Compare commits
No commits in common. 'c9' and 'i8c' have entirely different histories.
@ -1 +1 @@
|
||||
SOURCES/pcre-8.44.tar.bz2
|
||||
SOURCES/pcre-8.42.tar.bz2
|
||||
|
||||
@ -1 +1 @@
|
||||
8179b083053fce9b4a766513fa1f14807aabee42 SOURCES/pcre-8.44.tar.bz2
|
||||
df0d1c2ff04c359220cb902539a6e134af4497f4 SOURCES/pcre-8.42.tar.bz2
|
||||
|
||||
@ -1,39 +0,0 @@
|
||||
This is Philip Hazel's public GPG key.
|
||||
|
||||
-----BEGIN PGP PUBLIC KEY BLOCK-----
|
||||
Version: GnuPG v1.2.0 (SunOS)
|
||||
|
||||
mQEMBD2z9g0BCACpGWJ4K2mjR6K9nv7/K+HLjJtDO0E0ckudeNjg2JvnuUU7m1bK
|
||||
87nSklHTSMNiqmuVU5hFAuGiMGBMj31GPpce3VbABdm6TmAJB8ZndvJXyf8lC7Ko
|
||||
5iZX/gCYU6817UI5pFwYR3MTIUSSxTGqk1y62pYLKUq9keSU5Cew4q1KMwl46Bn3
|
||||
SwlfQeST4WGwL2w9b8mWNltBBY23bgzwYu7vFBD4GQnfzJsvMC9lj825wob2nKy5
|
||||
CL4bY6tiYUqALzDYaIVCBqJ1Ky1rpPzRcDZUeeF8Ta3njh18Y0KzPEj+Xu5Ksexg
|
||||
7LUy/2wFzVRtJgqvAFZo43KIOptsj/63zxJjAAkBAbQhUGhpbGlwIEhhemVsIDxw
|
||||
aDEwQGN1cy5jYW0uYWMudWs+iQEyBBMBAgAcBQI9s/YNAhsDBAsHAwIDFQIDAxYC
|
||||
AQIeAQIXgAAKCRCXZuCE+w9D2JbrCACBHrxybV+9t3TGR3+uk2QnUuct90a+udcB
|
||||
cbWgKQgX345ZgwIfyoPKAbG84mN/+SFUN/Ie4ZHZxVsbN//l2P/Osx0CNQBfem4e
|
||||
HUWegkFGzC5DxQhN5nsJ/GR6pnxG1l/DbNhupNY73oTajWFdi7K8jYfamU/zS/Da
|
||||
VCwUKxWtGqyEqOxvbdqcjsnTMmLVfXtefx7CbKyhaClPP8Pe4FL+eqyfHJF7uywK
|
||||
VWlUNmQa4E+ZAK8tkoK9sZAc0ImWwZMumiKZDEpr2D8Ty+Gg2umTS2OMIcbY8QF1
|
||||
r6DeubFabdPIe1kn0BGXtdAurhbdJCIbcAEQS0eEeWC4A4LiDprntB1QaGlsaXAg
|
||||
SGF6ZWwgPHBoMTBAY2FtLmFjLnVrPokBMgQTAQIAHAUCPbaC8QIbAwQLBwMCAxUC
|
||||
AwMWAgECHgECF4AACgkQl2bghPsPQ9ho5Qf+I2HUhZeXOUneeqSa+cITsIELJMrZ
|
||||
UWcydY1z+L68yOqURVHB1jF4aC6QSlq0pLjozwF6KfZO5AfN9FvkRQ0DzCEXH48W
|
||||
OXqzbjzgNxRMdaP5+49Axl0UQuhupGJg66T4FiGnSVBhK8kTURPCSpLLgkCJqNcY
|
||||
t5zuNwk3e7JvleT59EVpk/kw3a5p9oeKYBln57pHwq+HdPLSCdkedQBKdsbNrazy
|
||||
qduYEXen4ogsIoTpA9lLH1Xsi9wL+soLAlWXtl/GNa1h7Jpwn41zp1hqIZe1ebIM
|
||||
dSGbtMEaGJlqgDauYJSl0lkVgaPtZCTmfAf57TsGfD0IckN2XNGLuwb6DLkBCwQ9
|
||||
s/eJAQgAuXq2I1VTDB108O0NAB6QKfA2gMHafoj3OcwEUHnJgXrHqjYOpFr0r5lC
|
||||
YjwMRMvHO7r4jgtwKjGACI/1IE2hiGXkKFOWZFkCf7Qjoi13a78myC/VRwe4uEIs
|
||||
xdz++w8WfzuC1sYw8d/rlybVzHTfTsKfmFOQamuyLCTm3Kdx/wZyGM7JMmgVn2zF
|
||||
HWasdID0n7QJFZGR3yEfqis5zq1t3w28AaXlTZG7QtFj9V0cWIhZRjjjVe1biPA8
|
||||
Btp+eFkmcat79N9hHbiEiAY3u2qmQCcn1fiBBC0Og09cY4dRyX0/bXUo4n8lHysp
|
||||
JcL/sIZ7Ahd1LOfg9h+yQfoaey78LQAGKYkBHwQYAQIACQUCPbP3iQIbDAAKCRCX
|
||||
ZuCE+w9D2P0BCACXsbwFerd4wlnt8PxxQlRGXi1t4EGBFf2zSdgQvIXJ3ntPzcDU
|
||||
+8hnIWQJhCTZ3PfVmoqB6ZOM1RWI/IcP6b4qVb6vJGI/VxRICraAczw+4bdFxFiB
|
||||
n2+SyD4MxZg8t66FcH3fHugQqsApHTmtKmahv5BXB+0dPmP/BRUmzxYUv8hdnhHl
|
||||
91S4R4BDu84DvKYyJK9jlqUl06pvqXrBSNsbWKgAJnmnDe8FaLUk+p0IXeMSavUn
|
||||
T5E26MRIioxAz/G4FZI8IdO92GHTz82O7/mQMhveAdpspDvpV3M4wnldU425GlxZ
|
||||
nmq1bbjAU4QVuIS9nkk2GrGx5+N6cYMy8JqG
|
||||
=EHsz
|
||||
-----END PGP PUBLIC KEY BLOCK-----
|
||||
@ -1,161 +0,0 @@
|
||||
From f1e9a32ee7fad2263636a51536ce0f9f13f09949 Mon Sep 17 00:00:00 2001
|
||||
From: =?UTF-8?q?Petr=20P=C3=ADsa=C5=99?= <ppisar@redhat.com>
|
||||
Date: Wed, 23 Jan 2019 10:16:20 +0100
|
||||
Subject: [PATCH] Declare POSIX regex function names as macros to PCRE
|
||||
functions
|
||||
MIME-Version: 1.0
|
