2] == CHAR_LESS_THAN_SIGN))
+          break;
+
+        /* Most other conditions use OP_CREF (a couple change to OP_RREF
+        below), and all need to skip 1+IMM2_SIZE bytes at the start of the group. */
+
+        code[1+LINK_SIZE] = OP_CREF;
+        skipbytes = 1+IMM2_SIZE;
+        refsign = -1;
+
+        /* Check for a test for recursion in a named group. */
+
+        if (ptr[1] == CHAR_R && ptr[2] == CHAR_AMPERSAND)
+          {
+          terminator = -1;
+          ptr += 2;
+          code[1+LINK_SIZE] = OP_RREF;    /* Change the type of test */
+          }
+
+        /* Check for a test for a named group's having been set, using the Perl
+        syntax (?() or (?('name') */
+
+        else if (ptr[1] == CHAR_LESS_THAN_SIGN)
+          {
+          terminator = CHAR_GREATER_THAN_SIGN;
+          ptr++;
+          }
+        else if (ptr[1] == CHAR_APOSTROPHE)
+          {
+          terminator = CHAR_APOSTROPHE;
+          ptr++;
+          }
+        else
+          {
+          terminator = CHAR_NULL;
+          if (ptr[1] == CHAR_MINUS || ptr[1] == CHAR_PLUS) refsign = *(++ptr);
+          }
+
+        /* We now expect to read a name; any thing else is an error */
+
+        if (!MAX_255(ptr[1]) || (cd->ctypes[ptr[1]] & ctype_word) == 0)
+          {
+          ptr += 1;  /* To get the right offset */
+          *errorcodeptr = ERR28;
+          goto FAILED;
+          }
+
+        /* Read the name, but also get it as a number if it's all digits */
+
+        recno = 0;
+        name = ++ptr;
+        while (MAX_255(*ptr) && (cd->ctypes[*ptr] & ctype_word) != 0)
+          {
+          if (recno >= 0)
+            recno = (IS_DIGIT(*ptr))? recno * 10 + (int)(*ptr - CHAR_0) : -1;
+          ptr++;
+          }
+        namelen = (int)(ptr - name);
+
+        if ((terminator > 0 && *ptr++ != (pcre_uchar)terminator) ||
+            *ptr++ != CHAR_RIGHT_PARENTHESIS)
+          {
+          ptr--;      /* Error offset */
+          *errorcodeptr = ERR26;
+          goto FAILED;
+          }
+
+        /* Do no further checking in the pre-compile phase. */
+
+        if (lengthptr != NULL) break;
+
+        /* In the real compile we do the work of looking for the actual
+        reference. If the string started with "+" or "-" we require the rest to
+        be digits, in which case recno will be set. */
+
+        if (refsign > 0)
+          {
+          if (recno <= 0)
+            {
+            *errorcodeptr = ERR58;
+            goto FAILED;
+            }
+          recno = (refsign == CHAR_MINUS)?
+            cd->bracount - recno + 1 : recno +cd->bracount;
+          if (recno <= 0 || recno > cd->final_bracount)
+            {
+            *errorcodeptr = ERR15;
+            goto FAILED;
+            }
+          PUT2(code, 2+LINK_SIZE, recno);
+          break;
+          }
+
+        /* Otherwise (did not start with "+" or "-"), start by looking for the
+        name. If we find a name, add one to the opcode to change OP_CREF or
+        OP_RREF into OP_NCREF or OP_NRREF. These behave exactly the same,
+        except they record that the reference was originally to a name. The
+        information is used to check duplicate names. */
+
+        slot = cd->name_table;
+        for (i = 0; i < cd->names_found; i++)
+          {
+          if (STRNCMP_UC_UC(name, slot+IMM2_SIZE, namelen) == 0) break;
+          slot += cd->name_entry_size;
+          }
+
+        /* Found a previous named subpattern */
+
+        if (i < cd->names_found)
+          {
+          recno = GET2(slot, 0);
+          PUT2(code, 2+LINK_SIZE, recno);
+          code[1+LINK_SIZE]++;
+          }
+
+        /* Search the pattern for a forward reference */
+
+        else if ((i = find_parens(cd, name, namelen,
+                        (options & PCRE_EXTENDED) != 0, utf)) > 0)
+          {
+          PUT2(code, 2+LINK_SIZE, i);
+          code[1+LINK_SIZE]++;
+          }
+
+        /* If terminator == CHAR_NULL it means that the name followed directly
+        after the opening parenthesis [e.g. (?(abc)...] and in this case there
+        are some further alternatives to try. For the cases where terminator !=
+        0 [things like (?(... or (?('name')... or (?(R&name)... ] we have
+        now checked all the possibilities, so give an error. */
+
+        else if (terminator != CHAR_NULL)
+          {
+          *errorcodeptr = ERR15;
+          goto FAILED;
+          }
+
+        /* Check for (?(R) for recursion. Allow digits after R to specify a
+        specific group number. */
+
+        else if (*name == CHAR_R)
+          {
+          recno = 0;
+          for (i = 1; i < namelen; i++)
+            {
+            if (!IS_DIGIT(name[i]))
+              {
+              *errorcodeptr = ERR15;
+              goto FAILED;
+              }
+            recno = recno * 10 + name[i] - CHAR_0;
+            }
+          if (recno == 0) recno = RREF_ANY;
+          code[1+LINK_SIZE] = OP_RREF;      /* Change test type */
+          PUT2(code, 2+LINK_SIZE, recno);
+          }
+
+        /* Similarly, check for the (?(DEFINE) "condition", which is always
+        false. */
+
+        else if (namelen == 6 && STRNCMP_UC_C8(name, STRING_DEFINE, 6) == 0)
+          {
+          code[1+LINK_SIZE] = OP_DEF;
+          skipbytes = 1;
+          }
+
+        /* Check for the "name" actually being a subpattern number. We are
+        in the second pass here, so final_bracount is set. */
+
+        else if (recno > 0 && recno <= cd->final_bracount)
+          {
+          PUT2(code, 2+LINK_SIZE, recno);
+          }
+
+        /* Either an unidentified subpattern, or a reference to (?(0) */
+
+        else
+          {
+          *errorcodeptr = (recno == 0)? ERR35: ERR15;
+          goto FAILED;
+          }
+        break;
+
+
+        /* ------------------------------------------------------------ */
+        case CHAR_EQUALS_SIGN:                 /* Positive lookahead */
+        bravalue = OP_ASSERT;
+        cd->assert_depth += 1;
+        ptr++;
+        break;
+
+
+        /* ------------------------------------------------------------ */
+        case CHAR_EXCLAMATION_MARK:            /* Negative lookahead */
+        ptr++;
+        if (*ptr == CHAR_RIGHT_PARENTHESIS)    /* Optimize (?!) */
+          {
+          *code++ = OP_FAIL;
+          previous = NULL;
+          continue;
+          }
+        bravalue = OP_ASSERT_NOT;
+        cd->assert_depth += 1;
+        break;
+
+
+        /* ------------------------------------------------------------ */
+        case CHAR_LESS_THAN_SIGN:              /* Lookbehind or named define */
+        switch (ptr[1])
+          {
+          case CHAR_EQUALS_SIGN:               /* Positive lookbehind */
+          bravalue = OP_ASSERTBACK;
+          cd->assert_depth += 1;
+          ptr += 2;
+          break;
+
+          case CHAR_EXCLAMATION_MARK:          /* Negative lookbehind */
+          bravalue = OP_ASSERTBACK_NOT;
+          cd->assert_depth += 1;
+          ptr += 2;
+          break;
+
+          default:                /* Could be name define, else bad */
+          if (MAX_255(ptr[1]) && (cd->ctypes[ptr[1]] & ctype_word) != 0)
+            goto DEFINE_NAME;
+          ptr++;                  /* Correct offset for error */
+          *errorcodeptr = ERR24;
+          goto FAILED;
+          }
+        break;
+
+
+        /* ------------------------------------------------------------ */
+        case CHAR_GREATER_THAN_SIGN:           /* One-time brackets */
+        bravalue = OP_ONCE;
+        ptr++;
+        break;
+
+
+        /* ------------------------------------------------------------ */
+        case CHAR_C:                 /* Callout - may be followed by digits; */
+        previous_callout = code;     /* Save for later completion */
+        after_manual_callout = 1;    /* Skip one item before completing */
+        *code++ = OP_CALLOUT;
+          {
+          int n = 0;
+          ptr++;
+          while(IS_DIGIT(*ptr))
+            n = n * 10 + *ptr++ - CHAR_0;
+          if (*ptr != CHAR_RIGHT_PARENTHESIS)
+            {
+            *errorcodeptr = ERR39;
+            goto FAILED;
+            }
+          if (n > 255)
+            {
+            *errorcodeptr = ERR38;
+            goto FAILED;
+            }
+          *code++ = n;
+          PUT(code, 0, (int)(ptr - cd->start_pattern + 1)); /* Pattern offset */
+          PUT(code, LINK_SIZE, 0);                          /* Default length */
+          code += 2 * LINK_SIZE;
+          }
+        previous = NULL;
+        continue;
+
+
+        /* ------------------------------------------------------------ */
+        case CHAR_P:              /* Python-style named subpattern handling */
+        if (*(++ptr) == CHAR_EQUALS_SIGN ||
+            *ptr == CHAR_GREATER_THAN_SIGN)  /* Reference or recursion */
+          {
+          is_recurse = *ptr == CHAR_GREATER_THAN_SIGN;
+          terminator = CHAR_RIGHT_PARENTHESIS;
+          goto NAMED_REF_OR_RECURSE;
+          }
+        else if (*ptr != CHAR_LESS_THAN_SIGN)  /* Test for Python-style defn */
+          {
+          *errorcodeptr = ERR41;
+          goto FAILED;
+          }
+        /* Fall through to handle (?P< as (?< is handled */
+
+
+        /* ------------------------------------------------------------ */
+        DEFINE_NAME:    /* Come here from (?< handling */
+        case CHAR_APOSTROPHE:
+          {
+          terminator = (*ptr == CHAR_LESS_THAN_SIGN)?
+            CHAR_GREATER_THAN_SIGN : CHAR_APOSTROPHE;
+          name = ++ptr;
+
+          while (MAX_255(*ptr) && (cd->ctypes[*ptr] & ctype_word) != 0) ptr++;
+          namelen = (int)(ptr - name);
+
+          /* In the pre-compile phase, just do a syntax check. */
+
+          if (lengthptr != NULL)
+            {
+            if (*ptr != (pcre_uchar)terminator)
+              {
+              *errorcodeptr = ERR42;
+              goto FAILED;
+              }
+            if (cd->names_found >= MAX_NAME_COUNT)
+              {
+              *errorcodeptr = ERR49;
+              goto FAILED;
+              }
+            if (namelen + IMM2_SIZE + 1 > cd->name_entry_size)
+              {
+              cd->name_entry_size = namelen + IMM2_SIZE + 1;
+              if (namelen > MAX_NAME_SIZE)
+                {
+                *errorcodeptr = ERR48;
+                goto FAILED;
+                }
+              }
+            }
+
+          /* In the real compile, create the entry in the table, maintaining
+          alphabetical order. Duplicate names for different numbers are
+          permitted only if PCRE_DUPNAMES is set. Duplicate names for the same
+          number are always OK. (An existing number can be re-used if (?|
+          appears in the pattern.) In either event, a duplicate name results in
+          a duplicate entry in the table, even if the number is the same. This
+          is because the number of names, and hence the table size, is computed
+          in the pre-compile, and it affects various numbers and pointers which
+          would all have to be modified, and the compiled code moved down, if
+          duplicates with the same number were omitted from the table. This
+          doesn't seem worth the hassle. However, *different* names for the
+          same number are not permitted. */
+
+          else
+            {
+            BOOL dupname = FALSE;
+            slot = cd->name_table;
+
+            for (i = 0; i < cd->names_found; i++)
+              {
+              int crc = memcmp(name, slot+IMM2_SIZE, IN_UCHARS(namelen));
+              if (crc == 0)
+                {
+                if (slot[IMM2_SIZE+namelen] == 0)
+                  {
+                  if (GET2(slot, 0) != cd->bracount + 1 &&
+                      (options & PCRE_DUPNAMES) == 0)
+                    {
+                    *errorcodeptr = ERR43;
+                    goto FAILED;
+                    }
+                  else dupname = TRUE;
+                  }
+                else crc = -1;      /* Current name is a substring */
+                }
+
+              /* Make space in the table and break the loop for an earlier
+              name. For a duplicate or later name, carry on. We do this for
+              duplicates so that in the simple case (when ?(| is not used) they
+              are in order of their numbers. */
+
+              if (crc < 0)
+                {
+                memmove(slot + cd->name_entry_size, slot,
+                  IN_UCHARS((cd->names_found - i) * cd->name_entry_size));
+                break;
+                }
+
+              /* Continue the loop for a later or duplicate name */
+
+              slot += cd->name_entry_size;
+              }
+
+            /* For non-duplicate names, check for a duplicate number before
+            adding the new name. */
+
+            if (!dupname)
+              {
+              pcre_uchar *cslot = cd->name_table;
+              for (i = 0; i < cd->names_found; i++)
+                {
+                if (cslot != slot)
+                  {
+                  if (GET2(cslot, 0) == cd->bracount + 1)
+                    {
+                    *errorcodeptr = ERR65;
+                    goto FAILED;
+                    }
+                  }
+                else i--;
+                cslot += cd->name_entry_size;
+                }
+              }
+
+            PUT2(slot, 0, cd->bracount + 1);
+            memcpy(slot + IMM2_SIZE, name, IN_UCHARS(namelen));
+            slot[IMM2_SIZE + namelen] = 0;
+            }
+          }
+
+        /* In both pre-compile and compile, count the number of names we've
+        encountered. */
+
+        cd->names_found++;
+        ptr++;                    /* Move past > or ' */
+        goto NUMBERED_GROUP;
+
+
+        /* ------------------------------------------------------------ */
+        case CHAR_AMPERSAND:            /* Perl recursion/subroutine syntax */
+        terminator = CHAR_RIGHT_PARENTHESIS;
+        is_recurse = TRUE;
+        /* Fall through */
+
+        /* We come here from the Python syntax above that handles both
+        references (?P=name) and recursion (?P>name), as well as falling
+        through from the Perl recursion syntax (?&name). We also come here from
+        the Perl \k or \k'name' back reference syntax and the \k{name}
+        .NET syntax, and the Oniguruma \g<...> and \g'...' subroutine syntax. */
+
+        NAMED_REF_OR_RECURSE:
+        name = ++ptr;
+        while (MAX_255(*ptr) && (cd->ctypes[*ptr] & ctype_word) != 0) ptr++;
+        namelen = (int)(ptr - name);
+
+        /* In the pre-compile phase, do a syntax check. We used to just set
+        a dummy reference number, because it was not used in the first pass.
+        However, with the change of recursive back references to be atomic,
+        we have to look for the number so that this state can be identified, as
+        otherwise the incorrect length is computed. If it's not a backwards
+        reference, the dummy number will do. */
+
+        if (lengthptr != NULL)
+          {
+          const pcre_uchar *temp;
+
+          if (namelen == 0)
+            {
+            *errorcodeptr = ERR62;
+            goto FAILED;
+            }
+          if (*ptr != (pcre_uchar)terminator)
+            {
+            *errorcodeptr = ERR42;
+            goto FAILED;
+            }
+          if (namelen > MAX_NAME_SIZE)
+            {
+            *errorcodeptr = ERR48;
+            goto FAILED;
+            }
+
+          /* The name table does not exist in the first pass, so we cannot
+          do a simple search as in the code below. Instead, we have to scan the
+          pattern to find the number. It is important that we scan it only as
+          far as we have got because the syntax of named subpatterns has not
+          been checked for the rest of the pattern, and find_parens() assumes
+          correct syntax. In any case, it's a waste of resources to scan
+          further. We stop the scan at the current point by temporarily
+          adjusting the value of cd->endpattern. */
+
+          temp = cd->end_pattern;
+          cd->end_pattern = ptr;
+          recno = find_parens(cd, name, namelen,
+            (options & PCRE_EXTENDED) != 0, utf);
+          cd->end_pattern = temp;
+          if (recno < 0) recno = 0;    /* Forward ref; set dummy number */
+          }
+
+        /* In the real compile, seek the name in the table. We check the name
+        first, and then check that we have reached the end of the name in the
+        table. That way, if the name that is longer than any in the table,
+        the comparison will fail without reading beyond the table entry. */
+
+        else
+          {
+          slot = cd->name_table;
+          for (i = 0; i < cd->names_found; i++)
+            {
+            if (STRNCMP_UC_UC(name, slot+IMM2_SIZE, namelen) == 0 &&
+                slot[IMM2_SIZE+namelen] == 0)
+              break;
+            slot += cd->name_entry_size;
+            }
+
+          if (i < cd->names_found)         /* Back reference */
+            {
+            recno = GET2(slot, 0);
+            }
+          else if ((recno =                /* Forward back reference */
+                    find_parens(cd, name, namelen,
+                      (options & PCRE_EXTENDED) != 0, utf)) <= 0)
+            {
+            *errorcodeptr = ERR15;
+            goto FAILED;
+            }
+          }
+
+        /* In both phases, we can now go to the code than handles numerical
+        recursion or backreferences. */
+
+        if (is_recurse) goto HANDLE_RECURSION;
+          else goto HANDLE_REFERENCE;
+
+
+        /* ------------------------------------------------------------ */
+        case CHAR_R:              /* Recursion */
+        ptr++;                    /* Same as (?0)      */
+        /* Fall through */
+
+
+        /* ------------------------------------------------------------ */
+        case CHAR_MINUS: case CHAR_PLUS:  /* Recursion or subroutine */
+        case CHAR_0: case CHAR_1: case CHAR_2: case CHAR_3: case CHAR_4:
+        case CHAR_5: case CHAR_6: case CHAR_7: case CHAR_8: case CHAR_9:
+          {
+          const pcre_uchar *called;
+          terminator = CHAR_RIGHT_PARENTHESIS;
+
+          /* Come here from the \g<...> and \g'...' code (Oniguruma
+          compatibility). However, the syntax has been checked to ensure that
+          the ... are a (signed) number, so that neither ERR63 nor ERR29 will
+          be called on this path, nor with the jump to OTHER_CHAR_AFTER_QUERY
+          ever be taken. */
+
+          HANDLE_NUMERICAL_RECURSION:
+
+          if ((refsign = *ptr) == CHAR_PLUS)
+            {
+            ptr++;
+            if (!IS_DIGIT(*ptr))
+              {
+              *errorcodeptr = ERR63;
+              goto FAILED;
+              }
+            }
+          else if (refsign == CHAR_MINUS)
+            {
+            if (!IS_DIGIT(ptr[1]))
+              goto OTHER_CHAR_AFTER_QUERY;
+            ptr++;
+            }
+
+          recno = 0;
+          while(IS_DIGIT(*ptr))
+            recno = recno * 10 + *ptr++ - CHAR_0;
+
+          if (*ptr != (pcre_uchar)terminator)
+            {
+            *errorcodeptr = ERR29;
+            goto FAILED;
+            }
+
+          if (refsign == CHAR_MINUS)
+            {
+            if (recno == 0)
+              {
+              *errorcodeptr = ERR58;
+              goto FAILED;
+              }
+            recno = cd->bracount - recno + 1;
+            if (recno <= 0)
+              {
+              *errorcodeptr = ERR15;
+              goto FAILED;
+              }
+            }
+          else if (refsign == CHAR_PLUS)
+            {
+            if (recno == 0)
+              {
+              *errorcodeptr = ERR58;
+              goto FAILED;
+              }
+            recno += cd->bracount;
+            }
+
+          /* Come here from code above that handles a named recursion */
+
+          HANDLE_RECURSION:
+
+          previous = code;
+          called = cd->start_code;
+
+          /* When we are actually compiling, find the bracket that is being
+          referenced. Temporarily end the regex in case it doesn't exist before
+          this point. If we end up with a forward reference, first check that
+          the bracket does occur later so we can give the error (and position)
+          now. Then remember this forward reference in the workspace so it can
+          be filled in at the end. */
+
+          if (lengthptr == NULL)
+            {
+            *code = OP_END;
+            if (recno != 0)
+              called = PRIV(find_bracket)(cd->start_code, utf, recno);
+
+            /* Forward reference */
+
+            if (called == NULL)
+              {
+              if (find_parens(cd, NULL, recno,
+                    (options & PCRE_EXTENDED) != 0, utf) < 0)
+                {
+                *errorcodeptr = ERR15;
+                goto FAILED;
+                }
+
+              /* Fudge the value of "called" so that when it is inserted as an
+              offset below, what it actually inserted is the reference number
+              of the group. Then remember the forward reference. */
+
+              called = cd->start_code + recno;
+              if (cd->hwm >= cd->start_workspace + cd->workspace_size -
+                  WORK_SIZE_SAFETY_MARGIN)
+                {
+                *errorcodeptr = expand_workspace(cd);
+                if (*errorcodeptr != 0) goto FAILED;
+                }
+              PUTINC(cd->hwm, 0, (int)(code + 1 - cd->start_code));
+              }
+
+            /* If not a forward reference, and the subpattern is still open,
+            this is a recursive call. We check to see if this is a left
+            recursion that could loop for ever, and diagnose that case. We
+            must not, however, do this check if we are in a conditional
+            subpattern because the condition might be testing for recursion in
+            a pattern such as /(?(R)a+|(?R)b)/, which is perfectly valid.
+            Forever loops are also detected at runtime, so those that occur in
+            conditional subpatterns will be picked up then. */
+
+            else if (GET(called, 1) == 0 && cond_depth <= 0 &&
+                     could_be_empty(called, code, bcptr, utf, cd))
+              {
+              *errorcodeptr = ERR40;
+              goto FAILED;
+              }
+            }
+
+          /* Insert the recursion/subroutine item. It does not have a set first
+          character (relevant if it is repeated, because it will then be
+          wrapped with ONCE brackets). */
+
+          *code = OP_RECURSE;
+          PUT(code, 1, (int)(called - cd->start_code));
+          code += 1 + LINK_SIZE;
+          groupsetfirstchar = FALSE;
+          }
+
+        /* Can't determine a first byte now */
+
+        if (firstcharflags == REQ_UNSET) firstcharflags = REQ_NONE;
+        continue;
+
+
+        /* ------------------------------------------------------------ */
+        default:              /* Other characters: check option setting */
+        OTHER_CHAR_AFTER_QUERY:
+        set = unset = 0;
+        optset = &set;
+
+        while (*ptr != CHAR_RIGHT_PARENTHESIS && *ptr != CHAR_COLON)
+          {
+          switch (*ptr++)
+            {
+            case CHAR_MINUS: optset = &unset; break;
+
+            case CHAR_J:    /* Record that it changed in the external options */
+            *optset |= PCRE_DUPNAMES;
+            cd->external_flags |= PCRE_JCHANGED;
+            break;
+
+            case CHAR_i: *optset |= PCRE_CASELESS; break;
+            case CHAR_m: *optset |= PCRE_MULTILINE; break;
+            case CHAR_s: *optset |= PCRE_DOTALL; break;
+            case CHAR_x: *optset |= PCRE_EXTENDED; break;
+            case CHAR_U: *optset |= PCRE_UNGREEDY; break;
+            case CHAR_X: *optset |= PCRE_EXTRA; break;
+
+            default:  *errorcodeptr = ERR12;
+                      ptr--;    /* Correct the offset */
+                      goto FAILED;
+            }
+          }
+
+        /* Set up the changed option bits, but don't change anything yet. */
+
+        newoptions = (options | set) & (~unset);
+
+        /* If the options ended with ')' this is not the start of a nested
+        group with option changes, so the options change at this level. If this
+        item is right at the start of the pattern, the options can be
+        abstracted and made external in the pre-compile phase, and ignored in
+        the compile phase. This can be helpful when matching -- for instance in
+        caseless checking of required bytes.
+
+        If the code pointer is not (cd->start_code + 1 + LINK_SIZE), we are
+        definitely *not* at the start of the pattern because something has been
+        compiled. In the pre-compile phase, however, the code pointer can have
+        that value after the start, because it gets reset as code is discarded
+        during the pre-compile. However, this can happen only at top level - if
+        we are within parentheses, the starting BRA will still be present. At
+        any parenthesis level, the length value can be used to test if anything
+        has been compiled at that level. Thus, a test for both these conditions
+        is necessary to ensure we correctly detect the start of the pattern in
+        both phases.
+
+        If we are not at the pattern start, reset the greedy defaults and the
+        case value for firstchar and reqchar. */
+
+        if (*ptr == CHAR_RIGHT_PARENTHESIS)
+          {
+          if (code == cd->start_code + 1 + LINK_SIZE &&
+               (lengthptr == NULL || *lengthptr == 2 + 2*LINK_SIZE))
+            {
+            cd->external_options = newoptions;
+            }
+          else
+            {
+            greedy_default = ((newoptions & PCRE_UNGREEDY) != 0);
+            greedy_non_default = greedy_default ^ 1;
+            req_caseopt = ((newoptions & PCRE_CASELESS) != 0)? REQ_CASELESS:0;
+            }
+
+          /* Change options at this level, and pass them back for use
+          in subsequent branches. */
+
+          *optionsptr = options = newoptions;
+          previous = NULL;       /* This item can't be repeated */
+          continue;              /* It is complete */
+          }
+
+        /* If the options ended with ':' we are heading into a nested group
+        with possible change of options. Such groups are non-capturing and are
+        not assertions of any kind. All we need to do is skip over the ':';
+        the newoptions value is handled below. */
+
+        bravalue = OP_BRA;
+        ptr++;
+        }     /* End of switch for character following (? */
+      }       /* End of (? handling */
+
+    /* Opening parenthesis not followed by '*' or '?'. If PCRE_NO_AUTO_CAPTURE
+    is set, all unadorned brackets become non-capturing and behave like (?:...)
+    brackets. */
+
+    else if ((options & PCRE_NO_AUTO_CAPTURE) != 0)
+      {
+      bravalue = OP_BRA;
+      }
+
+    /* Else we have a capturing group. */
+
+    else
+      {
+      NUMBERED_GROUP:
+      cd->bracount += 1;
+      PUT2(code, 1+LINK_SIZE, cd->bracount);
+      skipbytes = IMM2_SIZE;
+      }
+
+    /* Process nested bracketed regex. Assertions used not to be repeatable,
+    but this was changed for Perl compatibility, so all kinds can now be
+    repeated. We copy code into a non-register variable (tempcode) in order to
+    be able to pass its address because some compilers complain otherwise. */
+
+    previous = code;                      /* For handling repetition */
+    *code = bravalue;
+    tempcode = code;
+    tempreqvary = cd->req_varyopt;        /* Save value before bracket */
+    tempbracount = cd->bracount;          /* Save value before bracket */
+    length_prevgroup = 0;                 /* Initialize for pre-compile phase */
+
+    if (!compile_regex(
+         newoptions,                      /* The complete new option state */
+         &tempcode,                       /* Where to put code (updated) */
+         &ptr,                            /* Input pointer (updated) */
+         errorcodeptr,                    /* Where to put an error message */
+         (bravalue == OP_ASSERTBACK ||
+          bravalue == OP_ASSERTBACK_NOT), /* TRUE if back assert */
+         reset_bracount,                  /* True if (?| group */
+         skipbytes,                       /* Skip over bracket number */
+         cond_depth +
+           ((bravalue == OP_COND)?1:0),   /* Depth of condition subpatterns */
+         &subfirstchar,                   /* For possible first char */
+         &subfirstcharflags,
+         &subreqchar,                     /* For possible last char */
+         &subreqcharflags,
+         bcptr,                           /* Current branch chain */
+         cd,                              /* Tables block */
+         (lengthptr == NULL)? NULL :      /* Actual compile phase */
+           &length_prevgroup              /* Pre-compile phase */
+         ))
+      goto FAILED;
+
+    /* If this was an atomic group and there are no capturing groups within it,
+    generate OP_ONCE_NC instead of OP_ONCE. */
+
+    if (bravalue == OP_ONCE && cd->bracount <= tempbracount)
+      *code = OP_ONCE_NC;
+
+    if (bravalue >= OP_ASSERT && bravalue <= OP_ASSERTBACK_NOT)
+      cd->assert_depth -= 1;
+
+    /* At the end of compiling, code is still pointing to the start of the
+    group, while tempcode has been updated to point past the end of the group.
+    The pattern pointer (ptr) is on the bracket.
+
+    If this is a conditional bracket, check that there are no more than
+    two branches in the group, or just one if it's a DEFINE group. We do this
+    in the real compile phase, not in the pre-pass, where the whole group may
+    not be available. */
+
+    if (bravalue == OP_COND && lengthptr == NULL)
+      {
+      pcre_uchar *tc = code;
+      int condcount = 0;
+
+      do {
+         condcount++;
+         tc += GET(tc,1);
+         }
+      while (*tc != OP_KET);
+
+      /* A DEFINE group is never obeyed inline (the "condition" is always
+      false). It must have only one branch. */
+
+      if (code[LINK_SIZE+1] == OP_DEF)
+        {
+        if (condcount > 1)
+          {
+          *errorcodeptr = ERR54;
+          goto FAILED;
+          }
+        bravalue = OP_DEF;   /* Just a flag to suppress char handling below */
+        }
+
+      /* A "normal" conditional group. If there is just one branch, we must not
+      make use of its firstchar or reqchar, because this is equivalent to an
+      empty second branch. */
+
+      else
+        {
+        if (condcount > 2)
+          {
+          *errorcodeptr = ERR27;
+          goto FAILED;
+          }
+        if (condcount == 1) subfirstcharflags = subreqcharflags = REQ_NONE;
+        }
+      }
+
+    /* Error if hit end of pattern */
+
+    if (*ptr != CHAR_RIGHT_PARENTHESIS)
+      {
+      *errorcodeptr = ERR14;
+      goto FAILED;
+      }
+
+    /* In the pre-compile phase, update the length by the length of the group,
+    less the brackets at either end. Then reduce the compiled code to just a
+    set of non-capturing brackets so that it doesn't use much memory if it is
+    duplicated by a quantifier.*/
+
+    if (lengthptr != NULL)
+      {
+      if (OFLOW_MAX - *lengthptr < length_prevgroup - 2 - 2*LINK_SIZE)
+        {
+        *errorcodeptr = ERR20;
+        goto FAILED;
+        }
+      *lengthptr += length_prevgroup - 2 - 2*LINK_SIZE;
+      code++;   /* This already contains bravalue */
+      PUTINC(code, 0, 1 + LINK_SIZE);
+      *code++ = OP_KET;
+      PUTINC(code, 0, 1 + LINK_SIZE);
+      break;    /* No need to waste time with special character handling */
+      }
+
+    /* Otherwise update the main code pointer to the end of the group. */
+
+    code = tempcode;
+
+    /* For a DEFINE group, required and first character settings are not
+    relevant. */
+
+    if (bravalue == OP_DEF) break;
+
+    /* Handle updating of the required and first characters for other types of
+    group. Update for normal brackets of all kinds, and conditions with two
+    branches (see code above). If the bracket is followed by a quantifier with
+    zero repeat, we have to back off. Hence the definition of zeroreqchar and
+    zerofirstchar outside the main loop so that they can be accessed for the
+    back off. */
+
+    zeroreqchar = reqchar;
+    zeroreqcharflags = reqcharflags;
+    zerofirstchar = firstchar;
+    zerofirstcharflags = firstcharflags;
+    groupsetfirstchar = FALSE;
+
+    if (bravalue >= OP_ONCE)
+      {
+      /* If we have not yet set a firstchar in this branch, take it from the
+      subpattern, remembering that it was set here so that a repeat of more
+      than one can replicate it as reqchar if necessary. If the subpattern has
+      no firstchar, set "none" for the whole branch. In both cases, a zero
+      repeat forces firstchar to "none". */
+
+      if (firstcharflags == REQ_UNSET)
+        {
+        if (subfirstcharflags >= 0)
+          {
+          firstchar = subfirstchar;
+          firstcharflags = subfirstcharflags;
+          groupsetfirstchar = TRUE;
+          }
+        else firstcharflags = REQ_NONE;
+        zerofirstcharflags = REQ_NONE;
+        }
+
+      /* If firstchar was previously set, convert the subpattern's firstchar
+      into reqchar if there wasn't one, using the vary flag that was in
+      existence beforehand. */
+
+      else if (subfirstcharflags >= 0 && subreqcharflags < 0)
+        {
+        subreqchar = subfirstchar;
+        subreqcharflags = subfirstcharflags | tempreqvary;
+        }
+
+      /* If the subpattern set a required byte (or set a first byte that isn't
+      really the first byte - see above), set it. */
+
+      if (subreqcharflags >= 0)
+        {
+        reqchar = subreqchar;
+        reqcharflags = subreqcharflags;
+        }
+      }
+
+    /* For a forward assertion, we take the reqchar, if set. This can be
+    helpful if the pattern that follows the assertion doesn't set a different
+    char. For example, it's useful for /(?=abcde).+/. We can't set firstchar
+    for an assertion, however because it leads to incorrect effect for patterns
+    such as /(?=a)a.+/ when the "real" "a" would then become a reqchar instead
+    of a firstchar. This is overcome by a scan at the end if there's no
+    firstchar, looking for an asserted first char. */
+
+    else if (bravalue == OP_ASSERT && subreqcharflags >= 0)
+      {
+      reqchar = subreqchar;
+      reqcharflags = subreqcharflags;
+      }
+    break;     /* End of processing '(' */
+
+
+    /* ===================================================================*/
+    /* Handle metasequences introduced by \. For ones like \d, the ESC_ values
+    are arranged to be the negation of the corresponding OP_values in the
+    default case when PCRE_UCP is not set. For the back references, the values
+    are negative the reference number. Only back references and those types
+    that consume a character may be repeated. We can test for values between
+    ESC_b and ESC_Z for the latter; this may have to change if any new ones are
+    ever created. */
+
+    case CHAR_BACKSLASH:
+    tempptr = ptr;
+    escape = check_escape(&ptr, &ec, errorcodeptr, cd->bracount, options, FALSE);
+    if (*errorcodeptr != 0) goto FAILED;
+
+    if (escape == 0)                  /* The escape coded a single character */
+      c = ec;
+    else
+      {
+      if (escape == ESC_Q)            /* Handle start of quoted string */
+        {
+        if (ptr[1] == CHAR_BACKSLASH && ptr[2] == CHAR_E)
+          ptr += 2;               /* avoid empty string */
+            else inescq = TRUE;
+        continue;
+        }
+
+      if (escape == ESC_E) continue;  /* Perl ignores an orphan \E */
+
+      /* For metasequences that actually match a character, we disable the
+      setting of a first character if it hasn't already been set. */
+
+      if (firstcharflags == REQ_UNSET && escape > ESC_b && escape < ESC_Z)
+        firstcharflags = REQ_NONE;
+
+      /* Set values to reset to if this is followed by a zero repeat. */
+
+      zerofirstchar = firstchar;
+      zerofirstcharflags = firstcharflags;
+      zeroreqchar = reqchar;
+      zeroreqcharflags = reqcharflags;
+
+      /* \g or \g'name' is a subroutine call by name and \g or \g'n'
+      is a subroutine call by number (Oniguruma syntax). In fact, the value
+      ESC_g is returned only for these cases. So we don't need to check for <
+      or ' if the value is ESC_g. For the Perl syntax \g{n} the value is
+      -n, and for the Perl syntax \g{name} the result is ESC_k (as
+      that is a synonym for a named back reference). */
+
+      if (escape == ESC_g)
+        {
+        const pcre_uchar *p;
+        save_hwm = cd->hwm;   /* Normally this is set when '(' is read */
+        terminator = (*(++ptr) == CHAR_LESS_THAN_SIGN)?
+          CHAR_GREATER_THAN_SIGN : CHAR_APOSTROPHE;
+
+        /* These two statements stop the compiler for warning about possibly
+        unset variables caused by the jump to HANDLE_NUMERICAL_RECURSION. In
+        fact, because we actually check for a number below, the paths that
+        would actually be in error are never taken. */
+
+        skipbytes = 0;
+        reset_bracount = FALSE;
+
+        /* Test for a name */
+
+        if (ptr[1] != CHAR_PLUS && ptr[1] != CHAR_MINUS)
+          {
+          BOOL is_a_number = TRUE;
+          for (p = ptr + 1; *p != CHAR_NULL && *p != (pcre_uchar)terminator; p++)
+            {
+            if (!MAX_255(*p)) { is_a_number = FALSE; break; }
+            if ((cd->ctypes[*p] & ctype_digit) == 0) is_a_number = FALSE;
+            if ((cd->ctypes[*p] & ctype_word) == 0) break;
+            }
+          if (*p != (pcre_uchar)terminator)
+            {
+            *errorcodeptr = ERR57;
+            break;
+            }
+          if (is_a_number)
+            {
+            ptr++;
+            goto HANDLE_NUMERICAL_RECURSION;
+            }
+          is_recurse = TRUE;
+          goto NAMED_REF_OR_RECURSE;
+          }
+
+        /* Test a signed number in angle brackets or quotes. */
+
+        p = ptr + 2;
+        while (IS_DIGIT(*p)) p++;
+        if (*p != (pcre_uchar)terminator)
+          {
+          *errorcodeptr = ERR57;
+          break;
+          }
+        ptr++;
+        goto HANDLE_NUMERICAL_RECURSION;
+        }
+
+      /* \k or \k'name' is a back reference by name (Perl syntax).
+      We also support \k{name} (.NET syntax).  */
+
+      if (escape == ESC_k)
+        {
+        if ((ptr[1] != CHAR_LESS_THAN_SIGN &&
+          ptr[1] != CHAR_APOSTROPHE && ptr[1] != CHAR_LEFT_CURLY_BRACKET))
+          {
+          *errorcodeptr = ERR69;
+          break;
+          }
+        is_recurse = FALSE;
+        terminator = (*(++ptr) == CHAR_LESS_THAN_SIGN)?
+          CHAR_GREATER_THAN_SIGN : (*ptr == CHAR_APOSTROPHE)?
+          CHAR_APOSTROPHE : CHAR_RIGHT_CURLY_BRACKET;
+        goto NAMED_REF_OR_RECURSE;
+        }
+
+      /* Back references are handled specially; must disable firstchar if
+      not set to cope with cases like (?=(\w+))\1: which would otherwise set
+      ':' later. */
+
+      if (escape < 0)
+        {
+        open_capitem *oc;
+        recno = -escape;
+
+        HANDLE_REFERENCE:    /* Come here from named backref handling */
+        if (firstcharflags == REQ_UNSET) firstcharflags = REQ_NONE;
+        previous = code;
+        *code++ = ((options & PCRE_CASELESS) != 0)? OP_REFI : OP_REF;
+        PUT2INC(code, 0, recno);
+        cd->backref_map |= (recno < 32)? (1 << recno) : 1;
+        if (recno > cd->top_backref) cd->top_backref = recno;
+
+        /* Check to see if this back reference is recursive, that it, it
+        is inside the group that it references. A flag is set so that the
+        group can be made atomic. */
+
+        for (oc = cd->open_caps; oc != NULL; oc = oc->next)
+          {
+          if (oc->number == recno)
+            {
+            oc->flag = TRUE;
+            break;
+            }
+          }
+        }
+
+      /* So are Unicode property matches, if supported. */
+
+#ifdef SUPPORT_UCP
+      else if (escape == ESC_P || escape == ESC_p)
+        {
+        BOOL negated;
+        unsigned int ptype = 0, pdata = 0;
+        if (!get_ucp(&ptr, &negated, &ptype, &pdata, errorcodeptr))
+          goto FAILED;
+        previous = code;
+        *code++ = ((escape == ESC_p) != negated)? OP_PROP : OP_NOTPROP;
+        *code++ = ptype;
+        *code++ = pdata;
+        }
+#else
+
+      /* If Unicode properties are not supported, \X, \P, and \p are not
+      allowed. */
+
+      else if (escape == ESC_X || escape == ESC_P || escape == ESC_p)
+        {
+        *errorcodeptr = ERR45;
+        goto FAILED;
+        }
+#endif
+
+      /* For the rest (including \X when Unicode properties are supported), we
+      can obtain the OP value by negating the escape value in the default
+      situation when PCRE_UCP is not set. When it *is* set, we substitute
+      Unicode property tests. Note that \b and \B do a one-character
+      lookbehind, and \A also behaves as if it does. */
+
+      else
+        {
+        if ((escape == ESC_b || escape == ESC_B || escape == ESC_A) &&
+             cd->max_lookbehind == 0)
+          cd->max_lookbehind = 1;
+#ifdef SUPPORT_UCP
+        if (escape >= ESC_DU && escape <= ESC_wu)
+          {
+          nestptr = ptr + 1;                   /* Where to resume */
+          ptr = substitutes[escape - ESC_DU] - 1;  /* Just before substitute */
+          }
+        else
+#endif
+        /* In non-UTF-8 mode, we turn \C into OP_ALLANY instead of OP_ANYBYTE
+        so that it works in DFA mode and in lookbehinds. */
+
+          {
+          previous = (escape > ESC_b && escape < ESC_Z)? code : NULL;
+          *code++ = (!utf && escape == ESC_C)? OP_ALLANY : escape;
+          }
+        }
+      continue;
+      }
+
+    /* We have a data character whose value is in c. In UTF-8 mode it may have
+    a value > 127. We set its representation in the length/buffer, and then
+    handle it as a data character. */
+
+#if defined SUPPORT_UTF && !defined COMPILE_PCRE32
+    if (utf && c > MAX_VALUE_FOR_SINGLE_CHAR)
+      mclength = PRIV(ord2utf)(c, mcbuffer);
+    else
+#endif
+
+     {
+     mcbuffer[0] = c;
+     mclength = 1;
+     }
+    goto ONE_CHAR;
+
+
+    /* ===================================================================*/
+    /* Handle a literal character. It is guaranteed not to be whitespace or #
+    when the extended flag is set. If we are in UTF-8 mode, it may be a
+    multi-byte literal character. */
+
+    default:
+    NORMAL_CHAR:
+    mclength = 1;
+    mcbuffer[0] = c;
+
+#ifdef SUPPORT_UTF
+    if (utf && HAS_EXTRALEN(c))
+      ACROSSCHAR(TRUE, ptr[1], mcbuffer[mclength++] = *(++ptr));
+#endif
+
+    /* At this point we have the character's bytes in mcbuffer, and the length
+    in mclength. When not in UTF-8 mode, the length is always 1. */
+
+    ONE_CHAR:
+    previous = code;
+
+    /* For caseless UTF-8 mode when UCP support is available, check whether
+    this character has more than one other case. If so, generate a special
+    OP_PROP item instead of OP_CHARI. */
+
+#ifdef SUPPORT_UCP
+    if (utf && (options & PCRE_CASELESS) != 0)
+      {
+      GETCHAR(c, mcbuffer);
+      if ((c = UCD_CASESET(c)) != 0)
+        {
+        *code++ = OP_PROP;
+        *code++ = PT_CLIST;
+        *code++ = c;
+        if (firstcharflags == REQ_UNSET) firstcharflags = zerofirstcharflags = REQ_NONE;
+        break;
+        }
+      }
+#endif
+
+    /* Caseful matches, or not one of the multicase characters. */
+
+    *code++ = ((options & PCRE_CASELESS) != 0)? OP_CHARI : OP_CHAR;
+    for (c = 0; c < mclength; c++) *code++ = mcbuffer[c];
+
+    /* Remember if \r or \n were seen */
+
+    if (mcbuffer[0] == CHAR_CR || mcbuffer[0] == CHAR_NL)
+      cd->external_flags |= PCRE_HASCRORLF;
+
+    /* Set the first and required bytes appropriately. If no previous first
+    byte, set it from this character, but revert to none on a zero repeat.
+    Otherwise, leave the firstchar value alone, and don't change it on a zero
+    repeat. */
+
+    if (firstcharflags == REQ_UNSET)
+      {
+      zerofirstcharflags = REQ_NONE;
+      zeroreqchar = reqchar;
+      zeroreqcharflags = reqcharflags;
+
+      /* If the character is more than one byte long, we can set firstchar
+      only if it is not to be matched caselessly. */
+
+      if (mclength == 1 || req_caseopt == 0)
+        {
+        firstchar = mcbuffer[0] | req_caseopt;
+        firstchar = mcbuffer[0];
+        firstcharflags = req_caseopt;
+
+        if (mclength != 1)
+          {
+          reqchar = code[-1];
+          reqcharflags = cd->req_varyopt;
+          }
+        }
+      else firstcharflags = reqcharflags = REQ_NONE;
+      }
+
+    /* firstchar was previously set; we can set reqchar only if the length is
+    1 or the matching is caseful. */
+
+    else
+      {
+      zerofirstchar = firstchar;
+      zerofirstcharflags = firstcharflags;
+      zeroreqchar = reqchar;
+      zeroreqcharflags = reqcharflags;
+      if (mclength == 1 || req_caseopt == 0)
+        {
+        reqchar = code[-1];
+        reqcharflags = req_caseopt | cd->req_varyopt;
+        }
+      }
+
+    break;            /* End of literal character handling */
+    }
+  }                   /* end of big loop */
+
+
+/* Control never reaches here by falling through, only by a goto for all the
+error states. Pass back the position in the pattern so that it can be displayed
+to the user for diagnosing the error. */
+
+FAILED:
+*ptrptr = ptr;
+return FALSE;
+}
+
+
+
+/*************************************************
+*     Compile sequence of alternatives           *
+*************************************************/
+
+/* On entry, ptr is pointing past the bracket character, but on return it
+points to the closing bracket, or vertical bar, or end of string. The code
+variable is pointing at the byte into which the BRA operator has been stored.
+This function is used during the pre-compile phase when we are trying to find
+out the amount of memory needed, as well as during the real compile phase. The
+value of lengthptr distinguishes the two phases.
+
+Arguments:
+  options        option bits, including any changes for this subpattern
+  codeptr        -> the address of the current code pointer
+  ptrptr         -> the address of the current pattern pointer
+  errorcodeptr   -> pointer to error code variable
+  lookbehind     TRUE if this is a lookbehind assertion
+  reset_bracount TRUE to reset the count for each branch
+  skipbytes      skip this many bytes at start (for brackets and OP_COND)
+  cond_depth     depth of nesting for conditional subpatterns
+  firstcharptr    place to put the first required character
+  firstcharflagsptr place to put the first character flags, or a negative number
+  reqcharptr     place to put the last required character
+  reqcharflagsptr place to put the last required character flags, or a negative number
+  bcptr          pointer to the chain of currently open branches
+  cd             points to the data block with tables pointers etc.
+  lengthptr      NULL during the real compile phase
+                 points to length accumulator during pre-compile phase
+
+Returns:         TRUE on success
+*/
+
+static BOOL
+compile_regex(int options, pcre_uchar **codeptr, const pcre_uchar **ptrptr,
+  int *errorcodeptr, BOOL lookbehind, BOOL reset_bracount, int skipbytes,
+  int cond_depth,
+  pcre_uint32 *firstcharptr, pcre_int32 *firstcharflagsptr,
+  pcre_uint32 *reqcharptr, pcre_int32 *reqcharflagsptr,
+  branch_chain *bcptr, compile_data *cd, int *lengthptr)
+{
+const pcre_uchar *ptr = *ptrptr;
+pcre_uchar *code = *codeptr;
+pcre_uchar *last_branch = code;
+pcre_uchar *start_bracket = code;
+pcre_uchar *reverse_count = NULL;
+open_capitem capitem;
+int capnumber = 0;
+pcre_uint32 firstchar, reqchar;
+pcre_int32 firstcharflags, reqcharflags;
+pcre_uint32 branchfirstchar, branchreqchar;
+pcre_int32 branchfirstcharflags, branchreqcharflags;
+int length;
+unsigned int orig_bracount;
+unsigned int max_bracount;
+branch_chain bc;
+
+bc.outer = bcptr;
+bc.current_branch = code;
+
+firstchar = reqchar = 0;
+firstcharflags = reqcharflags = REQ_UNSET;
+
+/* Accumulate the length for use in the pre-compile phase. Start with the
+length of the BRA and KET and any extra bytes that are required at the
+beginning. We accumulate in a local variable to save frequent testing of
+lenthptr for NULL. We cannot do this by looking at the value of code at the
+start and end of each alternative, because compiled items are discarded during
+the pre-compile phase so that the work space is not exceeded. */
+
+length = 2 + 2*LINK_SIZE + skipbytes;
+
+/* WARNING: If the above line is changed for any reason, you must also change
+the code that abstracts option settings at the start of the pattern and makes
+them global. It tests the value of length for (2 + 2*LINK_SIZE) in the
+pre-compile phase to find out whether anything has yet been compiled or not. */
+
+/* If this is a capturing subpattern, add to the chain of open capturing items
+so that we can detect them if (*ACCEPT) is encountered. This is also used to
+detect groups that contain recursive back references to themselves. Note that
+only OP_CBRA need be tested here; changing this opcode to one of its variants,
+e.g. OP_SCBRAPOS, happens later, after the group has been compiled. */
+
+if (*code == OP_CBRA)
+  {
+  capnumber = GET2(code, 1 + LINK_SIZE);
+  capitem.number = capnumber;
+  capitem.next = cd->open_caps;
+  capitem.flag = FALSE;
+  cd->open_caps = &capitem;
+  }
+
+/* Offset is set zero to mark that this bracket is still open */
+
+PUT(code, 1, 0);
+code += 1 + LINK_SIZE + skipbytes;
+
+/* Loop for each alternative branch */
+
+orig_bracount = max_bracount = cd->bracount;
+for (;;)
+  {
+  /* For a (?| group, reset the capturing bracket count so that each branch
+  uses the same numbers. */
+
+  if (reset_bracount) cd->bracount = orig_bracount;
+
+  /* Set up dummy OP_REVERSE if lookbehind assertion */
+
+  if (lookbehind)
+    {
+    *code++ = OP_REVERSE;
+    reverse_count = code;
+    PUTINC(code, 0, 0);
+    length += 1 + LINK_SIZE;
+    }
+
+  /* Now compile the branch; in the pre-compile phase its length gets added
+  into the length. */
+
+  if (!compile_branch(&options, &code, &ptr, errorcodeptr, &branchfirstchar,
+        &branchfirstcharflags, &branchreqchar, &branchreqcharflags, &bc,
+        cond_depth, cd, (lengthptr == NULL)? NULL : &length))
+    {
+    *ptrptr = ptr;
+    return FALSE;
+    }
+
+  /* Keep the highest bracket count in case (?| was used and some branch
+  has fewer than the rest. */
+
+  if (cd->bracount > max_bracount) max_bracount = cd->bracount;
+
+  /* In the real compile phase, there is some post-processing to be done. */
+
+  if (lengthptr == NULL)
+    {
+    /* If this is the first branch, the firstchar and reqchar values for the
+    branch become the values for the regex. */
+
+    if (*last_branch != OP_ALT)
+      {
+      firstchar = branchfirstchar;
+      firstcharflags = branchfirstcharflags;
+      reqchar = branchreqchar;
+      reqcharflags = branchreqcharflags;
+      }
+
+    /* If this is not the first branch, the first char and reqchar have to
+    match the values from all the previous branches, except that if the
+    previous value for reqchar didn't have REQ_VARY set, it can still match,
+    and we set REQ_VARY for the regex. */
+
+    else
+      {
+      /* If we previously had a firstchar, but it doesn't match the new branch,
+      we have to abandon the firstchar for the regex, but if there was
+      previously no reqchar, it takes on the value of the old firstchar. */
+
+      if (firstcharflags >= 0 &&
+          (firstcharflags != branchfirstcharflags || firstchar != branchfirstchar))
+        {
+        if (reqcharflags < 0)
+          {
+          reqchar = firstchar;
+          reqcharflags = firstcharflags;
+          }
+        firstcharflags = REQ_NONE;
+        }
+
+      /* If we (now or from before) have no firstchar, a firstchar from the
+      branch becomes a reqchar if there isn't a branch reqchar. */
+
+      if (firstcharflags < 0 && branchfirstcharflags >= 0 && branchreqcharflags < 0)
+        {
+        branchreqchar = branchfirstchar;
+        branchreqcharflags = branchfirstcharflags;
+        }
+
+      /* Now ensure that the reqchars match */
+
+      if (((reqcharflags & ~REQ_VARY) != (branchreqcharflags & ~REQ_VARY)) ||
+          reqchar != branchreqchar)
+        reqcharflags = REQ_NONE;
+      else
+        {
+        reqchar = branchreqchar;
+        reqcharflags |= branchreqcharflags; /* To "or" REQ_VARY */
+        }
+      }
+
+    /* If lookbehind, check that this branch matches a fixed-length string, and
+    put the length into the OP_REVERSE item. Temporarily mark the end of the
+    branch with OP_END. If the branch contains OP_RECURSE, the result is -3
+    because there may be forward references that we can't check here. Set a
+    flag to cause another lookbehind check at the end. Why not do it all at the
+    end? Because common, erroneous checks are picked up here and the offset of
+    the problem can be shown. */
+
+    if (lookbehind)
+      {
+      int fixed_length;
+      *code = OP_END;
+      fixed_length = find_fixedlength(last_branch,  (options & PCRE_UTF8) != 0,
+        FALSE, cd);
+      DPRINTF(("fixed length = %d\n", fixed_length));
+      if (fixed_length == -3)
+        {
+        cd->check_lookbehind = TRUE;
+        }
+      else if (fixed_length < 0)
+        {
+        *errorcodeptr = (fixed_length == -2)? ERR36 :
+                        (fixed_length == -4)? ERR70: ERR25;
+        *ptrptr = ptr;
+        return FALSE;
+        }
+      else
+        {
+        if (fixed_length > cd->max_lookbehind)
+          cd->max_lookbehind = fixed_length;
+        PUT(reverse_count, 0, fixed_length);
+        }
+      }
+    }
+
+  /* Reached end of expression, either ')' or end of pattern. In the real
+  compile phase, go back through the alternative branches and reverse the chain
+  of offsets, with the field in the BRA item now becoming an offset to the
+  first alternative. If there are no alternatives, it points to the end of the
+  group. The length in the terminating ket is always the length of the whole
+  bracketed item. Return leaving the pointer at the terminating char. */
+
+  if (*ptr != CHAR_VERTICAL_LINE)
+    {
+    if (lengthptr == NULL)
+      {
+      int branch_length = (int)(code - last_branch);
+      do
+        {
+        int prev_length = GET(last_branch, 1);
+        PUT(last_branch, 1, branch_length);
+        branch_length = prev_length;
+        last_branch -= branch_length;
+        }
+      while (branch_length > 0);
+      }
+
+    /* Fill in the ket */
+
+    *code = OP_KET;
+    PUT(code, 1, (int)(code - start_bracket));
+    code += 1 + LINK_SIZE;
+
+    /* If it was a capturing subpattern, check to see if it contained any
+    recursive back references. If so, we must wrap it in atomic brackets.
+    In any event, remove the block from the chain. */
+
+    if (capnumber > 0)
+      {
+      if (cd->open_caps->flag)
+        {
+        memmove(start_bracket + 1 + LINK_SIZE, start_bracket,
+          IN_UCHARS(code - start_bracket));
+        *start_bracket = OP_ONCE;
+        code += 1 + LINK_SIZE;
+        PUT(start_bracket, 1, (int)(code - start_bracket));
+        *code = OP_KET;
+        PUT(code, 1, (int)(code - start_bracket));
+        code += 1 + LINK_SIZE;
+        length += 2 + 2*LINK_SIZE;
+        }
+      cd->open_caps = cd->open_caps->next;
+      }
+
+    /* Retain the highest bracket number, in case resetting was used. */
+
+    cd->bracount = max_bracount;
+
+    /* Set values to pass back */
+
+    *codeptr = code;
+    *ptrptr = ptr;
+    *firstcharptr = firstchar;
+    *firstcharflagsptr = firstcharflags;
+    *reqcharptr = reqchar;
+    *reqcharflagsptr = reqcharflags;
+    if (lengthptr != NULL)
+      {
+      if (OFLOW_MAX - *lengthptr < length)
+        {
+        *errorcodeptr = ERR20;
+        return FALSE;
+        }
+      *lengthptr += length;
+      }
+    return TRUE;
+    }
+
+  /* Another branch follows. In the pre-compile phase, we can move the code
+  pointer back to where it was for the start of the first branch. (That is,
+  pretend that each branch is the only one.)
+
+  In the real compile phase, insert an ALT node. Its length field points back
+  to the previous branch while the bracket remains open. At the end the chain
+  is reversed. It's done like this so that the start of the bracket has a
+  zero offset until it is closed, making it possible to detect recursion. */
+
+  if (lengthptr != NULL)
+    {
+    code = *codeptr + 1 + LINK_SIZE + skipbytes;
+    length += 1 + LINK_SIZE;
+    }
+  else
+    {
+    *code = OP_ALT;
+    PUT(code, 1, (int)(code - last_branch));
+    bc.current_branch = last_branch = code;
+    code += 1 + LINK_SIZE;
+    }
+
+  ptr++;
+  }
+/* Control never reaches here */
+}
+
+
+
+
+/*************************************************
+*          Check for anchored expression         *
+*************************************************/
+
+/* Try to find out if this is an anchored regular expression. Consider each
+alternative branch. If they all start with OP_SOD or OP_CIRC, or with a bracket
+all of whose alternatives start with OP_SOD or OP_CIRC (recurse ad lib), then
+it's anchored. However, if this is a multiline pattern, then only OP_SOD will
+be found, because ^ generates OP_CIRCM in that mode.
+
+We can also consider a regex to be anchored if OP_SOM starts all its branches.
+This is the code for \G, which means "match at start of match position, taking
+into account the match offset".
+
+A branch is also implicitly anchored if it starts with .* and DOTALL is set,
+because that will try the rest of the pattern at all possible matching points,
+so there is no point trying again.... er ....
+
+.... except when the .* appears inside capturing parentheses, and there is a
+subsequent back reference to those parentheses. We haven't enough information
+to catch that case precisely.
+
+At first, the best we could do was to detect when .* was in capturing brackets
+and the highest back reference was greater than or equal to that level.
+However, by keeping a bitmap of the first 31 back references, we can catch some
+of the more common cases more precisely.
+
+... A second exception is when the .* appears inside an atomic group, because
+this prevents the number of characters it matches from being adjusted.
+
+Arguments:
+  code           points to start of expression (the bracket)
+  bracket_map    a bitmap of which brackets we are inside while testing; this
+                  handles up to substring 31; after that we just have to take
+                  the less precise approach
+  cd             points to the compile data block
+  atomcount      atomic group level
+
+Returns:     TRUE or FALSE
+*/
+
+static BOOL
+is_anchored(register const pcre_uchar *code, unsigned int bracket_map,
+  compile_data *cd, int atomcount)
+{
+do {
+   const pcre_uchar *scode = first_significant_code(
+     code + PRIV(OP_lengths)[*code], FALSE);
+   register int op = *scode;
+
+   /* Non-capturing brackets */
+
+   if (op == OP_BRA  || op == OP_BRAPOS ||
+       op == OP_SBRA || op == OP_SBRAPOS)
+     {
+     if (!is_anchored(scode, bracket_map, cd, atomcount)) return FALSE;
+     }
+
+   /* Capturing brackets */
+
+   else if (op == OP_CBRA  || op == OP_CBRAPOS ||
+            op == OP_SCBRA || op == OP_SCBRAPOS)
+     {
+     int n = GET2(scode, 1+LINK_SIZE);
+     int new_map = bracket_map | ((n < 32)? (1 << n) : 1);
+     if (!is_anchored(scode, new_map, cd, atomcount)) return FALSE;
+     }
+
+   /* Positive forward assertions and conditions */
+
+   else if (op == OP_ASSERT || op == OP_COND)
+     {
+     if (!is_anchored(scode, bracket_map, cd, atomcount)) return FALSE;
+     }
+
+   /* Atomic groups */
+
+   else if (op == OP_ONCE || op == OP_ONCE_NC)
+     {
+     if (!is_anchored(scode, bracket_map, cd, atomcount + 1))
+       return FALSE;
+     }
+
+   /* .* is not anchored unless DOTALL is set (which generates OP_ALLANY) and
+   it isn't in brackets that are or may be referenced or inside an atomic
+   group. */
+
+   else if ((op == OP_TYPESTAR || op == OP_TYPEMINSTAR ||
+             op == OP_TYPEPOSSTAR))
+     {
+     if (scode[1] != OP_ALLANY || (bracket_map & cd->backref_map) != 0 ||
+         atomcount > 0 || cd->had_pruneorskip)
+       return FALSE;
+     }
+
+   /* Check for explicit anchoring */
+
+   else if (op != OP_SOD && op != OP_SOM && op != OP_CIRC) return FALSE;
+
+   code += GET(code, 1);
+   }
+while (*code == OP_ALT);   /* Loop for each alternative */
+return TRUE;
+}
+
+
+
+/*************************************************
+*         Check for starting with ^ or .*        *
+*************************************************/
+
+/* This is called to find out if every branch starts with ^ or .* so that
+"first char" processing can be done to speed things up in multiline
+matching and for non-DOTALL patterns that start with .* (which must start at
+the beginning or after \n). As in the case of is_anchored() (see above), we
+have to take account of back references to capturing brackets that contain .*
+because in that case we can't make the assumption. Also, the appearance of .*
+inside atomic brackets or in a pattern that contains *PRUNE or *SKIP does not
+count, because once again the assumption no longer holds.
+
+Arguments:
+  code           points to start of expression (the bracket)
+  bracket_map    a bitmap of which brackets we are inside while testing; this
+                  handles up to substring 31; after that we just have to take
+                  the less precise approach
+  cd             points to the compile data
+  atomcount      atomic group level
+
+Returns:         TRUE or FALSE
+*/
+
+static BOOL
+is_startline(const pcre_uchar *code, unsigned int bracket_map,
+  compile_data *cd, int atomcount)
+{
+do {
+   const pcre_uchar *scode = first_significant_code(
+     code + PRIV(OP_lengths)[*code], FALSE);
+   register int op = *scode;
+
+   /* If we are at the start of a conditional assertion group, *both* the
+   conditional assertion *and* what follows the condition must satisfy the test
+   for start of line. Other kinds of condition fail. Note that there may be an
+   auto-callout at the start of a condition. */
+
+   if (op == OP_COND)
+     {
+     scode += 1 + LINK_SIZE;
+     if (*scode == OP_CALLOUT) scode += PRIV(OP_lengths)[OP_CALLOUT];
+     switch (*scode)
+       {
+       case OP_CREF:
+       case OP_NCREF:
+       case OP_RREF:
+       case OP_NRREF:
+       case OP_DEF:
+       return FALSE;
+
+       default:     /* Assertion */
+       if (!is_startline(scode, bracket_map, cd, atomcount)) return FALSE;
+       do scode += GET(scode, 1); while (*scode == OP_ALT);
+       scode += 1 + LINK_SIZE;
+       break;
+       }
+     scode = first_significant_code(scode, FALSE);
+     op = *scode;
+     }
+
+   /* Non-capturing brackets */
+
+   if (op == OP_BRA  || op == OP_BRAPOS ||
+       op == OP_SBRA || op == OP_SBRAPOS)
+     {
+     if (!is_startline(scode, bracket_map, cd, atomcount)) return FALSE;
+     }
+
+   /* Capturing brackets */
+
+   else if (op == OP_CBRA  || op == OP_CBRAPOS ||
+            op == OP_SCBRA || op == OP_SCBRAPOS)
+     {
+     int n = GET2(scode, 1+LINK_SIZE);
+     int new_map = bracket_map | ((n < 32)? (1 << n) : 1);
+     if (!is_startline(scode, new_map, cd, atomcount)) return FALSE;
+     }
+
+   /* Positive forward assertions */
+
+   else if (op == OP_ASSERT)
+     {
+     if (!is_startline(scode, bracket_map, cd, atomcount)) return FALSE;
+     }
+
+   /* Atomic brackets */
+
+   else if (op == OP_ONCE || op == OP_ONCE_NC)
+     {
+     if (!is_startline(scode, bracket_map, cd, atomcount + 1)) return FALSE;
+     }
+
+   /* .* means "start at start or after \n" if it isn't in atomic brackets or
+   brackets that may be referenced, as long as the pattern does not contain
+   *PRUNE or *SKIP, because these break the feature. Consider, for example,
+   /.*?a(*PRUNE)b/ with the subject "aab", which matches "ab", i.e. not at the
+   start of a line. */
+
+   else if (op == OP_TYPESTAR || op == OP_TYPEMINSTAR || op == OP_TYPEPOSSTAR)
+     {
+     if (scode[1] != OP_ANY || (bracket_map & cd->backref_map) != 0 ||
+         atomcount > 0 || cd->had_pruneorskip)
+       return FALSE;
+     }
+
+   /* Check for explicit circumflex; anything else gives a FALSE result. Note
+   in particular that this includes atomic brackets OP_ONCE and OP_ONCE_NC
+   because the number of characters matched by .* cannot be adjusted inside
+   them. */
+
+   else if (op != OP_CIRC && op != OP_CIRCM) return FALSE;
+
+   /* Move on to the next alternative */
+
+   code += GET(code, 1);
+   }
+while (*code == OP_ALT);  /* Loop for each alternative */
+return TRUE;
+}
+
+
+
+/*************************************************
+*       Check for asserted fixed first char      *
+*************************************************/
+
+/* During compilation, the "first char" settings from forward assertions are
+discarded, because they can cause conflicts with actual literals that follow.
+However, if we end up without a first char setting for an unanchored pattern,
+it is worth scanning the regex to see if there is an initial asserted first
+char. If all branches start with the same asserted char, or with a bracket all
+of whose alternatives start with the same asserted char (recurse ad lib), then
+we return that char, otherwise -1.
+
+Arguments:
+  code       points to start of expression (the bracket)
+  flags       points to the first char flags, or to REQ_NONE
+  inassert   TRUE if in an assertion
+
+Returns:     the fixed first char, or 0 with REQ_NONE in flags
+*/
+
+static pcre_uint32
+find_firstassertedchar(const pcre_uchar *code, pcre_int32 *flags,
+  BOOL inassert)
+{
+register pcre_uint32 c = 0;
+int cflags = REQ_NONE;
+
+*flags = REQ_NONE;
+do {
+   pcre_uint32 d;
+   int dflags;
+   int xl = (*code == OP_CBRA || *code == OP_SCBRA ||
+             *code == OP_CBRAPOS || *code == OP_SCBRAPOS)? IMM2_SIZE:0;
+   const pcre_uchar *scode = first_significant_code(code + 1+LINK_SIZE + xl,
+     TRUE);
+   register pcre_uchar op = *scode;
+
+   switch(op)
+     {
+     default:
+     return 0;
+
+     case OP_BRA:
+     case OP_BRAPOS:
+     case OP_CBRA:
+     case OP_SCBRA:
+     case OP_CBRAPOS:
+     case OP_SCBRAPOS:
+     case OP_ASSERT:
+     case OP_ONCE:
+     case OP_ONCE_NC:
+     case OP_COND:
+     d = find_firstassertedchar(scode, &dflags, op == OP_ASSERT);
+     if (dflags < 0)
+       return 0;
+     if (cflags < 0) { c = d; cflags = dflags; } else if (c != d || cflags != dflags) return 0;
+     break;
+
+     case OP_EXACT:
+     scode += IMM2_SIZE;
+     /* Fall through */
+
+     case OP_CHAR:
+     case OP_PLUS:
+     case OP_MINPLUS:
+     case OP_POSPLUS:
+     if (!inassert) return 0;
+     if (cflags < 0) { c = scode[1]; cflags = 0; }
+       else if (c != scode[1]) return 0;
+     break;
+
+     case OP_EXACTI:
+     scode += IMM2_SIZE;
+     /* Fall through */
+
+     case OP_CHARI:
+     case OP_PLUSI:
+     case OP_MINPLUSI:
+     case OP_POSPLUSI:
+     if (!inassert) return 0;
+     if (cflags < 0) { c = scode[1]; cflags = REQ_CASELESS; }
+       else if (c != scode[1]) return 0;
+     break;
+     }
+
+   code += GET(code, 1);
+   }
+while (*code == OP_ALT);
+
+*flags = cflags;
+return c;
+}
+
+
+
+/*************************************************
+*        Compile a Regular Expression            *
+*************************************************/
+
+/* This function takes a string and returns a pointer to a block of store
+holding a compiled version of the expression. The original API for this
+function had no error code return variable; it is retained for backwards
+compatibility. The new function is given a new name.
+
+Arguments:
+  pattern       the regular expression
+  options       various option bits
+  errorcodeptr  pointer to error code variable (pcre_compile2() only)
+                  can be NULL if you don't want a code value
+  errorptr      pointer to pointer to error text
+  erroroffset   ptr offset in pattern where error was detected
+  tables        pointer to character tables or NULL
+
+Returns:        pointer to compiled data block, or NULL on error,
+                with errorptr and erroroffset set
+*/
+
+#if defined COMPILE_PCRE8
+PCRE_EXP_DEFN pcre * PCRE_CALL_CONVENTION
+pcre_compile(const char *pattern, int options, const char **errorptr,
+  int *erroroffset, const unsigned char *tables)
+#elif defined COMPILE_PCRE16
+PCRE_EXP_DEFN pcre16 * PCRE_CALL_CONVENTION
+pcre16_compile(PCRE_SPTR16 pattern, int options, const char **errorptr,
+  int *erroroffset, const unsigned char *tables)
+#elif defined COMPILE_PCRE32
+PCRE_EXP_DEFN pcre32 * PCRE_CALL_CONVENTION
+pcre32_compile(PCRE_SPTR32 pattern, int options, const char **errorptr,
+  int *erroroffset, const unsigned char *tables)
+#endif
+{
+#if defined COMPILE_PCRE8
+return pcre_compile2(pattern, options, NULL, errorptr, erroroffset, tables);
+#elif defined COMPILE_PCRE16
+return pcre16_compile2(pattern, options, NULL, errorptr, erroroffset, tables);
+#elif defined COMPILE_PCRE32
+return pcre32_compile2(pattern, options, NULL, errorptr, erroroffset, tables);
+#endif
+}
+
+
+#if defined COMPILE_PCRE8
+PCRE_EXP_DEFN pcre * PCRE_CALL_CONVENTION
+pcre_compile2(const char *pattern, int options, int *errorcodeptr,
+  const char **errorptr, int *erroroffset, const unsigned char *tables)
+#elif defined COMPILE_PCRE16
+PCRE_EXP_DEFN pcre16 * PCRE_CALL_CONVENTION
+pcre16_compile2(PCRE_SPTR16 pattern, int options, int *errorcodeptr,
+  const char **errorptr, int *erroroffset, const unsigned char *tables)
+#elif defined COMPILE_PCRE32
+PCRE_EXP_DEFN pcre32 * PCRE_CALL_CONVENTION
+pcre32_compile2(PCRE_SPTR32 pattern, int options, int *errorcodeptr,
+  const char **errorptr, int *erroroffset, const unsigned char *tables)
+#endif
+{
+REAL_PCRE *re;
+int length = 1;  /* For final END opcode */
+pcre_int32 firstcharflags, reqcharflags;
+pcre_uint32 firstchar, reqchar;
+pcre_uint32 limit_match = PCRE_UINT32_MAX;
+pcre_uint32 limit_recursion = PCRE_UINT32_MAX;
+int newline;
+int errorcode = 0;
+int skipatstart = 0;
+BOOL utf;
+BOOL never_utf = FALSE;
+size_t size;
+pcre_uchar *code;
+const pcre_uchar *codestart;
+const pcre_uchar *ptr;
+compile_data compile_block;
+compile_data *cd = &compile_block;
+
+/* This space is used for "compiling" into during the first phase, when we are
+computing the amount of memory that is needed. Compiled items are thrown away
+as soon as possible, so that a fairly large buffer should be sufficient for
+this purpose. The same space is used in the second phase for remembering where
+to fill in forward references to subpatterns. That may overflow, in which case
+new memory is obtained from malloc(). */
+
+pcre_uchar cworkspace[COMPILE_WORK_SIZE];
+
+/* Set this early so that early errors get offset 0. */
+
+ptr = (const pcre_uchar *)pattern;
+
+/* We can't pass back an error message if errorptr is NULL; I guess the best we
+can do is just return NULL, but we can set a code value if there is a code
+pointer. */
+
+if (errorptr == NULL)
+  {
+  if (errorcodeptr != NULL) *errorcodeptr = 99;
+  return NULL;
+  }
+
+*errorptr = NULL;
+if (errorcodeptr != NULL) *errorcodeptr = ERR0;
+
+/* However, we can give a message for this error */
+
+if (erroroffset == NULL)
+  {
+  errorcode = ERR16;
+  goto PCRE_EARLY_ERROR_RETURN2;
+  }
+
+*erroroffset = 0;
+
+/* Set up pointers to the individual character tables */
+
+if (tables == NULL) tables = PRIV(default_tables);
+cd->lcc = tables + lcc_offset;
+cd->fcc = tables + fcc_offset;
+cd->cbits = tables + cbits_offset;
+cd->ctypes = tables + ctypes_offset;
+
+/* Check that all undefined public option bits are zero */
+
+if ((options & ~PUBLIC_COMPILE_OPTIONS) != 0)
+  {
+  errorcode = ERR17;
+  goto PCRE_EARLY_ERROR_RETURN;
+  }
+
+/* If PCRE_NEVER_UTF is set, remember it. */
+
+if ((options & PCRE_NEVER_UTF) != 0) never_utf = TRUE;
+
+/* Check for global one-time settings at the start of the pattern, and remember
+the offset for later. */
+
+cd->external_flags = 0;   /* Initialize here for LIMIT_MATCH/RECURSION */
+
+while (ptr[skipatstart] == CHAR_LEFT_PARENTHESIS &&
+       ptr[skipatstart+1] == CHAR_ASTERISK)
+  {
+  int newnl = 0;
+  int newbsr = 0;
+
+/* For completeness and backward compatibility, (*UTFn) is supported in the
+relevant libraries, but (*UTF) is generic and always supported. Note that
+PCRE_UTF8 == PCRE_UTF16 == PCRE_UTF32. */
+
+#ifdef COMPILE_PCRE8
+  if (STRNCMP_UC_C8(ptr+skipatstart+2, STRING_UTF8_RIGHTPAR, 5) == 0)
+    { skipatstart += 7; options |= PCRE_UTF8; continue; }
+#endif
+#ifdef COMPILE_PCRE16
+  if (STRNCMP_UC_C8(ptr+skipatstart+2, STRING_UTF16_RIGHTPAR, 6) == 0)
+    { skipatstart += 8; options |= PCRE_UTF16; continue; }
+#endif
+#ifdef COMPILE_PCRE32
+  if (STRNCMP_UC_C8(ptr+skipatstart+2, STRING_UTF32_RIGHTPAR, 6) == 0)
+    { skipatstart += 8; options |= PCRE_UTF32; continue; }
+#endif
+
+  else if (STRNCMP_UC_C8(ptr+skipatstart+2, STRING_UTF_RIGHTPAR, 4) == 0)
+    { skipatstart += 6; options |= PCRE_UTF8; continue; }
+  else if (STRNCMP_UC_C8(ptr+skipatstart+2, STRING_UCP_RIGHTPAR, 4) == 0)
+    { skipatstart += 6; options |= PCRE_UCP; continue; }
+  else if (STRNCMP_UC_C8(ptr+skipatstart+2, STRING_NO_START_OPT_RIGHTPAR, 13) == 0)
+    { skipatstart += 15; options |= PCRE_NO_START_OPTIMIZE; continue; }
+
+  else if (STRNCMP_UC_C8(ptr+skipatstart+2, STRING_LIMIT_MATCH_EQ, 12) == 0)
+    {
+    pcre_uint32 c = 0;
+    int p = skipatstart + 14;
+    while (isdigit(ptr[p]))
+      {
+      if (c > PCRE_UINT32_MAX / 10 - 1) break;   /* Integer overflow */
+      c = c*10 + ptr[p++] - CHAR_0;
+      }
+    if (ptr[p++] != CHAR_RIGHT_PARENTHESIS) break;
+    if (c < limit_match)
+      {
+      limit_match = c;
+      cd->external_flags |= PCRE_MLSET;
+      }
+    skipatstart = p;
+    continue;
+    }
+
+  else if (STRNCMP_UC_C8(ptr+skipatstart+2, STRING_LIMIT_RECURSION_EQ, 16) == 0)
+    {
+    pcre_uint32 c = 0;
+    int p = skipatstart + 18;
+    while (isdigit(ptr[p]))
+      {
+      if (c > PCRE_UINT32_MAX / 10 - 1) break;   /* Integer overflow check */
+      c = c*10 + ptr[p++] - CHAR_0;
+      }
+    if (ptr[p++] != CHAR_RIGHT_PARENTHESIS) break;
+    if (c < limit_recursion)
+      {
+      limit_recursion = c;
+      cd->external_flags |= PCRE_RLSET;
+      }
+    skipatstart = p;
+    continue;
+    }
+
+  if (STRNCMP_UC_C8(ptr+skipatstart+2, STRING_CR_RIGHTPAR, 3) == 0)
+    { skipatstart += 5; newnl = PCRE_NEWLINE_CR; }
+  else if (STRNCMP_UC_C8(ptr+skipatstart+2, STRING_LF_RIGHTPAR, 3)  == 0)
+    { skipatstart += 5; newnl = PCRE_NEWLINE_LF; }
+  else if (STRNCMP_UC_C8(ptr+skipatstart+2, STRING_CRLF_RIGHTPAR, 5)  == 0)
+    { skipatstart += 7; newnl = PCRE_NEWLINE_CR + PCRE_NEWLINE_LF; }
+  else if (STRNCMP_UC_C8(ptr+skipatstart+2, STRING_ANY_RIGHTPAR, 4) == 0)
+    { skipatstart += 6; newnl = PCRE_NEWLINE_ANY; }
+  else if (STRNCMP_UC_C8(ptr+skipatstart+2, STRING_ANYCRLF_RIGHTPAR, 8) == 0)
+    { skipatstart += 10; newnl = PCRE_NEWLINE_ANYCRLF; }
+
+  else if (STRNCMP_UC_C8(ptr+skipatstart+2, STRING_BSR_ANYCRLF_RIGHTPAR, 12) == 0)
+    { skipatstart += 14; newbsr = PCRE_BSR_ANYCRLF; }
+  else if (STRNCMP_UC_C8(ptr+skipatstart+2, STRING_BSR_UNICODE_RIGHTPAR, 12) == 0)
+    { skipatstart += 14; newbsr = PCRE_BSR_UNICODE; }
+
+  if (newnl != 0)
+    options = (options & ~PCRE_NEWLINE_BITS) | newnl;
+  else if (newbsr != 0)
+    options = (options & ~(PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE)) | newbsr;
+  else break;
+  }
+
+/* PCRE_UTF(16|32) have the same value as PCRE_UTF8. */
+utf = (options & PCRE_UTF8) != 0;
+if (utf && never_utf)
+  {
+  errorcode = ERR78;
+  goto PCRE_EARLY_ERROR_RETURN2;
+  }
+
+/* Can't support UTF unless PCRE has been compiled to include the code. The
+return of an error code from PRIV(valid_utf)() is a new feature, introduced in
+release 8.13. It is passed back from pcre_[dfa_]exec(), but at the moment is
+not used here. */
+
+#ifdef SUPPORT_UTF
+if (utf && (options & PCRE_NO_UTF8_CHECK) == 0 &&
+     (errorcode = PRIV(valid_utf)((PCRE_PUCHAR)pattern, -1, erroroffset)) != 0)
+  {
+#if defined COMPILE_PCRE8
+  errorcode = ERR44;
+#elif defined COMPILE_PCRE16
+  errorcode = ERR74;
+#elif defined COMPILE_PCRE32
+  errorcode = ERR77;
+#endif
+  goto PCRE_EARLY_ERROR_RETURN2;
+  }
+#else
+if (utf)
+  {
+  errorcode = ERR32;
+  goto PCRE_EARLY_ERROR_RETURN;
+  }
+#endif
+
+/* Can't support UCP unless PCRE has been compiled to include the code. */
+
+#ifndef SUPPORT_UCP
+if ((options & PCRE_UCP) != 0)
+  {
+  errorcode = ERR67;
+  goto PCRE_EARLY_ERROR_RETURN;
+  }
+#endif
+
+/* Check validity of \R options. */
+
+if ((options & (PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE)) ==
+     (PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE))
+  {
+  errorcode = ERR56;
+  goto PCRE_EARLY_ERROR_RETURN;
+  }
+
+/* Handle different types of newline. The three bits give seven cases. The
+current code allows for fixed one- or two-byte sequences, plus "any" and
+"anycrlf". */
+
+switch (options & PCRE_NEWLINE_BITS)
+  {
+  case 0: newline = NEWLINE; break;   /* Build-time default */
+  case PCRE_NEWLINE_CR: newline = CHAR_CR; break;
+  case PCRE_NEWLINE_LF: newline = CHAR_NL; break;
+  case PCRE_NEWLINE_CR+
+       PCRE_NEWLINE_LF: newline = (CHAR_CR << 8) | CHAR_NL; break;
+  case PCRE_NEWLINE_ANY: newline = -1; break;
+  case PCRE_NEWLINE_ANYCRLF: newline = -2; break;
+  default: errorcode = ERR56; goto PCRE_EARLY_ERROR_RETURN;
+  }
+
+if (newline == -2)
+  {
+  cd->nltype = NLTYPE_ANYCRLF;
+  }
+else if (newline < 0)
+  {
+  cd->nltype = NLTYPE_ANY;
+  }
+else
+  {
+  cd->nltype = NLTYPE_FIXED;
+  if (newline > 255)
+    {
+    cd->nllen = 2;
+    cd->nl[0] = (newline >> 8) & 255;
+    cd->nl[1] = newline & 255;
+    }
+  else
+    {
+    cd->nllen = 1;
+    cd->nl[0] = newline;
+    }
+  }
+
+/* Maximum back reference and backref bitmap. The bitmap records up to 31 back
+references to help in deciding whether (.*) can be treated as anchored or not.
+*/
+
+cd->top_backref = 0;
+cd->backref_map = 0;
+
+/* Reflect pattern for debugging output */
+
+DPRINTF(("------------------------------------------------------------------\n"));
+#ifdef PCRE_DEBUG
+print_puchar(stdout, (PCRE_PUCHAR)pattern);
+#endif
+DPRINTF(("\n"));
+
+/* Pretend to compile the pattern while actually just accumulating the length
+of memory required. This behaviour is triggered by passing a non-NULL final
+argument to compile_regex(). We pass a block of workspace (cworkspace) for it
+to compile parts of the pattern into; the compiled code is discarded when it is
+no longer needed, so hopefully this workspace will never overflow, though there
+is a test for its doing so. */
+
+cd->bracount = cd->final_bracount = 0;
+cd->names_found = 0;
+cd->name_entry_size = 0;
+cd->name_table = NULL;
+cd->start_code = cworkspace;
+cd->hwm = cworkspace;
+cd->start_workspace = cworkspace;
+cd->workspace_size = COMPILE_WORK_SIZE;
+cd->start_pattern = (const pcre_uchar *)pattern;
+cd->end_pattern = (const pcre_uchar *)(pattern + STRLEN_UC((const pcre_uchar *)pattern));
+cd->req_varyopt = 0;
+cd->assert_depth = 0;
+cd->max_lookbehind = 0;
+cd->external_options = options;
+cd->open_caps = NULL;
+
+/* Now do the pre-compile. On error, errorcode will be set non-zero, so we
+don't need to look at the result of the function here. The initial options have
+been put into the cd block so that they can be changed if an option setting is
+found within the regex right at the beginning. Bringing initial option settings
+outside can help speed up starting point checks. */
+
+ptr += skipatstart;
+code = cworkspace;
+*code = OP_BRA;
+(void)compile_regex(cd->external_options, &code, &ptr, &errorcode, FALSE,
+  FALSE, 0, 0, &firstchar, &firstcharflags, &reqchar, &reqcharflags, NULL,
+  cd, &length);
+if (errorcode != 0) goto PCRE_EARLY_ERROR_RETURN;
+
+DPRINTF(("end pre-compile: length=%d workspace=%d\n", length,
+  (int)(cd->hwm - cworkspace)));
+
+if (length > MAX_PATTERN_SIZE)
+  {
+  errorcode = ERR20;
+  goto PCRE_EARLY_ERROR_RETURN;
+  }
+
+/* Compute the size of data block needed and get it, either from malloc or
+externally provided function. Integer overflow should no longer be possible
+because nowadays we limit the maximum value of cd->names_found and
+cd->name_entry_size. */
+
+size = sizeof(REAL_PCRE) + (length + cd->names_found * cd->name_entry_size) * sizeof(pcre_uchar);
+re = (REAL_PCRE *)(PUBL(malloc))(size);
+
+if (re == NULL)
+  {
+  errorcode = ERR21;
+  goto PCRE_EARLY_ERROR_RETURN;
+  }
+
+/* Put in the magic number, and save the sizes, initial options, internal
+flags, and character table pointer. NULL is used for the default character
+tables. The nullpad field is at the end; it's there to help in the case when a
+regex compiled on a system with 4-byte pointers is run on another with 8-byte
+pointers. */
+
+re->magic_number = MAGIC_NUMBER;
+re->size = (int)size;
+re->options = cd->external_options;
+re->flags = cd->external_flags;
+re->limit_match = limit_match;
+re->limit_recursion = limit_recursion;
+re->first_char = 0;
+re->req_char = 0;
+re->name_table_offset = sizeof(REAL_PCRE) / sizeof(pcre_uchar);
+re->name_entry_size = cd->name_entry_size;
+re->name_count = cd->names_found;
+re->ref_count = 0;
+re->tables = (tables == PRIV(default_tables))? NULL : tables;
+re->nullpad = NULL;
+#ifdef COMPILE_PCRE32
+re->dummy = 0;
+#else
+re->dummy1 = re->dummy2 = re->dummy3 = 0;
+#endif
+
+/* The starting points of the name/number translation table and of the code are
+passed around in the compile data block. The start/end pattern and initial
+options are already set from the pre-compile phase, as is the name_entry_size
+field. Reset the bracket count and the names_found field. Also reset the hwm
+field; this time it's used for remembering forward references to subpatterns.
+*/
+
+cd->final_bracount = cd->bracount;  /* Save for checking forward references */
+cd->assert_depth = 0;
+cd->bracount = 0;
+cd->max_lookbehind = 0;
+cd->names_found = 0;
+cd->name_table = (pcre_uchar *)re + re->name_table_offset;
+codestart = cd->name_table + re->name_entry_size * re->name_count;
+cd->start_code = codestart;
+cd->hwm = (pcre_uchar *)(cd->start_workspace);
+cd->req_varyopt = 0;
+cd->had_accept = FALSE;
+cd->had_pruneorskip = FALSE;
+cd->check_lookbehind = FALSE;
+cd->open_caps = NULL;
+
+/* Set up a starting, non-extracting bracket, then compile the expression. On
+error, errorcode will be set non-zero, so we don't need to look at the result
+of the function here. */
+
+ptr = (const pcre_uchar *)pattern + skipatstart;
+code = (pcre_uchar *)codestart;
+*code = OP_BRA;
+(void)compile_regex(re->options, &code, &ptr, &errorcode, FALSE, FALSE, 0, 0,
+  &firstchar, &firstcharflags, &reqchar, &reqcharflags, NULL, cd, NULL);
+re->top_bracket = cd->bracount;
+re->top_backref = cd->top_backref;
+re->max_lookbehind = cd->max_lookbehind;
+re->flags = cd->external_flags | PCRE_MODE;
+
+if (cd->had_accept)
+  {
+  reqchar = 0;              /* Must disable after (*ACCEPT) */
+  reqcharflags = REQ_NONE;
+  }
+
+/* If not reached end of pattern on success, there's an excess bracket. */
+
+if (errorcode == 0 && *ptr != CHAR_NULL) errorcode = ERR22;
+
+/* Fill in the terminating state and check for disastrous overflow, but
+if debugging, leave the test till after things are printed out. */
+
+*code++ = OP_END;
+
+#ifndef PCRE_DEBUG
+if (code - codestart > length) errorcode = ERR23;
+#endif
+
+#ifdef SUPPORT_VALGRIND
+/* If the estimated length exceeds the really used length, mark the extra
+allocated memory as unaddressable, so that any out-of-bound reads can be
+detected. */
+VALGRIND_MAKE_MEM_NOACCESS(code, (length - (code - codestart)) * sizeof(pcre_uchar));
+#endif
+
+/* Fill in any forward references that are required. There may be repeated
+references; optimize for them, as searching a large regex takes time. */
+
+if (cd->hwm > cd->start_workspace)
+  {
+  int prev_recno = -1;
+  const pcre_uchar *groupptr = NULL;
+  while (errorcode == 0 && cd->hwm > cd->start_workspace)
+    {
+    int offset, recno;
+    cd->hwm -= LINK_SIZE;
+    offset = GET(cd->hwm, 0);
+    recno = GET(codestart, offset);
+    if (recno != prev_recno)
+      {
+      groupptr = PRIV(find_bracket)(codestart, utf, recno);
+      prev_recno = recno;
+      }
+    if (groupptr == NULL) errorcode = ERR53;
+      else PUT(((pcre_uchar *)codestart), offset, (int)(groupptr - codestart));
+    }
+  }
+
+/* If the workspace had to be expanded, free the new memory. */
+
+if (cd->workspace_size > COMPILE_WORK_SIZE)
+  (PUBL(free))((void *)cd->start_workspace);
+
+/* Give an error if there's back reference to a non-existent capturing
+subpattern. */
+
+if (errorcode == 0 && re->top_backref > re->top_bracket) errorcode = ERR15;
+
+/* If there were any lookbehind assertions that contained OP_RECURSE
+(recursions or subroutine calls), a flag is set for them to be checked here,
+because they may contain forward references. Actual recursions cannot be fixed
+length, but subroutine calls can. It is done like this so that those without
+OP_RECURSE that are not fixed length get a diagnosic with a useful offset. The
+exceptional ones forgo this. We scan the pattern to check that they are fixed
+length, and set their lengths. */
+
+if (cd->check_lookbehind)
+  {
+  pcre_uchar *cc = (pcre_uchar *)codestart;
+
+  /* Loop, searching for OP_REVERSE items, and process those that do not have
+  their length set. (Actually, it will also re-process any that have a length
+  of zero, but that is a pathological case, and it does no harm.) When we find
+  one, we temporarily terminate the branch it is in while we scan it. */
+
+  for (cc = (pcre_uchar *)PRIV(find_bracket)(codestart, utf, -1);
+       cc != NULL;
+       cc = (pcre_uchar *)PRIV(find_bracket)(cc, utf, -1))
+    {
+    if (GET(cc, 1) == 0)
+      {
+      int fixed_length;
+      pcre_uchar *be = cc - 1 - LINK_SIZE + GET(cc, -LINK_SIZE);
+      int end_op = *be;
+      *be = OP_END;
+      fixed_length = find_fixedlength(cc, (re->options & PCRE_UTF8) != 0, TRUE,
+        cd);
+      *be = end_op;
+      DPRINTF(("fixed length = %d\n", fixed_length));
+      if (fixed_length < 0)
+        {
+        errorcode = (fixed_length == -2)? ERR36 :
+                    (fixed_length == -4)? ERR70 : ERR25;
+        break;
+        }
+      if (fixed_length > cd->max_lookbehind) cd->max_lookbehind = fixed_length;
+      PUT(cc, 1, fixed_length);
+      }
+    cc += 1 + LINK_SIZE;
+    }
+  }
+
+/* Failed to compile, or error while post-processing */
+
+if (errorcode != 0)
+  {
+  (PUBL(free))(re);
+  PCRE_EARLY_ERROR_RETURN:
+  *erroroffset = (int)(ptr - (const pcre_uchar *)pattern);
+  PCRE_EARLY_ERROR_RETURN2:
+  *errorptr = find_error_text(errorcode);
+  if (errorcodeptr != NULL) *errorcodeptr = errorcode;
+  return NULL;
+  }
+
+/* If the anchored option was not passed, set the flag if we can determine that
+the pattern is anchored by virtue of ^ characters or \A or anything else, such
+as starting with non-atomic .* when DOTALL is set and there are no occurrences
+of *PRUNE or *SKIP.
+
+Otherwise, if we know what the first byte has to be, save it, because that
+speeds up unanchored matches no end. If not, see if we can set the
+PCRE_STARTLINE flag. This is helpful for multiline matches when all branches
+start with ^. and also when all branches start with non-atomic .* for
+non-DOTALL matches when *PRUNE and SKIP are not present. */
+
+if ((re->options & PCRE_ANCHORED) == 0)
+  {
+  if (is_anchored(codestart, 0, cd, 0)) re->options |= PCRE_ANCHORED;
+  else
+    {
+    if (firstcharflags < 0)
+      firstchar = find_firstassertedchar(codestart, &firstcharflags, FALSE);
+    if (firstcharflags >= 0)   /* Remove caseless flag for non-caseable chars */
+      {
+#if defined COMPILE_PCRE8
+      re->first_char = firstchar & 0xff;
+#elif defined COMPILE_PCRE16
+      re->first_char = firstchar & 0xffff;
+#elif defined COMPILE_PCRE32
+      re->first_char = firstchar;
+#endif
+      if ((firstcharflags & REQ_CASELESS) != 0)
+        {
+#if defined SUPPORT_UCP && !(defined COMPILE_PCRE8)
+        /* We ignore non-ASCII first chars in 8 bit mode. */
+        if (utf)
+          {
+          if (re->first_char < 128)
+            {
+            if (cd->fcc[re->first_char] != re->first_char)
+              re->flags |= PCRE_FCH_CASELESS;
+            }
+          else if (UCD_OTHERCASE(re->first_char) != re->first_char)
+            re->flags |= PCRE_FCH_CASELESS;
+          }
+        else
+#endif
+        if (MAX_255(re->first_char)
+            && cd->fcc[re->first_char] != re->first_char)
+          re->flags |= PCRE_FCH_CASELESS;
+        }
+
+      re->flags |= PCRE_FIRSTSET;
+      }
+
+    else if (is_startline(codestart, 0, cd, 0)) re->flags |= PCRE_STARTLINE;
+    }
+  }
+
+/* For an anchored pattern, we use the "required byte" only if it follows a
+variable length item in the regex. Remove the caseless flag for non-caseable
+bytes. */
+
+if (reqcharflags >= 0 &&
+     ((re->options & PCRE_ANCHORED) == 0 || (reqcharflags & REQ_VARY) != 0))
+  {
+#if defined COMPILE_PCRE8
+  re->req_char = reqchar & 0xff;
+#elif defined COMPILE_PCRE16
+  re->req_char = reqchar & 0xffff;
+#elif defined COMPILE_PCRE32
+  re->req_char = reqchar;
+#endif
+  if ((reqcharflags & REQ_CASELESS) != 0)
+    {
+#if defined SUPPORT_UCP && !(defined COMPILE_PCRE8)
+    /* We ignore non-ASCII first chars in 8 bit mode. */
+    if (utf)
+      {
+      if (re->req_char < 128)
+        {
+        if (cd->fcc[re->req_char] != re->req_char)
+          re->flags |= PCRE_RCH_CASELESS;
+        }
+      else if (UCD_OTHERCASE(re->req_char) != re->req_char)
+        re->flags |= PCRE_RCH_CASELESS;
+      }
+    else
+#endif
+    if (MAX_255(re->req_char) && cd->fcc[re->req_char] != re->req_char)
+      re->flags |= PCRE_RCH_CASELESS;
+    }
+
+  re->flags |= PCRE_REQCHSET;
+  }
+
+/* Print out the compiled data if debugging is enabled. This is never the
+case when building a production library. */
+
+#ifdef PCRE_DEBUG
+printf("Length = %d top_bracket = %d top_backref = %d\n",
+  length, re->top_bracket, re->top_backref);
+
+printf("Options=%08x\n", re->options);
+
+if ((re->flags & PCRE_FIRSTSET) != 0)
+  {
+  pcre_uchar ch = re->first_char;
+  const char *caseless =
+    ((re->flags & PCRE_FCH_CASELESS) == 0)? "" : " (caseless)";
+  if (PRINTABLE(ch)) printf("First char = %c%s\n", ch, caseless);
+    else printf("First char = \\x%02x%s\n", ch, caseless);
+  }
+
+if ((re->flags & PCRE_REQCHSET) != 0)
+  {
+  pcre_uchar ch = re->req_char;
+  const char *caseless =
+    ((re->flags & PCRE_RCH_CASELESS) == 0)? "" : " (caseless)";
+  if (PRINTABLE(ch)) printf("Req char = %c%s\n", ch, caseless);
+    else printf("Req char = \\x%02x%s\n", ch, caseless);
+  }
+
+#if defined COMPILE_PCRE8
+pcre_printint((pcre *)re, stdout, TRUE);
+#elif defined COMPILE_PCRE16
+pcre16_printint((pcre *)re, stdout, TRUE);
+#elif defined COMPILE_PCRE32
+pcre32_printint((pcre *)re, stdout, TRUE);
+#endif
+
+/* This check is done here in the debugging case so that the code that
+was compiled can be seen. */
+
+if (code - codestart > length)
+  {
+  (PUBL(free))(re);
+  *errorptr = find_error_text(ERR23);
+  *erroroffset = ptr - (pcre_uchar *)pattern;
+  if (errorcodeptr != NULL) *errorcodeptr = ERR23;
+  return NULL;
+  }
+#endif   /* PCRE_DEBUG */
+
+#if defined COMPILE_PCRE8
+return (pcre *)re;
+#elif defined COMPILE_PCRE16
+return (pcre16 *)re;
+#elif defined COMPILE_PCRE32
+return (pcre32 *)re;
+#endif
+}
+
+/* End of pcre_compile.c */
diff --git a/pcre/pcre_config.c b/pcre/pcre_config.c
new file mode 100644
index 00000000000..3d5689f62c3
--- /dev/null
+++ b/pcre/pcre_config.c
@@ -0,0 +1,186 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+
+/* This module contains the external function pcre_config(). */
+
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+/* Keep the original link size. */
+static int real_link_size = LINK_SIZE;
+
+#include "pcre_internal.h"
+
+
+/*************************************************
+* Return info about what features are configured *
+*************************************************/
+
+/* This function has an extensible interface so that additional items can be
+added compatibly.
+
+Arguments:
+  what             what information is required
+  where            where to put the information
+
+Returns:           0 if data returned, negative on error
+*/
+
+#if defined COMPILE_PCRE8
+PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
+pcre_config(int what, void *where)
+#elif defined COMPILE_PCRE16
+PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
+pcre16_config(int what, void *where)
+#elif defined COMPILE_PCRE32
+PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
+pcre32_config(int what, void *where)
+#endif
+{
+switch (what)
+  {
+  case PCRE_CONFIG_UTF8:
+#if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
+  *((int *)where) = 0;
+  return PCRE_ERROR_BADOPTION;
+#else
+#if defined SUPPORT_UTF
+  *((int *)where) = 1;
+#else
+  *((int *)where) = 0;
+#endif
+  break;
+#endif
+
+  case PCRE_CONFIG_UTF16:
+#if defined COMPILE_PCRE8 || defined COMPILE_PCRE32
+  *((int *)where) = 0;
+  return PCRE_ERROR_BADOPTION;
+#else
+#if defined SUPPORT_UTF
+  *((int *)where) = 1;
+#else
+  *((int *)where) = 0;
+#endif
+  break;
+#endif
+
+  case PCRE_CONFIG_UTF32:
+#if defined COMPILE_PCRE8 || defined COMPILE_PCRE16
+  *((int *)where) = 0;
+  return PCRE_ERROR_BADOPTION;
+#else
+#if defined SUPPORT_UTF
+  *((int *)where) = 1;
+#else
+  *((int *)where) = 0;
+#endif
+  break;
+#endif
+
+  case PCRE_CONFIG_UNICODE_PROPERTIES:
+#ifdef SUPPORT_UCP
+  *((int *)where) = 1;
+#else
+  *((int *)where) = 0;
+#endif
+  break;
+
+  case PCRE_CONFIG_JIT:
+#ifdef SUPPORT_JIT
+  *((int *)where) = 1;
+#else
+  *((int *)where) = 0;
+#endif
+  break;
+
+  case PCRE_CONFIG_JITTARGET:
+#ifdef SUPPORT_JIT
+  *((const char **)where) = PRIV(jit_get_target)();
+#else
+  *((const char **)where) = NULL;
+#endif
+  break;
+
+  case PCRE_CONFIG_NEWLINE:
+  *((int *)where) = NEWLINE;
+  break;
+
+  case PCRE_CONFIG_BSR:
+#ifdef BSR_ANYCRLF
+  *((int *)where) = 1;
+#else
+  *((int *)where) = 0;
+#endif
+  break;
+
+  case PCRE_CONFIG_LINK_SIZE:
+  *((int *)where) = real_link_size;
+  break;
+
+  case PCRE_CONFIG_POSIX_MALLOC_THRESHOLD:
+  *((int *)where) = POSIX_MALLOC_THRESHOLD;
+  break;
+
+  case PCRE_CONFIG_MATCH_LIMIT:
+  *((unsigned long int *)where) = MATCH_LIMIT;
+  break;
+
+  case PCRE_CONFIG_MATCH_LIMIT_RECURSION:
+  *((unsigned long int *)where) = MATCH_LIMIT_RECURSION;
+  break;
+
+  case PCRE_CONFIG_STACKRECURSE:
+#ifdef NO_RECURSE
+  *((int *)where) = 0;
+#else
+  *((int *)where) = 1;
+#endif
+  break;
+
+  default: return PCRE_ERROR_BADOPTION;
+  }
+
+return 0;
+}
+
+/* End of pcre_config.c */
diff --git a/pcre/pcre_dfa_exec.c b/pcre/pcre_dfa_exec.c
new file mode 100644
index 00000000000..02bd3f05534
--- /dev/null
+++ b/pcre/pcre_dfa_exec.c
@@ -0,0 +1,3608 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language (but see
+below for why this module is different).
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2013 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+/* This module contains the external function pcre_dfa_exec(), which is an
+alternative matching function that uses a sort of DFA algorithm (not a true
+FSM). This is NOT Perl-compatible, but it has advantages in certain
+applications. */
+
+
+/* NOTE ABOUT PERFORMANCE: A user of this function sent some code that improved
+the performance of his patterns greatly. I could not use it as it stood, as it
+was not thread safe, and made assumptions about pattern sizes. Also, it caused
+test 7 to loop, and test 9 to crash with a segfault.
+
+The issue is the check for duplicate states, which is done by a simple linear
+search up the state list. (Grep for "duplicate" below to find the code.) For
+many patterns, there will never be many states active at one time, so a simple
+linear search is fine. In patterns that have many active states, it might be a
+bottleneck. The suggested code used an indexing scheme to remember which states
+had previously been used for each character, and avoided the linear search when
+it knew there was no chance of a duplicate. This was implemented when adding
+states to the state lists.
+
+I wrote some thread-safe, not-limited code to try something similar at the time
+of checking for duplicates (instead of when adding states), using index vectors
+on the stack. It did give a 13% improvement with one specially constructed
+pattern for certain subject strings, but on other strings and on many of the
+simpler patterns in the test suite it did worse. The major problem, I think,
+was the extra time to initialize the index. This had to be done for each call
+of internal_dfa_exec(). (The supplied patch used a static vector, initialized
+only once - I suspect this was the cause of the problems with the tests.)
+
+Overall, I concluded that the gains in some cases did not outweigh the losses
+in others, so I abandoned this code. */
+
+
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#define NLBLOCK md             /* Block containing newline information */
+#define PSSTART start_subject  /* Field containing processed string start */
+#define PSEND   end_subject    /* Field containing processed string end */
+
+#include "pcre_internal.h"
+
+
+/* For use to indent debugging output */
+
+#define SP "                   "
+
+
+/*************************************************
+*      Code parameters and static tables         *
+*************************************************/
+
+/* These are offsets that are used to turn the OP_TYPESTAR and friends opcodes
+into others, under special conditions. A gap of 20 between the blocks should be
+enough. The resulting opcodes don't have to be less than 256 because they are
+never stored, so we push them well clear of the normal opcodes. */
+
+#define OP_PROP_EXTRA       300
+#define OP_EXTUNI_EXTRA     320
+#define OP_ANYNL_EXTRA      340
+#define OP_HSPACE_EXTRA     360
+#define OP_VSPACE_EXTRA     380
+
+
+/* This table identifies those opcodes that are followed immediately by a
+character that is to be tested in some way. This makes it possible to
+centralize the loading of these characters. In the case of Type * etc, the
+"character" is the opcode for \D, \d, \S, \s, \W, or \w, which will always be a
+small value. Non-zero values in the table are the offsets from the opcode where
+the character is to be found. ***NOTE*** If the start of this table is
+modified, the three tables that follow must also be modified. */
+
+static const pcre_uint8 coptable[] = {
+  0,                             /* End                                    */
+  0, 0, 0, 0, 0,                 /* \A, \G, \K, \B, \b                     */
+  0, 0, 0, 0, 0, 0,              /* \D, \d, \S, \s, \W, \w                 */
+  0, 0, 0,                       /* Any, AllAny, Anybyte                   */
+  0, 0,                          /* \P, \p                                 */
+  0, 0, 0, 0, 0,                 /* \R, \H, \h, \V, \v                     */
+  0,                             /* \X                                     */
+  0, 0, 0, 0, 0, 0,              /* \Z, \z, ^, ^M, $, $M                   */
+  1,                             /* Char                                   */
+  1,                             /* Chari                                  */
+  1,                             /* not                                    */
+  1,                             /* noti                                   */
+  /* Positive single-char repeats                                          */
+  1, 1, 1, 1, 1, 1,              /* *, *?, +, +?, ?, ??                    */
+  1+IMM2_SIZE, 1+IMM2_SIZE,      /* upto, minupto                          */
+  1+IMM2_SIZE,                   /* exact                                  */
+  1, 1, 1, 1+IMM2_SIZE,          /* *+, ++, ?+, upto+                      */
+  1, 1, 1, 1, 1, 1,              /* *I, *?I, +I, +?I, ?I, ??I              */
+  1+IMM2_SIZE, 1+IMM2_SIZE,      /* upto I, minupto I                      */
+  1+IMM2_SIZE,                   /* exact I                                */
+  1, 1, 1, 1+IMM2_SIZE,          /* *+I, ++I, ?+I, upto+I                  */
+  /* Negative single-char repeats - only for chars < 256                   */
+  1, 1, 1, 1, 1, 1,              /* NOT *, *?, +, +?, ?, ??                */
+  1+IMM2_SIZE, 1+IMM2_SIZE,      /* NOT upto, minupto                      */
+  1+IMM2_SIZE,                   /* NOT exact                              */
+  1, 1, 1, 1+IMM2_SIZE,          /* NOT *+, ++, ?+, upto+                  */
+  1, 1, 1, 1, 1, 1,              /* NOT *I, *?I, +I, +?I, ?I, ??I          */
+  1+IMM2_SIZE, 1+IMM2_SIZE,      /* NOT upto I, minupto I                  */
+  1+IMM2_SIZE,                   /* NOT exact I                            */
+  1, 1, 1, 1+IMM2_SIZE,          /* NOT *+I, ++I, ?+I, upto+I              */
+  /* Positive type repeats                                                 */
+  1, 1, 1, 1, 1, 1,              /* Type *, *?, +, +?, ?, ??               */
+  1+IMM2_SIZE, 1+IMM2_SIZE,      /* Type upto, minupto                     */
+  1+IMM2_SIZE,                   /* Type exact                             */
+  1, 1, 1, 1+IMM2_SIZE,          /* Type *+, ++, ?+, upto+                 */
+  /* Character class & ref repeats                                         */
+  0, 0, 0, 0, 0, 0,              /* *, *?, +, +?, ?, ??                    */
+  0, 0,                          /* CRRANGE, CRMINRANGE                    */
+  0,                             /* CLASS                                  */
+  0,                             /* NCLASS                                 */
+  0,                             /* XCLASS - variable length               */
+  0,                             /* REF                                    */
+  0,                             /* REFI                                   */
+  0,                             /* RECURSE                                */
+  0,                             /* CALLOUT                                */
+  0,                             /* Alt                                    */
+  0,                             /* Ket                                    */
+  0,                             /* KetRmax                                */
+  0,                             /* KetRmin                                */
+  0,                             /* KetRpos                                */
+  0,                             /* Reverse                                */
+  0,                             /* Assert                                 */
+  0,                             /* Assert not                             */
+  0,                             /* Assert behind                          */
+  0,                             /* Assert behind not                      */
+  0, 0,                          /* ONCE, ONCE_NC                          */
+  0, 0, 0, 0, 0,                 /* BRA, BRAPOS, CBRA, CBRAPOS, COND       */
+  0, 0, 0, 0, 0,                 /* SBRA, SBRAPOS, SCBRA, SCBRAPOS, SCOND  */
+  0, 0,                          /* CREF, NCREF                            */
+  0, 0,                          /* RREF, NRREF                            */
+  0,                             /* DEF                                    */
+  0, 0, 0,                       /* BRAZERO, BRAMINZERO, BRAPOSZERO        */
+  0, 0, 0,                       /* MARK, PRUNE, PRUNE_ARG                 */
+  0, 0, 0, 0,                    /* SKIP, SKIP_ARG, THEN, THEN_ARG         */
+  0, 0, 0, 0,                    /* COMMIT, FAIL, ACCEPT, ASSERT_ACCEPT    */
+  0, 0                           /* CLOSE, SKIPZERO  */
+};
+
+/* This table identifies those opcodes that inspect a character. It is used to
+remember the fact that a character could have been inspected when the end of
+the subject is reached. ***NOTE*** If the start of this table is modified, the
+two tables that follow must also be modified. */
+
+static const pcre_uint8 poptable[] = {
+  0,                             /* End                                    */
+  0, 0, 0, 1, 1,                 /* \A, \G, \K, \B, \b                     */
+  1, 1, 1, 1, 1, 1,              /* \D, \d, \S, \s, \W, \w                 */
+  1, 1, 1,                       /* Any, AllAny, Anybyte                   */
+  1, 1,                          /* \P, \p                                 */
+  1, 1, 1, 1, 1,                 /* \R, \H, \h, \V, \v                     */
+  1,                             /* \X                                     */
+  0, 0, 0, 0, 0, 0,              /* \Z, \z, ^, ^M, $, $M                   */
+  1,                             /* Char                                   */
+  1,                             /* Chari                                  */
+  1,                             /* not                                    */
+  1,                             /* noti                                   */
+  /* Positive single-char repeats                                          */
+  1, 1, 1, 1, 1, 1,              /* *, *?, +, +?, ?, ??                    */
+  1, 1, 1,                       /* upto, minupto, exact                   */
+  1, 1, 1, 1,                    /* *+, ++, ?+, upto+                      */
+  1, 1, 1, 1, 1, 1,              /* *I, *?I, +I, +?I, ?I, ??I              */
+  1, 1, 1,                       /* upto I, minupto I, exact I             */
+  1, 1, 1, 1,                    /* *+I, ++I, ?+I, upto+I                  */
+  /* Negative single-char repeats - only for chars < 256                   */
+  1, 1, 1, 1, 1, 1,              /* NOT *, *?, +, +?, ?, ??                */
+  1, 1, 1,                       /* NOT upto, minupto, exact               */
+  1, 1, 1, 1,                    /* NOT *+, ++, ?+, upto+                  */
+  1, 1, 1, 1, 1, 1,              /* NOT *I, *?I, +I, +?I, ?I, ??I          */
+  1, 1, 1,                       /* NOT upto I, minupto I, exact I         */
+  1, 1, 1, 1,                    /* NOT *+I, ++I, ?+I, upto+I              */
+  /* Positive type repeats                                                 */
+  1, 1, 1, 1, 1, 1,              /* Type *, *?, +, +?, ?, ??               */
+  1, 1, 1,                       /* Type upto, minupto, exact              */
+  1, 1, 1, 1,                    /* Type *+, ++, ?+, upto+                 */
+  /* Character class & ref repeats                                         */
+  1, 1, 1, 1, 1, 1,              /* *, *?, +, +?, ?, ??                    */
+  1, 1,                          /* CRRANGE, CRMINRANGE                    */
+  1,                             /* CLASS                                  */
+  1,                             /* NCLASS                                 */
+  1,                             /* XCLASS - variable length               */
+  0,                             /* REF                                    */
+  0,                             /* REFI                                   */
+  0,                             /* RECURSE                                */
+  0,                             /* CALLOUT                                */
+  0,                             /* Alt                                    */
+  0,                             /* Ket                                    */
+  0,                             /* KetRmax                                */
+  0,                             /* KetRmin                                */
+  0,                             /* KetRpos                                */
+  0,                             /* Reverse                                */
+  0,                             /* Assert                                 */
+  0,                             /* Assert not                             */
+  0,                             /* Assert behind                          */
+  0,                             /* Assert behind not                      */
+  0, 0,                          /* ONCE, ONCE_NC                          */
+  0, 0, 0, 0, 0,                 /* BRA, BRAPOS, CBRA, CBRAPOS, COND       */
+  0, 0, 0, 0, 0,                 /* SBRA, SBRAPOS, SCBRA, SCBRAPOS, SCOND  */
+  0, 0,                          /* CREF, NCREF                            */
+  0, 0,                          /* RREF, NRREF                            */
+  0,                             /* DEF                                    */
+  0, 0, 0,                       /* BRAZERO, BRAMINZERO, BRAPOSZERO        */
+  0, 0, 0,                       /* MARK, PRUNE, PRUNE_ARG                 */
+  0, 0, 0, 0,                    /* SKIP, SKIP_ARG, THEN, THEN_ARG         */
+  0, 0, 0, 0,                    /* COMMIT, FAIL, ACCEPT, ASSERT_ACCEPT    */
+  0, 0                           /* CLOSE, SKIPZERO                        */
+};
+
+/* These 2 tables allow for compact code for testing for \D, \d, \S, \s, \W,
+and \w */
+
+static const pcre_uint8 toptable1[] = {
+  0, 0, 0, 0, 0, 0,
+  ctype_digit, ctype_digit,
+  ctype_space, ctype_space,
+  ctype_word,  ctype_word,
+  0, 0                            /* OP_ANY, OP_ALLANY */
+};
+
+static const pcre_uint8 toptable2[] = {
+  0, 0, 0, 0, 0, 0,
+  ctype_digit, 0,
+  ctype_space, 0,
+  ctype_word,  0,
+  1, 1                            /* OP_ANY, OP_ALLANY */
+};
+
+
+/* Structure for holding data about a particular state, which is in effect the
+current data for an active path through the match tree. It must consist
+entirely of ints because the working vector we are passed, and which we put
+these structures in, is a vector of ints. */
+
+typedef struct stateblock {
+  int offset;                     /* Offset to opcode */
+  int count;                      /* Count for repeats */
+  int data;                       /* Some use extra data */
+} stateblock;
+
+#define INTS_PER_STATEBLOCK  (int)(sizeof(stateblock)/sizeof(int))
+
+
+#ifdef PCRE_DEBUG
+/*************************************************
+*             Print character string             *
+*************************************************/
+
+/* Character string printing function for debugging.
+
+Arguments:
+  p            points to string
+  length       number of bytes
+  f            where to print
+
+Returns:       nothing
+*/
+
+static void
+pchars(const pcre_uchar *p, int length, FILE *f)
+{
+pcre_uint32 c;
+while (length-- > 0)
+  {
+  if (isprint(c = *(p++)))
+    fprintf(f, "%c", c);
+  else
+    fprintf(f, "\\x{%02x}", c);
+  }
+}
+#endif
+
+
+
+/*************************************************
+*    Execute a Regular Expression - DFA engine   *
+*************************************************/
+
+/* This internal function applies a compiled pattern to a subject string,
+starting at a given point, using a DFA engine. This function is called from the
+external one, possibly multiple times if the pattern is not anchored. The
+function calls itself recursively for some kinds of subpattern.
+
+Arguments:
+  md                the match_data block with fixed information
+  this_start_code   the opening bracket of this subexpression's code
+  current_subject   where we currently are in the subject string
+  start_offset      start offset in the subject string
+  offsets           vector to contain the matching string offsets
+  offsetcount       size of same
+  workspace         vector of workspace
+  wscount           size of same
+  rlevel            function call recursion level
+
+Returns:            > 0 => number of match offset pairs placed in offsets
+                    = 0 => offsets overflowed; longest matches are present
+                     -1 => failed to match
+                   < -1 => some kind of unexpected problem
+
+The following macros are used for adding states to the two state vectors (one
+for the current character, one for the following character). */
+
+#define ADD_ACTIVE(x,y) \
+  if (active_count++ < wscount) \
+    { \
+    next_active_state->offset = (x); \
+    next_active_state->count  = (y); \
+    next_active_state++; \
+    DPRINTF(("%.*sADD_ACTIVE(%d,%d)\n", rlevel*2-2, SP, (x), (y))); \
+    } \
+  else return PCRE_ERROR_DFA_WSSIZE
+
+#define ADD_ACTIVE_DATA(x,y,z) \
+  if (active_count++ < wscount) \
+    { \
+    next_active_state->offset = (x); \
+    next_active_state->count  = (y); \
+    next_active_state->data   = (z); \
+    next_active_state++; \
+    DPRINTF(("%.*sADD_ACTIVE_DATA(%d,%d,%d)\n", rlevel*2-2, SP, (x), (y), (z))); \
+    } \
+  else return PCRE_ERROR_DFA_WSSIZE
+
+#define ADD_NEW(x,y) \
+  if (new_count++ < wscount) \
+    { \
+    next_new_state->offset = (x); \
+    next_new_state->count  = (y); \
+    next_new_state++; \
+    DPRINTF(("%.*sADD_NEW(%d,%d)\n", rlevel*2-2, SP, (x), (y))); \
+    } \
+  else return PCRE_ERROR_DFA_WSSIZE
+
+#define ADD_NEW_DATA(x,y,z) \
+  if (new_count++ < wscount) \
+    { \
+    next_new_state->offset = (x); \
+    next_new_state->count  = (y); \
+    next_new_state->data   = (z); \
+    next_new_state++; \
+    DPRINTF(("%.*sADD_NEW_DATA(%d,%d,%d) line %d\n", rlevel*2-2, SP, \
+      (x), (y), (z), __LINE__)); \
+    } \
+  else return PCRE_ERROR_DFA_WSSIZE
+
+/* And now, here is the code */
+
+static int
+internal_dfa_exec(
+  dfa_match_data *md,
+  const pcre_uchar *this_start_code,
+  const pcre_uchar *current_subject,
+  int start_offset,
+  int *offsets,
+  int offsetcount,
+  int *workspace,
+  int wscount,
+  int  rlevel)
+{
+stateblock *active_states, *new_states, *temp_states;
+stateblock *next_active_state, *next_new_state;
+
+const pcre_uint8 *ctypes, *lcc, *fcc;
+const pcre_uchar *ptr;
+const pcre_uchar *end_code, *first_op;
+
+dfa_recursion_info new_recursive;
+
+int active_count, new_count, match_count;
+
+/* Some fields in the md block are frequently referenced, so we load them into
+independent variables in the hope that this will perform better. */
+
+const pcre_uchar *start_subject = md->start_subject;
+const pcre_uchar *end_subject = md->end_subject;
+const pcre_uchar *start_code = md->start_code;
+
+#ifdef SUPPORT_UTF
+BOOL utf = (md->poptions & PCRE_UTF8) != 0;
+#else
+BOOL utf = FALSE;
+#endif
+
+BOOL reset_could_continue = FALSE;
+
+rlevel++;
+offsetcount &= (-2);
+
+wscount -= 2;
+wscount = (wscount - (wscount % (INTS_PER_STATEBLOCK * 2))) /
+          (2 * INTS_PER_STATEBLOCK);
+
+DPRINTF(("\n%.*s---------------------\n"
+  "%.*sCall to internal_dfa_exec f=%d\n",
+  rlevel*2-2, SP, rlevel*2-2, SP, rlevel));
+
+ctypes = md->tables + ctypes_offset;
+lcc = md->tables + lcc_offset;
+fcc = md->tables + fcc_offset;
+
+match_count = PCRE_ERROR_NOMATCH;   /* A negative number */
+
+active_states = (stateblock *)(workspace + 2);
+next_new_state = new_states = active_states + wscount;
+new_count = 0;
+
+first_op = this_start_code + 1 + LINK_SIZE +
+  ((*this_start_code == OP_CBRA || *this_start_code == OP_SCBRA ||
+    *this_start_code == OP_CBRAPOS || *this_start_code == OP_SCBRAPOS)
+    ? IMM2_SIZE:0);
+
+/* The first thing in any (sub) pattern is a bracket of some sort. Push all
+the alternative states onto the list, and find out where the end is. This
+makes is possible to use this function recursively, when we want to stop at a
+matching internal ket rather than at the end.
+
+If the first opcode in the first alternative is OP_REVERSE, we are dealing with
+a backward assertion. In that case, we have to find out the maximum amount to
+move back, and set up each alternative appropriately. */
+
+if (*first_op == OP_REVERSE)
+  {
+  int max_back = 0;
+  int gone_back;
+
+  end_code = this_start_code;
+  do
+    {
+    int back = GET(end_code, 2+LINK_SIZE);
+    if (back > max_back) max_back = back;
+    end_code += GET(end_code, 1);
+    }
+  while (*end_code == OP_ALT);
+
+  /* If we can't go back the amount required for the longest lookbehind
+  pattern, go back as far as we can; some alternatives may still be viable. */
+
+#ifdef SUPPORT_UTF
+  /* In character mode we have to step back character by character */
+
+  if (utf)
+    {
+    for (gone_back = 0; gone_back < max_back; gone_back++)
+      {
+      if (current_subject <= start_subject) break;
+      current_subject--;
+      ACROSSCHAR(current_subject > start_subject, *current_subject, current_subject--);
+      }
+    }
+  else
+#endif
+
+  /* In byte-mode we can do this quickly. */
+
+    {
+    gone_back = (current_subject - max_back < start_subject)?
+      (int)(current_subject - start_subject) : max_back;
+    current_subject -= gone_back;
+    }
+
+  /* Save the earliest consulted character */
+
+  if (current_subject < md->start_used_ptr)
+    md->start_used_ptr = current_subject;
+
+  /* Now we can process the individual branches. */
+
+  end_code = this_start_code;
+  do
+    {
+    int back = GET(end_code, 2+LINK_SIZE);
+    if (back <= gone_back)
+      {
+      int bstate = (int)(end_code - start_code + 2 + 2*LINK_SIZE);
+      ADD_NEW_DATA(-bstate, 0, gone_back - back);
+      }
+    end_code += GET(end_code, 1);
+    }
+  while (*end_code == OP_ALT);
+ }
+
+/* This is the code for a "normal" subpattern (not a backward assertion). The
+start of a whole pattern is always one of these. If we are at the top level,
+we may be asked to restart matching from the same point that we reached for a
+previous partial match. We still have to scan through the top-level branches to
+find the end state. */
+
+else
+  {
+  end_code = this_start_code;
+
+  /* Restarting */
+
+  if (rlevel == 1 && (md->moptions & PCRE_DFA_RESTART) != 0)
+    {
+    do { end_code += GET(end_code, 1); } while (*end_code == OP_ALT);
+    new_count = workspace[1];
+    if (!workspace[0])
+      memcpy(new_states, active_states, new_count * sizeof(stateblock));
+    }
+
+  /* Not restarting */
+
+  else
+    {
+    int length = 1 + LINK_SIZE +
+      ((*this_start_code == OP_CBRA || *this_start_code == OP_SCBRA ||
+        *this_start_code == OP_CBRAPOS || *this_start_code == OP_SCBRAPOS)
+        ? IMM2_SIZE:0);
+    do
+      {
+      ADD_NEW((int)(end_code - start_code + length), 0);
+      end_code += GET(end_code, 1);
+      length = 1 + LINK_SIZE;
+      }
+    while (*end_code == OP_ALT);
+    }
+  }
+
+workspace[0] = 0;    /* Bit indicating which vector is current */
+
+DPRINTF(("%.*sEnd state = %d\n", rlevel*2-2, SP, (int)(end_code - start_code)));
+
+/* Loop for scanning the subject */
+
+ptr = current_subject;
+for (;;)
+  {
+  int i, j;
+  int clen, dlen;
+  pcre_uint32 c, d;
+  int forced_fail = 0;
+  BOOL partial_newline = FALSE;
+  BOOL could_continue = reset_could_continue;
+  reset_could_continue = FALSE;
+
+  /* Make the new state list into the active state list and empty the
+  new state list. */
+
+  temp_states = active_states;
+  active_states = new_states;
+  new_states = temp_states;
+  active_count = new_count;
+  new_count = 0;
+
+  workspace[0] ^= 1;              /* Remember for the restarting feature */
+  workspace[1] = active_count;
+
+#ifdef PCRE_DEBUG
+  printf("%.*sNext character: rest of subject = \"", rlevel*2-2, SP);
+  pchars(ptr, STRLEN_UC(ptr), stdout);
+  printf("\"\n");
+
+  printf("%.*sActive states: ", rlevel*2-2, SP);
+  for (i = 0; i < active_count; i++)
+    printf("%d/%d ", active_states[i].offset, active_states[i].count);
+  printf("\n");
+#endif
+
+  /* Set the pointers for adding new states */
+
+  next_active_state = active_states + active_count;
+  next_new_state = new_states;
+
+  /* Load the current character from the subject outside the loop, as many
+  different states may want to look at it, and we assume that at least one
+  will. */
+
+  if (ptr < end_subject)
+    {
+    clen = 1;        /* Number of data items in the character */
+#ifdef SUPPORT_UTF
+    GETCHARLENTEST(c, ptr, clen);
+#else
+    c = *ptr;
+#endif  /* SUPPORT_UTF */
+    }
+  else
+    {
+    clen = 0;        /* This indicates the end of the subject */
+    c = NOTACHAR;    /* This value should never actually be used */
+    }
+
+  /* Scan up the active states and act on each one. The result of an action
+  may be to add more states to the currently active list (e.g. on hitting a
+  parenthesis) or it may be to put states on the new list, for considering
+  when we move the character pointer on. */
+
+  for (i = 0; i < active_count; i++)
+    {
+    stateblock *current_state = active_states + i;
+    BOOL caseless = FALSE;
+    const pcre_uchar *code;
+    int state_offset = current_state->offset;
+    int codevalue, rrc;
+    int count;
+
+#ifdef PCRE_DEBUG
+    printf ("%.*sProcessing state %d c=", rlevel*2-2, SP, state_offset);
+    if (clen == 0) printf("EOL\n");
+      else if (c > 32 && c < 127) printf("'%c'\n", c);
+        else printf("0x%02x\n", c);
+#endif
+
+    /* A negative offset is a special case meaning "hold off going to this
+    (negated) state until the number of characters in the data field have
+    been skipped". If the could_continue flag was passed over from a previous
+    state, arrange for it to passed on. */
+
+    if (state_offset < 0)
+      {
+      if (current_state->data > 0)
+        {
+        DPRINTF(("%.*sSkipping this character\n", rlevel*2-2, SP));
+        ADD_NEW_DATA(state_offset, current_state->count,
+          current_state->data - 1);
+        if (could_continue) reset_could_continue = TRUE;
+        continue;
+        }
+      else
+        {
+        current_state->offset = state_offset = -state_offset;
+        }
+      }
+
+    /* Check for a duplicate state with the same count, and skip if found.
+    See the note at the head of this module about the possibility of improving
+    performance here. */
+
+    for (j = 0; j < i; j++)
+      {
+      if (active_states[j].offset == state_offset &&
+          active_states[j].count == current_state->count)
+        {
+        DPRINTF(("%.*sDuplicate state: skipped\n", rlevel*2-2, SP));
+        goto NEXT_ACTIVE_STATE;
+        }
+      }
+
+    /* The state offset is the offset to the opcode */
+
+    code = start_code + state_offset;
+    codevalue = *code;
+
+    /* If this opcode inspects a character, but we are at the end of the
+    subject, remember the fact for use when testing for a partial match. */
+
+    if (clen == 0 && poptable[codevalue] != 0)
+      could_continue = TRUE;
+
+    /* If this opcode is followed by an inline character, load it. It is
+    tempting to test for the presence of a subject character here, but that
+    is wrong, because sometimes zero repetitions of the subject are
+    permitted.
+
+    We also use this mechanism for opcodes such as OP_TYPEPLUS that take an
+    argument that is not a data character - but is always one byte long because
+    the values are small. We have to take special action to deal with  \P, \p,
+    \H, \h, \V, \v and \X in this case. To keep the other cases fast, convert
+    these ones to new opcodes. */
+
+    if (coptable[codevalue] > 0)
+      {
+      dlen = 1;
+#ifdef SUPPORT_UTF
+      if (utf) { GETCHARLEN(d, (code + coptable[codevalue]), dlen); } else
+#endif  /* SUPPORT_UTF */
+      d = code[coptable[codevalue]];
+      if (codevalue >= OP_TYPESTAR)
+        {
+        switch(d)
+          {
+          case OP_ANYBYTE: return PCRE_ERROR_DFA_UITEM;
+          case OP_NOTPROP:
+          case OP_PROP: codevalue += OP_PROP_EXTRA; break;
+          case OP_ANYNL: codevalue += OP_ANYNL_EXTRA; break;
+          case OP_EXTUNI: codevalue += OP_EXTUNI_EXTRA; break;
+          case OP_NOT_HSPACE:
+          case OP_HSPACE: codevalue += OP_HSPACE_EXTRA; break;
+          case OP_NOT_VSPACE:
+          case OP_VSPACE: codevalue += OP_VSPACE_EXTRA; break;
+          default: break;
+          }
+        }
+      }
+    else
+      {
+      dlen = 0;         /* Not strictly necessary, but compilers moan */
+      d = NOTACHAR;     /* if these variables are not set. */
+      }
+
+
+    /* Now process the individual opcodes */
+
+    switch (codevalue)
+      {
+/* ========================================================================== */
+      /* These cases are never obeyed. This is a fudge that causes a compile-
+      time error if the vectors coptable or poptable, which are indexed by
+      opcode, are not the correct length. It seems to be the only way to do
+      such a check at compile time, as the sizeof() operator does not work
+      in the C preprocessor. */
+
+      case OP_TABLE_LENGTH:
+      case OP_TABLE_LENGTH +
+        ((sizeof(coptable) == OP_TABLE_LENGTH) &&
+         (sizeof(poptable) == OP_TABLE_LENGTH)):
+      break;
+
+/* ========================================================================== */
+      /* Reached a closing bracket. If not at the end of the pattern, carry
+      on with the next opcode. For repeating opcodes, also add the repeat
+      state. Note that KETRPOS will always be encountered at the end of the
+      subpattern, because the possessive subpattern repeats are always handled
+      using recursive calls. Thus, it never adds any new states.
+
+      At the end of the (sub)pattern, unless we have an empty string and
+      PCRE_NOTEMPTY is set, or PCRE_NOTEMPTY_ATSTART is set and we are at the
+      start of the subject, save the match data, shifting up all previous
+      matches so we always have the longest first. */
+
+      case OP_KET:
+      case OP_KETRMIN:
+      case OP_KETRMAX:
+      case OP_KETRPOS:
+      if (code != end_code)
+        {
+        ADD_ACTIVE(state_offset + 1 + LINK_SIZE, 0);
+        if (codevalue != OP_KET)
+          {
+          ADD_ACTIVE(state_offset - GET(code, 1), 0);
+          }
+        }
+      else
+        {
+        if (ptr > current_subject ||
+            ((md->moptions & PCRE_NOTEMPTY) == 0 &&
+              ((md->moptions & PCRE_NOTEMPTY_ATSTART) == 0 ||
+                current_subject > start_subject + md->start_offset)))
+          {
+          if (match_count < 0) match_count = (offsetcount >= 2)? 1 : 0;
+            else if (match_count > 0 && ++match_count * 2 > offsetcount)
+              match_count = 0;
+          count = ((match_count == 0)? offsetcount : match_count * 2) - 2;
+          if (count > 0) memmove(offsets + 2, offsets, count * sizeof(int));
+          if (offsetcount >= 2)
+            {
+            offsets[0] = (int)(current_subject - start_subject);
+            offsets[1] = (int)(ptr - start_subject);
+            DPRINTF(("%.*sSet matched string = \"%.*s\"\n", rlevel*2-2, SP,
+              offsets[1] - offsets[0], (char *)current_subject));
+            }
+          if ((md->moptions & PCRE_DFA_SHORTEST) != 0)
+            {
+            DPRINTF(("%.*sEnd of internal_dfa_exec %d: returning %d\n"
+              "%.*s---------------------\n\n", rlevel*2-2, SP, rlevel,
+              match_count, rlevel*2-2, SP));
+            return match_count;
+            }
+          }
+        }
+      break;
+
+/* ========================================================================== */
+      /* These opcodes add to the current list of states without looking
+      at the current character. */
+
+      /*-----------------------------------------------------------------*/
+      case OP_ALT:
+      do { code += GET(code, 1); } while (*code == OP_ALT);
+      ADD_ACTIVE((int)(code - start_code), 0);
+      break;
+
+      /*-----------------------------------------------------------------*/
+      case OP_BRA:
+      case OP_SBRA:
+      do
+        {
+        ADD_ACTIVE((int)(code - start_code + 1 + LINK_SIZE), 0);
+        code += GET(code, 1);
+        }
+      while (*code == OP_ALT);
+      break;
+
+      /*-----------------------------------------------------------------*/
+      case OP_CBRA:
+      case OP_SCBRA:
+      ADD_ACTIVE((int)(code - start_code + 1 + LINK_SIZE + IMM2_SIZE),  0);
+      code += GET(code, 1);
+      while (*code == OP_ALT)
+        {
+        ADD_ACTIVE((int)(code - start_code + 1 + LINK_SIZE),  0);
+        code += GET(code, 1);
+        }
+      break;
+
+      /*-----------------------------------------------------------------*/
+      case OP_BRAZERO:
+      case OP_BRAMINZERO:
+      ADD_ACTIVE(state_offset + 1, 0);
+      code += 1 + GET(code, 2);
+      while (*code == OP_ALT) code += GET(code, 1);
+      ADD_ACTIVE((int)(code - start_code + 1 + LINK_SIZE), 0);
+      break;
+
+      /*-----------------------------------------------------------------*/
+      case OP_SKIPZERO:
+      code += 1 + GET(code, 2);
+      while (*code == OP_ALT) code += GET(code, 1);
+      ADD_ACTIVE((int)(code - start_code + 1 + LINK_SIZE), 0);
+      break;
+
+      /*-----------------------------------------------------------------*/
+      case OP_CIRC:
+      if (ptr == start_subject && (md->moptions & PCRE_NOTBOL) == 0)
+        { ADD_ACTIVE(state_offset + 1, 0); }
+      break;
+
+      /*-----------------------------------------------------------------*/
+      case OP_CIRCM:
+      if ((ptr == start_subject && (md->moptions & PCRE_NOTBOL) == 0) ||
+          (ptr != end_subject && WAS_NEWLINE(ptr)))
+        { ADD_ACTIVE(state_offset + 1, 0); }
+      break;
+
+      /*-----------------------------------------------------------------*/
+      case OP_EOD:
+      if (ptr >= end_subject)
+        {
+        if ((md->moptions & PCRE_PARTIAL_HARD) != 0)
+          could_continue = TRUE;
+        else { ADD_ACTIVE(state_offset + 1, 0); }
+        }
+      break;
+
+      /*-----------------------------------------------------------------*/
+      case OP_SOD:
+      if (ptr == start_subject) { ADD_ACTIVE(state_offset + 1, 0); }
+      break;
+
+      /*-----------------------------------------------------------------*/
+      case OP_SOM:
+      if (ptr == start_subject + start_offset) { ADD_ACTIVE(state_offset + 1, 0); }
+      break;
+
+
+/* ========================================================================== */
+      /* These opcodes inspect the next subject character, and sometimes
+      the previous one as well, but do not have an argument. The variable
+      clen contains the length of the current character and is zero if we are
+      at the end of the subject. */
+
+      /*-----------------------------------------------------------------*/
+      case OP_ANY:
+      if (clen > 0 && !IS_NEWLINE(ptr))
+        {
+        if (ptr + 1 >= md->end_subject &&
+            (md->moptions & (PCRE_PARTIAL_HARD)) != 0 &&
+            NLBLOCK->nltype == NLTYPE_FIXED &&
+            NLBLOCK->nllen == 2 &&
+            c == NLBLOCK->nl[0])
+          {
+          could_continue = partial_newline = TRUE;
+          }
+        else
+          {
+          ADD_NEW(state_offset + 1, 0);
+          }
+        }
+      break;
+
+      /*-----------------------------------------------------------------*/
+      case OP_ALLANY:
+      if (clen > 0)
+        { ADD_NEW(state_offset + 1, 0); }
+      break;
+
+      /*-----------------------------------------------------------------*/
+      case OP_EODN:
+      if (clen == 0 && (md->moptions & PCRE_PARTIAL_HARD) != 0)
+        could_continue = TRUE;
+      else if (clen == 0 || (IS_NEWLINE(ptr) && ptr == end_subject - md->nllen))
+        { ADD_ACTIVE(state_offset + 1, 0); }
+      break;
+
+      /*-----------------------------------------------------------------*/
+      case OP_DOLL:
+      if ((md->moptions & PCRE_NOTEOL) == 0)
+        {
+        if (clen == 0 && (md->moptions & PCRE_PARTIAL_HARD) != 0)
+          could_continue = TRUE;
+        else if (clen == 0 ||
+            ((md->poptions & PCRE_DOLLAR_ENDONLY) == 0 && IS_NEWLINE(ptr) &&
+               (ptr == end_subject - md->nllen)
+            ))
+          { ADD_ACTIVE(state_offset + 1, 0); }
+        else if (ptr + 1 >= md->end_subject &&
+                 (md->moptions & (PCRE_PARTIAL_HARD|PCRE_PARTIAL_SOFT)) != 0 &&
+                 NLBLOCK->nltype == NLTYPE_FIXED &&
+                 NLBLOCK->nllen == 2 &&
+                 c == NLBLOCK->nl[0])
+          {
+          if ((md->moptions & PCRE_PARTIAL_HARD) != 0)
+            {
+            reset_could_continue = TRUE;
+            ADD_NEW_DATA(-(state_offset + 1), 0, 1);
+            }
+          else could_continue = partial_newline = TRUE;
+          }
+        }
+      break;
+
+      /*-----------------------------------------------------------------*/
+      case OP_DOLLM:
+      if ((md->moptions & PCRE_NOTEOL) == 0)
+        {
+        if (clen == 0 && (md->moptions & PCRE_PARTIAL_HARD) != 0)
+          could_continue = TRUE;
+        else if (clen == 0 ||
+            ((md->poptions & PCRE_DOLLAR_ENDONLY) == 0 && IS_NEWLINE(ptr)))
+          { ADD_ACTIVE(state_offset + 1, 0); }
+        else if (ptr + 1 >= md->end_subject &&
+                 (md->moptions & (PCRE_PARTIAL_HARD|PCRE_PARTIAL_SOFT)) != 0 &&
+                 NLBLOCK->nltype == NLTYPE_FIXED &&
+                 NLBLOCK->nllen == 2 &&
+                 c == NLBLOCK->nl[0])
+          {
+          if ((md->moptions & PCRE_PARTIAL_HARD) != 0)
+            {
+            reset_could_continue = TRUE;
+            ADD_NEW_DATA(-(state_offset + 1), 0, 1);
+            }
+          else could_continue = partial_newline = TRUE;
+          }
+        }
+      else if (IS_NEWLINE(ptr))
+        { ADD_ACTIVE(state_offset + 1, 0); }
+      break;
+
+      /*-----------------------------------------------------------------*/
+
+      case OP_DIGIT:
+      case OP_WHITESPACE:
+      case OP_WORDCHAR:
+      if (clen > 0 && c < 256 &&
+            ((ctypes[c] & toptable1[codevalue]) ^ toptable2[codevalue]) != 0)
+        { ADD_NEW(state_offset + 1, 0); }
+      break;
+
+      /*-----------------------------------------------------------------*/
+      case OP_NOT_DIGIT:
+      case OP_NOT_WHITESPACE:
+      case OP_NOT_WORDCHAR:
+      if (clen > 0 && (c >= 256 ||
+            ((ctypes[c] & toptable1[codevalue]) ^ toptable2[codevalue]) != 0))
+        { ADD_NEW(state_offset + 1, 0); }
+      break;
+
+      /*-----------------------------------------------------------------*/
+      case OP_WORD_BOUNDARY:
+      case OP_NOT_WORD_BOUNDARY:
+        {
+        int left_word, right_word;
+
+        if (ptr > start_subject)
+          {
+          const pcre_uchar *temp = ptr - 1;
+          if (temp < md->start_used_ptr) md->start_used_ptr = temp;
+#if defined SUPPORT_UTF && !defined COMPILE_PCRE32
+          if (utf) { BACKCHAR(temp); }
+#endif
+          GETCHARTEST(d, temp);
+#ifdef SUPPORT_UCP
+          if ((md->poptions & PCRE_UCP) != 0)
+            {
+            if (d == '_') left_word = TRUE; else
+              {
+              int cat = UCD_CATEGORY(d);
+              left_word = (cat == ucp_L || cat == ucp_N);
+              }
+            }
+          else
+#endif
+          left_word = d < 256 && (ctypes[d] & ctype_word) != 0;
+          }
+        else left_word = FALSE;
+
+        if (clen > 0)
+          {
+#ifdef SUPPORT_UCP
+          if ((md->poptions & PCRE_UCP) != 0)
+            {
+            if (c == '_') right_word = TRUE; else
+              {
+              int cat = UCD_CATEGORY(c);
+              right_word = (cat == ucp_L || cat == ucp_N);
+              }
+            }
+          else
+#endif
+          right_word = c < 256 && (ctypes[c] & ctype_word) != 0;
+          }
+        else right_word = FALSE;
+
+        if ((left_word == right_word) == (codevalue == OP_NOT_WORD_BOUNDARY))
+          { ADD_ACTIVE(state_offset + 1, 0); }
+        }
+      break;
+
+
+      /*-----------------------------------------------------------------*/
+      /* Check the next character by Unicode property. We will get here only
+      if the support is in the binary; otherwise a compile-time error occurs.
+      */
+
+#ifdef SUPPORT_UCP
+      case OP_PROP:
+      case OP_NOTPROP:
+      if (clen > 0)
+        {
+        BOOL OK;
+        const pcre_uint32 *cp;
+        const ucd_record * prop = GET_UCD(c);
+        switch(code[1])
+          {
+          case PT_ANY:
+          OK = TRUE;
+          break;
+
+          case PT_LAMP:
+          OK = prop->chartype == ucp_Lu || prop->chartype == ucp_Ll ||
+               prop->chartype == ucp_Lt;
+          break;
+
+          case PT_GC:
+          OK = PRIV(ucp_gentype)[prop->chartype] == code[2];
+          break;
+
+          case PT_PC:
+          OK = prop->chartype == code[2];
+          break;
+
+          case PT_SC:
+          OK = prop->script == code[2];
+          break;
+
+          /* These are specials for combination cases. */
+
+          case PT_ALNUM:
+          OK = PRIV(ucp_gentype)[prop->chartype] == ucp_L ||
+               PRIV(ucp_gentype)[prop->chartype] == ucp_N;
+          break;
+
+          case PT_SPACE:    /* Perl space */
+          OK = PRIV(ucp_gentype)[prop->chartype] == ucp_Z ||
+               c == CHAR_HT || c == CHAR_NL || c == CHAR_FF || c == CHAR_CR;
+          break;
+
+          case PT_PXSPACE:  /* POSIX space */
+          OK = PRIV(ucp_gentype)[prop->chartype] == ucp_Z ||
+               c == CHAR_HT || c == CHAR_NL || c == CHAR_VT ||
+               c == CHAR_FF || c == CHAR_CR;
+          break;
+
+          case PT_WORD:
+          OK = PRIV(ucp_gentype)[prop->chartype] == ucp_L ||
+               PRIV(ucp_gentype)[prop->chartype] == ucp_N ||
+               c == CHAR_UNDERSCORE;
+          break;
+
+          case PT_CLIST:
+          cp = PRIV(ucd_caseless_sets) + code[2];
+          for (;;)
+            {
+            if (c < *cp) { OK = FALSE; break; }
+            if (c == *cp++) { OK = TRUE; break; }
+            }
+          break;
+
+          case PT_UCNC:
+          OK = c == CHAR_DOLLAR_SIGN || c == CHAR_COMMERCIAL_AT ||
+               c == CHAR_GRAVE_ACCENT || (c >= 0xa0 && c <= 0xd7ff) ||
+               c >= 0xe000;
+          break;
+
+          /* Should never occur, but keep compilers from grumbling. */
+
+          default:
+          OK = codevalue != OP_PROP;
+          break;
+          }
+
+        if (OK == (codevalue == OP_PROP)) { ADD_NEW(state_offset + 3, 0); }
+        }
+      break;
+#endif
+
+
+
+/* ========================================================================== */
+      /* These opcodes likewise inspect the subject character, but have an
+      argument that is not a data character. It is one of these opcodes:
+      OP_ANY, OP_ALLANY, OP_DIGIT, OP_NOT_DIGIT, OP_WHITESPACE, OP_NOT_SPACE,
+      OP_WORDCHAR, OP_NOT_WORDCHAR. The value is loaded into d. */
+
+      case OP_TYPEPLUS:
+      case OP_TYPEMINPLUS:
+      case OP_TYPEPOSPLUS:
+      count = current_state->count;  /* Already matched */
+      if (count > 0) { ADD_ACTIVE(state_offset + 2, 0); }
+      if (clen > 0)
+        {
+        if (d == OP_ANY && ptr + 1 >= md->end_subject &&
+            (md->moptions & (PCRE_PARTIAL_HARD)) != 0 &&
+            NLBLOCK->nltype == NLTYPE_FIXED &&
+            NLBLOCK->nllen == 2 &&
+            c == NLBLOCK->nl[0])
+          {
+          could_continue = partial_newline = TRUE;
+          }
+        else if ((c >= 256 && d != OP_DIGIT && d != OP_WHITESPACE && d != OP_WORDCHAR) ||
+            (c < 256 &&
+              (d != OP_ANY || !IS_NEWLINE(ptr)) &&
+              ((ctypes[c] & toptable1[d]) ^ toptable2[d]) != 0))
+          {
+          if (count > 0 && codevalue == OP_TYPEPOSPLUS)
+            {
+            active_count--;            /* Remove non-match possibility */
+            next_active_state--;
+            }
+          count++;
+          ADD_NEW(state_offset, count);
+          }
+        }
+      break;
+
+      /*-----------------------------------------------------------------*/
+      case OP_TYPEQUERY:
+      case OP_TYPEMINQUERY:
+      case OP_TYPEPOSQUERY:
+      ADD_ACTIVE(state_offset + 2, 0);
+      if (clen > 0)
+        {
+        if (d == OP_ANY && ptr + 1 >= md->end_subject &&
+            (md->moptions & (PCRE_PARTIAL_HARD)) != 0 &&
+            NLBLOCK->nltype == NLTYPE_FIXED &&
+            NLBLOCK->nllen == 2 &&
+            c == NLBLOCK->nl[0])
+          {
+          could_continue = partial_newline = TRUE;
+          }
+        else if ((c >= 256 && d != OP_DIGIT && d != OP_WHITESPACE && d != OP_WORDCHAR) ||
+            (c < 256 &&
+              (d != OP_ANY || !IS_NEWLINE(ptr)) &&
+              ((ctypes[c] & toptable1[d]) ^ toptable2[d]) != 0))
+          {
+          if (codevalue == OP_TYPEPOSQUERY)
+            {
+            active_count--;            /* Remove non-match possibility */
+            next_active_state--;
+            }
+          ADD_NEW(state_offset + 2, 0);
+          }
+        }
+      break;
+
+      /*-----------------------------------------------------------------*/
+      case OP_TYPESTAR:
+      case OP_TYPEMINSTAR:
+      case OP_TYPEPOSSTAR:
+      ADD_ACTIVE(state_offset + 2, 0);
+      if (clen > 0)
+        {
+        if (d == OP_ANY && ptr + 1 >= md->end_subject &&
+            (md->moptions & (PCRE_PARTIAL_HARD)) != 0 &&
+            NLBLOCK->nltype == NLTYPE_FIXED &&
+            NLBLOCK->nllen == 2 &&
+            c == NLBLOCK->nl[0])
+          {
+          could_continue = partial_newline = TRUE;
+          }
+        else if ((c >= 256 && d != OP_DIGIT && d != OP_WHITESPACE && d != OP_WORDCHAR) ||
+            (c < 256 &&
+              (d != OP_ANY || !IS_NEWLINE(ptr)) &&
+              ((ctypes[c] & toptable1[d]) ^ toptable2[d]) != 0))
+          {
+          if (codevalue == OP_TYPEPOSSTAR)
+            {
+            active_count--;            /* Remove non-match possibility */
+            next_active_state--;
+            }
+          ADD_NEW(state_offset, 0);
+          }
+        }
+      break;
+
+      /*-----------------------------------------------------------------*/
+      case OP_TYPEEXACT:
+      count = current_state->count;  /* Number already matched */
+      if (clen > 0)
+        {
+        if (d == OP_ANY && ptr + 1 >= md->end_subject &&
+            (md->moptions & (PCRE_PARTIAL_HARD)) != 0 &&
+            NLBLOCK->nltype == NLTYPE_FIXED &&
+            NLBLOCK->nllen == 2 &&
+            c == NLBLOCK->nl[0])
+          {
+          could_continue = partial_newline = TRUE;
+          }
+        else if ((c >= 256 && d != OP_DIGIT && d != OP_WHITESPACE && d != OP_WORDCHAR) ||
+            (c < 256 &&
+              (d != OP_ANY || !IS_NEWLINE(ptr)) &&
+              ((ctypes[c] & toptable1[d]) ^ toptable2[d]) != 0))
+          {
+          if (++count >= (int)GET2(code, 1))
+            { ADD_NEW(state_offset + 1 + IMM2_SIZE + 1, 0); }
+          else
+            { ADD_NEW(state_offset, count); }
+          }
+        }
+      break;
+
+      /*-----------------------------------------------------------------*/
+      case OP_TYPEUPTO:
+      case OP_TYPEMINUPTO:
+      case OP_TYPEPOSUPTO:
+      ADD_ACTIVE(state_offset + 2 + IMM2_SIZE, 0);
+      count = current_state->count;  /* Number already matched */
+      if (clen > 0)
+        {
+        if (d == OP_ANY && ptr + 1 >= md->end_subject &&
+            (md->moptions & (PCRE_PARTIAL_HARD)) != 0 &&
+            NLBLOCK->nltype == NLTYPE_FIXED &&
+            NLBLOCK->nllen == 2 &&
+            c == NLBLOCK->nl[0])
+          {
+          could_continue = partial_newline = TRUE;
+          }
+        else if ((c >= 256 && d != OP_DIGIT && d != OP_WHITESPACE && d != OP_WORDCHAR) ||
+            (c < 256 &&
+              (d != OP_ANY || !IS_NEWLINE(ptr)) &&
+              ((ctypes[c] & toptable1[d]) ^ toptable2[d]) != 0))
+          {
+          if (codevalue == OP_TYPEPOSUPTO)
+            {
+            active_count--;           /* Remove non-match possibility */
+            next_active_state--;
+            }
+          if (++count >= (int)GET2(code, 1))
+            { ADD_NEW(state_offset + 2 + IMM2_SIZE, 0); }
+          else
+            { ADD_NEW(state_offset, count); }
+          }
+        }
+      break;
+
+/* ========================================================================== */
+      /* These are virtual opcodes that are used when something like
+      OP_TYPEPLUS has OP_PROP, OP_NOTPROP, OP_ANYNL, or OP_EXTUNI as its
+      argument. It keeps the code above fast for the other cases. The argument
+      is in the d variable. */
+
+#ifdef SUPPORT_UCP
+      case OP_PROP_EXTRA + OP_TYPEPLUS:
+      case OP_PROP_EXTRA + OP_TYPEMINPLUS:
+      case OP_PROP_EXTRA + OP_TYPEPOSPLUS:
+      count = current_state->count;           /* Already matched */
+      if (count > 0) { ADD_ACTIVE(state_offset + 4, 0); }
+      if (clen > 0)
+        {
+        BOOL OK;
+        const pcre_uint32 *cp;
+        const ucd_record * prop = GET_UCD(c);
+        switch(code[2])
+          {
+          case PT_ANY:
+          OK = TRUE;
+          break;
+
+          case PT_LAMP:
+          OK = prop->chartype == ucp_Lu || prop->chartype == ucp_Ll ||
+            prop->chartype == ucp_Lt;
+          break;
+
+          case PT_GC:
+          OK = PRIV(ucp_gentype)[prop->chartype] == code[3];
+          break;
+
+          case PT_PC:
+          OK = prop->chartype == code[3];
+          break;
+
+          case PT_SC:
+          OK = prop->script == code[3];
+          break;
+
+          /* These are specials for combination cases. */
+
+          case PT_ALNUM:
+          OK = PRIV(ucp_gentype)[prop->chartype] == ucp_L ||
+               PRIV(ucp_gentype)[prop->chartype] == ucp_N;
+          break;
+
+          case PT_SPACE:    /* Perl space */
+          OK = PRIV(ucp_gentype)[prop->chartype] == ucp_Z ||
+               c == CHAR_HT || c == CHAR_NL || c == CHAR_FF || c == CHAR_CR;
+          break;
+
+          case PT_PXSPACE:  /* POSIX space */
+          OK = PRIV(ucp_gentype)[prop->chartype] == ucp_Z ||
+               c == CHAR_HT || c == CHAR_NL || c == CHAR_VT ||
+               c == CHAR_FF || c == CHAR_CR;
+          break;
+
+          case PT_WORD:
+          OK = PRIV(ucp_gentype)[prop->chartype] == ucp_L ||
+               PRIV(ucp_gentype)[prop->chartype] == ucp_N ||
+               c == CHAR_UNDERSCORE;
+          break;
+
+          case PT_CLIST:
+          cp = PRIV(ucd_caseless_sets) + code[3];
+          for (;;)
+            {
+            if (c < *cp) { OK = FALSE; break; }
+            if (c == *cp++) { OK = TRUE; break; }
+            }
+          break;
+
+          case PT_UCNC:
+          OK = c == CHAR_DOLLAR_SIGN || c == CHAR_COMMERCIAL_AT ||
+               c == CHAR_GRAVE_ACCENT || (c >= 0xa0 && c <= 0xd7ff) ||
+               c >= 0xe000;
+          break;
+
+          /* Should never occur, but keep compilers from grumbling. */
+
+          default:
+          OK = codevalue != OP_PROP;
+          break;
+          }
+
+        if (OK == (d == OP_PROP))
+          {
+          if (count > 0 && codevalue == OP_PROP_EXTRA + OP_TYPEPOSPLUS)
+            {
+            active_count--;           /* Remove non-match possibility */
+            next_active_state--;
+            }
+          count++;
+          ADD_NEW(state_offset, count);
+          }
+        }
+      break;
+
+      /*-----------------------------------------------------------------*/
+      case OP_EXTUNI_EXTRA + OP_TYPEPLUS:
+      case OP_EXTUNI_EXTRA + OP_TYPEMINPLUS:
+      case OP_EXTUNI_EXTRA + OP_TYPEPOSPLUS:
+      count = current_state->count;  /* Already matched */
+      if (count > 0) { ADD_ACTIVE(state_offset + 2, 0); }
+      if (clen > 0)
+        {
+        int lgb, rgb;
+        const pcre_uchar *nptr = ptr + clen;
+        int ncount = 0;
+        if (count > 0 && codevalue == OP_EXTUNI_EXTRA + OP_TYPEPOSPLUS)
+          {
+          active_count--;           /* Remove non-match possibility */
+          next_active_state--;
+          }
+        lgb = UCD_GRAPHBREAK(c);
+        while (nptr < end_subject)
+          {
+          dlen = 1;
+          if (!utf) d = *nptr; else { GETCHARLEN(d, nptr, dlen); }
+          rgb = UCD_GRAPHBREAK(d);
+          if ((PRIV(ucp_gbtable)[lgb] & (1 << rgb)) == 0) break;
+          ncount++;
+          lgb = rgb;
+          nptr += dlen;
+          }
+        count++;
+        ADD_NEW_DATA(-state_offset, count, ncount);
+        }
+      break;
+#endif
+
+      /*-----------------------------------------------------------------*/
+      case OP_ANYNL_EXTRA + OP_TYPEPLUS:
+      case OP_ANYNL_EXTRA + OP_TYPEMINPLUS:
+      case OP_ANYNL_EXTRA + OP_TYPEPOSPLUS:
+      count = current_state->count;  /* Already matched */
+      if (count > 0) { ADD_ACTIVE(state_offset + 2, 0); }
+      if (clen > 0)
+        {
+        int ncount = 0;
+        switch (c)
+          {
+          case CHAR_VT:
+          case CHAR_FF:
+          case CHAR_NEL:
+#ifndef EBCDIC
+          case 0x2028:
+          case 0x2029:
+#endif  /* Not EBCDIC */
+          if ((md->moptions & PCRE_BSR_ANYCRLF) != 0) break;
+          goto ANYNL01;
+
+          case CHAR_CR:
+          if (ptr + 1 < end_subject && RAWUCHARTEST(ptr + 1) == CHAR_LF) ncount = 1;
+          /* Fall through */
+
+          ANYNL01:
+          case CHAR_LF:
+          if (count > 0 && codevalue == OP_ANYNL_EXTRA + OP_TYPEPOSPLUS)
+            {
+            active_count--;           /* Remove non-match possibility */
+            next_active_state--;
+            }
+          count++;
+          ADD_NEW_DATA(-state_offset, count, ncount);
+          break;
+
+          default:
+          break;
+          }
+        }
+      break;
+
+      /*-----------------------------------------------------------------*/
+      case OP_VSPACE_EXTRA + OP_TYPEPLUS:
+      case OP_VSPACE_EXTRA + OP_TYPEMINPLUS:
+      case OP_VSPACE_EXTRA + OP_TYPEPOSPLUS:
+      count = current_state->count;  /* Already matched */
+      if (count > 0) { ADD_ACTIVE(state_offset + 2, 0); }
+      if (clen > 0)
+        {
+        BOOL OK;
+        switch (c)
+          {
+          VSPACE_CASES:
+          OK = TRUE;
+          break;
+
+          default:
+          OK = FALSE;
+          break;
+          }
+
+        if (OK == (d == OP_VSPACE))
+          {
+          if (count > 0 && codevalue == OP_VSPACE_EXTRA + OP_TYPEPOSPLUS)
+            {
+            active_count--;           /* Remove non-match possibility */
+            next_active_state--;
+            }
+          count++;
+          ADD_NEW_DATA(-state_offset, count, 0);
+          }
+        }
+      break;
+
+      /*-----------------------------------------------------------------*/
+      case OP_HSPACE_EXTRA + OP_TYPEPLUS:
+      case OP_HSPACE_EXTRA + OP_TYPEMINPLUS:
+      case OP_HSPACE_EXTRA + OP_TYPEPOSPLUS:
+      count = current_state->count;  /* Already matched */
+      if (count > 0) { ADD_ACTIVE(state_offset + 2, 0); }
+      if (clen > 0)
+        {
+        BOOL OK;
+        switch (c)
+          {
+          HSPACE_CASES:
+          OK = TRUE;
+          break;
+
+          default:
+          OK = FALSE;
+          break;
+          }
+
+        if (OK == (d == OP_HSPACE))
+          {
+          if (count > 0 && codevalue == OP_HSPACE_EXTRA + OP_TYPEPOSPLUS)
+            {
+            active_count--;           /* Remove non-match possibility */
+            next_active_state--;
+            }
+          count++;
+          ADD_NEW_DATA(-state_offset, count, 0);
+          }
+        }
+      break;
+
+      /*-----------------------------------------------------------------*/
+#ifdef SUPPORT_UCP
+      case OP_PROP_EXTRA + OP_TYPEQUERY:
+      case OP_PROP_EXTRA + OP_TYPEMINQUERY:
+      case OP_PROP_EXTRA + OP_TYPEPOSQUERY:
+      count = 4;
+      goto QS1;
+
+      case OP_PROP_EXTRA + OP_TYPESTAR:
+      case OP_PROP_EXTRA + OP_TYPEMINSTAR:
+      case OP_PROP_EXTRA + OP_TYPEPOSSTAR:
+      count = 0;
+
+      QS1:
+
+      ADD_ACTIVE(state_offset + 4, 0);
+      if (clen > 0)
+        {
+        BOOL OK;
+        const pcre_uint32 *cp;
+        const ucd_record * prop = GET_UCD(c);
+        switch(code[2])
+          {
+          case PT_ANY:
+          OK = TRUE;
+          break;
+
+          case PT_LAMP:
+          OK = prop->chartype == ucp_Lu || prop->chartype == ucp_Ll ||
+            prop->chartype == ucp_Lt;
+          break;
+
+          case PT_GC:
+          OK = PRIV(ucp_gentype)[prop->chartype] == code[3];
+          break;
+
+          case PT_PC:
+          OK = prop->chartype == code[3];
+          break;
+
+          case PT_SC:
+          OK = prop->script == code[3];
+          break;
+
+          /* These are specials for combination cases. */
+
+          case PT_ALNUM:
+          OK = PRIV(ucp_gentype)[prop->chartype] == ucp_L ||
+               PRIV(ucp_gentype)[prop->chartype] == ucp_N;
+          break;
+
+          case PT_SPACE:    /* Perl space */
+          OK = PRIV(ucp_gentype)[prop->chartype] == ucp_Z ||
+               c == CHAR_HT || c == CHAR_NL || c == CHAR_FF || c == CHAR_CR;
+          break;
+
+          case PT_PXSPACE:  /* POSIX space */
+          OK = PRIV(ucp_gentype)[prop->chartype] == ucp_Z ||
+               c == CHAR_HT || c == CHAR_NL || c == CHAR_VT ||
+               c == CHAR_FF || c == CHAR_CR;
+          break;
+
+          case PT_WORD:
+          OK = PRIV(ucp_gentype)[prop->chartype] == ucp_L ||
+               PRIV(ucp_gentype)[prop->chartype] == ucp_N ||
+               c == CHAR_UNDERSCORE;
+          break;
+
+          case PT_CLIST:
+          cp = PRIV(ucd_caseless_sets) + code[3];
+          for (;;)
+            {
+            if (c < *cp) { OK = FALSE; break; }
+            if (c == *cp++) { OK = TRUE; break; }
+            }
+          break;
+
+          case PT_UCNC:
+          OK = c == CHAR_DOLLAR_SIGN || c == CHAR_COMMERCIAL_AT ||
+               c == CHAR_GRAVE_ACCENT || (c >= 0xa0 && c <= 0xd7ff) ||
+               c >= 0xe000;
+          break;
+
+          /* Should never occur, but keep compilers from grumbling. */
+
+          default:
+          OK = codevalue != OP_PROP;
+          break;
+          }
+
+        if (OK == (d == OP_PROP))
+          {
+          if (codevalue == OP_PROP_EXTRA + OP_TYPEPOSSTAR ||
+              codevalue == OP_PROP_EXTRA + OP_TYPEPOSQUERY)
+            {
+            active_count--;           /* Remove non-match possibility */
+            next_active_state--;
+            }
+          ADD_NEW(state_offset + count, 0);
+          }
+        }
+      break;
+
+      /*-----------------------------------------------------------------*/
+      case OP_EXTUNI_EXTRA + OP_TYPEQUERY:
+      case OP_EXTUNI_EXTRA + OP_TYPEMINQUERY:
+      case OP_EXTUNI_EXTRA + OP_TYPEPOSQUERY:
+      count = 2;
+      goto QS2;
+
+      case OP_EXTUNI_EXTRA + OP_TYPESTAR:
+      case OP_EXTUNI_EXTRA + OP_TYPEMINSTAR:
+      case OP_EXTUNI_EXTRA + OP_TYPEPOSSTAR:
+      count = 0;
+
+      QS2:
+
+      ADD_ACTIVE(state_offset + 2, 0);
+      if (clen > 0)
+        {
+        int lgb, rgb;
+        const pcre_uchar *nptr = ptr + clen;
+        int ncount = 0;
+        if (codevalue == OP_EXTUNI_EXTRA + OP_TYPEPOSSTAR ||
+            codevalue == OP_EXTUNI_EXTRA + OP_TYPEPOSQUERY)
+          {
+          active_count--;           /* Remove non-match possibility */
+          next_active_state--;
+          }
+        lgb = UCD_GRAPHBREAK(c);
+        while (nptr < end_subject)
+          {
+          dlen = 1;
+          if (!utf) d = *nptr; else { GETCHARLEN(d, nptr, dlen); }
+          rgb = UCD_GRAPHBREAK(d);
+          if ((PRIV(ucp_gbtable)[lgb] & (1 << rgb)) == 0) break;
+          ncount++;
+          lgb = rgb;
+          nptr += dlen;
+          }
+        ADD_NEW_DATA(-(state_offset + count), 0, ncount);
+        }
+      break;
+#endif
+
+      /*-----------------------------------------------------------------*/
+      case OP_ANYNL_EXTRA + OP_TYPEQUERY:
+      case OP_ANYNL_EXTRA + OP_TYPEMINQUERY:
+      case OP_ANYNL_EXTRA + OP_TYPEPOSQUERY:
+      count = 2;
+      goto QS3;
+
+      case OP_ANYNL_EXTRA + OP_TYPESTAR:
+      case OP_ANYNL_EXTRA + OP_TYPEMINSTAR:
+      case OP_ANYNL_EXTRA + OP_TYPEPOSSTAR:
+      count = 0;
+
+      QS3:
+      ADD_ACTIVE(state_offset + 2, 0);
+      if (clen > 0)
+        {
+        int ncount = 0;
+        switch (c)
+          {
+          case CHAR_VT:
+          case CHAR_FF:
+          case CHAR_NEL:
+#ifndef EBCDIC
+          case 0x2028:
+          case 0x2029:
+#endif  /* Not EBCDIC */
+          if ((md->moptions & PCRE_BSR_ANYCRLF) != 0) break;
+          goto ANYNL02;
+
+          case CHAR_CR:
+          if (ptr + 1 < end_subject && RAWUCHARTEST(ptr + 1) == CHAR_LF) ncount = 1;
+          /* Fall through */
+
+          ANYNL02:
+          case CHAR_LF:
+          if (codevalue == OP_ANYNL_EXTRA + OP_TYPEPOSSTAR ||
+              codevalue == OP_ANYNL_EXTRA + OP_TYPEPOSQUERY)
+            {
+            active_count--;           /* Remove non-match possibility */
+            next_active_state--;
+            }
+          ADD_NEW_DATA(-(state_offset + (int)count), 0, ncount);
+          break;
+
+          default:
+          break;
+          }
+        }
+      break;
+
+      /*-----------------------------------------------------------------*/
+      case OP_VSPACE_EXTRA + OP_TYPEQUERY:
+      case OP_VSPACE_EXTRA + OP_TYPEMINQUERY:
+      case OP_VSPACE_EXTRA + OP_TYPEPOSQUERY:
+      count = 2;
+      goto QS4;
+
+      case OP_VSPACE_EXTRA + OP_TYPESTAR:
+      case OP_VSPACE_EXTRA + OP_TYPEMINSTAR:
+      case OP_VSPACE_EXTRA + OP_TYPEPOSSTAR:
+      count = 0;
+
+      QS4:
+      ADD_ACTIVE(state_offset + 2, 0);
+      if (clen > 0)
+        {
+        BOOL OK;
+        switch (c)
+          {
+          VSPACE_CASES:
+          OK = TRUE;
+          break;
+
+          default:
+          OK = FALSE;
+          break;
+          }
+        if (OK == (d == OP_VSPACE))
+          {
+          if (codevalue == OP_VSPACE_EXTRA + OP_TYPEPOSSTAR ||
+              codevalue == OP_VSPACE_EXTRA + OP_TYPEPOSQUERY)
+            {
+            active_count--;           /* Remove non-match possibility */
+            next_active_state--;
+            }
+          ADD_NEW_DATA(-(state_offset + (int)count), 0, 0);
+          }
+        }
+      break;
+
+      /*-----------------------------------------------------------------*/
+      case OP_HSPACE_EXTRA + OP_TYPEQUERY:
+      case OP_HSPACE_EXTRA + OP_TYPEMINQUERY:
+      case OP_HSPACE_EXTRA + OP_TYPEPOSQUERY:
+      count = 2;
+      goto QS5;
+
+      case OP_HSPACE_EXTRA + OP_TYPESTAR:
+      case OP_HSPACE_EXTRA + OP_TYPEMINSTAR:
+      case OP_HSPACE_EXTRA + OP_TYPEPOSSTAR:
+      count = 0;
+
+      QS5:
+      ADD_ACTIVE(state_offset + 2, 0);
+      if (clen > 0)
+        {
+        BOOL OK;
+        switch (c)
+          {
+          HSPACE_CASES:
+          OK = TRUE;
+          break;
+
+          default:
+          OK = FALSE;
+          break;
+          }
+
+        if (OK == (d == OP_HSPACE))
+          {
+          if (codevalue == OP_HSPACE_EXTRA + OP_TYPEPOSSTAR ||
+              codevalue == OP_HSPACE_EXTRA + OP_TYPEPOSQUERY)
+            {
+            active_count--;           /* Remove non-match possibility */
+            next_active_state--;
+            }
+          ADD_NEW_DATA(-(state_offset + (int)count), 0, 0);
+          }
+        }
+      break;
+
+      /*-----------------------------------------------------------------*/
+#ifdef SUPPORT_UCP
+      case OP_PROP_EXTRA + OP_TYPEEXACT:
+      case OP_PROP_EXTRA + OP_TYPEUPTO:
+      case OP_PROP_EXTRA + OP_TYPEMINUPTO:
+      case OP_PROP_EXTRA + OP_TYPEPOSUPTO:
+      if (codevalue != OP_PROP_EXTRA + OP_TYPEEXACT)
+        { ADD_ACTIVE(state_offset + 1 + IMM2_SIZE + 3, 0); }
+      count = current_state->count;  /* Number already matched */
+      if (clen > 0)
+        {
+        BOOL OK;
+        const pcre_uint32 *cp;
+        const ucd_record * prop = GET_UCD(c);
+        switch(code[1 + IMM2_SIZE + 1])
+          {
+          case PT_ANY:
+          OK = TRUE;
+          break;
+
+          case PT_LAMP:
+          OK = prop->chartype == ucp_Lu || prop->chartype == ucp_Ll ||
+            prop->chartype == ucp_Lt;
+          break;
+
+          case PT_GC:
+          OK = PRIV(ucp_gentype)[prop->chartype] == code[1 + IMM2_SIZE + 2];
+          break;
+
+          case PT_PC:
+          OK = prop->chartype == code[1 + IMM2_SIZE + 2];
+          break;
+
+          case PT_SC:
+          OK = prop->script == code[1 + IMM2_SIZE + 2];
+          break;
+
+          /* These are specials for combination cases. */
+
+          case PT_ALNUM:
+          OK = PRIV(ucp_gentype)[prop->chartype] == ucp_L ||
+               PRIV(ucp_gentype)[prop->chartype] == ucp_N;
+          break;
+
+          case PT_SPACE:    /* Perl space */
+          OK = PRIV(ucp_gentype)[prop->chartype] == ucp_Z ||
+               c == CHAR_HT || c == CHAR_NL || c == CHAR_FF || c == CHAR_CR;
+          break;
+
+          case PT_PXSPACE:  /* POSIX space */
+          OK = PRIV(ucp_gentype)[prop->chartype] == ucp_Z ||
+               c == CHAR_HT || c == CHAR_NL || c == CHAR_VT ||
+               c == CHAR_FF || c == CHAR_CR;
+          break;
+
+          case PT_WORD:
+          OK = PRIV(ucp_gentype)[prop->chartype] == ucp_L ||
+               PRIV(ucp_gentype)[prop->chartype] == ucp_N ||
+               c == CHAR_UNDERSCORE;
+          break;
+
+          case PT_CLIST:
+          cp = PRIV(ucd_caseless_sets) + code[1 + IMM2_SIZE + 2];
+          for (;;)
+            {
+            if (c < *cp) { OK = FALSE; break; }
+            if (c == *cp++) { OK = TRUE; break; }
+            }
+          break;
+
+          case PT_UCNC:
+          OK = c == CHAR_DOLLAR_SIGN || c == CHAR_COMMERCIAL_AT ||
+               c == CHAR_GRAVE_ACCENT || (c >= 0xa0 && c <= 0xd7ff) ||
+               c >= 0xe000;
+          break;
+
+          /* Should never occur, but keep compilers from grumbling. */
+
+          default:
+          OK = codevalue != OP_PROP;
+          break;
+          }
+
+        if (OK == (d == OP_PROP))
+          {
+          if (codevalue == OP_PROP_EXTRA + OP_TYPEPOSUPTO)
+            {
+            active_count--;           /* Remove non-match possibility */
+            next_active_state--;
+            }
+          if (++count >= (int)GET2(code, 1))
+            { ADD_NEW(state_offset + 1 + IMM2_SIZE + 3, 0); }
+          else
+            { ADD_NEW(state_offset, count); }
+          }
+        }
+      break;
+
+      /*-----------------------------------------------------------------*/
+      case OP_EXTUNI_EXTRA + OP_TYPEEXACT:
+      case OP_EXTUNI_EXTRA + OP_TYPEUPTO:
+      case OP_EXTUNI_EXTRA + OP_TYPEMINUPTO:
+      case OP_EXTUNI_EXTRA + OP_TYPEPOSUPTO:
+      if (codevalue != OP_EXTUNI_EXTRA + OP_TYPEEXACT)
+        { ADD_ACTIVE(state_offset + 2 + IMM2_SIZE, 0); }
+      count = current_state->count;  /* Number already matched */
+      if (clen > 0)
+        {
+        int lgb, rgb;
+        const pcre_uchar *nptr = ptr + clen;
+        int ncount = 0;
+        if (codevalue == OP_EXTUNI_EXTRA + OP_TYPEPOSUPTO)
+          {
+          active_count--;           /* Remove non-match possibility */
+          next_active_state--;
+          }
+        lgb = UCD_GRAPHBREAK(c);
+        while (nptr < end_subject)
+          {
+          dlen = 1;
+          if (!utf) d = *nptr; else { GETCHARLEN(d, nptr, dlen); }
+          rgb = UCD_GRAPHBREAK(d);
+          if ((PRIV(ucp_gbtable)[lgb] & (1 << rgb)) == 0) break;
+          ncount++;
+          lgb = rgb;
+          nptr += dlen;
+          }
+        if (nptr >= end_subject && (md->moptions & PCRE_PARTIAL_HARD) != 0)
+            reset_could_continue = TRUE;
+        if (++count >= (int)GET2(code, 1))
+          { ADD_NEW_DATA(-(state_offset + 2 + IMM2_SIZE), 0, ncount); }
+        else
+          { ADD_NEW_DATA(-state_offset, count, ncount); }
+        }
+      break;
+#endif
+
+      /*-----------------------------------------------------------------*/
+      case OP_ANYNL_EXTRA + OP_TYPEEXACT:
+      case OP_ANYNL_EXTRA + OP_TYPEUPTO:
+      case OP_ANYNL_EXTRA + OP_TYPEMINUPTO:
+      case OP_ANYNL_EXTRA + OP_TYPEPOSUPTO:
+      if (codevalue != OP_ANYNL_EXTRA + OP_TYPEEXACT)
+        { ADD_ACTIVE(state_offset + 2 + IMM2_SIZE, 0); }
+      count = current_state->count;  /* Number already matched */
+      if (clen > 0)
+        {
+        int ncount = 0;
+        switch (c)
+          {
+          case CHAR_VT:
+          case CHAR_FF:
+          case CHAR_NEL:
+#ifndef EBCDIC
+          case 0x2028:
+          case 0x2029:
+#endif  /* Not EBCDIC */
+          if ((md->moptions & PCRE_BSR_ANYCRLF) != 0) break;
+          goto ANYNL03;
+
+          case CHAR_CR:
+          if (ptr + 1 < end_subject && RAWUCHARTEST(ptr + 1) == CHAR_LF) ncount = 1;
+          /* Fall through */
+
+          ANYNL03:
+          case CHAR_LF:
+          if (codevalue == OP_ANYNL_EXTRA + OP_TYPEPOSUPTO)
+            {
+            active_count--;           /* Remove non-match possibility */
+            next_active_state--;
+            }
+          if (++count >= (int)GET2(code, 1))
+            { ADD_NEW_DATA(-(state_offset + 2 + IMM2_SIZE), 0, ncount); }
+          else
+            { ADD_NEW_DATA(-state_offset, count, ncount); }
+          break;
+
+          default:
+          break;
+          }
+        }
+      break;
+
+      /*-----------------------------------------------------------------*/
+      case OP_VSPACE_EXTRA + OP_TYPEEXACT:
+      case OP_VSPACE_EXTRA + OP_TYPEUPTO:
+      case OP_VSPACE_EXTRA + OP_TYPEMINUPTO:
+      case OP_VSPACE_EXTRA + OP_TYPEPOSUPTO:
+      if (codevalue != OP_VSPACE_EXTRA + OP_TYPEEXACT)
+        { ADD_ACTIVE(state_offset + 2 + IMM2_SIZE, 0); }
+      count = current_state->count;  /* Number already matched */
+      if (clen > 0)
+        {
+        BOOL OK;
+        switch (c)
+          {
+          VSPACE_CASES:
+          OK = TRUE;
+          break;
+
+          default:
+          OK = FALSE;
+          }
+
+        if (OK == (d == OP_VSPACE))
+          {
+          if (codevalue == OP_VSPACE_EXTRA + OP_TYPEPOSUPTO)
+            {
+            active_count--;           /* Remove non-match possibility */
+            next_active_state--;
+            }
+          if (++count >= (int)GET2(code, 1))
+            { ADD_NEW_DATA(-(state_offset + 2 + IMM2_SIZE), 0, 0); }
+          else
+            { ADD_NEW_DATA(-state_offset, count, 0); }
+          }
+        }
+      break;
+
+      /*-----------------------------------------------------------------*/
+      case OP_HSPACE_EXTRA + OP_TYPEEXACT:
+      case OP_HSPACE_EXTRA + OP_TYPEUPTO:
+      case OP_HSPACE_EXTRA + OP_TYPEMINUPTO:
+      case OP_HSPACE_EXTRA + OP_TYPEPOSUPTO:
+      if (codevalue != OP_HSPACE_EXTRA + OP_TYPEEXACT)
+        { ADD_ACTIVE(state_offset + 2 + IMM2_SIZE, 0); }
+      count = current_state->count;  /* Number already matched */
+      if (clen > 0)
+        {
+        BOOL OK;
+        switch (c)
+          {
+          HSPACE_CASES:
+          OK = TRUE;
+          break;
+
+          default:
+          OK = FALSE;
+          break;
+          }
+
+        if (OK == (d == OP_HSPACE))
+          {
+          if (codevalue == OP_HSPACE_EXTRA + OP_TYPEPOSUPTO)
+            {
+            active_count--;           /* Remove non-match possibility */
+            next_active_state--;
+            }
+          if (++count >= (int)GET2(code, 1))
+            { ADD_NEW_DATA(-(state_offset + 2 + IMM2_SIZE), 0, 0); }
+          else
+            { ADD_NEW_DATA(-state_offset, count, 0); }
+          }
+        }
+      break;
+
+/* ========================================================================== */
+      /* These opcodes are followed by a character that is usually compared
+      to the current subject character; it is loaded into d. We still get
+      here even if there is no subject character, because in some cases zero
+      repetitions are permitted. */
+
+      /*-----------------------------------------------------------------*/
+      case OP_CHAR:
+      if (clen > 0 && c == d) { ADD_NEW(state_offset + dlen + 1, 0); }
+      break;
+
+      /*-----------------------------------------------------------------*/
+      case OP_CHARI:
+      if (clen == 0) break;
+
+#ifdef SUPPORT_UTF
+      if (utf)
+        {
+        if (c == d) { ADD_NEW(state_offset + dlen + 1, 0); } else
+          {
+          unsigned int othercase;
+          if (c < 128)
+            othercase = fcc[c];
+          else
+            /* If we have Unicode property support, we can use it to test the
+            other case of the character. */
+#ifdef SUPPORT_UCP
+            othercase = UCD_OTHERCASE(c);
+#else
+            othercase = NOTACHAR;
+#endif
+
+          if (d == othercase) { ADD_NEW(state_offset + dlen + 1, 0); }
+          }
+        }
+      else
+#endif  /* SUPPORT_UTF */
+      /* Not UTF mode */
+        {
+        if (TABLE_GET(c, lcc, c) == TABLE_GET(d, lcc, d))
+          { ADD_NEW(state_offset + 2, 0); }
+        }
+      break;
+
+
+#ifdef SUPPORT_UCP
+      /*-----------------------------------------------------------------*/
+      /* This is a tricky one because it can match more than one character.
+      Find out how many characters to skip, and then set up a negative state
+      to wait for them to pass before continuing. */
+
+      case OP_EXTUNI:
+      if (clen > 0)
+        {
+        int lgb, rgb;
+        const pcre_uchar *nptr = ptr + clen;
+        int ncount = 0;
+        lgb = UCD_GRAPHBREAK(c);
+        while (nptr < end_subject)
+          {
+          dlen = 1;
+          if (!utf) d = *nptr; else { GETCHARLEN(d, nptr, dlen); }
+          rgb = UCD_GRAPHBREAK(d);
+          if ((PRIV(ucp_gbtable)[lgb] & (1 << rgb)) == 0) break;
+          ncount++;
+          lgb = rgb;
+          nptr += dlen;
+          }
+        if (nptr >= end_subject && (md->moptions & PCRE_PARTIAL_HARD) != 0)
+            reset_could_continue = TRUE;
+        ADD_NEW_DATA(-(state_offset + 1), 0, ncount);
+        }
+      break;
+#endif
+
+      /*-----------------------------------------------------------------*/
+      /* This is a tricky like EXTUNI because it too can match more than one
+      character (when CR is followed by LF). In this case, set up a negative
+      state to wait for one character to pass before continuing. */
+
+      case OP_ANYNL:
+      if (clen > 0) switch(c)
+        {
+        case CHAR_VT:
+        case CHAR_FF:
+        case CHAR_NEL:
+#ifndef EBCDIC
+        case 0x2028:
+        case 0x2029:
+#endif  /* Not EBCDIC */
+        if ((md->moptions & PCRE_BSR_ANYCRLF) != 0) break;
+
+        case CHAR_LF:
+        ADD_NEW(state_offset + 1, 0);
+        break;
+
+        case CHAR_CR:
+        if (ptr + 1 >= end_subject)
+          {
+          ADD_NEW(state_offset + 1, 0);
+          if ((md->moptions & PCRE_PARTIAL_HARD) != 0)
+            reset_could_continue = TRUE;
+          }
+        else if (RAWUCHARTEST(ptr + 1) == CHAR_LF)
+          {
+          ADD_NEW_DATA(-(state_offset + 1), 0, 1);
+          }
+        else
+          {
+          ADD_NEW(state_offset + 1, 0);
+          }
+        break;
+        }
+      break;
+
+      /*-----------------------------------------------------------------*/
+      case OP_NOT_VSPACE:
+      if (clen > 0) switch(c)
+        {
+        VSPACE_CASES:
+        break;
+
+        default:
+        ADD_NEW(state_offset + 1, 0);
+        break;
+        }
+      break;
+
+      /*-----------------------------------------------------------------*/
+      case OP_VSPACE:
+      if (clen > 0) switch(c)
+        {
+        VSPACE_CASES:
+        ADD_NEW(state_offset + 1, 0);
+        break;
+
+        default:
+        break;
+        }
+      break;
+
+      /*-----------------------------------------------------------------*/
+      case OP_NOT_HSPACE:
+      if (clen > 0) switch(c)
+        {
+        HSPACE_CASES:
+        break;
+
+        default:
+        ADD_NEW(state_offset + 1, 0);
+        break;
+        }
+      break;
+
+      /*-----------------------------------------------------------------*/
+      case OP_HSPACE:
+      if (clen > 0) switch(c)
+        {
+        HSPACE_CASES:
+        ADD_NEW(state_offset + 1, 0);
+        break;
+
+        default:
+        break;
+        }
+      break;
+
+      /*-----------------------------------------------------------------*/
+      /* Match a negated single character casefully. */
+
+      case OP_NOT:
+      if (clen > 0 && c != d) { ADD_NEW(state_offset + dlen + 1, 0); }
+      break;
+
+      /*-----------------------------------------------------------------*/
+      /* Match a negated single character caselessly. */
+
+      case OP_NOTI:
+      if (clen > 0)
+        {
+        unsigned int otherd;
+#ifdef SUPPORT_UTF
+        if (utf && d >= 128)
+          {
+#ifdef SUPPORT_UCP
+          otherd = UCD_OTHERCASE(d);
+#endif  /* SUPPORT_UCP */
+          }
+        else
+#endif  /* SUPPORT_UTF */
+        otherd = TABLE_GET(d, fcc, d);
+        if (c != d && c != otherd)
+          { ADD_NEW(state_offset + dlen + 1, 0); }
+        }
+      break;
+
+      /*-----------------------------------------------------------------*/
+      case OP_PLUSI:
+      case OP_MINPLUSI:
+      case OP_POSPLUSI:
+      case OP_NOTPLUSI:
+      case OP_NOTMINPLUSI:
+      case OP_NOTPOSPLUSI:
+      caseless = TRUE;
+      codevalue -= OP_STARI - OP_STAR;
+
+      /* Fall through */
+      case OP_PLUS:
+      case OP_MINPLUS:
+      case OP_POSPLUS:
+      case OP_NOTPLUS:
+      case OP_NOTMINPLUS:
+      case OP_NOTPOSPLUS:
+      count = current_state->count;  /* Already matched */
+      if (count > 0) { ADD_ACTIVE(state_offset + dlen + 1, 0); }
+      if (clen > 0)
+        {
+        pcre_uint32 otherd = NOTACHAR;
+        if (caseless)
+          {
+#ifdef SUPPORT_UTF
+          if (utf && d >= 128)
+            {
+#ifdef SUPPORT_UCP
+            otherd = UCD_OTHERCASE(d);
+#endif  /* SUPPORT_UCP */
+            }
+          else
+#endif  /* SUPPORT_UTF */
+          otherd = TABLE_GET(d, fcc, d);
+          }
+        if ((c == d || c == otherd) == (codevalue < OP_NOTSTAR))
+          {
+          if (count > 0 &&
+              (codevalue == OP_POSPLUS || codevalue == OP_NOTPOSPLUS))
+            {
+            active_count--;             /* Remove non-match possibility */
+            next_active_state--;
+            }
+          count++;
+          ADD_NEW(state_offset, count);
+          }
+        }
+      break;
+
+      /*-----------------------------------------------------------------*/
+      case OP_QUERYI:
+      case OP_MINQUERYI:
+      case OP_POSQUERYI:
+      case OP_NOTQUERYI:
+      case OP_NOTMINQUERYI:
+      case OP_NOTPOSQUERYI:
+      caseless = TRUE;
+      codevalue -= OP_STARI - OP_STAR;
+      /* Fall through */
+      case OP_QUERY:
+      case OP_MINQUERY:
+      case OP_POSQUERY:
+      case OP_NOTQUERY:
+      case OP_NOTMINQUERY:
+      case OP_NOTPOSQUERY:
+      ADD_ACTIVE(state_offset + dlen + 1, 0);
+      if (clen > 0)
+        {
+        pcre_uint32 otherd = NOTACHAR;
+        if (caseless)
+          {
+#ifdef SUPPORT_UTF
+          if (utf && d >= 128)
+            {
+#ifdef SUPPORT_UCP
+            otherd = UCD_OTHERCASE(d);
+#endif  /* SUPPORT_UCP */
+            }
+          else
+#endif  /* SUPPORT_UTF */
+          otherd = TABLE_GET(d, fcc, d);
+          }
+        if ((c == d || c == otherd) == (codevalue < OP_NOTSTAR))
+          {
+          if (codevalue == OP_POSQUERY || codevalue == OP_NOTPOSQUERY)
+            {
+            active_count--;            /* Remove non-match possibility */
+            next_active_state--;
+            }
+          ADD_NEW(state_offset + dlen + 1, 0);
+          }
+        }
+      break;
+
+      /*-----------------------------------------------------------------*/
+      case OP_STARI:
+      case OP_MINSTARI:
+      case OP_POSSTARI:
+      case OP_NOTSTARI:
+      case OP_NOTMINSTARI:
+      case OP_NOTPOSSTARI:
+      caseless = TRUE;
+      codevalue -= OP_STARI - OP_STAR;
+      /* Fall through */
+      case OP_STAR:
+      case OP_MINSTAR:
+      case OP_POSSTAR:
+      case OP_NOTSTAR:
+      case OP_NOTMINSTAR:
+      case OP_NOTPOSSTAR:
+      ADD_ACTIVE(state_offset + dlen + 1, 0);
+      if (clen > 0)
+        {
+        pcre_uint32 otherd = NOTACHAR;
+        if (caseless)
+          {
+#ifdef SUPPORT_UTF
+          if (utf && d >= 128)
+            {
+#ifdef SUPPORT_UCP
+            otherd = UCD_OTHERCASE(d);
+#endif  /* SUPPORT_UCP */
+            }
+          else
+#endif  /* SUPPORT_UTF */
+          otherd = TABLE_GET(d, fcc, d);
+          }
+        if ((c == d || c == otherd) == (codevalue < OP_NOTSTAR))
+          {
+          if (codevalue == OP_POSSTAR || codevalue == OP_NOTPOSSTAR)
+            {
+            active_count--;            /* Remove non-match possibility */
+            next_active_state--;
+            }
+          ADD_NEW(state_offset, 0);
+          }
+        }
+      break;
+
+      /*-----------------------------------------------------------------*/
+      case OP_EXACTI:
+      case OP_NOTEXACTI:
+      caseless = TRUE;
+      codevalue -= OP_STARI - OP_STAR;
+      /* Fall through */
+      case OP_EXACT:
+      case OP_NOTEXACT:
+      count = current_state->count;  /* Number already matched */
+      if (clen > 0)
+        {
+        pcre_uint32 otherd = NOTACHAR;
+        if (caseless)
+          {
+#ifdef SUPPORT_UTF
+          if (utf && d >= 128)
+            {
+#ifdef SUPPORT_UCP
+            otherd = UCD_OTHERCASE(d);
+#endif  /* SUPPORT_UCP */
+            }
+          else
+#endif  /* SUPPORT_UTF */
+          otherd = TABLE_GET(d, fcc, d);
+          }
+        if ((c == d || c == otherd) == (codevalue < OP_NOTSTAR))
+          {
+          if (++count >= (int)GET2(code, 1))
+            { ADD_NEW(state_offset + dlen + 1 + IMM2_SIZE, 0); }
+          else
+            { ADD_NEW(state_offset, count); }
+          }
+        }
+      break;
+
+      /*-----------------------------------------------------------------*/
+      case OP_UPTOI:
+      case OP_MINUPTOI:
+      case OP_POSUPTOI:
+      case OP_NOTUPTOI:
+      case OP_NOTMINUPTOI:
+      case OP_NOTPOSUPTOI:
+      caseless = TRUE;
+      codevalue -= OP_STARI - OP_STAR;
+      /* Fall through */
+      case OP_UPTO:
+      case OP_MINUPTO:
+      case OP_POSUPTO:
+      case OP_NOTUPTO:
+      case OP_NOTMINUPTO:
+      case OP_NOTPOSUPTO:
+      ADD_ACTIVE(state_offset + dlen + 1 + IMM2_SIZE, 0);
+      count = current_state->count;  /* Number already matched */
+      if (clen > 0)
+        {
+        pcre_uint32 otherd = NOTACHAR;
+        if (caseless)
+          {
+#ifdef SUPPORT_UTF
+          if (utf && d >= 128)
+            {
+#ifdef SUPPORT_UCP
+            otherd = UCD_OTHERCASE(d);
+#endif  /* SUPPORT_UCP */
+            }
+          else
+#endif  /* SUPPORT_UTF */
+          otherd = TABLE_GET(d, fcc, d);
+          }
+        if ((c == d || c == otherd) == (codevalue < OP_NOTSTAR))
+          {
+          if (codevalue == OP_POSUPTO || codevalue == OP_NOTPOSUPTO)
+            {
+            active_count--;             /* Remove non-match possibility */
+            next_active_state--;
+            }
+          if (++count >= (int)GET2(code, 1))
+            { ADD_NEW(state_offset + dlen + 1 + IMM2_SIZE, 0); }
+          else
+            { ADD_NEW(state_offset, count); }
+          }
+        }
+      break;
+
+
+/* ========================================================================== */
+      /* These are the class-handling opcodes */
+
+      case OP_CLASS:
+      case OP_NCLASS:
+      case OP_XCLASS:
+        {
+        BOOL isinclass = FALSE;
+        int next_state_offset;
+        const pcre_uchar *ecode;
+
+        /* For a simple class, there is always just a 32-byte table, and we
+        can set isinclass from it. */
+
+        if (codevalue != OP_XCLASS)
+          {
+          ecode = code + 1 + (32 / sizeof(pcre_uchar));
+          if (clen > 0)
+            {
+            isinclass = (c > 255)? (codevalue == OP_NCLASS) :
+              ((((pcre_uint8 *)(code + 1))[c/8] & (1 << (c&7))) != 0);
+            }
+          }
+
+        /* An extended class may have a table or a list of single characters,
+        ranges, or both, and it may be positive or negative. There's a
+        function that sorts all this out. */
+
+        else
+         {
+         ecode = code + GET(code, 1);
+         if (clen > 0) isinclass = PRIV(xclass)(c, code + 1 + LINK_SIZE, utf);
+         }
+
+        /* At this point, isinclass is set for all kinds of class, and ecode
+        points to the byte after the end of the class. If there is a
+        quantifier, this is where it will be. */
+
+        next_state_offset = (int)(ecode - start_code);
+
+        switch (*ecode)
+          {
+          case OP_CRSTAR:
+          case OP_CRMINSTAR:
+          ADD_ACTIVE(next_state_offset + 1, 0);
+          if (isinclass) { ADD_NEW(state_offset, 0); }
+          break;
+
+          case OP_CRPLUS:
+          case OP_CRMINPLUS:
+          count = current_state->count;  /* Already matched */
+          if (count > 0) { ADD_ACTIVE(next_state_offset + 1, 0); }
+          if (isinclass) { count++; ADD_NEW(state_offset, count); }
+          break;
+
+          case OP_CRQUERY:
+          case OP_CRMINQUERY:
+          ADD_ACTIVE(next_state_offset + 1, 0);
+          if (isinclass) { ADD_NEW(next_state_offset + 1, 0); }
+          break;
+
+          case OP_CRRANGE:
+          case OP_CRMINRANGE:
+          count = current_state->count;  /* Already matched */
+          if (count >= (int)GET2(ecode, 1))
+            { ADD_ACTIVE(next_state_offset + 1 + 2 * IMM2_SIZE, 0); }
+          if (isinclass)
+            {
+            int max = (int)GET2(ecode, 1 + IMM2_SIZE);
+            if (++count >= max && max != 0)   /* Max 0 => no limit */
+              { ADD_NEW(next_state_offset + 1 + 2 * IMM2_SIZE, 0); }
+            else
+              { ADD_NEW(state_offset, count); }
+            }
+          break;
+
+          default:
+          if (isinclass) { ADD_NEW(next_state_offset, 0); }
+          break;
+          }
+        }
+      break;
+
+/* ========================================================================== */
+      /* These are the opcodes for fancy brackets of various kinds. We have
+      to use recursion in order to handle them. The "always failing" assertion
+      (?!) is optimised to OP_FAIL when compiling, so we have to support that,
+      though the other "backtracking verbs" are not supported. */
+
+      case OP_FAIL:
+      forced_fail++;    /* Count FAILs for multiple states */
+      break;
+
+      case OP_ASSERT:
+      case OP_ASSERT_NOT:
+      case OP_ASSERTBACK:
+      case OP_ASSERTBACK_NOT:
+        {
+        int rc;
+        int local_offsets[2];
+        int local_workspace[1000];
+        const pcre_uchar *endasscode = code + GET(code, 1);
+
+        while (*endasscode == OP_ALT) endasscode += GET(endasscode, 1);
+
+        rc = internal_dfa_exec(
+          md,                                   /* static match data */
+          code,                                 /* this subexpression's code */
+          ptr,                                  /* where we currently are */
+          (int)(ptr - start_subject),           /* start offset */
+          local_offsets,                        /* offset vector */
+          sizeof(local_offsets)/sizeof(int),    /* size of same */
+          local_workspace,                      /* workspace vector */
+          sizeof(local_workspace)/sizeof(int),  /* size of same */
+          rlevel);                              /* function recursion level */
+
+        if (rc == PCRE_ERROR_DFA_UITEM) return rc;
+        if ((rc >= 0) == (codevalue == OP_ASSERT || codevalue == OP_ASSERTBACK))
+            { ADD_ACTIVE((int)(endasscode + LINK_SIZE + 1 - start_code), 0); }
+        }
+      break;
+
+      /*-----------------------------------------------------------------*/
+      case OP_COND:
+      case OP_SCOND:
+        {
+        int local_offsets[1000];
+        int local_workspace[1000];
+        int codelink = GET(code, 1);
+        int condcode;
+
+        /* Because of the way auto-callout works during compile, a callout item
+        is inserted between OP_COND and an assertion condition. This does not
+        happen for the other conditions. */
+
+        if (code[LINK_SIZE+1] == OP_CALLOUT)
+          {
+          rrc = 0;
+          if (PUBL(callout) != NULL)
+            {
+            PUBL(callout_block) cb;
+            cb.version          = 1;   /* Version 1 of the callout block */
+            cb.callout_number   = code[LINK_SIZE+2];
+            cb.offset_vector    = offsets;
+#if defined COMPILE_PCRE8
+            cb.subject          = (PCRE_SPTR)start_subject;
+#elif defined COMPILE_PCRE16
+            cb.subject          = (PCRE_SPTR16)start_subject;
+#elif defined COMPILE_PCRE32
+            cb.subject          = (PCRE_SPTR32)start_subject;
+#endif
+            cb.subject_length   = (int)(end_subject - start_subject);
+            cb.start_match      = (int)(current_subject - start_subject);
+            cb.current_position = (int)(ptr - start_subject);
+            cb.pattern_position = GET(code, LINK_SIZE + 3);
+            cb.next_item_length = GET(code, 3 + 2*LINK_SIZE);
+            cb.capture_top      = 1;
+            cb.capture_last     = -1;
+            cb.callout_data     = md->callout_data;
+            cb.mark             = NULL;   /* No (*MARK) support */
+            if ((rrc = (*PUBL(callout))(&cb)) < 0) return rrc;   /* Abandon */
+            }
+          if (rrc > 0) break;                      /* Fail this thread */
+          code += PRIV(OP_lengths)[OP_CALLOUT];    /* Skip callout data */
+          }
+
+        condcode = code[LINK_SIZE+1];
+
+        /* Back reference conditions are not supported */
+
+        if (condcode == OP_CREF || condcode == OP_NCREF)
+          return PCRE_ERROR_DFA_UCOND;
+
+        /* The DEFINE condition is always false */
+
+        if (condcode == OP_DEF)
+          { ADD_ACTIVE(state_offset + codelink + LINK_SIZE + 1, 0); }
+
+        /* The only supported version of OP_RREF is for the value RREF_ANY,
+        which means "test if in any recursion". We can't test for specifically
+        recursed groups. */
+
+        else if (condcode == OP_RREF || condcode == OP_NRREF)
+          {
+          int value = GET2(code, LINK_SIZE + 2);
+          if (value != RREF_ANY) return PCRE_ERROR_DFA_UCOND;
+          if (md->recursive != NULL)
+            { ADD_ACTIVE(state_offset + LINK_SIZE + 2 + IMM2_SIZE, 0); }
+          else { ADD_ACTIVE(state_offset + codelink + LINK_SIZE + 1, 0); }
+          }
+
+        /* Otherwise, the condition is an assertion */
+
+        else
+          {
+          int rc;
+          const pcre_uchar *asscode = code + LINK_SIZE + 1;
+          const pcre_uchar *endasscode = asscode + GET(asscode, 1);
+
+          while (*endasscode == OP_ALT) endasscode += GET(endasscode, 1);
+
+          rc = internal_dfa_exec(
+            md,                                   /* fixed match data */
+            asscode,                              /* this subexpression's code */
+            ptr,                                  /* where we currently are */
+            (int)(ptr - start_subject),           /* start offset */
+            local_offsets,                        /* offset vector */
+            sizeof(local_offsets)/sizeof(int),    /* size of same */
+            local_workspace,                      /* workspace vector */
+            sizeof(local_workspace)/sizeof(int),  /* size of same */
+            rlevel);                              /* function recursion level */
+
+          if (rc == PCRE_ERROR_DFA_UITEM) return rc;
+          if ((rc >= 0) ==
+                (condcode == OP_ASSERT || condcode == OP_ASSERTBACK))
+            { ADD_ACTIVE((int)(endasscode + LINK_SIZE + 1 - start_code), 0); }
+          else
+            { ADD_ACTIVE(state_offset + codelink + LINK_SIZE + 1, 0); }
+          }
+        }
+      break;
+
+      /*-----------------------------------------------------------------*/
+      case OP_RECURSE:
+        {
+        dfa_recursion_info *ri;
+        int local_offsets[1000];
+        int local_workspace[1000];
+        const pcre_uchar *callpat = start_code + GET(code, 1);
+        int recno = (callpat == md->start_code)? 0 :
+          GET2(callpat, 1 + LINK_SIZE);
+        int rc;
+
+        DPRINTF(("%.*sStarting regex recursion\n", rlevel*2-2, SP));
+
+        /* Check for repeating a recursion without advancing the subject
+        pointer. This should catch convoluted mutual recursions. (Some simple
+        cases are caught at compile time.) */
+
+        for (ri = md->recursive; ri != NULL; ri = ri->prevrec)
+          if (recno == ri->group_num && ptr == ri->subject_position)
+            return PCRE_ERROR_RECURSELOOP;
+
+        /* Remember this recursion and where we started it so as to
+        catch infinite loops. */
+
+        new_recursive.group_num = recno;
+        new_recursive.subject_position = ptr;
+        new_recursive.prevrec = md->recursive;
+        md->recursive = &new_recursive;
+
+        rc = internal_dfa_exec(
+          md,                                   /* fixed match data */
+          callpat,                              /* this subexpression's code */
+          ptr,                                  /* where we currently are */
+          (int)(ptr - start_subject),           /* start offset */
+          local_offsets,                        /* offset vector */
+          sizeof(local_offsets)/sizeof(int),    /* size of same */
+          local_workspace,                      /* workspace vector */
+          sizeof(local_workspace)/sizeof(int),  /* size of same */
+          rlevel);                              /* function recursion level */
+
+        md->recursive = new_recursive.prevrec;  /* Done this recursion */
+
+        DPRINTF(("%.*sReturn from regex recursion: rc=%d\n", rlevel*2-2, SP,
+          rc));
+
+        /* Ran out of internal offsets */
+
+        if (rc == 0) return PCRE_ERROR_DFA_RECURSE;
+
+        /* For each successful matched substring, set up the next state with a
+        count of characters to skip before trying it. Note that the count is in
+        characters, not bytes. */
+
+        if (rc > 0)
+          {
+          for (rc = rc*2 - 2; rc >= 0; rc -= 2)
+            {
+            int charcount = local_offsets[rc+1] - local_offsets[rc];
+#if defined SUPPORT_UTF && !defined COMPILE_PCRE32
+            if (utf)
+              {
+              const pcre_uchar *p = start_subject + local_offsets[rc];
+              const pcre_uchar *pp = start_subject + local_offsets[rc+1];
+              while (p < pp) if (NOT_FIRSTCHAR(*p++)) charcount--;
+              }
+#endif
+            if (charcount > 0)
+              {
+              ADD_NEW_DATA(-(state_offset + LINK_SIZE + 1), 0, (charcount - 1));
+              }
+            else
+              {
+              ADD_ACTIVE(state_offset + LINK_SIZE + 1, 0);
+              }
+            }
+          }
+        else if (rc != PCRE_ERROR_NOMATCH) return rc;
+        }
+      break;
+
+      /*-----------------------------------------------------------------*/
+      case OP_BRAPOS:
+      case OP_SBRAPOS:
+      case OP_CBRAPOS:
+      case OP_SCBRAPOS:
+      case OP_BRAPOSZERO:
+        {
+        int charcount, matched_count;
+        const pcre_uchar *local_ptr = ptr;
+        BOOL allow_zero;
+
+        if (codevalue == OP_BRAPOSZERO)
+          {
+          allow_zero = TRUE;
+          codevalue = *(++code);  /* Codevalue will be one of above BRAs */
+          }
+        else allow_zero = FALSE;
+
+        /* Loop to match the subpattern as many times as possible as if it were
+        a complete pattern. */
+
+        for (matched_count = 0;; matched_count++)
+          {
+          int local_offsets[2];
+          int local_workspace[1000];
+
+          int rc = internal_dfa_exec(
+            md,                                   /* fixed match data */
+            code,                                 /* this subexpression's code */
+            local_ptr,                            /* where we currently are */
+            (int)(ptr - start_subject),           /* start offset */
+            local_offsets,                        /* offset vector */
+            sizeof(local_offsets)/sizeof(int),    /* size of same */
+            local_workspace,                      /* workspace vector */
+            sizeof(local_workspace)/sizeof(int),  /* size of same */
+            rlevel);                              /* function recursion level */
+
+          /* Failed to match */
+
+          if (rc < 0)
+            {
+            if (rc != PCRE_ERROR_NOMATCH) return rc;
+            break;
+            }
+
+          /* Matched: break the loop if zero characters matched. */
+
+          charcount = local_offsets[1] - local_offsets[0];
+          if (charcount == 0) break;
+          local_ptr += charcount;    /* Advance temporary position ptr */
+          }
+
+        /* At this point we have matched the subpattern matched_count
+        times, and local_ptr is pointing to the character after the end of the
+        last match. */
+
+        if (matched_count > 0 || allow_zero)
+          {
+          const pcre_uchar *end_subpattern = code;
+          int next_state_offset;
+
+          do { end_subpattern += GET(end_subpattern, 1); }
+            while (*end_subpattern == OP_ALT);
+          next_state_offset =
+            (int)(end_subpattern - start_code + LINK_SIZE + 1);
+
+          /* Optimization: if there are no more active states, and there
+          are no new states yet set up, then skip over the subject string
+          right here, to save looping. Otherwise, set up the new state to swing
+          into action when the end of the matched substring is reached. */
+
+          if (i + 1 >= active_count && new_count == 0)
+            {
+            ptr = local_ptr;
+            clen = 0;
+            ADD_NEW(next_state_offset, 0);
+            }
+          else
+            {
+            const pcre_uchar *p = ptr;
+            const pcre_uchar *pp = local_ptr;
+            charcount = (int)(pp - p);
+#if defined SUPPORT_UTF && !defined COMPILE_PCRE32
+            if (utf) while (p < pp) if (NOT_FIRSTCHAR(*p++)) charcount--;
+#endif
+            ADD_NEW_DATA(-next_state_offset, 0, (charcount - 1));
+            }
+          }
+        }
+      break;
+
+      /*-----------------------------------------------------------------*/
+      case OP_ONCE:
+      case OP_ONCE_NC:
+        {
+        int local_offsets[2];
+        int local_workspace[1000];
+
+        int rc = internal_dfa_exec(
+          md,                                   /* fixed match data */
+          code,                                 /* this subexpression's code */
+          ptr,                                  /* where we currently are */
+          (int)(ptr - start_subject),           /* start offset */
+          local_offsets,                        /* offset vector */
+          sizeof(local_offsets)/sizeof(int),    /* size of same */
+          local_workspace,                      /* workspace vector */
+          sizeof(local_workspace)/sizeof(int),  /* size of same */
+          rlevel);                              /* function recursion level */
+
+        if (rc >= 0)
+          {
+          const pcre_uchar *end_subpattern = code;
+          int charcount = local_offsets[1] - local_offsets[0];
+          int next_state_offset, repeat_state_offset;
+
+          do { end_subpattern += GET(end_subpattern, 1); }
+            while (*end_subpattern == OP_ALT);
+          next_state_offset =
+            (int)(end_subpattern - start_code + LINK_SIZE + 1);
+
+          /* If the end of this subpattern is KETRMAX or KETRMIN, we must
+          arrange for the repeat state also to be added to the relevant list.
+          Calculate the offset, or set -1 for no repeat. */
+
+          repeat_state_offset = (*end_subpattern == OP_KETRMAX ||
+                                 *end_subpattern == OP_KETRMIN)?
+            (int)(end_subpattern - start_code - GET(end_subpattern, 1)) : -1;
+
+          /* If we have matched an empty string, add the next state at the
+          current character pointer. This is important so that the duplicate
+          checking kicks in, which is what breaks infinite loops that match an
+          empty string. */
+
+          if (charcount == 0)
+            {
+            ADD_ACTIVE(next_state_offset, 0);
+            }
+
+          /* Optimization: if there are no more active states, and there
+          are no new states yet set up, then skip over the subject string
+          right here, to save looping. Otherwise, set up the new state to swing
+          into action when the end of the matched substring is reached. */
+
+          else if (i + 1 >= active_count && new_count == 0)
+            {
+            ptr += charcount;
+            clen = 0;
+            ADD_NEW(next_state_offset, 0);
+
+            /* If we are adding a repeat state at the new character position,
+            we must fudge things so that it is the only current state.
+            Otherwise, it might be a duplicate of one we processed before, and
+            that would cause it to be skipped. */
+
+            if (repeat_state_offset >= 0)
+              {
+              next_active_state = active_states;
+              active_count = 0;
+              i = -1;
+              ADD_ACTIVE(repeat_state_offset, 0);
+              }
+            }
+          else
+            {
+#if defined SUPPORT_UTF && !defined COMPILE_PCRE32
+            if (utf)
+              {
+              const pcre_uchar *p = start_subject + local_offsets[0];
+              const pcre_uchar *pp = start_subject + local_offsets[1];
+              while (p < pp) if (NOT_FIRSTCHAR(*p++)) charcount--;
+              }
+#endif
+            ADD_NEW_DATA(-next_state_offset, 0, (charcount - 1));
+            if (repeat_state_offset >= 0)
+              { ADD_NEW_DATA(-repeat_state_offset, 0, (charcount - 1)); }
+            }
+          }
+        else if (rc != PCRE_ERROR_NOMATCH) return rc;
+        }
+      break;
+
+
+/* ========================================================================== */
+      /* Handle callouts */
+
+      case OP_CALLOUT:
+      rrc = 0;
+      if (PUBL(callout) != NULL)
+        {
+        PUBL(callout_block) cb;
+        cb.version          = 1;   /* Version 1 of the callout block */
+        cb.callout_number   = code[1];
+        cb.offset_vector    = offsets;
+#if defined COMPILE_PCRE8
+        cb.subject          = (PCRE_SPTR)start_subject;
+#elif defined COMPILE_PCRE16
+        cb.subject          = (PCRE_SPTR16)start_subject;
+#elif defined COMPILE_PCRE32
+        cb.subject          = (PCRE_SPTR32)start_subject;
+#endif
+        cb.subject_length   = (int)(end_subject - start_subject);
+        cb.start_match      = (int)(current_subject - start_subject);
+        cb.current_position = (int)(ptr - start_subject);
+        cb.pattern_position = GET(code, 2);
+        cb.next_item_length = GET(code, 2 + LINK_SIZE);
+        cb.capture_top      = 1;
+        cb.capture_last     = -1;
+        cb.callout_data     = md->callout_data;
+        cb.mark             = NULL;   /* No (*MARK) support */
+        if ((rrc = (*PUBL(callout))(&cb)) < 0) return rrc;   /* Abandon */
+        }
+      if (rrc == 0)
+        { ADD_ACTIVE(state_offset + PRIV(OP_lengths)[OP_CALLOUT], 0); }
+      break;
+
+
+/* ========================================================================== */
+      default:        /* Unsupported opcode */
+      return PCRE_ERROR_DFA_UITEM;
+      }
+
+    NEXT_ACTIVE_STATE: continue;
+
+    }      /* End of loop scanning active states */
+
+  /* We have finished the processing at the current subject character. If no
+  new states have been set for the next character, we have found all the
+  matches that we are going to find. If we are at the top level and partial
+  matching has been requested, check for appropriate conditions.
+
+  The "forced_ fail" variable counts the number of (*F) encountered for the
+  character. If it is equal to the original active_count (saved in
+  workspace[1]) it means that (*F) was found on every active state. In this
+  case we don't want to give a partial match.
+
+  The "could_continue" variable is true if a state could have continued but
+  for the fact that the end of the subject was reached. */
+
+  if (new_count <= 0)
+    {
+    if (rlevel == 1 &&                               /* Top level, and */
+        could_continue &&                            /* Some could go on, and */
+        forced_fail != workspace[1] &&               /* Not all forced fail & */
+        (                                            /* either... */
+        (md->moptions & PCRE_PARTIAL_HARD) != 0      /* Hard partial */
+        ||                                           /* or... */
+        ((md->moptions & PCRE_PARTIAL_SOFT) != 0 &&  /* Soft partial and */
+         match_count < 0)                            /* no matches */
+        ) &&                                         /* And... */
+        (
+        partial_newline ||                           /* Either partial NL */
+          (                                          /* or ... */
+          ptr >= end_subject &&                /* End of subject and */
+          ptr > md->start_used_ptr)            /* Inspected non-empty string */
+          )
+        )
+      match_count = PCRE_ERROR_PARTIAL;
+    DPRINTF(("%.*sEnd of internal_dfa_exec %d: returning %d\n"
+      "%.*s---------------------\n\n", rlevel*2-2, SP, rlevel, match_count,
+      rlevel*2-2, SP));
+    break;        /* In effect, "return", but see the comment below */
+    }
+
+  /* One or more states are active for the next character. */
+
+  ptr += clen;    /* Advance to next subject character */
+  }               /* Loop to move along the subject string */
+
+/* Control gets here from "break" a few lines above. We do it this way because
+if we use "return" above, we have compiler trouble. Some compilers warn if
+there's nothing here because they think the function doesn't return a value. On
+the other hand, if we put a dummy statement here, some more clever compilers
+complain that it can't be reached. Sigh. */
+
+return match_count;
+}
+
+
+
+
+/*************************************************
+*    Execute a Regular Expression - DFA engine   *
+*************************************************/
+
+/* This external function applies a compiled re to a subject string using a DFA
+engine. This function calls the internal function multiple times if the pattern
+is not anchored.
+
+Arguments:
+  argument_re     points to the compiled expression
+  extra_data      points to extra data or is NULL
+  subject         points to the subject string
+  length          length of subject string (may contain binary zeros)
+  start_offset    where to start in the subject string
+  options         option bits
+  offsets         vector of match offsets
+  offsetcount     size of same
+  workspace       workspace vector
+  wscount         size of same
+
+Returns:          > 0 => number of match offset pairs placed in offsets
+                  = 0 => offsets overflowed; longest matches are present
+                   -1 => failed to match
+                 < -1 => some kind of unexpected problem
+*/
+
+#if defined COMPILE_PCRE8
+PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
+pcre_dfa_exec(const pcre *argument_re, const pcre_extra *extra_data,
+  const char *subject, int length, int start_offset, int options, int *offsets,
+  int offsetcount, int *workspace, int wscount)
+#elif defined COMPILE_PCRE16
+PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
+pcre16_dfa_exec(const pcre16 *argument_re, const pcre16_extra *extra_data,
+  PCRE_SPTR16 subject, int length, int start_offset, int options, int *offsets,
+  int offsetcount, int *workspace, int wscount)
+#elif defined COMPILE_PCRE32
+PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
+pcre32_dfa_exec(const pcre32 *argument_re, const pcre32_extra *extra_data,
+  PCRE_SPTR32 subject, int length, int start_offset, int options, int *offsets,
+  int offsetcount, int *workspace, int wscount)
+#endif
+{
+REAL_PCRE *re = (REAL_PCRE *)argument_re;
+dfa_match_data match_block;
+dfa_match_data *md = &match_block;
+BOOL utf, anchored, startline, firstline;
+const pcre_uchar *current_subject, *end_subject;
+const pcre_study_data *study = NULL;
+
+const pcre_uchar *req_char_ptr;
+const pcre_uint8 *start_bits = NULL;
+BOOL has_first_char = FALSE;
+BOOL has_req_char = FALSE;
+pcre_uchar first_char = 0;
+pcre_uchar first_char2 = 0;
+pcre_uchar req_char = 0;
+pcre_uchar req_char2 = 0;
+int newline;
+
+/* Plausibility checks */
+
+if ((options & ~PUBLIC_DFA_EXEC_OPTIONS) != 0) return PCRE_ERROR_BADOPTION;
+if (re == NULL || subject == NULL || workspace == NULL ||
+   (offsets == NULL && offsetcount > 0)) return PCRE_ERROR_NULL;
+if (offsetcount < 0) return PCRE_ERROR_BADCOUNT;
+if (wscount < 20) return PCRE_ERROR_DFA_WSSIZE;
+if (length < 0) return PCRE_ERROR_BADLENGTH;
+if (start_offset < 0 || start_offset > length) return PCRE_ERROR_BADOFFSET;
+
+/* Check that the first field in the block is the magic number. If it is not,
+return with PCRE_ERROR_BADMAGIC. However, if the magic number is equal to
+REVERSED_MAGIC_NUMBER we return with PCRE_ERROR_BADENDIANNESS, which
+means that the pattern is likely compiled with different endianness. */
+
+if (re->magic_number != MAGIC_NUMBER)
+  return re->magic_number == REVERSED_MAGIC_NUMBER?
+    PCRE_ERROR_BADENDIANNESS:PCRE_ERROR_BADMAGIC;
+if ((re->flags & PCRE_MODE) == 0) return PCRE_ERROR_BADMODE;
+
+/* If restarting after a partial match, do some sanity checks on the contents
+of the workspace. */
+
+if ((options & PCRE_DFA_RESTART) != 0)
+  {
+  if ((workspace[0] & (-2)) != 0 || workspace[1] < 1 ||
+    workspace[1] > (wscount - 2)/INTS_PER_STATEBLOCK)
+      return PCRE_ERROR_DFA_BADRESTART;
+  }
+
+/* Set up study, callout, and table data */
+
+md->tables = re->tables;
+md->callout_data = NULL;
+
+if (extra_data != NULL)
+  {
+  unsigned int flags = extra_data->flags;
+  if ((flags & PCRE_EXTRA_STUDY_DATA) != 0)
+    study = (const pcre_study_data *)extra_data->study_data;
+  if ((flags & PCRE_EXTRA_MATCH_LIMIT) != 0) return PCRE_ERROR_DFA_UMLIMIT;
+  if ((flags & PCRE_EXTRA_MATCH_LIMIT_RECURSION) != 0)
+    return PCRE_ERROR_DFA_UMLIMIT;
+  if ((flags & PCRE_EXTRA_CALLOUT_DATA) != 0)
+    md->callout_data = extra_data->callout_data;
+  if ((flags & PCRE_EXTRA_TABLES) != 0)
+    md->tables = extra_data->tables;
+  }
+
+/* Set some local values */
+
+current_subject = (const pcre_uchar *)subject + start_offset;
+end_subject = (const pcre_uchar *)subject + length;
+req_char_ptr = current_subject - 1;
+
+#ifdef SUPPORT_UTF
+/* PCRE_UTF(16|32) have the same value as PCRE_UTF8. */
+utf = (re->options & PCRE_UTF8) != 0;
+#else
+utf = FALSE;
+#endif
+
+anchored = (options & (PCRE_ANCHORED|PCRE_DFA_RESTART)) != 0 ||
+  (re->options & PCRE_ANCHORED) != 0;
+
+/* The remaining fixed data for passing around. */
+
+md->start_code = (const pcre_uchar *)argument_re +
+    re->name_table_offset + re->name_count * re->name_entry_size;
+md->start_subject = (const pcre_uchar *)subject;
+md->end_subject = end_subject;
+md->start_offset = start_offset;
+md->moptions = options;
+md->poptions = re->options;
+
+/* If the BSR option is not set at match time, copy what was set
+at compile time. */
+
+if ((md->moptions & (PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE)) == 0)
+  {
+  if ((re->options & (PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE)) != 0)
+    md->moptions |= re->options & (PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE);
+#ifdef BSR_ANYCRLF
+  else md->moptions |= PCRE_BSR_ANYCRLF;
+#endif
+  }
+
+/* Handle different types of newline. The three bits give eight cases. If
+nothing is set at run time, whatever was used at compile time applies. */
+
+switch ((((options & PCRE_NEWLINE_BITS) == 0)? re->options : (pcre_uint32)options) &
+         PCRE_NEWLINE_BITS)
+  {
+  case 0: newline = NEWLINE; break;   /* Compile-time default */
+  case PCRE_NEWLINE_CR: newline = CHAR_CR; break;
+  case PCRE_NEWLINE_LF: newline = CHAR_NL; break;
+  case PCRE_NEWLINE_CR+
+       PCRE_NEWLINE_LF: newline = (CHAR_CR << 8) | CHAR_NL; break;
+  case PCRE_NEWLINE_ANY: newline = -1; break;
+  case PCRE_NEWLINE_ANYCRLF: newline = -2; break;
+  default: return PCRE_ERROR_BADNEWLINE;
+  }
+
+if (newline == -2)
+  {
+  md->nltype = NLTYPE_ANYCRLF;
+  }
+else if (newline < 0)
+  {
+  md->nltype = NLTYPE_ANY;
+  }
+else
+  {
+  md->nltype = NLTYPE_FIXED;
+  if (newline > 255)
+    {
+    md->nllen = 2;
+    md->nl[0] = (newline >> 8) & 255;
+    md->nl[1] = newline & 255;
+    }
+  else
+    {
+    md->nllen = 1;
+    md->nl[0] = newline;
+    }
+  }
+
+/* Check a UTF-8 string if required. Unfortunately there's no way of passing
+back the character offset. */
+
+#ifdef SUPPORT_UTF
+if (utf && (options & PCRE_NO_UTF8_CHECK) == 0)
+  {
+  int erroroffset;
+  int errorcode = PRIV(valid_utf)((pcre_uchar *)subject, length, &erroroffset);
+  if (errorcode != 0)
+    {
+    if (offsetcount >= 2)
+      {
+      offsets[0] = erroroffset;
+      offsets[1] = errorcode;
+      }
+#if defined COMPILE_PCRE8
+    return (errorcode <= PCRE_UTF8_ERR5 && (options & PCRE_PARTIAL_HARD) != 0) ?
+      PCRE_ERROR_SHORTUTF8 : PCRE_ERROR_BADUTF8;
+#elif defined COMPILE_PCRE16
+    return (errorcode <= PCRE_UTF16_ERR1 && (options & PCRE_PARTIAL_HARD) != 0) ?
+      PCRE_ERROR_SHORTUTF16 : PCRE_ERROR_BADUTF16;
+#elif defined COMPILE_PCRE32
+    return PCRE_ERROR_BADUTF32;
+#endif
+    }
+#if defined COMPILE_PCRE8 || defined COMPILE_PCRE16
+  if (start_offset > 0 && start_offset < length &&
+        NOT_FIRSTCHAR(((PCRE_PUCHAR)subject)[start_offset]))
+    return PCRE_ERROR_BADUTF8_OFFSET;
+#endif
+  }
+#endif
+
+/* If the exec call supplied NULL for tables, use the inbuilt ones. This
+is a feature that makes it possible to save compiled regex and re-use them
+in other programs later. */
+
+if (md->tables == NULL) md->tables = PRIV(default_tables);
+
+/* The "must be at the start of a line" flags are used in a loop when finding
+where to start. */
+
+startline = (re->flags & PCRE_STARTLINE) != 0;
+firstline = (re->options & PCRE_FIRSTLINE) != 0;
+
+/* Set up the first character to match, if available. The first_byte value is
+never set for an anchored regular expression, but the anchoring may be forced
+at run time, so we have to test for anchoring. The first char may be unset for
+an unanchored pattern, of course. If there's no first char and the pattern was
+studied, there may be a bitmap of possible first characters. */
+
+if (!anchored)
+  {
+  if ((re->flags & PCRE_FIRSTSET) != 0)
+    {
+    has_first_char = TRUE;
+    first_char = first_char2 = (pcre_uchar)(re->first_char);
+    if ((re->flags & PCRE_FCH_CASELESS) != 0)
+      {
+      first_char2 = TABLE_GET(first_char, md->tables + fcc_offset, first_char);
+#if defined SUPPORT_UCP && !(defined COMPILE_PCRE8)
+      if (utf && first_char > 127)
+        first_char2 = UCD_OTHERCASE(first_char);
+#endif
+      }
+    }
+  else
+    {
+    if (!startline && study != NULL &&
+         (study->flags & PCRE_STUDY_MAPPED) != 0)
+      start_bits = study->start_bits;
+    }
+  }
+
+/* For anchored or unanchored matches, there may be a "last known required
+character" set. */
+
+if ((re->flags & PCRE_REQCHSET) != 0)
+  {
+  has_req_char = TRUE;
+  req_char = req_char2 = (pcre_uchar)(re->req_char);
+  if ((re->flags & PCRE_RCH_CASELESS) != 0)
+    {
+    req_char2 = TABLE_GET(req_char, md->tables + fcc_offset, req_char);
+#if defined SUPPORT_UCP && !(defined COMPILE_PCRE8)
+    if (utf && req_char > 127)
+      req_char2 = UCD_OTHERCASE(req_char);
+#endif
+    }
+  }
+
+/* Call the main matching function, looping for a non-anchored regex after a
+failed match. If not restarting, perform certain optimizations at the start of
+a match. */
+
+for (;;)
+  {
+  int rc;
+
+  if ((options & PCRE_DFA_RESTART) == 0)
+    {
+    const pcre_uchar *save_end_subject = end_subject;
+
+    /* If firstline is TRUE, the start of the match is constrained to the first
+    line of a multiline string. Implement this by temporarily adjusting
+    end_subject so that we stop scanning at a newline. If the match fails at
+    the newline, later code breaks this loop. */
+
+    if (firstline)
+      {
+      PCRE_PUCHAR t = current_subject;
+#ifdef SUPPORT_UTF
+      if (utf)
+        {
+        while (t < md->end_subject && !IS_NEWLINE(t))
+          {
+          t++;
+          ACROSSCHAR(t < end_subject, *t, t++);
+          }
+        }
+      else
+#endif
+      while (t < md->end_subject && !IS_NEWLINE(t)) t++;
+      end_subject = t;
+      }
+
+    /* There are some optimizations that avoid running the match if a known
+    starting point is not found. However, there is an option that disables
+    these, for testing and for ensuring that all callouts do actually occur.
+    The option can be set in the regex by (*NO_START_OPT) or passed in
+    match-time options. */
+
+    if (((options | re->options) & PCRE_NO_START_OPTIMIZE) == 0)
+      {
+      /* Advance to a known first char. */
+
+      if (has_first_char)
+        {
+        if (first_char != first_char2)
+          {
+          pcre_uchar csc;
+          while (current_subject < end_subject &&
+                 (csc = RAWUCHARTEST(current_subject)) != first_char && csc != first_char2)
+            current_subject++;
+          }
+        else
+          while (current_subject < end_subject &&
+                 RAWUCHARTEST(current_subject) != first_char)
+            current_subject++;
+        }
+
+      /* Or to just after a linebreak for a multiline match if possible */
+
+      else if (startline)
+        {
+        if (current_subject > md->start_subject + start_offset)
+          {
+#ifdef SUPPORT_UTF
+          if (utf)
+            {
+            while (current_subject < end_subject &&
+                   !WAS_NEWLINE(current_subject))
+              {
+              current_subject++;
+              ACROSSCHAR(current_subject < end_subject, *current_subject,
+                current_subject++);
+              }
+            }
+          else
+#endif
+          while (current_subject < end_subject && !WAS_NEWLINE(current_subject))
+            current_subject++;
+
+          /* If we have just passed a CR and the newline option is ANY or
+          ANYCRLF, and we are now at a LF, advance the match position by one
+          more character. */
+
+          if (RAWUCHARTEST(current_subject - 1) == CHAR_CR &&
+               (md->nltype == NLTYPE_ANY || md->nltype == NLTYPE_ANYCRLF) &&
+               current_subject < end_subject &&
+               RAWUCHARTEST(current_subject) == CHAR_NL)
+            current_subject++;
+          }
+        }
+
+      /* Or to a non-unique first char after study */
+
+      else if (start_bits != NULL)
+        {
+        while (current_subject < end_subject)
+          {
+          register pcre_uint32 c = RAWUCHARTEST(current_subject);
+#ifndef COMPILE_PCRE8
+          if (c > 255) c = 255;
+#endif
+          if ((start_bits[c/8] & (1 << (c&7))) == 0)
+            {
+            current_subject++;
+#if defined SUPPORT_UTF && defined COMPILE_PCRE8
+            /* In non 8-bit mode, the iteration will stop for
+            characters > 255 at the beginning or not stop at all. */
+            if (utf)
+              ACROSSCHAR(current_subject < end_subject, *current_subject,
+                current_subject++);
+#endif
+            }
+          else break;
+          }
+        }
+      }
+
+    /* Restore fudged end_subject */
+
+    end_subject = save_end_subject;
+
+    /* The following two optimizations are disabled for partial matching or if
+    disabling is explicitly requested (and of course, by the test above, this
+    code is not obeyed when restarting after a partial match). */
+
+    if (((options | re->options) & PCRE_NO_START_OPTIMIZE) == 0 &&
+        (options & (PCRE_PARTIAL_HARD|PCRE_PARTIAL_SOFT)) == 0)
+      {
+      /* If the pattern was studied, a minimum subject length may be set. This
+      is a lower bound; no actual string of that length may actually match the
+      pattern. Although the value is, strictly, in characters, we treat it as
+      bytes to avoid spending too much time in this optimization. */
+
+      if (study != NULL && (study->flags & PCRE_STUDY_MINLEN) != 0 &&
+          (pcre_uint32)(end_subject - current_subject) < study->minlength)
+        return PCRE_ERROR_NOMATCH;
+
+      /* If req_char is set, we know that that character must appear in the
+      subject for the match to succeed. If the first character is set, req_char
+      must be later in the subject; otherwise the test starts at the match
+      point. This optimization can save a huge amount of work in patterns with
+      nested unlimited repeats that aren't going to match. Writing separate
+      code for cased/caseless versions makes it go faster, as does using an
+      autoincrement and backing off on a match.
+
+      HOWEVER: when the subject string is very, very long, searching to its end
+      can take a long time, and give bad performance on quite ordinary
+      patterns. This showed up when somebody was matching /^C/ on a 32-megabyte
+      string... so we don't do this when the string is sufficiently long. */
+
+      if (has_req_char && end_subject - current_subject < REQ_BYTE_MAX)
+        {
+        register PCRE_PUCHAR p = current_subject + (has_first_char? 1:0);
+
+        /* We don't need to repeat the search if we haven't yet reached the
+        place we found it at last time. */
+
+        if (p > req_char_ptr)
+          {
+          if (req_char != req_char2)
+            {
+            while (p < end_subject)
+              {
+              register pcre_uint32 pp = RAWUCHARINCTEST(p);
+              if (pp == req_char || pp == req_char2) { p--; break; }
+              }
+            }
+          else
+            {
+            while (p < end_subject)
+              {
+              if (RAWUCHARINCTEST(p) == req_char) { p--; break; }
+              }
+            }
+
+          /* If we can't find the required character, break the matching loop,
+          which will cause a return or PCRE_ERROR_NOMATCH. */
+
+          if (p >= end_subject) break;
+
+          /* If we have found the required character, save the point where we
+          found it, so that we don't search again next time round the loop if
+          the start hasn't passed this character yet. */
+
+          req_char_ptr = p;
+          }
+        }
+      }
+    }   /* End of optimizations that are done when not restarting */
+
+  /* OK, now we can do the business */
+
+  md->start_used_ptr = current_subject;
+  md->recursive = NULL;
+
+  rc = internal_dfa_exec(
+    md,                                /* fixed match data */
+    md->start_code,                    /* this subexpression's code */
+    current_subject,                   /* where we currently are */
+    start_offset,                      /* start offset in subject */
+    offsets,                           /* offset vector */
+    offsetcount,                       /* size of same */
+    workspace,                         /* workspace vector */
+    wscount,                           /* size of same */
+    0);                                /* function recurse level */
+
+  /* Anything other than "no match" means we are done, always; otherwise, carry
+  on only if not anchored. */
+
+  if (rc != PCRE_ERROR_NOMATCH || anchored)
+    {
+    if (rc == PCRE_ERROR_PARTIAL && offsetcount >= 2)
+      {
+      offsets[0] = (int)(md->start_used_ptr - (PCRE_PUCHAR)subject);
+      offsets[1] = (int)(end_subject - (PCRE_PUCHAR)subject);
+      if (offsetcount > 2)
+        offsets[2] = (int)(current_subject - (PCRE_PUCHAR)subject);
+      }
+    return rc;
+    }
+
+  /* Advance to the next subject character unless we are at the end of a line
+  and firstline is set. */
+
+  if (firstline && IS_NEWLINE(current_subject)) break;
+  current_subject++;
+#ifdef SUPPORT_UTF
+  if (utf)
+    {
+    ACROSSCHAR(current_subject < end_subject, *current_subject,
+      current_subject++);
+    }
+#endif
+  if (current_subject > end_subject) break;
+
+  /* If we have just passed a CR and we are now at a LF, and the pattern does
+  not contain any explicit matches for \r or \n, and the newline option is CRLF
+  or ANY or ANYCRLF, advance the match position by one more character. */
+
+  if (RAWUCHARTEST(current_subject - 1) == CHAR_CR &&
+      current_subject < end_subject &&
+      RAWUCHARTEST(current_subject) == CHAR_NL &&
+      (re->flags & PCRE_HASCRORLF) == 0 &&
+        (md->nltype == NLTYPE_ANY ||
+         md->nltype == NLTYPE_ANYCRLF ||
+         md->nllen == 2))
+    current_subject++;
+
+  }   /* "Bumpalong" loop */
+
+return PCRE_ERROR_NOMATCH;
+}
+
+/* End of pcre_dfa_exec.c */
diff --git a/pcre/pcre_exec.c b/pcre/pcre_exec.c
new file mode 100644
index 00000000000..ab76d020d16
--- /dev/null
+++ b/pcre/pcre_exec.c
@@ -0,0 +1,7213 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2013 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+/* This module contains pcre_exec(), the externally visible function that does
+pattern matching using an NFA algorithm, trying to mimic Perl as closely as
+possible. There are also some static supporting functions. */
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#define NLBLOCK md             /* Block containing newline information */
+#define PSSTART start_subject  /* Field containing processed string start */
+#define PSEND   end_subject    /* Field containing processed string end */
+
+#include "pcre_internal.h"
+
+/* Undefine some potentially clashing cpp symbols */
+
+#undef min
+#undef max
+
+/* The md->capture_last field uses the lower 16 bits for the last captured
+substring (which can never be greater than 65535) and a bit in the top half
+to mean "capture vector overflowed". This odd way of doing things was
+implemented when it was realized that preserving and restoring the overflow bit
+whenever the last capture number was saved/restored made for a neater
+interface, and doing it this way saved on (a) another variable, which would
+have increased the stack frame size (a big NO-NO in PCRE) and (b) another
+separate set of save/restore instructions. The following defines are used in
+implementing this. */
+
+#define CAPLMASK    0x0000ffff    /* The bits used for last_capture */
+#define OVFLMASK    0xffff0000    /* The bits used for the overflow flag */
+#define OVFLBIT     0x00010000    /* The bit that is set for overflow */
+
+/* Values for setting in md->match_function_type to indicate two special types
+of call to match(). We do it this way to save on using another stack variable,
+as stack usage is to be discouraged. */
+
+#define MATCH_CONDASSERT     1  /* Called to check a condition assertion */
+#define MATCH_CBEGROUP       2  /* Could-be-empty unlimited repeat group */
+
+/* Non-error returns from the match() function. Error returns are externally
+defined PCRE_ERROR_xxx codes, which are all negative. */
+
+#define MATCH_MATCH        1
+#define MATCH_NOMATCH      0
+
+/* Special internal returns from the match() function. Make them sufficiently
+negative to avoid the external error codes. */
+
+#define MATCH_ACCEPT       (-999)
+#define MATCH_KETRPOS      (-998)
+#define MATCH_ONCE         (-997)
+/* The next 5 must be kept together and in sequence so that a test that checks
+for any one of them can use a range. */
+#define MATCH_COMMIT       (-996)
+#define MATCH_PRUNE        (-995)
+#define MATCH_SKIP         (-994)
+#define MATCH_SKIP_ARG     (-993)
+#define MATCH_THEN         (-992)
+#define MATCH_BACKTRACK_MAX MATCH_THEN
+#define MATCH_BACKTRACK_MIN MATCH_COMMIT
+
+/* Maximum number of ints of offset to save on the stack for recursive calls.
+If the offset vector is bigger, malloc is used. This should be a multiple of 3,
+because the offset vector is always a multiple of 3 long. */
+
+#define REC_STACK_SAVE_MAX 30
+
+/* Min and max values for the common repeats; for the maxima, 0 => infinity */
+
+static const char rep_min[] = { 0, 0, 1, 1, 0, 0 };
+static const char rep_max[] = { 0, 0, 0, 0, 1, 1 };
+
+#ifdef PCRE_DEBUG
+/*************************************************
+*        Debugging function to print chars       *
+*************************************************/
+
+/* Print a sequence of chars in printable format, stopping at the end of the
+subject if the requested.
+
+Arguments:
+  p           points to characters
+  length      number to print
+  is_subject  TRUE if printing from within md->start_subject
+  md          pointer to matching data block, if is_subject is TRUE
+
+Returns:     nothing
+*/
+
+static void
+pchars(const pcre_uchar *p, int length, BOOL is_subject, match_data *md)
+{
+pcre_uint32 c;
+BOOL utf = md->utf;
+if (is_subject && length > md->end_subject - p) length = md->end_subject - p;
+while (length-- > 0)
+  if (isprint(c = RAWUCHARINCTEST(p))) printf("%c", (char)c); else printf("\\x{%02x}", c);
+}
+#endif
+
+
+
+/*************************************************
+*          Match a back-reference                *
+*************************************************/
+
+/* Normally, if a back reference hasn't been set, the length that is passed is
+negative, so the match always fails. However, in JavaScript compatibility mode,
+the length passed is zero. Note that in caseless UTF-8 mode, the number of
+subject bytes matched may be different to the number of reference bytes.
+
+Arguments:
+  offset      index into the offset vector
+  eptr        pointer into the subject
+  length      length of reference to be matched (number of bytes)
+  md          points to match data block
+  caseless    TRUE if caseless
+
+Returns:      >= 0 the number of subject bytes matched
+              -1 no match
+              -2 partial match; always given if at end subject
+*/
+
+static int
+match_ref(int offset, register PCRE_PUCHAR eptr, int length, match_data *md,
+  BOOL caseless)
+{
+PCRE_PUCHAR eptr_start = eptr;
+register PCRE_PUCHAR p = md->start_subject + md->offset_vector[offset];
+#ifdef SUPPORT_UTF
+BOOL utf = md->utf;
+#endif
+
+#ifdef PCRE_DEBUG
+if (eptr >= md->end_subject)
+  printf("matching subject ");
+else
+  {
+  printf("matching subject ");
+  pchars(eptr, length, TRUE, md);
+  }
+printf(" against backref ");
+pchars(p, length, FALSE, md);
+printf("\n");
+#endif
+
+/* Always fail if reference not set (and not JavaScript compatible - in that
+case the length is passed as zero). */
+
+if (length < 0) return -1;
+
+/* Separate the caseless case for speed. In UTF-8 mode we can only do this
+properly if Unicode properties are supported. Otherwise, we can check only
+ASCII characters. */
+
+if (caseless)
+  {
+#ifdef SUPPORT_UTF
+#ifdef SUPPORT_UCP
+  if (utf)
+    {
+    /* Match characters up to the end of the reference. NOTE: the number of
+    data units matched may differ, because in UTF-8 there are some characters
+    whose upper and lower case versions code have different numbers of bytes.
+    For example, U+023A (2 bytes in UTF-8) is the upper case version of U+2C65
+    (3 bytes in UTF-8); a sequence of 3 of the former uses 6 bytes, as does a
+    sequence of two of the latter. It is important, therefore, to check the
+    length along the reference, not along the subject (earlier code did this
+    wrong). */
+
+    PCRE_PUCHAR endptr = p + length;
+    while (p < endptr)
+      {
+      pcre_uint32 c, d;
+      const ucd_record *ur;
+      if (eptr >= md->end_subject) return -2;   /* Partial match */
+      GETCHARINC(c, eptr);
+      GETCHARINC(d, p);
+      ur = GET_UCD(d);
+      if (c != d && c != d + ur->other_case)
+        {
+        const pcre_uint32 *pp = PRIV(ucd_caseless_sets) + ur->caseset;
+        for (;;)
+          {
+          if (c < *pp) return -1;
+          if (c == *pp++) break;
+          }
+        }
+      }
+    }
+  else
+#endif
+#endif
+
+  /* The same code works when not in UTF-8 mode and in UTF-8 mode when there
+  is no UCP support. */
+    {
+    while (length-- > 0)
+      {
+      pcre_uint32 cc, cp;
+      if (eptr >= md->end_subject) return -2;   /* Partial match */
+      cc = RAWUCHARTEST(eptr);
+      cp = RAWUCHARTEST(p);
+      if (TABLE_GET(cp, md->lcc, cp) != TABLE_GET(cc, md->lcc, cc)) return -1;
+      p++;
+      eptr++;
+      }
+    }
+  }
+
+/* In the caseful case, we can just compare the bytes, whether or not we
+are in UTF-8 mode. */
+
+else
+  {
+  while (length-- > 0)
+    {
+    if (eptr >= md->end_subject) return -2;   /* Partial match */
+    if (RAWUCHARINCTEST(p) != RAWUCHARINCTEST(eptr)) return -1;
+    }
+  }
+
+return (int)(eptr - eptr_start);
+}
+
+
+
+/***************************************************************************
+****************************************************************************
+                   RECURSION IN THE match() FUNCTION
+
+The match() function is highly recursive, though not every recursive call
+increases the recursive depth. Nevertheless, some regular expressions can cause
+it to recurse to a great depth. I was writing for Unix, so I just let it call
+itself recursively. This uses the stack for saving everything that has to be
+saved for a recursive call. On Unix, the stack can be large, and this works
+fine.
+
+It turns out that on some non-Unix-like systems there are problems with
+programs that use a lot of stack. (This despite the fact that every last chip
+has oodles of memory these days, and techniques for extending the stack have
+been known for decades.) So....
+
+There is a fudge, triggered by defining NO_RECURSE, which avoids recursive
+calls by keeping local variables that need to be preserved in blocks of memory
+obtained from malloc() instead instead of on the stack. Macros are used to
+achieve this so that the actual code doesn't look very different to what it
+always used to.
+
+The original heap-recursive code used longjmp(). However, it seems that this
+can be very slow on some operating systems. Following a suggestion from Stan
+Switzer, the use of longjmp() has been abolished, at the cost of having to
+provide a unique number for each call to RMATCH. There is no way of generating
+a sequence of numbers at compile time in C. I have given them names, to make
+them stand out more clearly.
+
+Crude tests on x86 Linux show a small speedup of around 5-8%. However, on
+FreeBSD, avoiding longjmp() more than halves the time taken to run the standard
+tests. Furthermore, not using longjmp() means that local dynamic variables
+don't have indeterminate values; this has meant that the frame size can be
+reduced because the result can be "passed back" by straight setting of the
+variable instead of being passed in the frame.
+****************************************************************************
+***************************************************************************/
+
+/* Numbers for RMATCH calls. When this list is changed, the code at HEAP_RETURN
+below must be updated in sync.  */
+
+enum { RM1=1, RM2,  RM3,  RM4,  RM5,  RM6,  RM7,  RM8,  RM9,  RM10,
+       RM11,  RM12, RM13, RM14, RM15, RM16, RM17, RM18, RM19, RM20,
+       RM21,  RM22, RM23, RM24, RM25, RM26, RM27, RM28, RM29, RM30,
+       RM31,  RM32, RM33, RM34, RM35, RM36, RM37, RM38, RM39, RM40,
+       RM41,  RM42, RM43, RM44, RM45, RM46, RM47, RM48, RM49, RM50,
+       RM51,  RM52, RM53, RM54, RM55, RM56, RM57, RM58, RM59, RM60,
+       RM61,  RM62, RM63, RM64, RM65, RM66, RM67, RM68 };
+
+/* These versions of the macros use the stack, as normal. There are debugging
+versions and production versions. Note that the "rw" argument of RMATCH isn't
+actually used in this definition. */
+
+#ifndef NO_RECURSE
+#define REGISTER register
+
+#ifdef PCRE_DEBUG
+#define RMATCH(ra,rb,rc,rd,re,rw) \
+  { \
+  printf("match() called in line %d\n", __LINE__); \
+  rrc = match(ra,rb,mstart,rc,rd,re,rdepth+1); \
+  printf("to line %d\n", __LINE__); \
+  }
+#define RRETURN(ra) \
+  { \
+  printf("match() returned %d from line %d\n", ra, __LINE__); \
+  return ra; \
+  }
+#else
+#define RMATCH(ra,rb,rc,rd,re,rw) \
+  rrc = match(ra,rb,mstart,rc,rd,re,rdepth+1)
+#define RRETURN(ra) return ra
+#endif
+
+#else
+
+
+/* These versions of the macros manage a private stack on the heap. Note that
+the "rd" argument of RMATCH isn't actually used in this definition. It's the md
+argument of match(), which never changes. */
+
+#define REGISTER
+
+#define RMATCH(ra,rb,rc,rd,re,rw)\
+  {\
+  heapframe *newframe = frame->Xnextframe;\
+  if (newframe == NULL)\
+    {\
+    newframe = (heapframe *)(PUBL(stack_malloc))(sizeof(heapframe));\
+    if (newframe == NULL) RRETURN(PCRE_ERROR_NOMEMORY);\
+    newframe->Xnextframe = NULL;\
+    frame->Xnextframe = newframe;\
+    }\
+  frame->Xwhere = rw;\
+  newframe->Xeptr = ra;\
+  newframe->Xecode = rb;\
+  newframe->Xmstart = mstart;\
+  newframe->Xoffset_top = rc;\
+  newframe->Xeptrb = re;\
+  newframe->Xrdepth = frame->Xrdepth + 1;\
+  newframe->Xprevframe = frame;\
+  frame = newframe;\
+  DPRINTF(("restarting from line %d\n", __LINE__));\
+  goto HEAP_RECURSE;\
+  L_##rw:\
+  DPRINTF(("jumped back to line %d\n", __LINE__));\
+  }
+
+#define RRETURN(ra)\
+  {\
+  heapframe *oldframe = frame;\
+  frame = oldframe->Xprevframe;\
+  if (frame != NULL)\
+    {\
+    rrc = ra;\
+    goto HEAP_RETURN;\
+    }\
+  return ra;\
+  }
+
+
+/* Structure for remembering the local variables in a private frame */
+
+typedef struct heapframe {
+  struct heapframe *Xprevframe;
+  struct heapframe *Xnextframe;
+
+  /* Function arguments that may change */
+
+  PCRE_PUCHAR Xeptr;
+  const pcre_uchar *Xecode;
+  PCRE_PUCHAR Xmstart;
+  int Xoffset_top;
+  eptrblock *Xeptrb;
+  unsigned int Xrdepth;
+
+  /* Function local variables */
+
+  PCRE_PUCHAR Xcallpat;
+#ifdef SUPPORT_UTF
+  PCRE_PUCHAR Xcharptr;
+#endif
+  PCRE_PUCHAR Xdata;
+  PCRE_PUCHAR Xnext;
+  PCRE_PUCHAR Xpp;
+  PCRE_PUCHAR Xprev;
+  PCRE_PUCHAR Xsaved_eptr;
+
+  recursion_info Xnew_recursive;
+
+  BOOL Xcur_is_word;
+  BOOL Xcondition;
+  BOOL Xprev_is_word;
+
+#ifdef SUPPORT_UCP
+  int Xprop_type;
+  unsigned int Xprop_value;
+  int Xprop_fail_result;
+  int Xoclength;
+  pcre_uchar Xocchars[6];
+#endif
+
+  int Xcodelink;
+  int Xctype;
+  unsigned int Xfc;
+  int Xfi;
+  int Xlength;
+  int Xmax;
+  int Xmin;
+  unsigned int Xnumber;
+  int Xoffset;
+  unsigned int Xop;
+  pcre_int32 Xsave_capture_last;
+  int Xsave_offset1, Xsave_offset2, Xsave_offset3;
+  int Xstacksave[REC_STACK_SAVE_MAX];
+
+  eptrblock Xnewptrb;
+
+  /* Where to jump back to */
+
+  int Xwhere;
+
+} heapframe;
+
+#endif
+
+
+/***************************************************************************
+***************************************************************************/
+
+
+
+/*************************************************
+*         Match from current position            *
+*************************************************/
+
+/* This function is called recursively in many circumstances. Whenever it
+returns a negative (error) response, the outer incarnation must also return the
+same response. */
+
+/* These macros pack up tests that are used for partial matching, and which
+appear several times in the code. We set the "hit end" flag if the pointer is
+at the end of the subject and also past the start of the subject (i.e.
+something has been matched). For hard partial matching, we then return
+immediately. The second one is used when we already know we are past the end of
+the subject. */
+
+#define CHECK_PARTIAL()\
+  if (md->partial != 0 && eptr >= md->end_subject && \
+      eptr > md->start_used_ptr) \
+    { \
+    md->hitend = TRUE; \
+    if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL); \
+    }
+
+#define SCHECK_PARTIAL()\
+  if (md->partial != 0 && eptr > md->start_used_ptr) \
+    { \
+    md->hitend = TRUE; \
+    if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL); \
+    }
+
+
+/* Performance note: It might be tempting to extract commonly used fields from
+the md structure (e.g. utf, end_subject) into individual variables to improve
+performance. Tests using gcc on a SPARC disproved this; in the first case, it
+made performance worse.
+
+Arguments:
+   eptr        pointer to current character in subject
+   ecode       pointer to current position in compiled code
+   mstart      pointer to the current match start position (can be modified
+                 by encountering \K)
+   offset_top  current top pointer
+   md          pointer to "static" info for the match
+   eptrb       pointer to chain of blocks containing eptr at start of
+                 brackets - for testing for empty matches
+   rdepth      the recursion depth
+
+Returns:       MATCH_MATCH if matched            )  these values are >= 0
+               MATCH_NOMATCH if failed to match  )
+               a negative MATCH_xxx value for PRUNE, SKIP, etc
+               a negative PCRE_ERROR_xxx value if aborted by an error condition
+                 (e.g. stopped by repeated call or recursion limit)
+*/
+
+static int
+match(REGISTER PCRE_PUCHAR eptr, REGISTER const pcre_uchar *ecode,
+  PCRE_PUCHAR mstart, int offset_top, match_data *md, eptrblock *eptrb,
+  unsigned int rdepth)
+{
+/* These variables do not need to be preserved over recursion in this function,
+so they can be ordinary variables in all cases. Mark some of them with
+"register" because they are used a lot in loops. */
+
+register int  rrc;         /* Returns from recursive calls */
+register int  i;           /* Used for loops not involving calls to RMATCH() */
+register pcre_uint32 c;    /* Character values not kept over RMATCH() calls */
+register BOOL utf;         /* Local copy of UTF flag for speed */
+
+BOOL minimize, possessive; /* Quantifier options */
+BOOL caseless;
+int condcode;
+
+/* When recursion is not being used, all "local" variables that have to be
+preserved over calls to RMATCH() are part of a "frame". We set up the top-level
+frame on the stack here; subsequent instantiations are obtained from the heap
+whenever RMATCH() does a "recursion". See the macro definitions above. Putting
+the top-level on the stack rather than malloc-ing them all gives a performance
+boost in many cases where there is not much "recursion". */
+
+#ifdef NO_RECURSE
+heapframe *frame = (heapframe *)md->match_frames_base;
+
+/* Copy in the original argument variables */
+
+frame->Xeptr = eptr;
+frame->Xecode = ecode;
+frame->Xmstart = mstart;
+frame->Xoffset_top = offset_top;
+frame->Xeptrb = eptrb;
+frame->Xrdepth = rdepth;
+
+/* This is where control jumps back to to effect "recursion" */
+
+HEAP_RECURSE:
+
+/* Macros make the argument variables come from the current frame */
+
+#define eptr               frame->Xeptr
+#define ecode              frame->Xecode
+#define mstart             frame->Xmstart
+#define offset_top         frame->Xoffset_top
+#define eptrb              frame->Xeptrb
+#define rdepth             frame->Xrdepth
+
+/* Ditto for the local variables */
+
+#ifdef SUPPORT_UTF
+#define charptr            frame->Xcharptr
+#endif
+#define callpat            frame->Xcallpat
+#define codelink           frame->Xcodelink
+#define data               frame->Xdata
+#define next               frame->Xnext
+#define pp                 frame->Xpp
+#define prev               frame->Xprev
+#define saved_eptr         frame->Xsaved_eptr
+
+#define new_recursive      frame->Xnew_recursive
+
+#define cur_is_word        frame->Xcur_is_word
+#define condition          frame->Xcondition
+#define prev_is_word       frame->Xprev_is_word
+
+#ifdef SUPPORT_UCP
+#define prop_type          frame->Xprop_type
+#define prop_value         frame->Xprop_value
+#define prop_fail_result   frame->Xprop_fail_result
+#define oclength           frame->Xoclength
+#define occhars            frame->Xocchars
+#endif
+
+#define ctype              frame->Xctype
+#define fc                 frame->Xfc
+#define fi                 frame->Xfi
+#define length             frame->Xlength
+#define max                frame->Xmax
+#define min                frame->Xmin
+#define number             frame->Xnumber
+#define offset             frame->Xoffset
+#define op                 frame->Xop
+#define save_capture_last  frame->Xsave_capture_last
+#define save_offset1       frame->Xsave_offset1
+#define save_offset2       frame->Xsave_offset2
+#define save_offset3       frame->Xsave_offset3
+#define stacksave          frame->Xstacksave
+
+#define newptrb            frame->Xnewptrb
+
+/* When recursion is being used, local variables are allocated on the stack and
+get preserved during recursion in the normal way. In this environment, fi and
+i, and fc and c, can be the same variables. */
+
+#else         /* NO_RECURSE not defined */
+#define fi i
+#define fc c
+
+/* Many of the following variables are used only in small blocks of the code.
+My normal style of coding would have declared them within each of those blocks.
+However, in order to accommodate the version of this code that uses an external
+"stack" implemented on the heap, it is easier to declare them all here, so the
+declarations can be cut out in a block. The only declarations within blocks
+below are for variables that do not have to be preserved over a recursive call
+to RMATCH(). */
+
+#ifdef SUPPORT_UTF
+const pcre_uchar *charptr;
+#endif
+const pcre_uchar *callpat;
+const pcre_uchar *data;
+const pcre_uchar *next;
+PCRE_PUCHAR       pp;
+const pcre_uchar *prev;
+PCRE_PUCHAR       saved_eptr;
+
+recursion_info new_recursive;
+
+BOOL cur_is_word;
+BOOL condition;
+BOOL prev_is_word;
+
+#ifdef SUPPORT_UCP
+int prop_type;
+unsigned int prop_value;
+int prop_fail_result;
+int oclength;
+pcre_uchar occhars[6];
+#endif
+
+int codelink;
+int ctype;
+int length;
+int max;
+int min;
+unsigned int number;
+int offset;
+unsigned int op;
+pcre_int32 save_capture_last;
+int save_offset1, save_offset2, save_offset3;
+int stacksave[REC_STACK_SAVE_MAX];
+
+eptrblock newptrb;
+
+/* There is a special fudge for calling match() in a way that causes it to
+measure the size of its basic stack frame when the stack is being used for
+recursion. The second argument (ecode) being NULL triggers this behaviour. It
+cannot normally ever be NULL. The return is the negated value of the frame
+size. */
+
+if (ecode == NULL)
+  {
+  if (rdepth == 0)
+    return match((PCRE_PUCHAR)&rdepth, NULL, NULL, 0, NULL, NULL, 1);
+  else
+    {
+    int len = (char *)&rdepth - (char *)eptr;
+    return (len > 0)? -len : len;
+    }
+  }
+#endif     /* NO_RECURSE */
+
+/* To save space on the stack and in the heap frame, I have doubled up on some
+of the local variables that are used only in localised parts of the code, but
+still need to be preserved over recursive calls of match(). These macros define
+the alternative names that are used. */
+
+#define allow_zero    cur_is_word
+#define cbegroup      condition
+#define code_offset   codelink
+#define condassert    condition
+#define matched_once  prev_is_word
+#define foc           number
+#define save_mark     data
+
+/* These statements are here to stop the compiler complaining about unitialized
+variables. */
+
+#ifdef SUPPORT_UCP
+prop_value = 0;
+prop_fail_result = 0;
+#endif
+
+
+/* This label is used for tail recursion, which is used in a few cases even
+when NO_RECURSE is not defined, in order to reduce the amount of stack that is
+used. Thanks to Ian Taylor for noticing this possibility and sending the
+original patch. */
+
+TAIL_RECURSE:
+
+/* OK, now we can get on with the real code of the function. Recursive calls
+are specified by the macro RMATCH and RRETURN is used to return. When
+NO_RECURSE is *not* defined, these just turn into a recursive call to match()
+and a "return", respectively (possibly with some debugging if PCRE_DEBUG is
+defined). However, RMATCH isn't like a function call because it's quite a
+complicated macro. It has to be used in one particular way. This shouldn't,
+however, impact performance when true recursion is being used. */
+
+#ifdef SUPPORT_UTF
+utf = md->utf;       /* Local copy of the flag */
+#else
+utf = FALSE;
+#endif
+
+/* First check that we haven't called match() too many times, or that we
+haven't exceeded the recursive call limit. */
+
+if (md->match_call_count++ >= md->match_limit) RRETURN(PCRE_ERROR_MATCHLIMIT);
+if (rdepth >= md->match_limit_recursion) RRETURN(PCRE_ERROR_RECURSIONLIMIT);
+
+/* At the start of a group with an unlimited repeat that may match an empty
+string, the variable md->match_function_type is set to MATCH_CBEGROUP. It is
+done this way to save having to use another function argument, which would take
+up space on the stack. See also MATCH_CONDASSERT below.
+
+When MATCH_CBEGROUP is set, add the current subject pointer to the chain of
+such remembered pointers, to be checked when we hit the closing ket, in order
+to break infinite loops that match no characters. When match() is called in
+other circumstances, don't add to the chain. The MATCH_CBEGROUP feature must
+NOT be used with tail recursion, because the memory block that is used is on
+the stack, so a new one may be required for each match(). */
+
+if (md->match_function_type == MATCH_CBEGROUP)
+  {
+  newptrb.epb_saved_eptr = eptr;
+  newptrb.epb_prev = eptrb;
+  eptrb = &newptrb;
+  md->match_function_type = 0;
+  }
+
+/* Now start processing the opcodes. */
+
+for (;;)
+  {
+  minimize = possessive = FALSE;
+  op = *ecode;
+
+  switch(op)
+    {
+    case OP_MARK:
+    md->nomatch_mark = ecode + 2;
+    md->mark = NULL;    /* In case previously set by assertion */
+    RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode] + ecode[1], offset_top, md,
+      eptrb, RM55);
+    if ((rrc == MATCH_MATCH || rrc == MATCH_ACCEPT) &&
+         md->mark == NULL) md->mark = ecode + 2;
+
+    /* A return of MATCH_SKIP_ARG means that matching failed at SKIP with an
+    argument, and we must check whether that argument matches this MARK's
+    argument. It is passed back in md->start_match_ptr (an overloading of that
+    variable). If it does match, we reset that variable to the current subject
+    position and return MATCH_SKIP. Otherwise, pass back the return code
+    unaltered. */
+
+    else if (rrc == MATCH_SKIP_ARG &&
+        STRCMP_UC_UC_TEST(ecode + 2, md->start_match_ptr) == 0)
+      {
+      md->start_match_ptr = eptr;
+      RRETURN(MATCH_SKIP);
+      }
+    RRETURN(rrc);
+
+    case OP_FAIL:
+    RRETURN(MATCH_NOMATCH);
+
+    case OP_COMMIT:
+    RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md,
+      eptrb, RM52);
+    if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+    RRETURN(MATCH_COMMIT);
+
+    case OP_PRUNE:
+    RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md,
+      eptrb, RM51);
+    if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+    RRETURN(MATCH_PRUNE);
+
+    case OP_PRUNE_ARG:
+    md->nomatch_mark = ecode + 2;
+    md->mark = NULL;    /* In case previously set by assertion */
+    RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode] + ecode[1], offset_top, md,
+      eptrb, RM56);
+    if ((rrc == MATCH_MATCH || rrc == MATCH_ACCEPT) &&
+         md->mark == NULL) md->mark = ecode + 2;
+    if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+    RRETURN(MATCH_PRUNE);
+
+    case OP_SKIP:
+    RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md,
+      eptrb, RM53);
+    if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+    md->start_match_ptr = eptr;   /* Pass back current position */
+    RRETURN(MATCH_SKIP);
+
+    /* Note that, for Perl compatibility, SKIP with an argument does NOT set
+    nomatch_mark. When a pattern match ends with a SKIP_ARG for which there was
+    not a matching mark, we have to re-run the match, ignoring the SKIP_ARG
+    that failed and any that precede it (either they also failed, or were not
+    triggered). To do this, we maintain a count of executed SKIP_ARGs. If a
+    SKIP_ARG gets to top level, the match is re-run with md->ignore_skip_arg
+    set to the count of the one that failed. */
+
+    case OP_SKIP_ARG:
+    md->skip_arg_count++;
+    if (md->skip_arg_count <= md->ignore_skip_arg)
+      {
+      ecode += PRIV(OP_lengths)[*ecode] + ecode[1];
+      break;
+      }
+    RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode] + ecode[1], offset_top, md,
+      eptrb, RM57);
+    if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+
+    /* Pass back the current skip name by overloading md->start_match_ptr and
+    returning the special MATCH_SKIP_ARG return code. This will either be
+    caught by a matching MARK, or get to the top, where it causes a rematch
+    with md->ignore_skip_arg set to the value of md->skip_arg_count. */
+
+    md->start_match_ptr = ecode + 2;
+    RRETURN(MATCH_SKIP_ARG);
+
+    /* For THEN (and THEN_ARG) we pass back the address of the opcode, so that
+    the branch in which it occurs can be determined. Overload the start of
+    match pointer to do this. */
+
+    case OP_THEN:
+    RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md,
+      eptrb, RM54);
+    if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+    md->start_match_ptr = ecode;
+    RRETURN(MATCH_THEN);
+
+    case OP_THEN_ARG:
+    md->nomatch_mark = ecode + 2;
+    md->mark = NULL;    /* In case previously set by assertion */
+    RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode] + ecode[1], offset_top,
+      md, eptrb, RM58);
+    if ((rrc == MATCH_MATCH || rrc == MATCH_ACCEPT) &&
+         md->mark == NULL) md->mark = ecode + 2;
+    if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+    md->start_match_ptr = ecode;
+    RRETURN(MATCH_THEN);
+
+    /* Handle an atomic group that does not contain any capturing parentheses.
+    This can be handled like an assertion. Prior to 8.13, all atomic groups
+    were handled this way. In 8.13, the code was changed as below for ONCE, so
+    that backups pass through the group and thereby reset captured values.
+    However, this uses a lot more stack, so in 8.20, atomic groups that do not
+    contain any captures generate OP_ONCE_NC, which can be handled in the old,
+    less stack intensive way.
+
+    Check the alternative branches in turn - the matching won't pass the KET
+    for this kind of subpattern. If any one branch matches, we carry on as at
+    the end of a normal bracket, leaving the subject pointer, but resetting
+    the start-of-match value in case it was changed by \K. */
+
+    case OP_ONCE_NC:
+    prev = ecode;
+    saved_eptr = eptr;
+    save_mark = md->mark;
+    do
+      {
+      RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, eptrb, RM64);
+      if (rrc == MATCH_MATCH)  /* Note: _not_ MATCH_ACCEPT */
+        {
+        mstart = md->start_match_ptr;
+        break;
+        }
+      if (rrc == MATCH_THEN)
+        {
+        next = ecode + GET(ecode,1);
+        if (md->start_match_ptr < next &&
+            (*ecode == OP_ALT || *next == OP_ALT))
+          rrc = MATCH_NOMATCH;
+        }
+
+      if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+      ecode += GET(ecode,1);
+      md->mark = save_mark;
+      }
+    while (*ecode == OP_ALT);
+
+    /* If hit the end of the group (which could be repeated), fail */
+
+    if (*ecode != OP_ONCE_NC && *ecode != OP_ALT) RRETURN(MATCH_NOMATCH);
+
+    /* Continue as from after the group, updating the offsets high water
+    mark, since extracts may have been taken. */
+
+    do ecode += GET(ecode, 1); while (*ecode == OP_ALT);
+
+    offset_top = md->end_offset_top;
+    eptr = md->end_match_ptr;
+
+    /* For a non-repeating ket, just continue at this level. This also
+    happens for a repeating ket if no characters were matched in the group.
+    This is the forcible breaking of infinite loops as implemented in Perl
+    5.005. */
+
+    if (*ecode == OP_KET || eptr == saved_eptr)
+      {
+      ecode += 1+LINK_SIZE;
+      break;
+      }
+
+    /* The repeating kets try the rest of the pattern or restart from the
+    preceding bracket, in the appropriate order. The second "call" of match()
+    uses tail recursion, to avoid using another stack frame. */
+
+    if (*ecode == OP_KETRMIN)
+      {
+      RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, eptrb, RM65);
+      if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+      ecode = prev;
+      goto TAIL_RECURSE;
+      }
+    else  /* OP_KETRMAX */
+      {
+      RMATCH(eptr, prev, offset_top, md, eptrb, RM66);
+      if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+      ecode += 1 + LINK_SIZE;
+      goto TAIL_RECURSE;
+      }
+    /* Control never gets here */
+
+    /* Handle a capturing bracket, other than those that are possessive with an
+    unlimited repeat. If there is space in the offset vector, save the current
+    subject position in the working slot at the top of the vector. We mustn't
+    change the current values of the data slot, because they may be set from a
+    previous iteration of this group, and be referred to by a reference inside
+    the group. A failure to match might occur after the group has succeeded,
+    if something later on doesn't match. For this reason, we need to restore
+    the working value and also the values of the final offsets, in case they
+    were set by a previous iteration of the same bracket.
+
+    If there isn't enough space in the offset vector, treat this as if it were
+    a non-capturing bracket. Don't worry about setting the flag for the error
+    case here; that is handled in the code for KET. */
+
+    case OP_CBRA:
+    case OP_SCBRA:
+    number = GET2(ecode, 1+LINK_SIZE);
+    offset = number << 1;
+
+#ifdef PCRE_DEBUG
+    printf("start bracket %d\n", number);
+    printf("subject=");
+    pchars(eptr, 16, TRUE, md);
+    printf("\n");
+#endif
+
+    if (offset < md->offset_max)
+      {
+      save_offset1 = md->offset_vector[offset];
+      save_offset2 = md->offset_vector[offset+1];
+      save_offset3 = md->offset_vector[md->offset_end - number];
+      save_capture_last = md->capture_last;
+      save_mark = md->mark;
+
+      DPRINTF(("saving %d %d %d\n", save_offset1, save_offset2, save_offset3));
+      md->offset_vector[md->offset_end - number] =
+        (int)(eptr - md->start_subject);
+
+      for (;;)
+        {
+        if (op >= OP_SBRA) md->match_function_type = MATCH_CBEGROUP;
+        RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md,
+          eptrb, RM1);
+        if (rrc == MATCH_ONCE) break;  /* Backing up through an atomic group */
+
+        /* If we backed up to a THEN, check whether it is within the current
+        branch by comparing the address of the THEN that is passed back with
+        the end of the branch. If it is within the current branch, and the
+        branch is one of two or more alternatives (it either starts or ends
+        with OP_ALT), we have reached the limit of THEN's action, so convert
+        the return code to NOMATCH, which will cause normal backtracking to
+        happen from now on. Otherwise, THEN is passed back to an outer
+        alternative. This implements Perl's treatment of parenthesized groups,
+        where a group not containing | does not affect the current alternative,
+        that is, (X) is NOT the same as (X|(*F)). */
+
+        if (rrc == MATCH_THEN)
+          {
+          next = ecode + GET(ecode,1);
+          if (md->start_match_ptr < next &&
+              (*ecode == OP_ALT || *next == OP_ALT))
+            rrc = MATCH_NOMATCH;
+          }
+
+        /* Anything other than NOMATCH is passed back. */
+
+        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+        md->capture_last = save_capture_last;
+        ecode += GET(ecode, 1);
+        md->mark = save_mark;
+        if (*ecode != OP_ALT) break;
+        }
+
+      DPRINTF(("bracket %d failed\n", number));
+      md->offset_vector[offset] = save_offset1;
+      md->offset_vector[offset+1] = save_offset2;
+      md->offset_vector[md->offset_end - number] = save_offset3;
+
+      /* At this point, rrc will be one of MATCH_ONCE or MATCH_NOMATCH. */
+
+      RRETURN(rrc);
+      }
+
+    /* FALL THROUGH ... Insufficient room for saving captured contents. Treat
+    as a non-capturing bracket. */
+
+    /* VVVVVVVVVVVVVVVVVVVVVVVVV */
+    /* VVVVVVVVVVVVVVVVVVVVVVVVV */
+
+    DPRINTF(("insufficient capture room: treat as non-capturing\n"));
+
+    /* VVVVVVVVVVVVVVVVVVVVVVVVV */
+    /* VVVVVVVVVVVVVVVVVVVVVVVVV */
+
+    /* Non-capturing or atomic group, except for possessive with unlimited
+    repeat and ONCE group with no captures. Loop for all the alternatives.
+
+    When we get to the final alternative within the brackets, we used to return
+    the result of a recursive call to match() whatever happened so it was
+    possible to reduce stack usage by turning this into a tail recursion,
+    except in the case of a possibly empty group. However, now that there is
+    the possiblity of (*THEN) occurring in the final alternative, this
+    optimization is no longer always possible.
+
+    We can optimize if we know there are no (*THEN)s in the pattern; at present
+    this is the best that can be done.
+
+    MATCH_ONCE is returned when the end of an atomic group is successfully
+    reached, but subsequent matching fails. It passes back up the tree (causing
+    captured values to be reset) until the original atomic group level is
+    reached. This is tested by comparing md->once_target with the start of the
+    group. At this point, the return is converted into MATCH_NOMATCH so that
+    previous backup points can be taken. */
+
+    case OP_ONCE:
+    case OP_BRA:
+    case OP_SBRA:
+    DPRINTF(("start non-capturing bracket\n"));
+
+    for (;;)
+      {
+      if (op >= OP_SBRA || op == OP_ONCE)
+        md->match_function_type = MATCH_CBEGROUP;
+
+      /* If this is not a possibly empty group, and there are no (*THEN)s in
+      the pattern, and this is the final alternative, optimize as described
+      above. */
+
+      else if (!md->hasthen && ecode[GET(ecode, 1)] != OP_ALT)
+        {
+        ecode += PRIV(OP_lengths)[*ecode];
+        goto TAIL_RECURSE;
+        }
+
+      /* In all other cases, we have to make another call to match(). */
+
+      save_mark = md->mark;
+      save_capture_last = md->capture_last;
+      RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md, eptrb,
+        RM2);
+
+      /* See comment in the code for capturing groups above about handling
+      THEN. */
+
+      if (rrc == MATCH_THEN)
+        {
+        next = ecode + GET(ecode,1);
+        if (md->start_match_ptr < next &&
+            (*ecode == OP_ALT || *next == OP_ALT))
+          rrc = MATCH_NOMATCH;
+        }
+
+      if (rrc != MATCH_NOMATCH)
+        {
+        if (rrc == MATCH_ONCE)
+          {
+          const pcre_uchar *scode = ecode;
+          if (*scode != OP_ONCE)           /* If not at start, find it */
+            {
+            while (*scode == OP_ALT) scode += GET(scode, 1);
+            scode -= GET(scode, 1);
+            }
+          if (md->once_target == scode) rrc = MATCH_NOMATCH;
+          }
+        RRETURN(rrc);
+        }
+      ecode += GET(ecode, 1);
+      md->mark = save_mark;
+      if (*ecode != OP_ALT) break;
+      md->capture_last = save_capture_last;
+      }
+
+    RRETURN(MATCH_NOMATCH);
+
+    /* Handle possessive capturing brackets with an unlimited repeat. We come
+    here from BRAZERO with allow_zero set TRUE. The offset_vector values are
+    handled similarly to the normal case above. However, the matching is
+    different. The end of these brackets will always be OP_KETRPOS, which
+    returns MATCH_KETRPOS without going further in the pattern. By this means
+    we can handle the group by iteration rather than recursion, thereby
+    reducing the amount of stack needed. */
+
+    case OP_CBRAPOS:
+    case OP_SCBRAPOS:
+    allow_zero = FALSE;
+
+    POSSESSIVE_CAPTURE:
+    number = GET2(ecode, 1+LINK_SIZE);
+    offset = number << 1;
+
+#ifdef PCRE_DEBUG
+    printf("start possessive bracket %d\n", number);
+    printf("subject=");
+    pchars(eptr, 16, TRUE, md);
+    printf("\n");
+#endif
+
+    if (offset < md->offset_max)
+      {
+      matched_once = FALSE;
+      code_offset = (int)(ecode - md->start_code);
+
+      save_offset1 = md->offset_vector[offset];
+      save_offset2 = md->offset_vector[offset+1];
+      save_offset3 = md->offset_vector[md->offset_end - number];
+      save_capture_last = md->capture_last;
+
+      DPRINTF(("saving %d %d %d\n", save_offset1, save_offset2, save_offset3));
+
+      /* Each time round the loop, save the current subject position for use
+      when the group matches. For MATCH_MATCH, the group has matched, so we
+      restart it with a new subject starting position, remembering that we had
+      at least one match. For MATCH_NOMATCH, carry on with the alternatives, as
+      usual. If we haven't matched any alternatives in any iteration, check to
+      see if a previous iteration matched. If so, the group has matched;
+      continue from afterwards. Otherwise it has failed; restore the previous
+      capture values before returning NOMATCH. */
+
+      for (;;)
+        {
+        md->offset_vector[md->offset_end - number] =
+          (int)(eptr - md->start_subject);
+        if (op >= OP_SBRA) md->match_function_type = MATCH_CBEGROUP;
+        RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md,
+          eptrb, RM63);
+        if (rrc == MATCH_KETRPOS)
+          {
+          offset_top = md->end_offset_top;
+          eptr = md->end_match_ptr;
+          ecode = md->start_code + code_offset;
+          save_capture_last = md->capture_last;
+          matched_once = TRUE;
+          continue;
+          }
+
+        /* See comment in the code for capturing groups above about handling
+        THEN. */
+
+        if (rrc == MATCH_THEN)
+          {
+          next = ecode + GET(ecode,1);
+          if (md->start_match_ptr < next &&
+              (*ecode == OP_ALT || *next == OP_ALT))
+            rrc = MATCH_NOMATCH;
+          }
+
+        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+        md->capture_last = save_capture_last;
+        ecode += GET(ecode, 1);
+        if (*ecode != OP_ALT) break;
+        }
+
+      if (!matched_once)
+        {
+        md->offset_vector[offset] = save_offset1;
+        md->offset_vector[offset+1] = save_offset2;
+        md->offset_vector[md->offset_end - number] = save_offset3;
+        }
+
+      if (allow_zero || matched_once)
+        {
+        ecode += 1 + LINK_SIZE;
+        break;
+        }
+
+      RRETURN(MATCH_NOMATCH);
+      }
+
+    /* FALL THROUGH ... Insufficient room for saving captured contents. Treat
+    as a non-capturing bracket. */
+
+    /* VVVVVVVVVVVVVVVVVVVVVVVVV */
+    /* VVVVVVVVVVVVVVVVVVVVVVVVV */
+
+    DPRINTF(("insufficient capture room: treat as non-capturing\n"));
+
+    /* VVVVVVVVVVVVVVVVVVVVVVVVV */
+    /* VVVVVVVVVVVVVVVVVVVVVVVVV */
+
+    /* Non-capturing possessive bracket with unlimited repeat. We come here
+    from BRAZERO with allow_zero = TRUE. The code is similar to the above,
+    without the capturing complication. It is written out separately for speed
+    and cleanliness. */
+
+    case OP_BRAPOS:
+    case OP_SBRAPOS:
+    allow_zero = FALSE;
+
+    POSSESSIVE_NON_CAPTURE:
+    matched_once = FALSE;
+    code_offset = (int)(ecode - md->start_code);
+    save_capture_last = md->capture_last;
+
+    for (;;)
+      {
+      if (op >= OP_SBRA) md->match_function_type = MATCH_CBEGROUP;
+      RMATCH(eptr, ecode + PRIV(OP_lengths)[*ecode], offset_top, md,
+        eptrb, RM48);
+      if (rrc == MATCH_KETRPOS)
+        {
+        offset_top = md->end_offset_top;
+        eptr = md->end_match_ptr;
+        ecode = md->start_code + code_offset;
+        matched_once = TRUE;
+        continue;
+        }
+
+      /* See comment in the code for capturing groups above about handling
+      THEN. */
+
+      if (rrc == MATCH_THEN)
+        {
+        next = ecode + GET(ecode,1);
+        if (md->start_match_ptr < next &&
+            (*ecode == OP_ALT || *next == OP_ALT))
+          rrc = MATCH_NOMATCH;
+        }
+
+      if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+      ecode += GET(ecode, 1);
+      if (*ecode != OP_ALT) break;
+      md->capture_last = save_capture_last;
+      }
+
+    if (matched_once || allow_zero)
+      {
+      ecode += 1 + LINK_SIZE;
+      break;
+      }
+    RRETURN(MATCH_NOMATCH);
+
+    /* Control never reaches here. */
+
+    /* Conditional group: compilation checked that there are no more than
+    two branches. If the condition is false, skipping the first branch takes us
+    past the end if there is only one branch, but that's OK because that is
+    exactly what going to the ket would do. */
+
+    case OP_COND:
+    case OP_SCOND:
+    codelink = GET(ecode, 1);
+
+    /* Because of the way auto-callout works during compile, a callout item is
+    inserted between OP_COND and an assertion condition. */
+
+    if (ecode[LINK_SIZE+1] == OP_CALLOUT)
+      {
+      if (PUBL(callout) != NULL)
+        {
+        PUBL(callout_block) cb;
+        cb.version          = 2;   /* Version 1 of the callout block */
+        cb.callout_number   = ecode[LINK_SIZE+2];
+        cb.offset_vector    = md->offset_vector;
+#if defined COMPILE_PCRE8
+        cb.subject          = (PCRE_SPTR)md->start_subject;
+#elif defined COMPILE_PCRE16
+        cb.subject          = (PCRE_SPTR16)md->start_subject;
+#elif defined COMPILE_PCRE32
+        cb.subject          = (PCRE_SPTR32)md->start_subject;
+#endif
+        cb.subject_length   = (int)(md->end_subject - md->start_subject);
+        cb.start_match      = (int)(mstart - md->start_subject);
+        cb.current_position = (int)(eptr - md->start_subject);
+        cb.pattern_position = GET(ecode, LINK_SIZE + 3);
+        cb.next_item_length = GET(ecode, 3 + 2*LINK_SIZE);
+        cb.capture_top      = offset_top/2;
+        cb.capture_last     = md->capture_last & CAPLMASK;
+        /* Internal change requires this for API compatibility. */
+        if (cb.capture_last == 0) cb.capture_last = -1;
+        cb.callout_data     = md->callout_data;
+        cb.mark             = md->nomatch_mark;
+        if ((rrc = (*PUBL(callout))(&cb)) > 0) RRETURN(MATCH_NOMATCH);
+        if (rrc < 0) RRETURN(rrc);
+        }
+      ecode += PRIV(OP_lengths)[OP_CALLOUT];
+      codelink -= PRIV(OP_lengths)[OP_CALLOUT];
+      }
+
+    condcode = ecode[LINK_SIZE+1];
+
+    /* Now see what the actual condition is */
+
+    if (condcode == OP_RREF || condcode == OP_NRREF)    /* Recursion test */
+      {
+      if (md->recursive == NULL)                /* Not recursing => FALSE */
+        {
+        condition = FALSE;
+        ecode += GET(ecode, 1);
+        }
+      else
+        {
+        unsigned int recno = GET2(ecode, LINK_SIZE + 2);   /* Recursion group number*/
+        condition = (recno == RREF_ANY || recno == md->recursive->group_num);
+
+        /* If the test is for recursion into a specific subpattern, and it is
+        false, but the test was set up by name, scan the table to see if the
+        name refers to any other numbers, and test them. The condition is true
+        if any one is set. */
+
+        if (!condition && condcode == OP_NRREF)
+          {
+          pcre_uchar *slotA = md->name_table;
+          for (i = 0; i < md->name_count; i++)
+            {
+            if (GET2(slotA, 0) == recno) break;
+            slotA += md->name_entry_size;
+            }
+
+          /* Found a name for the number - there can be only one; duplicate
+          names for different numbers are allowed, but not vice versa. First
+          scan down for duplicates. */
+
+          if (i < md->name_count)
+            {
+            pcre_uchar *slotB = slotA;
+            while (slotB > md->name_table)
+              {
+              slotB -= md->name_entry_size;
+              if (STRCMP_UC_UC(slotA + IMM2_SIZE, slotB + IMM2_SIZE) == 0)
+                {
+                condition = GET2(slotB, 0) == md->recursive->group_num;
+                if (condition) break;
+                }
+              else break;
+              }
+
+            /* Scan up for duplicates */
+
+            if (!condition)
+              {
+              slotB = slotA;
+              for (i++; i < md->name_count; i++)
+                {
+                slotB += md->name_entry_size;
+                if (STRCMP_UC_UC(slotA + IMM2_SIZE, slotB + IMM2_SIZE) == 0)
+                  {
+                  condition = GET2(slotB, 0) == md->recursive->group_num;
+                  if (condition) break;
+                  }
+                else break;
+                }
+              }
+            }
+          }
+
+        /* Chose branch according to the condition */
+
+        ecode += condition? 1 + IMM2_SIZE : GET(ecode, 1);
+        }
+      }
+
+    else if (condcode == OP_CREF || condcode == OP_NCREF)  /* Group used test */
+      {
+      offset = GET2(ecode, LINK_SIZE+2) << 1;  /* Doubled ref number */
+      condition = offset < offset_top && md->offset_vector[offset] >= 0;
+
+      /* If the numbered capture is unset, but the reference was by name,
+      scan the table to see if the name refers to any other numbers, and test
+      them. The condition is true if any one is set. This is tediously similar
+      to the code above, but not close enough to try to amalgamate. */
+
+      if (!condition && condcode == OP_NCREF)
+        {
+        unsigned int refno = offset >> 1;
+        pcre_uchar *slotA = md->name_table;
+
+        for (i = 0; i < md->name_count; i++)
+          {
+          if (GET2(slotA, 0) == refno) break;
+          slotA += md->name_entry_size;
+          }
+
+        /* Found a name for the number - there can be only one; duplicate names
+        for different numbers are allowed, but not vice versa. First scan down
+        for duplicates. */
+
+        if (i < md->name_count)
+          {
+          pcre_uchar *slotB = slotA;
+          while (slotB > md->name_table)
+            {
+            slotB -= md->name_entry_size;
+            if (STRCMP_UC_UC(slotA + IMM2_SIZE, slotB + IMM2_SIZE) == 0)
+              {
+              offset = GET2(slotB, 0) << 1;
+              condition = offset < offset_top &&
+                md->offset_vector[offset] >= 0;
+              if (condition) break;
+              }
+            else break;
+            }
+
+          /* Scan up for duplicates */
+
+          if (!condition)
+            {
+            slotB = slotA;
+            for (i++; i < md->name_count; i++)
+              {
+              slotB += md->name_entry_size;
+              if (STRCMP_UC_UC(slotA + IMM2_SIZE, slotB + IMM2_SIZE) == 0)
+                {
+                offset = GET2(slotB, 0) << 1;
+                condition = offset < offset_top &&
+                  md->offset_vector[offset] >= 0;
+                if (condition) break;
+                }
+              else break;
+              }
+            }
+          }
+        }
+
+      /* Chose branch according to the condition */
+
+      ecode += condition? 1 + IMM2_SIZE : GET(ecode, 1);
+      }
+
+    else if (condcode == OP_DEF)     /* DEFINE - always false */
+      {
+      condition = FALSE;
+      ecode += GET(ecode, 1);
+      }
+
+    /* The condition is an assertion. Call match() to evaluate it - setting
+    md->match_function_type to MATCH_CONDASSERT causes it to stop at the end of
+    an assertion. */
+
+    else
+      {
+      md->match_function_type = MATCH_CONDASSERT;
+      RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, NULL, RM3);
+      if (rrc == MATCH_MATCH)
+        {
+        if (md->end_offset_top > offset_top)
+          offset_top = md->end_offset_top;  /* Captures may have happened */
+        condition = TRUE;
+        ecode += 1 + LINK_SIZE + GET(ecode, LINK_SIZE + 2);
+        while (*ecode == OP_ALT) ecode += GET(ecode, 1);
+        }
+
+      /* PCRE doesn't allow the effect of (*THEN) to escape beyond an
+      assertion; it is therefore treated as NOMATCH. */
+
+      else if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN)
+        {
+        RRETURN(rrc);         /* Need braces because of following else */
+        }
+      else
+        {
+        condition = FALSE;
+        ecode += codelink;
+        }
+      }
+
+    /* We are now at the branch that is to be obeyed. As there is only one, can
+    use tail recursion to avoid using another stack frame, except when there is
+    unlimited repeat of a possibly empty group. In the latter case, a recursive
+    call to match() is always required, unless the second alternative doesn't
+    exist, in which case we can just plough on. Note that, for compatibility
+    with Perl, the | in a conditional group is NOT treated as creating two
+    alternatives. If a THEN is encountered in the branch, it propagates out to
+    the enclosing alternative (unless nested in a deeper set of alternatives,
+    of course). */
+
+    if (condition || *ecode == OP_ALT)
+      {
+      if (op != OP_SCOND)
+        {
+        ecode += 1 + LINK_SIZE;
+        goto TAIL_RECURSE;
+        }
+
+      md->match_function_type = MATCH_CBEGROUP;
+      RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, eptrb, RM49);
+      RRETURN(rrc);
+      }
+
+     /* Condition false & no alternative; continue after the group. */
+
+    else
+      {
+      ecode += 1 + LINK_SIZE;
+      }
+    break;
+
+
+    /* Before OP_ACCEPT there may be any number of OP_CLOSE opcodes,
+    to close any currently open capturing brackets. */
+
+    case OP_CLOSE:
+    number = GET2(ecode, 1);   /* Must be less than 65536 */
+    offset = number << 1;
+
+#ifdef PCRE_DEBUG
+      printf("end bracket %d at *ACCEPT", number);
+      printf("\n");
+#endif
+
+    md->capture_last = (md->capture_last & OVFLMASK) | number;
+    if (offset >= md->offset_max) md->capture_last |= OVFLBIT; else
+      {
+      md->offset_vector[offset] =
+        md->offset_vector[md->offset_end - number];
+      md->offset_vector[offset+1] = (int)(eptr - md->start_subject);
+      if (offset_top <= offset) offset_top = offset + 2;
+      }
+    ecode += 1 + IMM2_SIZE;
+    break;
+
+
+    /* End of the pattern, either real or forced. */
+
+    case OP_END:
+    case OP_ACCEPT:
+    case OP_ASSERT_ACCEPT:
+
+    /* If we have matched an empty string, fail if not in an assertion and not
+    in a recursion if either PCRE_NOTEMPTY is set, or if PCRE_NOTEMPTY_ATSTART
+    is set and we have matched at the start of the subject. In both cases,
+    backtracking will then try other alternatives, if any. */
+
+    if (eptr == mstart && op != OP_ASSERT_ACCEPT &&
+         md->recursive == NULL &&
+         (md->notempty ||
+           (md->notempty_atstart &&
+             mstart == md->start_subject + md->start_offset)))
+      RRETURN(MATCH_NOMATCH);
+
+    /* Otherwise, we have a match. */
+
+    md->end_match_ptr = eptr;           /* Record where we ended */
+    md->end_offset_top = offset_top;    /* and how many extracts were taken */
+    md->start_match_ptr = mstart;       /* and the start (\K can modify) */
+
+    /* For some reason, the macros don't work properly if an expression is
+    given as the argument to RRETURN when the heap is in use. */
+
+    rrc = (op == OP_END)? MATCH_MATCH : MATCH_ACCEPT;
+    RRETURN(rrc);
+
+    /* Assertion brackets. Check the alternative branches in turn - the
+    matching won't pass the KET for an assertion. If any one branch matches,
+    the assertion is true. Lookbehind assertions have an OP_REVERSE item at the
+    start of each branch to move the current point backwards, so the code at
+    this level is identical to the lookahead case. When the assertion is part
+    of a condition, we want to return immediately afterwards. The caller of
+    this incarnation of the match() function will have set MATCH_CONDASSERT in
+    md->match_function type, and one of these opcodes will be the first opcode
+    that is processed. We use a local variable that is preserved over calls to
+    match() to remember this case. */
+
+    case OP_ASSERT:
+    case OP_ASSERTBACK:
+    save_mark = md->mark;
+    if (md->match_function_type == MATCH_CONDASSERT)
+      {
+      condassert = TRUE;
+      md->match_function_type = 0;
+      }
+    else condassert = FALSE;
+
+    /* Loop for each branch */
+
+    do
+      {
+      RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, NULL, RM4);
+
+      /* A match means that the assertion is true; break out of the loop
+      that matches its alternatives. */
+
+      if (rrc == MATCH_MATCH || rrc == MATCH_ACCEPT)
+        {
+        mstart = md->start_match_ptr;   /* In case \K reset it */
+        break;
+        }
+
+      /* If not matched, restore the previous mark setting. */
+
+      md->mark = save_mark;
+
+      /* See comment in the code for capturing groups above about handling
+      THEN. */
+
+      if (rrc == MATCH_THEN)
+        {
+        next = ecode + GET(ecode,1);
+        if (md->start_match_ptr < next &&
+            (*ecode == OP_ALT || *next == OP_ALT))
+          rrc = MATCH_NOMATCH;
+        }
+
+      /* Anything other than NOMATCH causes the entire assertion to fail,
+      passing back the return code. This includes COMMIT, SKIP, PRUNE and an
+      uncaptured THEN, which means they take their normal effect. This
+      consistent approach does not always have exactly the same effect as in
+      Perl. */
+
+      if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+      ecode += GET(ecode, 1);
+      }
+    while (*ecode == OP_ALT);   /* Continue for next alternative */
+
+    /* If we have tried all the alternative branches, the assertion has
+    failed. If not, we broke out after a match. */
+
+    if (*ecode == OP_KET) RRETURN(MATCH_NOMATCH);
+
+    /* If checking an assertion for a condition, return MATCH_MATCH. */
+
+    if (condassert) RRETURN(MATCH_MATCH);
+
+    /* Continue from after a successful assertion, updating the offsets high
+    water mark, since extracts may have been taken during the assertion. */
+
+    do ecode += GET(ecode,1); while (*ecode == OP_ALT);
+    ecode += 1 + LINK_SIZE;
+    offset_top = md->end_offset_top;
+    continue;
+
+    /* Negative assertion: all branches must fail to match for the assertion to
+    succeed. */
+
+    case OP_ASSERT_NOT:
+    case OP_ASSERTBACK_NOT:
+    save_mark = md->mark;
+    if (md->match_function_type == MATCH_CONDASSERT)
+      {
+      condassert = TRUE;
+      md->match_function_type = 0;
+      }
+    else condassert = FALSE;
+
+    /* Loop for each alternative branch. */
+
+    do
+      {
+      RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, NULL, RM5);
+      md->mark = save_mark;   /* Always restore the mark setting */
+
+      switch(rrc)
+        {
+        case MATCH_MATCH:            /* A successful match means */
+        case MATCH_ACCEPT:           /* the assertion has failed. */
+        RRETURN(MATCH_NOMATCH);
+
+        case MATCH_NOMATCH:          /* Carry on with next branch */
+        break;
+
+        /* See comment in the code for capturing groups above about handling
+        THEN. */
+
+        case MATCH_THEN:
+        next = ecode + GET(ecode,1);
+        if (md->start_match_ptr < next &&
+            (*ecode == OP_ALT || *next == OP_ALT))
+          {
+          rrc = MATCH_NOMATCH;
+          break;
+          }
+        /* Otherwise fall through. */
+
+        /* COMMIT, SKIP, PRUNE, and an uncaptured THEN cause the whole
+        assertion to fail to match, without considering any more alternatives.
+        Failing to match means the assertion is true. This is a consistent
+        approach, but does not always have the same effect as in Perl. */
+
+        case MATCH_COMMIT:
+        case MATCH_SKIP:
+        case MATCH_SKIP_ARG:
+        case MATCH_PRUNE:
+        do ecode += GET(ecode,1); while (*ecode == OP_ALT);
+        goto NEG_ASSERT_TRUE;   /* Break out of alternation loop */
+
+        /* Anything else is an error */
+
+        default:
+        RRETURN(rrc);
+        }
+
+      /* Continue with next branch */
+
+      ecode += GET(ecode,1);
+      }
+    while (*ecode == OP_ALT);
+
+    /* All branches in the assertion failed to match. */
+
+    NEG_ASSERT_TRUE:
+    if (condassert) RRETURN(MATCH_MATCH);  /* Condition assertion */
+    ecode += 1 + LINK_SIZE;                /* Continue with current branch */
+    continue;
+
+    /* Move the subject pointer back. This occurs only at the start of
+    each branch of a lookbehind assertion. If we are too close to the start to
+    move back, this match function fails. When working with UTF-8 we move
+    back a number of characters, not bytes. */
+
+    case OP_REVERSE:
+#ifdef SUPPORT_UTF
+    if (utf)
+      {
+      i = GET(ecode, 1);
+      while (i-- > 0)
+        {
+        eptr--;
+        if (eptr < md->start_subject) RRETURN(MATCH_NOMATCH);
+        BACKCHAR(eptr);
+        }
+      }
+    else
+#endif
+
+    /* No UTF-8 support, or not in UTF-8 mode: count is byte count */
+
+      {
+      eptr -= GET(ecode, 1);
+      if (eptr < md->start_subject) RRETURN(MATCH_NOMATCH);
+      }
+
+    /* Save the earliest consulted character, then skip to next op code */
+
+    if (eptr < md->start_used_ptr) md->start_used_ptr = eptr;
+    ecode += 1 + LINK_SIZE;
+    break;
+
+    /* The callout item calls an external function, if one is provided, passing
+    details of the match so far. This is mainly for debugging, though the
+    function is able to force a failure. */
+
+    case OP_CALLOUT:
+    if (PUBL(callout) != NULL)
+      {
+      PUBL(callout_block) cb;
+      cb.version          = 2;   /* Version 1 of the callout block */
+      cb.callout_number   = ecode[1];
+      cb.offset_vector    = md->offset_vector;
+#if defined COMPILE_PCRE8
+      cb.subject          = (PCRE_SPTR)md->start_subject;
+#elif defined COMPILE_PCRE16
+      cb.subject          = (PCRE_SPTR16)md->start_subject;
+#elif defined COMPILE_PCRE32
+      cb.subject          = (PCRE_SPTR32)md->start_subject;
+#endif
+      cb.subject_length   = (int)(md->end_subject - md->start_subject);
+      cb.start_match      = (int)(mstart - md->start_subject);
+      cb.current_position = (int)(eptr - md->start_subject);
+      cb.pattern_position = GET(ecode, 2);
+      cb.next_item_length = GET(ecode, 2 + LINK_SIZE);
+      cb.capture_top      = offset_top/2;
+      cb.capture_last     = md->capture_last & CAPLMASK;
+      /* Internal change requires this for API compatibility. */
+      if (cb.capture_last == 0) cb.capture_last = -1;
+      cb.callout_data     = md->callout_data;
+      cb.mark             = md->nomatch_mark;
+      if ((rrc = (*PUBL(callout))(&cb)) > 0) RRETURN(MATCH_NOMATCH);
+      if (rrc < 0) RRETURN(rrc);
+      }
+    ecode += 2 + 2*LINK_SIZE;
+    break;
+
+    /* Recursion either matches the current regex, or some subexpression. The
+    offset data is the offset to the starting bracket from the start of the
+    whole pattern. (This is so that it works from duplicated subpatterns.)
+
+    The state of the capturing groups is preserved over recursion, and
+    re-instated afterwards. We don't know how many are started and not yet
+    finished (offset_top records the completed total) so we just have to save
+    all the potential data. There may be up to 65535 such values, which is too
+    large to put on the stack, but using malloc for small numbers seems
+    expensive. As a compromise, the stack is used when there are no more than
+    REC_STACK_SAVE_MAX values to store; otherwise malloc is used.
+
+    There are also other values that have to be saved. We use a chained
+    sequence of blocks that actually live on the stack. Thanks to Robin Houston
+    for the original version of this logic. It has, however, been hacked around
+    a lot, so he is not to blame for the current way it works. */
+
+    case OP_RECURSE:
+      {
+      recursion_info *ri;
+      unsigned int recno;
+
+      callpat = md->start_code + GET(ecode, 1);
+      recno = (callpat == md->start_code)? 0 :
+        GET2(callpat, 1 + LINK_SIZE);
+
+      /* Check for repeating a recursion without advancing the subject pointer.
+      This should catch convoluted mutual recursions. (Some simple cases are
+      caught at compile time.) */
+
+      for (ri = md->recursive; ri != NULL; ri = ri->prevrec)
+        if (recno == ri->group_num && eptr == ri->subject_position)
+          RRETURN(PCRE_ERROR_RECURSELOOP);
+
+      /* Add to "recursing stack" */
+
+      new_recursive.group_num = recno;
+      new_recursive.saved_capture_last = md->capture_last;
+      new_recursive.subject_position = eptr;
+      new_recursive.prevrec = md->recursive;
+      md->recursive = &new_recursive;
+
+      /* Where to continue from afterwards */
+
+      ecode += 1 + LINK_SIZE;
+
+      /* Now save the offset data */
+
+      new_recursive.saved_max = md->offset_end;
+      if (new_recursive.saved_max <= REC_STACK_SAVE_MAX)
+        new_recursive.offset_save = stacksave;
+      else
+        {
+        new_recursive.offset_save =
+          (int *)(PUBL(malloc))(new_recursive.saved_max * sizeof(int));
+        if (new_recursive.offset_save == NULL) RRETURN(PCRE_ERROR_NOMEMORY);
+        }
+      memcpy(new_recursive.offset_save, md->offset_vector,
+            new_recursive.saved_max * sizeof(int));
+
+      /* OK, now we can do the recursion. After processing each alternative,
+      restore the offset data and the last captured value. If there were nested
+      recursions, md->recursive might be changed, so reset it before looping.
+      */
+
+      DPRINTF(("Recursing into group %d\n", new_recursive.group_num));
+      cbegroup = (*callpat >= OP_SBRA);
+      do
+        {
+        if (cbegroup) md->match_function_type = MATCH_CBEGROUP;
+        RMATCH(eptr, callpat + PRIV(OP_lengths)[*callpat], offset_top,
+          md, eptrb, RM6);
+        memcpy(md->offset_vector, new_recursive.offset_save,
+            new_recursive.saved_max * sizeof(int));
+        md->capture_last = new_recursive.saved_capture_last;
+        md->recursive = new_recursive.prevrec;
+        if (rrc == MATCH_MATCH || rrc == MATCH_ACCEPT)
+          {
+          DPRINTF(("Recursion matched\n"));
+          if (new_recursive.offset_save != stacksave)
+            (PUBL(free))(new_recursive.offset_save);
+
+          /* Set where we got to in the subject, and reset the start in case
+          it was changed by \K. This *is* propagated back out of a recursion,
+          for Perl compatibility. */
+
+          eptr = md->end_match_ptr;
+          mstart = md->start_match_ptr;
+          goto RECURSION_MATCHED;        /* Exit loop; end processing */
+          }
+
+        /* PCRE does not allow THEN, SKIP, PRUNE or COMMIT to escape beyond a
+        recursion; they cause a NOMATCH for the entire recursion. These codes
+        are defined in a range that can be tested for. */
+
+        if (rrc >= MATCH_BACKTRACK_MIN && rrc <= MATCH_BACKTRACK_MAX)
+          RRETURN(MATCH_NOMATCH);
+
+        /* Any return code other than NOMATCH is an error. */
+
+        if (rrc != MATCH_NOMATCH)
+          {
+          DPRINTF(("Recursion gave error %d\n", rrc));
+          if (new_recursive.offset_save != stacksave)
+            (PUBL(free))(new_recursive.offset_save);
+          RRETURN(rrc);
+          }
+
+        md->recursive = &new_recursive;
+        callpat += GET(callpat, 1);
+        }
+      while (*callpat == OP_ALT);
+
+      DPRINTF(("Recursion didn't match\n"));
+      md->recursive = new_recursive.prevrec;
+      if (new_recursive.offset_save != stacksave)
+        (PUBL(free))(new_recursive.offset_save);
+      RRETURN(MATCH_NOMATCH);
+      }
+
+    RECURSION_MATCHED:
+    break;
+
+    /* An alternation is the end of a branch; scan along to find the end of the
+    bracketed group and go to there. */
+
+    case OP_ALT:
+    do ecode += GET(ecode,1); while (*ecode == OP_ALT);
+    break;
+
+    /* BRAZERO, BRAMINZERO and SKIPZERO occur just before a bracket group,
+    indicating that it may occur zero times. It may repeat infinitely, or not
+    at all - i.e. it could be ()* or ()? or even (){0} in the pattern. Brackets
+    with fixed upper repeat limits are compiled as a number of copies, with the
+    optional ones preceded by BRAZERO or BRAMINZERO. */
+
+    case OP_BRAZERO:
+    next = ecode + 1;
+    RMATCH(eptr, next, offset_top, md, eptrb, RM10);
+    if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+    do next += GET(next, 1); while (*next == OP_ALT);
+    ecode = next + 1 + LINK_SIZE;
+    break;
+
+    case OP_BRAMINZERO:
+    next = ecode + 1;
+    do next += GET(next, 1); while (*next == OP_ALT);
+    RMATCH(eptr, next + 1+LINK_SIZE, offset_top, md, eptrb, RM11);
+    if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+    ecode++;
+    break;
+
+    case OP_SKIPZERO:
+    next = ecode+1;
+    do next += GET(next,1); while (*next == OP_ALT);
+    ecode = next + 1 + LINK_SIZE;
+    break;
+
+    /* BRAPOSZERO occurs before a possessive bracket group. Don't do anything
+    here; just jump to the group, with allow_zero set TRUE. */
+
+    case OP_BRAPOSZERO:
+    op = *(++ecode);
+    allow_zero = TRUE;
+    if (op == OP_CBRAPOS || op == OP_SCBRAPOS) goto POSSESSIVE_CAPTURE;
+      goto POSSESSIVE_NON_CAPTURE;
+
+    /* End of a group, repeated or non-repeating. */
+
+    case OP_KET:
+    case OP_KETRMIN:
+    case OP_KETRMAX:
+    case OP_KETRPOS:
+    prev = ecode - GET(ecode, 1);
+
+    /* If this was a group that remembered the subject start, in order to break
+    infinite repeats of empty string matches, retrieve the subject start from
+    the chain. Otherwise, set it NULL. */
+
+    if (*prev >= OP_SBRA || *prev == OP_ONCE)
+      {
+      saved_eptr = eptrb->epb_saved_eptr;   /* Value at start of group */
+      eptrb = eptrb->epb_prev;              /* Backup to previous group */
+      }
+    else saved_eptr = NULL;
+
+    /* If we are at the end of an assertion group or a non-capturing atomic
+    group, stop matching and return MATCH_MATCH, but record the current high
+    water mark for use by positive assertions. We also need to record the match
+    start in case it was changed by \K. */
+
+    if ((*prev >= OP_ASSERT && *prev <= OP_ASSERTBACK_NOT) ||
+         *prev == OP_ONCE_NC)
+      {
+      md->end_match_ptr = eptr;      /* For ONCE_NC */
+      md->end_offset_top = offset_top;
+      md->start_match_ptr = mstart;
+      RRETURN(MATCH_MATCH);         /* Sets md->mark */
+      }
+
+    /* For capturing groups we have to check the group number back at the start
+    and if necessary complete handling an extraction by setting the offsets and
+    bumping the high water mark. Whole-pattern recursion is coded as a recurse
+    into group 0, so it won't be picked up here. Instead, we catch it when the
+    OP_END is reached. Other recursion is handled here. We just have to record
+    the current subject position and start match pointer and give a MATCH
+    return. */
+
+    if (*prev == OP_CBRA || *prev == OP_SCBRA ||
+        *prev == OP_CBRAPOS || *prev == OP_SCBRAPOS)
+      {
+      number = GET2(prev, 1+LINK_SIZE);
+      offset = number << 1;
+
+#ifdef PCRE_DEBUG
+      printf("end bracket %d", number);
+      printf("\n");
+#endif
+
+      /* Handle a recursively called group. */
+
+      if (md->recursive != NULL && md->recursive->group_num == number)
+        {
+        md->end_match_ptr = eptr;
+        md->start_match_ptr = mstart;
+        RRETURN(MATCH_MATCH);
+        }
+
+      /* Deal with capturing */
+
+      md->capture_last = (md->capture_last & OVFLMASK) | number;
+      if (offset >= md->offset_max) md->capture_last |= OVFLBIT; else
+        {
+        /* If offset is greater than offset_top, it means that we are
+        "skipping" a capturing group, and that group's offsets must be marked
+        unset. In earlier versions of PCRE, all the offsets were unset at the
+        start of matching, but this doesn't work because atomic groups and
+        assertions can cause a value to be set that should later be unset.
+        Example: matching /(?>(a))b|(a)c/ against "ac". This sets group 1 as
+        part of the atomic group, but this is not on the final matching path,
+        so must be unset when 2 is set. (If there is no group 2, there is no
+        problem, because offset_top will then be 2, indicating no capture.) */
+
+        if (offset > offset_top)
+          {
+          register int *iptr = md->offset_vector + offset_top;
+          register int *iend = md->offset_vector + offset;
+          while (iptr < iend) *iptr++ = -1;
+          }
+
+        /* Now make the extraction */
+
+        md->offset_vector[offset] =
+          md->offset_vector[md->offset_end - number];
+        md->offset_vector[offset+1] = (int)(eptr - md->start_subject);
+        if (offset_top <= offset) offset_top = offset + 2;
+        }
+      }
+
+    /* For an ordinary non-repeating ket, just continue at this level. This
+    also happens for a repeating ket if no characters were matched in the
+    group. This is the forcible breaking of infinite loops as implemented in
+    Perl 5.005. For a non-repeating atomic group that includes captures,
+    establish a backup point by processing the rest of the pattern at a lower
+    level. If this results in a NOMATCH return, pass MATCH_ONCE back to the
+    original OP_ONCE level, thereby bypassing intermediate backup points, but
+    resetting any captures that happened along the way. */
+
+    if (*ecode == OP_KET || eptr == saved_eptr)
+      {
+      if (*prev == OP_ONCE)
+        {
+        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, eptrb, RM12);
+        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+        md->once_target = prev;  /* Level at which to change to MATCH_NOMATCH */
+        RRETURN(MATCH_ONCE);
+        }
+      ecode += 1 + LINK_SIZE;    /* Carry on at this level */
+      break;
+      }
+
+    /* OP_KETRPOS is a possessive repeating ket. Remember the current position,
+    and return the MATCH_KETRPOS. This makes it possible to do the repeats one
+    at a time from the outer level, thus saving stack. */
+
+    if (*ecode == OP_KETRPOS)
+      {
+      md->end_match_ptr = eptr;
+      md->end_offset_top = offset_top;
+      RRETURN(MATCH_KETRPOS);
+      }
+
+    /* The normal repeating kets try the rest of the pattern or restart from
+    the preceding bracket, in the appropriate order. In the second case, we can
+    use tail recursion to avoid using another stack frame, unless we have an
+    an atomic group or an unlimited repeat of a group that can match an empty
+    string. */
+
+    if (*ecode == OP_KETRMIN)
+      {
+      RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, eptrb, RM7);
+      if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+      if (*prev == OP_ONCE)
+        {
+        RMATCH(eptr, prev, offset_top, md, eptrb, RM8);
+        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+        md->once_target = prev;  /* Level at which to change to MATCH_NOMATCH */
+        RRETURN(MATCH_ONCE);
+        }
+      if (*prev >= OP_SBRA)    /* Could match an empty string */
+        {
+        RMATCH(eptr, prev, offset_top, md, eptrb, RM50);
+        RRETURN(rrc);
+        }
+      ecode = prev;
+      goto TAIL_RECURSE;
+      }
+    else  /* OP_KETRMAX */
+      {
+      RMATCH(eptr, prev, offset_top, md, eptrb, RM13);
+      if (rrc == MATCH_ONCE && md->once_target == prev) rrc = MATCH_NOMATCH;
+      if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+      if (*prev == OP_ONCE)
+        {
+        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, eptrb, RM9);
+        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+        md->once_target = prev;
+        RRETURN(MATCH_ONCE);
+        }
+      ecode += 1 + LINK_SIZE;
+      goto TAIL_RECURSE;
+      }
+    /* Control never gets here */
+
+    /* Not multiline mode: start of subject assertion, unless notbol. */
+
+    case OP_CIRC:
+    if (md->notbol && eptr == md->start_subject) RRETURN(MATCH_NOMATCH);
+
+    /* Start of subject assertion */
+
+    case OP_SOD:
+    if (eptr != md->start_subject) RRETURN(MATCH_NOMATCH);
+    ecode++;
+    break;
+
+    /* Multiline mode: start of subject unless notbol, or after any newline. */
+
+    case OP_CIRCM:
+    if (md->notbol && eptr == md->start_subject) RRETURN(MATCH_NOMATCH);
+    if (eptr != md->start_subject &&
+        (eptr == md->end_subject || !WAS_NEWLINE(eptr)))
+      RRETURN(MATCH_NOMATCH);
+    ecode++;
+    break;
+
+    /* Start of match assertion */
+
+    case OP_SOM:
+    if (eptr != md->start_subject + md->start_offset) RRETURN(MATCH_NOMATCH);
+    ecode++;
+    break;
+
+    /* Reset the start of match point */
+
+    case OP_SET_SOM:
+    mstart = eptr;
+    ecode++;
+    break;
+
+    /* Multiline mode: assert before any newline, or before end of subject
+    unless noteol is set. */
+
+    case OP_DOLLM:
+    if (eptr < md->end_subject)
+      {
+      if (!IS_NEWLINE(eptr))
+        {
+        if (md->partial != 0 &&
+            eptr + 1 >= md->end_subject &&
+            NLBLOCK->nltype == NLTYPE_FIXED &&
+            NLBLOCK->nllen == 2 &&
+            RAWUCHARTEST(eptr) == NLBLOCK->nl[0])
+          {
+          md->hitend = TRUE;
+          if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);
+          }
+        RRETURN(MATCH_NOMATCH);
+        }
+      }
+    else
+      {
+      if (md->noteol) RRETURN(MATCH_NOMATCH);
+      SCHECK_PARTIAL();
+      }
+    ecode++;
+    break;
+
+    /* Not multiline mode: assert before a terminating newline or before end of
+    subject unless noteol is set. */
+
+    case OP_DOLL:
+    if (md->noteol) RRETURN(MATCH_NOMATCH);
+    if (!md->endonly) goto ASSERT_NL_OR_EOS;
+
+    /* ... else fall through for endonly */
+
+    /* End of subject assertion (\z) */
+
+    case OP_EOD:
+    if (eptr < md->end_subject) RRETURN(MATCH_NOMATCH);
+    SCHECK_PARTIAL();
+    ecode++;
+    break;
+
+    /* End of subject or ending \n assertion (\Z) */
+
+    case OP_EODN:
+    ASSERT_NL_OR_EOS:
+    if (eptr < md->end_subject &&
+        (!IS_NEWLINE(eptr) || eptr != md->end_subject - md->nllen))
+      {
+      if (md->partial != 0 &&
+          eptr + 1 >= md->end_subject &&
+          NLBLOCK->nltype == NLTYPE_FIXED &&
+          NLBLOCK->nllen == 2 &&
+          RAWUCHARTEST(eptr) == NLBLOCK->nl[0])
+        {
+        md->hitend = TRUE;
+        if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);
+        }
+      RRETURN(MATCH_NOMATCH);
+      }
+
+    /* Either at end of string or \n before end. */
+
+    SCHECK_PARTIAL();
+    ecode++;
+    break;
+
+    /* Word boundary assertions */
+
+    case OP_NOT_WORD_BOUNDARY:
+    case OP_WORD_BOUNDARY:
+      {
+
+      /* Find out if the previous and current characters are "word" characters.
+      It takes a bit more work in UTF-8 mode. Characters > 255 are assumed to
+      be "non-word" characters. Remember the earliest consulted character for
+      partial matching. */
+
+#ifdef SUPPORT_UTF
+      if (utf)
+        {
+        /* Get status of previous character */
+
+        if (eptr == md->start_subject) prev_is_word = FALSE; else
+          {
+          PCRE_PUCHAR lastptr = eptr - 1;
+          BACKCHAR(lastptr);
+          if (lastptr < md->start_used_ptr) md->start_used_ptr = lastptr;
+          GETCHAR(c, lastptr);
+#ifdef SUPPORT_UCP
+          if (md->use_ucp)
+            {
+            if (c == '_') prev_is_word = TRUE; else
+              {
+              int cat = UCD_CATEGORY(c);
+              prev_is_word = (cat == ucp_L || cat == ucp_N);
+              }
+            }
+          else
+#endif
+          prev_is_word = c < 256 && (md->ctypes[c] & ctype_word) != 0;
+          }
+
+        /* Get status of next character */
+
+        if (eptr >= md->end_subject)
+          {
+          SCHECK_PARTIAL();
+          cur_is_word = FALSE;
+          }
+        else
+          {
+          GETCHAR(c, eptr);
+#ifdef SUPPORT_UCP
+          if (md->use_ucp)
+            {
+            if (c == '_') cur_is_word = TRUE; else
+              {
+              int cat = UCD_CATEGORY(c);
+              cur_is_word = (cat == ucp_L || cat == ucp_N);
+              }
+            }
+          else
+#endif
+          cur_is_word = c < 256 && (md->ctypes[c] & ctype_word) != 0;
+          }
+        }
+      else
+#endif
+
+      /* Not in UTF-8 mode, but we may still have PCRE_UCP set, and for
+      consistency with the behaviour of \w we do use it in this case. */
+
+        {
+        /* Get status of previous character */
+
+        if (eptr == md->start_subject) prev_is_word = FALSE; else
+          {
+          if (eptr <= md->start_used_ptr) md->start_used_ptr = eptr - 1;
+#ifdef SUPPORT_UCP
+          if (md->use_ucp)
+            {
+            c = eptr[-1];
+            if (c == '_') prev_is_word = TRUE; else
+              {
+              int cat = UCD_CATEGORY(c);
+              prev_is_word = (cat == ucp_L || cat == ucp_N);
+              }
+            }
+          else
+#endif
+          prev_is_word = MAX_255(eptr[-1])
+            && ((md->ctypes[eptr[-1]] & ctype_word) != 0);
+          }
+
+        /* Get status of next character */
+
+        if (eptr >= md->end_subject)
+          {
+          SCHECK_PARTIAL();
+          cur_is_word = FALSE;
+          }
+        else
+#ifdef SUPPORT_UCP
+        if (md->use_ucp)
+          {
+          c = *eptr;
+          if (c == '_') cur_is_word = TRUE; else
+            {
+            int cat = UCD_CATEGORY(c);
+            cur_is_word = (cat == ucp_L || cat == ucp_N);
+            }
+          }
+        else
+#endif
+        cur_is_word = MAX_255(*eptr)
+          && ((md->ctypes[*eptr] & ctype_word) != 0);
+        }
+
+      /* Now see if the situation is what we want */
+
+      if ((*ecode++ == OP_WORD_BOUNDARY)?
+           cur_is_word == prev_is_word : cur_is_word != prev_is_word)
+        RRETURN(MATCH_NOMATCH);
+      }
+    break;
+
+    /* Match any single character type except newline; have to take care with
+    CRLF newlines and partial matching. */
+
+    case OP_ANY:
+    if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH);
+    if (md->partial != 0 &&
+        eptr + 1 >= md->end_subject &&
+        NLBLOCK->nltype == NLTYPE_FIXED &&
+        NLBLOCK->nllen == 2 &&
+        RAWUCHARTEST(eptr) == NLBLOCK->nl[0])
+      {
+      md->hitend = TRUE;
+      if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);
+      }
+
+    /* Fall through */
+
+    /* Match any single character whatsoever. */
+
+    case OP_ALLANY:
+    if (eptr >= md->end_subject)   /* DO NOT merge the eptr++ here; it must */
+      {                            /* not be updated before SCHECK_PARTIAL. */
+      SCHECK_PARTIAL();
+      RRETURN(MATCH_NOMATCH);
+      }
+    eptr++;
+#ifdef SUPPORT_UTF
+    if (utf) ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++);
+#endif
+    ecode++;
+    break;
+
+    /* Match a single byte, even in UTF-8 mode. This opcode really does match
+    any byte, even newline, independent of the setting of PCRE_DOTALL. */
+
+    case OP_ANYBYTE:
+    if (eptr >= md->end_subject)   /* DO NOT merge the eptr++ here; it must */
+      {                            /* not be updated before SCHECK_PARTIAL. */
+      SCHECK_PARTIAL();
+      RRETURN(MATCH_NOMATCH);
+      }
+    eptr++;
+    ecode++;
+    break;
+
+    case OP_NOT_DIGIT:
+    if (eptr >= md->end_subject)
+      {
+      SCHECK_PARTIAL();
+      RRETURN(MATCH_NOMATCH);
+      }
+    GETCHARINCTEST(c, eptr);
+    if (
+#if defined SUPPORT_UTF || !(defined COMPILE_PCRE8)
+       c < 256 &&
+#endif
+       (md->ctypes[c] & ctype_digit) != 0
+       )
+      RRETURN(MATCH_NOMATCH);
+    ecode++;
+    break;
+
+    case OP_DIGIT:
+    if (eptr >= md->end_subject)
+      {
+      SCHECK_PARTIAL();
+      RRETURN(MATCH_NOMATCH);
+      }
+    GETCHARINCTEST(c, eptr);
+    if (
+#if defined SUPPORT_UTF || !(defined COMPILE_PCRE8)
+       c > 255 ||
+#endif
+       (md->ctypes[c] & ctype_digit) == 0
+       )
+      RRETURN(MATCH_NOMATCH);
+    ecode++;
+    break;
+
+    case OP_NOT_WHITESPACE:
+    if (eptr >= md->end_subject)
+      {
+      SCHECK_PARTIAL();
+      RRETURN(MATCH_NOMATCH);
+      }
+    GETCHARINCTEST(c, eptr);
+    if (
+#if defined SUPPORT_UTF || !(defined COMPILE_PCRE8)
+       c < 256 &&
+#endif
+       (md->ctypes[c] & ctype_space) != 0
+       )
+      RRETURN(MATCH_NOMATCH);
+    ecode++;
+    break;
+
+    case OP_WHITESPACE:
+    if (eptr >= md->end_subject)
+      {
+      SCHECK_PARTIAL();
+      RRETURN(MATCH_NOMATCH);
+      }
+    GETCHARINCTEST(c, eptr);
+    if (
+#if defined SUPPORT_UTF || !(defined COMPILE_PCRE8)
+       c > 255 ||
+#endif
+       (md->ctypes[c] & ctype_space) == 0
+       )
+      RRETURN(MATCH_NOMATCH);
+    ecode++;
+    break;
+
+    case OP_NOT_WORDCHAR:
+    if (eptr >= md->end_subject)
+      {
+      SCHECK_PARTIAL();
+      RRETURN(MATCH_NOMATCH);
+      }
+    GETCHARINCTEST(c, eptr);
+    if (
+#if defined SUPPORT_UTF || !(defined COMPILE_PCRE8)
+       c < 256 &&
+#endif
+       (md->ctypes[c] & ctype_word) != 0
+       )
+      RRETURN(MATCH_NOMATCH);
+    ecode++;
+    break;
+
+    case OP_WORDCHAR:
+    if (eptr >= md->end_subject)
+      {
+      SCHECK_PARTIAL();
+      RRETURN(MATCH_NOMATCH);
+      }
+    GETCHARINCTEST(c, eptr);
+    if (
+#if defined SUPPORT_UTF || !(defined COMPILE_PCRE8)
+       c > 255 ||
+#endif
+       (md->ctypes[c] & ctype_word) == 0
+       )
+      RRETURN(MATCH_NOMATCH);
+    ecode++;
+    break;
+
+    case OP_ANYNL:
+    if (eptr >= md->end_subject)
+      {
+      SCHECK_PARTIAL();
+      RRETURN(MATCH_NOMATCH);
+      }
+    GETCHARINCTEST(c, eptr);
+    switch(c)
+      {
+      default: RRETURN(MATCH_NOMATCH);
+
+      case CHAR_CR:
+      if (eptr >= md->end_subject)
+        {
+        SCHECK_PARTIAL();
+        }
+      else if (RAWUCHARTEST(eptr) == CHAR_LF) eptr++;
+      break;
+
+      case CHAR_LF:
+      break;
+
+      case CHAR_VT:
+      case CHAR_FF:
+      case CHAR_NEL:
+#ifndef EBCDIC
+      case 0x2028:
+      case 0x2029:
+#endif  /* Not EBCDIC */
+      if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
+      break;
+      }
+    ecode++;
+    break;
+
+    case OP_NOT_HSPACE:
+    if (eptr >= md->end_subject)
+      {
+      SCHECK_PARTIAL();
+      RRETURN(MATCH_NOMATCH);
+      }
+    GETCHARINCTEST(c, eptr);
+    switch(c)
+      {
+      HSPACE_CASES: RRETURN(MATCH_NOMATCH);  /* Byte and multibyte cases */
+      default: break;
+      }
+    ecode++;
+    break;
+
+    case OP_HSPACE:
+    if (eptr >= md->end_subject)
+      {
+      SCHECK_PARTIAL();
+      RRETURN(MATCH_NOMATCH);
+      }
+    GETCHARINCTEST(c, eptr);
+    switch(c)
+      {
+      HSPACE_CASES: break;  /* Byte and multibyte cases */
+      default: RRETURN(MATCH_NOMATCH);
+      }
+    ecode++;
+    break;
+
+    case OP_NOT_VSPACE:
+    if (eptr >= md->end_subject)
+      {
+      SCHECK_PARTIAL();
+      RRETURN(MATCH_NOMATCH);
+      }
+    GETCHARINCTEST(c, eptr);
+    switch(c)
+      {
+      VSPACE_CASES: RRETURN(MATCH_NOMATCH);
+      default: break;
+      }
+    ecode++;
+    break;
+
+    case OP_VSPACE:
+    if (eptr >= md->end_subject)
+      {
+      SCHECK_PARTIAL();
+      RRETURN(MATCH_NOMATCH);
+      }
+    GETCHARINCTEST(c, eptr);
+    switch(c)
+      {
+      VSPACE_CASES: break;
+      default: RRETURN(MATCH_NOMATCH);
+      }
+    ecode++;
+    break;
+
+#ifdef SUPPORT_UCP
+    /* Check the next character by Unicode property. We will get here only
+    if the support is in the binary; otherwise a compile-time error occurs. */
+
+    case OP_PROP:
+    case OP_NOTPROP:
+    if (eptr >= md->end_subject)
+      {
+      SCHECK_PARTIAL();
+      RRETURN(MATCH_NOMATCH);
+      }
+    GETCHARINCTEST(c, eptr);
+      {
+      const pcre_uint32 *cp;
+      const ucd_record *prop = GET_UCD(c);
+
+      switch(ecode[1])
+        {
+        case PT_ANY:
+        if (op == OP_NOTPROP) RRETURN(MATCH_NOMATCH);
+        break;
+
+        case PT_LAMP:
+        if ((prop->chartype == ucp_Lu ||
+             prop->chartype == ucp_Ll ||
+             prop->chartype == ucp_Lt) == (op == OP_NOTPROP))
+          RRETURN(MATCH_NOMATCH);
+        break;
+
+        case PT_GC:
+        if ((ecode[2] != PRIV(ucp_gentype)[prop->chartype]) == (op == OP_PROP))
+          RRETURN(MATCH_NOMATCH);
+        break;
+
+        case PT_PC:
+        if ((ecode[2] != prop->chartype) == (op == OP_PROP))
+          RRETURN(MATCH_NOMATCH);
+        break;
+
+        case PT_SC:
+        if ((ecode[2] != prop->script) == (op == OP_PROP))
+          RRETURN(MATCH_NOMATCH);
+        break;
+
+        /* These are specials */
+
+        case PT_ALNUM:
+        if ((PRIV(ucp_gentype)[prop->chartype] == ucp_L ||
+             PRIV(ucp_gentype)[prop->chartype] == ucp_N) == (op == OP_NOTPROP))
+          RRETURN(MATCH_NOMATCH);
+        break;
+
+        case PT_SPACE:    /* Perl space */
+        if ((PRIV(ucp_gentype)[prop->chartype] == ucp_Z ||
+             c == CHAR_HT || c == CHAR_NL || c == CHAR_FF || c == CHAR_CR)
+               == (op == OP_NOTPROP))
+          RRETURN(MATCH_NOMATCH);
+        break;
+
+        case PT_PXSPACE:  /* POSIX space */
+        if ((PRIV(ucp_gentype)[prop->chartype] == ucp_Z ||
+             c == CHAR_HT || c == CHAR_NL || c == CHAR_VT ||
+             c == CHAR_FF || c == CHAR_CR)
+               == (op == OP_NOTPROP))
+          RRETURN(MATCH_NOMATCH);
+        break;
+
+        case PT_WORD:
+        if ((PRIV(ucp_gentype)[prop->chartype] == ucp_L ||
+             PRIV(ucp_gentype)[prop->chartype] == ucp_N ||
+             c == CHAR_UNDERSCORE) == (op == OP_NOTPROP))
+          RRETURN(MATCH_NOMATCH);
+        break;
+
+        case PT_CLIST:
+        cp = PRIV(ucd_caseless_sets) + ecode[2];
+        for (;;)
+          {
+          if (c < *cp)
+            { if (op == OP_PROP) { RRETURN(MATCH_NOMATCH); } else break; }
+          if (c == *cp++)
+            { if (op == OP_PROP) break; else { RRETURN(MATCH_NOMATCH); } }
+          }
+        break;
+
+        case PT_UCNC:
+        if ((c == CHAR_DOLLAR_SIGN || c == CHAR_COMMERCIAL_AT ||
+             c == CHAR_GRAVE_ACCENT || (c >= 0xa0 && c <= 0xd7ff) ||
+             c >= 0xe000) == (op == OP_NOTPROP))
+          RRETURN(MATCH_NOMATCH);
+        break;
+
+        /* This should never occur */
+
+        default:
+        RRETURN(PCRE_ERROR_INTERNAL);
+        }
+
+      ecode += 3;
+      }
+    break;
+
+    /* Match an extended Unicode sequence. We will get here only if the support
+    is in the binary; otherwise a compile-time error occurs. */
+
+    case OP_EXTUNI:
+    if (eptr >= md->end_subject)
+      {
+      SCHECK_PARTIAL();
+      RRETURN(MATCH_NOMATCH);
+      }
+    else
+      {
+      int lgb, rgb;
+      GETCHARINCTEST(c, eptr);
+      lgb = UCD_GRAPHBREAK(c);
+      while (eptr < md->end_subject)
+        {
+        int len = 1;
+        if (!utf) c = *eptr; else { GETCHARLEN(c, eptr, len); }
+        rgb = UCD_GRAPHBREAK(c);
+        if ((PRIV(ucp_gbtable)[lgb] & (1 << rgb)) == 0) break;
+        lgb = rgb;
+        eptr += len;
+        }
+      }
+    CHECK_PARTIAL();
+    ecode++;
+    break;
+#endif  /* SUPPORT_UCP */
+
+
+    /* Match a back reference, possibly repeatedly. Look past the end of the
+    item to see if there is repeat information following. The code is similar
+    to that for character classes, but repeated for efficiency. Then obey
+    similar code to character type repeats - written out again for speed.
+    However, if the referenced string is the empty string, always treat
+    it as matched, any number of times (otherwise there could be infinite
+    loops). */
+
+    case OP_REF:
+    case OP_REFI:
+    caseless = op == OP_REFI;
+    offset = GET2(ecode, 1) << 1;               /* Doubled ref number */
+    ecode += 1 + IMM2_SIZE;
+
+    /* If the reference is unset, there are two possibilities:
+
+    (a) In the default, Perl-compatible state, set the length negative;
+    this ensures that every attempt at a match fails. We can't just fail
+    here, because of the possibility of quantifiers with zero minima.
+
+    (b) If the JavaScript compatibility flag is set, set the length to zero
+    so that the back reference matches an empty string.
+
+    Otherwise, set the length to the length of what was matched by the
+    referenced subpattern. */
+
+    if (offset >= offset_top || md->offset_vector[offset] < 0)
+      length = (md->jscript_compat)? 0 : -1;
+    else
+      length = md->offset_vector[offset+1] - md->offset_vector[offset];
+
+    /* Set up for repetition, or handle the non-repeated case */
+
+    switch (*ecode)
+      {
+      case OP_CRSTAR:
+      case OP_CRMINSTAR:
+      case OP_CRPLUS:
+      case OP_CRMINPLUS:
+      case OP_CRQUERY:
+      case OP_CRMINQUERY:
+      c = *ecode++ - OP_CRSTAR;
+      minimize = (c & 1) != 0;
+      min = rep_min[c];                 /* Pick up values from tables; */
+      max = rep_max[c];                 /* zero for max => infinity */
+      if (max == 0) max = INT_MAX;
+      break;
+
+      case OP_CRRANGE:
+      case OP_CRMINRANGE:
+      minimize = (*ecode == OP_CRMINRANGE);
+      min = GET2(ecode, 1);
+      max = GET2(ecode, 1 + IMM2_SIZE);
+      if (max == 0) max = INT_MAX;
+      ecode += 1 + 2 * IMM2_SIZE;
+      break;
+
+      default:               /* No repeat follows */
+      if ((length = match_ref(offset, eptr, length, md, caseless)) < 0)
+        {
+        if (length == -2) eptr = md->end_subject;   /* Partial match */
+        CHECK_PARTIAL();
+        RRETURN(MATCH_NOMATCH);
+        }
+      eptr += length;
+      continue;              /* With the main loop */
+      }
+
+    /* Handle repeated back references. If the length of the reference is
+    zero, just continue with the main loop. If the length is negative, it
+    means the reference is unset in non-Java-compatible mode. If the minimum is
+    zero, we can continue at the same level without recursion. For any other
+    minimum, carrying on will result in NOMATCH. */
+
+    if (length == 0) continue;
+    if (length < 0 && min == 0) continue;
+
+    /* First, ensure the minimum number of matches are present. We get back
+    the length of the reference string explicitly rather than passing the
+    address of eptr, so that eptr can be a register variable. */
+
+    for (i = 1; i <= min; i++)
+      {
+      int slength;
+      if ((slength = match_ref(offset, eptr, length, md, caseless)) < 0)
+        {
+        if (slength == -2) eptr = md->end_subject;   /* Partial match */
+        CHECK_PARTIAL();
+        RRETURN(MATCH_NOMATCH);
+        }
+      eptr += slength;
+      }
+
+    /* If min = max, continue at the same level without recursion.
+    They are not both allowed to be zero. */
+
+    if (min == max) continue;
+
+    /* If minimizing, keep trying and advancing the pointer */
+
+    if (minimize)
+      {
+      for (fi = min;; fi++)
+        {
+        int slength;
+        RMATCH(eptr, ecode, offset_top, md, eptrb, RM14);
+        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+        if (fi >= max) RRETURN(MATCH_NOMATCH);
+        if ((slength = match_ref(offset, eptr, length, md, caseless)) < 0)
+          {
+          if (slength == -2) eptr = md->end_subject;   /* Partial match */
+          CHECK_PARTIAL();
+          RRETURN(MATCH_NOMATCH);
+          }
+        eptr += slength;
+        }
+      /* Control never gets here */
+      }
+
+    /* If maximizing, find the longest string and work backwards */
+
+    else
+      {
+      pp = eptr;
+      for (i = min; i < max; i++)
+        {
+        int slength;
+        if ((slength = match_ref(offset, eptr, length, md, caseless)) < 0)
+          {
+          /* Can't use CHECK_PARTIAL because we don't want to update eptr in
+          the soft partial matching case. */
+
+          if (slength == -2 && md->partial != 0 &&
+              md->end_subject > md->start_used_ptr)
+            {
+            md->hitend = TRUE;
+            if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);
+            }
+          break;
+          }
+        eptr += slength;
+        }
+
+      while (eptr >= pp)
+        {
+        RMATCH(eptr, ecode, offset_top, md, eptrb, RM15);
+        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+        eptr -= length;
+        }
+      RRETURN(MATCH_NOMATCH);
+      }
+    /* Control never gets here */
+
+    /* Match a bit-mapped character class, possibly repeatedly. This op code is
+    used when all the characters in the class have values in the range 0-255,
+    and either the matching is caseful, or the characters are in the range
+    0-127 when UTF-8 processing is enabled. The only difference between
+    OP_CLASS and OP_NCLASS occurs when a data character outside the range is
+    encountered.
+
+    First, look past the end of the item to see if there is repeat information
+    following. Then obey similar code to character type repeats - written out
+    again for speed. */
+
+    case OP_NCLASS:
+    case OP_CLASS:
+      {
+      /* The data variable is saved across frames, so the byte map needs to
+      be stored there. */
+#define BYTE_MAP ((pcre_uint8 *)data)
+      data = ecode + 1;                /* Save for matching */
+      ecode += 1 + (32 / sizeof(pcre_uchar)); /* Advance past the item */
+
+      switch (*ecode)
+        {
+        case OP_CRSTAR:
+        case OP_CRMINSTAR:
+        case OP_CRPLUS:
+        case OP_CRMINPLUS:
+        case OP_CRQUERY:
+        case OP_CRMINQUERY:
+        c = *ecode++ - OP_CRSTAR;
+        minimize = (c & 1) != 0;
+        min = rep_min[c];                 /* Pick up values from tables; */
+        max = rep_max[c];                 /* zero for max => infinity */
+        if (max == 0) max = INT_MAX;
+        break;
+
+        case OP_CRRANGE:
+        case OP_CRMINRANGE:
+        minimize = (*ecode == OP_CRMINRANGE);
+        min = GET2(ecode, 1);
+        max = GET2(ecode, 1 + IMM2_SIZE);
+        if (max == 0) max = INT_MAX;
+        ecode += 1 + 2 * IMM2_SIZE;
+        break;
+
+        default:               /* No repeat follows */
+        min = max = 1;
+        break;
+        }
+
+      /* First, ensure the minimum number of matches are present. */
+
+#ifdef SUPPORT_UTF
+      if (utf)
+        {
+        for (i = 1; i <= min; i++)
+          {
+          if (eptr >= md->end_subject)
+            {
+            SCHECK_PARTIAL();
+            RRETURN(MATCH_NOMATCH);
+            }
+          GETCHARINC(c, eptr);
+          if (c > 255)
+            {
+            if (op == OP_CLASS) RRETURN(MATCH_NOMATCH);
+            }
+          else
+            if ((BYTE_MAP[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH);
+          }
+        }
+      else
+#endif
+      /* Not UTF mode */
+        {
+        for (i = 1; i <= min; i++)
+          {
+          if (eptr >= md->end_subject)
+            {
+            SCHECK_PARTIAL();
+            RRETURN(MATCH_NOMATCH);
+            }
+          c = *eptr++;
+#ifndef COMPILE_PCRE8
+          if (c > 255)
+            {
+            if (op == OP_CLASS) RRETURN(MATCH_NOMATCH);
+            }
+          else
+#endif
+            if ((BYTE_MAP[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH);
+          }
+        }
+
+      /* If max == min we can continue with the main loop without the
+      need to recurse. */
+
+      if (min == max) continue;
+
+      /* If minimizing, keep testing the rest of the expression and advancing
+      the pointer while it matches the class. */
+
+      if (minimize)
+        {
+#ifdef SUPPORT_UTF
+        if (utf)
+          {
+          for (fi = min;; fi++)
+            {
+            RMATCH(eptr, ecode, offset_top, md, eptrb, RM16);
+            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+            if (fi >= max) RRETURN(MATCH_NOMATCH);
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              RRETURN(MATCH_NOMATCH);
+              }
+            GETCHARINC(c, eptr);
+            if (c > 255)
+              {
+              if (op == OP_CLASS) RRETURN(MATCH_NOMATCH);
+              }
+            else
+              if ((BYTE_MAP[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH);
+            }
+          }
+        else
+#endif
+        /* Not UTF mode */
+          {
+          for (fi = min;; fi++)
+            {
+            RMATCH(eptr, ecode, offset_top, md, eptrb, RM17);
+            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+            if (fi >= max) RRETURN(MATCH_NOMATCH);
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              RRETURN(MATCH_NOMATCH);
+              }
+            c = *eptr++;
+#ifndef COMPILE_PCRE8
+            if (c > 255)
+              {
+              if (op == OP_CLASS) RRETURN(MATCH_NOMATCH);
+              }
+            else
+#endif
+              if ((BYTE_MAP[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH);
+            }
+          }
+        /* Control never gets here */
+        }
+
+      /* If maximizing, find the longest possible run, then work backwards. */
+
+      else
+        {
+        pp = eptr;
+
+#ifdef SUPPORT_UTF
+        if (utf)
+          {
+          for (i = min; i < max; i++)
+            {
+            int len = 1;
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              break;
+              }
+            GETCHARLEN(c, eptr, len);
+            if (c > 255)
+              {
+              if (op == OP_CLASS) break;
+              }
+            else
+              if ((BYTE_MAP[c/8] & (1 << (c&7))) == 0) break;
+            eptr += len;
+            }
+          for (;;)
+            {
+            RMATCH(eptr, ecode, offset_top, md, eptrb, RM18);
+            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+            if (eptr-- == pp) break;        /* Stop if tried at original pos */
+            BACKCHAR(eptr);
+            }
+          }
+        else
+#endif
+          /* Not UTF mode */
+          {
+          for (i = min; i < max; i++)
+            {
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              break;
+              }
+            c = *eptr;
+#ifndef COMPILE_PCRE8
+            if (c > 255)
+              {
+              if (op == OP_CLASS) break;
+              }
+            else
+#endif
+              if ((BYTE_MAP[c/8] & (1 << (c&7))) == 0) break;
+            eptr++;
+            }
+          while (eptr >= pp)
+            {
+            RMATCH(eptr, ecode, offset_top, md, eptrb, RM19);
+            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+            eptr--;
+            }
+          }
+
+        RRETURN(MATCH_NOMATCH);
+        }
+#undef BYTE_MAP
+      }
+    /* Control never gets here */
+
+
+    /* Match an extended character class. This opcode is encountered only
+    when UTF-8 mode mode is supported. Nevertheless, we may not be in UTF-8
+    mode, because Unicode properties are supported in non-UTF-8 mode. */
+
+#if defined SUPPORT_UTF || !defined COMPILE_PCRE8
+    case OP_XCLASS:
+      {
+      data = ecode + 1 + LINK_SIZE;                /* Save for matching */
+      ecode += GET(ecode, 1);                      /* Advance past the item */
+
+      switch (*ecode)
+        {
+        case OP_CRSTAR:
+        case OP_CRMINSTAR:
+        case OP_CRPLUS:
+        case OP_CRMINPLUS:
+        case OP_CRQUERY:
+        case OP_CRMINQUERY:
+        c = *ecode++ - OP_CRSTAR;
+        minimize = (c & 1) != 0;
+        min = rep_min[c];                 /* Pick up values from tables; */
+        max = rep_max[c];                 /* zero for max => infinity */
+        if (max == 0) max = INT_MAX;
+        break;
+
+        case OP_CRRANGE:
+        case OP_CRMINRANGE:
+        minimize = (*ecode == OP_CRMINRANGE);
+        min = GET2(ecode, 1);
+        max = GET2(ecode, 1 + IMM2_SIZE);
+        if (max == 0) max = INT_MAX;
+        ecode += 1 + 2 * IMM2_SIZE;
+        break;
+
+        default:               /* No repeat follows */
+        min = max = 1;
+        break;
+        }
+
+      /* First, ensure the minimum number of matches are present. */
+
+      for (i = 1; i <= min; i++)
+        {
+        if (eptr >= md->end_subject)
+          {
+          SCHECK_PARTIAL();
+          RRETURN(MATCH_NOMATCH);
+          }
+        GETCHARINCTEST(c, eptr);
+        if (!PRIV(xclass)(c, data, utf)) RRETURN(MATCH_NOMATCH);
+        }
+
+      /* If max == min we can continue with the main loop without the
+      need to recurse. */
+
+      if (min == max) continue;
+
+      /* If minimizing, keep testing the rest of the expression and advancing
+      the pointer while it matches the class. */
+
+      if (minimize)
+        {
+        for (fi = min;; fi++)
+          {
+          RMATCH(eptr, ecode, offset_top, md, eptrb, RM20);
+          if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+          if (fi >= max) RRETURN(MATCH_NOMATCH);
+          if (eptr >= md->end_subject)
+            {
+            SCHECK_PARTIAL();
+            RRETURN(MATCH_NOMATCH);
+            }
+          GETCHARINCTEST(c, eptr);
+          if (!PRIV(xclass)(c, data, utf)) RRETURN(MATCH_NOMATCH);
+          }
+        /* Control never gets here */
+        }
+
+      /* If maximizing, find the longest possible run, then work backwards. */
+
+      else
+        {
+        pp = eptr;
+        for (i = min; i < max; i++)
+          {
+          int len = 1;
+          if (eptr >= md->end_subject)
+            {
+            SCHECK_PARTIAL();
+            break;
+            }
+#ifdef SUPPORT_UTF
+          GETCHARLENTEST(c, eptr, len);
+#else
+          c = *eptr;
+#endif
+          if (!PRIV(xclass)(c, data, utf)) break;
+          eptr += len;
+          }
+        for(;;)
+          {
+          RMATCH(eptr, ecode, offset_top, md, eptrb, RM21);
+          if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+          if (eptr-- == pp) break;        /* Stop if tried at original pos */
+#ifdef SUPPORT_UTF
+          if (utf) BACKCHAR(eptr);
+#endif
+          }
+        RRETURN(MATCH_NOMATCH);
+        }
+
+      /* Control never gets here */
+      }
+#endif    /* End of XCLASS */
+
+    /* Match a single character, casefully */
+
+    case OP_CHAR:
+#ifdef SUPPORT_UTF
+    if (utf)
+      {
+      length = 1;
+      ecode++;
+      GETCHARLEN(fc, ecode, length);
+      if (length > md->end_subject - eptr)
+        {
+        CHECK_PARTIAL();             /* Not SCHECK_PARTIAL() */
+        RRETURN(MATCH_NOMATCH);
+        }
+      while (length-- > 0) if (*ecode++ != RAWUCHARINC(eptr)) RRETURN(MATCH_NOMATCH);
+      }
+    else
+#endif
+    /* Not UTF mode */
+      {
+      if (md->end_subject - eptr < 1)
+        {
+        SCHECK_PARTIAL();            /* This one can use SCHECK_PARTIAL() */
+        RRETURN(MATCH_NOMATCH);
+        }
+      if (ecode[1] != *eptr++) RRETURN(MATCH_NOMATCH);
+      ecode += 2;
+      }
+    break;
+
+    /* Match a single character, caselessly. If we are at the end of the
+    subject, give up immediately. */
+
+    case OP_CHARI:
+    if (eptr >= md->end_subject)
+      {
+      SCHECK_PARTIAL();
+      RRETURN(MATCH_NOMATCH);
+      }
+
+#ifdef SUPPORT_UTF
+    if (utf)
+      {
+      length = 1;
+      ecode++;
+      GETCHARLEN(fc, ecode, length);
+
+      /* If the pattern character's value is < 128, we have only one byte, and
+      we know that its other case must also be one byte long, so we can use the
+      fast lookup table. We know that there is at least one byte left in the
+      subject. */
+
+      if (fc < 128)
+        {
+        pcre_uint32 cc = RAWUCHAR(eptr);
+        if (md->lcc[fc] != TABLE_GET(cc, md->lcc, cc)) RRETURN(MATCH_NOMATCH);
+        ecode++;
+        eptr++;
+        }
+
+      /* Otherwise we must pick up the subject character. Note that we cannot
+      use the value of "length" to check for sufficient bytes left, because the
+      other case of the character may have more or fewer bytes.  */
+
+      else
+        {
+        pcre_uint32 dc;
+        GETCHARINC(dc, eptr);
+        ecode += length;
+
+        /* If we have Unicode property support, we can use it to test the other
+        case of the character, if there is one. */
+
+        if (fc != dc)
+          {
+#ifdef SUPPORT_UCP
+          if (dc != UCD_OTHERCASE(fc))
+#endif
+            RRETURN(MATCH_NOMATCH);
+          }
+        }
+      }
+    else
+#endif   /* SUPPORT_UTF */
+
+    /* Not UTF mode */
+      {
+      if (TABLE_GET(ecode[1], md->lcc, ecode[1])
+          != TABLE_GET(*eptr, md->lcc, *eptr)) RRETURN(MATCH_NOMATCH);
+      eptr++;
+      ecode += 2;
+      }
+    break;
+
+    /* Match a single character repeatedly. */
+
+    case OP_EXACT:
+    case OP_EXACTI:
+    min = max = GET2(ecode, 1);
+    ecode += 1 + IMM2_SIZE;
+    goto REPEATCHAR;
+
+    case OP_POSUPTO:
+    case OP_POSUPTOI:
+    possessive = TRUE;
+    /* Fall through */
+
+    case OP_UPTO:
+    case OP_UPTOI:
+    case OP_MINUPTO:
+    case OP_MINUPTOI:
+    min = 0;
+    max = GET2(ecode, 1);
+    minimize = *ecode == OP_MINUPTO || *ecode == OP_MINUPTOI;
+    ecode += 1 + IMM2_SIZE;
+    goto REPEATCHAR;
+
+    case OP_POSSTAR:
+    case OP_POSSTARI:
+    possessive = TRUE;
+    min = 0;
+    max = INT_MAX;
+    ecode++;
+    goto REPEATCHAR;
+
+    case OP_POSPLUS:
+    case OP_POSPLUSI:
+    possessive = TRUE;
+    min = 1;
+    max = INT_MAX;
+    ecode++;
+    goto REPEATCHAR;
+
+    case OP_POSQUERY:
+    case OP_POSQUERYI:
+    possessive = TRUE;
+    min = 0;
+    max = 1;
+    ecode++;
+    goto REPEATCHAR;
+
+    case OP_STAR:
+    case OP_STARI:
+    case OP_MINSTAR:
+    case OP_MINSTARI:
+    case OP_PLUS:
+    case OP_PLUSI:
+    case OP_MINPLUS:
+    case OP_MINPLUSI:
+    case OP_QUERY:
+    case OP_QUERYI:
+    case OP_MINQUERY:
+    case OP_MINQUERYI:
+    c = *ecode++ - ((op < OP_STARI)? OP_STAR : OP_STARI);
+    minimize = (c & 1) != 0;
+    min = rep_min[c];                 /* Pick up values from tables; */
+    max = rep_max[c];                 /* zero for max => infinity */
+    if (max == 0) max = INT_MAX;
+
+    /* Common code for all repeated single-character matches. We first check
+    for the minimum number of characters. If the minimum equals the maximum, we
+    are done. Otherwise, if minimizing, check the rest of the pattern for a
+    match; if there isn't one, advance up to the maximum, one character at a
+    time.
+
+    If maximizing, advance up to the maximum number of matching characters,
+    until eptr is past the end of the maximum run. If possessive, we are
+    then done (no backing up). Otherwise, match at this position; anything
+    other than no match is immediately returned. For nomatch, back up one
+    character, unless we are matching \R and the last thing matched was
+    \r\n, in which case, back up two bytes. When we reach the first optional
+    character position, we can save stack by doing a tail recurse.
