Fix wrong first character when a backreference with a zero repeat is first

(apart from assersions) in a pattern.


git-svn-id: svn://vcs.exim.org/pcre/code/trunk@1719 2f5784b3-3f2a-0410-8824-cb99058d5e15

4 files changed, 30 insertions, 1 deletions

diff --git a/ChangeLog b/ChangeLog
index d90ae78..395b88d 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -31,6 +31,11 @@ were all int variables, causing overflow when files with more than 2147483647
 lines were processed (assuming 32-bit ints). They have all been changed to 
 unsigned long ints.
 
+7. If a backreference with a minimum repeat count of zero was first in a
+pattern, apart from assertions, an incorrect first matching character could be
+recorded. For example, for the pattern /(?=(a))\1?b/, "b" was incorrectly set
+as the first character of a match.
+
 
 Version 8.41 05-July-2017
 -------------------------
diff --git a/pcre_compile.c b/pcre_compile.c
index 42f204c..6dd8886 100644
--- a/pcre_compile.c
+++ b/pcre_compile.c
@@ -8060,7 +8060,7 @@ for (;; ptr++)
         single group (i.e. not to a duplicated name. */
 
         HANDLE_REFERENCE:
-        if (firstcharflags == REQ_UNSET) firstcharflags = REQ_NONE;
+        if (firstcharflags == REQ_UNSET) zerofirstcharflags = firstcharflags = REQ_NONE;
         previous = code;
         item_hwm_offset = cd->hwm - cd->start_workspace;
         *code++ = ((options & PCRE_CASELESS) != 0)? OP_REFI : OP_REF;
diff --git a/testdata/testinput2 b/testdata/testinput2
index 08c6f39..8ba4dc4 100644
--- a/testdata/testinput2
+++ b/testdata/testinput2
@@ -4249,4 +4249,12 @@ backtracking verbs. --/
 
 /(?=.*[A-Z])/I
 
+"(?<=(a))\1?b"
+    ab
+    aaab 
+
+"(?=(a))\1?b"
+    ab
+    aaab 
+
 /-- End of testinput2 --/
diff --git a/testdata/testoutput2 b/testdata/testoutput2
index 811bbef..61ed8d9 100644
--- a/testdata/testoutput2
+++ b/testdata/testoutput2
@@ -14705,4 +14705,20 @@ No options
 No first char
 No need char
 
+"(?<=(a))\1?b"
+    ab
+ 0: b
+ 1: a
+    aaab 
+ 0: ab
+ 1: a
+
+"(?=(a))\1?b"
+    ab
+ 0: ab
+ 1: a
+    aaab 
+ 0: ab
+ 1: a
+
 /-- End of testinput2 --/