rpms/pcre
6
0
Fork 0
Code Issues Pull Requests Releases Activity

8.21-RC1 bump

Browse Source
This commit is contained in:
Petr Písař 2011-12-06 19:29:30 +01:00
parent 789dda6d1e
commit 20e607628a
11 changed files with 78 additions and 949 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
.gitignore vendored
View File

@ -6,3 +6,4 @@ pcre-8.10.tar.bz2
/pcre-8.20-RC2.tar.bz2 /pcre-8.20-RC2.tar.bz2
/pcre-8.20-RC3.tar.bz2 /pcre-8.20-RC3.tar.bz2
/pcre-8.20.tar.bz2 /pcre-8.20.tar.bz2
/pcre-8.21-RC1.tar.bz2

38
pcre-8.10-multilib.patch
View File

@ -1,38 +0,0 @@
diff -Naur pcre-8.10.orig/pcre-config.in pcre-8.10/pcre-config.in
--- pcre-8.10.orig/pcre-config.in 2010-03-02 12:08:45.000000000 +0100
+++ pcre-8.10/pcre-config.in 2010-07-12 14:15:30.666700210 +0200
@@ -15,16 +15,6 @@
exit 1
fi
-libR=
-case `uname -s` in
- *SunOS*)
- libR=" -R@libdir@"
- ;;
- *BSD*)
- libR=" -Wl,-R@libdir@"
- ;;
-esac
-
while test $# -gt 0; do
case "$1" in
-*=*) optarg=`echo "$1" | sed 's/[-_a-zA-Z0-9]*=//'` ;;
@@ -58,14 +48,14 @@
echo $includes @PCRE_STATIC_CFLAG@
;;
--libs-posix)
- echo -L@libdir@$libR -lpcreposix -lpcre
+ echo -lpcreposix -lpcre
;;
--libs)
- echo -L@libdir@$libR -lpcre
+ echo -lpcre
;;
--libs-cpp)
if test @enable_cpp@ = yes ; then
- echo -L@libdir@$libR -lpcrecpp -lpcre
+ echo -lpcrecpp -lpcre
else
echo "${usage}" 1>&2
fi

150
pcre-8.20-caseless_different_length.patch
View File

@ -1,150 +0,0 @@
From 72a4bb52e09d46af0b00dd4064f93e9948fdad51 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Petr=20P=C3=ADsa=C5=99?= <ppisar@redhat.com>
Date: Fri, 2 Dec 2011 11:36:54 +0100
Subject: [PATCH] Fix caseless match if cases differ in encoding length
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
From:
r778 | ph10 | 2011-12-01 18:38:47 +0100 (Čt, 01 pro 2011) | 3 lines
Fix bug with caseless matching of characters of different lengths when
the shorter is right at the end of the subject.
Petr Pisar: Changelog entry removed.
---
pcre_exec.c | 32 ++++++++++++++++----------------
testdata/testinput6 | 14 ++++++++++++++
testdata/testoutput6 | 22 ++++++++++++++++++++++
3 files changed, 52 insertions(+), 16 deletions(-)
diff --git a/pcre_exec.c b/pcre_exec.c
index 2e763d1..9881bdd 100644
--- a/pcre_exec.c
+++ b/pcre_exec.c
@@ -427,7 +427,7 @@ returns a negative (error) response, the outer incarnation must also return the
same response. */
/* These macros pack up tests that are used for partial matching, and which
-appears several times in the code. We set the "hit end" flag if the pointer is
+appear several times in the code. We set the "hit end" flag if the pointer is
at the end of the subject and also past the start of the subject (i.e.
something has been matched). For hard partial matching, we then return
immediately. The second one is used when we already know we are past the end of
@@ -3039,31 +3039,36 @@ for (;;)
}
break;
- /* Match a single character, caselessly */
+ /* Match a single character, caselessly. If we are at the end of the
+ subject, give up immediately. */
case OP_CHARI:
+ if (eptr >= md->end_subject)
+ {
+ SCHECK_PARTIAL();
+ MRRETURN(MATCH_NOMATCH);
+ }
+
#ifdef SUPPORT_UTF8
if (utf8)
{
length = 1;
ecode++;
GETCHARLEN(fc, ecode, length);
-
- if (length > md->end_subject - eptr)
- {
- CHECK_PARTIAL(); /* Not SCHECK_PARTIAL() */
- MRRETURN(MATCH_NOMATCH);
- }
-
+
/* If the pattern character's value is < 128, we have only one byte, and
- can use the fast lookup table. */
+ we know that its other case must also be one byte long, so we can use the
+ fast lookup table. We know that there is at least one byte left in the
+ subject. */
if (fc < 128)
{
if (md->lcc[*ecode++] != md->lcc[*eptr++]) MRRETURN(MATCH_NOMATCH);
}
- /* Otherwise we must pick up the subject character */
+ /* Otherwise we must pick up the subject character. Note that we cannot
+ use the value of "length" to check for sufficient bytes left, because the
+ other case of the character may have more or fewer bytes. */
else
{
@@ -3088,11 +3093,6 @@ for (;;)
/* Non-UTF-8 mode */
{
- if (md->end_subject - eptr < 1)
- {
- SCHECK_PARTIAL(); /* This one can use SCHECK_PARTIAL() */
- MRRETURN(MATCH_NOMATCH);
- }
if (md->lcc[ecode[1]] != md->lcc[*eptr++]) MRRETURN(MATCH_NOMATCH);
ecode += 2;
}
diff --git a/testdata/testinput6 b/testdata/testinput6
index e5fc0e9..6b0d2f7 100644
--- a/testdata/testinput6
+++ b/testdata/testinput6
@@ -802,4 +802,18 @@
** Failers
a\xFCb
+/ⱥ/8i
+ ⱥ
+ Ⱥx
+ Ⱥ
+
+/[ⱥ]/8i
+ ⱥ
+ Ⱥx
+ Ⱥ
+
+/Ⱥ/8i
+ Ⱥ
+ ⱥ
+
/-- End of testinput6 --/
diff --git a/testdata/testoutput6 b/testdata/testoutput6
index 1acaa23..68c0a46 100644
--- a/testdata/testoutput6
+++ b/testdata/testoutput6
@@ -1353,4 +1353,26 @@ No match
a\xFCb
No match
+/ⱥ/8i
+ ⱥ
+ 0: \x{2c65}
+ Ⱥx
+ 0: \x{23a}
+ Ⱥ
+ 0: \x{23a}
+
+/[ⱥ]/8i
+ ⱥ
+ 0: \x{2c65}
+ Ⱥx
+ 0: \x{23a}
+ Ⱥ
+ 0: \x{23a}
+
+/Ⱥ/8i
+ Ⱥ
+ 0: \x{23a}
+ ⱥ
+ 0: \x{2c65}
+
/-- End of testinput6 --/
--
1.7.7.4