+
+    The various UTF/non-UTF and caseful/caseless cases are handled separately,
+    for speed. */
+
+    REPEATCHAR:
+#ifdef SUPPORT_UTF
+    if (utf)
+      {
+      length = 1;
+      charptr = ecode;
+      GETCHARLEN(fc, ecode, length);
+      ecode += length;
+
+      /* Handle multibyte character matching specially here. There is
+      support for caseless matching if UCP support is present. */
+
+      if (length > 1)
+        {
+#ifdef SUPPORT_UCP
+        pcre_uint32 othercase;
+        if (op >= OP_STARI &&     /* Caseless */
+            (othercase = UCD_OTHERCASE(fc)) != fc)
+          oclength = PRIV(ord2utf)(othercase, occhars);
+        else oclength = 0;
+#endif  /* SUPPORT_UCP */
+
+        for (i = 1; i <= min; i++)
+          {
+          if (eptr <= md->end_subject - length &&
+            memcmp(eptr, charptr, IN_UCHARS(length)) == 0) eptr += length;
+#ifdef SUPPORT_UCP
+          else if (oclength > 0 &&
+                   eptr <= md->end_subject - oclength &&
+                   memcmp(eptr, occhars, IN_UCHARS(oclength)) == 0) eptr += oclength;
+#endif  /* SUPPORT_UCP */
+          else
+            {
+            CHECK_PARTIAL();
+            RRETURN(MATCH_NOMATCH);
+            }
+          }
+
+        if (min == max) continue;
+
+        if (minimize)
+          {
+          for (fi = min;; fi++)
+            {
+            RMATCH(eptr, ecode, offset_top, md, eptrb, RM22);
+            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+            if (fi >= max) RRETURN(MATCH_NOMATCH);
+            if (eptr <= md->end_subject - length &&
+              memcmp(eptr, charptr, IN_UCHARS(length)) == 0) eptr += length;
+#ifdef SUPPORT_UCP
+            else if (oclength > 0 &&
+                     eptr <= md->end_subject - oclength &&
+                     memcmp(eptr, occhars, IN_UCHARS(oclength)) == 0) eptr += oclength;
+#endif  /* SUPPORT_UCP */
+            else
+              {
+              CHECK_PARTIAL();
+              RRETURN(MATCH_NOMATCH);
+              }
+            }
+          /* Control never gets here */
+          }
+
+        else  /* Maximize */
+          {
+          pp = eptr;
+          for (i = min; i < max; i++)
+            {
+            if (eptr <= md->end_subject - length &&
+                memcmp(eptr, charptr, IN_UCHARS(length)) == 0) eptr += length;
+#ifdef SUPPORT_UCP
+            else if (oclength > 0 &&
+                     eptr <= md->end_subject - oclength &&
+                     memcmp(eptr, occhars, IN_UCHARS(oclength)) == 0) eptr += oclength;
+#endif  /* SUPPORT_UCP */
+            else
+              {
+              CHECK_PARTIAL();
+              break;
+              }
+            }
+
+          if (possessive) continue;    /* No backtracking */
+          for(;;)
+            {
+            if (eptr == pp) goto TAIL_RECURSE;
+            RMATCH(eptr, ecode, offset_top, md, eptrb, RM23);
+            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+#ifdef SUPPORT_UCP
+            eptr--;
+            BACKCHAR(eptr);
+#else   /* without SUPPORT_UCP */
+            eptr -= length;
+#endif  /* SUPPORT_UCP */
+            }
+          }
+        /* Control never gets here */
+        }
+
+      /* If the length of a UTF-8 character is 1, we fall through here, and
+      obey the code as for non-UTF-8 characters below, though in this case the
+      value of fc will always be < 128. */
+      }
+    else
+#endif  /* SUPPORT_UTF */
+      /* When not in UTF-8 mode, load a single-byte character. */
+      fc = *ecode++;
+
+    /* The value of fc at this point is always one character, though we may
+    or may not be in UTF mode. The code is duplicated for the caseless and
+    caseful cases, for speed, since matching characters is likely to be quite
+    common. First, ensure the minimum number of matches are present. If min =
+    max, continue at the same level without recursing. Otherwise, if
+    minimizing, keep trying the rest of the expression and advancing one
+    matching character if failing, up to the maximum. Alternatively, if
+    maximizing, find the maximum number of characters and work backwards. */
+
+    DPRINTF(("matching %c{%d,%d} against subject %.*s\n", fc, min, max,
+      max, (char *)eptr));
+
+    if (op >= OP_STARI)  /* Caseless */
+      {
+#ifdef COMPILE_PCRE8
+      /* fc must be < 128 if UTF is enabled. */
+      foc = md->fcc[fc];
+#else
+#ifdef SUPPORT_UTF
+#ifdef SUPPORT_UCP
+      if (utf && fc > 127)
+        foc = UCD_OTHERCASE(fc);
+#else
+      if (utf && fc > 127)
+        foc = fc;
+#endif /* SUPPORT_UCP */
+      else
+#endif /* SUPPORT_UTF */
+        foc = TABLE_GET(fc, md->fcc, fc);
+#endif /* COMPILE_PCRE8 */
+
+      for (i = 1; i <= min; i++)
+        {
+        pcre_uint32 cc;                 /* Faster than pcre_uchar */
+        if (eptr >= md->end_subject)
+          {
+          SCHECK_PARTIAL();
+          RRETURN(MATCH_NOMATCH);
+          }
+        cc = RAWUCHARTEST(eptr);
+        if (fc != cc && foc != cc) RRETURN(MATCH_NOMATCH);
+        eptr++;
+        }
+      if (min == max) continue;
+      if (minimize)
+        {
+        for (fi = min;; fi++)
+          {
+          pcre_uint32 cc;               /* Faster than pcre_uchar */
+          RMATCH(eptr, ecode, offset_top, md, eptrb, RM24);
+          if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+          if (fi >= max) RRETURN(MATCH_NOMATCH);
+          if (eptr >= md->end_subject)
+            {
+            SCHECK_PARTIAL();
+            RRETURN(MATCH_NOMATCH);
+            }
+          cc = RAWUCHARTEST(eptr);
+          if (fc != cc && foc != cc) RRETURN(MATCH_NOMATCH);
+          eptr++;
+          }
+        /* Control never gets here */
+        }
+      else  /* Maximize */
+        {
+        pp = eptr;
+        for (i = min; i < max; i++)
+          {
+          pcre_uint32 cc;               /* Faster than pcre_uchar */
+          if (eptr >= md->end_subject)
+            {
+            SCHECK_PARTIAL();
+            break;
+            }
+          cc = RAWUCHARTEST(eptr);
+          if (fc != cc && foc != cc) break;
+          eptr++;
+          }
+
+        if (possessive) continue;       /* No backtracking */
+        for (;;)
+          {
+          if (eptr == pp) goto TAIL_RECURSE;
+          RMATCH(eptr, ecode, offset_top, md, eptrb, RM25);
+          eptr--;
+          if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+          }
+        RRETURN(MATCH_NOMATCH);
+        }
+      /* Control never gets here */
+      }
+
+    /* Caseful comparisons (includes all multi-byte characters) */
+
+    else
+      {
+      for (i = 1; i <= min; i++)
+        {
+        if (eptr >= md->end_subject)
+          {
+          SCHECK_PARTIAL();
+          RRETURN(MATCH_NOMATCH);
+          }
+        if (fc != RAWUCHARINCTEST(eptr)) RRETURN(MATCH_NOMATCH);
+        }
+
+      if (min == max) continue;
+
+      if (minimize)
+        {
+        for (fi = min;; fi++)
+          {
+          RMATCH(eptr, ecode, offset_top, md, eptrb, RM26);
+          if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+          if (fi >= max) RRETURN(MATCH_NOMATCH);
+          if (eptr >= md->end_subject)
+            {
+            SCHECK_PARTIAL();
+            RRETURN(MATCH_NOMATCH);
+            }
+          if (fc != RAWUCHARINCTEST(eptr)) RRETURN(MATCH_NOMATCH);
+          }
+        /* Control never gets here */
+        }
+      else  /* Maximize */
+        {
+        pp = eptr;
+        for (i = min; i < max; i++)
+          {
+          if (eptr >= md->end_subject)
+            {
+            SCHECK_PARTIAL();
+            break;
+            }
+          if (fc != RAWUCHARTEST(eptr)) break;
+          eptr++;
+          }
+        if (possessive) continue;    /* No backtracking */
+        for (;;)
+          {
+          if (eptr == pp) goto TAIL_RECURSE;
+          RMATCH(eptr, ecode, offset_top, md, eptrb, RM27);
+          eptr--;
+          if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+          }
+        RRETURN(MATCH_NOMATCH);
+        }
+      }
+    /* Control never gets here */
+
+    /* Match a negated single one-byte character. The character we are
+    checking can be multibyte. */
+
+    case OP_NOT:
+    case OP_NOTI:
+    if (eptr >= md->end_subject)
+      {
+      SCHECK_PARTIAL();
+      RRETURN(MATCH_NOMATCH);
+      }
+#ifdef SUPPORT_UTF
+    if (utf)
+      {
+      register pcre_uint32 ch, och;
+
+      ecode++;
+      GETCHARINC(ch, ecode);
+      GETCHARINC(c, eptr);
+
+      if (op == OP_NOT)
+        {
+        if (ch == c) RRETURN(MATCH_NOMATCH);
+        }
+      else
+        {
+#ifdef SUPPORT_UCP
+        if (ch > 127)
+          och = UCD_OTHERCASE(ch);
+#else
+        if (ch > 127)
+          och = ch;
+#endif /* SUPPORT_UCP */
+        else
+          och = TABLE_GET(ch, md->fcc, ch);
+        if (ch == c || och == c) RRETURN(MATCH_NOMATCH);
+        }
+      }
+    else
+#endif
+      {
+      register pcre_uint32 ch = ecode[1];
+      c = *eptr++;
+      if (ch == c || (op == OP_NOTI && TABLE_GET(ch, md->fcc, ch) == c))
+        RRETURN(MATCH_NOMATCH);
+      ecode += 2;
+      }
+    break;
+
+    /* Match a negated single one-byte character repeatedly. This is almost a
+    repeat of the code for a repeated single character, but I haven't found a
+    nice way of commoning these up that doesn't require a test of the
+    positive/negative option for each character match. Maybe that wouldn't add
+    very much to the time taken, but character matching *is* what this is all
+    about... */
+
+    case OP_NOTEXACT:
+    case OP_NOTEXACTI:
+    min = max = GET2(ecode, 1);
+    ecode += 1 + IMM2_SIZE;
+    goto REPEATNOTCHAR;
+
+    case OP_NOTUPTO:
+    case OP_NOTUPTOI:
+    case OP_NOTMINUPTO:
+    case OP_NOTMINUPTOI:
+    min = 0;
+    max = GET2(ecode, 1);
+    minimize = *ecode == OP_NOTMINUPTO || *ecode == OP_NOTMINUPTOI;
+    ecode += 1 + IMM2_SIZE;
+    goto REPEATNOTCHAR;
+
+    case OP_NOTPOSSTAR:
+    case OP_NOTPOSSTARI:
+    possessive = TRUE;
+    min = 0;
+    max = INT_MAX;
+    ecode++;
+    goto REPEATNOTCHAR;
+
+    case OP_NOTPOSPLUS:
+    case OP_NOTPOSPLUSI:
+    possessive = TRUE;
+    min = 1;
+    max = INT_MAX;
+    ecode++;
+    goto REPEATNOTCHAR;
+
+    case OP_NOTPOSQUERY:
+    case OP_NOTPOSQUERYI:
+    possessive = TRUE;
+    min = 0;
+    max = 1;
+    ecode++;
+    goto REPEATNOTCHAR;
+
+    case OP_NOTPOSUPTO:
+    case OP_NOTPOSUPTOI:
+    possessive = TRUE;
+    min = 0;
+    max = GET2(ecode, 1);
+    ecode += 1 + IMM2_SIZE;
+    goto REPEATNOTCHAR;
+
+    case OP_NOTSTAR:
+    case OP_NOTSTARI:
+    case OP_NOTMINSTAR:
+    case OP_NOTMINSTARI:
+    case OP_NOTPLUS:
+    case OP_NOTPLUSI:
+    case OP_NOTMINPLUS:
+    case OP_NOTMINPLUSI:
+    case OP_NOTQUERY:
+    case OP_NOTQUERYI:
+    case OP_NOTMINQUERY:
+    case OP_NOTMINQUERYI:
+    c = *ecode++ - ((op >= OP_NOTSTARI)? OP_NOTSTARI: OP_NOTSTAR);
+    minimize = (c & 1) != 0;
+    min = rep_min[c];                 /* Pick up values from tables; */
+    max = rep_max[c];                 /* zero for max => infinity */
+    if (max == 0) max = INT_MAX;
+
+    /* Common code for all repeated single-byte matches. */
+
+    REPEATNOTCHAR:
+    GETCHARINCTEST(fc, ecode);
+
+    /* The code is duplicated for the caseless and caseful cases, for speed,
+    since matching characters is likely to be quite common. First, ensure the
+    minimum number of matches are present. If min = max, continue at the same
+    level without recursing. Otherwise, if minimizing, keep trying the rest of
+    the expression and advancing one matching character if failing, up to the
+    maximum. Alternatively, if maximizing, find the maximum number of
+    characters and work backwards. */
+
+    DPRINTF(("negative matching %c{%d,%d} against subject %.*s\n", fc, min, max,
+      max, (char *)eptr));
+
+    if (op >= OP_NOTSTARI)     /* Caseless */
+      {
+#ifdef SUPPORT_UTF
+#ifdef SUPPORT_UCP
+      if (utf && fc > 127)
+        foc = UCD_OTHERCASE(fc);
+#else
+      if (utf && fc > 127)
+        foc = fc;
+#endif /* SUPPORT_UCP */
+      else
+#endif /* SUPPORT_UTF */
+        foc = TABLE_GET(fc, md->fcc, fc);
+
+#ifdef SUPPORT_UTF
+      if (utf)
+        {
+        register pcre_uint32 d;
+        for (i = 1; i <= min; i++)
+          {
+          if (eptr >= md->end_subject)
+            {
+            SCHECK_PARTIAL();
+            RRETURN(MATCH_NOMATCH);
+            }
+          GETCHARINC(d, eptr);
+          if (fc == d || (unsigned int)foc == d) RRETURN(MATCH_NOMATCH);
+          }
+        }
+      else
+#endif  /* SUPPORT_UTF */
+      /* Not UTF mode */
+        {
+        for (i = 1; i <= min; i++)
+          {
+          if (eptr >= md->end_subject)
+            {
+            SCHECK_PARTIAL();
+            RRETURN(MATCH_NOMATCH);
+            }
+          if (fc == *eptr || foc == *eptr) RRETURN(MATCH_NOMATCH);
+          eptr++;
+          }
+        }
+
+      if (min == max) continue;
+
+      if (minimize)
+        {
+#ifdef SUPPORT_UTF
+        if (utf)
+          {
+          register pcre_uint32 d;
+          for (fi = min;; fi++)
+            {
+            RMATCH(eptr, ecode, offset_top, md, eptrb, RM28);
+            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+            if (fi >= max) RRETURN(MATCH_NOMATCH);
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              RRETURN(MATCH_NOMATCH);
+              }
+            GETCHARINC(d, eptr);
+            if (fc == d || (unsigned int)foc == d) RRETURN(MATCH_NOMATCH);
+            }
+          }
+        else
+#endif  /*SUPPORT_UTF */
+        /* Not UTF mode */
+          {
+          for (fi = min;; fi++)
+            {
+            RMATCH(eptr, ecode, offset_top, md, eptrb, RM29);
+            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+            if (fi >= max) RRETURN(MATCH_NOMATCH);
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              RRETURN(MATCH_NOMATCH);
+              }
+            if (fc == *eptr || foc == *eptr) RRETURN(MATCH_NOMATCH);
+            eptr++;
+            }
+          }
+        /* Control never gets here */
+        }
+
+      /* Maximize case */
+
+      else
+        {
+        pp = eptr;
+
+#ifdef SUPPORT_UTF
+        if (utf)
+          {
+          register pcre_uint32 d;
+          for (i = min; i < max; i++)
+            {
+            int len = 1;
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              break;
+              }
+            GETCHARLEN(d, eptr, len);
+            if (fc == d || (unsigned int)foc == d) break;
+            eptr += len;
+            }
+          if (possessive) continue;    /* No backtracking */
+          for(;;)
+            {
+            if (eptr == pp) goto TAIL_RECURSE;
+            RMATCH(eptr, ecode, offset_top, md, eptrb, RM30);
+            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+            eptr--;
+            BACKCHAR(eptr);
+            }
+          }
+        else
+#endif  /* SUPPORT_UTF */
+        /* Not UTF mode */
+          {
+          for (i = min; i < max; i++)
+            {
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              break;
+              }
+            if (fc == *eptr || foc == *eptr) break;
+            eptr++;
+            }
+          if (possessive) continue;    /* No backtracking */
+          for (;;)
+            {
+            if (eptr == pp) goto TAIL_RECURSE;
+            RMATCH(eptr, ecode, offset_top, md, eptrb, RM31);
+            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+            eptr--;
+            }
+          }
+
+        RRETURN(MATCH_NOMATCH);
+        }
+      /* Control never gets here */
+      }
+
+    /* Caseful comparisons */
+
+    else
+      {
+#ifdef SUPPORT_UTF
+      if (utf)
+        {
+        register pcre_uint32 d;
+        for (i = 1; i <= min; i++)
+          {
+          if (eptr >= md->end_subject)
+            {
+            SCHECK_PARTIAL();
+            RRETURN(MATCH_NOMATCH);
+            }
+          GETCHARINC(d, eptr);
+          if (fc == d) RRETURN(MATCH_NOMATCH);
+          }
+        }
+      else
+#endif
+      /* Not UTF mode */
+        {
+        for (i = 1; i <= min; i++)
+          {
+          if (eptr >= md->end_subject)
+            {
+            SCHECK_PARTIAL();
+            RRETURN(MATCH_NOMATCH);
+            }
+          if (fc == *eptr++) RRETURN(MATCH_NOMATCH);
+          }
+        }
+
+      if (min == max) continue;
+
+      if (minimize)
+        {
+#ifdef SUPPORT_UTF
+        if (utf)
+          {
+          register pcre_uint32 d;
+          for (fi = min;; fi++)
+            {
+            RMATCH(eptr, ecode, offset_top, md, eptrb, RM32);
+            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+            if (fi >= max) RRETURN(MATCH_NOMATCH);
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              RRETURN(MATCH_NOMATCH);
+              }
+            GETCHARINC(d, eptr);
+            if (fc == d) RRETURN(MATCH_NOMATCH);
+            }
+          }
+        else
+#endif
+        /* Not UTF mode */
+          {
+          for (fi = min;; fi++)
+            {
+            RMATCH(eptr, ecode, offset_top, md, eptrb, RM33);
+            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+            if (fi >= max) RRETURN(MATCH_NOMATCH);
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              RRETURN(MATCH_NOMATCH);
+              }
+            if (fc == *eptr++) RRETURN(MATCH_NOMATCH);
+            }
+          }
+        /* Control never gets here */
+        }
+
+      /* Maximize case */
+
+      else
+        {
+        pp = eptr;
+
+#ifdef SUPPORT_UTF
+        if (utf)
+          {
+          register pcre_uint32 d;
+          for (i = min; i < max; i++)
+            {
+            int len = 1;
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              break;
+              }
+            GETCHARLEN(d, eptr, len);
+            if (fc == d) break;
+            eptr += len;
+            }
+          if (possessive) continue;    /* No backtracking */
+          for(;;)
+            {
+            if (eptr == pp) goto TAIL_RECURSE;
+            RMATCH(eptr, ecode, offset_top, md, eptrb, RM34);
+            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+            eptr--;
+            BACKCHAR(eptr);
+            }
+          }
+        else
+#endif
+        /* Not UTF mode */
+          {
+          for (i = min; i < max; i++)
+            {
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              break;
+              }
+            if (fc == *eptr) break;
+            eptr++;
+            }
+          if (possessive) continue;    /* No backtracking */
+          for (;;)
+            {
+            if (eptr == pp) goto TAIL_RECURSE;
+            RMATCH(eptr, ecode, offset_top, md, eptrb, RM35);
+            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+            eptr--;
+            }
+          }
+
+        RRETURN(MATCH_NOMATCH);
+        }
+      }
+    /* Control never gets here */
+
+    /* Match a single character type repeatedly; several different opcodes
+    share code. This is very similar to the code for single characters, but we
+    repeat it in the interests of efficiency. */
+
+    case OP_TYPEEXACT:
+    min = max = GET2(ecode, 1);
+    minimize = TRUE;
+    ecode += 1 + IMM2_SIZE;
+    goto REPEATTYPE;
+
+    case OP_TYPEUPTO:
+    case OP_TYPEMINUPTO:
+    min = 0;
+    max = GET2(ecode, 1);
+    minimize = *ecode == OP_TYPEMINUPTO;
+    ecode += 1 + IMM2_SIZE;
+    goto REPEATTYPE;
+
+    case OP_TYPEPOSSTAR:
+    possessive = TRUE;
+    min = 0;
+    max = INT_MAX;
+    ecode++;
+    goto REPEATTYPE;
+
+    case OP_TYPEPOSPLUS:
+    possessive = TRUE;
+    min = 1;
+    max = INT_MAX;
+    ecode++;
+    goto REPEATTYPE;
+
+    case OP_TYPEPOSQUERY:
+    possessive = TRUE;
+    min = 0;
+    max = 1;
+    ecode++;
+    goto REPEATTYPE;
+
+    case OP_TYPEPOSUPTO:
+    possessive = TRUE;
+    min = 0;
+    max = GET2(ecode, 1);
+    ecode += 1 + IMM2_SIZE;
+    goto REPEATTYPE;
+
+    case OP_TYPESTAR:
+    case OP_TYPEMINSTAR:
+    case OP_TYPEPLUS:
+    case OP_TYPEMINPLUS:
+    case OP_TYPEQUERY:
+    case OP_TYPEMINQUERY:
+    c = *ecode++ - OP_TYPESTAR;
+    minimize = (c & 1) != 0;
+    min = rep_min[c];                 /* Pick up values from tables; */
+    max = rep_max[c];                 /* zero for max => infinity */
+    if (max == 0) max = INT_MAX;
+
+    /* Common code for all repeated single character type matches. Note that
+    in UTF-8 mode, '.' matches a character of any length, but for the other
+    character types, the valid characters are all one-byte long. */
+
+    REPEATTYPE:
+    ctype = *ecode++;      /* Code for the character type */
+
+#ifdef SUPPORT_UCP
+    if (ctype == OP_PROP || ctype == OP_NOTPROP)
+      {
+      prop_fail_result = ctype == OP_NOTPROP;
+      prop_type = *ecode++;
+      prop_value = *ecode++;
+      }
+    else prop_type = -1;
+#endif
+
+    /* First, ensure the minimum number of matches are present. Use inline
+    code for maximizing the speed, and do the type test once at the start
+    (i.e. keep it out of the loop). Separate the UTF-8 code completely as that
+    is tidier. Also separate the UCP code, which can be the same for both UTF-8
+    and single-bytes. */
+
+    if (min > 0)
+      {
+#ifdef SUPPORT_UCP
+      if (prop_type >= 0)
+        {
+        switch(prop_type)
+          {
+          case PT_ANY:
+          if (prop_fail_result) RRETURN(MATCH_NOMATCH);
+          for (i = 1; i <= min; i++)
+            {
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              RRETURN(MATCH_NOMATCH);
+              }
+            GETCHARINCTEST(c, eptr);
+            }
+          break;
+
+          case PT_LAMP:
+          for (i = 1; i <= min; i++)
+            {
+            int chartype;
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              RRETURN(MATCH_NOMATCH);
+              }
+            GETCHARINCTEST(c, eptr);
+            chartype = UCD_CHARTYPE(c);
+            if ((chartype == ucp_Lu ||
+                 chartype == ucp_Ll ||
+                 chartype == ucp_Lt) == prop_fail_result)
+              RRETURN(MATCH_NOMATCH);
+            }
+          break;
+
+          case PT_GC:
+          for (i = 1; i <= min; i++)
+            {
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              RRETURN(MATCH_NOMATCH);
+              }
+            GETCHARINCTEST(c, eptr);
+            if ((UCD_CATEGORY(c) == prop_value) == prop_fail_result)
+              RRETURN(MATCH_NOMATCH);
+            }
+          break;
+
+          case PT_PC:
+          for (i = 1; i <= min; i++)
+            {
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              RRETURN(MATCH_NOMATCH);
+              }
+            GETCHARINCTEST(c, eptr);
+            if ((UCD_CHARTYPE(c) == prop_value) == prop_fail_result)
+              RRETURN(MATCH_NOMATCH);
+            }
+          break;
+
+          case PT_SC:
+          for (i = 1; i <= min; i++)
+            {
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              RRETURN(MATCH_NOMATCH);
+              }
+            GETCHARINCTEST(c, eptr);
+            if ((UCD_SCRIPT(c) == prop_value) == prop_fail_result)
+              RRETURN(MATCH_NOMATCH);
+            }
+          break;
+
+          case PT_ALNUM:
+          for (i = 1; i <= min; i++)
+            {
+            int category;
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              RRETURN(MATCH_NOMATCH);
+              }
+            GETCHARINCTEST(c, eptr);
+            category = UCD_CATEGORY(c);
+            if ((category == ucp_L || category == ucp_N) == prop_fail_result)
+              RRETURN(MATCH_NOMATCH);
+            }
+          break;
+
+          case PT_SPACE:    /* Perl space */
+          for (i = 1; i <= min; i++)
+            {
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              RRETURN(MATCH_NOMATCH);
+              }
+            GETCHARINCTEST(c, eptr);
+            if ((UCD_CATEGORY(c) == ucp_Z || c == CHAR_HT || c == CHAR_NL ||
+                 c == CHAR_FF || c == CHAR_CR)
+                   == prop_fail_result)
+              RRETURN(MATCH_NOMATCH);
+            }
+          break;
+
+          case PT_PXSPACE:  /* POSIX space */
+          for (i = 1; i <= min; i++)
+            {
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              RRETURN(MATCH_NOMATCH);
+              }
+            GETCHARINCTEST(c, eptr);
+            if ((UCD_CATEGORY(c) == ucp_Z || c == CHAR_HT || c == CHAR_NL ||
+                 c == CHAR_VT || c == CHAR_FF || c == CHAR_CR)
+                   == prop_fail_result)
+              RRETURN(MATCH_NOMATCH);
+            }
+          break;
+
+          case PT_WORD:
+          for (i = 1; i <= min; i++)
+            {
+            int category;
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              RRETURN(MATCH_NOMATCH);
+              }
+            GETCHARINCTEST(c, eptr);
+            category = UCD_CATEGORY(c);
+            if ((category == ucp_L || category == ucp_N || c == CHAR_UNDERSCORE)
+                   == prop_fail_result)
+              RRETURN(MATCH_NOMATCH);
+            }
+          break;
+
+          case PT_CLIST:
+          for (i = 1; i <= min; i++)
+            {
+            const pcre_uint32 *cp;
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              RRETURN(MATCH_NOMATCH);
+              }
+            GETCHARINCTEST(c, eptr);
+            cp = PRIV(ucd_caseless_sets) + prop_value;
+            for (;;)
+              {
+              if (c < *cp)
+                { if (prop_fail_result) break; else { RRETURN(MATCH_NOMATCH); } }
+              if (c == *cp++)
+                { if (prop_fail_result) { RRETURN(MATCH_NOMATCH); } else break; }
+              }
+            }
+          break;
+
+          case PT_UCNC:
+          for (i = 1; i <= min; i++)
+            {
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              RRETURN(MATCH_NOMATCH);
+              }
+            GETCHARINCTEST(c, eptr);
+            if ((c == CHAR_DOLLAR_SIGN || c == CHAR_COMMERCIAL_AT ||
+                 c == CHAR_GRAVE_ACCENT || (c >= 0xa0 && c <= 0xd7ff) ||
+                 c >= 0xe000) == prop_fail_result)
+              RRETURN(MATCH_NOMATCH);
+            }
+          break;
+
+          /* This should not occur */
+
+          default:
+          RRETURN(PCRE_ERROR_INTERNAL);
+          }
+        }
+
+      /* Match extended Unicode sequences. We will get here only if the
+      support is in the binary; otherwise a compile-time error occurs. */
+
+      else if (ctype == OP_EXTUNI)
+        {
+        for (i = 1; i <= min; i++)
+          {
+          if (eptr >= md->end_subject)
+            {
+            SCHECK_PARTIAL();
+            RRETURN(MATCH_NOMATCH);
+            }
+          else
+            {
+            int lgb, rgb;
+            GETCHARINCTEST(c, eptr);
+            lgb = UCD_GRAPHBREAK(c);
+           while (eptr < md->end_subject)
+              {
+              int len = 1;
+              if (!utf) c = *eptr; else { GETCHARLEN(c, eptr, len); }
+              rgb = UCD_GRAPHBREAK(c);
+              if ((PRIV(ucp_gbtable)[lgb] & (1 << rgb)) == 0) break;
+              lgb = rgb;
+              eptr += len;
+              }
+            }
+          CHECK_PARTIAL();
+          }
+        }
+
+      else
+#endif     /* SUPPORT_UCP */
+
+/* Handle all other cases when the coding is UTF-8 */
+
+#ifdef SUPPORT_UTF
+      if (utf) switch(ctype)
+        {
+        case OP_ANY:
+        for (i = 1; i <= min; i++)
+          {
+          if (eptr >= md->end_subject)
+            {
+            SCHECK_PARTIAL();
+            RRETURN(MATCH_NOMATCH);
+            }
+          if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH);
+          if (md->partial != 0 &&
+              eptr + 1 >= md->end_subject &&
+              NLBLOCK->nltype == NLTYPE_FIXED &&
+              NLBLOCK->nllen == 2 &&
+              RAWUCHAR(eptr) == NLBLOCK->nl[0])
+            {
+            md->hitend = TRUE;
+            if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);
+            }
+          eptr++;
+          ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++);
+          }
+        break;
+
+        case OP_ALLANY:
+        for (i = 1; i <= min; i++)
+          {
+          if (eptr >= md->end_subject)
+            {
+            SCHECK_PARTIAL();
+            RRETURN(MATCH_NOMATCH);
+            }
+          eptr++;
+          ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++);
+          }
+        break;
+
+        case OP_ANYBYTE:
+        if (eptr > md->end_subject - min) RRETURN(MATCH_NOMATCH);
+        eptr += min;
+        break;
+
+        case OP_ANYNL:
+        for (i = 1; i <= min; i++)
+          {
+          if (eptr >= md->end_subject)
+            {
+            SCHECK_PARTIAL();
+            RRETURN(MATCH_NOMATCH);
+            }
+          GETCHARINC(c, eptr);
+          switch(c)
+            {
+            default: RRETURN(MATCH_NOMATCH);
+
+            case CHAR_CR:
+            if (eptr < md->end_subject && RAWUCHAR(eptr) == CHAR_LF) eptr++;
+            break;
+
+            case CHAR_LF:
+            break;
+
+            case CHAR_VT:
+            case CHAR_FF:
+            case CHAR_NEL:
+#ifndef EBCDIC
+            case 0x2028:
+            case 0x2029:
+#endif  /* Not EBCDIC */
+            if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
+            break;
+            }
+          }
+        break;
+
+        case OP_NOT_HSPACE:
+        for (i = 1; i <= min; i++)
+          {
+          if (eptr >= md->end_subject)
+            {
+            SCHECK_PARTIAL();
+            RRETURN(MATCH_NOMATCH);
+            }
+          GETCHARINC(c, eptr);
+          switch(c)
+            {
+            HSPACE_CASES: RRETURN(MATCH_NOMATCH);  /* Byte and multibyte cases */
+            default: break;
+            }
+          }
+        break;
+
+        case OP_HSPACE:
+        for (i = 1; i <= min; i++)
+          {
+          if (eptr >= md->end_subject)
+            {
+            SCHECK_PARTIAL();
+            RRETURN(MATCH_NOMATCH);
+            }
+          GETCHARINC(c, eptr);
+          switch(c)
+            {
+            HSPACE_CASES: break;  /* Byte and multibyte cases */
+            default: RRETURN(MATCH_NOMATCH);
+            }
+          }
+        break;
+
+        case OP_NOT_VSPACE:
+        for (i = 1; i <= min; i++)
+          {
+          if (eptr >= md->end_subject)
+            {
+            SCHECK_PARTIAL();
+            RRETURN(MATCH_NOMATCH);
+            }
+          GETCHARINC(c, eptr);
+          switch(c)
+            {
+            VSPACE_CASES: RRETURN(MATCH_NOMATCH);
+            default: break;
+            }
+          }
+        break;
+
+        case OP_VSPACE:
+        for (i = 1; i <= min; i++)
+          {
+          if (eptr >= md->end_subject)
+            {
+            SCHECK_PARTIAL();
+            RRETURN(MATCH_NOMATCH);
+            }
+          GETCHARINC(c, eptr);
+          switch(c)
+            {
+            VSPACE_CASES: break;
+            default: RRETURN(MATCH_NOMATCH);
+            }
+          }
+        break;
+
+        case OP_NOT_DIGIT:
+        for (i = 1; i <= min; i++)
+          {
+          if (eptr >= md->end_subject)
+            {
+            SCHECK_PARTIAL();
+            RRETURN(MATCH_NOMATCH);
+            }
+          GETCHARINC(c, eptr);
+          if (c < 128 && (md->ctypes[c] & ctype_digit) != 0)
+            RRETURN(MATCH_NOMATCH);
+          }
+        break;
+
+        case OP_DIGIT:
+        for (i = 1; i <= min; i++)
+          {
+          pcre_uint32 cc;
+          if (eptr >= md->end_subject)
+            {
+            SCHECK_PARTIAL();
+            RRETURN(MATCH_NOMATCH);
+            }
+          cc = RAWUCHAR(eptr);
+          if (cc >= 128 || (md->ctypes[cc] & ctype_digit) == 0)
+            RRETURN(MATCH_NOMATCH);
+          eptr++;
+          /* No need to skip more bytes - we know it's a 1-byte character */
+          }
+        break;
+
+        case OP_NOT_WHITESPACE:
+        for (i = 1; i <= min; i++)
+          {
+          pcre_uint32 cc;
+          if (eptr >= md->end_subject)
+            {
+            SCHECK_PARTIAL();
+            RRETURN(MATCH_NOMATCH);
+            }
+          cc = RAWUCHAR(eptr);
+          if (cc < 128 && (md->ctypes[cc] & ctype_space) != 0)
+            RRETURN(MATCH_NOMATCH);
+          eptr++;
+          ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++);
+          }
+        break;
+
+        case OP_WHITESPACE:
+        for (i = 1; i <= min; i++)
+          {
+          pcre_uint32 cc;
+          if (eptr >= md->end_subject)
+            {
+            SCHECK_PARTIAL();
+            RRETURN(MATCH_NOMATCH);
+            }
+          cc = RAWUCHAR(eptr);
+          if (cc >= 128 || (md->ctypes[cc] & ctype_space) == 0)
+            RRETURN(MATCH_NOMATCH);
+          eptr++;
+          /* No need to skip more bytes - we know it's a 1-byte character */
+          }
+        break;
+
+        case OP_NOT_WORDCHAR:
+        for (i = 1; i <= min; i++)
+          {
+          pcre_uint32 cc;
+          if (eptr >= md->end_subject)
+            {
+            SCHECK_PARTIAL();
+            RRETURN(MATCH_NOMATCH);
+            }
+          cc = RAWUCHAR(eptr);
+          if (cc < 128 && (md->ctypes[cc] & ctype_word) != 0)
+            RRETURN(MATCH_NOMATCH);
+          eptr++;
+          ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++);
+          }
+        break;
+
+        case OP_WORDCHAR:
+        for (i = 1; i <= min; i++)
+          {
+          pcre_uint32 cc;
+          if (eptr >= md->end_subject)
+            {
+            SCHECK_PARTIAL();
+            RRETURN(MATCH_NOMATCH);
+            }
+          cc = RAWUCHAR(eptr);
+          if (cc >= 128 || (md->ctypes[cc] & ctype_word) == 0)
+            RRETURN(MATCH_NOMATCH);
+          eptr++;
+          /* No need to skip more bytes - we know it's a 1-byte character */
+          }
+        break;
+
+        default:
+        RRETURN(PCRE_ERROR_INTERNAL);
+        }  /* End switch(ctype) */
+
+      else
+#endif     /* SUPPORT_UTF */
+
+      /* Code for the non-UTF-8 case for minimum matching of operators other
+      than OP_PROP and OP_NOTPROP. */
+
+      switch(ctype)
+        {
+        case OP_ANY:
+        for (i = 1; i <= min; i++)
+          {
+          if (eptr >= md->end_subject)
+            {
+            SCHECK_PARTIAL();
+            RRETURN(MATCH_NOMATCH);
+            }
+          if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH);
+          if (md->partial != 0 &&
+              eptr + 1 >= md->end_subject &&
+              NLBLOCK->nltype == NLTYPE_FIXED &&
+              NLBLOCK->nllen == 2 &&
+              *eptr == NLBLOCK->nl[0])
+            {
+            md->hitend = TRUE;
+            if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);
+            }
+          eptr++;
+          }
+        break;
+
+        case OP_ALLANY:
+        if (eptr > md->end_subject - min)
+          {
+          SCHECK_PARTIAL();
+          RRETURN(MATCH_NOMATCH);
+          }
+        eptr += min;
+        break;
+
+        case OP_ANYBYTE:
+        if (eptr > md->end_subject - min)
+          {
+          SCHECK_PARTIAL();
+          RRETURN(MATCH_NOMATCH);
+          }
+        eptr += min;
+        break;
+
+        case OP_ANYNL:
+        for (i = 1; i <= min; i++)
+          {
+          if (eptr >= md->end_subject)
+            {
+            SCHECK_PARTIAL();
+            RRETURN(MATCH_NOMATCH);
+            }
+          switch(*eptr++)
+            {
+            default: RRETURN(MATCH_NOMATCH);
+
+            case CHAR_CR:
+            if (eptr < md->end_subject && *eptr == CHAR_LF) eptr++;
+            break;
+
+            case CHAR_LF:
+            break;
+
+            case CHAR_VT:
+            case CHAR_FF:
+            case CHAR_NEL:
+#if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
+            case 0x2028:
+            case 0x2029:
+#endif
+            if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
+            break;
+            }
+          }
+        break;
+
+        case OP_NOT_HSPACE:
+        for (i = 1; i <= min; i++)
+          {
+          if (eptr >= md->end_subject)
+            {
+            SCHECK_PARTIAL();
+            RRETURN(MATCH_NOMATCH);
+            }
+          switch(*eptr++)
+            {
+            default: break;
+            HSPACE_BYTE_CASES:
+#if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
+            HSPACE_MULTIBYTE_CASES:
+#endif
+            RRETURN(MATCH_NOMATCH);
+            }
+          }
+        break;
+
+        case OP_HSPACE:
+        for (i = 1; i <= min; i++)
+          {
+          if (eptr >= md->end_subject)
+            {
+            SCHECK_PARTIAL();
+            RRETURN(MATCH_NOMATCH);
+            }
+          switch(*eptr++)
+            {
+            default: RRETURN(MATCH_NOMATCH);
+            HSPACE_BYTE_CASES:
+#if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
+            HSPACE_MULTIBYTE_CASES:
+#endif
+            break;
+            }
+          }
+        break;
+
+        case OP_NOT_VSPACE:
+        for (i = 1; i <= min; i++)
+          {
+          if (eptr >= md->end_subject)
+            {
+            SCHECK_PARTIAL();
+            RRETURN(MATCH_NOMATCH);
+            }
+          switch(*eptr++)
+            {
+            VSPACE_BYTE_CASES:
+#if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
+            VSPACE_MULTIBYTE_CASES:
+#endif
+            RRETURN(MATCH_NOMATCH);
+            default: break;
+            }
+          }
+        break;
+
+        case OP_VSPACE:
+        for (i = 1; i <= min; i++)
+          {
+          if (eptr >= md->end_subject)
+            {
+            SCHECK_PARTIAL();
+            RRETURN(MATCH_NOMATCH);
+            }
+          switch(*eptr++)
+            {
+            default: RRETURN(MATCH_NOMATCH);
+            VSPACE_BYTE_CASES:
+#if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
+            VSPACE_MULTIBYTE_CASES:
+#endif
+            break;
+            }
+          }
+        break;
+
+        case OP_NOT_DIGIT:
+        for (i = 1; i <= min; i++)
+          {
+          if (eptr >= md->end_subject)
+            {
+            SCHECK_PARTIAL();
+            RRETURN(MATCH_NOMATCH);
+            }
+          if (MAX_255(*eptr) && (md->ctypes[*eptr] & ctype_digit) != 0)
+            RRETURN(MATCH_NOMATCH);
+          eptr++;
+          }
+        break;
+
+        case OP_DIGIT:
+        for (i = 1; i <= min; i++)
+          {
+          if (eptr >= md->end_subject)
+            {
+            SCHECK_PARTIAL();
+            RRETURN(MATCH_NOMATCH);
+            }
+          if (!MAX_255(*eptr) || (md->ctypes[*eptr] & ctype_digit) == 0)
+            RRETURN(MATCH_NOMATCH);
+          eptr++;
+          }
+        break;
+
+        case OP_NOT_WHITESPACE:
+        for (i = 1; i <= min; i++)
+          {
+          if (eptr >= md->end_subject)
+            {
+            SCHECK_PARTIAL();
+            RRETURN(MATCH_NOMATCH);
+            }
+          if (MAX_255(*eptr) && (md->ctypes[*eptr] & ctype_space) != 0)
+            RRETURN(MATCH_NOMATCH);
+          eptr++;
+          }
+        break;
+
+        case OP_WHITESPACE:
+        for (i = 1; i <= min; i++)
+          {
+          if (eptr >= md->end_subject)
+            {
+            SCHECK_PARTIAL();
+            RRETURN(MATCH_NOMATCH);
+            }
+          if (!MAX_255(*eptr) || (md->ctypes[*eptr] & ctype_space) == 0)
+            RRETURN(MATCH_NOMATCH);
+          eptr++;
+          }
+        break;
+
+        case OP_NOT_WORDCHAR:
+        for (i = 1; i <= min; i++)
+          {
+          if (eptr >= md->end_subject)
+            {
+            SCHECK_PARTIAL();
+            RRETURN(MATCH_NOMATCH);
+            }
+          if (MAX_255(*eptr) && (md->ctypes[*eptr] & ctype_word) != 0)
+            RRETURN(MATCH_NOMATCH);
+          eptr++;
+          }
+        break;
+
+        case OP_WORDCHAR:
+        for (i = 1; i <= min; i++)
+          {
+          if (eptr >= md->end_subject)
+            {
+            SCHECK_PARTIAL();
+            RRETURN(MATCH_NOMATCH);
+            }
+          if (!MAX_255(*eptr) || (md->ctypes[*eptr] & ctype_word) == 0)
+            RRETURN(MATCH_NOMATCH);
+          eptr++;
+          }
+        break;
+
+        default:
+        RRETURN(PCRE_ERROR_INTERNAL);
+        }
+      }
+
+    /* If min = max, continue at the same level without recursing */
+
+    if (min == max) continue;
+
+    /* If minimizing, we have to test the rest of the pattern before each
+    subsequent match. Again, separate the UTF-8 case for speed, and also
+    separate the UCP cases. */
+
+    if (minimize)
+      {
+#ifdef SUPPORT_UCP
+      if (prop_type >= 0)
+        {
+        switch(prop_type)
+          {
+          case PT_ANY:
+          for (fi = min;; fi++)
+            {
+            RMATCH(eptr, ecode, offset_top, md, eptrb, RM36);
+            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+            if (fi >= max) RRETURN(MATCH_NOMATCH);
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              RRETURN(MATCH_NOMATCH);
+              }
+            GETCHARINCTEST(c, eptr);
+            if (prop_fail_result) RRETURN(MATCH_NOMATCH);
+            }
+          /* Control never gets here */
+
+          case PT_LAMP:
+          for (fi = min;; fi++)
+            {
+            int chartype;
+            RMATCH(eptr, ecode, offset_top, md, eptrb, RM37);
+            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+            if (fi >= max) RRETURN(MATCH_NOMATCH);
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              RRETURN(MATCH_NOMATCH);
+              }
+            GETCHARINCTEST(c, eptr);
+            chartype = UCD_CHARTYPE(c);
+            if ((chartype == ucp_Lu ||
+                 chartype == ucp_Ll ||
+                 chartype == ucp_Lt) == prop_fail_result)
+              RRETURN(MATCH_NOMATCH);
+            }
+          /* Control never gets here */
+
+          case PT_GC:
+          for (fi = min;; fi++)
+            {
+            RMATCH(eptr, ecode, offset_top, md, eptrb, RM38);
+            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+            if (fi >= max) RRETURN(MATCH_NOMATCH);
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              RRETURN(MATCH_NOMATCH);
+              }
+            GETCHARINCTEST(c, eptr);
+            if ((UCD_CATEGORY(c) == prop_value) == prop_fail_result)
+              RRETURN(MATCH_NOMATCH);
+            }
+          /* Control never gets here */
+
+          case PT_PC:
+          for (fi = min;; fi++)
+            {
+            RMATCH(eptr, ecode, offset_top, md, eptrb, RM39);
+            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+            if (fi >= max) RRETURN(MATCH_NOMATCH);
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              RRETURN(MATCH_NOMATCH);
+              }
+            GETCHARINCTEST(c, eptr);
+            if ((UCD_CHARTYPE(c) == prop_value) == prop_fail_result)
+              RRETURN(MATCH_NOMATCH);
+            }
+          /* Control never gets here */
+
+          case PT_SC:
+          for (fi = min;; fi++)
+            {
+            RMATCH(eptr, ecode, offset_top, md, eptrb, RM40);
+            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+            if (fi >= max) RRETURN(MATCH_NOMATCH);
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              RRETURN(MATCH_NOMATCH);
+              }
+            GETCHARINCTEST(c, eptr);
+            if ((UCD_SCRIPT(c) == prop_value) == prop_fail_result)
+              RRETURN(MATCH_NOMATCH);
+            }
+          /* Control never gets here */
+
+          case PT_ALNUM:
+          for (fi = min;; fi++)
+            {
+            int category;
+            RMATCH(eptr, ecode, offset_top, md, eptrb, RM59);
+            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+            if (fi >= max) RRETURN(MATCH_NOMATCH);
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              RRETURN(MATCH_NOMATCH);
+              }
+            GETCHARINCTEST(c, eptr);
+            category = UCD_CATEGORY(c);
+            if ((category == ucp_L || category == ucp_N) == prop_fail_result)
+              RRETURN(MATCH_NOMATCH);
+            }
+          /* Control never gets here */
+
+          case PT_SPACE:    /* Perl space */
+          for (fi = min;; fi++)
+            {
+            RMATCH(eptr, ecode, offset_top, md, eptrb, RM60);
+            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+            if (fi >= max) RRETURN(MATCH_NOMATCH);
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              RRETURN(MATCH_NOMATCH);
+              }
+            GETCHARINCTEST(c, eptr);
+            if ((UCD_CATEGORY(c) == ucp_Z || c == CHAR_HT || c == CHAR_NL ||
+                 c == CHAR_FF || c == CHAR_CR)
+                   == prop_fail_result)
+              RRETURN(MATCH_NOMATCH);
+            }
+          /* Control never gets here */
+
+          case PT_PXSPACE:  /* POSIX space */
+          for (fi = min;; fi++)
+            {
+            RMATCH(eptr, ecode, offset_top, md, eptrb, RM61);
+            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+            if (fi >= max) RRETURN(MATCH_NOMATCH);
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              RRETURN(MATCH_NOMATCH);
+              }
+            GETCHARINCTEST(c, eptr);
+            if ((UCD_CATEGORY(c) == ucp_Z || c == CHAR_HT || c == CHAR_NL ||
+                 c == CHAR_VT || c == CHAR_FF || c == CHAR_CR)
+                   == prop_fail_result)
+              RRETURN(MATCH_NOMATCH);
+            }
+          /* Control never gets here */
+
+          case PT_WORD:
+          for (fi = min;; fi++)
+            {
+            int category;
+            RMATCH(eptr, ecode, offset_top, md, eptrb, RM62);
+            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+            if (fi >= max) RRETURN(MATCH_NOMATCH);
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              RRETURN(MATCH_NOMATCH);
+              }
+            GETCHARINCTEST(c, eptr);
+            category = UCD_CATEGORY(c);
+            if ((category == ucp_L ||
+                 category == ucp_N ||
+                 c == CHAR_UNDERSCORE)
+                   == prop_fail_result)
+              RRETURN(MATCH_NOMATCH);
+            }
+          /* Control never gets here */
+
+          case PT_CLIST:
+          for (fi = min;; fi++)
+            {
+            const pcre_uint32 *cp;
+            RMATCH(eptr, ecode, offset_top, md, eptrb, RM67);
+            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+            if (fi >= max) RRETURN(MATCH_NOMATCH);
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              RRETURN(MATCH_NOMATCH);
+              }
+            GETCHARINCTEST(c, eptr);
+            cp = PRIV(ucd_caseless_sets) + prop_value;
+            for (;;)
+              {
+              if (c < *cp)
+                { if (prop_fail_result) break; else { RRETURN(MATCH_NOMATCH); } }
+              if (c == *cp++)
+                { if (prop_fail_result) { RRETURN(MATCH_NOMATCH); } else break; }
+              }
+            }
+          /* Control never gets here */
+
+          case PT_UCNC:
+          for (fi = min;; fi++)
+            {
+            RMATCH(eptr, ecode, offset_top, md, eptrb, RM68);
+            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+            if (fi >= max) RRETURN(MATCH_NOMATCH);
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              RRETURN(MATCH_NOMATCH);
+              }
+            GETCHARINCTEST(c, eptr);
+            if ((c == CHAR_DOLLAR_SIGN || c == CHAR_COMMERCIAL_AT ||
+                 c == CHAR_GRAVE_ACCENT || (c >= 0xa0 && c <= 0xd7ff) ||
+                 c >= 0xe000) == prop_fail_result)
+              RRETURN(MATCH_NOMATCH);
+            }
+          /* Control never gets here */
+
+          /* This should never occur */
+          default:
+          RRETURN(PCRE_ERROR_INTERNAL);
+          }
+        }
+
+      /* Match extended Unicode sequences. We will get here only if the
+      support is in the binary; otherwise a compile-time error occurs. */
+
+      else if (ctype == OP_EXTUNI)
+        {
+        for (fi = min;; fi++)
+          {
+          RMATCH(eptr, ecode, offset_top, md, eptrb, RM41);
+          if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+          if (fi >= max) RRETURN(MATCH_NOMATCH);
+          if (eptr >= md->end_subject)
+            {
+            SCHECK_PARTIAL();
+            RRETURN(MATCH_NOMATCH);
+            }
+          else
+            {
+            int lgb, rgb;
+            GETCHARINCTEST(c, eptr);
+            lgb = UCD_GRAPHBREAK(c);
+            while (eptr < md->end_subject)
+              {
+              int len = 1;
+              if (!utf) c = *eptr; else { GETCHARLEN(c, eptr, len); }
+              rgb = UCD_GRAPHBREAK(c);
+              if ((PRIV(ucp_gbtable)[lgb] & (1 << rgb)) == 0) break;
+              lgb = rgb;
+              eptr += len;
+              }
+            }
+          CHECK_PARTIAL();
+          }
+        }
+      else
+#endif     /* SUPPORT_UCP */
+
+#ifdef SUPPORT_UTF
+      if (utf)
+        {
+        for (fi = min;; fi++)
+          {
+          RMATCH(eptr, ecode, offset_top, md, eptrb, RM42);
+          if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+          if (fi >= max) RRETURN(MATCH_NOMATCH);
+          if (eptr >= md->end_subject)
+            {
+            SCHECK_PARTIAL();
+            RRETURN(MATCH_NOMATCH);
+            }
+          if (ctype == OP_ANY && IS_NEWLINE(eptr))
+            RRETURN(MATCH_NOMATCH);
+          GETCHARINC(c, eptr);
+          switch(ctype)
+            {
+            case OP_ANY:               /* This is the non-NL case */
+            if (md->partial != 0 &&    /* Take care with CRLF partial */
+                eptr >= md->end_subject &&
+                NLBLOCK->nltype == NLTYPE_FIXED &&
+                NLBLOCK->nllen == 2 &&
+                c == NLBLOCK->nl[0])
+              {
+              md->hitend = TRUE;
+              if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);
+              }
+            break;
+
+            case OP_ALLANY:
+            case OP_ANYBYTE:
+            break;
+
+            case OP_ANYNL:
+            switch(c)
+              {
+              default: RRETURN(MATCH_NOMATCH);
+              case CHAR_CR:
+              if (eptr < md->end_subject && RAWUCHAR(eptr) == CHAR_LF) eptr++;
+              break;
+
+              case CHAR_LF:
+              break;
+
+              case CHAR_VT:
+              case CHAR_FF:
+              case CHAR_NEL:
+#ifndef EBCDIC
+              case 0x2028:
+              case 0x2029:
+#endif  /* Not EBCDIC */
+              if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
+              break;
+              }
+            break;
+
+            case OP_NOT_HSPACE:
+            switch(c)
+              {
+              HSPACE_CASES: RRETURN(MATCH_NOMATCH);
+              default: break;
+              }
+            break;
+
+            case OP_HSPACE:
+            switch(c)
+              {
+              HSPACE_CASES: break;
+              default: RRETURN(MATCH_NOMATCH);
+              }
+            break;
+
+            case OP_NOT_VSPACE:
+            switch(c)
+              {
+              VSPACE_CASES: RRETURN(MATCH_NOMATCH);
+              default: break;
+              }
+            break;
+
+            case OP_VSPACE:
+            switch(c)
+              {
+              VSPACE_CASES: break;
+              default: RRETURN(MATCH_NOMATCH);
+              }
+            break;
+
+            case OP_NOT_DIGIT:
+            if (c < 256 && (md->ctypes[c] & ctype_digit) != 0)
+              RRETURN(MATCH_NOMATCH);
+            break;
+
+            case OP_DIGIT:
+            if (c >= 256 || (md->ctypes[c] & ctype_digit) == 0)
+              RRETURN(MATCH_NOMATCH);
+            break;
+
+            case OP_NOT_WHITESPACE:
+            if (c < 256 && (md->ctypes[c] & ctype_space) != 0)
+              RRETURN(MATCH_NOMATCH);
+            break;
+
+            case OP_WHITESPACE:
+            if (c >= 256 || (md->ctypes[c] & ctype_space) == 0)
+              RRETURN(MATCH_NOMATCH);
+            break;
+
+            case OP_NOT_WORDCHAR:
+            if (c < 256 && (md->ctypes[c] & ctype_word) != 0)
+              RRETURN(MATCH_NOMATCH);
+            break;
+
+            case OP_WORDCHAR:
+            if (c >= 256 || (md->ctypes[c] & ctype_word) == 0)
+              RRETURN(MATCH_NOMATCH);
+            break;
+
+            default:
+            RRETURN(PCRE_ERROR_INTERNAL);
+            }
+          }
+        }
+      else
+#endif
+      /* Not UTF mode */
+        {
+        for (fi = min;; fi++)
+          {
+          RMATCH(eptr, ecode, offset_top, md, eptrb, RM43);
+          if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+          if (fi >= max) RRETURN(MATCH_NOMATCH);
+          if (eptr >= md->end_subject)
+            {
+            SCHECK_PARTIAL();
+            RRETURN(MATCH_NOMATCH);
+            }
+          if (ctype == OP_ANY && IS_NEWLINE(eptr))
+            RRETURN(MATCH_NOMATCH);
+          c = *eptr++;
+          switch(ctype)
+            {
+            case OP_ANY:               /* This is the non-NL case */
+            if (md->partial != 0 &&    /* Take care with CRLF partial */
+                eptr >= md->end_subject &&
+                NLBLOCK->nltype == NLTYPE_FIXED &&
+                NLBLOCK->nllen == 2 &&
+                c == NLBLOCK->nl[0])
+              {
+              md->hitend = TRUE;
+              if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);
+              }
+            break;
+
+            case OP_ALLANY:
+            case OP_ANYBYTE:
+            break;
+
+            case OP_ANYNL:
+            switch(c)
+              {
+              default: RRETURN(MATCH_NOMATCH);
+              case CHAR_CR:
+              if (eptr < md->end_subject && *eptr == CHAR_LF) eptr++;
+              break;
+
+              case CHAR_LF:
+              break;
+
+              case CHAR_VT:
+              case CHAR_FF:
+              case CHAR_NEL:
+#if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
+              case 0x2028:
+              case 0x2029:
+#endif
+              if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
+              break;
+              }
+            break;
+
+            case OP_NOT_HSPACE:
+            switch(c)
+              {
+              default: break;
+              HSPACE_BYTE_CASES:
+#if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
+              HSPACE_MULTIBYTE_CASES:
+#endif
+              RRETURN(MATCH_NOMATCH);
+              }
+            break;
+
+            case OP_HSPACE:
+            switch(c)
+              {
+              default: RRETURN(MATCH_NOMATCH);
+              HSPACE_BYTE_CASES:
+#if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
+              HSPACE_MULTIBYTE_CASES:
+#endif
+              break;
+              }
+            break;
+
+            case OP_NOT_VSPACE:
+            switch(c)
+              {
+              default: break;
+              VSPACE_BYTE_CASES:
+#if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
+              VSPACE_MULTIBYTE_CASES:
+#endif
+              RRETURN(MATCH_NOMATCH);
+              }
+            break;
+
+            case OP_VSPACE:
+            switch(c)
+              {
+              default: RRETURN(MATCH_NOMATCH);
+              VSPACE_BYTE_CASES:
+#if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
+              VSPACE_MULTIBYTE_CASES:
+#endif
+              break;
+              }
+            break;
+
+            case OP_NOT_DIGIT:
+            if (MAX_255(c) && (md->ctypes[c] & ctype_digit) != 0) RRETURN(MATCH_NOMATCH);
+            break;
+
+            case OP_DIGIT:
+            if (!MAX_255(c) || (md->ctypes[c] & ctype_digit) == 0) RRETURN(MATCH_NOMATCH);
+            break;
+
+            case OP_NOT_WHITESPACE:
+            if (MAX_255(c) && (md->ctypes[c] & ctype_space) != 0) RRETURN(MATCH_NOMATCH);
+            break;
+
+            case OP_WHITESPACE:
+            if (!MAX_255(c) || (md->ctypes[c] & ctype_space) == 0) RRETURN(MATCH_NOMATCH);
+            break;
+
+            case OP_NOT_WORDCHAR:
+            if (MAX_255(c) && (md->ctypes[c] & ctype_word) != 0) RRETURN(MATCH_NOMATCH);
+            break;
+
+            case OP_WORDCHAR:
+            if (!MAX_255(c) || (md->ctypes[c] & ctype_word) == 0) RRETURN(MATCH_NOMATCH);
+            break;
+
+            default:
+            RRETURN(PCRE_ERROR_INTERNAL);
+            }
+          }
+        }
+      /* Control never gets here */
+      }
+
+    /* If maximizing, it is worth using inline code for speed, doing the type
+    test once at the start (i.e. keep it out of the loop). Again, keep the
+    UTF-8 and UCP stuff separate. */
+
+    else
+      {
+      pp = eptr;  /* Remember where we started */
+
+#ifdef SUPPORT_UCP
+      if (prop_type >= 0)
+        {
+        switch(prop_type)
+          {
+          case PT_ANY:
+          for (i = min; i < max; i++)
+            {
+            int len = 1;
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              break;
+              }
+            GETCHARLENTEST(c, eptr, len);
+            if (prop_fail_result) break;
+            eptr+= len;
+            }
+          break;
+
+          case PT_LAMP:
+          for (i = min; i < max; i++)
+            {
+            int chartype;
+            int len = 1;
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              break;
+              }
+            GETCHARLENTEST(c, eptr, len);
+            chartype = UCD_CHARTYPE(c);
+            if ((chartype == ucp_Lu ||
+                 chartype == ucp_Ll ||
+                 chartype == ucp_Lt) == prop_fail_result)
+              break;
+            eptr+= len;
+            }
+          break;
+
+          case PT_GC:
+          for (i = min; i < max; i++)
+            {
+            int len = 1;
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              break;
+              }
+            GETCHARLENTEST(c, eptr, len);
+            if ((UCD_CATEGORY(c) == prop_value) == prop_fail_result) break;
+            eptr+= len;
+            }
+          break;
+
+          case PT_PC:
+          for (i = min; i < max; i++)
+            {
+            int len = 1;
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              break;
+              }
+            GETCHARLENTEST(c, eptr, len);
+            if ((UCD_CHARTYPE(c) == prop_value) == prop_fail_result) break;
+            eptr+= len;
+            }
+          break;
+
+          case PT_SC:
+          for (i = min; i < max; i++)
+            {
+            int len = 1;
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              break;
+              }
+            GETCHARLENTEST(c, eptr, len);
+            if ((UCD_SCRIPT(c) == prop_value) == prop_fail_result) break;
+            eptr+= len;
+            }
+          break;
+
+          case PT_ALNUM:
+          for (i = min; i < max; i++)
+            {
+            int category;
+            int len = 1;
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              break;
+              }
+            GETCHARLENTEST(c, eptr, len);
+            category = UCD_CATEGORY(c);
+            if ((category == ucp_L || category == ucp_N) == prop_fail_result)
+              break;
+            eptr+= len;
+            }
+          break;
+
+          case PT_SPACE:    /* Perl space */
+          for (i = min; i < max; i++)
+            {
+            int len = 1;
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              break;
+              }
+            GETCHARLENTEST(c, eptr, len);
+            if ((UCD_CATEGORY(c) == ucp_Z || c == CHAR_HT || c == CHAR_NL ||
+                 c == CHAR_FF || c == CHAR_CR)
+                 == prop_fail_result)
+              break;
+            eptr+= len;
+            }
+          break;
+
+          case PT_PXSPACE:  /* POSIX space */
+          for (i = min; i < max; i++)
+            {
+            int len = 1;
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              break;
+              }
+            GETCHARLENTEST(c, eptr, len);
+            if ((UCD_CATEGORY(c) == ucp_Z || c == CHAR_HT || c == CHAR_NL ||
+                 c == CHAR_VT || c == CHAR_FF || c == CHAR_CR)
+                 == prop_fail_result)
+              break;
+            eptr+= len;
+            }
+          break;
+
+          case PT_WORD:
+          for (i = min; i < max; i++)
+            {
+            int category;
+            int len = 1;
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              break;
+              }
+            GETCHARLENTEST(c, eptr, len);
+            category = UCD_CATEGORY(c);
+            if ((category == ucp_L || category == ucp_N ||
+                 c == CHAR_UNDERSCORE) == prop_fail_result)
+              break;
+            eptr+= len;
+            }
+          break;
+
+          case PT_CLIST:
+          for (i = min; i < max; i++)
+            {
+            const pcre_uint32 *cp;
+            int len = 1;
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              break;
+              }
+            GETCHARLENTEST(c, eptr, len);
+            cp = PRIV(ucd_caseless_sets) + prop_value;
+            for (;;)
+              {
+              if (c < *cp)
+                { if (prop_fail_result) break; else goto GOT_MAX; }
+              if (c == *cp++)
+                { if (prop_fail_result) goto GOT_MAX; else break; }
+              }
+            eptr += len;
+            }
+          GOT_MAX:
+          break;
+
+          case PT_UCNC:
+          for (i = min; i < max; i++)
+            {
+            int len = 1;
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              break;
+              }
+            GETCHARLENTEST(c, eptr, len);
+            if ((c == CHAR_DOLLAR_SIGN || c == CHAR_COMMERCIAL_AT ||
+                 c == CHAR_GRAVE_ACCENT || (c >= 0xa0 && c <= 0xd7ff) ||
+                 c >= 0xe000) == prop_fail_result)
+              break;
+            eptr += len;
+            }
+          break;
+
+          default:
+          RRETURN(PCRE_ERROR_INTERNAL);
+          }
+
+        /* eptr is now past the end of the maximum run */
+
+        if (possessive) continue;    /* No backtracking */
+        for(;;)
+          {
+          if (eptr == pp) goto TAIL_RECURSE;
+          RMATCH(eptr, ecode, offset_top, md, eptrb, RM44);
+          if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+          eptr--;
+          if (utf) BACKCHAR(eptr);
+          }
+        }
+
+      /* Match extended Unicode sequences. We will get here only if the
+      support is in the binary; otherwise a compile-time error occurs. */
+
+      else if (ctype == OP_EXTUNI)
+        {
+        for (i = min; i < max; i++)
+          {
+          if (eptr >= md->end_subject)
+            {
+            SCHECK_PARTIAL();
+            break;
+            }
+          else
+            {
+            int lgb, rgb;
+            GETCHARINCTEST(c, eptr);
+            lgb = UCD_GRAPHBREAK(c);
+            while (eptr < md->end_subject)
+              {
+              int len = 1;
+              if (!utf) c = *eptr; else { GETCHARLEN(c, eptr, len); }
+              rgb = UCD_GRAPHBREAK(c);
+              if ((PRIV(ucp_gbtable)[lgb] & (1 << rgb)) == 0) break;
+              lgb = rgb;
+              eptr += len;
+              }
+            }
+          CHECK_PARTIAL();
+          }
+
+        /* eptr is now past the end of the maximum run */
+
+        if (possessive) continue;    /* No backtracking */
+        for(;;)
+          {
+          if (eptr == pp) goto TAIL_RECURSE;
+          RMATCH(eptr, ecode, offset_top, md, eptrb, RM45);
+          if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+          eptr--;
+          for (;;)                        /* Move back over one extended */
+            {
+            if (!utf) c = *eptr; else
+              {
+              BACKCHAR(eptr);
+              GETCHAR(c, eptr);
+              }
+            if (UCD_CATEGORY(c) != ucp_M) break;
+            eptr--;
+            }
+          }
+        }
+
+      else
+#endif   /* SUPPORT_UCP */
+
+#ifdef SUPPORT_UTF
+      if (utf)
+        {
+        switch(ctype)
+          {
+          case OP_ANY:
+          if (max < INT_MAX)
+            {
+            for (i = min; i < max; i++)
+              {
+              if (eptr >= md->end_subject)
+                {
+                SCHECK_PARTIAL();
+                break;
+                }
+              if (IS_NEWLINE(eptr)) break;
+              if (md->partial != 0 &&    /* Take care with CRLF partial */
+                  eptr + 1 >= md->end_subject &&
+                  NLBLOCK->nltype == NLTYPE_FIXED &&
+                  NLBLOCK->nllen == 2 &&
+                  RAWUCHAR(eptr) == NLBLOCK->nl[0])
+                {
+                md->hitend = TRUE;
+                if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);
+                }
+              eptr++;
+              ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++);
+              }
+            }
+
+          /* Handle unlimited UTF-8 repeat */
+
+          else
+            {
+            for (i = min; i < max; i++)
+              {
+              if (eptr >= md->end_subject)
+                {
+                SCHECK_PARTIAL();
+                break;
+                }
+              if (IS_NEWLINE(eptr)) break;
+              if (md->partial != 0 &&    /* Take care with CRLF partial */
+                  eptr + 1 >= md->end_subject &&
+                  NLBLOCK->nltype == NLTYPE_FIXED &&
+                  NLBLOCK->nllen == 2 &&
+                  RAWUCHAR(eptr) == NLBLOCK->nl[0])
+                {
+                md->hitend = TRUE;
+                if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);
+                }
+              eptr++;
+              ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++);
+              }
+            }
+          break;
+
+          case OP_ALLANY:
+          if (max < INT_MAX)
+            {
+            for (i = min; i < max; i++)
+              {
+              if (eptr >= md->end_subject)
+                {
+                SCHECK_PARTIAL();
+                break;
+                }
+              eptr++;
+              ACROSSCHAR(eptr < md->end_subject, *eptr, eptr++);
+              }
+            }
+          else
+            {
+            eptr = md->end_subject;   /* Unlimited UTF-8 repeat */
+            SCHECK_PARTIAL();
+            }
+          break;
+
+          /* The byte case is the same as non-UTF8 */
+
+          case OP_ANYBYTE:
+          c = max - min;
+          if (c > (unsigned int)(md->end_subject - eptr))
+            {
+            eptr = md->end_subject;
+            SCHECK_PARTIAL();
+            }
+          else eptr += c;
+          break;
+
+          case OP_ANYNL:
+          for (i = min; i < max; i++)
+            {
+            int len = 1;
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              break;
+              }
+            GETCHARLEN(c, eptr, len);
+            if (c == CHAR_CR)
+              {
+              if (++eptr >= md->end_subject) break;
+              if (RAWUCHAR(eptr) == CHAR_LF) eptr++;
+              }
+            else
+              {
+              if (c != CHAR_LF &&
+                  (md->bsr_anycrlf ||
+                   (c != CHAR_VT && c != CHAR_FF && c != CHAR_NEL
+#ifndef EBCDIC
+                    && c != 0x2028 && c != 0x2029
+#endif  /* Not EBCDIC */
+                    )))
+                break;
+              eptr += len;
+              }
+            }
+          break;
+
+          case OP_NOT_HSPACE:
+          case OP_HSPACE:
+          for (i = min; i < max; i++)
+            {
+            BOOL gotspace;
+            int len = 1;
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              break;
+              }
+            GETCHARLEN(c, eptr, len);
+            switch(c)
+              {
+              HSPACE_CASES: gotspace = TRUE; break;
+              default: gotspace = FALSE; break;
+              }
+            if (gotspace == (ctype == OP_NOT_HSPACE)) break;
+            eptr += len;
+            }
+          break;
+
+          case OP_NOT_VSPACE:
+          case OP_VSPACE:
+          for (i = min; i < max; i++)
+            {
+            BOOL gotspace;
+            int len = 1;
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              break;
+              }
+            GETCHARLEN(c, eptr, len);
+            switch(c)
+              {
+              VSPACE_CASES: gotspace = TRUE; break;
+              default: gotspace = FALSE; break;
+              }
+            if (gotspace == (ctype == OP_NOT_VSPACE)) break;
+            eptr += len;
+            }
+          break;
+
+          case OP_NOT_DIGIT:
+          for (i = min; i < max; i++)
+            {
+            int len = 1;
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              break;
+              }
+            GETCHARLEN(c, eptr, len);
+            if (c < 256 && (md->ctypes[c] & ctype_digit) != 0) break;
+            eptr+= len;
+            }
+          break;
+
+          case OP_DIGIT:
+          for (i = min; i < max; i++)
+            {
+            int len = 1;
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              break;
+              }
+            GETCHARLEN(c, eptr, len);
+            if (c >= 256 ||(md->ctypes[c] & ctype_digit) == 0) break;
+            eptr+= len;
+            }
+          break;
+
+          case OP_NOT_WHITESPACE:
+          for (i = min; i < max; i++)
+            {
+            int len = 1;
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              break;
+              }
+            GETCHARLEN(c, eptr, len);
+            if (c < 256 && (md->ctypes[c] & ctype_space) != 0) break;
+            eptr+= len;
+            }
+          break;
+
+          case OP_WHITESPACE:
+          for (i = min; i < max; i++)
+            {
+            int len = 1;
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              break;
+              }
+            GETCHARLEN(c, eptr, len);
+            if (c >= 256 ||(md->ctypes[c] & ctype_space) == 0) break;
+            eptr+= len;
+            }
+          break;
+
+          case OP_NOT_WORDCHAR:
+          for (i = min; i < max; i++)
+            {
+            int len = 1;
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              break;
+              }
+            GETCHARLEN(c, eptr, len);
+            if (c < 256 && (md->ctypes[c] & ctype_word) != 0) break;
+            eptr+= len;
+            }
+          break;
+
+          case OP_WORDCHAR:
+          for (i = min; i < max; i++)
+            {
+            int len = 1;
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              break;
+              }
+            GETCHARLEN(c, eptr, len);
+            if (c >= 256 || (md->ctypes[c] & ctype_word) == 0) break;
+            eptr+= len;
+            }
+          break;
+
+          default:
+          RRETURN(PCRE_ERROR_INTERNAL);
+          }
+
+        if (possessive) continue;    /* No backtracking */
+        for(;;)
+          {
+          if (eptr == pp) goto TAIL_RECURSE;
+          RMATCH(eptr, ecode, offset_top, md, eptrb, RM46);
+          if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+          eptr--;
+          BACKCHAR(eptr);
+          if (ctype == OP_ANYNL && eptr > pp  && RAWUCHAR(eptr) == CHAR_NL &&
+              RAWUCHAR(eptr - 1) == CHAR_CR) eptr--;
+          }
+        }
+      else
+#endif  /* SUPPORT_UTF */
+      /* Not UTF mode */
+        {
+        switch(ctype)
+          {
+          case OP_ANY:
+          for (i = min; i < max; i++)
+            {
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              break;
+              }
+            if (IS_NEWLINE(eptr)) break;
+            if (md->partial != 0 &&    /* Take care with CRLF partial */
+                eptr + 1 >= md->end_subject &&
+                NLBLOCK->nltype == NLTYPE_FIXED &&
+                NLBLOCK->nllen == 2 &&
+                *eptr == NLBLOCK->nl[0])
+              {
+              md->hitend = TRUE;
+              if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);
+              }
+            eptr++;
+            }
+          break;
+
+          case OP_ALLANY:
+          case OP_ANYBYTE:
+          c = max - min;
+          if (c > (unsigned int)(md->end_subject - eptr))
+            {
+            eptr = md->end_subject;
+            SCHECK_PARTIAL();
+            }
+          else eptr += c;
+          break;
+
+          case OP_ANYNL:
+          for (i = min; i < max; i++)
+            {
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              break;
+              }
+            c = *eptr;
+            if (c == CHAR_CR)
+              {
+              if (++eptr >= md->end_subject) break;
+              if (*eptr == CHAR_LF) eptr++;
+              }
+            else
+              {
+              if (c != CHAR_LF && (md->bsr_anycrlf ||
+                 (c != CHAR_VT && c != CHAR_FF && c != CHAR_NEL
+#if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
+                 && c != 0x2028 && c != 0x2029
+#endif
+                 ))) break;
+              eptr++;
+              }
+            }
+          break;
+
+          case OP_NOT_HSPACE:
+          for (i = min; i < max; i++)
+            {
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              break;
+              }
+            switch(*eptr)
+              {
+              default: eptr++; break;
+              HSPACE_BYTE_CASES:
+#if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
+              HSPACE_MULTIBYTE_CASES:
+#endif
+              goto ENDLOOP00;
+              }
+            }
+          ENDLOOP00:
+          break;
+
+          case OP_HSPACE:
+          for (i = min; i < max; i++)
+            {
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              break;
+              }
+            switch(*eptr)
+              {
+              default: goto ENDLOOP01;
+              HSPACE_BYTE_CASES:
+#if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
+              HSPACE_MULTIBYTE_CASES:
+#endif
+              eptr++; break;
+              }
+            }
+          ENDLOOP01:
+          break;
+
+          case OP_NOT_VSPACE:
+          for (i = min; i < max; i++)
+            {
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              break;
+              }
+            switch(*eptr)
+              {
+              default: eptr++; break;
+              VSPACE_BYTE_CASES:
+#if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
+              VSPACE_MULTIBYTE_CASES:
+#endif
+              goto ENDLOOP02;
+              }
+            }
+          ENDLOOP02:
+          break;
+
+          case OP_VSPACE:
+          for (i = min; i < max; i++)
+            {
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              break;
+              }
+            switch(*eptr)
+              {
+              default: goto ENDLOOP03;
+              VSPACE_BYTE_CASES:
+#if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
+              VSPACE_MULTIBYTE_CASES:
+#endif
+              eptr++; break;
+              }
+            }
+          ENDLOOP03:
+          break;
+
+          case OP_NOT_DIGIT:
+          for (i = min; i < max; i++)
+            {
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              break;
+              }
+            if (MAX_255(*eptr) && (md->ctypes[*eptr] & ctype_digit) != 0) break;
+            eptr++;
+            }
+          break;
+
+          case OP_DIGIT:
+          for (i = min; i < max; i++)
+            {
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              break;
+              }
+            if (!MAX_255(*eptr) || (md->ctypes[*eptr] & ctype_digit) == 0) break;
+            eptr++;
+            }
+          break;
+
+          case OP_NOT_WHITESPACE:
+          for (i = min; i < max; i++)
+            {
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              break;
+              }
+            if (MAX_255(*eptr) && (md->ctypes[*eptr] & ctype_space) != 0) break;
+            eptr++;
+            }
+          break;
+
+          case OP_WHITESPACE:
+          for (i = min; i < max; i++)
+            {
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              break;
+              }
+            if (!MAX_255(*eptr) || (md->ctypes[*eptr] & ctype_space) == 0) break;
+            eptr++;
+            }
+          break;
+
+          case OP_NOT_WORDCHAR:
+          for (i = min; i < max; i++)
+            {
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              break;
+              }
+            if (MAX_255(*eptr) && (md->ctypes[*eptr] & ctype_word) != 0) break;
+            eptr++;
+            }
+          break;
+
+          case OP_WORDCHAR:
+          for (i = min; i < max; i++)
+            {
+            if (eptr >= md->end_subject)
+              {
+              SCHECK_PARTIAL();
+              break;
+              }
+            if (!MAX_255(*eptr) || (md->ctypes[*eptr] & ctype_word) == 0) break;
+            eptr++;
+            }
+          break;
+
+          default:
+          RRETURN(PCRE_ERROR_INTERNAL);
+          }
+
+        if (possessive) continue;    /* No backtracking */
+        for (;;)
+          {
+          if (eptr == pp) goto TAIL_RECURSE;
+          RMATCH(eptr, ecode, offset_top, md, eptrb, RM47);
+          if (rrc != MATCH_NOMATCH) RRETURN(rrc);
+          eptr--;
+          if (ctype == OP_ANYNL && eptr > pp  && *eptr == CHAR_LF &&
+              eptr[-1] == CHAR_CR) eptr--;
+          }
+        }
+
+      /* Get here if we can't make it match with any permitted repetitions */
+
+      RRETURN(MATCH_NOMATCH);
+      }
+    /* Control never gets here */
+
+    /* There's been some horrible disaster. Arrival here can only mean there is
+    something seriously wrong in the code above or the OP_xxx definitions. */
+
+    default:
+    DPRINTF(("Unknown opcode %d\n", *ecode));
+    RRETURN(PCRE_ERROR_UNKNOWN_OPCODE);
+    }
+
+  /* Do not stick any code in here without much thought; it is assumed
+  that "continue" in the code above comes out to here to repeat the main
+  loop. */
+
+  }             /* End of main loop */
+/* Control never reaches here */
+
+
+/* When compiling to use the heap rather than the stack for recursive calls to
+match(), the RRETURN() macro jumps here. The number that is saved in
+frame->Xwhere indicates which label we actually want to return to. */
+
+#ifdef NO_RECURSE
+#define LBL(val) case val: goto L_RM##val;
+HEAP_RETURN:
+switch (frame->Xwhere)
+  {
+  LBL( 1) LBL( 2) LBL( 3) LBL( 4) LBL( 5) LBL( 6) LBL( 7) LBL( 8)
+  LBL( 9) LBL(10) LBL(11) LBL(12) LBL(13) LBL(14) LBL(15) LBL(17)
+  LBL(19) LBL(24) LBL(25) LBL(26) LBL(27) LBL(29) LBL(31) LBL(33)
+  LBL(35) LBL(43) LBL(47) LBL(48) LBL(49) LBL(50) LBL(51) LBL(52)
+  LBL(53) LBL(54) LBL(55) LBL(56) LBL(57) LBL(58) LBL(63) LBL(64)
+  LBL(65) LBL(66)
+#if defined SUPPORT_UTF || !defined COMPILE_PCRE8
+  LBL(21)
+#endif
+#ifdef SUPPORT_UTF
+  LBL(16) LBL(18) LBL(20)
+  LBL(22) LBL(23) LBL(28) LBL(30)
+  LBL(32) LBL(34) LBL(42) LBL(46)
+#ifdef SUPPORT_UCP
+  LBL(36) LBL(37) LBL(38) LBL(39) LBL(40) LBL(41) LBL(44) LBL(45)
+  LBL(59) LBL(60) LBL(61) LBL(62) LBL(67) LBL(68)
+#endif  /* SUPPORT_UCP */
+#endif  /* SUPPORT_UTF */
+  default:
+  DPRINTF(("jump error in pcre match: label %d non-existent\n", frame->Xwhere));
+  return PCRE_ERROR_INTERNAL;
+  }
+#undef LBL
+#endif  /* NO_RECURSE */
+}
+
+
+/***************************************************************************
+****************************************************************************
+                   RECURSION IN THE match() FUNCTION
+
+Undefine all the macros that were defined above to handle this. */
+
+#ifdef NO_RECURSE
+#undef eptr
+#undef ecode
+#undef mstart
+#undef offset_top
+#undef eptrb
+#undef flags
+
+#undef callpat
+#undef charptr
+#undef data
+#undef next
+#undef pp
+#undef prev
+#undef saved_eptr
+
+#undef new_recursive
+
+#undef cur_is_word
+#undef condition
+#undef prev_is_word
+
+#undef ctype
+#undef length
+#undef max
+#undef min
+#undef number
+#undef offset
+#undef op
+#undef save_capture_last
+#undef save_offset1
+#undef save_offset2
+#undef save_offset3
+#undef stacksave
+
+#undef newptrb
+
+#endif
+
+/* These two are defined as macros in both cases */
+
+#undef fc
+#undef fi
+
+/***************************************************************************
+***************************************************************************/
+
+
+#ifdef NO_RECURSE
+/*************************************************
+*          Release allocated heap frames         *
+*************************************************/
+
+/* This function releases all the allocated frames. The base frame is on the
+machine stack, and so must not be freed.
+
+Argument: the address of the base frame
+Returns:  nothing
+*/
+
+static void
+release_match_heapframes (heapframe *frame_base)
+{
+heapframe *nextframe = frame_base->Xnextframe;
+while (nextframe != NULL)
+  {
+  heapframe *oldframe = nextframe;
+  nextframe = nextframe->Xnextframe;
+  (PUBL(stack_free))(oldframe);
+  }
+}
+#endif
+
+
+/*************************************************
+*         Execute a Regular Expression           *
+*************************************************/
+
+/* This function applies a compiled re to a subject string and picks out
+portions of the string if it matches. Two elements in the vector are set for
+each substring: the offsets to the start and end of the substring.
+
+Arguments:
+  argument_re     points to the compiled expression
+  extra_data      points to extra data or is NULL
+  subject         points to the subject string
+  length          length of subject string (may contain binary zeros)
+  start_offset    where to start in the subject string
+  options         option bits
+  offsets         points to a vector of ints to be filled in with offsets
+  offsetcount     the number of elements in the vector
+
+Returns:          > 0 => success; value is the number of elements filled in
+                  = 0 => success, but offsets is not big enough
+                   -1 => failed to match
+                 < -1 => some kind of unexpected problem
+*/
+
+#if defined COMPILE_PCRE8
+PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
+pcre_exec(const pcre *argument_re, const pcre_extra *extra_data,
+  PCRE_SPTR subject, int length, int start_offset, int options, int *offsets,
+  int offsetcount)
+#elif defined COMPILE_PCRE16
+PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
+pcre16_exec(const pcre16 *argument_re, const pcre16_extra *extra_data,
+  PCRE_SPTR16 subject, int length, int start_offset, int options, int *offsets,
+  int offsetcount)
+#elif defined COMPILE_PCRE32
+PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
+pcre32_exec(const pcre32 *argument_re, const pcre32_extra *extra_data,
+  PCRE_SPTR32 subject, int length, int start_offset, int options, int *offsets,
+  int offsetcount)
+#endif
+{
+int rc, ocount, arg_offset_max;
+int newline;
+BOOL using_temporary_offsets = FALSE;
+BOOL anchored;
+BOOL startline;
+BOOL firstline;
+BOOL utf;
+BOOL has_first_char = FALSE;
+BOOL has_req_char = FALSE;
+pcre_uchar first_char = 0;
+pcre_uchar first_char2 = 0;
+pcre_uchar req_char = 0;
+pcre_uchar req_char2 = 0;
+match_data match_block;
+match_data *md = &match_block;
+const pcre_uint8 *tables;
+const pcre_uint8 *start_bits = NULL;
+PCRE_PUCHAR start_match = (PCRE_PUCHAR)subject + start_offset;
+PCRE_PUCHAR end_subject;
+PCRE_PUCHAR start_partial = NULL;
+PCRE_PUCHAR match_partial;
+PCRE_PUCHAR req_char_ptr = start_match - 1;
+
+const pcre_study_data *study;
+const REAL_PCRE *re = (const REAL_PCRE *)argument_re;
+
+#ifdef NO_RECURSE
+heapframe frame_zero;
+frame_zero.Xprevframe = NULL;            /* Marks the top level */
+frame_zero.Xnextframe = NULL;            /* None are allocated yet */
+md->match_frames_base = &frame_zero;
+#endif
+
+/* Check for the special magic call that measures the size of the stack used
+per recursive call of match(). Without the funny casting for sizeof, a Windows
+compiler gave this error: "unary minus operator applied to unsigned type,
+result still unsigned". Hopefully the cast fixes that. */
+
+if (re == NULL && extra_data == NULL && subject == NULL && length == -999 &&
+    start_offset == -999)
+#ifdef NO_RECURSE
+  return -((int)sizeof(heapframe));
+#else
+  return match(NULL, NULL, NULL, 0, NULL, NULL, 0);
+#endif
+
+/* Plausibility checks */
+
+if ((options & ~PUBLIC_EXEC_OPTIONS) != 0) return PCRE_ERROR_BADOPTION;
+if (re == NULL || subject == NULL || (offsets == NULL && offsetcount > 0))
+  return PCRE_ERROR_NULL;
+if (offsetcount < 0) return PCRE_ERROR_BADCOUNT;
+if (length < 0) return PCRE_ERROR_BADLENGTH;
+if (start_offset < 0 || start_offset > length) return PCRE_ERROR_BADOFFSET;
+
+/* Check that the first field in the block is the magic number. If it is not,
+return with PCRE_ERROR_BADMAGIC. However, if the magic number is equal to
+REVERSED_MAGIC_NUMBER we return with PCRE_ERROR_BADENDIANNESS, which
+means that the pattern is likely compiled with different endianness. */
+
+if (re->magic_number != MAGIC_NUMBER)
+  return re->magic_number == REVERSED_MAGIC_NUMBER?
+    PCRE_ERROR_BADENDIANNESS:PCRE_ERROR_BADMAGIC;
+if ((re->flags & PCRE_MODE) == 0) return PCRE_ERROR_BADMODE;
+
+/* These two settings are used in the code for checking a UTF-8 string that
+follows immediately afterwards. Other values in the md block are used only
+during "normal" pcre_exec() processing, not when the JIT support is in use,
+so they are set up later. */
+
+/* PCRE_UTF16 has the same value as PCRE_UTF8. */
+utf = md->utf = (re->options & PCRE_UTF8) != 0;
+md->partial = ((options & PCRE_PARTIAL_HARD) != 0)? 2 :
+              ((options & PCRE_PARTIAL_SOFT) != 0)? 1 : 0;
+
+/* Check a UTF-8 string if required. Pass back the character offset and error
+code for an invalid string if a results vector is available. */
+
+#ifdef SUPPORT_UTF
+if (utf && (options & PCRE_NO_UTF8_CHECK) == 0)
+  {
+  int erroroffset;
+  int errorcode = PRIV(valid_utf)((PCRE_PUCHAR)subject, length, &erroroffset);
+  if (errorcode != 0)
+    {
+    if (offsetcount >= 2)
+      {
+      offsets[0] = erroroffset;
+      offsets[1] = errorcode;
+      }
+#if defined COMPILE_PCRE8
+    return (errorcode <= PCRE_UTF8_ERR5 && md->partial > 1)?
+      PCRE_ERROR_SHORTUTF8 : PCRE_ERROR_BADUTF8;
+#elif defined COMPILE_PCRE16
+    return (errorcode <= PCRE_UTF16_ERR1 && md->partial > 1)?
+      PCRE_ERROR_SHORTUTF16 : PCRE_ERROR_BADUTF16;
+#elif defined COMPILE_PCRE32
+    return PCRE_ERROR_BADUTF32;
+#endif
+    }
+#if defined COMPILE_PCRE8 || defined COMPILE_PCRE16
+  /* Check that a start_offset points to the start of a UTF character. */
+  if (start_offset > 0 && start_offset < length &&
+      NOT_FIRSTCHAR(((PCRE_PUCHAR)subject)[start_offset]))
+    return PCRE_ERROR_BADUTF8_OFFSET;
+#endif
+  }
+#endif
+
+/* If the pattern was successfully studied with JIT support, run the JIT
+executable instead of the rest of this function. Most options must be set at
+compile time for the JIT code to be usable. Fallback to the normal code path if
+an unsupported flag is set. */
+
+#ifdef SUPPORT_JIT
+if (extra_data != NULL
+    && (extra_data->flags & (PCRE_EXTRA_EXECUTABLE_JIT |
+                             PCRE_EXTRA_TABLES)) == PCRE_EXTRA_EXECUTABLE_JIT
+    && extra_data->executable_jit != NULL
+    && (options & ~PUBLIC_JIT_EXEC_OPTIONS) == 0)
+  {
+  rc = PRIV(jit_exec)(extra_data, (const pcre_uchar *)subject, length,
+       start_offset, options, offsets, offsetcount);
+
+  /* PCRE_ERROR_NULL means that the selected normal or partial matching
+  mode is not compiled. In this case we simply fallback to interpreter. */
+
+  if (rc != PCRE_ERROR_JIT_BADOPTION) return rc;
+  }
+#endif
+
+/* Carry on with non-JIT matching. This information is for finding all the
+numbers associated with a given name, for condition testing. */
+
+md->name_table = (pcre_uchar *)re + re->name_table_offset;
+md->name_count = re->name_count;
+md->name_entry_size = re->name_entry_size;
+
+/* Fish out the optional data from the extra_data structure, first setting
+the default values. */
+
+study = NULL;
+md->match_limit = MATCH_LIMIT;
+md->match_limit_recursion = MATCH_LIMIT_RECURSION;
+md->callout_data = NULL;
+
+/* The table pointer is always in native byte order. */
+
+tables = re->tables;
+
+/* The two limit values override the defaults, whatever their value. */
+
+if (extra_data != NULL)
+  {
+  register unsigned int flags = extra_data->flags;
+  if ((flags & PCRE_EXTRA_STUDY_DATA) != 0)
+    study = (const pcre_study_data *)extra_data->study_data;
+  if ((flags & PCRE_EXTRA_MATCH_LIMIT) != 0)
+    md->match_limit = extra_data->match_limit;
+  if ((flags & PCRE_EXTRA_MATCH_LIMIT_RECURSION) != 0)
+    md->match_limit_recursion = extra_data->match_limit_recursion;
+  if ((flags & PCRE_EXTRA_CALLOUT_DATA) != 0)
+    md->callout_data = extra_data->callout_data;
+  if ((flags & PCRE_EXTRA_TABLES) != 0) tables = extra_data->tables;
+  }
+
+/* Limits in the regex override only if they are smaller. */
+
+if ((re->flags & PCRE_MLSET) != 0 && re->limit_match < md->match_limit)
+  md->match_limit = re->limit_match;
+
+if ((re->flags & PCRE_RLSET) != 0 &&
+    re->limit_recursion < md->match_limit_recursion)
+  md->match_limit_recursion = re->limit_recursion;
+
+/* If the exec call supplied NULL for tables, use the inbuilt ones. This
+is a feature that makes it possible to save compiled regex and re-use them
+in other programs later. */
+
+if (tables == NULL) tables = PRIV(default_tables);
+
+/* Set up other data */
+
+anchored = ((re->options | options) & PCRE_ANCHORED) != 0;
+startline = (re->flags & PCRE_STARTLINE) != 0;
+firstline = (re->options & PCRE_FIRSTLINE) != 0;
+
+/* The code starts after the real_pcre block and the capture name table. */
+
+md->start_code = (const pcre_uchar *)re + re->name_table_offset +
+  re->name_count * re->name_entry_size;
+
+md->start_subject = (PCRE_PUCHAR)subject;
+md->start_offset = start_offset;
+md->end_subject = md->start_subject + length;
+end_subject = md->end_subject;
+
+md->endonly = (re->options & PCRE_DOLLAR_ENDONLY) != 0;
+md->use_ucp = (re->options & PCRE_UCP) != 0;
+md->jscript_compat = (re->options & PCRE_JAVASCRIPT_COMPAT) != 0;
+md->ignore_skip_arg = 0;
+
+/* Some options are unpacked into BOOL variables in the hope that testing
+them will be faster than individual option bits. */
+
+md->notbol = (options & PCRE_NOTBOL) != 0;
+md->noteol = (options & PCRE_NOTEOL) != 0;
+md->notempty = (options & PCRE_NOTEMPTY) != 0;
+md->notempty_atstart = (options & PCRE_NOTEMPTY_ATSTART) != 0;
+
+md->hitend = FALSE;
+md->mark = md->nomatch_mark = NULL;     /* In case never set */
+
+md->recursive = NULL;                   /* No recursion at top level */
+md->hasthen = (re->flags & PCRE_HASTHEN) != 0;
+
+md->lcc = tables + lcc_offset;
+md->fcc = tables + fcc_offset;
+md->ctypes = tables + ctypes_offset;
+
+/* Handle different \R options. */
+
+switch (options & (PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE))
+  {
+  case 0:
+  if ((re->options & (PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE)) != 0)
+    md->bsr_anycrlf = (re->options & PCRE_BSR_ANYCRLF) != 0;
+  else
+#ifdef BSR_ANYCRLF
+  md->bsr_anycrlf = TRUE;
+#else
+  md->bsr_anycrlf = FALSE;
+#endif
+  break;
+
+  case PCRE_BSR_ANYCRLF:
+  md->bsr_anycrlf = TRUE;
+  break;
+
+  case PCRE_BSR_UNICODE:
+  md->bsr_anycrlf = FALSE;
+  break;
+
+  default: return PCRE_ERROR_BADNEWLINE;
+  }
+
+/* Handle different types of newline. The three bits give eight cases. If
+nothing is set at run time, whatever was used at compile time applies. */
+
+switch ((((options & PCRE_NEWLINE_BITS) == 0)? re->options :
+        (pcre_uint32)options) & PCRE_NEWLINE_BITS)
+  {
+  case 0: newline = NEWLINE; break;   /* Compile-time default */
+  case PCRE_NEWLINE_CR: newline = CHAR_CR; break;
+  case PCRE_NEWLINE_LF: newline = CHAR_NL; break;
+  case PCRE_NEWLINE_CR+
+       PCRE_NEWLINE_LF: newline = (CHAR_CR << 8) | CHAR_NL; break;
+  case PCRE_NEWLINE_ANY: newline = -1; break;
+  case PCRE_NEWLINE_ANYCRLF: newline = -2; break;
+  default: return PCRE_ERROR_BADNEWLINE;
+  }
+
+if (newline == -2)
+  {
+  md->nltype = NLTYPE_ANYCRLF;
+  }
+else if (newline < 0)
+  {
+  md->nltype = NLTYPE_ANY;
+  }
+else
+  {
+  md->nltype = NLTYPE_FIXED;
+  if (newline > 255)
+    {
+    md->nllen = 2;
+    md->nl[0] = (newline >> 8) & 255;
+    md->nl[1] = newline & 255;
+    }
+  else
+    {
+    md->nllen = 1;
+    md->nl[0] = newline;
+    }
+  }
+
+/* Partial matching was originally supported only for a restricted set of
+regexes; from release 8.00 there are no restrictions, but the bits are still
+defined (though never set). So there's no harm in leaving this code. */
+
+if (md->partial && (re->flags & PCRE_NOPARTIAL) != 0)
+  return PCRE_ERROR_BADPARTIAL;
+
+/* If the expression has got more back references than the offsets supplied can
+hold, we get a temporary chunk of working store to use during the matching.
+Otherwise, we can use the vector supplied, rounding down its size to a multiple
+of 3. */
+
+ocount = offsetcount - (offsetcount % 3);
+arg_offset_max = (2*ocount)/3;
+
+if (re->top_backref > 0 && re->top_backref >= ocount/3)
+  {
+  ocount = re->top_backref * 3 + 3;
+  md->offset_vector = (int *)(PUBL(malloc))(ocount * sizeof(int));
+  if (md->offset_vector == NULL) return PCRE_ERROR_NOMEMORY;
+  using_temporary_offsets = TRUE;
+  DPRINTF(("Got memory to hold back references\n"));
+  }
+else md->offset_vector = offsets;
+md->offset_end = ocount;
+md->offset_max = (2*ocount)/3;
+md->capture_last = 0;
+
+/* Reset the working variable associated with each extraction. These should
+never be used unless previously set, but they get saved and restored, and so we
+initialize them to avoid reading uninitialized locations. Also, unset the
+offsets for the matched string. This is really just for tidiness with callouts,
+in case they inspect these fields. */
+
+if (md->offset_vector != NULL)
+  {
+  register int *iptr = md->offset_vector + ocount;
+  register int *iend = iptr - re->top_bracket;
+  if (iend < md->offset_vector + 2) iend = md->offset_vector + 2;
+  while (--iptr >= iend) *iptr = -1;
+  md->offset_vector[0] = md->offset_vector[1] = -1;
+  }
+
+/* Set up the first character to match, if available. The first_char value is
+never set for an anchored regular expression, but the anchoring may be forced
+at run time, so we have to test for anchoring. The first char may be unset for
+an unanchored pattern, of course. If there's no first char and the pattern was
+studied, there may be a bitmap of possible first characters. */
+
+if (!anchored)
+  {
+  if ((re->flags & PCRE_FIRSTSET) != 0)
+    {
+    has_first_char = TRUE;
+    first_char = first_char2 = (pcre_uchar)(re->first_char);
+    if ((re->flags & PCRE_FCH_CASELESS) != 0)
+      {
+      first_char2 = TABLE_GET(first_char, md->fcc, first_char);
+#if defined SUPPORT_UCP && !(defined COMPILE_PCRE8)
+      if (utf && first_char > 127)
+        first_char2 = UCD_OTHERCASE(first_char);
+#endif
+      }
+    }
+  else
+    if (!startline && study != NULL &&
+      (study->flags & PCRE_STUDY_MAPPED) != 0)
+        start_bits = study->start_bits;
+  }
+
+/* For anchored or unanchored matches, there may be a "last known required
+character" set. */
+
+if ((re->flags & PCRE_REQCHSET) != 0)
+  {
+  has_req_char = TRUE;
+  req_char = req_char2 = (pcre_uchar)(re->req_char);
+  if ((re->flags & PCRE_RCH_CASELESS) != 0)
+    {
+    req_char2 = TABLE_GET(req_char, md->fcc, req_char);
+#if defined SUPPORT_UCP && !(defined COMPILE_PCRE8)
+    if (utf && req_char > 127)
+      req_char2 = UCD_OTHERCASE(req_char);
+#endif
+    }
+  }
+
+
+/* ==========================================================================*/
+
+/* Loop for handling unanchored repeated matching attempts; for anchored regexs
+the loop runs just once. */
+
+for(;;)
+  {
+  PCRE_PUCHAR save_end_subject = end_subject;
+  PCRE_PUCHAR new_start_match;
+
+  /* If firstline is TRUE, the start of the match is constrained to the first
+  line of a multiline string. That is, the match must be before or at the first
+  newline. Implement this by temporarily adjusting end_subject so that we stop
+  scanning at a newline. If the match fails at the newline, later code breaks
+  this loop. */
+
+  if (firstline)
+    {
+    PCRE_PUCHAR t = start_match;
+#ifdef SUPPORT_UTF
+    if (utf)
+      {
+      while (t < md->end_subject && !IS_NEWLINE(t))
+        {
+        t++;
+        ACROSSCHAR(t < end_subject, *t, t++);
+        }
+      }
+    else
+#endif
+    while (t < md->end_subject && !IS_NEWLINE(t)) t++;
+    end_subject = t;
+    }
+
+  /* There are some optimizations that avoid running the match if a known
+  starting point is not found, or if a known later character is not present.
+  However, there is an option that disables these, for testing and for ensuring
+  that all callouts do actually occur. The option can be set in the regex by
+  (*NO_START_OPT) or passed in match-time options. */
+
+  if (((options | re->options) & PCRE_NO_START_OPTIMIZE) == 0)
+    {
+    /* Advance to a unique first char if there is one. */
+
+    if (has_first_char)
+      {
+      pcre_uchar smc;
+
+      if (first_char != first_char2)
+        while (start_match < end_subject &&
+          (smc = RAWUCHARTEST(start_match)) != first_char && smc != first_char2)
+          start_match++;
+      else
+        while (start_match < end_subject && RAWUCHARTEST(start_match) != first_char)
+          start_match++;
+      }
+
+    /* Or to just after a linebreak for a multiline match */
+
+    else if (startline)
+      {
+      if (start_match > md->start_subject + start_offset)
+        {
+#ifdef SUPPORT_UTF
+        if (utf)
+          {
+          while (start_match < end_subject && !WAS_NEWLINE(start_match))
+            {
+            start_match++;
+            ACROSSCHAR(start_match < end_subject, *start_match,
+              start_match++);
+            }
+          }
+        else
+#endif
+        while (start_match < end_subject && !WAS_NEWLINE(start_match))
+          start_match++;
+
+        /* If we have just passed a CR and the newline option is ANY or ANYCRLF,
+        and we are now at a LF, advance the match position by one more character.
+        */
+
+        if (start_match[-1] == CHAR_CR &&
+             (md->nltype == NLTYPE_ANY || md->nltype == NLTYPE_ANYCRLF) &&
+             start_match < end_subject &&
+             RAWUCHARTEST(start_match) == CHAR_NL)
+          start_match++;
+        }
+      }
+
+    /* Or to a non-unique first byte after study */
+
+    else if (start_bits != NULL)
+      {
+      while (start_match < end_subject)
+        {
+        register pcre_uint32 c = RAWUCHARTEST(start_match);
+#ifndef COMPILE_PCRE8
+        if (c > 255) c = 255;
+#endif
+        if ((start_bits[c/8] & (1 << (c&7))) == 0)
+          {
+          start_match++;
+#if defined SUPPORT_UTF && defined COMPILE_PCRE8
+          /* In non 8-bit mode, the iteration will stop for
+          characters > 255 at the beginning or not stop at all. */
+          if (utf)
+            ACROSSCHAR(start_match < end_subject, *start_match,
+              start_match++);
+#endif
+          }
+        else break;
+        }
+      }
+    }   /* Starting optimizations */
+
+  /* Restore fudged end_subject */
+
+  end_subject = save_end_subject;
+
+  /* The following two optimizations are disabled for partial matching or if
+  disabling is explicitly requested. */
+
+  if (((options | re->options) & PCRE_NO_START_OPTIMIZE) == 0 && !md->partial)
+    {
+    /* If the pattern was studied, a minimum subject length may be set. This is
+    a lower bound; no actual string of that length may actually match the
+    pattern. Although the value is, strictly, in characters, we treat it as
+    bytes to avoid spending too much time in this optimization. */
+
+    if (study != NULL && (study->flags & PCRE_STUDY_MINLEN) != 0 &&
+        (pcre_uint32)(end_subject - start_match) < study->minlength)
+      {
+      rc = MATCH_NOMATCH;
+      break;
+      }
+
+    /* If req_char is set, we know that that character must appear in the
+    subject for the match to succeed. If the first character is set, req_char
+    must be later in the subject; otherwise the test starts at the match point.
+    This optimization can save a huge amount of backtracking in patterns with
+    nested unlimited repeats that aren't going to match. Writing separate code
+    for cased/caseless versions makes it go faster, as does using an
+    autoincrement and backing off on a match.
+
+    HOWEVER: when the subject string is very, very long, searching to its end
+    can take a long time, and give bad performance on quite ordinary patterns.
+    This showed up when somebody was matching something like /^\d+C/ on a
+    32-megabyte string... so we don't do this when the string is sufficiently
+    long. */
+
+    if (has_req_char && end_subject - start_match < REQ_BYTE_MAX)
+      {
+      register PCRE_PUCHAR p = start_match + (has_first_char? 1:0);
+
+      /* We don't need to repeat the search if we haven't yet reached the
+      place we found it at last time. */
+
+      if (p > req_char_ptr)
+        {
+        if (req_char != req_char2)
+          {
+          while (p < end_subject)
+            {
+            register pcre_uint32 pp = RAWUCHARINCTEST(p);
+            if (pp == req_char || pp == req_char2) { p--; break; }
+            }
+          }
+        else
+          {
+          while (p < end_subject)
+            {
+            if (RAWUCHARINCTEST(p) == req_char) { p--; break; }
+            }
+          }
+
+        /* If we can't find the required character, break the matching loop,
+        forcing a match failure. */
+
+        if (p >= end_subject)
+          {
+          rc = MATCH_NOMATCH;
+          break;
+          }
+
+        /* If we have found the required character, save the point where we
+        found it, so that we don't search again next time round the loop if
+        the start hasn't passed this character yet. */
+
+        req_char_ptr = p;
+        }
+      }
+    }
+
+#ifdef PCRE_DEBUG  /* Sigh. Some compilers never learn. */
+  printf(">>>> Match against: ");
+  pchars(start_match, end_subject - start_match, TRUE, md);
+  printf("\n");
+#endif
+
+  /* OK, we can now run the match. If "hitend" is set afterwards, remember the
+  first starting point for which a partial match was found. */
+
+  md->start_match_ptr = start_match;
+  md->start_used_ptr = start_match;
+  md->match_call_count = 0;
+  md->match_function_type = 0;
+  md->end_offset_top = 0;
+  md->skip_arg_count = 0;
+  rc = match(start_match, md->start_code, start_match, 2, md, NULL, 0);
+  if (md->hitend && start_partial == NULL)
+    {
+    start_partial = md->start_used_ptr;
+    match_partial = start_match;
+    }
+
+  switch(rc)
+    {
+    /* If MATCH_SKIP_ARG reaches this level it means that a MARK that matched
+    the SKIP's arg was not found. In this circumstance, Perl ignores the SKIP
+    entirely. The only way we can do that is to re-do the match at the same
+    point, with a flag to force SKIP with an argument to be ignored. Just
+    treating this case as NOMATCH does not work because it does not check other
+    alternatives in patterns such as A(*SKIP:A)B|AC when the subject is AC. */
+
+    case MATCH_SKIP_ARG:
+    new_start_match = start_match;
+    md->ignore_skip_arg = md->skip_arg_count;
+    break;
+
+    /* SKIP passes back the next starting point explicitly, but if it is no
+    greater than the match we have just done, treat it as NOMATCH. */
+
+    case MATCH_SKIP:
+    if (md->start_match_ptr > start_match)
+      {
+      new_start_match = md->start_match_ptr;
+      break;
+      }
+    /* Fall through */
+
+    /* NOMATCH and PRUNE advance by one character. THEN at this level acts
+    exactly like PRUNE. Unset ignore SKIP-with-argument. */
+
+    case MATCH_NOMATCH:
+    case MATCH_PRUNE:
+    case MATCH_THEN:
+    md->ignore_skip_arg = 0;
+    new_start_match = start_match + 1;
+#ifdef SUPPORT_UTF
+    if (utf)
+      ACROSSCHAR(new_start_match < end_subject, *new_start_match,
+        new_start_match++);
+#endif
+    break;
+
+    /* COMMIT disables the bumpalong, but otherwise behaves as NOMATCH. */
+
+    case MATCH_COMMIT:
+    rc = MATCH_NOMATCH;
+    goto ENDLOOP;
+
+    /* Any other return is either a match, or some kind of error. */
+
+    default:
+    goto ENDLOOP;
+    }
+
+  /* Control reaches here for the various types of "no match at this point"
+  result. Reset the code to MATCH_NOMATCH for subsequent checking. */
+
+  rc = MATCH_NOMATCH;
+
+  /* If PCRE_FIRSTLINE is set, the match must happen before or at the first
+  newline in the subject (though it may continue over the newline). Therefore,
+  if we have just failed to match, starting at a newline, do not continue. */
+
+  if (firstline && IS_NEWLINE(start_match)) break;
+
+  /* Advance to new matching position */
+
+  start_match = new_start_match;
+
+  /* Break the loop if the pattern is anchored or if we have passed the end of
+  the subject. */
+
+  if (anchored || start_match > end_subject) break;
+
+  /* If we have just passed a CR and we are now at a LF, and the pattern does
+  not contain any explicit matches for \r or \n, and the newline option is CRLF
+  or ANY or ANYCRLF, advance the match position by one more character. In
+  normal matching start_match will aways be greater than the first position at
+  this stage, but a failed *SKIP can cause a return at the same point, which is
+  why the first test exists. */
+
+  if (start_match > (PCRE_PUCHAR)subject + start_offset &&
+      start_match[-1] == CHAR_CR &&
+      start_match < end_subject &&
+      *start_match == CHAR_NL &&
+      (re->flags & PCRE_HASCRORLF) == 0 &&
+        (md->nltype == NLTYPE_ANY ||
+         md->nltype == NLTYPE_ANYCRLF ||
+         md->nllen == 2))
+    start_match++;
+
+  md->mark = NULL;   /* Reset for start of next match attempt */
+  }                  /* End of for(;;) "bumpalong" loop */
+
+/* ==========================================================================*/
+
+/* We reach here when rc is not MATCH_NOMATCH, or if one of the stopping
+conditions is true:
+
+(1) The pattern is anchored or the match was failed by (*COMMIT);
+
+(2) We are past the end of the subject;
+
+(3) PCRE_FIRSTLINE is set and we have failed to match at a newline, because
+    this option requests that a match occur at or before the first newline in
+    the subject.
+
+When we have a match and the offset vector is big enough to deal with any
+backreferences, captured substring offsets will already be set up. In the case
+where we had to get some local store to hold offsets for backreference
+processing, copy those that we can. In this case there need not be overflow if
+certain parts of the pattern were not used, even though there are more
+capturing parentheses than vector slots. */
+
+ENDLOOP:
+
+if (rc == MATCH_MATCH || rc == MATCH_ACCEPT)
+  {
+  if (using_temporary_offsets)
+    {
+    if (arg_offset_max >= 4)
+      {
+      memcpy(offsets + 2, md->offset_vector + 2,
+        (arg_offset_max - 2) * sizeof(int));
+      DPRINTF(("Copied offsets from temporary memory\n"));
+      }
+    if (md->end_offset_top > arg_offset_max) md->capture_last |= OVFLBIT;
+    DPRINTF(("Freeing temporary memory\n"));
+    (PUBL(free))(md->offset_vector);
+    }
+
+  /* Set the return code to the number of captured strings, or 0 if there were
+  too many to fit into the vector. */
+
+  rc = ((md->capture_last & OVFLBIT) != 0 &&
+         md->end_offset_top >= arg_offset_max)?
+    0 : md->end_offset_top/2;
+
+  /* If there is space in the offset vector, set any unused pairs at the end of
+  the pattern to -1 for backwards compatibility. It is documented that this
+  happens. In earlier versions, the whole set of potential capturing offsets
+  was set to -1 each time round the loop, but this is handled differently now.
+  "Gaps" are set to -1 dynamically instead (this fixes a bug). Thus, it is only
+  those at the end that need unsetting here. We can't just unset them all at
+  the start of the whole thing because they may get set in one branch that is
+  not the final matching branch. */
+
+  if (md->end_offset_top/2 <= re->top_bracket && offsets != NULL)
+    {
+    register int *iptr, *iend;
+    int resetcount = 2 + re->top_bracket * 2;
+    if (resetcount > offsetcount) resetcount = offsetcount;
+    iptr = offsets + md->end_offset_top;
+    iend = offsets + resetcount;
+    while (iptr < iend) *iptr++ = -1;
+    }
+
+  /* If there is space, set up the whole thing as substring 0. The value of
+  md->start_match_ptr might be modified if \K was encountered on the success
+  matching path. */
+
+  if (offsetcount < 2) rc = 0; else
+    {
+    offsets[0] = (int)(md->start_match_ptr - md->start_subject);
+    offsets[1] = (int)(md->end_match_ptr - md->start_subject);
+    }
+
+  /* Return MARK data if requested */
+
+  if (extra_data != NULL && (extra_data->flags & PCRE_EXTRA_MARK) != 0)
+    *(extra_data->mark) = (pcre_uchar *)md->mark;
+  DPRINTF((">>>> returning %d\n", rc));
+#ifdef NO_RECURSE
+  release_match_heapframes(&frame_zero);
+#endif
+  return rc;
+  }
+
+/* Control gets here if there has been an error, or if the overall match
+attempt has failed at all permitted starting positions. */
+
+if (using_temporary_offsets)
+  {
+  DPRINTF(("Freeing temporary memory\n"));
+  (PUBL(free))(md->offset_vector);
+  }
+
+/* For anything other than nomatch or partial match, just return the code. */
+
+if (rc != MATCH_NOMATCH && rc != PCRE_ERROR_PARTIAL)
+  {
+  DPRINTF((">>>> error: returning %d\n", rc));
+#ifdef NO_RECURSE
+  release_match_heapframes(&frame_zero);
+#endif
+  return rc;
+  }
+
+/* Handle partial matches - disable any mark data */
+
+if (start_partial != NULL)
+  {
+  DPRINTF((">>>> returning PCRE_ERROR_PARTIAL\n"));
+  md->mark = NULL;
+  if (offsetcount > 1)
+    {
+    offsets[0] = (int)(start_partial - (PCRE_PUCHAR)subject);
+    offsets[1] = (int)(end_subject - (PCRE_PUCHAR)subject);
+    if (offsetcount > 2)
+      offsets[2] = (int)(match_partial - (PCRE_PUCHAR)subject);
+    }
+  rc = PCRE_ERROR_PARTIAL;
+  }
+
+/* This is the classic nomatch case */
+
+else
+  {
+  DPRINTF((">>>> returning PCRE_ERROR_NOMATCH\n"));
+  rc = PCRE_ERROR_NOMATCH;
+  }
+
+/* Return the MARK data if it has been requested. */
+
+if (extra_data != NULL && (extra_data->flags & PCRE_EXTRA_MARK) != 0)
+  *(extra_data->mark) = (pcre_uchar *)md->nomatch_mark;
+#ifdef NO_RECURSE
+  release_match_heapframes(&frame_zero);
+#endif
+return rc;
+}
+
+/* End of pcre_exec.c */
diff --git a/pcre/pcre_fullinfo.c b/pcre/pcre_fullinfo.c
new file mode 100644
index 00000000000..c4eb5c0e1ad
--- /dev/null
+++ b/pcre/pcre_fullinfo.c
@@ -0,0 +1,241 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2013 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+
+/* This module contains the external function pcre_fullinfo(), which returns
+information about a compiled pattern. */
+
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include "pcre_internal.h"
+
+
+/*************************************************
+*        Return info about compiled pattern      *
+*************************************************/
+
+/* This is a newer "info" function which has an extensible interface so
+that additional items can be added compatibly.
+
+Arguments:
+  argument_re      points to compiled code
+  extra_data       points extra data, or NULL
+  what             what information is required
+  where            where to put the information
+
+Returns:           0 if data returned, negative on error
+*/
+
+#if defined COMPILE_PCRE8
+PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
+pcre_fullinfo(const pcre *argument_re, const pcre_extra *extra_data,
+  int what, void *where)
+#elif defined COMPILE_PCRE16
+PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
+pcre16_fullinfo(const pcre16 *argument_re, const pcre16_extra *extra_data,
+  int what, void *where)
+#elif defined COMPILE_PCRE32
+PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
+pcre32_fullinfo(const pcre32 *argument_re, const pcre32_extra *extra_data,
+  int what, void *where)
+#endif
+{
+const REAL_PCRE *re = (const REAL_PCRE *)argument_re;
+const pcre_study_data *study = NULL;
+
+if (re == NULL || where == NULL) return PCRE_ERROR_NULL;
+
+if (extra_data != NULL && (extra_data->flags & PCRE_EXTRA_STUDY_DATA) != 0)
+  study = (const pcre_study_data *)extra_data->study_data;
+
+/* Check that the first field in the block is the magic number. If it is not,
+return with PCRE_ERROR_BADMAGIC. However, if the magic number is equal to
+REVERSED_MAGIC_NUMBER we return with PCRE_ERROR_BADENDIANNESS, which
+means that the pattern is likely compiled with different endianness. */
+
+if (re->magic_number != MAGIC_NUMBER)
+  return re->magic_number == REVERSED_MAGIC_NUMBER?
+    PCRE_ERROR_BADENDIANNESS:PCRE_ERROR_BADMAGIC;
+
+/* Check that this pattern was compiled in the correct bit mode */
+
+if ((re->flags & PCRE_MODE) == 0) return PCRE_ERROR_BADMODE;
+
+switch (what)
+  {
+  case PCRE_INFO_OPTIONS:
+  *((unsigned long int *)where) = re->options & PUBLIC_COMPILE_OPTIONS;
+  break;
+
+  case PCRE_INFO_SIZE:
+  *((size_t *)where) = re->size;
+  break;
+
+  case PCRE_INFO_STUDYSIZE:
+  *((size_t *)where) = (study == NULL)? 0 : study->size;
+  break;
+
+  case PCRE_INFO_JITSIZE:
+#ifdef SUPPORT_JIT
+  *((size_t *)where) =
+      (extra_data != NULL &&
+      (extra_data->flags & PCRE_EXTRA_EXECUTABLE_JIT) != 0 &&
+      extra_data->executable_jit != NULL)?
+    PRIV(jit_get_size)(extra_data->executable_jit) : 0;
+#else
+  *((size_t *)where) = 0;
+#endif
+  break;
+
+  case PCRE_INFO_CAPTURECOUNT:
+  *((int *)where) = re->top_bracket;
+  break;
+
+  case PCRE_INFO_BACKREFMAX:
+  *((int *)where) = re->top_backref;
+  break;
+
+  case PCRE_INFO_FIRSTBYTE:
+  *((int *)where) =
+    ((re->flags & PCRE_FIRSTSET) != 0)? (int)re->first_char :
+    ((re->flags & PCRE_STARTLINE) != 0)? -1 : -2;
+  break;
+
+  case PCRE_INFO_FIRSTCHARACTER:
+    *((pcre_uint32 *)where) =
+      (re->flags & PCRE_FIRSTSET) != 0 ? re->first_char : 0;
+    break;
+
+  case PCRE_INFO_FIRSTCHARACTERFLAGS:
+    *((int *)where) =
+      ((re->flags & PCRE_FIRSTSET) != 0) ? 1 :
+      ((re->flags & PCRE_STARTLINE) != 0) ? 2 : 0;
+    break;
+
+  /* Make sure we pass back the pointer to the bit vector in the external
+  block, not the internal copy (with flipped integer fields). */
+
+  case PCRE_INFO_FIRSTTABLE:
+  *((const pcre_uint8 **)where) =
+    (study != NULL && (study->flags & PCRE_STUDY_MAPPED) != 0)?
+      ((const pcre_study_data *)extra_data->study_data)->start_bits : NULL;
+  break;
+
+  case PCRE_INFO_MINLENGTH:
+  *((int *)where) =
+    (study != NULL && (study->flags & PCRE_STUDY_MINLEN) != 0)?
+      (int)(study->minlength) : -1;
+  break;
+
+  case PCRE_INFO_JIT:
+  *((int *)where) = extra_data != NULL &&
+                    (extra_data->flags & PCRE_EXTRA_EXECUTABLE_JIT) != 0 &&
+                    extra_data->executable_jit != NULL;
+  break;
+
+  case PCRE_INFO_LASTLITERAL:
+  *((int *)where) =
+    ((re->flags & PCRE_REQCHSET) != 0)? (int)re->req_char : -1;
+  break;
+
+  case PCRE_INFO_REQUIREDCHAR:
+    *((pcre_uint32 *)where) =
+      ((re->flags & PCRE_REQCHSET) != 0) ? re->req_char : 0;
+    break;
+
+  case PCRE_INFO_REQUIREDCHARFLAGS:
+    *((int *)where) =
+      ((re->flags & PCRE_REQCHSET) != 0);
+    break;
+
+  case PCRE_INFO_NAMEENTRYSIZE:
+  *((int *)where) = re->name_entry_size;
+  break;
+
+  case PCRE_INFO_NAMECOUNT:
+  *((int *)where) = re->name_count;
+  break;
+
+  case PCRE_INFO_NAMETABLE:
+  *((const pcre_uchar **)where) = (const pcre_uchar *)re + re->name_table_offset;
+  break;
+
+  case PCRE_INFO_DEFAULT_TABLES:
+  *((const pcre_uint8 **)where) = (const pcre_uint8 *)(PRIV(default_tables));
+  break;
+
+  /* From release 8.00 this will always return TRUE because NOPARTIAL is
+  no longer ever set (the restrictions have been removed). */
+
+  case PCRE_INFO_OKPARTIAL:
+  *((int *)where) = (re->flags & PCRE_NOPARTIAL) == 0;
+  break;
+
+  case PCRE_INFO_JCHANGED:
+  *((int *)where) = (re->flags & PCRE_JCHANGED) != 0;
+  break;
+
+  case PCRE_INFO_HASCRORLF:
+  *((int *)where) = (re->flags & PCRE_HASCRORLF) != 0;
+  break;
+
+  case PCRE_INFO_MAXLOOKBEHIND:
+  *((int *)where) = re->max_lookbehind;
+  break;
+
+  case PCRE_INFO_MATCHLIMIT:
+  if ((re->flags & PCRE_MLSET) == 0) return PCRE_ERROR_UNSET;
+  *((pcre_uint32 *)where) = re->limit_match;
+  break;
+
+  case PCRE_INFO_RECURSIONLIMIT:
+  if ((re->flags & PCRE_RLSET) == 0) return PCRE_ERROR_UNSET;
+  *((pcre_uint32 *)where) = re->limit_recursion;
+  break;
+
+  default: return PCRE_ERROR_BADOPTION;
+  }
+
+return 0;
+}
+
+/* End of pcre_fullinfo.c */
diff --git a/pcre/pcre_get.c b/pcre/pcre_get.c
new file mode 100644
index 00000000000..8094b34bbfb
--- /dev/null
+++ b/pcre/pcre_get.c
@@ -0,0 +1,662 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+
+/* This module contains some convenience functions for extracting substrings
+from the subject string after a regex match has succeeded. The original idea
+for these functions came from Scott Wimer. */
+
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include "pcre_internal.h"
+
+
+/*************************************************
+*           Find number for named string         *
+*************************************************/
+
+/* This function is used by the get_first_set() function below, as well
+as being generally available. It assumes that names are unique.
+
+Arguments:
+  code        the compiled regex
+  stringname  the name whose number is required
+
+Returns:      the number of the named parentheses, or a negative number
+                (PCRE_ERROR_NOSUBSTRING) if not found
+*/
+
+#if defined COMPILE_PCRE8
+PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
+pcre_get_stringnumber(const pcre *code, const char *stringname)
+#elif defined COMPILE_PCRE16
+PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
+pcre16_get_stringnumber(const pcre16 *code, PCRE_SPTR16 stringname)
+#elif defined COMPILE_PCRE32
+PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
+pcre32_get_stringnumber(const pcre32 *code, PCRE_SPTR32 stringname)
+#endif
+{
+int rc;
+int entrysize;
+int top, bot;
+pcre_uchar *nametable;
+
+#ifdef COMPILE_PCRE8
+if ((rc = pcre_fullinfo(code, NULL, PCRE_INFO_NAMECOUNT, &top)) != 0)
+  return rc;
+if (top <= 0) return PCRE_ERROR_NOSUBSTRING;
+
+if ((rc = pcre_fullinfo(code, NULL, PCRE_INFO_NAMEENTRYSIZE, &entrysize)) != 0)
+  return rc;
+if ((rc = pcre_fullinfo(code, NULL, PCRE_INFO_NAMETABLE, &nametable)) != 0)
+  return rc;
+#endif
+#ifdef COMPILE_PCRE16
+if ((rc = pcre16_fullinfo(code, NULL, PCRE_INFO_NAMECOUNT, &top)) != 0)
+  return rc;
+if (top <= 0) return PCRE_ERROR_NOSUBSTRING;
+
+if ((rc = pcre16_fullinfo(code, NULL, PCRE_INFO_NAMEENTRYSIZE, &entrysize)) != 0)
+  return rc;
+if ((rc = pcre16_fullinfo(code, NULL, PCRE_INFO_NAMETABLE, &nametable)) != 0)
+  return rc;
+#endif
+#ifdef COMPILE_PCRE32
+if ((rc = pcre32_fullinfo(code, NULL, PCRE_INFO_NAMECOUNT, &top)) != 0)
+  return rc;
+if (top <= 0) return PCRE_ERROR_NOSUBSTRING;
+
+if ((rc = pcre32_fullinfo(code, NULL, PCRE_INFO_NAMEENTRYSIZE, &entrysize)) != 0)
+  return rc;
+if ((rc = pcre32_fullinfo(code, NULL, PCRE_INFO_NAMETABLE, &nametable)) != 0)
+  return rc;
+#endif
+
+bot = 0;
+while (top > bot)
+  {
+  int mid = (top + bot) / 2;
+  pcre_uchar *entry = nametable + entrysize*mid;
+  int c = STRCMP_UC_UC((pcre_uchar *)stringname,
+    (pcre_uchar *)(entry + IMM2_SIZE));
+  if (c == 0) return GET2(entry, 0);
+  if (c > 0) bot = mid + 1; else top = mid;
+  }
+
+return PCRE_ERROR_NOSUBSTRING;
+}
+
+
+
+/*************************************************
+*     Find (multiple) entries for named string   *
+*************************************************/
+
+/* This is used by the get_first_set() function below, as well as being
+generally available. It is used when duplicated names are permitted.
+
+Arguments:
+  code        the compiled regex
+  stringname  the name whose entries required
+  firstptr    where to put the pointer to the first entry
+  lastptr     where to put the pointer to the last entry
+
+Returns:      the length of each entry, or a negative number
+                (PCRE_ERROR_NOSUBSTRING) if not found
+*/
+
+#if defined COMPILE_PCRE8
+PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
+pcre_get_stringtable_entries(const pcre *code, const char *stringname,
+  char **firstptr, char **lastptr)
+#elif defined COMPILE_PCRE16
+PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
+pcre16_get_stringtable_entries(const pcre16 *code, PCRE_SPTR16 stringname,
+  PCRE_UCHAR16 **firstptr, PCRE_UCHAR16 **lastptr)
+#elif defined COMPILE_PCRE32
+PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
+pcre32_get_stringtable_entries(const pcre32 *code, PCRE_SPTR32 stringname,
+  PCRE_UCHAR32 **firstptr, PCRE_UCHAR32 **lastptr)
+#endif
+{
+int rc;
+int entrysize;
+int top, bot;
+pcre_uchar *nametable, *lastentry;
+
+#ifdef COMPILE_PCRE8
+if ((rc = pcre_fullinfo(code, NULL, PCRE_INFO_NAMECOUNT, &top)) != 0)
+  return rc;
+if (top <= 0) return PCRE_ERROR_NOSUBSTRING;
+
+if ((rc = pcre_fullinfo(code, NULL, PCRE_INFO_NAMEENTRYSIZE, &entrysize)) != 0)
+  return rc;
+if ((rc = pcre_fullinfo(code, NULL, PCRE_INFO_NAMETABLE, &nametable)) != 0)
+  return rc;
+#endif
+#ifdef COMPILE_PCRE16
+if ((rc = pcre16_fullinfo(code, NULL, PCRE_INFO_NAMECOUNT, &top)) != 0)
+  return rc;
+if (top <= 0) return PCRE_ERROR_NOSUBSTRING;
+
+if ((rc = pcre16_fullinfo(code, NULL, PCRE_INFO_NAMEENTRYSIZE, &entrysize)) != 0)
+  return rc;
+if ((rc = pcre16_fullinfo(code, NULL, PCRE_INFO_NAMETABLE, &nametable)) != 0)
+  return rc;
+#endif
+#ifdef COMPILE_PCRE32
+if ((rc = pcre32_fullinfo(code, NULL, PCRE_INFO_NAMECOUNT, &top)) != 0)
+  return rc;
+if (top <= 0) return PCRE_ERROR_NOSUBSTRING;
+
+if ((rc = pcre32_fullinfo(code, NULL, PCRE_INFO_NAMEENTRYSIZE, &entrysize)) != 0)
+  return rc;
+if ((rc = pcre32_fullinfo(code, NULL, PCRE_INFO_NAMETABLE, &nametable)) != 0)
+  return rc;
+#endif
+
+lastentry = nametable + entrysize * (top - 1);
+bot = 0;
+while (top > bot)
+  {
+  int mid = (top + bot) / 2;
+  pcre_uchar *entry = nametable + entrysize*mid;
+  int c = STRCMP_UC_UC((pcre_uchar *)stringname,
+    (pcre_uchar *)(entry + IMM2_SIZE));
+  if (c == 0)
+    {
+    pcre_uchar *first = entry;
+    pcre_uchar *last = entry;
+    while (first > nametable)
+      {
+      if (STRCMP_UC_UC((pcre_uchar *)stringname,
+        (pcre_uchar *)(first - entrysize + IMM2_SIZE)) != 0) break;
+      first -= entrysize;
+      }
+    while (last < lastentry)
+      {
+      if (STRCMP_UC_UC((pcre_uchar *)stringname,
+        (pcre_uchar *)(last + entrysize + IMM2_SIZE)) != 0) break;
+      last += entrysize;
+      }
+#if defined COMPILE_PCRE8
+    *firstptr = (char *)first;
+    *lastptr = (char *)last;
+#elif defined COMPILE_PCRE16
+    *firstptr = (PCRE_UCHAR16 *)first;
+    *lastptr = (PCRE_UCHAR16 *)last;
+#elif defined COMPILE_PCRE32
+    *firstptr = (PCRE_UCHAR32 *)first;
+    *lastptr = (PCRE_UCHAR32 *)last;
+#endif
+    return entrysize;
+    }
+  if (c > 0) bot = mid + 1; else top = mid;
+  }
+
+return PCRE_ERROR_NOSUBSTRING;
+}
+
+
+
+/*************************************************
+*    Find first set of multiple named strings    *
+*************************************************/
+
+/* This function allows for duplicate names in the table of named substrings.
+It returns the number of the first one that was set in a pattern match.
+
+Arguments:
+  code         the compiled regex
+  stringname   the name of the capturing substring
+  ovector      the vector of matched substrings
+
+Returns:       the number of the first that is set,
+               or the number of the last one if none are set,
+               or a negative number on error
+*/
+
+#if defined COMPILE_PCRE8
+static int
+get_first_set(const pcre *code, const char *stringname, int *ovector)
+#elif defined COMPILE_PCRE16
+static int
+get_first_set(const pcre16 *code, PCRE_SPTR16 stringname, int *ovector)
+#elif defined COMPILE_PCRE32
+static int
+get_first_set(const pcre32 *code, PCRE_SPTR32 stringname, int *ovector)
+#endif
+{
+const REAL_PCRE *re = (const REAL_PCRE *)code;
+int entrysize;
+pcre_uchar *entry;
+#if defined COMPILE_PCRE8
+char *first, *last;
+#elif defined COMPILE_PCRE16
+PCRE_UCHAR16 *first, *last;
+#elif defined COMPILE_PCRE32
+PCRE_UCHAR32 *first, *last;
+#endif
+
+#if defined COMPILE_PCRE8
+if ((re->options & PCRE_DUPNAMES) == 0 && (re->flags & PCRE_JCHANGED) == 0)
+  return pcre_get_stringnumber(code, stringname);
+entrysize = pcre_get_stringtable_entries(code, stringname, &first, &last);
+#elif defined COMPILE_PCRE16
+if ((re->options & PCRE_DUPNAMES) == 0 && (re->flags & PCRE_JCHANGED) == 0)
+  return pcre16_get_stringnumber(code, stringname);
+entrysize = pcre16_get_stringtable_entries(code, stringname, &first, &last);
+#elif defined COMPILE_PCRE32
+if ((re->options & PCRE_DUPNAMES) == 0 && (re->flags & PCRE_JCHANGED) == 0)
+  return pcre32_get_stringnumber(code, stringname);
+entrysize = pcre32_get_stringtable_entries(code, stringname, &first, &last);
+#endif
+if (entrysize <= 0) return entrysize;
+for (entry = (pcre_uchar *)first; entry <= (pcre_uchar *)last; entry += entrysize)
+  {
+  int n = GET2(entry, 0);
+  if (ovector[n*2] >= 0) return n;
+  }
+return GET2(entry, 0);
+}
+
+
+
+
+/*************************************************
+*      Copy captured string to given buffer      *
+*************************************************/
+
+/* This function copies a single captured substring into a given buffer.
+Note that we use memcpy() rather than strncpy() in case there are binary zeros
+in the string.
+
+Arguments:
+  subject        the subject string that was matched
+  ovector        pointer to the offsets table
+  stringcount    the number of substrings that were captured
+                   (i.e. the yield of the pcre_exec call, unless
+                   that was zero, in which case it should be 1/3
+                   of the offset table size)
+  stringnumber   the number of the required substring
+  buffer         where to put the substring
+  size           the size of the buffer
+
+Returns:         if successful:
+                   the length of the copied string, not including the zero
+                   that is put on the end; can be zero
+                 if not successful:
+                   PCRE_ERROR_NOMEMORY (-6) buffer too small
+                   PCRE_ERROR_NOSUBSTRING (-7) no such captured substring
+*/
+
+#if defined COMPILE_PCRE8
+PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
+pcre_copy_substring(const char *subject, int *ovector, int stringcount,
+  int stringnumber, char *buffer, int size)
+#elif defined COMPILE_PCRE16
+PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
+pcre16_copy_substring(PCRE_SPTR16 subject, int *ovector, int stringcount,
+  int stringnumber, PCRE_UCHAR16 *buffer, int size)
+#elif defined COMPILE_PCRE32
+PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
+pcre32_copy_substring(PCRE_SPTR32 subject, int *ovector, int stringcount,
+  int stringnumber, PCRE_UCHAR32 *buffer, int size)
+#endif
+{
+int yield;
+if (stringnumber < 0 || stringnumber >= stringcount)
+  return PCRE_ERROR_NOSUBSTRING;
+stringnumber *= 2;
+yield = ovector[stringnumber+1] - ovector[stringnumber];
+if (size < yield + 1) return PCRE_ERROR_NOMEMORY;
+memcpy(buffer, subject + ovector[stringnumber], IN_UCHARS(yield));
+buffer[yield] = 0;
+return yield;
+}
+
+
+
+/*************************************************
+*   Copy named captured string to given buffer   *
+*************************************************/
+
+/* This function copies a single captured substring into a given buffer,
+identifying it by name. If the regex permits duplicate names, the first
+substring that is set is chosen.
+
+Arguments:
+  code           the compiled regex
+  subject        the subject string that was matched
+  ovector        pointer to the offsets table
+  stringcount    the number of substrings that were captured
+                   (i.e. the yield of the pcre_exec call, unless
+                   that was zero, in which case it should be 1/3
+                   of the offset table size)
+  stringname     the name of the required substring
+  buffer         where to put the substring
+  size           the size of the buffer
+
+Returns:         if successful:
+                   the length of the copied string, not including the zero
+                   that is put on the end; can be zero
+                 if not successful:
+                   PCRE_ERROR_NOMEMORY (-6) buffer too small
+                   PCRE_ERROR_NOSUBSTRING (-7) no such captured substring
+*/
+
+#if defined COMPILE_PCRE8
+PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
+pcre_copy_named_substring(const pcre *code, const char *subject,
+  int *ovector, int stringcount, const char *stringname,
+  char *buffer, int size)
+#elif defined COMPILE_PCRE16
+PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
+pcre16_copy_named_substring(const pcre16 *code, PCRE_SPTR16 subject,
+  int *ovector, int stringcount, PCRE_SPTR16 stringname,
+  PCRE_UCHAR16 *buffer, int size)
+#elif defined COMPILE_PCRE32
+PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
+pcre32_copy_named_substring(const pcre32 *code, PCRE_SPTR32 subject,
+  int *ovector, int stringcount, PCRE_SPTR32 stringname,
+  PCRE_UCHAR32 *buffer, int size)
+#endif
+{
+int n = get_first_set(code, stringname, ovector);
+if (n <= 0) return n;
+#if defined COMPILE_PCRE8
+return pcre_copy_substring(subject, ovector, stringcount, n, buffer, size);
+#elif defined COMPILE_PCRE16
+return pcre16_copy_substring(subject, ovector, stringcount, n, buffer, size);
+#elif defined COMPILE_PCRE32
+return pcre32_copy_substring(subject, ovector, stringcount, n, buffer, size);
+#endif
+}
+
+
+
+/*************************************************
+*      Copy all captured strings to new store    *
+*************************************************/
+
+/* This function gets one chunk of store and builds a list of pointers and all
+of the captured substrings in it. A NULL pointer is put on the end of the list.
+
+Arguments:
+  subject        the subject string that was matched
+  ovector        pointer to the offsets table
+  stringcount    the number of substrings that were captured
+                   (i.e. the yield of the pcre_exec call, unless
+                   that was zero, in which case it should be 1/3
+                   of the offset table size)
+  listptr        set to point to the list of pointers
+
+Returns:         if successful: 0
+                 if not successful:
+                   PCRE_ERROR_NOMEMORY (-6) failed to get store
+*/
+
+#if defined COMPILE_PCRE8
+PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
+pcre_get_substring_list(const char *subject, int *ovector, int stringcount,
+  const char ***listptr)
+#elif defined COMPILE_PCRE16
+PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
+pcre16_get_substring_list(PCRE_SPTR16 subject, int *ovector, int stringcount,
+  PCRE_SPTR16 **listptr)
+#elif defined COMPILE_PCRE32
+PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
+pcre32_get_substring_list(PCRE_SPTR32 subject, int *ovector, int stringcount,
+  PCRE_SPTR32 **listptr)
+#endif
+{
+int i;
+int size = sizeof(pcre_uchar *);
+int double_count = stringcount * 2;
+pcre_uchar **stringlist;
+pcre_uchar *p;
+
+for (i = 0; i < double_count; i += 2)
+  size += sizeof(pcre_uchar *) + IN_UCHARS(ovector[i+1] - ovector[i] + 1);
+
+stringlist = (pcre_uchar **)(PUBL(malloc))(size);
+if (stringlist == NULL) return PCRE_ERROR_NOMEMORY;
+
+#if defined COMPILE_PCRE8
+*listptr = (const char **)stringlist;
+#elif defined COMPILE_PCRE16
+*listptr = (PCRE_SPTR16 *)stringlist;
+#elif defined COMPILE_PCRE32
+*listptr = (PCRE_SPTR32 *)stringlist;
+#endif
+p = (pcre_uchar *)(stringlist + stringcount + 1);
+
+for (i = 0; i < double_count; i += 2)
+  {
+  int len = ovector[i+1] - ovector[i];
+  memcpy(p, subject + ovector[i], IN_UCHARS(len));
+  *stringlist++ = p;
+  p += len;
+  *p++ = 0;
+  }
+
+*stringlist = NULL;
+return 0;
+}
+
+
+
+/*************************************************
+*   Free store obtained by get_substring_list    *
+*************************************************/
+
+/* This function exists for the benefit of people calling PCRE from non-C
+programs that can call its functions, but not free() or (PUBL(free))()
+directly.
+
+Argument:   the result of a previous pcre_get_substring_list()
+Returns:    nothing
+*/
+
+#if defined COMPILE_PCRE8
+PCRE_EXP_DEFN void PCRE_CALL_CONVENTION
+pcre_free_substring_list(const char **pointer)
+#elif defined COMPILE_PCRE16
+PCRE_EXP_DEFN void PCRE_CALL_CONVENTION
+pcre16_free_substring_list(PCRE_SPTR16 *pointer)
+#elif defined COMPILE_PCRE32
+PCRE_EXP_DEFN void PCRE_CALL_CONVENTION
+pcre32_free_substring_list(PCRE_SPTR32 *pointer)
+#endif
+{
+(PUBL(free))((void *)pointer);
+}
+
+
+
+/*************************************************
+*      Copy captured string to new store         *
+*************************************************/
+
+/* This function copies a single captured substring into a piece of new
+store
+
+Arguments:
+  subject        the subject string that was matched
+  ovector        pointer to the offsets table
+  stringcount    the number of substrings that were captured
+                   (i.e. the yield of the pcre_exec call, unless
+                   that was zero, in which case it should be 1/3
+                   of the offset table size)
+  stringnumber   the number of the required substring
+  stringptr      where to put a pointer to the substring
+
+Returns:         if successful:
+                   the length of the string, not including the zero that
+                   is put on the end; can be zero
+                 if not successful:
+                   PCRE_ERROR_NOMEMORY (-6) failed to get store
+                   PCRE_ERROR_NOSUBSTRING (-7) substring not present
+*/
+
+#if defined COMPILE_PCRE8
+PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
+pcre_get_substring(const char *subject, int *ovector, int stringcount,
+  int stringnumber, const char **stringptr)
+#elif defined COMPILE_PCRE16
+PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
+pcre16_get_substring(PCRE_SPTR16 subject, int *ovector, int stringcount,
+  int stringnumber, PCRE_SPTR16 *stringptr)
+#elif defined COMPILE_PCRE32
+PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
+pcre32_get_substring(PCRE_SPTR32 subject, int *ovector, int stringcount,
+  int stringnumber, PCRE_SPTR32 *stringptr)
+#endif
+{
+int yield;
+pcre_uchar *substring;
+if (stringnumber < 0 || stringnumber >= stringcount)
+  return PCRE_ERROR_NOSUBSTRING;
+stringnumber *= 2;
+yield = ovector[stringnumber+1] - ovector[stringnumber];
+substring = (pcre_uchar *)(PUBL(malloc))(IN_UCHARS(yield + 1));
+if (substring == NULL) return PCRE_ERROR_NOMEMORY;
+memcpy(substring, subject + ovector[stringnumber], IN_UCHARS(yield));
+substring[yield] = 0;
+#if defined COMPILE_PCRE8
+*stringptr = (const char *)substring;
+#elif defined COMPILE_PCRE16
+*stringptr = (PCRE_SPTR16)substring;
+#elif defined COMPILE_PCRE32
+*stringptr = (PCRE_SPTR32)substring;
+#endif
+return yield;
+}
+
+
+
+/*************************************************
+*   Copy named captured string to new store      *
+*************************************************/
+
+/* This function copies a single captured substring, identified by name, into
+new store. If the regex permits duplicate names, the first substring that is
+set is chosen.
+
+Arguments:
+  code           the compiled regex
+  subject        the subject string that was matched
+  ovector        pointer to the offsets table
+  stringcount    the number of substrings that were captured
+                   (i.e. the yield of the pcre_exec call, unless
+                   that was zero, in which case it should be 1/3
+                   of the offset table size)
+  stringname     the name of the required substring
+  stringptr      where to put the pointer
+
+Returns:         if successful:
+                   the length of the copied string, not including the zero
+                   that is put on the end; can be zero
+                 if not successful:
+                   PCRE_ERROR_NOMEMORY (-6) couldn't get memory
+                   PCRE_ERROR_NOSUBSTRING (-7) no such captured substring
+*/
+
+#if defined COMPILE_PCRE8
+PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
+pcre_get_named_substring(const pcre *code, const char *subject,
+  int *ovector, int stringcount, const char *stringname,
+  const char **stringptr)
+#elif defined COMPILE_PCRE16
+PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
+pcre16_get_named_substring(const pcre16 *code, PCRE_SPTR16 subject,
+  int *ovector, int stringcount, PCRE_SPTR16 stringname,
+  PCRE_SPTR16 *stringptr)
+#elif defined COMPILE_PCRE32
+PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
+pcre32_get_named_substring(const pcre32 *code, PCRE_SPTR32 subject,
+  int *ovector, int stringcount, PCRE_SPTR32 stringname,
+  PCRE_SPTR32 *stringptr)
+#endif
+{
+int n = get_first_set(code, stringname, ovector);
+if (n <= 0) return n;
+#if defined COMPILE_PCRE8
+return pcre_get_substring(subject, ovector, stringcount, n, stringptr);
+#elif defined COMPILE_PCRE16
+return pcre16_get_substring(subject, ovector, stringcount, n, stringptr);
+#elif defined COMPILE_PCRE32
+return pcre32_get_substring(subject, ovector, stringcount, n, stringptr);
+#endif
+}
+
+
+
+
+/*************************************************
+*       Free store obtained by get_substring     *
+*************************************************/
+
+/* This function exists for the benefit of people calling PCRE from non-C
+programs that can call its functions, but not free() or (PUBL(free))()
+directly.
+
+Argument:   the result of a previous pcre_get_substring()
+Returns:    nothing
+*/
+
+#if defined COMPILE_PCRE8
+PCRE_EXP_DEFN void PCRE_CALL_CONVENTION
+pcre_free_substring(const char *pointer)
+#elif defined COMPILE_PCRE16
+PCRE_EXP_DEFN void PCRE_CALL_CONVENTION
+pcre16_free_substring(PCRE_SPTR16 pointer)
+#elif defined COMPILE_PCRE32
+PCRE_EXP_DEFN void PCRE_CALL_CONVENTION
+pcre32_free_substring(PCRE_SPTR32 pointer)
+#endif
+{
+(PUBL(free))((void *)pointer);
+}
+
+/* End of pcre_get.c */
diff --git a/pcre/pcre_globals.c b/pcre/pcre_globals.c
new file mode 100644
index 00000000000..36e6ddb3a89
--- /dev/null
+++ b/pcre/pcre_globals.c
@@ -0,0 +1,84 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+
+/* This module contains global variables that are exported by the PCRE library.
+PCRE is thread-clean and doesn't use any global variables in the normal sense.
+However, it calls memory allocation and freeing functions via the four
+indirections below, and it can optionally do callouts, using the fifth
+indirection. These values can be changed by the caller, but are shared between
+all threads.
+
+For MS Visual Studio and Symbian OS, there are problems in initializing these
+variables to non-local functions. In these cases, therefore, an indirection via
+a local function is used.
+
+Also, when compiling for Virtual Pascal, things are done differently, and
+global variables are not used. */
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include "pcre_internal.h"
+
+#if defined _MSC_VER || defined  __SYMBIAN32__
+static void* LocalPcreMalloc(size_t aSize)
+  {
+  return malloc(aSize);
+  }
+static void LocalPcreFree(void* aPtr)
+  {
+  free(aPtr);
+  }
+PCRE_EXP_DATA_DEFN void *(*PUBL(malloc))(size_t) = LocalPcreMalloc;
+PCRE_EXP_DATA_DEFN void  (*PUBL(free))(void *) = LocalPcreFree;
+PCRE_EXP_DATA_DEFN void *(*PUBL(stack_malloc))(size_t) = LocalPcreMalloc;
+PCRE_EXP_DATA_DEFN void  (*PUBL(stack_free))(void *) = LocalPcreFree;
+PCRE_EXP_DATA_DEFN int   (*PUBL(callout))(PUBL(callout_block) *) = NULL;
+
+#elif !defined VPCOMPAT
+PCRE_EXP_DATA_DEFN void *(*PUBL(malloc))(size_t) = malloc;
+PCRE_EXP_DATA_DEFN void  (*PUBL(free))(void *) = free;
+PCRE_EXP_DATA_DEFN void *(*PUBL(stack_malloc))(size_t) = malloc;
+PCRE_EXP_DATA_DEFN void  (*PUBL(stack_free))(void *) = free;
+PCRE_EXP_DATA_DEFN int   (*PUBL(callout))(PUBL(callout_block) *) = NULL;
+#endif
+
+/* End of pcre_globals.c */
diff --git a/pcre/pcre_internal.h b/pcre/pcre_internal.h
new file mode 100644
index 00000000000..307069ca9d6
--- /dev/null
+++ b/pcre/pcre_internal.h
@@ -0,0 +1,2771 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2013 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+/* This header contains definitions that are shared between the different
+modules, but which are not relevant to the exported API. This includes some
+functions whose names all begin with "_pcre_", "_pcre16_" or "_pcre32_"
+depending on the PRIV macro. */
+
+#ifndef PCRE_INTERNAL_H
+#define PCRE_INTERNAL_H
+
+/* Define PCRE_DEBUG to get debugging output on stdout. */
+
+#if 0
+#define PCRE_DEBUG
+#endif
+
+/* PCRE is compiled as an 8 bit library if it is not requested otherwise. */
+
+#if !defined COMPILE_PCRE16 && !defined COMPILE_PCRE32
+#define COMPILE_PCRE8
+#endif
+
+/* If SUPPORT_UCP is defined, SUPPORT_UTF must also be defined. The
+"configure" script ensures this, but not everybody uses "configure". */
+
+#if defined SUPPORT_UCP && !(defined SUPPORT_UTF)
+#define SUPPORT_UTF 1
+#endif
+
+/* We define SUPPORT_UTF if SUPPORT_UTF8 is enabled for compatibility
+reasons with existing code. */
+
+#if defined SUPPORT_UTF8 && !(defined SUPPORT_UTF)
+#define SUPPORT_UTF 1
+#endif
+
+/* Fixme: SUPPORT_UTF8 should be eventually disappear from the code.
+Until then we define it if SUPPORT_UTF is defined. */
+
+#if defined SUPPORT_UTF && !(defined SUPPORT_UTF8)
+#define SUPPORT_UTF8 1
+#endif
+
+/* We do not support both EBCDIC and UTF-8/16/32 at the same time. The "configure"
+script prevents both being selected, but not everybody uses "configure". */
+
+#if defined EBCDIC && defined SUPPORT_UTF
+#error The use of both EBCDIC and SUPPORT_UTF is not supported.
+#endif
+
+/* Use a macro for debugging printing, 'cause that eliminates the use of #ifdef
+inline, and there are *still* stupid compilers about that don't like indented
+pre-processor statements, or at least there were when I first wrote this. After
+all, it had only been about 10 years then...
+
+It turns out that the Mac Debugging.h header also defines the macro DPRINTF, so
+be absolutely sure we get our version. */
+
+#undef DPRINTF
+#ifdef PCRE_DEBUG
+#define DPRINTF(p) printf p
+#else
+#define DPRINTF(p) /* Nothing */
+#endif
+
+
+/* Standard C headers plus the external interface definition. The only time
+setjmp and stdarg are used is when NO_RECURSE is set. */
+
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+
+/* Valgrind (memcheck) support */
+
+#ifdef SUPPORT_VALGRIND
+#include 
+#endif
+
+/* When compiling a DLL for Windows, the exported symbols have to be declared
+using some MS magic. I found some useful information on this web page:
+http://msdn2.microsoft.com/en-us/library/y4h7bcy6(VS.80).aspx. According to the
+information there, using __declspec(dllexport) without "extern" we have a
+definition; with "extern" we have a declaration. The settings here override the
+setting in pcre.h (which is included below); it defines only PCRE_EXP_DECL,
+which is all that is needed for applications (they just import the symbols). We
+use:
+
+  PCRE_EXP_DECL       for declarations
+  PCRE_EXP_DEFN       for definitions of exported functions
+  PCRE_EXP_DATA_DEFN  for definitions of exported variables
+
+The reason for the two DEFN macros is that in non-Windows environments, one
+does not want to have "extern" before variable definitions because it leads to
+compiler warnings. So we distinguish between functions and variables. In
+Windows, the two should always be the same.
+
+The reason for wrapping this in #ifndef PCRE_EXP_DECL is so that pcretest,
+which is an application, but needs to import this file in order to "peek" at
+internals, can #include pcre.h first to get an application's-eye view.
+
+In principle, people compiling for non-Windows, non-Unix-like (i.e. uncommon,
+special-purpose environments) might want to stick other stuff in front of
+exported symbols. That's why, in the non-Windows case, we set PCRE_EXP_DEFN and
+PCRE_EXP_DATA_DEFN only if they are not already set. */
+
+#ifndef PCRE_EXP_DECL
+#  ifdef _WIN32
+#    ifndef PCRE_STATIC
+#      define PCRE_EXP_DECL       extern __declspec(dllexport)
+#      define PCRE_EXP_DEFN       __declspec(dllexport)
+#      define PCRE_EXP_DATA_DEFN  __declspec(dllexport)
+#    else
+#      define PCRE_EXP_DECL       extern
+#      define PCRE_EXP_DEFN
+#      define PCRE_EXP_DATA_DEFN
+#    endif
+#  else
+#    ifdef __cplusplus
+#      define PCRE_EXP_DECL       extern "C"
+#    else
+#      define PCRE_EXP_DECL       extern
+#    endif
+#    ifndef PCRE_EXP_DEFN
+#      define PCRE_EXP_DEFN       PCRE_EXP_DECL
+#    endif
+#    ifndef PCRE_EXP_DATA_DEFN
+#      define PCRE_EXP_DATA_DEFN
+#    endif
+#  endif
+#endif
+
+/* When compiling with the MSVC compiler, it is sometimes necessary to include
+a "calling convention" before exported function names. (This is secondhand
+information; I know nothing about MSVC myself). For example, something like
+
+  void __cdecl function(....)
+
+might be needed. In order so make this easy, all the exported functions have
+PCRE_CALL_CONVENTION just before their names. It is rarely needed; if not
+set, we ensure here that it has no effect. */
+
+#ifndef PCRE_CALL_CONVENTION
+#define PCRE_CALL_CONVENTION
+#endif
+
+/* We need to have types that specify unsigned 8, 16 and 32-bit integers. We
+cannot determine these outside the compilation (e.g. by running a program as
+part of "configure") because PCRE is often cross-compiled for use on other
+systems. Instead we make use of the maximum sizes that are available at
+preprocessor time in standard C environments. */
+
+typedef unsigned char pcre_uint8;
+
+#if USHRT_MAX == 65535
+typedef unsigned short pcre_uint16;
+typedef short pcre_int16;
+#define PCRE_UINT16_MAX USHRT_MAX
+#define PCRE_INT16_MAX SHRT_MAX
+#elif UINT_MAX == 65535
+typedef unsigned int pcre_uint16;
+typedef int pcre_int16;
+#define PCRE_UINT16_MAX UINT_MAX
+#define PCRE_INT16_MAX INT_MAX
+#else
+#error Cannot determine a type for 16-bit integers
+#endif
+
+#if UINT_MAX == 4294967295U
+typedef unsigned int pcre_uint32;
+typedef int pcre_int32;
+#define PCRE_UINT32_MAX UINT_MAX
+#define PCRE_INT32_MAX INT_MAX
+#elif ULONG_MAX == 4294967295UL
+typedef unsigned long int pcre_uint32;
+typedef long int pcre_int32;
+#define PCRE_UINT32_MAX ULONG_MAX
+#define PCRE_INT32_MAX LONG_MAX
+#else
+#error Cannot determine a type for 32-bit integers
+#endif
+
+/* When checking for integer overflow in pcre_compile(), we need to handle
+large integers. If a 64-bit integer type is available, we can use that.
+Otherwise we have to cast to double, which of course requires floating point
+arithmetic. Handle this by defining a macro for the appropriate type. If
+stdint.h is available, include it; it may define INT64_MAX. Systems that do not
+have stdint.h (e.g. Solaris) may have inttypes.h. The macro int64_t may be set
+by "configure". */
+
+#if defined HAVE_STDINT_H
+#include 
+#elif defined HAVE_INTTYPES_H
+#include 
+#endif
+
+#if defined INT64_MAX || defined int64_t
+#define INT64_OR_DOUBLE int64_t
+#else
+#define INT64_OR_DOUBLE double
+#endif
+
+/* All character handling must be done as unsigned characters. Otherwise there
+are problems with top-bit-set characters and functions such as isspace().
+However, we leave the interface to the outside world as char * or short *,
+because that should make things easier for callers. This character type is
+called pcre_uchar.
+
+The IN_UCHARS macro multiply its argument with the byte size of the current
+pcre_uchar type. Useful for memcpy and such operations, whose require the
+byte size of their input/output buffers.
+
+The MAX_255 macro checks whether its pcre_uchar input is less than 256.
+
+The TABLE_GET macro is designed for accessing elements of tables whose contain
+exactly 256 items. When the character is able to contain more than 256
+items, some check is needed before accessing these tables.
+*/
+
+#if defined COMPILE_PCRE8
+
+typedef unsigned char pcre_uchar;
+#define IN_UCHARS(x) (x)
+#define MAX_255(c) 1
+#define TABLE_GET(c, table, default) ((table)[c])
+
+#elif defined COMPILE_PCRE16
+
+#if USHRT_MAX != 65535
+/* This is a warning message. Change PCRE_UCHAR16 to a 16 bit data type in
+pcre.h(.in) and disable (comment out) this message. */
+#error Warning: PCRE_UCHAR16 is not a 16 bit data type.
+#endif
+
+typedef pcre_uint16 pcre_uchar;
+#define UCHAR_SHIFT (1)
+#define IN_UCHARS(x) ((x) << UCHAR_SHIFT)
+#define MAX_255(c) ((c) <= 255u)
+#define TABLE_GET(c, table, default) (MAX_255(c)? ((table)[c]):(default))
+
+#elif defined COMPILE_PCRE32
+
+typedef pcre_uint32 pcre_uchar;
+#define UCHAR_SHIFT (2)
+#define IN_UCHARS(x) ((x) << UCHAR_SHIFT)
+#define MAX_255(c) ((c) <= 255u)
+#define TABLE_GET(c, table, default) (MAX_255(c)? ((table)[c]):(default))
+
+#else
+#error Unsupported compiling mode
+#endif /* COMPILE_PCRE[8|16|32] */
+
+/* This is an unsigned int value that no character can ever have. UTF-8
+characters only go up to 0x7fffffff (though Unicode doesn't go beyond
+0x0010ffff). */
+
+#define NOTACHAR 0xffffffff
+
+/* PCRE is able to support several different kinds of newline (CR, LF, CRLF,
+"any" and "anycrlf" at present). The following macros are used to package up
+testing for newlines. NLBLOCK, PSSTART, and PSEND are defined in the various
+modules to indicate in which datablock the parameters exist, and what the
+start/end of string field names are. */
+
+#define NLTYPE_FIXED    0     /* Newline is a fixed length string */
+#define NLTYPE_ANY      1     /* Newline is any Unicode line ending */
+#define NLTYPE_ANYCRLF  2     /* Newline is CR, LF, or CRLF */
+
+/* This macro checks for a newline at the given position */
+
+#define IS_NEWLINE(p) \
+  ((NLBLOCK->nltype != NLTYPE_FIXED)? \
+    ((p) < NLBLOCK->PSEND && \
+     PRIV(is_newline)((p), NLBLOCK->nltype, NLBLOCK->PSEND, \
+       &(NLBLOCK->nllen), utf)) \
+    : \
+    ((p) <= NLBLOCK->PSEND - NLBLOCK->nllen && \
+     RAWUCHARTEST(p) == NLBLOCK->nl[0] && \
+     (NLBLOCK->nllen == 1 || RAWUCHARTEST(p+1) == NLBLOCK->nl[1])       \
+    ) \
+  )
+
+/* This macro checks for a newline immediately preceding the given position */
+
+#define WAS_NEWLINE(p) \
+  ((NLBLOCK->nltype != NLTYPE_FIXED)? \
+    ((p) > NLBLOCK->PSSTART && \
+     PRIV(was_newline)((p), NLBLOCK->nltype, NLBLOCK->PSSTART, \
+       &(NLBLOCK->nllen), utf)) \
+    : \
+    ((p) >= NLBLOCK->PSSTART + NLBLOCK->nllen && \
+     RAWUCHARTEST(p - NLBLOCK->nllen) == NLBLOCK->nl[0] &&              \
+     (NLBLOCK->nllen == 1 || RAWUCHARTEST(p - NLBLOCK->nllen + 1) == NLBLOCK->nl[1]) \
+    ) \
+  )
+
+/* When PCRE is compiled as a C++ library, the subject pointer can be replaced
+with a custom type. This makes it possible, for example, to allow pcre_exec()
+to process subject strings that are discontinuous by using a smart pointer
+class. It must always be possible to inspect all of the subject string in
+pcre_exec() because of the way it backtracks. Two macros are required in the
+normal case, for sign-unspecified and unsigned char pointers. The former is
+used for the external interface and appears in pcre.h, which is why its name
+must begin with PCRE_. */
+
+#ifdef CUSTOM_SUBJECT_PTR
+#define PCRE_PUCHAR CUSTOM_SUBJECT_PTR
+#else
+#define PCRE_PUCHAR const pcre_uchar *
+#endif
+
+/* Include the public PCRE header and the definitions of UCP character property
+values. */
+
+#include "pcre.h"
+#include "ucp.h"
+
+#ifdef COMPILE_PCRE32
+/* Assert that the public PCRE_UCHAR32 is a 32-bit type */
+typedef int __assert_pcre_uchar32_size[sizeof(PCRE_UCHAR32) == 4 ? 1 : -1];
+#endif
+
+/* When compiling for use with the Virtual Pascal compiler, these functions
+need to have their names changed. PCRE must be compiled with the -DVPCOMPAT
+option on the command line. */
+
+#ifdef VPCOMPAT
+#define strlen(s)        _strlen(s)
+#define strncmp(s1,s2,m) _strncmp(s1,s2,m)
+#define memcmp(s,c,n)    _memcmp(s,c,n)
+#define memcpy(d,s,n)    _memcpy(d,s,n)
+#define memmove(d,s,n)   _memmove(d,s,n)
+#define memset(s,c,n)    _memset(s,c,n)
+#else  /* VPCOMPAT */
+
+/* To cope with SunOS4 and other systems that lack memmove() but have bcopy(),
+define a macro for memmove() if HAVE_MEMMOVE is false, provided that HAVE_BCOPY
+is set. Otherwise, include an emulating function for those systems that have
+neither (there some non-Unix environments where this is the case). */
+
+#ifndef HAVE_MEMMOVE
+#undef  memmove        /* some systems may have a macro */
+#ifdef HAVE_BCOPY
+#define memmove(a, b, c) bcopy(b, a, c)
+#else  /* HAVE_BCOPY */
+static void *
+pcre_memmove(void *d, const void *s, size_t n)
+{
+size_t i;
+unsigned char *dest = (unsigned char *)d;
+const unsigned char *src = (const unsigned char *)s;
+if (dest > src)
+  {
+  dest += n;
+  src += n;
+  for (i = 0; i < n; ++i) *(--dest) = *(--src);
+  return (void *)dest;
+  }
+else
+  {
+  for (i = 0; i < n; ++i) *dest++ = *src++;
+  return (void *)(dest - n);
+  }
+}
+#define memmove(a, b, c) pcre_memmove(a, b, c)
+#endif   /* not HAVE_BCOPY */
+#endif   /* not HAVE_MEMMOVE */
+#endif   /* not VPCOMPAT */
+
+
+/* PCRE keeps offsets in its compiled code as 2-byte quantities (always stored
+in big-endian order) by default. These are used, for example, to link from the
+start of a subpattern to its alternatives and its end. The use of 2 bytes per
+offset limits the size of the compiled regex to around 64K, which is big enough
+for almost everybody. However, I received a request for an even bigger limit.
+For this reason, and also to make the code easier to maintain, the storing and
+loading of offsets from the byte string is now handled by the macros that are
+defined here.
+
+The macros are controlled by the value of LINK_SIZE. This defaults to 2 in
+the config.h file, but can be overridden by using -D on the command line. This
+is automated on Unix systems via the "configure" command. */
+
+#if defined COMPILE_PCRE8
+
+#if LINK_SIZE == 2
+
+#define PUT(a,n,d)   \
+  (a[n] = (d) >> 8), \
+  (a[(n)+1] = (d) & 255)
+
+#define GET(a,n) \
+  (((a)[n] << 8) | (a)[(n)+1])
+
+#define MAX_PATTERN_SIZE (1 << 16)
+
+
+#elif LINK_SIZE == 3
+
+#define PUT(a,n,d)       \
+  (a[n] = (d) >> 16),    \
+  (a[(n)+1] = (d) >> 8), \
+  (a[(n)+2] = (d) & 255)
+
+#define GET(a,n) \
+  (((a)[n] << 16) | ((a)[(n)+1] << 8) | (a)[(n)+2])
+
+#define MAX_PATTERN_SIZE (1 << 24)
+
+
+#elif LINK_SIZE == 4
+
+#define PUT(a,n,d)        \
+  (a[n] = (d) >> 24),     \
+  (a[(n)+1] = (d) >> 16), \
+  (a[(n)+2] = (d) >> 8),  \
+  (a[(n)+3] = (d) & 255)
+
+#define GET(a,n) \
+  (((a)[n] << 24) | ((a)[(n)+1] << 16) | ((a)[(n)+2] << 8) | (a)[(n)+3])
+
+/* Keep it positive */
+#define MAX_PATTERN_SIZE (1 << 30)
+
+#else
+#error LINK_SIZE must be either 2, 3, or 4
+#endif
+
+#elif defined COMPILE_PCRE16
+
+#if LINK_SIZE == 2
+
+/* Redefine LINK_SIZE as a multiple of sizeof(pcre_uchar) */
+#undef LINK_SIZE
+#define LINK_SIZE 1
+
+#define PUT(a,n,d)   \
+  (a[n] = (d))
+
+#define GET(a,n) \
+  (a[n])
+
+#define MAX_PATTERN_SIZE (1 << 16)
+
+#elif LINK_SIZE == 3 || LINK_SIZE == 4
+
+/* Redefine LINK_SIZE as a multiple of sizeof(pcre_uchar) */
+#undef LINK_SIZE
+#define LINK_SIZE 2
+
+#define PUT(a,n,d)   \
+  (a[n] = (d) >> 16), \
+  (a[(n)+1] = (d) & 65535)
+
+#define GET(a,n) \
+  (((a)[n] << 16) | (a)[(n)+1])
+
+/* Keep it positive */
+#define MAX_PATTERN_SIZE (1 << 30)
+
+#else
+#error LINK_SIZE must be either 2, 3, or 4
+#endif
+
+#elif defined COMPILE_PCRE32
+
+/* Only supported LINK_SIZE is 4 */
+/* Redefine LINK_SIZE as a multiple of sizeof(pcre_uchar) */
+#undef LINK_SIZE
+#define LINK_SIZE 1
+
+#define PUT(a,n,d)   \
+  (a[n] = (d))
+
+#define GET(a,n) \
+  (a[n])
+
+/* Keep it positive */
+#define MAX_PATTERN_SIZE (1 << 30)
+
+#else
+#error Unsupported compiling mode
+#endif /* COMPILE_PCRE[8|16|32] */
+
+/* Convenience macro defined in terms of the others */
+
+#define PUTINC(a,n,d)   PUT(a,n,d), a += LINK_SIZE
+
+
+/* PCRE uses some other 2-byte quantities that do not change when the size of
+offsets changes. There are used for repeat counts and for other things such as
+capturing parenthesis numbers in back references. */
+
+#if defined COMPILE_PCRE8
+
+#define IMM2_SIZE 2
+
+#define PUT2(a,n,d)   \
+  a[n] = (d) >> 8; \
+  a[(n)+1] = (d) & 255
+
+/* For reasons that I do not understand, the expression in this GET2 macro is
+treated by gcc as a signed expression, even when a is declared as unsigned. It
+seems that any kind of arithmetic results in a signed value. */
+
+#define GET2(a,n) \
+  (unsigned int)(((a)[n] << 8) | (a)[(n)+1])
+
+#elif defined COMPILE_PCRE16
+
+#define IMM2_SIZE 1
+
+#define PUT2(a,n,d)   \
+   a[n] = d
+
+#define GET2(a,n) \
+   a[n]
+
+#elif defined COMPILE_PCRE32
+
+#define IMM2_SIZE 1
+
+#define PUT2(a,n,d)   \
+   a[n] = d
+
+#define GET2(a,n) \
+   a[n]
+
+#else
+#error Unsupported compiling mode
+#endif /* COMPILE_PCRE[8|16|32] */
+
+#define PUT2INC(a,n,d)  PUT2(a,n,d), a += IMM2_SIZE
+
+/* The maximum length of a MARK name is currently one data unit; it may be
+changed in future to be a fixed number of bytes or to depend on LINK_SIZE. */
+
+#if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
+#define MAX_MARK ((1u << 16) - 1)
+#else
+#define MAX_MARK ((1u << 8) - 1)
+#endif
+
+/* When UTF encoding is being used, a character is no longer just a single
+byte. The macros for character handling generate simple sequences when used in
+character-mode, and more complicated ones for UTF characters. GETCHARLENTEST
+and other macros are not used when UTF is not supported, so they are not
+defined. To make sure they can never even appear when UTF support is omitted,
+we don't even define them. */
+
+#ifndef SUPPORT_UTF
+
+/* #define MAX_VALUE_FOR_SINGLE_CHAR */
+/* #define HAS_EXTRALEN(c) */
+/* #define GET_EXTRALEN(c) */
+/* #define NOT_FIRSTCHAR(c) */
+#define GETCHAR(c, eptr) c = *eptr;
+#define GETCHARTEST(c, eptr) c = *eptr;
+#define GETCHARINC(c, eptr) c = *eptr++;
+#define GETCHARINCTEST(c, eptr) c = *eptr++;
+#define GETCHARLEN(c, eptr, len) c = *eptr;
+#define RAWUCHAR(eptr) (*(eptr))
+#define RAWUCHARINC(eptr) (*(eptr)++)
+#define RAWUCHARTEST(eptr) (*(eptr))
+#define RAWUCHARINCTEST(eptr) (*(eptr)++)
+/* #define GETCHARLENTEST(c, eptr, len) */
+/* #define BACKCHAR(eptr) */
+/* #define FORWARDCHAR(eptr) */
+/* #define ACROSSCHAR(condition, eptr, action) */
+
+#else   /* SUPPORT_UTF */
+
+/* Tests whether the code point needs extra characters to decode. */
+
+#define HASUTF8EXTRALEN(c) ((c) >= 0xc0)
+
+/* Base macro to pick up the remaining bytes of a UTF-8 character, not
+advancing the pointer. */
+
+#define GETUTF8(c, eptr) \
+    { \
+    if ((c & 0x20) == 0) \
+      c = ((c & 0x1f) << 6) | (eptr[1] & 0x3f); \
+    else if ((c & 0x10) == 0) \
+      c = ((c & 0x0f) << 12) | ((eptr[1] & 0x3f) << 6) | (eptr[2] & 0x3f); \
+    else if ((c & 0x08) == 0) \
+      c = ((c & 0x07) << 18) | ((eptr[1] & 0x3f) << 12) | \
+      ((eptr[2] & 0x3f) << 6) | (eptr[3] & 0x3f); \
+    else if ((c & 0x04) == 0) \
+      c = ((c & 0x03) << 24) | ((eptr[1] & 0x3f) << 18) | \
+          ((eptr[2] & 0x3f) << 12) | ((eptr[3] & 0x3f) << 6) | \
+          (eptr[4] & 0x3f); \
+    else \
+      c = ((c & 0x01) << 30) | ((eptr[1] & 0x3f) << 24) | \
+          ((eptr[2] & 0x3f) << 18) | ((eptr[3] & 0x3f) << 12) | \
+          ((eptr[4] & 0x3f) << 6) | (eptr[5] & 0x3f); \
+    }
+
+/* Base macro to pick up the remaining bytes of a UTF-8 character, advancing
+the pointer. */
+
+#define GETUTF8INC(c, eptr) \
+    { \
+    if ((c & 0x20) == 0) \
+      c = ((c & 0x1f) << 6) | (*eptr++ & 0x3f); \
+    else if ((c & 0x10) == 0) \
+      { \
+      c = ((c & 0x0f) << 12) | ((*eptr & 0x3f) << 6) | (eptr[1] & 0x3f); \
+      eptr += 2; \
+      } \
+    else if ((c & 0x08) == 0) \
+      { \
+      c = ((c & 0x07) << 18) | ((*eptr & 0x3f) << 12) | \
+          ((eptr[1] & 0x3f) << 6) | (eptr[2] & 0x3f); \
+      eptr += 3; \
+      } \
+    else if ((c & 0x04) == 0) \
+      { \
+      c = ((c & 0x03) << 24) | ((*eptr & 0x3f) << 18) | \
+          ((eptr[1] & 0x3f) << 12) | ((eptr[2] & 0x3f) << 6) | \
+          (eptr[3] & 0x3f); \
+      eptr += 4; \
+      } \
+    else \
+      { \
+      c = ((c & 0x01) << 30) | ((*eptr & 0x3f) << 24) | \
+          ((eptr[1] & 0x3f) << 18) | ((eptr[2] & 0x3f) << 12) | \
+          ((eptr[3] & 0x3f) << 6) | (eptr[4] & 0x3f); \
+      eptr += 5; \
+      } \
+    }
+
+#if defined COMPILE_PCRE8
+
+/* These macros were originally written in the form of loops that used data
+from the tables whose names start with PRIV(utf8_table). They were rewritten by
+a user so as not to use loops, because in some environments this gives a
+significant performance advantage, and it seems never to do any harm. */
+
+/* Tells the biggest code point which can be encoded as a single character. */
+
+#define MAX_VALUE_FOR_SINGLE_CHAR 127
+
+/* Tests whether the code point needs extra characters to decode. */
+
+#define HAS_EXTRALEN(c) ((c) >= 0xc0)
+
+/* Returns with the additional number of characters if IS_MULTICHAR(c) is TRUE.
+Otherwise it has an undefined behaviour. */
+
+#define GET_EXTRALEN(c) (PRIV(utf8_table4)[(c) & 0x3f])
+
+/* Returns TRUE, if the given character is not the first character
+of a UTF sequence. */
+
+#define NOT_FIRSTCHAR(c) (((c) & 0xc0) == 0x80)
+
+/* Get the next UTF-8 character, not advancing the pointer. This is called when
+we know we are in UTF-8 mode. */
+
+#define GETCHAR(c, eptr) \
+  c = *eptr; \
+  if (c >= 0xc0) GETUTF8(c, eptr);
+
+/* Get the next UTF-8 character, testing for UTF-8 mode, and not advancing the
+pointer. */
+
+#define GETCHARTEST(c, eptr) \
+  c = *eptr; \
+  if (utf && c >= 0xc0) GETUTF8(c, eptr);
+
+/* Get the next UTF-8 character, advancing the pointer. This is called when we
+know we are in UTF-8 mode. */
+
+#define GETCHARINC(c, eptr) \
+  c = *eptr++; \
+  if (c >= 0xc0) GETUTF8INC(c, eptr);
+
+/* Get the next character, testing for UTF-8 mode, and advancing the pointer.
+This is called when we don't know if we are in UTF-8 mode. */
+
+#define GETCHARINCTEST(c, eptr) \
+  c = *eptr++; \
+  if (utf && c >= 0xc0) GETUTF8INC(c, eptr);
+
+/* Base macro to pick up the remaining bytes of a UTF-8 character, not
+advancing the pointer, incrementing the length. */
+
+#define GETUTF8LEN(c, eptr, len) \
+    { \
+    if ((c & 0x20) == 0) \
+      { \
+      c = ((c & 0x1f) << 6) | (eptr[1] & 0x3f); \
+      len++; \
+      } \
+    else if ((c & 0x10)  == 0) \
+      { \
+      c = ((c & 0x0f) << 12) | ((eptr[1] & 0x3f) << 6) | (eptr[2] & 0x3f); \
+      len += 2; \
+      } \
+    else if ((c & 0x08)  == 0) \
+      {\
+      c = ((c & 0x07) << 18) | ((eptr[1] & 0x3f) << 12) | \
+          ((eptr[2] & 0x3f) << 6) | (eptr[3] & 0x3f); \
+      len += 3; \
+      } \
+    else if ((c & 0x04)  == 0) \
+      { \
+      c = ((c & 0x03) << 24) | ((eptr[1] & 0x3f) << 18) | \
+          ((eptr[2] & 0x3f) << 12) | ((eptr[3] & 0x3f) << 6) | \
+          (eptr[4] & 0x3f); \
+      len += 4; \
+      } \
+    else \
+      {\
+      c = ((c & 0x01) << 30) | ((eptr[1] & 0x3f) << 24) | \
+          ((eptr[2] & 0x3f) << 18) | ((eptr[3] & 0x3f) << 12) | \
+          ((eptr[4] & 0x3f) << 6) | (eptr[5] & 0x3f); \
+      len += 5; \
+      } \
+    }
+
+/* Get the next UTF-8 character, not advancing the pointer, incrementing length
+if there are extra bytes. This is called when we know we are in UTF-8 mode. */
+
+#define GETCHARLEN(c, eptr, len) \
+  c = *eptr; \
+  if (c >= 0xc0) GETUTF8LEN(c, eptr, len);
+
+/* Get the next UTF-8 character, testing for UTF-8 mode, not advancing the
+pointer, incrementing length if there are extra bytes. This is called when we
+do not know if we are in UTF-8 mode. */
+
+#define GETCHARLENTEST(c, eptr, len) \
+  c = *eptr; \
+  if (utf && c >= 0xc0) GETUTF8LEN(c, eptr, len);
+
+/* Returns the next uchar, not advancing the pointer. This is called when
+we know we are in UTF mode. */
+
+#define RAWUCHAR(eptr) \
+  (*(eptr))
+
+/* Returns the next uchar, advancing the pointer. This is called when
+we know we are in UTF mode. */
+
+#define RAWUCHARINC(eptr) \
+  (*((eptr)++))
+
+/* Returns the next uchar, testing for UTF mode, and not advancing the
+pointer. */
+
+#define RAWUCHARTEST(eptr) \
+  (*(eptr))
+
+/* Returns the next uchar, testing for UTF mode, advancing the
+pointer. */
+
+#define RAWUCHARINCTEST(eptr) \
+  (*((eptr)++))
+
+/* If the pointer is not at the start of a character, move it back until
+it is. This is called only in UTF-8 mode - we don't put a test within the macro
+because almost all calls are already within a block of UTF-8 only code. */
+
+#define BACKCHAR(eptr) while((*eptr & 0xc0) == 0x80) eptr--
+
+/* Same as above, just in the other direction. */
+#define FORWARDCHAR(eptr) while((*eptr & 0xc0) == 0x80) eptr++
+
+/* Same as above, but it allows a fully customizable form. */
+#define ACROSSCHAR(condition, eptr, action) \
+  while((condition) && ((eptr) & 0xc0) == 0x80) action
+
+#elif defined COMPILE_PCRE16
+
+/* Tells the biggest code point which can be encoded as a single character. */
+
+#define MAX_VALUE_FOR_SINGLE_CHAR 65535
+
+/* Tests whether the code point needs extra characters to decode. */
+
+#define HAS_EXTRALEN(c) (((c) & 0xfc00) == 0xd800)
+
+/* Returns with the additional number of characters if IS_MULTICHAR(c) is TRUE.
+Otherwise it has an undefined behaviour. */
+
+#define GET_EXTRALEN(c) 1
+
+/* Returns TRUE, if the given character is not the first character
+of a UTF sequence. */
+
+#define NOT_FIRSTCHAR(c) (((c) & 0xfc00) == 0xdc00)
+
+/* Base macro to pick up the low surrogate of a UTF-16 character, not
+advancing the pointer. */
+
+#define GETUTF16(c, eptr) \
+   { c = (((c & 0x3ff) << 10) | (eptr[1] & 0x3ff)) + 0x10000; }
+
+/* Get the next UTF-16 character, not advancing the pointer. This is called when
+we know we are in UTF-16 mode. */
+
+#define GETCHAR(c, eptr) \
+  c = *eptr; \
+  if ((c & 0xfc00) == 0xd800) GETUTF16(c, eptr);
+
+/* Get the next UTF-16 character, testing for UTF-16 mode, and not advancing the
+pointer. */
+
+#define GETCHARTEST(c, eptr) \
+  c = *eptr; \
+  if (utf && (c & 0xfc00) == 0xd800) GETUTF16(c, eptr);
+
+/* Base macro to pick up the low surrogate of a UTF-16 character, advancing
+the pointer. */
+
+#define GETUTF16INC(c, eptr) \
+   { c = (((c & 0x3ff) << 10) | (*eptr++ & 0x3ff)) + 0x10000; }
+
+/* Get the next UTF-16 character, advancing the pointer. This is called when we
+know we are in UTF-16 mode. */
+
+#define GETCHARINC(c, eptr) \
+  c = *eptr++; \
+  if ((c & 0xfc00) == 0xd800) GETUTF16INC(c, eptr);
+
+/* Get the next character, testing for UTF-16 mode, and advancing the pointer.
+This is called when we don't know if we are in UTF-16 mode. */
+
+#define GETCHARINCTEST(c, eptr) \
+  c = *eptr++; \
+  if (utf && (c & 0xfc00) == 0xd800) GETUTF16INC(c, eptr);
+
+/* Base macro to pick up the low surrogate of a UTF-16 character, not
+advancing the pointer, incrementing the length. */
+
+#define GETUTF16LEN(c, eptr, len) \
+   { c = (((c & 0x3ff) << 10) | (eptr[1] & 0x3ff)) + 0x10000; len++; }
+
+/* Get the next UTF-16 character, not advancing the pointer, incrementing
+length if there is a low surrogate. This is called when we know we are in
+UTF-16 mode. */
+
+#define GETCHARLEN(c, eptr, len) \
+  c = *eptr; \
+  if ((c & 0xfc00) == 0xd800) GETUTF16LEN(c, eptr, len);
+
+/* Get the next UTF-816character, testing for UTF-16 mode, not advancing the
+pointer, incrementing length if there is a low surrogate. This is called when
+we do not know if we are in UTF-16 mode. */
+
+#define GETCHARLENTEST(c, eptr, len) \
+  c = *eptr; \
+  if (utf && (c & 0xfc00) == 0xd800) GETUTF16LEN(c, eptr, len);
+
+/* Returns the next uchar, not advancing the pointer. This is called when
+we know we are in UTF mode. */
+
+#define RAWUCHAR(eptr) \
+  (*(eptr))
+
+/* Returns the next uchar, advancing the pointer. This is called when
+we know we are in UTF mode. */
+
+#define RAWUCHARINC(eptr) \
+  (*((eptr)++))
+
+/* Returns the next uchar, testing for UTF mode, and not advancing the
+pointer. */
+
+#define RAWUCHARTEST(eptr) \
+  (*(eptr))
+
+/* Returns the next uchar, testing for UTF mode, advancing the
+pointer. */
+
+#define RAWUCHARINCTEST(eptr) \
+  (*((eptr)++))
+
+/* If the pointer is not at the start of a character, move it back until
+it is. This is called only in UTF-16 mode - we don't put a test within the
+macro because almost all calls are already within a block of UTF-16 only
+code. */
+
+#define BACKCHAR(eptr) if ((*eptr & 0xfc00) == 0xdc00) eptr--
+
+/* Same as above, just in the other direction. */
+#define FORWARDCHAR(eptr) if ((*eptr & 0xfc00) == 0xdc00) eptr++
+
+/* Same as above, but it allows a fully customizable form. */
+#define ACROSSCHAR(condition, eptr, action) \
+  if ((condition) && ((eptr) & 0xfc00) == 0xdc00) action
+
+#elif defined COMPILE_PCRE32
+
+/* These are trivial for the 32-bit library, since all UTF-32 characters fit
+into one pcre_uchar unit. */
+#define MAX_VALUE_FOR_SINGLE_CHAR (0x10ffffu)
+#define HAS_EXTRALEN(c) (0)
+#define GET_EXTRALEN(c) (0)
+#define NOT_FIRSTCHAR(c) (0)
+
+/* Get the next UTF-32 character, not advancing the pointer. This is called when
+we know we are in UTF-32 mode. */
+
+#define GETCHAR(c, eptr) \
+  c = *(eptr);
+
+/* Get the next UTF-32 character, testing for UTF-32 mode, and not advancing the
+pointer. */
+
+#define GETCHARTEST(c, eptr) \
+  c = *(eptr);
+
+/* Get the next UTF-32 character, advancing the pointer. This is called when we
+know we are in UTF-32 mode. */
+
+#define GETCHARINC(c, eptr) \
+  c = *((eptr)++);
+
+/* Get the next character, testing for UTF-32 mode, and advancing the pointer.
+This is called when we don't know if we are in UTF-32 mode. */
+
+#define GETCHARINCTEST(c, eptr) \
+  c = *((eptr)++);
+
+/* Get the next UTF-32 character, not advancing the pointer, not incrementing
+length (since all UTF-32 is of length 1). This is called when we know we are in
+UTF-32 mode. */
+
+#define GETCHARLEN(c, eptr, len) \
+  GETCHAR(c, eptr)
+
+/* Get the next UTF-32character, testing for UTF-32 mode, not advancing the
+pointer, not incrementing the length (since all UTF-32 is of length 1).
+This is called when we do not know if we are in UTF-32 mode. */
+
+#define GETCHARLENTEST(c, eptr, len) \
+  GETCHARTEST(c, eptr)
+
+/* Returns the next uchar, not advancing the pointer. This is called when
+we know we are in UTF mode. */
+
+#define RAWUCHAR(eptr) \
+  (*(eptr))
+
+/* Returns the next uchar, advancing the pointer. This is called when
+we know we are in UTF mode. */
+
+#define RAWUCHARINC(eptr) \
+  (*((eptr)++))
+
+/* Returns the next uchar, testing for UTF mode, and not advancing the
+pointer. */
+
+#define RAWUCHARTEST(eptr) \
+  (*(eptr))
+
+/* Returns the next uchar, testing for UTF mode, advancing the
+pointer. */
+
+#define RAWUCHARINCTEST(eptr) \
+  (*((eptr)++))
+
+/* If the pointer is not at the start of a character, move it back until
+it is. This is called only in UTF-32 mode - we don't put a test within the
+macro because almost all calls are already within a block of UTF-32 only
+code.
+These are all no-ops since all UTF-32 characters fit into one pcre_uchar. */
+
+#define BACKCHAR(eptr) do { } while (0)
+
+/* Same as above, just in the other direction. */
+#define FORWARDCHAR(eptr) do { } while (0)
+
+/* Same as above, but it allows a fully customizable form. */
+#define ACROSSCHAR(condition, eptr, action) do { } while (0)
+
+#else
+#error Unsupported compiling mode
+#endif /* COMPILE_PCRE[8|16|32] */
+
+#endif  /* SUPPORT_UTF */
+
+/* Tests for Unicode horizontal and vertical whitespace characters must check a
+number of different values. Using a switch statement for this generates the
+fastest code (no loop, no memory access), and there are several places in the
+interpreter code where this happens. In order to ensure that all the case lists
+remain in step, we use macros so that there is only one place where the lists
+are defined.
+
+These values are also required as lists in pcre_compile.c when processing \h,
+\H, \v and \V in a character class. The lists are defined in pcre_tables.c, but
+macros that define the values are here so that all the definitions are
+together. The lists must be in ascending character order, terminated by
+NOTACHAR (which is 0xffffffff).
+
+Any changes should ensure that the various macros are kept in step with each
+other. NOTE: The values also appear in pcre_jit_compile.c. */
+
+/* ------ ASCII/Unicode environments ------ */
+
+#ifndef EBCDIC
+
+#define HSPACE_LIST \
+  CHAR_HT, CHAR_SPACE, 0xa0, \
+  0x1680, 0x180e, 0x2000, 0x2001, 0x2002, 0x2003, 0x2004, 0x2005, \
+  0x2006, 0x2007, 0x2008, 0x2009, 0x200A, 0x202f, 0x205f, 0x3000, \
+  NOTACHAR
+
+#define HSPACE_MULTIBYTE_CASES \
+  case 0x1680:  /* OGHAM SPACE MARK */ \
+  case 0x180e:  /* MONGOLIAN VOWEL SEPARATOR */ \
+  case 0x2000:  /* EN QUAD */ \
+  case 0x2001:  /* EM QUAD */ \
+  case 0x2002:  /* EN SPACE */ \
+  case 0x2003:  /* EM SPACE */ \
+  case 0x2004:  /* THREE-PER-EM SPACE */ \
+  case 0x2005:  /* FOUR-PER-EM SPACE */ \
+  case 0x2006:  /* SIX-PER-EM SPACE */ \
+  case 0x2007:  /* FIGURE SPACE */ \
+  case 0x2008:  /* PUNCTUATION SPACE */ \
+  case 0x2009:  /* THIN SPACE */ \
+  case 0x200A:  /* HAIR SPACE */ \
+  case 0x202f:  /* NARROW NO-BREAK SPACE */ \
+  case 0x205f:  /* MEDIUM MATHEMATICAL SPACE */ \
+  case 0x3000   /* IDEOGRAPHIC SPACE */
+
+#define HSPACE_BYTE_CASES \
+  case CHAR_HT: \
+  case CHAR_SPACE: \
+  case 0xa0     /* NBSP */
+
+#define HSPACE_CASES \
+  HSPACE_BYTE_CASES: \
+  HSPACE_MULTIBYTE_CASES
+
+#define VSPACE_LIST \
+  CHAR_LF, CHAR_VT, CHAR_FF, CHAR_CR, CHAR_NEL, 0x2028, 0x2029, NOTACHAR
+
+#define VSPACE_MULTIBYTE_CASES \
+  case 0x2028:    /* LINE SEPARATOR */ \
+  case 0x2029     /* PARAGRAPH SEPARATOR */
+
+#define VSPACE_BYTE_CASES \
+  case CHAR_LF: \
+  case CHAR_VT: \
+  case CHAR_FF: \
+  case CHAR_CR: \
+  case CHAR_NEL
+
+#define VSPACE_CASES \
+  VSPACE_BYTE_CASES: \
+  VSPACE_MULTIBYTE_CASES
+
+/* ------ EBCDIC environments ------ */
+
+#else
+#define HSPACE_LIST CHAR_HT, CHAR_SPACE
+
+#define HSPACE_BYTE_CASES \
+  case CHAR_HT: \
+  case CHAR_SPACE
+
+#define HSPACE_CASES HSPACE_BYTE_CASES
+
+#ifdef EBCDIC_NL25
+#define VSPACE_LIST \
+  CHAR_VT, CHAR_FF, CHAR_CR, CHAR_NEL, CHAR_LF, NOTACHAR
+#else
+#define VSPACE_LIST \
+  CHAR_VT, CHAR_FF, CHAR_CR, CHAR_LF, CHAR_NEL, NOTACHAR
+#endif
+
+#define VSPACE_BYTE_CASES \
+  case CHAR_LF: \
+  case CHAR_VT: \
+  case CHAR_FF: \
+  case CHAR_CR: \
+  case CHAR_NEL
+
+#define VSPACE_CASES VSPACE_BYTE_CASES
+#endif  /* EBCDIC */
+
+/* ------ End of whitespace macros ------ */
+
+
+
+/* Private flags containing information about the compiled regex. They used to
+live at the top end of the options word, but that got almost full, so they were
+moved to a 16-bit flags word - which got almost full, so now they are in a
+32-bit flags word. From release 8.00, PCRE_NOPARTIAL is unused, as the
+restrictions on partial matching have been lifted. It remains for backwards
+compatibility. */
+
+#define PCRE_MODE8         0x00000001  /* compiled in 8 bit mode */
+#define PCRE_MODE16        0x00000002  /* compiled in 16 bit mode */
+#define PCRE_MODE32        0x00000004  /* compiled in 32 bit mode */
+#define PCRE_FIRSTSET      0x00000010  /* first_char is set */
+#define PCRE_FCH_CASELESS  0x00000020  /* caseless first char */
+#define PCRE_REQCHSET      0x00000040  /* req_byte is set */
+#define PCRE_RCH_CASELESS  0x00000080  /* caseless requested char */
+#define PCRE_STARTLINE     0x00000100  /* start after \n for multiline */
+#define PCRE_NOPARTIAL     0x00000200  /* can't use partial with this regex */
+#define PCRE_JCHANGED      0x00000400  /* j option used in regex */
+#define PCRE_HASCRORLF     0x00000800  /* explicit \r or \n in pattern */
+#define PCRE_HASTHEN       0x00001000  /* pattern contains (*THEN) */
+#define PCRE_MLSET         0x00002000  /* match limit set by regex */
+#define PCRE_RLSET         0x00004000  /* recursion limit set by regex */
+
+#if defined COMPILE_PCRE8
+#define PCRE_MODE          PCRE_MODE8
+#elif defined COMPILE_PCRE16
+#define PCRE_MODE          PCRE_MODE16
+#elif defined COMPILE_PCRE32
+#define PCRE_MODE          PCRE_MODE32
+#endif
+#define PCRE_MODE_MASK     (PCRE_MODE8 | PCRE_MODE16 | PCRE_MODE32)
+
+/* Flags for the "extra" block produced by pcre_study(). */
+
+#define PCRE_STUDY_MAPPED  0x0001  /* a map of starting chars exists */
+#define PCRE_STUDY_MINLEN  0x0002  /* a minimum length field exists */
+
+/* Masks for identifying the public options that are permitted at compile
+time, run time, or study time, respectively. */
+
+#define PCRE_NEWLINE_BITS (PCRE_NEWLINE_CR|PCRE_NEWLINE_LF|PCRE_NEWLINE_ANY| \
+                           PCRE_NEWLINE_ANYCRLF)
+
+#define PUBLIC_COMPILE_OPTIONS \
+  (PCRE_CASELESS|PCRE_EXTENDED|PCRE_ANCHORED|PCRE_MULTILINE| \
+   PCRE_DOTALL|PCRE_DOLLAR_ENDONLY|PCRE_EXTRA|PCRE_UNGREEDY|PCRE_UTF8| \
+   PCRE_NO_AUTO_CAPTURE|PCRE_NO_UTF8_CHECK|PCRE_AUTO_CALLOUT|PCRE_FIRSTLINE| \
+   PCRE_DUPNAMES|PCRE_NEWLINE_BITS|PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE| \
+   PCRE_JAVASCRIPT_COMPAT|PCRE_UCP|PCRE_NO_START_OPTIMIZE|PCRE_NEVER_UTF)
+
+#define PUBLIC_EXEC_OPTIONS \
+  (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NOTEMPTY_ATSTART| \
+   PCRE_NO_UTF8_CHECK|PCRE_PARTIAL_HARD|PCRE_PARTIAL_SOFT|PCRE_NEWLINE_BITS| \
+   PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE|PCRE_NO_START_OPTIMIZE)
+
+#define PUBLIC_DFA_EXEC_OPTIONS \
+  (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NOTEMPTY_ATSTART| \
+   PCRE_NO_UTF8_CHECK|PCRE_PARTIAL_HARD|PCRE_PARTIAL_SOFT|PCRE_DFA_SHORTEST| \
+   PCRE_DFA_RESTART|PCRE_NEWLINE_BITS|PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE| \
+   PCRE_NO_START_OPTIMIZE)
+
+#define PUBLIC_STUDY_OPTIONS \
+   (PCRE_STUDY_JIT_COMPILE|PCRE_STUDY_JIT_PARTIAL_SOFT_COMPILE| \
+    PCRE_STUDY_JIT_PARTIAL_HARD_COMPILE|PCRE_STUDY_EXTRA_NEEDED)
+
+#define PUBLIC_JIT_EXEC_OPTIONS \
+   (PCRE_NO_UTF8_CHECK|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|\
+    PCRE_NOTEMPTY_ATSTART|PCRE_PARTIAL_SOFT|PCRE_PARTIAL_HARD)
+
+/* Magic number to provide a small check against being handed junk. */
+
+#define MAGIC_NUMBER  0x50435245UL   /* 'PCRE' */
+
+/* This variable is used to detect a loaded regular expression
+in different endianness. */
+
+#define REVERSED_MAGIC_NUMBER  0x45524350UL   /* 'ERCP' */
+
+/* The maximum remaining length of subject we are prepared to search for a
+req_byte match. */
+
+#define REQ_BYTE_MAX 1000
+
+/* Miscellaneous definitions. The #ifndef is to pacify compiler warnings in
+environments where these macros are defined elsewhere. Unfortunately, there
+is no way to do the same for the typedef. */
+
+typedef int BOOL;
+
+#ifndef FALSE
+#define FALSE   0
+#define TRUE    1
+#endif
+
+/* If PCRE is to support UTF-8 on EBCDIC platforms, we cannot use normal
+character constants like '*' because the compiler would emit their EBCDIC code,
+which is different from their ASCII/UTF-8 code. Instead we define macros for
+the characters so that they always use the ASCII/UTF-8 code when UTF-8 support
+is enabled. When UTF-8 support is not enabled, the definitions use character
+literals. Both character and string versions of each character are needed, and
+there are some longer strings as well.
+
+This means that, on EBCDIC platforms, the PCRE library can handle either
+EBCDIC, or UTF-8, but not both. To support both in the same compiled library
+would need different lookups depending on whether PCRE_UTF8 was set or not.
+This would make it impossible to use characters in switch/case statements,
+which would reduce performance. For a theoretical use (which nobody has asked
+for) in a minority area (EBCDIC platforms), this is not sensible. Any
+application that did need both could compile two versions of the library, using
+macros to give the functions distinct names. */
+
+#ifndef SUPPORT_UTF
+
+/* UTF-8 support is not enabled; use the platform-dependent character literals
+so that PCRE works in both ASCII and EBCDIC environments, but only in non-UTF
+mode. Newline characters are problematic in EBCDIC. Though it has CR and LF
+characters, a common practice has been to use its NL (0x15) character as the
+line terminator in C-like processing environments. However, sometimes the LF
+(0x25) character is used instead, according to this Unicode document:
+
+http://unicode.org/standard/reports/tr13/tr13-5.html
+
+PCRE defaults EBCDIC NL to 0x15, but has a build-time option to select 0x25
+instead. Whichever is *not* chosen is defined as NEL.
+
+In both ASCII and EBCDIC environments, CHAR_NL and CHAR_LF are synonyms for the
+same code point. */
+
+#ifdef EBCDIC
+
+#ifndef EBCDIC_NL25
+#define CHAR_NL                     '\x15'
+#define CHAR_NEL                    '\x25'
+#define STR_NL                      "\x15"
+#define STR_NEL                     "\x25"
+#else
+#define CHAR_NL                     '\x25'
+#define CHAR_NEL                    '\x15'
+#define STR_NL                      "\x25"
+#define STR_NEL                     "\x15"
+#endif
+
+#define CHAR_LF                     CHAR_NL
+#define STR_LF                      STR_NL
+
+#define CHAR_ESC                    '\047'
+#define CHAR_DEL                    '\007'
+#define STR_ESC                     "\047"
+#define STR_DEL                     "\007"
+
+#else  /* Not EBCDIC */
+
+/* In ASCII/Unicode, linefeed is '\n' and we equate this to NL for
+compatibility. NEL is the Unicode newline character; make sure it is
+a positive value. */
+
+#define CHAR_LF                     '\n'
+#define CHAR_NL                     CHAR_LF
+#define CHAR_NEL                    ((unsigned char)'\x85')
+#define CHAR_ESC                    '\033'
+#define CHAR_DEL                    '\177'
+
+#define STR_LF                      "\n"
+#define STR_NL                      STR_LF
+#define STR_NEL                     "\x85"
+#define STR_ESC                     "\033"
+#define STR_DEL                     "\177"
+
+#endif  /* EBCDIC */
+
+/* The remaining definitions work in both environments. */
+
+#define CHAR_NULL                   '\0'
+#define CHAR_HT                     '\t'
+#define CHAR_VT                     '\v'
+#define CHAR_FF                     '\f'
+#define CHAR_CR                     '\r'
+#define CHAR_BS                     '\b'
+#define CHAR_BEL                    '\a'
+
+#define CHAR_SPACE                  ' '
+#define CHAR_EXCLAMATION_MARK       '!'
+#define CHAR_QUOTATION_MARK         '"'
+#define CHAR_NUMBER_SIGN            '#'
+#define CHAR_DOLLAR_SIGN            '$'
+#define CHAR_PERCENT_SIGN           '%'
+#define CHAR_AMPERSAND              '&'
+#define CHAR_APOSTROPHE             '\''
+#define CHAR_LEFT_PARENTHESIS       '('
+#define CHAR_RIGHT_PARENTHESIS      ')'
+#define CHAR_ASTERISK               '*'
+#define CHAR_PLUS                   '+'
+#define CHAR_COMMA                  ','
+#define CHAR_MINUS                  '-'
+#define CHAR_DOT                    '.'
+#define CHAR_SLASH                  '/'
+#define CHAR_0                      '0'
+#define CHAR_1                      '1'
+#define CHAR_2                      '2'
+#define CHAR_3                      '3'
+#define CHAR_4                      '4'
+#define CHAR_5                      '5'
+#define CHAR_6                      '6'
+#define CHAR_7                      '7'
+#define CHAR_8                      '8'
+#define CHAR_9                      '9'
+#define CHAR_COLON                  ':'
+#define CHAR_SEMICOLON              ';'
+#define CHAR_LESS_THAN_SIGN         '<'
+#define CHAR_EQUALS_SIGN            '='
+#define CHAR_GREATER_THAN_SIGN      '>'
+#define CHAR_QUESTION_MARK          '?'
+#define CHAR_COMMERCIAL_AT          '@'
+#define CHAR_A                      'A'
+#define CHAR_B                      'B'
+#define CHAR_C                      'C'
+#define CHAR_D                      'D'
+#define CHAR_E                      'E'
+#define CHAR_F                      'F'
+#define CHAR_G                      'G'
+#define CHAR_H                      'H'
+#define CHAR_I                      'I'
+#define CHAR_J                      'J'
+#define CHAR_K                      'K'
+#define CHAR_L                      'L'
+#define CHAR_M                      'M'
+#define CHAR_N                      'N'
+#define CHAR_O                      'O'
+#define CHAR_P                      'P'
+#define CHAR_Q                      'Q'
+#define CHAR_R                      'R'
+#define CHAR_S                      'S'
+#define CHAR_T                      'T'
+#define CHAR_U                      'U'
+#define CHAR_V                      'V'
+#define CHAR_W                      'W'
+#define CHAR_X                      'X'
+#define CHAR_Y                      'Y'
+#define CHAR_Z                      'Z'
+#define CHAR_LEFT_SQUARE_BRACKET    '['
+#define CHAR_BACKSLASH              '\\'
+#define CHAR_RIGHT_SQUARE_BRACKET   ']'
+#define CHAR_CIRCUMFLEX_ACCENT      '^'
+#define CHAR_UNDERSCORE             '_'
+#define CHAR_GRAVE_ACCENT           '`'
+#define CHAR_a                      'a'
+#define CHAR_b                      'b'
+#define CHAR_c                      'c'
+#define CHAR_d                      'd'
+#define CHAR_e                      'e'
+#define CHAR_f                      'f'
+#define CHAR_g                      'g'
+#define CHAR_h                      'h'
+#define CHAR_i                      'i'
+#define CHAR_j                      'j'
+#define CHAR_k                      'k'
+#define CHAR_l                      'l'
+#define CHAR_m                      'm'
+#define CHAR_n                      'n'
+#define CHAR_o                      'o'
+#define CHAR_p                      'p'
+#define CHAR_q                      'q'
+#define CHAR_r                      'r'
+#define CHAR_s                      's'
+#define CHAR_t                      't'
+#define CHAR_u                      'u'
+#define CHAR_v                      'v'
+#define CHAR_w                      'w'
+#define CHAR_x                      'x'
+#define CHAR_y                      'y'
+#define CHAR_z                      'z'
+#define CHAR_LEFT_CURLY_BRACKET     '{'
+#define CHAR_VERTICAL_LINE          '|'
+#define CHAR_RIGHT_CURLY_BRACKET    '}'
+#define CHAR_TILDE                  '~'
+
+#define STR_HT                      "\t"
+#define STR_VT                      "\v"
+#define STR_FF                      "\f"
+#define STR_CR                      "\r"
+#define STR_BS                      "\b"
+#define STR_BEL                     "\a"
+
+#define STR_SPACE                   " "
+#define STR_EXCLAMATION_MARK        "!"
+#define STR_QUOTATION_MARK          "\""
+#define STR_NUMBER_SIGN             "#"
+#define STR_DOLLAR_SIGN             "$"
+#define STR_PERCENT_SIGN            "%"
+#define STR_AMPERSAND               "&"
+#define STR_APOSTROPHE              "'"
+#define STR_LEFT_PARENTHESIS        "("
+#define STR_RIGHT_PARENTHESIS       ")"
+#define STR_ASTERISK                "*"
+#define STR_PLUS                    "+"
+#define STR_COMMA                   ","
+#define STR_MINUS                   "-"
+#define STR_DOT                     "."
+#define STR_SLASH                   "/"
+#define STR_0                       "0"
+#define STR_1                       "1"
+#define STR_2                       "2"
+#define STR_3                       "3"
+#define STR_4                       "4"
+#define STR_5                       "5"
+#define STR_6                       "6"
+#define STR_7                       "7"
+#define STR_8                       "8"
+#define STR_9                       "9"
+#define STR_COLON                   ":"
+#define STR_SEMICOLON               ";"
+#define STR_LESS_THAN_SIGN          "<"
+#define STR_EQUALS_SIGN             "="
+#define STR_GREATER_THAN_SIGN       ">"
+#define STR_QUESTION_MARK           "?"
+#define STR_COMMERCIAL_AT           "@"
+#define STR_A                       "A"
+#define STR_B                       "B"
+#define STR_C                       "C"
+#define STR_D                       "D"
+#define STR_E                       "E"
+#define STR_F                       "F"
+#define STR_G                       "G"
+#define STR_H                       "H"
+#define STR_I                       "I"
+#define STR_J                       "J"
+#define STR_K                       "K"
+#define STR_L                       "L"
+#define STR_M                       "M"
+#define STR_N                       "N"
+#define STR_O                       "O"
+#define STR_P                       "P"
+#define STR_Q                       "Q"
+#define STR_R                       "R"
+#define STR_S                       "S"
+#define STR_T                       "T"
+#define STR_U                       "U"
+#define STR_V                       "V"
+#define STR_W                       "W"
+#define STR_X                       "X"
+#define STR_Y                       "Y"
+#define STR_Z                       "Z"
+#define STR_LEFT_SQUARE_BRACKET     "["
+#define STR_BACKSLASH               "\\"
+#define STR_RIGHT_SQUARE_BRACKET    "]"
+#define STR_CIRCUMFLEX_ACCENT       "^"
+#define STR_UNDERSCORE              "_"
+#define STR_GRAVE_ACCENT            "`"
+#define STR_a                       "a"
+#define STR_b                       "b"
+#define STR_c                       "c"
+#define STR_d                       "d"
+#define STR_e                       "e"
+#define STR_f                       "f"
+#define STR_g                       "g"
+#define STR_h                       "h"
+#define STR_i                       "i"
+#define STR_j                       "j"
+#define STR_k                       "k"
+#define STR_l                       "l"
+#define STR_m                       "m"
+#define STR_n                       "n"
+#define STR_o                       "o"
+#define STR_p                       "p"
+#define STR_q                       "q"
+#define STR_r                       "r"
+#define STR_s                       "s"
+#define STR_t                       "t"
+#define STR_u                       "u"
+#define STR_v                       "v"
+#define STR_w                       "w"
+#define STR_x                       "x"
+#define STR_y                       "y"
+#define STR_z                       "z"
+#define STR_LEFT_CURLY_BRACKET      "{"
+#define STR_VERTICAL_LINE           "|"
+#define STR_RIGHT_CURLY_BRACKET     "}"
+#define STR_TILDE                   "~"
+
+#define STRING_ACCEPT0              "ACCEPT\0"
+#define STRING_COMMIT0              "COMMIT\0"
+#define STRING_F0                   "F\0"
+#define STRING_FAIL0                "FAIL\0"
+#define STRING_MARK0                "MARK\0"
+#define STRING_PRUNE0               "PRUNE\0"
+#define STRING_SKIP0                "SKIP\0"
+#define STRING_THEN                 "THEN"
+
+#define STRING_alpha0               "alpha\0"
+#define STRING_lower0               "lower\0"
+#define STRING_upper0               "upper\0"
+#define STRING_alnum0               "alnum\0"
+#define STRING_ascii0               "ascii\0"
+#define STRING_blank0               "blank\0"
+#define STRING_cntrl0               "cntrl\0"
+#define STRING_digit0               "digit\0"
+#define STRING_graph0               "graph\0"
+#define STRING_print0               "print\0"
+#define STRING_punct0               "punct\0"
+#define STRING_space0               "space\0"
+#define STRING_word0                "word\0"
+#define STRING_xdigit               "xdigit"
+
+#define STRING_DEFINE               "DEFINE"
+
+#define STRING_CR_RIGHTPAR             "CR)"
+#define STRING_LF_RIGHTPAR             "LF)"
+#define STRING_CRLF_RIGHTPAR           "CRLF)"
+#define STRING_ANY_RIGHTPAR            "ANY)"
+#define STRING_ANYCRLF_RIGHTPAR        "ANYCRLF)"
+#define STRING_BSR_ANYCRLF_RIGHTPAR    "BSR_ANYCRLF)"
+#define STRING_BSR_UNICODE_RIGHTPAR    "BSR_UNICODE)"
+#define STRING_UTF8_RIGHTPAR           "UTF8)"
+#define STRING_UTF16_RIGHTPAR          "UTF16)"
+#define STRING_UTF32_RIGHTPAR          "UTF32)"
+#define STRING_UTF_RIGHTPAR            "UTF)"
+#define STRING_UCP_RIGHTPAR            "UCP)"
+#define STRING_NO_START_OPT_RIGHTPAR   "NO_START_OPT)"
+#define STRING_LIMIT_MATCH_EQ          "LIMIT_MATCH="
+#define STRING_LIMIT_RECURSION_EQ      "LIMIT_RECURSION="
+
+#else  /* SUPPORT_UTF */
+
+/* UTF-8 support is enabled; always use UTF-8 (=ASCII) character codes. This
+works in both modes non-EBCDIC platforms, and on EBCDIC platforms in UTF-8 mode
+only. */
+
+#define CHAR_HT                     '\011'
+#define CHAR_VT                     '\013'
+#define CHAR_FF                     '\014'
+#define CHAR_CR                     '\015'
+#define CHAR_LF                     '\012'
+#define CHAR_NL                     CHAR_LF
+#define CHAR_NEL                    ((unsigned char)'\x85')
+#define CHAR_BS                     '\010'
+#define CHAR_BEL                    '\007'
+#define CHAR_ESC                    '\033'
+#define CHAR_DEL                    '\177'
+
+#define CHAR_NULL                   '\0'
+#define CHAR_SPACE                  '\040'
+#define CHAR_EXCLAMATION_MARK       '\041'
+#define CHAR_QUOTATION_MARK         '\042'
+#define CHAR_NUMBER_SIGN            '\043'
+#define CHAR_DOLLAR_SIGN            '\044'
+#define CHAR_PERCENT_SIGN           '\045'
+#define CHAR_AMPERSAND              '\046'
+#define CHAR_APOSTROPHE             '\047'
+#define CHAR_LEFT_PARENTHESIS       '\050'
+#define CHAR_RIGHT_PARENTHESIS      '\051'
+#define CHAR_ASTERISK               '\052'
+#define CHAR_PLUS                   '\053'
+#define CHAR_COMMA                  '\054'
+#define CHAR_MINUS                  '\055'
+#define CHAR_DOT                    '\056'
+#define CHAR_SLASH                  '\057'
+#define CHAR_0                      '\060'
+#define CHAR_1                      '\061'
+#define CHAR_2                      '\062'
+#define CHAR_3                      '\063'
+#define CHAR_4                      '\064'
+#define CHAR_5                      '\065'
+#define CHAR_6                      '\066'
+#define CHAR_7                      '\067'
+#define CHAR_8                      '\070'
+#define CHAR_9                      '\071'
+#define CHAR_COLON                  '\072'
+#define CHAR_SEMICOLON              '\073'
+#define CHAR_LESS_THAN_SIGN         '\074'
+#define CHAR_EQUALS_SIGN            '\075'
+#define CHAR_GREATER_THAN_SIGN      '\076'
+#define CHAR_QUESTION_MARK          '\077'
+#define CHAR_COMMERCIAL_AT          '\100'
+#define CHAR_A                      '\101'
+#define CHAR_B                      '\102'
+#define CHAR_C                      '\103'
+#define CHAR_D                      '\104'
+#define CHAR_E                      '\105'
+#define CHAR_F                      '\106'
+#define CHAR_G                      '\107'
+#define CHAR_H                      '\110'
+#define CHAR_I                      '\111'
+#define CHAR_J                      '\112'
+#define CHAR_K                      '\113'
+#define CHAR_L                      '\114'
+#define CHAR_M                      '\115'
+#define CHAR_N                      '\116'
+#define CHAR_O                      '\117'
+#define CHAR_P                      '\120'
+#define CHAR_Q                      '\121'
+#define CHAR_R                      '\122'
+#define CHAR_S                      '\123'
+#define CHAR_T                      '\124'
+#define CHAR_U                      '\125'
+#define CHAR_V                      '\126'
+#define CHAR_W                      '\127'
+#define CHAR_X                      '\130'
+#define CHAR_Y                      '\131'
+#define CHAR_Z                      '\132'
+#define CHAR_LEFT_SQUARE_BRACKET    '\133'
+#define CHAR_BACKSLASH              '\134'
+#define CHAR_RIGHT_SQUARE_BRACKET   '\135'
+#define CHAR_CIRCUMFLEX_ACCENT      '\136'
+#define CHAR_UNDERSCORE             '\137'
+#define CHAR_GRAVE_ACCENT           '\140'
+#define CHAR_a                      '\141'
+#define CHAR_b                      '\142'
+#define CHAR_c                      '\143'
+#define CHAR_d                      '\144'
+#define CHAR_e                      '\145'
+#define CHAR_f                      '\146'
+#define CHAR_g                      '\147'
+#define CHAR_h                      '\150'
+#define CHAR_i                      '\151'
+#define CHAR_j                      '\152'
+#define CHAR_k                      '\153'
+#define CHAR_l                      '\154'
+#define CHAR_m                      '\155'
+#define CHAR_n                      '\156'
+#define CHAR_o                      '\157'
+#define CHAR_p                      '\160'
+#define CHAR_q                      '\161'
+#define CHAR_r                      '\162'
+#define CHAR_s                      '\163'
+#define CHAR_t                      '\164'
+#define CHAR_u                      '\165'
+#define CHAR_v                      '\166'
+#define CHAR_w                      '\167'
+#define CHAR_x                      '\170'
+#define CHAR_y                      '\171'
+#define CHAR_z                      '\172'
+#define CHAR_LEFT_CURLY_BRACKET     '\173'
+#define CHAR_VERTICAL_LINE          '\174'
+#define CHAR_RIGHT_CURLY_BRACKET    '\175'
+#define CHAR_TILDE                  '\176'
+
+#define STR_HT                      "\011"
+#define STR_VT                      "\013"
+#define STR_FF                      "\014"
+#define STR_CR                      "\015"
+#define STR_NL                      "\012"
+#define STR_BS                      "\010"
+#define STR_BEL                     "\007"
+#define STR_ESC                     "\033"
+#define STR_DEL                     "\177"
+
+#define STR_SPACE                   "\040"
+#define STR_EXCLAMATION_MARK        "\041"
+#define STR_QUOTATION_MARK          "\042"
+#define STR_NUMBER_SIGN             "\043"
+#define STR_DOLLAR_SIGN             "\044"
+#define STR_PERCENT_SIGN            "\045"
+#define STR_AMPERSAND               "\046"
+#define STR_APOSTROPHE              "\047"
+#define STR_LEFT_PARENTHESIS        "\050"
+#define STR_RIGHT_PARENTHESIS       "\051"
+#define STR_ASTERISK                "\052"
+#define STR_PLUS                    "\053"
+#define STR_COMMA                   "\054"
+#define STR_MINUS                   "\055"
+#define STR_DOT                     "\056"
+#define STR_SLASH                   "\057"
+#define STR_0                       "\060"
+#define STR_1                       "\061"
+#define STR_2                       "\062"
+#define STR_3                       "\063"
+#define STR_4                       "\064"
+#define STR_5                       "\065"
+#define STR_6                       "\066"
+#define STR_7                       "\067"
+#define STR_8                       "\070"
+#define STR_9                       "\071"
+#define STR_COLON                   "\072"
+#define STR_SEMICOLON               "\073"
+#define STR_LESS_THAN_SIGN          "\074"
+#define STR_EQUALS_SIGN             "\075"
+#define STR_GREATER_THAN_SIGN       "\076"
+#define STR_QUESTION_MARK           "\077"
+#define STR_COMMERCIAL_AT           "\100"
+#define STR_A                       "\101"
+#define STR_B                       "\102"
+#define STR_C                       "\103"
+#define STR_D                       "\104"
+#define STR_E                       "\105"
+#define STR_F                       "\106"
+#define STR_G                       "\107"
+#define STR_H                       "\110"
+#define STR_I                       "\111"
+#define STR_J                       "\112"
+#define STR_K                       "\113"
+#define STR_L                       "\114"
+#define STR_M                       "\115"
+#define STR_N                       "\116"
+#define STR_O                       "\117"
+#define STR_P                       "\120"
+#define STR_Q                       "\121"
+#define STR_R                       "\122"
+#define STR_S                       "\123"
+#define STR_T                       "\124"
+#define STR_U                       "\125"
+#define STR_V                       "\126"
+#define STR_W                       "\127"
+#define STR_X                       "\130"
+#define STR_Y                       "\131"
+#define STR_Z                       "\132"
+#define STR_LEFT_SQUARE_BRACKET     "\133"
+#define STR_BACKSLASH               "\134"
+#define STR_RIGHT_SQUARE_BRACKET    "\135"
+#define STR_CIRCUMFLEX_ACCENT       "\136"
+#define STR_UNDERSCORE              "\137"
+#define STR_GRAVE_ACCENT            "\140"
+#define STR_a                       "\141"
+#define STR_b                       "\142"
+#define STR_c                       "\143"
+#define STR_d                       "\144"
+#define STR_e                       "\145"
+#define STR_f                       "\146"
+#define STR_g                       "\147"
+#define STR_h                       "\150"
+#define STR_i                       "\151"
+#define STR_j                       "\152"
+#define STR_k                       "\153"
+#define STR_l                       "\154"
+#define STR_m                       "\155"
+#define STR_n                       "\156"
+#define STR_o                       "\157"
+#define STR_p                       "\160"
+#define STR_q                       "\161"
+#define STR_r                       "\162"
+#define STR_s                       "\163"
+#define STR_t                       "\164"
+#define STR_u                       "\165"
+#define STR_v                       "\166"
+#define STR_w                       "\167"
+#define STR_x                       "\170"
+#define STR_y                       "\171"
+#define STR_z                       "\172"
+#define STR_LEFT_CURLY_BRACKET      "\173"
+#define STR_VERTICAL_LINE           "\174"
+#define STR_RIGHT_CURLY_BRACKET     "\175"
+#define STR_TILDE                   "\176"
+
+#define STRING_ACCEPT0              STR_A STR_C STR_C STR_E STR_P STR_T "\0"
+#define STRING_COMMIT0              STR_C STR_O STR_M STR_M STR_I STR_T "\0"
+#define STRING_F0                   STR_F "\0"
+#define STRING_FAIL0                STR_F STR_A STR_I STR_L "\0"
+#define STRING_MARK0                STR_M STR_A STR_R STR_K "\0"
+#define STRING_PRUNE0               STR_P STR_R STR_U STR_N STR_E "\0"
+#define STRING_SKIP0                STR_S STR_K STR_I STR_P "\0"
+#define STRING_THEN                 STR_T STR_H STR_E STR_N
+
+#define STRING_alpha0               STR_a STR_l STR_p STR_h STR_a "\0"
+#define STRING_lower0               STR_l STR_o STR_w STR_e STR_r "\0"
+#define STRING_upper0               STR_u STR_p STR_p STR_e STR_r "\0"
+#define STRING_alnum0               STR_a STR_l STR_n STR_u STR_m "\0"
+#define STRING_ascii0               STR_a STR_s STR_c STR_i STR_i "\0"
+#define STRING_blank0               STR_b STR_l STR_a STR_n STR_k "\0"
+#define STRING_cntrl0               STR_c STR_n STR_t STR_r STR_l "\0"
+#define STRING_digit0               STR_d STR_i STR_g STR_i STR_t "\0"
+#define STRING_graph0               STR_g STR_r STR_a STR_p STR_h "\0"
+#define STRING_print0               STR_p STR_r STR_i STR_n STR_t "\0"
+#define STRING_punct0               STR_p STR_u STR_n STR_c STR_t "\0"
+#define STRING_space0               STR_s STR_p STR_a STR_c STR_e "\0"
+#define STRING_word0                STR_w STR_o STR_r STR_d       "\0"
+#define STRING_xdigit               STR_x STR_d STR_i STR_g STR_i STR_t
+
+#define STRING_DEFINE               STR_D STR_E STR_F STR_I STR_N STR_E
+
+#define STRING_CR_RIGHTPAR             STR_C STR_R STR_RIGHT_PARENTHESIS
+#define STRING_LF_RIGHTPAR             STR_L STR_F STR_RIGHT_PARENTHESIS
+#define STRING_CRLF_RIGHTPAR           STR_C STR_R STR_L STR_F STR_RIGHT_PARENTHESIS
+#define STRING_ANY_RIGHTPAR            STR_A STR_N STR_Y STR_RIGHT_PARENTHESIS
+#define STRING_ANYCRLF_RIGHTPAR        STR_A STR_N STR_Y STR_C STR_R STR_L STR_F STR_RIGHT_PARENTHESIS
+#define STRING_BSR_ANYCRLF_RIGHTPAR    STR_B STR_S STR_R STR_UNDERSCORE STR_A STR_N STR_Y STR_C STR_R STR_L STR_F STR_RIGHT_PARENTHESIS
+#define STRING_BSR_UNICODE_RIGHTPAR    STR_B STR_S STR_R STR_UNDERSCORE STR_U STR_N STR_I STR_C STR_O STR_D STR_E STR_RIGHT_PARENTHESIS
+#define STRING_UTF8_RIGHTPAR           STR_U STR_T STR_F STR_8 STR_RIGHT_PARENTHESIS
+#define STRING_UTF16_RIGHTPAR          STR_U STR_T STR_F STR_1 STR_6 STR_RIGHT_PARENTHESIS
+#define STRING_UTF32_RIGHTPAR          STR_U STR_T STR_F STR_3 STR_2 STR_RIGHT_PARENTHESIS
+#define STRING_UTF_RIGHTPAR            STR_U STR_T STR_F STR_RIGHT_PARENTHESIS
+#define STRING_UCP_RIGHTPAR            STR_U STR_C STR_P STR_RIGHT_PARENTHESIS
+#define STRING_NO_START_OPT_RIGHTPAR   STR_N STR_O STR_UNDERSCORE STR_S STR_T STR_A STR_R STR_T STR_UNDERSCORE STR_O STR_P STR_T STR_RIGHT_PARENTHESIS
+#define STRING_LIMIT_MATCH_EQ          STR_L STR_I STR_M STR_I STR_T STR_UNDERSCORE STR_M STR_A STR_T STR_C STR_H STR_EQUALS_SIGN
+#define STRING_LIMIT_RECURSION_EQ      STR_L STR_I STR_M STR_I STR_T STR_UNDERSCORE STR_R STR_E STR_C STR_U STR_R STR_S STR_I STR_O STR_N STR_EQUALS_SIGN
+
+#endif  /* SUPPORT_UTF */
+
+/* Escape items that are just an encoding of a particular data value. */
+
+#ifndef ESC_e
+#define ESC_e CHAR_ESC
+#endif
+
+#ifndef ESC_f
+#define ESC_f CHAR_FF
+#endif
+
+#ifndef ESC_n
+#define ESC_n CHAR_LF
+#endif
+
+#ifndef ESC_r
+#define ESC_r CHAR_CR
+#endif
+
+/* We can't officially use ESC_t because it is a POSIX reserved identifier
+(presumably because of all the others like size_t). */
+
+#ifndef ESC_tee
+#define ESC_tee CHAR_HT
+#endif
+
+/* Codes for different types of Unicode property */
+
+#define PT_ANY        0    /* Any property - matches all chars */
+#define PT_LAMP       1    /* L& - the union of Lu, Ll, Lt */
+#define PT_GC         2    /* Specified general characteristic (e.g. L) */
+#define PT_PC         3    /* Specified particular characteristic (e.g. Lu) */
+#define PT_SC         4    /* Script (e.g. Han) */
+#define PT_ALNUM      5    /* Alphanumeric - the union of L and N */
+#define PT_SPACE      6    /* Perl space - Z plus 9,10,12,13 */
+#define PT_PXSPACE    7    /* POSIX space - Z plus 9,10,11,12,13 */
+#define PT_WORD       8    /* Word - L plus N plus underscore */
+#define PT_CLIST      9    /* Pseudo-property: match character list */
+#define PT_UCNC      10    /* Universal Character nameable character */
+
+/* Flag bits and data types for the extended class (OP_XCLASS) for classes that
+contain characters with values greater than 255. */
+
+#define XCL_NOT    0x01    /* Flag: this is a negative class */
+#define XCL_MAP    0x02    /* Flag: a 32-byte map is present */
+
+#define XCL_END       0    /* Marks end of individual items */
+#define XCL_SINGLE    1    /* Single item (one multibyte char) follows */
+#define XCL_RANGE     2    /* A range (two multibyte chars) follows */
+#define XCL_PROP      3    /* Unicode property (2-byte property code follows) */
+#define XCL_NOTPROP   4    /* Unicode inverted property (ditto) */
+
+/* These are escaped items that aren't just an encoding of a particular data
+value such as \n. They must have non-zero values, as check_escape() returns
+0 for a data character.  Also, they must appear in the same order as in the opcode
+definitions below, up to ESC_z. There's a dummy for OP_ALLANY because it
+corresponds to "." in DOTALL mode rather than an escape sequence. It is also
+used for [^] in JavaScript compatibility mode, and for \C in non-utf mode. In
+non-DOTALL mode, "." behaves like \N.
+
+The special values ESC_DU, ESC_du, etc. are used instead of ESC_D, ESC_d, etc.
+when PCRE_UCP is set and replacement of \d etc by \p sequences is required.
+They must be contiguous, and remain in order so that the replacements can be
+looked up from a table.
+
+Negative numbers are used to encode a backreference (\1, \2, \3, etc.) in
+check_escape(). There are two tests in the code for an escape
+greater than ESC_b and less than ESC_Z to detect the types that may be
+repeated. These are the types that consume characters. If any new escapes are
+put in between that don't consume a character, that code will have to change.
+*/
+
+enum { ESC_A = 1, ESC_G, ESC_K, ESC_B, ESC_b, ESC_D, ESC_d, ESC_S, ESC_s,
+       ESC_W, ESC_w, ESC_N, ESC_dum, ESC_C, ESC_P, ESC_p, ESC_R, ESC_H,
+       ESC_h, ESC_V, ESC_v, ESC_X, ESC_Z, ESC_z,
+       ESC_E, ESC_Q, ESC_g, ESC_k,
+       ESC_DU, ESC_du, ESC_SU, ESC_su, ESC_WU, ESC_wu };
+
+/* Opcode table: Starting from 1 (i.e. after OP_END), the values up to
+OP_EOD must correspond in order to the list of escapes immediately above.
+
+*** NOTE NOTE NOTE *** Whenever this list is updated, the two macro definitions
+that follow must also be updated to match. There are also tables called
+"coptable" and "poptable" in pcre_dfa_exec.c that must be updated. */
+
+enum {
+  OP_END,            /* 0 End of pattern */
+
+  /* Values corresponding to backslashed metacharacters */
+
+  OP_SOD,            /* 1 Start of data: \A */
+  OP_SOM,            /* 2 Start of match (subject + offset): \G */
+  OP_SET_SOM,        /* 3 Set start of match (\K) */
+  OP_NOT_WORD_BOUNDARY,  /*  4 \B */
+  OP_WORD_BOUNDARY,      /*  5 \b */
+  OP_NOT_DIGIT,          /*  6 \D */
+  OP_DIGIT,              /*  7 \d */
+  OP_NOT_WHITESPACE,     /*  8 \S */
+  OP_WHITESPACE,         /*  9 \s */
+  OP_NOT_WORDCHAR,       /* 10 \W */
+  OP_WORDCHAR,           /* 11 \w */
+
+  OP_ANY,            /* 12 Match any character except newline (\N) */
+  OP_ALLANY,         /* 13 Match any character */
+  OP_ANYBYTE,        /* 14 Match any byte (\C); different to OP_ANY for UTF-8 */
+  OP_NOTPROP,        /* 15 \P (not Unicode property) */
+  OP_PROP,           /* 16 \p (Unicode property) */
+  OP_ANYNL,          /* 17 \R (any newline sequence) */
+  OP_NOT_HSPACE,     /* 18 \H (not horizontal whitespace) */
+  OP_HSPACE,         /* 19 \h (horizontal whitespace) */
+  OP_NOT_VSPACE,     /* 20 \V (not vertical whitespace) */
+  OP_VSPACE,         /* 21 \v (vertical whitespace) */
+  OP_EXTUNI,         /* 22 \X (extended Unicode sequence */
+  OP_EODN,           /* 23 End of data or \n at end of data (\Z) */
+  OP_EOD,            /* 24 End of data (\z) */
+
+  OP_CIRC,           /* 25 Start of line - not multiline */
+  OP_CIRCM,          /* 26 Start of line - multiline */
+  OP_DOLL,           /* 27 End of line - not multiline */
+  OP_DOLLM,          /* 28 End of line - multiline */
+  OP_CHAR,           /* 29 Match one character, casefully */
+  OP_CHARI,          /* 30 Match one character, caselessly */
+  OP_NOT,            /* 31 Match one character, not the given one, casefully */
+  OP_NOTI,           /* 32 Match one character, not the given one, caselessly */
+
+  /* The following sets of 13 opcodes must always be kept in step because
+  the offset from the first one is used to generate the others. */
+
+  /**** Single characters, caseful, must precede the caseless ones ****/
+
+  OP_STAR,           /* 33 The maximizing and minimizing versions of */
+  OP_MINSTAR,        /* 34 these six opcodes must come in pairs, with */
+  OP_PLUS,           /* 35 the minimizing one second. */
+  OP_MINPLUS,        /* 36 */
+  OP_QUERY,          /* 37 */
+  OP_MINQUERY,       /* 38 */
+
+  OP_UPTO,           /* 39 From 0 to n matches of one character, caseful*/
+  OP_MINUPTO,        /* 40 */
+  OP_EXACT,          /* 41 Exactly n matches */
+
+  OP_POSSTAR,        /* 42 Possessified star, caseful */
+  OP_POSPLUS,        /* 43 Possessified plus, caseful */
+  OP_POSQUERY,       /* 44 Posesssified query, caseful */
+  OP_POSUPTO,        /* 45 Possessified upto, caseful */
+
+  /**** Single characters, caseless, must follow the caseful ones */
+
+  OP_STARI,          /* 46 */
+  OP_MINSTARI,       /* 47 */
+  OP_PLUSI,          /* 48 */
+  OP_MINPLUSI,       /* 49 */
+  OP_QUERYI,         /* 50 */
+  OP_MINQUERYI,      /* 51 */
+
+  OP_UPTOI,          /* 52 From 0 to n matches of one character, caseless */
+  OP_MINUPTOI,       /* 53 */
+  OP_EXACTI,         /* 54 */
+
+  OP_POSSTARI,       /* 55 Possessified star, caseless */
+  OP_POSPLUSI,       /* 56 Possessified plus, caseless */
+  OP_POSQUERYI,      /* 57 Posesssified query, caseless */
+  OP_POSUPTOI,       /* 58 Possessified upto, caseless */
+
+  /**** The negated ones must follow the non-negated ones, and match them ****/
+  /**** Negated single character, caseful; must precede the caseless ones ****/
+
+  OP_NOTSTAR,        /* 59 The maximizing and minimizing versions of */
+  OP_NOTMINSTAR,     /* 60 these six opcodes must come in pairs, with */
+  OP_NOTPLUS,        /* 61 the minimizing one second. They must be in */
+  OP_NOTMINPLUS,     /* 62 exactly the same order as those above. */
+  OP_NOTQUERY,       /* 63 */
+  OP_NOTMINQUERY,    /* 64 */
+
+  OP_NOTUPTO,        /* 65 From 0 to n matches, caseful */
+  OP_NOTMINUPTO,     /* 66 */
+  OP_NOTEXACT,       /* 67 Exactly n matches */
+
+  OP_NOTPOSSTAR,     /* 68 Possessified versions, caseful */
+  OP_NOTPOSPLUS,     /* 69 */
+  OP_NOTPOSQUERY,    /* 70 */
+  OP_NOTPOSUPTO,     /* 71 */
+
+  /**** Negated single character, caseless; must follow the caseful ones ****/
+
+  OP_NOTSTARI,       /* 72 */
+  OP_NOTMINSTARI,    /* 73 */
+  OP_NOTPLUSI,       /* 74 */
+  OP_NOTMINPLUSI,    /* 75 */
+  OP_NOTQUERYI,      /* 76 */
+  OP_NOTMINQUERYI,   /* 77 */
+
+  OP_NOTUPTOI,       /* 78 From 0 to n matches, caseless */
+  OP_NOTMINUPTOI,    /* 79 */
+  OP_NOTEXACTI,      /* 80 Exactly n matches */
+
+  OP_NOTPOSSTARI,    /* 81 Possessified versions, caseless */
+  OP_NOTPOSPLUSI,    /* 82 */
+  OP_NOTPOSQUERYI,   /* 83 */
+  OP_NOTPOSUPTOI,    /* 84 */
+
+  /**** Character types ****/
+
+  OP_TYPESTAR,       /* 85 The maximizing and minimizing versions of */
+  OP_TYPEMINSTAR,    /* 86 these six opcodes must come in pairs, with */
+  OP_TYPEPLUS,       /* 87 the minimizing one second. These codes must */
+  OP_TYPEMINPLUS,    /* 88 be in exactly the same order as those above. */
+  OP_TYPEQUERY,      /* 89 */
+  OP_TYPEMINQUERY,   /* 90 */
+
+  OP_TYPEUPTO,       /* 91 From 0 to n matches */
+  OP_TYPEMINUPTO,    /* 92 */
+  OP_TYPEEXACT,      /* 93 Exactly n matches */
+
+  OP_TYPEPOSSTAR,    /* 94 Possessified versions */
+  OP_TYPEPOSPLUS,    /* 95 */
+  OP_TYPEPOSQUERY,   /* 96 */
+  OP_TYPEPOSUPTO,    /* 97 */
+
+  /* These are used for character classes and back references; only the
+  first six are the same as the sets above. */
+
+  OP_CRSTAR,         /* 98 The maximizing and minimizing versions of */
+  OP_CRMINSTAR,      /* 99 all these opcodes must come in pairs, with */
+  OP_CRPLUS,         /* 100 the minimizing one second. These codes must */
+  OP_CRMINPLUS,      /* 101 be in exactly the same order as those above. */
+  OP_CRQUERY,        /* 102 */
+  OP_CRMINQUERY,     /* 103 */
+
+  OP_CRRANGE,        /* 104 These are different to the three sets above. */
+  OP_CRMINRANGE,     /* 105 */
+
+  /* End of quantifier opcodes */
+
+  OP_CLASS,          /* 106 Match a character class, chars < 256 only */
+  OP_NCLASS,         /* 107 Same, but the bitmap was created from a negative
+                              class - the difference is relevant only when a
+                              character > 255 is encountered. */
+  OP_XCLASS,         /* 108 Extended class for handling > 255 chars within the
+                              class. This does both positive and negative. */
+  OP_REF,            /* 109 Match a back reference, casefully */
+  OP_REFI,           /* 110 Match a back reference, caselessly */
+  OP_RECURSE,        /* 111 Match a numbered subpattern (possibly recursive) */
+  OP_CALLOUT,        /* 112 Call out to external function if provided */
+
+  OP_ALT,            /* 113 Start of alternation */
+  OP_KET,            /* 114 End of group that doesn't have an unbounded repeat */
+  OP_KETRMAX,        /* 115 These two must remain together and in this */
+  OP_KETRMIN,        /* 116 order. They are for groups the repeat for ever. */
+  OP_KETRPOS,        /* 117 Possessive unlimited repeat. */
+
+  /* The assertions must come before BRA, CBRA, ONCE, and COND, and the four
+  asserts must remain in order. */
+
+  OP_REVERSE,        /* 118 Move pointer back - used in lookbehind assertions */
+  OP_ASSERT,         /* 119 Positive lookahead */
+  OP_ASSERT_NOT,     /* 120 Negative lookahead */
+  OP_ASSERTBACK,     /* 121 Positive lookbehind */
+  OP_ASSERTBACK_NOT, /* 122 Negative lookbehind */
+
+  /* ONCE, ONCE_NC, BRA, BRAPOS, CBRA, CBRAPOS, and COND must come immediately
+  after the assertions, with ONCE first, as there's a test for >= ONCE for a
+  subpattern that isn't an assertion. The POS versions must immediately follow
+  the non-POS versions in each case. */
+
+  OP_ONCE,           /* 123 Atomic group, contains captures */
+  OP_ONCE_NC,        /* 124 Atomic group containing no captures */
+  OP_BRA,            /* 125 Start of non-capturing bracket */
+  OP_BRAPOS,         /* 126 Ditto, with unlimited, possessive repeat */
+  OP_CBRA,           /* 127 Start of capturing bracket */
+  OP_CBRAPOS,        /* 128 Ditto, with unlimited, possessive repeat */
+  OP_COND,           /* 129 Conditional group */
+
+  /* These five must follow the previous five, in the same order. There's a
+  check for >= SBRA to distinguish the two sets. */
+
+  OP_SBRA,           /* 130 Start of non-capturing bracket, check empty  */
+  OP_SBRAPOS,        /* 131 Ditto, with unlimited, possessive repeat */
+  OP_SCBRA,          /* 132 Start of capturing bracket, check empty */
+  OP_SCBRAPOS,       /* 133 Ditto, with unlimited, possessive repeat */
+  OP_SCOND,          /* 134 Conditional group, check empty */
+
+  /* The next two pairs must (respectively) be kept together. */
+
+  OP_CREF,           /* 135 Used to hold a capture number as condition */
+  OP_NCREF,          /* 136 Same, but generated by a name reference*/
+  OP_RREF,           /* 137 Used to hold a recursion number as condition */
+  OP_NRREF,          /* 138 Same, but generated by a name reference*/
+  OP_DEF,            /* 139 The DEFINE condition */
+
+  OP_BRAZERO,        /* 140 These two must remain together and in this */
+  OP_BRAMINZERO,     /* 141 order. */
+  OP_BRAPOSZERO,     /* 142 */
+
+  /* These are backtracking control verbs */
+
+  OP_MARK,           /* 143 always has an argument */
+  OP_PRUNE,          /* 144 */
+  OP_PRUNE_ARG,      /* 145 same, but with argument */
+  OP_SKIP,           /* 146 */
+  OP_SKIP_ARG,       /* 147 same, but with argument */
+  OP_THEN,           /* 148 */
+  OP_THEN_ARG,       /* 149 same, but with argument */
+  OP_COMMIT,         /* 150 */
+
+  /* These are forced failure and success verbs */
+
+  OP_FAIL,           /* 151 */
+  OP_ACCEPT,         /* 152 */
+  OP_ASSERT_ACCEPT,  /* 153 Used inside assertions */
+  OP_CLOSE,          /* 154 Used before OP_ACCEPT to close open captures */
+
+  /* This is used to skip a subpattern with a {0} quantifier */
+
+  OP_SKIPZERO,       /* 155 */
+
+  /* This is not an opcode, but is used to check that tables indexed by opcode
+  are the correct length, in order to catch updating errors - there have been
+  some in the past. */
+
+  OP_TABLE_LENGTH
+};
+
+/* *** NOTE NOTE NOTE *** Whenever the list above is updated, the two macro
+definitions that follow must also be updated to match. There are also tables
+called "coptable" and "poptable" in pcre_dfa_exec.c that must be updated. */
+
+
+/* This macro defines textual names for all the opcodes. These are used only
+for debugging, and some of them are only partial names. The macro is referenced
+only in pcre_printint.c, which fills out the full names in many cases (and in
+some cases doesn't actually use these names at all). */
+
+#define OP_NAME_LIST \
+  "End", "\\A", "\\G", "\\K", "\\B", "\\b", "\\D", "\\d",         \
+  "\\S", "\\s", "\\W", "\\w", "Any", "AllAny", "Anybyte",         \
+  "notprop", "prop", "\\R", "\\H", "\\h", "\\V", "\\v",           \
+  "extuni",  "\\Z", "\\z",                                        \
+  "^", "^", "$", "$", "char", "chari", "not", "noti",             \
+  "*", "*?", "+", "+?", "?", "??",                                \
+  "{", "{", "{",                                                  \
+  "*+","++", "?+", "{",                                           \
+  "*", "*?", "+", "+?", "?", "??",                                \
+  "{", "{", "{",                                                  \
+  "*+","++", "?+", "{",                                           \
+  "*", "*?", "+", "+?", "?", "??",                                \
+  "{", "{", "{",                                                  \
+  "*+","++", "?+", "{",                                           \
+  "*", "*?", "+", "+?", "?", "??",                                \
+  "{", "{", "{",                                                  \
+  "*+","++", "?+", "{",                                           \
+  "*", "*?", "+", "+?", "?", "??", "{", "{", "{",                 \
+  "*+","++", "?+", "{",                                           \
+  "*", "*?", "+", "+?", "?", "??", "{", "{",                      \
+  "class", "nclass", "xclass", "Ref", "Refi",                     \
+  "Recurse", "Callout",                                           \
+  "Alt", "Ket", "KetRmax", "KetRmin", "KetRpos",                  \
+  "Reverse", "Assert", "Assert not", "AssertB", "AssertB not",    \
+  "Once", "Once_NC",                                              \
+  "Bra", "BraPos", "CBra", "CBraPos",                             \
+  "Cond",                                                         \
+  "SBra", "SBraPos", "SCBra", "SCBraPos",                         \
+  "SCond",                                                        \
+  "Cond ref", "Cond nref", "Cond rec", "Cond nrec", "Cond def",   \
+  "Brazero", "Braminzero", "Braposzero",                          \
+  "*MARK", "*PRUNE", "*PRUNE", "*SKIP", "*SKIP",                  \
+  "*THEN", "*THEN", "*COMMIT", "*FAIL",                           \
+  "*ACCEPT", "*ASSERT_ACCEPT",                                    \
+  "Close", "Skip zero"
+
+
+/* This macro defines the length of fixed length operations in the compiled
+regex. The lengths are used when searching for specific things, and also in the
+debugging printing of a compiled regex. We use a macro so that it can be
+defined close to the definitions of the opcodes themselves.
+
+As things have been extended, some of these are no longer fixed lenths, but are
+minima instead. For example, the length of a single-character repeat may vary
+in UTF-8 mode. The code that uses this table must know about such things. */
+
+#define OP_LENGTHS \
+  1,                             /* End                                    */ \
+  1, 1, 1, 1, 1,                 /* \A, \G, \K, \B, \b                     */ \
+  1, 1, 1, 1, 1, 1,              /* \D, \d, \S, \s, \W, \w                 */ \
+  1, 1, 1,                       /* Any, AllAny, Anybyte                   */ \
+  3, 3,                          /* \P, \p                                 */ \
+  1, 1, 1, 1, 1,                 /* \R, \H, \h, \V, \v                     */ \
+  1,                             /* \X                                     */ \
+  1, 1, 1, 1, 1, 1,              /* \Z, \z, ^, ^M, $, $M                   */ \
+  2,                             /* Char  - the minimum length             */ \
+  2,                             /* Chari  - the minimum length            */ \
+  2,                             /* not                                    */ \
+  2,                             /* noti                                   */ \
+  /* Positive single-char repeats                             ** These are */ \
+  2, 2, 2, 2, 2, 2,              /* *, *?, +, +?, ?, ??       ** minima in */ \
+  2+IMM2_SIZE, 2+IMM2_SIZE,      /* upto, minupto             ** mode      */ \
+  2+IMM2_SIZE,                   /* exact                                  */ \
+  2, 2, 2, 2+IMM2_SIZE,          /* *+, ++, ?+, upto+                      */ \
+  2, 2, 2, 2, 2, 2,              /* *I, *?I, +I, +?I, ?I, ??I ** UTF-8     */ \
+  2+IMM2_SIZE, 2+IMM2_SIZE,      /* upto I, minupto I                      */ \
+  2+IMM2_SIZE,                   /* exact I                                */ \
+  2, 2, 2, 2+IMM2_SIZE,          /* *+I, ++I, ?+I, upto+I                  */ \
+  /* Negative single-char repeats - only for chars < 256                   */ \
+  2, 2, 2, 2, 2, 2,              /* NOT *, *?, +, +?, ?, ??                */ \
+  2+IMM2_SIZE, 2+IMM2_SIZE,      /* NOT upto, minupto                      */ \
+  2+IMM2_SIZE,                   /* NOT exact                              */ \
+  2, 2, 2, 2+IMM2_SIZE,          /* Possessive NOT *, +, ?, upto           */ \
+  2, 2, 2, 2, 2, 2,              /* NOT *I, *?I, +I, +?I, ?I, ??I          */ \
+  2+IMM2_SIZE, 2+IMM2_SIZE,      /* NOT upto I, minupto I                  */ \
+  2+IMM2_SIZE,                   /* NOT exact I                            */ \
+  2, 2, 2, 2+IMM2_SIZE,          /* Possessive NOT *I, +I, ?I, upto I      */ \
+  /* Positive type repeats                                                 */ \
+  2, 2, 2, 2, 2, 2,              /* Type *, *?, +, +?, ?, ??               */ \
+  2+IMM2_SIZE, 2+IMM2_SIZE,      /* Type upto, minupto                     */ \
+  2+IMM2_SIZE,                   /* Type exact                             */ \
+  2, 2, 2, 2+IMM2_SIZE,          /* Possessive *+, ++, ?+, upto+           */ \
+  /* Character class & ref repeats                                         */ \
+  1, 1, 1, 1, 1, 1,              /* *, *?, +, +?, ?, ??                    */ \
+  1+2*IMM2_SIZE, 1+2*IMM2_SIZE,  /* CRRANGE, CRMINRANGE                    */ \
+  1+(32/sizeof(pcre_uchar)),     /* CLASS                                  */ \
+  1+(32/sizeof(pcre_uchar)),     /* NCLASS                                 */ \
+  0,                             /* XCLASS - variable length               */ \
+  1+IMM2_SIZE,                   /* REF                                    */ \
+  1+IMM2_SIZE,                   /* REFI                                   */ \
+  1+LINK_SIZE,                   /* RECURSE                                */ \
+  2+2*LINK_SIZE,                 /* CALLOUT                                */ \
+  1+LINK_SIZE,                   /* Alt                                    */ \
+  1+LINK_SIZE,                   /* Ket                                    */ \
+  1+LINK_SIZE,                   /* KetRmax                                */ \
+  1+LINK_SIZE,                   /* KetRmin                                */ \
+  1+LINK_SIZE,                   /* KetRpos                                */ \
+  1+LINK_SIZE,                   /* Reverse                                */ \
+  1+LINK_SIZE,                   /* Assert                                 */ \
+  1+LINK_SIZE,                   /* Assert not                             */ \
+  1+LINK_SIZE,                   /* Assert behind                          */ \
+  1+LINK_SIZE,                   /* Assert behind not                      */ \
+  1+LINK_SIZE,                   /* ONCE                                   */ \
+  1+LINK_SIZE,                   /* ONCE_NC                                */ \
+  1+LINK_SIZE,                   /* BRA                                    */ \
+  1+LINK_SIZE,                   /* BRAPOS                                 */ \
+  1+LINK_SIZE+IMM2_SIZE,         /* CBRA                                   */ \
+  1+LINK_SIZE+IMM2_SIZE,         /* CBRAPOS                                */ \
+  1+LINK_SIZE,                   /* COND                                   */ \
+  1+LINK_SIZE,                   /* SBRA                                   */ \
+  1+LINK_SIZE,                   /* SBRAPOS                                */ \
+  1+LINK_SIZE+IMM2_SIZE,         /* SCBRA                                  */ \
+  1+LINK_SIZE+IMM2_SIZE,         /* SCBRAPOS                               */ \
+  1+LINK_SIZE,                   /* SCOND                                  */ \
+  1+IMM2_SIZE, 1+IMM2_SIZE,      /* CREF, NCREF                            */ \
+  1+IMM2_SIZE, 1+IMM2_SIZE,      /* RREF, NRREF                            */ \
+  1,                             /* DEF                                    */ \
+  1, 1, 1,                       /* BRAZERO, BRAMINZERO, BRAPOSZERO        */ \
+  3, 1, 3,                       /* MARK, PRUNE, PRUNE_ARG                 */ \
+  1, 3,                          /* SKIP, SKIP_ARG                         */ \
+  1, 3,                          /* THEN, THEN_ARG                         */ \
+  1, 1, 1, 1,                    /* COMMIT, FAIL, ACCEPT, ASSERT_ACCEPT    */ \
+  1+IMM2_SIZE, 1                 /* CLOSE, SKIPZERO                        */
+
+/* A magic value for OP_RREF and OP_NRREF to indicate the "any recursion"
+condition. */
+
+#define RREF_ANY  0xffff
+
+/* Compile time error code numbers. They are given names so that they can more
+easily be tracked. When a new number is added, the table called eint in
+pcreposix.c must be updated. */
+
+enum { ERR0,  ERR1,  ERR2,  ERR3,  ERR4,  ERR5,  ERR6,  ERR7,  ERR8,  ERR9,
+       ERR10, ERR11, ERR12, ERR13, ERR14, ERR15, ERR16, ERR17, ERR18, ERR19,
+       ERR20, ERR21, ERR22, ERR23, ERR24, ERR25, ERR26, ERR27, ERR28, ERR29,
+       ERR30, ERR31, ERR32, ERR33, ERR34, ERR35, ERR36, ERR37, ERR38, ERR39,
+       ERR40, ERR41, ERR42, ERR43, ERR44, ERR45, ERR46, ERR47, ERR48, ERR49,
+       ERR50, ERR51, ERR52, ERR53, ERR54, ERR55, ERR56, ERR57, ERR58, ERR59,
+       ERR60, ERR61, ERR62, ERR63, ERR64, ERR65, ERR66, ERR67, ERR68, ERR69,
+       ERR70, ERR71, ERR72, ERR73, ERR74, ERR75, ERR76, ERR77, ERR78, ERRCOUNT };
+
+/* JIT compiling modes. The function list is indexed by them. */
+enum { JIT_COMPILE, JIT_PARTIAL_SOFT_COMPILE, JIT_PARTIAL_HARD_COMPILE,
+       JIT_NUMBER_OF_COMPILE_MODES };
+
+/* The real format of the start of the pcre block; the index of names and the
+code vector run on as long as necessary after the end. We store an explicit
+offset to the name table so that if a regex is compiled on one host, saved, and
+then run on another where the size of pointers is different, all might still
+be well.
+
+The size of the structure must be a multiple of 8 bytes. For the case of
+compiled-on-4 and run-on-8, we include an extra pointer that is always NULL so
+that there are an even number of pointers which therefore are a multiple of 8
+bytes.
+
+It is necessary to fork the struct for the 32 bit library, since it needs to
+use pcre_uint32 for first_char and req_char. We can't put an ifdef inside the
+typedef because pcretest needs access to the struct of the 8-, 16- and 32-bit
+variants.
+
+*** WARNING ***
+When new fields are added to these structures, remember to adjust the code in
+pcre_byte_order.c that is concerned with swapping the byte order of the fields
+when a compiled regex is reloaded on a host with different endianness.
+*** WARNING ***
+There is also similar byte-flipping code in pcretest.c, which is used for
+testing the byte-flipping features. It must also be kept in step.
+*** WARNING ***
+*/
+
+typedef struct real_pcre8_or_16 {
+  pcre_uint32 magic_number;
+  pcre_uint32 size;               /* Total that was malloced */
+  pcre_uint32 options;            /* Public options */
+  pcre_uint32 flags;              /* Private flags */
+  pcre_uint32 limit_match;        /* Limit set from regex */
+  pcre_uint32 limit_recursion;    /* Limit set from regex */
+  pcre_uint16 first_char;         /* Starting character */
+  pcre_uint16 req_char;           /* This character must be seen */
+  pcre_uint16 max_lookbehind;     /* Longest lookbehind (characters) */
+  pcre_uint16 top_bracket;        /* Highest numbered group */
+  pcre_uint16 top_backref;        /* Highest numbered back reference */
+  pcre_uint16 name_table_offset;  /* Offset to name table that follows */
+  pcre_uint16 name_entry_size;    /* Size of any name items */
+  pcre_uint16 name_count;         /* Number of name items */
+  pcre_uint16 ref_count;          /* Reference count */
+  pcre_uint16 dummy1;             /* To ensure size is a multiple of 8 */
+  pcre_uint16 dummy2;             /* To ensure size is a multiple of 8 */
+  pcre_uint16 dummy3;             /* To ensure size is a multiple of 8 */
+  const pcre_uint8 *tables;       /* Pointer to tables or NULL for std */
+  void             *nullpad;      /* NULL padding */
+} real_pcre8_or_16;
+
+typedef struct real_pcre8_or_16 real_pcre;
+typedef struct real_pcre8_or_16 real_pcre16;
+
+typedef struct real_pcre32 {
+  pcre_uint32 magic_number;
+  pcre_uint32 size;               /* Total that was malloced */
+  pcre_uint32 options;            /* Public options */
+  pcre_uint32 flags;              /* Private flags */
+  pcre_uint32 limit_match;        /* Limit set from regex */
+  pcre_uint32 limit_recursion;    /* Limit set from regex */
+  pcre_uint32 first_char;         /* Starting character */
+  pcre_uint32 req_char;           /* This character must be seen */
+  pcre_uint16 max_lookbehind;     /* Longest lookbehind (characters) */
+  pcre_uint16 top_bracket;        /* Highest numbered group */
+  pcre_uint16 top_backref;        /* Highest numbered back reference */
+  pcre_uint16 name_table_offset;  /* Offset to name table that follows */
+  pcre_uint16 name_entry_size;    /* Size of any name items */
+  pcre_uint16 name_count;         /* Number of name items */
+  pcre_uint16 ref_count;          /* Reference count */
+  pcre_uint16 dummy;              /* To ensure size is a multiple of 8 */
+  const pcre_uint8 *tables;       /* Pointer to tables or NULL for std */
+  void             *nullpad;      /* NULL padding */
+} real_pcre32;
+
+#if defined COMPILE_PCRE8
+#define REAL_PCRE real_pcre
+#elif defined COMPILE_PCRE16
+#define REAL_PCRE real_pcre16
+#elif defined COMPILE_PCRE32
+#define REAL_PCRE real_pcre32
+#endif
+
+/* Assert that the size of REAL_PCRE is divisible by 8 */
+typedef int __assert_real_pcre_size_divisible_8[(sizeof(REAL_PCRE) % 8) == 0 ? 1 : -1];
+
+/* Needed in pcretest to access some fields in the real_pcre* structures
+ * directly. They're unified for 8/16/32 bits since the structs only differ
+ * after these fields; if that ever changes, need to fork those defines into
+ * 8/16 and 32 bit versions. */
+#define REAL_PCRE_MAGIC(re)     (((REAL_PCRE*)re)->magic_number)
+#define REAL_PCRE_SIZE(re)      (((REAL_PCRE*)re)->size)
+#define REAL_PCRE_OPTIONS(re)   (((REAL_PCRE*)re)->options)
+#define REAL_PCRE_FLAGS(re)     (((REAL_PCRE*)re)->flags)
+
+/* The format of the block used to store data from pcre_study(). The same
+remark (see NOTE above) about extending this structure applies. */
+
+typedef struct pcre_study_data {
+  pcre_uint32 size;               /* Total that was malloced */
+  pcre_uint32 flags;              /* Private flags */
+  pcre_uint8 start_bits[32];      /* Starting char bits */
+  pcre_uint32 minlength;          /* Minimum subject length */
+} pcre_study_data;
+
+/* Structure for building a chain of open capturing subpatterns during
+compiling, so that instructions to close them can be compiled when (*ACCEPT) is
+encountered. This is also used to identify subpatterns that contain recursive
+back references to themselves, so that they can be made atomic. */
+
+typedef struct open_capitem {
+  struct open_capitem *next;    /* Chain link */
+  pcre_uint16 number;           /* Capture number */
+  pcre_uint16 flag;             /* Set TRUE if recursive back ref */
+} open_capitem;
+
+/* Structure for passing "static" information around between the functions
+doing the compiling, so that they are thread-safe. */
+
+typedef struct compile_data {
+  const pcre_uint8 *lcc;            /* Points to lower casing table */
+  const pcre_uint8 *fcc;            /* Points to case-flipping table */
+  const pcre_uint8 *cbits;          /* Points to character type table */
+  const pcre_uint8 *ctypes;         /* Points to table of type maps */
+  const pcre_uchar *start_workspace;/* The start of working space */
+  const pcre_uchar *start_code;     /* The start of the compiled code */
+  const pcre_uchar *start_pattern;  /* The start of the pattern */
+  const pcre_uchar *end_pattern;    /* The end of the pattern */
+  open_capitem *open_caps;          /* Chain of open capture items */
+  pcre_uchar *hwm;                  /* High watermark of workspace */
+  pcre_uchar *name_table;           /* The name/number table */
+  int  names_found;                 /* Number of entries so far */
+  int  name_entry_size;             /* Size of each entry */
+  int  workspace_size;              /* Size of workspace */
+  unsigned int bracount;            /* Count of capturing parens as we compile */
+  int  final_bracount;              /* Saved value after first pass */
+  int  max_lookbehind;              /* Maximum lookbehind (characters) */
+  int  top_backref;                 /* Maximum back reference */
+  unsigned int backref_map;         /* Bitmap of low back refs */
+  int  assert_depth;                /* Depth of nested assertions */
+  pcre_uint32 external_options;     /* External (initial) options */
+  pcre_uint32 external_flags;       /* External flag bits to be set */
+  int  req_varyopt;                 /* "After variable item" flag for reqbyte */
+  BOOL had_accept;                  /* (*ACCEPT) encountered */
+  BOOL had_pruneorskip;             /* (*PRUNE) or (*SKIP) encountered */
+  BOOL check_lookbehind;            /* Lookbehinds need later checking */
+  int  nltype;                      /* Newline type */
+  int  nllen;                       /* Newline string length */
+  pcre_uchar nl[4];                 /* Newline string when fixed length */
+} compile_data;
+
+/* Structure for maintaining a chain of pointers to the currently incomplete
+branches, for testing for left recursion while compiling. */
+
+typedef struct branch_chain {
+  struct branch_chain *outer;
+  pcre_uchar *current_branch;
+} branch_chain;
+
+/* Structure for items in a linked list that represents an explicit recursive
+call within the pattern; used by pcre_exec(). */
+
+typedef struct recursion_info {
+  struct recursion_info *prevrec; /* Previous recursion record (or NULL) */
+  unsigned int group_num;         /* Number of group that was called */
+  int *offset_save;               /* Pointer to start of saved offsets */
+  int saved_max;                  /* Number of saved offsets */
+  int saved_capture_last;         /* Last capture number */
+  PCRE_PUCHAR subject_position;   /* Position at start of recursion */
+} recursion_info;
+
+/* A similar structure for pcre_dfa_exec(). */
+
+typedef struct dfa_recursion_info {
+  struct dfa_recursion_info *prevrec;
+  int group_num;
+  PCRE_PUCHAR subject_position;
+} dfa_recursion_info;
+
+/* Structure for building a chain of data for holding the values of the subject
+pointer at the start of each subpattern, so as to detect when an empty string
+has been matched by a subpattern - to break infinite loops; used by
+pcre_exec(). */
+
+typedef struct eptrblock {
+  struct eptrblock *epb_prev;
+  PCRE_PUCHAR epb_saved_eptr;
+} eptrblock;
+
+
+/* Structure for passing "static" information around between the functions
+doing traditional NFA matching, so that they are thread-safe. */
+
+typedef struct match_data {
+  unsigned long int match_call_count;      /* As it says */
+  unsigned long int match_limit;           /* As it says */
+  unsigned long int match_limit_recursion; /* As it says */
+  int   *offset_vector;           /* Offset vector */
+  int    offset_end;              /* One past the end */
+  int    offset_max;              /* The maximum usable for return data */
+  int    nltype;                  /* Newline type */
+  int    nllen;                   /* Newline string length */
+  int    name_count;              /* Number of names in name table */
+  int    name_entry_size;         /* Size of entry in names table */
+  unsigned int skip_arg_count;    /* For counting SKIP_ARGs */
+  unsigned int ignore_skip_arg;   /* For re-run when SKIP arg name not found */
+  pcre_uchar *name_table;         /* Table of names */
+  pcre_uchar nl[4];               /* Newline string when fixed */
+  const  pcre_uint8 *lcc;         /* Points to lower casing table */
+  const  pcre_uint8 *fcc;         /* Points to case-flipping table */
+  const  pcre_uint8 *ctypes;      /* Points to table of type maps */
+  BOOL   notbol;                  /* NOTBOL flag */
+  BOOL   noteol;                  /* NOTEOL flag */
+  BOOL   utf;                     /* UTF-8 / UTF-16 flag */
+  BOOL   jscript_compat;          /* JAVASCRIPT_COMPAT flag */
+  BOOL   use_ucp;                 /* PCRE_UCP flag */
+  BOOL   endonly;                 /* Dollar not before final \n */
+  BOOL   notempty;                /* Empty string match not wanted */
+  BOOL   notempty_atstart;        /* Empty string match at start not wanted */
+  BOOL   hitend;                  /* Hit the end of the subject at some point */
+  BOOL   bsr_anycrlf;             /* \R is just any CRLF, not full Unicode */
+  BOOL   hasthen;                 /* Pattern contains (*THEN) */
+  const  pcre_uchar *start_code;  /* For use when recursing */
+  PCRE_PUCHAR start_subject;      /* Start of the subject string */
+  PCRE_PUCHAR end_subject;        /* End of the subject string */
+  PCRE_PUCHAR start_match_ptr;    /* Start of matched string */
+  PCRE_PUCHAR end_match_ptr;      /* Subject position at end match */
+  PCRE_PUCHAR start_used_ptr;     /* Earliest consulted character */
+  int    partial;                 /* PARTIAL options */
+  int    end_offset_top;          /* Highwater mark at end of match */
+  pcre_int32 capture_last;        /* Most recent capture number + overflow flag */
+  int    start_offset;            /* The start offset value */
+  int    match_function_type;     /* Set for certain special calls of MATCH() */
+  eptrblock *eptrchain;           /* Chain of eptrblocks for tail recursions */
+  int    eptrn;                   /* Next free eptrblock */
+  recursion_info *recursive;      /* Linked list of recursion data */
+  void  *callout_data;            /* To pass back to callouts */
+  const  pcre_uchar *mark;        /* Mark pointer to pass back on success */
+  const  pcre_uchar *nomatch_mark;/* Mark pointer to pass back on failure */
+  const  pcre_uchar *once_target; /* Where to back up to for atomic groups */
+#ifdef NO_RECURSE
+  void  *match_frames_base;       /* For remembering malloc'd frames */
+#endif
+} match_data;
+
+/* A similar structure is used for the same purpose by the DFA matching
+functions. */
+
+typedef struct dfa_match_data {
+  const pcre_uchar *start_code;     /* Start of the compiled pattern */
+  const pcre_uchar *start_subject ; /* Start of the subject string */
+  const pcre_uchar *end_subject;    /* End of subject string */
+  const pcre_uchar *start_used_ptr; /* Earliest consulted character */
+  const pcre_uint8 *tables;         /* Character tables */
+  int   start_offset;               /* The start offset value */
+  int   moptions;                   /* Match options */
+  int   poptions;                   /* Pattern options */
+  int   nltype;                     /* Newline type */
+  int   nllen;                      /* Newline string length */
+  pcre_uchar nl[4];                 /* Newline string when fixed */
+  void *callout_data;               /* To pass back to callouts */
+  dfa_recursion_info *recursive;    /* Linked list of recursion data */
+} dfa_match_data;
+
+/* Bit definitions for entries in the pcre_ctypes table. */
+
+#define ctype_space   0x01
+#define ctype_letter  0x02
+#define ctype_digit   0x04
+#define ctype_xdigit  0x08
+#define ctype_word    0x10   /* alphanumeric or '_' */
+#define ctype_meta    0x80   /* regexp meta char or zero (end pattern) */
+
+/* Offsets for the bitmap tables in pcre_cbits. Each table contains a set
+of bits for a class map. Some classes are built by combining these tables. */
+
+#define cbit_space     0      /* [:space:] or \s */
+#define cbit_xdigit   32      /* [:xdigit:] */
+#define cbit_digit    64      /* [:digit:] or \d */
+#define cbit_upper    96      /* [:upper:] */
+#define cbit_lower   128      /* [:lower:] */
+#define cbit_word    160      /* [:word:] or \w */
+#define cbit_graph   192      /* [:graph:] */
+#define cbit_print   224      /* [:print:] */
+#define cbit_punct   256      /* [:punct:] */
+#define cbit_cntrl   288      /* [:cntrl:] */
+#define cbit_length  320      /* Length of the cbits table */
+
+/* Offsets of the various tables from the base tables pointer, and
+total length. */
+
+#define lcc_offset      0
+#define fcc_offset    256
+#define cbits_offset  512
+#define ctypes_offset (cbits_offset + cbit_length)
+#define tables_length (ctypes_offset + 256)
+
+/* Internal function and data prefixes. */
+
+#if defined COMPILE_PCRE8
+#ifndef PUBL
+#define PUBL(name) pcre_##name
+#endif
+#ifndef PRIV
+#define PRIV(name) _pcre_##name
+#endif
+#elif defined COMPILE_PCRE16
+#ifndef PUBL
+#define PUBL(name) pcre16_##name
+#endif
+#ifndef PRIV
+#define PRIV(name) _pcre16_##name
+#endif
+#elif defined COMPILE_PCRE32
+#ifndef PUBL
+#define PUBL(name) pcre32_##name
+#endif
+#ifndef PRIV
+#define PRIV(name) _pcre32_##name
+#endif
+#else
+#error Unsupported compiling mode
+#endif /* COMPILE_PCRE[8|16|32] */
+
+/* Layout of the UCP type table that translates property names into types and
+codes. Each entry used to point directly to a name, but to reduce the number of
+relocations in shared libraries, it now has an offset into a single string
+instead. */
+
+typedef struct {
+  pcre_uint16 name_offset;
+  pcre_uint16 type;
+  pcre_uint16 value;
+} ucp_type_table;
+
+
+/* Internal shared data tables. These are tables that are used by more than one
+of the exported public functions. They have to be "external" in the C sense,
+but are not part of the PCRE public API. The data for these tables is in the
+pcre_tables.c module. */
+
+#ifdef COMPILE_PCRE8
+extern const int            PRIV(utf8_table1)[];
+extern const int            PRIV(utf8_table1_size);
+extern const int            PRIV(utf8_table2)[];
+extern const int            PRIV(utf8_table3)[];
+extern const pcre_uint8     PRIV(utf8_table4)[];
+#endif /* COMPILE_PCRE8 */
+
+extern const char           PRIV(utt_names)[];
+extern const ucp_type_table PRIV(utt)[];
+extern const int            PRIV(utt_size);
+
+extern const pcre_uint8     PRIV(OP_lengths)[];
+extern const pcre_uint8     PRIV(default_tables)[];
+
+extern const pcre_uint32    PRIV(hspace_list)[];
+extern const pcre_uint32    PRIV(vspace_list)[];
+
+
+/* Internal shared functions. These are functions that are used by more than
+one of the exported public functions. They have to be "external" in the C
+sense, but are not part of the PCRE public API. */
+
+/* String comparison functions. */
+#if defined COMPILE_PCRE8
+
+#define STRCMP_UC_UC(str1, str2) \
+  strcmp((char *)(str1), (char *)(str2))
+#define STRCMP_UC_C8(str1, str2) \
+  strcmp((char *)(str1), (str2))
+#define STRNCMP_UC_UC(str1, str2, num) \
+  strncmp((char *)(str1), (char *)(str2), (num))
+#define STRNCMP_UC_C8(str1, str2, num) \
+  strncmp((char *)(str1), (str2), (num))
+#define STRLEN_UC(str) strlen((const char *)str)
+
+#elif defined COMPILE_PCRE16 || defined COMPILE_PCRE32
+
+extern int               PRIV(strcmp_uc_uc)(const pcre_uchar *,
+                           const pcre_uchar *);
+extern int               PRIV(strcmp_uc_c8)(const pcre_uchar *,
+                           const char *);
+extern int               PRIV(strncmp_uc_uc)(const pcre_uchar *,
+                           const pcre_uchar *, unsigned int num);
+extern int               PRIV(strncmp_uc_c8)(const pcre_uchar *,
+                           const char *, unsigned int num);
+extern unsigned int      PRIV(strlen_uc)(const pcre_uchar *str);
+
+#define STRCMP_UC_UC(str1, str2) \
+  PRIV(strcmp_uc_uc)((str1), (str2))
+#define STRCMP_UC_C8(str1, str2) \
+  PRIV(strcmp_uc_c8)((str1), (str2))
+#define STRNCMP_UC_UC(str1, str2, num) \
+  PRIV(strncmp_uc_uc)((str1), (str2), (num))
+#define STRNCMP_UC_C8(str1, str2, num) \
+  PRIV(strncmp_uc_c8)((str1), (str2), (num))
+#define STRLEN_UC(str) PRIV(strlen_uc)(str)
+
+#endif /* COMPILE_PCRE[8|16|32] */
+
+#if defined COMPILE_PCRE8 || defined COMPILE_PCRE16
+
+#define STRCMP_UC_UC_TEST(str1, str2) STRCMP_UC_UC(str1, str2)
+#define STRCMP_UC_C8_TEST(str1, str2) STRCMP_UC_C8(str1, str2)
+
+#elif defined COMPILE_PCRE32
+
+extern int               PRIV(strcmp_uc_uc_utf)(const pcre_uchar *,
+                           const pcre_uchar *);
+extern int               PRIV(strcmp_uc_c8_utf)(const pcre_uchar *,
+                           const char *);
+
+#define STRCMP_UC_UC_TEST(str1, str2) \
+  (utf ? PRIV(strcmp_uc_uc_utf)((str1), (str2)) : PRIV(strcmp_uc_uc)((str1), (str2)))
+#define STRCMP_UC_C8_TEST(str1, str2) \
+  (utf ? PRIV(strcmp_uc_c8_utf)((str1), (str2)) : PRIV(strcmp_uc_c8)((str1), (str2)))
+
+#endif /* COMPILE_PCRE[8|16|32] */
+
+extern const pcre_uchar *PRIV(find_bracket)(const pcre_uchar *, BOOL, int);
+extern BOOL              PRIV(is_newline)(PCRE_PUCHAR, int, PCRE_PUCHAR,
+                           int *, BOOL);
+extern unsigned int      PRIV(ord2utf)(pcre_uint32, pcre_uchar *);
+extern int               PRIV(valid_utf)(PCRE_PUCHAR, int, int *);
+extern BOOL              PRIV(was_newline)(PCRE_PUCHAR, int, PCRE_PUCHAR,
+                           int *, BOOL);
+extern BOOL              PRIV(xclass)(pcre_uint32, const pcre_uchar *, BOOL);
+
+#ifdef SUPPORT_JIT
+extern void              PRIV(jit_compile)(const REAL_PCRE *,
+                           PUBL(extra) *, int);
+extern int               PRIV(jit_exec)(const PUBL(extra) *,
+                           const pcre_uchar *, int, int, int, int *, int);
+extern void              PRIV(jit_free)(void *);
+extern int               PRIV(jit_get_size)(void *);
+extern const char*       PRIV(jit_get_target)(void);
+#endif
+
+/* Unicode character database (UCD) */
+
+typedef struct {
+  pcre_uint8 script;     /* ucp_Arabic, etc. */
+  pcre_uint8 chartype;   /* ucp_Cc, etc. (general categories) */
+  pcre_uint8 gbprop;     /* ucp_gbControl, etc. (grapheme break property) */
+  pcre_uint8 caseset;    /* offset to multichar other cases or zero */
+  pcre_int32 other_case; /* offset to other case, or zero if none */
+} ucd_record;
+
+extern const pcre_uint32 PRIV(ucd_caseless_sets)[];
+extern const ucd_record  PRIV(ucd_records)[];
+extern const pcre_uint8  PRIV(ucd_stage1)[];
+extern const pcre_uint16 PRIV(ucd_stage2)[];
+extern const pcre_uint32 PRIV(ucp_gentype)[];
+extern const pcre_uint32 PRIV(ucp_gbtable)[];
+#ifdef SUPPORT_JIT
+extern const int         PRIV(ucp_typerange)[];
+#endif
+
+#ifdef SUPPORT_UCP
+/* UCD access macros */
+
+#define UCD_BLOCK_SIZE 128
+#define GET_UCD(ch) (PRIV(ucd_records) + \
+        PRIV(ucd_stage2)[PRIV(ucd_stage1)[(int)(ch) / UCD_BLOCK_SIZE] * \
+        UCD_BLOCK_SIZE + (int)(ch) % UCD_BLOCK_SIZE])
+
+#define UCD_CHARTYPE(ch)    GET_UCD(ch)->chartype
+#define UCD_SCRIPT(ch)      GET_UCD(ch)->script
+#define UCD_CATEGORY(ch)    PRIV(ucp_gentype)[UCD_CHARTYPE(ch)]
+#define UCD_GRAPHBREAK(ch)  GET_UCD(ch)->gbprop
+#define UCD_CASESET(ch)     GET_UCD(ch)->caseset
+#define UCD_OTHERCASE(ch)   ((pcre_uint32)((int)ch + (int)(GET_UCD(ch)->other_case)))
+
+#endif /* SUPPORT_UCP */
+
+#endif
+
+/* End of pcre_internal.h */
diff --git a/pcre/pcre_jit_compile.c b/pcre/pcre_jit_compile.c
new file mode 100644
index 00000000000..201416dd264
--- /dev/null
+++ b/pcre/pcre_jit_compile.c
@@ -0,0 +1,9752 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2013 University of Cambridge
+
+  The machine code generator part (this module) was written by Zoltan Herczeg
+                      Copyright (c) 2010-2013
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include "pcre_internal.h"
+
+#if defined SUPPORT_JIT
+
+/* All-in-one: Since we use the JIT compiler only from here,
+we just include it. This way we don't need to touch the build
+system files. */
+
+#define SLJIT_MALLOC(size) (PUBL(malloc))(size)
+#define SLJIT_FREE(ptr) (PUBL(free))(ptr)
+#define SLJIT_CONFIG_AUTO 1
+#define SLJIT_CONFIG_STATIC 1
+#define SLJIT_VERBOSE 0
+#define SLJIT_DEBUG 0
+
+#include "sljit/sljitLir.c"
+
+#if defined SLJIT_CONFIG_UNSUPPORTED && SLJIT_CONFIG_UNSUPPORTED
+#error Unsupported architecture
+#endif
+
+/* Defines for debugging purposes. */
+
+/* 1 - Use unoptimized capturing brackets.
+   2 - Enable capture_last_ptr (includes option 1). */
+/* #define DEBUG_FORCE_UNOPTIMIZED_CBRAS 2 */
+
+/* 1 - Always have a control head. */
+/* #define DEBUG_FORCE_CONTROL_HEAD 1 */
+
+/* Allocate memory for the regex stack on the real machine stack.
+Fast, but limited size. */
+#define MACHINE_STACK_SIZE 32768
+
+/* Growth rate for stack allocated by the OS. Should be the multiply
+of page size. */
+#define STACK_GROWTH_RATE 8192
+
+/* Enable to check that the allocation could destroy temporaries. */
+#if defined SLJIT_DEBUG && SLJIT_DEBUG
+#define DESTROY_REGISTERS 1
+#endif
+
+/*
+Short summary about the backtracking mechanism empolyed by the jit code generator:
+
+The code generator follows the recursive nature of the PERL compatible regular
+expressions. The basic blocks of regular expressions are condition checkers
+whose execute different commands depending on the result of the condition check.
+The relationship between the operators can be horizontal (concatenation) and
+vertical (sub-expression) (See struct backtrack_common for more details).
+
+  'ab' - 'a' and 'b' regexps are concatenated
+  'a+' - 'a' is the sub-expression of the '+' operator
+
+The condition checkers are boolean (true/false) checkers. Machine code is generated
+for the checker itself and for the actions depending on the result of the checker.
+The 'true' case is called as the matching path (expected path), and the other is called as
+the 'backtrack' path. Branch instructions are expesive for all CPUs, so we avoid taken
+branches on the matching path.
+
+ Greedy star operator (*) :
+   Matching path: match happens.
+   Backtrack path: match failed.
+ Non-greedy star operator (*?) :
+   Matching path: no need to perform a match.
+   Backtrack path: match is required.
+
+The following example shows how the code generated for a capturing bracket
+with two alternatives. Let A, B, C, D are arbirary regular expressions, and
+we have the following regular expression:
+
+   A(B|C)D
+
+The generated code will be the following:
+
+ A matching path
+ '(' matching path (pushing arguments to the stack)
+ B matching path
+ ')' matching path (pushing arguments to the stack)
+ D matching path
+ return with successful match
+
+ D backtrack path
+ ')' backtrack path (If we arrived from "C" jump to the backtrack of "C")
+ B backtrack path
+ C expected path
+ jump to D matching path
+ C backtrack path
+ A backtrack path
+
+ Notice, that the order of backtrack code paths are the opposite of the fast
+ code paths. In this way the topmost value on the stack is always belong
+ to the current backtrack code path. The backtrack path must check
+ whether there is a next alternative. If so, it needs to jump back to
+ the matching path eventually. Otherwise it needs to clear out its own stack
+ frame and continue the execution on the backtrack code paths.
+*/
+
+/*
+Saved stack frames:
+
+Atomic blocks and asserts require reloading the values of private data
+when the backtrack mechanism performed. Because of OP_RECURSE, the data
+are not necessarly known in compile time, thus we need a dynamic restore
+mechanism.
+
+The stack frames are stored in a chain list, and have the following format:
+([ capturing bracket offset ][ start value ][ end value ])+ ... [ 0 ] [ previous head ]
+
+Thus we can restore the private data to a particular point in the stack.
+*/
+
+typedef struct jit_arguments {
+  /* Pointers first. */
+  struct sljit_stack *stack;
+  const pcre_uchar *str;
+  const pcre_uchar *begin;
+  const pcre_uchar *end;
+  int *offsets;
+  pcre_uchar *uchar_ptr;
+  pcre_uchar *mark_ptr;
+  void *callout_data;
+  /* Everything else after. */
+  pcre_uint32 limit_match;
+  int real_offset_count;
+  int offset_count;
+  pcre_uint8 notbol;
+  pcre_uint8 noteol;
+  pcre_uint8 notempty;
+  pcre_uint8 notempty_atstart;
+} jit_arguments;
+
+typedef struct executable_functions {
+  void *executable_funcs[JIT_NUMBER_OF_COMPILE_MODES];
+  PUBL(jit_callback) callback;
+  void *userdata;
+  pcre_uint32 top_bracket;
+  pcre_uint32 limit_match;
+  sljit_uw executable_sizes[JIT_NUMBER_OF_COMPILE_MODES];
+} executable_functions;
+
+typedef struct jump_list {
+  struct sljit_jump *jump;
+  struct jump_list *next;
+} jump_list;
+
+typedef struct stub_list {
+  struct sljit_jump *start;
+  struct sljit_label *quit;
+  struct stub_list *next;
+} stub_list;
+
+enum frame_types {
+  no_frame = -1,
+  no_stack = -2
+};
+
+enum control_types {
+  type_mark = 0,
+  type_then_trap = 1
+};
+
+typedef int (SLJIT_CALL *jit_function)(jit_arguments *args);
+
+/* The following structure is the key data type for the recursive
+code generator. It is allocated by compile_matchingpath, and contains
+the arguments for compile_backtrackingpath. Must be the first member
+of its descendants. */
+typedef struct backtrack_common {
+  /* Concatenation stack. */
+  struct backtrack_common *prev;
+  jump_list *nextbacktracks;
+  /* Internal stack (for component operators). */
+  struct backtrack_common *top;
+  jump_list *topbacktracks;
+  /* Opcode pointer. */
+  pcre_uchar *cc;
+} backtrack_common;
+
+typedef struct assert_backtrack {
+  backtrack_common common;
+  jump_list *condfailed;
+  /* Less than 0 if a frame is not needed. */
+  int framesize;
+  /* Points to our private memory word on the stack. */
+  int private_data_ptr;
+  /* For iterators. */
+  struct sljit_label *matchingpath;
+} assert_backtrack;
+
+typedef struct bracket_backtrack {
+  backtrack_common common;
+  /* Where to coninue if an alternative is successfully matched. */
+  struct sljit_label *alternative_matchingpath;
+  /* For rmin and rmax iterators. */
+  struct sljit_label *recursive_matchingpath;
+  /* For greedy ? operator. */
+  struct sljit_label *zero_matchingpath;
+  /* Contains the branches of a failed condition. */
+  union {
+    /* Both for OP_COND, OP_SCOND. */
+    jump_list *condfailed;
+    assert_backtrack *assert;
+    /* For OP_ONCE. Less than 0 if not needed. */
+    int framesize;
+  } u;
+  /* Points to our private memory word on the stack. */
+  int private_data_ptr;
+} bracket_backtrack;
+
+typedef struct bracketpos_backtrack {
+  backtrack_common common;
+  /* Points to our private memory word on the stack. */
+  int private_data_ptr;
+  /* Reverting stack is needed. */
+  int framesize;
+  /* Allocated stack size. */
+  int stacksize;
+} bracketpos_backtrack;
+
+typedef struct braminzero_backtrack {
+  backtrack_common common;
+  struct sljit_label *matchingpath;
+} braminzero_backtrack;
+
+typedef struct iterator_backtrack {
+  backtrack_common common;
+  /* Next iteration. */
+  struct sljit_label *matchingpath;
+} iterator_backtrack;
+
+typedef struct recurse_entry {
+  struct recurse_entry *next;
+  /* Contains the function entry. */
+  struct sljit_label *entry;
+  /* Collects the calls until the function is not created. */
+  jump_list *calls;
+  /* Points to the starting opcode. */
+  sljit_sw start;
+} recurse_entry;
+
+typedef struct recurse_backtrack {
+  backtrack_common common;
+  BOOL inlined_pattern;
+} recurse_backtrack;
+
+#define OP_THEN_TRAP OP_TABLE_LENGTH
+
+typedef struct then_trap_backtrack {
+  backtrack_common common;
+  /* If then_trap is not NULL, this structure contains the real
+  then_trap for the backtracking path. */
+  struct then_trap_backtrack *then_trap;
+  /* Points to the starting opcode. */
+  sljit_sw start;
+  /* Exit point for the then opcodes of this alternative. */
+  jump_list *quit;
+  /* Frame size of the current alternative. */
+  int framesize;
+} then_trap_backtrack;
+
+#define MAX_RANGE_SIZE 6
+
+typedef struct compiler_common {
+  /* The sljit ceneric compiler. */
+  struct sljit_compiler *compiler;
+  /* First byte code. */
+  pcre_uchar *start;
+  /* Maps private data offset to each opcode. */
+  sljit_si *private_data_ptrs;
+  /* Tells whether the capturing bracket is optimized. */
+  pcre_uint8 *optimized_cbracket;
+  /* Tells whether the starting offset is a target of then. */
+  pcre_uint8 *then_offsets;
+  /* Current position where a THEN must jump. */
+  then_trap_backtrack *then_trap;
+  /* Starting offset of private data for capturing brackets. */
+  int cbra_ptr;
+  /* Output vector starting point. Must be divisible by 2. */
+  int ovector_start;
+  /* Last known position of the requested byte. */
+  int req_char_ptr;
+  /* Head of the last recursion. */
+  int recursive_head_ptr;
+  /* First inspected character for partial matching. */
+  int start_used_ptr;
+  /* Starting pointer for partial soft matches. */
+  int hit_start;
+  /* End pointer of the first line. */
+  int first_line_end;
+  /* Points to the marked string. */
+  int mark_ptr;
+  /* Recursive control verb management chain. */
+  int control_head_ptr;
+  /* Points to the last matched capture block index. */
+  int capture_last_ptr;
+  /* Points to the starting position of the current match. */
+  int start_ptr;
+
+  /* Flipped and lower case tables. */
+  const pcre_uint8 *fcc;
+  sljit_sw lcc;
+  /* Mode can be PCRE_STUDY_JIT_COMPILE and others. */
+  int mode;
+  /* \K is found in the pattern. */
+  BOOL has_set_som;
+  /* (*SKIP:arg) is found in the pattern. */
+  BOOL has_skip_arg;
+  /* (*THEN) is found in the pattern. */
+  BOOL has_then;
+  /* Needs to know the start position anytime. */
+  BOOL needs_start_ptr;
+  /* Currently in recurse or negative assert. */
+  BOOL local_exit;
+  /* Currently in a positive assert. */
+  BOOL positive_assert;
+  /* Newline control. */
+  int nltype;
+  int newline;
+  int bsr_nltype;
+  /* Dollar endonly. */
+  int endonly;
+  /* Tables. */
+  sljit_sw ctypes;
+  int digits[2 + MAX_RANGE_SIZE];
+  /* Named capturing brackets. */
+  sljit_uw name_table;
+  sljit_sw name_count;
+  sljit_sw name_entry_size;
+
+  /* Labels and jump lists. */
+  struct sljit_label *partialmatchlabel;
+  struct sljit_label *quit_label;
+  struct sljit_label *forced_quit_label;
+  struct sljit_label *accept_label;
+  stub_list *stubs;
+  recurse_entry *entries;
+  recurse_entry *currententry;
+  jump_list *partialmatch;
+  jump_list *quit;
+  jump_list *positive_assert_quit;
+  jump_list *forced_quit;
+  jump_list *accept;
+  jump_list *calllimit;
+  jump_list *stackalloc;
+  jump_list *revertframes;
+  jump_list *wordboundary;
+  jump_list *anynewline;
+  jump_list *hspace;
+  jump_list *vspace;
+  jump_list *casefulcmp;
+  jump_list *caselesscmp;
+  jump_list *reset_match;
+  BOOL jscript_compat;
+#ifdef SUPPORT_UTF
+  BOOL utf;
+#ifdef SUPPORT_UCP
+  BOOL use_ucp;
+#endif
+#ifndef COMPILE_PCRE32
+  jump_list *utfreadchar;
+#endif
+#ifdef COMPILE_PCRE8
+  jump_list *utfreadtype8;
+#endif
+#endif /* SUPPORT_UTF */
+#ifdef SUPPORT_UCP
+  jump_list *getucd;
+#endif
+} compiler_common;
+
+/* For byte_sequence_compare. */
+
+typedef struct compare_context {
+  int length;
+  int sourcereg;
+#if defined SLJIT_UNALIGNED && SLJIT_UNALIGNED
+  int ucharptr;
+  union {
+    sljit_si asint;
+    sljit_uh asushort;
+#if defined COMPILE_PCRE8
+    sljit_ub asbyte;
+    sljit_ub asuchars[4];
+#elif defined COMPILE_PCRE16
+    sljit_uh asuchars[2];
+#elif defined COMPILE_PCRE32
+    sljit_ui asuchars[1];
+#endif
+  } c;
+  union {
+    sljit_si asint;
+    sljit_uh asushort;
+#if defined COMPILE_PCRE8
+    sljit_ub asbyte;
+    sljit_ub asuchars[4];
+#elif defined COMPILE_PCRE16
+    sljit_uh asuchars[2];
+#elif defined COMPILE_PCRE32
+    sljit_ui asuchars[1];
+#endif
+  } oc;
+#endif
+} compare_context;
+
+/* Undefine sljit macros. */
+#undef CMP
+
+/* Used for accessing the elements of the stack. */
+#define STACK(i)      ((-(i) - 1) * (int)sizeof(sljit_sw))
+
+#define TMP1          SLJIT_SCRATCH_REG1
+#define TMP2          SLJIT_SCRATCH_REG3
+#define TMP3          SLJIT_TEMPORARY_EREG2
+#define STR_PTR       SLJIT_SAVED_REG1
+#define STR_END       SLJIT_SAVED_REG2
+#define STACK_TOP     SLJIT_SCRATCH_REG2
+#define STACK_LIMIT   SLJIT_SAVED_REG3
+#define ARGUMENTS     SLJIT_SAVED_EREG1
+#define COUNT_MATCH   SLJIT_SAVED_EREG2
+#define RETURN_ADDR   SLJIT_TEMPORARY_EREG1
+
+/* Local space layout. */
+/* These two locals can be used by the current opcode. */
+#define LOCALS0          (0 * sizeof(sljit_sw))
+#define LOCALS1          (1 * sizeof(sljit_sw))
+/* Two local variables for possessive quantifiers (char1 cannot use them). */
+#define POSSESSIVE0      (2 * sizeof(sljit_sw))
+#define POSSESSIVE1      (3 * sizeof(sljit_sw))
+/* Max limit of recursions. */
+#define LIMIT_MATCH      (4 * sizeof(sljit_sw))
+/* The output vector is stored on the stack, and contains pointers
+to characters. The vector data is divided into two groups: the first
+group contains the start / end character pointers, and the second is
+the start pointers when the end of the capturing group has not yet reached. */
+#define OVECTOR_START    (common->ovector_start)
+#define OVECTOR(i)       (OVECTOR_START + (i) * sizeof(sljit_sw))
+#define OVECTOR_PRIV(i)  (common->cbra_ptr + (i) * sizeof(sljit_sw))
+#define PRIVATE_DATA(cc) (common->private_data_ptrs[(cc) - common->start])
+
+#if defined COMPILE_PCRE8
+#define MOV_UCHAR  SLJIT_MOV_UB
+#define MOVU_UCHAR SLJIT_MOVU_UB
+#elif defined COMPILE_PCRE16
+#define MOV_UCHAR  SLJIT_MOV_UH
+#define MOVU_UCHAR SLJIT_MOVU_UH
+#elif defined COMPILE_PCRE32
+#define MOV_UCHAR  SLJIT_MOV_UI
+#define MOVU_UCHAR SLJIT_MOVU_UI
+#else
+#error Unsupported compiling mode
+#endif
+
+/* Shortcuts. */
+#define DEFINE_COMPILER \
+  struct sljit_compiler *compiler = common->compiler
+#define OP1(op, dst, dstw, src, srcw) \
+  sljit_emit_op1(compiler, (op), (dst), (dstw), (src), (srcw))
+#define OP2(op, dst, dstw, src1, src1w, src2, src2w) \
+  sljit_emit_op2(compiler, (op), (dst), (dstw), (src1), (src1w), (src2), (src2w))
+#define LABEL() \
+  sljit_emit_label(compiler)
+#define JUMP(type) \
+  sljit_emit_jump(compiler, (type))
+#define JUMPTO(type, label) \
+  sljit_set_label(sljit_emit_jump(compiler, (type)), (label))
+#define JUMPHERE(jump) \
+  sljit_set_label((jump), sljit_emit_label(compiler))
+#define SET_LABEL(jump, label) \
+  sljit_set_label((jump), (label))
+#define CMP(type, src1, src1w, src2, src2w) \
+  sljit_emit_cmp(compiler, (type), (src1), (src1w), (src2), (src2w))
+#define CMPTO(type, src1, src1w, src2, src2w, label) \
+  sljit_set_label(sljit_emit_cmp(compiler, (type), (src1), (src1w), (src2), (src2w)), (label))
+#define OP_FLAGS(op, dst, dstw, src, srcw, type) \
+  sljit_emit_op_flags(compiler, (op), (dst), (dstw), (src), (srcw), (type))
+#define GET_LOCAL_BASE(dst, dstw, offset) \
+  sljit_get_local_base(compiler, (dst), (dstw), (offset))
+
+static pcre_uchar* bracketend(pcre_uchar* cc)
+{
+SLJIT_ASSERT((*cc >= OP_ASSERT && *cc <= OP_ASSERTBACK_NOT) || (*cc >= OP_ONCE && *cc <= OP_SCOND));
+do cc += GET(cc, 1); while (*cc == OP_ALT);
+SLJIT_ASSERT(*cc >= OP_KET && *cc <= OP_KETRPOS);
+cc += 1 + LINK_SIZE;
+return cc;
+}
+
+/* Functions whose might need modification for all new supported opcodes:
+ next_opcode
+ check_opcode_types
+ set_private_data_ptrs
+ get_framesize
+ init_frame
+ get_private_data_copy_length
+ copy_private_data
+ compile_matchingpath
+ compile_backtrackingpath
+*/
+
+static pcre_uchar *next_opcode(compiler_common *common, pcre_uchar *cc)
+{
+SLJIT_UNUSED_ARG(common);
+switch(*cc)
+  {
+  case OP_SOD:
+  case OP_SOM:
+  case OP_SET_SOM:
+  case OP_NOT_WORD_BOUNDARY:
+  case OP_WORD_BOUNDARY:
+  case OP_NOT_DIGIT:
+  case OP_DIGIT:
+  case OP_NOT_WHITESPACE:
+  case OP_WHITESPACE:
+  case OP_NOT_WORDCHAR:
+  case OP_WORDCHAR:
+  case OP_ANY:
+  case OP_ALLANY:
+  case OP_NOTPROP:
+  case OP_PROP:
+  case OP_ANYNL:
+  case OP_NOT_HSPACE:
+  case OP_HSPACE:
+  case OP_NOT_VSPACE:
+  case OP_VSPACE:
+  case OP_EXTUNI:
+  case OP_EODN:
+  case OP_EOD:
+  case OP_CIRC:
+  case OP_CIRCM:
+  case OP_DOLL:
+  case OP_DOLLM:
+  case OP_CRSTAR:
+  case OP_CRMINSTAR:
+  case OP_CRPLUS:
+  case OP_CRMINPLUS:
+  case OP_CRQUERY:
+  case OP_CRMINQUERY:
+  case OP_CRRANGE:
+  case OP_CRMINRANGE:
+  case OP_CLASS:
+  case OP_NCLASS:
+  case OP_REF:
+  case OP_REFI:
+  case OP_RECURSE:
+  case OP_CALLOUT:
+  case OP_ALT:
+  case OP_KET:
+  case OP_KETRMAX:
+  case OP_KETRMIN:
+  case OP_KETRPOS:
+  case OP_REVERSE:
+  case OP_ASSERT:
+  case OP_ASSERT_NOT:
+  case OP_ASSERTBACK:
+  case OP_ASSERTBACK_NOT:
+  case OP_ONCE:
+  case OP_ONCE_NC:
+  case OP_BRA:
+  case OP_BRAPOS:
+  case OP_CBRA:
+  case OP_CBRAPOS:
+  case OP_COND:
+  case OP_SBRA:
+  case OP_SBRAPOS:
+  case OP_SCBRA:
+  case OP_SCBRAPOS:
+  case OP_SCOND:
+  case OP_CREF:
+  case OP_NCREF:
+  case OP_RREF:
+  case OP_NRREF:
+  case OP_DEF:
+  case OP_BRAZERO:
+  case OP_BRAMINZERO:
+  case OP_BRAPOSZERO:
+  case OP_PRUNE:
+  case OP_SKIP:
+  case OP_THEN:
+  case OP_COMMIT:
+  case OP_FAIL:
+  case OP_ACCEPT:
+  case OP_ASSERT_ACCEPT:
+  case OP_CLOSE:
+  case OP_SKIPZERO:
+  return cc + PRIV(OP_lengths)[*cc];
+
+  case OP_CHAR:
+  case OP_CHARI:
+  case OP_NOT:
+  case OP_NOTI:
+  case OP_STAR:
+  case OP_MINSTAR:
+  case OP_PLUS:
+  case OP_MINPLUS:
+  case OP_QUERY:
+  case OP_MINQUERY:
+  case OP_UPTO:
+  case OP_MINUPTO:
+  case OP_EXACT:
+  case OP_POSSTAR:
+  case OP_POSPLUS:
+  case OP_POSQUERY:
+  case OP_POSUPTO:
+  case OP_STARI:
+  case OP_MINSTARI:
+  case OP_PLUSI:
+  case OP_MINPLUSI:
+  case OP_QUERYI:
+  case OP_MINQUERYI:
+  case OP_UPTOI:
+  case OP_MINUPTOI:
+  case OP_EXACTI:
+  case OP_POSSTARI:
+  case OP_POSPLUSI:
+  case OP_POSQUERYI:
+  case OP_POSUPTOI:
+  case OP_NOTSTAR:
+  case OP_NOTMINSTAR:
+  case OP_NOTPLUS:
+  case OP_NOTMINPLUS:
+  case OP_NOTQUERY:
+  case OP_NOTMINQUERY:
+  case OP_NOTUPTO:
+  case OP_NOTMINUPTO:
+  case OP_NOTEXACT:
+  case OP_NOTPOSSTAR:
+  case OP_NOTPOSPLUS:
+  case OP_NOTPOSQUERY:
+  case OP_NOTPOSUPTO:
+  case OP_NOTSTARI:
+  case OP_NOTMINSTARI:
+  case OP_NOTPLUSI:
+  case OP_NOTMINPLUSI:
+  case OP_NOTQUERYI:
+  case OP_NOTMINQUERYI:
+  case OP_NOTUPTOI:
+  case OP_NOTMINUPTOI:
+  case OP_NOTEXACTI:
+  case OP_NOTPOSSTARI:
+  case OP_NOTPOSPLUSI:
+  case OP_NOTPOSQUERYI:
+  case OP_NOTPOSUPTOI:
+  cc += PRIV(OP_lengths)[*cc];
+#ifdef SUPPORT_UTF
+  if (common->utf && HAS_EXTRALEN(cc[-1])) cc += GET_EXTRALEN(cc[-1]);
+#endif
+  return cc;
+
+  /* Special cases. */
+  case OP_TYPESTAR:
+  case OP_TYPEMINSTAR:
+  case OP_TYPEPLUS:
+  case OP_TYPEMINPLUS:
+  case OP_TYPEQUERY:
+  case OP_TYPEMINQUERY:
+  case OP_TYPEUPTO:
+  case OP_TYPEMINUPTO:
+  case OP_TYPEEXACT:
+  case OP_TYPEPOSSTAR:
+  case OP_TYPEPOSPLUS:
+  case OP_TYPEPOSQUERY:
+  case OP_TYPEPOSUPTO:
+  return cc + PRIV(OP_lengths)[*cc] - 1;
+
+  case OP_ANYBYTE:
+#ifdef SUPPORT_UTF
+  if (common->utf) return NULL;
+#endif
+  return cc + 1;
+
+#if defined SUPPORT_UTF || !defined COMPILE_PCRE8
+  case OP_XCLASS:
+  return cc + GET(cc, 1);
+#endif
+
+  case OP_MARK:
+  case OP_PRUNE_ARG:
+  case OP_SKIP_ARG:
+  case OP_THEN_ARG:
+  return cc + 1 + 2 + cc[1];
+
+  default:
+  /* All opcodes are supported now! */
+  SLJIT_ASSERT_STOP();
+  return NULL;
+  }
+}
+
+static BOOL check_opcode_types(compiler_common *common, pcre_uchar *cc, pcre_uchar *ccend)
+{
+pcre_uchar *name;
+pcre_uchar *name2;
+unsigned int cbra_index;
+int i;
+
+/* Calculate important variables (like stack size) and checks whether all opcodes are supported. */
+while (cc < ccend)
+  {
+  switch(*cc)
+    {
+    case OP_SET_SOM:
+    common->has_set_som = TRUE;
+    cc += 1;
+    break;
+
+    case OP_REF:
+    case OP_REFI:
+    common->optimized_cbracket[GET2(cc, 1)] = 0;
+    cc += 1 + IMM2_SIZE;
+    break;
+
+    case OP_CBRAPOS:
+    case OP_SCBRAPOS:
+    common->optimized_cbracket[GET2(cc, 1 + LINK_SIZE)] = 0;
+    cc += 1 + LINK_SIZE + IMM2_SIZE;
+    break;
+
+    case OP_COND:
+    case OP_SCOND:
+    /* Only AUTO_CALLOUT can insert this opcode. We do
+       not intend to support this case. */
+    if (cc[1 + LINK_SIZE] == OP_CALLOUT)
+      return FALSE;
+    cc += 1 + LINK_SIZE;
+    break;
+
+    case OP_CREF:
+    i = GET2(cc, 1);
+    common->optimized_cbracket[i] = 0;
+    cc += 1 + IMM2_SIZE;
+    break;
+
+    case OP_NCREF:
+    cbra_index = GET2(cc, 1);
+    name = (pcre_uchar *)common->name_table;
+    name2 = name;
+    for (i = 0; i < common->name_count; i++)
+      {
+      if (GET2(name, 0) == cbra_index) break;
+      name += common->name_entry_size;
+      }
+    SLJIT_ASSERT(i != common->name_count);
+
+    for (i = 0; i < common->name_count; i++)
+      {
+      if (STRCMP_UC_UC(name2 + IMM2_SIZE, name + IMM2_SIZE) == 0)
+        common->optimized_cbracket[GET2(name2, 0)] = 0;
+      name2 += common->name_entry_size;
+      }
+    cc += 1 + IMM2_SIZE;
+    break;
+
+    case OP_RECURSE:
+    /* Set its value only once. */
+    if (common->recursive_head_ptr == 0)
+      {
+      common->recursive_head_ptr = common->ovector_start;
+      common->ovector_start += sizeof(sljit_sw);
+      }
+    cc += 1 + LINK_SIZE;
+    break;
+
+    case OP_CALLOUT:
+    if (common->capture_last_ptr == 0)
+      {
+      common->capture_last_ptr = common->ovector_start;
+      common->ovector_start += sizeof(sljit_sw);
+      }
+    cc += 2 + 2 * LINK_SIZE;
+    break;
+
+    case OP_THEN_ARG:
+    common->has_then = TRUE;
+    common->control_head_ptr = 1;
+    /* Fall through. */
+
+    case OP_PRUNE_ARG:
+    common->needs_start_ptr = TRUE;
+    /* Fall through. */
+
+    case OP_MARK:
+    if (common->mark_ptr == 0)
+      {
+      common->mark_ptr = common->ovector_start;
+      common->ovector_start += sizeof(sljit_sw);
+      }
+    cc += 1 + 2 + cc[1];
+    break;
+
+    case OP_THEN:
+    common->has_then = TRUE;
+    common->control_head_ptr = 1;
+    /* Fall through. */
+
+    case OP_PRUNE:
+    case OP_SKIP:
+    common->needs_start_ptr = TRUE;
+    cc += 1;
+    break;
+
+    case OP_SKIP_ARG:
+    common->control_head_ptr = 1;
+    common->has_skip_arg = TRUE;
+    cc += 1 + 2 + cc[1];
+    break;
+
+    default:
+    cc = next_opcode(common, cc);
+    if (cc == NULL)
+      return FALSE;
+    break;
+    }
+  }
+return TRUE;
+}
+
+static int get_class_iterator_size(pcre_uchar *cc)
+{
+switch(*cc)
+  {
+  case OP_CRSTAR:
+  case OP_CRPLUS:
+  return 2;
+
+  case OP_CRMINSTAR:
+  case OP_CRMINPLUS:
+  case OP_CRQUERY:
+  case OP_CRMINQUERY:
+  return 1;
+
+  case OP_CRRANGE:
+  case OP_CRMINRANGE:
+  if (GET2(cc, 1) == GET2(cc, 1 + IMM2_SIZE))
+    return 0;
+  return 2;
+
+  default:
+  return 0;
+  }
+}
+
+static BOOL detect_repeat(compiler_common *common, pcre_uchar *begin)
+{
+pcre_uchar *end = bracketend(begin);
+pcre_uchar *next;
+pcre_uchar *next_end;
+pcre_uchar *max_end;
+pcre_uchar type;
+sljit_sw length = end - begin;
+int min, max, i;
+
+/* Detect fixed iterations first. */
+if (end[-(1 + LINK_SIZE)] != OP_KET)
+  return FALSE;
+
+/* Already detected repeat. */
+if (common->private_data_ptrs[end - common->start - LINK_SIZE] != 0)
+  return TRUE;
+
+next = end;
+min = 1;
+while (1)
+  {
+  if (*next != *begin)
+    break;
+  next_end = bracketend(next);
+  if (next_end - next != length || memcmp(begin, next, IN_UCHARS(length)) != 0)
+    break;
+  next = next_end;
+  min++;
+  }
+
+if (min == 2)
+  return FALSE;
+
+max = 0;
+max_end = next;
+if (*next == OP_BRAZERO || *next == OP_BRAMINZERO)
+  {
+  type = *next;
+  while (1)
+    {
+    if (next[0] != type || next[1] != OP_BRA || next[2 + LINK_SIZE] != *begin)
+      break;
+    next_end = bracketend(next + 2 + LINK_SIZE);
+    if (next_end - next != (length + 2 + LINK_SIZE) || memcmp(begin, next + 2 + LINK_SIZE, IN_UCHARS(length)) != 0)
+      break;
+    next = next_end;
+    max++;
+    }
+
+  if (next[0] == type && next[1] == *begin && max >= 1)
+    {
+    next_end = bracketend(next + 1);
+    if (next_end - next == (length + 1) && memcmp(begin, next + 1, IN_UCHARS(length)) == 0)
+      {
+      for (i = 0; i < max; i++, next_end += 1 + LINK_SIZE)
+        if (*next_end != OP_KET)
+          break;
+
+      if (i == max)
+        {
+        common->private_data_ptrs[max_end - common->start - LINK_SIZE] = next_end - max_end;
+        common->private_data_ptrs[max_end - common->start - LINK_SIZE + 1] = (type == OP_BRAZERO) ? OP_UPTO : OP_MINUPTO;
+        /* +2 the original and the last. */
+        common->private_data_ptrs[max_end - common->start - LINK_SIZE + 2] = max + 2;
+        if (min == 1)
+          return TRUE;
+        min--;
+        max_end -= (1 + LINK_SIZE) + GET(max_end, -LINK_SIZE);
+        }
+      }
+    }
+  }
+
+if (min >= 3)
+  {
+  common->private_data_ptrs[end - common->start - LINK_SIZE] = max_end - end;
+  common->private_data_ptrs[end - common->start - LINK_SIZE + 1] = OP_EXACT;
+  common->private_data_ptrs[end - common->start - LINK_SIZE + 2] = min;
+  return TRUE;
+  }
+
+return FALSE;
+}
+
+#define CASE_ITERATOR_PRIVATE_DATA_1 \
+    case OP_MINSTAR: \
+    case OP_MINPLUS: \
+    case OP_QUERY: \
+    case OP_MINQUERY: \
+    case OP_MINSTARI: \
+    case OP_MINPLUSI: \
+    case OP_QUERYI: \
+    case OP_MINQUERYI: \
+    case OP_NOTMINSTAR: \
+    case OP_NOTMINPLUS: \
+    case OP_NOTQUERY: \
+    case OP_NOTMINQUERY: \
+    case OP_NOTMINSTARI: \
+    case OP_NOTMINPLUSI: \
+    case OP_NOTQUERYI: \
+    case OP_NOTMINQUERYI:
+
+#define CASE_ITERATOR_PRIVATE_DATA_2A \
+    case OP_STAR: \
+    case OP_PLUS: \
+    case OP_STARI: \
+    case OP_PLUSI: \
+    case OP_NOTSTAR: \
+    case OP_NOTPLUS: \
+    case OP_NOTSTARI: \
+    case OP_NOTPLUSI:
+
+#define CASE_ITERATOR_PRIVATE_DATA_2B \
+    case OP_UPTO: \
+    case OP_MINUPTO: \
+    case OP_UPTOI: \
+    case OP_MINUPTOI: \
+    case OP_NOTUPTO: \
+    case OP_NOTMINUPTO: \
+    case OP_NOTUPTOI: \
+    case OP_NOTMINUPTOI:
+
+#define CASE_ITERATOR_TYPE_PRIVATE_DATA_1 \
+    case OP_TYPEMINSTAR: \
+    case OP_TYPEMINPLUS: \
+    case OP_TYPEQUERY: \
+    case OP_TYPEMINQUERY:
+
+#define CASE_ITERATOR_TYPE_PRIVATE_DATA_2A \
+    case OP_TYPESTAR: \
+    case OP_TYPEPLUS:
+
+#define CASE_ITERATOR_TYPE_PRIVATE_DATA_2B \
+    case OP_TYPEUPTO: \
+    case OP_TYPEMINUPTO:
+
+static void set_private_data_ptrs(compiler_common *common, int *private_data_start, pcre_uchar *ccend)
+{
+pcre_uchar *cc = common->start;
+pcre_uchar *alternative;
+pcre_uchar *end = NULL;
+int private_data_ptr = *private_data_start;
+int space, size, bracketlen;
+
+while (cc < ccend)
+  {
+  space = 0;
+  size = 0;
+  bracketlen = 0;
+  if (private_data_ptr > SLJIT_MAX_LOCAL_SIZE)
+    return;
+
+  if (*cc == OP_ONCE || *cc == OP_ONCE_NC || *cc == OP_BRA || *cc == OP_CBRA || *cc == OP_COND)
+    if (detect_repeat(common, cc))
+      {
+      /* These brackets are converted to repeats, so no global
+      based single character repeat is allowed. */
+      if (cc >= end)
+        end = bracketend(cc);
+      }
+
+  switch(*cc)
+    {
+    case OP_KET:
+    if (common->private_data_ptrs[cc + 1 - common->start] != 0)
+      {
+      common->private_data_ptrs[cc - common->start] = private_data_ptr;
+      private_data_ptr += sizeof(sljit_sw);
+      cc += common->private_data_ptrs[cc + 1 - common->start];
+      }
+    cc += 1 + LINK_SIZE;
+    break;
+
+    case OP_ASSERT:
+    case OP_ASSERT_NOT:
+    case OP_ASSERTBACK:
+    case OP_ASSERTBACK_NOT:
+    case OP_ONCE:
+    case OP_ONCE_NC:
+    case OP_BRAPOS:
+    case OP_SBRA:
+    case OP_SBRAPOS:
+    case OP_SCOND:
+    common->private_data_ptrs[cc - common->start] = private_data_ptr;
+    private_data_ptr += sizeof(sljit_sw);
+    bracketlen = 1 + LINK_SIZE;
+    break;
+
+    case OP_CBRAPOS:
+    case OP_SCBRAPOS:
+    common->private_data_ptrs[cc - common->start] = private_data_ptr;
+    private_data_ptr += sizeof(sljit_sw);
+    bracketlen = 1 + LINK_SIZE + IMM2_SIZE;
+    break;
+
+    case OP_COND:
+    /* Might be a hidden SCOND. */
+    alternative = cc + GET(cc, 1);
+    if (*alternative == OP_KETRMAX || *alternative == OP_KETRMIN)
+      {
+      common->private_data_ptrs[cc - common->start] = private_data_ptr;
+      private_data_ptr += sizeof(sljit_sw);
+      }
+    bracketlen = 1 + LINK_SIZE;
+    break;
+
+    case OP_BRA:
+    bracketlen = 1 + LINK_SIZE;
+    break;
+
+    case OP_CBRA:
+    case OP_SCBRA:
+    bracketlen = 1 + LINK_SIZE + IMM2_SIZE;
+    break;
+
+    CASE_ITERATOR_PRIVATE_DATA_1
+    space = 1;
+    size = -2;
+    break;
+
+    CASE_ITERATOR_PRIVATE_DATA_2A
+    space = 2;
+    size = -2;
+    break;
+
+    CASE_ITERATOR_PRIVATE_DATA_2B
+    space = 2;
+    size = -(2 + IMM2_SIZE);
+    break;
+
+    CASE_ITERATOR_TYPE_PRIVATE_DATA_1
+    space = 1;
+    size = 1;
+    break;
+
+    CASE_ITERATOR_TYPE_PRIVATE_DATA_2A
+    if (cc[1] != OP_ANYNL && cc[1] != OP_EXTUNI)
+      space = 2;
+    size = 1;
+    break;
+
+    CASE_ITERATOR_TYPE_PRIVATE_DATA_2B
+    if (cc[1 + IMM2_SIZE] != OP_ANYNL && cc[1 + IMM2_SIZE] != OP_EXTUNI)
+      space = 2;
+    size = 1 + IMM2_SIZE;
+    break;
+
+    case OP_CLASS:
+    case OP_NCLASS:
+    size += 1 + 32 / sizeof(pcre_uchar);
+    space = get_class_iterator_size(cc + size);
+    break;
+
+#if defined SUPPORT_UTF || !defined COMPILE_PCRE8
+    case OP_XCLASS:
+    size = GET(cc, 1);
+    space = get_class_iterator_size(cc + size);
+    break;
+#endif
+
+    default:
+    cc = next_opcode(common, cc);
+    SLJIT_ASSERT(cc != NULL);
+    break;
+    }
+
+  /* Character iterators, which are not inside a repeated bracket,
+     gets a private slot instead of allocating it on the stack. */
+  if (space > 0 && cc >= end)
+    {
+    common->private_data_ptrs[cc - common->start] = private_data_ptr;
+    private_data_ptr += sizeof(sljit_sw) * space;
+    }
+
+  if (size != 0)
+    {
+    if (size < 0)
+      {
+      cc += -size;
+#ifdef SUPPORT_UTF
+      if (common->utf && HAS_EXTRALEN(cc[-1])) cc += GET_EXTRALEN(cc[-1]);
+#endif
+      }
+    else
+      cc += size;
+    }
+
+  if (bracketlen > 0)
+    {
+    if (cc >= end)
+      {
+      end = bracketend(cc);
+      if (end[-1 - LINK_SIZE] == OP_KET)
+        end = NULL;
+      }
+    cc += bracketlen;
+    }
+  }
+*private_data_start = private_data_ptr;
+}
+
+/* Returns with a frame_types (always < 0) if no need for frame. */
+static int get_framesize(compiler_common *common, pcre_uchar *cc, pcre_uchar *ccend, BOOL recursive, BOOL* needs_control_head)
+{
+int length = 0;
+int possessive = 0;
+BOOL stack_restore = FALSE;
+BOOL setsom_found = recursive;
+BOOL setmark_found = recursive;
+/* The last capture is a local variable even for recursions. */
+BOOL capture_last_found = FALSE;
+
+#if defined DEBUG_FORCE_CONTROL_HEAD && DEBUG_FORCE_CONTROL_HEAD
+SLJIT_ASSERT(common->control_head_ptr != 0);
+*needs_control_head = TRUE;
+#else
+*needs_control_head = FALSE;
+#endif
+
+if (ccend == NULL)
+  {
+  ccend = bracketend(cc) - (1 + LINK_SIZE);
+  if (!recursive && (*cc == OP_CBRAPOS || *cc == OP_SCBRAPOS))
+    {
+    possessive = length = (common->capture_last_ptr != 0) ? 5 : 3;
+    /* This is correct regardless of common->capture_last_ptr. */
+    capture_last_found = TRUE;
+    }
+  cc = next_opcode(common, cc);
+  }
+
+SLJIT_ASSERT(cc != NULL);
+while (cc < ccend)
+  switch(*cc)
+    {
+    case OP_SET_SOM:
+    SLJIT_ASSERT(common->has_set_som);
+    stack_restore = TRUE;
+    if (!setsom_found)
+      {
+      length += 2;
+      setsom_found = TRUE;
+      }
+    cc += 1;
+    break;
+
+    case OP_MARK:
+    case OP_PRUNE_ARG:
+    case OP_THEN_ARG:
+    SLJIT_ASSERT(common->mark_ptr != 0);
+    stack_restore = TRUE;
+    if (!setmark_found)
+      {
+      length += 2;
+      setmark_found = TRUE;
+      }
+    if (common->control_head_ptr != 0)
+      *needs_control_head = TRUE;
+    cc += 1 + 2 + cc[1];
+    break;
+
+    case OP_RECURSE:
+    stack_restore = TRUE;
+    if (common->has_set_som && !setsom_found)
+      {
+      length += 2;
+      setsom_found = TRUE;
+      }
+    if (common->mark_ptr != 0 && !setmark_found)
+      {
+      length += 2;
+      setmark_found = TRUE;
+      }
+    if (common->capture_last_ptr != 0 && !capture_last_found)
+      {
+      length += 2;
+      capture_last_found = TRUE;
+      }
+    cc += 1 + LINK_SIZE;
+    break;
+
+    case OP_CBRA:
+    case OP_CBRAPOS:
+    case OP_SCBRA:
+    case OP_SCBRAPOS:
+    stack_restore = TRUE;
+    if (common->capture_last_ptr != 0 && !capture_last_found)
+      {
+      length += 2;
+      capture_last_found = TRUE;
+      }
+    length += 3;
+    cc += 1 + LINK_SIZE + IMM2_SIZE;
+    break;
+
+    default:
+    stack_restore = TRUE;
+    /* Fall through. */
+
+    case OP_NOT_WORD_BOUNDARY:
+    case OP_WORD_BOUNDARY:
+    case OP_NOT_DIGIT:
+    case OP_DIGIT:
+    case OP_NOT_WHITESPACE:
+    case OP_WHITESPACE:
+    case OP_NOT_WORDCHAR:
+    case OP_WORDCHAR:
+    case OP_ANY:
+    case OP_ALLANY:
+    case OP_ANYBYTE:
+    case OP_NOTPROP:
+    case OP_PROP:
+    case OP_ANYNL:
+    case OP_NOT_HSPACE:
+    case OP_HSPACE:
+    case OP_NOT_VSPACE:
+    case OP_VSPACE:
+    case OP_EXTUNI:
+    case OP_EODN:
+    case OP_EOD:
+    case OP_CIRC:
+    case OP_CIRCM:
+    case OP_DOLL:
+    case OP_DOLLM:
+    case OP_CHAR:
+    case OP_CHARI:
+    case OP_NOT:
+    case OP_NOTI:
+
+    case OP_EXACT:
+    case OP_POSSTAR:
+    case OP_POSPLUS:
+    case OP_POSQUERY:
+    case OP_POSUPTO:
+
+    case OP_EXACTI:
+    case OP_POSSTARI:
+    case OP_POSPLUSI:
+    case OP_POSQUERYI:
+    case OP_POSUPTOI:
+
+    case OP_NOTEXACT:
+    case OP_NOTPOSSTAR:
+    case OP_NOTPOSPLUS:
+    case OP_NOTPOSQUERY:
+    case OP_NOTPOSUPTO:
+
+    case OP_NOTEXACTI:
+    case OP_NOTPOSSTARI:
+    case OP_NOTPOSPLUSI:
+    case OP_NOTPOSQUERYI:
+    case OP_NOTPOSUPTOI:
+
+    case OP_TYPEEXACT:
+    case OP_TYPEPOSSTAR:
+    case OP_TYPEPOSPLUS:
+    case OP_TYPEPOSQUERY:
+    case OP_TYPEPOSUPTO:
+
+    case OP_CLASS:
+    case OP_NCLASS:
+    case OP_XCLASS:
+
+    cc = next_opcode(common, cc);
+    SLJIT_ASSERT(cc != NULL);
+    break;
+    }
+
+/* Possessive quantifiers can use a special case. */
+if (SLJIT_UNLIKELY(possessive == length))
+  return stack_restore ? no_frame : no_stack;
+
+if (length > 0)
+  return length + 1;
+return stack_restore ? no_frame : no_stack;
+}
+
+static void init_frame(compiler_common *common, pcre_uchar *cc, pcre_uchar *ccend, int stackpos, int stacktop, BOOL recursive)
+{
+DEFINE_COMPILER;
+BOOL setsom_found = recursive;
+BOOL setmark_found = recursive;
+/* The last capture is a local variable even for recursions. */
+BOOL capture_last_found = FALSE;
+int offset;
+
+/* >= 1 + shortest item size (2) */
+SLJIT_UNUSED_ARG(stacktop);
+SLJIT_ASSERT(stackpos >= stacktop + 2);
+
+stackpos = STACK(stackpos);
+if (ccend == NULL)
+  {
+  ccend = bracketend(cc) - (1 + LINK_SIZE);
+  if (recursive || (*cc != OP_CBRAPOS && *cc != OP_SCBRAPOS))
+    cc = next_opcode(common, cc);
+  }
+
+SLJIT_ASSERT(cc != NULL);
+while (cc < ccend)
+  switch(*cc)
+    {
+    case OP_SET_SOM:
+    SLJIT_ASSERT(common->has_set_som);
+    if (!setsom_found)
+      {
+      OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(0));
+      OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), stackpos, SLJIT_IMM, -OVECTOR(0));
+      stackpos += (int)sizeof(sljit_sw);
+      OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), stackpos, TMP1, 0);
+      stackpos += (int)sizeof(sljit_sw);
+      setsom_found = TRUE;
+      }
+    cc += 1;
+    break;
+
+    case OP_MARK:
+    case OP_PRUNE_ARG:
+    case OP_THEN_ARG:
+    SLJIT_ASSERT(common->mark_ptr != 0);
+    if (!setmark_found)
+      {
+      OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), common->mark_ptr);
+      OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), stackpos, SLJIT_IMM, -common->mark_ptr);
+      stackpos += (int)sizeof(sljit_sw);
+      OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), stackpos, TMP1, 0);
+      stackpos += (int)sizeof(sljit_sw);
+      setmark_found = TRUE;
+      }
+    cc += 1 + 2 + cc[1];
+    break;
+
+    case OP_RECURSE:
+    if (common->has_set_som && !setsom_found)
+      {
+      OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(0));
+      OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), stackpos, SLJIT_IMM, -OVECTOR(0));
+      stackpos += (int)sizeof(sljit_sw);
+      OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), stackpos, TMP1, 0);
+      stackpos += (int)sizeof(sljit_sw);
+      setsom_found = TRUE;
+      }
+    if (common->mark_ptr != 0 && !setmark_found)
+      {
+      OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), common->mark_ptr);
+      OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), stackpos, SLJIT_IMM, -common->mark_ptr);
+      stackpos += (int)sizeof(sljit_sw);
+      OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), stackpos, TMP1, 0);
+      stackpos += (int)sizeof(sljit_sw);
+      setmark_found = TRUE;
+      }
+    if (common->capture_last_ptr != 0 && !capture_last_found)
+      {
+      OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), common->capture_last_ptr);
+      OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), stackpos, SLJIT_IMM, -common->capture_last_ptr);
+      stackpos += (int)sizeof(sljit_sw);
+      OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), stackpos, TMP1, 0);
+      stackpos += (int)sizeof(sljit_sw);
+      capture_last_found = TRUE;
+      }
+    cc += 1 + LINK_SIZE;
+    break;
+
+    case OP_CBRA:
+    case OP_CBRAPOS:
+    case OP_SCBRA:
+    case OP_SCBRAPOS:
+    if (common->capture_last_ptr != 0 && !capture_last_found)
+      {
+      OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), common->capture_last_ptr);
+      OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), stackpos, SLJIT_IMM, -common->capture_last_ptr);
+      stackpos += (int)sizeof(sljit_sw);
+      OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), stackpos, TMP1, 0);
+      stackpos += (int)sizeof(sljit_sw);
+      capture_last_found = TRUE;
+      }
+    offset = (GET2(cc, 1 + LINK_SIZE)) << 1;
+    OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), stackpos, SLJIT_IMM, OVECTOR(offset));
+    stackpos += (int)sizeof(sljit_sw);
+    OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(offset));
+    OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(offset + 1));
+    OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), stackpos, TMP1, 0);
+    stackpos += (int)sizeof(sljit_sw);
+    OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), stackpos, TMP2, 0);
+    stackpos += (int)sizeof(sljit_sw);
+
+    cc += 1 + LINK_SIZE + IMM2_SIZE;
+    break;
+
+    default:
+    cc = next_opcode(common, cc);
+    SLJIT_ASSERT(cc != NULL);
+    break;
+    }
+
+OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), stackpos, SLJIT_IMM, 0);
+SLJIT_ASSERT(stackpos == STACK(stacktop));
+}
+
+static SLJIT_INLINE int get_private_data_copy_length(compiler_common *common, pcre_uchar *cc, pcre_uchar *ccend, BOOL needs_control_head)
+{
+int private_data_length = needs_control_head ? 3 : 2;
+int size;
+pcre_uchar *alternative;
+/* Calculate the sum of the private machine words. */
+while (cc < ccend)
+  {
+  size = 0;
+  switch(*cc)
+    {
+    case OP_KET:
+    if (PRIVATE_DATA(cc) != 0)
+      private_data_length++;
+    cc += 1 + LINK_SIZE;
+    break;
+
+    case OP_ASSERT:
+    case OP_ASSERT_NOT:
+    case OP_ASSERTBACK:
+    case OP_ASSERTBACK_NOT:
+    case OP_ONCE:
+    case OP_ONCE_NC:
+    case OP_BRAPOS:
+    case OP_SBRA:
+    case OP_SBRAPOS:
+    case OP_SCOND:
+    private_data_length++;
+    cc += 1 + LINK_SIZE;
+    break;
+
+    case OP_CBRA:
+    case OP_SCBRA:
+    if (common->optimized_cbracket[GET2(cc, 1 + LINK_SIZE)] == 0)
+      private_data_length++;
+    cc += 1 + LINK_SIZE + IMM2_SIZE;
+    break;
+
+    case OP_CBRAPOS:
+    case OP_SCBRAPOS:
+    private_data_length += 2;
+    cc += 1 + LINK_SIZE + IMM2_SIZE;
+    break;
+
+    case OP_COND:
+    /* Might be a hidden SCOND. */
+    alternative = cc + GET(cc, 1);
+    if (*alternative == OP_KETRMAX || *alternative == OP_KETRMIN)
+      private_data_length++;
+    cc += 1 + LINK_SIZE;
+    break;
+
+    CASE_ITERATOR_PRIVATE_DATA_1
+    if (PRIVATE_DATA(cc))
+      private_data_length++;
+    cc += 2;
+#ifdef SUPPORT_UTF
+    if (common->utf && HAS_EXTRALEN(cc[-1])) cc += GET_EXTRALEN(cc[-1]);
+#endif
+    break;
+
+    CASE_ITERATOR_PRIVATE_DATA_2A
+    if (PRIVATE_DATA(cc))
+      private_data_length += 2;
+    cc += 2;
+#ifdef SUPPORT_UTF
+    if (common->utf && HAS_EXTRALEN(cc[-1])) cc += GET_EXTRALEN(cc[-1]);
+#endif
+    break;
+
+    CASE_ITERATOR_PRIVATE_DATA_2B
+    if (PRIVATE_DATA(cc))
+      private_data_length += 2;
+    cc += 2 + IMM2_SIZE;
+#ifdef SUPPORT_UTF
+    if (common->utf && HAS_EXTRALEN(cc[-1])) cc += GET_EXTRALEN(cc[-1]);
+#endif
+    break;
+
+    CASE_ITERATOR_TYPE_PRIVATE_DATA_1
+    if (PRIVATE_DATA(cc))
+      private_data_length++;
+    cc += 1;
+    break;
+
+    CASE_ITERATOR_TYPE_PRIVATE_DATA_2A
+    if (PRIVATE_DATA(cc))
+      private_data_length += 2;
+    cc += 1;
+    break;
+
+    CASE_ITERATOR_TYPE_PRIVATE_DATA_2B
+    if (PRIVATE_DATA(cc))
+      private_data_length += 2;
+    cc += 1 + IMM2_SIZE;
+    break;
+
+    case OP_CLASS:
+    case OP_NCLASS:
+#if defined SUPPORT_UTF || !defined COMPILE_PCRE8
+    case OP_XCLASS:
+    size = (*cc == OP_XCLASS) ? GET(cc, 1) : 1 + 32 / (int)sizeof(pcre_uchar);
+#else
+    size = 1 + 32 / (int)sizeof(pcre_uchar);
+#endif
+    if (PRIVATE_DATA(cc))
+      private_data_length += get_class_iterator_size(cc + size);
+    cc += size;
+    break;
+
+    default:
+    cc = next_opcode(common, cc);
+    SLJIT_ASSERT(cc != NULL);
+    break;
+    }
+  }
+SLJIT_ASSERT(cc == ccend);
+return private_data_length;
+}
+
+static void copy_private_data(compiler_common *common, pcre_uchar *cc, pcre_uchar *ccend,
+  BOOL save, int stackptr, int stacktop, BOOL needs_control_head)
+{
+DEFINE_COMPILER;
+int srcw[2];
+int count, size;
+BOOL tmp1next = TRUE;
+BOOL tmp1empty = TRUE;
+BOOL tmp2empty = TRUE;
+pcre_uchar *alternative;
+enum {
+  start,
+  loop,
+  end
+} status;
+
+status = save ? start : loop;
+stackptr = STACK(stackptr - 2);
+stacktop = STACK(stacktop - 1);
+
+if (!save)
+  {
+  stackptr += (needs_control_head ? 2 : 1) * sizeof(sljit_sw);
+  if (stackptr < stacktop)
+    {
+    OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(STACK_TOP), stackptr);
+    stackptr += sizeof(sljit_sw);
+    tmp1empty = FALSE;
+    }
+  if (stackptr < stacktop)
+    {
+    OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(STACK_TOP), stackptr);
+    stackptr += sizeof(sljit_sw);
+    tmp2empty = FALSE;
+    }
+  /* The tmp1next must be TRUE in either way. */
+  }
+
+do
+  {
+  count = 0;
+  switch(status)
+    {
+    case start:
+    SLJIT_ASSERT(save && common->recursive_head_ptr != 0);
+    count = 1;
+    srcw[0] = common->recursive_head_ptr;
+    if (needs_control_head)
+      {
+      SLJIT_ASSERT(common->control_head_ptr != 0);
+      count = 2;
+      srcw[1] = common->control_head_ptr;
+      }
+    status = loop;
+    break;
+
+    case loop:
+    if (cc >= ccend)
+      {
+      status = end;
+      break;
+      }
+
+    switch(*cc)
+      {
+      case OP_KET:
+      if (PRIVATE_DATA(cc) != 0)
+        {
+        count = 1;
+        srcw[0] = PRIVATE_DATA(cc);
+        }
+      cc += 1 + LINK_SIZE;
+      break;
+
+      case OP_ASSERT:
+      case OP_ASSERT_NOT:
+      case OP_ASSERTBACK:
+      case OP_ASSERTBACK_NOT:
+      case OP_ONCE:
+      case OP_ONCE_NC:
+      case OP_BRAPOS:
+      case OP_SBRA:
+      case OP_SBRAPOS:
+      case OP_SCOND:
+      count = 1;
+      srcw[0] = PRIVATE_DATA(cc);
+      SLJIT_ASSERT(srcw[0] != 0);
+      cc += 1 + LINK_SIZE;
+      break;
+
+      case OP_CBRA:
+      case OP_SCBRA:
+      if (common->optimized_cbracket[GET2(cc, 1 + LINK_SIZE)] == 0)
+        {
+        count = 1;
+        srcw[0] = OVECTOR_PRIV(GET2(cc, 1 + LINK_SIZE));
+        }
+      cc += 1 + LINK_SIZE + IMM2_SIZE;
+      break;
+
+      case OP_CBRAPOS:
+      case OP_SCBRAPOS:
+      count = 2;
+      srcw[0] = PRIVATE_DATA(cc);
+      srcw[1] = OVECTOR_PRIV(GET2(cc, 1 + LINK_SIZE));
+      SLJIT_ASSERT(srcw[0] != 0 && srcw[1] != 0);
+      cc += 1 + LINK_SIZE + IMM2_SIZE;
+      break;
+
+      case OP_COND:
+      /* Might be a hidden SCOND. */
+      alternative = cc + GET(cc, 1);
+      if (*alternative == OP_KETRMAX || *alternative == OP_KETRMIN)
+        {
+        count = 1;
+        srcw[0] = PRIVATE_DATA(cc);
+        SLJIT_ASSERT(srcw[0] != 0);
+        }
+      cc += 1 + LINK_SIZE;
+      break;
+
+      CASE_ITERATOR_PRIVATE_DATA_1
+      if (PRIVATE_DATA(cc))
+        {
+        count = 1;
+        srcw[0] = PRIVATE_DATA(cc);
+        }
+      cc += 2;
+#ifdef SUPPORT_UTF
+      if (common->utf && HAS_EXTRALEN(cc[-1])) cc += GET_EXTRALEN(cc[-1]);
+#endif
+      break;
+
+      CASE_ITERATOR_PRIVATE_DATA_2A
+      if (PRIVATE_DATA(cc))
+        {
+        count = 2;
+        srcw[0] = PRIVATE_DATA(cc);
+        srcw[1] = PRIVATE_DATA(cc) + sizeof(sljit_sw);
+        }
+      cc += 2;
+#ifdef SUPPORT_UTF
+      if (common->utf && HAS_EXTRALEN(cc[-1])) cc += GET_EXTRALEN(cc[-1]);
+#endif
+      break;
+
+      CASE_ITERATOR_PRIVATE_DATA_2B
+      if (PRIVATE_DATA(cc))
+        {
+        count = 2;
+        srcw[0] = PRIVATE_DATA(cc);
+        srcw[1] = PRIVATE_DATA(cc) + sizeof(sljit_sw);
+        }
+      cc += 2 + IMM2_SIZE;
+#ifdef SUPPORT_UTF
+      if (common->utf && HAS_EXTRALEN(cc[-1])) cc += GET_EXTRALEN(cc[-1]);
+#endif
+      break;
+
+      CASE_ITERATOR_TYPE_PRIVATE_DATA_1
+      if (PRIVATE_DATA(cc))
+        {
+        count = 1;
+        srcw[0] = PRIVATE_DATA(cc);
+        }
+      cc += 1;
+      break;
+
+      CASE_ITERATOR_TYPE_PRIVATE_DATA_2A
+      if (PRIVATE_DATA(cc))
+        {
+        count = 2;
+        srcw[0] = PRIVATE_DATA(cc);
+        srcw[1] = srcw[0] + sizeof(sljit_sw);
+        }
+      cc += 1;
+      break;
+
+      CASE_ITERATOR_TYPE_PRIVATE_DATA_2B
+      if (PRIVATE_DATA(cc))
+        {
+        count = 2;
+        srcw[0] = PRIVATE_DATA(cc);
+        srcw[1] = srcw[0] + sizeof(sljit_sw);
+        }
+      cc += 1 + IMM2_SIZE;
+      break;
+
+      case OP_CLASS:
+      case OP_NCLASS:
+#if defined SUPPORT_UTF || !defined COMPILE_PCRE8
+      case OP_XCLASS:
+      size = (*cc == OP_XCLASS) ? GET(cc, 1) : 1 + 32 / (int)sizeof(pcre_uchar);
+#else
+      size = 1 + 32 / (int)sizeof(pcre_uchar);
+#endif
+      if (PRIVATE_DATA(cc))
+        switch(get_class_iterator_size(cc + size))
+          {
+          case 1:
+          count = 1;
+          srcw[0] = PRIVATE_DATA(cc);
+          break;
+
+          case 2:
+          count = 2;
+          srcw[0] = PRIVATE_DATA(cc);
+          srcw[1] = srcw[0] + sizeof(sljit_sw);
+          break;
+
+          default:
+          SLJIT_ASSERT_STOP();
+          break;
+          }
+      cc += size;
+      break;
+
+      default:
+      cc = next_opcode(common, cc);
+      SLJIT_ASSERT(cc != NULL);
+      break;
+      }
+    break;
+
+    case end:
+    SLJIT_ASSERT_STOP();
+    break;
+    }
+
+  while (count > 0)
+    {
+    count--;
+    if (save)
+      {
+      if (tmp1next)
+        {
+        if (!tmp1empty)
+          {
+          OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), stackptr, TMP1, 0);
+          stackptr += sizeof(sljit_sw);
+          }
+        OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), srcw[count]);
+        tmp1empty = FALSE;
+        tmp1next = FALSE;
+        }
+      else
+        {
+        if (!tmp2empty)
+          {
+          OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), stackptr, TMP2, 0);
+          stackptr += sizeof(sljit_sw);
+          }
+        OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), srcw[count]);
+        tmp2empty = FALSE;
+        tmp1next = TRUE;
+        }
+      }
+    else
+      {
+      if (tmp1next)
+        {
+        SLJIT_ASSERT(!tmp1empty);
+        OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), srcw[count], TMP1, 0);
+        tmp1empty = stackptr >= stacktop;
+        if (!tmp1empty)
+          {
+          OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(STACK_TOP), stackptr);
+          stackptr += sizeof(sljit_sw);
+          }
+        tmp1next = FALSE;
+        }
+      else
+        {
+        SLJIT_ASSERT(!tmp2empty);
+        OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), srcw[count], TMP2, 0);
+        tmp2empty = stackptr >= stacktop;
+        if (!tmp2empty)
+          {
+          OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(STACK_TOP), stackptr);
+          stackptr += sizeof(sljit_sw);
+          }
+        tmp1next = TRUE;
+        }
+      }
+    }
+  }
+while (status != end);
+
+if (save)
+  {
+  if (tmp1next)
+    {
+    if (!tmp1empty)
+      {
+      OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), stackptr, TMP1, 0);
+      stackptr += sizeof(sljit_sw);
+      }
+    if (!tmp2empty)
+      {
+      OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), stackptr, TMP2, 0);
+      stackptr += sizeof(sljit_sw);
+      }
+    }
+  else
+    {
+    if (!tmp2empty)
+      {
+      OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), stackptr, TMP2, 0);
+      stackptr += sizeof(sljit_sw);
+      }
+    if (!tmp1empty)
+      {
+      OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), stackptr, TMP1, 0);
+      stackptr += sizeof(sljit_sw);
+      }
+    }
+  }
+SLJIT_ASSERT(cc == ccend && stackptr == stacktop && (save || (tmp1empty && tmp2empty)));
+}
+
+static SLJIT_INLINE pcre_uchar *set_then_offsets(compiler_common *common, pcre_uchar *cc, pcre_uint8 *current_offset)
+{
+pcre_uchar *end = bracketend(cc);
+BOOL has_alternatives = cc[GET(cc, 1)] == OP_ALT;
+
+/* Assert captures then. */
+if (*cc >= OP_ASSERT && *cc <= OP_ASSERTBACK_NOT)
+  current_offset = NULL;
+/* Conditional block does not. */
+if (*cc == OP_COND || *cc == OP_SCOND)
+  has_alternatives = FALSE;
+
+cc = next_opcode(common, cc);
+if (has_alternatives)
+  current_offset = common->then_offsets + (cc - common->start);
+
+while (cc < end)
+  {
+  if ((*cc >= OP_ASSERT && *cc <= OP_ASSERTBACK_NOT) || (*cc >= OP_ONCE && *cc <= OP_SCOND))
+    cc = set_then_offsets(common, cc, current_offset);
+  else
+    {
+    if (*cc == OP_ALT && has_alternatives)
+      current_offset = common->then_offsets + (cc + 1 + LINK_SIZE - common->start);
+    if (*cc >= OP_THEN && *cc <= OP_THEN_ARG && current_offset != NULL)
+      *current_offset = 1;
+    cc = next_opcode(common, cc);
+    }
+  }
+
+return end;
+}
+
+#undef CASE_ITERATOR_PRIVATE_DATA_1
+#undef CASE_ITERATOR_PRIVATE_DATA_2A
+#undef CASE_ITERATOR_PRIVATE_DATA_2B
+#undef CASE_ITERATOR_TYPE_PRIVATE_DATA_1
+#undef CASE_ITERATOR_TYPE_PRIVATE_DATA_2A
+#undef CASE_ITERATOR_TYPE_PRIVATE_DATA_2B
+
+static SLJIT_INLINE BOOL is_powerof2(unsigned int value)
+{
+return (value & (value - 1)) == 0;
+}
+
+static SLJIT_INLINE void set_jumps(jump_list *list, struct sljit_label *label)
+{
+while (list)
+  {
+  /* sljit_set_label is clever enough to do nothing
+  if either the jump or the label is NULL. */
+  SET_LABEL(list->jump, label);
+  list = list->next;
+  }
+}
+
+static SLJIT_INLINE void add_jump(struct sljit_compiler *compiler, jump_list **list, struct sljit_jump* jump)
+{
+jump_list *list_item = sljit_alloc_memory(compiler, sizeof(jump_list));
+if (list_item)
+  {
+  list_item->next = *list;
+  list_item->jump = jump;
+  *list = list_item;
+  }
+}
+
+static void add_stub(compiler_common *common, struct sljit_jump *start)
+{
+DEFINE_COMPILER;
+stub_list* list_item = sljit_alloc_memory(compiler, sizeof(stub_list));
+
+if (list_item)
+  {
+  list_item->start = start;
+  list_item->quit = LABEL();
+  list_item->next = common->stubs;
+  common->stubs = list_item;
+  }
+}
+
+static void flush_stubs(compiler_common *common)
+{
+DEFINE_COMPILER;
+stub_list* list_item = common->stubs;
+
+while (list_item)
+  {
+  JUMPHERE(list_item->start);
+  add_jump(compiler, &common->stackalloc, JUMP(SLJIT_FAST_CALL));
+  JUMPTO(SLJIT_JUMP, list_item->quit);
+  list_item = list_item->next;
+  }
+common->stubs = NULL;
+}
+
+static SLJIT_INLINE void count_match(compiler_common *common)
+{
+DEFINE_COMPILER;
+
+OP2(SLJIT_SUB | SLJIT_SET_E, COUNT_MATCH, 0, COUNT_MATCH, 0, SLJIT_IMM, 1);
+add_jump(compiler, &common->calllimit, JUMP(SLJIT_C_ZERO));
+}
+
+static SLJIT_INLINE void allocate_stack(compiler_common *common, int size)
+{
+/* May destroy all locals and registers except TMP2. */
+DEFINE_COMPILER;
+
+OP2(SLJIT_ADD, STACK_TOP, 0, STACK_TOP, 0, SLJIT_IMM, size * sizeof(sljit_sw));
+#ifdef DESTROY_REGISTERS
+OP1(SLJIT_MOV, TMP1, 0, SLJIT_IMM, 12345);
+OP1(SLJIT_MOV, TMP3, 0, TMP1, 0);
+OP1(SLJIT_MOV, RETURN_ADDR, 0, TMP1, 0);
+OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), LOCALS0, TMP1, 0);
+OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), LOCALS1, TMP1, 0);
+#endif
+add_stub(common, CMP(SLJIT_C_GREATER, STACK_TOP, 0, STACK_LIMIT, 0));
+}
+
+static SLJIT_INLINE void free_stack(compiler_common *common, int size)
+{
+DEFINE_COMPILER;
+OP2(SLJIT_SUB, STACK_TOP, 0, STACK_TOP, 0, SLJIT_IMM, size * sizeof(sljit_sw));
+}
+
+static SLJIT_INLINE void reset_ovector(compiler_common *common, int length)
+{
+DEFINE_COMPILER;
+struct sljit_label *loop;
+int i;
+
+/* At this point we can freely use all temporary registers. */
+SLJIT_ASSERT(length > 1);
+/* TMP1 returns with begin - 1. */
+OP2(SLJIT_SUB, SLJIT_SCRATCH_REG1, 0, SLJIT_MEM1(SLJIT_SAVED_REG1), SLJIT_OFFSETOF(jit_arguments, begin), SLJIT_IMM, IN_UCHARS(1));
+if (length < 8)
+  {
+  for (i = 1; i < length; i++)
+    OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(i), SLJIT_SCRATCH_REG1, 0);
+  }
+else
+  {
+  GET_LOCAL_BASE(SLJIT_SCRATCH_REG2, 0, OVECTOR_START);
+  OP1(SLJIT_MOV, SLJIT_SCRATCH_REG3, 0, SLJIT_IMM, length - 1);
+  loop = LABEL();
+  OP1(SLJIT_MOVU, SLJIT_MEM1(SLJIT_SCRATCH_REG2), sizeof(sljit_sw), SLJIT_SCRATCH_REG1, 0);
+  OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_SCRATCH_REG3, 0, SLJIT_SCRATCH_REG3, 0, SLJIT_IMM, 1);
+  JUMPTO(SLJIT_C_NOT_ZERO, loop);
+  }
+}
+
+static SLJIT_INLINE void do_reset_match(compiler_common *common, int length)
+{
+DEFINE_COMPILER;
+struct sljit_label *loop;
+int i;
+
+SLJIT_ASSERT(length > 1);
+/* OVECTOR(1) contains the "string begin - 1" constant. */
+if (length > 2)
+  OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(1));
+if (length < 8)
+  {
+  for (i = 2; i < length; i++)
+    OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(i), TMP1, 0);
+  }
+else
+  {
+  GET_LOCAL_BASE(TMP2, 0, OVECTOR_START + sizeof(sljit_sw));
+  OP1(SLJIT_MOV, STACK_TOP, 0, SLJIT_IMM, length - 2);
+  loop = LABEL();
+  OP1(SLJIT_MOVU, SLJIT_MEM1(TMP2), sizeof(sljit_sw), TMP1, 0);
+  OP2(SLJIT_SUB | SLJIT_SET_E, STACK_TOP, 0, STACK_TOP, 0, SLJIT_IMM, 1);
+  JUMPTO(SLJIT_C_NOT_ZERO, loop);
+  }
+
+OP1(SLJIT_MOV, STACK_TOP, 0, ARGUMENTS, 0);
+if (common->mark_ptr != 0)
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->mark_ptr, SLJIT_IMM, 0);
+if (common->control_head_ptr != 0)
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->control_head_ptr, SLJIT_IMM, 0);
+OP1(SLJIT_MOV, STACK_TOP, 0, SLJIT_MEM1(STACK_TOP), SLJIT_OFFSETOF(jit_arguments, stack));
+OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), common->start_ptr);
+OP1(SLJIT_MOV, STACK_TOP, 0, SLJIT_MEM1(STACK_TOP), SLJIT_OFFSETOF(struct sljit_stack, base));
+}
+
+static sljit_sw SLJIT_CALL do_search_mark(sljit_sw *current, const pcre_uchar *skip_arg)
+{
+while (current != NULL)
+  {
+  switch (current[-2])
+    {
+    case type_then_trap:
+    break;
+
+    case type_mark:
+    if (STRCMP_UC_UC(skip_arg, (pcre_uchar *)current[-3]) == 0)
+      return current[-4];
+    break;
+
+    default:
+    SLJIT_ASSERT_STOP();
+    break;
+    }
+  current = (sljit_sw*)current[-1];
+  }
+return -1;
+}
+
+static SLJIT_INLINE void copy_ovector(compiler_common *common, int topbracket)
+{
+DEFINE_COMPILER;
+struct sljit_label *loop;
+struct sljit_jump *early_quit;
+
+/* At this point we can freely use all registers. */
+OP1(SLJIT_MOV, SLJIT_SAVED_REG3, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(1));
+OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(1), STR_PTR, 0);
+
+OP1(SLJIT_MOV, SLJIT_SCRATCH_REG1, 0, ARGUMENTS, 0);
+if (common->mark_ptr != 0)
+  OP1(SLJIT_MOV, SLJIT_SCRATCH_REG3, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), common->mark_ptr);
+OP1(SLJIT_MOV_SI, SLJIT_SCRATCH_REG2, 0, SLJIT_MEM1(SLJIT_SCRATCH_REG1), SLJIT_OFFSETOF(jit_arguments, offset_count));
+if (common->mark_ptr != 0)
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_SCRATCH_REG1), SLJIT_OFFSETOF(jit_arguments, mark_ptr), SLJIT_SCRATCH_REG3, 0);
+OP2(SLJIT_SUB, SLJIT_SCRATCH_REG3, 0, SLJIT_MEM1(SLJIT_SCRATCH_REG1), SLJIT_OFFSETOF(jit_arguments, offsets), SLJIT_IMM, sizeof(int));
+OP1(SLJIT_MOV, SLJIT_SCRATCH_REG1, 0, SLJIT_MEM1(SLJIT_SCRATCH_REG1), SLJIT_OFFSETOF(jit_arguments, begin));
+GET_LOCAL_BASE(SLJIT_SAVED_REG1, 0, OVECTOR_START);
+/* Unlikely, but possible */
+early_quit = CMP(SLJIT_C_EQUAL, SLJIT_SCRATCH_REG2, 0, SLJIT_IMM, 0);
+loop = LABEL();
+OP2(SLJIT_SUB, SLJIT_SAVED_REG2, 0, SLJIT_MEM1(SLJIT_SAVED_REG1), 0, SLJIT_SCRATCH_REG1, 0);
+OP2(SLJIT_ADD, SLJIT_SAVED_REG1, 0, SLJIT_SAVED_REG1, 0, SLJIT_IMM, sizeof(sljit_sw));
+/* Copy the integer value to the output buffer */
+#if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
+OP2(SLJIT_ASHR, SLJIT_SAVED_REG2, 0, SLJIT_SAVED_REG2, 0, SLJIT_IMM, UCHAR_SHIFT);
+#endif
+OP1(SLJIT_MOVU_SI, SLJIT_MEM1(SLJIT_SCRATCH_REG3), sizeof(int), SLJIT_SAVED_REG2, 0);
+OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_SCRATCH_REG2, 0, SLJIT_SCRATCH_REG2, 0, SLJIT_IMM, 1);
+JUMPTO(SLJIT_C_NOT_ZERO, loop);
+JUMPHERE(early_quit);
+
+/* Calculate the return value, which is the maximum ovector value. */
+if (topbracket > 1)
+  {
+  GET_LOCAL_BASE(SLJIT_SCRATCH_REG1, 0, OVECTOR_START + topbracket * 2 * sizeof(sljit_sw));
+  OP1(SLJIT_MOV, SLJIT_SCRATCH_REG2, 0, SLJIT_IMM, topbracket + 1);
+
+  /* OVECTOR(0) is never equal to SLJIT_SAVED_REG3. */
+  loop = LABEL();
+  OP1(SLJIT_MOVU, SLJIT_SCRATCH_REG3, 0, SLJIT_MEM1(SLJIT_SCRATCH_REG1), -(2 * (sljit_sw)sizeof(sljit_sw)));
+  OP2(SLJIT_SUB, SLJIT_SCRATCH_REG2, 0, SLJIT_SCRATCH_REG2, 0, SLJIT_IMM, 1);
+  CMPTO(SLJIT_C_EQUAL, SLJIT_SCRATCH_REG3, 0, SLJIT_SAVED_REG3, 0, loop);
+  OP1(SLJIT_MOV, SLJIT_RETURN_REG, 0, SLJIT_SCRATCH_REG2, 0);
+  }
+else
+  OP1(SLJIT_MOV, SLJIT_RETURN_REG, 0, SLJIT_IMM, 1);
+}
+
+static SLJIT_INLINE void return_with_partial_match(compiler_common *common, struct sljit_label *quit)
+{
+DEFINE_COMPILER;
+struct sljit_jump *jump;
+
+SLJIT_COMPILE_ASSERT(STR_END == SLJIT_SAVED_REG2, str_end_must_be_saved_reg2);
+SLJIT_ASSERT(common->start_used_ptr != 0 && common->start_ptr != 0
+  && (common->mode == JIT_PARTIAL_SOFT_COMPILE ? common->hit_start != 0 : common->hit_start == 0));
+
+OP1(SLJIT_MOV, SLJIT_SCRATCH_REG2, 0, ARGUMENTS, 0);
+OP1(SLJIT_MOV, SLJIT_RETURN_REG, 0, SLJIT_IMM, PCRE_ERROR_PARTIAL);
+OP1(SLJIT_MOV_SI, SLJIT_SCRATCH_REG3, 0, SLJIT_MEM1(SLJIT_SCRATCH_REG2), SLJIT_OFFSETOF(jit_arguments, real_offset_count));
+CMPTO(SLJIT_C_SIG_LESS, SLJIT_SCRATCH_REG3, 0, SLJIT_IMM, 2, quit);
+
+/* Store match begin and end. */
+OP1(SLJIT_MOV, SLJIT_SAVED_REG1, 0, SLJIT_MEM1(SLJIT_SCRATCH_REG2), SLJIT_OFFSETOF(jit_arguments, begin));
+OP1(SLJIT_MOV, SLJIT_SCRATCH_REG2, 0, SLJIT_MEM1(SLJIT_SCRATCH_REG2), SLJIT_OFFSETOF(jit_arguments, offsets));
+
+jump = CMP(SLJIT_C_SIG_LESS, SLJIT_SCRATCH_REG3, 0, SLJIT_IMM, 3);
+OP2(SLJIT_SUB, SLJIT_SCRATCH_REG3, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), common->mode == JIT_PARTIAL_HARD_COMPILE ? common->start_ptr : (common->hit_start + (int)sizeof(sljit_sw)), SLJIT_SAVED_REG1, 0);
+#if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
+OP2(SLJIT_ASHR, SLJIT_SCRATCH_REG3, 0, SLJIT_SCRATCH_REG3, 0, SLJIT_IMM, UCHAR_SHIFT);
+#endif
+OP1(SLJIT_MOV_SI, SLJIT_MEM1(SLJIT_SCRATCH_REG2), 2 * sizeof(int), SLJIT_SCRATCH_REG3, 0);
+JUMPHERE(jump);
+
+OP1(SLJIT_MOV, SLJIT_SCRATCH_REG3, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), common->mode == JIT_PARTIAL_HARD_COMPILE ? common->start_used_ptr : common->hit_start);
+OP2(SLJIT_SUB, SLJIT_SAVED_REG2, 0, STR_END, 0, SLJIT_SAVED_REG1, 0);
+#if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
+OP2(SLJIT_ASHR, SLJIT_SAVED_REG2, 0, SLJIT_SAVED_REG2, 0, SLJIT_IMM, UCHAR_SHIFT);
+#endif
+OP1(SLJIT_MOV_SI, SLJIT_MEM1(SLJIT_SCRATCH_REG2), sizeof(int), SLJIT_SAVED_REG2, 0);
+
+OP2(SLJIT_SUB, SLJIT_SCRATCH_REG3, 0, SLJIT_SCRATCH_REG3, 0, SLJIT_SAVED_REG1, 0);
+#if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
+OP2(SLJIT_ASHR, SLJIT_SCRATCH_REG3, 0, SLJIT_SCRATCH_REG3, 0, SLJIT_IMM, UCHAR_SHIFT);
+#endif
+OP1(SLJIT_MOV_SI, SLJIT_MEM1(SLJIT_SCRATCH_REG2), 0, SLJIT_SCRATCH_REG3, 0);
+
+JUMPTO(SLJIT_JUMP, quit);
+}
+
+static SLJIT_INLINE void check_start_used_ptr(compiler_common *common)
+{
+/* May destroy TMP1. */
+DEFINE_COMPILER;
+struct sljit_jump *jump;
+
+if (common->mode == JIT_PARTIAL_SOFT_COMPILE)
+  {
+  /* The value of -1 must be kept for start_used_ptr! */
+  OP2(SLJIT_ADD, TMP1, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), common->start_used_ptr, SLJIT_IMM, 1);
+  /* Jumps if start_used_ptr < STR_PTR, or start_used_ptr == -1. Although overwriting
+  is not necessary if start_used_ptr == STR_PTR, it does not hurt as well. */
+  jump = CMP(SLJIT_C_LESS_EQUAL, TMP1, 0, STR_PTR, 0);
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->start_used_ptr, STR_PTR, 0);
+  JUMPHERE(jump);
+  }
+else if (common->mode == JIT_PARTIAL_HARD_COMPILE)
+  {
+  jump = CMP(SLJIT_C_LESS_EQUAL, SLJIT_MEM1(SLJIT_LOCALS_REG), common->start_used_ptr, STR_PTR, 0);
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->start_used_ptr, STR_PTR, 0);
+  JUMPHERE(jump);
+  }
+}
+
+static SLJIT_INLINE BOOL char_has_othercase(compiler_common *common, pcre_uchar* cc)
+{
+/* Detects if the character has an othercase. */
+unsigned int c;
+
+#ifdef SUPPORT_UTF
+if (common->utf)
+  {
+  GETCHAR(c, cc);
+  if (c > 127)
+    {
+#ifdef SUPPORT_UCP
+    return c != UCD_OTHERCASE(c);
+#else
+    return FALSE;
+#endif
+    }
+#ifndef COMPILE_PCRE8
+  return common->fcc[c] != c;
+#endif
+  }
+else
+#endif
+  c = *cc;
+return MAX_255(c) ? common->fcc[c] != c : FALSE;
+}
+
+static SLJIT_INLINE unsigned int char_othercase(compiler_common *common, unsigned int c)
+{
+/* Returns with the othercase. */
+#ifdef SUPPORT_UTF
+if (common->utf && c > 127)
+  {
+#ifdef SUPPORT_UCP
+  return UCD_OTHERCASE(c);
+#else
+  return c;
+#endif
+  }
+#endif
+return TABLE_GET(c, common->fcc, c);
+}
+
+static unsigned int char_get_othercase_bit(compiler_common *common, pcre_uchar* cc)
+{
+/* Detects if the character and its othercase has only 1 bit difference. */
+unsigned int c, oc, bit;
+#if defined SUPPORT_UTF && defined COMPILE_PCRE8
+int n;
+#endif
+
+#ifdef SUPPORT_UTF
+if (common->utf)
+  {
+  GETCHAR(c, cc);
+  if (c <= 127)
+    oc = common->fcc[c];
+  else
+    {
+#ifdef SUPPORT_UCP
+    oc = UCD_OTHERCASE(c);
+#else
+    oc = c;
+#endif
+    }
+  }
+else
+  {
+  c = *cc;
+  oc = TABLE_GET(c, common->fcc, c);
+  }
+#else
+c = *cc;
+oc = TABLE_GET(c, common->fcc, c);
+#endif
+
+SLJIT_ASSERT(c != oc);
+
+bit = c ^ oc;
+/* Optimized for English alphabet. */
+if (c <= 127 && bit == 0x20)
+  return (0 << 8) | 0x20;
+
+/* Since c != oc, they must have at least 1 bit difference. */
+if (!is_powerof2(bit))
+  return 0;
+
+#if defined COMPILE_PCRE8
+
+#ifdef SUPPORT_UTF
+if (common->utf && c > 127)
+  {
+  n = GET_EXTRALEN(*cc);
+  while ((bit & 0x3f) == 0)
+    {
+    n--;
+    bit >>= 6;
+    }
+  return (n << 8) | bit;
+  }
+#endif /* SUPPORT_UTF */
+return (0 << 8) | bit;
+
+#elif defined COMPILE_PCRE16 || defined COMPILE_PCRE32
+
+#ifdef SUPPORT_UTF
+if (common->utf && c > 65535)
+  {
+  if (bit >= (1 << 10))
+    bit >>= 10;
+  else
+    return (bit < 256) ? ((2 << 8) | bit) : ((3 << 8) | (bit >> 8));
+  }
+#endif /* SUPPORT_UTF */
+return (bit < 256) ? ((0 << 8) | bit) : ((1 << 8) | (bit >> 8));
+
+#endif /* COMPILE_PCRE[8|16|32] */
+}
+
+static void check_partial(compiler_common *common, BOOL force)
+{
+/* Checks whether a partial matching is occurred. Does not modify registers. */
+DEFINE_COMPILER;
+struct sljit_jump *jump = NULL;
+
+SLJIT_ASSERT(!force || common->mode != JIT_COMPILE);
+
+if (common->mode == JIT_COMPILE)
+  return;
+
+if (!force)
+  jump = CMP(SLJIT_C_GREATER_EQUAL, SLJIT_MEM1(SLJIT_LOCALS_REG), common->start_used_ptr, STR_PTR, 0);
+else if (common->mode == JIT_PARTIAL_SOFT_COMPILE)
+  jump = CMP(SLJIT_C_EQUAL, SLJIT_MEM1(SLJIT_LOCALS_REG), common->start_used_ptr, SLJIT_IMM, -1);
+
+if (common->mode == JIT_PARTIAL_SOFT_COMPILE)
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->hit_start, SLJIT_IMM, 0);
+else
+  {
+  if (common->partialmatchlabel != NULL)
+    JUMPTO(SLJIT_JUMP, common->partialmatchlabel);
+  else
+    add_jump(compiler, &common->partialmatch, JUMP(SLJIT_JUMP));
+  }
+
+if (jump != NULL)
+  JUMPHERE(jump);
+}
+
+static void check_str_end(compiler_common *common, jump_list **end_reached)
+{
+/* Does not affect registers. Usually used in a tight spot. */
+DEFINE_COMPILER;
+struct sljit_jump *jump;
+
+if (common->mode == JIT_COMPILE)
+  {
+  add_jump(compiler, end_reached, CMP(SLJIT_C_GREATER_EQUAL, STR_PTR, 0, STR_END, 0));
+  return;
+  }
+
+jump = CMP(SLJIT_C_LESS, STR_PTR, 0, STR_END, 0);
+if (common->mode == JIT_PARTIAL_SOFT_COMPILE)
+  {
+  add_jump(compiler, end_reached, CMP(SLJIT_C_GREATER_EQUAL, SLJIT_MEM1(SLJIT_LOCALS_REG), common->start_used_ptr, STR_PTR, 0));
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->hit_start, SLJIT_IMM, 0);
+  add_jump(compiler, end_reached, JUMP(SLJIT_JUMP));
+  }
+else
+  {
+  add_jump(compiler, end_reached, CMP(SLJIT_C_GREATER_EQUAL, SLJIT_MEM1(SLJIT_LOCALS_REG), common->start_used_ptr, STR_PTR, 0));
+  if (common->partialmatchlabel != NULL)
+    JUMPTO(SLJIT_JUMP, common->partialmatchlabel);
+  else
+    add_jump(compiler, &common->partialmatch, JUMP(SLJIT_JUMP));
+  }
+JUMPHERE(jump);
+}
+
+static void detect_partial_match(compiler_common *common, jump_list **backtracks)
+{
+DEFINE_COMPILER;
+struct sljit_jump *jump;
+
+if (common->mode == JIT_COMPILE)
+  {
+  add_jump(compiler, backtracks, CMP(SLJIT_C_GREATER_EQUAL, STR_PTR, 0, STR_END, 0));
+  return;
+  }
+
+/* Partial matching mode. */
+jump = CMP(SLJIT_C_LESS, STR_PTR, 0, STR_END, 0);
+add_jump(compiler, backtracks, CMP(SLJIT_C_GREATER_EQUAL, SLJIT_MEM1(SLJIT_LOCALS_REG), common->start_used_ptr, STR_PTR, 0));
+if (common->mode == JIT_PARTIAL_SOFT_COMPILE)
+  {
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->hit_start, SLJIT_IMM, 0);
+  add_jump(compiler, backtracks, JUMP(SLJIT_JUMP));
+  }
+else
+  {
+  if (common->partialmatchlabel != NULL)
+    JUMPTO(SLJIT_JUMP, common->partialmatchlabel);
+  else
+    add_jump(compiler, &common->partialmatch, JUMP(SLJIT_JUMP));
+  }
+JUMPHERE(jump);
+}
+
+static void read_char(compiler_common *common)
+{
+/* Reads the character into TMP1, updates STR_PTR.
+Does not check STR_END. TMP2 Destroyed. */
+DEFINE_COMPILER;
+#if defined SUPPORT_UTF && !defined COMPILE_PCRE32
+struct sljit_jump *jump;
+#endif
+
+OP1(MOV_UCHAR, TMP1, 0, SLJIT_MEM1(STR_PTR), 0);
+#if defined SUPPORT_UTF && !defined COMPILE_PCRE32
+if (common->utf)
+  {
+#if defined COMPILE_PCRE8
+  jump = CMP(SLJIT_C_LESS, TMP1, 0, SLJIT_IMM, 0xc0);
+#elif defined COMPILE_PCRE16
+  jump = CMP(SLJIT_C_LESS, TMP1, 0, SLJIT_IMM, 0xd800);
+#endif /* COMPILE_PCRE[8|16] */
+  add_jump(compiler, &common->utfreadchar, JUMP(SLJIT_FAST_CALL));
+  JUMPHERE(jump);
+  }
+#endif /* SUPPORT_UTF && !COMPILE_PCRE32 */
+OP2(SLJIT_ADD, STR_PTR, 0, STR_PTR, 0, SLJIT_IMM, IN_UCHARS(1));
+}
+
+static void peek_char(compiler_common *common)
+{
+/* Reads the character into TMP1, keeps STR_PTR.
+Does not check STR_END. TMP2 Destroyed. */
+DEFINE_COMPILER;
+#if defined SUPPORT_UTF && !defined COMPILE_PCRE32
+struct sljit_jump *jump;
+#endif
+
+OP1(MOV_UCHAR, TMP1, 0, SLJIT_MEM1(STR_PTR), 0);
+#if defined SUPPORT_UTF && !defined COMPILE_PCRE32
+if (common->utf)
+  {
+#if defined COMPILE_PCRE8
+  jump = CMP(SLJIT_C_LESS, TMP1, 0, SLJIT_IMM, 0xc0);
+#elif defined COMPILE_PCRE16
+  jump = CMP(SLJIT_C_LESS, TMP1, 0, SLJIT_IMM, 0xd800);
+#endif /* COMPILE_PCRE[8|16] */
+  add_jump(compiler, &common->utfreadchar, JUMP(SLJIT_FAST_CALL));
+  OP2(SLJIT_SUB, STR_PTR, 0, STR_PTR, 0, TMP2, 0);
+  JUMPHERE(jump);
+  }
+#endif /* SUPPORT_UTF && !COMPILE_PCRE32 */
+}
+
+static void read_char8_type(compiler_common *common)
+{
+/* Reads the character type into TMP1, updates STR_PTR. Does not check STR_END. */
+DEFINE_COMPILER;
+#if defined SUPPORT_UTF || defined COMPILE_PCRE16 || defined COMPILE_PCRE32
+struct sljit_jump *jump;
+#endif
+
+#ifdef SUPPORT_UTF
+if (common->utf)
+  {
+  OP1(MOV_UCHAR, TMP2, 0, SLJIT_MEM1(STR_PTR), 0);
+  OP2(SLJIT_ADD, STR_PTR, 0, STR_PTR, 0, SLJIT_IMM, IN_UCHARS(1));
+#if defined COMPILE_PCRE8
+  /* This can be an extra read in some situations, but hopefully
+  it is needed in most cases. */
+  OP1(SLJIT_MOV_UB, TMP1, 0, SLJIT_MEM1(TMP2), common->ctypes);
+  jump = CMP(SLJIT_C_LESS, TMP2, 0, SLJIT_IMM, 0xc0);
+  add_jump(compiler, &common->utfreadtype8, JUMP(SLJIT_FAST_CALL));
+  JUMPHERE(jump);
+#elif defined COMPILE_PCRE16
+  OP1(SLJIT_MOV, TMP1, 0, SLJIT_IMM, 0);
+  jump = CMP(SLJIT_C_GREATER, TMP2, 0, SLJIT_IMM, 255);
+  OP1(SLJIT_MOV_UB, TMP1, 0, SLJIT_MEM1(TMP2), common->ctypes);
+  JUMPHERE(jump);
+  /* Skip low surrogate if necessary. */
+  OP2(SLJIT_AND, TMP2, 0, TMP2, 0, SLJIT_IMM, 0xfc00);
+  OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP2, 0, SLJIT_IMM, 0xd800);
+  OP_FLAGS(SLJIT_MOV, TMP2, 0, SLJIT_UNUSED, 0, SLJIT_C_EQUAL);
+  OP2(SLJIT_SHL, TMP2, 0, TMP2, 0, SLJIT_IMM, 1);
+  OP2(SLJIT_ADD, STR_PTR, 0, STR_PTR, 0, TMP2, 0);
+#elif defined COMPILE_PCRE32
+  OP1(SLJIT_MOV, TMP1, 0, SLJIT_IMM, 0);
+  jump = CMP(SLJIT_C_GREATER, TMP2, 0, SLJIT_IMM, 255);
+  OP1(SLJIT_MOV_UB, TMP1, 0, SLJIT_MEM1(TMP2), common->ctypes);
+  JUMPHERE(jump);
+#endif /* COMPILE_PCRE[8|16|32] */
+  return;
+  }
+#endif /* SUPPORT_UTF */
+OP1(MOV_UCHAR, TMP2, 0, SLJIT_MEM1(STR_PTR), 0);
+OP2(SLJIT_ADD, STR_PTR, 0, STR_PTR, 0, SLJIT_IMM, IN_UCHARS(1));
+#if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
+/* The ctypes array contains only 256 values. */
+OP1(SLJIT_MOV, TMP1, 0, SLJIT_IMM, 0);
+jump = CMP(SLJIT_C_GREATER, TMP2, 0, SLJIT_IMM, 255);
+#endif
+OP1(SLJIT_MOV_UB, TMP1, 0, SLJIT_MEM1(TMP2), common->ctypes);
+#if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
+JUMPHERE(jump);
+#endif
+}
+
+static void skip_char_back(compiler_common *common)
+{
+/* Goes one character back. Affects STR_PTR and TMP1. Does not check begin. */
+DEFINE_COMPILER;
+#if defined SUPPORT_UTF && !defined COMPILE_PCRE32
+#if defined COMPILE_PCRE8
+struct sljit_label *label;
+
+if (common->utf)
+  {
+  label = LABEL();
+  OP1(MOV_UCHAR, TMP1, 0, SLJIT_MEM1(STR_PTR), -IN_UCHARS(1));
+  OP2(SLJIT_SUB, STR_PTR, 0, STR_PTR, 0, SLJIT_IMM, IN_UCHARS(1));
+  OP2(SLJIT_AND, TMP1, 0, TMP1, 0, SLJIT_IMM, 0xc0);
+  CMPTO(SLJIT_C_EQUAL, TMP1, 0, SLJIT_IMM, 0x80, label);
+  return;
+  }
+#elif defined COMPILE_PCRE16
+if (common->utf)
+  {
+  OP1(MOV_UCHAR, TMP1, 0, SLJIT_MEM1(STR_PTR), -IN_UCHARS(1));
+  OP2(SLJIT_SUB, STR_PTR, 0, STR_PTR, 0, SLJIT_IMM, IN_UCHARS(1));
+  /* Skip low surrogate if necessary. */
+  OP2(SLJIT_AND, TMP1, 0, TMP1, 0, SLJIT_IMM, 0xfc00);
+  OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, 0xdc00);
+  OP_FLAGS(SLJIT_MOV, TMP1, 0, SLJIT_UNUSED, 0, SLJIT_C_EQUAL);
+  OP2(SLJIT_SHL, TMP1, 0, TMP1, 0, SLJIT_IMM, 1);
+  OP2(SLJIT_SUB, STR_PTR, 0, STR_PTR, 0, TMP1, 0);
+  return;
+  }
+#endif /* COMPILE_PCRE[8|16] */
+#endif /* SUPPORT_UTF && !COMPILE_PCRE32 */
+OP2(SLJIT_SUB, STR_PTR, 0, STR_PTR, 0, SLJIT_IMM, IN_UCHARS(1));
+}
+
+static void check_newlinechar(compiler_common *common, int nltype, jump_list **backtracks, BOOL jumpiftrue)
+{
+/* Character comes in TMP1. Checks if it is a newline. TMP2 may be destroyed. */
+DEFINE_COMPILER;
+
+if (nltype == NLTYPE_ANY)
+  {
+  add_jump(compiler, &common->anynewline, JUMP(SLJIT_FAST_CALL));
+  add_jump(compiler, backtracks, JUMP(jumpiftrue ? SLJIT_C_NOT_ZERO : SLJIT_C_ZERO));
+  }
+else if (nltype == NLTYPE_ANYCRLF)
+  {
+  OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, CHAR_CR);
+  OP_FLAGS(SLJIT_MOV, TMP2, 0, SLJIT_UNUSED, 0, SLJIT_C_EQUAL);
+  OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, CHAR_NL);
+  OP_FLAGS(SLJIT_OR | SLJIT_SET_E, TMP2, 0, TMP2, 0, SLJIT_C_EQUAL);
+  add_jump(compiler, backtracks, JUMP(jumpiftrue ? SLJIT_C_NOT_ZERO : SLJIT_C_ZERO));
+  }
+else
+  {
+  SLJIT_ASSERT(nltype == NLTYPE_FIXED && common->newline < 256);
+  add_jump(compiler, backtracks, CMP(jumpiftrue ? SLJIT_C_EQUAL : SLJIT_C_NOT_EQUAL, TMP1, 0, SLJIT_IMM, common->newline));
+  }
+}
+
+#ifdef SUPPORT_UTF
+
+#if defined COMPILE_PCRE8
+static void do_utfreadchar(compiler_common *common)
+{
+/* Fast decoding a UTF-8 character. TMP1 contains the first byte
+of the character (>= 0xc0). Return char value in TMP1, length - 1 in TMP2. */
+DEFINE_COMPILER;
+struct sljit_jump *jump;
+
+sljit_emit_fast_enter(compiler, RETURN_ADDR, 0);
+/* Searching for the first zero. */
+OP2(SLJIT_AND | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, 0x20);
+jump = JUMP(SLJIT_C_NOT_ZERO);
+/* Two byte sequence. */
+OP1(MOV_UCHAR, TMP2, 0, SLJIT_MEM1(STR_PTR), IN_UCHARS(1));
+OP2(SLJIT_ADD, STR_PTR, 0, STR_PTR, 0, SLJIT_IMM, IN_UCHARS(1));
+OP2(SLJIT_AND, TMP1, 0, TMP1, 0, SLJIT_IMM, 0x1f);
+OP2(SLJIT_SHL, TMP1, 0, TMP1, 0, SLJIT_IMM, 6);
+OP2(SLJIT_AND, TMP2, 0, TMP2, 0, SLJIT_IMM, 0x3f);
+OP2(SLJIT_OR, TMP1, 0, TMP1, 0, TMP2, 0);
+OP1(SLJIT_MOV, TMP2, 0, SLJIT_IMM, IN_UCHARS(1));
+sljit_emit_fast_return(compiler, RETURN_ADDR, 0);
+JUMPHERE(jump);
+
+OP2(SLJIT_AND | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, 0x10);
+jump = JUMP(SLJIT_C_NOT_ZERO);
+/* Three byte sequence. */
+OP1(MOV_UCHAR, TMP2, 0, SLJIT_MEM1(STR_PTR), IN_UCHARS(1));
+OP2(SLJIT_AND, TMP1, 0, TMP1, 0, SLJIT_IMM, 0x0f);
+OP2(SLJIT_SHL, TMP1, 0, TMP1, 0, SLJIT_IMM, 12);
+OP2(SLJIT_AND, TMP2, 0, TMP2, 0, SLJIT_IMM, 0x3f);
+OP2(SLJIT_SHL, TMP2, 0, TMP2, 0, SLJIT_IMM, 6);
+OP2(SLJIT_OR, TMP1, 0, TMP1, 0, TMP2, 0);
+OP1(MOV_UCHAR, TMP2, 0, SLJIT_MEM1(STR_PTR), IN_UCHARS(2));
+OP2(SLJIT_ADD, STR_PTR, 0, STR_PTR, 0, SLJIT_IMM, IN_UCHARS(2));
+OP2(SLJIT_AND, TMP2, 0, TMP2, 0, SLJIT_IMM, 0x3f);
+OP2(SLJIT_OR, TMP1, 0, TMP1, 0, TMP2, 0);
+OP1(SLJIT_MOV, TMP2, 0, SLJIT_IMM, IN_UCHARS(2));
+sljit_emit_fast_return(compiler, RETURN_ADDR, 0);
+JUMPHERE(jump);
+
+/* Four byte sequence. */
+OP1(MOV_UCHAR, TMP2, 0, SLJIT_MEM1(STR_PTR), IN_UCHARS(1));
+OP2(SLJIT_AND, TMP1, 0, TMP1, 0, SLJIT_IMM, 0x07);
+OP2(SLJIT_SHL, TMP1, 0, TMP1, 0, SLJIT_IMM, 18);
+OP2(SLJIT_AND, TMP2, 0, TMP2, 0, SLJIT_IMM, 0x3f);
+OP2(SLJIT_SHL, TMP2, 0, TMP2, 0, SLJIT_IMM, 12);
+OP2(SLJIT_OR, TMP1, 0, TMP1, 0, TMP2, 0);
+OP1(MOV_UCHAR, TMP2, 0, SLJIT_MEM1(STR_PTR), IN_UCHARS(2));
+OP2(SLJIT_AND, TMP2, 0, TMP2, 0, SLJIT_IMM, 0x3f);
+OP2(SLJIT_SHL, TMP2, 0, TMP2, 0, SLJIT_IMM, 6);
+OP2(SLJIT_OR, TMP1, 0, TMP1, 0, TMP2, 0);
+OP1(MOV_UCHAR, TMP2, 0, SLJIT_MEM1(STR_PTR), IN_UCHARS(3));
+OP2(SLJIT_ADD, STR_PTR, 0, STR_PTR, 0, SLJIT_IMM, IN_UCHARS(3));
+OP2(SLJIT_AND, TMP2, 0, TMP2, 0, SLJIT_IMM, 0x3f);
+OP2(SLJIT_OR, TMP1, 0, TMP1, 0, TMP2, 0);
+OP1(SLJIT_MOV, TMP2, 0, SLJIT_IMM, IN_UCHARS(3));
+sljit_emit_fast_return(compiler, RETURN_ADDR, 0);
+}
+
+static void do_utfreadtype8(compiler_common *common)
+{
+/* Fast decoding a UTF-8 character type. TMP2 contains the first byte
+of the character (>= 0xc0). Return value in TMP1. */
+DEFINE_COMPILER;
+struct sljit_jump *jump;
+struct sljit_jump *compare;
+
+sljit_emit_fast_enter(compiler, RETURN_ADDR, 0);
+
+OP2(SLJIT_AND | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP2, 0, SLJIT_IMM, 0x20);
+jump = JUMP(SLJIT_C_NOT_ZERO);
+/* Two byte sequence. */
+OP1(MOV_UCHAR, TMP1, 0, SLJIT_MEM1(STR_PTR), IN_UCHARS(0));
+OP2(SLJIT_ADD, STR_PTR, 0, STR_PTR, 0, SLJIT_IMM, IN_UCHARS(1));
+OP2(SLJIT_AND, TMP2, 0, TMP2, 0, SLJIT_IMM, 0x1f);
+OP2(SLJIT_SHL, TMP2, 0, TMP2, 0, SLJIT_IMM, 6);
+OP2(SLJIT_AND, TMP1, 0, TMP1, 0, SLJIT_IMM, 0x3f);
+OP2(SLJIT_OR, TMP2, 0, TMP2, 0, TMP1, 0);
+compare = CMP(SLJIT_C_GREATER, TMP2, 0, SLJIT_IMM, 255);
+OP1(SLJIT_MOV_UB, TMP1, 0, SLJIT_MEM1(TMP2), common->ctypes);
+sljit_emit_fast_return(compiler, RETURN_ADDR, 0);
+
+JUMPHERE(compare);
+OP1(SLJIT_MOV, TMP1, 0, SLJIT_IMM, 0);
+sljit_emit_fast_return(compiler, RETURN_ADDR, 0);
+JUMPHERE(jump);
+
+/* We only have types for characters less than 256. */
+OP1(SLJIT_MOV_UB, TMP1, 0, SLJIT_MEM1(TMP2), (sljit_sw)PRIV(utf8_table4) - 0xc0);
+OP2(SLJIT_ADD, STR_PTR, 0, STR_PTR, 0, TMP1, 0);
+OP1(SLJIT_MOV, TMP1, 0, SLJIT_IMM, 0);
+sljit_emit_fast_return(compiler, RETURN_ADDR, 0);
+}
+
+#elif defined COMPILE_PCRE16
+
+static void do_utfreadchar(compiler_common *common)
+{
+/* Fast decoding a UTF-16 character. TMP1 contains the first 16 bit char
+of the character (>= 0xd800). Return char value in TMP1, length - 1 in TMP2. */
+DEFINE_COMPILER;
+struct sljit_jump *jump;
+
+sljit_emit_fast_enter(compiler, RETURN_ADDR, 0);
+jump = CMP(SLJIT_C_LESS, TMP1, 0, SLJIT_IMM, 0xdc00);
+/* Do nothing, only return. */
+sljit_emit_fast_return(compiler, RETURN_ADDR, 0);
+
+JUMPHERE(jump);
+/* Combine two 16 bit characters. */
+OP1(MOV_UCHAR, TMP2, 0, SLJIT_MEM1(STR_PTR), IN_UCHARS(1));
+OP2(SLJIT_ADD, STR_PTR, 0, STR_PTR, 0, SLJIT_IMM, IN_UCHARS(1));
+OP2(SLJIT_AND, TMP1, 0, TMP1, 0, SLJIT_IMM, 0x3ff);
+OP2(SLJIT_SHL, TMP1, 0, TMP1, 0, SLJIT_IMM, 10);
+OP2(SLJIT_AND, TMP2, 0, TMP2, 0, SLJIT_IMM, 0x3ff);
+OP2(SLJIT_OR, TMP1, 0, TMP1, 0, TMP2, 0);
+OP1(SLJIT_MOV, TMP2, 0, SLJIT_IMM, IN_UCHARS(1));
+OP2(SLJIT_ADD, TMP1, 0, TMP1, 0, SLJIT_IMM, 0x10000);
+sljit_emit_fast_return(compiler, RETURN_ADDR, 0);
+}
+
+#endif /* COMPILE_PCRE[8|16] */
+
+#endif /* SUPPORT_UTF */
+
+#ifdef SUPPORT_UCP
+
+/* UCD_BLOCK_SIZE must be 128 (see the assert below). */
+#define UCD_BLOCK_MASK 127
+#define UCD_BLOCK_SHIFT 7
+
+static void do_getucd(compiler_common *common)
+{
+/* Search the UCD record for the character comes in TMP1.
+Returns chartype in TMP1 and UCD offset in TMP2. */
+DEFINE_COMPILER;
+
+SLJIT_ASSERT(UCD_BLOCK_SIZE == 128 && sizeof(ucd_record) == 8);
+
+sljit_emit_fast_enter(compiler, RETURN_ADDR, 0);
+OP2(SLJIT_LSHR, TMP2, 0, TMP1, 0, SLJIT_IMM, UCD_BLOCK_SHIFT);
+OP1(SLJIT_MOV_UB, TMP2, 0, SLJIT_MEM1(TMP2), (sljit_sw)PRIV(ucd_stage1));
+OP2(SLJIT_AND, TMP1, 0, TMP1, 0, SLJIT_IMM, UCD_BLOCK_MASK);
+OP2(SLJIT_SHL, TMP2, 0, TMP2, 0, SLJIT_IMM, UCD_BLOCK_SHIFT);
+OP2(SLJIT_ADD, TMP1, 0, TMP1, 0, TMP2, 0);
+OP1(SLJIT_MOV, TMP2, 0, SLJIT_IMM, (sljit_sw)PRIV(ucd_stage2));
+OP1(SLJIT_MOV_UH, TMP2, 0, SLJIT_MEM2(TMP2, TMP1), 1);
+OP1(SLJIT_MOV, TMP1, 0, SLJIT_IMM, (sljit_sw)PRIV(ucd_records) + SLJIT_OFFSETOF(ucd_record, chartype));
+OP1(SLJIT_MOV_UB, TMP1, 0, SLJIT_MEM2(TMP1, TMP2), 3);
+sljit_emit_fast_return(compiler, RETURN_ADDR, 0);
+}
+#endif
+
+static SLJIT_INLINE struct sljit_label *mainloop_entry(compiler_common *common, BOOL hascrorlf, BOOL firstline)
+{
+DEFINE_COMPILER;
+struct sljit_label *mainloop;
+struct sljit_label *newlinelabel = NULL;
+struct sljit_jump *start;
+struct sljit_jump *end = NULL;
+struct sljit_jump *nl = NULL;
+#if defined SUPPORT_UTF && !defined COMPILE_PCRE32
+struct sljit_jump *singlechar;
+#endif
+jump_list *newline = NULL;
+BOOL newlinecheck = FALSE;
+BOOL readuchar = FALSE;
+
+if (!(hascrorlf || firstline) && (common->nltype == NLTYPE_ANY ||
+    common->nltype == NLTYPE_ANYCRLF || common->newline > 255))
+  newlinecheck = TRUE;
+
+if (firstline)
+  {
+  /* Search for the end of the first line. */
+  SLJIT_ASSERT(common->first_line_end != 0);
+  OP1(SLJIT_MOV, TMP3, 0, STR_PTR, 0);
+
+  if (common->nltype == NLTYPE_FIXED && common->newline > 255)
+    {
+    mainloop = LABEL();
+    OP2(SLJIT_ADD, STR_PTR, 0, STR_PTR, 0, SLJIT_IMM, IN_UCHARS(1));
+    end = CMP(SLJIT_C_GREATER_EQUAL, STR_PTR, 0, STR_END, 0);
+    OP1(MOV_UCHAR, TMP1, 0, SLJIT_MEM1(STR_PTR), IN_UCHARS(-1));
+    OP1(MOV_UCHAR, TMP2, 0, SLJIT_MEM1(STR_PTR), IN_UCHARS(0));
+    CMPTO(SLJIT_C_NOT_EQUAL, TMP1, 0, SLJIT_IMM, (common->newline >> 8) & 0xff, mainloop);
+    CMPTO(SLJIT_C_NOT_EQUAL, TMP2, 0, SLJIT_IMM, common->newline & 0xff, mainloop);
+    JUMPHERE(end);
+    OP2(SLJIT_SUB, SLJIT_MEM1(SLJIT_LOCALS_REG), common->first_line_end, STR_PTR, 0, SLJIT_IMM, IN_UCHARS(1));
+    }
+  else
+    {
+    end = CMP(SLJIT_C_GREATER_EQUAL, STR_PTR, 0, STR_END, 0);
+    mainloop = LABEL();
+    /* Continual stores does not cause data dependency. */
+    OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->first_line_end, STR_PTR, 0);
+    read_char(common);
+    check_newlinechar(common, common->nltype, &newline, TRUE);
+    CMPTO(SLJIT_C_LESS, STR_PTR, 0, STR_END, 0, mainloop);
+    JUMPHERE(end);
+    OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->first_line_end, STR_PTR, 0);
+    set_jumps(newline, LABEL());
+    }
+
+  OP1(SLJIT_MOV, STR_PTR, 0, TMP3, 0);
+  }
+
+start = JUMP(SLJIT_JUMP);
+
+if (newlinecheck)
+  {
+  newlinelabel = LABEL();
+  OP2(SLJIT_ADD, STR_PTR, 0, STR_PTR, 0, SLJIT_IMM, IN_UCHARS(1));
+  end = CMP(SLJIT_C_GREATER_EQUAL, STR_PTR, 0, STR_END, 0);
+  OP1(MOV_UCHAR, TMP1, 0, SLJIT_MEM1(STR_PTR), 0);
+  OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, common->newline & 0xff);
+  OP_FLAGS(SLJIT_MOV, TMP1, 0, SLJIT_UNUSED, 0, SLJIT_C_EQUAL);
+#if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
+  OP2(SLJIT_SHL, TMP1, 0, TMP1, 0, SLJIT_IMM, UCHAR_SHIFT);
+#endif
+  OP2(SLJIT_ADD, STR_PTR, 0, STR_PTR, 0, TMP1, 0);
+  nl = JUMP(SLJIT_JUMP);
+  }
+
+mainloop = LABEL();
+
+/* Increasing the STR_PTR here requires one less jump in the most common case. */
+#ifdef SUPPORT_UTF
+if (common->utf) readuchar = TRUE;
+#endif
+if (newlinecheck) readuchar = TRUE;
+
+if (readuchar)
+  OP1(MOV_UCHAR, TMP1, 0, SLJIT_MEM1(STR_PTR), 0);
+
+if (newlinecheck)
+  CMPTO(SLJIT_C_EQUAL, TMP1, 0, SLJIT_IMM, (common->newline >> 8) & 0xff, newlinelabel);
+
+OP2(SLJIT_ADD, STR_PTR, 0, STR_PTR, 0, SLJIT_IMM, IN_UCHARS(1));
+#if defined SUPPORT_UTF && !defined COMPILE_PCRE32
+#if defined COMPILE_PCRE8
+if (common->utf)
+  {
+  singlechar = CMP(SLJIT_C_LESS, TMP1, 0, SLJIT_IMM, 0xc0);
+  OP1(SLJIT_MOV_UB, TMP1, 0, SLJIT_MEM1(TMP1), (sljit_sw)PRIV(utf8_table4) - 0xc0);
+  OP2(SLJIT_ADD, STR_PTR, 0, STR_PTR, 0, TMP1, 0);
+  JUMPHERE(singlechar);
+  }
+#elif defined COMPILE_PCRE16
+if (common->utf)
+  {
+  singlechar = CMP(SLJIT_C_LESS, TMP1, 0, SLJIT_IMM, 0xd800);
+  OP2(SLJIT_AND, TMP1, 0, TMP1, 0, SLJIT_IMM, 0xfc00);
+  OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, 0xd800);
+  OP_FLAGS(SLJIT_MOV, TMP1, 0, SLJIT_UNUSED, 0, SLJIT_C_EQUAL);
+  OP2(SLJIT_SHL, TMP1, 0, TMP1, 0, SLJIT_IMM, 1);
+  OP2(SLJIT_ADD, STR_PTR, 0, STR_PTR, 0, TMP1, 0);
+  JUMPHERE(singlechar);
+  }
+#endif /* COMPILE_PCRE[8|16] */
+#endif /* SUPPORT_UTF && !COMPILE_PCRE32 */
+JUMPHERE(start);
+
+if (newlinecheck)
+  {
+  JUMPHERE(end);
+  JUMPHERE(nl);
+  }
+
+return mainloop;
+}
+
+#define MAX_N_CHARS 3
+
+static SLJIT_INLINE BOOL fast_forward_first_n_chars(compiler_common *common, BOOL firstline)
+{
+DEFINE_COMPILER;
+struct sljit_label *start;
+struct sljit_jump *quit;
+pcre_uint32 chars[MAX_N_CHARS * 2];
+pcre_uchar *cc = common->start + 1 + LINK_SIZE;
+int location = 0;
+pcre_int32 len, c, bit, caseless;
+int must_stop;
+
+/* We do not support alternatives now. */
+if (*(common->start + GET(common->start, 1)) == OP_ALT)
+  return FALSE;
+
+while (TRUE)
+  {
+  caseless = 0;
+  must_stop = 1;
+  switch(*cc)
+    {
+    case OP_CHAR:
+    must_stop = 0;
+    cc++;
+    break;
+
+    case OP_CHARI:
+    caseless = 1;
+    must_stop = 0;
+    cc++;
+    break;
+
+    case OP_SOD:
+    case OP_SOM:
+    case OP_SET_SOM:
+    case OP_NOT_WORD_BOUNDARY:
+    case OP_WORD_BOUNDARY:
+    case OP_EODN:
+    case OP_EOD:
+    case OP_CIRC:
+    case OP_CIRCM:
+    case OP_DOLL:
+    case OP_DOLLM:
+    /* Zero width assertions. */
+    cc++;
+    continue;
+
+    case OP_PLUS:
+    case OP_MINPLUS:
+    case OP_POSPLUS:
+    cc++;
+    break;
+
+    case OP_EXACT:
+    cc += 1 + IMM2_SIZE;
+    break;
+
+    case OP_PLUSI:
+    case OP_MINPLUSI:
+    case OP_POSPLUSI:
+    caseless = 1;
+    cc++;
+    break;
+
+    case OP_EXACTI:
+    caseless = 1;
+    cc += 1 + IMM2_SIZE;
+    break;
+
+    default:
+    must_stop = 2;
+    break;
+    }
+
+  if (must_stop == 2)
+      break;
+
+  len = 1;
+#ifdef SUPPORT_UTF
+  if (common->utf && HAS_EXTRALEN(cc[0])) len += GET_EXTRALEN(cc[0]);
+#endif
+
+  if (caseless && char_has_othercase(common, cc))
+    {
+    caseless = char_get_othercase_bit(common, cc);
+    if (caseless == 0)
+      return FALSE;
+#ifdef COMPILE_PCRE8
+    caseless = ((caseless & 0xff) << 8) | (len - (caseless >> 8));
+#else
+    if ((caseless & 0x100) != 0)
+      caseless = ((caseless & 0xff) << 16) | (len - (caseless >> 9));
+    else
+      caseless = ((caseless & 0xff) << 8) | (len - (caseless >> 9));
+#endif
+    }
+  else
+    caseless = 0;
+
+  while (len > 0 && location < MAX_N_CHARS * 2)
+    {
+    c = *cc;
+    bit = 0;
+    if (len == (caseless & 0xff))
+      {
+      bit = caseless >> 8;
+      c |= bit;
+      }
+
+    chars[location] = c;
+    chars[location + 1] = bit;
+
+    len--;
+    location += 2;
+    cc++;
+    }
+
+  if (location >= MAX_N_CHARS * 2 || must_stop != 0)
+    break;
+  }
+
+/* At least two characters are required. */
+if (location < 2 * 2)
+    return FALSE;
+
+if (firstline)
+  {
+  SLJIT_ASSERT(common->first_line_end != 0);
+  OP1(SLJIT_MOV, TMP3, 0, STR_END, 0);
+  OP2(SLJIT_SUB, STR_END, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), common->first_line_end, SLJIT_IMM, IN_UCHARS((location >> 1) - 1));
+  }
+else
+  OP2(SLJIT_SUB, STR_END, 0, STR_END, 0, SLJIT_IMM, IN_UCHARS((location >> 1) - 1));
+
+start = LABEL();
+quit = CMP(SLJIT_C_GREATER_EQUAL, STR_PTR, 0, STR_END, 0);
+
+OP1(MOV_UCHAR, TMP1, 0, SLJIT_MEM1(STR_PTR), IN_UCHARS(0));
+OP1(MOV_UCHAR, TMP2, 0, SLJIT_MEM1(STR_PTR), IN_UCHARS(1));
+OP2(SLJIT_ADD, STR_PTR, 0, STR_PTR, 0, SLJIT_IMM, IN_UCHARS(1));
+if (chars[1] != 0)
+  OP2(SLJIT_OR, TMP1, 0, TMP1, 0, SLJIT_IMM, chars[1]);
+CMPTO(SLJIT_C_NOT_EQUAL, TMP1, 0, SLJIT_IMM, chars[0], start);
+if (location > 2 * 2)
+  OP1(MOV_UCHAR, TMP1, 0, SLJIT_MEM1(STR_PTR), IN_UCHARS(1));
+if (chars[3] != 0)
+  OP2(SLJIT_OR, TMP2, 0, TMP2, 0, SLJIT_IMM, chars[3]);
+CMPTO(SLJIT_C_NOT_EQUAL, TMP2, 0, SLJIT_IMM, chars[2], start);
+if (location > 2 * 2)
+  {
+  if (chars[5] != 0)
+    OP2(SLJIT_OR, TMP1, 0, TMP1, 0, SLJIT_IMM, chars[5]);
+  CMPTO(SLJIT_C_NOT_EQUAL, TMP1, 0, SLJIT_IMM, chars[4], start);
+  }
+OP2(SLJIT_SUB, STR_PTR, 0, STR_PTR, 0, SLJIT_IMM, IN_UCHARS(1));
+
+JUMPHERE(quit);
+
+if (firstline)
+  OP1(SLJIT_MOV, STR_END, 0, TMP3, 0);
+else
+  OP2(SLJIT_ADD, STR_END, 0, STR_END, 0, SLJIT_IMM, IN_UCHARS((location >> 1) - 1));
+return TRUE;
+}
+
+#undef MAX_N_CHARS
+
+static SLJIT_INLINE void fast_forward_first_char(compiler_common *common, pcre_uchar first_char, BOOL caseless, BOOL firstline)
+{
+DEFINE_COMPILER;
+struct sljit_label *start;
+struct sljit_jump *quit;
+struct sljit_jump *found;
+pcre_uchar oc, bit;
+
+if (firstline)
+  {
+  SLJIT_ASSERT(common->first_line_end != 0);
+  OP1(SLJIT_MOV, TMP3, 0, STR_END, 0);
+  OP1(SLJIT_MOV, STR_END, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), common->first_line_end);
+  }
+
+start = LABEL();
+quit = CMP(SLJIT_C_GREATER_EQUAL, STR_PTR, 0, STR_END, 0);
+OP1(MOV_UCHAR, TMP1, 0, SLJIT_MEM1(STR_PTR), 0);
+
+oc = first_char;
+if (caseless)
+  {
+  oc = TABLE_GET(first_char, common->fcc, first_char);
+#if defined SUPPORT_UCP && !(defined COMPILE_PCRE8)
+  if (first_char > 127 && common->utf)
+    oc = UCD_OTHERCASE(first_char);
+#endif
+  }
+if (first_char == oc)
+  found = CMP(SLJIT_C_EQUAL, TMP1, 0, SLJIT_IMM, first_char);
+else
+  {
+  bit = first_char ^ oc;
+  if (is_powerof2(bit))
+    {
+    OP2(SLJIT_OR, TMP2, 0, TMP1, 0, SLJIT_IMM, bit);
+    found = CMP(SLJIT_C_EQUAL, TMP2, 0, SLJIT_IMM, first_char | bit);
+    }
+  else
+    {
+    OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, first_char);
+    OP_FLAGS(SLJIT_MOV, TMP2, 0, SLJIT_UNUSED, 0, SLJIT_C_EQUAL);
+    OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, oc);
+    OP_FLAGS(SLJIT_OR | SLJIT_SET_E, TMP2, 0, TMP2, 0, SLJIT_C_EQUAL);
+    found = JUMP(SLJIT_C_NOT_ZERO);
+    }
+  }
+
+OP2(SLJIT_ADD, STR_PTR, 0, STR_PTR, 0, SLJIT_IMM, IN_UCHARS(1));
+JUMPTO(SLJIT_JUMP, start);
+JUMPHERE(found);
+JUMPHERE(quit);
+
+if (firstline)
+  OP1(SLJIT_MOV, STR_END, 0, TMP3, 0);
+}
+
+static SLJIT_INLINE void fast_forward_newline(compiler_common *common, BOOL firstline)
+{
+DEFINE_COMPILER;
+struct sljit_label *loop;
+struct sljit_jump *lastchar;
+struct sljit_jump *firstchar;
+struct sljit_jump *quit;
+struct sljit_jump *foundcr = NULL;
+struct sljit_jump *notfoundnl;
+jump_list *newline = NULL;
+
+if (firstline)
+  {
+  SLJIT_ASSERT(common->first_line_end != 0);
+  OP1(SLJIT_MOV, TMP3, 0, STR_END, 0);
+  OP1(SLJIT_MOV, STR_END, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), common->first_line_end);
+  }
+
+if (common->nltype == NLTYPE_FIXED && common->newline > 255)
+  {
+  lastchar = CMP(SLJIT_C_GREATER_EQUAL, STR_PTR, 0, STR_END, 0);
+  OP1(SLJIT_MOV, TMP1, 0, ARGUMENTS, 0);
+  OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(TMP1), SLJIT_OFFSETOF(jit_arguments, str));
+  OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(TMP1), SLJIT_OFFSETOF(jit_arguments, begin));
+  firstchar = CMP(SLJIT_C_LESS_EQUAL, STR_PTR, 0, TMP2, 0);
+
+  OP2(SLJIT_ADD, TMP1, 0, TMP1, 0, SLJIT_IMM, IN_UCHARS(2));
+  OP2(SLJIT_SUB | SLJIT_SET_U, SLJIT_UNUSED, 0, STR_PTR, 0, TMP1, 0);
+  OP_FLAGS(SLJIT_MOV, TMP2, 0, SLJIT_UNUSED, 0, SLJIT_C_GREATER_EQUAL);
+#if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
+  OP2(SLJIT_SHL, TMP2, 0, TMP2, 0, SLJIT_IMM, UCHAR_SHIFT);
+#endif
+  OP2(SLJIT_SUB, STR_PTR, 0, STR_PTR, 0, TMP2, 0);
+
+  loop = LABEL();
+  OP2(SLJIT_ADD, STR_PTR, 0, STR_PTR, 0, SLJIT_IMM, IN_UCHARS(1));
+  quit = CMP(SLJIT_C_GREATER_EQUAL, STR_PTR, 0, STR_END, 0);
+  OP1(MOV_UCHAR, TMP1, 0, SLJIT_MEM1(STR_PTR), IN_UCHARS(-2));
+  OP1(MOV_UCHAR, TMP2, 0, SLJIT_MEM1(STR_PTR), IN_UCHARS(-1));
+  CMPTO(SLJIT_C_NOT_EQUAL, TMP1, 0, SLJIT_IMM, (common->newline >> 8) & 0xff, loop);
+  CMPTO(SLJIT_C_NOT_EQUAL, TMP2, 0, SLJIT_IMM, common->newline & 0xff, loop);
+
+  JUMPHERE(quit);
+  JUMPHERE(firstchar);
+  JUMPHERE(lastchar);
+
+  if (firstline)
+    OP1(SLJIT_MOV, STR_END, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), POSSESSIVE0);
+  return;
+  }
+
+OP1(SLJIT_MOV, TMP1, 0, ARGUMENTS, 0);
+OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(TMP1), SLJIT_OFFSETOF(jit_arguments, str));
+firstchar = CMP(SLJIT_C_LESS_EQUAL, STR_PTR, 0, TMP2, 0);
+skip_char_back(common);
+
+loop = LABEL();
+read_char(common);
+lastchar = CMP(SLJIT_C_GREATER_EQUAL, STR_PTR, 0, STR_END, 0);
+if (common->nltype == NLTYPE_ANY || common->nltype == NLTYPE_ANYCRLF)
+  foundcr = CMP(SLJIT_C_EQUAL, TMP1, 0, SLJIT_IMM, CHAR_CR);
+check_newlinechar(common, common->nltype, &newline, FALSE);
+set_jumps(newline, loop);
+
+if (common->nltype == NLTYPE_ANY || common->nltype == NLTYPE_ANYCRLF)
+  {
+  quit = JUMP(SLJIT_JUMP);
+  JUMPHERE(foundcr);
+  notfoundnl = CMP(SLJIT_C_GREATER_EQUAL, STR_PTR, 0, STR_END, 0);
+  OP1(MOV_UCHAR, TMP1, 0, SLJIT_MEM1(STR_PTR), 0);
+  OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, CHAR_NL);
+  OP_FLAGS(SLJIT_MOV, TMP1, 0, SLJIT_UNUSED, 0, SLJIT_C_EQUAL);
+#if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
+  OP2(SLJIT_SHL, TMP1, 0, TMP1, 0, SLJIT_IMM, UCHAR_SHIFT);
+#endif
+  OP2(SLJIT_ADD, STR_PTR, 0, STR_PTR, 0, TMP1, 0);
+  JUMPHERE(notfoundnl);
+  JUMPHERE(quit);
+  }
+JUMPHERE(lastchar);
+JUMPHERE(firstchar);
+
+if (firstline)
+  OP1(SLJIT_MOV, STR_END, 0, TMP3, 0);
+}
+
+static BOOL check_class_ranges(compiler_common *common, const pcre_uint8 *bits, BOOL nclass, jump_list **backtracks);
+
+static SLJIT_INLINE void fast_forward_start_bits(compiler_common *common, sljit_uw start_bits, BOOL firstline)
+{
+DEFINE_COMPILER;
+struct sljit_label *start;
+struct sljit_jump *quit;
+struct sljit_jump *found = NULL;
+jump_list *matches = NULL;
+pcre_uint8 inverted_start_bits[32];
+int i;
+#ifndef COMPILE_PCRE8
+struct sljit_jump *jump;
+#endif
+
+for (i = 0; i < 32; ++i)
+  inverted_start_bits[i] = ~(((pcre_uint8*)start_bits)[i]);
+
+if (firstline)
+  {
+  SLJIT_ASSERT(common->first_line_end != 0);
+  OP1(SLJIT_MOV, RETURN_ADDR, 0, STR_END, 0);
+  OP1(SLJIT_MOV, STR_END, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), common->first_line_end);
+  }
+
+start = LABEL();
+quit = CMP(SLJIT_C_GREATER_EQUAL, STR_PTR, 0, STR_END, 0);
+OP1(MOV_UCHAR, TMP1, 0, SLJIT_MEM1(STR_PTR), 0);
+#ifdef SUPPORT_UTF
+if (common->utf)
+  OP1(SLJIT_MOV, TMP3, 0, TMP1, 0);
+#endif
+
+if (!check_class_ranges(common, inverted_start_bits, (inverted_start_bits[31] & 0x80) != 0, &matches))
+  {
+#ifndef COMPILE_PCRE8
+  jump = CMP(SLJIT_C_LESS, TMP1, 0, SLJIT_IMM, 255);
+  OP1(SLJIT_MOV, TMP1, 0, SLJIT_IMM, 255);
+  JUMPHERE(jump);
+#endif
+  OP2(SLJIT_AND, TMP2, 0, TMP1, 0, SLJIT_IMM, 0x7);
+  OP2(SLJIT_LSHR, TMP1, 0, TMP1, 0, SLJIT_IMM, 3);
+  OP1(SLJIT_MOV_UB, TMP1, 0, SLJIT_MEM1(TMP1), start_bits);
+  OP2(SLJIT_SHL, TMP2, 0, SLJIT_IMM, 1, TMP2, 0);
+  OP2(SLJIT_AND | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP1, 0, TMP2, 0);
+  found = JUMP(SLJIT_C_NOT_ZERO);
+  }
+
+#ifdef SUPPORT_UTF
+if (common->utf)
+  OP1(SLJIT_MOV, TMP1, 0, TMP3, 0);
+#endif
+OP2(SLJIT_ADD, STR_PTR, 0, STR_PTR, 0, SLJIT_IMM, IN_UCHARS(1));
+#ifdef SUPPORT_UTF
+#if defined COMPILE_PCRE8
+if (common->utf)
+  {
+  CMPTO(SLJIT_C_LESS, TMP1, 0, SLJIT_IMM, 0xc0, start);
+  OP1(SLJIT_MOV_UB, TMP1, 0, SLJIT_MEM1(TMP1), (sljit_sw)PRIV(utf8_table4) - 0xc0);
+  OP2(SLJIT_ADD, STR_PTR, 0, STR_PTR, 0, TMP1, 0);
+  }
+#elif defined COMPILE_PCRE16
+if (common->utf)
+  {
+  CMPTO(SLJIT_C_LESS, TMP1, 0, SLJIT_IMM, 0xd800, start);
+  OP2(SLJIT_AND, TMP1, 0, TMP1, 0, SLJIT_IMM, 0xfc00);
+  OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, 0xd800);
+  OP_FLAGS(SLJIT_MOV, TMP1, 0, SLJIT_UNUSED, 0, SLJIT_C_EQUAL);
+  OP2(SLJIT_SHL, TMP1, 0, TMP1, 0, SLJIT_IMM, 1);
+  OP2(SLJIT_ADD, STR_PTR, 0, STR_PTR, 0, TMP1, 0);
+  }
+#endif /* COMPILE_PCRE[8|16] */
+#endif /* SUPPORT_UTF */
+JUMPTO(SLJIT_JUMP, start);
+if (found != NULL)
+  JUMPHERE(found);
+if (matches != NULL)
+  set_jumps(matches, LABEL());
+JUMPHERE(quit);
+
+if (firstline)
+  OP1(SLJIT_MOV, STR_END, 0, RETURN_ADDR, 0);
+}
+
+static SLJIT_INLINE struct sljit_jump *search_requested_char(compiler_common *common, pcre_uchar req_char, BOOL caseless, BOOL has_firstchar)
+{
+DEFINE_COMPILER;
+struct sljit_label *loop;
+struct sljit_jump *toolong;
+struct sljit_jump *alreadyfound;
+struct sljit_jump *found;
+struct sljit_jump *foundoc = NULL;
+struct sljit_jump *notfound;
+pcre_uint32 oc, bit;
+
+SLJIT_ASSERT(common->req_char_ptr != 0);
+OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), common->req_char_ptr);
+OP2(SLJIT_ADD, TMP1, 0, STR_PTR, 0, SLJIT_IMM, REQ_BYTE_MAX);
+toolong = CMP(SLJIT_C_LESS, TMP1, 0, STR_END, 0);
+alreadyfound = CMP(SLJIT_C_LESS, STR_PTR, 0, TMP2, 0);
+
+if (has_firstchar)
+  OP2(SLJIT_ADD, TMP1, 0, STR_PTR, 0, SLJIT_IMM, IN_UCHARS(1));
+else
+  OP1(SLJIT_MOV, TMP1, 0, STR_PTR, 0);
+
+loop = LABEL();
+notfound = CMP(SLJIT_C_GREATER_EQUAL, TMP1, 0, STR_END, 0);
+
+OP1(MOV_UCHAR, TMP2, 0, SLJIT_MEM1(TMP1), 0);
+oc = req_char;
+if (caseless)
+  {
+  oc = TABLE_GET(req_char, common->fcc, req_char);
+#if defined SUPPORT_UCP && !(defined COMPILE_PCRE8)
+  if (req_char > 127 && common->utf)
+    oc = UCD_OTHERCASE(req_char);
+#endif
+  }
+if (req_char == oc)
+  found = CMP(SLJIT_C_EQUAL, TMP2, 0, SLJIT_IMM, req_char);
+else
+  {
+  bit = req_char ^ oc;
+  if (is_powerof2(bit))
+    {
+    OP2(SLJIT_OR, TMP2, 0, TMP2, 0, SLJIT_IMM, bit);
+    found = CMP(SLJIT_C_EQUAL, TMP2, 0, SLJIT_IMM, req_char | bit);
+    }
+  else
+    {
+    found = CMP(SLJIT_C_EQUAL, TMP2, 0, SLJIT_IMM, req_char);
+    foundoc = CMP(SLJIT_C_EQUAL, TMP2, 0, SLJIT_IMM, oc);
+    }
+  }
+OP2(SLJIT_ADD, TMP1, 0, TMP1, 0, SLJIT_IMM, IN_UCHARS(1));
+JUMPTO(SLJIT_JUMP, loop);
+
+JUMPHERE(found);
+if (foundoc)
+  JUMPHERE(foundoc);
+OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->req_char_ptr, TMP1, 0);
+JUMPHERE(alreadyfound);
+JUMPHERE(toolong);
+return notfound;
+}
+
+static void do_revertframes(compiler_common *common)
+{
+DEFINE_COMPILER;
+struct sljit_jump *jump;
+struct sljit_label *mainloop;
+
+sljit_emit_fast_enter(compiler, RETURN_ADDR, 0);
+OP1(SLJIT_MOV, TMP1, 0, STACK_TOP, 0);
+GET_LOCAL_BASE(TMP3, 0, 0);
+
+/* Drop frames until we reach STACK_TOP. */
+mainloop = LABEL();
+OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(TMP1), 0);
+OP2(SLJIT_SUB | SLJIT_SET_S, SLJIT_UNUSED, 0, TMP2, 0, SLJIT_IMM, 0);
+jump = JUMP(SLJIT_C_SIG_LESS_EQUAL);
+
+OP2(SLJIT_ADD, TMP2, 0, TMP2, 0, TMP3, 0);
+OP1(SLJIT_MOV, SLJIT_MEM1(TMP2), 0, SLJIT_MEM1(TMP1), sizeof(sljit_sw));
+OP1(SLJIT_MOV, SLJIT_MEM1(TMP2), sizeof(sljit_sw), SLJIT_MEM1(TMP1), 2 * sizeof(sljit_sw));
+OP2(SLJIT_ADD, TMP1, 0, TMP1, 0, SLJIT_IMM, 3 * sizeof(sljit_sw));
+JUMPTO(SLJIT_JUMP, mainloop);
+
+JUMPHERE(jump);
+jump = JUMP(SLJIT_C_SIG_LESS);
+/* End of dropping frames. */
+sljit_emit_fast_return(compiler, RETURN_ADDR, 0);
+
+JUMPHERE(jump);
+OP1(SLJIT_NEG, TMP2, 0, TMP2, 0);
+OP2(SLJIT_ADD, TMP2, 0, TMP2, 0, TMP3, 0);
+OP1(SLJIT_MOV, SLJIT_MEM1(TMP2), 0, SLJIT_MEM1(TMP1), sizeof(sljit_sw));
+OP2(SLJIT_ADD, TMP1, 0, TMP1, 0, SLJIT_IMM, 2 * sizeof(sljit_sw));
+JUMPTO(SLJIT_JUMP, mainloop);
+}
+
+static void check_wordboundary(compiler_common *common)
+{
+DEFINE_COMPILER;
+struct sljit_jump *skipread;
+jump_list *skipread_list = NULL;
+#if !(defined COMPILE_PCRE8) || defined SUPPORT_UTF
+struct sljit_jump *jump;
+#endif
+
+SLJIT_COMPILE_ASSERT(ctype_word == 0x10, ctype_word_must_be_16);
+
+sljit_emit_fast_enter(compiler, SLJIT_MEM1(SLJIT_LOCALS_REG), LOCALS0);
+/* Get type of the previous char, and put it to LOCALS1. */
+OP1(SLJIT_MOV, TMP1, 0, ARGUMENTS, 0);
+OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(TMP1), SLJIT_OFFSETOF(jit_arguments, begin));
+OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), LOCALS1, SLJIT_IMM, 0);
+skipread = CMP(SLJIT_C_LESS_EQUAL, STR_PTR, 0, TMP1, 0);
+skip_char_back(common);
+check_start_used_ptr(common);
+read_char(common);
+
+/* Testing char type. */
+#ifdef SUPPORT_UCP
+if (common->use_ucp)
+  {
+  OP1(SLJIT_MOV, TMP2, 0, SLJIT_IMM, 1);
+  jump = CMP(SLJIT_C_EQUAL, TMP1, 0, SLJIT_IMM, CHAR_UNDERSCORE);
+  add_jump(compiler, &common->getucd, JUMP(SLJIT_FAST_CALL));
+  OP2(SLJIT_SUB, TMP1, 0, TMP1, 0, SLJIT_IMM, ucp_Ll);
+  OP2(SLJIT_SUB | SLJIT_SET_U, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, ucp_Lu - ucp_Ll);
+  OP_FLAGS(SLJIT_MOV, TMP2, 0, SLJIT_UNUSED, 0, SLJIT_C_LESS_EQUAL);
+  OP2(SLJIT_SUB, TMP1, 0, TMP1, 0, SLJIT_IMM, ucp_Nd - ucp_Ll);
+  OP2(SLJIT_SUB | SLJIT_SET_U, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, ucp_No - ucp_Nd);
+  OP_FLAGS(SLJIT_OR, TMP2, 0, TMP2, 0, SLJIT_C_LESS_EQUAL);
+  JUMPHERE(jump);
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), LOCALS1, TMP2, 0);
+  }
+else
+#endif
+  {
+#ifndef COMPILE_PCRE8
+  jump = CMP(SLJIT_C_GREATER, TMP1, 0, SLJIT_IMM, 255);
+#elif defined SUPPORT_UTF
+  /* Here LOCALS1 has already been zeroed. */
+  jump = NULL;
+  if (common->utf)
+    jump = CMP(SLJIT_C_GREATER, TMP1, 0, SLJIT_IMM, 255);
+#endif /* COMPILE_PCRE8 */
+  OP1(SLJIT_MOV_UB, TMP1, 0, SLJIT_MEM1(TMP1), common->ctypes);
+  OP2(SLJIT_LSHR, TMP1, 0, TMP1, 0, SLJIT_IMM, 4 /* ctype_word */);
+  OP2(SLJIT_AND, TMP1, 0, TMP1, 0, SLJIT_IMM, 1);
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), LOCALS1, TMP1, 0);
+#ifndef COMPILE_PCRE8
+  JUMPHERE(jump);
+#elif defined SUPPORT_UTF
+  if (jump != NULL)
+    JUMPHERE(jump);
+#endif /* COMPILE_PCRE8 */
+  }
+JUMPHERE(skipread);
+
+OP1(SLJIT_MOV, TMP2, 0, SLJIT_IMM, 0);
+check_str_end(common, &skipread_list);
+peek_char(common);
+
+/* Testing char type. This is a code duplication. */
+#ifdef SUPPORT_UCP
+if (common->use_ucp)
+  {
+  OP1(SLJIT_MOV, TMP2, 0, SLJIT_IMM, 1);
+  jump = CMP(SLJIT_C_EQUAL, TMP1, 0, SLJIT_IMM, CHAR_UNDERSCORE);
+  add_jump(compiler, &common->getucd, JUMP(SLJIT_FAST_CALL));
+  OP2(SLJIT_SUB, TMP1, 0, TMP1, 0, SLJIT_IMM, ucp_Ll);
+  OP2(SLJIT_SUB | SLJIT_SET_U, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, ucp_Lu - ucp_Ll);
+  OP_FLAGS(SLJIT_MOV, TMP2, 0, SLJIT_UNUSED, 0, SLJIT_C_LESS_EQUAL);
+  OP2(SLJIT_SUB, TMP1, 0, TMP1, 0, SLJIT_IMM, ucp_Nd - ucp_Ll);
+  OP2(SLJIT_SUB | SLJIT_SET_U, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, ucp_No - ucp_Nd);
+  OP_FLAGS(SLJIT_OR, TMP2, 0, TMP2, 0, SLJIT_C_LESS_EQUAL);
+  JUMPHERE(jump);
+  }
+else
+#endif
+  {
+#ifndef COMPILE_PCRE8
+  /* TMP2 may be destroyed by peek_char. */
+  OP1(SLJIT_MOV, TMP2, 0, SLJIT_IMM, 0);
+  jump = CMP(SLJIT_C_GREATER, TMP1, 0, SLJIT_IMM, 255);
+#elif defined SUPPORT_UTF
+  OP1(SLJIT_MOV, TMP2, 0, SLJIT_IMM, 0);
+  jump = NULL;
+  if (common->utf)
+    jump = CMP(SLJIT_C_GREATER, TMP1, 0, SLJIT_IMM, 255);
+#endif
+  OP1(SLJIT_MOV_UB, TMP2, 0, SLJIT_MEM1(TMP1), common->ctypes);
+  OP2(SLJIT_LSHR, TMP2, 0, TMP2, 0, SLJIT_IMM, 4 /* ctype_word */);
+  OP2(SLJIT_AND, TMP2, 0, TMP2, 0, SLJIT_IMM, 1);
+#ifndef COMPILE_PCRE8
+  JUMPHERE(jump);
+#elif defined SUPPORT_UTF
+  if (jump != NULL)
+    JUMPHERE(jump);
+#endif /* COMPILE_PCRE8 */
+  }
+set_jumps(skipread_list, LABEL());
+
+OP2(SLJIT_XOR | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP2, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), LOCALS1);
+sljit_emit_fast_return(compiler, SLJIT_MEM1(SLJIT_LOCALS_REG), LOCALS0);
+}
+
+/*
+  range format:
+
+  ranges[0] = length of the range (max MAX_RANGE_SIZE, -1 means invalid range).