||||
Content-Type: text/plain; charset=UTF-8
|
||||
Content-Transfer-Encoding: 8bit
|
||||
|
||||
POSIX regex libraries differ in regex_t size. If a program includes
|
||||
<pcreposix.h>, but is not linked to pcreposix library at run-time
|
||||
(either in effect of --as-needed or a lazy binding in dlopen)
|
||||
other implementation touches memory out of the structure and the
|
||||
program can crash.
|
||||
|
||||
That means once a program includes <pcreposix.h>, it must link to the
|
||||
pcreposix library.
|
||||
|
||||
This patch replaces the POSIX regex declaration with macros to the
|
||||
PCRE uniqely-named function. This ensures that the PCRE's regex_t
|
||||
structure is always handled by the PCRE functions.
|
||||
|
||||
This patch still preserves the POSIX regex definitions in order to
|
||||
preseve ABI with application compiled before this change. The
|
||||
definition can be removed in the future.
|
||||
|
||||
Signed-off-by: Petr Písař <ppisar@redhat.com>
|
||||
---
|
||||
pcreposix.c | 50 +++++++++++++++++++++++++++++++++++++++++++++-----
|
||||
pcreposix.h | 20 ++++++++++++++------
|
||||
2 files changed, 59 insertions(+), 11 deletions(-)
|
||||
|
||||
diff --git a/pcreposix.c b/pcreposix.c
|
||||
index a76d6bf..3f2f3ef 100644
|
||||
--- a/pcreposix.c
|
||||
+++ b/pcreposix.c
|
||||
@@ -39,7 +39,10 @@ POSSIBILITY OF SUCH DAMAGE.
|
||||
|
||||
|
||||
/* This module is a wrapper that provides a POSIX API to the underlying PCRE
|
||||
-functions. */
|
||||
+functions. The operative functions are called pcre_regcomp(), etc., with
|
||||
+wrappers that use the plain POSIX names. This makes it easier for an
|
||||
+application to be sure it gets the PCRE versions in the presence of other
|
||||
+POSIX regex libraries. */
|
||||
|
||||
|
||||
#ifdef HAVE_CONFIG_H
|
||||
@@ -204,12 +207,49 @@ static const char *const pstring[] = {
|
||||
|
||||
|
||||
/*************************************************
|
||||
-* Translate error code to string *
|
||||
+* Wrappers with traditional POSIX names *
|
||||
*************************************************/
|
||||
|
||||
+/* Keep defining them to preseve ABI with application linked to pcreposix
|
||||
+ * library before they were changed into macros. */
|
||||
+
|
||||
+#undef regerror
|
||||
PCREPOSIX_EXP_DEFN size_t PCRE_CALL_CONVENTION
|
||||
regerror(int errcode, const regex_t *preg, char *errbuf, size_t errbuf_size)
|
||||
{
|
||||
+return pcre_regerror(errcode, preg, errbuf, errbuf_size);
|
||||
+}
|
||||
+
|
||||
+#undef regfree
|
||||
+PCREPOSIX_EXP_DEFN void PCRE_CALL_CONVENTION
|
||||
+regfree(regex_t *preg)
|
||||
+{
|
||||
+pcre_regfree(preg);
|
||||
+}
|
||||
+
|
||||
+#undef regcomp
|
||||
+PCREPOSIX_EXP_DEFN int PCRE_CALL_CONVENTION
|
||||
+regcomp(regex_t *preg, const char *pattern, int cflags)
|
||||
+{
|
||||
+return pcre_regcomp(preg, pattern, cflags);
|
||||
+}
|
||||
+
|
||||
+#undef regexec
|
||||
+PCREPOSIX_EXP_DEFN int PCRE_CALL_CONVENTION
|
||||
+regexec(const regex_t *preg, const char *string, size_t nmatch,
|
||||
+ regmatch_t pmatch[], int eflags)
|
||||
+{
|
||||
+return pcre_regexec(preg, string, nmatch, pmatch, eflags);
|
||||
+}
|
||||
+
|