69
pcre-8.20-forward_reference.patch
View File

@ -1,69 +0,0 @@
Fix repeated forward reference needed character bug
From:
r762 | ph10 | 2011-11-22 14:36:51 +0100 (Út, 22 lis 2011) | 2 lines
16. A repeated forward reference in a pattern such as (a)(?2){2}(.) was
incorrectly expecting the subject to contain another "a" after the start.
Petr Pisar: Changelog removed.
See <http://lists.pcre.org/lurker/message/20111121.203922.3cfd476e.en.html>.
Index: testdata/testoutput11
===================================================================
--- testdata/testoutput11 (revision 761)
+++ testdata/testoutput11 (revision 762)
@@ -1441,4 +1441,10 @@
xabcd
0: c
+/(a)(?2){2}(.)/
+ abcd
+ 0: abcd
+ 1: a
+ 2: d
+
/-- End of testinput11 --/
Index: testdata/testinput11
===================================================================
--- testdata/testinput11 (revision 761)
+++ testdata/testinput11 (revision 762)
@@ -800,4 +800,7 @@
/(?<=a(*THEN)b)c/
xabcd
+/(a)(?2){2}(.)/
+ abcd
+
/-- End of testinput11 --/
Index: pcre_compile.c
===================================================================
--- pcre_compile.c (revision 761)
+++ pcre_compile.c (revision 762)
@@ -4884,7 +4884,8 @@
*lengthptr += delta;
}
- /* This is compiling for real */
+ /* This is compiling for real. If there is a set first byte for
+ the group, and we have not yet set a "required byte", set it. */
else
{
@@ -6015,11 +6016,14 @@
}
}
- /* Insert the recursion/subroutine item. */
+ /* Insert the recursion/subroutine item. It does not have a set first
+ byte (relevant if it is repeated, because it will then be wrapped
+ with ONCE brackets). */
*code = OP_RECURSE;
PUT(code, 1, (int)(called - cd->start_code));
code += 1 + LINK_SIZE;
+ groupsetfirstbyte = FALSE;
}
/* Can't determine a first byte now */