+  ranges[1] = first bit (0 or 1)
+  ranges[2-length] = position of the bit change (when the current bit is not equal to the previous)
+*/
+
+static BOOL check_ranges(compiler_common *common, int *ranges, jump_list **backtracks, BOOL readch)
+{
+DEFINE_COMPILER;
+struct sljit_jump *jump;
+
+if (ranges[0] < 0)
+  return FALSE;
+
+switch(ranges[0])
+  {
+  case 1:
+  if (readch)
+    read_char(common);
+  add_jump(compiler, backtracks, CMP(ranges[1] == 0 ? SLJIT_C_LESS : SLJIT_C_GREATER_EQUAL, TMP1, 0, SLJIT_IMM, ranges[2]));
+  return TRUE;
+
+  case 2:
+  if (readch)
+    read_char(common);
+  OP2(SLJIT_SUB, TMP1, 0, TMP1, 0, SLJIT_IMM, ranges[2]);
+  add_jump(compiler, backtracks, CMP(ranges[1] != 0 ? SLJIT_C_LESS : SLJIT_C_GREATER_EQUAL, TMP1, 0, SLJIT_IMM, ranges[3] - ranges[2]));
+  return TRUE;
+
+  case 4:
+  if (ranges[2] + 1 == ranges[3] && ranges[4] + 1 == ranges[5])
+    {
+    if (readch)
+      read_char(common);
+    if (ranges[1] != 0)
+      {
+      add_jump(compiler, backtracks, CMP(SLJIT_C_EQUAL, TMP1, 0, SLJIT_IMM, ranges[2]));
+      add_jump(compiler, backtracks, CMP(SLJIT_C_EQUAL, TMP1, 0, SLJIT_IMM, ranges[4]));
+      }
+    else
+      {
+      jump = CMP(SLJIT_C_EQUAL, TMP1, 0, SLJIT_IMM, ranges[2]);
+      add_jump(compiler, backtracks, CMP(SLJIT_C_NOT_EQUAL, TMP1, 0, SLJIT_IMM, ranges[4]));
+      JUMPHERE(jump);
+      }
+    return TRUE;
+    }
+  if ((ranges[3] - ranges[2]) == (ranges[5] - ranges[4]) && is_powerof2(ranges[4] - ranges[2]))
+    {
+    if (readch)
+      read_char(common);
+    OP2(SLJIT_OR, TMP1, 0, TMP1, 0, SLJIT_IMM, ranges[4] - ranges[2]);
+    OP2(SLJIT_SUB, TMP1, 0, TMP1, 0, SLJIT_IMM, ranges[4]);
+    add_jump(compiler, backtracks, CMP(ranges[1] != 0 ? SLJIT_C_LESS : SLJIT_C_GREATER_EQUAL, TMP1, 0, SLJIT_IMM, ranges[5] - ranges[4]));
+    return TRUE;
+    }
+  return FALSE;
+
+  default:
+  return FALSE;
+  }
+}
+
+static void get_ctype_ranges(compiler_common *common, int flag, int *ranges)
+{
+int i, bit, length;
+const pcre_uint8 *ctypes = (const pcre_uint8*)common->ctypes;
+
+bit = ctypes[0] & flag;
+ranges[0] = -1;
+ranges[1] = bit != 0 ? 1 : 0;
+length = 0;
+
+for (i = 1; i < 256; i++)
+  if ((ctypes[i] & flag) != bit)
+    {
+    if (length >= MAX_RANGE_SIZE)
+      return;
+    ranges[2 + length] = i;
+    length++;
+    bit ^= flag;
+    }
+
+if (bit != 0)
+  {
+  if (length >= MAX_RANGE_SIZE)
+    return;
+  ranges[2 + length] = 256;
+  length++;
+  }
+ranges[0] = length;
+}
+
+static BOOL check_class_ranges(compiler_common *common, const pcre_uint8 *bits, BOOL nclass, jump_list **backtracks)
+{
+int ranges[2 + MAX_RANGE_SIZE];
+pcre_uint8 bit, cbit, all;
+int i, byte, length = 0;
+
+bit = bits[0] & 0x1;
+ranges[1] = bit;
+/* Can be 0 or 255. */
+all = -bit;
+
+for (i = 0; i < 256; )
+  {
+  byte = i >> 3;
+  if ((i & 0x7) == 0 && bits[byte] == all)
+    i += 8;
+  else
+    {
+    cbit = (bits[byte] >> (i & 0x7)) & 0x1;
+    if (cbit != bit)
+      {
+      if (length >= MAX_RANGE_SIZE)
+        return FALSE;
+      ranges[2 + length] = i;
+      length++;
+      bit = cbit;
+      all = -cbit;
+      }
+    i++;
+    }
+  }
+
+if (((bit == 0) && nclass) || ((bit == 1) && !nclass))
+  {
+  if (length >= MAX_RANGE_SIZE)
+    return FALSE;
+  ranges[2 + length] = 256;
+  length++;
+  }
+ranges[0] = length;
+
+return check_ranges(common, ranges, backtracks, FALSE);
+}
+
+static void check_anynewline(compiler_common *common)
+{
+/* Check whether TMP1 contains a newline character. TMP2 destroyed. */
+DEFINE_COMPILER;
+
+sljit_emit_fast_enter(compiler, RETURN_ADDR, 0);
+
+OP2(SLJIT_SUB, TMP1, 0, TMP1, 0, SLJIT_IMM, 0x0a);
+OP2(SLJIT_SUB | SLJIT_SET_U, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, 0x0d - 0x0a);
+OP_FLAGS(SLJIT_MOV, TMP2, 0, SLJIT_UNUSED, 0, SLJIT_C_LESS_EQUAL);
+OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, 0x85 - 0x0a);
+#if defined SUPPORT_UTF || defined COMPILE_PCRE16 || defined COMPILE_PCRE32
+#ifdef COMPILE_PCRE8
+if (common->utf)
+  {
+#endif
+  OP_FLAGS(SLJIT_OR, TMP2, 0, TMP2, 0, SLJIT_C_EQUAL);
+  OP2(SLJIT_OR, TMP1, 0, TMP1, 0, SLJIT_IMM, 0x1);
+  OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, 0x2029 - 0x0a);
+#ifdef COMPILE_PCRE8
+  }
+#endif
+#endif /* SUPPORT_UTF || COMPILE_PCRE16 || COMPILE_PCRE32 */
+OP_FLAGS(SLJIT_OR | SLJIT_SET_E, TMP2, 0, TMP2, 0, SLJIT_C_EQUAL);
+sljit_emit_fast_return(compiler, RETURN_ADDR, 0);
+}
+
+static void check_hspace(compiler_common *common)
+{
+/* Check whether TMP1 contains a newline character. TMP2 destroyed. */
+DEFINE_COMPILER;
+
+sljit_emit_fast_enter(compiler, RETURN_ADDR, 0);
+
+OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, 0x09);
+OP_FLAGS(SLJIT_MOV, TMP2, 0, SLJIT_UNUSED, 0, SLJIT_C_EQUAL);
+OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, 0x20);
+OP_FLAGS(SLJIT_OR, TMP2, 0, TMP2, 0, SLJIT_C_EQUAL);
+OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, 0xa0);
+#if defined SUPPORT_UTF || defined COMPILE_PCRE16 || defined COMPILE_PCRE32
+#ifdef COMPILE_PCRE8
+if (common->utf)
+  {
+#endif
+  OP_FLAGS(SLJIT_OR, TMP2, 0, TMP2, 0, SLJIT_C_EQUAL);
+  OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, 0x1680);
+  OP_FLAGS(SLJIT_OR, TMP2, 0, TMP2, 0, SLJIT_C_EQUAL);
+  OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, 0x180e);
+  OP_FLAGS(SLJIT_OR, TMP2, 0, TMP2, 0, SLJIT_C_EQUAL);
+  OP2(SLJIT_SUB, TMP1, 0, TMP1, 0, SLJIT_IMM, 0x2000);
+  OP2(SLJIT_SUB | SLJIT_SET_U, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, 0x200A - 0x2000);
+  OP_FLAGS(SLJIT_OR, TMP2, 0, TMP2, 0, SLJIT_C_LESS_EQUAL);
+  OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, 0x202f - 0x2000);
+  OP_FLAGS(SLJIT_OR, TMP2, 0, TMP2, 0, SLJIT_C_EQUAL);
+  OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, 0x205f - 0x2000);
+  OP_FLAGS(SLJIT_OR, TMP2, 0, TMP2, 0, SLJIT_C_EQUAL);
+  OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, 0x3000 - 0x2000);
+#ifdef COMPILE_PCRE8
+  }
+#endif
+#endif /* SUPPORT_UTF || COMPILE_PCRE16 || COMPILE_PCRE32 */
+OP_FLAGS(SLJIT_OR | SLJIT_SET_E, TMP2, 0, TMP2, 0, SLJIT_C_EQUAL);
+
+sljit_emit_fast_return(compiler, RETURN_ADDR, 0);
+}
+
+static void check_vspace(compiler_common *common)
+{
+/* Check whether TMP1 contains a newline character. TMP2 destroyed. */
+DEFINE_COMPILER;
+
+sljit_emit_fast_enter(compiler, RETURN_ADDR, 0);
+
+OP2(SLJIT_SUB, TMP1, 0, TMP1, 0, SLJIT_IMM, 0x0a);
+OP2(SLJIT_SUB | SLJIT_SET_U, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, 0x0d - 0x0a);
+OP_FLAGS(SLJIT_MOV, TMP2, 0, SLJIT_UNUSED, 0, SLJIT_C_LESS_EQUAL);
+OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, 0x85 - 0x0a);
+#if defined SUPPORT_UTF || defined COMPILE_PCRE16 || defined COMPILE_PCRE32
+#ifdef COMPILE_PCRE8
+if (common->utf)
+  {
+#endif
+  OP_FLAGS(SLJIT_OR | SLJIT_SET_E, TMP2, 0, TMP2, 0, SLJIT_C_EQUAL);
+  OP2(SLJIT_OR, TMP1, 0, TMP1, 0, SLJIT_IMM, 0x1);
+  OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, 0x2029 - 0x0a);
+#ifdef COMPILE_PCRE8
+  }
+#endif
+#endif /* SUPPORT_UTF || COMPILE_PCRE16 || COMPILE_PCRE32 */
+OP_FLAGS(SLJIT_OR | SLJIT_SET_E, TMP2, 0, TMP2, 0, SLJIT_C_EQUAL);
+
+sljit_emit_fast_return(compiler, RETURN_ADDR, 0);
+}
+
+#define CHAR1 STR_END
+#define CHAR2 STACK_TOP
+
+static void do_casefulcmp(compiler_common *common)
+{
+DEFINE_COMPILER;
+struct sljit_jump *jump;
+struct sljit_label *label;
+
+sljit_emit_fast_enter(compiler, RETURN_ADDR, 0);
+OP2(SLJIT_SUB, STR_PTR, 0, STR_PTR, 0, TMP2, 0);
+OP1(SLJIT_MOV, TMP3, 0, CHAR1, 0);
+OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), LOCALS0, CHAR2, 0);
+OP2(SLJIT_SUB, TMP1, 0, TMP1, 0, SLJIT_IMM, IN_UCHARS(1));
+OP2(SLJIT_SUB, STR_PTR, 0, STR_PTR, 0, SLJIT_IMM, IN_UCHARS(1));
+
+label = LABEL();
+OP1(MOVU_UCHAR, CHAR1, 0, SLJIT_MEM1(TMP1), IN_UCHARS(1));
+OP1(MOVU_UCHAR, CHAR2, 0, SLJIT_MEM1(STR_PTR), IN_UCHARS(1));
+jump = CMP(SLJIT_C_NOT_EQUAL, CHAR1, 0, CHAR2, 0);
+OP2(SLJIT_SUB | SLJIT_SET_E, TMP2, 0, TMP2, 0, SLJIT_IMM, IN_UCHARS(1));
+JUMPTO(SLJIT_C_NOT_ZERO, label);
+
+JUMPHERE(jump);
+OP2(SLJIT_ADD, STR_PTR, 0, STR_PTR, 0, SLJIT_IMM, IN_UCHARS(1));
+OP1(SLJIT_MOV, CHAR1, 0, TMP3, 0);
+OP1(SLJIT_MOV, CHAR2, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), LOCALS0);
+sljit_emit_fast_return(compiler, RETURN_ADDR, 0);
+}
+
+#define LCC_TABLE STACK_LIMIT
+
+static void do_caselesscmp(compiler_common *common)
+{
+DEFINE_COMPILER;
+struct sljit_jump *jump;
+struct sljit_label *label;
+
+sljit_emit_fast_enter(compiler, RETURN_ADDR, 0);
+OP2(SLJIT_SUB, STR_PTR, 0, STR_PTR, 0, TMP2, 0);
+
+OP1(SLJIT_MOV, TMP3, 0, LCC_TABLE, 0);
+OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), LOCALS0, CHAR1, 0);
+OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), LOCALS1, CHAR2, 0);
+OP1(SLJIT_MOV, LCC_TABLE, 0, SLJIT_IMM, common->lcc);
+OP2(SLJIT_SUB, TMP1, 0, TMP1, 0, SLJIT_IMM, IN_UCHARS(1));
+OP2(SLJIT_SUB, STR_PTR, 0, STR_PTR, 0, SLJIT_IMM, IN_UCHARS(1));
+
+label = LABEL();
+OP1(MOVU_UCHAR, CHAR1, 0, SLJIT_MEM1(TMP1), IN_UCHARS(1));
+OP1(MOVU_UCHAR, CHAR2, 0, SLJIT_MEM1(STR_PTR), IN_UCHARS(1));
+#ifndef COMPILE_PCRE8
+jump = CMP(SLJIT_C_GREATER, CHAR1, 0, SLJIT_IMM, 255);
+#endif
+OP1(SLJIT_MOV_UB, CHAR1, 0, SLJIT_MEM2(LCC_TABLE, CHAR1), 0);
+#ifndef COMPILE_PCRE8
+JUMPHERE(jump);
+jump = CMP(SLJIT_C_GREATER, CHAR2, 0, SLJIT_IMM, 255);
+#endif
+OP1(SLJIT_MOV_UB, CHAR2, 0, SLJIT_MEM2(LCC_TABLE, CHAR2), 0);
+#ifndef COMPILE_PCRE8
+JUMPHERE(jump);
+#endif
+jump = CMP(SLJIT_C_NOT_EQUAL, CHAR1, 0, CHAR2, 0);
+OP2(SLJIT_SUB | SLJIT_SET_E, TMP2, 0, TMP2, 0, SLJIT_IMM, IN_UCHARS(1));
+JUMPTO(SLJIT_C_NOT_ZERO, label);
+
+JUMPHERE(jump);
+OP2(SLJIT_ADD, STR_PTR, 0, STR_PTR, 0, SLJIT_IMM, IN_UCHARS(1));
+OP1(SLJIT_MOV, LCC_TABLE, 0, TMP3, 0);
+OP1(SLJIT_MOV, CHAR1, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), LOCALS0);
+OP1(SLJIT_MOV, CHAR2, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), LOCALS1);
+sljit_emit_fast_return(compiler, RETURN_ADDR, 0);
+}
+
+#undef LCC_TABLE
+#undef CHAR1
+#undef CHAR2
+
+#if defined SUPPORT_UTF && defined SUPPORT_UCP
+
+static const pcre_uchar * SLJIT_CALL do_utf_caselesscmp(pcre_uchar *src1, jit_arguments *args, pcre_uchar *end1)
+{
+/* This function would be ineffective to do in JIT level. */
+pcre_uint32 c1, c2;
+const pcre_uchar *src2 = args->uchar_ptr;
+const pcre_uchar *end2 = args->end;
+const ucd_record *ur;
+const pcre_uint32 *pp;
+
+while (src1 < end1)
+  {
+  if (src2 >= end2)
+    return (pcre_uchar*)1;
+  GETCHARINC(c1, src1);
+  GETCHARINC(c2, src2);
+  ur = GET_UCD(c2);
+  if (c1 != c2 && c1 != c2 + ur->other_case)
+    {
+    pp = PRIV(ucd_caseless_sets) + ur->caseset;
+    for (;;)
+      {
+      if (c1 < *pp) return NULL;
+      if (c1 == *pp++) break;
+      }
+    }
+  }
+return src2;
+}
+
+#endif /* SUPPORT_UTF && SUPPORT_UCP */
+
+static pcre_uchar *byte_sequence_compare(compiler_common *common, BOOL caseless, pcre_uchar *cc,
+    compare_context* context, jump_list **backtracks)
+{
+DEFINE_COMPILER;
+unsigned int othercasebit = 0;
+pcre_uchar *othercasechar = NULL;
+#ifdef SUPPORT_UTF
+int utflength;
+#endif
+
+if (caseless && char_has_othercase(common, cc))
+  {
+  othercasebit = char_get_othercase_bit(common, cc);
+  SLJIT_ASSERT(othercasebit);
+  /* Extracting bit difference info. */
+#if defined COMPILE_PCRE8
+  othercasechar = cc + (othercasebit >> 8);
+  othercasebit &= 0xff;
+#elif defined COMPILE_PCRE16 || defined COMPILE_PCRE32
+  /* Note that this code only handles characters in the BMP. If there
+  ever are characters outside the BMP whose othercase differs in only one
+  bit from itself (there currently are none), this code will need to be
+  revised for COMPILE_PCRE32. */
+  othercasechar = cc + (othercasebit >> 9);
+  if ((othercasebit & 0x100) != 0)
+    othercasebit = (othercasebit & 0xff) << 8;
+  else
+    othercasebit &= 0xff;
+#endif /* COMPILE_PCRE[8|16|32] */
+  }
+
+if (context->sourcereg == -1)
+  {
+#if defined COMPILE_PCRE8
+#if defined SLJIT_UNALIGNED && SLJIT_UNALIGNED
+  if (context->length >= 4)
+    OP1(SLJIT_MOV_SI, TMP1, 0, SLJIT_MEM1(STR_PTR), -context->length);
+  else if (context->length >= 2)
+    OP1(SLJIT_MOV_UH, TMP1, 0, SLJIT_MEM1(STR_PTR), -context->length);
+  else
+#endif
+    OP1(SLJIT_MOV_UB, TMP1, 0, SLJIT_MEM1(STR_PTR), -context->length);
+#elif defined COMPILE_PCRE16
+#if defined SLJIT_UNALIGNED && SLJIT_UNALIGNED
+  if (context->length >= 4)
+    OP1(SLJIT_MOV_SI, TMP1, 0, SLJIT_MEM1(STR_PTR), -context->length);
+  else
+#endif
+    OP1(MOV_UCHAR, TMP1, 0, SLJIT_MEM1(STR_PTR), -context->length);
+#elif defined COMPILE_PCRE32
+  OP1(MOV_UCHAR, TMP1, 0, SLJIT_MEM1(STR_PTR), -context->length);
+#endif /* COMPILE_PCRE[8|16|32] */
+  context->sourcereg = TMP2;
+  }
+
+#ifdef SUPPORT_UTF
+utflength = 1;
+if (common->utf && HAS_EXTRALEN(*cc))
+  utflength += GET_EXTRALEN(*cc);
+
+do
+  {
+#endif
+
+  context->length -= IN_UCHARS(1);
+#if (defined SLJIT_UNALIGNED && SLJIT_UNALIGNED) && (defined COMPILE_PCRE8 || defined COMPILE_PCRE16)
+
+  /* Unaligned read is supported. */
+  if (othercasebit != 0 && othercasechar == cc)
+    {
+    context->c.asuchars[context->ucharptr] = *cc | othercasebit;
+    context->oc.asuchars[context->ucharptr] = othercasebit;
+    }
+  else
+    {
+    context->c.asuchars[context->ucharptr] = *cc;
+    context->oc.asuchars[context->ucharptr] = 0;
+    }
+  context->ucharptr++;
+
+#if defined COMPILE_PCRE8
+  if (context->ucharptr >= 4 || context->length == 0 || (context->ucharptr == 2 && context->length == 1))
+#else
+  if (context->ucharptr >= 2 || context->length == 0)
+#endif
+    {
+    if (context->length >= 4)
+      OP1(SLJIT_MOV_SI, context->sourcereg, 0, SLJIT_MEM1(STR_PTR), -context->length);
+    else if (context->length >= 2)
+      OP1(SLJIT_MOV_UH, context->sourcereg, 0, SLJIT_MEM1(STR_PTR), -context->length);
+#if defined COMPILE_PCRE8
+    else if (context->length >= 1)
+      OP1(SLJIT_MOV_UB, context->sourcereg, 0, SLJIT_MEM1(STR_PTR), -context->length);
+#endif /* COMPILE_PCRE8 */
+    context->sourcereg = context->sourcereg == TMP1 ? TMP2 : TMP1;
+
+    switch(context->ucharptr)
+      {
+      case 4 / sizeof(pcre_uchar):
+      if (context->oc.asint != 0)
+        OP2(SLJIT_OR, context->sourcereg, 0, context->sourcereg, 0, SLJIT_IMM, context->oc.asint);
+      add_jump(compiler, backtracks, CMP(SLJIT_C_NOT_EQUAL, context->sourcereg, 0, SLJIT_IMM, context->c.asint | context->oc.asint));
+      break;
+
+      case 2 / sizeof(pcre_uchar):
+      if (context->oc.asushort != 0)
+        OP2(SLJIT_OR, context->sourcereg, 0, context->sourcereg, 0, SLJIT_IMM, context->oc.asushort);
+      add_jump(compiler, backtracks, CMP(SLJIT_C_NOT_EQUAL, context->sourcereg, 0, SLJIT_IMM, context->c.asushort | context->oc.asushort));
+      break;
+
+#ifdef COMPILE_PCRE8
+      case 1:
+      if (context->oc.asbyte != 0)
+        OP2(SLJIT_OR, context->sourcereg, 0, context->sourcereg, 0, SLJIT_IMM, context->oc.asbyte);
+      add_jump(compiler, backtracks, CMP(SLJIT_C_NOT_EQUAL, context->sourcereg, 0, SLJIT_IMM, context->c.asbyte | context->oc.asbyte));
+      break;
+#endif
+
+      default:
+      SLJIT_ASSERT_STOP();
+      break;
+      }
+    context->ucharptr = 0;
+    }
+
+#else
+
+  /* Unaligned read is unsupported or in 32 bit mode. */
+  if (context->length >= 1)
+    OP1(MOV_UCHAR, context->sourcereg, 0, SLJIT_MEM1(STR_PTR), -context->length);
+
+  context->sourcereg = context->sourcereg == TMP1 ? TMP2 : TMP1;
+
+  if (othercasebit != 0 && othercasechar == cc)
+    {
+    OP2(SLJIT_OR, context->sourcereg, 0, context->sourcereg, 0, SLJIT_IMM, othercasebit);
+    add_jump(compiler, backtracks, CMP(SLJIT_C_NOT_EQUAL, context->sourcereg, 0, SLJIT_IMM, *cc | othercasebit));
+    }
+  else
+    add_jump(compiler, backtracks, CMP(SLJIT_C_NOT_EQUAL, context->sourcereg, 0, SLJIT_IMM, *cc));
+
+#endif
+
+  cc++;
+#ifdef SUPPORT_UTF
+  utflength--;
+  }
+while (utflength > 0);
+#endif
+
+return cc;
+}
+
+#if defined SUPPORT_UTF || !defined COMPILE_PCRE8
+
+#define SET_TYPE_OFFSET(value) \
+  if ((value) != typeoffset) \
+    { \
+    if ((value) > typeoffset) \
+      OP2(SLJIT_SUB, typereg, 0, typereg, 0, SLJIT_IMM, (value) - typeoffset); \
+    else \
+      OP2(SLJIT_ADD, typereg, 0, typereg, 0, SLJIT_IMM, typeoffset - (value)); \
+    } \
+  typeoffset = (value);
+
+#define SET_CHAR_OFFSET(value) \
+  if ((value) != charoffset) \
+    { \
+    if ((value) > charoffset) \
+      OP2(SLJIT_SUB, TMP1, 0, TMP1, 0, SLJIT_IMM, (value) - charoffset); \
+    else \
+      OP2(SLJIT_ADD, TMP1, 0, TMP1, 0, SLJIT_IMM, charoffset - (value)); \
+    } \
+  charoffset = (value);
+
+static void compile_xclass_matchingpath(compiler_common *common, pcre_uchar *cc, jump_list **backtracks)
+{
+DEFINE_COMPILER;
+jump_list *found = NULL;
+jump_list **list = (*cc & XCL_NOT) == 0 ? &found : backtracks;
+pcre_int32 c, charoffset;
+const pcre_uint32 *other_cases;
+struct sljit_jump *jump = NULL;
+pcre_uchar *ccbegin;
+int compares, invertcmp, numberofcmps;
+#ifdef SUPPORT_UCP
+BOOL needstype = FALSE, needsscript = FALSE, needschar = FALSE;
+BOOL charsaved = FALSE;
+int typereg = TMP1, scriptreg = TMP1;
+pcre_int32 typeoffset;
+#endif
+
+/* Although SUPPORT_UTF must be defined, we are
+   not necessary in utf mode even in 8 bit mode. */
+detect_partial_match(common, backtracks);
+read_char(common);
+
+if ((*cc++ & XCL_MAP) != 0)
+  {
+  OP1(SLJIT_MOV, TMP3, 0, TMP1, 0);
+#ifndef COMPILE_PCRE8
+  jump = CMP(SLJIT_C_GREATER, TMP1, 0, SLJIT_IMM, 255);
+#elif defined SUPPORT_UTF
+  if (common->utf)
+    jump = CMP(SLJIT_C_GREATER, TMP1, 0, SLJIT_IMM, 255);
+#endif
+
+  if (!check_class_ranges(common, (const pcre_uint8 *)cc, TRUE, list))
+    {
+    OP2(SLJIT_AND, TMP2, 0, TMP1, 0, SLJIT_IMM, 0x7);
+    OP2(SLJIT_LSHR, TMP1, 0, TMP1, 0, SLJIT_IMM, 3);
+    OP1(SLJIT_MOV_UB, TMP1, 0, SLJIT_MEM1(TMP1), (sljit_sw)cc);
+    OP2(SLJIT_SHL, TMP2, 0, SLJIT_IMM, 1, TMP2, 0);
+    OP2(SLJIT_AND | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP1, 0, TMP2, 0);
+    add_jump(compiler, list, JUMP(SLJIT_C_NOT_ZERO));
+    }
+
+#ifndef COMPILE_PCRE8
+  JUMPHERE(jump);
+#elif defined SUPPORT_UTF
+  if (common->utf)
+    JUMPHERE(jump);
+#endif
+  OP1(SLJIT_MOV, TMP1, 0, TMP3, 0);
+#ifdef SUPPORT_UCP
+  charsaved = TRUE;
+#endif
+  cc += 32 / sizeof(pcre_uchar);
+  }
+
+/* Scanning the necessary info. */
+ccbegin = cc;
+compares = 0;
+while (*cc != XCL_END)
+  {
+  compares++;
+  if (*cc == XCL_SINGLE)
+    {
+    cc += 2;
+#ifdef SUPPORT_UTF
+    if (common->utf && HAS_EXTRALEN(cc[-1])) cc += GET_EXTRALEN(cc[-1]);
+#endif
+#ifdef SUPPORT_UCP
+    needschar = TRUE;
+#endif
+    }
+  else if (*cc == XCL_RANGE)
+    {
+    cc += 2;
+#ifdef SUPPORT_UTF
+    if (common->utf && HAS_EXTRALEN(cc[-1])) cc += GET_EXTRALEN(cc[-1]);
+#endif
+    cc++;
+#ifdef SUPPORT_UTF
+    if (common->utf && HAS_EXTRALEN(cc[-1])) cc += GET_EXTRALEN(cc[-1]);
+#endif
+#ifdef SUPPORT_UCP
+    needschar = TRUE;
+#endif
+    }
+#ifdef SUPPORT_UCP
+  else
+    {
+    SLJIT_ASSERT(*cc == XCL_PROP || *cc == XCL_NOTPROP);
+    cc++;
+    switch(*cc)
+      {
+      case PT_ANY:
+      break;
+
+      case PT_LAMP:
+      case PT_GC:
+      case PT_PC:
+      case PT_ALNUM:
+      needstype = TRUE;
+      break;
+
+      case PT_SC:
+      needsscript = TRUE;
+      break;
+
+      case PT_SPACE:
+      case PT_PXSPACE:
+      case PT_WORD:
+      needstype = TRUE;
+      needschar = TRUE;
+      break;
+
+      case PT_CLIST:
+      case PT_UCNC:
+      needschar = TRUE;
+      break;
+
+      default:
+      SLJIT_ASSERT_STOP();
+      break;
+      }
+    cc += 2;
+    }
+#endif
+  }
+
+#ifdef SUPPORT_UCP
+/* Simple register allocation. TMP1 is preferred if possible. */
+if (needstype || needsscript)
+  {
+  if (needschar && !charsaved)
+    OP1(SLJIT_MOV, TMP3, 0, TMP1, 0);
+  add_jump(compiler, &common->getucd, JUMP(SLJIT_FAST_CALL));
+  if (needschar)
+    {
+    if (needstype)
+      {
+      OP1(SLJIT_MOV, RETURN_ADDR, 0, TMP1, 0);
+      typereg = RETURN_ADDR;
+      }
+
+    if (needsscript)
+      scriptreg = TMP3;
+    OP1(SLJIT_MOV, TMP1, 0, TMP3, 0);
+    }
+  else if (needstype && needsscript)
+    scriptreg = TMP3;
+  /* In all other cases only one of them was specified, and that can goes to TMP1. */
+
+  if (needsscript)
+    {
+    if (scriptreg == TMP1)
+      {
+      OP1(SLJIT_MOV, scriptreg, 0, SLJIT_IMM, (sljit_sw)PRIV(ucd_records) + SLJIT_OFFSETOF(ucd_record, script));
+      OP1(SLJIT_MOV_UB, scriptreg, 0, SLJIT_MEM2(scriptreg, TMP2), 3);
+      }
+    else
+      {
+      OP2(SLJIT_SHL, TMP2, 0, TMP2, 0, SLJIT_IMM, 3);
+      OP2(SLJIT_ADD, TMP2, 0, TMP2, 0, SLJIT_IMM, (sljit_sw)PRIV(ucd_records) + SLJIT_OFFSETOF(ucd_record, script));
+      OP1(SLJIT_MOV_UB, scriptreg, 0, SLJIT_MEM1(TMP2), 0);
+      }
+    }
+  }
+#endif
+
+/* Generating code. */
+cc = ccbegin;
+charoffset = 0;
+numberofcmps = 0;
+#ifdef SUPPORT_UCP
+typeoffset = 0;
+#endif
+
+while (*cc != XCL_END)
+  {
+  compares--;
+  invertcmp = (compares == 0 && list != backtracks);
+  jump = NULL;
+
+  if (*cc == XCL_SINGLE)
+    {
+    cc ++;
+#ifdef SUPPORT_UTF
+    if (common->utf)
+      {
+      GETCHARINC(c, cc);
+      }
+    else
+#endif
+      c = *cc++;
+
+    if (numberofcmps < 3 && (*cc == XCL_SINGLE || *cc == XCL_RANGE))
+      {
+      OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, c - charoffset);
+      OP_FLAGS(numberofcmps == 0 ? SLJIT_MOV : SLJIT_OR, TMP2, 0, numberofcmps == 0 ? SLJIT_UNUSED : TMP2, 0, SLJIT_C_EQUAL);
+      numberofcmps++;
+      }
+    else if (numberofcmps > 0)
+      {
+      OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, c - charoffset);
+      OP_FLAGS(SLJIT_OR | SLJIT_SET_E, TMP2, 0, TMP2, 0, SLJIT_C_EQUAL);
+      jump = JUMP(SLJIT_C_NOT_ZERO ^ invertcmp);
+      numberofcmps = 0;
+      }
+    else
+      {
+      jump = CMP(SLJIT_C_EQUAL ^ invertcmp, TMP1, 0, SLJIT_IMM, c - charoffset);
+      numberofcmps = 0;
+      }
+    }
+  else if (*cc == XCL_RANGE)
+    {
+    cc ++;
+#ifdef SUPPORT_UTF
+    if (common->utf)
+      {
+      GETCHARINC(c, cc);
+      }
+    else
+#endif
+      c = *cc++;
+    SET_CHAR_OFFSET(c);
+#ifdef SUPPORT_UTF
+    if (common->utf)
+      {
+      GETCHARINC(c, cc);
+      }
+    else
+#endif
+      c = *cc++;
+    if (numberofcmps < 3 && (*cc == XCL_SINGLE || *cc == XCL_RANGE))
+      {
+      OP2(SLJIT_SUB | SLJIT_SET_U, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, c - charoffset);
+      OP_FLAGS(numberofcmps == 0 ? SLJIT_MOV : SLJIT_OR, TMP2, 0, numberofcmps == 0 ? SLJIT_UNUSED : TMP2, 0, SLJIT_C_LESS_EQUAL);
+      numberofcmps++;
+      }
+    else if (numberofcmps > 0)
+      {
+      OP2(SLJIT_SUB | SLJIT_SET_U, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, c - charoffset);
+      OP_FLAGS(SLJIT_OR | SLJIT_SET_E, TMP2, 0, TMP2, 0, SLJIT_C_LESS_EQUAL);
+      jump = JUMP(SLJIT_C_NOT_ZERO ^ invertcmp);
+      numberofcmps = 0;
+      }
+    else
+      {
+      jump = CMP(SLJIT_C_LESS_EQUAL ^ invertcmp, TMP1, 0, SLJIT_IMM, c - charoffset);
+      numberofcmps = 0;
+      }
+    }
+#ifdef SUPPORT_UCP
+  else
+    {
+    if (*cc == XCL_NOTPROP)
+      invertcmp ^= 0x1;
+    cc++;
+    switch(*cc)
+      {
+      case PT_ANY:
+      if (list != backtracks)
+        {
+        if ((cc[-1] == XCL_NOTPROP && compares > 0) || (cc[-1] == XCL_PROP && compares == 0))
+          continue;
+        }
+      else if (cc[-1] == XCL_NOTPROP)
+        continue;
+      jump = JUMP(SLJIT_JUMP);
+      break;
+
+      case PT_LAMP:
+      OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_UNUSED, 0, typereg, 0, SLJIT_IMM, ucp_Lu - typeoffset);
+      OP_FLAGS(SLJIT_MOV, TMP2, 0, SLJIT_UNUSED, 0, SLJIT_C_EQUAL);
+      OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_UNUSED, 0, typereg, 0, SLJIT_IMM, ucp_Ll - typeoffset);
+      OP_FLAGS(SLJIT_OR, TMP2, 0, TMP2, 0, SLJIT_C_EQUAL);
+      OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_UNUSED, 0, typereg, 0, SLJIT_IMM, ucp_Lt - typeoffset);
+      OP_FLAGS(SLJIT_OR | SLJIT_SET_E, TMP2, 0, TMP2, 0, SLJIT_C_EQUAL);
+      jump = JUMP(SLJIT_C_NOT_ZERO ^ invertcmp);
+      break;
+
+      case PT_GC:
+      c = PRIV(ucp_typerange)[(int)cc[1] * 2];
+      SET_TYPE_OFFSET(c);
+      jump = CMP(SLJIT_C_LESS_EQUAL ^ invertcmp, typereg, 0, SLJIT_IMM, PRIV(ucp_typerange)[(int)cc[1] * 2 + 1] - c);
+      break;
+
+      case PT_PC:
+      jump = CMP(SLJIT_C_EQUAL ^ invertcmp, typereg, 0, SLJIT_IMM, (int)cc[1] - typeoffset);
+      break;
+
+      case PT_SC:
+      jump = CMP(SLJIT_C_EQUAL ^ invertcmp, scriptreg, 0, SLJIT_IMM, (int)cc[1]);
+      break;
+
+      case PT_SPACE:
+      case PT_PXSPACE:
+      if (*cc == PT_SPACE)
+        {
+        OP1(SLJIT_MOV, TMP2, 0, SLJIT_IMM, 0);
+        jump = CMP(SLJIT_C_EQUAL, TMP1, 0, SLJIT_IMM, 11 - charoffset);
+        }
+      SET_CHAR_OFFSET(9);
+      OP2(SLJIT_SUB | SLJIT_SET_U, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, 13 - 9);
+      OP_FLAGS(SLJIT_MOV, TMP2, 0, SLJIT_UNUSED, 0, SLJIT_C_LESS_EQUAL);
+      if (*cc == PT_SPACE)
+        JUMPHERE(jump);
+
+      SET_TYPE_OFFSET(ucp_Zl);
+      OP2(SLJIT_SUB | SLJIT_SET_U, SLJIT_UNUSED, 0, typereg, 0, SLJIT_IMM, ucp_Zs - ucp_Zl);
+      OP_FLAGS(SLJIT_OR | SLJIT_SET_E, TMP2, 0, TMP2, 0, SLJIT_C_LESS_EQUAL);
+      jump = JUMP(SLJIT_C_NOT_ZERO ^ invertcmp);
+      break;
+
+      case PT_WORD:
+      OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, CHAR_UNDERSCORE - charoffset);
+      OP_FLAGS(SLJIT_MOV, TMP2, 0, SLJIT_UNUSED, 0, SLJIT_C_EQUAL);
+      /* Fall through. */
+
+      case PT_ALNUM:
+      SET_TYPE_OFFSET(ucp_Ll);
+      OP2(SLJIT_SUB | SLJIT_SET_U, SLJIT_UNUSED, 0, typereg, 0, SLJIT_IMM, ucp_Lu - ucp_Ll);
+      OP_FLAGS((*cc == PT_ALNUM) ? SLJIT_MOV : SLJIT_OR, TMP2, 0, (*cc == PT_ALNUM) ? SLJIT_UNUSED : TMP2, 0, SLJIT_C_LESS_EQUAL);
+      SET_TYPE_OFFSET(ucp_Nd);
+      OP2(SLJIT_SUB | SLJIT_SET_U, SLJIT_UNUSED, 0, typereg, 0, SLJIT_IMM, ucp_No - ucp_Nd);
+      OP_FLAGS(SLJIT_OR | SLJIT_SET_E, TMP2, 0, TMP2, 0, SLJIT_C_LESS_EQUAL);
+      jump = JUMP(SLJIT_C_NOT_ZERO ^ invertcmp);
+      break;
+
+      case PT_CLIST:
+      other_cases = PRIV(ucd_caseless_sets) + cc[1];
+
+      /* At least three characters are required.
+         Otherwise this case would be handled by the normal code path. */
+      SLJIT_ASSERT(other_cases[0] != NOTACHAR && other_cases[1] != NOTACHAR && other_cases[2] != NOTACHAR);
+      SLJIT_ASSERT(other_cases[0] < other_cases[1] && other_cases[1] < other_cases[2]);
+
+      /* Optimizing character pairs, if their difference is power of 2. */
+      if (is_powerof2(other_cases[1] ^ other_cases[0]))
+        {
+        if (charoffset == 0)
+          OP2(SLJIT_OR, TMP2, 0, TMP1, 0, SLJIT_IMM, other_cases[1] ^ other_cases[0]);
+        else
+          {
+          OP2(SLJIT_ADD, TMP2, 0, TMP1, 0, SLJIT_IMM, (sljit_sw)charoffset);
+          OP2(SLJIT_OR, TMP2, 0, TMP2, 0, SLJIT_IMM, other_cases[1] ^ other_cases[0]);
+          }
+        OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP2, 0, SLJIT_IMM, other_cases[1]);
+        OP_FLAGS(SLJIT_MOV, TMP2, 0, SLJIT_UNUSED, 0, SLJIT_C_EQUAL);
+        other_cases += 2;
+        }
+      else if (is_powerof2(other_cases[2] ^ other_cases[1]))
+        {
+        if (charoffset == 0)
+          OP2(SLJIT_OR, TMP2, 0, TMP1, 0, SLJIT_IMM, other_cases[2] ^ other_cases[1]);
+        else
+          {
+          OP2(SLJIT_ADD, TMP2, 0, TMP1, 0, SLJIT_IMM, (sljit_sw)charoffset);
+          OP2(SLJIT_OR, TMP2, 0, TMP2, 0, SLJIT_IMM, other_cases[1] ^ other_cases[0]);
+          }
+        OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP2, 0, SLJIT_IMM, other_cases[2]);
+        OP_FLAGS(SLJIT_MOV, TMP2, 0, SLJIT_UNUSED, 0, SLJIT_C_EQUAL);
+
+        OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, other_cases[0] - charoffset);
+        OP_FLAGS(SLJIT_OR | ((other_cases[3] == NOTACHAR) ? SLJIT_SET_E : 0), TMP2, 0, TMP2, 0, SLJIT_C_EQUAL);
+
+        other_cases += 3;
+        }
+      else
+        {
+        OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, *other_cases++ - charoffset);
+        OP_FLAGS(SLJIT_MOV, TMP2, 0, SLJIT_UNUSED, 0, SLJIT_C_EQUAL);
+        }
+
+      while (*other_cases != NOTACHAR)
+        {
+        OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, *other_cases++ - charoffset);
+        OP_FLAGS(SLJIT_OR | ((*other_cases == NOTACHAR) ? SLJIT_SET_E : 0), TMP2, 0, TMP2, 0, SLJIT_C_EQUAL);
+        }
+      jump = JUMP(SLJIT_C_NOT_ZERO ^ invertcmp);
+      break;
+
+      case PT_UCNC:
+      OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, CHAR_DOLLAR_SIGN - charoffset);
+      OP_FLAGS(SLJIT_MOV, TMP2, 0, SLJIT_UNUSED, 0, SLJIT_C_EQUAL);
+      OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, CHAR_COMMERCIAL_AT - charoffset);
+      OP_FLAGS(SLJIT_OR, TMP2, 0, TMP2, 0, SLJIT_C_EQUAL);
+      OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, CHAR_GRAVE_ACCENT - charoffset);
+      OP_FLAGS(SLJIT_OR, TMP2, 0, TMP2, 0, SLJIT_C_EQUAL);
+
+      SET_CHAR_OFFSET(0xa0);
+      OP2(SLJIT_SUB | SLJIT_SET_U, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, 0xd7ff - charoffset);
+      OP_FLAGS(SLJIT_OR, TMP2, 0, TMP2, 0, SLJIT_C_LESS_EQUAL);
+      SET_CHAR_OFFSET(0);
+      OP2(SLJIT_SUB | SLJIT_SET_U, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, 0xe000 - 0);
+      OP_FLAGS(SLJIT_OR | SLJIT_SET_E, TMP2, 0, TMP2, 0, SLJIT_C_GREATER_EQUAL);
+      jump = JUMP(SLJIT_C_NOT_ZERO ^ invertcmp);
+      break;
+      }
+    cc += 2;
+    }
+#endif
+
+  if (jump != NULL)
+    add_jump(compiler, compares > 0 ? list : backtracks, jump);
+  }
+
+if (found != NULL)
+  set_jumps(found, LABEL());
+}
+
+#undef SET_TYPE_OFFSET
+#undef SET_CHAR_OFFSET
+
+#endif
+
+static pcre_uchar *compile_char1_matchingpath(compiler_common *common, pcre_uchar type, pcre_uchar *cc, jump_list **backtracks)
+{
+DEFINE_COMPILER;
+int length;
+unsigned int c, oc, bit;
+compare_context context;
+struct sljit_jump *jump[4];
+jump_list *end_list;
+#ifdef SUPPORT_UTF
+struct sljit_label *label;
+#ifdef SUPPORT_UCP
+pcre_uchar propdata[5];
+#endif
+#endif
+
+switch(type)
+  {
+  case OP_SOD:
+  OP1(SLJIT_MOV, TMP1, 0, ARGUMENTS, 0);
+  OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(TMP1), SLJIT_OFFSETOF(jit_arguments, begin));
+  add_jump(compiler, backtracks, CMP(SLJIT_C_NOT_EQUAL, STR_PTR, 0, TMP1, 0));
+  return cc;
+
+  case OP_SOM:
+  OP1(SLJIT_MOV, TMP1, 0, ARGUMENTS, 0);
+  OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(TMP1), SLJIT_OFFSETOF(jit_arguments, str));
+  add_jump(compiler, backtracks, CMP(SLJIT_C_NOT_EQUAL, STR_PTR, 0, TMP1, 0));
+  return cc;
+
+  case OP_NOT_WORD_BOUNDARY:
+  case OP_WORD_BOUNDARY:
+  add_jump(compiler, &common->wordboundary, JUMP(SLJIT_FAST_CALL));
+  add_jump(compiler, backtracks, JUMP(type == OP_NOT_WORD_BOUNDARY ? SLJIT_C_NOT_ZERO : SLJIT_C_ZERO));
+  return cc;
+
+  case OP_NOT_DIGIT:
+  case OP_DIGIT:
+  /* Digits are usually 0-9, so it is worth to optimize them. */
+  if (common->digits[0] == -2)
+    get_ctype_ranges(common, ctype_digit, common->digits);
+  detect_partial_match(common, backtracks);
+  /* Flip the starting bit in the negative case. */
+  if (type == OP_NOT_DIGIT)
+    common->digits[1] ^= 1;
+  if (!check_ranges(common, common->digits, backtracks, TRUE))
+    {
+    read_char8_type(common);
+    OP2(SLJIT_AND | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, ctype_digit);
+    add_jump(compiler, backtracks, JUMP(type == OP_DIGIT ? SLJIT_C_ZERO : SLJIT_C_NOT_ZERO));
+    }
+  if (type == OP_NOT_DIGIT)
+    common->digits[1] ^= 1;
+  return cc;
+
+  case OP_NOT_WHITESPACE:
+  case OP_WHITESPACE:
+  detect_partial_match(common, backtracks);
+  read_char8_type(common);
+  OP2(SLJIT_AND | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, ctype_space);
+  add_jump(compiler, backtracks, JUMP(type == OP_WHITESPACE ? SLJIT_C_ZERO : SLJIT_C_NOT_ZERO));
+  return cc;
+
+  case OP_NOT_WORDCHAR:
+  case OP_WORDCHAR:
+  detect_partial_match(common, backtracks);
+  read_char8_type(common);
+  OP2(SLJIT_AND | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, ctype_word);
+  add_jump(compiler, backtracks, JUMP(type == OP_WORDCHAR ? SLJIT_C_ZERO : SLJIT_C_NOT_ZERO));
+  return cc;
+
+  case OP_ANY:
+  detect_partial_match(common, backtracks);
+  read_char(common);
+  if (common->nltype == NLTYPE_FIXED && common->newline > 255)
+    {
+    jump[0] = CMP(SLJIT_C_NOT_EQUAL, TMP1, 0, SLJIT_IMM, (common->newline >> 8) & 0xff);
+    end_list = NULL;
+    if (common->mode != JIT_PARTIAL_HARD_COMPILE)
+      add_jump(compiler, &end_list, CMP(SLJIT_C_GREATER_EQUAL, STR_PTR, 0, STR_END, 0));
+    else
+      check_str_end(common, &end_list);
+
+    OP1(MOV_UCHAR, TMP1, 0, SLJIT_MEM1(STR_PTR), 0);
+    add_jump(compiler, backtracks, CMP(SLJIT_C_EQUAL, TMP1, 0, SLJIT_IMM, common->newline & 0xff));
+    set_jumps(end_list, LABEL());
+    JUMPHERE(jump[0]);
+    }
+  else
+    check_newlinechar(common, common->nltype, backtracks, TRUE);
+  return cc;
+
+  case OP_ALLANY:
+  detect_partial_match(common, backtracks);
+#ifdef SUPPORT_UTF
+  if (common->utf)
+    {
+    OP1(MOV_UCHAR, TMP1, 0, SLJIT_MEM1(STR_PTR), 0);
+    OP2(SLJIT_ADD, STR_PTR, 0, STR_PTR, 0, SLJIT_IMM, IN_UCHARS(1));
+#if defined COMPILE_PCRE8 || defined COMPILE_PCRE16
+#if defined COMPILE_PCRE8
+    jump[0] = CMP(SLJIT_C_LESS, TMP1, 0, SLJIT_IMM, 0xc0);
+    OP1(SLJIT_MOV_UB, TMP1, 0, SLJIT_MEM1(TMP1), (sljit_sw)PRIV(utf8_table4) - 0xc0);
+    OP2(SLJIT_ADD, STR_PTR, 0, STR_PTR, 0, TMP1, 0);
+#elif defined COMPILE_PCRE16
+    jump[0] = CMP(SLJIT_C_LESS, TMP1, 0, SLJIT_IMM, 0xd800);
+    OP2(SLJIT_AND, TMP1, 0, TMP1, 0, SLJIT_IMM, 0xfc00);
+    OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, 0xd800);
+    OP_FLAGS(SLJIT_MOV, TMP1, 0, SLJIT_UNUSED, 0, SLJIT_C_EQUAL);
+    OP2(SLJIT_SHL, TMP1, 0, TMP1, 0, SLJIT_IMM, 1);
+    OP2(SLJIT_ADD, STR_PTR, 0, STR_PTR, 0, TMP1, 0);
+#endif
+    JUMPHERE(jump[0]);
+#endif /* COMPILE_PCRE[8|16] */
+    return cc;
+    }
+#endif
+  OP2(SLJIT_ADD, STR_PTR, 0, STR_PTR, 0, SLJIT_IMM, IN_UCHARS(1));
+  return cc;
+
+  case OP_ANYBYTE:
+  detect_partial_match(common, backtracks);
+  OP2(SLJIT_ADD, STR_PTR, 0, STR_PTR, 0, SLJIT_IMM, IN_UCHARS(1));
+  return cc;
+
+#ifdef SUPPORT_UTF
+#ifdef SUPPORT_UCP
+  case OP_NOTPROP:
+  case OP_PROP:
+  propdata[0] = 0;
+  propdata[1] = type == OP_NOTPROP ? XCL_NOTPROP : XCL_PROP;
+  propdata[2] = cc[0];
+  propdata[3] = cc[1];
+  propdata[4] = XCL_END;
+  compile_xclass_matchingpath(common, propdata, backtracks);
+  return cc + 2;
+#endif
+#endif
+
+  case OP_ANYNL:
+  detect_partial_match(common, backtracks);
+  read_char(common);
+  jump[0] = CMP(SLJIT_C_NOT_EQUAL, TMP1, 0, SLJIT_IMM, CHAR_CR);
+  /* We don't need to handle soft partial matching case. */
+  end_list = NULL;
+  if (common->mode != JIT_PARTIAL_HARD_COMPILE)
+    add_jump(compiler, &end_list, CMP(SLJIT_C_GREATER_EQUAL, STR_PTR, 0, STR_END, 0));
+  else
+    check_str_end(common, &end_list);
+  OP1(MOV_UCHAR, TMP1, 0, SLJIT_MEM1(STR_PTR), 0);
+  jump[1] = CMP(SLJIT_C_NOT_EQUAL, TMP1, 0, SLJIT_IMM, CHAR_NL);
+  OP2(SLJIT_ADD, STR_PTR, 0, STR_PTR, 0, SLJIT_IMM, IN_UCHARS(1));
+  jump[2] = JUMP(SLJIT_JUMP);
+  JUMPHERE(jump[0]);
+  check_newlinechar(common, common->bsr_nltype, backtracks, FALSE);
+  set_jumps(end_list, LABEL());
+  JUMPHERE(jump[1]);
+  JUMPHERE(jump[2]);
+  return cc;
+
+  case OP_NOT_HSPACE:
+  case OP_HSPACE:
+  detect_partial_match(common, backtracks);
+  read_char(common);
+  add_jump(compiler, &common->hspace, JUMP(SLJIT_FAST_CALL));
+  add_jump(compiler, backtracks, JUMP(type == OP_NOT_HSPACE ? SLJIT_C_NOT_ZERO : SLJIT_C_ZERO));
+  return cc;
+
+  case OP_NOT_VSPACE:
+  case OP_VSPACE:
+  detect_partial_match(common, backtracks);
+  read_char(common);
+  add_jump(compiler, &common->vspace, JUMP(SLJIT_FAST_CALL));
+  add_jump(compiler, backtracks, JUMP(type == OP_NOT_VSPACE ? SLJIT_C_NOT_ZERO : SLJIT_C_ZERO));
+  return cc;
+
+#ifdef SUPPORT_UCP
+  case OP_EXTUNI:
+  detect_partial_match(common, backtracks);
+  read_char(common);
+  add_jump(compiler, &common->getucd, JUMP(SLJIT_FAST_CALL));
+  OP1(SLJIT_MOV, TMP1, 0, SLJIT_IMM, (sljit_sw)PRIV(ucd_records) + SLJIT_OFFSETOF(ucd_record, gbprop));
+  /* Optimize register allocation: use a real register. */
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), LOCALS0, STACK_TOP, 0);
+  OP1(SLJIT_MOV_UB, STACK_TOP, 0, SLJIT_MEM2(TMP1, TMP2), 3);
+
+  label = LABEL();
+  jump[0] = CMP(SLJIT_C_GREATER_EQUAL, STR_PTR, 0, STR_END, 0);
+  OP1(SLJIT_MOV, TMP3, 0, STR_PTR, 0);
+  read_char(common);
+  add_jump(compiler, &common->getucd, JUMP(SLJIT_FAST_CALL));
+  OP1(SLJIT_MOV, TMP1, 0, SLJIT_IMM, (sljit_sw)PRIV(ucd_records) + SLJIT_OFFSETOF(ucd_record, gbprop));
+  OP1(SLJIT_MOV_UB, TMP2, 0, SLJIT_MEM2(TMP1, TMP2), 3);
+
+  OP2(SLJIT_SHL, STACK_TOP, 0, STACK_TOP, 0, SLJIT_IMM, 2);
+  OP1(SLJIT_MOV_UI, TMP1, 0, SLJIT_MEM1(STACK_TOP), (sljit_sw)PRIV(ucp_gbtable));
+  OP1(SLJIT_MOV, STACK_TOP, 0, TMP2, 0);
+  OP2(SLJIT_SHL, TMP2, 0, SLJIT_IMM, 1, TMP2, 0);
+  OP2(SLJIT_AND | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP1, 0, TMP2, 0);
+  JUMPTO(SLJIT_C_NOT_ZERO, label);
+
+  OP1(SLJIT_MOV, STR_PTR, 0, TMP3, 0);
+  JUMPHERE(jump[0]);
+  OP1(SLJIT_MOV, STACK_TOP, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), LOCALS0);
+
+  if (common->mode == JIT_PARTIAL_HARD_COMPILE)
+    {
+    jump[0] = CMP(SLJIT_C_LESS, STR_PTR, 0, STR_END, 0);
+    /* Since we successfully read a char above, partial matching must occure. */
+    check_partial(common, TRUE);
+    JUMPHERE(jump[0]);
+    }
+  return cc;
+#endif
+
+  case OP_EODN:
+  /* Requires rather complex checks. */
+  jump[0] = CMP(SLJIT_C_GREATER_EQUAL, STR_PTR, 0, STR_END, 0);
+  if (common->nltype == NLTYPE_FIXED && common->newline > 255)
+    {
+    OP2(SLJIT_ADD, TMP2, 0, STR_PTR, 0, SLJIT_IMM, IN_UCHARS(2));
+    OP1(MOV_UCHAR, TMP1, 0, SLJIT_MEM1(STR_PTR), IN_UCHARS(0));
+    if (common->mode == JIT_COMPILE)
+      add_jump(compiler, backtracks, CMP(SLJIT_C_NOT_EQUAL, TMP2, 0, STR_END, 0));
+    else
+      {
+      jump[1] = CMP(SLJIT_C_EQUAL, TMP2, 0, STR_END, 0);
+      OP2(SLJIT_SUB | SLJIT_SET_U, SLJIT_UNUSED, 0, TMP2, 0, STR_END, 0);
+      OP_FLAGS(SLJIT_MOV, TMP2, 0, SLJIT_UNUSED, 0, SLJIT_C_LESS);
+      OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, (common->newline >> 8) & 0xff);
+      OP_FLAGS(SLJIT_OR | SLJIT_SET_E, TMP2, 0, TMP2, 0, SLJIT_C_NOT_EQUAL);
+      add_jump(compiler, backtracks, JUMP(SLJIT_C_NOT_EQUAL));
+      check_partial(common, TRUE);
+      add_jump(compiler, backtracks, JUMP(SLJIT_JUMP));
+      JUMPHERE(jump[1]);
+      }
+    OP1(MOV_UCHAR, TMP2, 0, SLJIT_MEM1(STR_PTR), IN_UCHARS(1));
+    add_jump(compiler, backtracks, CMP(SLJIT_C_NOT_EQUAL, TMP1, 0, SLJIT_IMM, (common->newline >> 8) & 0xff));
+    add_jump(compiler, backtracks, CMP(SLJIT_C_NOT_EQUAL, TMP2, 0, SLJIT_IMM, common->newline & 0xff));
+    }
+  else if (common->nltype == NLTYPE_FIXED)
+    {
+    OP2(SLJIT_ADD, TMP2, 0, STR_PTR, 0, SLJIT_IMM, IN_UCHARS(1));
+    OP1(MOV_UCHAR, TMP1, 0, SLJIT_MEM1(STR_PTR), IN_UCHARS(0));
+    add_jump(compiler, backtracks, CMP(SLJIT_C_NOT_EQUAL, TMP2, 0, STR_END, 0));
+    add_jump(compiler, backtracks, CMP(SLJIT_C_NOT_EQUAL, TMP1, 0, SLJIT_IMM, common->newline));
+    }
+  else
+    {
+    OP1(MOV_UCHAR, TMP1, 0, SLJIT_MEM1(STR_PTR), IN_UCHARS(0));
+    jump[1] = CMP(SLJIT_C_NOT_EQUAL, TMP1, 0, SLJIT_IMM, CHAR_CR);
+    OP2(SLJIT_ADD, TMP2, 0, STR_PTR, 0, SLJIT_IMM, IN_UCHARS(2));
+    OP2(SLJIT_SUB | SLJIT_SET_U, SLJIT_UNUSED, 0, TMP2, 0, STR_END, 0);
+    jump[2] = JUMP(SLJIT_C_GREATER);
+    add_jump(compiler, backtracks, JUMP(SLJIT_C_LESS));
+    /* Equal. */
+    OP1(MOV_UCHAR, TMP1, 0, SLJIT_MEM1(STR_PTR), IN_UCHARS(1));
+    jump[3] = CMP(SLJIT_C_EQUAL, TMP1, 0, SLJIT_IMM, CHAR_NL);
+    add_jump(compiler, backtracks, JUMP(SLJIT_JUMP));
+
+    JUMPHERE(jump[1]);
+    if (common->nltype == NLTYPE_ANYCRLF)
+      {
+      OP2(SLJIT_ADD, TMP2, 0, STR_PTR, 0, SLJIT_IMM, IN_UCHARS(1));
+      add_jump(compiler, backtracks, CMP(SLJIT_C_LESS, TMP2, 0, STR_END, 0));
+      add_jump(compiler, backtracks, CMP(SLJIT_C_NOT_EQUAL, TMP1, 0, SLJIT_IMM, CHAR_NL));
+      }
+    else
+      {
+      OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), LOCALS1, STR_PTR, 0);
+      read_char(common);
+      add_jump(compiler, backtracks, CMP(SLJIT_C_NOT_EQUAL, STR_PTR, 0, STR_END, 0));
+      add_jump(compiler, &common->anynewline, JUMP(SLJIT_FAST_CALL));
+      add_jump(compiler, backtracks, JUMP(SLJIT_C_ZERO));
+      OP1(SLJIT_MOV, STR_PTR, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), LOCALS1);
+      }
+    JUMPHERE(jump[2]);
+    JUMPHERE(jump[3]);
+    }
+  JUMPHERE(jump[0]);
+  check_partial(common, FALSE);
+  return cc;
+
+  case OP_EOD:
+  add_jump(compiler, backtracks, CMP(SLJIT_C_LESS, STR_PTR, 0, STR_END, 0));
+  check_partial(common, FALSE);
+  return cc;
+
+  case OP_CIRC:
+  OP1(SLJIT_MOV, TMP2, 0, ARGUMENTS, 0);
+  OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(TMP2), SLJIT_OFFSETOF(jit_arguments, begin));
+  add_jump(compiler, backtracks, CMP(SLJIT_C_GREATER, STR_PTR, 0, TMP1, 0));
+  OP1(SLJIT_MOV_UB, TMP2, 0, SLJIT_MEM1(TMP2), SLJIT_OFFSETOF(jit_arguments, notbol));
+  add_jump(compiler, backtracks, CMP(SLJIT_C_NOT_EQUAL, TMP2, 0, SLJIT_IMM, 0));
+  return cc;
+
+  case OP_CIRCM:
+  OP1(SLJIT_MOV, TMP2, 0, ARGUMENTS, 0);
+  OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(TMP2), SLJIT_OFFSETOF(jit_arguments, begin));
+  jump[1] = CMP(SLJIT_C_GREATER, STR_PTR, 0, TMP1, 0);
+  OP1(SLJIT_MOV_UB, TMP2, 0, SLJIT_MEM1(TMP2), SLJIT_OFFSETOF(jit_arguments, notbol));
+  add_jump(compiler, backtracks, CMP(SLJIT_C_NOT_EQUAL, TMP2, 0, SLJIT_IMM, 0));
+  jump[0] = JUMP(SLJIT_JUMP);
+  JUMPHERE(jump[1]);
+
+  add_jump(compiler, backtracks, CMP(SLJIT_C_GREATER_EQUAL, STR_PTR, 0, STR_END, 0));
+  if (common->nltype == NLTYPE_FIXED && common->newline > 255)
+    {
+    OP2(SLJIT_SUB, TMP2, 0, STR_PTR, 0, SLJIT_IMM, IN_UCHARS(2));
+    add_jump(compiler, backtracks, CMP(SLJIT_C_LESS, TMP2, 0, TMP1, 0));
+    OP1(MOV_UCHAR, TMP1, 0, SLJIT_MEM1(STR_PTR), IN_UCHARS(-2));
+    OP1(MOV_UCHAR, TMP2, 0, SLJIT_MEM1(STR_PTR), IN_UCHARS(-1));
+    add_jump(compiler, backtracks, CMP(SLJIT_C_NOT_EQUAL, TMP1, 0, SLJIT_IMM, (common->newline >> 8) & 0xff));
+    add_jump(compiler, backtracks, CMP(SLJIT_C_NOT_EQUAL, TMP2, 0, SLJIT_IMM, common->newline & 0xff));
+    }
+  else
+    {
+    skip_char_back(common);
+    read_char(common);
+    check_newlinechar(common, common->nltype, backtracks, FALSE);
+    }
+  JUMPHERE(jump[0]);
+  return cc;
+
+  case OP_DOLL:
+  OP1(SLJIT_MOV, TMP2, 0, ARGUMENTS, 0);
+  OP1(SLJIT_MOV_UB, TMP2, 0, SLJIT_MEM1(TMP2), SLJIT_OFFSETOF(jit_arguments, noteol));
+  add_jump(compiler, backtracks, CMP(SLJIT_C_NOT_EQUAL, TMP2, 0, SLJIT_IMM, 0));
+
+  if (!common->endonly)
+    compile_char1_matchingpath(common, OP_EODN, cc, backtracks);
+  else
+    {
+    add_jump(compiler, backtracks, CMP(SLJIT_C_LESS, STR_PTR, 0, STR_END, 0));
+    check_partial(common, FALSE);
+    }
+  return cc;
+
+  case OP_DOLLM:
+  jump[1] = CMP(SLJIT_C_LESS, STR_PTR, 0, STR_END, 0);
+  OP1(SLJIT_MOV, TMP2, 0, ARGUMENTS, 0);
+  OP1(SLJIT_MOV_UB, TMP2, 0, SLJIT_MEM1(TMP2), SLJIT_OFFSETOF(jit_arguments, noteol));
+  add_jump(compiler, backtracks, CMP(SLJIT_C_NOT_EQUAL, TMP2, 0, SLJIT_IMM, 0));
+  check_partial(common, FALSE);
+  jump[0] = JUMP(SLJIT_JUMP);
+  JUMPHERE(jump[1]);
+
+  if (common->nltype == NLTYPE_FIXED && common->newline > 255)
+    {
+    OP2(SLJIT_ADD, TMP2, 0, STR_PTR, 0, SLJIT_IMM, IN_UCHARS(2));
+    OP1(MOV_UCHAR, TMP1, 0, SLJIT_MEM1(STR_PTR), IN_UCHARS(0));
+    if (common->mode == JIT_COMPILE)
+      add_jump(compiler, backtracks, CMP(SLJIT_C_GREATER, TMP2, 0, STR_END, 0));
+    else
+      {
+      jump[1] = CMP(SLJIT_C_LESS_EQUAL, TMP2, 0, STR_END, 0);
+      /* STR_PTR = STR_END - IN_UCHARS(1) */
+      add_jump(compiler, backtracks, CMP(SLJIT_C_NOT_EQUAL, TMP1, 0, SLJIT_IMM, (common->newline >> 8) & 0xff));
+      check_partial(common, TRUE);
+      add_jump(compiler, backtracks, JUMP(SLJIT_JUMP));
+      JUMPHERE(jump[1]);
+      }
+
+    OP1(MOV_UCHAR, TMP2, 0, SLJIT_MEM1(STR_PTR), IN_UCHARS(1));
+    add_jump(compiler, backtracks, CMP(SLJIT_C_NOT_EQUAL, TMP1, 0, SLJIT_IMM, (common->newline >> 8) & 0xff));
+    add_jump(compiler, backtracks, CMP(SLJIT_C_NOT_EQUAL, TMP2, 0, SLJIT_IMM, common->newline & 0xff));
+    }
+  else
+    {
+    peek_char(common);
+    check_newlinechar(common, common->nltype, backtracks, FALSE);
+    }
+  JUMPHERE(jump[0]);
+  return cc;
+
+  case OP_CHAR:
+  case OP_CHARI:
+  length = 1;
+#ifdef SUPPORT_UTF
+  if (common->utf && HAS_EXTRALEN(*cc)) length += GET_EXTRALEN(*cc);
+#endif
+  if (common->mode == JIT_COMPILE && (type == OP_CHAR || !char_has_othercase(common, cc) || char_get_othercase_bit(common, cc) != 0))
+    {
+    OP2(SLJIT_ADD, STR_PTR, 0, STR_PTR, 0, SLJIT_IMM, IN_UCHARS(length));
+    add_jump(compiler, backtracks, CMP(SLJIT_C_GREATER, STR_PTR, 0, STR_END, 0));
+
+    context.length = IN_UCHARS(length);
+    context.sourcereg = -1;
+#if defined SLJIT_UNALIGNED && SLJIT_UNALIGNED
+    context.ucharptr = 0;
+#endif
+    return byte_sequence_compare(common, type == OP_CHARI, cc, &context, backtracks);
+    }
+  detect_partial_match(common, backtracks);
+  read_char(common);
+#ifdef SUPPORT_UTF
+  if (common->utf)
+    {
+    GETCHAR(c, cc);
+    }
+  else
+#endif
+    c = *cc;
+  if (type == OP_CHAR || !char_has_othercase(common, cc))
+    {
+    add_jump(compiler, backtracks, CMP(SLJIT_C_NOT_EQUAL, TMP1, 0, SLJIT_IMM, c));
+    return cc + length;
+    }
+  oc = char_othercase(common, c);
+  bit = c ^ oc;
+  if (is_powerof2(bit))
+    {
+    OP2(SLJIT_OR, TMP1, 0, TMP1, 0, SLJIT_IMM, bit);
+    add_jump(compiler, backtracks, CMP(SLJIT_C_NOT_EQUAL, TMP1, 0, SLJIT_IMM, c | bit));
+    return cc + length;
+    }
+  OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, c);
+  OP_FLAGS(SLJIT_MOV, TMP2, 0, SLJIT_UNUSED, 0, SLJIT_C_EQUAL);
+  OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_IMM, oc);
+  OP_FLAGS(SLJIT_OR | SLJIT_SET_E, TMP2, 0, TMP2, 0, SLJIT_C_EQUAL);
+  add_jump(compiler, backtracks, JUMP(SLJIT_C_ZERO));
+  return cc + length;
+
+  case OP_NOT:
+  case OP_NOTI:
+  detect_partial_match(common, backtracks);
+  length = 1;
+#ifdef SUPPORT_UTF
+  if (common->utf)
+    {
+#ifdef COMPILE_PCRE8
+    c = *cc;
+    if (c < 128)
+      {
+      OP1(SLJIT_MOV_UB, TMP1, 0, SLJIT_MEM1(STR_PTR), 0);
+      if (type == OP_NOT || !char_has_othercase(common, cc))
+        add_jump(compiler, backtracks, CMP(SLJIT_C_EQUAL, TMP1, 0, SLJIT_IMM, c));
+      else
+        {
+        /* Since UTF8 code page is fixed, we know that c is in [a-z] or [A-Z] range. */
+        OP2(SLJIT_OR, TMP2, 0, TMP1, 0, SLJIT_IMM, 0x20);
+        add_jump(compiler, backtracks, CMP(SLJIT_C_EQUAL, TMP2, 0, SLJIT_IMM, c | 0x20));
+        }
+      /* Skip the variable-length character. */
+      OP2(SLJIT_ADD, STR_PTR, 0, STR_PTR, 0, SLJIT_IMM, IN_UCHARS(1));
+      jump[0] = CMP(SLJIT_C_LESS, TMP1, 0, SLJIT_IMM, 0xc0);
+      OP1(MOV_UCHAR, TMP1, 0, SLJIT_MEM1(TMP1), (sljit_sw)PRIV(utf8_table4) - 0xc0);
+      OP2(SLJIT_ADD, STR_PTR, 0, STR_PTR, 0, TMP1, 0);
+      JUMPHERE(jump[0]);
+      return cc + 1;
+      }
+    else
+#endif /* COMPILE_PCRE8 */
+      {
+      GETCHARLEN(c, cc, length);
+      read_char(common);
+      }
+    }
+  else
+#endif /* SUPPORT_UTF */
+    {
+    read_char(common);
+    c = *cc;
+    }
+
+  if (type == OP_NOT || !char_has_othercase(common, cc))
+    add_jump(compiler, backtracks, CMP(SLJIT_C_EQUAL, TMP1, 0, SLJIT_IMM, c));
+  else
+    {
+    oc = char_othercase(common, c);
+    bit = c ^ oc;
+    if (is_powerof2(bit))
+      {
+      OP2(SLJIT_OR, TMP1, 0, TMP1, 0, SLJIT_IMM, bit);
+      add_jump(compiler, backtracks, CMP(SLJIT_C_EQUAL, TMP1, 0, SLJIT_IMM, c | bit));
+      }
+    else
+      {
+      add_jump(compiler, backtracks, CMP(SLJIT_C_EQUAL, TMP1, 0, SLJIT_IMM, c));
+      add_jump(compiler, backtracks, CMP(SLJIT_C_EQUAL, TMP1, 0, SLJIT_IMM, oc));
+      }
+    }
+  return cc + length;
+
+  case OP_CLASS:
+  case OP_NCLASS:
+  detect_partial_match(common, backtracks);
+  read_char(common);
+  if (check_class_ranges(common, (const pcre_uint8 *)cc, type == OP_NCLASS, backtracks))
+    return cc + 32 / sizeof(pcre_uchar);
+
+#if defined SUPPORT_UTF || !defined COMPILE_PCRE8
+  jump[0] = NULL;
+#ifdef COMPILE_PCRE8
+  /* This check only affects 8 bit mode. In other modes, we
+  always need to compare the value with 255. */
+  if (common->utf)
+#endif /* COMPILE_PCRE8 */
+    {
+    jump[0] = CMP(SLJIT_C_GREATER, TMP1, 0, SLJIT_IMM, 255);
+    if (type == OP_CLASS)
+      {
+      add_jump(compiler, backtracks, jump[0]);
+      jump[0] = NULL;
+      }
+    }
+#endif /* SUPPORT_UTF || !COMPILE_PCRE8 */
+  OP2(SLJIT_AND, TMP2, 0, TMP1, 0, SLJIT_IMM, 0x7);
+  OP2(SLJIT_LSHR, TMP1, 0, TMP1, 0, SLJIT_IMM, 3);
+  OP1(SLJIT_MOV_UB, TMP1, 0, SLJIT_MEM1(TMP1), (sljit_sw)cc);
+  OP2(SLJIT_SHL, TMP2, 0, SLJIT_IMM, 1, TMP2, 0);
+  OP2(SLJIT_AND | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP1, 0, TMP2, 0);
+  add_jump(compiler, backtracks, JUMP(SLJIT_C_ZERO));
+#if defined SUPPORT_UTF || !defined COMPILE_PCRE8
+  if (jump[0] != NULL)
+    JUMPHERE(jump[0]);
+#endif /* SUPPORT_UTF || !COMPILE_PCRE8 */
+  return cc + 32 / sizeof(pcre_uchar);
+
+#if defined SUPPORT_UTF || defined COMPILE_PCRE16 || defined COMPILE_PCRE32
+  case OP_XCLASS:
+  compile_xclass_matchingpath(common, cc + LINK_SIZE, backtracks);
+  return cc + GET(cc, 0) - 1;
+#endif
+
+  case OP_REVERSE:
+  length = GET(cc, 0);
+  if (length == 0)
+    return cc + LINK_SIZE;
+  OP1(SLJIT_MOV, TMP1, 0, ARGUMENTS, 0);
+#ifdef SUPPORT_UTF
+  if (common->utf)
+    {
+    OP1(SLJIT_MOV, TMP3, 0, SLJIT_MEM1(TMP1), SLJIT_OFFSETOF(jit_arguments, begin));
+    OP1(SLJIT_MOV, TMP2, 0, SLJIT_IMM, length);
+    label = LABEL();
+    add_jump(compiler, backtracks, CMP(SLJIT_C_LESS_EQUAL, STR_PTR, 0, TMP3, 0));
+    skip_char_back(common);
+    OP2(SLJIT_SUB | SLJIT_SET_E, TMP2, 0, TMP2, 0, SLJIT_IMM, 1);
+    JUMPTO(SLJIT_C_NOT_ZERO, label);
+    }
+  else
+#endif
+    {
+    OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(TMP1), SLJIT_OFFSETOF(jit_arguments, begin));
+    OP2(SLJIT_SUB, STR_PTR, 0, STR_PTR, 0, SLJIT_IMM, IN_UCHARS(length));
+    add_jump(compiler, backtracks, CMP(SLJIT_C_LESS, STR_PTR, 0, TMP1, 0));
+    }
+  check_start_used_ptr(common);
+  return cc + LINK_SIZE;
+  }
+SLJIT_ASSERT_STOP();
+return cc;
+}
+
+static SLJIT_INLINE pcre_uchar *compile_charn_matchingpath(compiler_common *common, pcre_uchar *cc, pcre_uchar *ccend, jump_list **backtracks)
+{
+/* This function consumes at least one input character. */
+/* To decrease the number of length checks, we try to concatenate the fixed length character sequences. */
+DEFINE_COMPILER;
+pcre_uchar *ccbegin = cc;
+compare_context context;
+int size;
+
+context.length = 0;
+do
+  {
+  if (cc >= ccend)
+    break;
+
+  if (*cc == OP_CHAR)
+    {
+    size = 1;
+#ifdef SUPPORT_UTF
+    if (common->utf && HAS_EXTRALEN(cc[1]))
+      size += GET_EXTRALEN(cc[1]);
+#endif
+    }
+  else if (*cc == OP_CHARI)
+    {
+    size = 1;
+#ifdef SUPPORT_UTF
+    if (common->utf)
+      {
+      if (char_has_othercase(common, cc + 1) && char_get_othercase_bit(common, cc + 1) == 0)
+        size = 0;
+      else if (HAS_EXTRALEN(cc[1]))
+        size += GET_EXTRALEN(cc[1]);
+      }
+    else
+#endif
+    if (char_has_othercase(common, cc + 1) && char_get_othercase_bit(common, cc + 1) == 0)
+      size = 0;
+    }
+  else
+    size = 0;
+
+  cc += 1 + size;
+  context.length += IN_UCHARS(size);
+  }
+while (size > 0 && context.length <= 128);
+
+cc = ccbegin;
+if (context.length > 0)
+  {
+  /* We have a fixed-length byte sequence. */
+  OP2(SLJIT_ADD, STR_PTR, 0, STR_PTR, 0, SLJIT_IMM, context.length);
+  add_jump(compiler, backtracks, CMP(SLJIT_C_GREATER, STR_PTR, 0, STR_END, 0));
+
+  context.sourcereg = -1;
+#if defined SLJIT_UNALIGNED && SLJIT_UNALIGNED
+  context.ucharptr = 0;
+#endif
+  do cc = byte_sequence_compare(common, *cc == OP_CHARI, cc + 1, &context, backtracks); while (context.length > 0);
+  return cc;
+  }
+
+/* A non-fixed length character will be checked if length == 0. */
+return compile_char1_matchingpath(common, *cc, cc + 1, backtracks);
+}
+
+static struct sljit_jump *compile_ref_checks(compiler_common *common, pcre_uchar *cc, jump_list **backtracks)
+{
+DEFINE_COMPILER;
+int offset = GET2(cc, 1) << 1;
+
+OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(offset));
+if (!common->jscript_compat)
+  {
+  if (backtracks == NULL)
+    {
+    /* OVECTOR(1) contains the "string begin - 1" constant. */
+    OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(1));
+    OP_FLAGS(SLJIT_MOV, TMP2, 0, SLJIT_UNUSED, 0, SLJIT_C_EQUAL);
+    OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_UNUSED, 0, TMP1, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(offset + 1));
+    OP_FLAGS(SLJIT_OR | SLJIT_SET_E, TMP2, 0, TMP2, 0, SLJIT_C_EQUAL);
+    return JUMP(SLJIT_C_NOT_ZERO);
+    }
+  add_jump(compiler, backtracks, CMP(SLJIT_C_EQUAL, TMP1, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(1)));
+  }
+return CMP(SLJIT_C_EQUAL, TMP1, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(offset + 1));
+}
+
+/* Forward definitions. */
+static void compile_matchingpath(compiler_common *, pcre_uchar *, pcre_uchar *, backtrack_common *);
+static void compile_backtrackingpath(compiler_common *, struct backtrack_common *);
+
+#define PUSH_BACKTRACK(size, ccstart, error) \
+  do \
+    { \
+    backtrack = sljit_alloc_memory(compiler, (size)); \
+    if (SLJIT_UNLIKELY(sljit_get_compiler_error(compiler))) \
+      return error; \
+    memset(backtrack, 0, size); \
+    backtrack->prev = parent->top; \
+    backtrack->cc = (ccstart); \
+    parent->top = backtrack; \
+    } \
+  while (0)
+
+#define PUSH_BACKTRACK_NOVALUE(size, ccstart) \
+  do \
+    { \
+    backtrack = sljit_alloc_memory(compiler, (size)); \
+    if (SLJIT_UNLIKELY(sljit_get_compiler_error(compiler))) \
+      return; \
+    memset(backtrack, 0, size); \
+    backtrack->prev = parent->top; \
+    backtrack->cc = (ccstart); \
+    parent->top = backtrack; \
+    } \
+  while (0)
+
+#define BACKTRACK_AS(type) ((type *)backtrack)
+
+static pcre_uchar *compile_ref_matchingpath(compiler_common *common, pcre_uchar *cc, jump_list **backtracks, BOOL withchecks, BOOL emptyfail)
+{
+DEFINE_COMPILER;
+int offset = GET2(cc, 1) << 1;
+struct sljit_jump *jump = NULL;
+struct sljit_jump *partial;
+struct sljit_jump *nopartial;
+
+OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(offset));
+/* OVECTOR(1) contains the "string begin - 1" constant. */
+if (withchecks && !common->jscript_compat)
+  add_jump(compiler, backtracks, CMP(SLJIT_C_EQUAL, TMP1, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(1)));
+
+#if defined SUPPORT_UTF && defined SUPPORT_UCP
+if (common->utf && *cc == OP_REFI)
+  {
+  SLJIT_ASSERT(TMP1 == SLJIT_SCRATCH_REG1 && STACK_TOP == SLJIT_SCRATCH_REG2 && TMP2 == SLJIT_SCRATCH_REG3);
+  OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(offset + 1));
+  if (withchecks)
+    jump = CMP(SLJIT_C_EQUAL, TMP1, 0, TMP2, 0);
+
+  /* Needed to save important temporary registers. */
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), LOCALS0, STACK_TOP, 0);
+  OP1(SLJIT_MOV, SLJIT_SCRATCH_REG2, 0, ARGUMENTS, 0);
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_SCRATCH_REG2), SLJIT_OFFSETOF(jit_arguments, uchar_ptr), STR_PTR, 0);
+  sljit_emit_ijump(compiler, SLJIT_CALL3, SLJIT_IMM, SLJIT_FUNC_OFFSET(do_utf_caselesscmp));
+  OP1(SLJIT_MOV, STACK_TOP, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), LOCALS0);
+  if (common->mode == JIT_COMPILE)
+    add_jump(compiler, backtracks, CMP(SLJIT_C_LESS_EQUAL, SLJIT_RETURN_REG, 0, SLJIT_IMM, 1));
+  else
+    {
+    add_jump(compiler, backtracks, CMP(SLJIT_C_EQUAL, SLJIT_RETURN_REG, 0, SLJIT_IMM, 0));
+    nopartial = CMP(SLJIT_C_NOT_EQUAL, SLJIT_RETURN_REG, 0, SLJIT_IMM, 1);
+    check_partial(common, FALSE);
+    add_jump(compiler, backtracks, JUMP(SLJIT_JUMP));
+    JUMPHERE(nopartial);
+    }
+  OP1(SLJIT_MOV, STR_PTR, 0, SLJIT_RETURN_REG, 0);
+  }
+else
+#endif /* SUPPORT_UTF && SUPPORT_UCP */
+  {
+  OP2(SLJIT_SUB | SLJIT_SET_E, TMP2, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(offset + 1), TMP1, 0);
+  if (withchecks)
+    jump = JUMP(SLJIT_C_ZERO);
+
+  OP2(SLJIT_ADD, STR_PTR, 0, STR_PTR, 0, TMP2, 0);
+  partial = CMP(SLJIT_C_GREATER, STR_PTR, 0, STR_END, 0);
+  if (common->mode == JIT_COMPILE)
+    add_jump(compiler, backtracks, partial);
+
+  add_jump(compiler, *cc == OP_REF ? &common->casefulcmp : &common->caselesscmp, JUMP(SLJIT_FAST_CALL));
+  add_jump(compiler, backtracks, CMP(SLJIT_C_NOT_EQUAL, TMP2, 0, SLJIT_IMM, 0));
+
+  if (common->mode != JIT_COMPILE)
+    {
+    nopartial = JUMP(SLJIT_JUMP);
+    JUMPHERE(partial);
+    /* TMP2 -= STR_END - STR_PTR */
+    OP2(SLJIT_SUB, TMP2, 0, TMP2, 0, STR_PTR, 0);
+    OP2(SLJIT_ADD, TMP2, 0, TMP2, 0, STR_END, 0);
+    partial = CMP(SLJIT_C_EQUAL, TMP2, 0, SLJIT_IMM, 0);
+    OP1(SLJIT_MOV, STR_PTR, 0, STR_END, 0);
+    add_jump(compiler, *cc == OP_REF ? &common->casefulcmp : &common->caselesscmp, JUMP(SLJIT_FAST_CALL));
+    add_jump(compiler, backtracks, CMP(SLJIT_C_NOT_EQUAL, TMP2, 0, SLJIT_IMM, 0));
+    JUMPHERE(partial);
+    check_partial(common, FALSE);
+    add_jump(compiler, backtracks, JUMP(SLJIT_JUMP));
+    JUMPHERE(nopartial);
+    }
+  }
+
+if (jump != NULL)
+  {
+  if (emptyfail)
+    add_jump(compiler, backtracks, jump);
+  else
+    JUMPHERE(jump);
+  }
+return cc + 1 + IMM2_SIZE;
+}
+
+static SLJIT_INLINE pcre_uchar *compile_ref_iterator_matchingpath(compiler_common *common, pcre_uchar *cc, backtrack_common *parent)
+{
+DEFINE_COMPILER;
+backtrack_common *backtrack;
+pcre_uchar type;
+struct sljit_label *label;
+struct sljit_jump *zerolength;
+struct sljit_jump *jump = NULL;
+pcre_uchar *ccbegin = cc;
+int min = 0, max = 0;
+BOOL minimize;
+
+PUSH_BACKTRACK(sizeof(iterator_backtrack), cc, NULL);
+
+type = cc[1 + IMM2_SIZE];
+minimize = (type & 0x1) != 0;
+switch(type)
+  {
+  case OP_CRSTAR:
+  case OP_CRMINSTAR:
+  min = 0;
+  max = 0;
+  cc += 1 + IMM2_SIZE + 1;
+  break;
+  case OP_CRPLUS:
+  case OP_CRMINPLUS:
+  min = 1;
+  max = 0;
+  cc += 1 + IMM2_SIZE + 1;
+  break;
+  case OP_CRQUERY:
+  case OP_CRMINQUERY:
+  min = 0;
+  max = 1;
+  cc += 1 + IMM2_SIZE + 1;
+  break;
+  case OP_CRRANGE:
+  case OP_CRMINRANGE:
+  min = GET2(cc, 1 + IMM2_SIZE + 1);
+  max = GET2(cc, 1 + IMM2_SIZE + 1 + IMM2_SIZE);
+  cc += 1 + IMM2_SIZE + 1 + 2 * IMM2_SIZE;
+  break;
+  default:
+  SLJIT_ASSERT_STOP();
+  break;
+  }
+
+if (!minimize)
+  {
+  if (min == 0)
+    {
+    allocate_stack(common, 2);
+    OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(0), STR_PTR, 0);
+    OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(1), SLJIT_IMM, 0);
+    /* Temporary release of STR_PTR. */
+    OP2(SLJIT_SUB, STACK_TOP, 0, STACK_TOP, 0, SLJIT_IMM, sizeof(sljit_sw));
+    zerolength = compile_ref_checks(common, ccbegin, NULL);
+    /* Restore if not zero length. */
+    OP2(SLJIT_ADD, STACK_TOP, 0, STACK_TOP, 0, SLJIT_IMM, sizeof(sljit_sw));
+    }
+  else
+    {
+    allocate_stack(common, 1);
+    OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(0), SLJIT_IMM, 0);
+    zerolength = compile_ref_checks(common, ccbegin, &backtrack->topbacktracks);
+    }
+
+  if (min > 1 || max > 1)
+    OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), POSSESSIVE0, SLJIT_IMM, 0);
+
+  label = LABEL();
+  compile_ref_matchingpath(common, ccbegin, &backtrack->topbacktracks, FALSE, FALSE);
+
+  if (min > 1 || max > 1)
+    {
+    OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), POSSESSIVE0);
+    OP2(SLJIT_ADD, TMP1, 0, TMP1, 0, SLJIT_IMM, 1);
+    OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), POSSESSIVE0, TMP1, 0);
+    if (min > 1)
+      CMPTO(SLJIT_C_LESS, TMP1, 0, SLJIT_IMM, min, label);
+    if (max > 1)
+      {
+      jump = CMP(SLJIT_C_GREATER_EQUAL, TMP1, 0, SLJIT_IMM, max);
+      allocate_stack(common, 1);
+      OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(0), STR_PTR, 0);
+      JUMPTO(SLJIT_JUMP, label);
+      JUMPHERE(jump);
+      }
+    }
+
+  if (max == 0)
+    {
+    /* Includes min > 1 case as well. */
+    allocate_stack(common, 1);
+    OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(0), STR_PTR, 0);
+    JUMPTO(SLJIT_JUMP, label);
+    }
+
+  JUMPHERE(zerolength);
+  BACKTRACK_AS(iterator_backtrack)->matchingpath = LABEL();
+
+  count_match(common);
+  return cc;
+  }
+
+allocate_stack(common, 2);
+OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(0), SLJIT_IMM, 0);
+if (type != OP_CRMINSTAR)
+  OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(1), SLJIT_IMM, 0);
+
+if (min == 0)
+  {
+  zerolength = compile_ref_checks(common, ccbegin, NULL);
+  OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(0), STR_PTR, 0);
+  jump = JUMP(SLJIT_JUMP);
+  }
+else
+  zerolength = compile_ref_checks(common, ccbegin, &backtrack->topbacktracks);
+
+BACKTRACK_AS(iterator_backtrack)->matchingpath = LABEL();
+if (max > 0)
+  add_jump(compiler, &backtrack->topbacktracks, CMP(SLJIT_C_GREATER_EQUAL, SLJIT_MEM1(STACK_TOP), STACK(1), SLJIT_IMM, max));
+
+compile_ref_matchingpath(common, ccbegin, &backtrack->topbacktracks, TRUE, TRUE);
+OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(0), STR_PTR, 0);
+
+if (min > 1)
+  {
+  OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(STACK_TOP), STACK(1));
+  OP2(SLJIT_ADD, TMP1, 0, TMP1, 0, SLJIT_IMM, 1);
+  OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(1), TMP1, 0);
+  CMPTO(SLJIT_C_LESS, TMP1, 0, SLJIT_IMM, min, BACKTRACK_AS(iterator_backtrack)->matchingpath);
+  }
+else if (max > 0)
+  OP2(SLJIT_ADD, SLJIT_MEM1(STACK_TOP), STACK(1), SLJIT_MEM1(STACK_TOP), STACK(1), SLJIT_IMM, 1);
+
+if (jump != NULL)
+  JUMPHERE(jump);
+JUMPHERE(zerolength);
+
+count_match(common);
+return cc;
+}
+
+static SLJIT_INLINE pcre_uchar *compile_recurse_matchingpath(compiler_common *common, pcre_uchar *cc, backtrack_common *parent)
+{
+DEFINE_COMPILER;
+backtrack_common *backtrack;
+recurse_entry *entry = common->entries;
+recurse_entry *prev = NULL;
+sljit_sw start = GET(cc, 1);
+pcre_uchar *start_cc;
+BOOL needs_control_head;
+
+PUSH_BACKTRACK(sizeof(recurse_backtrack), cc, NULL);
+
+/* Inlining simple patterns. */
+if (get_framesize(common, common->start + start, NULL, TRUE, &needs_control_head) == no_stack)
+  {
+  start_cc = common->start + start;
+  compile_matchingpath(common, next_opcode(common, start_cc), bracketend(start_cc) - (1 + LINK_SIZE), backtrack);
+  BACKTRACK_AS(recurse_backtrack)->inlined_pattern = TRUE;
+  return cc + 1 + LINK_SIZE;
+  }
+
+while (entry != NULL)
+  {
+  if (entry->start == start)
+    break;
+  prev = entry;
+  entry = entry->next;
+  }
+
+if (entry == NULL)
+  {
+  entry = sljit_alloc_memory(compiler, sizeof(recurse_entry));
+  if (SLJIT_UNLIKELY(sljit_get_compiler_error(compiler)))
+    return NULL;
+  entry->next = NULL;
+  entry->entry = NULL;
+  entry->calls = NULL;
+  entry->start = start;
+
+  if (prev != NULL)
+    prev->next = entry;
+  else
+    common->entries = entry;
+  }
+
+if (common->has_set_som && common->mark_ptr != 0)
+  {
+  OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(0));
+  allocate_stack(common, 2);
+  OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), common->mark_ptr);
+  OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(0), TMP2, 0);
+  OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(1), TMP1, 0);
+  }
+else if (common->has_set_som || common->mark_ptr != 0)
+  {
+  OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), common->has_set_som ? (int)(OVECTOR(0)) : common->mark_ptr);
+  allocate_stack(common, 1);
+  OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(0), TMP2, 0);
+  }
+
+if (entry->entry == NULL)
+  add_jump(compiler, &entry->calls, JUMP(SLJIT_FAST_CALL));
+else
+  JUMPTO(SLJIT_FAST_CALL, entry->entry);
+/* Leave if the match is failed. */
+add_jump(compiler, &backtrack->topbacktracks, CMP(SLJIT_C_EQUAL, TMP1, 0, SLJIT_IMM, 0));
+return cc + 1 + LINK_SIZE;
+}
+
+static int SLJIT_CALL do_callout(struct jit_arguments* arguments, PUBL(callout_block) *callout_block, pcre_uchar **jit_ovector)
+{
+const pcre_uchar *begin = arguments->begin;
+int *offset_vector = arguments->offsets;
+int offset_count = arguments->offset_count;
+int i;
+
+if (PUBL(callout) == NULL)
+  return 0;
+
+callout_block->version = 2;
+callout_block->callout_data = arguments->callout_data;
+
+/* Offsets in subject. */
+callout_block->subject_length = arguments->end - arguments->begin;
+callout_block->start_match = (pcre_uchar*)callout_block->subject - arguments->begin;
+callout_block->current_position = (pcre_uchar*)callout_block->offset_vector - arguments->begin;
+#if defined COMPILE_PCRE8
+callout_block->subject = (PCRE_SPTR)begin;
+#elif defined COMPILE_PCRE16
+callout_block->subject = (PCRE_SPTR16)begin;
+#elif defined COMPILE_PCRE32
+callout_block->subject = (PCRE_SPTR32)begin;
+#endif
+
+/* Convert and copy the JIT offset vector to the offset_vector array. */
+callout_block->capture_top = 0;
+callout_block->offset_vector = offset_vector;
+for (i = 2; i < offset_count; i += 2)
+  {
+  offset_vector[i] = jit_ovector[i] - begin;
+  offset_vector[i + 1] = jit_ovector[i + 1] - begin;
+  if (jit_ovector[i] >= begin)
+    callout_block->capture_top = i;
+  }
+
+callout_block->capture_top = (callout_block->capture_top >> 1) + 1;
+if (offset_count > 0)
+  offset_vector[0] = -1;
+if (offset_count > 1)
+  offset_vector[1] = -1;
+return (*PUBL(callout))(callout_block);
+}
+
+/* Aligning to 8 byte. */
+#define CALLOUT_ARG_SIZE \
+    (((int)sizeof(PUBL(callout_block)) + 7) & ~7)
+
+#define CALLOUT_ARG_OFFSET(arg) \
+    (-CALLOUT_ARG_SIZE + SLJIT_OFFSETOF(PUBL(callout_block), arg))
+
+static SLJIT_INLINE pcre_uchar *compile_callout_matchingpath(compiler_common *common, pcre_uchar *cc, backtrack_common *parent)
+{
+DEFINE_COMPILER;
+backtrack_common *backtrack;
+
+PUSH_BACKTRACK(sizeof(backtrack_common), cc, NULL);
+
+allocate_stack(common, CALLOUT_ARG_SIZE / sizeof(sljit_sw));
+
+OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), common->capture_last_ptr);
+OP1(SLJIT_MOV, TMP1, 0, ARGUMENTS, 0);
+SLJIT_ASSERT(common->capture_last_ptr != 0);
+OP1(SLJIT_MOV_SI, SLJIT_MEM1(STACK_TOP), CALLOUT_ARG_OFFSET(callout_number), SLJIT_IMM, cc[1]);
+OP1(SLJIT_MOV_SI, SLJIT_MEM1(STACK_TOP), CALLOUT_ARG_OFFSET(capture_last), TMP2, 0);
+
+/* These pointer sized fields temporarly stores internal variables. */
+OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(0));
+OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), CALLOUT_ARG_OFFSET(offset_vector), STR_PTR, 0);
+OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), CALLOUT_ARG_OFFSET(subject), TMP2, 0);
+
+if (common->mark_ptr != 0)
+  OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(TMP1), SLJIT_OFFSETOF(jit_arguments, mark_ptr));
+OP1(SLJIT_MOV_SI, SLJIT_MEM1(STACK_TOP), CALLOUT_ARG_OFFSET(pattern_position), SLJIT_IMM, GET(cc, 2));
+OP1(SLJIT_MOV_SI, SLJIT_MEM1(STACK_TOP), CALLOUT_ARG_OFFSET(next_item_length), SLJIT_IMM, GET(cc, 2 + LINK_SIZE));
+OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), CALLOUT_ARG_OFFSET(mark), (common->mark_ptr != 0) ? TMP2 : SLJIT_IMM, 0);
+
+/* Needed to save important temporary registers. */
+OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), LOCALS0, STACK_TOP, 0);
+OP2(SLJIT_SUB, SLJIT_SCRATCH_REG2, 0, STACK_TOP, 0, SLJIT_IMM, CALLOUT_ARG_SIZE);
+GET_LOCAL_BASE(SLJIT_SCRATCH_REG3, 0, OVECTOR_START);
+sljit_emit_ijump(compiler, SLJIT_CALL3, SLJIT_IMM, SLJIT_FUNC_OFFSET(do_callout));
+OP1(SLJIT_MOV_SI, SLJIT_RETURN_REG, 0, SLJIT_RETURN_REG, 0);
+OP1(SLJIT_MOV, STACK_TOP, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), LOCALS0);
+free_stack(common, CALLOUT_ARG_SIZE / sizeof(sljit_sw));
+
+/* Check return value. */
+OP2(SLJIT_SUB | SLJIT_SET_S, SLJIT_UNUSED, 0, SLJIT_RETURN_REG, 0, SLJIT_IMM, 0);
+add_jump(compiler, &backtrack->topbacktracks, JUMP(SLJIT_C_SIG_GREATER));
+if (common->forced_quit_label == NULL)
+  add_jump(compiler, &common->forced_quit, JUMP(SLJIT_C_SIG_LESS));
+else
+  JUMPTO(SLJIT_C_SIG_LESS, common->forced_quit_label);
+return cc + 2 + 2 * LINK_SIZE;
+}
+
+#undef CALLOUT_ARG_SIZE
+#undef CALLOUT_ARG_OFFSET
+
+static pcre_uchar *compile_assert_matchingpath(compiler_common *common, pcre_uchar *cc, assert_backtrack *backtrack, BOOL conditional)
+{
+DEFINE_COMPILER;
+int framesize;
+int extrasize;
+BOOL needs_control_head;
+int private_data_ptr;
+backtrack_common altbacktrack;
+pcre_uchar *ccbegin;
+pcre_uchar opcode;
+pcre_uchar bra = OP_BRA;
+jump_list *tmp = NULL;
+jump_list **target = (conditional) ? &backtrack->condfailed : &backtrack->common.topbacktracks;
+jump_list **found;
+/* Saving previous accept variables. */
+BOOL save_local_exit = common->local_exit;
+BOOL save_positive_assert = common->positive_assert;
+then_trap_backtrack *save_then_trap = common->then_trap;
+struct sljit_label *save_quit_label = common->quit_label;
+struct sljit_label *save_accept_label = common->accept_label;
+jump_list *save_quit = common->quit;
+jump_list *save_positive_assert_quit = common->positive_assert_quit;
+jump_list *save_accept = common->accept;
+struct sljit_jump *jump;
+struct sljit_jump *brajump = NULL;
+
+/* Assert captures then. */
+common->then_trap = NULL;
+
+if (*cc == OP_BRAZERO || *cc == OP_BRAMINZERO)
+  {
+  SLJIT_ASSERT(!conditional);
+  bra = *cc;
+  cc++;
+  }
+private_data_ptr = PRIVATE_DATA(cc);
+SLJIT_ASSERT(private_data_ptr != 0);
+framesize = get_framesize(common, cc, NULL, FALSE, &needs_control_head);
+backtrack->framesize = framesize;
+backtrack->private_data_ptr = private_data_ptr;
+opcode = *cc;
+SLJIT_ASSERT(opcode >= OP_ASSERT && opcode <= OP_ASSERTBACK_NOT);
+found = (opcode == OP_ASSERT || opcode == OP_ASSERTBACK) ? &tmp : target;
+ccbegin = cc;
+cc += GET(cc, 1);
+
+if (bra == OP_BRAMINZERO)
+  {
+  /* This is a braminzero backtrack path. */
+  OP1(SLJIT_MOV, STR_PTR, 0, SLJIT_MEM1(STACK_TOP), STACK(0));
+  free_stack(common, 1);
+  brajump = CMP(SLJIT_C_EQUAL, STR_PTR, 0, SLJIT_IMM, 0);
+  }
+
+if (framesize < 0)
+  {
+  extrasize = needs_control_head ? 2 : 1;
+  if (framesize == no_frame)
+    OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr, STACK_TOP, 0);
+  allocate_stack(common, extrasize);
+  if (needs_control_head)
+    OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), common->control_head_ptr);
+  OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(0), STR_PTR, 0);
+  if (needs_control_head)
+    {
+    OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->control_head_ptr, SLJIT_IMM, 0);
+    OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(1), TMP1, 0);
+    }
+  }
+else
+  {
+  extrasize = needs_control_head ? 3 : 2;
+  allocate_stack(common, framesize + extrasize);
+  OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr);
+  OP2(SLJIT_SUB, TMP2, 0, STACK_TOP, 0, SLJIT_IMM, (framesize + extrasize) * sizeof(sljit_sw));
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr, TMP2, 0);
+  if (needs_control_head)
+    OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), common->control_head_ptr);
+  OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(0), STR_PTR, 0);
+  if (needs_control_head)
+    {
+    OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(2), TMP1, 0);
+    OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(1), TMP2, 0);
+    OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->control_head_ptr, SLJIT_IMM, 0);
+    }
+  else
+    OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(1), TMP1, 0);
+  init_frame(common, ccbegin, NULL, framesize + extrasize - 1, extrasize, FALSE);
+  }
+
+memset(&altbacktrack, 0, sizeof(backtrack_common));
+if (opcode == OP_ASSERT_NOT || opcode == OP_ASSERTBACK_NOT)
+  {
+  /* Negative assert is stronger than positive assert. */
+  common->local_exit = TRUE;
+  common->quit_label = NULL;
+  common->quit = NULL;
+  common->positive_assert = FALSE;
+  }
+else
+  common->positive_assert = TRUE;
+common->positive_assert_quit = NULL;
+
+while (1)
+  {
+  common->accept_label = NULL;
+  common->accept = NULL;
+  altbacktrack.top = NULL;
+  altbacktrack.topbacktracks = NULL;
+
+  if (*ccbegin == OP_ALT)
+    OP1(SLJIT_MOV, STR_PTR, 0, SLJIT_MEM1(STACK_TOP), STACK(0));
+
+  altbacktrack.cc = ccbegin;
+  compile_matchingpath(common, ccbegin + 1 + LINK_SIZE, cc, &altbacktrack);
+  if (SLJIT_UNLIKELY(sljit_get_compiler_error(compiler)))
+    {
+    if (opcode == OP_ASSERT_NOT || opcode == OP_ASSERTBACK_NOT)
+      {
+      common->local_exit = save_local_exit;
+      common->quit_label = save_quit_label;
+      common->quit = save_quit;
+      }
+    common->positive_assert = save_positive_assert;
+    common->then_trap = save_then_trap;
+    common->accept_label = save_accept_label;
+    common->positive_assert_quit = save_positive_assert_quit;
+    common->accept = save_accept;
+    return NULL;
+    }
+  common->accept_label = LABEL();
+  if (common->accept != NULL)
+    set_jumps(common->accept, common->accept_label);
+
+  /* Reset stack. */
+  if (framesize < 0)
+    {
+    if (framesize == no_frame)
+      OP1(SLJIT_MOV, STACK_TOP, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr);
+    else
+      free_stack(common, extrasize);
+    if (needs_control_head)
+      OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->control_head_ptr, SLJIT_MEM1(STACK_TOP), 0);
+    }
+  else
+    {
+    if ((opcode != OP_ASSERT_NOT && opcode != OP_ASSERTBACK_NOT) || conditional)
+      {
+      /* We don't need to keep the STR_PTR, only the previous private_data_ptr. */
+      OP2(SLJIT_ADD, STACK_TOP, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr, SLJIT_IMM, (framesize + 1) * sizeof(sljit_sw));
+      if (needs_control_head)
+        OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->control_head_ptr, SLJIT_MEM1(STACK_TOP), 0);
+      }
+    else
+      {
+      OP1(SLJIT_MOV, STACK_TOP, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr);
+      if (needs_control_head)
+        OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->control_head_ptr, SLJIT_MEM1(STACK_TOP), (framesize + 1) * sizeof(sljit_sw));
+      add_jump(compiler, &common->revertframes, JUMP(SLJIT_FAST_CALL));
+      }
+    }
+
+  if (opcode == OP_ASSERT_NOT || opcode == OP_ASSERTBACK_NOT)
+    {
+    /* We know that STR_PTR was stored on the top of the stack. */
+    if (conditional)
+      OP1(SLJIT_MOV, STR_PTR, 0, SLJIT_MEM1(STACK_TOP), needs_control_head ? sizeof(sljit_sw) : 0);
+    else if (bra == OP_BRAZERO)
+      {
+      if (framesize < 0)
+        OP1(SLJIT_MOV, STR_PTR, 0, SLJIT_MEM1(STACK_TOP), (extrasize - 1) * sizeof(sljit_sw));
+      else
+        {
+        OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(STACK_TOP), framesize * sizeof(sljit_sw));
+        OP1(SLJIT_MOV, STR_PTR, 0, SLJIT_MEM1(STACK_TOP), (framesize + extrasize - 1) * sizeof(sljit_sw));
+        OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr, TMP1, 0);
+        }
+      OP2(SLJIT_ADD, STACK_TOP, 0, STACK_TOP, 0, SLJIT_IMM, sizeof(sljit_sw));
+      OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(0), SLJIT_IMM, 0);
+      }
+    else if (framesize >= 0)
+      {
+      /* For OP_BRA and OP_BRAMINZERO. */
+      OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr, SLJIT_MEM1(STACK_TOP), framesize * sizeof(sljit_sw));
+      }
+    }
+  add_jump(compiler, found, JUMP(SLJIT_JUMP));
+
+  compile_backtrackingpath(common, altbacktrack.top);
+  if (SLJIT_UNLIKELY(sljit_get_compiler_error(compiler)))
+    {
+    if (opcode == OP_ASSERT_NOT || opcode == OP_ASSERTBACK_NOT)
+      {
+      common->local_exit = save_local_exit;
+      common->quit_label = save_quit_label;
+      common->quit = save_quit;
+      }
+    common->positive_assert = save_positive_assert;
+    common->then_trap = save_then_trap;
+    common->accept_label = save_accept_label;
+    common->positive_assert_quit = save_positive_assert_quit;
+    common->accept = save_accept;
+    return NULL;
+    }
+  set_jumps(altbacktrack.topbacktracks, LABEL());
+
+  if (*cc != OP_ALT)
+    break;
+
+  ccbegin = cc;
+  cc += GET(cc, 1);
+  }
+
+if (opcode == OP_ASSERT_NOT || opcode == OP_ASSERTBACK_NOT)
+  {
+  SLJIT_ASSERT(common->positive_assert_quit == NULL);
+  /* Makes the check less complicated below. */
+  common->positive_assert_quit = common->quit;
+  }
+
+/* None of them matched. */
+if (common->positive_assert_quit != NULL)
+  {
+  jump = JUMP(SLJIT_JUMP);
+  set_jumps(common->positive_assert_quit, LABEL());
+  SLJIT_ASSERT(framesize != no_stack);
+  if (framesize < 0)
+    OP2(SLJIT_ADD, STACK_TOP, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr, SLJIT_IMM, extrasize * sizeof(sljit_sw));
+  else
+    {
+    OP1(SLJIT_MOV, STACK_TOP, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr);
+    add_jump(compiler, &common->revertframes, JUMP(SLJIT_FAST_CALL));
+    OP2(SLJIT_ADD, STACK_TOP, 0, STACK_TOP, 0, SLJIT_IMM, (framesize + extrasize) * sizeof(sljit_sw));
+    }
+  JUMPHERE(jump);
+  }
+
+if (needs_control_head)
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->control_head_ptr, SLJIT_MEM1(STACK_TOP), STACK(1));
+
+if (opcode == OP_ASSERT || opcode == OP_ASSERTBACK)
+  {
+  /* Assert is failed. */
+  if (conditional || bra == OP_BRAZERO)
+    OP1(SLJIT_MOV, STR_PTR, 0, SLJIT_MEM1(STACK_TOP), STACK(0));
+
+  if (framesize < 0)
+    {
+    /* The topmost item should be 0. */
+    if (bra == OP_BRAZERO)
+      {
+      if (extrasize == 2)
+        free_stack(common, 1);
+      OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(0), SLJIT_IMM, 0);
+      }
+    else
+      free_stack(common, extrasize);
+    }
+  else
+    {
+    OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(STACK_TOP), STACK(extrasize - 1));
+    /* The topmost item should be 0. */
+    if (bra == OP_BRAZERO)
+      {
+      free_stack(common, framesize + extrasize - 1);
+      OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(0), SLJIT_IMM, 0);
+      }
+    else
+      free_stack(common, framesize + extrasize);
+    OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr, TMP1, 0);
+    }
+  jump = JUMP(SLJIT_JUMP);
+  if (bra != OP_BRAZERO)
+    add_jump(compiler, target, jump);
+
+  /* Assert is successful. */
+  set_jumps(tmp, LABEL());
+  if (framesize < 0)
+    {
+    /* We know that STR_PTR was stored on the top of the stack. */
+    OP1(SLJIT_MOV, STR_PTR, 0, SLJIT_MEM1(STACK_TOP), (extrasize - 1) * sizeof(sljit_sw));
+    /* Keep the STR_PTR on the top of the stack. */
+    if (bra == OP_BRAZERO)
+      {
+      OP2(SLJIT_ADD, STACK_TOP, 0, STACK_TOP, 0, SLJIT_IMM, sizeof(sljit_sw));
+      if (extrasize == 2)
+        OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(0), STR_PTR, 0);
+      }
+    else if (bra == OP_BRAMINZERO)
+      {
+      OP2(SLJIT_ADD, STACK_TOP, 0, STACK_TOP, 0, SLJIT_IMM, sizeof(sljit_sw));
+      OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(0), SLJIT_IMM, 0);
+      }
+    }
+  else
+    {
+    if (bra == OP_BRA)
+      {
+      /* We don't need to keep the STR_PTR, only the previous private_data_ptr. */
+      OP2(SLJIT_ADD, STACK_TOP, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr, SLJIT_IMM, (framesize + 1) * sizeof(sljit_sw));
+      OP1(SLJIT_MOV, STR_PTR, 0, SLJIT_MEM1(STACK_TOP), (extrasize - 2) * sizeof(sljit_sw));
+      }
+    else
+      {
+      /* We don't need to keep the STR_PTR, only the previous private_data_ptr. */
+      OP2(SLJIT_ADD, STACK_TOP, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr, SLJIT_IMM, (framesize + 2) * sizeof(sljit_sw));
+      if (extrasize == 2)
+        {
+        OP1(SLJIT_MOV, STR_PTR, 0, SLJIT_MEM1(STACK_TOP), STACK(0));
+        if (bra == OP_BRAMINZERO)
+          OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(0), SLJIT_IMM, 0);
+        }
+      else
+        {
+        OP1(SLJIT_MOV, STR_PTR, 0, SLJIT_MEM1(STACK_TOP), 0);
+        OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(0), bra == OP_BRAZERO ? STR_PTR : SLJIT_IMM, 0);
+        }
+      }
+    }
+
+  if (bra == OP_BRAZERO)
+    {
+    backtrack->matchingpath = LABEL();
+    SET_LABEL(jump, backtrack->matchingpath);
+    }
+  else if (bra == OP_BRAMINZERO)
+    {
+    JUMPTO(SLJIT_JUMP, backtrack->matchingpath);
+    JUMPHERE(brajump);
+    if (framesize >= 0)
+      {
+      OP1(SLJIT_MOV, STACK_TOP, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr);
+      add_jump(compiler, &common->revertframes, JUMP(SLJIT_FAST_CALL));
+      OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr, SLJIT_MEM1(STACK_TOP), framesize * sizeof(sljit_sw));
+      }
+    set_jumps(backtrack->common.topbacktracks, LABEL());
+    }
+  }
+else
+  {
+  /* AssertNot is successful. */
+  if (framesize < 0)
+    {
+    OP1(SLJIT_MOV, STR_PTR, 0, SLJIT_MEM1(STACK_TOP), STACK(0));
+    if (bra != OP_BRA)
+      {
+      if (extrasize == 2)
+        free_stack(common, 1);
+      OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(0), SLJIT_IMM, 0);
+      }
+    else
+      free_stack(common, extrasize);
+    }
+  else
+    {
+    OP1(SLJIT_MOV, STR_PTR, 0, SLJIT_MEM1(STACK_TOP), STACK(0));
+    OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(STACK_TOP), STACK(extrasize - 1));
+    /* The topmost item should be 0. */
+    if (bra != OP_BRA)
+      {
+      free_stack(common, framesize + extrasize - 1);
+      OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(0), SLJIT_IMM, 0);
+      }
+    else
+      free_stack(common, framesize + extrasize);
+    OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr, TMP1, 0);
+    }
+
+  if (bra == OP_BRAZERO)
+    backtrack->matchingpath = LABEL();
+  else if (bra == OP_BRAMINZERO)
+    {
+    JUMPTO(SLJIT_JUMP, backtrack->matchingpath);
+    JUMPHERE(brajump);
+    }
+
+  if (bra != OP_BRA)
+    {
+    SLJIT_ASSERT(found == &backtrack->common.topbacktracks);
+    set_jumps(backtrack->common.topbacktracks, LABEL());
+    backtrack->common.topbacktracks = NULL;
+    }
+  }
+
+if (opcode == OP_ASSERT_NOT || opcode == OP_ASSERTBACK_NOT)
+  {
+  common->local_exit = save_local_exit;
+  common->quit_label = save_quit_label;
+  common->quit = save_quit;
+  }
+common->positive_assert = save_positive_assert;
+common->then_trap = save_then_trap;
+common->accept_label = save_accept_label;
+common->positive_assert_quit = save_positive_assert_quit;
+common->accept = save_accept;
+return cc + 1 + LINK_SIZE;
+}
+
+static sljit_sw SLJIT_CALL do_searchovector(sljit_uw refno, sljit_sw* locals, pcre_uchar *name_table)
+{
+int condition = FALSE;
+pcre_uchar *slotA = name_table;
+pcre_uchar *slotB;
+sljit_sw name_count = locals[LOCALS0 / sizeof(sljit_sw)];
+sljit_sw name_entry_size = locals[LOCALS1 / sizeof(sljit_sw)];
+sljit_sw no_capture;
+int i;
+
+locals += refno & 0xff;
+refno >>= 8;
+no_capture = locals[1];
+
+for (i = 0; i < name_count; i++)
+  {
+  if (GET2(slotA, 0) == refno) break;
+  slotA += name_entry_size;
+  }
+
+if (i < name_count)
+  {
+  /* Found a name for the number - there can be only one; duplicate names
+  for different numbers are allowed, but not vice versa. First scan down
+  for duplicates. */
+
+  slotB = slotA;
+  while (slotB > name_table)
+    {
+    slotB -= name_entry_size;
+    if (STRCMP_UC_UC(slotA + IMM2_SIZE, slotB + IMM2_SIZE) == 0)
+      {
+      condition = locals[GET2(slotB, 0) << 1] != no_capture;
+      if (condition) break;
+      }
+    else break;
+    }
+
+  /* Scan up for duplicates */
+  if (!condition)
+    {
+    slotB = slotA;
+    for (i++; i < name_count; i++)
+      {
+      slotB += name_entry_size;
+      if (STRCMP_UC_UC(slotA + IMM2_SIZE, slotB + IMM2_SIZE) == 0)
+        {
+        condition = locals[GET2(slotB, 0) << 1] != no_capture;
+        if (condition) break;
+        }
+      else break;
+      }
+    }
+  }
+return condition;
+}
+
+static sljit_sw SLJIT_CALL do_searchgroups(sljit_uw recno, sljit_uw* locals, pcre_uchar *name_table)
+{
+int condition = FALSE;
+pcre_uchar *slotA = name_table;
+pcre_uchar *slotB;
+sljit_uw name_count = locals[LOCALS0 / sizeof(sljit_sw)];
+sljit_uw name_entry_size = locals[LOCALS1 / sizeof(sljit_sw)];
+sljit_uw group_num = locals[POSSESSIVE0 / sizeof(sljit_sw)];
+sljit_uw i;
+
+for (i = 0; i < name_count; i++)
+  {
+  if (GET2(slotA, 0) == recno) break;
+  slotA += name_entry_size;
+  }
+
+if (i < name_count)
+  {
+  /* Found a name for the number - there can be only one; duplicate
+  names for different numbers are allowed, but not vice versa. First
+  scan down for duplicates. */
+
+  slotB = slotA;
+  while (slotB > name_table)
+    {
+    slotB -= name_entry_size;
+    if (STRCMP_UC_UC(slotA + IMM2_SIZE, slotB + IMM2_SIZE) == 0)
+      {
+      condition = GET2(slotB, 0) == group_num;
+      if (condition) break;
+      }
+    else break;
+    }
+
+  /* Scan up for duplicates */
+  if (!condition)
+    {
+    slotB = slotA;
+    for (i++; i < name_count; i++)
+      {
+      slotB += name_entry_size;
+      if (STRCMP_UC_UC(slotA + IMM2_SIZE, slotB + IMM2_SIZE) == 0)
+        {
+        condition = GET2(slotB, 0) == group_num;
+        if (condition) break;
+        }
+      else break;
+      }
+    }
+  }
+return condition;
+}
+
+static SLJIT_INLINE void match_once_common(compiler_common *common, pcre_uchar ket, int framesize, int private_data_ptr, BOOL has_alternatives, BOOL needs_control_head)
+{
+DEFINE_COMPILER;
+int stacksize;
+
+if (framesize < 0)
+  {
+  if (framesize == no_frame)
+    OP1(SLJIT_MOV, STACK_TOP, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr);
+  else
+    {
+    stacksize = needs_control_head ? 1 : 0;
+    if (ket != OP_KET || has_alternatives)
+      stacksize++;
+    free_stack(common, stacksize);
+    }
+
+  if (needs_control_head)
+    OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(STACK_TOP), (ket != OP_KET || has_alternatives) ? sizeof(sljit_sw) : 0);
+
+  /* TMP2 which is set here used by OP_KETRMAX below. */
+  if (ket == OP_KETRMAX)
+    OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(STACK_TOP), 0);
+  else if (ket == OP_KETRMIN)
+    {
+    /* Move the STR_PTR to the private_data_ptr. */
+    OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr, SLJIT_MEM1(STACK_TOP), 0);
+    }
+  }
+else
+  {
+  stacksize = (ket != OP_KET || has_alternatives) ? 2 : 1;
+  OP2(SLJIT_ADD, STACK_TOP, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr, SLJIT_IMM, (framesize + stacksize) * sizeof(sljit_sw));
+  if (needs_control_head)
+    OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(STACK_TOP), 0);
+
+  if (ket == OP_KETRMAX)
+    {
+    /* TMP2 which is set here used by OP_KETRMAX below. */
+    OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(STACK_TOP), STACK(0));
+    }
+  }
+if (needs_control_head)
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->control_head_ptr, TMP1, 0);
+}
+
+static SLJIT_INLINE int match_capture_common(compiler_common *common, int stacksize, int offset, int private_data_ptr)
+{
+DEFINE_COMPILER;
+
+if (common->capture_last_ptr != 0)
+  {
+  OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), common->capture_last_ptr);
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->capture_last_ptr, SLJIT_IMM, offset >> 1);
+  OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(stacksize), TMP1, 0);
+  stacksize++;
+  }
+if (common->optimized_cbracket[offset >> 1] == 0)
+  {
+  OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(offset));
+  OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(offset + 1));
+  OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(stacksize), TMP1, 0);
+  OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr);
+  OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(stacksize + 1), TMP2, 0);
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(offset + 1), STR_PTR, 0);
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(offset), TMP1, 0);
+  stacksize += 2;
+  }
+return stacksize;
+}
+
+/*
+  Handling bracketed expressions is probably the most complex part.
+
+  Stack layout naming characters:
+    S - Push the current STR_PTR
+    0 - Push a 0 (NULL)
+    A - Push the current STR_PTR. Needed for restoring the STR_PTR
+        before the next alternative. Not pushed if there are no alternatives.
+    M - Any values pushed by the current alternative. Can be empty, or anything.
+    C - Push the previous OVECTOR(i), OVECTOR(i+1) and OVECTOR_PRIV(i) to the stack.
+    L - Push the previous local (pointed by localptr) to the stack
+   () - opional values stored on the stack
+  ()* - optonal, can be stored multiple times
+
+  The following list shows the regular expression templates, their PCRE byte codes
+  and stack layout supported by pcre-sljit.
+
+  (?:)                     OP_BRA     | OP_KET                A M
+  ()                       OP_CBRA    | OP_KET                C M
+  (?:)+                    OP_BRA     | OP_KETRMAX        0   A M S   ( A M S )*
+                           OP_SBRA    | OP_KETRMAX        0   L M S   ( L M S )*
+  (?:)+?                   OP_BRA     | OP_KETRMIN        0   A M S   ( A M S )*
+                           OP_SBRA    | OP_KETRMIN        0   L M S   ( L M S )*
+  ()+                      OP_CBRA    | OP_KETRMAX        0   C M S   ( C M S )*
+                           OP_SCBRA   | OP_KETRMAX        0   C M S   ( C M S )*
+  ()+?                     OP_CBRA    | OP_KETRMIN        0   C M S   ( C M S )*
+                           OP_SCBRA   | OP_KETRMIN        0   C M S   ( C M S )*
+  (?:)?    OP_BRAZERO    | OP_BRA     | OP_KET            S ( A M 0 )
+  (?:)??   OP_BRAMINZERO | OP_BRA     | OP_KET            S ( A M 0 )
+  ()?      OP_BRAZERO    | OP_CBRA    | OP_KET            S ( C M 0 )
+  ()??     OP_BRAMINZERO | OP_CBRA    | OP_KET            S ( C M 0 )
+  (?:)*    OP_BRAZERO    | OP_BRA     | OP_KETRMAX      S 0 ( A M S )*
+           OP_BRAZERO    | OP_SBRA    | OP_KETRMAX      S 0 ( L M S )*
+  (?:)*?   OP_BRAMINZERO | OP_BRA     | OP_KETRMIN      S 0 ( A M S )*
+           OP_BRAMINZERO | OP_SBRA    | OP_KETRMIN      S 0 ( L M S )*
+  ()*      OP_BRAZERO    | OP_CBRA    | OP_KETRMAX      S 0 ( C M S )*
+           OP_BRAZERO    | OP_SCBRA   | OP_KETRMAX      S 0 ( C M S )*
+  ()*?     OP_BRAMINZERO | OP_CBRA    | OP_KETRMIN      S 0 ( C M S )*
+           OP_BRAMINZERO | OP_SCBRA   | OP_KETRMIN      S 0 ( C M S )*
+
+
+  Stack layout naming characters:
+    A - Push the alternative index (starting from 0) on the stack.
+        Not pushed if there is no alternatives.
+    M - Any values pushed by the current alternative. Can be empty, or anything.
+
+  The next list shows the possible content of a bracket:
+  (|)     OP_*BRA    | OP_ALT ...         M A
+  (?()|)  OP_*COND   | OP_ALT             M A
+  (?>|)   OP_ONCE    | OP_ALT ...         [stack trace] M A
+  (?>|)   OP_ONCE_NC | OP_ALT ...         [stack trace] M A
+                                          Or nothing, if trace is unnecessary
+*/
+
+static pcre_uchar *compile_bracket_matchingpath(compiler_common *common, pcre_uchar *cc, backtrack_common *parent)
+{
+DEFINE_COMPILER;
+backtrack_common *backtrack;
+pcre_uchar opcode;
+int private_data_ptr = 0;
+int offset = 0;
+int stacksize;
+int repeat_ptr = 0, repeat_length = 0;
+int repeat_type = 0, repeat_count = 0;
+pcre_uchar *ccbegin;
+pcre_uchar *matchingpath;
+pcre_uchar bra = OP_BRA;
+pcre_uchar ket;
+assert_backtrack *assert;
+BOOL has_alternatives;
+BOOL needs_control_head = FALSE;
+struct sljit_jump *jump;
+struct sljit_jump *skip;
+struct sljit_label *rmax_label = NULL;
+struct sljit_jump *braminzero = NULL;
+
+PUSH_BACKTRACK(sizeof(bracket_backtrack), cc, NULL);
+
+if (*cc == OP_BRAZERO || *cc == OP_BRAMINZERO)
+  {
+  bra = *cc;
+  cc++;
+  opcode = *cc;
+  }
+
+opcode = *cc;
+ccbegin = cc;
+matchingpath = bracketend(cc) - 1 - LINK_SIZE;
+ket = *matchingpath;
+if (ket == OP_KET && PRIVATE_DATA(matchingpath) != 0)
+  {
+  repeat_ptr = PRIVATE_DATA(matchingpath);
+  repeat_length = PRIVATE_DATA(matchingpath + 1);
+  repeat_type = PRIVATE_DATA(matchingpath + 2);
+  repeat_count = PRIVATE_DATA(matchingpath + 3);
+  SLJIT_ASSERT(repeat_length != 0 && repeat_type != 0 && repeat_count != 0);
+  if (repeat_type == OP_UPTO)
+    ket = OP_KETRMAX;
+  if (repeat_type == OP_MINUPTO)
+    ket = OP_KETRMIN;
+  }
+
+if ((opcode == OP_COND || opcode == OP_SCOND) && cc[1 + LINK_SIZE] == OP_DEF)
+  {
+  /* Drop this bracket_backtrack. */
+  parent->top = backtrack->prev;
+  return matchingpath + 1 + LINK_SIZE + repeat_length;
+  }
+
+matchingpath = ccbegin + 1 + LINK_SIZE;
+SLJIT_ASSERT(ket == OP_KET || ket == OP_KETRMAX || ket == OP_KETRMIN);
+SLJIT_ASSERT(!((bra == OP_BRAZERO && ket == OP_KETRMIN) || (bra == OP_BRAMINZERO && ket == OP_KETRMAX)));
+cc += GET(cc, 1);
+
+has_alternatives = *cc == OP_ALT;
+if (SLJIT_UNLIKELY(opcode == OP_COND) || SLJIT_UNLIKELY(opcode == OP_SCOND))
+  {
+  has_alternatives = (*matchingpath == OP_RREF) ? FALSE : TRUE;
+  if (*matchingpath == OP_NRREF)
+    {
+    stacksize = GET2(matchingpath, 1);
+    if (common->currententry == NULL || stacksize == RREF_ANY)
+      has_alternatives = FALSE;
+    else if (common->currententry->start == 0)
+      has_alternatives = stacksize != 0;
+    else
+      has_alternatives = stacksize != (int)GET2(common->start, common->currententry->start + 1 + LINK_SIZE);
+    }
+  }
+
+if (SLJIT_UNLIKELY(opcode == OP_COND) && (*cc == OP_KETRMAX || *cc == OP_KETRMIN))
+  opcode = OP_SCOND;
+if (SLJIT_UNLIKELY(opcode == OP_ONCE_NC))
+  opcode = OP_ONCE;
+
+if (opcode == OP_CBRA || opcode == OP_SCBRA)
+  {
+  /* Capturing brackets has a pre-allocated space. */
+  offset = GET2(ccbegin, 1 + LINK_SIZE);
+  if (common->optimized_cbracket[offset] == 0)
+    {
+    private_data_ptr = OVECTOR_PRIV(offset);
+    offset <<= 1;
+    }
+  else
+    {
+    offset <<= 1;
+    private_data_ptr = OVECTOR(offset);
+    }
+  BACKTRACK_AS(bracket_backtrack)->private_data_ptr = private_data_ptr;
+  matchingpath += IMM2_SIZE;
+  }
+else if (opcode == OP_ONCE || opcode == OP_SBRA || opcode == OP_SCOND)
+  {
+  /* Other brackets simply allocate the next entry. */
+  private_data_ptr = PRIVATE_DATA(ccbegin);
+  SLJIT_ASSERT(private_data_ptr != 0);
+  BACKTRACK_AS(bracket_backtrack)->private_data_ptr = private_data_ptr;
+  if (opcode == OP_ONCE)
+    BACKTRACK_AS(bracket_backtrack)->u.framesize = get_framesize(common, ccbegin, NULL, FALSE, &needs_control_head);
+  }
+
+/* Instructions before the first alternative. */
+stacksize = 0;
+if (ket == OP_KETRMAX || (ket == OP_KETRMIN && bra != OP_BRAMINZERO))
+  stacksize++;
+if (bra == OP_BRAZERO)
+  stacksize++;
+
+if (stacksize > 0)
+  allocate_stack(common, stacksize);
+
+stacksize = 0;
+if (ket == OP_KETRMAX || (ket == OP_KETRMIN && bra != OP_BRAMINZERO))
+  {
+  OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(stacksize), SLJIT_IMM, 0);
+  stacksize++;
+  }
+
+if (bra == OP_BRAZERO)
+  OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(stacksize), STR_PTR, 0);
+
+if (bra == OP_BRAMINZERO)
+  {
+  /* This is a backtrack path! (Since the try-path of OP_BRAMINZERO matches to the empty string) */
+  OP1(SLJIT_MOV, STR_PTR, 0, SLJIT_MEM1(STACK_TOP), STACK(0));
+  if (ket != OP_KETRMIN)
+    {
+    free_stack(common, 1);
+    braminzero = CMP(SLJIT_C_EQUAL, STR_PTR, 0, SLJIT_IMM, 0);
+    }
+  else
+    {
+    if (opcode == OP_ONCE || opcode >= OP_SBRA)
+      {
+      jump = CMP(SLJIT_C_NOT_EQUAL, STR_PTR, 0, SLJIT_IMM, 0);
+      OP1(SLJIT_MOV, STR_PTR, 0, SLJIT_MEM1(STACK_TOP), STACK(1));
+      /* Nothing stored during the first run. */
+      skip = JUMP(SLJIT_JUMP);
+      JUMPHERE(jump);
+      /* Checking zero-length iteration. */
+      if (opcode != OP_ONCE || BACKTRACK_AS(bracket_backtrack)->u.framesize < 0)
+        {
+        /* When we come from outside, private_data_ptr contains the previous STR_PTR. */
+        braminzero = CMP(SLJIT_C_EQUAL, STR_PTR, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr);
+        }
+      else
+        {
+        /* Except when the whole stack frame must be saved. */
+        OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr);
+        braminzero = CMP(SLJIT_C_EQUAL, STR_PTR, 0, SLJIT_MEM1(TMP1), (BACKTRACK_AS(bracket_backtrack)->u.framesize + 1) * sizeof(sljit_sw));
+        }
+      JUMPHERE(skip);
+      }
+    else
+      {
+      jump = CMP(SLJIT_C_NOT_EQUAL, STR_PTR, 0, SLJIT_IMM, 0);
+      OP1(SLJIT_MOV, STR_PTR, 0, SLJIT_MEM1(STACK_TOP), STACK(1));
+      JUMPHERE(jump);
+      }
+    }
+  }
+
+if (repeat_type != 0)
+  {
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), repeat_ptr, SLJIT_IMM, repeat_count);
+  if (repeat_type == OP_EXACT)
+    rmax_label = LABEL();
+  }
+
+if (ket == OP_KETRMIN)
+  BACKTRACK_AS(bracket_backtrack)->recursive_matchingpath = LABEL();
+
+if (ket == OP_KETRMAX)
+  {
+  rmax_label = LABEL();
+  if (has_alternatives && opcode != OP_ONCE && opcode < OP_SBRA && repeat_type == 0)
+    BACKTRACK_AS(bracket_backtrack)->alternative_matchingpath = rmax_label;
+  }
+
+/* Handling capturing brackets and alternatives. */
+if (opcode == OP_ONCE)
+  {
+  stacksize = 0;
+  if (needs_control_head)
+    {
+    OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), common->control_head_ptr);
+    stacksize++;
+    }
+
+  if (BACKTRACK_AS(bracket_backtrack)->u.framesize < 0)
+    {
+    /* Neither capturing brackets nor recursions are found in the block. */
+    if (ket == OP_KETRMIN)
+      {
+      stacksize += 2;
+      if (!needs_control_head)
+        OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr);
+      }
+    else
+      {
+      if (BACKTRACK_AS(bracket_backtrack)->u.framesize == no_frame)
+        OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr, STACK_TOP, 0);
+      if (ket == OP_KETRMAX || has_alternatives)
+        stacksize++;
+      }
+
+    if (stacksize > 0)
+      allocate_stack(common, stacksize);
+
+    stacksize = 0;
+    if (needs_control_head)
+      {
+      stacksize++;
+      OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(0), TMP2, 0);
+      }
+
+    if (ket == OP_KETRMIN)
+      {
+      if (needs_control_head)
+        OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr);
+      OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(stacksize), STR_PTR, 0);
+      if (BACKTRACK_AS(bracket_backtrack)->u.framesize == no_frame)
+        OP2(SLJIT_SUB, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr, STACK_TOP, 0, SLJIT_IMM, needs_control_head ? (2 * sizeof(sljit_sw)) : sizeof(sljit_sw));
+      OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(stacksize + 1), TMP2, 0);
+      }
+    else if (ket == OP_KETRMAX || has_alternatives)
+      OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(stacksize), STR_PTR, 0);
+    }
+  else
+    {
+    if (ket != OP_KET || has_alternatives)
+      stacksize++;
+
+    stacksize += BACKTRACK_AS(bracket_backtrack)->u.framesize + 1;
+    allocate_stack(common, stacksize);
+
+    if (needs_control_head)
+      OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(0), TMP2, 0);
+
+    OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr);
+    OP2(SLJIT_SUB, TMP2, 0, STACK_TOP, 0, SLJIT_IMM, stacksize * sizeof(sljit_sw));
+
+    stacksize = needs_control_head ? 1 : 0;
+    if (ket != OP_KET || has_alternatives)
+      {
+      OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(stacksize), STR_PTR, 0);
+      OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr, TMP2, 0);
+      stacksize++;
+      OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(stacksize), TMP1, 0);
+      }
+    else
+      {
+      OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr, TMP2, 0);
+      OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(stacksize), TMP1, 0);
+      }
+    init_frame(common, ccbegin, NULL, BACKTRACK_AS(bracket_backtrack)->u.framesize + stacksize, stacksize + 1, FALSE);
+    }
+  }
+else if (opcode == OP_CBRA || opcode == OP_SCBRA)
+  {
+  /* Saving the previous values. */
+  if (common->optimized_cbracket[offset >> 1] != 0)
+    {
+    SLJIT_ASSERT(private_data_ptr == OVECTOR(offset));
+    allocate_stack(common, 2);
+    OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr);
+    OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr + sizeof(sljit_sw));
+    OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr, STR_PTR, 0);
+    OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(0), TMP1, 0);
+    OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(1), TMP2, 0);
+    }
+  else
+    {
+    OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr);
+    allocate_stack(common, 1);
+    OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr, STR_PTR, 0);
+    OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(0), TMP2, 0);
+    }
+  }
+else if (opcode == OP_SBRA || opcode == OP_SCOND)
+  {
+  /* Saving the previous value. */
+  OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr);
+  allocate_stack(common, 1);
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr, STR_PTR, 0);
+  OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(0), TMP2, 0);
+  }
+else if (has_alternatives)
+  {
+  /* Pushing the starting string pointer. */
+  allocate_stack(common, 1);
+  OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(0), STR_PTR, 0);
+  }
+
+/* Generating code for the first alternative. */
+if (opcode == OP_COND || opcode == OP_SCOND)
+  {
+  if (*matchingpath == OP_CREF)
+    {
+    SLJIT_ASSERT(has_alternatives);
+    add_jump(compiler, &(BACKTRACK_AS(bracket_backtrack)->u.condfailed),
+      CMP(SLJIT_C_EQUAL, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(GET2(matchingpath, 1) << 1), SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(1)));
+    matchingpath += 1 + IMM2_SIZE;
+    }
+  else if (*matchingpath == OP_NCREF)
+    {
+    SLJIT_ASSERT(has_alternatives);
+    stacksize = GET2(matchingpath, 1);
+    jump = CMP(SLJIT_C_NOT_EQUAL, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(stacksize << 1), SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(1));
+
+    OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), POSSESSIVE1, STACK_TOP, 0);
+    OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), LOCALS0, SLJIT_IMM, common->name_count);
+    OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), LOCALS1, SLJIT_IMM, common->name_entry_size);
+    OP1(SLJIT_MOV, SLJIT_SCRATCH_REG1, 0, SLJIT_IMM, (stacksize << 8) | (common->ovector_start / sizeof(sljit_sw)));
+    GET_LOCAL_BASE(SLJIT_SCRATCH_REG2, 0, 0);
+    OP1(SLJIT_MOV, SLJIT_SCRATCH_REG3, 0, SLJIT_IMM, common->name_table);
+    sljit_emit_ijump(compiler, SLJIT_CALL3, SLJIT_IMM, SLJIT_FUNC_OFFSET(do_searchovector));
+    OP1(SLJIT_MOV, STACK_TOP, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), POSSESSIVE1);
+    add_jump(compiler, &(BACKTRACK_AS(bracket_backtrack)->u.condfailed), CMP(SLJIT_C_EQUAL, SLJIT_SCRATCH_REG1, 0, SLJIT_IMM, 0));
+
+    JUMPHERE(jump);
+    matchingpath += 1 + IMM2_SIZE;
+    }
+  else if (*matchingpath == OP_RREF || *matchingpath == OP_NRREF)
+    {
+    /* Never has other case. */
+    BACKTRACK_AS(bracket_backtrack)->u.condfailed = NULL;
+
+    stacksize = GET2(matchingpath, 1);
+    if (common->currententry == NULL)
+      stacksize = 0;
+    else if (stacksize == RREF_ANY)
+      stacksize = 1;
+    else if (common->currententry->start == 0)
+      stacksize = stacksize == 0;
+    else
+      stacksize = stacksize == (int)GET2(common->start, common->currententry->start + 1 + LINK_SIZE);
+
+    if (*matchingpath == OP_RREF || stacksize || common->currententry == NULL)
+      {
+      SLJIT_ASSERT(!has_alternatives);
+      if (stacksize != 0)
+        matchingpath += 1 + IMM2_SIZE;
+      else
+        {
+        if (*cc == OP_ALT)
+          {
+          matchingpath = cc + 1 + LINK_SIZE;
+          cc += GET(cc, 1);
+          }
+        else
+          matchingpath = cc;
+        }
+      }
+    else
+      {
+      SLJIT_ASSERT(has_alternatives);
+
+      stacksize = GET2(matchingpath, 1);
+      OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), POSSESSIVE1, STACK_TOP, 0);
+      OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), LOCALS0, SLJIT_IMM, common->name_count);
+      OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), LOCALS1, SLJIT_IMM, common->name_entry_size);
+      OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), POSSESSIVE0, SLJIT_IMM, GET2(common->start, common->currententry->start + 1 + LINK_SIZE));
+      OP1(SLJIT_MOV, SLJIT_SCRATCH_REG1, 0, SLJIT_IMM, stacksize);
+      GET_LOCAL_BASE(SLJIT_SCRATCH_REG2, 0, 0);
+      OP1(SLJIT_MOV, SLJIT_SCRATCH_REG3, 0, SLJIT_IMM, common->name_table);
+      sljit_emit_ijump(compiler, SLJIT_CALL3, SLJIT_IMM, SLJIT_FUNC_OFFSET(do_searchgroups));
+      OP1(SLJIT_MOV, STACK_TOP, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), POSSESSIVE1);
+      add_jump(compiler, &(BACKTRACK_AS(bracket_backtrack)->u.condfailed), CMP(SLJIT_C_EQUAL, SLJIT_SCRATCH_REG1, 0, SLJIT_IMM, 0));
+      matchingpath += 1 + IMM2_SIZE;
+      }
+    }
+  else
+    {
+    SLJIT_ASSERT(has_alternatives && *matchingpath >= OP_ASSERT && *matchingpath <= OP_ASSERTBACK_NOT);
+    /* Similar code as PUSH_BACKTRACK macro. */
+    assert = sljit_alloc_memory(compiler, sizeof(assert_backtrack));
+    if (SLJIT_UNLIKELY(sljit_get_compiler_error(compiler)))
+      return NULL;
+    memset(assert, 0, sizeof(assert_backtrack));
+    assert->common.cc = matchingpath;
+    BACKTRACK_AS(bracket_backtrack)->u.assert = assert;
+    matchingpath = compile_assert_matchingpath(common, matchingpath, assert, TRUE);
+    }
+  }
+
+compile_matchingpath(common, matchingpath, cc, backtrack);
+if (SLJIT_UNLIKELY(sljit_get_compiler_error(compiler)))
+  return NULL;
+
+if (opcode == OP_ONCE)
+  match_once_common(common, ket, BACKTRACK_AS(bracket_backtrack)->u.framesize, private_data_ptr, has_alternatives, needs_control_head);
+
+stacksize = 0;
+if (repeat_type == OP_MINUPTO)
+  {
+  /* We need to preserve the counter. TMP2 will be used below. */
+  OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), repeat_ptr);
+  stacksize++;
+  }
+if (ket != OP_KET || bra != OP_BRA)
+  stacksize++;
+if (offset != 0)
+  {
+  if (common->capture_last_ptr != 0)
+    stacksize++;
+  if (common->optimized_cbracket[offset >> 1] == 0)
+    stacksize += 2;
+  }
+if (has_alternatives && opcode != OP_ONCE)
+  stacksize++;
+
+if (stacksize > 0)
+  allocate_stack(common, stacksize);
+
+stacksize = 0;
+if (repeat_type == OP_MINUPTO)
+  {
+  /* TMP2 was set above. */
+  OP2(SLJIT_SUB, SLJIT_MEM1(STACK_TOP), STACK(stacksize), TMP2, 0, SLJIT_IMM, 1);
+  stacksize++;
+  }
+
+if (ket != OP_KET || bra != OP_BRA)
+  {
+  if (ket != OP_KET)
+    OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(stacksize), STR_PTR, 0);
+  else
+    OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(stacksize), SLJIT_IMM, 0);
+  stacksize++;
+  }
+
+if (offset != 0)
+  stacksize = match_capture_common(common, stacksize, offset, private_data_ptr);
+
+if (has_alternatives)
+  {
+  if (opcode != OP_ONCE)
+    OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(stacksize), SLJIT_IMM, 0);
+  if (ket != OP_KETRMAX)
+    BACKTRACK_AS(bracket_backtrack)->alternative_matchingpath = LABEL();
+  }
+
+/* Must be after the matchingpath label. */
+if (offset != 0 && common->optimized_cbracket[offset >> 1] != 0)
+  {
+  SLJIT_ASSERT(private_data_ptr == OVECTOR(offset + 0));
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(offset + 1), STR_PTR, 0);
+  }
+
+if (ket == OP_KETRMAX)
+  {
+  if (repeat_type != 0)
+    {
+    if (has_alternatives)
+      BACKTRACK_AS(bracket_backtrack)->alternative_matchingpath = LABEL();
+    OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_MEM1(SLJIT_LOCALS_REG), repeat_ptr, SLJIT_MEM1(SLJIT_LOCALS_REG), repeat_ptr, SLJIT_IMM, 1);
+    JUMPTO(SLJIT_C_NOT_ZERO, rmax_label);
+    /* Drop STR_PTR for greedy plus quantifier. */
+    if (opcode != OP_ONCE)
+      free_stack(common, 1);
+    }
+  else if (opcode == OP_ONCE || opcode >= OP_SBRA)
+    {
+    if (has_alternatives)
+      BACKTRACK_AS(bracket_backtrack)->alternative_matchingpath = LABEL();
+    /* Checking zero-length iteration. */
+    if (opcode != OP_ONCE)
+      {
+      CMPTO(SLJIT_C_NOT_EQUAL, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr, STR_PTR, 0, rmax_label);
+      /* Drop STR_PTR for greedy plus quantifier. */
+      if (bra != OP_BRAZERO)
+        free_stack(common, 1);
+      }
+    else
+      /* TMP2 must contain the starting STR_PTR. */
+      CMPTO(SLJIT_C_NOT_EQUAL, TMP2, 0, STR_PTR, 0, rmax_label);
+    }
+  else
+    JUMPTO(SLJIT_JUMP, rmax_label);
+  BACKTRACK_AS(bracket_backtrack)->recursive_matchingpath = LABEL();
+  }
+
+if (repeat_type == OP_EXACT)
+  {
+  OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_MEM1(SLJIT_LOCALS_REG), repeat_ptr, SLJIT_MEM1(SLJIT_LOCALS_REG), repeat_ptr, SLJIT_IMM, 1);
+  JUMPTO(SLJIT_C_NOT_ZERO, rmax_label);
+  }
+else if (repeat_type == OP_UPTO)
+  {
+  /* We need to preserve the counter. */
+  OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), repeat_ptr);
+  allocate_stack(common, 1);
+  OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(0), TMP2, 0);
+  }
+
+if (bra == OP_BRAZERO)
+  BACKTRACK_AS(bracket_backtrack)->zero_matchingpath = LABEL();
+
+if (bra == OP_BRAMINZERO)
+  {
+  /* This is a backtrack path! (From the viewpoint of OP_BRAMINZERO) */
+  JUMPTO(SLJIT_JUMP, ((braminzero_backtrack *)parent)->matchingpath);
+  if (braminzero != NULL)
+    {
+    JUMPHERE(braminzero);
+    /* We need to release the end pointer to perform the
+    backtrack for the zero-length iteration. When
+    framesize is < 0, OP_ONCE will do the release itself. */
+    if (opcode == OP_ONCE && BACKTRACK_AS(bracket_backtrack)->u.framesize >= 0)
+      {
+      OP1(SLJIT_MOV, STACK_TOP, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr);
+      add_jump(compiler, &common->revertframes, JUMP(SLJIT_FAST_CALL));
+      }
+    else if (ket == OP_KETRMIN && opcode != OP_ONCE)
+      free_stack(common, 1);
+    }
+  /* Continue to the normal backtrack. */
+  }
+
+if ((ket != OP_KET && bra != OP_BRAMINZERO) || bra == OP_BRAZERO)
+  count_match(common);
+
+/* Skip the other alternatives. */
+while (*cc == OP_ALT)
+  cc += GET(cc, 1);
+cc += 1 + LINK_SIZE;
+
+/* Temporarily encoding the needs_control_head in framesize. */
+if (opcode == OP_ONCE)
+  BACKTRACK_AS(bracket_backtrack)->u.framesize = (BACKTRACK_AS(bracket_backtrack)->u.framesize << 1) | (needs_control_head ? 1 : 0);
+return cc + repeat_length;
+}
+
+static pcre_uchar *compile_bracketpos_matchingpath(compiler_common *common, pcre_uchar *cc, backtrack_common *parent)
+{
+DEFINE_COMPILER;
+backtrack_common *backtrack;
+pcre_uchar opcode;
+int private_data_ptr;
+int cbraprivptr = 0;
+BOOL needs_control_head;
+int framesize;
+int stacksize;
+int offset = 0;
+BOOL zero = FALSE;
+pcre_uchar *ccbegin = NULL;
+int stack; /* Also contains the offset of control head. */
+struct sljit_label *loop = NULL;
+struct jump_list *emptymatch = NULL;
+
+PUSH_BACKTRACK(sizeof(bracketpos_backtrack), cc, NULL);
+if (*cc == OP_BRAPOSZERO)
+  {
+  zero = TRUE;
+  cc++;
+  }
+
+opcode = *cc;
+private_data_ptr = PRIVATE_DATA(cc);
+SLJIT_ASSERT(private_data_ptr != 0);
+BACKTRACK_AS(bracketpos_backtrack)->private_data_ptr = private_data_ptr;
+switch(opcode)
+  {
+  case OP_BRAPOS:
+  case OP_SBRAPOS:
+  ccbegin = cc + 1 + LINK_SIZE;
+  break;
+
+  case OP_CBRAPOS:
+  case OP_SCBRAPOS:
+  offset = GET2(cc, 1 + LINK_SIZE);
+  /* This case cannot be optimized in the same was as
+  normal capturing brackets. */
+  SLJIT_ASSERT(common->optimized_cbracket[offset] == 0);
+  cbraprivptr = OVECTOR_PRIV(offset);
+  offset <<= 1;
+  ccbegin = cc + 1 + LINK_SIZE + IMM2_SIZE;
+  break;
+
+  default:
+  SLJIT_ASSERT_STOP();
+  break;
+  }
+
+framesize = get_framesize(common, cc, NULL, FALSE, &needs_control_head);
+BACKTRACK_AS(bracketpos_backtrack)->framesize = framesize;
+if (framesize < 0)
+  {
+  if (offset != 0)
+    {
+    stacksize = 2;
+    if (common->capture_last_ptr != 0)
+      stacksize++;
+    }
+  else
+    stacksize = 1;
+
+  if (needs_control_head)
+    stacksize++;
+  if (!zero)
+    stacksize++;
+
+  BACKTRACK_AS(bracketpos_backtrack)->stacksize = stacksize;
+  allocate_stack(common, stacksize);
+  if (framesize == no_frame)
+    OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr, STACK_TOP, 0);
+
+  stack = 0;
+  if (offset != 0)
+    {
+    stack = 2;
+    OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(offset));
+    OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(offset + 1));
+    OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(0), TMP1, 0);
+    if (common->capture_last_ptr != 0)
+      OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), common->capture_last_ptr);
+    OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(1), TMP2, 0);
+    if (needs_control_head)
+      OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), common->control_head_ptr);
+    if (common->capture_last_ptr != 0)
+      {
+      OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(2), TMP1, 0);
+      stack = 3;
+      }
+    }
+  else
+    {
+    if (needs_control_head)
+      OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), common->control_head_ptr);
+    OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(0), STR_PTR, 0);
+    stack = 1;
+    }
+
+  if (needs_control_head)
+    stack++;
+  if (!zero)
+    OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(stack), SLJIT_IMM, 1);
+  if (needs_control_head)
+    {
+    stack--;
+    OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(stack), TMP2, 0);
+    }
+  }
+else
+  {
+  stacksize = framesize + 1;
+  if (!zero)
+    stacksize++;
+  if (needs_control_head)
+    stacksize++;
+  if (offset == 0)
+    stacksize++;
+  BACKTRACK_AS(bracketpos_backtrack)->stacksize = stacksize;
+
+  allocate_stack(common, stacksize);
+  OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr);
+  if (needs_control_head)
+    OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), common->control_head_ptr);
+  OP2(SLJIT_SUB, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr, STACK_TOP, 0, SLJIT_IMM, -STACK(stacksize - 1));
+
+  stack = 0;
+  if (!zero)
+    {
+    OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(0), SLJIT_IMM, 1);
+    stack = 1;
+    }
+  if (needs_control_head)
+    {
+    OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(stack), TMP2, 0);
+    stack++;
+    }
+  if (offset == 0)
+    {
+    OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(stack), STR_PTR, 0);
+    stack++;
+    }
+  OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(stack), TMP1, 0);
+  init_frame(common, cc, NULL, stacksize - 1, stacksize - framesize, FALSE);
+  stack -= 1 + (offset == 0);
+  }
+
+if (offset != 0)
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), cbraprivptr, STR_PTR, 0);
+
+loop = LABEL();
+while (*cc != OP_KETRPOS)
+  {
+  backtrack->top = NULL;
+  backtrack->topbacktracks = NULL;
+  cc += GET(cc, 1);
+
+  compile_matchingpath(common, ccbegin, cc, backtrack);
+  if (SLJIT_UNLIKELY(sljit_get_compiler_error(compiler)))
+    return NULL;
+
+  if (framesize < 0)
+    {
+    if (framesize == no_frame)
+      OP1(SLJIT_MOV, STACK_TOP, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr);
+
+    if (offset != 0)
+      {
+      OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), cbraprivptr);
+      OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(offset + 1), STR_PTR, 0);
+      OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), cbraprivptr, STR_PTR, 0);
+      if (common->capture_last_ptr != 0)
+        OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->capture_last_ptr, SLJIT_IMM, offset >> 1);
+      OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(offset), TMP1, 0);
+      }
+    else
+      {
+      if (opcode == OP_SBRAPOS)
+        OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(STACK_TOP), STACK(0));
+      OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(0), STR_PTR, 0);
+      }
+
+    if (opcode == OP_SBRAPOS || opcode == OP_SCBRAPOS)
+      add_jump(compiler, &emptymatch, CMP(SLJIT_C_EQUAL, TMP1, 0, STR_PTR, 0));
+
+    if (!zero)
+      OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(stacksize - 1), SLJIT_IMM, 0);
+    }
+  else
+    {
+    if (offset != 0)
+      {
+      OP2(SLJIT_ADD, STACK_TOP, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr, SLJIT_IMM, stacksize * sizeof(sljit_sw));
+      OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), cbraprivptr);
+      OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(offset + 1), STR_PTR, 0);
+      OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), cbraprivptr, STR_PTR, 0);
+      if (common->capture_last_ptr != 0)
+        OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->capture_last_ptr, SLJIT_IMM, offset >> 1);
+      OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(offset), TMP1, 0);
+      }
+    else
+      {
+      OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr);
+      OP2(SLJIT_ADD, STACK_TOP, 0, TMP2, 0, SLJIT_IMM, stacksize * sizeof(sljit_sw));
+      if (opcode == OP_SBRAPOS)
+        OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(TMP2), (framesize + 1) * sizeof(sljit_sw));
+      OP1(SLJIT_MOV, SLJIT_MEM1(TMP2), (framesize + 1) * sizeof(sljit_sw), STR_PTR, 0);
+      }
+
+    if (opcode == OP_SBRAPOS || opcode == OP_SCBRAPOS)
+      add_jump(compiler, &emptymatch, CMP(SLJIT_C_EQUAL, TMP1, 0, STR_PTR, 0));
+
+    if (!zero)
+      {
+      if (framesize < 0)
+        OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(stacksize - 1), SLJIT_IMM, 0);
+      else
+        OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(0), SLJIT_IMM, 0);
+      }
+    }
+
+  if (needs_control_head)
+    OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->control_head_ptr, SLJIT_MEM1(STACK_TOP), STACK(stack));
+
+  JUMPTO(SLJIT_JUMP, loop);
+  flush_stubs(common);
+
+  compile_backtrackingpath(common, backtrack->top);
+  if (SLJIT_UNLIKELY(sljit_get_compiler_error(compiler)))
+    return NULL;
+  set_jumps(backtrack->topbacktracks, LABEL());
+
+  if (framesize < 0)
+    {
+    if (offset != 0)
+      OP1(SLJIT_MOV, STR_PTR, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), cbraprivptr);
+    else
+      OP1(SLJIT_MOV, STR_PTR, 0, SLJIT_MEM1(STACK_TOP), STACK(0));
+    }
+  else
+    {
+    if (offset != 0)
+      {
+      /* Last alternative. */
+      if (*cc == OP_KETRPOS)
+        OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr);
+      OP1(SLJIT_MOV, STR_PTR, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), cbraprivptr);
+      }
+    else
+      {
+      OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr);
+      OP1(SLJIT_MOV, STR_PTR, 0, SLJIT_MEM1(TMP2), (framesize + 1) * sizeof(sljit_sw));
+      }
+    }
+
+  if (*cc == OP_KETRPOS)
+    break;
+  ccbegin = cc + 1 + LINK_SIZE;
+  }
+
+/* We don't have to restore the control head in case of a failed match. */
+
+backtrack->topbacktracks = NULL;
+if (!zero)
+  {
+  if (framesize < 0)
+    add_jump(compiler, &backtrack->topbacktracks, CMP(SLJIT_C_NOT_EQUAL, SLJIT_MEM1(STACK_TOP), STACK(stacksize - 1), SLJIT_IMM, 0));
+  else /* TMP2 is set to [private_data_ptr] above. */
+    add_jump(compiler, &backtrack->topbacktracks, CMP(SLJIT_C_NOT_EQUAL, SLJIT_MEM1(TMP2), (stacksize - 1) * sizeof(sljit_sw), SLJIT_IMM, 0));
+  }
+
+/* None of them matched. */
+set_jumps(emptymatch, LABEL());
+count_match(common);
+return cc + 1 + LINK_SIZE;
+}
+
+static SLJIT_INLINE pcre_uchar *get_iterator_parameters(compiler_common *common, pcre_uchar *cc, pcre_uchar *opcode, pcre_uchar *type, int *arg1, int *arg2, pcre_uchar **end)
+{
+int class_len;
+
+*opcode = *cc;
+if (*opcode >= OP_STAR && *opcode <= OP_POSUPTO)
+  {
+  cc++;
+  *type = OP_CHAR;
+  }
+else if (*opcode >= OP_STARI && *opcode <= OP_POSUPTOI)
+  {
+  cc++;
+  *type = OP_CHARI;
+  *opcode -= OP_STARI - OP_STAR;
+  }
+else if (*opcode >= OP_NOTSTAR && *opcode <= OP_NOTPOSUPTO)
+  {
+  cc++;
+  *type = OP_NOT;
+  *opcode -= OP_NOTSTAR - OP_STAR;
+  }
+else if (*opcode >= OP_NOTSTARI && *opcode <= OP_NOTPOSUPTOI)
+  {
+  cc++;
+  *type = OP_NOTI;
+  *opcode -= OP_NOTSTARI - OP_STAR;
+  }
+else if (*opcode >= OP_TYPESTAR && *opcode <= OP_TYPEPOSUPTO)
+  {
+  cc++;
+  *opcode -= OP_TYPESTAR - OP_STAR;
+  *type = 0;
+  }
+else
+  {
+  SLJIT_ASSERT(*opcode >= OP_CLASS || *opcode <= OP_XCLASS);
+  *type = *opcode;
+  cc++;
+  class_len = (*type < OP_XCLASS) ? (int)(1 + (32 / sizeof(pcre_uchar))) : GET(cc, 0);
+  *opcode = cc[class_len - 1];
+  if (*opcode >= OP_CRSTAR && *opcode <= OP_CRMINQUERY)
+    {
+    *opcode -= OP_CRSTAR - OP_STAR;
+    if (end != NULL)
+      *end = cc + class_len;
+    }
+  else
+    {
+    SLJIT_ASSERT(*opcode == OP_CRRANGE || *opcode == OP_CRMINRANGE);
+    *arg1 = GET2(cc, (class_len + IMM2_SIZE));
+    *arg2 = GET2(cc, class_len);
+
+    if (*arg2 == 0)
+      {
+      SLJIT_ASSERT(*arg1 != 0);
+      *opcode = (*opcode == OP_CRRANGE) ? OP_UPTO : OP_MINUPTO;
+      }
+    if (*arg1 == *arg2)
+      *opcode = OP_EXACT;
+
+    if (end != NULL)
+      *end = cc + class_len + 2 * IMM2_SIZE;
+    }
+  return cc;
+  }
+
+if (*opcode == OP_UPTO || *opcode == OP_MINUPTO || *opcode == OP_EXACT || *opcode == OP_POSUPTO)
+  {
+  *arg1 = GET2(cc, 0);
+  cc += IMM2_SIZE;
+  }
+
+if (*type == 0)
+  {
+  *type = *cc;
+  if (end != NULL)
+    *end = next_opcode(common, cc);
+  cc++;
+  return cc;
+  }
+
+if (end != NULL)
+  {
+  *end = cc + 1;
+#ifdef SUPPORT_UTF
+  if (common->utf && HAS_EXTRALEN(*cc)) *end += GET_EXTRALEN(*cc);
+#endif
+  }
+return cc;
+}
+
+static pcre_uchar *compile_iterator_matchingpath(compiler_common *common, pcre_uchar *cc, backtrack_common *parent)
+{
+DEFINE_COMPILER;
+backtrack_common *backtrack;
+pcre_uchar opcode;
+pcre_uchar type;
+int arg1 = -1, arg2 = -1;
+pcre_uchar* end;
+jump_list *nomatch = NULL;
+struct sljit_jump *jump = NULL;
+struct sljit_label *label;
+int private_data_ptr = PRIVATE_DATA(cc);
+int base = (private_data_ptr == 0) ? SLJIT_MEM1(STACK_TOP) : SLJIT_MEM1(SLJIT_LOCALS_REG);
+int offset0 = (private_data_ptr == 0) ? STACK(0) : private_data_ptr;
+int offset1 = (private_data_ptr == 0) ? STACK(1) : private_data_ptr + (int)sizeof(sljit_sw);
+int tmp_base, tmp_offset;
+
+PUSH_BACKTRACK(sizeof(iterator_backtrack), cc, NULL);
+
+cc = get_iterator_parameters(common, cc, &opcode, &type, &arg1, &arg2, &end);
+
+switch(type)
+  {
+  case OP_NOT_DIGIT:
+  case OP_DIGIT:
+  case OP_NOT_WHITESPACE:
+  case OP_WHITESPACE:
+  case OP_NOT_WORDCHAR:
+  case OP_WORDCHAR:
+  case OP_ANY:
+  case OP_ALLANY:
+  case OP_ANYBYTE:
+  case OP_ANYNL:
+  case OP_NOT_HSPACE:
+  case OP_HSPACE:
+  case OP_NOT_VSPACE:
+  case OP_VSPACE:
+  case OP_CHAR:
+  case OP_CHARI:
+  case OP_NOT:
+  case OP_NOTI:
+  case OP_CLASS:
+  case OP_NCLASS:
+  tmp_base = TMP3;
+  tmp_offset = 0;
+  break;
+
+  default:
+  SLJIT_ASSERT_STOP();
+  /* Fall through. */
+
+  case OP_EXTUNI:
+  case OP_XCLASS:
+  case OP_NOTPROP:
+  case OP_PROP:
+  tmp_base = SLJIT_MEM1(SLJIT_LOCALS_REG);
+  tmp_offset = POSSESSIVE0;
+  break;
+  }
+
+switch(opcode)
+  {
+  case OP_STAR:
+  case OP_PLUS:
+  case OP_UPTO:
+  case OP_CRRANGE:
+  if (type == OP_ANYNL || type == OP_EXTUNI)
+    {
+    SLJIT_ASSERT(private_data_ptr == 0);
+    if (opcode == OP_STAR || opcode == OP_UPTO)
+      {
+      allocate_stack(common, 2);
+      OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(0), STR_PTR, 0);
+      OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(1), SLJIT_IMM, 0);
+      }
+    else
+      {
+      allocate_stack(common, 1);
+      OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(0), SLJIT_IMM, 0);
+      }
+
+    if (opcode == OP_UPTO || opcode == OP_CRRANGE)
+      OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), POSSESSIVE0, SLJIT_IMM, 0);
+
+    label = LABEL();
+    compile_char1_matchingpath(common, type, cc, &backtrack->topbacktracks);
+    if (opcode == OP_UPTO || opcode == OP_CRRANGE)
+      {
+      OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), POSSESSIVE0);
+      OP2(SLJIT_ADD, TMP1, 0, TMP1, 0, SLJIT_IMM, 1);
+      if (opcode == OP_CRRANGE && arg2 > 0)
+        CMPTO(SLJIT_C_LESS, TMP1, 0, SLJIT_IMM, arg2, label);
+      if (opcode == OP_UPTO || (opcode == OP_CRRANGE && arg1 > 0))
+        jump = CMP(SLJIT_C_GREATER_EQUAL, TMP1, 0, SLJIT_IMM, arg1);
+      OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), POSSESSIVE0, TMP1, 0);
+      }
+
+    /* We cannot use TMP3 because of this allocate_stack. */
+    allocate_stack(common, 1);
+    OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(0), STR_PTR, 0);
+    JUMPTO(SLJIT_JUMP, label);
+    if (jump != NULL)
+      JUMPHERE(jump);
+    }
+  else
+    {
+    if (opcode == OP_PLUS)
+      compile_char1_matchingpath(common, type, cc, &backtrack->topbacktracks);
+    if (private_data_ptr == 0)
+      allocate_stack(common, 2);
+    OP1(SLJIT_MOV, base, offset0, STR_PTR, 0);
+    if (opcode <= OP_PLUS)
+      OP1(SLJIT_MOV, base, offset1, STR_PTR, 0);
+    else
+      OP1(SLJIT_MOV, base, offset1, SLJIT_IMM, 1);
+    label = LABEL();
+    compile_char1_matchingpath(common, type, cc, &nomatch);
+    OP1(SLJIT_MOV, base, offset0, STR_PTR, 0);
+    if (opcode <= OP_PLUS)
+      JUMPTO(SLJIT_JUMP, label);
+    else if (opcode == OP_CRRANGE && arg1 == 0)
+      {
+      OP2(SLJIT_ADD, base, offset1, base, offset1, SLJIT_IMM, 1);
+      JUMPTO(SLJIT_JUMP, label);
+      }
+    else
+      {
+      OP1(SLJIT_MOV, TMP1, 0, base, offset1);
+      OP2(SLJIT_ADD, TMP1, 0, TMP1, 0, SLJIT_IMM, 1);
+      OP1(SLJIT_MOV, base, offset1, TMP1, 0);
+      CMPTO(SLJIT_C_LESS, TMP1, 0, SLJIT_IMM, arg1 + 1, label);
+      }
+    set_jumps(nomatch, LABEL());
+    if (opcode == OP_CRRANGE)
+      add_jump(compiler, &backtrack->topbacktracks, CMP(SLJIT_C_LESS, base, offset1, SLJIT_IMM, arg2 + 1));
+    OP1(SLJIT_MOV, STR_PTR, 0, base, offset0);
+    }
+  BACKTRACK_AS(iterator_backtrack)->matchingpath = LABEL();
+  break;
+
+  case OP_MINSTAR:
+  case OP_MINPLUS:
+  if (opcode == OP_MINPLUS)
+    compile_char1_matchingpath(common, type, cc, &backtrack->topbacktracks);
+  if (private_data_ptr == 0)
+    allocate_stack(common, 1);
+  OP1(SLJIT_MOV, base, offset0, STR_PTR, 0);
+  BACKTRACK_AS(iterator_backtrack)->matchingpath = LABEL();
+  break;
+
+  case OP_MINUPTO:
+  case OP_CRMINRANGE:
+  if (private_data_ptr == 0)
+    allocate_stack(common, 2);
+  OP1(SLJIT_MOV, base, offset0, STR_PTR, 0);
+  OP1(SLJIT_MOV, base, offset1, SLJIT_IMM, 1);
+  if (opcode == OP_CRMINRANGE)
+    add_jump(compiler, &backtrack->topbacktracks, JUMP(SLJIT_JUMP));
+  BACKTRACK_AS(iterator_backtrack)->matchingpath = LABEL();
+  break;
+
+  case OP_QUERY:
+  case OP_MINQUERY:
+  if (private_data_ptr == 0)
+    allocate_stack(common, 1);
+  OP1(SLJIT_MOV, base, offset0, STR_PTR, 0);
+  if (opcode == OP_QUERY)
+    compile_char1_matchingpath(common, type, cc, &backtrack->topbacktracks);
+  BACKTRACK_AS(iterator_backtrack)->matchingpath = LABEL();
+  break;
+
+  case OP_EXACT:
+  OP1(SLJIT_MOV, tmp_base, tmp_offset, SLJIT_IMM, arg1);
+  label = LABEL();
+  compile_char1_matchingpath(common, type, cc, &backtrack->topbacktracks);
+  OP2(SLJIT_SUB | SLJIT_SET_E, tmp_base, tmp_offset, tmp_base, tmp_offset, SLJIT_IMM, 1);
+  JUMPTO(SLJIT_C_NOT_ZERO, label);
+  break;
+
+  case OP_POSSTAR:
+  case OP_POSPLUS:
+  case OP_POSUPTO:
+  if (opcode == OP_POSPLUS)
+    compile_char1_matchingpath(common, type, cc, &backtrack->topbacktracks);
+  if (opcode == OP_POSUPTO)
+    OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), POSSESSIVE1, SLJIT_IMM, arg1);
+  OP1(SLJIT_MOV, tmp_base, tmp_offset, STR_PTR, 0);
+  label = LABEL();
+  compile_char1_matchingpath(common, type, cc, &nomatch);
+  OP1(SLJIT_MOV, tmp_base, tmp_offset, STR_PTR, 0);
+  if (opcode != OP_POSUPTO)
+    JUMPTO(SLJIT_JUMP, label);
+  else
+    {
+    OP2(SLJIT_SUB | SLJIT_SET_E, SLJIT_MEM1(SLJIT_LOCALS_REG), POSSESSIVE1, SLJIT_MEM1(SLJIT_LOCALS_REG), POSSESSIVE1, SLJIT_IMM, 1);
+    JUMPTO(SLJIT_C_NOT_ZERO, label);
+    }
+  set_jumps(nomatch, LABEL());
+  OP1(SLJIT_MOV, STR_PTR, 0, tmp_base, tmp_offset);
+  break;
+
+  case OP_POSQUERY:
+  OP1(SLJIT_MOV, tmp_base, tmp_offset, STR_PTR, 0);
+  compile_char1_matchingpath(common, type, cc, &nomatch);
+  OP1(SLJIT_MOV, tmp_base, tmp_offset, STR_PTR, 0);
+  set_jumps(nomatch, LABEL());
+  OP1(SLJIT_MOV, STR_PTR, 0, tmp_base, tmp_offset);
+  break;
+
+  default:
+  SLJIT_ASSERT_STOP();
+  break;
+  }
+
+count_match(common);
+return end;
+}
+
+static SLJIT_INLINE pcre_uchar *compile_fail_accept_matchingpath(compiler_common *common, pcre_uchar *cc, backtrack_common *parent)
+{
+DEFINE_COMPILER;
+backtrack_common *backtrack;
+
+PUSH_BACKTRACK(sizeof(backtrack_common), cc, NULL);
+
+if (*cc == OP_FAIL)
+  {
+  add_jump(compiler, &backtrack->topbacktracks, JUMP(SLJIT_JUMP));
+  return cc + 1;
+  }
+
+if (*cc == OP_ASSERT_ACCEPT || common->currententry != NULL)
+  {
+  /* No need to check notempty conditions. */
+  if (common->accept_label == NULL)
+    add_jump(compiler, &common->accept, JUMP(SLJIT_JUMP));
+  else
+    JUMPTO(SLJIT_JUMP, common->accept_label);
+  return cc + 1;
+  }
+
+if (common->accept_label == NULL)
+  add_jump(compiler, &common->accept, CMP(SLJIT_C_NOT_EQUAL, STR_PTR, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(0)));
+else
+  CMPTO(SLJIT_C_NOT_EQUAL, STR_PTR, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(0), common->accept_label);
+OP1(SLJIT_MOV, TMP1, 0, ARGUMENTS, 0);
+OP1(SLJIT_MOV_UB, TMP2, 0, SLJIT_MEM1(TMP1), SLJIT_OFFSETOF(jit_arguments, notempty));
+add_jump(compiler, &backtrack->topbacktracks, CMP(SLJIT_C_NOT_EQUAL, TMP2, 0, SLJIT_IMM, 0));
+OP1(SLJIT_MOV_UB, TMP2, 0, SLJIT_MEM1(TMP1), SLJIT_OFFSETOF(jit_arguments, notempty_atstart));
+if (common->accept_label == NULL)
+  add_jump(compiler, &common->accept, CMP(SLJIT_C_EQUAL, TMP2, 0, SLJIT_IMM, 0));
+else
+  CMPTO(SLJIT_C_EQUAL, TMP2, 0, SLJIT_IMM, 0, common->accept_label);
+OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(TMP1), SLJIT_OFFSETOF(jit_arguments, str));
+if (common->accept_label == NULL)
+  add_jump(compiler, &common->accept, CMP(SLJIT_C_NOT_EQUAL, TMP2, 0, STR_PTR, 0));
+else
+  CMPTO(SLJIT_C_NOT_EQUAL, TMP2, 0, STR_PTR, 0, common->accept_label);
+add_jump(compiler, &backtrack->topbacktracks, JUMP(SLJIT_JUMP));
+return cc + 1;
+}
+
+static SLJIT_INLINE pcre_uchar *compile_close_matchingpath(compiler_common *common, pcre_uchar *cc)
+{
+DEFINE_COMPILER;
+int offset = GET2(cc, 1);
+BOOL optimized_cbracket = common->optimized_cbracket[offset] != 0;
+
+/* Data will be discarded anyway... */
+if (common->currententry != NULL)
+  return cc + 1 + IMM2_SIZE;
+
+if (!optimized_cbracket)
+  OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR_PRIV(offset));
+offset <<= 1;
+OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(offset + 1), STR_PTR, 0);
+if (!optimized_cbracket)
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(offset), TMP1, 0);
+return cc + 1 + IMM2_SIZE;
+}
+
+static SLJIT_INLINE pcre_uchar *compile_control_verb_matchingpath(compiler_common *common, pcre_uchar *cc, backtrack_common *parent)
+{
+DEFINE_COMPILER;
+backtrack_common *backtrack;
+pcre_uchar opcode = *cc;
+pcre_uchar *ccend = cc + 1;
+
+if (opcode == OP_PRUNE_ARG || opcode == OP_SKIP_ARG || opcode == OP_THEN_ARG)
+  ccend += 2 + cc[1];
+
+PUSH_BACKTRACK(sizeof(backtrack_common), cc, NULL);
+
+if (opcode == OP_SKIP)
+  {
+  allocate_stack(common, 1);
+  OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(0), STR_PTR, 0);
+  return ccend;
+  }
+
+if (opcode == OP_PRUNE_ARG || opcode == OP_THEN_ARG)
+  {
+  OP1(SLJIT_MOV, TMP1, 0, ARGUMENTS, 0);
+  OP1(SLJIT_MOV, TMP2, 0, SLJIT_IMM, (sljit_sw)(cc + 2));
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->mark_ptr, TMP2, 0);
+  OP1(SLJIT_MOV, SLJIT_MEM1(TMP1), SLJIT_OFFSETOF(jit_arguments, mark_ptr), TMP2, 0);
+  }
+
+return ccend;
+}
+
+static pcre_uchar then_trap_opcode[1] = { OP_THEN_TRAP };
+
+static SLJIT_INLINE void compile_then_trap_matchingpath(compiler_common *common, pcre_uchar *cc, pcre_uchar *ccend, backtrack_common *parent)
+{
+DEFINE_COMPILER;
+backtrack_common *backtrack;
+BOOL needs_control_head;
+int size;
+
+PUSH_BACKTRACK_NOVALUE(sizeof(then_trap_backtrack), cc);
+common->then_trap = BACKTRACK_AS(then_trap_backtrack);
+BACKTRACK_AS(then_trap_backtrack)->common.cc = then_trap_opcode;
+BACKTRACK_AS(then_trap_backtrack)->start = (sljit_sw)(cc - common->start);
+BACKTRACK_AS(then_trap_backtrack)->framesize = get_framesize(common, cc, ccend, FALSE, &needs_control_head);
+
+size = BACKTRACK_AS(then_trap_backtrack)->framesize;
+size = 3 + (size < 0 ? 0 : size);
+
+OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), common->control_head_ptr);
+allocate_stack(common, size);
+if (size > 3)
+  OP2(SLJIT_SUB, SLJIT_MEM1(SLJIT_LOCALS_REG), common->control_head_ptr, STACK_TOP, 0, SLJIT_IMM, (size - 3) * sizeof(sljit_sw));
+else
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->control_head_ptr, STACK_TOP, 0);
+OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(size - 1), SLJIT_IMM, BACKTRACK_AS(then_trap_backtrack)->start);
+OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(size - 2), SLJIT_IMM, type_then_trap);
+OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(size - 3), TMP2, 0);
+
+size = BACKTRACK_AS(then_trap_backtrack)->framesize;
+if (size >= 0)
+  init_frame(common, cc, ccend, size - 1, 0, FALSE);
+}
+
+static void compile_matchingpath(compiler_common *common, pcre_uchar *cc, pcre_uchar *ccend, backtrack_common *parent)
+{
+DEFINE_COMPILER;
+backtrack_common *backtrack;
+BOOL has_then_trap = FALSE;
+then_trap_backtrack *save_then_trap = NULL;
+
+SLJIT_ASSERT(*ccend == OP_END || (*ccend >= OP_ALT && *ccend <= OP_KETRPOS));
+
+if (common->has_then && common->then_offsets[cc - common->start] != 0)
+  {
+  SLJIT_ASSERT(*ccend != OP_END && common->control_head_ptr != 0);
+  has_then_trap = TRUE;
+  save_then_trap = common->then_trap;
+  /* Tail item on backtrack. */
+  compile_then_trap_matchingpath(common, cc, ccend, parent);
+  }
+
+while (cc < ccend)
+  {
+  switch(*cc)
+    {
+    case OP_SOD:
+    case OP_SOM:
+    case OP_NOT_WORD_BOUNDARY:
+    case OP_WORD_BOUNDARY:
+    case OP_NOT_DIGIT:
+    case OP_DIGIT:
+    case OP_NOT_WHITESPACE:
+    case OP_WHITESPACE:
+    case OP_NOT_WORDCHAR:
+    case OP_WORDCHAR:
+    case OP_ANY:
+    case OP_ALLANY:
+    case OP_ANYBYTE:
+    case OP_NOTPROP:
+    case OP_PROP:
+    case OP_ANYNL:
+    case OP_NOT_HSPACE:
+    case OP_HSPACE:
+    case OP_NOT_VSPACE:
+    case OP_VSPACE:
+    case OP_EXTUNI:
+    case OP_EODN:
+    case OP_EOD:
+    case OP_CIRC:
+    case OP_CIRCM:
+    case OP_DOLL:
+    case OP_DOLLM:
+    case OP_NOT:
+    case OP_NOTI:
+    case OP_REVERSE:
+    cc = compile_char1_matchingpath(common, *cc, cc + 1, parent->top != NULL ? &parent->top->nextbacktracks : &parent->topbacktracks);
+    break;
+
+    case OP_SET_SOM:
+    PUSH_BACKTRACK_NOVALUE(sizeof(backtrack_common), cc);
+    OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(0));
+    allocate_stack(common, 1);
+    OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(0), STR_PTR, 0);
+    OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(0), TMP2, 0);
+    cc++;
+    break;
+
+    case OP_CHAR:
+    case OP_CHARI:
+    if (common->mode == JIT_COMPILE)
+      cc = compile_charn_matchingpath(common, cc, ccend, parent->top != NULL ? &parent->top->nextbacktracks : &parent->topbacktracks);
+    else
+      cc = compile_char1_matchingpath(common, *cc, cc + 1, parent->top != NULL ? &parent->top->nextbacktracks : &parent->topbacktracks);
+    break;
+
+    case OP_STAR:
+    case OP_MINSTAR:
+    case OP_PLUS:
+    case OP_MINPLUS:
+    case OP_QUERY:
+    case OP_MINQUERY:
+    case OP_UPTO:
+    case OP_MINUPTO:
+    case OP_EXACT:
+    case OP_POSSTAR:
+    case OP_POSPLUS:
+    case OP_POSQUERY:
+    case OP_POSUPTO:
+    case OP_STARI:
+    case OP_MINSTARI:
+    case OP_PLUSI:
+    case OP_MINPLUSI:
+    case OP_QUERYI:
+    case OP_MINQUERYI:
+    case OP_UPTOI:
+    case OP_MINUPTOI:
+    case OP_EXACTI:
+    case OP_POSSTARI:
+    case OP_POSPLUSI:
+    case OP_POSQUERYI:
+    case OP_POSUPTOI:
+    case OP_NOTSTAR:
+    case OP_NOTMINSTAR:
+    case OP_NOTPLUS:
+    case OP_NOTMINPLUS:
+    case OP_NOTQUERY:
+    case OP_NOTMINQUERY:
+    case OP_NOTUPTO:
+    case OP_NOTMINUPTO:
+    case OP_NOTEXACT:
+    case OP_NOTPOSSTAR:
+    case OP_NOTPOSPLUS:
+    case OP_NOTPOSQUERY:
+    case OP_NOTPOSUPTO:
+    case OP_NOTSTARI:
+    case OP_NOTMINSTARI:
+    case OP_NOTPLUSI:
+    case OP_NOTMINPLUSI:
+    case OP_NOTQUERYI:
+    case OP_NOTMINQUERYI:
+    case OP_NOTUPTOI:
+    case OP_NOTMINUPTOI:
+    case OP_NOTEXACTI:
+    case OP_NOTPOSSTARI:
+    case OP_NOTPOSPLUSI:
+    case OP_NOTPOSQUERYI:
+    case OP_NOTPOSUPTOI:
+    case OP_TYPESTAR:
+    case OP_TYPEMINSTAR:
+    case OP_TYPEPLUS:
+    case OP_TYPEMINPLUS:
+    case OP_TYPEQUERY:
+    case OP_TYPEMINQUERY:
+    case OP_TYPEUPTO:
+    case OP_TYPEMINUPTO:
+    case OP_TYPEEXACT:
+    case OP_TYPEPOSSTAR:
+    case OP_TYPEPOSPLUS:
+    case OP_TYPEPOSQUERY:
+    case OP_TYPEPOSUPTO:
+    cc = compile_iterator_matchingpath(common, cc, parent);
+    break;
+
+    case OP_CLASS:
+    case OP_NCLASS:
+    if (cc[1 + (32 / sizeof(pcre_uchar))] >= OP_CRSTAR && cc[1 + (32 / sizeof(pcre_uchar))] <= OP_CRMINRANGE)
+      cc = compile_iterator_matchingpath(common, cc, parent);
+    else
+      cc = compile_char1_matchingpath(common, *cc, cc + 1, parent->top != NULL ? &parent->top->nextbacktracks : &parent->topbacktracks);
+    break;
+
+#if defined SUPPORT_UTF || defined COMPILE_PCRE16 || defined COMPILE_PCRE32
+    case OP_XCLASS:
+    if (*(cc + GET(cc, 1)) >= OP_CRSTAR && *(cc + GET(cc, 1)) <= OP_CRMINRANGE)
+      cc = compile_iterator_matchingpath(common, cc, parent);
+    else
+      cc = compile_char1_matchingpath(common, *cc, cc + 1, parent->top != NULL ? &parent->top->nextbacktracks : &parent->topbacktracks);
+    break;
+#endif
+
+    case OP_REF:
+    case OP_REFI:
+    if (cc[1 + IMM2_SIZE] >= OP_CRSTAR && cc[1 + IMM2_SIZE] <= OP_CRMINRANGE)
+      cc = compile_ref_iterator_matchingpath(common, cc, parent);
+    else
+      cc = compile_ref_matchingpath(common, cc, parent->top != NULL ? &parent->top->nextbacktracks : &parent->topbacktracks, TRUE, FALSE);
+    break;
+
+    case OP_RECURSE:
+    cc = compile_recurse_matchingpath(common, cc, parent);
+    break;
+
+    case OP_CALLOUT:
+    cc = compile_callout_matchingpath(common, cc, parent);
+    break;
+
+    case OP_ASSERT:
+    case OP_ASSERT_NOT:
+    case OP_ASSERTBACK:
+    case OP_ASSERTBACK_NOT:
+    PUSH_BACKTRACK_NOVALUE(sizeof(assert_backtrack), cc);
+    cc = compile_assert_matchingpath(common, cc, BACKTRACK_AS(assert_backtrack), FALSE);
+    break;
+
+    case OP_BRAMINZERO:
+    PUSH_BACKTRACK_NOVALUE(sizeof(braminzero_backtrack), cc);
+    cc = bracketend(cc + 1);
+    if (*(cc - 1 - LINK_SIZE) != OP_KETRMIN)
+      {
+      allocate_stack(common, 1);
+      OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(0), STR_PTR, 0);
+      }
+    else
+      {
+      allocate_stack(common, 2);
+      OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(0), SLJIT_IMM, 0);
+      OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(1), STR_PTR, 0);
+      }
+    BACKTRACK_AS(braminzero_backtrack)->matchingpath = LABEL();
+    if (cc[1] > OP_ASSERTBACK_NOT)
+      count_match(common);
+    break;
+
+    case OP_ONCE:
+    case OP_ONCE_NC:
+    case OP_BRA:
+    case OP_CBRA:
+    case OP_COND:
+    case OP_SBRA:
+    case OP_SCBRA:
+    case OP_SCOND:
+    cc = compile_bracket_matchingpath(common, cc, parent);
+    break;
+
+    case OP_BRAZERO:
+    if (cc[1] > OP_ASSERTBACK_NOT)
+      cc = compile_bracket_matchingpath(common, cc, parent);
+    else
+      {
+      PUSH_BACKTRACK_NOVALUE(sizeof(assert_backtrack), cc);
+      cc = compile_assert_matchingpath(common, cc, BACKTRACK_AS(assert_backtrack), FALSE);
+      }
+    break;
+
+    case OP_BRAPOS:
+    case OP_CBRAPOS:
+    case OP_SBRAPOS:
+    case OP_SCBRAPOS:
+    case OP_BRAPOSZERO:
+    cc = compile_bracketpos_matchingpath(common, cc, parent);
+    break;
+
+    case OP_MARK:
+    PUSH_BACKTRACK_NOVALUE(sizeof(backtrack_common), cc);
+    SLJIT_ASSERT(common->mark_ptr != 0);
+    OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), common->mark_ptr);
+    allocate_stack(common, common->has_skip_arg ? 5 : 1);
+    OP1(SLJIT_MOV, TMP1, 0, ARGUMENTS, 0);
+    OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(common->has_skip_arg ? 4 : 0), TMP2, 0);
+    OP1(SLJIT_MOV, TMP2, 0, SLJIT_IMM, (sljit_sw)(cc + 2));
+    OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->mark_ptr, TMP2, 0);
+    OP1(SLJIT_MOV, SLJIT_MEM1(TMP1), SLJIT_OFFSETOF(jit_arguments, mark_ptr), TMP2, 0);
+    if (common->has_skip_arg)
+      {
+      OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), common->control_head_ptr);
+      OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->control_head_ptr, STACK_TOP, 0);
+      OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(1), SLJIT_IMM, type_mark);
+      OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(2), SLJIT_IMM, (sljit_sw)(cc + 2));
+      OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(3), STR_PTR, 0);
+      OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(0), TMP1, 0);
+      }
+    cc += 1 + 2 + cc[1];
+    break;
+
+    case OP_PRUNE:
+    case OP_PRUNE_ARG:
+    case OP_SKIP:
+    case OP_SKIP_ARG:
+    case OP_THEN:
+    case OP_THEN_ARG:
+    case OP_COMMIT:
+    cc = compile_control_verb_matchingpath(common, cc, parent);
+    break;
+
+    case OP_FAIL:
+    case OP_ACCEPT:
+    case OP_ASSERT_ACCEPT:
+    cc = compile_fail_accept_matchingpath(common, cc, parent);
+    break;
+
+    case OP_CLOSE:
+    cc = compile_close_matchingpath(common, cc);
+    break;
+
+    case OP_SKIPZERO:
+    cc = bracketend(cc + 1);
+    break;
+
+    default:
+    SLJIT_ASSERT_STOP();
+    return;
+    }
+  if (cc == NULL)
+    return;
+  }
+
+if (has_then_trap)
+  {
+  /* Head item on backtrack. */
+  PUSH_BACKTRACK_NOVALUE(sizeof(then_trap_backtrack), cc);
+  BACKTRACK_AS(then_trap_backtrack)->common.cc = then_trap_opcode;
+  BACKTRACK_AS(then_trap_backtrack)->then_trap = common->then_trap;
+  common->then_trap = save_then_trap;
+  }
+SLJIT_ASSERT(cc == ccend);
+}
+
+#undef PUSH_BACKTRACK
+#undef PUSH_BACKTRACK_NOVALUE
+#undef BACKTRACK_AS
+
+#define COMPILE_BACKTRACKINGPATH(current) \
+  do \
+    { \
+    compile_backtrackingpath(common, (current)); \
+    if (SLJIT_UNLIKELY(sljit_get_compiler_error(compiler))) \
+      return; \
+    } \
+  while (0)
+
+#define CURRENT_AS(type) ((type *)current)
+
+static void compile_iterator_backtrackingpath(compiler_common *common, struct backtrack_common *current)
+{
+DEFINE_COMPILER;
+pcre_uchar *cc = current->cc;
+pcre_uchar opcode;
+pcre_uchar type;
+int arg1 = -1, arg2 = -1;
+struct sljit_label *label = NULL;
+struct sljit_jump *jump = NULL;
+jump_list *jumplist = NULL;
+int private_data_ptr = PRIVATE_DATA(cc);
+int base = (private_data_ptr == 0) ? SLJIT_MEM1(STACK_TOP) : SLJIT_MEM1(SLJIT_LOCALS_REG);
+int offset0 = (private_data_ptr == 0) ? STACK(0) : private_data_ptr;
+int offset1 = (private_data_ptr == 0) ? STACK(1) : private_data_ptr + (int)sizeof(sljit_sw);
+
+cc = get_iterator_parameters(common, cc, &opcode, &type, &arg1, &arg2, NULL);
+
+switch(opcode)
+  {
+  case OP_STAR:
+  case OP_PLUS:
+  case OP_UPTO:
+  case OP_CRRANGE:
+  if (type == OP_ANYNL || type == OP_EXTUNI)
+    {
+    SLJIT_ASSERT(private_data_ptr == 0);
+    set_jumps(current->topbacktracks, LABEL());
+    OP1(SLJIT_MOV, STR_PTR, 0, SLJIT_MEM1(STACK_TOP), STACK(0));
+    free_stack(common, 1);
+    CMPTO(SLJIT_C_NOT_EQUAL, STR_PTR, 0, SLJIT_IMM, 0, CURRENT_AS(iterator_backtrack)->matchingpath);
+    }
+  else
+    {
+    if (opcode == OP_UPTO)
+      arg2 = 0;
+    if (opcode <= OP_PLUS)
+      {
+      OP1(SLJIT_MOV, STR_PTR, 0, base, offset0);
+      jump = CMP(SLJIT_C_LESS_EQUAL, STR_PTR, 0, base, offset1);
+      }
+    else
+      {
+      OP1(SLJIT_MOV, TMP1, 0, base, offset1);
+      OP1(SLJIT_MOV, STR_PTR, 0, base, offset0);
+      jump = CMP(SLJIT_C_LESS_EQUAL, TMP1, 0, SLJIT_IMM, arg2 + 1);
+      OP2(SLJIT_SUB, base, offset1, TMP1, 0, SLJIT_IMM, 1);
+      }
+    skip_char_back(common);
+    OP1(SLJIT_MOV, base, offset0, STR_PTR, 0);
+    JUMPTO(SLJIT_JUMP, CURRENT_AS(iterator_backtrack)->matchingpath);
+    if (opcode == OP_CRRANGE)
+      set_jumps(current->topbacktracks, LABEL());
+    JUMPHERE(jump);
+    if (private_data_ptr == 0)
+      free_stack(common, 2);
+    if (opcode == OP_PLUS)
+      set_jumps(current->topbacktracks, LABEL());
+    }
+  break;
+
+  case OP_MINSTAR:
+  case OP_MINPLUS:
+  OP1(SLJIT_MOV, STR_PTR, 0, base, offset0);
+  compile_char1_matchingpath(common, type, cc, &jumplist);
+  OP1(SLJIT_MOV, base, offset0, STR_PTR, 0);
+  JUMPTO(SLJIT_JUMP, CURRENT_AS(iterator_backtrack)->matchingpath);
+  set_jumps(jumplist, LABEL());
+  if (private_data_ptr == 0)
+    free_stack(common, 1);
+  if (opcode == OP_MINPLUS)
+    set_jumps(current->topbacktracks, LABEL());
+  break;
+
+  case OP_MINUPTO:
+  case OP_CRMINRANGE:
+  if (opcode == OP_CRMINRANGE)
+    {
+    label = LABEL();
+    set_jumps(current->topbacktracks, label);
+    }
+  OP1(SLJIT_MOV, STR_PTR, 0, base, offset0);
+  compile_char1_matchingpath(common, type, cc, &jumplist);
+
+  OP1(SLJIT_MOV, TMP1, 0, base, offset1);
+  OP1(SLJIT_MOV, base, offset0, STR_PTR, 0);
+  OP2(SLJIT_ADD, TMP1, 0, TMP1, 0, SLJIT_IMM, 1);
+  OP1(SLJIT_MOV, base, offset1, TMP1, 0);
+
+  if (opcode == OP_CRMINRANGE)
+    CMPTO(SLJIT_C_LESS, TMP1, 0, SLJIT_IMM, arg2 + 1, label);
+
+  if (opcode == OP_CRMINRANGE && arg1 == 0)
+    JUMPTO(SLJIT_JUMP, CURRENT_AS(iterator_backtrack)->matchingpath);
+  else
+    CMPTO(SLJIT_C_LESS, TMP1, 0, SLJIT_IMM, arg1 + 2, CURRENT_AS(iterator_backtrack)->matchingpath);
+
+  set_jumps(jumplist, LABEL());
+  if (private_data_ptr == 0)
+    free_stack(common, 2);
+  break;
+
+  case OP_QUERY:
+  OP1(SLJIT_MOV, STR_PTR, 0, base, offset0);
+  OP1(SLJIT_MOV, base, offset0, SLJIT_IMM, 0);
+  CMPTO(SLJIT_C_NOT_EQUAL, STR_PTR, 0, SLJIT_IMM, 0, CURRENT_AS(iterator_backtrack)->matchingpath);
+  jump = JUMP(SLJIT_JUMP);
+  set_jumps(current->topbacktracks, LABEL());
+  OP1(SLJIT_MOV, STR_PTR, 0, base, offset0);
+  OP1(SLJIT_MOV, base, offset0, SLJIT_IMM, 0);
+  JUMPTO(SLJIT_JUMP, CURRENT_AS(iterator_backtrack)->matchingpath);
+  JUMPHERE(jump);
+  if (private_data_ptr == 0)
+    free_stack(common, 1);
+  break;
+
+  case OP_MINQUERY:
+  OP1(SLJIT_MOV, STR_PTR, 0, base, offset0);
+  OP1(SLJIT_MOV, base, offset0, SLJIT_IMM, 0);
+  jump = CMP(SLJIT_C_EQUAL, STR_PTR, 0, SLJIT_IMM, 0);
+  compile_char1_matchingpath(common, type, cc, &jumplist);
+  JUMPTO(SLJIT_JUMP, CURRENT_AS(iterator_backtrack)->matchingpath);
+  set_jumps(jumplist, LABEL());
+  JUMPHERE(jump);
+  if (private_data_ptr == 0)
+    free_stack(common, 1);
+  break;
+
+  case OP_EXACT:
+  case OP_POSPLUS:
+  set_jumps(current->topbacktracks, LABEL());
+  break;
+
+  case OP_POSSTAR:
+  case OP_POSQUERY:
+  case OP_POSUPTO:
+  break;
+
+  default:
+  SLJIT_ASSERT_STOP();
+  break;
+  }
+}
+
+static SLJIT_INLINE void compile_ref_iterator_backtrackingpath(compiler_common *common, struct backtrack_common *current)
+{
+DEFINE_COMPILER;
+pcre_uchar *cc = current->cc;
+pcre_uchar type;
+
+type = cc[1 + IMM2_SIZE];
+if ((type & 0x1) == 0)
+  {
+  set_jumps(current->topbacktracks, LABEL());
+  OP1(SLJIT_MOV, STR_PTR, 0, SLJIT_MEM1(STACK_TOP), STACK(0));
+  free_stack(common, 1);
+  CMPTO(SLJIT_C_NOT_EQUAL, STR_PTR, 0, SLJIT_IMM, 0, CURRENT_AS(iterator_backtrack)->matchingpath);
+  return;
+  }
+
+OP1(SLJIT_MOV, STR_PTR, 0, SLJIT_MEM1(STACK_TOP), STACK(0));
+CMPTO(SLJIT_C_NOT_EQUAL, STR_PTR, 0, SLJIT_IMM, 0, CURRENT_AS(iterator_backtrack)->matchingpath);
+set_jumps(current->topbacktracks, LABEL());
+free_stack(common, 2);
+}
+
+static SLJIT_INLINE void compile_recurse_backtrackingpath(compiler_common *common, struct backtrack_common *current)
+{
+DEFINE_COMPILER;
+
+if (CURRENT_AS(recurse_backtrack)->inlined_pattern)
+  compile_backtrackingpath(common, current->top);
+set_jumps(current->topbacktracks, LABEL());
+if (CURRENT_AS(recurse_backtrack)->inlined_pattern)
+  return;
+
+if (common->has_set_som && common->mark_ptr != 0)
+  {
+  OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(STACK_TOP), STACK(0));
+  OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(STACK_TOP), STACK(1));
+  free_stack(common, 2);
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(0), TMP2, 0);
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->mark_ptr, TMP1, 0);
+  }
+else if (common->has_set_som || common->mark_ptr != 0)
+  {
+  OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(STACK_TOP), STACK(0));
+  free_stack(common, 1);
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->has_set_som ? (int)(OVECTOR(0)) : common->mark_ptr, TMP2, 0);
+  }
+}
+
+static void compile_assert_backtrackingpath(compiler_common *common, struct backtrack_common *current)
+{
+DEFINE_COMPILER;
+pcre_uchar *cc = current->cc;
+pcre_uchar bra = OP_BRA;
+struct sljit_jump *brajump = NULL;
+
+SLJIT_ASSERT(*cc != OP_BRAMINZERO);
+if (*cc == OP_BRAZERO)
+  {
+  bra = *cc;
+  cc++;
+  }
+
+if (bra == OP_BRAZERO)
+  {
+  SLJIT_ASSERT(current->topbacktracks == NULL);
+  OP1(SLJIT_MOV, STR_PTR, 0, SLJIT_MEM1(STACK_TOP), STACK(0));
+  }
+
+if (CURRENT_AS(assert_backtrack)->framesize < 0)
+  {
+  set_jumps(current->topbacktracks, LABEL());
+
+  if (bra == OP_BRAZERO)
+    {
+    OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(0), SLJIT_IMM, 0);
+    CMPTO(SLJIT_C_NOT_EQUAL, STR_PTR, 0, SLJIT_IMM, 0, CURRENT_AS(assert_backtrack)->matchingpath);
+    free_stack(common, 1);
+    }
+  return;
+  }
+
+if (bra == OP_BRAZERO)
+  {
+  if (*cc == OP_ASSERT_NOT || *cc == OP_ASSERTBACK_NOT)
+    {
+    OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(0), SLJIT_IMM, 0);
+    CMPTO(SLJIT_C_NOT_EQUAL, STR_PTR, 0, SLJIT_IMM, 0, CURRENT_AS(assert_backtrack)->matchingpath);
+    free_stack(common, 1);
+    return;
+    }
+  free_stack(common, 1);
+  brajump = CMP(SLJIT_C_EQUAL, STR_PTR, 0, SLJIT_IMM, 0);
+  }
+
+if (*cc == OP_ASSERT || *cc == OP_ASSERTBACK)
+  {
+  OP1(SLJIT_MOV, STACK_TOP, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), CURRENT_AS(assert_backtrack)->private_data_ptr);
+  add_jump(compiler, &common->revertframes, JUMP(SLJIT_FAST_CALL));
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), CURRENT_AS(assert_backtrack)->private_data_ptr, SLJIT_MEM1(STACK_TOP), CURRENT_AS(assert_backtrack)->framesize * sizeof(sljit_sw));
+
+  set_jumps(current->topbacktracks, LABEL());
+  }
+else
+  set_jumps(current->topbacktracks, LABEL());
+
+if (bra == OP_BRAZERO)
+  {
+  /* We know there is enough place on the stack. */
+  OP2(SLJIT_ADD, STACK_TOP, 0, STACK_TOP, 0, SLJIT_IMM, sizeof(sljit_sw));
+  OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(0), SLJIT_IMM, 0);
+  JUMPTO(SLJIT_JUMP, CURRENT_AS(assert_backtrack)->matchingpath);
+  JUMPHERE(brajump);
+  }
+}
+
+static void compile_bracket_backtrackingpath(compiler_common *common, struct backtrack_common *current)
+{
+DEFINE_COMPILER;
+int opcode, stacksize, count;
+int offset = 0;
+int private_data_ptr = CURRENT_AS(bracket_backtrack)->private_data_ptr;
+int repeat_ptr = 0, repeat_type = 0, repeat_count = 0;
+pcre_uchar *cc = current->cc;
+pcre_uchar *ccbegin;
+pcre_uchar *ccprev;
+jump_list *jumplist = NULL;
+jump_list *jumplistitem = NULL;
+pcre_uchar bra = OP_BRA;
+pcre_uchar ket;
+assert_backtrack *assert;
+BOOL has_alternatives;
+BOOL needs_control_head = FALSE;
+struct sljit_jump *brazero = NULL;
+struct sljit_jump *once = NULL;
+struct sljit_jump *cond = NULL;
+struct sljit_label *rmin_label = NULL;
+struct sljit_label *exact_label = NULL;
+
+if (*cc == OP_BRAZERO || *cc == OP_BRAMINZERO)
+  {
+  bra = *cc;
+  cc++;
+  }
+
+opcode = *cc;
+ccbegin = bracketend(cc) - 1 - LINK_SIZE;
+ket = *ccbegin;
+if (ket == OP_KET && PRIVATE_DATA(ccbegin) != 0)
+  {
+  repeat_ptr = PRIVATE_DATA(ccbegin);
+  repeat_type = PRIVATE_DATA(ccbegin + 2);
+  repeat_count = PRIVATE_DATA(ccbegin + 3);
+  SLJIT_ASSERT(repeat_type != 0 && repeat_count != 0);
+  if (repeat_type == OP_UPTO)
+    ket = OP_KETRMAX;
+  if (repeat_type == OP_MINUPTO)
+    ket = OP_KETRMIN;
+  }
+ccbegin = cc;
+cc += GET(cc, 1);
+has_alternatives = *cc == OP_ALT;
+if (SLJIT_UNLIKELY(opcode == OP_COND) || SLJIT_UNLIKELY(opcode == OP_SCOND))
+  has_alternatives = (ccbegin[1 + LINK_SIZE] >= OP_ASSERT && ccbegin[1 + LINK_SIZE] <= OP_ASSERTBACK_NOT) || CURRENT_AS(bracket_backtrack)->u.condfailed != NULL;
+if (opcode == OP_CBRA || opcode == OP_SCBRA)
+  offset = (GET2(ccbegin, 1 + LINK_SIZE)) << 1;
+if (SLJIT_UNLIKELY(opcode == OP_COND) && (*cc == OP_KETRMAX || *cc == OP_KETRMIN))
+  opcode = OP_SCOND;
+if (SLJIT_UNLIKELY(opcode == OP_ONCE_NC))
+  opcode = OP_ONCE;
+
+/* Decoding the needs_control_head in framesize. */
+if (opcode == OP_ONCE)
+  {
+  needs_control_head = (CURRENT_AS(bracket_backtrack)->u.framesize & 0x1) != 0;
+  CURRENT_AS(bracket_backtrack)->u.framesize >>= 1;
+  }
+
+if (ket != OP_KET && repeat_type != 0)
+  {
+  /* TMP1 is used in OP_KETRMIN below. */
+  OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(STACK_TOP), STACK(0));
+  free_stack(common, 1);
+  if (repeat_type == OP_UPTO)
+    OP2(SLJIT_ADD, SLJIT_MEM1(SLJIT_LOCALS_REG), repeat_ptr, TMP1, 0, SLJIT_IMM, 1);
+  else
+    OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), repeat_ptr, TMP1, 0);
+  }
+
+if (ket == OP_KETRMAX)
+  {
+  if (bra == OP_BRAZERO)
+    {
+    OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(STACK_TOP), STACK(0));
+    free_stack(common, 1);
+    brazero = CMP(SLJIT_C_EQUAL, TMP1, 0, SLJIT_IMM, 0);
+    }
+  }
+else if (ket == OP_KETRMIN)
+  {
+  if (bra != OP_BRAMINZERO)
+    {
+    OP1(SLJIT_MOV, STR_PTR, 0, SLJIT_MEM1(STACK_TOP), STACK(0));
+    if (repeat_type != 0)
+      {
+      /* TMP1 was set a few lines above. */
+      CMPTO(SLJIT_C_NOT_EQUAL, TMP1, 0, SLJIT_IMM, 0, CURRENT_AS(bracket_backtrack)->recursive_matchingpath);
+      /* Drop STR_PTR for non-greedy plus quantifier. */
+      if (opcode != OP_ONCE)
+        free_stack(common, 1);
+      }
+    else if (opcode >= OP_SBRA || opcode == OP_ONCE)
+      {
+      /* Checking zero-length iteration. */
+      if (opcode != OP_ONCE || CURRENT_AS(bracket_backtrack)->u.framesize < 0)
+        CMPTO(SLJIT_C_NOT_EQUAL, STR_PTR, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr, CURRENT_AS(bracket_backtrack)->recursive_matchingpath);
+      else
+        {
+        OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr);
+        CMPTO(SLJIT_C_NOT_EQUAL, STR_PTR, 0, SLJIT_MEM1(TMP1), (CURRENT_AS(bracket_backtrack)->u.framesize + 1) * sizeof(sljit_sw), CURRENT_AS(bracket_backtrack)->recursive_matchingpath);
+        }
+      /* Drop STR_PTR for non-greedy plus quantifier. */
+      if (opcode != OP_ONCE)
+        free_stack(common, 1);
+      }
+    else
+      JUMPTO(SLJIT_JUMP, CURRENT_AS(bracket_backtrack)->recursive_matchingpath);
+    }
+  rmin_label = LABEL();
+  if (repeat_type != 0)
+    OP2(SLJIT_ADD, SLJIT_MEM1(SLJIT_LOCALS_REG), repeat_ptr, SLJIT_MEM1(SLJIT_LOCALS_REG), repeat_ptr, SLJIT_IMM, 1);
+  }
+else if (bra == OP_BRAZERO)
+  {
+  OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(STACK_TOP), STACK(0));
+  free_stack(common, 1);
+  brazero = CMP(SLJIT_C_NOT_EQUAL, TMP1, 0, SLJIT_IMM, 0);
+  }
+else if (repeat_type == OP_EXACT)
+  {
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), repeat_ptr, SLJIT_IMM, 1);
+  exact_label = LABEL();
+  }
+
+if (offset != 0)
+  {
+  if (common->capture_last_ptr != 0)
+    {
+    SLJIT_ASSERT(common->optimized_cbracket[offset >> 1] == 0);
+    OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(STACK_TOP), STACK(0));
+    OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(STACK_TOP), STACK(1));
+    OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->capture_last_ptr, TMP1, 0);
+    OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(STACK_TOP), STACK(2));
+    free_stack(common, 3);
+    OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(offset), TMP2, 0);
+    OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(offset + 1), TMP1, 0);
+    }
+  else if (common->optimized_cbracket[offset >> 1] == 0)
+    {
+    OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(STACK_TOP), STACK(0));
+    OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(STACK_TOP), STACK(1));
+    free_stack(common, 2);
+    OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(offset), TMP1, 0);
+    OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(offset + 1), TMP2, 0);
+    }
+  }
+
+if (SLJIT_UNLIKELY(opcode == OP_ONCE))
+  {
+  if (CURRENT_AS(bracket_backtrack)->u.framesize >= 0)
+    {
+    OP1(SLJIT_MOV, STACK_TOP, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr);
+    add_jump(compiler, &common->revertframes, JUMP(SLJIT_FAST_CALL));
+    }
+  once = JUMP(SLJIT_JUMP);
+  }
+else if (SLJIT_UNLIKELY(opcode == OP_COND) || SLJIT_UNLIKELY(opcode == OP_SCOND))
+  {
+  if (has_alternatives)
+    {
+    /* Always exactly one alternative. */
+    OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(STACK_TOP), STACK(0));
+    free_stack(common, 1);
+
+    jumplistitem = sljit_alloc_memory(compiler, sizeof(jump_list));
+    if (SLJIT_UNLIKELY(!jumplistitem))
+      return;
+    jumplist = jumplistitem;
+    jumplistitem->next = NULL;
+    jumplistitem->jump = CMP(SLJIT_C_EQUAL, TMP1, 0, SLJIT_IMM, 1);
+    }
+  }
+else if (*cc == OP_ALT)
+  {
+  /* Build a jump list. Get the last successfully matched branch index. */
+  OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(STACK_TOP), STACK(0));
+  free_stack(common, 1);
+  count = 1;
+  do
+    {
+    /* Append as the last item. */
+    if (jumplist != NULL)
+      {
+      jumplistitem->next = sljit_alloc_memory(compiler, sizeof(jump_list));
+      jumplistitem = jumplistitem->next;
+      }
+    else
+      {
+      jumplistitem = sljit_alloc_memory(compiler, sizeof(jump_list));
+      jumplist = jumplistitem;
+      }
+
+    if (SLJIT_UNLIKELY(!jumplistitem))
+      return;
+
+    jumplistitem->next = NULL;
+    jumplistitem->jump = CMP(SLJIT_C_EQUAL, TMP1, 0, SLJIT_IMM, count++);
+    cc += GET(cc, 1);
+    }
+  while (*cc == OP_ALT);
+
+  cc = ccbegin + GET(ccbegin, 1);
+  }
+
+COMPILE_BACKTRACKINGPATH(current->top);
+if (current->topbacktracks)
+  set_jumps(current->topbacktracks, LABEL());
+
+if (SLJIT_UNLIKELY(opcode == OP_COND) || SLJIT_UNLIKELY(opcode == OP_SCOND))
+  {
+  /* Conditional block always has at most one alternative. */
+  if (ccbegin[1 + LINK_SIZE] >= OP_ASSERT && ccbegin[1 + LINK_SIZE] <= OP_ASSERTBACK_NOT)
+    {
+    SLJIT_ASSERT(has_alternatives);
+    assert = CURRENT_AS(bracket_backtrack)->u.assert;
+    if (assert->framesize >= 0 && (ccbegin[1 + LINK_SIZE] == OP_ASSERT || ccbegin[1 + LINK_SIZE] == OP_ASSERTBACK))
+      {
+      OP1(SLJIT_MOV, STACK_TOP, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), assert->private_data_ptr);
+      add_jump(compiler, &common->revertframes, JUMP(SLJIT_FAST_CALL));
+      OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), assert->private_data_ptr, SLJIT_MEM1(STACK_TOP), assert->framesize * sizeof(sljit_sw));
+      }
+    cond = JUMP(SLJIT_JUMP);
+    set_jumps(CURRENT_AS(bracket_backtrack)->u.assert->condfailed, LABEL());
+    }
+  else if (CURRENT_AS(bracket_backtrack)->u.condfailed != NULL)
+    {
+    SLJIT_ASSERT(has_alternatives);
+    cond = JUMP(SLJIT_JUMP);
+    set_jumps(CURRENT_AS(bracket_backtrack)->u.condfailed, LABEL());
+    }
+  else
+    SLJIT_ASSERT(!has_alternatives);
+  }
+
+if (has_alternatives)
+  {
+  count = 1;
+  do
+    {
+    current->top = NULL;
+    current->topbacktracks = NULL;
+    current->nextbacktracks = NULL;
+    /* Conditional blocks always have an additional alternative, even if it is empty. */
+    if (*cc == OP_ALT)
+      {
+      ccprev = cc + 1 + LINK_SIZE;
+      cc += GET(cc, 1);
+      if (opcode != OP_COND && opcode != OP_SCOND)
+        {
+        if (opcode != OP_ONCE)
+          {
+          if (private_data_ptr != 0)
+            OP1(SLJIT_MOV, STR_PTR, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr);
+          else
+            OP1(SLJIT_MOV, STR_PTR, 0, SLJIT_MEM1(STACK_TOP), STACK(0));
+          }
+        else
+          OP1(SLJIT_MOV, STR_PTR, 0, SLJIT_MEM1(STACK_TOP), STACK(needs_control_head ? 1 : 0));
+        }
+      compile_matchingpath(common, ccprev, cc, current);
+      if (SLJIT_UNLIKELY(sljit_get_compiler_error(compiler)))
+        return;
+      }
+
+    /* Instructions after the current alternative is successfully matched. */
+    /* There is a similar code in compile_bracket_matchingpath. */
+    if (opcode == OP_ONCE)
+      match_once_common(common, ket, CURRENT_AS(bracket_backtrack)->u.framesize, private_data_ptr, has_alternatives, needs_control_head);
+
+    stacksize = 0;
+    if (repeat_type == OP_MINUPTO)
+      {
+      /* We need to preserve the counter. TMP2 will be used below. */
+      OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), repeat_ptr);
+      stacksize++;
+      }
+    if (ket != OP_KET || bra != OP_BRA)
+      stacksize++;
+    if (offset != 0)
+      {
+      if (common->capture_last_ptr != 0)
+        stacksize++;
+      if (common->optimized_cbracket[offset >> 1] == 0)
+        stacksize += 2;
+      }
+    if (opcode != OP_ONCE)
+      stacksize++;
+
+    if (stacksize > 0)
+      allocate_stack(common, stacksize);
+
+    stacksize = 0;
+    if (repeat_type == OP_MINUPTO)
+      {
+      /* TMP2 was set above. */
+      OP2(SLJIT_SUB, SLJIT_MEM1(STACK_TOP), STACK(stacksize), TMP2, 0, SLJIT_IMM, 1);
+      stacksize++;
+      }
+
+    if (ket != OP_KET || bra != OP_BRA)
+      {
+      if (ket != OP_KET)
+        OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(stacksize), STR_PTR, 0);
+      else
+        OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(stacksize), SLJIT_IMM, 0);
+      stacksize++;
+      }
+
+    if (offset != 0)
+      stacksize = match_capture_common(common, stacksize, offset, private_data_ptr);
+
+    if (opcode != OP_ONCE)
+      OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(stacksize), SLJIT_IMM, count++);
+
+    if (offset != 0 && ket == OP_KETRMAX && common->optimized_cbracket[offset >> 1] != 0)
+      {
+      /* If ket is not OP_KETRMAX, this code path is executed after the jump to alternative_matchingpath. */
+      SLJIT_ASSERT(private_data_ptr == OVECTOR(offset + 0));
+      OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(offset + 1), STR_PTR, 0);
+      }
+
+    JUMPTO(SLJIT_JUMP, CURRENT_AS(bracket_backtrack)->alternative_matchingpath);
+
+    if (opcode != OP_ONCE)
+      {
+      SLJIT_ASSERT(jumplist);
+      JUMPHERE(jumplist->jump);
+      jumplist = jumplist->next;
+      }
+
+    COMPILE_BACKTRACKINGPATH(current->top);
+    if (current->topbacktracks)
+      set_jumps(current->topbacktracks, LABEL());
+    SLJIT_ASSERT(!current->nextbacktracks);
+    }
+  while (*cc == OP_ALT);
+  SLJIT_ASSERT(!jumplist);
+
+  if (cond != NULL)
+    {
+    SLJIT_ASSERT(opcode == OP_COND || opcode == OP_SCOND);
+    assert = CURRENT_AS(bracket_backtrack)->u.assert;
+    if ((ccbegin[1 + LINK_SIZE] == OP_ASSERT_NOT || ccbegin[1 + LINK_SIZE] == OP_ASSERTBACK_NOT) && assert->framesize >= 0)
+      {
+      OP1(SLJIT_MOV, STACK_TOP, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), assert->private_data_ptr);
+      add_jump(compiler, &common->revertframes, JUMP(SLJIT_FAST_CALL));
+      OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), assert->private_data_ptr, SLJIT_MEM1(STACK_TOP), assert->framesize * sizeof(sljit_sw));
+      }
+    JUMPHERE(cond);
+    }
+
+  /* Free the STR_PTR. */
+  if (private_data_ptr == 0)
+    free_stack(common, 1);
+  }
+
+if (offset != 0)
+  {
+  /* Using both tmp register is better for instruction scheduling. */
+  if (common->optimized_cbracket[offset >> 1] != 0)
+    {
+    OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(STACK_TOP), STACK(0));
+    OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(STACK_TOP), STACK(1));
+    free_stack(common, 2);
+    OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(offset), TMP1, 0);
+    OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(offset + 1), TMP2, 0);
+    }
+  else
+    {
+    OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(STACK_TOP), STACK(0));
+    free_stack(common, 1);
+    OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr, TMP1, 0);
+    }
+  }
+else if (opcode == OP_SBRA || opcode == OP_SCOND)
+  {
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr, SLJIT_MEM1(STACK_TOP), STACK(0));
+  free_stack(common, 1);
+  }
+else if (opcode == OP_ONCE)
+  {
+  cc = ccbegin + GET(ccbegin, 1);
+  stacksize = needs_control_head ? 1 : 0;
+
+  if (CURRENT_AS(bracket_backtrack)->u.framesize >= 0)
+    {
+    /* Reset head and drop saved frame. */
+    stacksize += CURRENT_AS(bracket_backtrack)->u.framesize + ((ket != OP_KET || *cc == OP_ALT) ? 2 : 1);
+    }
+  else if (ket == OP_KETRMAX || (*cc == OP_ALT && ket != OP_KETRMIN))
+    {
+    /* The STR_PTR must be released. */
+    stacksize++;
+    }
+  free_stack(common, stacksize);
+
+  JUMPHERE(once);
+  /* Restore previous private_data_ptr */
+  if (CURRENT_AS(bracket_backtrack)->u.framesize >= 0)
+    OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr, SLJIT_MEM1(STACK_TOP), CURRENT_AS(bracket_backtrack)->u.framesize * sizeof(sljit_sw));
+  else if (ket == OP_KETRMIN)
+    {
+    OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(STACK_TOP), STACK(1));
+    /* See the comment below. */
+    free_stack(common, 2);
+    OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), private_data_ptr, TMP1, 0);
+    }
+  }
+
+if (repeat_type == OP_EXACT)
+  {
+  OP2(SLJIT_ADD, TMP1, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), repeat_ptr, SLJIT_IMM, 1);
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), repeat_ptr, TMP1, 0);
+  CMPTO(SLJIT_C_LESS_EQUAL, TMP1, 0, SLJIT_IMM, repeat_count, exact_label);
+  }
+else if (ket == OP_KETRMAX)
+  {
+  OP1(SLJIT_MOV, STR_PTR, 0, SLJIT_MEM1(STACK_TOP), STACK(0));
+  if (bra != OP_BRAZERO)
+    free_stack(common, 1);
+
+  CMPTO(SLJIT_C_NOT_EQUAL, STR_PTR, 0, SLJIT_IMM, 0, CURRENT_AS(bracket_backtrack)->recursive_matchingpath);
+  if (bra == OP_BRAZERO)
+    {
+    OP1(SLJIT_MOV, STR_PTR, 0, SLJIT_MEM1(STACK_TOP), STACK(1));
+    JUMPTO(SLJIT_JUMP, CURRENT_AS(bracket_backtrack)->zero_matchingpath);
+    JUMPHERE(brazero);
+    free_stack(common, 1);
+    }
+  }
+else if (ket == OP_KETRMIN)
+  {
+  OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(STACK_TOP), STACK(0));
+
+  /* OP_ONCE removes everything in case of a backtrack, so we don't
+  need to explicitly release the STR_PTR. The extra release would
+  affect badly the free_stack(2) above. */
+  if (opcode != OP_ONCE)
+    free_stack(common, 1);
+  CMPTO(SLJIT_C_NOT_EQUAL, TMP1, 0, SLJIT_IMM, 0, rmin_label);
+  if (opcode == OP_ONCE)
+    free_stack(common, bra == OP_BRAMINZERO ? 2 : 1);
+  else if (bra == OP_BRAMINZERO)
+    free_stack(common, 1);
+  }
+else if (bra == OP_BRAZERO)
+  {
+  OP1(SLJIT_MOV, STR_PTR, 0, SLJIT_MEM1(STACK_TOP), STACK(0));
+  JUMPTO(SLJIT_JUMP, CURRENT_AS(bracket_backtrack)->zero_matchingpath);
+  JUMPHERE(brazero);
+  }
+}
+
+static SLJIT_INLINE void compile_bracketpos_backtrackingpath(compiler_common *common, struct backtrack_common *current)
+{
+DEFINE_COMPILER;
+int offset;
+struct sljit_jump *jump;
+
+if (CURRENT_AS(bracketpos_backtrack)->framesize < 0)
+  {
+  if (*current->cc == OP_CBRAPOS || *current->cc == OP_SCBRAPOS)
+    {
+    offset = (GET2(current->cc, 1 + LINK_SIZE)) << 1;
+    OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(STACK_TOP), STACK(0));
+    OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(STACK_TOP), STACK(1));
+    OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(offset), TMP1, 0);
+    if (common->capture_last_ptr != 0)
+      OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(STACK_TOP), STACK(2));
+    OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(offset + 1), TMP2, 0);
+    if (common->capture_last_ptr != 0)
+      OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->capture_last_ptr, TMP1, 0);
+    }
+  set_jumps(current->topbacktracks, LABEL());
+  free_stack(common, CURRENT_AS(bracketpos_backtrack)->stacksize);
+  return;
+  }
+
+OP1(SLJIT_MOV, STACK_TOP, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), CURRENT_AS(bracketpos_backtrack)->private_data_ptr);
+add_jump(compiler, &common->revertframes, JUMP(SLJIT_FAST_CALL));
+
+if (current->topbacktracks)
+  {
+  jump = JUMP(SLJIT_JUMP);
+  set_jumps(current->topbacktracks, LABEL());
+  /* Drop the stack frame. */
+  free_stack(common, CURRENT_AS(bracketpos_backtrack)->stacksize);
+  JUMPHERE(jump);
+  }
+OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), CURRENT_AS(bracketpos_backtrack)->private_data_ptr, SLJIT_MEM1(STACK_TOP), CURRENT_AS(bracketpos_backtrack)->framesize * sizeof(sljit_sw));
+}
+
+static SLJIT_INLINE void compile_braminzero_backtrackingpath(compiler_common *common, struct backtrack_common *current)
+{
+assert_backtrack backtrack;
+
+current->top = NULL;
+current->topbacktracks = NULL;
+current->nextbacktracks = NULL;
+if (current->cc[1] > OP_ASSERTBACK_NOT)
+  {
+  /* Manual call of compile_bracket_matchingpath and compile_bracket_backtrackingpath. */
+  compile_bracket_matchingpath(common, current->cc, current);
+  compile_bracket_backtrackingpath(common, current->top);
+  }
+else
+  {
+  memset(&backtrack, 0, sizeof(backtrack));
+  backtrack.common.cc = current->cc;
+  backtrack.matchingpath = CURRENT_AS(braminzero_backtrack)->matchingpath;
+  /* Manual call of compile_assert_matchingpath. */
+  compile_assert_matchingpath(common, current->cc, &backtrack, FALSE);
+  }
+SLJIT_ASSERT(!current->nextbacktracks && !current->topbacktracks);
+}
+
+static SLJIT_INLINE void compile_control_verb_backtrackingpath(compiler_common *common, struct backtrack_common *current)
+{
+DEFINE_COMPILER;
+pcre_uchar opcode = *current->cc;
+struct sljit_label *loop;
+struct sljit_jump *jump;
+
+if (opcode == OP_THEN || opcode == OP_THEN_ARG)
+  {
+  if (common->then_trap != NULL)
+    {
+    SLJIT_ASSERT(common->control_head_ptr != 0);
+
+    OP1(SLJIT_MOV, STACK_TOP, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), common->control_head_ptr);
+    OP1(SLJIT_MOV, TMP1, 0, SLJIT_IMM, type_then_trap);
+    OP1(SLJIT_MOV, TMP2, 0, SLJIT_IMM, common->then_trap->start);
+    jump = JUMP(SLJIT_JUMP);
+
+    loop = LABEL();
+    OP1(SLJIT_MOV, STACK_TOP, 0, SLJIT_MEM1(STACK_TOP), -(int)sizeof(sljit_sw));
+    JUMPHERE(jump);
+    CMPTO(SLJIT_C_NOT_EQUAL, SLJIT_MEM1(STACK_TOP), -(int)(2 * sizeof(sljit_sw)), TMP1, 0, loop);
+    CMPTO(SLJIT_C_NOT_EQUAL, SLJIT_MEM1(STACK_TOP), -(int)(3 * sizeof(sljit_sw)), TMP2, 0, loop);
+    add_jump(compiler, &common->then_trap->quit, JUMP(SLJIT_JUMP));
+    return;
+    }
+  else if (common->positive_assert)
+    {
+    add_jump(compiler, &common->positive_assert_quit, JUMP(SLJIT_JUMP));
+    return;
+    }
+  }
+
+if (common->local_exit)
+  {
+  if (common->quit_label == NULL)
+    add_jump(compiler, &common->quit, JUMP(SLJIT_JUMP));
+  else
+    JUMPTO(SLJIT_JUMP, common->quit_label);
+  return;
+  }
+
+if (opcode == OP_SKIP_ARG)
+  {
+  SLJIT_ASSERT(common->control_head_ptr != 0);
+  OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), common->control_head_ptr);
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), LOCALS0, STACK_TOP, 0);
+  OP1(SLJIT_MOV, STACK_TOP, 0, SLJIT_IMM, (sljit_sw)(current->cc + 2));
+  sljit_emit_ijump(compiler, SLJIT_CALL2, SLJIT_IMM, SLJIT_FUNC_OFFSET(do_search_mark));
+  OP1(SLJIT_MOV, STACK_TOP, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), LOCALS0);
+
+  OP1(SLJIT_MOV, STR_PTR, 0, TMP1, 0);
+  add_jump(compiler, &common->reset_match, CMP(SLJIT_C_NOT_EQUAL, STR_PTR, 0, SLJIT_IMM, -1));
+  return;
+  }
+
+if (opcode == OP_SKIP)
+  OP1(SLJIT_MOV, STR_PTR, 0, SLJIT_MEM1(STACK_TOP), STACK(0));
+else
+  OP1(SLJIT_MOV, STR_PTR, 0, SLJIT_IMM, 0);
+add_jump(compiler, &common->reset_match, JUMP(SLJIT_JUMP));
+}
+
+static SLJIT_INLINE void compile_then_trap_backtrackingpath(compiler_common *common, struct backtrack_common *current)
+{
+DEFINE_COMPILER;
+struct sljit_jump *jump;
+int size;
+
+if (CURRENT_AS(then_trap_backtrack)->then_trap)
+  {
+  common->then_trap = CURRENT_AS(then_trap_backtrack)->then_trap;
+  return;
+  }
+
+size = CURRENT_AS(then_trap_backtrack)->framesize;
+size = 3 + (size < 0 ? 0 : size);
+
+OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(STACK_TOP), STACK(size - 3));
+free_stack(common, size);
+jump = JUMP(SLJIT_JUMP);
+
+set_jumps(CURRENT_AS(then_trap_backtrack)->quit, LABEL());
+/* STACK_TOP is set by THEN. */
+if (CURRENT_AS(then_trap_backtrack)->framesize >= 0)
+  add_jump(compiler, &common->revertframes, JUMP(SLJIT_FAST_CALL));
+OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(STACK_TOP), STACK(0));
+free_stack(common, 3);
+
+JUMPHERE(jump);
+OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->control_head_ptr, TMP1, 0);
+}
+
+static void compile_backtrackingpath(compiler_common *common, struct backtrack_common *current)
+{
+DEFINE_COMPILER;
+then_trap_backtrack *save_then_trap = common->then_trap;
+
+while (current)
+  {
+  if (current->nextbacktracks != NULL)
+    set_jumps(current->nextbacktracks, LABEL());
+  switch(*current->cc)
+    {
+    case OP_SET_SOM:
+    OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(STACK_TOP), STACK(0));
+    free_stack(common, 1);
+    OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(0), TMP1, 0);
+    break;
+
+    case OP_STAR:
+    case OP_MINSTAR:
+    case OP_PLUS:
+    case OP_MINPLUS:
+    case OP_QUERY:
+    case OP_MINQUERY:
+    case OP_UPTO:
+    case OP_MINUPTO:
+    case OP_EXACT:
+    case OP_POSSTAR:
+    case OP_POSPLUS:
+    case OP_POSQUERY:
+    case OP_POSUPTO:
+    case OP_STARI:
+    case OP_MINSTARI:
+    case OP_PLUSI:
+    case OP_MINPLUSI:
+    case OP_QUERYI:
+    case OP_MINQUERYI:
+    case OP_UPTOI:
+    case OP_MINUPTOI:
+    case OP_EXACTI:
+    case OP_POSSTARI:
+    case OP_POSPLUSI:
+    case OP_POSQUERYI:
+    case OP_POSUPTOI:
+    case OP_NOTSTAR:
+    case OP_NOTMINSTAR:
+    case OP_NOTPLUS:
+    case OP_NOTMINPLUS:
+    case OP_NOTQUERY:
+    case OP_NOTMINQUERY:
+    case OP_NOTUPTO:
+    case OP_NOTMINUPTO:
+    case OP_NOTEXACT:
+    case OP_NOTPOSSTAR:
+    case OP_NOTPOSPLUS:
+    case OP_NOTPOSQUERY:
+    case OP_NOTPOSUPTO:
+    case OP_NOTSTARI:
+    case OP_NOTMINSTARI:
+    case OP_NOTPLUSI:
+    case OP_NOTMINPLUSI:
+    case OP_NOTQUERYI:
+    case OP_NOTMINQUERYI:
+    case OP_NOTUPTOI:
+    case OP_NOTMINUPTOI:
+    case OP_NOTEXACTI:
+    case OP_NOTPOSSTARI:
+    case OP_NOTPOSPLUSI:
+    case OP_NOTPOSQUERYI:
+    case OP_NOTPOSUPTOI:
+    case OP_TYPESTAR:
+    case OP_TYPEMINSTAR:
+    case OP_TYPEPLUS:
+    case OP_TYPEMINPLUS:
+    case OP_TYPEQUERY:
+    case OP_TYPEMINQUERY:
+    case OP_TYPEUPTO:
+    case OP_TYPEMINUPTO:
+    case OP_TYPEEXACT:
+    case OP_TYPEPOSSTAR:
+    case OP_TYPEPOSPLUS:
+    case OP_TYPEPOSQUERY:
+    case OP_TYPEPOSUPTO:
+    case OP_CLASS:
+    case OP_NCLASS:
+#if defined SUPPORT_UTF || !defined COMPILE_PCRE8
+    case OP_XCLASS:
+#endif
+    compile_iterator_backtrackingpath(common, current);
+    break;
+
+    case OP_REF:
+    case OP_REFI:
+    compile_ref_iterator_backtrackingpath(common, current);
+    break;
+
+    case OP_RECURSE:
+    compile_recurse_backtrackingpath(common, current);
+    break;
+
+    case OP_ASSERT:
+    case OP_ASSERT_NOT:
+    case OP_ASSERTBACK:
+    case OP_ASSERTBACK_NOT:
+    compile_assert_backtrackingpath(common, current);
+    break;
+
+    case OP_ONCE:
+    case OP_ONCE_NC:
+    case OP_BRA:
+    case OP_CBRA:
+    case OP_COND:
+    case OP_SBRA:
+    case OP_SCBRA:
+    case OP_SCOND:
+    compile_bracket_backtrackingpath(common, current);
+    break;
+
+    case OP_BRAZERO:
+    if (current->cc[1] > OP_ASSERTBACK_NOT)
+      compile_bracket_backtrackingpath(common, current);
+    else
+      compile_assert_backtrackingpath(common, current);
+    break;
+
+    case OP_BRAPOS:
+    case OP_CBRAPOS:
+    case OP_SBRAPOS:
+    case OP_SCBRAPOS:
+    case OP_BRAPOSZERO:
+    compile_bracketpos_backtrackingpath(common, current);
+    break;
+
+    case OP_BRAMINZERO:
+    compile_braminzero_backtrackingpath(common, current);
+    break;
+
+    case OP_MARK:
+    OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(STACK_TOP), STACK(common->has_skip_arg ? 4 : 0));
+    if (common->has_skip_arg)
+      OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(STACK_TOP), STACK(0));
+    free_stack(common, common->has_skip_arg ? 5 : 1);
+    OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->mark_ptr, TMP1, 0);
+    if (common->has_skip_arg)
+      OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->control_head_ptr, TMP2, 0);
+    break;
+
+    case OP_THEN:
+    case OP_THEN_ARG:
+    case OP_PRUNE:
+    case OP_PRUNE_ARG:
+    case OP_SKIP:
+    case OP_SKIP_ARG:
+    compile_control_verb_backtrackingpath(common, current);
+    break;
+
+    case OP_COMMIT:
+    if (!common->local_exit)
+      OP1(SLJIT_MOV, SLJIT_RETURN_REG, 0, SLJIT_IMM, PCRE_ERROR_NOMATCH);
+    if (common->quit_label == NULL)
+      add_jump(compiler, &common->quit, JUMP(SLJIT_JUMP));
+    else
+      JUMPTO(SLJIT_JUMP, common->quit_label);
+    break;
+
+    case OP_CALLOUT:
+    case OP_FAIL:
+    case OP_ACCEPT:
+    case OP_ASSERT_ACCEPT:
+    set_jumps(current->topbacktracks, LABEL());
+    break;
+
+    case OP_THEN_TRAP:
+    /* A virtual opcode for then traps. */
+    compile_then_trap_backtrackingpath(common, current);
+    break;
+
+    default:
+    SLJIT_ASSERT_STOP();
+    break;
+    }
+  current = current->prev;
+  }
+common->then_trap = save_then_trap;
+}
+
+static SLJIT_INLINE void compile_recurse(compiler_common *common)
+{
+DEFINE_COMPILER;
+pcre_uchar *cc = common->start + common->currententry->start;
+pcre_uchar *ccbegin = cc + 1 + LINK_SIZE + (*cc == OP_BRA ? 0 : IMM2_SIZE);
+pcre_uchar *ccend = bracketend(cc);
+BOOL needs_control_head;
+int framesize = get_framesize(common, cc, NULL, TRUE, &needs_control_head);
+int private_data_size = get_private_data_copy_length(common, ccbegin, ccend, needs_control_head);
+int alternativesize;
+BOOL needs_frame;
+backtrack_common altbacktrack;
+struct sljit_jump *jump;
+
+/* Recurse captures then. */
+common->then_trap = NULL;
+
+SLJIT_ASSERT(*cc == OP_BRA || *cc == OP_CBRA || *cc == OP_CBRAPOS || *cc == OP_SCBRA || *cc == OP_SCBRAPOS);
+needs_frame = framesize >= 0;
+if (!needs_frame)
+  framesize = 0;
+alternativesize = *(cc + GET(cc, 1)) == OP_ALT ? 1 : 0;
+
+SLJIT_ASSERT(common->currententry->entry == NULL && common->recursive_head_ptr != 0);
+common->currententry->entry = LABEL();
+set_jumps(common->currententry->calls, common->currententry->entry);
+
+sljit_emit_fast_enter(compiler, TMP2, 0);
+allocate_stack(common, private_data_size + framesize + alternativesize);
+OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(private_data_size + framesize + alternativesize - 1), TMP2, 0);
+copy_private_data(common, ccbegin, ccend, TRUE, private_data_size + framesize + alternativesize, framesize + alternativesize, needs_control_head);
+if (needs_control_head)
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->control_head_ptr, SLJIT_IMM, 0);
+OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->recursive_head_ptr, STACK_TOP, 0);
+if (needs_frame)
+  init_frame(common, cc, NULL, framesize + alternativesize - 1, alternativesize, TRUE);
+
+if (alternativesize > 0)
+  OP1(SLJIT_MOV, SLJIT_MEM1(STACK_TOP), STACK(0), STR_PTR, 0);
+
+memset(&altbacktrack, 0, sizeof(backtrack_common));
+common->quit_label = NULL;
+common->accept_label = NULL;
+common->quit = NULL;
+common->accept = NULL;
+altbacktrack.cc = ccbegin;
+cc += GET(cc, 1);
+while (1)
+  {
+  altbacktrack.top = NULL;
+  altbacktrack.topbacktracks = NULL;
+
+  if (altbacktrack.cc != ccbegin)
+    OP1(SLJIT_MOV, STR_PTR, 0, SLJIT_MEM1(STACK_TOP), STACK(0));
+
+  compile_matchingpath(common, altbacktrack.cc, cc, &altbacktrack);
+  if (SLJIT_UNLIKELY(sljit_get_compiler_error(compiler)))
+    return;
+
+  add_jump(compiler, &common->accept, JUMP(SLJIT_JUMP));
+
+  compile_backtrackingpath(common, altbacktrack.top);
+  if (SLJIT_UNLIKELY(sljit_get_compiler_error(compiler)))
+    return;
+  set_jumps(altbacktrack.topbacktracks, LABEL());
+
+  if (*cc != OP_ALT)
+    break;
+
+  altbacktrack.cc = cc + 1 + LINK_SIZE;
+  cc += GET(cc, 1);
+  }
+
+/* None of them matched. */
+OP1(SLJIT_MOV, TMP3, 0, SLJIT_IMM, 0);
+jump = JUMP(SLJIT_JUMP);
+
+if (common->quit != NULL)
+  {
+  set_jumps(common->quit, LABEL());
+  OP1(SLJIT_MOV, STACK_TOP, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), common->recursive_head_ptr);
+  if (needs_frame)
+    {
+    OP2(SLJIT_SUB, STACK_TOP, 0, STACK_TOP, 0, SLJIT_IMM, (framesize + alternativesize) * sizeof(sljit_sw));
+    add_jump(compiler, &common->revertframes, JUMP(SLJIT_FAST_CALL));
+    OP2(SLJIT_ADD, STACK_TOP, 0, STACK_TOP, 0, SLJIT_IMM, (framesize + alternativesize) * sizeof(sljit_sw));
+    }
+  OP1(SLJIT_MOV, TMP3, 0, SLJIT_IMM, 0);
+  common->quit = NULL;
+  add_jump(compiler, &common->quit, JUMP(SLJIT_JUMP));
+  }
+
+set_jumps(common->accept, LABEL());
+OP1(SLJIT_MOV, STACK_TOP, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), common->recursive_head_ptr);
+if (needs_frame)
+  {
+  OP2(SLJIT_SUB, STACK_TOP, 0, STACK_TOP, 0, SLJIT_IMM, (framesize + alternativesize) * sizeof(sljit_sw));
+  add_jump(compiler, &common->revertframes, JUMP(SLJIT_FAST_CALL));
+  OP2(SLJIT_ADD, STACK_TOP, 0, STACK_TOP, 0, SLJIT_IMM, (framesize + alternativesize) * sizeof(sljit_sw));
+  }
+OP1(SLJIT_MOV, TMP3, 0, SLJIT_IMM, 1);
+
+JUMPHERE(jump);
+if (common->quit != NULL)
+  set_jumps(common->quit, LABEL());
+copy_private_data(common, ccbegin, ccend, FALSE, private_data_size + framesize + alternativesize, framesize + alternativesize, needs_control_head);
+free_stack(common, private_data_size + framesize + alternativesize);
+if (needs_control_head)
+  {
+  OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(STACK_TOP), 2 * sizeof(sljit_sw));
+  OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(STACK_TOP), sizeof(sljit_sw));
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->recursive_head_ptr, TMP1, 0);
+  OP1(SLJIT_MOV, TMP1, 0, TMP3, 0);
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->control_head_ptr, TMP2, 0);
+  }
+else
+  {
+  OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(STACK_TOP), sizeof(sljit_sw));
+  OP1(SLJIT_MOV, TMP1, 0, TMP3, 0);
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->recursive_head_ptr, TMP2, 0);
+  }
+sljit_emit_fast_return(compiler, SLJIT_MEM1(STACK_TOP), 0);
+}
+
+#undef COMPILE_BACKTRACKINGPATH
+#undef CURRENT_AS
+
+void
+PRIV(jit_compile)(const REAL_PCRE *re, PUBL(extra) *extra, int mode)
+{
+struct sljit_compiler *compiler;
+backtrack_common rootbacktrack;
+compiler_common common_data;
+compiler_common *common = &common_data;
+const pcre_uint8 *tables = re->tables;
+pcre_study_data *study;
+int private_data_size;
+pcre_uchar *ccend;
+executable_functions *functions;
+void *executable_func;
+sljit_uw executable_size;
+struct sljit_label *mainloop_label = NULL;
+struct sljit_label *continue_match_label;
+struct sljit_label *empty_match_found_label;
+struct sljit_label *empty_match_backtrack_label;
+struct sljit_label *reset_match_label;
+struct sljit_jump *jump;
+struct sljit_jump *minlength_check_failed = NULL;
+struct sljit_jump *reqbyte_notfound = NULL;
+struct sljit_jump *empty_match;
+struct sljit_label *quit_label;
+
+SLJIT_ASSERT((extra->flags & PCRE_EXTRA_STUDY_DATA) != 0);
+study = extra->study_data;
+
+if (!tables)
+  tables = PRIV(default_tables);
+
+memset(&rootbacktrack, 0, sizeof(backtrack_common));
+memset(common, 0, sizeof(compiler_common));
+rootbacktrack.cc = (pcre_uchar *)re + re->name_table_offset + re->name_count * re->name_entry_size;
+
+common->start = rootbacktrack.cc;
+common->fcc = tables + fcc_offset;
+common->lcc = (sljit_sw)(tables + lcc_offset);
+common->mode = mode;
+common->nltype = NLTYPE_FIXED;
+switch(re->options & PCRE_NEWLINE_BITS)
+  {
+  case 0:
+  /* Compile-time default */
+  switch(NEWLINE)
+    {
+    case -1: common->newline = (CHAR_CR << 8) | CHAR_NL; common->nltype = NLTYPE_ANY; break;
+    case -2: common->newline = (CHAR_CR << 8) | CHAR_NL; common->nltype = NLTYPE_ANYCRLF; break;
+    default: common->newline = NEWLINE; break;
+    }
+  break;
+  case PCRE_NEWLINE_CR: common->newline = CHAR_CR; break;
+  case PCRE_NEWLINE_LF: common->newline = CHAR_NL; break;
+  case PCRE_NEWLINE_CR+
+       PCRE_NEWLINE_LF: common->newline = (CHAR_CR << 8) | CHAR_NL; break;
+  case PCRE_NEWLINE_ANY: common->newline = (CHAR_CR << 8) | CHAR_NL; common->nltype = NLTYPE_ANY; break;
+  case PCRE_NEWLINE_ANYCRLF: common->newline = (CHAR_CR << 8) | CHAR_NL; common->nltype = NLTYPE_ANYCRLF; break;
+  default: return;
+  }
+if ((re->options & PCRE_BSR_ANYCRLF) != 0)
+  common->bsr_nltype = NLTYPE_ANYCRLF;
+else if ((re->options & PCRE_BSR_UNICODE) != 0)
+  common->bsr_nltype = NLTYPE_ANY;
+else
+  {
+#ifdef BSR_ANYCRLF
+  common->bsr_nltype = NLTYPE_ANYCRLF;
+#else
+  common->bsr_nltype = NLTYPE_ANY;
+#endif
+  }
+common->endonly = (re->options & PCRE_DOLLAR_ENDONLY) != 0;
+common->ctypes = (sljit_sw)(tables + ctypes_offset);
+common->digits[0] = -2;
+common->name_table = (sljit_sw)((pcre_uchar *)re + re->name_table_offset);
+common->name_count = re->name_count;
+common->name_entry_size = re->name_entry_size;
+common->jscript_compat = (re->options & PCRE_JAVASCRIPT_COMPAT) != 0;
+#ifdef SUPPORT_UTF
+/* PCRE_UTF[16|32] have the same value as PCRE_UTF8. */
+common->utf = (re->options & PCRE_UTF8) != 0;
+#ifdef SUPPORT_UCP
+common->use_ucp = (re->options & PCRE_UCP) != 0;
+#endif
+#endif /* SUPPORT_UTF */
+ccend = bracketend(rootbacktrack.cc);
+
+/* Calculate the local space size on the stack. */
+common->ovector_start = LIMIT_MATCH + sizeof(sljit_sw);
+common->optimized_cbracket = (pcre_uint8 *)SLJIT_MALLOC(re->top_bracket + 1);
+if (!common->optimized_cbracket)
+  return;
+#if defined DEBUG_FORCE_UNOPTIMIZED_CBRAS && DEBUG_FORCE_UNOPTIMIZED_CBRAS == 1
+memset(common->optimized_cbracket, 0, re->top_bracket + 1);
+#else
+memset(common->optimized_cbracket, 1, re->top_bracket + 1);
+#endif
+
+SLJIT_ASSERT(*rootbacktrack.cc == OP_BRA && ccend[-(1 + LINK_SIZE)] == OP_KET);
+#if defined DEBUG_FORCE_UNOPTIMIZED_CBRAS && DEBUG_FORCE_UNOPTIMIZED_CBRAS == 2
+common->capture_last_ptr = common->ovector_start;
+common->ovector_start += sizeof(sljit_sw);
+#endif
+if (!check_opcode_types(common, rootbacktrack.cc, ccend))
+  {
+  SLJIT_FREE(common->optimized_cbracket);
+  return;
+  }
+
+/* Checking flags and updating ovector_start. */
+if (mode == JIT_COMPILE && (re->flags & PCRE_REQCHSET) != 0 && (re->options & PCRE_NO_START_OPTIMIZE) == 0)
+  {
+  common->req_char_ptr = common->ovector_start;
+  common->ovector_start += sizeof(sljit_sw);
+  }
+if (mode != JIT_COMPILE)
+  {
+  common->start_used_ptr = common->ovector_start;
+  common->ovector_start += sizeof(sljit_sw);
+  if (mode == JIT_PARTIAL_SOFT_COMPILE)
+    {
+    common->hit_start = common->ovector_start;
+    common->ovector_start += 2 * sizeof(sljit_sw);
+    }
+  else
+    {
+    SLJIT_ASSERT(mode == JIT_PARTIAL_HARD_COMPILE);
+    common->needs_start_ptr = TRUE;
+    }
+  }
+if ((re->options & PCRE_FIRSTLINE) != 0)
+  {
+  common->first_line_end = common->ovector_start;
+  common->ovector_start += sizeof(sljit_sw);
+  }
+#if defined DEBUG_FORCE_CONTROL_HEAD && DEBUG_FORCE_CONTROL_HEAD
+common->control_head_ptr = 1;
+#endif
+if (common->control_head_ptr != 0)
+  {
+  common->control_head_ptr = common->ovector_start;
+  common->ovector_start += sizeof(sljit_sw);
+  }
+if (common->needs_start_ptr && common->has_set_som)
+  {
+  /* Saving the real start pointer is necessary. */
+  common->start_ptr = common->ovector_start;
+  common->ovector_start += sizeof(sljit_sw);
+  }
+else
+  common->needs_start_ptr = FALSE;
+
+/* Aligning ovector to even number of sljit words. */
+if ((common->ovector_start & sizeof(sljit_sw)) != 0)
+  common->ovector_start += sizeof(sljit_sw);
+
+if (common->start_ptr == 0)
+  common->start_ptr = OVECTOR(0);
+
+/* Capturing brackets cannot be optimized if callouts are allowed. */
+if (common->capture_last_ptr != 0)
+  memset(common->optimized_cbracket, 0, re->top_bracket + 1);
+
+SLJIT_ASSERT(!(common->req_char_ptr != 0 && common->start_used_ptr != 0));
+common->cbra_ptr = OVECTOR_START + (re->top_bracket + 1) * 2 * sizeof(sljit_sw);
+
+common->private_data_ptrs = (int *)SLJIT_MALLOC((ccend - rootbacktrack.cc) * sizeof(sljit_si));
+if (!common->private_data_ptrs)
+  {
+  SLJIT_FREE(common->optimized_cbracket);
+  return;
+  }
+memset(common->private_data_ptrs, 0, (ccend - rootbacktrack.cc) * sizeof(int));
+
+private_data_size = common->cbra_ptr + (re->top_bracket + 1) * sizeof(sljit_sw);
+set_private_data_ptrs(common, &private_data_size, ccend);
+if (private_data_size > SLJIT_MAX_LOCAL_SIZE)
+  {
+  SLJIT_FREE(common->private_data_ptrs);
+  SLJIT_FREE(common->optimized_cbracket);
+  return;
+  }
+
+if (common->has_then)
+  {
+  common->then_offsets = (pcre_uint8 *)SLJIT_MALLOC(ccend - rootbacktrack.cc);
+  if (!common->then_offsets)
+    {
+    SLJIT_FREE(common->optimized_cbracket);
+    SLJIT_FREE(common->private_data_ptrs);
+    return;
+    }
+  memset(common->then_offsets, 0, ccend - rootbacktrack.cc);
+  set_then_offsets(common, rootbacktrack.cc, NULL);
+  }
+
+compiler = sljit_create_compiler();
+if (!compiler)
+  {
+  SLJIT_FREE(common->optimized_cbracket);
+  SLJIT_FREE(common->private_data_ptrs);
+  if (common->has_then)
+    SLJIT_FREE(common->then_offsets);
+  return;
+  }
+common->compiler = compiler;
+
+/* Main pcre_jit_exec entry. */
+sljit_emit_enter(compiler, 1, 5, 5, private_data_size);
+
+/* Register init. */
+reset_ovector(common, (re->top_bracket + 1) * 2);
+if (common->req_char_ptr != 0)
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->req_char_ptr, SLJIT_SCRATCH_REG1, 0);
+
+OP1(SLJIT_MOV, ARGUMENTS, 0, SLJIT_SAVED_REG1, 0);
+OP1(SLJIT_MOV, TMP1, 0, SLJIT_SAVED_REG1, 0);
+OP1(SLJIT_MOV, STR_PTR, 0, SLJIT_MEM1(TMP1), SLJIT_OFFSETOF(jit_arguments, str));
+OP1(SLJIT_MOV, STR_END, 0, SLJIT_MEM1(TMP1), SLJIT_OFFSETOF(jit_arguments, end));
+OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(TMP1), SLJIT_OFFSETOF(jit_arguments, stack));
+OP1(SLJIT_MOV_UI, TMP1, 0, SLJIT_MEM1(TMP1), SLJIT_OFFSETOF(jit_arguments, limit_match));
+OP1(SLJIT_MOV, STACK_TOP, 0, SLJIT_MEM1(TMP2), SLJIT_OFFSETOF(struct sljit_stack, base));
+OP1(SLJIT_MOV, STACK_LIMIT, 0, SLJIT_MEM1(TMP2), SLJIT_OFFSETOF(struct sljit_stack, limit));
+OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), LIMIT_MATCH, TMP1, 0);
+
+if (mode == JIT_PARTIAL_SOFT_COMPILE)
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->hit_start, SLJIT_IMM, -1);
+if (common->mark_ptr != 0)
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->mark_ptr, SLJIT_IMM, 0);
+if (common->control_head_ptr != 0)
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->control_head_ptr, SLJIT_IMM, 0);
+
+/* Main part of the matching */
+if ((re->options & PCRE_ANCHORED) == 0)
+  {
+  mainloop_label = mainloop_entry(common, (re->flags & PCRE_HASCRORLF) != 0, (re->options & PCRE_FIRSTLINE) != 0);
+  continue_match_label = LABEL();
+  /* Forward search if possible. */
+  if ((re->options & PCRE_NO_START_OPTIMIZE) == 0)
+    {
+    if (mode == JIT_COMPILE && fast_forward_first_n_chars(common, (re->options & PCRE_FIRSTLINE) != 0))
+      { /* Do nothing */ }
+    else if ((re->flags & PCRE_FIRSTSET) != 0)
+      fast_forward_first_char(common, (pcre_uchar)re->first_char, (re->flags & PCRE_FCH_CASELESS) != 0, (re->options & PCRE_FIRSTLINE) != 0);
+    else if ((re->flags & PCRE_STARTLINE) != 0)
+      fast_forward_newline(common, (re->options & PCRE_FIRSTLINE) != 0);
+    else if ((re->flags & PCRE_STARTLINE) == 0 && study != NULL && (study->flags & PCRE_STUDY_MAPPED) != 0)
+      fast_forward_start_bits(common, (sljit_uw)study->start_bits, (re->options & PCRE_FIRSTLINE) != 0);
+    }
+  }
+else
+  continue_match_label = LABEL();
+
+if (mode == JIT_COMPILE && study->minlength > 0 && (re->options & PCRE_NO_START_OPTIMIZE) == 0)
+  {
+  OP1(SLJIT_MOV, SLJIT_RETURN_REG, 0, SLJIT_IMM, PCRE_ERROR_NOMATCH);
+  OP2(SLJIT_ADD, TMP2, 0, STR_PTR, 0, SLJIT_IMM, IN_UCHARS(study->minlength));
+  minlength_check_failed = CMP(SLJIT_C_GREATER, TMP2, 0, STR_END, 0);
+  }
+if (common->req_char_ptr != 0)
+  reqbyte_notfound = search_requested_char(common, (pcre_uchar)re->req_char, (re->flags & PCRE_RCH_CASELESS) != 0, (re->flags & PCRE_FIRSTSET) != 0);
+
+/* Store the current STR_PTR in OVECTOR(0). */
+OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(0), STR_PTR, 0);
+/* Copy the limit of allowed recursions. */
+OP1(SLJIT_MOV, COUNT_MATCH, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), LIMIT_MATCH);
+if (common->capture_last_ptr != 0)
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->capture_last_ptr, SLJIT_IMM, -1);
+
+if (common->needs_start_ptr)
+  {
+  SLJIT_ASSERT(common->start_ptr != OVECTOR(0));
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->start_ptr, STR_PTR, 0);
+  }
+else
+  SLJIT_ASSERT(common->start_ptr == OVECTOR(0));
+
+/* Copy the beginning of the string. */
+if (mode == JIT_PARTIAL_SOFT_COMPILE)
+  {
+  jump = CMP(SLJIT_C_NOT_EQUAL, SLJIT_MEM1(SLJIT_LOCALS_REG), common->hit_start, SLJIT_IMM, -1);
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->start_used_ptr, STR_PTR, 0);
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->hit_start + sizeof(sljit_sw), STR_PTR, 0);
+  JUMPHERE(jump);
+  }
+else if (mode == JIT_PARTIAL_HARD_COMPILE)
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->start_used_ptr, STR_PTR, 0);
+
+compile_matchingpath(common, rootbacktrack.cc, ccend, &rootbacktrack);
+if (SLJIT_UNLIKELY(sljit_get_compiler_error(compiler)))
+  {
+  sljit_free_compiler(compiler);
+  SLJIT_FREE(common->optimized_cbracket);
+  SLJIT_FREE(common->private_data_ptrs);
+  if (common->has_then)
+    SLJIT_FREE(common->then_offsets);
+  return;
+  }
+
+empty_match = CMP(SLJIT_C_EQUAL, STR_PTR, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), OVECTOR(0));
+empty_match_found_label = LABEL();
+
+common->accept_label = LABEL();
+if (common->accept != NULL)
+  set_jumps(common->accept, common->accept_label);
+
+/* This means we have a match. Update the ovector. */
+copy_ovector(common, re->top_bracket + 1);
+common->quit_label = common->forced_quit_label = LABEL();
+if (common->quit != NULL)
+  set_jumps(common->quit, common->quit_label);
+if (common->forced_quit != NULL)
+  set_jumps(common->forced_quit, common->forced_quit_label);
+if (minlength_check_failed != NULL)
+  SET_LABEL(minlength_check_failed, common->forced_quit_label);
+sljit_emit_return(compiler, SLJIT_MOV, SLJIT_RETURN_REG, 0);
+
+if (mode != JIT_COMPILE)
+  {
+  common->partialmatchlabel = LABEL();
+  set_jumps(common->partialmatch, common->partialmatchlabel);
+  return_with_partial_match(common, common->quit_label);
+  }
+
+empty_match_backtrack_label = LABEL();
+compile_backtrackingpath(common, rootbacktrack.top);
+if (SLJIT_UNLIKELY(sljit_get_compiler_error(compiler)))
+  {
+  sljit_free_compiler(compiler);
+  SLJIT_FREE(common->optimized_cbracket);
+  SLJIT_FREE(common->private_data_ptrs);
+  if (common->has_then)
+    SLJIT_FREE(common->then_offsets);
+  return;
+  }
+
+SLJIT_ASSERT(rootbacktrack.prev == NULL);
+reset_match_label = LABEL();
+
+if (mode == JIT_PARTIAL_SOFT_COMPILE)
+  {
+  /* Update hit_start only in the first time. */
+  jump = CMP(SLJIT_C_NOT_EQUAL, SLJIT_MEM1(SLJIT_LOCALS_REG), common->hit_start, SLJIT_IMM, 0);
+  OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), common->start_used_ptr);
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->start_used_ptr, SLJIT_IMM, -1);
+  OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), common->hit_start, TMP1, 0);
+  JUMPHERE(jump);
+  }
+
+/* Check we have remaining characters. */
+if ((re->options & PCRE_ANCHORED) == 0 && (re->options & PCRE_FIRSTLINE) != 0)
+  {
+  SLJIT_ASSERT(common->first_line_end != 0);
+  OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), common->first_line_end);
+  }
+
+OP1(SLJIT_MOV, STR_PTR, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), common->start_ptr);
+
+if ((re->options & PCRE_ANCHORED) == 0)
+  {
+  if ((re->options & PCRE_FIRSTLINE) == 0)
+    CMPTO(SLJIT_C_LESS, STR_PTR, 0, STR_END, 0, mainloop_label);
+  else
+    CMPTO(SLJIT_C_LESS, STR_PTR, 0, TMP1, 0, mainloop_label);
+  }
+
+/* No more remaining characters. */
+if (reqbyte_notfound != NULL)
+  JUMPHERE(reqbyte_notfound);
+
+if (mode == JIT_PARTIAL_SOFT_COMPILE)
+  CMPTO(SLJIT_C_NOT_EQUAL, SLJIT_MEM1(SLJIT_LOCALS_REG), common->hit_start, SLJIT_IMM, -1, common->partialmatchlabel);
+
+OP1(SLJIT_MOV, SLJIT_RETURN_REG, 0, SLJIT_IMM, PCRE_ERROR_NOMATCH);
+JUMPTO(SLJIT_JUMP, common->quit_label);
+
+flush_stubs(common);
+
+JUMPHERE(empty_match);
+OP1(SLJIT_MOV, TMP1, 0, ARGUMENTS, 0);
+OP1(SLJIT_MOV_UB, TMP2, 0, SLJIT_MEM1(TMP1), SLJIT_OFFSETOF(jit_arguments, notempty));
+CMPTO(SLJIT_C_NOT_EQUAL, TMP2, 0, SLJIT_IMM, 0, empty_match_backtrack_label);
+OP1(SLJIT_MOV_UB, TMP2, 0, SLJIT_MEM1(TMP1), SLJIT_OFFSETOF(jit_arguments, notempty_atstart));
+CMPTO(SLJIT_C_EQUAL, TMP2, 0, SLJIT_IMM, 0, empty_match_found_label);
+OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(TMP1), SLJIT_OFFSETOF(jit_arguments, str));
+CMPTO(SLJIT_C_NOT_EQUAL, TMP2, 0, STR_PTR, 0, empty_match_found_label);
+JUMPTO(SLJIT_JUMP, empty_match_backtrack_label);
+
+common->currententry = common->entries;
+common->local_exit = TRUE;
+quit_label = common->quit_label;
+while (common->currententry != NULL)
+  {
+  /* Might add new entries. */
+  compile_recurse(common);
+  if (SLJIT_UNLIKELY(sljit_get_compiler_error(compiler)))
+    {
+    sljit_free_compiler(compiler);
+    SLJIT_FREE(common->optimized_cbracket);
+    SLJIT_FREE(common->private_data_ptrs);
+    if (common->has_then)
+      SLJIT_FREE(common->then_offsets);
+    return;
+    }
+  flush_stubs(common);
+  common->currententry = common->currententry->next;
+  }
+common->local_exit = FALSE;
+common->quit_label = quit_label;
+
+/* Allocating stack, returns with PCRE_ERROR_JIT_STACKLIMIT if fails. */
+/* This is a (really) rare case. */
+set_jumps(common->stackalloc, LABEL());
+/* RETURN_ADDR is not a saved register. */
+sljit_emit_fast_enter(compiler, SLJIT_MEM1(SLJIT_LOCALS_REG), LOCALS0);
+OP1(SLJIT_MOV, SLJIT_MEM1(SLJIT_LOCALS_REG), LOCALS1, TMP2, 0);
+OP1(SLJIT_MOV, TMP1, 0, ARGUMENTS, 0);
+OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(TMP1), SLJIT_OFFSETOF(jit_arguments, stack));
+OP1(SLJIT_MOV, SLJIT_MEM1(TMP1), SLJIT_OFFSETOF(struct sljit_stack, top), STACK_TOP, 0);
+OP2(SLJIT_ADD, TMP2, 0, SLJIT_MEM1(TMP1), SLJIT_OFFSETOF(struct sljit_stack, limit), SLJIT_IMM, STACK_GROWTH_RATE);
+
+sljit_emit_ijump(compiler, SLJIT_CALL2, SLJIT_IMM, SLJIT_FUNC_OFFSET(sljit_stack_resize));
+jump = CMP(SLJIT_C_NOT_EQUAL, SLJIT_RETURN_REG, 0, SLJIT_IMM, 0);
+OP1(SLJIT_MOV, TMP1, 0, ARGUMENTS, 0);
+OP1(SLJIT_MOV, TMP1, 0, SLJIT_MEM1(TMP1), SLJIT_OFFSETOF(jit_arguments, stack));
+OP1(SLJIT_MOV, STACK_TOP, 0, SLJIT_MEM1(TMP1), SLJIT_OFFSETOF(struct sljit_stack, top));
+OP1(SLJIT_MOV, STACK_LIMIT, 0, SLJIT_MEM1(TMP1), SLJIT_OFFSETOF(struct sljit_stack, limit));
+OP1(SLJIT_MOV, TMP2, 0, SLJIT_MEM1(SLJIT_LOCALS_REG), LOCALS1);
+sljit_emit_fast_return(compiler, SLJIT_MEM1(SLJIT_LOCALS_REG), LOCALS0);
+
+/* Allocation failed. */
+JUMPHERE(jump);
+/* We break the return address cache here, but this is a really rare case. */
+OP1(SLJIT_MOV, SLJIT_RETURN_REG, 0, SLJIT_IMM, PCRE_ERROR_JIT_STACKLIMIT);
+JUMPTO(SLJIT_JUMP, common->quit_label);
+
+/* Call limit reached. */
+set_jumps(common->calllimit, LABEL());
+OP1(SLJIT_MOV, SLJIT_RETURN_REG, 0, SLJIT_IMM, PCRE_ERROR_MATCHLIMIT);
+JUMPTO(SLJIT_JUMP, common->quit_label);
+
+if (common->revertframes != NULL)
+  {
+  set_jumps(common->revertframes, LABEL());
+  do_revertframes(common);
+  }
+if (common->wordboundary != NULL)
+  {
+  set_jumps(common->wordboundary, LABEL());
+  check_wordboundary(common);
+  }
+if (common->anynewline != NULL)
+  {
+  set_jumps(common->anynewline, LABEL());
+  check_anynewline(common);
+  }
+if (common->hspace != NULL)
+  {
+  set_jumps(common->hspace, LABEL());
+  check_hspace(common);
+  }
+if (common->vspace != NULL)
+  {
+  set_jumps(common->vspace, LABEL());
+  check_vspace(common);
+  }
+if (common->casefulcmp != NULL)
+  {
+  set_jumps(common->casefulcmp, LABEL());
+  do_casefulcmp(common);
+  }
+if (common->caselesscmp != NULL)
+  {
+  set_jumps(common->caselesscmp, LABEL());
+  do_caselesscmp(common);
+  }
+if (common->reset_match != NULL)
+  {
+  set_jumps(common->reset_match, LABEL());
+  do_reset_match(common, (re->top_bracket + 1) * 2);
+  CMPTO(SLJIT_C_GREATER, STR_PTR, 0, TMP1, 0, continue_match_label);
+  OP1(SLJIT_MOV, STR_PTR, 0, TMP1, 0);
+  JUMPTO(SLJIT_JUMP, reset_match_label);
+  }
+#ifdef SUPPORT_UTF
+#ifndef COMPILE_PCRE32
+if (common->utfreadchar != NULL)
+  {
+  set_jumps(common->utfreadchar, LABEL());
+  do_utfreadchar(common);
+  }
+#endif /* !COMPILE_PCRE32 */
+#ifdef COMPILE_PCRE8
+if (common->utfreadtype8 != NULL)
+  {
+  set_jumps(common->utfreadtype8, LABEL());
+  do_utfreadtype8(common);
+  }
+#endif /* COMPILE_PCRE8 */
+#endif /* SUPPORT_UTF */
+#ifdef SUPPORT_UCP
+if (common->getucd != NULL)
+  {
+  set_jumps(common->getucd, LABEL());
+  do_getucd(common);
+  }
+#endif
+
+SLJIT_FREE(common->optimized_cbracket);
+SLJIT_FREE(common->private_data_ptrs);
+if (common->has_then)
+  SLJIT_FREE(common->then_offsets);
+
+executable_func = sljit_generate_code(compiler);
+executable_size = sljit_get_generated_code_size(compiler);
+sljit_free_compiler(compiler);
+if (executable_func == NULL)
+  return;
+
+/* Reuse the function descriptor if possible. */
+if ((extra->flags & PCRE_EXTRA_EXECUTABLE_JIT) != 0 && extra->executable_jit != NULL)
+  functions = (executable_functions *)extra->executable_jit;
+else
+  {
+  /* Note: If your memory-checker has flagged the allocation below as a
+   * memory leak, it is probably because you either forgot to call
+   * pcre_free_study() (or pcre16_free_study()) on the pcre_extra (or
+   * pcre16_extra) object, or you called said function after having
+   * cleared the PCRE_EXTRA_EXECUTABLE_JIT bit from the "flags" field
+   * of the object. (The function will only free the JIT data if the
+   * bit remains set, as the bit indicates that the pointer to the data
+   * is valid.)
+   */
+  functions = SLJIT_MALLOC(sizeof(executable_functions));
+  if (functions == NULL)
+    {
+    /* This case is highly unlikely since we just recently
+    freed a lot of memory. Although not impossible. */
+    sljit_free_code(executable_func);
+    return;
+    }
+  memset(functions, 0, sizeof(executable_functions));
+  functions->top_bracket = (re->top_bracket + 1) * 2;
+  functions->limit_match = (re->flags & PCRE_MLSET) != 0 ? re->limit_match : 0;
+  extra->executable_jit = functions;
+  extra->flags |= PCRE_EXTRA_EXECUTABLE_JIT;
+  }
+
+functions->executable_funcs[mode] = executable_func;
+functions->executable_sizes[mode] = executable_size;
+}
+
+static int jit_machine_stack_exec(jit_arguments *arguments, void* executable_func)
+{
+union {
+   void* executable_func;
+   jit_function call_executable_func;
+} convert_executable_func;
+pcre_uint8 local_space[MACHINE_STACK_SIZE];
+struct sljit_stack local_stack;
+
+local_stack.top = (sljit_sw)&local_space;
+local_stack.base = local_stack.top;
+local_stack.limit = local_stack.base + MACHINE_STACK_SIZE;
+local_stack.max_limit = local_stack.limit;
+arguments->stack = &local_stack;
+convert_executable_func.executable_func = executable_func;
+return convert_executable_func.call_executable_func(arguments);
+}
+
+int
+PRIV(jit_exec)(const PUBL(extra) *extra_data, const pcre_uchar *subject,
+  int length, int start_offset, int options, int *offsets, int offset_count)
+{
+executable_functions *functions = (executable_functions *)extra_data->executable_jit;
+union {
+   void* executable_func;
+   jit_function call_executable_func;
+} convert_executable_func;
+jit_arguments arguments;
+int max_offset_count;
+int retval;
+int mode = JIT_COMPILE;
+
+if ((options & PCRE_PARTIAL_HARD) != 0)
+  mode = JIT_PARTIAL_HARD_COMPILE;
+else if ((options & PCRE_PARTIAL_SOFT) != 0)
+  mode = JIT_PARTIAL_SOFT_COMPILE;
+
+if (functions->executable_funcs[mode] == NULL)
+  return PCRE_ERROR_JIT_BADOPTION;
+
+/* Sanity checks should be handled by pcre_exec. */
+arguments.str = subject + start_offset;
+arguments.begin = subject;
+arguments.end = subject + length;
+arguments.mark_ptr = NULL;
+/* JIT decreases this value less frequently than the interpreter. */
+arguments.limit_match = ((extra_data->flags & PCRE_EXTRA_MATCH_LIMIT) == 0) ? MATCH_LIMIT : (pcre_uint32)(extra_data->match_limit);
+if (functions->limit_match != 0 && functions->limit_match < arguments.limit_match)
+  arguments.limit_match = functions->limit_match;
+arguments.notbol = (options & PCRE_NOTBOL) != 0;
+arguments.noteol = (options & PCRE_NOTEOL) != 0;
+arguments.notempty = (options & PCRE_NOTEMPTY) != 0;
+arguments.notempty_atstart = (options & PCRE_NOTEMPTY_ATSTART) != 0;
+arguments.offsets = offsets;
+arguments.callout_data = (extra_data->flags & PCRE_EXTRA_CALLOUT_DATA) != 0 ? extra_data->callout_data : NULL;
+arguments.real_offset_count = offset_count;
+
+/* pcre_exec() rounds offset_count to a multiple of 3, and then uses only 2/3 of
+the output vector for storing captured strings, with the remainder used as
+workspace. We don't need the workspace here. For compatibility, we limit the
+number of captured strings in the same way as pcre_exec(), so that the user
+gets the same result with and without JIT. */
+
+if (offset_count != 2)
+  offset_count = ((offset_count - (offset_count % 3)) * 2) / 3;
+max_offset_count = functions->top_bracket;
+if (offset_count > max_offset_count)
+  offset_count = max_offset_count;
+arguments.offset_count = offset_count;
+
+if (functions->callback)
+  arguments.stack = (struct sljit_stack *)functions->callback(functions->userdata);
+else
+  arguments.stack = (struct sljit_stack *)functions->userdata;
+
+if (arguments.stack == NULL)
+  retval = jit_machine_stack_exec(&arguments, functions->executable_funcs[mode]);
+else
+  {
+  convert_executable_func.executable_func = functions->executable_funcs[mode];
+  retval = convert_executable_func.call_executable_func(&arguments);
+  }
+
+if (retval * 2 > offset_count)
+  retval = 0;
+if ((extra_data->flags & PCRE_EXTRA_MARK) != 0)
+  *(extra_data->mark) = arguments.mark_ptr;
+
+return retval;
+}
+
+#if defined COMPILE_PCRE8
+PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
+pcre_jit_exec(const pcre *argument_re, const pcre_extra *extra_data,
+  PCRE_SPTR subject, int length, int start_offset, int options,
+  int *offsets, int offset_count, pcre_jit_stack *stack)
+#elif defined COMPILE_PCRE16
+PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
+pcre16_jit_exec(const pcre16 *argument_re, const pcre16_extra *extra_data,
+  PCRE_SPTR16 subject, int length, int start_offset, int options,
+  int *offsets, int offset_count, pcre16_jit_stack *stack)
+#elif defined COMPILE_PCRE32
+PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
+pcre32_jit_exec(const pcre32 *argument_re, const pcre32_extra *extra_data,
+  PCRE_SPTR32 subject, int length, int start_offset, int options,
+  int *offsets, int offset_count, pcre32_jit_stack *stack)
+#endif
+{
+pcre_uchar *subject_ptr = (pcre_uchar *)subject;
+executable_functions *functions = (executable_functions *)extra_data->executable_jit;
+union {
+   void* executable_func;
+   jit_function call_executable_func;
+} convert_executable_func;
+jit_arguments arguments;
+int max_offset_count;
+int retval;
+int mode = JIT_COMPILE;
+
+SLJIT_UNUSED_ARG(argument_re);
+
+/* Plausibility checks */
+if ((options & ~PUBLIC_JIT_EXEC_OPTIONS) != 0) return PCRE_ERROR_JIT_BADOPTION;
+
+if ((options & PCRE_PARTIAL_HARD) != 0)
+  mode = JIT_PARTIAL_HARD_COMPILE;
+else if ((options & PCRE_PARTIAL_SOFT) != 0)
+  mode = JIT_PARTIAL_SOFT_COMPILE;
+
+if (functions->executable_funcs[mode] == NULL)
+  return PCRE_ERROR_JIT_BADOPTION;
+
+/* Sanity checks should be handled by pcre_exec. */
+arguments.stack = (struct sljit_stack *)stack;
+arguments.str = subject_ptr + start_offset;
+arguments.begin = subject_ptr;
+arguments.end = subject_ptr + length;
+arguments.mark_ptr = NULL;
+/* JIT decreases this value less frequently than the interpreter. */
+arguments.limit_match = ((extra_data->flags & PCRE_EXTRA_MATCH_LIMIT) == 0) ? MATCH_LIMIT : (pcre_uint32)(extra_data->match_limit);
+if (functions->limit_match != 0 && functions->limit_match < arguments.limit_match)
+  arguments.limit_match = functions->limit_match;
+arguments.notbol = (options & PCRE_NOTBOL) != 0;
+arguments.noteol = (options & PCRE_NOTEOL) != 0;
+arguments.notempty = (options & PCRE_NOTEMPTY) != 0;
+arguments.notempty_atstart = (options & PCRE_NOTEMPTY_ATSTART) != 0;
+arguments.offsets = offsets;
+arguments.callout_data = (extra_data->flags & PCRE_EXTRA_CALLOUT_DATA) != 0 ? extra_data->callout_data : NULL;
+arguments.real_offset_count = offset_count;
+
+/* pcre_exec() rounds offset_count to a multiple of 3, and then uses only 2/3 of
+the output vector for storing captured strings, with the remainder used as
+workspace. We don't need the workspace here. For compatibility, we limit the
+number of captured strings in the same way as pcre_exec(), so that the user
+gets the same result with and without JIT. */
+
+if (offset_count != 2)
+  offset_count = ((offset_count - (offset_count % 3)) * 2) / 3;
+max_offset_count = functions->top_bracket;
+if (offset_count > max_offset_count)
+  offset_count = max_offset_count;
+arguments.offset_count = offset_count;
+
+convert_executable_func.executable_func = functions->executable_funcs[mode];
+retval = convert_executable_func.call_executable_func(&arguments);
+
+if (retval * 2 > offset_count)
+  retval = 0;
+if ((extra_data->flags & PCRE_EXTRA_MARK) != 0)
+  *(extra_data->mark) = arguments.mark_ptr;
+
+return retval;
+}
+
+void
+PRIV(jit_free)(void *executable_funcs)
+{
+int i;
+executable_functions *functions = (executable_functions *)executable_funcs;
+for (i = 0; i < JIT_NUMBER_OF_COMPILE_MODES; i++)
+  {
+  if (functions->executable_funcs[i] != NULL)
+    sljit_free_code(functions->executable_funcs[i]);
+  }
+SLJIT_FREE(functions);
+}
+
+int
+PRIV(jit_get_size)(void *executable_funcs)
+{
+int i;
+sljit_uw size = 0;
+sljit_uw *executable_sizes = ((executable_functions *)executable_funcs)->executable_sizes;
+for (i = 0; i < JIT_NUMBER_OF_COMPILE_MODES; i++)
+  size += executable_sizes[i];
+return (int)size;
+}
+
+const char*
+PRIV(jit_get_target)(void)
+{
+return sljit_get_platform_name();
+}
+
+#if defined COMPILE_PCRE8
+PCRE_EXP_DECL pcre_jit_stack *
+pcre_jit_stack_alloc(int startsize, int maxsize)
+#elif defined COMPILE_PCRE16
+PCRE_EXP_DECL pcre16_jit_stack *
+pcre16_jit_stack_alloc(int startsize, int maxsize)
+#elif defined COMPILE_PCRE32
+PCRE_EXP_DECL pcre32_jit_stack *
+pcre32_jit_stack_alloc(int startsize, int maxsize)
+#endif
+{
+if (startsize < 1 || maxsize < 1)
+  return NULL;
+if (startsize > maxsize)
+  startsize = maxsize;
+startsize = (startsize + STACK_GROWTH_RATE - 1) & ~(STACK_GROWTH_RATE - 1);
+maxsize = (maxsize + STACK_GROWTH_RATE - 1) & ~(STACK_GROWTH_RATE - 1);
+return (PUBL(jit_stack)*)sljit_allocate_stack(startsize, maxsize);
+}
+
+#if defined COMPILE_PCRE8
+PCRE_EXP_DECL void
+pcre_jit_stack_free(pcre_jit_stack *stack)
+#elif defined COMPILE_PCRE16
+PCRE_EXP_DECL void
+pcre16_jit_stack_free(pcre16_jit_stack *stack)
+#elif defined COMPILE_PCRE32
+PCRE_EXP_DECL void
+pcre32_jit_stack_free(pcre32_jit_stack *stack)
+#endif
+{
+sljit_free_stack((struct sljit_stack *)stack);
+}
+
+#if defined COMPILE_PCRE8
+PCRE_EXP_DECL void
+pcre_assign_jit_stack(pcre_extra *extra, pcre_jit_callback callback, void *userdata)
+#elif defined COMPILE_PCRE16
+PCRE_EXP_DECL void
+pcre16_assign_jit_stack(pcre16_extra *extra, pcre16_jit_callback callback, void *userdata)
+#elif defined COMPILE_PCRE32
+PCRE_EXP_DECL void
+pcre32_assign_jit_stack(pcre32_extra *extra, pcre32_jit_callback callback, void *userdata)
+#endif
+{
+executable_functions *functions;
+if (extra != NULL &&
+    (extra->flags & PCRE_EXTRA_EXECUTABLE_JIT) != 0 &&
+    extra->executable_jit != NULL)
+  {
+  functions = (executable_functions *)extra->executable_jit;
+  functions->callback = callback;
+  functions->userdata = userdata;
+  }
+}
+
+#else  /* SUPPORT_JIT */
+
+/* These are dummy functions to avoid linking errors when JIT support is not
+being compiled. */
+
+#if defined COMPILE_PCRE8
+PCRE_EXP_DECL pcre_jit_stack *
+pcre_jit_stack_alloc(int startsize, int maxsize)
+#elif defined COMPILE_PCRE16
+PCRE_EXP_DECL pcre16_jit_stack *
+pcre16_jit_stack_alloc(int startsize, int maxsize)
+#elif defined COMPILE_PCRE32
+PCRE_EXP_DECL pcre32_jit_stack *
+pcre32_jit_stack_alloc(int startsize, int maxsize)
+#endif
+{
+(void)startsize;
+(void)maxsize;
+return NULL;
+}
+
+#if defined COMPILE_PCRE8
+PCRE_EXP_DECL void
+pcre_jit_stack_free(pcre_jit_stack *stack)
+#elif defined COMPILE_PCRE16
+PCRE_EXP_DECL void
+pcre16_jit_stack_free(pcre16_jit_stack *stack)
+#elif defined COMPILE_PCRE32
+PCRE_EXP_DECL void
+pcre32_jit_stack_free(pcre32_jit_stack *stack)
+#endif
+{
+(void)stack;
+}
+
+#if defined COMPILE_PCRE8
+PCRE_EXP_DECL void
+pcre_assign_jit_stack(pcre_extra *extra, pcre_jit_callback callback, void *userdata)
+#elif defined COMPILE_PCRE16
+PCRE_EXP_DECL void
+pcre16_assign_jit_stack(pcre16_extra *extra, pcre16_jit_callback callback, void *userdata)
+#elif defined COMPILE_PCRE32
+PCRE_EXP_DECL void
+pcre32_assign_jit_stack(pcre32_extra *extra, pcre32_jit_callback callback, void *userdata)
+#endif
+{
+(void)extra;
+(void)callback;
+(void)userdata;
+}
+
+#endif
+
+/* End of pcre_jit_compile.c */
diff --git a/pcre/pcre_jit_test.c b/pcre/pcre_jit_test.c
new file mode 100644
index 00000000000..1c247c58117
--- /dev/null
+++ b/pcre/pcre_jit_test.c
@@ -0,0 +1,1691 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                  Main Library written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+  This JIT compiler regression test program was written by Zoltan Herczeg
+                      Copyright (c) 2010-2012
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include 
+#include 
+#include "pcre.h"
+
+
+#include "pcre_internal.h"
+
+#define PCRE_BUG 0x80000000
+
+/*
+ Letter characters:
+   \xe6\x92\xad = 0x64ad = 25773 (kanji)
+ Non-letter characters:
+   \xc2\xa1 = 0xa1 =  (Inverted Exclamation Mark)
+   \xf3\xa9\xb7\x80 = 0xe9dc0 = 957888
+   \xed\xa0\x80 = 55296 = 0xd800 (Invalid UTF character)
+   \xed\xb0\x80 = 56320 = 0xdc00 (Invalid UTF character)
+ Newlines:
+   \xc2\x85 = 0x85 = 133 (NExt Line = NEL)
+   \xe2\x80\xa8 = 0x2028 = 8232 (Line Separator)
+ Othercase pairs:
+   \xc3\xa9 = 0xe9 = 233 (e')
+      \xc3\x89 = 0xc9 = 201 (E')
+   \xc3\xa1 = 0xe1 = 225 (a')
+      \xc3\x81 = 0xc1 = 193 (A')
+   \xc8\xba = 0x23a = 570
+      \xe2\xb1\xa5 = 0x2c65 = 11365
+   \xe1\xbd\xb8 = 0x1f78 = 8056
+      \xe1\xbf\xb8 = 0x1ff8 = 8184
+   \xf0\x90\x90\x80 = 0x10400 = 66560
+      \xf0\x90\x90\xa8 = 0x10428 = 66600
+ Mark property:
+   \xcc\x8d = 0x30d = 781
+ Special:
+   \xdf\xbf = 0x7ff = 2047 (highest 2 byte character)
+   \xe0\xa0\x80 = 0x800 = 2048 (lowest 2 byte character)
+   \xef\xbf\xbf = 0xffff = 65535 (highest 3 byte character)
+   \xf0\x90\x80\x80 = 0x10000 = 65536 (lowest 4 byte character)
+   \xf4\x8f\xbf\xbf = 0x10ffff = 1114111 (highest allowed utf character)
+*/
+
+static int regression_tests(void);
+
+int main(void)
+{
+	int jit = 0;
+#if defined SUPPORT_PCRE8
+	pcre_config(PCRE_CONFIG_JIT, &jit);
+#elif defined SUPPORT_PCRE16
+	pcre16_config(PCRE_CONFIG_JIT, &jit);
+#elif defined SUPPORT_PCRE32
+	pcre32_config(PCRE_CONFIG_JIT, &jit);
+#endif
+	if (!jit) {
+		printf("JIT must be enabled to run pcre_jit_test\n");
+		return 1;
+	}
+	return regression_tests();
+}
+
+/* --------------------------------------------------------------------------------------- */
+
+#if !(defined SUPPORT_PCRE8) && !(defined SUPPORT_PCRE16) && !(defined SUPPORT_PCRE32)
+#error SUPPORT_PCRE8 or SUPPORT_PCRE16 or SUPPORT_PCRE32 must be defined
+#endif
+
+#define MUA	(PCRE_MULTILINE | PCRE_UTF8 | PCRE_NEWLINE_ANYCRLF)
+#define MUAP	(PCRE_MULTILINE | PCRE_UTF8 | PCRE_NEWLINE_ANYCRLF | PCRE_UCP)
+#define CMUA	(PCRE_CASELESS | PCRE_MULTILINE | PCRE_UTF8 | PCRE_NEWLINE_ANYCRLF)
+#define CMUAP	(PCRE_CASELESS | PCRE_MULTILINE | PCRE_UTF8 | PCRE_NEWLINE_ANYCRLF | PCRE_UCP)
+#define MA	(PCRE_MULTILINE | PCRE_NEWLINE_ANYCRLF)
+#define MAP	(PCRE_MULTILINE | PCRE_NEWLINE_ANYCRLF | PCRE_UCP)
+#define CMA	(PCRE_CASELESS | PCRE_MULTILINE | PCRE_NEWLINE_ANYCRLF)
+
+#define OFFSET_MASK	0x00ffff
+#define F_NO8		0x010000
+#define F_NO16		0x020000
+#define F_NO32		0x020000
+#define F_NOMATCH	0x040000
+#define F_DIFF		0x080000
+#define F_FORCECONV	0x100000
+#define F_PROPERTY	0x200000
+#define F_STUDY		0x400000
+
+struct regression_test_case {
+	int flags;
+	int start_offset;
+	const char *pattern;
+	const char *input;
+};
+
+static struct regression_test_case regression_test_cases[] = {
+	/* Constant strings. */
+	{ MUA, 0, "AbC", "AbAbC" },
+	{ MUA, 0, "ACCEPT", "AACACCACCEACCEPACCEPTACCEPTT" },
+	{ CMUA, 0, "aA#\xc3\xa9\xc3\x81", "aA#Aa#\xc3\x89\xc3\xa1" },
+	{ MA, 0, "[^a]", "aAbB" },
+	{ CMA, 0, "[^m]", "mMnN" },
+	{ MA, 0, "a[^b][^#]", "abacd" },
+	{ CMA, 0, "A[^B][^E]", "abacd" },
+	{ CMUA, 0, "[^x][^#]", "XxBll" },
+	{ MUA, 0, "[^a]", "aaa\xc3\xa1#Ab" },
+	{ CMUA, 0, "[^A]", "aA\xe6\x92\xad" },
+	{ MUA, 0, "\\W(\\W)?\\w", "\r\n+bc" },
+	{ MUA, 0, "\\W(\\W)?\\w", "\n\r+bc" },
+	{ MUA, 0, "\\W(\\W)?\\w", "\r\r+bc" },
+	{ MUA, 0, "\\W(\\W)?\\w", "\n\n+bc" },
+	{ MUA, 0, "[axd]", "sAXd" },
+	{ CMUA, 0, "[axd]", "sAXd" },
+	{ CMUA, 0 | F_NOMATCH, "[^axd]", "DxA" },
+	{ MUA, 0, "[a-dA-C]", "\xe6\x92\xad\xc3\xa9.B" },
+	{ MUA, 0, "[^a-dA-C]", "\xe6\x92\xad\xc3\xa9" },
+	{ CMUA, 0, "[^\xc3\xa9]", "\xc3\xa9\xc3\x89." },
+	{ MUA, 0, "[^\xc3\xa9]", "\xc3\xa9\xc3\x89." },
+	{ MUA, 0, "[^a]", "\xc2\x80[]" },
+	{ CMUA, 0, "\xf0\x90\x90\xa7", "\xf0\x90\x91\x8f" },
+	{ CMA, 0, "1a2b3c4", "1a2B3c51A2B3C4" },
+	{ PCRE_CASELESS, 0, "\xff#a", "\xff#\xff\xfe##\xff#A" },
+	{ PCRE_CASELESS, 0, "\xfe", "\xff\xfc#\xfe\xfe" },
+	{ PCRE_CASELESS, 0, "a1", "Aa1" },
+	{ MA, 0, "\\Ca", "cda" },
+	{ CMA, 0, "\\Ca", "CDA" },
+	{ MA, 0 | F_NOMATCH, "\\Cx", "cda" },
+	{ CMA, 0 | F_NOMATCH, "\\Cx", "CDA" },
+	{ CMUAP, 0, "\xf0\x90\x90\x80\xf0\x90\x90\xa8", "\xf0\x90\x90\xa8\xf0\x90\x90\x80" },
+	{ CMUAP, 0, "\xf0\x90\x90\x80{2}", "\xf0\x90\x90\x80#\xf0\x90\x90\xa8\xf0\x90\x90\x80" },
+	{ CMUAP, 0, "\xf0\x90\x90\xa8{2}", "\xf0\x90\x90\x80#\xf0\x90\x90\xa8\xf0\x90\x90\x80" },
+	{ CMUAP, 0, "\xe1\xbd\xb8\xe1\xbf\xb8", "\xe1\xbf\xb8\xe1\xbd\xb8" },
+
+	/* Assertions. */
+	{ MUA, 0, "\\b[^A]", "A_B#" },
+	{ MA, 0 | F_NOMATCH, "\\b\\W", "\n*" },
+	{ MUA, 0, "\\B[^,]\\b[^s]\\b", "#X" },
+	{ MAP, 0, "\\B", "_\xa1" },
+	{ MAP, 0, "\\b_\\b[,A]\\B", "_," },
+	{ MUAP, 0, "\\b", "\xe6\x92\xad!" },
+	{ MUAP, 0, "\\B", "_\xc2\xa1\xc3\xa1\xc2\x85" },
+	{ MUAP, 0, "\\b[^A]\\B[^c]\\b[^_]\\B", "_\xc3\xa1\xe2\x80\xa8" },
+	{ MUAP, 0, "\\b\\w+\\B", "\xc3\x89\xc2\xa1\xe6\x92\xad\xc3\x81\xc3\xa1" },
+	{ MUA, 0 | F_NOMATCH, "\\b.", "\xcd\xbe" },
+	{ CMUAP, 0, "\\By", "\xf0\x90\x90\xa8y" },
+	{ MA, 0 | F_NOMATCH, "\\R^", "\n" },
+	{ MA, 1 | F_NOMATCH, "^", "\n" },
+	{ 0, 0, "^ab", "ab" },
+	{ 0, 0 | F_NOMATCH, "^ab", "aab" },
+	{ PCRE_MULTILINE | PCRE_NEWLINE_CRLF, 0, "^a", "\r\raa\n\naa\r\naa" },
+	{ PCRE_MULTILINE | PCRE_UTF8 | PCRE_NEWLINE_ANYCRLF, 0, "^-", "\xe2\x80\xa8--\xc2\x85-\r\n-" },
+	{ PCRE_MULTILINE | PCRE_NEWLINE_ANY, 0, "^-", "a--b--\x85--" },
+	{ PCRE_MULTILINE | PCRE_UTF8 | PCRE_NEWLINE_ANY, 0, "^-", "a--\xe2\x80\xa8--" },
+	{ PCRE_MULTILINE | PCRE_UTF8 | PCRE_NEWLINE_ANY, 0, "^-", "a--\xc2\x85--" },
+	{ 0, 0, "ab$", "ab" },
+	{ 0, 0 | F_NOMATCH, "ab$", "abab\n\n" },
+	{ PCRE_DOLLAR_ENDONLY, 0 | F_NOMATCH, "ab$", "abab\r\n" },
+	{ PCRE_MULTILINE | PCRE_NEWLINE_CRLF, 0, "a$", "\r\raa\n\naa\r\naa" },
+	{ PCRE_MULTILINE | PCRE_NEWLINE_ANY, 0, "a$", "aaa" },
+	{ PCRE_MULTILINE | PCRE_UTF8 | PCRE_NEWLINE_ANYCRLF, 0, "#$", "#\xc2\x85###\r#" },
+	{ PCRE_MULTILINE | PCRE_UTF8 | PCRE_NEWLINE_ANY, 0, "#$", "#\xe2\x80\xa9" },
+	{ PCRE_NOTBOL | PCRE_NEWLINE_ANY, 0 | F_NOMATCH, "^a", "aa\naa" },
+	{ PCRE_NOTBOL | PCRE_MULTILINE | PCRE_NEWLINE_ANY, 0, "^a", "aa\naa" },
+	{ PCRE_NOTEOL | PCRE_NEWLINE_ANY, 0 | F_NOMATCH, "a$", "aa\naa" },
+	{ PCRE_NOTEOL | PCRE_NEWLINE_ANY, 0 | F_NOMATCH, "a$", "aa\r\n" },
+	{ PCRE_UTF8 | PCRE_DOLLAR_ENDONLY | PCRE_NEWLINE_ANY, 0 | F_PROPERTY, "\\p{Any}{2,}$", "aa\r\n" },
+	{ PCRE_NOTEOL | PCRE_MULTILINE | PCRE_NEWLINE_ANY, 0, "a$", "aa\naa" },
+	{ PCRE_NEWLINE_CR, 0, ".\\Z", "aaa" },
+	{ PCRE_NEWLINE_CR | PCRE_UTF8, 0, "a\\Z", "aaa\r" },
+	{ PCRE_NEWLINE_CR, 0, ".\\Z", "aaa\n" },
+	{ PCRE_NEWLINE_CRLF, 0, ".\\Z", "aaa\r" },
+	{ PCRE_NEWLINE_CRLF | PCRE_UTF8, 0, ".\\Z", "aaa\n" },
+	{ PCRE_NEWLINE_CRLF, 0, ".\\Z", "aaa\r\n" },
+	{ PCRE_NEWLINE_ANYCRLF | PCRE_UTF8, 0, ".\\Z", "aaa" },
+	{ PCRE_NEWLINE_ANYCRLF | PCRE_UTF8, 0, ".\\Z", "aaa\r" },
+	{ PCRE_NEWLINE_ANYCRLF | PCRE_UTF8, 0, ".\\Z", "aaa\n" },
+	{ PCRE_NEWLINE_ANYCRLF | PCRE_UTF8, 0, ".\\Z", "aaa\r\n" },
+	{ PCRE_NEWLINE_ANYCRLF | PCRE_UTF8, 0, ".\\Z", "aaa\xe2\x80\xa8" },
+	{ PCRE_NEWLINE_ANYCRLF | PCRE_UTF8, 0, ".\\Z", "aaa" },
+	{ PCRE_NEWLINE_ANYCRLF | PCRE_UTF8, 0, ".\\Z", "aaa\r" },
+	{ PCRE_NEWLINE_ANYCRLF | PCRE_UTF8, 0, ".\\Z", "aaa\n" },
+	{ PCRE_NEWLINE_ANYCRLF | PCRE_UTF8, 0, ".\\Z", "aaa\r\n" },
+	{ PCRE_NEWLINE_ANY | PCRE_UTF8, 0, ".\\Z", "aaa\xc2\x85" },
+	{ PCRE_NEWLINE_ANY | PCRE_UTF8, 0, ".\\Z", "aaa\xe2\x80\xa8" },
+	{ MA, 0, "\\Aa", "aaa" },
+	{ MA, 1 | F_NOMATCH, "\\Aa", "aaa" },
+	{ MA, 1, "\\Ga", "aaa" },
+	{ MA, 1 | F_NOMATCH, "\\Ga", "aba" },
+	{ MA, 0, "a\\z", "aaa" },
+	{ MA, 0 | F_NOMATCH, "a\\z", "aab" },
+
+	/* Brackets. */
+	{ MUA, 0, "(ab|bb|cd)", "bacde" },
+	{ MUA, 0, "(?:ab|a)(bc|c)", "ababc" },
+	{ MUA, 0, "((ab|(cc))|(bb)|(?:cd|efg))", "abac" },
+	{ CMUA, 0, "((aB|(Cc))|(bB)|(?:cd|EFg))", "AcCe" },
+	{ MUA, 0, "((ab|(cc))|(bb)|(?:cd|ebg))", "acebebg" },
+	{ MUA, 0, "(?:(a)|(?:b))(cc|(?:d|e))(a|b)k", "accabdbbccbk" },
+
+	/* Greedy and non-greedy ? operators. */
+	{ MUA, 0, "(?:a)?a", "laab" },
+	{ CMUA, 0, "(A)?A", "llaab" },
+	{ MUA, 0, "(a)?\?a", "aab" }, /* ?? is the prefix of trygraphs in GCC. */
+	{ MUA, 0, "(a)?a", "manm" },
+	{ CMUA, 0, "(a|b)?\?d((?:e)?)", "ABABdx" },
+	{ MUA, 0, "(a|b)?\?d((?:e)?)", "abcde" },
+	{ MUA, 0, "((?:ab)?\?g|b(?:g(nn|d)?\?)?)?\?(?:n)?m", "abgnbgnnbgdnmm" },
+
+	/* Greedy and non-greedy + operators */
+	{ MUA, 0, "(aa)+aa", "aaaaaaa" },
+	{ MUA, 0, "(aa)+?aa", "aaaaaaa" },
+	{ MUA, 0, "(?:aba|ab|a)+l", "ababamababal" },
+	{ MUA, 0, "(?:aba|ab|a)+?l", "ababamababal" },
+	{ MUA, 0, "(a(?:bc|cb|b|c)+?|ss)+e", "accssabccbcacbccbbXaccssabccbcacbccbbe" },
+	{ MUA, 0, "(a(?:bc|cb|b|c)+|ss)+?e", "accssabccbcacbccbbXaccssabccbcacbccbbe" },
+	{ MUA, 0, "(?:(b(c)+?)+)?\?(?:(bc)+|(cb)+)+(?:m)+", "bccbcccbcbccbcbPbccbcccbcbccbcbmmn" },
+
+	/* Greedy and non-greedy * operators */
+	{ CMUA, 0, "(?:AA)*AB", "aaaaaaamaaaaaaab" },
+	{ MUA, 0, "(?:aa)*?ab", "aaaaaaamaaaaaaab" },
+	{ MUA, 0, "(aa|ab)*ab", "aaabaaab" },
+	{ CMUA, 0, "(aa|Ab)*?aB", "aaabaaab" },
+	{ MUA, 0, "(a|b)*(?:a)*(?:b)*m", "abbbaaababanabbbaaababamm" },
+	{ MUA, 0, "(a|b)*?(?:a)*?(?:b)*?m", "abbbaaababanabbbaaababamm" },
+	{ MA, 0, "a(a(\\1*)a|(b)b+){0}a", "aa" },
+	{ MA, 0, "((?:a|)*){0}a", "a" },
+
+	/* Combining ? + * operators */
+	{ MUA, 0, "((bm)+)?\?(?:a)*(bm)+n|((am)+?)?(?:a)+(am)*n", "bmbmabmamaaamambmaman" },
+	{ MUA, 0, "(((ab)?cd)*ef)+g", "abcdcdefcdefefmabcdcdefcdefefgg" },
+	{ MUA, 0, "(((ab)?\?cd)*?ef)+?g", "abcdcdefcdefefmabcdcdefcdefefgg" },
+	{ MUA, 0, "(?:(ab)?c|(?:ab)+?d)*g", "ababcdccababddg" },
+	{ MUA, 0, "(?:(?:ab)?\?c|(ab)+d)*?g", "ababcdccababddg" },
+
+	/* Single character iterators. */
+	{ MUA, 0, "(a+aab)+aaaab", "aaaabcaaaabaabcaabcaaabaaaab" },
+	{ MUA, 0, "(a*a*aab)+x", "aaaaabaabaaabmaabx" },
+	{ MUA, 0, "(a*?(b|ab)a*?)+x", "aaaabcxbbaabaacbaaabaabax" },
+	{ MUA, 0, "(a+(ab|ad)a+)+x", "aaabaaaadaabaaabaaaadaaax" },
+	{ MUA, 0, "(a?(a)a?)+(aaa)", "abaaabaaaaaaaa" },
+	{ MUA, 0, "(a?\?(a)a?\?)+(b)", "aaaacaaacaacacbaaab" },
+	{ MUA, 0, "(a{0,4}(b))+d", "aaaaaabaabcaaaaabaaaaabd" },
+	{ MUA, 0, "(a{0,4}?[^b])+d+(a{0,4}[^b])d+", "aaaaadaaaacaadddaaddd" },
+	{ MUA, 0, "(ba{2})+c", "baabaaabacbaabaac" },
+	{ MUA, 0, "(a*+bc++)+", "aaabbcaaabcccab" },
+	{ MUA, 0, "(a?+[^b])+", "babaacacb" },
+	{ MUA, 0, "(a{0,3}+b)(a{0,3}+b)(a{0,3}+)[^c]", "abaabaaacbaabaaaac" },
+	{ CMUA, 0, "([a-c]+[d-f]+?)+?g", "aBdacdehAbDaFgA" },
+	{ CMUA, 0, "[c-f]+k", "DemmFke" },
+	{ MUA, 0, "([DGH]{0,4}M)+", "GGDGHDGMMHMDHHGHM" },
+	{ MUA, 0, "([a-c]{4,}s)+", "abasabbasbbaabsbba" },
+	{ CMUA, 0, "[ace]{3,7}", "AcbDAcEEcEd" },
+	{ CMUA, 0, "[ace]{3,7}?", "AcbDAcEEcEd" },
+	{ CMUA, 0, "[ace]{3,}", "AcbDAcEEcEd" },
+	{ CMUA, 0, "[ace]{3,}?", "AcbDAcEEcEd" },
+	{ MUA, 0, "[ckl]{2,}?g", "cdkkmlglglkcg" },
+	{ CMUA, 0, "[ace]{5}?", "AcCebDAcEEcEd" },
+	{ MUA, 0, "([AbC]{3,5}?d)+", "BACaAbbAEAACCbdCCbdCCAAbb" },
+	{ MUA, 0, "([^ab]{0,}s){2}", "abaabcdsABamsDDs" },
+	{ MUA, 0, "\\b\\w+\\B", "x,a_cd" },
+	{ MUAP, 0, "\\b[^\xc2\xa1]+\\B", "\xc3\x89\xc2\xa1\xe6\x92\xad\xc3\x81\xc3\xa1" },
+	{ CMUA, 0, "[^b]+(a*)([^c]?d{3})", "aaaaddd" },
+	{ CMUAP, 0, "\xe1\xbd\xb8{2}", "\xe1\xbf\xb8#\xe1\xbf\xb8\xe1\xbd\xb8" },
+	{ CMUA, 0, "[^\xf0\x90\x90\x80]{2,4}@", "\xf0\x90\x90\xa8\xf0\x90\x90\x80###\xf0\x90\x90\x80@@@" },
+	{ CMUA, 0, "[^\xe1\xbd\xb8][^\xc3\xa9]", "\xe1\xbd\xb8\xe1\xbf\xb8\xc3\xa9\xc3\x89#" },
+	{ MUA, 0, "[^\xe1\xbd\xb8][^\xc3\xa9]", "\xe1\xbd\xb8\xe1\xbf\xb8\xc3\xa9\xc3\x89#" },
+	{ MUA, 0, "[^\xe1\xbd\xb8]{3,}?", "##\xe1\xbd\xb8#\xe1\xbd\xb8#\xc3\x89#\xe1\xbd\xb8" },
+
+	/* Bracket repeats with limit. */
+	{ MUA, 0, "(?:(ab){2}){5}M", "abababababababababababM" },
+	{ MUA, 0, "(?:ab|abab){1,5}M", "abababababababababababM" },
+	{ MUA, 0, "(?>ab|abab){1,5}M", "abababababababababababM" },
+	{ MUA, 0, "(?:ab|abab){1,5}?M", "abababababababababababM" },
+	{ MUA, 0, "(?>ab|abab){1,5}?M", "abababababababababababM" },
+	{ MUA, 0, "(?:(ab){1,4}?){1,3}?M", "abababababababababababababM" },
+	{ MUA, 0, "(?:(ab){1,4}){1,3}abababababababababababM", "ababababababababababababM" },
+	{ MUA, 0 | F_NOMATCH, "(?:(ab){1,4}){1,3}abababababababababababM", "abababababababababababM" },
+	{ MUA, 0, "(ab){4,6}?M", "abababababababM" },
+
+	/* Basic character sets. */
+	{ MUA, 0, "(?:\\s)+(?:\\S)+", "ab \t\xc3\xa9\xe6\x92\xad " },
+	{ MUA, 0, "(\\w)*(k)(\\W)?\?", "abcdef abck11" },
+	{ MUA, 0, "\\((\\d)+\\)\\D", "a() (83 (8)2 (9)ab" },
+	{ MUA, 0, "\\w(\\s|(?:\\d)*,)+\\w\\wb", "a 5, 4,, bb 5, 4,, aab" },
+	{ MUA, 0, "(\\v+)(\\V+)", "\x0e\xc2\x85\xe2\x80\xa8\x0b\x09\xe2\x80\xa9" },
+	{ MUA, 0, "(\\h+)(\\H+)", "\xe2\x80\xa8\xe2\x80\x80\x20\xe2\x80\x8a\xe2\x81\x9f\xe3\x80\x80\x09\x20\xc2\xa0\x0a" },
+
+	/* Unicode properties. */
+	{ MUAP, 0, "[1-5\xc3\xa9\\w]", "\xc3\xa1_" },
+	{ MUAP, 0 | F_PROPERTY, "[\xc3\x81\\p{Ll}]", "A_\xc3\x89\xc3\xa1" },
+	{ MUAP, 0, "[\\Wd-h_x-z]+", "a\xc2\xa1#_yhzdxi" },
+	{ MUAP, 0 | F_NOMATCH | F_PROPERTY, "[\\P{Any}]", "abc" },
+	{ MUAP, 0 | F_NOMATCH | F_PROPERTY, "[^\\p{Any}]", "abc" },
+	{ MUAP, 0 | F_NOMATCH | F_PROPERTY, "[\\P{Any}\xc3\xa1-\xc3\xa8]", "abc" },
+	{ MUAP, 0 | F_NOMATCH | F_PROPERTY, "[^\\p{Any}\xc3\xa1-\xc3\xa8]", "abc" },
+	{ MUAP, 0 | F_NOMATCH | F_PROPERTY, "[\xc3\xa1-\xc3\xa8\\P{Any}]", "abc" },
+	{ MUAP, 0 | F_NOMATCH | F_PROPERTY, "[^\xc3\xa1-\xc3\xa8\\p{Any}]", "abc" },
+	{ MUAP, 0 | F_PROPERTY, "[\xc3\xa1-\xc3\xa8\\p{Any}]", "abc" },
+	{ MUAP, 0 | F_PROPERTY, "[^\xc3\xa1-\xc3\xa8\\P{Any}]", "abc" },
+	{ MUAP, 0, "[b-\xc3\xa9\\s]", "a\xc\xe6\x92\xad" },
+	{ CMUAP, 0, "[\xc2\x85-\xc2\x89\xc3\x89]", "\xc2\x84\xc3\xa9" },
+	{ MUAP, 0, "[^b-d^&\\s]{3,}", "db^ !a\xe2\x80\xa8_ae" },
+	{ MUAP, 0 | F_PROPERTY, "[^\\S\\P{Any}][\\sN]{1,3}[\\P{N}]{4}", "\xe2\x80\xaa\xa N\x9\xc3\xa9_0" },
+	{ MUA, 0 | F_PROPERTY, "[^\\P{L}\x9!D-F\xa]{2,3}", "\x9,.DF\xa.CG\xc3\x81" },
+	{ CMUAP, 0, "[\xc3\xa1-\xc3\xa9_\xe2\x80\xa0-\xe2\x80\xaf]{1,5}[^\xe2\x80\xa0-\xe2\x80\xaf]", "\xc2\xa1\xc3\x89\xc3\x89\xe2\x80\xaf_\xe2\x80\xa0" },
+	{ MUAP, 0 | F_PROPERTY, "[\xc3\xa2-\xc3\xa6\xc3\x81-\xc3\x84\xe2\x80\xa8-\xe2\x80\xa9\xe6\x92\xad\\p{Zs}]{2,}", "\xe2\x80\xa7\xe2\x80\xa9\xe6\x92\xad \xe6\x92\xae" },
+	{ MUAP, 0 | F_PROPERTY, "[\\P{L&}]{2}[^\xc2\x85-\xc2\x89\\p{Ll}\\p{Lu}]{2}", "\xc3\xa9\xe6\x92\xad.a\xe6\x92\xad|\xc2\x8a#" },
+	{ PCRE_UCP, 0, "[a-b\\s]{2,5}[^a]", "AB  baaa" },
+
+	/* Possible empty brackets. */
+	{ MUA, 0, "(?:|ab||bc|a)+d", "abcxabcabd" },
+	{ MUA, 0, "(|ab||bc|a)+d", "abcxabcabd" },
+	{ MUA, 0, "(?:|ab||bc|a)*d", "abcxabcabd" },
+	{ MUA, 0, "(|ab||bc|a)*d", "abcxabcabd" },
+	{ MUA, 0, "(?:|ab||bc|a)+?d", "abcxabcabd" },
+	{ MUA, 0, "(|ab||bc|a)+?d", "abcxabcabd" },
+	{ MUA, 0, "(?:|ab||bc|a)*?d", "abcxabcabd" },
+	{ MUA, 0, "(|ab||bc|a)*?d", "abcxabcabd" },
+	{ MUA, 0, "(((a)*?|(?:ba)+)+?|(?:|c|ca)*)*m", "abaacaccabacabalabaacaccabacabamm" },
+	{ MUA, 0, "(?:((?:a)*|(ba)+?)+|(|c|ca)*?)*?m", "abaacaccabacabalabaacaccabacabamm" },
+
+	/* Start offset. */
+	{ MUA, 3, "(\\d|(?:\\w)*\\w)+", "0ac01Hb" },
+	{ MUA, 4 | F_NOMATCH, "(\\w\\W\\w)+", "ab#d" },
+	{ MUA, 2 | F_NOMATCH, "(\\w\\W\\w)+", "ab#d" },
+	{ MUA, 1, "(\\w\\W\\w)+", "ab#d" },
+
+	/* Newline. */
+	{ PCRE_MULTILINE | PCRE_NEWLINE_CRLF, 0, "\\W{0,2}[^#]{3}", "\r\n#....." },
+	{ PCRE_MULTILINE | PCRE_NEWLINE_CR, 0, "\\W{0,2}[^#]{3}", "\r\n#....." },
+	{ PCRE_MULTILINE | PCRE_NEWLINE_CRLF, 0, "\\W{1,3}[^#]", "\r\n##...." },
+
+	/* Any character except newline or any newline. */
+	{ PCRE_NEWLINE_CRLF, 0, ".", "\r" },
+	{ PCRE_NEWLINE_CRLF | PCRE_UTF8, 0, ".(.).", "a\xc3\xa1\r\n\n\r\r" },
+	{ PCRE_NEWLINE_ANYCRLF, 0, ".(.)", "a\rb\nc\r\n\xc2\x85\xe2\x80\xa8" },
+	{ PCRE_NEWLINE_ANYCRLF | PCRE_UTF8, 0, ".(.)", "a\rb\nc\r\n\xc2\x85\xe2\x80\xa8" },
+	{ PCRE_NEWLINE_ANY | PCRE_UTF8, 0, "(.).", "a\rb\nc\r\n\xc2\x85\xe2\x80\xa9$de" },
+	{ PCRE_NEWLINE_ANYCRLF | PCRE_UTF8, 0 | F_NOMATCH, ".(.).", "\xe2\x80\xa8\nb\r" },
+	{ PCRE_NEWLINE_ANY, 0, "(.)(.)", "#\x85#\r#\n#\r\n#\x84" },
+	{ PCRE_NEWLINE_ANY | PCRE_UTF8, 0, "(.+)#", "#\rMn\xc2\x85#\n###" },
+	{ PCRE_BSR_ANYCRLF, 0, "\\R", "\r" },
+	{ PCRE_BSR_ANYCRLF, 0, "\\R", "\x85#\r\n#" },
+	{ PCRE_BSR_UNICODE | PCRE_UTF8, 0, "\\R", "ab\xe2\x80\xa8#c" },
+	{ PCRE_BSR_UNICODE | PCRE_UTF8, 0, "\\R", "ab\r\nc" },
+	{ PCRE_NEWLINE_CRLF | PCRE_BSR_UNICODE | PCRE_UTF8, 0, "(\\R.)+", "\xc2\x85\r\n#\xe2\x80\xa8\n\r\n\r" },
+	{ MUA, 0 | F_NOMATCH, "\\R+", "ab" },
+	{ MUA, 0, "\\R+", "ab\r\n\r" },
+	{ MUA, 0, "\\R*", "ab\r\n\r" },
+	{ MUA, 0, "\\R*", "\r\n\r" },
+	{ MUA, 0, "\\R{2,4}", "\r\nab\r\r" },
+	{ MUA, 0, "\\R{2,4}", "\r\nab\n\n\n\r\r\r" },
+	{ MUA, 0, "\\R{2,}", "\r\nab\n\n\n\r\r\r" },
+	{ MUA, 0, "\\R{0,3}", "\r\n\r\n\r\n\r\n\r\n" },
+	{ MUA, 0 | F_NOMATCH, "\\R+\\R\\R", "\r\n\r\n" },
+	{ MUA, 0, "\\R+\\R\\R", "\r\r\r" },
+	{ MUA, 0, "\\R*\\R\\R", "\n\r" },
+	{ MUA, 0 | F_NOMATCH, "\\R{2,4}\\R\\R", "\r\r\r" },
+	{ MUA, 0, "\\R{2,4}\\R\\R", "\r\r\r\r" },
+
+	/* Atomic groups (no fallback from "next" direction). */
+	{ MUA, 0 | F_NOMATCH, "(?>ab)ab", "bab" },
+	{ MUA, 0 | F_NOMATCH, "(?>(ab))ab", "bab" },
+	{ MUA, 0, "(?>ab)+abc(?>de)*def(?>gh)?ghe(?>ij)+?k(?>lm)*?n(?>op)?\?op",
+			"bababcdedefgheijijklmlmnop" },
+	{ MUA, 0, "(?>a(b)+a|(ab)?\?(b))an", "abban" },
+	{ MUA, 0, "(?>ab+a|(?:ab)?\?b)an", "abban" },
+	{ MUA, 0, "((?>ab|ad|)*?)(?>|c)*abad", "abababcababad" },
+	{ MUA, 0, "(?>(aa|b|)*+(?>(##)|###)*d|(aa)(?>(baa)?)m)", "aabaa#####da" },
+	{ MUA, 0, "((?>a|)+?)b", "aaacaaab" },
+	{ MUA, 0, "(?>x|)*$", "aaa" },
+	{ MUA, 0, "(?>(x)|)*$", "aaa" },
+	{ MUA, 0, "(?>x|())*$", "aaa" },
+	{ MUA, 0, "((?>[cxy]a|[a-d])*?)b", "aaa+ aaab" },
+	{ MUA, 0, "((?>[cxy](a)|[a-d])*?)b", "aaa+ aaab" },
+	{ MUA, 0, "(?>((?>(a+))))bab|(?>((?>(a+))))bb", "aaaabaaabaabab" },
+	{ MUA, 0, "(?>(?>a+))bab|(?>(?>a+))bb", "aaaabaaabaabab" },
+	{ MUA, 0, "(?>(a)c|(?>(c)|(a))a)b*?bab", "aaaabaaabaabab" },
+	{ MUA, 0, "(?>ac|(?>c|a)a)b*?bab", "aaaabaaabaabab" },
+	{ MUA, 0, "(?>(b)b|(a))*b(?>(c)|d)?x", "ababcaaabdbx" },
+	{ MUA, 0, "(?>bb|a)*b(?>c|d)?x", "ababcaaabdbx" },
+	{ MUA, 0, "(?>(bb)|a)*b(?>c|(d))?x", "ababcaaabdbx" },
+	{ MUA, 0, "(?>(a))*?(?>(a))+?(?>(a))??x", "aaaaaacccaaaaabax" },
+	{ MUA, 0, "(?>a)*?(?>a)+?(?>a)??x", "aaaaaacccaaaaabax" },
+	{ MUA, 0, "(?>(a)|)*?(?>(a)|)+?(?>(a)|)??x", "aaaaaacccaaaaabax" },
+	{ MUA, 0, "(?>a|)*?(?>a|)+?(?>a|)??x", "aaaaaacccaaaaabax" },
+	{ MUA, 0, "(?>a(?>(a{0,2}))*?b|aac)+b", "aaaaaaacaaaabaaaaacaaaabaacaaabb" },
+	{ CMA, 0, "(?>((?>a{32}|b+|(a*))?(?>c+|d*)?\?)+e)+?f", "aaccebbdde bbdaaaccebbdee bbdaaaccebbdeef" },
+	{ MUA, 0, "(?>(?:(?>aa|a||x)+?b|(?>aa|a||(x))+?c)?(?>[ad]{0,2})*?d)+d", "aaacdbaabdcabdbaaacd aacaabdbdcdcaaaadaabcbaadd" },
+	{ MUA, 0, "(?>(?:(?>aa|a||(x))+?b|(?>aa|a||x)+?c)?(?>[ad]{0,2})*?d)+d", "aaacdbaabdcabdbaaacd aacaabdbdcdcaaaadaabcbaadd" },
+	{ MUA, 0 | F_PROPERTY, "\\X", "\xcc\x8d\xcc\x8d" },
+	{ MUA, 0 | F_PROPERTY, "\\X", "\xcc\x8d\xcc\x8d#\xcc\x8d\xcc\x8d" },
+	{ MUA, 0 | F_PROPERTY, "\\X+..", "\xcc\x8d#\xcc\x8d#\xcc\x8d\xcc\x8d" },
+	{ MUA, 0 | F_PROPERTY, "\\X{2,4}", "abcdef" },
+	{ MUA, 0 | F_PROPERTY, "\\X{2,4}?", "abcdef" },
+	{ MUA, 0 | F_NOMATCH | F_PROPERTY, "\\X{2,4}..", "#\xcc\x8d##" },
+	{ MUA, 0 | F_PROPERTY, "\\X{2,4}..", "#\xcc\x8d#\xcc\x8d##" },
+	{ MUA, 0, "(c(ab)?+ab)+", "cabcababcab" },
+	{ MUA, 0, "(?>(a+)b)+aabab", "aaaabaaabaabab" },
+
+	/* Possessive quantifiers. */
+	{ MUA, 0, "(?:a|b)++m", "mababbaaxababbaam" },
+	{ MUA, 0, "(?:a|b)*+m", "mababbaaxababbaam" },
+	{ MUA, 0, "(?:a|b)*+m", "ababbaaxababbaam" },
+	{ MUA, 0, "(a|b)++m", "mababbaaxababbaam" },
+	{ MUA, 0, "(a|b)*+m", "mababbaaxababbaam" },
+	{ MUA, 0, "(a|b)*+m", "ababbaaxababbaam" },
+	{ MUA, 0, "(a|b(*ACCEPT))++m", "maaxab" },
+	{ MUA, 0, "(?:b*)++m", "bxbbxbbbxm" },
+	{ MUA, 0, "(?:b*)++m", "bxbbxbbbxbbm" },
+	{ MUA, 0, "(?:b*)*+m", "bxbbxbbbxm" },
+	{ MUA, 0, "(?:b*)*+m", "bxbbxbbbxbbm" },
+	{ MUA, 0, "(b*)++m", "bxbbxbbbxm" },
+	{ MUA, 0, "(b*)++m", "bxbbxbbbxbbm" },
+	{ MUA, 0, "(b*)*+m", "bxbbxbbbxm" },
+	{ MUA, 0, "(b*)*+m", "bxbbxbbbxbbm" },
+	{ MUA, 0, "(?:a|(b))++m", "mababbaaxababbaam" },
+	{ MUA, 0, "(?:(a)|b)*+m", "mababbaaxababbaam" },
+	{ MUA, 0, "(?:(a)|(b))*+m", "ababbaaxababbaam" },
+	{ MUA, 0, "(a|(b))++m", "mababbaaxababbaam" },
+	{ MUA, 0, "((a)|b)*+m", "mababbaaxababbaam" },
+	{ MUA, 0, "((a)|(b))*+m", "ababbaaxababbaam" },
+	{ MUA, 0, "(a|(b)(*ACCEPT))++m", "maaxab" },
+	{ MUA, 0, "(?:(b*))++m", "bxbbxbbbxm" },
+	{ MUA, 0, "(?:(b*))++m", "bxbbxbbbxbbm" },
+	{ MUA, 0, "(?:(b*))*+m", "bxbbxbbbxm" },
+	{ MUA, 0, "(?:(b*))*+m", "bxbbxbbbxbbm" },
+	{ MUA, 0, "((b*))++m", "bxbbxbbbxm" },
+	{ MUA, 0, "((b*))++m", "bxbbxbbbxbbm" },
+	{ MUA, 0, "((b*))*+m", "bxbbxbbbxm" },
+	{ MUA, 0, "((b*))*+m", "bxbbxbbbxbbm" },
+	{ MUA, 0 | F_NOMATCH, "(?>(b{2,4}))(?:(?:(aa|c))++m|(?:(aa|c))+n)", "bbaacaaccaaaacxbbbmbn" },
+	{ MUA, 0, "((?:b)++a)+(cd)*+m", "bbababbacdcdnbbababbacdcdm" },
+	{ MUA, 0, "((?:(b))++a)+((c)d)*+m", "bbababbacdcdnbbababbacdcdm" },
+	{ MUA, 0, "(?:(?:(?:ab)*+k)++(?:n(?:cd)++)*+)*+m", "ababkkXababkkabkncXababkkabkncdcdncdXababkkabkncdcdncdkkabkncdXababkkabkncdcdncdkkabkncdm" },
+	{ MUA, 0, "(?:((ab)*+(k))++(n(?:c(d))++)*+)*+m", "ababkkXababkkabkncXababkkabkncdcdncdXababkkabkncdcdncdkkabkncdXababkkabkncdcdncdkkabkncdm" },
+
+	/* Back references. */
+	{ MUA, 0, "(aa|bb)(\\1*)(ll|)(\\3*)bbbbbbc", "aaaaaabbbbbbbbc" },
+	{ CMUA, 0, "(aa|bb)(\\1+)(ll|)(\\3+)bbbbbbc", "bBbbBbCbBbbbBbbcbbBbbbBBbbC" },
+	{ CMA, 0, "(a{2,4})\\1", "AaAaaAaA" },
+	{ MUA, 0, "(aa|bb)(\\1?)aa(\\1?)(ll|)(\\4+)bbc", "aaaaaaaabbaabbbbaabbbbc" },
+	{ MUA, 0, "(aa|bb)(\\1{0,5})(ll|)(\\3{0,5})cc", "bbxxbbbbxxaaaaaaaaaaaaaaaacc" },
+	{ MUA, 0, "(aa|bb)(\\1{3,5})(ll|)(\\3{3,5})cc", "bbbbbbbbbbbbaaaaaaccbbbbbbbbbbbbbbcc" },
+	{ MUA, 0, "(aa|bb)(\\1{3,})(ll|)(\\3{3,})cc", "bbbbbbbbbbbbaaaaaaccbbbbbbbbbbbbbbcc" },
+	{ MUA, 0, "(\\w+)b(\\1+)c", "GabGaGaDbGaDGaDc" },
+	{ MUA, 0, "(?:(aa)|b)\\1?b", "bb" },
+	{ CMUA, 0, "(aa|bb)(\\1*?)aa(\\1+?)", "bBBbaaAAaaAAaa" },
+	{ MUA, 0, "(aa|bb)(\\1*?)(dd|)cc(\\3+?)", "aaaaaccdd" },
+	{ CMUA, 0, "(?:(aa|bb)(\\1?\?)cc){2}(\\1?\?)", "aAaABBbbAAaAcCaAcCaA" },
+	{ MUA, 0, "(?:(aa|bb)(\\1{3,5}?)){2}(dd|)(\\3{3,5}?)", "aaaaaabbbbbbbbbbaaaaaaaaaaaaaa" },
+	{ CMA, 0, "(?:(aa|bb)(\\1{3,}?)){2}(dd|)(\\3{3,}?)", "aaaaaabbbbbbbbbbaaaaaaaaaaaaaa" },
+	{ MUA, 0, "(?:(aa|bb)(\\1{0,3}?)){2}(dd|)(\\3{0,3}?)b(\\1{0,3}?)(\\1{0,3})", "aaaaaaaaaaaaaaabaaaaa" },
+	{ MUA, 0, "(a(?:\\1|)a){3}b", "aaaaaaaaaaab" },
+	{ MA, 0, "(a?)b(\\1\\1*\\1+\\1?\\1*?\\1+?\\1??\\1*+\\1++\\1?+\\1{4}\\1{3,5}\\1{4,}\\1{0,5}\\1{3,5}?\\1{4,}?\\1{0,5}?\\1{3,5}+\\1{4,}+\\1{0,5}+#){2}d", "bb#b##d" },
+	{ MUAP, 0 | F_PROPERTY, "(\\P{N})\\1{2,}", ".www." },
+	{ MUAP, 0 | F_PROPERTY, "(\\P{N})\\1{0,2}", "wwwww." },
+	{ MUAP, 0 | F_PROPERTY, "(\\P{N})\\1{1,2}ww", "wwww" },
+	{ MUAP, 0 | F_PROPERTY, "(\\P{N})\\1{1,2}ww", "wwwww" },
+	{ PCRE_UCP, 0 | F_PROPERTY, "(\\P{N})\\1{2,}", ".www." },
+	{ CMUAP, 0, "(\xf0\x90\x90\x80)\\1", "\xf0\x90\x90\xa8\xf0\x90\x90\xa8" },
+
+	/* Assertions. */
+	{ MUA, 0, "(?=xx|yy|zz)\\w{4}", "abczzdefg" },
+	{ MUA, 0, "(?=((\\w+)b){3}|ab)", "dbbbb ab" },
+	{ MUA, 0, "(?!ab|bc|cd)[a-z]{2}", "Xabcdef" },
+	{ MUA, 0, "(?<=aaa|aa|a)a", "aaa" },
+	{ MUA, 2, "(?<=aaa|aa|a)a", "aaa" },
+	{ MA, 0, "(?<=aaa|aa|a)a", "aaa" },
+	{ MA, 2, "(?<=aaa|aa|a)a", "aaa" },
+	{ MUA, 0, "(\\d{2})(?!\\w+c|(((\\w?)m){2}n)+|\\1)", "x5656" },
+	{ MUA, 0, "((?=((\\d{2,6}\\w){2,}))\\w{5,20}K){2,}", "567v09708K12l00M00 567v09708K12l00M00K45K" },
+	{ MUA, 0, "(?=(?:(?=\\S+a)\\w*(b)){3})\\w+\\d", "bba bbab nbbkba nbbkba0kl" },
+	{ MUA, 0, "(?>a(?>(b+))a(?=(..)))*?k", "acabbcabbaabacabaabbakk" },
+	{ MUA, 0, "((?(?=(a))a)+k)", "bbak" },
+	{ MUA, 0, "((?(?=a)a)+k)", "bbak" },
+	{ MUA, 0 | F_NOMATCH, "(?=(?>(a))m)amk", "a k" },
+	{ MUA, 0 | F_NOMATCH, "(?!(?>(a))m)amk", "a k" },
+	{ MUA, 0 | F_NOMATCH, "(?>(?=(a))am)amk", "a k" },
+	{ MUA, 0, "(?=(?>a|(?=(?>(b+))a|c)[a-c]+)*?m)[a-cm]+k", "aaam bbam baaambaam abbabba baaambaamk" },
+	{ MUA, 0, "(?> ?\?\\b(?(?=\\w{1,4}(a))m)\\w{0,8}bc){2,}?", "bca ssbc mabd ssbc mabc" },
+	{ MUA, 0, "(?:(?=ab)?[^n][^n])+m", "ababcdabcdcdabnababcdabcdcdabm" },
+	{ MUA, 0, "(?:(?=a(b))?[^n][^n])+m", "ababcdabcdcdabnababcdabcdcdabm" },
+	{ MUA, 0, "(?:(?=.(.))??\\1.)+m", "aabbbcbacccanaabbbcbacccam" },
+	{ MUA, 0, "(?:(?=.)??[a-c])+m", "abacdcbacacdcaccam" },
+	{ MUA, 0, "((?!a)?(?!([^a]))?)+$", "acbab" },
+	{ MUA, 0, "((?!a)?\?(?!([^a]))?\?)+$", "acbab" },
+
+	/* Not empty, ACCEPT, FAIL */
+	{ MUA | PCRE_NOTEMPTY, 0 | F_NOMATCH, "a*", "bcx" },
+	{ MUA | PCRE_NOTEMPTY, 0, "a*", "bcaad" },
+	{ MUA | PCRE_NOTEMPTY, 0, "a*?", "bcaad" },
+	{ MUA | PCRE_NOTEMPTY_ATSTART, 0, "a*", "bcaad" },
+	{ MUA, 0, "a(*ACCEPT)b", "ab" },
+	{ MUA | PCRE_NOTEMPTY, 0 | F_NOMATCH, "a*(*ACCEPT)b", "bcx" },
+	{ MUA | PCRE_NOTEMPTY, 0, "a*(*ACCEPT)b", "bcaad" },
+	{ MUA | PCRE_NOTEMPTY, 0, "a*?(*ACCEPT)b", "bcaad" },
+	{ MUA | PCRE_NOTEMPTY, 0 | F_NOMATCH, "(?:z|a*(*ACCEPT)b)", "bcx" },
+	{ MUA | PCRE_NOTEMPTY, 0, "(?:z|a*(*ACCEPT)b)", "bcaad" },
+	{ MUA | PCRE_NOTEMPTY, 0, "(?:z|a*?(*ACCEPT)b)", "bcaad" },
+	{ MUA | PCRE_NOTEMPTY_ATSTART, 0, "a*(*ACCEPT)b", "bcx" },
+	{ MUA | PCRE_NOTEMPTY_ATSTART, 0 | F_NOMATCH, "a*(*ACCEPT)b", "" },
+	{ MUA, 0, "((a(*ACCEPT)b))", "ab" },
+	{ MUA, 0, "(a(*FAIL)a|a)", "aaa" },
+	{ MUA, 0, "(?=ab(*ACCEPT)b)a", "ab" },
+	{ MUA, 0, "(?=(?:x|ab(*ACCEPT)b))", "ab" },
+	{ MUA, 0, "(?=(a(b(*ACCEPT)b)))a", "ab" },
+	{ MUA | PCRE_NOTEMPTY, 0, "(?=a*(*ACCEPT))c", "c" },
+
+	/* Conditional blocks. */
+	{ MUA, 0, "(?(?=(a))a|b)+k", "ababbalbbadabak" },
+	{ MUA, 0, "(?(?!(b))a|b)+k", "ababbalbbadabak" },
+	{ MUA, 0, "(?(?=a)a|b)+k", "ababbalbbadabak" },
+	{ MUA, 0, "(?(?!b)a|b)+k", "ababbalbbadabak" },
+	{ MUA, 0, "(?(?=(a))a*|b*)+k", "ababbalbbadabak" },
+	{ MUA, 0, "(?(?!(b))a*|b*)+k", "ababbalbbadabak" },
+	{ MUA, 0, "(?(?!(b))(?:aaaaaa|a)|(?:bbbbbb|b))+aaaak", "aaaaaaaaaaaaaa bbbbbbbbbbbbbbb aaaaaaak" },
+	{ MUA, 0, "(?(?!b)(?:aaaaaa|a)|(?:bbbbbb|b))+aaaak", "aaaaaaaaaaaaaa bbbbbbbbbbbbbbb aaaaaaak" },
+	{ MUA, 0 | F_DIFF, "(?(?!(b))(?:aaaaaa|a)|(?:bbbbbb|b))+bbbbk", "aaaaaaaaaaaaaa bbbbbbbbbbbbbbb bbbbbbbk" },
+	{ MUA, 0, "(?(?!b)(?:aaaaaa|a)|(?:bbbbbb|b))+bbbbk", "aaaaaaaaaaaaaa bbbbbbbbbbbbbbb bbbbbbbk" },
+	{ MUA, 0, "(?(?=a)a*|b*)+k", "ababbalbbadabak" },
+	{ MUA, 0, "(?(?!b)a*|b*)+k", "ababbalbbadabak" },
+	{ MUA, 0, "(?(?=a)ab)", "a" },
+	{ MUA, 0, "(?(?a)?(?Pb)?(?(Name)c|d)*l", "bc ddd abccabccl" },
+	{ MUA, 0, "(?Pa)?(?Pb)?(?(Name)c|d)+?dd", "bcabcacdb bdddd" },
+	{ MUA, 0, "(?Pa)?(?Pb)?(?(Name)c|d)+l", "ababccddabdbccd abcccl" },
+	{ MUA, 0, "((?:a|aa)(?(1)aaa))x", "aax" },
+
+	/* Set start of match. */
+	{ MUA, 0, "(?:\\Ka)*aaaab", "aaaaaaaa aaaaaaabb" },
+	{ MUA, 0, "(?>\\Ka\\Ka)*aaaab", "aaaaaaaa aaaaaaaaaabb" },
+	{ MUA, 0, "a+\\K(?<=\\Gaa)a", "aaaaaa" },
+	{ MUA | PCRE_NOTEMPTY, 0 | F_NOMATCH, "a\\K(*ACCEPT)b", "aa" },
+	{ MUA | PCRE_NOTEMPTY_ATSTART, 0, "a\\K(*ACCEPT)b", "aa" },
+
+	/* First line. */
+	{ MUA | PCRE_FIRSTLINE, 0 | F_PROPERTY, "\\p{Any}a", "bb\naaa" },
+	{ MUA | PCRE_FIRSTLINE, 0 | F_NOMATCH | F_PROPERTY, "\\p{Any}a", "bb\r\naaa" },
+	{ MUA | PCRE_FIRSTLINE, 0, "(?<=a)", "a" },
+	{ MUA | PCRE_FIRSTLINE, 0 | F_NOMATCH, "[^a][^b]", "ab" },
+	{ MUA | PCRE_FIRSTLINE, 0 | F_NOMATCH, "a", "\na" },
+	{ MUA | PCRE_FIRSTLINE, 0 | F_NOMATCH, "[abc]", "\na" },
+	{ MUA | PCRE_FIRSTLINE, 0 | F_NOMATCH, "^a", "\na" },
+	{ MUA | PCRE_FIRSTLINE, 0 | F_NOMATCH, "^(?<=\n)", "\na" },
+	{ MUA | PCRE_FIRSTLINE, 0, "\xf0\x90\x90\x80", "\xf0\x90\x90\x80" },
+	{ PCRE_MULTILINE | PCRE_UTF8 | PCRE_NEWLINE_ANY | PCRE_FIRSTLINE, 0 | F_NOMATCH, "#", "\xc2\x85#" },
+	{ PCRE_MULTILINE | PCRE_NEWLINE_ANY | PCRE_FIRSTLINE, 0 | F_NOMATCH, "#", "\x85#" },
+	{ PCRE_MULTILINE | PCRE_UTF8 | PCRE_NEWLINE_ANY | PCRE_FIRSTLINE, 0 | F_NOMATCH, "^#", "\xe2\x80\xa8#" },
+	{ PCRE_MULTILINE | PCRE_UTF8 | PCRE_NEWLINE_CRLF | PCRE_FIRSTLINE, 0 | F_PROPERTY, "\\p{Any}", "\r\na" },
+	{ PCRE_MULTILINE | PCRE_UTF8 | PCRE_NEWLINE_CRLF | PCRE_FIRSTLINE, 0, ".", "\r" },
+	{ PCRE_MULTILINE | PCRE_UTF8 | PCRE_NEWLINE_CRLF | PCRE_FIRSTLINE, 0, "a", "\ra" },
+	{ PCRE_MULTILINE | PCRE_UTF8 | PCRE_NEWLINE_CRLF | PCRE_FIRSTLINE, 0 | F_NOMATCH, "ba", "bbb\r\nba" },
+	{ PCRE_MULTILINE | PCRE_UTF8 | PCRE_NEWLINE_CRLF | PCRE_FIRSTLINE, 0 | F_NOMATCH | F_PROPERTY, "\\p{Any}{4}|a", "\r\na" },
+	{ PCRE_MULTILINE | PCRE_UTF8 | PCRE_NEWLINE_CRLF | PCRE_FIRSTLINE, 1, ".", "\r\n" },
+	{ PCRE_FIRSTLINE | PCRE_NEWLINE_LF | PCRE_DOTALL, 0 | F_NOMATCH, "ab.", "ab" },
+
+	/* Recurse. */
+	{ MUA, 0, "(a)(?1)", "aa" },
+	{ MUA, 0, "((a))(?1)", "aa" },
+	{ MUA, 0, "(b|a)(?1)", "aa" },
+	{ MUA, 0, "(b|(a))(?1)", "aa" },
+	{ MUA, 0 | F_NOMATCH, "((a)(b)(?:a*))(?1)", "aba" },
+	{ MUA, 0, "((a)(b)(?:a*))(?1)", "abab" },
+	{ MUA, 0, "((a+)c(?2))b(?1)", "aacaabaca" },
+	{ MUA, 0, "((?2)b|(a)){2}(?1)", "aabab" },
+	{ MUA, 0, "(?1)(a)*+(?2)(b(?1))", "aababa" },
+	{ MUA, 0, "(?1)(((a(*ACCEPT)))b)", "axaa" },
+	{ MUA, 0, "(?1)(?(DEFINE) (((ac(*ACCEPT)))b) )", "akaac" },
+	{ MUA, 0, "(a+)b(?1)b\\1", "abaaabaaaaa" },
+	{ MUA, 0 | F_NOMATCH, "(?(DEFINE)(aa|a))(?1)ab", "aab" },
+	{ MUA, 0, "(?(DEFINE)(a\\Kb))(?1)+ababc", "abababxabababc" },
+	{ MUA, 0, "(a\\Kb)(?1)+ababc", "abababxababababc" },
+	{ MUA, 0 | F_NOMATCH, "(a\\Kb)(?1)+ababc", "abababxababababxc" },
+	{ MUA, 0, "b|<(?R)*>", "<" },
+	{ MUA, 0, "(a\\K){0}(?:(?1)b|ac)", "ac" },
+	{ MUA, 0, "(?(DEFINE)(a(?2)|b)(b(?1)|(a)))(?:(?1)|(?2))m", "ababababnababababaam" },
+	{ MUA, 0, "(a)((?(R)a|b))(?2)", "aabbabaa" },
+	{ MUA, 0, "(a)((?(R2)a|b))(?2)", "aabbabaa" },
+	{ MUA, 0, "(a)((?(R1)a|b))(?2)", "ababba" },
+	{ MUA, 0, "(?(R0)aa|bb(?R))", "abba aabb bbaa" },
+	{ MUA, 0, "((?(R)(?:aaaa|a)|(?:(aaaa)|(a)))+)(?1)$", "aaaaaaaaaa aaaa" },
+	{ MUA, 0, "(?Pa(?(R&Name)a|b))(?1)", "aab abb abaa" },
+	{ MUA, 0, "((?(R)a|(?1)){3})", "XaaaaaaaaaX" },
+	{ MUA, 0, "((?:(?(R)a|(?1))){3})", "XaaaaaaaaaX" },
+	{ MUA, 0, "((?(R)a|(?1)){1,3})aaaaaa", "aaaaaaaaXaaaaaaaaa" },
+	{ MUA, 0, "((?(R)a|(?1)){1,3}?)M", "aaaM" },
+
+	/* 16 bit specific tests. */
+	{ CMA, 0 | F_FORCECONV, "\xc3\xa1", "\xc3\x81\xc3\xa1" },
+	{ CMA, 0 | F_FORCECONV, "\xe1\xbd\xb8", "\xe1\xbf\xb8\xe1\xbd\xb8" },
+	{ CMA, 0 | F_FORCECONV, "[\xc3\xa1]", "\xc3\x81\xc3\xa1" },
+	{ CMA, 0 | F_FORCECONV, "[\xe1\xbd\xb8]", "\xe1\xbf\xb8\xe1\xbd\xb8" },
+	{ CMA, 0 | F_FORCECONV, "[a-\xed\xb0\x80]", "A" },
+	{ CMA, 0 | F_NO8 | F_FORCECONV, "[a-\\x{dc00}]", "B" },
+	{ CMA, 0 | F_NO8 | F_NOMATCH | F_FORCECONV, "[b-\\x{dc00}]", "a" },
+	{ CMA, 0 | F_NO8 | F_FORCECONV, "\xed\xa0\x80\\x{d800}\xed\xb0\x80\\x{dc00}", "\xed\xa0\x80\xed\xa0\x80\xed\xb0\x80\xed\xb0\x80" },
+	{ CMA, 0 | F_NO8 | F_FORCECONV, "[\xed\xa0\x80\\x{d800}]{1,2}?[\xed\xb0\x80\\x{dc00}]{1,2}?#", "\xed\xa0\x80\xed\xa0\x80\xed\xb0\x80\xed\xb0\x80#" },
+	{ CMA, 0 | F_FORCECONV, "[\xed\xa0\x80\xed\xb0\x80#]{0,3}(?<=\xed\xb0\x80.)", "\xed\xa0\x80#\xed\xa0\x80##\xed\xb0\x80\xed\xa0\x80" },
+	{ CMA, 0 | F_FORCECONV, "[\xed\xa0\x80-\xed\xb3\xbf]", "\xed\x9f\xbf\xed\xa0\x83" },
+	{ CMA, 0 | F_FORCECONV, "[\xed\xa0\x80-\xed\xb3\xbf]", "\xed\xb4\x80\xed\xb3\xb0" },
+	{ CMA, 0 | F_NO8 | F_FORCECONV, "[\\x{d800}-\\x{dcff}]", "\xed\x9f\xbf\xed\xa0\x83" },
+	{ CMA, 0 | F_NO8 | F_FORCECONV, "[\\x{d800}-\\x{dcff}]", "\xed\xb4\x80\xed\xb3\xb0" },
+	{ CMA, 0 | F_FORCECONV, "[\xed\xa0\x80-\xef\xbf\xbf]+[\x1-\xed\xb0\x80]+#", "\xed\xa0\x85\xc3\x81\xed\xa0\x85\xef\xbf\xb0\xc2\x85\xed\xa9\x89#" },
+	{ CMA, 0 | F_FORCECONV, "[\xed\xa0\x80][\xed\xb0\x80]{2,}", "\xed\xa0\x80\xed\xb0\x80\xed\xa0\x80\xed\xb0\x80\xed\xb0\x80\xed\xb0\x80" },
+	{ MA, 0 | F_FORCECONV, "[^\xed\xb0\x80]{3,}?", "##\xed\xb0\x80#\xed\xb0\x80#\xc3\x89#\xed\xb0\x80" },
+	{ MA, 0 | F_NO8 | F_FORCECONV, "[^\\x{dc00}]{3,}?", "##\xed\xb0\x80#\xed\xb0\x80#\xc3\x89#\xed\xb0\x80" },
+	{ CMA, 0 | F_FORCECONV, ".\\B.", "\xed\xa0\x80\xed\xb0\x80" },
+	{ CMA, 0 | F_FORCECONV, "\\D+(?:\\d+|.)\\S+(?:\\s+|.)\\W+(?:\\w+|.)\xed\xa0\x80\xed\xa0\x80", "\xed\xa0\x80\xed\xa0\x80\xed\xa0\x80\xed\xa0\x80\xed\xa0\x80\xed\xa0\x80\xed\xa0\x80\xed\xa0\x80" },
+	{ CMA, 0 | F_FORCECONV, "\\d*\\s*\\w*\xed\xa0\x80\xed\xa0\x80", "\xed\xa0\x80\xed\xa0\x80" },
+	{ CMA, 0 | F_FORCECONV | F_NOMATCH, "\\d*?\\D*?\\s*?\\S*?\\w*?\\W*?##", "\xed\xa0\x80\xed\xa0\x80\xed\xa0\x80\xed\xa0\x80#" },
+	{ CMA | PCRE_EXTENDED, 0 | F_FORCECONV, "\xed\xa0\x80 \xed\xb0\x80 !", "\xed\xa0\x80\xed\xb0\x80!" },
+	{ CMA, 0 | F_FORCECONV, "\xed\xa0\x80+#[^#]+\xed\xa0\x80", "\xed\xa0\x80#a\xed\xa0\x80" },
+	{ CMA, 0 | F_FORCECONV, "(\xed\xa0\x80+)#\\1", "\xed\xa0\x80\xed\xa0\x80#\xed\xa0\x80\xed\xa0\x80" },
+	{ PCRE_MULTILINE | PCRE_NEWLINE_ANY, 0 | F_NO8 | F_FORCECONV, "^-", "a--\xe2\x80\xa8--" },
+	{ PCRE_BSR_UNICODE, 0 | F_NO8 | F_FORCECONV, "\\R", "ab\xe2\x80\xa8" },
+	{ 0, 0 | F_NO8 | F_FORCECONV, "\\v", "ab\xe2\x80\xa9" },
+	{ 0, 0 | F_NO8 | F_FORCECONV, "\\h", "ab\xe1\xa0\x8e" },
+	{ 0, 0 | F_NO8 | F_FORCECONV, "\\v+?\\V+?#", "\xe2\x80\xa9\xe2\x80\xa9\xef\xbf\xbf\xef\xbf\xbf#" },
+	{ 0, 0 | F_NO8 | F_FORCECONV, "\\h+?\\H+?#", "\xe1\xa0\x8e\xe1\xa0\x8e\xef\xbf\xbf\xef\xbf\xbf#" },
+
+	/* Partial matching. */
+	{ MUA | PCRE_PARTIAL_SOFT, 0, "ab", "a" },
+	{ MUA | PCRE_PARTIAL_SOFT, 0, "ab|a", "a" },
+	{ MUA | PCRE_PARTIAL_HARD, 0, "ab|a", "a" },
+	{ MUA | PCRE_PARTIAL_SOFT, 0, "\\b#", "a" },
+	{ MUA | PCRE_PARTIAL_SOFT, 0, "(?<=a)b", "a" },
+	{ MUA | PCRE_PARTIAL_SOFT, 0, "abc|(?<=xxa)bc", "xxab" },
+	{ MUA | PCRE_PARTIAL_SOFT, 0, "a\\B", "a" },
+	{ MUA | PCRE_PARTIAL_HARD, 0, "a\\b", "a" },
+
+	/* (*MARK) verb. */
+	{ MUA, 0, "a(*MARK:aa)a", "ababaa" },
+	{ MUA, 0 | F_NOMATCH, "a(*:aa)a", "abab" },
+	{ MUA, 0, "a(*:aa)(b(*:bb)b|bc)", "abc" },
+	{ MUA, 0 | F_NOMATCH, "a(*:1)x|b(*:2)y", "abc" },
+	{ MUA, 0, "(?>a(*:aa))b|ac", "ac" },
+	{ MUA, 0, "(?(DEFINE)(a(*:aa)))(?1)", "a" },
+	{ MUA, 0 | F_NOMATCH, "(?(DEFINE)((a)(*:aa)))(?1)b", "aa" },
+	{ MUA, 0, "(?(DEFINE)(a(*:aa)))a(?1)b|aac", "aac" },
+	{ MUA, 0, "(a(*:aa)){0}(?:b(?1)b|c)+c", "babbab cc" },
+	{ MUA, 0, "(a(*:aa)){0}(?:b(?1)b)+", "babba" },
+	{ MUA, 0 | F_NOMATCH | F_STUDY, "(a(*:aa)){0}(?:b(?1)b)+", "ba" },
+	{ MUA, 0, "(a\\K(*:aa)){0}(?:b(?1)b|c)+c", "babbab cc" },
+	{ MUA, 0, "(a\\K(*:aa)){0}(?:b(?1)b)+", "babba" },
+	{ MUA, 0 | F_NOMATCH | F_STUDY, "(a\\K(*:aa)){0}(?:b(?1)b)+", "ba" },
+	{ MUA, 0 | F_NOMATCH | F_STUDY, "(*:mark)m", "a" },
+
+	/* (*COMMIT) verb. */
+	{ MUA, 0 | F_NOMATCH, "a(*COMMIT)b", "ac" },
+	{ MUA, 0, "aa(*COMMIT)b", "xaxaab" },
+	{ MUA, 0 | F_NOMATCH, "a(*COMMIT)(*:msg)b|ac", "ac" },
+	{ MUA, 0 | F_NOMATCH, "(a(*COMMIT)b)++", "abac" },
+	{ MUA, 0 | F_NOMATCH, "((a)(*COMMIT)b)++", "abac" },
+	{ MUA, 0 | F_NOMATCH, "(?=a(*COMMIT)b)ab|ad", "ad" },
+
+	/* (*PRUNE) verb. */
+	{ MUA, 0, "aa\\K(*PRUNE)b", "aaab" },
+	{ MUA, 0, "aa(*PRUNE:bb)b|a", "aa" },
+	{ MUA, 0, "(a)(a)(*PRUNE)b|(a)", "aa" },
+	{ MUA, 0, "(a)(a)(a)(a)(a)(a)(a)(a)(*PRUNE)b|(a)", "aaaaaaaa" },
+	{ MUA | PCRE_PARTIAL_SOFT, 0, "a(*PRUNE)a|", "a" },
+	{ MUA | PCRE_PARTIAL_SOFT, 0, "a(*PRUNE)a|m", "a" },
+	{ MUA, 0 | F_NOMATCH, "(?=a(*PRUNE)b)ab|ad", "ad" },
+	{ MUA, 0, "a(*COMMIT)(*PRUNE)d|bc", "abc" },
+	{ MUA, 0, "(?=a(*COMMIT)b)a(*PRUNE)c|bc", "abc" },
+	{ MUA, 0 | F_NOMATCH, "(*COMMIT)(?=a(*COMMIT)b)a(*PRUNE)c|bc", "abc" },
+	{ MUA, 0, "(?=(a)(*COMMIT)b)a(*PRUNE)c|bc", "abc" },
+	{ MUA, 0 | F_NOMATCH, "(*COMMIT)(?=(a)(*COMMIT)b)a(*PRUNE)c|bc", "abc" },
+	{ MUA, 0, "(a(*COMMIT)b){0}a(?1)(*PRUNE)c|bc", "abc" },
+	{ MUA, 0 | F_NOMATCH, "(a(*COMMIT)b){0}a(*COMMIT)(?1)(*PRUNE)c|bc", "abc" },
+	{ MUA, 0, "(a(*COMMIT)b)++(*PRUNE)d|c", "ababc" },
+	{ MUA, 0 | F_NOMATCH, "(*COMMIT)(a(*COMMIT)b)++(*PRUNE)d|c", "ababc" },
+	{ MUA, 0, "((a)(*COMMIT)b)++(*PRUNE)d|c", "ababc" },
+	{ MUA, 0 | F_NOMATCH, "(*COMMIT)((a)(*COMMIT)b)++(*PRUNE)d|c", "ababc" },
+	{ MUA, 0, "(?>a(*COMMIT)b)*abab(*PRUNE)d|ba", "ababab" },
+	{ MUA, 0 | F_NOMATCH, "(*COMMIT)(?>a(*COMMIT)b)*abab(*PRUNE)d|ba", "ababab" },
+	{ MUA, 0, "(?>a(*COMMIT)b)+abab(*PRUNE)d|ba", "ababab" },
+	{ MUA, 0 | F_NOMATCH, "(*COMMIT)(?>a(*COMMIT)b)+abab(*PRUNE)d|ba", "ababab" },
+	{ MUA, 0, "(?>a(*COMMIT)b)?ab(*PRUNE)d|ba", "aba" },
+	{ MUA, 0 | F_NOMATCH, "(*COMMIT)(?>a(*COMMIT)b)?ab(*PRUNE)d|ba", "aba" },
+	{ MUA, 0, "(?>a(*COMMIT)b)*?n(*PRUNE)d|ba", "abababn" },
+	{ MUA, 0 | F_NOMATCH, "(*COMMIT)(?>a(*COMMIT)b)*?n(*PRUNE)d|ba", "abababn" },
+	{ MUA, 0, "(?>a(*COMMIT)b)+?n(*PRUNE)d|ba", "abababn" },
+	{ MUA, 0 | F_NOMATCH, "(*COMMIT)(?>a(*COMMIT)b)+?n(*PRUNE)d|ba", "abababn" },
+	{ MUA, 0, "(?>a(*COMMIT)b)??n(*PRUNE)d|bn", "abn" },
+	{ MUA, 0 | F_NOMATCH, "(*COMMIT)(?>a(*COMMIT)b)??n(*PRUNE)d|bn", "abn" },
+
+	/* (*SKIP) verb. */
+	{ MUA, 0 | F_NOMATCH, "(?=a(*SKIP)b)ab|ad", "ad" },
+
+	/* (*THEN) verb. */
+	{ MUA, 0, "((?:a(*THEN)|aab)(*THEN)c|a+)+m", "aabcaabcaabcaabcnacm" },
+	{ MUA, 0 | F_NOMATCH, "((?:a(*THEN)|aab)(*THEN)c|a+)+m", "aabcm" },
+	{ MUA, 0, "((?:a(*THEN)|aab)c|a+)+m", "aabcaabcnmaabcaabcm" },
+	{ MUA, 0, "((?:a|aab)(*THEN)c|a+)+m", "aam" },
+	{ MUA, 0, "((?:a(*COMMIT)|aab)(*THEN)c|a+)+m", "aam" },
+	{ MUA, 0, "(?(?=a(*THEN)b)ab|ad)", "ad" },
+	{ MUA, 0, "(?(?!a(*THEN)b)ad|add)", "add" },
+	{ MUA, 0 | F_NOMATCH, "(?(?=a)a(*THEN)b|ad)", "ad" },
+	{ MUA, 0, "(?!(?(?=a)ab|b(*THEN)d))bn|bnn", "bnn" },
+
+	/* Deep recursion. */
+	{ MUA, 0, "((((?:(?:(?:\\w)+)?)*|(?>\\w)+?)+|(?>\\w)?\?)*)?\\s", "aaaaa+ " },
+	{ MUA, 0, "(?:((?:(?:(?:\\w*?)+)??|(?>\\w)?|\\w*+)*)+)+?\\s", "aa+ " },
+	{ MUA, 0, "((a?)+)+b", "aaaaaaaaaaaa b" },
+
+	/* Deep recursion: Stack limit reached. */
+	{ MA, 0 | F_NOMATCH, "a?a?a?a?a?a?a?a?a?a?a?a?a?a?a?a?a?a?a?a?a?a?a?aaaaaaaaaaaaaaaaaaaaaaa", "aaaaaaaaaaaaaaaaaaaaaaa" },
+	{ MA, 0 | F_NOMATCH, "(?:a+)+b", "aaaaaaaaaaaaaaaaaaaaaaaa b" },
+	{ MA, 0 | F_NOMATCH, "(?:a+?)+?b", "aaaaaaaaaaaaaaaaaaaaaaaa b" },
+	{ MA, 0 | F_NOMATCH, "(?:a*)*b", "aaaaaaaaaaaaaaaaaaaaaaaa b" },
+	{ MA, 0 | F_NOMATCH, "(?:a*?)*?b", "aaaaaaaaaaaaaaaaaaaaaaaa b" },
+
+	{ 0, 0, NULL, NULL }
+};
+
+static const unsigned char *tables(int mode)
+{
+	/* The purpose of this function to allow valgrind
+	for reporting invalid reads and writes. */
+	static unsigned char *tables_copy;
+	const char *errorptr;
+	int erroroffset;
+	unsigned char *default_tables;
+#if defined SUPPORT_PCRE8
+	pcre *regex;
+	char null_str[1] = { 0 };
+#elif defined SUPPORT_PCRE16
+	pcre16 *regex;
+	PCRE_UCHAR16 null_str[1] = { 0 };
+#elif defined SUPPORT_PCRE32
+	pcre32 *regex;
+	PCRE_UCHAR32 null_str[1] = { 0 };
+#endif
+
+	if (mode) {
+		if (tables_copy)
+			free(tables_copy);
+		tables_copy = NULL;
+		return NULL;
+	}
+
+	if (tables_copy)
+		return tables_copy;
+
+	default_tables = NULL;
+#if defined SUPPORT_PCRE8
+	regex = pcre_compile(null_str, 0, &errorptr, &erroroffset, NULL);
+	if (regex) {
+		pcre_fullinfo(regex, NULL, PCRE_INFO_DEFAULT_TABLES, &default_tables);
+		pcre_free(regex);
+	}
+#elif defined SUPPORT_PCRE16
+	regex = pcre16_compile(null_str, 0, &errorptr, &erroroffset, NULL);
+	if (regex) {
+		pcre16_fullinfo(regex, NULL, PCRE_INFO_DEFAULT_TABLES, &default_tables);
+		pcre16_free(regex);
+	}
+#elif defined SUPPORT_PCRE32
+	regex = pcre32_compile(null_str, 0, &errorptr, &erroroffset, NULL);
+	if (regex) {
+		pcre32_fullinfo(regex, NULL, PCRE_INFO_DEFAULT_TABLES, &default_tables);
+		pcre32_free(regex);
+	}
+#endif
+	/* Shouldn't ever happen. */
+	if (!default_tables)
+		return NULL;
+
+	/* Unfortunately this value cannot get from pcre_fullinfo.
+	Since this is a test program, this is acceptable at the moment. */
+	tables_copy = (unsigned char *)malloc(1088);
+	if (!tables_copy)
+		return NULL;
+
+	memcpy(tables_copy, default_tables, 1088);
+	return tables_copy;
+}
+
+#ifdef SUPPORT_PCRE8
+static pcre_jit_stack* callback8(void *arg)
+{
+	return (pcre_jit_stack *)arg;
+}
+#endif
+
+#ifdef SUPPORT_PCRE16
+static pcre16_jit_stack* callback16(void *arg)
+{
+	return (pcre16_jit_stack *)arg;
+}
+#endif
+
+#ifdef SUPPORT_PCRE32
+static pcre32_jit_stack* callback32(void *arg)
+{
+	return (pcre32_jit_stack *)arg;
+}
+#endif
+
+#ifdef SUPPORT_PCRE8
+static pcre_jit_stack *stack8;
+
+static pcre_jit_stack *getstack8(void)
+{
+	if (!stack8)
+		stack8 = pcre_jit_stack_alloc(1, 1024 * 1024);
+	return stack8;
+}
+
+static void setstack8(pcre_extra *extra)
+{
+	if (!extra) {
+		if (stack8)
+			pcre_jit_stack_free(stack8);
+		stack8 = NULL;
+		return;
+	}
+
+	pcre_assign_jit_stack(extra, callback8, getstack8());
+}
+#endif /* SUPPORT_PCRE8 */
+
+#ifdef SUPPORT_PCRE16
+static pcre16_jit_stack *stack16;
+
+static pcre16_jit_stack *getstack16(void)
+{
+	if (!stack16)
+		stack16 = pcre16_jit_stack_alloc(1, 1024 * 1024);
+	return stack16;
+}
+
+static void setstack16(pcre16_extra *extra)
+{
+	if (!extra) {
+		if (stack16)
+			pcre16_jit_stack_free(stack16);
+		stack16 = NULL;
+		return;
+	}
+
+	pcre16_assign_jit_stack(extra, callback16, getstack16());
+}
+#endif /* SUPPORT_PCRE8 */
+
+#ifdef SUPPORT_PCRE32
+static pcre32_jit_stack *stack32;
+
+static pcre32_jit_stack *getstack32(void)
+{
+	if (!stack32)
+		stack32 = pcre32_jit_stack_alloc(1, 1024 * 1024);
+	return stack32;
+}
+
+static void setstack32(pcre32_extra *extra)
+{
+	if (!extra) {
+		if (stack32)
+			pcre32_jit_stack_free(stack32);
+		stack32 = NULL;
+		return;
+	}
+
+	pcre32_assign_jit_stack(extra, callback32, getstack32());
+}
+#endif /* SUPPORT_PCRE8 */
+
+#ifdef SUPPORT_PCRE16
+
+static int convert_utf8_to_utf16(const char *input, PCRE_UCHAR16 *output, int *offsetmap, int max_length)
+{
+	unsigned char *iptr = (unsigned char*)input;
+	PCRE_UCHAR16 *optr = output;
+	unsigned int c;
+
+	if (max_length == 0)
+		return 0;
+
+	while (*iptr && max_length > 1) {
+		c = 0;
+		if (offsetmap)
+			*offsetmap++ = (int)(iptr - (unsigned char*)input);
+
+		if (!(*iptr & 0x80))
+			c = *iptr++;
+		else if (!(*iptr & 0x20)) {
+			c = ((iptr[0] & 0x1f) << 6) | (iptr[1] & 0x3f);
+			iptr += 2;
+		} else if (!(*iptr & 0x10)) {
+			c = ((iptr[0] & 0x0f) << 12) | ((iptr[1] & 0x3f) << 6) | (iptr[2] & 0x3f);
+			iptr += 3;
+		} else if (!(*iptr & 0x08)) {
+			c = ((iptr[0] & 0x07) << 18) | ((iptr[1] & 0x3f) << 12) | ((iptr[2] & 0x3f) << 6) | (iptr[3] & 0x3f);
+			iptr += 4;
+		}
+
+		if (c < 65536) {
+			*optr++ = c;
+			max_length--;
+		} else if (max_length <= 2) {
+			*optr = '\0';
+			return (int)(optr - output);
+		} else {
+			c -= 0x10000;
+			*optr++ = 0xd800 | ((c >> 10) & 0x3ff);
+			*optr++ = 0xdc00 | (c & 0x3ff);
+			max_length -= 2;
+			if (offsetmap)
+				offsetmap++;
+		}
+	}
+	if (offsetmap)
+		*offsetmap = (int)(iptr - (unsigned char*)input);
+	*optr = '\0';
+	return (int)(optr - output);
+}
+
+static int copy_char8_to_char16(const char *input, PCRE_UCHAR16 *output, int max_length)
+{
+	unsigned char *iptr = (unsigned char*)input;
+	PCRE_UCHAR16 *optr = output;
+
+	if (max_length == 0)
+		return 0;
+
+	while (*iptr && max_length > 1) {
+		*optr++ = *iptr++;
+		max_length--;
+	}
+	*optr = '\0';
+	return (int)(optr - output);
+}
+
+#define REGTEST_MAX_LENGTH16 4096
+static PCRE_UCHAR16 regtest_buf16[REGTEST_MAX_LENGTH16];
+static int regtest_offsetmap16[REGTEST_MAX_LENGTH16];
+
+#endif /* SUPPORT_PCRE16 */
+
+#ifdef SUPPORT_PCRE32
+
+static int convert_utf8_to_utf32(const char *input, PCRE_UCHAR32 *output, int *offsetmap, int max_length)
+{
+	unsigned char *iptr = (unsigned char*)input;
+	PCRE_UCHAR32 *optr = output;
+	unsigned int c;
+
+	if (max_length == 0)
+		return 0;
+
+	while (*iptr && max_length > 1) {
+		c = 0;
+		if (offsetmap)
+			*offsetmap++ = (int)(iptr - (unsigned char*)input);
+
+		if (!(*iptr & 0x80))
+			c = *iptr++;
+		else if (!(*iptr & 0x20)) {
+			c = ((iptr[0] & 0x1f) << 6) | (iptr[1] & 0x3f);
+			iptr += 2;
+		} else if (!(*iptr & 0x10)) {
+			c = ((iptr[0] & 0x0f) << 12) | ((iptr[1] & 0x3f) << 6) | (iptr[2] & 0x3f);
+			iptr += 3;
+		} else if (!(*iptr & 0x08)) {
+			c = ((iptr[0] & 0x07) << 18) | ((iptr[1] & 0x3f) << 12) | ((iptr[2] & 0x3f) << 6) | (iptr[3] & 0x3f);
+			iptr += 4;
+		}
+
+		*optr++ = c;
+		max_length--;
+	}
+	if (offsetmap)
+		*offsetmap = (int)(iptr - (unsigned char*)input);
+	*optr = 0;
+	return (int)(optr - output);
+}
+
+static int copy_char8_to_char32(const char *input, PCRE_UCHAR32 *output, int max_length)
+{
+	unsigned char *iptr = (unsigned char*)input;
+	PCRE_UCHAR32 *optr = output;
+
+	if (max_length == 0)
+		return 0;
+
+	while (*iptr && max_length > 1) {
+		*optr++ = *iptr++;
+		max_length--;
+	}
+	*optr = '\0';
+	return (int)(optr - output);
+}
+
+#define REGTEST_MAX_LENGTH32 4096
+static PCRE_UCHAR32 regtest_buf32[REGTEST_MAX_LENGTH32];
+static int regtest_offsetmap32[REGTEST_MAX_LENGTH32];
+
+#endif /* SUPPORT_PCRE32 */
+
+static int check_ascii(const char *input)
+{
+	const unsigned char *ptr = (unsigned char *)input;
+	while (*ptr) {
+		if (*ptr > 127)
+			return 0;
+		ptr++;
+	}
+	return 1;
+}
+
+static int regression_tests(void)
+{
+	struct regression_test_case *current = regression_test_cases;
+	const char *error;
+	char *cpu_info;
+	int i, err_offs;
+	int is_successful, is_ascii_pattern, is_ascii_input;
+	int total = 0;
+	int successful = 0;
+	int successful_row = 0;
+	int counter = 0;
+	int study_mode;
+	int utf = 0, ucp = 0;
+	int disabled_flags = 0;
+#ifdef SUPPORT_PCRE8
+	pcre *re8;
+	pcre_extra *extra8;
+	pcre_extra dummy_extra8;
+	int ovector8_1[32];
+	int ovector8_2[32];
+	int return_value8[2];
+	unsigned char *mark8_1, *mark8_2;
+#endif
+#ifdef SUPPORT_PCRE16
+	pcre16 *re16;
+	pcre16_extra *extra16;
+	pcre16_extra dummy_extra16;
+	int ovector16_1[32];
+	int ovector16_2[32];
+	int return_value16[2];
+	PCRE_UCHAR16 *mark16_1, *mark16_2;
+	int length16;
+#endif
+#ifdef SUPPORT_PCRE32
+	pcre32 *re32;
+	pcre32_extra *extra32;
+	pcre32_extra dummy_extra32;
+	int ovector32_1[32];
+	int ovector32_2[32];
+	int return_value32[2];
+	PCRE_UCHAR32 *mark32_1, *mark32_2;
+	int length32;
+#endif
+
+	/* This test compares the behaviour of interpreter and JIT. Although disabling
+	utf or ucp may make tests fail, if the pcre_exec result is the SAME, it is
+	still considered successful from pcre_jit_test point of view. */
+
+#if defined SUPPORT_PCRE8
+	pcre_config(PCRE_CONFIG_JITTARGET, &cpu_info);
+#elif defined SUPPORT_PCRE16
+	pcre16_config(PCRE_CONFIG_JITTARGET, &cpu_info);
+#elif defined SUPPORT_PCRE32
+	pcre32_config(PCRE_CONFIG_JITTARGET, &cpu_info);
+#endif
+
+	printf("Running JIT regression tests\n");
+	printf("  target CPU of SLJIT compiler: %s\n", cpu_info);
+
+#if defined SUPPORT_PCRE8
+	pcre_config(PCRE_CONFIG_UTF8, &utf);
+	pcre_config(PCRE_CONFIG_UNICODE_PROPERTIES, &ucp);
+#elif defined SUPPORT_PCRE16
+	pcre16_config(PCRE_CONFIG_UTF16, &utf);
+	pcre16_config(PCRE_CONFIG_UNICODE_PROPERTIES, &ucp);
+#elif defined SUPPORT_PCRE16
+	pcre32_config(PCRE_CONFIG_UTF32, &utf);
+	pcre32_config(PCRE_CONFIG_UNICODE_PROPERTIES, &ucp);
+#endif
+
+	if (!utf)
+		disabled_flags |= PCRE_UTF8 | PCRE_UTF16 | PCRE_UTF32;
+	if (!ucp)
+		disabled_flags |= PCRE_UCP;
+#ifdef SUPPORT_PCRE8
+	printf("  in  8 bit mode with UTF-8  %s and ucp %s:\n", utf ? "enabled" : "disabled", ucp ? "enabled" : "disabled");
+#endif
+#ifdef SUPPORT_PCRE16
+	printf("  in 16 bit mode with UTF-16 %s and ucp %s:\n", utf ? "enabled" : "disabled", ucp ? "enabled" : "disabled");
+#endif
+#ifdef SUPPORT_PCRE32
+	printf("  in 32 bit mode with UTF-32 %s and ucp %s:\n", utf ? "enabled" : "disabled", ucp ? "enabled" : "disabled");
+#endif
+
+	while (current->pattern) {
+		/* printf("\nPattern: %s :\n", current->pattern); */
+		total++;
+		if (current->start_offset & F_PROPERTY) {
+			is_ascii_pattern = 0;
+			is_ascii_input = 0;
+		} else {
+			is_ascii_pattern = check_ascii(current->pattern);
+			is_ascii_input = check_ascii(current->input);
+		}
+
+		if (current->flags & PCRE_PARTIAL_SOFT)
+			study_mode = PCRE_STUDY_JIT_PARTIAL_SOFT_COMPILE;
+		else if (current->flags & PCRE_PARTIAL_HARD)
+			study_mode = PCRE_STUDY_JIT_PARTIAL_HARD_COMPILE;
+		else
+			study_mode = PCRE_STUDY_JIT_COMPILE;
+		error = NULL;
+#ifdef SUPPORT_PCRE8
+		re8 = NULL;
+		if (!(current->start_offset & F_NO8))
+			re8 = pcre_compile(current->pattern,
+				current->flags & ~(PCRE_NOTBOL | PCRE_NOTEOL | PCRE_NOTEMPTY | PCRE_NOTEMPTY_ATSTART | PCRE_PARTIAL_SOFT | PCRE_PARTIAL_HARD | disabled_flags),
+				&error, &err_offs, tables(0));
+
+		extra8 = NULL;
+		if (re8) {
+			error = NULL;
+			extra8 = pcre_study(re8, study_mode, &error);
+			if (!extra8) {
+				printf("\n8 bit: Cannot study pattern: %s\n", current->pattern);
+				pcre_free(re8);
+				re8 = NULL;
+			}
+			else if (!(extra8->flags & PCRE_EXTRA_EXECUTABLE_JIT)) {
+				printf("\n8 bit: JIT compiler does not support: %s\n", current->pattern);
+				pcre_free_study(extra8);
+				pcre_free(re8);
+				re8 = NULL;
+			}
+			extra8->flags |= PCRE_EXTRA_MARK;
+		} else if (((utf && ucp) || is_ascii_pattern) && !(current->start_offset & F_NO8))
+			printf("\n8 bit: Cannot compile pattern \"%s\": %s\n", current->pattern, error);
+#endif
+#ifdef SUPPORT_PCRE16
+		if ((current->flags & PCRE_UTF16) || (current->start_offset & F_FORCECONV))
+			convert_utf8_to_utf16(current->pattern, regtest_buf16, NULL, REGTEST_MAX_LENGTH16);
+		else
+			copy_char8_to_char16(current->pattern, regtest_buf16, REGTEST_MAX_LENGTH16);
+
+		re16 = NULL;
+		if (!(current->start_offset & F_NO16))
+			re16 = pcre16_compile(regtest_buf16,
+				current->flags & ~(PCRE_NOTBOL | PCRE_NOTEOL | PCRE_NOTEMPTY | PCRE_NOTEMPTY_ATSTART | PCRE_PARTIAL_SOFT | PCRE_PARTIAL_HARD | disabled_flags),
+				&error, &err_offs, tables(0));
+
+		extra16 = NULL;
+		if (re16) {
+			error = NULL;
+			extra16 = pcre16_study(re16, study_mode, &error);
+			if (!extra16) {
+				printf("\n16 bit: Cannot study pattern: %s\n", current->pattern);
+				pcre16_free(re16);
+				re16 = NULL;
+			}
+			else if (!(extra16->flags & PCRE_EXTRA_EXECUTABLE_JIT)) {
+				printf("\n16 bit: JIT compiler does not support: %s\n", current->pattern);
+				pcre16_free_study(extra16);
+				pcre16_free(re16);
+				re16 = NULL;
+			}
+			extra16->flags |= PCRE_EXTRA_MARK;
+		} else if (((utf && ucp) || is_ascii_pattern) && !(current->start_offset & F_NO16))
+			printf("\n16 bit: Cannot compile pattern \"%s\": %s\n", current->pattern, error);
+#endif
+#ifdef SUPPORT_PCRE32
+		if ((current->flags & PCRE_UTF32) || (current->start_offset & F_FORCECONV))
+			convert_utf8_to_utf32(current->pattern, regtest_buf32, NULL, REGTEST_MAX_LENGTH32);
+		else
+			copy_char8_to_char32(current->pattern, regtest_buf32, REGTEST_MAX_LENGTH32);
+
+		re32 = NULL;
+		if (!(current->start_offset & F_NO32))
+			re32 = pcre32_compile(regtest_buf32,
+				current->flags & ~(PCRE_NOTBOL | PCRE_NOTEOL | PCRE_NOTEMPTY | PCRE_NOTEMPTY_ATSTART | PCRE_PARTIAL_SOFT | PCRE_PARTIAL_HARD | disabled_flags),
+				&error, &err_offs, tables(0));
+
+		extra32 = NULL;
+		if (re32) {
+			error = NULL;
+			extra32 = pcre32_study(re32, study_mode, &error);
+			if (!extra32) {
+				printf("\n32 bit: Cannot study pattern: %s\n", current->pattern);
+				pcre32_free(re32);
+				re32 = NULL;
+			}
+			if (!(extra32->flags & PCRE_EXTRA_EXECUTABLE_JIT)) {
+				printf("\n32 bit: JIT compiler does not support: %s\n", current->pattern);
+				pcre32_free_study(extra32);
+				pcre32_free(re32);
+				re32 = NULL;
+			}
+			extra32->flags |= PCRE_EXTRA_MARK;
+		} else if (((utf && ucp) || is_ascii_pattern) && !(current->start_offset & F_NO32))
+			printf("\n32 bit: Cannot compile pattern \"%s\": %s\n", current->pattern, error);
+#endif
+
+		counter++;
+		if ((counter & 0x3) != 0) {
+#ifdef SUPPORT_PCRE8
+			setstack8(NULL);
+#endif
+#ifdef SUPPORT_PCRE16
+			setstack16(NULL);
+#endif
+#ifdef SUPPORT_PCRE32
+			setstack32(NULL);
+#endif
+		}
+
+#ifdef SUPPORT_PCRE8
+		return_value8[0] = -1000;
+		return_value8[1] = -1000;
+		for (i = 0; i < 32; ++i)
+			ovector8_1[i] = -2;
+		for (i = 0; i < 32; ++i)
+			ovector8_2[i] = -2;
+		if (re8) {
+			mark8_1 = NULL;
+			mark8_2 = NULL;
+			extra8->mark = &mark8_1;
+
+			if ((counter & 0x1) != 0) {
+				setstack8(extra8);
+				return_value8[0] = pcre_exec(re8, extra8, current->input, strlen(current->input), current->start_offset & OFFSET_MASK,
+					current->flags & (PCRE_NOTBOL | PCRE_NOTEOL | PCRE_NOTEMPTY | PCRE_NOTEMPTY_ATSTART | PCRE_PARTIAL_SOFT | PCRE_PARTIAL_HARD), ovector8_1, 32);
+			} else
+				return_value8[0] = pcre_jit_exec(re8, extra8, current->input, strlen(current->input), current->start_offset & OFFSET_MASK,
+					current->flags & (PCRE_NOTBOL | PCRE_NOTEOL | PCRE_NOTEMPTY | PCRE_NOTEMPTY_ATSTART | PCRE_PARTIAL_SOFT | PCRE_PARTIAL_HARD), ovector8_1, 32, getstack8());
+			memset(&dummy_extra8, 0, sizeof(pcre_extra));
+			dummy_extra8.flags = PCRE_EXTRA_MARK;
+			if (current->start_offset & F_STUDY) {
+				dummy_extra8.flags |= PCRE_EXTRA_STUDY_DATA;
+				dummy_extra8.study_data = extra8->study_data;
+			}
+			dummy_extra8.mark = &mark8_2;
+			return_value8[1] = pcre_exec(re8, &dummy_extra8, current->input, strlen(current->input), current->start_offset & OFFSET_MASK,
+				current->flags & (PCRE_NOTBOL | PCRE_NOTEOL | PCRE_NOTEMPTY | PCRE_NOTEMPTY_ATSTART | PCRE_PARTIAL_SOFT | PCRE_PARTIAL_HARD), ovector8_2, 32);
+		}
+#endif
+
+#ifdef SUPPORT_PCRE16
+		return_value16[0] = -1000;
+		return_value16[1] = -1000;
+		for (i = 0; i < 32; ++i)
+			ovector16_1[i] = -2;
+		for (i = 0; i < 32; ++i)
+			ovector16_2[i] = -2;
+		if (re16) {
+			mark16_1 = NULL;
+			mark16_2 = NULL;
+			if ((current->flags & PCRE_UTF16) || (current->start_offset & F_FORCECONV))
+				length16 = convert_utf8_to_utf16(current->input, regtest_buf16, regtest_offsetmap16, REGTEST_MAX_LENGTH16);
+			else
+				length16 = copy_char8_to_char16(current->input, regtest_buf16, REGTEST_MAX_LENGTH16);
+			extra16->mark = &mark16_1;
+			if ((counter & 0x1) != 0) {
+				setstack16(extra16);
+				return_value16[0] = pcre16_exec(re16, extra16, regtest_buf16, length16, current->start_offset & OFFSET_MASK,
+					current->flags & (PCRE_NOTBOL | PCRE_NOTEOL | PCRE_NOTEMPTY | PCRE_NOTEMPTY_ATSTART | PCRE_PARTIAL_SOFT | PCRE_PARTIAL_HARD), ovector16_1, 32);
+			} else
+				return_value16[0] = pcre16_jit_exec(re16, extra16, regtest_buf16, length16, current->start_offset & OFFSET_MASK,
+					current->flags & (PCRE_NOTBOL | PCRE_NOTEOL | PCRE_NOTEMPTY | PCRE_NOTEMPTY_ATSTART | PCRE_PARTIAL_SOFT | PCRE_PARTIAL_HARD), ovector16_1, 32, getstack16());
+			memset(&dummy_extra16, 0, sizeof(pcre16_extra));
+			dummy_extra16.flags = PCRE_EXTRA_MARK;
+			if (current->start_offset & F_STUDY) {
+				dummy_extra16.flags |= PCRE_EXTRA_STUDY_DATA;
+				dummy_extra16.study_data = extra16->study_data;
+			}
+			dummy_extra16.mark = &mark16_2;
+			return_value16[1] = pcre16_exec(re16, &dummy_extra16, regtest_buf16, length16, current->start_offset & OFFSET_MASK,
+				current->flags & (PCRE_NOTBOL | PCRE_NOTEOL | PCRE_NOTEMPTY | PCRE_NOTEMPTY_ATSTART | PCRE_PARTIAL_SOFT | PCRE_PARTIAL_HARD), ovector16_2, 32);
+		}
+#endif
+
+#ifdef SUPPORT_PCRE32
+		return_value32[0] = -1000;
+		return_value32[1] = -1000;
+		for (i = 0; i < 32; ++i)
+			ovector32_1[i] = -2;
+		for (i = 0; i < 32; ++i)
+			ovector32_2[i] = -2;
+		if (re32) {
+			mark32_1 = NULL;
+			mark32_2 = NULL;
+			if ((current->flags & PCRE_UTF32) || (current->start_offset & F_FORCECONV))
+				length32 = convert_utf8_to_utf32(current->input, regtest_buf32, regtest_offsetmap32, REGTEST_MAX_LENGTH32);
+			else
+				length32 = copy_char8_to_char32(current->input, regtest_buf32, REGTEST_MAX_LENGTH32);
+			extra32->mark = &mark32_1;
+			if ((counter & 0x1) != 0) {
+				setstack32(extra32);
+				return_value32[0] = pcre32_exec(re32, extra32, regtest_buf32, length32, current->start_offset & OFFSET_MASK,
+					current->flags & (PCRE_NOTBOL | PCRE_NOTEOL | PCRE_NOTEMPTY | PCRE_NOTEMPTY_ATSTART | PCRE_PARTIAL_SOFT | PCRE_PARTIAL_HARD), ovector32_1, 32);
+			} else
+				return_value32[0] = pcre32_jit_exec(re32, extra32, regtest_buf32, length32, current->start_offset & OFFSET_MASK,
+					current->flags & (PCRE_NOTBOL | PCRE_NOTEOL | PCRE_NOTEMPTY | PCRE_NOTEMPTY_ATSTART | PCRE_PARTIAL_SOFT | PCRE_PARTIAL_HARD), ovector32_1, 32, getstack32());
+			memset(&dummy_extra32, 0, sizeof(pcre32_extra));
+			dummy_extra32.flags = PCRE_EXTRA_MARK;
+			if (current->start_offset & F_STUDY) {
+				dummy_extra32.flags |= PCRE_EXTRA_STUDY_DATA;
+				dummy_extra32.study_data = extra32->study_data;
+			}
+			dummy_extra32.mark = &mark32_2;
+			return_value32[1] = pcre32_exec(re32, &dummy_extra32, regtest_buf32, length32, current->start_offset & OFFSET_MASK,
+				current->flags & (PCRE_NOTBOL | PCRE_NOTEOL | PCRE_NOTEMPTY | PCRE_NOTEMPTY_ATSTART | PCRE_PARTIAL_SOFT | PCRE_PARTIAL_HARD), ovector32_2, 32);
+		}
+#endif
+
+		/* printf("[%d-%d-%d|%d-%d|%d-%d|%d-%d]%s",
+			return_value8[0], return_value16[0],
+			ovector8_1[0], ovector8_1[1],
+			ovector16_1[0], ovector16_1[1],
+			ovector32_1[0], ovector32_1[1],
+			(current->flags & PCRE_CASELESS) ? "C" : ""); */
+
+		/* If F_DIFF is set, just run the test, but do not compare the results.
