rpms/pcre2
6
0
Fork 0
Code Issues Pull Requests Releases Activity

10.31-RC1 bump

Browse Source
This commit is contained in:
Petr Písař 2018-01-15 12:04:06 +01:00
parent 1c9da09ce5
commit fd6a3698c1
16 changed files with 23 additions and 1792 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
.gitignore vendored
View File

@ -9,3 +9,4 @@
/pcre2-10.23.tar.bz2
/pcre2-10.30-RC1.tar.bz2
/pcre2-10.30.tar.bz2
/pcre2-10.31-RC1.tar.bz2

137
pcre2-10.30-Change-pcre2grep-line-number-and-count-variables-to-.patch
View File

@ -1,137 +0,0 @@
From 43e13d367d3f442ace4ce1cf43d9c2b27e2226bb Mon Sep 17 00:00:00 2001
From: ph10 <ph10@6239d852-aaf2-0410-a92c-79f79f948069>
Date: Fri, 8 Dec 2017 10:25:49 +0000
Subject: [PATCH] Change pcre2grep line number and count variables to unsigned
long int.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
git-svn-id: svn://vcs.exim.org/pcre2/code/trunk@887 6239d852-aaf2-0410-a92c-79f79f948069
Petr Písař: Port to 10.30.
diff --git a/src/pcre2grep.c b/src/pcre2grep.c
index c7c95c5..e764313 100644
--- a/src/pcre2grep.c
+++ b/src/pcre2grep.c
@@ -196,8 +196,9 @@ static int bufthird = PCRE2GREP_BUFSIZE;
static int max_bufthird = PCRE2GREP_MAX_BUFSIZE;
static int bufsize = 3*PCRE2GREP_BUFSIZE;
static int endlinetype;
-static int total_count = 0;
-static int counts_printed = 0;
+
+static unsigned long int total_count = 0;
+static unsigned long int counts_printed = 0;
#ifdef WIN32
static int dee_action = dee_SKIP;
@@ -1606,8 +1607,8 @@ Returns: nothing
*/
static void
-do_after_lines(int lastmatchnumber, char *lastmatchrestart, char *endptr,
- const char *printname)
+do_after_lines(unsigned long int lastmatchnumber, char *lastmatchrestart,
+ char *endptr, const char *printname)
{
if (after_context > 0 && lastmatchnumber > 0)
{
@@ -1618,7 +1619,7 @@ if (after_context > 0 && lastmatchnumber > 0)
char *pp = end_of_line(lastmatchrestart, endptr, &ellength);
if (ellength == 0 && pp == main_buffer + bufsize) break;
if (printname != NULL) fprintf(stdout, "%s-", printname);
- if (number) fprintf(stdout, "%d-", lastmatchnumber++);
+ if (number) fprintf(stdout, "%lu-", lastmatchnumber++);
FWRITE_IGNORE(lastmatchrestart, 1, pp - lastmatchrestart, stdout);
lastmatchrestart = pp;
count++;
@@ -2309,10 +2310,10 @@ static int
pcre2grep(void *handle, int frtype, const char *filename, const char *printname)
{
int rc = 1;
-int linenumber = 1;
-int lastmatchnumber = 0;
-int count = 0;
int filepos = 0;
+unsigned long int linenumber = 1;
+unsigned long int lastmatchnumber = 0;
+unsigned long int count = 0;
char *lastmatchrestart = NULL;
char *ptr = main_buffer;
char *endptr;
@@ -2400,7 +2401,7 @@ while (ptr < endptr)
if (new_buffer == NULL)
{
fprintf(stderr,
- "pcre2grep: line %d%s%s is too long for the internal buffer\n"
+ "pcre2grep: line %lu%s%s is too long for the internal buffer\n"
"pcre2grep: not enough memory to increase the buffer size to %d\n",
linenumber,
(filename == NULL)? "" : " of file ",
@@ -2430,7 +2431,7 @@ while (ptr < endptr)
else
{
fprintf(stderr,
- "pcre2grep: line %d%s%s is too long for the internal buffer\n"
+ "pcre2grep: line %lu%s%s is too long for the internal buffer\n"
"pcre2grep: the maximum buffer size is %d\n"
"pcre2grep: use the --max-buffer-size option to change it\n",
linenumber,
@@ -2562,7 +2563,7 @@ while (ptr < endptr)
size_t oldstartoffset;
if (printname != NULL) fprintf(stdout, "%s:", printname);
- if (number) fprintf(stdout, "%d:", linenumber);
+ if (number) fprintf(stdout, "%lu:", linenumber);
/* Handle --line-offsets */
@@ -2682,7 +2683,7 @@ while (ptr < endptr)
{
char *pp = lastmatchrestart;
if (printname != NULL) fprintf(stdout, "%s-", printname);
- if (number) fprintf(stdout, "%d-", lastmatchnumber++);
+ if (number) fprintf(stdout, "%lu-", lastmatchnumber++);
pp = end_of_line(pp, endptr, &ellength);
FWRITE_IGNORE(lastmatchrestart, 1, pp - lastmatchrestart, stdout);
lastmatchrestart = pp;
@@ -2722,7 +2723,7 @@ while (ptr < endptr)
int ellength;
char *pp = p;
if (printname != NULL) fprintf(stdout, "%s-", printname);
- if (number) fprintf(stdout, "%d-", linenumber - linecount--);
+ if (number) fprintf(stdout, "%lu-", linenumber - linecount--);
pp = end_of_line(pp, endptr, &ellength);
FWRITE_IGNORE(p, 1, pp - p, stdout);
p = pp;
@@ -2736,7 +2737,7 @@ while (ptr < endptr)
endhyphenpending = TRUE;
if (printname != NULL) fprintf(stdout, "%s:", printname);
- if (number) fprintf(stdout, "%d:", linenumber);
+ if (number) fprintf(stdout, "%lu:", linenumber);
/* This extra option, for Jeffrey Friedl's debugging requirements,
replaces the matched string, or a specific captured string if it exists,
@@ -2918,7 +2919,7 @@ if (count_only && !quiet)
{
if (printname != NULL && filenames != FN_NONE)
fprintf(stdout, "%s:", printname);
- fprintf(stdout, "%d" STDOUT_NL, count);
+ fprintf(stdout, "%lu" STDOUT_NL, count);
counts_printed++;
}
}
@@ -4196,7 +4197,7 @@ if (show_total_count && counts_printed != 1 && filenames != FN_NOMATCH_ONLY)
{
if (counts_printed != 0 && filenames >= FN_DEFAULT)
fprintf(stdout, "TOTAL:");
- fprintf(stdout, "%d" STDOUT_NL, total_count);
+ fprintf(stdout, "%lu" STDOUT_NL, total_count);
}
EXIT:
--
2.13.6