473
pcre-8.20-lookbehind-2.patch
View File

@ -1,473 +0,0 @@
From 5d9a1b3aee83b5068ab2635e474c3d75a0277e1c Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Petr=20P=C3=ADsa=C5=99?= <ppisar@redhat.com>
Date: Wed, 16 Nov 2011 13:18:09 +0100
Subject: [PATCH] Fixed several items that were being incorrectly rejected as
"not fixed length" in lookbehinds.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
From SVN tree:
r747 | ph10 | 2011-11-15 18:35:10 +0100 (Út, 15 lis 2011) | 3 lines
While fixing 6 above, I noticed that a number of other items were being
incorrectly rejected as "not fixed length". This arose partly because newer
opcodes had not been added to the fixed-length checking code. I have (a)
corrected the bug and added tests for these items, and (b) arranged for an
error to occur if an unknown opcode is encountered while checking for fixed
length instead of just assuming "not fixed length". The items that were
rejected were: (*ACCEPT), (*COMMIT), (*FAIL), (*MARK), (*PRUNE), (*SKIP),
(*THEN), \h, \H, \v, \V, and single character negative classes with fixed
repetitions, e.g. [^a]{3}, with and without PCRE_CASELESS.
Petr Pisar: Remove change log entry.
See <https://lists.exim.org/lurker/message/20111115.175054.077a216c.en.html>.
---
pcre_compile.c | 154 ++++++++++++++++++++++++++++++++++++++++++-------
pcre_internal.h | 2 +-
pcreposix.c | 2 +
testdata/testinput1 | 24 ++++++++
testdata/testinput11 | 27 +++++++++
testdata/testoutput1 | 36 +++++++++++
testdata/testoutput11 | 41 +++++++++++++
7 files changed, 263 insertions(+), 23 deletions(-)
diff --git a/pcre_compile.c b/pcre_compile.c
index 588e981..27c8240 100644
--- a/pcre_compile.c
+++ b/pcre_compile.c
@@ -410,6 +410,8 @@ static const char error_texts[] =
"this version of PCRE is not compiled with PCRE_UCP support\0"
"\\c must be followed by an ASCII character\0"
"\\k is not followed by a braced, angle-bracketed, or quoted name\0"
+ /* 70 */
+ "internal error: unknown opcode in find_fixedlength()\0"
;
/* Table to identify digits and hex digits. This is used when compiling
@@ -1477,6 +1479,7 @@ Returns: the fixed length,
or -1 if there is no fixed length,
or -2 if \C was encountered
or -3 if an OP_RECURSE item was encountered and atend is FALSE
+ or -4 if an unknown opcode was encountered (internal error)
*/
static int
@@ -1500,8 +1503,7 @@ for (;;)
/* We only need to continue for OP_CBRA (normal capturing bracket) and
OP_BRA (normal non-capturing bracket) because the other variants of these
opcodes are all concerned with unlimited repeated groups, which of course
- are not of fixed length. They will cause a -1 response from the default
- case of this switch. */
+ are not of fixed length. */
case OP_CBRA:
case OP_BRA:
@@ -1515,15 +1517,17 @@ for (;;)
cc += 1 + LINK_SIZE;
break;
- /* Reached end of a branch; if it's a ket it is the end of a nested
- call. If it's ALT it is an alternation in a nested call. If it is
- END it's the end of the outer call. All can be handled by the same code.
- Note that we must not include the OP_KETRxxx opcodes here, because they
- all imply an unlimited repeat. */
+ /* Reached end of a branch; if it's a ket it is the end of a nested call.
+ If it's ALT it is an alternation in a nested call. An ACCEPT is effectively
+ an ALT. If it is END it's the end of the outer call. All can be handled by
+ the same code. Note that we must not include the OP_KETRxxx opcodes here,
+ because they all imply an unlimited repeat. */
case OP_ALT:
case OP_KET:
case OP_END:
+ case OP_ACCEPT:
+ case OP_ASSERT_ACCEPT:
if (length < 0) length = branchlength;
else if (length != branchlength) return -1;
if (*cc != OP_ALT) return length;
@@ -1557,23 +1561,36 @@ for (;;)
/* Skip over things that don't match chars */
- case OP_REVERSE:
- case OP_CREF:
- case OP_NCREF:
- case OP_RREF:
- case OP_NRREF:
- case OP_DEF:
+ case OP_MARK:
+ case OP_PRUNE_ARG:
+ case OP_SKIP_ARG:
+ case OP_THEN_ARG:
+ cc += cc[1] + _pcre_OP_lengths[*cc];
+ break;
+
case OP_CALLOUT:
- case OP_SOD:
- case OP_SOM:
- case OP_SET_SOM:
- case OP_EOD:
- case OP_EODN:
case OP_CIRC:
case OP_CIRCM:
+ case OP_CLOSE:
+ case OP_COMMIT:
+ case OP_CREF:
+ case OP_DEF:
case OP_DOLL:
case OP_DOLLM:
+ case OP_EOD:
+ case OP_EODN:
+ case OP_FAIL:
+ case OP_NCREF:
+ case OP_NRREF:
case OP_NOT_WORD_BOUNDARY:
+ case OP_PRUNE:
+ case OP_REVERSE:
+ case OP_RREF:
+ case OP_SET_SOM:
+ case OP_SKIP:
+ case OP_SOD:
+ case OP_SOM:
+ case OP_THEN:
case OP_WORD_BOUNDARY:
cc += _pcre_OP_lengths[*cc];
break;
@@ -1595,7 +1612,9 @@ for (;;)
need to skip over a multibyte character in UTF8 mode. */
case OP_EXACT:
- case OP_EXACTI:
+ case OP_EXACTI:
+ case OP_NOTEXACT:
+ case OP_NOTEXACTI:
branchlength += GET2(cc,1);
cc += 4;
#ifdef SUPPORT_UTF8
@@ -1616,6 +1635,10 @@ for (;;)
cc += 2;
/* Fall through */
+ case OP_HSPACE:
+ case OP_VSPACE:
+ case OP_NOT_HSPACE:
+ case OP_NOT_VSPACE:
case OP_NOT_DIGIT:
case OP_DIGIT:
case OP_NOT_WHITESPACE:
@@ -1647,6 +1670,8 @@ for (;;)
switch (*cc)
{
+ case OP_CRPLUS:
+ case OP_CRMINPLUS:
case OP_CRSTAR:
case OP_CRMINSTAR:
case OP_CRQUERY:
@@ -1667,8 +1692,91 @@ for (;;)
/* Anything else is variable length */
- default:
+ case OP_ANYNL:
+ case OP_BRAMINZERO:
+ case OP_BRAPOS:
+ case OP_BRAPOSZERO:
+ case OP_BRAZERO:
+ case OP_CBRAPOS:
+ case OP_EXTUNI:
+ case OP_KETRMAX:
+ case OP_KETRMIN:
+ case OP_KETRPOS:
+ case OP_MINPLUS:
+ case OP_MINPLUSI:
+ case OP_MINQUERY:
+ case OP_MINQUERYI:
+ case OP_MINSTAR:
+ case OP_MINSTARI:
+ case OP_MINUPTO:
+ case OP_MINUPTOI:
+ case OP_NOTMINPLUS:
+ case OP_NOTMINPLUSI:
+ case OP_NOTMINQUERY:
+ case OP_NOTMINQUERYI:
+ case OP_NOTMINSTAR:
+ case OP_NOTMINSTARI:
+ case OP_NOTMINUPTO:
+ case OP_NOTMINUPTOI:
+ case OP_NOTPLUS:
+ case OP_NOTPLUSI:
+ case OP_NOTPOSPLUS:
+ case OP_NOTPOSPLUSI:
+ case OP_NOTPOSQUERY:
+ case OP_NOTPOSQUERYI:
+ case OP_NOTPOSSTAR:
+ case OP_NOTPOSSTARI:
+ case OP_NOTPOSUPTO:
+ case OP_NOTPOSUPTOI:
+ case OP_NOTQUERY:
+ case OP_NOTQUERYI:
+ case OP_NOTSTAR:
+ case OP_NOTSTARI:
+ case OP_NOTUPTO:
+ case OP_NOTUPTOI:
+ case OP_PLUS:
+ case OP_PLUSI:
+ case OP_POSPLUS:
+ case OP_POSPLUSI:
+ case OP_POSQUERY:
+ case OP_POSQUERYI:
+ case OP_POSSTAR:
+ case OP_POSSTARI:
+ case OP_POSUPTO:
+ case OP_POSUPTOI:
+ case OP_QUERY:
+ case OP_QUERYI:
+ case OP_REF:
+ case OP_REFI:
+ case OP_SBRA:
+ case OP_SBRAPOS:
+ case OP_SCBRA:
+ case OP_SCBRAPOS:
+ case OP_SCOND:
+ case OP_SKIPZERO:
+ case OP_STAR:
+ case OP_STARI:
+ case OP_TYPEMINPLUS:
+ case OP_TYPEMINQUERY:
+ case OP_TYPEMINSTAR:
+ case OP_TYPEMINUPTO:
+ case OP_TYPEPLUS:
+ case OP_TYPEPOSPLUS:
+ case OP_TYPEPOSQUERY:
+ case OP_TYPEPOSSTAR:
+ case OP_TYPEPOSUPTO:
+ case OP_TYPEQUERY:
+ case OP_TYPESTAR:
+ case OP_TYPEUPTO:
+ case OP_UPTO:
+ case OP_UPTOI:
return -1;
+
+ /* Catch unrecognized opcodes so that when new ones are added they
+ are not forgotten, as has happened in the past. */
+
+ default:
+ return -4;
}
}
/* Control never gets here */
@@ -6564,7 +6672,8 @@ for (;;)
}
else if (fixed_length < 0)
{
- *errorcodeptr = (fixed_length == -2)? ERR36 : ERR25;
+ *errorcodeptr = (fixed_length == -2)? ERR36 :