+		Segfaults can still be captured. */
+
+		is_successful = 1;
+		if (!(current->start_offset & F_DIFF)) {
+#if defined SUPPORT_UTF && ((defined(SUPPORT_PCRE8) + defined(SUPPORT_PCRE16) + defined(SUPPORT_PCRE32)) >= 2)
+			if (!(current->start_offset & F_FORCECONV)) {
+				int return_value;
+
+				/* All results must be the same. */
+#ifdef SUPPORT_PCRE8
+				if ((return_value = return_value8[0]) != return_value8[1]) {
+					printf("\n8 bit: Return value differs(J8:%d,I8:%d): [%d] '%s' @ '%s'\n",
+						return_value8[0], return_value8[1], total, current->pattern, current->input);
+					is_successful = 0;
+				} else
+#endif
+#ifdef SUPPORT_PCRE16
+				if ((return_value = return_value16[0]) != return_value16[1]) {
+					printf("\n16 bit: Return value differs(J16:%d,I16:%d): [%d] '%s' @ '%s'\n",
+						return_value16[0], return_value16[1], total, current->pattern, current->input);
+					is_successful = 0;
+				} else
+#endif
+#ifdef SUPPORT_PCRE32
+				if ((return_value = return_value32[0]) != return_value32[1]) {
+					printf("\n32 bit: Return value differs(J32:%d,I32:%d): [%d] '%s' @ '%s'\n",
+						return_value32[0], return_value32[1], total, current->pattern, current->input);
+					is_successful = 0;
+				} else
+#endif
+#if defined SUPPORT_PCRE8 && defined SUPPORT_PCRE16
+				if (return_value8[0] != return_value16[0]) {
+					printf("\n8 and 16 bit: Return value differs(J8:%d,J16:%d): [%d] '%s' @ '%s'\n",
+						return_value8[0], return_value16[0],
+						total, current->pattern, current->input);
+					is_successful = 0;
+				} else
+#endif
+#if defined SUPPORT_PCRE8 && defined SUPPORT_PCRE32
+				if (return_value8[0] != return_value32[0]) {
+					printf("\n8 and 32 bit: Return value differs(J8:%d,J32:%d): [%d] '%s' @ '%s'\n",
+						return_value8[0], return_value32[0],
+						total, current->pattern, current->input);
+					is_successful = 0;
+				} else
+#endif
+#if defined SUPPORT_PCRE16 && defined SUPPORT_PCRE32
+				if (return_value16[0] != return_value32[0]) {
+					printf("\n16 and 32 bit: Return value differs(J16:%d,J32:%d): [%d] '%s' @ '%s'\n",
+						return_value16[0], return_value32[0],
+						total, current->pattern, current->input);
+					is_successful = 0;
+				} else
+#endif
+				if (return_value >= 0 || return_value == PCRE_ERROR_PARTIAL) {
+					if (return_value == PCRE_ERROR_PARTIAL) {
+						return_value = 2;
+					} else {
+						return_value *= 2;
+					}
+#ifdef SUPPORT_PCRE8
+					return_value8[0] = return_value;
+#endif
+#ifdef SUPPORT_PCRE16
+					return_value16[0] = return_value;
+#endif
+#ifdef SUPPORT_PCRE32
+					return_value32[0] = return_value;
+#endif
+					/* Transform back the results. */
+					if (current->flags & PCRE_UTF8) {
+#ifdef SUPPORT_PCRE16
+						for (i = 0; i < return_value; ++i) {
+							if (ovector16_1[i] >= 0)
+								ovector16_1[i] = regtest_offsetmap16[ovector16_1[i]];
+							if (ovector16_2[i] >= 0)
+								ovector16_2[i] = regtest_offsetmap16[ovector16_2[i]];
+						}
+#endif
+#ifdef SUPPORT_PCRE32
+						for (i = 0; i < return_value; ++i) {
+							if (ovector32_1[i] >= 0)
+								ovector32_1[i] = regtest_offsetmap32[ovector32_1[i]];
+							if (ovector32_2[i] >= 0)
+								ovector32_2[i] = regtest_offsetmap32[ovector32_2[i]];
+						}
+#endif
+					}
+
+					for (i = 0; i < return_value; ++i) {
+#if defined SUPPORT_PCRE8 && defined SUPPORT_PCRE16
+						if (ovector8_1[i] != ovector8_2[i] || ovector8_1[i] != ovector16_1[i] || ovector8_1[i] != ovector16_2[i]) {
+							printf("\n8 and 16 bit: Ovector[%d] value differs(J8:%d,I8:%d,J16:%d,I16:%d): [%d] '%s' @ '%s' \n",
+								i, ovector8_1[i], ovector8_2[i], ovector16_1[i], ovector16_2[i],
+								total, current->pattern, current->input);
+							is_successful = 0;
+						}
+#endif
+#if defined SUPPORT_PCRE8 && defined SUPPORT_PCRE32
+						if (ovector8_1[i] != ovector8_2[i] || ovector8_1[i] != ovector32_1[i] || ovector8_1[i] != ovector32_2[i]) {
+							printf("\n8 and 32 bit: Ovector[%d] value differs(J8:%d,I8:%d,J32:%d,I32:%d): [%d] '%s' @ '%s' \n",
+								i, ovector8_1[i], ovector8_2[i], ovector32_1[i], ovector32_2[i],
+								total, current->pattern, current->input);
+							is_successful = 0;
+						}
+#endif
+#if defined SUPPORT_PCRE16 && defined SUPPORT_PCRE16
+						if (ovector16_1[i] != ovector16_2[i] || ovector16_1[i] != ovector16_1[i] || ovector16_1[i] != ovector16_2[i]) {
+							printf("\n16 and 16 bit: Ovector[%d] value differs(J16:%d,I16:%d,J32:%d,I32:%d): [%d] '%s' @ '%s' \n",
+								i, ovector16_1[i], ovector16_2[i], ovector16_1[i], ovector16_2[i],
+								total, current->pattern, current->input);
+							is_successful = 0;
+						}
+#endif
+					}
+				}
+			} else
+#endif /* more than one of SUPPORT_PCRE8, SUPPORT_PCRE16 and SUPPORT_PCRE32 */
+			{
+				/* Only the 8 bit and 16 bit results must be equal. */
+#ifdef SUPPORT_PCRE8
+				if (return_value8[0] != return_value8[1]) {
+					printf("\n8 bit: Return value differs(%d:%d): [%d] '%s' @ '%s'\n",
+						return_value8[0], return_value8[1], total, current->pattern, current->input);
+					is_successful = 0;
+				} else if (return_value8[0] >= 0 || return_value8[0] == PCRE_ERROR_PARTIAL) {
+					if (return_value8[0] == PCRE_ERROR_PARTIAL)
+						return_value8[0] = 2;
+					else
+						return_value8[0] *= 2;
+
+					for (i = 0; i < return_value8[0]; ++i)
+						if (ovector8_1[i] != ovector8_2[i]) {
+							printf("\n8 bit: Ovector[%d] value differs(%d:%d): [%d] '%s' @ '%s'\n",
+								i, ovector8_1[i], ovector8_2[i], total, current->pattern, current->input);
+							is_successful = 0;
+						}
+				}
+#endif
+
+#ifdef SUPPORT_PCRE16
+				if (return_value16[0] != return_value16[1]) {
+					printf("\n16 bit: Return value differs(%d:%d): [%d] '%s' @ '%s'\n",
+						return_value16[0], return_value16[1], total, current->pattern, current->input);
+					is_successful = 0;
+				} else if (return_value16[0] >= 0 || return_value16[0] == PCRE_ERROR_PARTIAL) {
+					if (return_value16[0] == PCRE_ERROR_PARTIAL)
+						return_value16[0] = 2;
+					else
+						return_value16[0] *= 2;
+
+					for (i = 0; i < return_value16[0]; ++i)
+						if (ovector16_1[i] != ovector16_2[i]) {
+							printf("\n16 bit: Ovector[%d] value differs(%d:%d): [%d] '%s' @ '%s'\n",
+								i, ovector16_1[i], ovector16_2[i], total, current->pattern, current->input);
+							is_successful = 0;
+						}
+				}
+#endif
+
+#ifdef SUPPORT_PCRE32
+				if (return_value32[0] != return_value32[1]) {
+					printf("\n32 bit: Return value differs(%d:%d): [%d] '%s' @ '%s'\n",
+						return_value32[0], return_value32[1], total, current->pattern, current->input);
+					is_successful = 0;
+				} else if (return_value32[0] >= 0 || return_value32[0] == PCRE_ERROR_PARTIAL) {
+					if (return_value32[0] == PCRE_ERROR_PARTIAL)
+						return_value32[0] = 2;
+					else
+						return_value32[0] *= 2;
+
+					for (i = 0; i < return_value32[0]; ++i)
+						if (ovector32_1[i] != ovector32_2[i]) {
+							printf("\n32 bit: Ovector[%d] value differs(%d:%d): [%d] '%s' @ '%s'\n",
+								i, ovector32_1[i], ovector32_2[i], total, current->pattern, current->input);
+							is_successful = 0;
+						}
+				}
+#endif
+			}
+		}
+
+		if (is_successful) {
+#ifdef SUPPORT_PCRE8
+			if (!(current->start_offset & F_NO8) && ((utf && ucp) || is_ascii_input)) {
+				if (return_value8[0] < 0 && !(current->start_offset & F_NOMATCH)) {
+					printf("8 bit: Test should match: [%d] '%s' @ '%s'\n",
+						total, current->pattern, current->input);
+					is_successful = 0;
+				}
+
+				if (return_value8[0] >= 0 && (current->start_offset & F_NOMATCH)) {
+					printf("8 bit: Test should not match: [%d] '%s' @ '%s'\n",
+						total, current->pattern, current->input);
+					is_successful = 0;
+				}
+			}
+#endif
+#ifdef SUPPORT_PCRE16
+			if (!(current->start_offset & F_NO16) && ((utf && ucp) || is_ascii_input)) {
+				if (return_value16[0] < 0 && !(current->start_offset & F_NOMATCH)) {
+					printf("16 bit: Test should match: [%d] '%s' @ '%s'\n",
+						total, current->pattern, current->input);
+					is_successful = 0;
+				}
+
+				if (return_value16[0] >= 0 && (current->start_offset & F_NOMATCH)) {
+					printf("16 bit: Test should not match: [%d] '%s' @ '%s'\n",
+						total, current->pattern, current->input);
+					is_successful = 0;
+				}
+			}
+#endif
+#ifdef SUPPORT_PCRE32
+			if (!(current->start_offset & F_NO32) && ((utf && ucp) || is_ascii_input)) {
+				if (return_value32[0] < 0 && !(current->start_offset & F_NOMATCH)) {
+					printf("32 bit: Test should match: [%d] '%s' @ '%s'\n",
+						total, current->pattern, current->input);
+					is_successful = 0;
+				}
+
+				if (return_value32[0] >= 0 && (current->start_offset & F_NOMATCH)) {
+					printf("32 bit: Test should not match: [%d] '%s' @ '%s'\n",
+						total, current->pattern, current->input);
+					is_successful = 0;
+				}
+			}
+#endif
+		}
+
+		if (is_successful) {
+#ifdef SUPPORT_PCRE8
+			if (mark8_1 != mark8_2) {
+				printf("8 bit: Mark value mismatch: [%d] '%s' @ '%s'\n",
+					total, current->pattern, current->input);
+				is_successful = 0;
+			}
+#endif
+#ifdef SUPPORT_PCRE16
+			if (mark16_1 != mark16_2) {
+				printf("16 bit: Mark value mismatch: [%d] '%s' @ '%s'\n",
+					total, current->pattern, current->input);
+				is_successful = 0;
+			}
+#endif
+#ifdef SUPPORT_PCRE32
+			if (mark32_1 != mark32_2) {
+				printf("32 bit: Mark value mismatch: [%d] '%s' @ '%s'\n",
+					total, current->pattern, current->input);
+				is_successful = 0;
+			}
+#endif
+		}
+
+#ifdef SUPPORT_PCRE8
+		if (re8) {
+			pcre_free_study(extra8);
+			pcre_free(re8);
+		}
+#endif
+#ifdef SUPPORT_PCRE16
+		if (re16) {
+			pcre16_free_study(extra16);
+			pcre16_free(re16);
+		}
+#endif
+#ifdef SUPPORT_PCRE32
+		if (re32) {
+			pcre32_free_study(extra32);
+			pcre32_free(re32);
+		}
+#endif
+
+		if (is_successful) {
+			successful++;
+			successful_row++;
+			printf(".");
+			if (successful_row >= 60) {
+				successful_row = 0;
+				printf("\n");
+			}
+		} else
+			successful_row = 0;
+
+		fflush(stdout);
+		current++;
+	}
+	tables(1);
+#ifdef SUPPORT_PCRE8
+	setstack8(NULL);
+#endif
+#ifdef SUPPORT_PCRE16
+	setstack16(NULL);
+#endif
+#ifdef SUPPORT_PCRE32
+	setstack32(NULL);
+#endif
+
+	if (total == successful) {
+		printf("\nAll JIT regression tests are successfully passed.\n");
+		return 0;
+	} else {
+		printf("\nSuccessful test ratio: %d%% (%d failed)\n", successful * 100 / total, total - successful);
+		return 1;
+	}
+}
+
+/* End of pcre_jit_test.c */
diff --git a/pcre/pcre_maketables.c b/pcre/pcre_maketables.c
new file mode 100644
index 00000000000..610a669586b
--- /dev/null
+++ b/pcre/pcre_maketables.c
@@ -0,0 +1,151 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+
+/* This module contains the external function pcre_maketables(), which builds
+character tables for PCRE in the current locale. The file is compiled on its
+own as part of the PCRE library. However, it is also included in the
+compilation of dftables.c, in which case the macro DFTABLES is defined. */
+
+
+#ifndef DFTABLES
+#  ifdef HAVE_CONFIG_H
+#  include "config.h"
+#  endif
+#  include "pcre_internal.h"
+#endif
+
+
+/*************************************************
+*           Create PCRE character tables         *
+*************************************************/
+
+/* This function builds a set of character tables for use by PCRE and returns
+a pointer to them. They are build using the ctype functions, and consequently
+their contents will depend upon the current locale setting. When compiled as
+part of the library, the store is obtained via PUBL(malloc)(), but when
+compiled inside dftables, use malloc().
+
+Arguments:   none
+Returns:     pointer to the contiguous block of data
+*/
+
+#if defined COMPILE_PCRE8
+const unsigned char *
+pcre_maketables(void)
+#elif defined COMPILE_PCRE16
+const unsigned char *
+pcre16_maketables(void)
+#elif defined COMPILE_PCRE32
+const unsigned char *
+pcre32_maketables(void)
+#endif
+{
+unsigned char *yield, *p;
+int i;
+
+#ifndef DFTABLES
+yield = (unsigned char*)(PUBL(malloc))(tables_length);
+#else
+yield = (unsigned char*)malloc(tables_length);
+#endif
+
+if (yield == NULL) return NULL;
+p = yield;
+
+/* First comes the lower casing table */
+
+for (i = 0; i < 256; i++) *p++ = tolower(i);
+
+/* Next the case-flipping table */
+
+for (i = 0; i < 256; i++) *p++ = islower(i)? toupper(i) : tolower(i);
+
+/* Then the character class tables. Don't try to be clever and save effort on
+exclusive ones - in some locales things may be different. Note that the table
+for "space" includes everything "isspace" gives, including VT in the default
+locale. This makes it work for the POSIX class [:space:]. Note also that it is
+possible for a character to be alnum or alpha without being lower or upper,
+such as "male and female ordinals" (\xAA and \xBA) in the fr_FR locale (at
+least under Debian Linux's locales as of 12/2005). So we must test for alnum
+specially. */
+
+memset(p, 0, cbit_length);
+for (i = 0; i < 256; i++)
+  {
+  if (isdigit(i)) p[cbit_digit  + i/8] |= 1 << (i&7);
+  if (isupper(i)) p[cbit_upper  + i/8] |= 1 << (i&7);
+  if (islower(i)) p[cbit_lower  + i/8] |= 1 << (i&7);
+  if (isalnum(i)) p[cbit_word   + i/8] |= 1 << (i&7);
+  if (i == '_')   p[cbit_word   + i/8] |= 1 << (i&7);
+  if (isspace(i)) p[cbit_space  + i/8] |= 1 << (i&7);
+  if (isxdigit(i))p[cbit_xdigit + i/8] |= 1 << (i&7);
+  if (isgraph(i)) p[cbit_graph  + i/8] |= 1 << (i&7);
+  if (isprint(i)) p[cbit_print  + i/8] |= 1 << (i&7);
+  if (ispunct(i)) p[cbit_punct  + i/8] |= 1 << (i&7);
+  if (iscntrl(i)) p[cbit_cntrl  + i/8] |= 1 << (i&7);
+  }
+p += cbit_length;
+
+/* Finally, the character type table. In this, we exclude VT from the white
+space chars, because Perl doesn't recognize it as such for \s and for comments
+within regexes. */
+
+for (i = 0; i < 256; i++)
+  {
+  int x = 0;
+  if (i != CHAR_VT && isspace(i)) x += ctype_space;
+  if (isalpha(i)) x += ctype_letter;
+  if (isdigit(i)) x += ctype_digit;
+  if (isxdigit(i)) x += ctype_xdigit;
+  if (isalnum(i) || i == '_') x += ctype_word;
+
+  /* Note: strchr includes the terminating zero in the characters it considers.
+  In this instance, that is ok because we want binary zero to be flagged as a
+  meta-character, which in this sense is any character that terminates a run
+  of data characters. */
+
+  if (strchr("\\*+?{^.$|()[", i) != 0) x += ctype_meta;
+  *p++ = x;
+  }
+
+return yield;
+}
+
+/* End of pcre_maketables.c */
diff --git a/pcre/pcre_newline.c b/pcre/pcre_newline.c
new file mode 100644
index 00000000000..b8f5a4de19c
--- /dev/null
+++ b/pcre/pcre_newline.c
@@ -0,0 +1,210 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+
+/* This module contains internal functions for testing newlines when more than
+one kind of newline is to be recognized. When a newline is found, its length is
+returned. In principle, we could implement several newline "types", each
+referring to a different set of newline characters. At present, PCRE supports
+only NLTYPE_FIXED, which gets handled without these functions, NLTYPE_ANYCRLF,
+and NLTYPE_ANY. The full list of Unicode newline characters is taken from
+http://unicode.org/unicode/reports/tr18/. */
+
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include "pcre_internal.h"
+
+
+
+/*************************************************
+*      Check for newline at given position       *
+*************************************************/
+
+/* It is guaranteed that the initial value of ptr is less than the end of the
+string that is being processed.
+
+Arguments:
+  ptr          pointer to possible newline
+  type         the newline type
+  endptr       pointer to the end of the string
+  lenptr       where to return the length
+  utf          TRUE if in utf mode
+
+Returns:       TRUE or FALSE
+*/
+
+BOOL
+PRIV(is_newline)(PCRE_PUCHAR ptr, int type, PCRE_PUCHAR endptr, int *lenptr,
+  BOOL utf)
+{
+pcre_uint32 c;
+(void)utf;
+#ifdef SUPPORT_UTF
+if (utf)
+  {
+  GETCHAR(c, ptr);
+  }
+else
+#endif  /* SUPPORT_UTF */
+  c = *ptr;
+
+/* Note that this function is called only for ANY or ANYCRLF. */
+
+if (type == NLTYPE_ANYCRLF) switch(c)
+  {
+  case CHAR_LF: *lenptr = 1; return TRUE;
+  case CHAR_CR: *lenptr = (ptr < endptr - 1 && ptr[1] == CHAR_LF)? 2 : 1;
+               return TRUE;
+  default: return FALSE;
+  }
+
+/* NLTYPE_ANY */
+
+else switch(c)
+  {
+#ifdef EBCDIC
+  case CHAR_NEL:
+#endif
+  case CHAR_LF:
+  case CHAR_VT:
+  case CHAR_FF: *lenptr = 1; return TRUE;
+
+  case CHAR_CR:
+  *lenptr = (ptr < endptr - 1 && ptr[1] == CHAR_LF)? 2 : 1;
+  return TRUE;
+
+#ifndef EBCDIC
+#ifdef COMPILE_PCRE8
+  case CHAR_NEL: *lenptr = utf? 2 : 1; return TRUE;
+  case 0x2028:                                       /* LS */
+  case 0x2029: *lenptr = 3; return TRUE;             /* PS */
+#else /* COMPILE_PCRE16 || COMPILE_PCRE32 */
+  case CHAR_NEL:
+  case 0x2028:                                       /* LS */
+  case 0x2029: *lenptr = 1; return TRUE;             /* PS */
+#endif  /* COMPILE_PCRE8 */
+#endif  /* Not EBCDIC */
+
+  default: return FALSE;
+  }
+}
+
+
+
+/*************************************************
+*     Check for newline at previous position     *
+*************************************************/
+
+/* It is guaranteed that the initial value of ptr is greater than the start of
+the string that is being processed.
+
+Arguments:
+  ptr          pointer to possible newline
+  type         the newline type
+  startptr     pointer to the start of the string
+  lenptr       where to return the length
+  utf          TRUE if in utf mode
+
+Returns:       TRUE or FALSE
+*/
+
+BOOL
+PRIV(was_newline)(PCRE_PUCHAR ptr, int type, PCRE_PUCHAR startptr, int *lenptr,
+  BOOL utf)
+{
+pcre_uint32 c;
+(void)utf;
+ptr--;
+#ifdef SUPPORT_UTF
+if (utf)
+  {
+  BACKCHAR(ptr);
+  GETCHAR(c, ptr);
+  }
+else
+#endif  /* SUPPORT_UTF */
+  c = *ptr;
+
+/* Note that this function is called only for ANY or ANYCRLF. */
+
+if (type == NLTYPE_ANYCRLF) switch(c)
+  {
+  case CHAR_LF:
+  *lenptr = (ptr > startptr && ptr[-1] == CHAR_CR)? 2 : 1;
+  return TRUE;
+
+  case CHAR_CR: *lenptr = 1; return TRUE;
+  default: return FALSE;
+  }
+
+/* NLTYPE_ANY */
+
+else switch(c)
+  {
+  case CHAR_LF:
+  *lenptr = (ptr > startptr && ptr[-1] == CHAR_CR)? 2 : 1;
+  return TRUE;
+
+#ifdef EBCDIC
+  case CHAR_NEL:
+#endif
+  case CHAR_VT:
+  case CHAR_FF:
+  case CHAR_CR: *lenptr = 1; return TRUE;
+
+#ifndef EBCDIC
+#ifdef COMPILE_PCRE8
+  case CHAR_NEL: *lenptr = utf? 2 : 1; return TRUE;
+  case 0x2028:                                       /* LS */
+  case 0x2029: *lenptr = 3; return TRUE;             /* PS */
+#else /* COMPILE_PCRE16 || COMPILE_PCRE32 */
+  case CHAR_NEL:
+  case 0x2028:                                       /* LS */
+  case 0x2029: *lenptr = 1; return TRUE;             /* PS */
+#endif  /* COMPILE_PCRE8 */
+#endif  /* NotEBCDIC */
+
+  default: return FALSE;
+  }
+}
+
+/* End of pcre_newline.c */
diff --git a/pcre/pcre_ord2utf8.c b/pcre/pcre_ord2utf8.c
new file mode 100644
index 00000000000..95f1beb963e
--- /dev/null
+++ b/pcre/pcre_ord2utf8.c
@@ -0,0 +1,94 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+
+/* This file contains a private PCRE function that converts an ordinal
+character value into a UTF8 string. */
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#define COMPILE_PCRE8
+
+#include "pcre_internal.h"
+
+/*************************************************
+*       Convert character value to UTF-8         *
+*************************************************/
+
+/* This function takes an integer value in the range 0 - 0x10ffff
+and encodes it as a UTF-8 character in 1 to 4 pcre_uchars.
+
+Arguments:
+  cvalue     the character value
+  buffer     pointer to buffer for result - at least 6 pcre_uchars long
+
+Returns:     number of characters placed in the buffer
+*/
+
+unsigned
+int
+PRIV(ord2utf)(pcre_uint32 cvalue, pcre_uchar *buffer)
+{
+#ifdef SUPPORT_UTF
+
+register int i, j;
+
+for (i = 0; i < PRIV(utf8_table1_size); i++)
+  if ((int)cvalue <= PRIV(utf8_table1)[i]) break;
+buffer += i;
+for (j = i; j > 0; j--)
+ {
+ *buffer-- = 0x80 | (cvalue & 0x3f);
+ cvalue >>= 6;
+ }
+*buffer = PRIV(utf8_table2)[i] | cvalue;
+return i + 1;
+
+#else
+
+(void)(cvalue);  /* Keep compiler happy; this function won't ever be */
+(void)(buffer);  /* called when SUPPORT_UTF is not defined. */
+return 0;
+
+#endif
+}
+
+/* End of pcre_ord2utf8.c */
diff --git a/pcre/pcre_printint.c b/pcre/pcre_printint.c
new file mode 100644
index 00000000000..10b575426b3
--- /dev/null
+++ b/pcre/pcre_printint.c
@@ -0,0 +1,766 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+
+/* This module contains a PCRE private debugging function for printing out the
+internal form of a compiled regular expression, along with some supporting
+local functions. This source file is used in two places:
+
+(1) It is #included by pcre_compile.c when it is compiled in debugging mode
+(PCRE_DEBUG defined in pcre_internal.h). It is not included in production
+compiles. In this case PCRE_INCLUDED is defined.
+
+(2) It is also compiled separately and linked with pcretest.c, which can be
+asked to print out a compiled regex for debugging purposes. */
+
+#ifndef PCRE_INCLUDED
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+/* For pcretest program. */
+#define PRIV(name) name
+
+/* We have to include pcre_internal.h because we need the internal info for
+displaying the results of pcre_study() and we also need to know about the
+internal macros, structures, and other internal data values; pcretest has
+"inside information" compared to a program that strictly follows the PCRE API.
+
+Although pcre_internal.h does itself include pcre.h, we explicitly include it
+here before pcre_internal.h so that the PCRE_EXP_xxx macros get set
+appropriately for an application, not for building PCRE. */
+
+#include "pcre.h"
+#include "pcre_internal.h"
+
+/* These are the funtions that are contained within. It doesn't seem worth
+having a separate .h file just for this. */
+
+#endif /* PCRE_INCLUDED */
+
+#ifdef PCRE_INCLUDED
+static /* Keep the following function as private. */
+#endif
+
+#if defined COMPILE_PCRE8
+void pcre_printint(pcre *external_re, FILE *f, BOOL print_lengths);
+#elif defined COMPILE_PCRE16
+void pcre16_printint(pcre *external_re, FILE *f, BOOL print_lengths);
+#elif defined COMPILE_PCRE32
+void pcre32_printint(pcre *external_re, FILE *f, BOOL print_lengths);
+#endif
+
+/* Macro that decides whether a character should be output as a literal or in
+hexadecimal. We don't use isprint() because that can vary from system to system
+(even without the use of locales) and we want the output always to be the same,
+for testing purposes. */
+
+#ifdef EBCDIC
+#define PRINTABLE(c) ((c) >= 64 && (c) < 255)
+#else
+#define PRINTABLE(c) ((c) >= 32 && (c) < 127)
+#endif
+
+/* The table of operator names. */
+
+static const char *priv_OP_names[] = { OP_NAME_LIST };
+
+/* This table of operator lengths is not actually used by the working code,
+but its size is needed for a check that ensures it is the correct size for the
+number of opcodes (thus catching update omissions). */
+
+static const pcre_uint8 priv_OP_lengths[] = { OP_LENGTHS };
+
+
+
+/*************************************************
+*       Print single- or multi-byte character    *
+*************************************************/
+
+static unsigned int
+print_char(FILE *f, pcre_uchar *ptr, BOOL utf)
+{
+pcre_uint32 c = *ptr;
+
+#ifndef SUPPORT_UTF
+
+(void)utf;  /* Avoid compiler warning */
+if (PRINTABLE(c)) fprintf(f, "%c", (char)c);
+else if (c <= 0x80) fprintf(f, "\\x%02x", c);
+else fprintf(f, "\\x{%x}", c);
+return 0;
+
+#else
+
+#if defined COMPILE_PCRE8
+
+if (!utf || (c & 0xc0) != 0xc0)
+  {
+  if (PRINTABLE(c)) fprintf(f, "%c", (char)c);
+  else if (c < 0x80) fprintf(f, "\\x%02x", c);
+  else fprintf(f, "\\x{%02x}", c);
+  return 0;
+  }
+else
+  {
+  int i;
+  int a = PRIV(utf8_table4)[c & 0x3f];  /* Number of additional bytes */
+  int s = 6*a;
+  c = (c & PRIV(utf8_table3)[a]) << s;
+  for (i = 1; i <= a; i++)
+    {
+    /* This is a check for malformed UTF-8; it should only occur if the sanity
+    check has been turned off. Rather than swallow random bytes, just stop if
+    we hit a bad one. Print it with \X instead of \x as an indication. */
+
+    if ((ptr[i] & 0xc0) != 0x80)
+      {
+      fprintf(f, "\\X{%x}", c);
+      return i - 1;
+      }
+
+    /* The byte is OK */
+
+    s -= 6;
+    c |= (ptr[i] & 0x3f) << s;
+    }
+  fprintf(f, "\\x{%x}", c);
+  return a;
+  }
+
+#elif defined COMPILE_PCRE16
+
+if (!utf || (c & 0xfc00) != 0xd800)
+  {
+  if (PRINTABLE(c)) fprintf(f, "%c", (char)c);
+  else if (c <= 0x80) fprintf(f, "\\x%02x", c);
+  else fprintf(f, "\\x{%02x}", c);
+  return 0;
+  }
+else
+  {
+  /* This is a check for malformed UTF-16; it should only occur if the sanity
+  check has been turned off. Rather than swallow a low surrogate, just stop if
+  we hit a bad one. Print it with \X instead of \x as an indication. */
+
+  if ((ptr[1] & 0xfc00) != 0xdc00)
+    {
+    fprintf(f, "\\X{%x}", c);
+    return 0;
+    }
+
+  c = (((c & 0x3ff) << 10) | (ptr[1] & 0x3ff)) + 0x10000;
+  fprintf(f, "\\x{%x}", c);
+  return 1;
+  }
+
+#elif defined COMPILE_PCRE32
+
+if (!utf || (c & 0xfffff800u) != 0xd800u)
+  {
+  if (PRINTABLE(c)) fprintf(f, "%c", (char)c);
+  else if (c <= 0x80) fprintf(f, "\\x%02x", c);
+  else fprintf(f, "\\x{%x}", c);
+  return 0;
+  }
+else
+  {
+  /* This is a check for malformed UTF-32; it should only occur if the sanity
+  check has been turned off. Rather than swallow a surrogate, just stop if
+  we hit one. Print it with \X instead of \x as an indication. */
+  fprintf(f, "\\X{%x}", c);
+  return 0;
+  }
+
+#endif /* COMPILE_PCRE[8|16|32] */
+
+#endif /* SUPPORT_UTF */
+}
+
+/*************************************************
+*  Print uchar string (regardless of utf)        *
+*************************************************/
+
+static void
+print_puchar(FILE *f, PCRE_PUCHAR ptr)
+{
+while (*ptr != '\0')
+  {
+  register pcre_uint32 c = *ptr++;
+  if (PRINTABLE(c)) fprintf(f, "%c", c); else fprintf(f, "\\x{%x}", c);
+  }
+}
+
+/*************************************************
+*          Find Unicode property name            *
+*************************************************/
+
+static const char *
+get_ucpname(unsigned int ptype, unsigned int pvalue)
+{
+#ifdef SUPPORT_UCP
+int i;
+for (i = PRIV(utt_size) - 1; i >= 0; i--)
+  {
+  if (ptype == PRIV(utt)[i].type && pvalue == PRIV(utt)[i].value) break;
+  }
+return (i >= 0)? PRIV(utt_names) + PRIV(utt)[i].name_offset : "??";
+#else
+/* It gets harder and harder to shut off unwanted compiler warnings. */
+ptype = ptype * pvalue;
+return (ptype == pvalue)? "??" : "??";
+#endif
+}
+
+
+/*************************************************
+*       Print Unicode property value             *
+*************************************************/
+
+/* "Normal" properties can be printed from tables. The PT_CLIST property is a
+pseudo-property that contains a pointer to a list of case-equivalent
+characters. This is used only when UCP support is available and UTF mode is
+selected. It should never occur otherwise, but just in case it does, have
+something ready to print. */
+
+static void
+print_prop(FILE *f, pcre_uchar *code, const char *before, const char *after)
+{
+if (code[1] != PT_CLIST)
+  {
+  fprintf(f, "%s%s %s%s", before, priv_OP_names[*code], get_ucpname(code[1],
+    code[2]), after);
+  }
+else
+  {
+  const char *not = (*code == OP_PROP)? "" : "not ";
+#ifndef SUPPORT_UCP
+  fprintf(f, "%s%sclist %d%s", before, not, code[2], after);
+#else
+  const pcre_uint32 *p = PRIV(ucd_caseless_sets) + code[2];
+  fprintf (f, "%s%sclist", before, not);
+  while (*p < NOTACHAR) fprintf(f, " %04x", *p++);
+  fprintf(f, "%s", after);
+#endif
+  }
+}
+
+
+
+
+/*************************************************
+*         Print compiled regex                   *
+*************************************************/
+
+/* Make this function work for a regex with integers either byte order.
+However, we assume that what we are passed is a compiled regex. The
+print_lengths flag controls whether offsets and lengths of items are printed.
+They can be turned off from pcretest so that automatic tests on bytecode can be
+written that do not depend on the value of LINK_SIZE. */
+
+#ifdef PCRE_INCLUDED
+static /* Keep the following function as private. */
+#endif
+#if defined COMPILE_PCRE8
+void
+pcre_printint(pcre *external_re, FILE *f, BOOL print_lengths)
+#elif defined COMPILE_PCRE16
+void
+pcre16_printint(pcre *external_re, FILE *f, BOOL print_lengths)
+#elif defined COMPILE_PCRE32
+void
+pcre32_printint(pcre *external_re, FILE *f, BOOL print_lengths)
+#endif
+{
+REAL_PCRE *re = (REAL_PCRE *)external_re;
+pcre_uchar *codestart, *code;
+BOOL utf;
+
+unsigned int options = re->options;
+int offset = re->name_table_offset;
+int count = re->name_count;
+int size = re->name_entry_size;
+
+if (re->magic_number != MAGIC_NUMBER)
+  {
+  offset = ((offset << 8) & 0xff00) | ((offset >> 8) & 0xff);
+  count = ((count << 8) & 0xff00) | ((count >> 8) & 0xff);
+  size = ((size << 8) & 0xff00) | ((size >> 8) & 0xff);
+  options = ((options << 24) & 0xff000000) |
+            ((options <<  8) & 0x00ff0000) |
+            ((options >>  8) & 0x0000ff00) |
+            ((options >> 24) & 0x000000ff);
+  }
+
+code = codestart = (pcre_uchar *)re + offset + count * size;
+/* PCRE_UTF(16|32) have the same value as PCRE_UTF8. */
+utf = (options & PCRE_UTF8) != 0;
+
+for(;;)
+  {
+  pcre_uchar *ccode;
+  const char *flag = "  ";
+  pcre_uint32 c;
+  unsigned int extra = 0;
+
+  if (print_lengths)
+    fprintf(f, "%3d ", (int)(code - codestart));
+  else
+    fprintf(f, "    ");
+
+  switch(*code)
+    {
+/* ========================================================================== */
+      /* These cases are never obeyed. This is a fudge that causes a compile-
+      time error if the vectors OP_names or OP_lengths, which are indexed
+      by opcode, are not the correct length. It seems to be the only way to do
+      such a check at compile time, as the sizeof() operator does not work in
+      the C preprocessor. */
+
+      case OP_TABLE_LENGTH:
+      case OP_TABLE_LENGTH +
+        ((sizeof(priv_OP_names)/sizeof(const char *) == OP_TABLE_LENGTH) &&
+        (sizeof(priv_OP_lengths) == OP_TABLE_LENGTH)):
+      break;
+/* ========================================================================== */
+
+    case OP_END:
+    fprintf(f, "    %s\n", priv_OP_names[*code]);
+    fprintf(f, "------------------------------------------------------------------\n");
+    return;
+
+    case OP_CHAR:
+    fprintf(f, "    ");
+    do
+      {
+      code++;
+      code += 1 + print_char(f, code, utf);
+      }
+    while (*code == OP_CHAR);
+    fprintf(f, "\n");
+    continue;
+
+    case OP_CHARI:
+    fprintf(f, " /i ");
+    do
+      {
+      code++;
+      code += 1 + print_char(f, code, utf);
+      }
+    while (*code == OP_CHARI);
+    fprintf(f, "\n");
+    continue;
+
+    case OP_CBRA:
+    case OP_CBRAPOS:
+    case OP_SCBRA:
+    case OP_SCBRAPOS:
+    if (print_lengths) fprintf(f, "%3d ", GET(code, 1));
+      else fprintf(f, "    ");
+    fprintf(f, "%s %d", priv_OP_names[*code], GET2(code, 1+LINK_SIZE));
+    break;
+
+    case OP_BRA:
+    case OP_BRAPOS:
+    case OP_SBRA:
+    case OP_SBRAPOS:
+    case OP_KETRMAX:
+    case OP_KETRMIN:
+    case OP_KETRPOS:
+    case OP_ALT:
+    case OP_KET:
+    case OP_ASSERT:
+    case OP_ASSERT_NOT:
+    case OP_ASSERTBACK:
+    case OP_ASSERTBACK_NOT:
+    case OP_ONCE:
+    case OP_ONCE_NC:
+    case OP_COND:
+    case OP_SCOND:
+    case OP_REVERSE:
+    if (print_lengths) fprintf(f, "%3d ", GET(code, 1));
+      else fprintf(f, "    ");
+    fprintf(f, "%s", priv_OP_names[*code]);
+    break;
+
+    case OP_CLOSE:
+    fprintf(f, "    %s %d", priv_OP_names[*code], GET2(code, 1));
+    break;
+
+    case OP_CREF:
+    case OP_NCREF:
+    fprintf(f, "%3d %s", GET2(code,1), priv_OP_names[*code]);
+    break;
+
+    case OP_RREF:
+    c = GET2(code, 1);
+    if (c == RREF_ANY)
+      fprintf(f, "    Cond recurse any");
+    else
+      fprintf(f, "    Cond recurse %d", c);
+    break;
+
+    case OP_NRREF:
+    c = GET2(code, 1);
+    if (c == RREF_ANY)
+      fprintf(f, "    Cond nrecurse any");
+    else
+      fprintf(f, "    Cond nrecurse %d", c);
+    break;
+
+    case OP_DEF:
+    fprintf(f, "    Cond def");
+    break;
+
+    case OP_STARI:
+    case OP_MINSTARI:
+    case OP_POSSTARI:
+    case OP_PLUSI:
+    case OP_MINPLUSI:
+    case OP_POSPLUSI:
+    case OP_QUERYI:
+    case OP_MINQUERYI:
+    case OP_POSQUERYI:
+    flag = "/i";
+    /* Fall through */
+    case OP_STAR:
+    case OP_MINSTAR:
+    case OP_POSSTAR:
+    case OP_PLUS:
+    case OP_MINPLUS:
+    case OP_POSPLUS:
+    case OP_QUERY:
+    case OP_MINQUERY:
+    case OP_POSQUERY:
+    case OP_TYPESTAR:
+    case OP_TYPEMINSTAR:
+    case OP_TYPEPOSSTAR:
+    case OP_TYPEPLUS:
+    case OP_TYPEMINPLUS:
+    case OP_TYPEPOSPLUS:
+    case OP_TYPEQUERY:
+    case OP_TYPEMINQUERY:
+    case OP_TYPEPOSQUERY:
+    fprintf(f, " %s ", flag);
+    if (*code >= OP_TYPESTAR)
+      {
+      if (code[1] == OP_PROP || code[1] == OP_NOTPROP)
+        {
+        print_prop(f, code + 1, "", " ");
+        extra = 2;
+        }
+      else fprintf(f, "%s", priv_OP_names[code[1]]);
+      }
+    else extra = print_char(f, code+1, utf);
+    fprintf(f, "%s", priv_OP_names[*code]);
+    break;
+
+    case OP_EXACTI:
+    case OP_UPTOI:
+    case OP_MINUPTOI:
+    case OP_POSUPTOI:
+    flag = "/i";
+    /* Fall through */
+    case OP_EXACT:
+    case OP_UPTO:
+    case OP_MINUPTO:
+    case OP_POSUPTO:
+    fprintf(f, " %s ", flag);
+    extra = print_char(f, code + 1 + IMM2_SIZE, utf);
+    fprintf(f, "{");
+    if (*code != OP_EXACT && *code != OP_EXACTI) fprintf(f, "0,");
+    fprintf(f, "%d}", GET2(code,1));
+    if (*code == OP_MINUPTO || *code == OP_MINUPTOI) fprintf(f, "?");
+      else if (*code == OP_POSUPTO || *code == OP_POSUPTOI) fprintf(f, "+");
+    break;
+
+    case OP_TYPEEXACT:
+    case OP_TYPEUPTO:
+    case OP_TYPEMINUPTO:
+    case OP_TYPEPOSUPTO:
+    if (code[1 + IMM2_SIZE] == OP_PROP || code[1 + IMM2_SIZE] == OP_NOTPROP)
+      {
+      print_prop(f, code + IMM2_SIZE + 1, "    ", " ");
+      extra = 2;
+      }
+    else fprintf(f, "    %s", priv_OP_names[code[1 + IMM2_SIZE]]);
+    fprintf(f, "{");
+    if (*code != OP_TYPEEXACT) fprintf(f, "0,");
+    fprintf(f, "%d}", GET2(code,1));
+    if (*code == OP_TYPEMINUPTO) fprintf(f, "?");
+      else if (*code == OP_TYPEPOSUPTO) fprintf(f, "+");
+    break;
+
+    case OP_NOTI:
+    flag = "/i";
+    /* Fall through */
+    case OP_NOT:
+    fprintf(f, " %s [^", flag);
+    extra = print_char(f, code + 1, utf);
+    fprintf(f, "]");
+    break;
+
+    case OP_NOTSTARI:
+    case OP_NOTMINSTARI:
+    case OP_NOTPOSSTARI:
+    case OP_NOTPLUSI:
+    case OP_NOTMINPLUSI:
+    case OP_NOTPOSPLUSI:
+    case OP_NOTQUERYI:
+    case OP_NOTMINQUERYI:
+    case OP_NOTPOSQUERYI:
+    flag = "/i";
+    /* Fall through */
+
+    case OP_NOTSTAR:
+    case OP_NOTMINSTAR:
+    case OP_NOTPOSSTAR:
+    case OP_NOTPLUS:
+    case OP_NOTMINPLUS:
+    case OP_NOTPOSPLUS:
+    case OP_NOTQUERY:
+    case OP_NOTMINQUERY:
+    case OP_NOTPOSQUERY:
+    fprintf(f, " %s [^", flag);
+    extra = print_char(f, code + 1, utf);
+    fprintf(f, "]%s", priv_OP_names[*code]);
+    break;
+
+    case OP_NOTEXACTI:
+    case OP_NOTUPTOI:
+    case OP_NOTMINUPTOI:
+    case OP_NOTPOSUPTOI:
+    flag = "/i";
+    /* Fall through */
+
+    case OP_NOTEXACT:
+    case OP_NOTUPTO:
+    case OP_NOTMINUPTO:
+    case OP_NOTPOSUPTO:
+    fprintf(f, " %s [^", flag);
+    extra = print_char(f, code + 1 + IMM2_SIZE, utf);
+    fprintf(f, "]{");
+    if (*code != OP_NOTEXACT && *code != OP_NOTEXACTI) fprintf(f, "0,");
+    fprintf(f, "%d}", GET2(code,1));
+    if (*code == OP_NOTMINUPTO || *code == OP_NOTMINUPTOI) fprintf(f, "?");
+      else
+    if (*code == OP_NOTPOSUPTO || *code == OP_NOTPOSUPTOI) fprintf(f, "+");
+    break;
+
+    case OP_RECURSE:
+    if (print_lengths) fprintf(f, "%3d ", GET(code, 1));
+      else fprintf(f, "    ");
+    fprintf(f, "%s", priv_OP_names[*code]);
+    break;
+
+    case OP_REFI:
+    flag = "/i";
+    /* Fall through */
+    case OP_REF:
+    fprintf(f, " %s \\%d", flag, GET2(code,1));
+    ccode = code + priv_OP_lengths[*code];
+    goto CLASS_REF_REPEAT;
+
+    case OP_CALLOUT:
+    fprintf(f, "    %s %d %d %d", priv_OP_names[*code], code[1], GET(code,2),
+      GET(code, 2 + LINK_SIZE));
+    break;
+
+    case OP_PROP:
+    case OP_NOTPROP:
+    print_prop(f, code, "    ", "");
+    break;
+
+    /* OP_XCLASS can only occur in UTF or PCRE16 modes. However, there's no
+    harm in having this code always here, and it makes it less messy without
+    all those #ifdefs. */
+
+    case OP_CLASS:
+    case OP_NCLASS:
+    case OP_XCLASS:
+      {
+      int i;
+      unsigned int min, max;
+      BOOL printmap;
+      pcre_uint8 *map;
+
+      fprintf(f, "    [");
+
+      if (*code == OP_XCLASS)
+        {
+        extra = GET(code, 1);
+        ccode = code + LINK_SIZE + 1;
+        printmap = (*ccode & XCL_MAP) != 0;
+        if ((*ccode++ & XCL_NOT) != 0) fprintf(f, "^");
+        }
+      else
+        {
+        printmap = TRUE;
+        ccode = code + 1;
+        }
+
+      /* Print a bit map */
+
+      if (printmap)
+        {
+        map = (pcre_uint8 *)ccode;
+        for (i = 0; i < 256; i++)
+          {
+          if ((map[i/8] & (1 << (i&7))) != 0)
+            {
+            int j;
+            for (j = i+1; j < 256; j++)
+              if ((map[j/8] & (1 << (j&7))) == 0) break;
+            if (i == '-' || i == ']') fprintf(f, "\\");
+            if (PRINTABLE(i)) fprintf(f, "%c", i);
+              else fprintf(f, "\\x%02x", i);
+            if (--j > i)
+              {
+              if (j != i + 1) fprintf(f, "-");
+              if (j == '-' || j == ']') fprintf(f, "\\");
+              if (PRINTABLE(j)) fprintf(f, "%c", j);
+                else fprintf(f, "\\x%02x", j);
+              }
+            i = j;
+            }
+          }
+        ccode += 32 / sizeof(pcre_uchar);
+        }
+
+      /* For an XCLASS there is always some additional data */
+
+      if (*code == OP_XCLASS)
+        {
+        pcre_uchar ch;
+        while ((ch = *ccode++) != XCL_END)
+          {
+          if (ch == XCL_PROP)
+            {
+            unsigned int ptype = *ccode++;
+            unsigned int pvalue = *ccode++;
+            fprintf(f, "\\p{%s}", get_ucpname(ptype, pvalue));
+            }
+          else if (ch == XCL_NOTPROP)
+            {
+            unsigned int ptype = *ccode++;
+            unsigned int pvalue = *ccode++;
+            fprintf(f, "\\P{%s}", get_ucpname(ptype, pvalue));
+            }
+          else
+            {
+            ccode += 1 + print_char(f, ccode, utf);
+            if (ch == XCL_RANGE)
+              {
+              fprintf(f, "-");
+              ccode += 1 + print_char(f, ccode, utf);
+              }
+            }
+          }
+        }
+
+      /* Indicate a non-UTF class which was created by negation */
+
+      fprintf(f, "]%s", (*code == OP_NCLASS)? " (neg)" : "");
+
+      /* Handle repeats after a class or a back reference */
+
+      CLASS_REF_REPEAT:
+      switch(*ccode)
+        {
+        case OP_CRSTAR:
+        case OP_CRMINSTAR:
+        case OP_CRPLUS:
+        case OP_CRMINPLUS:
+        case OP_CRQUERY:
+        case OP_CRMINQUERY:
+        fprintf(f, "%s", priv_OP_names[*ccode]);
+        extra += priv_OP_lengths[*ccode];
+        break;
+
+        case OP_CRRANGE:
+        case OP_CRMINRANGE:
+        min = GET2(ccode,1);
+        max = GET2(ccode,1 + IMM2_SIZE);
+        if (max == 0) fprintf(f, "{%u,}", min);
+        else fprintf(f, "{%u,%u}", min, max);
+        if (*ccode == OP_CRMINRANGE) fprintf(f, "?");
+        extra += priv_OP_lengths[*ccode];
+        break;
+
+        /* Do nothing if it's not a repeat; this code stops picky compilers
+        warning about the lack of a default code path. */
+
+        default:
+        break;
+        }
+      }
+    break;
+
+    case OP_MARK:
+    case OP_PRUNE_ARG:
+    case OP_SKIP_ARG:
+    case OP_THEN_ARG:
+    fprintf(f, "    %s ", priv_OP_names[*code]);
+    print_puchar(f, code + 2);
+    extra += code[1];
+    break;
+
+    case OP_THEN:
+    fprintf(f, "    %s", priv_OP_names[*code]);
+    break;
+
+    case OP_CIRCM:
+    case OP_DOLLM:
+    flag = "/m";
+    /* Fall through */
+
+    /* Anything else is just an item with no data, but possibly a flag. */
+
+    default:
+    fprintf(f, " %s %s", flag, priv_OP_names[*code]);
+    break;
+    }
+
+  code += priv_OP_lengths[*code] + extra;
+  fprintf(f, "\n");
+  }
+}
+
+/* End of pcre_printint.src */
diff --git a/pcre/pcre_refcount.c b/pcre/pcre_refcount.c
new file mode 100644
index 00000000000..79efa90f216
--- /dev/null
+++ b/pcre/pcre_refcount.c
@@ -0,0 +1,92 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+
+/* This module contains the external function pcre_refcount(), which is an
+auxiliary function that can be used to maintain a reference count in a compiled
+pattern data block. This might be helpful in applications where the block is
+shared by different users. */
+
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include "pcre_internal.h"
+
+
+/*************************************************
+*           Maintain reference count             *
+*************************************************/
+
+/* The reference count is a 16-bit field, initialized to zero. It is not
+possible to transfer a non-zero count from one host to a different host that
+has a different byte order - though I can't see why anyone in their right mind
+would ever want to do that!
+
+Arguments:
+  argument_re   points to compiled code
+  adjust        value to add to the count
+
+Returns:        the (possibly updated) count value (a non-negative number), or
+                a negative error number
+*/
+
+#if defined COMPILE_PCRE8
+PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
+pcre_refcount(pcre *argument_re, int adjust)
+#elif defined COMPILE_PCRE16
+PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
+pcre16_refcount(pcre16 *argument_re, int adjust)
+#elif defined COMPILE_PCRE32
+PCRE_EXP_DEFN int PCRE_CALL_CONVENTION
+pcre32_refcount(pcre32 *argument_re, int adjust)
+#endif
+{
+REAL_PCRE *re = (REAL_PCRE *)argument_re;
+if (re == NULL) return PCRE_ERROR_NULL;
+if (re->magic_number != MAGIC_NUMBER) return PCRE_ERROR_BADMAGIC;
+if ((re->flags & PCRE_MODE) == 0) return PCRE_ERROR_BADMODE;
+re->ref_count = (-adjust > re->ref_count)? 0 :
+                (adjust + re->ref_count > 65535)? 65535 :
+                re->ref_count + adjust;
+return re->ref_count;
+}
+
+/* End of pcre_refcount.c */
diff --git a/pcre/pcre_scanner.cc b/pcre/pcre_scanner.cc
new file mode 100644
index 00000000000..6be2be6829b
--- /dev/null
+++ b/pcre/pcre_scanner.cc
@@ -0,0 +1,199 @@
+// Copyright (c) 2005, Google Inc.
+// All rights reserved.
+//
+// Redistribution and use in source and binary forms, with or without
+// modification, are permitted provided that the following conditions are
+// met:
+//
+//     * Redistributions of source code must retain the above copyright
+// notice, this list of conditions and the following disclaimer.
+//     * Redistributions in binary form must reproduce the above
+// copyright notice, this list of conditions and the following disclaimer
+// in the documentation and/or other materials provided with the
+// distribution.
+//     * Neither the name of Google Inc. nor the names of its
+// contributors may be used to endorse or promote products derived from
+// this software without specific prior written permission.
+//
+// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+//
+// Author: Sanjay Ghemawat
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include 
+#include 
+
+#include "pcrecpp_internal.h"
+#include "pcre_scanner.h"
+
+using std::vector;
+
+namespace pcrecpp {
+
+Scanner::Scanner()
+  : data_(),
+    input_(data_),
+    skip_(NULL),
+    should_skip_(false),
+    skip_repeat_(false),
+    save_comments_(false),
+    comments_(NULL),
+    comments_offset_(0) {
+}
+
+Scanner::Scanner(const string& in)
+  : data_(in),
+    input_(data_),
+    skip_(NULL),
+    should_skip_(false),
+    skip_repeat_(false),
+    save_comments_(false),
+    comments_(NULL),
+    comments_offset_(0) {
+}
+
+Scanner::~Scanner() {
+  delete skip_;
+  delete comments_;
+}
+
+void Scanner::SetSkipExpression(const char* re) {
+  delete skip_;
+  if (re != NULL) {
+    skip_ = new RE(re);
+    should_skip_ = true;
+    skip_repeat_ = true;
+    ConsumeSkip();
+  } else {
+    skip_ = NULL;
+    should_skip_ = false;
+    skip_repeat_ = false;
+  }
+}
+
+void Scanner::Skip(const char* re) {
+  delete skip_;
+  if (re != NULL) {
+    skip_ = new RE(re);
+    should_skip_ = true;
+    skip_repeat_ = false;
+    ConsumeSkip();
+  } else {
+    skip_ = NULL;
+    should_skip_ = false;
+    skip_repeat_ = false;
+  }
+}
+
+void Scanner::DisableSkip() {
+  assert(skip_ != NULL);
+  should_skip_ = false;
+}
+
+void Scanner::EnableSkip() {
+  assert(skip_ != NULL);
+  should_skip_ = true;
+  ConsumeSkip();
+}
+
+int Scanner::LineNumber() const {
+  // TODO: Make it more efficient by keeping track of the last point
+  // where we computed line numbers and counting newlines since then.
+  // We could use std:count, but not all systems have it. :-(
+  int count = 1;
+  for (const char* p = data_.data(); p < input_.data(); ++p)
+    if (*p == '\n')
+      ++count;
+  return count;
+}
+
+int Scanner::Offset() const {
+  return (int)(input_.data() - data_.c_str());
+}
+
+bool Scanner::LookingAt(const RE& re) const {
+  int consumed;
+  return re.DoMatch(input_, RE::ANCHOR_START, &consumed, 0, 0);
+}
+
+
+bool Scanner::Consume(const RE& re,
+                      const Arg& arg0,
+                      const Arg& arg1,
+                      const Arg& arg2) {
+  const bool result = re.Consume(&input_, arg0, arg1, arg2);
+  if (result && should_skip_) ConsumeSkip();
+  return result;
+}
+
+// helper function to consume *skip_ and honour save_comments_
+void Scanner::ConsumeSkip() {
+  const char* start_data = input_.data();
+  while (skip_->Consume(&input_)) {
+    if (!skip_repeat_) {
+      // Only one skip allowed.
+      break;
+    }
+  }
+  if (save_comments_) {
+    if (comments_ == NULL) {
+      comments_ = new vector;
+    }
+    // already pointing one past end, so no need to +1
+    int length = (int)(input_.data() - start_data);
+    if (length > 0) {
+      comments_->push_back(StringPiece(start_data, length));
+    }
+  }
+}
+
+
+void Scanner::GetComments(int start, int end, vector *ranges) {
+  // short circuit out if we've not yet initialized comments_
+  // (e.g., when save_comments is false)
+  if (!comments_) {
+    return;
+  }
+  // TODO: if we guarantee that comments_ will contain StringPieces
+  // that are ordered by their start, then we can do a binary search
+  // for the first StringPiece at or past start and then scan for the
+  // ones contained in the range, quit early (use equal_range or
+  // lower_bound)
+  for (vector::const_iterator it = comments_->begin();
+       it != comments_->end(); ++it) {
+    if ((it->data() >= data_.c_str() + start &&
+         it->data() + it->size() <= data_.c_str() + end)) {
+      ranges->push_back(*it);
+    }
+  }
+}
+
+
+void Scanner::GetNextComments(vector *ranges) {
+  // short circuit out if we've not yet initialized comments_
+  // (e.g., when save_comments is false)
+  if (!comments_) {
+    return;
+  }
+  for (vector::const_iterator it =
+         comments_->begin() + comments_offset_;
+       it != comments_->end(); ++it) {
+    ranges->push_back(*it);
+    ++comments_offset_;
+  }
+}
+
+}   // namespace pcrecpp
diff --git a/pcre/pcre_scanner.h b/pcre/pcre_scanner.h
new file mode 100644
index 00000000000..5617e4515cb
--- /dev/null
+++ b/pcre/pcre_scanner.h
@@ -0,0 +1,172 @@
+// Copyright (c) 2005, Google Inc.
+// All rights reserved.
+//
+// Redistribution and use in source and binary forms, with or without
+// modification, are permitted provided that the following conditions are
+// met:
+//
+//     * Redistributions of source code must retain the above copyright
+// notice, this list of conditions and the following disclaimer.
+//     * Redistributions in binary form must reproduce the above
+// copyright notice, this list of conditions and the following disclaimer
+// in the documentation and/or other materials provided with the
+// distribution.
+//     * Neither the name of Google Inc. nor the names of its
+// contributors may be used to endorse or promote products derived from
+// this software without specific prior written permission.
+//
+// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+//
+// Author: Sanjay Ghemawat
+//
+// Regular-expression based scanner for parsing an input stream.
+//
+// Example 1: parse a sequence of "var = number" entries from input:
+//
+//      Scanner scanner(input);
+//      string var;
+//      int number;
+//      scanner.SetSkipExpression("\\s+"); // Skip any white space we encounter
+//      while (scanner.Consume("(\\w+) = (\\d+)", &var, &number)) {
+//        ...;
+//      }
+
+#ifndef _PCRE_SCANNER_H
+#define _PCRE_SCANNER_H
+
+#include 
+#include 
+#include 
+
+#include 
+#include 
+
+namespace pcrecpp {
+
+class PCRECPP_EXP_DEFN Scanner {
+ public:
+  Scanner();
+  explicit Scanner(const std::string& input);
+  ~Scanner();
+
+  // Return current line number.  The returned line-number is
+  // one-based.  I.e. it returns 1 + the number of consumed newlines.
+  //
+  // Note: this method may be slow.  It may take time proportional to
+  // the size of the input.
+  int LineNumber() const;
+
+  // Return the byte-offset that the scanner is looking in the
+  // input data;
+  int Offset() const;
+
+  // Return true iff the start of the remaining input matches "re"
+  bool LookingAt(const RE& re) const;
+
+  // Return true iff all of the following are true
+  //    a. the start of the remaining input matches "re",
+  //    b. if any arguments are supplied, matched sub-patterns can be
+  //       parsed and stored into the arguments.
+  // If it returns true, it skips over the matched input and any
+  // following input that matches the "skip" regular expression.
+  bool Consume(const RE& re,
+               const Arg& arg0 = RE::no_arg,
+               const Arg& arg1 = RE::no_arg,
+               const Arg& arg2 = RE::no_arg
+               // TODO: Allow more arguments?
+               );
+
+  // Set the "skip" regular expression.  If after consuming some data,
+  // a prefix of the input matches this RE, it is automatically
+  // skipped.  For example, a programming language scanner would use
+  // a skip RE that matches white space and comments.
+  //
+  //    scanner.SetSkipExpression("\\s+|//.*|/[*](.|\n)*?[*]/");
+  //
+  // Skipping repeats as long as it succeeds.  We used to let people do
+  // this by writing "(...)*" in the regular expression, but that added
+  // up to lots of recursive calls within the pcre library, so now we
+  // control repetition explicitly via the function call API.
+  //
+  // You can pass NULL for "re" if you do not want any data to be skipped.
+  void Skip(const char* re);   // DEPRECATED; does *not* repeat
+  void SetSkipExpression(const char* re);
+
+  // Temporarily pause "skip"ing. This
+  //   Skip("Foo"); code ; DisableSkip(); code; EnableSkip()
+  // is similar to
+  //   Skip("Foo"); code ; Skip(NULL); code ; Skip("Foo");
+  // but avoids creating/deleting new RE objects.
+  void DisableSkip();
+
+  // Reenable previously paused skipping.  Any prefix of the input
+  // that matches the skip pattern is immediately dropped.
+  void EnableSkip();
+
+  /***** Special wrappers around SetSkip() for some common idioms *****/
+
+  // Arranges to skip whitespace, C comments, C++ comments.
+  // The overall RE is a disjunction of the following REs:
+  //    \\s                     whitespace
+  //    //.*\n                  C++ comment
+  //    /[*](.|\n)*?[*]/        C comment (x*? means minimal repetitions of x)
+  // We get repetition via the semantics of SetSkipExpression, not by using *
+  void SkipCXXComments() {
+    SetSkipExpression("\\s|//.*\n|/[*](?:\n|.)*?[*]/");
+  }
+
+  void set_save_comments(bool comments) {
+    save_comments_ = comments;
+  }
+
+  bool save_comments() {
+    return save_comments_;
+  }
+
+  // Append to vector ranges the comments found in the
+  // byte range [start,end] (inclusive) of the input data.
+  // Only comments that were extracted entirely within that
+  // range are returned: no range splitting of atomically-extracted
+  // comments is performed.
+  void GetComments(int start, int end, std::vector *ranges);
+
+  // Append to vector ranges the comments added
+  // since the last time this was called. This
+  // functionality is provided for efficiency when
+  // interleaving scanning with parsing.
+  void GetNextComments(std::vector *ranges);
+
+ private:
+  std::string   data_;          // All the input data
+  StringPiece   input_;         // Unprocessed input
+  RE*           skip_;          // If non-NULL, RE for skipping input
+  bool          should_skip_;   // If true, use skip_
+  bool          skip_repeat_;   // If true, repeat skip_ as long as it works
+  bool          save_comments_; // If true, aggregate the skip expression
+
+  // the skipped comments
+  // TODO: later consider requiring that the StringPieces be added
+  // in order by their start position
+  std::vector *comments_;
+
+  // the offset into comments_ that has been returned by GetNextComments
+  int           comments_offset_;
+
+  // helper function to consume *skip_ and honour
+  // save_comments_
+  void ConsumeSkip();
+};
+
+}   // namespace pcrecpp
+
+#endif /* _PCRE_SCANNER_H */
diff --git a/pcre/pcre_scanner_unittest.cc b/pcre/pcre_scanner_unittest.cc
new file mode 100644
index 00000000000..7de8d2e8d9a
--- /dev/null
+++ b/pcre/pcre_scanner_unittest.cc
@@ -0,0 +1,159 @@
+// Copyright (c) 2005, Google Inc.
+// All rights reserved.
+//
+// Redistribution and use in source and binary forms, with or without
+// modification, are permitted provided that the following conditions are
+// met:
+//
+//     * Redistributions of source code must retain the above copyright
+// notice, this list of conditions and the following disclaimer.
+//     * Redistributions in binary form must reproduce the above
+// copyright notice, this list of conditions and the following disclaimer
+// in the documentation and/or other materials provided with the
+// distribution.
+//     * Neither the name of Google Inc. nor the names of its
+// contributors may be used to endorse or promote products derived from
+// this software without specific prior written permission.
+//
+// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+//
+// Author: Greg J. Badros
+//
+// Unittest for scanner, especially GetNextComments and GetComments()
+// functionality.
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include 
+#include       /* for strchr */
+#include 
+#include 
+
+#include "pcrecpp.h"
+#include "pcre_stringpiece.h"
+#include "pcre_scanner.h"
+
+#define FLAGS_unittest_stack_size   49152
+
+// Dies with a fatal error if the two values are not equal.
+#define CHECK_EQ(a, b)  do {                                    \
+  if ( (a) != (b) ) {                                           \
+    fprintf(stderr, "%s:%d: Check failed because %s != %s\n",   \
+            __FILE__, __LINE__, #a, #b);                        \
+    exit(1);                                                    \
+  }                                                             \
+} while (0)
+
+using std::vector;
+using pcrecpp::StringPiece;
+using pcrecpp::Scanner;
+
+static void TestScanner() {
+  const char input[] = "\n"
+                       "alpha = 1; // this sets alpha\n"
+                       "bravo = 2; // bravo is set here\n"
+                       "gamma = 33; /* and here is gamma */\n";
+
+  const char *re = "(\\w+) = (\\d+);";
+
+  Scanner s(input);
+  string var;
+  int number;
+  s.SkipCXXComments();
+  s.set_save_comments(true);
+  vector comments;
+
+  s.Consume(re, &var, &number);
+  CHECK_EQ(var, "alpha");
+  CHECK_EQ(number, 1);
+  CHECK_EQ(s.LineNumber(), 3);
+  s.GetNextComments(&comments);
+  CHECK_EQ(comments.size(), 1);
+  CHECK_EQ(comments[0].as_string(), " // this sets alpha\n");
+  comments.resize(0);
+
+  s.Consume(re, &var, &number);
+  CHECK_EQ(var, "bravo");
+  CHECK_EQ(number, 2);
+  s.GetNextComments(&comments);
+  CHECK_EQ(comments.size(), 1);
+  CHECK_EQ(comments[0].as_string(), " // bravo is set here\n");
+  comments.resize(0);
+
+  s.Consume(re, &var, &number);
+  CHECK_EQ(var, "gamma");
+  CHECK_EQ(number, 33);
+  s.GetNextComments(&comments);
+  CHECK_EQ(comments.size(), 1);
+  CHECK_EQ(comments[0].as_string(), " /* and here is gamma */\n");
+  comments.resize(0);
+
+  s.GetComments(0, sizeof(input), &comments);
+  CHECK_EQ(comments.size(), 3);
+  CHECK_EQ(comments[0].as_string(), " // this sets alpha\n");
+  CHECK_EQ(comments[1].as_string(), " // bravo is set here\n");
+  CHECK_EQ(comments[2].as_string(), " /* and here is gamma */\n");
+  comments.resize(0);
+
+  s.GetComments(0, (int)(strchr(input, '/') - input), &comments);
+  CHECK_EQ(comments.size(), 0);
+  comments.resize(0);
+
+  s.GetComments((int)(strchr(input, '/') - input - 1), sizeof(input),
+                &comments);
+  CHECK_EQ(comments.size(), 3);
+  CHECK_EQ(comments[0].as_string(), " // this sets alpha\n");
+  CHECK_EQ(comments[1].as_string(), " // bravo is set here\n");
+  CHECK_EQ(comments[2].as_string(), " /* and here is gamma */\n");
+  comments.resize(0);
+
+  s.GetComments((int)(strchr(input, '/') - input - 1),
+                (int)(strchr(input + 1, '\n') - input + 1), &comments);
+  CHECK_EQ(comments.size(), 1);
+  CHECK_EQ(comments[0].as_string(), " // this sets alpha\n");
+  comments.resize(0);
+}
+
+static void TestBigComment() {
+  string input;
+  for (int i = 0; i < 1024; ++i) {
+    char buf[1024];  // definitely big enough
+    sprintf(buf, "    # Comment %d\n", i);
+    input += buf;
+  }
+  input += "name = value;\n";
+
+  Scanner s(input.c_str());
+  s.SetSkipExpression("\\s+|#.*\n");
+
+  string name;
+  string value;
+  s.Consume("(\\w+) = (\\w+);", &name, &value);
+  CHECK_EQ(name, "name");
+  CHECK_EQ(value, "value");
+}
+
+// TODO: also test scanner and big-comment in a thread with a
+//       small stack size
+
+int main(int argc, char** argv) {
+  TestScanner();
+  TestBigComment();
+
+  // Done
+  printf("OK\n");
+
+  return 0;
+}
diff --git a/pcre/pcre_string_utils.c b/pcre/pcre_string_utils.c
new file mode 100644
index 00000000000..10b53d5bcb4
--- /dev/null
+++ b/pcre/pcre_string_utils.c
@@ -0,0 +1,211 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2013 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+
+/* This module contains internal functions for comparing and finding the length
+of strings for different data item sizes. */
+
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include "pcre_internal.h"
+
+#ifndef COMPILE_PCRE8
+
+/*************************************************
+*           Compare string utilities             *
+*************************************************/
+
+/* The following two functions compares two strings. Basically a strcmp
+for non 8 bit characters.
+
+Arguments:
+  str1        first string
+  str2        second string
+
+Returns:      0 if both string are equal (like strcmp), 1 otherwise
+*/
+
+int
+PRIV(strcmp_uc_uc)(const pcre_uchar *str1, const pcre_uchar *str2)
+{
+pcre_uchar c1;
+pcre_uchar c2;
+
+while (*str1 != '\0' || *str2 != '\0')
+  {
+  c1 = *str1++;
+  c2 = *str2++;
+  if (c1 != c2)
+    return ((c1 > c2) << 1) - 1;
+  }
+/* Both length and characters must be equal. */
+return 0;
+}
+
+#ifdef COMPILE_PCRE32
+
+int
+PRIV(strcmp_uc_uc_utf)(const pcre_uchar *str1, const pcre_uchar *str2)
+{
+pcre_uchar c1;
+pcre_uchar c2;
+
+while (*str1 != '\0' || *str2 != '\0')
+  {
+  c1 = RAWUCHARINC(str1);
+  c2 = RAWUCHARINC(str2);
+  if (c1 != c2)
+    return ((c1 > c2) << 1) - 1;
+  }
+/* Both length and characters must be equal. */
+return 0;
+}
+
+#endif /* COMPILE_PCRE32 */
+
+int
+PRIV(strcmp_uc_c8)(const pcre_uchar *str1, const char *str2)
+{
+const pcre_uint8 *ustr2 = (pcre_uint8 *)str2;
+pcre_uchar c1;
+pcre_uchar c2;
+
+while (*str1 != '\0' || *ustr2 != '\0')
+  {
+  c1 = *str1++;
+  c2 = (pcre_uchar)*ustr2++;
+  if (c1 != c2)
+    return ((c1 > c2) << 1) - 1;
+  }
+/* Both length and characters must be equal. */
+return 0;
+}
+
+#ifdef COMPILE_PCRE32
+
+int
+PRIV(strcmp_uc_c8_utf)(const pcre_uchar *str1, const char *str2)
+{
+const pcre_uint8 *ustr2 = (pcre_uint8 *)str2;
+pcre_uchar c1;
+pcre_uchar c2;
+
+while (*str1 != '\0' || *ustr2 != '\0')
+  {
+  c1 = RAWUCHARINC(str1);
+  c2 = (pcre_uchar)*ustr2++;
+  if (c1 != c2)
+    return ((c1 > c2) << 1) - 1;
+  }
+/* Both length and characters must be equal. */
+return 0;
+}
+
+#endif /* COMPILE_PCRE32 */
+
+/* The following two functions compares two, fixed length
+strings. Basically an strncmp for non 8 bit characters.
+
+Arguments:
+  str1        first string
+  str2        second string
+  num         size of the string
+
+Returns:      0 if both string are equal (like strcmp), 1 otherwise
+*/
+
+int
+PRIV(strncmp_uc_uc)(const pcre_uchar *str1, const pcre_uchar *str2, unsigned int num)
+{
+pcre_uchar c1;
+pcre_uchar c2;
+
+while (num-- > 0)
+  {
+  c1 = *str1++;
+  c2 = *str2++;
+  if (c1 != c2)
+    return ((c1 > c2) << 1) - 1;
+  }
+/* Both length and characters must be equal. */
+return 0;
+}
+
+int
+PRIV(strncmp_uc_c8)(const pcre_uchar *str1, const char *str2, unsigned int num)
+{
+const pcre_uint8 *ustr2 = (pcre_uint8 *)str2;
+pcre_uchar c1;
+pcre_uchar c2;
+
+while (num-- > 0)
+  {
+  c1 = *str1++;
+  c2 = (pcre_uchar)*ustr2++;
+  if (c1 != c2)
+    return ((c1 > c2) << 1) - 1;
+  }
+/* Both length and characters must be equal. */
+return 0;
+}
+
+/* The following function returns with the length of
+a zero terminated string. Basically an strlen for non 8 bit characters.
+
+Arguments:
+  str         string
+
+Returns:      length of the string
+*/
+
+unsigned int
+PRIV(strlen_uc)(const pcre_uchar *str)
+{
+unsigned int len = 0;
+while (*str++ != 0)
+  len++;
+return len;
+}
+
+#endif /* !COMPILE_PCRE8 */
+
+/* End of pcre_string_utils.c */
diff --git a/pcre/pcre_stringpiece.cc b/pcre/pcre_stringpiece.cc
new file mode 100644
index 00000000000..67c0f1fc0e5
--- /dev/null
+++ b/pcre/pcre_stringpiece.cc
@@ -0,0 +1,43 @@
+// Copyright (c) 2005, Google Inc.
+// All rights reserved.
+//
+// Redistribution and use in source and binary forms, with or without
+// modification, are permitted provided that the following conditions are
+// met:
+//
+//     * Redistributions of source code must retain the above copyright
+// notice, this list of conditions and the following disclaimer.
+//     * Redistributions in binary form must reproduce the above
+// copyright notice, this list of conditions and the following disclaimer
+// in the documentation and/or other materials provided with the
+// distribution.
+//     * Neither the name of Google Inc. nor the names of its
+// contributors may be used to endorse or promote products derived from
+// this software without specific prior written permission.
+//
+// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+//
+// Author: wilsonh@google.com (Wilson Hsieh)
+//
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include 
+#include "pcrecpp_internal.h"
+#include "pcre_stringpiece.h"
+
+std::ostream& operator<<(std::ostream& o, const pcrecpp::StringPiece& piece) {
+  return (o << piece.as_string());
+}
diff --git a/pcre/pcre_stringpiece.h b/pcre/pcre_stringpiece.h
new file mode 100644
index 00000000000..4ea32f62261
--- /dev/null
+++ b/pcre/pcre_stringpiece.h
@@ -0,0 +1,179 @@
+// Copyright (c) 2005, Google Inc.
+// All rights reserved.
+//
+// Redistribution and use in source and binary forms, with or without
+// modification, are permitted provided that the following conditions are
+// met:
+//
+//     * Redistributions of source code must retain the above copyright
+// notice, this list of conditions and the following disclaimer.
+//     * Redistributions in binary form must reproduce the above
+// copyright notice, this list of conditions and the following disclaimer
+// in the documentation and/or other materials provided with the
+// distribution.
+//     * Neither the name of Google Inc. nor the names of its
+// contributors may be used to endorse or promote products derived from
+// this software without specific prior written permission.
+//
+// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+//
+// Author: Sanjay Ghemawat
+//
+// A string like object that points into another piece of memory.
+// Useful for providing an interface that allows clients to easily
+// pass in either a "const char*" or a "string".
+//
+// Arghh!  I wish C++ literals were automatically of type "string".
+
+#ifndef _PCRE_STRINGPIECE_H
+#define _PCRE_STRINGPIECE_H
+
+#include 
+#include 
+#include     // for ostream forward-declaration
+
+#if 0
+#define HAVE_TYPE_TRAITS
+#include 
+#elif 0
+#define HAVE_TYPE_TRAITS
+#include 
+#endif
+
+#include 
+
+using std::memcmp;
+using std::strlen;
+using std::string;
+
+namespace pcrecpp {
+
+class PCRECPP_EXP_DEFN StringPiece {
+ private:
+  const char*   ptr_;
+  int           length_;
+
+ public:
+  // We provide non-explicit singleton constructors so users can pass
+  // in a "const char*" or a "string" wherever a "StringPiece" is
+  // expected.
+  StringPiece()
+    : ptr_(NULL), length_(0) { }
+  StringPiece(const char* str)
+    : ptr_(str), length_(static_cast(strlen(ptr_))) { }
+  StringPiece(const unsigned char* str)
+    : ptr_(reinterpret_cast(str)),
+      length_(static_cast(strlen(ptr_))) { }
+  StringPiece(const string& str)
+    : ptr_(str.data()), length_(static_cast(str.size())) { }
+  StringPiece(const char* offset, int len)
+    : ptr_(offset), length_(len) { }
+
+  // data() may return a pointer to a buffer with embedded NULs, and the
+  // returned buffer may or may not be null terminated.  Therefore it is
+  // typically a mistake to pass data() to a routine that expects a NUL
+  // terminated string.  Use "as_string().c_str()" if you really need to do
+  // this.  Or better yet, change your routine so it does not rely on NUL
+  // termination.
+  const char* data() const { return ptr_; }
+  int size() const { return length_; }
+  bool empty() const { return length_ == 0; }
+
+  void clear() { ptr_ = NULL; length_ = 0; }
+  void set(const char* buffer, int len) { ptr_ = buffer; length_ = len; }
+  void set(const char* str) {
+    ptr_ = str;
+    length_ = static_cast(strlen(str));
+  }
+  void set(const void* buffer, int len) {
+    ptr_ = reinterpret_cast(buffer);
+    length_ = len;
+  }
+
+  char operator[](int i) const { return ptr_[i]; }
+
+  void remove_prefix(int n) {
+    ptr_ += n;
+    length_ -= n;
+  }
+
+  void remove_suffix(int n) {
+    length_ -= n;
+  }
+
+  bool operator==(const StringPiece& x) const {
+    return ((length_ == x.length_) &&
+            (memcmp(ptr_, x.ptr_, length_) == 0));
+  }
+  bool operator!=(const StringPiece& x) const {
+    return !(*this == x);
+  }
+
+#define STRINGPIECE_BINARY_PREDICATE(cmp,auxcmp)                             \
+  bool operator cmp (const StringPiece& x) const {                           \
+    int r = memcmp(ptr_, x.ptr_, length_ < x.length_ ? length_ : x.length_); \
+    return ((r auxcmp 0) || ((r == 0) && (length_ cmp x.length_)));          \
+  }
+  STRINGPIECE_BINARY_PREDICATE(<,  <);
+  STRINGPIECE_BINARY_PREDICATE(<=, <);
+  STRINGPIECE_BINARY_PREDICATE(>=, >);
+  STRINGPIECE_BINARY_PREDICATE(>,  >);
+#undef STRINGPIECE_BINARY_PREDICATE
+
+  int compare(const StringPiece& x) const {
+    int r = memcmp(ptr_, x.ptr_, length_ < x.length_ ? length_ : x.length_);
+    if (r == 0) {
+      if (length_ < x.length_) r = -1;
+      else if (length_ > x.length_) r = +1;
+    }
+    return r;
+  }
+
+  string as_string() const {
+    return string(data(), size());
+  }
+
+  void CopyToString(string* target) const {
+    target->assign(ptr_, length_);
+  }
+
+  // Does "this" start with "x"
+  bool starts_with(const StringPiece& x) const {
+    return ((length_ >= x.length_) && (memcmp(ptr_, x.ptr_, x.length_) == 0));
+  }
+};
+
+}   // namespace pcrecpp
+
+// ------------------------------------------------------------------
+// Functions used to create STL containers that use StringPiece
+//  Remember that a StringPiece's lifetime had better be less than
+//  that of the underlying string or char*.  If it is not, then you
+//  cannot safely store a StringPiece into an STL container
+// ------------------------------------------------------------------
+
+#ifdef HAVE_TYPE_TRAITS
+// This makes vector really fast for some STL implementations
+template<> struct __type_traits {
+  typedef __true_type    has_trivial_default_constructor;
+  typedef __true_type    has_trivial_copy_constructor;
+  typedef __true_type    has_trivial_assignment_operator;
+  typedef __true_type    has_trivial_destructor;
+  typedef __true_type    is_POD_type;
+};
+#endif
+
+// allow StringPiece to be logged
+std::ostream& operator<<(std::ostream& o, const pcrecpp::StringPiece& piece);
+
+#endif /* _PCRE_STRINGPIECE_H */
diff --git a/pcre/pcre_stringpiece.h.in b/pcre/pcre_stringpiece.h.in
new file mode 100644
index 00000000000..369c10f31b7
--- /dev/null
+++ b/pcre/pcre_stringpiece.h.in
@@ -0,0 +1,179 @@
+// Copyright (c) 2005, Google Inc.
+// All rights reserved.
+//
+// Redistribution and use in source and binary forms, with or without
+// modification, are permitted provided that the following conditions are
+// met:
+//
+//     * Redistributions of source code must retain the above copyright
+// notice, this list of conditions and the following disclaimer.
+//     * Redistributions in binary form must reproduce the above
+// copyright notice, this list of conditions and the following disclaimer
+// in the documentation and/or other materials provided with the
+// distribution.
+//     * Neither the name of Google Inc. nor the names of its
+// contributors may be used to endorse or promote products derived from
+// this software without specific prior written permission.
+//
+// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+//
+// Author: Sanjay Ghemawat
+//
+// A string like object that points into another piece of memory.
+// Useful for providing an interface that allows clients to easily
+// pass in either a "const char*" or a "string".
+//
+// Arghh!  I wish C++ literals were automatically of type "string".
+
+#ifndef _PCRE_STRINGPIECE_H
+#define _PCRE_STRINGPIECE_H
+
+#include 
+#include 
+#include     // for ostream forward-declaration
+
+#if @pcre_have_type_traits@
+#define HAVE_TYPE_TRAITS
+#include 
+#elif @pcre_have_bits_type_traits@
+#define HAVE_TYPE_TRAITS
+#include 
+#endif
+
+#include 
+
+using std::memcmp;
+using std::strlen;
+using std::string;
+
+namespace pcrecpp {
+
+class PCRECPP_EXP_DEFN StringPiece {
+ private:
+  const char*   ptr_;
+  int           length_;
+
+ public:
+  // We provide non-explicit singleton constructors so users can pass
+  // in a "const char*" or a "string" wherever a "StringPiece" is
+  // expected.
+  StringPiece()
+    : ptr_(NULL), length_(0) { }
+  StringPiece(const char* str)
+    : ptr_(str), length_(static_cast(strlen(ptr_))) { }
+  StringPiece(const unsigned char* str)
+    : ptr_(reinterpret_cast(str)),
+      length_(static_cast(strlen(ptr_))) { }
+  StringPiece(const string& str)
+    : ptr_(str.data()), length_(static_cast(str.size())) { }
+  StringPiece(const char* offset, int len)
+    : ptr_(offset), length_(len) { }
+
+  // data() may return a pointer to a buffer with embedded NULs, and the
+  // returned buffer may or may not be null terminated.  Therefore it is
+  // typically a mistake to pass data() to a routine that expects a NUL
+  // terminated string.  Use "as_string().c_str()" if you really need to do
+  // this.  Or better yet, change your routine so it does not rely on NUL
+  // termination.
+  const char* data() const { return ptr_; }
+  int size() const { return length_; }
+  bool empty() const { return length_ == 0; }
+
+  void clear() { ptr_ = NULL; length_ = 0; }
+  void set(const char* buffer, int len) { ptr_ = buffer; length_ = len; }
+  void set(const char* str) {
+    ptr_ = str;
+    length_ = static_cast(strlen(str));
+  }
+  void set(const void* buffer, int len) {
+    ptr_ = reinterpret_cast(buffer);
+    length_ = len;
+  }
+
+  char operator[](int i) const { return ptr_[i]; }
+
+  void remove_prefix(int n) {
+    ptr_ += n;
+    length_ -= n;
+  }
+
+  void remove_suffix(int n) {
+    length_ -= n;
+  }
+
+  bool operator==(const StringPiece& x) const {
+    return ((length_ == x.length_) &&
+            (memcmp(ptr_, x.ptr_, length_) == 0));
+  }
+  bool operator!=(const StringPiece& x) const {
+    return !(*this == x);
+  }
+
+#define STRINGPIECE_BINARY_PREDICATE(cmp,auxcmp)                             \
+  bool operator cmp (const StringPiece& x) const {                           \
+    int r = memcmp(ptr_, x.ptr_, length_ < x.length_ ? length_ : x.length_); \
+    return ((r auxcmp 0) || ((r == 0) && (length_ cmp x.length_)));          \
+  }
+  STRINGPIECE_BINARY_PREDICATE(<,  <);
+  STRINGPIECE_BINARY_PREDICATE(<=, <);
+  STRINGPIECE_BINARY_PREDICATE(>=, >);
+  STRINGPIECE_BINARY_PREDICATE(>,  >);
+#undef STRINGPIECE_BINARY_PREDICATE
+
+  int compare(const StringPiece& x) const {
+    int r = memcmp(ptr_, x.ptr_, length_ < x.length_ ? length_ : x.length_);
+    if (r == 0) {
+      if (length_ < x.length_) r = -1;
+      else if (length_ > x.length_) r = +1;
+    }
+    return r;
+  }
+
+  string as_string() const {
+    return string(data(), size());
+  }
+
+  void CopyToString(string* target) const {
+    target->assign(ptr_, length_);
+  }
+
+  // Does "this" start with "x"
+  bool starts_with(const StringPiece& x) const {
+    return ((length_ >= x.length_) && (memcmp(ptr_, x.ptr_, x.length_) == 0));
+  }
+};
+
+}   // namespace pcrecpp
+
+// ------------------------------------------------------------------
+// Functions used to create STL containers that use StringPiece
+//  Remember that a StringPiece's lifetime had better be less than
+//  that of the underlying string or char*.  If it is not, then you
+//  cannot safely store a StringPiece into an STL container
+// ------------------------------------------------------------------
+
+#ifdef HAVE_TYPE_TRAITS
+// This makes vector really fast for some STL implementations
+template<> struct __type_traits {
+  typedef __true_type    has_trivial_default_constructor;
+  typedef __true_type    has_trivial_copy_constructor;
+  typedef __true_type    has_trivial_assignment_operator;
+  typedef __true_type    has_trivial_destructor;
+  typedef __true_type    is_POD_type;
+};
+#endif
+
+// allow StringPiece to be logged
+std::ostream& operator<<(std::ostream& o, const pcrecpp::StringPiece& piece);
+
+#endif /* _PCRE_STRINGPIECE_H */
diff --git a/pcre/pcre_stringpiece_unittest.cc b/pcre/pcre_stringpiece_unittest.cc
new file mode 100644
index 00000000000..c58e028cefe
--- /dev/null
+++ b/pcre/pcre_stringpiece_unittest.cc
@@ -0,0 +1,150 @@
+// Copyright 2003 and onwards Google Inc.
+// Author: Sanjay Ghemawat
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include 
+#include 
+#include     // for make_pair
+
+#include "pcrecpp.h"
+#include "pcre_stringpiece.h"
+
+// CHECK dies with a fatal error if condition is not true.  It is *not*
+// controlled by NDEBUG, so the check will be executed regardless of
+// compilation mode.  Therefore, it is safe to do things like:
+//    CHECK(fp->Write(x) == 4)
+#define CHECK(condition) do {                           \
+  if (!(condition)) {                                   \
+    fprintf(stderr, "%s:%d: Check failed: %s\n",        \
+            __FILE__, __LINE__, #condition);            \
+    exit(1);                                            \
+  }                                                     \
+} while (0)
+
+using pcrecpp::StringPiece;
+
+static void CheckSTLComparator() {
+  string s1("foo");
+  string s2("bar");
+  string s3("baz");
+
+  StringPiece p1(s1);
+  StringPiece p2(s2);
+  StringPiece p3(s3);
+
+  typedef std::map TestMap;
+  TestMap map;
+
+  map.insert(std::make_pair(p1, 0));
+  map.insert(std::make_pair(p2, 1));
+  map.insert(std::make_pair(p3, 2));
+
+  CHECK(map.size() == 3);
+
+  TestMap::const_iterator iter = map.begin();
+  CHECK(iter->second == 1);
+  ++iter;
+  CHECK(iter->second == 2);
+  ++iter;
+  CHECK(iter->second == 0);
+  ++iter;
+  CHECK(iter == map.end());
+
+  TestMap::iterator new_iter = map.find("zot");
+  CHECK(new_iter == map.end());
+
+  new_iter = map.find("bar");
+  CHECK(new_iter != map.end());
+
+  map.erase(new_iter);
+  CHECK(map.size() == 2);
+
+  iter = map.begin();
+  CHECK(iter->second == 2);
+  ++iter;
+  CHECK(iter->second == 0);
+  ++iter;
+  CHECK(iter == map.end());
+}
+
+static void CheckComparisonOperators() {
+#define CMP_Y(op, x, y)                                         \
+  CHECK( (StringPiece((x)) op StringPiece((y))));               \
+  CHECK( (StringPiece((x)).compare(StringPiece((y))) op 0))
+
+#define CMP_N(op, x, y)                                         \
+  CHECK(!(StringPiece((x)) op StringPiece((y))));               \
+  CHECK(!(StringPiece((x)).compare(StringPiece((y))) op 0))
+
+  CMP_Y(==, "",   "");
+  CMP_Y(==, "a",  "a");
+  CMP_Y(==, "aa", "aa");
+  CMP_N(==, "a",  "");
+  CMP_N(==, "",   "a");
+  CMP_N(==, "a",  "b");
+  CMP_N(==, "a",  "aa");
+  CMP_N(==, "aa", "a");
+
+  CMP_N(!=, "",   "");
+  CMP_N(!=, "a",  "a");
+  CMP_N(!=, "aa", "aa");
+  CMP_Y(!=, "a",  "");
+  CMP_Y(!=, "",   "a");
+  CMP_Y(!=, "a",  "b");
+  CMP_Y(!=, "a",  "aa");
+  CMP_Y(!=, "aa", "a");
+
+  CMP_Y(<, "a",  "b");
+  CMP_Y(<, "a",  "aa");
+  CMP_Y(<, "aa", "b");
+  CMP_Y(<, "aa", "bb");
+  CMP_N(<, "a",  "a");
+  CMP_N(<, "b",  "a");
+  CMP_N(<, "aa", "a");
+  CMP_N(<, "b",  "aa");
+  CMP_N(<, "bb", "aa");
+
+  CMP_Y(<=, "a",  "a");
+  CMP_Y(<=, "a",  "b");
+  CMP_Y(<=, "a",  "aa");
+  CMP_Y(<=, "aa", "b");
+  CMP_Y(<=, "aa", "bb");
+  CMP_N(<=, "b",  "a");
+  CMP_N(<=, "aa", "a");
+  CMP_N(<=, "b",  "aa");
+  CMP_N(<=, "bb", "aa");
+
+  CMP_N(>=, "a",  "b");
+  CMP_N(>=, "a",  "aa");
+  CMP_N(>=, "aa", "b");
+  CMP_N(>=, "aa", "bb");
+  CMP_Y(>=, "a",  "a");
+  CMP_Y(>=, "b",  "a");
+  CMP_Y(>=, "aa", "a");
+  CMP_Y(>=, "b",  "aa");
+  CMP_Y(>=, "bb", "aa");
+
+  CMP_N(>, "a",  "a");
+  CMP_N(>, "a",  "b");
+  CMP_N(>, "a",  "aa");
+  CMP_N(>, "aa", "b");
+  CMP_N(>, "aa", "bb");
+  CMP_Y(>, "b",  "a");
+  CMP_Y(>, "aa", "a");
+  CMP_Y(>, "b",  "aa");
+  CMP_Y(>, "bb", "aa");
+
+#undef CMP_Y
+#undef CMP_N
+}
+
+int main(int argc, char** argv) {
+  CheckComparisonOperators();
+  CheckSTLComparator();
+
+  printf("OK\n");
+  return 0;
+}
diff --git a/pcre/pcre_study.c b/pcre/pcre_study.c
new file mode 100644
index 00000000000..12d2a66817b
--- /dev/null
+++ b/pcre/pcre_study.c
@@ -0,0 +1,1562 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+
+/* This module contains the external function pcre_study(), along with local
+supporting functions. */
+
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include "pcre_internal.h"
+
+#define SET_BIT(c) start_bits[c/8] |= (1 << (c&7))
+
+/* Returns from set_start_bits() */
+
+enum { SSB_FAIL, SSB_DONE, SSB_CONTINUE, SSB_UNKNOWN };
+
+
+
+/*************************************************
+*   Find the minimum subject length for a group  *
+*************************************************/
+
+/* Scan a parenthesized group and compute the minimum length of subject that
+is needed to match it. This is a lower bound; it does not mean there is a
+string of that length that matches. In UTF8 mode, the result is in characters
+rather than bytes.
+
+Arguments:
+  code            pointer to start of group (the bracket)
+  startcode       pointer to start of the whole pattern
+  options         the compiling options
+  int             RECURSE depth
+
+Returns:   the minimum length
+           -1 if \C in UTF-8 mode or (*ACCEPT) was encountered
+           -2 internal error (missing capturing bracket)
+           -3 internal error (opcode not listed)
+*/
+
+static int
+find_minlength(const pcre_uchar *code, const pcre_uchar *startcode, int options,
+  int recurse_depth)
+{
+int length = -1;
+/* PCRE_UTF16 has the same value as PCRE_UTF8. */
+BOOL utf = (options & PCRE_UTF8) != 0;
+BOOL had_recurse = FALSE;
+register int branchlength = 0;
+register pcre_uchar *cc = (pcre_uchar *)code + 1 + LINK_SIZE;
+
+if (*code == OP_CBRA || *code == OP_SCBRA ||
+    *code == OP_CBRAPOS || *code == OP_SCBRAPOS) cc += IMM2_SIZE;
+
+/* Scan along the opcodes for this branch. If we get to the end of the
+branch, check the length against that of the other branches. */
+
+for (;;)
+  {
+  int d, min;
+  pcre_uchar *cs, *ce;
+  register pcre_uchar op = *cc;
+
+  switch (op)
+    {
+    case OP_COND:
+    case OP_SCOND:
+
+    /* If there is only one branch in a condition, the implied branch has zero
+    length, so we don't add anything. This covers the DEFINE "condition"
+    automatically. */
+
+    cs = cc + GET(cc, 1);
+    if (*cs != OP_ALT)
+      {
+      cc = cs + 1 + LINK_SIZE;
+      break;
+      }
+
+    /* Otherwise we can fall through and treat it the same as any other
+    subpattern. */
+
+    case OP_CBRA:
+    case OP_SCBRA:
+    case OP_BRA:
+    case OP_SBRA:
+    case OP_CBRAPOS:
+    case OP_SCBRAPOS:
+    case OP_BRAPOS:
+    case OP_SBRAPOS:
+    case OP_ONCE:
+    case OP_ONCE_NC:
+    d = find_minlength(cc, startcode, options, recurse_depth);
+    if (d < 0) return d;
+    branchlength += d;
+    do cc += GET(cc, 1); while (*cc == OP_ALT);
+    cc += 1 + LINK_SIZE;
+    break;
+
+    /* ACCEPT makes things far too complicated; we have to give up. */
+
+    case OP_ACCEPT:
+    case OP_ASSERT_ACCEPT:
+    return -1;
+
+    /* Reached end of a branch; if it's a ket it is the end of a nested
+    call. If it's ALT it is an alternation in a nested call. If it is END it's
+    the end of the outer call. All can be handled by the same code. If an
+    ACCEPT was previously encountered, use the length that was in force at that
+    time, and pass back the shortest ACCEPT length. */
+
+    case OP_ALT:
+    case OP_KET:
+    case OP_KETRMAX:
+    case OP_KETRMIN:
+    case OP_KETRPOS:
+    case OP_END:
+    if (length < 0 || (!had_recurse && branchlength < length))
+      length = branchlength;
+    if (op != OP_ALT) return length;
+    cc += 1 + LINK_SIZE;
+    branchlength = 0;
+    had_recurse = FALSE;
+    break;
+
+    /* Skip over assertive subpatterns */
+
+    case OP_ASSERT:
+    case OP_ASSERT_NOT:
+    case OP_ASSERTBACK:
+    case OP_ASSERTBACK_NOT:
+    do cc += GET(cc, 1); while (*cc == OP_ALT);
+    /* Fall through */
+
+    /* Skip over things that don't match chars */
+
+    case OP_REVERSE:
+    case OP_CREF:
+    case OP_NCREF:
+    case OP_RREF:
+    case OP_NRREF:
+    case OP_DEF:
+    case OP_CALLOUT:
+    case OP_SOD:
+    case OP_SOM:
+    case OP_EOD:
+    case OP_EODN:
+    case OP_CIRC:
+    case OP_CIRCM:
+    case OP_DOLL:
+    case OP_DOLLM:
+    case OP_NOT_WORD_BOUNDARY:
+    case OP_WORD_BOUNDARY:
+    cc += PRIV(OP_lengths)[*cc];
+    break;
+
+    /* Skip over a subpattern that has a {0} or {0,x} quantifier */
+
+    case OP_BRAZERO:
+    case OP_BRAMINZERO:
+    case OP_BRAPOSZERO:
+    case OP_SKIPZERO:
+    cc += PRIV(OP_lengths)[*cc];
+    do cc += GET(cc, 1); while (*cc == OP_ALT);
+    cc += 1 + LINK_SIZE;
+    break;
+
+    /* Handle literal characters and + repetitions */
+
+    case OP_CHAR:
+    case OP_CHARI:
+    case OP_NOT:
+    case OP_NOTI:
+    case OP_PLUS:
+    case OP_PLUSI:
+    case OP_MINPLUS:
+    case OP_MINPLUSI:
+    case OP_POSPLUS:
+    case OP_POSPLUSI:
+    case OP_NOTPLUS:
+    case OP_NOTPLUSI:
+    case OP_NOTMINPLUS:
+    case OP_NOTMINPLUSI:
+    case OP_NOTPOSPLUS:
+    case OP_NOTPOSPLUSI:
+    branchlength++;
+    cc += 2;
+#ifdef SUPPORT_UTF
+    if (utf && HAS_EXTRALEN(cc[-1])) cc += GET_EXTRALEN(cc[-1]);
+#endif
+    break;
+
+    case OP_TYPEPLUS:
+    case OP_TYPEMINPLUS:
+    case OP_TYPEPOSPLUS:
+    branchlength++;
+    cc += (cc[1] == OP_PROP || cc[1] == OP_NOTPROP)? 4 : 2;
+    break;
+
+    /* Handle exact repetitions. The count is already in characters, but we
+    need to skip over a multibyte character in UTF8 mode.  */
+
+    case OP_EXACT:
+    case OP_EXACTI:
+    case OP_NOTEXACT:
+    case OP_NOTEXACTI:
+    branchlength += GET2(cc,1);
+    cc += 2 + IMM2_SIZE;
+#ifdef SUPPORT_UTF
+    if (utf && HAS_EXTRALEN(cc[-1])) cc += GET_EXTRALEN(cc[-1]);
+#endif
+    break;
+
+    case OP_TYPEEXACT:
+    branchlength += GET2(cc,1);
+    cc += 2 + IMM2_SIZE + ((cc[1 + IMM2_SIZE] == OP_PROP
+      || cc[1 + IMM2_SIZE] == OP_NOTPROP)? 2 : 0);
+    break;
+
+    /* Handle single-char non-literal matchers */
+
+    case OP_PROP:
+    case OP_NOTPROP:
+    cc += 2;
+    /* Fall through */
+
+    case OP_NOT_DIGIT:
+    case OP_DIGIT:
+    case OP_NOT_WHITESPACE:
+    case OP_WHITESPACE:
+    case OP_NOT_WORDCHAR:
+    case OP_WORDCHAR:
+    case OP_ANY:
+    case OP_ALLANY:
+    case OP_EXTUNI:
+    case OP_HSPACE:
+    case OP_NOT_HSPACE:
+    case OP_VSPACE:
+    case OP_NOT_VSPACE:
+    branchlength++;
+    cc++;
+    break;
+
+    /* "Any newline" might match two characters, but it also might match just
+    one. */
+
+    case OP_ANYNL:
+    branchlength += 1;
+    cc++;
+    break;
+
+    /* The single-byte matcher means we can't proceed in UTF-8 mode. (In
+    non-UTF-8 mode \C will actually be turned into OP_ALLANY, so won't ever
+    appear, but leave the code, just in case.) */
+
+    case OP_ANYBYTE:
+#ifdef SUPPORT_UTF
+    if (utf) return -1;
+#endif
+    branchlength++;
+    cc++;
+    break;
+
+    /* For repeated character types, we have to test for \p and \P, which have
+    an extra two bytes of parameters. */
+
+    case OP_TYPESTAR:
+    case OP_TYPEMINSTAR:
+    case OP_TYPEQUERY:
+    case OP_TYPEMINQUERY:
+    case OP_TYPEPOSSTAR:
+    case OP_TYPEPOSQUERY:
+    if (cc[1] == OP_PROP || cc[1] == OP_NOTPROP) cc += 2;
+    cc += PRIV(OP_lengths)[op];
+    break;
+
+    case OP_TYPEUPTO:
+    case OP_TYPEMINUPTO:
+    case OP_TYPEPOSUPTO:
+    if (cc[1 + IMM2_SIZE] == OP_PROP
+      || cc[1 + IMM2_SIZE] == OP_NOTPROP) cc += 2;
+    cc += PRIV(OP_lengths)[op];
+    break;
+
+    /* Check a class for variable quantification */
+
+    case OP_CLASS:
+    case OP_NCLASS:
+#if defined SUPPORT_UTF || defined COMPILE_PCRE16 || defined COMPILE_PCRE32
+    case OP_XCLASS:
+    /* The original code caused an unsigned overflow in 64 bit systems,
+    so now we use a conditional statement. */
+    if (op == OP_XCLASS)
+      cc += GET(cc, 1);
+    else
+      cc += PRIV(OP_lengths)[OP_CLASS];
+#else
+    cc += PRIV(OP_lengths)[OP_CLASS];
+#endif
+
+    switch (*cc)
+      {
+      case OP_CRPLUS:
+      case OP_CRMINPLUS:
+      branchlength++;
+      /* Fall through */
+
+      case OP_CRSTAR:
+      case OP_CRMINSTAR:
+      case OP_CRQUERY:
+      case OP_CRMINQUERY:
+      cc++;
+      break;
+
+      case OP_CRRANGE:
+      case OP_CRMINRANGE:
+      branchlength += GET2(cc,1);
+      cc += 1 + 2 * IMM2_SIZE;
+      break;
+
+      default:
+      branchlength++;
+      break;
+      }
+    break;
+
+    /* Backreferences and subroutine calls are treated in the same way: we find
+    the minimum length for the subpattern. A recursion, however, causes an
+    a flag to be set that causes the length of this branch to be ignored. The
+    logic is that a recursion can only make sense if there is another
+    alternation that stops the recursing. That will provide the minimum length
+    (when no recursion happens). A backreference within the group that it is
+    referencing behaves in the same way.
+
+    If PCRE_JAVASCRIPT_COMPAT is set, a backreference to an unset bracket
+    matches an empty string (by default it causes a matching failure), so in
+    that case we must set the minimum length to zero. */
+
+    case OP_REF:
+    case OP_REFI:
+    if ((options & PCRE_JAVASCRIPT_COMPAT) == 0)
+      {
+      ce = cs = (pcre_uchar *)PRIV(find_bracket)(startcode, utf, GET2(cc, 1));
+      if (cs == NULL) return -2;
+      do ce += GET(ce, 1); while (*ce == OP_ALT);
+      if (cc > cs && cc < ce)
+        {
+        d = 0;
+        had_recurse = TRUE;
+        }
+      else
+        {
+        d = find_minlength(cs, startcode, options, recurse_depth);
+        }
+      }
+    else d = 0;
+    cc += 1 + IMM2_SIZE;
+
+    /* Handle repeated back references */
+
+    switch (*cc)
+      {
+      case OP_CRSTAR:
+      case OP_CRMINSTAR:
+      case OP_CRQUERY:
+      case OP_CRMINQUERY:
+      min = 0;
+      cc++;
+      break;
+
+      case OP_CRPLUS:
+      case OP_CRMINPLUS:
+      min = 1;
+      cc++;
+      break;
+
+      case OP_CRRANGE:
+      case OP_CRMINRANGE:
+      min = GET2(cc, 1);
+      cc += 1 + 2 * IMM2_SIZE;
+      break;
+
+      default:
+      min = 1;
+      break;
+      }
+
+    branchlength += min * d;
+    break;
+
+    /* We can easily detect direct recursion, but not mutual recursion. This is
+    caught by a recursion depth count. */
+
+    case OP_RECURSE:
+    cs = ce = (pcre_uchar *)startcode + GET(cc, 1);
+    do ce += GET(ce, 1); while (*ce == OP_ALT);
+    if ((cc > cs && cc < ce) || recurse_depth > 10)
+      had_recurse = TRUE;
+    else
+      {
+      branchlength += find_minlength(cs, startcode, options, recurse_depth + 1);
+      }
+    cc += 1 + LINK_SIZE;
+    break;
+
+    /* Anything else does not or need not match a character. We can get the
+    item's length from the table, but for those that can match zero occurrences
+    of a character, we must take special action for UTF-8 characters. As it
+    happens, the "NOT" versions of these opcodes are used at present only for
+    ASCII characters, so they could be omitted from this list. However, in
+    future that may change, so we include them here so as not to leave a
+    gotcha for a future maintainer. */
+
+    case OP_UPTO:
+    case OP_UPTOI:
+    case OP_NOTUPTO:
+    case OP_NOTUPTOI:
+    case OP_MINUPTO:
+    case OP_MINUPTOI:
+    case OP_NOTMINUPTO:
+    case OP_NOTMINUPTOI:
+    case OP_POSUPTO:
+    case OP_POSUPTOI:
+    case OP_NOTPOSUPTO:
+    case OP_NOTPOSUPTOI:
+
+    case OP_STAR:
+    case OP_STARI:
+    case OP_NOTSTAR:
+    case OP_NOTSTARI:
+    case OP_MINSTAR:
+    case OP_MINSTARI:
+    case OP_NOTMINSTAR:
+    case OP_NOTMINSTARI:
+    case OP_POSSTAR:
+    case OP_POSSTARI:
+    case OP_NOTPOSSTAR:
+    case OP_NOTPOSSTARI:
+
+    case OP_QUERY:
+    case OP_QUERYI:
+    case OP_NOTQUERY:
+    case OP_NOTQUERYI:
+    case OP_MINQUERY:
+    case OP_MINQUERYI:
+    case OP_NOTMINQUERY:
+    case OP_NOTMINQUERYI:
+    case OP_POSQUERY:
+    case OP_POSQUERYI:
+    case OP_NOTPOSQUERY:
+    case OP_NOTPOSQUERYI:
+
+    cc += PRIV(OP_lengths)[op];
+#ifdef SUPPORT_UTF
+    if (utf && HAS_EXTRALEN(cc[-1])) cc += GET_EXTRALEN(cc[-1]);
+#endif
+    break;
+
+    /* Skip these, but we need to add in the name length. */
+
+    case OP_MARK:
+    case OP_PRUNE_ARG:
+    case OP_SKIP_ARG:
+    case OP_THEN_ARG:
+    cc += PRIV(OP_lengths)[op] + cc[1];
+    break;
+
+    /* The remaining opcodes are just skipped over. */
+
+    case OP_CLOSE:
+    case OP_COMMIT:
+    case OP_FAIL:
+    case OP_PRUNE:
+    case OP_SET_SOM:
+    case OP_SKIP:
+    case OP_THEN:
+    cc += PRIV(OP_lengths)[op];
+    break;
+
+    /* This should not occur: we list all opcodes explicitly so that when
+    new ones get added they are properly considered. */
+
+    default:
+    return -3;
+    }
+  }
+/* Control never gets here */
+}
+
+
+
+/*************************************************
+*      Set a bit and maybe its alternate case    *
+*************************************************/
+
+/* Given a character, set its first byte's bit in the table, and also the
+corresponding bit for the other version of a letter if we are caseless. In
+UTF-8 mode, for characters greater than 127, we can only do the caseless thing
+when Unicode property support is available.
+
+Arguments:
+  start_bits    points to the bit map
+  p             points to the character
+  caseless      the caseless flag
+  cd            the block with char table pointers
+  utf           TRUE for UTF-8 / UTF-16 / UTF-32 mode
+
+Returns:        pointer after the character
+*/
+
+static const pcre_uchar *
+set_table_bit(pcre_uint8 *start_bits, const pcre_uchar *p, BOOL caseless,
+  compile_data *cd, BOOL utf)
+{
+pcre_uint32 c = *p;
+
+#ifdef COMPILE_PCRE8
+SET_BIT(c);
+
+#ifdef SUPPORT_UTF
+if (utf && c > 127)
+  {
+  GETCHARINC(c, p);
+#ifdef SUPPORT_UCP
+  if (caseless)
+    {
+    pcre_uchar buff[6];
+    c = UCD_OTHERCASE(c);
+    (void)PRIV(ord2utf)(c, buff);
+    SET_BIT(buff[0]);
+    }
+#endif  /* Not SUPPORT_UCP */
+  return p;
+  }
+#else   /* Not SUPPORT_UTF */
+(void)(utf);   /* Stops warning for unused parameter */
+#endif  /* SUPPORT_UTF */
+
+/* Not UTF-8 mode, or character is less than 127. */
+
+if (caseless && (cd->ctypes[c] & ctype_letter) != 0) SET_BIT(cd->fcc[c]);
+return p + 1;
+#endif  /* COMPILE_PCRE8 */
+
+#if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
+if (c > 0xff)
+  {
+  c = 0xff;
+  caseless = FALSE;
+  }
+SET_BIT(c);
+
+#ifdef SUPPORT_UTF
+if (utf && c > 127)
+  {
+  GETCHARINC(c, p);
+#ifdef SUPPORT_UCP
+  if (caseless)
+    {
+    c = UCD_OTHERCASE(c);
+    if (c > 0xff)
+      c = 0xff;
+    SET_BIT(c);
+    }
+#endif  /* SUPPORT_UCP */
+  return p;
+  }
+#else   /* Not SUPPORT_UTF */
+(void)(utf);   /* Stops warning for unused parameter */
+#endif  /* SUPPORT_UTF */
+
+if (caseless && (cd->ctypes[c] & ctype_letter) != 0) SET_BIT(cd->fcc[c]);
+return p + 1;
+#endif
+}
+
+
+
+/*************************************************
+*     Set bits for a positive character type     *
+*************************************************/
+
+/* This function sets starting bits for a character type. In UTF-8 mode, we can
+only do a direct setting for bytes less than 128, as otherwise there can be
+confusion with bytes in the middle of UTF-8 characters. In a "traditional"
+environment, the tables will only recognize ASCII characters anyway, but in at
+least one Windows environment, some higher bytes bits were set in the tables.
+So we deal with that case by considering the UTF-8 encoding.
+
+Arguments:
+  start_bits     the starting bitmap
+  cbit type      the type of character wanted
+  table_limit    32 for non-UTF-8; 16 for UTF-8
+  cd             the block with char table pointers
+
+Returns:         nothing
+*/
+
+static void
+set_type_bits(pcre_uint8 *start_bits, int cbit_type, unsigned int table_limit,
+  compile_data *cd)
+{
+register pcre_uint32 c;
+for (c = 0; c < table_limit; c++) start_bits[c] |= cd->cbits[c+cbit_type];
+#if defined SUPPORT_UTF && defined COMPILE_PCRE8
+if (table_limit == 32) return;
+for (c = 128; c < 256; c++)
+  {
+  if ((cd->cbits[c/8] & (1 << (c&7))) != 0)
+    {
+    pcre_uchar buff[6];
+    (void)PRIV(ord2utf)(c, buff);
+    SET_BIT(buff[0]);
+    }
+  }
+#endif
+}
+
+
+/*************************************************
+*     Set bits for a negative character type     *
+*************************************************/
+
+/* This function sets starting bits for a negative character type such as \D.
+In UTF-8 mode, we can only do a direct setting for bytes less than 128, as
+otherwise there can be confusion with bytes in the middle of UTF-8 characters.
+Unlike in the positive case, where we can set appropriate starting bits for
+specific high-valued UTF-8 characters, in this case we have to set the bits for
+all high-valued characters. The lowest is 0xc2, but we overkill by starting at
+0xc0 (192) for simplicity.
+
+Arguments:
+  start_bits     the starting bitmap
+  cbit type      the type of character wanted
+  table_limit    32 for non-UTF-8; 16 for UTF-8
+  cd             the block with char table pointers
+
+Returns:         nothing
+*/
+
+static void
+set_nottype_bits(pcre_uint8 *start_bits, int cbit_type, unsigned int table_limit,
+  compile_data *cd)
+{
+register pcre_uint32 c;
+for (c = 0; c < table_limit; c++) start_bits[c] |= ~cd->cbits[c+cbit_type];
+#if defined SUPPORT_UTF && defined COMPILE_PCRE8
+if (table_limit != 32) for (c = 24; c < 32; c++) start_bits[c] = 0xff;
+#endif
+}
+
+
+
+/*************************************************
+*          Create bitmap of starting bytes       *
+*************************************************/
+
+/* This function scans a compiled unanchored expression recursively and
+attempts to build a bitmap of the set of possible starting bytes. As time goes
+by, we may be able to get more clever at doing this. The SSB_CONTINUE return is
+useful for parenthesized groups in patterns such as (a*)b where the group
+provides some optional starting bytes but scanning must continue at the outer
+level to find at least one mandatory byte. At the outermost level, this
+function fails unless the result is SSB_DONE.
+
+Arguments:
+  code         points to an expression
+  start_bits   points to a 32-byte table, initialized to 0
+  utf          TRUE if in UTF-8 / UTF-16 / UTF-32 mode
+  cd           the block with char table pointers
+
+Returns:       SSB_FAIL     => Failed to find any starting bytes
+               SSB_DONE     => Found mandatory starting bytes
+               SSB_CONTINUE => Found optional starting bytes
+               SSB_UNKNOWN  => Hit an unrecognized opcode
+*/
+
+static int
+set_start_bits(const pcre_uchar *code, pcre_uint8 *start_bits, BOOL utf,
+  compile_data *cd)
+{
+register pcre_uint32 c;
+int yield = SSB_DONE;
+#if defined SUPPORT_UTF && defined COMPILE_PCRE8
+int table_limit = utf? 16:32;
+#else
+int table_limit = 32;
+#endif
+
+#if 0
+/* ========================================================================= */
+/* The following comment and code was inserted in January 1999. In May 2006,
+when it was observed to cause compiler warnings about unused values, I took it
+out again. If anybody is still using OS/2, they will have to put it back
+manually. */
+
+/* This next statement and the later reference to dummy are here in order to
+trick the optimizer of the IBM C compiler for OS/2 into generating correct
+code. Apparently IBM isn't going to fix the problem, and we would rather not
+disable optimization (in this module it actually makes a big difference, and
+the pcre module can use all the optimization it can get). */
+
+volatile int dummy;
+/* ========================================================================= */
+#endif
+
+do
+  {
+  BOOL try_next = TRUE;
+  const pcre_uchar *tcode = code + 1 + LINK_SIZE;
+
+  if (*code == OP_CBRA || *code == OP_SCBRA ||
+      *code == OP_CBRAPOS || *code == OP_SCBRAPOS) tcode += IMM2_SIZE;
+
+  while (try_next)    /* Loop for items in this branch */
+    {
+    int rc;
+
+    switch(*tcode)
+      {
+      /* If we reach something we don't understand, it means a new opcode has
+      been created that hasn't been added to this code. Hopefully this problem
+      will be discovered during testing. */
+
+      default:
+      return SSB_UNKNOWN;
+
+      /* Fail for a valid opcode that implies no starting bits. */
+
+      case OP_ACCEPT:
+      case OP_ASSERT_ACCEPT:
+      case OP_ALLANY:
+      case OP_ANY:
+      case OP_ANYBYTE:
+      case OP_CIRC:
+      case OP_CIRCM:
+      case OP_CLOSE:
+      case OP_COMMIT:
+      case OP_COND:
+      case OP_CREF:
+      case OP_DEF:
+      case OP_DOLL:
+      case OP_DOLLM:
+      case OP_END:
+      case OP_EOD:
+      case OP_EODN:
+      case OP_EXTUNI:
+      case OP_FAIL:
+      case OP_MARK:
+      case OP_NCREF:
+      case OP_NOT:
+      case OP_NOTEXACT:
+      case OP_NOTEXACTI:
+      case OP_NOTI:
+      case OP_NOTMINPLUS:
+      case OP_NOTMINPLUSI:
+      case OP_NOTMINQUERY:
+      case OP_NOTMINQUERYI:
+      case OP_NOTMINSTAR:
+      case OP_NOTMINSTARI:
+      case OP_NOTMINUPTO:
+      case OP_NOTMINUPTOI:
+      case OP_NOTPLUS:
+      case OP_NOTPLUSI:
+      case OP_NOTPOSPLUS:
+      case OP_NOTPOSPLUSI:
+      case OP_NOTPOSQUERY:
+      case OP_NOTPOSQUERYI:
+      case OP_NOTPOSSTAR:
+      case OP_NOTPOSSTARI:
+      case OP_NOTPOSUPTO:
+      case OP_NOTPOSUPTOI:
+      case OP_NOTPROP:
+      case OP_NOTQUERY:
+      case OP_NOTQUERYI:
+      case OP_NOTSTAR:
+      case OP_NOTSTARI:
+      case OP_NOTUPTO:
+      case OP_NOTUPTOI:
+      case OP_NOT_HSPACE:
+      case OP_NOT_VSPACE:
+      case OP_NRREF:
+      case OP_PROP:
+      case OP_PRUNE:
+      case OP_PRUNE_ARG:
+      case OP_RECURSE:
+      case OP_REF:
+      case OP_REFI:
+      case OP_REVERSE:
+      case OP_RREF:
+      case OP_SCOND:
+      case OP_SET_SOM:
+      case OP_SKIP:
+      case OP_SKIP_ARG:
+      case OP_SOD:
+      case OP_SOM:
+      case OP_THEN:
+      case OP_THEN_ARG:
+#if defined SUPPORT_UTF || !defined COMPILE_PCRE8
+      case OP_XCLASS:
+#endif
+      return SSB_FAIL;
+
+      /* We can ignore word boundary tests. */
+
+      case OP_WORD_BOUNDARY:
+      case OP_NOT_WORD_BOUNDARY:
+      tcode++;
+      break;
+
+      /* If we hit a bracket or a positive lookahead assertion, recurse to set
+      bits from within the subpattern. If it can't find anything, we have to
+      give up. If it finds some mandatory character(s), we are done for this
+      branch. Otherwise, carry on scanning after the subpattern. */
+
+      case OP_BRA:
+      case OP_SBRA:
+      case OP_CBRA:
+      case OP_SCBRA:
+      case OP_BRAPOS:
+      case OP_SBRAPOS:
+      case OP_CBRAPOS:
+      case OP_SCBRAPOS:
+      case OP_ONCE:
+      case OP_ONCE_NC:
+      case OP_ASSERT:
+      rc = set_start_bits(tcode, start_bits, utf, cd);
+      if (rc == SSB_FAIL || rc == SSB_UNKNOWN) return rc;
+      if (rc == SSB_DONE) try_next = FALSE; else
+        {
+        do tcode += GET(tcode, 1); while (*tcode == OP_ALT);
+        tcode += 1 + LINK_SIZE;
+        }
+      break;
+
+      /* If we hit ALT or KET, it means we haven't found anything mandatory in
+      this branch, though we might have found something optional. For ALT, we
+      continue with the next alternative, but we have to arrange that the final
+      result from subpattern is SSB_CONTINUE rather than SSB_DONE. For KET,
+      return SSB_CONTINUE: if this is the top level, that indicates failure,
+      but after a nested subpattern, it causes scanning to continue. */
+
+      case OP_ALT:
+      yield = SSB_CONTINUE;
+      try_next = FALSE;
+      break;
+
+      case OP_KET:
+      case OP_KETRMAX:
+      case OP_KETRMIN:
+      case OP_KETRPOS:
+      return SSB_CONTINUE;
+
+      /* Skip over callout */
+
+      case OP_CALLOUT:
+      tcode += 2 + 2*LINK_SIZE;
+      break;
+
+      /* Skip over lookbehind and negative lookahead assertions */
+
+      case OP_ASSERT_NOT:
+      case OP_ASSERTBACK:
+      case OP_ASSERTBACK_NOT:
+      do tcode += GET(tcode, 1); while (*tcode == OP_ALT);
+      tcode += 1 + LINK_SIZE;
+      break;
+
+      /* BRAZERO does the bracket, but carries on. */
+
+      case OP_BRAZERO:
+      case OP_BRAMINZERO:
+      case OP_BRAPOSZERO:
+      rc = set_start_bits(++tcode, start_bits, utf, cd);
+      if (rc == SSB_FAIL || rc == SSB_UNKNOWN) return rc;
+/* =========================================================================
+      See the comment at the head of this function concerning the next line,
+      which was an old fudge for the benefit of OS/2.
+      dummy = 1;
+  ========================================================================= */
+      do tcode += GET(tcode,1); while (*tcode == OP_ALT);
+      tcode += 1 + LINK_SIZE;
+      break;
+
+      /* SKIPZERO skips the bracket. */
+
+      case OP_SKIPZERO:
+      tcode++;
+      do tcode += GET(tcode,1); while (*tcode == OP_ALT);
+      tcode += 1 + LINK_SIZE;
+      break;
+
+      /* Single-char * or ? sets the bit and tries the next item */
+
+      case OP_STAR:
+      case OP_MINSTAR:
+      case OP_POSSTAR:
+      case OP_QUERY:
+      case OP_MINQUERY:
+      case OP_POSQUERY:
+      tcode = set_table_bit(start_bits, tcode + 1, FALSE, cd, utf);
+      break;
+
+      case OP_STARI:
+      case OP_MINSTARI:
+      case OP_POSSTARI:
+      case OP_QUERYI:
+      case OP_MINQUERYI:
+      case OP_POSQUERYI:
+      tcode = set_table_bit(start_bits, tcode + 1, TRUE, cd, utf);
+      break;
+
+      /* Single-char upto sets the bit and tries the next */
+
+      case OP_UPTO:
+      case OP_MINUPTO:
+      case OP_POSUPTO:
+      tcode = set_table_bit(start_bits, tcode + 1 + IMM2_SIZE, FALSE, cd, utf);
+      break;
+
+      case OP_UPTOI:
+      case OP_MINUPTOI:
+      case OP_POSUPTOI:
+      tcode = set_table_bit(start_bits, tcode + 1 + IMM2_SIZE, TRUE, cd, utf);
+      break;
+
+      /* At least one single char sets the bit and stops */
+
+      case OP_EXACT:
+      tcode += IMM2_SIZE;
+      /* Fall through */
+      case OP_CHAR:
+      case OP_PLUS:
+      case OP_MINPLUS:
+      case OP_POSPLUS:
+      (void)set_table_bit(start_bits, tcode + 1, FALSE, cd, utf);
+      try_next = FALSE;
+      break;
+
+      case OP_EXACTI:
+      tcode += IMM2_SIZE;
+      /* Fall through */
+      case OP_CHARI:
+      case OP_PLUSI:
+      case OP_MINPLUSI:
+      case OP_POSPLUSI:
+      (void)set_table_bit(start_bits, tcode + 1, TRUE, cd, utf);
+      try_next = FALSE;
+      break;
+
+      /* Special spacing and line-terminating items. These recognize specific
+      lists of characters. The difference between VSPACE and ANYNL is that the
+      latter can match the two-character CRLF sequence, but that is not
+      relevant for finding the first character, so their code here is
+      identical. */
+
+      case OP_HSPACE:
+      SET_BIT(CHAR_HT);
+      SET_BIT(CHAR_SPACE);
+#ifdef SUPPORT_UTF
+      if (utf)
+        {
+#ifdef COMPILE_PCRE8
+        SET_BIT(0xC2);  /* For U+00A0 */
+        SET_BIT(0xE1);  /* For U+1680, U+180E */
+        SET_BIT(0xE2);  /* For U+2000 - U+200A, U+202F, U+205F */
+        SET_BIT(0xE3);  /* For U+3000 */
+#elif defined COMPILE_PCRE16 || defined COMPILE_PCRE32
+        SET_BIT(0xA0);
+        SET_BIT(0xFF);  /* For characters > 255 */
+#endif  /* COMPILE_PCRE[8|16|32] */
+        }
+      else
+#endif /* SUPPORT_UTF */
+        {
+#ifndef EBCDIC
+        SET_BIT(0xA0);
+#endif  /* Not EBCDIC */
+#if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
+        SET_BIT(0xFF);  /* For characters > 255 */
+#endif  /* COMPILE_PCRE[16|32] */
+        }
+      try_next = FALSE;
+      break;
+
+      case OP_ANYNL:
+      case OP_VSPACE:
+      SET_BIT(CHAR_LF);
+      SET_BIT(CHAR_VT);
+      SET_BIT(CHAR_FF);
+      SET_BIT(CHAR_CR);
+#ifdef SUPPORT_UTF
+      if (utf)
+        {
+#ifdef COMPILE_PCRE8
+        SET_BIT(0xC2);  /* For U+0085 */
+        SET_BIT(0xE2);  /* For U+2028, U+2029 */
+#elif defined COMPILE_PCRE16 || defined COMPILE_PCRE32
+        SET_BIT(CHAR_NEL);
+        SET_BIT(0xFF);  /* For characters > 255 */
+#endif  /* COMPILE_PCRE[8|16|32] */
+        }
+      else
+#endif /* SUPPORT_UTF */
+        {
+        SET_BIT(CHAR_NEL);
+#if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
+        SET_BIT(0xFF);  /* For characters > 255 */
+#endif
+        }
+      try_next = FALSE;
+      break;
+
+      /* Single character types set the bits and stop. Note that if PCRE_UCP
+      is set, we do not see these op codes because \d etc are converted to
+      properties. Therefore, these apply in the case when only characters less
+      than 256 are recognized to match the types. */
+
+      case OP_NOT_DIGIT:
+      set_nottype_bits(start_bits, cbit_digit, table_limit, cd);
+      try_next = FALSE;
+      break;
+
+      case OP_DIGIT:
+      set_type_bits(start_bits, cbit_digit, table_limit, cd);
+      try_next = FALSE;
+      break;
+
+      /* The cbit_space table has vertical tab as whitespace; we have to
+      ensure it is set as not whitespace. Luckily, the code value is the same
+      (0x0b) in ASCII and EBCDIC, so we can just adjust the appropriate bit. */
+
+      case OP_NOT_WHITESPACE:
+      set_nottype_bits(start_bits, cbit_space, table_limit, cd);
+      start_bits[1] |= 0x08;
+      try_next = FALSE;
+      break;
+
+      /* The cbit_space table has vertical tab as whitespace; we have to not
+      set it from the table. Luckily, the code value is the same (0x0b) in
+      ASCII and EBCDIC, so we can just adjust the appropriate bit. */
+
+      case OP_WHITESPACE:
+      c = start_bits[1];    /* Save in case it was already set */
+      set_type_bits(start_bits, cbit_space, table_limit, cd);
+      start_bits[1] = (start_bits[1] & ~0x08) | c;
+      try_next = FALSE;
+      break;
+
+      case OP_NOT_WORDCHAR:
+      set_nottype_bits(start_bits, cbit_word, table_limit, cd);
+      try_next = FALSE;
+      break;
+
+      case OP_WORDCHAR:
+      set_type_bits(start_bits, cbit_word, table_limit, cd);
+      try_next = FALSE;
+      break;
+
+      /* One or more character type fudges the pointer and restarts, knowing
+      it will hit a single character type and stop there. */
+
+      case OP_TYPEPLUS:
+      case OP_TYPEMINPLUS:
+      case OP_TYPEPOSPLUS:
+      tcode++;
+      break;
+
+      case OP_TYPEEXACT:
+      tcode += 1 + IMM2_SIZE;
+      break;
+
+      /* Zero or more repeats of character types set the bits and then
+      try again. */
+
+      case OP_TYPEUPTO:
+      case OP_TYPEMINUPTO:
+      case OP_TYPEPOSUPTO:
+      tcode += IMM2_SIZE;  /* Fall through */
+
+      case OP_TYPESTAR:
+      case OP_TYPEMINSTAR:
+      case OP_TYPEPOSSTAR:
+      case OP_TYPEQUERY:
+      case OP_TYPEMINQUERY:
+      case OP_TYPEPOSQUERY:
+      switch(tcode[1])
+        {
+        default:
+        case OP_ANY:
+        case OP_ALLANY:
+        return SSB_FAIL;
+
+        case OP_HSPACE:
+        SET_BIT(CHAR_HT);
+        SET_BIT(CHAR_SPACE);
+#ifdef SUPPORT_UTF
+        if (utf)
+          {
+#ifdef COMPILE_PCRE8
+          SET_BIT(0xC2);  /* For U+00A0 */
+          SET_BIT(0xE1);  /* For U+1680, U+180E */
+          SET_BIT(0xE2);  /* For U+2000 - U+200A, U+202F, U+205F */
+          SET_BIT(0xE3);  /* For U+3000 */
+#elif defined COMPILE_PCRE16 || defined COMPILE_PCRE32
+          SET_BIT(0xA0);
+          SET_BIT(0xFF);  /* For characters > 255 */
+#endif  /* COMPILE_PCRE[8|16|32] */
+          }
+        else
+#endif /* SUPPORT_UTF */
+#ifndef EBCDIC
+          SET_BIT(0xA0);
+#endif  /* Not EBCDIC */
+        break;
+
+        case OP_ANYNL:
+        case OP_VSPACE:
+        SET_BIT(CHAR_LF);
+        SET_BIT(CHAR_VT);
+        SET_BIT(CHAR_FF);
+        SET_BIT(CHAR_CR);
+#ifdef SUPPORT_UTF
+        if (utf)
+          {
+#ifdef COMPILE_PCRE8
+          SET_BIT(0xC2);  /* For U+0085 */
+          SET_BIT(0xE2);  /* For U+2028, U+2029 */
+#elif defined COMPILE_PCRE16 || defined COMPILE_PCRE32
+          SET_BIT(CHAR_NEL);
+          SET_BIT(0xFF);  /* For characters > 255 */
+#endif  /* COMPILE_PCRE16 */
+          }
+        else
+#endif /* SUPPORT_UTF */
+          SET_BIT(CHAR_NEL);
+        break;
+
+        case OP_NOT_DIGIT:
+        set_nottype_bits(start_bits, cbit_digit, table_limit, cd);
+        break;
+
+        case OP_DIGIT:
+        set_type_bits(start_bits, cbit_digit, table_limit, cd);
+        break;
+
+        /* The cbit_space table has vertical tab as whitespace; we have to
+        ensure it gets set as not whitespace. Luckily, the code value is the
+        same (0x0b) in ASCII and EBCDIC, so we can just adjust the appropriate
+        bit. */
+
+        case OP_NOT_WHITESPACE:
+        set_nottype_bits(start_bits, cbit_space, table_limit, cd);
+        start_bits[1] |= 0x08;
+        break;
+
+        /* The cbit_space table has vertical tab as whitespace; we have to
+        avoid setting it. Luckily, the code value is the same (0x0b) in ASCII
+        and EBCDIC, so we can just adjust the appropriate bit. */
+
+        case OP_WHITESPACE:
+        c = start_bits[1];    /* Save in case it was already set */
+        set_type_bits(start_bits, cbit_space, table_limit, cd);
+        start_bits[1] = (start_bits[1] & ~0x08) | c;
+        break;
+
+        case OP_NOT_WORDCHAR:
+        set_nottype_bits(start_bits, cbit_word, table_limit, cd);
+        break;
+
+        case OP_WORDCHAR:
+        set_type_bits(start_bits, cbit_word, table_limit, cd);
+        break;
+        }
+
+      tcode += 2;
+      break;
+
+      /* Character class where all the information is in a bit map: set the
+      bits and either carry on or not, according to the repeat count. If it was
+      a negative class, and we are operating with UTF-8 characters, any byte
+      with a value >= 0xc4 is a potentially valid starter because it starts a
+      character with a value > 255. */
+
+      case OP_NCLASS:
+#if defined SUPPORT_UTF && defined COMPILE_PCRE8
+      if (utf)
+        {
+        start_bits[24] |= 0xf0;              /* Bits for 0xc4 - 0xc8 */
+        memset(start_bits+25, 0xff, 7);      /* Bits for 0xc9 - 0xff */
+        }
+#endif
+#if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
+      SET_BIT(0xFF);                         /* For characters > 255 */
+#endif
+      /* Fall through */
+
+      case OP_CLASS:
+        {
+        pcre_uint8 *map;
+        tcode++;
+        map = (pcre_uint8 *)tcode;
+
+        /* In UTF-8 mode, the bits in a bit map correspond to character
+        values, not to byte values. However, the bit map we are constructing is
+        for byte values. So we have to do a conversion for characters whose
+        value is > 127. In fact, there are only two possible starting bytes for
+        characters in the range 128 - 255. */
+
+#if defined SUPPORT_UTF && defined COMPILE_PCRE8
+        if (utf)
+          {
+          for (c = 0; c < 16; c++) start_bits[c] |= map[c];
+          for (c = 128; c < 256; c++)
+            {
+            if ((map[c/8] && (1 << (c&7))) != 0)
+              {
+              int d = (c >> 6) | 0xc0;            /* Set bit for this starter */
+              start_bits[d/8] |= (1 << (d&7));    /* and then skip on to the */
+              c = (c & 0xc0) + 0x40 - 1;          /* next relevant character. */
+              }
+            }
+          }
+        else
+#endif
+          {
+          /* In non-UTF-8 mode, the two bit maps are completely compatible. */
+          for (c = 0; c < 32; c++) start_bits[c] |= map[c];
+          }
+
+        /* Advance past the bit map, and act on what follows. For a zero
+        minimum repeat, continue; otherwise stop processing. */
+
+        tcode += 32 / sizeof(pcre_uchar);
+        switch (*tcode)
+          {
+          case OP_CRSTAR:
+          case OP_CRMINSTAR:
+          case OP_CRQUERY:
+          case OP_CRMINQUERY:
+          tcode++;
+          break;
+
+          case OP_CRRANGE:
+          case OP_CRMINRANGE:
+          if (GET2(tcode, 1) == 0) tcode += 1 + 2 * IMM2_SIZE;
+            else try_next = FALSE;
+          break;
+
+          default:
+          try_next = FALSE;
+          break;
+          }
+        }
+      break; /* End of bitmap class handling */
+
+      }      /* End of switch */
+    }        /* End of try_next loop */
+
+  code += GET(code, 1);   /* Advance to next branch */
+  }
+while (*code == OP_ALT);
+return yield;
+}
+
+
+
+
+
+/*************************************************
+*          Study a compiled expression           *
+*************************************************/
+
+/* This function is handed a compiled expression that it must study to produce
+information that will speed up the matching. It returns a pcre[16]_extra block
+which then gets handed back to pcre_exec().
+
+Arguments:
+  re        points to the compiled expression
+  options   contains option bits
+  errorptr  points to where to place error messages;
+            set NULL unless error
+
+Returns:    pointer to a pcre[16]_extra block, with study_data filled in and
+              the appropriate flags set;
+            NULL on error or if no optimization possible
+*/
+
+#if defined COMPILE_PCRE8
+PCRE_EXP_DEFN pcre_extra * PCRE_CALL_CONVENTION
+pcre_study(const pcre *external_re, int options, const char **errorptr)
+#elif defined COMPILE_PCRE16
+PCRE_EXP_DEFN pcre16_extra * PCRE_CALL_CONVENTION
+pcre16_study(const pcre16 *external_re, int options, const char **errorptr)
+#elif defined COMPILE_PCRE32
+PCRE_EXP_DEFN pcre32_extra * PCRE_CALL_CONVENTION
+pcre32_study(const pcre32 *external_re, int options, const char **errorptr)
+#endif
+{
+int min;
+BOOL bits_set = FALSE;
+pcre_uint8 start_bits[32];
+PUBL(extra) *extra = NULL;
+pcre_study_data *study;
+const pcre_uint8 *tables;
+pcre_uchar *code;
+compile_data compile_block;
+const REAL_PCRE *re = (const REAL_PCRE *)external_re;
+
+*errorptr = NULL;
+
+if (re == NULL || re->magic_number != MAGIC_NUMBER)
+  {
+  *errorptr = "argument is not a compiled regular expression";
+  return NULL;
+  }
+
+if ((re->flags & PCRE_MODE) == 0)
+  {
+#if defined COMPILE_PCRE8
+  *errorptr = "argument not compiled in 8 bit mode";
+#elif defined COMPILE_PCRE16
+  *errorptr = "argument not compiled in 16 bit mode";
+#elif defined COMPILE_PCRE32
+  *errorptr = "argument not compiled in 32 bit mode";
+#endif
+  return NULL;
+  }
+
+if ((options & ~PUBLIC_STUDY_OPTIONS) != 0)
+  {
+  *errorptr = "unknown or incorrect option bit(s) set";
+  return NULL;
+  }
+
+code = (pcre_uchar *)re + re->name_table_offset +
+  (re->name_count * re->name_entry_size);
+
+/* For an anchored pattern, or an unanchored pattern that has a first char, or
+a multiline pattern that matches only at "line starts", there is no point in
+seeking a list of starting bytes. */
+
+if ((re->options & PCRE_ANCHORED) == 0 &&
+    (re->flags & (PCRE_FIRSTSET|PCRE_STARTLINE)) == 0)
+  {
+  int rc;
+
+  /* Set the character tables in the block that is passed around */
+
+  tables = re->tables;
+
+#if defined COMPILE_PCRE8
+  if (tables == NULL)
+    (void)pcre_fullinfo(external_re, NULL, PCRE_INFO_DEFAULT_TABLES,
+    (void *)(&tables));
+#elif defined COMPILE_PCRE16
+  if (tables == NULL)
+    (void)pcre16_fullinfo(external_re, NULL, PCRE_INFO_DEFAULT_TABLES,
+    (void *)(&tables));
+#elif defined COMPILE_PCRE32
+  if (tables == NULL)
+    (void)pcre32_fullinfo(external_re, NULL, PCRE_INFO_DEFAULT_TABLES,
+    (void *)(&tables));
+#endif
+
+  compile_block.lcc = tables + lcc_offset;
+  compile_block.fcc = tables + fcc_offset;
+  compile_block.cbits = tables + cbits_offset;
+  compile_block.ctypes = tables + ctypes_offset;
+
+  /* See if we can find a fixed set of initial characters for the pattern. */
+
+  memset(start_bits, 0, 32 * sizeof(pcre_uint8));
+  rc = set_start_bits(code, start_bits, (re->options & PCRE_UTF8) != 0,
+    &compile_block);
+  bits_set = rc == SSB_DONE;
+  if (rc == SSB_UNKNOWN)
+    {
+    *errorptr = "internal error: opcode not recognized";
+    return NULL;
+    }
+  }
+
+/* Find the minimum length of subject string. */
+
+switch(min = find_minlength(code, code, re->options, 0))
+  {
+  case -2: *errorptr = "internal error: missing capturing bracket"; return NULL;
+  case -3: *errorptr = "internal error: opcode not recognized"; return NULL;
+  default: break;
+  }
+
+/* If a set of starting bytes has been identified, or if the minimum length is
+greater than zero, or if JIT optimization has been requested, or if
+PCRE_STUDY_EXTRA_NEEDED is set, get a pcre[16]_extra block and a
+pcre_study_data block. The study data is put in the latter, which is pointed to
+by the former, which may also get additional data set later by the calling
+program. At the moment, the size of pcre_study_data is fixed. We nevertheless
+save it in a field for returning via the pcre_fullinfo() function so that if it
+becomes variable in the future, we don't have to change that code. */
+
+if (bits_set || min > 0 || (options & (
+#ifdef SUPPORT_JIT
+    PCRE_STUDY_JIT_COMPILE | PCRE_STUDY_JIT_PARTIAL_SOFT_COMPILE |
+    PCRE_STUDY_JIT_PARTIAL_HARD_COMPILE |
+#endif
+    PCRE_STUDY_EXTRA_NEEDED)) != 0)
+  {
+  extra = (PUBL(extra) *)(PUBL(malloc))
+    (sizeof(PUBL(extra)) + sizeof(pcre_study_data));
+  if (extra == NULL)
+    {
+    *errorptr = "failed to get memory";
+    return NULL;
+    }
+
+  study = (pcre_study_data *)((char *)extra + sizeof(PUBL(extra)));
+  extra->flags = PCRE_EXTRA_STUDY_DATA;
+  extra->study_data = study;
+
+  study->size = sizeof(pcre_study_data);
+  study->flags = 0;
+
+  /* Set the start bits always, to avoid unset memory errors if the
+  study data is written to a file, but set the flag only if any of the bits
+  are set, to save time looking when none are. */
+
+  if (bits_set)
+    {
+    study->flags |= PCRE_STUDY_MAPPED;
+    memcpy(study->start_bits, start_bits, sizeof(start_bits));
+    }
+  else memset(study->start_bits, 0, 32 * sizeof(pcre_uint8));
+
+#ifdef PCRE_DEBUG
+  if (bits_set)
+    {
+    pcre_uint8 *ptr = start_bits;
+    int i;
+
+    printf("Start bits:\n");
+    for (i = 0; i < 32; i++)
+      printf("%3d: %02x%s", i * 8, *ptr++, ((i + 1) & 0x7) != 0? " " : "\n");
+    }
+#endif
+
+  /* Always set the minlength value in the block, because the JIT compiler
+  makes use of it. However, don't set the bit unless the length is greater than
+  zero - the interpretive pcre_exec() and pcre_dfa_exec() needn't waste time
+  checking the zero case. */
+
+  if (min > 0)
+    {
+    study->flags |= PCRE_STUDY_MINLEN;
+    study->minlength = min;
+    }
+  else study->minlength = 0;
+
+  /* If JIT support was compiled and requested, attempt the JIT compilation.
+  If no starting bytes were found, and the minimum length is zero, and JIT
+  compilation fails, abandon the extra block and return NULL, unless
+  PCRE_STUDY_EXTRA_NEEDED is set. */
+
+#ifdef SUPPORT_JIT
+  extra->executable_jit = NULL;
+  if ((options & PCRE_STUDY_JIT_COMPILE) != 0)
+    PRIV(jit_compile)(re, extra, JIT_COMPILE);
+  if ((options & PCRE_STUDY_JIT_PARTIAL_SOFT_COMPILE) != 0)
+    PRIV(jit_compile)(re, extra, JIT_PARTIAL_SOFT_COMPILE);
+  if ((options & PCRE_STUDY_JIT_PARTIAL_HARD_COMPILE) != 0)
+    PRIV(jit_compile)(re, extra, JIT_PARTIAL_HARD_COMPILE);
+
+  if (study->flags == 0 && (extra->flags & PCRE_EXTRA_EXECUTABLE_JIT) == 0 &&
+      (options & PCRE_STUDY_EXTRA_NEEDED) == 0)
+    {
+#if defined COMPILE_PCRE8
+    pcre_free_study(extra);
+#elif defined COMPILE_PCRE16
+    pcre16_free_study(extra);
+#elif defined COMPILE_PCRE32
+    pcre32_free_study(extra);
+#endif
+    extra = NULL;
+    }
+#endif
+  }
+
+return extra;
+}
+
+
+/*************************************************
+*          Free the study data                   *
+*************************************************/
+
+/* This function frees the memory that was obtained by pcre_study().
+
+Argument:   a pointer to the pcre[16]_extra block
+Returns:    nothing
+*/
+
+#if defined COMPILE_PCRE8
+PCRE_EXP_DEFN void
+pcre_free_study(pcre_extra *extra)
+#elif defined COMPILE_PCRE16
+PCRE_EXP_DEFN void
+pcre16_free_study(pcre16_extra *extra)
+#elif defined COMPILE_PCRE32
+PCRE_EXP_DEFN void
+pcre32_free_study(pcre32_extra *extra)
+#endif
+{
+if (extra == NULL)
+  return;
+#ifdef SUPPORT_JIT
+if ((extra->flags & PCRE_EXTRA_EXECUTABLE_JIT) != 0 &&
+     extra->executable_jit != NULL)
+  PRIV(jit_free)(extra->executable_jit);
+#endif
+PUBL(free)(extra);
+}
+
+/* End of pcre_study.c */
diff --git a/pcre/pcre_tables.c b/pcre/pcre_tables.c
new file mode 100644
index 00000000000..f38ab52cbb8
--- /dev/null
+++ b/pcre/pcre_tables.c
@@ -0,0 +1,658 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+#ifndef PCRE_INCLUDED
+
+/* This module contains some fixed tables that are used by more than one of the
+PCRE code modules. The tables are also #included by the pcretest program, which
+uses macros to change their names from _pcre_xxx to xxxx, thereby avoiding name
+clashes with the library. */
+
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include "pcre_internal.h"
+
+#endif /* PCRE_INCLUDED */
+
+/* Table of sizes for the fixed-length opcodes. It's defined in a macro so that
+the definition is next to the definition of the opcodes in pcre_internal.h. */
+
+const pcre_uint8 PRIV(OP_lengths)[] = { OP_LENGTHS };
+
+/* Tables of horizontal and vertical whitespace characters, suitable for
+adding to classes. */
+
+const pcre_uint32 PRIV(hspace_list)[] = { HSPACE_LIST };
+const pcre_uint32 PRIV(vspace_list)[] = { VSPACE_LIST };
+
+
+
+/*************************************************
+*           Tables for UTF-8 support             *
+*************************************************/
+
+/* These are the breakpoints for different numbers of bytes in a UTF-8
+character. */
+
+#if (defined SUPPORT_UTF && defined COMPILE_PCRE8) \
+  || (defined PCRE_INCLUDED && (defined SUPPORT_PCRE16 || defined SUPPORT_PCRE32))
+
+/* These tables are also required by pcretest in 16- or 32-bit mode. */
+
+const int PRIV(utf8_table1)[] =
+  { 0x7f, 0x7ff, 0xffff, 0x1fffff, 0x3ffffff, 0x7fffffff};
+
+const int PRIV(utf8_table1_size) = sizeof(PRIV(utf8_table1)) / sizeof(int);
+
+/* These are the indicator bits and the mask for the data bits to set in the
+first byte of a character, indexed by the number of additional bytes. */
+
+const int PRIV(utf8_table2)[] = { 0,    0xc0, 0xe0, 0xf0, 0xf8, 0xfc};
+const int PRIV(utf8_table3)[] = { 0xff, 0x1f, 0x0f, 0x07, 0x03, 0x01};
+
+/* Table of the number of extra bytes, indexed by the first byte masked with
+0x3f. The highest number for a valid UTF-8 first byte is in fact 0x3d. */
+
+const pcre_uint8 PRIV(utf8_table4)[] = {
+  1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
+  1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
+  2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,
+  3,3,3,3,3,3,3,3,4,4,4,4,5,5,5,5 };
+
+#endif /* (SUPPORT_UTF && COMPILE_PCRE8) || (PCRE_INCLUDED && SUPPORT_PCRE[16|32])*/
+
+#ifdef SUPPORT_UTF
+
+/* Table to translate from particular type value to the general value. */
+
+const pcre_uint32 PRIV(ucp_gentype)[] = {
+  ucp_C, ucp_C, ucp_C, ucp_C, ucp_C,  /* Cc, Cf, Cn, Co, Cs */
+  ucp_L, ucp_L, ucp_L, ucp_L, ucp_L,  /* Ll, Lu, Lm, Lo, Lt */
+  ucp_M, ucp_M, ucp_M,                /* Mc, Me, Mn */
+  ucp_N, ucp_N, ucp_N,                /* Nd, Nl, No */
+  ucp_P, ucp_P, ucp_P, ucp_P, ucp_P,  /* Pc, Pd, Pe, Pf, Pi */
+  ucp_P, ucp_P,                       /* Ps, Po */
+  ucp_S, ucp_S, ucp_S, ucp_S,         /* Sc, Sk, Sm, So */
+  ucp_Z, ucp_Z, ucp_Z                 /* Zl, Zp, Zs */
+};
+
+/* This table encodes the rules for finding the end of an extended grapheme
+cluster. Every code point has a grapheme break property which is one of the
+ucp_gbXX values defined in ucp.h. The 2-dimensional table is indexed by the
+properties of two adjacent code points. The left property selects a word from
+the table, and the right property selects a bit from that word like this:
+
+  ucp_gbtable[left-property] & (1 << right-property)
+
+The value is non-zero if a grapheme break is NOT permitted between the relevant
+two code points. The breaking rules are as follows:
+
+1. Break at the start and end of text (pretty obviously).
+
+2. Do not break between a CR and LF; otherwise, break before and   after
+   controls.
+
+3. Do not break Hangul syllable sequences, the rules for which are:
+
+    L may be followed by L, V, LV or LVT
+    LV or V may be followed by V or T
+    LVT or T may be followed by T
+
+4. Do not break before extending characters.
+
+The next two rules are only for extended grapheme clusters (but that's what we
+are implementing).
+
+5. Do not break before SpacingMarks.
+
+6. Do not break after Prepend characters.
+
+7. Otherwise, break everywhere.
+*/
+
+const pcre_uint32 PRIV(ucp_gbtable[]) = {
+   (1< 0x10ffff is not permitted
+PCRE_UTF8_ERR14  3-byte character with value 0xd000-0xdfff is not permitted
+PCRE_UTF8_ERR15  Overlong 2-byte sequence
+PCRE_UTF8_ERR16  Overlong 3-byte sequence
+PCRE_UTF8_ERR17  Overlong 4-byte sequence
+PCRE_UTF8_ERR18  Overlong 5-byte sequence (won't ever occur)
+PCRE_UTF8_ERR19  Overlong 6-byte sequence (won't ever occur)
+PCRE_UTF8_ERR20  Isolated 0x80 byte (not within UTF-8 character)
+PCRE_UTF8_ERR21  Byte with the illegal value 0xfe or 0xff
+PCRE_UTF8_ERR22  Unused (was non-character)
+
+Arguments:
+  string       points to the string
+  length       length of string, or -1 if the string is zero-terminated
+  errp         pointer to an error position offset variable
+
+Returns:       = 0    if the string is a valid UTF-8 string
+               > 0    otherwise, setting the offset of the bad character
+*/
+
+int
+PRIV(valid_utf)(PCRE_PUCHAR string, int length, int *erroroffset)
+{
+#ifdef SUPPORT_UTF
+register PCRE_PUCHAR p;
+
+if (length < 0)
+  {
+  for (p = string; *p != 0; p++);
+  length = (int)(p - string);
+  }
+
+for (p = string; length-- > 0; p++)
+  {
+  register pcre_uchar ab, c, d;
+
+  c = *p;
+  if (c < 128) continue;                /* ASCII character */
+
+  if (c < 0xc0)                         /* Isolated 10xx xxxx byte */
+    {
+    *erroroffset = (int)(p - string);
+    return PCRE_UTF8_ERR20;
+    }
+
+  if (c >= 0xfe)                        /* Invalid 0xfe or 0xff bytes */
+    {
+    *erroroffset = (int)(p - string);
+    return PCRE_UTF8_ERR21;
+    }
+
+  ab = PRIV(utf8_table4)[c & 0x3f];     /* Number of additional bytes */
+  if (length < ab)
+    {
+    *erroroffset = (int)(p - string);          /* Missing bytes */
+    return ab - length;                 /* Codes ERR1 to ERR5 */
+    }
+  length -= ab;                         /* Length remaining */
+
+  /* Check top bits in the second byte */
+
+  if (((d = *(++p)) & 0xc0) != 0x80)
+    {
+    *erroroffset = (int)(p - string) - 1;
+    return PCRE_UTF8_ERR6;
+    }
+
+  /* For each length, check that the remaining bytes start with the 0x80 bit
+  set and not the 0x40 bit. Then check for an overlong sequence, and for the
+  excluded range 0xd800 to 0xdfff. */
+
+  switch (ab)
+    {
+    /* 2-byte character. No further bytes to check for 0x80. Check first byte
+    for for xx00 000x (overlong sequence). */
+
+    case 1: if ((c & 0x3e) == 0)
+      {
+      *erroroffset = (int)(p - string) - 1;
+      return PCRE_UTF8_ERR15;
+      }
+    break;
+
+    /* 3-byte character. Check third byte for 0x80. Then check first 2 bytes
+      for 1110 0000, xx0x xxxx (overlong sequence) or
+          1110 1101, 1010 xxxx (0xd800 - 0xdfff) */
+
+    case 2:
+    if ((*(++p) & 0xc0) != 0x80)     /* Third byte */
+      {
+      *erroroffset = (int)(p - string) - 2;
+      return PCRE_UTF8_ERR7;
+      }
+    if (c == 0xe0 && (d & 0x20) == 0)
+      {
+      *erroroffset = (int)(p - string) - 2;
+      return PCRE_UTF8_ERR16;
+      }
+    if (c == 0xed && d >= 0xa0)
+      {
+      *erroroffset = (int)(p - string) - 2;
+      return PCRE_UTF8_ERR14;
+      }
+    break;
+
+    /* 4-byte character. Check 3rd and 4th bytes for 0x80. Then check first 2
+       bytes for for 1111 0000, xx00 xxxx (overlong sequence), then check for a
+       character greater than 0x0010ffff (f4 8f bf bf) */
+
+    case 3:
+    if ((*(++p) & 0xc0) != 0x80)     /* Third byte */
+      {
+      *erroroffset = (int)(p - string) - 2;
+      return PCRE_UTF8_ERR7;
+      }
+    if ((*(++p) & 0xc0) != 0x80)     /* Fourth byte */
+      {
+      *erroroffset = (int)(p - string) - 3;
+      return PCRE_UTF8_ERR8;
+      }
+    if (c == 0xf0 && (d & 0x30) == 0)
+      {
+      *erroroffset = (int)(p - string) - 3;
+      return PCRE_UTF8_ERR17;
+      }
+    if (c > 0xf4 || (c == 0xf4 && d > 0x8f))
+      {
+      *erroroffset = (int)(p - string) - 3;
+      return PCRE_UTF8_ERR13;
+      }
+    break;
+
+    /* 5-byte and 6-byte characters are not allowed by RFC 3629, and will be
+    rejected by the length test below. However, we do the appropriate tests
+    here so that overlong sequences get diagnosed, and also in case there is
+    ever an option for handling these larger code points. */
+
+    /* 5-byte character. Check 3rd, 4th, and 5th bytes for 0x80. Then check for
+    1111 1000, xx00 0xxx */
+
+    case 4:
+    if ((*(++p) & 0xc0) != 0x80)     /* Third byte */
+      {
+      *erroroffset = (int)(p - string) - 2;
+      return PCRE_UTF8_ERR7;
+      }
+    if ((*(++p) & 0xc0) != 0x80)     /* Fourth byte */
+      {
+      *erroroffset = (int)(p - string) - 3;
+      return PCRE_UTF8_ERR8;
+      }
+    if ((*(++p) & 0xc0) != 0x80)     /* Fifth byte */
+      {
+      *erroroffset = (int)(p - string) - 4;
+      return PCRE_UTF8_ERR9;
+      }
+    if (c == 0xf8 && (d & 0x38) == 0)
+      {
+      *erroroffset = (int)(p - string) - 4;
+      return PCRE_UTF8_ERR18;
+      }
+    break;
+
+    /* 6-byte character. Check 3rd-6th bytes for 0x80. Then check for
+    1111 1100, xx00 00xx. */
+
+    case 5:
+    if ((*(++p) & 0xc0) != 0x80)     /* Third byte */
+      {
+      *erroroffset = (int)(p - string) - 2;
+      return PCRE_UTF8_ERR7;
+      }
+    if ((*(++p) & 0xc0) != 0x80)     /* Fourth byte */
+      {
+      *erroroffset = (int)(p - string) - 3;
+      return PCRE_UTF8_ERR8;
+      }
+    if ((*(++p) & 0xc0) != 0x80)     /* Fifth byte */
+      {
+      *erroroffset = (int)(p - string) - 4;
+      return PCRE_UTF8_ERR9;
+      }
+    if ((*(++p) & 0xc0) != 0x80)     /* Sixth byte */
+      {
+      *erroroffset = (int)(p - string) - 5;
+      return PCRE_UTF8_ERR10;
+      }
+    if (c == 0xfc && (d & 0x3c) == 0)
+      {
+      *erroroffset = (int)(p - string) - 5;
+      return PCRE_UTF8_ERR19;
+      }
+    break;
+    }
+
+  /* Character is valid under RFC 2279, but 4-byte and 5-byte characters are
+  excluded by RFC 3629. The pointer p is currently at the last byte of the
+  character. */
+
+  if (ab > 3)
+    {
+    *erroroffset = (int)(p - string) - ab;
+    return (ab == 4)? PCRE_UTF8_ERR11 : PCRE_UTF8_ERR12;
+    }
+  }
+
+#else  /* Not SUPPORT_UTF */
+(void)(string);  /* Keep picky compilers happy */
+(void)(length);
+(void)(erroroffset);
+#endif
+
+return PCRE_UTF8_ERR0;   /* This indicates success */
+}
+
+/* End of pcre_valid_utf8.c */
diff --git a/pcre/pcre_version.c b/pcre/pcre_version.c
new file mode 100644
index 00000000000..ae86ff28bc8
--- /dev/null
+++ b/pcre/pcre_version.c
@@ -0,0 +1,98 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+
+/* This module contains the external function pcre_version(), which returns a
+string that identifies the PCRE version that is in use. */
+
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include "pcre_internal.h"
+
+
+/*************************************************
+*          Return version string                 *
+*************************************************/
+
+/* These macros are the standard way of turning unquoted text into C strings.
+They allow macros like PCRE_MAJOR to be defined without quotes, which is
+convenient for user programs that want to test its value. */
+
+#define STRING(a)  # a
+#define XSTRING(s) STRING(s)
+
+/* A problem turned up with PCRE_PRERELEASE, which is defined empty for
+production releases. Originally, it was used naively in this code:
+
+  return XSTRING(PCRE_MAJOR)
+         "." XSTRING(PCRE_MINOR)
+             XSTRING(PCRE_PRERELEASE)
+         " " XSTRING(PCRE_DATE);
+
+However, when PCRE_PRERELEASE is empty, this leads to an attempted expansion of
+STRING(). The C standard states: "If (before argument substitution) any
+argument consists of no preprocessing tokens, the behavior is undefined." It
+turns out the gcc treats this case as a single empty string - which is what we
+really want - but Visual C grumbles about the lack of an argument for the
+macro. Unfortunately, both are within their rights. To cope with both ways of
+handling this, I had resort to some messy hackery that does a test at run time.
+I could find no way of detecting that a macro is defined as an empty string at
+pre-processor time. This hack uses a standard trick for avoiding calling
+the STRING macro with an empty argument when doing the test. */
+
+#if defined COMPILE_PCRE8
+PCRE_EXP_DEFN const char * PCRE_CALL_CONVENTION
+pcre_version(void)
+#elif defined COMPILE_PCRE16
+PCRE_EXP_DEFN const char * PCRE_CALL_CONVENTION
+pcre16_version(void)
+#elif defined COMPILE_PCRE32
+PCRE_EXP_DEFN const char * PCRE_CALL_CONVENTION
+pcre32_version(void)
+#endif
+{
+return (XSTRING(Z PCRE_PRERELEASE)[1] == 0)?
+  XSTRING(PCRE_MAJOR.PCRE_MINOR PCRE_DATE) :
+  XSTRING(PCRE_MAJOR.PCRE_MINOR) XSTRING(PCRE_PRERELEASE PCRE_DATE);
+}
+
+/* End of pcre_version.c */
diff --git a/pcre/pcre_xclass.c b/pcre/pcre_xclass.c
new file mode 100644
index 00000000000..d777acb57c9
--- /dev/null
+++ b/pcre/pcre_xclass.c
@@ -0,0 +1,212 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2013 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+
+/* This module contains an internal function that is used to match an extended
+class. It is used by both pcre_exec() and pcre_def_exec(). */
+
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include "pcre_internal.h"
+
+
+/*************************************************
+*       Match character against an XCLASS        *
+*************************************************/
+
+/* This function is called to match a character against an extended class that
+might contain values > 255 and/or Unicode properties.
+
+Arguments:
+  c           the character
+  data        points to the flag byte of the XCLASS data
+
+Returns:      TRUE if character matches, else FALSE
+*/
+
+BOOL
+PRIV(xclass)(pcre_uint32 c, const pcre_uchar *data, BOOL utf)
+{
+pcre_uchar t;
+BOOL negated = (*data & XCL_NOT) != 0;
+
+(void)utf;
+#ifdef COMPILE_PCRE8
+/* In 8 bit mode, this must always be TRUE. Help the compiler to know that. */
+utf = TRUE;
+#endif
+
+/* Character values < 256 are matched against a bitmap, if one is present. If
+not, we still carry on, because there may be ranges that start below 256 in the
+additional data. */
+
+if (c < 256)
+  {
+  if ((*data & XCL_MAP) != 0 &&
+    (((pcre_uint8 *)(data + 1))[c/8] & (1 << (c&7))) != 0)
+    return !negated; /* char found */
+  }
+
+/* First skip the bit map if present. Then match against the list of Unicode
+properties or large chars or ranges that end with a large char. We won't ever
+encounter XCL_PROP or XCL_NOTPROP when UCP support is not compiled. */
+
+if ((*data++ & XCL_MAP) != 0) data += 32 / sizeof(pcre_uchar);
+
+while ((t = *data++) != XCL_END)
+  {
+  pcre_uint32 x, y;
+  if (t == XCL_SINGLE)
+    {
+#ifdef SUPPORT_UTF
+    if (utf)
+      {
+      GETCHARINC(x, data); /* macro generates multiple statements */
+      }
+    else
+#endif
+      x = *data++;
+    if (c == x) return !negated;
+    }
+  else if (t == XCL_RANGE)
+    {
+#ifdef SUPPORT_UTF
+    if (utf)
+      {
+      GETCHARINC(x, data); /* macro generates multiple statements */
+      GETCHARINC(y, data); /* macro generates multiple statements */
+      }
+    else
+#endif
+      {
+      x = *data++;
+      y = *data++;
+      }
+    if (c >= x && c <= y) return !negated;
+    }
+
+#ifdef SUPPORT_UCP
+  else  /* XCL_PROP & XCL_NOTPROP */
+    {
+    const ucd_record *prop = GET_UCD(c);
+
+    switch(*data)
+      {
+      case PT_ANY:
+      if (t == XCL_PROP) return !negated;
+      break;
+
+      case PT_LAMP:
+      if ((prop->chartype == ucp_Lu || prop->chartype == ucp_Ll ||
+           prop->chartype == ucp_Lt) == (t == XCL_PROP)) return !negated;
+      break;
+
+      case PT_GC:
+      if ((data[1] == PRIV(ucp_gentype)[prop->chartype]) == (t == XCL_PROP))
+        return !negated;
+      break;
+
+      case PT_PC:
+      if ((data[1] == prop->chartype) == (t == XCL_PROP)) return !negated;
+      break;
+
+      case PT_SC:
+      if ((data[1] == prop->script) == (t == XCL_PROP)) return !negated;
+      break;
+
+      case PT_ALNUM:
+      if ((PRIV(ucp_gentype)[prop->chartype] == ucp_L ||
+           PRIV(ucp_gentype)[prop->chartype] == ucp_N) == (t == XCL_PROP))
+        return !negated;
+      break;
+
+      case PT_SPACE:    /* Perl space */
+      if ((PRIV(ucp_gentype)[prop->chartype] == ucp_Z ||
+           c == CHAR_HT || c == CHAR_NL || c == CHAR_FF || c == CHAR_CR)
+             == (t == XCL_PROP))
+        return !negated;
+      break;
+
+      case PT_PXSPACE:  /* POSIX space */
+      if ((PRIV(ucp_gentype)[prop->chartype] == ucp_Z ||
+           c == CHAR_HT || c == CHAR_NL || c == CHAR_VT ||
+           c == CHAR_FF || c == CHAR_CR) == (t == XCL_PROP))
+        return !negated;
+      break;
+
+      case PT_WORD:
+      if ((PRIV(ucp_gentype)[prop->chartype] == ucp_L ||
+           PRIV(ucp_gentype)[prop->chartype] == ucp_N || c == CHAR_UNDERSCORE)
+             == (t == XCL_PROP))
+        return !negated;
+      break;
+
+      case PT_UCNC:
+      if (c < 0xa0)
+        {
+        if ((c == CHAR_DOLLAR_SIGN || c == CHAR_COMMERCIAL_AT ||
+             c == CHAR_GRAVE_ACCENT) == (t == XCL_PROP))
+          return !negated;
+        }
+      else
+        {
+        if ((c < 0xd800 || c > 0xdfff) == (t == XCL_PROP))
+          return !negated;
+        }
+      break;
+
+      /* This should never occur, but compilers may mutter if there is no
+      default. */
+
+      default:
+      return FALSE;
+      }
+
+    data += 2;
+    }
+#endif  /* SUPPORT_UCP */
+  }
+
+return negated;   /* char did not match */
+}
+
+/* End of pcre_xclass.c */
diff --git a/pcre/pcrecpp.cc b/pcre/pcrecpp.cc
new file mode 100644
index 00000000000..c0ba9ca14e1
--- /dev/null
+++ b/pcre/pcrecpp.cc
@@ -0,0 +1,922 @@
+// Copyright (c) 2010, Google Inc.
+// All rights reserved.
+//
+// Redistribution and use in source and binary forms, with or without
+// modification, are permitted provided that the following conditions are
+// met:
+//
+//     * Redistributions of source code must retain the above copyright
+// notice, this list of conditions and the following disclaimer.
+//     * Redistributions in binary form must reproduce the above
+// copyright notice, this list of conditions and the following disclaimer
+// in the documentation and/or other materials provided with the
+// distribution.
+//     * Neither the name of Google Inc. nor the names of its
+// contributors may be used to endorse or promote products derived from
+// this software without specific prior written permission.
+//
+// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+//
+// Author: Sanjay Ghemawat
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include 
+#include 
+#include 
+#include       /* for SHRT_MIN, USHRT_MAX, etc */
+#include       /* for memcpy */
+#include 
+#include 
+#include 
+#include 
+
+#include "pcrecpp_internal.h"
+#include "pcre.h"
+#include "pcrecpp.h"
+#include "pcre_stringpiece.h"
+
+
+namespace pcrecpp {
+
+// Maximum number of args we can set
+static const int kMaxArgs = 16;
+static const int kVecSize = (1 + kMaxArgs) * 3;  // results + PCRE workspace
+
+// Special object that stands-in for no argument
+Arg RE::no_arg((void*)NULL);
+
+// This is for ABI compatibility with old versions of pcre (pre-7.6),
+// which defined a global no_arg variable instead of putting it in the
+// RE class.  This works on GCC >= 3, at least.  It definitely works
+// for ELF, but may not for other object formats (Mach-O, for
+// instance, does not support aliases.)  We could probably have a more
+// inclusive test if we ever needed it.  (Note that not only the
+// __attribute__ syntax, but also __USER_LABEL_PREFIX__, are
+// gnu-specific.)
+#if defined(__GNUC__) && __GNUC__ >= 3 && defined(__ELF__)
+# define ULP_AS_STRING(x)            ULP_AS_STRING_INTERNAL(x)
+# define ULP_AS_STRING_INTERNAL(x)   #x
+# define USER_LABEL_PREFIX_STR       ULP_AS_STRING(__USER_LABEL_PREFIX__)
+extern Arg no_arg
+  __attribute__((alias(USER_LABEL_PREFIX_STR "_ZN7pcrecpp2RE6no_argE")));
+#endif
+
+// If a regular expression has no error, its error_ field points here
+static const string empty_string;
+
+// If the user doesn't ask for any options, we just use this one
+static RE_Options default_options;
+
+void RE::Init(const string& pat, const RE_Options* options) {
+  pattern_ = pat;
+  if (options == NULL) {
+    options_ = default_options;
+  } else {
+    options_ = *options;
+  }
+  error_ = &empty_string;
+  re_full_ = NULL;
+  re_partial_ = NULL;
+
+  re_partial_ = Compile(UNANCHORED);
+  if (re_partial_ != NULL) {
+    re_full_ = Compile(ANCHOR_BOTH);
+  }
+}
+
+void RE::Cleanup() {
+  if (re_full_ != NULL)         (*pcre_free)(re_full_);
+  if (re_partial_ != NULL)      (*pcre_free)(re_partial_);
+  if (error_ != &empty_string)  delete error_;
+}
+
+
+RE::~RE() {
+  Cleanup();
+}
+
+
+pcre* RE::Compile(Anchor anchor) {
+  // First, convert RE_Options into pcre options
+  int pcre_options = 0;
+  pcre_options = options_.all_options();
+
+  // Special treatment for anchoring.  This is needed because at
+  // runtime pcre only provides an option for anchoring at the
+  // beginning of a string (unless you use offset).
+  //
+  // There are three types of anchoring we want:
+  //    UNANCHORED      Compile the original pattern, and use
+  //                    a pcre unanchored match.
+  //    ANCHOR_START    Compile the original pattern, and use
+  //                    a pcre anchored match.
+  //    ANCHOR_BOTH     Tack a "\z" to the end of the original pattern
+  //                    and use a pcre anchored match.
+
+  const char* compile_error;
+  int eoffset;
+  pcre* re;
+  if (anchor != ANCHOR_BOTH) {
+    re = pcre_compile(pattern_.c_str(), pcre_options,
+                      &compile_error, &eoffset, NULL);
+  } else {
+    // Tack a '\z' at the end of RE.  Parenthesize it first so that
+    // the '\z' applies to all top-level alternatives in the regexp.
+    string wrapped = "(?:";  // A non-counting grouping operator
+    wrapped += pattern_;
+    wrapped += ")\\z";
+    re = pcre_compile(wrapped.c_str(), pcre_options,
+                      &compile_error, &eoffset, NULL);
+  }
+  if (re == NULL) {
+    if (error_ == &empty_string) error_ = new string(compile_error);
+  }
+  return re;
+}
+
+/***** Matching interfaces *****/
+
+bool RE::FullMatch(const StringPiece& text,
+                   const Arg& ptr1,
+                   const Arg& ptr2,
+                   const Arg& ptr3,
+                   const Arg& ptr4,
+                   const Arg& ptr5,
+                   const Arg& ptr6,
+                   const Arg& ptr7,
+                   const Arg& ptr8,
+                   const Arg& ptr9,
+                   const Arg& ptr10,
+                   const Arg& ptr11,
+                   const Arg& ptr12,
+                   const Arg& ptr13,
+                   const Arg& ptr14,
+                   const Arg& ptr15,
+                   const Arg& ptr16) const {
+  const Arg* args[kMaxArgs];
+  int n = 0;
+  if (&ptr1  == &no_arg) goto done; args[n++] = &ptr1;
+  if (&ptr2  == &no_arg) goto done; args[n++] = &ptr2;
+  if (&ptr3  == &no_arg) goto done; args[n++] = &ptr3;
+  if (&ptr4  == &no_arg) goto done; args[n++] = &ptr4;
+  if (&ptr5  == &no_arg) goto done; args[n++] = &ptr5;
+  if (&ptr6  == &no_arg) goto done; args[n++] = &ptr6;
+  if (&ptr7  == &no_arg) goto done; args[n++] = &ptr7;
+  if (&ptr8  == &no_arg) goto done; args[n++] = &ptr8;
+  if (&ptr9  == &no_arg) goto done; args[n++] = &ptr9;
+  if (&ptr10 == &no_arg) goto done; args[n++] = &ptr10;
+  if (&ptr11 == &no_arg) goto done; args[n++] = &ptr11;
+  if (&ptr12 == &no_arg) goto done; args[n++] = &ptr12;
+  if (&ptr13 == &no_arg) goto done; args[n++] = &ptr13;
+  if (&ptr14 == &no_arg) goto done; args[n++] = &ptr14;
+  if (&ptr15 == &no_arg) goto done; args[n++] = &ptr15;
+  if (&ptr16 == &no_arg) goto done; args[n++] = &ptr16;
+ done:
+
+  int consumed;
+  int vec[kVecSize];
+  return DoMatchImpl(text, ANCHOR_BOTH, &consumed, args, n, vec, kVecSize);
+}
+
+bool RE::PartialMatch(const StringPiece& text,
+                      const Arg& ptr1,
+                      const Arg& ptr2,
+                      const Arg& ptr3,
+                      const Arg& ptr4,
+                      const Arg& ptr5,
+                      const Arg& ptr6,
+                      const Arg& ptr7,
+                      const Arg& ptr8,
+                      const Arg& ptr9,
+                      const Arg& ptr10,
+                      const Arg& ptr11,
+                      const Arg& ptr12,
+                      const Arg& ptr13,
+                      const Arg& ptr14,
+                      const Arg& ptr15,
+                      const Arg& ptr16) const {
+  const Arg* args[kMaxArgs];
+  int n = 0;
+  if (&ptr1  == &no_arg) goto done; args[n++] = &ptr1;
+  if (&ptr2  == &no_arg) goto done; args[n++] = &ptr2;
+  if (&ptr3  == &no_arg) goto done; args[n++] = &ptr3;
+  if (&ptr4  == &no_arg) goto done; args[n++] = &ptr4;
+  if (&ptr5  == &no_arg) goto done; args[n++] = &ptr5;
+  if (&ptr6  == &no_arg) goto done; args[n++] = &ptr6;
+  if (&ptr7  == &no_arg) goto done; args[n++] = &ptr7;
+  if (&ptr8  == &no_arg) goto done; args[n++] = &ptr8;
+  if (&ptr9  == &no_arg) goto done; args[n++] = &ptr9;
+  if (&ptr10 == &no_arg) goto done; args[n++] = &ptr10;
+  if (&ptr11 == &no_arg) goto done; args[n++] = &ptr11;
+  if (&ptr12 == &no_arg) goto done; args[n++] = &ptr12;
+  if (&ptr13 == &no_arg) goto done; args[n++] = &ptr13;
+  if (&ptr14 == &no_arg) goto done; args[n++] = &ptr14;
+  if (&ptr15 == &no_arg) goto done; args[n++] = &ptr15;
+  if (&ptr16 == &no_arg) goto done; args[n++] = &ptr16;
+ done:
+
+  int consumed;
+  int vec[kVecSize];
+  return DoMatchImpl(text, UNANCHORED, &consumed, args, n, vec, kVecSize);
+}
+
+bool RE::Consume(StringPiece* input,
+                 const Arg& ptr1,
+                 const Arg& ptr2,
+                 const Arg& ptr3,
+                 const Arg& ptr4,
+                 const Arg& ptr5,
+                 const Arg& ptr6,
+                 const Arg& ptr7,
+                 const Arg& ptr8,
+                 const Arg& ptr9,
+                 const Arg& ptr10,
+                 const Arg& ptr11,
+                 const Arg& ptr12,
+                 const Arg& ptr13,
+                 const Arg& ptr14,
+                 const Arg& ptr15,
+                 const Arg& ptr16) const {
+  const Arg* args[kMaxArgs];
+  int n = 0;
+  if (&ptr1  == &no_arg) goto done; args[n++] = &ptr1;
+  if (&ptr2  == &no_arg) goto done; args[n++] = &ptr2;
+  if (&ptr3  == &no_arg) goto done; args[n++] = &ptr3;
+  if (&ptr4  == &no_arg) goto done; args[n++] = &ptr4;
+  if (&ptr5  == &no_arg) goto done; args[n++] = &ptr5;
+  if (&ptr6  == &no_arg) goto done; args[n++] = &ptr6;
+  if (&ptr7  == &no_arg) goto done; args[n++] = &ptr7;
+  if (&ptr8  == &no_arg) goto done; args[n++] = &ptr8;
+  if (&ptr9  == &no_arg) goto done; args[n++] = &ptr9;
+  if (&ptr10 == &no_arg) goto done; args[n++] = &ptr10;
+  if (&ptr11 == &no_arg) goto done; args[n++] = &ptr11;
+  if (&ptr12 == &no_arg) goto done; args[n++] = &ptr12;
+  if (&ptr13 == &no_arg) goto done; args[n++] = &ptr13;
+  if (&ptr14 == &no_arg) goto done; args[n++] = &ptr14;
+  if (&ptr15 == &no_arg) goto done; args[n++] = &ptr15;
+  if (&ptr16 == &no_arg) goto done; args[n++] = &ptr16;
+ done:
+
+  int consumed;
+  int vec[kVecSize];
+  if (DoMatchImpl(*input, ANCHOR_START, &consumed,
+                  args, n, vec, kVecSize)) {
+    input->remove_prefix(consumed);
+    return true;
+  } else {
+    return false;
+  }
+}
+
+bool RE::FindAndConsume(StringPiece* input,
+                        const Arg& ptr1,
+                        const Arg& ptr2,
+                        const Arg& ptr3,
+                        const Arg& ptr4,
+                        const Arg& ptr5,
+                        const Arg& ptr6,
+                        const Arg& ptr7,
+                        const Arg& ptr8,
+                        const Arg& ptr9,
+                        const Arg& ptr10,
+                        const Arg& ptr11,
+                        const Arg& ptr12,
+                        const Arg& ptr13,
+                        const Arg& ptr14,
+                        const Arg& ptr15,
+                        const Arg& ptr16) const {
+  const Arg* args[kMaxArgs];
+  int n = 0;
+  if (&ptr1  == &no_arg) goto done; args[n++] = &ptr1;
+  if (&ptr2  == &no_arg) goto done; args[n++] = &ptr2;
+  if (&ptr3  == &no_arg) goto done; args[n++] = &ptr3;
+  if (&ptr4  == &no_arg) goto done; args[n++] = &ptr4;
+  if (&ptr5  == &no_arg) goto done; args[n++] = &ptr5;
+  if (&ptr6  == &no_arg) goto done; args[n++] = &ptr6;
+  if (&ptr7  == &no_arg) goto done; args[n++] = &ptr7;
+  if (&ptr8  == &no_arg) goto done; args[n++] = &ptr8;
+  if (&ptr9  == &no_arg) goto done; args[n++] = &ptr9;
+  if (&ptr10 == &no_arg) goto done; args[n++] = &ptr10;
+  if (&ptr11 == &no_arg) goto done; args[n++] = &ptr11;
+  if (&ptr12 == &no_arg) goto done; args[n++] = &ptr12;
+  if (&ptr13 == &no_arg) goto done; args[n++] = &ptr13;
+  if (&ptr14 == &no_arg) goto done; args[n++] = &ptr14;
+  if (&ptr15 == &no_arg) goto done; args[n++] = &ptr15;
+  if (&ptr16 == &no_arg) goto done; args[n++] = &ptr16;
+ done:
+
+  int consumed;
+  int vec[kVecSize];
+  if (DoMatchImpl(*input, UNANCHORED, &consumed,
+                  args, n, vec, kVecSize)) {
+    input->remove_prefix(consumed);
+    return true;
+  } else {
+    return false;
+  }
+}
+
+bool RE::Replace(const StringPiece& rewrite,
+                 string *str) const {
+  int vec[kVecSize];
+  int matches = TryMatch(*str, 0, UNANCHORED, true, vec, kVecSize);
+  if (matches == 0)
+    return false;
+
+  string s;
+  if (!Rewrite(&s, rewrite, *str, vec, matches))
+    return false;
+
+  assert(vec[0] >= 0);
+  assert(vec[1] >= 0);
+  str->replace(vec[0], vec[1] - vec[0], s);
+  return true;
+}
+
+// Returns PCRE_NEWLINE_CRLF, PCRE_NEWLINE_CR, or PCRE_NEWLINE_LF.
+// Note that PCRE_NEWLINE_CRLF is defined to be P_N_CR | P_N_LF.
+// Modified by PH to add PCRE_NEWLINE_ANY and PCRE_NEWLINE_ANYCRLF.
+
+static int NewlineMode(int pcre_options) {
+  // TODO: if we can make it threadsafe, cache this var
+  int newline_mode = 0;
+  /* if (newline_mode) return newline_mode; */  // do this once it's cached
+  if (pcre_options & (PCRE_NEWLINE_CRLF|PCRE_NEWLINE_CR|PCRE_NEWLINE_LF|
+                      PCRE_NEWLINE_ANY|PCRE_NEWLINE_ANYCRLF)) {
+    newline_mode = (pcre_options &
+                    (PCRE_NEWLINE_CRLF|PCRE_NEWLINE_CR|PCRE_NEWLINE_LF|
+                     PCRE_NEWLINE_ANY|PCRE_NEWLINE_ANYCRLF));
+  } else {
+    int newline;
+    pcre_config(PCRE_CONFIG_NEWLINE, &newline);
+    if (newline == 10)
+      newline_mode = PCRE_NEWLINE_LF;
+    else if (newline == 13)
+      newline_mode = PCRE_NEWLINE_CR;
+    else if (newline == 3338)
+      newline_mode = PCRE_NEWLINE_CRLF;
+    else if (newline == -1)
+      newline_mode = PCRE_NEWLINE_ANY;
+    else if (newline == -2)
+      newline_mode = PCRE_NEWLINE_ANYCRLF;
+    else
+      assert(NULL == "Unexpected return value from pcre_config(NEWLINE)");
+  }
+  return newline_mode;
+}
+
+int RE::GlobalReplace(const StringPiece& rewrite,
+                      string *str) const {
+  int count = 0;
+  int vec[kVecSize];
+  string out;
+  int start = 0;
+  bool last_match_was_empty_string = false;
+
+  while (start <= static_cast(str->length())) {
+    // If the previous match was for the empty string, we shouldn't
+    // just match again: we'll match in the same way and get an
+    // infinite loop.  Instead, we do the match in a special way:
+    // anchored -- to force another try at the same position --
+    // and with a flag saying that this time, ignore empty matches.
+    // If this special match returns, that means there's a non-empty
+    // match at this position as well, and we can continue.  If not,
+    // we do what perl does, and just advance by one.
+    // Notice that perl prints '@@@' for this;
+    //    perl -le '$_ = "aa"; s/b*|aa/@/g; print'
+    int matches;
+    if (last_match_was_empty_string) {
+      matches = TryMatch(*str, start, ANCHOR_START, false, vec, kVecSize);
+      if (matches <= 0) {
+        int matchend = start + 1;     // advance one character.
+        // If the current char is CR and we're in CRLF mode, skip LF too.
+        // Note it's better to call pcre_fullinfo() than to examine
+        // all_options(), since options_ could have changed bewteen
+        // compile-time and now, but this is simpler and safe enough.
+        // Modified by PH to add ANY and ANYCRLF.
+        if (matchend < static_cast(str->length()) &&
+            (*str)[start] == '\r' && (*str)[matchend] == '\n' &&
+            (NewlineMode(options_.all_options()) == PCRE_NEWLINE_CRLF ||
+             NewlineMode(options_.all_options()) == PCRE_NEWLINE_ANY ||
+             NewlineMode(options_.all_options()) == PCRE_NEWLINE_ANYCRLF)) {
+          matchend++;
+        }
+        // We also need to advance more than one char if we're in utf8 mode.
+#ifdef SUPPORT_UTF8
+        if (options_.utf8()) {
+          while (matchend < static_cast(str->length()) &&
+                 ((*str)[matchend] & 0xc0) == 0x80)
+            matchend++;
+        }
+#endif
+        if (start < static_cast(str->length()))
+          out.append(*str, start, matchend - start);
+        start = matchend;
+        last_match_was_empty_string = false;
+        continue;
+      }
+    } else {
+      matches = TryMatch(*str, start, UNANCHORED, true, vec, kVecSize);
+      if (matches <= 0)
+        break;
+    }
+    int matchstart = vec[0], matchend = vec[1];
+    assert(matchstart >= start);
+    assert(matchend >= matchstart);
+    out.append(*str, start, matchstart - start);
+    Rewrite(&out, rewrite, *str, vec, matches);
+    start = matchend;
+    count++;
+    last_match_was_empty_string = (matchstart == matchend);
+  }
+
+  if (count == 0)
+    return 0;
+
+  if (start < static_cast(str->length()))
+    out.append(*str, start, str->length() - start);
+  swap(out, *str);
+  return count;
+}
+
+bool RE::Extract(const StringPiece& rewrite,
+                 const StringPiece& text,
+                 string *out) const {
+  int vec[kVecSize];
+  int matches = TryMatch(text, 0, UNANCHORED, true, vec, kVecSize);
+  if (matches == 0)
+    return false;
+  out->erase();
+  return Rewrite(out, rewrite, text, vec, matches);
+}
+
+/*static*/ string RE::QuoteMeta(const StringPiece& unquoted) {
+  string result;
+
+  // Escape any ascii character not in [A-Za-z_0-9].
+  //
+  // Note that it's legal to escape a character even if it has no
+  // special meaning in a regular expression -- so this function does
+  // that.  (This also makes it identical to the perl function of the
+  // same name; see `perldoc -f quotemeta`.)  The one exception is
+  // escaping NUL: rather than doing backslash + NUL, like perl does,
+  // we do '\0', because pcre itself doesn't take embedded NUL chars.
+  for (int ii = 0; ii < unquoted.size(); ++ii) {
+    // Note that using 'isalnum' here raises the benchmark time from
+    // 32ns to 58ns:
+    if (unquoted[ii] == '\0') {
+      result += "\\0";
+    } else if ((unquoted[ii] < 'a' || unquoted[ii] > 'z') &&
+               (unquoted[ii] < 'A' || unquoted[ii] > 'Z') &&
+               (unquoted[ii] < '0' || unquoted[ii] > '9') &&
+               unquoted[ii] != '_' &&
+               // If this is the part of a UTF8 or Latin1 character, we need
+               // to copy this byte without escaping.  Experimentally this is
+               // what works correctly with the regexp library.
+               !(unquoted[ii] & 128)) {
+      result += '\\';
+      result += unquoted[ii];
+    } else {
+      result += unquoted[ii];
+    }
+  }
+
+  return result;
+}
+
+/***** Actual matching and rewriting code *****/
+
+int RE::TryMatch(const StringPiece& text,
+                 int startpos,
+                 Anchor anchor,
+                 bool empty_ok,
+                 int *vec,
+                 int vecsize) const {
+  pcre* re = (anchor == ANCHOR_BOTH) ? re_full_ : re_partial_;
+  if (re == NULL) {
+    //fprintf(stderr, "Matching against invalid re: %s\n", error_->c_str());
+    return 0;
+  }
+
+  pcre_extra extra = { 0, 0, 0, 0, 0, 0 };
+  if (options_.match_limit() > 0) {
+    extra.flags |= PCRE_EXTRA_MATCH_LIMIT;
+    extra.match_limit = options_.match_limit();
+  }
+  if (options_.match_limit_recursion() > 0) {
+    extra.flags |= PCRE_EXTRA_MATCH_LIMIT_RECURSION;
+    extra.match_limit_recursion = options_.match_limit_recursion();
+  }
+
+  // int options = 0;
+  // Changed by PH as a result of bugzilla #1288
+  int options = (options_.all_options() & PCRE_NO_UTF8_CHECK);
+
+  if (anchor != UNANCHORED)
+    options |= PCRE_ANCHORED;
+  if (!empty_ok)
+    options |= PCRE_NOTEMPTY;
+
+  int rc = pcre_exec(re,              // The regular expression object
+                     &extra,
+                     (text.data() == NULL) ? "" : text.data(),
+                     text.size(),
+                     startpos,
+                     options,
+                     vec,
+                     vecsize);
+
+  // Handle errors
+  if (rc == PCRE_ERROR_NOMATCH) {
+    return 0;
+  } else if (rc < 0) {
+    //fprintf(stderr, "Unexpected return code: %d when matching '%s'\n",
+    //        re, pattern_.c_str());
+    return 0;
+  } else if (rc == 0) {
+    // pcre_exec() returns 0 as a special case when the number of
+    // capturing subpatterns exceeds the size of the vector.
+    // When this happens, there is a match and the output vector
+    // is filled, but we miss out on the positions of the extra subpatterns.
+    rc = vecsize / 2;
+  }
+
+  return rc;
+}
+
+bool RE::DoMatchImpl(const StringPiece& text,
+                     Anchor anchor,
+                     int* consumed,
+                     const Arg* const* args,
+                     int n,
+                     int* vec,
+                     int vecsize) const {
+  assert((1 + n) * 3 <= vecsize);  // results + PCRE workspace
+  int matches = TryMatch(text, 0, anchor, true, vec, vecsize);
+  assert(matches >= 0);  // TryMatch never returns negatives
+  if (matches == 0)
+    return false;
+
+  *consumed = vec[1];
+
+  if (n == 0 || args == NULL) {
+    // We are not interested in results
+    return true;
+  }
+
+  if (NumberOfCapturingGroups() < n) {
+    // RE has fewer capturing groups than number of arg pointers passed in
+    return false;
+  }
+
+  // If we got here, we must have matched the whole pattern.
+  // We do not need (can not do) any more checks on the value of 'matches' here
+  // -- see the comment for TryMatch.
+  for (int i = 0; i < n; i++) {
+    const int start = vec[2*(i+1)];
+    const int limit = vec[2*(i+1)+1];
+    if (!args[i]->Parse(text.data() + start, limit-start)) {
+      // TODO: Should we indicate what the error was?
+      return false;
+    }
+  }
+
+  return true;
+}
+
+bool RE::DoMatch(const StringPiece& text,
+                 Anchor anchor,
+                 int* consumed,
+                 const Arg* const args[],
+                 int n) const {
+  assert(n >= 0);
+  size_t const vecsize = (1 + n) * 3;  // results + PCRE workspace
+                                       // (as for kVecSize)
+  int space[21];   // use stack allocation for small vecsize (common case)
+  int* vec = vecsize <= 21 ? space : new int[vecsize];
+  bool retval = DoMatchImpl(text, anchor, consumed, args, n, vec, (int)vecsize);
+  if (vec != space) delete [] vec;
+  return retval;
+}
+
+bool RE::Rewrite(string *out, const StringPiece &rewrite,
+                 const StringPiece &text, int *vec, int veclen) const {
+  for (const char *s = rewrite.data(), *end = s + rewrite.size();
+       s < end; s++) {
+    int c = *s;
+    if (c == '\\') {
+      c = *++s;
+      if (isdigit(c)) {
+        int n = (c - '0');
+        if (n >= veclen) {
+          //fprintf(stderr, requested group %d in regexp %.*s\n",
+          //        n, rewrite.size(), rewrite.data());
+          return false;
+        }
+        int start = vec[2 * n];
+        if (start >= 0)
+          out->append(text.data() + start, vec[2 * n + 1] - start);
+      } else if (c == '\\') {
+        *out += '\\';
+      } else {
+        //fprintf(stderr, "invalid rewrite pattern: %.*s\n",
+        //        rewrite.size(), rewrite.data());
+        return false;
+      }
+    } else {
+      *out += c;
+    }
+  }
+  return true;
+}
+
+// Return the number of capturing subpatterns, or -1 if the
+// regexp wasn't valid on construction.
+int RE::NumberOfCapturingGroups() const {
+  if (re_partial_ == NULL) return -1;
+
+  int result;
+  int pcre_retval = pcre_fullinfo(re_partial_,  // The regular expression object
+                                  NULL,         // We did not study the pattern
+                                  PCRE_INFO_CAPTURECOUNT,
+                                  &result);
+  assert(pcre_retval == 0);
+  return result;
+}
+
+/***** Parsers for various types *****/
+
+bool Arg::parse_null(const char* str, int n, void* dest) {
+  // We fail if somebody asked us to store into a non-NULL void* pointer
+  return (dest == NULL);
+}
+
+bool Arg::parse_string(const char* str, int n, void* dest) {
+  if (dest == NULL) return true;
+  reinterpret_cast(dest)->assign(str, n);
+  return true;
+}
+
+bool Arg::parse_stringpiece(const char* str, int n, void* dest) {
+  if (dest == NULL) return true;
+  reinterpret_cast(dest)->set(str, n);
+  return true;
+}
+
+bool Arg::parse_char(const char* str, int n, void* dest) {
+  if (n != 1) return false;
+  if (dest == NULL) return true;
+  *(reinterpret_cast(dest)) = str[0];
+  return true;
+}
+
+bool Arg::parse_uchar(const char* str, int n, void* dest) {
+  if (n != 1) return false;
+  if (dest == NULL) return true;
+  *(reinterpret_cast(dest)) = str[0];
+  return true;
+}
+
+// Largest number spec that we are willing to parse
+static const int kMaxNumberLength = 32;
+
+// REQUIRES "buf" must have length at least kMaxNumberLength+1
+// REQUIRES "n > 0"
+// Copies "str" into "buf" and null-terminates if necessary.
+// Returns one of:
+//      a. "str" if no termination is needed
+//      b. "buf" if the string was copied and null-terminated
+//      c. "" if the input was invalid and has no hope of being parsed
+static const char* TerminateNumber(char* buf, const char* str, int n) {
+  if ((n > 0) && isspace(*str)) {
+    // We are less forgiving than the strtoxxx() routines and do not
+    // allow leading spaces.
+    return "";
+  }
+
+  // See if the character right after the input text may potentially
+  // look like a digit.
+  if (isdigit(str[n]) ||
+      ((str[n] >= 'a') && (str[n] <= 'f')) ||
+      ((str[n] >= 'A') && (str[n] <= 'F'))) {
+    if (n > kMaxNumberLength) return ""; // Input too big to be a valid number
+    memcpy(buf, str, n);
+    buf[n] = '\0';
+    return buf;
+  } else {
+    // We can parse right out of the supplied string, so return it.
+    return str;
+  }
+}
+
+bool Arg::parse_long_radix(const char* str,
+                           int n,
+                           void* dest,
+                           int radix) {
+  if (n == 0) return false;
+  char buf[kMaxNumberLength+1];
+  str = TerminateNumber(buf, str, n);
+  char* end;
+  errno = 0;
+  long r = strtol(str, &end, radix);
+  if (end != str + n) return false;   // Leftover junk
+  if (errno) return false;
+  if (dest == NULL) return true;
+  *(reinterpret_cast(dest)) = r;
+  return true;
+}
+
+bool Arg::parse_ulong_radix(const char* str,
+                            int n,
+                            void* dest,
+                            int radix) {
+  if (n == 0) return false;
+  char buf[kMaxNumberLength+1];
+  str = TerminateNumber(buf, str, n);
+  if (str[0] == '-') return false;    // strtoul() on a negative number?!
+  char* end;
+  errno = 0;
+  unsigned long r = strtoul(str, &end, radix);
+  if (end != str + n) return false;   // Leftover junk
+  if (errno) return false;
+  if (dest == NULL) return true;
+  *(reinterpret_cast(dest)) = r;
+  return true;
+}
+
+bool Arg::parse_short_radix(const char* str,
+                            int n,
+                            void* dest,
+                            int radix) {
+  long r;
+  if (!parse_long_radix(str, n, &r, radix)) return false; // Could not parse
+  if (r < SHRT_MIN || r > SHRT_MAX) return false;       // Out of range
+  if (dest == NULL) return true;
+  *(reinterpret_cast(dest)) = static_cast(r);
+  return true;
+}
+
+bool Arg::parse_ushort_radix(const char* str,
+                             int n,
+                             void* dest,
+                             int radix) {
+  unsigned long r;
+  if (!parse_ulong_radix(str, n, &r, radix)) return false; // Could not parse
+  if (r > USHRT_MAX) return false;                      // Out of range
+  if (dest == NULL) return true;
+  *(reinterpret_cast(dest)) = static_cast(r);
+  return true;
+}
+
+bool Arg::parse_int_radix(const char* str,
+                          int n,
+                          void* dest,
+                          int radix) {
+  long r;
+  if (!parse_long_radix(str, n, &r, radix)) return false; // Could not parse
+  if (r < INT_MIN || r > INT_MAX) return false;         // Out of range
+  if (dest == NULL) return true;
+  *(reinterpret_cast(dest)) = r;
+  return true;
+}
+
+bool Arg::parse_uint_radix(const char* str,
+                           int n,
+                           void* dest,
+                           int radix) {
+  unsigned long r;
+  if (!parse_ulong_radix(str, n, &r, radix)) return false; // Could not parse
+  if (r > UINT_MAX) return false;                       // Out of range
+  if (dest == NULL) return true;
+  *(reinterpret_cast(dest)) = r;
+  return true;
+}
+
+bool Arg::parse_longlong_radix(const char* str,
+                               int n,
+                               void* dest,
+                               int radix) {
+#ifndef HAVE_LONG_LONG
+  return false;
+#else
+  if (n == 0) return false;
+  char buf[kMaxNumberLength+1];
+  str = TerminateNumber(buf, str, n);
+  char* end;
+  errno = 0;
+#if defined HAVE_STRTOQ
+  long long r = strtoq(str, &end, radix);
+#elif defined HAVE_STRTOLL
+  long long r = strtoll(str, &end, radix);
+#elif defined HAVE__STRTOI64
+  long long r = _strtoi64(str, &end, radix);
+#elif defined HAVE_STRTOIMAX
+  long long r = strtoimax(str, &end, radix);
+#else
+#error parse_longlong_radix: cannot convert input to a long-long
+#endif
+  if (end != str + n) return false;   // Leftover junk
+  if (errno) return false;
+  if (dest == NULL) return true;
+  *(reinterpret_cast(dest)) = r;
+  return true;
+#endif   /* HAVE_LONG_LONG */
+}
+
+bool Arg::parse_ulonglong_radix(const char* str,
+                                int n,
+                                void* dest,
+                                int radix) {
+#ifndef HAVE_UNSIGNED_LONG_LONG
+  return false;
+#else
+  if (n == 0) return false;
+  char buf[kMaxNumberLength+1];
+  str = TerminateNumber(buf, str, n);
+  if (str[0] == '-') return false;    // strtoull() on a negative number?!
+  char* end;
+  errno = 0;
+#if defined HAVE_STRTOQ
+  unsigned long long r = strtouq(str, &end, radix);
+#elif defined HAVE_STRTOLL
+  unsigned long long r = strtoull(str, &end, radix);
+#elif defined HAVE__STRTOI64
+  unsigned long long r = _strtoui64(str, &end, radix);
+#elif defined HAVE_STRTOIMAX
+  unsigned long long r = strtoumax(str, &end, radix);
+#else
+#error parse_ulonglong_radix: cannot convert input to a long-long
+#endif
+  if (end != str + n) return false;   // Leftover junk
+  if (errno) return false;
+  if (dest == NULL) return true;
+  *(reinterpret_cast(dest)) = r;
+  return true;
+#endif   /* HAVE_UNSIGNED_LONG_LONG */
+}
+
+bool Arg::parse_double(const char* str, int n, void* dest) {
+  if (n == 0) return false;
+  static const int kMaxLength = 200;
+  char buf[kMaxLength];
+  if (n >= kMaxLength) return false;
+  memcpy(buf, str, n);
+  buf[n] = '\0';
+  errno = 0;
+  char* end;
+  double r = strtod(buf, &end);
+  if (end != buf + n) return false;   // Leftover junk
+  if (errno) return false;
+  if (dest == NULL) return true;
+  *(reinterpret_cast(dest)) = r;
+  return true;
+}
+
+bool Arg::parse_float(const char* str, int n, void* dest) {
+  double r;
+  if (!parse_double(str, n, &r)) return false;
+  if (dest == NULL) return true;
+  *(reinterpret_cast(dest)) = static_cast(r);
+  return true;
+}
+
+
+#define DEFINE_INTEGER_PARSERS(name)                                    \
+  bool Arg::parse_##name(const char* str, int n, void* dest) {          \
+    return parse_##name##_radix(str, n, dest, 10);                      \
+  }                                                                     \
+  bool Arg::parse_##name##_hex(const char* str, int n, void* dest) {    \
+    return parse_##name##_radix(str, n, dest, 16);                      \
+  }                                                                     \
+  bool Arg::parse_##name##_octal(const char* str, int n, void* dest) {  \
+    return parse_##name##_radix(str, n, dest, 8);                       \
+  }                                                                     \
+  bool Arg::parse_##name##_cradix(const char* str, int n, void* dest) { \
+    return parse_##name##_radix(str, n, dest, 0);                       \
+  }
+
+DEFINE_INTEGER_PARSERS(short)      /*                                   */
+DEFINE_INTEGER_PARSERS(ushort)     /*                                   */
+DEFINE_INTEGER_PARSERS(int)        /* Don't use semicolons after these  */
+DEFINE_INTEGER_PARSERS(uint)       /* statements because they can cause */
+DEFINE_INTEGER_PARSERS(long)       /* compiler warnings if the checking */
+DEFINE_INTEGER_PARSERS(ulong)      /* level is turned up high enough.   */
+DEFINE_INTEGER_PARSERS(longlong)   /*                                   */
+DEFINE_INTEGER_PARSERS(ulonglong)  /*                                   */
+
+#undef DEFINE_INTEGER_PARSERS
+
+}   // namespace pcrecpp
diff --git a/pcre/pcrecpp.h b/pcre/pcrecpp.h
new file mode 100644
index 00000000000..3e594b0d439
--- /dev/null
+++ b/pcre/pcrecpp.h
@@ -0,0 +1,710 @@
+// Copyright (c) 2005, Google Inc.
+// All rights reserved.
+//
+// Redistribution and use in source and binary forms, with or without
+// modification, are permitted provided that the following conditions are
+// met:
+//
+//     * Redistributions of source code must retain the above copyright
+// notice, this list of conditions and the following disclaimer.
+//     * Redistributions in binary form must reproduce the above
+// copyright notice, this list of conditions and the following disclaimer
+// in the documentation and/or other materials provided with the
+// distribution.
+//     * Neither the name of Google Inc. nor the names of its
+// contributors may be used to endorse or promote products derived from
+// this software without specific prior written permission.
+//
+// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+//
+// Author: Sanjay Ghemawat
+// Support for PCRE_XXX modifiers added by Giuseppe Maxia, July 2005
+
+#ifndef _PCRECPP_H
+#define _PCRECPP_H
+
+// C++ interface to the pcre regular-expression library.  RE supports
+// Perl-style regular expressions (with extensions like \d, \w, \s,
+// ...).
+//
+// -----------------------------------------------------------------------
+// REGEXP SYNTAX:
+//
+// This module is part of the pcre library and hence supports its syntax
+// for regular expressions.
+//
+// The syntax is pretty similar to Perl's.  For those not familiar
+// with Perl's regular expressions, here are some examples of the most
+// commonly used extensions:
+//
+//   "hello (\\w+) world"  -- \w matches a "word" character
+//   "version (\\d+)"      -- \d matches a digit
+//   "hello\\s+world"      -- \s matches any whitespace character
+//   "\\b(\\w+)\\b"        -- \b matches empty string at a word boundary
+//   "(?i)hello"           -- (?i) turns on case-insensitive matching
+//   "/\\*(.*?)\\*/"       -- .*? matches . minimum no. of times possible
+//
+// -----------------------------------------------------------------------
+// MATCHING INTERFACE:
+//
+// The "FullMatch" operation checks that supplied text matches a
+// supplied pattern exactly.
+//
+// Example: successful match
+//    pcrecpp::RE re("h.*o");
+//    re.FullMatch("hello");
+//
+// Example: unsuccessful match (requires full match):
+//    pcrecpp::RE re("e");
+//    !re.FullMatch("hello");
+//
+// Example: creating a temporary RE object:
+//    pcrecpp::RE("h.*o").FullMatch("hello");
+//
+// You can pass in a "const char*" or a "string" for "text".  The
+// examples below tend to use a const char*.
+//
+// You can, as in the different examples above, store the RE object
+// explicitly in a variable or use a temporary RE object.  The
+// examples below use one mode or the other arbitrarily.  Either
+// could correctly be used for any of these examples.
+//
+// -----------------------------------------------------------------------
+// MATCHING WITH SUB-STRING EXTRACTION:
+//
+// You can supply extra pointer arguments to extract matched subpieces.
+//
+// Example: extracts "ruby" into "s" and 1234 into "i"
+//    int i;
+//    string s;
+//    pcrecpp::RE re("(\\w+):(\\d+)");
+//    re.FullMatch("ruby:1234", &s, &i);
+//
+// Example: does not try to extract any extra sub-patterns
+//    re.FullMatch("ruby:1234", &s);
+//
+// Example: does not try to extract into NULL
+//    re.FullMatch("ruby:1234", NULL, &i);
+//
+// Example: integer overflow causes failure
+//    !re.FullMatch("ruby:1234567891234", NULL, &i);
+//
+// Example: fails because there aren't enough sub-patterns:
+//    !pcrecpp::RE("\\w+:\\d+").FullMatch("ruby:1234", &s);
+//
+// Example: fails because string cannot be stored in integer
+//    !pcrecpp::RE("(.*)").FullMatch("ruby", &i);
+//
+// The provided pointer arguments can be pointers to any scalar numeric
+// type, or one of
+//    string        (matched piece is copied to string)
+//    StringPiece   (StringPiece is mutated to point to matched piece)
+//    T             (where "bool T::ParseFrom(const char*, int)" exists)
+//    NULL          (the corresponding matched sub-pattern is not copied)
+//
+// CAVEAT: An optional sub-pattern that does not exist in the matched
+// string is assigned the empty string.  Therefore, the following will
+// return false (because the empty string is not a valid number):
+//    int number;
+//    pcrecpp::RE::FullMatch("abc", "[a-z]+(\\d+)?", &number);
+//
+// -----------------------------------------------------------------------
+// DO_MATCH
+//
+// The matching interface supports at most 16 arguments per call.
+// If you need more, consider using the more general interface
+// pcrecpp::RE::DoMatch().  See pcrecpp.h for the signature for DoMatch.
+//
+// -----------------------------------------------------------------------
+// PARTIAL MATCHES
+//
+// You can use the "PartialMatch" operation when you want the pattern
+// to match any substring of the text.
+//
+// Example: simple search for a string:
+//    pcrecpp::RE("ell").PartialMatch("hello");
+//
+// Example: find first number in a string:
+//    int number;
+//    pcrecpp::RE re("(\\d+)");
+//    re.PartialMatch("x*100 + 20", &number);
+//    assert(number == 100);
+//
+// -----------------------------------------------------------------------
+// UTF-8 AND THE MATCHING INTERFACE:
+//
+// By default, pattern and text are plain text, one byte per character.
+// The UTF8 flag, passed to the constructor, causes both pattern
+// and string to be treated as UTF-8 text, still a byte stream but
+// potentially multiple bytes per character. In practice, the text
+// is likelier to be UTF-8 than the pattern, but the match returned
+// may depend on the UTF8 flag, so always use it when matching
+// UTF8 text.  E.g., "." will match one byte normally but with UTF8
+// set may match up to three bytes of a multi-byte character.
+//
+// Example:
+//    pcrecpp::RE_Options options;
+//    options.set_utf8();
+//    pcrecpp::RE re(utf8_pattern, options);
+//    re.FullMatch(utf8_string);
+//
+// Example: using the convenience function UTF8():
+//    pcrecpp::RE re(utf8_pattern, pcrecpp::UTF8());
+//    re.FullMatch(utf8_string);
+//
+// NOTE: The UTF8 option is ignored if pcre was not configured with the
+//       --enable-utf8 flag.
+//
+// -----------------------------------------------------------------------
+// PASSING MODIFIERS TO THE REGULAR EXPRESSION ENGINE
+//
+// PCRE defines some modifiers to change the behavior of the regular
+// expression engine.
+// The C++ wrapper defines an auxiliary class, RE_Options, as a vehicle
+// to pass such modifiers to a RE class.
+//
+// Currently, the following modifiers are supported
+//
+//    modifier              description               Perl corresponding
+//
+//    PCRE_CASELESS         case insensitive match    /i
+//    PCRE_MULTILINE        multiple lines match      /m
+//    PCRE_DOTALL           dot matches newlines      /s
+//    PCRE_DOLLAR_ENDONLY   $ matches only at end     N/A
+//    PCRE_EXTRA            strict escape parsing     N/A
+//    PCRE_EXTENDED         ignore whitespaces        /x
+//    PCRE_UTF8             handles UTF8 chars        built-in
+//    PCRE_UNGREEDY         reverses * and *?         N/A
+//    PCRE_NO_AUTO_CAPTURE  disables matching parens  N/A (*)
+//
+// (For a full account on how each modifier works, please check the
+// PCRE API reference manual).
+//
+// (*) Both Perl and PCRE allow non matching parentheses by means of the
+// "?:" modifier within the pattern itself. e.g. (?:ab|cd) does not
+// capture, while (ab|cd) does.
+//
+// For each modifier, there are two member functions whose name is made
+// out of the modifier in lowercase, without the "PCRE_" prefix. For
+// instance, PCRE_CASELESS is handled by
+//    bool caseless(),
+// which returns true if the modifier is set, and
+//    RE_Options & set_caseless(bool),
+// which sets or unsets the modifier.
+//
+// Moreover, PCRE_EXTRA_MATCH_LIMIT can be accessed through the
+// set_match_limit() and match_limit() member functions.
+// Setting match_limit to a non-zero value will limit the executation of
+// pcre to keep it from doing bad things like blowing the stack or taking
+// an eternity to return a result.  A value of 5000 is good enough to stop
+// stack blowup in a 2MB thread stack.  Setting match_limit to zero will
+// disable match limiting.  Alternately, you can set match_limit_recursion()
+// which uses PCRE_EXTRA_MATCH_LIMIT_RECURSION to limit how much pcre
+// recurses.  match_limit() caps the number of matches pcre does;
+// match_limit_recrusion() caps the depth of recursion.
+//
+// Normally, to pass one or more modifiers to a RE class, you declare
+// a RE_Options object, set the appropriate options, and pass this
+// object to a RE constructor. Example:
+//
+//    RE_options opt;
+//    opt.set_caseless(true);
+//
+//    if (RE("HELLO", opt).PartialMatch("hello world")) ...
+//
+// RE_options has two constructors. The default constructor takes no
+// arguments and creates a set of flags that are off by default.
+//
+// The optional parameter 'option_flags' is to facilitate transfer
+// of legacy code from C programs.  This lets you do
+//    RE(pattern, RE_Options(PCRE_CASELESS|PCRE_MULTILINE)).PartialMatch(str);
+//
+// But new code is better off doing
+//    RE(pattern,
+//      RE_Options().set_caseless(true).set_multiline(true)).PartialMatch(str);
+// (See below)
+//
+// If you are going to pass one of the most used modifiers, there are some
+// convenience functions that return a RE_Options class with the
+// appropriate modifier already set:
+// CASELESS(), UTF8(), MULTILINE(), DOTALL(), EXTENDED()
+//
+// If you need to set several options at once, and you don't want to go
+// through the pains of declaring a RE_Options object and setting several
+// options, there is a parallel method that give you such ability on the
+// fly. You can concatenate several set_xxxxx member functions, since each
+// of them returns a reference to its class object.  e.g.: to pass
+// PCRE_CASELESS, PCRE_EXTENDED, and PCRE_MULTILINE to a RE with one
+// statement, you may write
+//
+//    RE(" ^ xyz \\s+ .* blah$", RE_Options()
+//                            .set_caseless(true)
+//                            .set_extended(true)
+//                            .set_multiline(true)).PartialMatch(sometext);
+//
+// -----------------------------------------------------------------------
+// SCANNING TEXT INCREMENTALLY
+//
+// The "Consume" operation may be useful if you want to repeatedly
+// match regular expressions at the front of a string and skip over
+// them as they match.  This requires use of the "StringPiece" type,
+// which represents a sub-range of a real string.  Like RE, StringPiece
+// is defined in the pcrecpp namespace.
+//
+// Example: read lines of the form "var = value" from a string.
+//    string contents = ...;                 // Fill string somehow
+//    pcrecpp::StringPiece input(contents);  // Wrap in a StringPiece
+//
+//    string var;
+//    int value;
+//    pcrecpp::RE re("(\\w+) = (\\d+)\n");
+//    while (re.Consume(&input, &var, &value)) {
+//      ...;
+//    }
+//
+// Each successful call to "Consume" will set "var/value", and also
+// advance "input" so it points past the matched text.
+//
+// The "FindAndConsume" operation is similar to "Consume" but does not
+// anchor your match at the beginning of the string.  For example, you
+// could extract all words from a string by repeatedly calling
+//     pcrecpp::RE("(\\w+)").FindAndConsume(&input, &word)
+//
+// -----------------------------------------------------------------------
+// PARSING HEX/OCTAL/C-RADIX NUMBERS
+//
+// By default, if you pass a pointer to a numeric value, the
+// corresponding text is interpreted as a base-10 number.  You can
+// instead wrap the pointer with a call to one of the operators Hex(),
+// Octal(), or CRadix() to interpret the text in another base.  The
+// CRadix operator interprets C-style "0" (base-8) and "0x" (base-16)
+// prefixes, but defaults to base-10.
+//
+// Example:
+//   int a, b, c, d;
+//   pcrecpp::RE re("(.*) (.*) (.*) (.*)");
+//   re.FullMatch("100 40 0100 0x40",
+//                pcrecpp::Octal(&a), pcrecpp::Hex(&b),
+//                pcrecpp::CRadix(&c), pcrecpp::CRadix(&d));
+// will leave 64 in a, b, c, and d.
+//
+// -----------------------------------------------------------------------
+// REPLACING PARTS OF STRINGS
+//
+// You can replace the first match of "pattern" in "str" with
+// "rewrite".  Within "rewrite", backslash-escaped digits (\1 to \9)
+// can be used to insert text matching corresponding parenthesized
+// group from the pattern.  \0 in "rewrite" refers to the entire
+// matching text.  E.g.,
+//
+//   string s = "yabba dabba doo";
+//   pcrecpp::RE("b+").Replace("d", &s);
+//
+// will leave "s" containing "yada dabba doo".  The result is true if
+// the pattern matches and a replacement occurs, or false otherwise.
+//
+// GlobalReplace() is like Replace(), except that it replaces all
+// occurrences of the pattern in the string with the rewrite.
+// Replacements are not subject to re-matching.  E.g.,
+//
+//   string s = "yabba dabba doo";
+//   pcrecpp::RE("b+").GlobalReplace("d", &s);
+//
+// will leave "s" containing "yada dada doo".  It returns the number
+// of replacements made.
+//
+// Extract() is like Replace(), except that if the pattern matches,
+// "rewrite" is copied into "out" (an additional argument) with
+// substitutions.  The non-matching portions of "text" are ignored.
+// Returns true iff a match occurred and the extraction happened
+// successfully.  If no match occurs, the string is left unaffected.
+
+
+#include 
+#include 
+#include    // defines the Arg class
+// This isn't technically needed here, but we include it
+// anyway so folks who include pcrecpp.h don't have to.
+#include 
+
+namespace pcrecpp {
+
+#define PCRE_SET_OR_CLEAR(b, o) \
+    if (b) all_options_ |= (o); else all_options_ &= ~(o); \
+    return *this
+
+#define PCRE_IS_SET(o)  \
+        (all_options_ & o) == o
+
+/***** Compiling regular expressions: the RE class *****/
+
+// RE_Options allow you to set options to be passed along to pcre,
+// along with other options we put on top of pcre.
+// Only 9 modifiers, plus match_limit and match_limit_recursion,
+// are supported now.
+class PCRECPP_EXP_DEFN RE_Options {
+ public:
+  // constructor
+  RE_Options() : match_limit_(0), match_limit_recursion_(0), all_options_(0) {}
+
+  // alternative constructor.
+  // To facilitate transfer of legacy code from C programs
+  //
+  // This lets you do
+  //    RE(pattern, RE_Options(PCRE_CASELESS|PCRE_MULTILINE)).PartialMatch(str);
+  // But new code is better off doing
+  //    RE(pattern,
+  //      RE_Options().set_caseless(true).set_multiline(true)).PartialMatch(str);
+  RE_Options(int option_flags) : match_limit_(0), match_limit_recursion_(0),
+                                 all_options_(option_flags) {}
+  // we're fine with the default destructor, copy constructor, etc.
+
+  // accessors and mutators
+  int match_limit() const { return match_limit_; };
+  RE_Options &set_match_limit(int limit) {
+    match_limit_ = limit;
+    return *this;
+  }
+
+  int match_limit_recursion() const { return match_limit_recursion_; };
+  RE_Options &set_match_limit_recursion(int limit) {
+    match_limit_recursion_ = limit;
+    return *this;
+  }
+
+  bool caseless() const {
+    return PCRE_IS_SET(PCRE_CASELESS);
+  }
+  RE_Options &set_caseless(bool x) {
+    PCRE_SET_OR_CLEAR(x, PCRE_CASELESS);
+  }
+
+  bool multiline() const {
+    return PCRE_IS_SET(PCRE_MULTILINE);
+  }
+  RE_Options &set_multiline(bool x) {
+    PCRE_SET_OR_CLEAR(x, PCRE_MULTILINE);
+  }
+
+  bool dotall() const {
+    return PCRE_IS_SET(PCRE_DOTALL);
+  }
+  RE_Options &set_dotall(bool x) {
+    PCRE_SET_OR_CLEAR(x, PCRE_DOTALL);
+  }
+
+  bool extended() const {
+    return PCRE_IS_SET(PCRE_EXTENDED);
+  }
+  RE_Options &set_extended(bool x) {
+    PCRE_SET_OR_CLEAR(x, PCRE_EXTENDED);
+  }
+
+  bool dollar_endonly() const {
+    return PCRE_IS_SET(PCRE_DOLLAR_ENDONLY);
+  }
+  RE_Options &set_dollar_endonly(bool x) {
+    PCRE_SET_OR_CLEAR(x, PCRE_DOLLAR_ENDONLY);
+  }
+
+  bool extra() const {
+    return PCRE_IS_SET(PCRE_EXTRA);
+  }
+  RE_Options &set_extra(bool x) {
+    PCRE_SET_OR_CLEAR(x, PCRE_EXTRA);
+  }
+
+  bool ungreedy() const {
+    return PCRE_IS_SET(PCRE_UNGREEDY);
+  }
+  RE_Options &set_ungreedy(bool x) {
+    PCRE_SET_OR_CLEAR(x, PCRE_UNGREEDY);
+  }
+
+  bool utf8() const {
+    return PCRE_IS_SET(PCRE_UTF8);
+  }
+  RE_Options &set_utf8(bool x) {
+    PCRE_SET_OR_CLEAR(x, PCRE_UTF8);
+  }
+
+  bool no_auto_capture() const {
+    return PCRE_IS_SET(PCRE_NO_AUTO_CAPTURE);
+  }
+  RE_Options &set_no_auto_capture(bool x) {
+    PCRE_SET_OR_CLEAR(x, PCRE_NO_AUTO_CAPTURE);
+  }
+
+  RE_Options &set_all_options(int opt) {
+    all_options_ = opt;
+    return *this;
+  }
+  int all_options() const {
+    return all_options_ ;
+  }
+
+  // TODO: add other pcre flags
+
+ private:
+  int match_limit_;
+  int match_limit_recursion_;
+  int all_options_;
+};
+
+// These functions return some common RE_Options
+static inline RE_Options UTF8() {
+  return RE_Options().set_utf8(true);
+}
+
+static inline RE_Options CASELESS() {
+  return RE_Options().set_caseless(true);
+}
+static inline RE_Options MULTILINE() {
+  return RE_Options().set_multiline(true);
+}
+
+static inline RE_Options DOTALL() {
+  return RE_Options().set_dotall(true);
+}
+
+static inline RE_Options EXTENDED() {
+  return RE_Options().set_extended(true);
+}
+
+// Interface for regular expression matching.  Also corresponds to a
+// pre-compiled regular expression.  An "RE" object is safe for
+// concurrent use by multiple threads.
+class PCRECPP_EXP_DEFN RE {
+ public:
+  // We provide implicit conversions from strings so that users can
+  // pass in a string or a "const char*" wherever an "RE" is expected.
+  RE(const string& pat) { Init(pat, NULL); }
+  RE(const string& pat, const RE_Options& option) { Init(pat, &option); }
+  RE(const char* pat) { Init(pat, NULL); }
+  RE(const char* pat, const RE_Options& option) { Init(pat, &option); }
+  RE(const unsigned char* pat) {
+    Init(reinterpret_cast(pat), NULL);
+  }
+  RE(const unsigned char* pat, const RE_Options& option) {
+    Init(reinterpret_cast(pat), &option);
+  }
+
+  // Copy constructor & assignment - note that these are expensive
+  // because they recompile the expression.
+  RE(const RE& re) { Init(re.pattern_, &re.options_); }
+  const RE& operator=(const RE& re) {
+    if (this != &re) {
+      Cleanup();
+
+      // This is the code that originally came from Google
+      // Init(re.pattern_.c_str(), &re.options_);
+
+      // This is the replacement from Ari Pollak
+      Init(re.pattern_, &re.options_);
+    }
+    return *this;
+  }
+
+
+  ~RE();
+
+  // The string specification for this RE.  E.g.
+  //   RE re("ab*c?d+");
+  //   re.pattern();    // "ab*c?d+"
+  const string& pattern() const { return pattern_; }
+
+  // If RE could not be created properly, returns an error string.
+  // Else returns the empty string.
+  const string& error() const { return *error_; }
+
+  /***** The useful part: the matching interface *****/
+
+  // This is provided so one can do pattern.ReplaceAll() just as
+  // easily as ReplaceAll(pattern-text, ....)
+
+  bool FullMatch(const StringPiece& text,
+                 const Arg& ptr1 = no_arg,
+                 const Arg& ptr2 = no_arg,
+                 const Arg& ptr3 = no_arg,
+                 const Arg& ptr4 = no_arg,
+                 const Arg& ptr5 = no_arg,
+                 const Arg& ptr6 = no_arg,
+                 const Arg& ptr7 = no_arg,
+                 const Arg& ptr8 = no_arg,
+                 const Arg& ptr9 = no_arg,
+                 const Arg& ptr10 = no_arg,
+                 const Arg& ptr11 = no_arg,
+                 const Arg& ptr12 = no_arg,
+                 const Arg& ptr13 = no_arg,
+                 const Arg& ptr14 = no_arg,
+                 const Arg& ptr15 = no_arg,
+                 const Arg& ptr16 = no_arg) const;
+
+  bool PartialMatch(const StringPiece& text,
+                    const Arg& ptr1 = no_arg,
+                    const Arg& ptr2 = no_arg,
+                    const Arg& ptr3 = no_arg,
+                    const Arg& ptr4 = no_arg,
+                    const Arg& ptr5 = no_arg,
+                    const Arg& ptr6 = no_arg,
+                    const Arg& ptr7 = no_arg,
+                    const Arg& ptr8 = no_arg,
+                    const Arg& ptr9 = no_arg,
+                    const Arg& ptr10 = no_arg,
+                    const Arg& ptr11 = no_arg,
+                    const Arg& ptr12 = no_arg,
+                    const Arg& ptr13 = no_arg,
+                    const Arg& ptr14 = no_arg,
+                    const Arg& ptr15 = no_arg,
+                    const Arg& ptr16 = no_arg) const;
+
+  bool Consume(StringPiece* input,
+               const Arg& ptr1 = no_arg,
+               const Arg& ptr2 = no_arg,
+               const Arg& ptr3 = no_arg,
+               const Arg& ptr4 = no_arg,
+               const Arg& ptr5 = no_arg,
+               const Arg& ptr6 = no_arg,
+               const Arg& ptr7 = no_arg,
+               const Arg& ptr8 = no_arg,
+               const Arg& ptr9 = no_arg,
+               const Arg& ptr10 = no_arg,
+               const Arg& ptr11 = no_arg,
+               const Arg& ptr12 = no_arg,
+               const Arg& ptr13 = no_arg,
+               const Arg& ptr14 = no_arg,
+               const Arg& ptr15 = no_arg,
+               const Arg& ptr16 = no_arg) const;
+
+  bool FindAndConsume(StringPiece* input,
+                      const Arg& ptr1 = no_arg,
+                      const Arg& ptr2 = no_arg,
+                      const Arg& ptr3 = no_arg,
+                      const Arg& ptr4 = no_arg,
+                      const Arg& ptr5 = no_arg,
+                      const Arg& ptr6 = no_arg,
+                      const Arg& ptr7 = no_arg,
+                      const Arg& ptr8 = no_arg,
+                      const Arg& ptr9 = no_arg,
+                      const Arg& ptr10 = no_arg,
+                      const Arg& ptr11 = no_arg,
+                      const Arg& ptr12 = no_arg,
+                      const Arg& ptr13 = no_arg,
+                      const Arg& ptr14 = no_arg,
+                      const Arg& ptr15 = no_arg,
+                      const Arg& ptr16 = no_arg) const;
+
+  bool Replace(const StringPiece& rewrite,
+               string *str) const;
+
+  int GlobalReplace(const StringPiece& rewrite,
+                    string *str) const;
+
+  bool Extract(const StringPiece &rewrite,
+               const StringPiece &text,
+               string *out) const;
+
+  // Escapes all potentially meaningful regexp characters in
+  // 'unquoted'.  The returned string, used as a regular expression,
+  // will exactly match the original string.  For example,
+  //           1.5-2.0?
+  // may become:
+  //           1\.5\-2\.0\?
+  // Note QuoteMeta behaves the same as perl's QuoteMeta function,
+  // *except* that it escapes the NUL character (\0) as backslash + 0,
+  // rather than backslash + NUL.
+  static string QuoteMeta(const StringPiece& unquoted);
+
+
+  /***** Generic matching interface *****/
+
+  // Type of match (TODO: Should be restructured as part of RE_Options)
+  enum Anchor {
+    UNANCHORED,         // No anchoring
+    ANCHOR_START,       // Anchor at start only
+    ANCHOR_BOTH         // Anchor at start and end
+  };
+
+  // General matching routine.  Stores the length of the match in
+  // "*consumed" if successful.
+  bool DoMatch(const StringPiece& text,
+               Anchor anchor,
+               int* consumed,
+               const Arg* const* args, int n) const;
+
+  // Return the number of capturing subpatterns, or -1 if the
+  // regexp wasn't valid on construction.
+  int NumberOfCapturingGroups() const;
+
+  // The default value for an argument, to indicate the end of the argument
+  // list. This must be used only in optional argument defaults. It should NOT
+  // be passed explicitly. Some people have tried to use it like this:
+  //
+  //   FullMatch(x, y, &z, no_arg, &w);
+  //
+  // This is a mistake, and will not work.
+  static Arg no_arg;
+
+ private:
+
+  void Init(const string& pattern, const RE_Options* options);
+  void Cleanup();
+
+  // Match against "text", filling in "vec" (up to "vecsize" * 2/3) with
+  // pairs of integers for the beginning and end positions of matched
+  // text.  The first pair corresponds to the entire matched text;
+  // subsequent pairs correspond, in order, to parentheses-captured
+  // matches.  Returns the number of pairs (one more than the number of
+  // the last subpattern with a match) if matching was successful
+  // and zero if the match failed.
+  // I.e. for RE("(foo)|(bar)|(baz)") it will return 2, 3, and 4 when matching
+  // against "foo", "bar", and "baz" respectively.
+  // When matching RE("(foo)|hello") against "hello", it will return 1.
+  // But the values for all subpattern are filled in into "vec".
+  int TryMatch(const StringPiece& text,
+               int startpos,
+               Anchor anchor,
+               bool empty_ok,
+               int *vec,
+               int vecsize) const;
+
+  // Append the "rewrite" string, with backslash subsitutions from "text"
+  // and "vec", to string "out".
+  bool Rewrite(string *out,
+               const StringPiece& rewrite,
+               const StringPiece& text,
+               int *vec,
+               int veclen) const;
+
+  // internal implementation for DoMatch
+  bool DoMatchImpl(const StringPiece& text,
+                   Anchor anchor,
+                   int* consumed,
+                   const Arg* const args[],
+                   int n,
+                   int* vec,
+                   int vecsize) const;
+
+  // Compile the regexp for the specified anchoring mode
+  pcre* Compile(Anchor anchor);
+
+  string        pattern_;
+  RE_Options    options_;
+  pcre*         re_full_;       // For full matches
+  pcre*         re_partial_;    // For partial matches
+  const string* error_;         // Error indicator (or points to empty string)
+};
+
+}   // namespace pcrecpp
+
+#endif /* _PCRECPP_H */
diff --git a/pcre/pcrecpp_internal.h b/pcre/pcrecpp_internal.h
new file mode 100644
index 00000000000..827f9e04e2a
--- /dev/null
+++ b/pcre/pcrecpp_internal.h
@@ -0,0 +1,71 @@
+/*************************************************
+*       Perl-Compatible Regular Expressions      *
+*************************************************/
+
+/*
+Copyright (c) 2005, Google Inc.
+All rights reserved.
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+
+#ifndef PCRECPP_INTERNAL_H
+#define PCRECPP_INTERNAL_H
+
+/* When compiling a DLL for Windows, the exported symbols have to be declared
+using some MS magic. I found some useful information on this web page:
+http://msdn2.microsoft.com/en-us/library/y4h7bcy6(VS.80).aspx. According to the
+information there, using __declspec(dllexport) without "extern" we have a
+definition; with "extern" we have a declaration. The settings here override the
+setting in pcre.h. We use:
+
+  PCRECPP_EXP_DECL       for declarations
+  PCRECPP_EXP_DEFN       for definitions of exported functions
+
+*/
+
+#ifndef PCRECPP_EXP_DECL
+#  ifdef _WIN32
+#    ifndef PCRE_STATIC
+#      define PCRECPP_EXP_DECL       extern __declspec(dllexport)
+#      define PCRECPP_EXP_DEFN       __declspec(dllexport)
+#    else
+#      define PCRECPP_EXP_DECL       extern
+#      define PCRECPP_EXP_DEFN
+#    endif
+#  else
+#    define PCRECPP_EXP_DECL         extern
+#    define PCRECPP_EXP_DEFN
+#  endif
+#endif
+
+#endif  /* PCRECPP_INTERNAL_H */
+
+/* End of pcrecpp_internal.h */
diff --git a/pcre/pcrecpp_unittest.cc b/pcre/pcrecpp_unittest.cc
new file mode 100644
index 00000000000..92cae8fbea5
--- /dev/null
+++ b/pcre/pcrecpp_unittest.cc
@@ -0,0 +1,1291 @@
+// -*- coding: utf-8 -*-
+//
+// Copyright (c) 2005 - 2010, Google Inc.
+// All rights reserved.
+//
+// Redistribution and use in source and binary forms, with or without
+// modification, are permitted provided that the following conditions are
+// met:
+//
+//     * Redistributions of source code must retain the above copyright
+// notice, this list of conditions and the following disclaimer.
+//     * Redistributions in binary form must reproduce the above
+// copyright notice, this list of conditions and the following disclaimer
+// in the documentation and/or other materials provided with the
+// distribution.
+//     * Neither the name of Google Inc. nor the names of its
+// contributors may be used to endorse or promote products derived from
+// this software without specific prior written permission.
+//
+// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+//
+// Author: Sanjay Ghemawat
+//
+// TODO: Test extractions for PartialMatch/Consume
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include 
+#include       /* for memset and strcmp */
+#include 
+#include 
+#include "pcrecpp.h"
+
+using pcrecpp::StringPiece;
+using pcrecpp::RE;
+using pcrecpp::RE_Options;
+using pcrecpp::Hex;
+using pcrecpp::Octal;
+using pcrecpp::CRadix;
+
+static bool VERBOSE_TEST  = false;
+
+// CHECK dies with a fatal error if condition is not true.  It is *not*
+// controlled by NDEBUG, so the check will be executed regardless of
+// compilation mode.  Therefore, it is safe to do things like:
+//    CHECK_EQ(fp->Write(x), 4)
+#define CHECK(condition) do {                           \
+  if (!(condition)) {                                   \
+    fprintf(stderr, "%s:%d: Check failed: %s\n",        \
+            __FILE__, __LINE__, #condition);            \
+    exit(1);                                            \
+  }                                                     \
+} while (0)
+
+#define CHECK_EQ(a, b)   CHECK(a == b)
+
+static void Timing1(int num_iters) {
+  // Same pattern lots of times
+  RE pattern("ruby:\\d+");
+  StringPiece p("ruby:1234");
+  for (int j = num_iters; j > 0; j--) {
+    CHECK(pattern.FullMatch(p));
+  }
+}
+
+static void Timing2(int num_iters) {
+  // Same pattern lots of times
+  RE pattern("ruby:(\\d+)");
+  int i;
+  for (int j = num_iters; j > 0; j--) {
+    CHECK(pattern.FullMatch("ruby:1234", &i));
+    CHECK_EQ(i, 1234);
+  }
+}
+
+static void Timing3(int num_iters) {
+  string text_string;
+  for (int j = num_iters; j > 0; j--) {
+    text_string += "this is another line\n";
+  }
+
+  RE line_matcher(".*\n");
+  string line;
+  StringPiece text(text_string);
+  int counter = 0;
+  while (line_matcher.Consume(&text)) {
+    counter++;
+  }
+  printf("Matched %d lines\n", counter);
+}
+
+#if 0  // uncomment this if you have a way of defining VirtualProcessSize()
+
+static void LeakTest() {
+  // Check for memory leaks
+  unsigned long long initial_size = 0;
+  for (int i = 0; i < 100000; i++) {
+    if (i == 50000) {
+      initial_size = VirtualProcessSize();
+      printf("Size after 50000: %llu\n", initial_size);
+    }
+    char buf[100];  // definitely big enough
+    sprintf(buf, "pat%09d", i);
+    RE newre(buf);
+  }
+  uint64 final_size = VirtualProcessSize();
+  printf("Size after 100000: %llu\n", final_size);
+  const double growth = double(final_size - initial_size) / final_size;
+  printf("Growth: %0.2f%%", growth * 100);
+  CHECK(growth < 0.02);       // Allow < 2% growth
+}
+
+#endif
+
+static void RadixTests() {
+  printf("Testing hex\n");
+
+#define CHECK_HEX(type, value) \
+  do { \
+    type v; \
+    CHECK(RE("([0-9a-fA-F]+)[uUlL]*").FullMatch(#value, Hex(&v))); \
+    CHECK_EQ(v, 0x ## value); \
+    CHECK(RE("([0-9a-fA-FxX]+)[uUlL]*").FullMatch("0x" #value, CRadix(&v))); \
+    CHECK_EQ(v, 0x ## value); \
+  } while(0)
+
+  CHECK_HEX(short,              2bad);
+  CHECK_HEX(unsigned short,     2badU);
+  CHECK_HEX(int,                dead);
+  CHECK_HEX(unsigned int,       deadU);
+  CHECK_HEX(long,               7eadbeefL);
+  CHECK_HEX(unsigned long,      deadbeefUL);
+#ifdef HAVE_LONG_LONG
+  CHECK_HEX(long long,          12345678deadbeefLL);
+#endif
+#ifdef HAVE_UNSIGNED_LONG_LONG
+  CHECK_HEX(unsigned long long, cafebabedeadbeefULL);
+#endif
+
+#undef CHECK_HEX
+
+  printf("Testing octal\n");
+
+#define CHECK_OCTAL(type, value) \
+  do { \
+    type v; \
+    CHECK(RE("([0-7]+)[uUlL]*").FullMatch(#value, Octal(&v))); \
+    CHECK_EQ(v, 0 ## value); \
+    CHECK(RE("([0-9a-fA-FxX]+)[uUlL]*").FullMatch("0" #value, CRadix(&v))); \
+    CHECK_EQ(v, 0 ## value); \
+  } while(0)
+
+  CHECK_OCTAL(short,              77777);
+  CHECK_OCTAL(unsigned short,     177777U);
+  CHECK_OCTAL(int,                17777777777);
+  CHECK_OCTAL(unsigned int,       37777777777U);
+  CHECK_OCTAL(long,               17777777777L);
+  CHECK_OCTAL(unsigned long,      37777777777UL);
+#ifdef HAVE_LONG_LONG
+  CHECK_OCTAL(long long,          777777777777777777777LL);
+#endif
+#ifdef HAVE_UNSIGNED_LONG_LONG
+  CHECK_OCTAL(unsigned long long, 1777777777777777777777ULL);
+#endif
+
+#undef CHECK_OCTAL
+
+  printf("Testing decimal\n");
+
+#define CHECK_DECIMAL(type, value) \
+  do { \
+    type v; \
+    CHECK(RE("(-?[0-9]+)[uUlL]*").FullMatch(#value, &v)); \
+    CHECK_EQ(v, value); \
+    CHECK(RE("(-?[0-9a-fA-FxX]+)[uUlL]*").FullMatch(#value, CRadix(&v))); \
+    CHECK_EQ(v, value); \
+  } while(0)
+
+  CHECK_DECIMAL(short,              -1);
+  CHECK_DECIMAL(unsigned short,     9999);
+  CHECK_DECIMAL(int,                -1000);
+  CHECK_DECIMAL(unsigned int,       12345U);
+  CHECK_DECIMAL(long,               -10000000L);
+  CHECK_DECIMAL(unsigned long,      3083324652U);
+#ifdef HAVE_LONG_LONG
+  CHECK_DECIMAL(long long,          -100000000000000LL);
+#endif
+#ifdef HAVE_UNSIGNED_LONG_LONG
+  CHECK_DECIMAL(unsigned long long, 1234567890987654321ULL);
+#endif
+
+#undef CHECK_DECIMAL
+
+}
+
+static void TestReplace() {
+  printf("Testing Replace\n");
+
+  struct ReplaceTest {
+    const char *regexp;
+    const char *rewrite;
+    const char *original;
+    const char *single;
+    const char *global;
+    int global_count;         // the expected return value from ReplaceAll
+  };
+  static const ReplaceTest tests[] = {
+    { "(qu|[b-df-hj-np-tv-z]*)([a-z]+)",
+      "\\2\\1ay",
+      "the quick brown fox jumps over the lazy dogs.",
+      "ethay quick brown fox jumps over the lazy dogs.",
+      "ethay ickquay ownbray oxfay umpsjay overay ethay azylay ogsday.",
+      9 },
+    { "\\w+",
+      "\\0-NOSPAM",
+      "paul.haahr@google.com",
+      "paul-NOSPAM.haahr@google.com",
+      "paul-NOSPAM.haahr-NOSPAM@google-NOSPAM.com-NOSPAM",
+      4 },
+    { "^",
+      "(START)",
+      "foo",
+      "(START)foo",
+      "(START)foo",
+      1 },
+    { "^",
+      "(START)",
+      "",
+      "(START)",
+      "(START)",
+      1 },
+    { "$",
+      "(END)",
+      "",
+      "(END)",
+      "(END)",
+      1 },
+    { "b",
+      "bb",
+      "ababababab",
+      "abbabababab",
+      "abbabbabbabbabb",
+       5 },
+    { "b",
+      "bb",
+      "bbbbbb",
+      "bbbbbbb",
+      "bbbbbbbbbbbb",
+      6 },
+    { "b+",
+      "bb",
+      "bbbbbb",
+      "bb",
+      "bb",
+      1 },
+    { "b*",
+      "bb",
+      "bbbbbb",
+      "bb",
+      "bbbb",
+      2 },
+    { "b*",
+      "bb",
+      "aaaaa",
+      "bbaaaaa",
+      "bbabbabbabbabbabb",
+      6 },
+    { "b*",
+      "bb",
+      "aa\naa\n",
+      "bbaa\naa\n",
+      "bbabbabb\nbbabbabb\nbb",
+      7 },
+    { "b*",
+      "bb",
+      "aa\raa\r",
+      "bbaa\raa\r",
+      "bbabbabb\rbbabbabb\rbb",
+      7 },
+    { "b*",
+      "bb",
+      "aa\r\naa\r\n",
+      "bbaa\r\naa\r\n",
+      "bbabbabb\r\nbbabbabb\r\nbb",
+      7 },
+    // Check empty-string matching (it's tricky!)
+    { "aa|b*",
+      "@",
+      "aa",
+      "@",
+      "@@",
+      2 },
+    { "b*|aa",
+      "@",
+      "aa",
+      "@aa",
+      "@@@",
+      3 },
+#ifdef SUPPORT_UTF8
+    { "b*",
+      "bb",
+      "\xE3\x83\x9B\xE3\x83\xBC\xE3\x83\xA0\xE3\x81\xB8",   // utf8
+      "bb\xE3\x83\x9B\xE3\x83\xBC\xE3\x83\xA0\xE3\x81\xB8",
+      "bb\xE3\x83\x9B""bb""\xE3\x83\xBC""bb""\xE3\x83\xA0""bb""\xE3\x81\xB8""bb",
+      5 },
+    { "b*",
+      "bb",
+      "\xE3\x83\x9B\r\n\xE3\x83\xBC\r\xE3\x83\xA0\n\xE3\x81\xB8\r\n",   // utf8
+      "bb\xE3\x83\x9B\r\n\xE3\x83\xBC\r\xE3\x83\xA0\n\xE3\x81\xB8\r\n",
+      ("bb\xE3\x83\x9B""bb\r\nbb""\xE3\x83\xBC""bb\rbb""\xE3\x83\xA0"
+       "bb\nbb""\xE3\x81\xB8""bb\r\nbb"),
+      9 },
+#endif
+    { "", NULL, NULL, NULL, NULL, 0 }
+  };
+
+#ifdef SUPPORT_UTF8
+  const bool support_utf8 = true;
+#else
+  const bool support_utf8 = false;
+#endif
+
+  for (const ReplaceTest *t = tests; t->original != NULL; ++t) {
+    RE re(t->regexp, RE_Options(PCRE_NEWLINE_CRLF).set_utf8(support_utf8));
+    assert(re.error().empty());
+    string one(t->original);
+    CHECK(re.Replace(t->rewrite, &one));
+    CHECK_EQ(one, t->single);
+    string all(t->original);
+    const int replace_count = re.GlobalReplace(t->rewrite, &all);
+    CHECK_EQ(all, t->global);
+    CHECK_EQ(replace_count, t->global_count);
+  }
+
+  // One final test: test \r\n replacement when we're not in CRLF mode
+  {
+    RE re("b*", RE_Options(PCRE_NEWLINE_CR).set_utf8(support_utf8));
+    assert(re.error().empty());
+    string all("aa\r\naa\r\n");
+    CHECK_EQ(re.GlobalReplace("bb", &all), 9);
+    CHECK_EQ(all, string("bbabbabb\rbb\nbbabbabb\rbb\nbb"));
+  }
+  {
+    RE re("b*", RE_Options(PCRE_NEWLINE_LF).set_utf8(support_utf8));
+    assert(re.error().empty());
+    string all("aa\r\naa\r\n");
+    CHECK_EQ(re.GlobalReplace("bb", &all), 9);
+    CHECK_EQ(all, string("bbabbabb\rbb\nbbabbabb\rbb\nbb"));
+  }
+  // TODO: test what happens when no PCRE_NEWLINE_* flag is set.
+  //       Alas, the answer depends on how pcre was compiled.
+}
+
+static void TestExtract() {
+  printf("Testing Extract\n");
+
+  string s;
+
+  CHECK(RE("(.*)@([^.]*)").Extract("\\2!\\1", "boris@kremvax.ru", &s));
+  CHECK_EQ(s, "kremvax!boris");
+
+  // check the RE interface as well
+  CHECK(RE(".*").Extract("'\\0'", "foo", &s));
+  CHECK_EQ(s, "'foo'");
+  CHECK(!RE("bar").Extract("'\\0'", "baz", &s));
+  CHECK_EQ(s, "'foo'");
+}
+
+static void TestConsume() {
+  printf("Testing Consume\n");
+
+  string word;
+
+  string s("   aaa b!@#$@#$cccc");
+  StringPiece input(s);
+
+  RE r("\\s*(\\w+)");    // matches a word, possibly proceeded by whitespace
+  CHECK(r.Consume(&input, &word));
+  CHECK_EQ(word, "aaa");
+  CHECK(r.Consume(&input, &word));
+  CHECK_EQ(word, "b");
+  CHECK(! r.Consume(&input, &word));
+}
+
+static void TestFindAndConsume() {
+  printf("Testing FindAndConsume\n");
+
+  string word;
+
+  string s("   aaa b!@#$@#$cccc");
+  StringPiece input(s);
+
+  RE r("(\\w+)");      // matches a word
+  CHECK(r.FindAndConsume(&input, &word));
+  CHECK_EQ(word, "aaa");
+  CHECK(r.FindAndConsume(&input, &word));
+  CHECK_EQ(word, "b");
+  CHECK(r.FindAndConsume(&input, &word));
+  CHECK_EQ(word, "cccc");
+  CHECK(! r.FindAndConsume(&input, &word));
+}
+
+static void TestMatchNumberPeculiarity() {
+  printf("Testing match-number peculiarity\n");
+
+  string word1;
+  string word2;
+  string word3;
+
+  RE r("(foo)|(bar)|(baz)");
+  CHECK(r.PartialMatch("foo", &word1, &word2, &word3));
+  CHECK_EQ(word1, "foo");
+  CHECK_EQ(word2, "");
+  CHECK_EQ(word3, "");
+  CHECK(r.PartialMatch("bar", &word1, &word2, &word3));
+  CHECK_EQ(word1, "");
+  CHECK_EQ(word2, "bar");
+  CHECK_EQ(word3, "");
+  CHECK(r.PartialMatch("baz", &word1, &word2, &word3));
+  CHECK_EQ(word1, "");
+  CHECK_EQ(word2, "");
+  CHECK_EQ(word3, "baz");
+  CHECK(!r.PartialMatch("f", &word1, &word2, &word3));
+
+  string a;
+  CHECK(RE("(foo)|hello").FullMatch("hello", &a));
+  CHECK_EQ(a, "");
+}
+
+static void TestRecursion() {
+  printf("Testing recursion\n");
+
+  // Get one string that passes (sometimes), one that never does.
+  string text_good("abcdefghijk");
+  string text_bad("acdefghijkl");
+
+  // According to pcretest, matching text_good against (\w+)*b
+  // requires match_limit of at least 8192, and match_recursion_limit
+  // of at least 37.
+
+  RE_Options options_ml;
+  options_ml.set_match_limit(8192);
+  RE re("(\\w+)*b", options_ml);
+  CHECK(re.PartialMatch(text_good) == true);
+  CHECK(re.PartialMatch(text_bad) == false);
+  CHECK(re.FullMatch(text_good) == false);
+  CHECK(re.FullMatch(text_bad) == false);
+
+  options_ml.set_match_limit(1024);
+  RE re2("(\\w+)*b", options_ml);
+  CHECK(re2.PartialMatch(text_good) == false);   // because of match_limit
+  CHECK(re2.PartialMatch(text_bad) == false);
+  CHECK(re2.FullMatch(text_good) == false);
+  CHECK(re2.FullMatch(text_bad) == false);
+
+  RE_Options options_mlr;
+  options_mlr.set_match_limit_recursion(50);
+  RE re3("(\\w+)*b", options_mlr);
+  CHECK(re3.PartialMatch(text_good) == true);
+  CHECK(re3.PartialMatch(text_bad) == false);
+  CHECK(re3.FullMatch(text_good) == false);
+  CHECK(re3.FullMatch(text_bad) == false);
+
+  options_mlr.set_match_limit_recursion(10);
+  RE re4("(\\w+)*b", options_mlr);
+  CHECK(re4.PartialMatch(text_good) == false);
+  CHECK(re4.PartialMatch(text_bad) == false);
+  CHECK(re4.FullMatch(text_good) == false);
+  CHECK(re4.FullMatch(text_bad) == false);
+}
+
+// A meta-quoted string, interpreted as a pattern, should always match
+// the original unquoted string.
+static void TestQuoteMeta(string unquoted, RE_Options options = RE_Options()) {
+  string quoted = RE::QuoteMeta(unquoted);
+  RE re(quoted, options);
+  CHECK(re.FullMatch(unquoted));
+}
+
+// A string containing meaningful regexp characters, which is then meta-
+// quoted, should not generally match a string the unquoted string does.
+static void NegativeTestQuoteMeta(string unquoted, string should_not_match,
+                                  RE_Options options = RE_Options()) {
+  string quoted = RE::QuoteMeta(unquoted);
+  RE re(quoted, options);
+  CHECK(!re.FullMatch(should_not_match));
+}
+
+// Tests that quoted meta characters match their original strings,
+// and that a few things that shouldn't match indeed do not.
+static void TestQuotaMetaSimple() {
+  TestQuoteMeta("foo");
+  TestQuoteMeta("foo.bar");
+  TestQuoteMeta("foo\\.bar");
+  TestQuoteMeta("[1-9]");
+  TestQuoteMeta("1.5-2.0?");
+  TestQuoteMeta("\\d");
+  TestQuoteMeta("Who doesn't like ice cream?");
+  TestQuoteMeta("((a|b)c?d*e+[f-h]i)");
+  TestQuoteMeta("((?!)xxx).*yyy");
+  TestQuoteMeta("([");
+  TestQuoteMeta(string("foo\0bar", 7));
+}
+
+static void TestQuoteMetaSimpleNegative() {
+  NegativeTestQuoteMeta("foo", "bar");
+  NegativeTestQuoteMeta("...", "bar");
+  NegativeTestQuoteMeta("\\.", ".");
+  NegativeTestQuoteMeta("\\.", "..");
+  NegativeTestQuoteMeta("(a)", "a");
+  NegativeTestQuoteMeta("(a|b)", "a");
+  NegativeTestQuoteMeta("(a|b)", "(a)");
+  NegativeTestQuoteMeta("(a|b)", "a|b");
+  NegativeTestQuoteMeta("[0-9]", "0");
+  NegativeTestQuoteMeta("[0-9]", "0-9");
+  NegativeTestQuoteMeta("[0-9]", "[9]");
+  NegativeTestQuoteMeta("((?!)xxx)", "xxx");
+}
+
+static void TestQuoteMetaLatin1() {
+  TestQuoteMeta("3\xb2 = 9");
+}
+
+static void TestQuoteMetaUtf8() {
+#ifdef SUPPORT_UTF8
+  TestQuoteMeta("Pl\xc3\xa1\x63ido Domingo", pcrecpp::UTF8());
+  TestQuoteMeta("xyz", pcrecpp::UTF8());            // No fancy utf8
+  TestQuoteMeta("\xc2\xb0", pcrecpp::UTF8());       // 2-byte utf8 (degree symbol)
+  TestQuoteMeta("27\xc2\xb0 degrees", pcrecpp::UTF8());  // As a middle character
+  TestQuoteMeta("\xe2\x80\xb3", pcrecpp::UTF8());   // 3-byte utf8 (double prime)
+  TestQuoteMeta("\xf0\x9d\x85\x9f", pcrecpp::UTF8()); // 4-byte utf8 (music note)
+  TestQuoteMeta("27\xc2\xb0"); // Interpreted as Latin-1, but should still work
+  NegativeTestQuoteMeta("27\xc2\xb0",               // 2-byte utf (degree symbol)
+                        "27\\\xc2\\\xb0",
+                        pcrecpp::UTF8());
+#endif
+}
+
+static void TestQuoteMetaAll() {
+  printf("Testing QuoteMeta\n");
+  TestQuotaMetaSimple();
+  TestQuoteMetaSimpleNegative();
+  TestQuoteMetaLatin1();
+  TestQuoteMetaUtf8();
+}
+
+//
+// Options tests contributed by
+// Giuseppe Maxia, CTO, Stardata s.r.l.
+// July 2005
+//
+static void GetOneOptionResult(
+                const char *option_name,
+                const char *regex,
+                const char *str,
+                RE_Options options,
+                bool full,
+                string expected) {
+
+  printf("Testing Option <%s>\n", option_name);
+  if(VERBOSE_TEST)
+    printf("/%s/ finds \"%s\" within \"%s\" \n",
+                    regex,
+                    expected.c_str(),
+                    str);
+  string captured("");
+  if (full)
+    RE(regex,options).FullMatch(str, &captured);
+  else
+    RE(regex,options).PartialMatch(str, &captured);
+  CHECK_EQ(captured, expected);
+}
+
+static void TestOneOption(
+                const char *option_name,
+                const char *regex,
+                const char *str,
+                RE_Options options,
+                bool full,
+                bool assertive = true) {
+
+  printf("Testing Option <%s>\n", option_name);
+  if (VERBOSE_TEST)
+    printf("'%s' %s /%s/ \n",
+                  str,
+                  (assertive? "matches" : "doesn't match"),
+                  regex);
+  if (assertive) {
+    if (full)
+      CHECK(RE(regex,options).FullMatch(str));
+    else
+      CHECK(RE(regex,options).PartialMatch(str));
+  } else {
+    if (full)
+      CHECK(!RE(regex,options).FullMatch(str));
+    else
+      CHECK(!RE(regex,options).PartialMatch(str));
+  }
+}
+
+static void Test_CASELESS() {
+  RE_Options options;
+  RE_Options options2;
+
+  options.set_caseless(true);
+  TestOneOption("CASELESS (class)",  "HELLO",    "hello", options, false);
+  TestOneOption("CASELESS (class2)", "HELLO",    "hello", options2.set_caseless(true), false);
+  TestOneOption("CASELESS (class)",  "^[A-Z]+$", "Hello", options, false);
+
+  TestOneOption("CASELESS (function)", "HELLO",    "hello", pcrecpp::CASELESS(), false);
+  TestOneOption("CASELESS (function)", "^[A-Z]+$", "Hello", pcrecpp::CASELESS(), false);
+  options.set_caseless(false);
+  TestOneOption("no CASELESS", "HELLO",    "hello", options, false, false);
+}
+
+static void Test_MULTILINE() {
+  RE_Options options;
+  RE_Options options2;
+  const char *str = "HELLO\n" "cruel\n" "world\n";
+
+  options.set_multiline(true);
+  TestOneOption("MULTILINE (class)",    "^cruel$", str, options, false);
+  TestOneOption("MULTILINE (class2)",   "^cruel$", str, options2.set_multiline(true), false);
+  TestOneOption("MULTILINE (function)", "^cruel$", str, pcrecpp::MULTILINE(), false);
+  options.set_multiline(false);
+  TestOneOption("no MULTILINE", "^cruel$", str, options, false, false);
+}
+
+static void Test_DOTALL() {
+  RE_Options options;
+  RE_Options options2;
+  const char *str = "HELLO\n" "cruel\n" "world";
+
+  options.set_dotall(true);
+  TestOneOption("DOTALL (class)",    "HELLO.*world", str, options, true);
+  TestOneOption("DOTALL (class2)",   "HELLO.*world", str, options2.set_dotall(true), true);
+  TestOneOption("DOTALL (function)",    "HELLO.*world", str, pcrecpp::DOTALL(), true);
+  options.set_dotall(false);
+  TestOneOption("no DOTALL", "HELLO.*world", str, options, true, false);
+}
+
+static void Test_DOLLAR_ENDONLY() {
+  RE_Options options;
+  RE_Options options2;
+  const char *str = "HELLO world\n";
+
+  TestOneOption("no DOLLAR_ENDONLY", "world$", str, options, false);
+  options.set_dollar_endonly(true);
+  TestOneOption("DOLLAR_ENDONLY 1",    "world$", str, options, false, false);
+  TestOneOption("DOLLAR_ENDONLY 2",    "world$", str, options2.set_dollar_endonly(true), false, false);
+}
+
+static void Test_EXTRA() {
+  RE_Options options;
+  const char *str = "HELLO";
+
+  options.set_extra(true);
+  TestOneOption("EXTRA 1", "\\HELL\\O", str, options, true, false );
+  TestOneOption("EXTRA 2", "\\HELL\\O", str, RE_Options().set_extra(true), true, false );
+  options.set_extra(false);
+  TestOneOption("no EXTRA", "\\HELL\\O", str, options, true );
+}
+
+static void Test_EXTENDED() {
+  RE_Options options;
+  RE_Options options2;
+  const char *str = "HELLO world";
+
+  options.set_extended(true);
+  TestOneOption("EXTENDED (class)",    "HELLO world", str, options, false, false);
+  TestOneOption("EXTENDED (class2)",   "HELLO world", str, options2.set_extended(true), false, false);
+  TestOneOption("EXTENDED (class)",
+                    "^ HE L{2} O "
+                    "\\s+        "
+                    "\\w+ $      ",
+                    str,
+                    options,
+                    false);
+
+  TestOneOption("EXTENDED (function)",    "HELLO world", str, pcrecpp::EXTENDED(), false, false);
+  TestOneOption("EXTENDED (function)",
+                    "^ HE L{2} O "
+                    "\\s+        "
+                    "\\w+ $      ",
+                    str,
+                    pcrecpp::EXTENDED(),
+                    false);
+
+  options.set_extended(false);
+  TestOneOption("no EXTENDED", "HELLO world", str, options, false);
+}
+
+static void Test_NO_AUTO_CAPTURE() {
+  RE_Options options;
+  const char *str = "HELLO world";
+  string captured;
+
+  printf("Testing Option \n");
+  if (VERBOSE_TEST)
+    printf("parentheses capture text\n");
+  RE re("(world|universe)$", options);
+  CHECK(re.Extract("\\1", str , &captured));
+  CHECK_EQ(captured, "world");
+  options.set_no_auto_capture(true);
+  printf("testing Option \n");
+  if (VERBOSE_TEST)
+    printf("parentheses do not capture text\n");
+  re.Extract("\\1",str, &captured );
+  CHECK_EQ(captured, "world");
+}
+
+static void Test_UNGREEDY() {
+  RE_Options options;
+  const char *str = "HELLO, 'this' is the 'world'";
+
+  options.set_ungreedy(true);
+  GetOneOptionResult("UNGREEDY 1", "('.*')", str, options, false, "'this'" );
+  GetOneOptionResult("UNGREEDY 2", "('.*')", str, RE_Options().set_ungreedy(true), false, "'this'" );
+  GetOneOptionResult("UNGREEDY", "('.*?')", str, options, false, "'this' is the 'world'" );
+
+  options.set_ungreedy(false);
+  GetOneOptionResult("no UNGREEDY", "('.*')", str, options, false, "'this' is the 'world'" );
+  GetOneOptionResult("no UNGREEDY", "('.*?')", str, options, false, "'this'" );
+}
+
+static void Test_all_options() {
+  const char *str = "HELLO\n" "cruel\n" "world";
+  RE_Options options;
+  options.set_all_options(PCRE_CASELESS | PCRE_DOTALL);
+
+  TestOneOption("all_options (CASELESS|DOTALL)", "^hello.*WORLD", str , options, false);
+  options.set_all_options(0);
+  TestOneOption("all_options (0)", "^hello.*WORLD", str , options, false, false);
+  options.set_all_options(PCRE_MULTILINE | PCRE_EXTENDED);
+
+  TestOneOption("all_options (MULTILINE|EXTENDED)", " ^ c r u e l $ ", str, options, false);
+  TestOneOption("all_options (MULTILINE|EXTENDED) with constructor",
+                  " ^ c r u e l $ ",
+                  str,
+                  RE_Options(PCRE_MULTILINE | PCRE_EXTENDED),
+                  false);
+
+  TestOneOption("all_options (MULTILINE|EXTENDED) with concatenation",
+                  " ^ c r u e l $ ",
+                  str,
+                  RE_Options()
+                       .set_multiline(true)
+                       .set_extended(true),
+                  false);
+
+  options.set_all_options(0);
+  TestOneOption("all_options (0)", "^ c r u e l $", str, options, false, false);
+
+}
+
+static void TestOptions() {
+  printf("Testing Options\n");
+  Test_CASELESS();
+  Test_MULTILINE();
+  Test_DOTALL();
+  Test_DOLLAR_ENDONLY();
+  Test_EXTENDED();
+  Test_NO_AUTO_CAPTURE();
+  Test_UNGREEDY();
+  Test_EXTRA();
+  Test_all_options();
+}
+
+static void TestConstructors() {
+  printf("Testing constructors\n");
+
+  RE_Options options;
+  options.set_dotall(true);
+  const char *str = "HELLO\n" "cruel\n" "world";
+
+  RE orig("HELLO.*world", options);
+  CHECK(orig.FullMatch(str));
+
+  RE copy1(orig);
+  CHECK(copy1.FullMatch(str));
+
+  RE copy2("not a match");
+  CHECK(!copy2.FullMatch(str));
+  copy2 = copy1;
+  CHECK(copy2.FullMatch(str));
+  copy2 = orig;
+  CHECK(copy2.FullMatch(str));
+
+  // Make sure when we assign to ourselves, nothing bad happens
+  orig = orig;
+  copy1 = copy1;
+  copy2 = copy2;
+  CHECK(orig.FullMatch(str));
+  CHECK(copy1.FullMatch(str));
+  CHECK(copy2.FullMatch(str));
+}
+
+int main(int argc, char** argv) {
+  // Treat any flag as --help
+  if (argc > 1 && argv[1][0] == '-') {
+    printf("Usage: %s [timing1|timing2|timing3 num-iters]\n"
+           "       If 'timingX ###' is specified, run the given timing test\n"
+           "       with the given number of iterations, rather than running\n"
+           "       the default corectness test.\n", argv[0]);
+    return 0;
+  }
+
+  if (argc > 1) {
+    if ( argc == 2 || atoi(argv[2]) == 0) {
+      printf("timing mode needs a num-iters argument\n");
+      return 1;
+    }
+    if (!strcmp(argv[1], "timing1"))
+      Timing1(atoi(argv[2]));
+    else if (!strcmp(argv[1], "timing2"))
+      Timing2(atoi(argv[2]));
+    else if (!strcmp(argv[1], "timing3"))
+      Timing3(atoi(argv[2]));
+    else
+      printf("Unknown argument '%s'\n", argv[1]);
+    return 0;
+  }
+
+  printf("PCRE C++ wrapper tests\n");
+  printf("Testing FullMatch\n");
+
+  int i;
+  string s;
+
+  /***** FullMatch with no args *****/
+
+  CHECK(RE("h.*o").FullMatch("hello"));
+  CHECK(!RE("h.*o").FullMatch("othello"));     // Must be anchored at front
+  CHECK(!RE("h.*o").FullMatch("hello!"));      // Must be anchored at end
+  CHECK(RE("a*").FullMatch("aaaa"));           // Fullmatch with normal op
+  CHECK(RE("a*?").FullMatch("aaaa"));          // Fullmatch with nongreedy op
+  CHECK(RE("a*?\\z").FullMatch("aaaa"));       // Two unusual ops
+
+  /***** FullMatch with args *****/
+
+  // Zero-arg
+  CHECK(RE("\\d+").FullMatch("1001"));
+
+  // Single-arg
+  CHECK(RE("(\\d+)").FullMatch("1001",   &i));
+  CHECK_EQ(i, 1001);
+  CHECK(RE("(-?\\d+)").FullMatch("-123", &i));
+  CHECK_EQ(i, -123);
+  CHECK(!RE("()\\d+").FullMatch("10", &i));
+  CHECK(!RE("(\\d+)").FullMatch("1234567890123456789012345678901234567890",
+                                &i));
+
+  // Digits surrounding integer-arg
+  CHECK(RE("1(\\d*)4").FullMatch("1234", &i));
+  CHECK_EQ(i, 23);
+  CHECK(RE("(\\d)\\d+").FullMatch("1234", &i));
+  CHECK_EQ(i, 1);
+  CHECK(RE("(-\\d)\\d+").FullMatch("-1234", &i));
+  CHECK_EQ(i, -1);
+  CHECK(RE("(\\d)").PartialMatch("1234", &i));
+  CHECK_EQ(i, 1);
+  CHECK(RE("(-\\d)").PartialMatch("-1234", &i));
+  CHECK_EQ(i, -1);
+
+  // String-arg
+  CHECK(RE("h(.*)o").FullMatch("hello", &s));
+  CHECK_EQ(s, string("ell"));
+
+  // StringPiece-arg
+  StringPiece sp;
+  CHECK(RE("(\\w+):(\\d+)").FullMatch("ruby:1234", &sp, &i));
+  CHECK_EQ(sp.size(), 4);
+  CHECK(memcmp(sp.data(), "ruby", 4) == 0);
+  CHECK_EQ(i, 1234);
+
+  // Multi-arg
+  CHECK(RE("(\\w+):(\\d+)").FullMatch("ruby:1234", &s, &i));
+  CHECK_EQ(s, string("ruby"));
+  CHECK_EQ(i, 1234);
+
+  // Ignore non-void* NULL arg
+  CHECK(RE("he(.*)lo").FullMatch("hello", (char*)NULL));
+  CHECK(RE("h(.*)o").FullMatch("hello", (string*)NULL));
+  CHECK(RE("h(.*)o").FullMatch("hello", (StringPiece*)NULL));
+  CHECK(RE("(.*)").FullMatch("1234", (int*)NULL));
+#ifdef HAVE_LONG_LONG
+  CHECK(RE("(.*)").FullMatch("1234567890123456", (long long*)NULL));
+#endif
+  CHECK(RE("(.*)").FullMatch("123.4567890123456", (double*)NULL));
+  CHECK(RE("(.*)").FullMatch("123.4567890123456", (float*)NULL));
+
+  // Fail on non-void* NULL arg if the match doesn't parse for the given type.
+  CHECK(!RE("h(.*)lo").FullMatch("hello", &s, (char*)NULL));
+  CHECK(!RE("(.*)").FullMatch("hello", (int*)NULL));
+  CHECK(!RE("(.*)").FullMatch("1234567890123456", (int*)NULL));
+  CHECK(!RE("(.*)").FullMatch("hello", (double*)NULL));
+  CHECK(!RE("(.*)").FullMatch("hello", (float*)NULL));
+
+  // Ignored arg
+  CHECK(RE("(\\w+)(:)(\\d+)").FullMatch("ruby:1234", &s, (void*)NULL, &i));
+  CHECK_EQ(s, string("ruby"));
+  CHECK_EQ(i, 1234);
+
+  // Type tests
+  {
+    char c;
+    CHECK(RE("(H)ello").FullMatch("Hello", &c));
+    CHECK_EQ(c, 'H');
+  }
+  {
+    unsigned char c;
+    CHECK(RE("(H)ello").FullMatch("Hello", &c));
+    CHECK_EQ(c, static_cast('H'));
+  }
+  {
+    short v;
+    CHECK(RE("(-?\\d+)").FullMatch("100",     &v));    CHECK_EQ(v, 100);
+    CHECK(RE("(-?\\d+)").FullMatch("-100",    &v));    CHECK_EQ(v, -100);
+    CHECK(RE("(-?\\d+)").FullMatch("32767",   &v));    CHECK_EQ(v, 32767);
+    CHECK(RE("(-?\\d+)").FullMatch("-32768",  &v));    CHECK_EQ(v, -32768);
+    CHECK(!RE("(-?\\d+)").FullMatch("-32769", &v));
+    CHECK(!RE("(-?\\d+)").FullMatch("32768",  &v));
+  }
+  {
+    unsigned short v;
+    CHECK(RE("(\\d+)").FullMatch("100",     &v));    CHECK_EQ(v, 100);
+    CHECK(RE("(\\d+)").FullMatch("32767",   &v));    CHECK_EQ(v, 32767);
+    CHECK(RE("(\\d+)").FullMatch("65535",   &v));    CHECK_EQ(v, 65535);
+    CHECK(!RE("(\\d+)").FullMatch("65536",  &v));
+  }
+  {
+    int v;
+    static const int max_value = 0x7fffffff;
+    static const int min_value = -max_value - 1;
+    CHECK(RE("(-?\\d+)").FullMatch("100",         &v)); CHECK_EQ(v, 100);
+    CHECK(RE("(-?\\d+)").FullMatch("-100",        &v)); CHECK_EQ(v, -100);
+    CHECK(RE("(-?\\d+)").FullMatch("2147483647",  &v)); CHECK_EQ(v, max_value);
+    CHECK(RE("(-?\\d+)").FullMatch("-2147483648", &v)); CHECK_EQ(v, min_value);
+    CHECK(!RE("(-?\\d+)").FullMatch("-2147483649", &v));
+    CHECK(!RE("(-?\\d+)").FullMatch("2147483648",  &v));
+  }
+  {
+    unsigned int v;
+    static const unsigned int max_value = 0xfffffffful;
+    CHECK(RE("(\\d+)").FullMatch("100",         &v)); CHECK_EQ(v, 100);
+    CHECK(RE("(\\d+)").FullMatch("4294967295",  &v)); CHECK_EQ(v, max_value);
+    CHECK(!RE("(\\d+)").FullMatch("4294967296", &v));
+  }
+#ifdef HAVE_LONG_LONG
+# if defined(__MINGW__) || defined(__MINGW32__)
+#   define LLD "%I64d"
+#   define LLU "%I64u"
+# else
+#   define LLD "%lld"
+#   define LLU "%llu"
+# endif
+  {
+    long long v;
+    static const long long max_value = 0x7fffffffffffffffLL;
+    static const long long min_value = -max_value - 1;
+    char buf[32];  // definitely big enough for a long long
+
+    CHECK(RE("(-?\\d+)").FullMatch("100", &v)); CHECK_EQ(v, 100);
+    CHECK(RE("(-?\\d+)").FullMatch("-100",&v)); CHECK_EQ(v, -100);
+
+    sprintf(buf, LLD, max_value);
+    CHECK(RE("(-?\\d+)").FullMatch(buf,&v)); CHECK_EQ(v, max_value);
+
+    sprintf(buf, LLD, min_value);
+    CHECK(RE("(-?\\d+)").FullMatch(buf,&v)); CHECK_EQ(v, min_value);
+
+    sprintf(buf, LLD, max_value);
+    assert(buf[strlen(buf)-1] != '9');
+    buf[strlen(buf)-1]++;
+    CHECK(!RE("(-?\\d+)").FullMatch(buf, &v));
+
+    sprintf(buf, LLD, min_value);
+    assert(buf[strlen(buf)-1] != '9');
+    buf[strlen(buf)-1]++;
+    CHECK(!RE("(-?\\d+)").FullMatch(buf, &v));
+  }
+#endif
+#if defined HAVE_UNSIGNED_LONG_LONG && defined HAVE_LONG_LONG
+  {
+    unsigned long long v;
+    long long v2;
+    static const unsigned long long max_value = 0xffffffffffffffffULL;
+    char buf[32];  // definitely big enough for a unsigned long long
+
+    CHECK(RE("(-?\\d+)").FullMatch("100",&v)); CHECK_EQ(v, 100);
+    CHECK(RE("(-?\\d+)").FullMatch("-100",&v2)); CHECK_EQ(v2, -100);
+
+    sprintf(buf, LLU, max_value);
+    CHECK(RE("(-?\\d+)").FullMatch(buf,&v)); CHECK_EQ(v, max_value);
+
+    assert(buf[strlen(buf)-1] != '9');
+    buf[strlen(buf)-1]++;
+    CHECK(!RE("(-?\\d+)").FullMatch(buf, &v));
+  }
+#endif
+  {
+    float v;
+    CHECK(RE("(.*)").FullMatch("100", &v));
+    CHECK(RE("(.*)").FullMatch("-100.", &v));
+    CHECK(RE("(.*)").FullMatch("1e23", &v));
+  }
+  {
+    double v;
+    CHECK(RE("(.*)").FullMatch("100", &v));
+    CHECK(RE("(.*)").FullMatch("-100.", &v));
+    CHECK(RE("(.*)").FullMatch("1e23", &v));
+  }
+
+  // Check that matching is fully anchored
+  CHECK(!RE("(\\d+)").FullMatch("x1001",  &i));
+  CHECK(!RE("(\\d+)").FullMatch("1001x",  &i));
+  CHECK(RE("x(\\d+)").FullMatch("x1001", &i)); CHECK_EQ(i, 1001);
+  CHECK(RE("(\\d+)x").FullMatch("1001x", &i)); CHECK_EQ(i, 1001);
+
+  // Braces
+  CHECK(RE("[0-9a-f+.-]{5,}").FullMatch("0abcd"));
+  CHECK(RE("[0-9a-f+.-]{5,}").FullMatch("0abcde"));
+  CHECK(!RE("[0-9a-f+.-]{5,}").FullMatch("0abc"));
+
+  // Complicated RE
+  CHECK(RE("foo|bar|[A-Z]").FullMatch("foo"));
+  CHECK(RE("foo|bar|[A-Z]").FullMatch("bar"));
+  CHECK(RE("foo|bar|[A-Z]").FullMatch("X"));
+  CHECK(!RE("foo|bar|[A-Z]").FullMatch("XY"));
+
+  // Check full-match handling (needs '$' tacked on internally)
+  CHECK(RE("fo|foo").FullMatch("fo"));
+  CHECK(RE("fo|foo").FullMatch("foo"));
+  CHECK(RE("fo|foo$").FullMatch("fo"));
+  CHECK(RE("fo|foo$").FullMatch("foo"));
+  CHECK(RE("foo$").FullMatch("foo"));
+  CHECK(!RE("foo\\$").FullMatch("foo$bar"));
+  CHECK(!RE("fo|bar").FullMatch("fox"));
+
+  // Uncomment the following if we change the handling of '$' to
+  // prevent it from matching a trailing newline
+  if (false) {
+    // Check that we don't get bitten by pcre's special handling of a
+    // '\n' at the end of the string matching '$'
+    CHECK(!RE("foo$").PartialMatch("foo\n"));
+  }
+
+  // Number of args
+  int a[16];
+  CHECK(RE("").FullMatch(""));
+
+  memset(a, 0, sizeof(0));
+  CHECK(RE("(\\d){1}").FullMatch("1",
+                                 &a[0]));
+  CHECK_EQ(a[0], 1);
+
+  memset(a, 0, sizeof(0));
+  CHECK(RE("(\\d)(\\d)").FullMatch("12",
+                                   &a[0],  &a[1]));
+  CHECK_EQ(a[0], 1);
+  CHECK_EQ(a[1], 2);
+
+  memset(a, 0, sizeof(0));
+  CHECK(RE("(\\d)(\\d)(\\d)").FullMatch("123",
+                                        &a[0],  &a[1],  &a[2]));
+  CHECK_EQ(a[0], 1);
+  CHECK_EQ(a[1], 2);
+  CHECK_EQ(a[2], 3);
+
+  memset(a, 0, sizeof(0));
+  CHECK(RE("(\\d)(\\d)(\\d)(\\d)").FullMatch("1234",
+                                             &a[0],  &a[1],  &a[2],  &a[3]));
+  CHECK_EQ(a[0], 1);
+  CHECK_EQ(a[1], 2);
+  CHECK_EQ(a[2], 3);
+  CHECK_EQ(a[3], 4);
+
+  memset(a, 0, sizeof(0));
+  CHECK(RE("(\\d)(\\d)(\\d)(\\d)(\\d)").FullMatch("12345",
+                                                  &a[0],  &a[1],  &a[2],
+                                                  &a[3],  &a[4]));
+  CHECK_EQ(a[0], 1);
+  CHECK_EQ(a[1], 2);
+  CHECK_EQ(a[2], 3);
+  CHECK_EQ(a[3], 4);
+  CHECK_EQ(a[4], 5);
+
+  memset(a, 0, sizeof(0));
+  CHECK(RE("(\\d)(\\d)(\\d)(\\d)(\\d)(\\d)").FullMatch("123456",
+                                                       &a[0],  &a[1],  &a[2],
+                                                       &a[3],  &a[4],  &a[5]));
+  CHECK_EQ(a[0], 1);
+  CHECK_EQ(a[1], 2);
+  CHECK_EQ(a[2], 3);
+  CHECK_EQ(a[3], 4);
+  CHECK_EQ(a[4], 5);
+  CHECK_EQ(a[5], 6);
+
+  memset(a, 0, sizeof(0));
+  CHECK(RE("(\\d)(\\d)(\\d)(\\d)(\\d)(\\d)(\\d)").FullMatch("1234567",
+                                                            &a[0],  &a[1],  &a[2],  &a[3],
+                                                            &a[4],  &a[5],  &a[6]));
+  CHECK_EQ(a[0], 1);
+  CHECK_EQ(a[1], 2);
+  CHECK_EQ(a[2], 3);
+  CHECK_EQ(a[3], 4);
+  CHECK_EQ(a[4], 5);
+  CHECK_EQ(a[5], 6);
+  CHECK_EQ(a[6], 7);
+
+  memset(a, 0, sizeof(0));
+  CHECK(RE("(\\d)(\\d)(\\d)(\\d)(\\d)(\\d)(\\d)(\\d)"
+           "(\\d)(\\d)(\\d)(\\d)(\\d)(\\d)(\\d)(\\d)").FullMatch(
+               "1234567890123456",
+               &a[0],  &a[1],  &a[2],  &a[3],
+               &a[4],  &a[5],  &a[6],  &a[7],
+               &a[8],  &a[9],  &a[10], &a[11],
+               &a[12], &a[13], &a[14], &a[15]));
+  CHECK_EQ(a[0], 1);
+  CHECK_EQ(a[1], 2);
+  CHECK_EQ(a[2], 3);
+  CHECK_EQ(a[3], 4);
+  CHECK_EQ(a[4], 5);
+  CHECK_EQ(a[5], 6);
+  CHECK_EQ(a[6], 7);
+  CHECK_EQ(a[7], 8);
+  CHECK_EQ(a[8], 9);
+  CHECK_EQ(a[9], 0);
+  CHECK_EQ(a[10], 1);
+  CHECK_EQ(a[11], 2);
+  CHECK_EQ(a[12], 3);
+  CHECK_EQ(a[13], 4);
+  CHECK_EQ(a[14], 5);
+  CHECK_EQ(a[15], 6);
+
+  /***** PartialMatch *****/
+
+  printf("Testing PartialMatch\n");
+
+  CHECK(RE("h.*o").PartialMatch("hello"));
+  CHECK(RE("h.*o").PartialMatch("othello"));
+  CHECK(RE("h.*o").PartialMatch("hello!"));
+  CHECK(RE("((((((((((((((((((((x))))))))))))))))))))").PartialMatch("x"));
+
+  /***** other tests *****/
+
+  RadixTests();
+  TestReplace();
+  TestExtract();
+  TestConsume();
+  TestFindAndConsume();
+  TestQuoteMetaAll();
+  TestMatchNumberPeculiarity();
+
+  // Check the pattern() accessor
+  {
+    const string kPattern = "http://([^/]+)/.*";
+    const RE re(kPattern);
+    CHECK_EQ(kPattern, re.pattern());
+  }
+
+  // Check RE error field.
+  {
+    RE re("foo");
+    CHECK(re.error().empty());  // Must have no error
+  }
+
+#ifdef SUPPORT_UTF8
+  // Check UTF-8 handling
+  {
+    printf("Testing UTF-8 handling\n");
+
+    // Three Japanese characters (nihongo)
+    const unsigned char utf8_string[] = {
+         0xe6, 0x97, 0xa5, // 65e5
+         0xe6, 0x9c, 0xac, // 627c
+         0xe8, 0xaa, 0x9e, // 8a9e
+         0
+    };
+    const unsigned char utf8_pattern[] = {
+         '.',
+         0xe6, 0x9c, 0xac, // 627c
+         '.',
+         0
+    };
+
+    // Both should match in either mode, bytes or UTF-8
+    RE re_test1(".........");
+    CHECK(re_test1.FullMatch(utf8_string));
+    RE re_test2("...", pcrecpp::UTF8());
+    CHECK(re_test2.FullMatch(utf8_string));
+
+    // Check that '.' matches one byte or UTF-8 character
+    // according to the mode.
+    string ss;
+    RE re_test3("(.)");
+    CHECK(re_test3.PartialMatch(utf8_string, &ss));
+    CHECK_EQ(ss, string("\xe6"));
+    RE re_test4("(.)", pcrecpp::UTF8());
+    CHECK(re_test4.PartialMatch(utf8_string, &ss));
+    CHECK_EQ(ss, string("\xe6\x97\xa5"));
+
+    // Check that string matches itself in either mode
+    RE re_test5(utf8_string);
+    CHECK(re_test5.FullMatch(utf8_string));
+    RE re_test6(utf8_string, pcrecpp::UTF8());
+    CHECK(re_test6.FullMatch(utf8_string));
+
+    // Check that pattern matches string only in UTF8 mode
+    RE re_test7(utf8_pattern);
+    CHECK(!re_test7.FullMatch(utf8_string));
+    RE re_test8(utf8_pattern, pcrecpp::UTF8());
+    CHECK(re_test8.FullMatch(utf8_string));
+  }
+
+  // Check that ungreedy, UTF8 regular expressions don't match when they
+  // oughtn't -- see bug 82246.
+  {
+    // This code always worked.
+    const char* pattern = "\\w+X";
+    const string target = "a aX";
+    RE match_sentence(pattern);
+    RE match_sentence_re(pattern, pcrecpp::UTF8());
+
+    CHECK(!match_sentence.FullMatch(target));
+    CHECK(!match_sentence_re.FullMatch(target));
+  }
+
+  {
+    const char* pattern = "(?U)\\w+X";
+    const string target = "a aX";
+    RE match_sentence(pattern);
+    RE match_sentence_re(pattern, pcrecpp::UTF8());
+
+    CHECK(!match_sentence.FullMatch(target));
+    CHECK(!match_sentence_re.FullMatch(target));
+  }
+#endif  /* def SUPPORT_UTF8 */
+
+  printf("Testing error reporting\n");
+
+  { RE re("a\\1"); CHECK(!re.error().empty()); }
+  {
+    RE re("a[x");
+    CHECK(!re.error().empty());
+  }
+  {
+    RE re("a[z-a]");
+    CHECK(!re.error().empty());
+  }
+  {
+    RE re("a[[:foobar:]]");
+    CHECK(!re.error().empty());
+  }
+  {
+    RE re("a(b");
+    CHECK(!re.error().empty());
+  }
+  {
+    RE re("a\\");
+    CHECK(!re.error().empty());
+  }
+
+  // Test that recursion is stopped
+  TestRecursion();
+
+  // Test Options
+  if (getenv("VERBOSE_TEST") != NULL)
+    VERBOSE_TEST  = true;
+  TestOptions();
+
+  // Test the constructors
+  TestConstructors();
+
+  // Done
+  printf("OK\n");
+
+  return 0;
+}
diff --git a/pcre/pcrecpparg.h b/pcre/pcrecpparg.h
new file mode 100644
index 00000000000..b4f9c3f4989
--- /dev/null
+++ b/pcre/pcrecpparg.h
@@ -0,0 +1,174 @@
+// Copyright (c) 2005, Google Inc.
+// All rights reserved.
+//
+// Redistribution and use in source and binary forms, with or without
+// modification, are permitted provided that the following conditions are
+// met:
+//
+//     * Redistributions of source code must retain the above copyright
+// notice, this list of conditions and the following disclaimer.
+//     * Redistributions in binary form must reproduce the above
+// copyright notice, this list of conditions and the following disclaimer
+// in the documentation and/or other materials provided with the
+// distribution.
+//     * Neither the name of Google Inc. nor the names of its
+// contributors may be used to endorse or promote products derived from
+// this software without specific prior written permission.
+//
+// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+//
+// Author: Sanjay Ghemawat
+
+#ifndef _PCRECPPARG_H
+#define _PCRECPPARG_H
+
+#include     // for NULL
+#include 
+
+#include 
+
+namespace pcrecpp {
+
+class StringPiece;
+
+// Hex/Octal/Binary?
+
+// Special class for parsing into objects that define a ParseFrom() method
+template 
+class _RE_MatchObject {
+ public:
+  static inline bool Parse(const char* str, int n, void* dest) {
+    if (dest == NULL) return true;
+    T* object = reinterpret_cast(dest);
+    return object->ParseFrom(str, n);
+  }
+};
+
+class PCRECPP_EXP_DEFN Arg {
+ public:
+  // Empty constructor so we can declare arrays of Arg
+  Arg();
+
+  // Constructor specially designed for NULL arguments
+  Arg(void*);
+
+  typedef bool (*Parser)(const char* str, int n, void* dest);
+
+// Type-specific parsers
+#define PCRE_MAKE_PARSER(type,name)                             \
+  Arg(type* p) : arg_(p), parser_(name) { }                     \
+  Arg(type* p, Parser parser) : arg_(p), parser_(parser) { }
+
+
+  PCRE_MAKE_PARSER(char,               parse_char);
+  PCRE_MAKE_PARSER(unsigned char,      parse_uchar);
+  PCRE_MAKE_PARSER(short,              parse_short);
+  PCRE_MAKE_PARSER(unsigned short,     parse_ushort);
+  PCRE_MAKE_PARSER(int,                parse_int);
+  PCRE_MAKE_PARSER(unsigned int,       parse_uint);
+  PCRE_MAKE_PARSER(long,               parse_long);
+  PCRE_MAKE_PARSER(unsigned long,      parse_ulong);
+#if 1
+  PCRE_MAKE_PARSER(long long,          parse_longlong);
+#endif
+#if 1
+  PCRE_MAKE_PARSER(unsigned long long, parse_ulonglong);
+#endif
+  PCRE_MAKE_PARSER(float,              parse_float);
+  PCRE_MAKE_PARSER(double,             parse_double);
+  PCRE_MAKE_PARSER(std::string,        parse_string);
+  PCRE_MAKE_PARSER(StringPiece,        parse_stringpiece);
+
+#undef PCRE_MAKE_PARSER
+
+  // Generic constructor
+  template  Arg(T*, Parser parser);
+  // Generic constructor template
+  template  Arg(T* p)
+    : arg_(p), parser_(_RE_MatchObject::Parse) {
+  }
+
+  // Parse the data
+  bool Parse(const char* str, int n) const;
+
+ private:
+  void*         arg_;
+  Parser        parser_;
+
+  static bool parse_null          (const char* str, int n, void* dest);
+  static bool parse_char          (const char* str, int n, void* dest);
+  static bool parse_uchar         (const char* str, int n, void* dest);
+  static bool parse_float         (const char* str, int n, void* dest);
+  static bool parse_double        (const char* str, int n, void* dest);
+  static bool parse_string        (const char* str, int n, void* dest);
+  static bool parse_stringpiece   (const char* str, int n, void* dest);
+
+#define PCRE_DECLARE_INTEGER_PARSER(name)                                   \
+ private:                                                                   \
+  static bool parse_ ## name(const char* str, int n, void* dest);           \
+  static bool parse_ ## name ## _radix(                                     \
+    const char* str, int n, void* dest, int radix);                         \
+ public:                                                                    \
+  static bool parse_ ## name ## _hex(const char* str, int n, void* dest);   \
+  static bool parse_ ## name ## _octal(const char* str, int n, void* dest); \
+  static bool parse_ ## name ## _cradix(const char* str, int n, void* dest)
+
+  PCRE_DECLARE_INTEGER_PARSER(short);
+  PCRE_DECLARE_INTEGER_PARSER(ushort);
+  PCRE_DECLARE_INTEGER_PARSER(int);
+  PCRE_DECLARE_INTEGER_PARSER(uint);
+  PCRE_DECLARE_INTEGER_PARSER(long);
+  PCRE_DECLARE_INTEGER_PARSER(ulong);
+  PCRE_DECLARE_INTEGER_PARSER(longlong);
+  PCRE_DECLARE_INTEGER_PARSER(ulonglong);
+
+#undef PCRE_DECLARE_INTEGER_PARSER
+};
+
+inline Arg::Arg() : arg_(NULL), parser_(parse_null) { }
+inline Arg::Arg(void* p) : arg_(p), parser_(parse_null) { }
+
+inline bool Arg::Parse(const char* str, int n) const {
+  return (*parser_)(str, n, arg_);
+}
+
+// This part of the parser, appropriate only for ints, deals with bases
+#define MAKE_INTEGER_PARSER(type, name) \
+  inline Arg Hex(type* ptr) { \
+    return Arg(ptr, Arg::parse_ ## name ## _hex); } \
+  inline Arg Octal(type* ptr) { \
+    return Arg(ptr, Arg::parse_ ## name ## _octal); } \
+  inline Arg CRadix(type* ptr) { \
+    return Arg(ptr, Arg::parse_ ## name ## _cradix); }
+
+MAKE_INTEGER_PARSER(short,              short)     /*                        */
+MAKE_INTEGER_PARSER(unsigned short,     ushort)    /*                        */
+MAKE_INTEGER_PARSER(int,                int)       /* Don't use semicolons   */
+MAKE_INTEGER_PARSER(unsigned int,       uint)      /* after these statement  */
+MAKE_INTEGER_PARSER(long,               long)      /* because they can cause */
+MAKE_INTEGER_PARSER(unsigned long,      ulong)     /* compiler warnings if   */
+#if 1                          /* the checking level is  */
+MAKE_INTEGER_PARSER(long long,          longlong)  /* turned up high enough. */
+#endif                                             /*                        */
+#if 1                         /*                        */
+MAKE_INTEGER_PARSER(unsigned long long, ulonglong) /*                        */
+#endif
+
+#undef PCRE_IS_SET
+#undef PCRE_SET_OR_CLEAR
+#undef MAKE_INTEGER_PARSER
+
+}   // namespace pcrecpp
+
+
+#endif /* _PCRECPPARG_H */
diff --git a/pcre/pcrecpparg.h.in b/pcre/pcrecpparg.h.in
new file mode 100644
index 00000000000..61bcab5402c
--- /dev/null
+++ b/pcre/pcrecpparg.h.in
@@ -0,0 +1,174 @@
+// Copyright (c) 2005, Google Inc.
+// All rights reserved.
+//
+// Redistribution and use in source and binary forms, with or without
+// modification, are permitted provided that the following conditions are
+// met:
+//
+//     * Redistributions of source code must retain the above copyright
+// notice, this list of conditions and the following disclaimer.
+//     * Redistributions in binary form must reproduce the above
+// copyright notice, this list of conditions and the following disclaimer
+// in the documentation and/or other materials provided with the
+// distribution.
+//     * Neither the name of Google Inc. nor the names of its
+// contributors may be used to endorse or promote products derived from
+// this software without specific prior written permission.
+//
+// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+//
+// Author: Sanjay Ghemawat
+
+#ifndef _PCRECPPARG_H
+#define _PCRECPPARG_H
+
+#include     // for NULL
+#include 
+
+#include 
+
+namespace pcrecpp {
+
+class StringPiece;
+
+// Hex/Octal/Binary?
+
+// Special class for parsing into objects that define a ParseFrom() method
+template 
+class _RE_MatchObject {
+ public:
+  static inline bool Parse(const char* str, int n, void* dest) {
+    if (dest == NULL) return true;
+    T* object = reinterpret_cast(dest);
+    return object->ParseFrom(str, n);
+  }
+};
+
+class PCRECPP_EXP_DEFN Arg {
+ public:
+  // Empty constructor so we can declare arrays of Arg
+  Arg();
+
+  // Constructor specially designed for NULL arguments
+  Arg(void*);
+
+  typedef bool (*Parser)(const char* str, int n, void* dest);
+
+// Type-specific parsers
+#define PCRE_MAKE_PARSER(type,name)                             \
+  Arg(type* p) : arg_(p), parser_(name) { }                     \
+  Arg(type* p, Parser parser) : arg_(p), parser_(parser) { }
+
+
+  PCRE_MAKE_PARSER(char,               parse_char);
+  PCRE_MAKE_PARSER(unsigned char,      parse_uchar);
+  PCRE_MAKE_PARSER(short,              parse_short);
+  PCRE_MAKE_PARSER(unsigned short,     parse_ushort);
+  PCRE_MAKE_PARSER(int,                parse_int);
+  PCRE_MAKE_PARSER(unsigned int,       parse_uint);
+  PCRE_MAKE_PARSER(long,               parse_long);
+  PCRE_MAKE_PARSER(unsigned long,      parse_ulong);
+#if @pcre_have_long_long@
+  PCRE_MAKE_PARSER(long long,          parse_longlong);
+#endif
+#if @pcre_have_ulong_long@
+  PCRE_MAKE_PARSER(unsigned long long, parse_ulonglong);
+#endif
+  PCRE_MAKE_PARSER(float,              parse_float);
+  PCRE_MAKE_PARSER(double,             parse_double);
+  PCRE_MAKE_PARSER(std::string,        parse_string);
+  PCRE_MAKE_PARSER(StringPiece,        parse_stringpiece);
+
+#undef PCRE_MAKE_PARSER
+
+  // Generic constructor
+  template  Arg(T*, Parser parser);
+  // Generic constructor template
+  template  Arg(T* p)
+    : arg_(p), parser_(_RE_MatchObject::Parse) {
+  }
+
+  // Parse the data
+  bool Parse(const char* str, int n) const;
+
+ private:
+  void*         arg_;
+  Parser        parser_;
+
+  static bool parse_null          (const char* str, int n, void* dest);
+  static bool parse_char          (const char* str, int n, void* dest);
+  static bool parse_uchar         (const char* str, int n, void* dest);
+  static bool parse_float         (const char* str, int n, void* dest);
+  static bool parse_double        (const char* str, int n, void* dest);
+  static bool parse_string        (const char* str, int n, void* dest);
+  static bool parse_stringpiece   (const char* str, int n, void* dest);
+
+#define PCRE_DECLARE_INTEGER_PARSER(name)                                   \
+ private:                                                                   \
+  static bool parse_ ## name(const char* str, int n, void* dest);           \
+  static bool parse_ ## name ## _radix(                                     \
+    const char* str, int n, void* dest, int radix);                         \
+ public:                                                                    \
+  static bool parse_ ## name ## _hex(const char* str, int n, void* dest);   \
+  static bool parse_ ## name ## _octal(const char* str, int n, void* dest); \
+  static bool parse_ ## name ## _cradix(const char* str, int n, void* dest)
+
+  PCRE_DECLARE_INTEGER_PARSER(short);
+  PCRE_DECLARE_INTEGER_PARSER(ushort);
+  PCRE_DECLARE_INTEGER_PARSER(int);
+  PCRE_DECLARE_INTEGER_PARSER(uint);
+  PCRE_DECLARE_INTEGER_PARSER(long);
+  PCRE_DECLARE_INTEGER_PARSER(ulong);
+  PCRE_DECLARE_INTEGER_PARSER(longlong);
+  PCRE_DECLARE_INTEGER_PARSER(ulonglong);
+
+#undef PCRE_DECLARE_INTEGER_PARSER
+};
+
+inline Arg::Arg() : arg_(NULL), parser_(parse_null) { }
+inline Arg::Arg(void* p) : arg_(p), parser_(parse_null) { }
+
+inline bool Arg::Parse(const char* str, int n) const {
+  return (*parser_)(str, n, arg_);
+}
+
+// This part of the parser, appropriate only for ints, deals with bases
+#define MAKE_INTEGER_PARSER(type, name) \
+  inline Arg Hex(type* ptr) { \
+    return Arg(ptr, Arg::parse_ ## name ## _hex); } \
+  inline Arg Octal(type* ptr) { \
+    return Arg(ptr, Arg::parse_ ## name ## _octal); } \
+  inline Arg CRadix(type* ptr) { \
+    return Arg(ptr, Arg::parse_ ## name ## _cradix); }
+
+MAKE_INTEGER_PARSER(short,              short)     /*                        */
+MAKE_INTEGER_PARSER(unsigned short,     ushort)    /*                        */
+MAKE_INTEGER_PARSER(int,                int)       /* Don't use semicolons   */
+MAKE_INTEGER_PARSER(unsigned int,       uint)      /* after these statement  */
+MAKE_INTEGER_PARSER(long,               long)      /* because they can cause */
+MAKE_INTEGER_PARSER(unsigned long,      ulong)     /* compiler warnings if   */
+#if @pcre_have_long_long@                          /* the checking level is  */
+MAKE_INTEGER_PARSER(long long,          longlong)  /* turned up high enough. */
+#endif                                             /*                        */
+#if @pcre_have_ulong_long@                         /*                        */
+MAKE_INTEGER_PARSER(unsigned long long, ulonglong) /*                        */
+#endif
+
+#undef PCRE_IS_SET
+#undef PCRE_SET_OR_CLEAR
+#undef MAKE_INTEGER_PARSER
+
+}   // namespace pcrecpp
+
+
+#endif /* _PCRECPPARG_H */
diff --git a/pcre/pcredemo.c b/pcre/pcredemo.c
new file mode 100644
index 00000000000..946aba45cdc
--- /dev/null
+++ b/pcre/pcredemo.c
@@ -0,0 +1,406 @@
+/*************************************************
+*           PCRE DEMONSTRATION PROGRAM           *
+*************************************************/
+
+/* This is a demonstration program to illustrate the most straightforward ways
+of calling the PCRE regular expression library from a C program. See the
+pcresample documentation for a short discussion ("man pcresample" if you have
+the PCRE man pages installed).
+
+In Unix-like environments, if PCRE is installed in your standard system
+libraries, you should be able to compile this program using this command:
+
+gcc -Wall pcredemo.c -lpcre -o pcredemo
+
+If PCRE is not installed in a standard place, it is likely to be installed with
+support for the pkg-config mechanism. If you have pkg-config, you can compile
+this program using this command:
+
+gcc -Wall pcredemo.c `pkg-config --cflags --libs libpcre` -o pcredemo
+
+If you do not have pkg-config, you may have to use this:
+
+gcc -Wall pcredemo.c -I/usr/local/include -L/usr/local/lib \
+  -R/usr/local/lib -lpcre -o pcredemo
+
+Replace "/usr/local/include" and "/usr/local/lib" with wherever the include and
+library files for PCRE are installed on your system. Only some operating
+systems (e.g. Solaris) use the -R option.
+
+Building under Windows:
+
+If you want to statically link this program against a non-dll .a file, you must
+define PCRE_STATIC before including pcre.h, otherwise the pcre_malloc() and
+pcre_free() exported functions will be declared __declspec(dllimport), with
+unwanted results. So in this environment, uncomment the following line. */
+
+/* #define PCRE_STATIC */
+
+#include 
+#include 
+#include 
+
+#define OVECCOUNT 30    /* should be a multiple of 3 */
+
+
+int main(int argc, char **argv)
+{
+pcre *re;
+const char *error;
+char *pattern;
+char *subject;
+unsigned char *name_table;
+unsigned int option_bits;
+int erroffset;
+int find_all;
+int crlf_is_newline;
+int namecount;
+int name_entry_size;
+int ovector[OVECCOUNT];
+int subject_length;
+int rc, i;
+int utf8;
+
+
+/**************************************************************************
+* First, sort out the command line. There is only one possible option at  *
+* the moment, "-g" to request repeated matching to find all occurrences,  *
+* like Perl's /g option. We set the variable find_all to a non-zero value *
+* if the -g option is present. Apart from that, there must be exactly two *
+* arguments.                                                              *
+**************************************************************************/
+
+find_all = 0;
+for (i = 1; i < argc; i++)
+  {
+  if (strcmp(argv[i], "-g") == 0) find_all = 1;
+    else break;
+  }
+
+/* After the options, we require exactly two arguments, which are the pattern,
+and the subject string. */
+
+if (argc - i != 2)
+  {
+  printf("Two arguments required: a regex and a subject string\n");
+  return 1;
+  }
+
+pattern = argv[i];
+subject = argv[i+1];
+subject_length = (int)strlen(subject);
+
+
+/*************************************************************************
+* Now we are going to compile the regular expression pattern, and handle *
+* and errors that are detected.                                          *
+*************************************************************************/
+
+re = pcre_compile(
+  pattern,              /* the pattern */
+  0,                    /* default options */
+  &error,               /* for error message */
+  &erroffset,           /* for error offset */
+  NULL);                /* use default character tables */
+
+/* Compilation failed: print the error message and exit */
+
+if (re == NULL)
+  {
+  printf("PCRE compilation failed at offset %d: %s\n", erroffset, error);
+  return 1;
+  }
+
+
+/*************************************************************************
+* If the compilation succeeded, we call PCRE again, in order to do a     *
+* pattern match against the subject string. This does just ONE match. If *
+* further matching is needed, it will be done below.                     *
+*************************************************************************/
+
+rc = pcre_exec(
+  re,                   /* the compiled pattern */
+  NULL,                 /* no extra data - we didn't study the pattern */
+  subject,              /* the subject string */
+  subject_length,       /* the length of the subject */
+  0,                    /* start at offset 0 in the subject */
+  0,                    /* default options */
+  ovector,              /* output vector for substring information */
+  OVECCOUNT);           /* number of elements in the output vector */
+
+/* Matching failed: handle error cases */
+
+if (rc < 0)
+  {
+  switch(rc)
+    {
+    case PCRE_ERROR_NOMATCH: printf("No match\n"); break;
+    /*
+    Handle other special cases if you like
+    */
+    default: printf("Matching error %d\n", rc); break;
+    }
+  pcre_free(re);     /* Release memory used for the compiled pattern */
+  return 1;
+  }
+
+/* Match succeded */
+
+printf("\nMatch succeeded at offset %d\n", ovector[0]);
+
+
+/*************************************************************************
+* We have found the first match within the subject string. If the output *
+* vector wasn't big enough, say so. Then output any substrings that were *
+* captured.                                                              *
+*************************************************************************/
+
+/* The output vector wasn't big enough */
+
+if (rc == 0)
+  {
+  rc = OVECCOUNT/3;
+  printf("ovector only has room for %d captured substrings\n", rc - 1);
+  }
+
+/* Show substrings stored in the output vector by number. Obviously, in a real
+application you might want to do things other than print them. */
+
+for (i = 0; i < rc; i++)
+  {
+  char *substring_start = subject + ovector[2*i];
+  int substring_length = ovector[2*i+1] - ovector[2*i];
+  printf("%2d: %.*s\n", i, substring_length, substring_start);
+  }
+
+
+/**************************************************************************
+* That concludes the basic part of this demonstration program. We have    *
+* compiled a pattern, and performed a single match. The code that follows *
+* shows first how to access named substrings, and then how to code for    *
+* repeated matches on the same subject.                                   *
+**************************************************************************/
+
+/* See if there are any named substrings, and if so, show them by name. First
+we have to extract the count of named parentheses from the pattern. */
+
+(void)pcre_fullinfo(
+  re,                   /* the compiled pattern */
+  NULL,                 /* no extra data - we didn't study the pattern */
+  PCRE_INFO_NAMECOUNT,  /* number of named substrings */
+  &namecount);          /* where to put the answer */
+
+if (namecount <= 0) printf("No named substrings\n"); else
+  {
+  unsigned char *tabptr;
+  printf("Named substrings\n");
+
+  /* Before we can access the substrings, we must extract the table for
+  translating names to numbers, and the size of each entry in the table. */
+
+  (void)pcre_fullinfo(
+    re,                       /* the compiled pattern */
+    NULL,                     /* no extra data - we didn't study the pattern */
+    PCRE_INFO_NAMETABLE,      /* address of the table */
+    &name_table);             /* where to put the answer */
+
+  (void)pcre_fullinfo(
+    re,                       /* the compiled pattern */
+    NULL,                     /* no extra data - we didn't study the pattern */
+    PCRE_INFO_NAMEENTRYSIZE,  /* size of each entry in the table */
+    &name_entry_size);        /* where to put the answer */
+
+  /* Now we can scan the table and, for each entry, print the number, the name,
+  and the substring itself. */
+
+  tabptr = name_table;
+  for (i = 0; i < namecount; i++)
+    {
+    int n = (tabptr[0] << 8) | tabptr[1];
+    printf("(%d) %*s: %.*s\n", n, name_entry_size - 3, tabptr + 2,
+      ovector[2*n+1] - ovector[2*n], subject + ovector[2*n]);
+    tabptr += name_entry_size;
+    }
+  }
+
+
+/*************************************************************************
+* If the "-g" option was given on the command line, we want to continue  *
+* to search for additional matches in the subject string, in a similar   *
+* way to the /g option in Perl. This turns out to be trickier than you   *
+* might think because of the possibility of matching an empty string.    *
+* What happens is as follows:                                            *
+*                                                                        *
+* If the previous match was NOT for an empty string, we can just start   *
+* the next match at the end of the previous one.                         *
+*                                                                        *
+* If the previous match WAS for an empty string, we can't do that, as it *
+* would lead to an infinite loop. Instead, a special call of pcre_exec() *
+* is made with the PCRE_NOTEMPTY_ATSTART and PCRE_ANCHORED flags set.    *
+* The first of these tells PCRE that an empty string at the start of the *
+* subject is not a valid match; other possibilities must be tried. The   *
+* second flag restricts PCRE to one match attempt at the initial string  *
+* position. If this match succeeds, an alternative to the empty string   *
+* match has been found, and we can print it and proceed round the loop,  *
+* advancing by the length of whatever was found. If this match does not  *
+* succeed, we still stay in the loop, advancing by just one character.   *
+* In UTF-8 mode, which can be set by (*UTF8) in the pattern, this may be *
+* more than one byte.                                                    *
+*                                                                        *
+* However, there is a complication concerned with newlines. When the     *
+* newline convention is such that CRLF is a valid newline, we must       *
+* advance by two characters rather than one. The newline convention can  *
+* be set in the regex by (*CR), etc.; if not, we must find the default.  *
+*************************************************************************/
+
+if (!find_all)     /* Check for -g */
+  {
+  pcre_free(re);   /* Release the memory used for the compiled pattern */
+  return 0;        /* Finish unless -g was given */
+  }
+
+/* Before running the loop, check for UTF-8 and whether CRLF is a valid newline
+sequence. First, find the options with which the regex was compiled; extract
+the UTF-8 state, and mask off all but the newline options. */
+
+(void)pcre_fullinfo(re, NULL, PCRE_INFO_OPTIONS, &option_bits);
+utf8 = option_bits & PCRE_UTF8;
+option_bits &= PCRE_NEWLINE_CR|PCRE_NEWLINE_LF|PCRE_NEWLINE_CRLF|
+               PCRE_NEWLINE_ANY|PCRE_NEWLINE_ANYCRLF;
+
+/* If no newline options were set, find the default newline convention from the
+build configuration. */
+
+if (option_bits == 0)
+  {
+  int d;
+  (void)pcre_config(PCRE_CONFIG_NEWLINE, &d);
+  /* Note that these values are always the ASCII ones, even in
+  EBCDIC environments. CR = 13, NL = 10. */
+  option_bits = (d == 13)? PCRE_NEWLINE_CR :
+          (d == 10)? PCRE_NEWLINE_LF :
+          (d == (13<<8 | 10))? PCRE_NEWLINE_CRLF :
+          (d == -2)? PCRE_NEWLINE_ANYCRLF :
+          (d == -1)? PCRE_NEWLINE_ANY : 0;
+  }
+
+/* See if CRLF is a valid newline sequence. */
+
+crlf_is_newline =
+     option_bits == PCRE_NEWLINE_ANY ||
+     option_bits == PCRE_NEWLINE_CRLF ||
+     option_bits == PCRE_NEWLINE_ANYCRLF;
+
+/* Loop for second and subsequent matches */
+
+for (;;)
+  {
+  int options = 0;                 /* Normally no options */
+  int start_offset = ovector[1];   /* Start at end of previous match */
+
+  /* If the previous match was for an empty string, we are finished if we are
+  at the end of the subject. Otherwise, arrange to run another match at the
+  same point to see if a non-empty match can be found. */
+
+  if (ovector[0] == ovector[1])
+    {
+    if (ovector[0] == subject_length) break;
+    options = PCRE_NOTEMPTY_ATSTART | PCRE_ANCHORED;
+    }
+
+  /* Run the next matching operation */
+
+  rc = pcre_exec(
+    re,                   /* the compiled pattern */
+    NULL,                 /* no extra data - we didn't study the pattern */
+    subject,              /* the subject string */
+    subject_length,       /* the length of the subject */
+    start_offset,         /* starting offset in the subject */
+    options,              /* options */
+    ovector,              /* output vector for substring information */
+    OVECCOUNT);           /* number of elements in the output vector */
+
+  /* This time, a result of NOMATCH isn't an error. If the value in "options"
+  is zero, it just means we have found all possible matches, so the loop ends.
+  Otherwise, it means we have failed to find a non-empty-string match at a
+  point where there was a previous empty-string match. In this case, we do what
+  Perl does: advance the matching position by one character, and continue. We
+  do this by setting the "end of previous match" offset, because that is picked
+  up at the top of the loop as the point at which to start again.
+
+  There are two complications: (a) When CRLF is a valid newline sequence, and
+  the current position is just before it, advance by an extra byte. (b)
+  Otherwise we must ensure that we skip an entire UTF-8 character if we are in
+  UTF-8 mode. */
+
+  if (rc == PCRE_ERROR_NOMATCH)
+    {
+    if (options == 0) break;                    /* All matches found */
+    ovector[1] = start_offset + 1;              /* Advance one byte */
+    if (crlf_is_newline &&                      /* If CRLF is newline & */
+        start_offset < subject_length - 1 &&    /* we are at CRLF, */
+        subject[start_offset] == '\r' &&
+        subject[start_offset + 1] == '\n')
+      ovector[1] += 1;                          /* Advance by one more. */
+    else if (utf8)                              /* Otherwise, ensure we */
+      {                                         /* advance a whole UTF-8 */
+      while (ovector[1] < subject_length)       /* character. */
+        {
+        if ((subject[ovector[1]] & 0xc0) != 0x80) break;
+        ovector[1] += 1;
+        }
+      }
+    continue;    /* Go round the loop again */
+    }
+
+  /* Other matching errors are not recoverable. */
+
+  if (rc < 0)
+    {
+    printf("Matching error %d\n", rc);
+    pcre_free(re);    /* Release memory used for the compiled pattern */
+    return 1;
+    }
+
+  /* Match succeded */
+
+  printf("\nMatch succeeded again at offset %d\n", ovector[0]);
+
+  /* The match succeeded, but the output vector wasn't big enough. */
+
+  if (rc == 0)
+    {
+    rc = OVECCOUNT/3;
+    printf("ovector only has room for %d captured substrings\n", rc - 1);
+    }
+
+  /* As before, show substrings stored in the output vector by number, and then
+  also any named substrings. */
+
+  for (i = 0; i < rc; i++)
+    {
+    char *substring_start = subject + ovector[2*i];
+    int substring_length = ovector[2*i+1] - ovector[2*i];
+    printf("%2d: %.*s\n", i, substring_length, substring_start);
+    }
+
+  if (namecount <= 0) printf("No named substrings\n"); else
+    {
+    unsigned char *tabptr = name_table;
+    printf("Named substrings\n");
+    for (i = 0; i < namecount; i++)
+      {
+      int n = (tabptr[0] << 8) | tabptr[1];
+      printf("(%d) %*s: %.*s\n", n, name_entry_size - 3, tabptr + 2,
+        ovector[2*n+1] - ovector[2*n], subject + ovector[2*n]);
+      tabptr += name_entry_size;
+      }
+    }
+  }      /* End of loop to find second and subsequent matches */
+
+printf("\n");
+pcre_free(re);       /* Release memory used for the compiled pattern */
+return 0;
+}
+
+/* End of pcredemo.c */
diff --git a/pcre/pcregrep.c b/pcre/pcregrep.c
new file mode 100644
index 00000000000..64023483d2d
--- /dev/null
+++ b/pcre/pcregrep.c
@@ -0,0 +1,3186 @@
+/*************************************************
+*               pcregrep program                 *
+*************************************************/
+
+/* This is a grep program that uses the PCRE regular expression library to do
+its pattern matching. On a Unix or Win32 system it can recurse into
+directories.
+
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+
+#include 
+#include 
+
+#ifdef HAVE_UNISTD_H
+#include 
+#endif
+
+#ifdef SUPPORT_LIBZ
+#include 
+#endif
+
+#ifdef SUPPORT_LIBBZ2
+#include 
+#endif
+
+#include "pcre.h"
+
+#define FALSE 0
+#define TRUE 1
+
+typedef int BOOL;
+
+#define OFFSET_SIZE 99
+
+#if BUFSIZ > 8192
+#define MAXPATLEN BUFSIZ
+#else
+#define MAXPATLEN 8192
+#endif
+
+#define PATBUFSIZE (MAXPATLEN + 10)   /* Allows for prefix+suffix */
+
+/* Values for the "filenames" variable, which specifies options for file name
+output. The order is important; it is assumed that a file name is wanted for
+all values greater than FN_DEFAULT. */
+
+enum { FN_NONE, FN_DEFAULT, FN_MATCH_ONLY, FN_NOMATCH_ONLY, FN_FORCE };
+
+/* File reading styles */
+
+enum { FR_PLAIN, FR_LIBZ, FR_LIBBZ2 };
+
+/* Actions for the -d and -D options */
+
+enum { dee_READ, dee_SKIP, dee_RECURSE };
+enum { DEE_READ, DEE_SKIP };
+
+/* Actions for special processing options (flag bits) */
+
+#define PO_WORD_MATCH     0x0001
+#define PO_LINE_MATCH     0x0002
+#define PO_FIXED_STRINGS  0x0004
+
+/* Line ending types */
+
+enum { EL_LF, EL_CR, EL_CRLF, EL_ANY, EL_ANYCRLF };
+
+/* Binary file options */
+
+enum { BIN_BINARY, BIN_NOMATCH, BIN_TEXT };
+
+/* In newer versions of gcc, with FORTIFY_SOURCE set (the default in some
+environments), a warning is issued if the value of fwrite() is ignored.
+Unfortunately, casting to (void) does not suppress the warning. To get round
+this, we use a macro that compiles a fudge. Oddly, this does not also seem to
+apply to fprintf(). */
+
+#define FWRITE(a,b,c,d) if (fwrite(a,b,c,d)) {}
+
+
+
+/*************************************************
+*               Global variables                 *
+*************************************************/
+
+/* Jeffrey Friedl has some debugging requirements that are not part of the
+regular code. */
+
+#ifdef JFRIEDL_DEBUG
+static int S_arg = -1;
+static unsigned int jfriedl_XR = 0; /* repeat regex attempt this many times */
+static unsigned int jfriedl_XT = 0; /* replicate text this many times */
+static const char *jfriedl_prefix = "";
+static const char *jfriedl_postfix = "";
+#endif
+
+static int  endlinetype;
+
+static char *colour_string = (char *)"1;31";
+static char *colour_option = NULL;
+static char *dee_option = NULL;
+static char *DEE_option = NULL;
+static char *locale = NULL;
+static char *main_buffer = NULL;
+static char *newline = NULL;
+static char *om_separator = (char *)"";
+static char *stdin_name = (char *)"(standard input)";
+
+static const unsigned char *pcretables = NULL;
+
+static int after_context = 0;
+static int before_context = 0;
+static int binary_files = BIN_BINARY;
+static int both_context = 0;
+static int bufthird = PCREGREP_BUFSIZE;
+static int bufsize = 3*PCREGREP_BUFSIZE;
+
+#if defined HAVE_WINDOWS_H && HAVE_WINDOWS_H
+static int dee_action = dee_SKIP;
+#else
+static int dee_action = dee_READ;
+#endif
+
+static int DEE_action = DEE_READ;
+static int error_count = 0;
+static int filenames = FN_DEFAULT;
+static int pcre_options = 0;
+static int process_options = 0;
+
+#ifdef SUPPORT_PCREGREP_JIT
+static int study_options = PCRE_STUDY_JIT_COMPILE;
+#else
+static int study_options = 0;
+#endif
+
+static unsigned long int match_limit = 0;
+static unsigned long int match_limit_recursion = 0;
+
+static BOOL count_only = FALSE;
+static BOOL do_colour = FALSE;
+static BOOL file_offsets = FALSE;
+static BOOL hyphenpending = FALSE;
+static BOOL invert = FALSE;
+static BOOL line_buffered = FALSE;
+static BOOL line_offsets = FALSE;
+static BOOL multiline = FALSE;
+static BOOL number = FALSE;
+static BOOL omit_zero_count = FALSE;
+static BOOL resource_error = FALSE;
+static BOOL quiet = FALSE;
+static BOOL show_only_matching = FALSE;
+static BOOL silent = FALSE;
+static BOOL utf8 = FALSE;
+
+/* Structure for list of --only-matching capturing numbers. */
+
+typedef struct omstr {
+  struct omstr *next;
+  int groupnum;
+} omstr;
+
+static omstr *only_matching = NULL;
+static omstr *only_matching_last = NULL;
+
+/* Structure for holding the two variables that describe a number chain. */
+
+typedef struct omdatastr {
+  omstr **anchor;
+  omstr **lastptr;
+} omdatastr;
+
+static omdatastr only_matching_data = { &only_matching, &only_matching_last };
+
+/* Structure for list of file names (for -f and --{in,ex}clude-from) */
+
+typedef struct fnstr {
+  struct fnstr *next;
+  char *name;
+} fnstr;
+
+static fnstr *exclude_from = NULL;
+static fnstr *exclude_from_last = NULL;
+static fnstr *include_from = NULL;
+static fnstr *include_from_last = NULL;
+
+static fnstr *file_lists = NULL;
+static fnstr *file_lists_last = NULL;
+static fnstr *pattern_files = NULL;
+static fnstr *pattern_files_last = NULL;
+
+/* Structure for holding the two variables that describe a file name chain. */
+
+typedef struct fndatastr {
+  fnstr **anchor;
+  fnstr **lastptr;
+} fndatastr;
+
+static fndatastr exclude_from_data = { &exclude_from, &exclude_from_last };
+static fndatastr include_from_data = { &include_from, &include_from_last };
+static fndatastr file_lists_data = { &file_lists, &file_lists_last };
+static fndatastr pattern_files_data = { &pattern_files, &pattern_files_last };
+
+/* Structure for pattern and its compiled form; used for matching patterns and
+also for include/exclude patterns. */
+
+typedef struct patstr {
+  struct patstr *next;
+  char *string;
+  pcre *compiled;
+  pcre_extra *hint;
+} patstr;
+
+static patstr *patterns = NULL;
+static patstr *patterns_last = NULL;
+static patstr *include_patterns = NULL;
+static patstr *include_patterns_last = NULL;
+static patstr *exclude_patterns = NULL;
+static patstr *exclude_patterns_last = NULL;
+static patstr *include_dir_patterns = NULL;
+static patstr *include_dir_patterns_last = NULL;
+static patstr *exclude_dir_patterns = NULL;
+static patstr *exclude_dir_patterns_last = NULL;
+
+/* Structure holding the two variables that describe a pattern chain. A pointer
+to such structures is used for each appropriate option. */
+
+typedef struct patdatastr {
+  patstr **anchor;
+  patstr **lastptr;
+} patdatastr;
+
+static patdatastr match_patdata = { &patterns, &patterns_last };
+static patdatastr include_patdata = { &include_patterns, &include_patterns_last };
+static patdatastr exclude_patdata = { &exclude_patterns, &exclude_patterns_last };
+static patdatastr include_dir_patdata = { &include_dir_patterns, &include_dir_patterns_last };
+static patdatastr exclude_dir_patdata = { &exclude_dir_patterns, &exclude_dir_patterns_last };
+
+static patstr **incexlist[4] = { &include_patterns, &exclude_patterns,
+                                 &include_dir_patterns, &exclude_dir_patterns };
+
+static const char *incexname[4] = { "--include", "--exclude",
+                                    "--include-dir", "--exclude-dir" };
+
+/* Structure for options and list of them */
+
+enum { OP_NODATA, OP_STRING, OP_OP_STRING, OP_NUMBER, OP_LONGNUMBER,
+       OP_OP_NUMBER, OP_OP_NUMBERS, OP_PATLIST, OP_FILELIST, OP_BINFILES };
+
+typedef struct option_item {
+  int type;
+  int one_char;
+  void *dataptr;
+  const char *long_name;
+  const char *help_text;
+} option_item;
+
+/* Options without a single-letter equivalent get a negative value. This can be
+used to identify them. */
+
+#define N_COLOUR       (-1)
+#define N_EXCLUDE      (-2)
+#define N_EXCLUDE_DIR  (-3)
+#define N_HELP         (-4)
+#define N_INCLUDE      (-5)
+#define N_INCLUDE_DIR  (-6)
+#define N_LABEL        (-7)
+#define N_LOCALE       (-8)
+#define N_NULL         (-9)
+#define N_LOFFSETS     (-10)
+#define N_FOFFSETS     (-11)
+#define N_LBUFFER      (-12)
+#define N_M_LIMIT      (-13)
+#define N_M_LIMIT_REC  (-14)
+#define N_BUFSIZE      (-15)
+#define N_NOJIT        (-16)
+#define N_FILE_LIST    (-17)
+#define N_BINARY_FILES (-18)
+#define N_EXCLUDE_FROM (-19)
+#define N_INCLUDE_FROM (-20)
+#define N_OM_SEPARATOR (-21)
+
+static option_item optionlist[] = {
+  { OP_NODATA,     N_NULL,   NULL,              "",              "terminate options" },
+  { OP_NODATA,     N_HELP,   NULL,              "help",          "display this help and exit" },
+  { OP_NUMBER,     'A',      &after_context,    "after-context=number", "set number of following context lines" },
+  { OP_NODATA,     'a',      NULL,              "text",          "treat binary files as text" },
+  { OP_NUMBER,     'B',      &before_context,   "before-context=number", "set number of prior context lines" },
+  { OP_BINFILES,   N_BINARY_FILES, NULL,        "binary-files=word", "set treatment of binary files" },
+  { OP_NUMBER,     N_BUFSIZE,&bufthird,         "buffer-size=number", "set processing buffer size parameter" },
+  { OP_OP_STRING,  N_COLOUR, &colour_option,    "color=option",  "matched text color option" },
+  { OP_OP_STRING,  N_COLOUR, &colour_option,    "colour=option", "matched text colour option" },
+  { OP_NUMBER,     'C',      &both_context,     "context=number", "set number of context lines, before & after" },
+  { OP_NODATA,     'c',      NULL,              "count",         "print only a count of matching lines per FILE" },
+  { OP_STRING,     'D',      &DEE_option,       "devices=action","how to handle devices, FIFOs, and sockets" },
+  { OP_STRING,     'd',      &dee_option,       "directories=action", "how to handle directories" },
+  { OP_PATLIST,    'e',      &match_patdata,    "regex(p)=pattern", "specify pattern (may be used more than once)" },
+  { OP_NODATA,     'F',      NULL,              "fixed-strings", "patterns are sets of newline-separated strings" },
+  { OP_FILELIST,   'f',      &pattern_files_data, "file=path",   "read patterns from file" },
+  { OP_FILELIST,   N_FILE_LIST, &file_lists_data, "file-list=path","read files to search from file" },
+  { OP_NODATA,     N_FOFFSETS, NULL,            "file-offsets",  "output file offsets, not text" },
+  { OP_NODATA,     'H',      NULL,              "with-filename", "force the prefixing filename on output" },
+  { OP_NODATA,     'h',      NULL,              "no-filename",   "suppress the prefixing filename on output" },
+  { OP_NODATA,     'I',      NULL,              "",              "treat binary files as not matching (ignore)" },
+  { OP_NODATA,     'i',      NULL,              "ignore-case",   "ignore case distinctions" },
+#ifdef SUPPORT_PCREGREP_JIT
+  { OP_NODATA,     N_NOJIT,  NULL,              "no-jit",        "do not use just-in-time compiler optimization" },
+#else
+  { OP_NODATA,     N_NOJIT,  NULL,              "no-jit",        "ignored: this pcregrep does not support JIT" },
+#endif
+  { OP_NODATA,     'l',      NULL,              "files-with-matches", "print only FILE names containing matches" },
+  { OP_NODATA,     'L',      NULL,              "files-without-match","print only FILE names not containing matches" },
+  { OP_STRING,     N_LABEL,  &stdin_name,       "label=name",    "set name for standard input" },
+  { OP_NODATA,     N_LBUFFER, NULL,             "line-buffered", "use line buffering" },
+  { OP_NODATA,     N_LOFFSETS, NULL,            "line-offsets",  "output line numbers and offsets, not text" },
+  { OP_STRING,     N_LOCALE, &locale,           "locale=locale", "use the named locale" },
+  { OP_LONGNUMBER, N_M_LIMIT, &match_limit,     "match-limit=number", "set PCRE match limit option" },
+  { OP_LONGNUMBER, N_M_LIMIT_REC, &match_limit_recursion, "recursion-limit=number", "set PCRE match recursion limit option" },
+  { OP_NODATA,     'M',      NULL,              "multiline",     "run in multiline mode" },
+  { OP_STRING,     'N',      &newline,          "newline=type",  "set newline type (CR, LF, CRLF, ANYCRLF or ANY)" },
+  { OP_NODATA,     'n',      NULL,              "line-number",   "print line number with output lines" },
+  { OP_OP_NUMBERS, 'o',      &only_matching_data, "only-matching=n", "show only the part of the line that matched" },
+  { OP_STRING,     N_OM_SEPARATOR, &om_separator, "om-separator=text", "set separator for multiple -o output" },
+  { OP_NODATA,     'q',      NULL,              "quiet",         "suppress output, just set return code" },
+  { OP_NODATA,     'r',      NULL,              "recursive",     "recursively scan sub-directories" },
+  { OP_PATLIST,    N_EXCLUDE,&exclude_patdata,  "exclude=pattern","exclude matching files when recursing" },
+  { OP_PATLIST,    N_INCLUDE,&include_patdata,  "include=pattern","include matching files when recursing" },
+  { OP_PATLIST,    N_EXCLUDE_DIR,&exclude_dir_patdata, "exclude-dir=pattern","exclude matching directories when recursing" },
+  { OP_PATLIST,    N_INCLUDE_DIR,&include_dir_patdata, "include-dir=pattern","include matching directories when recursing" },
+  { OP_FILELIST,   N_EXCLUDE_FROM,&exclude_from_data, "exclude-from=path", "read exclude list from file" },
+  { OP_FILELIST,   N_INCLUDE_FROM,&include_from_data, "include-from=path", "read include list from file" },
+
+  /* These two were accidentally implemented with underscores instead of
+  hyphens in the option names. As this was not discovered for several releases,
+  the incorrect versions are left in the table for compatibility. However, the
+  --help function misses out any option that has an underscore in its name. */
+
+  { OP_PATLIST,   N_EXCLUDE_DIR,&exclude_dir_patdata, "exclude_dir=pattern","exclude matching directories when recursing" },
+  { OP_PATLIST,   N_INCLUDE_DIR,&include_dir_patdata, "include_dir=pattern","include matching directories when recursing" },
+
+#ifdef JFRIEDL_DEBUG
+  { OP_OP_NUMBER, 'S',      &S_arg,            "jeffS",         "replace matched (sub)string with X" },
+#endif
+  { OP_NODATA,    's',      NULL,              "no-messages",   "suppress error messages" },
+  { OP_NODATA,    'u',      NULL,              "utf-8",         "use UTF-8 mode" },
+  { OP_NODATA,    'V',      NULL,              "version",       "print version information and exit" },
+  { OP_NODATA,    'v',      NULL,              "invert-match",  "select non-matching lines" },
+  { OP_NODATA,    'w',      NULL,              "word-regex(p)", "force patterns to match only as words"  },
+  { OP_NODATA,    'x',      NULL,              "line-regex(p)", "force patterns to match only whole lines" },
+  { OP_NODATA,    0,        NULL,               NULL,            NULL }
+};
+
+/* Tables for prefixing and suffixing patterns, according to the -w, -x, and -F
+options. These set the 1, 2, and 4 bits in process_options, respectively. Note
+that the combination of -w and -x has the same effect as -x on its own, so we
+can treat them as the same. Note that the MAXPATLEN macro assumes the longest
+prefix+suffix is 10 characters; if anything longer is added, it must be
+adjusted. */
+
+static const char *prefix[] = {
+  "", "\\b", "^(?:", "^(?:", "\\Q", "\\b\\Q", "^(?:\\Q", "^(?:\\Q" };
+
+static const char *suffix[] = {
+  "", "\\b", ")$",   ")$",   "\\E", "\\E\\b", "\\E)$",   "\\E)$" };
+
+/* UTF-8 tables - used only when the newline setting is "any". */
+
+const int utf8_table3[] = { 0xff, 0x1f, 0x0f, 0x07, 0x03, 0x01};
+
+const char utf8_table4[] = {
+  1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
+  1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
+  2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,
+  3,3,3,3,3,3,3,3,4,4,4,4,5,5,5,5 };
+
+
+
+/*************************************************
+*         Exit from the program                  *
+*************************************************/
+
+/* If there has been a resource error, give a suitable message.
+
+Argument:  the return code
+Returns:   does not return
+*/
+
+static void
+pcregrep_exit(int rc)
+{
+if (resource_error)
+  {
+  fprintf(stderr, "pcregrep: Error %d, %d or %d means that a resource limit "
+    "was exceeded.\n", PCRE_ERROR_MATCHLIMIT, PCRE_ERROR_RECURSIONLIMIT,
+    PCRE_ERROR_JIT_STACKLIMIT);
+  fprintf(stderr, "pcregrep: Check your regex for nested unlimited loops.\n");
+  }
+exit(rc);
+}
+
+
+/*************************************************
+*          Add item to chain of patterns         *
+*************************************************/
+
+/* Used to add an item onto a chain, or just return an unconnected item if the
+"after" argument is NULL.
+
+Arguments:
+  s          pattern string to add
+  after      if not NULL points to item to insert after
+
+Returns:     new pattern block
+*/
+
+static patstr *
+add_pattern(char *s, patstr *after)
+{
+patstr *p = (patstr *)malloc(sizeof(patstr));
+if (p == NULL)
+  {
+  fprintf(stderr, "pcregrep: malloc failed\n");
+  pcregrep_exit(2);
+  }
+if (strlen(s) > MAXPATLEN)
+  {
+  fprintf(stderr, "pcregrep: pattern is too long (limit is %d bytes)\n",
+    MAXPATLEN);
+  return NULL;
+  }
+p->next = NULL;
+p->string = s;
+p->compiled = NULL;
+p->hint = NULL;
+
+if (after != NULL)
+  {
+  p->next = after->next;
+  after->next = p;
+  }
+return p;
+}
+
+
+/*************************************************
+*           Free chain of patterns               *
+*************************************************/
+
+/* Used for several chains of patterns.
+
+Argument: pointer to start of chain
+Returns:  nothing
+*/
+
+static void
+free_pattern_chain(patstr *pc)
+{
+while (pc != NULL)
+  {
+  patstr *p = pc;
+  pc = p->next;
+  if (p->hint != NULL) pcre_free_study(p->hint);
+  if (p->compiled != NULL) pcre_free(p->compiled);
+  free(p);
+  }
+}
+
+
+/*************************************************
+*           Free chain of file names             *
+*************************************************/
+
+/*
+Argument: pointer to start of chain
+Returns:  nothing
+*/
+
+static void
+free_file_chain(fnstr *fn)
+{
+while (fn != NULL)
+  {
+  fnstr *f = fn;
+  fn = f->next;
+  free(f);
+  }
+}
+
+
+/*************************************************
+*            OS-specific functions               *
+*************************************************/
+
+/* These functions are defined so that they can be made system specific,
+although at present the only ones are for Unix, Win32, and for "no support". */
+
+
+/************* Directory scanning in Unix ***********/
+
+#if defined HAVE_SYS_STAT_H && defined HAVE_DIRENT_H && defined HAVE_SYS_TYPES_H
+#include 
+#include 
+#include 
+
+typedef DIR directory_type;
+#define FILESEP '/'
+
+static int
+isdirectory(char *filename)
+{
+struct stat statbuf;
+if (stat(filename, &statbuf) < 0)
+  return 0;        /* In the expectation that opening as a file will fail */
+return (statbuf.st_mode & S_IFMT) == S_IFDIR;
+}
+
+static directory_type *
+opendirectory(char *filename)
+{
+return opendir(filename);
+}
+
+static char *
+readdirectory(directory_type *dir)
+{
+for (;;)
+  {
+  struct dirent *dent = readdir(dir);
+  if (dent == NULL) return NULL;
+  if (strcmp(dent->d_name, ".") != 0 && strcmp(dent->d_name, "..") != 0)
+    return dent->d_name;
+  }
+/* Control never reaches here */
+}
+
+static void
+closedirectory(directory_type *dir)
+{
+closedir(dir);
+}
+
+
+/************* Test for regular file in Unix **********/
+
+static int
+isregfile(char *filename)
+{
+struct stat statbuf;
+if (stat(filename, &statbuf) < 0)
+  return 1;        /* In the expectation that opening as a file will fail */
+return (statbuf.st_mode & S_IFMT) == S_IFREG;
+}
+
+
+/************* Test for a terminal in Unix **********/
+
+static BOOL
+is_stdout_tty(void)
+{
+return isatty(fileno(stdout));
+}
+
+static BOOL
+is_file_tty(FILE *f)
+{
+return isatty(fileno(f));
+}
+
+
+/************* Directory scanning in Win32 ***********/
+
+/* I (Philip Hazel) have no means of testing this code. It was contributed by
+Lionel Fourquaux. David Burgess added a patch to define INVALID_FILE_ATTRIBUTES
+when it did not exist. David Byron added a patch that moved the #include of
+ to before the INVALID_FILE_ATTRIBUTES definition rather than after.
+The double test below stops gcc 4.4.4 grumbling that HAVE_WINDOWS_H is
+undefined when it is indeed undefined. */
+
+#elif defined HAVE_WINDOWS_H && HAVE_WINDOWS_H
+
+#ifndef STRICT
+# define STRICT
+#endif
+#ifndef WIN32_LEAN_AND_MEAN
+# define WIN32_LEAN_AND_MEAN
+#endif
+
+#include 
+
+#ifndef INVALID_FILE_ATTRIBUTES
+#define INVALID_FILE_ATTRIBUTES 0xFFFFFFFF
+#endif
+
+typedef struct directory_type
+{
+HANDLE handle;
+BOOL first;
+WIN32_FIND_DATA data;
+} directory_type;
+
+#define FILESEP '/'
+
+int
+isdirectory(char *filename)
+{
+DWORD attr = GetFileAttributes(filename);
+if (attr == INVALID_FILE_ATTRIBUTES)
+  return 0;
+return (attr & FILE_ATTRIBUTE_DIRECTORY) != 0;
+}
+
+directory_type *
+opendirectory(char *filename)
+{
+size_t len;
+char *pattern;
+directory_type *dir;
+DWORD err;
+len = strlen(filename);
+pattern = (char *)malloc(len + 3);
+dir = (directory_type *)malloc(sizeof(*dir));
+if ((pattern == NULL) || (dir == NULL))
+  {
+  fprintf(stderr, "pcregrep: malloc failed\n");
+  pcregrep_exit(2);
+  }
+memcpy(pattern, filename, len);
+memcpy(&(pattern[len]), "\\*", 3);
+dir->handle = FindFirstFile(pattern, &(dir->data));
+if (dir->handle != INVALID_HANDLE_VALUE)
+  {
+  free(pattern);
+  dir->first = TRUE;
+  return dir;
+  }
+err = GetLastError();
+free(pattern);
+free(dir);
+errno = (err == ERROR_ACCESS_DENIED) ? EACCES : ENOENT;
+return NULL;
+}
+
+char *
+readdirectory(directory_type *dir)
+{
+for (;;)
+  {
+  if (!dir->first)
+    {
+    if (!FindNextFile(dir->handle, &(dir->data)))
+      return NULL;
+    }
+  else
+    {
+    dir->first = FALSE;
+    }
+  if (strcmp(dir->data.cFileName, ".") != 0 && strcmp(dir->data.cFileName, "..") != 0)
+    return dir->data.cFileName;
+  }
+#ifndef _MSC_VER
+return NULL;   /* Keep compiler happy; never executed */
+#endif
+}
+
+void
+closedirectory(directory_type *dir)
+{
+FindClose(dir->handle);
+free(dir);
+}
+
+
+/************* Test for regular file in Win32 **********/
+
+/* I don't know how to do this, or if it can be done; assume all paths are
+regular if they are not directories. */
+
+int isregfile(char *filename)
+{
+return !isdirectory(filename);
+}
+
+
+/************* Test for a terminal in Win32 **********/
+
+/* I don't know how to do this; assume never */
+
+static BOOL
+is_stdout_tty(void)
+{
+return FALSE;
+}
+
+static BOOL
+is_file_tty(FILE *f)
+{
+return FALSE;
+}
+
+
+/************* Directory scanning when we can't do it ***********/
+
+/* The type is void, and apart from isdirectory(), the functions do nothing. */
+
+#else
+
+#define FILESEP 0
+typedef void directory_type;
+
+int isdirectory(char *filename) { return 0; }
+directory_type * opendirectory(char *filename) { return (directory_type*)0;}
+char *readdirectory(directory_type *dir) { return (char*)0;}
+void closedirectory(directory_type *dir) {}
+
+
+/************* Test for regular when we can't do it **********/
+
+/* Assume all files are regular. */
+
+int isregfile(char *filename) { return 1; }
+
+
+/************* Test for a terminal when we can't do it **********/
+
+static BOOL
+is_stdout_tty(void)
+{
+return FALSE;
+}
+
+static BOOL
+is_file_tty(FILE *f)
+{
+return FALSE;
+}
+
+#endif
+
+
+
+#ifndef HAVE_STRERROR
+/*************************************************
+*     Provide strerror() for non-ANSI libraries  *
+*************************************************/
+
+/* Some old-fashioned systems still around (e.g. SunOS4) don't have strerror()
+in their libraries, but can provide the same facility by this simple
+alternative function. */
+
+extern int   sys_nerr;
+extern char *sys_errlist[];
+
+char *
+strerror(int n)
+{
+if (n < 0 || n >= sys_nerr) return "unknown error number";
+return sys_errlist[n];
+}
+#endif /* HAVE_STRERROR */
+
+
+
+/*************************************************
+*                Usage function                  *
+*************************************************/
+
+static int
+usage(int rc)
+{
+option_item *op;
+fprintf(stderr, "Usage: pcregrep [-");
+for (op = optionlist; op->one_char != 0; op++)
+  {
+  if (op->one_char > 0) fprintf(stderr, "%c", op->one_char);
+  }
+fprintf(stderr, "] [long options] [pattern] [files]\n");
+fprintf(stderr, "Type `pcregrep --help' for more information and the long "
+  "options.\n");
+return rc;
+}
+
+
+
+/*************************************************
+*                Help function                   *
+*************************************************/
+
+static void
+help(void)
+{
+option_item *op;
+
+printf("Usage: pcregrep [OPTION]... [PATTERN] [FILE1 FILE2 ...]\n");
+printf("Search for PATTERN in each FILE or standard input.\n");
+printf("PATTERN must be present if neither -e nor -f is used.\n");
+printf("\"-\" can be used as a file name to mean STDIN.\n");
+
+#ifdef SUPPORT_LIBZ
+printf("Files whose names end in .gz are read using zlib.\n");
+#endif
+
+#ifdef SUPPORT_LIBBZ2
+printf("Files whose names end in .bz2 are read using bzlib2.\n");
+#endif
+
+#if defined SUPPORT_LIBZ || defined SUPPORT_LIBBZ2
+printf("Other files and the standard input are read as plain files.\n\n");
+#else
+printf("All files are read as plain files, without any interpretation.\n\n");
+#endif
+
+printf("Example: pcregrep -i 'hello.*world' menu.h main.c\n\n");
+printf("Options:\n");
+
+for (op = optionlist; op->one_char != 0; op++)
+  {
+  int n;
+  char s[4];
+
+  /* Two options were accidentally implemented and documented with underscores
+  instead of hyphens in their names, something that was not noticed for quite a
+  few releases. When fixing this, I left the underscored versions in the list
+  in case people were using them. However, we don't want to display them in the
+  help data. There are no other options that contain underscores, and we do not
+  expect ever to implement such options. Therefore, just omit any option that
+  contains an underscore. */
+
+  if (strchr(op->long_name, '_') != NULL) continue;
+
+  if (op->one_char > 0 && (op->long_name)[0] == 0)
+    n = 31 - printf("  -%c", op->one_char);
+  else
+    {
+    if (op->one_char > 0) sprintf(s, "-%c,", op->one_char);
+      else strcpy(s, "   ");
+    n = 31 - printf("  %s --%s", s, op->long_name);
+    }
+
+  if (n < 1) n = 1;
+  printf("%.*s%s\n", n, "                           ", op->help_text);
+  }
+
+printf("\nNumbers may be followed by K or M, e.g. --buffer-size=100K.\n");
+printf("The default value for --buffer-size is %d.\n", PCREGREP_BUFSIZE);
+printf("When reading patterns or file names from a file, trailing white\n");
+printf("space is removed and blank lines are ignored.\n");
+printf("The maximum size of any pattern is %d bytes.\n", MAXPATLEN);
+
+printf("\nWith no FILEs, read standard input. If fewer than two FILEs given, assume -h.\n");
+printf("Exit status is 0 if any matches, 1 if no matches, and 2 if trouble.\n");
+}
+
+
+
+/*************************************************
+*            Test exclude/includes               *
+*************************************************/
+
+/* If any exclude pattern matches, the path is excluded. Otherwise, unless
+there are no includes, the path must match an include pattern.
+
+Arguments:
+  path      the path to be matched
+  ip        the chain of include patterns
+  ep        the chain of exclude patterns
+
+Returns:    TRUE if the path is not excluded
+*/
+
+static BOOL
+test_incexc(char *path, patstr *ip, patstr *ep)
+{
+int plen = strlen(path);
+
+for (; ep != NULL; ep = ep->next)
+  {
+  if (pcre_exec(ep->compiled, NULL, path, plen, 0, 0, NULL, 0) >= 0)
+    return FALSE;
+  }
+
+if (ip == NULL) return TRUE;
+
+for (; ip != NULL; ip = ip->next)
+  {
+  if (pcre_exec(ip->compiled, NULL, path, plen, 0, 0, NULL, 0) >= 0)
+    return TRUE;
+  }
+
+return FALSE;
+}
+
+
+
+/*************************************************
+*         Decode integer argument value          *
+*************************************************/
+
+/* Integer arguments can be followed by K or M. Avoid the use of strtoul()
+because SunOS4 doesn't have it. This is used only for unpicking arguments, so
+just keep it simple.
+
+Arguments:
+  option_data   the option data string
+  op            the option item (for error messages)
+  longop        TRUE if option given in long form
+
+Returns:        a long integer
+*/
+
+static long int
+decode_number(char *option_data, option_item *op, BOOL longop)
+{
+unsigned long int n = 0;
+char *endptr = option_data;
+while (*endptr != 0 && isspace((unsigned char)(*endptr))) endptr++;
+while (isdigit((unsigned char)(*endptr)))
+  n = n * 10 + (int)(*endptr++ - '0');
+if (toupper(*endptr) == 'K')
+  {
+  n *= 1024;
+  endptr++;
+  }
+else if (toupper(*endptr) == 'M')
+  {
+  n *= 1024*1024;
+  endptr++;
+  }
+
+if (*endptr != 0)   /* Error */
+  {
+  if (longop)
+    {
+    char *equals = strchr(op->long_name, '=');
+    int nlen = (equals == NULL)? (int)strlen(op->long_name) :
+      (int)(equals - op->long_name);
+    fprintf(stderr, "pcregrep: Malformed number \"%s\" after --%.*s\n",
+      option_data, nlen, op->long_name);
+    }
+  else
+    fprintf(stderr, "pcregrep: Malformed number \"%s\" after -%c\n",
+      option_data, op->one_char);
+  pcregrep_exit(usage(2));
+  }
+
+return n;
+}
+
+
+
+/*************************************************
+*       Add item to a chain of numbers           *
+*************************************************/
+
+/* Used to add an item onto a chain, or just return an unconnected item if the
+"after" argument is NULL.
+
+Arguments:
+  n          the number to add
+  after      if not NULL points to item to insert after
+
+Returns:     new number block
+*/
+
+static omstr *
+add_number(int n, omstr *after)
+{
+omstr *om = (omstr *)malloc(sizeof(omstr));
+
+if (om == NULL)
+  {
+  fprintf(stderr, "pcregrep: malloc failed\n");
+  pcregrep_exit(2);
+  }
+om->next = NULL;
+om->groupnum = n;
+
+if (after != NULL)
+  {
+  om->next = after->next;
+  after->next = om;
+  }
+return om;
+}
+
+
+
+/*************************************************
+*            Read one line of input              *
+*************************************************/
+
+/* Normally, input is read using fread() into a large buffer, so many lines may
+be read at once. However, doing this for tty input means that no output appears
+until a lot of input has been typed. Instead, tty input is handled line by
+line. We cannot use fgets() for this, because it does not stop at a binary
+zero, and therefore there is no way of telling how many characters it has read,
+because there may be binary zeros embedded in the data.
+
+Arguments:
+  buffer     the buffer to read into
+  length     the maximum number of characters to read
+  f          the file
+
+Returns:     the number of characters read, zero at end of file
+*/
+
+static unsigned int
+read_one_line(char *buffer, int length, FILE *f)
+{
+int c;
+int yield = 0;
+while ((c = fgetc(f)) != EOF)
+  {
+  buffer[yield++] = c;
+  if (c == '\n' || yield >= length) break;
+  }
+return yield;
+}
+
+
+
+/*************************************************
+*             Find end of line                   *
+*************************************************/
+
+/* The length of the endline sequence that is found is set via lenptr. This may
+be zero at the very end of the file if there is no line-ending sequence there.
+
+Arguments:
+  p         current position in line
+  endptr    end of available data
+  lenptr    where to put the length of the eol sequence
+
+Returns:    pointer after the last byte of the line,
+            including the newline byte(s)
+*/
+
+static char *
+end_of_line(char *p, char *endptr, int *lenptr)
+{
+switch(endlinetype)
+  {
+  default:      /* Just in case */
+  case EL_LF:
+  while (p < endptr && *p != '\n') p++;
+  if (p < endptr)
+    {
+    *lenptr = 1;
+    return p + 1;
+    }
+  *lenptr = 0;
+  return endptr;
+
+  case EL_CR:
+  while (p < endptr && *p != '\r') p++;
+  if (p < endptr)
+    {
+    *lenptr = 1;
+    return p + 1;
+    }
+  *lenptr = 0;
+  return endptr;
+
+  case EL_CRLF:
+  for (;;)
+    {
+    while (p < endptr && *p != '\r') p++;
+    if (++p >= endptr)
+      {
+      *lenptr = 0;
+      return endptr;
+      }
+    if (*p == '\n')
+      {
+      *lenptr = 2;
+      return p + 1;
+      }
+    }
+  break;
+
+  case EL_ANYCRLF:
+  while (p < endptr)
+    {
+    int extra = 0;
+    register int c = *((unsigned char *)p);
+
+    if (utf8 && c >= 0xc0)
+      {
+      int gcii, gcss;
+      extra = utf8_table4[c & 0x3f];  /* Number of additional bytes */
+      gcss = 6*extra;
+      c = (c & utf8_table3[extra]) << gcss;
+      for (gcii = 1; gcii <= extra; gcii++)
+        {
+        gcss -= 6;
+        c |= (p[gcii] & 0x3f) << gcss;
+        }
+      }
+
+    p += 1 + extra;
+
+    switch (c)
+      {
+      case '\n':
+      *lenptr = 1;
+      return p;
+
+      case '\r':
+      if (p < endptr && *p == '\n')
+        {
+        *lenptr = 2;
+        p++;
+        }
+      else *lenptr = 1;
+      return p;
+
+      default:
+      break;
+      }
+    }   /* End of loop for ANYCRLF case */
+
+  *lenptr = 0;  /* Must have hit the end */
+  return endptr;
+
+  case EL_ANY:
+  while (p < endptr)
+    {
+    int extra = 0;
+    register int c = *((unsigned char *)p);
+
+    if (utf8 && c >= 0xc0)
+      {
+      int gcii, gcss;
+      extra = utf8_table4[c & 0x3f];  /* Number of additional bytes */
+      gcss = 6*extra;
+      c = (c & utf8_table3[extra]) << gcss;
+      for (gcii = 1; gcii <= extra; gcii++)
+        {
+        gcss -= 6;
+        c |= (p[gcii] & 0x3f) << gcss;
+        }
+      }
+
+    p += 1 + extra;
+
+    switch (c)
+      {
+      case '\n':    /* LF */
+      case '\v':    /* VT */
+      case '\f':    /* FF */
+      *lenptr = 1;
+      return p;
+
+      case '\r':    /* CR */
+      if (p < endptr && *p == '\n')
+        {
+        *lenptr = 2;
+        p++;
+        }
+      else *lenptr = 1;
+      return p;
+
+#ifndef EBCDIC
+      case 0x85:    /* Unicode NEL */
+      *lenptr = utf8? 2 : 1;
+      return p;
+
+      case 0x2028:  /* Unicode LS */
+      case 0x2029:  /* Unicode PS */
+      *lenptr = 3;
+      return p;
+#endif  /* Not EBCDIC */
+
+      default:
+      break;
+      }
+    }   /* End of loop for ANY case */
+
+  *lenptr = 0;  /* Must have hit the end */
+  return endptr;
+  }     /* End of overall switch */
+}
+
+
+
+/*************************************************
+*         Find start of previous line            *
+*************************************************/
+
+/* This is called when looking back for before lines to print.
+
+Arguments:
+  p         start of the subsequent line
+  startptr  start of available data
+
+Returns:    pointer to the start of the previous line
+*/
+
+static char *
+previous_line(char *p, char *startptr)
+{
+switch(endlinetype)
+  {
+  default:      /* Just in case */
+  case EL_LF:
+  p--;
+  while (p > startptr && p[-1] != '\n') p--;
+  return p;
+
+  case EL_CR:
+  p--;
+  while (p > startptr && p[-1] != '\n') p--;
+  return p;
+
+  case EL_CRLF:
+  for (;;)
+    {
+    p -= 2;
+    while (p > startptr && p[-1] != '\n') p--;
+    if (p <= startptr + 1 || p[-2] == '\r') return p;
+    }
+  return p;   /* But control should never get here */
+
+  case EL_ANY:
+  case EL_ANYCRLF:
+  if (*(--p) == '\n' && p > startptr && p[-1] == '\r') p--;
+  if (utf8) while ((*p & 0xc0) == 0x80) p--;
+
+  while (p > startptr)
+    {
+    register unsigned int c;
+    char *pp = p - 1;
+
+    if (utf8)
+      {
+      int extra = 0;
+      while ((*pp & 0xc0) == 0x80) pp--;
+      c = *((unsigned char *)pp);
+      if (c >= 0xc0)
+        {
+        int gcii, gcss;
+        extra = utf8_table4[c & 0x3f];  /* Number of additional bytes */
+        gcss = 6*extra;
+        c = (c & utf8_table3[extra]) << gcss;
+        for (gcii = 1; gcii <= extra; gcii++)
+          {
+          gcss -= 6;
+          c |= (pp[gcii] & 0x3f) << gcss;
+          }
+        }
+      }
+    else c = *((unsigned char *)pp);
+
+    if (endlinetype == EL_ANYCRLF) switch (c)
+      {
+      case '\n':    /* LF */
+      case '\r':    /* CR */
+      return p;
+
+      default:
+      break;
+      }
+
+    else switch (c)
+      {
+      case '\n':    /* LF */
+      case '\v':    /* VT */
+      case '\f':    /* FF */
+      case '\r':    /* CR */
+#ifndef EBCDIE
+      case 0x85:    /* Unicode NEL */
+      case 0x2028:  /* Unicode LS */
+      case 0x2029:  /* Unicode PS */
+#endif  /* Not EBCDIC */
+      return p;
+
+      default:
+      break;
+      }
+
+    p = pp;  /* Back one character */
+    }        /* End of loop for ANY case */
+
+  return startptr;  /* Hit start of data */
+  }     /* End of overall switch */
+}
+
+
+
+
+
+/*************************************************
+*       Print the previous "after" lines         *
+*************************************************/
+
+/* This is called if we are about to lose said lines because of buffer filling,
+and at the end of the file. The data in the line is written using fwrite() so
+that a binary zero does not terminate it.
+
+Arguments:
+  lastmatchnumber   the number of the last matching line, plus one
+  lastmatchrestart  where we restarted after the last match
+  endptr            end of available data
+  printname         filename for printing
+
+Returns:            nothing
+*/
+
+static void
+do_after_lines(int lastmatchnumber, char *lastmatchrestart, char *endptr,
+  char *printname)
+{
+if (after_context > 0 && lastmatchnumber > 0)
+  {
+  int count = 0;
+  while (lastmatchrestart < endptr && count++ < after_context)
+    {
+    int ellength;
+    char *pp = lastmatchrestart;
+    if (printname != NULL) fprintf(stdout, "%s-", printname);
+    if (number) fprintf(stdout, "%d-", lastmatchnumber++);
+    pp = end_of_line(pp, endptr, &ellength);
+    FWRITE(lastmatchrestart, 1, pp - lastmatchrestart, stdout);
+    lastmatchrestart = pp;
+    }
+  hyphenpending = TRUE;
+  }
+}
+
+
+
+/*************************************************
+*   Apply patterns to subject till one matches   *
+*************************************************/
+
+/* This function is called to run through all patterns, looking for a match. It
+is used multiple times for the same subject when colouring is enabled, in order
+to find all possible matches.
+
+Arguments:
+  matchptr     the start of the subject
+  length       the length of the subject to match
+  options      options for pcre_exec
+  startoffset  where to start matching
+  offsets      the offets vector to fill in
+  mrc          address of where to put the result of pcre_exec()
+
+Returns:      TRUE if there was a match
+              FALSE if there was no match
+              invert if there was a non-fatal error
+*/
+
+static BOOL
+match_patterns(char *matchptr, size_t length, unsigned int options,
+  int startoffset, int *offsets, int *mrc)
+{
+int i;
+size_t slen = length;
+patstr *p = patterns;
+const char *msg = "this text:\n\n";
+
+if (slen > 200)
+  {
+  slen = 200;
+  msg = "text that starts:\n\n";
+  }
+for (i = 1; p != NULL; p = p->next, i++)
+  {
+  *mrc = pcre_exec(p->compiled, p->hint, matchptr, (int)length,
+    startoffset, options, offsets, OFFSET_SIZE);
+  if (*mrc >= 0) return TRUE;
+  if (*mrc == PCRE_ERROR_NOMATCH) continue;
+  fprintf(stderr, "pcregrep: pcre_exec() gave error %d while matching ", *mrc);
+  if (patterns->next != NULL) fprintf(stderr, "pattern number %d to ", i);
+  fprintf(stderr, "%s", msg);
+  FWRITE(matchptr, 1, slen, stderr);   /* In case binary zero included */
+  fprintf(stderr, "\n\n");
+  if (*mrc == PCRE_ERROR_MATCHLIMIT || *mrc == PCRE_ERROR_RECURSIONLIMIT ||
+      *mrc == PCRE_ERROR_JIT_STACKLIMIT)
+    resource_error = TRUE;
+  if (error_count++ > 20)
+    {
+    fprintf(stderr, "pcregrep: Too many errors - abandoned.\n");
+    pcregrep_exit(2);
+    }
+  return invert;    /* No more matching; don't show the line again */
+  }
+
+return FALSE;  /* No match, no errors */
+}
+
+
+
+/*************************************************
+*            Grep an individual file             *
+*************************************************/
+
+/* This is called from grep_or_recurse() below. It uses a buffer that is three
+times the value of bufthird. The matching point is never allowed to stray into
+the top third of the buffer, thus keeping more of the file available for
+context printing or for multiline scanning. For large files, the pointer will
+be in the middle third most of the time, so the bottom third is available for
+"before" context printing.
+
+Arguments:
+  handle       the fopened FILE stream for a normal file
+               the gzFile pointer when reading is via libz
+               the BZFILE pointer when reading is via libbz2
+  frtype       FR_PLAIN, FR_LIBZ, or FR_LIBBZ2
+  filename     the file name or NULL (for errors)
+  printname    the file name if it is to be printed for each match
+               or NULL if the file name is not to be printed
+               it cannot be NULL if filenames[_nomatch]_only is set
+
+Returns:       0 if there was at least one match
+               1 otherwise (no matches)
+               2 if an overlong line is encountered
+               3 if there is a read error on a .bz2 file
+*/
+
+static int
+pcregrep(void *handle, int frtype, char *filename, char *printname)
+{
+int rc = 1;
+int linenumber = 1;
+int lastmatchnumber = 0;
+int count = 0;
+int filepos = 0;
+int offsets[OFFSET_SIZE];
+char *lastmatchrestart = NULL;
+char *ptr = main_buffer;
+char *endptr;
+size_t bufflength;
+BOOL binary = FALSE;
+BOOL endhyphenpending = FALSE;
+BOOL input_line_buffered = line_buffered;
+FILE *in = NULL;                    /* Ensure initialized */
+
+#ifdef SUPPORT_LIBZ
+gzFile ingz = NULL;
+#endif
+
+#ifdef SUPPORT_LIBBZ2
+BZFILE *inbz2 = NULL;
+#endif
+
+
+/* Do the first read into the start of the buffer and set up the pointer to end
+of what we have. In the case of libz, a non-zipped .gz file will be read as a
+plain file. However, if a .bz2 file isn't actually bzipped, the first read will
+fail. */
+
+(void)frtype;
+
+#ifdef SUPPORT_LIBZ
+if (frtype == FR_LIBZ)
+  {
+  ingz = (gzFile)handle;
+  bufflength = gzread (ingz, main_buffer, bufsize);
+  }
+else
+#endif
+
+#ifdef SUPPORT_LIBBZ2
+if (frtype == FR_LIBBZ2)
+  {
+  inbz2 = (BZFILE *)handle;
+  bufflength = BZ2_bzread(inbz2, main_buffer, bufsize);
+  if ((int)bufflength < 0) return 2;   /* Gotcha: bufflength is size_t; */
+  }                                    /* without the cast it is unsigned. */
+else
+#endif
+
+  {
+  in = (FILE *)handle;
+  if (is_file_tty(in)) input_line_buffered = TRUE;
+  bufflength = input_line_buffered?
+    read_one_line(main_buffer, bufsize, in) :
+    fread(main_buffer, 1, bufsize, in);
+  }
+
+endptr = main_buffer + bufflength;
+
+/* Unless binary-files=text, see if we have a binary file. This uses the same
+rule as GNU grep, namely, a search for a binary zero byte near the start of the
+file. */
+
+if (binary_files != BIN_TEXT)
+  {
+  binary =
+    memchr(main_buffer, 0, (bufflength > 1024)? 1024 : bufflength) != NULL;
+  if (binary && binary_files == BIN_NOMATCH) return 1;
+  }
+
+/* Loop while the current pointer is not at the end of the file. For large
+files, endptr will be at the end of the buffer when we are in the middle of the
+file, but ptr will never get there, because as soon as it gets over 2/3 of the
+way, the buffer is shifted left and re-filled. */
+
+while (ptr < endptr)
+  {
+  int endlinelength;
+  int mrc = 0;
+  int startoffset = 0;
+  unsigned int options = 0;
+  BOOL match;
+  char *matchptr = ptr;
+  char *t = ptr;
+  size_t length, linelength;
+
+  /* At this point, ptr is at the start of a line. We need to find the length
+  of the subject string to pass to pcre_exec(). In multiline mode, it is the
+  length remainder of the data in the buffer. Otherwise, it is the length of
+  the next line, excluding the terminating newline. After matching, we always
+  advance by the length of the next line. In multiline mode the PCRE_FIRSTLINE
+  option is used for compiling, so that any match is constrained to be in the
+  first line. */
+
+  t = end_of_line(t, endptr, &endlinelength);
+  linelength = t - ptr - endlinelength;
+  length = multiline? (size_t)(endptr - ptr) : linelength;
+
+  /* Check to see if the line we are looking at extends right to the very end
+  of the buffer without a line terminator. This means the line is too long to
+  handle. */
+
+  if (endlinelength == 0 && t == main_buffer + bufsize)
+    {
+    fprintf(stderr, "pcregrep: line %d%s%s is too long for the internal buffer\n"
+                    "pcregrep: check the --buffer-size option\n",
+                    linenumber,
+                    (filename == NULL)? "" : " of file ",
+                    (filename == NULL)? "" : filename);
+    return 2;
+    }
+
+  /* Extra processing for Jeffrey Friedl's debugging. */
+
+#ifdef JFRIEDL_DEBUG
+  if (jfriedl_XT || jfriedl_XR)
+  {
+#     include 
+#     include 
+      struct timeval start_time, end_time;
+      struct timezone dummy;
+      int i;
+
+      if (jfriedl_XT)
+      {
+          unsigned long newlen = length * jfriedl_XT + strlen(jfriedl_prefix) + strlen(jfriedl_postfix);
+          const char *orig = ptr;
+          ptr = malloc(newlen + 1);
+          if (!ptr) {
+                  printf("out of memory");
+                  pcregrep_exit(2);
+          }
+          endptr = ptr;
+          strcpy(endptr, jfriedl_prefix); endptr += strlen(jfriedl_prefix);
+          for (i = 0; i < jfriedl_XT; i++) {
+                  strncpy(endptr, orig,  length);
+                  endptr += length;
+          }
+          strcpy(endptr, jfriedl_postfix); endptr += strlen(jfriedl_postfix);
+          length = newlen;
+      }
+
+      if (gettimeofday(&start_time, &dummy) != 0)
+              perror("bad gettimeofday");
+
+
+      for (i = 0; i < jfriedl_XR; i++)
+          match = (pcre_exec(patterns->compiled, patterns->hint, ptr, length, 0,
+              PCRE_NOTEMPTY, offsets, OFFSET_SIZE) >= 0);
+
+      if (gettimeofday(&end_time, &dummy) != 0)
+              perror("bad gettimeofday");
+
+      double delta = ((end_time.tv_sec + (end_time.tv_usec / 1000000.0))
+                      -
+                      (start_time.tv_sec + (start_time.tv_usec / 1000000.0)));
+
+      printf("%s TIMER[%.4f]\n", match ? "MATCH" : "FAIL", delta);
+      return 0;
+  }
+#endif
+
+  /* We come back here after a match when show_only_matching is set, in order
+  to find any further matches in the same line. This applies to
+  --only-matching, --file-offsets, and --line-offsets. */
+
+  ONLY_MATCHING_RESTART:
+
+  /* Run through all the patterns until one matches or there is an error other
+  than NOMATCH. This code is in a subroutine so that it can be re-used for
+  finding subsequent matches when colouring matched lines. After finding one
+  match, set PCRE_NOTEMPTY to disable any further matches of null strings in
+  this line. */
+
+  match = match_patterns(matchptr, length, options, startoffset, offsets, &mrc);
+  options = PCRE_NOTEMPTY;
+
+  /* If it's a match or a not-match (as required), do what's wanted. */
+
+  if (match != invert)
+    {
+    BOOL hyphenprinted = FALSE;
+
+    /* We've failed if we want a file that doesn't have any matches. */
+
+    if (filenames == FN_NOMATCH_ONLY) return 1;
+
+    /* Just count if just counting is wanted. */
+
+    if (count_only) count++;
+
+    /* When handling a binary file and binary-files==binary, the "binary"
+    variable will be set true (it's false in all other cases). In this
+    situation we just want to output the file name. No need to scan further. */
+
+    else if (binary)
+      {
+      fprintf(stdout, "Binary file %s matches\n", filename);
+      return 0;
+      }
+
+    /* If all we want is a file name, there is no need to scan any more lines
+    in the file. */
+
+    else if (filenames == FN_MATCH_ONLY)
+      {
+      fprintf(stdout, "%s\n", printname);
+      return 0;
+      }
+
+    /* Likewise, if all we want is a yes/no answer. */
+
+    else if (quiet) return 0;
+
+    /* The --only-matching option prints just the substring that matched,
+    and/or one or more captured portions of it, as long as these strings are
+    not empty. The --file-offsets and --line-offsets options output offsets for
+    the matching substring (all three set show_only_matching). None of these
+    mutually exclusive options prints any context. Afterwards, adjust the start
+    and then jump back to look for further matches in the same line. If we are
+    in invert mode, however, nothing is printed and we do not restart - this
+    could still be useful because the return code is set. */
+
+    else if (show_only_matching)
+      {
+      if (!invert)
+        {
+        if (printname != NULL) fprintf(stdout, "%s:", printname);
+        if (number) fprintf(stdout, "%d:", linenumber);
+
+        /* Handle --line-offsets */
+
+        if (line_offsets)
+          fprintf(stdout, "%d,%d\n", (int)(matchptr + offsets[0] - ptr),
+            offsets[1] - offsets[0]);
+
+        /* Handle --file-offsets */
+
+        else if (file_offsets)
+          fprintf(stdout, "%d,%d\n",
+            (int)(filepos + matchptr + offsets[0] - ptr),
+            offsets[1] - offsets[0]);
+
+        /* Handle --only-matching, which may occur many times */
+
+        else
+          {
+          BOOL printed = FALSE;
+          omstr *om;
+
+          for (om = only_matching; om != NULL; om = om->next)
+            {
+            int n = om->groupnum;
+            if (n < mrc)
+              {
+              int plen = offsets[2*n + 1] - offsets[2*n];
+              if (plen > 0)
+                {
+                if (printed) fprintf(stdout, "%s", om_separator);
+                if (do_colour) fprintf(stdout, "%c[%sm", 0x1b, colour_string);
+                FWRITE(matchptr + offsets[n*2], 1, plen, stdout);
+                if (do_colour) fprintf(stdout, "%c[00m", 0x1b);
+                printed = TRUE;
+                }
+              }
+            }
+
+          if (printed || printname != NULL || number) fprintf(stdout, "\n");
+          }
+
+        /* Prepare to repeat to find the next match */
+
+        match = FALSE;
+        if (line_buffered) fflush(stdout);
+        rc = 0;                      /* Had some success */
+        startoffset = offsets[1];    /* Restart after the match */
+        goto ONLY_MATCHING_RESTART;
+        }
+      }
+
+    /* This is the default case when none of the above options is set. We print
+    the matching lines(s), possibly preceded and/or followed by other lines of
+    context. */
+
+    else
+      {
+      /* See if there is a requirement to print some "after" lines from a
+      previous match. We never print any overlaps. */
+
+      if (after_context > 0 && lastmatchnumber > 0)
+        {
+        int ellength;
+        int linecount = 0;
+        char *p = lastmatchrestart;
+
+        while (p < ptr && linecount < after_context)
+          {
+          p = end_of_line(p, ptr, &ellength);
+          linecount++;
+          }
+
+        /* It is important to advance lastmatchrestart during this printing so
+        that it interacts correctly with any "before" printing below. Print
+        each line's data using fwrite() in case there are binary zeroes. */
+
+        while (lastmatchrestart < p)
+          {
+          char *pp = lastmatchrestart;
+          if (printname != NULL) fprintf(stdout, "%s-", printname);
+          if (number) fprintf(stdout, "%d-", lastmatchnumber++);
+          pp = end_of_line(pp, endptr, &ellength);
+          FWRITE(lastmatchrestart, 1, pp - lastmatchrestart, stdout);
+          lastmatchrestart = pp;
+          }
+        if (lastmatchrestart != ptr) hyphenpending = TRUE;
+        }
+
+      /* If there were non-contiguous lines printed above, insert hyphens. */
+
+      if (hyphenpending)
+        {
+        fprintf(stdout, "--\n");
+        hyphenpending = FALSE;
+        hyphenprinted = TRUE;
+        }
+
+      /* See if there is a requirement to print some "before" lines for this
+      match. Again, don't print overlaps. */
+
+      if (before_context > 0)
+        {
+        int linecount = 0;
+        char *p = ptr;
+
+        while (p > main_buffer && (lastmatchnumber == 0 || p > lastmatchrestart) &&
+               linecount < before_context)
+          {
+          linecount++;
+          p = previous_line(p, main_buffer);
+          }
+
+        if (lastmatchnumber > 0 && p > lastmatchrestart && !hyphenprinted)
+          fprintf(stdout, "--\n");
+
+        while (p < ptr)
+          {
+          int ellength;
+          char *pp = p;
+          if (printname != NULL) fprintf(stdout, "%s-", printname);
+          if (number) fprintf(stdout, "%d-", linenumber - linecount--);
+          pp = end_of_line(pp, endptr, &ellength);
+          FWRITE(p, 1, pp - p, stdout);
+          p = pp;
+          }
+        }
+
+      /* Now print the matching line(s); ensure we set hyphenpending at the end
+      of the file if any context lines are being output. */
+
+      if (after_context > 0 || before_context > 0)
+        endhyphenpending = TRUE;
+
+      if (printname != NULL) fprintf(stdout, "%s:", printname);
+      if (number) fprintf(stdout, "%d:", linenumber);
+
+      /* In multiline mode, we want to print to the end of the line in which
+      the end of the matched string is found, so we adjust linelength and the
+      line number appropriately, but only when there actually was a match
+      (invert not set). Because the PCRE_FIRSTLINE option is set, the start of
+      the match will always be before the first newline sequence. */
+
+      if (multiline & !invert)
+        {
+        char *endmatch = ptr + offsets[1];
+        t = ptr;
+        while (t < endmatch)
+          {
+          t = end_of_line(t, endptr, &endlinelength);
+          if (t < endmatch) linenumber++; else break;
+          }
+        linelength = t - ptr - endlinelength;
+        }
+
+      /*** NOTE: Use only fwrite() to output the data line, so that binary
+      zeroes are treated as just another data character. */
+
+      /* This extra option, for Jeffrey Friedl's debugging requirements,
+      replaces the matched string, or a specific captured string if it exists,
+      with X. When this happens, colouring is ignored. */
+
+#ifdef JFRIEDL_DEBUG
+      if (S_arg >= 0 && S_arg < mrc)
+        {
+        int first = S_arg * 2;
+        int last  = first + 1;
+        FWRITE(ptr, 1, offsets[first], stdout);
+        fprintf(stdout, "X");
+        FWRITE(ptr + offsets[last], 1, linelength - offsets[last], stdout);
+        }
+      else
+#endif
+
+      /* We have to split the line(s) up if colouring, and search for further
+      matches, but not of course if the line is a non-match. */
+
+      if (do_colour && !invert)
+        {
+        int plength;
+        FWRITE(ptr, 1, offsets[0], stdout);
+        fprintf(stdout, "%c[%sm", 0x1b, colour_string);
+        FWRITE(ptr + offsets[0], 1, offsets[1] - offsets[0], stdout);
+        fprintf(stdout, "%c[00m", 0x1b);
+        for (;;)
+          {
+          startoffset = offsets[1];
+          if (startoffset >= (int)linelength + endlinelength ||
+              !match_patterns(matchptr, length, options, startoffset, offsets,
+                &mrc))
+            break;
+          FWRITE(matchptr + startoffset, 1, offsets[0] - startoffset, stdout);
+          fprintf(stdout, "%c[%sm", 0x1b, colour_string);
+          FWRITE(matchptr + offsets[0], 1, offsets[1] - offsets[0], stdout);
+          fprintf(stdout, "%c[00m", 0x1b);
+          }
+
+        /* In multiline mode, we may have already printed the complete line
+        and its line-ending characters (if they matched the pattern), so there
+        may be no more to print. */
+
+        plength = (int)((linelength + endlinelength) - startoffset);
+        if (plength > 0) FWRITE(ptr + startoffset, 1, plength, stdout);
+        }
+
+      /* Not colouring; no need to search for further matches */
+
+      else FWRITE(ptr, 1, linelength + endlinelength, stdout);
+      }
+
+    /* End of doing what has to be done for a match. If --line-buffered was
+    given, flush the output. */
+
+    if (line_buffered) fflush(stdout);
+    rc = 0;    /* Had some success */
+
+    /* Remember where the last match happened for after_context. We remember
+    where we are about to restart, and that line's number. */
+
+    lastmatchrestart = ptr + linelength + endlinelength;
+    lastmatchnumber = linenumber + 1;
+    }
+
+  /* For a match in multiline inverted mode (which of course did not cause
+  anything to be printed), we have to move on to the end of the match before
+  proceeding. */
+
+  if (multiline && invert && match)
+    {
+    int ellength;
+    char *endmatch = ptr + offsets[1];
+    t = ptr;
+    while (t < endmatch)
+      {
+      t = end_of_line(t, endptr, &ellength);
+      if (t <= endmatch) linenumber++; else break;
+      }
+    endmatch = end_of_line(endmatch, endptr, &ellength);
+    linelength = endmatch - ptr - ellength;
+    }
+
+  /* Advance to after the newline and increment the line number. The file
+  offset to the current line is maintained in filepos. */
+
+  ptr += linelength + endlinelength;
+  filepos += (int)(linelength + endlinelength);
+  linenumber++;
+
+  /* If input is line buffered, and the buffer is not yet full, read another
+  line and add it into the buffer. */
+
+  if (input_line_buffered && bufflength < (size_t)bufsize)
+    {
+    int add = read_one_line(ptr, bufsize - (int)(ptr - main_buffer), in);
+    bufflength += add;
+    endptr += add;
+    }
+
+  /* If we haven't yet reached the end of the file (the buffer is full), and
+  the current point is in the top 1/3 of the buffer, slide the buffer down by
+  1/3 and refill it. Before we do this, if some unprinted "after" lines are
+  about to be lost, print them. */
+
+  if (bufflength >= (size_t)bufsize && ptr > main_buffer + 2*bufthird)
+    {
+    if (after_context > 0 &&
+        lastmatchnumber > 0 &&
+        lastmatchrestart < main_buffer + bufthird)
+      {
+      do_after_lines(lastmatchnumber, lastmatchrestart, endptr, printname);
+      lastmatchnumber = 0;
+      }
+
+    /* Now do the shuffle */
+
+    memmove(main_buffer, main_buffer + bufthird, 2*bufthird);
+    ptr -= bufthird;
+
+#ifdef SUPPORT_LIBZ
+    if (frtype == FR_LIBZ)
+      bufflength = 2*bufthird +
+        gzread (ingz, main_buffer + 2*bufthird, bufthird);
+    else
+#endif
+
+#ifdef SUPPORT_LIBBZ2
+    if (frtype == FR_LIBBZ2)
+      bufflength = 2*bufthird +
+        BZ2_bzread(inbz2, main_buffer + 2*bufthird, bufthird);
+    else
+#endif
+
+    bufflength = 2*bufthird +
+      (input_line_buffered?
+       read_one_line(main_buffer + 2*bufthird, bufthird, in) :
+       fread(main_buffer + 2*bufthird, 1, bufthird, in));
+    endptr = main_buffer + bufflength;
+
+    /* Adjust any last match point */
+
+    if (lastmatchnumber > 0) lastmatchrestart -= bufthird;
+    }
+  }     /* Loop through the whole file */
+
+/* End of file; print final "after" lines if wanted; do_after_lines sets
+hyphenpending if it prints something. */
+
+if (!show_only_matching && !count_only)
+  {
+  do_after_lines(lastmatchnumber, lastmatchrestart, endptr, printname);
+  hyphenpending |= endhyphenpending;
+  }
+
+/* Print the file name if we are looking for those without matches and there
+were none. If we found a match, we won't have got this far. */
+
+if (filenames == FN_NOMATCH_ONLY)
+  {
+  fprintf(stdout, "%s\n", printname);
+  return 0;
+  }
+
+/* Print the match count if wanted */
+
+if (count_only)
+  {
+  if (count > 0 || !omit_zero_count)
+    {
+    if (printname != NULL && filenames != FN_NONE)
+      fprintf(stdout, "%s:", printname);
+    fprintf(stdout, "%d\n", count);
+    }
+  }
+
+return rc;
+}
+
+
+
+/*************************************************
+*     Grep a file or recurse into a directory    *
+*************************************************/
+
+/* Given a path name, if it's a directory, scan all the files if we are
+recursing; if it's a file, grep it.
+
+Arguments:
+  pathname          the path to investigate
+  dir_recurse       TRUE if recursing is wanted (-r or -drecurse)
+  only_one_at_top   TRUE if the path is the only one at toplevel
+
+Returns:  -1 the file/directory was skipped
+           0 if there was at least one match
+           1 if there were no matches
+           2 there was some kind of error
+
+However, file opening failures are suppressed if "silent" is set.
+*/
+
+static int
+grep_or_recurse(char *pathname, BOOL dir_recurse, BOOL only_one_at_top)
+{
+int rc = 1;
+int frtype;
+void *handle;
+char *lastcomp;
+FILE *in = NULL;           /* Ensure initialized */
+
+#ifdef SUPPORT_LIBZ
+gzFile ingz = NULL;
+#endif
+
+#ifdef SUPPORT_LIBBZ2
+BZFILE *inbz2 = NULL;
+#endif
+
+#if defined SUPPORT_LIBZ || defined SUPPORT_LIBBZ2
+int pathlen;
+#endif
+
+/* If the file name is "-" we scan stdin */
+
+if (strcmp(pathname, "-") == 0)
+  {
+  return pcregrep(stdin, FR_PLAIN, stdin_name,
+    (filenames > FN_DEFAULT || (filenames == FN_DEFAULT && !only_one_at_top))?
+      stdin_name : NULL);
+  }
+
+/* Inclusion and exclusion: --include-dir and --exclude-dir apply only to
+directories, whereas --include and --exclude apply to everything else. The test
+is against the final component of the path. */
+
+lastcomp = strrchr(pathname, FILESEP);
+lastcomp = (lastcomp == NULL)? pathname : lastcomp + 1;
+
+/* If the file is a directory, skip if not recursing or if explicitly excluded.
+Otherwise, scan the directory and recurse for each path within it. The scanning
+code is localized so it can be made system-specific. */
+
+if (isdirectory(pathname))
+  {
+  if (dee_action == dee_SKIP ||
+      !test_incexc(lastcomp, include_dir_patterns, exclude_dir_patterns))
+    return -1;
+
+  if (dee_action == dee_RECURSE)
+    {
+    char buffer[1024];
+    char *nextfile;
+    directory_type *dir = opendirectory(pathname);
+
+    if (dir == NULL)
+      {
+      if (!silent)
+        fprintf(stderr, "pcregrep: Failed to open directory %s: %s\n", pathname,
+          strerror(errno));
+      return 2;
+      }
+
+    while ((nextfile = readdirectory(dir)) != NULL)
+      {
+      int frc;
+      sprintf(buffer, "%.512s%c%.128s", pathname, FILESEP, nextfile);
+      frc = grep_or_recurse(buffer, dir_recurse, FALSE);
+      if (frc > 1) rc = frc;
+       else if (frc == 0 && rc == 1) rc = 0;
+      }
+
+    closedirectory(dir);
+    return rc;
+    }
+  }
+
+/* If the file is not a directory and not a regular file, skip it if that's
+been requested. Otherwise, check for explicit include/exclude. */
+
+else if ((!isregfile(pathname) && DEE_action == DEE_SKIP) ||
+          !test_incexc(lastcomp, include_patterns, exclude_patterns))
+        return -1;
+
+/* Control reaches here if we have a regular file, or if we have a directory
+and recursion or skipping was not requested, or if we have anything else and
+skipping was not requested. The scan proceeds. If this is the first and only
+argument at top level, we don't show the file name, unless we are only showing
+the file name, or the filename was forced (-H). */
+
+#if defined SUPPORT_LIBZ || defined SUPPORT_LIBBZ2
+pathlen = (int)(strlen(pathname));
+#endif
+
+/* Open using zlib if it is supported and the file name ends with .gz. */
+
+#ifdef SUPPORT_LIBZ
+if (pathlen > 3 && strcmp(pathname + pathlen - 3, ".gz") == 0)
+  {
+  ingz = gzopen(pathname, "rb");
+  if (ingz == NULL)
+    {
+    if (!silent)
+      fprintf(stderr, "pcregrep: Failed to open %s: %s\n", pathname,
+        strerror(errno));
+    return 2;
+    }
+  handle = (void *)ingz;
+  frtype = FR_LIBZ;
+  }
+else
+#endif
+
+/* Otherwise open with bz2lib if it is supported and the name ends with .bz2. */
+
+#ifdef SUPPORT_LIBBZ2
+if (pathlen > 4 && strcmp(pathname + pathlen - 4, ".bz2") == 0)
+  {
+  inbz2 = BZ2_bzopen(pathname, "rb");
+  handle = (void *)inbz2;
+  frtype = FR_LIBBZ2;
+  }
+else
+#endif
+
+/* Otherwise use plain fopen(). The label is so that we can come back here if
+an attempt to read a .bz2 file indicates that it really is a plain file. */
+
+#ifdef SUPPORT_LIBBZ2
+PLAIN_FILE:
+#endif
+  {
+  in = fopen(pathname, "rb");
+  handle = (void *)in;
+  frtype = FR_PLAIN;
+  }
+
+/* All the opening methods return errno when they fail. */
+
+if (handle == NULL)
+  {
+  if (!silent)
+    fprintf(stderr, "pcregrep: Failed to open %s: %s\n", pathname,
+      strerror(errno));
+  return 2;
+  }
+
+/* Now grep the file */
+
+rc = pcregrep(handle, frtype, pathname, (filenames > FN_DEFAULT ||
+  (filenames == FN_DEFAULT && !only_one_at_top))? pathname : NULL);
+
+/* Close in an appropriate manner. */
+
+#ifdef SUPPORT_LIBZ
+if (frtype == FR_LIBZ)
+  gzclose(ingz);
+else
+#endif
+
+/* If it is a .bz2 file and the result is 3, it means that the first attempt to
+read failed. If the error indicates that the file isn't in fact bzipped, try
+again as a normal file. */
+
+#ifdef SUPPORT_LIBBZ2
+if (frtype == FR_LIBBZ2)
+  {
+  if (rc == 3)
+    {
+    int errnum;
+    const char *err = BZ2_bzerror(inbz2, &errnum);
+    if (errnum == BZ_DATA_ERROR_MAGIC)
+      {
+      BZ2_bzclose(inbz2);
+      goto PLAIN_FILE;
+      }
+    else if (!silent)
+      fprintf(stderr, "pcregrep: Failed to read %s using bzlib: %s\n",
+        pathname, err);
+    rc = 2;    /* The normal "something went wrong" code */
+    }
+  BZ2_bzclose(inbz2);
+  }
+else
+#endif
+
+/* Normal file close */
+
+fclose(in);
+
+/* Pass back the yield from pcregrep(). */
+
+return rc;
+}
+
+
+
+/*************************************************
+*    Handle a single-letter, no data option      *
+*************************************************/
+
+static int
+handle_option(int letter, int options)
+{
+switch(letter)
+  {
+  case N_FOFFSETS: file_offsets = TRUE; break;
+  case N_HELP: help(); pcregrep_exit(0);
+  case N_LBUFFER: line_buffered = TRUE; break;
+  case N_LOFFSETS: line_offsets = number = TRUE; break;
+  case N_NOJIT: study_options &= ~PCRE_STUDY_JIT_COMPILE; break;
+  case 'a': binary_files = BIN_TEXT; break;
+  case 'c': count_only = TRUE; break;
+  case 'F': process_options |= PO_FIXED_STRINGS; break;
+  case 'H': filenames = FN_FORCE; break;
+  case 'I': binary_files = BIN_NOMATCH; break;
+  case 'h': filenames = FN_NONE; break;
+  case 'i': options |= PCRE_CASELESS; break;
+  case 'l': omit_zero_count = TRUE; filenames = FN_MATCH_ONLY; break;
+  case 'L': filenames = FN_NOMATCH_ONLY; break;
+  case 'M': multiline = TRUE; options |= PCRE_MULTILINE|PCRE_FIRSTLINE; break;
+  case 'n': number = TRUE; break;
+
+  case 'o':
+  only_matching_last = add_number(0, only_matching_last);
+  if (only_matching == NULL) only_matching = only_matching_last;
+  break;
+
+  case 'q': quiet = TRUE; break;
+  case 'r': dee_action = dee_RECURSE; break;
+  case 's': silent = TRUE; break;
+  case 'u': options |= PCRE_UTF8; utf8 = TRUE; break;
+  case 'v': invert = TRUE; break;
+  case 'w': process_options |= PO_WORD_MATCH; break;
+  case 'x': process_options |= PO_LINE_MATCH; break;
+
+  case 'V':
+  fprintf(stdout, "pcregrep version %s\n", pcre_version());
+  pcregrep_exit(0);
+  break;
+
+  default:
+  fprintf(stderr, "pcregrep: Unknown option -%c\n", letter);
+  pcregrep_exit(usage(2));
+  }
+
+return options;
+}
+
+
+
+
+/*************************************************
+*          Construct printed ordinal             *
+*************************************************/
+
+/* This turns a number into "1st", "3rd", etc. */
+
+static char *
+ordin(int n)
+{
+static char buffer[8];
+char *p = buffer;
+sprintf(p, "%d", n);
+while (*p != 0) p++;
+switch (n%10)
+  {
+  case 1: strcpy(p, "st"); break;
+  case 2: strcpy(p, "nd"); break;
+  case 3: strcpy(p, "rd"); break;
+  default: strcpy(p, "th"); break;
+  }
+return buffer;
+}
+
+
+
+/*************************************************
+*          Compile a single pattern              *
+*************************************************/
+
+/* Do nothing if the pattern has already been compiled. This is the case for
+include/exclude patterns read from a file.
+
+When the -F option has been used, each "pattern" may be a list of strings,
+separated by line breaks. They will be matched literally. We split such a
+string and compile the first substring, inserting an additional block into the
+pattern chain.
+
+Arguments:
+  p              points to the pattern block
+  options        the PCRE options
+  popts          the processing options
+  fromfile       TRUE if the pattern was read from a file
+  fromtext       file name or identifying text (e.g. "include")
+  count          0 if this is the only command line pattern, or
+                 number of the command line pattern, or
+                 linenumber for a pattern from a file
+
+Returns:         TRUE on success, FALSE after an error
+*/
+
+static BOOL
+compile_pattern(patstr *p, int options, int popts, int fromfile,
+  const char *fromtext, int count)
+{
+char buffer[PATBUFSIZE];
+const char *error;
+char *ps = p->string;
+int patlen = strlen(ps);
+int errptr;
+
+if (p->compiled != NULL) return TRUE;
+
+if ((popts & PO_FIXED_STRINGS) != 0)
+  {
+  int ellength;
+  char *eop = ps + patlen;
+  char *pe = end_of_line(ps, eop, &ellength);
+
+  if (ellength != 0)
+    {
+    if (add_pattern(pe, p) == NULL) return FALSE;
+    patlen = (int)(pe - ps - ellength);
+    }
+  }
+
+sprintf(buffer, "%s%.*s%s", prefix[popts], patlen, ps, suffix[popts]);
+p->compiled = pcre_compile(buffer, options, &error, &errptr, pcretables);
+if (p->compiled != NULL) return TRUE;
+
+/* Handle compile errors */
+
+errptr -= (int)strlen(prefix[popts]);
+if (errptr > patlen) errptr = patlen;
+
+if (fromfile)
+  {
+  fprintf(stderr, "pcregrep: Error in regex in line %d of %s "
+    "at offset %d: %s\n", count, fromtext, errptr, error);
+  }
+else
+  {
+  if (count == 0)
+    fprintf(stderr, "pcregrep: Error in %s regex at offset %d: %s\n",
+      fromtext, errptr, error);
+  else
+    fprintf(stderr, "pcregrep: Error in %s %s regex at offset %d: %s\n",
+      ordin(count), fromtext, errptr, error);
+  }
+
+return FALSE;
+}
+
+
+
+/*************************************************
+*     Read and compile a file of patterns        *
+*************************************************/
+
+/* This is used for --filelist, --include-from, and --exclude-from.
+
+Arguments:
+  name         the name of the file; "-" is stdin
+  patptr       pointer to the pattern chain anchor
+  patlastptr   pointer to the last pattern pointer
+  popts        the process options to pass to pattern_compile()
+
+Returns:       TRUE if all went well
+*/
+
+static BOOL
+read_pattern_file(char *name, patstr **patptr, patstr **patlastptr, int popts)
+{
+int linenumber = 0;
+FILE *f;
+char *filename;
+char buffer[PATBUFSIZE];
+
+if (strcmp(name, "-") == 0)
+  {
+  f = stdin;
+  filename = stdin_name;
+  }
+else
+  {
+  f = fopen(name, "r");
+  if (f == NULL)
+    {
+    fprintf(stderr, "pcregrep: Failed to open %s: %s\n", name, strerror(errno));
+    return FALSE;
+    }
+  filename = name;
+  }
+
+while (fgets(buffer, PATBUFSIZE, f) != NULL)
+  {
+  char *s = buffer + (int)strlen(buffer);
+  while (s > buffer && isspace((unsigned char)(s[-1]))) s--;
+  *s = 0;
+  linenumber++;
+  if (buffer[0] == 0) continue;   /* Skip blank lines */
+
+  /* Note: this call to add_pattern() puts a pointer to the local variable
+  "buffer" into the pattern chain. However, that pointer is used only when
+  compiling the pattern, which happens immediately below, so we flatten it
+  afterwards, as a precaution against any later code trying to use it. */
+
+  *patlastptr = add_pattern(buffer, *patlastptr);
+  if (*patlastptr == NULL) return FALSE;
+  if (*patptr == NULL) *patptr = *patlastptr;
+
+  /* This loop is needed because compiling a "pattern" when -F is set may add
+  on additional literal patterns if the original contains a newline. In the
+  common case, it never will, because fgets() stops at a newline. However,
+  the -N option can be used to give pcregrep a different newline setting. */
+
+  for(;;)
+    {
+    if (!compile_pattern(*patlastptr, pcre_options, popts, TRUE, filename,
+        linenumber))
+      return FALSE;
+    (*patlastptr)->string = NULL;            /* Insurance */
+    if ((*patlastptr)->next == NULL) break;
+    *patlastptr = (*patlastptr)->next;
+    }
+  }
+
+if (f != stdin) fclose(f);
+return TRUE;
+}
+
+
+
+/*************************************************
+*                Main program                    *
+*************************************************/
+
+/* Returns 0 if something matched, 1 if nothing matched, 2 after an error. */
+
+int
+main(int argc, char **argv)
+{
+int i, j;
+int rc = 1;
+BOOL only_one_at_top;
+patstr *cp;
+fnstr *fn;
+const char *locale_from = "--locale";
+const char *error;
+
+#ifdef SUPPORT_PCREGREP_JIT
+pcre_jit_stack *jit_stack = NULL;
+#endif
+
+/* Set the default line ending value from the default in the PCRE library;
+"lf", "cr", "crlf", and "any" are supported. Anything else is treated as "lf".
