diff options
author | Glenn Randers-Pehrson <glennrp at users.sourceforge.net> | 2015-03-26 08:06:25 -0500 |
---|---|---|
committer | Glenn Randers-Pehrson <glennrp at users.sourceforge.net> | 2015-03-26 08:11:22 -0500 |
commit | 2b667e49232dd36aa7ea3a7382f265fd7f97d2cd (patch) | |
tree | 14913ecdb249da47133d5df68904994b2790a1d1 | |
parent | 070a616b8275277e18ef8ee91e2ca23f7bdc67d5 (diff) | |
download | libpng-2b667e49232dd36aa7ea3a7382f265fd7f97d2cd.tar.gz |
[master] Imported from libpng-1.6.17.tarv1.6.17-masterlibpng-1.6.17-master-signed
54 files changed, 1499 insertions, 703 deletions
@@ -1,4 +1,4 @@ -Libpng 1.6.16 - December 22, 2014 +Libpng 1.6.17 - March 26, 2015 This is a public release of libpng, intended for use in production codes. @@ -7,30 +7,61 @@ Files available for download: Source files with LF line endings (for Unix/Linux) and with a "configure" script - libpng-1.6.16.tar.xz (LZMA-compressed, recommended) - libpng-1.6.16.tar.gz + libpng-1.6.17.tar.xz (LZMA-compressed, recommended) + libpng-1.6.17.tar.gz Source files with CRLF line endings (for Windows), without the "configure" script - lpng1616.7z (LZMA-compressed, recommended) - lpng1616.zip + lpng1617.7z (LZMA-compressed, recommended) + lpng1617.zip Other information: - libpng-1.6.16-README.txt - libpng-1.6.16-LICENSE.txt - libpng-1.6.16-*.asc (armored detached GPG signatures) - -Changes since the last public release (1.6.15): - Added ".align 2" to arm/filter_neon.S to support old GAS assemblers that - don't do alignment correctly. - Revised Makefile.am and scripts/*.dfn to work with MinGW/MSYS; - renamed scripts/*.dfn to scripts/*.c (Bob Friesenhahn and John Bowler). - Quiet a "comparison always true" warning in pngstest.c (John Bowler). - Restored a test on width that was removed from png.c at libpng-1.6.9 - (Bug report by Alex Eubanks). - Fixed an overflow in png_combine_row with very wide interlaced images. + libpng-1.6.17-README.txt + libpng-1.6.17-LICENSE.txt + libpng-1.6.17-*.asc (armored detached GPG signatures) + +Changes since the last public release (1.6.16): + + Removed duplicate PNG_SAFE_LIMITS_SUPPORTED handling from pngconf.h + Corrected the width limit calculation in png_check_IHDR(). + Removed user limits from pngfix. Also pass NULL pointers to + png_read_row to skip the unnecessary row de-interlace stuff. + Added testing of png_set_packing() to pngvalid.c + Regenerated configure scripts in the *.tar distributions with libtool-2.4.4 + Implement previously untested cases of libpng transforms in pngvalid.c + Fixed byte order in 2-byte filler, in png_do_read_filler(). + Made the check for out-of-range values in png_set_tRNS() detect + values that are exactly 2^bit_depth, and work on 16-bit platforms. + Merged some parts of libpng-1.6.17beta01 and libpng-1.7.0beta47. + Added #ifndef __COVERITY__ where needed in png.c, pngrutil.c and + pngset.c to avoid warnings about dead code. + Do not build png_product2() when it is unused. + Display user limits in the output from pngtest. + Eliminated the PNG_SAFE_LIMITS macro and restored the 1-million-column + and 1-million-row default limits in pnglibconf.dfa, that can be reset + by the user at build time or run time. This provides a more robust + defense against DOS and as-yet undiscovered overflows. + Added PNG_WRITE_CUSTOMIZE_COMPRESSION_SUPPORTED macro, on by default. + Allow user to call png_get_IHDR() with NULL arguments (Reuben Hawkins). + Rebuilt configure scripts with automake-1.15 and libtool-2.4.6 + Moved png_set_filter() prototype into a PNG_WRITE_SUPPORTED block + of png.h. + Avoid runtime checks when converting integer to png_byte with + Visual Studio (Sergey Kosarevsky) + Removed some comments that the configure script did not handle + properly from scripts/pnglibconf.dfa and pnglibconf.h.prebuilt. + Free the unknown_chunks structure even when it contains no data. + Updated CMakeLists.txt to add OSX framework, change YES/NO to ON/OFF + for consistency, and remove some useless tests (Alexey Petruchik). + Remove pnglibconf.h, pnglibconf.c, pnglibconf.pre, pnglibconf.dfn, + and pnglibconf.out instead of pnglibconf.* in "make clean" (Cosmin). + Fixed simplified 8-bit-linear to sRGB alpha. The calculated alpha + value was wrong. It's not clear if this affected the final stored + value; in the obvious code path the upper and lower 8-bits of the + alpha value were identical and the alpha was truncated to 8-bits + rather than dividing by 257 (John Bowler). Send comments/corrections/commendations to png-mng-implement at lists.sf.net (subscription required; visit @@ -2102,7 +2102,7 @@ Version 1.4.0beta24 [July 25, 2008] png_decompress_chunk(), and remove "chunkdata" from parameter list. Put a call to png_check_chunk_name() in png_read_chunk_header(). Revised png_check_chunk_name() to reject a name with a lowercase 3rd byte. - Removed two calls to png_check_chunk_name() occuring later in the process. + Removed two calls to png_check_chunk_name() occurring later in the process. Define PNG_NO_ERROR_NUMBERS by default in pngconf.h Version 1.4.0beta25 [July 30, 2008] @@ -5114,17 +5114,97 @@ Version 1.6.16beta03 [December 21, 2014] Version 1.6.16rc01 [December 21, 2014] Restored a test on width that was removed from png.c at libpng-1.6.9 - (Bug report by Alex Eubanks). + (Bug report by Alex Eubanks, CVE-2015-0973). Version 1.6.16rc02 [December 21, 2014] Undid the update to pngrutil.c in 1.6.16rc01. Version 1.6.16rc03 [December 21, 2014] - Fixed an overflow in png_combine_row with very wide interlaced images. + Fixed an overflow in png_combine_row() with very wide interlaced images + (Bug report and fix by John Bowler, CVE-2014-9495). Version 1.6.16 [December 22, 2014] No changes. +Version 1.6.17beta01 [January 29, 2015] + Removed duplicate PNG_SAFE_LIMITS_SUPPORTED handling from pngconf.h + Corrected the width limit calculation in png_check_IHDR(). + Removed user limits from pngfix. Also pass NULL pointers to + png_read_row to skip the unnecessary row de-interlace stuff. + Added testing of png_set_packing() to pngvalid.c + Regenerated configure scripts in the *.tar distributions with libtool-2.4.4 + Implement previously untested cases of libpng transforms in pngvalid.c + Fixed byte order in 2-byte filler, in png_do_read_filler(). + Made the check for out-of-range values in png_set_tRNS() detect + values that are exactly 2^bit_depth, and work on 16-bit platforms. + Merged some parts of libpng-1.6.17beta01 and libpng-1.7.0beta47. + Added #ifndef __COVERITY__ where needed in png.c, pngrutil.c and + pngset.c to avoid warnings about dead code. + Added "& 0xff" to many instances of expressions that are typecast + to (png_byte), to avoid Coverity gripes. + +Version 1.6.17beta02 [February 7, 2015] + Work around one more Coverity-scan dead-code warning. + Do not build png_product2() when it is unused. + +Version 1.6.17beta03 [February 17, 2015] + Display user limits in the output from pngtest. + Eliminated the PNG_SAFE_LIMITS macro and restored the 1-million-column + and 1-million-row default limits in pnglibconf.dfa, that can be reset + by the user at build time or run time. This provides a more robust + defense against DOS and as-yet undiscovered overflows. + +Version 1.6.17beta04 [February 21, 2015] + Added PNG_WRITE_CUSTOMIZE_COMPRESSION_SUPPORTED macro, on by default. + Allow user to call png_get_IHDR() with NULL arguments (Reuben Hawkins). + Rebuilt configure scripts with automake-1.15 and libtool-2.4.6 + +Version 1.6.17beta05 [February 25, 2015] + Restored compiling of png_reciprocal2 with PNG_NO_16BIT. + +Version 1.6.17beta06 [February 27, 2015] + Moved png_set_filter() prototype into a PNG_WRITE_SUPPORTED block + of png.h. + Avoid runtime checks when converting integer to png_byte with + Visual Studio (Sergey Kosarevsky) + +Version 1.6.17rc01 [March 4, 2015] + No changes. + +Version 1.6.17rc02 [March 9, 2015] + Removed some comments that the configure script did not handle + properly from scripts/pnglibconf.dfa and pnglibconf.h.prebuilt. + Free the unknown_chunks structure even when it contains no data. + +Version 1.6.17rc03 [March 12, 2015] + Updated CMakeLists.txt to add OSX framework, change YES/NO to ON/OFF + for consistency, and remove some useless tests (Alexey Petruchik). + +Version 1.6.17rc04 [March 16, 2015] + Remove pnglibconf.h, pnglibconf.c, and pnglibconf.out instead of + pnglibconf.* in "make clean" (Cosmin). + Fix bug in calculation of maxbits, in png_write_sBIT, introduced + in libpng-1.6.17beta01 (John Bowler). + +Version 1.6.17rc05 [March 21, 2015] + Define PNG_FILTER_* and PNG_FILTER_VALUE_* in png.h even when WRITE + is not supported (John Bowler). This fixes an error introduced in + libpng-1.6.17beta06. + Reverted "& 0xff" additions of version 1.6.17beta01. Libpng passes + the Coverity scan without them. + +Version 1.6.17rc06 [March 23, 2015] + Remove pnglibconf.dfn and pnglibconf.pre with "make clean". + Reformatted some "&0xff" instances to "& 0xff". + Fixed simplified 8-bit-linear to sRGB alpha. The calculated alpha + value was wrong. It's not clear if this affected the final stored + value; in the obvious code path the upper and lower 8-bits of the + alpha value were identical and the alpha was truncated to 8-bits + rather than dividing by 257 (John Bowler). + +Version 1.6.17 [March 26, 2015] + No changes. + Send comments/corrections/commendations to png-mng-implement at lists.sf.net (subscription required; visit https://lists.sourceforge.net/lists/listinfo/png-mng-implement diff --git a/CMakeLists.txt b/CMakeLists.txt index 28d1ca7db..f062a23a1 100644 --- a/CMakeLists.txt +++ b/CMakeLists.txt @@ -1,6 +1,6 @@ # CMakeLists.txt -# Copyright (C) 2007-2014 Glenn Randers-Pehrson +# Copyright (C) 2007-2015 Glenn Randers-Pehrson # This code is released under the libpng license. # For conditions of distribution and use, see the disclaimer @@ -16,7 +16,7 @@ enable_testing() set(PNGLIB_MAJOR 1) set(PNGLIB_MINOR 6) -set(PNGLIB_RELEASE 16) +set(PNGLIB_RELEASE 17) set(PNGLIB_NAME libpng${PNGLIB_MAJOR}${PNGLIB_MINOR}) set(PNGLIB_VERSION ${PNGLIB_MAJOR}.${PNGLIB_MINOR}.${PNGLIB_RELEASE}) @@ -30,8 +30,7 @@ if(NOT WIN32) PATHS /usr/lib /usr/local/lib ) if(NOT M_LIBRARY) - message(STATUS - "math library 'libm' not found - floating point support disabled") + message(STATUS "math lib 'libm' not found; floating point support disabled") endif() else() # not needed on windows @@ -39,22 +38,14 @@ else() endif() # COMMAND LINE OPTIONS -if(DEFINED PNG_SHARED) - option(PNG_SHARED "Build shared lib" ${PNG_SHARED}) -else() - option(PNG_SHARED "Build shared lib" ON) -endif() -if(DEFINED PNG_STATIC) - option(PNG_STATIC "Build static lib" ${PNG_STATIC}) -else() - option(PNG_STATIC "Build static lib" ON) -endif() - -option(PNG_TESTS "Build libpng tests" YES) +option(PNG_SHARED "Build shared lib" ON) +option(PNG_STATIC "Build static lib" ON) +option(PNG_TESTS "Build libpng tests" ON) # Many more configuration options could be added here -option(PNG_DEBUG "Build with debug output" NO) -option(PNGARG "Disable ANSI-C prototypes" NO) +option(PNG_FRAMEWORK "Build OS X framework" OFF) +option(PNG_DEBUG "Build with debug output" OFF) +option(PNGARG "Disable ANSI-C prototypes" OFF) # SET LIBNAME set(PNG_LIB_NAME png${PNGLIB_MAJOR}${PNGLIB_MINOR}) @@ -134,7 +125,7 @@ if(PNG_SHARED) endif() if(PNG_STATIC) -# does not work without changing name + # does not work without changing name set(PNG_LIB_NAME_STATIC ${PNG_LIB_NAME}_static) add_library(${PNG_LIB_NAME_STATIC} STATIC ${libpng_sources}) list(APPEND PNG_LIB_TARGETS ${PNG_LIB_NAME_STATIC}) @@ -145,10 +136,27 @@ if(PNG_STATIC) target_link_libraries(${PNG_LIB_NAME_STATIC} ${ZLIB_LIBRARY} ${M_LIBRARY}) endif() +if(PNG_FRAMEWORK) + set(PNG_LIB_NAME_FRAMEWORK ${PNG_LIB_NAME}_framework) + add_library(${PNG_LIB_NAME_FRAMEWORK} SHARED ${libpng_sources}) + list(APPEND PNG_LIB_TARGETS ${PNG_LIB_NAME_FRAMEWORK}) + set_target_properties(${PNG_LIB_NAME_FRAMEWORK} PROPERTIES + FRAMEWORK TRUE + FRAMEWORK_VERSION ${PNGLIB_VERSION} + MACOSX_FRAMEWORK_SHORT_VERSION_STRING ${PNGLIB_MAJOR}.${PNGLIB_MINOR} + MACOSX_FRAMEWORK_BUNDLE_VERSION ${PNGLIB_VERSION} + MACOSX_FRAMEWORK_IDENTIFIER org.libpng.libpng + XCODE_ATTRIBUTE_INSTALL_PATH "@rpath" + PUBLIC_HEADER "${libpng_public_hdrs}" + OUTPUT_NAME png) + target_link_libraries(${PNG_LIB_NAME_FRAMEWORK} ${ZLIB_LIBRARY} ${M_LIBRARY}) +endif() + if(NOT PNG_LIB_TARGETS) message(SEND_ERROR "No library variant selected to build. " - "Please enable at least one of the following options: PNG_STATIC, PNG_SHARED") + "Please enable at least one of the following options: " + " PNG_STATIC, PNG_SHARED, PNG_FRAMEWORK") endif() if(PNG_SHARED AND WIN32) @@ -253,7 +261,7 @@ endif(NOT WIN32 OR CYGWIN OR MINGW) # SET UP LINKS if(PNG_SHARED) set_target_properties(${PNG_LIB_NAME} PROPERTIES -# VERSION 16.${PNGLIB_RELEASE}.1.6.16 +# VERSION 16.${PNGLIB_RELEASE}.1.6.17 VERSION 16.${PNGLIB_RELEASE}.0 SOVERSION 16 CLEAN_DIRECT_OUTPUT 1) @@ -284,7 +292,8 @@ if(NOT SKIP_INSTALL_LIBRARIES AND NOT SKIP_INSTALL_ALL ) ${PNG_EXPORT_RULE} RUNTIME DESTINATION bin LIBRARY DESTINATION ${CMAKE_INSTALL_LIBDIR} - ARCHIVE DESTINATION ${CMAKE_INSTALL_LIBDIR}) + ARCHIVE DESTINATION ${CMAKE_INSTALL_LIBDIR} + FRAMEWORK DESTINATION ${CMAKE_INSTALL_LIBDIR}) if(PNG_SHARED) # Create a symlink for libpng.dll.a => libpng16.dll.a on Cygwin @@ -17,7 +17,7 @@ VIII. Configuring libpng for 16-bit platforms XII. Configuring for compiler xxx: XIII. Removing unwanted object code XIV. Changes to the build and configuration of libpng in libpng-1.5.x - XV. Configuring libpng for multiprocessing + XV. Setjmp/longjmp issues XVI. Other sources of information about libpng I. Simple installation @@ -189,7 +189,7 @@ run "make install". VIII. Configuring libpng for 16-bit platforms You will want to look into zconf.h to tell zlib (and thus libpng) that -it cannot allocate more then 64K at a time. Even if you can, the memory +it cannot allocate more than 64K at a time. Even if you can, the memory won't be accessible. So limit zlib and libpng to 64K by defining MAXSEG_64K. IX. Configuring for DOS @@ -362,7 +362,7 @@ $PREFIX/include directory). Do not edit pnglibconf.h after you have built libpng, because than the settings would not accurately reflect the settings that were used to build libpng. -XV. Configuring libpng for multiprocessing +XV. Setjmp/longjmp issues Libpng uses setjmp()/longjmp() for error handling. Unfortunately setjmp() is known to be not thread-safe on some platforms and we don't know of @@ -375,6 +375,11 @@ configure libpng with PNG_NO_SETJMP in your pngusr.dfa file, with in your pnglibconf.h or pngusr.h. +Starting with libpng-1.6.0, the library included a "simplified API". +This requires setjmp/longjmp, so you must either build the library +with PNG_SETJMP_SUPPORTED defined, or with PNG_SIMPLIFIED_READ_SUPPORTED +and PNG_SIMPLIFIED_WRITE_SUPPORTED undefined. + XVI. Other sources of information about libpng: Further information can be found in the README and libpng-manual.txt @@ -10,8 +10,8 @@ this sentence. This code is released under the libpng license. -libpng versions 1.2.6, August 15, 2004, through 1.6.16, December 22, 2014, are -Copyright (c) 2004, 2006-2014 Glenn Randers-Pehrson, and are +libpng versions 1.2.6, August 15, 2004, through 1.6.17, March 26, 2015, are +Copyright (c) 2004, 2006-2015 Glenn Randers-Pehrson, and are distributed according to the same disclaimer and license as libpng-1.2.5 with the following individual added to the list of Contributing Authors @@ -108,4 +108,4 @@ certification mark of the Open Source Initiative. Glenn Randers-Pehrson glennrp at users.sourceforge.net -December 22, 2014 +March 26, 2015 diff --git a/Makefile.am b/Makefile.am index a9e8d3c3c..ac5f95cf7 100644 --- a/Makefile.am +++ b/Makefile.am @@ -129,11 +129,12 @@ EXTRA_DIST= \ $(TESTS) $(XFAIL_TESTS) tests/pngstest \ CMakeLists.txt example.c libpng-manual.txt -SCRIPT_CLEANFILES=scripts/*.out scripts/*.chk scripts/pnglibconf.c +SCRIPT_CLEANFILES=scripts/*.out scripts/*.chk CLEANFILES= *.tf? pngout.png libpng@PNGLIB_MAJOR@@PNGLIB_MINOR@.pc \ libpng@PNGLIB_MAJOR@@PNGLIB_MINOR@-config libpng.vers libpng.sym \ - check.new pnglibconf.* pngprefix.h symbols.new pngtest-log.txt \ + check.new pnglibconf.h pngprefix.h symbols.new pngtest-log.txt \ + pnglibconf.out pnglibconf.c pnglibconf.pre pnglibconf.dfn \ $(SCRIPT_CLEANFILES) MAINTAINERCLEANFILES = Makefile.in aclocal.m4 config.guess config.h.in \ @@ -1,4 +1,4 @@ -README for libpng version 1.6.16 - December 22, 2014 (shared library 16.0) +README for libpng version 1.6.17 - March 26, 2015 (shared library 16.0) See the note about version numbers near the top of png.h See INSTALL for instructions on how to install libpng. diff --git a/configure.ac b/configure.ac index c99a4d9ae..1c46ca235 100644 --- a/configure.ac +++ b/configure.ac @@ -18,7 +18,7 @@ AC_PREREQ([2.68]) dnl Version number stuff here: -AC_INIT([libpng],[1.6.16],[png-mng-implement@lists.sourceforge.net]) +AC_INIT([libpng],[1.6.17],[png-mng-implement@lists.sourceforge.net]) AC_CONFIG_MACRO_DIR([scripts]) # libpng does not follow GNU file name conventions (hence 'foreign') @@ -39,10 +39,10 @@ dnl automake, so the following is not necessary (and is not defined anyway): dnl AM_PREREQ([1.11.2]) dnl stop configure from automagically running automake -PNGLIB_VERSION=1.6.16 +PNGLIB_VERSION=1.6.17 PNGLIB_MAJOR=1 PNGLIB_MINOR=6 -PNGLIB_RELEASE=16 +PNGLIB_RELEASE=17 dnl End of version number stuff diff --git a/contrib/libtests/pngunknown.c b/contrib/libtests/pngunknown.c index b8c4899de..309ee604e 100644 --- a/contrib/libtests/pngunknown.c +++ b/contrib/libtests/pngunknown.c @@ -1242,4 +1242,4 @@ main(void) /* So the test is skipped: */ return 77; } -#endif /* SET_UNKNOWN_CHUNKS && READ*/ +#endif /* SET_UNKNOWN_CHUNKS && READ */ diff --git a/contrib/libtests/pngvalid.c b/contrib/libtests/pngvalid.c index a93fe65a6..79de125da 100644 --- a/contrib/libtests/pngvalid.c +++ b/contrib/libtests/pngvalid.c @@ -1,8 +1,8 @@ /* pngvalid.c - validate libpng by constructing then reading png files. * - * Last changed in libpng 1.6.14 [October 23, 2014] - * Copyright (c) 2014 Glenn Randers-Pehrson + * Last changed in libpng 1.6.17 [March 26, 2015] + * Copyright (c) 2014-2015 Glenn Randers-Pehrson * Written by John Cunningham Bowler * * This code is released under the libpng license. @@ -278,7 +278,8 @@ random_mod(unsigned int max) return x % max; /* 0 .. max-1 */ } -#ifdef PNG_READ_RGB_TO_GRAY_SUPPORTED +#if (defined PNG_READ_RGB_TO_GRAY_SUPPORTED) ||\ + (defined PNG_READ_FILLER_SUPPORTED) static int random_choice(void) { @@ -303,7 +304,7 @@ random_choice(void) #define COL_FROM_ID(id) ((png_byte)((id)& 0x7U)) #define DEPTH_FROM_ID(id) ((png_byte)(((id) >> 3) & 0x1fU)) #define PALETTE_FROM_ID(id) (((id) >> 8) & 0x1f) -#define INTERLACE_FROM_ID(id) ((int)(((id) >> 13) & 0x3)) +#define INTERLACE_FROM_ID(id) ((png_byte)(((id) >> 13) & 0x3)) #define DO_INTERLACE_FROM_ID(id) ((int)(((id)>>15) & 1)) #define WIDTH_FROM_ID(id) (((id)>>16) & 0xff) #define HEIGHT_FROM_ID(id) (((id)>>24) & 0xff) @@ -440,7 +441,7 @@ next_format(png_bytep colour_type, png_bytep bit_depth, #ifdef PNG_READ_TRANSFORMS_SUPPORTED static unsigned int sample(png_const_bytep row, png_byte colour_type, png_byte bit_depth, - png_uint_32 x, unsigned int sample_index) + png_uint_32 x, unsigned int sample_index, int swap16, int littleendian) { png_uint_32 bit_index, result; @@ -469,11 +470,23 @@ sample(png_const_bytep row, png_byte colour_type, png_byte bit_depth, return result; else if (bit_depth > 8) - return (result << 8) + *++row; + { + if (swap16) + return (*++row << 8) + result; + else + return (result << 8) + *++row; + } - /* Less than 8 bits per sample. */ + /* Less than 8 bits per sample. By default PNG has the big end of + * the egg on the left of the screen, but if littleendian is set + * then the big end is on the right. + */ bit_index &= 7; - return (result >> (8-bit_index-bit_depth)) & ((1U<<bit_depth)-1); + + if (!littleendian) + bit_index = 8-bit_index-bit_depth; + + return (result >> bit_index) & ((1U<<bit_depth)-1); } #endif /* PNG_READ_TRANSFORMS_SUPPORTED */ @@ -1565,7 +1578,7 @@ set_store_for_write(png_store *ps, png_infopp ppi, else # endif - ps->pwrite = png_create_write_struct(png_get_libpng_ver(NULL), + ps->pwrite = png_create_write_struct(PNG_LIBPNG_VER_STRING, ps, store_error, store_warning); png_set_write_fn(ps->pwrite, ps, store_write, store_flush); @@ -3172,10 +3185,10 @@ init_standard_palette(png_store *ps, png_structp pp, png_infop pi, int npalette, for (; i<256; ++i) tRNS[i] = 24; -# ifdef PNG_WRITE_tRNS_SUPPORTED +# ifdef PNG_WRITE_tRNS_SUPPORTED if (j > 0) png_set_tRNS(pp, pi, tRNS, j, 0/*color*/); -# endif +# endif } } @@ -4263,7 +4276,8 @@ typedef struct standard_display png_byte green_sBIT; png_byte blue_sBIT; png_byte alpha_sBIT; - int interlace_type; + png_byte interlace_type; + png_byte filler; /* Output has a filler */ png_uint_32 id; /* Calculated file ID */ png_uint_32 w; /* Width of image */ png_uint_32 h; /* Height of image */ @@ -4650,8 +4664,16 @@ standard_info_part2(standard_display *dp, png_const_structp pp, png_const_infop pi, int nImages) { /* Record cbRow now that it can be found. */ - dp->pixel_size = bit_size(pp, png_get_color_type(pp, pi), - png_get_bit_depth(pp, pi)); + { + png_byte ct = png_get_color_type(pp, pi); + png_byte bd = png_get_bit_depth(pp, pi); + + if (bd >= 8 && (ct == PNG_COLOR_TYPE_RGB || ct == PNG_COLOR_TYPE_GRAY) && + dp->filler) + ct |= 4; /* handle filler as faked alpha channel */ + + dp->pixel_size = bit_size(pp, ct, bd); + } dp->bit_width = png_get_image_width(pp, pi) * dp->pixel_size; dp->cbRow = png_get_rowbytes(pp, pi); @@ -5291,10 +5313,17 @@ typedef struct image_pixel */ unsigned int red, green, blue, alpha; /* For non-palette images. */ unsigned int palette_index; /* For a palette image. */ - png_byte colour_type; /* As in the spec. */ - png_byte bit_depth; /* Defines bit size in row */ - png_byte sample_depth; /* Scale of samples */ - int have_tRNS; /* tRNS chunk may need processing */ + png_byte colour_type; /* As in the spec. */ + png_byte bit_depth; /* Defines bit size in row */ + png_byte sample_depth; /* Scale of samples */ + unsigned int have_tRNS :1; /* tRNS chunk may need processing */ + unsigned int swap_rgb :1; /* RGB swapped to BGR */ + unsigned int alpha_first :1; /* Alpha at start, not end */ + unsigned int alpha_inverted :1; /* Alpha channel inverted */ + unsigned int mono_inverted :1; /* Gray channel inverted */ + unsigned int swap16 :1; /* Byte swap 16-bit components */ + unsigned int littleendian :1; /* High bits on right */ + unsigned int sig_bits :1; /* Pixel shifted (sig bits only) */ /* For checking the code calculates double precision floating point values * along with an error value, accumulated from the transforms. Because an @@ -5302,6 +5331,9 @@ typedef struct image_pixel * up to just less than +/-1 in the scaled value) the *lowest* sBIT for each * channel is stored. This sBIT value is folded in to the stored error value * at the end of the application of the transforms to the pixel. + * + * If sig_bits is set above the red, green, blue and alpha values have been + * scaled so they only contain the significant bits of the component values. */ double redf, greenf, bluef, alphaf; double rede, greene, bluee, alphae; @@ -5310,26 +5342,27 @@ typedef struct image_pixel /* Shared utility function, see below. */ static void -image_pixel_setf(image_pixel *this, unsigned int max) +image_pixel_setf(image_pixel *this, unsigned int rMax, unsigned int gMax, + unsigned int bMax, unsigned int aMax) { - this->redf = this->red / (double)max; - this->greenf = this->green / (double)max; - this->bluef = this->blue / (double)max; - this->alphaf = this->alpha / (double)max; + this->redf = this->red / (double)rMax; + this->greenf = this->green / (double)gMax; + this->bluef = this->blue / (double)bMax; + this->alphaf = this->alpha / (double)aMax; - if (this->red < max) + if (this->red < rMax) this->rede = this->redf * DBL_EPSILON; else this->rede = 0; - if (this->green < max) + if (this->green < gMax) this->greene = this->greenf * DBL_EPSILON; else this->greene = 0; - if (this->blue < max) + if (this->blue < bMax) this->bluee = this->bluef * DBL_EPSILON; else this->bluee = 0; - if (this->alpha < max) + if (this->alpha < aMax) this->alphae = this->alphaf * DBL_EPSILON; else this->alphae = 0; @@ -5341,18 +5374,22 @@ image_pixel_setf(image_pixel *this, unsigned int max) */ static void image_pixel_init(image_pixel *this, png_const_bytep row, png_byte colour_type, - png_byte bit_depth, png_uint_32 x, store_palette palette) + png_byte bit_depth, png_uint_32 x, store_palette palette, + PNG_CONST image_pixel *format /*from pngvalid transform of input*/) { PNG_CONST png_byte sample_depth = (png_byte)(colour_type == PNG_COLOR_TYPE_PALETTE ? 8 : bit_depth); PNG_CONST unsigned int max = (1U<<sample_depth)-1; + PNG_CONST int swap16 = (format != 0 && format->swap16); + PNG_CONST int littleendian = (format != 0 && format->littleendian); + PNG_CONST int sig_bits = (format != 0 && format->sig_bits); /* Initially just set everything to the same number and the alpha to opaque. * Note that this currently assumes a simple palette where entry x has colour * rgb(x,x,x)! */ this->palette_index = this->red = this->green = this->blue = - sample(row, colour_type, bit_depth, x, 0); + sample(row, colour_type, bit_depth, x, 0, swap16, littleendian); this->alpha = max; this->red_sBIT = this->green_sBIT = this->blue_sBIT = this->alpha_sBIT = sample_depth; @@ -5376,21 +5413,52 @@ image_pixel_init(image_pixel *this, png_const_bytep row, png_byte colour_type, { unsigned int i = 0; + if ((colour_type & 4) != 0 && format != 0 && format->alpha_first) + { + this->alpha = this->red; + /* This handles the gray case for 'AG' pixels */ + this->palette_index = this->red = this->green = this->blue = + sample(row, colour_type, bit_depth, x, 1, swap16, littleendian); + i = 1; + } + if (colour_type & 2) { - this->green = sample(row, colour_type, bit_depth, x, 1); - this->blue = sample(row, colour_type, bit_depth, x, 2); - i = 2; + /* Green is second for both BGR and RGB: */ + this->green = sample(row, colour_type, bit_depth, x, ++i, swap16, + littleendian); + + if (format != 0 && format->swap_rgb) /* BGR */ + this->red = sample(row, colour_type, bit_depth, x, ++i, swap16, + littleendian); + else + this->blue = sample(row, colour_type, bit_depth, x, ++i, swap16, + littleendian); + } + + else /* grayscale */ if (format != 0 && format->mono_inverted) + this->red = this->green = this->blue = this->red ^ max; + + if ((colour_type & 4) != 0) /* alpha */ + { + if (format == 0 || !format->alpha_first) + this->alpha = sample(row, colour_type, bit_depth, x, ++i, swap16, + littleendian); + + if (format != 0 && format->alpha_inverted) + this->alpha ^= max; } - if (colour_type & 4) - this->alpha = sample(row, colour_type, bit_depth, x, ++i); } /* Calculate the scaled values, these are simply the values divided by * 'max' and the error is initialized to the double precision epsilon value * from the header file. */ - image_pixel_setf(this, max); + image_pixel_setf(this, + sig_bits ? (1U << format->red_sBIT)-1 : max, + sig_bits ? (1U << format->green_sBIT)-1 : max, + sig_bits ? (1U << format->blue_sBIT)-1 : max, + sig_bits ? (1U << format->alpha_sBIT)-1 : max); /* Store the input information for use in the transforms - these will * modify the information. @@ -5399,6 +5467,13 @@ image_pixel_init(image_pixel *this, png_const_bytep row, png_byte colour_type, this->bit_depth = bit_depth; this->sample_depth = sample_depth; this->have_tRNS = 0; + this->swap_rgb = 0; + this->alpha_first = 0; + this->alpha_inverted = 0; + this->mono_inverted = 0; + this->swap16 = 0; + this->littleendian = 0; + this->sig_bits = 0; } /* Convert a palette image to an rgb image. This necessarily converts the tRNS @@ -5560,6 +5635,7 @@ typedef struct transform_display /* Local variables */ png_byte output_colour_type; png_byte output_bit_depth; + png_byte unpacked; /* Modifications (not necessarily used.) */ gama_modification gama_mod; @@ -5635,6 +5711,7 @@ image_transform_mod_end(PNG_CONST image_transform *this, image_pixel *that, png_const_structp pp, PNG_CONST transform_display *display) { PNG_CONST unsigned int scale = (1U<<that->sample_depth)-1; + PNG_CONST int sig_bits = that->sig_bits; UNUSED(this) UNUSED(pp) @@ -5649,6 +5726,13 @@ image_transform_mod_end(PNG_CONST image_transform *this, image_pixel *that, */ that->red = sample_scale(that->redf, scale); + /* This is a bit bogus; really the above calculation should use the red_sBIT + * value, not sample_depth, but because libpng does png_set_shift by just + * shifting the bits we get errors if we don't do it the same way. + */ + if (sig_bits && that->red_sBIT < that->sample_depth) + that->red >>= that->sample_depth - that->red_sBIT; + /* The error value is increased, at the end, according to the lowest sBIT * value seen. Common sense tells us that the intermediate integer * representations are no more accurate than +/- 0.5 in the integral values, @@ -5664,7 +5748,13 @@ image_transform_mod_end(PNG_CONST image_transform *this, image_pixel *that, if (that->colour_type & PNG_COLOR_MASK_COLOR) { that->green = sample_scale(that->greenf, scale); + if (sig_bits && that->green_sBIT < that->sample_depth) + that->green >>= that->sample_depth - that->green_sBIT; + that->blue = sample_scale(that->bluef, scale); + if (sig_bits && that->blue_sBIT < that->sample_depth) + that->blue >>= that->sample_depth - that->blue_sBIT; + that->greene += 1./(2*((1U<<that->green_sBIT)-1)); that->bluee += 1./(2*((1U<<that->blue_sBIT)-1)); } @@ -5684,9 +5774,12 @@ image_transform_mod_end(PNG_CONST image_transform *this, image_pixel *that, else { that->alpha = scale; /* opaque */ - that->alpha = 1; /* Override this. */ + that->alphaf = 1; /* Override this. */ that->alphae = 0; /* It's exact ;-) */ } + + if (sig_bits && that->alpha_sBIT < that->sample_depth) + that->alpha >>= that->sample_depth - that->alpha_sBIT; } /* Static 'end' structure: */ @@ -5724,6 +5817,7 @@ transform_display_init(transform_display *dp, png_modifier *pm, png_uint_32 id, /* Local variable fields */ dp->output_colour_type = 255; /* invalid */ dp->output_bit_depth = 255; /* invalid */ + dp->unpacked = 0; /* not unpacked */ } static void @@ -5751,6 +5845,13 @@ transform_info_imp(transform_display *dp, png_structp pp, png_infop pi) dp->output_colour_type = png_get_color_type(pp, pi); dp->output_bit_depth = png_get_bit_depth(pp, pi); + /* If png_set_filler is in action then fake the output color type to include + * an alpha channel where appropriate. + */ + if (dp->output_bit_depth >= 8 && (dp->output_colour_type == PNG_COLOR_TYPE_RGB || + dp->output_colour_type == PNG_COLOR_TYPE_GRAY) && dp->this.filler) + dp->output_colour_type |= 4; + /* Validate the combination of colour type and bit depth that we are getting * out of libpng; the semantics of something not in the PNG spec are, at * best, unclear. @@ -5785,7 +5886,8 @@ transform_info_imp(transform_display *dp, png_structp pp, png_infop pi) } /* Use a test pixel to check that the output agrees with what we expect - - * this avoids running the whole test if the output is unexpected. + * this avoids running the whole test if the output is unexpected. This also + * checks for internal errors. */ { image_pixel test_pixel; @@ -5800,7 +5902,7 @@ transform_info_imp(transform_display *dp, png_structp pp, png_infop pi) /* Don't need sBIT here, but it must be set to non-zero to avoid * arithmetic overflows. */ - test_pixel.have_tRNS = dp->this.is_transparent; + test_pixel.have_tRNS = dp->this.is_transparent != 0; test_pixel.red_sBIT = test_pixel.green_sBIT = test_pixel.blue_sBIT = test_pixel.alpha_sBIT = test_pixel.sample_depth; @@ -5831,22 +5933,41 @@ transform_info_imp(transform_display *dp, png_structp pp, png_infop pi) } /* If both bit depth and colour type are correct check the sample depth. - * I believe these are both internal errors. */ - if (test_pixel.colour_type == PNG_COLOR_TYPE_PALETTE) + if (test_pixel.colour_type == PNG_COLOR_TYPE_PALETTE && + test_pixel.sample_depth != 8) /* oops - internal error! */ + png_error(pp, "pngvalid: internal: palette sample depth not 8"); + else if (dp->unpacked && test_pixel.bit_depth != 8) + png_error(pp, "pngvalid: internal: bad unpacked pixel depth"); + else if (!dp->unpacked && test_pixel.colour_type != PNG_COLOR_TYPE_PALETTE + && test_pixel.bit_depth != test_pixel.sample_depth) { - if (test_pixel.sample_depth != 8) /* oops - internal error! */ - png_error(pp, "pngvalid: internal: palette sample depth not 8"); + char message[128]; + size_t pos = safecat(message, sizeof message, 0, + "internal: sample depth "); + + /* Because unless something has set 'unpacked' or the image is palette + * mapped we expect the transform to keep sample depth and bit depth + * the same. + */ + pos = safecatn(message, sizeof message, pos, test_pixel.sample_depth); + pos = safecat(message, sizeof message, pos, " expected "); + pos = safecatn(message, sizeof message, pos, test_pixel.bit_depth); + + png_error(pp, message); } - else if (test_pixel.sample_depth != dp->output_bit_depth) + else if (test_pixel.bit_depth != dp->output_bit_depth) { + /* This could be a libpng error too; libpng has not produced what we + * expect for the output bit depth. + */ char message[128]; size_t pos = safecat(message, sizeof message, 0, - "internal: sample depth "); + "internal: bit depth "); pos = safecatn(message, sizeof message, pos, dp->output_bit_depth); pos = safecat(message, sizeof message, pos, " expected "); - pos = safecatn(message, sizeof message, pos, test_pixel.sample_depth); + pos = safecatn(message, sizeof message, pos, test_pixel.bit_depth); png_error(pp, message); } @@ -5993,13 +6114,14 @@ transform_image_validate(transform_display *dp, png_const_structp pp, unsigned int r, g, b, a; /* Find out what we think the pixel should be: */ - image_pixel_init(&in_pixel, std, in_ct, in_bd, x, dp->this.palette); + image_pixel_init(&in_pixel, std, in_ct, in_bd, x, dp->this.palette, + NULL); in_pixel.red_sBIT = red_sBIT; in_pixel.green_sBIT = green_sBIT; in_pixel.blue_sBIT = blue_sBIT; in_pixel.alpha_sBIT = alpha_sBIT; - in_pixel.have_tRNS = have_tRNS; + in_pixel.have_tRNS = have_tRNS != 0; /* For error detection, below. */ r = in_pixel.red; @@ -6007,12 +6129,18 @@ transform_image_validate(transform_display *dp, png_const_structp pp, b = in_pixel.blue; a = in_pixel.alpha; + /* This applies the transforms to the input date, including output + * format operations which must be used when reading the output + * pixel that libpng produces. + */ dp->transform_list->mod(dp->transform_list, &in_pixel, pp, dp); /* Read the output pixel and compare it to what we got, we don't - * use the error field here, so no need to update sBIT. + * use the error field here, so no need to update sBIT. in_pixel + * says whether we expect libpng to change the output format. */ - image_pixel_init(&out_pixel, pRow, out_ct, out_bd, x, out_palette); + image_pixel_init(&out_pixel, pRow, out_ct, out_bd, x, out_palette, + &in_pixel); /* We don't expect changes to the index here even if the bit depth is * changed. @@ -6793,14 +6921,14 @@ image_transform_png_set_rgb_to_gray_ini(PNG_CONST image_transform *this, * conversion adds another +/-2 in the 16-bit case and * +/-(1<<(15-PNG_MAX_GAMMA_8)) in the 8-bit case. */ - that->pm->limit += -# if PNG_MAX_GAMMA_8 < 14 - pow((that->this.bit_depth == 16 ? - 8. : 6. + (1<<(15-PNG_MAX_GAMMA_8)))/65535, data.gamma); -# else - pow((that->this.bit_depth == 16 ? - 8. : 8. + (1<<(15-PNG_MAX_GAMMA_8)))/65535, data.gamma); -# endif + that->pm->limit += pow( +# if PNG_MAX_GAMMA_8 < 14 + (that->this.bit_depth == 16 ? 8. : + 6. + (1<<(15-PNG_MAX_GAMMA_8))) +# else + 8. +# endif + /65535, data.gamma); } else @@ -6812,18 +6940,18 @@ image_transform_png_set_rgb_to_gray_ini(PNG_CONST image_transform *this, * When DIGITIZE is set because a pre-1.7 version of libpng is being * tested allow a bigger slack. * - * NOTE: this magic number was determined by experiment to be 1.1 (when - * using fixed point arithmetic). There's no great merit to the value - * below, however it only affects the limit used for checking for - * internal calculation errors, not the actual limit imposed by - * pngvalid on the output errors. + * NOTE: this magic number was determined by experiment to be 1.25. + * There's no great merit to the value below, however it only affects + * the limit used for checking for internal calculation errors, not + * the actual limit imposed by pngvalid on the output errors. */ - that->pm->limit += + that->pm->limit += pow( # if DIGITIZE - pow(1.1 /255, data.gamma); + 1.25 # else - pow(1.0 /255, data.gamma); + 1.0 # endif + /255, data.gamma); } } @@ -7306,7 +7434,7 @@ image_transform_png_set_background_set(PNG_CONST image_transform *this, } image_pixel_init(&data, random_bytes, colour_type, - bit_depth, 0/*x*/, 0/*unused: palette*/); + bit_depth, 0/*x*/, 0/*unused: palette*/, NULL/*format*/); /* Extract the background colour from this image_pixel, but make sure the * unused fields of 'back' are garbage. @@ -7397,6 +7525,598 @@ IT(background); #define PT ITSTRUCT(background) #endif /* PNG_READ_BACKGROUND_SUPPORTED */ +/* png_set_quantize(png_structp, png_colorp palette, int num_palette, + * int maximum_colors, png_const_uint_16p histogram, int full_quantize) + * + * Very difficult to validate this! + */ +/*NOTE: TBD NYI */ + +/* The data layout transforms are handled by swapping our own channel data, + * necessarily these need to happen at the end of the transform list because the + * semantic of the channels changes after these are executed. Some of these, + * like set_shift and set_packing, can't be done at present because they change + * the layout of the data at the sub-sample level so sample() won't get the + * right answer. + */ +/* png_set_invert_alpha */ +#ifdef PNG_READ_INVERT_ALPHA_SUPPORTED +/* Invert the alpha channel + * + * png_set_invert_alpha(png_structrp png_ptr) + */ +static void +image_transform_png_set_invert_alpha_set(PNG_CONST image_transform *this, + transform_display *that, png_structp pp, png_infop pi) +{ + png_set_invert_alpha(pp); + this->next->set(this->next, that, pp, pi); +} + +static void +image_transform_png_set_invert_alpha_mod(PNG_CONST image_transform *this, + image_pixel *that, png_const_structp pp, + PNG_CONST transform_display *display) +{ + if (that->colour_type & 4) + that->alpha_inverted = 1; + + this->next->mod(this->next, that, pp, display); +} + +static int +image_transform_png_set_invert_alpha_add(image_transform *this, + PNG_CONST image_transform **that, png_byte colour_type, png_byte bit_depth) +{ + UNUSED(bit_depth) + + this->next = *that; + *that = this; + + /* Only has an effect on pixels with alpha: */ + return (colour_type & 4) != 0; +} + +IT(invert_alpha); +#undef PT +#define PT ITSTRUCT(invert_alpha) + +#endif /* PNG_READ_INVERT_ALPHA_SUPPORTED */ + +/* png_set_bgr */ +#ifdef PNG_READ_BGR_SUPPORTED +/* Swap R,G,B channels to order B,G,R. + * + * png_set_bgr(png_structrp png_ptr) + * + * This only has an effect on RGB and RGBA pixels. + */ +static void +image_transform_png_set_bgr_set(PNG_CONST image_transform *this, + transform_display *that, png_structp pp, png_infop pi) +{ + png_set_bgr(pp); + this->next->set(this->next, that, pp, pi); +} + +static void +image_transform_png_set_bgr_mod(PNG_CONST image_transform *this, + image_pixel *that, png_const_structp pp, + PNG_CONST transform_display *display) +{ + if (that->colour_type == PNG_COLOR_TYPE_RGB || + that->colour_type == PNG_COLOR_TYPE_RGBA) + that->swap_rgb = 1; + + this->next->mod(this->next, that, pp, display); +} + +static int +image_transform_png_set_bgr_add(image_transform *this, + PNG_CONST image_transform **that, png_byte colour_type, png_byte