||||
+
|
||||
+/*************************************************
|
||||
+* Translate error code to string *
|
||||
+*************************************************/
|
||||
+
|
||||
+PCREPOSIX_EXP_DEFN size_t PCRE_CALL_CONVENTION
|
||||
+pcre_regerror(int errcode, const regex_t *preg, char *errbuf, size_t errbuf_size)
|
||||
+{
|
||||
const char *message, *addmessage;
|
||||
size_t length, addlength;
|
||||
|
||||
@@ -243,7 +283,7 @@ return length + addlength;
|
||||
*************************************************/
|
||||
|
||||
PCREPOSIX_EXP_DEFN void PCRE_CALL_CONVENTION
|
||||
-regfree(regex_t *preg)
|
||||
+pcre_regfree(regex_t *preg)
|
||||
{
|
||||
(PUBL(free))(preg->re_pcre);
|
||||
}
|
||||
@@ -266,7 +306,7 @@ Returns: 0 on success
|
||||
*/
|
||||
|
||||
PCREPOSIX_EXP_DEFN int PCRE_CALL_CONVENTION
|
||||
-regcomp(regex_t *preg, const char *pattern, int cflags)
|
||||
+pcre_regcomp(regex_t *preg, const char *pattern, int cflags)
|
||||
{
|
||||
const char *errorptr;
|
||||
int erroffset;
|
||||
@@ -320,7 +360,7 @@ be set. When this is the case, the nmatch and pmatch arguments are ignored, and
|
||||
the only result is yes/no/error. */
|
||||
|
||||
PCREPOSIX_EXP_DEFN int PCRE_CALL_CONVENTION
|
||||
-regexec(const regex_t *preg, const char *string, size_t nmatch,
|
||||
+pcre_regexec(const regex_t *preg, const char *string, size_t nmatch,
|
||||
regmatch_t pmatch[], int eflags)
|
||||
{
|
||||
int rc, so, eo;
|
||||
diff --git a/pcreposix.h b/pcreposix.h
|
||||
index c77c0b0..6f108b8 100644
|
||||
--- a/pcreposix.h
|
||||
+++ b/pcreposix.h
|
||||
@@ -131,13 +131,21 @@ file. */
|
||||
# endif
|
||||
#endif
|
||||
|
||||
-/* The functions */
|
||||
-
|
||||
-PCREPOSIX_EXP_DECL int regcomp(regex_t *, const char *, int);
|
||||
-PCREPOSIX_EXP_DECL int regexec(const regex_t *, const char *, size_t,
|
||||
+/* The functions. The actual code is in functions with pcre_xxx names for
|
||||
+uniqueness. POSIX names are provided for API compatibility with POSIX regex
|
||||
+functions. It's done this way to ensure to they are always get from the
|
||||
+PCRE library and not by accident from elsewhere. (regex_t differs in size
|
||||
+elsewhere.) */
|
||||
+
|
||||
+PCREPOSIX_EXP_DECL int pcre_regcomp(regex_t *, const char *, int);
|
||||
+#define regcomp pcre_regcomp
|
||||
+PCREPOSIX_EXP_DECL int pcre_regexec(const regex_t *, const char *, size_t,
|
||||
regmatch_t *, int);
|
||||
-PCREPOSIX_EXP_DECL size_t regerror(int, const regex_t *, char *, size_t);
|
||||
-PCREPOSIX_EXP_DECL void regfree(regex_t *);
|
||||
+#define regexec pcre_regexec
|
||||
+PCREPOSIX_EXP_DECL size_t pcre_regerror(int, const regex_t *, char *, size_t);
|
||||
+#define regerror pcre_regerror
|
||||
+PCREPOSIX_EXP_DECL void pcre_regfree(regex_t *);
|
||||
+#define regfree pcre_regfree
|
||||
|
||||
#ifdef __cplusplus
|
||||
} /* extern "C" */
|
||||
--
|
||||
2.17.2
|
||||
|
||||
@ -0,0 +1,70 @@
|
||||
From 7abc4de8303e8908eeb96714dac53ae10ff465e3 Mon Sep 17 00:00:00 2001
|
||||
From: ph10 <ph10@2f5784b3-3f2a-0410-8824-cb99058d5e15>
|
||||
Date: Sun, 2 Sep 2018 17:05:38 +0000
|
||||
Subject: [PATCH] Fix anchoring bug in conditional subexpression.