+Note that the return values from pcre_config(), though derived from the ASCII
+codes, are the same in EBCDIC environments, so we must use the actual values
+rather than escapes such as as '\r'. */
+
+(void)pcre_config(PCRE_CONFIG_NEWLINE, &i);
+switch(i)
+  {
+  default:               newline = (char *)"lf"; break;
+  case 13:               newline = (char *)"cr"; break;
+  case (13 << 8) | 10:   newline = (char *)"crlf"; break;
+  case -1:               newline = (char *)"any"; break;
+  case -2:               newline = (char *)"anycrlf"; break;
+  }
+
+/* Process the options */
+
+for (i = 1; i < argc; i++)
+  {
+  option_item *op = NULL;
+  char *option_data = (char *)"";    /* default to keep compiler happy */
+  BOOL longop;
+  BOOL longopwasequals = FALSE;
+
+  if (argv[i][0] != '-') break;
+
+  /* If we hit an argument that is just "-", it may be a reference to STDIN,
+  but only if we have previously had -e or -f to define the patterns. */
+
+  if (argv[i][1] == 0)
+    {
+    if (pattern_files != NULL || patterns != NULL) break;
+      else pcregrep_exit(usage(2));
+    }
+
+  /* Handle a long name option, or -- to terminate the options */
+
+  if (argv[i][1] == '-')
+    {
+    char *arg = argv[i] + 2;
+    char *argequals = strchr(arg, '=');
+
+    if (*arg == 0)    /* -- terminates options */
+      {
+      i++;
+      break;                /* out of the options-handling loop */
+      }
+
+    longop = TRUE;
+
+    /* Some long options have data that follows after =, for example file=name.
+    Some options have variations in the long name spelling: specifically, we
+    allow "regexp" because GNU grep allows it, though I personally go along
+    with Jeffrey Friedl and Larry Wall in preferring "regex" without the "p".
+    These options are entered in the table as "regex(p)". Options can be in
+    both these categories. */
+
+    for (op = optionlist; op->one_char != 0; op++)
+      {
+      char *opbra = strchr(op->long_name, '(');
+      char *equals = strchr(op->long_name, '=');
+
+      /* Handle options with only one spelling of the name */
+
+      if (opbra == NULL)     /* Does not contain '(' */
+        {
+        if (equals == NULL)  /* Not thing=data case */
+          {
+          if (strcmp(arg, op->long_name) == 0) break;
+          }
+        else                 /* Special case xxx=data */
+          {
+          int oplen = (int)(equals - op->long_name);
+          int arglen = (argequals == NULL)?
+            (int)strlen(arg) : (int)(argequals - arg);
+          if (oplen == arglen && strncmp(arg, op->long_name, oplen) == 0)
+            {
+            option_data = arg + arglen;
+            if (*option_data == '=')
+              {
+              option_data++;
+              longopwasequals = TRUE;
+              }
+            break;
+            }
+          }
+        }
+
+      /* Handle options with an alternate spelling of the name */
+
+      else
+        {
+        char buff1[24];
+        char buff2[24];
+
+        int baselen = (int)(opbra - op->long_name);
+        int fulllen = (int)(strchr(op->long_name, ')') - op->long_name + 1);
+        int arglen = (argequals == NULL || equals == NULL)?
+          (int)strlen(arg) : (int)(argequals - arg);
+
+        sprintf(buff1, "%.*s", baselen, op->long_name);
+        sprintf(buff2, "%s%.*s", buff1, fulllen - baselen - 2, opbra + 1);
+
+        if (strncmp(arg, buff1, arglen) == 0 ||
+           strncmp(arg, buff2, arglen) == 0)
+          {
+          if (equals != NULL && argequals != NULL)
+            {
+            option_data = argequals;
+            if (*option_data == '=')
+              {
+              option_data++;
+              longopwasequals = TRUE;
+              }
+            }
+          break;
+          }
+        }
+      }
+
+    if (op->one_char == 0)
+      {
+      fprintf(stderr, "pcregrep: Unknown option %s\n", argv[i]);
+      pcregrep_exit(usage(2));
+      }
+    }
+
+  /* Jeffrey Friedl's debugging harness uses these additional options which
+  are not in the right form for putting in the option table because they use
+  only one hyphen, yet are more than one character long. By putting them
+  separately here, they will not get displayed as part of the help() output,
+  but I don't think Jeffrey will care about that. */
+
+#ifdef JFRIEDL_DEBUG
+  else if (strcmp(argv[i], "-pre") == 0) {
+          jfriedl_prefix = argv[++i];
+          continue;
+  } else if (strcmp(argv[i], "-post") == 0) {
+          jfriedl_postfix = argv[++i];
+          continue;
+  } else if (strcmp(argv[i], "-XT") == 0) {
+          sscanf(argv[++i], "%d", &jfriedl_XT);
+          continue;
+  } else if (strcmp(argv[i], "-XR") == 0) {
+          sscanf(argv[++i], "%d", &jfriedl_XR);
+          continue;
+  }
+#endif
+
+
+  /* One-char options; many that have no data may be in a single argument; we
+  continue till we hit the last one or one that needs data. */
+
+  else
+    {
+    char *s = argv[i] + 1;
+    longop = FALSE;
+
+    while (*s != 0)
+      {
+      for (op = optionlist; op->one_char != 0; op++)
+        {
+        if (*s == op->one_char) break;
+        }
+      if (op->one_char == 0)
+        {
+        fprintf(stderr, "pcregrep: Unknown option letter '%c' in \"%s\"\n",
+          *s, argv[i]);
+        pcregrep_exit(usage(2));
+        }
+
+      option_data = s+1;
+
+      /* Break out if this is the last character in the string; it's handled
+      below like a single multi-char option. */
+
+      if (*option_data == 0) break;
+
+      /* Check for a single-character option that has data: OP_OP_NUMBER(S)
+      are used for ones that either have a numerical number or defaults, i.e.
+      the data is optional. If a digit follows, there is data; if not, carry on
+      with other single-character options in the same string. */
+
+      if (op->type == OP_OP_NUMBER || op->type == OP_OP_NUMBERS)
+        {
+        if (isdigit((unsigned char)s[1])) break;
+        }
+      else   /* Check for an option with data */
+        {
+        if (op->type != OP_NODATA) break;
+        }
+
+      /* Handle a single-character option with no data, then loop for the
+      next character in the string. */
+
+      pcre_options = handle_option(*s++, pcre_options);
+      }
+    }
+
+  /* At this point we should have op pointing to a matched option. If the type
+  is NO_DATA, it means that there is no data, and the option might set
+  something in the PCRE options. */
+
+  if (op->type == OP_NODATA)
+    {
+    pcre_options = handle_option(op->one_char, pcre_options);
+    continue;
+    }
+
+  /* If the option type is OP_OP_STRING or OP_OP_NUMBER(S), it's an option that
+  either has a value or defaults to something. It cannot have data in a
+  separate item. At the moment, the only such options are "colo(u)r",
+  "only-matching", and Jeffrey Friedl's special -S debugging option. */
+
+  if (*option_data == 0 &&
+      (op->type == OP_OP_STRING || op->type == OP_OP_NUMBER ||
+       op->type == OP_OP_NUMBERS))
+    {
+    switch (op->one_char)
+      {
+      case N_COLOUR:
+      colour_option = (char *)"auto";
+      break;
+
+      case 'o':
+      only_matching_last = add_number(0, only_matching_last);
+      if (only_matching == NULL) only_matching = only_matching_last;
+      break;
+
+#ifdef JFRIEDL_DEBUG
+      case 'S':
+      S_arg = 0;
+      break;
+#endif
+      }
+    continue;
+    }
+
+  /* Otherwise, find the data string for the option. */
+
+  if (*option_data == 0)
+    {
+    if (i >= argc - 1 || longopwasequals)
+      {
+      fprintf(stderr, "pcregrep: Data missing after %s\n", argv[i]);
+      pcregrep_exit(usage(2));
+      }
+    option_data = argv[++i];
+    }
+
+  /* If the option type is OP_OP_NUMBERS, the value is a number that is to be
+  added to a chain of numbers. */
+
+  if (op->type == OP_OP_NUMBERS)
+    {
+    unsigned long int n = decode_number(option_data, op, longop);
+    omdatastr *omd = (omdatastr *)op->dataptr;
+    *(omd->lastptr) = add_number((int)n, *(omd->lastptr));
+    if (*(omd->anchor) == NULL) *(omd->anchor) = *(omd->lastptr);
+    }
+
+  /* If the option type is OP_PATLIST, it's the -e option, or one of the
+  include/exclude options, which can be called multiple times to create lists
+  of patterns. */
+
+  else if (op->type == OP_PATLIST)
+    {
+    patdatastr *pd = (patdatastr *)op->dataptr;
+    *(pd->lastptr) = add_pattern(option_data, *(pd->lastptr));
+    if (*(pd->lastptr) == NULL) goto EXIT2;
+    if (*(pd->anchor) == NULL) *(pd->anchor) = *(pd->lastptr);
+    }
+
+  /* If the option type is OP_FILELIST, it's one of the options that names a
+  file. */
+
+  else if (op->type == OP_FILELIST)
+    {
+    fndatastr *fd = (fndatastr *)op->dataptr;
+    fn = (fnstr *)malloc(sizeof(fnstr));
+    if (fn == NULL)
+      {
+      fprintf(stderr, "pcregrep: malloc failed\n");
+      goto EXIT2;
+      }
+    fn->next = NULL;
+    fn->name = option_data;
+    if (*(fd->anchor) == NULL)
+      *(fd->anchor) = fn;
+    else
+      (*(fd->lastptr))->next = fn;
+    *(fd->lastptr) = fn;
+    }
+
+  /* Handle OP_BINARY_FILES */
+
+  else if (op->type == OP_BINFILES)
+    {
+    if (strcmp(option_data, "binary") == 0)
+      binary_files = BIN_BINARY;
+    else if (strcmp(option_data, "without-match") == 0)
+      binary_files = BIN_NOMATCH;
+    else if (strcmp(option_data, "text") == 0)
+      binary_files = BIN_TEXT;
+    else
+      {
+      fprintf(stderr, "pcregrep: unknown value \"%s\" for binary-files\n",
+        option_data);
+      pcregrep_exit(usage(2));
+      }
+    }
+
+  /* Otherwise, deal with a single string or numeric data value. */
+
+  else if (op->type != OP_NUMBER && op->type != OP_LONGNUMBER &&
+           op->type != OP_OP_NUMBER)
+    {
+    *((char **)op->dataptr) = option_data;
+    }
+  else
+    {
+    unsigned long int n = decode_number(option_data, op, longop);
+    if (op->type == OP_LONGNUMBER) *((unsigned long int *)op->dataptr) = n;
+      else *((int *)op->dataptr) = n;
+    }
+  }
+
+/* Options have been decoded. If -C was used, its value is used as a default
+for -A and -B. */
+
+if (both_context > 0)
+  {
+  if (after_context == 0) after_context = both_context;
+  if (before_context == 0) before_context = both_context;
+  }
+
+/* Only one of --only-matching, --file-offsets, or --line-offsets is permitted.
+However, all three set show_only_matching because they display, each in their
+own way, only the data that has matched. */
+
+if ((only_matching != NULL && (file_offsets || line_offsets)) ||
+    (file_offsets && line_offsets))
+  {
+  fprintf(stderr, "pcregrep: Cannot mix --only-matching, --file-offsets "
+    "and/or --line-offsets\n");
+  pcregrep_exit(usage(2));
+  }
+
+if (only_matching != NULL || file_offsets || line_offsets)
+  show_only_matching = TRUE;
+
+/* If a locale has not been provided as an option, see if the LC_CTYPE or
+LC_ALL environment variable is set, and if so, use it. */
+
+if (locale == NULL)
+  {
+  locale = getenv("LC_ALL");
+  locale_from = "LCC_ALL";
+  }
+
+if (locale == NULL)
+  {
+  locale = getenv("LC_CTYPE");
+  locale_from = "LC_CTYPE";
+  }
+
+/* If a locale has been provided, set it, and generate the tables the PCRE
+needs. Otherwise, pcretables==NULL, which causes the use of default tables. */
+
+if (locale != NULL)
+  {
+  if (setlocale(LC_CTYPE, locale) == NULL)
+    {
+    fprintf(stderr, "pcregrep: Failed to set locale %s (obtained from %s)\n",
+      locale, locale_from);
+    return 2;
+    }
+  pcretables = pcre_maketables();
+  }
+
+/* Sort out colouring */
+
+if (colour_option != NULL && strcmp(colour_option, "never") != 0)
+  {
+  if (strcmp(colour_option, "always") == 0) do_colour = TRUE;
+  else if (strcmp(colour_option, "auto") == 0) do_colour = is_stdout_tty();
+  else
+    {
+    fprintf(stderr, "pcregrep: Unknown colour setting \"%s\"\n",
+      colour_option);
+    return 2;
+    }
+  if (do_colour)
+    {
+    char *cs = getenv("PCREGREP_COLOUR");
+    if (cs == NULL) cs = getenv("PCREGREP_COLOR");
+    if (cs != NULL) colour_string = cs;
+    }
+  }
+
+/* Interpret the newline type; the default settings are Unix-like. */
+
+if (strcmp(newline, "cr") == 0 || strcmp(newline, "CR") == 0)
+  {
+  pcre_options |= PCRE_NEWLINE_CR;
+  endlinetype = EL_CR;
+  }
+else if (strcmp(newline, "lf") == 0 || strcmp(newline, "LF") == 0)
+  {
+  pcre_options |= PCRE_NEWLINE_LF;
+  endlinetype = EL_LF;
+  }
+else if (strcmp(newline, "crlf") == 0 || strcmp(newline, "CRLF") == 0)
+  {
+  pcre_options |= PCRE_NEWLINE_CRLF;
+  endlinetype = EL_CRLF;
+  }
+else if (strcmp(newline, "any") == 0 || strcmp(newline, "ANY") == 0)
+  {
+  pcre_options |= PCRE_NEWLINE_ANY;
+  endlinetype = EL_ANY;
+  }
+else if (strcmp(newline, "anycrlf") == 0 || strcmp(newline, "ANYCRLF") == 0)
+  {
+  pcre_options |= PCRE_NEWLINE_ANYCRLF;
+  endlinetype = EL_ANYCRLF;
+  }
+else
+  {
+  fprintf(stderr, "pcregrep: Invalid newline specifier \"%s\"\n", newline);
+  return 2;
+  }
+
+/* Interpret the text values for -d and -D */
+
+if (dee_option != NULL)
+  {
+  if (strcmp(dee_option, "read") == 0) dee_action = dee_READ;
+  else if (strcmp(dee_option, "recurse") == 0) dee_action = dee_RECURSE;
+  else if (strcmp(dee_option, "skip") == 0) dee_action = dee_SKIP;
+  else
+    {
+    fprintf(stderr, "pcregrep: Invalid value \"%s\" for -d\n", dee_option);
+    return 2;
+    }
+  }
+
+if (DEE_option != NULL)
+  {
+  if (strcmp(DEE_option, "read") == 0) DEE_action = DEE_READ;
+  else if (strcmp(DEE_option, "skip") == 0) DEE_action = DEE_SKIP;
+  else
+    {
+    fprintf(stderr, "pcregrep: Invalid value \"%s\" for -D\n", DEE_option);
+    return 2;
+    }
+  }
+
+/* Check the values for Jeffrey Friedl's debugging options. */
+
+#ifdef JFRIEDL_DEBUG
+if (S_arg > 9)
+  {
+  fprintf(stderr, "pcregrep: bad value for -S option\n");
+  return 2;
+  }
+if (jfriedl_XT != 0 || jfriedl_XR != 0)
+  {
+  if (jfriedl_XT == 0) jfriedl_XT = 1;
+  if (jfriedl_XR == 0) jfriedl_XR = 1;
+  }
+#endif
+
+/* Get memory for the main buffer. */
+
+bufsize = 3*bufthird;
+main_buffer = (char *)malloc(bufsize);
+
+if (main_buffer == NULL)
+  {
+  fprintf(stderr, "pcregrep: malloc failed\n");
+  goto EXIT2;
+  }
+
+/* If no patterns were provided by -e, and there are no files provided by -f,
+the first argument is the one and only pattern, and it must exist. */
+
+if (patterns == NULL && pattern_files == NULL)
+  {
+  if (i >= argc) return usage(2);
+  patterns = patterns_last = add_pattern(argv[i++], NULL);
+  if (patterns == NULL) goto EXIT2;
+  }
+
+/* Compile the patterns that were provided on the command line, either by
+multiple uses of -e or as a single unkeyed pattern. We cannot do this until
+after all the command-line options are read so that we know which PCRE options
+to use. When -F is used, compile_pattern() may add another block into the
+chain, so we must not access the next pointer till after the compile. */
+
+for (j = 1, cp = patterns; cp != NULL; j++, cp = cp->next)
+  {
+  if (!compile_pattern(cp, pcre_options, process_options, FALSE, "command-line",
+       (j == 1 && patterns->next == NULL)? 0 : j))
+    goto EXIT2;
+  }
+
+/* Read and compile the regular expressions that are provided in files. */
+
+for (fn = pattern_files; fn != NULL; fn = fn->next)
+  {
+  if (!read_pattern_file(fn->name, &patterns, &patterns_last, process_options))
+    goto EXIT2;
+  }
+
+/* Study the regular expressions, as we will be running them many times. If an
+extra block is needed for a limit, set PCRE_STUDY_EXTRA_NEEDED so that one is
+returned, even if studying produces no data. */
+
+if (match_limit > 0 || match_limit_recursion > 0)
+  study_options |= PCRE_STUDY_EXTRA_NEEDED;
+
+/* Unless JIT has been explicitly disabled, arrange a stack for it to use. */
+
+#ifdef SUPPORT_PCREGREP_JIT
+if ((study_options & PCRE_STUDY_JIT_COMPILE) != 0)
+  jit_stack = pcre_jit_stack_alloc(32*1024, 1024*1024);
+#endif
+
+for (j = 1, cp = patterns; cp != NULL; j++, cp = cp->next)
+  {
+  cp->hint = pcre_study(cp->compiled, study_options, &error);
+  if (error != NULL)
+    {
+    char s[16];
+    if (patterns->next == NULL) s[0] = 0; else sprintf(s, " number %d", j);
+    fprintf(stderr, "pcregrep: Error while studying regex%s: %s\n", s, error);
+    goto EXIT2;
+    }
+#ifdef SUPPORT_PCREGREP_JIT
+  if (jit_stack != NULL && cp->hint != NULL)
+    pcre_assign_jit_stack(cp->hint, NULL, jit_stack);
+#endif
+  }
+
+/* If --match-limit or --recursion-limit was set, put the value(s) into the
+pcre_extra block for each pattern. There will always be an extra block because
+of the use of PCRE_STUDY_EXTRA_NEEDED above. */
+
+for (cp = patterns; cp != NULL; cp = cp->next)
+  {
+  if (match_limit > 0)
+    {
+    cp->hint->flags |= PCRE_EXTRA_MATCH_LIMIT;
+    cp->hint->match_limit = match_limit;
+    }
+
+  if (match_limit_recursion > 0)
+    {
+    cp->hint->flags |= PCRE_EXTRA_MATCH_LIMIT_RECURSION;
+    cp->hint->match_limit_recursion = match_limit_recursion;
+    }
+  }
+
+/* If there are include or exclude patterns read from the command line, compile
+them. -F, -w, and -x do not apply, so the third argument of compile_pattern is
+0. */
+
+for (j = 0; j < 4; j++)
+  {
+  int k;
+  for (k = 1, cp = *(incexlist[j]); cp != NULL; k++, cp = cp->next)
+    {
+    if (!compile_pattern(cp, pcre_options, 0, FALSE, incexname[j],
+         (k == 1 && cp->next == NULL)? 0 : k))
+      goto EXIT2;
+    }
+  }
+
+/* Read and compile include/exclude patterns from files. */
+
+for (fn = include_from; fn != NULL; fn = fn->next)
+  {
+  if (!read_pattern_file(fn->name, &include_patterns, &include_patterns_last, 0))
+    goto EXIT2;
+  }
+
+for (fn = exclude_from; fn != NULL; fn = fn->next)
+  {
+  if (!read_pattern_file(fn->name, &exclude_patterns, &exclude_patterns_last, 0))
+    goto EXIT2;
+  }
+
+/* If there are no files that contain lists of files to search, and there are
+no file arguments, search stdin, and then exit. */
+
+if (file_lists == NULL && i >= argc)
+  {
+  rc = pcregrep(stdin, FR_PLAIN, stdin_name,
+    (filenames > FN_DEFAULT)? stdin_name : NULL);
+  goto EXIT;
+  }
+
+/* If any files that contains a list of files to search have been specified,
+read them line by line and search the given files. */
+
+for (fn = file_lists; fn != NULL; fn = fn->next)
+  {
+  char buffer[PATBUFSIZE];
+  FILE *fl;
+  if (strcmp(fn->name, "-") == 0) fl = stdin; else
+    {
+    fl = fopen(fn->name, "rb");
+    if (fl == NULL)
+      {
+      fprintf(stderr, "pcregrep: Failed to open %s: %s\n", fn->name,
+        strerror(errno));
+      goto EXIT2;
+      }
+    }
+  while (fgets(buffer, PATBUFSIZE, fl) != NULL)
+    {
+    int frc;
+    char *end = buffer + (int)strlen(buffer);
+    while (end > buffer && isspace(end[-1])) end--;
+    *end = 0;
+    if (*buffer != 0)
+      {
+      frc = grep_or_recurse(buffer, dee_action == dee_RECURSE, FALSE);
+      if (frc > 1) rc = frc;
+        else if (frc == 0 && rc == 1) rc = 0;
+      }
+    }
+  if (fl != stdin) fclose(fl);
+  }
+
+/* After handling file-list, work through remaining arguments. Pass in the fact
+that there is only one argument at top level - this suppresses the file name if
+the argument is not a directory and filenames are not otherwise forced. */
+
+only_one_at_top = i == argc - 1 && file_lists == NULL;
+
+for (; i < argc; i++)
+  {
+  int frc = grep_or_recurse(argv[i], dee_action == dee_RECURSE,
+    only_one_at_top);
+  if (frc > 1) rc = frc;
+    else if (frc == 0 && rc == 1) rc = 0;
+  }
+
+EXIT:
+#ifdef SUPPORT_PCREGREP_JIT
+if (jit_stack != NULL) pcre_jit_stack_free(jit_stack);
+#endif
+
+if (main_buffer != NULL) free(main_buffer);
+
+free_pattern_chain(patterns);
+free_pattern_chain(include_patterns);
+free_pattern_chain(include_dir_patterns);
+free_pattern_chain(exclude_patterns);
+free_pattern_chain(exclude_dir_patterns);
+
+free_file_chain(exclude_from);
+free_file_chain(include_from);
+free_file_chain(pattern_files);
+free_file_chain(file_lists);
+
+while (only_matching != NULL)
+  {
+  omstr *this = only_matching;
+  only_matching = this->next;
+  free(this);
+  }
+
+pcregrep_exit(rc);
+
+EXIT2:
+rc = 2;
+goto EXIT;
+}
+
+/* End of pcregrep */
diff --git a/pcre/pcreposix.c b/pcre/pcreposix.c
new file mode 100644
index 00000000000..95fa39d6fe8
--- /dev/null
+++ b/pcre/pcreposix.c
@@ -0,0 +1,420 @@
+/*************************************************
+*      Perl-Compatible Regular Expressions       *
+*************************************************/
+
+/* PCRE is a library of functions to support regular expressions whose syntax
+and semantics are as close as possible to those of the Perl 5 language.
+
+                       Written by Philip Hazel
+           Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+
+/* This module is a wrapper that provides a POSIX API to the underlying PCRE
+functions. */
+
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+
+/* Ensure that the PCREPOSIX_EXP_xxx macros are set appropriately for
+compiling these functions. This must come before including pcreposix.h, where
+they are set for an application (using these functions) if they have not
+previously been set. */
+
+#if defined(_WIN32) && !defined(PCRE_STATIC)
+#  define PCREPOSIX_EXP_DECL extern __declspec(dllexport)
+#  define PCREPOSIX_EXP_DEFN __declspec(dllexport)
+#endif
+
+/* We include pcre.h before pcre_internal.h so that the PCRE library functions
+are declared as "import" for Windows by defining PCRE_EXP_DECL as "import".
+This is needed even though pcre_internal.h itself includes pcre.h, because it
+does so after it has set PCRE_EXP_DECL to "export" if it is not already set. */
+
+#include "pcre.h"
+#include "pcre_internal.h"
+#include "pcreposix.h"
+
+
+/* Table to translate PCRE compile time error codes into POSIX error codes. */
+
+static const int eint[] = {
+  0,           /* no error */
+  REG_EESCAPE, /* \ at end of pattern */
+  REG_EESCAPE, /* \c at end of pattern */
+  REG_EESCAPE, /* unrecognized character follows \ */
+  REG_BADBR,   /* numbers out of order in {} quantifier */
+  /* 5 */
+  REG_BADBR,   /* number too big in {} quantifier */
+  REG_EBRACK,  /* missing terminating ] for character class */
+  REG_ECTYPE,  /* invalid escape sequence in character class */
+  REG_ERANGE,  /* range out of order in character class */
+  REG_BADRPT,  /* nothing to repeat */
+  /* 10 */
+  REG_BADRPT,  /* operand of unlimited repeat could match the empty string */
+  REG_ASSERT,  /* internal error: unexpected repeat */
+  REG_BADPAT,  /* unrecognized character after (? */
+  REG_BADPAT,  /* POSIX named classes are supported only within a class */
+  REG_EPAREN,  /* missing ) */
+  /* 15 */
+  REG_ESUBREG, /* reference to non-existent subpattern */
+  REG_INVARG,  /* erroffset passed as NULL */
+  REG_INVARG,  /* unknown option bit(s) set */
+  REG_EPAREN,  /* missing ) after comment */
+  REG_ESIZE,   /* parentheses nested too deeply */
+  /* 20 */
+  REG_ESIZE,   /* regular expression too large */
+  REG_ESPACE,  /* failed to get memory */
+  REG_EPAREN,  /* unmatched parentheses */
+  REG_ASSERT,  /* internal error: code overflow */
+  REG_BADPAT,  /* unrecognized character after (?< */
+  /* 25 */
+  REG_BADPAT,  /* lookbehind assertion is not fixed length */
+  REG_BADPAT,  /* malformed number or name after (?( */
+  REG_BADPAT,  /* conditional group contains more than two branches */
+  REG_BADPAT,  /* assertion expected after (?( */
+  REG_BADPAT,  /* (?R or (?[+-]digits must be followed by ) */
+  /* 30 */
+  REG_ECTYPE,  /* unknown POSIX class name */
+  REG_BADPAT,  /* POSIX collating elements are not supported */
+  REG_INVARG,  /* this version of PCRE is not compiled with PCRE_UTF8 support */
+  REG_BADPAT,  /* spare error */
+  REG_BADPAT,  /* character value in \x{...} sequence is too large */
+  /* 35 */
+  REG_BADPAT,  /* invalid condition (?(0) */
+  REG_BADPAT,  /* \C not allowed in lookbehind assertion */
+  REG_EESCAPE, /* PCRE does not support \L, \l, \N, \U, or \u */
+  REG_BADPAT,  /* number after (?C is > 255 */
+  REG_BADPAT,  /* closing ) for (?C expected */
+  /* 40 */
+  REG_BADPAT,  /* recursive call could loop indefinitely */
+  REG_BADPAT,  /* unrecognized character after (?P */
+  REG_BADPAT,  /* syntax error in subpattern name (missing terminator) */
+  REG_BADPAT,  /* two named subpatterns have the same name */
+  REG_BADPAT,  /* invalid UTF-8 string */
+  /* 45 */
+  REG_BADPAT,  /* support for \P, \p, and \X has not been compiled */
+  REG_BADPAT,  /* malformed \P or \p sequence */
+  REG_BADPAT,  /* unknown property name after \P or \p */
+  REG_BADPAT,  /* subpattern name is too long (maximum 32 characters) */
+  REG_BADPAT,  /* too many named subpatterns (maximum 10,000) */
+  /* 50 */
+  REG_BADPAT,  /* repeated subpattern is too long */
+  REG_BADPAT,  /* octal value is greater than \377 (not in UTF-8 mode) */
+  REG_BADPAT,  /* internal error: overran compiling workspace */
+  REG_BADPAT,  /* internal error: previously-checked referenced subpattern not found */
+  REG_BADPAT,  /* DEFINE group contains more than one branch */
+  /* 55 */
+  REG_BADPAT,  /* repeating a DEFINE group is not allowed */
+  REG_INVARG,  /* inconsistent NEWLINE options */
+  REG_BADPAT,  /* \g is not followed followed by an (optionally braced) non-zero number */
+  REG_BADPAT,  /* a numbered reference must not be zero */
+  REG_BADPAT,  /* an argument is not allowed for (*ACCEPT), (*FAIL), or (*COMMIT) */
+  /* 60 */
+  REG_BADPAT,  /* (*VERB) not recognized */
+  REG_BADPAT,  /* number is too big */
+  REG_BADPAT,  /* subpattern name expected */
+  REG_BADPAT,  /* digit expected after (?+ */
+  REG_BADPAT,  /* ] is an invalid data character in JavaScript compatibility mode */
+  /* 65 */
+  REG_BADPAT,  /* different names for subpatterns of the same number are not allowed */
+  REG_BADPAT,  /* (*MARK) must have an argument */
+  REG_INVARG,  /* this version of PCRE is not compiled with PCRE_UCP support */
+  REG_BADPAT,  /* \c must be followed by an ASCII character */
+  REG_BADPAT,  /* \k is not followed by a braced, angle-bracketed, or quoted name */
+  /* 70 */
+  REG_BADPAT,  /* internal error: unknown opcode in find_fixedlength() */
+  REG_BADPAT,  /* \N is not supported in a class */
+  REG_BADPAT,  /* too many forward references */
+  REG_BADPAT,  /* disallowed UTF-8/16/32 code point (>= 0xd800 && <= 0xdfff) */
+  REG_BADPAT,  /* invalid UTF-16 string (should not occur) */
+  /* 75 */
+  REG_BADPAT,  /* overlong MARK name */
+  REG_BADPAT,  /* character value in \u.... sequence is too large */
+  REG_BADPAT,  /* invalid UTF-32 string (should not occur) */
+  REG_BADPAT   /* setting UTF is disabled by the application */
+};
+
+/* Table of texts corresponding to POSIX error codes */
+
+static const char *const pstring[] = {
+  "",                                /* Dummy for value 0 */
+  "internal error",                  /* REG_ASSERT */
+  "invalid repeat counts in {}",     /* BADBR      */
+  "pattern error",                   /* BADPAT     */
+  "? * + invalid",                   /* BADRPT     */
+  "unbalanced {}",                   /* EBRACE     */
+  "unbalanced []",                   /* EBRACK     */
+  "collation error - not relevant",  /* ECOLLATE   */
+  "bad class",                       /* ECTYPE     */
+  "bad escape sequence",             /* EESCAPE    */
+  "empty expression",                /* EMPTY      */
+  "unbalanced ()",                   /* EPAREN     */
+  "bad range inside []",             /* ERANGE     */
+  "expression too big",              /* ESIZE      */
+  "failed to get memory",            /* ESPACE     */
+  "bad back reference",              /* ESUBREG    */
+  "bad argument",                    /* INVARG     */
+  "match failed"                     /* NOMATCH    */
+};
+
+
+
+
+/*************************************************
+*          Translate error code to string        *
+*************************************************/
+
+PCREPOSIX_EXP_DEFN size_t PCRE_CALL_CONVENTION
+regerror(int errcode, const regex_t *preg, char *errbuf, size_t errbuf_size)
+{
+const char *message, *addmessage;
+size_t length, addlength;
+
+message = (errcode >= (int)(sizeof(pstring)/sizeof(char *)))?
+  "unknown error code" : pstring[errcode];
+length = strlen(message) + 1;
+
+addmessage = " at offset ";
+addlength = (preg != NULL && (int)preg->re_erroffset != -1)?
+  strlen(addmessage) + 6 : 0;
+
+if (errbuf_size > 0)
+  {
+  if (addlength > 0 && errbuf_size >= length + addlength)
+    sprintf(errbuf, "%s%s%-6d", message, addmessage, (int)preg->re_erroffset);
+  else
+    {
+    strncpy(errbuf, message, errbuf_size - 1);
+    errbuf[errbuf_size-1] = 0;
+    }
+  }
+
+return length + addlength;
+}
+
+
+
+
+/*************************************************
+*           Free store held by a regex           *
+*************************************************/
+
+PCREPOSIX_EXP_DEFN void PCRE_CALL_CONVENTION
+regfree(regex_t *preg)
+{
+(PUBL(free))(preg->re_pcre);
+}
+
+
+
+
+/*************************************************
+*            Compile a regular expression        *
+*************************************************/
+
+/*
+Arguments:
+  preg        points to a structure for recording the compiled expression
+  pattern     the pattern to compile
+  cflags      compilation flags
+
+Returns:      0 on success
+              various non-zero codes on failure
+*/
+
+PCREPOSIX_EXP_DEFN int PCRE_CALL_CONVENTION
+regcomp(regex_t *preg, const char *pattern, int cflags)
+{
+const char *errorptr;
+int erroffset;
+int errorcode;
+int options = 0;
+int re_nsub = 0;
+
+if ((cflags & REG_ICASE) != 0)    options |= PCRE_CASELESS;
+if ((cflags & REG_NEWLINE) != 0)  options |= PCRE_MULTILINE;
+if ((cflags & REG_DOTALL) != 0)   options |= PCRE_DOTALL;
+if ((cflags & REG_NOSUB) != 0)    options |= PCRE_NO_AUTO_CAPTURE;
+if ((cflags & REG_UTF8) != 0)     options |= PCRE_UTF8;
+if ((cflags & REG_UCP) != 0)      options |= PCRE_UCP;
+if ((cflags & REG_UNGREEDY) != 0) options |= PCRE_UNGREEDY;
+
+preg->re_pcre = pcre_compile2(pattern, options, &errorcode, &errorptr,
+  &erroffset, NULL);
+preg->re_erroffset = erroffset;
+
+/* Safety: if the error code is too big for the translation vector (which
+should not happen, but we all make mistakes), return REG_BADPAT. */
+
+if (preg->re_pcre == NULL)
+  {
+  return (errorcode < (int)(sizeof(eint)/sizeof(const int)))?
+    eint[errorcode] : REG_BADPAT;
+  }
+
+(void)pcre_fullinfo((const pcre *)preg->re_pcre, NULL, PCRE_INFO_CAPTURECOUNT,
+  &re_nsub);
+preg->re_nsub = (size_t)re_nsub;
+return 0;
+}
+
+
+
+
+/*************************************************
+*              Match a regular expression        *
+*************************************************/
+
+/* Unfortunately, PCRE requires 3 ints of working space for each captured
+substring, so we have to get and release working store instead of just using
+the POSIX structures as was done in earlier releases when PCRE needed only 2
+ints. However, if the number of possible capturing brackets is small, use a
+block of store on the stack, to reduce the use of malloc/free. The threshold is
+in a macro that can be changed at configure time.
+
+If REG_NOSUB was specified at compile time, the PCRE_NO_AUTO_CAPTURE flag will
+be set. When this is the case, the nmatch and pmatch arguments are ignored, and
+the only result is yes/no/error. */
+
+PCREPOSIX_EXP_DEFN int PCRE_CALL_CONVENTION
+regexec(const regex_t *preg, const char *string, size_t nmatch,
+  regmatch_t pmatch[], int eflags)
+{
+int rc, so, eo;
+int options = 0;
+int *ovector = NULL;
+int small_ovector[POSIX_MALLOC_THRESHOLD * 3];
+BOOL allocated_ovector = FALSE;
+BOOL nosub =
+  (REAL_PCRE_OPTIONS((const pcre *)preg->re_pcre) & PCRE_NO_AUTO_CAPTURE) != 0;
+
+if ((eflags & REG_NOTBOL) != 0) options |= PCRE_NOTBOL;
+if ((eflags & REG_NOTEOL) != 0) options |= PCRE_NOTEOL;
+if ((eflags & REG_NOTEMPTY) != 0) options |= PCRE_NOTEMPTY;
+
+((regex_t *)preg)->re_erroffset = (size_t)(-1);  /* Only has meaning after compile */
+
+/* When no string data is being returned, or no vector has been passed in which
+to put it, ensure that nmatch is zero. Otherwise, ensure the vector for holding
+the return data is large enough. */
+
+if (nosub || pmatch == NULL) nmatch = 0;
+
+else if (nmatch > 0)
+  {
+  if (nmatch <= POSIX_MALLOC_THRESHOLD)
+    {
+    ovector = &(small_ovector[0]);
+    }
+  else
+    {
+    if (nmatch > INT_MAX/(sizeof(int) * 3)) return REG_ESPACE;
+    ovector = (int *)malloc(sizeof(int) * nmatch * 3);
+    if (ovector == NULL) return REG_ESPACE;
+    allocated_ovector = TRUE;
+    }
+  }
+
+/* REG_STARTEND is a BSD extension, to allow for non-NUL-terminated strings.
+The man page from OS X says "REG_STARTEND affects only the location of the
+string, not how it is matched". That is why the "so" value is used to bump the
+start location rather than being passed as a PCRE "starting offset". */
+
+if ((eflags & REG_STARTEND) != 0)
+  {
+  so = pmatch[0].rm_so;
+  eo = pmatch[0].rm_eo;
+  }
+else
+  {
+  so = 0;
+  eo = (int)strlen(string);
+  }
+
+rc = pcre_exec((const pcre *)preg->re_pcre, NULL, string + so, (eo - so),
+  0, options, ovector, (int)(nmatch * 3));
+
+if (rc == 0) rc = (int)nmatch;    /* All captured slots were filled in */
+
+/* Successful match */
+
+if (rc >= 0)
+  {
+  size_t i;
+  if (!nosub)
+    {
+    for (i = 0; i < (size_t)rc; i++)
+      {
+      pmatch[i].rm_so = ovector[i*2];
+      pmatch[i].rm_eo = ovector[i*2+1];
+      }
+    if (allocated_ovector) free(ovector);
+    for (; i < nmatch; i++) pmatch[i].rm_so = pmatch[i].rm_eo = -1;
+    }
+  return 0;
+  }
+
+/* Unsuccessful match */
+
+if (allocated_ovector) free(ovector);
+switch(rc)
+  {
+/* ========================================================================== */
+  /* These cases are never obeyed. This is a fudge that causes a compile-time
+  error if the vector eint, which is indexed by compile-time error number, is
+  not the correct length. It seems to be the only way to do such a check at
+  compile time, as the sizeof() operator does not work in the C preprocessor.
+  As all the PCRE_ERROR_xxx values are negative, we can use 0 and 1. */
+
+  case 0:
+  case (sizeof(eint)/sizeof(int) == ERRCOUNT):
+  return REG_ASSERT;
+/* ========================================================================== */
+
+  case PCRE_ERROR_NOMATCH: return REG_NOMATCH;
+  case PCRE_ERROR_NULL: return REG_INVARG;
+  case PCRE_ERROR_BADOPTION: return REG_INVARG;
+  case PCRE_ERROR_BADMAGIC: return REG_INVARG;
+  case PCRE_ERROR_UNKNOWN_NODE: return REG_ASSERT;
+  case PCRE_ERROR_NOMEMORY: return REG_ESPACE;
+  case PCRE_ERROR_MATCHLIMIT: return REG_ESPACE;
+  case PCRE_ERROR_BADUTF8: return REG_INVARG;
+  case PCRE_ERROR_BADUTF8_OFFSET: return REG_INVARG;
+  case PCRE_ERROR_BADMODE: return REG_INVARG;
+  default: return REG_ASSERT;
+  }
+}
+
+/* End of pcreposix.c */
diff --git a/pcre/pcreposix.h b/pcre/pcreposix.h
new file mode 100644
index 00000000000..b900a9697f3
--- /dev/null
+++ b/pcre/pcreposix.h
@@ -0,0 +1,150 @@
+/*************************************************
+*       Perl-Compatible Regular Expressions      *
+*************************************************/
+
+#ifndef _PCREPOSIX_H
+#define _PCREPOSIX_H
+
+/* This is the header for the POSIX wrapper interface to the PCRE Perl-
+Compatible Regular Expression library. It defines the things POSIX says should
+be there. I hope.
+
+            Copyright (c) 1997-2012 University of Cambridge
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+/* Have to include stdlib.h in order to ensure that size_t is defined. */
+
+#include 
+
+/* Allow for C++ users */
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+/* Options, mostly defined by POSIX, but with some extras. */
+
+#define REG_ICASE     0x0001   /* Maps to PCRE_CASELESS */
+#define REG_NEWLINE   0x0002   /* Maps to PCRE_MULTILINE */
+#define REG_NOTBOL    0x0004   /* Maps to PCRE_NOTBOL */
+#define REG_NOTEOL    0x0008   /* Maps to PCRE_NOTEOL */
+#define REG_DOTALL    0x0010   /* NOT defined by POSIX; maps to PCRE_DOTALL */
+#define REG_NOSUB     0x0020   /* Maps to PCRE_NO_AUTO_CAPTURE */
+#define REG_UTF8      0x0040   /* NOT defined by POSIX; maps to PCRE_UTF8 */
+#define REG_STARTEND  0x0080   /* BSD feature: pass subject string by so,eo */
+#define REG_NOTEMPTY  0x0100   /* NOT defined by POSIX; maps to PCRE_NOTEMPTY */
+#define REG_UNGREEDY  0x0200   /* NOT defined by POSIX; maps to PCRE_UNGREEDY */
+#define REG_UCP       0x0400   /* NOT defined by POSIX; maps to PCRE_UCP */
+
+/*Defined in regex.h */
+#define REG_ATOI        255     /* convert name to number (!) */
+#define REG_ITOA        0400    /* convert number to name (!) */
+
+/* This is not used by PCRE, but by defining it we make it easier
+to slot PCRE into existing programs that make POSIX calls. */
+
+#define REG_EXTENDED  0
+
+/* Error values. Not all these are relevant or used by the wrapper. */
+
+enum {
+  REG_ASSERT = 1,  /* internal error ? */
+  REG_BADBR,       /* invalid repeat counts in {} */
+  REG_BADPAT,      /* pattern error */
+  REG_BADRPT,      /* ? * + invalid */
+  REG_EBRACE,      /* unbalanced {} */
+  REG_EBRACK,      /* unbalanced [] */
+  REG_ECOLLATE,    /* collation error - not relevant */
+  REG_ECTYPE,      /* bad class */
+  REG_EESCAPE,     /* bad escape sequence */
+  REG_EMPTY,       /* empty expression */
+  REG_EPAREN,      /* unbalanced () */
+  REG_ERANGE,      /* bad range inside [] */
+  REG_ESIZE,       /* expression too big */
+  REG_ESPACE,      /* failed to get memory */
+  REG_ESUBREG,     /* bad back reference */
+  REG_INVARG,      /* bad argument */
+  REG_NOMATCH      /* match failed */
+};
+
+
+/* The structure representing a compiled regular expression. */
+
+typedef struct {
+  void *re_pcre;
+  size_t re_nsub;
+  size_t re_erroffset;
+} regex_t;
+
+/* The structure in which a captured offset is returned. */
+
+typedef int regoff_t;
+
+typedef struct {
+  regoff_t rm_so;
+  regoff_t rm_eo;
+} regmatch_t;
+
+/* When an application links to a PCRE DLL in Windows, the symbols that are
+imported have to be identified as such. When building PCRE, the appropriate
+export settings are needed, and are set in pcreposix.c before including this
+file. */
+
+#if defined(_WIN32) && !defined(PCRE_STATIC) && !defined(PCREPOSIX_EXP_DECL)
+#  define PCREPOSIX_EXP_DECL  extern __declspec(dllimport)
+#  define PCREPOSIX_EXP_DEFN  __declspec(dllimport)
+#endif
+
+/* By default, we use the standard "extern" declarations. */
+
+#ifndef PCREPOSIX_EXP_DECL
+#  ifdef __cplusplus
+#    define PCREPOSIX_EXP_DECL  extern "C"
+#    define PCREPOSIX_EXP_DEFN  extern "C"
+#  else
+#    define PCREPOSIX_EXP_DECL  extern
+#    define PCREPOSIX_EXP_DEFN  extern
+#  endif
+#endif
+
+/* The functions */
+
+PCREPOSIX_EXP_DECL int regcomp(regex_t *, const char *, int);
+PCREPOSIX_EXP_DECL int regexec(const regex_t *, const char *, size_t,
+                     regmatch_t *, int);
+PCREPOSIX_EXP_DECL size_t regerror(int, const regex_t *, char *, size_t);
+PCREPOSIX_EXP_DECL void regfree(regex_t *);
+
+#ifdef __cplusplus
+}   /* extern "C" */
+#endif
+
+#endif /* End of pcreposix.h */
diff --git a/pcre/pcretest.c b/pcre/pcretest.c
new file mode 100644
index 00000000000..57ee041ef66
--- /dev/null
+++ b/pcre/pcretest.c
@@ -0,0 +1,5524 @@
+/*************************************************
+*             PCRE testing program               *
+*************************************************/
+
+/* This program was hacked up as a tester for PCRE. I really should have
+written it more tidily in the first place. Will I ever learn? It has grown and
+been extended and consequently is now rather, er, *very* untidy in places. The
+addition of 16-bit support has made it even worse. :-(
+
+-----------------------------------------------------------------------------
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the name of the University of Cambridge nor the names of its
+      contributors may be used to endorse or promote products derived from
+      this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+-----------------------------------------------------------------------------
+*/
+
+/* This program now supports the testing of all of the 8-bit, 16-bit, and
+32-bit PCRE libraries in a single program. This is different from the modules
+such as pcre_compile.c in the library itself, which are compiled separately for
+each mode. If two modes are enabled, for example, pcre_compile.c is compiled
+twice. By contrast, pcretest.c is compiled only once. Therefore, it must not
+make use of any of the macros from pcre_internal.h that depend on
+COMPILE_PCRE8, COMPILE_PCRE16, or COMPILE_PCRE32. It does, however, make use of
+SUPPORT_PCRE8, SUPPORT_PCRE16, and SUPPORT_PCRE32 to ensure that it calls only
+supported library functions. */
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+
+/* Both libreadline and libedit are optionally supported. The user-supplied
+original patch uses readline/readline.h for libedit, but in at least one system
+it is installed as editline/readline.h, so the configuration code now looks for
+that first, falling back to readline/readline.h. */
+
+#if defined(SUPPORT_LIBREADLINE) || defined(SUPPORT_LIBEDIT)
+#ifdef HAVE_UNISTD_H
+#include 
+#endif
+#if defined(SUPPORT_LIBREADLINE)
+#include 
+#include 
+#else
+#if defined(HAVE_EDITLINE_READLINE_H)
+#include 
+#else
+#include 
+#endif
+#endif
+#endif
+
+/* A number of things vary for Windows builds. Originally, pcretest opened its
+input and output without "b"; then I was told that "b" was needed in some
+environments, so it was added for release 5.0 to both the input and output. (It
+makes no difference on Unix-like systems.) Later I was told that it is wrong
+for the input on Windows. I've now abstracted the modes into two macros that
+are set here, to make it easier to fiddle with them, and removed "b" from the
+input mode under Windows. */
+
+#if defined(_WIN32) || defined(WIN32)
+#include                 /* For _setmode() */
+#include              /* For _O_BINARY */
+#define INPUT_MODE   "r"
+#define OUTPUT_MODE  "wb"
+
+#ifndef isatty
+#define isatty _isatty         /* This is what Windows calls them, I'm told, */
+#endif                         /* though in some environments they seem to   */
+                               /* be already defined, hence the #ifndefs.    */
+#ifndef fileno
+#define fileno _fileno
+#endif
+
+/* A user sent this fix for Borland Builder 5 under Windows. */
+
+#ifdef __BORLANDC__
+#define _setmode(handle, mode) setmode(handle, mode)
+#endif
+
+/* Not Windows */
+
+#else
+#include           /* These two includes are needed */
+#include       /* for setrlimit(). */
+#if defined NATIVE_ZOS         /* z/OS uses non-binary I/O */
+#define INPUT_MODE   "r"
+#define OUTPUT_MODE  "w"
+#else
+#define INPUT_MODE   "rb"
+#define OUTPUT_MODE  "wb"
+#endif
+#endif
+
+#ifdef __VMS
+#include 
+void vms_setsymbol( char *, char *, int );
+#endif
+
+
+#define PRIV(name) name
+
+/* We have to include pcre_internal.h because we need the internal info for
+displaying the results of pcre_study() and we also need to know about the
+internal macros, structures, and other internal data values; pcretest has
+"inside information" compared to a program that strictly follows the PCRE API.
+
+Although pcre_internal.h does itself include pcre.h, we explicitly include it
+here before pcre_internal.h so that the PCRE_EXP_xxx macros get set
+appropriately for an application, not for building PCRE. */
+
+#include "pcre.h"
+#include "pcre_internal.h"
+
+/* The pcre_printint() function, which prints the internal form of a compiled
+regex, is held in a separate file so that (a) it can be compiled in either
+8-, 16- or 32-bit mode, and (b) it can be #included directly in pcre_compile.c
+when that is compiled in debug mode. */
+
+#ifdef SUPPORT_PCRE8
+void pcre_printint(pcre *external_re, FILE *f, BOOL print_lengths);
+#endif
+#ifdef SUPPORT_PCRE16
+void pcre16_printint(pcre *external_re, FILE *f, BOOL print_lengths);
+#endif
+#ifdef SUPPORT_PCRE32
+void pcre32_printint(pcre *external_re, FILE *f, BOOL print_lengths);
+#endif
+
+/* We need access to some of the data tables that PCRE uses. So as not to have
+to keep two copies, we include the source files here, changing the names of the
+external symbols to prevent clashes. */
+
+#define PCRE_INCLUDED
+
+#include "pcre_tables.c"
+#include "pcre_ucd.c"
+
+/* The definition of the macro PRINTABLE, which determines whether to print an
+output character as-is or as a hex value when showing compiled patterns, is
+the same as in the printint.src file. We uses it here in cases when the locale
+has not been explicitly changed, so as to get consistent output from systems
+that differ in their output from isprint() even in the "C" locale. */
+
+#ifdef EBCDIC
+#define PRINTABLE(c) ((c) >= 64 && (c) < 255)
+#else
+#define PRINTABLE(c) ((c) >= 32 && (c) < 127)
+#endif
+
+#define PRINTOK(c) (locale_set? isprint(c) : PRINTABLE(c))
+
+/* Posix support is disabled in 16 or 32 bit only mode. */
+#if !defined SUPPORT_PCRE8 && !defined NOPOSIX
+#define NOPOSIX
+#endif
+
+/* It is possible to compile this test program without including support for
+testing the POSIX interface, though this is not available via the standard
+Makefile. */
+
+#if !defined NOPOSIX
+#include "pcreposix.h"
+#endif
+
+/* It is also possible, originally for the benefit of a version that was
+imported into Exim, to build pcretest without support for UTF8 or UTF16 (define
+NOUTF), without the interface to the DFA matcher (NODFA). In fact, we
+automatically cut out the UTF support if PCRE is built without it. */
+
+#ifndef SUPPORT_UTF
+#ifndef NOUTF
+#define NOUTF
+#endif
+#endif
+
+/* To make the code a bit tidier for 8/16/32-bit support, we define macros
+for all the pcre[16]_xxx functions (except pcre16_fullinfo, which is called
+only from one place and is handled differently). I couldn't dream up any way of
+using a single macro to do this in a generic way, because of the many different
+argument requirements. We know that at least one of SUPPORT_PCRE8 and
+SUPPORT_PCRE16 must be set. First define macros for each individual mode; then
+use these in the definitions of generic macros.
+
+**** Special note about the PCHARSxxx macros: the address of the string to be
+printed is always given as two arguments: a base address followed by an offset.
+The base address is cast to the correct data size for 8 or 16 bit data; the
+offset is in units of this size. If the string were given as base+offset in one
+argument, the casting might be incorrectly applied. */
+
+#ifdef SUPPORT_PCRE8
+
+#define PCHARS8(lv, p, offset, len, f) \
+  lv = pchars((pcre_uint8 *)(p) + offset, len, f)
+
+#define PCHARSV8(p, offset, len, f) \
+  (void)pchars((pcre_uint8 *)(p) + offset, len, f)
+
+#define READ_CAPTURE_NAME8(p, cn8, cn16, cn32, re) \
+  p = read_capture_name8(p, cn8, re)
+
+#define STRLEN8(p) ((int)strlen((char *)p))
+
+#define SET_PCRE_CALLOUT8(callout) \
+  pcre_callout = callout
+
+#define PCRE_ASSIGN_JIT_STACK8(extra, callback, userdata) \
+   pcre_assign_jit_stack(extra, callback, userdata)
+
+#define PCRE_COMPILE8(re, pat, options, error, erroffset, tables) \
+  re = pcre_compile((char *)pat, options, error, erroffset, tables)
+
+#define PCRE_COPY_NAMED_SUBSTRING8(rc, re, bptr, offsets, count, \
+    namesptr, cbuffer, size) \
+  rc = pcre_copy_named_substring(re, (char *)bptr, offsets, count, \
+    (char *)namesptr, cbuffer, size)
+
+#define PCRE_COPY_SUBSTRING8(rc, bptr, offsets, count, i, cbuffer, size) \
+  rc = pcre_copy_substring((char *)bptr, offsets, count, i, cbuffer, size)
+
+#define PCRE_DFA_EXEC8(count, re, extra, bptr, len, start_offset, options, \
+    offsets, size_offsets, workspace, size_workspace) \
+  count = pcre_dfa_exec(re, extra, (char *)bptr, len, start_offset, options, \
+    offsets, size_offsets, workspace, size_workspace)
+
+#define PCRE_EXEC8(count, re, extra, bptr, len, start_offset, options, \
+    offsets, size_offsets) \
+  count = pcre_exec(re, extra, (char *)bptr, len, start_offset, options, \
+    offsets, size_offsets)
+
+#define PCRE_FREE_STUDY8(extra) \
+  pcre_free_study(extra)
+
+#define PCRE_FREE_SUBSTRING8(substring) \
+  pcre_free_substring(substring)
+
+#define PCRE_FREE_SUBSTRING_LIST8(listptr) \
+  pcre_free_substring_list(listptr)
+
+#define PCRE_GET_NAMED_SUBSTRING8(rc, re, bptr, offsets, count, \
+    getnamesptr, subsptr) \
+  rc = pcre_get_named_substring(re, (char *)bptr, offsets, count, \
+    (char *)getnamesptr, subsptr)
+
+#define PCRE_GET_STRINGNUMBER8(n, rc, ptr) \
+  n = pcre_get_stringnumber(re, (char *)ptr)
+
+#define PCRE_GET_SUBSTRING8(rc, bptr, offsets, count, i, subsptr) \
+  rc = pcre_get_substring((char *)bptr, offsets, count, i, subsptr)
+
+#define PCRE_GET_SUBSTRING_LIST8(rc, bptr, offsets, count, listptr) \
+  rc = pcre_get_substring_list((const char *)bptr, offsets, count, listptr)
+
+#define PCRE_PATTERN_TO_HOST_BYTE_ORDER8(rc, re, extra, tables) \
+  rc = pcre_pattern_to_host_byte_order(re, extra, tables)
+
+#define PCRE_PRINTINT8(re, outfile, debug_lengths) \
+  pcre_printint(re, outfile, debug_lengths)
+
+#define PCRE_STUDY8(extra, re, options, error) \
+  extra = pcre_study(re, options, error)
+
+#define PCRE_JIT_STACK_ALLOC8(startsize, maxsize) \
+  pcre_jit_stack_alloc(startsize, maxsize)
+
+#define PCRE_JIT_STACK_FREE8(stack) \
+  pcre_jit_stack_free(stack)
+
+#define pcre8_maketables pcre_maketables
+
+#endif /* SUPPORT_PCRE8 */
+
+/* -----------------------------------------------------------*/
+
+#ifdef SUPPORT_PCRE16
+
+#define PCHARS16(lv, p, offset, len, f) \
+  lv = pchars16((PCRE_SPTR16)(p) + offset, len, f)
+
+#define PCHARSV16(p, offset, len, f) \
+  (void)pchars16((PCRE_SPTR16)(p) + offset, len, f)
+
+#define READ_CAPTURE_NAME16(p, cn8, cn16, cn32, re) \
+  p = read_capture_name16(p, cn16, re)
+
+#define STRLEN16(p) ((int)strlen16((PCRE_SPTR16)p))
+
+#define SET_PCRE_CALLOUT16(callout) \
+  pcre16_callout = (int (*)(pcre16_callout_block *))callout
+
+#define PCRE_ASSIGN_JIT_STACK16(extra, callback, userdata) \
+  pcre16_assign_jit_stack((pcre16_extra *)extra, \
+    (pcre16_jit_callback)callback, userdata)
+
+#define PCRE_COMPILE16(re, pat, options, error, erroffset, tables) \
+  re = (pcre *)pcre16_compile((PCRE_SPTR16)pat, options, error, erroffset, \
+    tables)
+
+#define PCRE_COPY_NAMED_SUBSTRING16(rc, re, bptr, offsets, count, \
+    namesptr, cbuffer, size) \
+  rc = pcre16_copy_named_substring((pcre16 *)re, (PCRE_SPTR16)bptr, offsets, \
+    count, (PCRE_SPTR16)namesptr, (PCRE_UCHAR16 *)cbuffer, size/2)
+
+#define PCRE_COPY_SUBSTRING16(rc, bptr, offsets, count, i, cbuffer, size) \
+  rc = pcre16_copy_substring((PCRE_SPTR16)bptr, offsets, count, i, \
+    (PCRE_UCHAR16 *)cbuffer, size/2)
+
+#define PCRE_DFA_EXEC16(count, re, extra, bptr, len, start_offset, options, \
+    offsets, size_offsets, workspace, size_workspace) \
+  count = pcre16_dfa_exec((pcre16 *)re, (pcre16_extra *)extra, \
+    (PCRE_SPTR16)bptr, len, start_offset, options, offsets, size_offsets, \
+    workspace, size_workspace)
+
+#define PCRE_EXEC16(count, re, extra, bptr, len, start_offset, options, \
+    offsets, size_offsets) \
+  count = pcre16_exec((pcre16 *)re, (pcre16_extra *)extra, (PCRE_SPTR16)bptr, \
+    len, start_offset, options, offsets, size_offsets)
+
+#define PCRE_FREE_STUDY16(extra) \
+  pcre16_free_study((pcre16_extra *)extra)
+
+#define PCRE_FREE_SUBSTRING16(substring) \
+  pcre16_free_substring((PCRE_SPTR16)substring)
+
+#define PCRE_FREE_SUBSTRING_LIST16(listptr) \
+  pcre16_free_substring_list((PCRE_SPTR16 *)listptr)
+
+#define PCRE_GET_NAMED_SUBSTRING16(rc, re, bptr, offsets, count, \
+    getnamesptr, subsptr) \
+  rc = pcre16_get_named_substring((pcre16 *)re, (PCRE_SPTR16)bptr, offsets, \
+    count, (PCRE_SPTR16)getnamesptr, (PCRE_SPTR16 *)(void*)subsptr)
+
+#define PCRE_GET_STRINGNUMBER16(n, rc, ptr) \
+  n = pcre16_get_stringnumber(re, (PCRE_SPTR16)ptr)
+
+#define PCRE_GET_SUBSTRING16(rc, bptr, offsets, count, i, subsptr) \
+  rc = pcre16_get_substring((PCRE_SPTR16)bptr, offsets, count, i, \
+    (PCRE_SPTR16 *)(void*)subsptr)
+
+#define PCRE_GET_SUBSTRING_LIST16(rc, bptr, offsets, count, listptr) \
+  rc = pcre16_get_substring_list((PCRE_SPTR16)bptr, offsets, count, \
+    (PCRE_SPTR16 **)(void*)listptr)
+
+#define PCRE_PATTERN_TO_HOST_BYTE_ORDER16(rc, re, extra, tables) \
+  rc = pcre16_pattern_to_host_byte_order((pcre16 *)re, (pcre16_extra *)extra, \
+    tables)
+
+#define PCRE_PRINTINT16(re, outfile, debug_lengths) \
+  pcre16_printint(re, outfile, debug_lengths)
+
+#define PCRE_STUDY16(extra, re, options, error) \
+  extra = (pcre_extra *)pcre16_study((pcre16 *)re, options, error)
+
+#define PCRE_JIT_STACK_ALLOC16(startsize, maxsize) \
+  (pcre_jit_stack *)pcre16_jit_stack_alloc(startsize, maxsize)
+
+#define PCRE_JIT_STACK_FREE16(stack) \
+  pcre16_jit_stack_free((pcre16_jit_stack *)stack)
+
+#endif /* SUPPORT_PCRE16 */
+
+/* -----------------------------------------------------------*/
+
+#ifdef SUPPORT_PCRE32
+
+#define PCHARS32(lv, p, offset, len, f) \
+  lv = pchars32((PCRE_SPTR32)(p) + offset, len, use_utf, f)
+
+#define PCHARSV32(p, offset, len, f)                \
+  (void)pchars32((PCRE_SPTR32)(p) + offset, len, use_utf, f)
+
+#define READ_CAPTURE_NAME32(p, cn8, cn16, cn32, re) \
+  p = read_capture_name32(p, cn32, re)
+
+#define STRLEN32(p) ((int)strlen32((PCRE_SPTR32)p))
+
+#define SET_PCRE_CALLOUT32(callout) \
+  pcre32_callout = (int (*)(pcre32_callout_block *))callout
+
+#define PCRE_ASSIGN_JIT_STACK32(extra, callback, userdata) \
+  pcre32_assign_jit_stack((pcre32_extra *)extra, \
+    (pcre32_jit_callback)callback, userdata)
+
+#define PCRE_COMPILE32(re, pat, options, error, erroffset, tables) \
+  re = (pcre *)pcre32_compile((PCRE_SPTR32)pat, options, error, erroffset, \
+    tables)
+
+#define PCRE_COPY_NAMED_SUBSTRING32(rc, re, bptr, offsets, count, \
+    namesptr, cbuffer, size) \
+  rc = pcre32_copy_named_substring((pcre32 *)re, (PCRE_SPTR32)bptr, offsets, \
+    count, (PCRE_SPTR32)namesptr, (PCRE_UCHAR32 *)cbuffer, size/2)
+
+#define PCRE_COPY_SUBSTRING32(rc, bptr, offsets, count, i, cbuffer, size) \
+  rc = pcre32_copy_substring((PCRE_SPTR32)bptr, offsets, count, i, \
+    (PCRE_UCHAR32 *)cbuffer, size/2)
+
+#define PCRE_DFA_EXEC32(count, re, extra, bptr, len, start_offset, options, \
+    offsets, size_offsets, workspace, size_workspace) \
+  count = pcre32_dfa_exec((pcre32 *)re, (pcre32_extra *)extra, \
+    (PCRE_SPTR32)bptr, len, start_offset, options, offsets, size_offsets, \
+    workspace, size_workspace)
+
+#define PCRE_EXEC32(count, re, extra, bptr, len, start_offset, options, \
+    offsets, size_offsets) \
+  count = pcre32_exec((pcre32 *)re, (pcre32_extra *)extra, (PCRE_SPTR32)bptr, \
+    len, start_offset, options, offsets, size_offsets)
+
+#define PCRE_FREE_STUDY32(extra) \
+  pcre32_free_study((pcre32_extra *)extra)
+
+#define PCRE_FREE_SUBSTRING32(substring) \
+  pcre32_free_substring((PCRE_SPTR32)substring)
+
+#define PCRE_FREE_SUBSTRING_LIST32(listptr) \
+  pcre32_free_substring_list((PCRE_SPTR32 *)listptr)
+
+#define PCRE_GET_NAMED_SUBSTRING32(rc, re, bptr, offsets, count, \
+    getnamesptr, subsptr) \
+  rc = pcre32_get_named_substring((pcre32 *)re, (PCRE_SPTR32)bptr, offsets, \
+    count, (PCRE_SPTR32)getnamesptr, (PCRE_SPTR32 *)(void*)subsptr)
+
+#define PCRE_GET_STRINGNUMBER32(n, rc, ptr) \
+  n = pcre32_get_stringnumber(re, (PCRE_SPTR32)ptr)
+
+#define PCRE_GET_SUBSTRING32(rc, bptr, offsets, count, i, subsptr) \
+  rc = pcre32_get_substring((PCRE_SPTR32)bptr, offsets, count, i, \
+    (PCRE_SPTR32 *)(void*)subsptr)
+
+#define PCRE_GET_SUBSTRING_LIST32(rc, bptr, offsets, count, listptr) \
+  rc = pcre32_get_substring_list((PCRE_SPTR32)bptr, offsets, count, \
+    (PCRE_SPTR32 **)(void*)listptr)
+
+#define PCRE_PATTERN_TO_HOST_BYTE_ORDER32(rc, re, extra, tables) \
+  rc = pcre32_pattern_to_host_byte_order((pcre32 *)re, (pcre32_extra *)extra, \
+    tables)
+
+#define PCRE_PRINTINT32(re, outfile, debug_lengths) \
+  pcre32_printint(re, outfile, debug_lengths)
+
+#define PCRE_STUDY32(extra, re, options, error) \
+  extra = (pcre_extra *)pcre32_study((pcre32 *)re, options, error)
+
+#define PCRE_JIT_STACK_ALLOC32(startsize, maxsize) \
+  (pcre_jit_stack *)pcre32_jit_stack_alloc(startsize, maxsize)
+
+#define PCRE_JIT_STACK_FREE32(stack) \
+  pcre32_jit_stack_free((pcre32_jit_stack *)stack)
+
+#endif /* SUPPORT_PCRE32 */
+
+
+/* ----- More than one mode is supported; a runtime test is needed, except for
+pcre_config(), and the JIT stack functions, when it doesn't matter which
+available version is called. ----- */
+
+enum {
+  PCRE8_MODE,
+  PCRE16_MODE,
+  PCRE32_MODE
+};
+
+#if (defined (SUPPORT_PCRE8) + defined (SUPPORT_PCRE16) + \
+     defined (SUPPORT_PCRE32)) >= 2
+
+#define CHAR_SIZE (1 << pcre_mode)
+
+/* There doesn't seem to be an easy way of writing these macros that can cope
+with the 3 pairs of bit sizes plus all three bit sizes. So just handle all the
+cases separately. */
+
+/* ----- All three modes supported ----- */
+
+#if defined(SUPPORT_PCRE8) && defined(SUPPORT_PCRE16) && defined(SUPPORT_PCRE32)
+
+#define PCHARS(lv, p, offset, len, f) \
+  if (pcre_mode == PCRE32_MODE) \
+    PCHARS32(lv, p, offset, len, f); \
+  else if (pcre_mode == PCRE16_MODE) \
+    PCHARS16(lv, p, offset, len, f); \
+  else \
+    PCHARS8(lv, p, offset, len, f)
+
+#define PCHARSV(p, offset, len, f) \
+  if (pcre_mode == PCRE32_MODE) \
+    PCHARSV32(p, offset, len, f); \
+  else if (pcre_mode == PCRE16_MODE) \
+    PCHARSV16(p, offset, len, f); \
+  else \
+    PCHARSV8(p, offset, len, f)
+
+#define READ_CAPTURE_NAME(p, cn8, cn16, cn32, re) \
+  if (pcre_mode == PCRE32_MODE) \
+    READ_CAPTURE_NAME32(p, cn8, cn16, cn32, re); \
+  else if (pcre_mode == PCRE16_MODE) \
+    READ_CAPTURE_NAME16(p, cn8, cn16, cn32, re); \
+  else \
+    READ_CAPTURE_NAME8(p, cn8, cn16, cn32, re)
+
+#define SET_PCRE_CALLOUT(callout) \
+  if (pcre_mode == PCRE32_MODE) \
+    SET_PCRE_CALLOUT32(callout); \
+  else if (pcre_mode == PCRE16_MODE) \
+    SET_PCRE_CALLOUT16(callout); \
+  else \
+    SET_PCRE_CALLOUT8(callout)
+
+#define STRLEN(p) (pcre_mode == PCRE32_MODE ? STRLEN32(p) : pcre_mode == PCRE16_MODE ? STRLEN16(p) : STRLEN8(p))
+
+#define PCRE_ASSIGN_JIT_STACK(extra, callback, userdata) \
+  if (pcre_mode == PCRE32_MODE) \
+    PCRE_ASSIGN_JIT_STACK32(extra, callback, userdata); \
+  else if (pcre_mode == PCRE16_MODE) \
+    PCRE_ASSIGN_JIT_STACK16(extra, callback, userdata); \
+  else \
+    PCRE_ASSIGN_JIT_STACK8(extra, callback, userdata)
+
+#define PCRE_COMPILE(re, pat, options, error, erroffset, tables) \
+  if (pcre_mode == PCRE32_MODE) \
+    PCRE_COMPILE32(re, pat, options, error, erroffset, tables); \
+  else if (pcre_mode == PCRE16_MODE) \
+    PCRE_COMPILE16(re, pat, options, error, erroffset, tables); \
+  else \
+    PCRE_COMPILE8(re, pat, options, error, erroffset, tables)
+
+#define PCRE_CONFIG pcre_config
+
+#define PCRE_COPY_NAMED_SUBSTRING(rc, re, bptr, offsets, count, \
+    namesptr, cbuffer, size) \
+  if (pcre_mode == PCRE32_MODE) \
+    PCRE_COPY_NAMED_SUBSTRING32(rc, re, bptr, offsets, count, \
+      namesptr, cbuffer, size); \
+  else if (pcre_mode == PCRE16_MODE) \
+    PCRE_COPY_NAMED_SUBSTRING16(rc, re, bptr, offsets, count, \
+      namesptr, cbuffer, size); \
+  else \
+    PCRE_COPY_NAMED_SUBSTRING8(rc, re, bptr, offsets, count, \
+      namesptr, cbuffer, size)
+
+#define PCRE_COPY_SUBSTRING(rc, bptr, offsets, count, i, cbuffer, size) \
+  if (pcre_mode == PCRE32_MODE) \
+    PCRE_COPY_SUBSTRING32(rc, bptr, offsets, count, i, cbuffer, size); \
+  else if (pcre_mode == PCRE16_MODE) \
+    PCRE_COPY_SUBSTRING16(rc, bptr, offsets, count, i, cbuffer, size); \
+  else \
+    PCRE_COPY_SUBSTRING8(rc, bptr, offsets, count, i, cbuffer, size)
+
+#define PCRE_DFA_EXEC(count, re, extra, bptr, len, start_offset, options, \
+    offsets, size_offsets, workspace, size_workspace) \
+  if (pcre_mode == PCRE32_MODE) \
+    PCRE_DFA_EXEC32(count, re, extra, bptr, len, start_offset, options, \
+      offsets, size_offsets, workspace, size_workspace); \
+  else if (pcre_mode == PCRE16_MODE) \
+    PCRE_DFA_EXEC16(count, re, extra, bptr, len, start_offset, options, \
+      offsets, size_offsets, workspace, size_workspace); \
+  else \
+    PCRE_DFA_EXEC8(count, re, extra, bptr, len, start_offset, options, \
+      offsets, size_offsets, workspace, size_workspace)
+
+#define PCRE_EXEC(count, re, extra, bptr, len, start_offset, options, \
+    offsets, size_offsets) \
+  if (pcre_mode == PCRE32_MODE) \
+    PCRE_EXEC32(count, re, extra, bptr, len, start_offset, options, \
+      offsets, size_offsets); \
+  else if (pcre_mode == PCRE16_MODE) \
+    PCRE_EXEC16(count, re, extra, bptr, len, start_offset, options, \
+      offsets, size_offsets); \
+  else \
+    PCRE_EXEC8(count, re, extra, bptr, len, start_offset, options, \
+      offsets, size_offsets)
+
+#define PCRE_FREE_STUDY(extra) \
+  if (pcre_mode == PCRE32_MODE) \
+    PCRE_FREE_STUDY32(extra); \
+  else if (pcre_mode == PCRE16_MODE) \
+    PCRE_FREE_STUDY16(extra); \
+  else \
+    PCRE_FREE_STUDY8(extra)
+
+#define PCRE_FREE_SUBSTRING(substring) \
+  if (pcre_mode == PCRE32_MODE) \
+    PCRE_FREE_SUBSTRING32(substring); \
+  else if (pcre_mode == PCRE16_MODE) \
+    PCRE_FREE_SUBSTRING16(substring); \
+  else \
+    PCRE_FREE_SUBSTRING8(substring)
+
+#define PCRE_FREE_SUBSTRING_LIST(listptr) \
+  if (pcre_mode == PCRE32_MODE) \
+    PCRE_FREE_SUBSTRING_LIST32(listptr); \
+  else if (pcre_mode == PCRE16_MODE) \
+    PCRE_FREE_SUBSTRING_LIST16(listptr); \
+  else \
+    PCRE_FREE_SUBSTRING_LIST8(listptr)
+
+#define PCRE_GET_NAMED_SUBSTRING(rc, re, bptr, offsets, count, \
+    getnamesptr, subsptr) \
+  if (pcre_mode == PCRE32_MODE) \
+    PCRE_GET_NAMED_SUBSTRING32(rc, re, bptr, offsets, count, \
+      getnamesptr, subsptr); \
+  else if (pcre_mode == PCRE16_MODE) \
+    PCRE_GET_NAMED_SUBSTRING16(rc, re, bptr, offsets, count, \
+      getnamesptr, subsptr); \
+  else \
+    PCRE_GET_NAMED_SUBSTRING8(rc, re, bptr, offsets, count, \
+      getnamesptr, subsptr)
+
+#define PCRE_GET_STRINGNUMBER(n, rc, ptr) \
+  if (pcre_mode == PCRE32_MODE) \
+    PCRE_GET_STRINGNUMBER32(n, rc, ptr); \
+  else if (pcre_mode == PCRE16_MODE) \
+    PCRE_GET_STRINGNUMBER16(n, rc, ptr); \
+  else \
+    PCRE_GET_STRINGNUMBER8(n, rc, ptr)
+
+#define PCRE_GET_SUBSTRING(rc, bptr, use_offsets, count, i, subsptr) \
+  if (pcre_mode == PCRE32_MODE) \
+    PCRE_GET_SUBSTRING32(rc, bptr, use_offsets, count, i, subsptr); \
+  else if (pcre_mode == PCRE16_MODE) \
+    PCRE_GET_SUBSTRING16(rc, bptr, use_offsets, count, i, subsptr); \
+  else \
+    PCRE_GET_SUBSTRING8(rc, bptr, use_offsets, count, i, subsptr)
+
+#define PCRE_GET_SUBSTRING_LIST(rc, bptr, offsets, count, listptr) \
+  if (pcre_mode == PCRE32_MODE) \
+    PCRE_GET_SUBSTRING_LIST32(rc, bptr, offsets, count, listptr); \
+  else if (pcre_mode == PCRE16_MODE) \
+    PCRE_GET_SUBSTRING_LIST16(rc, bptr, offsets, count, listptr); \
+  else \
+    PCRE_GET_SUBSTRING_LIST8(rc, bptr, offsets, count, listptr)
+
+#define PCRE_JIT_STACK_ALLOC(startsize, maxsize) \
+  (pcre_mode == PCRE32_MODE ? \
+     PCRE_JIT_STACK_ALLOC32(startsize, maxsize) \
+    : pcre_mode == PCRE16_MODE ? \
+      PCRE_JIT_STACK_ALLOC16(startsize, maxsize) \
+      : PCRE_JIT_STACK_ALLOC8(startsize, maxsize))
+
+#define PCRE_JIT_STACK_FREE(stack) \
+  if (pcre_mode == PCRE32_MODE) \
+    PCRE_JIT_STACK_FREE32(stack); \
+  else if (pcre_mode == PCRE16_MODE) \
+    PCRE_JIT_STACK_FREE16(stack); \
+  else \
+    PCRE_JIT_STACK_FREE8(stack)
+
+#define PCRE_MAKETABLES \
+  (pcre_mode == PCRE32_MODE ? pcre32_maketables() : pcre_mode == PCRE16_MODE ? pcre16_maketables() : pcre_maketables())
+
+#define PCRE_PATTERN_TO_HOST_BYTE_ORDER(rc, re, extra, tables) \
+  if (pcre_mode == PCRE32_MODE) \
+    PCRE_PATTERN_TO_HOST_BYTE_ORDER32(rc, re, extra, tables); \
+  else if (pcre_mode == PCRE16_MODE) \
+    PCRE_PATTERN_TO_HOST_BYTE_ORDER16(rc, re, extra, tables); \
+  else \
+    PCRE_PATTERN_TO_HOST_BYTE_ORDER8(rc, re, extra, tables)
+
+#define PCRE_PRINTINT(re, outfile, debug_lengths) \
+  if (pcre_mode == PCRE32_MODE) \
+    PCRE_PRINTINT32(re, outfile, debug_lengths); \
+  else if (pcre_mode == PCRE16_MODE) \
+    PCRE_PRINTINT16(re, outfile, debug_lengths); \
+  else \
+    PCRE_PRINTINT8(re, outfile, debug_lengths)
+
+#define PCRE_STUDY(extra, re, options, error) \
+  if (pcre_mode == PCRE32_MODE) \
+    PCRE_STUDY32(extra, re, options, error); \
+  else if (pcre_mode == PCRE16_MODE) \
+    PCRE_STUDY16(extra, re, options, error); \
+  else \
+    PCRE_STUDY8(extra, re, options, error)
+
+
+/* ----- Two out of three modes are supported ----- */
+
+#else
+
+/* We can use some macro trickery to make a single set of definitions work in
+the three different cases. */
+
+/* ----- 32-bit and 16-bit but not 8-bit supported ----- */
+
+#if defined(SUPPORT_PCRE32) && defined(SUPPORT_PCRE16)
+#define BITONE 32
+#define BITTWO 16
+
+/* ----- 32-bit and 8-bit but not 16-bit supported ----- */
+
+#elif defined(SUPPORT_PCRE32) && defined(SUPPORT_PCRE8)
+#define BITONE 32
+#define BITTWO 8
+
+/* ----- 16-bit and 8-bit but not 32-bit supported ----- */
+
+#else
+#define BITONE 16
+#define BITTWO 8
+#endif
+
+#define glue(a,b) a##b
+#define G(a,b) glue(a,b)
+
+
+/* ----- Common macros for two-mode cases ----- */
+
+#define PCHARS(lv, p, offset, len, f) \
+  if (pcre_mode == G(G(PCRE,BITONE),_MODE)) \
+    G(PCHARS,BITONE)(lv, p, offset, len, f); \
+  else \
+    G(PCHARS,BITTWO)(lv, p, offset, len, f)
+
+#define PCHARSV(p, offset, len, f) \
+  if (pcre_mode == G(G(PCRE,BITONE),_MODE)) \
+    G(PCHARSV,BITONE)(p, offset, len, f); \
+  else \
+    G(PCHARSV,BITTWO)(p, offset, len, f)
+
+#define READ_CAPTURE_NAME(p, cn8, cn16, cn32, re) \
+  if (pcre_mode == G(G(PCRE,BITONE),_MODE)) \
+    G(READ_CAPTURE_NAME,BITONE)(p, cn8, cn16, cn32, re); \
+  else \
+    G(READ_CAPTURE_NAME,BITTWO)(p, cn8, cn16, cn32, re)
+
+#define SET_PCRE_CALLOUT(callout) \
+  if (pcre_mode == G(G(PCRE,BITONE),_MODE)) \
+    G(SET_PCRE_CALLOUT,BITONE)(callout); \
+  else \
+    G(SET_PCRE_CALLOUT,BITTWO)(callout)
+
+#define STRLEN(p) ((pcre_mode == G(G(PCRE,BITONE),_MODE)) ? \
+  G(STRLEN,BITONE)(p) : G(STRLEN,BITTWO)(p))
+
+#define PCRE_ASSIGN_JIT_STACK(extra, callback, userdata) \
+  if (pcre_mode == G(G(PCRE,BITONE),_MODE)) \
+    G(PCRE_ASSIGN_JIT_STACK,BITONE)(extra, callback, userdata); \
+  else \
+    G(PCRE_ASSIGN_JIT_STACK,BITTWO)(extra, callback, userdata)
+
+#define PCRE_COMPILE(re, pat, options, error, erroffset, tables) \
+  if (pcre_mode == G(G(PCRE,BITONE),_MODE)) \
+    G(PCRE_COMPILE,BITONE)(re, pat, options, error, erroffset, tables); \
+  else \
+    G(PCRE_COMPILE,BITTWO)(re, pat, options, error, erroffset, tables)
+
+#define PCRE_CONFIG G(G(pcre,BITONE),_config)
+
+#define PCRE_COPY_NAMED_SUBSTRING(rc, re, bptr, offsets, count, \
+    namesptr, cbuffer, size) \
+  if (pcre_mode == G(G(PCRE,BITONE),_MODE)) \
+    G(PCRE_COPY_NAMED_SUBSTRING,BITONE)(rc, re, bptr, offsets, count, \
+      namesptr, cbuffer, size); \
+  else \
+    G(PCRE_COPY_NAMED_SUBSTRING,BITTWO)(rc, re, bptr, offsets, count, \
+      namesptr, cbuffer, size)
+
+#define PCRE_COPY_SUBSTRING(rc, bptr, offsets, count, i, cbuffer, size) \
+  if (pcre_mode == G(G(PCRE,BITONE),_MODE)) \
+    G(PCRE_COPY_SUBSTRING,BITONE)(rc, bptr, offsets, count, i, cbuffer, size); \
+  else \
+    G(PCRE_COPY_SUBSTRING,BITTWO)(rc, bptr, offsets, count, i, cbuffer, size)
+
+#define PCRE_DFA_EXEC(count, re, extra, bptr, len, start_offset, options, \
+    offsets, size_offsets, workspace, size_workspace) \
+  if (pcre_mode == G(G(PCRE,BITONE),_MODE)) \
+    G(PCRE_DFA_EXEC,BITONE)(count, re, extra, bptr, len, start_offset, options, \
+      offsets, size_offsets, workspace, size_workspace); \
+  else \
+    G(PCRE_DFA_EXEC,BITTWO)(count, re, extra, bptr, len, start_offset, options, \
+      offsets, size_offsets, workspace, size_workspace)
+
+#define PCRE_EXEC(count, re, extra, bptr, len, start_offset, options, \
+    offsets, size_offsets) \
+  if (pcre_mode == G(G(PCRE,BITONE),_MODE)) \
+    G(PCRE_EXEC,BITONE)(count, re, extra, bptr, len, start_offset, options, \
+      offsets, size_offsets); \
+  else \
+    G(PCRE_EXEC,BITTWO)(count, re, extra, bptr, len, start_offset, options, \
+      offsets, size_offsets)
+
+#define PCRE_FREE_STUDY(extra) \
+  if (pcre_mode == G(G(PCRE,BITONE),_MODE)) \
+    G(PCRE_FREE_STUDY,BITONE)(extra); \
+  else \
+    G(PCRE_FREE_STUDY,BITTWO)(extra)
+
+#define PCRE_FREE_SUBSTRING(substring) \
+  if (pcre_mode == G(G(PCRE,BITONE),_MODE)) \
+    G(PCRE_FREE_SUBSTRING,BITONE)(substring); \
+  else \
+    G(PCRE_FREE_SUBSTRING,BITTWO)(substring)
+
+#define PCRE_FREE_SUBSTRING_LIST(listptr) \
+  if (pcre_mode == G(G(PCRE,BITONE),_MODE)) \
+    G(PCRE_FREE_SUBSTRING_LIST,BITONE)(listptr); \
+  else \
+    G(PCRE_FREE_SUBSTRING_LIST,BITTWO)(listptr)
+
+#define PCRE_GET_NAMED_SUBSTRING(rc, re, bptr, offsets, count, \
+    getnamesptr, subsptr) \
+  if (pcre_mode == G(G(PCRE,BITONE),_MODE)) \
+    G(PCRE_GET_NAMED_SUBSTRING,BITONE)(rc, re, bptr, offsets, count, \
+      getnamesptr, subsptr); \
+  else \
+    G(PCRE_GET_NAMED_SUBSTRING,BITTWO)(rc, re, bptr, offsets, count, \
+      getnamesptr, subsptr)
+
+#define PCRE_GET_STRINGNUMBER(n, rc, ptr) \
+  if (pcre_mode == G(G(PCRE,BITONE),_MODE)) \
+    G(PCRE_GET_STRINGNUMBER,BITONE)(n, rc, ptr); \
+  else \
+    G(PCRE_GET_STRINGNUMBER,BITTWO)(n, rc, ptr)
+
+#define PCRE_GET_SUBSTRING(rc, bptr, use_offsets, count, i, subsptr) \
+  if (pcre_mode == G(G(PCRE,BITONE),_MODE)) \
+    G(PCRE_GET_SUBSTRING,BITONE)(rc, bptr, use_offsets, count, i, subsptr); \
+  else \
+    G(PCRE_GET_SUBSTRING,BITTWO)(rc, bptr, use_offsets, count, i, subsptr)
+
+#define PCRE_GET_SUBSTRING_LIST(rc, bptr, offsets, count, listptr) \
+  if (pcre_mode == G(G(PCRE,BITONE),_MODE)) \
+    G(PCRE_GET_SUBSTRING_LIST,BITONE)(rc, bptr, offsets, count, listptr); \
+  else \
+    G(PCRE_GET_SUBSTRING_LIST,BITTWO)(rc, bptr, offsets, count, listptr)
+
+#define PCRE_JIT_STACK_ALLOC(startsize, maxsize) \
+  (pcre_mode == G(G(PCRE,BITONE),_MODE)) ? \
+     G(PCRE_JIT_STACK_ALLOC,BITONE)(startsize, maxsize) \
+    : G(PCRE_JIT_STACK_ALLOC,BITTWO)(startsize, maxsize)
+
+#define PCRE_JIT_STACK_FREE(stack) \
+  if (pcre_mode == G(G(PCRE,BITONE),_MODE)) \
+    G(PCRE_JIT_STACK_FREE,BITONE)(stack); \
+  else \
+    G(PCRE_JIT_STACK_FREE,BITTWO)(stack)
+
+#define PCRE_MAKETABLES \
+  (pcre_mode == G(G(PCRE,BITONE),_MODE)) ? \
+    G(G(pcre,BITONE),_maketables)() : G(G(pcre,BITTWO),_maketables)()
+
+#define PCRE_PATTERN_TO_HOST_BYTE_ORDER(rc, re, extra, tables) \
+  if (pcre_mode == G(G(PCRE,BITONE),_MODE)) \
+    G(PCRE_PATTERN_TO_HOST_BYTE_ORDER,BITONE)(rc, re, extra, tables); \
+  else \
+    G(PCRE_PATTERN_TO_HOST_BYTE_ORDER,BITTWO)(rc, re, extra, tables)
+
+#define PCRE_PRINTINT(re, outfile, debug_lengths) \
+  if (pcre_mode == G(G(PCRE,BITONE),_MODE)) \
+    G(PCRE_PRINTINT,BITONE)(re, outfile, debug_lengths); \
+  else \
+    G(PCRE_PRINTINT,BITTWO)(re, outfile, debug_lengths)
+
+#define PCRE_STUDY(extra, re, options, error) \
+  if (pcre_mode == G(G(PCRE,BITONE),_MODE)) \
+    G(PCRE_STUDY,BITONE)(extra, re, options, error); \
+  else \
+    G(PCRE_STUDY,BITTWO)(extra, re, options, error)
+
+#endif  /* Two out of three modes */
+
+/* ----- End of cases where more than one mode is supported ----- */
+
+
+/* ----- Only 8-bit mode is supported ----- */
+
+#elif defined SUPPORT_PCRE8
+#define CHAR_SIZE                 1
+#define PCHARS                    PCHARS8
+#define PCHARSV                   PCHARSV8
+#define READ_CAPTURE_NAME         READ_CAPTURE_NAME8
+#define SET_PCRE_CALLOUT          SET_PCRE_CALLOUT8
+#define STRLEN                    STRLEN8
+#define PCRE_ASSIGN_JIT_STACK     PCRE_ASSIGN_JIT_STACK8
+#define PCRE_COMPILE              PCRE_COMPILE8
+#define PCRE_CONFIG               pcre_config
+#define PCRE_COPY_NAMED_SUBSTRING PCRE_COPY_NAMED_SUBSTRING8
+#define PCRE_COPY_SUBSTRING       PCRE_COPY_SUBSTRING8
+#define PCRE_DFA_EXEC             PCRE_DFA_EXEC8
+#define PCRE_EXEC                 PCRE_EXEC8
+#define PCRE_FREE_STUDY           PCRE_FREE_STUDY8
+#define PCRE_FREE_SUBSTRING       PCRE_FREE_SUBSTRING8
+#define PCRE_FREE_SUBSTRING_LIST  PCRE_FREE_SUBSTRING_LIST8
+#define PCRE_GET_NAMED_SUBSTRING  PCRE_GET_NAMED_SUBSTRING8
+#define PCRE_GET_STRINGNUMBER     PCRE_GET_STRINGNUMBER8
+#define PCRE_GET_SUBSTRING        PCRE_GET_SUBSTRING8
+#define PCRE_GET_SUBSTRING_LIST   PCRE_GET_SUBSTRING_LIST8
+#define PCRE_JIT_STACK_ALLOC      PCRE_JIT_STACK_ALLOC8
+#define PCRE_JIT_STACK_FREE       PCRE_JIT_STACK_FREE8
+#define PCRE_MAKETABLES           pcre_maketables()
+#define PCRE_PATTERN_TO_HOST_BYTE_ORDER PCRE_PATTERN_TO_HOST_BYTE_ORDER8
+#define PCRE_PRINTINT             PCRE_PRINTINT8
+#define PCRE_STUDY                PCRE_STUDY8
+
+/* ----- Only 16-bit mode is supported ----- */
+
+#elif defined SUPPORT_PCRE16
+#define CHAR_SIZE                 2
+#define PCHARS                    PCHARS16
+#define PCHARSV                   PCHARSV16
+#define READ_CAPTURE_NAME         READ_CAPTURE_NAME16
+#define SET_PCRE_CALLOUT          SET_PCRE_CALLOUT16
+#define STRLEN                    STRLEN16
+#define PCRE_ASSIGN_JIT_STACK     PCRE_ASSIGN_JIT_STACK16
+#define PCRE_COMPILE              PCRE_COMPILE16
+#define PCRE_CONFIG               pcre16_config
+#define PCRE_COPY_NAMED_SUBSTRING PCRE_COPY_NAMED_SUBSTRING16
+#define PCRE_COPY_SUBSTRING       PCRE_COPY_SUBSTRING16
+#define PCRE_DFA_EXEC             PCRE_DFA_EXEC16
+#define PCRE_EXEC                 PCRE_EXEC16
+#define PCRE_FREE_STUDY           PCRE_FREE_STUDY16
+#define PCRE_FREE_SUBSTRING       PCRE_FREE_SUBSTRING16
+#define PCRE_FREE_SUBSTRING_LIST  PCRE_FREE_SUBSTRING_LIST16
+#define PCRE_GET_NAMED_SUBSTRING  PCRE_GET_NAMED_SUBSTRING16
+#define PCRE_GET_STRINGNUMBER     PCRE_GET_STRINGNUMBER16
+#define PCRE_GET_SUBSTRING        PCRE_GET_SUBSTRING16
+#define PCRE_GET_SUBSTRING_LIST   PCRE_GET_SUBSTRING_LIST16
+#define PCRE_JIT_STACK_ALLOC      PCRE_JIT_STACK_ALLOC16
+#define PCRE_JIT_STACK_FREE       PCRE_JIT_STACK_FREE16
+#define PCRE_MAKETABLES           pcre16_maketables()
+#define PCRE_PATTERN_TO_HOST_BYTE_ORDER PCRE_PATTERN_TO_HOST_BYTE_ORDER16
+#define PCRE_PRINTINT             PCRE_PRINTINT16
+#define PCRE_STUDY                PCRE_STUDY16
+
+/* ----- Only 32-bit mode is supported ----- */
+
+#elif defined SUPPORT_PCRE32
+#define CHAR_SIZE                 4
+#define PCHARS                    PCHARS32
+#define PCHARSV                   PCHARSV32
+#define READ_CAPTURE_NAME         READ_CAPTURE_NAME32
+#define SET_PCRE_CALLOUT          SET_PCRE_CALLOUT32
+#define STRLEN                    STRLEN32
+#define PCRE_ASSIGN_JIT_STACK     PCRE_ASSIGN_JIT_STACK32
+#define PCRE_COMPILE              PCRE_COMPILE32
+#define PCRE_CONFIG               pcre32_config
+#define PCRE_COPY_NAMED_SUBSTRING PCRE_COPY_NAMED_SUBSTRING32
+#define PCRE_COPY_SUBSTRING       PCRE_COPY_SUBSTRING32
+#define PCRE_DFA_EXEC             PCRE_DFA_EXEC32
+#define PCRE_EXEC                 PCRE_EXEC32
+#define PCRE_FREE_STUDY           PCRE_FREE_STUDY32
+#define PCRE_FREE_SUBSTRING       PCRE_FREE_SUBSTRING32
+#define PCRE_FREE_SUBSTRING_LIST  PCRE_FREE_SUBSTRING_LIST32
+#define PCRE_GET_NAMED_SUBSTRING  PCRE_GET_NAMED_SUBSTRING32
+#define PCRE_GET_STRINGNUMBER     PCRE_GET_STRINGNUMBER32
+#define PCRE_GET_SUBSTRING        PCRE_GET_SUBSTRING32
+#define PCRE_GET_SUBSTRING_LIST   PCRE_GET_SUBSTRING_LIST32
+#define PCRE_JIT_STACK_ALLOC      PCRE_JIT_STACK_ALLOC32
+#define PCRE_JIT_STACK_FREE       PCRE_JIT_STACK_FREE32
+#define PCRE_MAKETABLES           pcre32_maketables()
+#define PCRE_PATTERN_TO_HOST_BYTE_ORDER PCRE_PATTERN_TO_HOST_BYTE_ORDER32
+#define PCRE_PRINTINT             PCRE_PRINTINT32
+#define PCRE_STUDY                PCRE_STUDY32
+
+#endif
+
+/* ----- End of mode-specific function call macros ----- */
+
+
+/* Other parameters */
+
+#ifndef CLOCKS_PER_SEC
+#ifdef CLK_TCK
+#define CLOCKS_PER_SEC CLK_TCK
+#else
+#define CLOCKS_PER_SEC 100
+#endif
+#endif
+
+#if !defined NODFA
+#define DFA_WS_DIMENSION 1000
+#endif
+
+/* This is the default loop count for timing. */
+
+#define LOOPREPEAT 500000
+
+/* Static variables */
+
+static FILE *outfile;
+static int log_store = 0;
+static int callout_count;
+static int callout_extra;
+static int callout_fail_count;
+static int callout_fail_id;
+static int debug_lengths;
+static int first_callout;
+static int jit_was_used;
+static int locale_set = 0;
+static int show_malloc;
+static int use_utf;
+static size_t gotten_store;
+static size_t first_gotten_store = 0;
+static const unsigned char *last_callout_mark = NULL;
+
+/* The buffers grow automatically if very long input lines are encountered. */
+
+static int buffer_size = 50000;
+static pcre_uint8 *buffer = NULL;
+static pcre_uint8 *pbuffer = NULL;
+
+/* Just as a safety check, make sure that COMPILE_PCRE[16|32] are *not* set. */
+
+#ifdef COMPILE_PCRE16
+#error COMPILE_PCRE16 must not be set when compiling pcretest.c
+#endif
+
+#ifdef COMPILE_PCRE32
+#error COMPILE_PCRE32 must not be set when compiling pcretest.c
+#endif
+
+/* We need buffers for building 16/32-bit strings, and the tables of operator
+lengths that are used for 16/32-bit compiling, in order to swap bytes in a
+pattern for saving/reloading testing. Luckily, the data for these tables is
+defined as a macro. However, we must ensure that LINK_SIZE and IMM2_SIZE (which
+are used in the tables) are adjusted appropriately for the 16/32-bit world.
+LINK_SIZE is also used later in this program. */
+
+#ifdef SUPPORT_PCRE16
+#undef IMM2_SIZE
+#define IMM2_SIZE 1
+
+#if LINK_SIZE == 2
+#undef LINK_SIZE
+#define LINK_SIZE 1
+#elif LINK_SIZE == 3 || LINK_SIZE == 4
+#undef LINK_SIZE
+#define LINK_SIZE 2
+#else
+#error LINK_SIZE must be either 2, 3, or 4
+#endif
+
+static int buffer16_size = 0;
+static pcre_uint16 *buffer16 = NULL;
+static const pcre_uint16 OP_lengths16[] = { OP_LENGTHS };
+#endif  /* SUPPORT_PCRE16 */
+
+#ifdef SUPPORT_PCRE32
+#undef IMM2_SIZE
+#define IMM2_SIZE 1
+#undef LINK_SIZE
+#define LINK_SIZE 1
+
+static int buffer32_size = 0;
+static pcre_uint32 *buffer32 = NULL;
+static const pcre_uint32 OP_lengths32[] = { OP_LENGTHS };
+#endif  /* SUPPORT_PCRE32 */
+
+/* If we have 8-bit support, default to it; if there is also 16-or 32-bit
+support, it can be changed by an option. If there is no 8-bit support, there
+must be 16-or 32-bit support, so default it to 1. */
+
+#if defined SUPPORT_PCRE8
+static int pcre_mode = PCRE8_MODE;
+#elif defined SUPPORT_PCRE16
+static int pcre_mode = PCRE16_MODE;
+#elif defined SUPPORT_PCRE32
+static int pcre_mode = PCRE32_MODE;
+#endif
+
+/* JIT study options for -s+n and /S+n where '1' <= n <= '7'. */
+
+static int jit_study_bits[] =
+  {
+  PCRE_STUDY_JIT_COMPILE,
+  PCRE_STUDY_JIT_PARTIAL_SOFT_COMPILE,
+  PCRE_STUDY_JIT_COMPILE + PCRE_STUDY_JIT_PARTIAL_SOFT_COMPILE,
+  PCRE_STUDY_JIT_PARTIAL_HARD_COMPILE,
+  PCRE_STUDY_JIT_COMPILE + PCRE_STUDY_JIT_PARTIAL_HARD_COMPILE,
+  PCRE_STUDY_JIT_PARTIAL_SOFT_COMPILE + PCRE_STUDY_JIT_PARTIAL_HARD_COMPILE,
+  PCRE_STUDY_JIT_COMPILE + PCRE_STUDY_JIT_PARTIAL_SOFT_COMPILE +
+    PCRE_STUDY_JIT_PARTIAL_HARD_COMPILE
+};
+
+#define PCRE_STUDY_ALLJIT (PCRE_STUDY_JIT_COMPILE | \
+  PCRE_STUDY_JIT_PARTIAL_SOFT_COMPILE | PCRE_STUDY_JIT_PARTIAL_HARD_COMPILE)
+
+/* Textual explanations for runtime error codes */
+
+static const char *errtexts[] = {
+  NULL,  /* 0 is no error */
+  NULL,  /* NOMATCH is handled specially */
+  "NULL argument passed",
+  "bad option value",
+  "magic number missing",
+  "unknown opcode - pattern overwritten?",
+  "no more memory",
+  NULL,  /* never returned by pcre_exec() or pcre_dfa_exec() */
+  "match limit exceeded",
+  "callout error code",
+  NULL,  /* BADUTF8/16 is handled specially */
+  NULL,  /* BADUTF8/16 offset is handled specially */
+  NULL,  /* PARTIAL is handled specially */
+  "not used - internal error",
+  "internal error - pattern overwritten?",
+  "bad count value",
+  "item unsupported for DFA matching",
+  "backreference condition or recursion test not supported for DFA matching",
+  "match limit not supported for DFA matching",
+  "workspace size exceeded in DFA matching",
+  "too much recursion for DFA matching",
+  "recursion limit exceeded",
+  "not used - internal error",
+  "invalid combination of newline options",
+  "bad offset value",
+  NULL,  /* SHORTUTF8/16 is handled specially */
+  "nested recursion at the same subject position",
+  "JIT stack limit reached",
+  "pattern compiled in wrong mode: 8-bit/16-bit error",
+  "pattern compiled with other endianness",
+  "invalid data in workspace for DFA restart",
+  "bad JIT option",
+  "bad length"
+};
+
+
+/*************************************************
+*         Alternate character tables             *
+*************************************************/
+
+/* By default, the "tables" pointer when calling PCRE is set to NULL, thereby
+using the default tables of the library. However, the T option can be used to
+select alternate sets of tables, for different kinds of testing. Note also that
+the L (locale) option also adjusts the tables. */
+
+/* This is the set of tables distributed as default with PCRE. It recognizes
+only ASCII characters. */
+
+static const pcre_uint8 tables0[] = {
+
+/* This table is a lower casing table. */
+
+    0,  1,  2,  3,  4,  5,  6,  7,
+    8,  9, 10, 11, 12, 13, 14, 15,
+   16, 17, 18, 19, 20, 21, 22, 23,
+   24, 25, 26, 27, 28, 29, 30, 31,
+   32, 33, 34, 35, 36, 37, 38, 39,
+   40, 41, 42, 43, 44, 45, 46, 47,
+   48, 49, 50, 51, 52, 53, 54, 55,
+   56, 57, 58, 59, 60, 61, 62, 63,
+   64, 97, 98, 99,100,101,102,103,
+  104,105,106,107,108,109,110,111,
+  112,113,114,115,116,117,118,119,
+  120,121,122, 91, 92, 93, 94, 95,
+   96, 97, 98, 99,100,101,102,103,
+  104,105,106,107,108,109,110,111,
+  112,113,114,115,116,117,118,119,
+  120,121,122,123,124,125,126,127,
+  128,129,130,131,132,133,134,135,
+  136,137,138,139,140,141,142,143,
+  144,145,146,147,148,149,150,151,
+  152,153,154,155,156,157,158,159,
+  160,161,162,163,164,165,166,167,
+  168,169,170,171,172,173,174,175,
+  176,177,178,179,180,181,182,183,
+  184,185,186,187,188,189,190,191,
+  192,193,194,195,196,197,198,199,
+  200,201,202,203,204,205,206,207,
+  208,209,210,211,212,213,214,215,
+  216,217,218,219,220,221,222,223,
+  224,225,226,227,228,229,230,231,
+  232,233,234,235,236,237,238,239,
+  240,241,242,243,244,245,246,247,
+  248,249,250,251,252,253,254,255,
+
+/* This table is a case flipping table. */
+
+    0,  1,  2,  3,  4,  5,  6,  7,
+    8,  9, 10, 11, 12, 13, 14, 15,
+   16, 17, 18, 19, 20, 21, 22, 23,
+   24, 25, 26, 27, 28, 29, 30, 31,
+   32, 33, 34, 35, 36, 37, 38, 39,
+   40, 41, 42, 43, 44, 45, 46, 47,
+   48, 49, 50, 51, 52, 53, 54, 55,
+   56, 57, 58, 59, 60, 61, 62, 63,
+   64, 97, 98, 99,100,101,102,103,
+  104,105,106,107,108,109,110,111,
+  112,113,114,115,116,117,118,119,
+  120,121,122, 91, 92, 93, 94, 95,
+   96, 65, 66, 67, 68, 69, 70, 71,
+   72, 73, 74, 75, 76, 77, 78, 79,
+   80, 81, 82, 83, 84, 85, 86, 87,
+   88, 89, 90,123,124,125,126,127,
+  128,129,130,131,132,133,134,135,
+  136,137,138,139,140,141,142,143,
+  144,145,146,147,148,149,150,151,
+  152,153,154,155,156,157,158,159,
+  160,161,162,163,164,165,166,167,
+  168,169,170,171,172,173,174,175,
+  176,177,178,179,180,181,182,183,
+  184,185,186,187,188,189,190,191,
+  192,193,194,195,196,197,198,199,
+  200,201,202,203,204,205,206,207,
+  208,209,210,211,212,213,214,215,
+  216,217,218,219,220,221,222,223,
+  224,225,226,227,228,229,230,231,
+  232,233,234,235,236,237,238,239,
+  240,241,242,243,244,245,246,247,
+  248,249,250,251,252,253,254,255,
+
+/* This table contains bit maps for various character classes. Each map is 32
+bytes long and the bits run from the least significant end of each byte. The
+classes that have their own maps are: space, xdigit, digit, upper, lower, word,
+graph, print, punct, and cntrl. Other classes are built from combinations. */
+
+  0x00,0x3e,0x00,0x00,0x01,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+
+  0x00,0x00,0x00,0x00,0x00,0x00,0xff,0x03,
+  0x7e,0x00,0x00,0x00,0x7e,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+
+  0x00,0x00,0x00,0x00,0x00,0x00,0xff,0x03,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+  0xfe,0xff,0xff,0x07,0x00,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0xfe,0xff,0xff,0x07,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+
+  0x00,0x00,0x00,0x00,0x00,0x00,0xff,0x03,
+  0xfe,0xff,0xff,0x87,0xfe,0xff,0xff,0x07,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+
+  0x00,0x00,0x00,0x00,0xfe,0xff,0xff,0xff,
+  0xff,0xff,0xff,0xff,0xff,0xff,0xff,0x7f,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+
+  0x00,0x00,0x00,0x00,0xff,0xff,0xff,0xff,
+  0xff,0xff,0xff,0xff,0xff,0xff,0xff,0x7f,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+
+  0x00,0x00,0x00,0x00,0xfe,0xff,0x00,0xfc,
+  0x01,0x00,0x00,0xf8,0x01,0x00,0x00,0x78,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+
+  0xff,0xff,0xff,0xff,0x00,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x80,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+
+/* This table identifies various classes of character by individual bits:
+  0x01   white space character
+  0x02   letter
+  0x04   decimal digit
+  0x08   hexadecimal digit
+  0x10   alphanumeric or '_'
+  0x80   regular expression metacharacter or binary zero
+*/
+
+  0x80,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*   0-  7 */
+  0x00,0x01,0x01,0x00,0x01,0x01,0x00,0x00, /*   8- 15 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*  16- 23 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*  24- 31 */
+  0x01,0x00,0x00,0x00,0x80,0x00,0x00,0x00, /*    - '  */
+  0x80,0x80,0x80,0x80,0x00,0x00,0x80,0x00, /*  ( - /  */
+  0x1c,0x1c,0x1c,0x1c,0x1c,0x1c,0x1c,0x1c, /*  0 - 7  */
+  0x1c,0x1c,0x00,0x00,0x00,0x00,0x00,0x80, /*  8 - ?  */
+  0x00,0x1a,0x1a,0x1a,0x1a,0x1a,0x1a,0x12, /*  @ - G  */
+  0x12,0x12,0x12,0x12,0x12,0x12,0x12,0x12, /*  H - O  */
+  0x12,0x12,0x12,0x12,0x12,0x12,0x12,0x12, /*  P - W  */
+  0x12,0x12,0x12,0x80,0x80,0x00,0x80,0x10, /*  X - _  */
+  0x00,0x1a,0x1a,0x1a,0x1a,0x1a,0x1a,0x12, /*  ` - g  */
+  0x12,0x12,0x12,0x12,0x12,0x12,0x12,0x12, /*  h - o  */
+  0x12,0x12,0x12,0x12,0x12,0x12,0x12,0x12, /*  p - w  */
+  0x12,0x12,0x12,0x80,0x80,0x00,0x00,0x00, /*  x -127 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 128-135 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 136-143 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 144-151 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 152-159 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 160-167 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 168-175 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 176-183 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 184-191 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 192-199 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 200-207 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 208-215 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 216-223 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 224-231 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 232-239 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 240-247 */
+  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00};/* 248-255 */
+
+/* This is a set of tables that came originally from a Windows user. It seems to
+be at least an approximation of ISO 8859. In particular, there are characters
+greater than 128 that are marked as spaces, letters, etc. */
+
+static const pcre_uint8 tables1[] = {
+0,1,2,3,4,5,6,7,
+8,9,10,11,12,13,14,15,
+16,17,18,19,20,21,22,23,
+24,25,26,27,28,29,30,31,
+32,33,34,35,36,37,38,39,
+40,41,42,43,44,45,46,47,
+48,49,50,51,52,53,54,55,
+56,57,58,59,60,61,62,63,
+64,97,98,99,100,101,102,103,
+104,105,106,107,108,109,110,111,
+112,113,114,115,116,117,118,119,
+120,121,122,91,92,93,94,95,
+96,97,98,99,100,101,102,103,
+104,105,106,107,108,109,110,111,
+112,113,114,115,116,117,118,119,
+120,121,122,123,124,125,126,127,
+128,129,130,131,132,133,134,135,
+136,137,138,139,140,141,142,143,
+144,145,146,147,148,149,150,151,
+152,153,154,155,156,157,158,159,
+160,161,162,163,164,165,166,167,
+168,169,170,171,172,173,174,175,
+176,177,178,179,180,181,182,183,
+184,185,186,187,188,189,190,191,
+224,225,226,227,228,229,230,231,
+232,233,234,235,236,237,238,239,
+240,241,242,243,244,245,246,215,
+248,249,250,251,252,253,254,223,
+224,225,226,227,228,229,230,231,
+232,233,234,235,236,237,238,239,
+240,241,242,243,244,245,246,247,
+248,249,250,251,252,253,254,255,
+0,1,2,3,4,5,6,7,
+8,9,10,11,12,13,14,15,
+16,17,18,19,20,21,22,23,
+24,25,26,27,28,29,30,31,
+32,33,34,35,36,37,38,39,
+40,41,42,43,44,45,46,47,
+48,49,50,51,52,53,54,55,
+56,57,58,59,60,61,62,63,
+64,97,98,99,100,101,102,103,
+104,105,106,107,108,109,110,111,
+112,113,114,115,116,117,118,119,
+120,121,122,91,92,93,94,95,
+96,65,66,67,68,69,70,71,
+72,73,74,75,76,77,78,79,
+80,81,82,83,84,85,86,87,
+88,89,90,123,124,125,126,127,
+128,129,130,131,132,133,134,135,
+136,137,138,139,140,141,142,143,
+144,145,146,147,148,149,150,151,
+152,153,154,155,156,157,158,159,
+160,161,162,163,164,165,166,167,
+168,169,170,171,172,173,174,175,
+176,177,178,179,180,181,182,183,
+184,185,186,187,188,189,190,191,
+224,225,226,227,228,229,230,231,
+232,233,234,235,236,237,238,239,
+240,241,242,243,244,245,246,215,
+248,249,250,251,252,253,254,223,
+192,193,194,195,196,197,198,199,
+200,201,202,203,204,205,206,207,
+208,209,210,211,212,213,214,247,
+216,217,218,219,220,221,222,255,
+0,62,0,0,1,0,0,0,
+0,0,0,0,0,0,0,0,
+32,0,0,0,1,0,0,0,
+0,0,0,0,0,0,0,0,
+0,0,0,0,0,0,255,3,
+126,0,0,0,126,0,0,0,
+0,0,0,0,0,0,0,0,
+0,0,0,0,0,0,0,0,
+0,0,0,0,0,0,255,3,
+0,0,0,0,0,0,0,0,
+0,0,0,0,0,0,12,2,
+0,0,0,0,0,0,0,0,
+0,0,0,0,0,0,0,0,
+254,255,255,7,0,0,0,0,
+0,0,0,0,0,0,0,0,
+255,255,127,127,0,0,0,0,
+0,0,0,0,0,0,0,0,
+0,0,0,0,254,255,255,7,
+0,0,0,0,0,4,32,4,
+0,0,0,128,255,255,127,255,
+0,0,0,0,0,0,255,3,
+254,255,255,135,254,255,255,7,
+0,0,0,0,0,4,44,6,
+255,255,127,255,255,255,127,255,
+0,0,0,0,254,255,255,255,
+255,255,255,255,255,255,255,127,
+0,0,0,0,254,255,255,255,
+255,255,255,255,255,255,255,255,
+0,2,0,0,255,255,255,255,
+255,255,255,255,255,255,255,127,
+0,0,0,0,255,255,255,255,
+255,255,255,255,255,255,255,255,
+0,0,0,0,254,255,0,252,
+1,0,0,248,1,0,0,120,
+0,0,0,0,254,255,255,255,
+0,0,128,0,0,0,128,0,
+255,255,255,255,0,0,0,0,
+0,0,0,0,0,0,0,128,
+255,255,255,255,0,0,0,0,
+0,0,0,0,0,0,0,0,
+128,0,0,0,0,0,0,0,
+0,1,1,0,1,1,0,0,
+0,0,0,0,0,0,0,0,
+0,0,0,0,0,0,0,0,
+1,0,0,0,128,0,0,0,
+128,128,128,128,0,0,128,0,
+28,28,28,28,28,28,28,28,
+28,28,0,0,0,0,0,128,
+0,26,26,26,26,26,26,18,
+18,18,18,18,18,18,18,18,
+18,18,18,18,18,18,18,18,
+18,18,18,128,128,0,128,16,
+0,26,26,26,26,26,26,18,
+18,18,18,18,18,18,18,18,
+18,18,18,18,18,18,18,18,
+18,18,18,128,128,0,0,0,
+0,0,0,0,0,1,0,0,
+0,0,0,0,0,0,0,0,
+0,0,0,0,0,0,0,0,
+0,0,0,0,0,0,0,0,
+1,0,0,0,0,0,0,0,
+0,0,18,0,0,0,0,0,
+0,0,20,20,0,18,0,0,
+0,20,18,0,0,0,0,0,
+18,18,18,18,18,18,18,18,
+18,18,18,18,18,18,18,18,
+18,18,18,18,18,18,18,0,
+18,18,18,18,18,18,18,18,
+18,18,18,18,18,18,18,18,
+18,18,18,18,18,18,18,18,
+18,18,18,18,18,18,18,0,
+18,18,18,18,18,18,18,18
+};
+
+
+
+
+#ifndef HAVE_STRERROR
+/*************************************************
+*     Provide strerror() for non-ANSI libraries  *
+*************************************************/
+
+/* Some old-fashioned systems still around (e.g. SunOS4) don't have strerror()
+in their libraries, but can provide the same facility by this simple
+alternative function. */
+
+extern int   sys_nerr;
+extern char *sys_errlist[];
+
+char *
+strerror(int n)
+{
+if (n < 0 || n >= sys_nerr) return "unknown error number";
+return sys_errlist[n];
+}
+#endif /* HAVE_STRERROR */
+
+
+
+/*************************************************
+*       Print newline configuration              *
+*************************************************/
+
+/*
+Arguments:
+  rc         the return code from PCRE_CONFIG_NEWLINE
+  isc        TRUE if called from "-C newline"
+Returns:     nothing
+*/
+
+static void
+print_newline_config(int rc, BOOL isc)
+{
+const char *s = NULL;
+if (!isc) printf("  Newline sequence is ");
+switch(rc)
+  {
+  case CHAR_CR: s = "CR"; break;
+  case CHAR_LF: s = "LF"; break;
+  case (CHAR_CR<<8 | CHAR_LF): s = "CRLF"; break;
+  case -1: s = "ANY"; break;
+  case -2: s = "ANYCRLF"; break;
+
+  default:
+  printf("a non-standard value: 0x%04x\n", rc);
+  return;
+  }
+
+printf("%s\n", s);
+}
+
+
+
+/*************************************************
+*         JIT memory callback                    *
+*************************************************/
+
+static pcre_jit_stack* jit_callback(void *arg)
+{
+jit_was_used = TRUE;
+return (pcre_jit_stack *)arg;
+}
+
+
+#if !defined NOUTF || defined SUPPORT_PCRE16 || defined SUPPORT_PCRE32
+/*************************************************
+*            Convert UTF-8 string to value       *
+*************************************************/
+
+/* This function takes one or more bytes that represents a UTF-8 character,
+and returns the value of the character.
+
+Argument:
+  utf8bytes   a pointer to the byte vector
+  vptr        a pointer to an int to receive the value
+
+Returns:      >  0 => the number of bytes consumed
+              -6 to 0 => malformed UTF-8 character at offset = (-return)
+*/
+
+static int
+utf82ord(pcre_uint8 *utf8bytes, pcre_uint32 *vptr)
+{
+pcre_uint32 c = *utf8bytes++;
+pcre_uint32 d = c;
+int i, j, s;
+
+for (i = -1; i < 6; i++)               /* i is number of additional bytes */
+  {
+  if ((d & 0x80) == 0) break;
+  d <<= 1;
+  }
+
+if (i == -1) { *vptr = c; return 1; }  /* ascii character */
+if (i == 0 || i == 6) return 0;        /* invalid UTF-8 */
+
+/* i now has a value in the range 1-5 */
+
+s = 6*i;
+d = (c & utf8_table3[i]) << s;
+
+for (j = 0; j < i; j++)
+  {
+  c = *utf8bytes++;
+  if ((c & 0xc0) != 0x80) return -(j+1);
+  s -= 6;
+  d |= (c & 0x3f) << s;
+  }
+
+/* Check that encoding was the correct unique one */
+
+for (j = 0; j < utf8_table1_size; j++)
+  if (d <= (pcre_uint32)utf8_table1[j]) break;
+if (j != i) return -(i+1);
+
+/* Valid value */
+
+*vptr = d;
+return i+1;
+}
+#endif /* NOUTF || SUPPORT_PCRE16 */
+
+
+
+#if defined SUPPORT_PCRE8 && !defined NOUTF
+/*************************************************
+*       Convert character value to UTF-8         *
+*************************************************/
+
+/* This function takes an integer value in the range 0 - 0x7fffffff
+and encodes it as a UTF-8 character in 0 to 6 bytes.
+
+Arguments:
+  cvalue     the character value
+  utf8bytes  pointer to buffer for result - at least 6 bytes long
+
+Returns:     number of characters placed in the buffer
+*/
+
+static int
+ord2utf8(pcre_uint32 cvalue, pcre_uint8 *utf8bytes)
+{
+register int i, j;
+if (cvalue > 0x7fffffffu)
+  return -1;
+for (i = 0; i < utf8_table1_size; i++)
+  if (cvalue <= (pcre_uint32)utf8_table1[i]) break;
+utf8bytes += i;
+for (j = i; j > 0; j--)
+ {
+ *utf8bytes-- = 0x80 | (cvalue & 0x3f);
+ cvalue >>= 6;
+ }
+*utf8bytes = utf8_table2[i] | cvalue;
+return i + 1;
+}
+#endif
+
+
+#ifdef SUPPORT_PCRE16
+/*************************************************
+*         Convert a string to 16-bit             *
+*************************************************/
+
+/* In non-UTF mode, the space needed for a 16-bit string is exactly double the
+8-bit size. For a UTF-8 string, the size needed for UTF-16 is no more than
+double, because up to 0xffff uses no more than 3 bytes in UTF-8 but possibly 4
+in UTF-16. Higher values use 4 bytes in UTF-8 and up to 4 bytes in UTF-16. The
+result is always left in buffer16.
+
+Note that this function does not object to surrogate values. This is
+deliberate; it makes it possible to construct UTF-16 strings that are invalid,
+for the purpose of testing that they are correctly faulted.
+
+Patterns to be converted are either plain ASCII or UTF-8; data lines are always
+in UTF-8 so that values greater than 255 can be handled.
+
+Arguments:
+  data       TRUE if converting a data line; FALSE for a regex
+  p          points to a byte string
+  utf        true if UTF-8 (to be converted to UTF-16)
+  len        number of bytes in the string (excluding trailing zero)
+
+Returns:     number of 16-bit data items used (excluding trailing zero)
+             OR -1 if a UTF-8 string is malformed
+             OR -2 if a value > 0x10ffff is encountered
+             OR -3 if a value > 0xffff is encountered when not in UTF mode
+*/
+
+static int
+to16(int data, pcre_uint8 *p, int utf, int len)
+{
+pcre_uint16 *pp;
+
+if (buffer16_size < 2*len + 2)
+  {
+  if (buffer16 != NULL) free(buffer16);
+  buffer16_size = 2*len + 2;
+  buffer16 = (pcre_uint16 *)malloc(buffer16_size);
+  if (buffer16 == NULL)
+    {
+    fprintf(stderr, "pcretest: malloc(%d) failed for buffer16\n", buffer16_size);
+    exit(1);
+    }
+  }
+
+pp = buffer16;
+
+if (!utf && !data)
+  {
+  while (len-- > 0) *pp++ = *p++;
+  }
+
+else
+  {
+  pcre_uint32 c = 0;
+  while (len > 0)
+    {
+    int chlen = utf82ord(p, &c);
+    if (chlen <= 0) return -1;
+    if (c > 0x10ffff) return -2;
+    p += chlen;
+    len -= chlen;
+    if (c < 0x10000) *pp++ = c; else
+      {
+      if (!utf) return -3;
+      c -= 0x10000;
+      *pp++ = 0xD800 | (c >> 10);
+      *pp++ = 0xDC00 | (c & 0x3ff);
+      }
+    }
+  }
+
+*pp = 0;
+return pp - buffer16;
+}
+#endif
+
+#ifdef SUPPORT_PCRE32
+/*************************************************
+*         Convert a string to 32-bit             *
+*************************************************/
+
+/* In non-UTF mode, the space needed for a 32-bit string is exactly four times the
+8-bit size. For a UTF-8 string, the size needed for UTF-32 is no more than four
+times, because up to 0xffff uses no more than 3 bytes in UTF-8 but possibly 4
+in UTF-32. Higher values use 4 bytes in UTF-8 and up to 4 bytes in UTF-32. The
+result is always left in buffer32.
+
+Note that this function does not object to surrogate values. This is
+deliberate; it makes it possible to construct UTF-32 strings that are invalid,
+for the purpose of testing that they are correctly faulted.
+
+Patterns to be converted are either plain ASCII or UTF-8; data lines are always
+in UTF-8 so that values greater than 255 can be handled.
+
+Arguments:
+  data       TRUE if converting a data line; FALSE for a regex
+  p          points to a byte string
+  utf        true if UTF-8 (to be converted to UTF-32)
+  len        number of bytes in the string (excluding trailing zero)
+
+Returns:     number of 32-bit data items used (excluding trailing zero)
+             OR -1 if a UTF-8 string is malformed
+             OR -2 if a value > 0x10ffff is encountered
+             OR -3 if an ill-formed value is encountered (i.e. a surrogate)
+*/
+
+static int
+to32(int data, pcre_uint8 *p, int utf, int len)
+{
+pcre_uint32 *pp;
+
+if (buffer32_size < 4*len + 4)
+  {
+  if (buffer32 != NULL) free(buffer32);
+  buffer32_size = 4*len + 4;
+  buffer32 = (pcre_uint32 *)malloc(buffer32_size);
+  if (buffer32 == NULL)
+    {
+    fprintf(stderr, "pcretest: malloc(%d) failed for buffer32\n", buffer32_size);
+    exit(1);
+    }
+  }
+
+pp = buffer32;
+
+if (!utf && !data)
+  {
+  while (len-- > 0) *pp++ = *p++;
+  }
+
+else
+  {
+  pcre_uint32 c = 0;
+  while (len > 0)
+    {
+    int chlen = utf82ord(p, &c);
+    if (chlen <= 0) return -1;
+    if (utf)
+      {
+      if (c > 0x10ffff) return -2;
+      if (!data && (c & 0xfffff800u) == 0xd800u) return -3;
+      }
+
+    p += chlen;
+    len -= chlen;
+    *pp++ = c;
+    }
+  }
+
+*pp = 0;
+return pp - buffer32;
+}
+
+/* Check that a 32-bit character string is valid UTF-32.
+
+Arguments:
+  string       points to the string
+  length       length of string, or -1 if the string is zero-terminated
+
+Returns:       TRUE  if the string is a valid UTF-32 string
+               FALSE otherwise
+*/
+
+#ifdef NEVER   /* Not used */
+#ifdef SUPPORT_UTF
+static BOOL
+valid_utf32(pcre_uint32 *string, int length)
+{
+register pcre_uint32 *p;
+register pcre_uint32 c;
+
+for (p = string; length-- > 0; p++)
+  {
+  c = *p;
+  if (c > 0x10ffffu) return FALSE;                 /* Too big */
+  if ((c & 0xfffff800u) == 0xd800u) return FALSE;  /* Surrogate */
+  }
+
+return TRUE;
+}
+#endif /* SUPPORT_UTF */
+#endif /* NEVER */
+#endif /* SUPPORT_PCRE32 */
+
+
+/*************************************************
+*        Read or extend an input line            *
+*************************************************/
+
+/* Input lines are read into buffer, but both patterns and data lines can be
+continued over multiple input lines. In addition, if the buffer fills up, we
+want to automatically expand it so as to be able to handle extremely large
+lines that are needed for certain stress tests. When the input buffer is
+expanded, the other two buffers must also be expanded likewise, and the
+contents of pbuffer, which are a copy of the input for callouts, must be
+preserved (for when expansion happens for a data line). This is not the most
+optimal way of handling this, but hey, this is just a test program!
+
+Arguments:
+  f            the file to read
+  start        where in buffer to start (this *must* be within buffer)
+  prompt       for stdin or readline()
+
+Returns:       pointer to the start of new data
+               could be a copy of start, or could be moved
+               NULL if no data read and EOF reached
+*/
+
+static pcre_uint8 *
+extend_inputline(FILE *f, pcre_uint8 *start, const char *prompt)
+{
+pcre_uint8 *here = start;
+
+for (;;)
+  {
+  size_t rlen = (size_t)(buffer_size - (here - buffer));
+
+  if (rlen > 1000)
+    {
+    int dlen;
+
+    /* If libreadline or libedit support is required, use readline() to read a
+    line if the input is a terminal. Note that readline() removes the trailing
+    newline, so we must put it back again, to be compatible with fgets(). */
+
+#if defined(SUPPORT_LIBREADLINE) || defined(SUPPORT_LIBEDIT)
+    if (isatty(fileno(f)))
+      {
+      size_t len;
+      char *s = readline(prompt);
+      if (s == NULL) return (here == start)? NULL : start;
+      len = strlen(s);
+      if (len > 0) add_history(s);
+      if (len > rlen - 1) len = rlen - 1;
+      memcpy(here, s, len);
+      here[len] = '\n';
+      here[len+1] = 0;
+      free(s);
+      }
+    else
+#endif
+
+    /* Read the next line by normal means, prompting if the file is stdin. */
+
+      {
+      if (f == stdin) printf("%s", prompt);
+      if (fgets((char *)here, rlen,  f) == NULL)
+        return (here == start)? NULL : start;
+      }
+
+    dlen = (int)strlen((char *)here);
+    if (dlen > 0 && here[dlen - 1] == '\n') return start;
+    here += dlen;
+    }
+
+  else
+    {
+    int new_buffer_size = 2*buffer_size;
+    pcre_uint8 *new_buffer = (pcre_uint8 *)malloc(new_buffer_size);
+    pcre_uint8 *new_pbuffer = (pcre_uint8 *)malloc(new_buffer_size);
+
+    if (new_buffer == NULL || new_pbuffer == NULL)
+      {
+      fprintf(stderr, "pcretest: malloc(%d) failed\n", new_buffer_size);
+      exit(1);
+      }
+
+    memcpy(new_buffer, buffer, buffer_size);
+    memcpy(new_pbuffer, pbuffer, buffer_size);
+
+    buffer_size = new_buffer_size;
+
+    start = new_buffer + (start - buffer);
+    here = new_buffer + (here - buffer);
+
+    free(buffer);
+    free(pbuffer);
+
+    buffer = new_buffer;
+    pbuffer = new_pbuffer;
+    }
+  }
+
+return NULL;  /* Control never gets here */
+}
+
+
+
+/*************************************************
+*          Read number from string               *
+*************************************************/
+
+/* We don't use strtoul() because SunOS4 doesn't have it. Rather than mess
+around with conditional compilation, just do the job by hand. It is only used
+for unpicking arguments, so just keep it simple.
+
+Arguments:
+  str           string to be converted
+  endptr        where to put the end pointer
+
+Returns:        the unsigned long
+*/
+
+static int
+get_value(pcre_uint8 *str, pcre_uint8 **endptr)
+{
+int result = 0;
+while(*str != 0 && isspace(*str)) str++;
+while (isdigit(*str)) result = result * 10 + (int)(*str++ - '0');
+*endptr = str;
+return(result);
+}
+
+
+
+/*************************************************
+*             Print one character                *
+*************************************************/
+
+/* Print a single character either literally, or as a hex escape. */
+
+static int pchar(pcre_uint32 c, FILE *f)
+{
+int n = 0;
+if (PRINTOK(c))
+  {
+  if (f != NULL) fprintf(f, "%c", c);
+  return 1;
+  }
+
+if (c < 0x100)
+  {
+  if (use_utf)
+    {
+    if (f != NULL) fprintf(f, "\\x{%02x}", c);
+    return 6;
+    }
+  else
+    {
+    if (f != NULL) fprintf(f, "\\x%02x", c);
+    return 4;
+    }
+  }
+
+if (f != NULL) n = fprintf(f, "\\x{%02x}", c);
+return n >= 0 ? n : 0;
+}
+
+
+
+#ifdef SUPPORT_PCRE8
+/*************************************************
+*         Print 8-bit character string           *
+*************************************************/
+
+/* Must handle UTF-8 strings in utf8 mode. Yields number of characters printed.
+If handed a NULL file, just counts chars without printing. */
+
+static int pchars(pcre_uint8 *p, int length, FILE *f)
+{
+pcre_uint32 c = 0;
+int yield = 0;
+
+if (length < 0)
+  length = strlen((char *)p);
+
+while (length-- > 0)
+  {
+#if !defined NOUTF
+  if (use_utf)
+    {
+    int rc = utf82ord(p, &c);
+    if (rc > 0 && rc <= length + 1)   /* Mustn't run over the end */
+      {
+      length -= rc - 1;
+      p += rc;
+      yield += pchar(c, f);
+      continue;
+      }
+    }
+#endif
+  c = *p++;
+  yield += pchar(c, f);
+  }
+
+return yield;
+}
+#endif
+
+
+
+#ifdef SUPPORT_PCRE16
+/*************************************************
+*    Find length of 0-terminated 16-bit string   *
+*************************************************/
+
+static int strlen16(PCRE_SPTR16 p)
+{
+int len = 0;
+while (*p++ != 0) len++;
+return len;
+}
+#endif  /* SUPPORT_PCRE16 */
+
+
+
+#ifdef SUPPORT_PCRE32
+/*************************************************
+*    Find length of 0-terminated 32-bit string   *
+*************************************************/
+
+static int strlen32(PCRE_SPTR32 p)
+{
+int len = 0;
+while (*p++ != 0) len++;
+return len;
+}
+#endif  /* SUPPORT_PCRE32 */
+
+
+
+#ifdef SUPPORT_PCRE16
+/*************************************************
+*           Print 16-bit character string        *
+*************************************************/
+
+/* Must handle UTF-16 strings in utf mode. Yields number of characters printed.
+If handed a NULL file, just counts chars without printing. */
+
+static int pchars16(PCRE_SPTR16 p, int length, FILE *f)
+{
+int yield = 0;
+
+if (length < 0)
+  length = strlen16(p);
+
+while (length-- > 0)
+  {
+  pcre_uint32 c = *p++ & 0xffff;
+#if !defined NOUTF
+  if (use_utf && c >= 0xD800 && c < 0xDC00 && length > 0)
+    {
+    int d = *p & 0xffff;
+    if (d >= 0xDC00 && d <= 0xDFFF)
+      {
+      c = ((c & 0x3ff) << 10) + (d & 0x3ff) + 0x10000;
+      length--;
+      p++;
+      }
+    }
+#endif
+  yield += pchar(c, f);
+  }
+
+return yield;
+}
+#endif  /* SUPPORT_PCRE16 */
+
+
+
+#ifdef SUPPORT_PCRE32
+/*************************************************
+*           Print 32-bit character string        *
+*************************************************/
+
+/* Must handle UTF-32 strings in utf mode. Yields number of characters printed.
+If handed a NULL file, just counts chars without printing. */
+
+static int pchars32(PCRE_SPTR32 p, int length, BOOL utf, FILE *f)
+{
+int yield = 0;
+
+(void)(utf);  /* Avoid compiler warning */
+
+if (length < 0)
+  length = strlen32(p);
+
+while (length-- > 0)
+  {
+  pcre_uint32 c = *p++;
+  yield += pchar(c, f);
+  }
+
+return yield;
+}
+#endif  /* SUPPORT_PCRE32 */
+
+
+
+#ifdef SUPPORT_PCRE8
+/*************************************************
+*     Read a capture name (8-bit) and check it   *
+*************************************************/
+
+static pcre_uint8 *
+read_capture_name8(pcre_uint8 *p, pcre_uint8 **pp, pcre *re)
+{
+pcre_uint8 *npp = *pp;
+while (isalnum(*p)) *npp++ = *p++;
+*npp++ = 0;
+*npp = 0;
+if (pcre_get_stringnumber(re, (char *)(*pp)) < 0)
+  {
+  fprintf(outfile, "no parentheses with name \"");
+  PCHARSV(*pp, 0, -1, outfile);
+  fprintf(outfile, "\"\n");
+  }
+
+*pp = npp;
+return p;
+}
+#endif  /* SUPPORT_PCRE8 */
+
+
+
+#ifdef SUPPORT_PCRE16
+/*************************************************
+*     Read a capture name (16-bit) and check it  *
+*************************************************/
+
+/* Note that the text being read is 8-bit. */
+
+static pcre_uint8 *
+read_capture_name16(pcre_uint8 *p, pcre_uint16 **pp, pcre *re)
+{
+pcre_uint16 *npp = *pp;
+while (isalnum(*p)) *npp++ = *p++;
+*npp++ = 0;
+*npp = 0;
+if (pcre16_get_stringnumber((pcre16 *)re, (PCRE_SPTR16)(*pp)) < 0)
+  {
+  fprintf(outfile, "no parentheses with name \"");
+  PCHARSV(*pp, 0, -1, outfile);
+  fprintf(outfile, "\"\n");
+  }
+*pp = npp;
+return p;
+}
+#endif  /* SUPPORT_PCRE16 */
+
+
+
+#ifdef SUPPORT_PCRE32
+/*************************************************
+*     Read a capture name (32-bit) and check it  *
+*************************************************/
+
+/* Note that the text being read is 8-bit. */
+
+static pcre_uint8 *
+read_capture_name32(pcre_uint8 *p, pcre_uint32 **pp, pcre *re)
+{
+pcre_uint32 *npp = *pp;
+while (isalnum(*p)) *npp++ = *p++;
+*npp++ = 0;
+*npp = 0;
+if (pcre32_get_stringnumber((pcre32 *)re, (PCRE_SPTR32)(*pp)) < 0)
+  {
+  fprintf(outfile, "no parentheses with name \"");
+  PCHARSV(*pp, 0, -1, outfile);
+  fprintf(outfile, "\"\n");
+  }
+*pp = npp;
+return p;
+}
+#endif  /* SUPPORT_PCRE32 */
+
+
+
+/*************************************************
+*              Callout function                  *
+*************************************************/
+
+/* Called from PCRE as a result of the (?C) item. We print out where we are in
+the match. Yield zero unless more callouts than the fail count, or the callout
+data is not zero. */
+
+static int callout(pcre_callout_block *cb)
+{
+FILE *f = (first_callout | callout_extra)? outfile : NULL;
+int i, pre_start, post_start, subject_length;
+
+if (callout_extra)
+  {
+  fprintf(f, "Callout %d: last capture = %d\n",
+    cb->callout_number, cb->capture_last);
+
+  for (i = 0; i < cb->capture_top * 2; i += 2)
+    {
+    if (cb->offset_vector[i] < 0)
+      fprintf(f, "%2d: \n", i/2);
+    else
+      {
+      fprintf(f, "%2d: ", i/2);
+      PCHARSV(cb->subject, cb->offset_vector[i],
+        cb->offset_vector[i+1] - cb->offset_vector[i], f);
+      fprintf(f, "\n");
+      }
+    }
+  }
+
+/* Re-print the subject in canonical form, the first time or if giving full
+datails. On subsequent calls in the same match, we use pchars just to find the
+printed lengths of the substrings. */
+
+if (f != NULL) fprintf(f, "--->");
+
+PCHARS(pre_start, cb->subject, 0, cb->start_match, f);
+PCHARS(post_start, cb->subject, cb->start_match,
+  cb->current_position - cb->start_match, f);
+
+PCHARS(subject_length, cb->subject, 0, cb->subject_length, NULL);
+
+PCHARSV(cb->subject, cb->current_position,
+  cb->subject_length - cb->current_position, f);
+
+if (f != NULL) fprintf(f, "\n");
+
+/* Always print appropriate indicators, with callout number if not already
+shown. For automatic callouts, show the pattern offset. */
+
+if (cb->callout_number == 255)
+  {
+  fprintf(outfile, "%+3d ", cb->pattern_position);
+  if (cb->pattern_position > 99) fprintf(outfile, "\n    ");
+  }
+else
+  {
+  if (callout_extra) fprintf(outfile, "    ");
+    else fprintf(outfile, "%3d ", cb->callout_number);
+  }
+
+for (i = 0; i < pre_start; i++) fprintf(outfile, " ");
+fprintf(outfile, "^");
+
+if (post_start > 0)
+  {
+  for (i = 0; i < post_start - 1; i++) fprintf(outfile, " ");
+  fprintf(outfile, "^");
+  }
+
+for (i = 0; i < subject_length - pre_start - post_start + 4; i++)
+  fprintf(outfile, " ");
+
+fprintf(outfile, "%.*s", (cb->next_item_length == 0)? 1 : cb->next_item_length,
+  pbuffer + cb->pattern_position);
+
+fprintf(outfile, "\n");
+first_callout = 0;
+
+if (cb->mark != last_callout_mark)
+  {
+  if (cb->mark == NULL)
+    fprintf(outfile, "Latest Mark: \n");
+  else
+    {
+    fprintf(outfile, "Latest Mark: ");
+    PCHARSV(cb->mark, 0, -1, outfile);
+    putc('\n', outfile);
+    }
+  last_callout_mark = cb->mark;
+  }
+
+if (cb->callout_data != NULL)
+  {
+  int callout_data = *((int *)(cb->callout_data));
+  if (callout_data != 0)
+    {
+    fprintf(outfile, "Callout data = %d\n", callout_data);
+    return callout_data;
+    }
+  }
+
+return (cb->callout_number != callout_fail_id)? 0 :
+       (++callout_count >= callout_fail_count)? 1 : 0;
+}
+
+
+/*************************************************
+*            Local malloc functions              *
+*************************************************/
+
+/* Alternative malloc function, to test functionality and save the size of a
+compiled re, which is the first store request that pcre_compile() makes. The
+show_malloc variable is set only during matching. */
+
+static void *new_malloc(size_t size)
+{
+void *block = malloc(size);
+gotten_store = size;
+if (first_gotten_store == 0) first_gotten_store = size;
+if (show_malloc)
+  fprintf(outfile, "malloc       %3d %p\n", (int)size, block);
+return block;
+}
+
+static void new_free(void *block)
+{
+if (show_malloc)
+  fprintf(outfile, "free             %p\n", block);
+free(block);
+}
+
+/* For recursion malloc/free, to test stacking calls */
+
+static void *stack_malloc(size_t size)
+{
+void *block = malloc(size);
+if (show_malloc)
+  fprintf(outfile, "stack_malloc %3d %p\n", (int)size, block);
+return block;
+}
+
+static void stack_free(void *block)
+{
+if (show_malloc)
+  fprintf(outfile, "stack_free       %p\n", block);
+free(block);
+}
+
+
+/*************************************************
+*          Call pcre_fullinfo()                  *
+*************************************************/
+
+/* Get one piece of information from the pcre_fullinfo() function. When only
+one of 8-, 16- or 32-bit is supported, pcre_mode should always have the correct
+value, but the code is defensive.
+
+Arguments:
+  re        compiled regex
+  study     study data
+  option    PCRE_INFO_xxx option
+  ptr       where to put the data
+
+Returns:    0 when OK, < 0 on error
+*/
+
+static int
+new_info(pcre *re, pcre_extra *study, int option, void *ptr)
+{
+int rc;
+
+if (pcre_mode == PCRE32_MODE)
+#ifdef SUPPORT_PCRE32
+  rc = pcre32_fullinfo((pcre32 *)re, (pcre32_extra *)study, option, ptr);
+#else
+  rc = PCRE_ERROR_BADMODE;
+#endif
+else if (pcre_mode == PCRE16_MODE)
+#ifdef SUPPORT_PCRE16
+  rc = pcre16_fullinfo((pcre16 *)re, (pcre16_extra *)study, option, ptr);
+#else
+  rc = PCRE_ERROR_BADMODE;
+#endif
+else
+#ifdef SUPPORT_PCRE8
+  rc = pcre_fullinfo(re, study, option, ptr);
+#else
+  rc = PCRE_ERROR_BADMODE;
+#endif
+
+if (rc < 0 && rc != PCRE_ERROR_UNSET)
+  {
+  fprintf(outfile, "Error %d from pcre%s_fullinfo(%d)\n", rc,
+    pcre_mode == PCRE32_MODE ? "32" : pcre_mode == PCRE16_MODE ? "16" : "", option);
+  if (rc == PCRE_ERROR_BADMODE)
+    fprintf(outfile, "Running in %d-bit mode but pattern was compiled in "
+      "%d-bit mode\n", 8 * CHAR_SIZE,
+      8 * (REAL_PCRE_FLAGS(re) & PCRE_MODE_MASK));
+  }
+
+return rc;
+}
+
+
+
+/*************************************************
+*             Swap byte functions                *
+*************************************************/
+
+/* The following functions swap the bytes of a pcre_uint16 and pcre_uint32
+value, respectively.
+
+Arguments:
+  value        any number
+
+Returns:       the byte swapped value
+*/
+
+static pcre_uint32
+swap_uint32(pcre_uint32 value)
+{
+return ((value & 0x000000ff) << 24) |
+       ((value & 0x0000ff00) <<  8) |
+       ((value & 0x00ff0000) >>  8) |
+       (value >> 24);
+}
+
+static pcre_uint16
+swap_uint16(pcre_uint16 value)
+{
+return (value >> 8) | (value << 8);
+}
+
+
+
+/*************************************************
+*        Flip bytes in a compiled pattern        *
+*************************************************/
+
+/* This function is called if the 'F' option was present on a pattern that is
+to be written to a file. We flip the bytes of all the integer fields in the
+regex data block and the study block. In 16-bit mode this also flips relevant
+bytes in the pattern itself. This is to make it possible to test PCRE's
+ability to reload byte-flipped patterns, e.g. those compiled on a different
+architecture. */
+
+#if defined SUPPORT_PCRE8 || defined SUPPORT_PCRE16
+static void
+regexflip8_or_16(pcre *ere, pcre_extra *extra)
+{
+real_pcre8_or_16 *re = (real_pcre8_or_16 *)ere;
+#ifdef SUPPORT_PCRE16
+int op;
+pcre_uint16 *ptr = (pcre_uint16 *)re + re->name_table_offset;
+int length = re->name_count * re->name_entry_size;
+#ifdef SUPPORT_UTF
+BOOL utf = (re->options & PCRE_UTF16) != 0;
+BOOL utf16_char = FALSE;
+#endif /* SUPPORT_UTF */
+#endif /* SUPPORT_PCRE16 */
+
+/* Always flip the bytes in the main data block and study blocks. */
+
+re->magic_number = REVERSED_MAGIC_NUMBER;
+re->size = swap_uint32(re->size);
+re->options = swap_uint32(re->options);
+re->flags = swap_uint32(re->flags);
+re->limit_match = swap_uint32(re->limit_match);
+re->limit_recursion = swap_uint32(re->limit_recursion);
+re->first_char = swap_uint16(re->first_char);
+re->req_char = swap_uint16(re->req_char);
+re->max_lookbehind = swap_uint16(re->max_lookbehind);
+re->top_bracket = swap_uint16(re->top_bracket);
+re->top_backref = swap_uint16(re->top_backref);
+re->name_table_offset = swap_uint16(re->name_table_offset);
+re->name_entry_size = swap_uint16(re->name_entry_size);
+re->name_count = swap_uint16(re->name_count);
+re->ref_count = swap_uint16(re->ref_count);
+
+if (extra != NULL)
+  {
+  pcre_study_data *rsd = (pcre_study_data *)(extra->study_data);
+  rsd->size = swap_uint32(rsd->size);
+  rsd->flags = swap_uint32(rsd->flags);
+  rsd->minlength = swap_uint32(rsd->minlength);
+  }
+
+/* In 8-bit mode, that is all we need to do. In 16-bit mode we must swap bytes
+in the name table, if present, and then in the pattern itself. */
+
+#ifdef SUPPORT_PCRE16
+if (pcre_mode != PCRE16_MODE) return;
+
+while(TRUE)
+  {
+  /* Swap previous characters. */
+  while (length-- > 0)
+    {
+    *ptr = swap_uint16(*ptr);
+    ptr++;
+    }
+#ifdef SUPPORT_UTF
+  if (utf16_char)
+    {
+    if ((ptr[-1] & 0xfc00) == 0xd800)
+      {
+      /* We know that there is only one extra character in UTF-16. */
+      *ptr = swap_uint16(*ptr);
+      ptr++;
+      }
+    }
+  utf16_char = FALSE;
+#endif /* SUPPORT_UTF */
+
+  /* Get next opcode. */
+
+  length = 0;
+  op = *ptr;
+  *ptr++ = swap_uint16(op);
+
+  switch (op)
+    {
+    case OP_END:
+    return;
+
+#ifdef SUPPORT_UTF
+    case OP_CHAR:
+    case OP_CHARI:
+    case OP_NOT:
+    case OP_NOTI:
+    case OP_STAR:
+    case OP_MINSTAR:
+    case OP_PLUS:
+    case OP_MINPLUS:
+    case OP_QUERY:
+    case OP_MINQUERY:
+    case OP_UPTO:
+    case OP_MINUPTO:
+    case OP_EXACT:
+    case OP_POSSTAR:
+    case OP_POSPLUS:
+    case OP_POSQUERY:
+    case OP_POSUPTO:
+    case OP_STARI:
+    case OP_MINSTARI:
+    case OP_PLUSI:
+    case OP_MINPLUSI:
+    case OP_QUERYI:
+    case OP_MINQUERYI:
+    case OP_UPTOI:
+    case OP_MINUPTOI:
+    case OP_EXACTI:
+    case OP_POSSTARI:
+    case OP_POSPLUSI:
+    case OP_POSQUERYI:
+    case OP_POSUPTOI:
+    case OP_NOTSTAR:
+    case OP_NOTMINSTAR:
+    case OP_NOTPLUS:
+    case OP_NOTMINPLUS:
+    case OP_NOTQUERY:
+    case OP_NOTMINQUERY:
+    case OP_NOTUPTO:
+    case OP_NOTMINUPTO:
+    case OP_NOTEXACT:
+    case OP_NOTPOSSTAR:
+    case OP_NOTPOSPLUS:
+    case OP_NOTPOSQUERY:
+    case OP_NOTPOSUPTO:
+    case OP_NOTSTARI:
+    case OP_NOTMINSTARI:
+    case OP_NOTPLUSI:
+    case OP_NOTMINPLUSI:
+    case OP_NOTQUERYI:
+    case OP_NOTMINQUERYI:
+    case OP_NOTUPTOI:
+    case OP_NOTMINUPTOI:
+    case OP_NOTEXACTI:
+    case OP_NOTPOSSTARI:
+    case OP_NOTPOSPLUSI:
+    case OP_NOTPOSQUERYI:
+    case OP_NOTPOSUPTOI:
+    if (utf) utf16_char = TRUE;
+#endif
+    /* Fall through. */
+
+    default:
+    length = OP_lengths16[op] - 1;
+    break;
+
+    case OP_CLASS:
+    case OP_NCLASS:
+    /* Skip the character bit map. */
+    ptr += 32/sizeof(pcre_uint16);
+    length = 0;
+    break;
+
+    case OP_XCLASS:
+    /* LINK_SIZE can be 1 or 2 in 16 bit mode. */
+    if (LINK_SIZE > 1)
+      length = (int)((((unsigned int)(ptr[0]) << 16) | (unsigned int)(ptr[1]))
+        - (1 + LINK_SIZE + 1));
+    else
+      length = (int)((unsigned int)(ptr[0]) - (1 + LINK_SIZE + 1));
+
+    /* Reverse the size of the XCLASS instance. */
+    *ptr = swap_uint16(*ptr);
+    ptr++;
+    if (LINK_SIZE > 1)
+      {
+      *ptr = swap_uint16(*ptr);
+      ptr++;
+      }
+
+    op = *ptr;
+    *ptr = swap_uint16(op);
+    ptr++;
+    if ((op & XCL_MAP) != 0)
+      {
+      /* Skip the character bit map. */
+      ptr += 32/sizeof(pcre_uint16);
+      length -= 32/sizeof(pcre_uint16);
+      }
+    break;
+    }
+  }
+/* Control should never reach here in 16 bit mode. */
+#endif /* SUPPORT_PCRE16 */
+}
+#endif /* SUPPORT_PCRE[8|16] */
+
+
+
+#if defined SUPPORT_PCRE32
+static void
+regexflip_32(pcre *ere, pcre_extra *extra)
+{
+real_pcre32 *re = (real_pcre32 *)ere;
+int op;
+pcre_uint32 *ptr = (pcre_uint32 *)re + re->name_table_offset;
+int length = re->name_count * re->name_entry_size;
+
+/* Always flip the bytes in the main data block and study blocks. */
+
+re->magic_number = REVERSED_MAGIC_NUMBER;
+re->size = swap_uint32(re->size);
+re->options = swap_uint32(re->options);
+re->flags = swap_uint32(re->flags);
+re->limit_match = swap_uint32(re->limit_match);
+re->limit_recursion = swap_uint32(re->limit_recursion);
+re->first_char = swap_uint32(re->first_char);
+re->req_char = swap_uint32(re->req_char);
+re->max_lookbehind = swap_uint16(re->max_lookbehind);
+re->top_bracket = swap_uint16(re->top_bracket);
+re->top_backref = swap_uint16(re->top_backref);
+re->name_table_offset = swap_uint16(re->name_table_offset);
+re->name_entry_size = swap_uint16(re->name_entry_size);
+re->name_count = swap_uint16(re->name_count);
+re->ref_count = swap_uint16(re->ref_count);
+
+if (extra != NULL)
+  {
+  pcre_study_data *rsd = (pcre_study_data *)(extra->study_data);
+  rsd->size = swap_uint32(rsd->size);
+  rsd->flags = swap_uint32(rsd->flags);
+  rsd->minlength = swap_uint32(rsd->minlength);
+  }
+
+/* In 32-bit mode we must swap bytes in the name table, if present, and then in
+the pattern itself. */
+
+while(TRUE)
+  {
+  /* Swap previous characters. */
+  while (length-- > 0)
+    {
+    *ptr = swap_uint32(*ptr);
+    ptr++;
+    }
+
+  /* Get next opcode. */
+
+  length = 0;
+  op = *ptr;
+  *ptr++ = swap_uint32(op);
+
+  switch (op)
+    {
+    case OP_END:
+    return;
+
+    default:
+    length = OP_lengths32[op] - 1;
+    break;
+
+    case OP_CLASS:
+    case OP_NCLASS:
+    /* Skip the character bit map. */
+    ptr += 32/sizeof(pcre_uint32);
+    length = 0;
+    break;
+
+    case OP_XCLASS:
+    /* LINK_SIZE can only be 1 in 32-bit mode. */
+    length = (int)((unsigned int)(ptr[0]) - (1 + LINK_SIZE + 1));
+
+    /* Reverse the size of the XCLASS instance. */
+    *ptr = swap_uint32(*ptr);
+    ptr++;
+
+    op = *ptr;
+    *ptr = swap_uint32(op);
+    ptr++;
+    if ((op & XCL_MAP) != 0)
+      {
+      /* Skip the character bit map. */
+      ptr += 32/sizeof(pcre_uint32);
+      length -= 32/sizeof(pcre_uint32);
+      }
+    break;
+    }
+  }
+/* Control should never reach here in 32 bit mode. */
+}
+
+#endif /* SUPPORT_PCRE32 */
+
+
+
+static void
+regexflip(pcre *ere, pcre_extra *extra)
+{
+#if defined SUPPORT_PCRE32
+  if (REAL_PCRE_FLAGS(ere) & PCRE_MODE32)
+    regexflip_32(ere, extra);
+#endif
+#if defined SUPPORT_PCRE8 || defined SUPPORT_PCRE16
+  if (REAL_PCRE_FLAGS(ere) & (PCRE_MODE8 | PCRE_MODE16))
+    regexflip8_or_16(ere, extra);
+#endif
+}
+
+
+
+/*************************************************
+*        Check match or recursion limit          *
+*************************************************/
+
+static int
+check_match_limit(pcre *re, pcre_extra *extra, pcre_uint8 *bptr, int len,
+  int start_offset, int options, int *use_offsets, int use_size_offsets,
+  int flag, unsigned long int *limit, int errnumber, const char *msg)
+{
+int count;
+int min = 0;
+int mid = 64;
+int max = -1;
+
+extra->flags |= flag;
+
+for (;;)
+  {
+  *limit = mid;
+
+  PCRE_EXEC(count, re, extra, bptr, len, start_offset, options,
+    use_offsets, use_size_offsets);
+
+  if (count == errnumber)
+    {
+    /* fprintf(outfile, "Testing %s limit = %d\n", msg, mid); */
+    min = mid;
+    mid = (mid == max - 1)? max : (max > 0)? (min + max)/2 : mid*2;
+    }
+
+  else if (count >= 0 || count == PCRE_ERROR_NOMATCH ||
+                         count == PCRE_ERROR_PARTIAL)
+    {
+    if (mid == min + 1)
+      {
+      fprintf(outfile, "Minimum %s limit = %d\n", msg, mid);
+      break;
+      }
+    /* fprintf(outfile, "Testing %s limit = %d\n", msg, mid); */
+    max = mid;
+    mid = (min + mid)/2;
+    }
+  else break;    /* Some other error */
+  }
+
+extra->flags &= ~flag;
+return count;
+}
+
+
+
+/*************************************************
+*         Case-independent strncmp() function    *
+*************************************************/
+
+/*
+Arguments:
+  s         first string
+  t         second string
+  n         number of characters to compare
+
+Returns:    < 0, = 0, or > 0, according to the comparison
+*/
+
+static int
+strncmpic(pcre_uint8 *s, pcre_uint8 *t, int n)
+{
+while (n--)
+  {
+  int c = tolower(*s++) - tolower(*t++);
+  if (c) return c;
+  }
+return 0;
+}
+
+
+
+/*************************************************
+*         Check newline indicator                *
+*************************************************/
+
+/* This is used both at compile and run-time to check for  escapes. Print
+a message and return 0 if there is no match.
+
+Arguments:
+  p           points after the leading '<'
+  f           file for error message
+
+Returns:      appropriate PCRE_NEWLINE_xxx flags, or 0
+*/
+
+static int
+check_newline(pcre_uint8 *p, FILE *f)
+{
+if (strncmpic(p, (pcre_uint8 *)"cr>", 3) == 0) return PCRE_NEWLINE_CR;
+if (strncmpic(p, (pcre_uint8 *)"lf>", 3) == 0) return PCRE_NEWLINE_LF;
+if (strncmpic(p, (pcre_uint8 *)"crlf>", 5) == 0) return PCRE_NEWLINE_CRLF;
+if (strncmpic(p, (pcre_uint8 *)"anycrlf>", 8) == 0) return PCRE_NEWLINE_ANYCRLF;
+if (strncmpic(p, (pcre_uint8 *)"any>", 4) == 0) return PCRE_NEWLINE_ANY;
+if (strncmpic(p, (pcre_uint8 *)"bsr_anycrlf>", 12) == 0) return PCRE_BSR_ANYCRLF;
+if (strncmpic(p, (pcre_uint8 *)"bsr_unicode>", 12) == 0) return PCRE_BSR_UNICODE;
+fprintf(f, "Unknown newline type at: <%s\n", p);
+return 0;
+}
+
+
+
+/*************************************************
+*             Usage function                     *
+*************************************************/
+
+static void
+usage(void)
+{
+printf("Usage:     pcretest [options] [ []]\n\n");
+printf("Input and output default to stdin and stdout.\n");
+#if defined(SUPPORT_LIBREADLINE) || defined(SUPPORT_LIBEDIT)
+printf("If input is a terminal, readline() is used to read from it.\n");
+#else
+printf("This version of pcretest is not linked with readline().\n");
+#endif
+printf("\nOptions:\n");
+#ifdef SUPPORT_PCRE16
+printf("  -16      use the 16-bit library\n");
+#endif
+#ifdef SUPPORT_PCRE32
+printf("  -32      use the 32-bit library\n");
+#endif
+printf("  -b       show compiled code\n");
+printf("  -C       show PCRE compile-time options and exit\n");
+printf("  -C arg   show a specific compile-time option\n");
+printf("           and exit with its value. The arg can be:\n");
+printf("     linksize     internal link size [2, 3, 4]\n");
+printf("     pcre8        8 bit library support enabled [0, 1]\n");
+printf("     pcre16       16 bit library support enabled [0, 1]\n");
+printf("     pcre32       32 bit library support enabled [0, 1]\n");
+printf("     utf          Unicode Transformation Format supported [0, 1]\n");
+printf("     ucp          Unicode Properties supported [0, 1]\n");
+printf("     jit          Just-in-time compiler supported [0, 1]\n");
+printf("     newline      Newline type [CR, LF, CRLF, ANYCRLF, ANY, ???]\n");
+printf("  -d       debug: show compiled code and information (-b and -i)\n");
+#if !defined NODFA
+printf("  -dfa     force DFA matching for all subjects\n");
+#endif
+printf("  -help    show usage information\n");
+printf("  -i       show information about compiled patterns\n"
+       "  -M       find MATCH_LIMIT minimum for each subject\n"
+       "  -m       output memory used information\n"
+       "  -o    set size of offsets vector to \n");
+#if !defined NOPOSIX
+printf("  -p       use POSIX interface\n");
+#endif
+printf("  -q       quiet: do not output PCRE version number at start\n");
+printf("  -S    set stack size to  megabytes\n");
+printf("  -s       force each pattern to be studied at basic level\n"
+       "  -s+      force each pattern to be studied, using JIT if available\n"
+       "  -s++     ditto, verifying when JIT was actually used\n"
+       "  -s+n     force each pattern to be studied, using JIT if available,\n"
+       "             where 1 <= n <= 7 selects JIT options\n"
+       "  -s++n    ditto, verifying when JIT was actually used\n"
+       "  -t       time compilation and execution\n");
+printf("  -t    time compilation and execution, repeating  times\n");
+printf("  -tm      time execution (matching) only\n");
+printf("  -tm   time execution (matching) only, repeating  times\n");
+}
+
+
+
+/*************************************************
+*                Main Program                    *
+*************************************************/
+
+/* Read lines from named file or stdin and write to named file or stdout; lines
+consist of a regular expression, in delimiters and optionally followed by
+options, followed by a set of test data, terminated by an empty line. */
+
+int main(int argc, char **argv)
+{
+FILE *infile = stdin;
+const char *version;
+int options = 0;
+int study_options = 0;
+int default_find_match_limit = FALSE;
+int op = 1;
+int timeit = 0;
+int timeitm = 0;
+int showinfo = 0;
+int showstore = 0;
+int force_study = -1;
+int force_study_options = 0;
+int quiet = 0;
+int size_offsets = 45;
+int size_offsets_max;
+int *offsets = NULL;
+int debug = 0;
+int done = 0;
+int all_use_dfa = 0;
+int verify_jit = 0;
+int yield = 0;
+int stack_size;
+pcre_uint8 *dbuffer = NULL;
+size_t dbuffer_size = 1u << 14;
+
+#if !defined NOPOSIX
+int posix = 0;
+#endif
+#if !defined NODFA
+int *dfa_workspace = NULL;
+#endif
+
+pcre_jit_stack *jit_stack = NULL;
+
+/* These vectors store, end-to-end, a list of zero-terminated captured
+substring names, each list itself being terminated by an empty name. Assume
+that 1024 is plenty long enough for the few names we'll be testing. It is
+easiest to keep separate 8-, 16- and 32-bit versions, using the 32-bit version
+for the actual memory, to ensure alignment. */
+
+pcre_uint32 copynames[1024];
+pcre_uint32 getnames[1024];
+
+#ifdef SUPPORT_PCRE32
+pcre_uint32 *cn32ptr;
+pcre_uint32 *gn32ptr;
+#endif
+
+#ifdef SUPPORT_PCRE16
+pcre_uint16 *copynames16 = (pcre_uint16 *)copynames;
+pcre_uint16 *getnames16 = (pcre_uint16 *)getnames;
+pcre_uint16 *cn16ptr;
+pcre_uint16 *gn16ptr;
+#endif
+
+#ifdef SUPPORT_PCRE8
+pcre_uint8 *copynames8 = (pcre_uint8 *)copynames;
+pcre_uint8 *getnames8 = (pcre_uint8 *)getnames;
+pcre_uint8 *cn8ptr;
+pcre_uint8 *gn8ptr;
+#endif
+
+/* Get buffers from malloc() so that valgrind will check their misuse when
+debugging. They grow automatically when very long lines are read. The 16-
+and 32-bit buffers (buffer16, buffer32) are obtained only if needed. */
+
+buffer = (pcre_uint8 *)malloc(buffer_size);
+pbuffer = (pcre_uint8 *)malloc(buffer_size);
+
+/* The outfile variable is static so that new_malloc can use it. */
+
+outfile = stdout;
+
+/* The following  _setmode() stuff is some Windows magic that tells its runtime
+library to translate CRLF into a single LF character. At least, that's what
+I've been told: never having used Windows I take this all on trust. Originally
+it set 0x8000, but then I was advised that _O_BINARY was better. */
+
+#if defined(_WIN32) || defined(WIN32)
+_setmode( _fileno( stdout ), _O_BINARY );
+#endif
+
+/* Get the version number: both pcre_version() and pcre16_version() give the
+same answer. We just need to ensure that we call one that is available. */
+
+#if defined SUPPORT_PCRE8
+version = pcre_version();
+#elif defined SUPPORT_PCRE16
+version = pcre16_version();
+#elif defined SUPPORT_PCRE32
+version = pcre32_version();
+#endif
+
+/* Scan options */
+
+while (argc > 1 && argv[op][0] == '-')
+  {
+  pcre_uint8 *endptr;
+  char *arg = argv[op];
+
+  if (strcmp(arg, "-m") == 0) showstore = 1;
+  else if (strcmp(arg, "-s") == 0) force_study = 0;
+
+  else if (strncmp(arg, "-s+", 3) == 0)
+    {
+    arg += 3;
+    if (*arg == '+') { arg++; verify_jit = TRUE; }
+    force_study = 1;
+    if (*arg == 0)
+      force_study_options = jit_study_bits[6];
+    else if (*arg >= '1' && *arg <= '7')
+      force_study_options = jit_study_bits[*arg - '1'];
+    else goto BAD_ARG;
+    }
+  else if (strcmp(arg, "-8") == 0)
+    {
+#ifdef SUPPORT_PCRE8
+    pcre_mode = PCRE8_MODE;
+#else
+    printf("** This version of PCRE was built without 8-bit support\n");
+    exit(1);
+#endif
+    }
+  else if (strcmp(arg, "-16") == 0)
+    {
+#ifdef SUPPORT_PCRE16
+    pcre_mode = PCRE16_MODE;
+#else
+    printf("** This version of PCRE was built without 16-bit support\n");
+    exit(1);
+#endif
+    }
+  else if (strcmp(arg, "-32") == 0)
+    {
+#ifdef SUPPORT_PCRE32
+    pcre_mode = PCRE32_MODE;
+#else
+    printf("** This version of PCRE was built without 32-bit support\n");
+    exit(1);
+#endif
+    }
+  else if (strcmp(arg, "-q") == 0) quiet = 1;
+  else if (strcmp(arg, "-b") == 0) debug = 1;
+  else if (strcmp(arg, "-i") == 0) showinfo = 1;
+  else if (strcmp(arg, "-d") == 0) showinfo = debug = 1;
+  else if (strcmp(arg, "-M") == 0) default_find_match_limit = TRUE;
+#if !defined NODFA
+  else if (strcmp(arg, "-dfa") == 0) all_use_dfa = 1;
+#endif
+  else if (strcmp(arg, "-o") == 0 && argc > 2 &&
+      ((size_offsets = get_value((pcre_uint8 *)argv[op+1], &endptr)),
+        *endptr == 0))
+    {
+    op++;
+    argc--;
+    }
+  else if (strcmp(arg, "-t") == 0 || strcmp(arg, "-tm") == 0)
+    {
+    int both = arg[2] == 0;
+    int temp;
+    if (argc > 2 && (temp = get_value((pcre_uint8 *)argv[op+1], &endptr),
+                     *endptr == 0))
+      {
+      timeitm = temp;
+      op++;
+      argc--;
+      }
+    else timeitm = LOOPREPEAT;
+    if (both) timeit = timeitm;
+    }
+  else if (strcmp(arg, "-S") == 0 && argc > 2 &&
+      ((stack_size = get_value((pcre_uint8 *)argv[op+1], &endptr)),
+        *endptr == 0))
+    {
+#if defined(_WIN32) || defined(WIN32) || defined(__minix) || defined(NATIVE_ZOS) || defined(__VMS)
+    printf("PCRE: -S not supported on this OS\n");
+    exit(1);
+#else
+    int rc;
+    struct rlimit rlim;
+    getrlimit(RLIMIT_STACK, &rlim);
+    rlim.rlim_cur = stack_size * 1024 * 1024;
+    rc = setrlimit(RLIMIT_STACK, &rlim);
+    if (rc != 0)
+      {
+    printf("PCRE: setrlimit() failed with error %d\n", rc);
+    exit(1);
+      }
+    op++;
+    argc--;
+#endif
+    }
+#if !defined NOPOSIX
+  else if (strcmp(arg, "-p") == 0) posix = 1;
+#endif
+  else if (strcmp(arg, "-C") == 0)
+    {
+    int rc;
+    unsigned long int lrc;
+
+    if (argc > 2)
+      {
+      if (strcmp(argv[op + 1], "linksize") == 0)
+        {
+        (void)PCRE_CONFIG(PCRE_CONFIG_LINK_SIZE, &rc);
+        printf("%d\n", rc);
+        yield = rc;
+
+#ifdef __VMS
+        vms_setsymbol("LINKSIZE",0,yield );
+#endif
+        }
+      else if (strcmp(argv[op + 1], "pcre8") == 0)
+        {
+#ifdef SUPPORT_PCRE8
+        printf("1\n");
+        yield = 1;
+#else
+        printf("0\n");
+        yield = 0;
+#endif
+#ifdef __VMS
+        vms_setsymbol("PCRE8",0,yield );
+#endif
+        }
+      else if (strcmp(argv[op + 1], "pcre16") == 0)
+        {
+#ifdef SUPPORT_PCRE16
+        printf("1\n");
+        yield = 1;
+#else
+        printf("0\n");
+        yield = 0;
+#endif
+#ifdef __VMS
+        vms_setsymbol("PCRE16",0,yield );
+#endif
+        }
+      else if (strcmp(argv[op + 1], "pcre32") == 0)
+        {
+#ifdef SUPPORT_PCRE32
+        printf("1\n");
+        yield = 1;
+#else
+        printf("0\n");
+        yield = 0;
+#endif
+#ifdef __VMS
+        vms_setsymbol("PCRE32",0,yield );
+#endif
+        }
+      else if (strcmp(argv[op + 1], "utf") == 0)
+        {
+#ifdef SUPPORT_PCRE8
+        if (pcre_mode == PCRE8_MODE)
+          (void)pcre_config(PCRE_CONFIG_UTF8, &rc);
+#endif
+#ifdef SUPPORT_PCRE16
+        if (pcre_mode == PCRE16_MODE)
+          (void)pcre16_config(PCRE_CONFIG_UTF16, &rc);
+#endif
+#ifdef SUPPORT_PCRE32
+        if (pcre_mode == PCRE32_MODE)
+          (void)pcre32_config(PCRE_CONFIG_UTF32, &rc);
+#endif
+        printf("%d\n", rc);
+        yield = rc;
+#ifdef __VMS
+        vms_setsymbol("UTF",0,yield );
+#endif
+        }
+      else if (strcmp(argv[op + 1], "ucp") == 0)
+        {
+        (void)PCRE_CONFIG(PCRE_CONFIG_UNICODE_PROPERTIES, &rc);
+        printf("%d\n", rc);
+        yield = rc;
+        }
+      else if (strcmp(argv[op + 1], "jit") == 0)
+        {
+        (void)PCRE_CONFIG(PCRE_CONFIG_JIT, &rc);
+        printf("%d\n", rc);
+        yield = rc;
+        }
+      else if (strcmp(argv[op + 1], "newline") == 0)
+        {
+        (void)PCRE_CONFIG(PCRE_CONFIG_NEWLINE, &rc);
+        print_newline_config(rc, TRUE);
+        }
+      else if (strcmp(argv[op + 1], "ebcdic") == 0)
+        {
+#ifdef EBCDIC
+        printf("1\n");
+        yield = 1;
+#else
+        printf("0\n");
+#endif
+        }
+      else if (strcmp(argv[op + 1], "ebcdic-nl") == 0)
+        {
+#ifdef EBCDIC
+        printf("0x%02x\n", CHAR_LF);
+#else
+        printf("0\n");
+#endif
+        }
+      else
+        {
+        printf("Unknown -C option: %s\n", argv[op + 1]);
+        }
+      goto EXIT;
+      }
+
+    /* No argument for -C: output all configuration information. */
+
+    printf("PCRE version %s\n", version);
+    printf("Compiled with\n");
+
+#ifdef EBCDIC
+    printf("  EBCDIC code support: LF is 0x%02x\n", CHAR_LF);
+#endif
+
+/* At least one of SUPPORT_PCRE8 and SUPPORT_PCRE16 will be set. If both
+are set, either both UTFs are supported or both are not supported. */
+
+#ifdef SUPPORT_PCRE8
+    printf("  8-bit support\n");
+    (void)pcre_config(PCRE_CONFIG_UTF8, &rc);
+      printf ("  %sUTF-8 support\n", rc ? "" : "No ");
+#endif
+#ifdef SUPPORT_PCRE16
+    printf("  16-bit support\n");
+    (void)pcre16_config(PCRE_CONFIG_UTF16, &rc);
+    printf ("  %sUTF-16 support\n", rc ? "" : "No ");
+#endif
+#ifdef SUPPORT_PCRE32
+    printf("  32-bit support\n");
+    (void)pcre32_config(PCRE_CONFIG_UTF32, &rc);
+    printf ("  %sUTF-32 support\n", rc ? "" : "No ");
+#endif
+
+    (void)PCRE_CONFIG(PCRE_CONFIG_UNICODE_PROPERTIES, &rc);
+    printf("  %sUnicode properties support\n", rc? "" : "No ");
+    (void)PCRE_CONFIG(PCRE_CONFIG_JIT, &rc);
+    if (rc)
+      {
+      const char *arch;
+      (void)PCRE_CONFIG(PCRE_CONFIG_JITTARGET, (void *)(&arch));
+      printf("  Just-in-time compiler support: %s\n", arch);
+      }
+    else
+      printf("  No just-in-time compiler support\n");
+    (void)PCRE_CONFIG(PCRE_CONFIG_NEWLINE, &rc);
+    print_newline_config(rc, FALSE);
+    (void)PCRE_CONFIG(PCRE_CONFIG_BSR, &rc);
+    printf("  \\R matches %s\n", rc? "CR, LF, or CRLF only" :
+                                     "all Unicode newlines");
+    (void)PCRE_CONFIG(PCRE_CONFIG_LINK_SIZE, &rc);
+    printf("  Internal link size = %d\n", rc);
+    (void)PCRE_CONFIG(PCRE_CONFIG_POSIX_MALLOC_THRESHOLD, &rc);
+    printf("  POSIX malloc threshold = %d\n", rc);
+    (void)PCRE_CONFIG(PCRE_CONFIG_MATCH_LIMIT, &lrc);
+    printf("  Default match limit = %ld\n", lrc);
+    (void)PCRE_CONFIG(PCRE_CONFIG_MATCH_LIMIT_RECURSION, &lrc);
+    printf("  Default recursion depth limit = %ld\n", lrc);
+    (void)PCRE_CONFIG(PCRE_CONFIG_STACKRECURSE, &rc);
+    printf("  Match recursion uses %s", rc? "stack" : "heap");
+    if (showstore)
+      {
+      PCRE_EXEC(stack_size, NULL, NULL, NULL, -999, -999, 0, NULL, 0);
+      printf(": %sframe size = %d bytes", rc? "approximate " : "", -stack_size);
+      }
+    printf("\n");
+    goto EXIT;
+    }
+  else if (strcmp(arg, "-help") == 0 ||
+           strcmp(arg, "--help") == 0)
+    {
+    usage();
+    goto EXIT;
+    }
+  else
+    {
+    BAD_ARG:
+    printf("** Unknown or malformed option %s\n", arg);
+    usage();
+    yield = 1;
+    goto EXIT;
+    }
+  op++;
+  argc--;
+  }
+
+/* Get the store for the offsets vector, and remember what it was */
+
+size_offsets_max = size_offsets;
+offsets = (int *)malloc(size_offsets_max * sizeof(int));
+if (offsets == NULL)
+  {
+  printf("** Failed to get %d bytes of memory for offsets vector\n",
+    (int)(size_offsets_max * sizeof(int)));
+  yield = 1;
+  goto EXIT;
+  }
+
+/* Sort out the input and output files */
+
+if (argc > 1)
+  {
+  infile = fopen(argv[op], INPUT_MODE);
+  if (infile == NULL)
+    {
+    printf("** Failed to open %s\n", argv[op]);
+    yield = 1;
+    goto EXIT;
+    }
+  }
+
+if (argc > 2)
+  {
+  outfile = fopen(argv[op+1], OUTPUT_MODE);
+  if (outfile == NULL)
+    {
+    printf("** Failed to open %s\n", argv[op+1]);
+    yield = 1;
+    goto EXIT;
+    }
+  }
+
+/* Set alternative malloc function */
+
+#ifdef SUPPORT_PCRE8
+pcre_malloc = new_malloc;
+pcre_free = new_free;
+pcre_stack_malloc = stack_malloc;
+pcre_stack_free = stack_free;
+#endif
+
+#ifdef SUPPORT_PCRE16
+pcre16_malloc = new_malloc;
+pcre16_free = new_free;
+pcre16_stack_malloc = stack_malloc;
+pcre16_stack_free = stack_free;
+#endif
+
+#ifdef SUPPORT_PCRE32
+pcre32_malloc = new_malloc;
+pcre32_free = new_free;
+pcre32_stack_malloc = stack_malloc;
+pcre32_stack_free = stack_free;
+#endif
+
+/* Heading line unless quiet, then prompt for first regex if stdin */
+
+if (!quiet) fprintf(outfile, "PCRE version %s\n\n", version);
+
+/* Main loop */
+
+while (!done)
+  {
+  pcre *re = NULL;
+  pcre_extra *extra = NULL;
+
+#if !defined NOPOSIX  /* There are still compilers that require no indent */
+  regex_t preg;
+  int do_posix = 0;
+#endif
+
+  const char *error;
+  pcre_uint8 *markptr;
+  pcre_uint8 *p, *pp, *ppp;
+  pcre_uint8 *to_file = NULL;
+  const pcre_uint8 *tables = NULL;
+  unsigned long int get_options;
+  unsigned long int true_size, true_study_size = 0;
+  size_t size, regex_gotten_store;
+  int do_allcaps = 0;
+  int do_mark = 0;
+  int do_study = 0;
+  int no_force_study = 0;
+  int do_debug = debug;
+  int do_G = 0;
+  int do_g = 0;
+  int do_showinfo = showinfo;
+  int do_showrest = 0;
+  int do_showcaprest = 0;
+  int do_flip = 0;
+  int erroroffset, len, delimiter, poffset;
+
+#if !defined NODFA
+  int dfa_matched = 0;
+#endif
+
+  use_utf = 0;
+  debug_lengths = 1;
+
+  if (extend_inputline(infile, buffer, "  re> ") == NULL) break;
+  if (infile != stdin) fprintf(outfile, "%s", (char *)buffer);
+  fflush(outfile);
+
+  p = buffer;
+  while (isspace(*p)) p++;
+  if (*p == 0) continue;
+
+  /* See if the pattern is to be loaded pre-compiled from a file. */
+
+  if (*p == '<' && strchr((char *)(p+1), '<') == NULL)
+    {
+    pcre_uint32 magic;
+    pcre_uint8 sbuf[8];
+    FILE *f;
+
+    p++;
+    if (*p == '!')
+      {
+      do_debug = TRUE;
+      do_showinfo = TRUE;
+      p++;
+      }
+
+    pp = p + (int)strlen((char *)p);
+    while (isspace(pp[-1])) pp--;
+    *pp = 0;
+
+    f = fopen((char *)p, "rb");
+    if (f == NULL)
+      {
+      fprintf(outfile, "Failed to open %s: %s\n", p, strerror(errno));
+      continue;
+      }
+
+    first_gotten_store = 0;
+    if (fread(sbuf, 1, 8, f) != 8) goto FAIL_READ;
+
+    true_size =
+      (sbuf[0] << 24) | (sbuf[1] << 16) | (sbuf[2] << 8) | sbuf[3];
+    true_study_size =
+      (sbuf[4] << 24) | (sbuf[5] << 16) | (sbuf[6] << 8) | sbuf[7];
+
+    re = (pcre *)new_malloc(true_size);
+    if (re == NULL)
+      {
+      printf("** Failed to get %d bytes of memory for pcre object\n",
+        (int)true_size);
+      yield = 1;
+      goto EXIT;
+      }
+    regex_gotten_store = first_gotten_store;
+
+    if (fread(re, 1, true_size, f) != true_size) goto FAIL_READ;
+
+    magic = REAL_PCRE_MAGIC(re);
+    if (magic != MAGIC_NUMBER)
+      {
+      if (swap_uint32(magic) == MAGIC_NUMBER)
+        {
+        do_flip = 1;
+        }
+      else
+        {
+        fprintf(outfile, "Data in %s is not a compiled PCRE regex\n", p);
+        new_free(re);
+        fclose(f);
+        continue;
+        }
+      }
+
+    /* We hide the byte-invert info for little and big endian tests. */
+    fprintf(outfile, "Compiled pattern%s loaded from %s\n",
+      do_flip && (p[-1] == '<') ? " (byte-inverted)" : "", p);
+
+    /* Now see if there is any following study data. */
+
+    if (true_study_size != 0)
+      {
+      pcre_study_data *psd;
+
+      extra = (pcre_extra *)new_malloc(sizeof(pcre_extra) + true_study_size);
+      extra->flags = PCRE_EXTRA_STUDY_DATA;
+
+      psd = (pcre_study_data *)(((char *)extra) + sizeof(pcre_extra));
+      extra->study_data = psd;
+
+      if (fread(psd, 1, true_study_size, f) != true_study_size)
+        {
+        FAIL_READ:
+        fprintf(outfile, "Failed to read data from %s\n", p);
+        if (extra != NULL)
+          {
+          PCRE_FREE_STUDY(extra);
+          }
+        new_free(re);
+        fclose(f);
+        continue;
+        }
+      fprintf(outfile, "Study data loaded from %s\n", p);
+      do_study = 1;     /* To get the data output if requested */
+      }
+    else fprintf(outfile, "No study data\n");
+
+    /* Flip the necessary bytes. */
+    if (do_flip)
+      {
+      int rc;
+      PCRE_PATTERN_TO_HOST_BYTE_ORDER(rc, re, extra, NULL);
+      if (rc == PCRE_ERROR_BADMODE)
+        {
+        pcre_uint32 flags_in_host_byte_order;
+        if (REAL_PCRE_MAGIC(re) == MAGIC_NUMBER)
+          flags_in_host_byte_order = REAL_PCRE_FLAGS(re);
+        else
+          flags_in_host_byte_order = swap_uint32(REAL_PCRE_FLAGS(re));
+        /* Simulate the result of the function call below. */
+        fprintf(outfile, "Error %d from pcre%s_fullinfo(%d)\n", rc,
+          pcre_mode == PCRE32_MODE ? "32" : pcre_mode == PCRE16_MODE ? "16" : "",
+          PCRE_INFO_OPTIONS);
+        fprintf(outfile, "Running in %d-bit mode but pattern was compiled in "
+          "%d-bit mode\n", 8 * CHAR_SIZE, 8 * (flags_in_host_byte_order & PCRE_MODE_MASK));
+        new_free(re);
+        fclose(f);
+        continue;
+        }
+      }
+
+    /* Need to know if UTF-8 for printing data strings. */
+
+    if (new_info(re, NULL, PCRE_INFO_OPTIONS, &get_options) < 0)
+      {
+      new_free(re);
+      fclose(f);
+      continue;
+      }
+    use_utf = (get_options & PCRE_UTF8) != 0;
+
+    fclose(f);
+    goto SHOW_INFO;
+    }
+
+  /* In-line pattern (the usual case). Get the delimiter and seek the end of
+  the pattern; if it isn't complete, read more. */
+
+  delimiter = *p++;
+
+  if (isalnum(delimiter) || delimiter == '\\')
+    {
+    fprintf(outfile, "** Delimiter must not be alphanumeric or \\\n");
+    goto SKIP_DATA;
+    }
+
+  pp = p;
+  poffset = (int)(p - buffer);
+
+  for(;;)
+    {
+    while (*pp != 0)
+      {
+      if (*pp == '\\' && pp[1] != 0) pp++;
+        else if (*pp == delimiter) break;
+      pp++;
+      }
+    if (*pp != 0) break;
+    if ((pp = extend_inputline(infile, pp, "    > ")) == NULL)
+      {
+      fprintf(outfile, "** Unexpected EOF\n");
+      done = 1;
+      goto CONTINUE;
+      }
+    if (infile != stdin) fprintf(outfile, "%s", (char *)pp);
+    }
+
+  /* The buffer may have moved while being extended; reset the start of data
+  pointer to the correct relative point in the buffer. */
+
+  p = buffer + poffset;
+
+  /* If the first character after the delimiter is backslash, make
+  the pattern end with backslash. This is purely to provide a way
+  of testing for the error message when a pattern ends with backslash. */
+
+  if (pp[1] == '\\') *pp++ = '\\';
+
+  /* Terminate the pattern at the delimiter, and save a copy of the pattern
+  for callouts. */
+
+  *pp++ = 0;
+  strcpy((char *)pbuffer, (char *)p);
+
+  /* Look for options after final delimiter */
+
+  options = 0;
+  study_options = force_study_options;
+  log_store = showstore;  /* default from command line */
+
+  while (*pp != 0)
+    {
+    switch (*pp++)
+      {
+      case 'f': options |= PCRE_FIRSTLINE; break;
+      case 'g': do_g = 1; break;
+      case 'i': options |= PCRE_CASELESS; break;
+      case 'm': options |= PCRE_MULTILINE; break;
+      case 's': options |= PCRE_DOTALL; break;
+      case 'x': options |= PCRE_EXTENDED; break;
+
+      case '+':
+      if (do_showrest) do_showcaprest = 1; else do_showrest = 1;
+      break;
+
+      case '=': do_allcaps = 1; break;
+      case 'A': options |= PCRE_ANCHORED; break;
+      case 'B': do_debug = 1; break;
+      case 'C': options |= PCRE_AUTO_CALLOUT; break;
+      case 'D': do_debug = do_showinfo = 1; break;
+      case 'E': options |= PCRE_DOLLAR_ENDONLY; break;
+      case 'F': do_flip = 1; break;
+      case 'G': do_G = 1; break;
+      case 'I': do_showinfo = 1; break;
+      case 'J': options |= PCRE_DUPNAMES; break;
+      case 'K': do_mark = 1; break;
+      case 'M': log_store = 1; break;
+      case 'N': options |= PCRE_NO_AUTO_CAPTURE; break;
+
+#if !defined NOPOSIX
+      case 'P': do_posix = 1; break;
+#endif
+
+      case 'S':
+      do_study = 1;
+      for (;;)
+        {
+        switch (*pp++)
+          {
+          case 'S':
+          do_study = 0;
+          no_force_study = 1;
+          break;
+
+          case '!':
+          study_options |= PCRE_STUDY_EXTRA_NEEDED;
+          break;
+
+          case '+':
+          if (*pp == '+')
+            {
+            verify_jit = TRUE;
+            pp++;
+            }
+          if (*pp >= '1' && *pp <= '7')
+            study_options |= jit_study_bits[*pp++ - '1'];
+          else
+            study_options |= jit_study_bits[6];
+          break;
+
+          case '-':
+          study_options &= ~PCRE_STUDY_ALLJIT;
+          break;
+
+          default:
+          pp--;
+          goto ENDLOOP;
+          }
+        }
+      ENDLOOP:
+      break;
+
+      case 'U': options |= PCRE_UNGREEDY; break;
+      case 'W': options |= PCRE_UCP; break;
+      case 'X': options |= PCRE_EXTRA; break;
+      case 'Y': options |= PCRE_NO_START_OPTIMISE; break;
+      case 'Z': debug_lengths = 0; break;
+      case '8': options |= PCRE_UTF8; use_utf = 1; break;
+      case '9': options |= PCRE_NEVER_UTF; break;
+      case '?': options |= PCRE_NO_UTF8_CHECK; break;
+
+      case 'T':
+      switch (*pp++)
+        {
+        case '0': tables = tables0; break;
+        case '1': tables = tables1; break;
+
+        case '\r':
+        case '\n':
+        case ' ':
+        case 0:
+        fprintf(outfile, "** Missing table number after /T\n");
+        goto SKIP_DATA;
+
+        default:
+        fprintf(outfile, "** Bad table number \"%c\" after /T\n", pp[-1]);
+        goto SKIP_DATA;
+        }
+      break;
+
+      case 'L':
+      ppp = pp;
+      /* The '\r' test here is so that it works on Windows. */
+      /* The '0' test is just in case this is an unterminated line. */
+      while (*ppp != 0 && *ppp != '\n' && *ppp != '\r' && *ppp != ' ') ppp++;
+      *ppp = 0;
+      if (setlocale(LC_CTYPE, (const char *)pp) == NULL)
+        {
+        fprintf(outfile, "** Failed to set locale \"%s\"\n", pp);
+        goto SKIP_DATA;
+        }
+      locale_set = 1;
+      tables = PCRE_MAKETABLES;
+      pp = ppp;
+      break;
+
+      case '>':
+      to_file = pp;
+      while (*pp != 0) pp++;
+      while (isspace(pp[-1])) pp--;
+      *pp = 0;
+      break;
+
+      case '<':
+        {
+        if (strncmpic(pp, (pcre_uint8 *)"JS>", 3) == 0)
+          {
+          options |= PCRE_JAVASCRIPT_COMPAT;
+          pp += 3;
+          }
+        else
+          {
+          int x = check_newline(pp, outfile);
+          if (x == 0) goto SKIP_DATA;
+          options |= x;
+          while (*pp++ != '>');
+          }
+        }
+      break;
+
+      case '\r':                      /* So that it works in Windows */
+      case '\n':
+      case ' ':
+      break;
+
+      default:
+      fprintf(outfile, "** Unknown option '%c'\n", pp[-1]);
+      goto SKIP_DATA;
+      }
+    }
+
+  /* Handle compiling via the POSIX interface, which doesn't support the
+  timing, showing, or debugging options, nor the ability to pass over
+  local character tables. Neither does it have 16-bit support. */
+
+#if !defined NOPOSIX
+  if (posix || do_posix)
+    {
+    int rc;
+    int cflags = 0;
+
+    if ((options & PCRE_CASELESS) != 0) cflags |= REG_ICASE;
+    if ((options & PCRE_MULTILINE) != 0) cflags |= REG_NEWLINE;
+    if ((options & PCRE_DOTALL) != 0) cflags |= REG_DOTALL;
+    if ((options & PCRE_NO_AUTO_CAPTURE) != 0) cflags |= REG_NOSUB;
+    if ((options & PCRE_UTF8) != 0) cflags |= REG_UTF8;
+    if ((options & PCRE_UCP) != 0) cflags |= REG_UCP;
+    if ((options & PCRE_UNGREEDY) != 0) cflags |= REG_UNGREEDY;
+
+    first_gotten_store = 0;
+    rc = regcomp(&preg, (char *)p, cflags);
+
+    /* Compilation failed; go back for another re, skipping to blank line
+    if non-interactive. */
+
+    if (rc != 0)
+      {
+      (void)regerror(rc, &preg, (char *)buffer, buffer_size);
+      fprintf(outfile, "Failed: POSIX code %d: %s\n", rc, buffer);
+      goto SKIP_DATA;
+      }
+    }
+
+  /* Handle compiling via the native interface */
+
+  else
+#endif  /* !defined NOPOSIX */
+
+    {
+    /* In 16- or 32-bit mode, convert the input. */
+
+#ifdef SUPPORT_PCRE16
+    if (pcre_mode == PCRE16_MODE)
+      {
+      switch(to16(FALSE, p, options & PCRE_UTF8, (int)strlen((char *)p)))
+        {
+        case -1:
+        fprintf(outfile, "**Failed: invalid UTF-8 string cannot be "
+          "converted to UTF-16\n");
+        goto SKIP_DATA;
+
+        case -2:
+        fprintf(outfile, "**Failed: character value greater than 0x10ffff "
+          "cannot be converted to UTF-16\n");
+        goto SKIP_DATA;
+
+        case -3: /* "Impossible error" when to16 is called arg1 FALSE */
+        fprintf(outfile, "**Failed: character value greater than 0xffff "
+          "cannot be converted to 16-bit in non-UTF mode\n");
+        goto SKIP_DATA;
+
+        default:
+        break;
+        }
+      p = (pcre_uint8 *)buffer16;
+      }
+#endif
+
+#ifdef SUPPORT_PCRE32
+    if (pcre_mode == PCRE32_MODE)
+      {
+      switch(to32(FALSE, p, options & PCRE_UTF32, (int)strlen((char *)p)))
+        {
+        case -1:
+        fprintf(outfile, "**Failed: invalid UTF-8 string cannot be "
+          "converted to UTF-32\n");
+        goto SKIP_DATA;
+
+        case -2:
+        fprintf(outfile, "**Failed: character value greater than 0x10ffff "
+          "cannot be converted to UTF-32\n");
+        goto SKIP_DATA;
+
+        case -3:
+        fprintf(outfile, "**Failed: character value is ill-formed UTF-32\n");
+        goto SKIP_DATA;
+
+        default:
+        break;
+        }
+      p = (pcre_uint8 *)buffer32;
+      }
+#endif
+
+    /* Compile many times when timing */
+
+    if (timeit > 0)
+      {
+      register int i;
+      clock_t time_taken;
+      clock_t start_time = clock();
+      for (i = 0; i < timeit; i++)
+        {
+        PCRE_COMPILE(re, p, options, &error, &erroroffset, tables);
+        if (re != NULL) free(re);
+        }
+      time_taken = clock() - start_time;
+      fprintf(outfile, "Compile time %.4f milliseconds\n",
+        (((double)time_taken * 1000.0) / (double)timeit) /
+          (double)CLOCKS_PER_SEC);
+      }
+
+    first_gotten_store = 0;
+    PCRE_COMPILE(re, p, options, &error, &erroroffset, tables);
+
+    /* Compilation failed; go back for another re, skipping to blank line
+    if non-interactive. */
+
+    if (re == NULL)
+      {
+      fprintf(outfile, "Failed: %s at offset %d\n", error, erroroffset);
+      SKIP_DATA:
+      if (infile != stdin)
+        {
+        for (;;)
+          {
+          if (extend_inputline(infile, buffer, NULL) == NULL)
+            {
+            done = 1;
+            goto CONTINUE;
+            }
+          len = (int)strlen((char *)buffer);
+          while (len > 0 && isspace(buffer[len-1])) len--;
+          if (len == 0) break;
+          }
+        fprintf(outfile, "\n");
+        }
+      goto CONTINUE;
+      }
+
+    /* Compilation succeeded. It is now possible to set the UTF-8 option from
+    within the regex; check for this so that we know how to process the data
+    lines. */
+
+    if (new_info(re, NULL, PCRE_INFO_OPTIONS, &get_options) < 0)
+      goto SKIP_DATA;
+    if ((get_options & PCRE_UTF8) != 0) use_utf = 1;
+
+    /* Extract the size for possible writing before possibly flipping it,
+    and remember the store that was got. */
+
+    true_size = REAL_PCRE_SIZE(re);
+    regex_gotten_store = first_gotten_store;
+
+    /* Output code size information if requested */
+
+    if (log_store)
+      {
+      int name_count, name_entry_size, real_pcre_size;
+
+      new_info(re, NULL, PCRE_INFO_NAMECOUNT, &name_count);
+      new_info(re, NULL, PCRE_INFO_NAMEENTRYSIZE, &name_entry_size);
+      real_pcre_size = 0;
+#ifdef SUPPORT_PCRE8
+      if (REAL_PCRE_FLAGS(re) & PCRE_MODE8)
+        real_pcre_size = sizeof(real_pcre);
+#endif
+#ifdef SUPPORT_PCRE16
+      if (REAL_PCRE_FLAGS(re) & PCRE_MODE16)
+        real_pcre_size = sizeof(real_pcre16);
+#endif
+#ifdef SUPPORT_PCRE32
+      if (REAL_PCRE_FLAGS(re) & PCRE_MODE32)
+        real_pcre_size = sizeof(real_pcre32);
+#endif
+      fprintf(outfile, "Memory allocation (code space): %d\n",
+        (int)(first_gotten_store - real_pcre_size - name_count * name_entry_size));
+      }
+
+    /* If -s or /S was present, study the regex to generate additional info to
+    help with the matching, unless the pattern has the SS option, which
+    suppresses the effect of /S (used for a few test patterns where studying is
+    never sensible). */
+
+    if (do_study || (force_study >= 0 && !no_force_study))
+      {
+      if (timeit > 0)
+        {
+        register int i;
+        clock_t time_taken;
+        clock_t start_time = clock();
+        for (i = 0; i < timeit; i++)
+          {
+          PCRE_STUDY(extra, re, study_options, &error);
+          }
+        time_taken = clock() - start_time;
+        if (extra != NULL)
+          {
+          PCRE_FREE_STUDY(extra);
+          }
+        fprintf(outfile, "  Study time %.4f milliseconds\n",
+          (((double)time_taken * 1000.0) / (double)timeit) /
+            (double)CLOCKS_PER_SEC);
+        }
+      PCRE_STUDY(extra, re, study_options, &error);
+      if (error != NULL)
+        fprintf(outfile, "Failed to study: %s\n", error);
+      else if (extra != NULL)
+        {
+        true_study_size = ((pcre_study_data *)(extra->study_data))->size;
+        if (log_store)
+          {
+          size_t jitsize;
+          if (new_info(re, extra, PCRE_INFO_JITSIZE, &jitsize) == 0 &&
+              jitsize != 0)
+            fprintf(outfile, "Memory allocation (JIT code): %d\n", (int)jitsize);
+          }
+        }
+      }
+
+    /* If /K was present, we set up for handling MARK data. */
+
+    if (do_mark)
+      {
+      if (extra == NULL)
+        {
+        extra = (pcre_extra *)malloc(sizeof(pcre_extra));
+        extra->flags = 0;
+        }
+      extra->mark = &markptr;
+      extra->flags |= PCRE_EXTRA_MARK;
+      }
+
+    /* Extract and display information from the compiled data if required. */
+
+    SHOW_INFO:
+
+    if (do_debug)
+      {
+      fprintf(outfile, "------------------------------------------------------------------\n");
+      PCRE_PRINTINT(re, outfile, debug_lengths);
+      }
+
+    /* We already have the options in get_options (see above) */
+
+    if (do_showinfo)
+      {
+      unsigned long int all_options;
+      pcre_uint32 first_char, need_char;
+      pcre_uint32 match_limit, recursion_limit;
+      int count, backrefmax, first_char_set, need_char_set, okpartial, jchanged,
+        hascrorlf, maxlookbehind;
+      int nameentrysize, namecount;
+      const pcre_uint8 *nametable;
+
+      if (new_info(re, NULL, PCRE_INFO_SIZE, &size) +
+          new_info(re, NULL, PCRE_INFO_CAPTURECOUNT, &count) +
+          new_info(re, NULL, PCRE_INFO_BACKREFMAX, &backrefmax) +
+          new_info(re, NULL, PCRE_INFO_FIRSTCHARACTER, &first_char) +
+          new_info(re, NULL, PCRE_INFO_FIRSTCHARACTERFLAGS, &first_char_set) +
+          new_info(re, NULL, PCRE_INFO_REQUIREDCHAR, &need_char) +
+          new_info(re, NULL, PCRE_INFO_REQUIREDCHARFLAGS, &need_char_set) +
+          new_info(re, NULL, PCRE_INFO_NAMEENTRYSIZE, &nameentrysize) +
+          new_info(re, NULL, PCRE_INFO_NAMECOUNT, &namecount) +
+          new_info(re, NULL, PCRE_INFO_NAMETABLE, (void *)&nametable) +
+          new_info(re, NULL, PCRE_INFO_OKPARTIAL, &okpartial) +
+          new_info(re, NULL, PCRE_INFO_JCHANGED, &jchanged) +
+          new_info(re, NULL, PCRE_INFO_HASCRORLF, &hascrorlf) +
+          new_info(re, NULL, PCRE_INFO_MAXLOOKBEHIND, &maxlookbehind)
+          != 0)
+        goto SKIP_DATA;
+
+      if (size != regex_gotten_store) fprintf(outfile,
+        "Size disagreement: pcre_fullinfo=%d call to malloc for %d\n",
+        (int)size, (int)regex_gotten_store);
+
+      fprintf(outfile, "Capturing subpattern count = %d\n", count);
+
+      if (backrefmax > 0)
+        fprintf(outfile, "Max back reference = %d\n", backrefmax);
+
+      if (maxlookbehind > 0)
+        fprintf(outfile, "Max lookbehind = %d\n", maxlookbehind);
+
+      if (new_info(re, NULL, PCRE_INFO_MATCHLIMIT, &match_limit) == 0)
+        fprintf(outfile, "Match limit = %u\n", match_limit);
+
+      if (new_info(re, NULL, PCRE_INFO_RECURSIONLIMIT, &recursion_limit) == 0)
+        fprintf(outfile, "Recursion limit = %u\n", recursion_limit);
+
+      if (namecount > 0)
+        {
+        fprintf(outfile, "Named capturing subpatterns:\n");
+        while (namecount-- > 0)
+          {
+          int imm2_size = pcre_mode == PCRE8_MODE ? 2 : 1;
+          int length = (int)STRLEN(nametable + imm2_size);
+          fprintf(outfile, "  ");
+          PCHARSV(nametable, imm2_size, length, outfile);
+          while (length++ < nameentrysize - imm2_size) putc(' ', outfile);
+#ifdef SUPPORT_PCRE32
+          if (pcre_mode == PCRE32_MODE)
+            fprintf(outfile, "%3d\n", (int)(((PCRE_SPTR32)nametable)[0]));
+#endif
+#ifdef SUPPORT_PCRE16
+          if (pcre_mode == PCRE16_MODE)
+            fprintf(outfile, "%3d\n", (int)(((PCRE_SPTR16)nametable)[0]));
+#endif
+#ifdef SUPPORT_PCRE8
+          if (pcre_mode == PCRE8_MODE)
+            fprintf(outfile, "%3d\n", ((int)nametable[0] << 8) | (int)nametable[1]);
+#endif
+          nametable += nameentrysize * CHAR_SIZE;
+          }
+        }
+
+      if (!okpartial) fprintf(outfile, "Partial matching not supported\n");
+      if (hascrorlf) fprintf(outfile, "Contains explicit CR or LF match\n");
+
+      all_options = REAL_PCRE_OPTIONS(re);
+      if (do_flip) all_options = swap_uint32(all_options);
+
+      if (get_options == 0) fprintf(outfile, "No options\n");
+        else fprintf(outfile, "Options:%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s\n",
+          ((get_options & PCRE_ANCHORED) != 0)? " anchored" : "",
+          ((get_options & PCRE_CASELESS) != 0)? " caseless" : "",
+          ((get_options & PCRE_EXTENDED) != 0)? " extended" : "",
+          ((get_options & PCRE_MULTILINE) != 0)? " multiline" : "",
+          ((get_options & PCRE_FIRSTLINE) != 0)? " firstline" : "",
+          ((get_options & PCRE_DOTALL) != 0)? " dotall" : "",
+          ((get_options & PCRE_BSR_ANYCRLF) != 0)? " bsr_anycrlf" : "",
+          ((get_options & PCRE_BSR_UNICODE) != 0)? " bsr_unicode" : "",
+          ((get_options & PCRE_DOLLAR_ENDONLY) != 0)? " dollar_endonly" : "",
+          ((get_options & PCRE_EXTRA) != 0)? " extra" : "",
+          ((get_options & PCRE_UNGREEDY) != 0)? " ungreedy" : "",
+          ((get_options & PCRE_NO_AUTO_CAPTURE) != 0)? " no_auto_capture" : "",
+          ((get_options & PCRE_UTF8) != 0)? " utf" : "",
+          ((get_options & PCRE_UCP) != 0)? " ucp" : "",
+          ((get_options & PCRE_NO_UTF8_CHECK) != 0)? " no_utf_check" : "",
+          ((get_options & PCRE_NO_START_OPTIMIZE) != 0)? " no_start_optimize" : "",
+          ((get_options & PCRE_DUPNAMES) != 0)? " dupnames" : "",
+          ((get_options & PCRE_NEVER_UTF) != 0)? " never_utf" : "");
+
+      if (jchanged) fprintf(outfile, "Duplicate name status changes\n");
+
+      switch (get_options & PCRE_NEWLINE_BITS)
+        {
+        case PCRE_NEWLINE_CR:
+        fprintf(outfile, "Forced newline sequence: CR\n");
+        break;
+
+        case PCRE_NEWLINE_LF:
+        fprintf(outfile, "Forced newline sequence: LF\n");
+        break;
+
+        case PCRE_NEWLINE_CRLF:
+        fprintf(outfile, "Forced newline sequence: CRLF\n");
+        break;
+
+        case PCRE_NEWLINE_ANYCRLF:
+        fprintf(outfile, "Forced newline sequence: ANYCRLF\n");
+        break;
+
+        case PCRE_NEWLINE_ANY:
+        fprintf(outfile, "Forced newline sequence: ANY\n");
+        break;
+
+        default:
+        break;
+        }
+
+      if (first_char_set == 2)
+        {
+        fprintf(outfile, "First char at start or follows newline\n");
+        }
+      else if (first_char_set == 1)
+        {
+        const char *caseless =
+          ((REAL_PCRE_FLAGS(re) & PCRE_FCH_CASELESS) == 0)?
+          "" : " (caseless)";
+
+        if (PRINTOK(first_char))
+          fprintf(outfile, "First char = \'%c\'%s\n", first_char, caseless);
+        else
+          {
+          fprintf(outfile, "First char = ");
+          pchar(first_char, outfile);
+          fprintf(outfile, "%s\n", caseless);
+          }
+        }
+      else
+        {
+        fprintf(outfile, "No first char\n");
+        }
+
+      if (need_char_set == 0)
+        {
+        fprintf(outfile, "No need char\n");
+        }
+      else
+        {
+        const char *caseless =
+          ((REAL_PCRE_FLAGS(re) & PCRE_RCH_CASELESS) == 0)?
+          "" : " (caseless)";
+
+        if (PRINTOK(need_char))
+          fprintf(outfile, "Need char = \'%c\'%s\n", need_char, caseless);
+        else
+          {
+          fprintf(outfile, "Need char = ");
+          pchar(need_char, outfile);
+          fprintf(outfile, "%s\n", caseless);
+          }
+        }
+
+      /* Don't output study size; at present it is in any case a fixed
+      value, but it varies, depending on the computer architecture, and
+      so messes up the test suite. (And with the /F option, it might be
+      flipped.) If study was forced by an external -s, don't show this
+      information unless -i or -d was also present. This means that, except
+      when auto-callouts are involved, the output from runs with and without
+      -s should be identical. */
+
+      if (do_study || (force_study >= 0 && showinfo && !no_force_study))
+        {
+        if (extra == NULL)
+          fprintf(outfile, "Study returned NULL\n");
+        else
+          {
+          pcre_uint8 *start_bits = NULL;
+          int minlength;
+
+          if (new_info(re, extra, PCRE_INFO_MINLENGTH, &minlength) == 0)
+            fprintf(outfile, "Subject length lower bound = %d\n", minlength);
+
+          if (new_info(re, extra, PCRE_INFO_FIRSTTABLE, &start_bits) == 0)
+            {
+            if (start_bits == NULL)
+              fprintf(outfile, "No set of starting bytes\n");
+            else
+              {
+              int i;
+              int c = 24;
+              fprintf(outfile, "Starting byte set: ");
+              for (i = 0; i < 256; i++)
+                {
+                if ((start_bits[i/8] & (1<<(i&7))) != 0)
+                  {
+                  if (c > 75)
+                    {
+                    fprintf(outfile, "\n  ");
+                    c = 2;
+                    }
+                  if (PRINTOK(i) && i != ' ')
+                    {
+                    fprintf(outfile, "%c ", i);
+                    c += 2;
+                    }
+                  else
+                    {
+                    fprintf(outfile, "\\x%02x ", i);
+                    c += 5;
+                    }
+                  }
+                }
+              fprintf(outfile, "\n");
+              }
+            }
+          }
+
+        /* Show this only if the JIT was set by /S, not by -s. */
+
+        if ((study_options & PCRE_STUDY_ALLJIT) != 0 &&
+            (force_study_options & PCRE_STUDY_ALLJIT) == 0)
+          {
+          int jit;
+          if (new_info(re, extra, PCRE_INFO_JIT, &jit) == 0)
+            {
+            if (jit)
+              fprintf(outfile, "JIT study was successful\n");
+            else
+#ifdef SUPPORT_JIT
+              fprintf(outfile, "JIT study was not successful\n");
+#else
+              fprintf(outfile, "JIT support is not available in this version of PCRE\n");
+#endif
+            }
+          }
+        }
+      }
+
+    /* If the '>' option was present, we write out the regex to a file, and
+    that is all. The first 8 bytes of the file are the regex length and then
+    the study length, in big-endian order. */
+
+    if (to_file != NULL)
+      {
+      FILE *f = fopen((char *)to_file, "wb");
+      if (f == NULL)
+        {
+        fprintf(outfile, "Unable to open %s: %s\n", to_file, strerror(errno));
+        }
+      else
+        {
+        pcre_uint8 sbuf[8];
+
+        if (do_flip) regexflip(re, extra);
+        sbuf[0] = (pcre_uint8)((true_size >> 24) & 255);
+        sbuf[1] = (pcre_uint8)((true_size >> 16) & 255);
+        sbuf[2] = (pcre_uint8)((true_size >>  8) & 255);
+        sbuf[3] = (pcre_uint8)((true_size) & 255);
+        sbuf[4] = (pcre_uint8)((true_study_size >> 24) & 255);
+        sbuf[5] = (pcre_uint8)((true_study_size >> 16) & 255);
+        sbuf[6] = (pcre_uint8)((true_study_size >>  8) & 255);
+        sbuf[7] = (pcre_uint8)((true_study_size) & 255);
+
+        if (fwrite(sbuf, 1, 8, f) < 8 ||
+            fwrite(re, 1, true_size, f) < true_size)
+          {
+          fprintf(outfile, "Write error on %s: %s\n", to_file, strerror(errno));
+          }
+        else
+          {
+          fprintf(outfile, "Compiled pattern written to %s\n", to_file);
+
+          /* If there is study data, write it. */
+
+          if (extra != NULL)
+            {
+            if (fwrite(extra->study_data, 1, true_study_size, f) <
+                true_study_size)
+              {
+              fprintf(outfile, "Write error on %s: %s\n", to_file,
+                strerror(errno));
+              }
+            else fprintf(outfile, "Study data written to %s\n", to_file);
+            }
+          }
+        fclose(f);
+        }
+
+      new_free(re);
+      if (extra != NULL)
+        {
+        PCRE_FREE_STUDY(extra);
+        }
+      if (locale_set)
+        {
+        new_free((void *)tables);
+        setlocale(LC_CTYPE, "C");
+        locale_set = 0;
+        }
+      continue;  /* With next regex */
+      }
+    }        /* End of non-POSIX compile */
+
+  /* Read data lines and test them */
+
+  for (;;)
+    {
+#ifdef SUPPORT_PCRE8
+    pcre_uint8 *q8;
+#endif
+#ifdef SUPPORT_PCRE16
+    pcre_uint16 *q16;
+#endif
+#ifdef SUPPORT_PCRE32
+    pcre_uint32 *q32;
+#endif
+    pcre_uint8 *bptr;
+    int *use_offsets = offsets;
+    int use_size_offsets = size_offsets;
+    int callout_data = 0;
+    int callout_data_set = 0;
+    int count;
+    pcre_uint32 c;
+    int copystrings = 0;
+    int find_match_limit = default_find_match_limit;
+    int getstrings = 0;
+    int getlist = 0;
+    int gmatched = 0;
+    int start_offset = 0;
+    int start_offset_sign = 1;
+    int g_notempty = 0;
+    int use_dfa = 0;
+
+    *copynames = 0;
+    *getnames = 0;
+
+#ifdef SUPPORT_PCRE32
+    cn32ptr = copynames;
+    gn32ptr = getnames;
+#endif
+#ifdef SUPPORT_PCRE16
+    cn16ptr = copynames16;
+    gn16ptr = getnames16;
+#endif
+#ifdef SUPPORT_PCRE8
+    cn8ptr = copynames8;
+    gn8ptr = getnames8;
+#endif
+
+    SET_PCRE_CALLOUT(callout);
+    first_callout = 1;
+    last_callout_mark = NULL;
+    callout_extra = 0;
+    callout_count = 0;
+    callout_fail_count = 999999;
+    callout_fail_id = -1;
+    show_malloc = 0;
+    options = 0;
+
+    if (extra != NULL) extra->flags &=
+      ~(PCRE_EXTRA_MATCH_LIMIT|PCRE_EXTRA_MATCH_LIMIT_RECURSION);
+
+    len = 0;
+    for (;;)
+      {
+      if (extend_inputline(infile, buffer + len, "data> ") == NULL)
+        {
+        if (len > 0)    /* Reached EOF without hitting a newline */
+          {
+          fprintf(outfile, "\n");
+          break;
+          }
+        done = 1;
+        goto CONTINUE;
+        }
+      if (infile != stdin) fprintf(outfile, "%s", (char *)buffer);
+      len = (int)strlen((char *)buffer);
+      if (buffer[len-1] == '\n') break;
+      }
+
+    while (len > 0 && isspace(buffer[len-1])) len--;
+    buffer[len] = 0;
+    if (len == 0) break;
+
+    p = buffer;
+    while (isspace(*p)) p++;
+
+#ifndef NOUTF
+    /* Check that the data is well-formed UTF-8 if we're in UTF mode. To create
+    invalid input to pcre_exec, you must use \x?? or \x{} sequences. */
+
+    if (use_utf)
+      {
+      pcre_uint8 *q;
+      pcre_uint32 cc;
+      int n = 1;
+
+      for (q = p; n > 0 && *q; q += n) n = utf82ord(q, &cc);
+      if (n <= 0)
+        {
+        fprintf(outfile, "**Failed: invalid UTF-8 string cannot be used as input in UTF mode\n");
+        goto NEXT_DATA;
+        }
+      }
+#endif
+
+#ifdef SUPPORT_VALGRIND
+    /* Mark the dbuffer as addressable but undefined again. */
+
+    if (dbuffer != NULL)
+      {
+      VALGRIND_MAKE_MEM_UNDEFINED(dbuffer, dbuffer_size * CHAR_SIZE);
+      }
+#endif
+
+    /* Allocate a buffer to hold the data line; len+1 is an upper bound on
+    the number of pcre_uchar units that will be needed. */
+
+    while (dbuffer == NULL || (size_t)len >= dbuffer_size)
+      {
+      dbuffer_size *= 2;
+      dbuffer = (pcre_uint8 *)realloc(dbuffer, dbuffer_size * CHAR_SIZE);
+      if (dbuffer == NULL)
+        {
+        fprintf(stderr, "pcretest: realloc(%d) failed\n", (int)dbuffer_size);
+        exit(1);
+        }
+      }
+
+#ifdef SUPPORT_PCRE8
+    q8 = (pcre_uint8 *) dbuffer;
+#endif
+#ifdef SUPPORT_PCRE16
+    q16 = (pcre_uint16 *) dbuffer;
+#endif
+#ifdef SUPPORT_PCRE32
+    q32 = (pcre_uint32 *) dbuffer;
+#endif
+
+    while ((c = *p++) != 0)
+      {
+      int i = 0;
+      int n = 0;
+
+      /* In UTF mode, input can be UTF-8, so just copy all non-backslash bytes.
+      In non-UTF mode, allow the value of the byte to fall through to later,
+      where values greater than 127 are turned into UTF-8 when running in
+      16-bit or 32-bit mode. */
+
+      if (c != '\\')
+        {
+#ifndef NOUTF
+        if (use_utf && HASUTF8EXTRALEN(c)) { GETUTF8INC(c, p); }
+#endif
+        }
+
+      /* Handle backslash escapes */
+
+      else switch ((c = *p++))
+        {
+        case 'a': c =    7; break;
+        case 'b': c = '\b'; break;
+        case 'e': c =   27; break;
+        case 'f': c = '\f'; break;
+        case 'n': c = '\n'; break;
+        case 'r': c = '\r'; break;
+        case 't': c = '\t'; break;
+        case 'v': c = '\v'; break;
+
+        case '0': case '1': case '2': case '3':
+        case '4': case '5': case '6': case '7':
+        c -= '0';
+        while (i++ < 2 && isdigit(*p) && *p != '8' && *p != '9')
+          c = c * 8 + *p++ - '0';
+        break;
+
+        case 'x':
+        if (*p == '{')
+          {
+          pcre_uint8 *pt = p;
+          c = 0;
+
+          /* We used to have "while (isxdigit(*(++pt)))" here, but it fails
+          when isxdigit() is a macro that refers to its argument more than
+          once. This is banned by the C Standard, but apparently happens in at
+          least one MacOS environment. */
+
+          for (pt++; isxdigit(*pt); pt++)
+            {
+            if (++i == 9)
+              fprintf(outfile, "** Too many hex digits in \\x{...} item; "
+                               "using only the first eight.\n");
+            else c = c * 16 + tolower(*pt) - ((isdigit(*pt))? '0' : 'a' - 10);
+            }
+          if (*pt == '}')
+            {
+            p = pt + 1;
+            break;
+            }
+          /* Not correct form for \x{...}; fall through */
+          }
+
+        /* \x without {} always defines just one byte in 8-bit mode. This
+        allows UTF-8 characters to be constructed byte by byte, and also allows
+        invalid UTF-8 sequences to be made. Just copy the byte in UTF mode.
+        Otherwise, pass it down to later code so that it can be turned into
+        UTF-8 when running in 16/32-bit mode. */
+
+        c = 0;
+        while (i++ < 2 && isxdigit(*p))
+          {
+          c = c * 16 + tolower(*p) - ((isdigit(*p))? '0' : 'a' - 10);
+          p++;
+          }
+#if !defined NOUTF && defined SUPPORT_PCRE8
+        if (use_utf && (pcre_mode == PCRE8_MODE))
+          {
+          *q8++ = c;
+          continue;
+          }
+#endif
+        break;
+
+        case 0:   /* \ followed by EOF allows for an empty line */
+        p--;
+        continue;
+
+        case '>':
+        if (*p == '-')
+          {
+          start_offset_sign = -1;
+          p++;
+          }
+        while(isdigit(*p)) start_offset = start_offset * 10 + *p++ - '0';
+        start_offset *= start_offset_sign;
+        continue;
+
+        case 'A':  /* Option setting */
+        options |= PCRE_ANCHORED;
+        continue;
+
+        case 'B':
+        options |= PCRE_NOTBOL;
+        continue;
+
+        case 'C':
+        if (isdigit(*p))    /* Set copy string */
+          {
+          while(isdigit(*p)) n = n * 10 + *p++ - '0';
+          copystrings |= 1 << n;
+          }
+        else if (isalnum(*p))
+          {
+          READ_CAPTURE_NAME(p, &cn8ptr, &cn16ptr, &cn32ptr, re);
+          }
+        else if (*p == '+')
+          {
+          callout_extra = 1;
+          p++;
+          }
+        else if (*p == '-')
+          {
+          SET_PCRE_CALLOUT(NULL);
+          p++;
+          }
+        else if (*p == '!')
+          {
+          callout_fail_id = 0;
+          p++;
+          while(isdigit(*p))
+            callout_fail_id = callout_fail_id * 10 + *p++ - '0';
+          callout_fail_count = 0;
+          if (*p == '!')
+            {
+            p++;
+            while(isdigit(*p))
+              callout_fail_count = callout_fail_count * 10 + *p++ - '0';
+            }
+          }
+        else if (*p == '*')
+          {
+          int sign = 1;
+          callout_data = 0;
+          if (*(++p) == '-') { sign = -1; p++; }
+          while(isdigit(*p))
+            callout_data = callout_data * 10 + *p++ - '0';
+          callout_data *= sign;
+          callout_data_set = 1;
+          }
+        continue;
+
+#if !defined NODFA
+        case 'D':
+#if !defined NOPOSIX
+        if (posix || do_posix)
+          printf("** Can't use dfa matching in POSIX mode: \\D ignored\n");
+        else
+#endif
+          use_dfa = 1;
+        continue;
+#endif
+
+#if !defined NODFA
+        case 'F':
+        options |= PCRE_DFA_SHORTEST;
+        continue;
+#endif
+
+        case 'G':
+        if (isdigit(*p))
+          {
+          while(isdigit(*p)) n = n * 10 + *p++ - '0';
+          getstrings |= 1 << n;
+          }
+        else if (isalnum(*p))
+          {
+          READ_CAPTURE_NAME(p, &gn8ptr, &gn16ptr, &gn32ptr, re);
+          }
+        continue;
+
+        case 'J':
+        while(isdigit(*p)) n = n * 10 + *p++ - '0';
+        if (extra != NULL
+            && (extra->flags & PCRE_EXTRA_EXECUTABLE_JIT) != 0
+            && extra->executable_jit != NULL)
+          {
+          if (jit_stack != NULL) { PCRE_JIT_STACK_FREE(jit_stack); }
+          jit_stack = PCRE_JIT_STACK_ALLOC(1, n * 1024);
+          PCRE_ASSIGN_JIT_STACK(extra, jit_callback, jit_stack);
+          }
+        continue;
+
+        case 'L':
+        getlist = 1;
+        continue;
+
+        case 'M':
+        find_match_limit = 1;
+        continue;
+
+        case 'N':
+        if ((options & PCRE_NOTEMPTY) != 0)
+          options = (options & ~PCRE_NOTEMPTY) | PCRE_NOTEMPTY_ATSTART;
+        else
+          options |= PCRE_NOTEMPTY;
+        continue;
+
+        case 'O':
+        while(isdigit(*p)) n = n * 10 + *p++ - '0';
+        if (n > size_offsets_max)
+          {
+          size_offsets_max = n;
+          free(offsets);
+          use_offsets = offsets = (int *)malloc(size_offsets_max * sizeof(int));
+          if (offsets == NULL)
+            {
+            printf("** Failed to get %d bytes of memory for offsets vector\n",
+              (int)(size_offsets_max * sizeof(int)));
+            yield = 1;
+            goto EXIT;
+            }
+          }
+        use_size_offsets = n;
+        if (n == 0) use_offsets = NULL;   /* Ensures it can't write to it */
+          else use_offsets = offsets + size_offsets_max - n;  /* To catch overruns */
+        continue;
+
+        case 'P':
+        options |= ((options & PCRE_PARTIAL_SOFT) == 0)?
+          PCRE_PARTIAL_SOFT : PCRE_PARTIAL_HARD;
+        continue;
+
+        case 'Q':
+        while(isdigit(*p)) n = n * 10 + *p++ - '0';
+        if (extra == NULL)
+          {
+          extra = (pcre_extra *)malloc(sizeof(pcre_extra));
+          extra->flags = 0;
+          }
+        extra->flags |= PCRE_EXTRA_MATCH_LIMIT_RECURSION;
+        extra->match_limit_recursion = n;
+        continue;
+
+        case 'q':
+        while(isdigit(*p)) n = n * 10 + *p++ - '0';
+        if (extra == NULL)
+          {
+          extra = (pcre_extra *)malloc(sizeof(pcre_extra));
+          extra->flags = 0;
+          }
+        extra->flags |= PCRE_EXTRA_MATCH_LIMIT;
+        extra->match_limit = n;
+        continue;
+
+#if !defined NODFA
+        case 'R':
+        options |= PCRE_DFA_RESTART;
+        continue;
+#endif
+
+        case 'S':
+        show_malloc = 1;
+        continue;
+
+        case 'Y':
+        options |= PCRE_NO_START_OPTIMIZE;
+        continue;
+
+        case 'Z':
+        options |= PCRE_NOTEOL;
+        continue;
+
+        case '?':
+        options |= PCRE_NO_UTF8_CHECK;
+        continue;
+
+        case '<':
+          {
+          int x = check_newline(p, outfile);
+          if (x == 0) goto NEXT_DATA;
+          options |= x;
+          while (*p++ != '>');
+          }
+        continue;
+        }
+
+      /* We now have a character value in c that may be greater than 255.
+      In 8-bit mode we convert to UTF-8 if we are in UTF mode. Values greater
+      than 127 in UTF mode must have come from \x{...} or octal constructs
+      because values from \x.. get this far only in non-UTF mode. */
+
+#ifdef SUPPORT_PCRE8
+      if (pcre_mode == PCRE8_MODE)
+        {
+#ifndef NOUTF
+        if (use_utf)
+          {
+          if (c > 0x7fffffff)
+            {
+            fprintf(outfile, "** Character \\x{%x} is greater than 0x7fffffff "
+              "and so cannot be converted to UTF-8\n", c);
+            goto NEXT_DATA;
+            }
+          q8 += ord2utf8(c, q8);
+          }
+        else
+#endif
+          {
+          if (c > 0xffu)
+            {
+            fprintf(outfile, "** Character \\x{%x} is greater than 255 "
+              "and UTF-8 mode is not enabled.\n", c);
+            fprintf(outfile, "** Truncation will probably give the wrong "
+              "result.\n");
+            }
+          *q8++ = c;
+          }
+        }
+#endif
+#ifdef SUPPORT_PCRE16
+      if (pcre_mode == PCRE16_MODE)
+        {
+#ifndef NOUTF
+        if (use_utf)
+          {
+          if (c > 0x10ffffu)
+            {
+            fprintf(outfile, "** Failed: character \\x{%x} is greater than "
+              "0x10ffff and so cannot be converted to UTF-16\n", c);
+            goto NEXT_DATA;
+            }
+          else if (c >= 0x10000u)
+            {
+            c-= 0x10000u;
+            *q16++ = 0xD800 | (c >> 10);
+            *q16++ = 0xDC00 | (c & 0x3ff);
+            }
+          else
+            *q16++ = c;
+          }
+        else
+#endif
+          {
+          if (c > 0xffffu)
+            {
+            fprintf(outfile, "** Character \\x{%x} is greater than 0xffff "
+              "and UTF-16 mode is not enabled.\n", c);
+            fprintf(outfile, "** Truncation will probably give the wrong "
+              "result.\n");
+            }
+
+          *q16++ = c;
+          }
+        }
+#endif
+#ifdef SUPPORT_PCRE32
+      if (pcre_mode == PCRE32_MODE)
+        {
+        *q32++ = c;
+        }
+#endif
+
+      }
+
+    /* Reached end of subject string */
+
+#ifdef SUPPORT_PCRE8
+    if (pcre_mode == PCRE8_MODE)
+    {
+      *q8 = 0;
+      len = (int)(q8 - (pcre_uint8 *)dbuffer);
+    }
+#endif
+#ifdef SUPPORT_PCRE16
+    if (pcre_mode == PCRE16_MODE)
+    {
+      *q16 = 0;
+      len = (int)(q16 - (pcre_uint16 *)dbuffer);
+    }
+#endif
+#ifdef SUPPORT_PCRE32
+    if (pcre_mode == PCRE32_MODE)
+    {
+      *q32 = 0;
+      len = (int)(q32 - (pcre_uint32 *)dbuffer);
+    }
+#endif
+
+    /* If we're compiling with explicit valgrind support, Mark the data from after
+    its end to the end of the buffer as unaddressable, so that a read over the end
+    of the buffer will be seen by valgrind, even if it doesn't cause a crash.
+    If we're not building with valgrind support, at least move the data to the end
+    of the buffer so that it might at least cause a crash.
+    If we are using the POSIX interface, we must include the terminating zero. */
+
+    bptr = dbuffer;
+
+#if !defined NOPOSIX
+    if (posix || do_posix)
+      {
+#ifdef SUPPORT_VALGRIND
+      VALGRIND_MAKE_MEM_NOACCESS(dbuffer + len + 1, dbuffer_size - (len + 1));
+#else
+      memmove(bptr + dbuffer_size - len - 1, bptr, len + 1);
+      bptr += dbuffer_size - len - 1;
+#endif
+      }
+    else
+#endif
+      {
+#ifdef SUPPORT_VALGRIND
+      VALGRIND_MAKE_MEM_NOACCESS(dbuffer + len * CHAR_SIZE, (dbuffer_size - len) * CHAR_SIZE);
+#else
+      bptr = memmove(bptr + (dbuffer_size - len) * CHAR_SIZE, bptr, len * CHAR_SIZE);
+#endif
+      }
+
+    if ((all_use_dfa || use_dfa) && find_match_limit)
+      {
+      printf("**Match limit not relevant for DFA matching: ignored\n");
+      find_match_limit = 0;
+      }
+
+    /* Handle matching via the POSIX interface, which does not
+    support timing or playing with the match limit or callout data. */
+
+#if !defined NOPOSIX
+    if (posix || do_posix)
+      {
+      int rc;
+      int eflags = 0;
+      regmatch_t *pmatch = NULL;
+      if (use_size_offsets > 0)
+        pmatch = (regmatch_t *)malloc(sizeof(regmatch_t) * use_size_offsets);
+      if ((options & PCRE_NOTBOL) != 0) eflags |= REG_NOTBOL;
+      if ((options & PCRE_NOTEOL) != 0) eflags |= REG_NOTEOL;
+      if ((options & PCRE_NOTEMPTY) != 0) eflags |= REG_NOTEMPTY;
+
+      rc = regexec(&preg, (const char *)bptr, use_size_offsets, pmatch, eflags);
+
+      if (rc != 0)
+        {
+        (void)regerror(rc, &preg, (char *)buffer, buffer_size);
+        fprintf(outfile, "No match: POSIX code %d: %s\n", rc, buffer);
+        }
+      else if ((REAL_PCRE_OPTIONS(preg.re_pcre) & PCRE_NO_AUTO_CAPTURE) != 0)
+        {
+        fprintf(outfile, "Matched with REG_NOSUB\n");
+        }
+      else
+        {
+        size_t i;
+        for (i = 0; i < (size_t)use_size_offsets; i++)
+          {
+          if (pmatch[i].rm_so >= 0)
+            {
+            fprintf(outfile, "%2d: ", (int)i);
+            PCHARSV(dbuffer, pmatch[i].rm_so,
+              pmatch[i].rm_eo - pmatch[i].rm_so, outfile);
+            fprintf(outfile, "\n");
+            if (do_showcaprest || (i == 0 && do_showrest))
+              {
+              fprintf(outfile, "%2d+ ", (int)i);
+              PCHARSV(dbuffer, pmatch[i].rm_eo, len - pmatch[i].rm_eo,
+                outfile);
+              fprintf(outfile, "\n");
+              }
+            }
+          }
+        }
+      free(pmatch);
+      goto NEXT_DATA;
+      }
+
+#endif  /* !defined NOPOSIX */
+
+    /* Handle matching via the native interface - repeats for /g and /G */
+
+    /* Ensure that there is a JIT callback if we want to verify that JIT was
+    actually used. If jit_stack == NULL, no stack has yet been assigned. */
+
+    if (verify_jit && jit_stack == NULL && extra != NULL)
+       { PCRE_ASSIGN_JIT_STACK(extra, jit_callback, jit_stack); }
+
+    for (;; gmatched++)    /* Loop for /g or /G */
+      {
+      markptr = NULL;
+      jit_was_used = FALSE;
+
+      if (timeitm > 0)
+        {
+        register int i;
+        clock_t time_taken;
+        clock_t start_time = clock();
+
+#if !defined NODFA
+        if (all_use_dfa || use_dfa)
+          {
+          if ((options & PCRE_DFA_RESTART) != 0)
+            {
+            fprintf(outfile, "Timing DFA restarts is not supported\n");
+            break;
+            }
+          if (dfa_workspace == NULL)
+            dfa_workspace = (int *)malloc(DFA_WS_DIMENSION*sizeof(int));
+          for (i = 0; i < timeitm; i++)
+            {
+            PCRE_DFA_EXEC(count, re, extra, bptr, len, start_offset,
+              (options | g_notempty), use_offsets, use_size_offsets,
+              dfa_workspace, DFA_WS_DIMENSION);
+            }
+          }
+        else
+#endif
+
+        for (i = 0; i < timeitm; i++)
+          {
+          PCRE_EXEC(count, re, extra, bptr, len, start_offset,
+            (options | g_notempty), use_offsets, use_size_offsets);
+          }
+        time_taken = clock() - start_time;
+        fprintf(outfile, "Execute time %.4f milliseconds\n",
+          (((double)time_taken * 1000.0) / (double)timeitm) /
+            (double)CLOCKS_PER_SEC);
+        }
+
+      /* If find_match_limit is set, we want to do repeated matches with
+      varying limits in order to find the minimum value for the match limit and
+      for the recursion limit. The match limits are relevant only to the normal
+      running of pcre_exec(), so disable the JIT optimization. This makes it
+      possible to run the same set of tests with and without JIT externally
+      requested. */
+
+      if (find_match_limit)
+        {
+        if (extra != NULL) { PCRE_FREE_STUDY(extra); }
+        extra = (pcre_extra *)malloc(sizeof(pcre_extra));
+        extra->flags = 0;
+
+        (void)check_match_limit(re, extra, bptr, len, start_offset,
+          options|g_notempty, use_offsets, use_size_offsets,
+          PCRE_EXTRA_MATCH_LIMIT, &(extra->match_limit),
+          PCRE_ERROR_MATCHLIMIT, "match()");
+
+        count = check_match_limit(re, extra, bptr, len, start_offset,
+          options|g_notempty, use_offsets, use_size_offsets,
+          PCRE_EXTRA_MATCH_LIMIT_RECURSION, &(extra->match_limit_recursion),
+          PCRE_ERROR_RECURSIONLIMIT, "match() recursion");
+        }
+
+      /* If callout_data is set, use the interface with additional data */
+
+      else if (callout_data_set)
+        {
+        if (extra == NULL)
+          {
+          extra = (pcre_extra *)malloc(sizeof(pcre_extra));
+          extra->flags = 0;
+          }
+        extra->flags |= PCRE_EXTRA_CALLOUT_DATA;
+        extra->callout_data = &callout_data;
+        PCRE_EXEC(count, re, extra, bptr, len, start_offset,
+          options | g_notempty, use_offsets, use_size_offsets);
+        extra->flags &= ~PCRE_EXTRA_CALLOUT_DATA;
+        }
+
+      /* The normal case is just to do the match once, with the default
+      value of match_limit. */
+
+#if !defined NODFA
+      else if (all_use_dfa || use_dfa)
+        {
+        if (dfa_workspace == NULL)
+          dfa_workspace = (int *)malloc(DFA_WS_DIMENSION*sizeof(int));
+        if (dfa_matched++ == 0)
+          dfa_workspace[0] = -1;  /* To catch bad restart */
+        PCRE_DFA_EXEC(count, re, extra, bptr, len, start_offset,
+          (options | g_notempty), use_offsets, use_size_offsets, dfa_workspace,
+          DFA_WS_DIMENSION);
+        if (count == 0)
+          {
+          fprintf(outfile, "Matched, but offsets vector is too small to show all matches\n");
+          count = use_size_offsets/2;
+          }
+        }
+#endif
+
+      else
+        {
+        PCRE_EXEC(count, re, extra, bptr, len, start_offset,
+          options | g_notempty, use_offsets, use_size_offsets);
+        if (count == 0)
+          {
+          fprintf(outfile, "Matched, but too many substrings\n");
+          /* 2 is a special case; match can be returned */
+          count = (use_size_offsets == 2)? 1 : use_size_offsets/3;
+          }
+        }
+
+      /* Matched */
+
+      if (count >= 0)
+        {
+        int i, maxcount;
+        void *cnptr, *gnptr;
+
+#if !defined NODFA
+        if (all_use_dfa || use_dfa) maxcount = use_size_offsets/2; else
+#endif
+          /* 2 is a special case; match can be returned */
+          maxcount = (use_size_offsets == 2)? 1 : use_size_offsets/3;
+
+        /* This is a check against a lunatic return value. */
+
+        if (count > maxcount)
+          {
+          fprintf(outfile,
+            "** PCRE error: returned count %d is too big for offset size %d\n",
+            count, use_size_offsets);
+          count = use_size_offsets/3;
+          if (do_g || do_G)
+            {
+            fprintf(outfile, "** /%c loop abandoned\n", do_g? 'g' : 'G');
+            do_g = do_G = FALSE;        /* Break g/G loop */
+            }
+          }
+
+        /* do_allcaps requests showing of all captures in the pattern, to check
+        unset ones at the end. */
+
+        if (do_allcaps)
+          {
+          if (new_info(re, NULL, PCRE_INFO_CAPTURECOUNT, &count) < 0)
+            goto SKIP_DATA;
+          count++;   /* Allow for full match */
+          if (count * 2 > use_size_offsets) count = use_size_offsets/2;
+          }
+
+        /* Output the captured substrings */
+
+        for (i = 0; i < count * 2; i += 2)
+          {
+          if (use_offsets[i] < 0)
+            {
+            if (use_offsets[i] != -1)
+              fprintf(outfile, "ERROR: bad negative value %d for offset %d\n",
+                use_offsets[i], i);
+            if (use_offsets[i+1] != -1)
+              fprintf(outfile, "ERROR: bad negative value %d for offset %d\n",
+                use_offsets[i+1], i+1);
+            fprintf(outfile, "%2d: \n", i/2);
+            }
+          else
+            {
+            fprintf(outfile, "%2d: ", i/2);
+            PCHARSV(bptr, use_offsets[i],
+              use_offsets[i+1] - use_offsets[i], outfile);
+            if (verify_jit && jit_was_used) fprintf(outfile, " (JIT)");
+            fprintf(outfile, "\n");
+            if (do_showcaprest || (i == 0 && do_showrest))
+              {
+              fprintf(outfile, "%2d+ ", i/2);
+              PCHARSV(bptr, use_offsets[i+1], len - use_offsets[i+1],
+                outfile);
+              fprintf(outfile, "\n");
+              }
+            }
+          }
+
+        if (markptr != NULL)
+          {
+          fprintf(outfile, "MK: ");
+          PCHARSV(markptr, 0, -1, outfile);
+          fprintf(outfile, "\n");
+          }
+
+        for (i = 0; i < 32; i++)
+          {
+          if ((copystrings & (1 << i)) != 0)
+            {
+            int rc;
+            char copybuffer[256];
+            PCRE_COPY_SUBSTRING(rc, bptr, use_offsets, count, i,
+              copybuffer, sizeof(copybuffer));
+            if (rc < 0)
+              fprintf(outfile, "copy substring %d failed %d\n", i, rc);
+            else
+              {
+              fprintf(outfile, "%2dC ", i);
+              PCHARSV(copybuffer, 0, rc, outfile);
+              fprintf(outfile, " (%d)\n", rc);
+              }
+            }
+          }
+
+        cnptr = copynames;
+        for (;;)
+          {
+          int rc;
+          char copybuffer[256];
+
+#ifdef SUPPORT_PCRE32
+          if (pcre_mode == PCRE32_MODE)
+            {
+            if (*(pcre_uint32 *)cnptr == 0) break;
+            }
+#endif
+#ifdef SUPPORT_PCRE16
+          if (pcre_mode == PCRE16_MODE)
+            {
+            if (*(pcre_uint16 *)cnptr == 0) break;
+            }
+#endif
+#ifdef SUPPORT_PCRE8
+          if (pcre_mode == PCRE8_MODE)
+            {
+            if (*(pcre_uint8 *)cnptr == 0) break;
+            }
+#endif
+
+          PCRE_COPY_NAMED_SUBSTRING(rc, re, bptr, use_offsets, count,
+            cnptr, copybuffer, sizeof(copybuffer));
+
+          if (rc < 0)
+            {
+            fprintf(outfile, "copy substring ");
+            PCHARSV(cnptr, 0, -1, outfile);
+            fprintf(outfile, " failed %d\n", rc);
+            }
+          else
+            {
+            fprintf(outfile, "  C ");
+            PCHARSV(copybuffer, 0, rc, outfile);
+            fprintf(outfile, " (%d) ", rc);
+            PCHARSV(cnptr, 0, -1, outfile);
+            putc('\n', outfile);
+            }
+
+          cnptr = (char *)cnptr + (STRLEN(cnptr) + 1) * CHAR_SIZE;
+          }
+
+        for (i = 0; i < 32; i++)
+          {
+          if ((getstrings & (1 << i)) != 0)
+            {
+            int rc;
+            const char *substring;
+            PCRE_GET_SUBSTRING(rc, bptr, use_offsets, count, i, &substring);
+            if (rc < 0)
+              fprintf(outfile, "get substring %d failed %d\n", i, rc);
+            else
+              {
+              fprintf(outfile, "%2dG ", i);
+              PCHARSV(substring, 0, rc, outfile);
+              fprintf(outfile, " (%d)\n", rc);
+              PCRE_FREE_SUBSTRING(substring);
+              }
+            }
+          }
+
+        gnptr = getnames;
+        for (;;)
+          {
+          int rc;
+          const char *substring;
+
+#ifdef SUPPORT_PCRE32
+          if (pcre_mode == PCRE32_MODE)
+            {
+            if (*(pcre_uint32 *)gnptr == 0) break;
+            }
+#endif
+#ifdef SUPPORT_PCRE16
+          if (pcre_mode == PCRE16_MODE)
+            {
+            if (*(pcre_uint16 *)gnptr == 0) break;
+            }
+#endif
+#ifdef SUPPORT_PCRE8
+          if (pcre_mode == PCRE8_MODE)
+            {
+            if (*(pcre_uint8 *)gnptr == 0) break;
+            }
+#endif
+
+          PCRE_GET_NAMED_SUBSTRING(rc, re, bptr, use_offsets, count,
+            gnptr, &substring);
+          if (rc < 0)
+            {
+            fprintf(outfile, "get substring ");
+            PCHARSV(gnptr, 0, -1, outfile);
+            fprintf(outfile, " failed %d\n", rc);
+            }
+          else
+            {
+            fprintf(outfile, "  G ");
+            PCHARSV(substring, 0, rc, outfile);
+            fprintf(outfile, " (%d) ", rc);
+            PCHARSV(gnptr, 0, -1, outfile);
+            PCRE_FREE_SUBSTRING(substring);
+            putc('\n', outfile);
+            }
+
+          gnptr = (char *)gnptr + (STRLEN(gnptr) + 1) * CHAR_SIZE;
+          }
+
+        if (getlist)
+          {
+          int rc;
+          const char **stringlist;
+          PCRE_GET_SUBSTRING_LIST(rc, bptr, use_offsets, count, &stringlist);
+          if (rc < 0)
+            fprintf(outfile, "get substring list failed %d\n", rc);
+          else
+            {
+            for (i = 0; i < count; i++)
+              {
+              fprintf(outfile, "%2dL ", i);
+              PCHARSV(stringlist[i], 0, -1, outfile);
+              putc('\n', outfile);
+              }
+            if (stringlist[i] != NULL)
+              fprintf(outfile, "string list not terminated by NULL\n");
+            PCRE_FREE_SUBSTRING_LIST(stringlist);
+            }
+          }
+        }
+
+      /* There was a partial match. If the bumpalong point is not the same as
+      the first inspected character, show the offset explicitly. */
+
+      else if (count == PCRE_ERROR_PARTIAL)
+        {
+        fprintf(outfile, "Partial match");
+        if (use_size_offsets > 2 && use_offsets[0] != use_offsets[2])
+          fprintf(outfile, " at offset %d", use_offsets[2]);
+        if (markptr != NULL)
+          {
+          fprintf(outfile, ", mark=");
+          PCHARSV(markptr, 0, -1, outfile);
+          }
+        if (use_size_offsets > 1)
+          {
+          fprintf(outfile, ": ");
+          PCHARSV(bptr, use_offsets[0], use_offsets[1] - use_offsets[0],
+            outfile);
+          }
+        if (verify_jit && jit_was_used) fprintf(outfile, " (JIT)");
+        fprintf(outfile, "\n");
+        break;  /* Out of the /g loop */
+        }
+
+      /* Failed to match. If this is a /g or /G loop and we previously set
+      g_notempty after a null match, this is not necessarily the end. We want
+      to advance the start offset, and continue. We won't be at the end of the
+      string - that was checked before setting g_notempty.
+
+      Complication arises in the case when the newline convention is "any",
+      "crlf", or "anycrlf". If the previous match was at the end of a line
+      terminated by CRLF, an advance of one character just passes the \r,
+      whereas we should prefer the longer newline sequence, as does the code in
+      pcre_exec(). Fudge the offset value to achieve this. We check for a
+      newline setting in the pattern; if none was set, use PCRE_CONFIG() to
+      find the default.
+
+      Otherwise, in the case of UTF-8 matching, the advance must be one
+      character, not one byte. */
+
+      else
+        {
+        if (g_notempty != 0)
+          {
+          int onechar = 1;
+          unsigned int obits = REAL_PCRE_OPTIONS(re);
+          use_offsets[0] = start_offset;
+          if ((obits & PCRE_NEWLINE_BITS) == 0)
+            {
+            int d;
+            (void)PCRE_CONFIG(PCRE_CONFIG_NEWLINE, &d);
+            /* Note that these values are always the ASCII ones, even in
+            EBCDIC environments. CR = 13, NL = 10. */
+            obits = (d == 13)? PCRE_NEWLINE_CR :
+                    (d == 10)? PCRE_NEWLINE_LF :
+                    (d == (13<<8 | 10))? PCRE_NEWLINE_CRLF :
+                    (d == -2)? PCRE_NEWLINE_ANYCRLF :
+                    (d == -1)? PCRE_NEWLINE_ANY : 0;
+            }
+          if (((obits & PCRE_NEWLINE_BITS) == PCRE_NEWLINE_ANY ||
+               (obits & PCRE_NEWLINE_BITS) == PCRE_NEWLINE_CRLF ||
+               (obits & PCRE_NEWLINE_BITS) == PCRE_NEWLINE_ANYCRLF)
+              &&
+              start_offset < len - 1 && (
+#ifdef SUPPORT_PCRE8
+              (pcre_mode == PCRE8_MODE &&
+               bptr[start_offset] == '\r' &&
+               bptr[start_offset + 1] == '\n') ||
+#endif
+#ifdef SUPPORT_PCRE16
+              (pcre_mode == PCRE16_MODE &&
+               ((PCRE_SPTR16)bptr)[start_offset] == '\r' &&
+               ((PCRE_SPTR16)bptr)[start_offset + 1] == '\n') ||
+#endif
+#ifdef SUPPORT_PCRE32
+              (pcre_mode == PCRE32_MODE &&
+               ((PCRE_SPTR32)bptr)[start_offset] == '\r' &&
+               ((PCRE_SPTR32)bptr)[start_offset + 1] == '\n') ||
+#endif
+              0))
+            onechar++;
+          else if (use_utf)
+            {
+            while (start_offset + onechar < len)
+              {
+              if ((bptr[start_offset+onechar] & 0xc0) != 0x80) break;
+              onechar++;
+              }
+            }
+          use_offsets[1] = start_offset + onechar;
+          }
+        else
+          {
+          switch(count)
+            {
+            case PCRE_ERROR_NOMATCH:
+            if (gmatched == 0)
+              {
+              if (markptr == NULL)
+                {
+                fprintf(outfile, "No match");
+                }
+              else
+                {
+                fprintf(outfile, "No match, mark = ");
+                PCHARSV(markptr, 0, -1, outfile);
+                }
+              if (verify_jit && jit_was_used) fprintf(outfile, " (JIT)");
+              putc('\n', outfile);
+              }
+            break;
+
+            case PCRE_ERROR_BADUTF8:
+            case PCRE_ERROR_SHORTUTF8:
+            fprintf(outfile, "Error %d (%s UTF-%d string)", count,
+              (count == PCRE_ERROR_BADUTF8)? "bad" : "short",
+              8 * CHAR_SIZE);
+            if (use_size_offsets >= 2)
+              fprintf(outfile, " offset=%d reason=%d", use_offsets[0],
+                use_offsets[1]);
+            fprintf(outfile, "\n");
+            break;
+
+            case PCRE_ERROR_BADUTF8_OFFSET:
+            fprintf(outfile, "Error %d (bad UTF-%d offset)\n", count,
+              8 * CHAR_SIZE);
+            break;
+
+            default:
+            if (count < 0 &&
+                (-count) < (int)(sizeof(errtexts)/sizeof(const char *)))
+              fprintf(outfile, "Error %d (%s)\n", count, errtexts[-count]);
+            else
+              fprintf(outfile, "Error %d (Unexpected value)\n", count);
+            break;
+            }
+
+          break;  /* Out of the /g loop */
+          }
+        }
+
+      /* If not /g or /G we are done */
+
+      if (!do_g && !do_G) break;
+
+      /* If we have matched an empty string, first check to see if we are at
+      the end of the subject. If so, the /g loop is over. Otherwise, mimic what
+      Perl's /g options does. This turns out to be rather cunning. First we set
+      PCRE_NOTEMPTY_ATSTART and PCRE_ANCHORED and try the match again at the
+      same point. If this fails (picked up above) we advance to the next
+      character. */
+
+      g_notempty = 0;
+
+      if (use_offsets[0] == use_offsets[1])
+        {
+        if (use_offsets[0] == len) break;
+        g_notempty = PCRE_NOTEMPTY_ATSTART | PCRE_ANCHORED;
+        }
+
+      /* For /g, update the start offset, leaving the rest alone */
+
+      if (do_g) start_offset = use_offsets[1];
+
+      /* For /G, update the pointer and length */
+
+      else
+        {
+        bptr += use_offsets[1] * CHAR_SIZE;
+        len -= use_offsets[1];
+        }
+      }  /* End of loop for /g and /G */
+
+    NEXT_DATA: continue;
+    }    /* End of loop for data lines */
+
+  CONTINUE:
+
+#if !defined NOPOSIX
+  if (posix || do_posix) regfree(&preg);
+#endif
+
+  if (re != NULL) new_free(re);
+  if (extra != NULL)
+    {
+    PCRE_FREE_STUDY(extra);
+    }
+  if (locale_set)
+    {
+    new_free((void *)tables);
+    setlocale(LC_CTYPE, "C");
+    locale_set = 0;
+    }
+  if (jit_stack != NULL)
+    {
+    PCRE_JIT_STACK_FREE(jit_stack);
+    jit_stack = NULL;
+    }
+  }
+
+if (infile == stdin) fprintf(outfile, "\n");
+
+EXIT:
+
+if (infile != NULL && infile != stdin) fclose(infile);
+if (outfile != NULL && outfile != stdout) fclose(outfile);
+
+free(buffer);
+free(dbuffer);
+free(pbuffer);
+free(offsets);
+
+#ifdef SUPPORT_PCRE16
+if (buffer16 != NULL) free(buffer16);
+#endif
+#ifdef SUPPORT_PCRE32
+if (buffer32 != NULL) free(buffer32);
+#endif
+
+#if !defined NODFA
+if (dfa_workspace != NULL)
+  free(dfa_workspace);
+#endif
+
+#if defined(__VMS)
+  yield = SS$_NORMAL;  /* Return values via DCL symbols */
+#endif
+
+return yield;
+}
+
+/* End of pcretest.c */
diff --git a/pcre/testdata/grepbinary b/pcre/testdata/grepbinary
new file mode 100644
index 00000000000..5efa1302049
Binary files /dev/null and b/pcre/testdata/grepbinary differ
diff --git a/pcre/testdata/grepfilelist b/pcre/testdata/grepfilelist
new file mode 100644
index 00000000000..dd73ec7f91d
--- /dev/null
+++ b/pcre/testdata/grepfilelist
@@ -0,0 +1,3 @@
+testdata/grepinputv
+
+testdata/grepinputx
diff --git a/pcre/testdata/grepinput b/pcre/testdata/grepinput
new file mode 100644
index 00000000000..0f00edd93ca
--- /dev/null
+++ b/pcre/testdata/grepinput
@@ -0,0 +1,611 @@
+This is a file of miscellaneous text that is used as test data for checking
+that the pcregrep command is working correctly. The file must be more than 24K
+long so that it needs more than a single read() call to process it. New
+features should be added at the end, because some of the tests involve the
+output of line numbers, and we don't want these to change.
+
+PATTERN at the start of a line.
+In the middle of a line, PATTERN appears.
+
+This pattern is in lower case.
+
+Here follows a whole lot of stuff that makes the file over 24K long.
+
+-------------------------------------------------------------------------------
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+
+The quick brown fox jumps over the lazy dog. The quick brown fox jumps over the
+lazy dog. The quick brown fox jumps over the lazy dog. The quick brown fox
+jumps over the lazy dog. The quick brown fox jumps over the lazy dog. The quick
+brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+-------------------------------------------------------------------------------
+
+aaaaa0
+aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+bbbbbb
+cccccccccccccccccccccccccccccccccccccccccc
+aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+eeeee
+aaaaa2
+ffffffffff
+
+This is a line before the binary zero.
+This line contains a binary zero here >< for testing.
+This is a line after the binary zero.
+
+ABOVE the elephant 
+ABOVE
+ABOVE theatre
+AB.VE
+AB.VE the turtle
+
+010203040506
+
+PUT NEW DATA ABOVE THIS LINE.
+=============================
+
+Check up on PATTERN near the end.
+This is the last line of this file.
diff --git a/pcre/testdata/grepinput3 b/pcre/testdata/grepinput3
new file mode 100644
index 00000000000..7409cfc07a6
--- /dev/null
+++ b/pcre/testdata/grepinput3
@@ -0,0 +1,15 @@
+triple:	t1_txt	s1_tag	s_txt	p_tag	p_txt	o_tag	o_txt
+
+triple:	t2_txt	s1_tag	s_txt	p_tag	p_txt	o_tag	
+Lorem ipsum dolor sit amet, consectetur adipisicing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur. Excepteur sint occaecat cupidatat non proident, sunt in culpa qui officia deserunt mollit anim id est laborum.
+
+triple:	t3_txt	s2_tag	s_txt	p_tag	p_txt	o_tag	o_txt
+
+triple:	t4_txt	s1_tag	s_txt	p_tag	p_txt	o_tag	o_txt
+
+triple:	t5_txt	s1_tag	s_txt	p_tag	p_txt	o_tag	
+o_txt
+
+triple:	t6_txt	s2_tag	s_txt	p_tag	p_txt	o_tag	o_txt
+
+triple:	t7_txt	s1_tag	s_txt	p_tag	p_txt	o_tag	o_txt
diff --git a/pcre/testdata/grepinput8 b/pcre/testdata/grepinput8
new file mode 100644
index 00000000000..c4b8c440f23
--- /dev/null
+++ b/pcre/testdata/grepinput8
@@ -0,0 +1,11 @@
+X one
+X twoX threeX four
X five
+X six
+X seven…X eight
X nine
X ten
+
+Before 111
+Before 222
Before 333…Match
+After 111
+After 222
After 333
+And so on and so on
+And so on and so on
diff --git a/pcre/testdata/grepinputv b/pcre/testdata/grepinputv
new file mode 100644
index 00000000000..d33d326b01d
--- /dev/null
+++ b/pcre/testdata/grepinputv
@@ -0,0 +1,4 @@
+The quick brown
+fox jumps
+over the lazy dog.
+This time it jumps and jumps and jumps.
diff --git a/pcre/testdata/grepinputx b/pcre/testdata/grepinputx
new file mode 100644
index 00000000000..730cc8a0d05
--- /dev/null
+++ b/pcre/testdata/grepinputx
@@ -0,0 +1,43 @@
+This is a second file of input for the pcregrep tests.
+
+Here is the pattern again.
+
+Pattern
+That time it was on a line by itself.
+
+To pat or not to pat, that is the question.
+
+complete pair
+of lines
+
+That was a complete pair
+of lines all by themselves.
+
+complete pair
+of lines
+
+And there they were again, to check line numbers.
+
+one
+two
+three
+four
+five
+six
+seven
+eight
+nine
+ten
+eleven
+twelve
+thirteen
+fourteen
+fifteen
+sixteen
+seventeen
+eighteen
+nineteen
+twenty
+
+This line contains pattern not on a line by itself.
+This is the last line of this file.
diff --git a/pcre/testdata/greplist b/pcre/testdata/greplist
new file mode 100644
index 00000000000..1434ae96f61
--- /dev/null
+++ b/pcre/testdata/greplist
@@ -0,0 +1,7 @@
+This is a file of patterns for testing the -f option. Don't include any blank
+lines because they will match everything! This is no longer true, so have one.
+
+pattern
+line by itself
+
+End of the list of patterns.
diff --git a/pcre/testdata/grepoutput b/pcre/testdata/grepoutput
new file mode 100644
index 00000000000..cf04091f4a2
--- /dev/null
+++ b/pcre/testdata/grepoutput
@@ -0,0 +1,742 @@
+---------------------------- Test 1 ------------------------------
+PATTERN at the start of a line.
+In the middle of a line, PATTERN appears.
+Check up on PATTERN near the end.
+RC=0
+---------------------------- Test 2 ------------------------------
+PATTERN at the start of a line.
+RC=0
+---------------------------- Test 3 ------------------------------
+7:PATTERN at the start of a line.
+8:In the middle of a line, PATTERN appears.
+10:This pattern is in lower case.
+610:Check up on PATTERN near the end.
+RC=0
+---------------------------- Test 4 ------------------------------
+4
+RC=0
+---------------------------- Test 5 ------------------------------
+./testdata/grepinput:7:PATTERN at the start of a line.
+./testdata/grepinput:8:In the middle of a line, PATTERN appears.
+./testdata/grepinput:10:This pattern is in lower case.
+./testdata/grepinput:610:Check up on PATTERN near the end.
+./testdata/grepinputx:3:Here is the pattern again.
+./testdata/grepinputx:5:Pattern
+./testdata/grepinputx:42:This line contains pattern not on a line by itself.
+RC=0
+---------------------------- Test 6 ------------------------------
+7:PATTERN at the start of a line.
+8:In the middle of a line, PATTERN appears.
+10:This pattern is in lower case.
+610:Check up on PATTERN near the end.
+3:Here is the pattern again.
+5:Pattern
+42:This line contains pattern not on a line by itself.
+RC=0
+---------------------------- Test 7 ------------------------------
+./testdata/grepinput
+./testdata/grepinputx
+RC=0
+---------------------------- Test 8 ------------------------------
+./testdata/grepinput
+RC=0
+---------------------------- Test 9 ------------------------------
+RC=0
+---------------------------- Test 10 -----------------------------
+RC=1
+---------------------------- Test 11 -----------------------------
+1:This is a second file of input for the pcregrep tests.
+2:
+4:
+5:Pattern
+6:That time it was on a line by itself.
+7:
+8:To pat or not to pat, that is the question.
+9:
+10:complete pair
+11:of lines
+12:
+13:That was a complete pair
+14:of lines all by themselves.
+15:
+16:complete pair
+17:of lines
+18:
+19:And there they were again, to check line numbers.
+20:
+21:one
+22:two
+23:three
+24:four
+25:five
+26:six
+27:seven
+28:eight
+29:nine
+30:ten
+31:eleven
+32:twelve
+33:thirteen
+34:fourteen
+35:fifteen
+36:sixteen
+37:seventeen
+38:eighteen
+39:nineteen
+40:twenty
+41:
+43:This is the last line of this file.
+RC=0
+---------------------------- Test 12 -----------------------------
+Pattern
+RC=0
+---------------------------- Test 13 -----------------------------
+Here is the pattern again.
+That time it was on a line by itself.
+seventeen
+This line contains pattern not on a line by itself.
+RC=0
+---------------------------- Test 14 -----------------------------
+./testdata/grepinputx:To pat or not to pat, that is the question.
+RC=0
+---------------------------- Test 15 -----------------------------
+pcregrep: Error in command-line regex at offset 4: nothing to repeat
+RC=2
+---------------------------- Test 16 -----------------------------
+pcregrep: Failed to open ./testdata/nonexistfile: No such file or directory
+RC=2
+---------------------------- Test 17 -----------------------------
+features should be added at the end, because some of the tests involve the
+output of line numbers, and we don't want these to change.
+RC=0
+---------------------------- Test 18 -----------------------------
+4:features should be added at the end, because some of the tests involve the
+output of line numbers, and we don't want these to change.
+583:brown fox jumps over the lazy dog. The quick brown fox jumps over the lazy dog.
+-------------------------------------------------------------------------------
+RC=0
+---------------------------- Test 19 -----------------------------
+Pattern
+RC=0
+---------------------------- Test 20 -----------------------------
+10:complete pair
+of lines
+16:complete pair
+of lines
+RC=0
+---------------------------- Test 21 -----------------------------
+24:four
+25-five
+26-six
+27-seven
+--
+34:fourteen
+35-fifteen
+36-sixteen
+37-seventeen
+RC=0
+---------------------------- Test 22 -----------------------------
+21-one
+22-two
+23-three
+24:four
+--
+31-eleven
+32-twelve
+33-thirteen
+34:fourteen
+RC=0
+---------------------------- Test 23 -----------------------------
+one
+two
+three
+four
+five
+six
+seven
+--
+eleven
+twelve
+thirteen
+fourteen
+fifteen
+sixteen
+seventeen
+RC=0
+---------------------------- Test 24 -----------------------------
+four
+five
+six
+seven
+eight
+nine
+ten
+eleven
+twelve
+thirteen
+fourteen
+fifteen
+sixteen
+seventeen
+eighteen
+nineteen
+twenty
+
+This line contains pattern not on a line by itself.
+This is the last line of this file.
+RC=0
+---------------------------- Test 25 -----------------------------
+15-
+16-complete pair
+17-of lines
+18-
+19-And there they were again, to check line numbers.
+20-
+21-one
+22-two
+23-three
+24:four
+25-five
+26-six
+27-seven
+28-eight
+29-nine
+30-ten
+31-eleven
+32-twelve
+33-thirteen
+34:fourteen
+RC=0
+---------------------------- Test 26 -----------------------------
+
+complete pair
+of lines
+
+And there they were again, to check line numbers.
+
+one
+two
+three
+four
+five
+six
+seven
+eight
+nine
+ten
+eleven
+twelve
+thirteen
+fourteen
+fifteen
+sixteen
+seventeen
+eighteen
+nineteen
+twenty
+
+This line contains pattern not on a line by itself.
+This is the last line of this file.
+RC=0
+---------------------------- Test 27 -----------------------------
+four
+five
+six
+seven
+eight
+nine
+ten
+eleven
+twelve
+thirteen
+fourteen
+fifteen
+sixteen
+seventeen
+eighteen
+nineteen
+twenty
+
+This line contains pattern not on a line by itself.
+This is the last line of this file.
+RC=0
+---------------------------- Test 28 -----------------------------
+14-of lines all by themselves.
+15-
+16-complete pair
+17-of lines
+18-
+19-And there they were again, to check line numbers.
+20-
+21-one
+22-two
+23-three
+24:four
+25-five
+26-six
+27-seven
+28-eight
+29-nine
+30-ten
+31-eleven
+32-twelve
+33-thirteen
+34:fourteen
+RC=0
+---------------------------- Test 29 -----------------------------
+of lines all by themselves.
+
+complete pair
+of lines
+
+And there they were again, to check line numbers.
+
+one
+two
+three
+four
+five
+six
+seven
+eight
+nine
+ten
+eleven
+twelve
+thirteen
+fourteen
+fifteen
+sixteen
+seventeen
+eighteen
+nineteen
+twenty
+
+This line contains pattern not on a line by itself.
+This is the last line of this file.
+RC=0
+---------------------------- Test 30 -----------------------------
+./testdata/grepinput-4-features should be added at the end, because some of the tests involve the
+./testdata/grepinput-5-output of line numbers, and we don't want these to change.
+./testdata/grepinput-6-
+./testdata/grepinput:7:PATTERN at the start of a line.
+./testdata/grepinput:8:In the middle of a line, PATTERN appears.
+./testdata/grepinput-9-
+./testdata/grepinput:10:This pattern is in lower case.
+--
+./testdata/grepinput-607-PUT NEW DATA ABOVE THIS LINE.
+./testdata/grepinput-608-=============================
+./testdata/grepinput-609-
+./testdata/grepinput:610:Check up on PATTERN near the end.
+--
+./testdata/grepinputx-1-This is a second file of input for the pcregrep tests.
+./testdata/grepinputx-2-
+./testdata/grepinputx:3:Here is the pattern again.
+./testdata/grepinputx-4-
+./testdata/grepinputx:5:Pattern
+--
+./testdata/grepinputx-39-nineteen
+./testdata/grepinputx-40-twenty
+./testdata/grepinputx-41-
+./testdata/grepinputx:42:This line contains pattern not on a line by itself.
+RC=0
+---------------------------- Test 31 -----------------------------
+./testdata/grepinput:7:PATTERN at the start of a line.
+./testdata/grepinput:8:In the middle of a line, PATTERN appears.
+./testdata/grepinput-9-
+./testdata/grepinput:10:This pattern is in lower case.
+./testdata/grepinput-11-
+./testdata/grepinput-12-Here follows a whole lot of stuff that makes the file over 24K long.
+./testdata/grepinput-13-
+--
+./testdata/grepinput:610:Check up on PATTERN near the end.
+./testdata/grepinput-611-This is the last line of this file.
+--
+./testdata/grepinputx:3:Here is the pattern again.
+./testdata/grepinputx-4-
+./testdata/grepinputx:5:Pattern
+./testdata/grepinputx-6-That time it was on a line by itself.
+./testdata/grepinputx-7-
+./testdata/grepinputx-8-To pat or not to pat, that is the question.
+--
+./testdata/grepinputx:42:This line contains pattern not on a line by itself.
+./testdata/grepinputx-43-This is the last line of this file.
+RC=0
+---------------------------- Test 32 -----------------------------
+./testdata/grepinputx
+RC=0
+---------------------------- Test 33 -----------------------------
+pcregrep: Failed to open ./testdata/grepnonexist: No such file or directory
+RC=2
+---------------------------- Test 34 -----------------------------
+RC=2
+---------------------------- Test 35 -----------------------------
+./testdata/grepinput8
+./testdata/grepinputx
+RC=0
+---------------------------- Test 36 -----------------------------
+./testdata/grepinput3
+./testdata/grepinputx
+RC=0
+---------------------------- Test 37 -----------------------------
+aaaaa0
+aaaaa2
+010203040506
+RC=0
+======== STDERR ========
+pcregrep: pcre_exec() gave error -8 while matching this text:
+
+aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+
+pcregrep: pcre_exec() gave error -8 while matching this text:
+
+aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+
+pcregrep: Error -8, -21 or -27 means that a resource limit was exceeded.
+pcregrep: Check your regex for nested unlimited loops.
+---------------------------- Test 38 ------------------------------
+This line contains a binary zero here >< for testing.
+RC=0
+---------------------------- Test 39 ------------------------------
+This is a line before the binary zero.
+This line contains a binary zero here >< for testing.
+RC=0
+---------------------------- Test 40 ------------------------------
+This line contains a binary zero here >< for testing.
+This is a line after the binary zero.
+RC=0
+---------------------------- Test 41 ------------------------------
+before the binary zero
+after the binary zero
+RC=0
+---------------------------- Test 42 ------------------------------
+./testdata/grepinput:595:before the binary zero
+./testdata/grepinput:597:after the binary zero
+RC=0
+---------------------------- Test 43 ------------------------------
+595:before
+595:zero
+596:zero
+597:after
+597:zero
+RC=0
+---------------------------- Test 44 ------------------------------
+595:before
+595:zero
+596:zero
+597:zero
+RC=0
+---------------------------- Test 45 ------------------------------
+10:pattern
+595:binary
+596:binary
+597:binary
+RC=0
+---------------------------- Test 46 ------------------------------
+pcregrep: Error in 2nd command-line regex at offset 9: missing )
+RC=2
+---------------------------- Test 47 ------------------------------
+AB.VE
+RC=0
+---------------------------- Test 48 ------------------------------
+ABOVE the elephant 
+AB.VE
+AB.VE the turtle
+RC=0
+---------------------------- Test 49 ------------------------------
+ABOVE the elephant 
+AB.VE
+AB.VE the turtle
+PUT NEW DATA ABOVE THIS LINE.
+RC=0
+---------------------------- Test 50 ------------------------------
+RC=1
+---------------------------- Test 51 ------------------------------
+over the lazy dog.
+This time it jumps and jumps and jumps.
+RC=0
+---------------------------- Test 52 ------------------------------
+fox jumps
+This time it jumps and jumps and jumps.
+RC=0
+---------------------------- Test 53 ------------------------------
+36972,6
+36990,4
+37024,4
+37066,5
+37083,4
+RC=0
+---------------------------- Test 54 ------------------------------
+595:15,6
+595:33,4
+596:28,4
+597:15,5
+597:32,4
+RC=0
+---------------------------- Test 55 -----------------------------
+Here is the pattern again.
+That time it was on a line by itself.
+This line contains pattern not on a line by itself.
+RC=0
+---------------------------- Test 56 -----------------------------
+./testdata/grepinput:456
+./testdata/grepinput3:0
+./testdata/grepinput8:0
+./testdata/grepinputv:1
+./testdata/grepinputx:0
+RC=0
+---------------------------- Test 57 -----------------------------
+./testdata/grepinput:456
+./testdata/grepinputv:1
+RC=0
+---------------------------- Test 58 -----------------------------
+PATTERN at the start of a line.
+In the middle of a line, PATTERN appears.
+Check up on PATTERN near the end.
+RC=0
+---------------------------- Test 59 -----------------------------
+PATTERN at the start of a line.
+In the middle of a line, PATTERN appears.
+Check up on PATTERN near the end.
+RC=0
+---------------------------- Test 60 -----------------------------
+PATTERN at the start of a line.
+In the middle of a line, PATTERN appears.
+Check up on PATTERN near the end.
+RC=0
+---------------------------- Test 61 -----------------------------
+PATTERN at the start of a line.
+In the middle of a line, PATTERN appears.
+Check up on PATTERN near the end.
+RC=0
+---------------------------- Test 62 -----------------------------
+pcregrep: pcre_exec() gave error -8 while matching text that starts:
+
+This is a file of miscellaneous text that is used as test data for checking
+that the pcregrep command is working correctly. The file must be more than 24K
+long so that it needs more than a single read
+
+pcregrep: Error -8, -21 or -27 means that a resource limit was exceeded.
+pcregrep: Check your regex for nested unlimited loops.
+RC=1
+---------------------------- Test 63 -----------------------------
+pcregrep: pcre_exec() gave error -21 while matching text that starts:
+
+This is a file of miscellaneous text that is used as test data for checking
+that the pcregrep command is working correctly. The file must be more than 24K
+long so that it needs more than a single read
+
+pcregrep: Error -8, -21 or -27 means that a resource limit was exceeded.
+pcregrep: Check your regex for nested unlimited loops.
+RC=1
+---------------------------- Test 64 ------------------------------
+appears
+RC=0
+---------------------------- Test 65 ------------------------------
+pear
+RC=0
+---------------------------- Test 66 ------------------------------
+RC=0
+---------------------------- Test 67 ------------------------------
+RC=0
+---------------------------- Test 68 ------------------------------
+pear
+RC=0
+---------------------------- Test 69 -----------------------------
+1:This is a second file of input for the pcregrep tests.
+2:
+4:
+5:Pattern
+6:That time it was on a line by itself.
+7:
+8:To pat or not to pat, that is the question.
+9:
+10:complete pair
+11:of lines
+12:
+13:That was a complete pair
+14:of lines all by themselves.
+15:
+16:complete pair
+17:of lines
+18:
+19:And there they were again, to check line numbers.
+20:
+21:one
+22:two
+23:three
+24:four
+25:five
+26:six
+27:seven
+28:eight
+29:nine
+30:ten
+31:eleven
+32:twelve
+33:thirteen
+34:fourteen
+35:fifteen
+36:sixteen
+37:seventeen
+38:eighteen
+39:nineteen
+40:twenty
+41:
+43:This is the last line of this file.
+RC=0
+---------------------------- Test 70 -----------------------------
+triple:	t1_txt	s1_tag	s_txt	p_tag	p_txt	o_tag	o_txt
+
+triple:	t3_txt	s2_tag	s_txt	p_tag	p_txt	o_tag	o_txt
+
+triple:	t4_txt	s1_tag	s_txt	p_tag	p_txt	o_tag	o_txt
+
+triple:	t6_txt	s2_tag	s_txt	p_tag	p_txt	o_tag	o_txt
+
+RC=0
+---------------------------- Test 71 -----------------------------
+01
+RC=0
+---------------------------- Test 72 -----------------------------
+010203040506
+RC=0
+---------------------------- Test 73 -----------------------------
+01
+RC=0
+---------------------------- Test 74 -----------------------------
+01
+02
+RC=0
+---------------------------- Test 75 -----------------------------
+010203040506
+RC=0
+---------------------------- Test 76 -----------------------------
+01
+02
+RC=0
+---------------------------- Test 77 -----------------------------
+01
+03
+RC=0
+---------------------------- Test 78 -----------------------------
+010203040506
+RC=0
+---------------------------- Test 79 -----------------------------
+01
+03
+RC=0
+---------------------------- Test 80 -----------------------------
+01
+RC=0
+---------------------------- Test 81 -----------------------------
+010203040506
+RC=0
+---------------------------- Test 82 -----------------------------
+01
+RC=0
+---------------------------- Test 83 -----------------------------
+pcregrep: line 4 of file ./testdata/grepinput3 is too long for the internal buffer
+pcregrep: check the --buffer-size option
+RC=2
+---------------------------- Test 84 -----------------------------
+testdata/grepinputv:fox jumps
+testdata/grepinputx:complete pair
+testdata/grepinputx:That was a complete pair
+testdata/grepinputx:complete pair
+testdata/grepinput3:triple:	t7_txt	s1_tag	s_txt	p_tag	p_txt	o_tag	o_txt
+RC=0
+---------------------------- Test 85 -----------------------------
+./testdata/grepinput3:Lorem ipsum dolor sit amet, consectetur adipisicing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur. Excepteur sint occaecat cupidatat non proident, sunt in culpa qui officia deserunt mollit anim id est laborum.
+RC=0
+---------------------------- Test 86 -----------------------------
+Binary file ./testdata/grepbinary matches
+RC=0
+---------------------------- Test 87 -----------------------------
+RC=1
+---------------------------- Test 88 -----------------------------
+Binary file ./testdata/grepbinary matches
+RC=0
+---------------------------- Test 89 -----------------------------
+RC=1
+---------------------------- Test 90 -----------------------------
+RC=1
+---------------------------- Test 91 -----------------------------
+The quick brown fx jumps over the lazy dog.
+RC=0
+---------------------------- Test 92 -----------------------------
+The quick brown fx jumps over the lazy dog.
+RC=0
+---------------------------- Test 93 -----------------------------
+The quick brown fx jumps over the lazy dog.
+RC=0
+---------------------------- Test 94 -----------------------------
+./testdata/grepinput8
+./testdata/grepinputx
+RC=0
+---------------------------- Test 95 -----------------------------
+testdata/grepinputx:complete pair
+testdata/grepinputx:That was a complete pair
+testdata/grepinputx:complete pair
+RC=0
+---------------------------- Test 96 -----------------------------
+./testdata/grepinput3
+./testdata/grepinput8
+./testdata/grepinputx
+RC=0
+---------------------------- Test 97 -----------------------------
+./testdata/grepinput3
+./testdata/grepinputx
+RC=0
+---------------------------- Test 98 -----------------------------
+./testdata/grepinputx
+RC=0
+---------------------------- Test 99 -----------------------------
+./testdata/grepinput3
+./testdata/grepinputx
+RC=0
+---------------------------- Test 100 ------------------------------
+./testdata/grepinput:zerothe.
+./testdata/grepinput:zeroa
+./testdata/grepinput:zerothe.
+RC=0
+---------------------------- Test 101 ------------------------------
+./testdata/grepinput:.|zero|the|.
+./testdata/grepinput:zero|a
+./testdata/grepinput:.|zero|the|.
+RC=0
+---------------------------- Test 102 -----------------------------
+2:
+5:
+7:
+9:
+12:
+14:
+RC=0
+---------------------------- Test 103 -----------------------------
+RC=0
+---------------------------- Test 104 -----------------------------
+2:
+5:
+7:
+9:
+12:
+14:
+RC=0
+---------------------------- Test 105 -----------------------------
+triple:	t1_txt	s1_tag	s_txt	p_tag	p_txt	o_tag	o_txt
+
+triple:	t2_txt	s1_tag	s_txt	p_tag	p_txt	o_tag	
+Lorem ipsum dolor sit amet, consectetur adipisicing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur. Excepteur sint occaecat cupidatat non proident, sunt in culpa qui officia deserunt mollit anim id est laborum.
+
+triple:	t3_txt	s2_tag	s_txt	p_tag	p_txt	o_tag	o_txt
+
+triple:	t4_txt	s1_tag	s_txt	p_tag	p_txt	o_tag	o_txt
+
+triple:	t5_txt	s1_tag	s_txt	p_tag	p_txt	o_tag	
+o_txt
+
+triple:	t6_txt	s2_tag	s_txt	p_tag	p_txt	o_tag	o_txt
+
+triple:	t7_txt	s1_tag	s_txt	p_tag	p_txt	o_tag	o_txt
+RC=0
diff --git a/pcre/testdata/grepoutput8 b/pcre/testdata/grepoutput8
new file mode 100644
index 00000000000..91493bdcf07
--- /dev/null
+++ b/pcre/testdata/grepoutput8
@@ -0,0 +1,12 @@
+---------------------------- Test U1 ------------------------------
+1:X one
+2:X two3:X three4:X four
5:X five
+6:X six
+7:X seven…8:X eight
9:X nine
10:X ten
+RC=0
+---------------------------- Test U2 ------------------------------
+12-Before 111
+13-Before 222
14-Before 333…15:Match
+16-After 111
+17-After 222
18-After 333
+RC=0
diff --git a/pcre/testdata/grepoutputN b/pcre/testdata/grepoutputN
new file mode 100644
index 00000000000..1f9f8801e3b
--- /dev/null
+++ b/pcre/testdata/grepoutputN
@@ -0,0 +1,16 @@
+---------------------------- Test N1 ------------------------------
+1:abc
2:def
---------------------------- Test N2 ------------------------------
+1:abc
def
+2:ghi
+jkl---------------------------- Test N3 ------------------------------
+2:def
3:
+ghi
+jkl---------------------------- Test N4 ------------------------------
+2:ghi
+jkl---------------------------- Test N5 ------------------------------
+1:abc
2:def
+3:ghi
+4:jkl---------------------------- Test N6 ------------------------------
+1:abc
2:def
+3:ghi
+4:jkl
\ No newline at end of file
diff --git a/pcre/testdata/greppatN4 b/pcre/testdata/greppatN4
new file mode 100644
index 00000000000..ea1bfc78ac4
--- /dev/null
+++ b/pcre/testdata/greppatN4
@@ -0,0 +1,2 @@
+xxx
+jkl
\ No newline at end of file
diff --git a/pcre/testdata/saved16 b/pcre/testdata/saved16
new file mode 100644
index 00000000000..583c4342862
Binary files /dev/null and b/pcre/testdata/saved16 differ
diff --git a/pcre/testdata/saved16BE-1 b/pcre/testdata/saved16BE-1
new file mode 100644
index 00000000000..e2e807d4fde
Binary files /dev/null and b/pcre/testdata/saved16BE-1 differ
diff --git a/pcre/testdata/saved16BE-2 b/pcre/testdata/saved16BE-2
new file mode 100644
index 00000000000..cc2718aa1ea
Binary files /dev/null and b/pcre/testdata/saved16BE-2 differ
diff --git a/pcre/testdata/saved16LE-1 b/pcre/testdata/saved16LE-1
new file mode 100644
index 00000000000..b037d4903e1
Binary files /dev/null and b/pcre/testdata/saved16LE-1 differ
diff --git a/pcre/testdata/saved16LE-2 b/pcre/testdata/saved16LE-2
new file mode 100644
index 00000000000..d7034f757fe
Binary files /dev/null and b/pcre/testdata/saved16LE-2 differ
diff --git a/pcre/testdata/saved32 b/pcre/testdata/saved32
new file mode 100644
index 00000000000..5b6fe34c948
Binary files /dev/null and b/pcre/testdata/saved32 differ
diff --git a/pcre/testdata/saved32BE-1 b/pcre/testdata/saved32BE-1
new file mode 100644
index 00000000000..ebe62ca7c66
Binary files /dev/null and b/pcre/testdata/saved32BE-1 differ
diff --git a/pcre/testdata/saved32BE-2 b/pcre/testdata/saved32BE-2
new file mode 100644
index 00000000000..8168343931c
Binary files /dev/null and b/pcre/testdata/saved32BE-2 differ
diff --git a/pcre/testdata/saved32LE-1 b/pcre/testdata/saved32LE-1
new file mode 100644
index 00000000000..e008f3a5a7c
Binary files /dev/null and b/pcre/testdata/saved32LE-1 differ
diff --git a/pcre/testdata/saved32LE-2 b/pcre/testdata/saved32LE-2
new file mode 100644
index 00000000000..cf3bd731928
Binary files /dev/null and b/pcre/testdata/saved32LE-2 differ
diff --git a/pcre/testdata/saved8 b/pcre/testdata/saved8
new file mode 100644
index 00000000000..37d733eedf3
Binary files /dev/null and b/pcre/testdata/saved8 differ
diff --git a/pcre/testdata/testinput1 b/pcre/testdata/testinput1
new file mode 100644
index 00000000000..e3050ef0a99
--- /dev/null
+++ b/pcre/testdata/testinput1
@@ -0,0 +1,5611 @@
+/-- This set of tests is for features that are compatible with all versions of
+    Perl >= 5.10, in non-UTF-8 mode. It should run clean for both the 8-bit and
+    16-bit PCRE libraries. --/
+
+/the quick brown fox/
+    the quick brown fox
+    The quick brown FOX
+    What do you know about the quick brown fox?
+    What do you know about THE QUICK BROWN FOX?
+
+/The quick brown fox/i
+    the quick brown fox
+    The quick brown FOX
+    What do you know about the quick brown fox?
+    What do you know about THE QUICK BROWN FOX?
+
+/abcd\t\n\r\f\a\e\071\x3b\$\\\?caxyz/
+    abcd\t\n\r\f\a\e9;\$\\?caxyz
+
+/a*abc?xyz+pqr{3}ab{2,}xy{4,5}pq{0,6}AB{0,}zz/
+    abxyzpqrrrabbxyyyypqAzz
+    abxyzpqrrrabbxyyyypqAzz
+    aabxyzpqrrrabbxyyyypqAzz
+    aaabxyzpqrrrabbxyyyypqAzz
+    aaaabxyzpqrrrabbxyyyypqAzz
+    abcxyzpqrrrabbxyyyypqAzz
+    aabcxyzpqrrrabbxyyyypqAzz
+    aaabcxyzpqrrrabbxyyyypAzz
+    aaabcxyzpqrrrabbxyyyypqAzz
+    aaabcxyzpqrrrabbxyyyypqqAzz
+    aaabcxyzpqrrrabbxyyyypqqqAzz
+    aaabcxyzpqrrrabbxyyyypqqqqAzz
+    aaabcxyzpqrrrabbxyyyypqqqqqAzz
+    aaabcxyzpqrrrabbxyyyypqqqqqqAzz
+    aaaabcxyzpqrrrabbxyyyypqAzz
+    abxyzzpqrrrabbxyyyypqAzz
+    aabxyzzzpqrrrabbxyyyypqAzz
+    aaabxyzzzzpqrrrabbxyyyypqAzz
+    aaaabxyzzzzpqrrrabbxyyyypqAzz
+    abcxyzzpqrrrabbxyyyypqAzz
+    aabcxyzzzpqrrrabbxyyyypqAzz
+    aaabcxyzzzzpqrrrabbxyyyypqAzz
+    aaaabcxyzzzzpqrrrabbxyyyypqAzz
+    aaaabcxyzzzzpqrrrabbbxyyyypqAzz
+    aaaabcxyzzzzpqrrrabbbxyyyyypqAzz
+    aaabcxyzpqrrrabbxyyyypABzz
+    aaabcxyzpqrrrabbxyyyypABBzz
+    >>>aaabxyzpqrrrabbxyyyypqAzz
+    >aaaabxyzpqrrrabbxyyyypqAzz
+    >>>>abcxyzpqrrrabbxyyyypqAzz
+    *** Failers
+    abxyzpqrrabbxyyyypqAzz
+    abxyzpqrrrrabbxyyyypqAzz
+    abxyzpqrrrabxyyyypqAzz
+    aaaabcxyzzzzpqrrrabbbxyyyyyypqAzz
+    aaaabcxyzzzzpqrrrabbbxyyypqAzz
+    aaabcxyzpqrrrabbxyyyypqqqqqqqAzz
+
+/^(abc){1,2}zz/
+    abczz
+    abcabczz
+    *** Failers
+    zz
+    abcabcabczz
+    >>abczz
+
+/^(b+?|a){1,2}?c/
+    bc
+    bbc
+    bbbc
+    bac
+    bbac
+    aac
+    abbbbbbbbbbbc
+    bbbbbbbbbbbac
+    *** Failers
+    aaac
+    abbbbbbbbbbbac
+
+/^(b+|a){1,2}c/
+    bc
+    bbc
+    bbbc
+    bac
+    bbac
+    aac
+    abbbbbbbbbbbc
+    bbbbbbbbbbbac
+    *** Failers
+    aaac
+    abbbbbbbbbbbac
+
+/^(b+|a){1,2}?bc/
+    bbc
+
+/^(b*|ba){1,2}?bc/
+    babc
+    bbabc
+    bababc
+    *** Failers
+    bababbc
+    babababc
+
+/^(ba|b*){1,2}?bc/
+    babc
+    bbabc
+    bababc
+    *** Failers
+    bababbc
+    babababc
+
+/^\ca\cA\c[\c{\c:/
+    \x01\x01\e;z
+
+/^[ab\]cde]/
+    athing
+    bthing
+    ]thing
+    cthing
+    dthing
+    ething
+    *** Failers
+    fthing
+    [thing
+    \\thing
+
+/^[]cde]/
+    ]thing
+    cthing
+    dthing
+    ething
+    *** Failers
+    athing
+    fthing
+
+/^[^ab\]cde]/
+    fthing
+    [thing
+    \\thing
+    *** Failers
+    athing
+    bthing
+    ]thing
+    cthing
+    dthing
+    ething
+
+/^[^]cde]/
+    athing
+    fthing
+    *** Failers
+    ]thing
+    cthing
+    dthing
+    ething
+
+/^\/
+    
+
+/^/
+    
+
+/^[0-9]+$/
+    0
+    1
+    2
+    3
+    4
+    5
+    6
+    7
+    8
+    9
+    10
+    100
+    *** Failers
+    abc
+
+/^.*nter/
+    enter
+    inter
+    uponter
+
+/^xxx[0-9]+$/
+    xxx0
+    xxx1234
+    *** Failers
+    xxx
+
+/^.+[0-9][0-9][0-9]$/
+    x123
+    xx123
+    123456
+    *** Failers
+    123
+    x1234
+
+/^.+?[0-9][0-9][0-9]$/
+    x123
+    xx123
+    123456
+    *** Failers
+    123
+    x1234
+
+/^([^!]+)!(.+)=apquxz\.ixr\.zzz\.ac\.uk$/
+    abc!pqr=apquxz.ixr.zzz.ac.uk
+    *** Failers
+    !pqr=apquxz.ixr.zzz.ac.uk
+    abc!=apquxz.ixr.zzz.ac.uk
+    abc!pqr=apquxz:ixr.zzz.ac.uk
+    abc!pqr=apquxz.ixr.zzz.ac.ukk
+
+/:/
+    Well, we need a colon: somewhere
+    *** Fail if we don't
+
+/([\da-f:]+)$/i
+    0abc
+    abc
+    fed
+    E
+    ::
+    5f03:12C0::932e
+    fed def
+    Any old stuff
+    *** Failers
+    0zzz
+    gzzz
+    fed\x20
+    Any old rubbish
+
+/^.*\.(\d{1,3})\.(\d{1,3})\.(\d{1,3})$/
+    .1.2.3
+    A.12.123.0
+    *** Failers
+    .1.2.3333
+    1.2.3
+    1234.2.3
+
+/^(\d+)\s+IN\s+SOA\s+(\S+)\s+(\S+)\s*\(\s*$/
+    1 IN SOA non-sp1 non-sp2(
+    1    IN    SOA    non-sp1    non-sp2   (
+    *** Failers
+    1IN SOA non-sp1 non-sp2(
+
+/^[a-zA-Z\d][a-zA-Z\d\-]*(\.[a-zA-Z\d][a-zA-z\d\-]*)*\.$/
+    a.
+    Z.
+    2.
+    ab-c.pq-r.
+    sxk.zzz.ac.uk.
+    x-.y-.
+    *** Failers
+    -abc.peq.
+
+/^\*\.[a-z]([a-z\-\d]*[a-z\d]+)?(\.[a-z]([a-z\-\d]*[a-z\d]+)?)*$/
+    *.a
+    *.b0-a
+    *.c3-b.c
+    *.c-a.b-c
+    *** Failers
+    *.0
+    *.a-
+    *.a-b.c-
+    *.c-a.0-c
+
+/^(?=ab(de))(abd)(e)/
+    abde
+
+/^(?!(ab)de|x)(abd)(f)/
+    abdf
+
+/^(?=(ab(cd)))(ab)/
+    abcd
+
+/^[\da-f](\.[\da-f])*$/i
+    a.b.c.d
+    A.B.C.D
+    a.b.c.1.2.3.C
+
+/^\".*\"\s*(;.*)?$/
+    \"1234\"
+    \"abcd\" ;
+    \"\" ; rhubarb
+    *** Failers
+    \"1234\" : things
+
+/^$/
+    \
+    *** Failers
+
+/   ^    a   (?# begins with a)  b\sc (?# then b c) $ (?# then end)/x
+    ab c
+    *** Failers
+    abc
+    ab cde
+
+/(?x)   ^    a   (?# begins with a)  b\sc (?# then b c) $ (?# then end)/
+    ab c
+    *** Failers
+    abc
+    ab cde
+
+/^   a\ b[c ]d       $/x
+    a bcd
+    a b d
+    *** Failers
+    abcd
+    ab d
+
+/^(a(b(c)))(d(e(f)))(h(i(j)))(k(l(m)))$/
+    abcdefhijklm
+
+/^(?:a(b(c)))(?:d(e(f)))(?:h(i(j)))(?:k(l(m)))$/
+    abcdefhijklm
+
+/^[\w][\W][\s][\S][\d][\D][\b][\n][\c]][\022]/
+    a+ Z0+\x08\n\x1d\x12
+
+/^[.^$|()*+?{,}]+/
+    .^\$(*+)|{?,?}
+
+/^a*\w/
+    z
+    az
+    aaaz
+    a
+    aa
+    aaaa
+    a+
+    aa+
+
+/^a*?\w/
+    z
+    az
+    aaaz
+    a
+    aa
+    aaaa
+    a+
+    aa+
+
+/^a+\w/
+    az
+    aaaz
+    aa
+    aaaa
+    aa+
+
+/^a+?\w/
+    az
+    aaaz
+    aa
+    aaaa
+    aa+
+
+/^\d{8}\w{2,}/
+    1234567890
+    12345678ab
+    12345678__
+    *** Failers
+    1234567
+
+/^[aeiou\d]{4,5}$/
+    uoie
+    1234
+    12345
+    aaaaa
+    *** Failers
+    123456
+
+/^[aeiou\d]{4,5}?/
+    uoie
+    1234
+    12345
+    aaaaa
+    123456
+
+/\A(abc|def)=(\1){2,3}\Z/
+    abc=abcabc
+    def=defdefdef
+    *** Failers
+    abc=defdef
+
+/^(a)(b)(c)(d)(e)(f)(g)(h)(i)(j)(k)\11*(\3\4)\1(?#)2$/
+    abcdefghijkcda2
+    abcdefghijkkkkcda2
+
+/(cat(a(ract|tonic)|erpillar)) \1()2(3)/
+    cataract cataract23
+    catatonic catatonic23
+    caterpillar caterpillar23
+
+
+/^From +([^ ]+) +[a-zA-Z][a-zA-Z][a-zA-Z] +[a-zA-Z][a-zA-Z][a-zA-Z] +[0-9]?[0-9] +[0-9][0-9]:[0-9][0-9]/
+    From abcd  Mon Sep 01 12:33:02 1997
+
+/^From\s+\S+\s+([a-zA-Z]{3}\s+){2}\d{1,2}\s+\d\d:\d\d/
+    From abcd  Mon Sep 01 12:33:02 1997
+    From abcd  Mon Sep  1 12:33:02 1997
+    *** Failers
+    From abcd  Sep 01 12:33:02 1997
+
+/^12.34/s
+    12\n34
+    12\r34
+
+/\w+(?=\t)/
+    the quick brown\t fox
+
+/foo(?!bar)(.*)/
+    foobar is foolish see?
+
+/(?:(?!foo)...|^.{0,2})bar(.*)/
+    foobar crowbar etc
+    barrel
+    2barrel
+    A barrel
+
+/^(\D*)(?=\d)(?!123)/
+    abc456
+    *** Failers
+    abc123
+
+/^1234(?# test newlines
+  inside)/
+    1234
+
+/^1234 #comment in extended re
+  /x
+    1234
+
+/#rhubarb
+  abcd/x
+    abcd
+
+/^abcd#rhubarb/x
+    abcd
+
+/^(a)\1{2,3}(.)/
+    aaab
+    aaaab
+    aaaaab
+    aaaaaab
+
+/(?!^)abc/
+    the abc
+    *** Failers
+    abc
+
+/(?=^)abc/
+    abc
+    *** Failers
+    the abc
+
+/^[ab]{1,3}(ab*|b)/
+    aabbbbb
+
+/^[ab]{1,3}?(ab*|b)/
+    aabbbbb
+
+/^[ab]{1,3}?(ab*?|b)/
+    aabbbbb
+
+/^[ab]{1,3}(ab*?|b)/
+    aabbbbb
+
+/  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*                          # optional leading comment
+(?:    (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+)                    # initial word
+(?:  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+)  )* # further okay, if led by a period
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  @  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*    (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                           # initial subdomain
+(?:                                  #
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.                        # if led by a period...
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                     #   ...further okay
+)*
+# address
+|                     #  or
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+)             # one word, optionally followed by....
+(?:
+[^()<>@,;:".\\\[\]\x80-\xff\000-\010\012-\037]  |  # atom and space parts, or...
+\(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)       |  # comments, or...
+
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+# quoted strings
+)*
+<  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*                     # leading <
+(?:  @  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*    (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                           # initial subdomain
+(?:                                  #
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.                        # if led by a period...
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                     #   ...further okay
+)*
+
+(?:  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  ,  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  @  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*    (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                           # initial subdomain
+(?:                                  #
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.                        # if led by a period...
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                     #   ...further okay
+)*
+)* # further okay, if led by comma
+:                                # closing colon
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  )? #       optional route
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+)                    # initial word
+(?:  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+)  )* # further okay, if led by a period
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  @  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*    (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                           # initial subdomain
+(?:                                  #
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.                        # if led by a period...
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                     #   ...further okay
+)*
+#       address spec
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  > #                  trailing >
+# name and address
+)  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*                       # optional trailing comment
+/x
+    Alan Other 
+    
+    user\@dom.ain
+    \"A. Other\"  (a comment)
+    A. Other  (a comment)
+    \"/s=user/ou=host/o=place/prmd=uu.yy/admd= /c=gb/\"\@x400-re.lay
+    A missing angle @,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+# Atom
+|                       #  or
+"                                     # "
+[^\\\x80-\xff\n\015"] *                            #   normal
+(?:  \\ [^\x80-\xff]  [^\\\x80-\xff\n\015"] * )*        #   ( special normal* )*
+"                                     #        "
+# Quoted string
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+\.
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+# Atom
+|                       #  or
+"                                     # "
+[^\\\x80-\xff\n\015"] *                            #   normal
+(?:  \\ [^\x80-\xff]  [^\\\x80-\xff\n\015"] * )*        #   ( special normal* )*
+"                                     #        "
+# Quoted string
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# additional words
+)*
+@
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+\[                            # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*     #    stuff
+\]                           #           ]
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# optional trailing comments
+(?:
+\.
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+\[                            # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*     #    stuff
+\]                           #           ]
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# optional trailing comments
+)*
+# address
+|                             #  or
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+# Atom
+|                       #  or
+"                                     # "
+[^\\\x80-\xff\n\015"] *                            #   normal
+(?:  \\ [^\x80-\xff]  [^\\\x80-\xff\n\015"] * )*        #   ( special normal* )*
+"                                     #        "
+# Quoted string
+)
+# leading word
+[^()<>@,;:".\\\[\]\x80-\xff\000-\010\012-\037] *               # "normal" atoms and or spaces
+(?:
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+|
+"                                     # "
+[^\\\x80-\xff\n\015"] *                            #   normal
+(?:  \\ [^\x80-\xff]  [^\\\x80-\xff\n\015"] * )*        #   ( special normal* )*
+"                                     #        "
+) # "special" comment or quoted string
+[^()<>@,;:".\\\[\]\x80-\xff\000-\010\012-\037] *            #  more "normal"
+)*
+<
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# <
+(?:
+@
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+\[                            # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*     #    stuff
+\]                           #           ]
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# optional trailing comments
+(?:
+\.
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+\[                            # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*     #    stuff
+\]                           #           ]
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# optional trailing comments
+)*
+(?: ,
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+@
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+\[                            # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*     #    stuff
+\]                           #           ]
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# optional trailing comments
+(?:
+\.
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+\[                            # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*     #    stuff
+\]                           #           ]
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# optional trailing comments
+)*
+)*  # additional domains
+:
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# optional trailing comments
+)?     #       optional route
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+# Atom
+|                       #  or
+"                                     # "
+[^\\\x80-\xff\n\015"] *                            #   normal
+(?:  \\ [^\x80-\xff]  [^\\\x80-\xff\n\015"] * )*        #   ( special normal* )*
+"                                     #        "
+# Quoted string
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+\.
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+# Atom
+|                       #  or
+"                                     # "
+[^\\\x80-\xff\n\015"] *                            #   normal
+(?:  \\ [^\x80-\xff]  [^\\\x80-\xff\n\015"] * )*        #   ( special normal* )*
+"                                     #        "
+# Quoted string
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# additional words
+)*
+@
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+\[                            # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*     #    stuff
+\]                           #           ]
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# optional trailing comments
+(?:
+\.
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+\[                            # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*     #    stuff
+\]                           #           ]
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# optional trailing comments
+)*
+#       address spec
+>                    #                 >
+# name and address
+)
+/x
+    Alan Other 
+    
+    user\@dom.ain
+    \"A. Other\"  (a comment)
+    A. Other  (a comment)
+    \"/s=user/ou=host/o=place/prmd=uu.yy/admd= /c=gb/\"\@x400-re.lay
+    A missing angle ]{0,})>]{0,})>([\d]{0,}\.)(.*)((
([\w\W\s\d][^<>]{0,})|[\s]{0,}))<\/a><\/TD>]{0,})>([\w\W\s\d][^<>]{0,})<\/TD>]{0,})>([\w\W\s\d][^<>]{0,})<\/TD><\/TR>/is
+  43.Word Processor
(N-1286)Lega lstaff.comCA - Statewide
+
+/a[^a]b/
+    acb
+    a\nb
+    
+/a.b/
+    acb
+    *** Failers 
+    a\nb   
+    
+/a[^a]b/s
+    acb
+    a\nb  
+    
+/a.b/s
+    acb
+    a\nb  
+
+/^(b+?|a){1,2}?c/
+    bac
+    bbac
+    bbbac
+    bbbbac
+    bbbbbac 
+
+/^(b+|a){1,2}?c/
+    bac
+    bbac
+    bbbac
+    bbbbac
+    bbbbbac 
+    
+/(?!\A)x/m
+    x\nb\n
+    a\bx\n  
+    
+/\x0{ab}/
+    \0{ab} 
+
+/(A|B)*?CD/
+    CD 
+    
+/(A|B)*CD/
+    CD 
+
+/(AB)*?\1/
+    ABABAB
+
+/(AB)*\1/
+    ABABAB
+    
+/(?.*/)foo"
+    /this/is/a/very/long/line/in/deed/with/very/many/slashes/in/it/you/see/
+
+"(?>.*/)foo"
+    /this/is/a/very/long/line/in/deed/with/very/many/slashes/in/and/foo
+
+/(?>(\.\d\d[1-9]?))\d+/
+    1.230003938
+    1.875000282
+    *** Failers 
+    1.235 
+
+/^((?>\w+)|(?>\s+))*$/
+    now is the time for all good men to come to the aid of the party
+    *** Failers
+    this is not a line with only words and spaces!
+    
+/(\d+)(\w)/
+    12345a
+    12345+ 
+
+/((?>\d+))(\w)/
+    12345a
+    *** Failers
+    12345+ 
+
+/(?>a+)b/
+    aaab
+
+/((?>a+)b)/
+    aaab
+
+/(?>(a+))b/
+    aaab
+
+/(?>b)+/
+    aaabbbccc
+
+/(?>a+|b+|c+)*c/
+    aaabbbbccccd
+
+/((?>[^()]+)|\([^()]*\))+/
+    ((abc(ade)ufh()()x
+    
+/\(((?>[^()]+)|\([^()]+\))+\)/ 
+    (abc)
+    (abc(def)xyz)
+    *** Failers
+    ((()aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa   
+
+/a(?-i)b/i
+    ab
+    Ab
+    *** Failers 
+    aB
+    AB
+        
+/(a (?x)b c)d e/
+    a bcd e
+    *** Failers
+    a b cd e
+    abcd e   
+    a bcde 
+ 
+/(a b(?x)c d (?-x)e f)/
+    a bcde f
+    *** Failers
+    abcdef  
+
+/(a(?i)b)c/
+    abc
+    aBc
+    *** Failers
+    abC
+    aBC  
+    Abc
+    ABc
+    ABC
+    AbC
+    
+/a(?i:b)c/
+    abc
+    aBc
+    *** Failers 
+    ABC
+    abC
+    aBC
+    
+/a(?i:b)*c/
+    aBc
+    aBBc
+    *** Failers 
+    aBC
+    aBBC
+    
+/a(?=b(?i)c)\w\wd/
+    abcd
+    abCd
+    *** Failers
+    aBCd
+    abcD     
+    
+/(?s-i:more.*than).*million/i
+    more than million
+    more than MILLION
+    more \n than Million 
+    *** Failers
+    MORE THAN MILLION    
+    more \n than \n million 
+
+/(?:(?s-i)more.*than).*million/i
+    more than million
+    more than MILLION
+    more \n than Million 
+    *** Failers
+    MORE THAN MILLION    
+    more \n than \n million 
+    
+/(?>a(?i)b+)+c/ 
+    abc
+    aBbc
+    aBBc 
+    *** Failers
+    Abc
+    abAb    
+    abbC 
+    
+/(?=a(?i)b)\w\wc/
+    abc
+    aBc
+    *** Failers
+    Ab 
+    abC
+    aBC     
+    
+/(?<=a(?i)b)(\w\w)c/
+    abxxc
+    aBxxc
+    *** Failers
+    Abxxc
+    ABxxc
+    abxxC      
+
+/(?:(a)|b)(?(1)A|B)/
+    aA
+    bB
+    *** Failers
+    aB
+    bA    
+
+/^(a)?(?(1)a|b)+$/
+    aa
+    b
+    bb  
+    *** Failers
+    ab   
+
+/^(?(?=abc)\w{3}:|\d\d)$/
+    abc:
+    12
+    *** Failers
+    123
+    xyz    
+
+/^(?(?!abc)\d\d|\w{3}:)$/
+    abc:
+    12
+    *** Failers
+    123
+    xyz    
+    
+/(?(?<=foo)bar|cat)/
+    foobar
+    cat
+    fcat
+    focat   
+    *** Failers
+    foocat  
+
+/(?(?a*)*/
+    a
+    aa
+    aaaa
+    
+/(abc|)+/
+    abc
+    abcabc
+    abcabcabc
+    xyz      
+
+/([a]*)*/
+    a
+    aaaaa 
+ 
+/([ab]*)*/
+    a
+    b
+    ababab
+    aaaabcde
+    bbbb    
+ 
+/([^a]*)*/
+    b
+    bbbb
+    aaa   
+ 
+/([^ab]*)*/
+    cccc
+    abab  
+ 
+/([a]*?)*/
+    a
+    aaaa 
+ 
+/([ab]*?)*/
+    a
+    b
+    abab
+    baba   
+ 
+/([^a]*?)*/
+    b
+    bbbb
+    aaa   
+ 
+/([^ab]*?)*/
+    c
+    cccc
+    baba   
+ 
+/(?>a*)*/
+    a
+    aaabcde 
+ 
+/((?>a*))*/
+    aaaaa
+    aabbaa 
+ 
+/((?>a*?))*/
+    aaaaa
+    aabbaa 
+
+/(?(?=[^a-z]+[a-z])  \d{2}-[a-z]{3}-\d{2}  |  \d{2}-\d{2}-\d{2} ) /x
+    12-sep-98
+    12-09-98
+    *** Failers
+    sep-12-98
+        
+/(?<=(foo))bar\1/
+    foobarfoo
+    foobarfootling 
+    *** Failers
+    foobar
+    barfoo   
+
+/(?i:saturday|sunday)/
+    saturday
+    sunday
+    Saturday
+    Sunday
+    SATURDAY
+    SUNDAY
+    SunDay
+    
+/(a(?i)bc|BB)x/
+    abcx
+    aBCx
+    bbx
+    BBx
+    *** Failers
+    abcX
+    aBCX
+    bbX
+    BBX               
+
+/^([ab](?i)[cd]|[ef])/
+    ac
+    aC
+    bD
+    elephant
+    Europe 
+    frog
+    France
+    *** Failers
+    Africa     
+
+/^(ab|a(?i)[b-c](?m-i)d|x(?i)y|z)/
+    ab
+    aBd
+    xy
+    xY
+    zebra
+    Zambesi
+    *** Failers
+    aCD  
+    XY  
+
+/(?<=foo\n)^bar/m
+    foo\nbar
+    *** Failers
+    bar
+    baz\nbar   
+
+/(?<=(?]&/
+    <&OUT
+
+/^(a\1?){4}$/
+    aaaaaaaaaa
+    *** Failers
+    AB
+    aaaaaaaaa
+    aaaaaaaaaaa
+
+/^(a(?(1)\1)){4}$/
+    aaaaaaaaaa
+    *** Failers
+    aaaaaaaaa
+    aaaaaaaaaaa
+
+/(?:(f)(o)(o)|(b)(a)(r))*/
+    foobar
+
+/(?<=a)b/
+    ab
+    *** Failers
+    cb
+    b
+
+/(?a+)ab/
+
+/(?>a+)b/
+    aaab
+
+/([[:]+)/
+    a:[b]:
+
+/([[=]+)/
+    a=[b]=
+
+/([[.]+)/
+    a.[b].
+
+/((?>a+)b)/
+    aaab
+
+/(?>(a+))b/
+    aaab
+
+/((?>[^()]+)|\([^()]*\))+/
+    ((abc(ade)ufh()()x
+
+/a\Z/
+    *** Failers
+    aaab
+    a\nb\n
+
+/b\Z/
+    a\nb\n
+
+/b\z/
+
+/b\Z/
+    a\nb
+
+/b\z/
+    a\nb
+    *** Failers
+    
+/^(?>(?(1)\.|())[^\W_](?>[a-z0-9-]*[^\W_])?)+$/
+    a
+    abc
+    a-b
+    0-9 
+    a.b
+    5.6.7  
+    the.quick.brown.fox
+    a100.b200.300c  
+    12-ab.1245 
+    *** Failers
+    \
+    .a
+    -a
+    a-
+    a.  
+    a_b 
+    a.-
+    a..  
+    ab..bc 
+    the.quick.brown.fox-
+    the.quick.brown.fox.
+    the.quick.brown.fox_
+    the.quick.brown.fox+       
+
+/(?>.*)(?<=(abcd|wxyz))/
+    alphabetabcd
+    endingwxyz
+    *** Failers
+    a rather long string that doesn't end with one of them
+
+/word (?>(?:(?!otherword)[a-zA-Z0-9]+ ){0,30})otherword/
+    word cat dog elephant mussel cow horse canary baboon snake shark otherword
+    word cat dog elephant mussel cow horse canary baboon snake shark
+  
+/word (?>[a-zA-Z0-9]+ ){0,30}otherword/
+    word cat dog elephant mussel cow horse canary baboon snake shark the quick brown fox and the lazy dog and several other words getting close to thirty by now I hope
+
+/(?<=\d{3}(?!999))foo/
+    999foo
+    123999foo 
+    *** Failers
+    123abcfoo
+    
+/(?<=(?!...999)\d{3})foo/
+    999foo
+    123999foo 
+    *** Failers
+    123abcfoo
+
+/(?<=\d{3}(?!999)...)foo/
+    123abcfoo
+    123456foo 
+    *** Failers
+    123999foo  
+    
+/(?<=\d{3}...)(?\s*)=(?>\s*)        # find Z)+|A)*/
+    ZABCDEFG
+
+/((?>)+|A)*/
+    ZABCDEFG
+
+/a*/g
+    abbab
+
+/^[a-\d]/
+    abcde
+    -things
+    0digit
+    *** Failers
+    bcdef    
+
+/^[\d-a]/
+    abcde
+    -things
+    0digit
+    *** Failers
+    bcdef    
+    
+/[[:space:]]+/
+    > \x09\x0a\x0c\x0d\x0b<
+     
+/[[:blank:]]+/
+    > \x09\x0a\x0c\x0d\x0b<
+     
+/[\s]+/
+    > \x09\x0a\x0c\x0d\x0b<
+     
+/\s+/
+    > \x09\x0a\x0c\x0d\x0b<
+     
+/ab/x
+    ab
+
+/(?!\A)x/m
+  a\nxb\n
+
+/(?!^)x/m
+  a\nxb\n
+
+/abc\Qabc\Eabc/
+    abcabcabc
+    
+/abc\Q(*+|\Eabc/
+    abc(*+|abc 
+
+/   abc\Q abc\Eabc/x
+    abc abcabc
+    *** Failers
+    abcabcabc  
+    
+/abc#comment
+    \Q#not comment
+    literal\E/x
+    abc#not comment\n    literal     
+
+/abc#comment
+    \Q#not comment
+    literal/x
+    abc#not comment\n    literal     
+
+/abc#comment
+    \Q#not comment
+    literal\E #more comment
+    /x
+    abc#not comment\n    literal     
+
+/abc#comment
+    \Q#not comment
+    literal\E #more comment/x
+    abc#not comment\n    literal     
+
+/\Qabc\$xyz\E/
+    abc\\\$xyz
+
+/\Qabc\E\$\Qxyz\E/
+    abc\$xyz
+
+/\Gabc/
+    abc
+    *** Failers
+    xyzabc  
+
+/\Gabc./g
+    abc1abc2xyzabc3
+
+/abc./g
+    abc1abc2xyzabc3 
+
+/a(?x: b c )d/
+    XabcdY
+    *** Failers 
+    Xa b c d Y 
+
+/((?x)x y z | a b c)/
+    XabcY
+    AxyzB 
+
+/(?i)AB(?-i)C/
+    XabCY
+    *** Failers
+    XabcY  
+
+/((?i)AB(?-i)C|D)E/
+    abCE
+    DE
+    *** Failers
+    abcE
+    abCe  
+    dE
+    De    
+
+/(.*)\d+\1/
+    abc123abc
+    abc123bc 
+
+/(.*)\d+\1/s
+    abc123abc
+    abc123bc 
+    
+/((.*))\d+\1/
+    abc123abc
+    abc123bc  
+
+/-- This tests for an IPv6 address in the form where it can have up to
+    eight components, one and only one of which is empty. This must be
+    an internal component. --/
+
+/^(?!:)                       # colon disallowed at start
+  (?:                         # start of item
+    (?: [0-9a-f]{1,4} |       # 1-4 hex digits or
+    (?(1)0 | () ) )           # if null previously matched, fail; else null
+    :                         # followed by colon
+  ){1,7}                      # end item; 1-7 of them required               
+  [0-9a-f]{1,4} $             # final hex number at end of string
+  (?(1)|.)                    # check that there was an empty component
+  /xi
+    a123::a123
+    a123:b342::abcd
+    a123:b342::324e:abcd
+    a123:ddde:b342::324e:abcd
+    a123:ddde:b342::324e:dcba:abcd
+    a123:ddde:9999:b342::324e:dcba:abcd
+    *** Failers
+    1:2:3:4:5:6:7:8
+    a123:bce:ddde:9999:b342::324e:dcba:abcd
+    a123::9999:b342::324e:dcba:abcd
+    abcde:2:3:4:5:6:7:8
+    ::1
+    abcd:fee0:123::   
+    :1
+    1:  
+
+/[z\Qa-d]\E]/
+    z
+    a
+    -
+    d
+    ] 
+    *** Failers
+    b     
+
+/[\z\C]/
+    z
+    C 
+    
+/\M/
+    M 
+    
+/(a+)*b/
+    aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa 
+    
+/(?i)reg(?:ul(?:[a]|ae)r|ex)/
+    REGular
+    regulaer
+    Regex  
+    regulr 
+
+/[--]+/
+    
+    
+    
+    
+
+/(?<=Z)X./
+  \x84XAZXB
+
+/ab cd (?x) de fg/
+    ab cd defg
+
+/ab cd(?x) de fg/
+    ab cddefg
+    ** Failers 
+    abcddefg
+
+/(?a|)*\d/
+  aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+  aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa4
+
+/(?:a|)*\d/
+  aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+  aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa4
+
+/\Z/g
+  abc\n
+  
+/^(?s)(?>.*)(?(a))b|(a)c/
+    ac
+
+/(?=(a))ab|(a)c/
+    ac
+
+/((?>(a))b|(a)c)/
+    ac
+
+/((?>(a))b|(a)c)++/
+    ac
+
+/(?:(?>(a))b|(a)c)++/
+    ac
+
+/(?=(?>(a))b|(a)c)(..)/
+    ac
+
+/(?>(?>(a))b|(a)c)/
+    ac
+
+/(?:(?>([ab])))+a=/+
+    =ba=
+
+/(?>([ab]))+a=/+
+    =ba=
+
+/((?>(a+)b)+(aabab))/
+    aaaabaaabaabab
+
+/(?>a+|ab)+?c/
+    aabc
+
+/(?>a+|ab)+c/
+    aabc
+
+/(?:a+|ab)+c/
+    aabc
+
+/(?(?=(a))a)/
+    a
+
+/(?(?=(a))a)(b)/
+    ab
+
+/^(?:a|ab)++c/
+    aaaabc
+
+/^(?>a|ab)++c/
+    aaaabc
+
+/^(?:a|ab)+c/
+    aaaabc
+
+/(?=abc){3}abc/+
+    abcabcabc
+    ** Failers
+    xyz  
+    
+/(?=abc)+abc/+
+    abcabcabc
+    ** Failers
+    xyz  
+    
+/(?=abc)++abc/+
+    abcabcabc
+    ** Failers
+    xyz  
+    
+/(?=abc){0}xyz/
+    xyz 
+
+/(?=abc){1}xyz/
+    ** Failers
+    xyz 
+    
+/(?=(a))?./
+    ab
+    bc
+      
+/(?=(a))??./
+    ab
+    bc
+
+/^(?=(a)){0}b(?1)/
+    backgammon
+
+/^(?=(?1))?[az]([abc])d/
+    abd 
+    zcdxx 
+
+/^(?!a){0}\w+/
+    aaaaa
+
+/(?<=(abc))?xyz/
+    abcxyz
+    pqrxyz 
+
+/^[\g]+/
+    ggg<<>>
+    ** Failers
+    \\ga  
+    
+/^[\ga]+/
+    gggagagaxyz 
+    
+/^[:a[:digit:]]+/
+    aaaa444:::Z 
+
+/^[:a[:digit:]:b]+/
+    aaaa444:::bbbZ 
+
+/[:a]xxx[b:]/
+     :xxx:
+     
+/(?<=a{2})b/i
+    xaabc
+    ** Failers
+    xabc  
+
+/(?XNNNYZ
+    >  X NYQZ
+    ** Failers
+    >XYZ   
+    >  X NY Z
+
+/\v*X\v?Y\v+Z\V*\x0a\V+\x0b\V{2,3}\x0c/
+    >XY\x0aZ\x0aA\x0bNN\x0c
+    >\x0a\x0dX\x0aY\x0a\x0bZZZ\x0aAAA\x0bNNN\x0c
+
+/(foo)\Kbar/
+    foobar
+   
+/(foo)(\Kbar|baz)/
+    foobar
+    foobaz 
+
+/(foo\Kbar)baz/
+    foobarbaz
+
+/abc\K|def\K/g+
+    Xabcdefghi
+
+/ab\Kc|de\Kf/g+
+    Xabcdefghi
+    
+/(?=C)/g+
+    ABCDECBA
+    
+/^abc\K/+
+    abcdef
+    ** Failers
+    defabcxyz   
+
+/^(a(b))\1\g1\g{1}\g-1\g{-1}\g{-02}Z/
+    ababababbbabZXXXX
+
+/(?tom|bon)-\g{A}/
+    tom-tom
+    bon-bon 
+    
+/(^(a|b\g{-1}))/
+    bacxxx
+
+/(?|(abc)|(xyz))\1/
+    abcabc
+    xyzxyz 
+    ** Failers
+    abcxyz
+    xyzabc   
+    
+/(?|(abc)|(xyz))(?1)/
+    abcabc
+    xyzabc 
+    ** Failers 
+    xyzxyz 
+ 
+/^X(?5)(a)(?|(b)|(q))(c)(d)(Y)/
+    XYabcdY
+
+/^X(?7)(a)(?|(b|(r)(s))|(q))(c)(d)(Y)/
+    XYabcdY
+
+/^X(?7)(a)(?|(b|(?|(r)|(t))(s))|(q))(c)(d)(Y)/
+    XYabcdY
+
+/(?'abc'\w+):\k{2}/
+    a:aaxyz
+    ab:ababxyz
+    ** Failers
+    a:axyz
+    ab:abxyz
+
+/(?'abc'\w+):\g{abc}{2}/
+    a:aaxyz
+    ab:ababxyz
+    ** Failers
+    a:axyz
+    ab:abxyz
+
+/^(?a)? (?()b|c) (?('ab')d|e)/x
+    abd
+    ce
+
+/^(a.)\g-1Z/
+    aXaXZ
+
+/^(a.)\g{-1}Z/
+    aXaXZ
+
+/^(?(DEFINE) (? a) (? b) )  (?&A) (?&B) /x
+    abcd
+
+/(?(?&NAME_PAT))\s+(?(?&ADDRESS_PAT))
+  (?(DEFINE)
+  (?[a-z]+)
+  (?\d+)
+  )/x
+    metcalfe 33
+
+/(?(DEFINE)(?2[0-4]\d|25[0-5]|1\d\d|[1-9]?\d))\b(?&byte)(\.(?&byte)){3}/
+    1.2.3.4
+    131.111.10.206
+    10.0.0.0
+    ** Failers
+    10.6
+    455.3.4.5
+
+/\b(?&byte)(\.(?&byte)){3}(?(DEFINE)(?2[0-4]\d|25[0-5]|1\d\d|[1-9]?\d))/
+    1.2.3.4
+    131.111.10.206
+    10.0.0.0
+    ** Failers
+    10.6
+    455.3.4.5
+
+/^(\w++|\s++)*$/
+    now is the time for all good men to come to the aid of the party
+    *** Failers
+    this is not a line with only words and spaces!
+
+/(\d++)(\w)/
+    12345a
+    *** Failers
+    12345+
+
+/a++b/
+    aaab
+
+/(a++b)/
+    aaab
+
+/(a++)b/
+    aaab
+
+/([^()]++|\([^()]*\))+/
+    ((abc(ade)ufh()()x
+
+/\(([^()]++|\([^()]+\))+\)/
+    (abc)
+    (abc(def)xyz)
+    *** Failers
+    ((()aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+
+/^([^()]|\((?1)*\))*$/
+    abc
+    a(b)c
+    a(b(c))d
+    *** Failers)
+    a(b(c)d
+
+/^>abc>([^()]|\((?1)*\))*abc>123abc>1(2)3abc>(1(2)3)]*+) | (?2)) * >))/x
+    <>
+    
+     hij>
+     hij>
+    def>
+    
+    *** Failers
+    a)(?<=b(?&X))/
+    baz
+
+/^(?|(abc)|(def))\1/
+    abcabc
+    defdef 
+    ** Failers
+    abcdef
+    defabc   
+    
+/^(?|(abc)|(def))(?1)/
+    abcabc
+    defabc
+    ** Failers
+    defdef
+    abcdef    
+
+/(?:a(? (?')|(?")) |b(? (?')|(?")) ) (?('quote')[a-z]+|[0-9]+)/xJ
+    a\"aaaaa
+    b\"aaaaa 
+    ** Failers 
+    b\"11111
+
+/(?:(?1)|B)(A(*F)|C)/
+    ABCD
+    CCD
+    ** Failers
+    CAD   
+
+/^(?:(?1)|B)(A(*F)|C)/
+    CCD
+    BCD 
+    ** Failers
+    ABCD
+    CAD
+    BAD    
+
+/(?:(?1)|B)(A(*ACCEPT)XX|C)D/
+    AAD
+    ACD
+    BAD
+    BCD
+    BAX  
+    ** Failers
+    ACX
+    ABC   
+
+/(?(DEFINE)(A))B(?1)C/
+    BAC
+
+/(?(DEFINE)((A)\2))B(?1)C/
+    BAAC
+
+/(? \( ( [^()]++ | (?&pn) )* \) )/x
+    (ab(cd)ef)
+
+/^(?=a(*SKIP)b|ac)/
+    ** Failers
+    ac
+    
+/^(?=a(*PRUNE)b)/
+    ab  
+    ** Failers 
+    ac
+
+/^(?=a(*ACCEPT)b)/
+    ac
+
+/(?>a\Kb)/
+    ab
+
+/((?>a\Kb))/
+    ab
+
+/(a\Kb)/
+    ab
+    
+/^a\Kcz|ac/
+    ac
+    
+/(?>a\Kbz|ab)/
+    ab 
+
+/^(?&t)(?(DEFINE)(?a\Kb))$/
+    ab
+
+/^([^()]|\((?1)*\))*$/
+    a(b)c
+    a(b(c)d)e 
+
+/(?P(?P0)(?P>L1)|(?P>L2))/
+    0
+    00
+    0000  
+
+/(?P(?P0)|(?P>L2)(?P>L1))/
+    0
+    00
+    0000  
+
+/--- This one does fail, as expected, in Perl. It needs the complex item at the
+     end of the pattern. A single letter instead of (B|D) makes it not fail,
+     which I think is a Perl bug. --- /
+
+/A(*COMMIT)(B|D)/
+    ACABX
+
+/--- Check the use of names for failure ---/
+
+/^(A(*PRUNE:A)B|C(*PRUNE:B)D)/K
+    ** Failers
+    AC
+    CB    
+    
+/--- Force no study, otherwise mark is not seen. The studied version is in
+     test 2 because it isn't Perl-compatible. ---/
+
+/(*MARK:A)(*SKIP:B)(C|X)/KSS
+    C
+    D
+     
+/^(A(*THEN:A)B|C(*THEN:B)D)/K
+    ** Failers
+    CB    
+
+/^(?:A(*THEN:A)B|C(*THEN:B)D)/K
+    CB    
+    
+/^(?>A(*THEN:A)B|C(*THEN:B)D)/K
+    CB    
+    
+/--- This should succeed, as the skip causes bump to offset 1 (the mark). Note
+that we have to have something complicated such as (B|Z) at the end because,
+for Perl, a simple character somehow causes an unwanted optimization to mess
+with the handling of backtracking verbs. ---/
+
+/A(*MARK:A)A+(*SKIP:A)(B|Z) | AC/xK
+    AAAC
+    
+/--- Test skipping over a non-matching mark. ---/
+
+/A(*MARK:A)A+(*MARK:B)(*SKIP:A)(B|Z) | AC/xK
+    AAAC
+    
+/--- Check shorthand for MARK ---/
+
+/A(*:A)A+(*SKIP:A)(B|Z) | AC/xK
+    AAAC
+
+/--- Don't loop! Force no study, otherwise mark is not seen. ---/
+
+/(*:A)A+(*SKIP:A)(B|Z)/KSS
+    AAAC
+
+/--- This should succeed, as a non-existent skip name disables the skip ---/ 
+
+/A(*MARK:A)A+(*SKIP:B)(B|Z) | AC/xK
+    AAAC
+
+/A(*MARK:A)A+(*SKIP:B)(B|Z) | AC(*:B)/xK
+    AAAC
+
+/--- COMMIT at the start of a pattern should act like an anchor. Again, 
+however, we need the complication for Perl. ---/
+
+/(*COMMIT)(A|P)(B|P)(C|P)/
+    ABCDEFG
+    ** Failers
+    DEFGABC  
+
+/--- COMMIT inside an atomic group can't stop backtracking over the group. ---/
+
+/(\w+)(?>b(*COMMIT))\w{2}/
+    abbb
+
+/(\w+)b(*COMMIT)\w{2}/
+    abbb
+
+/--- Check opening parens in comment when seeking forward reference. ---/ 
+
+/(?&t)(?#()(?(DEFINE)(?a))/
+    bac
+
+/--- COMMIT should override THEN ---/
+
+/(?>(*COMMIT)(?>yes|no)(*THEN)(*F))?/
+  yes
+
+/(?>(*COMMIT)(yes|no)(*THEN)(*F))?/
+  yes
+
+/b?(*SKIP)c/
+    bc
+    abc
+   
+/(*SKIP)bc/
+    a
+
+/(*SKIP)b/
+    a 
+
+/(?P(?P=abn)xxx|)+/
+    xxx
+
+/(?i:([^b]))(?1)/
+    aa
+    aA     
+    ** Failers
+    ab
+    aB
+    Ba
+    ba
+
+/^(?&t)*+(?(DEFINE)(?a))\w$/
+    aaaaaaX
+    ** Failers 
+    aaaaaa 
+
+/^(?&t)*(?(DEFINE)(?a))\w$/
+    aaaaaaX
+    aaaaaa 
+
+/^(a)*+(\w)/
+    aaaaX
+    YZ 
+    ** Failers 
+    aaaa
+
+/^(?:a)*+(\w)/
+    aaaaX
+    YZ 
+    ** Failers 
+    aaaa
+
+/^(a)++(\w)/
+    aaaaX
+    ** Failers 
+    aaaa
+    YZ 
+
+/^(?:a)++(\w)/
+    aaaaX
+    ** Failers 
+    aaaa
+    YZ 
+
+/^(a)?+(\w)/
+    aaaaX
+    YZ 
+
+/^(?:a)?+(\w)/
+    aaaaX
+    YZ 
+
+/^(a){2,}+(\w)/
+    aaaaX
+    ** Failers
+    aaa
+    YZ 
+
+/^(?:a){2,}+(\w)/
+    aaaaX
+    ** Failers
+    aaa
+    YZ 
+
+/(a|)*(?1)b/
+    b
+    ab
+    aab  
+
+/(a)++(?1)b/
+    ** Failers
+    ab 
+    aab
+
+/(a)*+(?1)b/
+    ** Failers
+    ab
+    aab  
+
+/(?1)(?:(b)){0}/
+    b
+
+/(foo ( \( ((?:(?> [^()]+ )|(?2))*) \) ) )/x
+    foo(bar(baz)+baz(bop))
+
+/(A (A|B(*ACCEPT)|C) D)(E)/x
+    AB
+
+/\A.*?(a|bc)/
+    ba
+
+/\A.*?(?:a|bc)++/
+    ba
+
+/\A.*?(a|bc)++/
+    ba
+
+/\A.*?(?:a|bc|d)/
+    ba
+
+/(?:(b))++/
+    beetle
+
+/(?(?=(a(*ACCEPT)z))a)/
+    a
+
+/^(a)(?1)+ab/
+    aaaab
+    
+/^(a)(?1)++ab/
+    aaaab
+
+/^(?=a(*:M))aZ/K
+    aZbc
+
+/^(?!(*:M)b)aZ/K
+    aZbc
+
+/(?(DEFINE)(a))?b(?1)/
+    backgammon
+
+/^\N+/
+    abc\ndef
+    
+/^\N{1,}/
+    abc\ndef 
+
+/(?(R)a+|(?R)b)/
+    aaaabcde
+
+/(?(R)a+|((?R))b)/
+    aaaabcde
+
+/((?(R)a+|(?1)b))/
+    aaaabcde
+
+/((?(R1)a+|(?1)b))/
+    aaaabcde
+
+/a(*:any 
+name)/K
+    abc
+    
+/(?>(?&t)c|(?&t))(?(DEFINE)(?a|b(*PRUNE)c))/
+    a
+    ba
+    bba 
+    
+/--- Checking revised (*THEN) handling ---/ 
+
+/--- Capture ---/
+
+/^.*? (a(*THEN)b) c/x
+    aabc
+
+/^.*? (a(*THEN)b|(*F)) c/x
+    aabc
+
+/^.*? ( (a(*THEN)b) | (*F) ) c/x
+    aabc
+
+/^.*? ( (a(*THEN)b) ) c/x
+    aabc
+
+/--- Non-capture ---/
+
+/^.*? (?:a(*THEN)b) c/x
+    aabc
+
+/^.*? (?:a(*THEN)b|(*F)) c/x
+    aabc
+
+/^.*? (?: (?:a(*THEN)b) | (*F) ) c/x
+    aabc
+
+/^.*? (?: (?:a(*THEN)b) ) c/x
+    aabc
+
+/--- Atomic ---/
+
+/^.*? (?>a(*THEN)b) c/x
+    aabc
+
+/^.*? (?>a(*THEN)b|(*F)) c/x
+    aabc
+
+/^.*? (?> (?>a(*THEN)b) | (*F) ) c/x
+    aabc
+
+/^.*? (?> (?>a(*THEN)b) ) c/x
+    aabc
+
+/--- Possessive capture ---/
+
+/^.*? (a(*THEN)b)++ c/x
+    aabc
+
+/^.*? (a(*THEN)b|(*F))++ c/x
+    aabc
+
+/^.*? ( (a(*THEN)b)++ | (*F) )++ c/x
+    aabc
+
+/^.*? ( (a(*THEN)b)++ )++ c/x
+    aabc
+
+/--- Possessive non-capture ---/
+
+/^.*? (?:a(*THEN)b)++ c/x
+    aabc
+
+/^.*? (?:a(*THEN)b|(*F))++ c/x
+    aabc
+
+/^.*? (?: (?:a(*THEN)b)++ | (*F) )++ c/x
+    aabc
+
+/^.*? (?: (?:a(*THEN)b)++ )++ c/x
+    aabc
+    
+/--- Condition assertion ---/
+
+/^(?(?=a(*THEN)b)ab|ac)/
+    ac
+ 
+/--- Condition ---/
+
+/^.*?(?(?=a)a|b(*THEN)c)/
+    ba
+
+/^.*?(?:(?(?=a)a|b(*THEN)c)|d)/
+    ba
+
+/^.*?(?(?=a)a(*THEN)b|c)/
+    ac
+
+/--- Assertion ---/
+
+/^.*(?=a(*THEN)b)/ 
+    aabc
+
+/------------------------------/
+
+/(?>a(*:m))/imsxSK 
+    a
+
+/(?>(a)(*:m))/imsxSK 
+    a
+
+/(?<=a(*ACCEPT)b)c/
+    xacd
+
+/(?<=(a(*ACCEPT)b))c/
+    xacd
+
+/(?<=(a(*COMMIT)b))c/
+    xabcd
+    ** Failers 
+    xacd
+    
+/(?a?)*)*c/
+  aac   
+
+/(?>.*?a)(?<=ba)/
+    aba
+
+/(?:.*?a)(?<=ba)/
+    aba
+
+/.*?a(*PRUNE)b/
+    aab
+
+/.*?a(*PRUNE)b/s
+    aab
+
+/^a(*PRUNE)b/s
+    aab
+
+/.*?a(*SKIP)b/
+    aab
+
+/(?>.*?a)b/s
+    aab
+
+/(?>.*?a)b/
+    aab
+
+/(?>^a)b/s
+    aab
+
+/(?>.*?)(?<=(abcd)|(wxyz))/
+    alphabetabcd
+    endingwxyz 
+
+/(?>.*)(?<=(abcd)|(wxyz))/
+    alphabetabcd
+    endingwxyz 
+
+"(?>.*)foo"
+    abcdfooxyz
+    
+"(?>.*?)foo"
+    abcdfooxyz
+
+/(?:(a(*PRUNE)b)){0}(?:(?1)|ac)/
+    ac
+    
+/(?:(a(*SKIP)b)){0}(?:(?1)|ac)/
+    ac 
+
+/(?<=(*SKIP)ac)a/
+    aa
+
+/A(*MARK:A)A+(*SKIP:B)(B|Z) | AC/xK
+    AAAC
+
+/a(*SKIP:m)x|ac(*:n)(*SKIP:n)d|ac/K
+    acacd
+
+/A(*SKIP:m)x|A(*SKIP:n)x|AB/K
+    AB
+
+/((*SKIP:r)d){0}a(*SKIP:m)x|ac(*:n)|ac/K
+    acacd
+
+/-- Tests that try to figure out how Perl works. My hypothesis is that the
+    first verb that is backtracked onto is the one that acts. This seems to be
+    the case almost all the time, but there is one exception that is perhaps a 
+    bug. --/
+
+/-- This matches "aaaac"; each PRUNE advances one character until the subject
+    no longer starts with 5 'a's. --/
+
+/aaaaa(*PRUNE)b|a+c/
+    aaaaaac
+
+/-- Putting SKIP in front of PRUNE makes no difference, as it is never 
+backtracked onto, whether or not it has a label. --/
+
+/aaaaa(*SKIP)(*PRUNE)b|a+c/
+    aaaaaac
+
+/aaaaa(*SKIP:N)(*PRUNE)b|a+c/
+    aaaaaac
+
+/aaaa(*:N)a(*SKIP:N)(*PRUNE)b|a+c/
+    aaaaaac
+
+/-- Putting THEN in front makes no difference. */
+    
+/aaaaa(*THEN)(*PRUNE)b|a+c/
+    aaaaaac
+ 
+/-- However, putting COMMIT in front of the prune changes it to "no match". I 
+    think this is inconsistent and possibly a bug. For the moment, running this
+    test is moved out of the Perl-compatible file. --/
+
+/aaaaa(*COMMIT)(*PRUNE)b|a+c/
+    
+
+/---- OK, lets play the same game again using SKIP instead of PRUNE. ----/
+
+/-- This matches "ac" because SKIP forces the next match to start on the
+    sixth "a". --/
+
+/aaaaa(*SKIP)b|a+c/
+    aaaaaac
+ 
+/-- Putting PRUNE in front makes no difference. --/
+
+/aaaaa(*PRUNE)(*SKIP)b|a+c/
+    aaaaaac
+
+/-- Putting THEN in front makes no difference. --/
+
+/aaaaa(*THEN)(*SKIP)b|a+c/
+    aaaaaac
+
+/-- In this case, neither does COMMIT. This still matches "ac". --/
+
+/aaaaa(*COMMIT)(*SKIP)b|a+c/
+    aaaaaac
+    
+/-- This gives "no match", as expected. --/
+
+/aaaaa(*COMMIT)b|a+c/
+    aaaaaac
+    
+
+/------ Tests using THEN ------/
+
+/-- This matches "aaaaaac", as expected. --/
+
+/aaaaa(*THEN)b|a+c/
+    aaaaaac
+
+/-- Putting SKIP in front makes no difference. --/
+
+/aaaaa(*SKIP)(*THEN)b|a+c/
+    aaaaaac
+    
+/-- Putting PRUNE in front makes no difference. --/
+
+/aaaaa(*PRUNE)(*THEN)b|a+c/
+    aaaaaac
+    
+/-- Putting COMMIT in front makes no difference. --/
+
+/aaaaa(*COMMIT)(*THEN)b|a+c/
+    aaaaaac
+    
+/-- End of "priority" tests --/ 
+
+/aaaaa(*:m)(*PRUNE:m)(*SKIP:m)m|a+/
+    aaaaaa
+
+/aaaaa(*:m)(*MARK:m)(*PRUNE)(*SKIP:m)m|a+/
+    aaaaaa
+
+/aaaaa(*:n)(*PRUNE:m)(*SKIP:m)m|a+/
+    aaaaaa
+
+/aaaaa(*:n)(*MARK:m)(*PRUNE)(*SKIP:m)m|a+/
+    aaaaaa
+
+/a(*MARK:A)aa(*PRUNE:A)a(*SKIP:A)b|a+c/
+    aaaac
+
+/a(*MARK:A)aa(*MARK:A)a(*SKIP:A)b|a+c/
+    aaaac
+
+/aaa(*PRUNE:A)a(*SKIP:A)b|a+c/
+    aaaac
+
+/aaa(*MARK:A)a(*SKIP:A)b|a+c/
+    aaaac
+
+/a(*:m)a(*COMMIT)(*SKIP:m)b|a+c/K
+    aaaaaac
+
+/.?(a|b(*THEN)c)/
+    ba
+
+/(a(*COMMIT)b)c|abd/
+    abc
+    abd
+
+/(?=a(*COMMIT)b)abc|abd/
+    abc
+    abd
+
+/(?>a(*COMMIT)b)c|abd/
+    abc
+    abd
+
+/a(?=b(*COMMIT)c)[^d]|abd/
+    abd
+    abc 
+
+/a(?=bc).|abd/
+    abd
+    abc 
+    
+/a(?>b(*COMMIT)c)d|abd/
+    abceabd 
+
+/a(?>bc)d|abd/
+    abceabd 
+
+/(?>a(*COMMIT)b)c|abd/
+    abd
+
+/(?>a(*COMMIT)c)d|abd/
+    abd
+
+/((?=a(*COMMIT)b)ab|ac){0}(?:(?1)|a(c))/
+    ac 
+    
+/-- These tests were formerly in test 2, but changes in PCRE and Perl have
+    made them compatible. --/
+    
+/^(a)?(?(1)a|b)+$/
+    *** Failers
+    a
+
+/(?=a\Kb)ab/
+    ab 
+
+/(?!a\Kb)ac/
+    ac 
+    
+/^abc(?<=b\Kc)d/
+    abcd
+
+/^abc(?b))/K
+    abc
+
+/(*PRUNE:m(m)(?&y)(?(DEFINE)(?b))/K
+    abc
+
+/(*SKIP:m(m)(?&y)(?(DEFINE)(?b))/K
+    abc
+
+/(*THEN:m(m)(?&y)(?(DEFINE)(?b))/K
+    abc
+
+/^\d*\w{4}/
+    1234
+    123 
+
+/^[^b]*\w{4}/
+    aaaa
+    aaa     
+
+/^[^b]*\w{4}/i
+    aaaa
+    aaa     
+
+/^a*\w{4}/
+    aaaa
+    aaa     
+
+/^a*\w{4}/i
+    aaaa
+    aaa     
+
+/-- End of testinput1 --/
diff --git a/pcre/testdata/testinput10 b/pcre/testdata/testinput10
new file mode 100644
index 00000000000..7f522e6af8f
--- /dev/null
+++ b/pcre/testdata/testinput10
@@ -0,0 +1,1401 @@
+/-- This set of tests check Unicode property support with the DFA matching 
+    functionality of pcre_dfa_exec(). The -dfa flag must be used with pcretest
+    when running it. --/
+
+/\pL\P{Nd}/8
+    AB
+    *** Failers
+    A0
+    00   
+
+/\X./8
+    AB
+    A\x{300}BC 
+    A\x{300}\x{301}\x{302}BC 
+    *** Failers
+    \x{300}  
+
+/\X\X/8
+    ABC
+    A\x{300}B\x{300}\x{301}C 
+    A\x{300}\x{301}\x{302}BC 
+    *** Failers
+    \x{300}  
+
+/^\pL+/8
+    abcd
+    a 
+    *** Failers 
+
+/^\PL+/8
+    1234
+    = 
+    *** Failers 
+    abcd 
+
+/^\X+/8
+    abcdA\x{300}\x{301}\x{302}
+    A\x{300}\x{301}\x{302}
+    A\x{300}\x{301}\x{302}A\x{300}\x{301}\x{302}
+    a 
+    *** Failers 
+    \x{300}\x{301}\x{302}
+
+/\X?abc/8
+    abc
+    A\x{300}abc
+    A\x{300}\x{301}\x{302}A\x{300}A\x{300}A\x{300}abcxyz
+    \x{300}abc  
+    *** Failers
+
+/^\X?abc/8
+    abc
+    A\x{300}abc
+    *** Failers
+    A\x{300}\x{301}\x{302}A\x{300}A\x{300}A\x{300}abcxyz
+    \x{300}abc  
+
+/\X*abc/8
+    abc
+    A\x{300}abc
+    A\x{300}\x{301}\x{302}A\x{300}A\x{300}A\x{300}abcxyz
+    \x{300}abc  
+    *** Failers
+
+/^\X*abc/8
+    abc
+    A\x{300}abc
+    A\x{300}\x{301}\x{302}A\x{300}A\x{300}A\x{300}abcxyz
+    *** Failers
+    \x{300}abc  
+
+/^\pL?=./8
+    A=b
+    =c 
+    *** Failers
+    1=2 
+    AAAA=b  
+
+/^\pL*=./8
+    AAAA=b
+    =c 
+    *** Failers
+    1=2  
+
+/^\X{2,3}X/8
+    A\x{300}\x{301}\x{302}A\x{300}\x{301}\x{302}X
+    A\x{300}\x{301}\x{302}A\x{300}\x{301}\x{302}A\x{300}\x{301}\x{302}X 
+    *** Failers
+    X
+    A\x{300}\x{301}\x{302}X
+    A\x{300}\x{301}\x{302}A\x{300}\x{301}\x{302}A\x{300}\x{301}\x{302}A\x{300}\x{301}\x{302}X
+
+/^\pC\pL\pM\pN\pP\pS\pZ\p{Xsp}/8
+    >\x{1680}\x{2028}\x{0b}
+    ** Failers
+    \x{0b} 
+
+/^>\p{Xsp}+/8
+    > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+
+/^>\p{Xsp}*/8
+    > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+    
+/^>\p{Xsp}{2,9}/8
+    > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+    
+/^>[\p{Xsp}]/8
+    >\x{2028}\x{0b}
+ 
+/^>[\p{Xsp}]+/8
+    > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+
+/^>\p{Xps}/8
+    >\x{1680}\x{2028}\x{0b}
+    >\x{a0} 
+    ** Failers
+    \x{0b} 
+
+/^>\p{Xps}+/8
+    > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+
+/^>\p{Xps}+?/8
+    >\x{1680}\x{2028}\x{0b}
+
+/^>\p{Xps}*/8
+    > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+    
+/^>\p{Xps}{2,9}/8
+    > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+    
+/^>\p{Xps}{2,9}?/8
+    > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+    
+/^>[\p{Xps}]/8
+    >\x{2028}\x{0b}
+ 
+/^>[\p{Xps}]+/8
+    > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+
+/^\p{Xwd}/8
+    ABCD
+    1234
+    \x{6ca}
+    \x{a6c}
+    \x{10a7}
+    _ABC    
+    ** Failers
+    [] 
+
+/^\p{Xwd}+/8
+    ABCD1234\x{6ca}\x{a6c}\x{10a7}_
+
+/^\p{Xwd}*/8
+    ABCD1234\x{6ca}\x{a6c}\x{10a7}_
+    
+/^\p{Xwd}{2,9}/8
+    A_12\x{6ca}\x{a6c}\x{10a7}
+    
+/^[\p{Xwd}]/8
+    ABCD1234_
+    1234abcd_
+    \x{6ca}
+    \x{a6c}
+    \x{10a7}   
+    _ABC 
+    ** Failers
+    []   
+ 
+/^[\p{Xwd}]+/8
+    ABCD1234\x{6ca}\x{a6c}\x{10a7}_
+
+/-- Unicode properties for \b abd \B --/
+
+/\b...\B/8W
+    abc_
+    \x{37e}abc\x{376} 
+    \x{37e}\x{376}\x{371}\x{393}\x{394} 
+    !\x{c0}++\x{c1}\x{c2} 
+    !\x{c0}+++++ 
+
+/-- Without PCRE_UCP, non-ASCII always fail, even if < 256  --/
+
+/\b...\B/8
+    abc_
+    ** Failers 
+    \x{37e}abc\x{376} 
+    \x{37e}\x{376}\x{371}\x{393}\x{394} 
+    !\x{c0}++\x{c1}\x{c2} 
+    !\x{c0}+++++ 
+
+/-- With PCRE_UCP, non-UTF8 chars that are < 256 still check properties  --/
+
+/\b...\B/W
+    abc_
+    !\x{c0}++\x{c1}\x{c2} 
+    !\x{c0}+++++ 
+    
+/-- Caseless single negated characters > 127 need UCP support --/
+
+/[^\x{100}]/8i
+    \x{100}\x{101}X
+
+/[^\x{100}]+/8i
+    \x{100}\x{101}XX
+
+/^\X/8
+    A\P
+    A\P\P 
+    A\x{300}\x{301}\P
+    A\x{300}\x{301}\P\P  
+    A\x{301}\P
+    A\x{301}\P\P  
+    
+/^\X{2,3}/8
+    A\P
+    A\P\P 
+    AA\P
+    AA\P\P  
+    A\x{300}\x{301}\P
+    A\x{300}\x{301}\P\P  
+    A\x{300}\x{301}A\x{300}\x{301}\P
+    A\x{300}\x{301}A\x{300}\x{301}\P\P  
+
+/^\X{2}/8
+    AA\P
+    AA\P\P  
+    A\x{300}\x{301}A\x{300}\x{301}\P
+    A\x{300}\x{301}A\x{300}\x{301}\P\P  
+    
+/^\X+/8
+    AA\P
+    AA\P\P  
+
+/^\X+?Z/8
+    AA\P
+    AA\P\P 
+
+/-- These are tests for extended grapheme clusters --/ 
+
+/^\X/8+
+    G\x{34e}\x{34e}X
+    \x{34e}\x{34e}X
+    \x04X
+    \x{1100}X
+    \x{1100}\x{34e}X
+    \x{1b04}\x{1b04}X 
+    *These match up to the roman letters
+    \x{1111}\x{1111}L,L
+    \x{1111}\x{1111}\x{1169}L,L,V
+    \x{1111}\x{ae4c}L, LV
+    \x{1111}\x{ad89}L, LVT
+    \x{1111}\x{ae4c}\x{1169}L, LV, V
+    \x{1111}\x{ae4c}\x{1169}\x{1169}L, LV, V, V
+    \x{1111}\x{ae4c}\x{1169}\x{11fe}L, LV, V, T
+    \x{1111}\x{ad89}\x{11fe}L, LVT, T
+    \x{1111}\x{ad89}\x{11fe}\x{11fe}L, LVT, T, T
+    \x{ad89}\x{11fe}\x{11fe}LVT, T, T
+    *These match just the first codepoint (invalid sequence)
+    \x{1111}\x{11fe}L, T
+    \x{ae4c}\x{1111}LV, L
+    \x{ae4c}\x{ae4c}LV, LV
+    \x{ae4c}\x{ad89}LV, LVT
+    \x{1169}\x{1111}V, L
+    \x{1169}\x{ae4c}V, LV
+    \x{1169}\x{ad89}V, LVT
+    \x{ad89}\x{1111}LVT, L
+    \x{ad89}\x{1169}LVT, V
+    \x{ad89}\x{ae4c}LVT, LV
+    \x{ad89}\x{ad89}LVT, LVT
+    \x{11fe}\x{1111}T, L
+    \x{11fe}\x{1169}T, V
+    \x{11fe}\x{ae4c}T, LV
+    \x{11fe}\x{ad89}T, LVT
+    *Test extend and spacing mark
+    \x{1111}\x{ae4c}\x{0711}L, LV, extend
+    \x{1111}\x{ae4c}\x{1b04}L, LV, spacing mark
+    \x{1111}\x{ae4c}\x{1b04}\x{0711}\x{1b04}L, LV, spacing mark, extend, spacing mark
+    *Test CR, LF, and control
+    \x0d\x{0711}CR, extend
+    \x0d\x{1b04}CR, spacingmark
+    \x0a\x{0711}LF, extend
+    \x0a\x{1b04}LF, spacingmark
+    \x0b\x{0711}Control, extend
+    \x09\x{1b04}Control, spacingmark
+    *There are no Prepend characters, so we can't test Prepend, CR
+    
+/^(?>\X{2})X/8+
+    \x{1111}\x{ae4c}\x{1111}\x{ae4c}X
+    
+/^\X{2,4}X/8+
+    \x{1111}\x{ae4c}\x{1111}\x{ae4c}X
+    \x{1111}\x{ae4c}\x{1111}\x{ae4c}\x{1111}\x{ae4c}X
+    \x{1111}\x{ae4c}\x{1111}\x{ae4c}\x{1111}\x{ae4c}\x{1111}\x{ae4c}X
+
+/^\X{2,4}?X/8+
+    \x{1111}\x{ae4c}\x{1111}\x{ae4c}X
+    \x{1111}\x{ae4c}\x{1111}\x{ae4c}\x{1111}\x{ae4c}X
+    \x{1111}\x{ae4c}\x{1111}\x{ae4c}\x{1111}\x{ae4c}\x{1111}\x{ae4c}X
+
+/-- --/
+
+/\x{1e9e}+/8i
+    \x{1e9e}\x{00df}
+
+/[z\x{1e9e}]+/8i
+    \x{1e9e}\x{00df}
+
+/\x{00df}+/8i
+    \x{1e9e}\x{00df}
+
+/[z\x{00df}]+/8i
+    \x{1e9e}\x{00df}
+
+/\x{1f88}+/8i
+    \x{1f88}\x{1f80} 
+
+/[z\x{1f88}]+/8i
+    \x{1f88}\x{1f80} 
+
+/-- Perl matches these --/
+
+/\x{00b5}+/8i
+    \x{00b5}\x{039c}\x{03bc}
+
+/\x{039c}+/8i
+    \x{00b5}\x{039c}\x{03bc}
+
+/\x{03bc}+/8i
+    \x{00b5}\x{039c}\x{03bc}
+
+
+/\x{00c5}+/8i
+    \x{00c5}\x{00e5}\x{212b}
+
+/\x{00e5}+/8i
+    \x{00c5}\x{00e5}\x{212b}
+
+/\x{212b}+/8i
+    \x{00c5}\x{00e5}\x{212b}
+
+
+/\x{01c4}+/8i
+    \x{01c4}\x{01c5}\x{01c6}
+
+/\x{01c5}+/8i
+    \x{01c4}\x{01c5}\x{01c6}
+
+/\x{01c6}+/8i
+    \x{01c4}\x{01c5}\x{01c6}
+
+
+/\x{01c7}+/8i
+    \x{01c7}\x{01c8}\x{01c9}
+
+/\x{01c8}+/8i
+    \x{01c7}\x{01c8}\x{01c9}
+
+/\x{01c9}+/8i
+    \x{01c7}\x{01c8}\x{01c9}
+
+
+/\x{01ca}+/8i
+    \x{01ca}\x{01cb}\x{01cc}
+
+/\x{01cb}+/8i
+    \x{01ca}\x{01cb}\x{01cc}
+
+/\x{01cc}+/8i
+    \x{01ca}\x{01cb}\x{01cc}
+
+
+/\x{01f1}+/8i
+    \x{01f1}\x{01f2}\x{01f3}
+
+/\x{01f2}+/8i
+    \x{01f1}\x{01f2}\x{01f3}
+
+/\x{01f3}+/8i
+    \x{01f1}\x{01f2}\x{01f3}
+
+
+/\x{0345}+/8i
+    \x{0345}\x{0399}\x{03b9}\x{1fbe}
+
+/\x{0399}+/8i
+    \x{0345}\x{0399}\x{03b9}\x{1fbe}
+
+/\x{03b9}+/8i
+    \x{0345}\x{0399}\x{03b9}\x{1fbe}
+
+/\x{1fbe}+/8i
+    \x{0345}\x{0399}\x{03b9}\x{1fbe}
+
+
+/\x{0392}+/8i
+    \x{0392}\x{03b2}\x{03d0}
+
+/\x{03b2}+/8i
+    \x{0392}\x{03b2}\x{03d0}
+
+/\x{03d0}+/8i
+    \x{0392}\x{03b2}\x{03d0}
+    
+
+/\x{0395}+/8i
+    \x{0395}\x{03b5}\x{03f5}
+
+/\x{03b5}+/8i
+    \x{0395}\x{03b5}\x{03f5}
+
+/\x{03f5}+/8i
+    \x{0395}\x{03b5}\x{03f5}
+
+
+/\x{0398}+/8i
+    \x{0398}\x{03b8}\x{03d1}\x{03f4}
+
+/\x{03b8}+/8i
+    \x{0398}\x{03b8}\x{03d1}\x{03f4}
+
+/\x{03d1}+/8i
+    \x{0398}\x{03b8}\x{03d1}\x{03f4}
+
+/\x{03f4}+/8i
+    \x{0398}\x{03b8}\x{03d1}\x{03f4}
+    
+
+/\x{039a}+/8i
+    \x{039a}\x{03ba}\x{03f0}
+
+/\x{03ba}+/8i
+    \x{039a}\x{03ba}\x{03f0}
+
+/\x{03f0}+/8i
+    \x{039a}\x{03ba}\x{03f0}
+    
+
+/\x{03a0}+/8i
+    \x{03a0}\x{03c0}\x{03d6} 
+
+/\x{03c0}+/8i
+    \x{03a0}\x{03c0}\x{03d6} 
+
+/\x{03d6}+/8i
+    \x{03a0}\x{03c0}\x{03d6} 
+
+
+/\x{03a1}+/8i
+    \x{03a1}\x{03c1}\x{03f1}
+
+/\x{03c1}+/8i
+    \x{03a1}\x{03c1}\x{03f1}
+
+/\x{03f1}+/8i
+    \x{03a1}\x{03c1}\x{03f1}
+
+
+/\x{03a3}+/8i
+    \x{03A3}\x{03C2}\x{03C3}
+
+/\x{03c2}+/8i
+    \x{03A3}\x{03C2}\x{03C3}
+
+/\x{03c3}+/8i
+    \x{03A3}\x{03C2}\x{03C3}
+    
+
+/\x{03a6}+/8i
+    \x{03a6}\x{03c6}\x{03d5} 
+
+/\x{03c6}+/8i
+    \x{03a6}\x{03c6}\x{03d5} 
+
+/\x{03d5}+/8i
+    \x{03a6}\x{03c6}\x{03d5} 
+
+
+/\x{03c9}+/8i
+    \x{03c9}\x{03a9}\x{2126}
+
+/\x{03a9}+/8i
+    \x{03c9}\x{03a9}\x{2126}
+
+/\x{2126}+/8i
+    \x{03c9}\x{03a9}\x{2126}
+    
+
+/\x{1e60}+/8i
+    \x{1e60}\x{1e61}\x{1e9b}
+
+/\x{1e61}+/8i
+    \x{1e60}\x{1e61}\x{1e9b}
+
+/\x{1e9b}+/8i
+    \x{1e60}\x{1e61}\x{1e9b}
+    
+
+/\x{1e9e}+/8i
+    \x{1e9e}\x{00df}
+
+/\x{00df}+/8i
+    \x{1e9e}\x{00df}
+    
+
+/\x{1f88}+/8i
+    \x{1f88}\x{1f80} 
+
+/\x{1f80}+/8i
+    \x{1f88}\x{1f80} 
+
+/\x{004b}+/8i
+    \x{004b}\x{006b}\x{212a}
+
+/\x{006b}+/8i
+    \x{004b}\x{006b}\x{212a}
+
+/\x{212a}+/8i
+    \x{004b}\x{006b}\x{212a}
+
+
+/\x{0053}+/8i
+    \x{0053}\x{0073}\x{017f}
+
+/\x{0073}+/8i
+    \x{0053}\x{0073}\x{017f}
+
+/\x{017f}+/8i
+    \x{0053}\x{0073}\x{017f}
+
+/ist/8i
+    ikt
+
+/is+t/8i
+    iSs\x{17f}t
+    ikt
+
+/is+?t/8i
+    ikt
+
+/is?t/8i
+    ikt
+
+/is{2}t/8i
+    iskt
+
+/^\p{Xuc}/8
+    $abc
+    @abc
+    `abc
+    \x{1234}abc
+    ** Failers
+    abc     
+
+/^\p{Xuc}+/8
+    $@`\x{a0}\x{1234}\x{e000}**
+    ** Failers
+    \x{9f}
+
+/^\p{Xuc}+?/8
+    $@`\x{a0}\x{1234}\x{e000}**
+    ** Failers
+    \x{9f}
+
+/^\p{Xuc}+?\*/8
+    $@`\x{a0}\x{1234}\x{e000}**
+    ** Failers
+    \x{9f}
+
+/^\p{Xuc}++/8
+    $@`\x{a0}\x{1234}\x{e000}**
+    ** Failers
+    \x{9f}
+
+/^\p{Xuc}{3,5}/8
+    $@`\x{a0}\x{1234}\x{e000}**
+    ** Failers
+    \x{9f}
+
+/^\p{Xuc}{3,5}?/8
+    $@`\x{a0}\x{1234}\x{e000}**
+    ** Failers
+    \x{9f}
+
+/^[\p{Xuc}]/8
+    $@`\x{a0}\x{1234}\x{e000}**
+    ** Failers
+    \x{9f}
+
+/^[\p{Xuc}]+/8
+    $@`\x{a0}\x{1234}\x{e000}**
+    ** Failers
+    \x{9f}
+
+/^\P{Xuc}/8
+    abc
+    ** Failers
+    $abc
+    @abc
+    `abc
+    \x{1234}abc
+
+/^[\P{Xuc}]/8
+    abc
+    ** Failers
+    $abc
+    @abc
+    `abc
+    \x{1234}abc
+
+/-- End of testinput10 --/ 
diff --git a/pcre/testdata/testinput11 b/pcre/testdata/testinput11
new file mode 100644
index 00000000000..391ada7aa8c
--- /dev/null
+++ b/pcre/testdata/testinput11
@@ -0,0 +1,135 @@
+/-- These are a few representative patterns whose lengths and offsets are to be 
+shown when the link size is 2. This is just a doublecheck test to ensure the 
+sizes don't go horribly wrong when something is changed. The pattern contents 
+are all themselves checked in other tests. Unicode, including property support, 
+is required for these tests. --/
+
+/((?i)b)/BM
+
+/(?s)(.*X|^B)/BM
+
+/(?s:.*X|^B)/BM
+
+/^[[:alnum:]]/BM
+
+/#/IxMD
+
+/a#/IxMD
+
+/x?+/BM
+
+/x++/BM
+
+/x{1,3}+/BM 
+
+/(x)*+/BM
+
+/^((a+)(?U)([ab]+)(?-U)([bc]+)(\w*))/BM
+
+|8J\$WE\<\.rX\+ix\[d1b\!H\#\?vV0vrK\:ZH1\=2M\>iV\;\?aPhFB\<\*vW\@QW\@sO9\}cfZA\-i\'w\%hKd6gt1UJP\,15_\#QY\$M\^Mss_U\/\]\&LK9\[5vQub\^w\[KDD\qmj\;2\}YWFdYx\.Ap\]hjCPTP\(n28k\+3\;o\&WXqs\/gOXdr\$\:r\'do0\;b4c\(f_Gr\=\"\\4\)\[01T7ajQJvL\$W\~mL_sS\/4h\:x\*\[ZN\=KLs\&L5zX\/\/\>it\,o\:aU\(\;Z\>pW\&T7oP\'2K\^E\:x9\'c\[\%z\-\,64JQ5AeH_G\#KijUKghQw\^\\vea3a\?kka_G\$8\#\`\*kynsxzBLru\'\]k_\[7FrVx\}\^\=\$blx\>s\-N\%j\;D\*aZDnsw\:YKZ\%Q\.Kne9\#hP\?\+b3\(SOvL\,\^\;\&u5\@\?5C5Bhb\=m\-vEh_L15Jl\]U\)0RP6\{q\%L\^_z5E\'Dw6X\b|BM
+
+|\$\<\.X\+ix\[d1b\!H\#\?vV0vrK\:ZH1\=2M\>iV\;\?aPhFB\<\*vW\@QW\@sO9\}cfZA\-i\'w\%hKd6gt1UJP\,15_\#QY\$M\^Mss_U\/\]\&LK9\[5vQub\^w\[KDD\qmj\;2\}YWFdYx\.Ap\]hjCPTP\(n28k\+3\;o\&WXqs\/gOXdr\$\:r\'do0\;b4c\(f_Gr\=\"\\4\)\[01T7ajQJvL\$W\~mL_sS\/4h\:x\*\[ZN\=KLs\&L5zX\/\/\>it\,o\:aU\(\;Z\>pW\&T7oP\'2K\^E\:x9\'c\[\%z\-\,64JQ5AeH_G\#KijUKghQw\^\\vea3a\?kka_G\$8\#\`\*kynsxzBLru\'\]k_\[7FrVx\}\^\=\$blx\>s\-N\%j\;D\*aZDnsw\:YKZ\%Q\.Kne9\#hP\?\+b3\(SOvL\,\^\;\&u5\@\?5C5Bhb\=m\-vEh_L15Jl\]U\)0RP6\{q\%L\^_z5E\'Dw6X\b|BM
+
+/(a(?1)b)/BM
+
+/(a(?1)+b)/BM
+
+/a(?Pb|c)d(?Pe)/BM
+
+/(?:a(?Pc(?Pd)))(?Pa)/BM
+
+/(?Pa)...(?P=a)bbb(?P>a)d/BM
+
+/abc(?C255)de(?C)f/BM
+
+/abcde/CBM
+
+/\x{100}/8BM
+
+/\x{1000}/8BM
+
+/\x{10000}/8BM
+
+/\x{100000}/8BM
+
+/\x{10ffff}/8BM
+
+/\x{110000}/8BM
+
+/[\x{ff}]/8BM
+
+/[\x{100}]/8BM
+
+/\x80/8BM
+
+/\xff/8BM
+
+/\x{0041}\x{2262}\x{0391}\x{002e}/D8M
+    
+/\x{D55c}\x{ad6d}\x{C5B4}/D8M 
+
+/\x{65e5}\x{672c}\x{8a9e}/D8M
+
+/[\x{100}]/8BM
+
+/[Z\x{100}]/8BM
+
+/^[\x{100}\E-\Q\E\x{150}]/B8M
+
+/^[\QĀ\E-\QŐ\E]/B8M
+
+/^[\QĀ\E-\QŐ\E/B8M
+
+/[\p{L}]/BM
+
+/[\p{^L}]/BM
+
+/[\P{L}]/BM
+
+/[\P{^L}]/BM
+
+/[abc\p{L}\x{0660}]/8BM
+
+/[\p{Nd}]/8BM
+
+/[\p{Nd}+-]+/8BM
+
+/A\x{391}\x{10427}\x{ff3a}\x{1fb0}/8iBM
+
+/A\x{391}\x{10427}\x{ff3a}\x{1fb0}/8BM
+
+/[\x{105}-\x{109}]/8iBM
+
+/( ( (?(1)0|) )*   )/xBM
+
+/(  (?(1)0|)*   )/xBM
+
+/[a]/BM
+
+/[a]/8BM
+
+/[\xaa]/BM
+
+/[\xaa]/8BM
+
+/[^a]/BM
+
+/[^a]/8BM
+
+/[^\xaa]/BM
+
+/[^\xaa]/8BM
+
+/[^\d]/8WB
+
+/[[:^alpha:][:^cntrl:]]+/8WB
+
+/[[:^cntrl:][:^alpha:]]+/8WB
+
+/[[:alpha:]]+/8WB
+
+/[[:^alpha:]\S]+/8WB
+
+/abc(d|e)(*THEN)x(123(*THEN)4|567(b|q)(*THEN)xx)/B
+
+/-- End of testinput11 --/
diff --git a/pcre/testdata/testinput12 b/pcre/testdata/testinput12
new file mode 100644
index 00000000000..5d727af26af
--- /dev/null
+++ b/pcre/testdata/testinput12
@@ -0,0 +1,90 @@
+/-- This test is run only when JIT support is available. It checks for a
+successful and an unsuccessful JIT compile and save and restore behaviour,
+and a couple of things that are different with JIT. --/
+
+/abc/S+I
+
+/(?(?C1)(?=a)a)/S+I
+
+/(?(?C1)(?=a)a)/S!+I
+
+/abc/S+I>testsavedregex
+
+b)c/PN
+    abc
+
+/a?|b?/P
+    abc
+    ** Failers
+    ddd\N   
+
+/\w+A/P
+   CDAAAAB 
+
+/\w+A/PU
+   CDAAAAB 
+   
+/\Biss\B/I+P
+    Mississippi
+
+/abc/\P
+
+/-- End of POSIX tests --/ 
+
+/a\Cb/
+    aXb
+    a\nb
+    ** Failers (too big char) 
+    A\x{123}B 
+  
+/\x{100}/I
+
+/  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*                          # optional leading comment
+(?:    (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+)                    # initial word
+(?:  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+)  )* # further okay, if led by a period
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  @  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*    (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                           # initial subdomain
+(?:                                  #
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.                        # if led by a period...
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                     #   ...further okay
+)*
+# address
+|                     #  or
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+)             # one word, optionally followed by....
+(?:
+[^()<>@,;:".\\\[\]\x80-\xff\000-\010\012-\037]  |  # atom and space parts, or...
+\(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)       |  # comments, or...
+
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+# quoted strings
+)*
+<  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*                     # leading <
+(?:  @  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*    (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                           # initial subdomain
+(?:                                  #
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.                        # if led by a period...
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                     #   ...further okay
+)*
+
+(?:  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  ,  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  @  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*    (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                           # initial subdomain
+(?:                                  #
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.                        # if led by a period...
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                     #   ...further okay
+)*
+)* # further okay, if led by comma
+:                                # closing colon
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  )? #       optional route
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+)                    # initial word
+(?:  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+)  )* # further okay, if led by a period
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  @  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*    (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                           # initial subdomain
+(?:                                  #
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.                        # if led by a period...
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                     #   ...further okay
+)*
+#       address spec
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  > #                  trailing >
+# name and address
+)  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*                       # optional trailing comment
+/xSI
+
+/-- Although this saved pattern was compiled with link-size=2, it does no harm
+to run this test with other link sizes because it is going to generated a
+"compiled in wrong mode" error as soon as it is loaded, so the link size does 
+not matter. --/
+
+\x09<
+
+/[\h]+/BZ
+    >\x09\x20\xa0<
+
+/[\v]/BZ
+
+/[\H]/BZ
+
+/[^\h]/BZ
+
+/[\V]/BZ
+
+/[\x0a\V]/BZ
+
+/\777/I
+
+/(*:0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF)XX/K
+    XX
+     
+/(*:0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDE)XX/K
+    XX
+
+/\u0100/
+
+/[\u0100-\u0200]/
+
+/-- End of testinput14 --/
diff --git a/pcre/testdata/testinput15 b/pcre/testdata/testinput15
new file mode 100644
index 00000000000..9aa8f6836a7
--- /dev/null
+++ b/pcre/testdata/testinput15
@@ -0,0 +1,364 @@
+/-- This set of tests is for UTF-8 support, and is relevant only to the 8-bit 
+    library. --/
+
+/X(\C{3})/8
+    X\x{1234}
+
+/X(\C{4})/8
+    X\x{1234}YZ
+    
+/X\C*/8
+    XYZabcdce
+    
+/X\C*?/8
+    XYZabcde
+    
+/X\C{3,5}/8
+    Xabcdefg   
+    X\x{1234} 
+    X\x{1234}YZ
+    X\x{1234}\x{512}  
+    X\x{1234}\x{512}YZ
+
+/X\C{3,5}?/8
+    Xabcdefg   
+    X\x{1234} 
+    X\x{1234}YZ
+    X\x{1234}\x{512}  
+
+/a\Cb/8
+    aXb
+    a\nb
+    
+/a\C\Cb/8 
+    a\x{100}b 
+
+/ab\Cde/8
+    abXde
+
+/a\C\Cb/8 
+    a\x{100}b
+    ** Failers 
+    a\x{12257}b
+
+/[]/8
+
+//8
+
+/xxx/8
+
+/xxx/8?DZSS
+
+/badutf/8
+    \xdf
+    \xef
+    \xef\x80
+    \xf7
+    \xf7\x80
+    \xf7\x80\x80
+    \xfb
+    \xfb\x80
+    \xfb\x80\x80
+    \xfb\x80\x80\x80
+    \xfd
+    \xfd\x80
+    \xfd\x80\x80
+    \xfd\x80\x80\x80
+    \xfd\x80\x80\x80\x80
+    \xdf\x7f
+    \xef\x7f\x80
+    \xef\x80\x7f
+    \xf7\x7f\x80\x80
+    \xf7\x80\x7f\x80
+    \xf7\x80\x80\x7f
+    \xfb\x7f\x80\x80\x80
+    \xfb\x80\x7f\x80\x80
+    \xfb\x80\x80\x7f\x80
+    \xfb\x80\x80\x80\x7f
+    \xfd\x7f\x80\x80\x80\x80
+    \xfd\x80\x7f\x80\x80\x80
+    \xfd\x80\x80\x7f\x80\x80
+    \xfd\x80\x80\x80\x7f\x80
+    \xfd\x80\x80\x80\x80\x7f
+    \xed\xa0\x80
+    \xc0\x8f
+    \xe0\x80\x8f
+    \xf0\x80\x80\x8f
+    \xf8\x80\x80\x80\x8f
+    \xfc\x80\x80\x80\x80\x8f
+    \x80
+    \xfe
+    \xff
+
+/badutf/8
+    \xfb\x80\x80\x80\x80
+    \xfd\x80\x80\x80\x80\x80
+    \xf7\xbf\xbf\xbf
+
+/shortutf/8
+    \P\P\xdf
+    \P\P\xef
+    \P\P\xef\x80
+    \P\P\xf7
+    \P\P\xf7\x80
+    \P\P\xf7\x80\x80
+    \P\P\xfb
+    \P\P\xfb\x80
+    \P\P\xfb\x80\x80
+    \P\P\xfb\x80\x80\x80
+    \P\P\xfd
+    \P\P\xfd\x80
+    \P\P\xfd\x80\x80
+    \P\P\xfd\x80\x80\x80
+    \P\P\xfd\x80\x80\x80\x80
+
+/anything/8
+    \xc0\x80
+    \xc1\x8f 
+    \xe0\x9f\x80
+    \xf0\x8f\x80\x80 
+    \xf8\x87\x80\x80\x80  
+    \xfc\x83\x80\x80\x80\x80
+    \xfe\x80\x80\x80\x80\x80  
+    \xff\x80\x80\x80\x80\x80  
+    \xc3\x8f
+    \xe0\xaf\x80
+    \xe1\x80\x80
+    \xf0\x9f\x80\x80 
+    \xf1\x8f\x80\x80 
+    \xf8\x88\x80\x80\x80  
+    \xf9\x87\x80\x80\x80  
+    \xfc\x84\x80\x80\x80\x80
+    \xfd\x83\x80\x80\x80\x80
+    \?\xf8\x88\x80\x80\x80  
+    \?\xf9\x87\x80\x80\x80  
+    \?\xfc\x84\x80\x80\x80\x80
+    \?\xfd\x83\x80\x80\x80\x80
+
+/\x{100}/8DZ
+
+/\x{1000}/8DZ
+
+/\x{10000}/8DZ
+
+/\x{100000}/8DZ
+
+/\x{10ffff}/8DZ
+
+/[\x{ff}]/8DZ
+
+/[\x{100}]/8DZ
+
+/\x80/8DZ
+
+/\xff/8DZ
+
+/\x{D55c}\x{ad6d}\x{C5B4}/DZ8 
+    \x{D55c}\x{ad6d}\x{C5B4} 
+
+/\x{65e5}\x{672c}\x{8a9e}/DZ8
+    \x{65e5}\x{672c}\x{8a9e}
+
+/\x{80}/DZ8
+
+/\x{084}/DZ8
+
+/\x{104}/DZ8
+
+/\x{861}/DZ8
+
+/\x{212ab}/DZ8
+
+/-- This one is here not because it's different to Perl, but because the way
+the captured single-byte is displayed. (In Perl it becomes a character, and you
+can't tell the difference.) --/
+    
+/X(\C)(.*)/8
+    X\x{1234}
+    X\nabc 
+
+/-- This one is here because Perl gives out a grumbly error message (quite 
+correctly, but that messes up comparisons). --/
+    
+/a\Cb/8
+    *** Failers 
+    a\x{100}b 
+    
+/[^ab\xC0-\xF0]/8SDZ
+    \x{f1}
+    \x{bf}
+    \x{100}
+    \x{1000}   
+    *** Failers
+    \x{c0} 
+    \x{f0} 
+
+/Ā{3,4}/8SDZ
+  \x{100}\x{100}\x{100}\x{100\x{100}
+
+/(\x{100}+|x)/8SDZ
+
+/(\x{100}*a|x)/8SDZ
+
+/(\x{100}{0,2}a|x)/8SDZ
+
+/(\x{100}{1,2}a|x)/8SDZ
+
+/\x{100}/8DZ
+
+/a\x{100}\x{101}*/8DZ
+
+/a\x{100}\x{101}+/8DZ
+
+/[^\x{c4}]/DZ
+
+/[\x{100}]/8DZ
+    \x{100}
+    Z\x{100}
+    \x{100}Z
+    *** Failers 
+
+/[\xff]/DZ8
+    >\x{ff}<
+
+/[^\xff]/8DZ
+
+/\x{100}abc(xyz(?1))/8DZ
+
+/a\x{1234}b/P8
+    a\x{1234}b
+
+/\777/8I
+  \x{1ff}
+  \777 
+  
+/\x{100}+\x{200}/8DZ
+
+/\x{100}+X/8DZ
+
+/^[\QĀ\E-\QŐ\E/BZ8
+
+/-- This tests the stricter UTF-8 check according to RFC 3629. --/ 
+    
+/X/8
+    \x{d800}
+    \x{d800}\?
+    \x{da00}
+    \x{da00}\?
+    \x{dfff}
+    \x{dfff}\?
+    \x{110000}    
+    \x{110000}\?    
+    \x{2000000} 
+    \x{2000000}\? 
+    \x{7fffffff} 
+    \x{7fffffff}\? 
+
+/(*UTF8)\x{1234}/
+  abcd\x{1234}pqr
+
+/(*CRLF)(*UTF)(*BSR_UNICODE)a\Rb/I
+
+/\h/SI8
+    ABC\x{09}
+    ABC\x{20}
+    ABC\x{a0}
+    ABC\x{1680}
+    ABC\x{180e}
+    ABC\x{2000}
+    ABC\x{202f} 
+    ABC\x{205f} 
+    ABC\x{3000} 
+
+/\v/SI8
+    ABC\x{0a}
+    ABC\x{0b}
+    ABC\x{0c}
+    ABC\x{0d}
+    ABC\x{85}
+    ABC\x{2028}
+
+/\h*A/SI8
+    CDBABC
+    
+/\v+A/SI8
+
+/\s?xxx\s/8SI
+
+/\sxxx\s/I8ST1
+    AB\x{85}xxx\x{a0}XYZ
+    AB\x{a0}xxx\x{85}XYZ
+
+/\S \S/I8ST1
+    \x{a2} \x{84} 
+    A Z 
+
+/a+/8
+    a\x{123}aa\>1
+    a\x{123}aa\>2
+    a\x{123}aa\>3
+    a\x{123}aa\>4
+    a\x{123}aa\>5
+    a\x{123}aa\>6
+
+/\x{1234}+/iS8I
+
+/\x{1234}+?/iS8I
+
+/\x{1234}++/iS8I
+
+/\x{1234}{2}/iS8I
+
+/[^\x{c4}]/8DZ
+
+/X+\x{200}/8DZ
+
+/\R/SI8
+
+/\777/8DZ
+
+/\w+\x{C4}/8BZ
+    a\x{C4}\x{C4}
+
+/\w+\x{C4}/8BZT1
+    a\x{C4}\x{C4}
+    
+/\W+\x{C4}/8BZ
+    !\x{C4}
+ 
+/\W+\x{C4}/8BZT1
+    !\x{C4}
+
+/\W+\x{A1}/8BZ
+    !\x{A1}
+ 
+/\W+\x{A1}/8BZT1
+    !\x{A1}
+
+/X\s+\x{A0}/8BZ
+    X\x20\x{A0}\x{A0}
+
+/X\s+\x{A0}/8BZT1
+    X\x20\x{A0}\x{A0}
+
+/\S+\x{A0}/8BZ
+    X\x{A0}\x{A0}
+
+/\S+\x{A0}/8BZT1
+    X\x{A0}\x{A0}
+
+/\x{a0}+\s!/8BZ
+    \x{a0}\x20!
+
+/\x{a0}+\s!/8BZT1
+    \x{a0}\x20!
+
+/A/8
+  \x{ff000041}
+  \x{7f000041} 
+
+/(*UTF8)abc/9
+
+/abc/89
+
+/-- End of testinput15 --/
diff --git a/pcre/testdata/testinput16 b/pcre/testdata/testinput16
new file mode 100644
index 00000000000..e7a05ae0d45
--- /dev/null
+++ b/pcre/testdata/testinput16
@@ -0,0 +1,35 @@
+/-- This set of tests is run only with the 8-bit library when Unicode property 
+    support is available. It starts with tests of the POSIX interface, because
+    that is supported only with the 8-bit library. --/
+
+/\w/P
+    +++\x{c2}
+
+/\w/WP
+    +++\x{c2}
+    
+/A\x{391}\x{10427}\x{ff3a}\x{1fb0}/8iDZ
+
+/A\x{391}\x{10427}\x{ff3a}\x{1fb0}/8DZ
+
+/AB\x{1fb0}/8DZ
+
+/AB\x{1fb0}/8DZi
+
+/\x{401}\x{420}\x{421}\x{422}\x{423}\x{424}\x{425}\x{426}\x{427}\x{428}\x{429}\x{42a}\x{42b}\x{42c}\x{42d}\x{42e}\x{42f}/8iSI
+    \x{401}\x{420}\x{421}\x{422}\x{423}\x{424}\x{425}\x{426}\x{427}\x{428}\x{429}\x{42a}\x{42b}\x{42c}\x{42d}\x{42e}\x{42f}
+    \x{451}\x{440}\x{441}\x{442}\x{443}\x{444}\x{445}\x{446}\x{447}\x{448}\x{449}\x{44a}\x{44b}\x{44c}\x{44d}\x{44e}\x{44f}
+
+/[ⱥ]/8iBZ
+
+/[^ⱥ]/8iBZ
+
+/\h/SI
+
+/\v/SI
+
+/\R/SI
+
+/[[:blank:]]/WBZ
+
+/-- End of testinput16 --/
diff --git a/pcre/testdata/testinput17 b/pcre/testdata/testinput17
new file mode 100644
index 00000000000..ba80318857b
--- /dev/null
+++ b/pcre/testdata/testinput17
@@ -0,0 +1,296 @@
+/-- This set of tests is for the 16- and 32-bit library's basic (non-UTF-16 
+    or -32) features that are not compatible with the 8-bit library, or which 
+    give different output in 16- or 32-bit mode. --/
+
+/a\Cb/
+    aXb
+    a\nb
+  
+/[^\x{c4}]/DZ
+  
+/\x{100}/I
+
+/  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*                          # optional leading comment
+(?:    (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+)                    # initial word
+(?:  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+)  )* # further okay, if led by a period
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  @  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*    (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                           # initial subdomain
+(?:                                  #
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.                        # if led by a period...
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                     #   ...further okay
+)*
+# address
+|                     #  or
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+)             # one word, optionally followed by....
+(?:
+[^()<>@,;:".\\\[\]\x80-\xff\000-\010\012-\037]  |  # atom and space parts, or...
+\(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)       |  # comments, or...
+
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+# quoted strings
+)*
+<  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*                     # leading <
+(?:  @  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*    (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                           # initial subdomain
+(?:                                  #
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.                        # if led by a period...
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                     #   ...further okay
+)*
+
+(?:  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  ,  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  @  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*    (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                           # initial subdomain
+(?:                                  #
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.                        # if led by a period...
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                     #   ...further okay
+)*
+)* # further okay, if led by comma
+:                                # closing colon
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  )? #       optional route
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+)                    # initial word
+(?:  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+)  )* # further okay, if led by a period
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  @  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*    (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                           # initial subdomain
+(?:                                  #
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.                        # if led by a period...
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                     #   ...further okay
+)*
+#       address spec
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  > #                  trailing >
+# name and address
+)  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*                       # optional trailing comment
+/xSI
+
+/[\h]/BZ
+    >\x09<
+
+/[\h]+/BZ
+    >\x09\x20\xa0<
+
+/[\v]/BZ
+
+/[^\h]/BZ
+
+/\h+/SI
+    \x{1681}\x{200b}\x{1680}\x{2000}\x{202f}\x{3000}
+    \x{3001}\x{2fff}\x{200a}\xa0\x{2000}
+
+/[\h\x{dc00}]+/BZSI
+    \x{1681}\x{200b}\x{1680}\x{2000}\x{202f}\x{3000}
+    \x{3001}\x{2fff}\x{200a}\xa0\x{2000}
+
+/\H+/SI
+    \x{1680}\x{180e}\x{167f}\x{1681}\x{180d}\x{180f}
+    \x{2000}\x{200a}\x{1fff}\x{200b}
+    \x{202f}\x{205f}\x{202e}\x{2030}\x{205e}\x{2060}
+    \xa0\x{3000}\x9f\xa1\x{2fff}\x{3001}
+
+/[\H\x{d800}]+/
+    \x{1680}\x{180e}\x{167f}\x{1681}\x{180d}\x{180f}
+    \x{2000}\x{200a}\x{1fff}\x{200b}
+    \x{202f}\x{205f}\x{202e}\x{2030}\x{205e}\x{2060}
+    \xa0\x{3000}\x9f\xa1\x{2fff}\x{3001}
+
+/\v+/SI
+    \x{2027}\x{2030}\x{2028}\x{2029}
+    \x09\x0e\x84\x86\x85\x0a\x0b\x0c\x0d
+
+/[\v\x{dc00}]+/BZSI
+    \x{2027}\x{2030}\x{2028}\x{2029}
+    \x09\x0e\x84\x86\x85\x0a\x0b\x0c\x0d
+
+/\V+/SI
+    \x{2028}\x{2029}\x{2027}\x{2030}
+    \x85\x0a\x0b\x0c\x0d\x09\x0e\x84\x86
+
+/[\V\x{d800}]+/
+    \x{2028}\x{2029}\x{2027}\x{2030}
+    \x85\x0a\x0b\x0c\x0d\x09\x0e\x84\x86
+
+/\R+/SI
+    \x{2027}\x{2030}\x{2028}\x{2029}
+    \x09\x0e\x84\x86\x85\x0a\x0b\x0c\x0d
+
+/\x{d800}\x{d7ff}\x{dc00}\x{dc00}\x{dcff}\x{dd00}/I
+    \x{d800}\x{d7ff}\x{dc00}\x{dc00}\x{dcff}\x{dd00}
+
+/[^\x{80}][^\x{ff}][^\x{100}][^\x{1000}][^\x{ffff}]/BZ
+
+/[^\x{80}][^\x{ff}][^\x{100}][^\x{1000}][^\x{ffff}]/BZi
+
+/[^\x{100}]*[^\x{1000}]+[^\x{ffff}]??[^\x{8000}]{4,}[^\x{7fff}]{2,9}?[^\x{100}]{5,6}+/BZ
+
+/[^\x{100}]*[^\x{1000}]+[^\x{ffff}]??[^\x{8000}]{4,}[^\x{7fff}]{2,9}?[^\x{100}]{5,6}+/BZi
+
+/(*:0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF)XX/K
+    XX
+     
+/(*:0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDE)XX/K
+    XX
+
+/\u0100/BZ
+
+/[\u0100-\u0200]/BZ
+
+/\ud800/BZ
+
+/^\x{ffff}+/i
+    \x{ffff}
+
+/^\x{ffff}?/i
+    \x{ffff}
+
+/^\x{ffff}*/i
+    \x{ffff}
+
+/^\x{ffff}{3}/i
+    \x{ffff}\x{ffff}\x{ffff}
+
+/^\x{ffff}{0,3}/i
+    \x{ffff}
+
+/-- End of testinput17 --/
diff --git a/pcre/testdata/testinput18 b/pcre/testdata/testinput18
new file mode 100644
index 00000000000..42bad1fa418
--- /dev/null
+++ b/pcre/testdata/testinput18
@@ -0,0 +1,298 @@
+/-- This set of tests is for UTF-16 and UTF-32 support, and is relevant only to the
+    16- and 32-bit library. --/
+
+/xxx/8?DZSS
+
+/abc/8
+    ]
+
+/X(\C{3})/8
+    X\x{11234}Y
+    X\x{11234}YZ
+
+/X(\C{4})/8
+    X\x{11234}YZ
+    X\x{11234}YZW
+
+/X\C*/8
+    XYZabcdce
+
+/X\C*?/8
+    XYZabcde
+
+/X\C{3,5}/8
+    Xabcdefg
+    X\x{11234}Y
+    X\x{11234}YZ
+    X\x{11234}\x{512}
+    X\x{11234}\x{512}YZ
+    X\x{11234}\x{512}\x{11234}Z
+
+/X\C{3,5}?/8
+    Xabcdefg
+    X\x{11234}Y
+    X\x{11234}YZ
+    X\x{11234}\x{512}YZ
+    *** Failers
+    X\x{11234}
+
+/a\Cb/8
+    aXb
+    a\nb
+
+/a\C\Cb/8
+    a\x{12257}b
+    a\x{12257}\x{11234}b
+    ** Failers
+    a\x{100}b
+
+/ab\Cde/8
+    abXde
+
+/-- Check maximum character size --/
+
+/\x{ffff}/8DZ
+
+/\x{10000}/8DZ
+
+/\x{100}/8DZ
+
+/\x{1000}/8DZ
+
+/\x{10000}/8DZ
+
+/\x{100000}/8DZ
+
+/\x{10ffff}/8DZ
+
+/[\x{ff}]/8DZ
+
+/[\x{100}]/8DZ
+
+/\x80/8DZ
+
+/\xff/8DZ
+
+/\x{D55c}\x{ad6d}\x{C5B4}/DZ8
+    \x{D55c}\x{ad6d}\x{C5B4}
+
+/\x{65e5}\x{672c}\x{8a9e}/DZ8
+    \x{65e5}\x{672c}\x{8a9e}
+
+/\x{80}/DZ8
+
+/\x{084}/DZ8
+
+/\x{104}/DZ8
+
+/\x{861}/DZ8
+
+/\x{212ab}/DZ8
+
+/-- This one is here not because it's different to Perl, but because the way
+the captured single-byte is displayed. (In Perl it becomes a character, and you
+can't tell the difference.) --/
+
+/X(\C)(.*)/8
+    X\x{1234}
+    X\nabc
+
+/-- This one is here because Perl gives out a grumbly error message (quite
+correctly, but that messes up comparisons). --/
+
+/a\Cb/8
+    *** Failers
+    a\x{100}b
+
+/[^ab\xC0-\xF0]/8SDZ
+    \x{f1}
+    \x{bf}
+    \x{100}
+    \x{1000}
+    *** Failers
+    \x{c0}
+    \x{f0}
+
+/Ā{3,4}/8SDZ
+  \x{100}\x{100}\x{100}\x{100\x{100}
+
+/(\x{100}+|x)/8SDZ
+
+/(\x{100}*a|x)/8SDZ
+
+/(\x{100}{0,2}a|x)/8SDZ
+
+/(\x{100}{1,2}a|x)/8SDZ
+
+/\x{100}/8DZ
+
+/a\x{100}\x{101}*/8DZ
+
+/a\x{100}\x{101}+/8DZ
+
+/[^\x{c4}]/DZ
+
+/[\x{100}]/8DZ
+    \x{100}
+    Z\x{100}
+    \x{100}Z
+    *** Failers
+
+/[\xff]/DZ8
+    >\x{ff}<
+
+/[^\xff]/8DZ
+
+/\x{100}abc(xyz(?1))/8DZ
+
+/\777/8I
+  \x{1ff}
+  \777
+
+/\x{100}+\x{200}/8DZ
+
+/\x{100}+X/8DZ
+
+/^[\QĀ\E-\QŐ\E/BZ8
+
+/X/8
+    \x{d800}
+    \x{d800}\?