bit_depth) +{ + UNUSED(bit_depth) + + this->next = *that; + *that = this; + + return colour_type == PNG_COLOR_TYPE_RGB || + colour_type == PNG_COLOR_TYPE_RGBA; +} + +IT(bgr); +#undef PT +#define PT ITSTRUCT(bgr) + +#endif /* PNG_READ_BGR_SUPPORTED */ + +/* png_set_swap_alpha */ +#ifdef PNG_READ_SWAP_ALPHA_SUPPORTED +/* Put the alpha channel first. + * + * png_set_swap_alpha(png_structrp png_ptr) + * + * This only has an effect on GA and RGBA pixels. + */ +static void +image_transform_png_set_swap_alpha_set(PNG_CONST image_transform *this, + transform_display *that, png_structp pp, png_infop pi) +{ + png_set_swap_alpha(pp); + this->next->set(this->next, that, pp, pi); +} + +static void +image_transform_png_set_swap_alpha_mod(PNG_CONST image_transform *this, + image_pixel *that, png_const_structp pp, + PNG_CONST transform_display *display) +{ + if (that->colour_type == PNG_COLOR_TYPE_GA || + that->colour_type == PNG_COLOR_TYPE_RGBA) + that->alpha_first = 1; + + this->next->mod(this->next, that, pp, display); +} + +static int +image_transform_png_set_swap_alpha_add(image_transform *this, + PNG_CONST image_transform **that, png_byte colour_type, png_byte bit_depth) +{ + UNUSED(bit_depth) + + this->next = *that; + *that = this; + + return colour_type == PNG_COLOR_TYPE_GA || + colour_type == PNG_COLOR_TYPE_RGBA; +} + +IT(swap_alpha); +#undef PT +#define PT ITSTRUCT(swap_alpha) + +#endif /* PNG_READ_SWAP_ALPHA_SUPPORTED */ + +/* png_set_swap */ +#ifdef PNG_READ_SWAP_SUPPORTED +/* Byte swap 16-bit components. + * + * png_set_swap(png_structrp png_ptr) + */ +static void +image_transform_png_set_swap_set(PNG_CONST image_transform *this, + transform_display *that, png_structp pp, png_infop pi) +{ + png_set_swap(pp); + this->next->set(this->next, that, pp, pi); +} + +static void +image_transform_png_set_swap_mod(PNG_CONST image_transform *this, + image_pixel *that, png_const_structp pp, + PNG_CONST transform_display *display) +{ + if (that->bit_depth == 16) + that->swap16 = 1; + + this->next->mod(this->next, that, pp, display); +} + +static int +image_transform_png_set_swap_add(image_transform *this, + PNG_CONST image_transform **that, png_byte colour_type, png_byte bit_depth) +{ + UNUSED(colour_type) + + this->next = *that; + *that = this; + + return bit_depth == 16; +} + +IT(swap); +#undef PT +#define PT ITSTRUCT(swap) + +#endif /* PNG_READ_SWAP_SUPPORTED */ + +#ifdef PNG_READ_FILLER_SUPPORTED +/* Add a filler byte to 8-bit Gray or 24-bit RGB images. + * + * png_set_filler, (png_structp png_ptr, png_uint_32 filler, int flags)); + * + * Flags: + * + * PNG_FILLER_BEFORE + * PNG_FILLER_AFTER + */ +#define data ITDATA(filler) +static struct +{ + png_uint_32 filler; + int flags; +} data; + +static void +image_transform_png_set_filler_set(PNG_CONST image_transform *this, + transform_display *that, png_structp pp, png_infop pi) +{ + /* Need a random choice for 'before' and 'after' as well as for the + * filler. The 'filler' value has all 32 bits set, but only bit_depth + * will be used. At this point we don't know bit_depth. + */ + RANDOMIZE(data.filler); + data.flags = random_choice(); + + png_set_filler(pp, data.filler, data.flags); + + /* The standard display handling stuff also needs to know that + * there is a filler, so set that here. + */ + that->this.filler = 1; + + this->next->set(this->next, that, pp, pi); +} + +static void +image_transform_png_set_filler_mod(PNG_CONST image_transform *this, + image_pixel *that, png_const_structp pp, + PNG_CONST transform_display *display) +{ + if (that->bit_depth >= 8 && + (that->colour_type == PNG_COLOR_TYPE_RGB || + that->colour_type == PNG_COLOR_TYPE_GRAY)) + { + PNG_CONST unsigned int max = (1U << that->bit_depth)-1; + that->alpha = data.filler & max; + that->alphaf = ((double)that->alpha) / max; + that->alphae = 0; + + /* The filler has been stored in the alpha channel, we must record + * that this has been done for the checking later on, the color + * type is faked to have an alpha channel, but libpng won't report + * this; the app has to know the extra channel is there and this + * was recording in standard_display::filler above. + */ + that->colour_type |= 4; /* alpha added */ + that->alpha_first = data.flags == PNG_FILLER_BEFORE; + } + + this->next->mod(this->next, that, pp, display); +} + +static int +image_transform_png_set_filler_add(image_transform *this, + PNG_CONST image_transform **that, png_byte colour_type, png_byte bit_depth) +{ + this->next = *that; + *that = this; + + return bit_depth >= 8 && (colour_type == PNG_COLOR_TYPE_RGB || + colour_type == PNG_COLOR_TYPE_GRAY); +} + +#undef data +IT(filler); +#undef PT +#define PT ITSTRUCT(filler) + +/* png_set_add_alpha, (png_structp png_ptr, png_uint_32 filler, int flags)); */ +/* Add an alpha byte to 8-bit Gray or 24-bit RGB images. */ +#define data ITDATA(add_alpha) +static struct +{ + png_uint_32 filler; + int flags; +} data; + +static void +image_transform_png_set_add_alpha_set(PNG_CONST image_transform *this, + transform_display *that, png_structp pp, png_infop pi) +{ + /* Need a random choice for 'before' and 'after' as well as for the + * filler. The 'filler' value has all 32 bits set, but only bit_depth + * will be used. At this point we don't know bit_depth. + */ + RANDOMIZE(data.filler); + data.flags = random_choice(); + + png_set_add_alpha(pp, data.filler, data.flags); + this->next->set(this->next, that, pp, pi); +} + +static void +image_transform_png_set_add_alpha_mod(PNG_CONST image_transform *this, + image_pixel *that, png_const_structp pp, + PNG_CONST transform_display *display) +{ + if (that->bit_depth >= 8 && + (that->colour_type == PNG_COLOR_TYPE_RGB || + that->colour_type == PNG_COLOR_TYPE_GRAY)) + { + PNG_CONST unsigned int max = (1U << that->bit_depth)-1; + that->alpha = data.filler & max; + that->alphaf = ((double)that->alpha) / max; + that->alphae = 0; + + that->colour_type |= 4; /* alpha added */ + that->alpha_first = data.flags == PNG_FILLER_BEFORE; + } + + this->next->mod(this->next, that, pp, display); +} + +static int +image_transform_png_set_add_alpha_add(image_transform *this, + PNG_CONST image_transform **that, png_byte colour_type, png_byte bit_depth) +{ + this->next = *that; + *that = this; + + return bit_depth >= 8 && (colour_type == PNG_COLOR_TYPE_RGB || + colour_type == PNG_COLOR_TYPE_GRAY); +} + +#undef data +IT(add_alpha); +#undef PT +#define PT ITSTRUCT(add_alpha) + +#endif /* PNG_READ_FILLER_SUPPORTED */ + +/* png_set_packing */ +#ifdef PNG_READ_PACK_SUPPORTED +/* Use 1 byte per pixel in 1, 2, or 4-bit depth files. + * + * png_set_packing(png_structrp png_ptr) + * + * This should only affect grayscale and palette images with less than 8 bits + * per pixel. + */ +static void +image_transform_png_set_packing_set(PNG_CONST image_transform *this, + transform_display *that, png_structp pp, png_infop pi) +{ + png_set_packing(pp); + that->unpacked = 1; + this->next->set(this->next, that, pp, pi); +} + +static void +image_transform_png_set_packing_mod(PNG_CONST image_transform *this, + image_pixel *that, png_const_structp pp, + PNG_CONST transform_display *display) +{ + /* The general expand case depends on what the colour type is, + * low bit-depth pixel values are unpacked into bytes without + * scaling, so sample_depth is not changed. + */ + if (that->bit_depth < 8) /* grayscale or palette */ + that->bit_depth = 8; + + this->next->mod(this->next, that, pp, display); +} + +static int +image_transform_png_set_packing_add(image_transform *this, + PNG_CONST image_transform **that, png_byte colour_type, png_byte bit_depth) +{ + UNUSED(colour_type) + + this->next = *that; + *that = this; + + /* Nothing should happen unless the bit depth is less than 8: */ + return bit_depth < 8; +} + +IT(packing); +#undef PT +#define PT ITSTRUCT(packing) + +#endif /* PNG_READ_PACK_SUPPORTED */ + +/* png_set_packswap */ +#ifdef PNG_READ_PACKSWAP_SUPPORTED +/* Swap pixels packed into bytes; reverses the order on screen so that + * the high order bits correspond to the rightmost pixels. + * + * png_set_packswap(png_structrp png_ptr) + */ +static void +image_transform_png_set_packswap_set(PNG_CONST image_transform *this, + transform_display *that, png_structp pp, png_infop pi) +{ + png_set_packswap(pp); + this->next->set(this->next, that, pp, pi); +} + +static void +image_transform_png_set_packswap_mod(PNG_CONST image_transform *this, + image_pixel *that, png_const_structp pp, + PNG_CONST transform_display *display) +{ + if (that->bit_depth < 8) + that->littleendian = 1; + + this->next->mod(this->next, that, pp, display); +} + +static int +image_transform_png_set_packswap_add(image_transform *this, + PNG_CONST image_transform **that, png_byte colour_type, png_byte bit_depth) +{ + UNUSED(colour_type) + + this->next = *that; + *that = this; + + return bit_depth < 8; +} + +IT(packswap); +#undef PT +#define PT ITSTRUCT(packswap) + +#endif /* PNG_READ_PACKSWAP_SUPPORTED */ + + +/* png_set_invert_mono */ +#ifdef PNG_READ_INVERT_MONO_SUPPORTED +/* Invert the gray channel + * + * png_set_invert_mono(png_structrp png_ptr) + */ +static void +image_transform_png_set_invert_mono_set(PNG_CONST image_transform *this, + transform_display *that, png_structp pp, png_infop pi) +{ + png_set_invert_mono(pp); + this->next->set(this->next, that, pp, pi); +} + +static void +image_transform_png_set_invert_mono_mod(PNG_CONST image_transform *this, + image_pixel *that, png_const_structp pp, + PNG_CONST transform_display *display) +{ + if (that->colour_type & 4) + that->mono_inverted = 1; + + this->next->mod(this->next, that, pp, display); +} + +static int +image_transform_png_set_invert_mono_add(image_transform *this, + PNG_CONST image_transform **that, png_byte colour_type, png_byte bit_depth) +{ + UNUSED(bit_depth) + + this->next = *that; + *that = this; + + /* Only has an effect on pixels with no colour: */ + return (colour_type & 2) == 0; +} + +IT(invert_mono); +#undef PT +#define PT ITSTRUCT(invert_mono) + +#endif /* PNG_READ_INVERT_MONO_SUPPORTED */ + +#ifdef PNG_READ_SHIFT_SUPPORTED +/* png_set_shift(png_structp, png_const_color_8p true_bits) + * + * The output pixels will be shifted by the given true_bits + * values. + */ +#define data ITDATA(shift) +static png_color_8 data; + +static void +image_transform_png_set_shift_set(PNG_CONST image_transform *this, + transform_display *that, png_structp pp, png_infop pi) +{ + /* Get a random set of shifts. The shifts need to do something + * to test the transform, so they are limited to the bit depth + * of the input image. Notice that in the following the 'gray' + * field is randomized independently. This acts as a check that + * libpng does use the correct field. + */ + PNG_CONST unsigned int depth = that->this.bit_depth; + + data.red = (png_byte)/*SAFE*/(random_mod(depth)+1); + data.green = (png_byte)/*SAFE*/(random_mod(depth)+1); + data.blue = (png_byte)/*SAFE*/(random_mod(depth)+1); + data.gray = (png_byte)/*SAFE*/(random_mod(depth)+1); + data.alpha = (png_byte)/*SAFE*/(random_mod(depth)+1); + + png_set_shift(pp, &data); + this->next->set(this->next, that, pp, pi); +} + +static void +image_transform_png_set_shift_mod(PNG_CONST image_transform *this, + image_pixel *that, png_const_structp pp, + PNG_CONST transform_display *display) +{ + /* Copy the correct values into the sBIT fields, libpng does not do + * anything to palette data: + */ + if (that->colour_type != PNG_COLOR_TYPE_PALETTE) + { + that->sig_bits = 1; + + /* The sBIT fields are reset to the values previously sent to + * png_set_shift according to the colour type. + * does. + */ + if (that->colour_type & 2) /* RGB channels */ + { + that->red_sBIT = data.red; + that->green_sBIT = data.green; + that->blue_sBIT = data.blue; + } + + else /* One grey channel */ + that->red_sBIT = that->green_sBIT = that->blue_sBIT = data.gray; + + that->alpha_sBIT = data.alpha; + } + + this->next->mod(this->next, that, pp, display); +} + +static int +image_transform_png_set_shift_add(image_transform *this, + PNG_CONST image_transform **that, png_byte colour_type, png_byte bit_depth) +{ + UNUSED(bit_depth) + + this->next = *that; + *that = this; + + return colour_type != PNG_COLOR_TYPE_PALETTE; +} + +IT(shift); +#undef PT +#define PT ITSTRUCT(shift) + +#endif /* PNG_READ_SHIFT_SUPPORTED */ + +#ifdef THIS_IS_THE_PROFORMA +static void +image_transform_png_set_@_set(PNG_CONST image_transform *this, + transform_display *that, png_structp pp, png_infop pi) +{ + png_set_@(pp); + this->next->set(this->next, that, pp, pi); +} + +static void +image_transform_png_set_@_mod(PNG_CONST image_transform *this, + image_pixel *that, png_const_structp pp, + PNG_CONST transform_display *display) +{ + this->next->mod(this->next, that, pp, display); +} + +static int +image_transform_png_set_@_add(image_transform *this, + PNG_CONST image_transform **that, png_byte colour_type, png_byte bit_depth) +{ + this->next = *that; + *that = this; + + return 1; +} + +IT(@); +#endif + + /* This may just be 'end' if all the transforms are disabled! */ static image_transform *PNG_CONST image_transform_first = &PT; @@ -7574,83 +8294,6 @@ image_transform_add(PNG_CONST image_transform **this, unsigned int max, } } -#ifdef THIS_IS_THE_PROFORMA -static void -image_transform_png_set_@_set(PNG_CONST image_transform *this, - transform_display *that, png_structp pp, png_infop pi) -{ - png_set_@(pp); - this->next->set(this->next, that, pp, pi); -} - -static void -image_transform_png_set_@_mod(PNG_CONST image_transform *this, - image_pixel *that, png_const_structp pp, - PNG_CONST transform_display *display) -{ - this->next->mod(this->next, that, pp, display); -} - -static int -image_transform_png_set_@_add(image_transform *this, - PNG_CONST image_transform **that, char *name, size_t sizeof_name, - size_t *pos, png_byte colour_type, png_byte bit_depth) -{ - this->next = *that; - *that = this; - - *pos = safecat(name, sizeof_name, *pos, " +@"); - - return 1; -} - -IT(@); -#endif - -/* png_set_quantize(png_structp, png_colorp palette, int num_palette, - * int maximum_colors, png_const_uint_16p histogram, int full_quantize) - * - * Very difficult to validate this! - */ -/*NOTE: TBD NYI */ - -/* The data layout transforms are handled by swapping our own channel data, - * necessarily these need to happen at the end of the transform list because the - * semantic of the channels changes after these are executed. Some of these, - * like set_shift and set_packing, can't be done at present because they change - * the layout of the data at the sub-sample level so sample() won't get the - * right answer. - */ -/* png_set_invert_alpha */ -/*NOTE: TBD NYI */ - -/* png_set_bgr */ -/*NOTE: TBD NYI */ - -/* png_set_swap_alpha */ -/*NOTE: TBD NYI */ - -/* png_set_swap */ -/*NOTE: TBD NYI */ - -/* png_set_filler, (png_structp png_ptr, png_uint_32 filler, int flags)); */ -/*NOTE: TBD NYI */ - -/* png_set_add_alpha, (png_structp png_ptr, png_uint_32 filler, int flags)); */ -/*NOTE: TBD NYI */ - -/* png_set_packing */ -/*NOTE: TBD NYI */ - -/* png_set_packswap */ -/*NOTE: TBD NYI */ - -/* png_set_invert_mono */ -/*NOTE: TBD NYI */ - -/* png_set_shift(png_structp, png_const_color_8p true_bits) */ -/*NOTE: TBD NYI */ - static void perform_transform_test(png_modifier *pm) { @@ -8753,9 +9396,9 @@ gamma_image_validate(gamma_display *dp, png_const_structp pp, /* Record the palette index for index images. */ PNG_CONST unsigned int in_index = - in_ct == 3 ? sample(std, 3, in_bd, x, 0) : 256; + in_ct == 3 ? sample(std, 3, in_bd, x, 0, 0, 0) : 256; PNG_CONST unsigned int out_index = - out_ct == 3 ? sample(std, 3, out_bd, x, 0) : 256; + out_ct == 3 ? sample(std, 3, out_bd, x, 0, 0, 0) : 256; /* Handle input alpha - png_set_background will cause the output * alpha to disappear so there is nothing to check. @@ -8765,7 +9408,7 @@ gamma_image_validate(gamma_display *dp, png_const_structp pp, { PNG_CONST unsigned int input_alpha = in_ct == 3 ? dp->this.palette[in_index].alpha : - sample(std, in_ct, in_bd, x, samples_per_pixel); + sample(std, in_ct, in_bd, x, samples_per_pixel, 0, 0); unsigned int output_alpha = 65536 /* as a flag value */; @@ -8777,7 +9420,7 @@ gamma_image_validate(gamma_display *dp, png_const_structp pp, else if ((out_ct & PNG_COLOR_MASK_ALPHA) != 0) output_alpha = sample(pRow, out_ct, out_bd, x, - samples_per_pixel); + samples_per_pixel, 0, 0); if (output_alpha != 65536) alpha = gamma_component_validate("alpha", &vi, input_alpha, @@ -8796,30 +9439,30 @@ gamma_image_validate(gamma_display *dp, png_const_structp pp, /* Handle grayscale or RGB components. */ if ((in_ct & PNG_COLOR_MASK_COLOR) == 0) /* grayscale */ (void)gamma_component_validate("gray", &vi, - sample(std, in_ct, in_bd, x, 0), - sample(pRow, out_ct, out_bd, x, 0), alpha/*component*/, - vi.background_red); + sample(std, in_ct, in_bd, x, 0, 0, 0), + sample(pRow, out_ct, out_bd, x, 0, 0, 0), + alpha/*component*/, vi.background_red); else /* RGB or palette */ { (void)gamma_component_validate("red", &vi, in_ct == 3 ? in_palette[in_index].red : - sample(std, in_ct, in_bd, x, 0), + sample(std, in_ct, in_bd, x, 0, 0, 0), out_ct == 3 ? out_palette[out_index].red : - sample(pRow, out_ct, out_bd, x, 0), + sample(pRow, out_ct, out_bd, x, 0, 0, 0), alpha/*component*/, vi.background_red); (void)gamma_component_validate("green", &vi, in_ct == 3 ? in_palette[in_index].green : - sample(std, in_ct, in_bd, x, 1), + sample(std, in_ct, in_bd, x, 1, 0, 0), out_ct == 3 ? out_palette[out_index].green : - sample(pRow, out_ct, out_bd, x, 1), + sample(pRow, out_ct, out_bd, x, 1, 0, 0), alpha/*component*/, vi.background_green); (void)gamma_component_validate("blue", &vi, in_ct == 3 ? in_palette[in_index].blue : - sample(std, in_ct, in_bd, x, 2), + sample(std, in_ct, in_bd, x, 2, 0, 0), out_ct == 3 ? out_palette[out_index].blue : - sample(pRow, out_ct, out_bd, x, 2), + sample(pRow, out_ct, out_bd, x, 2, 0, 0), alpha/*component*/, vi.background_blue); } } diff --git a/contrib/tools/pngfix.c b/contrib/tools/pngfix.c index 65422f600..67bab6aec 100644 --- a/contrib/tools/pngfix.c +++ b/contrib/tools/pngfix.c @@ -1,8 +1,8 @@ /* pngfix.c * - * Copyright (c) 2014 John Cunningham Bowler + * Copyright (c) 2014-2015 John Cunningham Bowler * - * Last changed in libpng 1.6.14 [October 23, 2014] + * Last changed in libpng 1.6.17 [March 26, 2015] * * This code is released under the libpng license. * For conditions of distribution and use, see the disclaimer @@ -2138,7 +2138,7 @@ zlib_end(struct zlib *zlib) * * z-rc is the zlib failure code; message is the error message with * spaces replaced by '-'. The compressed byte count indicates where - * in the zlib stream the error occured. + * in the zlib stream the error occurred. */ type_name(zlib->chunk->chunk_type, stdout); printf(" SKP %s %d %s ", zlib_flevel(zlib), zlib->file_bits, @@ -3577,10 +3577,9 @@ read_png(struct control *control) { png_structp png_ptr; png_infop info_ptr = NULL; - volatile png_bytep row = NULL, display = NULL; volatile int rc; - png_ptr = png_create_read_struct(png_get_libpng_ver(NULL), control, + png_ptr = png_create_read_struct(PNG_LIBPNG_VER_STRING, control, error_handler, warning_handler); if (png_ptr == NULL) @@ -3594,6 +3593,16 @@ read_png(struct control *control) rc = setjmp(control->file.jmpbuf); if (rc == 0) { +# ifdef PNG_SET_USER_LIMITS_SUPPORTED + /* Remove any limits on the size of PNG files that can be read, + * without this we may reject files based on built-in safety + * limits. + */ + png_set_user_limits(png_ptr, 0x7fffffff, 0x7fffffff); + png_set_chunk_cache_max(png_ptr, 0); + png_set_chunk_malloc_max(png_ptr, 0); +# endif + png_set_read_fn(png_ptr, control, read_callback); info_ptr = png_create_info_struct(png_ptr); @@ -3606,32 +3615,22 @@ read_png(struct control *control) png_read_info(png_ptr, info_ptr); { - png_size_t rowbytes = png_get_rowbytes(png_ptr, info_ptr); - - row = png_voidcast(png_byte*, malloc(rowbytes)); - display = png_voidcast(png_byte*, malloc(rowbytes)); - - if (row == NULL || display == NULL) - png_error(png_ptr, "OOM allocating row buffers"); - - { - png_uint_32 height = png_get_image_height(png_ptr, info_ptr); - int passes = png_set_interlace_handling(png_ptr); - int pass; - - png_start_read_image(png_ptr); - - for (pass = 0; pass < passes; ++pass) - { - png_uint_32 y = height; - - /* NOTE: this trashes the row each time; interlace handling won't - * work, but this avoids memory thrashing for speed testing. - */ - while (y-- > 0) - png_read_row(png_ptr, row, display); - } - } + png_uint_32 height = png_get_image_height(png_ptr, info_ptr); + int passes = png_set_interlace_handling(png_ptr); + int pass; + + png_start_read_image(png_ptr); + + for (pass = 0; pass < passes; ++pass) + { + png_uint_32 y = height; + + /* NOTE: this skips asking libpng to return either version of + * the image row, but libpng still reads the rows. + */ + while (y-- > 0) + png_read_row(png_ptr, NULL, NULL); + } } if (control->file.global->verbose) @@ -3642,8 +3641,6 @@ read_png(struct control *control) } png_destroy_read_struct(&png_ptr, &info_ptr, NULL); - if (row != NULL) free(row); - if (display != NULL) free(display); return rc; } @@ -3705,21 +3702,21 @@ usage(const char *prog) " gAMA, sRGB [all]: These specify the gamma encoding used for the pixel", " values.", " cHRM, iCCP [color]: These specify how colors are encoded. iCCP also", -" specifies the exact encoding of a pixel value however in practice", -" most programs will ignore it.", +" specifies the exact encoding of a pixel value; however, in", +" practice most programs will ignore it.", " bKGD [transform]: This is used by libpng transforms." " --max=<number>:", " Use IDAT chunks sized <number>. If no number is given the the IDAT", " chunks will be the maximum size permitted; 2^31-1 bytes. If the option", " is omitted the original chunk sizes will not be changed. When the", -" option is given --strip=unsafe is set automatically, this may be", +" option is given --strip=unsafe is set automatically. This may be", " cancelled if you know that all unknown unsafe-to-copy chunks really are", " safe to copy across an IDAT size change. This is true of all chunks", " that have ever been formally proposed as PNG extensions.", " MESSAGES", " By default the program only outputs summaries for each file.", " --quiet (-q):", -" Do not output the summaries except for files which cannot be read. With", +" Do not output the summaries except for files that cannot be read. With", " two --quiets these are not output either.", " --errors (-e):", " Output errors from libpng and the program (except too-far-back).", @@ -3752,7 +3749,7 @@ usage(const char *prog) " the following codes. Notice that the results for each file are combined", " together - check one file at a time to get a meaningful error code!", " 0x01: The zlib too-far-back error existed in at least one chunk.", -" 0x02: At least once chunk had a CRC error.", +" 0x02: At least one chunk had a CRC error.", " 0x04: A chunk length was incorrect.", " 0x08: The file was truncated.", " Errors less than 16 are potentially recoverable, for a single file if the", @@ -3760,7 +3757,7 @@ usage(const char *prog) " non-zero code is returned).", " 0x10: The file could not be read, even with corrections.", " 0x20: The output file could not be written.", -" 0x40: An unexpected, potentially internal, error occured.", +" 0x40: An unexpected, potentially internal, error occurred.", " If the command line arguments are incorrect the program exits with exit", " 255. Some older operating systems only support 7-bit exit codes, on those", " systems it is suggested that this program is first tested by supplying", @@ -3820,7 +3817,7 @@ usage(const char *prog) " SKP: The chunk was skipped because of a zlib issue (zlib-rc) with", " explanation 'message'", " ERR: The read of the file was aborted. The parameters explain why.", -"$3 status: For 'ERR' the accumulate status code from 'EXIT CODES' above.", +"$3 status: For 'ERR' the accumulated status code from 'EXIT CODES' above.", " This is printed as a 2 digit hexadecimal value", " comp-level: The recorded compression level (FLEVEL) of a zlib stream", " expressed as a string {supfast,stdfast,default,maximum}", diff --git a/libpng-manual.txt b/libpng-manual.txt index 3aa1f5787..f0dae987f 100644 --- a/libpng-manual.txt +++ b/libpng-manual.txt @@ -1,9 +1,9 @@ libpng-manual.txt - A description on how to use and modify libpng - libpng version 1.6.16 - December 22, 2014 + libpng version 1.6.17 - March 26, 2015 Updated and distributed by Glenn Randers-Pehrson <glennrp at users.sourceforge.net> - Copyright (c) 1998-2014 Glenn Randers-Pehrson + Copyright (c) 1998-2015 Glenn Randers-Pehrson This document is released under the libpng license. For conditions of distribution and use, see the disclaimer @@ -11,9 +11,9 @@ libpng-manual.txt - A description on how to use and modify libpng Based on: - libpng versions 0.97, January 1998, through 1.6.16 - December 22, 2014 + libpng versions 0.97, January 1998, through 1.6.17 - March 26, 2015 Updated and distributed by Glenn Randers-Pehrson - Copyright (c) 1998-2014 Glenn Randers-Pehrson + Copyright (c) 1998-2015 Glenn Randers-Pehrson libpng 1.0 beta 6 - version 0.96 - May 28, 1997 Updated and distributed by Andreas Dilger @@ -336,7 +336,7 @@ prediction. If you are intending to keep the file pointer open for use in libpng, you must ensure you don't read more than 8 bytes from the beginning -of the file, and you also have to make a call to png_set_sig_bytes_read() +of the file, and you also have to make a call to png_set_sig_bytes() with the number of bytes you read from the beginning. Libpng will then only check the bytes (if any) that your program didn't read. @@ -344,22 +344,23 @@ then only check the bytes (if any) that your program didn't read. to replace them with custom functions. See the discussion under Customizing libpng. - FILE *fp = fopen(file_name, "rb"); if (!fp) { return (ERROR); } - fread(header, 1, number, fp); - is_png = !png_sig_cmp(header, 0, number); + if (fread(header, 1, number, fp) != number) + { + return (ERROR); + } + is_png = !png_sig_cmp(header, 0, number); if (!is_png) { return (NOT_PNG); } - Next, png_struct and png_info need to be allocated and initialized. In order to ensure that the size of these structures is correct even with a dynamically linked libpng, there are functions to initialize and @@ -649,7 +650,7 @@ User limits The PNG specification allows the width and height of an image to be as large as 2^31-1 (0x7fffffff), or about 2.147 billion rows and columns. Larger images will be rejected immediately with a png_error() call. If -you wish to reduce these limits, you can use +you wish to change these limits, you can use png_set_user_limits(png_ptr, width_max, height_max); @@ -1264,13 +1265,13 @@ in until png_read_end() has read the chunk data following the image. the PNG datastream is embedded in a MNG-1.0 datastream) - Any or all of interlace_type, compression_type, or - filter_method can be NULL if you are - not interested in their values. + Any of width, height, color_type, bit_depth, + interlace_type, compression_type, or filter_method can + be NULL if you are not interested in their values. Note that png_get_IHDR() returns 32-bit data into the application's width and height variables. - This is an unsafe situation if these are 16-bit + This is an unsafe situation if these are not png_uint_32 variables. In such situations, the png_get_image_width() and png_get_image_height() functions described below are safer. @@ -1993,7 +1994,7 @@ value when you call it in this position: png_set_gamma(png_ptr, screen_gamma, 0.45455); If you need to reduce an RGB file to a paletted file, or if a paletted -file has more entries then will fit on your screen, png_set_quantize() +file has more entries than will fit on your screen, png_set_quantize() will do that. Note that this is a simple match quantization that merely finds the closest color available. This should work fairly well with optimized palettes, but fairly badly with linear color cubes. If you @@ -2546,7 +2547,7 @@ png_infop info_ptr; 64K. The library seems to run fine with sizes of 4K. Although you can give it much less if necessary (I assume you can give it chunks of - 1 byte, I haven't tried less then 256 bytes + 1 byte, I haven't tried less than 256 bytes yet). When this function returns, you may want to display any rows that were generated in the row callback if you don't already do @@ -5031,6 +5032,10 @@ The signatures of many exported functions were changed, such that png_infop became png_inforp or png_const_inforp where "rp" indicates a "restricted pointer". +The support for FAR/far types has been eliminated and the definition of +png_alloc_size_t is now controlled by a flag so that 'small size_t' systems +can select it if necessary. + Error detection in some chunks has improved; in particular the iCCP chunk reader now does pretty complete validation of the basic format. Some bad profiles that were previously accepted are now accepted with a warning or @@ -5045,7 +5050,7 @@ means of PNG_OPTION_ON); #endif -It's not a good idea to do this if you are using the new "simplified API", +It's not a good idea to do this if you are using the "simplified API", which needs to be able to recognize sRGB profiles conveyed via the iCCP chunk. @@ -5274,13 +5279,13 @@ Other rules can be inferred by inspecting the libpng source. XVI. Y2K Compliance in libpng -December 22, 2014 +March 26, 2015 Since the PNG Development group is an ad-hoc body, we can't make an official declaration. This is your unofficial assurance that libpng from version 0.71 and -upward through 1.6.16 are Y2K compliant. It is my belief that earlier +upward through 1.6.17 are Y2K compliant. It is my belief that earlier versions were also Y2K compliant. Libpng only has two year fields. One is a 2-byte unsigned integer @@ -1,6 +1,6 @@ -.TH LIBPNG 3 "December 22, 2014" +.TH LIBPNG 3 "March 26, 2015" .SH NAME -libpng \- Portable Network Graphics (PNG) Reference Library 1.6.16 +libpng \- Portable Network Graphics (PNG) Reference Library 1.6.17 .SH SYNOPSIS \fB #include <png.h>\fP @@ -504,10 +504,10 @@ Following is a copy of the libpng-manual.txt file that accompanies libpng. .SH LIBPNG.TXT libpng-manual.txt - A description on how to use and modify libpng - libpng version 1.6.16 - December 22, 2014 + libpng version 1.6.17 - March 26, 2015 Updated and distributed by Glenn Randers-Pehrson <glennrp at users.sourceforge.net> - Copyright (c) 1998-2014 Glenn Randers-Pehrson + Copyright (c) 1998-2015 Glenn Randers-Pehrson This document is released under the libpng license. For conditions of distribution and use, see the disclaimer @@ -515,9 +515,9 @@ libpng-manual.txt - A description on how to use and modify libpng Based on: - libpng versions 0.97, January 1998, through 1.6.16 - December 22, 2014 + libpng versions 0.97, January 1998, through 1.6.17 - March 26, 2015 Updated and distributed by Glenn Randers-Pehrson - Copyright (c) 1998-2014 Glenn Randers-Pehrson + Copyright (c) 1998-2015 Glenn Randers-Pehrson libpng 1.0 beta 6 - version 0.96 - May 28, 1997 Updated and distributed by Andreas Dilger @@ -840,7 +840,7 @@ prediction. If you are intending to keep the file pointer open for use in libpng, you must ensure you don't read more than 8 bytes from the beginning -of the file, and you also have to make a call to png_set_sig_bytes_read() +of the file, and you also have to make a call to png_set_sig_bytes() with the number of bytes you read from the beginning. Libpng will then only check the bytes (if any) that your program didn't read. @@ -848,22 +848,23 @@ then only check the bytes (if any) that your program didn't read. to replace them with custom functions. See the discussion under Customizing libpng. - FILE *fp = fopen(file_name, "rb"); if (!fp) { return (ERROR); } - fread(header, 1, number, fp); - is_png = !png_sig_cmp(header, 0, number); + if (fread(header, 1, number, fp) != number) + { + return (ERROR); + } + is_png = !png_sig_cmp(header, 0, number); if (!is_png) { return (NOT_PNG); } - Next, png_struct and png_info need to be allocated and initialized. In order to ensure that the size of these structures is correct even with a dynamically linked libpng, there are functions to initialize and @@ -1153,7 +1154,7 @@ callback function: The PNG specification allows the width and height of an image to be as large as 2^(31\-1 (0x7fffffff), or about 2.147 billion rows and columns. Larger images will be rejected immediately with a png_error() call. If -you wish to reduce these limits, you can use +you wish to change these limits, you can use png_set_user_limits(png_ptr, width_max, height_max); @@ -1768,13 +1769,13 @@ in until png_read_end() has read the chunk data following the image. the PNG datastream is embedded in a MNG-1.0 datastream) - Any or all of interlace_type, compression_type, or - filter_method can be NULL if you are - not interested in their values. + Any of width, height, color_type, bit_depth, + interlace_type, compression_type, or filter_method can + be NULL if you are not interested in their values. Note that png_get_IHDR() returns 32-bit data into the application's width and height variables. - This is an unsafe situation if these are 16-bit + This is an unsafe situation if these are not png_uint_32 variables. In such situations, the png_get_image_width() and png_get_image_height() functions described below are safer. @@ -2497,7 +2498,7 @@ value when you call it in this position: png_set_gamma(png_ptr, screen_gamma, 0.45455); If you need to reduce an RGB file to a paletted file, or if a paletted -file has more entries then will fit on your screen, png_set_quantize() +file has more entries than will fit on your screen, png_set_quantize() will do that. Note that this is a simple match quantization that merely finds the closest color available. This should work fairly well with optimized palettes, but fairly badly with linear color cubes. If you @@ -3050,7 +3051,7 @@ png_infop info_ptr; 64K. The library seems to run fine with sizes of 4K. Although you can give it much less if necessary (I assume you can give it chunks of - 1 byte, I haven't tried less then 256 bytes + 1 byte, I haven't tried less than 256 bytes yet). When this function returns, you may want to display any rows that were generated in the row callback if you don't already do @@ -5535,6 +5536,10 @@ The signatures of many exported functions were changed, such that png_infop became png_inforp or png_const_inforp where "rp" indicates a "restricted pointer". +The support for FAR/far types has been eliminated and the definition of +png_alloc_size_t is now controlled by a flag so that 'small size_t' systems +can select it if necessary. + Error detection in some chunks has improved; in particular the iCCP chunk reader now does pretty complete validation of the basic format. Some bad profiles that were previously accepted are now accepted with a warning or @@ -5549,7 +5554,7 @@ means of PNG_OPTION_ON); #endif -It's not a good idea to do this if you are using the new "simplified API", +It's not a good idea to do this if you are using the "simplified API", which needs to be able to recognize sRGB profiles conveyed via the iCCP chunk. @@ -5778,13 +5783,13 @@ Other rules can be inferred by inspecting the libpng source. .SH XVI. Y2K Compliance in libpng -December 22, 2014 +March 26, 2015 Since the PNG Development group is an ad-hoc body, we can't make an official declaration. This is your unofficial assurance that libpng from version 0.71 and -upward through 1.6.16 are Y2K compliant. It is my belief that earlier +upward through 1.6.17 are Y2K compliant. It is my belief that earlier versions were also Y2K compliant. Libpng only has two year fields. One is a 2-byte unsigned integer @@ -6036,6 +6041,9 @@ the first widely used release: 1.6.16beta01-03 16 10616 16.so.16.16[.0] 1.6.16rc01-02 16 10616 16.so.16.16[.0] 1.6.16 16 10616 16.so.16.16[.0] + 1.6.17beta01-06 16 10617 16.so.16.17[.0] + 1.6.17rc01-06 16 10617 16.so.16.17[.0] + 1.6.17 16 10617 16.so.16.17[.0] Henceforth the source version will match the shared-library minor and patch numbers; the shared-library major version number will be @@ -6092,7 +6100,7 @@ possible without all of you. Thanks to Frank J. T. Wojcik for helping with the documentation. -Libpng version 1.6.16 - December 22, 2014: +Libpng version 1.6.17 - March 26, 2015: Initially created in 1995 by Guy Eric Schalnat, then of Group 42, Inc. Currently maintained by Glenn Randers-Pehrson (glennrp at users.sourceforge.net). @@ -6115,8 +6123,8 @@ this sentence. This code is released under the libpng license. -libpng versions 1.2.6, August 15, 2004, through 1.6.16, December 22, 2014, are -Copyright (c) 2004,2006-2014 Glenn Randers-Pehrson, and are +libpng versions 1.2.6, August 15, 2004, through 1.6.17, March 26, 2015, are +Copyright (c) 2004,2006-2015 Glenn Randers-Pehrson, and are distributed according to the same disclaimer and license as libpng-1.2.5 with the following individual added to the list of Contributing Authors @@ -6214,7 +6222,7 @@ certification mark of the Open Source Initiative. Glenn Randers-Pehrson glennrp at users.sourceforge.net -December 22, 2014 +March 26, 2015 .\" end of man page diff --git a/libpngpf.3 b/libpngpf.3 index 7d489130b..d0dd5be57 100644 --- a/libpngpf.3 +++ b/libpngpf.3 @@ -1,6 +1,6 @@ -.TH LIBPNGPF 3 "December 22, 2014" +.TH LIBPNGPF 3 "March 26, 2015" .SH NAME -libpng \- Portable Network Graphics (PNG) Reference Library 1.6.16 +libpng \- Portable Network Graphics (PNG) Reference Library 1.6.17 (private functions) .SH SYNOPSIS \fB#include \fI"pngpriv.h" @@ -1,4 +1,4 @@ -.TH PNG 5 "December 22, 2014" +.TH PNG 5 "March 26, 2015" .SH NAME png \- Portable Network Graphics (PNG) format .SH DESCRIPTION @@ -1,8 +1,8 @@ /* png.c - location for general purpose libpng functions * - * Last changed in libpng 1.6.16 [December 22, 2014] - * Copyright (c) 1998-2014 Glenn Randers-Pehrson + * Last changed in libpng 1.6.17 [March 26, 2015] + * Copyright (c) 1998-2015 Glenn Randers-Pehrson * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger) * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.) * @@ -14,7 +14,7 @@ #include "pngpriv.h" /* Generate a compiler error if there is an old png.h in the search path. */ -typedef png_libpng_version_1_6_16 Your_png_h_is_not_version_1_6_16; +typedef png_libpng_version_1_6_17 Your_png_h_is_not_version_1_6_17; /* Tells libpng that we have already handled the first "num_bytes" bytes * of the PNG file signature. If the PNG data is embedded into another @@ -34,7 +34,7 @@ png_set_sig_bytes(png_structrp png_ptr, int num_bytes) if (num_bytes > 8) png_error(png_ptr, "Too many bytes for PNG signature"); - png_ptr->sig_bytes = (png_byte)(num_bytes < 0 ? 0 : num_bytes); + png_ptr->sig_bytes = (png_byte)((num_bytes < 0 ? 