|
||||
MIME-Version: 1.0
|
||||
Content-Type: text/plain; charset=UTF-8
|
||||
Content-Transfer-Encoding: 8bit
|
||||
|
||||
git-svn-id: svn://vcs.exim.org/pcre/code/trunk@1739 2f5784b3-3f2a-0410-8824-cb99058d5e15
|
||||
Petr Písař: Ported to 8.42
|
||||
---
|
||||
pcre_compile.c | 12 ++++++++++--
|
||||
testdata/testinput2 | 3 +++
|
||||
testdata/testoutput2 | 4 ++++
|
||||
|
||||
Version 8.42 20-March-2018
|
||||
--------------------------
|
||||
diff --git a/pcre_compile.c b/pcre_compile.c
|
||||
index 3991d6c..6141fb3 100644
|
||||
--- a/pcre_compile.c
|
||||
+++ b/pcre_compile.c
|
||||
@@ -8682,10 +8682,18 @@ do {
|
||||
if (!is_anchored(scode, new_map, cd, atomcount)) return FALSE;
|
||||
}
|
||||
|
||||
- /* Positive forward assertions and conditions */
|
||||
+ /* Positive forward assertion */
|
||||
|
||||
- else if (op == OP_ASSERT || op == OP_COND)
|
||||
+ else if (op == OP_ASSERT)
|
||||
+ {
|
||||
+ if (!is_anchored(scode, bracket_map, cd, atomcount)) return FALSE;
|
||||
+ }
|
||||
+
|
||||
+ /* Condition; not anchored if no second branch */
|
||||
+
|
||||
+ else if (op == OP_COND)
|
||||
{
|
||||
+ if (scode[GET(scode,1)] != OP_ALT) return FALSE;
|
||||
if (!is_anchored(scode, bracket_map, cd, atomcount)) return FALSE;
|
||||
}
|
||||
|
||||
diff --git a/testdata/testinput2 b/testdata/testinput2
|
||||
index 8ba4dc4..3528de1 100644
|
||||
--- a/testdata/testinput2
|
||||
+++ b/testdata/testinput2
|
||||
@@ -4257,4 +4257,7 @@ backtracking verbs. --/
|
||||
ab
|
||||
aaab
|
||||
|
||||
+/(?(?=^))b/
|
||||
+ abc
|
||||
+
|
||||
/-- End of testinput2 --/
|
||||
diff --git a/testdata/testoutput2 b/testdata/testoutput2
|
||||
index 61ed8d9..4ccda27 100644
|
||||
--- a/testdata/testoutput2
|
||||
+++ b/testdata/testoutput2
|
||||
@@ -14721,4 +14721,8 @@ No need char
|
||||
0: ab
|
||||
1: a
|
||||
|
||||
+/(?(?=^))b/
|
||||
+ abc
|
||||
+ 0: b
|
||||
+
|
||||
/-- End of testinput2 --/
|
||||
--
|
||||
2.14.4
|
||||
|
||||
@ -0,0 +1,29 @@
|
||||
From 1aa76cb33f04fcea3127a0859450e5d18369e5e2 Mon Sep 17 00:00:00 2001
|
||||
From: zherczeg <zherczeg@2f5784b3-3f2a-0410-8824-cb99058d5e15>
|
||||
Date: Fri, 21 Sep 2018 07:34:10 +0000
|
||||
Subject: [PATCH] Fix subject buffer overread in JIT.
|
||||
MIME-Version: 1.0
|
||||
Content-Type: text/plain; charset=UTF-8
|
||||
Content-Transfer-Encoding: 8bit
|
||||
|
||||
git-svn-id: svn://vcs.exim.org/pcre/code/trunk@1740 2f5784b3-3f2a-0410-8824-cb99058d5e15
|
||||
Petr Písař: Ported to 8.42.
|
||||
---
|
||||
pcre_jit_compile.c | 2 +-
|
||||
|
||||
diff --git a/pcre_jit_compile.c b/pcre_jit_compile.c
|
||||
index 2bad74b..bc5f9c0 100644
|
||||
--- a/pcre_jit_compile.c
|
||||
+++ b/pcre_jit_compile.c
|
||||
@@ -9002,7 +9002,7 @@ if (exact > 1)
|
||||
#ifdef SUPPORT_UTF
|
||||
&& !common->utf
|
||||
#endif
|
||||
- )
|
||||
+ && type != OP_ANYNL && type != OP_EXTUNI)
|
||||
{
|
||||
OP2(SLJIT_ADD, TMP1, 0, STR_PTR, 0, SLJIT_IMM, IN_UCHARS(exact));
|
||||
add_jump(compiler, &backtrack->topbacktracks, CMP(SLJIT_GREATER, TMP1, 0, STR_END, 0));
|
||||
--
|
||||
2.17.2
|
||||
|
||||
@ -0,0 +1,178 @@
|
||||
From 2ede5a4b4a98add3bbf982f5805e015e8c61c565 Mon Sep 17 00:00:00 2001
|
||||
From: ph10 <ph10@2f5784b3-3f2a-0410-8824-cb99058d5e15>
|
||||
Date: Tue, 26 Jun 2018 16:51:43 +0000
|
||||
Subject: [PATCH] Fix two C++ wrapper bugs, unnoticed for years.
|
||||
MIME-Version: 1.0
|
||||
Content-Type: text/plain; charset=UTF-8
|
||||
Content-Transfer-Encoding: 8bit
|
||||
|
||||
git-svn-id: svn://vcs.exim.org/pcre/code/trunk@1735 2f5784b3-3f2a-0410-8824-cb99058d5e15
|
||||
|
||||
Petr Písař: Ported to 8.42.