+    \x{da00}
+    \x{da00}\?
+    \x{dc00}
+    \x{dc00}\?
+    \x{de00}
+    \x{de00}\?
+    \x{dfff}
+    \x{dfff}\?
+    \x{110000}
+    \x{d800}\x{1234}
+
+/(*UTF16)\x{11234}/
+  abcd\x{11234}pqr
+
+/(*UTF)\x{11234}/I
+  abcd\x{11234}pqr
+
+/(*UTF-32)\x{11234}/
+  abcd\x{11234}pqr
+
+/(*CRLF)(*UTF16)(*BSR_UNICODE)a\Rb/I
+
+/(*CRLF)(*UTF32)(*BSR_UNICODE)a\Rb/I
+
+/\h/SI8
+    ABC\x{09}
+    ABC\x{20}
+    ABC\x{a0}
+    ABC\x{1680}
+    ABC\x{180e}
+    ABC\x{2000}
+    ABC\x{202f}
+    ABC\x{205f}
+    ABC\x{3000}
+
+/\v/SI8
+    ABC\x{0a}
+    ABC\x{0b}
+    ABC\x{0c}
+    ABC\x{0d}
+    ABC\x{85}
+    ABC\x{2028}
+
+/\h*A/SI8
+    CDBABC
+    \x{2000}ABC 
+
+/\R*A/SI8
+    CDBABC
+    \x{2028}A  
+
+/\v+A/SI8
+
+/\s?xxx\s/8SI
+
+/\sxxx\s/I8ST1
+    AB\x{85}xxx\x{a0}XYZ
+    AB\x{a0}xxx\x{85}XYZ
+
+/\S \S/I8ST1
+    \x{a2} \x{84}
+    A Z
+
+/a+/8
+    a\x{123}aa\>1
+    a\x{123}aa\>2
+    a\x{123}aa\>3
+    a\x{123}aa\>4
+    a\x{123}aa\>5
+    a\x{123}aa\>6
+
+/\x{1234}+/iS8I
+
+/\x{1234}+?/iS8I
+
+/\x{1234}++/iS8I
+
+/\x{1234}{2}/iS8I
+
+/[^\x{c4}]/8DZ
+
+/X+\x{200}/8DZ
+
+/\R/SI8
+
+/-- Check bad offset --/
+
+/a/8
+    \x{10000}\>1
+    \x{10000}ab\>1
+    \x{10000}ab\>2
+    \x{10000}ab\>3
+    \x{10000}ab\>4
+    \x{10000}ab\>5
+
+//8
+
+/\w+\x{C4}/8BZ
+    a\x{C4}\x{C4}
+
+/\w+\x{C4}/8BZT1
+    a\x{C4}\x{C4}
+    
+/\W+\x{C4}/8BZ
+    !\x{C4}
+ 
+/\W+\x{C4}/8BZT1
+    !\x{C4}
+
+/\W+\x{A1}/8BZ
+    !\x{A1}
+ 
+/\W+\x{A1}/8BZT1
+    !\x{A1}
+
+/X\s+\x{A0}/8BZ
+    X\x20\x{A0}\x{A0}
+
+/X\s+\x{A0}/8BZT1
+    X\x20\x{A0}\x{A0}
+
+/\S+\x{A0}/8BZ
+    X\x{A0}\x{A0}
+
+/\S+\x{A0}/8BZT1
+    X\x{A0}\x{A0}
+
+/\x{a0}+\s!/8BZ
+    \x{a0}\x20!
+
+/\x{a0}+\s!/8BZT1
+    \x{a0}\x20!
+
+/(*UTF)abc/9
+
+/abc/89
+
+/-- End of testinput18 --/
diff --git a/pcre/testdata/testinput19 b/pcre/testdata/testinput19
new file mode 100644
index 00000000000..00d80203f00
--- /dev/null
+++ b/pcre/testdata/testinput19
@@ -0,0 +1,22 @@
+/-- This set of tests is for Unicode property support, relevant only to the
+    16- and 32-bit library. --/
+    
+/A\x{391}\x{10427}\x{ff3a}\x{1fb0}/8iDZ
+
+/A\x{391}\x{10427}\x{ff3a}\x{1fb0}/8DZ
+
+/AB\x{1fb0}/8DZ
+
+/AB\x{1fb0}/8DZi
+
+/\x{401}\x{420}\x{421}\x{422}\x{423}\x{424}\x{425}\x{426}\x{427}\x{428}\x{429}\x{42a}\x{42b}\x{42c}\x{42d}\x{42e}\x{42f}/8iSI
+    \x{401}\x{420}\x{421}\x{422}\x{423}\x{424}\x{425}\x{426}\x{427}\x{428}\x{429}\x{42a}\x{42b}\x{42c}\x{42d}\x{42e}\x{42f}
+    \x{451}\x{440}\x{441}\x{442}\x{443}\x{444}\x{445}\x{446}\x{447}\x{448}\x{449}\x{44a}\x{44b}\x{44c}\x{44d}\x{44e}\x{44f}
+
+/[ⱥ]/8iBZ
+
+/[^ⱥ]/8iBZ
+
+/[[:blank:]]/WBZ
+
+/-- End of testinput19 --/ 
diff --git a/pcre/testdata/testinput2 b/pcre/testdata/testinput2
new file mode 100644
index 00000000000..6c82fb56df6
--- /dev/null
+++ b/pcre/testdata/testinput2
@@ -0,0 +1,3840 @@
+/-- This set of tests is not Perl-compatible. It checks on special features
+    of PCRE's API, error diagnostics, and the compiled code of some patterns.
+    It also checks the non-Perl syntax the PCRE supports (Python, .NET, 
+    Oniguruma). Finally, there are some tests where PCRE and Perl differ, 
+    either because PCRE can't be compatible, or there is a possible Perl 
+    bug.
+    
+    NOTE: This is a non-UTF set of tests. When UTF support is needed, use
+    test 5, and if Unicode Property Support is needed, use test 7. --/
+  
+/(a)b|/I
+
+/abc/I
+    abc
+    defabc
+    \Aabc
+    *** Failers
+    \Adefabc
+    ABC
+
+/^abc/I
+    abc
+    \Aabc
+    *** Failers
+    defabc
+    \Adefabc
+
+/a+bc/I
+
+/a*bc/I
+
+/a{3}bc/I
+
+/(abc|a+z)/I
+
+/^abc$/I
+    abc
+    *** Failers
+    def\nabc
+
+/ab\idef/X
+
+/(?X)ab\idef/X
+
+/x{5,4}/
+
+/z{65536}/
+
+/[abcd/
+
+/(?X)[\B]/
+
+/(?X)[\R]/
+
+/(?X)[\X]/
+
+/[\B]/BZ
+
+/[\R]/BZ
+
+/[\X]/BZ
+
+/[z-a]/
+
+/^*/
+
+/(abc/
+
+/(?# abc/
+
+/(?z)abc/
+
+/.*b/I
+
+/.*?b/I
+
+/cat|dog|elephant/I
+    this sentence eventually mentions a cat
+    this sentences rambles on and on for a while and then reaches elephant
+
+/cat|dog|elephant/IS
+    this sentence eventually mentions a cat
+    this sentences rambles on and on for a while and then reaches elephant
+
+/cat|dog|elephant/IiS
+    this sentence eventually mentions a CAT cat
+    this sentences rambles on and on for a while to elephant ElePhant
+
+/a|[bcd]/IS
+
+/(a|[^\dZ])/IS
+
+/(a|b)*[\s]/IS
+
+/(ab\2)/
+
+/{4,5}abc/
+
+/(a)(b)(c)\2/I
+    abcb
+    \O0abcb
+    \O3abcb
+    \O6abcb
+    \O9abcb
+    \O12abcb
+
+/(a)bc|(a)(b)\2/I
+    abc
+    \O0abc
+    \O3abc
+    \O6abc
+    aba
+    \O0aba
+    \O3aba
+    \O6aba
+    \O9aba
+    \O12aba
+
+/abc$/IE
+    abc
+    *** Failers
+    abc\n
+    abc\ndef
+
+/(a)(b)(c)(d)(e)\6/
+
+/the quick brown fox/I
+    the quick brown fox
+    this is a line with the quick brown fox
+
+/the quick brown fox/IA
+    the quick brown fox
+    *** Failers
+    this is a line with the quick brown fox
+
+/ab(?z)cd/
+
+/^abc|def/I
+    abcdef
+    abcdef\B
+
+/.*((abc)$|(def))/I
+    defabc
+    \Zdefabc
+
+/)/
+
+/a[]b/
+
+/[^aeiou ]{3,}/I
+    co-processors, and for
+
+/<.*>/I
+    abcghinop
+
+/<.*?>/I
+    abcghinop
+
+/<.*>/IU
+    abcghinop
+
+/(?U)<.*>/I
+    abcghinop
+
+/<.*?>/IU
+    abcghinop
+
+/={3,}/IU
+    abc========def
+
+/(?U)={3,}?/I
+    abc========def
+
+/(?^abc)/Im
+    abc
+    def\nabc
+    *** Failers
+    defabc
+
+/(?<=ab(c+)d)ef/
+
+/(?<=ab(?<=c+)d)ef/
+
+/(?<=ab(c|de)f)g/
+
+/The next three are in testinput2 because they have variable length branches/
+
+/(?<=bullock|donkey)-cart/I
+    the bullock-cart
+    a donkey-cart race
+    *** Failers
+    cart
+    horse-and-cart
+
+/(?<=ab(?i)x|y|z)/I
+
+/(?>.*)(?<=(abcd)|(xyz))/I
+    alphabetabcd
+    endingxyz
+
+/(?<=ab(?i)x(?-i)y|(?i)z|b)ZZ/I
+    abxyZZ
+    abXyZZ
+    ZZZ
+    zZZ
+    bZZ
+    BZZ
+    *** Failers
+    ZZ
+    abXYZZ
+    zzz
+    bzz
+
+/(?[^()]+)    # Either a sequence of non-brackets (no backtracking)
+  |             # Or
+  (?R)          # Recurse - i.e. nested bracketed string
+  )*            # Zero or more contents
+  \)            # Closing )
+  /Ix
+    (abcd)
+    (abcd)xyz
+    xyz(abcd)
+    (ab(xy)cd)pqr
+    (ab(xycd)pqr
+    () abc ()
+    12(abcde(fsh)xyz(foo(bar))lmno)89
+    *** Failers
+    abcd
+    abcd)
+    (abcd
+
+/\(  ( (?>[^()]+) | (?R) )* \) /Ixg
+    (ab(xy)cd)pqr
+    1(abcd)(x(y)z)pqr
+
+/\(  (?: (?>[^()]+) | (?R) ) \) /Ix
+    (abcd)
+    (ab(xy)cd)
+    (a(b(c)d)e)
+    ((ab))
+    *** Failers
+    ()
+
+/\(  (?: (?>[^()]+) | (?R) )? \) /Ix
+    ()
+    12(abcde(fsh)xyz(foo(bar))lmno)89
+
+/\(  ( (?>[^()]+) | (?R) )* \) /Ix
+    (ab(xy)cd)
+
+/\( ( ( (?>[^()]+) | (?R) )* ) \) /Ix
+    (ab(xy)cd)
+
+/\( (123)? ( ( (?>[^()]+) | (?R) )* ) \) /Ix
+    (ab(xy)cd)
+    (123ab(xy)cd)
+
+/\( ( (123)? ( (?>[^()]+) | (?R) )* ) \) /Ix
+    (ab(xy)cd)
+    (123ab(xy)cd)
+
+/\( (((((((((( ( (?>[^()]+) | (?R) )* )))))))))) \) /Ix
+    (ab(xy)cd)
+
+/\( ( ( (?>[^()<>]+) | ((?>[^()]+)) | (?R) )* ) \) /Ix
+    (abcd(xyz
qrs)123)
+
+/\( ( ( (?>[^()]+) | ((?R)) )* ) \) /Ix
+    (ab(cd)ef)
+    (ab(cd(ef)gh)ij)
+
+/^[[:alnum:]]/DZ
+
+/^[[:^alnum:]]/DZ
+
+/^[[:alpha:]]/DZ
+
+/^[[:^alpha:]]/DZ
+
+/[_[:alpha:]]/IS
+
+/^[[:ascii:]]/DZ
+
+/^[[:^ascii:]]/DZ
+
+/^[[:blank:]]/DZ
+
+/^[[:^blank:]]/DZ
+
+/[\n\x0b\x0c\x0d[:blank:]]/IS
+
+/^[[:cntrl:]]/DZ
+
+/^[[:digit:]]/DZ
+
+/^[[:graph:]]/DZ
+
+/^[[:lower:]]/DZ
+
+/^[[:print:]]/DZ
+
+/^[[:punct:]]/DZ
+
+/^[[:space:]]/DZ
+
+/^[[:upper:]]/DZ
+
+/^[[:xdigit:]]/DZ
+
+/^[[:word:]]/DZ
+
+/^[[:^cntrl:]]/DZ
+
+/^[12[:^digit:]]/DZ
+
+/^[[:^blank:]]/DZ
+
+/[01[:alpha:]%]/DZ
+
+/[[.ch.]]/I
+
+/[[=ch=]]/I
+
+/[[:rhubarb:]]/I
+
+/[[:upper:]]/Ii
+    A
+    a
+
+/[[:lower:]]/Ii
+    A
+    a
+
+/((?-i)[[:lower:]])[[:lower:]]/Ii
+    ab
+    aB
+    *** Failers
+    Ab
+    AB
+
+/[\200-\110]/I
+
+/^(?(0)f|b)oo/I
+
+/This one's here because of the large output vector needed/I
+
+/(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\w+)\s+(\270)/I
+    \O900 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 ABC ABC
+
+/This one's here because Perl does this differently and PCRE can't at present/I
+
+/(main(O)?)+/I
+    mainmain
+    mainOmain
+
+/These are all cases where Perl does it differently (nested captures)/I
+
+/^(a(b)?)+$/I
+    aba
+
+/^(aa(bb)?)+$/I
+    aabbaa
+
+/^(aa|aa(bb))+$/I
+    aabbaa
+
+/^(aa(bb)??)+$/I
+    aabbaa
+
+/^(?:aa(bb)?)+$/I
+    aabbaa
+
+/^(aa(b(b))?)+$/I
+    aabbaa
+
+/^(?:aa(b(b))?)+$/I
+    aabbaa
+
+/^(?:aa(b(?:b))?)+$/I
+    aabbaa
+
+/^(?:aa(bb(?:b))?)+$/I
+    aabbbaa
+
+/^(?:aa(b(?:bb))?)+$/I
+    aabbbaa
+
+/^(?:aa(?:b(b))?)+$/I
+    aabbaa
+
+/^(?:aa(?:b(bb))?)+$/I
+    aabbbaa
+
+/^(aa(b(bb))?)+$/I
+    aabbbaa
+
+/^(aa(bb(bb))?)+$/I
+    aabbbbaa
+
+/--------------------------------------------------------------------/I
+
+/#/IxDZ
+
+/a#/IxDZ
+
+/[\s]/DZ
+
+/[\S]/DZ
+
+/a(?i)b/DZ
+    ab
+    aB
+    *** Failers
+    AB
+
+/(a(?i)b)/DZ
+    ab
+    aB
+    *** Failers
+    AB
+
+/   (?i)abc/IxDZ
+
+/#this is a comment
+  (?i)abc/IxDZ
+
+/123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890/DZ
+
+/\Q123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890/DZ
+
+/\Q\E/DZ
+    \
+
+/\Q\Ex/DZ
+
+/ \Q\E/DZ
+
+/a\Q\E/DZ
+  abc
+  bca
+  bac
+
+/a\Q\Eb/DZ
+  abc
+
+/\Q\Eabc/DZ
+
+/x*+\w/DZ
+    *** Failers
+    xxxxx
+
+/x?+/DZ
+
+/x++/DZ
+
+/x{1,3}+/DZ
+
+/(x)*+/DZ
+
+/^(\w++|\s++)*$/I
+    now is the time for all good men to come to the aid of the party
+    *** Failers
+    this is not a line with only words and spaces!
+
+/(\d++)(\w)/I
+    12345a
+    *** Failers
+    12345+
+
+/a++b/I
+    aaab
+
+/(a++b)/I
+    aaab
+
+/(a++)b/I
+    aaab
+
+/([^()]++|\([^()]*\))+/I
+    ((abc(ade)ufh()()x
+
+/\(([^()]++|\([^()]+\))+\)/I
+    (abc)
+    (abc(def)xyz)
+    *** Failers
+    ((()aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+
+/(abc){1,3}+/DZ
+
+/a+?+/I
+
+/a{2,3}?+b/I
+
+/(?U)a+?+/I
+
+/a{2,3}?+b/IU
+
+/x(?U)a++b/DZ
+    xaaaab
+
+/(?U)xa++b/DZ
+    xaaaab
+
+/^((a+)(?U)([ab]+)(?-U)([bc]+)(\w*))/DZ
+
+/^x(?U)a+b/DZ
+
+/^x(?U)(a+)b/DZ
+
+/[.x.]/I
+
+/[=x=]/I
+
+/[:x:]/I
+
+/\l/I
+
+/\L/I
+
+/\N{name}/I
+
+/\u/I
+
+/\U/I
+
+/[/I
+
+/[a-/I
+
+/[[:space:]/I
+
+/[\s]/IDZ
+
+/[[:space:]]/IDZ
+
+/[[:space:]abcde]/IDZ
+
+/< (?: (?(R) \d++  | [^<>]*+) | (?R)) * >/Ix
+    <>
+    
+     hij>
+     hij>
+    def>
+    
+    *** Failers
+    iV\;\?aPhFB\<\*vW\@QW\@sO9\}cfZA\-i\'w\%hKd6gt1UJP\,15_\#QY\$M\^Mss_U\/\]\&LK9\[5vQub\^w\[KDD\qmj\;2\}YWFdYx\.Ap\]hjCPTP\(n28k\+3\;o\&WXqs\/gOXdr\$\:r\'do0\;b4c\(f_Gr\=\"\\4\)\[01T7ajQJvL\$W\~mL_sS\/4h\:x\*\[ZN\=KLs\&L5zX\/\/\>it\,o\:aU\(\;Z\>pW\&T7oP\'2K\^E\:x9\'c\[\%z\-\,64JQ5AeH_G\#KijUKghQw\^\\vea3a\?kka_G\$8\#\`\*kynsxzBLru\'\]k_\[7FrVx\}\^\=\$blx\>s\-N\%j\;D\*aZDnsw\:YKZ\%Q\.Kne9\#hP\?\+b3\(SOvL\,\^\;\&u5\@\?5C5Bhb\=m\-vEh_L15Jl\]U\)0RP6\{q\%L\^_z5E\'Dw6X\b|IDZ
+
+|\$\<\.X\+ix\[d1b\!H\#\?vV0vrK\:ZH1\=2M\>iV\;\?aPhFB\<\*vW\@QW\@sO9\}cfZA\-i\'w\%hKd6gt1UJP\,15_\#QY\$M\^Mss_U\/\]\&LK9\[5vQub\^w\[KDD\qmj\;2\}YWFdYx\.Ap\]hjCPTP\(n28k\+3\;o\&WXqs\/gOXdr\$\:r\'do0\;b4c\(f_Gr\=\"\\4\)\[01T7ajQJvL\$W\~mL_sS\/4h\:x\*\[ZN\=KLs\&L5zX\/\/\>it\,o\:aU\(\;Z\>pW\&T7oP\'2K\^E\:x9\'c\[\%z\-\,64JQ5AeH_G\#KijUKghQw\^\\vea3a\?kka_G\$8\#\`\*kynsxzBLru\'\]k_\[7FrVx\}\^\=\$blx\>s\-N\%j\;D\*aZDnsw\:YKZ\%Q\.Kne9\#hP\?\+b3\(SOvL\,\^\;\&u5\@\?5C5Bhb\=m\-vEh_L15Jl\]U\)0RP6\{q\%L\^_z5E\'Dw6X\b|IDZ
+
+/(.*)\d+\1/I
+
+/(.*)\d+/I
+
+/(.*)\d+\1/Is
+
+/(.*)\d+/Is
+
+/(.*(xyz))\d+\2/I
+
+/((.*))\d+\1/I
+    abc123bc
+
+/a[b]/I
+
+/(?=a).*/I
+
+/(?=abc).xyz/IiI
+
+/(?=abc)(?i).xyz/I
+
+/(?=a)(?=b)/I
+
+/(?=.)a/I
+
+/((?=abcda)a)/I
+
+/((?=abcda)ab)/I
+
+/()a/I
+
+/(?(1)ab|ac)(.)/I
+
+/(?(1)abz|acz)(.)/I
+
+/(?(1)abz)(.)/I
+
+/(?(1)abz)(1)23/I
+
+/(a)+/I
+
+/(a){2,3}/I
+
+/(a)*/I
+
+/[a]/I
+
+/[ab]/I
+
+/[ab]/IS
+
+/[^a]/I
+
+/\d456/I
+
+/\d456/IS
+
+/a^b/I
+
+/^a/Im
+  abcde
+  xy\nabc
+  *** Failers
+  xyabc
+
+/c|abc/I
+
+/(?i)[ab]/IS
+
+/[ab](?i)cd/IS
+
+/abc(?C)def/I
+    abcdef
+    1234abcdef
+    *** Failers
+    abcxyz
+    abcxyzf
+
+/abc(?C)de(?C1)f/I
+    123abcdef
+
+/(?C1)\dabc(?C2)def/IS
+    1234abcdef
+    *** Failers
+    abcdef
+
+/(?C1)\dabc(?C2)def/ISS
+    1234abcdef
+    *** Failers
+    abcdef
+
+/(?C255)ab/I
+
+/(?C256)ab/I
+
+/(?Cab)xx/I
+
+/(?C12vr)x/I
+
+/abc(?C)def/I
+    *** Failers
+    \x83\x0\x61bcdef
+
+/(abc)(?C)de(?C1)f/I
+    123abcdef
+    123abcdef\C+
+    123abcdef\C-
+    *** Failers
+    123abcdef\C!1
+
+/(?C0)(abc(?C1))*/I
+    abcabcabc
+    abcabc\C!1!3
+    *** Failers
+    abcabcabc\C!1!3
+
+/(\d{3}(?C))*/I
+    123\C+
+    123456\C+
+    123456789\C+
+
+/((xyz)(?C)p|(?C1)xyzabc)/I
+    xyzabc\C+
+
+/(X)((xyz)(?C)p|(?C1)xyzabc)/I
+    Xxyzabc\C+
+
+/(?=(abc))(?C)abcdef/I
+    abcdef\C+
+
+/(?!(abc)(?C1)d)(?C2)abcxyz/I
+    abcxyz\C+
+
+/(?<=(abc)(?C))xyz/I
+   abcxyz\C+
+
+/a(b+)(c*)(?C1)/I
+    abbbbbccc\C*1
+
+/a(b+?)(c*?)(?C1)/I
+    abbbbbccc\C*1
+
+/(?C)abc/I
+
+/(?C)^abc/I
+
+/(?C)a|b/IS
+
+/(?R)/I
+
+/(a|(?R))/I
+
+/(ab|(bc|(de|(?R))))/I
+
+/x(ab|(bc|(de|(?R))))/I
+    xab
+    xbc
+    xde
+    xxab
+    xxxab
+    *** Failers
+    xyab
+
+/(ab|(bc|(de|(?1))))/I
+
+/x(ab|(bc|(de|(?1)x)x)x)/I
+
+/^([^()]|\((?1)*\))*$/I
+    abc
+    a(b)c
+    a(b(c))d
+    *** Failers)
+    a(b(c)d
+
+/^>abc>([^()]|\((?1)*\))*abc>123abc>1(2)3abc>(1(2)3)]*+) | (?2)) * >))/Ix
+    <>
+    
+     hij>
+     hij>
+    def>
+    
+    *** Failers
+    b|c)d(?Pe)/DZ
+    abde
+    acde
+
+/(?:a(?Pc(?Pd)))(?Pa)/DZ
+
+/(?Pa)...(?P=a)bbb(?P>a)d/DZ
+
+/^\W*(?:(?P(?P.)\W*(?P>one)\W*(?P=two)|)|(?P(?P.)\W*(?P>three)\W*(?P=four)|\W*.\W*))\W*$/Ii
+    1221
+    Satan, oscillate my metallic sonatas!
+    A man, a plan, a canal: Panama!
+    Able was I ere I saw Elba.
+    *** Failers
+    The quick brown fox
+
+/((?(R)a|b))\1(?1)?/I
+  bb
+  bbaa
+
+/(.*)a/Is
+
+/(.*)a\1/Is
+
+/(.*)a(b)\2/Is
+
+/((.*)a|(.*)b)z/Is
+
+/((.*)a|(.*)b)z\1/Is
+
+/((.*)a|(.*)b)z\2/Is
+
+/((.*)a|(.*)b)z\3/Is
+
+/((.*)a|^(.*)b)z\3/Is
+
+/(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)a/Is
+
+/(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)a\31/Is
+
+/(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)a\32/Is
+
+/(a)(bc)/INDZ
+  abc
+
+/(?Pa)(bc)/INDZ
+  abc
+
+/(a)(?Pbc)/INDZ
+
+/(a+)*zz/I
+  aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaazzbbbbbb\M
+  aaaaaaaaaaaaaz\M
+
+/(aaa(?C1)bbb|ab)/I
+   aaabbb
+   aaabbb\C*0
+   aaabbb\C*1
+   aaabbb\C*-1
+
+/ab(?Pcd)ef(?Pgh)/I
+    abcdefgh
+    abcdefgh\C1\Gtwo
+    abcdefgh\Cone\Ctwo
+    abcdefgh\Cthree
+
+/(?P)(?P)/DZ
+
+/(?P)(?P)/DZ
+
+/(?Pzz)(?Paa)/I
+    zzaa\CZ
+    zzaa\CA
+
+/(?Peks)(?Peccs)/I
+
+/(?Pabc(?Pdef)(?Pxyz))/I
+
+"\[((?P\d+)(,(?P>elem))*)\]"I
+    [10,20,30,5,5,4,4,2,43,23,4234]
+    *** Failers
+    []
+
+"\[((?P\d+)(,(?P>elem))*)?\]"I
+    [10,20,30,5,5,4,4,2,43,23,4234]
+    []
+
+/(a(b(?2)c))?/DZ
+
+/(a(b(?2)c))*/DZ
+
+/(a(b(?2)c)){0,2}/DZ
+
+/[ab]{1}+/DZ
+
+/((w\/|-|with)*(free|immediate)*.*?shipping\s*[!.-]*)/Ii
+     Baby Bjorn Active Carrier - With free SHIPPING!!
+
+/((w\/|-|with)*(free|immediate)*.*?shipping\s*[!.-]*)/IiS
+     Baby Bjorn Active Carrier - With free SHIPPING!!
+
+/a*.*b/ISDZ
+
+/(a|b)*.?c/ISDZ
+
+/abc(?C255)de(?C)f/DZ
+
+/abcde/ICDZ
+  abcde
+  abcdfe
+
+/a*b/ICDZS
+  ab
+  aaaab
+  aaaacb
+
+/a*b/ICDZSS
+  ab
+  aaaab
+  aaaacb
+
+/a+b/ICDZ
+  ab
+  aaaab
+  aaaacb
+
+/(abc|def)x/ICDZS
+  abcx
+  defx
+  ** Failers 
+  abcdefzx
+
+/(abc|def)x/ICDZSS
+  abcx
+  defx
+  ** Failers 
+  abcdefzx
+
+/(ab|cd){3,4}/IC
+  ababab
+  abcdabcd
+  abcdcdcdcdcd
+
+/([ab]{,4}c|xy)/ICDZS
+    Note: that { does NOT introduce a quantifier
+
+/([ab]{,4}c|xy)/ICDZSS
+    Note: that { does NOT introduce a quantifier
+
+/([ab]{1,4}c|xy){4,5}?123/ICDZ
+    aacaacaacaacaac123
+
+/\b.*/I
+  ab cd\>1
+
+/\b.*/Is
+  ab cd\>1
+
+/(?!.bcd).*/I
+  Xbcd12345
+
+/abcde/I
+    ab\P
+    abc\P
+    abcd\P
+    abcde\P
+    the quick brown abc\P
+    ** Failers\P
+    the quick brown abxyz fox\P
+
+"^(0?[1-9]|[12][0-9]|3[01])/(0?[1-9]|1[012])/(20)?\d\d$"I
+    13/05/04\P
+    13/5/2004\P
+    02/05/09\P
+    1\P
+    1/2\P
+    1/2/0\P
+    1/2/04\P
+    0\P
+    02/\P
+    02/0\P
+    02/1\P
+    ** Failers\P
+    \P
+    123\P
+    33/4/04\P
+    3/13/04\P
+    0/1/2003\P
+    0/\P
+    02/0/\P
+    02/13\P
+
+/0{0,2}ABC/I
+
+/\d{3,}ABC/I
+
+/\d*ABC/I
+
+/[abc]+DE/I
+
+/[abc]?123/I
+    123\P
+    a\P
+    b\P
+    c\P
+    c12\P
+    c123\P
+
+/^(?:\d){3,5}X/I
+    1\P
+    123\P
+    123X
+    1234\P
+    1234X
+    12345\P
+    12345X
+    *** Failers
+    1X
+    123456\P
+
+/abc/IS>testsavedregex
+testsavedregex
+testsavedregex
+testsavedregex
+testsavedregex
+testsavedregex
+testsavedregex
+testsavedregex
+(.)*~smgI
+    \J1024\n\n\nPartner der LCO\nde\nPartner der LINEAS Consulting\nGmbH\nLINEAS Consulting GmbH Hamburg\nPartnerfirmen\n30 days\nindex,follow\n\nja\n3\nPartner\n\n\nLCO\nLINEAS Consulting\n15.10.2003\n\n\n\n\nDie Partnerfirmen der LINEAS Consulting\nGmbH\n\n\n \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
+
+/^a/IF
+
+/line\nbreak/I
+    this is a line\nbreak
+    line one\nthis is a line\nbreak in the second line
+
+/line\nbreak/If
+    this is a line\nbreak
+    ** Failers
+    line one\nthis is a line\nbreak in the second line
+
+/line\nbreak/Imf
+    this is a line\nbreak
+    ** Failers
+    line one\nthis is a line\nbreak in the second line
+
+/(?i)(?-i)AbCd/I
+    AbCd
+    ** Failers
+    abcd
+
+/a{11111111111111111111}/I
+
+/(){64294967295}/I
+
+/(){2,4294967295}/I
+
+"(?i:a)(?i:b)(?i:c)(?i:d)(?i:e)(?i:f)(?i:g)(?i:h)(?i:i)(?i:j)(k)(?i:l)A\1B"I
+    abcdefghijklAkB
+
+"(?Pa)(?Pb)(?Pc)(?Pd)(?Pe)(?Pf)(?Pg)(?Ph)(?Pi)(?Pj)(?Pk)(?Pl)A\11B"I
+    abcdefghijklAkB
+
+"(a)(b)(c)(d)(e)(f)(g)(h)(i)(j)(k)(l)A\11B"I
+    abcdefghijklAkB
+
+"(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)"I
+    aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+
+"(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)"I
+    aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+
+/[^()]*(?:\((?R)\)[^()]*)*/I
+    (this(and)that
+    (this(and)that)
+    (this(and)that)stuff
+
+/[^()]*(?:\((?>(?R))\)[^()]*)*/I
+    (this(and)that
+    (this(and)that)
+
+/[^()]*(?:\((?R)\))*[^()]*/I
+    (this(and)that
+    (this(and)that)
+
+/(?:\((?R)\))*[^()]*/I
+    (this(and)that
+    (this(and)that)
+    ((this))
+
+/(?:\((?R)\))|[^()]*/I
+    (this(and)that
+    (this(and)that)
+    (this)
+    ((this))
+
+/\x{0000ff}/I
+
+/^((?Pa1)|(?Pa2)b)/I
+
+/^((?Pa1)|(?Pa2)b)/IJ
+    a1b\CA
+    a2b\CA
+    ** Failers
+    a1b\CZ\CA
+
+/^(?Pa)(?Pb)/IJ
+    ab\CA
+
+/^(?Pa)(?Pb)|cd/IJ
+    ab\CA
+    cd\CA
+
+/^(?Pa)(?Pb)|cd(?Pef)(?Pgh)/IJ
+    cdefgh\CA
+
+/^((?Pa1)|(?Pa2)b)/IJ
+    a1b\GA
+    a2b\GA
+    ** Failers
+    a1b\GZ\GA
+
+/^(?Pa)(?Pb)/IJ
+    ab\GA
+
+/^(?Pa)(?Pb)|cd/IJ
+    ab\GA
+    cd\GA
+
+/^(?Pa)(?Pb)|cd(?Pef)(?Pgh)/IJ
+    cdefgh\GA
+
+/(?J)^((?Pa1)|(?Pa2)b)/I
+    a1b\CA
+    a2b\CA
+
+/^(?Pa) (?J:(?Pb)(?Pc)) (?Pd)/I
+
+/ In this next test, J is not set at the outer level; consequently it isn't
+set in the pattern's options; consequently pcre_get_named_substring() produces
+a random value. /Ix
+
+/^(?Pa) (?J:(?Pb)(?Pc)) (?Pd)/I
+    a bc d\CA\CB\CC
+
+/^(?Pa)?(?(A)a|b)/I
+    aabc
+    bc
+    ** Failers
+    abc
+
+/(?:(?(ZZ)a|b)(?PX))+/I
+    bXaX
+
+/(?:(?(2y)a|b)(X))+/I
+
+/(?:(?(ZA)a|b)(?PX))+/I
+
+/(?:(?(ZZ)a|b)(?(ZZ)a|b)(?PX))+/I
+    bbXaaX
+
+/(?:(?(ZZ)a|\(b\))\\(?PX))+/I
+    (b)\\Xa\\X
+
+/(?PX|Y))+/I
+    bXXaYYaY
+    bXYaXXaX
+
+/()()()()()()()()()(?:(?(A)(?P=A)a|b)(?PX|Y))+/I
+    bXXaYYaY
+
+/\s*,\s*/IS
+    \x0b,\x0b
+    \x0c,\x0d
+
+/^abc/Im
+    xyz\nabc
+    xyz\nabc\
+    xyz\r\nabc\
+    xyz\rabc\
+    xyz\r\nabc\
+    ** Failers
+    xyz\nabc\
+    xyz\r\nabc\
+    xyz\nabc\
+    xyz\rabc\
+    xyz\rabc\
+
+/abc$/Im
+    xyzabc
+    xyzabc\n
+    xyzabc\npqr
+    xyzabc\r\
+    xyzabc\rpqr\
+    xyzabc\r\n\
+    xyzabc\r\npqr\
+    ** Failers
+    xyzabc\r
+    xyzabc\rpqr
+    xyzabc\r\n
+    xyzabc\r\npqr
+
+/^abc/Im
+    xyz\rabcdef
+    xyz\nabcdef\
+    ** Failers
+    xyz\nabcdef
+
+/^abc/Im
+    xyz\nabcdef
+    xyz\rabcdef\
+    ** Failers
+    xyz\rabcdef
+
+/^abc/Im
+    xyz\r\nabcdef
+    xyz\rabcdef\
+    ** Failers
+    xyz\rabcdef
+
+/^abc/Im
+
+/abc/I
+    xyz\rabc\
+    abc
+
+/.*/I
+    abc\ndef
+    abc\rdef
+    abc\r\ndef
+    \abc\ndef
+    \abc\rdef
+    \abc\r\ndef
+    \abc\ndef
+    \abc\rdef
+    \abc\r\ndef
+
+/\w+(.)(.)?def/Is
+    abc\ndef
+    abc\rdef
+    abc\r\ndef
+
++((?:\s|//.*\\n|/[*](?:\\n|.)*?[*]/)*)+I
+   /* this is a C style comment */\M
+
+/(?P25[0-5]|2[0-4]\d|[01]?\d?\d)(?:\.(?P>B)){3}/I
+
+/()()()()()()()()()()()()()()()()()()()()
+ ()()()()()()()()()()()()()()()()()()()()
+ ()()()()()()()()()()()()()()()()()()()()
+ ()()()()()()()()()()()()()()()()()()()()
+ ()()()()()()()()()()()()()()()()()()()()
+ (.(.))/Ix
+    XY\O400
+
+/(a*b|(?i:c*(?-i)d))/IS
+
+/()[ab]xyz/IS
+
+/(|)[ab]xyz/IS
+
+/(|c)[ab]xyz/IS
+
+/(|c?)[ab]xyz/IS
+
+/(d?|c?)[ab]xyz/IS
+
+/(d?|c)[ab]xyz/IS
+
+/^a*b\d/DZ
+
+/^a*+b\d/DZ
+
+/^a*?b\d/DZ
+
+/^a+A\d/DZ
+    aaaA5
+    ** Failers
+    aaaa5
+
+/^a*A\d/IiDZ
+    aaaA5
+    aaaa5
+
+/(a*|b*)[cd]/IS
+
+/(a+|b*)[cd]/IS
+
+/(a*|b+)[cd]/IS
+
+/(a+|b+)[cd]/IS
+
+/((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((
+ ((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((
+ (((
+ a
+ ))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))
+ ))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))
+ )))
+/Ix
+  large nest
+
+/a*\d/BZ
+
+/a*\D/BZ
+
+/0*\d/BZ
+
+/0*\D/BZ
+
+/a*\s/BZ
+
+/a*\S/BZ
+
+/ *\s/BZ
+
+/ *\S/BZ
+
+/a*\w/BZ
+
+/a*\W/BZ
+
+/=*\w/BZ
+
+/=*\W/BZ
+
+/\d*a/BZ
+
+/\d*2/BZ
+
+/\d*\d/BZ
+
+/\d*\D/BZ
+
+/\d*\s/BZ
+
+/\d*\S/BZ
+
+/\d*\w/BZ
+
+/\d*\W/BZ
+
+/\D*a/BZ
+
+/\D*2/BZ
+
+/\D*\d/BZ
+
+/\D*\D/BZ
+
+/\D*\s/BZ
+
+/\D*\S/BZ
+
+/\D*\w/BZ
+
+/\D*\W/BZ
+
+/\s*a/BZ
+
+/\s*2/BZ
+
+/\s*\d/BZ
+
+/\s*\D/BZ
+
+/\s*\s/BZ
+
+/\s*\S/BZ
+
+/\s*\w/BZ
+
+/\s*\W/BZ
+
+/\S*a/BZ
+
+/\S*2/BZ
+
+/\S*\d/BZ
+
+/\S*\D/BZ
+
+/\S*\s/BZ
+
+/\S*\S/BZ
+
+/\S*\w/BZ
+
+/\S*\W/BZ
+
+/\w*a/BZ
+
+/\w*2/BZ
+
+/\w*\d/BZ
+
+/\w*\D/BZ
+
+/\w*\s/BZ
+
+/\w*\S/BZ
+
+/\w*\w/BZ
+
+/\w*\W/BZ
+
+/\W*a/BZ
+
+/\W*2/BZ
+
+/\W*\d/BZ
+
+/\W*\D/BZ
+
+/\W*\s/BZ
+
+/\W*\S/BZ
+
+/\W*\w/BZ
+
+/\W*\W/BZ
+
+/[^a]+a/BZ
+
+/[^a]+a/BZi
+
+/[^a]+A/BZi
+
+/[^a]+b/BZ
+
+/[^a]+\d/BZ
+
+/a*[^a]/BZ
+
+/(?Px)(?Py)/I
+    xy\Cabc\Cxyz
+
+/(?x)(?'xyz'y)/I
+    xy\Cabc\Cxyz
+
+/(?x)(?'xyz>y)/I
+
+/(?P'abc'x)(?Py)/I
+
+/^(?:(?(ZZ)a|b)(?X))+/
+    bXaX
+    bXbX
+    ** Failers
+    aXaX
+    aXbX
+
+/^(?P>abc)(?xxx)/
+
+/^(?P>abc)(?x|y)/
+    xx
+    xy
+    yy
+    yx
+
+/^(?P>abc)(?Px|y)/
+    xx
+    xy
+    yy
+    yx
+
+/^((?(abc)a|b)(?x|y))+/
+    bxay
+    bxby
+    ** Failers
+    axby
+
+/^(((?P=abc)|X)(?x|y))+/
+    XxXxxx
+    XxXyyx
+    XxXyxx
+    ** Failers
+    x
+
+/^(?1)(abc)/
+    abcabc
+
+/^(?:(?:\1|X)(a|b))+/
+    Xaaa
+    Xaba
+
+/^[\E\Qa\E-\Qz\E]+/BZ
+
+/^[a\Q]bc\E]/BZ
+
+/^[a-\Q\E]/BZ
+
+/^(?P>abc)[()](?)/BZ
+
+/^((?(abc)y)[()](?Px))+/BZ
+    (xy)x
+
+/^(?P>abc)\Q()\E(?)/BZ
+
+/^(?P>abc)[a\Q(]\E(](?)/BZ
+
+/^(?P>abc) # this is (a comment)
+  (?)/BZx
+
+/^\W*(?:(?(?.)\W*(?&one)\W*\k|)|(?(?.)\W*(?&three)\W*\k'four'|\W*.\W*))\W*$/Ii
+    1221
+    Satan, oscillate my metallic sonatas!
+    A man, a plan, a canal: Panama!
+    Able was I ere I saw Elba.
+    *** Failers
+    The quick brown fox
+
+/(?=(\w+))\1:/I
+    abcd:
+
+/(?=(?'abc'\w+))\k:/I
+    abcd:
+
+/(?'abc'a|b)(?d|e)\k{2}/J
+    adaa
+    ** Failers
+    addd
+    adbb
+
+/(?'abc'a|b)(?d|e)(?&abc){2}/J
+    bdaa
+    bdab
+    ** Failers
+    bddd
+
+/(?( (?'B' abc (?(R) (?(R&A)1) (?(R&B)2) X  |  (?1)  (?2)   (?R) ))) /x
+    abcabc1Xabc2XabcXabcabc
+
+/(? (?'B' abc (?(R) (?(R&1)1) (?(R&B)2) X  |  (?1)  (?2)   (?R) ))) /x
+
+/(?<1> (?'B' abc (?(R) (?(R&1)1) (?(R&B)2) X  |  (?1)  (?2)   (?R) ))) /x
+    abcabc1Xabc2XabcXabcabc
+
+/^(?(DEFINE) abc | xyz ) /x
+
+/(?(DEFINE) abc) xyz/xI
+
+/(a|)*\d/
+  \O0aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+  \O0aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa4
+
+/^a.b/
+    a\rb
+    a\nb\
+    a\x85b\ 
+    ** Failers
+    a\nb
+    a\nb\
+    a\rb\
+    a\rb\
+    a\x85b\ 
+    a\rb\
+
+/^abc./mgx
+    abc1 \x0aabc2 \x0babc3xx \x0cabc4 \x0dabc5xx \x0d\x0aabc6 \x85abc7 JUNK
+
+/abc.$/mgx
+    abc1\x0a abc2\x0b abc3\x0c abc4\x0d abc5\x0d\x0a abc6\x85 abc7 abc9
+
+/a/
+
+/a/
+
+/^a\Rb/
+    a\nb
+    a\rb
+    a\r\nb
+    a\x0bb
+    a\x0cb
+    a\x85b
+    ** Failers
+    a\n\rb
+
+/^a\R*b/
+    ab
+    a\nb
+    a\rb
+    a\r\nb
+    a\x0bb
+    a\x0cb
+    a\x85b
+    a\n\rb
+    a\n\r\x85\x0cb
+
+/^a\R+b/
+    a\nb
+    a\rb
+    a\r\nb
+    a\x0bb
+    a\x0cb
+    a\x85b
+    a\n\rb
+    a\n\r\x85\x0cb
+    ** Failers
+    ab
+
+/^a\R{1,3}b/
+    a\nb
+    a\n\rb
+    a\n\r\x85b
+    a\r\n\r\nb
+    a\r\n\r\n\r\nb
+    a\n\r\n\rb
+    a\n\n\r\nb
+    ** Failers
+    a\n\n\n\rb
+    a\r
+
+/^a[\R]b/
+    aRb
+    ** Failers
+    a\nb
+
+/(?&abc)X(?P)/I
+    abcPXP123
+
+/(?1)X(?P)/I
+    abcPXP123
+
+/(?:a(?&abc)b)*(?x)/
+    123axbaxbaxbx456
+    123axbaxbaxb456
+
+/(?:a(?&abc)b){1,5}(?x)/
+    123axbaxbaxbx456
+
+/(?:a(?&abc)b){2,5}(?x)/
+    123axbaxbaxbx456
+
+/(?:a(?&abc)b){2,}(?x)/
+    123axbaxbaxbx456
+
+/(abc)(?i:(?1))/
+   defabcabcxyz
+   DEFabcABCXYZ
+
+/(abc)(?:(?i)(?1))/
+   defabcabcxyz
+   DEFabcABCXYZ
+
+/^(a)\g-2/
+
+/^(a)\g/
+
+/^(a)\g{0}/
+
+/^(a)\g{3/
+
+/^(a)\g{4a}/
+
+/^a.b/
+    a\rb
+    *** Failers
+    a\nb
+
+/.+foo/
+    afoo
+    ** Failers
+    \r\nfoo
+    \nfoo
+
+/.+foo/
+    afoo
+    \nfoo
+    ** Failers
+    \r\nfoo
+
+/.+foo/
+    afoo
+    ** Failers
+    \nfoo
+    \r\nfoo
+
+/.+foo/s
+    afoo
+    \r\nfoo
+    \nfoo
+    
+/^$/mg
+    abc\r\rxyz
+    abc\n\rxyz  
+    ** Failers 
+    abc\r\nxyz
+
+/(?m)^$/g+
+    abc\r\n\r\n
+
+/(?m)^$|^\r\n/g+ 
+    abc\r\n\r\n
+    
+/(?m)$/g+ 
+    abc\r\n\r\n
+
+/abc.$/mgx
+    abc1\x0a abc2\x0b abc3\x0c abc4\x0d abc5\x0d\x0a abc6\x85 abc9
+
+/^X/m
+    XABC
+    ** Failers 
+    XABC\B
+
+/(ab|c)(?-1)/BZ
+    abc
+
+/xy(?+1)(abc)/BZ
+    xyabcabc
+    ** Failers
+    xyabc  
+    
+/x(?-0)y/
+
+/x(?-1)y/
+
+/x(?+0)y/ 
+
+/x(?+1)y/
+
+/^(abc)?(?(-1)X|Y)/BZ
+    abcX
+    Y
+    ** Failers
+    abcY   
+    
+/^((?(+1)X|Y)(abc))+/BZ 
+    YabcXabc
+    YabcXabcXabc
+    ** Failers
+    XabcXabc  
+
+/(?(-1)a)/BZ
+
+/((?(-1)a))/BZ
+
+/((?(-2)a))/BZ
+
+/^(?(+1)X|Y)(.)/BZ
+    Y!
+
+/(?tom|bon)-\k{A}/
+    tom-tom
+    bon-bon 
+    ** Failers
+    tom-bon  
+
+/\g{A/ 
+
+/(?|(abc)|(xyz))/BZ
+   >abc<
+   >xyz< 
+
+/(x)(?|(abc)|(xyz))(x)/BZ
+    xabcx
+    xxyzx 
+
+/(x)(?|(abc)(pqr)|(xyz))(x)/BZ
+    xabcpqrx
+    xxyzx 
+
+/\H++X/BZ
+    ** Failers
+    XXXX
+    
+/\H+\hY/BZ
+    XXXX Y 
+
+/\H+ Y/BZ
+
+/\h+A/BZ
+
+/\v*B/BZ
+
+/\V+\x0a/BZ
+
+/A+\h/BZ
+
+/ *\H/BZ
+
+/A*\v/BZ
+
+/\x0b*\V/BZ
+
+/\d+\h/BZ
+
+/\d*\v/BZ
+
+/S+\h\S+\v/BZ
+
+/\w{3,}\h\w+\v/BZ
+
+/\h+\d\h+\w\h+\S\h+\H/BZ
+
+/\v+\d\v+\w\v+\S\v+\V/BZ
+
+/\H+\h\H+\d/BZ
+
+/\V+\v\V+\w/BZ
+
+/\( (?: [^()]* | (?R) )* \)/x
+\J1024(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(00)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)
+
+/[\E]AAA/
+
+/[\Q\E]AAA/
+
+/[^\E]AAA/
+
+/[^\Q\E]AAA/
+
+/[\E^]AAA/
+
+/[\Q\E^]AAA/
+
+/A(*PRUNE)B(*SKIP)C(*THEN)D(*COMMIT)E(*F)F(*FAIL)G(?!)H(*ACCEPT)I/BZ
+
+/^a+(*FAIL)/C
+    aaaaaa
+    
+/a+b?c+(*FAIL)/C
+    aaabccc
+
+/a+b?(*PRUNE)c+(*FAIL)/C
+    aaabccc
+
+/a+b?(*COMMIT)c+(*FAIL)/C
+    aaabccc
+    
+/a+b?(*SKIP)c+(*FAIL)/C
+    aaabcccaaabccc
+
+/a+b?(*THEN)c+(*FAIL)/C
+    aaabccc
+    
+/a(*MARK)b/ 
+
+/(?i:A{1,}\6666666666)/
+
+/\g6666666666/
+
+/[\g6666666666]/BZ
+
+/(?1)\c[/
+
+/.+A/
+    \r\nA
+    
+/\nA/
+    \r\nA 
+
+/[\r\n]A/
+    \r\nA 
+
+/(\r|\n)A/
+    \r\nA 
+
+/a(*CR)b/
+
+/(*CR)a.b/
+    a\nb
+    ** Failers
+    a\rb  
+
+/(*CR)a.b/
+    a\nb
+    ** Failers
+    a\rb  
+
+/(*LF)a.b/
+    a\rb
+    ** Failers
+    a\nb  
+
+/(*CRLF)a.b/
+    a\rb
+    a\nb  
+    ** Failers
+    a\r\nb  
+
+/(*ANYCRLF)a.b/
+    ** Failers
+    a\rb
+    a\nb  
+    a\r\nb  
+
+/(*ANY)a.b/
+    ** Failers
+    a\rb
+    a\nb  
+    a\r\nb  
+    a\x85b 
+    
+/(*ANY).*/g
+    abc\r\ndef
+
+/(*ANYCRLF).*/g
+    abc\r\ndef
+
+/(*CRLF).*/g
+    abc\r\ndef
+
+/a\Rb/I
+    a\rb
+    a\nb
+    a\r\nb
+    ** Failers
+    a\x85b
+    a\x0bb     
+
+/a\Rb/I
+    a\rb
+    a\nb
+    a\r\nb
+    a\x85b
+    a\x0bb     
+    ** Failers 
+    a\x85b\
+    a\x0bb\
+    
+/a\R?b/I
+    a\rb
+    a\nb
+    a\r\nb
+    ** Failers
+    a\x85b
+    a\x0bb     
+
+/a\R?b/I
+    a\rb
+    a\nb
+    a\r\nb
+    a\x85b
+    a\x0bb     
+    ** Failers 
+    a\x85b\
+    a\x0bb\
+    
+/a\R{2,4}b/I
+    a\r\n\nb
+    a\n\r\rb
+    a\r\n\r\n\r\n\r\nb
+    ** Failers
+    a\x85\85b
+    a\x0b\0bb     
+
+/a\R{2,4}b/I
+    a\r\rb
+    a\n\n\nb
+    a\r\n\n\r\rb
+    a\x85\85b
+    a\x0b\0bb     
+    ** Failers 
+    a\r\r\r\r\rb 
+    a\x85\85b\
+    a\x0b\0bb\
+ 
+/(*BSR_ANYCRLF)a\Rb/I
+    a\nb
+    a\rb 
+
+/(*BSR_UNICODE)a\Rb/I
+    a\x85b
+
+/(*BSR_ANYCRLF)(*CRLF)a\Rb/I
+    a\nb
+    a\rb 
+
+/(*CRLF)(*BSR_UNICODE)a\Rb/I
+    a\x85b
+
+/(*CRLF)(*BSR_ANYCRLF)(*CR)ab/I
+
+/(?)(?&)/
+
+/(?)(?&a)/
+
+/(?)(?&aaaaaaaaaaaaaaaaaaaaaaa)/
+
+/(?+-a)/
+
+/(?-+a)/
+
+/(?(-1))/
+
+/(?(+10))/
+
+/(?(10))/
+
+/(?(+2))()()/
+
+/(?(2))()()/
+
+/\k''/
+
+/\k<>/
+
+/\k{}/
+
+/\k/
+
+/\kabc/
+
+/(?P=)/
+
+/(?P>)/
+
+/(?!\w)(?R)/
+
+/(?=\w)(?R)/
+
+/(?x|y){0}z/
+    xzxx
+    yzyy 
+    ** Failers
+    xxz  
+
+/(\3)(\1)(a)/
+    cat
+
+/(\3)(\1)(a)/
+    cat
+    
+/TA]/
+    The ACTA] comes 
+
+/TA]/
+    The ACTA] comes 
+
+/(?2)[]a()b](abc)/
+    abcbabc
+
+/(?2)[^]a()b](abc)/
+    abcbabc
+
+/(?1)[]a()b](abc)/
+    abcbabc
+    ** Failers 
+    abcXabc
+
+/(?1)[^]a()b](abc)/
+    abcXabc
+    ** Failers 
+    abcbabc
+
+/(?2)[]a()b](abc)(xyz)/
+    xyzbabcxyz
+
+/(?&N)[]a(?)](?abc)/
+   abc)](abc)/
+   abc
+    ** Failers
+    ab
+
+/a[]+b/
+    ** Failers
+    ab 
+
+/a[]*+b/
+    ** Failers
+    ab 
+
+/a[^]b/
+    aXb
+    a\nb 
+    ** Failers
+    ab  
+    
+/a[^]+b/ 
+    aXb
+    a\nX\nXb 
+    ** Failers
+    ab  
+
+/a(?!)+b/
+
+/a(*FAIL)+b/
+
+/(abc|pqr|123){0}[xyz]/SI
+
+/(?(?=.*b)b|^)/CI
+   adc
+   abc 
+   
+/(?(?=b).*b|^d)/I
+
+/(?(?=.*b).*b|^d)/I
+
+/xyz/C
+  xyz 
+  abcxyz 
+  abcxyz\Y
+  ** Failers 
+  abc
+  abc\Y
+  abcxypqr  
+  abcxypqr\Y  
+  
+/(*NO_START_OPT)xyz/C
+  abcxyz 
+
+/xyz/CY
+  abcxyz 
+
+/^"((?(?=[a])[^"])|b)*"$/C
+    "ab"
+
+/^"((?(?=[a])[^"])|b)*"$/
+    "ab"
+
+/^X(?5)(a)(?|(b)|(q))(c)(d)Y/
+    XYabcdY
+
+/^X(?&N)(a)(?|(b)|(q))(c)(d)(?Y)/
+    XYabcdY
+ 
+/Xa{2,4}b/
+    X\P
+    Xa\P
+    Xaa\P 
+    Xaaa\P
+    Xaaaa\P 
+    
+/Xa{2,4}?b/
+    X\P
+    Xa\P
+    Xaa\P 
+    Xaaa\P
+    Xaaaa\P 
+    
+/Xa{2,4}+b/
+    X\P
+    Xa\P
+    Xaa\P 
+    Xaaa\P
+    Xaaaa\P 
+    
+/X\d{2,4}b/
+    X\P
+    X3\P
+    X33\P 
+    X333\P
+    X3333\P 
+    
+/X\d{2,4}?b/
+    X\P
+    X3\P
+    X33\P 
+    X333\P
+    X3333\P 
+    
+/X\d{2,4}+b/
+    X\P
+    X3\P
+    X33\P 
+    X333\P
+    X3333\P 
+    
+/X\D{2,4}b/
+    X\P
+    Xa\P
+    Xaa\P 
+    Xaaa\P
+    Xaaaa\P 
+    
+/X\D{2,4}?b/
+    X\P
+    Xa\P
+    Xaa\P 
+    Xaaa\P
+    Xaaaa\P 
+    
+/X\D{2,4}+b/
+    X\P
+    Xa\P
+    Xaa\P 
+    Xaaa\P
+    Xaaaa\P 
+    
+/X[abc]{2,4}b/
+    X\P
+    Xa\P
+    Xaa\P 
+    Xaaa\P
+    Xaaaa\P 
+    
+/X[abc]{2,4}?b/
+    X\P
+    Xa\P
+    Xaa\P 
+    Xaaa\P
+    Xaaaa\P 
+    
+/X[abc]{2,4}+b/
+    X\P
+    Xa\P
+    Xaa\P 
+    Xaaa\P
+    Xaaaa\P 
+    
+/X[^a]{2,4}b/
+    X\P
+    Xz\P
+    Xzz\P 
+    Xzzz\P
+    Xzzzz\P 
+    
+/X[^a]{2,4}?b/
+    X\P
+    Xz\P
+    Xzz\P 
+    Xzzz\P
+    Xzzzz\P 
+    
+/X[^a]{2,4}+b/
+    X\P
+    Xz\P
+    Xzz\P 
+    Xzzz\P
+    Xzzzz\P 
+    
+/(Y)X\1{2,4}b/
+    YX\P
+    YXY\P
+    YXYY\P 
+    YXYYY\P
+    YXYYYY\P 
+    
+/(Y)X\1{2,4}?b/
+    YX\P
+    YXY\P
+    YXYY\P 
+    YXYYY\P
+    YXYYYY\P 
+    
+/(Y)X\1{2,4}+b/
+    YX\P
+    YXY\P
+    YXYY\P 
+    YXYYY\P
+    YXYYYY\P 
+    
+/\++\KZ|\d+X|9+Y/
+    ++++123999\P
+    ++++123999Y\P
+    ++++Z1234\P 
+
+/Z(*F)/
+    Z\P
+    ZA\P 
+    
+/Z(?!)/
+    Z\P 
+    ZA\P 
+
+/dog(sbody)?/
+    dogs\P
+    dogs\P\P 
+    
+/dog(sbody)??/
+    dogs\P
+    dogs\P\P 
+
+/dog|dogsbody/
+    dogs\P
+    dogs\P\P 
+ 
+/dogsbody|dog/
+    dogs\P
+    dogs\P\P 
+
+/\bthe cat\b/
+    the cat\P
+    the cat\P\P
+
+/abc/
+   abc\P
+   abc\P\P
+   
+/abc\K123/
+    xyzabc123pqr
+    xyzabc12\P
+    xyzabc12\P\P
+    
+/(?<=abc)123/
+    xyzabc123pqr 
+    xyzabc12\P
+    xyzabc12\P\P
+
+/\babc\b/
+    +++abc+++
+    +++ab\P
+    +++ab\P\P  
+
+/(?&word)(?&element)(?(DEFINE)(?<[^m][^>]>[^<])(?\w*+))/BZ
+
+/(?&word)(?&element)(?(DEFINE)(?<[^\d][^>]>[^<])(?\w*+))/BZ
+
+/(ab)(x(y)z(cd(*ACCEPT)))pq/BZ
+
+/abc\K/+
+    abcdef
+    abcdef\N\N
+    xyzabcdef\N\N
+    ** Failers
+    abcdef\N 
+    xyzabcdef\N
+    
+/^(?:(?=abc)|abc\K)/+
+    abcdef
+    abcdef\N\N 
+    ** Failers 
+    abcdef\N 
+
+/a?b?/+
+    xyz
+    xyzabc
+    xyzabc\N
+    xyzabc\N\N
+    xyz\N\N    
+    ** Failers 
+    xyz\N 
+
+/^a?b?/+
+    xyz
+    xyzabc
+    ** Failers 
+    xyzabc\N
+    xyzabc\N\N
+    xyz\N\N    
+    xyz\N 
+    
+/^(?a|b\gc)/
+    aaaa
+    bacxxx
+    bbaccxxx 
+    bbbacccxx
+
+/^(?a|b\g'name'c)/
+    aaaa
+    bacxxx
+    bbaccxxx 
+    bbbacccxx
+
+/^(a|b\g<1>c)/
+    aaaa
+    bacxxx
+    bbaccxxx 
+    bbbacccxx
+
+/^(a|b\g'1'c)/
+    aaaa
+    bacxxx
+    bbaccxxx 
+    bbbacccxx
+
+/^(a|b\g'-1'c)/
+    aaaa
+    bacxxx
+    bbaccxxx 
+    bbbacccxx
+
+/(^(a|b\g<-1>c))/
+    aaaa
+    bacxxx
+    bbaccxxx 
+    bbbacccxx
+
+/(?-i:\g)(?i:(?a))/
+    XaaX
+    XAAX 
+
+/(?i:\g)(?-i:(?a))/
+    XaaX
+    ** Failers 
+    XAAX 
+
+/(?-i:\g<+1>)(?i:(a))/
+    XaaX
+    XAAX 
+
+/(?=(?(?#simplesyntax)\$(?[a-zA-Z_\x{7f}-\x{ff}][a-zA-Z0-9_\x{7f}-\x{ff}]*)(?:\[(?[a-zA-Z0-9_\x{7f}-\x{ff}]+|\$\g)\]|->\g(\(.*?\))?)?|(?#simple syntax withbraces)\$\{(?:\g(?\[(?:\g|'(?:\\.|[^'\\])*'|"(?:\g|\\.|[^"\\])*")\])?|\g|\$\{\g\})\}|(?#complexsyntax)\{(?\$(?\g(\g*|\(.*?\))?)(?:->\g)*|\$\g|\$\{\g\})\}))\{/
+
+/(?a|b|c)\g*/
+   abc
+   accccbbb 
+
+/^X(?7)(a)(?|(b)|(q)(r)(s))(c)(d)(Y)/
+    XYabcdY
+
+/(?<=b(?1)|zzz)(a)/
+    xbaax
+    xzzzax 
+
+/(a)(?<=b\1)/
+
+/(a)(?<=b+(?1))/
+
+/(a+)(?<=b(?1))/
+
+/(a(?<=b(?1)))/
+
+/(?<=b(?1))xyz/
+
+/(?<=b(?1))xyz(b+)pqrstuvew/
+
+/(a|bc)\1/SI
+
+/(a|bc)\1{2,3}/SI
+
+/(a|bc)(?1)/SI
+
+/(a|b\1)(a|b\1)/SI
+
+/(a|b\1){2}/SI
+
+/(a|bbbb\1)(a|bbbb\1)/SI
+
+/(a|bbbb\1){2}/SI
+
+/^From +([^ ]+) +[a-zA-Z][a-zA-Z][a-zA-Z] +[a-zA-Z][a-zA-Z][a-zA-Z] +[0-9]?[0-9] +[0-9][0-9]:[0-9][0-9]/SI
+
+/]{0,})>]{0,})>([\d]{0,}\.)(.*)((
([\w\W\s\d][^<>]{0,})|[\s]{0,}))<\/a><\/TD>]{0,})>([\w\W\s\d][^<>]{0,})<\/TD>]{0,})>([\w\W\s\d][^<>]{0,})<\/TD><\/TR>/isIS
+
+"(?>.*/)foo"SI
+
+/(?(?=[^a-z]+[a-z])  \d{2}-[a-z]{3}-\d{2}  |  \d{2}-\d{2}-\d{2} ) /xSI
+
+/(?:(?:(?:(?:(?:(?:(?:(?:(?:(a|b|c))))))))))/iSI
+
+/(?:c|d)(?:)(?:aaaaaaaa(?:)(?:bbbbbbbb)(?:bbbbbbbb(?:))(?:bbbbbbbb(?:)(?:bbbbbbbb)))/SI
+
+/A)|(?B))/I
+    AB\Ca
+    BA\Ca
+
+/(?|(?A)|(?B))/ 
+
+/(?:a(? (?')|(?")) |
+    b(? (?')|(?")) ) 
+    (?('quote')[a-z]+|[0-9]+)/JIx
+    a"aaaaa
+    b"aaaaa 
+    ** Failers 
+    b"11111
+    a"11111 
+    
+/^(?|(a)(b)(c)(?d)|(?e)) (?('D')X|Y)/JDZx
+    abcdX
+    eX
+    ** Failers
+    abcdY
+    ey     
+    
+/(?a) (b)(c)  (?d  (?(R&A)$ | (?4)) )/JDZx
+    abcdd
+    ** Failers
+    abcdde  
+
+/abcd*/
+    xxxxabcd\P
+    xxxxabcd\P\P
+
+/abcd*/i
+    xxxxabcd\P
+    xxxxabcd\P\P
+    XXXXABCD\P
+    XXXXABCD\P\P
+
+/abc\d*/
+    xxxxabc1\P
+    xxxxabc1\P\P
+
+/(a)bc\1*/
+    xxxxabca\P
+    xxxxabca\P\P
+
+/abc[de]*/
+    xxxxabcde\P
+    xxxxabcde\P\P
+
+/-- This is not in the Perl-compatible test because Perl seems currently to be
+    broken and not behaving as specified in that it *does* bumpalong after
+    hitting (*COMMIT). --/
+
+/(?1)(A(*COMMIT)|B)D/
+    ABD
+    XABD
+    BAD
+    ABXABD  
+    ** Failers 
+    ABX 
+    BAXBAD  
+
+/(\3)(\1)(a)/
+    cat
+
+/(\3)(\1)(a)/SI
+    cat
+
+/(\3)(\1)(a)/SI
+    cat
+
+/i(?(DEFINE)(?a))/SI
+    i
+    
+/()i(?(1)a)/SI 
+    ia
+
+/(?i)a(?-i)b|c/BZ
+    XabX
+    XAbX
+    CcC 
+    ** Failers
+    XABX   
+
+/(?i)a(?s)b|c/BZ
+
+/(?i)a(?s-i)b|c/BZ
+
+/^(ab(c\1)d|x){2}$/BZ
+    xabcxd
+    
+/^(?&t)*+(?(DEFINE)(?.))$/BZ
+
+/^(?&t)*(?(DEFINE)(?.))$/BZ
+
+/ -- This one is here because Perl gives the match as "b" rather than "ab". I
+     believe this to be a Perl bug. --/  
+      
+/(?>a\Kb)z|(ab)/
+    ab 
+
+/(?P(?P0|)|(?P>L2)(?P>L1))/
+
+/abc(*MARK:)pqr/
+
+/abc(*:)pqr/
+
+/abc(*FAIL:123)xyz/
+
+/--- This should, and does, fail. In Perl, it does not, which I think is a 
+     bug because replacing the B in the pattern by (B|D) does make it fail. ---/
+
+/A(*COMMIT)B/+K
+    ACABX
+
+/--- These should be different, but in Perl they are not, which I think
+     is a bug in Perl. ---/
+
+/A(*THEN)B|A(*THEN)C/K
+    AC
+
+/A(*PRUNE)B|A(*PRUNE)C/K
+    AC
+    
+/--- Mark names can be duplicated. Perl doesn't give a mark for this one,
+though PCRE does. ---/
+
+/^A(*:A)B|^X(*:A)Y/K
+    ** Failers
+    XAQQ
+    
+/--- COMMIT at the start of a pattern should be the same as an anchor. Perl 
+optimizations defeat this. So does the PCRE optimization unless we disable it 
+with \Y. ---/
+
+/(*COMMIT)ABC/
+    ABCDEFG
+    ** Failers
+    DEFGABC\Y  
+    
+/^(ab (c+(*THEN)cd) | xyz)/x
+    abcccd  
+
+/^(ab (c+(*PRUNE)cd) | xyz)/x
+    abcccd  
+
+/^(ab (c+(*FAIL)cd) | xyz)/x
+    abcccd  
+    
+/--- Perl gets some of these wrong ---/ 
+
+/(?>.(*ACCEPT))*?5/
+    abcde
+
+/(.(*ACCEPT))*?5/
+    abcde
+
+/(.(*ACCEPT))5/
+    abcde
+
+/(.(*ACCEPT))*5/
+    abcde
+
+/A\NB./BZ
+    ACBD
+    *** Failers
+    A\nB
+    ACB\n   
+
+/A\NB./sBZ
+    ACBD
+    ACB\n 
+    *** Failers
+    A\nB  
+  
+/A\NB/
+    A\nB
+    A\rB
+    ** Failers
+    A\r\nB    
+
+/\R+b/BZ
+
+/\R+\n/BZ
+
+/\R+\d/BZ
+
+/\d*\R/BZ
+
+/\s*\R/BZ
+    \x20\x0a
+    \x20\x0d
+    \x20\x0d\x0a
+
+/\S*\R/BZ
+    a\x0a
+
+/X\h*\R/BZ
+    X\x20\x0a
+
+/X\H*\R/BZ
+    X\x0d\x0a
+
+/X\H+\R/BZ
+    X\x0d\x0a
+
+/X\H++\R/BZ
+    X\x0d\x0a
+
+/(?<=abc)def/
+    abc\P\P
+
+/abc$/
+    abc
+    abc\P
+    abc\P\P
+
+/abc$/m
+    abc
+    abc\n
+    abc\P\P
+    abc\n\P\P 
+    abc\P
+    abc\n\P
+
+/abc\z/
+    abc
+    abc\P
+    abc\P\P
+
+/abc\Z/
+    abc
+    abc\P
+    abc\P\P
+
+/abc\b/
+    abc
+    abc\P
+    abc\P\P
+
+/abc\B/
+    abc
+    abc\P
+    abc\P\P
+
+/.+/
+    abc\>0
+    abc\>1
+    abc\>2
+    abc\>3
+    abc\>4
+    abc\>-4 
+
+/^\cģ/
+
+/(?P(?P=abn)xxx)/BZ
+
+/(a\1z)/BZ
+
+/(?P(?P=abn)(?(?P=axn)xxx)/BZ
+
+/(?P(?P=axn)xxx)(?yy)/BZ
+
+/-- These tests are here because Perl gets the first one wrong. --/
+
+/(\R*)(.)/s
+    \r\n
+    \r\r\n\n\r 
+    \r\r\n\n\r\n 
+
+/(\R)*(.)/s
+    \r\n
+    \r\r\n\n\r 
+    \r\r\n\n\r\n 
+
+/((?>\r\n|\n|\x0b|\f|\r|\x85)*)(.)/s
+    \r\n
+    \r\r\n\n\r 
+    \r\r\n\n\r\n 
+
+/-- --/
+
+/^abc$/BZ
+
+/^abc$/BZm
+
+/^(a)*+(\w)/S
+    aaaaX
+    ** Failers 
+    aaaa
+
+/^(?:a)*+(\w)/S
+    aaaaX
+    ** Failers 
+    aaaa
+
+/(a)++1234/SDZ
+
+/([abc])++1234/SI
+
+/(?<=(abc)+)X/
+
+/(^ab)/I
+
+/(^ab)++/I
+
+/(^ab|^)+/I
+
+/(^ab|^)++/I
+
+/(?:^ab)/I
+
+/(?:^ab)++/I
+
+/(?:^ab|^)+/I
+
+/(?:^ab|^)++/I
+
+/(.*ab)/I
+
+/(.*ab)++/I
+
+/(.*ab|.*)+/I
+
+/(.*ab|.*)++/I
+
+/(?:.*ab)/I
+
+/(?:.*ab)++/I
+
+/(?:.*ab|.*)+/I
+
+/(?:.*ab|.*)++/I
+
+/(?=a)[bcd]/I
+
+/((?=a))[bcd]/I
+
+/((?=a))+[bcd]/I
+
+/((?=a))++[bcd]/I
+
+/(?=a+)[bcd]/iI
+
+/(?=a+?)[bcd]/iI
+
+/(?=a++)[bcd]/iI
+
+/(?=a{3})[bcd]/iI
+
+/(abc)\1+/S
+
+/-- Perl doesn't get these right IMO (the 3rd is PCRE-specific) --/
+
+/(?1)(?:(b(*ACCEPT))){0}/
+    b
+
+/(?1)(?:(b(*ACCEPT))){0}c/
+    bc
+    ** Failers 
+    b 
+
+/(?1)(?:((*ACCEPT))){0}c/
+    c
+    c\N 
+
+/^.*?(?(?=a)a|b(*THEN)c)/
+    ba
+
+/^.*?(?(?=a)a|bc)/
+    ba
+
+/^.*?(?(?=a)a(*THEN)b|c)/
+    ac
+
+/^.*?(?(?=a)a(*THEN)b)c/
+    ac
+
+/^.*?(a(*THEN)b)c/
+    aabc
+    
+/^.*? (?1) c (?(DEFINE)(a(*THEN)b))/x
+    aabc
+
+/^.*?(a(*THEN)b|z)c/
+    aabc
+
+/^.*?(z|a(*THEN)b)c/
+    aabc
+
+/-- --/
+
+/-- These studied versions are here because they are not Perl-compatible; the
+    studying means the mark is not seen. --/
+
+/(*MARK:A)(*SKIP:B)(C|X)/KS
+    C
+    D
+     
+/(*:A)A+(*SKIP:A)(B|Z)/KS
+    AAAC
+
+/-- --/
+
+"(?=a*(*ACCEPT)b)c"
+    c
+    c\N 
+    
+/(?1)c(?(DEFINE)((*ACCEPT)b))/
+    c
+    c\N  
+    
+/(?>(*ACCEPT)b)c/
+    c
+    c\N  
+
+/(?:(?>(a)))+a%/++
+    %aa%
+
+/(a)b|ac/++SS
+    ac\O3
+    
+/(a)(b)x|abc/++
+     abc\O6
+
+/(a)bc|(a)(b)\2/
+    \O3abc
+    \O4abc 
+
+/(?(DEFINE)(a(?2)|b)(b(?1)|a))(?:(?1)|(?2))/SI
+
+/(a(?2)|b)(b(?1)|a)(?:(?1)|(?2))/SI
+
+/(a(?2)|b)(b(?1)|a)(?1)(?2)/SI
+
+/(abc)(?1)/SI
+
+/^(?>a)++/
+    aa\M
+    aaaaaaaaa\M 
+    
+/(a)(?1)++/
+    aa\M
+    aaaaaaaaa\M  
+
+/(?:(foo)|(bar)|(baz))X/SS=
+    bazfooX
+    foobazbarX
+    barfooX
+    bazX
+    foobarbazX    
+    bazfooX\O0
+    bazfooX\O2
+    bazfooX\O4
+    bazfooX\O6
+    bazfooX\O8
+    bazfooX\O10
+
+/(?=abc){3}abc/BZ
+
+/(?=abc)+abc/BZ
+
+/(?=abc)++abc/BZ
+
+/(?=abc){0}xyz/BZ
+
+/(?=(a))?./BZ
+
+/(?=(a))??./BZ
+
+/^(?=(a)){0}b(?1)/BZ
+
+/(?(DEFINE)(a))?b(?1)/BZ
+
+/^(?=(?1))?[az]([abc])d/BZ
+
+/^(?!a){0}\w+/BZ
+
+/(?<=(abc))?xyz/BZ
+
+/[:a[:abc]b:]/BZ
+
+/((?2))((?1))/SS
+    abc
+
+/((?(R2)a+|(?1)b))/SS
+    aaaabcde
+
+/(?(R)a*(?1)|((?R))b)/SS
+    aaaabcde
+
+/(a+|(?R)b)/
+
+/^(a(*:A)(d|e(*:B))z|aeq)/C
+    adz
+    aez
+    aeqwerty
+
+/.(*F)/
+    \P\Pabc
+
+/\btype\b\W*?\btext\b\W*?\bjavascript\b/IS
+
+/\btype\b\W*?\btext\b\W*?\bjavascript\b|\burl\b\W*?\bshell:|a+)(?>(z+))\w/BZ
+    aaaazzzzb
+    ** Failers
+    aazz  
+
+/(.)(\1|a(?2))/
+    bab
+    
+/\1|(.)(?R)\1/
+    cbbbc
+    
+/(.)((?(1)c|a)|a(?2))/
+    baa  
+
+/(?P(?P=abn)xxx)/BZ
+
+/(a\1z)/BZ
+
+/^(?>a+)(?>b+)(?>c+)(?>d+)(?>e+)/
+     \Maabbccddee
+
+/^(?>(a+))(?>(b+))(?>(c+))(?>(d+))(?>(e+))/
+     \Maabbccddee
+
+/^(?>(a+))(?>b+)(?>(c+))(?>d+)(?>(e+))/
+     \Maabbccddee
+
+/^a\x41z/
+    aAz
+    *** Failers
+    ax41z
+
+/^a[m\x41]z/
+    aAz
+
+/^a\x1z/
+    ax1z
+
+/^a\u0041z/
+    aAz
+    *** Failers
+    au0041z
+
+/^a[m\u0041]z/
+    aAz
+
+/^a\u041z/
+    au041z
+    *** Failers
+    aAz
+
+/^a\U0041z/
+    aU0041z
+    *** Failers
+    aAz
+
+/(?(?=c)c|d)++Y/BZ
+
+/(?(?=c)c|d)*+Y/BZ
+
+/a[\NB]c/
+    aNc
+    
+/a[B-\Nc]/ 
+
+/(a)(?2){0,1999}?(b)/
+
+/(a)(?(DEFINE)(b))(?2){0,1999}?(?2)/
+
+/--- This test, with something more complicated than individual letters, causes
+different behaviour in Perl. Perhaps it disables some optimization; no tag is
+passed back for the failures, whereas in PCRE there is a tag. ---/
+    
+/(A|P)(*:A)(B|P) | (X|P)(X|P)(*:B)(Y|P)/xK
+    AABC
+    XXYZ 
+    ** Failers
+    XAQQ  
+    XAQQXZZ  
+    AXQQQ 
+    AXXQQQ 
+
+/-- Perl doesn't give marks for these, though it does if the alternatives are
+replaced by single letters. --/
+    
+/(b|q)(*:m)f|a(*:n)w/K
+    aw 
+    ** Failers 
+    abc
+
+/(q|b)(*:m)f|a(*:n)w/K
+    aw 
+    ** Failers 
+    abc
+
+/-- After a partial match, the behaviour is as for a failure. --/
+
+/^a(*:X)bcde/K
+   abc\P
+   
+/-- These are here because Perl doesn't return a mark, except for the first --/
+
+/(?=(*:x))(q|)/K+
+    abc
+
+/(?=(*:x))((*:y)q|)/K+
+    abc
+
+/(?=(*:x))(?:(*:y)q|)/K+
+    abc
+
+/(?=(*:x))(?>(*:y)q|)/K+
+    abc
+
+/(?=a(*:x))(?!a(*:y)c)/K+
+    ab
+
+/(?=a(*:x))(?=a(*:y)c|)/K+
+    ab
+
+/(..)\1/
+    ab\P
+    aba\P
+    abab\P
+
+/(..)\1/i
+    ab\P
+    abA\P
+    aBAb\P
+
+/(..)\1{2,}/
+    ab\P
+    aba\P
+    abab\P
+    ababa\P
+    ababab\P
+    ababab\P\P
+    abababa\P
+    abababa\P\P
+
+/(..)\1{2,}/i
+    ab\P
+    aBa\P
+    aBAb\P
+    AbaBA\P
+    abABAb\P
+    aBAbaB\P\P
+    abABabA\P
+    abaBABa\P\P
+
+/(..)\1{2,}?x/i
+    ab\P
+    abA\P
+    aBAb\P
+    abaBA\P
+    abAbaB\P
+    abaBabA\P
+    abAbABaBx\P
+
+/^(..)\1/
+    aba\P
+
+/^(..)\1{2,3}x/
+    aba\P
+    ababa\P
+    ababa\P\P
+    abababx
+    ababababx  
+
+/^(..)\1{2,3}?x/
+    aba\P
+    ababa\P
+    ababa\P\P
+    abababx
+    ababababx  
+    
+/^(..)(\1{2,3})ab/
+    abababab
+
+/^\R/
+    \r\P
+    \r\P\P
+    
+/^\R{2,3}x/
+    \r\P
+    \r\P\P
+    \r\r\P 
+    \r\r\P\P
+    \r\r\r\P  
+    \r\r\r\P\P
+    \r\rx
+    \r\r\rx    
+
+/^\R{2,3}?x/
+    \r\P
+    \r\P\P
+    \r\r\P 
+    \r\r\P\P
+    \r\r\r\P  
+    \r\r\r\P\P
+    \r\rx
+    \r\r\rx    
+    
+/^\R?x/
+    \r\P
+    \r\P\P 
+    x
+    \rx  
+
+/^\R+x/
+    \r\P
+    \r\P\P 
+    \r\n\P
+    \r\n\P\P  
+    \rx  
+
+/^a$/
+    a\r\P
+    a\r\P\P 
+
+/^a$/m
+    a\r\P
+    a\r\P\P 
+
+/^(a$|a\r)/
+    a\r\P
+    a\r\P\P 
+
+/^(a$|a\r)/m
+    a\r\P
+    a\r\P\P 
+
+/./
+    \r\P
+    \r\P\P 
+  
+/.{2,3}/
+    \r\P 
+    \r\P\P
+    \r\r\P
+    \r\r\P\P
+    \r\r\r\P
+    \r\r\r\P\P     
+
+/.{2,3}?/
+    \r\P 
+    \r\P\P
+    \r\r\P
+    \r\r\P\P
+    \r\r\r\P
+    \r\r\r\P\P     
+
+"AB(C(D))(E(F))?(?(?=\2)(?=\4))"
+    ABCDGHI\O03
+    
+/-- These are all run as real matches in test 1; here we are just checking the
+settings of the anchored and startline bits. --/ 
+
+/(?>.*?a)(?<=ba)/I
+
+/(?:.*?a)(?<=ba)/I
+
+/.*?a(*PRUNE)b/I
+
+/.*?a(*PRUNE)b/sI
+
+/^a(*PRUNE)b/sI
+
+/.*?a(*SKIP)b/I
+
+/(?>.*?a)b/sI
+
+/(?>.*?a)b/I
+
+/(?>^a)b/sI
+
+/(?>.*?)(?<=(abcd)|(wxyz))/I
+
+/(?>.*)(?<=(abcd)|(wxyz))/I
+
+"(?>.*)foo"I
+
+"(?>.*?)foo"I
+
+/(?>^abc)/mI
+
+/(?>.*abc)/mI
+
+/(?:.*abc)/mI
+
+/-- Check PCRE_STUDY_EXTRA_NEEDED --/
+
+/.?/S-I
+
+/.?/S!I
+
+/(?:(a)+(?C1)bb|aa(?C2)b)/
+    aab\C+
+   
+/(?:(a)++(?C1)bb|aa(?C2)b)/
+    aab\C+ 
+    
+/(?:(?>(a))(?C1)bb|aa(?C2)b)/
+    aab\C+ 
+
+/(?:(?1)(?C1)x|ab(?C2))((a)){0}/                                                
+    aab\C+ 
+
+/(?1)(?C1)((a)(?C2)){0}/   
+    aab\C+ 
+
+/(?:(a)+(?C1)bb|aa(?C2)b)++/
+    aab\C+
+    aab\C+\O2
+
+/(ab)x|ab/
+    ab\O3
+    ab\O2 
+  
+/(ab)/
+    ab\O3
+    ab\O2 
+    
+/(?<=123)(*MARK:xx)abc/K
+    xxxx123a\P\P
+    xxxx123a\P
+    
+/123\Kabc/
+    xxxx123a\P\P
+    xxxx123a\P
+
+/^(?(?=a)aa|bb)/C
+    bb
+
+/(?C1)^(?C2)(?(?C99)(?=(?C3)a(?C4))(?C5)a(?C6)a(?C7)|(?C8)b(?C9)b(?C10))(?C11)/
+    bb
+
+/-- Perl seems to have a bug with this one --/
+
+/aaaaa(*COMMIT)(*PRUNE)b|a+c/
+    aaaaaac
+    
+/-- Here are some that Perl treats differently because of the way it handles
+backtracking verbs. --/
+
+ /(?!a(*COMMIT)b)ac|ad/
+     ac
+     ad 
+
+/^(?!a(*THEN)b|ac)../
+     ac
+     ad 
+
+/^(?=a(*THEN)b|ac)/
+    ac
+    
+/\A.*?(?:a|b(*THEN)c)/
+    ba
+
+/\A.*?(?:a|b(*THEN)c)++/
+    ba
+
+/\A.*?(?:a|b(*THEN)c|d)/
+    ba
+
+/(?:(a(*MARK:X)a+(*SKIP:X)b)){0}(?:(?1)|aac)/
+    aac 
+
+/\A.*?(a|b(*THEN)c)/
+    ba
+
+/^(A(*THEN)B|A(*THEN)D)/
+    AD           
+    
+/(?!b(*THEN)a)bn|bnn/
+    bnn
+
+/(?(?=b(*SKIP)a)bn|bnn)/
+    bnn
+
+/(?=b(*THEN)a|)bn|bnn/
+    bnn
+
+/-------------------------/ 
+
+/(*LIMIT_MATCH=12bc)abc/
+
+/(*LIMIT_MATCH=4294967290)abc/
+
+/(*LIMIT_RECURSION=4294967280)abc/I
+
+/(a+)*zz/
+    aaaaaaaaaaaaaz
+    aaaaaaaaaaaaaz\q3000
+
+/(a+)*zz/S-
+    aaaaaaaaaaaaaz\Q10 
+
+/(*LIMIT_MATCH=3000)(a+)*zz/I
+    aaaaaaaaaaaaaz
+    aaaaaaaaaaaaaz\q60000
+
+/(*LIMIT_MATCH=60000)(*LIMIT_MATCH=3000)(a+)*zz/I
+    aaaaaaaaaaaaaz
+
+/(*LIMIT_MATCH=60000)(a+)*zz/I
+    aaaaaaaaaaaaaz
+    aaaaaaaaaaaaaz\q3000
+
+/(*LIMIT_RECURSION=10)(a+)*zz/IS-
+    aaaaaaaaaaaaaz
+    aaaaaaaaaaaaaz\Q1000
+
+/(*LIMIT_RECURSION=10)(*LIMIT_RECURSION=1000)(a+)*zz/IS-
+    aaaaaaaaaaaaaz
+
+/(*LIMIT_RECURSION=1000)(a+)*zz/IS-
+    aaaaaaaaaaaaaz
+    aaaaaaaaaaaaaz\Q10
+
+/-- End of testinput2 --/
diff --git a/pcre/testdata/testinput20 b/pcre/testdata/testinput20
new file mode 100644
index 00000000000..2a6b8f23f4c
--- /dev/null
+++ b/pcre/testdata/testinput20
@@ -0,0 +1,19 @@
+/-- These DFA tests are for the handling of characters greater than 255 in
+    16- or 32-bit, non-UTF mode. --/
+
+/^\x{ffff}+/i
+    \x{ffff}
+
+/^\x{ffff}?/i
+    \x{ffff}
+
+/^\x{ffff}*/i
+    \x{ffff}
+
+/^\x{ffff}{3}/i
+    \x{ffff}\x{ffff}\x{ffff}
+
+/^\x{ffff}{0,3}/i
+    \x{ffff}
+
+/-- End of testinput20 --/
diff --git a/pcre/testdata/testinput21 b/pcre/testdata/testinput21
new file mode 100644
index 00000000000..b285d637526
--- /dev/null
+++ b/pcre/testdata/testinput21
@@ -0,0 +1,22 @@
+/-- Tests for reloading pre-compiled patterns. The first one gives an error
+right away, and can be any old pattern compiled in 8-bit mode ("abc" is
+typical). The others require the link size to be 2. */x
+
+(?:[AaLl]+)[^xX-]*?)(?P[\x{150}-\x{250}\x{300}]|[^\x{800}aAs-uS-U\x{d800}-\x{dfff}])++[^#\b\x{500}\x{1000}]{3,5}$
+    In 16-bit mode with options:  S>testdata/saved16LE-1
+                                 FS>testdata/saved16BE-1
+    In 32-bit mode with options:  S>testdata/saved32LE-1
+                                 FS>testdata/saved32BE-1
+--%x
+
+[aZ\x{400}-\x{10ffff}]{4,}[\x{f123}\x{10039}\x{20000}-\x{21234}]?|[A-Cx-z\x{100000}-\x{1000a7}\x{101234}])(?[^az])
+    In 16-bit mode with options:  S8>testdata/saved16LE-1
+                                 FS8>testdata/saved16BE-1
+    In 32-bit mode with options:  S8>testdata/saved32LE-1
+                                 FS8testdata/saved32BE-1
+--%8x
+
+>>\xaa<<<
+    >>>\xba<<< 
+
+/[\W]+/Lfr_FR
+    >>>\xaa<<<
+    >>>\xba<<< 
+
+/[^[:alpha:]]+/Lfr_FR
+    >>>\xaa<<<
+    >>>\xba<<< 
+
+/\w+/Lfr_FR
+    >>>\xaa<<<
+    >>>\xba<<< 
+
+/[\w]+/Lfr_FR
+    >>>\xaa<<<
+    >>>\xba<<< 
+
+/[[:alpha:]]+/Lfr_FR
+    >>>\xaa<<<
+    >>>\xba<<< 
+    
+/[[:alpha:]][[:lower:]][[:upper:]]/DZLfr_FR 
+
+/-- End of testinput3 --/
diff --git a/pcre/testdata/testinput4 b/pcre/testdata/testinput4
new file mode 100644
index 00000000000..b9f1a72e4f2
--- /dev/null
+++ b/pcre/testdata/testinput4
@@ -0,0 +1,714 @@
+/-- This set of tests is for UTF support, excluding Unicode properties. It is
+    compatible with all versions of Perl >= 5.10 and both the 8-bit and 16-bit
+    PCRE libraries. --/
+   
+/a.b/8
+    acb
+    a\x7fb
+    a\x{100}b 
+    *** Failers
+    a\nb  
+
+/a(.{3})b/8
+    a\x{4000}xyb 
+    a\x{4000}\x7fyb 
+    a\x{4000}\x{100}yb 
+    *** Failers
+    a\x{4000}b 
+    ac\ncb 
+
+/a(.*?)(.)/
+    a\xc0\x88b
+
+/a(.*?)(.)/8
+    a\x{100}b
+
+/a(.*)(.)/
+    a\xc0\x88b
+
+/a(.*)(.)/8
+    a\x{100}b
+
+/a(.)(.)/
+    a\xc0\x92bcd
+
+/a(.)(.)/8
+    a\x{240}bcd
+
+/a(.?)(.)/
+    a\xc0\x92bcd
+
+/a(.?)(.)/8
+    a\x{240}bcd
+
+/a(.??)(.)/
+    a\xc0\x92bcd
+
+/a(.??)(.)/8
+    a\x{240}bcd
+
+/a(.{3})b/8
+    a\x{1234}xyb 
+    a\x{1234}\x{4321}yb 
+    a\x{1234}\x{4321}\x{3412}b 
+    *** Failers
+    a\x{1234}b 
+    ac\ncb 
+
+/a(.{3,})b/8
+    a\x{1234}xyb 
+    a\x{1234}\x{4321}yb 
+    a\x{1234}\x{4321}\x{3412}b 
+    axxxxbcdefghijb 
+    a\x{1234}\x{4321}\x{3412}\x{3421}b 
+    *** Failers
+    a\x{1234}b 
+
+/a(.{3,}?)b/8
+    a\x{1234}xyb 
+    a\x{1234}\x{4321}yb 
+    a\x{1234}\x{4321}\x{3412}b 
+    axxxxbcdefghijb 
+    a\x{1234}\x{4321}\x{3412}\x{3421}b 
+    *** Failers
+    a\x{1234}b 
+
+/a(.{3,5})b/8
+    a\x{1234}xyb 
+    a\x{1234}\x{4321}yb 
+    a\x{1234}\x{4321}\x{3412}b 
+    axxxxbcdefghijb 
+    a\x{1234}\x{4321}\x{3412}\x{3421}b 
+    axbxxbcdefghijb 
+    axxxxxbcdefghijb 
+    *** Failers
+    a\x{1234}b 
+    axxxxxxbcdefghijb 
+
+/a(.{3,5}?)b/8
+    a\x{1234}xyb 
+    a\x{1234}\x{4321}yb 
+    a\x{1234}\x{4321}\x{3412}b 
+    axxxxbcdefghijb 
+    a\x{1234}\x{4321}\x{3412}\x{3421}b 
+    axbxxbcdefghijb 
+    axxxxxbcdefghijb 
+    *** Failers
+    a\x{1234}b 
+    axxxxxxbcdefghijb 
+
+/^[a\x{c0}]/8
+    *** Failers
+    \x{100}
+
+/(?<=aXb)cd/8
+    aXbcd
+
+/(?<=a\x{100}b)cd/8
+    a\x{100}bcd
+
+/(?<=a\x{100000}b)cd/8
+    a\x{100000}bcd
+    
+/(?:\x{100}){3}b/8
+    \x{100}\x{100}\x{100}b
+    *** Failers 
+    \x{100}\x{100}b
+
+/\x{ab}/8
+    \x{ab} 
+    \xc2\xab
+    *** Failers 
+    \x00{ab}
+
+/(?<=(.))X/8
+    WXYZ
+    \x{256}XYZ 
+    *** Failers
+    XYZ 
+
+/[^a]+/8g
+    bcd
+    \x{100}aY\x{256}Z 
+    
+/^[^a]{2}/8
+    \x{100}bc
+ 
+/^[^a]{2,}/8
+    \x{100}bcAa
+
+/^[^a]{2,}?/8
+    \x{100}bca
+
+/[^a]+/8ig
+    bcd
+    \x{100}aY\x{256}Z 
+    
+/^[^a]{2}/8i
+    \x{100}bc
+ 
+/^[^a]{2,}/8i
+    \x{100}bcAa
+
+/^[^a]{2,}?/8i
+    \x{100}bca
+
+/\x{100}{0,0}/8
+    abcd
+ 
+/\x{100}?/8
+    abcd
+    \x{100}\x{100} 
+
+/\x{100}{0,3}/8 
+    \x{100}\x{100} 
+    \x{100}\x{100}\x{100}\x{100} 
+    
+/\x{100}*/8
+    abce
+    \x{100}\x{100}\x{100}\x{100} 
+
+/\x{100}{1,1}/8
+    abcd\x{100}\x{100}\x{100}\x{100} 
+
+/\x{100}{1,3}/8
+    abcd\x{100}\x{100}\x{100}\x{100} 
+
+/\x{100}+/8
+    abcd\x{100}\x{100}\x{100}\x{100} 
+
+/\x{100}{3}/8
+    abcd\x{100}\x{100}\x{100}XX
+
+/\x{100}{3,5}/8
+    abcd\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}XX
+
+/\x{100}{3,}/8
+    abcd\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}XX
+
+/(?<=a\x{100}{2}b)X/8+
+    Xyyya\x{100}\x{100}bXzzz
+
+/\D*/8
+  aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+
+/\D*/8
+  \x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}
+
+/\D/8
+    1X2
+    1\x{100}2 
+  
+/>\S/8
+    > >X Y
+    > >\x{100} Y
+  
+/\d/8
+    \x{100}3
+    
+/\s/8
+    \x{100} X
+    
+/\D+/8
+    12abcd34
+    *** Failers
+    1234  
+
+/\D{2,3}/8
+    12abcd34
+    12ab34
+    *** Failers  
+    1234
+    12a34  
+
+/\D{2,3}?/8
+    12abcd34
+    12ab34
+    *** Failers  
+    1234
+    12a34  
+
+/\d+/8
+    12abcd34
+    *** Failers
+
+/\d{2,3}/8
+    12abcd34
+    1234abcd
+    *** Failers  
+    1.4 
+
+/\d{2,3}?/8
+    12abcd34
+    1234abcd
+    *** Failers  
+    1.4 
+
+/\S+/8
+    12abcd34
+    *** Failers
+    \    \ 
+
+/\S{2,3}/8
+    12abcd34
+    1234abcd
+    *** Failers
+    \     \  
+
+/\S{2,3}?/8
+    12abcd34
+    1234abcd
+    *** Failers
+    \     \  
+
+/>\s+      <34
+    *** Failers
+
+/>\s{2,3}         \s{2,3}?         \xff<
+
+/[\xff]/8
+    >\x{ff}<
+
+/[^\xFF]/
+    XYZ
+
+/[^\xff]/8
+    XYZ
+    \x{123} 
+
+/^[ac]*b/8
+  xb
+
+/^[ac\x{100}]*b/8
+  xb
+
+/^[^x]*b/8i
+  xb
+
+/^[^x]*b/8
+  xb
+  
+/^\d*b/8
+  xb 
+
+/(|a)/g8
+    catac
+    a\x{256}a 
+
+/^\x{85}$/8i
+    \x{85}
+
+/^ሴ/8
+    ሴ 
+
+/^\ሴ/8
+    ሴ 
+
+"(?s)(.{1,5})"8
+    abcdefg
+    ab
+
+/a*\x{100}*\w/8
+    a 
+
+/\S\S/8g
+    A\x{a3}BC
+    
+/\S{2}/8g
+    A\x{a3}BC
+    
+/\W\W/8g
+    +\x{a3}== 
+
+/\W{2}/8g
+    +\x{a3}== 
+
+/\S/8g
+    \x{442}\x{435}\x{441}\x{442}
+
+/[\S]/8g
+    \x{442}\x{435}\x{441}\x{442}
+
+/\D/8g
+    \x{442}\x{435}\x{441}\x{442}
+
+/[\D]/8g
+    \x{442}\x{435}\x{441}\x{442}
+
+/\W/8g
+    \x{2442}\x{2435}\x{2441}\x{2442}
+
+/[\W]/8g
+    \x{2442}\x{2435}\x{2441}\x{2442}
+    
+/[\S\s]*/8
+    abc\n\r\x{442}\x{435}\x{441}\x{442}xyz 
+
+/[\x{41f}\S]/8g
+    \x{442}\x{435}\x{441}\x{442}
+
+/.[^\S]./8g
+    abc def\x{442}\x{443}xyz\npqr
+
+/.[^\S\n]./8g
+    abc def\x{442}\x{443}xyz\npqr
+
+/[[:^alnum:]]/8g  
+    +\x{2442}
+    
+/[[:^alpha:]]/8g 
+    +\x{2442}
+    
+/[[:^ascii:]]/8g 
+    A\x{442}
+    
+/[[:^blank:]]/8g 
+    A\x{442}
+    
+/[[:^cntrl:]]/8g 
+    A\x{442}
+    
+/[[:^digit:]]/8g 
+    A\x{442}
+    
+/[[:^graph:]]/8g 
+    \x19\x{e01ff}
+    
+/[[:^lower:]]/8g 
+    A\x{422}
+    
+/[[:^print:]]/8g 
+    \x{19}\x{e01ff}
+    
+/[[:^punct:]]/8g 
+    A\x{442}
+    
+/[[:^space:]]/8g 
+    A\x{442}
+    
+/[[:^upper:]]/8g 
+    a\x{442}
+    
+/[[:^word:]]/8g  
+    +\x{2442}
+    
+/[[:^xdigit:]]/8g
+    M\x{442}
+
+/[^ABCDEFGHIJKLMNOPQRSTUVWXYZÀÁÂÃÄÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖØÙÚÛÜÝÞĀĂĄĆĈĊČĎĐĒĔĖĘĚĜĞĠĢĤĦĨĪĬĮİIJĴĶĹĻĽĿŁŃŅŇŊŌŎŐŒŔŖŘŚŜŞŠŢŤŦŨŪŬŮŰŲŴŶŸŹŻŽƁƂƄƆƇƉƊƋƎƏƐƑƓƔƖƗƘƜƝƟƠƢƤƦƧƩƬƮƯƱƲƳƵƷƸƼDŽLJNJǍǏǑǓǕǗǙǛǞǠǢǤǦǨǪǬǮDZǴǶǷǸǺǼǾȀȂȄȆȈȊȌȎȐȒȔȖȘȚȜȞȠȢȤȦȨȪȬȮȰȲȺȻȽȾɁΆΈΉΊΌΎΏΑΒΓΔΕΖΗΘΙΚΛΜΝΞΟΠΡΣΤΥΦΧΨΩΪΫϒϓϔϘϚϜϞϠϢϤϦϨϪϬϮϴϷϹϺϽϾϿЀЁЂЃЄЅІЇЈЉЊЋЌЍЎЏАБВГДЕЖЗИЙКЛМНОПРСТУФХЦЧШЩЪЫЬЭЮЯѠѢѤѦѨѪѬѮѰѲѴѶѸѺѼѾҀҊҌҎҐҒҔҖҘҚҜҞҠҢҤҦҨҪҬҮҰҲҴҶҸҺҼҾӀӁӃӅӇӉӋӍӐӒӔӖӘӚӜӞӠӢӤӦӨӪӬӮӰӲӴӶӸԀԂԄԆԈԊԌԎԱԲԳԴԵԶԷԸԹԺԻԼԽԾԿՀՁՂՃՄՅՆՇՈՉՊՋՌՍՎՏՐՑՒՓՔՕՖႠႡႢႣႤႥႦႧႨႩႪႫႬႭႮႯႰႱႲႳႴႵႶႷႸႹႺႻႼႽႾႿჀჁჂჃჄჅḀḂḄḆḈḊḌḎḐḒḔḖḘḚḜḞḠḢḤḦḨḪḬḮḰḲḴḶḸḺḼḾṀṂṄṆṈṊṌṎṐṒṔṖṘṚṜṞṠṢṤṦṨṪṬṮṰṲṴṶṸṺṼṾẀẂẄẆẈẊẌẎẐẒẔẠẢẤẦẨẪẬẮẰẲẴẶẸẺẼẾỀỂỄỆỈỊỌỎỐỒỔỖỘỚỜỞỠỢỤỦỨỪỬỮỰỲỴỶỸἈἉἊἋἌἍἎἏἘἙἚἛἜἝἨἩἪἫἬἭἮἯἸἹἺἻἼἽἾἿὈὉὊὋὌὍὙὛὝὟὨὩὪὫὬὭὮὯᾸᾹᾺΆῈΈῊΉῘῙῚΊῨῩῪΎῬῸΌῺΏabcdefghijklmnopqrstuvwxyzªµºßàáâãäåæçèéêëìíîïðñòóôõöøùúûüýþÿāăąćĉċčďđēĕėęěĝğġģĥħĩīĭįıijĵķĸĺļľŀłńņňʼnŋōŏőœŕŗřśŝşšţťŧũūŭůűųŵŷźżžſƀƃƅƈƌƍƒƕƙƚƛƞơƣƥƨƪƫƭưƴƶƹƺƽƾƿdžljnjǎǐǒǔǖǘǚǜǝǟǡǣǥǧǩǫǭǯǰdzǵǹǻǽǿȁȃȅȇȉȋȍȏȑȓȕȗșțȝȟȡȣȥȧȩȫȭȯȱȳȴȵȶȷȸȹȼȿɀɐɑɒɓɔɕɖɗɘəɚɛɜɝɞɟɠɡɢɣɤɥɦɧɨɩɪɫɬɭɮɯɰɱɲɳɴɵɶɷɸɹɺɻɼɽɾɿʀʁʂʃʄʅʆʇʈʉʊʋʌʍʎʏʐʑʒʓʔʕʖʗʘʙʚʛʜʝʞʟʠʡʢʣʤʥʦʧʨʩʪʫʬʭʮʯΐάέήίΰαβγδεζηθικλμνξοπρςστυφχψωϊϋόύώϐϑϕϖϗϙϛϝϟϡϣϥϧϩϫϭϯϰϱϲϳϵϸϻϼабвгдежзийклмнопрстуфхцчшщъыьэюяѐёђѓєѕіїјљњћќѝўџѡѣѥѧѩѫѭѯѱѳѵѷѹѻѽѿҁҋҍҏґғҕҗҙқҝҟҡңҥҧҩҫҭүұҳҵҷҹһҽҿӂӄӆӈӊӌӎӑӓӕӗәӛӝӟӡӣӥӧөӫӭӯӱӳӵӷӹԁԃԅԇԉԋԍԏաբգդեզէըթժիլխծկհձղճմյնշոչպջռսվտրցւփքօֆևᴀᴁᴂᴃᴄᴅᴆᴇᴈᴉᴊᴋᴌᴍᴎᴏᴐᴑᴒᴓᴔᴕᴖᴗᴘᴙᴚᴛᴜᴝᴞᴟᴠᴡᴢᴣᴤᴥᴦᴧᴨᴩᴪᴫᵢᵣᵤᵥᵦᵧᵨᵩᵪᵫᵬᵭᵮᵯᵰᵱᵲᵳᵴᵵᵶᵷᵹᵺᵻᵼᵽᵾᵿᶀᶁᶂᶃᶄᶅᶆᶇᶈᶉᶊᶋᶌᶍᶎᶏᶐᶑᶒᶓᶔᶕᶖᶗᶘᶙᶚḁḃḅḇḉḋḍḏḑḓḕḗḙḛḝḟḡḣḥḧḩḫḭḯḱḳḵḷḹḻḽḿṁṃṅṇṉṋṍṏṑṓṕṗṙṛṝṟṡṣṥṧṩṫṭṯṱṳṵṷṹṻṽṿẁẃẅẇẉẋẍẏẑẓẕẖẗẘẙẚẛạảấầẩẫậắằẳẵặẹẻẽếềểễệỉịọỏốồổỗộớờởỡợụủứừửữựỳỵỷỹἀἁἂἃἄἅἆἇἐἑἒἓἔἕἠἡἢἣἤἥἦἧἰἱἲἳἴἵἶἷὀὁὂὃὄὅὐὑὒὓὔὕὖὗὠὡὢὣὤὥὦὧὰάὲέὴήὶίὸόὺύὼώᾀᾁᾂᾃᾄᾅᾆᾇᾐᾑᾒᾓᾔᾕᾖᾗᾠᾡᾢᾣᾤᾥᾦᾧᾰᾱᾲᾳᾴᾶᾷιῂῃῄῆῇῐῑῒΐῖῗῠῡῢΰῤῥῦῧῲῳῴῶῷⲁⲃⲅⲇⲉⲋⲍⲏⲑⲓⲕⲗⲙⲛⲝⲟⲡⲣⲥⲧⲩⲫⲭⲯⲱⲳⲵⲷⲹⲻⲽⲿⳁⳃⳅⳇⳉⳋⳍⳏⳑⳓⳕⳗⳙⳛⳝⳟⳡⳣⳤⴀⴁⴂⴃⴄⴅⴆⴇⴈⴉⴊⴋⴌⴍⴎⴏⴐⴑⴒⴓⴔⴕⴖⴗⴘⴙⴚⴛⴜⴝⴞⴟⴠⴡⴢⴣⴤⴥfffiflffifflſtstﬓﬔﬕﬖﬗ\d-_^]/8
+
+/^[^d]*?$/
+    abc
+
+/^[^d]*?$/8
+    abc
+
+/^[^d]*?$/i
+    abc
+
+/^[^d]*?$/8i
+    abc
+
+/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8
+
+/^[a\x{c0}]b/8
+    \x{c0}b
+    
+/^([a\x{c0}]*?)aa/8
+    a\x{c0}aaaa/ 
+
+/^([a\x{c0}]*?)aa/8
+    a\x{c0}aaaa/ 
+    a\x{c0}a\x{c0}aaa/ 
+
+/^([a\x{c0}]*)aa/8
+    a\x{c0}aaaa/ 
+    a\x{c0}a\x{c0}aaa/ 
+
+/^([a\x{c0}]*)a\x{c0}/8
+    a\x{c0}aaaa/ 
+    a\x{c0}a\x{c0}aaa/ 
+
+/A*/g8
+    AAB\x{123}BAA
+
+/(abc)\1/8i
+   abc
+
+/(abc)\1/8
+   abc
+
+/a(*:a\x{1234}b)/8K
+    abc
+
+/a(*:a£b)/8K 
+    abc
+
+/-- Noncharacters --/
+
+/./8
+    \x{fffe}
+    \x{ffff}
+    \x{1fffe}
+    \x{1ffff}
+    \x{2fffe}
+    \x{2ffff}
+    \x{3fffe}
+    \x{3ffff}
+    \x{4fffe}
+    \x{4ffff}
+    \x{5fffe}
+    \x{5ffff}
+    \x{6fffe}
+    \x{6ffff}
+    \x{7fffe}
+    \x{7ffff}
+    \x{8fffe}
+    \x{8ffff}
+    \x{9fffe}
+    \x{9ffff}
+    \x{afffe}
+    \x{affff}
+    \x{bfffe}
+    \x{bffff}
+    \x{cfffe}
+    \x{cffff}
+    \x{dfffe}
+    \x{dffff}
+    \x{efffe}
+    \x{effff}
+    \x{ffffe}
+    \x{fffff}
+    \x{10fffe}
+    \x{10ffff}
+    \x{fdd0}
+    \x{fdd1}
+    \x{fdd2}
+    \x{fdd3}
+    \x{fdd4}
+    \x{fdd5}
+    \x{fdd6}
+    \x{fdd7}
+    \x{fdd8}
+    \x{fdd9}
+    \x{fdda}
+    \x{fddb}
+    \x{fddc}
+    \x{fddd}
+    \x{fdde}
+    \x{fddf}
+    \x{fde0}
+    \x{fde1}
+    \x{fde2}
+    \x{fde3}
+    \x{fde4}
+    \x{fde5}
+    \x{fde6}
+    \x{fde7}
+    \x{fde8}
+    \x{fde9}
+    \x{fdea}
+    \x{fdeb}
+    \x{fdec}
+    \x{fded}
+    \x{fdee}
+    \x{fdef}
+
+/^\d*\w{4}/8
+    1234
+    123 
+    
+/^[^b]*\w{4}/8
+    aaaa
+    aaa  
+ 
+/^[^b]*\w{4}/8i
+    aaaa
+    aaa  
+ 
+/^\x{100}*.{4}/8
+    \x{100}\x{100}\x{100}\x{100}
+    \x{100}\x{100}\x{100}
+
+/^\x{100}*.{4}/8i
+    \x{100}\x{100}\x{100}\x{100}
+    \x{100}\x{100}\x{100}
+
+/-- End of testinput4 --/
diff --git a/pcre/testdata/testinput5 b/pcre/testdata/testinput5
new file mode 100644
index 00000000000..642749c315c
--- /dev/null
+++ b/pcre/testdata/testinput5
@@ -0,0 +1,772 @@
+/-- This set of tests checks the API, internals, and non-Perl stuff for UTF
+    support, excluding Unicode properties. However, tests that give different
+    results in 8-bit and 16-bit modes are excluded (see tests 16 and 17). --/
+
+/\x{110000}/8DZ
+
+/\x{ffffffff}/8
+
+/\x{100000000}/8
+
+/\x{d800}/8
+
+/\x{dfff}/8
+
+/\x{d7ff}/8
+
+/\x{e000}/8
+
+/^\x{100}a\x{1234}/8
+    \x{100}a\x{1234}bcd
+
+/\x{0041}\x{2262}\x{0391}\x{002e}/DZ8
+    \x{0041}\x{2262}\x{0391}\x{002e}
+    
+/.{3,5}X/DZ8
+    \x{212ab}\x{212ab}\x{212ab}\x{861}X
+
+/.{3,5}?/DZ8
+    \x{212ab}\x{212ab}\x{212ab}\x{861}
+
+/(?<=\C)X/8
+    Should produce an error diagnostic
+    
+/^[ab]/8DZ
+    bar
+    *** Failers
+    c
+    \x{ff}
+    \x{100}  
+
+/^[^ab]/8DZ
+    c
+    \x{ff}
+    \x{100}  
+    *** Failers 
+    aaa
+  
+/\x{100}*(\d+|"(?1)")/8
+    1234
+    "1234" 
+    \x{100}1234
+    "\x{100}1234"  
+    \x{100}\x{100}12ab 
+    \x{100}\x{100}"12" 
+    *** Failers 
+    \x{100}\x{100}abcd
+
+/\x{100}*/8DZ
+
+/a\x{100}*/8DZ
+
+/ab\x{100}*/8DZ
+
+/\x{100}*A/8DZ
+    A
+
+/\x{100}*\d(?R)/8DZ
+
+/[Z\x{100}]/8DZ
+    Z\x{100}
+    \x{100}
+    \x{100}Z
+    *** Failers 
+
+/[\x{200}-\x{100}]/8
+
+/[Ā-Ą]/8
+    \x{100}
+    \x{104}
+    *** Failers
+    \x{105}
+    \x{ff}    
+
+/[z-\x{100}]/8DZ
+
+/[z\Qa-d]Ā\E]/8DZ
+    \x{100}
+    Ā 
+
+/[\xFF]/DZ
+    >\xff<
+
+/[^\xFF]/DZ
+
+/[Ä-Ü]/8
+    Ö # Matches without Study
+    \x{d6}
+    
+/[Ä-Ü]/8S
+    Ö <-- Same with Study
+    \x{d6}
+    
+/[\x{c4}-\x{dc}]/8 
+    Ö # Matches without Study
+    \x{d6} 
+
+/[\x{c4}-\x{dc}]/8S
+    Ö <-- Same with Study
+    \x{d6} 
+
+/[^\x{100}]abc(xyz(?1))/8DZ
+
+/[ab\x{100}]abc(xyz(?1))/8DZ
+
+/(\x{100}(b(?2)c))?/DZ8
+
+/(\x{100}(b(?2)c)){0,2}/DZ8
+
+/(\x{100}(b(?1)c))?/DZ8
+
+/(\x{100}(b(?1)c)){0,2}/DZ8
+
+/\W/8
+    A.B
+    A\x{100}B 
+  
+/\w/8
+    \x{100}X   
+
+/^\ሴ/8DZ
+
+/\x{100}*\d/8DZ
+
+/\x{100}*\s/8DZ
+
+/\x{100}*\w/8DZ
+
+/\x{100}*\D/8DZ
+
+/\x{100}*\S/8DZ
+
+/\x{100}*\W/8DZ
+
+/()()()()()()()()()()
+ ()()()()()()()()()()
+ ()()()()()()()()()()
+ ()()()()()()()()()()
+ A (x) (?41) B/8x
+    AxxB     
+
+/^[\x{100}\E-\Q\E\x{150}]/BZ8
+
+/^[\QĀ\E-\QŐ\E]/BZ8
+
+/^abc./mgx8
+    abc1 \x0aabc2 \x0babc3xx \x0cabc4 \x0dabc5xx \x0d\x0aabc6 \x{0085}abc7 \x{2028}abc8 \x{2029}abc9 JUNK
+
+/abc.$/mgx8
+    abc1\x0a abc2\x0b abc3\x0c abc4\x0d abc5\x0d\x0a abc6\x{0085} abc7\x{2028} abc8\x{2029} abc9
+
+/^a\Rb/8
+    a\nb
+    a\rb
+    a\r\nb
+    a\x0bb
+    a\x0cb
+    a\x{85}b   
+    a\x{2028}b 
+    a\x{2029}b 
+    ** Failers
+    a\n\rb    
+
+/^a\R*b/8
+    ab
+    a\nb
+    a\rb
+    a\r\nb
+    a\x0bb
+    a\x0c\x{2028}\x{2029}b
+    a\x{85}b   
+    a\n\rb    
+    a\n\r\x{85}\x0cb 
+
+/^a\R+b/8
+    a\nb
+    a\rb
+    a\r\nb
+    a\x0bb
+    a\x0c\x{2028}\x{2029}b
+    a\x{85}b   
+    a\n\rb    
+    a\n\r\x{85}\x0cb 
+    ** Failers
+    ab  
+
+/^a\R{1,3}b/8
+    a\nb
+    a\n\rb
+    a\n\r\x{85}b
+    a\r\n\r\nb 
+    a\r\n\r\n\r\nb 
+    a\n\r\n\rb
+    a\n\n\r\nb 
+    ** Failers
+    a\n\n\n\rb
+    a\r
+
+/\H\h\V\v/8
+    X X\x0a
+    X\x09X\x0b
+    ** Failers
+    \x{a0} X\x0a   
+    
+/\H*\h+\V?\v{3,4}/8 
+    \x09\x20\x{a0}X\x0a\x0b\x0c\x0d\x0a
+    \x09\x20\x{a0}\x0a\x0b\x0c\x0d\x0a
+    \x09\x20\x{a0}\x0a\x0b\x0c
+    ** Failers 
+    \x09\x20\x{a0}\x0a\x0b
+     
+/\H\h\V\v/8
+    \x{3001}\x{3000}\x{2030}\x{2028}
+    X\x{180e}X\x{85}
+    ** Failers
+    \x{2009} X\x0a   
+    
+/\H*\h+\V?\v{3,4}/8 
+    \x{1680}\x{180e}\x{2007}X\x{2028}\x{2029}\x0c\x0d\x0a
+    \x09\x{205f}\x{a0}\x0a\x{2029}\x0c\x{2028}\x0a
+    \x09\x20\x{202f}\x0a\x0b\x0c
+    ** Failers 
+    \x09\x{200a}\x{a0}\x{2028}\x0b
+     
+/[\h]/8BZ
+    >\x{1680}
+
+/[\h]{3,}/8BZ
+    >\x{1680}\x{180e}\x{2000}\x{2003}\x{200a}\x{202f}\x{205f}\x{3000}<
+
+/[\v]/8BZ
+
+/[\H]/8BZ
+
+/[\V]/8BZ
+
+/.*$/8
+    \x{1ec5} 
+    
+/a\Rb/I8
+    a\rb
+    a\nb
+    a\r\nb
+    ** Failers
+    a\x{85}b
+    a\x0bb     
+
+/a\Rb/I8
+    a\rb
+    a\nb
+    a\r\nb
+    a\x{85}b
+    a\x0bb     
+    ** Failers 
+    a\x{85}b\
+    a\x0bb\
+    
+/a\R?b/I8
+    a\rb
+    a\nb
+    a\r\nb
+    ** Failers
+    a\x{85}b
+    a\x0bb     
+
+/a\R?b/I8
+    a\rb
+    a\nb
+    a\r\nb
+    a\x{85}b
+    a\x0bb     
+    ** Failers 
+    a\x{85}b\
+    a\x0bb\
+ 
+/.*a.*=.b.*/8
+    QQQ\x{2029}ABCaXYZ=!bPQR
+    ** Failers
+    a\x{2029}b
+    \x61\xe2\x80\xa9\x62 
+
+/[[:a\x{100}b:]]/8
+
+/a[^]b/8
+    a\x{1234}b
+    a\nb 
+    ** Failers
+    ab  
+    
+/a[^]+b/8
+    aXb
+    a\nX\nX\x{1234}b 
+    ** Failers
+    ab  
+
+/(\x{de})\1/
+    \x{de}\x{de}
+
+/X/8f 
+    A\x{1ec5}ABCXYZ
+
+/Xa{2,4}b/8
+    X\P
+    Xa\P
+    Xaa\P 
+    Xaaa\P
+    Xaaaa\P 
+    
+/Xa{2,4}?b/8
+    X\P
+    Xa\P
+    Xaa\P 
+    Xaaa\P
+    Xaaaa\P 
+    
+/Xa{2,4}+b/8
+    X\P
+    Xa\P
+    Xaa\P 
+    Xaaa\P
+    Xaaaa\P 
+    
+/X\x{123}{2,4}b/8
+    X\P
+    X\x{123}\P
+    X\x{123}\x{123}\P 
+    X\x{123}\x{123}\x{123}\P
+    X\x{123}\x{123}\x{123}\x{123}\P 
+    
+/X\x{123}{2,4}?b/8
+    X\P
+    X\x{123}\P
+    X\x{123}\x{123}\P 
+    X\x{123}\x{123}\x{123}\P
+    X\x{123}\x{123}\x{123}\x{123}\P 
+    
+/X\x{123}{2,4}+b/8
+    X\P
+    X\x{123}\P
+    X\x{123}\x{123}\P 
+    X\x{123}\x{123}\x{123}\P
+    X\x{123}\x{123}\x{123}\x{123}\P 
+    
+/X\x{123}{2,4}b/8
+    Xx\P
+    X\x{123}x\P
+    X\x{123}\x{123}x\P 
+    X\x{123}\x{123}\x{123}x\P
+    X\x{123}\x{123}\x{123}\x{123}x\P 
+    
+/X\x{123}{2,4}?b/8
+    Xx\P
+    X\x{123}x\P
+    X\x{123}\x{123}x\P 
+    X\x{123}\x{123}\x{123}x\P
+    X\x{123}\x{123}\x{123}\x{123}x\P 
+    
+/X\x{123}{2,4}+b/8
+    Xx\P
+    X\x{123}x\P
+    X\x{123}\x{123}x\P 
+    X\x{123}\x{123}\x{123}x\P
+    X\x{123}\x{123}\x{123}\x{123}x\P 
+    
+/X\d{2,4}b/8
+    X\P
+    X3\P
+    X33\P 
+    X333\P
+    X3333\P 
+    
+/X\d{2,4}?b/8
+    X\P
+    X3\P
+    X33\P 
+    X333\P
+    X3333\P 
+    
+/X\d{2,4}+b/8
+    X\P
+    X3\P
+    X33\P 
+    X333\P
+    X3333\P 
+
+/X\D{2,4}b/8
+    X\P
+    Xa\P
+    Xaa\P 
+    Xaaa\P
+    Xaaaa\P 
+    
+/X\D{2,4}?b/8
+    X\P
+    Xa\P
+    Xaa\P 
+    Xaaa\P
+    Xaaaa\P 
+    
+/X\D{2,4}+b/8
+    X\P
+    Xa\P
+    Xaa\P 
+    Xaaa\P
+    Xaaaa\P 
+
+/X\D{2,4}b/8
+    X\P
+    X\x{123}\P
+    X\x{123}\x{123}\P 
+    X\x{123}\x{123}\x{123}\P
+    X\x{123}\x{123}\x{123}\x{123}\P 
+    
+/X\D{2,4}?b/8
+    X\P
+    X\x{123}\P
+    X\x{123}\x{123}\P 
+    X\x{123}\x{123}\x{123}\P
+    X\x{123}\x{123}\x{123}\x{123}\P 
+    
+/X\D{2,4}+b/8
+    X\P
+    X\x{123}\P
+    X\x{123}\x{123}\P 
+    X\x{123}\x{123}\x{123}\P
+    X\x{123}\x{123}\x{123}\x{123}\P 
+
+/X[abc]{2,4}b/8
+    X\P
+    Xa\P
+    Xaa\P 
+    Xaaa\P
+    Xaaaa\P 
+    
+/X[abc]{2,4}?b/8
+    X\P
+    Xa\P
+    Xaa\P 
+    Xaaa\P
+    Xaaaa\P 
+    
+/X[abc]{2,4}+b/8
+    X\P
+    Xa\P
+    Xaa\P 
+    Xaaa\P
+    Xaaaa\P 
+
+/X[abc\x{123}]{2,4}b/8
+    X\P
+    X\x{123}\P
+    X\x{123}\x{123}\P 
+    X\x{123}\x{123}\x{123}\P
+    X\x{123}\x{123}\x{123}\x{123}\P 
+    
+/X[abc\x{123}]{2,4}?b/8
+    X\P
+    X\x{123}\P
+    X\x{123}\x{123}\P 
+    X\x{123}\x{123}\x{123}\P
+    X\x{123}\x{123}\x{123}\x{123}\P 
+    
+/X[abc\x{123}]{2,4}+b/8
+    X\P
+    X\x{123}\P
+    X\x{123}\x{123}\P 
+    X\x{123}\x{123}\x{123}\P
+    X\x{123}\x{123}\x{123}\x{123}\P 
+
+/X[^a]{2,4}b/8
+    X\P
+    Xz\P
+    Xzz\P 
+    Xzzz\P
+    Xzzzz\P 
+    
+/X[^a]{2,4}?b/8
+    X\P
+    Xz\P
+    Xzz\P 
+    Xzzz\P
+    Xzzzz\P 
+    
+/X[^a]{2,4}+b/8
+    X\P
+    Xz\P
+    Xzz\P 
+    Xzzz\P
+    Xzzzz\P 
+
+/X[^a]{2,4}b/8
+    X\P
+    X\x{123}\P
+    X\x{123}\x{123}\P 
+    X\x{123}\x{123}\x{123}\P
+    X\x{123}\x{123}\x{123}\x{123}\P 
+    
+/X[^a]{2,4}?b/8
+    X\P
+    X\x{123}\P
+    X\x{123}\x{123}\P 
+    X\x{123}\x{123}\x{123}\P
+    X\x{123}\x{123}\x{123}\x{123}\P 
+    
+/X[^a]{2,4}+b/8
+    X\P
+    X\x{123}\P
+    X\x{123}\x{123}\P 
+    X\x{123}\x{123}\x{123}\P
+    X\x{123}\x{123}\x{123}\x{123}\P 
+
+/(Y)X\1{2,4}b/8
+    YX\P
+    YXY\P
+    YXYY\P 
+    YXYYY\P
+    YXYYYY\P 
+    
+/(Y)X\1{2,4}?b/8
+    YX\P
+    YXY\P
+    YXYY\P 
+    YXYYY\P
+    YXYYYY\P 
+    
+/(Y)X\1{2,4}+b/8
+    YX\P
+    YXY\P
+    YXYY\P 
+    YXYYY\P
+    YXYYYY\P 
+
+/(\x{123})X\1{2,4}b/8
+    \x{123}X\P
+    \x{123}X\x{123}\P
+    \x{123}X\x{123}\x{123}\P 
+    \x{123}X\x{123}\x{123}\x{123}\P
+    \x{123}X\x{123}\x{123}\x{123}\x{123}\P 
+    
+/(\x{123})X\1{2,4}?b/8
+    \x{123}X\P
+    \x{123}X\x{123}\P
+    \x{123}X\x{123}\x{123}\P 
+    \x{123}X\x{123}\x{123}\x{123}\P
+    \x{123}X\x{123}\x{123}\x{123}\x{123}\P 
+    
+/(\x{123})X\1{2,4}+b/8
+    \x{123}X\P
+    \x{123}X\x{123}\P
+    \x{123}X\x{123}\x{123}\P 
+    \x{123}X\x{123}\x{123}\x{123}\P
+    \x{123}X\x{123}\x{123}\x{123}\x{123}\P 
+
+/\bthe cat\b/8
+    the cat\P
+    the cat\P\P
+
+/abcd*/8
+    xxxxabcd\P
+    xxxxabcd\P\P
+
+/abcd*/i8
+    xxxxabcd\P
+    xxxxabcd\P\P
+    XXXXABCD\P
+    XXXXABCD\P\P
+
+/abc\d*/8
+    xxxxabc1\P
+    xxxxabc1\P\P
+
+/(a)bc\1*/8
+    xxxxabca\P
+    xxxxabca\P\P
+
+/abc[de]*/8
+    xxxxabcde\P
+    xxxxabcde\P\P
+
+/X\W{3}X/8
+    \PX
+
+/\sxxx\s/8T1
+    AB\x{85}xxx\x{a0}XYZ
+    AB\x{a0}xxx\x{85}XYZ
+
+/\S \S/8T1
+    \x{a2} \x{84} 
+
+'A#хц'8xBZ
+
+'A#хц
+  PQ'8xBZ
+  
+/a+#хaa
+  z#XX?/8xBZ 
+
+/a+#хaa
+  z#х?/8xBZ 
+
+/\g{A}xxx#bXX(?'A'123)
(?'A'456)/8xBZ
+
+/\g{A}xxx#bх(?'A'123)
(?'A'456)/8xBZ
+
+/^\cģ/8
+
+/(\R*)(.)/s8
+    \r\n
+    \r\r\n\n\r 
+    \r\r\n\n\r\n 
+
+/(\R)*(.)/s8
+    \r\n
+    \r\r\n\n\r 
+    \r\r\n\n\r\n 
+
+/[^\x{1234}]+/iS8I   
+
+/[^\x{1234}]+?/iS8I   
+
+/[^\x{1234}]++/iS8I   
+
+/[^\x{1234}]{2}/iS8I
+
+//
+
+/f.*/
+    \P\Pfor
+
+/f.*/s
+    \P\Pfor
+
+/f.*/8
+    \P\Pfor
+
+/f.*/8s
+    \P\Pfor
+    
+/\x{d7ff}\x{e000}/8
+
+/\x{d800}/8
+
+/\x{dfff}/8 
+
+/\h+/8
+    \x{1681}\x{200b}\x{1680}\x{2000}\x{202f}\x{3000}
+    \x{3001}\x{2fff}\x{200a}\x{a0}\x{2000}
+
+/[\h\x{e000}]+/8BZ
+    \x{1681}\x{200b}\x{1680}\x{2000}\x{202f}\x{3000}
+    \x{3001}\x{2fff}\x{200a}\x{a0}\x{2000}
+
+/\H+/8
+    \x{1680}\x{180e}\x{167f}\x{1681}\x{180d}\x{180f}
+    \x{2000}\x{200a}\x{1fff}\x{200b}
+    \x{202f}\x{205f}\x{202e}\x{2030}\x{205e}\x{2060}
+    \x{a0}\x{3000}\x{9f}\x{a1}\x{2fff}\x{3001}
+
+/[\H\x{d7ff}]+/8BZ
+    \x{1680}\x{180e}\x{167f}\x{1681}\x{180d}\x{180f}
+    \x{2000}\x{200a}\x{1fff}\x{200b}
+    \x{202f}\x{205f}\x{202e}\x{2030}\x{205e}\x{2060}
+    \x{a0}\x{3000}\x{9f}\x{a1}\x{2fff}\x{3001}
+
+/\v+/8
+    \x{2027}\x{2030}\x{2028}\x{2029}
+    \x09\x0e\x{84}\x{86}\x{85}\x0a\x0b\x0c\x0d
+
+/[\v\x{e000}]+/8BZ
+    \x{2027}\x{2030}\x{2028}\x{2029}
+    \x09\x0e\x{84}\x{86}\x{85}\x0a\x0b\x0c\x0d
+
+/\V+/8
+    \x{2028}\x{2029}\x{2027}\x{2030}
+    \x{85}\x0a\x0b\x0c\x0d\x09\x0e\x{84}\x{86}
+
+/[\V\x{d7ff}]+/8BZ
+    \x{2028}\x{2029}\x{2027}\x{2030}
+    \x{85}\x0a\x0b\x0c\x0d\x09\x0e\x{84}\x{86}
+
+/\R+/8
+    \x{2027}\x{2030}\x{2028}\x{2029}
+    \x09\x0e\x{84}\x{86}\x{85}\x0a\x0b\x0c\x0d
+
+/(..)\1/8
+    ab\P
+    aba\P
+    abab\P
+
+/(..)\1/8i
+    ab\P
+    abA\P
+    aBAb\P
+
+/(..)\1{2,}/8
+    ab\P
+    aba\P
+    abab\P
+    ababa\P
+    ababab\P
+    ababab\P\P
+    abababa\P
+    abababa\P\P
+
+/(..)\1{2,}/8i
+    ab\P
+    aBa\P
+    aBAb\P
+    AbaBA\P
+    abABAb\P
+    aBAbaB\P\P
+    abABabA\P
+    abaBABa\P\P
+
+/(..)\1{2,}?x/8i
+    ab\P
+    abA\P
+    aBAb\P
+    abaBA\P
+    abAbaB\P
+    abaBabA\P
+    abAbABaBx\P
+
+/./8
+    \r\P
+    \r\P\P 
+  
+/.{2,3}/8
+    \r\P 
+    \r\P\P
+    \r\r\P
+    \r\r\P\P
+    \r\r\r\P
+    \r\r\r\P\P     
+
+/.{2,3}?/8
+    \r\P 
+    \r\P\P
+    \r\r\P
+    \r\r\P\P
+    \r\r\r\P
+    \r\r\r\P\P     
+
+/[^\x{100}][^\x{1234}][^\x{ffff}][^\x{10000}][^\x{10ffff}]/8BZ
+
+/[^\x{100}][^\x{1234}][^\x{ffff}][^\x{10000}][^\x{10ffff}]/8BZi
+
+/[^\x{100}]*[^\x{10000}]+[^\x{10ffff}]??[^\x{8000}]{4,}[^\x{7fff}]{2,9}?[^\x{fffff}]{5,6}+/8BZ
+
+/[^\x{100}]*[^\x{10000}]+[^\x{10ffff}]??[^\x{8000}]{4,}[^\x{7fff}]{2,9}?[^\x{fffff}]{5,6}+/8BZi
+
+/(?<=\x{1234}\x{1234})\bxy/I8
+
+/(?8BZ
+
+/[\u0100-\u0200]/8BZ
+
+/\ud800/8
+
+/-- End of testinput5 --/
diff --git a/pcre/testdata/testinput6 b/pcre/testdata/testinput6
new file mode 100644
index 00000000000..adbf621a98e
--- /dev/null
+++ b/pcre/testdata/testinput6
@@ -0,0 +1,1327 @@
+/-- This set of tests is for Unicode property support. It is compatible with
+    Perl >= 5.15. --/
+
+/^\pC\pL\pM\pN\pP\pS\pZ\s+/8W
+    >\x{20}\x{a0}\x{1680}\x{2028}\x{2029}\x{202f}\x{9}\x{b} 
+  
+/^>\pZ+/8W
+    >\x{20}\x{a0}\x{1680}\x{2028}\x{2029}\x{202f}\x{9}\x{b} 
+  
+/^>[[:space:]]*/8W
+    >\x{20}\x{a0}\x{1680}\x{2028}\x{2029}\x{202f}\x{9}\x{b} 
+
+/^>[[:blank:]]*/8W
+    >\x{20}\x{a0}\x{1680}\x{180e}\x{2000}\x{202f}\x{9}\x{b}\x{2028} 
+
+/^[[:alpha:]]*/8W
+    Az\x{aa}\x{c0}\x{1c5}\x{2b0}\x{3b6}\x{1d7c9}\x{2fa1d}
+
+/^[[:alnum:]]*/8W
+    Az\x{aa}\x{c0}\x{1c5}\x{2b0}\x{3b6}\x{1d7c9}\x{2fa1d}1\x{660}\x{bef}\x{16ee}
+
+/^[[:cntrl:]]*/8W
+    \x{0}\x{09}\x{1f}\x{7f}\x{9f} 
+
+/^[[:graph:]]*/8W
+    A\x{a1}\x{a0}
+
+/^[[:print:]]*/8W
+    A z\x{a0}\x{a1}
+
+/^[[:punct:]]*/8W
+    .+\x{a1}\x{a0}
+
+/\p{Zs}*?\R/
+    ** Failers
+    a\xFCb   
+
+/\p{Zs}*\R/                                                                    
+    ** Failers 
+    a\xFCb   
+
+/ⱥ/8i
+    ⱥ
+    Ⱥx 
+    Ⱥ 
+
+/[ⱥ]/8i
+    ⱥ
+    Ⱥx 
+    Ⱥ 
+
+/Ⱥ/8i
+    Ⱥ
+    ⱥ
+    
+/-- These are tests for extended grapheme clusters --/ 
+
+/^\X/8+
+    G\x{34e}\x{34e}X
+    \x{34e}\x{34e}X
+    \x04X
+    \x{1100}X
+    \x{1100}\x{34e}X
+    \x{1b04}\x{1b04}X 
+    *These match up to the roman letters
+    \x{1111}\x{1111}L,L
+    \x{1111}\x{1111}\x{1169}L,L,V
+    \x{1111}\x{ae4c}L, LV
+    \x{1111}\x{ad89}L, LVT
+    \x{1111}\x{ae4c}\x{1169}L, LV, V
+    \x{1111}\x{ae4c}\x{1169}\x{1169}L, LV, V, V
+    \x{1111}\x{ae4c}\x{1169}\x{11fe}L, LV, V, T
+    \x{1111}\x{ad89}\x{11fe}L, LVT, T
+    \x{1111}\x{ad89}\x{11fe}\x{11fe}L, LVT, T, T
+    \x{ad89}\x{11fe}\x{11fe}LVT, T, T
+    *These match just the first codepoint (invalid sequence)
+    \x{1111}\x{11fe}L, T
+    \x{ae4c}\x{1111}LV, L
+    \x{ae4c}\x{ae4c}LV, LV
+    \x{ae4c}\x{ad89}LV, LVT
+    \x{1169}\x{1111}V, L
+    \x{1169}\x{ae4c}V, LV
+    \x{1169}\x{ad89}V, LVT
+    \x{ad89}\x{1111}LVT, L
+    \x{ad89}\x{1169}LVT, V
+    \x{ad89}\x{ae4c}LVT, LV
+    \x{ad89}\x{ad89}LVT, LVT
+    \x{11fe}\x{1111}T, L
+    \x{11fe}\x{1169}T, V
+    \x{11fe}\x{ae4c}T, LV
+    \x{11fe}\x{ad89}T, LVT
+    *Test extend and spacing mark
+    \x{1111}\x{ae4c}\x{0711}L, LV, extend
+    \x{1111}\x{ae4c}\x{1b04}L, LV, spacing mark
+    \x{1111}\x{ae4c}\x{1b04}\x{0711}\x{1b04}L, LV, spacing mark, extend, spacing mark
+    *Test CR, LF, and control
+    \x0d\x{0711}CR, extend
+    \x0d\x{1b04}CR, spacingmark
+    \x0a\x{0711}LF, extend
+    \x0a\x{1b04}LF, spacingmark
+    \x0b\x{0711}Control, extend
+    \x09\x{1b04}Control, spacingmark
+    *There are no Prepend characters, so we can't test Prepend, CR
+    
+/^(?>\X{2})X/8+
+    \x{1111}\x{ae4c}\x{1111}\x{ae4c}X
+    
+/^\X{2,4}X/8+
+    \x{1111}\x{ae4c}\x{1111}\x{ae4c}X
+    \x{1111}\x{ae4c}\x{1111}\x{ae4c}\x{1111}\x{ae4c}X
+    \x{1111}\x{ae4c}\x{1111}\x{ae4c}\x{1111}\x{ae4c}\x{1111}\x{ae4c}X
+
+/^\X{2,4}?X/8+
+    \x{1111}\x{ae4c}\x{1111}\x{ae4c}X
+    \x{1111}\x{ae4c}\x{1111}\x{ae4c}\x{1111}\x{ae4c}X
+    \x{1111}\x{ae4c}\x{1111}\x{ae4c}\x{1111}\x{ae4c}\x{1111}\x{ae4c}X
+
+/-- --/
+
+/\x{1e9e}+/8i
+    \x{1e9e}\x{00df}
+
+/[z\x{1e9e}]+/8i
+    \x{1e9e}\x{00df}
+
+/\x{00df}+/8i
+    \x{1e9e}\x{00df}
+
+/[z\x{00df}]+/8i
+    \x{1e9e}\x{00df}
+
+/\x{1f88}+/8i
+    \x{1f88}\x{1f80} 
+
+/[z\x{1f88}]+/8i
+    \x{1f88}\x{1f80} 
+    
+/-- Characters with more than one other case; test in classes --/
+
+/[z\x{00b5}]+/8i
+    \x{00b5}\x{039c}\x{03bc}
+
+/[z\x{039c}]+/8i
+    \x{00b5}\x{039c}\x{03bc}
+
+/[z\x{03bc}]+/8i
+    \x{00b5}\x{039c}\x{03bc}
+
+/[z\x{00c5}]+/8i
+    \x{00c5}\x{00e5}\x{212b}
+
+/[z\x{00e5}]+/8i
+    \x{00c5}\x{00e5}\x{212b}
+
+/[z\x{212b}]+/8i
+    \x{00c5}\x{00e5}\x{212b}
+
+/[z\x{01c4}]+/8i
+    \x{01c4}\x{01c5}\x{01c6}
+
+/[z\x{01c5}]+/8i
+    \x{01c4}\x{01c5}\x{01c6}
+
+/[z\x{01c6}]+/8i
+    \x{01c4}\x{01c5}\x{01c6}
+
+/[z\x{01c7}]+/8i
+    \x{01c7}\x{01c8}\x{01c9}
+
+/[z\x{01c8}]+/8i
+    \x{01c7}\x{01c8}\x{01c9}
+
+/[z\x{01c9}]+/8i
+    \x{01c7}\x{01c8}\x{01c9}
+
+/[z\x{01ca}]+/8i
+    \x{01ca}\x{01cb}\x{01cc}
+
+/[z\x{01cb}]+/8i
+    \x{01ca}\x{01cb}\x{01cc}
+
+/[z\x{01cc}]+/8i
+    \x{01ca}\x{01cb}\x{01cc}
+
+/[z\x{01f1}]+/8i
+    \x{01f1}\x{01f2}\x{01f3}
+
+/[z\x{01f2}]+/8i
+    \x{01f1}\x{01f2}\x{01f3}
+
+/[z\x{01f3}]+/8i
+    \x{01f1}\x{01f2}\x{01f3}
+
+/[z\x{0345}]+/8i
+    \x{0345}\x{0399}\x{03b9}\x{1fbe}
+
+/[z\x{0399}]+/8i
+    \x{0345}\x{0399}\x{03b9}\x{1fbe}
+
+/[z\x{03b9}]+/8i
+    \x{0345}\x{0399}\x{03b9}\x{1fbe}
+
+/[z\x{1fbe}]+/8i
+    \x{0345}\x{0399}\x{03b9}\x{1fbe}
+
+/[z\x{0392}]+/8i
+    \x{0392}\x{03b2}\x{03d0}
+
+/[z\x{03b2}]+/8i
+    \x{0392}\x{03b2}\x{03d0}
+
+/[z\x{03d0}]+/8i
+    \x{0392}\x{03b2}\x{03d0}
+
+/[z\x{0395}]+/8i
+    \x{0395}\x{03b5}\x{03f5}
+
+/[z\x{03b5}]+/8i
+    \x{0395}\x{03b5}\x{03f5}
+
+/[z\x{03f5}]+/8i
+    \x{0395}\x{03b5}\x{03f5}
+
+/[z\x{0398}]+/8i
+    \x{0398}\x{03b8}\x{03d1}\x{03f4}
+
+/[z\x{03b8}]+/8i
+    \x{0398}\x{03b8}\x{03d1}\x{03f4}
+
+/[z\x{03d1}]+/8i
+    \x{0398}\x{03b8}\x{03d1}\x{03f4}
+
+/[z\x{03f4}]+/8i
+    \x{0398}\x{03b8}\x{03d1}\x{03f4}
+
+/[z\x{039a}]+/8i
+    \x{039a}\x{03ba}\x{03f0}
+
+/[z\x{03ba}]+/8i
+    \x{039a}\x{03ba}\x{03f0}
+
+/[z\x{03f0}]+/8i
+    \x{039a}\x{03ba}\x{03f0}
+
+/[z\x{03a0}]+/8i
+    \x{03a0}\x{03c0}\x{03d6} 
+
+/[z\x{03c0}]+/8i
+    \x{03a0}\x{03c0}\x{03d6} 
+
+/[z\x{03d6}]+/8i
+    \x{03a0}\x{03c0}\x{03d6} 
+
+/[z\x{03a1}]+/8i
+    \x{03a1}\x{03c1}\x{03f1}
+
+/[z\x{03c1}]+/8i
+    \x{03a1}\x{03c1}\x{03f1}
+
+/[z\x{03f1}]+/8i
+    \x{03a1}\x{03c1}\x{03f1}
+
+/[z\x{03a3}]+/8i
+    \x{03A3}\x{03C2}\x{03C3}
+
+/[z\x{03c2}]+/8i
+    \x{03A3}\x{03C2}\x{03C3}
+
+/[z\x{03c3}]+/8i
+    \x{03A3}\x{03C2}\x{03C3}
+
+/[z\x{03a6}]+/8i
+    \x{03a6}\x{03c6}\x{03d5} 
+
+/[z\x{03c6}]+/8i
+    \x{03a6}\x{03c6}\x{03d5} 
+
+/[z\x{03d5}]+/8i
+    \x{03a6}\x{03c6}\x{03d5} 
+
+/[z\x{03c9}]+/8i
+    \x{03c9}\x{03a9}\x{2126}
+
+/[z\x{03a9}]+/8i
+    \x{03c9}\x{03a9}\x{2126}
+
+/[z\x{2126}]+/8i
+    \x{03c9}\x{03a9}\x{2126}
+
+/[z\x{1e60}]+/8i
+    \x{1e60}\x{1e61}\x{1e9b}
+
+/[z\x{1e61}]+/8i
+    \x{1e60}\x{1e61}\x{1e9b}
+
+/[z\x{1e9b}]+/8i
+    \x{1e60}\x{1e61}\x{1e9b}
+
+/-- Perl 5.12.4 gets these wrong, but 5.15.3 is OK --/
+
+/[z\x{004b}]+/8i
+    \x{004b}\x{006b}\x{212a}
+
+/[z\x{006b}]+/8i
+    \x{004b}\x{006b}\x{212a}
+
+/[z\x{212a}]+/8i
+    \x{004b}\x{006b}\x{212a}
+
+/[z\x{0053}]+/8i
+    \x{0053}\x{0073}\x{017f}
+
+/[z\x{0073}]+/8i
+    \x{0053}\x{0073}\x{017f}
+
+/[z\x{017f}]+/8i
+    \x{0053}\x{0073}\x{017f}
+    
+/-- --/ 
+
+/(ΣΆΜΟΣ) \1/8i
+    ΣΆΜΟΣ ΣΆΜΟΣ
+    ΣΆΜΟΣ σάμος
+    σάμος σάμος
+    σάμος σάμοσ
+    σάμος ΣΆΜΟΣ  
+
+/(σάμος) \1/8i
+    ΣΆΜΟΣ ΣΆΜΟΣ
+    ΣΆΜΟΣ σάμος
+    σάμος σάμος
+    σάμος σάμοσ
+    σάμος ΣΆΜΟΣ  
+
+/(ΣΆΜΟΣ) \1*/8i
+    ΣΆΜΟΣ\x20
+    ΣΆΜΟΣ ΣΆΜΟΣσάμοςσάμος
+
+/-- Perl matches these --/
+
+/\x{00b5}+/8i
+    \x{00b5}\x{039c}\x{03bc}
+
+/\x{039c}+/8i
+    \x{00b5}\x{039c}\x{03bc}
+
+/\x{03bc}+/8i
+    \x{00b5}\x{039c}\x{03bc}
+
+
+/\x{00c5}+/8i
+    \x{00c5}\x{00e5}\x{212b}
+
+/\x{00e5}+/8i
+    \x{00c5}\x{00e5}\x{212b}
+
+/\x{212b}+/8i
+    \x{00c5}\x{00e5}\x{212b}
+
+
+/\x{01c4}+/8i
+    \x{01c4}\x{01c5}\x{01c6}
+
+/\x{01c5}+/8i
+    \x{01c4}\x{01c5}\x{01c6}
+
+/\x{01c6}+/8i
+    \x{01c4}\x{01c5}\x{01c6}
+
+
+/\x{01c7}+/8i
+    \x{01c7}\x{01c8}\x{01c9}
+
+/\x{01c8}+/8i
+    \x{01c7}\x{01c8}\x{01c9}
+
+/\x{01c9}+/8i
+    \x{01c7}\x{01c8}\x{01c9}
+
+
+/\x{01ca}+/8i
+    \x{01ca}\x{01cb}\x{01cc}
+
+/\x{01cb}+/8i
+    \x{01ca}\x{01cb}\x{01cc}
+
+/\x{01cc}+/8i
+    \x{01ca}\x{01cb}\x{01cc}
+
+
+/\x{01f1}+/8i
+    \x{01f1}\x{01f2}\x{01f3}
+
+/\x{01f2}+/8i
+    \x{01f1}\x{01f2}\x{01f3}
+
+/\x{01f3}+/8i
+    \x{01f1}\x{01f2}\x{01f3}
+
+
+/\x{0345}+/8i
+    \x{0345}\x{0399}\x{03b9}\x{1fbe}
+
+/\x{0399}+/8i
+    \x{0345}\x{0399}\x{03b9}\x{1fbe}
+
+/\x{03b9}+/8i
+    \x{0345}\x{0399}\x{03b9}\x{1fbe}
+
+/\x{1fbe}+/8i
+    \x{0345}\x{0399}\x{03b9}\x{1fbe}
+
+
+/\x{0392}+/8i
+    \x{0392}\x{03b2}\x{03d0}
+
+/\x{03b2}+/8i
+    \x{0392}\x{03b2}\x{03d0}
+
+/\x{03d0}+/8i
+    \x{0392}\x{03b2}\x{03d0}
+    
+
+/\x{0395}+/8i
+    \x{0395}\x{03b5}\x{03f5}
+
+/\x{03b5}+/8i
+    \x{0395}\x{03b5}\x{03f5}
+
+/\x{03f5}+/8i
+    \x{0395}\x{03b5}\x{03f5}
+
+
+/\x{0398}+/8i
+    \x{0398}\x{03b8}\x{03d1}\x{03f4}
+
+/\x{03b8}+/8i
+    \x{0398}\x{03b8}\x{03d1}\x{03f4}
+
+/\x{03d1}+/8i
+    \x{0398}\x{03b8}\x{03d1}\x{03f4}
+
+/\x{03f4}+/8i
+    \x{0398}\x{03b8}\x{03d1}\x{03f4}
+    
+
+/\x{039a}+/8i
+    \x{039a}\x{03ba}\x{03f0}
+
+/\x{03ba}+/8i
+    \x{039a}\x{03ba}\x{03f0}
+
+/\x{03f0}+/8i
+    \x{039a}\x{03ba}\x{03f0}
+    
+
+/\x{03a0}+/8i
+    \x{03a0}\x{03c0}\x{03d6} 
+
+/\x{03c0}+/8i
+    \x{03a0}\x{03c0}\x{03d6} 
+
+/\x{03d6}+/8i
+    \x{03a0}\x{03c0}\x{03d6} 
+
+
+/\x{03a1}+/8i
+    \x{03a1}\x{03c1}\x{03f1}
+
+/\x{03c1}+/8i
+    \x{03a1}\x{03c1}\x{03f1}
+
+/\x{03f1}+/8i
+    \x{03a1}\x{03c1}\x{03f1}
+
+
+/\x{03a3}+/8i
+    \x{03A3}\x{03C2}\x{03C3}
+
+/\x{03c2}+/8i
+    \x{03A3}\x{03C2}\x{03C3}
+
+/\x{03c3}+/8i
+    \x{03A3}\x{03C2}\x{03C3}
+    
+
+/\x{03a6}+/8i
+    \x{03a6}\x{03c6}\x{03d5} 
+
+/\x{03c6}+/8i
+    \x{03a6}\x{03c6}\x{03d5} 
+
+/\x{03d5}+/8i
+    \x{03a6}\x{03c6}\x{03d5} 
+
+
+/\x{03c9}+/8i
+    \x{03c9}\x{03a9}\x{2126}
+
+/\x{03a9}+/8i
+    \x{03c9}\x{03a9}\x{2126}
+
+/\x{2126}+/8i
+    \x{03c9}\x{03a9}\x{2126}
+    
+
+/\x{1e60}+/8i
+    \x{1e60}\x{1e61}\x{1e9b}
+
+/\x{1e61}+/8i
+    \x{1e60}\x{1e61}\x{1e9b}
+
+/\x{1e9b}+/8i
+    \x{1e60}\x{1e61}\x{1e9b}
+    
+
+/\x{1e9e}+/8i
+    \x{1e9e}\x{00df}
+
+/\x{00df}+/8i
+    \x{1e9e}\x{00df}
+    
+
+/\x{1f88}+/8i
+    \x{1f88}\x{1f80} 
+
+/\x{1f80}+/8i
+    \x{1f88}\x{1f80} 
+
+
+/-- Perl 5.12.4 gets these wrong, but 5.15.3 is OK --/
+
+/\x{004b}+/8i
+    \x{004b}\x{006b}\x{212a}
+
+/\x{006b}+/8i
+    \x{004b}\x{006b}\x{212a}
+
+/\x{212a}+/8i
+    \x{004b}\x{006b}\x{212a}
+
+
+/\x{0053}+/8i
+    \x{0053}\x{0073}\x{017f}
+
+/\x{0073}+/8i
+    \x{0053}\x{0073}\x{017f}
+
+/\x{017f}+/8i
+    \x{0053}\x{0073}\x{017f}
+
+/^\p{Any}*\d{4}/8
+    1234
+    123 
+ 
+/^\X*\w{4}/8
+    1234
+    123  
+    
+/-- End of testinput6 --/
diff --git a/pcre/testdata/testinput7 b/pcre/testdata/testinput7
new file mode 100644
index 00000000000..24c00d278b4
--- /dev/null
+++ b/pcre/testdata/testinput7
@@ -0,0 +1,738 @@
+/-- These tests for Unicode property support test PCRE's API and show some of
+    the compiled code. They are not Perl-compatible. --/
+
+/[\p{L}]/DZ
+
+/[\p{^L}]/DZ
+
+/[\P{L}]/DZ
+
+/[\P{^L}]/DZ
+
+/[abc\p{L}\x{0660}]/8DZ
+
+/[\p{Nd}]/8DZ
+    1234
+
+/[\p{Nd}+-]+/8DZ
+    1234
+    12-34
+    12+\x{661}-34  
+    ** Failers
+    abcd  
+
+/[\x{105}-\x{109}]/8iDZ
+    \x{104}
+    \x{105}
+    \x{109}  
+    ** Failers
+    \x{100}
+    \x{10a} 
+    
+/[z-\x{100}]/8iDZ 
+    Z
+    z
+    \x{39c}
+    \x{178}
+    |
+    \x{80}
+    \x{ff}
+    \x{100}
+    \x{101} 
+    ** Failers
+    \x{102}
+    Y
+    y           
+
+/[z-\x{100}]/8DZi
+
+/(?:[\PPa*]*){8,}/
+
+/[\P{Any}]/BZ
+
+/[\P{Any}\E]/BZ
+
+/(\P{Yi}+\277)/
+
+/(\P{Yi}+\277)?/
+
+/(?<=\P{Yi}{3}A)X/
+
+/\p{Yi}+(\P{Yi}+)(?1)/
+
+/(\P{Yi}{2}\277)?/
+
+/[\P{Yi}A]/
+
+/[\P{Yi}\P{Yi}\P{Yi}A]/
+
+/[^\P{Yi}A]/
+
+/[^\P{Yi}\P{Yi}\P{Yi}A]/
+
+/(\P{Yi}*\277)*/
+
+/(\P{Yi}*?\277)*/
+
+/(\p{Yi}*+\277)*/
+
+/(\P{Yi}?\277)*/
+
+/(\P{Yi}??\277)*/
+
+/(\p{Yi}?+\277)*/
+
+/(\P{Yi}{0,3}\277)*/
+
+/(\P{Yi}{0,3}?\277)*/
+
+/(\p{Yi}{0,3}+\277)*/
+
+/\p{Zl}{2,3}+/8BZ
+    


+    \x{2028}\x{2028}\x{2028}
+    
+/\p{Zl}/8BZ
+
+/\p{Lu}{3}+/8BZ
+
+/\pL{2}+/8BZ
+
+/\p{Cc}{2}+/8BZ
+
+/^\p{Cs}/8
+    \?\x{dfff}
+    ** Failers
+    \x{09f} 
+  
+/^\p{Sc}+/8
+    $\x{a2}\x{a3}\x{a4}\x{a5}\x{a6}
+    \x{9f2}
+    ** Failers
+    X
+    \x{2c2}
+  
+/^\p{Zs}/8
+    \ \
+    \x{a0}
+    \x{1680}
+    \x{180e}
+    \x{2000}
+    \x{2001}     
+    ** Failers
+    \x{2028}
+    \x{200d} 
+  
+/-- These four are here rather than in test 6 because Perl has problems with
+    the negative versions of the properties. --/
+      
+/\p{^Lu}/8i
+    1234
+    ** Failers
+    ABC 
+
+/\P{Lu}/8i
+    1234
+    ** Failers
+    ABC 
+
+/\p{Ll}/8i 
+    a
+    Az
+    ** Failers
+    ABC   
+
+/\p{Lu}/8i
+    A
+    a\x{10a0}B 
+    ** Failers 
+    a
+    \x{1d00}  
+
+/[\x{c0}\x{391}]/8i
+    \x{c0}
+    \x{e0} 
+
+/-- The next two are special cases where the lengths of the different cases of
+the same character differ. The first went wrong with heap frame storage; the
+second was broken in all cases. --/
+
+/^\x{023a}+?(\x{0130}+)/8i
+  \x{023a}\x{2c65}\x{0130}
+  
+/^\x{023a}+([^X])/8i
+  \x{023a}\x{2c65}X
+
+/\x{c0}+\x{116}+/8i
+    \x{c0}\x{e0}\x{116}\x{117}
+
+/[\x{c0}\x{116}]+/8i
+    \x{c0}\x{e0}\x{116}\x{117}
+
+/(\x{de})\1/8i
+    \x{de}\x{de}
+    \x{de}\x{fe}
+    \x{fe}\x{fe}
+    \x{fe}\x{de}
+
+/^\x{c0}$/8i
+    \x{c0}
+    \x{e0} 
+
+/^\x{e0}$/8i
+    \x{c0}
+    \x{e0} 
+
+/-- The next two should be Perl-compatible, but it fails to match \x{e0}. PCRE
+will match it only with UCP support, because without that it has no notion
+of case for anything other than the ASCII letters. --/ 
+
+/((?i)[\x{c0}])/8
+    \x{c0}
+    \x{e0} 
+
+/(?i:[\x{c0}])/8
+    \x{c0}
+    \x{e0} 
+
+/-- These are PCRE's extra properties to help with Unicodizing \d etc. --/
+
+/^\p{Xan}/8
+    ABCD
+    1234
+    \x{6ca}
+    \x{a6c}
+    \x{10a7}   
+    ** Failers
+    _ABC   
+
+/^\p{Xan}+/8
+    ABCD1234\x{6ca}\x{a6c}\x{10a7}_
+    ** Failers
+    _ABC   
+
+/^\p{Xan}+?/8
+    \x{6ca}\x{a6c}\x{10a7}_
+
+/^\p{Xan}*/8
+    ABCD1234\x{6ca}\x{a6c}\x{10a7}_
+    
+/^\p{Xan}{2,9}/8
+    ABCD1234\x{6ca}\x{a6c}\x{10a7}_
+    
+/^\p{Xan}{2,9}?/8
+    \x{6ca}\x{a6c}\x{10a7}_
+    
+/^[\p{Xan}]/8
+    ABCD1234_
+    1234abcd_
+    \x{6ca}
+    \x{a6c}
+    \x{10a7}   
+    ** Failers
+    _ABC   
+ 
+/^[\p{Xan}]+/8
+    ABCD1234\x{6ca}\x{a6c}\x{10a7}_
+    ** Failers
+    _ABC   
+
+/^>\p{Xsp}/8
+    >\x{1680}\x{2028}\x{0b}
+    >\x{a0} 
+    ** Failers
+    \x{0b} 
+
+/^>\p{Xsp}+/8
+    > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+
+/^>\p{Xsp}+?/8
+    >\x{1680}\x{2028}\x{0b}
+
+/^>\p{Xsp}*/8
+    > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+    
+/^>\p{Xsp}{2,9}/8
+    > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+    
+/^>\p{Xsp}{2,9}?/8
+    > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+    
+/^>[\p{Xsp}]/8
+    >\x{2028}\x{0b}
+ 
+/^>[\p{Xsp}]+/8
+    > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+
+/^>\p{Xps}/8
+    >\x{1680}\x{2028}\x{0b}
+    >\x{a0} 
+    ** Failers
+    \x{0b} 
+
+/^>\p{Xps}+/8
+    > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+
+/^>\p{Xps}+?/8
+    >\x{1680}\x{2028}\x{0b}
+
+/^>\p{Xps}*/8
+    > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+    
+/^>\p{Xps}{2,9}/8
+    > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+    
+/^>\p{Xps}{2,9}?/8
+    > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+    
+/^>[\p{Xps}]/8
+    >\x{2028}\x{0b}
+ 
+/^>[\p{Xps}]+/8
+    > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+
+/^\p{Xwd}/8
+    ABCD
+    1234
+    \x{6ca}
+    \x{a6c}
+    \x{10a7}
+    _ABC    
+    ** Failers
+    [] 
+
+/^\p{Xwd}+/8
+    ABCD1234\x{6ca}\x{a6c}\x{10a7}_
+
+/^\p{Xwd}+?/8
+    \x{6ca}\x{a6c}\x{10a7}_
+
+/^\p{Xwd}*/8
+    ABCD1234\x{6ca}\x{a6c}\x{10a7}_
+    
+/^\p{Xwd}{2,9}/8
+    A_B12\x{6ca}\x{a6c}\x{10a7}
+    
+/^\p{Xwd}{2,9}?/8
+    \x{6ca}\x{a6c}\x{10a7}_
+    
+/^[\p{Xwd}]/8
+    ABCD1234_
+    1234abcd_
+    \x{6ca}
+    \x{a6c}
+    \x{10a7}   
+    _ABC 
+    ** Failers
+    []   
+ 
+/^[\p{Xwd}]+/8
+    ABCD1234\x{6ca}\x{a6c}\x{10a7}_
+
+/-- A check not in UTF-8 mode --/
+
+/^[\p{Xwd}]+/
+    ABCD1234_
+    
+/-- Some negative checks --/
+
+/^[\P{Xwd}]+/8
+    !.+\x{019}\x{35a}AB
+
+/^[\p{^Xwd}]+/8
+    !.+\x{019}\x{35a}AB
+
+/[\D]/WBZ8
+    1\x{3c8}2
+
+/[\d]/WBZ8
+    >\x{6f4}<
+
+/[\S]/WBZ8
+    \x{1680}\x{6f4}\x{1680}
+
+/[\s]/WBZ8
+    >\x{1680}<
+
+/[\W]/WBZ8
+    A\x{1712}B
+
+/[\w]/WBZ8
+    >\x{1723}<
+
+/\D/WBZ8
+    1\x{3c8}2
+
+/\d/WBZ8
+    >\x{6f4}<
+
+/\S/WBZ8
+    \x{1680}\x{6f4}\x{1680}
+
+/\s/WBZ8
+    >\x{1680}>
+
+/\W/WBZ8
+    A\x{1712}B
+
+/\w/WBZ8
+    >\x{1723}<
+
+/[[:alpha:]]/WBZ
+
+/[[:lower:]]/WBZ
+
+/[[:upper:]]/WBZ
+
+/[[:alnum:]]/WBZ
+
+/[[:ascii:]]/WBZ
+
+/[[:cntrl:]]/WBZ
+
+/[[:digit:]]/WBZ
+
+/[[:graph:]]/WBZ
+
+/[[:print:]]/WBZ
+
+/[[:punct:]]/WBZ
+
+/[[:space:]]/WBZ
+
+/[[:word:]]/WBZ
+
+/[[:xdigit:]]/WBZ
+
+/-- Unicode properties for \b abd \B --/
+
+/\b...\B/8W
+    abc_
+    \x{37e}abc\x{376} 
+    \x{37e}\x{376}\x{371}\x{393}\x{394} 
+    !\x{c0}++\x{c1}\x{c2} 
+    !\x{c0}+++++ 
+
+/-- Without PCRE_UCP, non-ASCII always fail, even if < 256  --/
+
+/\b...\B/8
+    abc_
+    ** Failers 
+    \x{37e}abc\x{376} 
+    \x{37e}\x{376}\x{371}\x{393}\x{394} 
+    !\x{c0}++\x{c1}\x{c2} 
+    !\x{c0}+++++ 
+
+/-- With PCRE_UCP, non-UTF8 chars that are < 256 still check properties  --/
+
+/\b...\B/W
+    abc_
+    !\x{c0}++\x{c1}\x{c2} 
+    !\x{c0}+++++ 
+
+/-- Some of these are silly, but they check various combinations --/
+
+/[[:^alpha:][:^cntrl:]]+/8WBZ
+    123
+    abc 
+
+/[[:^cntrl:][:^alpha:]]+/8WBZ
+    123
+    abc 
+
+/[[:alpha:]]+/8WBZ
+    abc
+
+/[[:^alpha:]\S]+/8WBZ
+    123
+    abc 
+
+/[^\d]+/8WBZ
+    abc123
+    abc\x{123}
+    \x{660}abc   
+
+/\p{Lu}+9\p{Lu}+B\p{Lu}+b/BZ
+
+/\p{^Lu}+9\p{^Lu}+B\p{^Lu}+b/BZ
+
+/\P{Lu}+9\P{Lu}+B\P{Lu}+b/BZ
+
+/\p{Han}+X\p{Greek}+\x{370}/BZ8
+
+/\p{Xan}+!\p{Xan}+A/BZ
+
+/\p{Xsp}+!\p{Xsp}\t/BZ
+
+/\p{Xps}+!\p{Xps}\t/BZ
+
+/\p{Xwd}+!\p{Xwd}_/BZ
+
+/A+\p{N}A+\dB+\p{N}*B+\d*/WBZ
+
+/-- These behaved oddly in Perl, so they are kept in this test --/
+
+/(\x{23a}\x{23a}\x{23a})?\1/8i
+    \x{23a}\x{23a}\x{23a}\x{2c65}\x{2c65}
+
+/(ȺȺȺ)?\1/8i
+    ȺȺȺⱥⱥ
+
+/(\x{23a}\x{23a}\x{23a})?\1/8i
+    \x{23a}\x{23a}\x{23a}\x{2c65}\x{2c65}\x{2c65}
+
+/(ȺȺȺ)?\1/8i
+    ȺȺȺⱥⱥⱥ
+
+/(\x{23a}\x{23a}\x{23a})\1/8i
+    \x{23a}\x{23a}\x{23a}\x{2c65}\x{2c65}
+
+/(ȺȺȺ)\1/8i
+    ȺȺȺⱥⱥ
+
+/(\x{23a}\x{23a}\x{23a})\1/8i
+    \x{23a}\x{23a}\x{23a}\x{2c65}\x{2c65}\x{2c65}
+
+/(ȺȺȺ)\1/8i
+    ȺȺȺⱥⱥⱥ
+
+/(\x{2c65}\x{2c65})\1/8i
+    \x{2c65}\x{2c65}\x{23a}\x{23a}
+    
+/(ⱥⱥ)\1/8i
+    ⱥⱥȺȺ 
+    
+/(\x{23a}\x{23a}\x{23a})\1Y/8i
+    X\x{23a}\x{23a}\x{23a}\x{2c65}\x{2c65}\x{2c65}YZ
+
+/(\x{2c65}\x{2c65})\1Y/8i
+    X\x{2c65}\x{2c65}\x{23a}\x{23a}YZ
+
+/-- --/ 
+
+/-- These scripts weren't yet in Perl when I added Unicode 6.0.0 to PCRE --/
+
+/^[\p{Batak}]/8
+    \x{1bc0}
+    \x{1bff}
+    ** Failers
+    \x{1bf4}
+    
+/^[\p{Brahmi}]/8
+    \x{11000}
+    \x{1106f}
+    ** Failers
+    \x{1104e}
+    
+/^[\p{Mandaic}]/8
+    \x{840}
+    \x{85e}
+    ** Failers
+    \x{85c}
+    \x{85d}    
+
+/-- --/ 
+
+/(\X*)(.)/s8
+    A\x{300}
+
+/^S(\X*)e(\X*)$/8
+    Stéréo
+    
+/^\X/8 
+    ́réo
+
+/^a\X41z/
+    aX41z
+    *** Failers
+    aAz
+
+/(?<=ab\Cde)X/8
+
+/\X/
+    a\P
+    a\P\P
+
+/\Xa/
+    aa\P
+    aa\P\P
+
+/\X{2}/
+    aa\P
+    aa\P\P
+
+/\X+a/
+    a\P
+    aa\P
+    aa\P\P
+
+/\X+?a/
+    a\P
+    ab\P
+    aa\P
+    aa\P\P
+    aba\P
+    
+/-- These Unicode 6.1.0 scripts are not known to Perl. --/ 
+
+/\p{Chakma}\d/8W
+    \x{11100}\x{1113c}
+
+/\p{Takri}\d/8W
+    \x{11680}\x{116c0}
+
+/^\X/8
+    A\P
+    A\P\P 
+    A\x{300}\x{301}\P
+    A\x{300}\x{301}\P\P  
+    A\x{301}\P
+    A\x{301}\P\P  
+    
+/^\X{2,3}/8
+    A\P
+    A\P\P 
+    AA\P
+    AA\P\P  
+    A\x{300}\x{301}\P
+    A\x{300}\x{301}\P\P  
+    A\x{300}\x{301}A\x{300}\x{301}\P
+    A\x{300}\x{301}A\x{300}\x{301}\P\P  
+
+/^\X{2}/8
+    AA\P
+    AA\P\P  
+    A\x{300}\x{301}A\x{300}\x{301}\P
+    A\x{300}\x{301}A\x{300}\x{301}\P\P  
+    
+/^\X+/8
+    AA\P
+    AA\P\P  
+
+/^\X+?Z/8
+    AA\P
+    AA\P\P 
+
+/A\x{3a3}B/8iDZ
+
+/\x{3a3}B/8iDZ
+
+/[\x{3a3}]/8iBZ
+
+/[^\x{3a3}]/8iBZ
+
+/[\x{3a3}]+/8iBZ
+
+/[^\x{3a3}]+/8iBZ
+
+/a*\x{3a3}/8iBZ
+
+/\x{3a3}+a/8iBZ
+
+/\x{3a3}*\x{3c2}/8iBZ
+
+/\x{3a3}{3}/8i+
+    \x{3a3}\x{3c3}\x{3c2}\x{3a3}\x{3c3}\x{3c2}
+
+/\x{3a3}{2,4}/8i+
+    \x{3a3}\x{3c3}\x{3c2}\x{3a3}\x{3c3}\x{3c2}
+
+/\x{3a3}{2,4}?/8i+
+    \x{3a3}\x{3c3}\x{3c2}\x{3a3}\x{3c3}\x{3c2}
+
+/\x{3a3}+./8i+
+    \x{3a3}\x{3c3}\x{3c2}\x{3a3}\x{3c3}\x{3c2}
+
+/\x{3a3}++./8i+
+    ** Failers
+    \x{3a3}\x{3c3}\x{3c2}\x{3a3}\x{3c3}\x{3c2}
+
+/\x{3a3}*\x{3c2}/8iBZ
+
+/[^\x{3a3}]*\x{3c2}/8iBZ
+
+/[^a]*\x{3c2}/8iBZ
+
+/ist/8iBZ
+    ikt
+
+/is+t/8i
+    iSs\x{17f}t
+    ikt
+
+/is+?t/8i
+    ikt
+
+/is?t/8i
+    ikt
+
+/is{2}t/8i
+    iskt
+    
+/-- This property is a PCRE special --/
+
+/^\p{Xuc}/8
+    $abc
+    @abc
+    `abc
+    \x{1234}abc
+    ** Failers
+    abc     
+
+/^\p{Xuc}+/8
+    $@`\x{a0}\x{1234}\x{e000}**
+    ** Failers
+    \x{9f}
+
+/^\p{Xuc}+?/8
+    $@`\x{a0}\x{1234}\x{e000}**
+    ** Failers
+    \x{9f}
+
+/^\p{Xuc}+?\*/8
+    $@`\x{a0}\x{1234}\x{e000}**
+    ** Failers
+    \x{9f}
+
+/^\p{Xuc}++/8
+    $@`\x{a0}\x{1234}\x{e000}**
+    ** Failers
+    \x{9f}
+
+/^\p{Xuc}{3,5}/8
+    $@`\x{a0}\x{1234}\x{e000}**
+    ** Failers
+    \x{9f}
+
+/^\p{Xuc}{3,5}?/8
+    $@`\x{a0}\x{1234}\x{e000}**
+    ** Failers
+    \x{9f}
+
+/^[\p{Xuc}]/8
+    $@`\x{a0}\x{1234}\x{e000}**
+    ** Failers
+    \x{9f}
+
+/^[\p{Xuc}]+/8
+    $@`\x{a0}\x{1234}\x{e000}**
+    ** Failers
+    \x{9f}
+
+/^\P{Xuc}/8
+    abc
+    ** Failers
+    $abc
+    @abc
+    `abc
+    \x{1234}abc
+
+/^[\P{Xuc}]/8
+    abc
+    ** Failers
+    $abc
+    @abc
+    `abc
+    \x{1234}abc
+
+/-- End of testinput7 --/
diff --git a/pcre/testdata/testinput8 b/pcre/testdata/testinput8
new file mode 100644
index 00000000000..d91013bb2be
--- /dev/null
+++ b/pcre/testdata/testinput8
@@ -0,0 +1,4804 @@
+/-- This set of tests check the DFA matching functionality of pcre_dfa_exec().
+    The -dfa flag must be used with pcretest when running it. --/
+     
+/abc/
+    abc
+    
+/ab*c/
+    abc
+    abbbbc
+    ac
+    
+/ab+c/
+    abc
+    abbbbbbc
+    *** Failers 
+    ac
+    ab
+    
+/a*/
+    a
+    aaaaaaaaaaaaaaaaa
+    aaaaaaaaaaaaaaaaaaaaaaaaaaaaaa 
+    aaaaaaaaaaaaaaaaaaaaaaaaaaaaaa\F 
+    
+/(a|abcd|african)/
+    a
+    abcd
+    african
+    
+/^abc/
+    abcdef
+    *** Failers
+    xyzabc
+    xyz\nabc    
+    
+/^abc/m
+    abcdef
+    xyz\nabc    
+    *** Failers
+    xyzabc
+    
+/\Aabc/
+    abcdef
+    *** Failers
+    xyzabc
+    xyz\nabc    
+    
+/\Aabc/m
+    abcdef
+    *** Failers
+    xyzabc
+    xyz\nabc    
+    
+/\Gabc/
+    abcdef
+    xyzabc\>3
+    *** Failers
+    xyzabc    
+    xyzabc\>2 
+    
+/x\dy\Dz/
+    x9yzz
+    x0y+z
+    *** Failers
+    xyz
+    xxy0z     
+    
+/x\sy\Sz/
+    x yzz
+    x y+z
+    *** Failers
+    xyz
+    xxyyz
+    
+/x\wy\Wz/
+    xxy+z
+    *** Failers
+    xxy0z
+    x+y+z         
+    
+/x.y/
+    x+y
+    x-y
+    *** Failers
+    x\ny
+    
+/x.y/s
+    x+y
+    x-y
+    x\ny
+
+/(a.b(?s)c.d|x.y)p.q/
+    a+bc+dp+q
+    a+bc\ndp+q
+    x\nyp+q 
+    *** Failers 
+    a\nbc\ndp+q
+    a+bc\ndp\nq
+    x\nyp\nq 
+
+/a\d\z/
+    ba0
+    *** Failers
+    ba0\n
+    ba0\ncd   
+
+/a\d\z/m
+    ba0
+    *** Failers
+    ba0\n
+    ba0\ncd   
+
+/a\d\Z/
+    ba0
+    ba0\n
+    *** Failers
+    ba0\ncd   
+
+/a\d\Z/m
+    ba0
+    ba0\n
+    *** Failers
+    ba0\ncd   
+
+/a\d$/
+    ba0
+    ba0\n
+    *** Failers
+    ba0\ncd   
+
+/a\d$/m
+    ba0
+    ba0\n
+    ba0\ncd   
+    *** Failers
+
+/abc/i
+    abc
+    aBc
+    ABC
+    
+/[^a]/
+    abcd
+    
+/ab?\w/
+    abz
+    abbz
+    azz  
+
+/x{0,3}yz/
+    ayzq
+    axyzq
+    axxyz
+    axxxyzq
+    axxxxyzq
+    *** Failers
+    ax
+    axx     
+      
+/x{3}yz/
+    axxxyzq
+    axxxxyzq
+    *** Failers
+    ax
+    axx     
+    ayzq
+    axyzq
+    axxyz
+      
+/x{2,3}yz/
+    axxyz
+    axxxyzq
+    axxxxyzq
+    *** Failers
+    ax
+    axx     
+    ayzq
+    axyzq
+      
+/[^a]+/
+    bac
+    bcdefax
+    *** Failers
+    aaaaa   
+
+/[^a]*/
+    bac
+    bcdefax
+    *** Failers
+    aaaaa   
+    
+/[^a]{3,5}/
+    xyz
+    awxyza
+    abcdefa
+    abcdefghijk
+    *** Failers
+    axya
+    axa
+    aaaaa         
+
+/\d*/
+    1234b567
+    xyz
+    
+/\D*/
+    a1234b567
+    xyz
+     
+/\d+/
+    ab1234c56
+    *** Failers
+    xyz
+    
+/\D+/
+    ab123c56
+    *** Failers
+    789
+    
+/\d?A/
+    045ABC
+    ABC
+    *** Failers
+    XYZ
+    
+/\D?A/
+    ABC
+    BAC
+    9ABC             
+    *** Failers
+
+/a+/
+    aaaa
+
+/^.*xyz/
+    xyz
+    ggggggggxyz
+    
+/^.+xyz/
+    abcdxyz
+    axyz
+    *** Failers
+    xyz
+    
+/^.?xyz/
+    xyz
+    cxyz       
+
+/^\d{2,3}X/
+    12X
+    123X
+    *** Failers
+    X
+    1X
+    1234X     
+
+/^[abcd]\d/
+    a45
+    b93
+    c99z
+    d04
+    *** Failers
+    e45
+    abcd      
+    abcd1234
+    1234  
+
+/^[abcd]*\d/
+    a45
+    b93
+    c99z
+    d04
+    abcd1234
+    1234  
+    *** Failers
+    e45
+    abcd      
+
+/^[abcd]+\d/
+    a45
+    b93
+    c99z
+    d04
+    abcd1234
+    *** Failers
+    1234  
+    e45
+    abcd      
+
+/^a+X/
+    aX
+    aaX 
+
+/^[abcd]?\d/
+    a45
+    b93
+    c99z
+    d04
+    1234  
+    *** Failers
+    abcd1234
+    e45
+
+/^[abcd]{2,3}\d/
+    ab45
+    bcd93
+    *** Failers
+    1234 
+    a36 
+    abcd1234
+    ee45
+
+/^(abc)*\d/
+    abc45
+    abcabcabc45
+    42xyz 
+    *** Failers
+
+/^(abc)+\d/
+    abc45
+    abcabcabc45
+    *** Failers
+    42xyz 
+
+/^(abc)?\d/
+    abc45
+    42xyz 
+    *** Failers
+    abcabcabc45
+
+/^(abc){2,3}\d/
+    abcabc45
+    abcabcabc45
+    *** Failers
+    abcabcabcabc45
+    abc45
+    42xyz 
+
+/1(abc|xyz)2(?1)3/
+    1abc2abc3456
+    1abc2xyz3456 
+
+/^(a*\w|ab)=(a*\w|ab)/
+    ab=ab
+
+/^(a*\w|ab)=(?1)/
+    ab=ab
+
+/^([^()]|\((?1)*\))*$/
+    abc
+    a(b)c
+    a(b(c))d  
+    *** Failers)
+    a(b(c)d  
+
+/^>abc>([^()]|\((?1)*\))*abc>123abc>1(2)3abc>(1(2)3)a*)\d/
+    aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa9876
+    *** Failers 
+    aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+
+/< (?: (?(R) \d++  | [^<>]*+) | (?R)) * >/x
+    <>
+    
+     hij>
+     hij>
+    def> 
+          
+    *** Failers
+    3 
+    *** Failers 
+    defabcxyz
+
+/^abcdef/
+    ab\P
+    abcde\P
+    abcdef\P
+    *** Failers
+    abx\P    
+
+/^a{2,4}\d+z/
+    a\P
+    aa\P
+    aa2\P 
+    aaa\P
+    aaa23\P 
+    aaaa12345\P
+    aa0z\P
+    aaaa4444444444444z\P 
+    *** Failers
+    az\P 
+    aaaaa\P 
+    a56\P 
+
+/^abcdef/
+   abc\P
+   def\R 
+   
+/(?<=foo)bar/
+   xyzfo\P 
+   foob\P\>2 
+   foobar...\R\P\>4 
+   xyzfo\P
+   foobar\>2  
+   *** Failers
+   xyzfo\P
+   obar\R   
+
+/(ab*(cd|ef))+X/
+    adfadadaklhlkalkajhlkjahdfasdfasdfladsfjkj\P\Z
+    lkjhlkjhlkjhlkjhabbbbbbcdaefabbbbbbbefa\P\B\Z
+    cdabbbbbbbb\P\R\B\Z
+    efabbbbbbbbbbbbbbbb\P\R\B\Z
+    bbbbbbbbbbbbcdXyasdfadf\P\R\B\Z    
+
+/(a|b)/SF>testsavedregex
+>>aaabxyzpqrrrabbxyyyypqAzz
+    >aaaabxyzpqrrrabbxyyyypqAzz
+    >>>>abcxyzpqrrrabbxyyyypqAzz
+    *** Failers
+    abxyzpqrrabbxyyyypqAzz
+    abxyzpqrrrrabbxyyyypqAzz
+    abxyzpqrrrabxyyyypqAzz
+    aaaabcxyzzzzpqrrrabbbxyyyyyypqAzz
+    aaaabcxyzzzzpqrrrabbbxyyypqAzz
+    aaabcxyzpqrrrabbxyyyypqqqqqqqAzz
+
+/^(abc){1,2}zz/
+    abczz
+    abcabczz
+    *** Failers
+    zz
+    abcabcabczz
+    >>abczz
+
+/^(b+?|a){1,2}?c/
+    bc
+    bbc
+    bbbc
+    bac
+    bbac
+    aac
+    abbbbbbbbbbbc
+    bbbbbbbbbbbac
+    *** Failers
+    aaac
+    abbbbbbbbbbbac
+
+/^(b+|a){1,2}c/
+    bc
+    bbc
+    bbbc
+    bac
+    bbac
+    aac
+    abbbbbbbbbbbc
+    bbbbbbbbbbbac
+    *** Failers
+    aaac
+    abbbbbbbbbbbac
+
+/^(b+|a){1,2}?bc/
+    bbc
+
+/^(b*|ba){1,2}?bc/
+    babc
+    bbabc
+    bababc
+    *** Failers
+    bababbc
+    babababc
+
+/^(ba|b*){1,2}?bc/
+    babc
+    bbabc
+    bababc
+    *** Failers
+    bababbc
+    babababc
+
+/^\ca\cA\c[\c{\c:/
+    \x01\x01\e;z
+
+/^[ab\]cde]/
+    athing
+    bthing
+    ]thing
+    cthing
+    dthing
+    ething
+    *** Failers
+    fthing
+    [thing
+    \\thing
+
+/^[]cde]/
+    ]thing
+    cthing
+    dthing
+    ething
+    *** Failers
+    athing
+    fthing
+
+/^[^ab\]cde]/
+    fthing
+    [thing
+    \\thing
+    *** Failers
+    athing
+    bthing
+    ]thing
+    cthing
+    dthing
+    ething
+
+/^[^]cde]/
+    athing
+    fthing
+    *** Failers
+    ]thing
+    cthing
+    dthing
+    ething
+
+/^\/
+    
+
+/^/
+    
+
+/^[0-9]+$/
+    0
+    1
+    2
+    3
+    4
+    5
+    6
+    7
+    8
+    9
+    10
+    100
+    *** Failers
+    abc
+
+/^.*nter/
+    enter
+    inter
+    uponter
+
+/^xxx[0-9]+$/
+    xxx0
+    xxx1234
+    *** Failers
+    xxx
+
+/^.+[0-9][0-9][0-9]$/
+    x123
+    xx123
+    123456
+    *** Failers
+    123
+    x1234
+
+/^.+?[0-9][0-9][0-9]$/
+    x123
+    xx123
+    123456
+    *** Failers
+    123
+    x1234
+
+/^([^!]+)!(.+)=apquxz\.ixr\.zzz\.ac\.uk$/
+    abc!pqr=apquxz.ixr.zzz.ac.uk
+    *** Failers
+    !pqr=apquxz.ixr.zzz.ac.uk
+    abc!=apquxz.ixr.zzz.ac.uk
+    abc!pqr=apquxz:ixr.zzz.ac.uk
+    abc!pqr=apquxz.ixr.zzz.ac.ukk
+
+/:/
+    Well, we need a colon: somewhere
+    *** Fail if we don't
+
+/([\da-f:]+)$/i
+    0abc
+    abc
+    fed
+    E
+    ::
+    5f03:12C0::932e
+    fed def
+    Any old stuff
+    *** Failers
+    0zzz
+    gzzz
+    fed\x20
+    Any old rubbish
+
+/^.*\.(\d{1,3})\.(\d{1,3})\.(\d{1,3})$/
+    .1.2.3
+    A.12.123.0
+    *** Failers
+    .1.2.3333
+    1.2.3
+    1234.2.3
+
+/^(\d+)\s+IN\s+SOA\s+(\S+)\s+(\S+)\s*\(\s*$/
+    1 IN SOA non-sp1 non-sp2(
+    1    IN    SOA    non-sp1    non-sp2   (
+    *** Failers
+    1IN SOA non-sp1 non-sp2(
+
+/^[a-zA-Z\d][a-zA-Z\d\-]*(\.[a-zA-Z\d][a-zA-z\d\-]*)*\.$/
+    a.
+    Z.
+    2.
+    ab-c.pq-r.
+    sxk.zzz.ac.uk.
+    x-.y-.
+    *** Failers
+    -abc.peq.
+
+/^\*\.[a-z]([a-z\-\d]*[a-z\d]+)?(\.[a-z]([a-z\-\d]*[a-z\d]+)?)*$/
+    *.a
+    *.b0-a
+    *.c3-b.c
+    *.c-a.b-c
+    *** Failers
+    *.0
+    *.a-
+    *.a-b.c-
+    *.c-a.0-c
+
+/^(?=ab(de))(abd)(e)/
+    abde
+
+/^(?!(ab)de|x)(abd)(f)/
+    abdf
+
+/^(?=(ab(cd)))(ab)/
+    abcd
+
+/^[\da-f](\.[\da-f])*$/i
+    a.b.c.d
+    A.B.C.D
+    a.b.c.1.2.3.C
+
+/^\".*\"\s*(;.*)?$/
+    \"1234\"
+    \"abcd\" ;
+    \"\" ; rhubarb
+    *** Failers
+    \"1234\" : things
+
+/^$/
+    \
+    *** Failers
+
+/   ^    a   (?# begins with a)  b\sc (?# then b c) $ (?# then end)/x
+    ab c
+    *** Failers
+    abc
+    ab cde
+
+/(?x)   ^    a   (?# begins with a)  b\sc (?# then b c) $ (?# then end)/
+    ab c
+    *** Failers
+    abc
+    ab cde
+
+/^   a\ b[c ]d       $/x
+    a bcd
+    a b d
+    *** Failers
+    abcd
+    ab d
+
+/^(a(b(c)))(d(e(f)))(h(i(j)))(k(l(m)))$/
+    abcdefhijklm
+
+/^(?:a(b(c)))(?:d(e(f)))(?:h(i(j)))(?:k(l(m)))$/
+    abcdefhijklm
+
+/^[\w][\W][\s][\S][\d][\D][\b][\n][\c]][\022]/
+    a+ Z0+\x08\n\x1d\x12
+
+/^[.^$|()*+?{,}]+/
+    .^\$(*+)|{?,?}
+
+/^a*\w/
+    z
+    az
+    aaaz
+    a
+    aa
+    aaaa
+    a+
+    aa+
+
+/^a*?\w/
+    z
+    az
+    aaaz
+    a
+    aa
+    aaaa
+    a+
+    aa+
+
+/^a+\w/
+    az
+    aaaz
+    aa
+    aaaa
+    aa+
+
+/^a+?\w/
+    az
+    aaaz
+    aa
+    aaaa
+    aa+
+
+/^\d{8}\w{2,}/
+    1234567890
+    12345678ab
+    12345678__
+    *** Failers
+    1234567
+
+/^[aeiou\d]{4,5}$/
+    uoie
+    1234
+    12345
+    aaaaa
+    *** Failers
+    123456
+
+/^[aeiou\d]{4,5}?/
+    uoie
+    1234
+    12345
+    aaaaa
+    123456
+
+/^From +([^ ]+) +[a-zA-Z][a-zA-Z][a-zA-Z] +[a-zA-Z][a-zA-Z][a-zA-Z] +[0-9]?[0-9] +[0-9][0-9]:[0-9][0-9]/
+    From abcd  Mon Sep 01 12:33:02 1997
+
+/^From\s+\S+\s+([a-zA-Z]{3}\s+){2}\d{1,2}\s+\d\d:\d\d/
+    From abcd  Mon Sep 01 12:33:02 1997
+    From abcd  Mon Sep  1 12:33:02 1997
+    *** Failers
+    From abcd  Sep 01 12:33:02 1997
+
+/^12.34/s
+    12\n34
+    12\r34
+
+/\w+(?=\t)/
+    the quick brown\t fox
+
+/foo(?!bar)(.*)/
+    foobar is foolish see?
+
+/(?:(?!foo)...|^.{0,2})bar(.*)/
+    foobar crowbar etc
+    barrel
+    2barrel
+    A barrel
+
+/^(\D*)(?=\d)(?!123)/
+    abc456
+    *** Failers
+    abc123
+
+/^1234(?# test newlines
+  inside)/
+    1234
+
+/^1234 #comment in extended re
+  /x
+    1234
+
+/#rhubarb
+  abcd/x
+    abcd
+
+/^abcd#rhubarb/x
+    abcd
+
+/(?!^)abc/
+    the abc
+    *** Failers
+    abc
+
+/(?=^)abc/
+    abc
+    *** Failers
+    the abc
+
+/^[ab]{1,3}(ab*|b)/
+    aabbbbb
+
+/^[ab]{1,3}?(ab*|b)/
+    aabbbbb
+
+/^[ab]{1,3}?(ab*?|b)/
+    aabbbbb
+
+/^[ab]{1,3}(ab*?|b)/
+    aabbbbb
+
+/  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*                          # optional leading comment
+(?:    (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+)                    # initial word
+(?:  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+)  )* # further okay, if led by a period
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  @  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*    (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                           # initial subdomain
+(?:                                  #
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.                        # if led by a period...
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                     #   ...further okay
+)*
+# address
+|                     #  or
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+)             # one word, optionally followed by....
+(?:
+[^()<>@,;:".\\\[\]\x80-\xff\000-\010\012-\037]  |  # atom and space parts, or...
+\(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)       |  # comments, or...
+
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+# quoted strings
+)*
+<  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*                     # leading <
+(?:  @  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*    (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                           # initial subdomain
+(?:                                  #
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.                        # if led by a period...
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                     #   ...further okay
+)*
+
+(?:  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  ,  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  @  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*    (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                           # initial subdomain
+(?:                                  #
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.                        # if led by a period...
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                     #   ...further okay
+)*
+)* # further okay, if led by comma
+:                                # closing colon
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  )? #       optional route
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+)                    # initial word
+(?:  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+)  )* # further okay, if led by a period
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  @  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*    (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                           # initial subdomain
+(?:                                  #
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.                        # if led by a period...
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                     #   ...further okay
+)*
+#       address spec
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  > #                  trailing >
+# name and address
+)  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*                       # optional trailing comment
+/x
+    Alan Other 
+    
+    user\@dom.ain
+    \"A. Other\"  (a comment)
+    A. Other  (a comment)
+    \"/s=user/ou=host/o=place/prmd=uu.yy/admd= /c=gb/\"\@x400-re.lay
+    A missing angle @,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+# Atom
+|                       #  or
+"                                     # "
+[^\\\x80-\xff\n\015"] *                            #   normal
+(?:  \\ [^\x80-\xff]  [^\\\x80-\xff\n\015"] * )*        #   ( special normal* )*
+"                                     #        "
+# Quoted string
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+\.
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+# Atom
+|                       #  or
+"                                     # "
+[^\\\x80-\xff\n\015"] *                            #   normal
+(?:  \\ [^\x80-\xff]  [^\\\x80-\xff\n\015"] * )*        #   ( special normal* )*
+"                                     #        "
+# Quoted string
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# additional words
+)*
+@
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+\[                            # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*     #    stuff
+\]                           #           ]
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# optional trailing comments
+(?:
+\.
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+\[                            # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*     #    stuff
+\]                           #           ]
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# optional trailing comments
+)*
+# address
+|                             #  or
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+# Atom
+|                       #  or
+"                                     # "
+[^\\\x80-\xff\n\015"] *                            #   normal
+(?:  \\ [^\x80-\xff]  [^\\\x80-\xff\n\015"] * )*        #   ( special normal* )*
+"                                     #        "
+# Quoted string
+)
+# leading word
+[^()<>@,;:".\\\[\]\x80-\xff\000-\010\012-\037] *               # "normal" atoms and or spaces
+(?:
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+|
+"                                     # "
+[^\\\x80-\xff\n\015"] *                            #   normal
+(?:  \\ [^\x80-\xff]  [^\\\x80-\xff\n\015"] * )*        #   ( special normal* )*
+"                                     #        "
+) # "special" comment or quoted string
+[^()<>@,;:".\\\[\]\x80-\xff\000-\010\012-\037] *            #  more "normal"
+)*
+<
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# <
+(?:
+@
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+\[                            # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*     #    stuff
+\]                           #           ]
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# optional trailing comments
+(?:
+\.
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+\[                            # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*     #    stuff
+\]                           #           ]
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# optional trailing comments
+)*
+(?: ,
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+@
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+\[                            # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*     #    stuff
+\]                           #           ]
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# optional trailing comments
+(?:
+\.
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+\[                            # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*     #    stuff
+\]                           #           ]
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# optional trailing comments
+)*
+)*  # additional domains
+:
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# optional trailing comments
+)?     #       optional route
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+# Atom
+|                       #  or
+"                                     # "
+[^\\\x80-\xff\n\015"] *                            #   normal
+(?:  \\ [^\x80-\xff]  [^\\\x80-\xff\n\015"] * )*        #   ( special normal* )*
+"                                     #        "
+# Quoted string
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+\.
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+# Atom
+|                       #  or
+"                                     # "
+[^\\\x80-\xff\n\015"] *                            #   normal
+(?:  \\ [^\x80-\xff]  [^\\\x80-\xff\n\015"] * )*        #   ( special normal* )*
+"                                     #        "
+# Quoted string
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# additional words
+)*
+@
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+\[                            # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*     #    stuff
+\]                           #           ]
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# optional trailing comments
+(?:
+\.
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+\[                            # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*     #    stuff
+\]                           #           ]
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# optional trailing comments
+)*
+#       address spec
+>                    #                 >
+# name and address
+)
+/x
+    Alan Other 
+    
+    user\@dom.ain
+    \"A. Other\"  (a comment)
+    A. Other  (a comment)
+    \"/s=user/ou=host/o=place/prmd=uu.yy/admd= /c=gb/\"\@x400-re.lay
+    A missing angle 
+    a\rb
+    *** Failers
+    a\nb
+
+/abc$/
+    abc
+    abc\n
+    *** Failers
+    abc\ndef
+
+/(abc)\123/
+    abc\x53
+
+/(abc)\223/
+    abc\x93
+
+/(abc)\323/
+    abc\xd3
+
+/(abc)\100/
+    abc\x40
+    abc\100
+
+/(abc)\1000/
+    abc\x400
+    abc\x40\x30
+    abc\1000
+    abc\100\x30
+    abc\100\060
+    abc\100\60
+
+/abc\81/
+    abc\081
+    abc\0\x38\x31
+
+/abc\91/
+    abc\091
+    abc\0\x39\x31
+
+/(a)(b)(c)(d)(e)(f)(g)(h)(i)(j)(k)\12\123/
+    abcdefghijk\12S
+
+/ab\idef/
+    abidef
+
+/a{0}bc/
+    bc
+
+/(a|(bc)){0,0}?xyz/
+    xyz
+
+/abc[\10]de/
+    abc\010de
+
+/abc[\1]de/
+    abc\1de
+
+/(abc)[\1]de/
+    abc\1de
+
+/(?s)a.b/
+    a\nb
+
+/^([^a])([^\b])([^c]*)([^d]{3,4})/
+    baNOTccccd
+    baNOTcccd
+    baNOTccd
+    bacccd
+    *** Failers
+    anything
+    b\bc   
+    baccd
+
+/[^a]/
+    Abc
+  
+/[^a]/i
+    Abc 
+
+/[^a]+/
+    AAAaAbc
+  
+/[^a]+/i
+    AAAaAbc 
+
+/[^a]+/
+    bbb\nccc
+   
+/[^k]$/
+    abc
+    *** Failers
+    abk   
+   
+/[^k]{2,3}$/
+    abc
+    kbc
+    kabc 
+    *** Failers
+    abk
+    akb
+    akk 
+
+/^\d{8,}\@.+[^k]$/
+    12345678\@a.b.c.d
+    123456789\@x.y.z
+    *** Failers
+    12345678\@x.y.uk
+    1234567\@a.b.c.d       
+
+/[^a]/
+    aaaabcd
+    aaAabcd 
+
+/[^a]/i
+    aaaabcd
+    aaAabcd 
+
+/[^az]/
+    aaaabcd
+    aaAabcd 
+
+/[^az]/i
+    aaaabcd
+    aaAabcd 
+
+/\000\001\002\003\004\005\006\007\010\011\012\013\014\015\016\017\020\021\022\023\024\025\026\027\030\031\032\033\034\035\036\037\040\041\042\043\044\045\046\047\050\051\052\053\054\055\056\057\060\061\062\063\064\065\066\067\070\071\072\073\074\075\076\077\100\101\102\103\104\105\106\107\110\111\112\113\114\115\116\117\120\121\122\123\124\125\126\127\130\131\132\133\134\135\136\137\140\141\142\143\144\145\146\147\150\151\152\153\154\155\156\157\160\161\162\163\164\165\166\167\170\171\172\173\174\175\176\177\200\201\202\203\204\205\206\207\210\211\212\213\214\215\216\217\220\221\222\223\224\225\226\227\230\231\232\233\234\235\236\237\240\241\242\243\244\245\246\247\250\251\252\253\254\255\256\257\260\261\262\263\264\265\266\267\270\271\272\273\274\275\276\277\300\301\302\303\304\305\306\307\310\311\312\313\314\315\316\317\320\321\322\323\324\325\326\327\330\331\332\333\334\335\336\337\340\341\342\343\344\345\346\347\350\351\352\353\354\355\356\357\360\361\362\363\364\365\366\367\370\371\372\373\374\375\376\377/
+ \000\001\002\003\004\005\006\007\010\011\012\013\014\015\016\017\020\021\022\023\024\025\026\027\030\031\032\033\034\035\036\037\040\041\042\043\044\045\046\047\050\051\052\053\054\055\056\057\060\061\062\063\064\065\066\067\070\071\072\073\074\075\076\077\100\101\102\103\104\105\106\107\110\111\112\113\114\115\116\117\120\121\122\123\124\125\126\127\130\131\132\133\134\135\136\137\140\141\142\143\144\145\146\147\150\151\152\153\154\155\156\157\160\161\162\163\164\165\166\167\170\171\172\173\174\175\176\177\200\201\202\203\204\205\206\207\210\211\212\213\214\215\216\217\220\221\222\223\224\225\226\227\230\231\232\233\234\235\236\237\240\241\242\243\244\245\246\247\250\251\252\253\254\255\256\257\260\261\262\263\264\265\266\267\270\271\272\273\274\275\276\277\300\301\302\303\304\305\306\307\310\311\312\313\314\315\316\317\320\321\322\323\324\325\326\327\330\331\332\333\334\335\336\337\340\341\342\343\344\345\346\347\350\351\352\353\354\355\356\357\360\361\362\363\364\365\366\367\370\371\372\373\374\375\376\377
+
+/P[^*]TAIRE[^*]{1,6}?LL/
+    xxxxxxxxxxxPSTAIREISLLxxxxxxxxx
+
+/P[^*]TAIRE[^*]{1,}?LL/
+    xxxxxxxxxxxPSTAIREISLLxxxxxxxxx
+
+/(\.\d\d[1-9]?)\d+/
+    1.230003938
+    1.875000282   
+    1.235  
+                  
+/(\.\d\d((?=0)|\d(?=\d)))/
+    1.230003938      
+    1.875000282
+    *** Failers 
+    1.235 
+    
+/a(?)b/
+    ab 
+ 
+/\b(foo)\s+(\w+)/i
+    Food is on the foo table
+    
+/foo(.*)bar/
+    The food is under the bar in the barn.
+    
+/foo(.*?)bar/  
+    The food is under the bar in the barn.
+
+/(.*)(\d*)/
+    I have 2 numbers: 53147
+    
+/(.*)(\d+)/
+    I have 2 numbers: 53147
+ 
+/(.*?)(\d*)/
+    I have 2 numbers: 53147
+
+/(.*?)(\d+)/
+    I have 2 numbers: 53147
+
+/(.*)(\d+)$/
+    I have 2 numbers: 53147
+
+/(.*?)(\d+)$/
+    I have 2 numbers: 53147
+
+/(.*)\b(\d+)$/
+    I have 2 numbers: 53147
+
+/(.*\D)(\d+)$/
+    I have 2 numbers: 53147
+
+/^\D*(?!123)/
+    ABC123
+     
+/^(\D*)(?=\d)(?!123)/
+    ABC445
+    *** Failers
+    ABC123
+    
+/^[W-]46]/
+    W46]789 
+    -46]789
+    *** Failers
+    Wall
+    Zebra
+    42
+    [abcd] 
+    ]abcd[
+       
+/^[W-\]46]/
+    W46]789 
+    Wall
+    Zebra
+    Xylophone  
+    42
+    [abcd] 
+    ]abcd[
+    \\backslash 
+    *** Failers
+    -46]789
+    well
+    
+/\d\d\/\d\d\/\d\d\d\d/
+    01/01/2000
+
+/word (?:[a-zA-Z0-9]+ ){0,10}otherword/
+  word cat dog elephant mussel cow horse canary baboon snake shark otherword
+  word cat dog elephant mussel cow horse canary baboon snake shark
+
+/word (?:[a-zA-Z0-9]+ ){0,300}otherword/
+  word cat dog elephant mussel cow horse canary baboon snake shark the quick brown fox and the lazy dog and several other words getting close to thirty by now I hope
+
+/^(a){0,0}/
+    bcd
+    abc
+    aab     
+
+/^(a){0,1}/
+    bcd
+    abc
+    aab  
+
+/^(a){0,2}/
+    bcd
+    abc
+    aab  
+
+/^(a){0,3}/
+    bcd
+    abc
+    aab
+    aaa   
+
+/^(a){0,}/
+    bcd
+    abc
+    aab
+    aaa
+    aaaaaaaa    
+
+/^(a){1,1}/
+    bcd
+    abc
+    aab  
+
+/^(a){1,2}/
+    bcd
+    abc
+    aab  
+
+/^(a){1,3}/
+    bcd
+    abc
+    aab
+    aaa   
+
+/^(a){1,}/
+    bcd
+    abc
+    aab
+    aaa
+    aaaaaaaa    
+
+/.*\.gif/
+    borfle\nbib.gif\nno
+
+/.{0,}\.gif/
+    borfle\nbib.gif\nno
+
+/.*\.gif/m
+    borfle\nbib.gif\nno
+
+/.*\.gif/s
+    borfle\nbib.gif\nno
+
+/.*\.gif/ms
+    borfle\nbib.gif\nno
+    
+/.*$/
+    borfle\nbib.gif\nno
+
+/.*$/m
+    borfle\nbib.gif\nno
+
+/.*$/s
+    borfle\nbib.gif\nno
+
+/.*$/ms
+    borfle\nbib.gif\nno
+    
+/.*$/
+    borfle\nbib.gif\nno\n
+
+/.*$/m
+    borfle\nbib.gif\nno\n
+
+/.*$/s
+    borfle\nbib.gif\nno\n
+
+/.*$/ms
+    borfle\nbib.gif\nno\n
+    
+/(.*X|^B)/
+    abcde\n1234Xyz
+    BarFoo 
+    *** Failers
+    abcde\nBar  
+
+/(.*X|^B)/m
+    abcde\n1234Xyz
+    BarFoo 
+    abcde\nBar  
+
+/(.*X|^B)/s
+    abcde\n1234Xyz
+    BarFoo 
+    *** Failers
+    abcde\nBar  
+
+/(.*X|^B)/ms
+    abcde\n1234Xyz
+    BarFoo 
+    abcde\nBar  
+
+/(?s)(.*X|^B)/
+    abcde\n1234Xyz
+    BarFoo 
+    *** Failers 
+    abcde\nBar  
+
+/(?s:.*X|^B)/
+    abcde\n1234Xyz
+    BarFoo 
+    *** Failers 
+    abcde\nBar  
+
+/^.*B/
+    **** Failers
+    abc\nB
+     
+/(?s)^.*B/
+    abc\nB
+
+/(?m)^.*B/
+    abc\nB
+     
+/(?ms)^.*B/
+    abc\nB
+
+/(?ms)^B/
+    abc\nB
+
+/(?s)B$/
+    B\n
+
+/^[0-9][0-9][0-9][0-9][0-9][0-9][0-9][0-9][0-9][0-9][0-9][0-9]/
+    123456654321
+  
+/^\d\d\d\d\d\d\d\d\d\d\d\d/
+    123456654321 
+
+/^[\d][\d][\d][\d][\d][\d][\d][\d][\d][\d][\d][\d]/
+    123456654321
+  
+/^[abc]{12}/
+    abcabcabcabc
+    
+/^[a-c]{12}/
+    abcabcabcabc
+    
+/^(a|b|c){12}/
+    abcabcabcabc 
+
+/^[abcdefghijklmnopqrstuvwxy0123456789]/
+    n
+    *** Failers 
+    z 
+
+/abcde{0,0}/
+    abcd
+    *** Failers
+    abce  
+
+/ab[cd]{0,0}e/
+    abe
+    *** Failers
+    abcde 
+    
+/ab(c){0,0}d/
+    abd
+    *** Failers
+    abcd   
+
+/a(b*)/
+    a
+    ab
+    abbbb
+    *** Failers
+    bbbbb    
+    
+/ab\d{0}e/
+    abe
+    *** Failers
+    ab1e   
+    
+/"([^\\"]+|\\.)*"/
+    the \"quick\" brown fox
+    \"the \\\"quick\\\" brown fox\" 
+
+/.*?/g+
+    abc
+  
+/\b/g+
+    abc 
+
+/\b/+g
+    abc 
+
+//g
+    abc
+
+/]{0,})>]{0,})>([\d]{0,}\.)(.*)((
([\w\W\s\d][^<>]{0,})|[\s]{0,}))<\/a><\/TD>]{0,})>([\w\W\s\d][^<>]{0,})<\/TD>]{0,})>([\w\W\s\d][^<>]{0,})<\/TD><\/TR>/is
+  43.Word Processor
(N-1286)Lega lstaff.comCA - Statewide
+
+/a[^a]b/
+    acb
+    a\nb
+    
+/a.b/
+    acb
+    *** Failers 
+    a\nb   
+    
+/a[^a]b/s
+    acb
+    a\nb  
+    
+/a.b/s
+    acb
+    a\nb  
+
+/^(b+?|a){1,2}?c/
+    bac
+    bbac
+    bbbac
+    bbbbac
+    bbbbbac 
+
+/^(b+|a){1,2}?c/
+    bac
+    bbac
+    bbbac
+    bbbbac
+    bbbbbac 
+    
+/(?!\A)x/m
+    x\nb\n
+    a\bx\n  
+    
+/\x0{ab}/
+    \0{ab} 
+
+/(A|B)*?CD/
+    CD 
+    
+/(A|B)*CD/
+    CD 
+
+/(?.*/)foo"
+    /this/is/a/very/long/line/in/deed/with/very/many/slashes/in/it/you/see/
+
+"(?>.*/)foo"
+    /this/is/a/very/long/line/in/deed/with/very/many/slashes/in/and/foo
+
+/(?>(\.\d\d[1-9]?))\d+/
+    1.230003938
+    1.875000282
+    *** Failers 
+    1.235 
+
+/^((?>\w+)|(?>\s+))*$/
+    now is the time for all good men to come to the aid of the party
+    *** Failers
+    this is not a line with only words and spaces!
+    
+/(\d+)(\w)/
+    12345a
+    12345+ 
+
+/((?>\d+))(\w)/
+    12345a
+    *** Failers
+    12345+ 
+
+/(?>a+)b/
+    aaab
+
+/((?>a+)b)/
+    aaab
+
+/(?>(a+))b/
+    aaab
+
+/(?>b)+/
+    aaabbbccc
+
+/(?>a+|b+|c+)*c/
+    aaabbbbccccd
+    
+/(a+|b+|c+)*c/
+    aaabbbbccccd
+
+/((?>[^()]+)|\([^()]*\))+/
+    ((abc(ade)ufh()()x
+    
+/\(((?>[^()]+)|\([^()]+\))+\)/ 
+    (abc)
+    (abc(def)xyz)
+    *** Failers
+    ((()aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa   
+
+/a(?-i)b/i
+    ab
+    Ab
+    *** Failers 
+    aB
+    AB
+        
+/(a (?x)b c)d e/
+    a bcd e
+    *** Failers
+    a b cd e
+    abcd e   
+    a bcde 
+ 
+/(a b(?x)c d (?-x)e f)/
+    a bcde f
+    *** Failers
+    abcdef  
+
+/(a(?i)b)c/
+    abc
+    aBc
+    *** Failers
+    abC
+    aBC  
+    Abc
+    ABc
+    ABC
+    AbC
+    
+/a(?i:b)c/
+    abc
+    aBc
+    *** Failers 
+    ABC
+    abC
+    aBC
+    
+/a(?i:b)*c/
+    aBc
+    aBBc
+    *** Failers 
+    aBC
+    aBBC
+    
+/a(?=b(?i)c)\w\wd/
+    abcd
+    abCd
+    *** Failers
+    aBCd
+    abcD     
+    
+/(?s-i:more.*than).*million/i
+    more than million
+    more than MILLION
+    more \n than Million 
+    *** Failers
+    MORE THAN MILLION    
+    more \n than \n million 
+
+/(?:(?s-i)more.*than).*million/i
+    more than million
+    more than MILLION
+    more \n than Million 
+    *** Failers
+    MORE THAN MILLION    
+    more \n than \n million 
+    
+/(?>a(?i)b+)+c/ 
+    abc
+    aBbc
+    aBBc 
+    *** Failers
+    Abc
+    abAb    
+    abbC 
+    
+/(?=a(?i)b)\w\wc/
+    abc
+    aBc
+    *** Failers
+    Ab 
+    abC
+    aBC     
+    
+/(?<=a(?i)b)(\w\w)c/
+    abxxc
+    aBxxc
+    *** Failers
+    Abxxc
+    ABxxc
+    abxxC      
+
+/^(?(?=abc)\w{3}:|\d\d)$/
+    abc:
+    12
+    *** Failers
+    123
+    xyz    
+
+/^(?(?!abc)\d\d|\w{3}:)$/
+    abc:
+    12
+    *** Failers
+    123
+    xyz    
+    
+/(?(?<=foo)bar|cat)/
+    foobar
+    cat
+    fcat
+    focat   
+    *** Failers
+    foocat  
+
+/(?(?a*)*/
+    a
+    aa
+    aaaa
+    
+/(abc|)+/
+    abc
+    abcabc
+    abcabcabc
+    xyz      
+
+/([a]*)*/
+    a
+    aaaaa 
+ 
+/([ab]*)*/
+    a
+    b
+    ababab
+    aaaabcde
+    bbbb    
+ 
+/([^a]*)*/
+    b
+    bbbb
+    aaa   
+ 
+/([^ab]*)*/
+    cccc
+    abab  
+ 
+/([a]*?)*/
+    a
+    aaaa 
+ 
+/([ab]*?)*/
+    a
+    b
+    abab
+    baba   
+ 
+/([^a]*?)*/
+    b
+    bbbb
+    aaa   
+ 
+/([^ab]*?)*/
+    c
+    cccc
+    baba   
+ 
+/(?>a*)*/
+    a
+    aaabcde 
+ 
+/((?>a*))*/
+    aaaaa
+    aabbaa 
+ 
+/((?>a*?))*/
+    aaaaa
+    aabbaa 
+
+/(?(?=[^a-z]+[a-z])  \d{2}-[a-z]{3}-\d{2}  |  \d{2}-\d{2}-\d{2} ) /x
+    12-sep-98
+    12-09-98
+    *** Failers
+    sep-12-98
+        
+/(?i:saturday|sunday)/
+    saturday
+    sunday
+    Saturday
+    Sunday
+    SATURDAY
+    SUNDAY
+    SunDay
+    
+/(a(?i)bc|BB)x/
+    abcx
+    aBCx
+    bbx
+    BBx
+    *** Failers
+    abcX
+    aBCX
+    bbX
+    BBX               
+
+/^([ab](?i)[cd]|[ef])/
+    ac
+    aC
+    bD
+    elephant
+    Europe 
+    frog
+    France
+    *** Failers
+    Africa     
+
+/^(ab|a(?i)[b-c](?m-i)d|x(?i)y|z)/
+    ab
+    aBd
+    xy
+    xY
+    zebra
+    Zambesi
+    *** Failers
+    aCD  
+    XY  
+
+/(?<=foo\n)^bar/m
+    foo\nbar
+    *** Failers
+    bar
+    baz\nbar   
+
+/(?<=(?]&/
+    <&OUT
+
+/(?:(f)(o)(o)|(b)(a)(r))*/
+    foobar
+
+/(?<=a)b/
+    ab
+    *** Failers
+    cb
+    b
+
+/(?a+)ab/
+
+/(?>a+)b/
+    aaab
+
+/([[:]+)/
+    a:[b]:
+
+/([[=]+)/
+    a=[b]=
+
+/([[.]+)/
+    a.[b].
+
+/((?>a+)b)/
+    aaab
+
+/(?>(a+))b/
+    aaab
+
+/((?>[^()]+)|\([^()]*\))+/
+    ((abc(ade)ufh()()x
+
+/a\Z/
+    *** Failers
+    aaab
+    a\nb\n
+
+/b\Z/
+    a\nb\n
+
+/b\z/
+
+/b\Z/
+    a\nb
+
+/b\z/
+    a\nb
+    *** Failers
+    
+/(?>.*)(?<=(abcd|wxyz))/
+    alphabetabcd
+    endingwxyz
+    *** Failers
+    a rather long string that doesn't end with one of them
+
+/word (?>(?:(?!otherword)[a-zA-Z0-9]+ ){0,30})otherword/
+    word cat dog elephant mussel cow horse canary baboon snake shark otherword
+    word cat dog elephant mussel cow horse canary baboon snake shark
+  
+/word (?>[a-zA-Z0-9]+ ){0,30}otherword/
+    word cat dog elephant mussel cow horse canary baboon snake shark the quick brown fox and the lazy dog and several other words getting close to thirty by now I hope
+
+/(?<=\d{3}(?!999))foo/
+    999foo
+    123999foo 
+    *** Failers
+    123abcfoo
+    
+/(?<=(?!...999)\d{3})foo/
+    999foo
+    123999foo 
+    *** Failers
+    123abcfoo
+
+/(?<=\d{3}(?!999)...)foo/
+    123abcfoo
+    123456foo 
+    *** Failers
+    123999foo  
+    
+/(?<=\d{3}...)(?Z)+|A)*/
+    ZABCDEFG
+
+/((?>)+|A)*/
+    ZABCDEFG
+
+/a*/g
+    abbab
+
+/^[a-\d]/
+    abcde
+    -things
+    0digit
+    *** Failers
+    bcdef    
+
+/^[\d-a]/
+    abcde
+    -things
+    0digit
+    *** Failers
+    bcdef    
+    
+/[[:space:]]+/
+    > \x09\x0a\x0c\x0d\x0b<
+     
+/[[:blank:]]+/
+    > \x09\x0a\x0c\x0d\x0b<
+     
+/[\s]+/
+    > \x09\x0a\x0c\x0d\x0b<
+     
+/\s+/
+    > \x09\x0a\x0c\x0d\x0b<
+     
+/ab/x
+    ab
+
+/(?!\A)x/m
+  a\nxb\n
+
+/(?!^)x/m
+  a\nxb\n
+
+/abc\Qabc\Eabc/
+    abcabcabc
+    
+/abc\Q(*+|\Eabc/
+    abc(*+|abc 
+
+/   abc\Q abc\Eabc/x
+    abc abcabc
+    *** Failers
+    abcabcabc  
+    
+/abc#comment
+    \Q#not comment
+    literal\E/x
+    abc#not comment\n    literal     
+
+/abc#comment
+    \Q#not comment
+    literal/x
+    abc#not comment\n    literal     
+
+/abc#comment
+    \Q#not comment
+    literal\E #more comment
+    /x
+    abc#not comment\n    literal     
+
+/abc#comment
+    \Q#not comment
+    literal\E #more comment/x
+    abc#not comment\n    literal     
+
+/\Qabc\$xyz\E/
+    abc\\\$xyz
+
+/\Qabc\E\$\Qxyz\E/
+    abc\$xyz
+
+/\Gabc/
+    abc
+    *** Failers
+    xyzabc  
+
+/\Gabc./g
+    abc1abc2xyzabc3
+
+/abc./g
+    abc1abc2xyzabc3 
+
+/a(?x: b c )d/
+    XabcdY
+    *** Failers 
+    Xa b c d Y 
+
+/((?x)x y z | a b c)/
+    XabcY
+    AxyzB 
+
+/(?i)AB(?-i)C/
+    XabCY
+    *** Failers
+    XabcY  
+
+/((?i)AB(?-i)C|D)E/
+    abCE
+    DE
+    *** Failers
+    abcE
+    abCe  
+    dE
+    De    
+
+/[z\Qa-d]\E]/
+    z
+    a
+    -
+    d
+    ] 
+    *** Failers
+    b     
+
+/[\z\C]/
+    z
+    C 
+    
+/\M/
+    M 
+    
+/(a+)*b/
+    aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa 
+    
+/(?i)reg(?:ul(?:[a]|ae)r|ex)/
+    REGular
+    regulaer
+    Regex  
+    regulr 
+
+/[--]+/
+    
+    
+    
+    
+
+/(?<=Z)X./
+    \x84XAZXB
+
+/^(?(2)a|(1)(2))+$/
+    123a
+
+/(?<=a|bbbb)c/
+    ac
+    bbbbc
+
+/abc/SS>testsavedregex
+testsavedregex
+testsavedregex
+testsavedregex
+
+    xyz\r\nabc\
+    xyz\rabc\
+    xyz\r\nabc\
+    ** Failers 
+    xyz\nabc\
+    xyz\r\nabc\
+    xyz\nabc\
+    xyz\rabc\
+    xyz\rabc\
+    
+/abc$/m
+    xyzabc
+    xyzabc\n 
+    xyzabc\npqr 
+    xyzabc\r\ 
+    xyzabc\rpqr\ 
+    xyzabc\r\n\ 
+    xyzabc\r\npqr\ 
+    ** Failers
+    xyzabc\r 
+    xyzabc\rpqr 
+    xyzabc\r\n 
+    xyzabc\r\npqr 
+    
+/^abc/m
+    xyz\rabcdef
+    xyz\nabcdef\
+    ** Failers  
+    xyz\nabcdef
+       
+/^abc/m
+    xyz\nabcdef
+    xyz\rabcdef\
+    ** Failers  
+    xyz\rabcdef
+       
+/^abc/m
+    xyz\r\nabcdef
+    xyz\rabcdef\
+    ** Failers  
+    xyz\rabcdef
+    
+/.*/
+    abc\ndef
+    abc\rdef
+    abc\r\ndef
+    \abc\ndef
+    \abc\rdef
+    \abc\r\ndef
+    \abc\ndef
+    \abc\rdef
+    \abc\r\ndef
+
+/\w+(.)(.)?def/s
+    abc\ndef
+    abc\rdef
+    abc\r\ndef
+
+/^\w+=.*(\\\n.*)*/
+    abc=xyz\\\npqr
+
+/^(a()*)*/
+    aaaa
+
+/^(?:a(?:(?:))*)*/
+    aaaa
+
+/^(a()+)+/
+    aaaa
+
+/^(?:a(?:(?:))+)+/
+    aaaa
+
+/(a|)*\d/
+  aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+  aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa4
+
+/(?>a|)*\d/
+  aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+  aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa4
+
+/(?:a|)*\d/
+  aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+  aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa4
+
+/^a.b/
+    a\rb
+    a\nb\ 
+    ** Failers
+    a\nb
+    a\nb\
+    a\rb\   
+    a\rb\   
+
+/^abc./mgx
+    abc1 \x0aabc2 \x0babc3xx \x0cabc4 \x0dabc5xx \x0d\x0aabc6 \x85abc7 JUNK
+
+/abc.$/mgx
+    abc1\x0a abc2\x0b abc3\x0c abc4\x0d abc5\x0d\x0a abc6\x85 abc9
+
+/^a\Rb/
+    a\nb
+    a\rb
+    a\r\nb
+    a\x0bb
+    a\x0cb
+    a\x85b   
+    ** Failers
+    a\n\rb    
+
+/^a\R*b/
+    ab
+    a\nb
+    a\rb
+    a\r\nb
+    a\x0bb
+    a\x0cb
+    a\x85b   
+    a\n\rb    
+    a\n\r\x85\x0cb 
+
+/^a\R+b/
+    a\nb
+    a\rb
+    a\r\nb
+    a\x0bb
+    a\x0cb
+    a\x85b   
+    a\n\rb    
+    a\n\r\x85\x0cb 
+    ** Failers
+    ab  
+    
+/^a\R{1,3}b/
+    a\nb
+    a\n\rb
+    a\n\r\x85b
+    a\r\n\r\nb 
+    a\r\n\r\n\r\nb 
+    a\n\r\n\rb
+    a\n\n\r\nb 
+    ** Failers
+    a\n\n\n\rb
+    a\r
+
+/^a[\R]b/
+    aRb
+    ** Failers
+    a\nb  
+
+/.+foo/
+    afoo
+    ** Failers 
+    \r\nfoo 
+    \nfoo 
+
+/.+foo/
+    afoo
+    \nfoo 
+    ** Failers 
+    \r\nfoo 
+
+/.+foo/
+    afoo
+    ** Failers 
+    \nfoo 
+    \r\nfoo 
+
+/.+foo/s
+    afoo
+    \r\nfoo 
+    \nfoo 
+
+/^$/mg
+    abc\r\rxyz
+    abc\n\rxyz  
+    ** Failers 
+    abc\r\nxyz
+
+/^X/m
+    XABC
+    ** Failers 
+    XABC\B
+
+/(?m)^$/g+
+    abc\r\n\r\n
+
+/(?m)^$|^\r\n/g+ 
+    abc\r\n\r\n
+    
+/(?m)$/g+ 
+    abc\r\n\r\n
+
+/(?|(abc)|(xyz))/
+   >abc<
+   >xyz< 
+
+/(x)(?|(abc)|(xyz))(x)/
+    xabcx
+    xxyzx 
+
+/(x)(?|(abc)(pqr)|(xyz))(x)/
+    xabcpqrx
+    xxyzx 
+
+/(?|(abc)|(xyz))(?1)/
+    abcabc
+    xyzabc 
+    ** Failers 
+    xyzxyz 
+ 
+/\H\h\V\v/
+    X X\x0a
+    X\x09X\x0b
+    ** Failers
+    \xa0 X\x0a   
+    
+/\H*\h+\V?\v{3,4}/ 
+    \x09\x20\xa0X\x0a\x0b\x0c\x0d\x0a
+    \x09\x20\xa0\x0a\x0b\x0c\x0d\x0a
+    \x09\x20\xa0\x0a\x0b\x0c
+    ** Failers 
+    \x09\x20\xa0\x0a\x0b
+     
+/\H{3,4}/
+    XY  ABCDE
+    XY  PQR ST 
+    
+/.\h{3,4}./
+    XY  AB    PQRS
+
+/\h*X\h?\H+Y\H?Z/
+    >XNNNYZ
+    >  X NYQZ
+    ** Failers
+    >XYZ   
+    >  X NY Z
+
+/\v*X\v?Y\v+Z\V*\x0a\V+\x0b\V{2,3}\x0c/
+    >XY\x0aZ\x0aA\x0bNN\x0c
+    >\x0a\x0dX\x0aY\x0a\x0bZZZ\x0aAAA\x0bNNN\x0c
+
+/.+A/
+    \r\nA
+    
+/\nA/
+    \r\nA 
+
+/[\r\n]A/
+    \r\nA 
+
+/(\r|\n)A/
+    \r\nA 
+
+/a\Rb/I
+    a\rb
+    a\nb
+    a\r\nb
+    ** Failers
+    a\x85b
+    a\x0bb     
+
+/a\Rb/I
+    a\rb
+    a\nb
+    a\r\nb
+    a\x85b
+    a\x0bb     
+    ** Failers 
+    a\x85b\
+    a\x0bb\
+    
+/a\R?b/I
+    a\rb
+    a\nb
+    a\r\nb
+    ** Failers
+    a\x85b
+    a\x0bb     
+
+/a\R?b/I
+    a\rb
+    a\nb
+    a\r\nb
+    a\x85b
+    a\x0bb     
+    ** Failers 
+    a\x85b\
+    a\x0bb\
+    
+/a\R{2,4}b/I
+    a\r\n\nb
+    a\n\r\rb
+    a\r\n\r\n\r\n\r\nb
+    ** Failers
+    a\x85\85b
+    a\x0b\0bb     
+
+/a\R{2,4}b/I
+    a\r\rb
+    a\n\n\nb
+    a\r\n\n\r\rb
+    a\x85\85b
+    a\x0b\0bb     
+    ** Failers 
+    a\r\r\r\r\rb 
+    a\x85\85b\
+    a\x0b\0bb\
+    
+/a(?!)|\wbc/
+    abc 
+
+/a[]b/
+    ** Failers
+    ab
+
+/a[]+b/
+    ** Failers
+    ab 
+
+/a[]*+b/
+    ** Failers
+    ab 
+
+/a[^]b/
+    aXb
+    a\nb 
+    ** Failers
+    ab  
+    
+/a[^]+b/ 
+    aXb
+    a\nX\nXb 
+    ** Failers
+    ab  
+
+/X$/E
+    X
+    ** Failers 
+    X\n 
+
+/X$/
+    X
+    X\n 
+
+/xyz/C
+  xyz 
+  abcxyz 
+  abcxyz\Y
+  ** Failers 
+  abc
+  abc\Y
+  abcxypqr  
+  abcxypqr\Y  
+
+/(*NO_START_OPT)xyz/C
+  abcxyz 
+  
+/(?C)ab/
+  ab
+  \C-ab
+  
+/ab/C
+  ab
+  \C-ab    
+
+/^"((?(?=[a])[^"])|b)*"$/C
+    "ab"
+    \C-"ab"
+
+/\d+X|9+Y/
+    ++++123999\P
+    ++++123999Y\P
+
+/Z(*F)/
+    Z\P
+    ZA\P 
+    
+/Z(?!)/
+    Z\P 
+    ZA\P 
+
+/dog(sbody)?/
+    dogs\P
+    dogs\P\P 
+    
+/dog(sbody)??/
+    dogs\P
+    dogs\P\P 
+
+/dog|dogsbody/
+    dogs\P
+    dogs\P\P 
+ 
+/dogsbody|dog/
+    dogs\P
+    dogs\P\P 
+
+/Z(*F)Q|ZXY/
+    Z\P
+    ZA\P 
+    X\P 
+
+/\bthe cat\b/
+    the cat\P
+    the cat\P\P
+
+/dog(sbody)?/
+    dogs\D\P
+    body\D\R
+
+/dog(sbody)?/
+    dogs\D\P\P
+    body\D\R
+
+/abc/
+   abc\P
+   abc\P\P
+
+/abc\K123/
+    xyzabc123pqr
+    
+/(?<=abc)123/
+    xyzabc123pqr 
+    xyzabc12\P
+    xyzabc12\P\P
+
+/\babc\b/
+    +++abc+++
+    +++ab\P
+    +++ab\P\P  
+
+/(?=C)/g+
+    ABCDECBA
+
+/(abc|def|xyz)/I
+    terhjk;abcdaadsfe
+    the quick xyz brown fox 
+    \Yterhjk;abcdaadsfe
+    \Ythe quick xyz brown fox 
+    ** Failers
+    thejk;adlfj aenjl;fda asdfasd ehj;kjxyasiupd
+    \Ythejk;adlfj aenjl;fda asdfasd ehj;kjxyasiupd
+
+/(abc|def|xyz)/SI
+    terhjk;abcdaadsfe
+    the quick xyz brown fox 
+    \Yterhjk;abcdaadsfe
+    \Ythe quick xyz brown fox 
+    ** Failers
+    thejk;adlfj aenjl;fda asdfasd ehj;kjxyasiupd
+    \Ythejk;adlfj aenjl;fda asdfasd ehj;kjxyasiupd
+
+/abcd*/+
+    xxxxabcd\P
+    xxxxabcd\P\P
+    dddxxx\R 
+    xxxxabcd\P\P
+    xxx\R 
+
+/abcd*/i
+    xxxxabcd\P
+    xxxxabcd\P\P
+    XXXXABCD\P
+    XXXXABCD\P\P
+
+/abc\d*/
+    xxxxabc1\P
+    xxxxabc1\P\P
+
+/abc[de]*/
+    xxxxabcde\P
+    xxxxabcde\P\P
+
+/(?:(?1)|B)(A(*F)|C)/
+    ABCD
+    CCD
+    ** Failers
+    CAD   
+
+/^(?:(?1)|B)(A(*F)|C)/
+    CCD
+    BCD 
+    ** Failers
+    ABCD
+    CAD
+    BAD    
+
+/^(?!a(*SKIP)b)/
+    ac
+    
+/^(?=a(*SKIP)b|ac)/
+    ** Failers
+    ac
+    
+/^(?=a(*THEN)b|ac)/
+    ac
+    
+/^(?=a(*PRUNE)b)/
+    ab  
+    ** Failers 
+    ac
+
+/^(?(?!a(*SKIP)b))/
+    ac
+
+/(?<=abc)def/
+    abc\P\P
+
+/abc$/
+    abc
+    abc\P
+    abc\P\P
+
+/abc$/m
+    abc
+    abc\n
+    abc\P\P
+    abc\n\P\P 
+    abc\P
+    abc\n\P
+
+/abc\z/
+    abc
+    abc\P
+    abc\P\P
+
+/abc\Z/
+    abc
+    abc\P
+    abc\P\P
+
+/abc\b/
+    abc
+    abc\P
+    abc\P\P
+
+/abc\B/
+    abc
+    abc\P
+    abc\P\P
+
+/.+/
+    abc\>0
+    abc\>1
+    abc\>2
+    abc\>3
+    abc\>4
+    abc\>-4 
+
+/^(?:a)++\w/
+     aaaab
+     ** Failers 
+     aaaa 
+     bbb 
+
+/^(?:aa|(?:a)++\w)/
+     aaaab
+     aaaa 
+     ** Failers 
+     bbb 
+
+/^(?:a)*+\w/
+     aaaab
+     bbb 
+     ** Failers 
+     aaaa 
+
+/^(a)++\w/
+     aaaab
+     ** Failers 
+     aaaa 
+     bbb 
+
+/^(a|)++\w/
+     aaaab
+     ** Failers 
+     aaaa 
+     bbb 
+
+/(?=abc){3}abc/+
+    abcabcabc
+    ** Failers
+    xyz  
+    
+/(?=abc)+abc/+
+    abcabcabc
+    ** Failers
+    xyz  
+    
+/(?=abc)++abc/+
+    abcabcabc
+    ** Failers
+    xyz  
+    
+/(?=abc){0}xyz/
+    xyz 
+
+/(?=abc){1}xyz/
+    ** Failers
+    xyz 
+    
+/(?=(a))?./
+    ab
+    bc
+      
+/(?=(a))??./
+    ab
+    bc
+
+/^(?=(a)){0}b(?1)/
+    backgammon
+
+/^(?=(?1))?[az]([abc])d/
+    abd 
+    zcdxx 
+
+/^(?!a){0}\w+/
+    aaaaa
+
+/(?<=(abc))?xyz/
+    abcxyz
+    pqrxyz 
+
+/((?2))((?1))/
+    abc
+
+/(?(R)a+|(?R)b)/
+    aaaabcde
+
+/(?(R)a+|((?R))b)/
+    aaaabcde
+
+/((?(R)a+|(?1)b))/
+    aaaabcde
+
+/((?(R2)a+|(?1)b))/
+    aaaabcde
+
+/(?(R)a*(?1)|((?R))b)/
+    aaaabcde
+
+/(a+)/
+    \O6aaaa
+    \O8aaaa
+
+/ab\Cde/
+    abXde
+    
+/(?<=ab\Cde)X/
+    abZdeX
+
+/^\R/
+    \r\P
+    \r\P\P
+    
+/^\R{2,3}x/
+    \r\P
+    \r\P\P
+    \r\r\P 
+    \r\r\P\P
+    \r\r\r\P  
+    \r\r\r\P\P
+    \r\rx
+    \r\r\rx    
+
+/^\R{2,3}?x/
+    \r\P
+    \r\P\P
+    \r\r\P 
+    \r\r\P\P
+    \r\r\r\P  
+    \r\r\r\P\P
+    \r\rx
+    \r\r\rx    
+    
+/^\R?x/
+    \r\P
+    \r\P\P 
+    x
+    \rx  
+
+/^\R+x/
+    \r\P
+    \r\P\P 
+    \r\n\P
+    \r\n\P\P  
+    \rx  
+
+/^a$/
+    a\r\P
+    a\r\P\P 
+
+/^a$/m
+    a\r\P
+    a\r\P\P 
+
+/^(a$|a\r)/
+    a\r\P
+    a\r\P\P 
+
+/^(a$|a\r)/m
+    a\r\P
+    a\r\P\P 
+
+/./
+    \r\P
+    \r\P\P 
+  
+/.{2,3}/
+    \r\P 
+    \r\P\P
+    \r\r\P
+    \r\r\P\P
+    \r\r\r\P
+    \r\r\r\P\P     
+
+/.{2,3}?/
+    \r\P 
+    \r\P\P
+    \r\r\P
+    \r\r\P\P
+    \r\r\r\P
+    \r\r\r\P\P     
+
+/-- Test simple validity check for restarts --/
+
+/abcdef/
+   abc\R
+
+/)(.)|(?R))++)*F>/
+    text  text xxxxx  text F> text2  more text.
+
+/^(?>.{4})abc|^\w\w.xabcd/
+    xxxxabcd
+    xx\xa0xabcd 
+
+/^(.{4}){2}+abc|^\w\w.x\w\w\w\wabcd/
+    xxxxxxxxabcd
+    xx\xa0xxxxxabcd 
+
+/abcd/
+    abcd\O0
+
+/-- End of testinput8 --/
diff --git a/pcre/testdata/testinput9 b/pcre/testdata/testinput9
new file mode 100644
index 00000000000..766c9c0fb6f
--- /dev/null
+++ b/pcre/testdata/testinput9
@@ -0,0 +1,717 @@
+/-- This set of tests checks UTF-8 support with the DFA matching functionality
+    of pcre_dfa_exec(). The -dfa flag must be used with pcretest when running 
+    it. --/
+
+/\x{100}ab/8
+  \x{100}ab
+  
+/a\x{100}*b/8
+    ab
+    a\x{100}b  
+    a\x{100}\x{100}b  
+    
+/a\x{100}+b/8
+    a\x{100}b  
+    a\x{100}\x{100}b  
+    *** Failers 
+    ab
+     
+/\bX/8
+    Xoanon
+    +Xoanon
+    \x{300}Xoanon 
+    *** Failers 
+    YXoanon  
+    
+/\BX/8
+    YXoanon
+    *** Failers
+    Xoanon
+    +Xoanon    
+    \x{300}Xoanon 
+
+/X\b/8
+    X+oanon
+    ZX\x{300}oanon 
+    FAX 
+    *** Failers 
+    Xoanon  
+    
+/X\B/8
+    Xoanon  
+    *** Failers
+    X+oanon
+    ZX\x{300}oanon 
+    FAX 
+    
+/[^a]/8
+    abcd
+    a\x{100}   
+
+/^[abc\x{123}\x{400}-\x{402}]{2,3}\d/8
+    ab99
+    \x{123}\x{123}45
+    \x{400}\x{401}\x{402}6  
+    *** Failers
+    d99
+    \x{123}\x{122}4   
+    \x{400}\x{403}6  
+    \x{400}\x{401}\x{402}\x{402}6  
+
+/a.b/8
+    acb
+    a\x7fb
+    a\x{100}b 
+    *** Failers
+    a\nb  
+
+/a(.{3})b/8
+    a\x{4000}xyb 
+    a\x{4000}\x7fyb 
+    a\x{4000}\x{100}yb 
+    *** Failers
+    a\x{4000}b 
+    ac\ncb 
+
+/a(.*?)(.)/
+    a\xc0\x88b
+
+/a(.*?)(.)/8
+    a\x{100}b
+
+/a(.*)(.)/
+    a\xc0\x88b
+
+/a(.*)(.)/8
+    a\x{100}b
+
+/a(.)(.)/
+    a\xc0\x92bcd
+
+/a(.)(.)/8
+    a\x{240}bcd
+
+/a(.?)(.)/
+    a\xc0\x92bcd
+
+/a(.?)(.)/8
+    a\x{240}bcd
+
+/a(.??)(.)/
+    a\xc0\x92bcd
+
+/a(.??)(.)/8
+    a\x{240}bcd
+
+/a(.{3})b/8
+    a\x{1234}xyb 
+    a\x{1234}\x{4321}yb 
+    a\x{1234}\x{4321}\x{3412}b 
+    *** Failers
+    a\x{1234}b 
+    ac\ncb 
+
+/a(.{3,})b/8
+    a\x{1234}xyb 
+    a\x{1234}\x{4321}yb 
+    a\x{1234}\x{4321}\x{3412}b 
+    axxxxbcdefghijb 
+    a\x{1234}\x{4321}\x{3412}\x{3421}b 
+    *** Failers
+    a\x{1234}b 
+
+/a(.{3,}?)b/8
+    a\x{1234}xyb 
+    a\x{1234}\x{4321}yb 
+    a\x{1234}\x{4321}\x{3412}b 
+    axxxxbcdefghijb 
+    a\x{1234}\x{4321}\x{3412}\x{3421}b 
+    *** Failers
+    a\x{1234}b 
+
+/a(.{3,5})b/8
+    a\x{1234}xyb 
+    a\x{1234}\x{4321}yb 
+    a\x{1234}\x{4321}\x{3412}b 
+    axxxxbcdefghijb 
+    a\x{1234}\x{4321}\x{3412}\x{3421}b 
+    axbxxbcdefghijb 
+    axxxxxbcdefghijb 
+    *** Failers
+    a\x{1234}b 
+    axxxxxxbcdefghijb 
+
+/a(.{3,5}?)b/8
+    a\x{1234}xyb 
+    a\x{1234}\x{4321}yb 
+    a\x{1234}\x{4321}\x{3412}b 
+    axxxxbcdefghijb 
+    a\x{1234}\x{4321}\x{3412}\x{3421}b 
+    axbxxbcdefghijb 
+    axxxxxbcdefghijb 
+    *** Failers
+    a\x{1234}b 
+    axxxxxxbcdefghijb 
+
+/^[a\x{c0}]/8
+    *** Failers
+    \x{100}
+
+/(?<=aXb)cd/8
+    aXbcd
+
+/(?<=a\x{100}b)cd/8
+    a\x{100}bcd
+
+/(?<=a\x{100000}b)cd/8
+    a\x{100000}bcd
+    
+/(?:\x{100}){3}b/8
+    \x{100}\x{100}\x{100}b
+    *** Failers 
+    \x{100}\x{100}b
+
+/\x{ab}/8
+    \x{ab} 
+    \xc2\xab
+    *** Failers 
+    \x00{ab}
+
+/(?<=(.))X/8
+    WXYZ
+    \x{256}XYZ 
+    *** Failers
+    XYZ 
+
+/[^a]+/8g
+    bcd
+    \x{100}aY\x{256}Z 
+    
+/^[^a]{2}/8
+    \x{100}bc
+ 
+/^[^a]{2,}/8
+    \x{100}bcAa
+
+/^[^a]{2,}?/8
+    \x{100}bca
+
+/[^a]+/8ig
+    bcd
+    \x{100}aY\x{256}Z 
+    
+/^[^a]{2}/8i
+    \x{100}bc
+ 
+/^[^a]{2,}/8i
+    \x{100}bcAa
+
+/^[^a]{2,}?/8i
+    \x{100}bca
+
+/\x{100}{0,0}/8
+    abcd
+ 
+/\x{100}?/8
+    abcd
+    \x{100}\x{100} 
+
+/\x{100}{0,3}/8 
+    \x{100}\x{100} 
+    \x{100}\x{100}\x{100}\x{100} 
+    
+/\x{100}*/8
+    abce
+    \x{100}\x{100}\x{100}\x{100} 
+
+/\x{100}{1,1}/8
+    abcd\x{100}\x{100}\x{100}\x{100} 
+
+/\x{100}{1,3}/8
+    abcd\x{100}\x{100}\x{100}\x{100} 
+
+/\x{100}+/8
+    abcd\x{100}\x{100}\x{100}\x{100} 
+
+/\x{100}{3}/8
+    abcd\x{100}\x{100}\x{100}XX
+
+/\x{100}{3,5}/8
+    abcd\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}XX
+
+/\x{100}{3,}/8
+    abcd\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}XX
+
+/(?<=a\x{100}{2}b)X/8
+    Xyyya\x{100}\x{100}bXzzz
+
+/\D*/8
+  aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+
+/\D*/8
+  \x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}
+
+/\D/8
+    1X2
+    1\x{100}2 
+  
+/>\S/8
+    > >X Y
+    > >\x{100} Y
+  
+/\d/8
+    \x{100}3
+    
+/\s/8
+    \x{100} X
+    
+/\D+/8
+    12abcd34
+    *** Failers
+    1234  
+
+/\D{2,3}/8
+    12abcd34
+    12ab34
+    *** Failers  
+    1234
+    12a34  
+
+/\D{2,3}?/8
+    12abcd34
+    12ab34
+    *** Failers  
+    1234
+    12a34  
+
+/\d+/8
+    12abcd34
+    *** Failers
+
+/\d{2,3}/8
+    12abcd34
+    1234abcd
+    *** Failers  
+    1.4 
+
+/\d{2,3}?/8
+    12abcd34
+    1234abcd
+    *** Failers  
+    1.4 
+
+/\S+/8
+    12abcd34
+    *** Failers
+    \    \ 
+
+/\S{2,3}/8
+    12abcd34
+    1234abcd
+    *** Failers
+    \     \  
+
+/\S{2,3}?/8
+    12abcd34
+    1234abcd
+    *** Failers
+    \     \  
+
+/>\s+      <34
+    *** Failers
+
+/>\s{2,3}         \s{2,3}?         \xff<
+
+/[\xff]/8
+    >\x{ff}<
+
+/[^\xFF]/
+    XYZ
+
+/[^\xff]/8
+    XYZ
+    \x{123} 
+
+/^[ac]*b/8
+  xb
+
+/^[ac\x{100}]*b/8
+  xb
+
+/^[^x]*b/8i
+  xb
+
+/^[^x]*b/8
+  xb
+  
+/^\d*b/8
+  xb 
+
+/(|a)/g8
+    catac
+    a\x{256}a 
+
+/^\x{85}$/8i
+    \x{85}
+
+/^abc./mgx8
+    abc1 \x0aabc2 \x0babc3xx \x0cabc4 \x0dabc5xx \x0d\x0aabc6 \x{0085}abc7 \x{2028}abc8 \x{2029}abc9 JUNK
+
+/abc.$/mgx8
+    abc1\x0a abc2\x0b abc3\x0c abc4\x0d abc5\x0d\x0a abc6\x{0085} abc7\x{2028} abc8\x{2029} abc9
+
+/^a\Rb/8
+    a\nb
+    a\rb
+    a\r\nb
+    a\x0bb
+    a\x0cb
+    a\x{85}b   
+    a\x{2028}b 
+    a\x{2029}b 
+    ** Failers
+    a\n\rb    
+
+/^a\R*b/8
+    ab
+    a\nb
+    a\rb
+    a\r\nb
+    a\x0bb
+    a\x0c\x{2028}\x{2029}b
+    a\x{85}b   
+    a\n\rb    
+    a\n\r\x{85}\x0cb 
+
+/^a\R+b/8
+    a\nb
+    a\rb
+    a\r\nb
+    a\x0bb
+    a\x0c\x{2028}\x{2029}b
+    a\x{85}b   
+    a\n\rb    
+    a\n\r\x{85}\x0cb 
+    ** Failers
+    ab  
+
+/^a\R{1,3}b/8
+    a\nb
+    a\n\rb
+    a\n\r\x{85}b
+    a\r\n\r\nb 
+    a\r\n\r\n\r\nb 
+    a\n\r\n\rb
+    a\n\n\r\nb 
+    ** Failers
+    a\n\n\n\rb
+    a\r
+
+/\h+\V?\v{3,4}/8 
+    \x09\x20\x{a0}X\x0a\x0b\x0c\x0d\x0a
+
+/\V?\v{3,4}/8 
+    \x20\x{a0}X\x0a\x0b\x0c\x0d\x0a
+
+/\h+\V?\v{3,4}/8
+    >\x09\x20\x{a0}X\x0a\x0a\x0a<
+
+/\V?\v{3,4}/8
+    >\x09\x20\x{a0}X\x0a\x0a\x0a<
+
+/\H\h\V\v/8
+    X X\x0a
+    X\x09X\x0b
+    ** Failers
+    \x{a0} X\x0a   
+    
+/\H*\h+\V?\v{3,4}/8 
+    \x09\x20\x{a0}X\x0a\x0b\x0c\x0d\x0a
+    \x09\x20\x{a0}\x0a\x0b\x0c\x0d\x0a
+    \x09\x20\x{a0}\x0a\x0b\x0c
+    ** Failers 
+    \x09\x20\x{a0}\x0a\x0b
+     
+/\H\h\V\v/8
+    \x{3001}\x{3000}\x{2030}\x{2028}
+    X\x{180e}X\x{85}
+    ** Failers
+    \x{2009} X\x0a   
+    
+/\H*\h+\V?\v{3,4}/8 
+    \x{1680}\x{180e}\x{2007}X\x{2028}\x{2029}\x0c\x0d\x0a
+    \x09\x{205f}\x{a0}\x0a\x{2029}\x0c\x{2028}\x0a
+    \x09\x20\x{202f}\x0a\x0b\x0c
+    ** Failers 
+    \x09\x{200a}\x{a0}\x{2028}\x0b
+     
+/a\Rb/I8
+    a\rb
+    a\nb
+    a\r\nb
+    ** Failers
+    a\x{85}b
+    a\x0bb     
+
+/a\Rb/I8
+    a\rb
+    a\nb
+    a\r\nb
+    a\x{85}b
+    a\x0bb     
+    ** Failers 
+    a\x{85}b\
+    a\x0bb\
+    
+/a\R?b/I8
+    a\rb
+    a\nb
+    a\r\nb
+    ** Failers
+    a\x{85}b
+    a\x0bb     
+
+/a\R?b/I8
+    a\rb
+    a\nb
+    a\r\nb
+    a\x{85}b
+    a\x0bb     
+    ** Failers 
+    a\x{85}b\
+    a\x0bb\
+ 
+/X/8f 
+    A\x{1ec5}ABCXYZ
+
+/abcd*/8
+    xxxxabcd\P
+    xxxxabcd\P\P
+
+/abcd*/i8
+    xxxxabcd\P
+    xxxxabcd\P\P
+    XXXXABCD\P
+    XXXXABCD\P\P
+
+/abc\d*/8
+    xxxxabc1\P
+    xxxxabc1\P\P
+
+/abc[de]*/8
+    xxxxabcde\P
+    xxxxabcde\P\P
+
+/\bthe cat\b/8
+    the cat\P
+    the cat\P\P
+
+/ab\Cde/8
+    abXde
+
+/(?<=ab\Cde)X/8
+
+/./8
+    \r\P
+    \r\P\P 
+  
+/.{2,3}/8
+    \r\P 
+    \r\P\P
+    \r\r\P
+    \r\r\P\P
+    \r\r\r\P
+    \r\r\r\P\P     
+
+/.{2,3}?/8
+    \r\P 
+    \r\P\P
+    \r\r\P
+    \r\r\P\P
+    \r\r\r\P
+    \r\r\r\P\P     
+
+/[^\x{100}]/8
+    \x{100}\x{101}X
+
+/[^\x{100}]+/8
+    \x{100}\x{101}X
+
+/-- End of testinput9 --/ 
diff --git a/pcre/testdata/testinputEBC b/pcre/testdata/testinputEBC
new file mode 100644
index 00000000000..56efcd00aac
--- /dev/null
+++ b/pcre/testdata/testinputEBC
@@ -0,0 +1,121 @@
+/-- This is a specialized test for checking, when PCRE is compiled with the
+EBCDIC option but in an ASCII environment, that newline and white space
+functionality is working. It catches cases where explicit values such as 0x0a
+have been used instead of names like CHAR_LF. Needless to say, it is not a
+genuine EBCDIC test! In patterns, alphabetic characters that follow a backslash
+must be in EBCDIC code. In data, newlines and other spacing characters must be
+in EBCDIC, but can be specified as escapes. --/
+
+/-- Test default newline and variations --/
+
+/^A/m
+    ABC
+    12\x15ABC
+
+/^A/m
+    12\x15ABC
+    12\x0dABC
+    12\x0d\x15ABC
+    12\x25ABC
+
+/^A/m
+    12\x15ABC
+    12\x0dABC
+    12\x0d\x15ABC
+    ** Fail
+    12\x25ABC
+
+/-- Test \h --/
+
+/^A\/
+    A B
+
+/-- Test \H --/
+
+/^A\/
+    AB
+    ** Fail
+    A B
+
+/-- Test \R --/
+
+/^A\/
+    A\x15B
+    A\x0dB
+    A\x25B
+    A\x0bB
+    A\x0cB
+    ** Fail
+    A B
+
+/-- Test \v --/
+
+/^A\/
+    A\x15B
+    A\x0dB
+    A\x25B
+    A\x0bB
+    A\x0cB
+    ** Fail
+    A B
+
+/-- Test \V --/
+
+/^A\/
+    A B
+    ** Fail
+    A\x15B
+    A\x0dB
+    A\x25B
+    A\x0bB
+    A\x0cB
+    
+/-- For repeated items, use an atomic group so that the output is the same
+for DFA matching (otherwise it may show multiple matches). --/ 
+
+/-- Test \h+ --/
+
+/^A(?>\+)/
+    A B
+
+/-- Test \H+ --/
+
+/^A(?>\+)/
+    AB
+    ** Fail
+    A B
+
+/-- Test \R+ --/
+
+/^A(?>\+)/
+    A\x15B
+    A\x0dB
+    A\x25B
+    A\x0bB
+    A\x0cB
+    ** Fail
+    A B
+
+/-- Test \v+ --/
+
+/^A(?>\+)/
+    A\x15B
+    A\x0dB
+    A\x25B
+    A\x0bB
+    A\x0cB
+    ** Fail
+    A B
+
+/-- Test \V+ --/
+
+/^A(?>\+)/
+    A B
+    ** Fail
+    A\x15B
+    A\x0dB
+    A\x25B
+    A\x0bB
+    A\x0cB
+
+/-- End --/
diff --git a/pcre/testdata/testoutput1 b/pcre/testdata/testoutput1
new file mode 100644
index 00000000000..3ed635146a4
--- /dev/null
+++ b/pcre/testdata/testoutput1
@@ -0,0 +1,9203 @@
+/-- This set of tests is for features that are compatible with all versions of
+    Perl >= 5.10, in non-UTF-8 mode. It should run clean for both the 8-bit and
+    16-bit PCRE libraries. --/
+
+/the quick brown fox/
+    the quick brown fox
+ 0: the quick brown fox
+    The quick brown FOX
+No match
+    What do you know about the quick brown fox?
+ 0: the quick brown fox
+    What do you know about THE QUICK BROWN FOX?
+No match
+
+/The quick brown fox/i
+    the quick brown fox
+ 0: the quick brown fox
+    The quick brown FOX
+ 0: The quick brown FOX
+    What do you know about the quick brown fox?
+ 0: the quick brown fox
+    What do you know about THE QUICK BROWN FOX?
+ 0: THE QUICK BROWN FOX
+
+/abcd\t\n\r\f\a\e\071\x3b\$\\\?caxyz/
+    abcd\t\n\r\f\a\e9;\$\\?caxyz
+ 0: abcd\x09\x0a\x0d\x0c\x07\x1b9;$\?caxyz
+
+/a*abc?xyz+pqr{3}ab{2,}xy{4,5}pq{0,6}AB{0,}zz/
+    abxyzpqrrrabbxyyyypqAzz
+ 0: abxyzpqrrrabbxyyyypqAzz
+    abxyzpqrrrabbxyyyypqAzz
+ 0: abxyzpqrrrabbxyyyypqAzz
+    aabxyzpqrrrabbxyyyypqAzz
+ 0: aabxyzpqrrrabbxyyyypqAzz
+    aaabxyzpqrrrabbxyyyypqAzz
+ 0: aaabxyzpqrrrabbxyyyypqAzz
+    aaaabxyzpqrrrabbxyyyypqAzz
+ 0: aaaabxyzpqrrrabbxyyyypqAzz
+    abcxyzpqrrrabbxyyyypqAzz
+ 0: abcxyzpqrrrabbxyyyypqAzz
+    aabcxyzpqrrrabbxyyyypqAzz
+ 0: aabcxyzpqrrrabbxyyyypqAzz
+    aaabcxyzpqrrrabbxyyyypAzz
+ 0: aaabcxyzpqrrrabbxyyyypAzz
+    aaabcxyzpqrrrabbxyyyypqAzz
+ 0: aaabcxyzpqrrrabbxyyyypqAzz
+    aaabcxyzpqrrrabbxyyyypqqAzz
+ 0: aaabcxyzpqrrrabbxyyyypqqAzz
+    aaabcxyzpqrrrabbxyyyypqqqAzz
+ 0: aaabcxyzpqrrrabbxyyyypqqqAzz
+    aaabcxyzpqrrrabbxyyyypqqqqAzz
+ 0: aaabcxyzpqrrrabbxyyyypqqqqAzz
+    aaabcxyzpqrrrabbxyyyypqqqqqAzz
+ 0: aaabcxyzpqrrrabbxyyyypqqqqqAzz
+    aaabcxyzpqrrrabbxyyyypqqqqqqAzz
+ 0: aaabcxyzpqrrrabbxyyyypqqqqqqAzz
+    aaaabcxyzpqrrrabbxyyyypqAzz
+ 0: aaaabcxyzpqrrrabbxyyyypqAzz
+    abxyzzpqrrrabbxyyyypqAzz
+ 0: abxyzzpqrrrabbxyyyypqAzz
+    aabxyzzzpqrrrabbxyyyypqAzz
+ 0: aabxyzzzpqrrrabbxyyyypqAzz
+    aaabxyzzzzpqrrrabbxyyyypqAzz
+ 0: aaabxyzzzzpqrrrabbxyyyypqAzz
+    aaaabxyzzzzpqrrrabbxyyyypqAzz
+ 0: aaaabxyzzzzpqrrrabbxyyyypqAzz
+    abcxyzzpqrrrabbxyyyypqAzz
+ 0: abcxyzzpqrrrabbxyyyypqAzz
+    aabcxyzzzpqrrrabbxyyyypqAzz
+ 0: aabcxyzzzpqrrrabbxyyyypqAzz
+    aaabcxyzzzzpqrrrabbxyyyypqAzz
+ 0: aaabcxyzzzzpqrrrabbxyyyypqAzz
+    aaaabcxyzzzzpqrrrabbxyyyypqAzz
+ 0: aaaabcxyzzzzpqrrrabbxyyyypqAzz
+    aaaabcxyzzzzpqrrrabbbxyyyypqAzz
+ 0: aaaabcxyzzzzpqrrrabbbxyyyypqAzz
+    aaaabcxyzzzzpqrrrabbbxyyyyypqAzz
+ 0: aaaabcxyzzzzpqrrrabbbxyyyyypqAzz
+    aaabcxyzpqrrrabbxyyyypABzz
+ 0: aaabcxyzpqrrrabbxyyyypABzz
+    aaabcxyzpqrrrabbxyyyypABBzz
+ 0: aaabcxyzpqrrrabbxyyyypABBzz
+    >>>aaabxyzpqrrrabbxyyyypqAzz
+ 0: aaabxyzpqrrrabbxyyyypqAzz
+    >aaaabxyzpqrrrabbxyyyypqAzz
+ 0: aaaabxyzpqrrrabbxyyyypqAzz
+    >>>>abcxyzpqrrrabbxyyyypqAzz
+ 0: abcxyzpqrrrabbxyyyypqAzz
+    *** Failers
+No match
+    abxyzpqrrabbxyyyypqAzz
+No match
+    abxyzpqrrrrabbxyyyypqAzz
+No match
+    abxyzpqrrrabxyyyypqAzz
+No match
+    aaaabcxyzzzzpqrrrabbbxyyyyyypqAzz
+No match
+    aaaabcxyzzzzpqrrrabbbxyyypqAzz
+No match
+    aaabcxyzpqrrrabbxyyyypqqqqqqqAzz
+No match
+
+/^(abc){1,2}zz/
+    abczz
+ 0: abczz
+ 1: abc
+    abcabczz
+ 0: abcabczz
+ 1: abc
+    *** Failers
+No match
+    zz
+No match
+    abcabcabczz
+No match
+    >>abczz
+No match
+
+/^(b+?|a){1,2}?c/
+    bc
+ 0: bc
+ 1: b
+    bbc
+ 0: bbc
+ 1: b
+    bbbc
+ 0: bbbc
+ 1: bb
+    bac
+ 0: bac
+ 1: a
+    bbac
+ 0: bbac
+ 1: a
+    aac
+ 0: aac
+ 1: a
+    abbbbbbbbbbbc
+ 0: abbbbbbbbbbbc
+ 1: bbbbbbbbbbb
+    bbbbbbbbbbbac
+ 0: bbbbbbbbbbbac
+ 1: a
+    *** Failers
+No match
+    aaac
+No match
+    abbbbbbbbbbbac
+No match
+
+/^(b+|a){1,2}c/
+    bc
+ 0: bc
+ 1: b
+    bbc
+ 0: bbc
+ 1: bb
+    bbbc
+ 0: bbbc
+ 1: bbb
+    bac
+ 0: bac
+ 1: a
+    bbac
+ 0: bbac
+ 1: a
+    aac
+ 0: aac
+ 1: a
+    abbbbbbbbbbbc
+ 0: abbbbbbbbbbbc
+ 1: bbbbbbbbbbb
+    bbbbbbbbbbbac
+ 0: bbbbbbbbbbbac
+ 1: a
+    *** Failers
+No match
+    aaac
+No match
+    abbbbbbbbbbbac
+No match
+
+/^(b+|a){1,2}?bc/
+    bbc
+ 0: bbc
+ 1: b
+
+/^(b*|ba){1,2}?bc/
+    babc
+ 0: babc
+ 1: ba
+    bbabc
+ 0: bbabc
+ 1: ba
+    bababc
+ 0: bababc
+ 1: ba
+    *** Failers
+No match
+    bababbc
+No match
+    babababc
+No match
+
+/^(ba|b*){1,2}?bc/
+    babc
+ 0: babc
+ 1: ba
+    bbabc
+ 0: bbabc
+ 1: ba
+    bababc
+ 0: bababc
+ 1: ba
+    *** Failers
+No match
+    bababbc
+No match
+    babababc
+No match
+
+/^\ca\cA\c[\c{\c:/
+    \x01\x01\e;z
+ 0: \x01\x01\x1b;z
+
+/^[ab\]cde]/
+    athing
+ 0: a
+    bthing
+ 0: b
+    ]thing
+ 0: ]
+    cthing
+ 0: c
+    dthing
+ 0: d
+    ething
+ 0: e
+    *** Failers
+No match
+    fthing
+No match
+    [thing
+No match
+    \\thing
+No match
+
+/^[]cde]/
+    ]thing
+ 0: ]
+    cthing
+ 0: c
+    dthing
+ 0: d
+    ething
+ 0: e
+    *** Failers
+No match
+    athing
+No match
+    fthing
+No match
+
+/^[^ab\]cde]/
+    fthing
+ 0: f
+    [thing
+ 0: [
+    \\thing
+ 0: \
+    *** Failers
+ 0: *
+    athing
+No match
+    bthing
+No match
+    ]thing
+No match
+    cthing
+No match
+    dthing
+No match
+    ething
+No match
+
+/^[^]cde]/
+    athing
+ 0: a
+    fthing
+ 0: f
+    *** Failers
+ 0: *
+    ]thing
+No match
+    cthing
+No match
+    dthing
+No match
+    ething
+No match
+
+/^\/
+    
+ 0: \x81
+
+/^/
+    
+ 0: \xff
+
+/^[0-9]+$/
+    0
+ 0: 0
+    1
+ 0: 1
+    2
+ 0: 2
+    3
+ 0: 3
+    4
+ 0: 4
+    5
+ 0: 5
+    6
+ 0: 6
+    7
+ 0: 7
+    8
+ 0: 8
+    9
+ 0: 9
+    10
+ 0: 10
+    100
+ 0: 100
+    *** Failers
+No match
+    abc
+No match
+
+/^.*nter/
+    enter
+ 0: enter
+    inter
+ 0: inter
+    uponter
+ 0: uponter
+
+/^xxx[0-9]+$/
+    xxx0
+ 0: xxx0
+    xxx1234
+ 0: xxx1234
+    *** Failers
+No match
+    xxx
+No match
+
+/^.+[0-9][0-9][0-9]$/
+    x123
+ 0: x123
+    xx123
+ 0: xx123
+    123456
+ 0: 123456
+    *** Failers
+No match
+    123
+No match
+    x1234
+ 0: x1234
+
+/^.+?[0-9][0-9][0-9]$/
+    x123
+ 0: x123
+    xx123
+ 0: xx123
+    123456
+ 0: 123456
+    *** Failers
+No match
+    123
+No match
+    x1234
+ 0: x1234
+
+/^([^!]+)!(.+)=apquxz\.ixr\.zzz\.ac\.uk$/
+    abc!pqr=apquxz.ixr.zzz.ac.uk
+ 0: abc!pqr=apquxz.ixr.zzz.ac.uk
+ 1: abc
+ 2: pqr
+    *** Failers
+No match
+    !pqr=apquxz.ixr.zzz.ac.uk
+No match
+    abc!=apquxz.ixr.zzz.ac.uk
+No match
+    abc!pqr=apquxz:ixr.zzz.ac.uk
+No match
+    abc!pqr=apquxz.ixr.zzz.ac.ukk
+No match
+
+/:/
+    Well, we need a colon: somewhere
+ 0: :
+    *** Fail if we don't
+No match
+
+/([\da-f:]+)$/i
+    0abc
+ 0: 0abc
+ 1: 0abc
+    abc
+ 0: abc
+ 1: abc
+    fed
+ 0: fed
+ 1: fed
+    E
+ 0: E
+ 1: E
+    ::
+ 0: ::
+ 1: ::
+    5f03:12C0::932e
+ 0: 5f03:12C0::932e
+ 1: 5f03:12C0::932e
+    fed def
+ 0: def
+ 1: def
+    Any old stuff
+ 0: ff
+ 1: ff
+    *** Failers
+No match
+    0zzz
+No match
+    gzzz
+No match
+    fed\x20
+No match
+    Any old rubbish
+No match
+
+/^.*\.(\d{1,3})\.(\d{1,3})\.(\d{1,3})$/
+    .1.2.3
+ 0: .1.2.3
+ 1: 1
+ 2: 2
+ 3: 3
+    A.12.123.0
+ 0: A.12.123.0
+ 1: 12
+ 2: 123
+ 3: 0
+    *** Failers
+No match
+    .1.2.3333
+No match
+    1.2.3
+No match
+    1234.2.3
+No match
+
+/^(\d+)\s+IN\s+SOA\s+(\S+)\s+(\S+)\s*\(\s*$/
+    1 IN SOA non-sp1 non-sp2(
+ 0: 1 IN SOA non-sp1 non-sp2(
+ 1: 1
+ 2: non-sp1
+ 3: non-sp2
+    1    IN    SOA    non-sp1    non-sp2   (
+ 0: 1    IN    SOA    non-sp1    non-sp2   (
+ 1: 1
+ 2: non-sp1
+ 3: non-sp2
+    *** Failers
+No match
+    1IN SOA non-sp1 non-sp2(
+No match
+
+/^[a-zA-Z\d][a-zA-Z\d\-]*(\.[a-zA-Z\d][a-zA-z\d\-]*)*\.$/
+    a.
+ 0: a.
+    Z.
+ 0: Z.
+    2.
+ 0: 2.
+    ab-c.pq-r.
+ 0: ab-c.pq-r.
+ 1: .pq-r
+    sxk.zzz.ac.uk.
+ 0: sxk.zzz.ac.uk.
+ 1: .uk
+    x-.y-.
+ 0: x-.y-.
+ 1: .y-
+    *** Failers
+No match
+    -abc.peq.
+No match
+
+/^\*\.[a-z]([a-z\-\d]*[a-z\d]+)?(\.[a-z]([a-z\-\d]*[a-z\d]+)?)*$/
+    *.a
+ 0: *.a
+    *.b0-a
+ 0: *.b0-a
+ 1: 0-a
+    *.c3-b.c
+ 0: *.c3-b.c
+ 1: 3-b
+ 2: .c
+    *.c-a.b-c
+ 0: *.c-a.b-c
+ 1: -a
+ 2: .b-c
+ 3: -c
+    *** Failers
+No match
+    *.0
+No match
+    *.a-
+No match
+    *.a-b.c-
+No match
+    *.c-a.0-c
+No match
+
+/^(?=ab(de))(abd)(e)/
+    abde
+ 0: abde
+ 1: de
+ 2: abd
+ 3: e
+
+/^(?!(ab)de|x)(abd)(f)/
+    abdf
+ 0: abdf
+ 1: 
+ 2: abd
+ 3: f
+
+/^(?=(ab(cd)))(ab)/
+    abcd
+ 0: ab
+ 1: abcd
+ 2: cd
+ 3: ab
+
+/^[\da-f](\.[\da-f])*$/i
+    a.b.c.d
+ 0: a.b.c.d
+ 1: .d
+    A.B.C.D
+ 0: A.B.C.D
+ 1: .D
+    a.b.c.1.2.3.C
+ 0: a.b.c.1.2.3.C
+ 1: .C
+
+/^\".*\"\s*(;.*)?$/
+    \"1234\"
+ 0: "1234"
+    \"abcd\" ;
+ 0: "abcd" ;
+ 1: ;
+    \"\" ; rhubarb
+ 0: "" ; rhubarb
+ 1: ; rhubarb
+    *** Failers
+No match
+    \"1234\" : things
+No match
+
+/^$/
+    \
+ 0: 
+    *** Failers
+No match
+
+/   ^    a   (?# begins with a)  b\sc (?# then b c) $ (?# then end)/x
+    ab c
+ 0: ab c
+    *** Failers
+No match
+    abc
+No match
+    ab cde
+No match
+
+/(?x)   ^    a   (?# begins with a)  b\sc (?# then b c) $ (?# then end)/
+    ab c
+ 0: ab c
+    *** Failers
+No match
+    abc
+No match
+    ab cde
+No match
+
+/^   a\ b[c ]d       $/x
+    a bcd
+ 0: a bcd
+    a b d
+ 0: a b d
+    *** Failers
+No match
+    abcd
+No match
+    ab d
+No match
+
+/^(a(b(c)))(d(e(f)))(h(i(j)))(k(l(m)))$/
+    abcdefhijklm
+ 0: abcdefhijklm
+ 1: abc
+ 2: bc
+ 3: c
+ 4: def
+ 5: ef
+ 6: f
+ 7: hij
+ 8: ij
+ 9: j
+10: klm
+11: lm
+12: m
+
+/^(?:a(b(c)))(?:d(e(f)))(?:h(i(j)))(?:k(l(m)))$/
+    abcdefhijklm
+ 0: abcdefhijklm
+ 1: bc
+ 2: c
+ 3: ef
+ 4: f
+ 5: ij
+ 6: j
+ 7: lm
+ 8: m
+
+/^[\w][\W][\s][\S][\d][\D][\b][\n][\c]][\022]/
+    a+ Z0+\x08\n\x1d\x12
+ 0: a+ Z0+\x08\x0a\x1d\x12
+
+/^[.^$|()*+?{,}]+/
+    .^\$(*+)|{?,?}
+ 0: .^$(*+)|{?,?}
+
+/^a*\w/
+    z
+ 0: z
+    az
+ 0: az
+    aaaz
+ 0: aaaz
+    a
+ 0: a
+    aa
+ 0: aa
+    aaaa
+ 0: aaaa
+    a+
+ 0: a
+    aa+
+ 0: aa
+
+/^a*?\w/
+    z
+ 0: z
+    az
+ 0: a
+    aaaz
+ 0: a
+    a
+ 0: a
+    aa
+ 0: a
+    aaaa
+ 0: a
+    a+
+ 0: a
+    aa+
+ 0: a
+
+/^a+\w/
+    az
+ 0: az
+    aaaz
+ 0: aaaz
+    aa
+ 0: aa
+    aaaa
+ 0: aaaa
+    aa+
+ 0: aa
+
+/^a+?\w/
+    az
+ 0: az
+    aaaz
+ 0: aa
+    aa
+ 0: aa
+    aaaa
+ 0: aa
+    aa+
+ 0: aa
+
+/^\d{8}\w{2,}/
+    1234567890
+ 0: 1234567890
+    12345678ab
+ 0: 12345678ab
+    12345678__
+ 0: 12345678__
+    *** Failers
+No match
+    1234567
+No match
+
+/^[aeiou\d]{4,5}$/
+    uoie
+ 0: uoie
+    1234
+ 0: 1234
+    12345
+ 0: 12345
+    aaaaa
+ 0: aaaaa
+    *** Failers
+No match
+    123456
+No match
+
+/^[aeiou\d]{4,5}?/
+    uoie
+ 0: uoie
+    1234
+ 0: 1234
+    12345
+ 0: 1234
+    aaaaa
+ 0: aaaa
+    123456
+ 0: 1234
+
+/\A(abc|def)=(\1){2,3}\Z/
+    abc=abcabc
+ 0: abc=abcabc
+ 1: abc
+ 2: abc
+    def=defdefdef
+ 0: def=defdefdef
+ 1: def
+ 2: def
+    *** Failers
+No match
+    abc=defdef
+No match
+
+/^(a)(b)(c)(d)(e)(f)(g)(h)(i)(j)(k)\11*(\3\4)\1(?#)2$/
+    abcdefghijkcda2
+ 0: abcdefghijkcda2
+ 1: a
+ 2: b
+ 3: c
+ 4: d
+ 5: e
+ 6: f
+ 7: g
+ 8: h
+ 9: i
+10: j
+11: k
+12: cd
+    abcdefghijkkkkcda2
+ 0: abcdefghijkkkkcda2
+ 1: a
+ 2: b
+ 3: c
+ 4: d
+ 5: e
+ 6: f
+ 7: g
+ 8: h
+ 9: i
+10: j
+11: k
+12: cd
+
+/(cat(a(ract|tonic)|erpillar)) \1()2(3)/
+    cataract cataract23
+ 0: cataract cataract23
+ 1: cataract
+ 2: aract
+ 3: ract
+ 4: 
+ 5: 3
+    catatonic catatonic23
+ 0: catatonic catatonic23
+ 1: catatonic
+ 2: atonic
+ 3: tonic
+ 4: 
+ 5: 3
+    caterpillar caterpillar23
+ 0: caterpillar caterpillar23
+ 1: caterpillar
+ 2: erpillar
+ 3: 
+ 4: 
+ 5: 3
+
+
+/^From +([^ ]+) +[a-zA-Z][a-zA-Z][a-zA-Z] +[a-zA-Z][a-zA-Z][a-zA-Z] +[0-9]?[0-9] +[0-9][0-9]:[0-9][0-9]/
+    From abcd  Mon Sep 01 12:33:02 1997
+ 0: From abcd  Mon Sep 01 12:33
+ 1: abcd
+
+/^From\s+\S+\s+([a-zA-Z]{3}\s+){2}\d{1,2}\s+\d\d:\d\d/
+    From abcd  Mon Sep 01 12:33:02 1997
+ 0: From abcd  Mon Sep 01 12:33
+ 1: Sep 
+    From abcd  Mon Sep  1 12:33:02 1997
+ 0: From abcd  Mon Sep  1 12:33
+ 1: Sep  
+    *** Failers
+No match
+    From abcd  Sep 01 12:33:02 1997
+No match
+
+/^12.34/s
+    12\n34
+ 0: 12\x0a34
+    12\r34
+ 0: 12\x0d34
+
+/\w+(?=\t)/
+    the quick brown\t fox
+ 0: brown
+
+/foo(?!bar)(.*)/
+    foobar is foolish see?
+ 0: foolish see?
+ 1: lish see?
+
+/(?:(?!foo)...|^.{0,2})bar(.*)/
+    foobar crowbar etc
+ 0: rowbar etc
+ 1:  etc
+    barrel
+ 0: barrel
+ 1: rel
+    2barrel
+ 0: 2barrel
+ 1: rel
+    A barrel
+ 0: A barrel
+ 1: rel
+
+/^(\D*)(?=\d)(?!123)/
+    abc456
+ 0: abc
+ 1: abc
+    *** Failers
+No match
+    abc123
+No match
+
+/^1234(?# test newlines
+  inside)/
+    1234
+ 0: 1234
+
+/^1234 #comment in extended re
+  /x
+    1234
+ 0: 1234
+
+/#rhubarb
+  abcd/x
+    abcd
+ 0: abcd
+
+/^abcd#rhubarb/x
+    abcd
+ 0: abcd
+
+/^(a)\1{2,3}(.)/
+    aaab
+ 0: aaab
+ 1: a
+ 2: b
+    aaaab
+ 0: aaaab
+ 1: a
+ 2: b
+    aaaaab
+ 0: aaaaa
+ 1: a
+ 2: a
+    aaaaaab
+ 0: aaaaa
+ 1: a
+ 2: a
+
+/(?!^)abc/
+    the abc
+ 0: abc
+    *** Failers
+No match
+    abc
+No match
+
+/(?=^)abc/
+    abc
+ 0: abc
+    *** Failers
+No match
+    the abc
+No match
+
+/^[ab]{1,3}(ab*|b)/
+    aabbbbb
+ 0: aabb
+ 1: b
+
+/^[ab]{1,3}?(ab*|b)/
+    aabbbbb
+ 0: aabbbbb
+ 1: abbbbb
+
+/^[ab]{1,3}?(ab*?|b)/
+    aabbbbb
+ 0: aa
+ 1: a
+
+/^[ab]{1,3}(ab*?|b)/
+    aabbbbb
+ 0: aabb
+ 1: b
+
+/  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*                          # optional leading comment
+(?:    (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+)                    # initial word
+(?:  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+)  )* # further okay, if led by a period
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  @  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*    (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                           # initial subdomain
+(?:                                  #
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.                        # if led by a period...
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                     #   ...further okay
+)*
+# address
+|                     #  or
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+)             # one word, optionally followed by....
+(?:
+[^()<>@,;:".\\\[\]\x80-\xff\000-\010\012-\037]  |  # atom and space parts, or...
+\(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)       |  # comments, or...
+
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+# quoted strings
+)*
+<  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*                     # leading <
+(?:  @  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*    (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                           # initial subdomain
+(?:                                  #
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.                        # if led by a period...
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                     #   ...further okay
+)*
+
+(?:  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  ,  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  @  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*    (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                           # initial subdomain
+(?:                                  #
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.                        # if led by a period...
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                     #   ...further okay
+)*
+)* # further okay, if led by comma
+:                                # closing colon
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  )? #       optional route
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+)                    # initial word
+(?:  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+)  )* # further okay, if led by a period
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  @  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*    (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                           # initial subdomain
+(?:                                  #
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.                        # if led by a period...
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                     #   ...further okay
+)*
+#       address spec
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  > #                  trailing >
+# name and address
+)  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*                       # optional trailing comment
+/x
+    Alan Other 
+ 0: Alan Other 
+    
+ 0: user@dom.ain
+    user\@dom.ain
+ 0: user@dom.ain
+    \"A. Other\"  (a comment)
+ 0: "A. Other"  (a comment)
+    A. Other  (a comment)
+ 0:  Other  (a comment)
+    \"/s=user/ou=host/o=place/prmd=uu.yy/admd= /c=gb/\"\@x400-re.lay
+ 0: "/s=user/ou=host/o=place/prmd=uu.yy/admd= /c=gb/"@x400-re.lay
+    A missing angle @,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+# Atom
+|                       #  or
+"                                     # "
+[^\\\x80-\xff\n\015"] *                            #   normal
+(?:  \\ [^\x80-\xff]  [^\\\x80-\xff\n\015"] * )*        #   ( special normal* )*
+"                                     #        "
+# Quoted string
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+\.
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+# Atom
+|                       #  or
+"                                     # "
+[^\\\x80-\xff\n\015"] *                            #   normal
+(?:  \\ [^\x80-\xff]  [^\\\x80-\xff\n\015"] * )*        #   ( special normal* )*
+"                                     #        "
+# Quoted string
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# additional words
+)*
+@
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+\[                            # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*     #    stuff
+\]                           #           ]
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# optional trailing comments
+(?:
+\.
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+\[                            # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*     #    stuff
+\]                           #           ]
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# optional trailing comments
+)*
+# address
+|                             #  or
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+# Atom
+|                       #  or
+"                                     # "
+[^\\\x80-\xff\n\015"] *                            #   normal
+(?:  \\ [^\x80-\xff]  [^\\\x80-\xff\n\015"] * )*        #   ( special normal* )*
+"                                     #        "
+# Quoted string
+)
+# leading word
+[^()<>@,;:".\\\[\]\x80-\xff\000-\010\012-\037] *               # "normal" atoms and or spaces
+(?:
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+|
+"                                     # "
+[^\\\x80-\xff\n\015"] *                            #   normal
+(?:  \\ [^\x80-\xff]  [^\\\x80-\xff\n\015"] * )*        #   ( special normal* )*
+"                                     #        "
+) # "special" comment or quoted string
+[^()<>@,;:".\\\[\]\x80-\xff\000-\010\012-\037] *            #  more "normal"
+)*
+<
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# <
+(?:
+@
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+\[                            # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*     #    stuff
+\]                           #           ]
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# optional trailing comments
+(?:
+\.
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+\[                            # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*     #    stuff
+\]                           #           ]
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# optional trailing comments
+)*
+(?: ,
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+@
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+\[                            # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*     #    stuff
+\]                           #           ]
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# optional trailing comments
+(?:
+\.
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+\[                            # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*     #    stuff
+\]                           #           ]
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# optional trailing comments
+)*
+)*  # additional domains
+:
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# optional trailing comments
+)?     #       optional route
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+# Atom
+|                       #  or
+"                                     # "
+[^\\\x80-\xff\n\015"] *                            #   normal
+(?:  \\ [^\x80-\xff]  [^\\\x80-\xff\n\015"] * )*        #   ( special normal* )*
+"                                     #        "
+# Quoted string
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+\.
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+# Atom
+|                       #  or
+"                                     # "
+[^\\\x80-\xff\n\015"] *                            #   normal
+(?:  \\ [^\x80-\xff]  [^\\\x80-\xff\n\015"] * )*        #   ( special normal* )*
+"                                     #        "
+# Quoted string
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# additional words
+)*
+@
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+\[                            # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*     #    stuff
+\]                           #           ]
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# optional trailing comments
+(?:
+\.
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+\[                            # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*     #    stuff
+\]                           #           ]
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# optional trailing comments
+)*
+#       address spec
+>                    #                 >
+# name and address
+)
+/x
+    Alan Other 
+ 0: Alan Other 
+    
+ 0: user@dom.ain
+    user\@dom.ain
+ 0: user@dom.ain
+    \"A. Other\"  (a comment)
+ 0: "A. Other" 
+    A. Other  (a comment)
+ 0:  Other 
+    \"/s=user/ou=host/o=place/prmd=uu.yy/admd= /c=gb/\"\@x400-re.lay
+ 0: "/s=user/ou=host/o=place/prmd=uu.yy/admd= /c=gb/"@x400-re.lay
+    A missing angle ?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~\x7f\x80\x81\x82\x83\x84\x85\x86\x87\x88\x89\x8a\x8b\x8c\x8d\x8e\x8f\x90\x91\x92\x93\x94\x95\x96\x97\x98\x99\x9a\x9b\x9c\x9d\x9e\x9f\xa0\xa1\xa2\xa3\xa4\xa5\xa6\xa7\xa8\xa9\xaa\xab\xac\xad\xae\xaf\xb0\xb1\xb2\xb3\xb4\xb5\xb6\xb7\xb8\xb9\xba\xbb\xbc\xbd\xbe\xbf\xc0\xc1\xc2\xc3\xc4\xc5\xc6\xc7\xc8\xc9\xca\xcb\xcc\xcd\xce\xcf\xd0\xd1\xd2\xd3\xd4\xd5\xd6\xd7\xd8\xd9\xda\xdb\xdc\xdd\xde\xdf\xe0\xe1\xe2\xe3\xe4\xe5\xe6\xe7\xe8\xe9\xea\xeb\xec\xed\xee\xef\xf0\xf1\xf2\xf3\xf4\xf5\xf6\xf7\xf8\xf9\xfa\xfb\xfc\xfd\xfe\xff
+
+/P[^*]TAIRE[^*]{1,6}?LL/
+    xxxxxxxxxxxPSTAIREISLLxxxxxxxxx
+ 0: PSTAIREISLL
+
+/P[^*]TAIRE[^*]{1,}?LL/
+    xxxxxxxxxxxPSTAIREISLLxxxxxxxxx
+ 0: PSTAIREISLL
+
+/(\.\d\d[1-9]?)\d+/
+    1.230003938
+ 0: .230003938
+ 1: .23
+    1.875000282   
+ 0: .875000282
+ 1: .875
+    1.235  
+ 0: .235
+ 1: .23
+                  
+/(\.\d\d((?=0)|\d(?=\d)))/
+    1.230003938      
+ 0: .23
+ 1: .23
+ 2: 
+    1.875000282
+ 0: .875
+ 1: .875
+ 2: 5
+    *** Failers 
+No match
+    1.235 
+No match
+    
+/a(?)b/
+    ab 
+ 0: ab
+ 
+/\b(foo)\s+(\w+)/i
+    Food is on the foo table
+ 0: foo table
+ 1: foo
+ 2: table
+    
+/foo(.*)bar/
+    The food is under the bar in the barn.
+ 0: food is under the bar in the bar
+ 1: d is under the bar in the 
+    
+/foo(.*?)bar/  
+    The food is under the bar in the barn.
+ 0: food is under the bar
+ 1: d is under the 
+
+/(.*)(\d*)/
+    I have 2 numbers: 53147
+ 0: I have 2 numbers: 53147
+ 1: I have 2 numbers: 53147
+ 2: 
+    
+/(.*)(\d+)/
+    I have 2 numbers: 53147
+ 0: I have 2 numbers: 53147
+ 1: I have 2 numbers: 5314
+ 2: 7
+ 
+/(.*?)(\d*)/
+    I have 2 numbers: 53147
+ 0: 
+ 1: 
+ 2: 
+
+/(.*?)(\d+)/
+    I have 2 numbers: 53147
+ 0: I have 2
+ 1: I have 
+ 2: 2
+
+/(.*)(\d+)$/
+    I have 2 numbers: 53147
+ 0: I have 2 numbers: 53147
+ 1: I have 2 numbers: 5314
+ 2: 7
+
+/(.*?)(\d+)$/
+    I have 2 numbers: 53147
+ 0: I have 2 numbers: 53147
+ 1: I have 2 numbers: 
+ 2: 53147
+
+/(.*)\b(\d+)$/
+    I have 2 numbers: 53147
+ 0: I have 2 numbers: 53147
+ 1: I have 2 numbers: 
+ 2: 53147
+
+/(.*\D)(\d+)$/
+    I have 2 numbers: 53147
+ 0: I have 2 numbers: 53147
+ 1: I have 2 numbers: 
+ 2: 53147
+
+/^\D*(?!123)/
+    ABC123
+ 0: AB
+     
+/^(\D*)(?=\d)(?!123)/
+    ABC445
+ 0: ABC
+ 1: ABC
+    *** Failers
+No match
+    ABC123
+No match
+    
+/^[W-]46]/
+    W46]789 
+ 0: W46]
+    -46]789
+ 0: -46]
+    *** Failers
+No match
+    Wall
+No match
+    Zebra
+No match
+    42
+No match
+    [abcd] 
+No match
+    ]abcd[
+No match
+       
+/^[W-\]46]/
+    W46]789 
+ 0: W
+    Wall
+ 0: W
+    Zebra
+ 0: Z
+    Xylophone  
+ 0: X
+    42
+ 0: 4
+    [abcd] 
+ 0: [
+    ]abcd[
+ 0: ]
+    \\backslash 
+ 0: \
+    *** Failers
+No match
+    -46]789
+No match
+    well
+No match
+    
+/\d\d\/\d\d\/\d\d\d\d/
+    01/01/2000
+ 0: 01/01/2000
+
+/word (?:[a-zA-Z0-9]+ ){0,10}otherword/
+  word cat dog elephant mussel cow horse canary baboon snake shark otherword
+ 0: word cat dog elephant mussel cow horse canary baboon snake shark otherword
+  word cat dog elephant mussel cow horse canary baboon snake shark
+No match
+
+/word (?:[a-zA-Z0-9]+ ){0,300}otherword/
+  word cat dog elephant mussel cow horse canary baboon snake shark the quick brown fox and the lazy dog and several other words getting close to thirty by now I hope
+No match
+
+/^(a){0,0}/
+    bcd
+ 0: 
+    abc
+ 0: 
+    aab     
+ 0: 
+
+/^(a){0,1}/
+    bcd
+ 0: 
+    abc
+ 0: a
+ 1: a
+    aab  
+ 0: a
+ 1: a
+
+/^(a){0,2}/
+    bcd
+ 0: 
+    abc
+ 0: a
+ 1: a
+    aab  
+ 0: aa
+ 1: a
+
+/^(a){0,3}/
+    bcd
+ 0: 
+    abc
+ 0: a
+ 1: a
+    aab
+ 0: aa
+ 1: a
+    aaa   
+ 0: aaa
+ 1: a
+
+/^(a){0,}/
+    bcd
+ 0: 
+    abc
+ 0: a
+ 1: a
+    aab
+ 0: aa
+ 1: a
+    aaa
+ 0: aaa
+ 1: a
+    aaaaaaaa    
+ 0: aaaaaaaa
+ 1: a
+
+/^(a){1,1}/
+    bcd
+No match
+    abc
+ 0: a
+ 1: a
+    aab  
+ 0: a
+ 1: a
+
+/^(a){1,2}/
+    bcd
+No match
+    abc
+ 0: a
+ 1: a
+    aab  
+ 0: aa
+ 1: a
+
+/^(a){1,3}/
+    bcd
+No match
+    abc
+ 0: a
+ 1: a
+    aab
+ 0: aa
+ 1: a
+    aaa   
+ 0: aaa
+ 1: a
+
+/^(a){1,}/
+    bcd
+No match
+    abc
+ 0: a
+ 1: a
+    aab
+ 0: aa
+ 1: a
+    aaa
+ 0: aaa
+ 1: a
+    aaaaaaaa    
+ 0: aaaaaaaa
+ 1: a
+
+/.*\.gif/
+    borfle\nbib.gif\nno
+ 0: bib.gif
+
+/.{0,}\.gif/
+    borfle\nbib.gif\nno
+ 0: bib.gif
+
+/.*\.gif/m
+    borfle\nbib.gif\nno
+ 0: bib.gif
+
+/.*\.gif/s
+    borfle\nbib.gif\nno
+ 0: borfle\x0abib.gif
+
+/.*\.gif/ms
+    borfle\nbib.gif\nno
+ 0: borfle\x0abib.gif
+    
+/.*$/
+    borfle\nbib.gif\nno
+ 0: no
+
+/.*$/m
+    borfle\nbib.gif\nno
+ 0: borfle
+
+/.*$/s
+    borfle\nbib.gif\nno
+ 0: borfle\x0abib.gif\x0ano
+
+/.*$/ms
+    borfle\nbib.gif\nno
+ 0: borfle\x0abib.gif\x0ano
+    
+/.*$/
+    borfle\nbib.gif\nno\n
+ 0: no
+
+/.*$/m
+    borfle\nbib.gif\nno\n
+ 0: borfle
+
+/.*$/s
+    borfle\nbib.gif\nno\n
+ 0: borfle\x0abib.gif\x0ano\x0a
+
+/.*$/ms
+    borfle\nbib.gif\nno\n
+ 0: borfle\x0abib.gif\x0ano\x0a
+    
+/(.*X|^B)/
+    abcde\n1234Xyz
+ 0: 1234X
+ 1: 1234X
+    BarFoo 
+ 0: B
+ 1: B
+    *** Failers
+No match
+    abcde\nBar  
+No match
+
+/(.*X|^B)/m
+    abcde\n1234Xyz
+ 0: 1234X
+ 1: 1234X
+    BarFoo 
+ 0: B
+ 1: B
+    abcde\nBar  
+ 0: B
+ 1: B
+
+/(.*X|^B)/s
+    abcde\n1234Xyz
+ 0: abcde\x0a1234X
+ 1: abcde\x0a1234X
+    BarFoo 
+ 0: B
+ 1: B
+    *** Failers
+No match
+    abcde\nBar  
+No match
+
+/(.*X|^B)/ms
+    abcde\n1234Xyz
+ 0: abcde\x0a1234X
+ 1: abcde\x0a1234X
+    BarFoo 
+ 0: B
+ 1: B
+    abcde\nBar  
+ 0: B
+ 1: B
+
+/(?s)(.*X|^B)/
+    abcde\n1234Xyz
+ 0: abcde\x0a1234X
+ 1: abcde\x0a1234X
+    BarFoo 
+ 0: B
+ 1: B
+    *** Failers 
+No match
+    abcde\nBar  
+No match
+
+/(?s:.*X|^B)/
+    abcde\n1234Xyz
+ 0: abcde\x0a1234X
+    BarFoo 
+ 0: B
+    *** Failers 
+No match
+    abcde\nBar  
+No match
+
+/^.*B/
+    **** Failers
+No match
+    abc\nB
+No match
+     
+/(?s)^.*B/
+    abc\nB
+ 0: abc\x0aB
+
+/(?m)^.*B/
+    abc\nB
+ 0: B
+     
+/(?ms)^.*B/
+    abc\nB
+ 0: abc\x0aB
+
+/(?ms)^B/
+    abc\nB
+ 0: B
+
+/(?s)B$/
+    B\n
+ 0: B
+
+/^[0-9][0-9][0-9][0-9][0-9][0-9][0-9][0-9][0-9][0-9][0-9][0-9]/
+    123456654321
+ 0: 123456654321
+  
+/^\d\d\d\d\d\d\d\d\d\d\d\d/
+    123456654321 
+ 0: 123456654321
+
+/^[\d][\d][\d][\d][\d][\d][\d][\d][\d][\d][\d][\d]/
+    123456654321
+ 0: 123456654321
+  
+/^[abc]{12}/
+    abcabcabcabc
+ 0: abcabcabcabc
+    
+/^[a-c]{12}/
+    abcabcabcabc
+ 0: abcabcabcabc
+    
+/^(a|b|c){12}/
+    abcabcabcabc 
+ 0: abcabcabcabc
+ 1: c
+
+/^[abcdefghijklmnopqrstuvwxy0123456789]/
+    n
+ 0: n
+    *** Failers 
+No match
+    z 
+No match
+
+/abcde{0,0}/
+    abcd
+ 0: abcd
+    *** Failers
+No match
+    abce  
+No match
+
+/ab[cd]{0,0}e/
+    abe
+ 0: abe
+    *** Failers
+No match
+    abcde 
+No match
+    
+/ab(c){0,0}d/
+    abd
+ 0: abd
+    *** Failers
+No match
+    abcd   
+No match
+
+/a(b*)/
+    a
+ 0: a
+ 1: 
+    ab
+ 0: ab
+ 1: b
+    abbbb
+ 0: abbbb
+ 1: bbbb
+    *** Failers
+ 0: a
+ 1: 
+    bbbbb    
+No match
+    
+/ab\d{0}e/
+    abe
+ 0: abe
+    *** Failers
+No match
+    ab1e   
+No match
+    
+/"([^\\"]+|\\.)*"/
+    the \"quick\" brown fox
+ 0: "quick"
+ 1: quick
+    \"the \\\"quick\\\" brown fox\" 
+ 0: "the \"quick\" brown fox"
+ 1:  brown fox
+
+/.*?/g+
+    abc
+ 0: 
+ 0+ abc
+ 0: a
+ 0+ bc
+ 0: 
+ 0+ bc
+ 0: b
+ 0+ c
+ 0: 
+ 0+ c
+ 0: c
+ 0+ 
+ 0: 
+ 0+ 
+  
+/\b/g+
+    abc 
+ 0: 
+ 0+ abc
+ 0: 
+ 0+ 
+
+/\b/+g
+    abc 
+ 0: 
+ 0+ abc
+ 0: 
+ 0+ 
+
+//g
+    abc
+ 0: 
+ 0: 
+ 0: 
+ 0: 
+
+/]{0,})>]{0,})>([\d]{0,}\.)(.*)((
([\w\W\s\d][^<>]{0,})|[\s]{0,}))<\/a><\/TD>]{0,})>([\w\W\s\d][^<>]{0,})<\/TD>]{0,})>([\w\W\s\d][^<>]{0,})<\/TD><\/TR>/is
+  43.Word Processor
(N-1286)Lega lstaff.comCA - Statewide
+ 0: 43.Word Processor
(N-1286)Lega lstaff.comCA - Statewide
+ 1:  BGCOLOR='#DBE9E9'
+ 2:  align=left valign=top
+ 3: 43.
+ 4: Word Processor
(N-1286)
+ 5: 
+ 6: 
+ 7: 
+ 8:  align=left valign=top
+ 9: Lega lstaff.com
+10:  align=left valign=top
+11: CA - Statewide
+
+/a[^a]b/
+    acb
+ 0: acb
+    a\nb
+ 0: a\x0ab
+    
+/a.b/
+    acb
+ 0: acb
+    *** Failers 
+No match
+    a\nb   
+No match
+    
+/a[^a]b/s
+    acb
+ 0: acb
+    a\nb  
+ 0: a\x0ab
+    
+/a.b/s
+    acb
+ 0: acb
+    a\nb  
+ 0: a\x0ab
+
+/^(b+?|a){1,2}?c/
+    bac
+ 0: bac
+ 1: a
+    bbac
+ 0: bbac
+ 1: a
+    bbbac
+ 0: bbbac
+ 1: a
+    bbbbac
+ 0: bbbbac
+ 1: a
+    bbbbbac 
+ 0: bbbbbac
+ 1: a
+
+/^(b+|a){1,2}?c/
+    bac
+ 0: bac
+ 1: a
+    bbac
+ 0: bbac
+ 1: a
+    bbbac
+ 0: bbbac
+ 1: a
+    bbbbac
+ 0: bbbbac
+ 1: a
+    bbbbbac 
+ 0: bbbbbac
+ 1: a
+    
+/(?!\A)x/m
+    x\nb\n
+No match
+    a\bx\n  
+ 0: x
+    
+/\x0{ab}/
+    \0{ab} 
+ 0: \x00{ab}
+
+/(A|B)*?CD/
+    CD 
+ 0: CD
+    
+/(A|B)*CD/
+    CD 
+ 0: CD
+
+/(AB)*?\1/
+    ABABAB
+ 0: ABAB
+ 1: AB
+
+/(AB)*\1/
+    ABABAB
+ 0: ABABAB
+ 1: AB
+    
+/(?.*/)foo"
+    /this/is/a/very/long/line/in/deed/with/very/many/slashes/in/it/you/see/
+No match
+
+"(?>.*/)foo"
+    /this/is/a/very/long/line/in/deed/with/very/many/slashes/in/and/foo
+ 0: /this/is/a/very/long/line/in/deed/with/very/many/slashes/in/and/foo
+
+/(?>(\.\d\d[1-9]?))\d+/
+    1.230003938
+ 0: .230003938
+ 1: .23
+    1.875000282
+ 0: .875000282
+ 1: .875
+    *** Failers 
+No match
+    1.235 
+No match
+
+/^((?>\w+)|(?>\s+))*$/
+    now is the time for all good men to come to the aid of the party
+ 0: now is the time for all good men to come to the aid of the party
+ 1: party
+    *** Failers
+No match
+    this is not a line with only words and spaces!
+No match
+    
+/(\d+)(\w)/
+    12345a
+ 0: 12345a
+ 1: 12345
+ 2: a
+    12345+ 
+ 0: 12345
+ 1: 1234
+ 2: 5
+
+/((?>\d+))(\w)/
+    12345a
+ 0: 12345a
+ 1: 12345
+ 2: a
+    *** Failers
+No match
+    12345+ 
+No match
+
+/(?>a+)b/
+    aaab
+ 0: aaab
+
+/((?>a+)b)/
+    aaab
+ 0: aaab
+ 1: aaab
+
+/(?>(a+))b/
+    aaab
+ 0: aaab
+ 1: aaa
+
+/(?>b)+/
+    aaabbbccc
+ 0: bbb
+
+/(?>a+|b+|c+)*c/
+    aaabbbbccccd
+ 0: aaabbbbc
+
+/((?>[^()]+)|\([^()]*\))+/
+    ((abc(ade)ufh()()x
+ 0: abc(ade)ufh()()x
+ 1: x
+    
+/\(((?>[^()]+)|\([^()]+\))+\)/ 
+    (abc)
+ 0: (abc)
+ 1: abc
+    (abc(def)xyz)
+ 0: (abc(def)xyz)
+ 1: xyz
+    *** Failers
+No match
+    ((()aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa   
+No match
+
+/a(?-i)b/i
+    ab
+ 0: ab
+    Ab
+ 0: Ab
+    *** Failers 
+No match
+    aB
+No match
+    AB
+No match
+        
+/(a (?x)b c)d e/
+    a bcd e
+ 0: a bcd e
+ 1: a bc
+    *** Failers
+No match
+    a b cd e
+No match
+    abcd e   
+No match
+    a bcde 
+No match
+ 
+/(a b(?x)c d (?-x)e f)/
+    a bcde f
+ 0: a bcde f
+ 1: a bcde f
+    *** Failers
+No match
+    abcdef  
+No match
+
+/(a(?i)b)c/
+    abc
+ 0: abc
+ 1: ab
+    aBc
+ 0: aBc
+ 1: aB
+    *** Failers
+No match
+    abC
+No match
+    aBC  
+No match
+    Abc
+No match
+    ABc
+No match
+    ABC
+No match
+    AbC
+No match
+    
+/a(?i:b)c/
+    abc
+ 0: abc
+    aBc
+ 0: aBc
+    *** Failers 
+No match
+    ABC
+No match
+    abC
+No match
+    aBC
+No match
+    
+/a(?i:b)*c/
+    aBc
+ 0: aBc
+    aBBc
+ 0: aBBc
+    *** Failers 
+No match
+    aBC
+No match
+    aBBC
+No match
+    
+/a(?=b(?i)c)\w\wd/
+    abcd
+ 0: abcd
+    abCd
+ 0: abCd
+    *** Failers
+No match
+    aBCd
+No match
+    abcD     
+No match
+    
+/(?s-i:more.*than).*million/i
+    more than million
+ 0: more than million
+    more than MILLION
+ 0: more than MILLION
+    more \n than Million 
+ 0: more \x0a than Million
+    *** Failers
+No match
+    MORE THAN MILLION    
+No match
+    more \n than \n million 
+No match
+
+/(?:(?s-i)more.*than).*million/i
+    more than million
+ 0: more than million
+    more than MILLION
+ 0: more than MILLION
+    more \n than Million 
+ 0: more \x0a than Million
+    *** Failers
+No match
+    MORE THAN MILLION    
+No match
+    more \n than \n million 
+No match
+    
+/(?>a(?i)b+)+c/ 
+    abc
+ 0: abc
+    aBbc
+ 0: aBbc
+    aBBc 
+ 0: aBBc
+    *** Failers
+No match
+    Abc
+No match
+    abAb    
+No match
+    abbC 
+No match
+    
+/(?=a(?i)b)\w\wc/
+    abc
+ 0: abc
+    aBc
+ 0: aBc
+    *** Failers
+No match
+    Ab 
+No match
+    abC
+No match
+    aBC     
+No match
+    
+/(?<=a(?i)b)(\w\w)c/
+    abxxc
+ 0: xxc
+ 1: xx
+    aBxxc
+ 0: xxc
+ 1: xx
+    *** Failers
+No match
+    Abxxc
+No match
+    ABxxc
+No match
+    abxxC      
+No match
+
+/(?:(a)|b)(?(1)A|B)/
+    aA
+ 0: aA
+ 1: a
+    bB
+ 0: bB
+    *** Failers
+No match
+    aB
+No match
+    bA    
+No match
+
+/^(a)?(?(1)a|b)+$/
+    aa
+ 0: aa
+ 1: a
+    b
+ 0: b
+    bb  
+ 0: bb
+    *** Failers
+No match
+    ab   
+No match
+
+/^(?(?=abc)\w{3}:|\d\d)$/
+    abc:
+ 0: abc:
+    12
+ 0: 12
+    *** Failers
+No match
+    123
+No match
+    xyz    
+No match
+
+/^(?(?!abc)\d\d|\w{3}:)$/
+    abc:
+ 0: abc:
+    12
+ 0: 12
+    *** Failers
+No match
+    123
+No match
+    xyz    
+No match
+    
+/(?(?<=foo)bar|cat)/
+    foobar
+ 0: bar
+    cat
+ 0: cat
+    fcat
+ 0: cat
+    focat   
+ 0: cat
+    *** Failers
+No match
+    foocat  
+No match
+
+/(?(?a*)*/
+    a
+ 0: a
+    aa
+ 0: aa
+    aaaa
+ 0: aaaa
+    
+/(abc|)+/
+    abc
+ 0: abc
+ 1: 
+    abcabc
+ 0: abcabc
+ 1: 
+    abcabcabc
+ 0: abcabcabc
+ 1: 
+    xyz      
+ 0: 
+ 1: 
+
+/([a]*)*/
+    a
+ 0: a
+ 1: 
+    aaaaa 
+ 0: aaaaa
+ 1: 
+ 
+/([ab]*)*/
+    a
+ 0: a
+ 1: 
+    b
+ 0: b
+ 1: 
+    ababab
+ 0: ababab
+ 1: 
+    aaaabcde
+ 0: aaaab
+ 1: 
+    bbbb    
+ 0: bbbb
+ 1: 
+ 
+/([^a]*)*/
+    b
+ 0: b
+ 1: 
+    bbbb
+ 0: bbbb
+ 1: 
+    aaa   
+ 0: 
+ 1: 
+ 
+/([^ab]*)*/
+    cccc
+ 0: cccc
+ 1: 
+    abab  
+ 0: 
+ 1: 
+ 
+/([a]*?)*/
+    a
+ 0: 
+ 1: 
+    aaaa 
+ 0: 
+ 1: 
+ 
+/([ab]*?)*/
+    a
+ 0: 
+ 1: 
+    b
+ 0: 
+ 1: 
+    abab
+ 0: 
+ 1: 
+    baba   
+ 0: 
+ 1: 
+ 
+/([^a]*?)*/
+    b
+ 0: 
+ 1: 
+    bbbb
+ 0: 
+ 1: 
+    aaa   
+ 0: 
+ 1: 
+ 
+/([^ab]*?)*/
+    c
+ 0: 
+ 1: 
+    cccc
+ 0: 
+ 1: 
+    baba   
+ 0: 
+ 1: 
+ 
+/(?>a*)*/
+    a
+ 0: a
+    aaabcde 
+ 0: aaa
+ 
+/((?>a*))*/
+    aaaaa
+ 0: aaaaa
+ 1: 
+    aabbaa 
+ 0: aa
+ 1: 
+ 
+/((?>a*?))*/
+    aaaaa
+ 0: 
+ 1: 
+    aabbaa 
+ 0: 
+ 1: 
+
+/(?(?=[^a-z]+[a-z])  \d{2}-[a-z]{3}-\d{2}  |  \d{2}-\d{2}-\d{2} ) /x
+    12-sep-98
+ 0: 12-sep-98
+    12-09-98
+ 0: 12-09-98
+    *** Failers
+No match
+    sep-12-98
+No match
+        
+/(?<=(foo))bar\1/
+    foobarfoo
+ 0: barfoo
+ 1: foo
+    foobarfootling 
+ 0: barfoo
+ 1: foo
+    *** Failers
+No match
+    foobar
+No match
+    barfoo   
+No match
+
+/(?i:saturday|sunday)/
+    saturday
+ 0: saturday
+    sunday
+ 0: sunday
+    Saturday
+ 0: Saturday
+    Sunday
+ 0: Sunday
+    SATURDAY
+ 0: SATURDAY
+    SUNDAY
+ 0: SUNDAY
+    SunDay
+ 0: SunDay
+    
+/(a(?i)bc|BB)x/
+    abcx
+ 0: abcx
+ 1: abc
+    aBCx
+ 0: aBCx
+ 1: aBC
+    bbx
+ 0: bbx
+ 1: bb
+    BBx
+ 0: BBx
+ 1: BB
+    *** Failers
+No match
+    abcX
+No match
+    aBCX
+No match
+    bbX
+No match
+    BBX               
+No match
+
+/^([ab](?i)[cd]|[ef])/
+    ac
+ 0: ac
+ 1: ac
+    aC
+ 0: aC
+ 1: aC
+    bD
+ 0: bD
+ 1: bD
+    elephant
+ 0: e
+ 1: e
+    Europe 
+ 0: E
+ 1: E
+    frog
+ 0: f
+ 1: f
+    France
+ 0: F
+ 1: F
+    *** Failers
+No match
+    Africa     
+No match
+
+/^(ab|a(?i)[b-c](?m-i)d|x(?i)y|z)/
+    ab
+ 0: ab
+ 1: ab
+    aBd
+ 0: aBd
+ 1: aBd
+    xy
+ 0: xy
+ 1: xy
+    xY
+ 0: xY
+ 1: xY
+    zebra
+ 0: z
+ 1: z
+    Zambesi
+ 0: Z
+ 1: Z
+    *** Failers
+No match
+    aCD  
+No match
+    XY  
+No match
+
+/(?<=foo\n)^bar/m
+    foo\nbar
+ 0: bar
+    *** Failers
+No match
+    bar
+No match
+    baz\nbar   
+No match
+
+/(?<=(?]&/
+    <&OUT
+ 0: <&
+
+/^(a\1?){4}$/
+    aaaaaaaaaa
+ 0: aaaaaaaaaa
+ 1: aaaa
+    *** Failers
+No match
+    AB
+No match
+    aaaaaaaaa
+No match
+    aaaaaaaaaaa
+No match
+
+/^(a(?(1)\1)){4}$/
+    aaaaaaaaaa
+ 0: aaaaaaaaaa
+ 1: aaaa
+    *** Failers
+No match
+    aaaaaaaaa
+No match
+    aaaaaaaaaaa
+No match
+
+/(?:(f)(o)(o)|(b)(a)(r))*/
+    foobar
+ 0: foobar
+ 1: f
+ 2: o
+ 3: o
+ 4: b
+ 5: a
+ 6: r
+
+/(?<=a)b/
+    ab
+ 0: b
+    *** Failers
+No match
+    cb
+No match
+    b
+No match
+
+/(?
+ 2: abcd
+    xy:z:::abcd
+ 0: xy:z:::abcd
+ 1: xy:z:::
+ 2: abcd
+
+/^[^bcd]*(c+)/
+    aexycd
+ 0: aexyc
+ 1: c
+
+/(a*)b+/
+    caab
+ 0: aab
+ 1: aa
+
+/([\w:]+::)?(\w+)$/
+    abcd
+ 0: abcd
+ 1: 
+ 2: abcd
+    xy:z:::abcd
+ 0: xy:z:::abcd
+ 1: xy:z:::
+ 2: abcd
+    *** Failers
+ 0: Failers
+ 1: 
+ 2: Failers
+    abcd:
+No match
+    abcd:
+No match
+
+/^[^bcd]*(c+)/
+    aexycd
+ 0: aexyc
+ 1: c
+
+/(>a+)ab/
+
+/(?>a+)b/
+    aaab
+ 0: aaab
+
+/([[:]+)/
+    a:[b]:
+ 0: :[
+ 1: :[
+
+/([[=]+)/
+    a=[b]=
+ 0: =[
+ 1: =[
+
+/([[.]+)/
+    a.[b].
+ 0: .[
+ 1: .[
+
+/((?>a+)b)/
+    aaab
+ 0: aaab
+ 1: aaab
+
+/(?>(a+))b/
+    aaab
+ 0: aaab
+ 1: aaa
+
+/((?>[^()]+)|\([^()]*\))+/
+    ((abc(ade)ufh()()x
+ 0: abc(ade)ufh()()x
+ 1: x
+
+/a\Z/
+    *** Failers
+No match
+    aaab
+No match
+    a\nb\n
+No match
+
+/b\Z/
+    a\nb\n
+ 0: b
+
+/b\z/
+
+/b\Z/
+    a\nb
+ 0: b
+
+/b\z/
+    a\nb
+ 0: b
+    *** Failers
+No match
+    
+/^(?>(?(1)\.|())[^\W_](?>[a-z0-9-]*[^\W_])?)+$/
+    a
+ 0: a
+ 1: 
+    abc
+ 0: abc
+ 1: 
+    a-b
+ 0: a-b
+ 1: 
+    0-9 
+ 0: 0-9
+ 1: 
+    a.b
+ 0: a.b
+ 1: 
+    5.6.7  
+ 0: 5.6.7
+ 1: 
+    the.quick.brown.fox
+ 0: the.quick.brown.fox
+ 1: 
+    a100.b200.300c  
+ 0: a100.b200.300c
+ 1: 
+    12-ab.1245 
+ 0: 12-ab.1245
+ 1: 
+    *** Failers
+No match
+    \
+No match
+    .a
+No match
+    -a
+No match
+    a-
+No match
+    a.  
+No match
+    a_b 
+No match
+    a.-
+No match
+    a..  
+No match
+    ab..bc 
+No match
+    the.quick.brown.fox-
+No match
+    the.quick.brown.fox.
+No match
+    the.quick.brown.fox_
+No match
+    the.quick.brown.fox+       
+No match
+
+/(?>.*)(?<=(abcd|wxyz))/
+    alphabetabcd
+ 0: alphabetabcd
+ 1: abcd
+    endingwxyz
+ 0: endingwxyz
+ 1: wxyz
+    *** Failers
+No match
+    a rather long string that doesn't end with one of them
+No match
+
+/word (?>(?:(?!otherword)[a-zA-Z0-9]+ ){0,30})otherword/
+    word cat dog elephant mussel cow horse canary baboon snake shark otherword
+ 0: word cat dog elephant mussel cow horse canary baboon snake shark otherword
+    word cat dog elephant mussel cow horse canary baboon snake shark
+No match
+  
+/word (?>[a-zA-Z0-9]+ ){0,30}otherword/
+    word cat dog elephant mussel cow horse canary baboon snake shark the quick brown fox and the lazy dog and several other words getting close to thirty by now I hope
+No match
+
+/(?<=\d{3}(?!999))foo/
+    999foo
+ 0: foo
+    123999foo 
+ 0: foo
+    *** Failers
+No match
+    123abcfoo
+No match
+    
+/(?<=(?!...999)\d{3})foo/
+    999foo
+ 0: foo
+    123999foo 
+ 0: foo
+    *** Failers
+No match
+    123abcfoo
+No match
+
+/(?<=\d{3}(?!999)...)foo/
+    123abcfoo
+ 0: foo
+    123456foo 
+ 0: foo
+    *** Failers
+No match
+    123999foo  
+No match
+    
+/(?<=\d{3}...)(?
+ 2: 
+ 3: abcd
+    
+ 2: 
+ 3: abcd
+    \s*)=(?>\s*)        # find 
+ 2: 
+ 3: abcd
+    Z)+|A)*/
+    ZABCDEFG
+ 0: ZA
+ 1: A
+
+/((?>)+|A)*/
+    ZABCDEFG
+ 0: 
+ 1: 
+
+/a*/g
+    abbab
+ 0: a
+ 0: 
+ 0: 
+ 0: a
+ 0: 
+ 0: 
+
+/^[a-\d]/
+    abcde
+ 0: a
+    -things
+ 0: -
+    0digit
+ 0: 0
+    *** Failers
+No match
+    bcdef    
+No match
+
+/^[\d-a]/
+    abcde
+ 0: a
+    -things
+ 0: -
+    0digit
+ 0: 0
+    *** Failers
+No match
+    bcdef    
+No match
+    
+/[[:space:]]+/
+    > \x09\x0a\x0c\x0d\x0b<
+ 0:  \x09\x0a\x0c\x0d\x0b
+     
+/[[:blank:]]+/
+    > \x09\x0a\x0c\x0d\x0b<
+ 0:  \x09
+     
+/[\s]+/
+    > \x09\x0a\x0c\x0d\x0b<
+ 0:  \x09\x0a\x0c\x0d
+     
+/\s+/
+    > \x09\x0a\x0c\x0d\x0b<
+ 0:  \x09\x0a\x0c\x0d
+     
+/ab/x
+    ab
+No match
+
+/(?!\A)x/m
+  a\nxb\n
+ 0: x
+
+/(?!^)x/m
+  a\nxb\n
+No match
+
+/abc\Qabc\Eabc/
+    abcabcabc
+ 0: abcabcabc
+    
+/abc\Q(*+|\Eabc/
+    abc(*+|abc 
+ 0: abc(*+|abc
+
+/   abc\Q abc\Eabc/x
+    abc abcabc
+ 0: abc abcabc
+    *** Failers
+No match
+    abcabcabc  
+No match
+    
+/abc#comment
+    \Q#not comment
+    literal\E/x
+    abc#not comment\n    literal     
+ 0: abc#not comment\x0a    literal
+
+/abc#comment
+    \Q#not comment
+    literal/x
+    abc#not comment\n    literal     
+ 0: abc#not comment\x0a    literal
+
+/abc#comment
+    \Q#not comment
+    literal\E #more comment
+    /x
+    abc#not comment\n    literal     
+ 0: abc#not comment\x0a    literal
+
+/abc#comment
+    \Q#not comment
+    literal\E #more comment/x
+    abc#not comment\n    literal     
+ 0: abc#not comment\x0a    literal
+
+/\Qabc\$xyz\E/
+    abc\\\$xyz
+ 0: abc\$xyz
+
+/\Qabc\E\$\Qxyz\E/
+    abc\$xyz
+ 0: abc$xyz
+
+/\Gabc/
+    abc
+ 0: abc
+    *** Failers
+No match
+    xyzabc  
+No match
+
+/\Gabc./g
+    abc1abc2xyzabc3
+ 0: abc1
+ 0: abc2
+
+/abc./g
+    abc1abc2xyzabc3 
+ 0: abc1
+ 0: abc2
+ 0: abc3
+
+/a(?x: b c )d/
+    XabcdY
+ 0: abcd
+    *** Failers 
+No match
+    Xa b c d Y 
+No match
+
+/((?x)x y z | a b c)/
+    XabcY
+ 0: abc
+ 1: abc
+    AxyzB 
+ 0: xyz
+ 1: xyz
+
+/(?i)AB(?-i)C/
+    XabCY
+ 0: abC
+    *** Failers
+No match
+    XabcY  
+No match
+
+/((?i)AB(?-i)C|D)E/
+    abCE
+ 0: abCE
+ 1: abC
+    DE
+ 0: DE
+ 1: D
+    *** Failers
+No match
+    abcE
+No match
+    abCe  
+No match
+    dE
+No match
+    De    
+No match
+
+/(.*)\d+\1/
+    abc123abc
+ 0: abc123abc
+ 1: abc
+    abc123bc 
+ 0: bc123bc
+ 1: bc
+
+/(.*)\d+\1/s
+    abc123abc
+ 0: abc123abc
+ 1: abc
+    abc123bc 
+ 0: bc123bc
+ 1: bc
+    
+/((.*))\d+\1/
+    abc123abc
+ 0: abc123abc
+ 1: abc
+ 2: abc
+    abc123bc  
+ 0: bc123bc
+ 1: bc
+ 2: bc
+
+/-- This tests for an IPv6 address in the form where it can have up to
+    eight components, one and only one of which is empty. This must be
+    an internal component. --/
+
+/^(?!:)                       # colon disallowed at start
+  (?:                         # start of item
+    (?: [0-9a-f]{1,4} |       # 1-4 hex digits or
+    (?(1)0 | () ) )           # if null previously matched, fail; else null
+    :                         # followed by colon
+  ){1,7}                      # end item; 1-7 of them required               
+  [0-9a-f]{1,4} $             # final hex number at end of string
+  (?(1)|.)                    # check that there was an empty component
+  /xi
+    a123::a123
+ 0: a123::a123
+ 1: 
+    a123:b342::abcd
+ 0: a123:b342::abcd
+ 1: 
+    a123:b342::324e:abcd
+ 0: a123:b342::324e:abcd
+ 1: 
+    a123:ddde:b342::324e:abcd
+ 0: a123:ddde:b342::324e:abcd
+ 1: 
+    a123:ddde:b342::324e:dcba:abcd
+ 0: a123:ddde:b342::324e:dcba:abcd
+ 1: 
+    a123:ddde:9999:b342::324e:dcba:abcd
+ 0: a123:ddde:9999:b342::324e:dcba:abcd
+ 1: 
+    *** Failers
+No match
+    1:2:3:4:5:6:7:8
+No match
+    a123:bce:ddde:9999:b342::324e:dcba:abcd
+No match
+    a123::9999:b342::324e:dcba:abcd
+No match
+    abcde:2:3:4:5:6:7:8
+No match
+    ::1
+No match
+    abcd:fee0:123::   
+No match
+    :1
+No match
+    1:  
+No match
+
+/[z\Qa-d]\E]/
+    z
+ 0: z
+    a
+ 0: a
+    -
+ 0: -
+    d
+ 0: d
+    ] 
+ 0: ]
+    *** Failers
+ 0: a
+    b     
+No match
+
+/[\z\C]/
+    z
+ 0: z
+    C 
+ 0: C
+    
+/\M/
+    M 
+ 0: M
+    
+/(a+)*b/
+    aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa 
+No match
+    
+/(?i)reg(?:ul(?:[a]|ae)r|ex)/
+    REGular
+ 0: REGular
+    regulaer
+ 0: regulaer
+    Regex  
+ 0: Regex
+    regulr 
+ 0: regul\xe4r
+
+/[--]+/
+    
+ 0: \xc5\xe6\xe5\xe4\xe0
+    
+ 0: \xc5\xe6\xe5\xe4\xff
+    
+ 0: \xc5\xe6\xe5\xe4\xc0
+    
+ 0: \xc5\xe6\xe5\xe4\xdf
+
+/(?<=Z)X./
+  \x84XAZXB
+ 0: XB
+
+/ab cd (?x) de fg/
+    ab cd defg
+ 0: ab cd defg
+
+/ab cd(?x) de fg/
+    ab cddefg
+ 0: ab cddefg
+    ** Failers 
+No match
+    abcddefg
+No match
+
+/(?
+ 2: 
+    D  
+ 0: D
+ 1: 
+ 2: 
+
+/(a|)*\d/
+  aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+No match
+  aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa4
+ 0: aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa4
+ 1: 
+
+/(?>a|)*\d/
+  aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+No match
+  aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa4
+ 0: aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa4
+
+/(?:a|)*\d/
+  aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+No match
+  aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa4
+ 0: aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa4
+
+/\Z/g
+  abc\n
+ 0: 
+ 0: 
+  
+/^(?s)(?>.*)(?
+ 2: a
+
+/(?>(a))b|(a)c/
+    ac
+ 0: ac
+ 1: 
+ 2: a
+
+/(?=(a))ab|(a)c/
+    ac
+ 0: ac
+ 1: 
+ 2: a
+
+/((?>(a))b|(a)c)/
+    ac
+ 0: ac
+ 1: ac
+ 2: 
+ 3: a
+
+/((?>(a))b|(a)c)++/
+    ac
+ 0: ac
+ 1: ac
+ 2: 
+ 3: a
+
+/(?:(?>(a))b|(a)c)++/
+    ac
+ 0: ac
+ 1: 
+ 2: a
+
+/(?=(?>(a))b|(a)c)(..)/
+    ac
+ 0: ac
+ 1: 
+ 2: a
+ 3: ac
+
+/(?>(?>(a))b|(a)c)/
+    ac
+ 0: ac
+ 1: 
+ 2: a
+
+/(?:(?>([ab])))+a=/+
+    =ba=
+ 0: ba=
+ 0+ 
+ 1: b
+
+/(?>([ab]))+a=/+
+    =ba=
+ 0: ba=
+ 0+ 
+ 1: b
+
+/((?>(a+)b)+(aabab))/
+    aaaabaaabaabab
+ 0: aaaabaaabaabab
+ 1: aaaabaaabaabab
+ 2: aaa
+ 3: aabab
+
+/(?>a+|ab)+?c/
+    aabc
+No match
+
+/(?>a+|ab)+c/
+    aabc
+No match
+
+/(?:a+|ab)+c/
+    aabc
+ 0: aabc
+
+/(?(?=(a))a)/
+    a
+ 0: a
+ 1: a
+
+/(?(?=(a))a)(b)/
+    ab
+ 0: ab
+ 1: a
+ 2: b
+
+/^(?:a|ab)++c/
+    aaaabc
+No match
+
+/^(?>a|ab)++c/
+    aaaabc
+No match
+
+/^(?:a|ab)+c/
+    aaaabc
+ 0: aaaabc
+
+/(?=abc){3}abc/+
+    abcabcabc
+ 0: abc
+ 0+ abcabc
+    ** Failers
+No match
+    xyz  
+No match
+    
+/(?=abc)+abc/+
+    abcabcabc
+ 0: abc
+ 0+ abcabc
+    ** Failers
+No match
+    xyz  
+No match
+    
+/(?=abc)++abc/+
+    abcabcabc
+ 0: abc
+ 0+ abcabc
+    ** Failers
+No match
+    xyz  
+No match
+    
+/(?=abc){0}xyz/
+    xyz 
+ 0: xyz
+
+/(?=abc){1}xyz/
+    ** Failers
+No match
+    xyz 
+No match
+    
+/(?=(a))?./
+    ab
+ 0: a
+ 1: a
+    bc
+ 0: b
+      
+/(?=(a))??./
+    ab
+ 0: a
+    bc
+ 0: b
+
+/^(?=(a)){0}b(?1)/
+    backgammon
+ 0: ba
+
+/^(?=(?1))?[az]([abc])d/
+    abd 
+ 0: abd
+ 1: b
+    zcdxx 
+ 0: zcd
+ 1: c
+
+/^(?!a){0}\w+/
+    aaaaa
+ 0: aaaaa
+
+/(?<=(abc))?xyz/
+    abcxyz
+ 0: xyz
+ 1: abc
+    pqrxyz 
+ 0: xyz
+
+/^[\g]+/
+    ggg<<>>
+ 0: ggg<<>>
+    ** Failers
+No match
+    \\ga  
+No match
+    
+/^[\ga]+/
+    gggagagaxyz 
+ 0: gggagaga
+    
+/^[:a[:digit:]]+/
+    aaaa444:::Z 
+ 0: aaaa444:::
+
+/^[:a[:digit:]:b]+/
+    aaaa444:::bbbZ 
+ 0: aaaa444:::bbb
+
+/[:a]xxx[b:]/
+     :xxx:
+ 0: :xxx:
+     
+/(?<=a{2})b/i
+    xaabc
+ 0: b
+    ** Failers
+No match
+    xabc  
+No match
+
+/(?XNNNYZ
+ 0: XNNNYZ
+    >  X NYQZ
+ 0:   X NYQZ
+    ** Failers
+No match
+    >XYZ   
+No match
+    >  X NY Z
+No match
+
+/\v*X\v?Y\v+Z\V*\x0a\V+\x0b\V{2,3}\x0c/
+    >XY\x0aZ\x0aA\x0bNN\x0c
+ 0: XY\x0aZ\x0aA\x0bNN\x0c
+    >\x0a\x0dX\x0aY\x0a\x0bZZZ\x0aAAA\x0bNNN\x0c
+ 0: \x0a\x0dX\x0aY\x0a\x0bZZZ\x0aAAA\x0bNNN\x0c
+
+/(foo)\Kbar/
+    foobar
+ 0: bar
+ 1: foo
+   
+/(foo)(\Kbar|baz)/
+    foobar
+ 0: bar
+ 1: foo
+ 2: bar
+    foobaz 
+ 0: foobaz
+ 1: foo
+ 2: baz
+
+/(foo\Kbar)baz/
+    foobarbaz
+ 0: barbaz
+ 1: foobar
+
+/abc\K|def\K/g+
+    Xabcdefghi
+ 0: 
+ 0+ defghi
+ 0: 
+ 0+ ghi
+
+/ab\Kc|de\Kf/g+
+    Xabcdefghi
+ 0: c
+ 0+ defghi
+ 0: f
+ 0+ ghi
+    
+/(?=C)/g+
+    ABCDECBA
+ 0: 
+ 0+ CDECBA
+ 0: 
+ 0+ CBA
+    
+/^abc\K/+
+    abcdef
+ 0: 
+ 0+ def
+    ** Failers
+No match
+    defabcxyz   
+No match
+
+/^(a(b))\1\g1\g{1}\g-1\g{-1}\g{-02}Z/
+    ababababbbabZXXXX
+ 0: ababababbbabZ
+ 1: ab
+ 2: b
+
+/(?tom|bon)-\g{A}/
+    tom-tom
+ 0: tom-tom
+ 1: tom
+    bon-bon 
+ 0: bon-bon
+ 1: bon
+    
+/(^(a|b\g{-1}))/
+    bacxxx
+No match
+
+/(?|(abc)|(xyz))\1/
+    abcabc
+ 0: abcabc
+ 1: abc
+    xyzxyz 
+ 0: xyzxyz
+ 1: xyz
+    ** Failers
+No match
+    abcxyz
+No match
+    xyzabc   
+No match
+    
+/(?|(abc)|(xyz))(?1)/
+    abcabc
+ 0: abcabc
+ 1: abc
+    xyzabc 
+ 0: xyzabc
+ 1: xyz
+    ** Failers 
+No match
+    xyzxyz 
+No match
+ 
+/^X(?5)(a)(?|(b)|(q))(c)(d)(Y)/
+    XYabcdY
+ 0: XYabcdY
+ 1: a
+ 2: b
+ 3: c
+ 4: d
+ 5: Y
+
+/^X(?7)(a)(?|(b|(r)(s))|(q))(c)(d)(Y)/
+    XYabcdY
+ 0: XYabcdY
+ 1: a
+ 2: b
+ 3: 
+ 4: 
+ 5: c
+ 6: d
+ 7: Y
+
+/^X(?7)(a)(?|(b|(?|(r)|(t))(s))|(q))(c)(d)(Y)/
+    XYabcdY
+ 0: XYabcdY
+ 1: a
+ 2: b
+ 3: 
+ 4: 
+ 5: c
+ 6: d
+ 7: Y
+
+/(?'abc'\w+):\k{2}/
+    a:aaxyz
+ 0: a:aa
+ 1: a
+    ab:ababxyz
+ 0: ab:abab
+ 1: ab
+    ** Failers
+No match
+    a:axyz
+No match
+    ab:abxyz
+No match
+
+/(?'abc'\w+):\g{abc}{2}/
+    a:aaxyz
+ 0: a:aa
+ 1: a
+    ab:ababxyz
+ 0: ab:abab
+ 1: ab
+    ** Failers
+No match
+    a:axyz
+No match
+    ab:abxyz
+No match
+
+/^(?a)? (?()b|c) (?('ab')d|e)/x
+    abd
+ 0: abd
+ 1: a
+    ce
+ 0: ce
+
+/^(a.)\g-1Z/
+    aXaXZ
+ 0: aXaXZ
+ 1: aX
+
+/^(a.)\g{-1}Z/
+    aXaXZ
+ 0: aXaXZ
+ 1: aX
+
+/^(?(DEFINE) (? a) (? b) )  (?&A) (?&B) /x
+    abcd
+ 0: ab
+
+/(?(?&NAME_PAT))\s+(?(?&ADDRESS_PAT))
+  (?(DEFINE)
+  (?[a-z]+)
+  (?\d+)
+  )/x
+    metcalfe 33
+ 0: metcalfe 33
+ 1: metcalfe
+ 2: 33
+
+/(?(DEFINE)(?2[0-4]\d|25[0-5]|1\d\d|[1-9]?\d))\b(?&byte)(\.(?&byte)){3}/
+    1.2.3.4
+ 0: 1.2.3.4
+ 1: 
+ 2: .4
+    131.111.10.206
+ 0: 131.111.10.206
+ 1: 
+ 2: .206
+    10.0.0.0
+ 0: 10.0.0.0
+ 1: 
+ 2: .0
+    ** Failers
+No match
+    10.6
+No match
+    455.3.4.5
+No match
+
+/\b(?&byte)(\.(?&byte)){3}(?(DEFINE)(?2[0-4]\d|25[0-5]|1\d\d|[1-9]?\d))/
+    1.2.3.4
+ 0: 1.2.3.4
+ 1: .4
+    131.111.10.206
+ 0: 131.111.10.206
+ 1: .206
+    10.0.0.0
+ 0: 10.0.0.0
+ 1: .0
+    ** Failers
+No match
+    10.6
+No match
+    455.3.4.5
+No match
+
+/^(\w++|\s++)*$/
+    now is the time for all good men to come to the aid of the party
+ 0: now is the time for all good men to come to the aid of the party
+ 1: party
+    *** Failers
+No match
+    this is not a line with only words and spaces!
+No match
+
+/(\d++)(\w)/
+    12345a
+ 0: 12345a
+ 1: 12345
+ 2: a
+    *** Failers
+No match
+    12345+
+No match
+
+/a++b/
+    aaab
+ 0: aaab
+
+/(a++b)/
+    aaab
+ 0: aaab
+ 1: aaab
+
+/(a++)b/
+    aaab
+ 0: aaab
+ 1: aaa
+
+/([^()]++|\([^()]*\))+/
+    ((abc(ade)ufh()()x
+ 0: abc(ade)ufh()()x
+ 1: x
+
+/\(([^()]++|\([^()]+\))+\)/
+    (abc)
+ 0: (abc)
+ 1: abc
+    (abc(def)xyz)
+ 0: (abc(def)xyz)
+ 1: xyz
+    *** Failers
+No match
+    ((()aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+No match
+
+/^([^()]|\((?1)*\))*$/
+    abc
+ 0: abc
+ 1: c
+    a(b)c
+ 0: a(b)c
+ 1: c
+    a(b(c))d
+ 0: a(b(c))d
+ 1: d
+    *** Failers)
+No match
+    a(b(c)d
+No match
+
+/^>abc>([^()]|\((?1)*\))*abc>123abc>123abc>1(2)3abc>1(2)3abc>(1(2)3)abc>(1(2)3)
+ 2: 
+ 3: Satanoscillatemymetallicsonatas
+ 4: S
+    AmanaplanacanalPanama
+ 0: AmanaplanacanalPanama
+ 1: 
+ 2: 
+ 3: AmanaplanacanalPanama
+ 4: A
+    AblewasIereIsawElba
+ 0: AblewasIereIsawElba
+ 1: 
+ 2: 
+ 3: AblewasIereIsawElba
+ 4: A
+    *** Failers
+No match
+    Thequickbrownfox
+No match
+
+/^(\d+|\((?1)([+*-])(?1)\)|-(?1))$/
+    12
+ 0: 12
+ 1: 12
+    (((2+2)*-3)-7)
+ 0: (((2+2)*-3)-7)
+ 1: (((2+2)*-3)-7)
+ 2: -
+    -12
+ 0: -12
+ 1: -12
+    *** Failers
+No match
+    ((2+2)*-3)-7)
+No match
+
+/^(x(y|(?1){2})z)/
+    xyz
+ 0: xyz
+ 1: xyz
+ 2: y
+    xxyzxyzz
+ 0: xxyzxyzz
+ 1: xxyzxyzz
+ 2: xyzxyz
+    *** Failers
+No match
+    xxyzz
+No match
+    xxyzxyzxyzz
+No match
+
+/((< (?: (?(R) \d++  | [^<>]*+) | (?2)) * >))/x
+    <>
+ 0: <>
+ 1: <>
+ 2: <>
+    
+ 0: 
+ 1: 
+ 2: 
+     hij>
+ 0:  hij>
+ 1:  hij>
+ 2:  hij>
+     hij>
+ 0: 
+ 1: 
+ 2: 
+    def>
+ 0: def>
+ 1: def>
+ 2: def>
+    
+ 0: <>
+ 1: <>
+ 2: <>
+    *** Failers
+No match
+    
+ 2: 
+ 3: Satan, oscillate my metallic sonatas
+ 4: S
+    A man, a plan, a canal: Panama!
+ 0: A man, a plan, a canal: Panama!
+ 1: 
+ 2: 
+ 3: A man, a plan, a canal: Panama
+ 4: A
+    Able was I ere I saw Elba.
+ 0: Able was I ere I saw Elba.
+ 1: 
+ 2: 
+ 3: Able was I ere I saw Elba
+ 4: A
+    *** Failers
+No match
+    The quick brown fox
+No match
+
+/^((.)(?1)\2|.)$/
+    a
+ 0: a
+ 1: a
+    aba
+ 0: aba
+ 1: aba
+ 2: a
+    aabaa  
+ 0: aabaa
+ 1: aabaa
+ 2: a
+    abcdcba 
+ 0: abcdcba
+ 1: abcdcba
+ 2: a
+    pqaabaaqp  
+ 0: pqaabaaqp
+ 1: pqaabaaqp
+ 2: p
+    ablewasiereisawelba
+ 0: ablewasiereisawelba
+ 1: ablewasiereisawelba
+ 2: a
+    rhubarb
+No match
+    the quick brown fox  
+No match
+
+/(a)(?<=b(?1))/
+    baz
+ 0: a
+ 1: a
+    ** Failers
+No match
+    caz  
+No match
+    
+/(?<=b(?1))(a)/
+    zbaaz
+ 0: a
+ 1: a
+    ** Failers
+No match
+    aaa  
+No match
+    
+/(?a)(?<=b(?&X))/
+    baz
+ 0: a
+ 1: a
+
+/^(?|(abc)|(def))\1/
+    abcabc
+ 0: abcabc
+ 1: abc
+    defdef 
+ 0: defdef
+ 1: def
+    ** Failers
+No match
+    abcdef
+No match
+    defabc   
+No match
+    
+/^(?|(abc)|(def))(?1)/
+    abcabc
+ 0: abcabc
+ 1: abc
+    defabc
+ 0: defabc
+ 1: def
+    ** Failers
+No match
+    defdef
+No match
+    abcdef    
+No match
+
+/(?:a(? (?')|(?")) |b(? (?')|(?")) ) (?('quote')[a-z]+|[0-9]+)/xJ
+    a\"aaaaa
+ 0: a"aaaaa
+ 1: "
+ 2: 
+ 3: "
+    b\"aaaaa 
+ 0: b"aaaaa
+ 1: 
+ 2: 
+ 3: 
+ 4: "
+ 5: 
+ 6: "
+    ** Failers 
+No match
+    b\"11111
+No match
+
+/(?:(?1)|B)(A(*F)|C)/
+    ABCD
+ 0: BC
+ 1: C
+    CCD
+ 0: CC
+ 1: C
+    ** Failers
+No match
+    CAD   
+No match
+
+/^(?:(?1)|B)(A(*F)|C)/
+    CCD
+ 0: CC
+ 1: C
+    BCD 
+ 0: BC
+ 1: C
+    ** Failers
+No match
+    ABCD
+No match
+    CAD
+No match
+    BAD    
+No match
+
+/(?:(?1)|B)(A(*ACCEPT)XX|C)D/
+    AAD
+ 0: AA
+ 1: A
+    ACD
+ 0: ACD
+ 1: C
+    BAD
+ 0: BA
+ 1: A
+    BCD
+ 0: BCD
+ 1: C
+    BAX  
+ 0: BA
+ 1: A
+    ** Failers
+No match
+    ACX
+No match
+    ABC   
+No match
+
+/(?(DEFINE)(A))B(?1)C/
+    BAC
+ 0: BAC
+
+/(?(DEFINE)((A)\2))B(?1)C/
+    BAAC
+ 0: BAAC
+
+/(? \( ( [^()]++ | (?&pn) )* \) )/x
+    (ab(cd)ef)
+ 0: (ab(cd)ef)
+ 1: (ab(cd)ef)
+ 2: ef
+
+/^(?=a(*SKIP)b|ac)/
+    ** Failers
+No match
+    ac
+No match
+    
+/^(?=a(*PRUNE)b)/
+    ab  
+ 0: 
+    ** Failers 
+No match
+    ac
+No match
+
+/^(?=a(*ACCEPT)b)/
+    ac
+ 0: 
+
+/(?>a\Kb)/
+    ab
+ 0: b
+
+/((?>a\Kb))/
+    ab
+ 0: b
+ 1: ab
+
+/(a\Kb)/
+    ab
+ 0: b
+ 1: ab
+    
+/^a\Kcz|ac/
+    ac
+ 0: ac
+    
+/(?>a\Kbz|ab)/
+    ab 
+ 0: ab
+
+/^(?&t)(?(DEFINE)(?a\Kb))$/
+    ab
+ 0: b
+
+/^([^()]|\((?1)*\))*$/
+    a(b)c
+ 0: a(b)c
+ 1: c
+    a(b(c)d)e 
+ 0: a(b(c)d)e
+ 1: e
+
+/(?P(?P0)(?P>L1)|(?P>L2))/
+    0
+ 0: 0
+ 1: 0
+    00
+ 0: 00
+ 1: 00
+ 2: 0
+    0000  
+ 0: 0000
+ 1: 0000
+ 2: 0
+
+/(?P(?P0)|(?P>L2)(?P>L1))/
+    0
+ 0: 0
+ 1: 0
+ 2: 0
+    00
+ 0: 0
+ 1: 0
+ 2: 0
+    0000  
+ 0: 0
+ 1: 0
+ 2: 0
+
+/--- This one does fail, as expected, in Perl. It needs the complex item at the
+     end of the pattern. A single letter instead of (B|D) makes it not fail,
+     which I think is a Perl bug. --- /
+
+/A(*COMMIT)(B|D)/
+    ACABX
+No match
+
+/--- Check the use of names for failure ---/
+
+/^(A(*PRUNE:A)B|C(*PRUNE:B)D)/K
+    ** Failers
+No match
+    AC
+No match, mark = A
+    CB    
+No match, mark = B
+    
+/--- Force no study, otherwise mark is not seen. The studied version is in
+     test 2 because it isn't Perl-compatible. ---/
+
+/(*MARK:A)(*SKIP:B)(C|X)/KSS
+    C
+ 0: C
+ 1: C
+MK: A
+    D
+No match, mark = A
+     
+/^(A(*THEN:A)B|C(*THEN:B)D)/K
+    ** Failers
+No match
+    CB    
+No match, mark = B
+
+/^(?:A(*THEN:A)B|C(*THEN:B)D)/K
+    CB    
+No match, mark = B
+    
+/^(?>A(*THEN:A)B|C(*THEN:B)D)/K
+    CB    
+No match, mark = B
+    
+/--- This should succeed, as the skip causes bump to offset 1 (the mark). Note
+that we have to have something complicated such as (B|Z) at the end because,
+for Perl, a simple character somehow causes an unwanted optimization to mess
+with the handling of backtracking verbs. ---/
+
+/A(*MARK:A)A+(*SKIP:A)(B|Z) | AC/xK
+    AAAC
+ 0: AC
+    
+/--- Test skipping over a non-matching mark. ---/
+
+/A(*MARK:A)A+(*MARK:B)(*SKIP:A)(B|Z) | AC/xK
+    AAAC
+ 0: AC
+    
+/--- Check shorthand for MARK ---/
+
+/A(*:A)A+(*SKIP:A)(B|Z) | AC/xK
+    AAAC
+ 0: AC
+
+/--- Don't loop! Force no study, otherwise mark is not seen. ---/
+
+/(*:A)A+(*SKIP:A)(B|Z)/KSS
+    AAAC
+No match, mark = A
+
+/--- This should succeed, as a non-existent skip name disables the skip ---/ 
+
+/A(*MARK:A)A+(*SKIP:B)(B|Z) | AC/xK
+    AAAC
+ 0: AC
+
+/A(*MARK:A)A+(*SKIP:B)(B|Z) | AC(*:B)/xK
+    AAAC
+ 0: AC
+MK: B
+
+/--- COMMIT at the start of a pattern should act like an anchor. Again, 
+however, we need the complication for Perl. ---/
+
+/(*COMMIT)(A|P)(B|P)(C|P)/
+    ABCDEFG
+ 0: ABC
+ 1: A
+ 2: B
+ 3: C
+    ** Failers
+No match
+    DEFGABC  
+No match
+
+/--- COMMIT inside an atomic group can't stop backtracking over the group. ---/
+
+/(\w+)(?>b(*COMMIT))\w{2}/
+    abbb
+ 0: abbb
+ 1: a
+
+/(\w+)b(*COMMIT)\w{2}/
+    abbb
+No match
+
+/--- Check opening parens in comment when seeking forward reference. ---/ 
+
+/(?&t)(?#()(?(DEFINE)(?a))/
+    bac
+ 0: a
+
+/--- COMMIT should override THEN ---/
+
+/(?>(*COMMIT)(?>yes|no)(*THEN)(*F))?/
+  yes
+No match
+
+/(?>(*COMMIT)(yes|no)(*THEN)(*F))?/
+  yes
+No match
+
+/b?(*SKIP)c/
+    bc
+ 0: bc
+    abc
+ 0: bc
+   
+/(*SKIP)bc/
+    a
+No match
+
+/(*SKIP)b/
+    a 
+No match
+
+/(?P(?P=abn)xxx|)+/
+    xxx
+ 0: 
+ 1: 
+
+/(?i:([^b]))(?1)/
+    aa
+ 0: aa
+ 1: a
+    aA     
+ 0: aA
+ 1: a
+    ** Failers
+ 0: **
+ 1: *
+    ab
+No match
+    aB
+No match
+    Ba
+No match
+    ba
+No match
+
+/^(?&t)*+(?(DEFINE)(?a))\w$/
+    aaaaaaX
+ 0: aaaaaaX
+    ** Failers 
+No match
+    aaaaaa 
+No match
+
+/^(?&t)*(?(DEFINE)(?a))\w$/
+    aaaaaaX
+ 0: aaaaaaX
+    aaaaaa 
+ 0: aaaaaa
+
+/^(a)*+(\w)/
+    aaaaX
+ 0: aaaaX
+ 1: a
+ 2: X
+    YZ 
+ 0: Y
+ 1: 
+ 2: Y
+    ** Failers 
+No match
+    aaaa
+No match
+
+/^(?:a)*+(\w)/
+    aaaaX
+ 0: aaaaX
+ 1: X
+    YZ 
+ 0: Y
+ 1: Y
+    ** Failers 
+No match
+    aaaa
+No match
+
+/^(a)++(\w)/
+    aaaaX
+ 0: aaaaX
+ 1: a
+ 2: X
+    ** Failers 
+No match
+    aaaa
+No match
+    YZ 
+No match
+
+/^(?:a)++(\w)/
+    aaaaX
+ 0: aaaaX
+ 1: X
+    ** Failers 
+No match
+    aaaa
+No match
+    YZ 
+No match
+
+/^(a)?+(\w)/
+    aaaaX
+ 0: aa
+ 1: a
+ 2: a
+    YZ 
+ 0: Y
+ 1: 
+ 2: Y
+
+/^(?:a)?+(\w)/
+    aaaaX
+ 0: aa
+ 1: a
+    YZ 
+ 0: Y
+ 1: Y
+
+/^(a){2,}+(\w)/
+    aaaaX
+ 0: aaaaX
+ 1: a
+ 2: X
+    ** Failers
+No match
+    aaa
+No match
+    YZ 
+No match
+
+/^(?:a){2,}+(\w)/
+    aaaaX
+ 0: aaaaX
+ 1: X
+    ** Failers
+No match
+    aaa
+No match
+    YZ 
+No match
+
+/(a|)*(?1)b/
+    b
+ 0: b
+ 1: 
+    ab
+ 0: ab
+ 1: 
+    aab  
+ 0: aab
+ 1: 
+
+/(a)++(?1)b/
+    ** Failers
+No match
+    ab 
+No match
+    aab
+No match
+
+/(a)*+(?1)b/
+    ** Failers
+No match
+    ab
+No match
+    aab  
+No match
+
+/(?1)(?:(b)){0}/
+    b
+ 0: b
+
+/(foo ( \( ((?:(?> [^()]+ )|(?2))*) \) ) )/x
+    foo(bar(baz)+baz(bop))
+ 0: foo(bar(baz)+baz(bop))
+ 1: foo(bar(baz)+baz(bop))
+ 2: (bar(baz)+baz(bop))
+ 3: bar(baz)+baz(bop)
+
+/(A (A|B(*ACCEPT)|C) D)(E)/x
+    AB
+ 0: AB
+ 1: AB
+ 2: B
+
+/\A.*?(a|bc)/
+    ba
+ 0: ba
+ 1: a
+
+/\A.*?(?:a|bc)++/
+    ba
+ 0: ba
+
+/\A.*?(a|bc)++/
+    ba
+ 0: ba
+ 1: a
+
+/\A.*?(?:a|bc|d)/
+    ba
+ 0: ba
+
+/(?:(b))++/
+    beetle
+ 0: b
+ 1: b
+
+/(?(?=(a(*ACCEPT)z))a)/
+    a
+ 0: a
+ 1: a
+
+/^(a)(?1)+ab/
+    aaaab
+ 0: aaaab
+ 1: a
+    
+/^(a)(?1)++ab/
+    aaaab
+No match
+
+/^(?=a(*:M))aZ/K
+    aZbc
+ 0: aZ
+MK: M
+
+/^(?!(*:M)b)aZ/K
+    aZbc
+ 0: aZ
+
+/(?(DEFINE)(a))?b(?1)/
+    backgammon
+ 0: ba
+
+/^\N+/
+    abc\ndef
+ 0: abc
+    
+/^\N{1,}/
+    abc\ndef 
+ 0: abc
+
+/(?(R)a+|(?R)b)/
+    aaaabcde
+ 0: aaaab
+
+/(?(R)a+|((?R))b)/
+    aaaabcde
+ 0: aaaab
+ 1: aaaa
+
+/((?(R)a+|(?1)b))/
+    aaaabcde
+ 0: aaaab
+ 1: aaaab
+
+/((?(R1)a+|(?1)b))/
+    aaaabcde
+ 0: aaaab
+ 1: aaaab
+
+/a(*:any 
+name)/K
+    abc
+ 0: a
+MK: any \x0aname
+    
+/(?>(?&t)c|(?&t))(?(DEFINE)(?a|b(*PRUNE)c))/
+    a
+ 0: a
+    ba
+ 0: a
+    bba 
+ 0: a
+    
+/--- Checking revised (*THEN) handling ---/ 
+
+/--- Capture ---/
+
+/^.*? (a(*THEN)b) c/x
+    aabc
+No match
+
+/^.*? (a(*THEN)b|(*F)) c/x
+    aabc
+ 0: aabc
+ 1: ab
+
+/^.*? ( (a(*THEN)b) | (*F) ) c/x
+    aabc
+ 0: aabc
+ 1: ab
+ 2: ab
+
+/^.*? ( (a(*THEN)b) ) c/x
+    aabc
+No match
+
+/--- Non-capture ---/
+
+/^.*? (?:a(*THEN)b) c/x
+    aabc
+No match
+
+/^.*? (?:a(*THEN)b|(*F)) c/x
+    aabc
+ 0: aabc
+
+/^.*? (?: (?:a(*THEN)b) | (*F) ) c/x
+    aabc
+ 0: aabc
+
+/^.*? (?: (?:a(*THEN)b) ) c/x
+    aabc
+No match
+
+/--- Atomic ---/
+
+/^.*? (?>a(*THEN)b) c/x
+    aabc
+No match
+
+/^.*? (?>a(*THEN)b|(*F)) c/x
+    aabc
+ 0: aabc
+
+/^.*? (?> (?>a(*THEN)b) | (*F) ) c/x
+    aabc
+ 0: aabc
+
+/^.*? (?> (?>a(*THEN)b) ) c/x
+    aabc
+No match
+
+/--- Possessive capture ---/
+
+/^.*? (a(*THEN)b)++ c/x
+    aabc
+No match
+
+/^.*? (a(*THEN)b|(*F))++ c/x
+    aabc
+ 0: aabc
+ 1: ab
+
+/^.*? ( (a(*THEN)b)++ | (*F) )++ c/x
+    aabc
+ 0: aabc
+ 1: ab
+ 2: ab
+
+/^.*? ( (a(*THEN)b)++ )++ c/x
+    aabc
+No match
+
+/--- Possessive non-capture ---/
+
+/^.*? (?:a(*THEN)b)++ c/x
+    aabc
+No match
+
+/^.*? (?:a(*THEN)b|(*F))++ c/x
+    aabc
+ 0: aabc
+
+/^.*? (?: (?:a(*THEN)b)++ | (*F) )++ c/x
+    aabc
+ 0: aabc
+
+/^.*? (?: (?:a(*THEN)b)++ )++ c/x
+    aabc
+No match
+    
+/--- Condition assertion ---/
+
+/^(?(?=a(*THEN)b)ab|ac)/
+    ac
+ 0: ac
+ 
+/--- Condition ---/
+
+/^.*?(?(?=a)a|b(*THEN)c)/
+    ba
+No match
+
+/^.*?(?:(?(?=a)a|b(*THEN)c)|d)/
+    ba
+ 0: ba
+
+/^.*?(?(?=a)a(*THEN)b|c)/
+    ac
+No match
+
+/--- Assertion ---/
+
+/^.*(?=a(*THEN)b)/ 
+    aabc
+ 0: a
+
+/------------------------------/
+
+/(?>a(*:m))/imsxSK 
+    a
+ 0: a
+MK: m
+
+/(?>(a)(*:m))/imsxSK 
+    a
+ 0: a
+ 1: a
+MK: m
+
+/(?<=a(*ACCEPT)b)c/
+    xacd
+ 0: c
+
+/(?<=(a(*ACCEPT)b))c/
+    xacd
+ 0: c
+ 1: a
+
+/(?<=(a(*COMMIT)b))c/
+    xabcd
+ 0: c
+ 1: ab
+    ** Failers 
+No match
+    xacd
+No match
+    
+/(?
+ 2: 
+
+/(another)?(\1+)test/
+    hello world test
+No match
+
+/(a(*COMMIT)b){0}a(?1)|aac/
+    aac
+ 0: aac
+
+/((?:a?)*)*c/
+  aac   
+ 0: aac
+ 1: 
+
+/((?>a?)*)*c/
+  aac   
+ 0: aac
+ 1: 
+
+/(?>.*?a)(?<=ba)/
+    aba
+ 0: ba
+
+/(?:.*?a)(?<=ba)/
+    aba
+ 0: aba
+
+/.*?a(*PRUNE)b/
+    aab
+ 0: ab
+
+/.*?a(*PRUNE)b/s
+    aab
+ 0: ab
+
+/^a(*PRUNE)b/s
+    aab
+No match
+
+/.*?a(*SKIP)b/
+    aab
+ 0: ab
+
+/(?>.*?a)b/s
+    aab
+ 0: ab
+
+/(?>.*?a)b/
+    aab
+ 0: ab
+
+/(?>^a)b/s
+    aab
+No match
+
+/(?>.*?)(?<=(abcd)|(wxyz))/
+    alphabetabcd
+ 0: 
+ 1: abcd
+    endingwxyz 
+ 0: 
+ 1: 
+ 2: wxyz
+
+/(?>.*)(?<=(abcd)|(wxyz))/
+    alphabetabcd
+ 0: alphabetabcd
+ 1: abcd
+    endingwxyz 
+ 0: endingwxyz
+ 1: 
+ 2: wxyz
+
+"(?>.*)foo"
+    abcdfooxyz
+No match
+    
+"(?>.*?)foo"
+    abcdfooxyz
+ 0: foo
+
+/(?:(a(*PRUNE)b)){0}(?:(?1)|ac)/
+    ac
+ 0: ac
+    
+/(?:(a(*SKIP)b)){0}(?:(?1)|ac)/
+    ac 
+ 0: ac
+
+/(?<=(*SKIP)ac)a/
+    aa
+No match
+
+/A(*MARK:A)A+(*SKIP:B)(B|Z) | AC/xK
+    AAAC
+ 0: AC
+
+/a(*SKIP:m)x|ac(*:n)(*SKIP:n)d|ac/K
+    acacd
+ 0: acd
+MK: n
+
+/A(*SKIP:m)x|A(*SKIP:n)x|AB/K
+    AB
+ 0: AB
+
+/((*SKIP:r)d){0}a(*SKIP:m)x|ac(*:n)|ac/K
+    acacd
+ 0: ac
+MK: n
+
+/-- Tests that try to figure out how Perl works. My hypothesis is that the
+    first verb that is backtracked onto is the one that acts. This seems to be
+    the case almost all the time, but there is one exception that is perhaps a 
+    bug. --/
+
+/-- This matches "aaaac"; each PRUNE advances one character until the subject
+    no longer starts with 5 'a's. --/
+
+/aaaaa(*PRUNE)b|a+c/
+    aaaaaac
+ 0: aaaac
+
+/-- Putting SKIP in front of PRUNE makes no difference, as it is never 
+backtracked onto, whether or not it has a label. --/
+
+/aaaaa(*SKIP)(*PRUNE)b|a+c/
+    aaaaaac
+ 0: aaaac
+
+/aaaaa(*SKIP:N)(*PRUNE)b|a+c/
+    aaaaaac
+ 0: aaaac
+
+/aaaa(*:N)a(*SKIP:N)(*PRUNE)b|a+c/
+    aaaaaac
+ 0: aaaac
+
+/-- Putting THEN in front makes no difference. */
+    
+/aaaaa(*THEN)(*PRUNE)b|a+c/
+    aaaaaac
+ 0: aaaac
+ 
+/-- However, putting COMMIT in front of the prune changes it to "no match". I 
+    think this is inconsistent and possibly a bug. For the moment, running this
+    test is moved out of the Perl-compatible file. --/
+
+/aaaaa(*COMMIT)(*PRUNE)b|a+c/
+    
+
+/---- OK, lets play the same game again using SKIP instead of PRUNE. ----/
+
+/-- This matches "ac" because SKIP forces the next match to start on the
+    sixth "a". --/
+
+/aaaaa(*SKIP)b|a+c/
+    aaaaaac
+ 0: ac
+ 
+/-- Putting PRUNE in front makes no difference. --/
+
+/aaaaa(*PRUNE)(*SKIP)b|a+c/
+    aaaaaac
+ 0: ac
+
+/-- Putting THEN in front makes no difference. --/
+
+/aaaaa(*THEN)(*SKIP)b|a+c/
+    aaaaaac
+ 0: ac
+
+/-- In this case, neither does COMMIT. This still matches "ac". --/
+
+/aaaaa(*COMMIT)(*SKIP)b|a+c/
+    aaaaaac
+ 0: ac
+    
+/-- This gives "no match", as expected. --/
+
+/aaaaa(*COMMIT)b|a+c/
+    aaaaaac
+No match
+    
+
+/------ Tests using THEN ------/
+
+/-- This matches "aaaaaac", as expected. --/
+
+/aaaaa(*THEN)b|a+c/
+    aaaaaac
+ 0: aaaaaac
+
+/-- Putting SKIP in front makes no difference. --/
+
+/aaaaa(*SKIP)(*THEN)b|a+c/
+    aaaaaac
+ 0: aaaaaac
+    
+/-- Putting PRUNE in front makes no difference. --/
+
+/aaaaa(*PRUNE)(*THEN)b|a+c/
+    aaaaaac
+ 0: aaaaaac
+    
+/-- Putting COMMIT in front makes no difference. --/
+
+/aaaaa(*COMMIT)(*THEN)b|a+c/
+    aaaaaac
+ 0: aaaaaac
+    
+/-- End of "priority" tests --/ 
+
+/aaaaa(*:m)(*PRUNE:m)(*SKIP:m)m|a+/
+    aaaaaa
+ 0: a
+
+/aaaaa(*:m)(*MARK:m)(*PRUNE)(*SKIP:m)m|a+/
+    aaaaaa
+ 0: a
+
+/aaaaa(*:n)(*PRUNE:m)(*SKIP:m)m|a+/
+    aaaaaa
+ 0: aaaa
+
+/aaaaa(*:n)(*MARK:m)(*PRUNE)(*SKIP:m)m|a+/
+    aaaaaa
+ 0: a
+
+/a(*MARK:A)aa(*PRUNE:A)a(*SKIP:A)b|a+c/
+    aaaac
+ 0: aac
+
+/a(*MARK:A)aa(*MARK:A)a(*SKIP:A)b|a+c/
+    aaaac
+ 0: ac
+
+/aaa(*PRUNE:A)a(*SKIP:A)b|a+c/
+    aaaac
+ 0: aac
+
+/aaa(*MARK:A)a(*SKIP:A)b|a+c/
+    aaaac
+ 0: ac
+
+/a(*:m)a(*COMMIT)(*SKIP:m)b|a+c/K
+    aaaaaac
+ 0: ac
+
+/.?(a|b(*THEN)c)/
+    ba
+ 0: ba
+ 1: a
+
+/(a(*COMMIT)b)c|abd/
+    abc
+ 0: abc
+ 1: ab
+    abd
+No match
+
+/(?=a(*COMMIT)b)abc|abd/
+    abc
+ 0: abc
+    abd
+ 0: abd
+
+/(?>a(*COMMIT)b)c|abd/
+    abc
+ 0: abc
+    abd
+ 0: abd
+
+/a(?=b(*COMMIT)c)[^d]|abd/
+    abd
+No match
+    abc 
+ 0: ab
+
+/a(?=bc).|abd/
+    abd
+ 0: abd
+    abc 
+ 0: ab
+    
+/a(?>b(*COMMIT)c)d|abd/
+    abceabd 
+No match
+
+/a(?>bc)d|abd/
+    abceabd 
+ 0: abd
+
+/(?>a(*COMMIT)b)c|abd/
+    abd
+ 0: abd
+
+/(?>a(*COMMIT)c)d|abd/
+    abd
+No match
+
+/((?=a(*COMMIT)b)ab|ac){0}(?:(?1)|a(c))/
+    ac 
+ 0: ac
+ 1: 
+ 2: c
+    
+/-- These tests were formerly in test 2, but changes in PCRE and Perl have
+    made them compatible. --/
+    
+/^(a)?(?(1)a|b)+$/
+    *** Failers
+No match
+    a
+No match
+
+/(?=a\Kb)ab/
+    ab 
+ 0: b
+
+/(?!a\Kb)ac/
+    ac 
+ 0: ac
+    
+/^abc(?<=b\Kc)d/
+    abcd
+ 0: cd
+
+/^abc(?b))/K
+    abc
+ 0: b
+MK: m(m
+
+/(*PRUNE:m(m)(?&y)(?(DEFINE)(?b))/K
+    abc
+ 0: b
+MK: m(m
+
+/(*SKIP:m(m)(?&y)(?(DEFINE)(?b))/K
+    abc
+ 0: b
+
+/(*THEN:m(m)(?&y)(?(DEFINE)(?b))/K
+    abc
+ 0: b
+MK: m(m
+
+/^\d*\w{4}/
+    1234
+ 0: 1234
+    123 
+No match
+
+/^[^b]*\w{4}/
+    aaaa
+ 0: aaaa
+    aaa     
+No match
+
+/^[^b]*\w{4}/i
+    aaaa
+ 0: aaaa
+    aaa     
+No match
+
+/^a*\w{4}/
+    aaaa
+ 0: aaaa
+    aaa     
+No match
+
+/^a*\w{4}/i
+    aaaa
+ 0: aaaa
+    aaa     
+No match
+
+/-- End of testinput1 --/
diff --git a/pcre/testdata/testoutput10 b/pcre/testdata/testoutput10
new file mode 100644
index 00000000000..0e04205bb88
--- /dev/null
+++ b/pcre/testdata/testoutput10
@@ -0,0 +1,2851 @@
+/-- This set of tests check Unicode property support with the DFA matching 
+    functionality of pcre_dfa_exec(). The -dfa flag must be used with pcretest
+    when running it. --/
+
+/\pL\P{Nd}/8
+    AB
+ 0: AB
+    *** Failers
+ 0: Fa
+    A0
+No match
+    00   
+No match
+
+/\X./8
+    AB
+ 0: AB
+    A\x{300}BC 
+ 0: A\x{300}B
+    A\x{300}\x{301}\x{302}BC 
+ 0: A\x{300}\x{301}\x{302}B
+    *** Failers
+ 0: **
+    \x{300}  
+No match
+
+/\X\X/8
+    ABC
+ 0: AB
+    A\x{300}B\x{300}\x{301}C 
+ 0: A\x{300}B\x{300}\x{301}
+    A\x{300}\x{301}\x{302}BC 
+ 0: A\x{300}\x{301}\x{302}B
+    *** Failers
+ 0: **
+    \x{300}  
+No match
+
+/^\pL+/8
+    abcd
+ 0: abcd
+ 1: abc
+ 2: ab
+ 3: a
+    a 
+ 0: a
+    *** Failers 
+No match
+
+/^\PL+/8
+    1234
+ 0: 1234
+ 1: 123
+ 2: 12
+ 3: 1
+    = 
+ 0: =
+    *** Failers 
+ 0: *** 
+ 1: ***
+ 2: **
+ 3: *
+    abcd 
+No match
+
+/^\X+/8
+    abcdA\x{300}\x{301}\x{302}
+ 0: abcdA\x{300}\x{301}\x{302}
+ 1: abcd
+ 2: abc
+ 3: ab
+ 4: a
+    A\x{300}\x{301}\x{302}
+ 0: A\x{300}\x{301}\x{302}
+    A\x{300}\x{301}\x{302}A\x{300}\x{301}\x{302}
+ 0: A\x{300}\x{301}\x{302}A\x{300}\x{301}\x{302}
+ 1: A\x{300}\x{301}\x{302}
+    a 
+ 0: a
+    *** Failers 
+ 0: *** Failers
+ 1: *** Failer
+ 2: *** Faile
+ 3: *** Fail
+ 4: *** Fai
+ 5: *** Fa
+ 6: *** F
+ 7: *** 
+ 8: ***
+ 9: **
+10: *
+    \x{300}\x{301}\x{302}
+ 0: \x{300}\x{301}\x{302}
+
+/\X?abc/8
+    abc
+ 0: abc
+    A\x{300}abc
+ 0: A\x{300}abc
+    A\x{300}\x{301}\x{302}A\x{300}A\x{300}A\x{300}abcxyz
+ 0: A\x{300}abc
+    \x{300}abc  
+ 0: \x{300}abc
+    *** Failers
+No match
+
+/^\X?abc/8
+    abc
+ 0: abc
+    A\x{300}abc
+ 0: A\x{300}abc
+    *** Failers
+No match
+    A\x{300}\x{301}\x{302}A\x{300}A\x{300}A\x{300}abcxyz
+No match
+    \x{300}abc  
+ 0: \x{300}abc
+
+/\X*abc/8
+    abc
+ 0: abc
+    A\x{300}abc
+ 0: A\x{300}abc
+    A\x{300}\x{301}\x{302}A\x{300}A\x{300}A\x{300}abcxyz
+ 0: A\x{300}\x{301}\x{302}A\x{300}A\x{300}A\x{300}abc
+    \x{300}abc  
+ 0: \x{300}abc
+    *** Failers
+No match
+
+/^\X*abc/8
+    abc
+ 0: abc
+    A\x{300}abc
+ 0: A\x{300}abc
+    A\x{300}\x{301}\x{302}A\x{300}A\x{300}A\x{300}abcxyz
+ 0: A\x{300}\x{301}\x{302}A\x{300}A\x{300}A\x{300}abc
+    *** Failers
+No match
+    \x{300}abc  
+ 0: \x{300}abc
+
+/^\pL?=./8
+    A=b
+ 0: A=b
+    =c 
+ 0: =c
+    *** Failers
+No match
+    1=2 
+No match
+    AAAA=b  
+No match
+
+/^\pL*=./8
+    AAAA=b
+ 0: AAAA=b
+    =c 
+ 0: =c
+    *** Failers
+No match
+    1=2  
+No match
+
+/^\X{2,3}X/8
+    A\x{300}\x{301}\x{302}A\x{300}\x{301}\x{302}X
+ 0: A\x{300}\x{301}\x{302}A\x{300}\x{301}\x{302}X
+    A\x{300}\x{301}\x{302}A\x{300}\x{301}\x{302}A\x{300}\x{301}\x{302}X 
+ 0: A\x{300}\x{301}\x{302}A\x{300}\x{301}\x{302}A\x{300}\x{301}\x{302}X
+    *** Failers
+No match
+    X
+No match
+    A\x{300}\x{301}\x{302}X
+No match
+    A\x{300}\x{301}\x{302}A\x{300}\x{301}\x{302}A\x{300}\x{301}\x{302}A\x{300}\x{301}\x{302}X
+No match
+
+/^\pC\pL\pM\pN\pP\pS\pZ\p{Xsp}/8
+    >\x{1680}\x{2028}\x{0b}
+ 0: >\x{1680}
+    ** Failers
+No match
+    \x{0b} 
+No match
+
+/^>\p{Xsp}+/8
+    > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+ 0: > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}
+ 1: > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}
+ 2: > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}
+ 3: > \x{09}\x{0a}\x{0c}\x{0d}
+ 4: > \x{09}\x{0a}\x{0c}
+ 5: > \x{09}\x{0a}
+ 6: > \x{09}
+ 7: > 
+
+/^>\p{Xsp}*/8
+    > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+ 0: > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}
+ 1: > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}
+ 2: > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}
+ 3: > \x{09}\x{0a}\x{0c}\x{0d}
+ 4: > \x{09}\x{0a}\x{0c}
+ 5: > \x{09}\x{0a}
+ 6: > \x{09}
+ 7: > 
+ 8: >
+    
+/^>\p{Xsp}{2,9}/8
+    > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+ 0: > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}
+ 1: > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}
+ 2: > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}
+ 3: > \x{09}\x{0a}\x{0c}\x{0d}
+ 4: > \x{09}\x{0a}\x{0c}
+ 5: > \x{09}\x{0a}
+ 6: > \x{09}
+    
+/^>[\p{Xsp}]/8
+    >\x{2028}\x{0b}
+ 0: >\x{2028}
+ 
+/^>[\p{Xsp}]+/8
+    > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+ 0: > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}
+ 1: > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}
+ 2: > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}
+ 3: > \x{09}\x{0a}\x{0c}\x{0d}
+ 4: > \x{09}\x{0a}\x{0c}
+ 5: > \x{09}\x{0a}
+ 6: > \x{09}
+ 7: > 
+
+/^>\p{Xps}/8
+    >\x{1680}\x{2028}\x{0b}
+ 0: >\x{1680}
+    >\x{a0} 
+ 0: >\x{a0}
+    ** Failers
+No match
+    \x{0b} 
+No match
+
+/^>\p{Xps}+/8
+    > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+ 0: > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+ 1: > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}
+ 2: > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}
+ 3: > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}
+ 4: > \x{09}\x{0a}\x{0c}\x{0d}
+ 5: > \x{09}\x{0a}\x{0c}
+ 6: > \x{09}\x{0a}
+ 7: > \x{09}
+ 8: > 
+
+/^>\p{Xps}+?/8
+    >\x{1680}\x{2028}\x{0b}
+ 0: >\x{1680}\x{2028}\x{0b}
+ 1: >\x{1680}\x{2028}
+ 2: >\x{1680}
+
+/^>\p{Xps}*/8
+    > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+ 0: > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+ 1: > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}
+ 2: > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}
+ 3: > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}
+ 4: > \x{09}\x{0a}\x{0c}\x{0d}
+ 5: > \x{09}\x{0a}\x{0c}
+ 6: > \x{09}\x{0a}
+ 7: > \x{09}
+ 8: > 
+ 9: >
+    
+/^>\p{Xps}{2,9}/8
+    > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+ 0: > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+ 1: > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}
+ 2: > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}
+ 3: > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}
+ 4: > \x{09}\x{0a}\x{0c}\x{0d}
+ 5: > \x{09}\x{0a}\x{0c}
+ 6: > \x{09}\x{0a}
+ 7: > \x{09}
+    
+/^>\p{Xps}{2,9}?/8
+    > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+ 0: > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+ 1: > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}
+ 2: > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}
+ 3: > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}
+ 4: > \x{09}\x{0a}\x{0c}\x{0d}
+ 5: > \x{09}\x{0a}\x{0c}
+ 6: > \x{09}\x{0a}
+ 7: > \x{09}
+    
+/^>[\p{Xps}]/8
+    >\x{2028}\x{0b}
+ 0: >\x{2028}
+ 
+/^>[\p{Xps}]+/8
+    > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+ 0: > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+ 1: > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}
+ 2: > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}
+ 3: > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}
+ 4: > \x{09}\x{0a}\x{0c}\x{0d}
+ 5: > \x{09}\x{0a}\x{0c}
+ 6: > \x{09}\x{0a}
+ 7: > \x{09}
+ 8: > 
+
+/^\p{Xwd}/8
+    ABCD
+ 0: A
+    1234
+ 0: 1
+    \x{6ca}
+ 0: \x{6ca}
+    \x{a6c}
+ 0: \x{a6c}
+    \x{10a7}
+ 0: \x{10a7}
+    _ABC    
+ 0: _
+    ** Failers
+No match
+    [] 
+No match
+
+/^\p{Xwd}+/8
+    ABCD1234\x{6ca}\x{a6c}\x{10a7}_
+ 0: ABCD1234\x{6ca}\x{a6c}\x{10a7}_
+ 1: ABCD1234\x{6ca}\x{a6c}\x{10a7}
+ 2: ABCD1234\x{6ca}\x{a6c}
+ 3: ABCD1234\x{6ca}
+ 4: ABCD1234
+ 5: ABCD123
+ 6: ABCD12
+ 7: ABCD1
+ 8: ABCD
+ 9: ABC
+10: AB
+11: A
+
+/^\p{Xwd}*/8
+    ABCD1234\x{6ca}\x{a6c}\x{10a7}_
+ 0: ABCD1234\x{6ca}\x{a6c}\x{10a7}_
+ 1: ABCD1234\x{6ca}\x{a6c}\x{10a7}
+ 2: ABCD1234\x{6ca}\x{a6c}
+ 3: ABCD1234\x{6ca}
+ 4: ABCD1234
+ 5: ABCD123
+ 6: ABCD12
+ 7: ABCD1
+ 8: ABCD
+ 9: ABC
+10: AB
+11: A
+12: 
+    
+/^\p{Xwd}{2,9}/8
+    A_12\x{6ca}\x{a6c}\x{10a7}
+ 0: A_12\x{6ca}\x{a6c}\x{10a7}
+ 1: A_12\x{6ca}\x{a6c}
+ 2: A_12\x{6ca}
+ 3: A_12
+ 4: A_1
+ 5: A_
+    
+/^[\p{Xwd}]/8
+    ABCD1234_
+ 0: A
+    1234abcd_
+ 0: 1
+    \x{6ca}
+ 0: \x{6ca}
+    \x{a6c}
+ 0: \x{a6c}
+    \x{10a7}   
+ 0: \x{10a7}
+    _ABC 
+ 0: _
+    ** Failers
+No match
+    []   
+No match
+ 
+/^[\p{Xwd}]+/8
+    ABCD1234\x{6ca}\x{a6c}\x{10a7}_
+ 0: ABCD1234\x{6ca}\x{a6c}\x{10a7}_
+ 1: ABCD1234\x{6ca}\x{a6c}\x{10a7}
+ 2: ABCD1234\x{6ca}\x{a6c}
+ 3: ABCD1234\x{6ca}
+ 4: ABCD1234
+ 5: ABCD123
+ 6: ABCD12
+ 7: ABCD1
+ 8: ABCD
+ 9: ABC
+10: AB
+11: A
+
+/-- Unicode properties for \b abd \B --/
+
+/\b...\B/8W
+    abc_
+ 0: abc
+    \x{37e}abc\x{376} 
+ 0: abc
+    \x{37e}\x{376}\x{371}\x{393}\x{394} 
+ 0: \x{376}\x{371}\x{393}
+    !\x{c0}++\x{c1}\x{c2} 
+ 0: ++\x{c1}
+    !\x{c0}+++++ 
+ 0: \x{c0}++
+
+/-- Without PCRE_UCP, non-ASCII always fail, even if < 256  --/
+
+/\b...\B/8
+    abc_
+ 0: abc
+    ** Failers 
+ 0: Fai
+    \x{37e}abc\x{376} 
+No match
+    \x{37e}\x{376}\x{371}\x{393}\x{394} 
+No match
+    !\x{c0}++\x{c1}\x{c2} 
+No match
+    !\x{c0}+++++ 
+No match
+
+/-- With PCRE_UCP, non-UTF8 chars that are < 256 still check properties  --/
+
+/\b...\B/W
+    abc_
+ 0: abc
+    !\x{c0}++\x{c1}\x{c2} 
+ 0: ++\xc1
+    !\x{c0}+++++ 
+ 0: \xc0++
+    
+/-- Caseless single negated characters > 127 need UCP support --/
+
+/[^\x{100}]/8i
+    \x{100}\x{101}X
+ 0: X
+
+/[^\x{100}]+/8i
+    \x{100}\x{101}XX
+ 0: XX
+ 1: X
+
+/^\X/8
+    A\P
+ 0: A
+    A\P\P 
+Partial match: A
+    A\x{300}\x{301}\P
+ 0: A\x{300}\x{301}
+    A\x{300}\x{301}\P\P  
+Partial match: A\x{300}\x{301}
+    A\x{301}\P
+ 0: A\x{301}
+    A\x{301}\P\P  
+Partial match: A\x{301}
+    
+/^\X{2,3}/8
+    A\P
+Partial match: A
+    A\P\P 
+Partial match: A
+    AA\P
+ 0: AA
+    AA\P\P  
+Partial match: AA
+    A\x{300}\x{301}\P
+Partial match: A\x{300}\x{301}
+    A\x{300}\x{301}\P\P  
+Partial match: A\x{300}\x{301}
+    A\x{300}\x{301}A\x{300}\x{301}\P
+ 0: A\x{300}\x{301}A\x{300}\x{301}
+    A\x{300}\x{301}A\x{300}\x{301}\P\P  
+Partial match: A\x{300}\x{301}A\x{300}\x{301}
+
+/^\X{2}/8
+    AA\P
+ 0: AA
+    AA\P\P  
+Partial match: AA
+    A\x{300}\x{301}A\x{300}\x{301}\P
+ 0: A\x{300}\x{301}A\x{300}\x{301}
+    A\x{300}\x{301}A\x{300}\x{301}\P\P  
+Partial match: A\x{300}\x{301}A\x{300}\x{301}
+    
+/^\X+/8
+    AA\P
+ 0: AA
+ 1: A
+    AA\P\P  
+Partial match: AA
+
+/^\X+?Z/8
+    AA\P
+Partial match: AA
+    AA\P\P 
+Partial match: AA
+
+/-- These are tests for extended grapheme clusters --/ 
+
+/^\X/8+
+    G\x{34e}\x{34e}X
+ 0: G\x{34e}\x{34e}
+ 0+ X
+    \x{34e}\x{34e}X
+ 0: \x{34e}\x{34e}
+ 0+ X
+    \x04X
+ 0: \x{04}
+ 0+ X
+    \x{1100}X
+ 0: \x{1100}
+ 0+ X
+    \x{1100}\x{34e}X
+ 0: \x{1100}\x{34e}
+ 0+ X
+    \x{1b04}\x{1b04}X 
+ 0: \x{1b04}\x{1b04}
+ 0+ X
+    *These match up to the roman letters
+ 0: *
+ 0+ These match up to the roman letters
+    \x{1111}\x{1111}L,L
+ 0: \x{1111}\x{1111}
+ 0+ L,L
+    \x{1111}\x{1111}\x{1169}L,L,V
+ 0: \x{1111}\x{1111}\x{1169}
+ 0+ L,L,V
+    \x{1111}\x{ae4c}L, LV
+ 0: \x{1111}\x{ae4c}
+ 0+ L, LV
+    \x{1111}\x{ad89}L, LVT
+ 0: \x{1111}\x{ad89}
+ 0+ L, LVT
+    \x{1111}\x{ae4c}\x{1169}L, LV, V
+ 0: \x{1111}\x{ae4c}\x{1169}
+ 0+ L, LV, V
+    \x{1111}\x{ae4c}\x{1169}\x{1169}L, LV, V, V
+ 0: \x{1111}\x{ae4c}\x{1169}\x{1169}
+ 0+ L, LV, V, V
+    \x{1111}\x{ae4c}\x{1169}\x{11fe}L, LV, V, T
+ 0: \x{1111}\x{ae4c}\x{1169}\x{11fe}
+ 0+ L, LV, V, T
+    \x{1111}\x{ad89}\x{11fe}L, LVT, T
+ 0: \x{1111}\x{ad89}\x{11fe}
+ 0+ L, LVT, T
+    \x{1111}\x{ad89}\x{11fe}\x{11fe}L, LVT, T, T
+ 0: \x{1111}\x{ad89}\x{11fe}\x{11fe}
+ 0+ L, LVT, T, T
+    \x{ad89}\x{11fe}\x{11fe}LVT, T, T
+ 0: \x{ad89}\x{11fe}\x{11fe}
+ 0+ LVT, T, T
+    *These match just the first codepoint (invalid sequence)
+ 0: *
+ 0+ These match just the first codepoint (invalid sequence)
+    \x{1111}\x{11fe}L, T
+ 0: \x{1111}
+ 0+ \x{11fe}L, T
+    \x{ae4c}\x{1111}LV, L
+ 0: \x{ae4c}
+ 0+ \x{1111}LV, L
+    \x{ae4c}\x{ae4c}LV, LV
+ 0: \x{ae4c}
+ 0+ \x{ae4c}LV, LV
+    \x{ae4c}\x{ad89}LV, LVT
+ 0: \x{ae4c}
+ 0+ \x{ad89}LV, LVT
+    \x{1169}\x{1111}V, L
+ 0: \x{1169}
+ 0+ \x{1111}V, L
+    \x{1169}\x{ae4c}V, LV
+ 0: \x{1169}
+ 0+ \x{ae4c}V, LV
+    \x{1169}\x{ad89}V, LVT
+ 0: \x{1169}
+ 0+ \x{ad89}V, LVT
+    \x{ad89}\x{1111}LVT, L
+ 0: \x{ad89}
+ 0+ \x{1111}LVT, L
+    \x{ad89}\x{1169}LVT, V
+ 0: \x{ad89}
+ 0+ \x{1169}LVT, V
+    \x{ad89}\x{ae4c}LVT, LV
+ 0: \x{ad89}
+ 0+ \x{ae4c}LVT, LV
+    \x{ad89}\x{ad89}LVT, LVT
+ 0: \x{ad89}
+ 0+ \x{ad89}LVT, LVT
+    \x{11fe}\x{1111}T, L
+ 0: \x{11fe}
+ 0+ \x{1111}T, L
+    \x{11fe}\x{1169}T, V
+ 0: \x{11fe}
+ 0+ \x{1169}T, V
+    \x{11fe}\x{ae4c}T, LV
+ 0: \x{11fe}
+ 0+ \x{ae4c}T, LV
+    \x{11fe}\x{ad89}T, LVT
+ 0: \x{11fe}
+ 0+ \x{ad89}T, LVT
+    *Test extend and spacing mark
+ 0: *
+ 0+ Test extend and spacing mark
+    \x{1111}\x{ae4c}\x{0711}L, LV, extend
+ 0: \x{1111}\x{ae4c}\x{711}
+ 0+ L, LV, extend
+    \x{1111}\x{ae4c}\x{1b04}L, LV, spacing mark
+ 0: \x{1111}\x{ae4c}\x{1b04}
+ 0+ L, LV, spacing mark
+    \x{1111}\x{ae4c}\x{1b04}\x{0711}\x{1b04}L, LV, spacing mark, extend, spacing mark
+ 0: \x{1111}\x{ae4c}\x{1b04}\x{711}\x{1b04}
+ 0+ L, LV, spacing mark, extend, spacing mark
+    *Test CR, LF, and control
+ 0: *
+ 0+ Test CR, LF, and control
+    \x0d\x{0711}CR, extend
+ 0: \x{0d}
+ 0+ \x{711}CR, extend
+    \x0d\x{1b04}CR, spacingmark
+ 0: \x{0d}
+ 0+ \x{1b04}CR, spacingmark
+    \x0a\x{0711}LF, extend
+ 0: \x{0a}
+ 0+ \x{711}LF, extend
+    \x0a\x{1b04}LF, spacingmark
+ 0: \x{0a}
+ 0+ \x{1b04}LF, spacingmark
+    \x0b\x{0711}Control, extend
+ 0: \x{0b}
+ 0+ \x{711}Control, extend
+    \x09\x{1b04}Control, spacingmark
+ 0: \x{09}
+ 0+ \x{1b04}Control, spacingmark
+    *There are no Prepend characters, so we can't test Prepend, CR
+ 0: *
+ 0+ There are no Prepend characters, so we can't test Prepend, CR
+    
+/^(?>\X{2})X/8+
+    \x{1111}\x{ae4c}\x{1111}\x{ae4c}X
+ 0: \x{1111}\x{ae4c}\x{1111}\x{ae4c}X
+ 0+ 
+    
+/^\X{2,4}X/8+
+    \x{1111}\x{ae4c}\x{1111}\x{ae4c}X
+ 0: \x{1111}\x{ae4c}\x{1111}\x{ae4c}X
+ 0+ 
+    \x{1111}\x{ae4c}\x{1111}\x{ae4c}\x{1111}\x{ae4c}X
+ 0: \x{1111}\x{ae4c}\x{1111}\x{ae4c}\x{1111}\x{ae4c}X
+ 0+ 
+    \x{1111}\x{ae4c}\x{1111}\x{ae4c}\x{1111}\x{ae4c}\x{1111}\x{ae4c}X
+ 0: \x{1111}\x{ae4c}\x{1111}\x{ae4c}\x{1111}\x{ae4c}\x{1111}\x{ae4c}X
+ 0+ 
+
+/^\X{2,4}?X/8+
+    \x{1111}\x{ae4c}\x{1111}\x{ae4c}X
+ 0: \x{1111}\x{ae4c}\x{1111}\x{ae4c}X
+ 0+ 
+    \x{1111}\x{ae4c}\x{1111}\x{ae4c}\x{1111}\x{ae4c}X
+ 0: \x{1111}\x{ae4c}\x{1111}\x{ae4c}\x{1111}\x{ae4c}X
+ 0+ 
+    \x{1111}\x{ae4c}\x{1111}\x{ae4c}\x{1111}\x{ae4c}\x{1111}\x{ae4c}X
+ 0: \x{1111}\x{ae4c}\x{1111}\x{ae4c}\x{1111}\x{ae4c}\x{1111}\x{ae4c}X
+ 0+ 
+
+/-- --/
+
+/\x{1e9e}+/8i
+    \x{1e9e}\x{00df}
+ 0: \x{1e9e}\x{df}
+ 1: \x{1e9e}
+
+/[z\x{1e9e}]+/8i
+    \x{1e9e}\x{00df}
+ 0: \x{1e9e}\x{df}
+ 1: \x{1e9e}
+
+/\x{00df}+/8i
+    \x{1e9e}\x{00df}
+ 0: \x{1e9e}\x{df}
+ 1: \x{1e9e}
+
+/[z\x{00df}]+/8i
+    \x{1e9e}\x{00df}
+ 0: \x{1e9e}\x{df}
+ 1: \x{1e9e}
+
+/\x{1f88}+/8i
+    \x{1f88}\x{1f80} 
+ 0: \x{1f88}\x{1f80}
+ 1: \x{1f88}
+
+/[z\x{1f88}]+/8i
+    \x{1f88}\x{1f80} 
+ 0: \x{1f88}\x{1f80}
+ 1: \x{1f88}
+
+/-- Perl matches these --/
+
+/\x{00b5}+/8i
+    \x{00b5}\x{039c}\x{03bc}
+ 0: \x{b5}\x{39c}\x{3bc}
+ 1: \x{b5}\x{39c}
+ 2: \x{b5}
+
+/\x{039c}+/8i
+    \x{00b5}\x{039c}\x{03bc}
+ 0: \x{b5}\x{39c}\x{3bc}
+ 1: \x{b5}\x{39c}
+ 2: \x{b5}
+
+/\x{03bc}+/8i
+    \x{00b5}\x{039c}\x{03bc}
+ 0: \x{b5}\x{39c}\x{3bc}
+ 1: \x{b5}\x{39c}
+ 2: \x{b5}
+
+
+/\x{00c5}+/8i
+    \x{00c5}\x{00e5}\x{212b}
+ 0: \x{c5}\x{e5}\x{212b}
+ 1: \x{c5}\x{e5}
+ 2: \x{c5}
+
+/\x{00e5}+/8i
+    \x{00c5}\x{00e5}\x{212b}
+ 0: \x{c5}\x{e5}\x{212b}
+ 1: \x{c5}\x{e5}
+ 2: \x{c5}
+
+/\x{212b}+/8i
+    \x{00c5}\x{00e5}\x{212b}
+ 0: \x{c5}\x{e5}\x{212b}
+ 1: \x{c5}\x{e5}
+ 2: \x{c5}
+
+
+/\x{01c4}+/8i
+    \x{01c4}\x{01c5}\x{01c6}
+ 0: \x{1c4}\x{1c5}\x{1c6}
+ 1: \x{1c4}\x{1c5}
+ 2: \x{1c4}
+
+/\x{01c5}+/8i
+    \x{01c4}\x{01c5}\x{01c6}
+ 0: \x{1c4}\x{1c5}\x{1c6}
+ 1: \x{1c4}\x{1c5}
+ 2: \x{1c4}
+
+/\x{01c6}+/8i
+    \x{01c4}\x{01c5}\x{01c6}
+ 0: \x{1c4}\x{1c5}\x{1c6}
+ 1: \x{1c4}\x{1c5}
+ 2: \x{1c4}
+
+
+/\x{01c7}+/8i
+    \x{01c7}\x{01c8}\x{01c9}
+ 0: \x{1c7}\x{1c8}\x{1c9}
+ 1: \x{1c7}\x{1c8}
+ 2: \x{1c7}
+
+/\x{01c8}+/8i
+    \x{01c7}\x{01c8}\x{01c9}
+ 0: \x{1c7}\x{1c8}\x{1c9}
+ 1: \x{1c7}\x{1c8}
+ 2: \x{1c7}
+
+/\x{01c9}+/8i
+    \x{01c7}\x{01c8}\x{01c9}
+ 0: \x{1c7}\x{1c8}\x{1c9}
+ 1: \x{1c7}\x{1c8}
+ 2: \x{1c7}
+
+
+/\x{01ca}+/8i
+    \x{01ca}\x{01cb}\x{01cc}
+ 0: \x{1ca}\x{1cb}\x{1cc}
+ 1: \x{1ca}\x{1cb}
+ 2: \x{1ca}
+
+/\x{01cb}+/8i
+    \x{01ca}\x{01cb}\x{01cc}
+ 0: \x{1ca}\x{1cb}\x{1cc}
+ 1: \x{1ca}\x{1cb}
+ 2: \x{1ca}
+
+/\x{01cc}+/8i
+    \x{01ca}\x{01cb}\x{01cc}
+ 0: \x{1ca}\x{1cb}\x{1cc}
+ 1: \x{1ca}\x{1cb}
+ 2: \x{1ca}
+
+
+/\x{01f1}+/8i
+    \x{01f1}\x{01f2}\x{01f3}
+ 0: \x{1f1}\x{1f2}\x{1f3}
+ 1: \x{1f1}\x{1f2}
+ 2: \x{1f1}
+
+/\x{01f2}+/8i
+    \x{01f1}\x{01f2}\x{01f3}
+ 0: \x{1f1}\x{1f2}\x{1f3}
+ 1: \x{1f1}\x{1f2}
+ 2: \x{1f1}
+
+/\x{01f3}+/8i
+    \x{01f1}\x{01f2}\x{01f3}
+ 0: \x{1f1}\x{1f2}\x{1f3}
+ 1: \x{1f1}\x{1f2}
+ 2: \x{1f1}
+
+
+/\x{0345}+/8i
+    \x{0345}\x{0399}\x{03b9}\x{1fbe}
+ 0: \x{345}\x{399}\x{3b9}\x{1fbe}
+ 1: \x{345}\x{399}\x{3b9}
+ 2: \x{345}\x{399}
+ 3: \x{345}
+
+/\x{0399}+/8i
+    \x{0345}\x{0399}\x{03b9}\x{1fbe}
+ 0: \x{345}\x{399}\x{3b9}\x{1fbe}
+ 1: \x{345}\x{399}\x{3b9}
+ 2: \x{345}\x{399}
+ 3: \x{345}
+
+/\x{03b9}+/8i
+    \x{0345}\x{0399}\x{03b9}\x{1fbe}
+ 0: \x{345}\x{399}\x{3b9}\x{1fbe}
+ 1: \x{345}\x{399}\x{3b9}
+ 2: \x{345}\x{399}
+ 3: \x{345}
+
+/\x{1fbe}+/8i
+    \x{0345}\x{0399}\x{03b9}\x{1fbe}
+ 0: \x{345}\x{399}\x{3b9}\x{1fbe}
+ 1: \x{345}\x{399}\x{3b9}
+ 2: \x{345}\x{399}
+ 3: \x{345}
+
+
+/\x{0392}+/8i
+    \x{0392}\x{03b2}\x{03d0}
+ 0: \x{392}\x{3b2}\x{3d0}
+ 1: \x{392}\x{3b2}
+ 2: \x{392}
+
+/\x{03b2}+/8i
+    \x{0392}\x{03b2}\x{03d0}
+ 0: \x{392}\x{3b2}\x{3d0}
+ 1: \x{392}\x{3b2}
+ 2: \x{392}
+
+/\x{03d0}+/8i
+    \x{0392}\x{03b2}\x{03d0}
+ 0: \x{392}\x{3b2}\x{3d0}
+ 1: \x{392}\x{3b2}
+ 2: \x{392}
+    
+
+/\x{0395}+/8i
+    \x{0395}\x{03b5}\x{03f5}
+ 0: \x{395}\x{3b5}\x{3f5}
+ 1: \x{395}\x{3b5}
+ 2: \x{395}
+
+/\x{03b5}+/8i
+    \x{0395}\x{03b5}\x{03f5}
+ 0: \x{395}\x{3b5}\x{3f5}
+ 1: \x{395}\x{3b5}
+ 2: \x{395}
+
+/\x{03f5}+/8i
+    \x{0395}\x{03b5}\x{03f5}
+ 0: \x{395}\x{3b5}\x{3f5}
+ 1: \x{395}\x{3b5}
+ 2: \x{395}
+
+
+/\x{0398}+/8i
+    \x{0398}\x{03b8}\x{03d1}\x{03f4}
+ 0: \x{398}\x{3b8}\x{3d1}\x{3f4}
+ 1: \x{398}\x{3b8}\x{3d1}
+ 2: \x{398}\x{3b8}
+ 3: \x{398}
+
+/\x{03b8}+/8i
+    \x{0398}\x{03b8}\x{03d1}\x{03f4}
+ 0: \x{398}\x{3b8}\x{3d1}\x{3f4}
+ 1: \x{398}\x{3b8}\x{3d1}
+ 2: \x{398}\x{3b8}
+ 3: \x{398}
+
+/\x{03d1}+/8i
+    \x{0398}\x{03b8}\x{03d1}\x{03f4}
+ 0: \x{398}\x{3b8}\x{3d1}\x{3f4}
+ 1: \x{398}\x{3b8}\x{3d1}
+ 2: \x{398}\x{3b8}
+ 3: \x{398}
+
+/\x{03f4}+/8i
+    \x{0398}\x{03b8}\x{03d1}\x{03f4}
+ 0: \x{398}\x{3b8}\x{3d1}\x{3f4}
+ 1: \x{398}\x{3b8}\x{3d1}
+ 2: \x{398}\x{3b8}
+ 3: \x{398}
+    
+
+/\x{039a}+/8i
+    \x{039a}\x{03ba}\x{03f0}
+ 0: \x{39a}\x{3ba}\x{3f0}
+ 1: \x{39a}\x{3ba}
+ 2: \x{39a}
+
+/\x{03ba}+/8i
+    \x{039a}\x{03ba}\x{03f0}
+ 0: \x{39a}\x{3ba}\x{3f0}
+ 1: \x{39a}\x{3ba}
+ 2: \x{39a}
+
+/\x{03f0}+/8i
+    \x{039a}\x{03ba}\x{03f0}
+ 0: \x{39a}\x{3ba}\x{3f0}
+ 1: \x{39a}\x{3ba}
+ 2: \x{39a}
+    
+
+/\x{03a0}+/8i
+    \x{03a0}\x{03c0}\x{03d6} 
+ 0: \x{3a0}\x{3c0}\x{3d6}
+ 1: \x{3a0}\x{3c0}
+ 2: \x{3a0}
+
+/\x{03c0}+/8i
+    \x{03a0}\x{03c0}\x{03d6} 
+ 0: \x{3a0}\x{3c0}\x{3d6}
+ 1: \x{3a0}\x{3c0}
+ 2: \x{3a0}
+
+/\x{03d6}+/8i
+    \x{03a0}\x{03c0}\x{03d6} 
+ 0: \x{3a0}\x{3c0}\x{3d6}
+ 1: \x{3a0}\x{3c0}
+ 2: \x{3a0}
+
+
+/\x{03a1}+/8i
+    \x{03a1}\x{03c1}\x{03f1}
+ 0: \x{3a1}\x{3c1}\x{3f1}
+ 1: \x{3a1}\x{3c1}
+ 2: \x{3a1}
+
+/\x{03c1}+/8i
+    \x{03a1}\x{03c1}\x{03f1}
+ 0: \x{3a1}\x{3c1}\x{3f1}
+ 1: \x{3a1}\x{3c1}
+ 2: \x{3a1}
+
+/\x{03f1}+/8i
+    \x{03a1}\x{03c1}\x{03f1}
+ 0: \x{3a1}\x{3c1}\x{3f1}
+ 1: \x{3a1}\x{3c1}
+ 2: \x{3a1}
+
+
+/\x{03a3}+/8i
+    \x{03A3}\x{03C2}\x{03C3}
+ 0: \x{3a3}\x{3c2}\x{3c3}
+ 1: \x{3a3}\x{3c2}
+ 2: \x{3a3}
+
+/\x{03c2}+/8i
+    \x{03A3}\x{03C2}\x{03C3}
+ 0: \x{3a3}\x{3c2}\x{3c3}
+ 1: \x{3a3}\x{3c2}
+ 2: \x{3a3}
+
+/\x{03c3}+/8i
+    \x{03A3}\x{03C2}\x{03C3}
+ 0: \x{3a3}\x{3c2}\x{3c3}
+ 1: \x{3a3}\x{3c2}
+ 2: \x{3a3}
+    
+
+/\x{03a6}+/8i
+    \x{03a6}\x{03c6}\x{03d5} 
+ 0: \x{3a6}\x{3c6}\x{3d5}
+ 1: \x{3a6}\x{3c6}
+ 2: \x{3a6}
+
+/\x{03c6}+/8i
+    \x{03a6}\x{03c6}\x{03d5} 
+ 0: \x{3a6}\x{3c6}\x{3d5}
+ 1: \x{3a6}\x{3c6}
+ 2: \x{3a6}
+
+/\x{03d5}+/8i
+    \x{03a6}\x{03c6}\x{03d5} 
+ 0: \x{3a6}\x{3c6}\x{3d5}
+ 1: \x{3a6}\x{3c6}
+ 2: \x{3a6}
+
+
+/\x{03c9}+/8i
+    \x{03c9}\x{03a9}\x{2126}
+ 0: \x{3c9}\x{3a9}\x{2126}
+ 1: \x{3c9}\x{3a9}
+ 2: \x{3c9}
+
+/\x{03a9}+/8i
+    \x{03c9}\x{03a9}\x{2126}
+ 0: \x{3c9}\x{3a9}\x{2126}
+ 1: \x{3c9}\x{3a9}
+ 2: \x{3c9}
+
+/\x{2126}+/8i
+    \x{03c9}\x{03a9}\x{2126}
+ 0: \x{3c9}\x{3a9}\x{2126}
+ 1: \x{3c9}\x{3a9}
+ 2: \x{3c9}
+    
+
+/\x{1e60}+/8i
+    \x{1e60}\x{1e61}\x{1e9b}
+ 0: \x{1e60}\x{1e61}\x{1e9b}
+ 1: \x{1e60}\x{1e61}
+ 2: \x{1e60}
+
+/\x{1e61}+/8i
+    \x{1e60}\x{1e61}\x{1e9b}
+ 0: \x{1e60}\x{1e61}\x{1e9b}
+ 1: \x{1e60}\x{1e61}
+ 2: \x{1e60}
+
+/\x{1e9b}+/8i
+    \x{1e60}\x{1e61}\x{1e9b}
+ 0: \x{1e60}\x{1e61}\x{1e9b}
+ 1: \x{1e60}\x{1e61}
+ 2: \x{1e60}
+    
+
+/\x{1e9e}+/8i
+    \x{1e9e}\x{00df}
+ 0: \x{1e9e}\x{df}
+ 1: \x{1e9e}
+
+/\x{00df}+/8i
+    \x{1e9e}\x{00df}
+ 0: \x{1e9e}\x{df}
+ 1: \x{1e9e}
+    
+
+/\x{1f88}+/8i
+    \x{1f88}\x{1f80} 
+ 0: \x{1f88}\x{1f80}
+ 1: \x{1f88}
+
+/\x{1f80}+/8i
+    \x{1f88}\x{1f80} 
+ 0: \x{1f88}\x{1f80}
+ 1: \x{1f88}
+
+/\x{004b}+/8i
+    \x{004b}\x{006b}\x{212a}
+ 0: Kk\x{212a}
+ 1: Kk
+ 2: K
+
+/\x{006b}+/8i
+    \x{004b}\x{006b}\x{212a}
+ 0: Kk\x{212a}
+ 1: Kk
+ 2: K
+
+/\x{212a}+/8i
+    \x{004b}\x{006b}\x{212a}
+ 0: Kk\x{212a}
+ 1: Kk
+ 2: K
+
+
+/\x{0053}+/8i
+    \x{0053}\x{0073}\x{017f}
+ 0: Ss\x{17f}
+ 1: Ss
+ 2: S
+
+/\x{0073}+/8i
+    \x{0053}\x{0073}\x{017f}
+ 0: Ss\x{17f}
+ 1: Ss
+ 2: S
+
+/\x{017f}+/8i
+    \x{0053}\x{0073}\x{017f}
+ 0: Ss\x{17f}
+ 1: Ss
+ 2: S
+
+/ist/8i
+    ikt
+No match
+
+/is+t/8i
+    iSs\x{17f}t
+ 0: iSs\x{17f}t
+    ikt
+No match
+
+/is+?t/8i
+    ikt
+No match
+
+/is?t/8i
+    ikt
+No match
+
+/is{2}t/8i
+    iskt
+No match
+
+/^\p{Xuc}/8
+    $abc
+ 0: $
+    @abc
+ 0: @
+    `abc
+ 0: `
+    \x{1234}abc
+ 0: \x{1234}
+    ** Failers
+No match
+    abc     
+No match
+
+/^\p{Xuc}+/8
+    $@`\x{a0}\x{1234}\x{e000}**
+ 0: $@`\x{a0}\x{1234}\x{e000}
+ 1: $@`\x{a0}\x{1234}
+ 2: $@`\x{a0}
+ 3: $@`
+ 4: $@
+ 5: $
+    ** Failers
+No match
+    \x{9f}
+No match
+
+/^\p{Xuc}+?/8
+    $@`\x{a0}\x{1234}\x{e000}**
+ 0: $@`\x{a0}\x{1234}\x{e000}
+ 1: $@`\x{a0}\x{1234}
+ 2: $@`\x{a0}
+ 3: $@`
+ 4: $@
+ 5: $
+    ** Failers
+No match
+    \x{9f}
+No match
+
+/^\p{Xuc}+?\*/8
+    $@`\x{a0}\x{1234}\x{e000}**
+ 0: $@`\x{a0}\x{1234}\x{e000}*
+    ** Failers
+No match
+    \x{9f}
+No match
+
+/^\p{Xuc}++/8
+    $@`\x{a0}\x{1234}\x{e000}**
+ 0: $@`\x{a0}\x{1234}\x{e000}
+    ** Failers
+No match
+    \x{9f}
+No match
+
+/^\p{Xuc}{3,5}/8
+    $@`\x{a0}\x{1234}\x{e000}**
+ 0: $@`\x{a0}\x{1234}
+ 1: $@`\x{a0}
+ 2: $@`
+    ** Failers
+No match
+    \x{9f}
+No match
+
+/^\p{Xuc}{3,5}?/8
+    $@`\x{a0}\x{1234}\x{e000}**
+ 0: $@`\x{a0}\x{1234}
+ 1: $@`\x{a0}
+ 2: $@`
+    ** Failers
+No match
+    \x{9f}
+No match
+
+/^[\p{Xuc}]/8
+    $@`\x{a0}\x{1234}\x{e000}**
+ 0: $
+    ** Failers
+No match
+    \x{9f}
+No match
+
+/^[\p{Xuc}]+/8
+    $@`\x{a0}\x{1234}\x{e000}**
+ 0: $@`\x{a0}\x{1234}\x{e000}
+ 1: $@`\x{a0}\x{1234}
+ 2: $@`\x{a0}
+ 3: $@`
+ 4: $@
+ 5: $
+    ** Failers
+No match
+    \x{9f}
+No match
+
+/^\P{Xuc}/8
+    abc
+ 0: a
+    ** Failers
+ 0: *
+    $abc
+No match
+    @abc
+No match
+    `abc
+No match
+    \x{1234}abc
+No match
+
+/^[\P{Xuc}]/8
+    abc
+ 0: a
+    ** Failers
+ 0: *
+    $abc
+No match
+    @abc
+No match
+    `abc
+No match
+    \x{1234}abc
+No match
+
+/-- End of testinput10 --/ 
diff --git a/pcre/testdata/testoutput11-16 b/pcre/testdata/testoutput11-16
new file mode 100644
index 00000000000..dff72b9477c
--- /dev/null
+++ b/pcre/testdata/testoutput11-16
@@ -0,0 +1,713 @@
+/-- These are a few representative patterns whose lengths and offsets are to be 
+shown when the link size is 2. This is just a doublecheck test to ensure the 
+sizes don't go horribly wrong when something is changed. The pattern contents 
+are all themselves checked in other tests. Unicode, including property support, 
+is required for these tests. --/
+
+/((?i)b)/BM
+Memory allocation (code space): 24
+------------------------------------------------------------------
+  0   9 Bra
+  2   5 CBra 1
+  5  /i b
+  7   5 Ket
+  9   9 Ket
+ 11     End
+------------------------------------------------------------------
+
+/(?s)(.*X|^B)/BM
+Memory allocation (code space): 38
+------------------------------------------------------------------
+  0  16 Bra
+  2   7 CBra 1
+  5     AllAny*
+  7     X
+  9   5 Alt
+ 11     ^
+ 12     B
+ 14  12 Ket
+ 16  16 Ket
+ 18     End
+------------------------------------------------------------------
+
+/(?s:.*X|^B)/BM
+Memory allocation (code space): 36
+------------------------------------------------------------------
+  0  15 Bra
+  2   6 Bra
+  4     AllAny*
+  6     X
+  8   5 Alt
+ 10     ^
+ 11     B
+ 13  11 Ket
+ 15  15 Ket
+ 17     End
+------------------------------------------------------------------
+
+/^[[:alnum:]]/BM
+Memory allocation (code space): 46
+------------------------------------------------------------------
+  0  20 Bra
+  2     ^
+  3     [0-9A-Za-z]
+ 20  20 Ket
+ 22     End
+------------------------------------------------------------------
+
+/#/IxMD
+Memory allocation (code space): 10
+------------------------------------------------------------------
+  0   2 Bra
+  2   2 Ket
+  4     End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: extended
+No first char
+No need char
+
+/a#/IxMD
+Memory allocation (code space): 14
+------------------------------------------------------------------
+  0   4 Bra
+  2     a
+  4   4 Ket
+  6     End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: extended
+First char = 'a'
+No need char
+
+/x?+/BM
+Memory allocation (code space): 14
+------------------------------------------------------------------
+  0   4 Bra
+  2     x?+
+  4   4 Ket
+  6     End
+------------------------------------------------------------------
+
+/x++/BM
+Memory allocation (code space): 14
+------------------------------------------------------------------
+  0   4 Bra
+  2     x++
+  4   4 Ket
+  6     End
+------------------------------------------------------------------
+
+/x{1,3}+/BM 
+Memory allocation (code space): 28
+------------------------------------------------------------------
+  0  11 Bra
+  2   7 Once
+  4     x
+  6     x{0,2}
+  9   7 Ket
+ 11  11 Ket
+ 13     End
+------------------------------------------------------------------
+
+/(x)*+/BM
+Memory allocation (code space): 26
+------------------------------------------------------------------
+  0  10 Bra
+  2     Braposzero
+  3   5 CBraPos 1
+  6     x
+  8   5 KetRpos
+ 10  10 Ket
+ 12     End
+------------------------------------------------------------------
+
+/^((a+)(?U)([ab]+)(?-U)([bc]+)(\w*))/BM
+Memory allocation (code space): 142
+------------------------------------------------------------------
+  0  68 Bra
+  2     ^
+  3  63 CBra 1
+  6   5 CBra 2
+  9     a+
+ 11   5 Ket
+ 13  21 CBra 3
+ 16     [ab]+?