0 : num_bytes) & 0xff); } /* Checks whether the supplied bytes match the PNG signature. We allow @@ -140,8 +140,10 @@ png_calculate_crc(png_structrp png_ptr, png_const_bytep ptr, png_size_t length) do { uInt safe_length = (uInt)length; +#ifndef __COVERITY__ if (safe_length == 0) safe_length = (uInt)-1; /* evil, but safe */ +#endif crc = crc32(crc, ptr, safe_length); @@ -476,9 +478,10 @@ png_free_data(png_const_structrp png_ptr, png_inforp info_ptr, png_uint_32 mask, /* Free any tRNS entry */ if (((mask & PNG_FREE_TRNS) & info_ptr->free_me) != 0) { + info_ptr->valid &= ~PNG_INFO_tRNS; png_free(png_ptr, info_ptr->trans_alpha); info_ptr->trans_alpha = NULL; - info_ptr->valid &= ~PNG_INFO_tRNS; + info_ptr->num_trans = 0; } #endif @@ -544,20 +547,17 @@ png_free_data(png_const_structrp png_ptr, png_inforp info_ptr, png_uint_32 mask, else { - if (info_ptr->splt_palettes_num != 0) - { - int i; - - for (i = 0; i < info_ptr->splt_palettes_num; i++) - { - png_free(png_ptr, info_ptr->splt_palettes[i].name); - png_free(png_ptr, info_ptr->splt_palettes[i].entries); - } + int i; - png_free(png_ptr, info_ptr->splt_palettes); - info_ptr->splt_palettes = NULL; - info_ptr->splt_palettes_num = 0; + for (i = 0; i < info_ptr->splt_palettes_num; i++) + { + png_free(png_ptr, info_ptr->splt_palettes[i].name); + png_free(png_ptr, info_ptr->splt_palettes[i].entries); } + + png_free(png_ptr, info_ptr->splt_palettes); + info_ptr->splt_palettes = NULL; + info_ptr->splt_palettes_num = 0; info_ptr->valid &= ~PNG_INFO_sPLT; } } @@ -577,15 +577,12 @@ png_free_data(png_const_structrp png_ptr, png_inforp info_ptr, png_uint_32 mask, { int i; - if (info_ptr->unknown_chunks_num != 0) - { - for (i = 0; i < info_ptr->unknown_chunks_num; i++) - png_free(png_ptr, info_ptr->unknown_chunks[i].data); + for (i = 0; i < info_ptr->unknown_chunks_num; i++) + png_free(png_ptr, info_ptr->unknown_chunks[i].data); - png_free(png_ptr, info_ptr->unknown_chunks); - info_ptr->unknown_chunks = NULL; - info_ptr->unknown_chunks_num = 0; - } + png_free(png_ptr, info_ptr->unknown_chunks); + info_ptr->unknown_chunks = NULL; + info_ptr->unknown_chunks_num = 0; } } #endif @@ -666,7 +663,7 @@ png_init_io(png_structrp png_ptr, png_FILE_p fp) } # endif -#ifdef PNG_SAVE_INT_32_SUPPORTED +# ifdef PNG_SAVE_INT_32_SUPPORTED /* The png_save_int_32 function assumes integers are stored in two's * complement format. If this isn't the case, then this routine needs to * be modified to write data in two's complement format. Note that, @@ -681,7 +678,7 @@ png_save_int_32(png_bytep buf, png_int_32 i) buf[2] = (png_byte)((i >> 8) & 0xff); buf[3] = (png_byte)(i & 0xff); } -#endif +# endif # ifdef PNG_TIME_RFC1123_SUPPORTED /* Convert the supplied time into an RFC 1123 string suitable for use in @@ -734,7 +731,7 @@ png_convert_to_rfc1123_buffer(char out[29], png_const_timep ptime) return 1; } -# if PNG_LIBPNG_VER < 10700 +# if PNG_LIBPNG_VER < 10700 /* To do: remove the following from libpng-1.7 */ /* Original API that uses a private buffer in png_struct. * Deprecated because it causes png_struct to carry a spurious temporary @@ -755,7 +752,7 @@ png_convert_to_rfc1123(png_structrp png_ptr, png_const_timep ptime) return NULL; } -# endif +# endif /* LIBPNG_VER < 10700 */ # endif /* TIME_RFC1123 */ #endif /* READ || WRITE */ @@ -769,14 +766,14 @@ png_get_copyright(png_const_structrp png_ptr) #else # ifdef __STDC__ return PNG_STRING_NEWLINE \ - "libpng version 1.6.16 - December 22, 2014" PNG_STRING_NEWLINE \ - "Copyright (c) 1998-2014 Glenn Randers-Pehrson" PNG_STRING_NEWLINE \ + "libpng version 1.6.17 - March 26, 2015" PNG_STRING_NEWLINE \ + "Copyright (c) 1998-2015 Glenn Randers-Pehrson" PNG_STRING_NEWLINE \ "Copyright (c) 1996-1997 Andreas Dilger" PNG_STRING_NEWLINE \ "Copyright (c) 1995-1996 Guy Eric Schalnat, Group 42, Inc." \ PNG_STRING_NEWLINE; # else - return "libpng version 1.6.16 - December 22, 2014\ - Copyright (c) 1998-2014 Glenn Randers-Pehrson\ + return "libpng version 1.6.17 - March 26, 2015\ + Copyright (c) 1998-2015 Glenn Randers-Pehrson\ Copyright (c) 1996-1997 Andreas Dilger\ Copyright (c) 1995-1996 Guy Eric Schalnat, Group 42, Inc."; # endif @@ -872,9 +869,9 @@ png_build_grayscale_palette(int bit_depth, png_colorp palette) for (i = 0, v = 0; i < num_palette; i++, v += color_inc) { - palette[i].red = (png_byte)v; - palette[i].green = (png_byte)v; - palette[i].blue = (png_byte)v; + palette[i].red = (png_byte)(v & 0xff); + palette[i].green = (png_byte)(v & 0xff); + palette[i].blue = (png_byte)(v & 0xff); } } #endif @@ -947,8 +944,6 @@ png_access_version_number(void) return((png_uint_32)PNG_LIBPNG_VER); } - - #if defined(PNG_READ_SUPPORTED) || defined(PNG_WRITE_SUPPORTED) /* Ensure that png_ptr->zstream.msg holds some appropriate error message string. * If it doesn't 'ret' is used to set it to something appropriate, even in cases @@ -1181,7 +1176,7 @@ png_colorspace_sync(png_const_structrp png_ptr, png_inforp info_ptr) png_colorspace_sync_info(png_ptr, info_ptr); } #endif -#endif +#endif /* GAMMA */ #ifdef PNG_COLORSPACE_SUPPORTED /* Added at libpng-1.5.5 to support read and write of true CIEXYZ values for @@ -2166,7 +2161,8 @@ png_icc_check_tag_table(png_const_structrp png_ptr, png_colorspacerp colorspace, return 1; /* success, maybe with warnings */ } -#if defined(PNG_sRGB_SUPPORTED) && PNG_sRGB_PROFILE_CHECKS >= 0 +#ifdef PNG_sRGB_SUPPORTED +#if PNG_sRGB_PROFILE_CHECKS >= 0 /* Information about the known ICC sRGB profiles */ static const struct { @@ -2324,8 +2320,8 @@ png_compare_ICC_profile_with_sRGB(png_const_structrp png_ptr, */ else if (png_sRGB_checks[i].have_md5 == 0) { - png_chunk_report(png_ptr, "out-of-date sRGB profile with" - " no signature", + png_chunk_report(png_ptr, + "out-of-date sRGB profile with no signature", PNG_CHUNK_WARNING); } @@ -2338,8 +2334,8 @@ png_compare_ICC_profile_with_sRGB(png_const_structrp png_ptr, * way. This probably indicates a data error or uninformed hacking. * Fall through to "no match". */ - png_chunk_report(png_ptr, "Not recognizing known sRGB profile that" - " has been edited", + png_chunk_report(png_ptr, + "Not recognizing known sRGB profile that has been edited", PNG_CHUNK_WARNING); break; # endif @@ -2349,9 +2345,8 @@ png_compare_ICC_profile_with_sRGB(png_const_structrp png_ptr, return 0; /* no match */ } -#endif +#endif /* PNG_sRGB_PROFILE_CHECKS >= 0 */ -#ifdef PNG_sRGB_SUPPORTED void /* PRIVATE */ png_icc_set_sRGB(png_const_structrp png_ptr, png_colorspacerp colorspace, png_const_bytep profile, uLong adler) @@ -2365,7 +2360,7 @@ png_icc_set_sRGB(png_const_structrp png_ptr, (void)png_colorspace_set_sRGB(png_ptr, colorspace, (int)/*already checked*/png_get_uint_32(profile+64)); } -#endif /* READ_sRGB */ +#endif /* sRGB */ int /* PRIVATE */ png_colorspace_set_ICC(png_const_structrp png_ptr, png_colorspacerp colorspace, @@ -2457,7 +2452,7 @@ png_colorspace_set_rgb_coefficients(png_structrp png_ptr) png_error(png_ptr, "internal error handling cHRM->XYZ"); } } -#endif +#endif /* READ_RGB_TO_GRAY */ #endif /* COLORSPACE */ @@ -2486,18 +2481,19 @@ png_check_IHDR(png_const_structrp png_ptr, png_warning(png_ptr, "Image width is zero in IHDR"); error = 1; } - else if (width > PNG_UINT_31_MAX) + + if (width > PNG_UINT_31_MAX) { png_warning(png_ptr, "Invalid image width in IHDR"); error = 1; } - else if (png_gt(width, - (PNG_SIZE_MAX >> 3) /* 8-byte RGBA pixels */ - - 48 /* big_row_buf hack */ - - 1 /* filter byte */ - - 7*8 /* rounding width to multiple of 8 pix */ - - 8)) /* extra max_pixel_depth pad */ + if (png_gt(((width + 7) & (~7)), + ((PNG_SIZE_MAX + - 48 /* big_row_buf hack */ + - 1) /* filter byte */ + / 8) /* 8-byte RGBA pixels */ + - 1)) /* extra max_pixel_depth pad */ { /* The size of the row must be within the limits of this architecture. * Because the read code can perform arbitrary transformations the @@ -2513,17 +2509,15 @@ png_check_IHDR(png_const_structrp png_ptr, png_warning(png_ptr, "Image width is too large for this architecture"); error = 1; } - else + +#ifdef PNG_SET_USER_LIMITS_SUPPORTED + if (width > png_ptr->user_width_max) +#else + if (width > PNG_USER_WIDTH_MAX) +#endif { -# ifdef PNG_SET_USER_LIMITS_SUPPORTED - if (width > png_ptr->user_width_max) -# else - if (width > PNG_USER_WIDTH_MAX) -# endif - { - png_warning(png_ptr, "Image width exceeds user limit in IHDR"); - error = 1; - } + png_warning(png_ptr, "Image width exceeds user limit in IHDR"); + error = 1; } if (height == 0) @@ -2531,22 +2525,21 @@ png_check_IHDR(png_const_structrp png_ptr, png_warning(png_ptr, "Image height is zero in IHDR"); error = 1; } - else if (height > PNG_UINT_31_MAX) + + if (height > PNG_UINT_31_MAX) { png_warning(png_ptr, "Invalid image height in IHDR"); error = 1; } - else + +#ifdef PNG_SET_USER_LIMITS_SUPPORTED + if (height > png_ptr->user_height_max) +#else + if (height > PNG_USER_HEIGHT_MAX) +#endif { -# ifdef PNG_SET_USER_LIMITS_SUPPORTED - if (height > png_ptr->user_height_max) -# else - if (height > PNG_USER_HEIGHT_MAX) -# endif - { - png_warning(png_ptr, "Image height exceeds user limit in IHDR"); - error = 1; - } + png_warning(png_ptr, "Image height exceeds user limit in IHDR"); + error = 1; } /* Check other values */ @@ -2585,7 +2578,7 @@ png_check_IHDR(png_const_structrp png_ptr, error = 1; } -# ifdef PNG_MNG_FEATURES_SUPPORTED +#ifdef PNG_MNG_FEATURES_SUPPORTED /* Accept filter_method 64 (intrapixel differencing) only if * 1. Libpng was compiled with PNG_MNG_FEATURES_SUPPORTED and * 2. Libpng did not read a PNG signature (this filter_method is only @@ -2618,13 +2611,13 @@ png_check_IHDR(png_const_structrp png_ptr, } } -# else +#else if (filter_type != PNG_FILTER_TYPE_BASE) { png_warning(png_ptr, "Unknown filter method in IHDR"); error = 1; } -# endif +#endif if (error == 1) png_error(png_ptr, "Invalid IHDR data"); @@ -3041,7 +3034,7 @@ png_ascii_from_fp(png_const_structrp png_ptr, png_charp ascii, png_size_t size, /* Check for an exponent, if we don't need one we are * done and just need to terminate the string. At * this point exp_b10==(-1) is effectively if flag - it got - * to '-1' because of the decrement after outputing + * to '-1' because of the decrement after outputting * the decimal point above (the exponent required is * *not* -1!) */ @@ -3049,7 +3042,7 @@ png_ascii_from_fp(png_const_structrp png_ptr, png_charp ascii, png_size_t size, { /* The following only happens if we didn't output the * leading zeros above for negative exponent, so this - * doest add to the digit requirement. Note that the + * doesn't add to the digit requirement. Note that the * two zeros here can only be output if the two leading * zeros were *not* output, so this doesn't increase * the output count. @@ -3206,7 +3199,7 @@ png_ascii_from_fixed(png_const_structrp png_ptr, png_charp ascii, png_error(png_ptr, "ASCII conversion buffer too small"); } # endif /* FIXED_POINT */ -#endif /* READ_SCAL */ +#endif /* SCAL */ #if defined(PNG_FLOATING_POINT_SUPPORTED) && \ !defined(PNG_FIXED_POINT_MACRO_SUPPORTED) && \ @@ -3224,7 +3217,7 @@ png_fixed(png_const_structrp png_ptr, double fp, png_const_charp text) png_fixed_error(png_ptr, text); # ifndef PNG_ERROR_TEXT_SUPPORTED - PNG_UNUSED(text) + PNG_UNUSED(text) # endif return (png_fixed_point)r; @@ -3405,29 +3398,29 @@ png_gamma_significant(png_fixed_point gamma_val) #endif #ifdef PNG_READ_GAMMA_SUPPORTED -#if defined(PNG_16BIT_SUPPORTED) || !defined(PNG_FLOATING_ARITHMETIC_SUPPORTED) +#ifdef PNG_16BIT_SUPPORTED /* A local convenience routine. */ static png_fixed_point png_product2(png_fixed_point a, png_fixed_point b) { /* The required result is 1/a * 1/b; the following preserves accuracy. */ -# ifdef PNG_FLOATING_ARITHMETIC_SUPPORTED +#ifdef PNG_FLOATING_ARITHMETIC_SUPPORTED double r = a * 1E-5; r *= b; r = floor(r+.5); if (r <= 2147483647. && r >= -2147483648.) return (png_fixed_point)r; -# else +#else png_fixed_point res; if (png_muldiv(&res, a, b, 100000) != 0) return res; -# endif +#endif return 0; /* overflow */ } -#endif /* 16BIT || !FLOATING_ARITHMETIC */ +#endif /* 16BIT */ /* The inverse of the above. */ png_fixed_point @@ -3435,12 +3428,15 @@ png_reciprocal2(png_fixed_point a, png_fixed_point b) { /* The required result is 1/a * 1/b; the following preserves accuracy. */ #ifdef PNG_FLOATING_ARITHMETIC_SUPPORTED - double r = 1E15/a; - r /= b; - r = floor(r+.5); + if (a != 0 && b != 0) + { + double r = 1E15/a; + r /= b; + r = floor(r+.5); - if (r <= 2147483647. && r >= -2147483648.) - return (png_fixed_point)r; + if (r <= 2147483647. && r >= -2147483648.) + return (png_fixed_point)r; + } #else /* This may overflow because the range of png_fixed_point isn't symmetric, * but this API is only used for the product of file and screen gamma so it @@ -3731,7 +3727,7 @@ png_exp8bit(png_fixed_point lg2) * step. */ x -= x >> 8; - return (png_byte)((x + 0x7fffffU) >> 24); + return (png_byte)(((x + 0x7fffffU) >> 24) & 0xff); } #ifdef PNG_16BIT_SUPPORTED @@ -3792,7 +3788,7 @@ png_gamma_8bit_correct(unsigned int value, png_fixed_point gamma_val) # endif } - return (png_byte)value; + return (png_byte)(value & 0xff); } #ifdef PNG_16BIT_SUPPORTED @@ -4014,7 +4010,7 @@ png_build_8bit_table(png_structrp png_ptr, png_bytepp ptable, else for (i=0; i<256; ++i) - table[i] = (png_byte)i; + table[i] = (png_byte)(i & 0xff); } /* Used from png_read_destroy and below to release the memory used by the gamma @@ -4154,7 +4150,8 @@ png_build_gamma_table(png_structrp png_ptr, int bit_depth) * */ if (sig_bit > 0 && sig_bit < 16U) - shift = (png_byte)(16U - sig_bit); /* shift == insignificant bits */ + /* shift == insignificant bits */ + shift = (png_byte)((16U - sig_bit) & 0xff); else shift = 0; /* keep all 16 bits */ @@ -4223,7 +4220,7 @@ png_set_option(png_structrp png_ptr, int option, int onoff) int setting = (2 + (onoff != 0)) << option; int current = png_ptr->options; - png_ptr->options = (png_byte)((current & ~mask) | setting); + png_ptr->options = (png_byte)(((current & ~mask) | setting) & 0xff); return (current & mask) >> option; } @@ -1,8 +1,9 @@ /* png.h - header file for PNG reference library * - * libpng version 1.6.16, December 22, 2014 - * Copyright (c) 1998-2014 Glenn Randers-Pehrson + * libpng version 1.6.17, March 26, 2015 + * + * Copyright (c) 1998-2015 Glenn Randers-Pehrson * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger) * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.) * @@ -11,7 +12,7 @@ * Authors and maintainers: * libpng versions 0.71, May 1995, through 0.88, January 1996: Guy Schalnat * libpng versions 0.89c, June 1996, through 0.96, May 1997: Andreas Dilger - * libpng versions 0.97, January 1998, through 1.6.16, December 22, 2014: Glenn + * libpng versions 0.97, January 1998, through 1.6.17, March 26, 2015: Glenn * See also "Contributing Authors", below. * * Note about libpng version numbers: @@ -212,6 +213,9 @@ * 1.6.16beta01-03 16 10616 16.so.16.16[.0] * 1.6.16rc01-02 16 10616 16.so.16.16[.0] * 1.6.16 16 10616 16.so.16.16[.0] + * 1.6.17beta01-06 16 10617 16.so.16.17[.0] + * 1.6.17rc01-06 16 10617 16.so.16.17[.0] + * 1.6.17 16 10617 16.so.16.17[.0] * * Henceforth the source version will match the shared-library major * and minor numbers; the shared-library major version number will be @@ -243,8 +247,8 @@ * * This code is released under the libpng license. * - * libpng versions 1.2.6, August 15, 2004, through 1.6.16, December 22, 2014, are - * Copyright (c) 2004, 2006-2014 Glenn Randers-Pehrson, and are + * libpng versions 1.2.6, August 15, 2004, through 1.6.17, March 26, 2015, are + * Copyright (c) 2004, 2006-2015 Glenn Randers-Pehrson, and are * distributed according to the same disclaimer and license as libpng-1.2.5 * with the following individual added to the list of Contributing Authors: * @@ -355,13 +359,13 @@ * Y2K compliance in libpng: * ========================= * - * December 22, 2014 + * March 26, 2015 * * Since the PNG Development group is an ad-hoc body, we can't make * an official declaration. * * This is your unofficial assurance that libpng from version 0.71 and - * upward through 1.6.16 are Y2K compliant. It is my belief that + * upward through 1.6.17 are Y2K compliant. It is my belief that * earlier versions were also Y2K compliant. * * Libpng only has two year fields. One is a 2-byte unsigned integer @@ -423,9 +427,9 @@ */ /* Version information for png.h - this should match the version in png.c */ -#define PNG_LIBPNG_VER_STRING "1.6.16" +#define PNG_LIBPNG_VER_STRING "1.6.17" #define PNG_HEADER_VERSION_STRING \ - " libpng version 1.6.16 - December 22, 2014\n" + " libpng version 1.6.17 - March 26, 2015\n" #define PNG_LIBPNG_VER_SONUM 16 #define PNG_LIBPNG_VER_DLLNUM 16 @@ -433,7 +437,7 @@ /* These should match the first 3 components of PNG_LIBPNG_VER_STRING: */ #define PNG_LIBPNG_VER_MAJOR 1 #define PNG_LIBPNG_VER_MINOR 6 -#define PNG_LIBPNG_VER_RELEASE 16 +#define PNG_LIBPNG_VER_RELEASE 17 /* This should match the numeric part of the final component of * PNG_LIBPNG_VER_STRING, omitting any leading zero: @@ -464,7 +468,7 @@ * version 1.0.0 was mis-numbered 100 instead of 10000). From * version 1.0.1 it's xxyyzz, where x=major, y=minor, z=release */ -#define PNG_LIBPNG_VER 10616 /* 1.6.16 */ +#define PNG_LIBPNG_VER 10617 /* 1.6.17 */ /* Library configuration: these options cannot be changed after * the library has been built. @@ -569,7 +573,7 @@ extern "C" { /* This triggers a compiler error in png.c, if png.c and png.h * do not agree upon the version number. */ -typedef char* png_libpng_version_1_6_16; +typedef char* png_libpng_version_1_6_17; /* Basic control structions. Read libpng-manual.txt or libpng.3 for more info. * @@ -1578,6 +1582,7 @@ PNG_EXPORT(66, void, png_set_crc_action, (png_structrp png_ptr, int crit_action, #define PNG_CRC_QUIET_USE 4 /* quiet/use data quiet/use data */ #define PNG_CRC_NO_CHANGE 5 /* use current value use current value */ +#ifdef PNG_WRITE_SUPPORTED /* These functions give the user control over the scan-line filtering in * libpng and the compression methods used by zlib. These functions are * mainly useful for testing, as the defaults should work with most users. @@ -1591,6 +1596,7 @@ PNG_EXPORT(66, void, png_set_crc_action, (png_structrp png_ptr, int crit_action, */ PNG_EXPORT(67, void, png_set_filter, (png_structrp png_ptr, int method, int filters)); +#endif /* WRITE */ /* Flags for png_set_filter() to say which filters to use. The flags * are chosen so that they don't conflict with real filter types @@ -1616,6 +1622,7 @@ PNG_EXPORT(67, void, png_set_filter, (png_structrp png_ptr, int method, #define PNG_FILTER_VALUE_PAETH 4 #define PNG_FILTER_VALUE_LAST 5 +#ifdef PNG_WRITE_SUPPORTED #ifdef PNG_WRITE_WEIGHTED_FILTER_SUPPORTED /* EXPERIMENTAL */ /* The "heuristic_method" is given by one of the PNG_FILTER_HEURISTIC_ * defines, either the default (minimum-sum-of-absolute-differences), or @@ -1662,7 +1669,6 @@ PNG_FIXED_EXPORT(209, void, png_set_filter_heuristics_fixed, #define PNG_FILTER_HEURISTIC_WEIGHTED 2 /* Experimental feature */ #define PNG_FILTER_HEURISTIC_LAST 3 /* Not a valid value */ -#ifdef PNG_WRITE_SUPPORTED /* Set the library compression level. Currently, valid values range from * 0 - 9, corresponding directly to the zlib compression levels 0 - 9 * (0 - no compression, 9 - "maximal" compression). Note that tests have @@ -1670,6 +1676,7 @@ PNG_FIXED_EXPORT(209, void, png_set_filter_heuristics_fixed, * for PNG images, and do considerably fewer caclulations. In the future, * these values may not correspond directly to the zlib compression levels. */ +#ifdef PNG_WRITE_CUSTOMIZE_COMPRESSION_SUPPORTED PNG_EXPORT(69, void, png_set_compression_level, (png_structrp png_ptr, int level)); @@ -1687,7 +1694,7 @@ PNG_EXPORT(72, void, png_set_compression_window_bits, (png_structrp png_ptr, PNG_EXPORT(73, void, png_set_compression_method, (png_structrp png_ptr, int method)); -#endif +#endif /* WRITE_CUSTOMIZE_COMPRESSION */ #ifdef PNG_WRITE_CUSTOMIZE_ZTXT_COMPRESSION_SUPPORTED /* Also set zlib parameters for compressing non-IDAT chunks */ @@ -1709,6 +1716,7 @@ PNG_EXPORT(225, void, png_set_text_compression_window_bits, PNG_EXPORT(226, void, png_set_text_compression_method, (png_structrp png_ptr, int method)); #endif /* WRITE_CUSTOMIZE_ZTXT_COMPRESSION */ +#endif /* WRITE */ /* These next functions are called for input/output, memory, and error * handling. They are in the file pngrio.c, pngwio.c, and pngerror.c, @@ -1819,7 +1827,7 @@ PNG_EXPORT(218, png_byte, png_get_current_pass_number, (png_const_structrp)); * * The integer return from the callback function is interpreted thus: * - * negative: An error occured, png_chunk_error will be called. + * negative: An error occurred; png_chunk_error will be called. * zero: The chunk was not handled, the chunk will be saved. A critical * chunk will cause an error at this point unless it is to be saved. * positive: The chunk was handled, libpng will ignore/discard it. @@ -2664,26 +2672,28 @@ PNG_EXPORT(216, png_uint_32, png_get_io_chunk_type, * (png_uint_16)(alpha) \ + (png_uint_16)(bg)*(png_uint_16)(255 \ - (png_uint_16)(alpha)) + 128); \ - (composite) = (png_byte)((temp + (temp >> 8)) >> 8); } + (composite) = (png_byte)(((temp + (temp >> 8)) >> 8) & 0xff); } # define png_composite_16(composite, fg, alpha, bg) \ { png_uint_32 temp = (png_uint_32)((png_uint_32)(fg) \ * (png_uint_32)(alpha) \ + (png_uint_32)(bg)*(65535 \ - (png_uint_32)(alpha)) + 32768); \ - (composite) = (png_uint_16)((temp + (temp >> 16)) >> 16); } + (composite) = (png_uint_16)(0xffff & ((temp + (temp >> 16)) >> 16)); } #else /* Standard method using integer division */ -# define png_composite(composite, fg, alpha, bg) \ - (composite) = (png_byte)(((png_uint_16)(fg) * (png_uint_16)(alpha) + \ - (png_uint_16)(bg) * (png_uint_16)(255 - (png_uint_16)(alpha)) + \ - 127) / 255) +# define png_composite(composite, fg, alpha, bg) \ + (composite) = \ + (png_byte)(0xff & (((png_uint_16)(fg) * (png_uint_16)(alpha) + \ + (png_uint_16)(bg) * (png_uint_16)(255 - (png_uint_16)(alpha)) + \ + 127) / 255)) # define png_composite_16(composite, fg, alpha, bg) \ - (composite) = (png_uint_16)(((png_uint_32)(fg) * (png_uint_32)(alpha) + \ - (png_uint_32)(bg)*(png_uint_32)(65535 - (png_uint_32)(alpha)) + \ - 32767) / 65535) + (composite) = \ + (png_uint_16)(0xffff & (((png_uint_32)(fg) * (png_uint_32)(alpha) + \ + (png_uint_32)(bg)*(png_uint_32)(65535 - (png_uint_32)(alpha)) + \ + 32767) / 65535)) #endif /* READ_COMPOSITE_NODIV */ #ifdef PNG_READ_INT_FUNCTIONS_SUPPORTED @@ -1,9 +1,9 @@ /* pngconf.h - machine configurable file for libpng * - * libpng version 1.6.16,December 22, 2014 + * libpng version 1.6.17, March 26, 2015 * - * Copyright (c) 1998-2014 Glenn Randers-Pehrson + * Copyright (c) 1998-2015 Glenn Randers-Pehrson * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger) * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.) * @@ -11,9 +11,7 @@ * For conditions of distribution and use, see the disclaimer * and license in png.h * - */ - -/* Any machine specific code is near the front of this file, so if you + * Any machine specific code is near the front of this file, so if you * are configuring libpng for a machine, you may want to read the section * starting here down to where it starts to typedef png_color, png_text, * and png_info. @@ -22,26 +20,6 @@ #ifndef PNGCONF_H #define PNGCONF_H -/* To do: Do all of this in scripts/pnglibconf.dfa */ -#ifdef PNG_SAFE_LIMITS_SUPPORTED -# ifdef PNG_USER_WIDTH_MAX -# undef PNG_USER_WIDTH_MAX -# define PNG_USER_WIDTH_MAX 1000000L -# endif -# ifdef PNG_USER_HEIGHT_MAX -# undef PNG_USER_HEIGHT_MAX -# define PNG_USER_HEIGHT_MAX 1000000L -# endif -# ifdef PNG_USER_CHUNK_MALLOC_MAX -# undef PNG_USER_CHUNK_MALLOC_MAX -# define PNG_USER_CHUNK_MALLOC_MAX 4000000L -# endif -# ifdef PNG_USER_CHUNK_CACHE_MAX -# undef PNG_USER_CHUNK_CACHE_MAX -# define PNG_USER_CHUNK_CACHE_MAX 128 -# endif -#endif - #ifndef PNG_BUILDING_SYMBOL_TABLE /* else includes may cause problems */ /* From libpng 1.6.0 libpng requires an ANSI X3.159-1989 ("ISOC90") compliant C diff --git a/pngdebug.h b/pngdebug.h index b43c59cde..6a01b106e 100644 --- a/pngdebug.h +++ b/pngdebug.h @@ -1,12 +1,11 @@ /* pngdebug.h - Debugging macros for libpng, also used in pngtest.c * + * Last changed in libpng 1.6.8 [December 19, 2013] * Copyright (c) 1998-2013 Glenn Randers-Pehrson * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger) * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.) * - * Last changed in libpng 1.6.8 [December 19, 2013] - * * This code is released under the libpng license. * For conditions of distribution and use, see the disclaimer * and license in png.h @@ -1,8 +1,8 @@ /* pngget.c - retrieval of values from info struct * - * Last changed in libpng 1.6.15 [November 20, 2014] - * Copyright (c) 1998-2014 Glenn Randers-Pehrson + * Last changed in libpng 1.6.17 [March 26, 2015] + * Copyright (c) 1998-2015 Glenn Randers-Pehrson * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger) * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.) * @@ -799,14 +799,20 @@ png_get_IHDR(png_const_structrp png_ptr, png_const_inforp info_ptr, { png_debug1(1, "in %s retrieval function", "IHDR"); - if (png_ptr == NULL || info_ptr == NULL || width == NULL || - height == NULL || bit_depth == NULL || color_type == NULL) + if (png_ptr == NULL || info_ptr == NULL) return (0); - *width = info_ptr->width; - *height = info_ptr->height; - *bit_depth = info_ptr->bit_depth; - *color_type = info_ptr->color_type; + if (width != NULL) + *width = info_ptr->width; + + if (height != NULL) + *height = info_ptr->height; + + if (bit_depth != NULL) + *bit_depth = info_ptr->bit_depth; + + if (color_type != NULL) + *color_type = info_ptr->color_type; if (compression_type != NULL) *compression_type = info_ptr->compression_type; @@ -1135,21 +1141,21 @@ png_get_compression_buffer_size(png_const_structrp png_ptr) if (png_ptr == NULL) return 0; -# ifdef PNG_WRITE_SUPPORTED +#ifdef PNG_WRITE_SUPPORTED if ((png_ptr->mode & PNG_IS_READ_STRUCT) != 0) -# endif +#endif { -# ifdef PNG_SEQUENTIAL_READ_SUPPORTED +#ifdef PNG_SEQUENTIAL_READ_SUPPORTED return png_ptr->IDAT_read_size; -# else +#else return PNG_IDAT_READ_SIZE; -# endif +#endif } -# ifdef PNG_WRITE_SUPPORTED +#ifdef PNG_WRITE_SUPPORTED else return png_ptr->zbuffer_size; -# endif +#endif } #ifdef PNG_SET_USER_LIMITS_SUPPORTED @@ -1,12 +1,11 @@ /* pnginfo.h - header file for PNG reference library * + * Last changed in libpng 1.6.1 [March 28, 2013] * Copyright (c) 1998-2013 Glenn Randers-Pehrson * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger) * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.) * - * Last changed in libpng 1.6.1 [March 28, 2013] - * * This code is released under the libpng license. * For conditions of distribution and use, see the disclaimer * and license in png.h @@ -41,7 +41,7 @@ png_destroy_png_struct(png_structrp png_ptr) } /* Allocate memory. For reasonable files, size should never exceed - * 64K. However, zlib may allocate more then 64K if you don't tell + * 64K. However, zlib may allocate more than 64K if you don't tell * it not to. See zconf.h and png.h for more information. zlib does * need to allocate exactly 64K, so whatever you call here must * have the ability to do that. diff --git a/pngpread.c b/pngpread.c index fb40e7db3..823dcad8c 100644 --- a/pngpread.c +++ b/pngpread.c @@ -1,8 +1,8 @@ /* pngpread.c - read a png file in push mode * - * Last changed in libpng 1.6.15 [November 20, 2014] - * Copyright (c) 1998-2014 Glenn Randers-Pehrson + * Last changed in libpng 1.6.17 [March 26, 2015] + * Copyright (c) 1998-2015 Glenn Randers-Pehrson * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger) * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.) * @@ -726,7 +726,7 @@ png_process_IDAT_data(png_structrp png_ptr, png_bytep buffer, * or the stream marked as finished. */ while (png_ptr->zstream.avail_in > 0 && - !(png_ptr->flags & PNG_FLAG_ZSTREAM_ENDED)) + (png_ptr->flags & PNG_FLAG_ZSTREAM_ENDED) == 0) { int ret; @@ -1036,6 +1036,7 @@ png_push_process_row(png_structrp png_ptr) } } else +#endif { png_push_have_row(png_ptr, png_ptr->row_buf + 1); png_read_push_finish_row(png_ptr); @@ -1045,6 +1046,7 @@ png_push_process_row(png_structrp png_ptr) void /* PRIVATE */ png_read_push_finish_row(png_structrp png_ptr) { +#ifdef PNG_READ_INTERLACING_SUPPORTED /* Arrays to facilitate easy interlacing - use pass (0 - 6) as index */ /* Start of interlace block */ @@ -1069,6 +1071,7 @@ png_read_push_finish_row(png_structrp png_ptr) if (png_ptr->row_number < png_ptr->num_rows) return; +#ifdef PNG_READ_INTERLACING_SUPPORTED if (png_ptr->interlaced != 0) { png_ptr->row_number = 0; @@ -1103,6 +1106,7 @@ png_read_push_finish_row(png_structrp png_ptr) } while (png_ptr->iwidth == 0 || png_ptr->num_rows == 0); } +#endif /* READ_INTERLACING */ } void /* PRIVATE */ @@ -1127,6 +1131,7 @@ png_push_have_row(png_structrp png_ptr, png_bytep row) (int)png_ptr->pass); } +#ifdef PNG_READ_INTERLACING_SUPPORTED void PNGAPI png_progressive_combine_row(png_const_structrp png_ptr, png_bytep old_row, png_const_bytep new_row) @@ -1141,6 +1146,7 @@ png_progressive_combine_row(png_const_structrp png_ptr, png_bytep old_row, if (new_row != NULL) png_combine_row(png_ptr, old_row, 1/*blocky display*/); } +#endif /* READ_INTERLACING */ void PNGAPI png_set_progressive_read_fn(png_structrp png_ptr, png_voidp progressive_ptr, @@ -1,13 +1,11 @@ /* pngpriv.h - private declarations for use inside libpng * - * For conditions of distribution and use, see copyright notice in png.h - * Copyright (c) 1998-2014 Glenn Randers-Pehrson + * Last changed in libpng 1.6.17 [March 26, 2015] + * Copyright (c) 1998-2015 Glenn Randers-Pehrson * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger) * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.) * - * Last changed in libpng 1.6.10 [March 6, 1014]] - * * This code is released under the libpng license. * For conditions of distribution and use, see the disclaimer * and license in png.h @@ -300,46 +298,9 @@ /* SECURITY and SAFETY: * - * By default libpng is built without any internal limits on image size, - * individual heap (png_malloc) allocations or the total amount of memory used. - * If PNG_SAFE_LIMITS_SUPPORTED is defined, however, the limits below are used - * (unless individually overridden). These limits are believed to be fairly - * safe, but builders of secure systems should verify the values against the - * real system capabilities. - */ -#ifdef PNG_SAFE_LIMITS_SUPPORTED - /* 'safe' limits */ -# ifndef PNG_USER_WIDTH_MAX -# define PNG_USER_WIDTH_MAX 1000000 -# endif -# ifndef PNG_USER_HEIGHT_MAX -# define PNG_USER_HEIGHT_MAX 1000000 -# endif -# ifndef PNG_USER_CHUNK_CACHE_MAX -# define PNG_USER_CHUNK_CACHE_MAX 128 -# endif -# ifndef PNG_USER_CHUNK_MALLOC_MAX -# define PNG_USER_CHUNK_MALLOC_MAX 8000000 -# endif -#else - /* values for no limits */ -# ifndef PNG_USER_WIDTH_MAX -# define PNG_USER_WIDTH_MAX 0x7fffffff -# endif -# ifndef PNG_USER_HEIGHT_MAX -# define PNG_USER_HEIGHT_MAX 0x7fffffff -# endif -# ifndef PNG_USER_CHUNK_CACHE_MAX -# define PNG_USER_CHUNK_CACHE_MAX 0 -# endif -# ifndef PNG_USER_CHUNK_MALLOC_MAX -# define PNG_USER_CHUNK_MALLOC_MAX 0 -# endif -#endif - -/* Moved to pngpriv.h at libpng-1.5.0 */ -/* NOTE: some of these may have been used in external applications as - * these definitions were exposed in pngconf.h prior to 1.5. + * libpng is built with support for internal limits on image dimensions and + * memory usage. These are documented in scripts/pnglibconf.dfa of the + * source and recorded in the machine generated header file pnglibconf.h. */ /* If you are running on a machine where you cannot allocate more @@ -582,13 +543,13 @@ #define PNG_RGB_TO_GRAY_WARN 0x400000 #define PNG_RGB_TO_GRAY 0x600000 /* two bits, RGB_TO_GRAY_ERR|WARN */ #define PNG_ENCODE_ALPHA 0x800000 /* Added to libpng-1.5.4 */ -#define PNG_ADD_ALPHA 0x1000000 /* Added to libpng-1.2.7 */ -#define PNG_EXPAND_tRNS 0x2000000 /* Added to libpng-1.2.9 */ -#define PNG_SCALE_16_TO_8 0x4000000 /* Added to libpng-1.5.4 */ - /* 0x8000000 unused */ - /* 0x10000000 unused */ - /* 0x20000000 unused */ - /* 0x40000000 unused */ +#define PNG_ADD_ALPHA 0x1000000 /* Added to libpng-1.2.7 */ +#define PNG_EXPAND_tRNS 0x2000000 /* Added to libpng-1.2.9 */ +#define PNG_SCALE_16_TO_8 0x4000000 /* Added to libpng-1.5.4 */ + /* 0x8000000 unused */ + /* 0x10000000 unused */ + /* 0x20000000 unused */ + /* 0x40000000 unused */ /* Flags for png_create_struct */ #define PNG_STRUCT_PNG 0x0001 #define PNG_STRUCT_INFO 0x0002 @@ -780,15 +741,17 @@ * macro will fail on top-bit-set values because of the sign extension. */ #define PNG_CHUNK_FROM_STRING(s)\ - PNG_U32(0xff&(s)[0], 0xff&(s)[1], 0xff&(s)[2], 0xff&(s)[3]) + PNG_U32(0xff & (s)[0], 0xff & (s)[1], 0xff & (s)[2], 0xff & (s)[3]) /* This uses (char), not (png_byte) to avoid warnings on systems where (char) is * signed and the argument is a (char[]) This macro will fail miserably on * systems where (char) is more than 8 bits. */ #define PNG_STRING_FROM_CHUNK(s,c)\ - (void)(((char*)(s))[0]=(char)((c)>>24), ((char*)(s))[1]=(char)((c)>>16),\ - ((char*)(s))[2]=(char)((c)>>8), ((char*)(s))[3]=(char)((c))) + (void)(((char*)(s))[0]=(char)(((c)>>24) & 0xff), \ + ((char*)(s))[1]=(char)(((c)>>16) & 0xff),\ + ((char*)(s))[2]=(char)(((c)>>8) & 0xff), \ + ((char*)(s))[3]=(char)((c & 0xff))) /* Do the same but terminate with a null character. */ #define PNG_CSTRING_FROM_CHUNK(s,c)\ @@ -850,8 +813,9 @@ PNG_INTERNAL_DATA(const png_uint_16, png_sRGB_table, [256]); PNG_INTERNAL_DATA(const png_uint_16, png_sRGB_base, [512]); PNG_INTERNAL_DATA(const png_byte, png_sRGB_delta, [512]); -#define PNG_sRGB_FROM_LINEAR(linear) ((png_byte)((png_sRGB_base[(linear)>>15] +\ - ((((linear)&0x7fff)*png_sRGB_delta[(linear)>>15])>>12)) >> 8)) +#define PNG_sRGB_FROM_LINEAR(linear) \ + ((png_byte)(0xff & ((png_sRGB_base[(linear)>>15] \ + + ((((linear) & 0x7fff)*png_sRGB_delta[(linear)>>15])>>12)) >> 8))) /* Given a value 'linear' in the range 0..255*65535 calculate the 8-bit sRGB * encoded value with maximum error 0.646365. Note that the input is not a * 16-bit value; it has been multiplied by 255! */ @@ -910,7 +874,7 @@ PNG_INTERNAL_FUNCTION(png_voidp,png_malloc_base,(png_const_structrp png_ptr, #if defined(PNG_TEXT_SUPPORTED) || defined(PNG_sPLT_SUPPORTED) ||\ defined(PNG_STORE_UNKNOWN_CHUNKS_SUPPORTED) /* Internal array allocator, outputs no error or warning messages on failure, - * just returns NULL. + * just returns NULL. */ PNG_INTERNAL_FUNCTION(png_voidp,png_malloc_array,(png_const_structrp png_ptr, int nelements, size_t element_size),PNG_ALLOCATED); @@ -1,8 +1,8 @@ /* pngread.c - read a PNG file * - * Last changed in libpng 1.6.15 [November 20, 2014] - * Copyright (c) 1998-2014 Glenn Randers-Pehrson + * Last changed in libpng 1.6.17 [March 26, 2015] + * Copyright (c) 1998-2015 Glenn Randers-Pehrson * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger) * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.) * @@ -814,8 +814,7 @@ png_read_end(png_structrp png_ptr, png_inforp info_ptr) /* Zero length IDATs are legal after the last IDAT has been * read, but not after other chunks have been read. */ - if ((length > 0) || - (png_ptr->mode & PNG_HAVE_CHUNK_AFTER_IDAT) != 0) + if ((length > 0) || (png_ptr->mode & PNG_HAVE_CHUNK_AFTER_IDAT) != 0) png_benign_error(png_ptr, "Too many IDATs found"); png_crc_finish(png_ptr, length); @@ -1824,6 +1823,7 @@ png_create_colormap_entry(png_image_read_control *display, y = (y + 128) >> 8; y *= 255; y = PNG_sRGB_FROM_LINEAR((y + 64) >> 7); + alpha = PNG_DIV257(alpha); encoding = P_sRGB; } @@ -2286,8 +2286,14 @@ png_image_read_colormap(png_voidp argument) output_processing = PNG_CMAP_NONE; break; } - +#ifdef __COVERITY__ + /* Coverity claims that output_encoding cannot be 2 (P_LINEAR) + * here. + */ + back_alpha = 255; +#else back_alpha = output_encoding == P_LINEAR ? 65535 : 255; +#endif } /* output_processing means that the libpng-processed row will be @@ -2412,7 +2418,14 @@ png_image_read_colormap(png_voidp argument) */ background_index = i; png_create_colormap_entry(display, i++, back_r, back_g, back_b, - output_encoding == P_LINEAR ? 65535U : 255U, output_encoding); +#ifdef __COVERITY__ + /* Coverity claims that output_encoding cannot be 2 (P_LINEAR) + * here. + */ 255U, +#else + output_encoding == P_LINEAR ? 65535U : 255U, +#endif + output_encoding); /* For non-opaque input composite on the sRGB background - this * requires inverting the encoding for each component. The input @@ -3246,7 +3259,7 @@ png_image_read_composite(png_voidp argument) png_uint_32 width = image->width; ptrdiff_t step_row = display->row_bytes; unsigned int channels = - (image->format & PNG_FORMAT_FLAG_COLOR) != 0 ? 3 : 1; + (image->format & PNG_FORMAT_FLAG_COLOR) != 0 ? 3 : 1; int pass; for (pass = 0; pass < passes; ++pass) @@ -1,8 +1,8 @@ /* pngrio.c - functions for data input * - * Last changed in libpng 1.6.15 [November 20, 2014] - * Copyright (c) 1998-2014 Glenn Randers-Pehrson + * Last changed in libpng 1.6.17 [March 26, 2015] + * Copyright (c) 1998-2015 Glenn Randers-Pehrson * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger) * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.) * @@ -26,7 +26,7 @@ * reads from a file pointer. Note that this routine sometimes gets called * with very small lengths, so you should implement some kind of simple * buffering if you are using unbuffered reads. This should never be asked - * to read more then 64K on a 16 bit machine. + * to read more than 64K on a 16 bit machine. */ void /* PRIVATE */ png_read_data(png_structrp png_ptr, png_bytep data, png_size_t length) diff --git a/pngrtran.c b/pngrtran.c index e6fe9fdbf..cad7a8daa 100644 --- a/pngrtran.c +++ b/pngrtran.c @@ -1,8 +1,8 @@ /* pngrtran.c - transforms the data in a row for PNG readers * - * Last changed in libpng 1.6.15 [November 20, 2014] - * Copyright (c) 1998-2014 Glenn Randers-Pehrson + * Last changed in libpng 1.6.17 [March 26, 2015] + * Copyright (c) 1998-2015 Glenn Randers-Pehrson * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger) * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.) * @@ -394,7 +394,7 @@ png_set_alpha_mode(png_structrp png_ptr, int mode, double output_gamma) /* Dither file to 8-bit. Supply a palette, the current number * of elements in the palette, the maximum number of elements * allowed, and a histogram if possible. If the current number - * of colors is greater then the maximum number, the palette will be + * of colors is greater than the maximum number, the palette will be * modified to fit in the maximum number. "full_quantize" indicates * whether we need a quantizing cube set up for RGB images, or if we * simply are reducing the number of colors in a paletted image. @@ -2357,7 +2357,7 @@ png_do_unshift(png_row_infop row_info, png_bytep row, if (++channel >= channels) channel = 0; *bp++ = (png_byte)(value >> 8); - *bp++ = (png_byte)(value & 0xff); + *bp++ = (png_byte)value; } break; } @@ -2662,9 +2662,9 @@ png_do_read_filler(png_row_infop row_info, png_bytep row, png_uint_32 row_width = row_info->width; #ifdef PNG_READ_16BIT_SUPPORTED - png_byte hi_filler = (png_byte)((filler>>8) & 0xff); + png_byte hi_filler = (png_byte)(filler>>8); #endif - png_byte lo_filler = (png_byte)(filler & 0xff); + png_byte lo_filler = (png_byte)filler; png_debug(1, "in png_do_read_filler"); @@ -2715,13 +2715,13 @@ png_do_read_filler(png_row_infop row_info, png_bytep row, png_bytep dp = sp + (png_size_t)row_width * 2; for (i = 1; i < row_width; i++) { - *(--dp) = hi_filler; *(--dp) = lo_filler; + *(--dp) = hi_filler; *(--dp) = *(--sp); *(--dp) = *(--sp); } - *(--dp) = hi_filler; *(--dp) = lo_filler; + *(--dp) = hi_filler; row_info->channels = 2; row_info->pixel_depth = 32; row_info->rowbytes = row_width * 4; @@ -2736,8 +2736,8 @@ png_do_read_filler(png_row_infop row_info, png_bytep row, { *(--dp) = *(--sp); *(--dp) = *(--sp); - *(--dp) = hi_filler; *(--dp) = lo_filler; + *(--dp) = hi_filler; } row_info->channels = 2; row_info->pixel_depth = 32; @@ -2796,8 +2796,8 @@ png_do_read_filler(png_row_infop row_info, png_bytep row, png_bytep dp = sp + (png_size_t)row_width * 2; for (i = 1; i < row_width; i++) { - *(--dp) = hi_filler; *(--dp) = lo_filler; + *(--dp) = hi_filler; *(--dp) = *(--sp); *(--dp) = *(--sp); *(--dp) = *(--sp); @@ -2805,8 +2805,8 @@ png_do_read_filler(png_row_infop row_info, png_bytep row, *(--dp) = *(--sp); *(--dp) = *(--sp); } - *(--dp) = hi_filler; *(--dp) = lo_filler; + *(--dp) = hi_filler; row_info->channels = 4; row_info->pixel_depth = 64; row_info->rowbytes = row_width * 8; @@ -2825,8 +2825,8 @@ png_do_read_filler(png_row_infop row_info, png_bytep row, *(--dp) = *(--sp); *(--dp) = *(--sp); *(--dp) = *(--sp); - *(--dp) = hi_filler; *(--dp) = lo_filler; + *(--dp) = hi_filler; } row_info->channels = 4; @@ -3087,10 +3087,11 @@ png_do_rgb_to_gray(png_structrp png_ptr, png_row_infop