174
pcre2-10.30-Documentation-update.patch
View File

@ -1,174 +0,0 @@
From c26d49eda45dd8a26e1de65a4430e84116266227 Mon Sep 17 00:00:00 2001
From: ph10 <ph10@6239d852-aaf2-0410-a92c-79f79f948069>
Date: Sun, 24 Dec 2017 10:27:13 +0000
Subject: [PATCH 2/3] Documentation update.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
git-svn-id: svn://vcs.exim.org/pcre2/code/trunk@895 6239d852-aaf2-0410-a92c-79f79f948069
Signed-off-by: Petr Písař <ppisar@redhat.com>
---
doc/html/pcre2demo.html | 53 +++++++++++++++++++++++++++++++++++++++++++++++++
doc/pcre2demo.3 | 53 +++++++++++++++++++++++++++++++++++++++++++++++++
2 files changed, 106 insertions(+)
diff --git a/doc/html/pcre2demo.html b/doc/html/pcre2demo.html
index d64e16b..72754d3 100644
--- a/doc/html/pcre2demo.html
+++ b/doc/html/pcre2demo.html
@@ -228,6 +228,21 @@ pcre2_match_data_create_from_pattern() above. */
if (rc == 0)
printf("ovector was not big enough for all the captured substrings\n");
+/* We must guard against patterns such as /(?=.\K)/ that use \K in an assertion
+to set the start of a match later than its end. In this demonstration program,
+we just detect this case and give up. */
+
+if (ovector[0] &gt; ovector[1])
+ {
+ printf("\\K was used in an assertion to set the match start after its end.\n"
+ "From end to start the match was: %.*s\n", (int)(ovector[0] - ovector[1]),
+ (char *)(subject + ovector[1]));
+ printf("Run abandoned\n");
+ pcre2_match_data_free(match_data);
+ pcre2_code_free(re);
+ return 1;
+ }
+
/* Show substrings stored in the output vector by number. Obviously, in a real
application you might want to do things other than print them. */
@@ -355,6 +370,29 @@ for (;;)
options = PCRE2_NOTEMPTY_ATSTART | PCRE2_ANCHORED;
}
+ /* If the previous match was not an empty string, there is one tricky case to
+ consider. If a pattern contains \K within a lookbehind assertion at the
+ start, the end of the matched string can be at the offset where the match
+ started. Without special action, this leads to a loop that keeps on matching
+ the same substring. We must detect this case and arrange to move the start on
+ by one character. The pcre2_get_startchar() function returns the starting
+ offset that was passed to pcre2_match(). */
+
+ else
+ {
+ PCRE2_SIZE startchar = pcre2_get_startchar(match_data);
+ if (start_offset &lt;= startchar)
+ {
+ if (startchar &gt;= subject_length) break; /* Reached end of subject. */
+ start_offset = startchar + 1; /* Advance by one character. */
+ if (utf8) /* If UTF-8, it may be more */
+ { /* than one code unit. */
+ for (; start_offset &lt; subject_length; start_offset++)
+ if ((subject[start_offset] &amp; 0xc0) != 0x80) break;
+ }
+ }
+ }
+
/* Run the next matching operation */
rc = pcre2_match(
@@ -419,6 +457,21 @@ for (;;)
if (rc == 0)
printf("ovector was not big enough for all the captured substrings\n");
+ /* We must guard against patterns such as /(?=.\K)/ that use \K in an
+ assertion to set the start of a match later than its end. In this
+ demonstration program, we just detect this case and give up. */
+
+ if (ovector[0] &gt; ovector[1])
+ {
+ printf("\\K was used in an assertion to set the match start after its end.\n"
+ "From end to start the match was: %.*s\n", (int)(ovector[0] - ovector[1]),
+ (char *)(subject + ovector[1]));
+ printf("Run abandoned\n");
+ pcre2_match_data_free(match_data);
+ pcre2_code_free(re);
+ return 1;
+ }
+
/* As before, show substrings stored in the output vector by number, and then
also any named substrings. */
diff --git a/doc/pcre2demo.3 b/doc/pcre2demo.3
index c02dcd9..a9e58e2 100644
--- a/doc/pcre2demo.3
+++ b/doc/pcre2demo.3
@@ -228,6 +228,21 @@ pcre2_match_data_create_from_pattern() above. */
if (rc == 0)
printf("ovector was not big enough for all the captured substrings\en");
+/* We must guard against patterns such as /(?=.\eK)/ that use \eK in an assertion
+to set the start of a match later than its end. In this demonstration program,
+we just detect this case and give up. */
+
+if (ovector[0] > ovector[1])
+ {
+ printf("\e\eK was used in an assertion to set the match start after its end.\en"
+ "From end to start the match was: %.*s\en", (int)(ovector[0] - ovector[1]),
+ (char *)(subject + ovector[1]));
+ printf("Run abandoned\en");
+ pcre2_match_data_free(match_data);
+ pcre2_code_free(re);
+ return 1;
+ }
+
/* Show substrings stored in the output vector by number. Obviously, in a real
application you might want to do things other than print them. */
@@ -355,6 +370,29 @@ for (;;)
options = PCRE2_NOTEMPTY_ATSTART | PCRE2_ANCHORED;
}
+ /* If the previous match was not an empty string, there is one tricky case to
+ consider. If a pattern contains \eK within a lookbehind assertion at the
+ start, the end of the matched string can be at the offset where the match
+ started. Without special action, this leads to a loop that keeps on matching
+ the same substring. We must detect this case and arrange to move the start on
+ by one character. The pcre2_get_startchar() function returns the starting
+ offset that was passed to pcre2_match(). */
+
+ else
+ {
+ PCRE2_SIZE startchar = pcre2_get_startchar(match_data);
+ if (start_offset <= startchar)
+ {
+ if (startchar >= subject_length) break; /* Reached end of subject. */
+ start_offset = startchar + 1; /* Advance by one character. */
+ if (utf8) /* If UTF-8, it may be more */
+ { /* than one code unit. */
+ for (; start_offset < subject_length; start_offset++)
+ if ((subject[start_offset] & 0xc0) != 0x80) break;
+ }
+ }
+ }
+
/* Run the next matching operation */
rc = pcre2_match(
@@ -419,6 +457,21 @@ for (;;)
if (rc == 0)
printf("ovector was not big enough for all the captured substrings\en");
+ /* We must guard against patterns such as /(?=.\eK)/ that use \eK in an
+ assertion to set the start of a match later than its end. In this
+ demonstration program, we just detect this case and give up. */
+
+ if (ovector[0] > ovector[1])
+ {
+ printf("\e\eK was used in an assertion to set the match start after its end.\en"
+ "From end to start the match was: %.*s\en", (int)(ovector[0] - ovector[1]),
+ (char *)(subject + ovector[1]));
+ printf("Run abandoned\en");
+ pcre2_match_data_free(match_data);
+ pcre2_code_free(re);
+ return 1;
+ }
+
/* As before, show substrings stored in the output vector by number, and then
also any named substrings. */
--
2.13.6

61
pcre2-10.30-FIRSTLINE_documentation-update.patch
View File

@ -1,61 +0,0 @@
From 3fdced6eef96f50ac5bd287426db0aa699be3edc Mon Sep 17 00:00:00 2001
From: ph10 <ph10@6239d852-aaf2-0410-a92c-79f79f948069>
Date: Sun, 31 Dec 2017 17:44:12 +0000
Subject: [PATCH] Documentation update.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
git-svn-id: svn://vcs.exim.org/pcre2/code/trunk@898 6239d852-aaf2-0410-a92c-79f79f948069
Petr Písař: Ported to 10.30.
Signed-off-by: Petr Písař <ppisar@redhat.com>
---
doc/pcre2api.3 | 24 +++++++++++++++---------
1 file changed, 15 insertions(+), 9 deletions(-)
diff --git a/doc/pcre2api.3 b/doc/pcre2api.3
index f80ae58..d55debf 100644
--- a/doc/pcre2api.3
+++ b/doc/pcre2api.3
@@ -871,10 +871,11 @@ compiled. If a match is started with a non-default match limit when
PCRE2_USE_OFFSET_LIMIT is not set, an error is generated.
.P
The offset limit facility can be used to track progress when searching large
-subject strings. See also the PCRE2_FIRSTLINE option, which requires a match to
-start within the first line of the subject. If this 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.
+subject strings. See also the PCRE2_FIRSTLINE option, which requires a match
+to start before or at the first newline that follows the start of matching in
+the subject. If this 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.
.sp
.nf
.B int pcre2_set_heap_limit(pcre2_match_context *\fImcontext\fP,
@@ -1423,11 +1424,16 @@ changed within a pattern by a (?xx) option setting.
PCRE2_FIRSTLINE
.sp
If this option is set, the start of an unanchored pattern match must be before
-or at the first newline in the subject string, though the matched text may
-continue over the newline. 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.
+or at the first newline in the subject string following the start of matching,
+though the matched text may continue over the newline. If \fIstartoffset\fP is
+non-zero, the limiting newline is not necessarily the first newline in the
+subject. For example, if the subject string is "abc\enxyz" (where \en
+represents a single-character newline) a pattern match for "yz" succeeds with
+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.
.sp
PCRE2_LITERAL
.sp
--
2.13.6