+ 34  21 Ket
+ 36  21 CBra 4
+ 39     [bc]+
+ 57  21 Ket
+ 59   5 CBra 5
+ 62     \w*
+ 64   5 Ket
+ 66  63 Ket
+ 68  68 Ket
+ 70     End
+------------------------------------------------------------------
+
+|8J\$WE\<\.rX\+ix\[d1b\!H\#\?vV0vrK\:ZH1\=2M\>iV\;\?aPhFB\<\*vW\@QW\@sO9\}cfZA\-i\'w\%hKd6gt1UJP\,15_\#QY\$M\^Mss_U\/\]\&LK9\[5vQub\^w\[KDD\qmj\;2\}YWFdYx\.Ap\]hjCPTP\(n28k\+3\;o\&WXqs\/gOXdr\$\:r\'do0\;b4c\(f_Gr\=\"\\4\)\[01T7ajQJvL\$W\~mL_sS\/4h\:x\*\[ZN\=KLs\&L5zX\/\/\>it\,o\:aU\(\;Z\>pW\&T7oP\'2K\^E\:x9\'c\[\%z\-\,64JQ5AeH_G\#KijUKghQw\^\\vea3a\?kka_G\$8\#\`\*kynsxzBLru\'\]k_\[7FrVx\}\^\=\$blx\>s\-N\%j\;D\*aZDnsw\:YKZ\%Q\.Kne9\#hP\?\+b3\(SOvL\,\^\;\&u5\@\?5C5Bhb\=m\-vEh_L15Jl\]U\)0RP6\{q\%L\^_z5E\'Dw6X\b|BM
+Memory allocation (code space): 1648
+------------------------------------------------------------------
+  0 821 Bra
+  2     8J$WE<.rX+ix[d1b!H#?vV0vrK:ZH1=2M>iV;?aPhFB<*vW@QW@sO9}cfZA-i'w%hKd6gt1UJP,15_#QY$M^Mss_U/]&LK9[5vQub^w[KDDqmj;2}YWFdYx.Ap]hjCPTP(n28k+3;o&WXqs/gOXdr$:r'do0;b4c(f_Gr="\4)[01T7ajQJvL$W~mL_sS/4h:x*[ZN=KLs&L5zX//>it,o:aU(;Z>pW&T7oP'2K^E:x9'c[%z-,64JQ5AeH_G#KijUKghQw^\vea3a?kka_G$8#`*kynsxzBLru']k_[7FrVx}^=$blx>s-N%j;D*aZDnsw:YKZ%Q.Kne9#hP?+b3(SOvL,^;&u5@?5C5Bhb=m-vEh_L15Jl]U)0RP6{q%L^_z5E'Dw6X
+820     \b
+821 821 Ket
+823     End
+------------------------------------------------------------------
+
+|\$\<\.X\+ix\[d1b\!H\#\?vV0vrK\:ZH1\=2M\>iV\;\?aPhFB\<\*vW\@QW\@sO9\}cfZA\-i\'w\%hKd6gt1UJP\,15_\#QY\$M\^Mss_U\/\]\&LK9\[5vQub\^w\[KDD\qmj\;2\}YWFdYx\.Ap\]hjCPTP\(n28k\+3\;o\&WXqs\/gOXdr\$\:r\'do0\;b4c\(f_Gr\=\"\\4\)\[01T7ajQJvL\$W\~mL_sS\/4h\:x\*\[ZN\=KLs\&L5zX\/\/\>it\,o\:aU\(\;Z\>pW\&T7oP\'2K\^E\:x9\'c\[\%z\-\,64JQ5AeH_G\#KijUKghQw\^\\vea3a\?kka_G\$8\#\`\*kynsxzBLru\'\]k_\[7FrVx\}\^\=\$blx\>s\-N\%j\;D\*aZDnsw\:YKZ\%Q\.Kne9\#hP\?\+b3\(SOvL\,\^\;\&u5\@\?5C5Bhb\=m\-vEh_L15Jl\]U\)0RP6\{q\%L\^_z5E\'Dw6X\b|BM
+Memory allocation (code space): 1628
+------------------------------------------------------------------
+  0 811 Bra
+  2     $<.X+ix[d1b!H#?vV0vrK:ZH1=2M>iV;?aPhFB<*vW@QW@sO9}cfZA-i'w%hKd6gt1UJP,15_#QY$M^Mss_U/]&LK9[5vQub^w[KDDqmj;2}YWFdYx.Ap]hjCPTP(n28k+3;o&WXqs/gOXdr$:r'do0;b4c(f_Gr="\4)[01T7ajQJvL$W~mL_sS/4h:x*[ZN=KLs&L5zX//>it,o:aU(;Z>pW&T7oP'2K^E:x9'c[%z-,64JQ5AeH_G#KijUKghQw^\vea3a?kka_G$8#`*kynsxzBLru']k_[7FrVx}^=$blx>s-N%j;D*aZDnsw:YKZ%Q.Kne9#hP?+b3(SOvL,^;&u5@?5C5Bhb=m-vEh_L15Jl]U)0RP6{q%L^_z5E'Dw6X
+810     \b
+811 811 Ket
+813     End
+------------------------------------------------------------------
+
+/(a(?1)b)/BM
+Memory allocation (code space): 32
+------------------------------------------------------------------
+  0  13 Bra
+  2   9 CBra 1
+  5     a
+  7   2 Recurse
+  9     b
+ 11   9 Ket
+ 13  13 Ket
+ 15     End
+------------------------------------------------------------------
+
+/(a(?1)+b)/BM
+Memory allocation (code space): 40
+------------------------------------------------------------------
+  0  17 Bra
+  2  13 CBra 1
+  5     a
+  7   4 Once
+  9   2 Recurse
+ 11   4 KetRmax
+ 13     b
+ 15  13 Ket
+ 17  17 Ket
+ 19     End
+------------------------------------------------------------------
+
+/a(?Pb|c)d(?Pe)/BM
+Memory allocation (code space): 80
+------------------------------------------------------------------
+  0  24 Bra
+  2     a
+  4   5 CBra 1
+  7     b
+  9   4 Alt
+ 11     c
+ 13   9 Ket
+ 15     d
+ 17   5 CBra 2
+ 20     e
+ 22   5 Ket
+ 24  24 Ket
+ 26     End
+------------------------------------------------------------------
+
+/(?:a(?Pc(?Pd)))(?Pa)/BM
+Memory allocation (code space): 73
+------------------------------------------------------------------
+  0  29 Bra
+  2  18 Bra
+  4     a
+  6  12 CBra 1
+  9     c
+ 11   5 CBra 2
+ 14     d
+ 16   5 Ket
+ 18  12 Ket
+ 20  18 Ket
+ 22   5 CBra 3
+ 25     a
+ 27   5 Ket
+ 29  29 Ket
+ 31     End
+------------------------------------------------------------------
+
+/(?Pa)...(?P=a)bbb(?P>a)d/BM
+Memory allocation (code space): 57
+------------------------------------------------------------------
+  0  24 Bra
+  2   5 CBra 1
+  5     a
+  7   5 Ket
+  9     Any
+ 10     Any
+ 11     Any
+ 12     \1
+ 14     bbb
+ 20   2 Recurse
+ 22     d
+ 24  24 Ket
+ 26     End
+------------------------------------------------------------------
+
+/abc(?C255)de(?C)f/BM
+Memory allocation (code space): 50
+------------------------------------------------------------------
+  0  22 Bra
+  2     abc
+  8     Callout 255 10 1
+ 12     de
+ 16     Callout 0 16 1
+ 20     f
+ 22  22 Ket
+ 24     End
+------------------------------------------------------------------
+
+/abcde/CBM
+Memory allocation (code space): 78
+------------------------------------------------------------------
+  0  36 Bra
+  2     Callout 255 0 1
+  6     a
+  8     Callout 255 1 1
+ 12     b
+ 14     Callout 255 2 1
+ 18     c
+ 20     Callout 255 3 1
+ 24     d
+ 26     Callout 255 4 1
+ 30     e
+ 32     Callout 255 5 0
+ 36  36 Ket
+ 38     End
+------------------------------------------------------------------
+
+/\x{100}/8BM
+Memory allocation (code space): 14
+------------------------------------------------------------------
+  0   4 Bra
+  2     \x{100}
+  4   4 Ket
+  6     End
+------------------------------------------------------------------
+
+/\x{1000}/8BM
+Memory allocation (code space): 14
+------------------------------------------------------------------
+  0   4 Bra
+  2     \x{1000}
+  4   4 Ket
+  6     End
+------------------------------------------------------------------
+
+/\x{10000}/8BM
+Memory allocation (code space): 16
+------------------------------------------------------------------
+  0   5 Bra
+  2     \x{10000}
+  5   5 Ket
+  7     End
+------------------------------------------------------------------
+
+/\x{100000}/8BM
+Memory allocation (code space): 16
+------------------------------------------------------------------
+  0   5 Bra
+  2     \x{100000}
+  5   5 Ket
+  7     End
+------------------------------------------------------------------
+
+/\x{10ffff}/8BM
+Memory allocation (code space): 16
+------------------------------------------------------------------
+  0   5 Bra
+  2     \x{10ffff}
+  5   5 Ket
+  7     End
+------------------------------------------------------------------
+
+/\x{110000}/8BM
+Failed: character value in \x{...} sequence is too large at offset 9
+
+/[\x{ff}]/8BM
+Memory allocation (code space): 14
+------------------------------------------------------------------
+  0   4 Bra
+  2     \x{ff}
+  4   4 Ket
+  6     End
+------------------------------------------------------------------
+
+/[\x{100}]/8BM
+Memory allocation (code space): 14
+------------------------------------------------------------------
+  0   4 Bra
+  2     \x{100}
+  4   4 Ket
+  6     End
+------------------------------------------------------------------
+
+/\x80/8BM
+Memory allocation (code space): 14
+------------------------------------------------------------------
+  0   4 Bra
+  2     \x80
+  4   4 Ket
+  6     End
+------------------------------------------------------------------
+
+/\xff/8BM
+Memory allocation (code space): 14
+------------------------------------------------------------------
+  0   4 Bra
+  2     \x{ff}
+  4   4 Ket
+  6     End
+------------------------------------------------------------------
+
+/\x{0041}\x{2262}\x{0391}\x{002e}/D8M
+Memory allocation (code space): 26
+------------------------------------------------------------------
+  0  10 Bra
+  2     A\x{2262}\x{391}.
+ 10  10 Ket
+ 12     End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = 'A'
+Need char = '.'
+    
+/\x{D55c}\x{ad6d}\x{C5B4}/D8M 
+Memory allocation (code space): 22
+------------------------------------------------------------------
+  0   8 Bra
+  2     \x{d55c}\x{ad6d}\x{c5b4}
+  8   8 Ket
+ 10     End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{d55c}
+Need char = \x{c5b4}
+
+/\x{65e5}\x{672c}\x{8a9e}/D8M
+Memory allocation (code space): 22
+------------------------------------------------------------------
+  0   8 Bra
+  2     \x{65e5}\x{672c}\x{8a9e}
+  8   8 Ket
+ 10     End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{65e5}
+Need char = \x{8a9e}
+
+/[\x{100}]/8BM
+Memory allocation (code space): 14
+------------------------------------------------------------------
+  0   4 Bra
+  2     \x{100}
+  4   4 Ket
+  6     End
+------------------------------------------------------------------
+
+/[Z\x{100}]/8BM
+Memory allocation (code space): 54
+------------------------------------------------------------------
+  0  24 Bra
+  2     [Z\x{100}]
+ 24  24 Ket
+ 26     End
+------------------------------------------------------------------
+
+/^[\x{100}\E-\Q\E\x{150}]/B8M
+Memory allocation (code space): 26
+------------------------------------------------------------------
+  0  10 Bra
+  2     ^
+  3     [\x{100}-\x{150}]
+ 10  10 Ket
+ 12     End
+------------------------------------------------------------------
+
+/^[\QĀ\E-\QŐ\E]/B8M
+Memory allocation (code space): 26
+------------------------------------------------------------------
+  0  10 Bra
+  2     ^
+  3     [\x{100}-\x{150}]
+ 10  10 Ket
+ 12     End
+------------------------------------------------------------------
+
+/^[\QĀ\E-\QŐ\E/B8M
+Failed: missing terminating ] for character class at offset 13
+
+/[\p{L}]/BM
+Memory allocation (code space): 24
+------------------------------------------------------------------
+  0   9 Bra
+  2     [\p{L}]
+  9   9 Ket
+ 11     End
+------------------------------------------------------------------
+
+/[\p{^L}]/BM
+Memory allocation (code space): 24
+------------------------------------------------------------------
+  0   9 Bra
+  2     [\P{L}]
+  9   9 Ket
+ 11     End
+------------------------------------------------------------------
+
+/[\P{L}]/BM
+Memory allocation (code space): 24
+------------------------------------------------------------------
+  0   9 Bra
+  2     [\P{L}]
+  9   9 Ket
+ 11     End
+------------------------------------------------------------------
+
+/[\P{^L}]/BM
+Memory allocation (code space): 24
+------------------------------------------------------------------
+  0   9 Bra
+  2     [\p{L}]
+  9   9 Ket
+ 11     End
+------------------------------------------------------------------
+
+/[abc\p{L}\x{0660}]/8BM
+Memory allocation (code space): 60
+------------------------------------------------------------------
+  0  27 Bra
+  2     [a-c\p{L}\x{660}]
+ 27  27 Ket
+ 29     End
+------------------------------------------------------------------
+
+/[\p{Nd}]/8BM
+Memory allocation (code space): 24
+------------------------------------------------------------------
+  0   9 Bra
+  2     [\p{Nd}]
+  9   9 Ket
+ 11     End
+------------------------------------------------------------------
+
+/[\p{Nd}+-]+/8BM
+Memory allocation (code space): 58
+------------------------------------------------------------------
+  0  26 Bra
+  2     [+\-\p{Nd}]+
+ 26  26 Ket
+ 28     End
+------------------------------------------------------------------
+
+/A\x{391}\x{10427}\x{ff3a}\x{1fb0}/8iBM
+Memory allocation (code space): 32
+------------------------------------------------------------------
+  0  13 Bra
+  2  /i A\x{391}\x{10427}\x{ff3a}\x{1fb0}
+ 13  13 Ket
+ 15     End
+------------------------------------------------------------------
+
+/A\x{391}\x{10427}\x{ff3a}\x{1fb0}/8BM
+Memory allocation (code space): 32
+------------------------------------------------------------------
+  0  13 Bra
+  2     A\x{391}\x{10427}\x{ff3a}\x{1fb0}
+ 13  13 Ket
+ 15     End
+------------------------------------------------------------------
+
+/[\x{105}-\x{109}]/8iBM
+Memory allocation (code space): 24
+------------------------------------------------------------------
+  0   9 Bra
+  2     [\x{104}-\x{109}]
+  9   9 Ket
+ 11     End
+------------------------------------------------------------------
+
+/( ( (?(1)0|) )*   )/xBM
+Memory allocation (code space): 52
+------------------------------------------------------------------
+  0  23 Bra
+  2  19 CBra 1
+  5     Brazero
+  6  13 SCBra 2
+  9   6 Cond
+ 11   1 Cond ref
+ 13     0
+ 15   2 Alt
+ 17   8 Ket
+ 19  13 KetRmax
+ 21  19 Ket
+ 23  23 Ket
+ 25     End
+------------------------------------------------------------------
+
+/(  (?(1)0|)*   )/xBM
+Memory allocation (code space): 42
+------------------------------------------------------------------
+  0  18 Bra
+  2  14 CBra 1
+  5     Brazero
+  6   6 SCond
+  8   1 Cond ref
+ 10     0
+ 12   2 Alt
+ 14   8 KetRmax
+ 16  14 Ket
+ 18  18 Ket
+ 20     End
+------------------------------------------------------------------
+
+/[a]/BM
+Memory allocation (code space): 14
+------------------------------------------------------------------
+  0   4 Bra
+  2     a
+  4   4 Ket
+  6     End
+------------------------------------------------------------------
+
+/[a]/8BM
+Memory allocation (code space): 14
+------------------------------------------------------------------
+  0   4 Bra
+  2     a
+  4   4 Ket
+  6     End
+------------------------------------------------------------------
+
+/[\xaa]/BM
+Memory allocation (code space): 14
+------------------------------------------------------------------
+  0   4 Bra
+  2     \x{aa}
+  4   4 Ket
+  6     End
+------------------------------------------------------------------
+
+/[\xaa]/8BM
+Memory allocation (code space): 14
+------------------------------------------------------------------
+  0   4 Bra
+  2     \x{aa}
+  4   4 Ket
+  6     End
+------------------------------------------------------------------
+
+/[^a]/BM
+Memory allocation (code space): 14
+------------------------------------------------------------------
+  0   4 Bra
+  2     [^a]
+  4   4 Ket
+  6     End
+------------------------------------------------------------------
+
+/[^a]/8BM
+Memory allocation (code space): 14
+------------------------------------------------------------------
+  0   4 Bra
+  2     [^a]
+  4   4 Ket
+  6     End
+------------------------------------------------------------------
+
+/[^\xaa]/BM
+Memory allocation (code space): 14
+------------------------------------------------------------------
+  0   4 Bra
+  2     [^\x{aa}]
+  4   4 Ket
+  6     End
+------------------------------------------------------------------
+
+/[^\xaa]/8BM
+Memory allocation (code space): 14
+------------------------------------------------------------------
+  0   4 Bra
+  2     [^\x{aa}]
+  4   4 Ket
+  6     End
+------------------------------------------------------------------
+
+/[^\d]/8WB
+------------------------------------------------------------------
+  0   9 Bra
+  2     [^\p{Nd}]
+  9   9 Ket
+ 11     End
+------------------------------------------------------------------
+
+/[[:^alpha:][:^cntrl:]]+/8WB
+------------------------------------------------------------------
+  0  26 Bra
+  2     [ -~\x80-\xff\P{L}]+
+ 26  26 Ket
+ 28     End
+------------------------------------------------------------------
+
+/[[:^cntrl:][:^alpha:]]+/8WB
+------------------------------------------------------------------
+  0  26 Bra
+  2     [ -~\x80-\xff\P{L}]+
+ 26  26 Ket
+ 28     End
+------------------------------------------------------------------
+
+/[[:alpha:]]+/8WB
+------------------------------------------------------------------
+  0  10 Bra
+  2     [\p{L}]+
+ 10  10 Ket
+ 12     End
+------------------------------------------------------------------
+
+/[[:^alpha:]\S]+/8WB
+------------------------------------------------------------------
+  0  13 Bra
+  2     [\P{L}\P{Xsp}]+
+ 13  13 Ket
+ 15     End
+------------------------------------------------------------------
+
+/abc(d|e)(*THEN)x(123(*THEN)4|567(b|q)(*THEN)xx)/B
+------------------------------------------------------------------
+  0  60 Bra
+  2     abc
+  8   5 CBra 1
+ 11     d
+ 13   4 Alt
+ 15     e
+ 17   9 Ket
+ 19     *THEN
+ 20     x
+ 22  12 CBra 2
+ 25     123
+ 31     *THEN
+ 32     4
+ 34  24 Alt
+ 36     567
+ 42   5 CBra 3
+ 45     b
+ 47   4 Alt
+ 49     q
+ 51   9 Ket
+ 53     *THEN
+ 54     xx
+ 58  36 Ket
+ 60  60 Ket
+ 62     End
+------------------------------------------------------------------
+
+/-- End of testinput11 --/
diff --git a/pcre/testdata/testoutput11-32 b/pcre/testdata/testoutput11-32
new file mode 100644
index 00000000000..8335fb82418
--- /dev/null
+++ b/pcre/testdata/testoutput11-32
@@ -0,0 +1,713 @@
+/-- These are a few representative patterns whose lengths and offsets are to be 
+shown when the link size is 2. This is just a doublecheck test to ensure the 
+sizes don't go horribly wrong when something is changed. The pattern contents 
+are all themselves checked in other tests. Unicode, including property support, 
+is required for these tests. --/
+
+/((?i)b)/BM
+Memory allocation (code space): 48
+------------------------------------------------------------------
+  0   9 Bra
+  2   5 CBra 1
+  5  /i b
+  7   5 Ket
+  9   9 Ket
+ 11     End
+------------------------------------------------------------------
+
+/(?s)(.*X|^B)/BM
+Memory allocation (code space): 76
+------------------------------------------------------------------
+  0  16 Bra
+  2   7 CBra 1
+  5     AllAny*
+  7     X
+  9   5 Alt
+ 11     ^
+ 12     B
+ 14  12 Ket
+ 16  16 Ket
+ 18     End
+------------------------------------------------------------------
+
+/(?s:.*X|^B)/BM
+Memory allocation (code space): 72
+------------------------------------------------------------------
+  0  15 Bra
+  2   6 Bra
+  4     AllAny*
+  6     X
+  8   5 Alt
+ 10     ^
+ 11     B
+ 13  11 Ket
+ 15  15 Ket
+ 17     End
+------------------------------------------------------------------
+
+/^[[:alnum:]]/BM
+Memory allocation (code space): 60
+------------------------------------------------------------------
+  0  12 Bra
+  2     ^
+  3     [0-9A-Za-z]
+ 12  12 Ket
+ 14     End
+------------------------------------------------------------------
+
+/#/IxMD
+Memory allocation (code space): 20
+------------------------------------------------------------------
+  0   2 Bra
+  2   2 Ket
+  4     End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: extended
+No first char
+No need char
+
+/a#/IxMD
+Memory allocation (code space): 28
+------------------------------------------------------------------
+  0   4 Bra
+  2     a
+  4   4 Ket
+  6     End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: extended
+First char = 'a'
+No need char
+
+/x?+/BM
+Memory allocation (code space): 28
+------------------------------------------------------------------
+  0   4 Bra
+  2     x?+
+  4   4 Ket
+  6     End
+------------------------------------------------------------------
+
+/x++/BM
+Memory allocation (code space): 28
+------------------------------------------------------------------
+  0   4 Bra
+  2     x++
+  4   4 Ket
+  6     End
+------------------------------------------------------------------
+
+/x{1,3}+/BM 
+Memory allocation (code space): 56
+------------------------------------------------------------------
+  0  11 Bra
+  2   7 Once
+  4     x
+  6     x{0,2}
+  9   7 Ket
+ 11  11 Ket
+ 13     End
+------------------------------------------------------------------
+
+/(x)*+/BM
+Memory allocation (code space): 52
+------------------------------------------------------------------
+  0  10 Bra
+  2     Braposzero
+  3   5 CBraPos 1
+  6     x
+  8   5 KetRpos
+ 10  10 Ket
+ 12     End
+------------------------------------------------------------------
+
+/^((a+)(?U)([ab]+)(?-U)([bc]+)(\w*))/BM
+Memory allocation (code space): 220
+------------------------------------------------------------------
+  0  52 Bra
+  2     ^
+  3  47 CBra 1
+  6   5 CBra 2
+  9     a+
+ 11   5 Ket
+ 13  13 CBra 3
+ 16     [ab]+?
+ 26  13 Ket
+ 28  13 CBra 4
+ 31     [bc]+
+ 41  13 Ket
+ 43   5 CBra 5
+ 46     \w*
+ 48   5 Ket
+ 50  47 Ket
+ 52  52 Ket
+ 54     End
+------------------------------------------------------------------
+
+|8J\$WE\<\.rX\+ix\[d1b\!H\#\?vV0vrK\:ZH1\=2M\>iV\;\?aPhFB\<\*vW\@QW\@sO9\}cfZA\-i\'w\%hKd6gt1UJP\,15_\#QY\$M\^Mss_U\/\]\&LK9\[5vQub\^w\[KDD\qmj\;2\}YWFdYx\.Ap\]hjCPTP\(n28k\+3\;o\&WXqs\/gOXdr\$\:r\'do0\;b4c\(f_Gr\=\"\\4\)\[01T7ajQJvL\$W\~mL_sS\/4h\:x\*\[ZN\=KLs\&L5zX\/\/\>it\,o\:aU\(\;Z\>pW\&T7oP\'2K\^E\:x9\'c\[\%z\-\,64JQ5AeH_G\#KijUKghQw\^\\vea3a\?kka_G\$8\#\`\*kynsxzBLru\'\]k_\[7FrVx\}\^\=\$blx\>s\-N\%j\;D\*aZDnsw\:YKZ\%Q\.Kne9\#hP\?\+b3\(SOvL\,\^\;\&u5\@\?5C5Bhb\=m\-vEh_L15Jl\]U\)0RP6\{q\%L\^_z5E\'Dw6X\b|BM
+Memory allocation (code space): 3296
+------------------------------------------------------------------
+  0 821 Bra
+  2     8J$WE<.rX+ix[d1b!H#?vV0vrK:ZH1=2M>iV;?aPhFB<*vW@QW@sO9}cfZA-i'w%hKd6gt1UJP,15_#QY$M^Mss_U/]&LK9[5vQub^w[KDDqmj;2}YWFdYx.Ap]hjCPTP(n28k+3;o&WXqs/gOXdr$:r'do0;b4c(f_Gr="\4)[01T7ajQJvL$W~mL_sS/4h:x*[ZN=KLs&L5zX//>it,o:aU(;Z>pW&T7oP'2K^E:x9'c[%z-,64JQ5AeH_G#KijUKghQw^\vea3a?kka_G$8#`*kynsxzBLru']k_[7FrVx}^=$blx>s-N%j;D*aZDnsw:YKZ%Q.Kne9#hP?+b3(SOvL,^;&u5@?5C5Bhb=m-vEh_L15Jl]U)0RP6{q%L^_z5E'Dw6X
+820     \b
+821 821 Ket
+823     End
+------------------------------------------------------------------
+
+|\$\<\.X\+ix\[d1b\!H\#\?vV0vrK\:ZH1\=2M\>iV\;\?aPhFB\<\*vW\@QW\@sO9\}cfZA\-i\'w\%hKd6gt1UJP\,15_\#QY\$M\^Mss_U\/\]\&LK9\[5vQub\^w\[KDD\qmj\;2\}YWFdYx\.Ap\]hjCPTP\(n28k\+3\;o\&WXqs\/gOXdr\$\:r\'do0\;b4c\(f_Gr\=\"\\4\)\[01T7ajQJvL\$W\~mL_sS\/4h\:x\*\[ZN\=KLs\&L5zX\/\/\>it\,o\:aU\(\;Z\>pW\&T7oP\'2K\^E\:x9\'c\[\%z\-\,64JQ5AeH_G\#KijUKghQw\^\\vea3a\?kka_G\$8\#\`\*kynsxzBLru\'\]k_\[7FrVx\}\^\=\$blx\>s\-N\%j\;D\*aZDnsw\:YKZ\%Q\.Kne9\#hP\?\+b3\(SOvL\,\^\;\&u5\@\?5C5Bhb\=m\-vEh_L15Jl\]U\)0RP6\{q\%L\^_z5E\'Dw6X\b|BM
+Memory allocation (code space): 3256
+------------------------------------------------------------------
+  0 811 Bra
+  2     $<.X+ix[d1b!H#?vV0vrK:ZH1=2M>iV;?aPhFB<*vW@QW@sO9}cfZA-i'w%hKd6gt1UJP,15_#QY$M^Mss_U/]&LK9[5vQub^w[KDDqmj;2}YWFdYx.Ap]hjCPTP(n28k+3;o&WXqs/gOXdr$:r'do0;b4c(f_Gr="\4)[01T7ajQJvL$W~mL_sS/4h:x*[ZN=KLs&L5zX//>it,o:aU(;Z>pW&T7oP'2K^E:x9'c[%z-,64JQ5AeH_G#KijUKghQw^\vea3a?kka_G$8#`*kynsxzBLru']k_[7FrVx}^=$blx>s-N%j;D*aZDnsw:YKZ%Q.Kne9#hP?+b3(SOvL,^;&u5@?5C5Bhb=m-vEh_L15Jl]U)0RP6{q%L^_z5E'Dw6X
+810     \b
+811 811 Ket
+813     End
+------------------------------------------------------------------
+
+/(a(?1)b)/BM
+Memory allocation (code space): 64
+------------------------------------------------------------------
+  0  13 Bra
+  2   9 CBra 1
+  5     a
+  7   2 Recurse
+  9     b
+ 11   9 Ket
+ 13  13 Ket
+ 15     End
+------------------------------------------------------------------
+
+/(a(?1)+b)/BM
+Memory allocation (code space): 80
+------------------------------------------------------------------
+  0  17 Bra
+  2  13 CBra 1
+  5     a
+  7   4 Once
+  9   2 Recurse
+ 11   4 KetRmax
+ 13     b
+ 15  13 Ket
+ 17  17 Ket
+ 19     End
+------------------------------------------------------------------
+
+/a(?Pb|c)d(?Pe)/BM
+Memory allocation (code space): 186
+------------------------------------------------------------------
+  0  24 Bra
+  2     a
+  4   5 CBra 1
+  7     b
+  9   4 Alt
+ 11     c
+ 13   9 Ket
+ 15     d
+ 17   5 CBra 2
+ 20     e
+ 22   5 Ket
+ 24  24 Ket
+ 26     End
+------------------------------------------------------------------
+
+/(?:a(?Pc(?Pd)))(?Pa)/BM
+Memory allocation (code space): 155
+------------------------------------------------------------------
+  0  29 Bra
+  2  18 Bra
+  4     a
+  6  12 CBra 1
+  9     c
+ 11   5 CBra 2
+ 14     d
+ 16   5 Ket
+ 18  12 Ket
+ 20  18 Ket
+ 22   5 CBra 3
+ 25     a
+ 27   5 Ket
+ 29  29 Ket
+ 31     End
+------------------------------------------------------------------
+
+/(?Pa)...(?P=a)bbb(?P>a)d/BM
+Memory allocation (code space): 117
+------------------------------------------------------------------
+  0  24 Bra
+  2   5 CBra 1
+  5     a
+  7   5 Ket
+  9     Any
+ 10     Any
+ 11     Any
+ 12     \1
+ 14     bbb
+ 20   2 Recurse
+ 22     d
+ 24  24 Ket
+ 26     End
+------------------------------------------------------------------
+
+/abc(?C255)de(?C)f/BM
+Memory allocation (code space): 100
+------------------------------------------------------------------
+  0  22 Bra
+  2     abc
+  8     Callout 255 10 1
+ 12     de
+ 16     Callout 0 16 1
+ 20     f
+ 22  22 Ket
+ 24     End
+------------------------------------------------------------------
+
+/abcde/CBM
+Memory allocation (code space): 156
+------------------------------------------------------------------
+  0  36 Bra
+  2     Callout 255 0 1
+  6     a
+  8     Callout 255 1 1
+ 12     b
+ 14     Callout 255 2 1
+ 18     c
+ 20     Callout 255 3 1
+ 24     d
+ 26     Callout 255 4 1
+ 30     e
+ 32     Callout 255 5 0
+ 36  36 Ket
+ 38     End
+------------------------------------------------------------------
+
+/\x{100}/8BM
+Memory allocation (code space): 28
+------------------------------------------------------------------
+  0   4 Bra
+  2     \x{100}
+  4   4 Ket
+  6     End
+------------------------------------------------------------------
+
+/\x{1000}/8BM
+Memory allocation (code space): 28
+------------------------------------------------------------------
+  0   4 Bra
+  2     \x{1000}
+  4   4 Ket
+  6     End
+------------------------------------------------------------------
+
+/\x{10000}/8BM
+Memory allocation (code space): 28
+------------------------------------------------------------------
+  0   4 Bra
+  2     \x{10000}
+  4   4 Ket
+  6     End
+------------------------------------------------------------------
+
+/\x{100000}/8BM
+Memory allocation (code space): 28
+------------------------------------------------------------------
+  0   4 Bra
+  2     \x{100000}
+  4   4 Ket
+  6     End
+------------------------------------------------------------------
+
+/\x{10ffff}/8BM
+Memory allocation (code space): 28
+------------------------------------------------------------------
+  0   4 Bra
+  2     \x{10ffff}
+  4   4 Ket
+  6     End
+------------------------------------------------------------------
+
+/\x{110000}/8BM
+Failed: character value in \x{...} sequence is too large at offset 9
+
+/[\x{ff}]/8BM
+Memory allocation (code space): 28
+------------------------------------------------------------------
+  0   4 Bra
+  2     \x{ff}
+  4   4 Ket
+  6     End
+------------------------------------------------------------------
+
+/[\x{100}]/8BM
+Memory allocation (code space): 28
+------------------------------------------------------------------
+  0   4 Bra
+  2     \x{100}
+  4   4 Ket
+  6     End
+------------------------------------------------------------------
+
+/\x80/8BM
+Memory allocation (code space): 28
+------------------------------------------------------------------
+  0   4 Bra
+  2     \x80
+  4   4 Ket
+  6     End
+------------------------------------------------------------------
+
+/\xff/8BM
+Memory allocation (code space): 28
+------------------------------------------------------------------
+  0   4 Bra
+  2     \x{ff}
+  4   4 Ket
+  6     End
+------------------------------------------------------------------
+
+/\x{0041}\x{2262}\x{0391}\x{002e}/D8M
+Memory allocation (code space): 52
+------------------------------------------------------------------
+  0  10 Bra
+  2     A\x{2262}\x{391}.
+ 10  10 Ket
+ 12     End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = 'A'
+Need char = '.'
+    
+/\x{D55c}\x{ad6d}\x{C5B4}/D8M 
+Memory allocation (code space): 44
+------------------------------------------------------------------
+  0   8 Bra
+  2     \x{d55c}\x{ad6d}\x{c5b4}
+  8   8 Ket
+ 10     End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{d55c}
+Need char = \x{c5b4}
+
+/\x{65e5}\x{672c}\x{8a9e}/D8M
+Memory allocation (code space): 44
+------------------------------------------------------------------
+  0   8 Bra
+  2     \x{65e5}\x{672c}\x{8a9e}
+  8   8 Ket
+ 10     End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{65e5}
+Need char = \x{8a9e}
+
+/[\x{100}]/8BM
+Memory allocation (code space): 28
+------------------------------------------------------------------
+  0   4 Bra
+  2     \x{100}
+  4   4 Ket
+  6     End
+------------------------------------------------------------------
+
+/[Z\x{100}]/8BM
+Memory allocation (code space): 76
+------------------------------------------------------------------
+  0  16 Bra
+  2     [Z\x{100}]
+ 16  16 Ket
+ 18     End
+------------------------------------------------------------------
+
+/^[\x{100}\E-\Q\E\x{150}]/B8M
+Memory allocation (code space): 52
+------------------------------------------------------------------
+  0  10 Bra
+  2     ^
+  3     [\x{100}-\x{150}]
+ 10  10 Ket
+ 12     End
+------------------------------------------------------------------
+
+/^[\QĀ\E-\QŐ\E]/B8M
+Memory allocation (code space): 52
+------------------------------------------------------------------
+  0  10 Bra
+  2     ^
+  3     [\x{100}-\x{150}]
+ 10  10 Ket
+ 12     End
+------------------------------------------------------------------
+
+/^[\QĀ\E-\QŐ\E/B8M
+Failed: missing terminating ] for character class at offset 13
+
+/[\p{L}]/BM
+Memory allocation (code space): 48
+------------------------------------------------------------------
+  0   9 Bra
+  2     [\p{L}]
+  9   9 Ket
+ 11     End
+------------------------------------------------------------------
+
+/[\p{^L}]/BM
+Memory allocation (code space): 48
+------------------------------------------------------------------
+  0   9 Bra
+  2     [\P{L}]
+  9   9 Ket
+ 11     End
+------------------------------------------------------------------
+
+/[\P{L}]/BM
+Memory allocation (code space): 48
+------------------------------------------------------------------
+  0   9 Bra
+  2     [\P{L}]
+  9   9 Ket
+ 11     End
+------------------------------------------------------------------
+
+/[\P{^L}]/BM
+Memory allocation (code space): 48
+------------------------------------------------------------------
+  0   9 Bra
+  2     [\p{L}]
+  9   9 Ket
+ 11     End
+------------------------------------------------------------------
+
+/[abc\p{L}\x{0660}]/8BM
+Memory allocation (code space): 88
+------------------------------------------------------------------
+  0  19 Bra
+  2     [a-c\p{L}\x{660}]
+ 19  19 Ket
+ 21     End
+------------------------------------------------------------------
+
+/[\p{Nd}]/8BM
+Memory allocation (code space): 48
+------------------------------------------------------------------
+  0   9 Bra
+  2     [\p{Nd}]
+  9   9 Ket
+ 11     End
+------------------------------------------------------------------
+
+/[\p{Nd}+-]+/8BM
+Memory allocation (code space): 84
+------------------------------------------------------------------
+  0  18 Bra
+  2     [+\-\p{Nd}]+
+ 18  18 Ket
+ 20     End
+------------------------------------------------------------------
+
+/A\x{391}\x{10427}\x{ff3a}\x{1fb0}/8iBM
+Memory allocation (code space): 60
+------------------------------------------------------------------
+  0  12 Bra
+  2  /i A\x{391}\x{10427}\x{ff3a}\x{1fb0}
+ 12  12 Ket
+ 14     End
+------------------------------------------------------------------
+
+/A\x{391}\x{10427}\x{ff3a}\x{1fb0}/8BM
+Memory allocation (code space): 60
+------------------------------------------------------------------
+  0  12 Bra
+  2     A\x{391}\x{10427}\x{ff3a}\x{1fb0}
+ 12  12 Ket
+ 14     End
+------------------------------------------------------------------
+
+/[\x{105}-\x{109}]/8iBM
+Memory allocation (code space): 48
+------------------------------------------------------------------
+  0   9 Bra
+  2     [\x{104}-\x{109}]
+  9   9 Ket
+ 11     End
+------------------------------------------------------------------
+
+/( ( (?(1)0|) )*   )/xBM
+Memory allocation (code space): 104
+------------------------------------------------------------------
+  0  23 Bra
+  2  19 CBra 1
+  5     Brazero
+  6  13 SCBra 2
+  9   6 Cond
+ 11   1 Cond ref
+ 13     0
+ 15   2 Alt
+ 17   8 Ket
+ 19  13 KetRmax
+ 21  19 Ket
+ 23  23 Ket
+ 25     End
+------------------------------------------------------------------
+
+/(  (?(1)0|)*   )/xBM
+Memory allocation (code space): 84
+------------------------------------------------------------------
+  0  18 Bra
+  2  14 CBra 1
+  5     Brazero
+  6   6 SCond
+  8   1 Cond ref
+ 10     0
+ 12   2 Alt
+ 14   8 KetRmax
+ 16  14 Ket
+ 18  18 Ket
+ 20     End
+------------------------------------------------------------------
+
+/[a]/BM
+Memory allocation (code space): 28
+------------------------------------------------------------------
+  0   4 Bra
+  2     a
+  4   4 Ket
+  6     End
+------------------------------------------------------------------
+
+/[a]/8BM
+Memory allocation (code space): 28
+------------------------------------------------------------------
+  0   4 Bra
+  2     a
+  4   4 Ket
+  6     End
+------------------------------------------------------------------
+
+/[\xaa]/BM
+Memory allocation (code space): 28
+------------------------------------------------------------------
+  0   4 Bra
+  2     \x{aa}
+  4   4 Ket
+  6     End
+------------------------------------------------------------------
+
+/[\xaa]/8BM
+Memory allocation (code space): 28
+------------------------------------------------------------------
+  0   4 Bra
+  2     \x{aa}
+  4   4 Ket
+  6     End
+------------------------------------------------------------------
+
+/[^a]/BM
+Memory allocation (code space): 28
+------------------------------------------------------------------
+  0   4 Bra
+  2     [^a]
+  4   4 Ket
+  6     End
+------------------------------------------------------------------
+
+/[^a]/8BM
+Memory allocation (code space): 28
+------------------------------------------------------------------
+  0   4 Bra
+  2     [^a]
+  4   4 Ket
+  6     End
+------------------------------------------------------------------
+
+/[^\xaa]/BM
+Memory allocation (code space): 28
+------------------------------------------------------------------
+  0   4 Bra
+  2     [^\x{aa}]
+  4   4 Ket
+  6     End
+------------------------------------------------------------------
+
+/[^\xaa]/8BM
+Memory allocation (code space): 28
+------------------------------------------------------------------
+  0   4 Bra
+  2     [^\x{aa}]
+  4   4 Ket
+  6     End
+------------------------------------------------------------------
+
+/[^\d]/8WB
+------------------------------------------------------------------
+  0   9 Bra
+  2     [^\p{Nd}]
+  9   9 Ket
+ 11     End
+------------------------------------------------------------------
+
+/[[:^alpha:][:^cntrl:]]+/8WB
+------------------------------------------------------------------
+  0  18 Bra
+  2     [ -~\x80-\xff\P{L}]+
+ 18  18 Ket
+ 20     End
+------------------------------------------------------------------
+
+/[[:^cntrl:][:^alpha:]]+/8WB
+------------------------------------------------------------------
+  0  18 Bra
+  2     [ -~\x80-\xff\P{L}]+
+ 18  18 Ket
+ 20     End
+------------------------------------------------------------------
+
+/[[:alpha:]]+/8WB
+------------------------------------------------------------------
+  0  10 Bra
+  2     [\p{L}]+
+ 10  10 Ket
+ 12     End
+------------------------------------------------------------------
+
+/[[:^alpha:]\S]+/8WB
+------------------------------------------------------------------
+  0  13 Bra
+  2     [\P{L}\P{Xsp}]+
+ 13  13 Ket
+ 15     End
+------------------------------------------------------------------
+
+/abc(d|e)(*THEN)x(123(*THEN)4|567(b|q)(*THEN)xx)/B
+------------------------------------------------------------------
+  0  60 Bra
+  2     abc
+  8   5 CBra 1
+ 11     d
+ 13   4 Alt
+ 15     e
+ 17   9 Ket
+ 19     *THEN
+ 20     x
+ 22  12 CBra 2
+ 25     123
+ 31     *THEN
+ 32     4
+ 34  24 Alt
+ 36     567
+ 42   5 CBra 3
+ 45     b
+ 47   4 Alt
+ 49     q
+ 51   9 Ket
+ 53     *THEN
+ 54     xx
+ 58  36 Ket
+ 60  60 Ket
+ 62     End
+------------------------------------------------------------------
+
+/-- End of testinput11 --/
diff --git a/pcre/testdata/testoutput11-8 b/pcre/testdata/testoutput11-8
new file mode 100644
index 00000000000..c1c85f9a3b6
--- /dev/null
+++ b/pcre/testdata/testoutput11-8
@@ -0,0 +1,713 @@
+/-- These are a few representative patterns whose lengths and offsets are to be 
+shown when the link size is 2. This is just a doublecheck test to ensure the 
+sizes don't go horribly wrong when something is changed. The pattern contents 
+are all themselves checked in other tests. Unicode, including property support, 
+is required for these tests. --/
+
+/((?i)b)/BM
+Memory allocation (code space): 17
+------------------------------------------------------------------
+  0  13 Bra
+  3   7 CBra 1
+  8  /i b
+ 10   7 Ket
+ 13  13 Ket
+ 16     End
+------------------------------------------------------------------
+
+/(?s)(.*X|^B)/BM
+Memory allocation (code space): 25
+------------------------------------------------------------------
+  0  21 Bra
+  3   9 CBra 1
+  8     AllAny*
+ 10     X
+ 12   6 Alt
+ 15     ^
+ 16     B
+ 18  15 Ket
+ 21  21 Ket
+ 24     End
+------------------------------------------------------------------
+
+/(?s:.*X|^B)/BM
+Memory allocation (code space): 23
+------------------------------------------------------------------
+  0  19 Bra
+  3   7 Bra
+  6     AllAny*
+  8     X
+ 10   6 Alt
+ 13     ^
+ 14     B
+ 16  13 Ket
+ 19  19 Ket
+ 22     End
+------------------------------------------------------------------
+
+/^[[:alnum:]]/BM
+Memory allocation (code space): 41
+------------------------------------------------------------------
+  0  37 Bra
+  3     ^
+  4     [0-9A-Za-z]
+ 37  37 Ket
+ 40     End
+------------------------------------------------------------------
+
+/#/IxMD
+Memory allocation (code space): 7
+------------------------------------------------------------------
+  0   3 Bra
+  3   3 Ket
+  6     End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: extended
+No first char
+No need char
+
+/a#/IxMD
+Memory allocation (code space): 9
+------------------------------------------------------------------
+  0   5 Bra
+  3     a
+  5   5 Ket
+  8     End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: extended
+First char = 'a'
+No need char
+
+/x?+/BM
+Memory allocation (code space): 9
+------------------------------------------------------------------
+  0   5 Bra
+  3     x?+
+  5   5 Ket
+  8     End
+------------------------------------------------------------------
+
+/x++/BM
+Memory allocation (code space): 9
+------------------------------------------------------------------
+  0   5 Bra
+  3     x++
+  5   5 Ket
+  8     End
+------------------------------------------------------------------
+
+/x{1,3}+/BM 
+Memory allocation (code space): 19
+------------------------------------------------------------------
+  0  15 Bra
+  3   9 Once
+  6     x
+  8     x{0,2}
+ 12   9 Ket
+ 15  15 Ket
+ 18     End
+------------------------------------------------------------------
+
+/(x)*+/BM
+Memory allocation (code space): 18
+------------------------------------------------------------------
+  0  14 Bra
+  3     Braposzero
+  4   7 CBraPos 1
+  9     x
+ 11   7 KetRpos
+ 14  14 Ket
+ 17     End
+------------------------------------------------------------------
+
+/^((a+)(?U)([ab]+)(?-U)([bc]+)(\w*))/BM
+Memory allocation (code space): 120
+------------------------------------------------------------------
+  0 116 Bra
+  3     ^
+  4 109 CBra 1
+  9   7 CBra 2
+ 14     a+
+ 16   7 Ket
+ 19  39 CBra 3
+ 24     [ab]+?
+ 58  39 Ket
+ 61  39 CBra 4
+ 66     [bc]+
+100  39 Ket
+103   7 CBra 5
+108     \w*
+110   7 Ket
+113 109 Ket
+116 116 Ket
+119     End
+------------------------------------------------------------------
+
+|8J\$WE\<\.rX\+ix\[d1b\!H\#\?vV0vrK\:ZH1\=2M\>iV\;\?aPhFB\<\*vW\@QW\@sO9\}cfZA\-i\'w\%hKd6gt1UJP\,15_\#QY\$M\^Mss_U\/\]\&LK9\[5vQub\^w\[KDD\qmj\;2\}YWFdYx\.Ap\]hjCPTP\(n28k\+3\;o\&WXqs\/gOXdr\$\:r\'do0\;b4c\(f_Gr\=\"\\4\)\[01T7ajQJvL\$W\~mL_sS\/4h\:x\*\[ZN\=KLs\&L5zX\/\/\>it\,o\:aU\(\;Z\>pW\&T7oP\'2K\^E\:x9\'c\[\%z\-\,64JQ5AeH_G\#KijUKghQw\^\\vea3a\?kka_G\$8\#\`\*kynsxzBLru\'\]k_\[7FrVx\}\^\=\$blx\>s\-N\%j\;D\*aZDnsw\:YKZ\%Q\.Kne9\#hP\?\+b3\(SOvL\,\^\;\&u5\@\?5C5Bhb\=m\-vEh_L15Jl\]U\)0RP6\{q\%L\^_z5E\'Dw6X\b|BM
+Memory allocation (code space): 826
+------------------------------------------------------------------
+  0 822 Bra
+  3     8J$WE<.rX+ix[d1b!H#?vV0vrK:ZH1=2M>iV;?aPhFB<*vW@QW@sO9}cfZA-i'w%hKd6gt1UJP,15_#QY$M^Mss_U/]&LK9[5vQub^w[KDDqmj;2}YWFdYx.Ap]hjCPTP(n28k+3;o&WXqs/gOXdr$:r'do0;b4c(f_Gr="\4)[01T7ajQJvL$W~mL_sS/4h:x*[ZN=KLs&L5zX//>it,o:aU(;Z>pW&T7oP'2K^E:x9'c[%z-,64JQ5AeH_G#KijUKghQw^\vea3a?kka_G$8#`*kynsxzBLru']k_[7FrVx}^=$blx>s-N%j;D*aZDnsw:YKZ%Q.Kne9#hP?+b3(SOvL,^;&u5@?5C5Bhb=m-vEh_L15Jl]U)0RP6{q%L^_z5E'Dw6X
+821     \b
+822 822 Ket
+825     End
+------------------------------------------------------------------
+
+|\$\<\.X\+ix\[d1b\!H\#\?vV0vrK\:ZH1\=2M\>iV\;\?aPhFB\<\*vW\@QW\@sO9\}cfZA\-i\'w\%hKd6gt1UJP\,15_\#QY\$M\^Mss_U\/\]\&LK9\[5vQub\^w\[KDD\qmj\;2\}YWFdYx\.Ap\]hjCPTP\(n28k\+3\;o\&WXqs\/gOXdr\$\:r\'do0\;b4c\(f_Gr\=\"\\4\)\[01T7ajQJvL\$W\~mL_sS\/4h\:x\*\[ZN\=KLs\&L5zX\/\/\>it\,o\:aU\(\;Z\>pW\&T7oP\'2K\^E\:x9\'c\[\%z\-\,64JQ5AeH_G\#KijUKghQw\^\\vea3a\?kka_G\$8\#\`\*kynsxzBLru\'\]k_\[7FrVx\}\^\=\$blx\>s\-N\%j\;D\*aZDnsw\:YKZ\%Q\.Kne9\#hP\?\+b3\(SOvL\,\^\;\&u5\@\?5C5Bhb\=m\-vEh_L15Jl\]U\)0RP6\{q\%L\^_z5E\'Dw6X\b|BM
+Memory allocation (code space): 816
+------------------------------------------------------------------
+  0 812 Bra
+  3     $<.X+ix[d1b!H#?vV0vrK:ZH1=2M>iV;?aPhFB<*vW@QW@sO9}cfZA-i'w%hKd6gt1UJP,15_#QY$M^Mss_U/]&LK9[5vQub^w[KDDqmj;2}YWFdYx.Ap]hjCPTP(n28k+3;o&WXqs/gOXdr$:r'do0;b4c(f_Gr="\4)[01T7ajQJvL$W~mL_sS/4h:x*[ZN=KLs&L5zX//>it,o:aU(;Z>pW&T7oP'2K^E:x9'c[%z-,64JQ5AeH_G#KijUKghQw^\vea3a?kka_G$8#`*kynsxzBLru']k_[7FrVx}^=$blx>s-N%j;D*aZDnsw:YKZ%Q.Kne9#hP?+b3(SOvL,^;&u5@?5C5Bhb=m-vEh_L15Jl]U)0RP6{q%L^_z5E'Dw6X
+811     \b
+812 812 Ket
+815     End
+------------------------------------------------------------------
+
+/(a(?1)b)/BM
+Memory allocation (code space): 22
+------------------------------------------------------------------
+  0  18 Bra
+  3  12 CBra 1
+  8     a
+ 10   3 Recurse
+ 13     b
+ 15  12 Ket
+ 18  18 Ket
+ 21     End
+------------------------------------------------------------------
+
+/(a(?1)+b)/BM
+Memory allocation (code space): 28
+------------------------------------------------------------------
+  0  24 Bra
+  3  18 CBra 1
+  8     a
+ 10   6 Once
+ 13   3 Recurse
+ 16   6 KetRmax
+ 19     b
+ 21  18 Ket
+ 24  24 Ket
+ 27     End
+------------------------------------------------------------------
+
+/a(?Pb|c)d(?Pe)/BM
+Memory allocation (code space): 36
+------------------------------------------------------------------
+  0  32 Bra
+  3     a
+  5   7 CBra 1
+ 10     b
+ 12   5 Alt
+ 15     c
+ 17  12 Ket
+ 20     d
+ 22   7 CBra 2
+ 27     e
+ 29   7 Ket
+ 32  32 Ket
+ 35     End
+------------------------------------------------------------------
+
+/(?:a(?Pc(?Pd)))(?Pa)/BM
+Memory allocation (code space): 45
+------------------------------------------------------------------
+  0  41 Bra
+  3  25 Bra
+  6     a
+  8  17 CBra 1
+ 13     c
+ 15   7 CBra 2
+ 20     d
+ 22   7 Ket
+ 25  17 Ket
+ 28  25 Ket
+ 31   7 CBra 3
+ 36     a
+ 38   7 Ket
+ 41  41 Ket
+ 44     End
+------------------------------------------------------------------
+
+/(?Pa)...(?P=a)bbb(?P>a)d/BM
+Memory allocation (code space): 34
+------------------------------------------------------------------
+  0  30 Bra
+  3   7 CBra 1
+  8     a
+ 10   7 Ket
+ 13     Any
+ 14     Any
+ 15     Any
+ 16     \1
+ 19     bbb
+ 25   3 Recurse
+ 28     d
+ 30  30 Ket
+ 33     End
+------------------------------------------------------------------
+
+/abc(?C255)de(?C)f/BM
+Memory allocation (code space): 31
+------------------------------------------------------------------
+  0  27 Bra
+  3     abc
+  9     Callout 255 10 1
+ 15     de
+ 19     Callout 0 16 1
+ 25     f
+ 27  27 Ket
+ 30     End
+------------------------------------------------------------------
+
+/abcde/CBM
+Memory allocation (code space): 53
+------------------------------------------------------------------
+  0  49 Bra
+  3     Callout 255 0 1
+  9     a
+ 11     Callout 255 1 1
+ 17     b
+ 19     Callout 255 2 1
+ 25     c
+ 27     Callout 255 3 1
+ 33     d
+ 35     Callout 255 4 1
+ 41     e
+ 43     Callout 255 5 0
+ 49  49 Ket
+ 52     End
+------------------------------------------------------------------
+
+/\x{100}/8BM
+Memory allocation (code space): 10
+------------------------------------------------------------------
+  0   6 Bra
+  3     \x{100}
+  6   6 Ket
+  9     End
+------------------------------------------------------------------
+
+/\x{1000}/8BM
+Memory allocation (code space): 11
+------------------------------------------------------------------
+  0   7 Bra
+  3     \x{1000}
+  7   7 Ket
+ 10     End
+------------------------------------------------------------------
+
+/\x{10000}/8BM
+Memory allocation (code space): 12
+------------------------------------------------------------------
+  0   8 Bra
+  3     \x{10000}
+  8   8 Ket
+ 11     End
+------------------------------------------------------------------
+
+/\x{100000}/8BM
+Memory allocation (code space): 12
+------------------------------------------------------------------
+  0   8 Bra
+  3     \x{100000}
+  8   8 Ket
+ 11     End
+------------------------------------------------------------------
+
+/\x{10ffff}/8BM
+Memory allocation (code space): 12
+------------------------------------------------------------------
+  0   8 Bra
+  3     \x{10ffff}
+  8   8 Ket
+ 11     End
+------------------------------------------------------------------
+
+/\x{110000}/8BM
+Failed: character value in \x{...} sequence is too large at offset 9
+
+/[\x{ff}]/8BM
+Memory allocation (code space): 10
+------------------------------------------------------------------
+  0   6 Bra
+  3     \x{ff}
+  6   6 Ket
+  9     End
+------------------------------------------------------------------
+
+/[\x{100}]/8BM
+Memory allocation (code space): 10
+------------------------------------------------------------------
+  0   6 Bra
+  3     \x{100}
+  6   6 Ket
+  9     End
+------------------------------------------------------------------
+
+/\x80/8BM
+Memory allocation (code space): 10
+------------------------------------------------------------------
+  0   6 Bra
+  3     \x{80}
+  6   6 Ket
+  9     End
+------------------------------------------------------------------
+
+/\xff/8BM
+Memory allocation (code space): 10
+------------------------------------------------------------------
+  0   6 Bra
+  3     \x{ff}
+  6   6 Ket
+  9     End
+------------------------------------------------------------------
+
+/\x{0041}\x{2262}\x{0391}\x{002e}/D8M
+Memory allocation (code space): 18
+------------------------------------------------------------------
+  0  14 Bra
+  3     A\x{2262}\x{391}.
+ 14  14 Ket
+ 17     End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = 'A'
+Need char = '.'
+    
+/\x{D55c}\x{ad6d}\x{C5B4}/D8M 
+Memory allocation (code space): 19
+------------------------------------------------------------------
+  0  15 Bra
+  3     \x{d55c}\x{ad6d}\x{c5b4}
+ 15  15 Ket
+ 18     End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{ed}
+Need char = \x{b4}
+
+/\x{65e5}\x{672c}\x{8a9e}/D8M
+Memory allocation (code space): 19
+------------------------------------------------------------------
+  0  15 Bra
+  3     \x{65e5}\x{672c}\x{8a9e}
+ 15  15 Ket
+ 18     End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{e6}
+Need char = \x{9e}
+
+/[\x{100}]/8BM
+Memory allocation (code space): 10
+------------------------------------------------------------------
+  0   6 Bra
+  3     \x{100}
+  6   6 Ket
+  9     End
+------------------------------------------------------------------
+
+/[Z\x{100}]/8BM
+Memory allocation (code space): 47
+------------------------------------------------------------------
+  0  43 Bra
+  3     [Z\x{100}]
+ 43  43 Ket
+ 46     End
+------------------------------------------------------------------
+
+/^[\x{100}\E-\Q\E\x{150}]/B8M
+Memory allocation (code space): 18
+------------------------------------------------------------------
+  0  14 Bra
+  3     ^
+  4     [\x{100}-\x{150}]
+ 14  14 Ket
+ 17     End
+------------------------------------------------------------------
+
+/^[\QĀ\E-\QŐ\E]/B8M
+Memory allocation (code space): 18
+------------------------------------------------------------------
+  0  14 Bra
+  3     ^
+  4     [\x{100}-\x{150}]
+ 14  14 Ket
+ 17     End
+------------------------------------------------------------------
+
+/^[\QĀ\E-\QŐ\E/B8M
+Failed: missing terminating ] for character class at offset 15
+
+/[\p{L}]/BM
+Memory allocation (code space): 15
+------------------------------------------------------------------
+  0  11 Bra
+  3     [\p{L}]
+ 11  11 Ket
+ 14     End
+------------------------------------------------------------------
+
+/[\p{^L}]/BM
+Memory allocation (code space): 15
+------------------------------------------------------------------
+  0  11 Bra
+  3     [\P{L}]
+ 11  11 Ket
+ 14     End
+------------------------------------------------------------------
+
+/[\P{L}]/BM
+Memory allocation (code space): 15
+------------------------------------------------------------------
+  0  11 Bra
+  3     [\P{L}]
+ 11  11 Ket
+ 14     End
+------------------------------------------------------------------
+
+/[\P{^L}]/BM
+Memory allocation (code space): 15
+------------------------------------------------------------------
+  0  11 Bra
+  3     [\p{L}]
+ 11  11 Ket
+ 14     End
+------------------------------------------------------------------
+
+/[abc\p{L}\x{0660}]/8BM
+Memory allocation (code space): 50
+------------------------------------------------------------------
+  0  46 Bra
+  3     [a-c\p{L}\x{660}]
+ 46  46 Ket
+ 49     End
+------------------------------------------------------------------
+
+/[\p{Nd}]/8BM
+Memory allocation (code space): 15
+------------------------------------------------------------------
+  0  11 Bra
+  3     [\p{Nd}]
+ 11  11 Ket
+ 14     End
+------------------------------------------------------------------
+
+/[\p{Nd}+-]+/8BM
+Memory allocation (code space): 48
+------------------------------------------------------------------
+  0  44 Bra
+  3     [+\-\p{Nd}]+
+ 44  44 Ket
+ 47     End
+------------------------------------------------------------------
+
+/A\x{391}\x{10427}\x{ff3a}\x{1fb0}/8iBM
+Memory allocation (code space): 25
+------------------------------------------------------------------
+  0  21 Bra
+  3  /i A\x{391}\x{10427}\x{ff3a}\x{1fb0}
+ 21  21 Ket
+ 24     End
+------------------------------------------------------------------
+
+/A\x{391}\x{10427}\x{ff3a}\x{1fb0}/8BM
+Memory allocation (code space): 25
+------------------------------------------------------------------
+  0  21 Bra
+  3     A\x{391}\x{10427}\x{ff3a}\x{1fb0}
+ 21  21 Ket
+ 24     End
+------------------------------------------------------------------
+
+/[\x{105}-\x{109}]/8iBM
+Memory allocation (code space): 17
+------------------------------------------------------------------
+  0  13 Bra
+  3     [\x{104}-\x{109}]
+ 13  13 Ket
+ 16     End
+------------------------------------------------------------------
+
+/( ( (?(1)0|) )*   )/xBM
+Memory allocation (code space): 38
+------------------------------------------------------------------
+  0  34 Bra
+  3  28 CBra 1
+  8     Brazero
+  9  19 SCBra 2
+ 14   8 Cond
+ 17   1 Cond ref
+ 20     0
+ 22   3 Alt
+ 25  11 Ket
+ 28  19 KetRmax
+ 31  28 Ket
+ 34  34 Ket
+ 37     End
+------------------------------------------------------------------
+
+/(  (?(1)0|)*   )/xBM
+Memory allocation (code space): 30
+------------------------------------------------------------------
+  0  26 Bra
+  3  20 CBra 1
+  8     Brazero
+  9   8 SCond
+ 12   1 Cond ref
+ 15     0
+ 17   3 Alt
+ 20  11 KetRmax
+ 23  20 Ket
+ 26  26 Ket
+ 29     End
+------------------------------------------------------------------
+
+/[a]/BM
+Memory allocation (code space): 9
+------------------------------------------------------------------
+  0   5 Bra
+  3     a
+  5   5 Ket
+  8     End
+------------------------------------------------------------------
+
+/[a]/8BM
+Memory allocation (code space): 9
+------------------------------------------------------------------
+  0   5 Bra
+  3     a
+  5   5 Ket
+  8     End
+------------------------------------------------------------------
+
+/[\xaa]/BM
+Memory allocation (code space): 9
+------------------------------------------------------------------
+  0   5 Bra
+  3     \x{aa}
+  5   5 Ket
+  8     End
+------------------------------------------------------------------
+
+/[\xaa]/8BM
+Memory allocation (code space): 10
+------------------------------------------------------------------
+  0   6 Bra
+  3     \x{aa}
+  6   6 Ket
+  9     End
+------------------------------------------------------------------
+
+/[^a]/BM
+Memory allocation (code space): 9
+------------------------------------------------------------------
+  0   5 Bra
+  3     [^a]
+  5   5 Ket
+  8     End
+------------------------------------------------------------------
+
+/[^a]/8BM
+Memory allocation (code space): 9
+------------------------------------------------------------------
+  0   5 Bra
+  3     [^a]
+  5   5 Ket
+  8     End
+------------------------------------------------------------------
+
+/[^\xaa]/BM
+Memory allocation (code space): 9
+------------------------------------------------------------------
+  0   5 Bra
+  3     [^\x{aa}]
+  5   5 Ket
+  8     End
+------------------------------------------------------------------
+
+/[^\xaa]/8BM
+Memory allocation (code space): 10
+------------------------------------------------------------------
+  0   6 Bra
+  3     [^\x{aa}]
+  6   6 Ket
+  9     End
+------------------------------------------------------------------
+
+/[^\d]/8WB
+------------------------------------------------------------------
+  0  11 Bra
+  3     [^\p{Nd}]
+ 11  11 Ket
+ 14     End
+------------------------------------------------------------------
+
+/[[:^alpha:][:^cntrl:]]+/8WB
+------------------------------------------------------------------
+  0  44 Bra
+  3     [ -~\x80-\xff\P{L}]+
+ 44  44 Ket
+ 47     End
+------------------------------------------------------------------
+
+/[[:^cntrl:][:^alpha:]]+/8WB
+------------------------------------------------------------------
+  0  44 Bra
+  3     [ -~\x80-\xff\P{L}]+
+ 44  44 Ket
+ 47     End
+------------------------------------------------------------------
+
+/[[:alpha:]]+/8WB
+------------------------------------------------------------------
+  0  12 Bra
+  3     [\p{L}]+
+ 12  12 Ket
+ 15     End
+------------------------------------------------------------------
+
+/[[:^alpha:]\S]+/8WB
+------------------------------------------------------------------
+  0  15 Bra
+  3     [\P{L}\P{Xsp}]+
+ 15  15 Ket
+ 18     End
+------------------------------------------------------------------
+
+/abc(d|e)(*THEN)x(123(*THEN)4|567(b|q)(*THEN)xx)/B
+------------------------------------------------------------------
+  0  73 Bra
+  3     abc
+  9   7 CBra 1
+ 14     d
+ 16   5 Alt
+ 19     e
+ 21  12 Ket
+ 24     *THEN
+ 25     x
+ 27  14 CBra 2
+ 32     123
+ 38     *THEN
+ 39     4
+ 41  29 Alt
+ 44     567
+ 50   7 CBra 3
+ 55     b
+ 57   5 Alt
+ 60     q
+ 62  12 Ket
+ 65     *THEN
+ 66     xx
+ 70  43 Ket
+ 73  73 Ket
+ 76     End
+------------------------------------------------------------------
+
+/-- End of testinput11 --/
diff --git a/pcre/testdata/testoutput12 b/pcre/testdata/testoutput12
new file mode 100644
index 00000000000..6c986d1185d
--- /dev/null
+++ b/pcre/testdata/testoutput12
@@ -0,0 +1,176 @@
+/-- This test is run only when JIT support is available. It checks for a
+successful and an unsuccessful JIT compile and save and restore behaviour,
+and a couple of things that are different with JIT. --/
+
+/abc/S+I
+Capturing subpattern count = 0
+No options
+First char = 'a'
+Need char = 'c'
+Subject length lower bound = 3
+No set of starting bytes
+JIT study was successful
+
+/(?(?C1)(?=a)a)/S+I
+Capturing subpattern count = 0
+No options
+First char = 'a'
+No need char
+Study returned NULL
+JIT study was not successful
+
+/(?(?C1)(?=a)a)/S!+I
+Capturing subpattern count = 0
+No options
+First char = 'a'
+No need char
+Subject length lower bound = -1
+No set of starting bytes
+JIT study was not successful
+
+/abc/S+I>testsavedregex
+Capturing subpattern count = 0
+No options
+First char = 'a'
+Need char = 'c'
+Subject length lower bound = 3
+No set of starting bytes
+JIT study was successful
+Compiled pattern written to testsavedregex
+Study data written to testsavedregex
+
+b)c/PN
+    abc
+Matched with REG_NOSUB
+
+/a?|b?/P
+    abc
+ 0: a
+    ** Failers
+ 0: 
+    ddd\N   
+No match: POSIX code 17: match failed
+
+/\w+A/P
+   CDAAAAB 
+ 0: CDAAAA
+
+/\w+A/PU
+   CDAAAAB 
+ 0: CDA
+   
+/\Biss\B/I+P
+    Mississippi
+ 0: iss
+ 0+ issippi
+
+/abc/\P
+Failed: POSIX code 9: bad escape sequence at offset 4     
+
+/-- End of POSIX tests --/ 
+
+/a\Cb/
+    aXb
+ 0: aXb
+    a\nb
+ 0: a\x0ab
+    ** Failers (too big char) 
+No match
+    A\x{123}B 
+** Character \x{123} is greater than 255 and UTF-8 mode is not enabled.
+** Truncation will probably give the wrong result.
+No match
+  
+/\x{100}/I
+Failed: character value in \x{...} sequence is too large at offset 6
+
+/  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*                          # optional leading comment
+(?:    (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+)                    # initial word
+(?:  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+)  )* # further okay, if led by a period
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  @  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*    (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                           # initial subdomain
+(?:                                  #
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.                        # if led by a period...
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                     #   ...further okay
+)*
+# address
+|                     #  or
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+)             # one word, optionally followed by....
+(?:
+[^()<>@,;:".\\\[\]\x80-\xff\000-\010\012-\037]  |  # atom and space parts, or...
+\(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)       |  # comments, or...
+
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+# quoted strings
+)*
+<  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*                     # leading <
+(?:  @  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*    (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                           # initial subdomain
+(?:                                  #
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.                        # if led by a period...
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                     #   ...further okay
+)*
+
+(?:  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  ,  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  @  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*    (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                           # initial subdomain
+(?:                                  #
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.                        # if led by a period...
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                     #   ...further okay
+)*
+)* # further okay, if led by comma
+:                                # closing colon
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  )? #       optional route
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+)                    # initial word
+(?:  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+)  )* # further okay, if led by a period
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  @  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*    (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                           # initial subdomain
+(?:                                  #
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.                        # if led by a period...
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                     #   ...further okay
+)*
+#       address spec
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  > #                  trailing >
+# name and address
+)  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*                       # optional trailing comment
+/xSI
+Capturing subpattern count = 0
+Contains explicit CR or LF match
+Options: extended
+No first char
+No need char
+Subject length lower bound = 3
+Starting byte set: \x09 \x20 ! " # $ % & ' ( * + - / 0 1 2 3 4 5 6 7 8 
+  9 = ? A B C D E F G H I J K L M N O P Q R S T U V W X Y Z ^ _ ` a b c d e 
+  f g h i j k l m n o p q r s t u v w x y z { | } ~ \x7f 
+
+/-- Although this saved pattern was compiled with link-size=2, it does no harm
+to run this test with other link sizes because it is going to generated a
+"compiled in wrong mode" error as soon as it is loaded, so the link size does 
+not matter. --/
+
+\x09<
+ 0: \x09
+
+/[\h]+/BZ
+------------------------------------------------------------------
+        Bra
+        [\x09 \xa0]+
+        Ket
+        End
+------------------------------------------------------------------
+    >\x09\x20\xa0<
+ 0: \x09 \xa0
+
+/[\v]/BZ
+------------------------------------------------------------------
+        Bra
+        [\x0a-\x0d\x85]
+        Ket
+        End
+------------------------------------------------------------------
+
+/[\H]/BZ
+------------------------------------------------------------------
+        Bra
+        [\x00-\x08\x0a-\x1f!-\x9f\xa1-\xff]
+        Ket
+        End
+------------------------------------------------------------------
+
+/[^\h]/BZ
+------------------------------------------------------------------
+        Bra
+        [\x00-\x08\x0a-\x1f!-\x9f\xa1-\xff] (neg)
+        Ket
+        End
+------------------------------------------------------------------
+
+/[\V]/BZ
+------------------------------------------------------------------
+        Bra
+        [\x00-\x09\x0e-\x84\x86-\xff]
+        Ket
+        End
+------------------------------------------------------------------
+
+/[\x0a\V]/BZ
+------------------------------------------------------------------
+        Bra
+        [\x00-\x0a\x0e-\x84\x86-\xff]
+        Ket
+        End
+------------------------------------------------------------------
+
+/\777/I
+Failed: octal value is greater than \377 in 8-bit non-UTF-8 mode at offset 3
+
+/(*:0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF)XX/K
+Failed: name is too long in (*MARK), (*PRUNE), (*SKIP), or (*THEN) at offset 259
+
+/(*:0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDE)XX/K
+    XX
+ 0: XX
+MK: 0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDE
+
+/\u0100/
+Failed: character value in \u.... sequence is too large at offset 5
+
+/[\u0100-\u0200]/
+Failed: character value in \u.... sequence is too large at offset 6
+
+/-- End of testinput14 --/
diff --git a/pcre/testdata/testoutput15 b/pcre/testdata/testoutput15
new file mode 100644
index 00000000000..1541637de8d
--- /dev/null
+++ b/pcre/testdata/testoutput15
@@ -0,0 +1,1137 @@
+/-- This set of tests is for UTF-8 support, and is relevant only to the 8-bit 
+    library. --/
+
+/X(\C{3})/8
+    X\x{1234}
+ 0: X\x{1234}
+ 1: \x{1234}
+
+/X(\C{4})/8
+    X\x{1234}YZ
+ 0: X\x{1234}Y
+ 1: \x{1234}Y
+    
+/X\C*/8
+    XYZabcdce
+ 0: XYZabcdce
+    
+/X\C*?/8
+    XYZabcde
+ 0: X
+    
+/X\C{3,5}/8
+    Xabcdefg   
+ 0: Xabcde
+    X\x{1234} 
+ 0: X\x{1234}
+    X\x{1234}YZ
+ 0: X\x{1234}YZ
+    X\x{1234}\x{512}  
+ 0: X\x{1234}\x{512}
+    X\x{1234}\x{512}YZ
+ 0: X\x{1234}\x{512}
+
+/X\C{3,5}?/8
+    Xabcdefg   
+ 0: Xabc
+    X\x{1234} 
+ 0: X\x{1234}
+    X\x{1234}YZ
+ 0: X\x{1234}
+    X\x{1234}\x{512}  
+ 0: X\x{1234}
+
+/a\Cb/8
+    aXb
+ 0: aXb
+    a\nb
+ 0: a\x{0a}b
+    
+/a\C\Cb/8 
+    a\x{100}b 
+ 0: a\x{100}b
+
+/ab\Cde/8
+    abXde
+ 0: abXde
+
+/a\C\Cb/8 
+    a\x{100}b
+ 0: a\x{100}b
+    ** Failers 
+No match
+    a\x{12257}b
+No match
+
+/[]/8
+Failed: invalid UTF-8 string at offset 1
+
+//8
+Failed: invalid UTF-8 string at offset 0
+
+/xxx/8
+Failed: invalid UTF-8 string at offset 0
+
+/xxx/8?DZSS
+------------------------------------------------------------------
+        Bra
+        \X{c0}\X{c0}\X{c0}xxx
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf no_utf_check
+First char = \x{c3}
+Need char = 'x'
+
+/badutf/8
+    \xdf
+Error -10 (bad UTF-8 string) offset=0 reason=1
+    \xef
+Error -10 (bad UTF-8 string) offset=0 reason=2
+    \xef\x80
+Error -10 (bad UTF-8 string) offset=0 reason=1
+    \xf7
+Error -10 (bad UTF-8 string) offset=0 reason=3
+    \xf7\x80
+Error -10 (bad UTF-8 string) offset=0 reason=2
+    \xf7\x80\x80
+Error -10 (bad UTF-8 string) offset=0 reason=1
+    \xfb
+Error -10 (bad UTF-8 string) offset=0 reason=4
+    \xfb\x80
+Error -10 (bad UTF-8 string) offset=0 reason=3
+    \xfb\x80\x80
+Error -10 (bad UTF-8 string) offset=0 reason=2
+    \xfb\x80\x80\x80
+Error -10 (bad UTF-8 string) offset=0 reason=1
+    \xfd
+Error -10 (bad UTF-8 string) offset=0 reason=5
+    \xfd\x80
+Error -10 (bad UTF-8 string) offset=0 reason=4
+    \xfd\x80\x80
+Error -10 (bad UTF-8 string) offset=0 reason=3
+    \xfd\x80\x80\x80
+Error -10 (bad UTF-8 string) offset=0 reason=2
+    \xfd\x80\x80\x80\x80
+Error -10 (bad UTF-8 string) offset=0 reason=1
+    \xdf\x7f
+Error -10 (bad UTF-8 string) offset=0 reason=6
+    \xef\x7f\x80
+Error -10 (bad UTF-8 string) offset=0 reason=6
+    \xef\x80\x7f
+Error -10 (bad UTF-8 string) offset=0 reason=7
+    \xf7\x7f\x80\x80
+Error -10 (bad UTF-8 string) offset=0 reason=6
+    \xf7\x80\x7f\x80
+Error -10 (bad UTF-8 string) offset=0 reason=7
+    \xf7\x80\x80\x7f
+Error -10 (bad UTF-8 string) offset=0 reason=8
+    \xfb\x7f\x80\x80\x80
+Error -10 (bad UTF-8 string) offset=0 reason=6
+    \xfb\x80\x7f\x80\x80
+Error -10 (bad UTF-8 string) offset=0 reason=7
+    \xfb\x80\x80\x7f\x80
+Error -10 (bad UTF-8 string) offset=0 reason=8
+    \xfb\x80\x80\x80\x7f
+Error -10 (bad UTF-8 string) offset=0 reason=9
+    \xfd\x7f\x80\x80\x80\x80
+Error -10 (bad UTF-8 string) offset=0 reason=6
+    \xfd\x80\x7f\x80\x80\x80
+Error -10 (bad UTF-8 string) offset=0 reason=7
+    \xfd\x80\x80\x7f\x80\x80
+Error -10 (bad UTF-8 string) offset=0 reason=8
+    \xfd\x80\x80\x80\x7f\x80
+Error -10 (bad UTF-8 string) offset=0 reason=9
+    \xfd\x80\x80\x80\x80\x7f
+Error -10 (bad UTF-8 string) offset=0 reason=10
+    \xed\xa0\x80
+Error -10 (bad UTF-8 string) offset=0 reason=14
+    \xc0\x8f
+Error -10 (bad UTF-8 string) offset=0 reason=15
+    \xe0\x80\x8f
+Error -10 (bad UTF-8 string) offset=0 reason=16
+    \xf0\x80\x80\x8f
+Error -10 (bad UTF-8 string) offset=0 reason=17
+    \xf8\x80\x80\x80\x8f
+Error -10 (bad UTF-8 string) offset=0 reason=18
+    \xfc\x80\x80\x80\x80\x8f
+Error -10 (bad UTF-8 string) offset=0 reason=19
+    \x80
+Error -10 (bad UTF-8 string) offset=0 reason=20
+    \xfe
+Error -10 (bad UTF-8 string) offset=0 reason=21
+    \xff
+Error -10 (bad UTF-8 string) offset=0 reason=21
+
+/badutf/8
+    \xfb\x80\x80\x80\x80
+Error -10 (bad UTF-8 string) offset=0 reason=11
+    \xfd\x80\x80\x80\x80\x80
+Error -10 (bad UTF-8 string) offset=0 reason=12
+    \xf7\xbf\xbf\xbf
+Error -10 (bad UTF-8 string) offset=0 reason=13
+
+/shortutf/8
+    \P\P\xdf
+Error -25 (short UTF-8 string) offset=0 reason=1
+    \P\P\xef
+Error -25 (short UTF-8 string) offset=0 reason=2
+    \P\P\xef\x80
+Error -25 (short UTF-8 string) offset=0 reason=1
+    \P\P\xf7
+Error -25 (short UTF-8 string) offset=0 reason=3
+    \P\P\xf7\x80
+Error -25 (short UTF-8 string) offset=0 reason=2
+    \P\P\xf7\x80\x80
+Error -25 (short UTF-8 string) offset=0 reason=1
+    \P\P\xfb
+Error -25 (short UTF-8 string) offset=0 reason=4
+    \P\P\xfb\x80
+Error -25 (short UTF-8 string) offset=0 reason=3
+    \P\P\xfb\x80\x80
+Error -25 (short UTF-8 string) offset=0 reason=2
+    \P\P\xfb\x80\x80\x80
+Error -25 (short UTF-8 string) offset=0 reason=1
+    \P\P\xfd
+Error -25 (short UTF-8 string) offset=0 reason=5
+    \P\P\xfd\x80
+Error -25 (short UTF-8 string) offset=0 reason=4
+    \P\P\xfd\x80\x80
+Error -25 (short UTF-8 string) offset=0 reason=3
+    \P\P\xfd\x80\x80\x80
+Error -25 (short UTF-8 string) offset=0 reason=2
+    \P\P\xfd\x80\x80\x80\x80
+Error -25 (short UTF-8 string) offset=0 reason=1
+
+/anything/8
+    \xc0\x80
+Error -10 (bad UTF-8 string) offset=0 reason=15
+    \xc1\x8f 
+Error -10 (bad UTF-8 string) offset=0 reason=15
+    \xe0\x9f\x80
+Error -10 (bad UTF-8 string) offset=0 reason=16
+    \xf0\x8f\x80\x80 
+Error -10 (bad UTF-8 string) offset=0 reason=17
+    \xf8\x87\x80\x80\x80  
+Error -10 (bad UTF-8 string) offset=0 reason=18
+    \xfc\x83\x80\x80\x80\x80
+Error -10 (bad UTF-8 string) offset=0 reason=19
+    \xfe\x80\x80\x80\x80\x80  
+Error -10 (bad UTF-8 string) offset=0 reason=21
+    \xff\x80\x80\x80\x80\x80  
+Error -10 (bad UTF-8 string) offset=0 reason=21
+    \xc3\x8f
+No match
+    \xe0\xaf\x80
+No match
+    \xe1\x80\x80
+No match
+    \xf0\x9f\x80\x80 
+No match
+    \xf1\x8f\x80\x80 
+No match
+    \xf8\x88\x80\x80\x80  
+Error -10 (bad UTF-8 string) offset=0 reason=11
+    \xf9\x87\x80\x80\x80  
+Error -10 (bad UTF-8 string) offset=0 reason=11
+    \xfc\x84\x80\x80\x80\x80
+Error -10 (bad UTF-8 string) offset=0 reason=12
+    \xfd\x83\x80\x80\x80\x80
+Error -10 (bad UTF-8 string) offset=0 reason=12
+    \?\xf8\x88\x80\x80\x80  
+No match
+    \?\xf9\x87\x80\x80\x80  
+No match
+    \?\xfc\x84\x80\x80\x80\x80
+No match
+    \?\xfd\x83\x80\x80\x80\x80
+No match
+
+/\x{100}/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{100}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{c4}
+Need char = \x{80}
+
+/\x{1000}/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{1000}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{e1}
+Need char = \x{80}
+
+/\x{10000}/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{10000}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{f0}
+Need char = \x{80}
+
+/\x{100000}/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{100000}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{f4}
+Need char = \x{80}
+
+/\x{10ffff}/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{10ffff}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{f4}
+Need char = \x{bf}
+
+/[\x{ff}]/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{ff}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{c3}
+Need char = \x{bf}
+
+/[\x{100}]/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{100}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{c4}
+Need char = \x{80}
+
+/\x80/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{80}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{c2}
+Need char = \x{80}
+
+/\xff/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{ff}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{c3}
+Need char = \x{bf}
+
+/\x{D55c}\x{ad6d}\x{C5B4}/DZ8 
+------------------------------------------------------------------
+        Bra
+        \x{d55c}\x{ad6d}\x{c5b4}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{ed}
+Need char = \x{b4}
+    \x{D55c}\x{ad6d}\x{C5B4} 
+ 0: \x{d55c}\x{ad6d}\x{c5b4}
+
+/\x{65e5}\x{672c}\x{8a9e}/DZ8
+------------------------------------------------------------------
+        Bra
+        \x{65e5}\x{672c}\x{8a9e}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{e6}
+Need char = \x{9e}
+    \x{65e5}\x{672c}\x{8a9e}
+ 0: \x{65e5}\x{672c}\x{8a9e}
+
+/\x{80}/DZ8
+------------------------------------------------------------------
+        Bra
+        \x{80}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{c2}
+Need char = \x{80}
+
+/\x{084}/DZ8
+------------------------------------------------------------------
+        Bra
+        \x{84}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{c2}
+Need char = \x{84}
+
+/\x{104}/DZ8
+------------------------------------------------------------------
+        Bra
+        \x{104}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{c4}
+Need char = \x{84}
+
+/\x{861}/DZ8
+------------------------------------------------------------------
+        Bra
+        \x{861}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{e0}
+Need char = \x{a1}
+
+/\x{212ab}/DZ8
+------------------------------------------------------------------
+        Bra
+        \x{212ab}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{f0}
+Need char = \x{ab}
+
+/-- This one is here not because it's different to Perl, but because the way
+the captured single-byte is displayed. (In Perl it becomes a character, and you
+can't tell the difference.) --/
+    
+/X(\C)(.*)/8
+    X\x{1234}
+ 0: X\x{1234}
+ 1: \x{e1}
+ 2: \x{88}\x{b4}
+    X\nabc 
+ 0: X\x{0a}abc
+ 1: \x{0a}
+ 2: abc
+
+/-- This one is here because Perl gives out a grumbly error message (quite 
+correctly, but that messes up comparisons). --/
+    
+/a\Cb/8
+    *** Failers 
+No match
+    a\x{100}b 
+No match
+    
+/[^ab\xC0-\xF0]/8SDZ
+------------------------------------------------------------------
+        Bra
+        [\x00-`c-\xbf\xf1-\xff] (neg)
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: \x00 \x01 \x02 \x03 \x04 \x05 \x06 \x07 \x08 \x09 \x0a 
+  \x0b \x0c \x0d \x0e \x0f \x10 \x11 \x12 \x13 \x14 \x15 \x16 \x17 \x18 \x19 
+  \x1a \x1b \x1c \x1d \x1e \x1f \x20 ! " # $ % & ' ( ) * + , - . / 0 1 2 3 4 
+  5 6 7 8 9 : ; < = > ? @ A B C D E F G H I J K L M N O P Q R S T U V W X Y 
+  Z [ \ ] ^ _ ` c d e f g h i j k l m n o p q r s t u v w x y z { | } ~ \x7f 
+  \xc2 \xc3 \xc4 \xc5 \xc6 \xc7 \xc8 \xc9 \xca \xcb \xcc \xcd \xce \xcf \xd0 
+  \xd1 \xd2 \xd3 \xd4 \xd5 \xd6 \xd7 \xd8 \xd9 \xda \xdb \xdc \xdd \xde \xdf 
+  \xe0 \xe1 \xe2 \xe3 \xe4 \xe5 \xe6 \xe7 \xe8 \xe9 \xea \xeb \xec \xed \xee 
+  \xef \xf0 \xf1 \xf2 \xf3 \xf4 \xf5 \xf6 \xf7 \xf8 \xf9 \xfa \xfb \xfc \xfd 
+  \xfe \xff 
+    \x{f1}
+ 0: \x{f1}
+    \x{bf}
+ 0: \x{bf}
+    \x{100}
+ 0: \x{100}
+    \x{1000}   
+ 0: \x{1000}
+    *** Failers
+ 0: *
+    \x{c0} 
+No match
+    \x{f0} 
+No match
+
+/Ā{3,4}/8SDZ
+------------------------------------------------------------------
+        Bra
+        \x{100}{3}
+        \x{100}?
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{c4}
+Need char = \x{80}
+Subject length lower bound = 3
+No set of starting bytes
+  \x{100}\x{100}\x{100}\x{100\x{100}
+ 0: \x{100}\x{100}\x{100}
+
+/(\x{100}+|x)/8SDZ
+------------------------------------------------------------------
+        Bra
+        CBra 1
+        \x{100}+
+        Alt
+        x
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 1
+Options: utf
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: x \xc4 
+
+/(\x{100}*a|x)/8SDZ
+------------------------------------------------------------------
+        Bra
+        CBra 1
+        \x{100}*+
+        a
+        Alt
+        x
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 1
+Options: utf
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: a x \xc4 
+
+/(\x{100}{0,2}a|x)/8SDZ
+------------------------------------------------------------------
+        Bra
+        CBra 1
+        \x{100}{0,2}
+        a
+        Alt
+        x
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 1
+Options: utf
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: a x \xc4 
+
+/(\x{100}{1,2}a|x)/8SDZ
+------------------------------------------------------------------
+        Bra
+        CBra 1
+        \x{100}
+        \x{100}{0,1}
+        a
+        Alt
+        x
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 1
+Options: utf
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: x \xc4 
+
+/\x{100}/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{100}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{c4}
+Need char = \x{80}
+
+/a\x{100}\x{101}*/8DZ
+------------------------------------------------------------------
+        Bra
+        a\x{100}
+        \x{101}*
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = 'a'
+Need char = \x{80}
+
+/a\x{100}\x{101}+/8DZ
+------------------------------------------------------------------
+        Bra
+        a\x{100}
+        \x{101}+
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = 'a'
+Need char = \x{81}
+
+/[^\x{c4}]/DZ
+------------------------------------------------------------------
+        Bra
+        [^\x{c4}]
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+
+/[\x{100}]/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{100}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{c4}
+Need char = \x{80}
+    \x{100}
+ 0: \x{100}
+    Z\x{100}
+ 0: \x{100}
+    \x{100}Z
+ 0: \x{100}
+    *** Failers 
+No match
+
+/[\xff]/DZ8
+------------------------------------------------------------------
+        Bra
+        \x{ff}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{c3}
+Need char = \x{bf}
+    >\x{ff}<
+ 0: \x{ff}
+
+/[^\xff]/8DZ
+------------------------------------------------------------------
+        Bra
+        [^\x{ff}]
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+No first char
+No need char
+
+/\x{100}abc(xyz(?1))/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{100}abc
+        CBra 1
+        xyz
+        Recurse
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 1
+Options: utf
+First char = \x{c4}
+Need char = 'z'
+
+/a\x{1234}b/P8
+    a\x{1234}b
+ 0: a\x{1234}b
+
+/\777/8I
+Capturing subpattern count = 0
+Options: utf
+First char = \x{c7}
+Need char = \x{bf}
+  \x{1ff}
+ 0: \x{1ff}
+  \777 
+ 0: \x{1ff}
+  
+/\x{100}+\x{200}/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{100}++
+        \x{200}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{c4}
+Need char = \x{80}
+
+/\x{100}+X/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{100}++
+        X
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{c4}
+Need char = 'X'
+
+/^[\QĀ\E-\QŐ\E/BZ8
+Failed: missing terminating ] for character class at offset 15
+
+/-- This tests the stricter UTF-8 check according to RFC 3629. --/ 
+    
+/X/8
+    \x{d800}
+Error -10 (bad UTF-8 string) offset=0 reason=14
+    \x{d800}\?
+No match
+    \x{da00}
+Error -10 (bad UTF-8 string) offset=0 reason=14
+    \x{da00}\?
+No match
+    \x{dfff}
+Error -10 (bad UTF-8 string) offset=0 reason=14
+    \x{dfff}\?
+No match
+    \x{110000}    
+Error -10 (bad UTF-8 string) offset=0 reason=13
+    \x{110000}\?    
+No match
+    \x{2000000} 
+Error -10 (bad UTF-8 string) offset=0 reason=11
+    \x{2000000}\? 
+No match
+    \x{7fffffff} 
+Error -10 (bad UTF-8 string) offset=0 reason=12
+    \x{7fffffff}\? 
+No match
+
+/(*UTF8)\x{1234}/
+  abcd\x{1234}pqr
+ 0: \x{1234}
+
+/(*CRLF)(*UTF)(*BSR_UNICODE)a\Rb/I
+Capturing subpattern count = 0
+Options: bsr_unicode utf
+Forced newline sequence: CRLF
+First char = 'a'
+Need char = 'b'
+
+/\h/SI8
+Capturing subpattern count = 0
+Options: utf
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: \x09 \x20 \xc2 \xe1 \xe2 \xe3 
+    ABC\x{09}
+ 0: \x{09}
+    ABC\x{20}
+ 0:  
+    ABC\x{a0}
+ 0: \x{a0}
+    ABC\x{1680}
+ 0: \x{1680}
+    ABC\x{180e}
+ 0: \x{180e}
+    ABC\x{2000}
+ 0: \x{2000}
+    ABC\x{202f} 
+ 0: \x{202f}
+    ABC\x{205f} 
+ 0: \x{205f}
+    ABC\x{3000} 
+ 0: \x{3000}
+
+/\v/SI8
+Capturing subpattern count = 0
+Options: utf
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: \x0a \x0b \x0c \x0d \xc2 \xe2 
+    ABC\x{0a}
+ 0: \x{0a}
+    ABC\x{0b}
+ 0: \x{0b}
+    ABC\x{0c}
+ 0: \x{0c}
+    ABC\x{0d}
+ 0: \x{0d}
+    ABC\x{85}
+ 0: \x{85}
+    ABC\x{2028}
+ 0: \x{2028}
+
+/\h*A/SI8
+Capturing subpattern count = 0
+Options: utf
+No first char
+Need char = 'A'
+Subject length lower bound = 1
+Starting byte set: \x09 \x20 A \xc2 \xe1 \xe2 \xe3 
+    CDBABC
+ 0: A
+    
+/\v+A/SI8
+Capturing subpattern count = 0
+Options: utf
+No first char
+Need char = 'A'
+Subject length lower bound = 2
+Starting byte set: \x0a \x0b \x0c \x0d \xc2 \xe2 
+
+/\s?xxx\s/8SI
+Capturing subpattern count = 0
+Options: utf
+No first char
+Need char = 'x'
+Subject length lower bound = 4
+Starting byte set: \x09 \x0a \x0c \x0d \x20 x 
+
+/\sxxx\s/I8ST1
+Capturing subpattern count = 0
+Options: utf
+No first char
+Need char = 'x'
+Subject length lower bound = 5
+Starting byte set: \x09 \x0a \x0c \x0d \x20 \xc2 
+    AB\x{85}xxx\x{a0}XYZ
+ 0: \x{85}xxx\x{a0}
+    AB\x{a0}xxx\x{85}XYZ
+ 0: \x{a0}xxx\x{85}
+
+/\S \S/I8ST1
+Capturing subpattern count = 0
+Options: utf
+No first char
+Need char = ' '
+Subject length lower bound = 3
+Starting byte set: \x00 \x01 \x02 \x03 \x04 \x05 \x06 \x07 \x08 \x0b \x0e 
+  \x0f \x10 \x11 \x12 \x13 \x14 \x15 \x16 \x17 \x18 \x19 \x1a \x1b \x1c \x1d 
+  \x1e \x1f ! " # $ % & ' ( ) * + , - . / 0 1 2 3 4 5 6 7 8 9 : ; < = > ? @ 
+  A B C D E F G H I J K L M N O P Q R S T U V W X Y Z [ \ ] ^ _ ` a b c d e 
+  f g h i j k l m n o p q r s t u v w x y z { | } ~ \x7f \xc0 \xc1 \xc2 \xc3 
+  \xc4 \xc5 \xc6 \xc7 \xc8 \xc9 \xca \xcb \xcc \xcd \xce \xcf \xd0 \xd1 \xd2 
+  \xd3 \xd4 \xd5 \xd6 \xd7 \xd8 \xd9 \xda \xdb \xdc \xdd \xde \xdf \xe0 \xe1 
+  \xe2 \xe3 \xe4 \xe5 \xe6 \xe7 \xe8 \xe9 \xea \xeb \xec \xed \xee \xef \xf0 
+  \xf1 \xf2 \xf3 \xf4 \xf5 \xf6 \xf7 \xf8 \xf9 \xfa \xfb \xfc \xfd \xfe \xff 
+    \x{a2} \x{84} 
+ 0: \x{a2} \x{84}
+    A Z 
+ 0: A Z
+
+/a+/8
+    a\x{123}aa\>1
+ 0: aa
+    a\x{123}aa\>2
+Error -11 (bad UTF-8 offset)
+    a\x{123}aa\>3
+ 0: aa
+    a\x{123}aa\>4
+ 0: a
+    a\x{123}aa\>5
+No match
+    a\x{123}aa\>6
+Error -24 (bad offset value)
+
+/\x{1234}+/iS8I
+Capturing subpattern count = 0
+Options: caseless utf
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: \xe1 
+
+/\x{1234}+?/iS8I
+Capturing subpattern count = 0
+Options: caseless utf
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: \xe1 
+
+/\x{1234}++/iS8I
+Capturing subpattern count = 0
+Options: caseless utf
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: \xe1 
+
+/\x{1234}{2}/iS8I
+Capturing subpattern count = 0
+Options: caseless utf
+No first char
+No need char
+Subject length lower bound = 2
+Starting byte set: \xe1 
+
+/[^\x{c4}]/8DZ
+------------------------------------------------------------------
+        Bra
+        [^\x{c4}]
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+No first char
+No need char
+
+/X+\x{200}/8DZ
+------------------------------------------------------------------
+        Bra
+        X++
+        \x{200}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = 'X'
+Need char = \x{80}
+
+/\R/SI8
+Capturing subpattern count = 0
+Options: utf
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: \x0a \x0b \x0c \x0d \xc2 \xe2 
+
+/\777/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{1ff}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{c7}
+Need char = \x{bf}
+
+/\w+\x{C4}/8BZ
+------------------------------------------------------------------
+        Bra
+        \w++
+        \x{c4}
+        Ket
+        End
+------------------------------------------------------------------
+    a\x{C4}\x{C4}
+ 0: a\x{c4}
+
+/\w+\x{C4}/8BZT1
+------------------------------------------------------------------
+        Bra
+        \w+
+        \x{c4}
+        Ket
+        End
+------------------------------------------------------------------
+    a\x{C4}\x{C4}
+ 0: a\x{c4}\x{c4}
+    
+/\W+\x{C4}/8BZ
+------------------------------------------------------------------
+        Bra
+        \W+
+        \x{c4}
+        Ket
+        End
+------------------------------------------------------------------
+    !\x{C4}
+ 0: !\x{c4}
+ 
+/\W+\x{C4}/8BZT1
+------------------------------------------------------------------
+        Bra
+        \W++
+        \x{c4}
+        Ket
+        End
+------------------------------------------------------------------
+    !\x{C4}
+ 0: !\x{c4}
+
+/\W+\x{A1}/8BZ
+------------------------------------------------------------------
+        Bra
+        \W+
+        \x{a1}
+        Ket
+        End
+------------------------------------------------------------------
+    !\x{A1}
+ 0: !\x{a1}
+ 
+/\W+\x{A1}/8BZT1
+------------------------------------------------------------------
+        Bra
+        \W+
+        \x{a1}
+        Ket
+        End
+------------------------------------------------------------------
+    !\x{A1}
+ 0: !\x{a1}
+
+/X\s+\x{A0}/8BZ
+------------------------------------------------------------------
+        Bra
+        X
+        \s++
+        \x{a0}
+        Ket
+        End
+------------------------------------------------------------------
+    X\x20\x{A0}\x{A0}
+ 0: X \x{a0}
+
+/X\s+\x{A0}/8BZT1
+------------------------------------------------------------------
+        Bra
+        X
+        \s+
+        \x{a0}
+        Ket
+        End
+------------------------------------------------------------------
+    X\x20\x{A0}\x{A0}
+ 0: X \x{a0}\x{a0}
+
+/\S+\x{A0}/8BZ
+------------------------------------------------------------------
+        Bra
+        \S+
+        \x{a0}
+        Ket
+        End
+------------------------------------------------------------------
+    X\x{A0}\x{A0}
+ 0: X\x{a0}\x{a0}
+
+/\S+\x{A0}/8BZT1
+------------------------------------------------------------------
+        Bra
+        \S++
+        \x{a0}
+        Ket
+        End
+------------------------------------------------------------------
+    X\x{A0}\x{A0}
+ 0: X\x{a0}
+
+/\x{a0}+\s!/8BZ
+------------------------------------------------------------------
+        Bra
+        \x{a0}++
+        \s
+        !
+        Ket
+        End
+------------------------------------------------------------------
+    \x{a0}\x20!
+ 0: \x{a0} !
+
+/\x{a0}+\s!/8BZT1
+------------------------------------------------------------------
+        Bra
+        \x{a0}+
+        \s
+        !
+        Ket
+        End
+------------------------------------------------------------------
+    \x{a0}\x20!
+ 0: \x{a0} !
+
+/A/8
+  \x{ff000041}
+** Character \x{ff000041} is greater than 0x7fffffff and so cannot be converted to UTF-8
+  \x{7f000041} 
+Error -10 (bad UTF-8 string) offset=0 reason=12
+
+/(*UTF8)abc/9
+Failed: setting UTF is disabled by the application at offset 0
+
+/abc/89
+Failed: setting UTF is disabled by the application at offset 0
+
+/-- End of testinput15 --/
diff --git a/pcre/testdata/testoutput16 b/pcre/testdata/testoutput16
new file mode 100644
index 00000000000..1d5f31d929a
--- /dev/null
+++ b/pcre/testdata/testoutput16
@@ -0,0 +1,121 @@
+/-- This set of tests is run only with the 8-bit library when Unicode property 
+    support is available. It starts with tests of the POSIX interface, because
+    that is supported only with the 8-bit library. --/
+
+/\w/P
+    +++\x{c2}
+No match: POSIX code 17: match failed
+
+/\w/WP
+    +++\x{c2}
+ 0: \xc2
+    
+/A\x{391}\x{10427}\x{ff3a}\x{1fb0}/8iDZ
+------------------------------------------------------------------
+        Bra
+     /i A\x{391}\x{10427}\x{ff3a}\x{1fb0}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: caseless utf
+First char = 'A' (caseless)
+No need char
+
+/A\x{391}\x{10427}\x{ff3a}\x{1fb0}/8DZ
+------------------------------------------------------------------
+        Bra
+        A\x{391}\x{10427}\x{ff3a}\x{1fb0}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = 'A'
+Need char = \x{b0}
+
+/AB\x{1fb0}/8DZ
+------------------------------------------------------------------
+        Bra
+        AB\x{1fb0}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = 'A'
+Need char = \x{b0}
+
+/AB\x{1fb0}/8DZi
+------------------------------------------------------------------
+        Bra
+     /i AB\x{1fb0}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: caseless utf
+First char = 'A' (caseless)
+Need char = 'B' (caseless)
+
+/\x{401}\x{420}\x{421}\x{422}\x{423}\x{424}\x{425}\x{426}\x{427}\x{428}\x{429}\x{42a}\x{42b}\x{42c}\x{42d}\x{42e}\x{42f}/8iSI
+Capturing subpattern count = 0
+Options: caseless utf
+No first char
+No need char
+Subject length lower bound = 17
+Starting byte set: \xd0 \xd1 
+    \x{401}\x{420}\x{421}\x{422}\x{423}\x{424}\x{425}\x{426}\x{427}\x{428}\x{429}\x{42a}\x{42b}\x{42c}\x{42d}\x{42e}\x{42f}
+ 0: \x{401}\x{420}\x{421}\x{422}\x{423}\x{424}\x{425}\x{426}\x{427}\x{428}\x{429}\x{42a}\x{42b}\x{42c}\x{42d}\x{42e}\x{42f}
+    \x{451}\x{440}\x{441}\x{442}\x{443}\x{444}\x{445}\x{446}\x{447}\x{448}\x{449}\x{44a}\x{44b}\x{44c}\x{44d}\x{44e}\x{44f}
+ 0: \x{451}\x{440}\x{441}\x{442}\x{443}\x{444}\x{445}\x{446}\x{447}\x{448}\x{449}\x{44a}\x{44b}\x{44c}\x{44d}\x{44e}\x{44f}
+
+/[ⱥ]/8iBZ
+------------------------------------------------------------------
+        Bra
+     /i \x{2c65}
+        Ket
+        End
+------------------------------------------------------------------
+
+/[^ⱥ]/8iBZ
+------------------------------------------------------------------
+        Bra
+     /i [^\x{2c65}]
+        Ket
+        End
+------------------------------------------------------------------
+
+/\h/SI
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: \x09 \x20 \xa0 
+
+/\v/SI
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: \x0a \x0b \x0c \x0d \x85 
+
+/\R/SI
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: \x0a \x0b \x0c \x0d \x85 
+
+/[[:blank:]]/WBZ
+------------------------------------------------------------------
+        Bra
+        [\x09 \xa0]
+        Ket
+        End
+------------------------------------------------------------------
+
+/-- End of testinput16 --/
diff --git a/pcre/testdata/testoutput17 b/pcre/testdata/testoutput17
new file mode 100644
index 00000000000..ef82dc9d0d8
--- /dev/null
+++ b/pcre/testdata/testoutput17
@@ -0,0 +1,505 @@
+/-- This set of tests is for the 16- and 32-bit library's basic (non-UTF-16 
+    or -32) features that are not compatible with the 8-bit library, or which 
+    give different output in 16- or 32-bit mode. --/
+
+/a\Cb/
+    aXb
+ 0: aXb
+    a\nb
+ 0: a\x0ab
+  
+/[^\x{c4}]/DZ
+------------------------------------------------------------------
+        Bra
+        [^\x{c4}]
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+  
+/\x{100}/I
+Capturing subpattern count = 0
+No options
+First char = \x{100}
+No need char
+
+/  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*                          # optional leading comment
+(?:    (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+)                    # initial word
+(?:  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+)  )* # further okay, if led by a period
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  @  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*    (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                           # initial subdomain
+(?:                                  #
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.                        # if led by a period...
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                     #   ...further okay
+)*
+# address
+|                     #  or
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+)             # one word, optionally followed by....
+(?:
+[^()<>@,;:".\\\[\]\x80-\xff\000-\010\012-\037]  |  # atom and space parts, or...
+\(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)       |  # comments, or...
+
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+# quoted strings
+)*
+<  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*                     # leading <
+(?:  @  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*    (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                           # initial subdomain
+(?:                                  #
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.                        # if led by a period...
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                     #   ...further okay
+)*
+
+(?:  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  ,  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  @  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*    (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                           # initial subdomain
+(?:                                  #
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.                        # if led by a period...
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                     #   ...further okay
+)*
+)* # further okay, if led by comma
+:                                # closing colon
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  )? #       optional route
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+)                    # initial word
+(?:  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+)  )* # further okay, if led by a period
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  @  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*    (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                           # initial subdomain
+(?:                                  #
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.                        # if led by a period...
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                     #   ...further okay
+)*
+#       address spec
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  > #                  trailing >
+# name and address
+)  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*                       # optional trailing comment
+/xSI
+Capturing subpattern count = 0
+Contains explicit CR or LF match
+Options: extended
+No first char
+No need char
+Subject length lower bound = 3
+Starting byte set: \x09 \x20 ! " # $ % & ' ( * + - / 0 1 2 3 4 5 6 7 8 
+  9 = ? A B C D E F G H I J K L M N O P Q R S T U V W X Y Z ^ _ ` a b c d e 
+  f g h i j k l m n o p q r s t u v w x y z { | } ~ \x7f \xff 
+
+/[\h]/BZ
+------------------------------------------------------------------
+        Bra
+        [\x09 \xa0\x{1680}\x{180e}\x{2000}-\x{200a}\x{202f}\x{205f}\x{3000}]
+        Ket
+        End
+------------------------------------------------------------------
+    >\x09<
+ 0: \x09
+
+/[\h]+/BZ
+------------------------------------------------------------------
+        Bra
+        [\x09 \xa0\x{1680}\x{180e}\x{2000}-\x{200a}\x{202f}\x{205f}\x{3000}]+
+        Ket
+        End
+------------------------------------------------------------------
+    >\x09\x20\xa0<
+ 0: \x09 \xa0
+
+/[\v]/BZ
+------------------------------------------------------------------
+        Bra
+        [\x0a-\x0d\x85\x{2028}-\x{2029}]
+        Ket
+        End
+------------------------------------------------------------------
+
+/[^\h]/BZ
+------------------------------------------------------------------
+        Bra
+        [^\x09 \xa0\x{1680}\x{180e}\x{2000}-\x{200a}\x{202f}\x{205f}\x{3000}]
+        Ket
+        End
+------------------------------------------------------------------
+
+/\h+/SI
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: \x09 \x20 \xa0 \xff 
+    \x{1681}\x{200b}\x{1680}\x{2000}\x{202f}\x{3000}
+ 0: \x{1680}\x{2000}\x{202f}\x{3000}
+    \x{3001}\x{2fff}\x{200a}\xa0\x{2000}
+ 0: \x{200a}\xa0\x{2000}
+
+/[\h\x{dc00}]+/BZSI
+------------------------------------------------------------------
+        Bra
+        [\x09 \xa0\x{1680}\x{180e}\x{2000}-\x{200a}\x{202f}\x{205f}\x{3000}\x{dc00}]+
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+Subject length lower bound = 1
+No set of starting bytes
+    \x{1681}\x{200b}\x{1680}\x{2000}\x{202f}\x{3000}
+ 0: \x{1680}\x{2000}\x{202f}\x{3000}
+    \x{3001}\x{2fff}\x{200a}\xa0\x{2000}
+ 0: \x{200a}\xa0\x{2000}
+
+/\H+/SI
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+Subject length lower bound = 1
+No set of starting bytes
+    \x{1680}\x{180e}\x{167f}\x{1681}\x{180d}\x{180f}
+ 0: \x{167f}\x{1681}\x{180d}\x{180f}
+    \x{2000}\x{200a}\x{1fff}\x{200b}
+ 0: \x{1fff}\x{200b}
+    \x{202f}\x{205f}\x{202e}\x{2030}\x{205e}\x{2060}
+ 0: \x{202e}\x{2030}\x{205e}\x{2060}
+    \xa0\x{3000}\x9f\xa1\x{2fff}\x{3001}
+ 0: \x9f\xa1\x{2fff}\x{3001}
+
+/[\H\x{d800}]+/
+    \x{1680}\x{180e}\x{167f}\x{1681}\x{180d}\x{180f}
+ 0: \x{167f}\x{1681}\x{180d}\x{180f}
+    \x{2000}\x{200a}\x{1fff}\x{200b}
+ 0: \x{1fff}\x{200b}
+    \x{202f}\x{205f}\x{202e}\x{2030}\x{205e}\x{2060}
+ 0: \x{202e}\x{2030}\x{205e}\x{2060}
+    \xa0\x{3000}\x9f\xa1\x{2fff}\x{3001}
+ 0: \x9f\xa1\x{2fff}\x{3001}
+
+/\v+/SI
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: \x0a \x0b \x0c \x0d \x85 \xff 
+    \x{2027}\x{2030}\x{2028}\x{2029}
+ 0: \x{2028}\x{2029}
+    \x09\x0e\x84\x86\x85\x0a\x0b\x0c\x0d
+ 0: \x85\x0a\x0b\x0c\x0d
+
+/[\v\x{dc00}]+/BZSI
+------------------------------------------------------------------
+        Bra
+        [\x0a-\x0d\x85\x{2028}-\x{2029}\x{dc00}]+
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+Subject length lower bound = 1
+No set of starting bytes
+    \x{2027}\x{2030}\x{2028}\x{2029}
+ 0: \x{2028}\x{2029}
+    \x09\x0e\x84\x86\x85\x0a\x0b\x0c\x0d
+ 0: \x85\x0a\x0b\x0c\x0d
+
+/\V+/SI
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+Subject length lower bound = 1
+No set of starting bytes
+    \x{2028}\x{2029}\x{2027}\x{2030}
+ 0: \x{2027}\x{2030}
+    \x85\x0a\x0b\x0c\x0d\x09\x0e\x84\x86
+ 0: \x09\x0e\x84\x86
+
+/[\V\x{d800}]+/
+    \x{2028}\x{2029}\x{2027}\x{2030}
+ 0: \x{2027}\x{2030}
+    \x85\x0a\x0b\x0c\x0d\x09\x0e\x84\x86
+ 0: \x09\x0e\x84\x86
+
+/\R+/SI
+Capturing subpattern count = 0
+Options: bsr_unicode
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: \x0a \x0b \x0c \x0d \x85 \xff 
+    \x{2027}\x{2030}\x{2028}\x{2029}
+ 0: \x{2028}\x{2029}
+    \x09\x0e\x84\x86\x85\x0a\x0b\x0c\x0d
+ 0: \x85\x0a\x0b\x0c\x0d
+
+/\x{d800}\x{d7ff}\x{dc00}\x{dc00}\x{dcff}\x{dd00}/I
+Capturing subpattern count = 0
+No options
+First char = \x{d800}
+Need char = \x{dd00}
+    \x{d800}\x{d7ff}\x{dc00}\x{dc00}\x{dcff}\x{dd00}
+ 0: \x{d800}\x{d7ff}\x{dc00}\x{dc00}\x{dcff}\x{dd00}
+
+/[^\x{80}][^\x{ff}][^\x{100}][^\x{1000}][^\x{ffff}]/BZ
+------------------------------------------------------------------
+        Bra
+        [^\x80]
+        [^\x{ff}]
+        [^\x{100}]
+        [^\x{1000}]
+        [^\x{ffff}]
+        Ket
+        End
+------------------------------------------------------------------
+
+/[^\x{80}][^\x{ff}][^\x{100}][^\x{1000}][^\x{ffff}]/BZi
+------------------------------------------------------------------
+        Bra
+     /i [^\x80]
+     /i [^\x{ff}]
+     /i [^\x{100}]
+     /i [^\x{1000}]
+     /i [^\x{ffff}]
+        Ket
+        End
+------------------------------------------------------------------
+
+/[^\x{100}]*[^\x{1000}]+[^\x{ffff}]??[^\x{8000}]{4,}[^\x{7fff}]{2,9}?[^\x{100}]{5,6}+/BZ
+------------------------------------------------------------------
+        Bra
+        [^\x{100}]*
+        [^\x{1000}]+
+        [^\x{ffff}]??
+        [^\x{8000}]{4}
+        [^\x{8000}]*
+        [^\x{7fff}]{2}
+        [^\x{7fff}]{0,7}?
+        [^\x{100}]{5}
+        [^\x{100}]?+
+        Ket
+        End
+------------------------------------------------------------------
+
+/[^\x{100}]*[^\x{1000}]+[^\x{ffff}]??[^\x{8000}]{4,}[^\x{7fff}]{2,9}?[^\x{100}]{5,6}+/BZi
+------------------------------------------------------------------
+        Bra
+     /i [^\x{100}]*
+     /i [^\x{1000}]+
+     /i [^\x{ffff}]??
+     /i [^\x{8000}]{4}
+     /i [^\x{8000}]*
+     /i [^\x{7fff}]{2}
+     /i [^\x{7fff}]{0,7}?
+        Once
+     /i [^\x{100}]{5}
+     /i [^\x{100}]?
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+
+/(*:0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF)XX/K
+    XX
+ 0: XX
+MK: 0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF
+     
+/(*:0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDE)XX/K
+    XX
+ 0: XX
+MK: 0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDE
+
+/\u0100/BZ
+------------------------------------------------------------------
+        Bra
+        \x{100}
+        Ket
+        End
+------------------------------------------------------------------
+
+/[\u0100-\u0200]/BZ
+------------------------------------------------------------------
+        Bra
+        [\x{100}-\x{200}]
+        Ket
+        End
+------------------------------------------------------------------
+
+/\ud800/BZ
+------------------------------------------------------------------
+        Bra
+        \x{d800}
+        Ket
+        End
+------------------------------------------------------------------
+
+/^\x{ffff}+/i
+    \x{ffff}
+ 0: \x{ffff}
+
+/^\x{ffff}?/i
+    \x{ffff}
+ 0: \x{ffff}
+
+/^\x{ffff}*/i
+    \x{ffff}
+ 0: \x{ffff}
+
+/^\x{ffff}{3}/i
+    \x{ffff}\x{ffff}\x{ffff}
+ 0: \x{ffff}\x{ffff}\x{ffff}
+
+/^\x{ffff}{0,3}/i
+    \x{ffff}
+ 0: \x{ffff}
+
+/-- End of testinput17 --/
diff --git a/pcre/testdata/testoutput18-16 b/pcre/testdata/testoutput18-16
new file mode 100644
index 00000000000..3414a661808
--- /dev/null
+++ b/pcre/testdata/testoutput18-16
@@ -0,0 +1,1024 @@
+/-- This set of tests is for UTF-16 and UTF-32 support, and is relevant only to the
+    16- and 32-bit library. --/
+
+/xxx/8?DZSS
+**Failed: invalid UTF-8 string cannot be converted to UTF-16
+
+/abc/8
+    ]
+**Failed: invalid UTF-8 string cannot be used as input in UTF mode
+
+/X(\C{3})/8
+    X\x{11234}Y
+ 0: X\x{11234}Y
+ 1: \x{11234}Y
+    X\x{11234}YZ
+ 0: X\x{11234}Y
+ 1: \x{11234}Y
+
+/X(\C{4})/8
+    X\x{11234}YZ
+ 0: X\x{11234}YZ
+ 1: \x{11234}YZ
+    X\x{11234}YZW
+ 0: X\x{11234}YZ
+ 1: \x{11234}YZ
+
+/X\C*/8
+    XYZabcdce
+ 0: XYZabcdce
+
+/X\C*?/8
+    XYZabcde
+ 0: X
+
+/X\C{3,5}/8
+    Xabcdefg
+ 0: Xabcde
+    X\x{11234}Y
+ 0: X\x{11234}Y
+    X\x{11234}YZ
+ 0: X\x{11234}YZ
+    X\x{11234}\x{512}
+ 0: X\x{11234}\x{512}
+    X\x{11234}\x{512}YZ
+ 0: X\x{11234}\x{512}YZ
+    X\x{11234}\x{512}\x{11234}Z
+ 0: X\x{11234}\x{512}\x{11234}
+
+/X\C{3,5}?/8
+    Xabcdefg
+ 0: Xabc
+    X\x{11234}Y
+ 0: X\x{11234}Y
+    X\x{11234}YZ
+ 0: X\x{11234}Y
+    X\x{11234}\x{512}YZ
+ 0: X\x{11234}\x{512}
+    *** Failers
+No match
+    X\x{11234}
+No match
+
+/a\Cb/8
+    aXb
+ 0: aXb
+    a\nb
+ 0: a\x{0a}b
+
+/a\C\Cb/8
+    a\x{12257}b
+ 0: a\x{12257}b
+    a\x{12257}\x{11234}b
+No match
+    ** Failers
+No match
+    a\x{100}b
+No match
+
+/ab\Cde/8
+    abXde
+ 0: abXde
+
+/-- Check maximum character size --/
+
+/\x{ffff}/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{ffff}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{ffff}
+No need char
+
+/\x{10000}/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{10000}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{d800}
+Need char = \x{dc00}
+
+/\x{100}/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{100}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{100}
+No need char
+
+/\x{1000}/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{1000}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{1000}
+No need char
+
+/\x{10000}/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{10000}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{d800}
+Need char = \x{dc00}
+
+/\x{100000}/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{100000}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{dbc0}
+Need char = \x{dc00}
+
+/\x{10ffff}/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{10ffff}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{dbff}
+Need char = \x{dfff}
+
+/[\x{ff}]/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{ff}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{ff}
+No need char
+
+/[\x{100}]/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{100}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{100}
+No need char
+
+/\x80/8DZ
+------------------------------------------------------------------
+        Bra
+        \x80
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{80}
+No need char
+
+/\xff/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{ff}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{ff}
+No need char
+
+/\x{D55c}\x{ad6d}\x{C5B4}/DZ8
+------------------------------------------------------------------
+        Bra
+        \x{d55c}\x{ad6d}\x{c5b4}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{d55c}
+Need char = \x{c5b4}
+    \x{D55c}\x{ad6d}\x{C5B4}
+ 0: \x{d55c}\x{ad6d}\x{c5b4}
+
+/\x{65e5}\x{672c}\x{8a9e}/DZ8
+------------------------------------------------------------------
+        Bra
+        \x{65e5}\x{672c}\x{8a9e}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{65e5}
+Need char = \x{8a9e}
+    \x{65e5}\x{672c}\x{8a9e}
+ 0: \x{65e5}\x{672c}\x{8a9e}
+
+/\x{80}/DZ8
+------------------------------------------------------------------
+        Bra
+        \x80
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{80}
+No need char
+
+/\x{084}/DZ8
+------------------------------------------------------------------
+        Bra
+        \x{84}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{84}
+No need char
+
+/\x{104}/DZ8
+------------------------------------------------------------------
+        Bra
+        \x{104}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{104}
+No need char
+
+/\x{861}/DZ8
+------------------------------------------------------------------
+        Bra
+        \x{861}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{861}
+No need char
+
+/\x{212ab}/DZ8
+------------------------------------------------------------------
+        Bra
+        \x{212ab}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{d844}
+Need char = \x{deab}
+
+/-- This one is here not because it's different to Perl, but because the way
+the captured single-byte is displayed. (In Perl it becomes a character, and you
+can't tell the difference.) --/
+
+/X(\C)(.*)/8
+    X\x{1234}
+ 0: X\x{1234}
+ 1: \x{1234}
+ 2: 
+    X\nabc
+ 0: X\x{0a}abc
+ 1: \x{0a}
+ 2: abc
+
+/-- This one is here because Perl gives out a grumbly error message (quite
+correctly, but that messes up comparisons). --/
+
+/a\Cb/8
+    *** Failers
+No match
+    a\x{100}b
+ 0: a\x{100}b
+
+/[^ab\xC0-\xF0]/8SDZ
+------------------------------------------------------------------
+        Bra
+        [\x00-`c-\xbf\xf1-\xff] (neg)
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: \x00 \x01 \x02 \x03 \x04 \x05 \x06 \x07 \x08 \x09 \x0a 
+  \x0b \x0c \x0d \x0e \x0f \x10 \x11 \x12 \x13 \x14 \x15 \x16 \x17 \x18 \x19 
+  \x1a \x1b \x1c \x1d \x1e \x1f \x20 ! " # $ % & ' ( ) * + , - . / 0 1 2 3 4 
+  5 6 7 8 9 : ; < = > ? @ A B C D E F G H I J K L M N O P Q R S T U V W X Y 
+  Z [ \ ] ^ _ ` c d e f g h i j k l m n o p q r s t u v w x y z { | } ~ \x7f 
+  \x80 \x81 \x82 \x83 \x84 \x85 \x86 \x87 \x88 \x89 \x8a \x8b \x8c \x8d \x8e 
+  \x8f \x90 \x91 \x92 \x93 \x94 \x95 \x96 \x97 \x98 \x99 \x9a \x9b \x9c \x9d 
+  \x9e \x9f \xa0 \xa1 \xa2 \xa3 \xa4 \xa5 \xa6 \xa7 \xa8 \xa9 \xaa \xab \xac 
+  \xad \xae \xaf \xb0 \xb1 \xb2 \xb3 \xb4 \xb5 \xb6 \xb7 \xb8 \xb9 \xba \xbb 
+  \xbc \xbd \xbe \xbf \xf1 \xf2 \xf3 \xf4 \xf5 \xf6 \xf7 \xf8 \xf9 \xfa \xfb 
+  \xfc \xfd \xfe \xff 
+    \x{f1}
+ 0: \x{f1}
+    \x{bf}
+ 0: \x{bf}
+    \x{100}
+ 0: \x{100}
+    \x{1000}
+ 0: \x{1000}
+    *** Failers
+ 0: *
+    \x{c0}
+No match
+    \x{f0}
+No match
+
+/Ā{3,4}/8SDZ
+------------------------------------------------------------------
+        Bra
+        \x{100}{3}
+        \x{100}?
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{100}
+Need char = \x{100}
+Subject length lower bound = 3
+No set of starting bytes
+  \x{100}\x{100}\x{100}\x{100\x{100}
+ 0: \x{100}\x{100}\x{100}
+
+/(\x{100}+|x)/8SDZ
+------------------------------------------------------------------
+        Bra
+        CBra 1
+        \x{100}+
+        Alt
+        x
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 1
+Options: utf
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: x \xff 
+
+/(\x{100}*a|x)/8SDZ
+------------------------------------------------------------------
+        Bra
+        CBra 1
+        \x{100}*+
+        a
+        Alt
+        x
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 1
+Options: utf
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: a x \xff 
+
+/(\x{100}{0,2}a|x)/8SDZ
+------------------------------------------------------------------
+        Bra
+        CBra 1
+        \x{100}{0,2}
+        a
+        Alt
+        x
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 1
+Options: utf
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: a x \xff 
+
+/(\x{100}{1,2}a|x)/8SDZ
+------------------------------------------------------------------
+        Bra
+        CBra 1
+        \x{100}
+        \x{100}{0,1}
+        a
+        Alt
+        x
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 1
+Options: utf
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: x \xff 
+
+/\x{100}/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{100}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{100}
+No need char
+
+/a\x{100}\x{101}*/8DZ
+------------------------------------------------------------------
+        Bra
+        a\x{100}
+        \x{101}*
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = 'a'
+Need char = \x{100}
+
+/a\x{100}\x{101}+/8DZ
+------------------------------------------------------------------
+        Bra
+        a\x{100}
+        \x{101}+
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = 'a'
+Need char = \x{101}
+
+/[^\x{c4}]/DZ
+------------------------------------------------------------------
+        Bra
+        [^\x{c4}]
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+
+/[\x{100}]/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{100}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{100}
+No need char
+    \x{100}
+ 0: \x{100}
+    Z\x{100}
+ 0: \x{100}
+    \x{100}Z
+ 0: \x{100}
+    *** Failers
+No match
+
+/[\xff]/DZ8
+------------------------------------------------------------------
+        Bra
+        \x{ff}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{ff}
+No need char
+    >\x{ff}<
+ 0: \x{ff}
+
+/[^\xff]/8DZ
+------------------------------------------------------------------
+        Bra
+        [^\x{ff}]
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+No first char
+No need char
+
+/\x{100}abc(xyz(?1))/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{100}abc
+        CBra 1
+        xyz
+        Recurse
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 1
+Options: utf
+First char = \x{100}
+Need char = 'z'
+
+/\777/8I
+Capturing subpattern count = 0
+Options: utf
+First char = \x{1ff}
+No need char
+  \x{1ff}
+ 0: \x{1ff}
+  \777
+ 0: \x{1ff}
+
+/\x{100}+\x{200}/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{100}++
+        \x{200}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{100}
+Need char = \x{200}
+
+/\x{100}+X/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{100}++
+        X
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{100}
+Need char = 'X'
+
+/^[\QĀ\E-\QŐ\E/BZ8
+Failed: missing terminating ] for character class at offset 13
+
+/X/8
+    \x{d800}
+Error -10 (bad UTF-16 string) offset=0 reason=1
+    \x{d800}\?
+No match
+    \x{da00}
+Error -10 (bad UTF-16 string) offset=0 reason=1
+    \x{da00}\?
+No match
+    \x{dc00}
+Error -10 (bad UTF-16 string) offset=0 reason=3
+    \x{dc00}\?
+No match
+    \x{de00}
+Error -10 (bad UTF-16 string) offset=0 reason=3
+    \x{de00}\?
+No match
+    \x{dfff}
+Error -10 (bad UTF-16 string) offset=0 reason=3
+    \x{dfff}\?
+No match
+    \x{110000}
+** Failed: character \x{110000} is greater than 0x10ffff and so cannot be converted to UTF-16
+    \x{d800}\x{1234}
+Error -10 (bad UTF-16 string) offset=1 reason=2
+
+/(*UTF16)\x{11234}/
+  abcd\x{11234}pqr
+ 0: \x{11234}
+
+/(*UTF)\x{11234}/I
+Capturing subpattern count = 0
+Options: utf
+First char = \x{d804}
+Need char = \x{de34}
+  abcd\x{11234}pqr
+ 0: \x{11234}
+
+/(*UTF-32)\x{11234}/
+Failed: (*VERB) not recognized or malformed at offset 5
+
+/(*CRLF)(*UTF16)(*BSR_UNICODE)a\Rb/I
+Capturing subpattern count = 0
+Options: bsr_unicode utf
+Forced newline sequence: CRLF
+First char = 'a'
+Need char = 'b'
+
+/(*CRLF)(*UTF32)(*BSR_UNICODE)a\Rb/I
+Failed: (*VERB) not recognized or malformed at offset 12
+
+/\h/SI8
+Capturing subpattern count = 0
+Options: utf
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: \x09 \x20 \xa0 \xff 
+    ABC\x{09}
+ 0: \x{09}
+    ABC\x{20}
+ 0:  
+    ABC\x{a0}
+ 0: \x{a0}
+    ABC\x{1680}
+ 0: \x{1680}
+    ABC\x{180e}
+ 0: \x{180e}
+    ABC\x{2000}
+ 0: \x{2000}
+    ABC\x{202f}
+ 0: \x{202f}
+    ABC\x{205f}
+ 0: \x{205f}
+    ABC\x{3000}
+ 0: \x{3000}
+
+/\v/SI8
+Capturing subpattern count = 0
+Options: utf
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: \x0a \x0b \x0c \x0d \x85 \xff 
+    ABC\x{0a}
+ 0: \x{0a}
+    ABC\x{0b}
+ 0: \x{0b}
+    ABC\x{0c}
+ 0: \x{0c}
+    ABC\x{0d}
+ 0: \x{0d}
+    ABC\x{85}
+ 0: \x{85}
+    ABC\x{2028}
+ 0: \x{2028}
+
+/\h*A/SI8
+Capturing subpattern count = 0
+Options: utf
+No first char
+Need char = 'A'
+Subject length lower bound = 1
+Starting byte set: \x09 \x20 A \xa0 \xff 
+    CDBABC
+ 0: A
+    \x{2000}ABC 
+ 0: \x{2000}A
+
+/\R*A/SI8
+Capturing subpattern count = 0
+Options: utf
+No first char
+Need char = 'A'
+Subject length lower bound = 1
+Starting byte set: \x0a \x0b \x0c \x0d A \x85 \xff 
+    CDBABC
+ 0: A
+    \x{2028}A  
+ 0: \x{2028}A
+
+/\v+A/SI8
+Capturing subpattern count = 0
+Options: utf
+No first char
+Need char = 'A'
+Subject length lower bound = 2
+Starting byte set: \x0a \x0b \x0c \x0d \x85 \xff 
+
+/\s?xxx\s/8SI
+Capturing subpattern count = 0
+Options: utf
+No first char
+Need char = 'x'
+Subject length lower bound = 4
+Starting byte set: \x09 \x0a \x0c \x0d \x20 x 
+
+/\sxxx\s/I8ST1
+Capturing subpattern count = 0
+Options: utf
+No first char
+Need char = 'x'
+Subject length lower bound = 5
+Starting byte set: \x09 \x0a \x0c \x0d \x20 \x85 \xa0 
+    AB\x{85}xxx\x{a0}XYZ
+ 0: \x{85}xxx\x{a0}
+    AB\x{a0}xxx\x{85}XYZ
+ 0: \x{a0}xxx\x{85}
+
+/\S \S/I8ST1
+Capturing subpattern count = 0
+Options: utf
+No first char
+Need char = ' '
+Subject length lower bound = 3
+Starting byte set: \x00 \x01 \x02 \x03 \x04 \x05 \x06 \x07 \x08 \x0b \x0e 
+  \x0f \x10 \x11 \x12 \x13 \x14 \x15 \x16 \x17 \x18 \x19 \x1a \x1b \x1c \x1d 
+  \x1e \x1f ! " # $ % & ' ( ) * + , - . / 0 1 2 3 4 5 6 7 8 9 : ; < = > ? @ 
+  A B C D E F G H I J K L M N O P Q R S T U V W X Y Z [ \ ] ^ _ ` a b c d e 
+  f g h i j k l m n o p q r s t u v w x y z { | } ~ \x7f \x80 \x81 \x82 \x83 
+  \x84 \x86 \x87 \x88 \x89 \x8a \x8b \x8c \x8d \x8e \x8f \x90 \x91 \x92 \x93 
+  \x94 \x95 \x96 \x97 \x98 \x99 \x9a \x9b \x9c \x9d \x9e \x9f \xa1 \xa2 \xa3 
+  \xa4 \xa5 \xa6 \xa7 \xa8 \xa9 \xaa \xab \xac \xad \xae \xaf \xb0 \xb1 \xb2 
+  \xb3 \xb4 \xb5 \xb6 \xb7 \xb8 \xb9 \xba \xbb \xbc \xbd \xbe \xbf \xc0 \xc1 
+  \xc2 \xc3 \xc4 \xc5 \xc6 \xc7 \xc8 \xc9 \xca \xcb \xcc \xcd \xce \xcf \xd0 
+  \xd1 \xd2 \xd3 \xd4 \xd5 \xd6 \xd7 \xd8 \xd9 \xda \xdb \xdc \xdd \xde \xdf 
+  \xe0 \xe1 \xe2 \xe3 \xe4 \xe5 \xe6 \xe7 \xe8 \xe9 \xea \xeb \xec \xed \xee 
+  \xef \xf0 \xf1 \xf2 \xf3 \xf4 \xf5 \xf6 \xf7 \xf8 \xf9 \xfa \xfb \xfc \xfd 
+  \xfe \xff 
+    \x{a2} \x{84}
+ 0: \x{a2} \x{84}
+    A Z
+ 0: A Z
+
+/a+/8
+    a\x{123}aa\>1
+ 0: aa
+    a\x{123}aa\>2
+ 0: aa
+    a\x{123}aa\>3
+ 0: a
+    a\x{123}aa\>4
+No match
+    a\x{123}aa\>5
+Error -24 (bad offset value)
+    a\x{123}aa\>6
+Error -24 (bad offset value)
+
+/\x{1234}+/iS8I
+Capturing subpattern count = 0
+Options: caseless utf
+First char = \x{1234}
+No need char
+Subject length lower bound = 1
+No set of starting bytes
+
+/\x{1234}+?/iS8I
+Capturing subpattern count = 0
+Options: caseless utf
+First char = \x{1234}
+No need char
+Subject length lower bound = 1
+No set of starting bytes
+
+/\x{1234}++/iS8I
+Capturing subpattern count = 0
+Options: caseless utf
+First char = \x{1234}
+No need char
+Subject length lower bound = 1
+No set of starting bytes
+
+/\x{1234}{2}/iS8I
+Capturing subpattern count = 0
+Options: caseless utf
+First char = \x{1234}
+Need char = \x{1234}
+Subject length lower bound = 2
+No set of starting bytes
+
+/[^\x{c4}]/8DZ
+------------------------------------------------------------------
+        Bra
+        [^\x{c4}]
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+No first char
+No need char
+
+/X+\x{200}/8DZ
+------------------------------------------------------------------
+        Bra
+        X++
+        \x{200}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = 'X'
+Need char = \x{200}
+
+/\R/SI8
+Capturing subpattern count = 0
+Options: utf
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: \x0a \x0b \x0c \x0d \x85 \xff 
+
+/-- Check bad offset --/
+
+/a/8
+    \x{10000}\>1
+Error -11 (bad UTF-16 offset)
+    \x{10000}ab\>1
+Error -11 (bad UTF-16 offset)
+    \x{10000}ab\>2
+ 0: a
+    \x{10000}ab\>3
+No match
+    \x{10000}ab\>4
+No match
+    \x{10000}ab\>5
+Error -24 (bad offset value)
+
+//8
+Failed: invalid UTF-16 string at offset 0
+
+/\w+\x{C4}/8BZ
+------------------------------------------------------------------
+        Bra
+        \w++
+        \x{c4}
+        Ket
+        End
+------------------------------------------------------------------
+    a\x{C4}\x{C4}
+ 0: a\x{c4}
+
+/\w+\x{C4}/8BZT1
+------------------------------------------------------------------
+        Bra
+        \w+
+        \x{c4}
+        Ket
+        End
+------------------------------------------------------------------
+    a\x{C4}\x{C4}
+ 0: a\x{c4}\x{c4}
+    
+/\W+\x{C4}/8BZ
+------------------------------------------------------------------
+        Bra
+        \W+
+        \x{c4}
+        Ket
+        End
+------------------------------------------------------------------
+    !\x{C4}
+ 0: !\x{c4}
+ 
+/\W+\x{C4}/8BZT1
+------------------------------------------------------------------
+        Bra
+        \W++
+        \x{c4}
+        Ket
+        End
+------------------------------------------------------------------
+    !\x{C4}
+ 0: !\x{c4}
+
+/\W+\x{A1}/8BZ
+------------------------------------------------------------------
+        Bra
+        \W+
+        \x{a1}
+        Ket
+        End
+------------------------------------------------------------------
+    !\x{A1}
+ 0: !\x{a1}
+ 
+/\W+\x{A1}/8BZT1
+------------------------------------------------------------------
+        Bra
+        \W+
+        \x{a1}
+        Ket
+        End
+------------------------------------------------------------------
+    !\x{A1}
+ 0: !\x{a1}
+
+/X\s+\x{A0}/8BZ
+------------------------------------------------------------------
+        Bra
+        X
+        \s++
+        \x{a0}
+        Ket
+        End
+------------------------------------------------------------------
+    X\x20\x{A0}\x{A0}
+ 0: X \x{a0}
+
+/X\s+\x{A0}/8BZT1
+------------------------------------------------------------------
+        Bra
+        X
+        \s+
+        \x{a0}
+        Ket
+        End
+------------------------------------------------------------------
+    X\x20\x{A0}\x{A0}
+ 0: X \x{a0}\x{a0}
+
+/\S+\x{A0}/8BZ
+------------------------------------------------------------------
+        Bra
+        \S+
+        \x{a0}
+        Ket
+        End
+------------------------------------------------------------------
+    X\x{A0}\x{A0}
+ 0: X\x{a0}\x{a0}
+
+/\S+\x{A0}/8BZT1
+------------------------------------------------------------------
+        Bra
+        \S++
+        \x{a0}
+        Ket
+        End
+------------------------------------------------------------------
+    X\x{A0}\x{A0}
+ 0: X\x{a0}
+
+/\x{a0}+\s!/8BZ
+------------------------------------------------------------------
+        Bra
+        \x{a0}++
+        \s
+        !
+        Ket
+        End
+------------------------------------------------------------------
+    \x{a0}\x20!
+ 0: \x{a0} !
+
+/\x{a0}+\s!/8BZT1
+------------------------------------------------------------------
+        Bra
+        \x{a0}+
+        \s
+        !
+        Ket
+        End
+------------------------------------------------------------------
+    \x{a0}\x20!
+ 0: \x{a0} !
+
+/(*UTF)abc/9
+Failed: setting UTF is disabled by the application at offset 0
+
+/abc/89
+Failed: setting UTF is disabled by the application at offset 0
+
+/-- End of testinput18 --/
diff --git a/pcre/testdata/testoutput18-32 b/pcre/testdata/testoutput18-32
new file mode 100644
index 00000000000..e917e3baec4
--- /dev/null
+++ b/pcre/testdata/testoutput18-32
@@ -0,0 +1,1021 @@
+/-- This set of tests is for UTF-16 and UTF-32 support, and is relevant only to the
+    16- and 32-bit library. --/
+
+/xxx/8?DZSS
+**Failed: invalid UTF-8 string cannot be converted to UTF-32
+
+/abc/8
+    ]
+**Failed: invalid UTF-8 string cannot be used as input in UTF mode
+
+/X(\C{3})/8
+    X\x{11234}Y
+No match
+    X\x{11234}YZ
+ 0: X\x{11234}YZ
+ 1: \x{11234}YZ
+
+/X(\C{4})/8
+    X\x{11234}YZ
+No match
+    X\x{11234}YZW
+ 0: X\x{11234}YZW
+ 1: \x{11234}YZW
+
+/X\C*/8
+    XYZabcdce
+ 0: XYZabcdce
+
+/X\C*?/8
+    XYZabcde
+ 0: X
+
+/X\C{3,5}/8
+    Xabcdefg
+ 0: Xabcde
+    X\x{11234}Y
+No match
+    X\x{11234}YZ
+ 0: X\x{11234}YZ
+    X\x{11234}\x{512}
+No match
+    X\x{11234}\x{512}YZ
+ 0: X\x{11234}\x{512}YZ
+    X\x{11234}\x{512}\x{11234}Z
+ 0: X\x{11234}\x{512}\x{11234}Z
+
+/X\C{3,5}?/8
+    Xabcdefg
+ 0: Xabc
+    X\x{11234}Y
+No match
+    X\x{11234}YZ
+ 0: X\x{11234}YZ
+    X\x{11234}\x{512}YZ
+ 0: X\x{11234}\x{512}Y
+    *** Failers
+No match
+    X\x{11234}
+No match
+
+/a\Cb/8
+    aXb
+ 0: aXb
+    a\nb
+ 0: a\x{0a}b
+
+/a\C\Cb/8
+    a\x{12257}b
+No match
+    a\x{12257}\x{11234}b
+ 0: a\x{12257}\x{11234}b
+    ** Failers
+No match
+    a\x{100}b
+No match
+
+/ab\Cde/8
+    abXde
+ 0: abXde
+
+/-- Check maximum character size --/
+
+/\x{ffff}/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{ffff}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{ffff}
+No need char
+
+/\x{10000}/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{10000}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{10000}
+No need char
+
+/\x{100}/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{100}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{100}
+No need char
+
+/\x{1000}/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{1000}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{1000}
+No need char
+
+/\x{10000}/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{10000}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{10000}
+No need char
+
+/\x{100000}/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{100000}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{100000}
+No need char
+
+/\x{10ffff}/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{10ffff}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{10ffff}
+No need char
+
+/[\x{ff}]/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{ff}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{ff}
+No need char
+
+/[\x{100}]/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{100}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{100}
+No need char
+
+/\x80/8DZ
+------------------------------------------------------------------
+        Bra
+        \x80
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{80}
+No need char
+
+/\xff/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{ff}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{ff}
+No need char
+
+/\x{D55c}\x{ad6d}\x{C5B4}/DZ8
+------------------------------------------------------------------
+        Bra
+        \x{d55c}\x{ad6d}\x{c5b4}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{d55c}
+Need char = \x{c5b4}
+    \x{D55c}\x{ad6d}\x{C5B4}
+ 0: \x{d55c}\x{ad6d}\x{c5b4}
+
+/\x{65e5}\x{672c}\x{8a9e}/DZ8
+------------------------------------------------------------------
+        Bra
+        \x{65e5}\x{672c}\x{8a9e}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{65e5}
+Need char = \x{8a9e}
+    \x{65e5}\x{672c}\x{8a9e}
+ 0: \x{65e5}\x{672c}\x{8a9e}
+
+/\x{80}/DZ8
+------------------------------------------------------------------
+        Bra
+        \x80
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{80}
+No need char
+
+/\x{084}/DZ8
+------------------------------------------------------------------
+        Bra
+        \x{84}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{84}
+No need char
+
+/\x{104}/DZ8
+------------------------------------------------------------------
+        Bra
+        \x{104}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{104}
+No need char
+
+/\x{861}/DZ8
+------------------------------------------------------------------
+        Bra
+        \x{861}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{861}
+No need char
+
+/\x{212ab}/DZ8
+------------------------------------------------------------------
+        Bra
+        \x{212ab}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{212ab}
+No need char
+
+/-- This one is here not because it's different to Perl, but because the way
+the captured single-byte is displayed. (In Perl it becomes a character, and you
+can't tell the difference.) --/
+
+/X(\C)(.*)/8
+    X\x{1234}
+ 0: X\x{1234}
+ 1: \x{1234}
+ 2: 
+    X\nabc
+ 0: X\x{0a}abc
+ 1: \x{0a}
+ 2: abc
+
+/-- This one is here because Perl gives out a grumbly error message (quite
+correctly, but that messes up comparisons). --/
+
+/a\Cb/8
+    *** Failers
+No match
+    a\x{100}b
+ 0: a\x{100}b
+
+/[^ab\xC0-\xF0]/8SDZ
+------------------------------------------------------------------
+        Bra
+        [\x00-`c-\xbf\xf1-\xff] (neg)
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: \x00 \x01 \x02 \x03 \x04 \x05 \x06 \x07 \x08 \x09 \x0a 
+  \x0b \x0c \x0d \x0e \x0f \x10 \x11 \x12 \x13 \x14 \x15 \x16 \x17 \x18 \x19 
+  \x1a \x1b \x1c \x1d \x1e \x1f \x20 ! " # $ % & ' ( ) * + , - . / 0 1 2 3 4 
+  5 6 7 8 9 : ; < = > ? @ A B C D E F G H I J K L M N O P Q R S T U V W X Y 
+  Z [ \ ] ^ _ ` c d e f g h i j k l m n o p q r s t u v w x y z { | } ~ \x7f 
+  \x80 \x81 \x82 \x83 \x84 \x85 \x86 \x87 \x88 \x89 \x8a \x8b \x8c \x8d \x8e 
+  \x8f \x90 \x91 \x92 \x93 \x94 \x95 \x96 \x97 \x98 \x99 \x9a \x9b \x9c \x9d 
+  \x9e \x9f \xa0 \xa1 \xa2 \xa3 \xa4 \xa5 \xa6 \xa7 \xa8 \xa9 \xaa \xab \xac 
+  \xad \xae \xaf \xb0 \xb1 \xb2 \xb3 \xb4 \xb5 \xb6 \xb7 \xb8 \xb9 \xba \xbb 
+  \xbc \xbd \xbe \xbf \xf1 \xf2 \xf3 \xf4 \xf5 \xf6 \xf7 \xf8 \xf9 \xfa \xfb 
+  \xfc \xfd \xfe \xff 
+    \x{f1}
+ 0: \x{f1}
+    \x{bf}
+ 0: \x{bf}
+    \x{100}
+ 0: \x{100}
+    \x{1000}
+ 0: \x{1000}
+    *** Failers
+ 0: *
+    \x{c0}
+No match
+    \x{f0}
+No match
+
+/Ā{3,4}/8SDZ
+------------------------------------------------------------------
+        Bra
+        \x{100}{3}
+        \x{100}?
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{100}
+Need char = \x{100}
+Subject length lower bound = 3
+No set of starting bytes
+  \x{100}\x{100}\x{100}\x{100\x{100}
+ 0: \x{100}\x{100}\x{100}
+
+/(\x{100}+|x)/8SDZ
+------------------------------------------------------------------
+        Bra
+        CBra 1
+        \x{100}+
+        Alt
+        x
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 1
+Options: utf
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: x \xff 
+
+/(\x{100}*a|x)/8SDZ
+------------------------------------------------------------------
+        Bra
+        CBra 1
+        \x{100}*+
+        a
+        Alt
+        x
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 1
+Options: utf
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: a x \xff 
+
+/(\x{100}{0,2}a|x)/8SDZ
+------------------------------------------------------------------
+        Bra
+        CBra 1
+        \x{100}{0,2}
+        a
+        Alt
+        x
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 1
+Options: utf
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: a x \xff 
+
+/(\x{100}{1,2}a|x)/8SDZ
+------------------------------------------------------------------
+        Bra
+        CBra 1
+        \x{100}
+        \x{100}{0,1}
+        a
+        Alt
+        x
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 1
+Options: utf
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: x \xff 
+
+/\x{100}/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{100}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{100}
+No need char
+
+/a\x{100}\x{101}*/8DZ
+------------------------------------------------------------------
+        Bra
+        a\x{100}
+        \x{101}*
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = 'a'
+Need char = \x{100}
+
+/a\x{100}\x{101}+/8DZ
+------------------------------------------------------------------
+        Bra
+        a\x{100}
+        \x{101}+
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = 'a'
+Need char = \x{101}
+
+/[^\x{c4}]/DZ
+------------------------------------------------------------------
+        Bra
+        [^\x{c4}]
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+
+/[\x{100}]/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{100}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{100}
+No need char
+    \x{100}
+ 0: \x{100}
+    Z\x{100}
+ 0: \x{100}
+    \x{100}Z
+ 0: \x{100}
+    *** Failers
+No match
+
+/[\xff]/DZ8
+------------------------------------------------------------------
+        Bra
+        \x{ff}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{ff}
+No need char
+    >\x{ff}<
+ 0: \x{ff}
+
+/[^\xff]/8DZ
+------------------------------------------------------------------
+        Bra
+        [^\x{ff}]
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+No first char
+No need char
+
+/\x{100}abc(xyz(?1))/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{100}abc
+        CBra 1
+        xyz
+        Recurse
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 1
+Options: utf
+First char = \x{100}
+Need char = 'z'
+
+/\777/8I
+Capturing subpattern count = 0
+Options: utf
+First char = \x{1ff}
+No need char
+  \x{1ff}
+ 0: \x{1ff}
+  \777
+ 0: \x{1ff}
+
+/\x{100}+\x{200}/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{100}++
+        \x{200}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{100}
+Need char = \x{200}
+
+/\x{100}+X/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{100}++
+        X
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = \x{100}
+Need char = 'X'
+
+/^[\QĀ\E-\QŐ\E/BZ8
+Failed: missing terminating ] for character class at offset 13
+
+/X/8
+    \x{d800}
+Error -10 (bad UTF-32 string) offset=0 reason=1
+    \x{d800}\?
+No match
+    \x{da00}
+Error -10 (bad UTF-32 string) offset=0 reason=1
+    \x{da00}\?
+No match
+    \x{dc00}
+Error -10 (bad UTF-32 string) offset=0 reason=1
+    \x{dc00}\?
+No match
+    \x{de00}
+Error -10 (bad UTF-32 string) offset=0 reason=1
+    \x{de00}\?
+No match
+    \x{dfff}
+Error -10 (bad UTF-32 string) offset=0 reason=1
+    \x{dfff}\?
+No match
+    \x{110000}
+Error -10 (bad UTF-32 string) offset=0 reason=3
+    \x{d800}\x{1234}
+Error -10 (bad UTF-32 string) offset=0 reason=1
+
+/(*UTF16)\x{11234}/
+Failed: (*VERB) not recognized or malformed at offset 5
+
+/(*UTF)\x{11234}/I
+Capturing subpattern count = 0
+Options: utf
+First char = \x{11234}
+No need char
+  abcd\x{11234}pqr
+ 0: \x{11234}
+
+/(*UTF-32)\x{11234}/
+Failed: (*VERB) not recognized or malformed at offset 5
+
+/(*CRLF)(*UTF16)(*BSR_UNICODE)a\Rb/I
+Failed: (*VERB) not recognized or malformed at offset 12
+
+/(*CRLF)(*UTF32)(*BSR_UNICODE)a\Rb/I
+Capturing subpattern count = 0
+Options: bsr_unicode utf
+Forced newline sequence: CRLF
+First char = 'a'
+Need char = 'b'
+
+/\h/SI8
+Capturing subpattern count = 0
+Options: utf
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: \x09 \x20 \xa0 \xff 
+    ABC\x{09}
+ 0: \x{09}
+    ABC\x{20}
+ 0:  
+    ABC\x{a0}
+ 0: \x{a0}
+    ABC\x{1680}
+ 0: \x{1680}
+    ABC\x{180e}
+ 0: \x{180e}
+    ABC\x{2000}
+ 0: \x{2000}
+    ABC\x{202f}
+ 0: \x{202f}
+    ABC\x{205f}
+ 0: \x{205f}
+    ABC\x{3000}
+ 0: \x{3000}
+
+/\v/SI8
+Capturing subpattern count = 0
+Options: utf
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: \x0a \x0b \x0c \x0d \x85 \xff 
+    ABC\x{0a}
+ 0: \x{0a}
+    ABC\x{0b}
+ 0: \x{0b}
+    ABC\x{0c}
+ 0: \x{0c}
+    ABC\x{0d}
+ 0: \x{0d}
+    ABC\x{85}
+ 0: \x{85}
+    ABC\x{2028}
+ 0: \x{2028}
+
+/\h*A/SI8
+Capturing subpattern count = 0
+Options: utf
+No first char
+Need char = 'A'
+Subject length lower bound = 1
+Starting byte set: \x09 \x20 A \xa0 \xff 
+    CDBABC
+ 0: A
+    \x{2000}ABC 
+ 0: \x{2000}A
+
+/\R*A/SI8
+Capturing subpattern count = 0
+Options: utf
+No first char
+Need char = 'A'
+Subject length lower bound = 1
+Starting byte set: \x0a \x0b \x0c \x0d A \x85 \xff 
+    CDBABC
+ 0: A
+    \x{2028}A  
+ 0: \x{2028}A
+
+/\v+A/SI8
+Capturing subpattern count = 0
+Options: utf
+No first char
+Need char = 'A'
+Subject length lower bound = 2
+Starting byte set: \x0a \x0b \x0c \x0d \x85 \xff 
+
+/\s?xxx\s/8SI
+Capturing subpattern count = 0
+Options: utf
+No first char
+Need char = 'x'
+Subject length lower bound = 4
+Starting byte set: \x09 \x0a \x0c \x0d \x20 x 
+
+/\sxxx\s/I8ST1
+Capturing subpattern count = 0
+Options: utf
+No first char
+Need char = 'x'
+Subject length lower bound = 5
+Starting byte set: \x09 \x0a \x0c \x0d \x20 \x85 \xa0 
+    AB\x{85}xxx\x{a0}XYZ
+ 0: \x{85}xxx\x{a0}
+    AB\x{a0}xxx\x{85}XYZ
+ 0: \x{a0}xxx\x{85}
+
+/\S \S/I8ST1
+Capturing subpattern count = 0
+Options: utf
+No first char
+Need char = ' '
+Subject length lower bound = 3
+Starting byte set: \x00 \x01 \x02 \x03 \x04 \x05 \x06 \x07 \x08 \x0b \x0e 
+  \x0f \x10 \x11 \x12 \x13 \x14 \x15 \x16 \x17 \x18 \x19 \x1a \x1b \x1c \x1d 
+  \x1e \x1f ! " # $ % & ' ( ) * + , - . / 0 1 2 3 4 5 6 7 8 9 : ; < = > ? @ 
+  A B C D E F G H I J K L M N O P Q R S T U V W X Y Z [ \ ] ^ _ ` a b c d e 
+  f g h i j k l m n o p q r s t u v w x y z { | } ~ \x7f \x80 \x81 \x82 \x83 
+  \x84 \x86 \x87 \x88 \x89 \x8a \x8b \x8c \x8d \x8e \x8f \x90 \x91 \x92 \x93 
+  \x94 \x95 \x96 \x97 \x98 \x99 \x9a \x9b \x9c \x9d \x9e \x9f \xa1 \xa2 \xa3 
+  \xa4 \xa5 \xa6 \xa7 \xa8 \xa9 \xaa \xab \xac \xad \xae \xaf \xb0 \xb1 \xb2 
+  \xb3 \xb4 \xb5 \xb6 \xb7 \xb8 \xb9 \xba \xbb \xbc \xbd \xbe \xbf \xc0 \xc1 
+  \xc2 \xc3 \xc4 \xc5 \xc6 \xc7 \xc8 \xc9 \xca \xcb \xcc \xcd \xce \xcf \xd0 
+  \xd1 \xd2 \xd3 \xd4 \xd5 \xd6 \xd7 \xd8 \xd9 \xda \xdb \xdc \xdd \xde \xdf 
+  \xe0 \xe1 \xe2 \xe3 \xe4 \xe5 \xe6 \xe7 \xe8 \xe9 \xea \xeb \xec \xed \xee 
+  \xef \xf0 \xf1 \xf2 \xf3 \xf4 \xf5 \xf6 \xf7 \xf8 \xf9 \xfa \xfb \xfc \xfd 
+  \xfe \xff 
+    \x{a2} \x{84}
+ 0: \x{a2} \x{84}
+    A Z
+ 0: A Z
+
+/a+/8
+    a\x{123}aa\>1
+ 0: aa
+    a\x{123}aa\>2
+ 0: aa
+    a\x{123}aa\>3
+ 0: a
+    a\x{123}aa\>4
+No match
+    a\x{123}aa\>5
+Error -24 (bad offset value)
+    a\x{123}aa\>6
+Error -24 (bad offset value)
+
+/\x{1234}+/iS8I
+Capturing subpattern count = 0
+Options: caseless utf
+First char = \x{1234}
+No need char
+Subject length lower bound = 1
+No set of starting bytes
+
+/\x{1234}+?/iS8I
+Capturing subpattern count = 0
+Options: caseless utf
+First char = \x{1234}
+No need char
+Subject length lower bound = 1
+No set of starting bytes
+
+/\x{1234}++/iS8I
+Capturing subpattern count = 0
+Options: caseless utf
+First char = \x{1234}
+No need char
+Subject length lower bound = 1
+No set of starting bytes
+
+/\x{1234}{2}/iS8I
+Capturing subpattern count = 0
+Options: caseless utf
+First char = \x{1234}
+Need char = \x{1234}
+Subject length lower bound = 2
+No set of starting bytes
+
+/[^\x{c4}]/8DZ
+------------------------------------------------------------------
+        Bra
+        [^\x{c4}]
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+No first char
+No need char
+
+/X+\x{200}/8DZ
+------------------------------------------------------------------
+        Bra
+        X++
+        \x{200}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = 'X'
+Need char = \x{200}
+
+/\R/SI8
+Capturing subpattern count = 0
+Options: utf
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: \x0a \x0b \x0c \x0d \x85 \xff 
+
+/-- Check bad offset --/
+
+/a/8
+    \x{10000}\>1
+No match
+    \x{10000}ab\>1
+ 0: a
+    \x{10000}ab\>2
+No match
+    \x{10000}ab\>3
+No match
+    \x{10000}ab\>4
+Error -24 (bad offset value)
+    \x{10000}ab\>5
+Error -24 (bad offset value)
+
+//8
+**Failed: character value is ill-formed UTF-32
+
+/\w+\x{C4}/8BZ
+------------------------------------------------------------------
+        Bra
+        \w++
+        \x{c4}
+        Ket
+        End
+------------------------------------------------------------------
+    a\x{C4}\x{C4}
+ 0: a\x{c4}
+
+/\w+\x{C4}/8BZT1
+------------------------------------------------------------------
+        Bra
+        \w+
+        \x{c4}
+        Ket
+        End
+------------------------------------------------------------------
+    a\x{C4}\x{C4}
+ 0: a\x{c4}\x{c4}
+    
+/\W+\x{C4}/8BZ
+------------------------------------------------------------------
+        Bra
+        \W+
+        \x{c4}
+        Ket
+        End
+------------------------------------------------------------------
+    !\x{C4}
+ 0: !\x{c4}
+ 
+/\W+\x{C4}/8BZT1
+------------------------------------------------------------------
+        Bra
+        \W++
+        \x{c4}
+        Ket
+        End
+------------------------------------------------------------------
+    !\x{C4}
+ 0: !\x{c4}
+
+/\W+\x{A1}/8BZ
+------------------------------------------------------------------
+        Bra
+        \W+
+        \x{a1}
+        Ket
+        End
+------------------------------------------------------------------
+    !\x{A1}
+ 0: !\x{a1}
+ 
+/\W+\x{A1}/8BZT1
+------------------------------------------------------------------
+        Bra
+        \W+
+        \x{a1}
+        Ket
+        End
+------------------------------------------------------------------
+    !\x{A1}
+ 0: !\x{a1}
+
+/X\s+\x{A0}/8BZ
+------------------------------------------------------------------
+        Bra
+        X
+        \s++
+        \x{a0}
+        Ket
+        End
+------------------------------------------------------------------
+    X\x20\x{A0}\x{A0}
+ 0: X \x{a0}
+
+/X\s+\x{A0}/8BZT1
+------------------------------------------------------------------
+        Bra
+        X
+        \s+
+        \x{a0}
+        Ket
+        End
+------------------------------------------------------------------
+    X\x20\x{A0}\x{A0}
+ 0: X \x{a0}\x{a0}
+
+/\S+\x{A0}/8BZ
+------------------------------------------------------------------
+        Bra
+        \S+
+        \x{a0}
+        Ket
+        End
+------------------------------------------------------------------
+    X\x{A0}\x{A0}
+ 0: X\x{a0}\x{a0}
+
+/\S+\x{A0}/8BZT1
+------------------------------------------------------------------
+        Bra
+        \S++
+        \x{a0}
+        Ket
+        End
+------------------------------------------------------------------
+    X\x{A0}\x{A0}
+ 0: X\x{a0}
+
+/\x{a0}+\s!/8BZ
+------------------------------------------------------------------
+        Bra
+        \x{a0}++
+        \s
+        !
+        Ket
+        End
+------------------------------------------------------------------
+    \x{a0}\x20!
+ 0: \x{a0} !
+
+/\x{a0}+\s!/8BZT1
+------------------------------------------------------------------
+        Bra
+        \x{a0}+
+        \s
+        !
+        Ket
+        End
+------------------------------------------------------------------
+    \x{a0}\x20!
+ 0: \x{a0} !
+
+/(*UTF)abc/9
+Failed: setting UTF is disabled by the application at offset 0
+
+/abc/89
+Failed: setting UTF is disabled by the application at offset 0
+
+/-- End of testinput18 --/
diff --git a/pcre/testdata/testoutput19 b/pcre/testdata/testoutput19
new file mode 100644
index 00000000000..ccc198cc153
--- /dev/null
+++ b/pcre/testdata/testoutput19
@@ -0,0 +1,88 @@
+/-- This set of tests is for Unicode property support, relevant only to the
+    16- and 32-bit library. --/
+    
+/A\x{391}\x{10427}\x{ff3a}\x{1fb0}/8iDZ
+------------------------------------------------------------------
+        Bra
+     /i A\x{391}\x{10427}\x{ff3a}\x{1fb0}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: caseless utf
+First char = 'A' (caseless)
+Need char = \x{1fb0} (caseless)
+
+/A\x{391}\x{10427}\x{ff3a}\x{1fb0}/8DZ
+------------------------------------------------------------------
+        Bra
+        A\x{391}\x{10427}\x{ff3a}\x{1fb0}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = 'A'
+Need char = \x{1fb0}
+
+/AB\x{1fb0}/8DZ
+------------------------------------------------------------------
+        Bra
+        AB\x{1fb0}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = 'A'
+Need char = \x{1fb0}
+
+/AB\x{1fb0}/8DZi
+------------------------------------------------------------------
+        Bra
+     /i AB\x{1fb0}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: caseless utf
+First char = 'A' (caseless)
+Need char = \x{1fb0} (caseless)
+
+/\x{401}\x{420}\x{421}\x{422}\x{423}\x{424}\x{425}\x{426}\x{427}\x{428}\x{429}\x{42a}\x{42b}\x{42c}\x{42d}\x{42e}\x{42f}/8iSI
+Capturing subpattern count = 0
+Options: caseless utf
+First char = \x{401} (caseless)
+Need char = \x{42f} (caseless)
+Subject length lower bound = 17
+No set of starting bytes
+    \x{401}\x{420}\x{421}\x{422}\x{423}\x{424}\x{425}\x{426}\x{427}\x{428}\x{429}\x{42a}\x{42b}\x{42c}\x{42d}\x{42e}\x{42f}
+ 0: \x{401}\x{420}\x{421}\x{422}\x{423}\x{424}\x{425}\x{426}\x{427}\x{428}\x{429}\x{42a}\x{42b}\x{42c}\x{42d}\x{42e}\x{42f}
+    \x{451}\x{440}\x{441}\x{442}\x{443}\x{444}\x{445}\x{446}\x{447}\x{448}\x{449}\x{44a}\x{44b}\x{44c}\x{44d}\x{44e}\x{44f}
+ 0: \x{451}\x{440}\x{441}\x{442}\x{443}\x{444}\x{445}\x{446}\x{447}\x{448}\x{449}\x{44a}\x{44b}\x{44c}\x{44d}\x{44e}\x{44f}
+
+/[ⱥ]/8iBZ
+------------------------------------------------------------------
+        Bra
+     /i \x{2c65}
+        Ket
+        End
+------------------------------------------------------------------
+
+/[^ⱥ]/8iBZ
+------------------------------------------------------------------
+        Bra
+     /i [^\x{2c65}]
+        Ket
+        End
+------------------------------------------------------------------
+
+/[[:blank:]]/WBZ
+------------------------------------------------------------------
+        Bra
+        [\x09 \xa0\x{1680}\x{180e}\x{2000}-\x{200a}\x{202f}\x{205f}\x{3000}]
+        Ket
+        End
+------------------------------------------------------------------
+
+/-- End of testinput19 --/ 
diff --git a/pcre/testdata/testoutput2 b/pcre/testdata/testoutput2
new file mode 100644
index 00000000000..fd958c2eb7e
--- /dev/null
+++ b/pcre/testdata/testoutput2
@@ -0,0 +1,12668 @@
+/-- This set of tests is not Perl-compatible. It checks on special features
+    of PCRE's API, error diagnostics, and the compiled code of some patterns.
+    It also checks the non-Perl syntax the PCRE supports (Python, .NET, 
+    Oniguruma). Finally, there are some tests where PCRE and Perl differ, 
+    either because PCRE can't be compatible, or there is a possible Perl 
+    bug.
+    
+    NOTE: This is a non-UTF set of tests. When UTF support is needed, use
+    test 5, and if Unicode Property Support is needed, use test 7. --/
+  
+/(a)b|/I
+Capturing subpattern count = 1
+No options
+No first char
+No need char
+
+/abc/I
+Capturing subpattern count = 0
+No options
+First char = 'a'
+Need char = 'c'
+    abc
+ 0: abc
+    defabc
+ 0: abc
+    \Aabc
+ 0: abc
+    *** Failers
+No match
+    \Adefabc
+No match
+    ABC
+No match
+
+/^abc/I
+Capturing subpattern count = 0
+Options: anchored
+No first char
+No need char
+    abc
+ 0: abc
+    \Aabc
+ 0: abc
+    *** Failers
+No match
+    defabc
+No match
+    \Adefabc
+No match
+
+/a+bc/I
+Capturing subpattern count = 0
+No options
+First char = 'a'
+Need char = 'c'
+
+/a*bc/I
+Capturing subpattern count = 0
+No options
+No first char
+Need char = 'c'
+
+/a{3}bc/I
+Capturing subpattern count = 0
+No options
+First char = 'a'
+Need char = 'c'
+
+/(abc|a+z)/I
+Capturing subpattern count = 1
+No options
+First char = 'a'
+No need char
+
+/^abc$/I
+Capturing subpattern count = 0
+Options: anchored
+No first char
+No need char
+    abc
+ 0: abc
+    *** Failers
+No match
+    def\nabc
+No match
+
+/ab\idef/X
+Failed: unrecognized character follows \ at offset 3
+
+/(?X)ab\idef/X
+Failed: unrecognized character follows \ at offset 7
+
+/x{5,4}/
+Failed: numbers out of order in {} quantifier at offset 5
+
+/z{65536}/
+Failed: number too big in {} quantifier at offset 7
+
+/[abcd/
+Failed: missing terminating ] for character class at offset 5
+
+/(?X)[\B]/
+Failed: invalid escape sequence in character class at offset 6
+
+/(?X)[\R]/
+Failed: invalid escape sequence in character class at offset 6
+
+/(?X)[\X]/
+Failed: invalid escape sequence in character class at offset 6
+
+/[\B]/BZ
+------------------------------------------------------------------
+        Bra
+        B
+        Ket
+        End
+------------------------------------------------------------------
+
+/[\R]/BZ
+------------------------------------------------------------------
+        Bra
+        R
+        Ket
+        End
+------------------------------------------------------------------
+
+/[\X]/BZ
+------------------------------------------------------------------
+        Bra
+        X
+        Ket
+        End
+------------------------------------------------------------------
+
+/[z-a]/
+Failed: range out of order in character class at offset 3
+
+/^*/
+Failed: nothing to repeat at offset 1
+
+/(abc/
+Failed: missing ) at offset 4
+
+/(?# abc/
+Failed: missing ) after comment at offset 7
+
+/(?z)abc/
+Failed: unrecognized character after (? or (?- at offset 2
+
+/.*b/I
+Capturing subpattern count = 0
+No options
+First char at start or follows newline
+Need char = 'b'
+
+/.*?b/I
+Capturing subpattern count = 0
+No options
+First char at start or follows newline
+Need char = 'b'
+
+/cat|dog|elephant/I
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+    this sentence eventually mentions a cat
+ 0: cat
+    this sentences rambles on and on for a while and then reaches elephant
+ 0: elephant
+
+/cat|dog|elephant/IS
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+Subject length lower bound = 3
+Starting byte set: c d e 
+    this sentence eventually mentions a cat
+ 0: cat
+    this sentences rambles on and on for a while and then reaches elephant
+ 0: elephant
+
+/cat|dog|elephant/IiS
+Capturing subpattern count = 0
+Options: caseless
+No first char
+No need char
+Subject length lower bound = 3
+Starting byte set: C D E c d e 
+    this sentence eventually mentions a CAT cat
+ 0: CAT
+    this sentences rambles on and on for a while to elephant ElePhant
+ 0: elephant
+
+/a|[bcd]/IS
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: a b c d 
+
+/(a|[^\dZ])/IS
+Capturing subpattern count = 1
+No options
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: \x00 \x01 \x02 \x03 \x04 \x05 \x06 \x07 \x08 \x09 \x0a 
+  \x0b \x0c \x0d \x0e \x0f \x10 \x11 \x12 \x13 \x14 \x15 \x16 \x17 \x18 \x19 
+  \x1a \x1b \x1c \x1d \x1e \x1f \x20 ! " # $ % & ' ( ) * + , - . / : ; < = > 
+  ? @ A B C D E F G H I J K L M N O P Q R S T U V W X Y [ \ ] ^ _ ` a b c d 
+  e f g h i j k l m n o p q r s t u v w x y z { | } ~ \x7f \x80 \x81 \x82 \x83 
+  \x84 \x85 \x86 \x87 \x88 \x89 \x8a \x8b \x8c \x8d \x8e \x8f \x90 \x91 \x92 
+  \x93 \x94 \x95 \x96 \x97 \x98 \x99 \x9a \x9b \x9c \x9d \x9e \x9f \xa0 \xa1 
+  \xa2 \xa3 \xa4 \xa5 \xa6 \xa7 \xa8 \xa9 \xaa \xab \xac \xad \xae \xaf \xb0 
+  \xb1 \xb2 \xb3 \xb4 \xb5 \xb6 \xb7 \xb8 \xb9 \xba \xbb \xbc \xbd \xbe \xbf 
+  \xc0 \xc1 \xc2 \xc3 \xc4 \xc5 \xc6 \xc7 \xc8 \xc9 \xca \xcb \xcc \xcd \xce 
+  \xcf \xd0 \xd1 \xd2 \xd3 \xd4 \xd5 \xd6 \xd7 \xd8 \xd9 \xda \xdb \xdc \xdd 
+  \xde \xdf \xe0 \xe1 \xe2 \xe3 \xe4 \xe5 \xe6 \xe7 \xe8 \xe9 \xea \xeb \xec 
+  \xed \xee \xef \xf0 \xf1 \xf2 \xf3 \xf4 \xf5 \xf6 \xf7 \xf8 \xf9 \xfa \xfb 
+  \xfc \xfd \xfe \xff 
+
+/(a|b)*[\s]/IS
+Capturing subpattern count = 1
+No options
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: \x09 \x0a \x0c \x0d \x20 a b 
+
+/(ab\2)/
+Failed: reference to non-existent subpattern at offset 6
+
+/{4,5}abc/
+Failed: nothing to repeat at offset 4
+
+/(a)(b)(c)\2/I
+Capturing subpattern count = 3
+Max back reference = 2
+No options
+First char = 'a'
+Need char = 'c'
+    abcb
+ 0: abcb
+ 1: a
+ 2: b
+ 3: c
+    \O0abcb
+Matched, but too many substrings
+    \O3abcb
+Matched, but too many substrings
+ 0: abcb
+    \O6abcb
+Matched, but too many substrings
+ 0: abcb
+ 1: a
+    \O9abcb
+Matched, but too many substrings
+ 0: abcb
+ 1: a
+ 2: b
+    \O12abcb
+ 0: abcb
+ 1: a
+ 2: b
+ 3: c
+
+/(a)bc|(a)(b)\2/I
+Capturing subpattern count = 3
+Max back reference = 2
+No options
+First char = 'a'
+No need char
+    abc
+ 0: abc
+ 1: a
+    \O0abc
+Matched, but too many substrings
+    \O3abc
+Matched, but too many substrings
+ 0: abc
+    \O6abc
+ 0: abc
+ 1: a
+    aba
+ 0: aba
+ 1: 
+ 2: a
+ 3: b
+    \O0aba
+Matched, but too many substrings
+    \O3aba
+Matched, but too many substrings
+ 0: aba
+    \O6aba
+Matched, but too many substrings
+ 0: aba
+ 1: 
+    \O9aba
+Matched, but too many substrings
+ 0: aba
+ 1: 
+ 2: a
+    \O12aba
+ 0: aba
+ 1: 
+ 2: a
+ 3: b
+
+/abc$/IE
+Capturing subpattern count = 0
+Options: dollar_endonly
+First char = 'a'
+Need char = 'c'
+    abc
+ 0: abc
+    *** Failers
+No match
+    abc\n
+No match
+    abc\ndef
+No match
+
+/(a)(b)(c)(d)(e)\6/
+Failed: reference to non-existent subpattern at offset 17
+
+/the quick brown fox/I
+Capturing subpattern count = 0
+No options
+First char = 't'
+Need char = 'x'
+    the quick brown fox
+ 0: the quick brown fox
+    this is a line with the quick brown fox
+ 0: the quick brown fox
+
+/the quick brown fox/IA
+Capturing subpattern count = 0
+Options: anchored
+No first char
+No need char
+    the quick brown fox
+ 0: the quick brown fox
+    *** Failers
+No match
+    this is a line with the quick brown fox
+No match
+
+/ab(?z)cd/
+Failed: unrecognized character after (? or (?- at offset 4
+
+/^abc|def/I
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+    abcdef
+ 0: abc
+    abcdef\B
+ 0: def
+
+/.*((abc)$|(def))/I
+Capturing subpattern count = 3
+No options
+First char at start or follows newline
+No need char
+    defabc
+ 0: defabc
+ 1: abc
+ 2: abc
+    \Zdefabc
+ 0: def
+ 1: def
+ 2: 
+ 3: def
+
+/)/
+Failed: unmatched parentheses at offset 0
+
+/a[]b/
+Failed: missing terminating ] for character class at offset 4
+
+/[^aeiou ]{3,}/I
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+    co-processors, and for
+ 0: -pr
+
+/<.*>/I
+Capturing subpattern count = 0
+No options
+First char = '<'
+Need char = '>'
+    abcghinop
+ 0: ghi
+
+/<.*?>/I
+Capturing subpattern count = 0
+No options
+First char = '<'
+Need char = '>'
+    abcghinop
+ 0: 
+
+/<.*>/IU
+Capturing subpattern count = 0
+Options: ungreedy
+First char = '<'
+Need char = '>'
+    abcghinop
+ 0: 
+
+/(?U)<.*>/I
+Capturing subpattern count = 0
+Options: ungreedy
+First char = '<'
+Need char = '>'
+    abcghinop
+ 0: 
+
+/<.*?>/IU
+Capturing subpattern count = 0
+Options: ungreedy
+First char = '<'
+Need char = '>'
+    abcghinop
+ 0: ghi
+
+/={3,}/IU
+Capturing subpattern count = 0
+Options: ungreedy
+First char = '='
+Need char = '='
+    abc========def
+ 0: ===
+
+/(?U)={3,}?/I
+Capturing subpattern count = 0
+Options: ungreedy
+First char = '='
+Need char = '='
+    abc========def
+ 0: ========
+
+/(?^abc)/Im
+Capturing subpattern count = 0
+Options: multiline
+First char at start or follows newline
+Need char = 'c'
+    abc
+ 0: abc
+    def\nabc
+ 0: abc
+    *** Failers
+No match
+    defabc
+No match
+
+/(?<=ab(c+)d)ef/
+Failed: lookbehind assertion is not fixed length at offset 11
+
+/(?<=ab(?<=c+)d)ef/
+Failed: lookbehind assertion is not fixed length at offset 12
+
+/(?<=ab(c|de)f)g/
+Failed: lookbehind assertion is not fixed length at offset 13
+
+/The next three are in testinput2 because they have variable length branches/
+
+/(?<=bullock|donkey)-cart/I
+Capturing subpattern count = 0
+Max lookbehind = 7
+No options
+First char = '-'
+Need char = 't'
+    the bullock-cart
+ 0: -cart
+    a donkey-cart race
+ 0: -cart
+    *** Failers
+No match
+    cart
+No match
+    horse-and-cart
+No match
+
+/(?<=ab(?i)x|y|z)/I
+Capturing subpattern count = 0
+Max lookbehind = 3
+No options
+No first char
+No need char
+
+/(?>.*)(?<=(abcd)|(xyz))/I
+Capturing subpattern count = 2
+Max lookbehind = 4
+No options
+No first char
+No need char
+    alphabetabcd
+ 0: alphabetabcd
+ 1: abcd
+    endingxyz
+ 0: endingxyz
+ 1: 
+ 2: xyz
+
+/(?<=ab(?i)x(?-i)y|(?i)z|b)ZZ/I
+Capturing subpattern count = 0
+Max lookbehind = 4
+No options
+First char = 'Z'
+Need char = 'Z'
+    abxyZZ
+ 0: ZZ
+    abXyZZ
+ 0: ZZ
+    ZZZ
+ 0: ZZ
+    zZZ
+ 0: ZZ
+    bZZ
+ 0: ZZ
+    BZZ
+ 0: ZZ
+    *** Failers
+No match
+    ZZ
+No match
+    abXYZZ
+No match
+    zzz
+No match
+    bzz
+No match
+
+/(?
+ 3: f
+ 1G a (1)
+ 2G  (0)
+ 3G f (1)
+get substring 4 failed -7
+ 0L adef
+ 1L a
+ 2L 
+ 3L f
+    bcdef\G1\G2\G3\G4\L
+ 0: bcdef
+ 1: bc
+ 2: bc
+ 3: f
+ 1G bc (2)
+ 2G bc (2)
+ 3G f (1)
+get substring 4 failed -7
+ 0L bcdef
+ 1L bc
+ 2L bc
+ 3L f
+    adefghijk\C0
+ 0: adef
+ 1: a
+ 2: 
+ 3: f
+ 0C adef (4)
+
+/^abc\00def/I
+Capturing subpattern count = 0
+Options: anchored
+No first char
+No need char
+    abc\00def\L\C0
+ 0: abc\x00def
+ 0C abc\x00def (7)
+ 0L abc
+
+/word ((?:[a-zA-Z0-9]+ )((?:[a-zA-Z0-9]+ )((?:[a-zA-Z0-9]+ )((?:[a-zA-Z0-9]+
+)((?:[a-zA-Z0-9]+ )((?:[a-zA-Z0-9]+ )((?:[a-zA-Z0-9]+ )((?:[a-zA-Z0-9]+
+)?)?)?)?)?)?)?)?)?otherword/I
+Capturing subpattern count = 8
+Contains explicit CR or LF match
+No options
+First char = 'w'
+Need char = 'd'
+
+/.*X/IDZ
+------------------------------------------------------------------
+        Bra
+        Any*
+        X
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+No options
+First char at start or follows newline
+Need char = 'X'
+
+/.*X/IDZs
+------------------------------------------------------------------
+        Bra
+        AllAny*
+        X
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: anchored dotall
+No first char
+Need char = 'X'
+
+/(.*X|^B)/IDZ
+------------------------------------------------------------------
+        Bra
+        CBra 1
+        Any*
+        X
+        Alt
+        ^
+        B
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 1
+No options
+First char at start or follows newline
+No need char
+
+/(.*X|^B)/IDZs
+------------------------------------------------------------------
+        Bra
+        CBra 1
+        AllAny*
+        X
+        Alt
+        ^
+        B
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 1
+Options: anchored dotall
+No first char
+No need char
+
+/(?s)(.*X|^B)/IDZ
+------------------------------------------------------------------
+        Bra
+        CBra 1
+        AllAny*
+        X
+        Alt
+        ^
+        B
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 1
+Options: anchored dotall
+No first char
+No need char
+
+/(?s:.*X|^B)/IDZ
+------------------------------------------------------------------
+        Bra
+        Bra
+        AllAny*
+        X
+        Alt
+        ^
+        B
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: anchored
+No first char
+No need char
+
+/\Biss\B/I+
+Capturing subpattern count = 0
+Max lookbehind = 1
+No options
+First char = 'i'
+Need char = 's'
+    Mississippi
+ 0: iss
+ 0+ issippi
+
+/iss/IG+
+Capturing subpattern count = 0
+No options
+First char = 'i'
+Need char = 's'
+    Mississippi
+ 0: iss
+ 0+ issippi
+ 0: iss
+ 0+ ippi
+
+/\Biss\B/IG+
+Capturing subpattern count = 0
+Max lookbehind = 1
+No options
+First char = 'i'
+Need char = 's'
+    Mississippi
+ 0: iss
+ 0+ issippi
+
+/\Biss\B/Ig+
+Capturing subpattern count = 0
+Max lookbehind = 1
+No options
+First char = 'i'
+Need char = 's'
+    Mississippi
+ 0: iss
+ 0+ issippi
+ 0: iss
+ 0+ ippi
+    *** Failers
+No match
+    Mississippi\A
+No match
+
+/(?<=[Ms])iss/Ig+
+Capturing subpattern count = 0
+Max lookbehind = 1
+No options
+First char = 'i'
+Need char = 's'
+    Mississippi
+ 0: iss
+ 0+ issippi
+ 0: iss
+ 0+ ippi
+
+/(?<=[Ms])iss/IG+
+Capturing subpattern count = 0
+Max lookbehind = 1
+No options
+First char = 'i'
+Need char = 's'
+    Mississippi
+ 0: iss
+ 0+ issippi
+
+/^iss/Ig+
+Capturing subpattern count = 0
+Options: anchored
+No first char
+No need char
+    ississippi
+ 0: iss
+ 0+ issippi
+
+/.*iss/Ig+
+Capturing subpattern count = 0
+No options
+First char at start or follows newline
+Need char = 's'
+    abciss\nxyzisspqr
+ 0: abciss
+ 0+ \x0axyzisspqr
+ 0: xyziss
+ 0+ pqr
+
+/.i./I+g
+Capturing subpattern count = 0
+No options
+No first char
+Need char = 'i'
+    Mississippi
+ 0: Mis
+ 0+ sissippi
+ 0: sis
+ 0+ sippi
+ 0: sip
+ 0+ pi
+    Mississippi\A
+ 0: Mis
+ 0+ sissippi
+ 0: sis
+ 0+ sippi
+ 0: sip
+ 0+ pi
+    Missouri river
+ 0: Mis
+ 0+ souri river
+ 0: ri 
+ 0+ river
+ 0: riv
+ 0+ er
+    Missouri river\A
+ 0: Mis
+ 0+ souri river
+
+/^.is/I+g
+Capturing subpattern count = 0
+Options: anchored
+No first char
+No need char
+    Mississippi
+ 0: Mis
+ 0+ sissippi
+
+/^ab\n/Ig+
+Capturing subpattern count = 0
+Contains explicit CR or LF match
+Options: anchored
+No first char
+No need char
+    ab\nab\ncd
+ 0: ab\x0a
+ 0+ ab\x0acd
+
+/^ab\n/Img+
+Capturing subpattern count = 0
+Contains explicit CR or LF match
+Options: multiline
+First char at start or follows newline
+Need char = \x0a
+    ab\nab\ncd
+ 0: ab\x0a
+ 0+ ab\x0acd
+ 0: ab\x0a
+ 0+ cd
+
+/abc/I
+Capturing subpattern count = 0
+No options
+First char = 'a'
+Need char = 'c'
+
+/abc|bac/I
+Capturing subpattern count = 0
+No options
+No first char
+Need char = 'c'
+
+/(abc|bac)/I
+Capturing subpattern count = 1
+No options
+No first char
+Need char = 'c'
+
+/(abc|(c|dc))/I
+Capturing subpattern count = 2
+No options
+No first char
+Need char = 'c'
+
+/(abc|(d|de)c)/I
+Capturing subpattern count = 2
+No options
+No first char
+Need char = 'c'
+
+/a*/I
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+
+/a+/I
+Capturing subpattern count = 0
+No options
+First char = 'a'
+No need char
+
+/(baa|a+)/I
+Capturing subpattern count = 1
+No options
+No first char
+Need char = 'a'
+
+/a{0,3}/I
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+
+/baa{3,}/I
+Capturing subpattern count = 0
+No options
+First char = 'b'
+Need char = 'a'
+
+/"([^\\"]+|\\.)*"/I
+Capturing subpattern count = 1
+No options
+First char = '"'
+Need char = '"'
+
+/(abc|ab[cd])/I
+Capturing subpattern count = 1
+No options
+First char = 'a'
+No need char
+
+/(a|.)/I
+Capturing subpattern count = 1
+No options
+No first char
+No need char
+
+/a|ba|\w/I
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+
+/abc(?=pqr)/I
+Capturing subpattern count = 0
+No options
+First char = 'a'
+Need char = 'r'
+
+/...(?<=abc)/I
+Capturing subpattern count = 0
+Max lookbehind = 3
+No options
+No first char
+No need char
+
+/abc(?!pqr)/I
+Capturing subpattern count = 0
+No options
+First char = 'a'
+Need char = 'c'
+
+/ab./I
+Capturing subpattern count = 0
+No options
+First char = 'a'
+Need char = 'b'
+
+/ab[xyz]/I
+Capturing subpattern count = 0
+No options
+First char = 'a'
+Need char = 'b'
+
+/abc*/I
+Capturing subpattern count = 0
+No options
+First char = 'a'
+Need char = 'b'
+
+/ab.c*/I
+Capturing subpattern count = 0
+No options
+First char = 'a'
+Need char = 'b'
+
+/a.c*/I
+Capturing subpattern count = 0
+No options
+First char = 'a'
+No need char
+
+/.c*/I
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+
+/ac*/I
+Capturing subpattern count = 0
+No options
+First char = 'a'
+No need char
+
+/(a.c*|b.c*)/I
+Capturing subpattern count = 1
+No options
+No first char
+No need char
+
+/a.c*|aba/I
+Capturing subpattern count = 0
+No options
+First char = 'a'
+No need char
+
+/.+a/I
+Capturing subpattern count = 0
+No options
+No first char
+Need char = 'a'
+
+/(?=abcda)a.*/I
+Capturing subpattern count = 0
+No options
+First char = 'a'
+Need char = 'a'
+
+/(?=a)a.*/I
+Capturing subpattern count = 0
+No options
+First char = 'a'
+No need char
+
+/a(b)*/I
+Capturing subpattern count = 1
+No options
+First char = 'a'
+No need char
+
+/a\d*/I
+Capturing subpattern count = 0
+No options
+First char = 'a'
+No need char
+
+/ab\d*/I
+Capturing subpattern count = 0
+No options
+First char = 'a'
+Need char = 'b'
+
+/a(\d)*/I
+Capturing subpattern count = 1
+No options
+First char = 'a'
+No need char
+
+/abcde{0,0}/I
+Capturing subpattern count = 0
+No options
+First char = 'a'
+Need char = 'd'
+
+/ab\d+/I
+Capturing subpattern count = 0
+No options
+First char = 'a'
+Need char = 'b'
+
+/a(?(1)b)(.)/I
+Capturing subpattern count = 1
+No options
+First char = 'a'
+No need char
+
+/a(?(1)bag|big)(.)/I
+Capturing subpattern count = 1
+No options
+First char = 'a'
+Need char = 'g'
+
+/a(?(1)bag|big)*(.)/I
+Capturing subpattern count = 1
+No options
+First char = 'a'
+No need char
+
+/a(?(1)bag|big)+(.)/I
+Capturing subpattern count = 1
+No options
+First char = 'a'
+Need char = 'g'
+
+/a(?(1)b..|b..)(.)/I
+Capturing subpattern count = 1
+No options
+First char = 'a'
+Need char = 'b'
+
+/ab\d{0}e/I
+Capturing subpattern count = 0
+No options
+First char = 'a'
+Need char = 'e'
+
+/a?b?/I
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+    a
+ 0: a
+    b
+ 0: b
+    ab
+ 0: ab
+    \
+ 0: 
+    *** Failers
+ 0: 
+    \N
+No match
+
+/|-/I
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+    abcd
+ 0: 
+    -abc
+ 0: 
+    \Nab-c
+ 0: -
+    *** Failers
+ 0: 
+    \Nabc
+No match
+
+/^.?abcd/IS
+Capturing subpattern count = 0
+Options: anchored
+No first char
+Need char = 'd'
+Subject length lower bound = 4
+No set of starting bytes
+
+/\(             # ( at start
+  (?:           # Non-capturing bracket
+  (?>[^()]+)    # Either a sequence of non-brackets (no backtracking)
+  |             # Or
+  (?R)          # Recurse - i.e. nested bracketed string
+  )*            # Zero or more contents
+  \)            # Closing )
+  /Ix
+Capturing subpattern count = 0
+Options: extended
+First char = '('
+Need char = ')'
+    (abcd)
+ 0: (abcd)
+    (abcd)xyz
+ 0: (abcd)
+    xyz(abcd)
+ 0: (abcd)
+    (ab(xy)cd)pqr
+ 0: (ab(xy)cd)
+    (ab(xycd)pqr
+ 0: (xycd)
+    () abc ()
+ 0: ()
+    12(abcde(fsh)xyz(foo(bar))lmno)89
+ 0: (abcde(fsh)xyz(foo(bar))lmno)
+    *** Failers
+No match
+    abcd
+No match
+    abcd)
+No match
+    (abcd
+No match
+
+/\(  ( (?>[^()]+) | (?R) )* \) /Ixg
+Capturing subpattern count = 1
+Options: extended
+First char = '('
+Need char = ')'
+    (ab(xy)cd)pqr
+ 0: (ab(xy)cd)
+ 1: cd
+    1(abcd)(x(y)z)pqr
+ 0: (abcd)
+ 1: abcd
+ 0: (x(y)z)
+ 1: z
+
+/\(  (?: (?>[^()]+) | (?R) ) \) /Ix
+Capturing subpattern count = 0
+Options: extended
+First char = '('
+Need char = ')'
+    (abcd)
+ 0: (abcd)
+    (ab(xy)cd)
+ 0: (xy)
+    (a(b(c)d)e)
+ 0: (c)
+    ((ab))
+ 0: ((ab))
+    *** Failers
+No match
+    ()
+No match
+
+/\(  (?: (?>[^()]+) | (?R) )? \) /Ix
+Capturing subpattern count = 0
+Options: extended
+First char = '('
+Need char = ')'
+    ()
+ 0: ()
+    12(abcde(fsh)xyz(foo(bar))lmno)89
+ 0: (fsh)
+
+/\(  ( (?>[^()]+) | (?R) )* \) /Ix
+Capturing subpattern count = 1
+Options: extended
+First char = '('
+Need char = ')'
+    (ab(xy)cd)
+ 0: (ab(xy)cd)
+ 1: cd
+
+/\( ( ( (?>[^()]+) | (?R) )* ) \) /Ix
+Capturing subpattern count = 2
+Options: extended
+First char = '('
+Need char = ')'
+    (ab(xy)cd)
+ 0: (ab(xy)cd)
+ 1: ab(xy)cd
+ 2: cd
+
+/\( (123)? ( ( (?>[^()]+) | (?R) )* ) \) /Ix
+Capturing subpattern count = 3
+Options: extended
+First char = '('
+Need char = ')'
+    (ab(xy)cd)
+ 0: (ab(xy)cd)
+ 1: 
+ 2: ab(xy)cd
+ 3: cd
+    (123ab(xy)cd)
+ 0: (123ab(xy)cd)
+ 1: 123
+ 2: ab(xy)cd
+ 3: cd
+
+/\( ( (123)? ( (?>[^()]+) | (?R) )* ) \) /Ix
+Capturing subpattern count = 3
+Options: extended
+First char = '('
+Need char = ')'
+    (ab(xy)cd)
+ 0: (ab(xy)cd)
+ 1: ab(xy)cd
+ 2: 
+ 3: cd
+    (123ab(xy)cd)
+ 0: (123ab(xy)cd)
+ 1: 123ab(xy)cd
+ 2: 123
+ 3: cd
+
+/\( (((((((((( ( (?>[^()]+) | (?R) )* )))))))))) \) /Ix
+Capturing subpattern count = 11
+Options: extended
+First char = '('
+Need char = ')'
+    (ab(xy)cd)
+ 0: (ab(xy)cd)
+ 1: ab(xy)cd
+ 2: ab(xy)cd
+ 3: ab(xy)cd
+ 4: ab(xy)cd
+ 5: ab(xy)cd
+ 6: ab(xy)cd
+ 7: ab(xy)cd
+ 8: ab(xy)cd
+ 9: ab(xy)cd
+10: ab(xy)cd
+11: cd
+
+/\( ( ( (?>[^()<>]+) | ((?>[^()]+)) | (?R) )* ) \) /Ix
+Capturing subpattern count = 3
+Options: extended
+First char = '('
+Need char = ')'
+    (abcd(xyz
qrs)123)
+ 0: (abcd(xyz
qrs)123)
+ 1: abcd(xyz
qrs)123
+ 2: 123
+
+/\( ( ( (?>[^()]+) | ((?R)) )* ) \) /Ix
+Capturing subpattern count = 3
+Options: extended
+First char = '('
+Need char = ')'
+    (ab(cd)ef)
+ 0: (ab(cd)ef)
+ 1: ab(cd)ef
+ 2: ef
+ 3: (cd)
+    (ab(cd(ef)gh)ij)
+ 0: (ab(cd(ef)gh)ij)
+ 1: ab(cd(ef)gh)ij
+ 2: ij
+ 3: (cd(ef)gh)
+
+/^[[:alnum:]]/DZ
+------------------------------------------------------------------
+        Bra
+        ^
+        [0-9A-Za-z]
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: anchored
+No first char
+No need char
+
+/^[[:^alnum:]]/DZ
+------------------------------------------------------------------
+        Bra
+        ^
+        [\x00-/:-@[-`{-\xff] (neg)
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: anchored
+No first char
+No need char
+
+/^[[:alpha:]]/DZ
+------------------------------------------------------------------
+        Bra
+        ^
+        [A-Za-z]
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: anchored
+No first char
+No need char
+
+/^[[:^alpha:]]/DZ
+------------------------------------------------------------------
+        Bra
+        ^
+        [\x00-@[-`{-\xff] (neg)
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: anchored
+No first char
+No need char
+
+/[_[:alpha:]]/IS
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: A B C D E F G H I J K L M N O P Q R S T U V W X Y Z 
+  _ a b c d e f g h i j k l m n o p q r s t u v w x y z 
+
+/^[[:ascii:]]/DZ
+------------------------------------------------------------------
+        Bra
+        ^
+        [\x00-\x7f]
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: anchored
+No first char
+No need char
+
+/^[[:^ascii:]]/DZ
+------------------------------------------------------------------
+        Bra
+        ^
+        [\x80-\xff] (neg)
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: anchored
+No first char
+No need char
+
+/^[[:blank:]]/DZ
+------------------------------------------------------------------
+        Bra
+        ^
+        [\x09 ]
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: anchored
+No first char
+No need char
+
+/^[[:^blank:]]/DZ
+------------------------------------------------------------------
+        Bra
+        ^
+        [\x00-\x08\x0a-\x1f!-\xff] (neg)
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: anchored
+No first char
+No need char
+
+/[\n\x0b\x0c\x0d[:blank:]]/IS
+Capturing subpattern count = 0
+Contains explicit CR or LF match
+No options
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: \x09 \x0a \x0b \x0c \x0d \x20 
+
+/^[[:cntrl:]]/DZ
+------------------------------------------------------------------
+        Bra
+        ^
+        [\x00-\x1f\x7f]
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: anchored
+No first char
+No need char
+
+/^[[:digit:]]/DZ
+------------------------------------------------------------------
+        Bra
+        ^
+        [0-9]
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: anchored
+No first char
+No need char
+
+/^[[:graph:]]/DZ
+------------------------------------------------------------------
+        Bra
+        ^
+        [!-~]
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: anchored
+No first char
+No need char
+
+/^[[:lower:]]/DZ
+------------------------------------------------------------------
+        Bra
+        ^
+        [a-z]
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: anchored
+No first char
+No need char
+
+/^[[:print:]]/DZ
+------------------------------------------------------------------
+        Bra
+        ^
+        [ -~]
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: anchored
+No first char
+No need char
+
+/^[[:punct:]]/DZ
+------------------------------------------------------------------
+        Bra
+        ^
+        [!-/:-@[-`{-~]
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: anchored
+No first char
+No need char
+
+/^[[:space:]]/DZ
+------------------------------------------------------------------
+        Bra
+        ^
+        [\x09-\x0d ]
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: anchored
+No first char
+No need char
+
+/^[[:upper:]]/DZ
+------------------------------------------------------------------
+        Bra
+        ^
+        [A-Z]
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: anchored
+No first char
+No need char
+
+/^[[:xdigit:]]/DZ
+------------------------------------------------------------------
+        Bra
+        ^
+        [0-9A-Fa-f]
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: anchored
+No first char
+No need char
+
+/^[[:word:]]/DZ
+------------------------------------------------------------------
+        Bra
+        ^
+        [0-9A-Z_a-z]
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: anchored
+No first char
+No need char
+
+/^[[:^cntrl:]]/DZ
+------------------------------------------------------------------
+        Bra
+        ^
+        [ -~\x80-\xff] (neg)
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: anchored
+No first char
+No need char
+
+/^[12[:^digit:]]/DZ
+------------------------------------------------------------------
+        Bra
+        ^
+        [\x00-/12:-\xff] (neg)
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: anchored
+No first char
+No need char
+
+/^[[:^blank:]]/DZ
+------------------------------------------------------------------
+        Bra
+        ^
+        [\x00-\x08\x0a-\x1f!-\xff] (neg)
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: anchored
+No first char
+No need char
+
+/[01[:alpha:]%]/DZ
+------------------------------------------------------------------
+        Bra
+        [%01A-Za-z]
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+
+/[[.ch.]]/I
+Failed: POSIX collating elements are not supported at offset 1
+
+/[[=ch=]]/I
+Failed: POSIX collating elements are not supported at offset 1
+
+/[[:rhubarb:]]/I
+Failed: unknown POSIX class name at offset 3
+
+/[[:upper:]]/Ii
+Capturing subpattern count = 0
+Options: caseless
+No first char
+No need char
+    A
+ 0: A
+    a
+ 0: a
+
+/[[:lower:]]/Ii
+Capturing subpattern count = 0
+Options: caseless
+No first char
+No need char
+    A
+ 0: A
+    a
+ 0: a
+
+/((?-i)[[:lower:]])[[:lower:]]/Ii
+Capturing subpattern count = 1
+Options: caseless
+No first char
+No need char
+    ab
+ 0: ab
+ 1: a
+    aB
+ 0: aB
+ 1: a
+    *** Failers
+ 0: ai
+ 1: a
+    Ab
+No match
+    AB
+No match
+
+/[\200-\110]/I
+Failed: range out of order in character class at offset 9
+
+/^(?(0)f|b)oo/I
+Failed: invalid condition (?(0) at offset 6
+
+/This one's here because of the large output vector needed/I
+Capturing subpattern count = 0
+No options
+First char = 'T'
+Need char = 'd'
+
+/(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\d+(?:\s|$))(\w+)\s+(\270)/I
+Capturing subpattern count = 271
+Max back reference = 270
+No options
+No first char
+No need char
+    \O900 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 ABC ABC
+ 0: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 ABC ABC
+ 1: 1 
+ 2: 2 
+ 3: 3 
+ 4: 4 
+ 5: 5 
+ 6: 6 
+ 7: 7 
+ 8: 8 
+ 9: 9 
+10: 10 
+11: 11 
+12: 12 
+13: 13 
+14: 14 
+15: 15 
+16: 16 
+17: 17 
+18: 18 
+19: 19 
+20: 20 
+21: 21 
+22: 22 
+23: 23 
+24: 24 
+25: 25 
+26: 26 
+27: 27 
+28: 28 
+29: 29 
+30: 30 
+31: 31 
+32: 32 
+33: 33 
+34: 34 
+35: 35 
+36: 36 
+37: 37 
+38: 38 
+39: 39 
+40: 40 
+41: 41 
+42: 42 
+43: 43 
+44: 44 
+45: 45 
+46: 46 
+47: 47 
+48: 48 
+49: 49 
+50: 50 
+51: 51 
+52: 52 
+53: 53 
+54: 54 
+55: 55 
+56: 56 
+57: 57 
+58: 58 
+59: 59 
+60: 60 
+61: 61 
+62: 62 
+63: 63 
+64: 64 
+65: 65 
+66: 66 
+67: 67 
+68: 68 
+69: 69 
+70: 70 
+71: 71 
+72: 72 
+73: 73 
+74: 74 
+75: 75 
+76: 76 
+77: 77 
+78: 78 
+79: 79 
+80: 80 
+81: 81 
+82: 82 
+83: 83 
+84: 84 
+85: 85 
+86: 86 
+87: 87 
+88: 88 
+89: 89 
+90: 90 
+91: 91 
+92: 92 
+93: 93 
+94: 94 
+95: 95 
+96: 96 
+97: 97 
+98: 98 
+99: 99 
+100: 100 
+101: 101 
+102: 102 
+103: 103 
+104: 104 
+105: 105 
+106: 106 
+107: 107 
+108: 108 
+109: 109 
+110: 110 
+111: 111 
+112: 112 
+113: 113 
+114: 114 
+115: 115 
+116: 116 
+117: 117 
+118: 118 
+119: 119 
+120: 120 
+121: 121 
+122: 122 
+123: 123 
+124: 124 
+125: 125 
+126: 126 
+127: 127 
+128: 128 
+129: 129 
+130: 130 
+131: 131 
+132: 132 
+133: 133 
+134: 134 
+135: 135 
+136: 136 
+137: 137 
+138: 138 
+139: 139 
+140: 140 
+141: 141 
+142: 142 
+143: 143 
+144: 144 
+145: 145 
+146: 146 
+147: 147 
+148: 148 
+149: 149 
+150: 150 
+151: 151 
+152: 152 
+153: 153 
+154: 154 
+155: 155 
+156: 156 
+157: 157 
+158: 158 
+159: 159 
+160: 160 
+161: 161 
+162: 162 
+163: 163 
+164: 164 
+165: 165 
+166: 166 
+167: 167 
+168: 168 
+169: 169 
+170: 170 
+171: 171 
+172: 172 
+173: 173 
+174: 174 
+175: 175 
+176: 176 
+177: 177 
+178: 178 
+179: 179 
+180: 180 
+181: 181 
+182: 182 
+183: 183 
+184: 184 
+185: 185 
+186: 186 
+187: 187 
+188: 188 
+189: 189 
+190: 190 
+191: 191 
+192: 192 
+193: 193 
+194: 194 
+195: 195 
+196: 196 
+197: 197 
+198: 198 
+199: 199 
+200: 200 
+201: 201 
+202: 202 
+203: 203 
+204: 204 
+205: 205 
+206: 206 
+207: 207 
+208: 208 
+209: 209 
+210: 210 
+211: 211 
+212: 212 
+213: 213 
+214: 214 
+215: 215 
+216: 216 
+217: 217 
+218: 218 
+219: 219 
+220: 220 
+221: 221 
+222: 222 
+223: 223 
+224: 224 
+225: 225 
+226: 226 
+227: 227 
+228: 228 
+229: 229 
+230: 230 
+231: 231 
+232: 232 
+233: 233 
+234: 234 
+235: 235 
+236: 236 
+237: 237 
+238: 238 
+239: 239 
+240: 240 
+241: 241 
+242: 242 
+243: 243 
+244: 244 
+245: 245 
+246: 246 
+247: 247 
+248: 248 
+249: 249 
+250: 250 
+251: 251 
+252: 252 
+253: 253 
+254: 254 
+255: 255 
+256: 256 
+257: 257 
+258: 258 
+259: 259 
+260: 260 
+261: 261 
+262: 262 
+263: 263 
+264: 264 
+265: 265 
+266: 266 
+267: 267 
+268: 268 
+269: 269 
+270: ABC
+271: ABC
+
+/This one's here because Perl does this differently and PCRE can't at present/I
+Capturing subpattern count = 0
+No options
+First char = 'T'
+Need char = 't'
+
+/(main(O)?)+/I
+Capturing subpattern count = 2
+No options
+First char = 'm'
+Need char = 'n'
+    mainmain
+ 0: mainmain
+ 1: main
+    mainOmain
+ 0: mainOmain
+ 1: main
+ 2: O
+
+/These are all cases where Perl does it differently (nested captures)/I
+Capturing subpattern count = 1
+No options
+First char = 'T'
+Need char = 's'
+
+/^(a(b)?)+$/I
+Capturing subpattern count = 2
+Options: anchored
+No first char
+No need char
+    aba
+ 0: aba
+ 1: a
+ 2: b
+
+/^(aa(bb)?)+$/I
+Capturing subpattern count = 2
+Options: anchored
+No first char
+No need char
+    aabbaa
+ 0: aabbaa
+ 1: aa
+ 2: bb
+
+/^(aa|aa(bb))+$/I
+Capturing subpattern count = 2
+Options: anchored
+No first char
+No need char
+    aabbaa
+ 0: aabbaa
+ 1: aa
+ 2: bb
+
+/^(aa(bb)??)+$/I
+Capturing subpattern count = 2
+Options: anchored
+No first char
+No need char
+    aabbaa
+ 0: aabbaa
+ 1: aa
+ 2: bb
+
+/^(?:aa(bb)?)+$/I
+Capturing subpattern count = 1
+Options: anchored
+No first char
+No need char
+    aabbaa
+ 0: aabbaa
+ 1: bb
+
+/^(aa(b(b))?)+$/I
+Capturing subpattern count = 3
+Options: anchored
+No first char
+No need char
+    aabbaa
+ 0: aabbaa
+ 1: aa
+ 2: bb
+ 3: b
+
+/^(?:aa(b(b))?)+$/I
+Capturing subpattern count = 2
+Options: anchored
+No first char
+No need char
+    aabbaa
+ 0: aabbaa
+ 1: bb
+ 2: b
+
+/^(?:aa(b(?:b))?)+$/I
+Capturing subpattern count = 1
+Options: anchored
+No first char
+No need char
+    aabbaa
+ 0: aabbaa
+ 1: bb
+
+/^(?:aa(bb(?:b))?)+$/I
+Capturing subpattern count = 1
+Options: anchored
+No first char
+No need char
+    aabbbaa
+ 0: aabbbaa
+ 1: bbb
+
+/^(?:aa(b(?:bb))?)+$/I
+Capturing subpattern count = 1
+Options: anchored
+No first char
+No need char
+    aabbbaa
+ 0: aabbbaa
+ 1: bbb
+
+/^(?:aa(?:b(b))?)+$/I
+Capturing subpattern count = 1
+Options: anchored
+No first char
+No need char
+    aabbaa
+ 0: aabbaa
+ 1: b
+
+/^(?:aa(?:b(bb))?)+$/I
+Capturing subpattern count = 1
+Options: anchored
+No first char
+No need char
+    aabbbaa
+ 0: aabbbaa
+ 1: bb
+
+/^(aa(b(bb))?)+$/I
+Capturing subpattern count = 3
+Options: anchored
+No first char
+No need char
+    aabbbaa
+ 0: aabbbaa
+ 1: aa
+ 2: bbb
+ 3: bb
+
+/^(aa(bb(bb))?)+$/I
+Capturing subpattern count = 3
+Options: anchored
+No first char
+No need char
+    aabbbbaa
+ 0: aabbbbaa
+ 1: aa
+ 2: bbbb
+ 3: bb
+
+/--------------------------------------------------------------------/I
+Capturing subpattern count = 0
+No options
+First char = '-'
+Need char = '-'
+
+/#/IxDZ
+------------------------------------------------------------------
+        Bra
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: extended
+No first char
+No need char
+
+/a#/IxDZ
+------------------------------------------------------------------
+        Bra
+        a
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: extended
+First char = 'a'
+No need char
+
+/[\s]/DZ
+------------------------------------------------------------------
+        Bra
+        [\x09\x0a\x0c\x0d ]
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+
+/[\S]/DZ
+------------------------------------------------------------------
+        Bra
+        [\x00-\x08\x0b\x0e-\x1f!-\xff] (neg)
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+
+/a(?i)b/DZ
+------------------------------------------------------------------
+        Bra
+        a
+     /i b
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+No options
+First char = 'a'
+Need char = 'b' (caseless)
+    ab
+ 0: ab
+    aB
+ 0: aB
+    *** Failers
+No match
+    AB
+No match
+
+/(a(?i)b)/DZ
+------------------------------------------------------------------
+        Bra
+        CBra 1
+        a
+     /i b
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 1
+No options
+First char = 'a'
+Need char = 'b' (caseless)
+    ab
+ 0: ab
+ 1: ab
+    aB
+ 0: aB
+ 1: aB
+    *** Failers
+No match
+    AB
+No match
+
+/   (?i)abc/IxDZ
+------------------------------------------------------------------
+        Bra
+     /i abc
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: caseless extended
+First char = 'a' (caseless)
+Need char = 'c' (caseless)
+
+/#this is a comment
+  (?i)abc/IxDZ
+------------------------------------------------------------------
+        Bra
+     /i abc
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: caseless extended
+First char = 'a' (caseless)
+Need char = 'c' (caseless)
+
+/123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890/DZ
+------------------------------------------------------------------
+        Bra
+        123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+No options
+First char = '1'
+Need char = '0'
+
+/\Q123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890/DZ
+------------------------------------------------------------------
+        Bra
+        123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+No options
+First char = '1'
+Need char = '0'
+
+/\Q\E/DZ
+------------------------------------------------------------------
+        Bra
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+    \
+ 0: 
+
+/\Q\Ex/DZ
+------------------------------------------------------------------
+        Bra
+        x
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+No options
+First char = 'x'
+No need char
+
+/ \Q\E/DZ
+------------------------------------------------------------------
+        Bra
+         
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+No options
+First char = ' '
+No need char
+
+/a\Q\E/DZ
+------------------------------------------------------------------
+        Bra
+        a
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+No options
+First char = 'a'
+No need char
+  abc
+ 0: a
+  bca
+ 0: a
+  bac
+ 0: a
+
+/a\Q\Eb/DZ
+------------------------------------------------------------------
+        Bra
+        ab
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+No options
+First char = 'a'
+Need char = 'b'
+  abc
+ 0: ab
+
+/\Q\Eabc/DZ
+------------------------------------------------------------------
+        Bra
+        abc
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+No options
+First char = 'a'
+Need char = 'c'
+
+/x*+\w/DZ
+------------------------------------------------------------------
+        Bra
+        x*+
+        \w
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+    *** Failers
+ 0: F
+    xxxxx
+No match
+
+/x?+/DZ
+------------------------------------------------------------------
+        Bra
+        x?+
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+
+/x++/DZ
+------------------------------------------------------------------
+        Bra
+        x++
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+No options
+First char = 'x'
+No need char
+
+/x{1,3}+/DZ
+------------------------------------------------------------------
+        Bra
+        Once
+        x
+        x{0,2}
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+No options
+First char = 'x'
+No need char
+
+/(x)*+/DZ
+------------------------------------------------------------------
+        Bra
+        Braposzero
+        CBraPos 1
+        x
+        KetRpos
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 1
+No options
+No first char
+No need char
+
+/^(\w++|\s++)*$/I
+Capturing subpattern count = 1
+Options: anchored
+No first char
+No need char
+    now is the time for all good men to come to the aid of the party
+ 0: now is the time for all good men to come to the aid of the party
+ 1: party
+    *** Failers
+No match
+    this is not a line with only words and spaces!
+No match
+
+/(\d++)(\w)/I
+Capturing subpattern count = 2
+No options
+No first char
+No need char
+    12345a
+ 0: 12345a
+ 1: 12345
+ 2: a
+    *** Failers
+No match
+    12345+
+No match
+
+/a++b/I
+Capturing subpattern count = 0
+No options
+First char = 'a'
+Need char = 'b'
+    aaab
+ 0: aaab
+
+/(a++b)/I
+Capturing subpattern count = 1
+No options
+First char = 'a'
+Need char = 'b'
+    aaab
+ 0: aaab
+ 1: aaab
+
+/(a++)b/I
+Capturing subpattern count = 1
+No options
+First char = 'a'
+Need char = 'b'
+    aaab
+ 0: aaab
+ 1: aaa
+
+/([^()]++|\([^()]*\))+/I
+Capturing subpattern count = 1
+No options
+No first char
+No need char
+    ((abc(ade)ufh()()x
+ 0: abc(ade)ufh()()x
+ 1: x
+
+/\(([^()]++|\([^()]+\))+\)/I
+Capturing subpattern count = 1
+No options
+First char = '('
+Need char = ')'
+    (abc)
+ 0: (abc)
+ 1: abc
+    (abc(def)xyz)
+ 0: (abc(def)xyz)
+ 1: xyz
+    *** Failers
+No match
+    ((()aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+No match
+
+/(abc){1,3}+/DZ
+------------------------------------------------------------------
+        Bra
+        Once
+        CBra 1
+        abc
+        Ket
+        Brazero
+        Bra
+        CBra 1
+        abc
+        Ket
+        Brazero
+        CBra 1
+        abc
+        Ket
+        Ket
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 1
+No options
+First char = 'a'
+Need char = 'c'
+
+/a+?+/I
+Failed: nothing to repeat at offset 3
+
+/a{2,3}?+b/I
+Failed: nothing to repeat at offset 7
+
+/(?U)a+?+/I
+Failed: nothing to repeat at offset 7
+
+/a{2,3}?+b/IU
+Failed: nothing to repeat at offset 7
+
+/x(?U)a++b/DZ
+------------------------------------------------------------------
+        Bra
+        x
+        a++
+        b
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+No options
+First char = 'x'
+Need char = 'b'
+    xaaaab
+ 0: xaaaab
+
+/(?U)xa++b/DZ
+------------------------------------------------------------------
+        Bra
+        x
+        a++
+        b
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: ungreedy
+First char = 'x'
+Need char = 'b'
+    xaaaab
+ 0: xaaaab
+
+/^((a+)(?U)([ab]+)(?-U)([bc]+)(\w*))/DZ
+------------------------------------------------------------------
+        Bra
+        ^
+        CBra 1
+        CBra 2
+        a+
+        Ket
+        CBra 3
+        [ab]+?
+        Ket
+        CBra 4
+        [bc]+
+        Ket
+        CBra 5
+        \w*
+        Ket
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 5
+Options: anchored
+No first char
+No need char
+
+/^x(?U)a+b/DZ
+------------------------------------------------------------------
+        Bra
+        ^
+        x
+        a++
+        b
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: anchored
+No first char
+Need char = 'b'
+
+/^x(?U)(a+)b/DZ
+------------------------------------------------------------------
+        Bra
+        ^
+        x
+        CBra 1
+        a+?
+        Ket
+        b
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 1
+Options: anchored
+No first char
+Need char = 'b'
+
+/[.x.]/I
+Failed: POSIX collating elements are not supported at offset 0
+
+/[=x=]/I
+Failed: POSIX collating elements are not supported at offset 0
+
+/[:x:]/I
+Failed: POSIX named classes are supported only within a class at offset 0
+
+/\l/I
+Failed: PCRE does not support \L, \l, \N{name}, \U, or \u at offset 1
+
+/\L/I
+Failed: PCRE does not support \L, \l, \N{name}, \U, or \u at offset 1
+
+/\N{name}/I
+Failed: PCRE does not support \L, \l, \N{name}, \U, or \u at offset 1
+
+/\u/I
+Failed: PCRE does not support \L, \l, \N{name}, \U, or \u at offset 1
+
+/\U/I
+Failed: PCRE does not support \L, \l, \N{name}, \U, or \u at offset 1
+
+/[/I
+Failed: missing terminating ] for character class at offset 1
+
+/[a-/I
+Failed: missing terminating ] for character class at offset 3
+
+/[[:space:]/I
+Failed: missing terminating ] for character class at offset 10
+
+/[\s]/IDZ
+------------------------------------------------------------------
+        Bra
+        [\x09\x0a\x0c\x0d ]
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+
+/[[:space:]]/IDZ
+------------------------------------------------------------------
+        Bra
+        [\x09-\x0d ]
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+
+/[[:space:]abcde]/IDZ
+------------------------------------------------------------------
+        Bra
+        [\x09-\x0d a-e]
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+
+/< (?: (?(R) \d++  | [^<>]*+) | (?R)) * >/Ix
+Capturing subpattern count = 0
+Options: extended
+First char = '<'
+Need char = '>'
+    <>
+ 0: <>
+    
+ 0: 
+     hij>
+ 0:  hij>
+     hij>
+ 0: 
+    def>
+ 0: def>
+    
+ 0: <>
+    *** Failers
+No match
+    iV\;\?aPhFB\<\*vW\@QW\@sO9\}cfZA\-i\'w\%hKd6gt1UJP\,15_\#QY\$M\^Mss_U\/\]\&LK9\[5vQub\^w\[KDD\qmj\;2\}YWFdYx\.Ap\]hjCPTP\(n28k\+3\;o\&WXqs\/gOXdr\$\:r\'do0\;b4c\(f_Gr\=\"\\4\)\[01T7ajQJvL\$W\~mL_sS\/4h\:x\*\[ZN\=KLs\&L5zX\/\/\>it\,o\:aU\(\;Z\>pW\&T7oP\'2K\^E\:x9\'c\[\%z\-\,64JQ5AeH_G\#KijUKghQw\^\\vea3a\?kka_G\$8\#\`\*kynsxzBLru\'\]k_\[7FrVx\}\^\=\$blx\>s\-N\%j\;D\*aZDnsw\:YKZ\%Q\.Kne9\#hP\?\+b3\(SOvL\,\^\;\&u5\@\?5C5Bhb\=m\-vEh_L15Jl\]U\)0RP6\{q\%L\^_z5E\'Dw6X\b|IDZ
+------------------------------------------------------------------
+        Bra
+        8J$WE<.rX+ix[d1b!H#?vV0vrK:ZH1=2M>iV;?aPhFB<*vW@QW@sO9}cfZA-i'w%hKd6gt1UJP,15_#QY$M^Mss_U/]&LK9[5vQub^w[KDDqmj;2}YWFdYx.Ap]hjCPTP(n28k+3;o&WXqs/gOXdr$:r'do0;b4c(f_Gr="\4)[01T7ajQJvL$W~mL_sS/4h:x*[ZN=KLs&L5zX//>it,o:aU(;Z>pW&T7oP'2K^E:x9'c[%z-,64JQ5AeH_G#KijUKghQw^\vea3a?kka_G$8#`*kynsxzBLru']k_[7FrVx}^=$blx>s-N%j;D*aZDnsw:YKZ%Q.Kne9#hP?+b3(SOvL,^;&u5@?5C5Bhb=m-vEh_L15Jl]U)0RP6{q%L^_z5E'Dw6X
+        \b
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Max lookbehind = 1
+No options
+First char = '8'
+Need char = 'X'
+
+|\$\<\.X\+ix\[d1b\!H\#\?vV0vrK\:ZH1\=2M\>iV\;\?aPhFB\<\*vW\@QW\@sO9\}cfZA\-i\'w\%hKd6gt1UJP\,15_\#QY\$M\^Mss_U\/\]\&LK9\[5vQub\^w\[KDD\qmj\;2\}YWFdYx\.Ap\]hjCPTP\(n28k\+3\;o\&WXqs\/gOXdr\$\:r\'do0\;b4c\(f_Gr\=\"\\4\)\[01T7ajQJvL\$W\~mL_sS\/4h\:x\*\[ZN\=KLs\&L5zX\/\/\>it\,o\:aU\(\;Z\>pW\&T7oP\'2K\^E\:x9\'c\[\%z\-\,64JQ5AeH_G\#KijUKghQw\^\\vea3a\?kka_G\$8\#\`\*kynsxzBLru\'\]k_\[7FrVx\}\^\=\$blx\>s\-N\%j\;D\*aZDnsw\:YKZ\%Q\.Kne9\#hP\?\+b3\(SOvL\,\^\;\&u5\@\?5C5Bhb\=m\-vEh_L15Jl\]U\)0RP6\{q\%L\^_z5E\'Dw6X\b|IDZ
+------------------------------------------------------------------
+        Bra
+        $<.X+ix[d1b!H#?vV0vrK:ZH1=2M>iV;?aPhFB<*vW@QW@sO9}cfZA-i'w%hKd6gt1UJP,15_#QY$M^Mss_U/]&LK9[5vQub^w[KDDqmj;2}YWFdYx.Ap]hjCPTP(n28k+3;o&WXqs/gOXdr$:r'do0;b4c(f_Gr="\4)[01T7ajQJvL$W~mL_sS/4h:x*[ZN=KLs&L5zX//>it,o:aU(;Z>pW&T7oP'2K^E:x9'c[%z-,64JQ5AeH_G#KijUKghQw^\vea3a?kka_G$8#`*kynsxzBLru']k_[7FrVx}^=$blx>s-N%j;D*aZDnsw:YKZ%Q.Kne9#hP?+b3(SOvL,^;&u5@?5C5Bhb=m-vEh_L15Jl]U)0RP6{q%L^_z5E'Dw6X
+        \b
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Max lookbehind = 1
+No options
+First char = '$'
+Need char = 'X'
+
+/(.*)\d+\1/I
+Capturing subpattern count = 1
+Max back reference = 1
+No options
+No first char
+No need char
+
+/(.*)\d+/I
+Capturing subpattern count = 1
+No options
+First char at start or follows newline
+No need char
+
+/(.*)\d+\1/Is
+Capturing subpattern count = 1
+Max back reference = 1
+Options: dotall
+No first char
+No need char
+
+/(.*)\d+/Is
+Capturing subpattern count = 1
+Options: anchored dotall
+No first char
+No need char
+
+/(.*(xyz))\d+\2/I
+Capturing subpattern count = 2
+Max back reference = 2
+No options
+First char at start or follows newline
+Need char = 'z'
+
+/((.*))\d+\1/I
+Capturing subpattern count = 2
+Max back reference = 1
+No options
+No first char
+No need char
+    abc123bc
+ 0: bc123bc
+ 1: bc
+ 2: bc
+
+/a[b]/I
+Capturing subpattern count = 0
+No options
+First char = 'a'
+Need char = 'b'
+
+/(?=a).*/I
+Capturing subpattern count = 0
+No options
+First char = 'a'
+No need char
+
+/(?=abc).xyz/IiI
+Capturing subpattern count = 0
+Options: caseless
+First char = 'a' (caseless)
+Need char = 'z' (caseless)
+
+/(?=abc)(?i).xyz/I
+Capturing subpattern count = 0
+No options
+First char = 'a'
+Need char = 'z' (caseless)
+
+/(?=a)(?=b)/I
+Capturing subpattern count = 0
+No options
+First char = 'a'
+No need char
+
+/(?=.)a/I
+Capturing subpattern count = 0
+No options
+First char = 'a'
+No need char
+
+/((?=abcda)a)/I
+Capturing subpattern count = 1
+No options
+First char = 'a'
+Need char = 'a'
+
+/((?=abcda)ab)/I
+Capturing subpattern count = 1
+No options
+First char = 'a'
+Need char = 'b'
+
+/()a/I
+Capturing subpattern count = 1
+No options
+No first char
+Need char = 'a'
+
+/(?(1)ab|ac)(.)/I
+Capturing subpattern count = 1
+No options
+First char = 'a'
+No need char
+
+/(?(1)abz|acz)(.)/I
+Capturing subpattern count = 1
+No options
+First char = 'a'
+Need char = 'z'
+
+/(?(1)abz)(.)/I
+Capturing subpattern count = 1
+No options
+No first char
+No need char
+
+/(?(1)abz)(1)23/I
+Capturing subpattern count = 1
+No options
+No first char
+Need char = '3'
+
+/(a)+/I
+Capturing subpattern count = 1
+No options
+First char = 'a'
+No need char
+
+/(a){2,3}/I
+Capturing subpattern count = 1
+No options
+First char = 'a'
+Need char = 'a'
+
+/(a)*/I
+Capturing subpattern count = 1
+No options
+No first char
+No need char
+
+/[a]/I
+Capturing subpattern count = 0
+No options
+First char = 'a'
+No need char
+
+/[ab]/I
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+
+/[ab]/IS
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: a b 
+
+/[^a]/I
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+
+/\d456/I
+Capturing subpattern count = 0
+No options
+No first char
+Need char = '6'
+
+/\d456/IS
+Capturing subpattern count = 0
+No options
+No first char
+Need char = '6'
+Subject length lower bound = 4
+Starting byte set: 0 1 2 3 4 5 6 7 8 9 
+
+/a^b/I
+Capturing subpattern count = 0
+No options
+First char = 'a'
+Need char = 'b'
+
+/^a/Im
+Capturing subpattern count = 0
+Options: multiline
+First char at start or follows newline
+Need char = 'a'
+  abcde
+ 0: a
+  xy\nabc
+ 0: a
+  *** Failers
+No match
+  xyabc
+No match
+
+/c|abc/I
+Capturing subpattern count = 0
+No options
+No first char
+Need char = 'c'
+
+/(?i)[ab]/IS
+Capturing subpattern count = 0
+Options: caseless
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: A B a b 
+
+/[ab](?i)cd/IS
+Capturing subpattern count = 0
+No options
+No first char
+Need char = 'd' (caseless)
+Subject length lower bound = 3
+Starting byte set: a b 
+
+/abc(?C)def/I
+Capturing subpattern count = 0
+No options
+First char = 'a'
+Need char = 'f'
+    abcdef
+--->abcdef
+  0 ^  ^       d
+ 0: abcdef
+    1234abcdef
+--->1234abcdef
+  0     ^  ^       d
+ 0: abcdef
+    *** Failers
+No match
+    abcxyz
+No match
+    abcxyzf
+--->abcxyzf
+  0 ^  ^        d
+No match
+
+/abc(?C)de(?C1)f/I
+Capturing subpattern count = 0
+No options
+First char = 'a'
+Need char = 'f'
+    123abcdef
+--->123abcdef
+  0    ^  ^       d
+  1    ^    ^     f
+ 0: abcdef
+
+/(?C1)\dabc(?C2)def/IS
+Capturing subpattern count = 0
+No options
+No first char
+Need char = 'f'
+Subject length lower bound = 7
+Starting byte set: 0 1 2 3 4 5 6 7 8 9 
+    1234abcdef
+--->1234abcdef
+  1 ^              \d
+  1  ^             \d
+  1   ^            \d
+  1    ^           \d
+  2    ^   ^       d
+ 0: 4abcdef
+    *** Failers
+No match
+    abcdef
+No match
+
+/(?C1)\dabc(?C2)def/ISS
+Capturing subpattern count = 0
+No options
+No first char
+Need char = 'f'
+    1234abcdef
+--->1234abcdef
+  1 ^              \d
+  1  ^             \d
+  1   ^            \d
+  1    ^           \d
+  2    ^   ^       d
+ 0: 4abcdef
+    *** Failers
+No match
+    abcdef
+--->abcdef
+  1 ^          \d
+  1  ^         \d
+  1   ^        \d
+  1    ^       \d
+  1     ^      \d
+  1      ^     \d
+No match
+
+/(?C255)ab/I
+Capturing subpattern count = 0
+No options
+First char = 'a'
+Need char = 'b'
+
+/(?C256)ab/I
+Failed: number after (?C is > 255 at offset 6
+
+/(?Cab)xx/I
+Failed: closing ) for (?C expected at offset 3
+
+/(?C12vr)x/I
+Failed: closing ) for (?C expected at offset 5
+
+/abc(?C)def/I
+Capturing subpattern count = 0
+No options
+First char = 'a'
+Need char = 'f'
+    *** Failers
+No match
+    \x83\x0\x61bcdef
+--->\x83\x00abcdef
+  0         ^  ^       d
+ 0: abcdef
+
+/(abc)(?C)de(?C1)f/I
+Capturing subpattern count = 1
+No options
+First char = 'a'
+Need char = 'f'
+    123abcdef
+--->123abcdef
+  0    ^  ^       d
+  1    ^    ^     f
+ 0: abcdef
+ 1: abc
+    123abcdef\C+
+Callout 0: last capture = 1
+ 0: 
+ 1: abc
+--->123abcdef
+       ^  ^       d
+Callout 1: last capture = 1
+ 0: 
+ 1: abc
+--->123abcdef
+       ^    ^     f
+ 0: abcdef
+ 1: abc
+    123abcdef\C-
+ 0: abcdef
+ 1: abc
+    *** Failers
+No match
+    123abcdef\C!1
+--->123abcdef
+  0    ^  ^       d
+  1    ^    ^     f
+No match
+
+/(?C0)(abc(?C1))*/I
+Capturing subpattern count = 1
+No options
+No first char
+No need char
+    abcabcabc
+--->abcabcabc
+  0 ^             (abc(?C1))*
+  1 ^  ^          )
+  1 ^     ^       )
+  1 ^        ^    )
+ 0: abcabcabc
+ 1: abc
+    abcabc\C!1!3
+--->abcabc
+  0 ^          (abc(?C1))*
+  1 ^  ^       )
+  1 ^     ^    )
+ 0: abcabc
+ 1: abc
+    *** Failers
+--->*** Failers
+  0 ^               (abc(?C1))*
+ 0: 
+    abcabcabc\C!1!3
+--->abcabcabc
+  0 ^             (abc(?C1))*
+  1 ^  ^          )
+  1 ^     ^       )
+  1 ^        ^    )
+ 0: abcabc
+ 1: abc
+
+/(\d{3}(?C))*/I
+Capturing subpattern count = 1
+No options
+No first char
+No need char
+    123\C+
+Callout 0: last capture = -1
+ 0: 
+--->123
+    ^  ^    )
+ 0: 123
+ 1: 123
+    123456\C+
+Callout 0: last capture = -1
+ 0: 
+--->123456
+    ^  ^       )
+Callout 0: last capture = 1
+ 0: 
+ 1: 123
+--->123456
+    ^     ^    )
+ 0: 123456
+ 1: 456
+    123456789\C+
+Callout 0: last capture = -1
+ 0: 
+--->123456789
+    ^  ^          )
+Callout 0: last capture = 1
+ 0: 
+ 1: 123
+--->123456789
+    ^     ^       )
+Callout 0: last capture = 1
+ 0: 
+ 1: 456
+--->123456789
+    ^        ^    )
+ 0: 123456789
+ 1: 789
+
+/((xyz)(?C)p|(?C1)xyzabc)/I
+Capturing subpattern count = 2
+No options
+First char = 'x'
+No need char
+    xyzabc\C+
+Callout 0: last capture = 2
+ 0: 
+ 1: 
+ 2: xyz
+--->xyzabc
+    ^  ^       p
+Callout 1: last capture = -1
+ 0: 
+--->xyzabc
+    ^          x
+ 0: xyzabc
+ 1: xyzabc
+
+/(X)((xyz)(?C)p|(?C1)xyzabc)/I
+Capturing subpattern count = 3
+No options
+First char = 'X'
+Need char = 'x'
+    Xxyzabc\C+
+Callout 0: last capture = 3
+ 0: 
+ 1: X
+ 2: 
+ 3: xyz
+--->Xxyzabc
+    ^   ^       p
+Callout 1: last capture = 1
+ 0: 
+ 1: X
+--->Xxyzabc
+    ^^          x
+ 0: Xxyzabc
+ 1: X
+ 2: xyzabc
+
+/(?=(abc))(?C)abcdef/I
+Capturing subpattern count = 1
+No options
+First char = 'a'
+Need char = 'f'
+    abcdef\C+
+Callout 0: last capture = 1
+ 0: 
+ 1: abc
+--->abcdef
+    ^          a
+ 0: abcdef
+ 1: abc
+
+/(?!(abc)(?C1)d)(?C2)abcxyz/I
+Capturing subpattern count = 1
+No options
+First char = 'a'
+Need char = 'z'
+    abcxyz\C+
+Callout 1: last capture = 1
+ 0: 
+ 1: abc
+--->abcxyz
+    ^  ^       d
+Callout 2: last capture = -1
+ 0: 
+--->abcxyz
+    ^          a
+ 0: abcxyz
+
+/(?<=(abc)(?C))xyz/I
+Capturing subpattern count = 1
+Max lookbehind = 3
+No options
+First char = 'x'
+Need char = 'z'
+   abcxyz\C+
+Callout 0: last capture = 1
+ 0: 
+ 1: abc
+--->abcxyz
+       ^       )
+ 0: xyz
+ 1: abc
+
+/a(b+)(c*)(?C1)/I
+Capturing subpattern count = 2
+No options
+First char = 'a'
+Need char = 'b'
+    abbbbbccc\C*1
+--->abbbbbccc
+  1 ^        ^    
+Callout data = 1
+  1 ^       ^     
+Callout data = 1
+  1 ^      ^      
+Callout data = 1
+  1 ^     ^       
+Callout data = 1
+  1 ^    ^        
+Callout data = 1
+  1 ^   ^         
+Callout data = 1
+  1 ^  ^          
+Callout data = 1
+  1 ^ ^           
+Callout data = 1
+No match
+
+/a(b+?)(c*?)(?C1)/I
+Capturing subpattern count = 2
+No options
+First char = 'a'
+Need char = 'b'
+    abbbbbccc\C*1
+--->abbbbbccc
+  1 ^ ^           
+Callout data = 1
+  1 ^  ^          
+Callout data = 1
+  1 ^   ^         
+Callout data = 1
+  1 ^    ^        
+Callout data = 1
+  1 ^     ^       
+Callout data = 1
+  1 ^      ^      
+Callout data = 1
+  1 ^       ^     
+Callout data = 1
+  1 ^        ^    
+Callout data = 1
+No match
+
+/(?C)abc/I
+Capturing subpattern count = 0
+No options
+First char = 'a'
+Need char = 'c'
+
+/(?C)^abc/I
+Capturing subpattern count = 0
+Options: anchored
+No first char
+No need char
+
+/(?C)a|b/IS
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: a b 
+
+/(?R)/I
+Failed: recursive call could loop indefinitely at offset 3
+
+/(a|(?R))/I
+Failed: recursive call could loop indefinitely at offset 6
+
+/(ab|(bc|(de|(?R))))/I
+Failed: recursive call could loop indefinitely at offset 15
+
+/x(ab|(bc|(de|(?R))))/I
+Capturing subpattern count = 3
+No options
+First char = 'x'
+No need char
+    xab
+ 0: xab
+ 1: ab
+    xbc
+ 0: xbc
+ 1: bc
+ 2: bc
+    xde
+ 0: xde
+ 1: de
+ 2: de
+ 3: de
+    xxab
+ 0: xxab
+ 1: xab
+ 2: xab
+ 3: xab
+    xxxab
+ 0: xxxab
+ 1: xxab
+ 2: xxab
+ 3: xxab
+    *** Failers
+No match
+    xyab
+No match
+
+/(ab|(bc|(de|(?1))))/I
+Failed: recursive call could loop indefinitely at offset 15
+
+/x(ab|(bc|(de|(?1)x)x)x)/I
+Failed: recursive call could loop indefinitely at offset 16
+
+/^([^()]|\((?1)*\))*$/I
+Capturing subpattern count = 1
+Options: anchored
+No first char
+No need char
+    abc
+ 0: abc
+ 1: c
+    a(b)c
+ 0: a(b)c
+ 1: c
+    a(b(c))d
+ 0: a(b(c))d
+ 1: d
+    *** Failers)
+No match
+    a(b(c)d
+No match
+
+/^>abc>([^()]|\((?1)*\))*abc>123abc>123abc>1(2)3abc>1(2)3abc>(1(2)3)abc>(1(2)3)]*+) | (?2)) * >))/Ix
+Capturing subpattern count = 2
+Options: extended
+First char = '<'
+Need char = '>'
+    <>
+ 0: <>
+ 1: <>
+ 2: <>
+    
+ 0: 
+ 1: 
+ 2: 
+     hij>
+ 0:  hij>
+ 1:  hij>
+ 2:  hij>
+     hij>
+ 0: 
+ 1: 
+ 2: 
+    def>
+ 0: def>
+ 1: def>
+ 2: def>
+    
+ 0: <>
+ 1: <>
+ 2: <>
+    *** Failers
+No match
+    b|c)d(?Pe)/DZ
+------------------------------------------------------------------
+        Bra
+        a
+        CBra 1
+        b
+        Alt
+        c
+        Ket
+        d
+        CBra 2
+        e
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 2
+Named capturing subpatterns:
+  longername2   2
+  name1         1
+No options
+First char = 'a'
+Need char = 'e'
+    abde
+ 0: abde
+ 1: b
+ 2: e
+    acde
+ 0: acde
+ 1: c
+ 2: e
+
+/(?:a(?Pc(?Pd)))(?Pa)/DZ
+------------------------------------------------------------------
+        Bra
+        Bra
+        a
+        CBra 1
+        c
+        CBra 2
+        d
+        Ket
+        Ket
+        Ket
+        CBra 3
+        a
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 3
+Named capturing subpatterns:
+  a   3
+  c   1
+  d   2
+No options
+First char = 'a'
+Need char = 'a'
+
+/(?Pa)...(?P=a)bbb(?P>a)d/DZ
+------------------------------------------------------------------
+        Bra
+        CBra 1
+        a
+        Ket
+        Any
+        Any
+        Any
+        \1
+        bbb
+        Recurse
+        d
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 1
+Max back reference = 1
+Named capturing subpatterns:
+  a   1
+No options
+First char = 'a'
+Need char = 'd'
+
+/^\W*(?:(?P(?P.)\W*(?P>one)\W*(?P=two)|)|(?P(?P.)\W*(?P>three)\W*(?P=four)|\W*.\W*))\W*$/Ii
+Capturing subpattern count = 4
+Max back reference = 4
+Named capturing subpatterns:
+  four    4
+  one     1
+  three   3
+  two     2
+Options: anchored caseless
+No first char
+No need char
+    1221
+ 0: 1221
+ 1: 1221
+ 2: 1
+    Satan, oscillate my metallic sonatas!
+ 0: Satan, oscillate my metallic sonatas!
+ 1: 
+ 2: 
+ 3: Satan, oscillate my metallic sonatas
+ 4: S
+    A man, a plan, a canal: Panama!
+ 0: A man, a plan, a canal: Panama!
+ 1: 
+ 2: 
+ 3: A man, a plan, a canal: Panama
+ 4: A
+    Able was I ere I saw Elba.
+ 0: Able was I ere I saw Elba.
+ 1: 
+ 2: 
+ 3: Able was I ere I saw Elba
+ 4: A
+    *** Failers
+No match
+    The quick brown fox
+No match
+
+/((?(R)a|b))\1(?1)?/I
+Capturing subpattern count = 1
+Max back reference = 1
+No options
+No first char
+No need char
+  bb
+ 0: bb
+ 1: b
+  bbaa
+ 0: bba
+ 1: b
+
+/(.*)a/Is
+Capturing subpattern count = 1
+Options: anchored dotall
+No first char
+Need char = 'a'
+
+/(.*)a\1/Is
+Capturing subpattern count = 1
+Max back reference = 1
+Options: dotall
+No first char
+Need char = 'a'
+
+/(.*)a(b)\2/Is
+Capturing subpattern count = 2
+Max back reference = 2
+Options: anchored dotall
+No first char
+Need char = 'b'
+
+/((.*)a|(.*)b)z/Is
+Capturing subpattern count = 3
+Options: anchored dotall
+No first char
+Need char = 'z'
+
+/((.*)a|(.*)b)z\1/Is
+Capturing subpattern count = 3
+Max back reference = 1
+Options: dotall
+No first char
+Need char = 'z'
+
+/((.*)a|(.*)b)z\2/Is
+Capturing subpattern count = 3
+Max back reference = 2
+Options: dotall
+No first char
+Need char = 'z'
+
+/((.*)a|(.*)b)z\3/Is
+Capturing subpattern count = 3
+Max back reference = 3
+Options: dotall
+No first char
+Need char = 'z'
+
+/((.*)a|^(.*)b)z\3/Is
+Capturing subpattern count = 3
+Max back reference = 3
+Options: anchored dotall
+No first char
+Need char = 'z'
+
+/(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)a/Is
+Capturing subpattern count = 31
+Options: anchored dotall
+No first char
+No need char
+
+/(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)a\31/Is
+Capturing subpattern count = 31
+Max back reference = 31
+Options: dotall
+No first char
+No need char
+
+/(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)|(.*)a\32/Is
+Capturing subpattern count = 32
+Max back reference = 32
+Options: dotall
+No first char
+No need char
+
+/(a)(bc)/INDZ
+------------------------------------------------------------------
+        Bra
+        Bra
+        a
+        Ket
+        Bra
+        bc
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: no_auto_capture
+First char = 'a'
+Need char = 'c'
+  abc
+ 0: abc
+
+/(?Pa)(bc)/INDZ
+------------------------------------------------------------------
+        Bra
+        CBra 1
+        a
+        Ket
+        Bra
+        bc
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 1
+Named capturing subpatterns:
+  one   1
+Options: no_auto_capture
+First char = 'a'
+Need char = 'c'
+  abc
+ 0: abc
+ 1: a
+
+/(a)(?Pbc)/INDZ
+------------------------------------------------------------------
+        Bra
+        Bra
+        a
+        Ket
+        CBra 1
+        bc
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 1
+Named capturing subpatterns:
+  named   1
+Options: no_auto_capture
+First char = 'a'
+Need char = 'c'
+
+/(a+)*zz/I
+Capturing subpattern count = 1
+No options
+No first char
+Need char = 'z'
+  aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaazzbbbbbb\M
+Minimum match() limit = 8
+Minimum match() recursion limit = 6
+ 0: aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaazz
+ 1: aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+  aaaaaaaaaaaaaz\M
+Minimum match() limit = 32768
+Minimum match() recursion limit = 29
+No match
+
+/(aaa(?C1)bbb|ab)/I
+Capturing subpattern count = 1
+No options
+First char = 'a'
+Need char = 'b'
+   aaabbb
+--->aaabbb
+  1 ^  ^       b
+ 0: aaabbb
+ 1: aaabbb
+   aaabbb\C*0
+--->aaabbb
+  1 ^  ^       b
+ 0: aaabbb
+ 1: aaabbb
+   aaabbb\C*1
+--->aaabbb
+  1 ^  ^       b
+Callout data = 1
+ 0: ab
+ 1: ab
+   aaabbb\C*-1
+--->aaabbb
+  1 ^  ^       b
+Callout data = -1
+No match
+
+/ab(?Pcd)ef(?Pgh)/I
+Capturing subpattern count = 2
+Named capturing subpatterns:
+  one   1
+  two   2
+No options
+First char = 'a'
+Need char = 'h'
+    abcdefgh
+ 0: abcdefgh
+ 1: cd
+ 2: gh
+    abcdefgh\C1\Gtwo
+ 0: abcdefgh
+ 1: cd
+ 2: gh
+ 1C cd (2)
+  G gh (2) two
+    abcdefgh\Cone\Ctwo
+ 0: abcdefgh
+ 1: cd
+ 2: gh
+  C cd (2) one
+  C gh (2) two
+    abcdefgh\Cthree
+no parentheses with name "three"
+ 0: abcdefgh
+ 1: cd
+ 2: gh
+copy substring three failed -7
+
+/(?P)(?P)/DZ
+------------------------------------------------------------------
+        Bra
+        CBra 1
+        Ket
+        CBra 2
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 2
+Named capturing subpatterns:
+  Tes    1
+  Test   2
+No options
+No first char
+No need char
+
+/(?P)(?P)/DZ
+------------------------------------------------------------------
+        Bra
+        CBra 1
+        Ket
+        CBra 2
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 2
+Named capturing subpatterns:
+  Tes    2
+  Test   1
+No options
+No first char
+No need char
+
+/(?Pzz)(?Paa)/I
+Capturing subpattern count = 2
+Named capturing subpatterns:
+  A   2
+  Z   1
+No options
+First char = 'z'
+Need char = 'a'
+    zzaa\CZ
+ 0: zzaa
+ 1: zz
+ 2: aa
+  C zz (2) Z
+    zzaa\CA
+ 0: zzaa
+ 1: zz
+ 2: aa
+  C aa (2) A
+
+/(?Peks)(?Peccs)/I
+Failed: two named subpatterns have the same name at offset 15
+
+/(?Pabc(?Pdef)(?Pxyz))/I
+Failed: two named subpatterns have the same name at offset 30
+
+"\[((?P\d+)(,(?P>elem))*)\]"I
+Capturing subpattern count = 3
+Named capturing subpatterns:
+  elem   2
+No options
+First char = '['
+Need char = ']'
+    [10,20,30,5,5,4,4,2,43,23,4234]
+ 0: [10,20,30,5,5,4,4,2,43,23,4234]
+ 1: 10,20,30,5,5,4,4,2,43,23,4234
+ 2: 10
+ 3: ,4234
+    *** Failers
+No match
+    []
+No match
+
+"\[((?P\d+)(,(?P>elem))*)?\]"I
+Capturing subpattern count = 3
+Named capturing subpatterns:
+  elem   2
+No options
+First char = '['
+Need char = ']'
+    [10,20,30,5,5,4,4,2,43,23,4234]
+ 0: [10,20,30,5,5,4,4,2,43,23,4234]
+ 1: 10,20,30,5,5,4,4,2,43,23,4234
+ 2: 10
+ 3: ,4234
+    []
+ 0: []
+
+/(a(b(?2)c))?/DZ
+------------------------------------------------------------------
+        Bra
+        Brazero
+        CBra 1
+        a
+        CBra 2
+        b
+        Recurse
+        c
+        Ket
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 2
+No options
+No first char
+No need char
+
+/(a(b(?2)c))*/DZ
+------------------------------------------------------------------
+        Bra
+        Brazero
+        CBra 1
+        a
+        CBra 2
+        b
+        Recurse
+        c
+        Ket
+        KetRmax
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 2
+No options
+No first char
+No need char
+
+/(a(b(?2)c)){0,2}/DZ
+------------------------------------------------------------------
+        Bra
+        Brazero
+        Bra
+        CBra 1
+        a
+        CBra 2
+        b
+        Recurse
+        c
+        Ket
+        Ket
+        Brazero
+        CBra 1
+        a
+        CBra 2
+        b
+        Recurse
+        c
+        Ket
+        Ket
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 2
+No options
+No first char
+No need char
+
+/[ab]{1}+/DZ
+------------------------------------------------------------------
+        Bra
+        Once
+        [ab]{1,1}
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+
+/((w\/|-|with)*(free|immediate)*.*?shipping\s*[!.-]*)/Ii
+Capturing subpattern count = 3
+Options: caseless
+No first char
+Need char = 'g' (caseless)
+     Baby Bjorn Active Carrier - With free SHIPPING!!
+ 0: Baby Bjorn Active Carrier - With free SHIPPING!!
+ 1: Baby Bjorn Active Carrier - With free SHIPPING!!
+
+/((w\/|-|with)*(free|immediate)*.*?shipping\s*[!.-]*)/IiS
+Capturing subpattern count = 3
+Options: caseless
+No first char
+Need char = 'g' (caseless)
+Subject length lower bound = 8
+No set of starting bytes
+     Baby Bjorn Active Carrier - With free SHIPPING!!
+ 0: Baby Bjorn Active Carrier - With free SHIPPING!!
+ 1: Baby Bjorn Active Carrier - With free SHIPPING!!
+
+/a*.*b/ISDZ
+------------------------------------------------------------------
+        Bra
+        a*
+        Any*
+        b
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+No options
+No first char
+Need char = 'b'
+Subject length lower bound = 1
+No set of starting bytes
+
+/(a|b)*.?c/ISDZ
+------------------------------------------------------------------
+        Bra
+        Brazero
+        CBra 1
+        a
+        Alt
+        b
+        KetRmax
+        Any?
+        c
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 1
+No options
+No first char
+Need char = 'c'
+Subject length lower bound = 1
+No set of starting bytes
+
+/abc(?C255)de(?C)f/DZ
+------------------------------------------------------------------
+        Bra
+        abc
+        Callout 255 10 1
+        de
+        Callout 0 16 1
+        f
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+No options
+First char = 'a'
+Need char = 'f'
+
+/abcde/ICDZ
+------------------------------------------------------------------
+        Bra
+        Callout 255 0 1
+        a
+        Callout 255 1 1
+        b
+        Callout 255 2 1
+        c
+        Callout 255 3 1
+        d
+        Callout 255 4 1
+        e
+        Callout 255 5 0
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options:
+First char = 'a'
+Need char = 'e'
+  abcde
+--->abcde
+ +0 ^         a
+ +1 ^^        b
+ +2 ^ ^       c
+ +3 ^  ^      d
+ +4 ^   ^     e
+ +5 ^    ^    
+ 0: abcde
+  abcdfe
+--->abcdfe
+ +0 ^          a
+ +1 ^^         b
+ +2 ^ ^        c
+ +3 ^  ^       d
+ +4 ^   ^      e
+No match
+
+/a*b/ICDZS
+------------------------------------------------------------------
+        Bra
+        Callout 255 0 2
+        a*+
+        Callout 255 2 1
+        b
+        Callout 255 3 0
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options:
+No first char
+Need char = 'b'
+Subject length lower bound = 1
+Starting byte set: a b 
+  ab
+--->ab
+ +0 ^      a*
+ +2 ^^     b
+ +3 ^ ^    
+ 0: ab
+  aaaab
+--->aaaab
+ +0 ^         a*
+ +2 ^   ^     b
+ +3 ^    ^    
+ 0: aaaab
+  aaaacb
+--->aaaacb
+ +0 ^          a*
+ +2 ^   ^      b
+ +0  ^         a*
+ +2  ^  ^      b
+ +0   ^        a*
+ +2   ^ ^      b
+ +0    ^       a*
+ +2    ^^      b
+ +0      ^     a*
+ +2      ^     b
+ +3      ^^    
+ 0: b
+
+/a*b/ICDZSS
+------------------------------------------------------------------
+        Bra
+        Callout 255 0 2
+        a*+
+        Callout 255 2 1
+        b
+        Callout 255 3 0
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options:
+No first char
+Need char = 'b'
+  ab
+--->ab
+ +0 ^      a*
+ +2 ^^     b
+ +3 ^ ^    
+ 0: ab
+  aaaab
+--->aaaab
+ +0 ^         a*
+ +2 ^   ^     b
+ +3 ^    ^    
+ 0: aaaab
+  aaaacb
+--->aaaacb
+ +0 ^          a*
+ +2 ^   ^      b
+ +0  ^         a*
+ +2  ^  ^      b
+ +0   ^        a*
+ +2   ^ ^      b
+ +0    ^       a*
+ +2    ^^      b
+ +0     ^      a*
+ +2     ^      b
+ +0      ^     a*
+ +2      ^     b
+ +3      ^^    
+ 0: b
+
+/a+b/ICDZ
+------------------------------------------------------------------
+        Bra
+        Callout 255 0 2
+        a++
+        Callout 255 2 1
+        b
+        Callout 255 3 0
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options:
+First char = 'a'
+Need char = 'b'
+  ab
+--->ab
+ +0 ^      a+
+ +2 ^^     b
+ +3 ^ ^    
+ 0: ab
+  aaaab
+--->aaaab
+ +0 ^         a+
+ +2 ^   ^     b
+ +3 ^    ^    
+ 0: aaaab
+  aaaacb
+--->aaaacb
+ +0 ^          a+
+ +2 ^   ^      b
+ +0  ^         a+
+ +2  ^  ^      b
+ +0   ^        a+
+ +2   ^ ^      b
+ +0    ^       a+
+ +2    ^^      b
+No match
+
+/(abc|def)x/ICDZS
+------------------------------------------------------------------
+        Bra
+        Callout 255 0 9
+        CBra 1
+        Callout 255 1 1
+        a
+        Callout 255 2 1
+        b
+        Callout 255 3 1
+        c
+        Callout 255 4 0
+        Alt
+        Callout 255 5 1
+        d
+        Callout 255 6 1
+        e
+        Callout 255 7 1
+        f
+        Callout 255 8 0
+        Ket
+        Callout 255 9 1
+        x
+        Callout 255 10 0
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 1
+Options:
+No first char
+Need char = 'x'
+Subject length lower bound = 4
+Starting byte set: a d 
+  abcx
+--->abcx
+ +0 ^        (abc|def)
+ +1 ^        a
+ +2 ^^       b
+ +3 ^ ^      c
+ +4 ^  ^     |
+ +9 ^  ^     x
++10 ^   ^    
+ 0: abcx
+ 1: abc
+  defx
+--->defx
+ +0 ^        (abc|def)
+ +1 ^        a
+ +5 ^        d
+ +6 ^^       e
+ +7 ^ ^      f
+ +8 ^  ^     )
+ +9 ^  ^     x
++10 ^   ^    
+ 0: defx
+ 1: def
+  ** Failers 
+No match
+  abcdefzx
+--->abcdefzx
+ +0 ^            (abc|def)
+ +1 ^            a
+ +2 ^^           b
+ +3 ^ ^          c
+ +4 ^  ^         |
+ +9 ^  ^         x
+ +5 ^            d
+ +0    ^         (abc|def)
+ +1    ^         a
+ +5    ^         d
+ +6    ^^        e
+ +7    ^ ^       f
+ +8    ^  ^      )
+ +9    ^  ^      x
+No match
+
+/(abc|def)x/ICDZSS
+------------------------------------------------------------------
+        Bra
+        Callout 255 0 9
+        CBra 1
+        Callout 255 1 1
+        a
+        Callout 255 2 1
+        b
+        Callout 255 3 1
+        c
+        Callout 255 4 0
+        Alt
+        Callout 255 5 1
+        d
+        Callout 255 6 1
+        e
+        Callout 255 7 1
+        f
+        Callout 255 8 0
+        Ket
+        Callout 255 9 1
+        x
+        Callout 255 10 0
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 1
+Options:
+No first char
+Need char = 'x'
+  abcx
+--->abcx
+ +0 ^        (abc|def)
+ +1 ^        a
+ +2 ^^       b
+ +3 ^ ^      c
+ +4 ^  ^     |
+ +9 ^  ^     x
++10 ^   ^    
+ 0: abcx
+ 1: abc
+  defx
+--->defx
+ +0 ^        (abc|def)
+ +1 ^        a
+ +5 ^        d
+ +6 ^^       e
+ +7 ^ ^      f
+ +8 ^  ^     )
+ +9 ^  ^     x
++10 ^   ^    
+ 0: defx
+ 1: def
+  ** Failers 
+No match
+  abcdefzx
+--->abcdefzx
+ +0 ^            (abc|def)
+ +1 ^            a
+ +2 ^^           b
+ +3 ^ ^          c
+ +4 ^  ^         |
+ +9 ^  ^         x
+ +5 ^            d
+ +0  ^           (abc|def)
+ +1  ^           a
+ +5  ^           d
+ +0   ^          (abc|def)
+ +1   ^          a
+ +5   ^          d
+ +0    ^         (abc|def)
+ +1    ^         a
+ +5    ^         d
+ +6    ^^        e
+ +7    ^ ^       f
+ +8    ^  ^      )
+ +9    ^  ^      x
+ +0     ^        (abc|def)
+ +1     ^        a
+ +5     ^        d
+ +0      ^       (abc|def)
+ +1      ^       a
+ +5      ^       d
+ +0       ^      (abc|def)
+ +1       ^      a
+ +5       ^      d
+ +0        ^     (abc|def)
+ +1        ^     a
+ +5        ^     d
+No match
+
+/(ab|cd){3,4}/IC
+Capturing subpattern count = 1
+Options:
+No first char
+No need char
+  ababab
+--->ababab
+ +0 ^          (ab|cd){3,4}
+ +1 ^          a
+ +2 ^^         b
+ +3 ^ ^        |
+ +1 ^ ^        a
+ +2 ^  ^       b
+ +3 ^   ^      |
+ +1 ^   ^      a
+ +2 ^    ^     b
+ +3 ^     ^    |
+ +1 ^     ^    a
+ +4 ^     ^    c
++12 ^     ^    
+ 0: ababab
+ 1: ab
+  abcdabcd
+--->abcdabcd
+ +0 ^            (ab|cd){3,4}
+ +1 ^            a
+ +2 ^^           b
+ +3 ^ ^          |
+ +1 ^ ^          a
+ +4 ^ ^          c
+ +5 ^  ^         d
+ +6 ^   ^        )
+ +1 ^   ^        a
+ +2 ^    ^       b
+ +3 ^     ^      |
+ +1 ^     ^      a
+ +4 ^     ^      c
+ +5 ^      ^     d
+ +6 ^       ^    )
++12 ^       ^    
+ 0: abcdabcd
+ 1: cd
+  abcdcdcdcdcd
+--->abcdcdcdcdcd
+ +0 ^                (ab|cd){3,4}
+ +1 ^                a
+ +2 ^^               b
+ +3 ^ ^              |
+ +1 ^ ^              a
+ +4 ^ ^              c
+ +5 ^  ^             d
+ +6 ^   ^            )
+ +1 ^   ^            a
+ +4 ^   ^            c
+ +5 ^    ^           d
+ +6 ^     ^          )
+ +1 ^     ^          a
+ +4 ^     ^          c
+ +5 ^      ^         d
+ +6 ^       ^        )
++12 ^       ^        
+ 0: abcdcdcd
+ 1: cd
+
+/([ab]{,4}c|xy)/ICDZS
+------------------------------------------------------------------
+        Bra
+        Callout 255 0 14
+        CBra 1
+        Callout 255 1 4
+        [ab]
+        Callout 255 5 1
+        {
+        Callout 255 6 1
+        ,
+        Callout 255 7 1
+        4
+        Callout 255 8 1
+        }
+        Callout 255 9 1
+        c
+        Callout 255 10 0
+        Alt
+        Callout 255 11 1
+        x
+        Callout 255 12 1
+        y
+        Callout 255 13 0
+        Ket
+        Callout 255 14 0
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 1
+Options:
+No first char
+No need char
+Subject length lower bound = 2
+Starting byte set: a b x 
+    Note: that { does NOT introduce a quantifier
+--->Note: that { does NOT introduce a quantifier
+ +0         ^                                        ([ab]{,4}c|xy)
+ +1         ^                                        [ab]
+ +5         ^^                                       {
++11         ^                                        x
+ +0                                 ^                ([ab]{,4}c|xy)
+ +1                                 ^                [ab]
+ +5                                 ^^               {
++11                                 ^                x
+ +0                                     ^            ([ab]{,4}c|xy)
+ +1                                     ^            [ab]
+ +5                                     ^^           {
++11                                     ^            x
+No match
+
+/([ab]{,4}c|xy)/ICDZSS
+------------------------------------------------------------------
+        Bra
+        Callout 255 0 14
+        CBra 1
+        Callout 255 1 4
+        [ab]
+        Callout 255 5 1
+        {
+        Callout 255 6 1
+        ,
+        Callout 255 7 1
+        4
+        Callout 255 8 1
+        }
+        Callout 255 9 1
+        c
+        Callout 255 10 0
+        Alt
+        Callout 255 11 1
+        x
+        Callout 255 12 1
+        y
+        Callout 255 13 0
+        Ket
+        Callout 255 14 0
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 1
+Options:
+No first char
+No need char
+    Note: that { does NOT introduce a quantifier
+--->Note: that { does NOT introduce a quantifier
+ +0 ^                                                ([ab]{,4}c|xy)
+ +1 ^                                                [ab]
++11 ^                                                x
+ +0  ^                                               ([ab]{,4}c|xy)
+ +1  ^                                               [ab]
++11  ^                                               x
+ +0   ^                                              ([ab]{,4}c|xy)
+ +1   ^                                              [ab]
++11   ^                                              x
+ +0    ^                                             ([ab]{,4}c|xy)
+ +1    ^                                             [ab]
++11    ^                                             x
+ +0     ^                                            ([ab]{,4}c|xy)
+ +1     ^                                            [ab]
++11     ^                                            x
+ +0      ^                                           ([ab]{,4}c|xy)
+ +1      ^                                           [ab]
++11      ^                                           x
+ +0       ^                                          ([ab]{,4}c|xy)
+ +1       ^                                          [ab]
++11       ^                                          x
+ +0        ^                                         ([ab]{,4}c|xy)
+ +1        ^                                         [ab]
++11        ^                                         x
+ +0         ^                                        ([ab]{,4}c|xy)
+ +1         ^                                        [ab]
+ +5         ^^                                       {
++11         ^                                        x
+ +0          ^                                       ([ab]{,4}c|xy)
+ +1          ^                                       [ab]
++11          ^                                       x
+ +0           ^                                      ([ab]{,4}c|xy)
+ +1           ^                                      [ab]
++11           ^                                      x
+ +0            ^                                     ([ab]{,4}c|xy)
+ +1            ^                                     [ab]
++11            ^                                     x
+ +0             ^                                    ([ab]{,4}c|xy)
+ +1             ^                                    [ab]
++11             ^                                    x
+ +0              ^                                   ([ab]{,4}c|xy)
+ +1              ^                                   [ab]
++11              ^                                   x
+ +0               ^                                  ([ab]{,4}c|xy)
+ +1               ^                                  [ab]
++11               ^                                  x
+ +0                ^                                 ([ab]{,4}c|xy)
+ +1                ^                                 [ab]
++11                ^                                 x
+ +0                 ^                                ([ab]{,4}c|xy)
+ +1                 ^                                [ab]
++11                 ^                                x
+ +0                  ^                               ([ab]{,4}c|xy)
+ +1                  ^                               [ab]
++11                  ^                               x
+ +0                   ^                              ([ab]{,4}c|xy)
+ +1                   ^                              [ab]
++11                   ^                              x
+ +0                    ^                             ([ab]{,4}c|xy)
+ +1                    ^                             [ab]
++11                    ^                             x
+ +0                     ^                            ([ab]{,4}c|xy)
+ +1                     ^                            [ab]
++11                     ^                            x
+ +0                      ^                           ([ab]{,4}c|xy)
+ +1                      ^                           [ab]
++11                      ^                           x
+ +0                       ^                          ([ab]{,4}c|xy)
+ +1                       ^                          [ab]
++11                       ^                          x
+ +0                        ^                         ([ab]{,4}c|xy)
+ +1                        ^                         [ab]
++11                        ^                         x
+ +0                         ^                        ([ab]{,4}c|xy)
+ +1                         ^                        [ab]
++11                         ^                        x
+ +0                          ^                       ([ab]{,4}c|xy)
+ +1                          ^                       [ab]
++11                          ^                       x
+ +0                           ^                      ([ab]{,4}c|xy)
+ +1                           ^                      [ab]
++11                           ^                      x
+ +0                            ^                     ([ab]{,4}c|xy)
+ +1                            ^                     [ab]
++11                            ^                     x
+ +0                             ^                    ([ab]{,4}c|xy)
+ +1                             ^                    [ab]
++11                             ^                    x
+ +0                              ^                   ([ab]{,4}c|xy)
+ +1                              ^                   [ab]
++11                              ^                   x
+ +0                               ^                  ([ab]{,4}c|xy)
+ +1                               ^                  [ab]
++11                               ^                  x
+ +0                                ^                 ([ab]{,4}c|xy)
+ +1                                ^                 [ab]
++11                                ^                 x
+ +0                                 ^                ([ab]{,4}c|xy)
+ +1                                 ^                [ab]
+ +5                                 ^^               {
++11                                 ^                x
+ +0                                  ^               ([ab]{,4}c|xy)
+ +1                                  ^               [ab]
++11                                  ^               x
+ +0                                   ^              ([ab]{,4}c|xy)
+ +1                                   ^              [ab]
++11                                   ^              x
+ +0                                    ^             ([ab]{,4}c|xy)
+ +1                                    ^             [ab]
++11                                    ^             x
+ +0                                     ^            ([ab]{,4}c|xy)
+ +1                                     ^            [ab]
+ +5                                     ^^           {
++11                                     ^            x
+ +0                                      ^           ([ab]{,4}c|xy)
+ +1                                      ^           [ab]
++11                                      ^           x
+ +0                                       ^          ([ab]{,4}c|xy)
+ +1                                       ^          [ab]
++11                                       ^          x
+ +0                                        ^         ([ab]{,4}c|xy)
+ +1                                        ^         [ab]
++11                                        ^         x
+ +0                                         ^        ([ab]{,4}c|xy)
+ +1                                         ^        [ab]
++11                                         ^        x
+ +0                                          ^       ([ab]{,4}c|xy)
+ +1                                          ^       [ab]
++11                                          ^       x
+ +0                                           ^      ([ab]{,4}c|xy)
+ +1                                           ^      [ab]
++11                                           ^      x
+ +0                                            ^     ([ab]{,4}c|xy)
+ +1                                            ^     [ab]
++11                                            ^     x
+ +0                                             ^    ([ab]{,4}c|xy)
+ +1                                             ^    [ab]
++11                                             ^    x
+No match
+
+/([ab]{1,4}c|xy){4,5}?123/ICDZ
+------------------------------------------------------------------
+        Bra
+        Callout 255 0 21
+        CBra 1
+        Callout 255 1 9
+        [ab]{1,4}
+        Callout 255 10 1
+        c
+        Callout 255 11 0
+        Alt
+        Callout 255 12 1
+        x
+        Callout 255 13 1
+        y
+        Callout 255 14 0
+        Ket
+        CBra 1
+        Callout 255 1 9
+        [ab]{1,4}
+        Callout 255 10 1
+        c
+        Callout 255 11 0
+        Alt
+        Callout 255 12 1
+        x
+        Callout 255 13 1
+        y
+        Callout 255 14 0
+        Ket
+        CBra 1
+        Callout 255 1 9
+        [ab]{1,4}
+        Callout 255 10 1
+        c
+        Callout 255 11 0
+        Alt
+        Callout 255 12 1
+        x
+        Callout 255 13 1
+        y
+        Callout 255 14 0
+        Ket
+        CBra 1
+        Callout 255 1 9
+        [ab]{1,4}
+        Callout 255 10 1
+        c
+        Callout 255 11 0
+        Alt
+        Callout 255 12 1
+        x
+        Callout 255 13 1
+        y
+        Callout 255 14 0
+        Ket
+        Braminzero
+        CBra 1
+        Callout 255 1 9
+        [ab]{1,4}
+        Callout 255 10 1
+        c
+        Callout 255 11 0
+        Alt
+        Callout 255 12 1
+        x
+        Callout 255 13 1
+        y
+        Callout 255 14 0
+        Ket
+        Callout 255 21 1
+        1
+        Callout 255 22 1
+        2
+        Callout 255 23 1
+        3
+        Callout 255 24 0
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 1
+Options:
+No first char
+Need char = '3'
+    aacaacaacaacaac123
+--->aacaacaacaacaac123
+ +0 ^                      ([ab]{1,4}c|xy){4,5}?
+ +1 ^                      [ab]{1,4}
++10 ^ ^                    c
++11 ^  ^                   |
+ +1 ^  ^                   [ab]{1,4}
++10 ^    ^                 c
++11 ^     ^                |
+ +1 ^     ^                [ab]{1,4}
++10 ^       ^              c
++11 ^        ^             |
+ +1 ^        ^             [ab]{1,4}
++10 ^          ^           c
++11 ^           ^          |
++21 ^           ^          1
+ +1 ^           ^          [ab]{1,4}
++10 ^             ^        c
++11 ^              ^       |
++21 ^              ^       1
++22 ^               ^      2
++23 ^                ^     3
++24 ^                 ^    
+ 0: aacaacaacaacaac123
+ 1: aac
+
+/\b.*/I
+Capturing subpattern count = 0
+Max lookbehind = 1
+No options
+No first char
+No need char
+  ab cd\>1
+ 0:  cd
+
+/\b.*/Is
+Capturing subpattern count = 0
+Max lookbehind = 1
+Options: dotall
+No first char
+No need char
+  ab cd\>1
+ 0:  cd
+
+/(?!.bcd).*/I
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+  Xbcd12345
+ 0: bcd12345
+
+/abcde/I
+Capturing subpattern count = 0
+No options
+First char = 'a'
+Need char = 'e'
+    ab\P
+Partial match: ab
+    abc\P
+Partial match: abc
+    abcd\P
+Partial match: abcd
+    abcde\P
+ 0: abcde
+    the quick brown abc\P
+Partial match: abc
+    ** Failers\P
+No match
+    the quick brown abxyz fox\P
+No match
+
+"^(0?[1-9]|[12][0-9]|3[01])/(0?[1-9]|1[012])/(20)?\d\d$"I
+Capturing subpattern count = 3
+Options: anchored
+No first char
+Need char = '/'
+    13/05/04\P
+ 0: 13/05/04
+ 1: 13
+ 2: 05
+    13/5/2004\P
+ 0: 13/5/2004
+ 1: 13
+ 2: 5
+ 3: 20
+    02/05/09\P
+ 0: 02/05/09
+ 1: 02
+ 2: 05
+    1\P
+Partial match: 1
+    1/2\P
+Partial match: 1/2
+    1/2/0\P
+Partial match: 1/2/0
+    1/2/04\P
+ 0: 1/2/04
+ 1: 1
+ 2: 2
+    0\P
+Partial match: 0
+    02/\P
+Partial match: 02/
+    02/0\P
+Partial match: 02/0
+    02/1\P
+Partial match: 02/1
+    ** Failers\P
+No match
+    \P
+No match
+    123\P
+No match
+    33/4/04\P
+No match
+    3/13/04\P
+No match
+    0/1/2003\P
+No match
+    0/\P
+No match
+    02/0/\P
+No match
+    02/13\P
+No match
+
+/0{0,2}ABC/I
+Capturing subpattern count = 0
+No options
+No first char
+Need char = 'C'
+
+/\d{3,}ABC/I
+Capturing subpattern count = 0
+No options
+No first char
+Need char = 'C'
+
+/\d*ABC/I
+Capturing subpattern count = 0
+No options
+No first char
+Need char = 'C'
+
+/[abc]+DE/I
+Capturing subpattern count = 0
+No options
+No first char
+Need char = 'E'
+
+/[abc]?123/I
+Capturing subpattern count = 0
+No options
+No first char
+Need char = '3'
+    123\P
+ 0: 123
+    a\P
+Partial match: a
+    b\P
+Partial match: b
+    c\P
+Partial match: c
+    c12\P
+Partial match: c12
+    c123\P
+ 0: c123
+
+/^(?:\d){3,5}X/I
+Capturing subpattern count = 0
+Options: anchored
+No first char
+Need char = 'X'
+    1\P
+Partial match: 1
+    123\P
+Partial match: 123
+    123X
+ 0: 123X
+    1234\P
+Partial match: 1234
+    1234X
+ 0: 1234X
+    12345\P
+Partial match: 12345
+    12345X
+ 0: 12345X
+    *** Failers
+No match
+    1X
+No match
+    123456\P
+No match
+
+/abc/IS>testsavedregex
+Capturing subpattern count = 0
+No options
+First char = 'a'
+Need char = 'c'
+Subject length lower bound = 3
+No set of starting bytes
+Compiled pattern written to testsavedregex
+Study data written to testsavedregex
+testsavedregex
+Capturing subpattern count = 0
+No options
+First char = 'a'
+Need char = 'c'
+Compiled pattern written to testsavedregex
+testsavedregex
+Capturing subpattern count = 0
+No options
+First char = 'a'
+Need char = 'c'
+Subject length lower bound = 3
+No set of starting bytes
+Compiled pattern written to testsavedregex
+Study data written to testsavedregex
+testsavedregex
+Capturing subpattern count = 0
+No options
+First char = 'a'
+Need char = 'c'
+Compiled pattern written to testsavedregex
+testsavedregex
+Capturing subpattern count = 1
+No options
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: a b 
+Compiled pattern written to testsavedregex
+Study data written to testsavedregex
+testsavedregex
+Capturing subpattern count = 1
+No options
+No first char
+No need char
+Compiled pattern written to testsavedregex
+testsavedregex
+Capturing subpattern count = 1
+No options
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: a b 
+Compiled pattern written to testsavedregex
+Study data written to testsavedregex
+testsavedregex
+Capturing subpattern count = 1
+No options
+No first char
+No need char
+Compiled pattern written to testsavedregex
+(.)*~smgI
+Capturing subpattern count = 3
+Max back reference = 1
+Options: multiline dotall
+First char = '<'
+Need char = '>'
+    \J1024\n\n\nPartner der LCO\nde\nPartner der LINEAS Consulting\nGmbH\nLINEAS Consulting GmbH Hamburg\nPartnerfirmen\n30 days\nindex,follow\n\nja\n3\nPartner\n\n\nLCO\nLINEAS Consulting\n15.10.2003\n\n\n\n\nDie Partnerfirmen der LINEAS Consulting\nGmbH\n\n\n \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
+ 0: \x0a\x0aPartner der LCO\x0ade\x0aPartner der LINEAS Consulting\x0aGmbH\x0aLINEAS Consulting GmbH Hamburg\x0aPartnerfirmen\x0a30 days\x0aindex,follow\x0a\x0aja\x0a3\x0aPartner\x0a\x0a\x0aLCO\x0aLINEAS Consulting\x0a15.10.2003\x0a\x0a\x0a\x0a\x0aDie Partnerfirmen der LINEAS Consulting\x0aGmbH\x0a\x0a\x0a \x0a\x0a\x0a\x0a\x0a\x0a\x0a\x0a\x0a\x0a\x0a\x0a\x0a\x0a\x0a\x0a\x0a\x0a\x0a\x0a\x0a\x0a
+ 1: seite
+ 2: \x0a
+ 3: seite
+
+/^a/IF
+Capturing subpattern count = 0
+Options: anchored
+No first char
+No need char
+
+/line\nbreak/I
+Capturing subpattern count = 0
+Contains explicit CR or LF match
+No options
+First char = 'l'
+Need char = 'k'
+    this is a line\nbreak
+ 0: line\x0abreak
+    line one\nthis is a line\nbreak in the second line
+ 0: line\x0abreak
+
+/line\nbreak/If
+Capturing subpattern count = 0
+Contains explicit CR or LF match
+Options: firstline
+First char = 'l'
+Need char = 'k'
+    this is a line\nbreak
+ 0: line\x0abreak
+    ** Failers
+No match
+    line one\nthis is a line\nbreak in the second line
+No match
+
+/line\nbreak/Imf
+Capturing subpattern count = 0
+Contains explicit CR or LF match
+Options: multiline firstline
+First char = 'l'
+Need char = 'k'
+    this is a line\nbreak
+ 0: line\x0abreak
+    ** Failers
+No match
+    line one\nthis is a line\nbreak in the second line
+No match
+
+/(?i)(?-i)AbCd/I
+Capturing subpattern count = 0
+No options
+First char = 'A'
+Need char = 'd'
+    AbCd
+ 0: AbCd
+    ** Failers
+No match
+    abcd
+No match
+
+/a{11111111111111111111}/I
+Failed: number too big in {} quantifier at offset 22
+
+/(){64294967295}/I
+Failed: number too big in {} quantifier at offset 14
+
+/(){2,4294967295}/I
+Failed: number too big in {} quantifier at offset 15
+
+"(?i:a)(?i:b)(?i:c)(?i:d)(?i:e)(?i:f)(?i:g)(?i:h)(?i:i)(?i:j)(k)(?i:l)A\1B"I
+Capturing subpattern count = 1
+Max back reference = 1
+No options
+First char = 'a' (caseless)
+Need char = 'B'
+    abcdefghijklAkB
+ 0: abcdefghijklAkB
+ 1: k
+
+"(?Pa)(?Pb)(?Pc)(?Pd)(?Pe)(?Pf)(?Pg)(?Ph)(?Pi)(?Pj)(?Pk)(?Pl)A\11B"I
+Capturing subpattern count = 12
+Max back reference = 11
+Named capturing subpatterns:
+  n0    1
+  n1    2
+  n10  11
+  n11  12
+  n2    3
+  n3    4
+  n4    5
+  n5    6
+  n6    7
+  n7    8
+  n8    9
+  n9   10
+No options
+First char = 'a'
+Need char = 'B'
+    abcdefghijklAkB
+ 0: abcdefghijklAkB
+ 1: a
+ 2: b
+ 3: c
+ 4: d
+ 5: e
+ 6: f
+ 7: g
+ 8: h
+ 9: i
+10: j
+11: k
+12: l
+
+"(a)(b)(c)(d)(e)(f)(g)(h)(i)(j)(k)(l)A\11B"I
+Capturing subpattern count = 12
+Max back reference = 11
+No options
+First char = 'a'
+Need char = 'B'
+    abcdefghijklAkB
+ 0: abcdefghijklAkB
+ 1: a
+ 2: b
+ 3: c
+ 4: d
+ 5: e
+ 6: f
+ 7: g
+ 8: h
+ 9: i
+10: j
+11: k
+12: l
+
+"(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)(?Pa)"I
+Capturing subpattern count = 101
+Named capturing subpatterns:
+  name0     1
+  name1     2
+  name10   11
+  name100 101
+  name11   12
+  name12   13
+  name13   14
+  name14   15
+  name15   16
+  name16   17
+  name17   18
+  name18   19
+  name19   20
+  name2     3
+  name20   21
+  name21   22
+  name22   23
+  name23   24
+  name24   25
+  name25   26
+  name26   27
+  name27   28
+  name28   29
+  name29   30
+  name3     4
+  name30   31
+  name31   32
+  name32   33
+  name33   34
+  name34   35
+  name35   36
+  name36   37
+  name37   38
+  name38   39
+  name39   40
+  name4     5
+  name40   41
+  name41   42
+  name42   43
+  name43   44
+  name44   45
+  name45   46
+  name46   47
+  name47   48
+  name48   49
+  name49   50
+  name5     6
+  name50   51
+  name51   52
+  name52   53
+  name53   54
+  name54   55
+  name55   56
+  name56   57
+  name57   58
+  name58   59
+  name59   60
+  name6     7
+  name60   61
+  name61   62
+  name62   63
+  name63   64
+  name64   65
+  name65   66
+  name66   67
+  name67   68
+  name68   69
+  name69   70
+  name7     8
+  name70   71
+  name71   72
+  name72   73
+  name73   74
+  name74   75
+  name75   76
+  name76   77
+  name77   78
+  name78   79
+  name79   80
+  name8     9
+  name80   81
+  name81   82
+  name82   83
+  name83   84
+  name84   85
+  name85   86
+  name86   87
+  name87   88
+  name88   89
+  name89   90
+  name9    10
+  name90   91
+  name91   92
+  name92   93
+  name93   94
+  name94   95
+  name95   96
+  name96   97
+  name97   98
+  name98   99
+  name99  100
+No options
+First char = 'a'
+Need char = 'a'
+    aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+Matched, but too many substrings
+ 0: aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+ 1: a
+ 2: a
+ 3: a
+ 4: a
+ 5: a
+ 6: a
+ 7: a
+ 8: a
+ 9: a
+10: a
+11: a
+12: a
+13: a
+14: a
+
+"(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)(a)"I
+Capturing subpattern count = 101
+No options
+First char = 'a'
+Need char = 'a'
+    aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+Matched, but too many substrings
+ 0: aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+ 1: a
+ 2: a
+ 3: a
+ 4: a
+ 5: a
+ 6: a
+ 7: a
+ 8: a
+ 9: a
+10: a
+11: a
+12: a
+13: a
+14: a
+
+/[^()]*(?:\((?R)\)[^()]*)*/I
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+    (this(and)that
+ 0: 
+    (this(and)that)
+ 0: (this(and)that)
+    (this(and)that)stuff
+ 0: (this(and)that)stuff
+
+/[^()]*(?:\((?>(?R))\)[^()]*)*/I
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+    (this(and)that
+ 0: 
+    (this(and)that)
+ 0: (this(and)that)
+
+/[^()]*(?:\((?R)\))*[^()]*/I
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+    (this(and)that
+ 0: 
+    (this(and)that)
+ 0: (this(and)that)
+
+/(?:\((?R)\))*[^()]*/I
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+    (this(and)that
+ 0: 
+    (this(and)that)
+ 0: 
+    ((this))
+ 0: ((this))
+
+/(?:\((?R)\))|[^()]*/I
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+    (this(and)that
+ 0: 
+    (this(and)that)
+ 0: 
+    (this)
+ 0: (this)
+    ((this))
+ 0: ((this))
+
+/\x{0000ff}/I
+Capturing subpattern count = 0
+No options
+First char = \xff
+No need char
+
+/^((?Pa1)|(?Pa2)b)/I
+Failed: two named subpatterns have the same name at offset 17
+
+/^((?Pa1)|(?Pa2)b)/IJ
+Capturing subpattern count = 3
+Named capturing subpatterns:
+  A   2
+  A   3
+Options: anchored dupnames
+No first char
+No need char
+    a1b\CA
+ 0: a1
+ 1: a1
+ 2: a1
+  C a1 (2) A
+    a2b\CA
+ 0: a2b
+ 1: a2b
+ 2: 
+ 3: a2
+  C a2 (2) A
+    ** Failers
+No match
+    a1b\CZ\CA
+no parentheses with name "Z"
+ 0: a1
+ 1: a1
+ 2: a1
+copy substring Z failed -7
+  C a1 (2) A
+
+/^(?Pa)(?Pb)/IJ
+Capturing subpattern count = 2
+Named capturing subpatterns:
+  A   1
+  A   2
+Options: anchored dupnames
+No first char
+No need char
+    ab\CA
+ 0: ab
+ 1: a
+ 2: b
+  C a (1) A
+
+/^(?Pa)(?Pb)|cd/IJ
+Capturing subpattern count = 2
+Named capturing subpatterns:
+  A   1
+  A   2
+Options: dupnames
+No first char
+No need char
+    ab\CA
+ 0: ab
+ 1: a
+ 2: b
+  C a (1) A
+    cd\CA
+ 0: cd
+copy substring A failed -7
+
+/^(?Pa)(?Pb)|cd(?Pef)(?Pgh)/IJ
+Capturing subpattern count = 4
+Named capturing subpatterns:
+  A   1
+  A   2
+  A   3
+  A   4
+Options: dupnames
+No first char
+No need char
+    cdefgh\CA
+ 0: cdefgh
+ 1: 
+ 2: 
+ 3: ef
+ 4: gh
+  C ef (2) A
+
+/^((?Pa1)|(?Pa2)b)/IJ
+Capturing subpattern count = 3
+Named capturing subpatterns:
+  A   2
+  A   3
+Options: anchored dupnames
+No first char
+No need char
+    a1b\GA
+ 0: a1
+ 1: a1
+ 2: a1
+  G a1 (2) A
+    a2b\GA
+ 0: a2b
+ 1: a2b
+ 2: 
+ 3: a2
+  G a2 (2) A
+    ** Failers
+No match
+    a1b\GZ\GA
+no parentheses with name "Z"
+ 0: a1
+ 1: a1
+ 2: a1
+get substring Z failed -7
+  G a1 (2) A
+
+/^(?Pa)(?Pb)/IJ
+Capturing subpattern count = 2
+Named capturing subpatterns:
+  A   1
+  A   2
+Options: anchored dupnames
+No first char
+No need char
+    ab\GA
+ 0: ab
+ 1: a
+ 2: b
+  G a (1) A
+
+/^(?Pa)(?Pb)|cd/IJ
+Capturing subpattern count = 2
+Named capturing subpatterns:
+  A   1
+  A   2
+Options: dupnames
+No first char
+No need char
+    ab\GA
+ 0: ab
+ 1: a
+ 2: b
+  G a (1) A
+    cd\GA
+ 0: cd
+get substring A failed -7
+
+/^(?Pa)(?Pb)|cd(?Pef)(?Pgh)/IJ
+Capturing subpattern count = 4
+Named capturing subpatterns:
+  A   1
+  A   2
+  A   3
+  A   4
+Options: dupnames
+No first char
+No need char
+    cdefgh\GA
+ 0: cdefgh
+ 1: 
+ 2: 
+ 3: ef
+ 4: gh
+  G ef (2) A
+
+/(?J)^((?Pa1)|(?Pa2)b)/I
+Capturing subpattern count = 3
+Named capturing subpatterns:
+  A   2
+  A   3
+Options: anchored dupnames
+Duplicate name status changes
+No first char
+No need char
+    a1b\CA
+ 0: a1
+ 1: a1
+ 2: a1
+  C a1 (2) A
+    a2b\CA
+ 0: a2b
+ 1: a2b
+ 2: 
+ 3: a2
+  C a2 (2) A
+
+/^(?Pa) (?J:(?Pb)(?Pc)) (?Pd)/I
+Failed: two named subpatterns have the same name at offset 37
+
+/ In this next test, J is not set at the outer level; consequently it isn't
+set in the pattern's options; consequently pcre_get_named_substring() produces
+a random value. /Ix
+Capturing subpattern count = 1
+Options: extended
+First char = 'I'
+Need char = 'e'
+
+/^(?Pa) (?J:(?Pb)(?Pc)) (?Pd)/I
+Capturing subpattern count = 4
+Named capturing subpatterns:
+  A   1
+  B   2
+  B   3
+  C   4
+Options: anchored
+Duplicate name status changes
+No first char
+No need char
+    a bc d\CA\CB\CC
+ 0: a bc d
+ 1: a
+ 2: b
+ 3: c
+ 4: d
+  C a (1) A
+  C b (1) B
+  C d (1) C
+
+/^(?Pa)?(?(A)a|b)/I
+Capturing subpattern count = 1
+Named capturing subpatterns:
+  A   1
+Options: anchored
+No first char
+No need char
+    aabc
+ 0: aa
+ 1: a
+    bc
+ 0: b
+    ** Failers
+No match
+    abc
+No match
+
+/(?:(?(ZZ)a|b)(?PX))+/I
+Capturing subpattern count = 1
+Named capturing subpatterns:
+  ZZ   1
+No options
+No first char
+Need char = 'X'
+    bXaX
+ 0: bXaX
+ 1: X
+
+/(?:(?(2y)a|b)(X))+/I
+Failed: reference to non-existent subpattern at offset 9
+
+/(?:(?(ZA)a|b)(?PX))+/I
+Failed: reference to non-existent subpattern at offset 9
+
+/(?:(?(ZZ)a|b)(?(ZZ)a|b)(?PX))+/I
+Capturing subpattern count = 1
+Named capturing subpatterns:
+  ZZ   1
+No options
+No first char
+Need char = 'X'
+    bbXaaX
+ 0: bbXaaX
+ 1: X
+
+/(?:(?(ZZ)a|\(b\))\\(?PX))+/I
+Capturing subpattern count = 1
+Named capturing subpatterns:
+  ZZ   1
+No options
+No first char
+Need char = 'X'
+    (b)\\Xa\\X
+ 0: (b)\Xa\X
+ 1: X
+
+/(?PX|Y))+/I
+Capturing subpattern count = 1
+Max back reference = 1
+Named capturing subpatterns:
+  A   1
+No options
+No first char
+No need char
+    bXXaYYaY
+ 0: bXXaYYaY
+ 1: Y
+    bXYaXXaX
+ 0: bX
+ 1: X
+
+/()()()()()()()()()(?:(?(A)(?P=A)a|b)(?PX|Y))+/I
+Capturing subpattern count = 10
+Max back reference = 10
+Named capturing subpatterns:
+  A  10
+No options
+No first char
+No need char
+    bXXaYYaY
+ 0: bXXaYYaY
+ 1: 
+ 2: 
+ 3: 
+ 4: 
+ 5: 
+ 6: 
+ 7: 
+ 8: 
+ 9: 
+10: Y
+
+/\s*,\s*/IS
+Capturing subpattern count = 0
+No options
+No first char
+Need char = ','
+Subject length lower bound = 1
+Starting byte set: \x09 \x0a \x0c \x0d \x20 , 
+    \x0b,\x0b
+ 0: ,
+    \x0c,\x0d
+ 0: \x0c,\x0d
+
+/^abc/Im
+Capturing subpattern count = 0
+Options: multiline
+First char at start or follows newline
+Need char = 'c'
+    xyz\nabc
+ 0: abc
+    xyz\nabc\
+ 0: abc
+    xyz\r\nabc\
+ 0: abc
+    xyz\rabc\
+ 0: abc
+    xyz\r\nabc\
+ 0: abc
+    ** Failers
+No match
+    xyz\nabc\
+No match
+    xyz\r\nabc\
+No match
+    xyz\nabc\
+No match
+    xyz\rabc\
+No match
+    xyz\rabc\
+No match
+
+/abc$/Im
+Capturing subpattern count = 0
+Options: multiline
+Forced newline sequence: LF
+First char = 'a'
+Need char = 'c'
+    xyzabc
+ 0: abc
+    xyzabc\n
+ 0: abc
+    xyzabc\npqr
+ 0: abc
+    xyzabc\r\
+ 0: abc
+    xyzabc\rpqr\
+ 0: abc
+    xyzabc\r\n\
+ 0: abc
+    xyzabc\r\npqr\
+ 0: abc
+    ** Failers
+No match
+    xyzabc\r
+No match
+    xyzabc\rpqr
+No match
+    xyzabc\r\n
+No match
+    xyzabc\r\npqr
+No match
+
+/^abc/Im
+Capturing subpattern count = 0
+Options: multiline
+Forced newline sequence: CR
+First char at start or follows newline
+Need char = 'c'
+    xyz\rabcdef
+ 0: abc
+    xyz\nabcdef\
+ 0: abc
+    ** Failers
+No match
+    xyz\nabcdef
+No match
+
+/^abc/Im
+Capturing subpattern count = 0
+Options: multiline
+Forced newline sequence: LF
+First char at start or follows newline
+Need char = 'c'
+    xyz\nabcdef
+ 0: abc
+    xyz\rabcdef\
+ 0: abc
+    ** Failers
+No match
+    xyz\rabcdef
+No match
+
+/^abc/Im
+Capturing subpattern count = 0
+Options: multiline
+Forced newline sequence: CRLF
+First char at start or follows newline
+Need char = 'c'
+    xyz\r\nabcdef
+ 0: abc
+    xyz\rabcdef\
+ 0: abc
+    ** Failers
+No match
+    xyz\rabcdef
+No match
+
+/^abc/Im
+Unknown newline type at: 
+
+
+/abc/I
+Capturing subpattern count = 0
+No options
+First char = 'a'
+Need char = 'c'
+    xyz\rabc\
+Unknown newline type at: 
+    abc
+ 0: abc
+
+/.*/I
+Capturing subpattern count = 0
+Options:
+Forced newline sequence: LF
+First char at start or follows newline
+No need char
+    abc\ndef
+ 0: abc
+    abc\rdef
+ 0: abc\x0ddef
+    abc\r\ndef
+ 0: abc\x0d
+    \abc\ndef
+ 0: abc\x0adef
+    \abc\rdef
+ 0: abc
+    \abc\r\ndef
+ 0: abc
+    \abc\ndef
+ 0: abc\x0adef
+    \abc\rdef
+ 0: abc\x0ddef
+    \abc\r\ndef
+ 0: abc
+
+/\w+(.)(.)?def/Is
+Capturing subpattern count = 2
+Options: dotall
+No first char
+Need char = 'f'
+    abc\ndef
+ 0: abc\x0adef
+ 1: \x0a
+    abc\rdef
+ 0: abc\x0ddef
+ 1: \x0d
+    abc\r\ndef
+ 0: abc\x0d\x0adef
+ 1: \x0d
+ 2: \x0a
+
++((?:\s|//.*\\n|/[*](?:\\n|.)*?[*]/)*)+I
+Capturing subpattern count = 1
+No options
+No first char
+No need char
+   /* this is a C style comment */\M
+Minimum match() limit = 120
+Minimum match() recursion limit = 6
+ 0: /* this is a C style comment */
+ 1: /* this is a C style comment */
+
+/(?P25[0-5]|2[0-4]\d|[01]?\d?\d)(?:\.(?P>B)){3}/I
+Capturing subpattern count = 1
+Named capturing subpatterns:
+  B   1
+No options
+No first char
+Need char = '.'
+
+/()()()()()()()()()()()()()()()()()()()()
+ ()()()()()()()()()()()()()()()()()()()()
+ ()()()()()()()()()()()()()()()()()()()()
+ ()()()()()()()()()()()()()()()()()()()()
+ ()()()()()()()()()()()()()()()()()()()()
+ (.(.))/Ix
+Capturing subpattern count = 102
+Options: extended
+No first char
+No need char
+    XY\O400
+ 0: XY
+ 1: 
+ 2: 
+ 3: 
+ 4: 
+ 5: 
+ 6: 
+ 7: 
+ 8: 
+ 9: 
+10: 
+11: 
+12: 
+13: 
+14: 
+15: 
+16: 
+17: 
+18: 
+19: 
+20: 
+21: 
+22: 
+23: 
+24: 
+25: 
+26: 
+27: 
+28: 
+29: 
+30: 
+31: 
+32: 
+33: 
+34: 
+35: 
+36: 
+37: 
+38: 
+39: 
+40: 
+41: 
+42: 
+43: 
+44: 
+45: 
+46: 
+47: 
+48: 
+49: 
+50: 
+51: 
+52: 
+53: 
+54: 
+55: 
+56: 
+57: 
+58: 
+59: 
+60: 
+61: 
+62: 
+63: 
+64: 
+65: 
+66: 
+67: 
+68: 
+69: 
+70: 
+71: 
+72: 
+73: 
+74: 
+75: 
+76: 
+77: 
+78: 
+79: 
+80: 
+81: 
+82: 
+83: 
+84: 
+85: 
+86: 
+87: 
+88: 
+89: 
+90: 
+91: 
+92: 
+93: 
+94: 
+95: 
+96: 
+97: 
+98: 
+99: 
+100: 
+101: XY
+102: Y
+
+/(a*b|(?i:c*(?-i)d))/IS
+Capturing subpattern count = 1
+No options
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: C a b c d 
+
+/()[ab]xyz/IS
+Capturing subpattern count = 1
+No options
+No first char
+Need char = 'z'
+Subject length lower bound = 4
+Starting byte set: a b 
+
+/(|)[ab]xyz/IS
+Capturing subpattern count = 1
+No options
+No first char
+Need char = 'z'
+Subject length lower bound = 4
+Starting byte set: a b 
+
+/(|c)[ab]xyz/IS
+Capturing subpattern count = 1
+No options
+No first char
+Need char = 'z'
+Subject length lower bound = 4
+Starting byte set: a b c 
+
+/(|c?)[ab]xyz/IS
+Capturing subpattern count = 1
+No options
+No first char
+Need char = 'z'
+Subject length lower bound = 4
+Starting byte set: a b c 
+
+/(d?|c?)[ab]xyz/IS
+Capturing subpattern count = 1
+No options
+No first char
+Need char = 'z'
+Subject length lower bound = 4
+Starting byte set: a b c d 
+
+/(d?|c)[ab]xyz/IS
+Capturing subpattern count = 1
+No options
+No first char
+Need char = 'z'
+Subject length lower bound = 4
+Starting byte set: a b c d 
+
+/^a*b\d/DZ
+------------------------------------------------------------------
+        Bra
+        ^
+        a*+
+        b
+        \d
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: anchored
+No first char
+Need char = 'b'
+
+/^a*+b\d/DZ
+------------------------------------------------------------------
+        Bra
+        ^
+        a*+
+        b
+        \d
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: anchored
+No first char
+Need char = 'b'
+
+/^a*?b\d/DZ
+------------------------------------------------------------------
+        Bra
+        ^
+        a*+
+        b
+        \d
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: anchored
+No first char
+Need char = 'b'
+
+/^a+A\d/DZ
+------------------------------------------------------------------
+        Bra
+        ^
+        a++
+        A
+        \d
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: anchored
+No first char
+Need char = 'A'
+    aaaA5
+ 0: aaaA5
+    ** Failers
+No match
+    aaaa5
+No match
+
+/^a*A\d/IiDZ
+------------------------------------------------------------------
+        Bra
+        ^
+     /i a*
+     /i A
+        \d
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: anchored caseless
+No first char
+Need char = 'A' (caseless)
+    aaaA5
+ 0: aaaA5
+    aaaa5
+ 0: aaaa5
+
+/(a*|b*)[cd]/IS
+Capturing subpattern count = 1
+No options
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: a b c d 
+
+/(a+|b*)[cd]/IS
+Capturing subpattern count = 1
+No options
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: a b c d 
+
+/(a*|b+)[cd]/IS
+Capturing subpattern count = 1
+No options
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: a b c d 
+
+/(a+|b+)[cd]/IS
+Capturing subpattern count = 1
+No options
+No first char
+No need char
+Subject length lower bound = 2
+Starting byte set: a b 
+
+/((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((
+ ((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((
+ (((
+ a
+ ))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))
+ ))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))
+ )))
+/Ix
+Capturing subpattern count = 203
+Options: extended
+First char = 'a'
+No need char
+  large nest
+Matched, but too many substrings
+ 0: a
+ 1: a
+ 2: a
+ 3: a
+ 4: a
+ 5: a
+ 6: a
+ 7: a
+ 8: a
+ 9: a
+10: a
+11: a
+12: a
+13: a
+14: a
+
+/a*\d/BZ
+------------------------------------------------------------------
+        Bra
+        a*+
+        \d
+        Ket
+        End
+------------------------------------------------------------------
+
+/a*\D/BZ
+------------------------------------------------------------------
+        Bra
+        a*
+        \D
+        Ket
+        End
+------------------------------------------------------------------
+
+/0*\d/BZ
+------------------------------------------------------------------
+        Bra
+        0*
+        \d
+        Ket
+        End
+------------------------------------------------------------------
+
+/0*\D/BZ
+------------------------------------------------------------------
+        Bra
+        0*+
+        \D
+        Ket
+        End
+------------------------------------------------------------------
+
+/a*\s/BZ
+------------------------------------------------------------------
+        Bra
+        a*+
+        \s
+        Ket
+        End
+------------------------------------------------------------------
+
+/a*\S/BZ
+------------------------------------------------------------------
+        Bra
+        a*
+        \S
+        Ket
+        End
+------------------------------------------------------------------
+
+/ *\s/BZ
+------------------------------------------------------------------
+        Bra
+         *
+        \s
+        Ket
+        End
+------------------------------------------------------------------
+
+/ *\S/BZ
+------------------------------------------------------------------
+        Bra
+         *+
+        \S
+        Ket
+        End
+------------------------------------------------------------------
+
+/a*\w/BZ
+------------------------------------------------------------------
+        Bra
+        a*
+        \w
+        Ket
+        End
+------------------------------------------------------------------
+
+/a*\W/BZ
+------------------------------------------------------------------
+        Bra
+        a*+
+        \W
+        Ket
+        End
+------------------------------------------------------------------
+
+/=*\w/BZ
+------------------------------------------------------------------
+        Bra
+        =*+
+        \w
+        Ket
+        End
+------------------------------------------------------------------
+
+/=*\W/BZ
+------------------------------------------------------------------
+        Bra
+        =*
+        \W
+        Ket
+        End
+------------------------------------------------------------------
+
+/\d*a/BZ
+------------------------------------------------------------------
+        Bra
+        \d*+
+        a
+        Ket
+        End
+------------------------------------------------------------------
+
+/\d*2/BZ
+------------------------------------------------------------------
+        Bra
+        \d*
+        2
+        Ket
+        End
+------------------------------------------------------------------
+
+/\d*\d/BZ
+------------------------------------------------------------------
+        Bra
+        \d*
+        \d
+        Ket
+        End
+------------------------------------------------------------------
+
+/\d*\D/BZ
+------------------------------------------------------------------
+        Bra
+        \d*+
+        \D
+        Ket
+        End
+------------------------------------------------------------------
+
+/\d*\s/BZ
+------------------------------------------------------------------
+        Bra
+        \d*+
+        \s
+        Ket
+        End
+------------------------------------------------------------------
+
+/\d*\S/BZ
+------------------------------------------------------------------
+        Bra
+        \d*
+        \S
+        Ket
+        End
+------------------------------------------------------------------
+
+/\d*\w/BZ
+------------------------------------------------------------------
+        Bra
+        \d*
+        \w
+        Ket
+        End
+------------------------------------------------------------------
+
+/\d*\W/BZ
+------------------------------------------------------------------
+        Bra
+        \d*+
+        \W
+        Ket
+        End
+------------------------------------------------------------------
+
+/\D*a/BZ
+------------------------------------------------------------------
+        Bra
+        \D*
+        a
+        Ket
+        End
+------------------------------------------------------------------
+
+/\D*2/BZ
+------------------------------------------------------------------
+        Bra
+        \D*+
+        2
+        Ket
+        End
+------------------------------------------------------------------
+
+/\D*\d/BZ
+------------------------------------------------------------------
+        Bra
+        \D*+
+        \d
+        Ket
+        End
+------------------------------------------------------------------
+
+/\D*\D/BZ
+------------------------------------------------------------------
+        Bra
+        \D*
+        \D
+        Ket
+        End
+------------------------------------------------------------------
+
+/\D*\s/BZ
+------------------------------------------------------------------
+        Bra
+        \D*
+        \s
+        Ket
+        End
+------------------------------------------------------------------
+
+/\D*\S/BZ
+------------------------------------------------------------------
+        Bra
+        \D*
+        \S
+        Ket
+        End
+------------------------------------------------------------------
+
+/\D*\w/BZ
+------------------------------------------------------------------
+        Bra
+        \D*
+        \w
+        Ket
+        End
+------------------------------------------------------------------
+
+/\D*\W/BZ
+------------------------------------------------------------------
+        Bra
+        \D*
+        \W
+        Ket
+        End
+------------------------------------------------------------------
+
+/\s*a/BZ
+------------------------------------------------------------------
+        Bra
+        \s*+
+        a
+        Ket
+        End
+------------------------------------------------------------------
+
+/\s*2/BZ
+------------------------------------------------------------------
+        Bra
+        \s*+
+        2
+        Ket
+        End
+------------------------------------------------------------------
+
+/\s*\d/BZ
+------------------------------------------------------------------
+        Bra
+        \s*+
+        \d
+        Ket
+        End
+------------------------------------------------------------------
+
+/\s*\D/BZ
+------------------------------------------------------------------
+        Bra
+        \s*
+        \D
+        Ket
+        End
+------------------------------------------------------------------
+
+/\s*\s/BZ
+------------------------------------------------------------------
+        Bra
+        \s*
+        \s
+        Ket
+        End
+------------------------------------------------------------------
+
+/\s*\S/BZ
+------------------------------------------------------------------
+        Bra
+        \s*+
+        \S
+        Ket
+        End
+------------------------------------------------------------------
+
+/\s*\w/BZ
+------------------------------------------------------------------
+        Bra
+        \s*+
+        \w
+        Ket
+        End
+------------------------------------------------------------------
+
+/\s*\W/BZ
+------------------------------------------------------------------
+        Bra
+        \s*
+        \W
+        Ket
+        End
+------------------------------------------------------------------
+
+/\S*a/BZ
+------------------------------------------------------------------
+        Bra
+        \S*
+        a
+        Ket
+        End
+------------------------------------------------------------------
+
+/\S*2/BZ
+------------------------------------------------------------------
+        Bra
+        \S*
+        2
+        Ket
+        End
+------------------------------------------------------------------
+
+/\S*\d/BZ
+------------------------------------------------------------------
+        Bra
+        \S*
+        \d
+        Ket
+        End
+------------------------------------------------------------------
+
+/\S*\D/BZ
+------------------------------------------------------------------
+        Bra
+        \S*
+        \D
+        Ket
+        End
+------------------------------------------------------------------
+
+/\S*\s/BZ
+------------------------------------------------------------------
+        Bra
+        \S*+
+        \s
+        Ket
+        End
+------------------------------------------------------------------
+
+/\S*\S/BZ
+------------------------------------------------------------------
+        Bra
+        \S*
+        \S
+        Ket
+        End
+------------------------------------------------------------------
+
+/\S*\w/BZ
+------------------------------------------------------------------
+        Bra
+        \S*
+        \w
+        Ket
+        End
+------------------------------------------------------------------
+
+/\S*\W/BZ
+------------------------------------------------------------------
+        Bra
+        \S*
+        \W
+        Ket
+        End
+------------------------------------------------------------------
+
+/\w*a/BZ
+------------------------------------------------------------------
+        Bra
+        \w*
+        a
+        Ket
+        End
+------------------------------------------------------------------
+
+/\w*2/BZ
+------------------------------------------------------------------
+        Bra
+        \w*
+        2
+        Ket
+        End
+------------------------------------------------------------------
+
+/\w*\d/BZ
+------------------------------------------------------------------
+        Bra
+        \w*
+        \d
+        Ket
+        End
+------------------------------------------------------------------
+
+/\w*\D/BZ
+------------------------------------------------------------------
+        Bra
+        \w*
+        \D
+        Ket
+        End
+------------------------------------------------------------------
+
+/\w*\s/BZ
+------------------------------------------------------------------
+        Bra
+        \w*+
+        \s
+        Ket
+        End
+------------------------------------------------------------------
+
+/\w*\S/BZ
+------------------------------------------------------------------
+        Bra
+        \w*
+        \S
+        Ket
+        End
+------------------------------------------------------------------
+
+/\w*\w/BZ
+------------------------------------------------------------------
+        Bra
+        \w*
+        \w
+        Ket
+        End
+------------------------------------------------------------------
+
+/\w*\W/BZ
+------------------------------------------------------------------
+        Bra
+        \w*+
+        \W
+        Ket
+        End
+------------------------------------------------------------------
+
+/\W*a/BZ
+------------------------------------------------------------------
+        Bra
+        \W*+
+        a
+        Ket
+        End
+------------------------------------------------------------------
+
+/\W*2/BZ
+------------------------------------------------------------------
+        Bra
+        \W*+
+        2
+        Ket
+        End
+------------------------------------------------------------------
+
+/\W*\d/BZ
+------------------------------------------------------------------
+        Bra
+        \W*+
+        \d
+        Ket
+        End
+------------------------------------------------------------------
+
+/\W*\D/BZ
+------------------------------------------------------------------
+        Bra
+        \W*
+        \D
+        Ket
+        End
+------------------------------------------------------------------
+
+/\W*\s/BZ
+------------------------------------------------------------------
+        Bra
+        \W*
+        \s
+        Ket
+        End
+------------------------------------------------------------------
+
+/\W*\S/BZ
+------------------------------------------------------------------
+        Bra
+        \W*
+        \S
+        Ket
+        End
+------------------------------------------------------------------
+
+/\W*\w/BZ
+------------------------------------------------------------------
+        Bra
+        \W*+
+        \w
+        Ket
+        End
+------------------------------------------------------------------
+
+/\W*\W/BZ
+------------------------------------------------------------------
+        Bra
+        \W*
+        \W
+        Ket
+        End
+------------------------------------------------------------------
+
+/[^a]+a/BZ
+------------------------------------------------------------------
+        Bra
+        [^a]++
+        a
+        Ket
+        End
+------------------------------------------------------------------
+
+/[^a]+a/BZi
+------------------------------------------------------------------
+        Bra
+     /i [^a]++
+     /i a
+        Ket
+        End
+------------------------------------------------------------------
+
+/[^a]+A/BZi
+------------------------------------------------------------------
+        Bra
+     /i [^a]++
+     /i A
+        Ket
+        End
+------------------------------------------------------------------
+
+/[^a]+b/BZ
+------------------------------------------------------------------
+        Bra
+        [^a]+
+        b
+        Ket
+        End
+------------------------------------------------------------------
+
+/[^a]+\d/BZ
+------------------------------------------------------------------
+        Bra
+        [^a]+
+        \d
+        Ket
+        End
+------------------------------------------------------------------
+
+/a*[^a]/BZ
+------------------------------------------------------------------
+        Bra
+        a*
+        [^a]
+        Ket
+        End
+------------------------------------------------------------------
+
+/(?Px)(?Py)/I
+Capturing subpattern count = 2
+Named capturing subpatterns:
+  abc   1
+  xyz   2
+No options
+First char = 'x'
+Need char = 'y'
+    xy\Cabc\Cxyz
+ 0: xy
+ 1: x
+ 2: y
+  C x (1) abc
+  C y (1) xyz
+
+/(?x)(?'xyz'y)/I
+Capturing subpattern count = 2
+Named capturing subpatterns:
+  abc   1
+  xyz   2
+No options
+First char = 'x'
+Need char = 'y'
+    xy\Cabc\Cxyz
+ 0: xy
+ 1: x
+ 2: y
+  C x (1) abc
+  C y (1) xyz
+
+/(?x)(?'xyz>y)/I
+Failed: syntax error in subpattern name (missing terminator) at offset 15
+
+/(?P'abc'x)(?Py)/I
+Failed: unrecognized character after (?P at offset 3
+
+/^(?:(?(ZZ)a|b)(?X))+/
+    bXaX
+ 0: bXaX
+ 1: X
+    bXbX
+ 0: bX
+ 1: X
+    ** Failers
+No match
+    aXaX
+No match
+    aXbX
+No match
+
+/^(?P>abc)(?xxx)/
+Failed: reference to non-existent subpattern at offset 8
+
+/^(?P>abc)(?x|y)/
+    xx
+ 0: xx
+ 1: x
+    xy
+ 0: xy
+ 1: y
+    yy
+ 0: yy
+ 1: y
+    yx
+ 0: yx
+ 1: x
+
+/^(?P>abc)(?Px|y)/
+    xx
+ 0: xx
+ 1: x
+    xy
+ 0: xy
+ 1: y
+    yy
+ 0: yy
+ 1: y
+    yx
+ 0: yx
+ 1: x
+
+/^((?(abc)a|b)(?x|y))+/
+    bxay
+ 0: bxay
+ 1: ay
+ 2: y
+    bxby
+ 0: bx
+ 1: bx
+ 2: x
+    ** Failers
+No match
+    axby
+No match
+
+/^(((?P=abc)|X)(?x|y))+/
+    XxXxxx
+ 0: XxXxxx
+ 1: xx
+ 2: x
+ 3: x
+    XxXyyx
+ 0: XxXyyx
+ 1: yx
+ 2: y
+ 3: x
+    XxXyxx
+ 0: XxXy
+ 1: Xy
+ 2: X
+ 3: y
+    ** Failers
+No match
+    x
+No match
+
+/^(?1)(abc)/
+    abcabc
+ 0: abcabc
+ 1: abc
+
+/^(?:(?:\1|X)(a|b))+/
+    Xaaa
+ 0: Xaaa
+ 1: a
+    Xaba
+ 0: Xa
+ 1: a
+
+/^[\E\Qa\E-\Qz\E]+/BZ
+------------------------------------------------------------------
+        Bra
+        ^
+        [a-z]+
+        Ket
+        End
+------------------------------------------------------------------
+
+/^[a\Q]bc\E]/BZ
+------------------------------------------------------------------
+        Bra
+        ^
+        [\]a-c]
+        Ket
+        End
+------------------------------------------------------------------
+
+/^[a-\Q\E]/BZ
+------------------------------------------------------------------
+        Bra
+        ^
+        [\-a]
+        Ket
+        End
+------------------------------------------------------------------
+
+/^(?P>abc)[()](?)/BZ
+------------------------------------------------------------------
+        Bra
+        ^
+        Recurse
+        [()]
+        CBra 1
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+
+/^((?(abc)y)[()](?Px))+/BZ
+------------------------------------------------------------------
+        Bra
+        ^
+        CBra 1
+        Cond
+      2 Cond nref
+        y
+        Ket
+        [()]
+        CBra 2
+        x
+        Ket
+        KetRmax
+        Ket
+        End
+------------------------------------------------------------------
+    (xy)x
+ 0: (xy)x
+ 1: y)x
+ 2: x
+
+/^(?P>abc)\Q()\E(?)/BZ
+------------------------------------------------------------------
+        Bra
+        ^
+        Recurse
+        ()
+        CBra 1
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+
+/^(?P>abc)[a\Q(]\E(](?)/BZ
+------------------------------------------------------------------
+        Bra
+        ^
+        Recurse
+        [(\]a]
+        CBra 1
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+
+/^(?P>abc) # this is (a comment)
+  (?)/BZx
+------------------------------------------------------------------
+        Bra
+        ^
+        Recurse
+        CBra 1
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+
+/^\W*(?:(?(?.)\W*(?&one)\W*\k|)|(?(?.)\W*(?&three)\W*\k'four'|\W*.\W*))\W*$/Ii
+Capturing subpattern count = 4
+Max back reference = 4
+Named capturing subpatterns:
+  four    4
+  one     1
+  three   3
+  two     2
+Options: anchored caseless
+No first char
+No need char
+    1221
+ 0: 1221
+ 1: 1221
+ 2: 1
+    Satan, oscillate my metallic sonatas!
+ 0: Satan, oscillate my metallic sonatas!
+ 1: 
+ 2: 
+ 3: Satan, oscillate my metallic sonatas
+ 4: S
+    A man, a plan, a canal: Panama!
+ 0: A man, a plan, a canal: Panama!
+ 1: 
+ 2: 
+ 3: A man, a plan, a canal: Panama
+ 4: A
+    Able was I ere I saw Elba.
+ 0: Able was I ere I saw Elba.
+ 1: 
+ 2: 
+ 3: Able was I ere I saw Elba
+ 4: A
+    *** Failers
+No match
+    The quick brown fox
+No match
+
+/(?=(\w+))\1:/I
+Capturing subpattern count = 1
+Max back reference = 1
+No options
+No first char
+Need char = ':'
+    abcd:
+ 0: abcd:
+ 1: abcd
+
+/(?=(?'abc'\w+))\k:/I
+Capturing subpattern count = 1
+Max back reference = 1
+Named capturing subpatterns:
+  abc   1
+No options
+No first char
+Need char = ':'
+    abcd:
+ 0: abcd:
+ 1: abcd
+
+/(?'abc'a|b)(?d|e)\k{2}/J
+    adaa
+ 0: adaa
+ 1: a
+ 2: d
+    ** Failers
+No match
+    addd
+No match
+    adbb
+No match
+
+/(?'abc'a|b)(?d|e)(?&abc){2}/J
+    bdaa
+ 0: bdaa
+ 1: b
+ 2: d
+    bdab
+ 0: bdab
+ 1: b
+ 2: d
+    ** Failers
+No match
+    bddd
+No match
+
+/(?( (?'B' abc (?(R) (?(R&A)1) (?(R&B)2) X  |  (?1)  (?2)   (?R) ))) /x
+    abcabc1Xabc2XabcXabcabc
+ 0: abcabc1Xabc2XabcX
+ 1: abcabc1Xabc2XabcX
+ 2: abcabc1Xabc2XabcX
+
+/(? (?'B' abc (?(R) (?(R&1)1) (?(R&B)2) X  |  (?1)  (?2)   (?R) ))) /x
+Failed: reference to non-existent subpattern at offset 29
+
+/(?<1> (?'B' abc (?(R) (?(R&1)1) (?(R&B)2) X  |  (?1)  (?2)   (?R) ))) /x
+    abcabc1Xabc2XabcXabcabc
+ 0: abcabc1Xabc2XabcX
+ 1: abcabc1Xabc2XabcX
+ 2: abcabc1Xabc2XabcX
+
+/^(?(DEFINE) abc | xyz ) /x
+Failed: DEFINE group contains more than one branch at offset 22
+
+/(?(DEFINE) abc) xyz/xI
+Capturing subpattern count = 0
+Options: extended
+First char = 'x'
+Need char = 'z'
+
+/(a|)*\d/
+  \O0aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+No match
+  \O0aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa4
+Matched, but too many substrings
+
+/^a.b/
+    a\rb
+ 0: a\x0db
+    a\nb\
+ 0: a\x0ab
+    a\x85b\ 
+ 0: a\x85b
+    ** Failers
+No match
+    a\nb
+No match
+    a\nb\
+No match
+    a\rb\
+No match
+    a\rb\
+No match
+    a\x85b\ 
+No match
+    a\rb\
+No match
+
+/^abc./mgx
+    abc1 \x0aabc2 \x0babc3xx \x0cabc4 \x0dabc5xx \x0d\x0aabc6 \x85abc7 JUNK
+ 0: abc1
+ 0: abc2
+ 0: abc3
+ 0: abc4
+ 0: abc5
+ 0: abc6
+ 0: abc7
+
+/abc.$/mgx
+    abc1\x0a abc2\x0b abc3\x0c abc4\x0d abc5\x0d\x0a abc6\x85 abc7 abc9
+ 0: abc1
+ 0: abc2
+ 0: abc3
+ 0: abc4
+ 0: abc5
+ 0: abc6
+ 0: abc9
+
+/a/
+
+/a/
+Failed: inconsistent NEWLINE options at offset 0
+
+/^a\Rb/
+    a\nb
+ 0: a\x0ab
+    a\rb
+ 0: a\x0db
+    a\r\nb
+ 0: a\x0d\x0ab
+    a\x0bb
+ 0: a\x0bb
+    a\x0cb
+ 0: a\x0cb
+    a\x85b
+ 0: a\x85b
+    ** Failers
+No match
+    a\n\rb
+No match
+
+/^a\R*b/
+    ab
+ 0: ab
+    a\nb
+ 0: a\x0ab
+    a\rb
+ 0: a\x0db
+    a\r\nb
+ 0: a\x0d\x0ab
+    a\x0bb
+ 0: a\x0bb
+    a\x0cb
+ 0: a\x0cb
+    a\x85b
+ 0: a\x85b
+    a\n\rb
+ 0: a\x0a\x0db
+    a\n\r\x85\x0cb
+ 0: a\x0a\x0d\x85\x0cb
+
+/^a\R+b/
+    a\nb
+ 0: a\x0ab
+    a\rb
+ 0: a\x0db
+    a\r\nb
+ 0: a\x0d\x0ab
+    a\x0bb
+ 0: a\x0bb
+    a\x0cb
+ 0: a\x0cb
+    a\x85b
+ 0: a\x85b
+    a\n\rb
+ 0: a\x0a\x0db
+    a\n\r\x85\x0cb
+ 0: a\x0a\x0d\x85\x0cb
+    ** Failers
+No match
+    ab
+No match
+
+/^a\R{1,3}b/
+    a\nb
+ 0: a\x0ab
+    a\n\rb
+ 0: a\x0a\x0db
+    a\n\r\x85b
+ 0: a\x0a\x0d\x85b
+    a\r\n\r\nb
+ 0: a\x0d\x0a\x0d\x0ab
+    a\r\n\r\n\r\nb
+ 0: a\x0d\x0a\x0d\x0a\x0d\x0ab
+    a\n\r\n\rb
+ 0: a\x0a\x0d\x0a\x0db
+    a\n\n\r\nb
+ 0: a\x0a\x0a\x0d\x0ab
+    ** Failers
+No match
+    a\n\n\n\rb
+No match
+    a\r
+No match
+
+/^a[\R]b/
+    aRb
+ 0: aRb
+    ** Failers
+No match
+    a\nb
+No match
+
+/(?&abc)X(?P)/I
+Capturing subpattern count = 1
+Named capturing subpatterns:
+  abc   1
+No options
+No first char
+Need char = 'P'
+    abcPXP123
+ 0: PXP
+ 1: P
+
+/(?1)X(?P)/I
+Capturing subpattern count = 1
+Named capturing subpatterns:
+  abc   1
+No options
+No first char
+Need char = 'P'
+    abcPXP123
+ 0: PXP
+ 1: P
+
+/(?:a(?&abc)b)*(?x)/
+    123axbaxbaxbx456
+ 0: axbaxbaxbx
+ 1: x
+    123axbaxbaxb456
+ 0: x
+ 1: x
+
+/(?:a(?&abc)b){1,5}(?x)/
+    123axbaxbaxbx456
+ 0: axbaxbaxbx
+ 1: x
+
+/(?:a(?&abc)b){2,5}(?x)/
+    123axbaxbaxbx456
+ 0: axbaxbaxbx
+ 1: x
+
+/(?:a(?&abc)b){2,}(?x)/
+    123axbaxbaxbx456
+ 0: axbaxbaxbx
+ 1: x
+
+/(abc)(?i:(?1))/
+   defabcabcxyz
+ 0: abcabc
+ 1: abc
+   DEFabcABCXYZ
+No match
+
+/(abc)(?:(?i)(?1))/
+   defabcabcxyz
+ 0: abcabc
+ 1: abc
+   DEFabcABCXYZ
+No match
+
+/^(a)\g-2/
+Failed: reference to non-existent subpattern at offset 7
+
+/^(a)\g/
+Failed: a numbered reference must not be zero at offset 5
+
+/^(a)\g{0}/
+Failed: a numbered reference must not be zero at offset 8
+
+/^(a)\g{3/
+Failed: \g is not followed by a braced, angle-bracketed, or quoted name/number or by a plain number at offset 8
+
+/^(a)\g{4a}/
+Failed: reference to non-existent subpattern at offset 9
+
+/^a.b/
+    a\rb
+ 0: a\x0db
+    *** Failers
+No match
+    a\nb
+No match
+
+/.+foo/
+    afoo
+ 0: afoo
+    ** Failers
+No match
+    \r\nfoo
+No match
+    \nfoo
+No match
+
+/.+foo/
+    afoo
+ 0: afoo
+    \nfoo
+ 0: \x0afoo
+    ** Failers
+No match
+    \r\nfoo
+No match
+
+/.+foo/
+    afoo
+ 0: afoo
+    ** Failers
+No match
+    \nfoo
+No match
+    \r\nfoo
+No match
+
+/.+foo/s
+    afoo
+ 0: afoo
+    \r\nfoo
+ 0: \x0d\x0afoo
+    \nfoo
+ 0: \x0afoo
+    
+/^$/mg
+    abc\r\rxyz
+ 0: 
+    abc\n\rxyz  
+ 0: 
+    ** Failers 
+No match
+    abc\r\nxyz
+No match
+
+/(?m)^$/g+
+    abc\r\n\r\n
+ 0: 
+ 0+ \x0d\x0a
+
+/(?m)^$|^\r\n/g+ 
+    abc\r\n\r\n
+ 0: 
+ 0+ \x0d\x0a
+ 0: \x0d\x0a
+ 0+ 
+    
+/(?m)$/g+ 
+    abc\r\n\r\n
+ 0: 
+ 0+ \x0d\x0a\x0d\x0a
+ 0: 
+ 0+ \x0d\x0a
+ 0: 
+ 0+ 
+
+/abc.$/mgx
+    abc1\x0a abc2\x0b abc3\x0c abc4\x0d abc5\x0d\x0a abc6\x85 abc9
+ 0: abc1
+ 0: abc4
+ 0: abc5
+ 0: abc9
+
+/^X/m
+    XABC
+ 0: X
+    ** Failers 
+No match
+    XABC\B
+No match
+
+/(ab|c)(?-1)/BZ
+------------------------------------------------------------------
+        Bra
+        CBra 1
+        ab
+        Alt
+        c
+        Ket
+        Recurse
+        Ket
+        End
+------------------------------------------------------------------
+    abc
+ 0: abc
+ 1: ab
+
+/xy(?+1)(abc)/BZ
+------------------------------------------------------------------
+        Bra
+        xy
+        Recurse
+        CBra 1
+        abc
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+    xyabcabc
+ 0: xyabcabc
+ 1: abc
+    ** Failers
+No match
+    xyabc  
+No match
+    
+/x(?-0)y/
+Failed: a numbered reference must not be zero at offset 5
+
+/x(?-1)y/
+Failed: reference to non-existent subpattern at offset 5
+
+/x(?+0)y/ 
+Failed: a numbered reference must not be zero at offset 5
+
+/x(?+1)y/
+Failed: reference to non-existent subpattern at offset 5
+
+/^(abc)?(?(-1)X|Y)/BZ
+------------------------------------------------------------------
+        Bra
+        ^
+        Brazero
+        CBra 1
+        abc
+        Ket
+        Cond
+      1 Cond ref
+        X
+        Alt
+        Y
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+    abcX
+ 0: abcX
+ 1: abc
+    Y
+ 0: Y
+    ** Failers
+No match
+    abcY   
+No match
+    
+/^((?(+1)X|Y)(abc))+/BZ 
+------------------------------------------------------------------
+        Bra
+        ^
+        CBra 1
+        Cond
+      2 Cond ref
+        X
+        Alt
+        Y
+        Ket
+        CBra 2
+        abc
+        Ket
+        KetRmax
+        Ket
+        End
+------------------------------------------------------------------
+    YabcXabc
+ 0: YabcXabc
+ 1: Xabc
+ 2: abc
+    YabcXabcXabc
+ 0: YabcXabcXabc
+ 1: Xabc
+ 2: abc
+    ** Failers
+No match
+    XabcXabc  
+No match
+
+/(?(-1)a)/BZ
+Failed: reference to non-existent subpattern at offset 6
+
+/((?(-1)a))/BZ
+------------------------------------------------------------------
+        Bra
+        CBra 1
+        Cond
+      1 Cond ref
+        a
+        Ket
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+
+/((?(-2)a))/BZ
+Failed: reference to non-existent subpattern at offset 7
+
+/^(?(+1)X|Y)(.)/BZ
+------------------------------------------------------------------
+        Bra
+        ^
+        Cond
+      1 Cond ref
+        X
+        Alt
+        Y
+        Ket
+        CBra 1
+        Any
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+    Y!
+ 0: Y!
+ 1: !
+
+/(?tom|bon)-\k{A}/
+    tom-tom
+ 0: tom-tom
+ 1: tom
+    bon-bon 
+ 0: bon-bon
+ 1: bon
+    ** Failers
+No match
+    tom-bon  
+No match
+
+/\g{A/ 
+Failed: syntax error in subpattern name (missing terminator) at offset 4
+
+/(?|(abc)|(xyz))/BZ
+------------------------------------------------------------------
+        Bra
+        Bra
+        CBra 1
+        abc
+        Ket
+        Alt
+        CBra 1
+        xyz
+        Ket
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+   >abc<
+ 0: abc
+ 1: abc
+   >xyz< 
+ 0: xyz
+ 1: xyz
+
+/(x)(?|(abc)|(xyz))(x)/BZ
+------------------------------------------------------------------
+        Bra
+        CBra 1
+        x
+        Ket
+        Bra
+        CBra 2
+        abc
+        Ket
+        Alt
+        CBra 2
+        xyz
+        Ket
+        Ket
+        CBra 3
+        x
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+    xabcx
+ 0: xabcx
+ 1: x
+ 2: abc
+ 3: x
+    xxyzx 
+ 0: xxyzx
+ 1: x
+ 2: xyz
+ 3: x
+
+/(x)(?|(abc)(pqr)|(xyz))(x)/BZ
+------------------------------------------------------------------
+        Bra
+        CBra 1
+        x
+        Ket
+        Bra
+        CBra 2
+        abc
+        Ket
+        CBra 3
+        pqr
+        Ket
+        Alt
+        CBra 2
+        xyz
+        Ket
+        Ket
+        CBra 4
+        x
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+    xabcpqrx
+ 0: xabcpqrx
+ 1: x
+ 2: abc
+ 3: pqr
+ 4: x
+    xxyzx 
+ 0: xxyzx
+ 1: x
+ 2: xyz
+ 3: 
+ 4: x
+
+/\H++X/BZ
+------------------------------------------------------------------
+        Bra
+        \H++
+        X
+        Ket
+        End
+------------------------------------------------------------------
+    ** Failers
+No match
+    XXXX
+No match
+    
+/\H+\hY/BZ
+------------------------------------------------------------------
+        Bra
+        \H++
+        \h
+        Y
+        Ket
+        End
+------------------------------------------------------------------
+    XXXX Y 
+ 0: XXXX Y
+
+/\H+ Y/BZ
+------------------------------------------------------------------
+        Bra
+        \H++
+         Y
+        Ket
+        End
+------------------------------------------------------------------
+
+/\h+A/BZ
+------------------------------------------------------------------
+        Bra
+        \h++
+        A
+        Ket
+        End
+------------------------------------------------------------------
+
+/\v*B/BZ
+------------------------------------------------------------------
+        Bra
+        \v*+
+        B
+        Ket
+        End
+------------------------------------------------------------------
+
+/\V+\x0a/BZ
+------------------------------------------------------------------
+        Bra
+        \V++
+        \x0a
+        Ket
+        End
+------------------------------------------------------------------
+
+/A+\h/BZ
+------------------------------------------------------------------
+        Bra
+        A++
+        \h
+        Ket
+        End
+------------------------------------------------------------------
+
+/ *\H/BZ
+------------------------------------------------------------------
+        Bra
+         *+
+        \H
+        Ket
+        End
+------------------------------------------------------------------
+
+/A*\v/BZ
+------------------------------------------------------------------
+        Bra
+        A*+
+        \v
+        Ket
+        End
+------------------------------------------------------------------
+
+/\x0b*\V/BZ
+------------------------------------------------------------------
+        Bra
+        \x0b*+
+        \V
+        Ket
+        End
+------------------------------------------------------------------
+
+/\d+\h/BZ
+------------------------------------------------------------------
+        Bra
+        \d++
+        \h
+        Ket
+        End
+------------------------------------------------------------------
+
+/\d*\v/BZ
+------------------------------------------------------------------
+        Bra
+        \d*+
+        \v
+        Ket
+        End
+------------------------------------------------------------------
+
+/S+\h\S+\v/BZ
+------------------------------------------------------------------
+        Bra
+        S++
+        \h
+        \S++
+        \v
+        Ket
+        End
+------------------------------------------------------------------
+
+/\w{3,}\h\w+\v/BZ
+------------------------------------------------------------------
+        Bra
+        \w{3}
+        \w*+
+        \h
+        \w++
+        \v
+        Ket
+        End
+------------------------------------------------------------------
+
+/\h+\d\h+\w\h+\S\h+\H/BZ
+------------------------------------------------------------------
+        Bra
+        \h++
+        \d
+        \h++
+        \w
+        \h++
+        \S
+        \h++
+        \H
+        Ket
+        End
+------------------------------------------------------------------
+
+/\v+\d\v+\w\v+\S\v+\V/BZ
+------------------------------------------------------------------
+        Bra
+        \v++
+        \d
+        \v++
+        \w
+        \v+
+        \S
+        \v++
+        \V
+        Ket
+        End
+------------------------------------------------------------------
+
+/\H+\h\H+\d/BZ
+------------------------------------------------------------------
+        Bra
+        \H++
+        \h
+        \H+
+        \d
+        Ket
+        End
+------------------------------------------------------------------
+
+/\V+\v\V+\w/BZ
+------------------------------------------------------------------
+        Bra
+        \V++
+        \v
+        \V+
+        \w
+        Ket
+        End
+------------------------------------------------------------------
+
+/\( (?: [^()]* | (?R) )* \)/x
+\J1024(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(00)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)
+ 0: (0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(0(00)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)0)
+
+/[\E]AAA/
+Failed: missing terminating ] for character class at offset 7
+
+/[\Q\E]AAA/
+Failed: missing terminating ] for character class at offset 9
+
+/[^\E]AAA/
+Failed: missing terminating ] for character class at offset 8
+
+/[^\Q\E]AAA/
+Failed: missing terminating ] for character class at offset 10
+
+/[\E^]AAA/
+Failed: missing terminating ] for character class at offset 8
+
+/[\Q\E^]AAA/
+Failed: missing terminating ] for character class at offset 10
+
+/A(*PRUNE)B(*SKIP)C(*THEN)D(*COMMIT)E(*F)F(*FAIL)G(?!)H(*ACCEPT)I/BZ
+------------------------------------------------------------------
+        Bra
+        A
+        *PRUNE
+        B
+        *SKIP
+        C
+        *THEN
+        D
+        *COMMIT
+        E
+        *FAIL
+        F
+        *FAIL
+        G
+        *FAIL
+        H
+        *ACCEPT
+        I
+        Ket
+        End
+------------------------------------------------------------------
+
+/^a+(*FAIL)/C
+    aaaaaa
+--->aaaaaa
+ +0 ^          ^
+ +1 ^          a+
+ +3 ^     ^    (*FAIL)
+ +3 ^    ^     (*FAIL)
+ +3 ^   ^      (*FAIL)
+ +3 ^  ^       (*FAIL)
+ +3 ^ ^        (*FAIL)
+ +3 ^^         (*FAIL)
+No match
+    
+/a+b?c+(*FAIL)/C
+    aaabccc
+--->aaabccc
+ +0 ^           a+
+ +2 ^  ^        b?
+ +4 ^   ^       c+
+ +6 ^      ^    (*FAIL)
+ +6 ^     ^     (*FAIL)
+ +6 ^    ^      (*FAIL)
+ +4 ^  ^        c+
+ +2 ^ ^         b?
+ +4 ^ ^         c+
+ +2 ^^          b?
+ +4 ^^          c+
+ +0  ^          a+
+ +2  ^ ^        b?
+ +4  ^  ^       c+
+ +6  ^     ^    (*FAIL)
+ +6  ^    ^     (*FAIL)
+ +6  ^   ^      (*FAIL)
+ +4  ^ ^        c+
+ +2  ^^         b?
+ +4  ^^         c+
+ +0   ^         a+
+ +2   ^^        b?
+ +4   ^ ^       c+
+ +6   ^    ^    (*FAIL)
+ +6   ^   ^     (*FAIL)
+ +6   ^  ^      (*FAIL)
+ +4   ^^        c+
+No match
+
+/a+b?(*PRUNE)c+(*FAIL)/C
+    aaabccc
+--->aaabccc
+ +0 ^           a+
+ +2 ^  ^        b?
+ +4 ^   ^       (*PRUNE)
++12 ^   ^       c+
++14 ^      ^    (*FAIL)
++14 ^     ^     (*FAIL)
++14 ^    ^      (*FAIL)
+ +0  ^          a+
+ +2  ^ ^        b?
+ +4  ^  ^       (*PRUNE)
++12  ^  ^       c+
++14  ^     ^    (*FAIL)
++14  ^    ^     (*FAIL)
++14  ^   ^      (*FAIL)
+ +0   ^         a+
+ +2   ^^        b?
+ +4   ^ ^       (*PRUNE)
++12   ^ ^       c+
++14   ^    ^    (*FAIL)
++14   ^   ^     (*FAIL)
++14   ^  ^      (*FAIL)
+No match
+
+/a+b?(*COMMIT)c+(*FAIL)/C
+    aaabccc
+--->aaabccc
+ +0 ^           a+
+ +2 ^  ^        b?
+ +4 ^   ^       (*COMMIT)
++13 ^   ^       c+
++15 ^      ^    (*FAIL)
++15 ^     ^     (*FAIL)
++15 ^    ^      (*FAIL)
+No match
+    
+/a+b?(*SKIP)c+(*FAIL)/C
+    aaabcccaaabccc
+--->aaabcccaaabccc
+ +0 ^                  a+
+ +2 ^  ^               b?
+ +4 ^   ^              (*SKIP)
++11 ^   ^              c+
++13 ^      ^           (*FAIL)
++13 ^     ^            (*FAIL)
++13 ^    ^             (*FAIL)
+ +0        ^           a+
+ +2        ^  ^        b?
+ +4        ^   ^       (*SKIP)
++11        ^   ^       c+
++13        ^      ^    (*FAIL)
++13        ^     ^     (*FAIL)
++13        ^    ^      (*FAIL)
+No match
+
+/a+b?(*THEN)c+(*FAIL)/C
+    aaabccc
+--->aaabccc
+ +0 ^           a+
+ +2 ^  ^        b?
+ +4 ^   ^       (*THEN)
++11 ^   ^       c+
++13 ^      ^    (*FAIL)
++13 ^     ^     (*FAIL)
++13 ^    ^      (*FAIL)
+ +0  ^          a+
+ +2  ^ ^        b?
+ +4  ^  ^       (*THEN)
++11  ^  ^       c+
++13  ^     ^    (*FAIL)
++13  ^    ^     (*FAIL)
++13  ^   ^      (*FAIL)
+ +0   ^         a+
+ +2   ^^        b?