+ (fixed_length == -4)? ERR70: ERR25;
*ptrptr = ptr;
return FALSE;
}
@@ -7363,7 +7472,8 @@ if (cd->check_lookbehind)
DPRINTF(("fixed length = %d\n", fixed_length));
if (fixed_length < 0)
{
- errorcode = (fixed_length == -2)? ERR36 : ERR25;
+ errorcode = (fixed_length == -2)? ERR36 :
+ (fixed_length == -4)? ERR70 : ERR25;
break;
}
PUT(cc, 1, fixed_length);
diff --git a/pcre_internal.h b/pcre_internal.h
index faf1b76..2d02e5d 100644
--- a/pcre_internal.h
+++ b/pcre_internal.h
@@ -1665,7 +1665,7 @@ enum { ERR0, ERR1, ERR2, ERR3, ERR4, ERR5, ERR6, ERR7, ERR8, ERR9,
ERR40, ERR41, ERR42, ERR43, ERR44, ERR45, ERR46, ERR47, ERR48, ERR49,
ERR50, ERR51, ERR52, ERR53, ERR54, ERR55, ERR56, ERR57, ERR58, ERR59,
ERR60, ERR61, ERR62, ERR63, ERR64, ERR65, ERR66, ERR67, ERR68, ERR69,
- ERRCOUNT };
+ ERR70, ERRCOUNT };
/* The real format of the start of the pcre block; the index of names and the
code vector run on as long as necessary after the end. We store an explicit
diff --git a/pcreposix.c b/pcreposix.c
index 2061be0..648254b 100644
--- a/pcreposix.c
+++ b/pcreposix.c
@@ -153,6 +153,8 @@ static const int eint[] = {
REG_INVARG, /* this version of PCRE is not compiled with PCRE_UCP support */
REG_BADPAT, /* \c must be followed by an ASCII character */
REG_BADPAT, /* \k is not followed by a braced, angle-bracketed, or quoted name */
+ /* 70 */
+ REG_BADPAT, /* internal error: unknown opcode in find_fixedlength() */
};
/* Table of texts corresponding to POSIX error codes */
diff --git a/testdata/testinput1 b/testdata/testinput1
index aa9ce42..b24f900 100644
--- a/testdata/testinput1
+++ b/testdata/testinput1
@@ -4261,4 +4261,28 @@
** Failers
xaabc
+/(?<=a\h)c/
+ xa c
+
+/(?<=[^a]{2})b/
+ axxbc
+ aAAbc
+ ** Failers
+ xaabc
+
+/(?<=[^a]{2})b/i
+ axxbc
+ ** Failers
+ aAAbc
+ xaabc
+
+/(?<=a\H)c/
+ abc
+
+/(?<=a\V)c/
+ abc
+
+/(?<=a\v)c/
+ a\nc
+
/-- End of testinput1 --/
diff --git a/testdata/testinput11 b/testdata/testinput11
index 198dbf2..37ee38b 100644
--- a/testdata/testinput11
+++ b/testdata/testinput11
@@ -767,4 +767,31 @@ name)/K
/------------------------------/
+/(?<=a(*ACCEPT)b)c/
+ xacd
+
+/(?<=(a(*ACCEPT)b))c/
+ xacd
+
+/(?<=(a(*COMMIT)b))c/
+ xabcd
+ ** Failers
+ xacd
+
+/(?<!a(*FAIL)b)c/
+ xcd
+ acd
+
+/(?<=a(*:N)b)c/K
+ xabcd
+
+/(?<=a(*PRUNE)b)c/
+ xabcd
+
+/(?<=a(*SKIP)b)c/
+ xabcd
+
+/(?<=a(*THEN)b)c/
+ xabcd
+
/-- End of testinput11 --/
diff --git a/testdata/testoutput1 b/testdata/testoutput1
index 5a025e2..0c2e84e 100644
--- a/testdata/testoutput1
+++ b/testdata/testoutput1
@@ -6968,4 +6968,40 @@ No match
xaabc
No match
+/(?<=a\h)c/
+ xa c
+ 0: c
+
+/(?<=[^a]{2})b/
+ axxbc
+ 0: b
+ aAAbc
+ 0: b
+ ** Failers
+No match
+ xaabc
+No match
+
+/(?<=[^a]{2})b/i
+ axxbc
+ 0: b
+ ** Failers
+No match
+ aAAbc
+No match
+ xaabc
+No match
+
+/(?<=a\H)c/
+ abc
+ 0: c
+
+/(?<=a\V)c/
+ abc
+ 0: c
+
+/(?<=a\v)c/
+ a\nc
+ 0: c
+
/-- End of testinput1 --/
diff --git a/testdata/testoutput11 b/testdata/testoutput11
index 4af2e92..8a9d6c2 100644
--- a/testdata/testoutput11
+++ b/testdata/testoutput11
@@ -1389,4 +1389,45 @@ No match
/------------------------------/
+/(?<=a(*ACCEPT)b)c/
+ xacd
+ 0: c
+
+/(?<=(a(*ACCEPT)b))c/
+ xacd
+ 0: c
+ 1: a
+
+/(?<=(a(*COMMIT)b))c/
+ xabcd
+ 0: c
+ 1: ab
+ ** Failers
+No match
+ xacd
+No match
+
+/(?<!a(*FAIL)b)c/
+ xcd
+ 0: c
+ acd
+ 0: c
+
+/(?<=a(*:N)b)c/K
+ xabcd
+ 0: c
+MK: N
+
+/(?<=a(*PRUNE)b)c/
+ xabcd
+ 0: c
+
+/(?<=a(*SKIP)b)c/
+ xabcd
+ 0: c
+
+/(?<=a(*THEN)b)c/
+ xabcd
+ 0: c
+
/-- End of testinput11 --/
--
1.7.6.4