row_info, png_bytep row) for (i = 0; i < row_width; i++) { png_uint_16 red, green, blue, w; + png_byte hi,lo; - red = (png_uint_16)(((*(sp)) << 8) | *(sp + 1)); sp += 2; - green = (png_uint_16)(((*(sp)) << 8) | *(sp + 1)); sp += 2; - blue = (png_uint_16)(((*(sp)) << 8) | *(sp + 1)); sp += 2; + hi=*(sp)++; lo=*(sp)++; red = (png_uint_16)((hi << 8) | (lo)); + hi=*(sp)++; lo=*(sp)++; green = (png_uint_16)((hi << 8) | (lo)); + hi=*(sp)++; lo=*(sp)++; blue = (png_uint_16)((hi << 8) | (lo)); if (red == green && red == blue) { @@ -3104,16 +3105,16 @@ png_do_rgb_to_gray(png_structrp png_ptr, png_row_infop row_info, png_bytep row) else { - png_uint_16 red_1 = png_ptr->gamma_16_to_1[(red&0xff) + png_uint_16 red_1 = png_ptr->gamma_16_to_1[(red & 0xff) >> png_ptr->gamma_shift][red>>8]; png_uint_16 green_1 = - png_ptr->gamma_16_to_1[(green&0xff) >> + png_ptr->gamma_16_to_1[(green & 0xff) >> png_ptr->gamma_shift][green>>8]; - png_uint_16 blue_1 = png_ptr->gamma_16_to_1[(blue&0xff) + png_uint_16 blue_1 = png_ptr->gamma_16_to_1[(blue & 0xff) >> png_ptr->gamma_shift][blue>>8]; png_uint_16 gray16 = (png_uint_16)((rc*red_1 + gc*green_1 + bc*blue_1 + 16384)>>15); - w = png_ptr->gamma_16_from_1[(gray16&0xff) >> + w = png_ptr->gamma_16_from_1[(gray16 & 0xff) >> png_ptr->gamma_shift][gray16 >> 8]; rgb_error |= 1; } @@ -3138,10 +3139,11 @@ png_do_rgb_to_gray(png_structrp png_ptr, png_row_infop row_info, png_bytep row) for (i = 0; i < row_width; i++) { png_uint_16 red, green, blue, gray16; + png_byte hi,lo; - red = (png_uint_16)(((*(sp)) << 8) | *(sp + 1)); sp += 2; - green = (png_uint_16)(((*(sp)) << 8) | *(sp + 1)); sp += 2; - blue = (png_uint_16)(((*(sp)) << 8) | *(sp + 1)); sp += 2; + hi=*(sp)++; lo=*(sp)++; red = (png_uint_16)((hi << 8) | (lo)); + hi=*(sp)++; lo=*(sp)++; green = (png_uint_16)((hi << 8) | (lo)); + hi=*(sp)++; lo=*(sp)++; blue = (png_uint_16)((hi << 8) | (lo)); if (red != green || red != blue) rgb_error |= 1; @@ -3667,7 +3669,8 @@ png_do_compose(png_row_infop row_info, png_bytep row, png_structrp png_ptr) if (optimize != 0) w = v; else - w = gamma_16_from_1[(v&0xff) >> gamma_shift][v >> 8]; + w = gamma_16_from_1[(v & 0xff) >> + gamma_shift][v >> 8]; *sp = (png_byte)((w >> 8) & 0xff); *(sp + 1) = (png_byte)(w & 0xff); } @@ -3831,7 +3834,7 @@ png_do_compose(png_row_infop row_info, png_bytep row, png_structrp png_ptr) v = gamma_16_to_1[*(sp + 1) >> gamma_shift][*sp]; png_composite_16(w, v, a, png_ptr->background_1.red); if (optimize == 0) - w = gamma_16_from_1[((w&0xff) >> gamma_shift)][w >> + w = gamma_16_from_1[((w & 0xff) >> gamma_shift)][w >> 8]; *sp = (png_byte)((w >> 8) & 0xff); *(sp + 1) = (png_byte)(w & 0xff); @@ -3839,7 +3842,7 @@ png_do_compose(png_row_infop row_info, png_bytep row, png_structrp png_ptr) v = gamma_16_to_1[*(sp + 3) >> gamma_shift][*(sp + 2)]; png_composite_16(w, v, a, png_ptr->background_1.green); if (optimize == 0) - w = gamma_16_from_1[((w&0xff) >> gamma_shift)][w >> + w = gamma_16_from_1[((w & 0xff) >> gamma_shift)][w >> 8]; *(sp + 2) = (png_byte)((w >> 8) & 0xff); @@ -3848,7 +3851,7 @@ png_do_compose(png_row_infop row_info, png_bytep row, png_structrp png_ptr) v = gamma_16_to_1[*(sp + 5) >> gamma_shift][*(sp + 4)]; png_composite_16(w, v, a, png_ptr->background_1.blue); if (optimize == 0) - w = gamma_16_from_1[((w&0xff) >> gamma_shift)][w >> + w = gamma_16_from_1[((w & 0xff) >> gamma_shift)][w >> 8]; *(sp + 4) = (png_byte)((w >> 8) & 0xff); @@ -4837,7 +4840,7 @@ png_do_read_transformations(png_structrp png_ptr, png_row_infop row_info) /* Because PNG_COMPOSE does the gamma transform if there is something to * do (if there is an alpha channel or transparency.) */ - !((png_ptr->transformations & PNG_COMPOSE) && + !((png_ptr->transformations & PNG_COMPOSE) != 0 && ((png_ptr->num_trans != 0) || (png_ptr->color_type & PNG_COLOR_MASK_ALPHA) != 0)) && #endif diff --git a/pngrutil.c b/pngrutil.c index 4c26be48c..6c5c37526 100644 --- a/pngrutil.c +++ b/pngrutil.c @@ -1,8 +1,8 @@ /* pngrutil.c - utilities to read a PNG file * - * Last changed in libpng 1.6.15 [November 20, 2014] - * Copyright (c) 1998-2014 Glenn Randers-Pehrson + * Last changed in libpng 1.6.17 [March 26, 2015] + * Copyright (c) 1998-2015 Glenn Randers-Pehrson * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger) * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.) * @@ -1515,8 +1515,10 @@ png_handle_iCCP(png_structrp png_ptr, png_inforp info_ptr, png_uint_32 length) else if (size > 0) errmsg = "truncated"; +#ifndef __COVERITY__ else errmsg = png_ptr->zstream.msg; +#endif } /* else png_icc_check_tag_table output an error */ @@ -2804,7 +2806,7 @@ png_handle_unknown(png_structrp png_ptr, png_inforp info_ptr, &png_ptr->unknown_chunk); /* ret is: - * negative: An error occured, png_chunk_error will be called. + * negative: An error occurred; png_chunk_error will be called. * zero: The chunk was not handled, the chunk will be discarded * unless png_set_keep_unknown_chunks has been used to set * a 'keep' behavior for this particular chunk, in which @@ -1,8 +1,8 @@ /* pngset.c - storage of image information into info struct * - * Last changed in libpng 1.6.15 [November 20, 2014] - * Copyright (c) 1998-2014 Glenn Randers-Pehrson + * Last changed in libpng 1.6.17 [March 26, 2015] + * Copyright (c) 1998-2015 Glenn Randers-Pehrson * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger) * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.) * @@ -190,6 +190,7 @@ png_set_hIST(png_const_structrp png_ptr, png_inforp info_ptr, if (info_ptr->hist == NULL) { png_warning(png_ptr, "Insufficient memory for hIST chunk data"); + return; } @@ -271,7 +272,7 @@ png_set_pCAL(png_const_structrp png_ptr, png_inforp info_ptr, png_debug1(1, "in %s storage function", "pCAL"); if (png_ptr == NULL || info_ptr == NULL || purpose == NULL || units == NULL - || (nparams > 0 && params == NULL)) + || (nparams > 0 && params == NULL)) return; length = strlen(purpose) + 1; @@ -301,6 +302,7 @@ png_set_pCAL(png_const_structrp png_ptr, png_inforp info_ptr, if (info_ptr->pcal_purpose == NULL) { png_warning(png_ptr, "Insufficient memory for pCAL purpose"); + return; } @@ -322,6 +324,7 @@ png_set_pCAL(png_const_structrp png_ptr, png_inforp info_ptr, if (info_ptr->pcal_units == NULL) { png_warning(png_ptr, "Insufficient memory for pCAL units"); + return; } @@ -333,6 +336,7 @@ png_set_pCAL(png_const_structrp png_ptr, png_inforp info_ptr, if (info_ptr->pcal_params == NULL) { png_warning(png_ptr, "Insufficient memory for pCAL params"); + return; } @@ -349,6 +353,7 @@ png_set_pCAL(png_const_structrp png_ptr, png_inforp info_ptr, if (info_ptr->pcal_params[i] == NULL) { png_warning(png_ptr, "Insufficient memory for pCAL parameter"); + return; } @@ -398,6 +403,7 @@ png_set_sCAL_s(png_const_structrp png_ptr, png_inforp info_ptr, if (info_ptr->scal_s_width == NULL) { png_warning(png_ptr, "Memory allocation failed while processing sCAL"); + return; } @@ -416,6 +422,7 @@ png_set_sCAL_s(png_const_structrp png_ptr, png_inforp info_ptr, info_ptr->scal_s_width = NULL; png_warning(png_ptr, "Memory allocation failed while processing sCAL"); + return; } @@ -519,6 +526,7 @@ png_set_PLTE(png_structrp png_ptr, png_inforp info_ptr, else { png_warning(png_ptr, "Invalid palette length"); + return; } } @@ -531,7 +539,6 @@ png_set_PLTE(png_structrp png_ptr, png_inforp info_ptr, )) { png_error(png_ptr, "Invalid palette"); - return; } /* It may not actually be necessary to set png_ptr->palette here; @@ -655,6 +662,7 @@ png_set_iCCP(png_const_structrp png_ptr, png_inforp info_ptr, if (new_iccp_name == NULL) { png_benign_error(png_ptr, "Insufficient memory to process iCCP chunk"); + return; } @@ -668,6 +676,7 @@ png_set_iCCP(png_const_structrp png_ptr, png_inforp info_ptr, new_iccp_name = NULL; png_benign_error(png_ptr, "Insufficient memory to process iCCP profile"); + return; } @@ -743,6 +752,7 @@ png_set_text_2(png_const_structrp png_ptr, png_inforp info_ptr, { png_chunk_report(png_ptr, "too many text chunks", PNG_CHUNK_WRITE_ERROR); + return 1; } @@ -798,7 +808,7 @@ png_set_text_2(png_const_structrp png_ptr, png_inforp info_ptr, else lang_key_len = 0; } -# else /* PNG_iTXt_SUPPORTED */ +# else /* iTXt */ { png_chunk_report(png_ptr, "iTXt chunk not supported", PNG_CHUNK_WRITE_ERROR); @@ -831,6 +841,7 @@ png_set_text_2(png_const_structrp png_ptr, png_inforp info_ptr, { png_chunk_report(png_ptr, "text chunk: out of memory", PNG_CHUNK_WRITE_ERROR); + return 1; } @@ -904,6 +915,7 @@ png_set_tIME(png_const_structrp png_ptr, png_inforp info_ptr, mod_time->second > 60) { png_warning(png_ptr, "Ignoring invalid time value"); + return; } @@ -920,6 +932,7 @@ png_set_tRNS(png_structrp png_ptr, png_inforp info_ptr, png_debug1(1, "in %s storage function", "tRNS"); if (png_ptr == NULL || info_ptr == NULL) + return; if (trans_alpha != NULL) @@ -945,16 +958,21 @@ png_set_tRNS(png_structrp png_ptr, png_inforp info_ptr, if (trans_color != NULL) { - int sample_max = (1 << info_ptr->bit_depth); - - if ((info_ptr->color_type == PNG_COLOR_TYPE_GRAY && - trans_color->gray > sample_max) || - (info_ptr->color_type == PNG_COLOR_TYPE_RGB && - (trans_color->red > sample_max || - trans_color->green > sample_max || - trans_color->blue > sample_max))) - png_warning(png_ptr, - "tRNS chunk has out-of-range samples for bit_depth"); +#ifdef PNG_WARNINGS_SUPPORTED + if (info_ptr->bit_depth < 16) + { + int sample_max = (1 << info_ptr->bit_depth) - 1; + + if ((info_ptr->color_type == PNG_COLOR_TYPE_GRAY && + trans_color->gray > sample_max) || + (info_ptr->color_type == PNG_COLOR_TYPE_RGB && + (trans_color->red > sample_max || + trans_color->green > sample_max || + trans_color->blue > sample_max))) + png_warning(png_ptr, + "tRNS chunk has out-of-range samples for bit_depth"); + } +#endif info_ptr->trans_color = *trans_color; @@ -1001,6 +1019,7 @@ png_set_sPLT(png_const_structrp png_ptr, { /* Out of memory or too many chunks */ png_chunk_report(png_ptr, "too many sPLT chunks", PNG_CHUNK_WRITE_ERROR); + return; } @@ -1116,7 +1135,7 @@ png_set_unknown_chunks(png_const_structrp png_ptr, png_unknown_chunkp np; if (png_ptr == NULL || info_ptr == NULL || num_unknowns <= 0 || - unknowns == NULL) + unknowns == NULL) return; /* Check for the failure cases where support has been disabled at compile @@ -1130,6 +1149,7 @@ png_set_unknown_chunks(png_const_structrp png_ptr, if ((png_ptr->mode & PNG_IS_READ_STRUCT) != 0) { png_app_error(png_ptr, "no unknown chunk support on read"); + return; } # endif @@ -1138,6 +1158,7 @@ png_set_unknown_chunks(png_const_structrp png_ptr, if ((png_ptr->mode & PNG_IS_READ_STRUCT) == 0) { png_app_error(png_ptr, "no unknown chunk support on write"); + return; } # endif @@ -1155,6 +1176,7 @@ png_set_unknown_chunks(png_const_structrp png_ptr, { png_chunk_report(png_ptr, "too many unknown chunks", PNG_CHUNK_WRITE_ERROR); + return; } @@ -1232,8 +1254,7 @@ png_set_unknown_chunk_location(png_const_structrp png_ptr, png_inforp info_ptr, check_location(png_ptr, location); } } -#endif - +#endif /* STORE_UNKNOWN_CHUNKS */ #ifdef PNG_MNG_FEATURES_SUPPORTED png_uint_32 PNGAPI @@ -1264,6 +1285,7 @@ add_one_chunk(png_bytep list, unsigned int count, png_const_bytep add, int keep) if (memcmp(list, add, 4) == 0) { list[4] = (png_byte)keep; + return count; } } @@ -1291,6 +1313,7 @@ png_set_keep_unknown_chunks(png_structrp png_ptr, int keep, if (keep < 0 || keep >= PNG_HANDLE_CHUNK_LAST) { png_app_error(png_ptr, "png_set_keep_unknown_chunks: invalid keep"); + return; } @@ -1340,6 +1363,7 @@ png_set_keep_unknown_chunks(png_structrp png_ptr, int keep, * which can be switched off. */ png_app_error(png_ptr, "png_set_keep_unknown_chunks: no chunk list"); + return; } @@ -1355,6 +1379,7 @@ png_set_keep_unknown_chunks(png_structrp png_ptr, int keep, if (num_chunks + old_num_chunks > UINT_MAX/5) { png_app_error(png_ptr, "png_set_keep_unknown_chunks: too many chunks"); + return; } @@ -1492,23 +1517,27 @@ png_set_compression_buffer_size(png_structrp png_ptr, png_size_t size) { png_warning(png_ptr, "Compression buffer size cannot be changed because it is in use"); + return; } +#ifndef __COVERITY__ if (size > ZLIB_IO_MAX) { png_warning(png_ptr, "Compression buffer size limited to system maximum"); size = ZLIB_IO_MAX; /* must fit */ } +#endif - else if (size < 6) + if (size < 6) { /* Deflate will potentially go into an infinite loop on a SYNC_FLUSH * if this is permitted. */ png_warning(png_ptr, "Compression buffer size cannot be reduced below 6"); + return; } @@ -1550,8 +1579,8 @@ png_set_user_limits (png_structrp png_ptr, png_uint_32 user_width_max, void PNGAPI png_set_chunk_cache_max (png_structrp png_ptr, png_uint_32 user_chunk_cache_max) { - if (png_ptr != NULL) - png_ptr->user_chunk_cache_max = user_chunk_cache_max; + if (png_ptr != NULL) + png_ptr->user_chunk_cache_max = user_chunk_cache_max; } /* This function was added to libpng 1.4.1 */ diff --git a/pngstruct.h b/pngstruct.h index c1dd026a9..8420d0997 100644 --- a/pngstruct.h +++ b/pngstruct.h @@ -1,12 +1,11 @@ /* pngstruct.h - header file for PNG reference library * + * Last changed in libpng 1.6.1 [March 28, 2013] * Copyright (c) 1998-2013 Glenn Randers-Pehrson * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger) * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.) * - * Last changed in libpng 1.6.1 [March 28, 2013] - * * This code is released under the libpng license. * For conditions of distribution and use, see the disclaimer * and license in png.h @@ -101,7 +100,7 @@ typedef struct png_XYZ #endif /* COLORSPACE */ #if defined(PNG_COLORSPACE_SUPPORTED) || defined(PNG_GAMMA_SUPPORTED) -/* A colorspace is all the above plus, potentially, profile information, +/* A colorspace is all the above plus, potentially, profile information; * however at present libpng does not use the profile internally so it is only * stored in the png_info struct (if iCCP is supported.) The rendering intent * is retained here and is checked. @@ -1,8 +1,8 @@ /* pngtest.c - a simple test program to test libpng * - * Last changed in libpng 1.6.15 [November 20, 2014] - * Copyright (c) 1998-2014 Glenn Randers-Pehrson + * Last changed in libpng 1.6.17 [March 26, 2015] + * Copyright (c) 1998-2015 Glenn Randers-Pehrson * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger) * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.) * @@ -469,7 +469,7 @@ pngtest_error(png_structp png_ptr, png_const_charp message) #if defined(PNG_USER_MEM_SUPPORTED) && PNG_DEBUG /* Allocate memory. For reasonable files, size should never exceed - * 64K. However, zlib may allocate more then 64K if you don't tell + * 64K. However, zlib may allocate more than 64K if you don't tell * it not to. See zconf.h and png.h for more information. zlib does * need to allocate exactly 64K, so whatever you call here must * have the ability to do that. @@ -1701,6 +1701,8 @@ main(int argc, char *argv[]) int multiple = 0; int ierror = 0; + png_structp dummy_ptr; + fprintf(STDERR, "\n Testing libpng version %s\n", PNG_LIBPNG_VER_STRING); fprintf(STDERR, " with zlib version %s\n", ZLIB_VERSION); fprintf(STDERR, "%s", png_get_copyright(NULL)); @@ -1994,6 +1996,24 @@ main(int argc, char *argv[]) else fprintf(STDERR, " libpng FAILS test\n"); + dummy_ptr = png_create_read_struct(PNG_LIBPNG_VER_STRING, NULL, NULL, NULL); + fprintf(STDERR, " Default limits:\n"); + fprintf(STDERR, " width_max = %lu\n", + (unsigned long) png_get_user_width_max(dummy_ptr)); + fprintf(STDERR, " height_max = %lu\n", + (unsigned long) png_get_user_height_max(dummy_ptr)); + if (png_get_chunk_cache_max(dummy_ptr) == 0) + fprintf(STDERR, " cache_max = unlimited\n"); + else + fprintf(STDERR, " cache_max = %lu\n", + (unsigned long) png_get_chunk_cache_max(dummy_ptr)); + if (png_get_chunk_malloc_max(dummy_ptr) == 0) + fprintf(STDERR, " malloc_max = unlimited\n"); + else + fprintf(STDERR, " malloc_max = %lu\n", + (unsigned long) png_get_chunk_malloc_max(dummy_ptr)); + png_destroy_read_struct(&dummy_ptr, NULL, NULL); + return (int)(ierror != 0); } #else @@ -2008,4 +2028,4 @@ main(void) #endif /* Generate a compiler error if there is an old png.h in the search path. */ -typedef png_libpng_version_1_6_16 Your_png_h_is_not_version_1_6_16; +typedef png_libpng_version_1_6_17 Your_png_h_is_not_version_1_6_17; diff --git a/pngtrans.c b/pngtrans.c index 56856b480..cd3a79b6f 100644 --- a/pngtrans.c +++ b/pngtrans.c @@ -1,8 +1,8 @@ /* pngtrans.c - transforms the data in a row (used by both readers and writers) * - * Last changed in libpng 1.6.15 [November 20, 2014] - * Copyright (c) 1998-2014 Glenn Randers-Pehrson + * Last changed in libpng 1.6.17 [March 26, 2015] + * Copyright (c) 1998-2015 Glenn Randers-Pehrson * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger) * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.) * diff --git a/pngusr.dfa b/pngusr.dfa index 9d39564b5..83067c38c 100644 --- a/pngusr.dfa +++ b/pngusr.dfa @@ -8,7 +8,7 @@ # here by entering the appropriate values as #defines preceded by '@' (to cause, # them to be passed through to the build of pnglibconf.h), for example: # -# @# define PNG_USER_WIDTH_MAX 1000000 -# @# define PNG_USER_HEIGHT_MAX 1000000 -# @# define PNG_USER_CHUNK_CACHE_MAX 128 -# @# define PNG_USER_CHUNK_MALLOC_MAX 8000000 +# @# define PNG_USER_WIDTH_MAX 65535 +# @# define PNG_USER_HEIGHT_MAX 65535 +# @# define PNG_USER_CHUNK_CACHE_MAX 256 +# @# define PNG_USER_CHUNK_MALLOC_MAX 640000 diff --git a/pngwrite.c b/pngwrite.c index 94548877b..e3c203496 100644 --- a/pngwrite.c +++ b/pngwrite.c @@ -1,8 +1,8 @@ /* pngwrite.c - general routines to write a PNG file * - * Last changed in libpng 1.6.15 [November 20, 2014] - * Copyright (c) 1998-2014 Glenn Randers-Pehrson + * Last changed in libpng 1.6.17 [March 26, 2015] + * Copyright (c) 1998-2015 Glenn Randers-Pehrson * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger) * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.) * @@ -638,8 +638,8 @@ png_do_write_intrapixel(png_row_infop row_info, png_bytep row) for (i = 0, rp = row; i < row_width; i++, rp += bytes_per_pixel) { - *(rp) = (png_byte)((*rp - *(rp + 1)) & 0xff); - *(rp + 2) = (png_byte)((*(rp + 2) - *(rp + 1)) & 0xff); + *(rp) = (png_byte)(*rp - *(rp + 1)); + *(rp + 2) = (png_byte)(*(rp + 2) - *(rp + 1)); } } @@ -665,10 +665,10 @@ png_do_write_intrapixel(png_row_infop row_info, png_bytep row) png_uint_32 s2 = (*(rp + 4) << 8) | *(rp + 5); png_uint_32 red = (png_uint_32)((s0 - s1) & 0xffffL); png_uint_32 blue = (png_uint_32)((s2 - s1) & 0xffffL); - *(rp ) = (png_byte)((red >> 8) & 0xff); - *(rp + 1) = (png_byte)(red & 0xff); - *(rp + 4) = (png_byte)((blue >> 8) & 0xff); - *(rp + 5) = (png_byte)(blue & 0xff); + *(rp ) = (png_byte)(red >> 8); + *(rp + 1) = (png_byte)red; + *(rp + 4) = (png_byte)(blue >> 8); + *(rp + 5) = (png_byte)blue; } } #endif /* WRITE_16BIT */ @@ -1050,8 +1050,8 @@ png_set_filter(png_structrp png_ptr, int method, int filters) * it is too late to start using the filters that need it, since we * will be missing the data in the previous row. If an application * wants to start and stop using particular filters during compression, - * it should start out with all of the filters, and then add and - * remove them after the start of compression. + * it should start out with all of the filters, and then remove them + * or add them back after the start of compression. */ if (png_ptr->row_buf != NULL) { @@ -1372,6 +1372,7 @@ png_set_filter_heuristics_fixed(png_structrp png_ptr, int heuristic_method, #endif /* FIXED_POINT */ #endif /* WRITE_WEIGHTED_FILTER */ +#ifdef PNG_WRITE_CUSTOMIZE_COMPRESSION_SUPPORTED void PNGAPI png_set_compression_level(png_structrp png_ptr, int level) { @@ -1417,8 +1418,8 @@ png_set_compression_window_bits(png_structrp png_ptr, int window_bits) if (png_ptr == NULL) return; - /* Prior to 1.6.0 this would warn but then set the window_bits value, this - * meant that negative window bits values could be selected which would cause + /* Prior to 1.6.0 this would warn but then set the window_bits