|
||||
|
||||
diff --git a/pcrecpp.cc b/pcrecpp.cc
|
||||
index d09c9ab..77a2fed 100644
|
||||
--- a/pcrecpp.cc
|
||||
+++ b/pcrecpp.cc
|
||||
@@ -80,6 +80,24 @@ static const string empty_string;
|
||||
// If the user doesn't ask for any options, we just use this one
|
||||
static RE_Options default_options;
|
||||
|
||||
+// Specials for the start of patterns. See comments where start_options is used
|
||||
+// below. (PH June 2018)
|
||||
+static const char *start_options[] = {
|
||||
+ "(*UTF8)",
|
||||
+ "(*UTF)",
|
||||
+ "(*UCP)",
|
||||
+ "(*NO_START_OPT)",
|
||||
+ "(*NO_AUTO_POSSESS)",
|
||||
+ "(*LIMIT_RECURSION=",
|
||||
+ "(*LIMIT_MATCH=",
|
||||
+ "(*CRLF)",
|
||||
+ "(*CR)",
|
||||
+ "(*BSR_UNICODE)",
|
||||
+ "(*BSR_ANYCRLF)",
|
||||
+ "(*ANYCRLF)",
|
||||
+ "(*ANY)",
|
||||
+ "" };
|
||||
+
|
||||
void RE::Init(const string& pat, const RE_Options* options) {
|
||||
pattern_ = pat;
|
||||
if (options == NULL) {
|
||||
@@ -135,7 +153,49 @@ pcre* RE::Compile(Anchor anchor) {
|
||||
} else {
|
||||
// Tack a '\z' at the end of RE. Parenthesize it first so that
|
||||
// the '\z' applies to all top-level alternatives in the regexp.
|
||||
- string wrapped = "(?:"; // A non-counting grouping operator
|
||||
+
|
||||
+ /* When this code was written (for PCRE 6.0) it was enough just to
|
||||
+ parenthesize the entire pattern. Unfortunately, when the feature of
|
||||
+ starting patterns with (*UTF8) or (*CR) etc. was added to PCRE patterns,
|
||||
+ this code was never updated. This bug was not noticed till 2018, long after
|
||||
+ PCRE became obsolescent and its maintainer no longer around. Since PCRE is
|
||||
+ frozen, I have added a hack to check for all the existing "start of
|
||||
+ pattern" specials - knowing that no new ones will ever be added. I am not a
|
||||
+ C++ programmer, so the code style is no doubt crude. It is also
|
||||
+ inefficient, but is only run when the pattern starts with "(*".
|
||||
+ PH June 2018. */
|
||||
+
|
||||
+ string wrapped = "";
|
||||
+
|
||||
+ if (pattern_.c_str()[0] == '(' && pattern_.c_str()[1] == '*') {
|
||||
+ int kk, klen, kmat;
|
||||
+ for (;;) { // Loop for any number of leading items
|
||||
+
|
||||
+ for (kk = 0; start_options[kk][0] != 0; kk++) {
|
||||
+ klen = strlen(start_options[kk]);
|
||||
+ kmat = strncmp(pattern_.c_str(), start_options[kk], klen);
|
||||
+ if (kmat >= 0) break;
|
||||
+ }
|
||||
+ if (kmat != 0) break; // Not found
|
||||
+
|
||||
+ // If the item ended in "=" we must copy digits up to ")".
|
||||
+
|
||||
+ if (start_options[kk][klen-1] == '=') {
|
||||
+ while (isdigit(pattern_.c_str()[klen])) klen++;
|
||||
+ if (pattern_.c_str()[klen] != ')') break; // Syntax error
|
||||
+ klen++;
|
||||
+ }
|
||||
+
|
||||
+ // Move the item from the pattern to the start of the wrapped string.
|
||||
+
|
||||
+ wrapped += pattern_.substr(0, klen);
|
||||
+ pattern_.erase(0, klen);
|
||||
+ }
|
||||
+ }
|
||||
+
|
||||
+ // Wrap the rest of the pattern.
|
||||
+
|
||||
+ wrapped += "(?:"; // A non-counting grouping operator
|
||||
wrapped += pattern_;
|
||||
wrapped += ")\\z";
|
||||
re = pcre_compile(wrapped.c_str(), pcre_options,
|
||||
@@ -415,7 +475,7 @@ int RE::GlobalReplace(const StringPiece& rewrite,
|
||||
matchend++;
|
||||
}
|
||||
// We also need to advance more than one char if we're in utf8 mode.