149
pcre2-10.30-Fix-K-issues-in-pcre2grep.patch
View File

@ -1,149 +0,0 @@
From a109c9e35a040a7a8032c12ce7396bc949f3f735 Mon Sep 17 00:00:00 2001
From: ph10 <ph10@6239d852-aaf2-0410-a92c-79f79f948069>
Date: Tue, 26 Dec 2017 15:10:04 +0000
Subject: [PATCH 3/3] Fix \K issues in pcre2grep.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
git-svn-id: svn://vcs.exim.org/pcre2/code/trunk@896 6239d852-aaf2-0410-a92c-79f79f948069
Petr Písař: Ported to 10.30.
---
RunGrepTest | 11 +++++++++++
src/pcre2grep.c | 49 ++++++++++++++++++++++++++++++++++++++++++++-----
testdata/grepoutput | 9 +++++++++
diff --git a/RunGrepTest b/RunGrepTest
index a7496cb..a26f677 100755
--- a/RunGrepTest
+++ b/RunGrepTest
@@ -630,6 +630,17 @@ echo "RC=$?" >>testtrygrep
(cd $srcdir; $valgrind $vjs $pcre2grep -Mn -A2 'start[\s]+end' testdata/grepinputM) >>testtrygrep
echo "RC=$?" >>testtrygrep
+echo "---------------------------- Test 125 -----------------------------" >>testtrygrep
+printf "abcd\n" >testNinputgrep
+$valgrind $vjs $pcre2grep --colour=always '(?<=\K.)' testNinputgrep >>testtrygrep
+echo "RC=$?" >>testtrygrep
+$valgrind $vjs $pcre2grep --colour=always '(?=.\K)' testNinputgrep >>testtrygrep
+echo "RC=$?" >>testtrygrep
+$valgrind $vjs $pcre2grep --colour=always '(?<=\K[ac])' testNinputgrep >>testtrygrep
+echo "RC=$?" >>testtrygrep
+$valgrind $vjs $pcre2grep --colour=always '(?=[ac]\K)' testNinputgrep >>testtrygrep
+echo "RC=$?" >>testtrygrep
+
# Now compare the results.
diff --git a/src/pcre2grep.c b/src/pcre2grep.c
index e764313..02339f5 100644
--- a/src/pcre2grep.c
+++ b/src/pcre2grep.c
@@ -1607,7 +1607,7 @@ Returns: nothing
*/
static void
-do_after_lines(unsigned long int lastmatchnumber, char *lastmatchrestart,
+do_after_lines(unsigned long int lastmatchnumber, char *lastmatchrestart,
char *endptr, const char *printname)
{
if (after_context > 0 && lastmatchnumber > 0)
@@ -2764,11 +2764,38 @@ while (ptr < endptr)
if ((multiline || do_colour) && !invert)
{
int plength;
+ PCRE2_SIZE endprevious;
+
+ /* The use of \K may make the end offset earlier than the start. In
+ this situation, swap them round. */
+
+ if (offsets[0] > offsets[1])
+ {
+ PCRE2_SIZE temp = offsets[0];
+ offsets[0] = offsets[1];
+ offsets[1] = temp;
+ }
+
FWRITE_IGNORE(ptr, 1, offsets[0], stdout);
print_match(ptr + offsets[0], offsets[1] - offsets[0]);
+
for (;;)
{
- startoffset = offsets[1]; /* Advance after previous match. */
+ PCRE2_SIZE oldstartoffset = pcre2_get_startchar(match_data);
+
+ endprevious = offsets[1];
+ startoffset = endprevious; /* Advance after previous match. */
+
+ /* If the pattern contained a lookbehind that included \K, it is
+ possible that the end of the match might be at or before the actual
+ starting offset we have just used. In this case, start one character
+ further on. */
+
+ if (startoffset <= oldstartoffset)
+ {
+ startoffset = oldstartoffset + 1;
+ if (utf) while ((ptr[startoffset] & 0xc0) == 0x80) startoffset++;
+ }
/* If the current match ended past the end of the line (only possible
in multiline mode), we must move on to the line in which it did end
@@ -2782,6 +2809,7 @@ while (ptr < endptr)
filepos += (int)(linelength + endlinelength);
linenumber++;
startoffset -= (int)(linelength + endlinelength);
+ endprevious -= (int)(linelength + endlinelength);
t = end_of_line(ptr, endptr, &endlinelength);
linelength = t - ptr - endlinelength;
length = (size_t)(endptr - ptr);
@@ -2797,7 +2825,18 @@ while (ptr < endptr)
loop for any that may follow. */
if (!match_patterns(ptr, length, options, startoffset, &mrc)) break;
- FWRITE_IGNORE(ptr + startoffset, 1, offsets[0] - startoffset, stdout);
+
+ /* The use of \K may make the end offset earlier than the start. In
+ this situation, swap them round. */
+
+ if (offsets[0] > offsets[1])
+ {
+ PCRE2_SIZE temp = offsets[0];
+ offsets[0] = offsets[1];
+ offsets[1] = temp;
+ }
+
+ FWRITE_IGNORE(ptr + endprevious, 1, offsets[0] - endprevious, stdout);
print_match(ptr + offsets[0], offsets[1] - offsets[0]);
}
@@ -2805,8 +2844,8 @@ while (ptr < endptr)
and its line-ending characters (if they matched the pattern), so there
may be no more to print. */
- plength = (int)((linelength + endlinelength) - startoffset);
- if (plength > 0) FWRITE_IGNORE(ptr + startoffset, 1, plength, stdout);
+ plength = (int)((linelength + endlinelength) - endprevious);
+ if (plength > 0) FWRITE_IGNORE(ptr + endprevious, 1, plength, stdout);
}
/* Not colouring or multiline; no need to search for further matches. */
diff --git a/testdata/grepoutput b/testdata/grepoutput
index 7e963fb..e49c2b2 100644
--- a/testdata/grepoutput
+++ b/testdata/grepoutput
@@ -936,3 +936,12 @@ end
16:start end in between start
end
RC=0
+---------------------------- Test 125 -----------------------------
+abcd
+RC=0
+abcd
+RC=0
+abcd
+RC=0
+abcd
+RC=0
--
2.13.6

206
pcre2-10.30-Fix-PCRE2_FIRSTLINE-bug-when-a-pattern-match-starts-.patch
View File

@ -1,206 +0,0 @@
From f3b22988611cca57770a705f05c0d9ef583d605a Mon Sep 17 00:00:00 2001
From: ph10 <ph10@6239d852-aaf2-0410-a92c-79f79f948069>
Date: Mon, 1 Jan 2018 14:12:35 +0000
Subject: [PATCH] Fix PCRE2_FIRSTLINE bug when a pattern match starts with the
first code unit of a newline sequence.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
git-svn-id: svn://vcs.exim.org/pcre2/code/trunk@899 6239d852-aaf2-0410-a92c-79f79f948069
Petr Písař: Ported to 10.32
Signed-off-by: Petr Písař <ppisar@redhat.com>
---
src/pcre2_dfa_match.c | 21 +++++++++++++++------
src/pcre2_match.c | 19 ++++++++++++++-----
testdata/testinput2 | 10 ++++++++++
testdata/testinput6 | 10 ++++++++++
testdata/testoutput2 | 13 +++++++++++++
testdata/testoutput6 | 13 +++++++++++++
6 files changed, 75 insertions(+), 11 deletions(-)
diff --git a/src/pcre2_dfa_match.c b/src/pcre2_dfa_match.c
index 5ae1394..7bbd6d3 100644
--- a/src/pcre2_dfa_match.c
+++ b/src/pcre2_dfa_match.c
@@ -7,7 +7,7 @@ and semantics are as close as possible to those of the Perl 5 language.
Written by Philip Hazel
Original API code Copyright (c) 1997-2012 University of Cambridge
- New API code Copyright (c) 2016-2017 University of Cambridge
+ New API code Copyright (c) 2016-2018 University of Cambridge
-----------------------------------------------------------------------------
Redistribution and use in source and binary forms, with or without
@@ -3558,9 +3558,11 @@ for (;;)
/* If firstline is TRUE, the start of the match is constrained to the first
line of a multiline string. That is, the match must be before or at the
- first newline. Implement this by temporarily adjusting end_subject so that
- we stop the optimization scans for a first code unit at a newline. If the
- match fails at the newline, later code breaks this loop. */
+ first newline following the start of matching. Temporarily adjust
+ end_subject so that we stop the optimization scans for a first code unit
+ immediately after the first character of a newline (the first code unit can
+ legitimately be a newline). If the match fails at the newline, later code
+ breaks this loop. */
if (firstline)
{
@@ -3568,7 +3570,7 @@ for (;;)
#ifdef SUPPORT_UNICODE
if (utf)
{
- while (t < mb->end_subject && !IS_NEWLINE(t))
+ while (t < end_subject && !IS_NEWLINE(t))
{
t++;
ACROSSCHAR(t < end_subject, *t, t++);
@@ -3576,7 +3578,14 @@ for (;;)
}
else
#endif
- while (t < mb->end_subject && !IS_NEWLINE(t)) t++;
+ while (t < end_subject && !IS_NEWLINE(t)) t++;
+
+ /* Note that we only need to advance by one code unit if we found a
+ newline. If the newline is CRLF, a first code unit of LF should not
+ match, because it is not at or before the newline. Similarly, only the
+ first code unit of a Unicode newline might be relevant. */
+
+ if (t < end_subject) t++;
end_subject = t;
}
diff --git a/src/pcre2_match.c b/src/pcre2_match.c
index 050b7e9..8afb0d8 100644
--- a/src/pcre2_match.c
+++ b/src/pcre2_match.c
@@ -6541,9 +6541,11 @@ for(;;)
/* If firstline is TRUE, the start of the match is constrained to the first
line of a multiline string. That is, the match must be before or at the
- first newline. Implement this by temporarily adjusting end_subject so that
- we stop the optimization scans for a first code unit at a newline. If the
- match fails at the newline, later code breaks this loop. */
+ first newline following the start of matching. Temporarily adjust
+ end_subject so that we stop the optimization scans for a first code unit
+ immediately after the first character of a newline (the first code unit can
+ legitimately be a newline). If the match fails at the newline, later code
+ breaks this loop. */
if (firstline)
{
@@ -6551,7 +6553,7 @@ for(;;)
#ifdef SUPPORT_UNICODE
if (utf)
{
- while (t < mb->end_subject && !IS_NEWLINE(t))
+ while (t < end_subject && !IS_NEWLINE(t))
{
t++;
ACROSSCHAR(t < end_subject, *t, t++);
@@ -6559,7 +6561,14 @@ for(;;)
}
else
#endif
- while (t < mb->end_subject && !IS_NEWLINE(t)) t++;
+ while (t < end_subject && !IS_NEWLINE(t)) t++;
+
+ /* Note that we only need to advance by one code unit if we found a
+ newline. If the newline is CRLF, a first code unit of LF should not
+ match, because it is not at or before the newline. Similarly, only the
+ first code unit of a Unicode newline might be relevant. */
+
+ if (t < end_subject) t++;
end_subject = t;
}
diff --git a/testdata/testinput2 b/testdata/testinput2
index 695f0a4..b173fe0 100644
--- a/testdata/testinput2
+++ b/testdata/testinput2
@@ -5385,4 +5385,14 @@ a)"xI
ab
aaab
+/\n/firstline
+ xyz\nabc
+
+/\nabc/firstline
+ xyz\nabc
+
+/\x{0a}abc/firstline,newline=crlf
+\= Expect no match
+ xyz\r\nabc
+
# End of testinput2
diff --git a/testdata/testinput6 b/testdata/testinput6
index ce2e082..614c3a0 100644
--- a/testdata/testinput6
+++ b/testdata/testinput6
@@ -4932,4 +4932,14 @@
/(*LIMIT_MATCH=100).*(?![|H]?.*(?![|H]?););.*(?![|H]?.*(?![|H]?););\x00\x00\x00\x00\x00\x00\x00(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?![|);)?.*(![|H]?);)?.*(?![|H]?);)?.*(?![|H]?);)?.*(?![|H]););![|H]?););[|H]?);|H]?);)\x00\x00\x00 \x00\x00\x00H]?););?![|H]?);)?.*(?![|H]?););[||H]?);)?.*(?![|H]?););[|H]?);(?![|H]?););![|H]?););[|H]?);|H]?);)?.*(?![|H]?););;[\x00\x00\x00\x00\x00\x00\x00![|H]?););![|H]?););[|H]?);|H]?);)?.*(?![|H]?););/no_dotstar_anchor
.*(?![|H]?.*(?![|H]?););.*(?![|H]?.*(?![|H]?););\x00\x00\x00\x00\x00\x00\x00(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?![|);)?.*(![|H]?);)?.*(?![|H]?);)?.*(?![|H]?);)?.*(?![|H]););![|H]?););[|H]?);|H]?);)\x00\x00\x00 \x00\x00\x00H]?););?![|H]?);)?.*(?![|H]?););[||H]?);)?.*(?![|H]?););[|H]?);(?![|H]?););![|H]?););[|H]?);|H]?);)?.*(?![|H]?););;[\x00\x00\x00\x00\x00\x00\x00![|H]?););![|H]?););[|H]?);|H]?);)?.*(?![|H]?););
+/\n/firstline
+ xyz\nabc
+
+/\nabc/firstline
+ xyz\nabc
+
+/\x{0a}abc/firstline,newline=crlf
+\= Expect no match
+ xyz\r\nabc
+
# End of testinput6
diff --git a/testdata/testoutput2 b/testdata/testoutput2
index 31ccfbe..c19c270 100644
--- a/testdata/testoutput2
+++ b/testdata/testoutput2
@@ -16368,6 +16368,19 @@ Subject length lower bound = 1
0: ab
1: a
+/\n/firstline
+ xyz\nabc
+ 0: \x0a
+
+/\nabc/firstline
+ xyz\nabc
+ 0: \x0aabc
+
+/\x{0a}abc/firstline,newline=crlf
+\= Expect no match
+ xyz\r\nabc
+No match
+
# End of testinput2
Error -65: PCRE2_ERROR_BADDATA (unknown error number)
Error -62: bad serialized data
diff --git a/testdata/testoutput6 b/testdata/testoutput6
index b912944..2d321d5 100644
--- a/testdata/testoutput6
+++ b/testdata/testoutput6
@@ -7753,4 +7753,17 @@ No match
.*(?![|H]?.*(?![|H]?););.*(?![|H]?.*(?![|H]?););\x00\x00\x00\x00\x00\x00\x00(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?!(?![|);)?.*(![|H]?);)?.*(?![|H]?);)?.*(?![|H]?);)?.*(?![|H]););![|H]?););[|H]?);|H]?);)\x00\x00\x00 \x00\x00\x00H]?););?![|H]?);)?.*(?![|H]?););[||H]?);)?.*(?![|H]?););[|H]?);(?![|H]?););![|H]?););[|H]?);|H]?);)?.*(?![|H]?););;[\x00\x00\x00\x00\x00\x00\x00![|H]?););![|H]?););[|H]?);|H]?);)?.*(?![|H]?););
Failed: error -47: match limit exceeded
+/\n/firstline
+ xyz\nabc
+ 0: \x0a
+
+/\nabc/firstline
+ xyz\nabc
+ 0: \x0aabc
+
+/\x{0a}abc/firstline,newline=crlf
+\= Expect no match
+ xyz\r\nabc
+No match
+
# End of testinput6
--
2.13.6