+ +4   ^ ^       (*THEN)
++11   ^ ^       c+
++13   ^    ^    (*FAIL)
++13   ^   ^     (*FAIL)
++13   ^  ^      (*FAIL)
+No match
+    
+/a(*MARK)b/ 
+Failed: (*MARK) must have an argument at offset 7
+
+/(?i:A{1,}\6666666666)/
+Failed: number is too big at offset 19
+
+/\g6666666666/
+Failed: number is too big at offset 11
+
+/[\g6666666666]/BZ
+------------------------------------------------------------------
+        Bra
+        [6g]
+        Ket
+        End
+------------------------------------------------------------------
+
+/(?1)\c[/
+Failed: reference to non-existent subpattern at offset 3
+
+/.+A/
+    \r\nA
+No match
+    
+/\nA/
+    \r\nA 
+ 0: \x0aA
+
+/[\r\n]A/
+    \r\nA 
+ 0: \x0aA
+
+/(\r|\n)A/
+    \r\nA 
+ 0: \x0aA
+ 1: \x0a
+
+/a(*CR)b/
+Failed: (*VERB) not recognized or malformed at offset 5
+
+/(*CR)a.b/
+    a\nb
+ 0: a\x0ab
+    ** Failers
+No match
+    a\rb  
+No match
+
+/(*CR)a.b/
+    a\nb
+ 0: a\x0ab
+    ** Failers
+No match
+    a\rb  
+No match
+
+/(*LF)a.b/
+    a\rb
+ 0: a\x0db
+    ** Failers
+No match
+    a\nb  
+No match
+
+/(*CRLF)a.b/
+    a\rb
+ 0: a\x0db
+    a\nb  
+ 0: a\x0ab
+    ** Failers
+No match
+    a\r\nb  
+No match
+
+/(*ANYCRLF)a.b/
+    ** Failers
+No match
+    a\rb
+No match
+    a\nb  
+No match
+    a\r\nb  
+No match
+
+/(*ANY)a.b/
+    ** Failers
+No match
+    a\rb
+No match
+    a\nb  
+No match
+    a\r\nb  
+No match
+    a\x85b 
+No match
+    
+/(*ANY).*/g
+    abc\r\ndef
+ 0: abc
+ 0: 
+ 0: def
+ 0: 
+
+/(*ANYCRLF).*/g
+    abc\r\ndef
+ 0: abc
+ 0: 
+ 0: def
+ 0: 
+
+/(*CRLF).*/g
+    abc\r\ndef
+ 0: abc
+ 0: 
+ 0: def
+ 0: 
+
+/a\Rb/I
+Capturing subpattern count = 0
+Options: bsr_anycrlf
+First char = 'a'
+Need char = 'b'
+    a\rb
+ 0: a\x0db
+    a\nb
+ 0: a\x0ab
+    a\r\nb
+ 0: a\x0d\x0ab
+    ** Failers
+No match
+    a\x85b
+No match
+    a\x0bb     
+No match
+
+/a\Rb/I
+Capturing subpattern count = 0
+Options: bsr_unicode
+First char = 'a'
+Need char = 'b'
+    a\rb
+ 0: a\x0db
+    a\nb
+ 0: a\x0ab
+    a\r\nb
+ 0: a\x0d\x0ab
+    a\x85b
+ 0: a\x85b
+    a\x0bb     
+ 0: a\x0bb
+    ** Failers 
+No match
+    a\x85b\
+No match
+    a\x0bb\
+No match
+    
+/a\R?b/I
+Capturing subpattern count = 0
+Options: bsr_anycrlf
+First char = 'a'
+Need char = 'b'
+    a\rb
+ 0: a\x0db
+    a\nb
+ 0: a\x0ab
+    a\r\nb
+ 0: a\x0d\x0ab
+    ** Failers
+No match
+    a\x85b
+No match
+    a\x0bb     
+No match
+
+/a\R?b/I
+Capturing subpattern count = 0
+Options: bsr_unicode
+First char = 'a'
+Need char = 'b'
+    a\rb
+ 0: a\x0db
+    a\nb
+ 0: a\x0ab
+    a\r\nb
+ 0: a\x0d\x0ab
+    a\x85b
+ 0: a\x85b
+    a\x0bb     
+ 0: a\x0bb
+    ** Failers 
+No match
+    a\x85b\
+No match
+    a\x0bb\
+No match
+    
+/a\R{2,4}b/I
+Capturing subpattern count = 0
+Options: bsr_anycrlf
+First char = 'a'
+Need char = 'b'
+    a\r\n\nb
+ 0: a\x0d\x0a\x0ab
+    a\n\r\rb
+ 0: a\x0a\x0d\x0db
+    a\r\n\r\n\r\n\r\nb
+ 0: a\x0d\x0a\x0d\x0a\x0d\x0a\x0d\x0ab
+    ** Failers
+No match
+    a\x85\85b
+No match
+    a\x0b\0bb     
+No match
+
+/a\R{2,4}b/I
+Capturing subpattern count = 0
+Options: bsr_unicode
+First char = 'a'
+Need char = 'b'
+    a\r\rb
+ 0: a\x0d\x0db
+    a\n\n\nb
+ 0: a\x0a\x0a\x0ab
+    a\r\n\n\r\rb
+ 0: a\x0d\x0a\x0a\x0d\x0db
+    a\x85\85b
+No match
+    a\x0b\0bb     
+No match
+    ** Failers 
+No match
+    a\r\r\r\r\rb 
+No match
+    a\x85\85b\
+No match
+    a\x0b\0bb\
+No match
+ 
+/(*BSR_ANYCRLF)a\Rb/I
+Capturing subpattern count = 0
+Options: bsr_anycrlf
+First char = 'a'
+Need char = 'b'
+    a\nb
+ 0: a\x0ab
+    a\rb 
+ 0: a\x0db
+
+/(*BSR_UNICODE)a\Rb/I
+Capturing subpattern count = 0
+Options: bsr_unicode
+First char = 'a'
+Need char = 'b'
+    a\x85b
+ 0: a\x85b
+
+/(*BSR_ANYCRLF)(*CRLF)a\Rb/I
+Capturing subpattern count = 0
+Options: bsr_anycrlf
+Forced newline sequence: CRLF
+First char = 'a'
+Need char = 'b'
+    a\nb
+ 0: a\x0ab
+    a\rb 
+ 0: a\x0db
+
+/(*CRLF)(*BSR_UNICODE)a\Rb/I
+Capturing subpattern count = 0
+Options: bsr_unicode
+Forced newline sequence: CRLF
+First char = 'a'
+Need char = 'b'
+    a\x85b
+ 0: a\x85b
+
+/(*CRLF)(*BSR_ANYCRLF)(*CR)ab/I
+Capturing subpattern count = 0
+Options: bsr_anycrlf
+Forced newline sequence: CR
+First char = 'a'
+Need char = 'b'
+
+/(?)(?&)/
+Failed: subpattern name expected at offset 9
+
+/(?)(?&a)/
+Failed: reference to non-existent subpattern at offset 12
+
+/(?)(?&aaaaaaaaaaaaaaaaaaaaaaa)/
+Failed: reference to non-existent subpattern at offset 32
+
+/(?+-a)/
+Failed: digit expected after (?+ at offset 3
+
+/(?-+a)/
+Failed: unrecognized character after (? or (?- at offset 3
+
+/(?(-1))/
+Failed: reference to non-existent subpattern at offset 6
+
+/(?(+10))/
+Failed: reference to non-existent subpattern at offset 7
+
+/(?(10))/
+Failed: reference to non-existent subpattern at offset 6
+
+/(?(+2))()()/
+
+/(?(2))()()/
+
+/\k''/
+Failed: subpattern name expected at offset 3
+
+/\k<>/
+Failed: subpattern name expected at offset 3
+
+/\k{}/
+Failed: subpattern name expected at offset 3
+
+/\k/
+Failed: \k is not followed by a braced, angle-bracketed, or quoted name at offset 2
+
+/\kabc/
+Failed: \k is not followed by a braced, angle-bracketed, or quoted name at offset 5
+
+/(?P=)/
+Failed: subpattern name expected at offset 4
+
+/(?P>)/
+Failed: subpattern name expected at offset 4
+
+/(?!\w)(?R)/
+Failed: recursive call could loop indefinitely at offset 9
+
+/(?=\w)(?R)/
+Failed: recursive call could loop indefinitely at offset 9
+
+/(?x|y){0}z/
+    xzxx
+ 0: xz
+    yzyy 
+ 0: yz
+    ** Failers
+No match
+    xxz  
+No match
+
+/(\3)(\1)(a)/
+    cat
+No match
+
+/(\3)(\1)(a)/
+    cat
+ 0: a
+ 1: 
+ 2: 
+ 3: a
+    
+/TA]/
+    The ACTA] comes 
+ 0: TA]
+
+/TA]/
+Failed: ] is an invalid data character in JavaScript compatibility mode at offset 2
+
+/(?2)[]a()b](abc)/
+Failed: reference to non-existent subpattern at offset 3
+
+/(?2)[^]a()b](abc)/
+Failed: reference to non-existent subpattern at offset 3
+
+/(?1)[]a()b](abc)/
+    abcbabc
+ 0: abcbabc
+ 1: abc
+    ** Failers 
+No match
+    abcXabc
+No match
+
+/(?1)[^]a()b](abc)/
+    abcXabc
+ 0: abcXabc
+ 1: abc
+    ** Failers 
+No match
+    abcbabc
+No match
+
+/(?2)[]a()b](abc)(xyz)/
+    xyzbabcxyz
+ 0: xyzbabcxyz
+ 1: abc
+ 2: xyz
+
+/(?&N)[]a(?)](?abc)/
+Failed: reference to non-existent subpattern at offset 4
+
+/(?&N)[]a(?)](abc)/
+Failed: reference to non-existent subpattern at offset 4
+
+/a[]b/
+Failed: missing terminating ] for character class at offset 4
+
+/a[^]b/
+Failed: missing terminating ] for character class at offset 5
+
+/a[]b/
+    ** Failers
+No match
+    ab
+No match
+
+/a[]+b/
+    ** Failers
+No match
+    ab 
+No match
+
+/a[]*+b/
+    ** Failers
+No match
+    ab 
+No match
+
+/a[^]b/
+    aXb
+ 0: aXb
+    a\nb 
+ 0: a\x0ab
+    ** Failers
+No match
+    ab  
+No match
+    
+/a[^]+b/ 
+    aXb
+ 0: aXb
+    a\nX\nXb 
+ 0: a\x0aX\x0aXb
+    ** Failers
+No match
+    ab  
+No match
+
+/a(?!)+b/
+Failed: nothing to repeat at offset 5
+
+/a(*FAIL)+b/
+Failed: nothing to repeat at offset 8
+
+/(abc|pqr|123){0}[xyz]/SI
+Capturing subpattern count = 1
+No options
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: x y z 
+
+/(?(?=.*b)b|^)/CI
+Capturing subpattern count = 0
+Options:
+No first char
+No need char
+   adc
+--->adc
+ +0 ^       (?(?=.*b)b|^)
+ +2 ^       (?=.*b)
+ +5 ^       .*
+ +7 ^  ^    b
+ +7 ^ ^     b
+ +7 ^^      b
+ +7 ^       b
++11 ^       ^
++12 ^       )
++13 ^       
+ 0: 
+   abc 
+--->abc
+ +0 ^       (?(?=.*b)b|^)
+ +2 ^       (?=.*b)
+ +5 ^       .*
+ +7 ^  ^    b
+ +7 ^ ^     b
+ +7 ^^      b
+ +8 ^ ^     )
+ +9 ^       b
+ +0  ^      (?(?=.*b)b|^)
+ +2  ^      (?=.*b)
+ +5  ^      .*
+ +7  ^ ^    b
+ +7  ^^     b
+ +7  ^      b
+ +8  ^^     )
+ +9  ^      b
++10  ^^     |
++13  ^^     
+ 0: b
+   
+/(?(?=b).*b|^d)/I
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+
+/(?(?=.*b).*b|^d)/I
+Capturing subpattern count = 0
+No options
+First char at start or follows newline
+No need char
+
+/xyz/C
+  xyz 
+--->xyz
+ +0 ^       x
+ +1 ^^      y
+ +2 ^ ^     z
+ +3 ^  ^    
+ 0: xyz
+  abcxyz 
+--->abcxyz
+ +0    ^       x
+ +1    ^^      y
+ +2    ^ ^     z
+ +3    ^  ^    
+ 0: xyz
+  abcxyz\Y
+--->abcxyz
+ +0 ^          x
+ +0  ^         x
+ +0   ^        x
+ +0    ^       x
+ +1    ^^      y
+ +2    ^ ^     z
+ +3    ^  ^    
+ 0: xyz
+  ** Failers 
+No match
+  abc
+No match
+  abc\Y
+--->abc
+ +0 ^       x
+ +0  ^      x
+ +0   ^     x
+ +0    ^    x
+No match
+  abcxypqr  
+No match
+  abcxypqr\Y  
+--->abcxypqr
+ +0 ^            x
+ +0  ^           x
+ +0   ^          x
+ +0    ^         x
+ +1    ^^        y
+ +2    ^ ^       z
+ +0     ^        x
+ +0      ^       x
+ +0       ^      x
+ +0        ^     x
+ +0         ^    x
+No match
+  
+/(*NO_START_OPT)xyz/C
+  abcxyz 
+--->abcxyz
++15 ^          x
++15  ^         x
++15   ^        x
++15    ^       x
++16    ^^      y
++17    ^ ^     z
++18    ^  ^    
+ 0: xyz
+
+/xyz/CY
+  abcxyz 
+--->abcxyz
+ +0 ^          x
+ +0  ^         x
+ +0   ^        x
+ +0    ^       x
+ +1    ^^      y
+ +2    ^ ^     z
+ +3    ^  ^    
+ 0: xyz
+
+/^"((?(?=[a])[^"])|b)*"$/C
+    "ab"
+--->"ab"
+ +0 ^        ^
+ +1 ^        "
+ +2 ^^       ((?(?=[a])[^"])|b)*
+ +3 ^^       (?(?=[a])[^"])
+ +5 ^^       (?=[a])
+ +8 ^^       [a]
++11 ^ ^      )
++12 ^^       [^"]
++16 ^ ^      )
++17 ^ ^      |
+ +3 ^ ^      (?(?=[a])[^"])
+ +5 ^ ^      (?=[a])
+ +8 ^ ^      [a]
++17 ^ ^      |
++21 ^ ^      "
++18 ^ ^      b
++19 ^  ^     )
+ +3 ^  ^     (?(?=[a])[^"])
+ +5 ^  ^     (?=[a])
+ +8 ^  ^     [a]
++17 ^  ^     |
++21 ^  ^     "
++22 ^   ^    $
++23 ^   ^    
+ 0: "ab"
+ 1: 
+
+/^"((?(?=[a])[^"])|b)*"$/
+    "ab"
+ 0: "ab"
+ 1: 
+
+/^X(?5)(a)(?|(b)|(q))(c)(d)Y/
+Failed: reference to non-existent subpattern at offset 5
+
+/^X(?&N)(a)(?|(b)|(q))(c)(d)(?Y)/
+    XYabcdY
+ 0: XYabcdY
+ 1: a
+ 2: b
+ 3: c
+ 4: d
+ 5: Y
+ 
+/Xa{2,4}b/
+    X\P
+Partial match: X
+    Xa\P
+Partial match: Xa
+    Xaa\P 
+Partial match: Xaa
+    Xaaa\P
+Partial match: Xaaa
+    Xaaaa\P 
+Partial match: Xaaaa
+    
+/Xa{2,4}?b/
+    X\P
+Partial match: X
+    Xa\P
+Partial match: Xa
+    Xaa\P 
+Partial match: Xaa
+    Xaaa\P
+Partial match: Xaaa
+    Xaaaa\P 
+Partial match: Xaaaa
+    
+/Xa{2,4}+b/
+    X\P
+Partial match: X
+    Xa\P
+Partial match: Xa
+    Xaa\P 
+Partial match: Xaa
+    Xaaa\P
+Partial match: Xaaa
+    Xaaaa\P 
+Partial match: Xaaaa
+    
+/X\d{2,4}b/
+    X\P
+Partial match: X
+    X3\P
+Partial match: X3
+    X33\P 
+Partial match: X33
+    X333\P
+Partial match: X333
+    X3333\P 
+Partial match: X3333
+    
+/X\d{2,4}?b/
+    X\P
+Partial match: X
+    X3\P
+Partial match: X3
+    X33\P 
+Partial match: X33
+    X333\P
+Partial match: X333
+    X3333\P 
+Partial match: X3333
+    
+/X\d{2,4}+b/
+    X\P
+Partial match: X
+    X3\P
+Partial match: X3
+    X33\P 
+Partial match: X33
+    X333\P
+Partial match: X333
+    X3333\P 
+Partial match: X3333
+    
+/X\D{2,4}b/
+    X\P
+Partial match: X
+    Xa\P
+Partial match: Xa
+    Xaa\P 
+Partial match: Xaa
+    Xaaa\P
+Partial match: Xaaa
+    Xaaaa\P 
+Partial match: Xaaaa
+    
+/X\D{2,4}?b/
+    X\P
+Partial match: X
+    Xa\P
+Partial match: Xa
+    Xaa\P 
+Partial match: Xaa
+    Xaaa\P
+Partial match: Xaaa
+    Xaaaa\P 
+Partial match: Xaaaa
+    
+/X\D{2,4}+b/
+    X\P
+Partial match: X
+    Xa\P
+Partial match: Xa
+    Xaa\P 
+Partial match: Xaa
+    Xaaa\P
+Partial match: Xaaa
+    Xaaaa\P 
+Partial match: Xaaaa
+    
+/X[abc]{2,4}b/
+    X\P
+Partial match: X
+    Xa\P
+Partial match: Xa
+    Xaa\P 
+Partial match: Xaa
+    Xaaa\P
+Partial match: Xaaa
+    Xaaaa\P 
+Partial match: Xaaaa
+    
+/X[abc]{2,4}?b/
+    X\P
+Partial match: X
+    Xa\P
+Partial match: Xa
+    Xaa\P 
+Partial match: Xaa
+    Xaaa\P
+Partial match: Xaaa
+    Xaaaa\P 
+Partial match: Xaaaa
+    
+/X[abc]{2,4}+b/
+    X\P
+Partial match: X
+    Xa\P
+Partial match: Xa
+    Xaa\P 
+Partial match: Xaa
+    Xaaa\P
+Partial match: Xaaa
+    Xaaaa\P 
+Partial match: Xaaaa
+    
+/X[^a]{2,4}b/
+    X\P
+Partial match: X
+    Xz\P
+Partial match: Xz
+    Xzz\P 
+Partial match: Xzz
+    Xzzz\P
+Partial match: Xzzz
+    Xzzzz\P 
+Partial match: Xzzzz
+    
+/X[^a]{2,4}?b/
+    X\P
+Partial match: X
+    Xz\P
+Partial match: Xz
+    Xzz\P 
+Partial match: Xzz
+    Xzzz\P
+Partial match: Xzzz
+    Xzzzz\P 
+Partial match: Xzzzz
+    
+/X[^a]{2,4}+b/
+    X\P
+Partial match: X
+    Xz\P
+Partial match: Xz
+    Xzz\P 
+Partial match: Xzz
+    Xzzz\P
+Partial match: Xzzz
+    Xzzzz\P 
+Partial match: Xzzzz
+    
+/(Y)X\1{2,4}b/
+    YX\P
+Partial match: YX
+    YXY\P
+Partial match: YXY
+    YXYY\P 
+Partial match: YXYY
+    YXYYY\P
+Partial match: YXYYY
+    YXYYYY\P 
+Partial match: YXYYYY
+    
+/(Y)X\1{2,4}?b/
+    YX\P
+Partial match: YX
+    YXY\P
+Partial match: YXY
+    YXYY\P 
+Partial match: YXYY
+    YXYYY\P
+Partial match: YXYYY
+    YXYYYY\P 
+Partial match: YXYYYY
+    
+/(Y)X\1{2,4}+b/
+    YX\P
+Partial match: YX
+    YXY\P
+Partial match: YXY
+    YXYY\P 
+Partial match: YXYY
+    YXYYY\P
+Partial match: YXYYY
+    YXYYYY\P 
+Partial match: YXYYYY
+    
+/\++\KZ|\d+X|9+Y/
+    ++++123999\P
+Partial match: 123999
+    ++++123999Y\P
+ 0: 999Y
+    ++++Z1234\P 
+ 0: Z
+
+/Z(*F)/
+    Z\P
+No match
+    ZA\P 
+No match
+    
+/Z(?!)/
+    Z\P 
+No match
+    ZA\P 
+No match
+
+/dog(sbody)?/
+    dogs\P
+ 0: dog
+    dogs\P\P 
+Partial match: dogs
+    
+/dog(sbody)??/
+    dogs\P
+ 0: dog
+    dogs\P\P 
+ 0: dog
+
+/dog|dogsbody/
+    dogs\P
+ 0: dog
+    dogs\P\P 
+ 0: dog
+ 
+/dogsbody|dog/
+    dogs\P
+ 0: dog
+    dogs\P\P 
+Partial match: dogs
+
+/\bthe cat\b/
+    the cat\P
+ 0: the cat
+    the cat\P\P
+Partial match: the cat
+
+/abc/
+   abc\P
+ 0: abc
+   abc\P\P
+ 0: abc
+   
+/abc\K123/
+    xyzabc123pqr
+ 0: 123
+    xyzabc12\P
+Partial match: abc12
+    xyzabc12\P\P
+Partial match: abc12
+    
+/(?<=abc)123/
+    xyzabc123pqr 
+ 0: 123
+    xyzabc12\P
+Partial match at offset 6: abc12
+    xyzabc12\P\P
+Partial match at offset 6: abc12
+
+/\babc\b/
+    +++abc+++
+ 0: abc
+    +++ab\P
+Partial match at offset 3: +ab
+    +++ab\P\P  
+Partial match at offset 3: +ab
+
+/(?&word)(?&element)(?(DEFINE)(?<[^m][^>]>[^<])(?\w*+))/BZ
+------------------------------------------------------------------
+        Bra
+        Recurse
+        Recurse
+        Cond
+        Cond def
+        CBra 1
+        <
+        [^m]
+        [^>]
+        >
+        [^<]
+        Ket
+        CBra 2
+        \w*+
+        Ket
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+
+/(?&word)(?&element)(?(DEFINE)(?<[^\d][^>]>[^<])(?\w*+))/BZ
+------------------------------------------------------------------
+        Bra
+        Recurse
+        Recurse
+        Cond
+        Cond def
+        CBra 1
+        <
+        [\x00-/:-\xff] (neg)
+        [^>]
+        >
+        [^<]
+        Ket
+        CBra 2
+        \w*+
+        Ket
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+
+/(ab)(x(y)z(cd(*ACCEPT)))pq/BZ
+------------------------------------------------------------------
+        Bra
+        CBra 1
+        ab
+        Ket
+        CBra 2
+        x
+        CBra 3
+        y
+        Ket
+        z
+        CBra 4
+        cd
+        Close 4
+        Close 2
+        *ACCEPT
+        Ket
+        Ket
+        pq
+        Ket
+        End
+------------------------------------------------------------------
+
+/abc\K/+
+    abcdef
+ 0: 
+ 0+ def
+    abcdef\N\N
+ 0: 
+ 0+ def
+    xyzabcdef\N\N
+ 0: 
+ 0+ def
+    ** Failers
+No match
+    abcdef\N 
+No match
+    xyzabcdef\N
+No match
+    
+/^(?:(?=abc)|abc\K)/+
+    abcdef
+ 0: 
+ 0+ abcdef
+    abcdef\N\N 
+ 0: 
+ 0+ def
+    ** Failers 
+No match
+    abcdef\N 
+No match
+
+/a?b?/+
+    xyz
+ 0: 
+ 0+ xyz
+    xyzabc
+ 0: 
+ 0+ xyzabc
+    xyzabc\N
+ 0: ab
+ 0+ c
+    xyzabc\N\N
+ 0: 
+ 0+ yzabc
+    xyz\N\N    
+ 0: 
+ 0+ yz
+    ** Failers 
+ 0: 
+ 0+ ** Failers
+    xyz\N 
+No match
+
+/^a?b?/+
+    xyz
+ 0: 
+ 0+ xyz
+    xyzabc
+ 0: 
+ 0+ xyzabc
+    ** Failers 
+ 0: 
+ 0+ ** Failers
+    xyzabc\N
+No match
+    xyzabc\N\N
+No match
+    xyz\N\N    
+No match
+    xyz\N 
+No match
+    
+/^(?a|b\gc)/
+    aaaa
+ 0: a
+ 1: a
+    bacxxx
+ 0: bac
+ 1: bac
+    bbaccxxx 
+ 0: bbacc
+ 1: bbacc
+    bbbacccxx
+ 0: bbbaccc
+ 1: bbbaccc
+
+/^(?a|b\g'name'c)/
+    aaaa
+ 0: a
+ 1: a
+    bacxxx
+ 0: bac
+ 1: bac
+    bbaccxxx 
+ 0: bbacc
+ 1: bbacc
+    bbbacccxx
+ 0: bbbaccc
+ 1: bbbaccc
+
+/^(a|b\g<1>c)/
+    aaaa
+ 0: a
+ 1: a
+    bacxxx
+ 0: bac
+ 1: bac
+    bbaccxxx 
+ 0: bbacc
+ 1: bbacc
+    bbbacccxx
+ 0: bbbaccc
+ 1: bbbaccc
+
+/^(a|b\g'1'c)/
+    aaaa
+ 0: a
+ 1: a
+    bacxxx
+ 0: bac
+ 1: bac
+    bbaccxxx 
+ 0: bbacc
+ 1: bbacc
+    bbbacccxx
+ 0: bbbaccc
+ 1: bbbaccc
+
+/^(a|b\g'-1'c)/
+    aaaa
+ 0: a
+ 1: a
+    bacxxx
+ 0: bac
+ 1: bac
+    bbaccxxx 
+ 0: bbacc
+ 1: bbacc
+    bbbacccxx
+ 0: bbbaccc
+ 1: bbbaccc
+
+/(^(a|b\g<-1>c))/
+    aaaa
+ 0: a
+ 1: a
+ 2: a
+    bacxxx
+ 0: bac
+ 1: bac
+ 2: bac
+    bbaccxxx 
+ 0: bbacc
+ 1: bbacc
+ 2: bbacc
+    bbbacccxx
+ 0: bbbaccc
+ 1: bbbaccc
+ 2: bbbaccc
+
+/(?-i:\g)(?i:(?a))/
+    XaaX
+ 0: aa
+ 1: a
+    XAAX 
+ 0: AA
+ 1: A
+
+/(?i:\g)(?-i:(?a))/
+    XaaX
+ 0: aa
+ 1: a
+    ** Failers 
+No match
+    XAAX 
+No match
+
+/(?-i:\g<+1>)(?i:(a))/
+    XaaX
+ 0: aa
+ 1: a
+    XAAX 
+ 0: AA
+ 1: A
+
+/(?=(?(?#simplesyntax)\$(?[a-zA-Z_\x{7f}-\x{ff}][a-zA-Z0-9_\x{7f}-\x{ff}]*)(?:\[(?[a-zA-Z0-9_\x{7f}-\x{ff}]+|\$\g)\]|->\g(\(.*?\))?)?|(?#simple syntax withbraces)\$\{(?:\g(?\[(?:\g|'(?:\\.|[^'\\])*'|"(?:\g|\\.|[^"\\])*")\])?|\g|\$\{\g\})\}|(?#complexsyntax)\{(?\$(?\g(\g*|\(.*?\))?)(?:->\g)*|\$\g|\$\{\g\})\}))\{/
+
+/(?a|b|c)\g*/
+   abc
+ 0: abc
+ 1: a
+   accccbbb 
+ 0: accccbbb
+ 1: a
+
+/^X(?7)(a)(?|(b)|(q)(r)(s))(c)(d)(Y)/
+    XYabcdY
+ 0: XYabcdY
+ 1: a
+ 2: b
+ 3: 
+ 4: 
+ 5: c
+ 6: d
+ 7: Y
+
+/(?<=b(?1)|zzz)(a)/
+    xbaax
+ 0: a
+ 1: a
+    xzzzax 
+ 0: a
+ 1: a
+
+/(a)(?<=b\1)/
+Failed: lookbehind assertion is not fixed length at offset 10
+
+/(a)(?<=b+(?1))/
+Failed: lookbehind assertion is not fixed length at offset 13
+
+/(a+)(?<=b(?1))/
+Failed: lookbehind assertion is not fixed length at offset 14
+
+/(a(?<=b(?1)))/
+Failed: lookbehind assertion is not fixed length at offset 13
+
+/(?<=b(?1))xyz/
+Failed: reference to non-existent subpattern at offset 8
+
+/(?<=b(?1))xyz(b+)pqrstuvew/
+Failed: lookbehind assertion is not fixed length at offset 26
+
+/(a|bc)\1/SI
+Capturing subpattern count = 1
+Max back reference = 1
+No options
+No first char
+No need char
+Subject length lower bound = 2
+Starting byte set: a b 
+
+/(a|bc)\1{2,3}/SI
+Capturing subpattern count = 1
+Max back reference = 1
+No options
+No first char
+No need char
+Subject length lower bound = 3
+Starting byte set: a b 
+
+/(a|bc)(?1)/SI
+Capturing subpattern count = 1
+No options
+No first char
+No need char
+Subject length lower bound = 2
+Starting byte set: a b 
+
+/(a|b\1)(a|b\1)/SI
+Capturing subpattern count = 2
+Max back reference = 1
+No options
+No first char
+No need char
+Subject length lower bound = 2
+Starting byte set: a b 
+
+/(a|b\1){2}/SI
+Capturing subpattern count = 1
+Max back reference = 1
+No options
+No first char
+No need char
+Subject length lower bound = 2
+Starting byte set: a b 
+
+/(a|bbbb\1)(a|bbbb\1)/SI
+Capturing subpattern count = 2
+Max back reference = 1
+No options
+No first char
+No need char
+Subject length lower bound = 2
+Starting byte set: a b 
+
+/(a|bbbb\1){2}/SI
+Capturing subpattern count = 1
+Max back reference = 1
+No options
+No first char
+No need char
+Subject length lower bound = 2
+Starting byte set: a b 
+
+/^From +([^ ]+) +[a-zA-Z][a-zA-Z][a-zA-Z] +[a-zA-Z][a-zA-Z][a-zA-Z] +[0-9]?[0-9] +[0-9][0-9]:[0-9][0-9]/SI
+Capturing subpattern count = 1
+Options: anchored
+No first char
+Need char = ':'
+Subject length lower bound = 22
+No set of starting bytes
+
+/]{0,})>]{0,})>([\d]{0,}\.)(.*)((
([\w\W\s\d][^<>]{0,})|[\s]{0,}))<\/a><\/TD>]{0,})>([\w\W\s\d][^<>]{0,})<\/TD>]{0,})>([\w\W\s\d][^<>]{0,})<\/TD><\/TR>/isIS
+Capturing subpattern count = 11
+Options: caseless dotall
+First char = '<'
+Need char = '>'
+Subject length lower bound = 47
+No set of starting bytes
+
+"(?>.*/)foo"SI
+Capturing subpattern count = 0
+No options
+No first char
+Need char = 'o'
+Subject length lower bound = 4
+No set of starting bytes
+
+/(?(?=[^a-z]+[a-z])  \d{2}-[a-z]{3}-\d{2}  |  \d{2}-\d{2}-\d{2} ) /xSI
+Capturing subpattern count = 0
+Options: extended
+No first char
+Need char = '-'
+Subject length lower bound = 8
+No set of starting bytes
+
+/(?:(?:(?:(?:(?:(?:(?:(?:(?:(a|b|c))))))))))/iSI
+Capturing subpattern count = 1
+Options: caseless
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: A B C a b c 
+
+/(?:c|d)(?:)(?:aaaaaaaa(?:)(?:bbbbbbbb)(?:bbbbbbbb(?:))(?:bbbbbbbb(?:)(?:bbbbbbbb)))/SI
+Capturing subpattern count = 0
+No options
+No first char
+Need char = 'b'
+Subject length lower bound = 41
+Starting byte set: c d 
+
+/A)|(?B))/I
+Capturing subpattern count = 1
+Named capturing subpatterns:
+  a   1
+  a   1
+No options
+No first char
+No need char
+    AB\Ca
+ 0: A
+ 1: A
+  C A (1) a
+    BA\Ca
+ 0: B
+ 1: B
+  C B (1) a
+
+/(?|(?A)|(?B))/ 
+Failed: different names for subpatterns of the same number are not allowed at offset 15
+
+/(?:a(? (?')|(?")) |
+    b(? (?')|(?")) ) 
+    (?('quote')[a-z]+|[0-9]+)/JIx
+Capturing subpattern count = 6
+Named capturing subpatterns:
+  apostrophe   2
+  apostrophe   5
+  quote        1
+  quote        4
+  realquote    3
+  realquote    6
+Options: extended dupnames
+No first char
+No need char
+    a"aaaaa
+ 0: a"aaaaa
+ 1: "
+ 2: 
+ 3: "
+    b"aaaaa 
+ 0: b"aaaaa
+ 1: 
+ 2: 
+ 3: 
+ 4: "
+ 5: 
+ 6: "
+    ** Failers 
+No match
+    b"11111
+No match
+    a"11111 
+No match
+    
+/^(?|(a)(b)(c)(?d)|(?e)) (?('D')X|Y)/JDZx
+------------------------------------------------------------------
+        Bra
+        ^
+        Bra
+        CBra 1
+        a
+        Ket
+        CBra 2
+        b
+        Ket
+        CBra 3
+        c
+        Ket
+        CBra 4
+        d
+        Ket
+        Alt
+        CBra 1
+        e
+        Ket
+        Ket
+        Cond
+      4 Cond nref
+        X
+        Alt
+        Y
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 4
+Named capturing subpatterns:
+  D   4
+  D   1
+Options: anchored extended dupnames
+No first char
+No need char
+    abcdX
+ 0: abcdX
+ 1: a
+ 2: b
+ 3: c
+ 4: d
+    eX
+ 0: eX
+ 1: e
+    ** Failers
+No match
+    abcdY
+No match
+    ey     
+No match
+    
+/(?a) (b)(c)  (?d  (?(R&A)$ | (?4)) )/JDZx
+------------------------------------------------------------------
+        Bra
+        CBra 1
+        a
+        Ket
+        CBra 2
+        b
+        Ket
+        CBra 3
+        c
+        Ket
+        CBra 4
+        d
+        Cond
+        Cond nrecurse 1
+        $
+        Alt
+        Recurse
+        Ket
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 4
+Named capturing subpatterns:
+  A   1
+  A   4
+Options: extended dupnames
+First char = 'a'
+Need char = 'd'
+    abcdd
+ 0: abcdd
+ 1: a
+ 2: b
+ 3: c
+ 4: dd
+    ** Failers
+No match
+    abcdde  
+No match
+
+/abcd*/
+    xxxxabcd\P
+ 0: abcd
+    xxxxabcd\P\P
+Partial match: abcd
+
+/abcd*/i
+    xxxxabcd\P
+ 0: abcd
+    xxxxabcd\P\P
+Partial match: abcd
+    XXXXABCD\P
+ 0: ABCD
+    XXXXABCD\P\P
+Partial match: ABCD
+
+/abc\d*/
+    xxxxabc1\P
+ 0: abc1
+    xxxxabc1\P\P
+Partial match: abc1
+
+/(a)bc\1*/
+    xxxxabca\P
+ 0: abca
+ 1: a
+    xxxxabca\P\P
+Partial match: abca
+
+/abc[de]*/
+    xxxxabcde\P
+ 0: abcde
+    xxxxabcde\P\P
+Partial match: abcde
+
+/-- This is not in the Perl-compatible test because Perl seems currently to be
+    broken and not behaving as specified in that it *does* bumpalong after
+    hitting (*COMMIT). --/
+
+/(?1)(A(*COMMIT)|B)D/
+    ABD
+ 0: ABD
+ 1: B
+    XABD
+ 0: ABD
+ 1: B
+    BAD
+ 0: BAD
+ 1: A
+    ABXABD  
+ 0: ABD
+ 1: B
+    ** Failers 
+No match
+    ABX 
+No match
+    BAXBAD  
+No match
+
+/(\3)(\1)(a)/
+    cat
+ 0: a
+ 1: 
+ 2: 
+ 3: a
+
+/(\3)(\1)(a)/SI
+Capturing subpattern count = 3
+Max back reference = 3
+Options:
+No first char
+Need char = 'a'
+Subject length lower bound = 1
+No set of starting bytes
+    cat
+ 0: a
+ 1: 
+ 2: 
+ 3: a
+
+/(\3)(\1)(a)/SI
+Capturing subpattern count = 3
+Max back reference = 3
+No options
+No first char
+Need char = 'a'
+Subject length lower bound = 3
+No set of starting bytes
+    cat
+No match
+
+/i(?(DEFINE)(?a))/SI
+Capturing subpattern count = 1
+Named capturing subpatterns:
+  s   1
+No options
+First char = 'i'
+No need char
+Subject length lower bound = 1
+No set of starting bytes
+    i
+ 0: i
+    
+/()i(?(1)a)/SI 
+Capturing subpattern count = 1
+No options
+No first char
+Need char = 'i'
+Subject length lower bound = 1
+Starting byte set: i 
+    ia
+ 0: ia
+ 1: 
+
+/(?i)a(?-i)b|c/BZ
+------------------------------------------------------------------
+        Bra
+     /i a
+        b
+        Alt
+        c
+        Ket
+        End
+------------------------------------------------------------------
+    XabX
+ 0: ab
+    XAbX
+ 0: Ab
+    CcC 
+ 0: c
+    ** Failers
+No match
+    XABX   
+No match
+
+/(?i)a(?s)b|c/BZ
+------------------------------------------------------------------
+        Bra
+     /i ab
+        Alt
+     /i c
+        Ket
+        End
+------------------------------------------------------------------
+
+/(?i)a(?s-i)b|c/BZ
+------------------------------------------------------------------
+        Bra
+     /i a
+        b
+        Alt
+        c
+        Ket
+        End
+------------------------------------------------------------------
+
+/^(ab(c\1)d|x){2}$/BZ
+------------------------------------------------------------------
+        Bra
+        ^
+        Once
+        CBra 1
+        ab
+        CBra 2
+        c
+        \1
+        Ket
+        d
+        Alt
+        x
+        Ket
+        Ket
+        Once
+        CBra 1
+        ab
+        CBra 2
+        c
+        \1
+        Ket
+        d
+        Alt
+        x
+        Ket
+        Ket
+        $
+        Ket
+        End
+------------------------------------------------------------------
+    xabcxd
+ 0: xabcxd
+ 1: abcxd
+ 2: cx
+    
+/^(?&t)*+(?(DEFINE)(?.))$/BZ
+------------------------------------------------------------------
+        Bra
+        ^
+        Braposzero
+        SBraPos
+        Recurse
+        KetRpos
+        Cond
+        Cond def
+        CBra 1
+        Any
+        Ket
+        Ket
+        $
+        Ket
+        End
+------------------------------------------------------------------
+
+/^(?&t)*(?(DEFINE)(?.))$/BZ
+------------------------------------------------------------------
+        Bra
+        ^
+        Brazero
+        Once
+        Recurse
+        KetRmax
+        Cond
+        Cond def
+        CBra 1
+        Any
+        Ket
+        Ket
+        $
+        Ket
+        End
+------------------------------------------------------------------
+
+/ -- This one is here because Perl gives the match as "b" rather than "ab". I
+     believe this to be a Perl bug. --/  
+      
+/(?>a\Kb)z|(ab)/
+    ab 
+ 0: ab
+ 1: ab
+
+/(?P(?P0|)|(?P>L2)(?P>L1))/
+Failed: recursive call could loop indefinitely at offset 31
+
+/abc(*MARK:)pqr/
+Failed: (*MARK) must have an argument at offset 10
+
+/abc(*:)pqr/
+Failed: (*MARK) must have an argument at offset 6
+
+/abc(*FAIL:123)xyz/
+Failed: an argument is not allowed for (*ACCEPT), (*FAIL), or (*COMMIT) at offset 13
+
+/--- This should, and does, fail. In Perl, it does not, which I think is a 
+     bug because replacing the B in the pattern by (B|D) does make it fail. ---/
+
+/A(*COMMIT)B/+K
+    ACABX
+No match
+
+/--- These should be different, but in Perl they are not, which I think
+     is a bug in Perl. ---/
+
+/A(*THEN)B|A(*THEN)C/K
+    AC
+ 0: AC
+
+/A(*PRUNE)B|A(*PRUNE)C/K
+    AC
+No match
+    
+/--- Mark names can be duplicated. Perl doesn't give a mark for this one,
+though PCRE does. ---/
+
+/^A(*:A)B|^X(*:A)Y/K
+    ** Failers
+No match
+    XAQQ
+No match, mark = A
+    
+/--- COMMIT at the start of a pattern should be the same as an anchor. Perl 
+optimizations defeat this. So does the PCRE optimization unless we disable it 
+with \Y. ---/
+
+/(*COMMIT)ABC/
+    ABCDEFG
+ 0: ABC
+    ** Failers
+No match
+    DEFGABC\Y  
+No match
+    
+/^(ab (c+(*THEN)cd) | xyz)/x
+    abcccd  
+No match
+
+/^(ab (c+(*PRUNE)cd) | xyz)/x
+    abcccd  
+No match
+
+/^(ab (c+(*FAIL)cd) | xyz)/x
+    abcccd  
+No match
+    
+/--- Perl gets some of these wrong ---/ 
+
+/(?>.(*ACCEPT))*?5/
+    abcde
+ 0: a
+
+/(.(*ACCEPT))*?5/
+    abcde
+ 0: a
+ 1: a
+
+/(.(*ACCEPT))5/
+    abcde
+ 0: a
+ 1: a
+
+/(.(*ACCEPT))*5/
+    abcde
+ 0: a
+ 1: a
+
+/A\NB./BZ
+------------------------------------------------------------------
+        Bra
+        A
+        Any
+        B
+        Any
+        Ket
+        End
+------------------------------------------------------------------
+    ACBD
+ 0: ACBD
+    *** Failers
+No match
+    A\nB
+No match
+    ACB\n   
+No match
+
+/A\NB./sBZ
+------------------------------------------------------------------
+        Bra
+        A
+        Any
+        B
+        AllAny
+        Ket
+        End
+------------------------------------------------------------------
+    ACBD
+ 0: ACBD
+    ACB\n 
+ 0: ACB\x0a
+    *** Failers
+No match
+    A\nB  
+No match
+  
+/A\NB/
+    A\nB
+ 0: A\x0aB
+    A\rB
+ 0: A\x0dB
+    ** Failers
+No match
+    A\r\nB    
+No match
+
+/\R+b/BZ
+------------------------------------------------------------------
+        Bra
+        \R++
+        b
+        Ket
+        End
+------------------------------------------------------------------
+
+/\R+\n/BZ
+------------------------------------------------------------------
+        Bra
+        \R+
+        \x0a
+        Ket
+        End
+------------------------------------------------------------------
+
+/\R+\d/BZ
+------------------------------------------------------------------
+        Bra
+        \R++
+        \d
+        Ket
+        End
+------------------------------------------------------------------
+
+/\d*\R/BZ
+------------------------------------------------------------------
+        Bra
+        \d*+
+        \R
+        Ket
+        End
+------------------------------------------------------------------
+
+/\s*\R/BZ
+------------------------------------------------------------------
+        Bra
+        \s*
+        \R
+        Ket
+        End
+------------------------------------------------------------------
+    \x20\x0a
+ 0:  \x0a
+    \x20\x0d
+ 0:  \x0d
+    \x20\x0d\x0a
+ 0:  \x0d\x0a
+
+/\S*\R/BZ
+------------------------------------------------------------------
+        Bra
+        \S*+
+        \R
+        Ket
+        End
+------------------------------------------------------------------
+    a\x0a
+ 0: a\x0a
+
+/X\h*\R/BZ
+------------------------------------------------------------------
+        Bra
+        X
+        \h*+
+        \R
+        Ket
+        End
+------------------------------------------------------------------
+    X\x20\x0a
+ 0: X \x0a
+
+/X\H*\R/BZ
+------------------------------------------------------------------
+        Bra
+        X
+        \H*
+        \R
+        Ket
+        End
+------------------------------------------------------------------
+    X\x0d\x0a
+ 0: X\x0d\x0a
+
+/X\H+\R/BZ
+------------------------------------------------------------------
+        Bra
+        X
+        \H+
+        \R
+        Ket
+        End
+------------------------------------------------------------------
+    X\x0d\x0a
+ 0: X\x0d\x0a
+
+/X\H++\R/BZ
+------------------------------------------------------------------
+        Bra
+        X
+        \H++
+        \R
+        Ket
+        End
+------------------------------------------------------------------
+    X\x0d\x0a
+No match
+
+/(?<=abc)def/
+    abc\P\P
+Partial match at offset 3: abc
+
+/abc$/
+    abc
+ 0: abc
+    abc\P
+ 0: abc
+    abc\P\P
+Partial match: abc
+
+/abc$/m
+    abc
+ 0: abc
+    abc\n
+ 0: abc
+    abc\P\P
+Partial match: abc
+    abc\n\P\P 
+ 0: abc
+    abc\P
+ 0: abc
+    abc\n\P
+ 0: abc
+
+/abc\z/
+    abc
+ 0: abc
+    abc\P
+ 0: abc
+    abc\P\P
+Partial match: abc
+
+/abc\Z/
+    abc
+ 0: abc
+    abc\P
+ 0: abc
+    abc\P\P
+Partial match: abc
+
+/abc\b/
+    abc
+ 0: abc
+    abc\P
+ 0: abc
+    abc\P\P
+Partial match: abc
+
+/abc\B/
+    abc
+No match
+    abc\P
+Partial match: abc
+    abc\P\P
+Partial match: abc
+
+/.+/
+    abc\>0
+ 0: abc
+    abc\>1
+ 0: bc
+    abc\>2
+ 0: c
+    abc\>3
+No match
+    abc\>4
+Error -24 (bad offset value)
+    abc\>-4 
+Error -24 (bad offset value)
+
+/^\cģ/
+Failed: \c must be followed by an ASCII character at offset 3
+
+/(?P(?P=abn)xxx)/BZ
+------------------------------------------------------------------
+        Bra
+        Once
+        CBra 1
+        \1
+        xxx
+        Ket
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+
+/(a\1z)/BZ
+------------------------------------------------------------------
+        Bra
+        Once
+        CBra 1
+        a
+        \1
+        z
+        Ket
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+
+/(?P(?P=abn)(?(?P=axn)xxx)/BZ
+Failed: reference to non-existent subpattern at offset 15
+
+/(?P(?P=axn)xxx)(?yy)/BZ
+------------------------------------------------------------------
+        Bra
+        CBra 1
+        \2
+        xxx
+        Ket
+        CBra 2
+        yy
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+
+/-- These tests are here because Perl gets the first one wrong. --/
+
+/(\R*)(.)/s
+    \r\n
+ 0: \x0d
+ 1: 
+ 2: \x0d
+    \r\r\n\n\r 
+ 0: \x0d\x0d\x0a\x0a\x0d
+ 1: \x0d\x0d\x0a\x0a
+ 2: \x0d
+    \r\r\n\n\r\n 
+ 0: \x0d\x0d\x0a\x0a\x0d
+ 1: \x0d\x0d\x0a\x0a
+ 2: \x0d
+
+/(\R)*(.)/s
+    \r\n
+ 0: \x0d
+ 1: 
+ 2: \x0d
+    \r\r\n\n\r 
+ 0: \x0d\x0d\x0a\x0a\x0d
+ 1: \x0a
+ 2: \x0d
+    \r\r\n\n\r\n 
+ 0: \x0d\x0d\x0a\x0a\x0d
+ 1: \x0a
+ 2: \x0d
+
+/((?>\r\n|\n|\x0b|\f|\r|\x85)*)(.)/s
+    \r\n
+ 0: \x0d
+ 1: 
+ 2: \x0d
+    \r\r\n\n\r 
+ 0: \x0d\x0d\x0a\x0a\x0d
+ 1: \x0d\x0d\x0a\x0a
+ 2: \x0d
+    \r\r\n\n\r\n 
+ 0: \x0d\x0d\x0a\x0a\x0d
+ 1: \x0d\x0d\x0a\x0a
+ 2: \x0d
+
+/-- --/
+
+/^abc$/BZ
+------------------------------------------------------------------
+        Bra
+        ^
+        abc
+        $
+        Ket
+        End
+------------------------------------------------------------------
+
+/^abc$/BZm
+------------------------------------------------------------------
+        Bra
+     /m ^
+        abc
+     /m $
+        Ket
+        End
+------------------------------------------------------------------
+
+/^(a)*+(\w)/S
+    aaaaX
+ 0: aaaaX
+ 1: a
+ 2: X
+    ** Failers 
+No match
+    aaaa
+No match
+
+/^(?:a)*+(\w)/S
+    aaaaX
+ 0: aaaaX
+ 1: X
+    ** Failers 
+No match
+    aaaa
+No match
+
+/(a)++1234/SDZ
+------------------------------------------------------------------
+        Bra
+        CBraPos 1
+        a
+        KetRpos
+        1234
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 1
+No options
+First char = 'a'
+Need char = '4'
+Subject length lower bound = 5
+No set of starting bytes
+
+/([abc])++1234/SI
+Capturing subpattern count = 1
+No options
+No first char
+Need char = '4'
+Subject length lower bound = 5
+Starting byte set: a b c 
+
+/(?<=(abc)+)X/
+Failed: lookbehind assertion is not fixed length at offset 10
+
+/(^ab)/I
+Capturing subpattern count = 1
+Options: anchored
+No first char
+No need char
+
+/(^ab)++/I
+Capturing subpattern count = 1
+Options: anchored
+No first char
+No need char
+
+/(^ab|^)+/I
+Capturing subpattern count = 1
+Options: anchored
+No first char
+No need char
+
+/(^ab|^)++/I
+Capturing subpattern count = 1
+Options: anchored
+No first char
+No need char
+
+/(?:^ab)/I
+Capturing subpattern count = 0
+Options: anchored
+No first char
+No need char
+
+/(?:^ab)++/I
+Capturing subpattern count = 0
+Options: anchored
+No first char
+No need char
+
+/(?:^ab|^)+/I
+Capturing subpattern count = 0
+Options: anchored
+No first char
+No need char
+
+/(?:^ab|^)++/I
+Capturing subpattern count = 0
+Options: anchored
+No first char
+No need char
+
+/(.*ab)/I
+Capturing subpattern count = 1
+No options
+First char at start or follows newline
+Need char = 'b'
+
+/(.*ab)++/I
+Capturing subpattern count = 1
+No options
+First char at start or follows newline
+Need char = 'b'
+
+/(.*ab|.*)+/I
+Capturing subpattern count = 1
+No options
+First char at start or follows newline
+No need char
+
+/(.*ab|.*)++/I
+Capturing subpattern count = 1
+No options
+First char at start or follows newline
+No need char
+
+/(?:.*ab)/I
+Capturing subpattern count = 0
+No options
+First char at start or follows newline
+Need char = 'b'
+
+/(?:.*ab)++/I
+Capturing subpattern count = 0
+No options
+First char at start or follows newline
+Need char = 'b'
+
+/(?:.*ab|.*)+/I
+Capturing subpattern count = 0
+No options
+First char at start or follows newline
+No need char
+
+/(?:.*ab|.*)++/I
+Capturing subpattern count = 0
+No options
+First char at start or follows newline
+No need char
+
+/(?=a)[bcd]/I
+Capturing subpattern count = 0
+No options
+First char = 'a'
+No need char
+
+/((?=a))[bcd]/I
+Capturing subpattern count = 1
+No options
+First char = 'a'
+No need char
+
+/((?=a))+[bcd]/I
+Capturing subpattern count = 1
+No options
+First char = 'a'
+No need char
+
+/((?=a))++[bcd]/I
+Capturing subpattern count = 1
+No options
+First char = 'a'
+No need char
+
+/(?=a+)[bcd]/iI
+Capturing subpattern count = 0
+Options: caseless
+First char = 'a' (caseless)
+No need char
+
+/(?=a+?)[bcd]/iI
+Capturing subpattern count = 0
+Options: caseless
+First char = 'a' (caseless)
+No need char
+
+/(?=a++)[bcd]/iI
+Capturing subpattern count = 0
+Options: caseless
+First char = 'a' (caseless)
+No need char
+
+/(?=a{3})[bcd]/iI
+Capturing subpattern count = 0
+Options: caseless
+First char = 'a' (caseless)
+Need char = 'a' (caseless)
+
+/(abc)\1+/S
+
+/-- Perl doesn't get these right IMO (the 3rd is PCRE-specific) --/
+
+/(?1)(?:(b(*ACCEPT))){0}/
+    b
+ 0: b
+
+/(?1)(?:(b(*ACCEPT))){0}c/
+    bc
+ 0: bc
+    ** Failers 
+No match
+    b 
+No match
+
+/(?1)(?:((*ACCEPT))){0}c/
+    c
+ 0: c
+    c\N 
+ 0: c
+
+/^.*?(?(?=a)a|b(*THEN)c)/
+    ba
+No match
+
+/^.*?(?(?=a)a|bc)/
+    ba
+ 0: ba
+
+/^.*?(?(?=a)a(*THEN)b|c)/
+    ac
+No match
+
+/^.*?(?(?=a)a(*THEN)b)c/
+    ac
+No match
+
+/^.*?(a(*THEN)b)c/
+    aabc
+No match
+    
+/^.*? (?1) c (?(DEFINE)(a(*THEN)b))/x
+    aabc
+ 0: aabc
+
+/^.*?(a(*THEN)b|z)c/
+    aabc
+ 0: aabc
+ 1: ab
+
+/^.*?(z|a(*THEN)b)c/
+    aabc
+ 0: aabc
+ 1: ab
+
+/-- --/
+
+/-- These studied versions are here because they are not Perl-compatible; the
+    studying means the mark is not seen. --/
+
+/(*MARK:A)(*SKIP:B)(C|X)/KS
+    C
+ 0: C
+ 1: C
+MK: A
+    D
+No match, mark = A
+     
+/(*:A)A+(*SKIP:A)(B|Z)/KS
+    AAAC
+No match, mark = A
+
+/-- --/
+
+"(?=a*(*ACCEPT)b)c"
+    c
+ 0: c
+    c\N 
+ 0: c
+    
+/(?1)c(?(DEFINE)((*ACCEPT)b))/
+    c
+ 0: c
+    c\N  
+ 0: c
+    
+/(?>(*ACCEPT)b)c/
+    c
+ 0: 
+    c\N  
+No match
+
+/(?:(?>(a)))+a%/++
+    %aa%
+ 0: aa%
+ 0+ 
+ 1: a
+ 1+ a%
+
+/(a)b|ac/++SS
+    ac\O3
+ 0: ac
+ 0+ 
+    
+/(a)(b)x|abc/++
+     abc\O6
+ 0: abc
+ 0+ 
+
+/(a)bc|(a)(b)\2/
+    \O3abc
+Matched, but too many substrings
+ 0: abc
+    \O4abc 
+Matched, but too many substrings
+ 0: abc
+
+/(?(DEFINE)(a(?2)|b)(b(?1)|a))(?:(?1)|(?2))/SI
+Capturing subpattern count = 2
+No options
+No first char
+No need char
+Subject length lower bound = 1
+No set of starting bytes
+
+/(a(?2)|b)(b(?1)|a)(?:(?1)|(?2))/SI
+Capturing subpattern count = 2
+No options
+No first char
+No need char
+Subject length lower bound = 3
+Starting byte set: a b 
+
+/(a(?2)|b)(b(?1)|a)(?1)(?2)/SI
+Capturing subpattern count = 2
+No options
+No first char
+No need char
+Subject length lower bound = 4
+Starting byte set: a b 
+
+/(abc)(?1)/SI
+Capturing subpattern count = 1
+No options
+First char = 'a'
+Need char = 'c'
+Subject length lower bound = 6
+No set of starting bytes
+
+/^(?>a)++/
+    aa\M
+Minimum match() limit = 5
+Minimum match() recursion limit = 2
+ 0: aa
+    aaaaaaaaa\M 
+Minimum match() limit = 12
+Minimum match() recursion limit = 2
+ 0: aaaaaaaaa
+    
+/(a)(?1)++/
+    aa\M
+Minimum match() limit = 7
+Minimum match() recursion limit = 4
+ 0: aa
+ 1: a
+    aaaaaaaaa\M  
+Minimum match() limit = 21
+Minimum match() recursion limit = 4
+ 0: aaaaaaaaa
+ 1: a
+
+/(?:(foo)|(bar)|(baz))X/SS=
+    bazfooX
+ 0: fooX
+ 1: foo
+ 2: 
+ 3: 
+    foobazbarX
+ 0: barX
+ 1: 
+ 2: bar
+ 3: 
+    barfooX
+ 0: fooX
+ 1: foo
+ 2: 
+ 3: 
+    bazX
+ 0: bazX
+ 1: 
+ 2: 
+ 3: baz
+    foobarbazX    
+ 0: bazX
+ 1: 
+ 2: 
+ 3: baz
+    bazfooX\O0
+Matched, but too many substrings
+    bazfooX\O2
+Matched, but too many substrings
+ 0: fooX
+    bazfooX\O4
+Matched, but too many substrings
+ 0: fooX
+ 1: 
+    bazfooX\O6
+Matched, but too many substrings
+ 0: fooX
+ 1: foo
+ 2: 
+    bazfooX\O8
+Matched, but too many substrings
+ 0: fooX
+ 1: foo
+ 2: 
+ 3: 
+    bazfooX\O10
+ 0: fooX
+ 1: foo
+ 2: 
+ 3: 
+
+/(?=abc){3}abc/BZ
+------------------------------------------------------------------
+        Bra
+        Assert
+        abc
+        Ket
+        abc
+        Ket
+        End
+------------------------------------------------------------------
+
+/(?=abc)+abc/BZ
+------------------------------------------------------------------
+        Bra
+        Assert
+        abc
+        Ket
+        abc
+        Ket
+        End
+------------------------------------------------------------------
+
+/(?=abc)++abc/BZ
+------------------------------------------------------------------
+        Bra
+        Assert
+        abc
+        Ket
+        abc
+        Ket
+        End
+------------------------------------------------------------------
+
+/(?=abc){0}xyz/BZ
+------------------------------------------------------------------
+        Bra
+        Skip zero
+        Assert
+        abc
+        Ket
+        xyz
+        Ket
+        End
+------------------------------------------------------------------
+
+/(?=(a))?./BZ
+------------------------------------------------------------------
+        Bra
+        Brazero
+        Assert
+        CBra 1
+        a
+        Ket
+        Ket
+        Any
+        Ket
+        End
+------------------------------------------------------------------
+
+/(?=(a))??./BZ
+------------------------------------------------------------------
+        Bra
+        Braminzero
+        Assert
+        CBra 1
+        a
+        Ket
+        Ket
+        Any
+        Ket
+        End
+------------------------------------------------------------------
+
+/^(?=(a)){0}b(?1)/BZ
+------------------------------------------------------------------
+        Bra
+        ^
+        Skip zero
+        Assert
+        CBra 1
+        a
+        Ket
+        Ket
+        b
+        Recurse
+        Ket
+        End
+------------------------------------------------------------------
+
+/(?(DEFINE)(a))?b(?1)/BZ
+------------------------------------------------------------------
+        Bra
+        Cond
+        Cond def
+        CBra 1
+        a
+        Ket
+        Ket
+        b
+        Recurse
+        Ket
+        End
+------------------------------------------------------------------
+
+/^(?=(?1))?[az]([abc])d/BZ
+------------------------------------------------------------------
+        Bra
+        ^
+        Brazero
+        Assert
+        Recurse
+        Ket
+        [az]
+        CBra 1
+        [a-c]
+        Ket
+        d
+        Ket
+        End
+------------------------------------------------------------------
+
+/^(?!a){0}\w+/BZ
+------------------------------------------------------------------
+        Bra
+        ^
+        Skip zero
+        Assert not
+        a
+        Ket
+        \w+
+        Ket
+        End
+------------------------------------------------------------------
+
+/(?<=(abc))?xyz/BZ
+------------------------------------------------------------------
+        Bra
+        Brazero
+        AssertB
+        Reverse
+        CBra 1
+        abc
+        Ket
+        Ket
+        xyz
+        Ket
+        End
+------------------------------------------------------------------
+
+/[:a[:abc]b:]/BZ
+------------------------------------------------------------------
+        Bra
+        [:[a-c]
+        b:]
+        Ket
+        End
+------------------------------------------------------------------
+
+/((?2))((?1))/SS
+    abc
+Error -26 (nested recursion at the same subject position)
+
+/((?(R2)a+|(?1)b))/SS
+    aaaabcde
+Error -26 (nested recursion at the same subject position)
+
+/(?(R)a*(?1)|((?R))b)/SS
+    aaaabcde
+Error -26 (nested recursion at the same subject position)
+
+/(a+|(?R)b)/
+Failed: recursive call could loop indefinitely at offset 7
+
+/^(a(*:A)(d|e(*:B))z|aeq)/C
+    adz
+--->adz
+ +0 ^       ^
+ +1 ^       (a(*:A)(d|e(*:B))z|aeq)
+ +2 ^       a
+ +3 ^^      (*:A)
+ +8 ^^      (d|e(*:B))
+Latest Mark: A
+ +9 ^^      d
++10 ^ ^     |
++18 ^ ^     z
++19 ^  ^    |
++24 ^  ^    
+ 0: adz
+ 1: adz
+ 2: d
+    aez
+--->aez
+ +0 ^       ^
+ +1 ^       (a(*:A)(d|e(*:B))z|aeq)
+ +2 ^       a
+ +3 ^^      (*:A)
+ +8 ^^      (d|e(*:B))
+Latest Mark: A
+ +9 ^^      d
++11 ^^      e
++12 ^ ^     (*:B)
++17 ^ ^     )
+Latest Mark: B
++18 ^ ^     z
++19 ^  ^    |
++24 ^  ^    
+ 0: aez
+ 1: aez
+ 2: e
+    aeqwerty
+--->aeqwerty
+ +0 ^            ^
+ +1 ^            (a(*:A)(d|e(*:B))z|aeq)
+ +2 ^            a
+ +3 ^^           (*:A)
+ +8 ^^           (d|e(*:B))
+Latest Mark: A
+ +9 ^^           d
++11 ^^           e
++12 ^ ^          (*:B)
++17 ^ ^          )
+Latest Mark: B
++18 ^ ^          z
++20 ^            a
++21 ^^           e
++22 ^ ^          q
++23 ^  ^         )
++24 ^  ^         
+ 0: aeq
+ 1: aeq
+
+/.(*F)/
+    \P\Pabc
+No match
+
+/\btype\b\W*?\btext\b\W*?\bjavascript\b/IS
+Capturing subpattern count = 0
+Max lookbehind = 1
+No options
+First char = 't'
+Need char = 't'
+Subject length lower bound = 18
+No set of starting bytes
+
+/\btype\b\W*?\btext\b\W*?\bjavascript\b|\burl\b\W*?\bshell:|a+)(?>(z+))\w/BZ
+------------------------------------------------------------------
+        Bra
+        ^
+        Once_NC
+        a+
+        Ket
+        Once
+        CBra 1
+        z+
+        Ket
+        Ket
+        \w
+        Ket
+        End
+------------------------------------------------------------------
+    aaaazzzzb
+ 0: aaaazzzzb
+ 1: zzzz
+    ** Failers
+No match
+    aazz  
+No match
+
+/(.)(\1|a(?2))/
+    bab
+ 0: bab
+ 1: b
+ 2: ab
+    
+/\1|(.)(?R)\1/
+    cbbbc
+ 0: cbbbc
+ 1: c
+    
+/(.)((?(1)c|a)|a(?2))/
+    baa  
+No match
+
+/(?P(?P=abn)xxx)/BZ
+------------------------------------------------------------------
+        Bra
+        Once
+        CBra 1
+        \1
+        xxx
+        Ket
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+
+/(a\1z)/BZ
+------------------------------------------------------------------
+        Bra
+        Once
+        CBra 1
+        a
+        \1
+        z
+        Ket
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+
+/^(?>a+)(?>b+)(?>c+)(?>d+)(?>e+)/
+     \Maabbccddee
+Minimum match() limit = 12
+Minimum match() recursion limit = 3
+ 0: aabbccddee
+
+/^(?>(a+))(?>(b+))(?>(c+))(?>(d+))(?>(e+))/
+     \Maabbccddee
+Minimum match() limit = 22
+Minimum match() recursion limit = 21
+ 0: aabbccddee
+ 1: aa
+ 2: bb
+ 3: cc
+ 4: dd
+ 5: ee
+
+/^(?>(a+))(?>b+)(?>(c+))(?>d+)(?>(e+))/
+     \Maabbccddee
+Minimum match() limit = 18
+Minimum match() recursion limit = 13
+ 0: aabbccddee
+ 1: aa
+ 2: cc
+ 3: ee
+
+/^a\x41z/
+    aAz
+ 0: aAz
+    *** Failers
+No match
+    ax41z
+No match
+
+/^a[m\x41]z/
+    aAz
+ 0: aAz
+
+/^a\x1z/
+    ax1z
+ 0: ax1z
+
+/^a\u0041z/
+    aAz
+ 0: aAz
+    *** Failers
+No match
+    au0041z
+No match
+
+/^a[m\u0041]z/
+    aAz
+ 0: aAz
+
+/^a\u041z/
+    au041z
+ 0: au041z
+    *** Failers
+No match
+    aAz
+No match
+
+/^a\U0041z/
+    aU0041z
+ 0: aU0041z
+    *** Failers
+No match
+    aAz
+No match
+
+/(?(?=c)c|d)++Y/BZ
+------------------------------------------------------------------
+        Bra
+        BraPos
+        Cond
+        Assert
+        c
+        Ket
+        c
+        Alt
+        d
+        Ket
+        KetRpos
+        Y
+        Ket
+        End
+------------------------------------------------------------------
+
+/(?(?=c)c|d)*+Y/BZ
+------------------------------------------------------------------
+        Bra
+        Braposzero
+        BraPos
+        Cond
+        Assert
+        c
+        Ket
+        c
+        Alt
+        d
+        Ket
+        KetRpos
+        Y
+        Ket
+        End
+------------------------------------------------------------------
+
+/a[\NB]c/
+Failed: \N is not supported in a class at offset 3
+
+/a[B-\Nc]/ 
+Failed: \N is not supported in a class at offset 5
+
+/(a)(?2){0,1999}?(b)/
+
+/(a)(?(DEFINE)(b))(?2){0,1999}?(?2)/
+
+/--- This test, with something more complicated than individual letters, causes
+different behaviour in Perl. Perhaps it disables some optimization; no tag is
+passed back for the failures, whereas in PCRE there is a tag. ---/
+    
+/(A|P)(*:A)(B|P) | (X|P)(X|P)(*:B)(Y|P)/xK
+    AABC
+ 0: AB
+ 1: A
+ 2: B
+MK: A
+    XXYZ 
+ 0: XXY
+ 1: 
+ 2: 
+ 3: X
+ 4: X
+ 5: Y
+MK: B
+    ** Failers
+No match
+    XAQQ  
+No match, mark = A
+    XAQQXZZ  
+No match, mark = A
+    AXQQQ 
+No match, mark = A
+    AXXQQQ 
+No match, mark = B
+
+/-- Perl doesn't give marks for these, though it does if the alternatives are
+replaced by single letters. --/
+    
+/(b|q)(*:m)f|a(*:n)w/K
+    aw 
+ 0: aw
+MK: n
+    ** Failers 
+No match, mark = n
+    abc
+No match, mark = m
+
+/(q|b)(*:m)f|a(*:n)w/K
+    aw 
+ 0: aw
+MK: n
+    ** Failers 
+No match, mark = n
+    abc
+No match, mark = m
+
+/-- After a partial match, the behaviour is as for a failure. --/
+
+/^a(*:X)bcde/K
+   abc\P
+Partial match, mark=X: abc
+   
+/-- These are here because Perl doesn't return a mark, except for the first --/
+
+/(?=(*:x))(q|)/K+
+    abc
+ 0: 
+ 0+ abc
+ 1: 
+MK: x
+
+/(?=(*:x))((*:y)q|)/K+
+    abc
+ 0: 
+ 0+ abc
+ 1: 
+MK: x
+
+/(?=(*:x))(?:(*:y)q|)/K+
+    abc
+ 0: 
+ 0+ abc
+MK: x
+
+/(?=(*:x))(?>(*:y)q|)/K+
+    abc
+ 0: 
+ 0+ abc
+MK: x
+
+/(?=a(*:x))(?!a(*:y)c)/K+
+    ab
+ 0: 
+ 0+ ab
+MK: x
+
+/(?=a(*:x))(?=a(*:y)c|)/K+
+    ab
+ 0: 
+ 0+ ab
+MK: x
+
+/(..)\1/
+    ab\P
+Partial match: ab
+    aba\P
+Partial match: aba
+    abab\P
+ 0: abab
+ 1: ab
+
+/(..)\1/i
+    ab\P
+Partial match: ab
+    abA\P
+Partial match: abA
+    aBAb\P
+ 0: aBAb
+ 1: aB
+
+/(..)\1{2,}/
+    ab\P
+Partial match: ab
+    aba\P
+Partial match: aba
+    abab\P
+Partial match: abab
+    ababa\P
+Partial match: ababa
+    ababab\P
+ 0: ababab
+ 1: ab
+    ababab\P\P
+Partial match: ababab
+    abababa\P
+ 0: ababab
+ 1: ab
+    abababa\P\P
+Partial match: abababa
+
+/(..)\1{2,}/i
+    ab\P
+Partial match: ab
+    aBa\P
+Partial match: aBa
+    aBAb\P
+Partial match: aBAb
+    AbaBA\P
+Partial match: AbaBA
+    abABAb\P
+ 0: abABAb
+ 1: ab
+    aBAbaB\P\P
+Partial match: aBAbaB
+    abABabA\P
+ 0: abABab
+ 1: ab
+    abaBABa\P\P
+Partial match: abaBABa
+
+/(..)\1{2,}?x/i
+    ab\P
+Partial match: ab
+    abA\P
+Partial match: abA
+    aBAb\P
+Partial match: aBAb
+    abaBA\P
+Partial match: abaBA
+    abAbaB\P
+Partial match: abAbaB
+    abaBabA\P
+Partial match: abaBabA
+    abAbABaBx\P
+ 0: abAbABaBx
+ 1: ab
+
+/^(..)\1/
+    aba\P
+Partial match: aba
+
+/^(..)\1{2,3}x/
+    aba\P
+Partial match: aba
+    ababa\P
+Partial match: ababa
+    ababa\P\P
+Partial match: ababa
+    abababx
+ 0: abababx
+ 1: ab
+    ababababx  
+ 0: ababababx
+ 1: ab
+
+/^(..)\1{2,3}?x/
+    aba\P
+Partial match: aba
+    ababa\P
+Partial match: ababa
+    ababa\P\P
+Partial match: ababa
+    abababx
+ 0: abababx
+ 1: ab
+    ababababx  
+ 0: ababababx
+ 1: ab
+    
+/^(..)(\1{2,3})ab/
+    abababab
+ 0: abababab
+ 1: ab
+ 2: abab
+
+/^\R/
+    \r\P
+ 0: \x0d
+    \r\P\P
+Partial match: \x0d
+    
+/^\R{2,3}x/
+    \r\P
+Partial match: \x0d
+    \r\P\P
+Partial match: \x0d
+    \r\r\P 
+Partial match: \x0d\x0d
+    \r\r\P\P
+Partial match: \x0d\x0d
+    \r\r\r\P  
+Partial match: \x0d\x0d\x0d
+    \r\r\r\P\P
+Partial match: \x0d\x0d\x0d
+    \r\rx
+ 0: \x0d\x0dx
+    \r\r\rx    
+ 0: \x0d\x0d\x0dx
+
+/^\R{2,3}?x/
+    \r\P
+Partial match: \x0d
+    \r\P\P
+Partial match: \x0d
+    \r\r\P 
+Partial match: \x0d\x0d
+    \r\r\P\P
+Partial match: \x0d\x0d
+    \r\r\r\P  
+Partial match: \x0d\x0d\x0d
+    \r\r\r\P\P
+Partial match: \x0d\x0d\x0d
+    \r\rx
+ 0: \x0d\x0dx
+    \r\r\rx    
+ 0: \x0d\x0d\x0dx
+    
+/^\R?x/
+    \r\P
+Partial match: \x0d
+    \r\P\P 
+Partial match: \x0d
+    x
+ 0: x
+    \rx  
+ 0: \x0dx
+
+/^\R+x/
+    \r\P
+Partial match: \x0d
+    \r\P\P 
+Partial match: \x0d
+    \r\n\P
+Partial match: \x0d\x0a
+    \r\n\P\P  
+Partial match: \x0d\x0a
+    \rx  
+ 0: \x0dx
+
+/^a$/
+    a\r\P
+Partial match: a\x0d
+    a\r\P\P 
+Partial match: a\x0d
+
+/^a$/m
+    a\r\P
+Partial match: a\x0d
+    a\r\P\P 
+Partial match: a\x0d
+
+/^(a$|a\r)/
+    a\r\P
+ 0: a\x0d
+ 1: a\x0d
+    a\r\P\P 
+Partial match: a\x0d
+
+/^(a$|a\r)/m
+    a\r\P
+ 0: a\x0d
+ 1: a\x0d
+    a\r\P\P 
+Partial match: a\x0d
+
+/./
+    \r\P
+ 0: \x0d
+    \r\P\P 
+Partial match: \x0d
+  
+/.{2,3}/
+    \r\P 
+Partial match: \x0d
+    \r\P\P
+Partial match: \x0d
+    \r\r\P
+ 0: \x0d\x0d
+    \r\r\P\P
+Partial match: \x0d\x0d
+    \r\r\r\P
+ 0: \x0d\x0d\x0d
+    \r\r\r\P\P     
+Partial match: \x0d\x0d\x0d
+
+/.{2,3}?/
+    \r\P 
+Partial match: \x0d
+    \r\P\P
+Partial match: \x0d
+    \r\r\P
+ 0: \x0d\x0d
+    \r\r\P\P
+Partial match: \x0d\x0d
+    \r\r\r\P
+ 0: \x0d\x0d
+    \r\r\r\P\P     
+ 0: \x0d\x0d
+
+"AB(C(D))(E(F))?(?(?=\2)(?=\4))"
+    ABCDGHI\O03
+Matched, but too many substrings
+ 0: ABCD
+    
+/-- These are all run as real matches in test 1; here we are just checking the
+settings of the anchored and startline bits. --/ 
+
+/(?>.*?a)(?<=ba)/I
+Capturing subpattern count = 0
+Max lookbehind = 2
+No options
+No first char
+Need char = 'a'
+
+/(?:.*?a)(?<=ba)/I
+Capturing subpattern count = 0
+Max lookbehind = 2
+No options
+First char at start or follows newline
+Need char = 'a'
+
+/.*?a(*PRUNE)b/I
+Capturing subpattern count = 0
+No options
+No first char
+Need char = 'b'
+
+/.*?a(*PRUNE)b/sI
+Capturing subpattern count = 0
+Options: dotall
+No first char
+Need char = 'b'
+
+/^a(*PRUNE)b/sI
+Capturing subpattern count = 0
+Options: anchored dotall
+No first char
+No need char
+
+/.*?a(*SKIP)b/I
+Capturing subpattern count = 0
+No options
+No first char
+Need char = 'b'
+
+/(?>.*?a)b/sI
+Capturing subpattern count = 0
+Options: dotall
+No first char
+Need char = 'b'
+
+/(?>.*?a)b/I
+Capturing subpattern count = 0
+No options
+No first char
+Need char = 'b'
+
+/(?>^a)b/sI
+Capturing subpattern count = 0
+Options: anchored dotall
+No first char
+No need char
+
+/(?>.*?)(?<=(abcd)|(wxyz))/I
+Capturing subpattern count = 2
+Max lookbehind = 4
+No options
+No first char
+No need char
+
+/(?>.*)(?<=(abcd)|(wxyz))/I
+Capturing subpattern count = 2
+Max lookbehind = 4
+No options
+No first char
+No need char
+
+"(?>.*)foo"I
+Capturing subpattern count = 0
+No options
+No first char
+Need char = 'o'
+
+"(?>.*?)foo"I
+Capturing subpattern count = 0
+No options
+No first char
+Need char = 'o'
+
+/(?>^abc)/mI
+Capturing subpattern count = 0
+Options: multiline
+First char at start or follows newline
+Need char = 'c'
+
+/(?>.*abc)/mI
+Capturing subpattern count = 0
+Options: multiline
+No first char
+Need char = 'c'
+
+/(?:.*abc)/mI
+Capturing subpattern count = 0
+Options: multiline
+First char at start or follows newline
+Need char = 'c'
+
+/-- Check PCRE_STUDY_EXTRA_NEEDED --/
+
+/.?/S-I
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+Study returned NULL
+
+/.?/S!I
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+Subject length lower bound = -1
+No set of starting bytes
+
+/(?:(a)+(?C1)bb|aa(?C2)b)/
+    aab\C+
+Callout 1: last capture = 1
+ 0: 
+ 1: a
+--->aab
+    ^ ^     b
+Callout 1: last capture = 1
+ 0: 
+ 1: a
+--->aab
+    ^^      b
+Callout 2: last capture = -1
+ 0: 
+--->aab
+    ^ ^     b
+ 0: aab
+   
+/(?:(a)++(?C1)bb|aa(?C2)b)/
+    aab\C+ 
+Callout 1: last capture = 1
+ 0: 
+ 1: a
+--->aab
+    ^ ^     b
+Callout 2: last capture = -1
+ 0: 
+--->aab
+    ^ ^     b
+ 0: aab
+    
+/(?:(?>(a))(?C1)bb|aa(?C2)b)/
+    aab\C+ 
+Callout 1: last capture = 1
+ 0: 
+ 1: a
+--->aab
+    ^^      b
+Callout 2: last capture = -1
+ 0: 
+--->aab
+    ^ ^     b
+ 0: aab
+
+/(?:(?1)(?C1)x|ab(?C2))((a)){0}/                                                
+    aab\C+ 
+Callout 1: last capture = -1
+ 0: 
+--->aab
+    ^^      x
+Callout 1: last capture = -1
+ 0: 
+--->aab
+     ^^     x
+Callout 2: last capture = -1
+ 0: 
+--->aab
+     ^ ^    )
+ 0: ab
+
+/(?1)(?C1)((a)(?C2)){0}/   
+    aab\C+ 
+Callout 2: last capture = 2
+ 0: 
+ 1: 
+ 2: a
+--->aab
+    ^^      )
+Callout 1: last capture = -1
+ 0: 
+--->aab
+    ^^      ((a)(?C2)){0}
+ 0: a
+
+/(?:(a)+(?C1)bb|aa(?C2)b)++/
+    aab\C+
+Callout 1: last capture = 1
+ 0: 
+ 1: a
+--->aab
+    ^ ^     b
+Callout 1: last capture = 1
+ 0: 
+ 1: a
+--->aab
+    ^^      b
+Callout 2: last capture = -1
+ 0: 
+--->aab
+    ^ ^     b
+ 0: aab
+    aab\C+\O2
+Callout 1: last capture = 1
+ 0: 
+--->aab
+    ^ ^     b
+Callout 1: last capture = 1
+ 0: 
+--->aab
+    ^^      b
+Callout 2: last capture = -1
+ 0: 
+--->aab
+    ^ ^     b
+ 0: aab
+
+/(ab)x|ab/
+    ab\O3
+ 0: ab
+    ab\O2 
+ 0: ab
+  
+/(ab)/
+    ab\O3
+Matched, but too many substrings
+ 0: ab
+    ab\O2 
+Matched, but too many substrings
+ 0: ab
+    
+/(?<=123)(*MARK:xx)abc/K
+    xxxx123a\P\P
+Partial match at offset 7, mark=xx: 123a
+    xxxx123a\P
+Partial match at offset 7, mark=xx: 123a
+    
+/123\Kabc/
+    xxxx123a\P\P
+Partial match: 123a
+    xxxx123a\P
+Partial match: 123a
+
+/^(?(?=a)aa|bb)/C
+    bb
+--->bb
+ +0 ^      ^
+ +1 ^      (?(?=a)aa|bb)
+ +3 ^      (?=a)
+ +6 ^      a
++11 ^      b
++12 ^^     b
++13 ^ ^    )
++14 ^ ^    
+ 0: bb
+
+/(?C1)^(?C2)(?(?C99)(?=(?C3)a(?C4))(?C5)a(?C6)a(?C7)|(?C8)b(?C9)b(?C10))(?C11)/
+    bb
+--->bb
+  1 ^      ^
+  2 ^      (?(?C99)(?=(?C3)a(?C4))(?C5)a(?C6)a(?C7)|(?C8)b(?C9)b(?C10))
+ 99 ^      (?=(?C3)a(?C4))
+  3 ^      a
+  8 ^      b
+  9 ^^     b
+ 10 ^ ^    )
+ 11 ^ ^    
+ 0: bb
+
+/-- Perl seems to have a bug with this one --/
+
+/aaaaa(*COMMIT)(*PRUNE)b|a+c/
+    aaaaaac
+ 0: aaaac
+    
+/-- Here are some that Perl treats differently because of the way it handles
+backtracking verbs. --/
+
+ /(?!a(*COMMIT)b)ac|ad/
+     ac
+ 0: ac
+     ad 
+ 0: ad
+
+/^(?!a(*THEN)b|ac)../
+     ac
+No match
+     ad 
+ 0: ad
+
+/^(?=a(*THEN)b|ac)/
+    ac
+ 0: 
+    
+/\A.*?(?:a|b(*THEN)c)/
+    ba
+ 0: ba
+
+/\A.*?(?:a|b(*THEN)c)++/
+    ba
+ 0: ba
+
+/\A.*?(?:a|b(*THEN)c|d)/
+    ba
+ 0: ba
+
+/(?:(a(*MARK:X)a+(*SKIP:X)b)){0}(?:(?1)|aac)/
+    aac 
+ 0: aac
+
+/\A.*?(a|b(*THEN)c)/
+    ba
+ 0: ba
+ 1: a
+
+/^(A(*THEN)B|A(*THEN)D)/
+    AD           
+ 0: AD
+ 1: AD
+    
+/(?!b(*THEN)a)bn|bnn/
+    bnn
+ 0: bn
+
+/(?(?=b(*SKIP)a)bn|bnn)/
+    bnn
+No match
+
+/(?=b(*THEN)a|)bn|bnn/
+    bnn
+ 0: bn
+
+/-------------------------/ 
+
+/(*LIMIT_MATCH=12bc)abc/
+Failed: (*VERB) not recognized or malformed at offset 7
+
+/(*LIMIT_MATCH=4294967290)abc/
+Failed: (*VERB) not recognized or malformed at offset 7
+
+/(*LIMIT_RECURSION=4294967280)abc/I
+Capturing subpattern count = 0
+Recursion limit = 4294967280
+No options
+First char = 'a'
+Need char = 'c'
+
+/(a+)*zz/
+    aaaaaaaaaaaaaz
+No match
+    aaaaaaaaaaaaaz\q3000
+Error -8 (match limit exceeded)
+
+/(a+)*zz/S-
+    aaaaaaaaaaaaaz\Q10 
+Error -21 (recursion limit exceeded)
+
+/(*LIMIT_MATCH=3000)(a+)*zz/I
+Capturing subpattern count = 1
+Match limit = 3000
+No options
+No first char
+Need char = 'z'
+    aaaaaaaaaaaaaz
+Error -8 (match limit exceeded)
+    aaaaaaaaaaaaaz\q60000
+Error -8 (match limit exceeded)
+
+/(*LIMIT_MATCH=60000)(*LIMIT_MATCH=3000)(a+)*zz/I
+Capturing subpattern count = 1
+Match limit = 3000
+No options
+No first char
+Need char = 'z'
+    aaaaaaaaaaaaaz
+Error -8 (match limit exceeded)
+
+/(*LIMIT_MATCH=60000)(a+)*zz/I
+Capturing subpattern count = 1
+Match limit = 60000
+No options
+No first char
+Need char = 'z'
+    aaaaaaaaaaaaaz
+No match
+    aaaaaaaaaaaaaz\q3000
+Error -8 (match limit exceeded)
+
+/(*LIMIT_RECURSION=10)(a+)*zz/IS-
+Capturing subpattern count = 1
+Recursion limit = 10
+No options
+No first char
+Need char = 'z'
+Subject length lower bound = 2
+Starting byte set: a z 
+    aaaaaaaaaaaaaz
+Error -21 (recursion limit exceeded)
+    aaaaaaaaaaaaaz\Q1000
+Error -21 (recursion limit exceeded)
+
+/(*LIMIT_RECURSION=10)(*LIMIT_RECURSION=1000)(a+)*zz/IS-
+Capturing subpattern count = 1
+Recursion limit = 10
+No options
+No first char
+Need char = 'z'
+Subject length lower bound = 2
+Starting byte set: a z 
+    aaaaaaaaaaaaaz
+Error -21 (recursion limit exceeded)
+
+/(*LIMIT_RECURSION=1000)(a+)*zz/IS-
+Capturing subpattern count = 1
+Recursion limit = 1000
+No options
+No first char
+Need char = 'z'
+Subject length lower bound = 2
+Starting byte set: a z 
+    aaaaaaaaaaaaaz
+No match
+    aaaaaaaaaaaaaz\Q10
+Error -21 (recursion limit exceeded)
+
+/-- End of testinput2 --/
diff --git a/pcre/testdata/testoutput20 b/pcre/testdata/testoutput20
new file mode 100644
index 00000000000..7753be2db6a
--- /dev/null
+++ b/pcre/testdata/testoutput20
@@ -0,0 +1,27 @@
+/-- These DFA tests are for the handling of characters greater than 255 in
+    16- or 32-bit, non-UTF mode. --/
+
+/^\x{ffff}+/i
+    \x{ffff}
+ 0: \x{ffff}
+
+/^\x{ffff}?/i
+    \x{ffff}
+ 0: \x{ffff}
+ 1: 
+
+/^\x{ffff}*/i
+    \x{ffff}
+ 0: \x{ffff}
+ 1: 
+
+/^\x{ffff}{3}/i
+    \x{ffff}\x{ffff}\x{ffff}
+ 0: \x{ffff}\x{ffff}\x{ffff}
+
+/^\x{ffff}{0,3}/i
+    \x{ffff}
+ 0: \x{ffff}
+ 1: 
+
+/-- End of testinput20 --/
diff --git a/pcre/testdata/testoutput21-16 b/pcre/testdata/testoutput21-16
new file mode 100644
index 00000000000..e831888a9e4
--- /dev/null
+++ b/pcre/testdata/testoutput21-16
@@ -0,0 +1,96 @@
+/-- Tests for reloading pre-compiled patterns. The first one gives an error
+right away, and can be any old pattern compiled in 8-bit mode ("abc" is
+typical). The others require the link size to be 2. */x
+
+(?:[AaLl]+)[^xX-]*?)(?P[\x{150}-\x{250}\x{300}]|[^\x{800}aAs-uS-U\x{d800}-\x{dfff}])++[^#\b\x{500}\x{1000}]{3,5}$
+    In 16-bit mode with options:  S>testdata/saved16LE-1
+                                 FS>testdata/saved16BE-1
+    In 32-bit mode with options:  S>testdata/saved32LE-1
+                                 FS>testdata/saved32BE-1
+--%x
+
+(?:[AaLl]+)[^xX-]*?)(?P[\x{150}-\x{250}\x{300}]|[^\x{800}aAs-uS-U\x{d800}-\x{dfff}])++[^#\b\x{500}\x{1000}]{3,5}$
+    In 16-bit mode with options:  S>testdata/saved16LE-1
+                                 FS>testdata/saved16BE-1
+    In 32-bit mode with options:  S>testdata/saved32LE-1
+                                 FS>testdata/saved32BE-1
+--%x
+
+[aZ\x{400}-\x{10ffff}]{4,}[\x{f123}\x{10039}\x{20000}-\x{21234}]?|[A-Cx-z\x{100000}-\x{1000a7}\x{101234}])(?[^az])
+    In 16-bit mode with options:  S8>testdata/saved16LE-1
+                                 FS8>testdata/saved16BE-1
+    In 32-bit mode with options:  S8>testdata/saved32LE-1
+                                 FS8testdata/saved32BE-1
+--%8x
+
+[aZ\x{400}-\x{10ffff}]{4,}[\x{f123}\x{10039}\x{20000}-\x{21234}]?|[A-Cx-z\x{100000}-\x{1000a7}\x{101234}])(?[^az])
+    In 16-bit mode with options:  S8>testdata/saved16LE-1
+                                 FS8>testdata/saved16BE-1
+    In 32-bit mode with options:  S8>testdata/saved32LE-1
+                                 FS8testdata/saved32BE-1
+--%8x
+
+>>\xaa<<<
+ 0: >>>
+    >>>\xba<<< 
+ 0: >>>
+
+/[\W]+/Lfr_FR
+    >>>\xaa<<<
+ 0: >>>
+    >>>\xba<<< 
+ 0: >>>
+
+/[^[:alpha:]]+/Lfr_FR
+    >>>\xaa<<<
+ 0: >>>
+    >>>\xba<<< 
+ 0: >>>
+
+/\w+/Lfr_FR
+    >>>\xaa<<<
+ 0: 
+    >>>\xba<<< 
+ 0: 
+
+/[\w]+/Lfr_FR
+    >>>\xaa<<<
+ 0: 
+    >>>\xba<<< 
+ 0: 
+
+/[[:alpha:]]+/Lfr_FR
+    >>>\xaa<<<
+ 0: 
+    >>>\xba<<< 
+ 0: 
+    
+/[[:alpha:]][[:lower:]][[:upper:]]/DZLfr_FR 
+------------------------------------------------------------------
+        Bra
+        [A-Za-z\xaa\xb5\xba\xc0-\xd6\xd8-\xf6\xf8-\xff]
+        [a-z\xb5\xdf-\xf6\xf8-\xff]
+        [A-Z\xc0-\xd6\xd8-\xde]
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+
+/-- End of testinput3 --/
diff --git a/pcre/testdata/testoutput4 b/pcre/testdata/testoutput4
new file mode 100644
index 00000000000..6694111fb55
--- /dev/null
+++ b/pcre/testdata/testoutput4
@@ -0,0 +1,1260 @@
+/-- This set of tests is for UTF support, excluding Unicode properties. It is
+    compatible with all versions of Perl >= 5.10 and both the 8-bit and 16-bit
+    PCRE libraries. --/
+   
+/a.b/8
+    acb
+ 0: acb
+    a\x7fb
+ 0: a\x{7f}b
+    a\x{100}b 
+ 0: a\x{100}b
+    *** Failers
+No match
+    a\nb  
+No match
+
+/a(.{3})b/8
+    a\x{4000}xyb 
+ 0: a\x{4000}xyb
+ 1: \x{4000}xy
+    a\x{4000}\x7fyb 
+ 0: a\x{4000}\x{7f}yb
+ 1: \x{4000}\x{7f}y
+    a\x{4000}\x{100}yb 
+ 0: a\x{4000}\x{100}yb
+ 1: \x{4000}\x{100}y
+    *** Failers
+No match
+    a\x{4000}b 
+No match
+    ac\ncb 
+No match
+
+/a(.*?)(.)/
+    a\xc0\x88b
+ 0: a\xc0
+ 1: 
+ 2: \xc0
+
+/a(.*?)(.)/8
+    a\x{100}b
+ 0: a\x{100}
+ 1: 
+ 2: \x{100}
+
+/a(.*)(.)/
+    a\xc0\x88b
+ 0: a\xc0\x88b
+ 1: \xc0\x88
+ 2: b
+
+/a(.*)(.)/8
+    a\x{100}b
+ 0: a\x{100}b
+ 1: \x{100}
+ 2: b
+
+/a(.)(.)/
+    a\xc0\x92bcd
+ 0: a\xc0\x92
+ 1: \xc0
+ 2: \x92
+
+/a(.)(.)/8
+    a\x{240}bcd
+ 0: a\x{240}b
+ 1: \x{240}
+ 2: b
+
+/a(.?)(.)/
+    a\xc0\x92bcd
+ 0: a\xc0\x92
+ 1: \xc0
+ 2: \x92
+
+/a(.?)(.)/8
+    a\x{240}bcd
+ 0: a\x{240}b
+ 1: \x{240}
+ 2: b
+
+/a(.??)(.)/
+    a\xc0\x92bcd
+ 0: a\xc0
+ 1: 
+ 2: \xc0
+
+/a(.??)(.)/8
+    a\x{240}bcd
+ 0: a\x{240}
+ 1: 
+ 2: \x{240}
+
+/a(.{3})b/8
+    a\x{1234}xyb 
+ 0: a\x{1234}xyb
+ 1: \x{1234}xy
+    a\x{1234}\x{4321}yb 
+ 0: a\x{1234}\x{4321}yb
+ 1: \x{1234}\x{4321}y
+    a\x{1234}\x{4321}\x{3412}b 
+ 0: a\x{1234}\x{4321}\x{3412}b
+ 1: \x{1234}\x{4321}\x{3412}
+    *** Failers
+No match
+    a\x{1234}b 
+No match
+    ac\ncb 
+No match
+
+/a(.{3,})b/8
+    a\x{1234}xyb 
+ 0: a\x{1234}xyb
+ 1: \x{1234}xy
+    a\x{1234}\x{4321}yb 
+ 0: a\x{1234}\x{4321}yb
+ 1: \x{1234}\x{4321}y
+    a\x{1234}\x{4321}\x{3412}b 
+ 0: a\x{1234}\x{4321}\x{3412}b
+ 1: \x{1234}\x{4321}\x{3412}
+    axxxxbcdefghijb 
+ 0: axxxxbcdefghijb
+ 1: xxxxbcdefghij
+    a\x{1234}\x{4321}\x{3412}\x{3421}b 
+ 0: a\x{1234}\x{4321}\x{3412}\x{3421}b
+ 1: \x{1234}\x{4321}\x{3412}\x{3421}
+    *** Failers
+No match
+    a\x{1234}b 
+No match
+
+/a(.{3,}?)b/8
+    a\x{1234}xyb 
+ 0: a\x{1234}xyb
+ 1: \x{1234}xy
+    a\x{1234}\x{4321}yb 
+ 0: a\x{1234}\x{4321}yb
+ 1: \x{1234}\x{4321}y
+    a\x{1234}\x{4321}\x{3412}b 
+ 0: a\x{1234}\x{4321}\x{3412}b
+ 1: \x{1234}\x{4321}\x{3412}
+    axxxxbcdefghijb 
+ 0: axxxxb
+ 1: xxxx
+    a\x{1234}\x{4321}\x{3412}\x{3421}b 
+ 0: a\x{1234}\x{4321}\x{3412}\x{3421}b
+ 1: \x{1234}\x{4321}\x{3412}\x{3421}
+    *** Failers
+No match
+    a\x{1234}b 
+No match
+
+/a(.{3,5})b/8
+    a\x{1234}xyb 
+ 0: a\x{1234}xyb
+ 1: \x{1234}xy
+    a\x{1234}\x{4321}yb 
+ 0: a\x{1234}\x{4321}yb
+ 1: \x{1234}\x{4321}y
+    a\x{1234}\x{4321}\x{3412}b 
+ 0: a\x{1234}\x{4321}\x{3412}b
+ 1: \x{1234}\x{4321}\x{3412}
+    axxxxbcdefghijb 
+ 0: axxxxb
+ 1: xxxx
+    a\x{1234}\x{4321}\x{3412}\x{3421}b 
+ 0: a\x{1234}\x{4321}\x{3412}\x{3421}b
+ 1: \x{1234}\x{4321}\x{3412}\x{3421}
+    axbxxbcdefghijb 
+ 0: axbxxb
+ 1: xbxx
+    axxxxxbcdefghijb 
+ 0: axxxxxb
+ 1: xxxxx
+    *** Failers
+No match
+    a\x{1234}b 
+No match
+    axxxxxxbcdefghijb 
+No match
+
+/a(.{3,5}?)b/8
+    a\x{1234}xyb 
+ 0: a\x{1234}xyb
+ 1: \x{1234}xy
+    a\x{1234}\x{4321}yb 
+ 0: a\x{1234}\x{4321}yb
+ 1: \x{1234}\x{4321}y
+    a\x{1234}\x{4321}\x{3412}b 
+ 0: a\x{1234}\x{4321}\x{3412}b
+ 1: \x{1234}\x{4321}\x{3412}
+    axxxxbcdefghijb 
+ 0: axxxxb
+ 1: xxxx
+    a\x{1234}\x{4321}\x{3412}\x{3421}b 
+ 0: a\x{1234}\x{4321}\x{3412}\x{3421}b
+ 1: \x{1234}\x{4321}\x{3412}\x{3421}
+    axbxxbcdefghijb 
+ 0: axbxxb
+ 1: xbxx
+    axxxxxbcdefghijb 
+ 0: axxxxxb
+ 1: xxxxx
+    *** Failers
+No match
+    a\x{1234}b 
+No match
+    axxxxxxbcdefghijb 
+No match
+
+/^[a\x{c0}]/8
+    *** Failers
+No match
+    \x{100}
+No match
+
+/(?<=aXb)cd/8
+    aXbcd
+ 0: cd
+
+/(?<=a\x{100}b)cd/8
+    a\x{100}bcd
+ 0: cd
+
+/(?<=a\x{100000}b)cd/8
+    a\x{100000}bcd
+ 0: cd
+    
+/(?:\x{100}){3}b/8
+    \x{100}\x{100}\x{100}b
+ 0: \x{100}\x{100}\x{100}b
+    *** Failers 
+No match
+    \x{100}\x{100}b
+No match
+
+/\x{ab}/8
+    \x{ab} 
+ 0: \x{ab}
+    \xc2\xab
+ 0: \x{ab}
+    *** Failers 
+No match
+    \x00{ab}
+No match
+
+/(?<=(.))X/8
+    WXYZ
+ 0: X
+ 1: W
+    \x{256}XYZ 
+ 0: X
+ 1: \x{256}
+    *** Failers
+No match
+    XYZ 
+No match
+
+/[^a]+/8g
+    bcd
+ 0: bcd
+    \x{100}aY\x{256}Z 
+ 0: \x{100}
+ 0: Y\x{256}Z
+    
+/^[^a]{2}/8
+    \x{100}bc
+ 0: \x{100}b
+ 
+/^[^a]{2,}/8
+    \x{100}bcAa
+ 0: \x{100}bcA
+
+/^[^a]{2,}?/8
+    \x{100}bca
+ 0: \x{100}b
+
+/[^a]+/8ig
+    bcd
+ 0: bcd
+    \x{100}aY\x{256}Z 
+ 0: \x{100}
+ 0: Y\x{256}Z
+    
+/^[^a]{2}/8i
+    \x{100}bc
+ 0: \x{100}b
+ 
+/^[^a]{2,}/8i
+    \x{100}bcAa
+ 0: \x{100}bc
+
+/^[^a]{2,}?/8i
+    \x{100}bca
+ 0: \x{100}b
+
+/\x{100}{0,0}/8
+    abcd
+ 0: 
+ 
+/\x{100}?/8
+    abcd
+ 0: 
+    \x{100}\x{100} 
+ 0: \x{100}
+
+/\x{100}{0,3}/8 
+    \x{100}\x{100} 
+ 0: \x{100}\x{100}
+    \x{100}\x{100}\x{100}\x{100} 
+ 0: \x{100}\x{100}\x{100}
+    
+/\x{100}*/8
+    abce
+ 0: 
+    \x{100}\x{100}\x{100}\x{100} 
+ 0: \x{100}\x{100}\x{100}\x{100}
+
+/\x{100}{1,1}/8
+    abcd\x{100}\x{100}\x{100}\x{100} 
+ 0: \x{100}
+
+/\x{100}{1,3}/8
+    abcd\x{100}\x{100}\x{100}\x{100} 
+ 0: \x{100}\x{100}\x{100}
+
+/\x{100}+/8
+    abcd\x{100}\x{100}\x{100}\x{100} 
+ 0: \x{100}\x{100}\x{100}\x{100}
+
+/\x{100}{3}/8
+    abcd\x{100}\x{100}\x{100}XX
+ 0: \x{100}\x{100}\x{100}
+
+/\x{100}{3,5}/8
+    abcd\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}XX
+ 0: \x{100}\x{100}\x{100}\x{100}\x{100}
+
+/\x{100}{3,}/8
+    abcd\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}XX
+ 0: \x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}
+
+/(?<=a\x{100}{2}b)X/8+
+    Xyyya\x{100}\x{100}bXzzz
+ 0: X
+ 0+ zzz
+
+/\D*/8
+  aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+ 0: aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+
+/\D*/8
+  \x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}
+ 0: \x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}
+
+/\D/8
+    1X2
+ 0: X
+    1\x{100}2 
+ 0: \x{100}
+  
+/>\S/8
+    > >X Y
+ 0: >X
+    > >\x{100} Y
+ 0: >\x{100}
+  
+/\d/8
+    \x{100}3
+ 0: 3
+    
+/\s/8
+    \x{100} X
+ 0:  
+    
+/\D+/8
+    12abcd34
+ 0: abcd
+    *** Failers
+ 0: *** Failers
+    1234  
+No match
+
+/\D{2,3}/8
+    12abcd34
+ 0: abc
+    12ab34
+ 0: ab
+    *** Failers  
+ 0: ***
+    1234
+No match
+    12a34  
+No match
+
+/\D{2,3}?/8
+    12abcd34
+ 0: ab
+    12ab34
+ 0: ab
+    *** Failers  
+ 0: **
+    1234
+No match
+    12a34  
+No match
+
+/\d+/8
+    12abcd34
+ 0: 12
+    *** Failers
+No match
+
+/\d{2,3}/8
+    12abcd34
+ 0: 12
+    1234abcd
+ 0: 123
+    *** Failers  
+No match
+    1.4 
+No match
+
+/\d{2,3}?/8
+    12abcd34
+ 0: 12
+    1234abcd
+ 0: 12
+    *** Failers  
+No match
+    1.4 
+No match
+
+/\S+/8
+    12abcd34
+ 0: 12abcd34
+    *** Failers
+ 0: ***
+    \    \ 
+No match
+
+/\S{2,3}/8
+    12abcd34
+ 0: 12a
+    1234abcd
+ 0: 123
+    *** Failers
+ 0: ***
+    \     \  
+No match
+
+/\S{2,3}?/8
+    12abcd34
+ 0: 12
+    1234abcd
+ 0: 12
+    *** Failers
+ 0: **
+    \     \  
+No match
+
+/>\s+      <34
+ 0: >      <
+ 0+ 34
+    *** Failers
+No match
+
+/>\s{2,3}    <
+ 0+ cd
+    ab>      <
+ 0+ ce
+    *** Failers
+No match
+    ab>    \s{2,3}?    <
+ 0+ cd
+    ab>      <
+ 0+ ce
+    *** Failers
+No match
+    ab>    \xff<
+ 0: \xff
+
+/[\xff]/8
+    >\x{ff}<
+ 0: \x{ff}
+
+/[^\xFF]/
+    XYZ
+ 0: X
+
+/[^\xff]/8
+    XYZ
+ 0: X
+    \x{123} 
+ 0: \x{123}
+
+/^[ac]*b/8
+  xb
+No match
+
+/^[ac\x{100}]*b/8
+  xb
+No match
+
+/^[^x]*b/8i
+  xb
+No match
+
+/^[^x]*b/8
+  xb
+No match
+  
+/^\d*b/8
+  xb 
+No match
+
+/(|a)/g8
+    catac
+ 0: 
+ 1: 
+ 0: 
+ 1: 
+ 0: a
+ 1: a
+ 0: 
+ 1: 
+ 0: 
+ 1: 
+ 0: a
+ 1: a
+ 0: 
+ 1: 
+ 0: 
+ 1: 
+    a\x{256}a 
+ 0: 
+ 1: 
+ 0: a
+ 1: a
+ 0: 
+ 1: 
+ 0: 
+ 1: 
+ 0: a
+ 1: a
+ 0: 
+ 1: 
+
+/^\x{85}$/8i
+    \x{85}
+ 0: \x{85}
+
+/^ሴ/8
+    ሴ 
+ 0: \x{1234}
+
+/^\ሴ/8
+    ሴ 
+ 0: \x{1234}
+
+"(?s)(.{1,5})"8
+    abcdefg
+ 0: abcde
+ 1: abcde
+    ab
+ 0: ab
+ 1: ab
+
+/a*\x{100}*\w/8
+    a 
+ 0: a
+
+/\S\S/8g
+    A\x{a3}BC
+ 0: A\x{a3}
+ 0: BC
+    
+/\S{2}/8g
+    A\x{a3}BC
+ 0: A\x{a3}
+ 0: BC
+    
+/\W\W/8g
+    +\x{a3}== 
+ 0: +\x{a3}
+ 0: ==
+
+/\W{2}/8g
+    +\x{a3}== 
+ 0: +\x{a3}
+ 0: ==
+
+/\S/8g
+    \x{442}\x{435}\x{441}\x{442}
+ 0: \x{442}
+ 0: \x{435}
+ 0: \x{441}
+ 0: \x{442}
+
+/[\S]/8g
+    \x{442}\x{435}\x{441}\x{442}
+ 0: \x{442}
+ 0: \x{435}
+ 0: \x{441}
+ 0: \x{442}
+
+/\D/8g
+    \x{442}\x{435}\x{441}\x{442}
+ 0: \x{442}
+ 0: \x{435}
+ 0: \x{441}
+ 0: \x{442}
+
+/[\D]/8g
+    \x{442}\x{435}\x{441}\x{442}
+ 0: \x{442}
+ 0: \x{435}
+ 0: \x{441}
+ 0: \x{442}
+
+/\W/8g
+    \x{2442}\x{2435}\x{2441}\x{2442}
+ 0: \x{2442}
+ 0: \x{2435}
+ 0: \x{2441}
+ 0: \x{2442}
+
+/[\W]/8g
+    \x{2442}\x{2435}\x{2441}\x{2442}
+ 0: \x{2442}
+ 0: \x{2435}
+ 0: \x{2441}
+ 0: \x{2442}
+    
+/[\S\s]*/8
+    abc\n\r\x{442}\x{435}\x{441}\x{442}xyz 
+ 0: abc\x{0a}\x{0d}\x{442}\x{435}\x{441}\x{442}xyz
+
+/[\x{41f}\S]/8g
+    \x{442}\x{435}\x{441}\x{442}
+ 0: \x{442}
+ 0: \x{435}
+ 0: \x{441}
+ 0: \x{442}
+
+/.[^\S]./8g
+    abc def\x{442}\x{443}xyz\npqr
+ 0: c d
+ 0: z\x{0a}p
+
+/.[^\S\n]./8g
+    abc def\x{442}\x{443}xyz\npqr
+ 0: c d
+
+/[[:^alnum:]]/8g  
+    +\x{2442}
+ 0: +
+ 0: \x{2442}
+    
+/[[:^alpha:]]/8g 
+    +\x{2442}
+ 0: +
+ 0: \x{2442}
+    
+/[[:^ascii:]]/8g 
+    A\x{442}
+ 0: \x{442}
+    
+/[[:^blank:]]/8g 
+    A\x{442}
+ 0: A
+ 0: \x{442}
+    
+/[[:^cntrl:]]/8g 
+    A\x{442}
+ 0: A
+ 0: \x{442}
+    
+/[[:^digit:]]/8g 
+    A\x{442}
+ 0: A
+ 0: \x{442}
+    
+/[[:^graph:]]/8g 
+    \x19\x{e01ff}
+ 0: \x{19}
+ 0: \x{e01ff}
+    
+/[[:^lower:]]/8g 
+    A\x{422}
+ 0: A
+ 0: \x{422}
+    
+/[[:^print:]]/8g 
+    \x{19}\x{e01ff}
+ 0: \x{19}
+ 0: \x{e01ff}
+    
+/[[:^punct:]]/8g 
+    A\x{442}
+ 0: A
+ 0: \x{442}
+    
+/[[:^space:]]/8g 
+    A\x{442}
+ 0: A
+ 0: \x{442}
+    
+/[[:^upper:]]/8g 
+    a\x{442}
+ 0: a
+ 0: \x{442}
+    
+/[[:^word:]]/8g  
+    +\x{2442}
+ 0: +
+ 0: \x{2442}
+    
+/[[:^xdigit:]]/8g
+    M\x{442}
+ 0: M
+ 0: \x{442}
+
+/[^ABCDEFGHIJKLMNOPQRSTUVWXYZÀÁÂÃÄÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖØÙÚÛÜÝÞĀĂĄĆĈĊČĎĐĒĔĖĘĚĜĞĠĢĤĦĨĪĬĮİIJĴĶĹĻĽĿŁŃŅŇŊŌŎŐŒŔŖŘŚŜŞŠŢŤŦŨŪŬŮŰŲŴŶŸŹŻŽƁƂƄƆƇƉƊƋƎƏƐƑƓƔƖƗƘƜƝƟƠƢƤƦƧƩƬƮƯƱƲƳƵƷƸƼDŽLJNJǍǏǑǓǕǗǙǛǞǠǢǤǦǨǪǬǮDZǴǶǷǸǺǼǾȀȂȄȆȈȊȌȎȐȒȔȖȘȚȜȞȠȢȤȦȨȪȬȮȰȲȺȻȽȾɁΆΈΉΊΌΎΏΑΒΓΔΕΖΗΘΙΚΛΜΝΞΟΠΡΣΤΥΦΧΨΩΪΫϒϓϔϘϚϜϞϠϢϤϦϨϪϬϮϴϷϹϺϽϾϿЀЁЂЃЄЅІЇЈЉЊЋЌЍЎЏАБВГДЕЖЗИЙКЛМНОПРСТУФХЦЧШЩЪЫЬЭЮЯѠѢѤѦѨѪѬѮѰѲѴѶѸѺѼѾҀҊҌҎҐҒҔҖҘҚҜҞҠҢҤҦҨҪҬҮҰҲҴҶҸҺҼҾӀӁӃӅӇӉӋӍӐӒӔӖӘӚӜӞӠӢӤӦӨӪӬӮӰӲӴӶӸԀԂԄԆԈԊԌԎԱԲԳԴԵԶԷԸԹԺԻԼԽԾԿՀՁՂՃՄՅՆՇՈՉՊՋՌՍՎՏՐՑՒՓՔՕՖႠႡႢႣႤႥႦႧႨႩႪႫႬႭႮႯႰႱႲႳႴႵႶႷႸႹႺႻႼႽႾႿჀჁჂჃჄჅḀḂḄḆḈḊḌḎḐḒḔḖḘḚḜḞḠḢḤḦḨḪḬḮḰḲḴḶḸḺḼḾṀṂṄṆṈṊṌṎṐṒṔṖṘṚṜṞṠṢṤṦṨṪṬṮṰṲṴṶṸṺṼṾẀẂẄẆẈẊẌẎẐẒẔẠẢẤẦẨẪẬẮẰẲẴẶẸẺẼẾỀỂỄỆỈỊỌỎỐỒỔỖỘỚỜỞỠỢỤỦỨỪỬỮỰỲỴỶỸἈἉἊἋἌἍἎἏἘἙἚἛἜἝἨἩἪἫἬἭἮἯἸἹἺἻἼἽἾἿὈὉὊὋὌὍὙὛὝὟὨὩὪὫὬὭὮὯᾸᾹᾺΆῈΈῊΉῘῙῚΊῨῩῪΎῬῸΌῺΏabcdefghijklmnopqrstuvwxyzªµºßàáâãäåæçèéêëìíîïðñòóôõöøùúûüýþÿāăąćĉċčďđēĕėęěĝğġģĥħĩīĭįıijĵķĸĺļľŀłńņňʼnŋōŏőœŕŗřśŝşšţťŧũūŭůűųŵŷźżžſƀƃƅƈƌƍƒƕƙƚƛƞơƣƥƨƪƫƭưƴƶƹƺƽƾƿdžljnjǎǐǒǔǖǘǚǜǝǟǡǣǥǧǩǫǭǯǰdzǵǹǻǽǿȁȃȅȇȉȋȍȏȑȓȕȗșțȝȟȡȣȥȧȩȫȭȯȱȳȴȵȶȷȸȹȼȿɀɐɑɒɓɔɕɖɗɘəɚɛɜɝɞɟɠɡɢɣɤɥɦɧɨɩɪɫɬɭɮɯɰɱɲɳɴɵɶɷɸɹɺɻɼɽɾɿʀʁʂʃʄʅʆʇʈʉʊʋʌʍʎʏʐʑʒʓʔʕʖʗʘʙʚʛʜʝʞʟʠʡʢʣʤʥʦʧʨʩʪʫʬʭʮʯΐάέήίΰαβγδεζηθικλμνξοπρςστυφχψωϊϋόύώϐϑϕϖϗϙϛϝϟϡϣϥϧϩϫϭϯϰϱϲϳϵϸϻϼабвгдежзийклмнопрстуфхцчшщъыьэюяѐёђѓєѕіїјљњћќѝўџѡѣѥѧѩѫѭѯѱѳѵѷѹѻѽѿҁҋҍҏґғҕҗҙқҝҟҡңҥҧҩҫҭүұҳҵҷҹһҽҿӂӄӆӈӊӌӎӑӓӕӗәӛӝӟӡӣӥӧөӫӭӯӱӳӵӷӹԁԃԅԇԉԋԍԏաբգդեզէըթժիլխծկհձղճմյնշոչպջռսվտրցւփքօֆևᴀᴁᴂᴃᴄᴅᴆᴇᴈᴉᴊᴋᴌᴍᴎᴏᴐᴑᴒᴓᴔᴕᴖᴗᴘᴙᴚᴛᴜᴝᴞᴟᴠᴡᴢᴣᴤᴥᴦᴧᴨᴩᴪᴫᵢᵣᵤᵥᵦᵧᵨᵩᵪᵫᵬᵭᵮᵯᵰᵱᵲᵳᵴᵵᵶᵷᵹᵺᵻᵼᵽᵾᵿᶀᶁᶂᶃᶄᶅᶆᶇᶈᶉᶊᶋᶌᶍᶎᶏᶐᶑᶒᶓᶔᶕᶖᶗᶘᶙᶚḁḃḅḇḉḋḍḏḑḓḕḗḙḛḝḟḡḣḥḧḩḫḭḯḱḳḵḷḹḻḽḿṁṃṅṇṉṋṍṏṑṓṕṗṙṛṝṟṡṣṥṧṩṫṭṯṱṳṵṷṹṻṽṿẁẃẅẇẉẋẍẏẑẓẕẖẗẘẙẚẛạảấầẩẫậắằẳẵặẹẻẽếềểễệỉịọỏốồổỗộớờởỡợụủứừửữựỳỵỷỹἀἁἂἃἄἅἆἇἐἑἒἓἔἕἠἡἢἣἤἥἦἧἰἱἲἳἴἵἶἷὀὁὂὃὄὅὐὑὒὓὔὕὖὗὠὡὢὣὤὥὦὧὰάὲέὴήὶίὸόὺύὼώᾀᾁᾂᾃᾄᾅᾆᾇᾐᾑᾒᾓᾔᾕᾖᾗᾠᾡᾢᾣᾤᾥᾦᾧᾰᾱᾲᾳᾴᾶᾷιῂῃῄῆῇῐῑῒΐῖῗῠῡῢΰῤῥῦῧῲῳῴῶῷⲁⲃⲅⲇⲉⲋⲍⲏⲑⲓⲕⲗⲙⲛⲝⲟⲡⲣⲥⲧⲩⲫⲭⲯⲱⲳⲵⲷⲹⲻⲽⲿⳁⳃⳅⳇⳉⳋⳍⳏⳑⳓⳕⳗⳙⳛⳝⳟⳡⳣⳤⴀⴁⴂⴃⴄⴅⴆⴇⴈⴉⴊⴋⴌⴍⴎⴏⴐⴑⴒⴓⴔⴕⴖⴗⴘⴙⴚⴛⴜⴝⴞⴟⴠⴡⴢⴣⴤⴥfffiflffifflſtstﬓﬔﬕﬖﬗ\d-_^]/8
+
+/^[^d]*?$/
+    abc
+ 0: abc
+
+/^[^d]*?$/8
+    abc
+ 0: abc
+
+/^[^d]*?$/i
+    abc
+ 0: abc
+
+/^[^d]*?$/8i
+    abc
+ 0: abc
+
+/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8
+
+/^[a\x{c0}]b/8
+    \x{c0}b
+ 0: \x{c0}b
+    
+/^([a\x{c0}]*?)aa/8
+    a\x{c0}aaaa/ 
+ 0: a\x{c0}aa
+ 1: a\x{c0}
+
+/^([a\x{c0}]*?)aa/8
+    a\x{c0}aaaa/ 
+ 0: a\x{c0}aa
+ 1: a\x{c0}
+    a\x{c0}a\x{c0}aaa/ 
+ 0: a\x{c0}a\x{c0}aa
+ 1: a\x{c0}a\x{c0}
+
+/^([a\x{c0}]*)aa/8
+    a\x{c0}aaaa/ 
+ 0: a\x{c0}aaaa
+ 1: a\x{c0}aa
+    a\x{c0}a\x{c0}aaa/ 
+ 0: a\x{c0}a\x{c0}aaa
+ 1: a\x{c0}a\x{c0}a
+
+/^([a\x{c0}]*)a\x{c0}/8
+    a\x{c0}aaaa/ 
+ 0: a\x{c0}
+ 1: 
+    a\x{c0}a\x{c0}aaa/ 
+ 0: a\x{c0}a\x{c0}
+ 1: a\x{c0}
+
+/A*/g8
+    AAB\x{123}BAA
+ 0: AA
+ 0: 
+ 0: 
+ 0: 
+ 0: AA
+ 0: 
+
+/(abc)\1/8i
+   abc
+No match
+
+/(abc)\1/8
+   abc
+No match
+
+/a(*:a\x{1234}b)/8K
+    abc
+ 0: a
+MK: a\x{1234}b
+
+/a(*:a£b)/8K 
+    abc
+ 0: a
+MK: a\x{a3}b
+
+/-- Noncharacters --/
+
+/./8
+    \x{fffe}
+ 0: \x{fffe}
+    \x{ffff}
+ 0: \x{ffff}
+    \x{1fffe}
+ 0: \x{1fffe}
+    \x{1ffff}
+ 0: \x{1ffff}
+    \x{2fffe}
+ 0: \x{2fffe}
+    \x{2ffff}
+ 0: \x{2ffff}
+    \x{3fffe}
+ 0: \x{3fffe}
+    \x{3ffff}
+ 0: \x{3ffff}
+    \x{4fffe}
+ 0: \x{4fffe}
+    \x{4ffff}
+ 0: \x{4ffff}
+    \x{5fffe}
+ 0: \x{5fffe}
+    \x{5ffff}
+ 0: \x{5ffff}
+    \x{6fffe}
+ 0: \x{6fffe}
+    \x{6ffff}
+ 0: \x{6ffff}
+    \x{7fffe}
+ 0: \x{7fffe}
+    \x{7ffff}
+ 0: \x{7ffff}
+    \x{8fffe}
+ 0: \x{8fffe}
+    \x{8ffff}
+ 0: \x{8ffff}
+    \x{9fffe}
+ 0: \x{9fffe}
+    \x{9ffff}
+ 0: \x{9ffff}
+    \x{afffe}
+ 0: \x{afffe}
+    \x{affff}
+ 0: \x{affff}
+    \x{bfffe}
+ 0: \x{bfffe}
+    \x{bffff}
+ 0: \x{bffff}
+    \x{cfffe}
+ 0: \x{cfffe}
+    \x{cffff}
+ 0: \x{cffff}
+    \x{dfffe}
+ 0: \x{dfffe}
+    \x{dffff}
+ 0: \x{dffff}
+    \x{efffe}
+ 0: \x{efffe}
+    \x{effff}
+ 0: \x{effff}
+    \x{ffffe}
+ 0: \x{ffffe}
+    \x{fffff}
+ 0: \x{fffff}
+    \x{10fffe}
+ 0: \x{10fffe}
+    \x{10ffff}
+ 0: \x{10ffff}
+    \x{fdd0}
+ 0: \x{fdd0}
+    \x{fdd1}
+ 0: \x{fdd1}
+    \x{fdd2}
+ 0: \x{fdd2}
+    \x{fdd3}
+ 0: \x{fdd3}
+    \x{fdd4}
+ 0: \x{fdd4}
+    \x{fdd5}
+ 0: \x{fdd5}
+    \x{fdd6}
+ 0: \x{fdd6}
+    \x{fdd7}
+ 0: \x{fdd7}
+    \x{fdd8}
+ 0: \x{fdd8}
+    \x{fdd9}
+ 0: \x{fdd9}
+    \x{fdda}
+ 0: \x{fdda}
+    \x{fddb}
+ 0: \x{fddb}
+    \x{fddc}
+ 0: \x{fddc}
+    \x{fddd}
+ 0: \x{fddd}
+    \x{fdde}
+ 0: \x{fdde}
+    \x{fddf}
+ 0: \x{fddf}
+    \x{fde0}
+ 0: \x{fde0}
+    \x{fde1}
+ 0: \x{fde1}
+    \x{fde2}
+ 0: \x{fde2}
+    \x{fde3}
+ 0: \x{fde3}
+    \x{fde4}
+ 0: \x{fde4}
+    \x{fde5}
+ 0: \x{fde5}
+    \x{fde6}
+ 0: \x{fde6}
+    \x{fde7}
+ 0: \x{fde7}
+    \x{fde8}
+ 0: \x{fde8}
+    \x{fde9}
+ 0: \x{fde9}
+    \x{fdea}
+ 0: \x{fdea}
+    \x{fdeb}
+ 0: \x{fdeb}
+    \x{fdec}
+ 0: \x{fdec}
+    \x{fded}
+ 0: \x{fded}
+    \x{fdee}
+ 0: \x{fdee}
+    \x{fdef}
+ 0: \x{fdef}
+
+/^\d*\w{4}/8
+    1234
+ 0: 1234
+    123 
+No match
+    
+/^[^b]*\w{4}/8
+    aaaa
+ 0: aaaa
+    aaa  
+No match
+ 
+/^[^b]*\w{4}/8i
+    aaaa
+ 0: aaaa
+    aaa  
+No match
+ 
+/^\x{100}*.{4}/8
+    \x{100}\x{100}\x{100}\x{100}
+ 0: \x{100}\x{100}\x{100}\x{100}
+    \x{100}\x{100}\x{100}
+No match
+
+/^\x{100}*.{4}/8i
+    \x{100}\x{100}\x{100}\x{100}
+ 0: \x{100}\x{100}\x{100}\x{100}
+    \x{100}\x{100}\x{100}
+No match
+
+/-- End of testinput4 --/
diff --git a/pcre/testdata/testoutput5 b/pcre/testdata/testoutput5
new file mode 100644
index 00000000000..d583119dd9b
--- /dev/null
+++ b/pcre/testdata/testoutput5
@@ -0,0 +1,1849 @@
+/-- This set of tests checks the API, internals, and non-Perl stuff for UTF
+    support, excluding Unicode properties. However, tests that give different
+    results in 8-bit and 16-bit modes are excluded (see tests 16 and 17). --/
+
+/\x{110000}/8DZ
+Failed: character value in \x{...} sequence is too large at offset 9
+
+/\x{ffffffff}/8
+Failed: character value in \x{...} sequence is too large at offset 11
+
+/\x{100000000}/8
+Failed: character value in \x{...} sequence is too large at offset 12
+
+/\x{d800}/8
+Failed: disallowed Unicode code point (>= 0xd800 && <= 0xdfff) at offset 7
+
+/\x{dfff}/8
+Failed: disallowed Unicode code point (>= 0xd800 && <= 0xdfff) at offset 7
+
+/\x{d7ff}/8
+
+/\x{e000}/8
+
+/^\x{100}a\x{1234}/8
+    \x{100}a\x{1234}bcd
+ 0: \x{100}a\x{1234}
+
+/\x{0041}\x{2262}\x{0391}\x{002e}/DZ8
+------------------------------------------------------------------
+        Bra
+        A\x{2262}\x{391}.
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = 'A'
+Need char = '.'
+    \x{0041}\x{2262}\x{0391}\x{002e}
+ 0: A\x{2262}\x{391}.
+    
+/.{3,5}X/DZ8
+------------------------------------------------------------------
+        Bra
+        Any{3}
+        Any{0,2}
+        X
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+No first char
+Need char = 'X'
+    \x{212ab}\x{212ab}\x{212ab}\x{861}X
+ 0: \x{212ab}\x{212ab}\x{212ab}\x{861}X
+
+/.{3,5}?/DZ8
+------------------------------------------------------------------
+        Bra
+        Any{3}
+        Any{0,2}?
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+No first char
+No need char
+    \x{212ab}\x{212ab}\x{212ab}\x{861}
+ 0: \x{212ab}\x{212ab}\x{212ab}
+
+/(?<=\C)X/8
+Failed: \C not allowed in lookbehind assertion at offset 6
+
+/^[ab]/8DZ
+------------------------------------------------------------------
+        Bra
+        ^
+        [ab]
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: anchored utf
+No first char
+No need char
+    bar
+ 0: b
+    *** Failers
+No match
+    c
+No match
+    \x{ff}
+No match
+    \x{100}  
+No match
+
+/^[^ab]/8DZ
+------------------------------------------------------------------
+        Bra
+        ^
+        [\x00-`c-\xff] (neg)
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: anchored utf
+No first char
+No need char
+    c
+ 0: c
+    \x{ff}
+ 0: \x{ff}
+    \x{100}  
+ 0: \x{100}
+    *** Failers 
+ 0: *
+    aaa
+No match
+  
+/\x{100}*(\d+|"(?1)")/8
+    1234
+ 0: 1234
+ 1: 1234
+    "1234" 
+ 0: "1234"
+ 1: "1234"
+    \x{100}1234
+ 0: \x{100}1234
+ 1: 1234
+    "\x{100}1234"  
+ 0: \x{100}1234
+ 1: 1234
+    \x{100}\x{100}12ab 
+ 0: \x{100}\x{100}12
+ 1: 12
+    \x{100}\x{100}"12" 
+ 0: \x{100}\x{100}"12"
+ 1: "12"
+    *** Failers 
+No match
+    \x{100}\x{100}abcd
+No match
+
+/\x{100}*/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{100}*
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+No first char
+No need char
+
+/a\x{100}*/8DZ
+------------------------------------------------------------------
+        Bra
+        a
+        \x{100}*
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = 'a'
+No need char
+
+/ab\x{100}*/8DZ
+------------------------------------------------------------------
+        Bra
+        ab
+        \x{100}*
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+First char = 'a'
+Need char = 'b'
+
+/\x{100}*A/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{100}*+
+        A
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+No first char
+Need char = 'A'
+    A
+ 0: A
+
+/\x{100}*\d(?R)/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{100}*+
+        \d
+        Recurse
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+No first char
+No need char
+
+/[Z\x{100}]/8DZ
+------------------------------------------------------------------
+        Bra
+        [Z\x{100}]
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+No first char
+No need char
+    Z\x{100}
+ 0: Z
+    \x{100}
+ 0: \x{100}
+    \x{100}Z
+ 0: \x{100}
+    *** Failers 
+No match
+
+/[\x{200}-\x{100}]/8
+Failed: range out of order in character class at offset 15
+
+/[Ā-Ą]/8
+    \x{100}
+ 0: \x{100}
+    \x{104}
+ 0: \x{104}
+    *** Failers
+No match
+    \x{105}
+No match
+    \x{ff}    
+No match
+
+/[z-\x{100}]/8DZ
+------------------------------------------------------------------
+        Bra
+        [z-\x{100}]
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+No first char
+No need char
+
+/[z\Qa-d]Ā\E]/8DZ
+------------------------------------------------------------------
+        Bra
+        [\-\]adz\x{100}]
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+No first char
+No need char
+    \x{100}
+ 0: \x{100}
+    Ā 
+ 0: \x{100}
+
+/[\xFF]/DZ
+------------------------------------------------------------------
+        Bra
+        \x{ff}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+No options
+First char = \xff
+No need char
+    >\xff<
+ 0: \xff
+
+/[^\xFF]/DZ
+------------------------------------------------------------------
+        Bra
+        [^\x{ff}]
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+
+/[Ä-Ü]/8
+    Ö # Matches without Study
+ 0: \x{d6}
+    \x{d6}
+ 0: \x{d6}
+    
+/[Ä-Ü]/8S
+    Ö <-- Same with Study
+ 0: \x{d6}
+    \x{d6}
+ 0: \x{d6}
+    
+/[\x{c4}-\x{dc}]/8 
+    Ö # Matches without Study
+ 0: \x{d6}
+    \x{d6} 
+ 0: \x{d6}
+
+/[\x{c4}-\x{dc}]/8S
+    Ö <-- Same with Study
+ 0: \x{d6}
+    \x{d6} 
+ 0: \x{d6}
+
+/[^\x{100}]abc(xyz(?1))/8DZ
+------------------------------------------------------------------
+        Bra
+        [^\x{100}]
+        abc
+        CBra 1
+        xyz
+        Recurse
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 1
+Options: utf
+No first char
+Need char = 'z'
+
+/[ab\x{100}]abc(xyz(?1))/8DZ
+------------------------------------------------------------------
+        Bra
+        [ab\x{100}]
+        abc
+        CBra 1
+        xyz
+        Recurse
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 1
+Options: utf
+No first char
+Need char = 'z'
+
+/(\x{100}(b(?2)c))?/DZ8
+------------------------------------------------------------------
+        Bra
+        Brazero
+        CBra 1
+        \x{100}
+        CBra 2
+        b
+        Recurse
+        c
+        Ket
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 2
+Options: utf
+No first char
+No need char
+
+/(\x{100}(b(?2)c)){0,2}/DZ8
+------------------------------------------------------------------
+        Bra
+        Brazero
+        Bra
+        CBra 1
+        \x{100}
+        CBra 2
+        b
+        Recurse
+        c
+        Ket
+        Ket
+        Brazero
+        CBra 1
+        \x{100}
+        CBra 2
+        b
+        Recurse
+        c
+        Ket
+        Ket
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 2
+Options: utf
+No first char
+No need char
+
+/(\x{100}(b(?1)c))?/DZ8
+------------------------------------------------------------------
+        Bra
+        Brazero
+        CBra 1
+        \x{100}
+        CBra 2
+        b
+        Recurse
+        c
+        Ket
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 2
+Options: utf
+No first char
+No need char
+
+/(\x{100}(b(?1)c)){0,2}/DZ8
+------------------------------------------------------------------
+        Bra
+        Brazero
+        Bra
+        CBra 1
+        \x{100}
+        CBra 2
+        b
+        Recurse
+        c
+        Ket
+        Ket
+        Brazero
+        CBra 1
+        \x{100}
+        CBra 2
+        b
+        Recurse
+        c
+        Ket
+        Ket
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 2
+Options: utf
+No first char
+No need char
+
+/\W/8
+    A.B
+ 0: .
+    A\x{100}B 
+ 0: \x{100}
+  
+/\w/8
+    \x{100}X   
+ 0: X
+
+/^\ሴ/8DZ
+------------------------------------------------------------------
+        Bra
+        ^
+        \x{1234}
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: anchored utf
+No first char
+No need char
+
+/\x{100}*\d/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{100}*+
+        \d
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+No first char
+No need char
+
+/\x{100}*\s/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{100}*+
+        \s
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+No first char
+No need char
+
+/\x{100}*\w/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{100}*+
+        \w
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+No first char
+No need char
+
+/\x{100}*\D/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{100}*
+        \D
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+No first char
+No need char
+
+/\x{100}*\S/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{100}*
+        \S
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+No first char
+No need char
+
+/\x{100}*\W/8DZ
+------------------------------------------------------------------
+        Bra
+        \x{100}*
+        \W
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+No first char
+No need char
+
+/()()()()()()()()()()
+ ()()()()()()()()()()
+ ()()()()()()()()()()
+ ()()()()()()()()()()
+ A (x) (?41) B/8x
+    AxxB     
+Matched, but too many substrings
+ 0: AxxB
+ 1: 
+ 2: 
+ 3: 
+ 4: 
+ 5: 
+ 6: 
+ 7: 
+ 8: 
+ 9: 
+10: 
+11: 
+12: 
+13: 
+14: 
+
+/^[\x{100}\E-\Q\E\x{150}]/BZ8
+------------------------------------------------------------------
+        Bra
+        ^
+        [\x{100}-\x{150}]
+        Ket
+        End
+------------------------------------------------------------------
+
+/^[\QĀ\E-\QŐ\E]/BZ8
+------------------------------------------------------------------
+        Bra
+        ^
+        [\x{100}-\x{150}]
+        Ket
+        End
+------------------------------------------------------------------
+
+/^abc./mgx8
+    abc1 \x0aabc2 \x0babc3xx \x0cabc4 \x0dabc5xx \x0d\x0aabc6 \x{0085}abc7 \x{2028}abc8 \x{2029}abc9 JUNK
+ 0: abc1
+ 0: abc2
+ 0: abc3
+ 0: abc4
+ 0: abc5
+ 0: abc6
+ 0: abc7
+ 0: abc8
+ 0: abc9
+
+/abc.$/mgx8
+    abc1\x0a abc2\x0b abc3\x0c abc4\x0d abc5\x0d\x0a abc6\x{0085} abc7\x{2028} abc8\x{2029} abc9
+ 0: abc1
+ 0: abc2
+ 0: abc3
+ 0: abc4
+ 0: abc5
+ 0: abc6
+ 0: abc7
+ 0: abc8
+ 0: abc9
+
+/^a\Rb/8
+    a\nb
+ 0: a\x{0a}b
+    a\rb
+ 0: a\x{0d}b
+    a\r\nb
+ 0: a\x{0d}\x{0a}b
+    a\x0bb
+ 0: a\x{0b}b
+    a\x0cb
+ 0: a\x{0c}b
+    a\x{85}b   
+ 0: a\x{85}b
+    a\x{2028}b 
+ 0: a\x{2028}b
+    a\x{2029}b 
+ 0: a\x{2029}b
+    ** Failers
+No match
+    a\n\rb    
+No match
+
+/^a\R*b/8
+    ab
+ 0: ab
+    a\nb
+ 0: a\x{0a}b
+    a\rb
+ 0: a\x{0d}b
+    a\r\nb
+ 0: a\x{0d}\x{0a}b
+    a\x0bb
+ 0: a\x{0b}b
+    a\x0c\x{2028}\x{2029}b
+ 0: a\x{0c}\x{2028}\x{2029}b
+    a\x{85}b   
+ 0: a\x{85}b
+    a\n\rb    
+ 0: a\x{0a}\x{0d}b
+    a\n\r\x{85}\x0cb 
+ 0: a\x{0a}\x{0d}\x{85}\x{0c}b
+
+/^a\R+b/8
+    a\nb
+ 0: a\x{0a}b
+    a\rb
+ 0: a\x{0d}b
+    a\r\nb
+ 0: a\x{0d}\x{0a}b
+    a\x0bb
+ 0: a\x{0b}b
+    a\x0c\x{2028}\x{2029}b
+ 0: a\x{0c}\x{2028}\x{2029}b
+    a\x{85}b   
+ 0: a\x{85}b
+    a\n\rb    
+ 0: a\x{0a}\x{0d}b
+    a\n\r\x{85}\x0cb 
+ 0: a\x{0a}\x{0d}\x{85}\x{0c}b
+    ** Failers
+No match
+    ab  
+No match
+
+/^a\R{1,3}b/8
+    a\nb
+ 0: a\x{0a}b
+    a\n\rb
+ 0: a\x{0a}\x{0d}b
+    a\n\r\x{85}b
+ 0: a\x{0a}\x{0d}\x{85}b
+    a\r\n\r\nb 
+ 0: a\x{0d}\x{0a}\x{0d}\x{0a}b
+    a\r\n\r\n\r\nb 
+ 0: a\x{0d}\x{0a}\x{0d}\x{0a}\x{0d}\x{0a}b
+    a\n\r\n\rb
+ 0: a\x{0a}\x{0d}\x{0a}\x{0d}b
+    a\n\n\r\nb 
+ 0: a\x{0a}\x{0a}\x{0d}\x{0a}b
+    ** Failers
+No match
+    a\n\n\n\rb
+No match
+    a\r
+No match
+
+/\H\h\V\v/8
+    X X\x0a
+ 0: X X\x{0a}
+    X\x09X\x0b
+ 0: X\x{09}X\x{0b}
+    ** Failers
+No match
+    \x{a0} X\x0a   
+No match
+    
+/\H*\h+\V?\v{3,4}/8 
+    \x09\x20\x{a0}X\x0a\x0b\x0c\x0d\x0a
+ 0: \x{09} \x{a0}X\x{0a}\x{0b}\x{0c}\x{0d}
+    \x09\x20\x{a0}\x0a\x0b\x0c\x0d\x0a
+ 0: \x{09} \x{a0}\x{0a}\x{0b}\x{0c}\x{0d}
+    \x09\x20\x{a0}\x0a\x0b\x0c
+ 0: \x{09} \x{a0}\x{0a}\x{0b}\x{0c}
+    ** Failers 
+No match
+    \x09\x20\x{a0}\x0a\x0b
+No match
+     
+/\H\h\V\v/8
+    \x{3001}\x{3000}\x{2030}\x{2028}
+ 0: \x{3001}\x{3000}\x{2030}\x{2028}
+    X\x{180e}X\x{85}
+ 0: X\x{180e}X\x{85}
+    ** Failers
+No match
+    \x{2009} X\x0a   
+No match
+    
+/\H*\h+\V?\v{3,4}/8 
+    \x{1680}\x{180e}\x{2007}X\x{2028}\x{2029}\x0c\x0d\x0a
+ 0: \x{1680}\x{180e}\x{2007}X\x{2028}\x{2029}\x{0c}\x{0d}
+    \x09\x{205f}\x{a0}\x0a\x{2029}\x0c\x{2028}\x0a
+ 0: \x{09}\x{205f}\x{a0}\x{0a}\x{2029}\x{0c}\x{2028}
+    \x09\x20\x{202f}\x0a\x0b\x0c
+ 0: \x{09} \x{202f}\x{0a}\x{0b}\x{0c}
+    ** Failers 
+No match
+    \x09\x{200a}\x{a0}\x{2028}\x0b
+No match
+     
+/[\h]/8BZ
+------------------------------------------------------------------
+        Bra
+        [\x09 \xa0\x{1680}\x{180e}\x{2000}-\x{200a}\x{202f}\x{205f}\x{3000}]
+        Ket
+        End
+------------------------------------------------------------------
+    >\x{1680}
+ 0: \x{1680}
+
+/[\h]{3,}/8BZ
+------------------------------------------------------------------
+        Bra
+        [\x09 \xa0\x{1680}\x{180e}\x{2000}-\x{200a}\x{202f}\x{205f}\x{3000}]{3,}
+        Ket
+        End
+------------------------------------------------------------------
+    >\x{1680}\x{180e}\x{2000}\x{2003}\x{200a}\x{202f}\x{205f}\x{3000}<
+ 0: \x{1680}\x{180e}\x{2000}\x{2003}\x{200a}\x{202f}\x{205f}\x{3000}
+
+/[\v]/8BZ
+------------------------------------------------------------------
+        Bra
+        [\x0a-\x0d\x85\x{2028}-\x{2029}]
+        Ket
+        End
+------------------------------------------------------------------
+
+/[\H]/8BZ
+------------------------------------------------------------------
+        Bra
+        [\x00-\x08\x0a-\x1f!-\x9f\x{a1}-\x{167f}\x{1681}-\x{180d}\x{180f}-\x{1fff}\x{200b}-\x{202e}\x{2030}-\x{205e}\x{2060}-\x{2fff}\x{3001}-\x{10ffff}]
+        Ket
+        End
+------------------------------------------------------------------
+
+/[\V]/8BZ
+------------------------------------------------------------------
+        Bra
+        [\x00-\x09\x0e-\x84\x{86}-\x{2027}\x{202a}-\x{10ffff}]
+        Ket
+        End
+------------------------------------------------------------------
+
+/.*$/8
+    \x{1ec5} 
+ 0: \x{1ec5}
+    
+/a\Rb/I8
+Capturing subpattern count = 0
+Options: bsr_anycrlf utf
+First char = 'a'
+Need char = 'b'
+    a\rb
+ 0: a\x{0d}b
+    a\nb
+ 0: a\x{0a}b
+    a\r\nb
+ 0: a\x{0d}\x{0a}b
+    ** Failers
+No match
+    a\x{85}b
+No match
+    a\x0bb     
+No match
+
+/a\Rb/I8
+Capturing subpattern count = 0
+Options: bsr_unicode utf
+First char = 'a'
+Need char = 'b'
+    a\rb
+ 0: a\x{0d}b
+    a\nb
+ 0: a\x{0a}b
+    a\r\nb
+ 0: a\x{0d}\x{0a}b
+    a\x{85}b
+ 0: a\x{85}b
+    a\x0bb     
+ 0: a\x{0b}b
+    ** Failers 
+No match
+    a\x{85}b\
+No match
+    a\x0bb\
+No match
+    
+/a\R?b/I8
+Capturing subpattern count = 0
+Options: bsr_anycrlf utf
+First char = 'a'
+Need char = 'b'
+    a\rb
+ 0: a\x{0d}b
+    a\nb
+ 0: a\x{0a}b
+    a\r\nb
+ 0: a\x{0d}\x{0a}b
+    ** Failers
+No match
+    a\x{85}b
+No match
+    a\x0bb     
+No match
+
+/a\R?b/I8
+Capturing subpattern count = 0
+Options: bsr_unicode utf
+First char = 'a'
+Need char = 'b'
+    a\rb
+ 0: a\x{0d}b
+    a\nb
+ 0: a\x{0a}b
+    a\r\nb
+ 0: a\x{0d}\x{0a}b
+    a\x{85}b
+ 0: a\x{85}b
+    a\x0bb     
+ 0: a\x{0b}b
+    ** Failers 
+No match
+    a\x{85}b\
+No match
+    a\x0bb\
+No match
+ 
+/.*a.*=.b.*/8
+    QQQ\x{2029}ABCaXYZ=!bPQR
+ 0: ABCaXYZ=!bPQR
+    ** Failers
+No match
+    a\x{2029}b
+No match
+    \x61\xe2\x80\xa9\x62 
+No match
+
+/[[:a\x{100}b:]]/8
+Failed: unknown POSIX class name at offset 3
+
+/a[^]b/8
+    a\x{1234}b
+ 0: a\x{1234}b
+    a\nb 
+ 0: a\x{0a}b
+    ** Failers
+No match
+    ab  
+No match
+    
+/a[^]+b/8
+    aXb
+ 0: aXb
+    a\nX\nX\x{1234}b 
+ 0: a\x{0a}X\x{0a}X\x{1234}b
+    ** Failers
+No match
+    ab  
+No match
+
+/(\x{de})\1/
+    \x{de}\x{de}
+ 0: \xde\xde
+ 1: \xde
+
+/X/8f 
+    A\x{1ec5}ABCXYZ
+ 0: X
+
+/Xa{2,4}b/8
+    X\P
+Partial match: X
+    Xa\P
+Partial match: Xa
+    Xaa\P 
+Partial match: Xaa
+    Xaaa\P
+Partial match: Xaaa
+    Xaaaa\P 
+Partial match: Xaaaa
+    
+/Xa{2,4}?b/8
+    X\P
+Partial match: X
+    Xa\P
+Partial match: Xa
+    Xaa\P 
+Partial match: Xaa
+    Xaaa\P
+Partial match: Xaaa
+    Xaaaa\P 
+Partial match: Xaaaa
+    
+/Xa{2,4}+b/8
+    X\P
+Partial match: X
+    Xa\P
+Partial match: Xa
+    Xaa\P 
+Partial match: Xaa
+    Xaaa\P
+Partial match: Xaaa
+    Xaaaa\P 
+Partial match: Xaaaa
+    
+/X\x{123}{2,4}b/8
+    X\P
+Partial match: X
+    X\x{123}\P
+Partial match: X\x{123}
+    X\x{123}\x{123}\P 
+Partial match: X\x{123}\x{123}
+    X\x{123}\x{123}\x{123}\P
+Partial match: X\x{123}\x{123}\x{123}
+    X\x{123}\x{123}\x{123}\x{123}\P 
+Partial match: X\x{123}\x{123}\x{123}\x{123}
+    
+/X\x{123}{2,4}?b/8
+    X\P
+Partial match: X
+    X\x{123}\P
+Partial match: X\x{123}
+    X\x{123}\x{123}\P 
+Partial match: X\x{123}\x{123}
+    X\x{123}\x{123}\x{123}\P
+Partial match: X\x{123}\x{123}\x{123}
+    X\x{123}\x{123}\x{123}\x{123}\P 
+Partial match: X\x{123}\x{123}\x{123}\x{123}
+    
+/X\x{123}{2,4}+b/8
+    X\P
+Partial match: X
+    X\x{123}\P
+Partial match: X\x{123}
+    X\x{123}\x{123}\P 
+Partial match: X\x{123}\x{123}
+    X\x{123}\x{123}\x{123}\P
+Partial match: X\x{123}\x{123}\x{123}
+    X\x{123}\x{123}\x{123}\x{123}\P 
+Partial match: X\x{123}\x{123}\x{123}\x{123}
+    
+/X\x{123}{2,4}b/8
+    Xx\P
+No match
+    X\x{123}x\P
+No match
+    X\x{123}\x{123}x\P 
+No match
+    X\x{123}\x{123}\x{123}x\P
+No match
+    X\x{123}\x{123}\x{123}\x{123}x\P 
+No match
+    
+/X\x{123}{2,4}?b/8
+    Xx\P
+No match
+    X\x{123}x\P
+No match
+    X\x{123}\x{123}x\P 
+No match
+    X\x{123}\x{123}\x{123}x\P
+No match
+    X\x{123}\x{123}\x{123}\x{123}x\P 
+No match
+    
+/X\x{123}{2,4}+b/8
+    Xx\P
+No match
+    X\x{123}x\P
+No match
+    X\x{123}\x{123}x\P 
+No match
+    X\x{123}\x{123}\x{123}x\P
+No match
+    X\x{123}\x{123}\x{123}\x{123}x\P 
+No match
+    
+/X\d{2,4}b/8
+    X\P
+Partial match: X
+    X3\P
+Partial match: X3
+    X33\P 
+Partial match: X33
+    X333\P
+Partial match: X333
+    X3333\P 
+Partial match: X3333
+    
+/X\d{2,4}?b/8
+    X\P
+Partial match: X
+    X3\P
+Partial match: X3
+    X33\P 
+Partial match: X33
+    X333\P
+Partial match: X333
+    X3333\P 
+Partial match: X3333
+    
+/X\d{2,4}+b/8
+    X\P
+Partial match: X
+    X3\P
+Partial match: X3
+    X33\P 
+Partial match: X33
+    X333\P
+Partial match: X333
+    X3333\P 
+Partial match: X3333
+
+/X\D{2,4}b/8
+    X\P
+Partial match: X
+    Xa\P
+Partial match: Xa
+    Xaa\P 
+Partial match: Xaa
+    Xaaa\P
+Partial match: Xaaa
+    Xaaaa\P 
+Partial match: Xaaaa
+    
+/X\D{2,4}?b/8
+    X\P
+Partial match: X
+    Xa\P
+Partial match: Xa
+    Xaa\P 
+Partial match: Xaa
+    Xaaa\P
+Partial match: Xaaa
+    Xaaaa\P 
+Partial match: Xaaaa
+    
+/X\D{2,4}+b/8
+    X\P
+Partial match: X
+    Xa\P
+Partial match: Xa
+    Xaa\P 
+Partial match: Xaa
+    Xaaa\P
+Partial match: Xaaa
+    Xaaaa\P 
+Partial match: Xaaaa
+
+/X\D{2,4}b/8
+    X\P
+Partial match: X
+    X\x{123}\P
+Partial match: X\x{123}
+    X\x{123}\x{123}\P 
+Partial match: X\x{123}\x{123}
+    X\x{123}\x{123}\x{123}\P
+Partial match: X\x{123}\x{123}\x{123}
+    X\x{123}\x{123}\x{123}\x{123}\P 
+Partial match: X\x{123}\x{123}\x{123}\x{123}
+    
+/X\D{2,4}?b/8
+    X\P
+Partial match: X
+    X\x{123}\P
+Partial match: X\x{123}
+    X\x{123}\x{123}\P 
+Partial match: X\x{123}\x{123}
+    X\x{123}\x{123}\x{123}\P
+Partial match: X\x{123}\x{123}\x{123}
+    X\x{123}\x{123}\x{123}\x{123}\P 
+Partial match: X\x{123}\x{123}\x{123}\x{123}
+    
+/X\D{2,4}+b/8
+    X\P
+Partial match: X
+    X\x{123}\P
+Partial match: X\x{123}
+    X\x{123}\x{123}\P 
+Partial match: X\x{123}\x{123}
+    X\x{123}\x{123}\x{123}\P
+Partial match: X\x{123}\x{123}\x{123}
+    X\x{123}\x{123}\x{123}\x{123}\P 
+Partial match: X\x{123}\x{123}\x{123}\x{123}
+
+/X[abc]{2,4}b/8
+    X\P
+Partial match: X
+    Xa\P
+Partial match: Xa
+    Xaa\P 
+Partial match: Xaa
+    Xaaa\P
+Partial match: Xaaa
+    Xaaaa\P 
+Partial match: Xaaaa
+    
+/X[abc]{2,4}?b/8
+    X\P
+Partial match: X
+    Xa\P
+Partial match: Xa
+    Xaa\P 
+Partial match: Xaa
+    Xaaa\P
+Partial match: Xaaa
+    Xaaaa\P 
+Partial match: Xaaaa
+    
+/X[abc]{2,4}+b/8
+    X\P
+Partial match: X
+    Xa\P
+Partial match: Xa
+    Xaa\P 
+Partial match: Xaa
+    Xaaa\P
+Partial match: Xaaa
+    Xaaaa\P 
+Partial match: Xaaaa
+
+/X[abc\x{123}]{2,4}b/8
+    X\P
+Partial match: X
+    X\x{123}\P
+Partial match: X\x{123}
+    X\x{123}\x{123}\P 
+Partial match: X\x{123}\x{123}
+    X\x{123}\x{123}\x{123}\P
+Partial match: X\x{123}\x{123}\x{123}
+    X\x{123}\x{123}\x{123}\x{123}\P 
+Partial match: X\x{123}\x{123}\x{123}\x{123}
+    
+/X[abc\x{123}]{2,4}?b/8
+    X\P
+Partial match: X
+    X\x{123}\P
+Partial match: X\x{123}
+    X\x{123}\x{123}\P 
+Partial match: X\x{123}\x{123}
+    X\x{123}\x{123}\x{123}\P
+Partial match: X\x{123}\x{123}\x{123}
+    X\x{123}\x{123}\x{123}\x{123}\P 
+Partial match: X\x{123}\x{123}\x{123}\x{123}
+    
+/X[abc\x{123}]{2,4}+b/8
+    X\P
+Partial match: X
+    X\x{123}\P
+Partial match: X\x{123}
+    X\x{123}\x{123}\P 
+Partial match: X\x{123}\x{123}
+    X\x{123}\x{123}\x{123}\P
+Partial match: X\x{123}\x{123}\x{123}
+    X\x{123}\x{123}\x{123}\x{123}\P 
+Partial match: X\x{123}\x{123}\x{123}\x{123}
+
+/X[^a]{2,4}b/8
+    X\P
+Partial match: X
+    Xz\P
+Partial match: Xz
+    Xzz\P 
+Partial match: Xzz
+    Xzzz\P
+Partial match: Xzzz
+    Xzzzz\P 
+Partial match: Xzzzz
+    
+/X[^a]{2,4}?b/8
+    X\P
+Partial match: X
+    Xz\P
+Partial match: Xz
+    Xzz\P 
+Partial match: Xzz
+    Xzzz\P
+Partial match: Xzzz
+    Xzzzz\P 
+Partial match: Xzzzz
+    
+/X[^a]{2,4}+b/8
+    X\P
+Partial match: X
+    Xz\P
+Partial match: Xz
+    Xzz\P 
+Partial match: Xzz
+    Xzzz\P
+Partial match: Xzzz
+    Xzzzz\P 
+Partial match: Xzzzz
+
+/X[^a]{2,4}b/8
+    X\P
+Partial match: X
+    X\x{123}\P
+Partial match: X\x{123}
+    X\x{123}\x{123}\P 
+Partial match: X\x{123}\x{123}
+    X\x{123}\x{123}\x{123}\P
+Partial match: X\x{123}\x{123}\x{123}
+    X\x{123}\x{123}\x{123}\x{123}\P 
+Partial match: X\x{123}\x{123}\x{123}\x{123}
+    
+/X[^a]{2,4}?b/8
+    X\P
+Partial match: X
+    X\x{123}\P
+Partial match: X\x{123}
+    X\x{123}\x{123}\P 
+Partial match: X\x{123}\x{123}
+    X\x{123}\x{123}\x{123}\P
+Partial match: X\x{123}\x{123}\x{123}
+    X\x{123}\x{123}\x{123}\x{123}\P 
+Partial match: X\x{123}\x{123}\x{123}\x{123}
+    
+/X[^a]{2,4}+b/8
+    X\P
+Partial match: X
+    X\x{123}\P
+Partial match: X\x{123}
+    X\x{123}\x{123}\P 
+Partial match: X\x{123}\x{123}
+    X\x{123}\x{123}\x{123}\P
+Partial match: X\x{123}\x{123}\x{123}
+    X\x{123}\x{123}\x{123}\x{123}\P 
+Partial match: X\x{123}\x{123}\x{123}\x{123}
+
+/(Y)X\1{2,4}b/8
+    YX\P
+Partial match: YX
+    YXY\P
+Partial match: YXY
+    YXYY\P 
+Partial match: YXYY
+    YXYYY\P
+Partial match: YXYYY
+    YXYYYY\P 
+Partial match: YXYYYY
+    
+/(Y)X\1{2,4}?b/8
+    YX\P
+Partial match: YX
+    YXY\P
+Partial match: YXY
+    YXYY\P 
+Partial match: YXYY
+    YXYYY\P
+Partial match: YXYYY
+    YXYYYY\P 
+Partial match: YXYYYY
+    
+/(Y)X\1{2,4}+b/8
+    YX\P
+Partial match: YX
+    YXY\P
+Partial match: YXY
+    YXYY\P 
+Partial match: YXYY
+    YXYYY\P
+Partial match: YXYYY
+    YXYYYY\P 
+Partial match: YXYYYY
+
+/(\x{123})X\1{2,4}b/8
+    \x{123}X\P
+Partial match: \x{123}X
+    \x{123}X\x{123}\P
+Partial match: \x{123}X\x{123}
+    \x{123}X\x{123}\x{123}\P 
+Partial match: \x{123}X\x{123}\x{123}
+    \x{123}X\x{123}\x{123}\x{123}\P
+Partial match: \x{123}X\x{123}\x{123}\x{123}
+    \x{123}X\x{123}\x{123}\x{123}\x{123}\P 
+Partial match: \x{123}X\x{123}\x{123}\x{123}\x{123}
+    
+/(\x{123})X\1{2,4}?b/8
+    \x{123}X\P
+Partial match: \x{123}X
+    \x{123}X\x{123}\P
+Partial match: \x{123}X\x{123}
+    \x{123}X\x{123}\x{123}\P 
+Partial match: \x{123}X\x{123}\x{123}
+    \x{123}X\x{123}\x{123}\x{123}\P
+Partial match: \x{123}X\x{123}\x{123}\x{123}
+    \x{123}X\x{123}\x{123}\x{123}\x{123}\P 
+Partial match: \x{123}X\x{123}\x{123}\x{123}\x{123}
+    
+/(\x{123})X\1{2,4}+b/8
+    \x{123}X\P
+Partial match: \x{123}X
+    \x{123}X\x{123}\P
+Partial match: \x{123}X\x{123}
+    \x{123}X\x{123}\x{123}\P 
+Partial match: \x{123}X\x{123}\x{123}
+    \x{123}X\x{123}\x{123}\x{123}\P
+Partial match: \x{123}X\x{123}\x{123}\x{123}
+    \x{123}X\x{123}\x{123}\x{123}\x{123}\P 
+Partial match: \x{123}X\x{123}\x{123}\x{123}\x{123}
+
+/\bthe cat\b/8
+    the cat\P
+ 0: the cat
+    the cat\P\P
+Partial match: the cat
+
+/abcd*/8
+    xxxxabcd\P
+ 0: abcd
+    xxxxabcd\P\P
+Partial match: abcd
+
+/abcd*/i8
+    xxxxabcd\P
+ 0: abcd
+    xxxxabcd\P\P
+Partial match: abcd
+    XXXXABCD\P
+ 0: ABCD
+    XXXXABCD\P\P
+Partial match: ABCD
+
+/abc\d*/8
+    xxxxabc1\P
+ 0: abc1
+    xxxxabc1\P\P
+Partial match: abc1
+
+/(a)bc\1*/8
+    xxxxabca\P
+ 0: abca
+ 1: a
+    xxxxabca\P\P
+Partial match: abca
+
+/abc[de]*/8
+    xxxxabcde\P
+ 0: abcde
+    xxxxabcde\P\P
+Partial match: abcde
+
+/X\W{3}X/8
+    \PX
+Partial match: X
+
+/\sxxx\s/8T1
+    AB\x{85}xxx\x{a0}XYZ
+ 0: \x{85}xxx\x{a0}
+    AB\x{a0}xxx\x{85}XYZ
+ 0: \x{a0}xxx\x{85}
+
+/\S \S/8T1
+    \x{a2} \x{84} 
+ 0: \x{a2} \x{84}
+
+'A#хц'8xBZ
+------------------------------------------------------------------
+        Bra
+        A
+        Ket
+        End
+------------------------------------------------------------------
+
+'A#хц
+  PQ'8xBZ
+------------------------------------------------------------------
+        Bra
+        APQ
+        Ket
+        End
+------------------------------------------------------------------
+  
+/a+#хaa
+  z#XX?/8xBZ 
+------------------------------------------------------------------
+        Bra
+        a++
+        z
+        Ket
+        End
+------------------------------------------------------------------
+
+/a+#хaa
+  z#х?/8xBZ 
+------------------------------------------------------------------
+        Bra
+        a++
+        z
+        Ket
+        End
+------------------------------------------------------------------
+
+/\g{A}xxx#bXX(?'A'123)
(?'A'456)/8xBZ
+------------------------------------------------------------------
+        Bra
+        \1
+        xxx
+        CBra 1
+        456
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+
+/\g{A}xxx#bх(?'A'123)
(?'A'456)/8xBZ
+------------------------------------------------------------------
+        Bra
+        \1
+        xxx
+        CBra 1
+        456
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+
+/^\cģ/8
+Failed: \c must be followed by an ASCII character at offset 3
+
+/(\R*)(.)/s8
+    \r\n
+ 0: \x{0d}
+ 1: 
+ 2: \x{0d}
+    \r\r\n\n\r 
+ 0: \x{0d}\x{0d}\x{0a}\x{0a}\x{0d}
+ 1: \x{0d}\x{0d}\x{0a}\x{0a}
+ 2: \x{0d}
+    \r\r\n\n\r\n 
+ 0: \x{0d}\x{0d}\x{0a}\x{0a}\x{0d}
+ 1: \x{0d}\x{0d}\x{0a}\x{0a}
+ 2: \x{0d}
+
+/(\R)*(.)/s8
+    \r\n
+ 0: \x{0d}
+ 1: 
+ 2: \x{0d}
+    \r\r\n\n\r 
+ 0: \x{0d}\x{0d}\x{0a}\x{0a}\x{0d}
+ 1: \x{0a}
+ 2: \x{0d}
+    \r\r\n\n\r\n 
+ 0: \x{0d}\x{0d}\x{0a}\x{0a}\x{0d}
+ 1: \x{0a}
+ 2: \x{0d}
+
+/[^\x{1234}]+/iS8I   
+Capturing subpattern count = 0
+Options: caseless utf
+No first char
+No need char
+Subject length lower bound = 1
+No set of starting bytes
+
+/[^\x{1234}]+?/iS8I   
+Capturing subpattern count = 0
+Options: caseless utf
+No first char
+No need char
+Subject length lower bound = 1
+No set of starting bytes
+
+/[^\x{1234}]++/iS8I   
+Capturing subpattern count = 0
+Options: caseless utf
+No first char
+No need char
+Subject length lower bound = 1
+No set of starting bytes
+
+/[^\x{1234}]{2}/iS8I
+Capturing subpattern count = 0
+Options: caseless utf
+No first char
+No need char
+Subject length lower bound = 2
+No set of starting bytes
+
+//
+Failed: inconsistent NEWLINE options at offset 0
+
+/f.*/
+    \P\Pfor
+Partial match: for
+
+/f.*/s
+    \P\Pfor
+Partial match: for
+
+/f.*/8
+    \P\Pfor
+Partial match: for
+
+/f.*/8s
+    \P\Pfor
+Partial match: for
+    
+/\x{d7ff}\x{e000}/8
+
+/\x{d800}/8
+Failed: disallowed Unicode code point (>= 0xd800 && <= 0xdfff) at offset 7
+
+/\x{dfff}/8 
+Failed: disallowed Unicode code point (>= 0xd800 && <= 0xdfff) at offset 7
+
+/\h+/8
+    \x{1681}\x{200b}\x{1680}\x{2000}\x{202f}\x{3000}
+ 0: \x{1680}\x{2000}\x{202f}\x{3000}
+    \x{3001}\x{2fff}\x{200a}\x{a0}\x{2000}
+ 0: \x{200a}\x{a0}\x{2000}
+
+/[\h\x{e000}]+/8BZ
+------------------------------------------------------------------
+        Bra
+        [\x09 \xa0\x{1680}\x{180e}\x{2000}-\x{200a}\x{202f}\x{205f}\x{3000}\x{e000}]+
+        Ket
+        End
+------------------------------------------------------------------
+    \x{1681}\x{200b}\x{1680}\x{2000}\x{202f}\x{3000}
+ 0: \x{1680}\x{2000}\x{202f}\x{3000}
+    \x{3001}\x{2fff}\x{200a}\x{a0}\x{2000}
+ 0: \x{200a}\x{a0}\x{2000}
+
+/\H+/8
+    \x{1680}\x{180e}\x{167f}\x{1681}\x{180d}\x{180f}
+ 0: \x{167f}\x{1681}\x{180d}\x{180f}
+    \x{2000}\x{200a}\x{1fff}\x{200b}
+ 0: \x{1fff}\x{200b}
+    \x{202f}\x{205f}\x{202e}\x{2030}\x{205e}\x{2060}
+ 0: \x{202e}\x{2030}\x{205e}\x{2060}
+    \x{a0}\x{3000}\x{9f}\x{a1}\x{2fff}\x{3001}
+ 0: \x{9f}\x{a1}\x{2fff}\x{3001}
+
+/[\H\x{d7ff}]+/8BZ
+------------------------------------------------------------------
+        Bra
+        [\x00-\x08\x0a-\x1f!-\x9f\x{a1}-\x{167f}\x{1681}-\x{180d}\x{180f}-\x{1fff}\x{200b}-\x{202e}\x{2030}-\x{205e}\x{2060}-\x{2fff}\x{3001}-\x{10ffff}\x{d7ff}]+
+        Ket
+        End
+------------------------------------------------------------------
+    \x{1680}\x{180e}\x{167f}\x{1681}\x{180d}\x{180f}
+ 0: \x{167f}\x{1681}\x{180d}\x{180f}
+    \x{2000}\x{200a}\x{1fff}\x{200b}
+ 0: \x{1fff}\x{200b}
+    \x{202f}\x{205f}\x{202e}\x{2030}\x{205e}\x{2060}
+ 0: \x{202e}\x{2030}\x{205e}\x{2060}
+    \x{a0}\x{3000}\x{9f}\x{a1}\x{2fff}\x{3001}
+ 0: \x{9f}\x{a1}\x{2fff}\x{3001}
+
+/\v+/8
+    \x{2027}\x{2030}\x{2028}\x{2029}
+ 0: \x{2028}\x{2029}
+    \x09\x0e\x{84}\x{86}\x{85}\x0a\x0b\x0c\x0d
+ 0: \x{85}\x{0a}\x{0b}\x{0c}\x{0d}
+
+/[\v\x{e000}]+/8BZ
+------------------------------------------------------------------
+        Bra
+        [\x0a-\x0d\x85\x{2028}-\x{2029}\x{e000}]+
+        Ket
+        End
+------------------------------------------------------------------
+    \x{2027}\x{2030}\x{2028}\x{2029}
+ 0: \x{2028}\x{2029}
+    \x09\x0e\x{84}\x{86}\x{85}\x0a\x0b\x0c\x0d
+ 0: \x{85}\x{0a}\x{0b}\x{0c}\x{0d}
+
+/\V+/8
+    \x{2028}\x{2029}\x{2027}\x{2030}
+ 0: \x{2027}\x{2030}
+    \x{85}\x0a\x0b\x0c\x0d\x09\x0e\x{84}\x{86}
+ 0: \x{09}\x{0e}\x{84}\x{86}
+
+/[\V\x{d7ff}]+/8BZ
+------------------------------------------------------------------
+        Bra
+        [\x00-\x09\x0e-\x84\x{86}-\x{2027}\x{202a}-\x{10ffff}\x{d7ff}]+
+        Ket
+        End
+------------------------------------------------------------------
+    \x{2028}\x{2029}\x{2027}\x{2030}
+ 0: \x{2027}\x{2030}
+    \x{85}\x0a\x0b\x0c\x0d\x09\x0e\x{84}\x{86}
+ 0: \x{09}\x{0e}\x{84}\x{86}
+
+/\R+/8
+    \x{2027}\x{2030}\x{2028}\x{2029}
+ 0: \x{2028}\x{2029}
+    \x09\x0e\x{84}\x{86}\x{85}\x0a\x0b\x0c\x0d
+ 0: \x{85}\x{0a}\x{0b}\x{0c}\x{0d}
+
+/(..)\1/8
+    ab\P
+Partial match: ab
+    aba\P
+Partial match: aba
+    abab\P
+ 0: abab
+ 1: ab
+
+/(..)\1/8i
+    ab\P
+Partial match: ab
+    abA\P
+Partial match: abA
+    aBAb\P
+ 0: aBAb
+ 1: aB
+
+/(..)\1{2,}/8
+    ab\P
+Partial match: ab
+    aba\P
+Partial match: aba
+    abab\P
+Partial match: abab
+    ababa\P
+Partial match: ababa
+    ababab\P
+ 0: ababab
+ 1: ab
+    ababab\P\P
+Partial match: ababab
+    abababa\P
+ 0: ababab
+ 1: ab
+    abababa\P\P
+Partial match: abababa
+
+/(..)\1{2,}/8i
+    ab\P
+Partial match: ab
+    aBa\P
+Partial match: aBa
+    aBAb\P
+Partial match: aBAb
+    AbaBA\P
+Partial match: AbaBA
+    abABAb\P
+ 0: abABAb
+ 1: ab
+    aBAbaB\P\P
+Partial match: aBAbaB
+    abABabA\P
+ 0: abABab
+ 1: ab
+    abaBABa\P\P
+Partial match: abaBABa
+
+/(..)\1{2,}?x/8i
+    ab\P
+Partial match: ab
+    abA\P
+Partial match: abA
+    aBAb\P
+Partial match: aBAb
+    abaBA\P
+Partial match: abaBA
+    abAbaB\P
+Partial match: abAbaB
+    abaBabA\P
+Partial match: abaBabA
+    abAbABaBx\P
+ 0: abAbABaBx
+ 1: ab
+
+/./8
+    \r\P
+ 0: \x{0d}
+    \r\P\P 
+Partial match: \x{0d}
+  
+/.{2,3}/8
+    \r\P 
+Partial match: \x{0d}
+    \r\P\P
+Partial match: \x{0d}
+    \r\r\P
+ 0: \x{0d}\x{0d}
+    \r\r\P\P
+Partial match: \x{0d}\x{0d}
+    \r\r\r\P
+ 0: \x{0d}\x{0d}\x{0d}
+    \r\r\r\P\P     
+Partial match: \x{0d}\x{0d}\x{0d}
+
+/.{2,3}?/8
+    \r\P 
+Partial match: \x{0d}
+    \r\P\P
+Partial match: \x{0d}
+    \r\r\P
+ 0: \x{0d}\x{0d}
+    \r\r\P\P
+Partial match: \x{0d}\x{0d}
+    \r\r\r\P
+ 0: \x{0d}\x{0d}
+    \r\r\r\P\P     
+ 0: \x{0d}\x{0d}
+
+/[^\x{100}][^\x{1234}][^\x{ffff}][^\x{10000}][^\x{10ffff}]/8BZ
+------------------------------------------------------------------
+        Bra
+        [^\x{100}]
+        [^\x{1234}]
+        [^\x{ffff}]
+        [^\x{10000}]
+        [^\x{10ffff}]
+        Ket
+        End
+------------------------------------------------------------------
+
+/[^\x{100}][^\x{1234}][^\x{ffff}][^\x{10000}][^\x{10ffff}]/8BZi
+------------------------------------------------------------------
+        Bra
+     /i [^\x{100}]
+     /i [^\x{1234}]
+     /i [^\x{ffff}]
+     /i [^\x{10000}]
+     /i [^\x{10ffff}]
+        Ket
+        End
+------------------------------------------------------------------
+
+/[^\x{100}]*[^\x{10000}]+[^\x{10ffff}]??[^\x{8000}]{4,}[^\x{7fff}]{2,9}?[^\x{fffff}]{5,6}+/8BZ
+------------------------------------------------------------------
+        Bra
+        [^\x{100}]*
+        [^\x{10000}]+
+        [^\x{10ffff}]??
+        [^\x{8000}]{4}
+        [^\x{8000}]*
+        [^\x{7fff}]{2}
+        [^\x{7fff}]{0,7}?
+        [^\x{fffff}]{5}
+        [^\x{fffff}]?+
+        Ket
+        End
+------------------------------------------------------------------
+
+/[^\x{100}]*[^\x{10000}]+[^\x{10ffff}]??[^\x{8000}]{4,}[^\x{7fff}]{2,9}?[^\x{fffff}]{5,6}+/8BZi
+------------------------------------------------------------------
+        Bra
+     /i [^\x{100}]*
+     /i [^\x{10000}]+
+     /i [^\x{10ffff}]??
+     /i [^\x{8000}]{4}
+     /i [^\x{8000}]*
+     /i [^\x{7fff}]{2}
+     /i [^\x{7fff}]{0,7}?
+        Once
+     /i [^\x{fffff}]{5}
+     /i [^\x{fffff}]?
+        Ket
+        Ket
+        End
+------------------------------------------------------------------
+
+/(?<=\x{1234}\x{1234})\bxy/I8
+Capturing subpattern count = 0
+Max lookbehind = 2
+Options: utf
+First char = 'x'
+Need char = 'y'
+
+/(?8BZ
+------------------------------------------------------------------
+        Bra
+        \x{100}
+        Ket
+        End
+------------------------------------------------------------------
+
+/[\u0100-\u0200]/8BZ
+------------------------------------------------------------------
+        Bra
+        [\x{100}-\x{200}]
+        Ket
+        End
+------------------------------------------------------------------
+
+/\ud800/8
+Failed: disallowed Unicode code point (>= 0xd800 && <= 0xdfff) at offset 5
+
+/-- End of testinput5 --/
diff --git a/pcre/testdata/testoutput6 b/pcre/testdata/testoutput6
new file mode 100644
index 00000000000..b1d45799261
--- /dev/null
+++ b/pcre/testdata/testoutput6
@@ -0,0 +1,2149 @@
+/-- This set of tests is for Unicode property support. It is compatible with
+    Perl >= 5.15. --/
+
+/^\pC\pL\pM\pN\pP\pS\pZ\s+/8W
+    >\x{20}\x{a0}\x{1680}\x{2028}\x{2029}\x{202f}\x{9}\x{b} 
+ 0: > \x{a0}\x{1680}\x{2028}\x{2029}\x{202f}\x{09}
+  
+/^>\pZ+/8W
+    >\x{20}\x{a0}\x{1680}\x{2028}\x{2029}\x{202f}\x{9}\x{b} 
+ 0: > \x{a0}\x{1680}\x{2028}\x{2029}\x{202f}
+  
+/^>[[:space:]]*/8W
+    >\x{20}\x{a0}\x{1680}\x{2028}\x{2029}\x{202f}\x{9}\x{b} 
+ 0: > \x{a0}\x{1680}\x{2028}\x{2029}\x{202f}\x{09}\x{0b}
+
+/^>[[:blank:]]*/8W
+    >\x{20}\x{a0}\x{1680}\x{180e}\x{2000}\x{202f}\x{9}\x{b}\x{2028} 
+ 0: > \x{a0}\x{1680}\x{180e}\x{2000}\x{202f}\x{09}
+
+/^[[:alpha:]]*/8W
+    Az\x{aa}\x{c0}\x{1c5}\x{2b0}\x{3b6}\x{1d7c9}\x{2fa1d}
+ 0: Az\x{aa}\x{c0}\x{1c5}\x{2b0}\x{3b6}\x{1d7c9}\x{2fa1d}
+
+/^[[:alnum:]]*/8W
+    Az\x{aa}\x{c0}\x{1c5}\x{2b0}\x{3b6}\x{1d7c9}\x{2fa1d}1\x{660}\x{bef}\x{16ee}
+ 0: Az\x{aa}\x{c0}\x{1c5}\x{2b0}\x{3b6}\x{1d7c9}\x{2fa1d}1\x{660}\x{bef}\x{16ee}
+
+/^[[:cntrl:]]*/8W
+    \x{0}\x{09}\x{1f}\x{7f}\x{9f} 
+ 0: \x{00}\x{09}\x{1f}\x{7f}
+
+/^[[:graph:]]*/8W
+    A\x{a1}\x{a0}
+ 0: A
+
+/^[[:print:]]*/8W
+    A z\x{a0}\x{a1}
+ 0: A z
+
+/^[[:punct:]]*/8W
+    .+\x{a1}\x{a0}
+ 0: .+
+
+/\p{Zs}*?\R/
+    ** Failers
+No match
+    a\xFCb   
+No match
+
+/\p{Zs}*\R/                                                                    
+    ** Failers 
+No match
+    a\xFCb   
+No match
+
+/ⱥ/8i
+    ⱥ
+ 0: \x{2c65}
+    Ⱥx 
+ 0: \x{23a}
+    Ⱥ 
+ 0: \x{23a}
+
+/[ⱥ]/8i
+    ⱥ
+ 0: \x{2c65}
+    Ⱥx 
+ 0: \x{23a}
+    Ⱥ 
+ 0: \x{23a}
+
+/Ⱥ/8i
+    Ⱥ
+ 0: \x{23a}
+    ⱥ
+ 0: \x{2c65}
+    
+/-- These are tests for extended grapheme clusters --/ 
+
+/^\X/8+
+    G\x{34e}\x{34e}X
+ 0: G\x{34e}\x{34e}
+ 0+ X
+    \x{34e}\x{34e}X
+ 0: \x{34e}\x{34e}
+ 0+ X
+    \x04X
+ 0: \x{04}
+ 0+ X
+    \x{1100}X
+ 0: \x{1100}
+ 0+ X
+    \x{1100}\x{34e}X
+ 0: \x{1100}\x{34e}
+ 0+ X
+    \x{1b04}\x{1b04}X 
+ 0: \x{1b04}\x{1b04}
+ 0+ X
+    *These match up to the roman letters
+ 0: *
+ 0+ These match up to the roman letters
+    \x{1111}\x{1111}L,L
+ 0: \x{1111}\x{1111}
+ 0+ L,L
+    \x{1111}\x{1111}\x{1169}L,L,V
+ 0: \x{1111}\x{1111}\x{1169}
+ 0+ L,L,V
+    \x{1111}\x{ae4c}L, LV
+ 0: \x{1111}\x{ae4c}
+ 0+ L, LV
+    \x{1111}\x{ad89}L, LVT
+ 0: \x{1111}\x{ad89}
+ 0+ L, LVT
+    \x{1111}\x{ae4c}\x{1169}L, LV, V
+ 0: \x{1111}\x{ae4c}\x{1169}
+ 0+ L, LV, V
+    \x{1111}\x{ae4c}\x{1169}\x{1169}L, LV, V, V
+ 0: \x{1111}\x{ae4c}\x{1169}\x{1169}
+ 0+ L, LV, V, V
+    \x{1111}\x{ae4c}\x{1169}\x{11fe}L, LV, V, T
+ 0: \x{1111}\x{ae4c}\x{1169}\x{11fe}
+ 0+ L, LV, V, T
+    \x{1111}\x{ad89}\x{11fe}L, LVT, T
+ 0: \x{1111}\x{ad89}\x{11fe}
+ 0+ L, LVT, T
+    \x{1111}\x{ad89}\x{11fe}\x{11fe}L, LVT, T, T
+ 0: \x{1111}\x{ad89}\x{11fe}\x{11fe}
+ 0+ L, LVT, T, T
+    \x{ad89}\x{11fe}\x{11fe}LVT, T, T
+ 0: \x{ad89}\x{11fe}\x{11fe}
+ 0+ LVT, T, T
+    *These match just the first codepoint (invalid sequence)
+ 0: *
+ 0+ These match just the first codepoint (invalid sequence)
+    \x{1111}\x{11fe}L, T
+ 0: \x{1111}
+ 0+ \x{11fe}L, T
+    \x{ae4c}\x{1111}LV, L
+ 0: \x{ae4c}
+ 0+ \x{1111}LV, L
+    \x{ae4c}\x{ae4c}LV, LV
+ 0: \x{ae4c}
+ 0+ \x{ae4c}LV, LV
+    \x{ae4c}\x{ad89}LV, LVT
+ 0: \x{ae4c}
+ 0+ \x{ad89}LV, LVT
+    \x{1169}\x{1111}V, L
+ 0: \x{1169}
+ 0+ \x{1111}V, L
+    \x{1169}\x{ae4c}V, LV
+ 0: \x{1169}
+ 0+ \x{ae4c}V, LV
+    \x{1169}\x{ad89}V, LVT
+ 0: \x{1169}
+ 0+ \x{ad89}V, LVT
+    \x{ad89}\x{1111}LVT, L
+ 0: \x{ad89}
+ 0+ \x{1111}LVT, L
+    \x{ad89}\x{1169}LVT, V
+ 0: \x{ad89}
+ 0+ \x{1169}LVT, V
+    \x{ad89}\x{ae4c}LVT, LV
+ 0: \x{ad89}
+ 0+ \x{ae4c}LVT, LV
+    \x{ad89}\x{ad89}LVT, LVT
+ 0: \x{ad89}
+ 0+ \x{ad89}LVT, LVT
+    \x{11fe}\x{1111}T, L
+ 0: \x{11fe}
+ 0+ \x{1111}T, L
+    \x{11fe}\x{1169}T, V
+ 0: \x{11fe}
+ 0+ \x{1169}T, V
+    \x{11fe}\x{ae4c}T, LV
+ 0: \x{11fe}
+ 0+ \x{ae4c}T, LV
+    \x{11fe}\x{ad89}T, LVT
+ 0: \x{11fe}
+ 0+ \x{ad89}T, LVT
+    *Test extend and spacing mark
+ 0: *
+ 0+ Test extend and spacing mark
+    \x{1111}\x{ae4c}\x{0711}L, LV, extend
+ 0: \x{1111}\x{ae4c}\x{711}
+ 0+ L, LV, extend
+    \x{1111}\x{ae4c}\x{1b04}L, LV, spacing mark
+ 0: \x{1111}\x{ae4c}\x{1b04}
+ 0+ L, LV, spacing mark
+    \x{1111}\x{ae4c}\x{1b04}\x{0711}\x{1b04}L, LV, spacing mark, extend, spacing mark
+ 0: \x{1111}\x{ae4c}\x{1b04}\x{711}\x{1b04}
+ 0+ L, LV, spacing mark, extend, spacing mark
+    *Test CR, LF, and control
+ 0: *
+ 0+ Test CR, LF, and control
+    \x0d\x{0711}CR, extend
+ 0: \x{0d}
+ 0+ \x{711}CR, extend
+    \x0d\x{1b04}CR, spacingmark
+ 0: \x{0d}
+ 0+ \x{1b04}CR, spacingmark
+    \x0a\x{0711}LF, extend
+ 0: \x{0a}
+ 0+ \x{711}LF, extend
+    \x0a\x{1b04}LF, spacingmark
+ 0: \x{0a}
+ 0+ \x{1b04}LF, spacingmark
+    \x0b\x{0711}Control, extend
+ 0: \x{0b}
+ 0+ \x{711}Control, extend
+    \x09\x{1b04}Control, spacingmark
+ 0: \x{09}
+ 0+ \x{1b04}Control, spacingmark
+    *There are no Prepend characters, so we can't test Prepend, CR
+ 0: *
+ 0+ There are no Prepend characters, so we can't test Prepend, CR
+    
+/^(?>\X{2})X/8+
+    \x{1111}\x{ae4c}\x{1111}\x{ae4c}X
+ 0: \x{1111}\x{ae4c}\x{1111}\x{ae4c}X
+ 0+ 
+    
+/^\X{2,4}X/8+
+    \x{1111}\x{ae4c}\x{1111}\x{ae4c}X
+ 0: \x{1111}\x{ae4c}\x{1111}\x{ae4c}X
+ 0+ 
+    \x{1111}\x{ae4c}\x{1111}\x{ae4c}\x{1111}\x{ae4c}X
+ 0: \x{1111}\x{ae4c}\x{1111}\x{ae4c}\x{1111}\x{ae4c}X
+ 0+ 
+    \x{1111}\x{ae4c}\x{1111}\x{ae4c}\x{1111}\x{ae4c}\x{1111}\x{ae4c}X
+ 0: \x{1111}\x{ae4c}\x{1111}\x{ae4c}\x{1111}\x{ae4c}\x{1111}\x{ae4c}X
+ 0+ 
+
+/^\X{2,4}?X/8+
+    \x{1111}\x{ae4c}\x{1111}\x{ae4c}X
+ 0: \x{1111}\x{ae4c}\x{1111}\x{ae4c}X
+ 0+ 
+    \x{1111}\x{ae4c}\x{1111}\x{ae4c}\x{1111}\x{ae4c}X
+ 0: \x{1111}\x{ae4c}\x{1111}\x{ae4c}\x{1111}\x{ae4c}X
+ 0+ 
+    \x{1111}\x{ae4c}\x{1111}\x{ae4c}\x{1111}\x{ae4c}\x{1111}\x{ae4c}X
+ 0: \x{1111}\x{ae4c}\x{1111}\x{ae4c}\x{1111}\x{ae4c}\x{1111}\x{ae4c}X
+ 0+ 
+
+/-- --/
+
+/\x{1e9e}+/8i
+    \x{1e9e}\x{00df}
+ 0: \x{1e9e}\x{df}
+
+/[z\x{1e9e}]+/8i
+    \x{1e9e}\x{00df}
+ 0: \x{1e9e}\x{df}
+
+/\x{00df}+/8i
+    \x{1e9e}\x{00df}
+ 0: \x{1e9e}\x{df}
+
+/[z\x{00df}]+/8i
+    \x{1e9e}\x{00df}
+ 0: \x{1e9e}\x{df}
+
+/\x{1f88}+/8i
+    \x{1f88}\x{1f80} 
+ 0: \x{1f88}\x{1f80}
+
+/[z\x{1f88}]+/8i
+    \x{1f88}\x{1f80} 
+ 0: \x{1f88}\x{1f80}
+    
+/-- Characters with more than one other case; test in classes --/
+
+/[z\x{00b5}]+/8i
+    \x{00b5}\x{039c}\x{03bc}
+ 0: \x{b5}\x{39c}\x{3bc}
+
+/[z\x{039c}]+/8i
+    \x{00b5}\x{039c}\x{03bc}
+ 0: \x{b5}\x{39c}\x{3bc}
+
+/[z\x{03bc}]+/8i
+    \x{00b5}\x{039c}\x{03bc}
+ 0: \x{b5}\x{39c}\x{3bc}
+
+/[z\x{00c5}]+/8i
+    \x{00c5}\x{00e5}\x{212b}
+ 0: \x{c5}\x{e5}\x{212b}
+
+/[z\x{00e5}]+/8i
+    \x{00c5}\x{00e5}\x{212b}
+ 0: \x{c5}\x{e5}\x{212b}
+
+/[z\x{212b}]+/8i
+    \x{00c5}\x{00e5}\x{212b}
+ 0: \x{c5}\x{e5}\x{212b}
+
+/[z\x{01c4}]+/8i
+    \x{01c4}\x{01c5}\x{01c6}
+ 0: \x{1c4}\x{1c5}\x{1c6}
+
+/[z\x{01c5}]+/8i
+    \x{01c4}\x{01c5}\x{01c6}
+ 0: \x{1c4}\x{1c5}\x{1c6}
+
+/[z\x{01c6}]+/8i
+    \x{01c4}\x{01c5}\x{01c6}
+ 0: \x{1c4}\x{1c5}\x{1c6}
+
+/[z\x{01c7}]+/8i
+    \x{01c7}\x{01c8}\x{01c9}
+ 0: \x{1c7}\x{1c8}\x{1c9}
+
+/[z\x{01c8}]+/8i
+    \x{01c7}\x{01c8}\x{01c9}
+ 0: \x{1c7}\x{1c8}\x{1c9}
+
+/[z\x{01c9}]+/8i
+    \x{01c7}\x{01c8}\x{01c9}
+ 0: \x{1c7}\x{1c8}\x{1c9}
+
+/[z\x{01ca}]+/8i
+    \x{01ca}\x{01cb}\x{01cc}
+ 0: \x{1ca}\x{1cb}\x{1cc}
+
+/[z\x{01cb}]+/8i
+    \x{01ca}\x{01cb}\x{01cc}
+ 0: \x{1ca}\x{1cb}\x{1cc}
+
+/[z\x{01cc}]+/8i
+    \x{01ca}\x{01cb}\x{01cc}
+ 0: \x{1ca}\x{1cb}\x{1cc}
+
+/[z\x{01f1}]+/8i
+    \x{01f1}\x{01f2}\x{01f3}
+ 0: \x{1f1}\x{1f2}\x{1f3}
+
+/[z\x{01f2}]+/8i
+    \x{01f1}\x{01f2}\x{01f3}
+ 0: \x{1f1}\x{1f2}\x{1f3}
+
+/[z\x{01f3}]+/8i
+    \x{01f1}\x{01f2}\x{01f3}
+ 0: \x{1f1}\x{1f2}\x{1f3}
+
+/[z\x{0345}]+/8i
+    \x{0345}\x{0399}\x{03b9}\x{1fbe}
+ 0: \x{345}\x{399}\x{3b9}\x{1fbe}
+
+/[z\x{0399}]+/8i
+    \x{0345}\x{0399}\x{03b9}\x{1fbe}
+ 0: \x{345}\x{399}\x{3b9}\x{1fbe}
+
+/[z\x{03b9}]+/8i
+    \x{0345}\x{0399}\x{03b9}\x{1fbe}
+ 0: \x{345}\x{399}\x{3b9}\x{1fbe}
+
+/[z\x{1fbe}]+/8i
+    \x{0345}\x{0399}\x{03b9}\x{1fbe}
+ 0: \x{345}\x{399}\x{3b9}\x{1fbe}
+
+/[z\x{0392}]+/8i
+    \x{0392}\x{03b2}\x{03d0}
+ 0: \x{392}\x{3b2}\x{3d0}
+
+/[z\x{03b2}]+/8i
+    \x{0392}\x{03b2}\x{03d0}
+ 0: \x{392}\x{3b2}\x{3d0}
+
+/[z\x{03d0}]+/8i
+    \x{0392}\x{03b2}\x{03d0}
+ 0: \x{392}\x{3b2}\x{3d0}
+
+/[z\x{0395}]+/8i
+    \x{0395}\x{03b5}\x{03f5}
+ 0: \x{395}\x{3b5}\x{3f5}
+
+/[z\x{03b5}]+/8i
+    \x{0395}\x{03b5}\x{03f5}
+ 0: \x{395}\x{3b5}\x{3f5}
+
+/[z\x{03f5}]+/8i
+    \x{0395}\x{03b5}\x{03f5}
+ 0: \x{395}\x{3b5}\x{3f5}
+
+/[z\x{0398}]+/8i
+    \x{0398}\x{03b8}\x{03d1}\x{03f4}
+ 0: \x{398}\x{3b8}\x{3d1}\x{3f4}
+
+/[z\x{03b8}]+/8i
+    \x{0398}\x{03b8}\x{03d1}\x{03f4}
+ 0: \x{398}\x{3b8}\x{3d1}\x{3f4}
+
+/[z\x{03d1}]+/8i
+    \x{0398}\x{03b8}\x{03d1}\x{03f4}
+ 0: \x{398}\x{3b8}\x{3d1}\x{3f4}
+
+/[z\x{03f4}]+/8i
+    \x{0398}\x{03b8}\x{03d1}\x{03f4}
+ 0: \x{398}\x{3b8}\x{3d1}\x{3f4}
+
+/[z\x{039a}]+/8i
+    \x{039a}\x{03ba}\x{03f0}
+ 0: \x{39a}\x{3ba}\x{3f0}
+
+/[z\x{03ba}]+/8i
+    \x{039a}\x{03ba}\x{03f0}
+ 0: \x{39a}\x{3ba}\x{3f0}
+
+/[z\x{03f0}]+/8i
+    \x{039a}\x{03ba}\x{03f0}
+ 0: \x{39a}\x{3ba}\x{3f0}
+
+/[z\x{03a0}]+/8i
+    \x{03a0}\x{03c0}\x{03d6} 
+ 0: \x{3a0}\x{3c0}\x{3d6}
+
+/[z\x{03c0}]+/8i
+    \x{03a0}\x{03c0}\x{03d6} 
+ 0: \x{3a0}\x{3c0}\x{3d6}
+
+/[z\x{03d6}]+/8i
+    \x{03a0}\x{03c0}\x{03d6} 
+ 0: \x{3a0}\x{3c0}\x{3d6}
+
+/[z\x{03a1}]+/8i
+    \x{03a1}\x{03c1}\x{03f1}
+ 0: \x{3a1}\x{3c1}\x{3f1}
+
+/[z\x{03c1}]+/8i
+    \x{03a1}\x{03c1}\x{03f1}
+ 0: \x{3a1}\x{3c1}\x{3f1}
+
+/[z\x{03f1}]+/8i
+    \x{03a1}\x{03c1}\x{03f1}
+ 0: \x{3a1}\x{3c1}\x{3f1}
+
+/[z\x{03a3}]+/8i
+    \x{03A3}\x{03C2}\x{03C3}
+ 0: \x{3a3}\x{3c2}\x{3c3}
+
+/[z\x{03c2}]+/8i
+    \x{03A3}\x{03C2}\x{03C3}
+ 0: \x{3a3}\x{3c2}\x{3c3}
+
+/[z\x{03c3}]+/8i
+    \x{03A3}\x{03C2}\x{03C3}
+ 0: \x{3a3}\x{3c2}\x{3c3}
+
+/[z\x{03a6}]+/8i
+    \x{03a6}\x{03c6}\x{03d5} 
+ 0: \x{3a6}\x{3c6}\x{3d5}
+
+/[z\x{03c6}]+/8i
+    \x{03a6}\x{03c6}\x{03d5} 
+ 0: \x{3a6}\x{3c6}\x{3d5}
+
+/[z\x{03d5}]+/8i
+    \x{03a6}\x{03c6}\x{03d5} 
+ 0: \x{3a6}\x{3c6}\x{3d5}
+
+/[z\x{03c9}]+/8i
+    \x{03c9}\x{03a9}\x{2126}
+ 0: \x{3c9}\x{3a9}\x{2126}
+
+/[z\x{03a9}]+/8i
+    \x{03c9}\x{03a9}\x{2126}
+ 0: \x{3c9}\x{3a9}\x{2126}
+
+/[z\x{2126}]+/8i
+    \x{03c9}\x{03a9}\x{2126}
+ 0: \x{3c9}\x{3a9}\x{2126}
+
+/[z\x{1e60}]+/8i
+    \x{1e60}\x{1e61}\x{1e9b}
+ 0: \x{1e60}\x{1e61}\x{1e9b}
+
+/[z\x{1e61}]+/8i
+    \x{1e60}\x{1e61}\x{1e9b}
+ 0: \x{1e60}\x{1e61}\x{1e9b}
+
+/[z\x{1e9b}]+/8i
+    \x{1e60}\x{1e61}\x{1e9b}
+ 0: \x{1e60}\x{1e61}\x{1e9b}
+
+/-- Perl 5.12.4 gets these wrong, but 5.15.3 is OK --/
+
+/[z\x{004b}]+/8i
+    \x{004b}\x{006b}\x{212a}
+ 0: Kk\x{212a}
+
+/[z\x{006b}]+/8i
+    \x{004b}\x{006b}\x{212a}
+ 0: Kk\x{212a}
+
+/[z\x{212a}]+/8i
+    \x{004b}\x{006b}\x{212a}
+ 0: Kk\x{212a}
+
+/[z\x{0053}]+/8i
+    \x{0053}\x{0073}\x{017f}
+ 0: Ss\x{17f}
+
+/[z\x{0073}]+/8i
+    \x{0053}\x{0073}\x{017f}
+ 0: Ss\x{17f}
+
+/[z\x{017f}]+/8i
+    \x{0053}\x{0073}\x{017f}
+ 0: Ss\x{17f}
+    
+/-- --/ 
+
+/(ΣΆΜΟΣ) \1/8i
+    ΣΆΜΟΣ ΣΆΜΟΣ
+ 0: \x{3a3}\x{386}\x{39c}\x{39f}\x{3a3} \x{3a3}\x{386}\x{39c}\x{39f}\x{3a3}
+ 1: \x{3a3}\x{386}\x{39c}\x{39f}\x{3a3}
+    ΣΆΜΟΣ σάμος
+ 0: \x{3a3}\x{386}\x{39c}\x{39f}\x{3a3} \x{3c3}\x{3ac}\x{3bc}\x{3bf}\x{3c2}
+ 1: \x{3a3}\x{386}\x{39c}\x{39f}\x{3a3}
+    σάμος σάμος
+ 0: \x{3c3}\x{3ac}\x{3bc}\x{3bf}\x{3c2} \x{3c3}\x{3ac}\x{3bc}\x{3bf}\x{3c2}
+ 1: \x{3c3}\x{3ac}\x{3bc}\x{3bf}\x{3c2}
+    σάμος σάμοσ
+ 0: \x{3c3}\x{3ac}\x{3bc}\x{3bf}\x{3c2} \x{3c3}\x{3ac}\x{3bc}\x{3bf}\x{3c3}
+ 1: \x{3c3}\x{3ac}\x{3bc}\x{3bf}\x{3c2}
+    σάμος ΣΆΜΟΣ  
+ 0: \x{3c3}\x{3ac}\x{3bc}\x{3bf}\x{3c2} \x{3a3}\x{386}\x{39c}\x{39f}\x{3a3}
+ 1: \x{3c3}\x{3ac}\x{3bc}\x{3bf}\x{3c2}
+
+/(σάμος) \1/8i
+    ΣΆΜΟΣ ΣΆΜΟΣ
+ 0: \x{3a3}\x{386}\x{39c}\x{39f}\x{3a3} \x{3a3}\x{386}\x{39c}\x{39f}\x{3a3}
+ 1: \x{3a3}\x{386}\x{39c}\x{39f}\x{3a3}
+    ΣΆΜΟΣ σάμος
+ 0: \x{3a3}\x{386}\x{39c}\x{39f}\x{3a3} \x{3c3}\x{3ac}\x{3bc}\x{3bf}\x{3c2}
+ 1: \x{3a3}\x{386}\x{39c}\x{39f}\x{3a3}
+    σάμος σάμος
+ 0: \x{3c3}\x{3ac}\x{3bc}\x{3bf}\x{3c2} \x{3c3}\x{3ac}\x{3bc}\x{3bf}\x{3c2}
+ 1: \x{3c3}\x{3ac}\x{3bc}\x{3bf}\x{3c2}
+    σάμος σάμοσ
+ 0: \x{3c3}\x{3ac}\x{3bc}\x{3bf}\x{3c2} \x{3c3}\x{3ac}\x{3bc}\x{3bf}\x{3c3}
+ 1: \x{3c3}\x{3ac}\x{3bc}\x{3bf}\x{3c2}
+    σάμος ΣΆΜΟΣ  
+ 0: \x{3c3}\x{3ac}\x{3bc}\x{3bf}\x{3c2} \x{3a3}\x{386}\x{39c}\x{39f}\x{3a3}
+ 1: \x{3c3}\x{3ac}\x{3bc}\x{3bf}\x{3c2}
+
+/(ΣΆΜΟΣ) \1*/8i
+    ΣΆΜΟΣ\x20
+ 0: \x{3a3}\x{386}\x{39c}\x{39f}\x{3a3} 
+ 1: \x{3a3}\x{386}\x{39c}\x{39f}\x{3a3}
+    ΣΆΜΟΣ ΣΆΜΟΣσάμοςσάμος
+ 0: \x{3a3}\x{386}\x{39c}\x{39f}\x{3a3} \x{3a3}\x{386}\x{39c}\x{39f}\x{3a3}\x{3c3}\x{3ac}\x{3bc}\x{3bf}\x{3c2}\x{3c3}\x{3ac}\x{3bc}\x{3bf}\x{3c2}
+ 1: \x{3a3}\x{386}\x{39c}\x{39f}\x{3a3}
+
+/-- Perl matches these --/
+
+/\x{00b5}+/8i
+    \x{00b5}\x{039c}\x{03bc}
+ 0: \x{b5}\x{39c}\x{3bc}
+
+/\x{039c}+/8i
+    \x{00b5}\x{039c}\x{03bc}
+ 0: \x{b5}\x{39c}\x{3bc}
+
+/\x{03bc}+/8i
+    \x{00b5}\x{039c}\x{03bc}
+ 0: \x{b5}\x{39c}\x{3bc}
+
+
+/\x{00c5}+/8i
+    \x{00c5}\x{00e5}\x{212b}
+ 0: \x{c5}\x{e5}\x{212b}
+
+/\x{00e5}+/8i
+    \x{00c5}\x{00e5}\x{212b}
+ 0: \x{c5}\x{e5}\x{212b}
+
+/\x{212b}+/8i
+    \x{00c5}\x{00e5}\x{212b}
+ 0: \x{c5}\x{e5}\x{212b}
+
+
+/\x{01c4}+/8i
+    \x{01c4}\x{01c5}\x{01c6}
+ 0: \x{1c4}\x{1c5}\x{1c6}
+
+/\x{01c5}+/8i
+    \x{01c4}\x{01c5}\x{01c6}
+ 0: \x{1c4}\x{1c5}\x{1c6}
+
+/\x{01c6}+/8i
+    \x{01c4}\x{01c5}\x{01c6}
+ 0: \x{1c4}\x{1c5}\x{1c6}
+
+
+/\x{01c7}+/8i
+    \x{01c7}\x{01c8}\x{01c9}
+ 0: \x{1c7}\x{1c8}\x{1c9}
+
+/\x{01c8}+/8i
+    \x{01c7}\x{01c8}\x{01c9}
+ 0: \x{1c7}\x{1c8}\x{1c9}
+
+/\x{01c9}+/8i
+    \x{01c7}\x{01c8}\x{01c9}
+ 0: \x{1c7}\x{1c8}\x{1c9}
+
+
+/\x{01ca}+/8i
+    \x{01ca}\x{01cb}\x{01cc}
+ 0: \x{1ca}\x{1cb}\x{1cc}
+
+/\x{01cb}+/8i
+    \x{01ca}\x{01cb}\x{01cc}
+ 0: \x{1ca}\x{1cb}\x{1cc}
+
+/\x{01cc}+/8i
+    \x{01ca}\x{01cb}\x{01cc}
+ 0: \x{1ca}\x{1cb}\x{1cc}
+
+
+/\x{01f1}+/8i
+    \x{01f1}\x{01f2}\x{01f3}
+ 0: \x{1f1}\x{1f2}\x{1f3}
+
+/\x{01f2}+/8i
+    \x{01f1}\x{01f2}\x{01f3}
+ 0: \x{1f1}\x{1f2}\x{1f3}
+
+/\x{01f3}+/8i
+    \x{01f1}\x{01f2}\x{01f3}
+ 0: \x{1f1}\x{1f2}\x{1f3}
+
+
+/\x{0345}+/8i
+    \x{0345}\x{0399}\x{03b9}\x{1fbe}
+ 0: \x{345}\x{399}\x{3b9}\x{1fbe}
+
+/\x{0399}+/8i
+    \x{0345}\x{0399}\x{03b9}\x{1fbe}
+ 0: \x{345}\x{399}\x{3b9}\x{1fbe}
+
+/\x{03b9}+/8i
+    \x{0345}\x{0399}\x{03b9}\x{1fbe}
+ 0: \x{345}\x{399}\x{3b9}\x{1fbe}
+
+/\x{1fbe}+/8i
+    \x{0345}\x{0399}\x{03b9}\x{1fbe}
+ 0: \x{345}\x{399}\x{3b9}\x{1fbe}
+
+
+/\x{0392}+/8i
+    \x{0392}\x{03b2}\x{03d0}
+ 0: \x{392}\x{3b2}\x{3d0}
+
+/\x{03b2}+/8i
+    \x{0392}\x{03b2}\x{03d0}
+ 0: \x{392}\x{3b2}\x{3d0}
+
+/\x{03d0}+/8i
+    \x{0392}\x{03b2}\x{03d0}
+ 0: \x{392}\x{3b2}\x{3d0}
+    
+
+/\x{0395}+/8i
+    \x{0395}\x{03b5}\x{03f5}
+ 0: \x{395}\x{3b5}\x{3f5}
+
+/\x{03b5}+/8i
+    \x{0395}\x{03b5}\x{03f5}
+ 0: \x{395}\x{3b5}\x{3f5}
+
+/\x{03f5}+/8i
+    \x{0395}\x{03b5}\x{03f5}
+ 0: \x{395}\x{3b5}\x{3f5}
+
+
+/\x{0398}+/8i
+    \x{0398}\x{03b8}\x{03d1}\x{03f4}
+ 0: \x{398}\x{3b8}\x{3d1}\x{3f4}
+
+/\x{03b8}+/8i
+    \x{0398}\x{03b8}\x{03d1}\x{03f4}
+ 0: \x{398}\x{3b8}\x{3d1}\x{3f4}
+
+/\x{03d1}+/8i
+    \x{0398}\x{03b8}\x{03d1}\x{03f4}
+ 0: \x{398}\x{3b8}\x{3d1}\x{3f4}
+
+/\x{03f4}+/8i
+    \x{0398}\x{03b8}\x{03d1}\x{03f4}
+ 0: \x{398}\x{3b8}\x{3d1}\x{3f4}
+    
+
+/\x{039a}+/8i
+    \x{039a}\x{03ba}\x{03f0}
+ 0: \x{39a}\x{3ba}\x{3f0}
+
+/\x{03ba}+/8i
+    \x{039a}\x{03ba}\x{03f0}
+ 0: \x{39a}\x{3ba}\x{3f0}
+
+/\x{03f0}+/8i
+    \x{039a}\x{03ba}\x{03f0}
+ 0: \x{39a}\x{3ba}\x{3f0}
+    
+
+/\x{03a0}+/8i
+    \x{03a0}\x{03c0}\x{03d6} 
+ 0: \x{3a0}\x{3c0}\x{3d6}
+
+/\x{03c0}+/8i
+    \x{03a0}\x{03c0}\x{03d6} 
+ 0: \x{3a0}\x{3c0}\x{3d6}
+
+/\x{03d6}+/8i
+    \x{03a0}\x{03c0}\x{03d6} 
+ 0: \x{3a0}\x{3c0}\x{3d6}
+
+
+/\x{03a1}+/8i
+    \x{03a1}\x{03c1}\x{03f1}
+ 0: \x{3a1}\x{3c1}\x{3f1}
+
+/\x{03c1}+/8i
+    \x{03a1}\x{03c1}\x{03f1}
+ 0: \x{3a1}\x{3c1}\x{3f1}
+
+/\x{03f1}+/8i
+    \x{03a1}\x{03c1}\x{03f1}
+ 0: \x{3a1}\x{3c1}\x{3f1}
+
+
+/\x{03a3}+/8i
+    \x{03A3}\x{03C2}\x{03C3}
+ 0: \x{3a3}\x{3c2}\x{3c3}
+
+/\x{03c2}+/8i
+    \x{03A3}\x{03C2}\x{03C3}
+ 0: \x{3a3}\x{3c2}\x{3c3}
+
+/\x{03c3}+/8i
+    \x{03A3}\x{03C2}\x{03C3}
+ 0: \x{3a3}\x{3c2}\x{3c3}
+    
+
+/\x{03a6}+/8i
+    \x{03a6}\x{03c6}\x{03d5} 
+ 0: \x{3a6}\x{3c6}\x{3d5}
+
+/\x{03c6}+/8i
+    \x{03a6}\x{03c6}\x{03d5} 
+ 0: \x{3a6}\x{3c6}\x{3d5}
+
+/\x{03d5}+/8i
+    \x{03a6}\x{03c6}\x{03d5} 
+ 0: \x{3a6}\x{3c6}\x{3d5}
+
+
+/\x{03c9}+/8i
+    \x{03c9}\x{03a9}\x{2126}
+ 0: \x{3c9}\x{3a9}\x{2126}
+
+/\x{03a9}+/8i
+    \x{03c9}\x{03a9}\x{2126}
+ 0: \x{3c9}\x{3a9}\x{2126}
+
+/\x{2126}+/8i
+    \x{03c9}\x{03a9}\x{2126}
+ 0: \x{3c9}\x{3a9}\x{2126}
+    
+
+/\x{1e60}+/8i
+    \x{1e60}\x{1e61}\x{1e9b}
+ 0: \x{1e60}\x{1e61}\x{1e9b}
+
+/\x{1e61}+/8i
+    \x{1e60}\x{1e61}\x{1e9b}
+ 0: \x{1e60}\x{1e61}\x{1e9b}
+
+/\x{1e9b}+/8i
+    \x{1e60}\x{1e61}\x{1e9b}
+ 0: \x{1e60}\x{1e61}\x{1e9b}
+    
+
+/\x{1e9e}+/8i
+    \x{1e9e}\x{00df}
+ 0: \x{1e9e}\x{df}
+
+/\x{00df}+/8i
+    \x{1e9e}\x{00df}
+ 0: \x{1e9e}\x{df}
+    
+
+/\x{1f88}+/8i
+    \x{1f88}\x{1f80} 
+ 0: \x{1f88}\x{1f80}
+
+/\x{1f80}+/8i
+    \x{1f88}\x{1f80} 
+ 0: \x{1f88}\x{1f80}
+
+
+/-- Perl 5.12.4 gets these wrong, but 5.15.3 is OK --/
+
+/\x{004b}+/8i
+    \x{004b}\x{006b}\x{212a}
+ 0: Kk\x{212a}
+
+/\x{006b}+/8i
+    \x{004b}\x{006b}\x{212a}
+ 0: Kk\x{212a}
+
+/\x{212a}+/8i
+    \x{004b}\x{006b}\x{212a}
+ 0: Kk\x{212a}
+
+
+/\x{0053}+/8i
+    \x{0053}\x{0073}\x{017f}
+ 0: Ss\x{17f}
+
+/\x{0073}+/8i
+    \x{0053}\x{0073}\x{017f}
+ 0: Ss\x{17f}
+
+/\x{017f}+/8i
+    \x{0053}\x{0073}\x{017f}
+ 0: Ss\x{17f}
+
+/^\p{Any}*\d{4}/8
+    1234
+ 0: 1234
+    123 
+No match
+ 
+/^\X*\w{4}/8
+    1234
+ 0: 1234
+    123  
+No match
+    
+/-- End of testinput6 --/
diff --git a/pcre/testdata/testoutput7 b/pcre/testdata/testoutput7
new file mode 100644
index 00000000000..ddd96fc2edc
--- /dev/null
+++ b/pcre/testdata/testoutput7
@@ -0,0 +1,1581 @@
+/-- These tests for Unicode property support test PCRE's API and show some of
+    the compiled code. They are not Perl-compatible. --/
+
+/[\p{L}]/DZ
+------------------------------------------------------------------
+        Bra
+        [\p{L}]
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+
+/[\p{^L}]/DZ
+------------------------------------------------------------------
+        Bra
+        [\P{L}]
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+
+/[\P{L}]/DZ
+------------------------------------------------------------------
+        Bra
+        [\P{L}]
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+
+/[\P{^L}]/DZ
+------------------------------------------------------------------
+        Bra
+        [\p{L}]
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+
+/[abc\p{L}\x{0660}]/8DZ
+------------------------------------------------------------------
+        Bra
+        [a-c\p{L}\x{660}]
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+No first char
+No need char
+
+/[\p{Nd}]/8DZ
+------------------------------------------------------------------
+        Bra
+        [\p{Nd}]
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+No first char
+No need char
+    1234
+ 0: 1
+
+/[\p{Nd}+-]+/8DZ
+------------------------------------------------------------------
+        Bra
+        [+\-\p{Nd}]+
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: utf
+No first char
+No need char
+    1234
+ 0: 1234
+    12-34
+ 0: 12-34
+    12+\x{661}-34  
+ 0: 12+\x{661}-34
+    ** Failers
+No match
+    abcd  
+No match
+
+/[\x{105}-\x{109}]/8iDZ
+------------------------------------------------------------------
+        Bra
+        [\x{104}-\x{109}]
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: caseless utf
+No first char
+No need char
+    \x{104}
+ 0: \x{104}
+    \x{105}
+ 0: \x{105}
+    \x{109}  
+ 0: \x{109}
+    ** Failers
+No match
+    \x{100}
+No match
+    \x{10a} 
+No match
+    
+/[z-\x{100}]/8iDZ 
+------------------------------------------------------------------
+        Bra
+        [Z\x{39c}\x{3bc}\x{1e9e}\x{178}z-\x{101}]
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: caseless utf
+No first char
+No need char
+    Z
+ 0: Z
+    z
+ 0: z
+    \x{39c}
+ 0: \x{39c}
+    \x{178}
+ 0: \x{178}
+    |
+ 0: |
+    \x{80}
+ 0: \x{80}
+    \x{ff}
+ 0: \x{ff}
+    \x{100}
+ 0: \x{100}
+    \x{101} 
+ 0: \x{101}
+    ** Failers
+No match
+    \x{102}
+No match
+    Y
+No match
+    y           
+No match
+
+/[z-\x{100}]/8DZi
+------------------------------------------------------------------
+        Bra
+        [Z\x{39c}\x{3bc}\x{1e9e}\x{178}z-\x{101}]
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: caseless utf
+No first char
+No need char
+
+/(?:[\PPa*]*){8,}/
+
+/[\P{Any}]/BZ
+------------------------------------------------------------------
+        Bra
+        [\P{Any}]
+        Ket
+        End
+------------------------------------------------------------------
+
+/[\P{Any}\E]/BZ
+------------------------------------------------------------------
+        Bra
+        [\P{Any}]
+        Ket
+        End
+------------------------------------------------------------------
+
+/(\P{Yi}+\277)/
+
+/(\P{Yi}+\277)?/
+
+/(?<=\P{Yi}{3}A)X/
+
+/\p{Yi}+(\P{Yi}+)(?1)/
+
+/(\P{Yi}{2}\277)?/
+
+/[\P{Yi}A]/
+
+/[\P{Yi}\P{Yi}\P{Yi}A]/
+
+/[^\P{Yi}A]/
+
+/[^\P{Yi}\P{Yi}\P{Yi}A]/
+
+/(\P{Yi}*\277)*/
+
+/(\P{Yi}*?\277)*/
+
+/(\p{Yi}*+\277)*/
+
+/(\P{Yi}?\277)*/
+
+/(\P{Yi}??\277)*/
+
+/(\p{Yi}?+\277)*/
+
+/(\P{Yi}{0,3}\277)*/
+
+/(\P{Yi}{0,3}?\277)*/
+
+/(\p{Yi}{0,3}+\277)*/
+
+/\p{Zl}{2,3}+/8BZ
+------------------------------------------------------------------
+        Bra
+        prop Zl {2}
+        prop Zl ?+
+        Ket
+        End
+------------------------------------------------------------------
+    


+ 0: \x{2028}\x{2028}
+    \x{2028}\x{2028}\x{2028}
+ 0: \x{2028}\x{2028}\x{2028}
+    
+/\p{Zl}/8BZ
+------------------------------------------------------------------
+        Bra
+        prop Zl
+        Ket
+        End
+------------------------------------------------------------------
+
+/\p{Lu}{3}+/8BZ
+------------------------------------------------------------------
+        Bra
+        prop Lu {3}
+        Ket
+        End
+------------------------------------------------------------------
+
+/\pL{2}+/8BZ
+------------------------------------------------------------------
+        Bra
+        prop L {2}
+        Ket
+        End
+------------------------------------------------------------------
+
+/\p{Cc}{2}+/8BZ
+------------------------------------------------------------------
+        Bra
+        prop Cc {2}
+        Ket
+        End
+------------------------------------------------------------------
+
+/^\p{Cs}/8
+    \?\x{dfff}
+ 0: \x{dfff}
+    ** Failers
+No match
+    \x{09f} 
+No match
+  
+/^\p{Sc}+/8
+    $\x{a2}\x{a3}\x{a4}\x{a5}\x{a6}
+ 0: $\x{a2}\x{a3}\x{a4}\x{a5}
+    \x{9f2}
+ 0: \x{9f2}
+    ** Failers
+No match
+    X
+No match
+    \x{2c2}
+No match
+  
+/^\p{Zs}/8
+    \ \
+ 0:  
+    \x{a0}
+ 0: \x{a0}
+    \x{1680}
+ 0: \x{1680}
+    \x{180e}
+ 0: \x{180e}
+    \x{2000}
+ 0: \x{2000}
+    \x{2001}     
+ 0: \x{2001}
+    ** Failers
+No match
+    \x{2028}
+No match
+    \x{200d} 
+No match
+  
+/-- These four are here rather than in test 6 because Perl has problems with
+    the negative versions of the properties. --/
+      
+/\p{^Lu}/8i
+    1234
+ 0: 1
+    ** Failers
+ 0: *
+    ABC 
+No match
+
+/\P{Lu}/8i
+    1234
+ 0: 1
+    ** Failers
+ 0: *
+    ABC 
+No match
+
+/\p{Ll}/8i 
+    a
+ 0: a
+    Az
+ 0: z
+    ** Failers
+ 0: a
+    ABC   
+No match
+
+/\p{Lu}/8i
+    A
+ 0: A
+    a\x{10a0}B 
+ 0: \x{10a0}
+    ** Failers 
+ 0: F
+    a
+No match
+    \x{1d00}  
+No match
+
+/[\x{c0}\x{391}]/8i
+    \x{c0}
+ 0: \x{c0}
+    \x{e0} 
+ 0: \x{e0}
+
+/-- The next two are special cases where the lengths of the different cases of
+the same character differ. The first went wrong with heap frame storage; the
+second was broken in all cases. --/
+
+/^\x{023a}+?(\x{0130}+)/8i
+  \x{023a}\x{2c65}\x{0130}
+ 0: \x{23a}\x{2c65}\x{130}
+ 1: \x{130}
+  
+/^\x{023a}+([^X])/8i
+  \x{023a}\x{2c65}X
+ 0: \x{23a}\x{2c65}
+ 1: \x{2c65}
+
+/\x{c0}+\x{116}+/8i
+    \x{c0}\x{e0}\x{116}\x{117}
+ 0: \x{c0}\x{e0}\x{116}\x{117}
+
+/[\x{c0}\x{116}]+/8i
+    \x{c0}\x{e0}\x{116}\x{117}
+ 0: \x{c0}\x{e0}\x{116}\x{117}
+
+/(\x{de})\1/8i
+    \x{de}\x{de}
+ 0: \x{de}\x{de}
+ 1: \x{de}
+    \x{de}\x{fe}
+ 0: \x{de}\x{fe}
+ 1: \x{de}
+    \x{fe}\x{fe}
+ 0: \x{fe}\x{fe}
+ 1: \x{fe}
+    \x{fe}\x{de}
+ 0: \x{fe}\x{de}
+ 1: \x{fe}
+
+/^\x{c0}$/8i
+    \x{c0}
+ 0: \x{c0}
+    \x{e0} 
+ 0: \x{e0}
+
+/^\x{e0}$/8i
+    \x{c0}
+ 0: \x{c0}
+    \x{e0} 
+ 0: \x{e0}
+
+/-- The next two should be Perl-compatible, but it fails to match \x{e0}. PCRE
+will match it only with UCP support, because without that it has no notion
+of case for anything other than the ASCII letters. --/ 
+
+/((?i)[\x{c0}])/8
+    \x{c0}
+ 0: \x{c0}
+ 1: \x{c0}
+    \x{e0} 
+ 0: \x{e0}
+ 1: \x{e0}
+
+/(?i:[\x{c0}])/8
+    \x{c0}
+ 0: \x{c0}
+    \x{e0} 
+ 0: \x{e0}
+
+/-- These are PCRE's extra properties to help with Unicodizing \d etc. --/
+
+/^\p{Xan}/8
+    ABCD
+ 0: A
+    1234
+ 0: 1
+    \x{6ca}
+ 0: \x{6ca}
+    \x{a6c}
+ 0: \x{a6c}
+    \x{10a7}   
+ 0: \x{10a7}
+    ** Failers
+No match
+    _ABC   
+No match
+
+/^\p{Xan}+/8
+    ABCD1234\x{6ca}\x{a6c}\x{10a7}_
+ 0: ABCD1234\x{6ca}\x{a6c}\x{10a7}
+    ** Failers
+No match
+    _ABC   
+No match
+
+/^\p{Xan}+?/8
+    \x{6ca}\x{a6c}\x{10a7}_
+ 0: \x{6ca}
+
+/^\p{Xan}*/8
+    ABCD1234\x{6ca}\x{a6c}\x{10a7}_
+ 0: ABCD1234\x{6ca}\x{a6c}\x{10a7}
+    
+/^\p{Xan}{2,9}/8
+    ABCD1234\x{6ca}\x{a6c}\x{10a7}_
+ 0: ABCD1234\x{6ca}
+    
+/^\p{Xan}{2,9}?/8
+    \x{6ca}\x{a6c}\x{10a7}_
+ 0: \x{6ca}\x{a6c}
+    
+/^[\p{Xan}]/8
+    ABCD1234_
+ 0: A
+    1234abcd_
+ 0: 1
+    \x{6ca}
+ 0: \x{6ca}
+    \x{a6c}
+ 0: \x{a6c}
+    \x{10a7}   
+ 0: \x{10a7}
+    ** Failers
+No match
+    _ABC   
+No match
+ 
+/^[\p{Xan}]+/8
+    ABCD1234\x{6ca}\x{a6c}\x{10a7}_
+ 0: ABCD1234\x{6ca}\x{a6c}\x{10a7}
+    ** Failers
+No match
+    _ABC   
+No match
+
+/^>\p{Xsp}/8
+    >\x{1680}\x{2028}\x{0b}
+ 0: >\x{1680}
+    >\x{a0} 
+ 0: >\x{a0}
+    ** Failers
+No match
+    \x{0b} 
+No match
+
+/^>\p{Xsp}+/8
+    > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+ 0: > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}
+
+/^>\p{Xsp}+?/8
+    >\x{1680}\x{2028}\x{0b}
+ 0: >\x{1680}
+
+/^>\p{Xsp}*/8
+    > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+ 0: > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}
+    
+/^>\p{Xsp}{2,9}/8
+    > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+ 0: > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}
+    
+/^>\p{Xsp}{2,9}?/8
+    > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+ 0: > \x{09}
+    
+/^>[\p{Xsp}]/8
+    >\x{2028}\x{0b}
+ 0: >\x{2028}
+ 
+/^>[\p{Xsp}]+/8
+    > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+ 0: > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}
+
+/^>\p{Xps}/8
+    >\x{1680}\x{2028}\x{0b}
+ 0: >\x{1680}
+    >\x{a0} 
+ 0: >\x{a0}
+    ** Failers
+No match
+    \x{0b} 
+No match
+
+/^>\p{Xps}+/8
+    > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+ 0: > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+
+/^>\p{Xps}+?/8
+    >\x{1680}\x{2028}\x{0b}
+ 0: >\x{1680}
+
+/^>\p{Xps}*/8
+    > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+ 0: > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+    
+/^>\p{Xps}{2,9}/8
+    > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+ 0: > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+    
+/^>\p{Xps}{2,9}?/8
+    > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+ 0: > \x{09}
+    
+/^>[\p{Xps}]/8
+    >\x{2028}\x{0b}
+ 0: >\x{2028}
+ 
+/^>[\p{Xps}]+/8
+    > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+ 0: > \x{09}\x{0a}\x{0c}\x{0d}\x{a0}\x{1680}\x{2028}\x{0b}
+
+/^\p{Xwd}/8
+    ABCD
+ 0: A
+    1234
+ 0: 1
+    \x{6ca}
+ 0: \x{6ca}
+    \x{a6c}
+ 0: \x{a6c}
+    \x{10a7}
+ 0: \x{10a7}
+    _ABC    
+ 0: _
+    ** Failers
+No match
+    [] 
+No match
+
+/^\p{Xwd}+/8
+    ABCD1234\x{6ca}\x{a6c}\x{10a7}_
+ 0: ABCD1234\x{6ca}\x{a6c}\x{10a7}_
+
+/^\p{Xwd}+?/8
+    \x{6ca}\x{a6c}\x{10a7}_
+ 0: \x{6ca}
+
+/^\p{Xwd}*/8
+    ABCD1234\x{6ca}\x{a6c}\x{10a7}_
+ 0: ABCD1234\x{6ca}\x{a6c}\x{10a7}_
+    
+/^\p{Xwd}{2,9}/8
+    A_B12\x{6ca}\x{a6c}\x{10a7}
+ 0: A_B12\x{6ca}\x{a6c}\x{10a7}
+    
+/^\p{Xwd}{2,9}?/8
+    \x{6ca}\x{a6c}\x{10a7}_
+ 0: \x{6ca}\x{a6c}
+    
+/^[\p{Xwd}]/8
+    ABCD1234_
+ 0: A
+    1234abcd_
+ 0: 1
+    \x{6ca}
+ 0: \x{6ca}
+    \x{a6c}
+ 0: \x{a6c}
+    \x{10a7}   
+ 0: \x{10a7}
+    _ABC 
+ 0: _
+    ** Failers
+No match
+    []   
+No match
+ 
+/^[\p{Xwd}]+/8
+    ABCD1234\x{6ca}\x{a6c}\x{10a7}_
+ 0: ABCD1234\x{6ca}\x{a6c}\x{10a7}_
+
+/-- A check not in UTF-8 mode --/
+
+/^[\p{Xwd}]+/
+    ABCD1234_
+ 0: ABCD1234_
+    
+/-- Some negative checks --/
+
+/^[\P{Xwd}]+/8
+    !.+\x{019}\x{35a}AB
+ 0: !.+\x{19}\x{35a}
+
+/^[\p{^Xwd}]+/8
+    !.+\x{019}\x{35a}AB
+ 0: !.+\x{19}\x{35a}
+
+/[\D]/WBZ8
+------------------------------------------------------------------
+        Bra
+        [\P{Nd}]
+        Ket
+        End
+------------------------------------------------------------------
+    1\x{3c8}2
+ 0: \x{3c8}
+
+/[\d]/WBZ8
+------------------------------------------------------------------
+        Bra
+        [\p{Nd}]
+        Ket
+        End
+------------------------------------------------------------------
+    >\x{6f4}<
+ 0: \x{6f4}
+
+/[\S]/WBZ8
+------------------------------------------------------------------
+        Bra
+        [\P{Xsp}]
+        Ket
+        End
+------------------------------------------------------------------
+    \x{1680}\x{6f4}\x{1680}
+ 0: \x{6f4}
+
+/[\s]/WBZ8
+------------------------------------------------------------------
+        Bra
+        [\p{Xsp}]
+        Ket
+        End
+------------------------------------------------------------------
+    >\x{1680}<
+ 0: \x{1680}
+
+/[\W]/WBZ8
+------------------------------------------------------------------
+        Bra
+        [\P{Xwd}]
+        Ket
+        End
+------------------------------------------------------------------
+    A\x{1712}B
+ 0: \x{1712}
+
+/[\w]/WBZ8
+------------------------------------------------------------------
+        Bra
+        [\p{Xwd}]
+        Ket
+        End
+------------------------------------------------------------------
+    >\x{1723}<
+ 0: \x{1723}
+
+/\D/WBZ8
+------------------------------------------------------------------
+        Bra
+        notprop Nd
+        Ket
+        End
+------------------------------------------------------------------
+    1\x{3c8}2
+ 0: \x{3c8}
+
+/\d/WBZ8
+------------------------------------------------------------------
+        Bra
+        prop Nd
+        Ket
+        End
+------------------------------------------------------------------
+    >\x{6f4}<
+ 0: \x{6f4}
+
+/\S/WBZ8
+------------------------------------------------------------------
+        Bra
+        notprop Xsp
+        Ket
+        End
+------------------------------------------------------------------
+    \x{1680}\x{6f4}\x{1680}
+ 0: \x{6f4}
+
+/\s/WBZ8
+------------------------------------------------------------------
+        Bra
+        prop Xsp
+        Ket
+        End
+------------------------------------------------------------------
+    >\x{1680}>
+ 0: \x{1680}
+
+/\W/WBZ8
+------------------------------------------------------------------
+        Bra
+        notprop Xwd
+        Ket
+        End
+------------------------------------------------------------------
+    A\x{1712}B
+ 0: \x{1712}
+
+/\w/WBZ8
+------------------------------------------------------------------
+        Bra
+        prop Xwd
+        Ket
+        End
+------------------------------------------------------------------
+    >\x{1723}<
+ 0: \x{1723}
+
+/[[:alpha:]]/WBZ
+------------------------------------------------------------------
+        Bra
+        [\p{L}]
+        Ket
+        End
+------------------------------------------------------------------
+
+/[[:lower:]]/WBZ
+------------------------------------------------------------------
+        Bra
+        [\p{Ll}]
+        Ket
+        End
+------------------------------------------------------------------
+
+/[[:upper:]]/WBZ
+------------------------------------------------------------------
+        Bra
+        [\p{Lu}]
+        Ket
+        End
+------------------------------------------------------------------
+
+/[[:alnum:]]/WBZ
+------------------------------------------------------------------
+        Bra
+        [\p{Xan}]
+        Ket
+        End
+------------------------------------------------------------------
+
+/[[:ascii:]]/WBZ
+------------------------------------------------------------------
+        Bra
+        [\x00-\x7f]
+        Ket
+        End
+------------------------------------------------------------------
+
+/[[:cntrl:]]/WBZ
+------------------------------------------------------------------
+        Bra
+        [\x00-\x1f\x7f]
+        Ket
+        End
+------------------------------------------------------------------
+
+/[[:digit:]]/WBZ
+------------------------------------------------------------------
+        Bra
+        [\p{Nd}]
+        Ket
+        End
+------------------------------------------------------------------
+
+/[[:graph:]]/WBZ
+------------------------------------------------------------------
+        Bra
+        [!-~]
+        Ket
+        End
+------------------------------------------------------------------
+
+/[[:print:]]/WBZ
+------------------------------------------------------------------
+        Bra
+        [ -~]
+        Ket
+        End
+------------------------------------------------------------------
+
+/[[:punct:]]/WBZ
+------------------------------------------------------------------
+        Bra
+        [!-/:-@[-`{-~]
+        Ket
+        End
+------------------------------------------------------------------
+
+/[[:space:]]/WBZ
+------------------------------------------------------------------
+        Bra
+        [\p{Xps}]
+        Ket
+        End
+------------------------------------------------------------------
+
+/[[:word:]]/WBZ
+------------------------------------------------------------------
+        Bra
+        [\p{Xwd}]
+        Ket
+        End
+------------------------------------------------------------------
+
+/[[:xdigit:]]/WBZ
+------------------------------------------------------------------
+        Bra
+        [0-9A-Fa-f]
+        Ket
+        End
+------------------------------------------------------------------
+
+/-- Unicode properties for \b abd \B --/
+
+/\b...\B/8W
+    abc_
+ 0: abc
+    \x{37e}abc\x{376} 
+ 0: abc
+    \x{37e}\x{376}\x{371}\x{393}\x{394} 
+ 0: \x{376}\x{371}\x{393}
+    !\x{c0}++\x{c1}\x{c2} 
+ 0: ++\x{c1}
+    !\x{c0}+++++ 
+ 0: \x{c0}++
+
+/-- Without PCRE_UCP, non-ASCII always fail, even if < 256  --/
+
+/\b...\B/8
+    abc_
+ 0: abc
+    ** Failers 
+ 0: Fai
+    \x{37e}abc\x{376} 
+No match
+    \x{37e}\x{376}\x{371}\x{393}\x{394} 
+No match
+    !\x{c0}++\x{c1}\x{c2} 
+No match
+    !\x{c0}+++++ 
+No match
+
+/-- With PCRE_UCP, non-UTF8 chars that are < 256 still check properties  --/
+
+/\b...\B/W
+    abc_
+ 0: abc
+    !\x{c0}++\x{c1}\x{c2} 
+ 0: ++\xc1
+    !\x{c0}+++++ 
+ 0: \xc0++
+
+/-- Some of these are silly, but they check various combinations --/
+
+/[[:^alpha:][:^cntrl:]]+/8WBZ
+------------------------------------------------------------------
+        Bra
+        [ -~\x80-\xff\P{L}]+
+        Ket
+        End
+------------------------------------------------------------------
+    123
+ 0: 123
+    abc 
+ 0: abc
+
+/[[:^cntrl:][:^alpha:]]+/8WBZ
+------------------------------------------------------------------
+        Bra
+        [ -~\x80-\xff\P{L}]+
+        Ket
+        End
+------------------------------------------------------------------
+    123
+ 0: 123
+    abc 
+ 0: abc
+
+/[[:alpha:]]+/8WBZ
+------------------------------------------------------------------
+        Bra
+        [\p{L}]+
+        Ket
+        End
+------------------------------------------------------------------
+    abc
+ 0: abc
+
+/[[:^alpha:]\S]+/8WBZ
+------------------------------------------------------------------
+        Bra
+        [\P{L}\P{Xsp}]+
+        Ket
+        End
+------------------------------------------------------------------
+    123
+ 0: 123
+    abc 
+ 0: abc
+
+/[^\d]+/8WBZ
+------------------------------------------------------------------
+        Bra
+        [^\p{Nd}]+
+        Ket
+        End
+------------------------------------------------------------------
+    abc123
+ 0: abc
+    abc\x{123}
+ 0: abc\x{123}
+    \x{660}abc   
+ 0: abc
+
+/\p{Lu}+9\p{Lu}+B\p{Lu}+b/BZ
+------------------------------------------------------------------
+        Bra
+        prop Lu ++
+        9
+        prop Lu +
+        B
+        prop Lu ++
+        b
+        Ket
+        End
+------------------------------------------------------------------
+
+/\p{^Lu}+9\p{^Lu}+B\p{^Lu}+b/BZ
+------------------------------------------------------------------
+        Bra
+        notprop Lu +
+        9
+        notprop Lu ++
+        B
+        notprop Lu +
+        b
+        Ket
+        End
+------------------------------------------------------------------
+
+/\P{Lu}+9\P{Lu}+B\P{Lu}+b/BZ
+------------------------------------------------------------------
+        Bra
+        notprop Lu +
+        9
+        notprop Lu ++
+        B
+        notprop Lu +
+        b
+        Ket
+        End
+------------------------------------------------------------------
+
+/\p{Han}+X\p{Greek}+\x{370}/BZ8
+------------------------------------------------------------------
+        Bra
+        prop Han ++
+        X
+        prop Greek +
+        \x{370}
+        Ket
+        End
+------------------------------------------------------------------
+
+/\p{Xan}+!\p{Xan}+A/BZ
+------------------------------------------------------------------
+        Bra
+        prop Xan ++
+        !
+        prop Xan +
+        A
+        Ket
+        End
+------------------------------------------------------------------
+
+/\p{Xsp}+!\p{Xsp}\t/BZ
+------------------------------------------------------------------
+        Bra
+        prop Xsp ++
+        !
+        prop Xsp
+        \x09
+        Ket
+        End
+------------------------------------------------------------------
+
+/\p{Xps}+!\p{Xps}\t/BZ
+------------------------------------------------------------------
+        Bra
+        prop Xps ++
+        !
+        prop Xps
+        \x09
+        Ket
+        End
+------------------------------------------------------------------
+
+/\p{Xwd}+!\p{Xwd}_/BZ
+------------------------------------------------------------------
+        Bra
+        prop Xwd ++
+        !
+        prop Xwd
+        _
+        Ket
+        End
+------------------------------------------------------------------
+
+/A+\p{N}A+\dB+\p{N}*B+\d*/WBZ
+------------------------------------------------------------------
+        Bra
+        A++
+        prop N
+        A++
+        prop Nd
+        B+
+        prop N *+
+        B+
+        prop Nd *
+        Ket
+        End
+------------------------------------------------------------------
+
+/-- These behaved oddly in Perl, so they are kept in this test --/
+
+/(\x{23a}\x{23a}\x{23a})?\1/8i
+    \x{23a}\x{23a}\x{23a}\x{2c65}\x{2c65}
+No match
+
+/(ȺȺȺ)?\1/8i
+    ȺȺȺⱥⱥ
+No match
+
+/(\x{23a}\x{23a}\x{23a})?\1/8i
+    \x{23a}\x{23a}\x{23a}\x{2c65}\x{2c65}\x{2c65}
+ 0: \x{23a}\x{23a}\x{23a}\x{2c65}\x{2c65}\x{2c65}
+ 1: \x{23a}\x{23a}\x{23a}
+
+/(ȺȺȺ)?\1/8i
+    ȺȺȺⱥⱥⱥ
+ 0: \x{23a}\x{23a}\x{23a}\x{2c65}\x{2c65}\x{2c65}
+ 1: \x{23a}\x{23a}\x{23a}
+
+/(\x{23a}\x{23a}\x{23a})\1/8i
+    \x{23a}\x{23a}\x{23a}\x{2c65}\x{2c65}
+No match
+
+/(ȺȺȺ)\1/8i
+    ȺȺȺⱥⱥ
+No match
+
+/(\x{23a}\x{23a}\x{23a})\1/8i
+    \x{23a}\x{23a}\x{23a}\x{2c65}\x{2c65}\x{2c65}
+ 0: \x{23a}\x{23a}\x{23a}\x{2c65}\x{2c65}\x{2c65}
+ 1: \x{23a}\x{23a}\x{23a}
+
+/(ȺȺȺ)\1/8i
+    ȺȺȺⱥⱥⱥ
+ 0: \x{23a}\x{23a}\x{23a}\x{2c65}\x{2c65}\x{2c65}
+ 1: \x{23a}\x{23a}\x{23a}
+
+/(\x{2c65}\x{2c65})\1/8i
+    \x{2c65}\x{2c65}\x{23a}\x{23a}
+ 0: \x{2c65}\x{2c65}\x{23a}\x{23a}
+ 1: \x{2c65}\x{2c65}
+    
+/(ⱥⱥ)\1/8i
+    ⱥⱥȺȺ 
+ 0: \x{2c65}\x{2c65}\x{23a}\x{23a}
+ 1: \x{2c65}\x{2c65}
+    
+/(\x{23a}\x{23a}\x{23a})\1Y/8i
+    X\x{23a}\x{23a}\x{23a}\x{2c65}\x{2c65}\x{2c65}YZ
+ 0: \x{23a}\x{23a}\x{23a}\x{2c65}\x{2c65}\x{2c65}Y
+ 1: \x{23a}\x{23a}\x{23a}
+
+/(\x{2c65}\x{2c65})\1Y/8i
+    X\x{2c65}\x{2c65}\x{23a}\x{23a}YZ
+ 0: \x{2c65}\x{2c65}\x{23a}\x{23a}Y
+ 1: \x{2c65}\x{2c65}
+
+/-- --/ 
+
+/-- These scripts weren't yet in Perl when I added Unicode 6.0.0 to PCRE --/
+
+/^[\p{Batak}]/8
+    \x{1bc0}
+ 0: \x{1bc0}
+    \x{1bff}
+ 0: \x{1bff}
+    ** Failers
+No match
+    \x{1bf4}
+No match
+    
+/^[\p{Brahmi}]/8
+    \x{11000}
+ 0: \x{11000}
+    \x{1106f}
+ 0: \x{1106f}
+    ** Failers
+No match
+    \x{1104e}
+No match
+    
+/^[\p{Mandaic}]/8
+    \x{840}
+ 0: \x{840}
+    \x{85e}
+ 0: \x{85e}
+    ** Failers
+No match
+    \x{85c}
+No match
+    \x{85d}    
+No match
+
+/-- --/ 
+
+/(\X*)(.)/s8
+    A\x{300}
+ 0: A
+ 1: 
+ 2: A
+
+/^S(\X*)e(\X*)$/8
+    Stéréo
+ 0: Ste\x{301}re\x{301}o
+ 1: te\x{301}r
+ 2: \x{301}o
+    
+/^\X/8 
+    ́réo
+ 0: \x{301}
+
+/^a\X41z/
+    aX41z
+ 0: aX41z
+    *** Failers
+No match
+    aAz
+No match
+
+/(?<=ab\Cde)X/8
+Failed: \C not allowed in lookbehind assertion at offset 10
+
+/\X/
+    a\P
+ 0: a
+    a\P\P
+Partial match: a
+
+/\Xa/
+    aa\P
+ 0: aa
+    aa\P\P
+ 0: aa
+
+/\X{2}/
+    aa\P
+ 0: aa
+    aa\P\P
+Partial match: aa
+
+/\X+a/
+    a\P
+Partial match: a
+    aa\P
+ 0: aa
+    aa\P\P
+Partial match: aa
+
+/\X+?a/
+    a\P
+Partial match: a
+    ab\P
+Partial match: ab
+    aa\P
+ 0: aa
+    aa\P\P
+ 0: aa
+    aba\P
+ 0: aba
+    
+/-- These Unicode 6.1.0 scripts are not known to Perl. --/ 
+
+/\p{Chakma}\d/8W
+    \x{11100}\x{1113c}
+ 0: \x{11100}\x{1113c}
+
+/\p{Takri}\d/8W
+    \x{11680}\x{116c0}
+ 0: \x{11680}\x{116c0}
+
+/^\X/8
+    A\P
+ 0: A
+    A\P\P 
+Partial match: A
+    A\x{300}\x{301}\P
+ 0: A\x{300}\x{301}
+    A\x{300}\x{301}\P\P  
+Partial match: A\x{300}\x{301}
+    A\x{301}\P
+ 0: A\x{301}
+    A\x{301}\P\P  
+Partial match: A\x{301}
+    
+/^\X{2,3}/8
+    A\P
+Partial match: A
+    A\P\P 
+Partial match: A
+    AA\P
+ 0: AA
+    AA\P\P  
+Partial match: AA
+    A\x{300}\x{301}\P
+Partial match: A\x{300}\x{301}
+    A\x{300}\x{301}\P\P  
+Partial match: A\x{300}\x{301}
+    A\x{300}\x{301}A\x{300}\x{301}\P
+ 0: A\x{300}\x{301}A\x{300}\x{301}
+    A\x{300}\x{301}A\x{300}\x{301}\P\P  
+Partial match: A\x{300}\x{301}A\x{300}\x{301}
+
+/^\X{2}/8
+    AA\P
+ 0: AA
+    AA\P\P  
+Partial match: AA
+    A\x{300}\x{301}A\x{300}\x{301}\P
+ 0: A\x{300}\x{301}A\x{300}\x{301}
+    A\x{300}\x{301}A\x{300}\x{301}\P\P  
+Partial match: A\x{300}\x{301}A\x{300}\x{301}
+    
+/^\X+/8
+    AA\P
+ 0: AA
+    AA\P\P  
+Partial match: AA
+
+/^\X+?Z/8
+    AA\P
+Partial match: AA
+    AA\P\P 
+Partial match: AA
+
+/A\x{3a3}B/8iDZ
+------------------------------------------------------------------
+        Bra
+     /i A
+        clist 03a3 03c2 03c3
+     /i B
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: caseless utf
+First char = 'A' (caseless)
+Need char = 'B' (caseless)
+
+/\x{3a3}B/8iDZ
+------------------------------------------------------------------
+        Bra
+        clist 03a3 03c2 03c3
+     /i B
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+Options: caseless utf
+No first char
+Need char = 'B' (caseless)
+
+/[\x{3a3}]/8iBZ
+------------------------------------------------------------------
+        Bra
+        clist 03a3 03c2 03c3
+        Ket
+        End
+------------------------------------------------------------------
+
+/[^\x{3a3}]/8iBZ
+------------------------------------------------------------------
+        Bra
+        not clist 03a3 03c2 03c3
+        Ket
+        End
+------------------------------------------------------------------
+
+/[\x{3a3}]+/8iBZ
+------------------------------------------------------------------
+        Bra
+        clist 03a3 03c2 03c3 +
+        Ket
+        End
+------------------------------------------------------------------
+
+/[^\x{3a3}]+/8iBZ
+------------------------------------------------------------------
+        Bra
+        not clist 03a3 03c2 03c3 +
+        Ket
+        End
+------------------------------------------------------------------
+
+/a*\x{3a3}/8iBZ
+------------------------------------------------------------------
+        Bra
+     /i a*+
+        clist 03a3 03c2 03c3
+        Ket
+        End
+------------------------------------------------------------------
+
+/\x{3a3}+a/8iBZ
+------------------------------------------------------------------
+        Bra
+        clist 03a3 03c2 03c3 ++
+     /i a
+        Ket
+        End
+------------------------------------------------------------------
+
+/\x{3a3}*\x{3c2}/8iBZ
+------------------------------------------------------------------
+        Bra
+        clist 03a3 03c2 03c3 *
+        clist 03a3 03c2 03c3
+        Ket
+        End
+------------------------------------------------------------------
+
+/\x{3a3}{3}/8i+
+    \x{3a3}\x{3c3}\x{3c2}\x{3a3}\x{3c3}\x{3c2}
+ 0: \x{3a3}\x{3c3}\x{3c2}
+ 0+ \x{3a3}\x{3c3}\x{3c2}
+
+/\x{3a3}{2,4}/8i+
+    \x{3a3}\x{3c3}\x{3c2}\x{3a3}\x{3c3}\x{3c2}
+ 0: \x{3a3}\x{3c3}\x{3c2}\x{3a3}
+ 0+ \x{3c3}\x{3c2}
+
+/\x{3a3}{2,4}?/8i+
+    \x{3a3}\x{3c3}\x{3c2}\x{3a3}\x{3c3}\x{3c2}
+ 0: \x{3a3}\x{3c3}
+ 0+ \x{3c2}\x{3a3}\x{3c3}\x{3c2}
+
+/\x{3a3}+./8i+
+    \x{3a3}\x{3c3}\x{3c2}\x{3a3}\x{3c3}\x{3c2}
+ 0: \x{3a3}\x{3c3}\x{3c2}\x{3a3}\x{3c3}\x{3c2}
+ 0+ 
+
+/\x{3a3}++./8i+
+    ** Failers
+No match
+    \x{3a3}\x{3c3}\x{3c2}\x{3a3}\x{3c3}\x{3c2}
+No match
+
+/\x{3a3}*\x{3c2}/8iBZ
+------------------------------------------------------------------
+        Bra
+        clist 03a3 03c2 03c3 *
+        clist 03a3 03c2 03c3
+        Ket
+        End
+------------------------------------------------------------------
+
+/[^\x{3a3}]*\x{3c2}/8iBZ
+------------------------------------------------------------------
+        Bra
+        not clist 03a3 03c2 03c3 *+
+        clist 03a3 03c2 03c3
+        Ket
+        End
+------------------------------------------------------------------
+
+/[^a]*\x{3c2}/8iBZ
+------------------------------------------------------------------
+        Bra
+     /i [^a]*
+        clist 03a3 03c2 03c3
+        Ket
+        End
+------------------------------------------------------------------
+
+/ist/8iBZ
+------------------------------------------------------------------
+        Bra
+     /i i
+        clist 0053 0073 017f
+     /i t
+        Ket
+        End
+------------------------------------------------------------------
+    ikt
+No match
+
+/is+t/8i
+    iSs\x{17f}t
+ 0: iSs\x{17f}t
+    ikt
+No match
+
+/is+?t/8i
+    ikt
+No match
+
+/is?t/8i
+    ikt
+No match
+
+/is{2}t/8i
+    iskt
+No match
+    
+/-- This property is a PCRE special --/
+
+/^\p{Xuc}/8
+    $abc
+ 0: $
+    @abc
+ 0: @
+    `abc
+ 0: `
+    \x{1234}abc
+ 0: \x{1234}
+    ** Failers
+No match
+    abc     
+No match
+
+/^\p{Xuc}+/8
+    $@`\x{a0}\x{1234}\x{e000}**
+ 0: $@`\x{a0}\x{1234}\x{e000}
+    ** Failers
+No match
+    \x{9f}
+No match
+
+/^\p{Xuc}+?/8
+    $@`\x{a0}\x{1234}\x{e000}**
+ 0: $
+    ** Failers
+No match
+    \x{9f}
+No match
+
+/^\p{Xuc}+?\*/8
+    $@`\x{a0}\x{1234}\x{e000}**
+ 0: $@`\x{a0}\x{1234}\x{e000}*
+    ** Failers
+No match
+    \x{9f}
+No match
+
+/^\p{Xuc}++/8
+    $@`\x{a0}\x{1234}\x{e000}**
+ 0: $@`\x{a0}\x{1234}\x{e000}
+    ** Failers
+No match
+    \x{9f}
+No match
+
+/^\p{Xuc}{3,5}/8
+    $@`\x{a0}\x{1234}\x{e000}**
+ 0: $@`\x{a0}\x{1234}
+    ** Failers
+No match
+    \x{9f}
+No match
+
+/^\p{Xuc}{3,5}?/8
+    $@`\x{a0}\x{1234}\x{e000}**
+ 0: $@`
+    ** Failers
+No match
+    \x{9f}
+No match
+
+/^[\p{Xuc}]/8
+    $@`\x{a0}\x{1234}\x{e000}**
+ 0: $
+    ** Failers
+No match
+    \x{9f}
+No match
+
+/^[\p{Xuc}]+/8
+    $@`\x{a0}\x{1234}\x{e000}**
+ 0: $@`\x{a0}\x{1234}\x{e000}
+    ** Failers
+No match
+    \x{9f}
+No match
+
+/^\P{Xuc}/8
+    abc
+ 0: a
+    ** Failers
+ 0: *
+    $abc
+No match
+    @abc
+No match
+    `abc
+No match
+    \x{1234}abc
+No match
+
+/^[\P{Xuc}]/8
+    abc
+ 0: a
+    ** Failers
+ 0: *
+    $abc
+No match
+    @abc
+No match
+    `abc
+No match
+    \x{1234}abc
+No match
+
+/-- End of testinput7 --/
diff --git a/pcre/testdata/testoutput8 b/pcre/testdata/testoutput8
new file mode 100644
index 00000000000..75affbe2d4c
--- /dev/null
+++ b/pcre/testdata/testoutput8
@@ -0,0 +1,8023 @@
+/-- This set of tests check the DFA matching functionality of pcre_dfa_exec().
+    The -dfa flag must be used with pcretest when running it. --/
+     
+/abc/
+    abc
+ 0: abc
+    
+/ab*c/
+    abc
+ 0: abc
+    abbbbc
+ 0: abbbbc
+    ac
+ 0: ac
+    
+/ab+c/
+    abc
+ 0: abc
+    abbbbbbc
+ 0: abbbbbbc
+    *** Failers 
+No match
+    ac
+No match
+    ab
+No match
+    
+/a*/
+    a
+ 0: a
+ 1: 
+    aaaaaaaaaaaaaaaaa
+ 0: aaaaaaaaaaaaaaaaa
+ 1: aaaaaaaaaaaaaaaa
+ 2: aaaaaaaaaaaaaaa
+ 3: aaaaaaaaaaaaaa
+ 4: aaaaaaaaaaaaa
+ 5: aaaaaaaaaaaa
+ 6: aaaaaaaaaaa
+ 7: aaaaaaaaaa
+ 8: aaaaaaaaa
+ 9: aaaaaaaa
+10: aaaaaaa
+11: aaaaaa
+12: aaaaa
+13: aaaa
+14: aaa
+15: aa
+16: a
+17: 
+    aaaaaaaaaaaaaaaaaaaaaaaaaaaaaa 
+Matched, but offsets vector is too small to show all matches
+ 0: aaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+ 1: aaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+ 2: aaaaaaaaaaaaaaaaaaaaaaaaaaaa
+ 3: aaaaaaaaaaaaaaaaaaaaaaaaaaa
+ 4: aaaaaaaaaaaaaaaaaaaaaaaaaa
+ 5: aaaaaaaaaaaaaaaaaaaaaaaaa
+ 6: aaaaaaaaaaaaaaaaaaaaaaaa
+ 7: aaaaaaaaaaaaaaaaaaaaaaa
+ 8: aaaaaaaaaaaaaaaaaaaaaa
+ 9: aaaaaaaaaaaaaaaaaaaaa
+10: aaaaaaaaaaaaaaaaaaaa
+11: aaaaaaaaaaaaaaaaaaa
+12: aaaaaaaaaaaaaaaaaa
+13: aaaaaaaaaaaaaaaaa
+14: aaaaaaaaaaaaaaaa
+15: aaaaaaaaaaaaaaa
+16: aaaaaaaaaaaaaa
+17: aaaaaaaaaaaaa
+18: aaaaaaaaaaaa
+19: aaaaaaaaaaa
+20: aaaaaaaaaa
+21: aaaaaaaaa
+    aaaaaaaaaaaaaaaaaaaaaaaaaaaaaa\F 
+ 0: 
+    
+/(a|abcd|african)/
+    a
+ 0: a
+    abcd
+ 0: abcd
+ 1: a
+    african
+ 0: african
+ 1: a
+    
+/^abc/
+    abcdef
+ 0: abc
+    *** Failers
+No match
+    xyzabc
+No match
+    xyz\nabc    
+No match
+    
+/^abc/m
+    abcdef
+ 0: abc
+    xyz\nabc    
+ 0: abc
+    *** Failers
+No match
+    xyzabc
+No match
+    
+/\Aabc/
+    abcdef
+ 0: abc
+    *** Failers
+No match
+    xyzabc
+No match
+    xyz\nabc    
+No match
+    
+/\Aabc/m
+    abcdef
+ 0: abc
+    *** Failers
+No match
+    xyzabc
+No match
+    xyz\nabc    
+No match
+    
+/\Gabc/
+    abcdef
+ 0: abc
+    xyzabc\>3
+ 0: abc
+    *** Failers
+No match
+    xyzabc    
+No match
+    xyzabc\>2 
+No match
+    
+/x\dy\Dz/
+    x9yzz
+ 0: x9yzz
+    x0y+z
+ 0: x0y+z
+    *** Failers
+No match
+    xyz
+No match
+    xxy0z     
+No match
+    
+/x\sy\Sz/
+    x yzz
+ 0: x yzz
+    x y+z
+ 0: x y+z
+    *** Failers
+No match
+    xyz
+No match
+    xxyyz
+No match
+    
+/x\wy\Wz/
+    xxy+z
+ 0: xxy+z
+    *** Failers
+No match
+    xxy0z
+No match
+    x+y+z         
+No match
+    
+/x.y/
+    x+y
+ 0: x+y
+    x-y
+ 0: x-y
+    *** Failers
+No match
+    x\ny
+No match
+    
+/x.y/s
+    x+y
+ 0: x+y
+    x-y
+ 0: x-y
+    x\ny
+ 0: x\x0ay
+
+/(a.b(?s)c.d|x.y)p.q/
+    a+bc+dp+q
+ 0: a+bc+dp+q
+    a+bc\ndp+q
+ 0: a+bc\x0adp+q
+    x\nyp+q 
+ 0: x\x0ayp+q
+    *** Failers 
+No match
+    a\nbc\ndp+q
+No match
+    a+bc\ndp\nq
+No match
+    x\nyp\nq 
+No match
+
+/a\d\z/
+    ba0
+ 0: a0
+    *** Failers
+No match
+    ba0\n
+No match
+    ba0\ncd   
+No match
+
+/a\d\z/m
+    ba0
+ 0: a0
+    *** Failers
+No match
+    ba0\n
+No match
+    ba0\ncd   
+No match
+
+/a\d\Z/
+    ba0
+ 0: a0
+    ba0\n
+ 0: a0
+    *** Failers
+No match
+    ba0\ncd   
+No match
+
+/a\d\Z/m
+    ba0
+ 0: a0
+    ba0\n
+ 0: a0
+    *** Failers
+No match
+    ba0\ncd   
+No match
+
+/a\d$/
+    ba0
+ 0: a0
+    ba0\n
+ 0: a0
+    *** Failers
+No match
+    ba0\ncd   
+No match
+
+/a\d$/m
+    ba0
+ 0: a0
+    ba0\n
+ 0: a0
+    ba0\ncd   
+ 0: a0
+    *** Failers
+No match
+
+/abc/i
+    abc
+ 0: abc
+    aBc
+ 0: aBc
+    ABC
+ 0: ABC
+    
+/[^a]/
+    abcd
+ 0: b
+    
+/ab?\w/
+    abz
+ 0: abz
+ 1: ab
+    abbz
+ 0: abb
+ 1: ab
+    azz  
+ 0: az
+
+/x{0,3}yz/
+    ayzq
+ 0: yz
+    axyzq
+ 0: xyz
+    axxyz
+ 0: xxyz
+    axxxyzq
+ 0: xxxyz
+    axxxxyzq
+ 0: xxxyz
+    *** Failers
+No match
+    ax
+No match
+    axx     
+No match
+      
+/x{3}yz/
+    axxxyzq
+ 0: xxxyz
+    axxxxyzq
+ 0: xxxyz
+    *** Failers
+No match
+    ax
+No match
+    axx     
+No match
+    ayzq
+No match
+    axyzq
+No match
+    axxyz
+No match
+      
+/x{2,3}yz/
+    axxyz
+ 0: xxyz
+    axxxyzq
+ 0: xxxyz
+    axxxxyzq
+ 0: xxxyz
+    *** Failers
+No match
+    ax
+No match
+    axx     
+No match
+    ayzq
+No match
+    axyzq
+No match
+      
+/[^a]+/
+    bac
+ 0: b
+    bcdefax
+ 0: bcdef
+ 1: bcde
+ 2: bcd
+ 3: bc
+ 4: b
+    *** Failers
+ 0: *** F
+ 1: *** 
+ 2: ***
+ 3: **
+ 4: *
+    aaaaa   
+No match
+
+/[^a]*/
+    bac
+ 0: b
+ 1: 
+    bcdefax
+ 0: bcdef
+ 1: bcde
+ 2: bcd
+ 3: bc
+ 4: b
+ 5: 
+    *** Failers
+ 0: *** F
+ 1: *** 
+ 2: ***
+ 3: **
+ 4: *
+ 5: 
+    aaaaa   
+ 0: 
+    
+/[^a]{3,5}/
+    xyz
+ 0: xyz
+    awxyza
+ 0: wxyz
+ 1: wxy
+    abcdefa
+ 0: bcdef
+ 1: bcde
+ 2: bcd
+    abcdefghijk
+ 0: bcdef
+ 1: bcde
+ 2: bcd
+    *** Failers
+ 0: *** F
+ 1: *** 
+ 2: ***
+    axya
+No match
+    axa
+No match
+    aaaaa         
+No match
+
+/\d*/
+    1234b567
+ 0: 1234
+ 1: 123
+ 2: 12
+ 3: 1
+ 4: 
+    xyz
+ 0: 
+    
+/\D*/
+    a1234b567
+ 0: a
+ 1: 
+    xyz
+ 0: xyz
+ 1: xy
+ 2: x
+ 3: 
+     
+/\d+/
+    ab1234c56
+ 0: 1234
+ 1: 123
+ 2: 12
+ 3: 1
+    *** Failers
+No match
+    xyz
+No match
+    
+/\D+/
+    ab123c56
+ 0: ab
+ 1: a
+    *** Failers
+ 0: *** Failers
+ 1: *** Failer
+ 2: *** Faile
+ 3: *** Fail
+ 4: *** Fai
+ 5: *** Fa
+ 6: *** F
+ 7: *** 
+ 8: ***
+ 9: **
+10: *
+    789
+No match
+    
+/\d?A/
+    045ABC
+ 0: 5A
+    ABC
+ 0: A
+    *** Failers
+No match
+    XYZ
+No match
+    
+/\D?A/
+    ABC
+ 0: A
+    BAC
+ 0: BA
+    9ABC             
+ 0: A
+    *** Failers
+No match
+
+/a+/
+    aaaa
+ 0: aaaa
+ 1: aaa
+ 2: aa
+ 3: a
+
+/^.*xyz/
+    xyz
+ 0: xyz
+    ggggggggxyz
+ 0: ggggggggxyz
+    
+/^.+xyz/
+    abcdxyz
+ 0: abcdxyz
+    axyz
+ 0: axyz
+    *** Failers
+No match
+    xyz
+No match
+    
+/^.?xyz/
+    xyz
+ 0: xyz
+    cxyz       
+ 0: cxyz
+
+/^\d{2,3}X/
+    12X
+ 0: 12X
+    123X
+ 0: 123X
+    *** Failers
+No match
+    X
+No match
+    1X
+No match
+    1234X     
+No match
+
+/^[abcd]\d/
+    a45
+ 0: a4
+    b93
+ 0: b9
+    c99z
+ 0: c9
+    d04
+ 0: d0
+    *** Failers
+No match
+    e45
+No match
+    abcd      
+No match
+    abcd1234
+No match
+    1234  
+No match
+
+/^[abcd]*\d/
+    a45
+ 0: a4
+    b93
+ 0: b9
+    c99z
+ 0: c9
+    d04
+ 0: d0
+    abcd1234
+ 0: abcd1
+    1234  
+ 0: 1
+    *** Failers
+No match
+    e45
+No match
+    abcd      
+No match
+
+/^[abcd]+\d/
+    a45
+ 0: a4
+    b93
+ 0: b9
+    c99z
+ 0: c9
+    d04
+ 0: d0
+    abcd1234
+ 0: abcd1
+    *** Failers
+No match
+    1234  
+No match
+    e45
+No match
+    abcd      
+No match
+
+/^a+X/
+    aX
+ 0: aX
+    aaX 
+ 0: aaX
+
+/^[abcd]?\d/
+    a45
+ 0: a4
+    b93
+ 0: b9
+    c99z
+ 0: c9
+    d04
+ 0: d0
+    1234  
+ 0: 1
+    *** Failers
+No match
+    abcd1234
+No match
+    e45
+No match
+
+/^[abcd]{2,3}\d/
+    ab45
+ 0: ab4
+    bcd93
+ 0: bcd9
+    *** Failers
+No match
+    1234 
+No match
+    a36 
+No match
+    abcd1234
+No match
+    ee45
+No match
+
+/^(abc)*\d/
+    abc45
+ 0: abc4
+    abcabcabc45
+ 0: abcabcabc4
+    42xyz 
+ 0: 4
+    *** Failers
+No match
+
+/^(abc)+\d/
+    abc45
+ 0: abc4
+    abcabcabc45
+ 0: abcabcabc4
+    *** Failers
+No match
+    42xyz 
+No match
+
+/^(abc)?\d/
+    abc45
+ 0: abc4
+    42xyz 
+ 0: 4
+    *** Failers
+No match
+    abcabcabc45
+No match
+
+/^(abc){2,3}\d/
+    abcabc45
+ 0: abcabc4
+    abcabcabc45
+ 0: abcabcabc4
+    *** Failers
+No match
+    abcabcabcabc45
+No match
+    abc45
+No match
+    42xyz 
+No match
+
+/1(abc|xyz)2(?1)3/
+    1abc2abc3456
+ 0: 1abc2abc3
+    1abc2xyz3456 
+ 0: 1abc2xyz3
+
+/^(a*\w|ab)=(a*\w|ab)/
+    ab=ab
+ 0: ab=ab
+ 1: ab=a
+
+/^(a*\w|ab)=(?1)/
+    ab=ab
+ 0: ab=ab
+ 1: ab=a
+
+/^([^()]|\((?1)*\))*$/
+    abc
+ 0: abc
+    a(b)c
+ 0: a(b)c
+    a(b(c))d  
+ 0: a(b(c))d
+    *** Failers)
+No match
+    a(b(c)d  
+No match
+
+/^>abc>([^()]|\((?1)*\))*abc>123abc>123abc>1(2)3abc>1(2)3abc>(1(2)3)abc>(1(2)3)a*)\d/
+    aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa9876
+ 0: aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa9
+    *** Failers 
+No match
+    aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+No match
+
+/< (?: (?(R) \d++  | [^<>]*+) | (?R)) * >/x
+    <>
+ 0: <>
+    
+ 0: 
+     hij>
+ 0:  hij>
+     hij>
+ 0: 
+    def> 
+ 0: def>
+          
+ 0: <>
+    *** Failers
+No match
+    abcxyz
+  1 ^  ^       x
+ 0: abcxyz
+    123abcxyz999 
+--->123abcxyz999
+  1    ^  ^          x
+ 0: abcxyz
+
+/(ab|cd){3,4}/C
+  ababab
+--->ababab
+ +0 ^          (ab|cd){3,4}
+ +1 ^          a
+ +4 ^          c
+ +2 ^^         b
+ +3 ^ ^        |
+ +1 ^ ^        a
+ +4 ^ ^        c
+ +2 ^  ^       b
+ +3 ^   ^      |
+ +1 ^   ^      a
+ +4 ^   ^      c
+ +2 ^    ^     b
+ +3 ^     ^    |
++12 ^     ^    
+ +1 ^     ^    a
+ +4 ^     ^    c
+ 0: ababab
+  abcdabcd
+--->abcdabcd
+ +0 ^            (ab|cd){3,4}
+ +1 ^            a
+ +4 ^            c
+ +2 ^^           b
+ +3 ^ ^          |
+ +1 ^ ^          a
+ +4 ^ ^          c
+ +5 ^  ^         d
+ +6 ^   ^        )
+ +1 ^   ^        a
+ +4 ^   ^        c
+ +2 ^    ^       b
+ +3 ^     ^      |
++12 ^     ^      
+ +1 ^     ^      a
+ +4 ^     ^      c
+ +5 ^      ^     d
+ +6 ^       ^    )
++12 ^       ^    
+ 0: abcdabcd
+ 1: abcdab
+  abcdcdcdcdcd  
+--->abcdcdcdcdcd
+ +0 ^                (ab|cd){3,4}
+ +1 ^                a
+ +4 ^                c
+ +2 ^^               b
+ +3 ^ ^              |
+ +1 ^ ^              a
+ +4 ^ ^              c
+ +5 ^  ^             d
+ +6 ^   ^            )
+ +1 ^   ^            a
+ +4 ^   ^            c
+ +5 ^    ^           d
+ +6 ^     ^          )
++12 ^     ^          
+ +1 ^     ^          a
+ +4 ^     ^          c
+ +5 ^      ^         d
+ +6 ^       ^        )
++12 ^       ^        
+ 0: abcdcdcd
+ 1: abcdcd
+
+/^abc/
+    abcdef
+ 0: abc
+    *** Failers
+No match
+    abcdef\B  
+No match
+
+/^(a*|xyz)/
+    bcd
+ 0: 
+    aaabcd
+ 0: aaa
+ 1: aa
+ 2: a
+ 3: 
+    xyz
+ 0: xyz
+ 1: 
+    xyz\N  
+ 0: xyz
+    *** Failers
+ 0: 
+    bcd\N   
+No match
+    
+/xyz$/
+    xyz
+ 0: xyz
+    xyz\n
+ 0: xyz
+    *** Failers
+No match
+    xyz\Z
+No match
+    xyz\n\Z    
+No match
+    
+/xyz$/m
+    xyz
+ 0: xyz
+    xyz\n 
+ 0: xyz
+    abcxyz\npqr 
+ 0: xyz
+    abcxyz\npqr\Z 
+ 0: xyz
+    xyz\n\Z    
+ 0: xyz
+    *** Failers
+No match
+    xyz\Z
+No match
+
+/\Gabc/
+    abcdef
+ 0: abc
+    defabcxyz\>3 
+ 0: abc
+    *** Failers 
+No match
+    defabcxyz
+No match
+
+/^abcdef/
+    ab\P
+Partial match: ab
+    abcde\P
+Partial match: abcde
+    abcdef\P
+ 0: abcdef
+    *** Failers
+No match
+    abx\P    
+No match
+
+/^a{2,4}\d+z/
+    a\P
+Partial match: a
+    aa\P
+Partial match: aa
+    aa2\P 
+Partial match: aa2
+    aaa\P
+Partial match: aaa
+    aaa23\P 
+Partial match: aaa23
+    aaaa12345\P
+Partial match: aaaa12345
+    aa0z\P
+ 0: aa0z
+    aaaa4444444444444z\P 
+ 0: aaaa4444444444444z
+    *** Failers
+No match
+    az\P 
+No match
+    aaaaa\P 
+No match
+    a56\P 
+No match
+
+/^abcdef/
+   abc\P
+Partial match: abc
+   def\R 
+ 0: def
+   
+/(?<=foo)bar/
+   xyzfo\P 
+No match
+   foob\P\>2 
+Partial match at offset 3: foob
+   foobar...\R\P\>4 
+ 0: ar
+   xyzfo\P
+No match
+   foobar\>2  
+ 0: bar
+   *** Failers
+No match
+   xyzfo\P
+No match
+   obar\R   
+No match
+
+/(ab*(cd|ef))+X/
+    adfadadaklhlkalkajhlkjahdfasdfasdfladsfjkj\P\Z
+No match
+    lkjhlkjhlkjhlkjhabbbbbbcdaefabbbbbbbefa\P\B\Z
+Partial match: abbbbbbcdaefabbbbbbbefa
+    cdabbbbbbbb\P\R\B\Z
+Partial match: cdabbbbbbbb
+    efabbbbbbbbbbbbbbbb\P\R\B\Z
+Partial match: efabbbbbbbbbbbbbbbb
+    bbbbbbbbbbbbcdXyasdfadf\P\R\B\Z    
+ 0: bbbbbbbbbbbbcdX
+
+/(a|b)/SF>testsavedregex
+Compiled pattern written to testsavedregex
+Study data written to testsavedregex
+>>aaabxyzpqrrrabbxyyyypqAzz
+ 0: aaabxyzpqrrrabbxyyyypqAzz
+    >aaaabxyzpqrrrabbxyyyypqAzz
+ 0: aaaabxyzpqrrrabbxyyyypqAzz
+    >>>>abcxyzpqrrrabbxyyyypqAzz
+ 0: abcxyzpqrrrabbxyyyypqAzz
+    *** Failers
+No match
+    abxyzpqrrabbxyyyypqAzz
+No match
+    abxyzpqrrrrabbxyyyypqAzz
+No match
+    abxyzpqrrrabxyyyypqAzz
+No match
+    aaaabcxyzzzzpqrrrabbbxyyyyyypqAzz
+No match
+    aaaabcxyzzzzpqrrrabbbxyyypqAzz
+No match
+    aaabcxyzpqrrrabbxyyyypqqqqqqqAzz
+No match
+
+/^(abc){1,2}zz/
+    abczz
+ 0: abczz
+    abcabczz
+ 0: abcabczz
+    *** Failers
+No match
+    zz
+No match
+    abcabcabczz
+No match
+    >>abczz
+No match
+
+/^(b+?|a){1,2}?c/
+    bc
+ 0: bc
+    bbc
+ 0: bbc
+    bbbc
+ 0: bbbc
+    bac
+ 0: bac
+    bbac
+ 0: bbac
+    aac
+ 0: aac
+    abbbbbbbbbbbc
+ 0: abbbbbbbbbbbc
+    bbbbbbbbbbbac
+ 0: bbbbbbbbbbbac
+    *** Failers
+No match
+    aaac
+No match
+    abbbbbbbbbbbac
+No match
+
+/^(b+|a){1,2}c/
+    bc
+ 0: bc
+    bbc
+ 0: bbc
+    bbbc
+ 0: bbbc
+    bac
+ 0: bac
+    bbac
+ 0: bbac
+    aac
+ 0: aac
+    abbbbbbbbbbbc
+ 0: abbbbbbbbbbbc
+    bbbbbbbbbbbac
+ 0: bbbbbbbbbbbac
+    *** Failers
+No match
+    aaac
+No match
+    abbbbbbbbbbbac
+No match
+
+/^(b+|a){1,2}?bc/
+    bbc
+ 0: bbc
+
+/^(b*|ba){1,2}?bc/
+    babc
+ 0: babc
+    bbabc
+ 0: bbabc
+    bababc
+ 0: bababc
+    *** Failers
+No match
+    bababbc
+No match
+    babababc
+No match
+
+/^(ba|b*){1,2}?bc/
+    babc
+ 0: babc
+    bbabc
+ 0: bbabc
+    bababc
+ 0: bababc
+    *** Failers
+No match
+    bababbc
+No match
+    babababc
+No match
+
+/^\ca\cA\c[\c{\c:/
+    \x01\x01\e;z
+ 0: \x01\x01\x1b;z
+
+/^[ab\]cde]/
+    athing
+ 0: a
+    bthing
+ 0: b
+    ]thing
+ 0: ]
+    cthing
+ 0: c
+    dthing
+ 0: d
+    ething
+ 0: e
+    *** Failers
+No match
+    fthing
+No match
+    [thing
+No match
+    \\thing
+No match
+
+/^[]cde]/
+    ]thing
+ 0: ]
+    cthing
+ 0: c
+    dthing
+ 0: d
+    ething
+ 0: e
+    *** Failers
+No match
+    athing
+No match
+    fthing
+No match
+
+/^[^ab\]cde]/
+    fthing
+ 0: f
+    [thing
+ 0: [
+    \\thing
+ 0: \
+    *** Failers
+ 0: *
+    athing
+No match
+    bthing
+No match
+    ]thing
+No match
+    cthing
+No match
+    dthing
+No match
+    ething
+No match
+
+/^[^]cde]/
+    athing
+ 0: a
+    fthing
+ 0: f
+    *** Failers
+ 0: *
+    ]thing
+No match
+    cthing
+No match
+    dthing
+No match
+    ething
+No match
+
+/^\/
+    
+ 0: \x81
+
+/^/
+    
+ 0: \xff
+
+/^[0-9]+$/
+    0
+ 0: 0
+    1
+ 0: 1
+    2
+ 0: 2
+    3
+ 0: 3
+    4
+ 0: 4
+    5
+ 0: 5
+    6
+ 0: 6
+    7
+ 0: 7
+    8
+ 0: 8
+    9
+ 0: 9
+    10
+ 0: 10
+    100
+ 0: 100
+    *** Failers
+No match
+    abc
+No match
+
+/^.*nter/
+    enter
+ 0: enter
+    inter
+ 0: inter
+    uponter
+ 0: uponter
+
+/^xxx[0-9]+$/
+    xxx0
+ 0: xxx0
+    xxx1234
+ 0: xxx1234
+    *** Failers
+No match
+    xxx
+No match
+
+/^.+[0-9][0-9][0-9]$/
+    x123
+ 0: x123
+    xx123
+ 0: xx123
+    123456
+ 0: 123456
+    *** Failers
+No match
+    123
+No match
+    x1234
+ 0: x1234
+
+/^.+?[0-9][0-9][0-9]$/
+    x123
+ 0: x123
+    xx123
+ 0: xx123
+    123456
+ 0: 123456
+    *** Failers
+No match
+    123
+No match
+    x1234
+ 0: x1234
+
+/^([^!]+)!(.+)=apquxz\.ixr\.zzz\.ac\.uk$/
+    abc!pqr=apquxz.ixr.zzz.ac.uk
+ 0: abc!pqr=apquxz.ixr.zzz.ac.uk
+    *** Failers
+No match
+    !pqr=apquxz.ixr.zzz.ac.uk
+No match
+    abc!=apquxz.ixr.zzz.ac.uk
+No match
+    abc!pqr=apquxz:ixr.zzz.ac.uk
+No match
+    abc!pqr=apquxz.ixr.zzz.ac.ukk
+No match
+
+/:/
+    Well, we need a colon: somewhere
+ 0: :
+    *** Fail if we don't
+No match
+
+/([\da-f:]+)$/i
+    0abc
+ 0: 0abc
+    abc
+ 0: abc
+    fed
+ 0: fed
+    E
+ 0: E
+    ::
+ 0: ::
+    5f03:12C0::932e
+ 0: 5f03:12C0::932e
+    fed def
+ 0: def
+    Any old stuff
+ 0: ff
+    *** Failers
+No match
+    0zzz
+No match
+    gzzz
+No match
+    fed\x20
+No match
+    Any old rubbish
+No match
+
+/^.*\.(\d{1,3})\.(\d{1,3})\.(\d{1,3})$/
+    .1.2.3
+ 0: .1.2.3
+    A.12.123.0
+ 0: A.12.123.0
+    *** Failers
+No match
+    .1.2.3333
+No match
+    1.2.3
+No match
+    1234.2.3
+No match
+
+/^(\d+)\s+IN\s+SOA\s+(\S+)\s+(\S+)\s*\(\s*$/
+    1 IN SOA non-sp1 non-sp2(
+ 0: 1 IN SOA non-sp1 non-sp2(
+    1    IN    SOA    non-sp1    non-sp2   (
+ 0: 1    IN    SOA    non-sp1    non-sp2   (
+    *** Failers
+No match
+    1IN SOA non-sp1 non-sp2(
+No match
+
+/^[a-zA-Z\d][a-zA-Z\d\-]*(\.[a-zA-Z\d][a-zA-z\d\-]*)*\.$/
+    a.
+ 0: a.
+    Z.
+ 0: Z.
+    2.
+ 0: 2.
+    ab-c.pq-r.
+ 0: ab-c.pq-r.
+    sxk.zzz.ac.uk.
+ 0: sxk.zzz.ac.uk.
+    x-.y-.
+ 0: x-.y-.
+    *** Failers
+No match
+    -abc.peq.
+No match
+
+/^\*\.[a-z]([a-z\-\d]*[a-z\d]+)?(\.[a-z]([a-z\-\d]*[a-z\d]+)?)*$/
+    *.a
+ 0: *.a
+    *.b0-a
+ 0: *.b0-a
+    *.c3-b.c
+ 0: *.c3-b.c
+    *.c-a.b-c
+ 0: *.c-a.b-c
+    *** Failers
+No match
+    *.0
+No match
+    *.a-
+No match
+    *.a-b.c-
+No match
+    *.c-a.0-c
+No match
+
+/^(?=ab(de))(abd)(e)/
+    abde
+ 0: abde
+
+/^(?!(ab)de|x)(abd)(f)/
+    abdf
+ 0: abdf
+
+/^(?=(ab(cd)))(ab)/
+    abcd
+ 0: ab
+
+/^[\da-f](\.[\da-f])*$/i
+    a.b.c.d
+ 0: a.b.c.d
+    A.B.C.D
+ 0: A.B.C.D
+    a.b.c.1.2.3.C
+ 0: a.b.c.1.2.3.C
+
+/^\".*\"\s*(;.*)?$/
+    \"1234\"
+ 0: "1234"
+    \"abcd\" ;
+ 0: "abcd" ;
+    \"\" ; rhubarb
+ 0: "" ; rhubarb
+    *** Failers
+No match
+    \"1234\" : things
+No match
+
+/^$/
+    \
+ 0: 
+    *** Failers
+No match
+
+/   ^    a   (?# begins with a)  b\sc (?# then b c) $ (?# then end)/x
+    ab c
+ 0: ab c
+    *** Failers
+No match
+    abc
+No match
+    ab cde
+No match
+
+/(?x)   ^    a   (?# begins with a)  b\sc (?# then b c) $ (?# then end)/
+    ab c
+ 0: ab c
+    *** Failers
+No match
+    abc
+No match
+    ab cde
+No match
+
+/^   a\ b[c ]d       $/x
+    a bcd
+ 0: a bcd
+    a b d
+ 0: a b d
+    *** Failers
+No match
+    abcd
+No match
+    ab d
+No match
+
+/^(a(b(c)))(d(e(f)))(h(i(j)))(k(l(m)))$/
+    abcdefhijklm
+ 0: abcdefhijklm
+
+/^(?:a(b(c)))(?:d(e(f)))(?:h(i(j)))(?:k(l(m)))$/
+    abcdefhijklm
+ 0: abcdefhijklm
+
+/^[\w][\W][\s][\S][\d][\D][\b][\n][\c]][\022]/
+    a+ Z0+\x08\n\x1d\x12
+ 0: a+ Z0+\x08\x0a\x1d\x12
+
+/^[.^$|()*+?{,}]+/
+    .^\$(*+)|{?,?}
+ 0: .^$(*+)|{?,?}
+ 1: .^$(*+)|{?,?
+ 2: .^$(*+)|{?,
+ 3: .^$(*+)|{?
+ 4: .^$(*+)|{
+ 5: .^$(*+)|
+ 6: .^$(*+)
+ 7: .^$(*+
+ 8: .^$(*
+ 9: .^$(
+10: .^$
+11: .^
+12: .
+
+/^a*\w/
+    z
+ 0: z
+    az
+ 0: az
+ 1: a
+    aaaz
+ 0: aaaz
+ 1: aaa
+ 2: aa
+ 3: a
+    a
+ 0: a
+    aa
+ 0: aa
+ 1: a
+    aaaa
+ 0: aaaa
+ 1: aaa
+ 2: aa
+ 3: a
+    a+
+ 0: a
+    aa+
+ 0: aa
+ 1: a
+
+/^a*?\w/
+    z
+ 0: z
+    az
+ 0: az
+ 1: a
+    aaaz
+ 0: aaaz
+ 1: aaa
+ 2: aa
+ 3: a
+    a
+ 0: a
+    aa
+ 0: aa
+ 1: a
+    aaaa
+ 0: aaaa
+ 1: aaa
+ 2: aa
+ 3: a
+    a+
+ 0: a
+    aa+
+ 0: aa
+ 1: a
+
+/^a+\w/
+    az
+ 0: az
+    aaaz
+ 0: aaaz
+ 1: aaa
+ 2: aa
+    aa
+ 0: aa
+    aaaa
+ 0: aaaa
+ 1: aaa
+ 2: aa
+    aa+
+ 0: aa
+
+/^a+?\w/
+    az
+ 0: az
+    aaaz
+ 0: aaaz
+ 1: aaa
+ 2: aa
+    aa
+ 0: aa
+    aaaa
+ 0: aaaa
+ 1: aaa
+ 2: aa
+    aa+
+ 0: aa
+
+/^\d{8}\w{2,}/
+    1234567890
+ 0: 1234567890
+    12345678ab
+ 0: 12345678ab
+    12345678__
+ 0: 12345678__
+    *** Failers
+No match
+    1234567
+No match
+
+/^[aeiou\d]{4,5}$/
+    uoie
+ 0: uoie
+    1234
+ 0: 1234
+    12345
+ 0: 12345
+    aaaaa
+ 0: aaaaa
+    *** Failers
+No match
+    123456
+No match
+
+/^[aeiou\d]{4,5}?/
+    uoie
+ 0: uoie
+    1234
+ 0: 1234
+    12345
+ 0: 12345
+ 1: 1234
+    aaaaa
+ 0: aaaaa
+ 1: aaaa
+    123456
+ 0: 12345
+ 1: 1234
+
+/^From +([^ ]+) +[a-zA-Z][a-zA-Z][a-zA-Z] +[a-zA-Z][a-zA-Z][a-zA-Z] +[0-9]?[0-9] +[0-9][0-9]:[0-9][0-9]/
+    From abcd  Mon Sep 01 12:33:02 1997
+ 0: From abcd  Mon Sep 01 12:33
+
+/^From\s+\S+\s+([a-zA-Z]{3}\s+){2}\d{1,2}\s+\d\d:\d\d/
+    From abcd  Mon Sep 01 12:33:02 1997
+ 0: From abcd  Mon Sep 01 12:33
+    From abcd  Mon Sep  1 12:33:02 1997
+ 0: From abcd  Mon Sep  1 12:33
+    *** Failers
+No match
+    From abcd  Sep 01 12:33:02 1997
+No match
+
+/^12.34/s
+    12\n34
+ 0: 12\x0a34
+    12\r34
+ 0: 12\x0d34
+
+/\w+(?=\t)/
+    the quick brown\t fox
+ 0: brown
+
+/foo(?!bar)(.*)/
+    foobar is foolish see?
+ 0: foolish see?
+ 1: foolish see
+ 2: foolish se
+ 3: foolish s
+ 4: foolish 
+ 5: foolish
+ 6: foolis
+ 7: fooli
+ 8: fool
+ 9: foo
+
+/(?:(?!foo)...|^.{0,2})bar(.*)/
+    foobar crowbar etc
+ 0: rowbar etc
+ 1: rowbar et
+ 2: rowbar e
+ 3: rowbar 
+ 4: rowbar
+    barrel
+ 0: barrel
+ 1: barre
+ 2: barr
+ 3: bar
+    2barrel
+ 0: 2barrel
+ 1: 2barre
+ 2: 2barr
+ 3: 2bar
+    A barrel
+ 0: A barrel
+ 1: A barre
+ 2: A barr
+ 3: A bar
+
+/^(\D*)(?=\d)(?!123)/
+    abc456
+ 0: abc
+    *** Failers
+No match
+    abc123
+No match
+
+/^1234(?# test newlines
+  inside)/
+    1234
+ 0: 1234
+
+/^1234 #comment in extended re
+  /x
+    1234
+ 0: 1234
+
+/#rhubarb
+  abcd/x
+    abcd
+ 0: abcd
+
+/^abcd#rhubarb/x
+    abcd
+ 0: abcd
+
+/(?!^)abc/
+    the abc
+ 0: abc
+    *** Failers
+No match
+    abc
+No match
+
+/(?=^)abc/
+    abc
+ 0: abc
+    *** Failers
+No match
+    the abc
+No match
+
+/^[ab]{1,3}(ab*|b)/
+    aabbbbb
+ 0: aabbbbb
+ 1: aabbbb
+ 2: aabbb
+ 3: aabb
+ 4: aab
+ 5: aa
+
+/^[ab]{1,3}?(ab*|b)/
+    aabbbbb
+ 0: aabbbbb
+ 1: aabbbb
+ 2: aabbb
+ 3: aabb
+ 4: aab
+ 5: aa
+
+/^[ab]{1,3}?(ab*?|b)/
+    aabbbbb
+ 0: aabbbbb
+ 1: aabbbb
+ 2: aabbb
+ 3: aabb
+ 4: aab
+ 5: aa
+
+/^[ab]{1,3}(ab*?|b)/
+    aabbbbb
+ 0: aabbbbb
+ 1: aabbbb
+ 2: aabbb
+ 3: aabb
+ 4: aab
+ 5: aa
+
+/  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*                          # optional leading comment
+(?:    (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+)                    # initial word
+(?:  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+)  )* # further okay, if led by a period
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  @  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*    (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                           # initial subdomain
+(?:                                  #
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.                        # if led by a period...
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                     #   ...further okay
+)*
+# address
+|                     #  or
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+)             # one word, optionally followed by....
+(?:
+[^()<>@,;:".\\\[\]\x80-\xff\000-\010\012-\037]  |  # atom and space parts, or...
+\(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)       |  # comments, or...
+
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+# quoted strings
+)*
+<  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*                     # leading <
+(?:  @  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*    (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                           # initial subdomain
+(?:                                  #
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.                        # if led by a period...
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                     #   ...further okay
+)*
+
+(?:  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  ,  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  @  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*    (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                           # initial subdomain
+(?:                                  #
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.                        # if led by a period...
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                     #   ...further okay
+)*
+)* # further okay, if led by comma
+:                                # closing colon
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  )? #       optional route
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+)                    # initial word
+(?:  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+" (?:                      # opening quote...
+[^\\\x80-\xff\n\015"]                #   Anything except backslash and quote
+|                     #    or
+\\ [^\x80-\xff]           #   Escaped something (something != CR)
+)* "  # closing quote
+)  )* # further okay, if led by a period
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  @  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*    (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                           # initial subdomain
+(?:                                  #
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  \.                        # if led by a period...
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*   (?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|   \[                         # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*    #    stuff
+\]                        #           ]
+)                     #   ...further okay
+)*
+#       address spec
+(?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*  > #                  trailing >
+# name and address
+)  (?: [\040\t] |  \(
+(?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  |  \( (?:  [^\\\x80-\xff\n\015()]  |  \\ [^\x80-\xff]  )* \)  )*
+\)  )*                       # optional trailing comment
+/x
+    Alan Other 
+ 0: Alan Other 
+    
+ 0: user@dom.ain
+ 1: user@dom
+    user\@dom.ain
+ 0: user@dom.ain
+ 1: user@dom
+    \"A. Other\"  (a comment)
+ 0: "A. Other"  (a comment)
+ 1: "A. Other"  
+ 2: "A. Other" 
+    A. Other  (a comment)
+ 0:  Other  (a comment)
+ 1:  Other  
+ 2:  Other 
+    \"/s=user/ou=host/o=place/prmd=uu.yy/admd= /c=gb/\"\@x400-re.lay
+ 0: "/s=user/ou=host/o=place/prmd=uu.yy/admd= /c=gb/"@x400-re.lay
+ 1: "/s=user/ou=host/o=place/prmd=uu.yy/admd= /c=gb/"@x400-re
+    A missing angle @,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+# Atom
+|                       #  or
+"                                     # "
+[^\\\x80-\xff\n\015"] *                            #   normal
+(?:  \\ [^\x80-\xff]  [^\\\x80-\xff\n\015"] * )*        #   ( special normal* )*
+"                                     #        "
+# Quoted string
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+\.
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+# Atom
+|                       #  or
+"                                     # "
+[^\\\x80-\xff\n\015"] *                            #   normal
+(?:  \\ [^\x80-\xff]  [^\\\x80-\xff\n\015"] * )*        #   ( special normal* )*
+"                                     #        "
+# Quoted string
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# additional words
+)*
+@
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+\[                            # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*     #    stuff
+\]                           #           ]
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# optional trailing comments
+(?:
+\.
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+\[                            # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*     #    stuff
+\]                           #           ]
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# optional trailing comments
+)*
+# address
+|                             #  or
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+# Atom
+|                       #  or
+"                                     # "
+[^\\\x80-\xff\n\015"] *                            #   normal
+(?:  \\ [^\x80-\xff]  [^\\\x80-\xff\n\015"] * )*        #   ( special normal* )*
+"                                     #        "
+# Quoted string
+)
+# leading word
+[^()<>@,;:".\\\[\]\x80-\xff\000-\010\012-\037] *               # "normal" atoms and or spaces
+(?:
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+|
+"                                     # "
+[^\\\x80-\xff\n\015"] *                            #   normal
+(?:  \\ [^\x80-\xff]  [^\\\x80-\xff\n\015"] * )*        #   ( special normal* )*
+"                                     #        "
+) # "special" comment or quoted string
+[^()<>@,;:".\\\[\]\x80-\xff\000-\010\012-\037] *            #  more "normal"
+)*
+<
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# <
+(?:
+@
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+\[                            # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*     #    stuff
+\]                           #           ]
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# optional trailing comments
+(?:
+\.
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+\[                            # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*     #    stuff
+\]                           #           ]
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# optional trailing comments
+)*
+(?: ,
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+@
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+\[                            # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*     #    stuff
+\]                           #           ]
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# optional trailing comments
+(?:
+\.
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+\[                            # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*     #    stuff
+\]                           #           ]
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# optional trailing comments
+)*
+)*  # additional domains
+:
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# optional trailing comments
+)?     #       optional route
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+# Atom
+|                       #  or
+"                                     # "
+[^\\\x80-\xff\n\015"] *                            #   normal
+(?:  \\ [^\x80-\xff]  [^\\\x80-\xff\n\015"] * )*        #   ( special normal* )*
+"                                     #        "
+# Quoted string
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+\.
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+# Atom
+|                       #  or
+"                                     # "
+[^\\\x80-\xff\n\015"] *                            #   normal
+(?:  \\ [^\x80-\xff]  [^\\\x80-\xff\n\015"] * )*        #   ( special normal* )*
+"                                     #        "
+# Quoted string
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# additional words
+)*
+@
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+\[                            # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*     #    stuff
+\]                           #           ]
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# optional trailing comments
+(?:
+\.
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+(?:
+[^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]+    # some number of atom characters...
+(?![^(\040)<>@,;:".\\\[\]\000-\037\x80-\xff]) # ..not followed by something that could be part of an atom
+|
+\[                            # [
+(?: [^\\\x80-\xff\n\015\[\]] |  \\ [^\x80-\xff]  )*     #    stuff
+\]                           #           ]
+)
+[\040\t]*                    # Nab whitespace.
+(?:
+\(                              #  (
+[^\\\x80-\xff\n\015()] *                             #     normal*
+(?:                                 #       (
+(?:  \\ [^\x80-\xff]  |
+\(                            #  (
+[^\\\x80-\xff\n\015()] *                            #     normal*
+(?:  \\ [^\x80-\xff]   [^\\\x80-\xff\n\015()] * )*        #     (special normal*)*
+\)                           #                       )
+)    #         special
+[^\\\x80-\xff\n\015()] *                         #         normal*
+)*                                  #            )*
+\)                             #                )
+[\040\t]* )*    # If comment found, allow more spaces.
+# optional trailing comments
+)*
+#       address spec
+>                    #                 >
+# name and address
+)
+/x
+    Alan Other 
+ 0: Alan Other 
+    
+ 0: user@dom.ain
+ 1: user@dom
+    user\@dom.ain
+ 0: user@dom.ain
+ 1: user@dom
+    \"A. Other\"  (a comment)
+ 0: "A. Other" 
+    A. Other  (a comment)
+ 0:  Other 
+    \"/s=user/ou=host/o=place/prmd=uu.yy/admd= /c=gb/\"\@x400-re.lay
+ 0: "/s=user/ou=host/o=place/prmd=uu.yy/admd= /c=gb/"@x400-re.lay
+ 1: "/s=user/ou=host/o=place/prmd=uu.yy/admd= /c=gb/"@x400-re
+    A missing angle 
+    a\rb
+ 0: a\x0db
+    *** Failers
+No match
+    a\nb
+No match
+
+/abc$/
+    abc
+ 0: abc
+    abc\n
+ 0: abc
+    *** Failers
+No match
+    abc\ndef
+No match
+
+/(abc)\123/
+    abc\x53
+ 0: abcS
+
+/(abc)\223/
+    abc\x93
+ 0: abc\x93
+
+/(abc)\323/
+    abc\xd3
+ 0: abc\xd3
+
+/(abc)\100/
+    abc\x40
+ 0: abc@
+    abc\100
+ 0: abc@
+
+/(abc)\1000/
+    abc\x400
+ 0: abc@0
+    abc\x40\x30
+ 0: abc@0
+    abc\1000
+ 0: abc@0
+    abc\100\x30
+ 0: abc@0
+    abc\100\060
+ 0: abc@0
+    abc\100\60
+ 0: abc@0
+
+/abc\81/
+    abc\081
+ 0: abc\x0081
+    abc\0\x38\x31
+ 0: abc\x0081
+
+/abc\91/
+    abc\091
+ 0: abc\x0091
+    abc\0\x39\x31
+ 0: abc\x0091
+
+/(a)(b)(c)(d)(e)(f)(g)(h)(i)(j)(k)\12\123/
+    abcdefghijk\12S
+ 0: abcdefghijk\x0aS
+
+/ab\idef/
+    abidef
+ 0: abidef
+
+/a{0}bc/
+    bc
+ 0: bc
+
+/(a|(bc)){0,0}?xyz/
+    xyz
+ 0: xyz
+
+/abc[\10]de/
+    abc\010de
+ 0: abc\x08de
+
+/abc[\1]de/
+    abc\1de
+ 0: abc\x01de
+
+/(abc)[\1]de/
+    abc\1de
+ 0: abc\x01de
+
+/(?s)a.b/
+    a\nb
+ 0: a\x0ab
+
+/^([^a])([^\b])([^c]*)([^d]{3,4})/
+    baNOTccccd
+ 0: baNOTcccc
+ 1: baNOTccc
+ 2: baNOTcc
+ 3: baNOTc
+ 4: baNOT
+    baNOTcccd
+ 0: baNOTccc
+ 1: baNOTcc
+ 2: baNOTc
+ 3: baNOT
+    baNOTccd
+ 0: baNOTcc
+ 1: baNOTc
+ 2: baNOT
+    bacccd
+ 0: baccc
+    *** Failers
+ 0: *** Failers
+ 1: *** Failer
+ 2: *** Faile
+ 3: *** Fail
+ 4: *** Fai
+ 5: *** Fa
+ 6: *** F
+    anything
+No match
+    b\bc   
+No match
+    baccd
+No match
+
+/[^a]/
+    Abc
+ 0: A
+  
+/[^a]/i
+    Abc 
+ 0: b
+
+/[^a]+/
+    AAAaAbc
+ 0: AAA
+ 1: AA
+ 2: A
+  
+/[^a]+/i
+    AAAaAbc 
+ 0: bc
+ 1: b
+
+/[^a]+/
+    bbb\nccc
+ 0: bbb\x0accc
+ 1: bbb\x0acc
+ 2: bbb\x0ac
+ 3: bbb\x0a
+ 4: bbb
+ 5: bb
+ 6: b
+   
+/[^k]$/
+    abc
+ 0: c
+    *** Failers
+ 0: s
+    abk   
+No match
+   
+/[^k]{2,3}$/
+    abc
+ 0: abc
+    kbc
+ 0: bc
+    kabc 
+ 0: abc
+    *** Failers
+ 0: ers
+    abk
+No match
+    akb
+No match
+    akk 
+No match
+
+/^\d{8,}\@.+[^k]$/
+    12345678\@a.b.c.d
+ 0: 12345678@a.b.c.d
+    123456789\@x.y.z
+ 0: 123456789@x.y.z
+    *** Failers
+No match
+    12345678\@x.y.uk
+No match
+    1234567\@a.b.c.d       
+No match
+
+/[^a]/
+    aaaabcd
+ 0: b
+    aaAabcd 
+ 0: A
+
+/[^a]/i
+    aaaabcd
+ 0: b
+    aaAabcd 
+ 0: b
+
+/[^az]/
+    aaaabcd
+ 0: b
+    aaAabcd 
+ 0: A
+
+/[^az]/i
+    aaaabcd
+ 0: b
+    aaAabcd 
+ 0: b
+
+/\000\001\002\003\004\005\006\007\010\011\012\013\014\015\016\017\020\021\022\023\024\025\026\027\030\031\032\033\034\035\036\037\040\041\042\043\044\045\046\047\050\051\052\053\054\055\056\057\060\061\062\063\064\065\066\067\070\071\072\073\074\075\076\077\100\101\102\103\104\105\106\107\110\111\112\113\114\115\116\117\120\121\122\123\124\125\126\127\130\131\132\133\134\135\136\137\140\141\142\143\144\145\146\147\150\151\152\153\154\155\156\157\160\161\162\163\164\165\166\167\170\171\172\173\174\175\176\177\200\201\202\203\204\205\206\207\210\211\212\213\214\215\216\217\220\221\222\223\224\225\226\227\230\231\232\233\234\235\236\237\240\241\242\243\244\245\246\247\250\251\252\253\254\255\256\257\260\261\262\263\264\265\266\267\270\271\272\273\274\275\276\277\300\301\302\303\304\305\306\307\310\311\312\313\314\315\316\317\320\321\322\323\324\325\326\327\330\331\332\333\334\335\336\337\340\341\342\343\344\345\346\347\350\351\352\353\354\355\356\357\360\361\362\363\364\365\366\367\370\371\372\373\374\375\376\377/
+ \000\001\002\003\004\005\006\007\010\011\012\013\014\015\016\017\020\021\022\023\024\025\026\027\030\031\032\033\034\035\036\037\040\041\042\043\044\045\046\047\050\051\052\053\054\055\056\057\060\061\062\063\064\065\066\067\070\071\072\073\074\075\076\077\100\101\102\103\104\105\106\107\110\111\112\113\114\115\116\117\120\121\122\123\124\125\126\127\130\131\132\133\134\135\136\137\140\141\142\143\144\145\146\147\150\151\152\153\154\155\156\157\160\161\162\163\164\165\166\167\170\171\172\173\174\175\176\177\200\201\202\203\204\205\206\207\210\211\212\213\214\215\216\217\220\221\222\223\224\225\226\227\230\231\232\233\234\235\236\237\240\241\242\243\244\245\246\247\250\251\252\253\254\255\256\257\260\261\262\263\264\265\266\267\270\271\272\273\274\275\276\277\300\301\302\303\304\305\306\307\310\311\312\313\314\315\316\317\320\321\322\323\324\325\326\327\330\331\332\333\334\335\336\337\340\341\342\343\344\345\346\347\350\351\352\353\354\355\356\357\360\361\362\363\364\365\366\367\370\371\372\373\374\375\376\377
+ 0: \x00\x01\x02\x03\x04\x05\x06\x07\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f\x10\x11\x12\x13\x14\x15\x16\x17\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~\x7f\x80\x81\x82\x83\x84\x85\x86\x87\x88\x89\x8a\x8b\x8c\x8d\x8e\x8f\x90\x91\x92\x93\x94\x95\x96\x97\x98\x99\x9a\x9b\x9c\x9d\x9e\x9f\xa0\xa1\xa2\xa3\xa4\xa5\xa6\xa7\xa8\xa9\xaa\xab\xac\xad\xae\xaf\xb0\xb1\xb2\xb3\xb4\xb5\xb6\xb7\xb8\xb9\xba\xbb\xbc\xbd\xbe\xbf\xc0\xc1\xc2\xc3\xc4\xc5\xc6\xc7\xc8\xc9\xca\xcb\xcc\xcd\xce\xcf\xd0\xd1\xd2\xd3\xd4\xd5\xd6\xd7\xd8\xd9\xda\xdb\xdc\xdd\xde\xdf\xe0\xe1\xe2\xe3\xe4\xe5\xe6\xe7\xe8\xe9\xea\xeb\xec\xed\xee\xef\xf0\xf1\xf2\xf3\xf4\xf5\xf6\xf7\xf8\xf9\xfa\xfb\xfc\xfd\xfe\xff
+
+/P[^*]TAIRE[^*]{1,6}?LL/
+    xxxxxxxxxxxPSTAIREISLLxxxxxxxxx
+ 0: PSTAIREISLL
+
+/P[^*]TAIRE[^*]{1,}?LL/
+    xxxxxxxxxxxPSTAIREISLLxxxxxxxxx
+ 0: PSTAIREISLL
+
+/(\.\d\d[1-9]?)\d+/
+    1.230003938
+ 0: .230003938
+ 1: .23000393
+ 2: .2300039
+ 3: .230003
+ 4: .23000
+ 5: .2300
+ 6: .230
+    1.875000282   
+ 0: .875000282
+ 1: .87500028
+ 2: .8750002
+ 3: .875000
+ 4: .87500
+ 5: .8750
+ 6: .875
+    1.235  
+ 0: .235
+                  
+/(\.\d\d((?=0)|\d(?=\d)))/
+    1.230003938      
+ 0: .230
+ 1: .23
+    1.875000282
+ 0: .875
+    *** Failers 
+No match
+    1.235 
+No match
+    
+/a(?)b/
+    ab 
+ 0: ab
+ 
+/\b(foo)\s+(\w+)/i
+    Food is on the foo table
+ 0: foo table
+ 1: foo tabl
+ 2: foo tab
+ 3: foo ta
+ 4: foo t
+    
+/foo(.*)bar/
+    The food is under the bar in the barn.
+ 0: food is under the bar in the bar
+ 1: food is under the bar
+    
+/foo(.*?)bar/  
+    The food is under the bar in the barn.
+ 0: food is under the bar in the bar
+ 1: food is under the bar
+
+/(.*)(\d*)/
+    I have 2 numbers: 53147
+Matched, but offsets vector is too small to show all matches
+ 0: I have 2 numbers: 53147
+ 1: I have 2 numbers: 5314
+ 2: I have 2 numbers: 531
+ 3: I have 2 numbers: 53
+ 4: I have 2 numbers: 5
+ 5: I have 2 numbers: 
+ 6: I have 2 numbers:
+ 7: I have 2 numbers
+ 8: I have 2 number
+ 9: I have 2 numbe
+10: I have 2 numb
+11: I have 2 num
+12: I have 2 nu
+13: I have 2 n
+14: I have 2 
+15: I have 2
+16: I have 
+17: I have
+18: I hav
+19: I ha
+20: I h
+21: I 
+    
+/(.*)(\d+)/
+    I have 2 numbers: 53147
+ 0: I have 2 numbers: 53147
+ 1: I have 2 numbers: 5314
+ 2: I have 2 numbers: 531
+ 3: I have 2 numbers: 53
+ 4: I have 2 numbers: 5
+ 5: I have 2
+ 
+/(.*?)(\d*)/
+    I have 2 numbers: 53147
+Matched, but offsets vector is too small to show all matches
+ 0: I have 2 numbers: 53147
+ 1: I have 2 numbers: 5314
+ 2: I have 2 numbers: 531
+ 3: I have 2 numbers: 53
+ 4: I have 2 numbers: 5
+ 5: I have 2 numbers: 
+ 6: I have 2 numbers:
+ 7: I have 2 numbers
+ 8: I have 2 number
+ 9: I have 2 numbe
+10: I have 2 numb
+11: I have 2 num
+12: I have 2 nu
+13: I have 2 n
+14: I have 2 
+15: I have 2
+16: I have 
+17: I have
+18: I hav
+19: I ha
+20: I h
+21: I 
+
+/(.*?)(\d+)/
+    I have 2 numbers: 53147
+ 0: I have 2 numbers: 53147
+ 1: I have 2 numbers: 5314
+ 2: I have 2 numbers: 531
+ 3: I have 2 numbers: 53
+ 4: I have 2 numbers: 5
+ 5: I have 2
+
+/(.*)(\d+)$/
+    I have 2 numbers: 53147
+ 0: I have 2 numbers: 53147
+
+/(.*?)(\d+)$/
+    I have 2 numbers: 53147
+ 0: I have 2 numbers: 53147
+
+/(.*)\b(\d+)$/
+    I have 2 numbers: 53147
+ 0: I have 2 numbers: 53147
+
+/(.*\D)(\d+)$/
+    I have 2 numbers: 53147
+ 0: I have 2 numbers: 53147
+
+/^\D*(?!123)/
+    ABC123
+ 0: AB
+ 1: A
+ 2: 
+     
+/^(\D*)(?=\d)(?!123)/
+    ABC445
+ 0: ABC
+    *** Failers
+No match
+    ABC123
+No match
+    
+/^[W-]46]/
+    W46]789 
+ 0: W46]
+    -46]789
+ 0: -46]
+    *** Failers
+No match
+    Wall
+No match
+    Zebra
+No match
+    42
+No match
+    [abcd] 
+No match
+    ]abcd[
+No match
+       
+/^[W-\]46]/
+    W46]789 
+ 0: W
+    Wall
+ 0: W
+    Zebra
+ 0: Z
+    Xylophone  
+ 0: X
+    42
+ 0: 4
+    [abcd] 
+ 0: [
+    ]abcd[
+ 0: ]
+    \\backslash 
+ 0: \
+    *** Failers
+No match
+    -46]789
+No match
+    well
+No match
+    
+/\d\d\/\d\d\/\d\d\d\d/
+    01/01/2000
+ 0: 01/01/2000
+
+/word (?:[a-zA-Z0-9]+ ){0,10}otherword/
+  word cat dog elephant mussel cow horse canary baboon snake shark otherword
+ 0: word cat dog elephant mussel cow horse canary baboon snake shark otherword
+  word cat dog elephant mussel cow horse canary baboon snake shark
+No match
+
+/word (?:[a-zA-Z0-9]+ ){0,300}otherword/
+  word cat dog elephant mussel cow horse canary baboon snake shark the quick brown fox and the lazy dog and several other words getting close to thirty by now I hope
+No match
+
+/^(a){0,0}/
+    bcd
+ 0: 
+    abc
+ 0: 
+    aab     
+ 0: 
+
+/^(a){0,1}/
+    bcd
+ 0: 
+    abc
+ 0: a
+ 1: 
+    aab  
+ 0: a
+ 1: 
+
+/^(a){0,2}/
+    bcd
+ 0: 
+    abc
+ 0: a
+ 1: 
+    aab  
+ 0: aa
+ 1: a
+ 2: 
+
+/^(a){0,3}/
+    bcd
+ 0: 
+    abc
+ 0: a
+ 1: 
+    aab
+ 0: aa
+ 1: a
+ 2: 
+    aaa   
+ 0: aaa
+ 1: aa
+ 2: a
+ 3: 
+
+/^(a){0,}/
+    bcd
+ 0: 
+    abc
+ 0: a
+ 1: 
+    aab
+ 0: aa
+ 1: a
+ 2: 
+    aaa
+ 0: aaa
+ 1: aa
+ 2: a
+ 3: 
+    aaaaaaaa    
+ 0: aaaaaaaa
+ 1: aaaaaaa
+ 2: aaaaaa
+ 3: aaaaa
+ 4: aaaa
+ 5: aaa
+ 6: aa
+ 7: a
+ 8: 
+
+/^(a){1,1}/
+    bcd
+No match
+    abc
+ 0: a
+    aab  
+ 0: a
+
+/^(a){1,2}/
+    bcd
+No match
+    abc
+ 0: a
+    aab  
+ 0: aa
+ 1: a
+
+/^(a){1,3}/
+    bcd
+No match
+    abc
+ 0: a
+    aab
+ 0: aa
+ 1: a
+    aaa   
+ 0: aaa
+ 1: aa
+ 2: a
+
+/^(a){1,}/
+    bcd
+No match
+    abc
+ 0: a
+    aab
+ 0: aa
+ 1: a
+    aaa
+ 0: aaa
+ 1: aa
+ 2: a
+    aaaaaaaa    
+ 0: aaaaaaaa
+ 1: aaaaaaa
+ 2: aaaaaa
+ 3: aaaaa
+ 4: aaaa
+ 5: aaa
+ 6: aa
+ 7: a
+
+/.*\.gif/
+    borfle\nbib.gif\nno
+ 0: bib.gif
+
+/.{0,}\.gif/
+    borfle\nbib.gif\nno
+ 0: bib.gif
+
+/.*\.gif/m
+    borfle\nbib.gif\nno
+ 0: bib.gif
+
+/.*\.gif/s
+    borfle\nbib.gif\nno
+ 0: borfle\x0abib.gif
+
+/.*\.gif/ms
+    borfle\nbib.gif\nno
+ 0: borfle\x0abib.gif
+    
+/.*$/
+    borfle\nbib.gif\nno
+ 0: no
+
+/.*$/m
+    borfle\nbib.gif\nno
+ 0: borfle
+
+/.*$/s
+    borfle\nbib.gif\nno
+ 0: borfle\x0abib.gif\x0ano
+
+/.*$/ms
+    borfle\nbib.gif\nno
+ 0: borfle\x0abib.gif\x0ano
+ 1: borfle\x0abib.gif
+ 2: borfle
+    
+/.*$/
+    borfle\nbib.gif\nno\n
+ 0: no
+
+/.*$/m
+    borfle\nbib.gif\nno\n
+ 0: borfle
+
+/.*$/s
+    borfle\nbib.gif\nno\n
+ 0: borfle\x0abib.gif\x0ano\x0a
+ 1: borfle\x0abib.gif\x0ano
+
+/.*$/ms
+    borfle\nbib.gif\nno\n
+ 0: borfle\x0abib.gif\x0ano\x0a
+ 1: borfle\x0abib.gif\x0ano
+ 2: borfle\x0abib.gif
+ 3: borfle
+    
+/(.*X|^B)/
+    abcde\n1234Xyz
+ 0: 1234X
+    BarFoo 
+ 0: B
+    *** Failers
+No match
+    abcde\nBar  
+No match
+
+/(.*X|^B)/m
+    abcde\n1234Xyz
+ 0: 1234X
+    BarFoo 
+ 0: B
+    abcde\nBar  
+ 0: B
+
+/(.*X|^B)/s
+    abcde\n1234Xyz
+ 0: abcde\x0a1234X
+    BarFoo 
+ 0: B
+    *** Failers
+No match
+    abcde\nBar  
+No match
+
+/(.*X|^B)/ms
+    abcde\n1234Xyz
+ 0: abcde\x0a1234X
+    BarFoo 
+ 0: B
+    abcde\nBar  
+ 0: B
+
+/(?s)(.*X|^B)/
+    abcde\n1234Xyz
+ 0: abcde\x0a1234X
+    BarFoo 
+ 0: B
+    *** Failers 
+No match
+    abcde\nBar  
+No match
+
+/(?s:.*X|^B)/
+    abcde\n1234Xyz
+ 0: abcde\x0a1234X
+    BarFoo 
+ 0: B
+    *** Failers 
+No match
+    abcde\nBar  
+No match
+
+/^.*B/
+    **** Failers
+No match
+    abc\nB
+No match
+     
+/(?s)^.*B/
+    abc\nB
+ 0: abc\x0aB
+
+/(?m)^.*B/
+    abc\nB
+ 0: B
+     
+/(?ms)^.*B/
+    abc\nB
+ 0: abc\x0aB
+
+/(?ms)^B/
+    abc\nB
+ 0: B
+
+/(?s)B$/
+    B\n
+ 0: B
+
+/^[0-9][0-9][0-9][0-9][0-9][0-9][0-9][0-9][0-9][0-9][0-9][0-9]/
+    123456654321
+ 0: 123456654321
+  
+/^\d\d\d\d\d\d\d\d\d\d\d\d/
+    123456654321 
+ 0: 123456654321
+
+/^[\d][\d][\d][\d][\d][\d][\d][\d][\d][\d][\d][\d]/
+    123456654321
+ 0: 123456654321
+  
+/^[abc]{12}/
+    abcabcabcabc
+ 0: abcabcabcabc
+    
+/^[a-c]{12}/
+    abcabcabcabc
+ 0: abcabcabcabc
+    
+/^(a|b|c){12}/
+    abcabcabcabc 
+ 0: abcabcabcabc
+
+/^[abcdefghijklmnopqrstuvwxy0123456789]/
+    n
+ 0: n
+    *** Failers 
+No match
+    z 
+No match
+
+/abcde{0,0}/
+    abcd
+ 0: abcd
+    *** Failers
+No match
+    abce  
+No match
+
+/ab[cd]{0,0}e/
+    abe
+ 0: abe
+    *** Failers
+No match
+    abcde 
+No match
+    
+/ab(c){0,0}d/
+    abd
+ 0: abd
+    *** Failers
+No match
+    abcd   
+No match
+
+/a(b*)/
+    a
+ 0: a
+    ab
+ 0: ab
+ 1: a
+    abbbb
+ 0: abbbb
+ 1: abbb
+ 2: abb
+ 3: ab
+ 4: a
+    *** Failers
+ 0: a
+    bbbbb    
+No match
+    
+/ab\d{0}e/
+    abe
+ 0: abe
+    *** Failers
+No match
+    ab1e   
+No match
+    
+/"([^\\"]+|\\.)*"/
+    the \"quick\" brown fox
+ 0: "quick"
+    \"the \\\"quick\\\" brown fox\" 
+ 0: "the \"quick\" brown fox"
+
+/.*?/g+
+    abc
+ 0: abc
+ 0+ 
+ 1: ab
+ 2: a
+ 3: 
+ 0: 
+ 0+ 
+  
+/\b/g+
+    abc 
+ 0: 
+ 0+ abc
+ 0: 
+ 0+ 
+
+/\b/+g
+    abc 
+ 0: 
+ 0+ abc
+ 0: 
+ 0+ 
+
+//g
+    abc
+ 0: 
+ 0: 
+ 0: 
+ 0: 
+
+/]{0,})>]{0,})>([\d]{0,}\.)(.*)((
([\w\W\s\d][^<>]{0,})|[\s]{0,}))<\/a><\/TD>]{0,})>([\w\W\s\d][^<>]{0,})<\/TD>]{0,})>([\w\W\s\d][^<>]{0,})<\/TD><\/TR>/is
+  43.Word Processor
(N-1286)Lega lstaff.comCA - Statewide
+ 0: 43.Word Processor
(N-1286)Lega lstaff.comCA - Statewide
+
+/a[^a]b/
+    acb
+ 0: acb
+    a\nb
+ 0: a\x0ab
+    
+/a.b/
+    acb
+ 0: acb
+    *** Failers 
+No match
+    a\nb   
+No match
+    
+/a[^a]b/s
+    acb
+ 0: acb
+    a\nb  
+ 0: a\x0ab
+    
+/a.b/s
+    acb
+ 0: acb
+    a\nb  
+ 0: a\x0ab
+
+/^(b+?|a){1,2}?c/
+    bac
+ 0: bac
+    bbac
+ 0: bbac
+    bbbac
+ 0: bbbac
+    bbbbac
+ 0: bbbbac
+    bbbbbac 
+ 0: bbbbbac
+
+/^(b+|a){1,2}?c/
+    bac
+ 0: bac
+    bbac
+ 0: bbac
+    bbbac
+ 0: bbbac
+    bbbbac
+ 0: bbbbac
+    bbbbbac 
+ 0: bbbbbac
+    
+/(?!\A)x/m
+    x\nb\n
+No match
+    a\bx\n  
+ 0: x
+    
+/\x0{ab}/
+    \0{ab} 
+ 0: \x00{ab}
+
+/(A|B)*?CD/
+    CD 
+ 0: CD
+    
+/(A|B)*CD/
+    CD 
+ 0: CD
+
+/(?.*/)foo"
+    /this/is/a/very/long/line/in/deed/with/very/many/slashes/in/it/you/see/
+No match
+
+"(?>.*/)foo"
+    /this/is/a/very/long/line/in/deed/with/very/many/slashes/in/and/foo
+ 0: /this/is/a/very/long/line/in/deed/with/very/many/slashes/in/and/foo
+
+/(?>(\.\d\d[1-9]?))\d+/
+    1.230003938
+ 0: .230003938
+ 1: .23000393
+ 2: .2300039
+ 3: .230003
+ 4: .23000
+ 5: .2300
+ 6: .230
+    1.875000282
+ 0: .875000282
+ 1: .87500028
+ 2: .8750002
+ 3: .875000
+ 4: .87500
+ 5: .8750
+    *** Failers 
+No match
+    1.235 
+No match
+
+/^((?>\w+)|(?>\s+))*$/
+    now is the time for all good men to come to the aid of the party
+ 0: now is the time for all good men to come to the aid of the party
+    *** Failers
+No match
+    this is not a line with only words and spaces!
+No match
+    
+/(\d+)(\w)/
+    12345a
+ 0: 12345a
+ 1: 12345
+ 2: 1234
+ 3: 123
+ 4: 12
+    12345+ 
+ 0: 12345
+ 1: 1234
+ 2: 123
+ 3: 12
+
+/((?>\d+))(\w)/
+    12345a
+ 0: 12345a
+    *** Failers
+No match
+    12345+ 
+No match
+
+/(?>a+)b/
+    aaab
+ 0: aaab
+
+/((?>a+)b)/
+    aaab
+ 0: aaab
+
+/(?>(a+))b/
+    aaab
+ 0: aaab
+
+/(?>b)+/
+    aaabbbccc
+ 0: bbb
+ 1: bb
+ 2: b
+
+/(?>a+|b+|c+)*c/
+    aaabbbbccccd
+ 0: aaabbbbcccc
+ 1: aaabbbbc
+    
+/(a+|b+|c+)*c/
+    aaabbbbccccd
+ 0: aaabbbbcccc
+ 1: aaabbbbccc
+ 2: aaabbbbcc
+ 3: aaabbbbc
+
+/((?>[^()]+)|\([^()]*\))+/
+    ((abc(ade)ufh()()x
+ 0: abc(ade)ufh()()x
+ 1: abc(ade)ufh()()
+ 2: abc(ade)ufh()
+ 3: abc(ade)ufh
+ 4: abc(ade)
+ 5: abc
+    
+/\(((?>[^()]+)|\([^()]+\))+\)/ 
+    (abc)
+ 0: (abc)
+    (abc(def)xyz)
+ 0: (abc(def)xyz)
+    *** Failers
+No match
+    ((()aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa   
+No match
+
+/a(?-i)b/i
+    ab
+ 0: ab
+    Ab
+ 0: Ab
+    *** Failers 
+No match
+    aB
+No match
+    AB
+No match
+        
+/(a (?x)b c)d e/
+    a bcd e
+ 0: a bcd e
+    *** Failers
+No match
+    a b cd e
+No match
+    abcd e   
+No match
+    a bcde 
+No match
+ 
+/(a b(?x)c d (?-x)e f)/
+    a bcde f
+ 0: a bcde f
+    *** Failers
+No match
+    abcdef  
+No match
+
+/(a(?i)b)c/
+    abc
+ 0: abc
+    aBc
+ 0: aBc
+    *** Failers
+No match
+    abC
+No match
+    aBC  
+No match
+    Abc
+No match
+    ABc
+No match
+    ABC
+No match
+    AbC
+No match
+    
+/a(?i:b)c/
+    abc
+ 0: abc
+    aBc
+ 0: aBc
+    *** Failers 
+No match
+    ABC
+No match
+    abC
+No match
+    aBC
+No match
+    
+/a(?i:b)*c/
+    aBc
+ 0: aBc
+    aBBc
+ 0: aBBc
+    *** Failers 
+No match
+    aBC
+No match
+    aBBC
+No match
+    
+/a(?=b(?i)c)\w\wd/
+    abcd
+ 0: abcd
+    abCd
+ 0: abCd
+    *** Failers
+No match
+    aBCd
+No match
+    abcD     
+No match
+    
+/(?s-i:more.*than).*million/i
+    more than million
+ 0: more than million
+    more than MILLION
+ 0: more than MILLION
+    more \n than Million 
+ 0: more \x0a than Million
+    *** Failers
+No match
+    MORE THAN MILLION    
+No match
+    more \n than \n million 
+No match
+
+/(?:(?s-i)more.*than).*million/i
+    more than million
+ 0: more than million
+    more than MILLION
+ 0: more than MILLION
+    more \n than Million 
+ 0: more \x0a than Million
+    *** Failers
+No match
+    MORE THAN MILLION    
+No match
+    more \n than \n million 
+No match
+    
+/(?>a(?i)b+)+c/ 
+    abc
+ 0: abc
+    aBbc
+ 0: aBbc
+    aBBc 
+ 0: aBBc
+    *** Failers
+No match
+    Abc
+No match
+    abAb    
+No match
+    abbC 
+No match
+    
+/(?=a(?i)b)\w\wc/
+    abc
+ 0: abc
+    aBc
+ 0: aBc
+    *** Failers
+No match
+    Ab 
+No match
+    abC
+No match
+    aBC     
+No match
+    
+/(?<=a(?i)b)(\w\w)c/
+    abxxc
+ 0: xxc
+    aBxxc
+ 0: xxc
+    *** Failers
+No match
+    Abxxc
+No match
+    ABxxc
+No match
+    abxxC      
+No match
+
+/^(?(?=abc)\w{3}:|\d\d)$/
+    abc:
+ 0: abc:
+    12
+ 0: 12
+    *** Failers
+No match
+    123
+No match
+    xyz    
+No match
+
+/^(?(?!abc)\d\d|\w{3}:)$/
+    abc:
+ 0: abc:
+    12
+ 0: 12
+    *** Failers
+No match
+    123
+No match
+    xyz    
+No match
+    
+/(?(?<=foo)bar|cat)/
+    foobar
+ 0: bar
+    cat
+ 0: cat
+    fcat
+ 0: cat
+    focat   
+ 0: cat
+    *** Failers
+No match
+    foocat  
+No match
+
+/(?(?a*)*/
+    a
+ 0: a
+ 1: 
+    aa
+ 0: aa
+ 1: 
+    aaaa
+ 0: aaaa
+ 1: 
+    
+/(abc|)+/
+    abc
+ 0: abc
+ 1: 
+    abcabc
+ 0: abcabc
+ 1: abc
+ 2: 
+    abcabcabc
+ 0: abcabcabc
+ 1: abcabc
+ 2: abc
+ 3: 
+    xyz      
+ 0: 
+
+/([a]*)*/
+    a
+ 0: a
+ 1: 
+    aaaaa 
+ 0: aaaaa
+ 1: aaaa
+ 2: aaa
+ 3: aa
+ 4: a
+ 5: 
+ 
+/([ab]*)*/
+    a
+ 0: a
+ 1: 
+    b
+ 0: b
+ 1: 
+    ababab
+ 0: ababab
+ 1: ababa
+ 2: abab
+ 3: aba
+ 4: ab
+ 5: a
+ 6: 
+    aaaabcde
+ 0: aaaab
+ 1: aaaa
+ 2: aaa
+ 3: aa
+ 4: a
+ 5: 
+    bbbb    
+ 0: bbbb
+ 1: bbb
+ 2: bb
+ 3: b
+ 4: 
+ 
+/([^a]*)*/
+    b
+ 0: b
+ 1: 
+    bbbb
+ 0: bbbb
+ 1: bbb
+ 2: bb
+ 3: b
+ 4: 
+    aaa   
+ 0: 
+ 
+/([^ab]*)*/
+    cccc
+ 0: cccc
+ 1: ccc
+ 2: cc
+ 3: c
+ 4: 
+    abab  
+ 0: 
+ 
+/([a]*?)*/
+    a
+ 0: a
+ 1: 
+    aaaa 
+ 0: aaaa
+ 1: aaa
+ 2: aa
+ 3: a
+ 4: 
+ 
+/([ab]*?)*/
+    a
+ 0: a
+ 1: 
+    b
+ 0: b
+ 1: 
+    abab
+ 0: abab
+ 1: aba
+ 2: ab
+ 3: a
+ 4: 
+    baba   
+ 0: baba
+ 1: bab
+ 2: ba
+ 3: b
+ 4: 
+ 
+/([^a]*?)*/
+    b
+ 0: b
+ 1: 
+    bbbb
+ 0: bbbb
+ 1: bbb
+ 2: bb
+ 3: b
+ 4: 
+    aaa   
+ 0: 
+ 
+/([^ab]*?)*/
+    c
+ 0: c
+ 1: 
+    cccc
+ 0: cccc
+ 1: ccc
+ 2: cc
+ 3: c
+ 4: 
+    baba   
+ 0: 
+ 
+/(?>a*)*/
+    a
+ 0: a
+ 1: 
+    aaabcde 
+ 0: aaa
+ 1: 
+ 
+/((?>a*))*/
+    aaaaa
+ 0: aaaaa
+ 1: 
+    aabbaa 
+ 0: aa
+ 1: 
+ 
+/((?>a*?))*/
+    aaaaa
+ 0: aaaaa
+ 1: 
+    aabbaa 
+ 0: aa
+ 1: 
+
+/(?(?=[^a-z]+[a-z])  \d{2}-[a-z]{3}-\d{2}  |  \d{2}-\d{2}-\d{2} ) /x
+    12-sep-98
+ 0: 12-sep-98
+    12-09-98
+ 0: 12-09-98
+    *** Failers
+No match
+    sep-12-98
+No match
+        
+/(?i:saturday|sunday)/
+    saturday
+ 0: saturday
+    sunday
+ 0: sunday
+    Saturday
+ 0: Saturday
+    Sunday
+ 0: Sunday
+    SATURDAY
+ 0: SATURDAY
+    SUNDAY
+ 0: SUNDAY
+    SunDay
+ 0: SunDay
+    
+/(a(?i)bc|BB)x/
+    abcx
+ 0: abcx
+    aBCx
+ 0: aBCx
+    bbx
+ 0: bbx
+    BBx
+ 0: BBx
+    *** Failers
+No match
+    abcX
+No match
+    aBCX
+No match
+    bbX
+No match
+    BBX               
+No match
+
+/^([ab](?i)[cd]|[ef])/
+    ac
+ 0: ac
+    aC
+ 0: aC
+    bD
+ 0: bD
+    elephant
+ 0: e
+    Europe 
+ 0: E
+    frog
+ 0: f
+    France
+ 0: F
+    *** Failers
+No match
+    Africa     
+No match
+
+/^(ab|a(?i)[b-c](?m-i)d|x(?i)y|z)/
+    ab
+ 0: ab
+    aBd
+ 0: aBd
+    xy
+ 0: xy
+    xY
+ 0: xY
+    zebra
+ 0: z
+    Zambesi
+ 0: Z
+    *** Failers
+No match
+    aCD  
+No match
+    XY  
+No match
+
+/(?<=foo\n)^bar/m
+    foo\nbar
+ 0: bar
+    *** Failers
+No match
+    bar
+No match
+    baz\nbar   
+No match
+
+/(?<=(?]&/
+    <&OUT
+ 0: <&
+
+/(?:(f)(o)(o)|(b)(a)(r))*/
+    foobar
+ 0: foobar
+ 1: foo
+ 2: 
+
+/(?<=a)b/
+    ab
+ 0: b
+    *** Failers
+No match
+    cb
+No match
+    b
+No match
+
+/(?a+)ab/
+
+/(?>a+)b/
+    aaab
+ 0: aaab
+
+/([[:]+)/
+    a:[b]:
+ 0: :[
+ 1: :
+
+/([[=]+)/
+    a=[b]=
+ 0: =[
+ 1: =
+
+/([[.]+)/
+    a.[b].
+ 0: .[
+ 1: .
+
+/((?>a+)b)/
+    aaab
+ 0: aaab
+
+/(?>(a+))b/
+    aaab
+ 0: aaab
+
+/((?>[^()]+)|\([^()]*\))+/
+    ((abc(ade)ufh()()x
+ 0: abc(ade)ufh()()x
+ 1: abc(ade)ufh()()
+ 2: abc(ade)ufh()
+ 3: abc(ade)ufh
+ 4: abc(ade)
+ 5: abc
+
+/a\Z/
+    *** Failers
+No match
+    aaab
+No match
+    a\nb\n
+No match
+
+/b\Z/
+    a\nb\n
+ 0: b
+
+/b\z/
+
+/b\Z/
+    a\nb
+ 0: b
+
+/b\z/
+    a\nb
+ 0: b
+    *** Failers
+No match
+    
+/(?>.*)(?<=(abcd|wxyz))/
+    alphabetabcd
+ 0: alphabetabcd
+    endingwxyz
+ 0: endingwxyz
+    *** Failers
+No match
+    a rather long string that doesn't end with one of them
+No match
+
+/word (?>(?:(?!otherword)[a-zA-Z0-9]+ ){0,30})otherword/
+    word cat dog elephant mussel cow horse canary baboon snake shark otherword
+ 0: word cat dog elephant mussel cow horse canary baboon snake shark otherword
+    word cat dog elephant mussel cow horse canary baboon snake shark
+No match
+  
+/word (?>[a-zA-Z0-9]+ ){0,30}otherword/
+    word cat dog elephant mussel cow horse canary baboon snake shark the quick brown fox and the lazy dog and several other words getting close to thirty by now I hope
+No match
+
+/(?<=\d{3}(?!999))foo/
+    999foo
+ 0: foo
+    123999foo 
+ 0: foo
+    *** Failers
+No match
+    123abcfoo
+No match
+    
+/(?<=(?!...999)\d{3})foo/
+    999foo
+ 0: foo
+    123999foo 
+ 0: foo
+    *** Failers
+No match
+    123abcfoo
+No match
+
+/(?<=\d{3}(?!999)...)foo/
+    123abcfoo
+ 0: foo
+    123456foo 
+ 0: foo
+    *** Failers
+No match
+    123999foo  
+No match
+    
+/(?<=\d{3}...)(?Z)+|A)*/
+    ZABCDEFG
+ 0: ZA
+ 1: Z
+ 2: 
+
+/((?>)+|A)*/
+    ZABCDEFG
+ 0: 
+
+/a*/g
+    abbab
+ 0: a
+ 1: 
+ 0: 
+ 0: 
+ 0: a
+ 1: 
+ 0: 
+ 0: 
+
+/^[a-\d]/
+    abcde
+ 0: a
+    -things
+ 0: -
+    0digit
+ 0: 0
+    *** Failers
+No match
+    bcdef    
+No match
+
+/^[\d-a]/
+    abcde
+ 0: a
+    -things
+ 0: -
+    0digit
+ 0: 0
+    *** Failers
+No match
+    bcdef    
+No match
+    
+/[[:space:]]+/
+    > \x09\x0a\x0c\x0d\x0b<
+ 0:  \x09\x0a\x0c\x0d\x0b
+ 1:  \x09\x0a\x0c\x0d
+ 2:  \x09\x0a\x0c
+ 3:  \x09\x0a
+ 4:  \x09
+ 5:  
+     
+/[[:blank:]]+/
+    > \x09\x0a\x0c\x0d\x0b<
+ 0:  \x09
+ 1:  
+     
+/[\s]+/
+    > \x09\x0a\x0c\x0d\x0b<
+ 0:  \x09\x0a\x0c\x0d
+ 1:  \x09\x0a\x0c
+ 2:  \x09\x0a
+ 3:  \x09
+ 4:  
+     
+/\s+/
+    > \x09\x0a\x0c\x0d\x0b<
+ 0:  \x09\x0a\x0c\x0d
+ 1:  \x09\x0a\x0c
+ 2:  \x09\x0a
+ 3:  \x09
+ 4:  
+     
+/ab/x
+    ab
+No match
+
+/(?!\A)x/m
+  a\nxb\n
+ 0: x
+
+/(?!^)x/m
+  a\nxb\n
+No match
+
+/abc\Qabc\Eabc/
+    abcabcabc
+ 0: abcabcabc
+    
+/abc\Q(*+|\Eabc/
+    abc(*+|abc 
+ 0: abc(*+|abc
+
+/   abc\Q abc\Eabc/x
+    abc abcabc
+ 0: abc abcabc
+    *** Failers
+No match
+    abcabcabc  
+No match
+    
+/abc#comment
+    \Q#not comment
+    literal\E/x
+    abc#not comment\n    literal     
+ 0: abc#not comment\x0a    literal
+
+/abc#comment
+    \Q#not comment
+    literal/x
+    abc#not comment\n    literal     
+ 0: abc#not comment\x0a    literal
+
+/abc#comment
+    \Q#not comment
+    literal\E #more comment
+    /x
+    abc#not comment\n    literal     
+ 0: abc#not comment\x0a    literal
+
+/abc#comment
+    \Q#not comment
+    literal\E #more comment/x
+    abc#not comment\n    literal     
+ 0: abc#not comment\x0a    literal
+
+/\Qabc\$xyz\E/
+    abc\\\$xyz
+ 0: abc\$xyz
+
+/\Qabc\E\$\Qxyz\E/
+    abc\$xyz
+ 0: abc$xyz
+
+/\Gabc/
+    abc
+ 0: abc
+    *** Failers
+No match
+    xyzabc  
+No match
+
+/\Gabc./g
+    abc1abc2xyzabc3
+ 0: abc1
+ 0: abc2
+
+/abc./g
+    abc1abc2xyzabc3 
+ 0: abc1
+ 0: abc2
+ 0: abc3
+
+/a(?x: b c )d/
+    XabcdY
+ 0: abcd
+    *** Failers 
+No match
+    Xa b c d Y 
+No match
+
+/((?x)x y z | a b c)/
+    XabcY
+ 0: abc
+    AxyzB 
+ 0: xyz
+
+/(?i)AB(?-i)C/
+    XabCY
+ 0: abC
+    *** Failers
+No match
+    XabcY  
+No match
+
+/((?i)AB(?-i)C|D)E/
+    abCE
+ 0: abCE
+    DE
+ 0: DE
+    *** Failers
+No match
+    abcE
+No match
+    abCe  
+No match
+    dE
+No match
+    De    
+No match
+
+/[z\Qa-d]\E]/
+    z
+ 0: z
+    a
+ 0: a
+    -
+ 0: -
+    d
+ 0: d
+    ] 
+ 0: ]
+    *** Failers
+ 0: a
+    b     
+No match
+
+/[\z\C]/
+    z
+ 0: z
+    C 
+ 0: C
+    
+/\M/
+    M 
+ 0: M
+    
+/(a+)*b/
+    aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa 
+No match
+    
+/(?i)reg(?:ul(?:[a]|ae)r|ex)/
+    REGular
+ 0: REGular
+    regulaer
+ 0: regulaer
+    Regex  
+ 0: Regex
+    regulr 
+ 0: regul\xe4r
+
+/[--]+/
+    
+ 0: \xc5\xe6\xe5\xe4\xe0
+    
+ 0: \xc5\xe6\xe5\xe4\xff
+    
+ 0: \xc5\xe6\xe5\xe4\xc0
+    
+ 0: \xc5\xe6\xe5\xe4\xdf
+
+/(?<=Z)X./
+    \x84XAZXB
+ 0: XB
+
+/^(?(2)a|(1)(2))+$/
+    123a
+Error -17 (backreference condition or recursion test not supported for DFA matching)
+
+/(?<=a|bbbb)c/
+    ac
+ 0: c
+    bbbbc
+ 0: c
+
+/abc/SS>testsavedregex
+Compiled pattern written to testsavedregex
+testsavedregex
+Compiled pattern written to testsavedregex
+testsavedregex
+Compiled pattern written to testsavedregex
+Study data written to testsavedregex
+testsavedregex
+Compiled pattern written to testsavedregex
+Study data written to testsavedregex
+
+ 0: abc
+    xyz\r\nabc\
+ 0: abc
+    xyz\rabc\
+ 0: abc
+    xyz\r\nabc\
+ 0: abc
+    ** Failers 
+No match
+    xyz\nabc\
+No match
+    xyz\r\nabc\
+No match
+    xyz\nabc\
+No match
+    xyz\rabc\
+No match
+    xyz\rabc\
+No match
+    
+/abc$/m
+    xyzabc
+ 0: abc
+    xyzabc\n 
+ 0: abc
+    xyzabc\npqr 
+ 0: abc
+    xyzabc\r\ 
+ 0: abc
+    xyzabc\rpqr\ 
+ 0: abc
+    xyzabc\r\n\ 
+ 0: abc
+    xyzabc\r\npqr\ 
+ 0: abc
+    ** Failers
+No match
+    xyzabc\r 
+No match
+    xyzabc\rpqr 
+No match
+    xyzabc\r\n 
+No match
+    xyzabc\r\npqr 
+No match
+    
+/^abc/m
+    xyz\rabcdef
+ 0: abc
+    xyz\nabcdef\
+ 0: abc
+    ** Failers  
+No match
+    xyz\nabcdef
+No match
+       
+/^abc/m
+    xyz\nabcdef
+ 0: abc
+    xyz\rabcdef\
+ 0: abc
+    ** Failers  
+No match
+    xyz\rabcdef
+No match
+       
+/^abc/m
+    xyz\r\nabcdef
+ 0: abc
+    xyz\rabcdef\
+ 0: abc
+    ** Failers  
+No match
+    xyz\rabcdef
+No match
+    
+/.*/
+    abc\ndef
+ 0: abc
+ 1: ab
+ 2: a
+ 3: 
+    abc\rdef
+ 0: abc\x0ddef
+ 1: abc\x0dde
+ 2: abc\x0dd
+ 3: abc\x0d
+ 4: abc
+ 5: ab
+ 6: a
+ 7: 
+    abc\r\ndef
+ 0: abc\x0d
+ 1: abc
+ 2: ab
+ 3: a
+ 4: 
+    \abc\ndef
+ 0: abc\x0adef
+ 1: abc\x0ade
+ 2: abc\x0ad
+ 3: abc\x0a
+ 4: abc
+ 5: ab
+ 6: a
+ 7: 
+    \abc\rdef
+ 0: abc
+ 1: ab
+ 2: a
+ 3: 
+    \abc\r\ndef
+ 0: abc
+ 1: ab
+ 2: a
+ 3: 
+    \abc\ndef
+ 0: abc\x0adef
+ 1: abc\x0ade
+ 2: abc\x0ad
+ 3: abc\x0a
+ 4: abc
+ 5: ab
+ 6: a
+ 7: 
+    \abc\rdef
+ 0: abc\x0ddef
+ 1: abc\x0dde
+ 2: abc\x0dd
+ 3: abc\x0d
+ 4: abc
+ 5: ab
+ 6: a
+ 7: 
+    \abc\r\ndef
+ 0: abc
+ 1: ab
+ 2: a
+ 3: 
+
+/\w+(.)(.)?def/s
+    abc\ndef
+ 0: abc\x0adef
+    abc\rdef
+ 0: abc\x0ddef
+    abc\r\ndef
+ 0: abc\x0d\x0adef
+
+/^\w+=.*(\\\n.*)*/
+    abc=xyz\\\npqr
+ 0: abc=xyz\\x0apqr
+ 1: abc=xyz\\x0apq
+ 2: abc=xyz\\x0ap
+ 3: abc=xyz\\x0a
+ 4: abc=xyz\
+ 5: abc=xyz
+ 6: abc=xy
+ 7: abc=x
+ 8: abc=
+
+/^(a()*)*/
+    aaaa
+ 0: aaaa
+ 1: aaa
+ 2: aa
+ 3: a
+ 4: 
+
+/^(?:a(?:(?:))*)*/
+    aaaa
+ 0: aaaa
+ 1: aaa
+ 2: aa
+ 3: a
+ 4: 
+
+/^(a()+)+/
+    aaaa
+ 0: aaaa
+ 1: aaa
+ 2: aa
+ 3: a
+
+/^(?:a(?:(?:))+)+/
+    aaaa
+ 0: aaaa
+ 1: aaa
+ 2: aa
+ 3: a
+
+/(a|)*\d/
+  aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+No match
+  aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa4
+ 0: aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa4
+
+/(?>a|)*\d/
+  aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+No match
+  aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa4
+ 0: aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa4
+
+/(?:a|)*\d/
+  aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+No match
+  aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa4
+ 0: aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa4
+
+/^a.b/
+    a\rb
+ 0: a\x0db
+    a\nb\ 
+ 0: a\x0ab
+    ** Failers
+No match
+    a\nb
+No match
+    a\nb\
+No match
+    a\rb\   
+No match
+    a\rb\   
+No match
+
+/^abc./mgx
+    abc1 \x0aabc2 \x0babc3xx \x0cabc4 \x0dabc5xx \x0d\x0aabc6 \x85abc7 JUNK
+ 0: abc1
+ 0: abc2
+ 0: abc3
+ 0: abc4
+ 0: abc5
+ 0: abc6
+ 0: abc7
+
+/abc.$/mgx
+    abc1\x0a abc2\x0b abc3\x0c abc4\x0d abc5\x0d\x0a abc6\x85 abc9
+ 0: abc1
+ 0: abc2
+ 0: abc3
+ 0: abc4
+ 0: abc5
+ 0: abc6
+ 0: abc9
+
+/^a\Rb/
+    a\nb
+ 0: a\x0ab
+    a\rb
+ 0: a\x0db
+    a\r\nb
+ 0: a\x0d\x0ab
+    a\x0bb
+ 0: a\x0bb
+    a\x0cb
+ 0: a\x0cb
+    a\x85b   
+ 0: a\x85b
+    ** Failers
+No match
+    a\n\rb    
+No match
+
+/^a\R*b/
+    ab
+ 0: ab
+    a\nb
+ 0: a\x0ab
+    a\rb
+ 0: a\x0db
+    a\r\nb
+ 0: a\x0d\x0ab
+    a\x0bb
+ 0: a\x0bb
+    a\x0cb
+ 0: a\x0cb
+    a\x85b   
+ 0: a\x85b
+    a\n\rb    
+ 0: a\x0a\x0db
+    a\n\r\x85\x0cb 
+ 0: a\x0a\x0d\x85\x0cb
+
+/^a\R+b/
+    a\nb
+ 0: a\x0ab
+    a\rb
+ 0: a\x0db
+    a\r\nb
+ 0: a\x0d\x0ab
+    a\x0bb
+ 0: a\x0bb
+    a\x0cb
+ 0: a\x0cb
+    a\x85b   
+ 0: a\x85b
+    a\n\rb    
+ 0: a\x0a\x0db
+    a\n\r\x85\x0cb 
+ 0: a\x0a\x0d\x85\x0cb
+    ** Failers
+No match
+    ab  
+No match
+    
+/^a\R{1,3}b/
+    a\nb
+ 0: a\x0ab
+    a\n\rb
+ 0: a\x0a\x0db
+    a\n\r\x85b
+ 0: a\x0a\x0d\x85b
+    a\r\n\r\nb 
+ 0: a\x0d\x0a\x0d\x0ab
+    a\r\n\r\n\r\nb 
+ 0: a\x0d\x0a\x0d\x0a\x0d\x0ab
+    a\n\r\n\rb
+ 0: a\x0a\x0d\x0a\x0db
+    a\n\n\r\nb 
+ 0: a\x0a\x0a\x0d\x0ab
+    ** Failers
+No match
+    a\n\n\n\rb
+No match
+    a\r
+No match
+
+/^a[\R]b/
+    aRb
+ 0: aRb
+    ** Failers
+No match
+    a\nb  
+No match
+
+/.+foo/
+    afoo
+ 0: afoo
+    ** Failers 
+No match
+    \r\nfoo 
+No match
+    \nfoo 
+No match
+
+/.+foo/
+    afoo
+ 0: afoo
+    \nfoo 
+ 0: \x0afoo
+    ** Failers 
+No match
+    \r\nfoo 
+No match
+
+/.+foo/
+    afoo
+ 0: afoo
+    ** Failers 
+No match
+    \nfoo 
+No match
+    \r\nfoo 
+No match
+
+/.+foo/s
+    afoo
+ 0: afoo
+    \r\nfoo 
+ 0: \x0d\x0afoo
+    \nfoo 
+ 0: \x0afoo
+
+/^$/mg
+    abc\r\rxyz
+ 0: 
+    abc\n\rxyz  
+ 0: 
+    ** Failers 
+No match
+    abc\r\nxyz
+No match
+
+/^X/m
+    XABC
+ 0: X
+    ** Failers 
+No match
+    XABC\B
+No match
+
+/(?m)^$/g+
+    abc\r\n\r\n
+ 0: 
+ 0+ \x0d\x0a
+
+/(?m)^$|^\r\n/g+ 
+    abc\r\n\r\n
+ 0: \x0d\x0a
+ 0+ 
+ 1: 
+    
+/(?m)$/g+ 
+    abc\r\n\r\n
+ 0: 
+ 0+ \x0d\x0a\x0d\x0a
+ 0: 
+ 0+ \x0d\x0a
+ 0: 
+ 0+ 
+
+/(?|(abc)|(xyz))/
+   >abc<
+ 0: abc
+   >xyz< 
+ 0: xyz
+
+/(x)(?|(abc)|(xyz))(x)/
+    xabcx
+ 0: xabcx
+    xxyzx 
+ 0: xxyzx
+
+/(x)(?|(abc)(pqr)|(xyz))(x)/
+    xabcpqrx
+ 0: xabcpqrx
+    xxyzx 
+ 0: xxyzx
+
+/(?|(abc)|(xyz))(?1)/
+    abcabc
+ 0: abcabc
+    xyzabc 
+ 0: xyzabc
+    ** Failers 
+No match
+    xyzxyz 
+No match
+ 
+/\H\h\V\v/
+    X X\x0a
+ 0: X X\x0a
+    X\x09X\x0b
+ 0: X\x09X\x0b
+    ** Failers
+No match
+    \xa0 X\x0a   
+No match
+    
+/\H*\h+\V?\v{3,4}/ 
+    \x09\x20\xa0X\x0a\x0b\x0c\x0d\x0a
+ 0: \x09 \xa0X\x0a\x0b\x0c\x0d
+ 1: \x09 \xa0X\x0a\x0b\x0c
+    \x09\x20\xa0\x0a\x0b\x0c\x0d\x0a
+ 0: \x09 \xa0\x0a\x0b\x0c\x0d
+ 1: \x09 \xa0\x0a\x0b\x0c
+    \x09\x20\xa0\x0a\x0b\x0c
+ 0: \x09 \xa0\x0a\x0b\x0c
+    ** Failers 
+No match
+    \x09\x20\xa0\x0a\x0b
+No match
+     
+/\H{3,4}/
+    XY  ABCDE
+ 0: ABCD
+ 1: ABC
+    XY  PQR ST 
+ 0: PQR
+    
+/.\h{3,4}./
+    XY  AB    PQRS
+ 0: B    P
+ 1: B    
+
+/\h*X\h?\H+Y\H?Z/
+    >XNNNYZ
+ 0: XNNNYZ
+    >  X NYQZ
+ 0:   X NYQZ
+    ** Failers
+No match
+    >XYZ   
+No match
+    >  X NY Z
+No match
+
+/\v*X\v?Y\v+Z\V*\x0a\V+\x0b\V{2,3}\x0c/
+    >XY\x0aZ\x0aA\x0bNN\x0c
+ 0: XY\x0aZ\x0aA\x0bNN\x0c
+    >\x0a\x0dX\x0aY\x0a\x0bZZZ\x0aAAA\x0bNNN\x0c
+ 0: \x0a\x0dX\x0aY\x0a\x0bZZZ\x0aAAA\x0bNNN\x0c
+
+/.+A/
+    \r\nA
+No match
+    
+/\nA/
+    \r\nA 
+ 0: \x0aA
+
+/[\r\n]A/
+    \r\nA 
+ 0: \x0aA
+
+/(\r|\n)A/
+    \r\nA 
+ 0: \x0aA
+
+/a\Rb/I
+Capturing subpattern count = 0
+Options: bsr_anycrlf
+First char = 'a'
+Need char = 'b'
+    a\rb
+ 0: a\x0db
+    a\nb
+ 0: a\x0ab
+    a\r\nb
+ 0: a\x0d\x0ab
+    ** Failers
+No match
+    a\x85b
+No match
+    a\x0bb     
+No match
+
+/a\Rb/I
+Capturing subpattern count = 0
+Options: bsr_unicode
+First char = 'a'
+Need char = 'b'
+    a\rb
+ 0: a\x0db
+    a\nb
+ 0: a\x0ab
+    a\r\nb
+ 0: a\x0d\x0ab
+    a\x85b
+ 0: a\x85b
+    a\x0bb     
+ 0: a\x0bb
+    ** Failers 
+No match
+    a\x85b\
+No match
+    a\x0bb\
+No match
+    
+/a\R?b/I
+Capturing subpattern count = 0
+Options: bsr_anycrlf
+First char = 'a'
+Need char = 'b'
+    a\rb
+ 0: a\x0db
+    a\nb
+ 0: a\x0ab
+    a\r\nb
+ 0: a\x0d\x0ab
+    ** Failers
+No match
+    a\x85b
+No match
+    a\x0bb     
+No match
+
+/a\R?b/I
+Capturing subpattern count = 0
+Options: bsr_unicode
+First char = 'a'
+Need char = 'b'
+    a\rb
+ 0: a\x0db
+    a\nb
+ 0: a\x0ab
+    a\r\nb
+ 0: a\x0d\x0ab
+    a\x85b
+ 0: a\x85b
+    a\x0bb     
+ 0: a\x0bb
+    ** Failers 
+No match
+    a\x85b\
+No match
+    a\x0bb\
+No match
+    
+/a\R{2,4}b/I
+Capturing subpattern count = 0
+Options: bsr_anycrlf
+First char = 'a'
+Need char = 'b'
+    a\r\n\nb
+ 0: a\x0d\x0a\x0ab
+    a\n\r\rb
+ 0: a\x0a\x0d\x0db
+    a\r\n\r\n\r\n\r\nb
+ 0: a\x0d\x0a\x0d\x0a\x0d\x0a\x0d\x0ab
+    ** Failers
+No match
+    a\x85\85b
+No match
+    a\x0b\0bb     
+No match
+
+/a\R{2,4}b/I
+Capturing subpattern count = 0
+Options: bsr_unicode
+First char = 'a'
+Need char = 'b'
+    a\r\rb
+ 0: a\x0d\x0db
+    a\n\n\nb
+ 0: a\x0a\x0a\x0ab
+    a\r\n\n\r\rb
+ 0: a\x0d\x0a\x0a\x0d\x0db
+    a\x85\85b
+No match
+    a\x0b\0bb     
+No match
+    ** Failers 
+No match
+    a\r\r\r\r\rb 
+No match
+    a\x85\85b\
+No match
+    a\x0b\0bb\
+No match
+    
+/a(?!)|\wbc/
+    abc 
+ 0: abc
+
+/a[]b/
+    ** Failers
+No match
+    ab
+No match
+
+/a[]+b/
+    ** Failers
+No match
+    ab 
+No match
+
+/a[]*+b/
+    ** Failers
+No match
+    ab 
+No match
+
+/a[^]b/
+    aXb
+ 0: aXb
+    a\nb 
+ 0: a\x0ab
+    ** Failers
+No match
+    ab  
+No match
+    
+/a[^]+b/ 
+    aXb
+ 0: aXb
+    a\nX\nXb 
+ 0: a\x0aX\x0aXb
+    ** Failers
+No match
+    ab  
+No match
+
+/X$/E
+    X
+ 0: X
+    ** Failers 
+No match
+    X\n 
+No match
+
+/X$/
+    X
+ 0: X
+    X\n 
+ 0: X
+
+/xyz/C
+  xyz 
+--->xyz
+ +0 ^       x
+ +1 ^^      y
+ +2 ^ ^     z
+ +3 ^  ^    
+ 0: xyz
+  abcxyz 
+--->abcxyz
+ +0    ^       x
+ +1    ^^      y
+ +2    ^ ^     z
+ +3    ^  ^    
+ 0: xyz
+  abcxyz\Y
+--->abcxyz
+ +0 ^          x
+ +0  ^         x
+ +0   ^        x
+ +0    ^       x
+ +1    ^^      y
+ +2    ^ ^     z
+ +3    ^  ^    
+ 0: xyz
+  ** Failers 
+No match
+  abc
+No match
+  abc\Y
+--->abc
+ +0 ^       x
+ +0  ^      x
+ +0   ^     x
+ +0    ^    x
+No match
+  abcxypqr  
+No match
+  abcxypqr\Y  
+--->abcxypqr
+ +0 ^            x
+ +0  ^           x
+ +0   ^          x
+ +0    ^         x
+ +1    ^^        y
+ +2    ^ ^       z
+ +0     ^        x
+ +0      ^       x
+ +0       ^      x
+ +0        ^     x
+ +0         ^    x
+No match
+
+/(*NO_START_OPT)xyz/C
+  abcxyz 
+--->abcxyz
++15 ^          x
++15  ^         x
++15   ^        x
++15    ^       x
++16    ^^      y
++17    ^ ^     z
++18    ^  ^    
+ 0: xyz
+  
+/(?C)ab/
+  ab
+--->ab
+  0 ^      a
+ 0: ab
+  \C-ab
+ 0: ab
+  
+/ab/C
+  ab
+--->ab
+ +0 ^      a
+ +1 ^^     b
+ +2 ^ ^    
+ 0: ab
+  \C-ab    
+ 0: ab
+
+/^"((?(?=[a])[^"])|b)*"$/C
+    "ab"
+--->"ab"
+ +0 ^        ^
+ +1 ^        "
+ +2 ^^       ((?(?=[a])[^"])|b)*
++21 ^^       "
+ +3 ^^       (?(?=[a])[^"])
++18 ^^       b
+ +5 ^^       (?=[a])
+ +8  ^       [a]
++11  ^^      )
++12 ^^       [^"]
++16 ^ ^      )
++17 ^ ^      |
++21 ^ ^      "
+ +3 ^ ^      (?(?=[a])[^"])
++18 ^ ^      b
+ +5 ^ ^      (?=[a])
+ +8   ^      [a]
++19 ^  ^     )
++21 ^  ^     "
+ +3 ^  ^     (?(?=[a])[^"])
++18 ^  ^     b
+ +5 ^  ^     (?=[a])
+ +8    ^     [a]
++17 ^  ^     |
++22 ^   ^    $
++23 ^   ^    
+ 0: "ab"
+    \C-"ab"
+ 0: "ab"
+
+/\d+X|9+Y/
+    ++++123999\P
+Partial match: 123999
+    ++++123999Y\P
+ 0: 999Y
+
+/Z(*F)/
+    Z\P
+No match
+    ZA\P 
+No match
+    
+/Z(?!)/
+    Z\P 
+No match
+    ZA\P 
+No match
+
+/dog(sbody)?/
+    dogs\P
+ 0: dog
+    dogs\P\P 
+Partial match: dogs
+    
+/dog(sbody)??/
+    dogs\P
+ 0: dog
+    dogs\P\P 
+Partial match: dogs
+
+/dog|dogsbody/
+    dogs\P
+ 0: dog
+    dogs\P\P 
+Partial match: dogs
+ 
+/dogsbody|dog/
+    dogs\P
+ 0: dog
+    dogs\P\P 
+Partial match: dogs
+
+/Z(*F)Q|ZXY/
+    Z\P
+Partial match: Z
+    ZA\P 
+No match
+    X\P 
+No match
+
+/\bthe cat\b/
+    the cat\P
+ 0: the cat
+    the cat\P\P
+Partial match: the cat
+
+/dog(sbody)?/
+    dogs\D\P
+ 0: dog
+    body\D\R
+ 0: body
+
+/dog(sbody)?/
+    dogs\D\P\P
+Partial match: dogs
+    body\D\R
+ 0: body
+
+/abc/
+   abc\P
+ 0: abc
+   abc\P\P
+ 0: abc
+
+/abc\K123/
+    xyzabc123pqr
+Error -16 (item unsupported for DFA matching)
+    
+/(?<=abc)123/
+    xyzabc123pqr 
+ 0: 123
+    xyzabc12\P
+Partial match at offset 6: abc12
+    xyzabc12\P\P
+Partial match at offset 6: abc12
+
+/\babc\b/
+    +++abc+++
+ 0: abc
+    +++ab\P
+Partial match at offset 3: +ab
+    +++ab\P\P  
+Partial match at offset 3: +ab
+
+/(?=C)/g+
+    ABCDECBA
+ 0: 
+ 0+ CDECBA
+ 0: 
+ 0+ CBA
+
+/(abc|def|xyz)/I
+Capturing subpattern count = 1
+No options
+No first char
+No need char
+    terhjk;abcdaadsfe
+ 0: abc
+    the quick xyz brown fox 
+ 0: xyz
+    \Yterhjk;abcdaadsfe
+ 0: abc
+    \Ythe quick xyz brown fox 
+ 0: xyz
+    ** Failers
+No match
+    thejk;adlfj aenjl;fda asdfasd ehj;kjxyasiupd
+No match
+    \Ythejk;adlfj aenjl;fda asdfasd ehj;kjxyasiupd
+No match
+
+/(abc|def|xyz)/SI
+Capturing subpattern count = 1
+No options
+No first char
+No need char
+Subject length lower bound = 3
+Starting byte set: a d x 
+    terhjk;abcdaadsfe
+ 0: abc
+    the quick xyz brown fox 
+ 0: xyz
+    \Yterhjk;abcdaadsfe
+ 0: abc
+    \Ythe quick xyz brown fox 
+ 0: xyz
+    ** Failers
+No match
+    thejk;adlfj aenjl;fda asdfasd ehj;kjxyasiupd
+No match
+    \Ythejk;adlfj aenjl;fda asdfasd ehj;kjxyasiupd
+No match
+
+/abcd*/+
+    xxxxabcd\P
+ 0: abcd
+ 0+ 
+ 1: abc
+    xxxxabcd\P\P
+Partial match: abcd
+    dddxxx\R 
+ 0: ddd
+ 0+ xxx
+ 1: dd
+ 2: d
+ 3: 
+    xxxxabcd\P\P
+Partial match: abcd
+    xxx\R 
+ 0: 
+ 0+ xxx
+
+/abcd*/i
+    xxxxabcd\P
+ 0: abcd
+ 1: abc
+    xxxxabcd\P\P
+Partial match: abcd
+    XXXXABCD\P
+ 0: ABCD
+ 1: ABC
+    XXXXABCD\P\P
+Partial match: ABCD
+
+/abc\d*/
+    xxxxabc1\P
+ 0: abc1
+ 1: abc
+    xxxxabc1\P\P
+Partial match: abc1
+
+/abc[de]*/
+    xxxxabcde\P
+ 0: abcde
+ 1: abcd
+ 2: abc
+    xxxxabcde\P\P
+Partial match: abcde
+
+/(?:(?1)|B)(A(*F)|C)/
+    ABCD
+ 0: BC
+    CCD
+ 0: CC
+    ** Failers
+No match
+    CAD   
+No match
+
+/^(?:(?1)|B)(A(*F)|C)/
+    CCD
+ 0: CC
+    BCD 
+ 0: BC
+    ** Failers
+No match
+    ABCD
+No match
+    CAD
+No match
+    BAD    
+No match
+
+/^(?!a(*SKIP)b)/
+    ac
+Error -16 (item unsupported for DFA matching)
+    
+/^(?=a(*SKIP)b|ac)/
+    ** Failers
+No match
+    ac
+Error -16 (item unsupported for DFA matching)
+    
+/^(?=a(*THEN)b|ac)/
+    ac
+Error -16 (item unsupported for DFA matching)
+    
+/^(?=a(*PRUNE)b)/
+    ab  
+Error -16 (item unsupported for DFA matching)
+    ** Failers 
+No match
+    ac
+Error -16 (item unsupported for DFA matching)
+
+/^(?(?!a(*SKIP)b))/
+    ac
+Error -16 (item unsupported for DFA matching)
+
+/(?<=abc)def/
+    abc\P\P
+Partial match at offset 3: abc
+
+/abc$/
+    abc
+ 0: abc
+    abc\P
+ 0: abc
+    abc\P\P
+Partial match: abc
+
+/abc$/m
+    abc
+ 0: abc
+    abc\n
+ 0: abc
+    abc\P\P
+Partial match: abc
+    abc\n\P\P 
+ 0: abc
+    abc\P
+ 0: abc
+    abc\n\P
+ 0: abc
+
+/abc\z/
+    abc
+ 0: abc
+    abc\P
+ 0: abc
+    abc\P\P
+Partial match: abc
+
+/abc\Z/
+    abc
+ 0: abc
+    abc\P
+ 0: abc
+    abc\P\P
+Partial match: abc
+
+/abc\b/
+    abc
+ 0: abc
+    abc\P
+ 0: abc
+    abc\P\P
+Partial match: abc
+
+/abc\B/
+    abc
+No match
+    abc\P
+Partial match: abc
+    abc\P\P
+Partial match: abc
+
+/.+/
+    abc\>0
+ 0: abc
+ 1: ab
+ 2: a
+    abc\>1
+ 0: bc
+ 1: b
+    abc\>2
+ 0: c
+    abc\>3
+No match
+    abc\>4
+Error -24 (bad offset value)
+    abc\>-4 
+Error -24 (bad offset value)
+
+/^(?:a)++\w/
+     aaaab
+ 0: aaaab
+     ** Failers 
+No match
+     aaaa 
+No match
+     bbb 
+No match
+
+/^(?:aa|(?:a)++\w)/
+     aaaab
+ 0: aaaab
+ 1: aa
+     aaaa 
+ 0: aa
+     ** Failers 
+No match
+     bbb 
+No match
+
+/^(?:a)*+\w/
+     aaaab
+ 0: aaaab
+     bbb 
+ 0: b
+     ** Failers 
+No match
+     aaaa 
+No match
+
+/^(a)++\w/
+     aaaab
+ 0: aaaab
+     ** Failers 
+No match
+     aaaa 
+No match
+     bbb 
+No match
+
+/^(a|)++\w/
+     aaaab
+ 0: aaaab
+     ** Failers 
+No match
+     aaaa 
+No match
+     bbb 
+No match
+
+/(?=abc){3}abc/+
+    abcabcabc
+ 0: abc
+ 0+ abcabc
+    ** Failers
+No match
+    xyz  
+No match
+    
+/(?=abc)+abc/+
+    abcabcabc
+ 0: abc
+ 0+ abcabc
+    ** Failers
+No match
+    xyz  
+No match
+    
+/(?=abc)++abc/+
+    abcabcabc
+ 0: abc
+ 0+ abcabc
+    ** Failers
+No match
+    xyz  
+No match
+    
+/(?=abc){0}xyz/
+    xyz 
+ 0: xyz
+
+/(?=abc){1}xyz/
+    ** Failers
+No match
+    xyz 
+No match
+    
+/(?=(a))?./
+    ab
+ 0: a
+    bc
+ 0: b
+      
+/(?=(a))??./
+    ab
+ 0: a
+    bc
+ 0: b
+
+/^(?=(a)){0}b(?1)/
+    backgammon
+ 0: ba
+
+/^(?=(?1))?[az]([abc])d/
+    abd 
+ 0: abd
+    zcdxx 
+ 0: zcd
+
+/^(?!a){0}\w+/
+    aaaaa
+ 0: aaaaa
+ 1: aaaa
+ 2: aaa
+ 3: aa
+ 4: a
+
+/(?<=(abc))?xyz/
+    abcxyz
+ 0: xyz
+    pqrxyz 
+ 0: xyz
+
+/((?2))((?1))/
+    abc
+Error -26 (nested recursion at the same subject position)
+
+/(?(R)a+|(?R)b)/
+    aaaabcde
+ 0: aaaab
+
+/(?(R)a+|((?R))b)/
+    aaaabcde
+ 0: aaaab
+
+/((?(R)a+|(?1)b))/
+    aaaabcde
+ 0: aaaab
+
+/((?(R2)a+|(?1)b))/
+    aaaabcde
+Error -17 (backreference condition or recursion test not supported for DFA matching)
+
+/(?(R)a*(?1)|((?R))b)/
+    aaaabcde
+Error -26 (nested recursion at the same subject position)
+
+/(a+)/
+    \O6aaaa
+Matched, but offsets vector is too small to show all matches
+ 0: aaaa
+ 1: aaa
+ 2: aa
+    \O8aaaa
+ 0: aaaa
+ 1: aaa
+ 2: aa
+ 3: a
+
+/ab\Cde/
+    abXde
+ 0: abXde
+    
+/(?<=ab\Cde)X/
+    abZdeX
+ 0: X
+
+/^\R/
+    \r\P
+ 0: \x0d
+    \r\P\P
+Partial match: \x0d
+    
+/^\R{2,3}x/
+    \r\P
+Partial match: \x0d
+    \r\P\P
+Partial match: \x0d
+    \r\r\P 
+Partial match: \x0d\x0d
+    \r\r\P\P
+Partial match: \x0d\x0d
+    \r\r\r\P  
+Partial match: \x0d\x0d\x0d
+    \r\r\r\P\P
+Partial match: \x0d\x0d\x0d
+    \r\rx
+ 0: \x0d\x0dx
+    \r\r\rx    
+ 0: \x0d\x0d\x0dx
+
+/^\R{2,3}?x/
+    \r\P
+Partial match: \x0d
+    \r\P\P
+Partial match: \x0d
+    \r\r\P 
+Partial match: \x0d\x0d
+    \r\r\P\P
+Partial match: \x0d\x0d
+    \r\r\r\P  
+Partial match: \x0d\x0d\x0d
+    \r\r\r\P\P
+Partial match: \x0d\x0d\x0d
+    \r\rx
+ 0: \x0d\x0dx
+    \r\r\rx    
+ 0: \x0d\x0d\x0dx
+    
+/^\R?x/
+    \r\P
+Partial match: \x0d
+    \r\P\P 
+Partial match: \x0d
+    x
+ 0: x
+    \rx  
+ 0: \x0dx
+
+/^\R+x/
+    \r\P
+Partial match: \x0d
+    \r\P\P 
+Partial match: \x0d
+    \r\n\P
+Partial match: \x0d\x0a
+    \r\n\P\P  
+Partial match: \x0d\x0a
+    \rx  
+ 0: \x0dx
+
+/^a$/
+    a\r\P
+Partial match: a\x0d
+    a\r\P\P 
+Partial match: a\x0d
+
+/^a$/m
+    a\r\P
+Partial match: a\x0d
+    a\r\P\P 
+Partial match: a\x0d
+
+/^(a$|a\r)/
+    a\r\P
+ 0: a\x0d
+    a\r\P\P 
+Partial match: a\x0d
+
+/^(a$|a\r)/m
+    a\r\P
+ 0: a\x0d
+    a\r\P\P 
+Partial match: a\x0d
+
+/./
+    \r\P
+ 0: \x0d
+    \r\P\P 
+Partial match: \x0d
+  
+/.{2,3}/
+    \r\P 
+Partial match: \x0d
+    \r\P\P
+Partial match: \x0d
+    \r\r\P
+ 0: \x0d\x0d
+    \r\r\P\P
+Partial match: \x0d\x0d
+    \r\r\r\P
+ 0: \x0d\x0d\x0d
+ 1: \x0d\x0d
+    \r\r\r\P\P     
+Partial match: \x0d\x0d\x0d
+
+/.{2,3}?/
+    \r\P 
+Partial match: \x0d
+    \r\P\P
+Partial match: \x0d
+    \r\r\P
+ 0: \x0d\x0d
+    \r\r\P\P
+Partial match: \x0d\x0d
+    \r\r\r\P
+ 0: \x0d\x0d\x0d
+ 1: \x0d\x0d
+    \r\r\r\P\P     
+Partial match: \x0d\x0d\x0d
+
+/-- Test simple validity check for restarts --/
+
+/abcdef/
+   abc\R
+Error -30 (invalid data in workspace for DFA restart)
+
+/)(.)|(?R))++)*F>/
+    text  text xxxxx  text F> text2  more text.
+ 0:  text xxxxx  text F>
+
+/^(?>.{4})abc|^\w\w.xabcd/
+    xxxxabcd
+ 0: xxxxabcd
+ 1: xxxxabc
+    xx\xa0xabcd 
+ 0: xx\xa0xabcd
+ 1: xx\xa0xabc
+
+/^(.{4}){2}+abc|^\w\w.x\w\w\w\wabcd/
+    xxxxxxxxabcd
+ 0: xxxxxxxxabcd
+ 1: xxxxxxxxabc
+    xx\xa0xxxxxabcd 
+ 0: xx\xa0xxxxxabcd
+ 1: xx\xa0xxxxxabc
+
+/abcd/
+    abcd\O0
+Matched, but offsets vector is too small to show all matches
+
+/-- End of testinput8 --/
diff --git a/pcre/testdata/testoutput9 b/pcre/testdata/testoutput9
new file mode 100644
index 00000000000..0bb101ad614
--- /dev/null
+++ b/pcre/testdata/testoutput9
@@ -0,0 +1,1371 @@
+/-- This set of tests checks UTF-8 support with the DFA matching functionality
+    of pcre_dfa_exec(). The -dfa flag must be used with pcretest when running 
+    it. --/
+
+/\x{100}ab/8
+  \x{100}ab
+ 0: \x{100}ab
+  
+/a\x{100}*b/8
+    ab
+ 0: ab
+    a\x{100}b  
+ 0: a\x{100}b
+    a\x{100}\x{100}b  
+ 0: a\x{100}\x{100}b
+    
+/a\x{100}+b/8
+    a\x{100}b  
+ 0: a\x{100}b
+    a\x{100}\x{100}b  
+ 0: a\x{100}\x{100}b
+    *** Failers 
+No match
+    ab
+No match
+     
+/\bX/8
+    Xoanon
+ 0: X
+    +Xoanon
+ 0: X
+    \x{300}Xoanon 
+ 0: X
+    *** Failers 
+No match
+    YXoanon  
+No match
+    
+/\BX/8
+    YXoanon
+ 0: X
+    *** Failers
+No match
+    Xoanon
+No match
+    +Xoanon    
+No match
+    \x{300}Xoanon 
+No match
+
+/X\b/8
+    X+oanon
+ 0: X
+    ZX\x{300}oanon 
+ 0: X
+    FAX 
+ 0: X
+    *** Failers 
+No match
+    Xoanon  
+No match
+    
+/X\B/8
+    Xoanon  
+ 0: X
+    *** Failers
+No match
+    X+oanon
+No match
+    ZX\x{300}oanon 
+No match
+    FAX 
+No match
+    
+/[^a]/8
+    abcd
+ 0: b
+    a\x{100}   
+ 0: \x{100}
+
+/^[abc\x{123}\x{400}-\x{402}]{2,3}\d/8
+    ab99
+ 0: ab9
+    \x{123}\x{123}45
+ 0: \x{123}\x{123}4
+    \x{400}\x{401}\x{402}6  
+ 0: \x{400}\x{401}\x{402}6
+    *** Failers
+No match
+    d99
+No match
+    \x{123}\x{122}4   
+No match
+    \x{400}\x{403}6  
+No match
+    \x{400}\x{401}\x{402}\x{402}6  
+No match
+
+/a.b/8
+    acb
+ 0: acb
+    a\x7fb
+ 0: a\x{7f}b
+    a\x{100}b 
+ 0: a\x{100}b
+    *** Failers
+No match
+    a\nb  
+No match
+
+/a(.{3})b/8
+    a\x{4000}xyb 
+ 0: a\x{4000}xyb
+    a\x{4000}\x7fyb 
+ 0: a\x{4000}\x{7f}yb
+    a\x{4000}\x{100}yb 
+ 0: a\x{4000}\x{100}yb
+    *** Failers
+No match
+    a\x{4000}b 
+No match
+    ac\ncb 
+No match
+
+/a(.*?)(.)/
+    a\xc0\x88b
+ 0: a\xc0\x88b
+ 1: a\xc0\x88
+ 2: a\xc0
+
+/a(.*?)(.)/8
+    a\x{100}b
+ 0: a\x{100}b
+ 1: a\x{100}
+
+/a(.*)(.)/
+    a\xc0\x88b
+ 0: a\xc0\x88b
+ 1: a\xc0\x88
+ 2: a\xc0
+
+/a(.*)(.)/8
+    a\x{100}b
+ 0: a\x{100}b
+ 1: a\x{100}
+
+/a(.)(.)/
+    a\xc0\x92bcd
+ 0: a\xc0\x92
+
+/a(.)(.)/8
+    a\x{240}bcd
+ 0: a\x{240}b
+
+/a(.?)(.)/
+    a\xc0\x92bcd
+ 0: a\xc0\x92
+ 1: a\xc0
+
+/a(.?)(.)/8
+    a\x{240}bcd
+ 0: a\x{240}b
+ 1: a\x{240}
+
+/a(.??)(.)/
+    a\xc0\x92bcd
+ 0: a\xc0\x92
+ 1: a\xc0
+
+/a(.??)(.)/8
+    a\x{240}bcd
+ 0: a\x{240}b
+ 1: a\x{240}
+
+/a(.{3})b/8
+    a\x{1234}xyb 
+ 0: a\x{1234}xyb
+    a\x{1234}\x{4321}yb 
+ 0: a\x{1234}\x{4321}yb
+    a\x{1234}\x{4321}\x{3412}b 
+ 0: a\x{1234}\x{4321}\x{3412}b
+    *** Failers
+No match
+    a\x{1234}b 
+No match
+    ac\ncb 
+No match
+
+/a(.{3,})b/8
+    a\x{1234}xyb 
+ 0: a\x{1234}xyb
+    a\x{1234}\x{4321}yb 
+ 0: a\x{1234}\x{4321}yb
+    a\x{1234}\x{4321}\x{3412}b 
+ 0: a\x{1234}\x{4321}\x{3412}b
+    axxxxbcdefghijb 
+ 0: axxxxbcdefghijb
+ 1: axxxxb
+    a\x{1234}\x{4321}\x{3412}\x{3421}b 
+ 0: a\x{1234}\x{4321}\x{3412}\x{3421}b
+    *** Failers
+No match
+    a\x{1234}b 
+No match
+
+/a(.{3,}?)b/8
+    a\x{1234}xyb 
+ 0: a\x{1234}xyb
+    a\x{1234}\x{4321}yb 
+ 0: a\x{1234}\x{4321}yb
+    a\x{1234}\x{4321}\x{3412}b 
+ 0: a\x{1234}\x{4321}\x{3412}b
+    axxxxbcdefghijb 
+ 0: axxxxbcdefghijb
+ 1: axxxxb
+    a\x{1234}\x{4321}\x{3412}\x{3421}b 
+ 0: a\x{1234}\x{4321}\x{3412}\x{3421}b
+    *** Failers
+No match
+    a\x{1234}b 
+No match
+
+/a(.{3,5})b/8
+    a\x{1234}xyb 
+ 0: a\x{1234}xyb
+    a\x{1234}\x{4321}yb 
+ 0: a\x{1234}\x{4321}yb
+    a\x{1234}\x{4321}\x{3412}b 
+ 0: a\x{1234}\x{4321}\x{3412}b
+    axxxxbcdefghijb 
+ 0: axxxxb
+    a\x{1234}\x{4321}\x{3412}\x{3421}b 
+ 0: a\x{1234}\x{4321}\x{3412}\x{3421}b
+    axbxxbcdefghijb 
+ 0: axbxxb
+    axxxxxbcdefghijb 
+ 0: axxxxxb
+    *** Failers
+No match
+    a\x{1234}b 
+No match
+    axxxxxxbcdefghijb 
+No match
+
+/a(.{3,5}?)b/8
+    a\x{1234}xyb 
+ 0: a\x{1234}xyb
+    a\x{1234}\x{4321}yb 
+ 0: a\x{1234}\x{4321}yb
+    a\x{1234}\x{4321}\x{3412}b 
+ 0: a\x{1234}\x{4321}\x{3412}b
+    axxxxbcdefghijb 
+ 0: axxxxb
+    a\x{1234}\x{4321}\x{3412}\x{3421}b 
+ 0: a\x{1234}\x{4321}\x{3412}\x{3421}b
+    axbxxbcdefghijb 
+ 0: axbxxb
+    axxxxxbcdefghijb 
+ 0: axxxxxb
+    *** Failers
+No match
+    a\x{1234}b 
+No match
+    axxxxxxbcdefghijb 
+No match
+
+/^[a\x{c0}]/8
+    *** Failers
+No match
+    \x{100}
+No match
+
+/(?<=aXb)cd/8
+    aXbcd
+ 0: cd
+
+/(?<=a\x{100}b)cd/8
+    a\x{100}bcd
+ 0: cd
+
+/(?<=a\x{100000}b)cd/8
+    a\x{100000}bcd
+ 0: cd
+    
+/(?:\x{100}){3}b/8
+    \x{100}\x{100}\x{100}b
+ 0: \x{100}\x{100}\x{100}b
+    *** Failers 
+No match
+    \x{100}\x{100}b
+No match
+
+/\x{ab}/8
+    \x{ab} 
+ 0: \x{ab}
+    \xc2\xab
+ 0: \x{ab}
+    *** Failers 
+No match
+    \x00{ab}
+No match
+
+/(?<=(.))X/8
+    WXYZ
+ 0: X
+    \x{256}XYZ 
+ 0: X
+    *** Failers
+No match
+    XYZ 
+No match
+
+/[^a]+/8g
+    bcd
+ 0: bcd
+ 1: bc
+ 2: b
+    \x{100}aY\x{256}Z 
+ 0: \x{100}
+ 0: Y\x{256}Z
+ 1: Y\x{256}
+ 2: Y
+    
+/^[^a]{2}/8
+    \x{100}bc
+ 0: \x{100}b
+ 
+/^[^a]{2,}/8
+    \x{100}bcAa
+ 0: \x{100}bcA
+ 1: \x{100}bc
+ 2: \x{100}b
+
+/^[^a]{2,}?/8
+    \x{100}bca
+ 0: \x{100}bc
+ 1: \x{100}b
+
+/[^a]+/8ig
+    bcd
+ 0: bcd
+ 1: bc
+ 2: b
+    \x{100}aY\x{256}Z 
+ 0: \x{100}
+ 0: Y\x{256}Z
+ 1: Y\x{256}
+ 2: Y
+    
+/^[^a]{2}/8i
+    \x{100}bc
+ 0: \x{100}b
+ 
+/^[^a]{2,}/8i
+    \x{100}bcAa
+ 0: \x{100}bc
+ 1: \x{100}b
+
+/^[^a]{2,}?/8i
+    \x{100}bca
+ 0: \x{100}bc
+ 1: \x{100}b
+
+/\x{100}{0,0}/8
+    abcd
+ 0: 
+ 
+/\x{100}?/8
+    abcd
+ 0: 
+    \x{100}\x{100} 
+ 0: \x{100}
+ 1: 
+
+/\x{100}{0,3}/8 
+    \x{100}\x{100} 
+ 0: \x{100}\x{100}
+ 1: \x{100}
+ 2: 
+    \x{100}\x{100}\x{100}\x{100} 
+ 0: \x{100}\x{100}\x{100}
+ 1: \x{100}\x{100}
+ 2: \x{100}
+ 3: 
+    
+/\x{100}*/8
+    abce
+ 0: 
+    \x{100}\x{100}\x{100}\x{100} 
+ 0: \x{100}\x{100}\x{100}\x{100}
+ 1: \x{100}\x{100}\x{100}
+ 2: \x{100}\x{100}
+ 3: \x{100}
+ 4: 
+
+/\x{100}{1,1}/8
+    abcd\x{100}\x{100}\x{100}\x{100} 
+ 0: \x{100}
+
+/\x{100}{1,3}/8
+    abcd\x{100}\x{100}\x{100}\x{100} 
+ 0: \x{100}\x{100}\x{100}
+ 1: \x{100}\x{100}
+ 2: \x{100}
+
+/\x{100}+/8
+    abcd\x{100}\x{100}\x{100}\x{100} 
+ 0: \x{100}\x{100}\x{100}\x{100}
+ 1: \x{100}\x{100}\x{100}
+ 2: \x{100}\x{100}
+ 3: \x{100}
+
+/\x{100}{3}/8
+    abcd\x{100}\x{100}\x{100}XX
+ 0: \x{100}\x{100}\x{100}
+
+/\x{100}{3,5}/8
+    abcd\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}XX
+ 0: \x{100}\x{100}\x{100}\x{100}\x{100}
+ 1: \x{100}\x{100}\x{100}\x{100}
+ 2: \x{100}\x{100}\x{100}
+
+/\x{100}{3,}/8
+    abcd\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}XX
+ 0: \x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}
+ 1: \x{100}\x{100}\x{100}\x{100}\x{100}\x{100}
+ 2: \x{100}\x{100}\x{100}\x{100}\x{100}
+ 3: \x{100}\x{100}\x{100}\x{100}
+ 4: \x{100}\x{100}\x{100}
+
+/(?<=a\x{100}{2}b)X/8
+    Xyyya\x{100}\x{100}bXzzz
+ 0: X
+
+/\D*/8
+  aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+Matched, but offsets vector is too small to show all matches
+ 0: aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+ 1: aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+ 2: aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+ 3: aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+ 4: aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+ 5: aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+ 6: aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+ 7: aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+ 8: aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+ 9: aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+10: aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+11: aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+12: aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+13: aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+14: aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+15: aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+16: aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+17: aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+18: aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+19: aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+20: aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+21: aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+
+/\D*/8
+  \x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}
+Matched, but offsets vector is too small to show all matches
+ 0: \x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}
+ 1: \x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}
+ 2: \x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}
+ 3: \x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}
+ 4: \x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}
+ 5: \x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}
+ 6: \x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}
+ 7: \x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}
+ 8: \x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}
+ 9: \x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}
+10: \x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}
+11: \x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}
+12: \x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}
+13: \x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}
+14: \x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}
+15: \x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}
+16: \x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}
+17: \x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}
+18: \x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}
+19: \x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}
+20: \x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}
+21: \x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}\x{100}
+
+/\D/8
+    1X2
+ 0: X
+    1\x{100}2 
+ 0: \x{100}
+  
+/>\S/8
+    > >X Y
+ 0: >X
+    > >\x{100} Y
+ 0: >\x{100}
+  
+/\d/8
+    \x{100}3
+ 0: 3
+    
+/\s/8
+    \x{100} X
+ 0:  
+    
+/\D+/8
+    12abcd34
+ 0: abcd
+ 1: abc
+ 2: ab
+ 3: a
+    *** Failers
+ 0: *** Failers
+ 1: *** Failer
+ 2: *** Faile
+ 3: *** Fail
+ 4: *** Fai
+ 5: *** Fa
+ 6: *** F
+ 7: *** 
+ 8: ***
+ 9: **
+10: *
+    1234  
+No match
+
+/\D{2,3}/8
+    12abcd34
+ 0: abc
+ 1: ab
+    12ab34
+ 0: ab
+    *** Failers  
+ 0: ***
+ 1: **
+    1234
+No match
+    12a34  
+No match
+
+/\D{2,3}?/8
+    12abcd34
+ 0: abc
+ 1: ab
+    12ab34
+ 0: ab
+    *** Failers  
+ 0: ***
+ 1: **
+    1234
+No match
+    12a34  
+No match
+
+/\d+/8
+    12abcd34
+ 0: 12
+ 1: 1
+    *** Failers
+No match
+
+/\d{2,3}/8
+    12abcd34
+ 0: 12
+    1234abcd
+ 0: 123
+ 1: 12
+    *** Failers  
+No match
+    1.4 
+No match
+
+/\d{2,3}?/8
+    12abcd34
+ 0: 12
+    1234abcd
+ 0: 123
+ 1: 12
+    *** Failers  
+No match
+    1.4 
+No match
+
+/\S+/8
+    12abcd34
+ 0: 12abcd34
+ 1: 12abcd3
+ 2: 12abcd
+ 3: 12abc
+ 4: 12ab
+ 5: 12a
+ 6: 12
+ 7: 1
+    *** Failers
+ 0: ***
+ 1: **
+ 2: *
+    \    \ 
+No match
+
+/\S{2,3}/8
+    12abcd34
+ 0: 12a
+ 1: 12
+    1234abcd
+ 0: 123
+ 1: 12
+    *** Failers
+ 0: ***
+ 1: **
+    \     \  
+No match
+
+/\S{2,3}?/8
+    12abcd34
+ 0: 12a
+ 1: 12
+    1234abcd
+ 0: 123
+ 1: 12
+    *** Failers
+ 0: ***
+ 1: **
+    \     \  
+No match
+
+/>\s+      <34
+ 0: >      <
+    *** Failers
+No match
+
+/>\s{2,3}    <
+    ab>      <
+    *** Failers
+No match
+    ab>    \s{2,3}?    <
+    ab>      <
+    *** Failers
+No match
+    ab>    \xff<
+ 0: \xff
+
+/[\xff]/8
+    >\x{ff}<
+ 0: \x{ff}
+
+/[^\xFF]/
+    XYZ
+ 0: X
+
+/[^\xff]/8
+    XYZ
+ 0: X
+    \x{123} 
+ 0: \x{123}
+
+/^[ac]*b/8
+  xb
+No match
+
+/^[ac\x{100}]*b/8
+  xb
+No match
+
+/^[^x]*b/8i
+  xb
+No match
+
+/^[^x]*b/8
+  xb
+No match
+  
+/^\d*b/8
+  xb 
+No match
+
+/(|a)/g8
+    catac
+ 0: 
+ 0: a
+ 1: 
+ 0: 
+ 0: a
+ 1: 
+ 0: 
+ 0: 
+    a\x{256}a 
+ 0: a
+ 1: 
+ 0: 
+ 0: a
+ 1: 
+ 0: 
+
+/^\x{85}$/8i
+    \x{85}
+ 0: \x{85}
+
+/^abc./mgx8
+    abc1 \x0aabc2 \x0babc3xx \x0cabc4 \x0dabc5xx \x0d\x0aabc6 \x{0085}abc7 \x{2028}abc8 \x{2029}abc9 JUNK
+ 0: abc1
+ 0: abc2
+ 0: abc3
+ 0: abc4
+ 0: abc5
+ 0: abc6
+ 0: abc7
+ 0: abc8
+ 0: abc9
+
+/abc.$/mgx8
+    abc1\x0a abc2\x0b abc3\x0c abc4\x0d abc5\x0d\x0a abc6\x{0085} abc7\x{2028} abc8\x{2029} abc9
+ 0: abc1
+ 0: abc2
+ 0: abc3
+ 0: abc4
+ 0: abc5
+ 0: abc6
+ 0: abc7
+ 0: abc8
+ 0: abc9
+
+/^a\Rb/8
+    a\nb
+ 0: a\x{0a}b
+    a\rb
+ 0: a\x{0d}b
+    a\r\nb
+ 0: a\x{0d}\x{0a}b
+    a\x0bb
+ 0: a\x{0b}b
+    a\x0cb
+ 0: a\x{0c}b
+    a\x{85}b   
+ 0: a\x{85}b
+    a\x{2028}b 
+ 0: a\x{2028}b
+    a\x{2029}b 
+ 0: a\x{2029}b
+    ** Failers
+No match
+    a\n\rb    
+No match
+
+/^a\R*b/8
+    ab
+ 0: ab
+    a\nb
+ 0: a\x{0a}b
+    a\rb
+ 0: a\x{0d}b
+    a\r\nb
+ 0: a\x{0d}\x{0a}b
+    a\x0bb
+ 0: a\x{0b}b
+    a\x0c\x{2028}\x{2029}b
+ 0: a\x{0c}\x{2028}\x{2029}b
+    a\x{85}b   
+ 0: a\x{85}b
+    a\n\rb    
+ 0: a\x{0a}\x{0d}b
+    a\n\r\x{85}\x0cb 
+ 0: a\x{0a}\x{0d}\x{85}\x{0c}b
+
+/^a\R+b/8
+    a\nb
+ 0: a\x{0a}b
+    a\rb
+ 0: a\x{0d}b
+    a\r\nb
+ 0: a\x{0d}\x{0a}b
+    a\x0bb
+ 0: a\x{0b}b
+    a\x0c\x{2028}\x{2029}b
+ 0: a\x{0c}\x{2028}\x{2029}b
+    a\x{85}b   
+ 0: a\x{85}b
+    a\n\rb    
+ 0: a\x{0a}\x{0d}b
+    a\n\r\x{85}\x0cb 
+ 0: a\x{0a}\x{0d}\x{85}\x{0c}b
+    ** Failers
+No match
+    ab  
+No match
+
+/^a\R{1,3}b/8
+    a\nb
+ 0: a\x{0a}b
+    a\n\rb
+ 0: a\x{0a}\x{0d}b
+    a\n\r\x{85}b
+ 0: a\x{0a}\x{0d}\x{85}b
+    a\r\n\r\nb 
+ 0: a\x{0d}\x{0a}\x{0d}\x{0a}b
+    a\r\n\r\n\r\nb 
+ 0: a\x{0d}\x{0a}\x{0d}\x{0a}\x{0d}\x{0a}b
+    a\n\r\n\rb
+ 0: a\x{0a}\x{0d}\x{0a}\x{0d}b
+    a\n\n\r\nb 
+ 0: a\x{0a}\x{0a}\x{0d}\x{0a}b
+    ** Failers
+No match
+    a\n\n\n\rb
+No match
+    a\r
+No match
+
+/\h+\V?\v{3,4}/8 
+    \x09\x20\x{a0}X\x0a\x0b\x0c\x0d\x0a
+ 0: \x{09} \x{a0}X\x{0a}\x{0b}\x{0c}\x{0d}
+ 1: \x{09} \x{a0}X\x{0a}\x{0b}\x{0c}
+
+/\V?\v{3,4}/8 
+    \x20\x{a0}X\x0a\x0b\x0c\x0d\x0a
+ 0: X\x{0a}\x{0b}\x{0c}\x{0d}
+ 1: X\x{0a}\x{0b}\x{0c}
+
+/\h+\V?\v{3,4}/8
+    >\x09\x20\x{a0}X\x0a\x0a\x0a<
+ 0: \x{09} \x{a0}X\x{0a}\x{0a}\x{0a}
+
+/\V?\v{3,4}/8
+    >\x09\x20\x{a0}X\x0a\x0a\x0a<
+ 0: X\x{0a}\x{0a}\x{0a}
+
+/\H\h\V\v/8
+    X X\x0a
+ 0: X X\x{0a}
+    X\x09X\x0b
+ 0: X\x{09}X\x{0b}
+    ** Failers
+No match
+    \x{a0} X\x0a   
+No match
+    
+/\H*\h+\V?\v{3,4}/8 
+    \x09\x20\x{a0}X\x0a\x0b\x0c\x0d\x0a
+ 0: \x{09} \x{a0}X\x{0a}\x{0b}\x{0c}\x{0d}
+ 1: \x{09} \x{a0}X\x{0a}\x{0b}\x{0c}
+    \x09\x20\x{a0}\x0a\x0b\x0c\x0d\x0a
+ 0: \x{09} \x{a0}\x{0a}\x{0b}\x{0c}\x{0d}
+ 1: \x{09} \x{a0}\x{0a}\x{0b}\x{0c}
+    \x09\x20\x{a0}\x0a\x0b\x0c
+ 0: \x{09} \x{a0}\x{0a}\x{0b}\x{0c}
+    ** Failers 
+No match
+    \x09\x20\x{a0}\x0a\x0b
+No match
+     
+/\H\h\V\v/8
+    \x{3001}\x{3000}\x{2030}\x{2028}
+ 0: \x{3001}\x{3000}\x{2030}\x{2028}
+    X\x{180e}X\x{85}
+ 0: X\x{180e}X\x{85}
+    ** Failers
+No match
+    \x{2009} X\x0a   
+No match
+    
+/\H*\h+\V?\v{3,4}/8 
+    \x{1680}\x{180e}\x{2007}X\x{2028}\x{2029}\x0c\x0d\x0a
+ 0: \x{1680}\x{180e}\x{2007}X\x{2028}\x{2029}\x{0c}\x{0d}
+ 1: \x{1680}\x{180e}\x{2007}X\x{2028}\x{2029}\x{0c}
+    \x09\x{205f}\x{a0}\x0a\x{2029}\x0c\x{2028}\x0a
+ 0: \x{09}\x{205f}\x{a0}\x{0a}\x{2029}\x{0c}\x{2028}
+ 1: \x{09}\x{205f}\x{a0}\x{0a}\x{2029}\x{0c}
+    \x09\x20\x{202f}\x0a\x0b\x0c
+ 0: \x{09} \x{202f}\x{0a}\x{0b}\x{0c}
+    ** Failers 
+No match
+    \x09\x{200a}\x{a0}\x{2028}\x0b
+No match
+     
+/a\Rb/I8
+Capturing subpattern count = 0
+Options: bsr_anycrlf utf
+First char = 'a'
+Need char = 'b'
+    a\rb
+ 0: a\x{0d}b
+    a\nb
+ 0: a\x{0a}b
+    a\r\nb
+ 0: a\x{0d}\x{0a}b
+    ** Failers
+No match
+    a\x{85}b
+No match
+    a\x0bb     
+No match
+
+/a\Rb/I8
+Capturing subpattern count = 0
+Options: bsr_unicode utf
+First char = 'a'
+Need char = 'b'
+    a\rb
+ 0: a\x{0d}b
+    a\nb
+ 0: a\x{0a}b
+    a\r\nb
+ 0: a\x{0d}\x{0a}b
+    a\x{85}b
+ 0: a\x{85}b
+    a\x0bb     
+ 0: a\x{0b}b
+    ** Failers 
+No match
+    a\x{85}b\
+No match
+    a\x0bb\
+No match
+    
+/a\R?b/I8
+Capturing subpattern count = 0
+Options: bsr_anycrlf utf
+First char = 'a'
+Need char = 'b'
+    a\rb
+ 0: a\x{0d}b
+    a\nb
+ 0: a\x{0a}b
+    a\r\nb
+ 0: a\x{0d}\x{0a}b
+    ** Failers
+No match
+    a\x{85}b
+No match
+    a\x0bb     
+No match
+
+/a\R?b/I8
+Capturing subpattern count = 0
+Options: bsr_unicode utf
+First char = 'a'
+Need char = 'b'
+    a\rb
+ 0: a\x{0d}b
+    a\nb
+ 0: a\x{0a}b
+    a\r\nb
+ 0: a\x{0d}\x{0a}b
+    a\x{85}b
+ 0: a\x{85}b
+    a\x0bb     
+ 0: a\x{0b}b
+    ** Failers 
+No match
+    a\x{85}b\
+No match
+    a\x0bb\
+No match
+ 
+/X/8f 
+    A\x{1ec5}ABCXYZ
+ 0: X
+
+/abcd*/8
+    xxxxabcd\P
+ 0: abcd
+ 1: abc
+    xxxxabcd\P\P
+Partial match: abcd
+
+/abcd*/i8
+    xxxxabcd\P
+ 0: abcd
+ 1: abc
+    xxxxabcd\P\P
+Partial match: abcd
+    XXXXABCD\P
+ 0: ABCD
+ 1: ABC
+    XXXXABCD\P\P
+Partial match: ABCD
+
+/abc\d*/8
+    xxxxabc1\P
+ 0: abc1
+ 1: abc
+    xxxxabc1\P\P
+Partial match: abc1
+
+/abc[de]*/8
+    xxxxabcde\P
+ 0: abcde
+ 1: abcd
+ 2: abc
+    xxxxabcde\P\P
+Partial match: abcde
+
+/\bthe cat\b/8
+    the cat\P
+ 0: the cat
+    the cat\P\P
+Partial match: the cat
+
+/ab\Cde/8
+    abXde
+Error -16 (item unsupported for DFA matching)
+
+/(?<=ab\Cde)X/8
+Failed: \C not allowed in lookbehind assertion at offset 10
+
+/./8
+    \r\P
+ 0: \x{0d}
+    \r\P\P 
+Partial match: \x{0d}
+  
+/.{2,3}/8
+    \r\P 
+Partial match: \x{0d}
+    \r\P\P
+Partial match: \x{0d}
+    \r\r\P
+ 0: \x{0d}\x{0d}
+    \r\r\P\P
+Partial match: \x{0d}\x{0d}
+    \r\r\r\P
+ 0: \x{0d}\x{0d}\x{0d}
+ 1: \x{0d}\x{0d}
+    \r\r\r\P\P     
+Partial match: \x{0d}\x{0d}\x{0d}
+
+/.{2,3}?/8
+    \r\P 
+Partial match: \x{0d}
+    \r\P\P
+Partial match: \x{0d}
+    \r\r\P
+ 0: \x{0d}\x{0d}
+    \r\r\P\P
+Partial match: \x{0d}\x{0d}
+    \r\r\r\P
+ 0: \x{0d}\x{0d}\x{0d}
+ 1: \x{0d}\x{0d}
+    \r\r\r\P\P     
+Partial match: \x{0d}\x{0d}\x{0d}
+
+/[^\x{100}]/8
+    \x{100}\x{101}X
+ 0: \x{101}
+
+/[^\x{100}]+/8
+    \x{100}\x{101}X
+ 0: \x{101}X
+ 1: \x{101}
+
+/-- End of testinput9 --/ 
diff --git a/pcre/testdata/testoutputEBC b/pcre/testdata/testoutputEBC
new file mode 100644
index 00000000000..abbfdc43fc7
--- /dev/null
+++ b/pcre/testdata/testoutputEBC
@@ -0,0 +1,182 @@
+/-- This is a specialized test for checking, when PCRE is compiled with the
+EBCDIC option but in an ASCII environment, that newline and white space
+functionality is working. It catches cases where explicit values such as 0x0a
+have been used instead of names like CHAR_LF. Needless to say, it is not a
+genuine EBCDIC test! In patterns, alphabetic characters that follow a backslash
+must be in EBCDIC code. In data, newlines and other spacing characters must be
+in EBCDIC, but can be specified as escapes. --/
+
+/-- Test default newline and variations --/
+
+/^A/m
+    ABC
+ 0: A
+    12\x15ABC
+ 0: A
+
+/^A/m
+    12\x15ABC
+ 0: A
+    12\x0dABC
+ 0: A
+    12\x0d\x15ABC
+ 0: A
+    12\x25ABC
+ 0: A
+
+/^A/m
+    12\x15ABC
+ 0: A
+    12\x0dABC
+ 0: A
+    12\x0d\x15ABC
+ 0: A
+    ** Fail
+No match
+    12\x25ABC
+No match
+
+/-- Test \h --/
+
+/^A\/
+    A B
+ 0: A\x20
+
+/-- Test \H --/
+
+/^A\/
+    AB
+ 0: AB
+    ** Fail
+No match
+    A B
+No match
+
+/-- Test \R --/
+
+/^A\/
+    A\x15B
+ 0: A\x15
+    A\x0dB
+ 0: A\x0d
+    A\x25B
+ 0: A\x25
+    A\x0bB
+ 0: A\x0b
+    A\x0cB
+ 0: A\x0c
+    ** Fail
+No match
+    A B
+No match
+
+/-- Test \v --/
+
+/^A\/
+    A\x15B
+ 0: A\x15
+    A\x0dB
+ 0: A\x0d
+    A\x25B
+ 0: A\x25
+    A\x0bB
+ 0: A\x0b
+    A\x0cB
+ 0: A\x0c
+    ** Fail
+No match
+    A B
+No match
+
+/-- Test \V --/
+
+/^A\/
+    A B
+ 0: A\x20
+    ** Fail
+No match
+    A\x15B
+No match
+    A\x0dB
+No match
+    A\x25B
+No match
+    A\x0bB
+No match
+    A\x0cB
+No match
+    
+/-- For repeated items, use an atomic group so that the output is the same
+for DFA matching (otherwise it may show multiple matches). --/ 
+
+/-- Test \h+ --/
+
+/^A(?>\+)/
+    A B
+ 0: A\x20
+
+/-- Test \H+ --/
+
+/^A(?>\+)/
+    AB
+ 0: AB
+    ** Fail
+No match
+    A B
+No match
+
+/-- Test \R+ --/
+
+/^A(?>\+)/
+    A\x15B
+ 0: A\x15
+    A\x0dB
+ 0: A\x0d
+    A\x25B
+ 0: A\x25
+    A\x0bB
+ 0: A\x0b
+    A\x0cB
+ 0: A\x0c
+    ** Fail
+No match
+    A B
+No match
+
+/-- Test \v+ --/
+
+/^A(?>\+)/
+    A\x15B
+ 0: A\x15
+    A\x0dB
+ 0: A\x0d
+    A\x25B
+ 0: A\x25
+    A\x0bB
+ 0: A\x0b
+    A\x0cB
+ 0: A\x0c
+    ** Fail
+No match
+    A B
+No match
+
+/-- Test \V+ --/
+
+/^A(?>\+)/
+    A B
+ 0: A\x20B
+    ** Fail
+No match
+    A\x15B
+No match
+    A\x0dB
+No match
+    A\x25B
+No match
+    A\x0bB
+No match
+    A\x0cB
+No match
+
+/-- End --/
diff --git a/pcre/testdata/wintestinput3 b/pcre/testdata/wintestinput3
new file mode 100644
index 00000000000..04e76a6dc85
--- /dev/null
+++ b/pcre/testdata/wintestinput3
@@ -0,0 +1,91 @@
+/^[\w]+/
+    *** Failers
+    cole
+
+/^[\w]+/Lfrench
+    cole
+
+/^[\w]+/
+    *** Failers
+    cole
+
+/^[\W]+/
+    cole
+
+/^[\W]+/Lfrench
+    *** Failers
+    cole
+
+/[\b]/
+    \b
+    *** Failers
+    a
+
+/[\b]/Lfrench
+    \b
+    *** Failers
+    a
+
+/^\w+/
+    *** Failers
+    cole
+
+/^\w+/Lfrench
+    cole
+
+/(.+)\b(.+)/
+    cole
+
+/(.+)\b(.+)/Lfrench
+    *** Failers
+    cole
+
+/cole/i
+    cole
+    *** Failers
+    cole
+
+/cole/iLfrench
+    cole
+    cole
+
+/\w/IS
+
+/\w/ISLfrench
+
+/^[\xc8-\xc9]/iLfrench
+    cole
+    cole
+
+/^[\xc8-\xc9]/Lfrench
+    cole
+    *** Failers 
+    cole
+
+/\W+/Lfrench
+    >>>\xaa<<<
+    >>>\xba<<< 
+
+/[\W]+/Lfrench
+    >>>\xaa<<<
+    >>>\xba<<< 
+
+/[^[:alpha:]]+/Lfrench
+    >>>\xaa<<<
+    >>>\xba<<< 
+
+/\w+/Lfrench
+    >>>\xaa<<<
+    >>>\xba<<< 
+
+/[\w]+/Lfrench
+    >>>\xaa<<<
+    >>>\xba<<< 
+
+/[[:alpha:]]+/Lfrench
+    >>>\xaa<<<
+    >>>\xba<<< 
+    
+/[[:alpha:]][[:lower:]][[:upper:]]/DZLfrench 
+
+/ End of testinput3 /
diff --git a/pcre/testdata/wintestoutput3 b/pcre/testdata/wintestoutput3
new file mode 100644
index 00000000000..00880070670
--- /dev/null
+++ b/pcre/testdata/wintestoutput3
@@ -0,0 +1,166 @@
+/^[\w]+/
+    *** Failers
+No match
+    cole
+No match
+
+/^[\w]+/Lfrench
+    cole
+ 0: cole
+
+/^[\w]+/
+    *** Failers
+No match
+    cole
+No match
+
+/^[\W]+/
+    cole
+ 0: \xc9
+
+/^[\W]+/Lfrench
+    *** Failers
+ 0: *** 
+    cole
+No match
+
+/[\b]/
+    \b
+ 0: \x08
+    *** Failers
+No match
+    a
+No match
+
+/[\b]/Lfrench
+    \b
+ 0: \x08
+    *** Failers
+No match
+    a
+No match
+
+/^\w+/
+    *** Failers
+No match
+    cole
+No match
+
+/^\w+/Lfrench
+    cole
+ 0: cole
+
+/(.+)\b(.+)/
+    cole
+ 0: \xc9cole
+ 1: \xc9
+ 2: cole
+
+/(.+)\b(.+)/Lfrench
+    *** Failers
+ 0: *** Failers
+ 1: *** 
+ 2: Failers
+    cole
+No match
+
+/cole/i
+    cole
+ 0: \xc9cole
+    *** Failers
+No match
+    cole
+No match
+
+/cole/iLfrench
+    cole
+ 0: cole
+    cole
+ 0: cole
+
+/\w/IS
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: 0 1 2 3 4 5 6 7 8 9 A B C D E F G H I J K L M N O P 
+  Q R S T U V W X Y Z _ a b c d e f g h i j k l m n o p q r s t u v w x y z 
+
+/\w/ISLfrench
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+Subject length lower bound = 1
+Starting byte set: 0 1 2 3 4 5 6 7 8 9 A B C D E F G H I J K L M N O P 
+  Q R S T U V W X Y Z _ a b c d e f g h i j k l m n o p q r s t u v w x y z 
+                                       
+                                       
+    
+
+/^[\xc8-\xc9]/iLfrench
+    cole
+ 0: 
+    cole
+ 0: 
+
+/^[\xc8-\xc9]/Lfrench
+    cole
+ 0: 
+    *** Failers 
+No match
+    cole
+No match
+
+/\W+/Lfrench
+    >>>\xaa<<<
+ 0: >>>
+    >>>\xba<<< 
+ 0: >>>
+
+/[\W]+/Lfrench
+    >>>\xaa<<<
+ 0: >>>
+    >>>\xba<<< 
+ 0: >>>
+
+/[^[:alpha:]]+/Lfrench
+    >>>\xaa<<<
+ 0: >>>
+    >>>\xba<<< 
+ 0: >>>
+
+/\w+/Lfrench
+    >>>\xaa<<<
+ 0: 
+    >>>\xba<<< 
+ 0: 
+
+/[\w]+/Lfrench
+    >>>\xaa<<<
+ 0: 
+    >>>\xba<<< 
+ 0: 
+
+/[[:alpha:]]+/Lfrench
+    >>>\xaa<<<
+ 0: 
+    >>>\xba<<< 
+ 0: 
+    
+/[[:alpha:]][[:lower:]][[:upper:]]/DZLfrench 
+------------------------------------------------------------------
+        Bra
+        [A-Za-z\x83\x8a\x8c\x8e\x9a\x9c\x9e\x9f\xaa\xb5\xba\xc0-\xd6\xd8-\xf6\xf8-\xff]
+        [a-z\x83\x9a\x9c\x9e\xaa\xb5\xba\xdf-\xf6\xf8-\xff]
+        [A-Z\x8a\x8c\x8e\x9f\xc0-\xd6\xd8-\xde]
+        Ket
+        End
+------------------------------------------------------------------
+Capturing subpattern count = 0
+No options
+No first char
+No need char
+
+/ End of testinput3 /
diff --git a/pcre/ucp.h b/pcre/ucp.h
new file mode 100644
index 00000000000..21039106e58
--- /dev/null
+++ b/pcre/ucp.h
@@ -0,0 +1,197 @@
+/*************************************************
+*          Unicode Property Table handler        *
+*************************************************/
+
+#ifndef _UCP_H
+#define _UCP_H
+
+/* This file contains definitions of the property values that are returned by
+the UCD access macros. New values that are added for new releases of Unicode
+should always be at the end of each enum, for backwards compatibility.
+
+IMPORTANT: Note also that the specific numeric values of the enums have to be
+the same as the values that are generated by the maint/MultiStage2.py script,
+where the equivalent property descriptive names are listed in vectors. */
+
+/* These are the general character categories. */
+
+enum {
+  ucp_C,     /* Other */
+  ucp_L,     /* Letter */
+  ucp_M,     /* Mark */
+  ucp_N,     /* Number */
+  ucp_P,     /* Punctuation */
+  ucp_S,     /* Symbol */
+  ucp_Z      /* Separator */
+};
+
+/* These are the particular character categories. */
+
+enum {
+  ucp_Cc,    /* Control */
+  ucp_Cf,    /* Format */
+  ucp_Cn,    /* Unassigned */
+  ucp_Co,    /* Private use */
+  ucp_Cs,    /* Surrogate */
+  ucp_Ll,    /* Lower case letter */
+  ucp_Lm,    /* Modifier letter */
+  ucp_Lo,    /* Other letter */
+  ucp_Lt,    /* Title case letter */
+  ucp_Lu,    /* Upper case letter */
+  ucp_Mc,    /* Spacing mark */
+  ucp_Me,    /* Enclosing mark */
+  ucp_Mn,    /* Non-spacing mark */
+  ucp_Nd,    /* Decimal number */
+  ucp_Nl,    /* Letter number */
+  ucp_No,    /* Other number */
+  ucp_Pc,    /* Connector punctuation */
+  ucp_Pd,    /* Dash punctuation */
+  ucp_Pe,    /* Close punctuation */
+  ucp_Pf,    /* Final punctuation */
+  ucp_Pi,    /* Initial punctuation */
+  ucp_Po,    /* Other punctuation */
+  ucp_Ps,    /* Open punctuation */
+  ucp_Sc,    /* Currency symbol */
+  ucp_Sk,    /* Modifier symbol */
+  ucp_Sm,    /* Mathematical symbol */
+  ucp_So,    /* Other symbol */
+  ucp_Zl,    /* Line separator */
+  ucp_Zp,    /* Paragraph separator */
+  ucp_Zs     /* Space separator */
+};
+
+/* These are grapheme break properties. Note that the code for processing them
+assumes that the values are less than 16. If more values are added that take
+the number to 16 or more, the code will have to be rewritten. */
+
+enum {
+  ucp_gbCR,                /*  0 */
+  ucp_gbLF,                /*  1 */
+  ucp_gbControl,           /*  2 */
+  ucp_gbExtend,            /*  3 */
+  ucp_gbPrepend,           /*  4 */
+  ucp_gbSpacingMark,       /*  5 */
+  ucp_gbL,                 /*  6 Hangul syllable type L */
+  ucp_gbV,                 /*  7 Hangul syllable type V */
+  ucp_gbT,                 /*  8 Hangul syllable type T */
+  ucp_gbLV,                /*  9 Hangul syllable type LV */
+  ucp_gbLVT,               /* 10 Hangul syllable type LVT */
+  ucp_gbRegionalIndicator, /* 11 */
+  ucp_gbOther              /* 12 */
+};
+
+/* These are the script identifications. */
+
+enum {
+  ucp_Arabic,
+  ucp_Armenian,
+  ucp_Bengali,
+  ucp_Bopomofo,
+  ucp_Braille,
+  ucp_Buginese,
+  ucp_Buhid,
+  ucp_Canadian_Aboriginal,
+  ucp_Cherokee,
+  ucp_Common,
+  ucp_Coptic,
+  ucp_Cypriot,
+  ucp_Cyrillic,
+  ucp_Deseret,
+  ucp_Devanagari,
+  ucp_Ethiopic,
+  ucp_Georgian,
+  ucp_Glagolitic,
+  ucp_Gothic,
+  ucp_Greek,
+  ucp_Gujarati,
+  ucp_Gurmukhi,
+  ucp_Han,
+  ucp_Hangul,
+  ucp_Hanunoo,
+  ucp_Hebrew,
+  ucp_Hiragana,
+  ucp_Inherited,
+  ucp_Kannada,
+  ucp_Katakana,
+  ucp_Kharoshthi,
+  ucp_Khmer,
+  ucp_Lao,
+  ucp_Latin,
+  ucp_Limbu,
+  ucp_Linear_B,
+  ucp_Malayalam,
+  ucp_Mongolian,
+  ucp_Myanmar,
+  ucp_New_Tai_Lue,
+  ucp_Ogham,
+  ucp_Old_Italic,
+  ucp_Old_Persian,
+  ucp_Oriya,
+  ucp_Osmanya,
+  ucp_Runic,
+  ucp_Shavian,
+  ucp_Sinhala,
+  ucp_Syloti_Nagri,
+  ucp_Syriac,
+  ucp_Tagalog,
+  ucp_Tagbanwa,
+  ucp_Tai_Le,
+  ucp_Tamil,
+  ucp_Telugu,
+  ucp_Thaana,
+  ucp_Thai,
+  ucp_Tibetan,
+  ucp_Tifinagh,
+  ucp_Ugaritic,
+  ucp_Yi,
+  /* New for Unicode 5.0: */
+  ucp_Balinese,
+  ucp_Cuneiform,
+  ucp_Nko,
+  ucp_Phags_Pa,
+  ucp_Phoenician,
+  /* New for Unicode 5.1: */
+  ucp_Carian,
+  ucp_Cham,
+  ucp_Kayah_Li,
+  ucp_Lepcha,
+  ucp_Lycian,
+  ucp_Lydian,
+  ucp_Ol_Chiki,
+  ucp_Rejang,
+  ucp_Saurashtra,
+  ucp_Sundanese,
+  ucp_Vai,
+  /* New for Unicode 5.2: */
+  ucp_Avestan,
+  ucp_Bamum,
+  ucp_Egyptian_Hieroglyphs,
+  ucp_Imperial_Aramaic,
+  ucp_Inscriptional_Pahlavi,
+  ucp_Inscriptional_Parthian,
+  ucp_Javanese,
+  ucp_Kaithi,
+  ucp_Lisu,
+  ucp_Meetei_Mayek,
+  ucp_Old_South_Arabian,
+  ucp_Old_Turkic,
+  ucp_Samaritan,
+  ucp_Tai_Tham,
+  ucp_Tai_Viet,
+  /* New for Unicode 6.0.0: */
+  ucp_Batak,
+  ucp_Brahmi,
+  ucp_Mandaic,
+  /* New for Unicode 6.1.0: */
+  ucp_Chakma,
+  ucp_Meroitic_Cursive,
+  ucp_Meroitic_Hieroglyphs,
+  ucp_Miao,
+  ucp_Sharada,
+  ucp_Sora_Sompeng,
+  ucp_Takri
+};
+
+#endif
+
+/* End of ucp.h */
-- 
cgit v1.2.1