From 52041d82e9a0b95f510f6f300b7f3ca43c14b035 Mon Sep 17 00:00:00 2001 From: Philip Hazel Date: Sat, 11 Nov 2023 15:56:07 +0000 Subject: [PATCH] Fix misbehaviour of pcre2_match() and pcre2_dfa_match() when PCRE2_FIRSTLINE was set for an anchored pattern. --- ChangeLog | 4 ++++ doc/html/pcre2api.html | 6 +++--- doc/pcre2.txt | 7 ++++--- doc/pcre2api.3 | 8 ++++---- doc/pcre2demo.3 | 2 +- src/pcre2_dfa_match.c | 2 +- src/pcre2_match.c | 2 +- testdata/testinput2 | 13 +++++++++++++ testdata/testinput6 | 13 +++++++++++++ testdata/testoutput2 | 17 +++++++++++++++++ testdata/testoutput6 | 17 +++++++++++++++++ 11 files changed, 78 insertions(+), 13 deletions(-) diff --git a/ChangeLog b/ChangeLog index ea2e07daf..20a6a49b2 100644 --- a/ChangeLog +++ b/ChangeLog @@ -142,6 +142,10 @@ above because \b and \B are defined in terms of \w. option, and (?aP) also sets (?aT) so that (?-aP) disables all ASCII restrictions on POSIX classes. +37. If PCRE2_FIRSTLINE was set on an anchored pattern, pcre2_match() and +pcre2_dfa_match() misbehaved. PCRE2_FIRSTLINE is now ignored for anchored +patterns. + Version 10.42 11-December-2022 ------------------------------ diff --git a/doc/html/pcre2api.html b/doc/html/pcre2api.html index f99beb166..22cf4b0d4 100644 --- a/doc/html/pcre2api.html +++ b/doc/html/pcre2api.html @@ -1686,7 +1686,7 @@

pcre2api man page

PCRE2_USE_OFFSET_LIMIT, which provides a more general limiting facility. If PCRE2_FIRSTLINE is set with an offset limit, a match must occur in the first line and also within the offset limit. In other words, whichever limit comes -first is used. +first is used. This option has no effect for anchored patterns. 
   PCRE2_LITERAL
@@ -2021,7 +2021,7 @@

pcre2api man page

This option forces all the POSIX character classes, including [:digit:] and [:xdigit:], to match only ASCII characters, even when PCRE2_UCP is set. It can -be changed within a pattern by means of the (?aP) option setting, but note that +be changed within a pattern by means of the (?aP) option setting, but note that this also sets PCRE2_EXTRA_ASCII_DIGIT in order to ensure that (?-aP) unsets all ASCII restrictions for POSIX classes. 
@@ -4140,7 +4140,7 @@ 
pcre2api man page

 

 
REVISION

 

-Last updated: 12 October 2023
+Last updated: 11 November 2023
 

 Copyright © 1997-2023 University of Cambridge.
 

diff --git a/doc/pcre2.txt b/doc/pcre2.txt
index 8c82b1c29..b0cad0cb3 100644
--- a/doc/pcre2.txt
+++ b/doc/pcre2.txt
@@ -1653,7 +1653,8 @@ COMPILING A PATTERN
        greater than 3. See also PCRE2_USE_OFFSET_LIMIT, which provides a  more
        general  limiting  facility.  If  PCRE2_FIRSTLINE is set with an offset
        limit, a match must occur in the first line and also within the  offset
-       limit. In other words, whichever limit comes first is used.
+       limit. In other words, whichever limit comes first is used. This option
+       has no effect for anchored patterns.
 
          PCRE2_LITERAL
 
@@ -3975,11 +3976,11 @@ AUTHOR
 
 REVISION
 
-       Last updated: 12 October 2023
+       Last updated: 11 November 2023
        Copyright (c) 1997-2023 University of Cambridge.
 
 
-PCRE2 10.43                     12 October 2023                    PCRE2API(3)
+PCRE2 10.43                    11 November 2023                    PCRE2API(3)
 ------------------------------------------------------------------------------
  