91
pcre2-10.30-Fix-incorrect-first-matching-character-when-a-backre.patch
View File

@ -1,91 +0,0 @@
From b5343d4a647d25640e16bfa1f813c39a7f2059a6 Mon Sep 17 00:00:00 2001
From: ph10 <ph10@6239d852-aaf2-0410-a92c-79f79f948069>
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).
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
git-svn-id: svn://vcs.exim.org/pcre2/code/trunk@888 6239d852-aaf2-0410-a92c-79f79f948069
Petr Písař: Ported to 10.30.
---
src/pcre2_compile.c | 2 +-
testdata/testinput2 | 10 ++++++++++
testdata/testoutput2 | 28 ++++++++++++++++++++++++++++
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
--
2.13.6

32
pcre2-10.30-Fix-memory-leak-issue-introduced-in-last-bug-fix-in-.patch
View File

@ -1,32 +0,0 @@
From 12dd78fc2ff2d941e724c7af72fa22daa8528041 Mon Sep 17 00:00:00 2001
From: ph10 <ph10@6239d852-aaf2-0410-a92c-79f79f948069>
Date: Sun, 22 Oct 2017 16:17:44 +0000
Subject: [PATCH 2/2] Fix memory leak issue introduced in last bug fix in
pcre2grep.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
git-svn-id: svn://vcs.exim.org/pcre2/code/trunk@874 6239d852-aaf2-0410-a92c-79f79f948069
Signed-off-by: Petr Písař <ppisar@redhat.com>
---
src/pcre2grep.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/src/pcre2grep.c b/src/pcre2grep.c
index d7410b4..d75917c 100644
--- a/src/pcre2grep.c
+++ b/src/pcre2grep.c
@@ -3051,7 +3051,8 @@ if (isdirectory(pathname))
if (fnlength > FNBUFSIZ)
{
fprintf(stderr, "pcre2grep: recursive filename is too long\n");
- return 2;
+ rc = 2;
+ break;
}
sprintf(buffer, "%s%c%s", pathname, FILESEP, nextfile);
frc = grep_or_recurse(buffer, dir_recurse, FALSE);
--
2.13.6

383
pcre2-10.30-Fix-multiple-multiline-matching-issues-in-pcre2grep.patch
View File

