From 59d85d7b5519cb5dd513d4a77b4a58de7c819f73 Mon Sep 17 00:00:00 2001
From: "Philip.Hazel" <Philip.Hazel@gmail.com>
Date: Tue, 12 Dec 2017 15:01:51 +0000
Subject: [PATCH] Fix incorrect first matching character when a backreference
 with zero minimum repeat starts a pattern (possibly after assertions).

---
 ChangeLog            |  5 +++++
 src/pcre2_compile.c  |  2 +-
 testdata/testinput2  | 10 ++++++++++
 testdata/testoutput2 | 28 ++++++++++++++++++++++++++++
 4 files changed, 44 insertions(+), 1 deletion(-)

diff --git a/ChangeLog b/ChangeLog
index 40f6761..7af58de 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -65,6 +65,11 @@ 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.
 
+17. 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 10.30 14-August-2017
 ----------------------------
diff --git a/src/pcre2_compile.c b/src/pcre2_compile.c
index 0b91d14..ad17338 100644
--- a/src/pcre2_compile.c
+++ b/src/pcre2_compile.c
@@ -7135,7 +7135,7 @@ for (;; pptr++)
     later. */
 
     HANDLE_SINGLE_REFERENCE:
-    if (firstcuflags == REQ_UNSET) firstcuflags = REQ_NONE;
+    if (firstcuflags == REQ_UNSET) zerofirstcuflags = firstcuflags = REQ_NONE;
     *code++ = ((options & PCRE2_CASELESS) != 0)? OP_REFI : OP_REF;
     PUT2INC(code, 0, meta_arg);
 
diff --git a/testdata/testinput2 b/testdata/testinput2
index 022df20..695f0a4 100644
--- a/testdata/testinput2
+++ b/testdata/testinput2
@@ -5375,4 +5375,14 @@ a)"xI
 
 /[\d-[:print:]]/
 
+# Perl gets the second of these wrong, giving no match.
+
+"(?<=(a))\1?b"I
+    ab
+    aaab 
+
+"(?=(a))\1?b"I
+    ab
+    aaab 
+
 # End of testinput2
diff --git a/testdata/testoutput2 b/testdata/testoutput2
index 2d9e347..31ccfbe 100644
--- a/testdata/testoutput2
+++ b/testdata/testoutput2
@@ -16340,6 +16340,34 @@ Failed: error 150 at offset 3: invalid range in character class
 /[\d-[:print:]]/
 Failed: error 150 at offset 3: invalid range in character class
 
+# Perl gets the second of these wrong, giving no match.
+
+"(?<=(a))\1?b"I
+Capturing subpattern count = 1
+Max back reference = 1
+Max lookbehind = 1
+Last code unit = 'b'
+Subject length lower bound = 1
+    ab
+ 0: b
+ 1: a
+    aaab 
+ 0: ab
+ 1: a
+
+"(?=(a))\1?b"I
+Capturing subpattern count = 1
+Max back reference = 1
+Starting code units: a 
+Last code unit = 'b'
+Subject length lower bound = 1
+    ab
+ 0: ab
+ 1: a
+    aaab 
+ 0: ab
+ 1: a
+
 # End of testinput2
 Error -65: PCRE2_ERROR_BADDATA (unknown error number)
 Error -62: bad serialized data