|
||||
-#ifdef SUPPORT_UTF8
|
||||
+#ifdef SUPPORT_UTF
|
||||
if (options_.utf8()) {
|
||||
while (matchend < static_cast<int>(str->length()) &&
|
||||
((*str)[matchend] & 0xc0) == 0x80)
|
||||
diff --git a/pcrecpp_unittest.cc b/pcrecpp_unittest.cc
|
||||
index 4b15fbe..255066f 100644
|
||||
--- a/pcrecpp_unittest.cc
|
||||
+++ b/pcrecpp_unittest.cc
|
||||
@@ -309,7 +309,7 @@ static void TestReplace() {
|
||||
"@aa",
|
||||
"@@@",
|
||||
3 },
|
||||
-#ifdef SUPPORT_UTF8
|
||||
+#ifdef SUPPORT_UTF
|
||||
{ "b*",
|
||||
"bb",
|
||||
"\xE3\x83\x9B\xE3\x83\xBC\xE3\x83\xA0\xE3\x81\xB8", // utf8
|
||||
@@ -327,7 +327,7 @@ static void TestReplace() {
|
||||
{ "", NULL, NULL, NULL, NULL, 0 }
|
||||
};
|
||||
|
||||
-#ifdef SUPPORT_UTF8
|
||||
+#ifdef SUPPORT_UTF
|
||||
const bool support_utf8 = true;
|
||||
#else
|
||||
const bool support_utf8 = false;
|
||||
@@ -535,7 +535,7 @@ static void TestQuoteMetaLatin1() {
|
||||
}
|
||||
|
||||
static void TestQuoteMetaUtf8() {
|
||||
-#ifdef SUPPORT_UTF8
|
||||
+#ifdef SUPPORT_UTF
|
||||
TestQuoteMeta("Pl\xc3\xa1\x63ido Domingo", pcrecpp::UTF8());
|
||||
TestQuoteMeta("xyz", pcrecpp::UTF8()); // No fancy utf8
|
||||
TestQuoteMeta("\xc2\xb0", pcrecpp::UTF8()); // 2-byte utf8 (degree symbol)
|
||||
@@ -1178,7 +1178,7 @@ int main(int argc, char** argv) {
|
||||
CHECK(re.error().empty()); // Must have no error
|
||||
}
|
||||
|
||||
-#ifdef SUPPORT_UTF8
|
||||
+#ifdef SUPPORT_UTF
|
||||
// Check UTF-8 handling
|
||||
{
|
||||
printf("Testing UTF-8 handling\n");
|
||||
@@ -1202,6 +1202,24 @@ int main(int argc, char** argv) {
|
||||
CHECK(re_test1.FullMatch(utf8_string));
|
||||
RE re_test2("...", pcrecpp::UTF8());
|
||||
CHECK(re_test2.FullMatch(utf8_string));
|
||||
+
|
||||
+ // PH added these tests for leading option settings
|
||||
+
|
||||
+ RE re_testZ1("(*UTF8)...");
|
||||
+ CHECK(re_testZ1.FullMatch(utf8_string));
|
||||
+
|
||||
+ RE re_testZ2("(*UTF)...");
|
||||
+ CHECK(re_testZ2.FullMatch(utf8_string));
|
||||
+
|
||||
+ RE re_testZ3("(*UCP)(*UTF)...");
|
||||
+ CHECK(re_testZ3.FullMatch(utf8_string));
|
||||
+
|
||||
+ RE re_testZ4("(*UCP)(*LIMIT_MATCH=1000)(*UTF)...");
|
||||
+ CHECK(re_testZ4.FullMatch(utf8_string));
|
||||
+
|
||||
+ RE re_testZ5("(*UCP)(*LIMIT_MATCH=1000)(*ANY)(*UTF)...");
|
||||
+ CHECK(re_testZ5.FullMatch(utf8_string));
|
||||
+
|
||||
|
||||
// Check that '.' matches one byte or UTF-8 character
|
||||
// according to the mode.
|
||||
@@ -1248,7 +1266,7 @@ int main(int argc, char** argv) {
|
||||
CHECK(!match_sentence.FullMatch(target));
|
||||
CHECK(!match_sentence_re.FullMatch(target));
|
||||
}
|
||||
-#endif /* def SUPPORT_UTF8 */
|
||||
+#endif /* def SUPPORT_UTF */
|
||||
|
||||
printf("Testing error reporting\n");
|
||||
|
||||
--
|
||||
2.14.4
|
||||
|
||||
@ -0,0 +1,36 @@
|
||||
From 6d8fa98d5167bdd915ce1e32bcb6aaed078e5938 Mon Sep 17 00:00:00 2001
|
||||
From: ph10 <ph10@2f5784b3-3f2a-0410-8824-cb99058d5e15>
|
||||
Date: Tue, 26 Jun 2018 17:04:02 +0000
|
||||
Subject: [PATCH] Fix typos in pcrgrep.
|
||||
MIME-Version: 1.0
|
||||
Content-Type: text/plain; charset=UTF-8
|
||||
Content-Transfer-Encoding: 8bit
|
||||
|
||||
git-svn-id: svn://vcs.exim.org/pcre/code/trunk@1736 2f5784b3-3f2a-0410-8824-cb99058d5e15
|
||||
Petr Písař: Ported to 8.42.
|
||||
|
||||
diff --git a/pcregrep.c b/pcregrep.c
|
||||
index a406be9..5982406 100644
|
||||
--- a/pcregrep.c
|
||||
+++ b/pcregrep.c
|
||||
@@ -2252,7 +2252,7 @@ if (isdirectory(pathname))
|
||||
int fnlength = strlen(pathname) + strlen(nextfile) + 2;
|
||||
if (fnlength > 2048)
|
||||
{
|
||||
- fprintf(stderr, "pcre2grep: recursive filename is too long\n");
|
||||
+ fprintf(stderr, "pcregrep: recursive filename is too long\n");
|
||||
rc = 2;
|
||||
break;
|
||||
}
|
||||
@@ -3034,7 +3034,7 @@ LC_ALL environment variable is set, and if so, use it. */
|
||||
if (locale == NULL)
|
||||
{
|
||||
locale = getenv("LC_ALL");
|
||||
- locale_from = "LCC_ALL";
|
||||
+ locale_from = "LC_ALL";
|
||||
}
|
||||
|
||||
if (locale == NULL)
|
||||
--
|
||||
2.14.4
|
||||
|
||||
@ -0,0 +1,55 @@
|
||||
From 3a9026509f9c1745f378595e55e5024361ad152d Mon Sep 17 00:00:00 2001
|
||||
From: ph10 <ph10@2f5784b3-3f2a-0410-8824-cb99058d5e15>
|
||||
Date: Mon, 10 Feb 2020 17:17:34 +0000
|
||||
Subject: [PATCH] Check the size of the number after (?C as it is read, in
|
||||
order to avoid integer overflow.