@ -1,383 +0,0 @@
From eff8c9e117259b1192919b85c2ee03a27b164f1a Mon Sep 17 00:00:00 2001
From: ph10 <ph10@6239d852-aaf2-0410-a92c-79f79f948069>
Date: Mon, 13 Nov 2017 16:52:39 +0000
Subject: [PATCH] Fix multiple multiline matching issues in pcre2grep.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
git-svn-id: svn://vcs.exim.org/pcre2/code/trunk@878 6239d852-aaf2-0410-a92c-79f79f948069
Petr Písař: Ported to 10.30.
diff --git a/RunGrepTest b/RunGrepTest
index cf88c78..a7496cb 100755
--- a/RunGrepTest
+++ b/RunGrepTest
@@ -248,7 +248,7 @@ echo "---------------------------- Test 35 -----------------------------" >>test
echo "RC=$?" >>testtrygrep
echo "---------------------------- Test 36 -----------------------------" >>testtrygrep
-(cd $srcdir; $valgrind $vjs $pcre2grep -L -r --include=grepinput --exclude 'grepinput$' --exclude=grepinput8 --exclude-dir='^\.' 'fox' ./testdata | sort) >>testtrygrep
+(cd $srcdir; $valgrind $vjs $pcre2grep -L -r --include=grepinput --exclude 'grepinput$' --exclude=grepinput8 --exclude=grepinputM --exclude-dir='^\.' 'fox' ./testdata | sort) >>testtrygrep
echo "RC=$?" >>testtrygrep
echo "---------------------------- Test 37 -----------------------------" >>testtrygrep
@@ -391,6 +391,12 @@ echo "RC=$?" >>testtrygrep
echo "---------------------------- Test 70 -----------------------------" >>testtrygrep
(cd $srcdir; $valgrind $vjs $pcre2grep --color=always -M "triple:\t.*\n\n" ./testdata/grepinput3) >>testtrygrep
echo "RC=$?" >>testtrygrep
+(cd $srcdir; $valgrind $vjs $pcre2grep --color=always -M -n "triple:\t.*\n\n" ./testdata/grepinput3) >>testtrygrep
+echo "RC=$?" >>testtrygrep
+(cd $srcdir; $valgrind $vjs $pcre2grep -M "triple:\t.*\n\n" ./testdata/grepinput3) >>testtrygrep
+echo "RC=$?" >>testtrygrep
+(cd $srcdir; $valgrind $vjs $pcre2grep -M -n "triple:\t.*\n\n" ./testdata/grepinput3) >>testtrygrep
+echo "RC=$?" >>testtrygrep
echo "---------------------------- Test 71 -----------------------------" >>testtrygrep
(cd $srcdir; $valgrind $vjs $pcre2grep -o "^01|^02|^03" ./testdata/grepinput) >>testtrygrep
@@ -494,25 +500,25 @@ echo "---------------------------- Test 95 -----------------------------" >>test
echo "RC=$?" >>testtrygrep
echo "---------------------------- Test 96 -----------------------------" >>testtrygrep
-(cd $srcdir; $valgrind $vjs $pcre2grep -L -r --include-dir=testdata --exclude '^(?!grepinput)' 'fox' ./test* | sort) >>testtrygrep
+(cd $srcdir; $valgrind $vjs $pcre2grep -L -r --include-dir=testdata --exclude '^(?!grepinput)' --exclude=grepinputM 'fox' ./test* | sort) >>testtrygrep
echo "RC=$?" >>testtrygrep
echo "---------------------------- Test 97 -----------------------------" >>testtrygrep
echo "grepinput$" >testtemp1grep
echo "grepinput8" >>testtemp1grep
-(cd $srcdir; $valgrind $vjs $pcre2grep -L -r --include=grepinput --exclude-from $builddir/testtemp1grep --exclude-dir='^\.' 'fox' ./testdata | sort) >>testtrygrep
+(cd $srcdir; $valgrind $vjs $pcre2grep -L -r --include=grepinput --exclude=grepinputM --exclude-from $builddir/testtemp1grep --exclude-dir='^\.' 'fox' ./testdata | sort) >>testtrygrep
echo "RC=$?" >>testtrygrep
echo "---------------------------- Test 98 -----------------------------" >>testtrygrep
echo "grepinput$" >testtemp1grep
echo "grepinput8" >>testtemp1grep
-(cd $srcdir; $valgrind $vjs $pcre2grep -L -r --exclude=grepinput3 --include=grepinput --exclude-from $builddir/testtemp1grep --exclude-dir='^\.' 'fox' ./testdata | sort) >>testtrygrep
+(cd $srcdir; $valgrind $vjs $pcre2grep -L -r --exclude=grepinput3 --exclude=grepinputM --include=grepinput --exclude-from $builddir/testtemp1grep --exclude-dir='^\.' 'fox' ./testdata | sort) >>testtrygrep
echo "RC=$?" >>testtrygrep
echo "---------------------------- Test 99 -----------------------------" >>testtrygrep
echo "grepinput$" >testtemp1grep
echo "grepinput8" >testtemp2grep
-(cd $srcdir; $valgrind $vjs $pcre2grep -L -r --include grepinput --exclude-from $builddir/testtemp1grep --exclude-from=$builddir/testtemp2grep --exclude-dir='^\.' 'fox' ./testdata | sort) >>testtrygrep
+(cd $srcdir; $valgrind $vjs $pcre2grep -L -r --include grepinput --exclude=grepinputM --exclude-from $builddir/testtemp1grep --exclude-from=$builddir/testtemp2grep --exclude-dir='^\.' 'fox' ./testdata | sort) >>testtrygrep
echo "RC=$?" >>testtrygrep
echo "---------------------------- Test 100 ------------------------------" >>testtrygrep
@@ -582,7 +588,7 @@ echo "---------------------------- Test 115 -----------------------------" >>tes
echo "RC=$?" >>testtrygrep
echo "---------------------------- Test 116 -----------------------------" >>testtrygrep
-(cd $srcdir; $valgrind $vjs $pcre2grep -th 'the' testdata/grepinput*) >>testtrygrep
+(cd $srcdir; $valgrind $vjs $pcre2grep --exclude=grepinputM -th 'the' testdata/grepinput*) >>testtrygrep
echo "RC=$?" >>testtrygrep
echo "---------------------------- Test 117 -----------------------------" >>testtrygrep
@@ -610,10 +616,20 @@ echo "---------------------------- Test 122 -----------------------------" >>tes
(cd $srcdir; $valgrind $vjs $pcre2grep -w 'cat|dog' testdata/grepinputv) >>testtrygrep
echo "RC=$?" >>testtrygrep
-echo "---------------------------- Test 122 -----------------------------" >>testtrygrep
+echo "---------------------------- Test 123 -----------------------------" >>testtrygrep
(cd $srcdir; $valgrind $vjs $pcre2grep -w 'dog|cat' testdata/grepinputv) >>testtrygrep
echo "RC=$?" >>testtrygrep
+echo "---------------------------- Test 124 -----------------------------" >>testtrygrep
+(cd $srcdir; $valgrind $vjs $pcre2grep -Mn --colour=always 'start[\s]+end' testdata/grepinputM) >>testtrygrep
+echo "RC=$?" >>testtrygrep
+(cd $srcdir; $valgrind $vjs $pcre2grep -Mn --colour=always -A2 'start[\s]+end' testdata/grepinputM) >>testtrygrep
+echo "RC=$?" >>testtrygrep
+(cd $srcdir; $valgrind $vjs $pcre2grep -Mn 'start[\s]+end' testdata/grepinputM) >>testtrygrep
+echo "RC=$?" >>testtrygrep
+(cd $srcdir; $valgrind $vjs $pcre2grep -Mn -A2 'start[\s]+end' testdata/grepinputM) >>testtrygrep
+echo "RC=$?" >>testtrygrep
+
# Now compare the results.
diff --git a/src/pcre2grep.c b/src/pcre2grep.c
index 1649d5a..bec07e1 100644
--- a/src/pcre2grep.c
+++ b/src/pcre2grep.c
@@ -2505,7 +2505,10 @@ while (ptr < endptr)
match = match_patterns(ptr, length, options, startoffset, &mrc);
options = PCRE2_NOTEMPTY;
- /* If it's a match or a not-match (as required), do what's wanted. */
+ /* If it's a match or a not-match (as required), do what's wanted. NOTE: Use
+ only FWRITE_IGNORE() - which is just a packaged fwrite() that ignores its
+ return code - to output data lines, so that binary zeroes are treated as just
+ another data character. */
if (match != invert)
{
@@ -2734,27 +2737,6 @@ while (ptr < endptr)
if (printname != NULL) fprintf(stdout, "%s:", printname);
if (number) fprintf(stdout, "%d:", linenumber);
- /* In multiline mode, we want to print to the end of the line in which
- the end of the matched string is found, so we adjust linelength and the
- line number appropriately, but only when there actually was a match
- (invert not set). Because the PCRE2_FIRSTLINE option is set, the start of
- the match will always be before the first newline sequence. */
-
- if (multiline & !invert)
- {
- char *endmatch = ptr + offsets[1];
- t = ptr;
- while (t <= endmatch)
- {
- t = end_of_line(t, endptr, &endlinelength);
- if (t < endmatch) linenumber++; else break;
- }
- linelength = t - ptr - endlinelength;
- }
-
- /*** NOTE: Use only fwrite() to output the data line, so that binary
- zeroes are treated as just another data character. */
-
/* This extra option, for Jeffrey Friedl's debugging requirements,
replaces the matched string, or a specific captured string if it exists,
with X. When this happens, colouring is ignored. */
@@ -2771,20 +2753,48 @@ while (ptr < endptr)
else
#endif
- /* We have to split the line(s) up if colouring, and search for further
- matches, but not of course if the line is a non-match. */
+ /* In multiline mode, or if colouring, we have to split the line(s) up
+ and search for further matches, but not of course if the line is a
+ non-match. In multiline mode this is necessary in case there is another
+ match that spans the end of the current line. When colouring we want to
+ colour all matches. */
- if (do_colour && !invert)
+ if ((multiline || do_colour) && !invert)
{
int plength;
FWRITE_IGNORE(ptr, 1, offsets[0], stdout);
print_match(ptr + offsets[0], offsets[1] - offsets[0]);
for (;;)
{
- startoffset = offsets[1];
- if (startoffset >= linelength + endlinelength ||
- !match_patterns(ptr, length, options, startoffset, &mrc))
- break;
+ startoffset = offsets[1]; /* Advance after previous match. */
+
+ /* If the current match ended past the end of the line (only possible
+ in multiline mode), we must move on to the line in which it did end
+ before searching for more matches. Because the PCRE2_FIRSTLINE option
+ is set, the start of the match will always be before the first
+ newline sequence. */
+
+ while (startoffset > linelength + endlinelength)
+ {
+ ptr += linelength + endlinelength;
+ filepos += (int)(linelength + endlinelength);
+ linenumber++;
+ startoffset -= (int)(linelength + endlinelength);
+ t = end_of_line(ptr, endptr, &endlinelength);
+ linelength = t - ptr - endlinelength;
+ length = (size_t)(endptr - ptr);
+ }
+
+ /* If startoffset is at the exact end of the line it means this
+ complete line was the final part of the match, so there is nothing
+ more to do. */
+
+ if (startoffset == linelength + endlinelength) break;
+
+ /* Otherwise, run a match from within the final line, and if found,
+ loop for any that may follow. */
+
+ if (!match_patterns(ptr, length, options, startoffset, &mrc)) break;
FWRITE_IGNORE(ptr + startoffset, 1, offsets[0] - startoffset, stdout);
print_match(ptr + offsets[0], offsets[1] - offsets[0]);
}
@@ -2797,7 +2807,7 @@ while (ptr < endptr)
if (plength > 0) FWRITE_IGNORE(ptr + startoffset, 1, plength, stdout);
}
- /* Not colouring; no need to search for further matches */
+ /* Not colouring or multiline; no need to search for further matches. */
else FWRITE_IGNORE(ptr, 1, linelength + endlinelength, stdout);
}
diff --git a/testdata/grepinputM b/testdata/grepinputM
new file mode 100644
index 0000000..9119e3d
--- /dev/null
+++ b/testdata/grepinputM
@@ -0,0 +1,17 @@
+Data file for multiline tests of multiple matches.
+
+start end in between start
+end and following
+Other stuff
+
+start end in between start
+end and following start
+end other stuff
+
+start end in between start
+
+end
+
+** These two lines must be last.
+start end in between start
+end
diff --git a/testdata/grepoutput b/testdata/grepoutput
index 52e0d17..7e963fb 100644
--- a/testdata/grepoutput
+++ b/testdata/grepoutput
@@ -487,6 +487,7 @@ RC=0
./testdata/grepinput:456
./testdata/grepinput3:0
./testdata/grepinput8:0
+./testdata/grepinputM:0
./testdata/grepinputv:1
./testdata/grepinputx:0
RC=0
@@ -600,6 +601,33 @@ RC=0
triple: t6_txt s2_tag s_txt p_tag p_txt o_tag o_txt
RC=0
+1:triple: t1_txt s1_tag s_txt p_tag p_txt o_tag o_txt
+
+6:triple: t3_txt s2_tag s_txt p_tag p_txt o_tag o_txt
+
+8:triple: t4_txt s1_tag s_txt p_tag p_txt o_tag o_txt
+
+13:triple: t6_txt s2_tag s_txt p_tag p_txt o_tag o_txt
+
+RC=0
+triple: t1_txt s1_tag s_txt p_tag p_txt o_tag o_txt
+
+triple: t3_txt s2_tag s_txt p_tag p_txt o_tag o_txt
+
+triple: t4_txt s1_tag s_txt p_tag p_txt o_tag o_txt
+
+triple: t6_txt s2_tag s_txt p_tag p_txt o_tag o_txt
+
+RC=0
+1:triple: t1_txt s1_tag s_txt p_tag p_txt o_tag o_txt
+
+6:triple: t3_txt s2_tag s_txt p_tag p_txt o_tag o_txt
+
+8:triple: t4_txt s1_tag s_txt p_tag p_txt o_tag o_txt
+
+13:triple: t6_txt s2_tag s_txt p_tag p_txt o_tag o_txt
+
+RC=0
---------------------------- Test 71 -----------------------------
01
RC=0
@@ -793,21 +821,23 @@ RC=0
37216,12
RC=0
---------------------------- Test 113 -----------------------------
-478
+480
RC=0
---------------------------- Test 114 -----------------------------
testdata/grepinput:469
testdata/grepinput3:0
testdata/grepinput8:0
+testdata/grepinputM:2
testdata/grepinputv:3
testdata/grepinputx:6
-TOTAL:478
+TOTAL:480
RC=0
---------------------------- Test 115 -----------------------------
testdata/grepinput:469
+testdata/grepinputM:2
testdata/grepinputv:3
testdata/grepinputx:6
-TOTAL:478
+TOTAL:480
RC=0
---------------------------- Test 116 -----------------------------
478
@@ -816,9 +846,10 @@ RC=0
469
0
0
+2
3
6
-478
+480
RC=0
---------------------------- Test 118 -----------------------------
testdata/grepinput3
@@ -846,7 +877,62 @@ RC=0
over the lazy dog.
The word is cat in this line
RC=0
----------------------------- Test 122 -----------------------------
+---------------------------- Test 123 -----------------------------
over the lazy dog.
The word is cat in this line
RC=0
+---------------------------- Test 124 -----------------------------
+3:start end in between start
+end and following
+7:start end in between start
+end and following start
+end other stuff
+11:start end in between start
+
+end
+16:start end in between start
+end
+RC=0
+3:start end in between start
+end and following
+5-Other stuff
+6-
+7:start end in between start
+end and following start
+end other stuff
+10-
+11:start end in between start
+
+end
+14-
+15-** These two lines must be last.
+16:start end in between start
+end
+RC=0
+3:start end in between start
+end and following
+7:start end in between start
+end and following start
+end other stuff
+11:start end in between start
+
+end
+16:start end in between start
+end
+RC=0
+3:start end in between start
+end and following
+5-Other stuff
+6-
+7:start end in between start
+end and following start
+end other stuff
+10-
+11:start end in between start
+
+end
+14-
+15-** These two lines must be last.
+16:start end in between start
+end
+RC=0
--
2.13.6