value. This + * meant that negative window bits values could be selected that would cause * libpng to write a non-standard PNG file with raw deflate or gzip * compressed IDAT or ancillary chunks. Such files can be read and there is * no warning on read, so this seems like a very bad idea. @@ -1454,6 +1455,7 @@ png_set_compression_method(png_structrp png_ptr, int method) png_ptr->zlib_method = method; } +#endif /* WRITE_CUSTOMIZE_COMPRESSION */ /* The following were added to libpng-1.5.4 */ #ifdef PNG_WRITE_CUSTOMIZE_ZTXT_COMPRESSION_SUPPORTED @@ -2287,7 +2289,9 @@ png_image_write_main(png_voidp argument) * it about 50 times. The speed-up in pngstest was about 10-20% of the * total (user) time on a heavily loaded system. */ +#ifdef PNG_WRITE_CUSTOMIZE_COMPRESSION_SUPPORTED png_set_compression_level(png_ptr, 3); +#endif } /* Check for the cases that currently require a pre-transform on the row diff --git a/pngwtran.c b/pngwtran.c index 09562a787..db82e27bc 100644 --- a/pngwtran.c +++ b/pngwtran.c @@ -1,8 +1,8 @@ /* pngwtran.c - transforms the data in a row for PNG writers * - * Last changed in libpng 1.6.15 [November 20, 2014] - * Copyright (c) 1998-2014 Glenn Randers-Pehrson + * Last changed in libpng 1.6.17 [March 26, 2015] + * Copyright (c) 1998-2015 Glenn Randers-Pehrson * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger) * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.) * diff --git a/pngwutil.c b/pngwutil.c index 10c1edffa..b289671d1 100644 --- a/pngwutil.c +++ b/pngwutil.c @@ -1,8 +1,8 @@ /* pngwutil.c - utilities to write a PNG file * - * Last changed in libpng 1.6.15 [November 20, 2014] - * Copyright (c) 1998-2014 Glenn Randers-Pehrson + * Last changed in libpng 1.6.17 [March 26, 2015] + * Copyright (c) 1998-2015 Glenn Randers-Pehrson * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger) * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.) * @@ -23,10 +23,10 @@ void PNGAPI png_save_uint_32(png_bytep buf, png_uint_32 i) { - buf[0] = (png_byte)((i >> 24) & 0xff); - buf[1] = (png_byte)((i >> 16) & 0xff); - buf[2] = (png_byte)((i >> 8) & 0xff); - buf[3] = (png_byte)(i & 0xff); + buf[0] = (png_byte)(i >> 24); + buf[1] = (png_byte)(i >> 16); + buf[2] = (png_byte)(i >> 8); + buf[3] = (png_byte)(i ); } /* Place a 16-bit number into a buffer in PNG byte order. @@ -36,8 +36,8 @@ png_save_uint_32(png_bytep buf, png_uint_32 i) void PNGAPI png_save_uint_16(png_bytep buf, unsigned int i) { - buf[0] = (png_byte)((i >> 8) & 0xff); - buf[1] = (png_byte)(i & 0xff); + buf[0] = (png_byte)(i >> 8); + buf[1] = (png_byte)(i ); } #endif @@ -695,7 +695,7 @@ png_check_keyword(png_structrp png_ptr, png_const_charp key, png_bytep new_key) while (*key && key_len < 79) { - png_byte ch = (png_byte)(0xff & *key++); + png_byte ch = (png_byte)*key++; if ((ch > 32 && ch <= 126) || (ch >= 161 /*&& ch <= 255*/)) *new_key++ = ch, ++key_len, space = 0; @@ -871,7 +871,7 @@ png_write_IHDR(png_structrp png_ptr, png_uint_32 width, png_uint_32 height, interlace_type=PNG_INTERLACE_NONE; #endif - /* Save the relevent information */ + /* Save the relevant information */ png_ptr->bit_depth = (png_byte)bit_depth; png_ptr->color_type = (png_byte)color_type; png_ptr->interlaced = (png_byte)interlace_type; @@ -1765,7 +1765,7 @@ png_write_iTXt(png_structrp png_ptr, int compression, png_const_charp key, png_write_compressed_data_out(png_ptr, &comp); else - png_write_chunk_data(png_ptr, (png_const_bytep)text, comp.input_len); + png_write_chunk_data(png_ptr, (png_const_bytep)text, comp.output_len); png_write_chunk_end(png_ptr); } @@ -2314,7 +2314,7 @@ png_do_write_interlace(png_row_infop row_info, png_bytep row, int pass) * been specified by the application, and then writes the row out with the * chosen filter. */ -static void +static void /* PRIVATE */ png_write_filtered_row(png_structrp png_ptr, png_bytep filtered_row, png_size_t row_bytes); @@ -2453,7 +2453,7 @@ png_write_find_filter(png_structrp png_ptr, png_row_infop row_info) for (lp = row_buf + 1; i < row_bytes; i++, rp++, lp++, dp++) { - *dp = (png_byte)(((int)*rp - (int)*lp) & 0xff); + *dp = (png_byte)((int)*rp - (int)*lp); } best_row = png_ptr->sub_row; @@ -2515,7 +2515,7 @@ png_write_find_filter(png_structrp png_ptr, png_row_infop row_info) for (lp = row_buf + 1; i < row_bytes; i++, rp++, lp++, dp++) { - v = *dp = (png_byte)(((int)*rp - (int)*lp) & 0xff); + v = *dp = (png_byte)((int)*rp - (int)*lp); sum += (v < 128) ? v : 256 - v; @@ -2574,7 +2574,7 @@ png_write_find_filter(png_structrp png_ptr, png_row_infop row_info) pp = prev_row + 1; i < row_bytes; i++, rp++, pp++, dp++) { - *dp = (png_byte)(((int)*rp - (int)*pp) & 0xff); + *dp = (png_byte)((int)*rp - (int)*pp); } best_row = png_ptr->up_row; @@ -2625,7 +2625,7 @@ png_write_find_filter(png_structrp png_ptr, png_row_infop row_info) for (i = 0, rp = row_buf + 1, dp = png_ptr->up_row + 1, pp = prev_row + 1; i < row_bytes; i++) { - v = *dp++ = (png_byte)(((int)*rp++ - (int)*pp++) & 0xff); + v = *dp++ = (png_byte)((int)*rp++ - (int)*pp++); sum += (v < 128) ? v : 256 - v; @@ -2683,13 +2683,13 @@ png_write_find_filter(png_structrp png_ptr, png_row_infop row_info) for (i = 0, rp = row_buf + 1, dp = png_ptr->avg_row + 1, pp = prev_row + 1; i < bpp; i++) { - *dp++ = (png_byte)(((int)*rp++ - ((int)*pp++ / 2)) & 0xff); + *dp++ = (png_byte)((int)*rp++ - ((int)*pp++ / 2)); } for (lp = row_buf + 1; i < row_bytes; i++) { - *dp++ = (png_byte)(((int)*rp++ - (((int)*pp++ + (int)*lp++) / 2)) - & 0xff); + *dp++ = + (png_byte)((int)*rp++ - (((int)*pp++ + (int)*lp++) / 2)); } best_row = png_ptr->avg_row; } @@ -2738,7 +2738,7 @@ png_write_find_filter(png_structrp png_ptr, png_row_infop row_info) for (i = 0, rp = row_buf + 1, dp = png_ptr->avg_row + 1, pp = prev_row + 1; i < bpp; i++) { - v = *dp++ = (png_byte)(((int)*rp++ - ((int)*pp++ / 2)) & 0xff); + v = *dp++ = (png_byte)((int)*rp++ - ((int)*pp++ / 2)); sum += (v < 128) ? v : 256 - v; } @@ -2746,7 +2746,7 @@ png_write_find_filter(png_structrp png_ptr, png_row_infop row_info) for (lp = row_buf + 1; i < row_bytes; i++) { v = *dp++ = - (png_byte)(((int)*rp++ - (((int)*pp++ + (int)*lp++) / 2)) & 0xff); + (png_byte)(((int)*rp++ - ((int)*pp++ + (int)*lp++) / 2)); sum += (v < 128) ? v : 256 - v; @@ -2804,7 +2804,7 @@ png_write_find_filter(png_structrp png_ptr, png_row_infop row_info) for (i = 0, rp = row_buf + 1, dp = png_ptr->paeth_row + 1, pp = prev_row + 1; i < bpp; i++) { - *dp++ = (png_byte)(((int)*rp++ - (int)*pp++) & 0xff); + *dp++ = (png_byte)((int)*rp++ - (int)*pp++); } for (lp = row_buf + 1, cp = prev_row + 1; i < row_bytes; i++) @@ -2830,7 +2830,7 @@ png_write_find_filter(png_structrp png_ptr, png_row_infop row_info) p = (pa <= pb && pa <=pc) ? a : (pb <= pc) ? b : c; - *dp++ = (png_byte)(((int)*rp++ - p) & 0xff); + *dp++ = (png_byte)((int)*rp++ - p); } best_row = png_ptr->paeth_row; } @@ -2879,7 +2879,7 @@ png_write_find_filter(png_structrp png_ptr, png_row_infop row_info) for (i = 0, rp = row_buf + 1, dp = png_ptr->paeth_row + 1, pp = prev_row + 1; i < bpp; i++) { - v = *dp++ = (png_byte)(((int)*rp++ - (int)*pp++) & 0xff); + v = *dp++ = (png_byte)((int)*rp++ - (int)*pp++); sum += (v < 128) ? v : 256 - v; } @@ -2921,7 +2921,7 @@ png_write_find_filter(png_structrp png_ptr, png_row_infop row_info) p = c; #endif /* SLOW_PAETH */ - v = *dp++ = (png_byte)(((int)*rp++ - p) & 0xff); + v = *dp++ = (png_byte)((int)*rp++ - p); sum += (v < 128) ? v : 256 - v; @@ -3003,6 +3003,7 @@ png_write_filtered_row(png_structrp png_ptr, png_bytep filtered_row, png_compress_IDAT(png_ptr, filtered_row, full_row_length, Z_NO_FLUSH); +#ifdef PNG_WRITE_FILTER_SUPPORTED /* Swap the current and previous rows */ if (png_ptr->prev_row != NULL) { @@ -3012,6 +3013,7 @@ png_write_filtered_row(png_structrp png_ptr, png_bytep filtered_row, png_ptr->prev_row = png_ptr->row_buf; png_ptr->row_buf = tptr; } +#endif /* WRITE_FILTER */ /* Finish row - updates counters and flushes zlib if last row */ png_write_finish_row(png_ptr); diff --git a/projects/vstudio/readme.txt b/projects/vstudio/readme.txt index e7a02d876..a2521f205 100644 --- a/projects/vstudio/readme.txt +++ b/projects/vstudio/readme.txt @@ -1,7 +1,7 @@ VisualStudio instructions -libpng version 1.6.16 - December 22, 2014 +libpng version 1.6.17 - March 26, 2015 Copyright (c) 1998-2010 Glenn Randers-Pehrson diff --git a/projects/vstudio/zlib.props b/projects/vstudio/zlib.props index adbfea825..5d6d69b72 100644 --- a/projects/vstudio/zlib.props +++ b/projects/vstudio/zlib.props @@ -2,7 +2,7 @@ <!-- * zlib.props - location of zlib source * - * libpng version 1.6.16 - December 22, 2014 + * libpng version 1.6.17 - March 26, 2015 * * Copyright (c) 1998-2011 Glenn Randers-Pehrson * diff --git a/scripts/README.txt b/scripts/README.txt index b689f8178..f6a9cf38c 100644 --- a/scripts/README.txt +++ b/scripts/README.txt @@ -1,9 +1,9 @@ -Makefiles for libpng version 1.6.16 - December 22, 2014 +Makefiles for libpng version 1.6.17 - March 26, 2015 pnglibconf.h.prebuilt => Stores configuration settings makefile.linux => Linux/ELF makefile - (gcc, creates libpng16.so.16.1.6.16) + (gcc, creates libpng16.so.16.1.6.17) makefile.gcc => Generic makefile (gcc, creates static libpng.a) makefile.knr => Archaic UNIX Makefile that converts files with ansi2knr (Requires ansi2knr.c from @@ -33,12 +33,12 @@ pnglibconf.h.prebuilt => Stores configuration settings makefile.os2 => OS/2 Makefile (gcc and emx, requires libpng.def) makefile.sco => For SCO OSr5 ELF and Unixware 7 with Native cc makefile.sggcc => Silicon Graphics (gcc, - creates libpng16.so.16.1.6.16) + creates libpng16.so.16.1.6.17) makefile.sgi => Silicon Graphics IRIX makefile (cc, creates static lib) makefile.solaris => Solaris 2.X makefile (gcc, - creates libpng16.so.16.1.6.16) + creates libpng16.so.16.1.6.17) makefile.so9 => Solaris 9 makefile (gcc, - creates libpng16.so.16.1.6.16) + creates libpng16.so.16.1.6.17) makefile.std => Generic UNIX makefile (cc, creates static libpng.a) makefile.sunos => Sun makefile makefile.32sunu => Sun Ultra 32-bit makefile diff --git a/scripts/def.c b/scripts/def.c index 19b796d2a..9c88f7743 100644 --- a/scripts/def.c +++ b/scripts/def.c @@ -21,7 +21,7 @@ PNG_DFN "OS2 DESCRIPTION "PNG image compression library"" PNG_DFN "OS2 CODE PRELOAD MOVEABLE DISCARDABLE" PNG_DFN "" PNG_DFN "EXPORTS" -PNG_DFN ";Version 1.6.16" +PNG_DFN ";Version 1.6.17" #define PNG_EXPORTA(ordinal, type, name, args, attributes)\ PNG_DFN "@" SYMBOL_PREFIX "@@" name "@" diff --git a/scripts/dfn.awk b/scripts/dfn.awk index 675829d47..346b9db7d 100755 --- a/scripts/dfn.awk +++ b/scripts/dfn.awk @@ -17,7 +17,7 @@ BEGIN{ out="/dev/null" # as a flag out_count=0 # count of output lines - err=0 # set if an error occured + err=0 # set if an error occurred sort=0 # sort the output array[""]="" } diff --git a/scripts/libpng-config-head.in b/scripts/libpng-config-head.in index 0d96b65f8..d19010df4 100644 --- a/scripts/libpng-config-head.in +++ b/scripts/libpng-config-head.in @@ -11,7 +11,7 @@ # Modeled after libxml-config. -version=1.6.16 +version=1.6.17 prefix="" libdir="" libs="" diff --git a/scripts/libpng.pc.in b/scripts/libpng.pc.in index 2c2653964..ce61a3e98 100644 --- a/scripts/libpng.pc.in +++ b/scripts/libpng.pc.in @@ -5,6 +5,6 @@ includedir=@includedir@/libpng16 Name: libpng Description: Loads and saves PNG files -Version: 1.6.16 +Version: 1.6.17 Libs: -L${libdir} -lpng16 Cflags: -I${includedir} diff --git a/scripts/makefile.cegcc b/scripts/makefile.cegcc index 06515fdc4..cc56297cd 100644 --- a/scripts/makefile.cegcc +++ b/scripts/makefile.cegcc @@ -23,7 +23,7 @@ VERMAJ = 1 VERMIN = 6 -VERMIC = 16 +VERMIC = 17 VER = $(VERMAJ).$(VERMIN).$(VERMIC) NAME = libpng PACKAGE = $(NAME)-$(VER) diff --git a/scripts/makefile.linux b/scripts/makefile.linux index 06f773aa9..86a397f4c 100644 --- a/scripts/makefile.linux +++ b/scripts/makefile.linux @@ -10,7 +10,7 @@ # Library name: LIBNAME = libpng16 PNGMAJ = 16 -RELEASE = 16 +RELEASE = 17 # Shared library names: LIBSO=$(LIBNAME).so diff --git a/scripts/makefile.msys b/scripts/makefile.msys index 106fb8a5b..30595a083 100644 --- a/scripts/makefile.msys +++ b/scripts/makefile.msys @@ -18,7 +18,7 @@ exec_prefix=$(prefix) # Library name: LIBNAME = libpng16 PNGMAJ = 16 -RELEASE = 16 +RELEASE = 17 # Shared library names: LIBSO=$(LIBNAME).dll diff --git a/scripts/makefile.ne12bsd b/scripts/makefile.ne12bsd index e3b7baa0c..d5be81cb7 100644 --- a/scripts/makefile.ne12bsd +++ b/scripts/makefile.ne12bsd @@ -17,7 +17,7 @@ INCSDIR=${LOCALBASE}/include/libpng16 LIB= png16 SHLIB_MAJOR= 0 -SHLIB_MINOR= 1.6.16 +SHLIB_MINOR= 1.6.17 SRCS= png.c pngset.c pngget.c pngrutil.c pngtrans.c pngwutil.c \ pngread.c pngrio.c pngwio.c pngwrite.c pngrtran.c \ pngwtran.c pngmem.c pngerror.c pngpread.c diff --git a/scripts/makefile.netbsd b/scripts/makefile.netbsd index 2cc7a17a9..58c48844d 100644 --- a/scripts/makefile.netbsd +++ b/scripts/makefile.netbsd @@ -17,7 +17,7 @@ INCSDIR=${LOCALBASE}/include LIB= png SHLIB_MAJOR= 16 -SHLIB_MINOR= 1.6.16 +SHLIB_MINOR= 1.6.17 SRCS= png.c pngset.c pngget.c pngrutil.c pngtrans.c pngwutil.c \ pngread.c pngrio.c pngwio.c pngwrite.c pngrtran.c \ pngwtran.c pngmem.c pngerror.c pngpread.c diff --git a/scripts/makefile.openbsd b/scripts/makefile.openbsd index debc2a22c..0ef62a4e4 100644 --- a/scripts/makefile.openbsd +++ b/scripts/makefile.openbsd @@ -11,7 +11,7 @@ LIBDIR= ${PREFIX}/lib MANDIR= ${PREFIX}/man/cat SHLIB_MAJOR= 16 -SHLIB_MINOR= 1.6.16 +SHLIB_MINOR= 1.6.17 LIB= png SRCS= png.c pngerror.c pngget.c pngmem.c pngpread.c \ diff --git a/scripts/makefile.std b/scripts/makefile.std index 75bb2de62..ee6a86835 100644 --- a/scripts/makefile.std +++ b/scripts/makefile.std @@ -106,7 +106,8 @@ install: libpng.a pnglibconf.h chmod 644 $(DESTDIR)$(LIBPATH)/libpng.a clean: - $(RM_F) *.o libpng.a pngtest pngout.png pnglibconf.* + $(RM_F) *.o libpng.a pngtest pngout.png pnglibconf.h pnglibconf.c \ + pnglibconf.out DOCS = ANNOUNCE CHANGES INSTALL KNOWNBUG LICENSE README TODO Y2KINFO writelock: diff --git a/scripts/pnglibconf.dfa b/scripts/pnglibconf.dfa index d98d27437..e9a82f303 100644 --- a/scripts/pnglibconf.dfa +++ b/scripts/pnglibconf.dfa @@ -350,36 +350,15 @@ option USER_MEM option IO_STATE -# This is only for PowerPC big-endian and 680x0 systems -# some testing, not enabled by default. -# NO LONGER USED - -#option READ_BIG_ENDIAN disabled - -# Allow users to control limits on what the READ code will -# read: - -# Added at libpng-1.2.43; adds limit fields to png_struct, -# allows some usages of these fields - -option USER_LIMITS - -# Added at libpng-1.2.6; adds setting APIs, allows additional -# usage of this field (UTSL) - -option SET_USER_LIMITS requires USER_LIMITS - -# Feature added at libpng-1.4.0, this flag added at 1.4.1 -option SET_USER_LIMITS enables SET_CHUNK_CACHE_LIMIT -# Feature added at libpng-1.4.1, this flag added at 1.4.1 +# Libpng limits: limit the size of images and data on read. +# +# If this option is disabled all the limit checking code will be disabled: -option SET_USER_LIMITS enables SET_CHUNK_MALLOC_LIMIT +option USER_LIMITS requires READ -# Libpng limits. -# -# If these settings are *not* set libpng will not limit the size of -# images or the size of data in ancilliary chunks. This does lead to -# security issues if PNG files come from untrusted sources. Settings have the +# The default settings given below for the limits mean that libpng will +# limit the size of images or the size of data in ancilliary chunks to less +# than the specification or implementation limits. Settings have the # following interpretations: # # USER_WIDTH_MAX: maximum width of an image that will be read @@ -388,18 +367,18 @@ option SET_USER_LIMITS enables SET_CHUNK_MALLOC_LIMIT # USER_CHUNK_CACHE_MAX: maximum number of chunks to be cached # # Only chunks that are variable in number are counted towards the -# USER_CHUNK_CACHE_MAX limit -setting USER_WIDTH_MAX -setting USER_HEIGHT_MAX -setting USER_CHUNK_CACHE_MAX -setting USER_CHUNK_MALLOC_MAX - -# To default all these settings to values that are large but probably -# safe turn the SAFE_LIMITS option on; this will cause the value in -# pngpriv.h to be used. Individual values can also be set, simply set -# them in pngusr.dfa with '@#define PNG_setting value' lines. -option SAFE_LIMITS enables USER_LIMITS disabled -= SAFE_LIMITS SAFE_LIMITS + +# Use 0x7fffffff for unlimited +setting USER_WIDTH_MAX default 1000000 +setting USER_HEIGHT_MAX default 1000000 + +# Use 0 for unlimited +setting USER_CHUNK_CACHE_MAX default 1000 +setting USER_CHUNK_MALLOC_MAX default 8000000 + +# If this option is enabled APIs to set the above limits at run time are added; +# without this the hardwired (compile time) limits will be used. +option SET_USER_LIMITS requires USER_LIMITS # All of the following options relate to code capabilities for # processing image data before creating a PNG or after reading one. @@ -541,6 +520,7 @@ option USER_TRANSFORM_INFO if READ_USER_TRANSFORM, WRITE_USER_TRANSFORM # non-IDAT chunks (zTXt, iTXt, iCCP, and unknown chunks). This feature # was added at libpng-1.5.3. option WRITE_CUSTOMIZE_ZTXT_COMPRESSION requires WRITE +option WRITE_CUSTOMIZE_COMPRESSION requires WRITE # Any chunks you are not interested in, you can undef here. The # ones that allocate memory may be expecially important (hIST, @@ -666,7 +646,7 @@ setting QUANTIZE_BLUE_BITS default 5 # are only interested in 8 bits anyway. Increasing this value # results in more memory being used, and more pow() functions # being called to fill in the gamma tables. Don't set this value -# less then 8, and even that may not work (I haven't tested it). +# less than 8, and even that may not work (I haven't tested it). setting MAX_GAMMA_8 default 11 diff --git a/scripts/pnglibconf.h.prebuilt b/scripts/pnglibconf.h.prebuilt index da3b22921..b4ec3c31e 100644 --- a/scripts/pnglibconf.h.prebuilt +++ b/scripts/pnglibconf.h.prebuilt @@ -1,8 +1,8 @@ -/* libpng 1.6.16 STANDARD API DEFINITION */ +/* libpng 1.6.17 STANDARD API DEFINITION */ /* pnglibconf.h - library build configuration */ -/* Libpng version 1.6.16 - December 22, 2014 */ +/* Libpng version 1.6.17 - March 26, 2015 */ /* Copyright (c) 1998-2014 Glenn Randers-Pehrson */ @@ -97,7 +97,6 @@ #define PNG_READ_tIME_SUPPORTED #define PNG_READ_tRNS_SUPPORTED #define PNG_READ_zTXt_SUPPORTED -/*#undef PNG_SAFE_LIMITS_SUPPORTED*/ #define PNG_SAVE_INT_32_SUPPORTED #define PNG_SAVE_UNKNOWN_CHUNKS_SUPPORTED #define PNG_SEQUENTIAL_READ_SUPPORTED @@ -129,6 +128,7 @@ #define PNG_WRITE_BGR_SUPPORTED #define PNG_WRITE_CHECK_FOR_INVALID_INDEX_SUPPORTED #define PNG_WRITE_COMPRESSED_TEXT_SUPPORTED +#define PNG_WRITE_CUSTOMIZE_COMPRESSION_SUPPORTED #define PNG_WRITE_CUSTOMIZE_ZTXT_COMPRESSION_SUPPORTED #define PNG_WRITE_FILLER_SUPPORTED #define PNG_WRITE_FILTER_SUPPORTED @@ -198,6 +198,10 @@ #define PNG_QUANTIZE_RED_BITS 5 #define PNG_TEXT_Z_DEFAULT_COMPRESSION (-1) #define PNG_TEXT_Z_DEFAULT_STRATEGY 0 +#define PNG_USER_CHUNK_CACHE_MAX 1000 +#define PNG_USER_CHUNK_MALLOC_MAX 8000000 +#define PNG_USER_HEIGHT_MAX 1000000 +#define PNG_USER_WIDTH_MAX 1000000 #define PNG_WEIGHT_SHIFT 8 #define PNG_ZBUF_SIZE 8192 #define PNG_ZLIB_VERNUM 0 /* unknown */ diff --git a/scripts/pnglibconf.mak b/scripts/pnglibconf.mak index c7117ddc4..c125ca2c9 100755 --- a/scripts/pnglibconf.mak +++ b/scripts/pnglibconf.mak @@ -49,6 +49,7 @@ pnglibconf.dfn: $(srcdir)/scripts/pnglibconf.dfa $(srcdir)/scripts/options.awk $ $(MOVE) pnglibconf.tmp $@ clean-pnglibconf: - $(DELETE) pnglibconf.* + $(DELETE) pnglibconf.h pnglibconf.c pnglibconf.out pnglibconf.pre \ + pnglibconf.dfn clean: clean-pnglibconf diff --git a/scripts/symbols.def b/scripts/symbols.def index 128374342..5bdcef20f 100644 --- a/scripts/symbols.def +++ b/scripts/symbols.def @@ -1,4 +1,4 @@ -;Version 1.6.16 +;Version 1.6.17 ;-------------------------------------------------------------- ; LIBPNG symbol list as a Win32 DEF file ; Contains all the symbols that can be exported from libpng |