|
||||
MIME-Version: 1.0
|
||||
Content-Type: text/plain; charset=UTF-8
|
||||
Content-Transfer-Encoding: 8bit
|
||||
|
||||
git-svn-id: svn://vcs.exim.org/pcre/code/trunk@1761 2f5784b3-3f2a-0410-8824-cb99058d5e15
|
||||
Petr Písař: Ported to 8.43.
|
||||
---
|
||||
pcre_compile.c | 14 ++++++++------
|
||||
|
||||
diff --git a/pcre_compile.c b/pcre_compile.c
|
||||
index 079d30a..1e3d6c3 100644
|
||||
--- a/pcre_compile.c
|
||||
+++ b/pcre_compile.c
|
||||
@@ -6,7 +6,7 @@
|
||||
and semantics are as close as possible to those of the Perl 5 language.
|
||||
|
||||
Written by Philip Hazel
|
||||
- Copyright (c) 1997-2018 University of Cambridge
|
||||
+ Copyright (c) 1997-2020 University of Cambridge
|
||||
|
||||
-----------------------------------------------------------------------------
|
||||
Redistribution and use in source and binary forms, with or without
|
||||
@@ -7130,17 +7130,19 @@ for (;; ptr++)
|
||||
int n = 0;
|
||||
ptr++;
|
||||
while(IS_DIGIT(*ptr))
|
||||
+ {
|
||||
n = n * 10 + *ptr++ - CHAR_0;
|
||||
+ if (n > 255)
|
||||
+ {
|
||||
+ *errorcodeptr = ERR38;
|
||||
+ goto FAILED;
|
||||
+ }
|
||||
+ }
|
||||
if (*ptr != CHAR_RIGHT_PARENTHESIS)
|
||||
{
|
||||
*errorcodeptr = ERR39;
|
||||
goto FAILED;
|
||||
}
|
||||
- if (n > 255)
|
||||
- {
|
||||
- *errorcodeptr = ERR38;
|
||||
- goto FAILED;
|
||||
- }
|
||||
*code++ = n;
|
||||
PUT(code, 0, (int)(ptr - cd->start_pattern + 1)); /* Pattern offset */
|
||||
PUT(code, LINK_SIZE, 0); /* Default length */
|
||||
--
|
||||
2.21.1
|
||||
|
||||
@ -1,44 +0,0 @@
|
||||
From f0bb9e8baf3157e0a84f484f194984295b2db23a Mon Sep 17 00:00:00 2001
|
||||
From: =?UTF-8?q?Petr=20P=C3=ADsa=C5=99?= <ppisar@redhat.com>
|
||||
Date: Mon, 19 Oct 2020 16:15:14 +0200
|
||||
Subject: [PATCH] Inicialize name table memory region
|
||||
MIME-Version: 1.0
|
||||
Content-Type: text/plain; charset=UTF-8
|
||||
Content-Transfer-Encoding: 8bit
|
||||
|
||||
Name table entry values are accessed past their ends in add_name()
|
||||
when comparing the values. Also a size of the entries could grow
|
||||
later. It's safer to initialize just after the allocation than to hunt
|
||||
the gaps later.
|
||||
|
||||
Reproducer:
|
||||
|
||||
pcre_compile2("(?<f>)(?<fir>)", PCRE_NO_AUTO_CAPTURE | PCRE_CASELESS, &ec, &eb, &eo, NULL);
|
||||
|
||||
built with clang++ -fsanitize=memory -fsanitize=fuzzer-no-link.
|
||||
|
||||
https://bugs.exim.org/show_bug.cgi?id=2661
|
||||
Signed-off-by: Petr Písař <ppisar@redhat.com>
|
||||
---
|
||||
pcre_compile.c | 5 +++++
|
||||
1 file changed, 5 insertions(+)
|
||||
|
||||
diff --git a/pcre_compile.c b/pcre_compile.c
|
||||
index 3be0fbf..75309e0 100644
|
||||
--- a/pcre_compile.c
|
||||
+++ b/pcre_compile.c
|
||||
@@ -9423,6 +9423,11 @@ if (re == NULL)
|
||||
goto PCRE_EARLY_ERROR_RETURN;
|
||||
}
|
||||
|
||||
+/* Initialize the memory. Name table entry values are accessed past their ends
|
||||
+ * (e.g. in add_name()) when comparing the values. Also a size of the entry can
|
||||
+ * grow later. It's safer to initialize here than to hunt the gaps later. */
|
||||
+memset(re, 0, size);
|
||||
+
|
||||
/* Put in the magic number, and save the sizes, initial options, internal
|
||||
flags, and character table pointer. NULL is used for the default character
|
||||
tables. The nullpad field is at the end; it's there to help in the case when a
|
||||
--
|
||||
2.25.4
|
||||
|
||||
File diff suppressed because it is too large
Load Diff
@ -1,98 +0,0 @@
|
||||
From 70fcff23652339438cabef86f2d3d9c645678687 Mon Sep 17 00:00:00 2001
|
||||
From: "H.J. Lu" <hjl.tools@gmail.com>
|
||||
Date: Sun, 20 Dec 2020 13:55:40 -0800
|
||||
Subject: [PATCH 2/2] Pass -mshstk to the compiler when Intel CET is enable
|
||||
|
||||
Copied from PCRE2.