28
pcre2-10.30-Fix-pcre2_jit_match-early-check.patch
View File

@ -1,28 +0,0 @@
From 67aa24cc00470222301cb5fe3c53bc319105d092 Mon Sep 17 00:00:00 2001
From: zherczeg <zherczeg@6239d852-aaf2-0410-a92c-79f79f948069>
Date: Thu, 23 Nov 2017 07:54:39 +0000
Subject: [PATCH] Fix pcre2_jit_match early check.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
git-svn-id: svn://vcs.exim.org/pcre2/code/trunk@884 6239d852-aaf2-0410-a92c-79f79f948069
Petr Písař: Ported to 10.30.
diff --git a/src/pcre2_jit_match.c b/src/pcre2_jit_match.c
index 4cad754..2f840ab 100644
--- a/src/pcre2_jit_match.c
+++ b/src/pcre2_jit_match.c
@@ -118,7 +118,7 @@ if ((options & PCRE2_PARTIAL_HARD) != 0)
else if ((options & PCRE2_PARTIAL_SOFT) != 0)
index = 1;
-if (functions->executable_funcs[index] == NULL)
+if (functions->executable_funcs == NULL || functions->executable_funcs[index] == NULL)
return PCRE2_ERROR_JIT_BADOPTION;
/* Sanity checks should be handled by pcre_exec. */
--
2.13.6

52
pcre2-10.30-Fix-pcre2grep-recursive-file-name-length-issue.patch
View File

@ -1,52 +0,0 @@
From 386fd5ed13bd694a9940365602be6fcefd52d295 Mon Sep 17 00:00:00 2001
From: ph10 <ph10@6239d852-aaf2-0410-a92c-79f79f948069>
Date: Fri, 20 Oct 2017 16:51:59 +0000
Subject: [PATCH 1/2] Fix pcre2grep recursive file name length issue.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
git-svn-id: svn://vcs.exim.org/pcre2/code/trunk@873 6239d852-aaf2-0410-a92c-79f79f948069
Petr Písař: Ported to 10.30.
diff --git a/src/pcre2grep.c b/src/pcre2grep.c
index bd86998..d7410b4 100644
--- a/src/pcre2grep.c
+++ b/src/pcre2grep.c
@@ -109,7 +109,7 @@ typedef int BOOL;
#define MAXPATLEN 8192
#endif
-#define FNBUFSIZ 1024
+#define FNBUFSIZ 2048
#define ERRBUFSIZ 256
/* Values for the "filenames" variable, which specifies options for file name
@@ -3032,7 +3032,7 @@ if (isdirectory(pathname))
if (dee_action == dee_RECURSE)
{
- char buffer[1024];
+ char buffer[FNBUFSIZ];
char *nextfile;
directory_type *dir = opendirectory(pathname);
@@ -3047,7 +3047,13 @@ if (isdirectory(pathname))
while ((nextfile = readdirectory(dir)) != NULL)
{
int frc;
- sprintf(buffer, "%.512s%c%.128s", pathname, FILESEP, nextfile);
+ int fnlength = strlen(pathname) + strlen(nextfile) + 2;
+ if (fnlength > FNBUFSIZ)
+ {
+ fprintf(stderr, "pcre2grep: recursive filename is too long\n");
+ return 2;
+ }
+ sprintf(buffer, "%s%c%s", pathname, FILESEP, nextfile);
frc = grep_or_recurse(buffer, dir_recurse, FALSE);
if (frc > 1) rc = frc;
else if (frc == 0 && rc == 1) rc = 0;
--
2.13.6