84
pcre-8.20-lookbehind.patch
View File

@ -1,84 +0,0 @@
From ad81574f3cd7aef3e7d9106d6fa86901556b1731 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Petr=20P=C3=ADsa=C5=99?= <ppisar@redhat.com>
Date: Tue, 15 Nov 2011 16:34:01 +0100
Subject: [PATCH] Caseless matching of backreferences with fixed length
repetitions was broken.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
From development SVN tree:
r746 | ph10 | 2011-11-15 16:07:02 +0100 (Út, 15 lis 2011) | 2 lines
6. Lookbehinds such as (?<=a{2}b) that contained a fixed repetition were
erroneously being rejected as "not fixed length" if PCRE_CASELESS was set.
This bug was probably introduced by change 9 of 8.13.
Petr Pisar: Changelog entry removed.
See <https://lists.exim.org/lurker/thread/20111115.150810.f771a73e.en.html>.
---
pcre_compile.c | 1 +
testdata/testinput1 | 10 ++++++++++
testdata/testoutput1 | 16 ++++++++++++++++
3 files changed, 27 insertions(+), 0 deletions(-)
diff --git a/pcre_compile.c b/pcre_compile.c
index 3fc7c82..588e981 100644
--- a/pcre_compile.c
+++ b/pcre_compile.c
@@ -1595,6 +1595,7 @@ for (;;)
need to skip over a multibyte character in UTF8 mode. */
case OP_EXACT:
+ case OP_EXACTI:
branchlength += GET2(cc,1);
cc += 4;
#ifdef SUPPORT_UTF8
diff --git a/testdata/testinput1 b/testdata/testinput1
index c913ee4..aa9ce42 100644
--- a/testdata/testinput1
+++ b/testdata/testinput1
@@ -4251,4 +4251,14 @@
/[:a]xxx[b:]/
:xxx:
+/(?<=a{2})b/i
+ xaabc
+ ** Failers
+ xabc
+
+/(?<!a{2})b/i
+ xabc
+ ** Failers
+ xaabc
+
/-- End of testinput1 --/
diff --git a/testdata/testoutput1 b/testdata/testoutput1
index 030f04f..5a025e2 100644
--- a/testdata/testoutput1
+++ b/testdata/testoutput1
@@ -6952,4 +6952,20 @@ No match
:xxx:
0: :xxx:
+/(?<=a{2})b/i
+ xaabc
+ 0: b
+ ** Failers
+No match
+ xabc
+No match
+
+/(?<!a{2})b/i
+ xabc
+ 0: b
+ ** Failers
+No match
+ xaabc
+No match
+
/-- End of testinput1 --/
--
1.7.6.4