  
diff --git a/doc/pcre2api.3 b/doc/pcre2api.3
index c183aeeb3..0edddbdc8 100644
--- a/doc/pcre2api.3
+++ b/doc/pcre2api.3
@@ -1,4 +1,4 @@
-.TH PCRE2API 3 "12 October 2023" "PCRE2 10.43"
+.TH PCRE2API 3 "11 November 2023" "PCRE2 10.43"
 .SH NAME
 PCRE2 - Perl-compatible regular expressions (revised API)
 .sp
@@ -1628,7 +1628,7 @@ PCRE2_FIRSTLINE if \fIstartoffset\fP is greater than 3. See also
 PCRE2_USE_OFFSET_LIMIT, which provides a more general limiting facility. If
 PCRE2_FIRSTLINE is set with an offset limit, a match must occur in the first
 line and also within the offset limit. In other words, whichever limit comes
-first is used.
+first is used. This option has no effect for anchored patterns.
 .sp
   PCRE2_LITERAL
 .sp
@@ -1979,7 +1979,7 @@ a pattern by means of the (?aT) option setting.
 .sp
 This option forces all the POSIX character classes, including [:digit:] and
 [:xdigit:], to match only ASCII characters, even when PCRE2_UCP is set. It can
-be changed within a pattern by means of the (?aP) option setting, but note that 
+be changed within a pattern by means of the (?aP) option setting, but note that
 this also sets PCRE2_EXTRA_ASCII_DIGIT in order to ensure that (?-aP) unsets
 all ASCII restrictions for POSIX classes.
 .sp
@@ -4148,6 +4148,6 @@ Cambridge, England.
 .rs
 .sp
 .nf
-Last updated: 12 October 2023
+Last updated: 11 November 2023
 Copyright (c) 1997-2023 University of Cambridge.
 .fi
diff --git a/doc/pcre2demo.3 b/doc/pcre2demo.3
index c40096eae..e5283605a 100644
--- a/doc/pcre2demo.3
+++ b/doc/pcre2demo.3
@@ -1,4 +1,4 @@
-.TH PCRE2DEMO 3 "12 October 2023" "PCRE2 10.43-DEV"
+.TH PCRE2DEMO 3 "11 November 2023" "PCRE2 10.43-DEV"
 .\"AUTOMATICALLY GENERATED BY PrepareRelease - do not EDIT!
 .SH NAME
 // - A demonstration C program for PCRE2 - //
diff --git a/src/pcre2_dfa_match.c b/src/pcre2_dfa_match.c
index 27b6ef042..acab75c42 100644
--- a/src/pcre2_dfa_match.c
+++ b/src/pcre2_dfa_match.c
@@ -3443,7 +3443,7 @@ anchored = (options & (PCRE2_ANCHORED|PCRE2_DFA_RESTART)) != 0 ||
 where to start. */
 
 startline = (re->flags & PCRE2_STARTLINE) != 0;
-firstline = (re->overall_options & PCRE2_FIRSTLINE) != 0;
+firstline = !anchored && (re->overall_options & PCRE2_FIRSTLINE) != 0;
 bumpalong_limit = end_subject;
 
 /* Initialize and set up the fixed fields in the callout block, with a pointer
diff --git a/src/pcre2_match.c b/src/pcre2_match.c
index d4d5121c4..8ae9c4999 100644
--- a/src/pcre2_match.c
+++ b/src/pcre2_match.c
@@ -6836,7 +6836,7 @@ if (mcontext == NULL)
 else mb->memctl = mcontext->memctl;
 
 anchored = ((re->overall_options | options) & PCRE2_ANCHORED) != 0;
-firstline = (re->overall_options & PCRE2_FIRSTLINE) != 0;
+firstline = !anchored && (re->overall_options & PCRE2_FIRSTLINE) != 0;
 startline = (re->flags & PCRE2_STARTLINE) != 0;
 bumpalong_limit = (mcontext->offset_limit == PCRE2_UNSET)?
   true_end_subject : subject + mcontext->offset_limit;
diff --git a/testdata/testinput2 b/testdata/testinput2
index 39e12d48a..0bb658542 100644
--- a/testdata/testinput2
+++ b/testdata/testinput2
@@ -6028,4 +6028,17 @@ a)"xI
 
 # --------
 
+/
+/anchored, firstline
+    \x0a
+
+/
+/anchored,firstline,no_start_optimize
+    \x0a
+
+/
+/firstline
+    \x0a
+    abc\x0adef
+
 # End of testinput2
diff --git a/testdata/testinput6 b/testdata/testinput6
index 0ca0d23c4..b15fe0631 100644
--- a/testdata/testinput6
+++ b/testdata/testinput6
@@ -5026,4 +5026,17 @@
 /c*+/
     ab\=ph,offset=2
 
+/
+/anchored, firstline
+    \x0a
+
+/
+/anchored,firstline,no_start_optimize
+    \x0a
+
+/
+/firstline
+    \x0a
+    abc\x0adef
+
 # End of testinput6
diff --git a/testdata/testoutput2 b/testdata/testoutput2
index 5411a401c..675d99c81 100644
--- a/testdata/testoutput2
+++ b/testdata/testoutput2
@@ -17901,6 +17901,23 @@ No match
 
 # --------
 
+/
+/anchored, firstline
+    \x0a
+ 0: \x0a
+
+/
+/anchored,firstline,no_start_optimize
+    \x0a
+ 0: \x0a
+
+/
+/firstline
+    \x0a
+ 0: \x0a
+    abc\x0adef
+ 0: \x0a
+
 # End of testinput2
 Error -70: PCRE2_ERROR_BADDATA (unknown error number)
 Error -62: bad serialized data
diff --git a/testdata/testoutput6 b/testdata/testoutput6
index 9d02ea85e..bd65130a2 100644
--- a/testdata/testoutput6
+++ b/testdata/testoutput6
@@ -7895,4 +7895,21 @@ Partial match:
     ab\=ph,offset=2
 Partial match: 
 
+/
+/anchored, firstline
+    \x0a
+ 0: \x0a
+
+/
+/anchored,firstline,no_start_optimize
+    \x0a
+ 0: \x0a
+
+/
+/firstline
+    \x0a
+ 0: \x0a
+    abc\x0adef
+ 0: \x0a
+
 # End of testinput6