218
pcre2-10.30-Previous-FIRSTLINE-patch-was-broken.-Fix-it.patch
View File

@ -1,218 +0,0 @@
From 1b5d77c6edc5ee8e8fe5c96bf9cad5798d6ce36c Mon Sep 17 00:00:00 2001
From: ph10 <ph10@6239d852-aaf2-0410-a92c-79f79f948069>
Date: Mon, 1 Jan 2018 14:54:06 +0000
Subject: [PATCH 3/3] Previous FIRSTLINE patch was broken. Fix it.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
git-svn-id: svn://vcs.exim.org/pcre2/code/trunk@900 6239d852-aaf2-0410-a92c-79f79f948069
Signed-off-by: Petr Písař <ppisar@redhat.com>
---
src/pcre2_dfa_match.c | 27 +++++++++++----------------
src/pcre2_match.c | 37 +++++++++++++++----------------------
testdata/testinput2 | 4 ++++
testdata/testinput6 | 4 ++++
testdata/testoutput2 | 5 +++++
testdata/testoutput6 | 5 +++++
6 files changed, 44 insertions(+), 38 deletions(-)
diff --git a/src/pcre2_dfa_match.c b/src/pcre2_dfa_match.c
index 9c1d805..65243bf 100644
--- a/src/pcre2_dfa_match.c
+++ b/src/pcre2_dfa_match.c
@@ -3363,8 +3363,6 @@ for (;;)
if ((re->overall_options & PCRE2_NO_START_OPTIMIZE) == 0 &&
(options & PCRE2_DFA_RESTART) == 0)
{
- PCRE2_SPTR save_end_subject = end_subject;
-
/* If firstline is TRUE, the start of the match is constrained to the first
line of a multiline string. That is, the match must be before or at the
first newline following the start of matching. Temporarily adjust
@@ -3388,13 +3386,6 @@ for (;;)
else
#endif
while (t < end_subject && !IS_NEWLINE(t)) t++;
-
- /* Note that we only need to advance by one code unit if we found a
- newline. If the newline is CRLF, a first code unit of LF should not
- match, because it is not at or before the newline. Similarly, only the
- first code unit of a Unicode newline might be relevant. */
-
- if (t < end_subject) t++;
end_subject = t;
}
@@ -3466,14 +3457,18 @@ for (;;)
#endif
}
- /* If we can't find the required code unit, break the bumpalong loop,
- to force a match failure, except when doing partial matching, when we
- let the next cycle run at the end of the subject. To see why, consider
- the pattern /(?<=abc)def/, which partially matches "abc", even though
- the string does not contain the starting character "d". */
+ /* If we can't find the required code unit, having reached the true end
+ of the subject, break the bumpalong loop, to force a match failure,
+ except when doing partial matching, when we let the next cycle run at
+ the end of the subject. To see why, consider the pattern /(?<=abc)def/,
+ which partially matches "abc", even though the string does not contain
+ the starting character "d". If we have not reached the true end of the
+ subject (PCRE2_FIRSTLINE caused end_subject to be temporarily modified)
+ we also let the cycle run, because the matching string is legitimately
+ allowed to start with the first code unit of a newline. */
if ((mb->moptions & (PCRE2_PARTIAL_HARD|PCRE2_PARTIAL_SOFT)) == 0 &&
- start_match >= end_subject)
+ start_match >= mb->end_subject)
break;
}
@@ -3532,7 +3527,7 @@ for (;;)
/* Restore fudged end_subject */
- end_subject = save_end_subject;
+ end_subject = mb->end_subject;
/* The following two optimizations are disabled for partial matching. */
diff --git a/src/pcre2_match.c b/src/pcre2_match.c
index 8872345..c6b6975 100644
--- a/src/pcre2_match.c
+++ b/src/pcre2_match.c
@@ -7,7 +7,7 @@ and semantics are as close as possible to those of the Perl 5 language.
Written by Philip Hazel
Original API code Copyright (c) 1997-2012 University of Cambridge
- New API code Copyright (c) 2015-2017 University of Cambridge
+ New API code Copyright (c) 2015-2018 University of Cambridge
-----------------------------------------------------------------------------
Redistribution and use in source and binary forms, with or without
@@ -6363,15 +6363,11 @@ for(;;)
if ((re->overall_options & PCRE2_NO_START_OPTIMIZE) == 0)
{
- PCRE2_SPTR save_end_subject = end_subject;
-
/* If firstline is TRUE, the start of the match is constrained to the first
line of a multiline string. That is, the match must be before or at the
first newline following the start of matching. Temporarily adjust
- end_subject so that we stop the optimization scans for a first code unit
- immediately after the first character of a newline (the first code unit can
- legitimately be a newline). If the match fails at the newline, later code
- breaks this loop. */
+ end_subject so that we stop the scans for a first code unit at a newline.
+ If the match fails at the newline, later code breaks the loop. */
if (firstline)
{
@@ -6388,13 +6384,6 @@ for(;;)
else
#endif
while (t < end_subject && !IS_NEWLINE(t)) t++;
-
- /* Note that we only need to advance by one code unit if we found a
- newline. If the newline is CRLF, a first code unit of LF should not
- match, because it is not at or before the newline. Similarly, only the
- first code unit of a Unicode newline might be relevant. */
-
- if (t < end_subject) t++;
end_subject = t;
}
@@ -6470,13 +6459,17 @@ for(;;)
#endif
}
- /* If we can't find the required code unit, break the bumpalong loop,
- to force a match failure, except when doing partial matching, when we
- let the next cycle run at the end of the subject. To see why, consider
- the pattern /(?<=abc)def/, which partially matches "abc", even though
- the string does not contain the starting character "d". */
-
- if (!mb->partial && start_match >= end_subject)
+ /* If we can't find the required code unit, having reached the true end
+ of the subject, break the bumpalong loop, to force a match failure,
+ except when doing partial matching, when we let the next cycle run at
+ the end of the subject. To see why, consider the pattern /(?<=abc)def/,
+ which partially matches "abc", even though the string does not contain
+ the starting character "d". If we have not reached the true end of the
+ subject (PCRE2_FIRSTLINE caused end_subject to be temporarily modified)
+ we also let the cycle run, because the matching string is legitimately
+ allowed to start with the first code unit of a newline. */
+
+ if (!mb->partial && start_match >= mb->end_subject)
{
rc = MATCH_NOMATCH;
break;
@@ -6538,7 +6531,7 @@ for(;;)
/* Restore fudged end_subject */
- end_subject = save_end_subject;
+ end_subject = mb->end_subject;
/* The following two optimizations must be disabled for partial matching. */
diff --git a/testdata/testinput2 b/testdata/testinput2
index fe8efbf..36e4454 100644
--- a/testdata/testinput2
+++ b/testdata/testinput2
@@ -5405,4 +5405,8 @@ a)"xI
\= Expect no match
xyz\r\nabc
+/[abc]/firstline
+\= Expect no match
+ \na
+
# End of testinput2
diff --git a/testdata/testinput6 b/testdata/testinput6
index 614c3a0..e2f00c0 100644
--- a/testdata/testinput6
+++ b/testdata/testinput6
@@ -4942,4 +4942,8 @@
\= Expect no match
xyz\r\nabc
+/[abc]/firstline
+\= Expect no match
+ \na
+
# End of testinput6
diff --git a/testdata/testoutput2 b/testdata/testoutput2
index 62ec12f..f146c0c 100644
--- a/testdata/testoutput2
+++ b/testdata/testoutput2
@@ -16453,6 +16453,11 @@ No match
xyz\r\nabc
No match
+/[abc]/firstline
+\= Expect no match
+ \na
+No match
+
# End of testinput2
Error -65: PCRE2_ERROR_BADDATA (unknown error number)
Error -62: bad serialized data
diff --git a/testdata/testoutput6 b/testdata/testoutput6
index 998f20b..b409fe0 100644
--- a/testdata/testoutput6
+++ b/testdata/testoutput6
@@ -7766,4 +7766,9 @@ Failed: error -47: match limit exceeded
xyz\r\nabc
No match
+/[abc]/firstline
+\= Expect no match
+ \na
+No match
+
# End of testinput6
--
2.13.6

105
pcre2-10.30-Remove-superflous-variable.patch
View File

@ -1,105 +0,0 @@
From 5e964db12e04a84c9b74751d5c22ae4ff4bb416c Mon Sep 17 00:00:00 2001
From: ph10 <ph10@6239d852-aaf2-0410-a92c-79f79f948069>
Date: Thu, 9 Nov 2017 17:50:59 +0000
Subject: [PATCH] Remove superflous variable.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
git-svn-id: svn://vcs.exim.org/pcre2/code/trunk@877 6239d852-aaf2-0410-a92c-79f79f948069
Signed-off-by: Petr Písař <ppisar@redhat.com>
---
src/pcre2grep.c | 21 ++++++++++-----------
1 file changed, 10 insertions(+), 11 deletions(-)
diff --git a/src/pcre2grep.c b/src/pcre2grep.c
index d75917c..1649d5a 100644
--- a/src/pcre2grep.c
+++ b/src/pcre2grep.c
@@ -2365,7 +2365,6 @@ while (ptr < endptr)
int mrc = 0;
unsigned int options = 0;
BOOL match;
- char *matchptr = ptr;
char *t = ptr;
size_t length, linelength;
size_t startoffset = 0;
@@ -2503,7 +2502,7 @@ while (ptr < endptr)
match, set PCRE2_NOTEMPTY to disable any further matches of null strings in
this line. */
- match = match_patterns(matchptr, length, options, startoffset, &mrc);
+ match = match_patterns(ptr, length, options, startoffset, &mrc);
options = PCRE2_NOTEMPTY;
/* If it's a match or a not-match (as required), do what's wanted. */
@@ -2564,14 +2563,14 @@ while (ptr < endptr)
/* Handle --line-offsets */
if (line_offsets)
- fprintf(stdout, "%d,%d" STDOUT_NL, (int)(matchptr + offsets[0] - ptr),
+ fprintf(stdout, "%d,%d" STDOUT_NL, (int)(ptr + offsets[0] - ptr),
(int)(offsets[1] - offsets[0]));
/* Handle --file-offsets */
else if (file_offsets)
fprintf(stdout, "%d,%d" STDOUT_NL,
- (int)(filepos + matchptr + offsets[0] - ptr),
+ (int)(filepos + ptr + offsets[0] - ptr),
(int)(offsets[1] - offsets[0]));
/* Handle --output (which has already been syntax checked) */
@@ -2579,7 +2578,7 @@ while (ptr < endptr)
else if (output_text != NULL)
{
if (display_output_text((PCRE2_SPTR)output_text, FALSE,
- (PCRE2_SPTR)matchptr, offsets, mrc) || printname != NULL ||
+ (PCRE2_SPTR)ptr, offsets, mrc) || printname != NULL ||
number)
fprintf(stdout, STDOUT_NL);
}
@@ -2601,7 +2600,7 @@ while (ptr < endptr)
{
if (printed && om_separator != NULL)
fprintf(stdout, "%s", om_separator);
- print_match(matchptr + offsets[n*2], plen);
+ print_match(ptr + offsets[n*2], plen);
printed = TRUE;
}
}
@@ -2628,7 +2627,7 @@ while (ptr < endptr)
{
if (startoffset >= length) goto END_ONE_MATCH; /* Were at end */
startoffset = oldstartoffset + 1;
- if (utf) while ((matchptr[startoffset] & 0xc0) == 0x80) startoffset++;
+ if (utf) while ((ptr[startoffset] & 0xc0) == 0x80) startoffset++;
}
/* If the current match ended past the end of the line (only possible
@@ -2637,7 +2636,7 @@ while (ptr < endptr)
while (startoffset > linelength)
{
- matchptr = ptr += linelength + endlinelength;
+ ptr += linelength + endlinelength;
filepos += (int)(linelength + endlinelength);
linenumber++;
startoffset -= (int)(linelength + endlinelength);
@@ -2784,10 +2783,10 @@ while (ptr < endptr)
{
startoffset = offsets[1];
if (startoffset >= linelength + endlinelength ||
- !match_patterns(matchptr, length, options, startoffset, &mrc))
+ !match_patterns(ptr, length, options, startoffset, &mrc))
break;
- FWRITE_IGNORE(matchptr + startoffset, 1, offsets[0] - startoffset, stdout);
- print_match(matchptr + offsets[0], offsets[1] - offsets[0]);
+ FWRITE_IGNORE(ptr + startoffset, 1, offsets[0] - startoffset, stdout);
+ print_match(ptr + offsets[0], offsets[1] - offsets[0]);
}
/* In multiline mode, we may have already printed the complete line
--
2.13.6