112
pcre-8.20-ppcjit.patch
View File

@ -1,112 +0,0 @@
Fix cache-flush on PPC
From:
r742 | zherczeg | 2011-11-06 09:05:33 +0100 (Ne, 06 lis 2011) | 3 lines
Fix cache-flush issue on PowerPC, adding some comments and a check for
disabled PCRE_EXTRA_TABLES.
Fix cache-flush issue on PowerPC (It is still an experimental JIT port).
PCRE_EXTRA_TABLES is not suported by JIT, and should be checked before
calling _pcre_jit_exec. Some extra comments are added.
Petr Pisar: Changelog removed
Index: pcre_exec.c
===================================================================
--- pcre_exec.c (revision 741)
+++ pcre_exec.c (revision 742)
@@ -6011,6 +6011,7 @@
if (extra_data != NULL
&& (extra_data->flags & PCRE_EXTRA_EXECUTABLE_JIT) != 0
&& extra_data->executable_jit != NULL
+ && (extra_data->flags & PCRE_EXTRA_TABLES) == 0
&& (options & ~(PCRE_NO_UTF8_CHECK | PCRE_NOTBOL | PCRE_NOTEOL |
PCRE_NOTEMPTY | PCRE_NOTEMPTY_ATSTART)) == 0)
return _pcre_jit_exec(re, extra_data->executable_jit, subject, length,
Index: sljit/sljitLir.h
===================================================================
--- sljit/sljitLir.h (revision 741)
+++ sljit/sljitLir.h (revision 742)
@@ -56,6 +56,9 @@
- mainly position independent code
- Optimizations (perhaps later)
- Only for basic blocks (when no labels inserted between LIR instructions)
+
+ For valgrind users:
+ - pass --smc-check=all argument to valgrind, since JIT is a "self-modifying code"
*/
#if !(defined SLJIT_NO_DEFAULT_CONFIG && SLJIT_NO_DEFAULT_CONFIG)
@@ -87,6 +90,7 @@
#define SLJIT_UNUSED 0
+/* Temporary (scratch) registers may not preserve their values across function calls. */
#define SLJIT_TEMPORARY_REG1 1
#define SLJIT_TEMPORARY_REG2 2
#define SLJIT_TEMPORARY_REG3 3
@@ -95,6 +99,7 @@
#define SLJIT_TEMPORARY_EREG1 4
#define SLJIT_TEMPORARY_EREG2 5
+/* General (saved) registers preserve their values across function calls. */
#define SLJIT_GENERAL_REG1 6
#define SLJIT_GENERAL_REG2 7
#define SLJIT_GENERAL_REG3 8
Index: sljit/sljitNativePPC_common.c
===================================================================
--- sljit/sljitNativePPC_common.c (revision 741)
+++ sljit/sljitNativePPC_common.c (revision 742)
@@ -37,6 +37,18 @@
Both for ppc-32 and ppc-64. */
typedef sljit_ui sljit_ins;
+static void ppc_cache_flush(sljit_ins *from, sljit_ins *to)
+{
+ while (from < to) {
+#ifdef __GNUC__
+ asm volatile ( "icbi 0, %0" : : "r"(from) );
+#else
+#error "Must implement icbi"
+#endif
+ from++;
+ }
+}
+
#define TMP_REG1 (SLJIT_NO_REGISTERS + 1)
#define TMP_REG2 (SLJIT_NO_REGISTERS + 2)
#define TMP_REG3 (SLJIT_NO_REGISTERS + 3)
Index: sljit/sljitConfigInternal.h
===================================================================
--- sljit/sljitConfigInternal.h (revision 741)
+++ sljit/sljitConfigInternal.h (revision 742)
@@ -178,13 +178,23 @@
#ifndef SLJIT_CACHE_FLUSH
-#if !(defined SLJIT_CONFIG_X86_32 && SLJIT_CONFIG_X86_32) && !(defined SLJIT_CONFIG_X86_64 && SLJIT_CONFIG_X86_64)
- /* Just call __ARM_NR_cacheflush on Linux. */
+#if (defined SLJIT_CONFIG_PPC_32 && SLJIT_CONFIG_PPC_32) || (defined SLJIT_CONFIG_PPC_64 && SLJIT_CONFIG_PPC_64)
+
+/* The __clear_cache() implementation of GCC is a dummy function on PowerPC. */
#define SLJIT_CACHE_FLUSH(from, to) \
+ ppc_cache_flush((from), (to))
+
+#elif (defined SLJIT_CONFIG_X86_32 && SLJIT_CONFIG_X86_32) || (defined SLJIT_CONFIG_X86_64 && SLJIT_CONFIG_X86_64)
+
+/* Not required to implement on archs with unified caches. */
+#define SLJIT_CACHE_FLUSH(from, to)
+
+#else
+
+/* Calls __ARM_NR_cacheflush on ARM-Linux. */
+#define SLJIT_CACHE_FLUSH(from, to) \
__clear_cache((char*)(from), (char*)(to))
-#else
- /* Not required to implement on archs with unified caches. */
-#define SLJIT_CACHE_FLUSH(from, to)
+
#endif
#endif /* !SLJIT_CACHE_FLUSH */

25
pcre-8.21-Make-pcre-config-libs-cpp-aware-of-cross-compilation.patch Normal file
View File

@ -0,0 +1,25 @@
From abedf373e5a7032af6a37cc7f96aac15bf5c26dd Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Petr=20P=C3=ADsa=C5=99?= <ppisar@redhat.com>
Date: Tue, 6 Dec 2011 19:13:31 +0100
Subject: [PATCH] Make pcre-config --libs-cpp aware of cross-compilation
---
pcre-config.in | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/pcre-config.in b/pcre-config.in
index aeee182..00c5395 100644
--- a/pcre-config.in
+++ b/pcre-config.in
@@ -58,7 +58,7 @@ while test $# -gt 0; do
;;
--libs-cpp)
if test @enable_cpp@ = yes ; then
- echo -L@libdir@$libR -lpcrecpp -lpcre
+ echo $libS$libR -lpcrecpp -lpcre
else
echo "${usage}" 1>&2
fi
--
1.7.7.4

37
pcre-8.21-multilib.patch Normal file
View File

@ -0,0 +1,37 @@
From 1d5ce3e2e6341b01609aefab20786a8638fee17a Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Petr=20P=C3=ADsa=C5=99?= <ppisar@redhat.com>
Date: Tue, 6 Dec 2011 18:44:11 +0100
Subject: [PATCH] Fix multilib
Do not set RPATH nor add explicit -L path to compiler.
---
pcre-config.in | 12 ------------
1 files changed, 0 insertions(+), 12 deletions(-)
diff --git a/pcre-config.in b/pcre-config.in
index ccbf210..aeee182 100644
--- a/pcre-config.in
+++ b/pcre-config.in
@@ -16,19 +16,7 @@ if test $# -eq 0; then
fi
libR=
-case `uname -s` in
- *SunOS*)
- libR=" -R@libdir@"
- ;;
- *BSD*)
- libR=" -Wl,-R@libdir@"
- ;;
-esac
-
libS=
-if test @libdir@ != /usr/lib ; then
- libS=-L@libdir@
-fi
while test $# -gt 0; do
case "$1" in
--
1.7.7.4