|
||||
---
|
||||
CMakeLists.txt | 18 ++++++++++++++++++
|
||||
Makefile.am | 3 +++
|
||||
configure.ac | 15 +++++++++++++++
|
||||
3 files changed, 36 insertions(+)
|
||||
|
||||
diff --git a/CMakeLists.txt b/CMakeLists.txt
|
||||
index 858a34b..74c27e9 100644
|
||||
--- a/CMakeLists.txt
|
||||
+++ b/CMakeLists.txt
|
||||
@@ -112,6 +112,24 @@ CHECK_FUNCTION_EXISTS(_strtoi64 HAVE__STRTOI64)
|
||||
CHECK_TYPE_SIZE("long long" LONG_LONG)
|
||||
CHECK_TYPE_SIZE("unsigned long long" UNSIGNED_LONG_LONG)
|
||||
|
||||
+# Check whether Intel CET is enabled, and if so, adjust compiler flags. This
|
||||
+# code was written by PH, trying to imitate the logic from the autotools
|
||||
+# configuration.
|
||||
+
|
||||
+CHECK_C_SOURCE_COMPILES(
|
||||
+ "#ifndef __CET__
|
||||
+ #error CET is not enabled
|
||||
+ #endif
|
||||
+ int main() { return 0; }"
|
||||
+ INTEL_CET_ENABLED
|
||||
+)
|
||||
+
|
||||
+IF (INTEL_CET_ENABLED)
|
||||
+ SET(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -mshstk")
|
||||
+ENDIF(INTEL_CET_ENABLED)
|
||||
+
|
||||
+
|
||||
+
|
||||
# User-configurable options
|
||||
#
|
||||
# (Note: CMakeSetup displays these in alphabetical order, regardless of
|
||||
diff --git a/Makefile.am b/Makefile.am
|
||||
index 22b6947..984c686 100644
|
||||
--- a/Makefile.am
|
||||
+++ b/Makefile.am
|
||||
@@ -250,6 +250,7 @@ libpcre_la_SOURCES = \
|
||||
|
||||
libpcre_la_CFLAGS = \
|
||||
$(VISIBILITY_CFLAGS) \
|
||||
+ $(CET_CFLAGS) \
|
||||
$(AM_CFLAGS)
|
||||
|
||||
libpcre_la_LIBADD =
|
||||
@@ -289,6 +290,7 @@ libpcre16_la_SOURCES = \
|
||||
|
||||
libpcre16_la_CFLAGS = \
|
||||
$(VISIBILITY_CFLAGS) \
|
||||
+ $(CET_CFLAGS) \
|
||||
$(AM_CFLAGS)
|
||||
|
||||
libpcre16_la_LIBADD =
|
||||
@@ -328,6 +330,7 @@ libpcre32_la_SOURCES = \
|
||||
|
||||
libpcre32_la_CFLAGS = \
|
||||
$(VISIBILITY_CFLAGS) \
|
||||
+ $(CET_CFLAGS) \
|
||||
$(AM_CFLAGS)
|
||||
|
||||
libpcre32_la_LIBADD =
|
||||
diff --git a/configure.ac b/configure.ac
|
||||
index 6a3e5f3..15f37d1 100644
|
||||
--- a/configure.ac
|
||||
+++ b/configure.ac
|
||||
@@ -1043,6 +1043,21 @@ fi # enable_coverage
|
||||
|
||||
AM_CONDITIONAL([WITH_GCOV],[test "x$enable_coverage" = "xyes"])
|
||||
|
||||
+AC_MSG_CHECKING([whether Intel CET is enabled])
|
||||
+AC_LANG_PUSH([C])
|
||||
+AC_COMPILE_IFELSE([AC_LANG_PROGRAM(,
|
||||
+ [[#ifndef __CET__
|
||||
+# error CET is not enabled
|
||||
+#endif]])],
|
||||
+ [intel_cet_enabled=yes],
|
||||
+ [intel_cet_enabled=no])
|
||||
+AC_MSG_RESULT([$intel_cet_enabled])
|
||||
+if test "$intel_cet_enabled" = yes; then
|
||||
+ CET_CFLAGS="-mshstk"
|
||||
+ AC_SUBST([CET_CFLAGS])
|
||||
+fi
|
||||
+AC_LANG_POP([C])
|
||||
+
|
||||
# Produce these files, in addition to config.h.
|
||||
AC_CONFIG_FILES(
|
||||
Makefile
|
||||
--
|
||||
2.29.2
|
||||
|
||||
Binary file not shown.
Loading…
Reference in new issue