36
pcre.spec
View File

@ -1,27 +1,20 @@
# This is stable release: %%global rcversion RC3 # This is stable release:
%global rcversion RC1
Name: pcre Name: pcre
Version: 8.20 Version: 8.21
Release: %{?rcversion:0.}7%{?rcversion:.%rcversion}%{?dist} Release: %{?rcversion:0.}1%{?rcversion:.%rcversion}%{?dist}
%global myversion %{version}%{?rcversion:-%rcversion} %global myversion %{version}%{?rcversion:-%rcversion}
Summary: Perl-compatible regular expression library Summary: Perl-compatible regular expression library
Group: System Environment/Libraries Group: System Environment/Libraries
License: BSD License: BSD
URL: http://www.pcre.org/ URL: http://www.pcre.org/
Source: ftp://ftp.csx.cam.ac.uk/pub/software/programming/%{name}/%{?rcversion:Testing/}%{name}-%{myversion}.tar.bz2 Source: ftp://ftp.csx.cam.ac.uk/pub/software/programming/%{name}/%{?rcversion:Testing/}%{name}-%{myversion}.tar.bz2
Patch0: pcre-8.10-multilib.patch # Upstream thinks RPATH is good idea.
Patch0: pcre-8.21-multilib.patch
# Consolidate pcre-config --libs-cpp, sent to upstream.
Patch1: pcre-8.21-Make-pcre-config-libs-cpp-aware-of-cross-compilation.patch
# Refused by upstream, bug #675477 # Refused by upstream, bug #675477
Patch1: pcre-8.20-refused_spelling_terminated.patch Patch2: pcre-8.20-refused_spelling_terminated.patch
# Fix look-behind regression, in upstream after 8.20.
Patch2: pcre-8.20-lookbehind.patch
# Fix other look-behind regression, in upstream after 8.20.
Patch3: pcre-8.20-lookbehind-2.patch
# Fix repeated forward reference, in upstream after 8.20.
Patch4: pcre-8.20-forward_reference.patch
# Fix cache-flush in JIT on PPC, in upstream after 8.20.
Patch5: pcre-8.20-ppcjit.patch
# Fix case-less match if cases differ in encoding length, in upstream after
# 8.20.
Patch6: pcre-8.20-caseless_different_length.patch
BuildRequires: readline-devel BuildRequires: readline-devel
# New libtool to get rid of rpath # New libtool to get rid of rpath
BuildRequires: autoconf, automake, libtool BuildRequires: autoconf, automake, libtool
@ -60,13 +53,9 @@ Utilities demonstrating PCRE capabilities like pcregrep or pcretest.
%setup -q -n %{name}-%{myversion} %setup -q -n %{name}-%{myversion}
# Get rid of rpath # Get rid of rpath
%patch0 -p1 -b .multilib %patch0 -p1 -b .multilib
%patch1 -p1 -b .cpp
libtoolize --copy --force && autoreconf libtoolize --copy --force && autoreconf
%patch1 -p1 -b .terminated_typos %patch2 -p1 -b .terminated_typos
%patch2 -p1 -b .lookbehind
%patch3 -p1 -b .lookbehind2
%patch4 -p0 -b .forward_reference
%patch5 -p0 -b .ppcjit
%patch6 -p1 -b .caseless_different_length
# One contributor's name is non-UTF-8 # One contributor's name is non-UTF-8
for F in ChangeLog; do for F in ChangeLog; do
iconv -f latin1 -t utf8 "$F" >"${F}.utf8" iconv -f latin1 -t utf8 "$F" >"${F}.utf8"
@ -137,6 +126,9 @@ make check
%{_mandir}/man1/pcretest.* %{_mandir}/man1/pcretest.*
%changelog %changelog
* Tue Dec 06 2011 Petr Pisar <ppisar@redhat.com> - 8.21-0.1.RC1
- 8.21-RC1 bump
* Fri Dec 02 2011 Petr Pisar <ppisar@redhat.com> - 8.20-7 * Fri Dec 02 2011 Petr Pisar <ppisar@redhat.com> - 8.20-7
- Fix case-less match if cases differ in encoding length (bug #756675) - Fix case-less match if cases differ in encoding length (bug #756675)

2
sources
View File

@ -1 +1 @@
a1931c70e1273e3450d5036fe273d25c pcre-8.20.tar.bz2 3aaead13e66d3ad4caa2795a2080e08e pcre-8.21-RC1.tar.bz2