96
pcre2-10.30-Update-pcre2demo-to-deal-with-various-K-inside-asser.patch
View File

@ -1,96 +0,0 @@
From f442210323e228a407dfda75b6bb7a62e91111ee Mon Sep 17 00:00:00 2001
From: ph10 <ph10@6239d852-aaf2-0410-a92c-79f79f948069>
Date: Sat, 23 Dec 2017 17:15:51 +0000
Subject: [PATCH 1/3] Update pcre2demo to deal with various \K inside assertion
anomalies.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
git-svn-id: svn://vcs.exim.org/pcre2/code/trunk@894 6239d852-aaf2-0410-a92c-79f79f948069
Petr Písař: Ported to 10.30.
---
src/pcre2demo.c | 53 +++++++++++++++++++++++++++++++++++++++++++++++++++++
diff --git a/src/pcre2demo.c b/src/pcre2demo.c
index 8ae49f1..5d9b321 100644
--- a/src/pcre2demo.c
+++ b/src/pcre2demo.c
@@ -211,6 +211,21 @@ pcre2_match_data_create_from_pattern() above. */
if (rc == 0)
printf("ovector was not big enough for all the captured substrings\n");
+/* We must guard against patterns such as /(?=.\K)/ that use \K in an assertion
+to set the start of a match later than its end. In this demonstration program,
+we just detect this case and give up. */
+
+if (ovector[0] > ovector[1])
+ {
+ printf("\\K was used in an assertion to set the match start after its end.\n"
+ "From end to start the match was: %.*s\n", (int)(ovector[0] - ovector[1]),
+ (char *)(subject + ovector[1]));
+ printf("Run abandoned\n");
+ pcre2_match_data_free(match_data);
+ pcre2_code_free(re);
+ return 1;
+ }
+
/* Show substrings stored in the output vector by number. Obviously, in a real
application you might want to do things other than print them. */
@@ -338,6 +353,29 @@ for (;;)
options = PCRE2_NOTEMPTY_ATSTART | PCRE2_ANCHORED;
}
+ /* If the previous match was not an empty string, there is one tricky case to
+ consider. If a pattern contains \K within a lookbehind assertion at the
+ start, the end of the matched string can be at the offset where the match
+ started. Without special action, this leads to a loop that keeps on matching
+ the same substring. We must detect this case and arrange to move the start on
+ by one character. The pcre2_get_startchar() function returns the starting
+ offset that was passed to pcre2_match(). */
+
+ else
+ {
+ PCRE2_SIZE startchar = pcre2_get_startchar(match_data);
+ if (start_offset <= startchar)
+ {
+ if (startchar >= subject_length) break; /* Reached end of subject. */
+ start_offset = startchar + 1; /* Advance by one character. */
+ if (utf8) /* If UTF-8, it may be more */
+ { /* than one code unit. */
+ for (; start_offset < subject_length; start_offset++)
+ if ((subject[start_offset] & 0xc0) != 0x80) break;
+ }
+ }
+ }
+
/* Run the next matching operation */
rc = pcre2_match(
@@ -402,6 +440,21 @@ for (;;)
if (rc == 0)
printf("ovector was not big enough for all the captured substrings\n");
+ /* We must guard against patterns such as /(?=.\K)/ that use \K in an
+ assertion to set the start of a match later than its end. In this
+ demonstration program, we just detect this case and give up. */
+
+ if (ovector[0] > ovector[1])
+ {
+ printf("\\K was used in an assertion to set the match start after its end.\n"
+ "From end to start the match was: %.*s\n", (int)(ovector[0] - ovector[1]),
+ (char *)(subject + ovector[1]));
+ printf("Run abandoned\n");
+ pcre2_match_data_free(match_data);
+ pcre2_code_free(re);
+ return 1;
+ }
+
/* As before, show substrings stored in the output vector by number, and then
also any named substrings. */
--
2.13.6

80
pcre2.spec
View File

@ -1,12 +1,15 @@
# Add readline edditing in pcre2test tool
%bcond_without pcre2_enables_readline
# Disable SELinux-frindly JIT allocator because it seems not to be fork-safe,
# https://bugs.exim.org/show_bug.cgi?id=1749#c45
%{bcond_with pcre2_enables_sealloc}
%bcond_with pcre2_enables_sealloc
# This is stable release:
#%%global rcversion RC1
%global rcversion RC1
Name: pcre2
Version: 10.30
Release: %{?rcversion:0.}5%{?rcversion:.%rcversion}%{?dist}
Version: 10.31
Release: %{?rcversion:0.}1%{?rcversion:.%rcversion}%{?dist}
%global myversion %{version}%{?rcversion:-%rcversion}
Summary: Perl-compatible regular expression library
# the library: BSD with exceptions
@ -17,23 +20,26 @@ Summary: Perl-compatible regular expression library
# for testdata
#Bundled
# src/sljit: BSD
#Not distributed in binary package
#Not distributed in any binary package
# aclocal.m4: FSFULLR and GPLv2+ with exception
# ar-lib: GPLv2+ with exception
# autotools: GPLv3+ with exception
# cmake/COPYING-CMAKE-SCRIPTS: BSD
# compile: GPLv2+ with exception
# config.sub: GPLv3+ with exception
# configure: FSFUL and GPLv2+ with exception
# depcomp: GPLv2+ with exception
# install-sh: MIT
# ltmain.sh: GPLv2+ with exception and GPLv3+ with
# exception and GPLv3+
# ltmain.sh: GPLv2+ with exception and (MIT or GPLv3+)
# m4/ax_pthread.m4: GPLv3+ with exception
# m4/libtool.m4: FSFULLR and GPLv2+ with exception
# m4/libtool.m4: FSFUL and FSFULLR and
# GPLv2+ with exception
# m4/ltoptions.m4: FSFULLR
# m4/ltsugar.m4: FSFULLR
# m4/ltversion.m4: FSFULLR
# m4/lt~obsolete.m4: FSFULLR
# m4/pcre2_visibility.m4: FSFULLR
# Makefile.in: FSFULLR
# missing: GPLv2+ with exception
# test-driver: GPLv2+ with exception
# testdata: Public Domain
@ -42,44 +48,6 @@ URL: http://www.pcre.org/
Source: ftp://ftp.csx.cam.ac.uk/pub/software/programming/pcre/%{?rcversion:Testing/}%{name}-%{myversion}.tar.bz2
# Do no set RPATH if libdir is not /usr/lib
Patch0: pcre2-10.10-Fix-multilib.patch
# 1/2 Accept files names longer than 128 bytes in recursive mode of pcre2grep,
# upstream bug #2177, in upstream after 10.30
Patch1: pcre2-10.30-Fix-pcre2grep-recursive-file-name-length-issue.patch
# 2/2 Accept files names longer than 128 bytes in recursive mode of pcre2grep,
# upstream bug #2177, in upstream after 10.30
Patch2: pcre2-10.30-Fix-memory-leak-issue-introduced-in-last-bug-fix-in-.patch
# Required for Fix-multiple-multiline-matching-issues-in-pcre2grep.patch
Patch3: pcre2-10.30-Remove-superflous-variable.patch
# Fix multi-line matching in pcre2grep tool, upstream bug #2187,
# in upstream after 10.30
Patch4: pcre2-10.30-Fix-multiple-multiline-matching-issues-in-pcre2grep.patch
# Fix pcre2_jit_match() to properly check the pattern was JIT-compiled,
# in upstream after 10.30
Patch5: pcre2-10.30-Fix-pcre2_jit_match-early-check.patch
# Allow pcre2grep match counter to handle values larger than 2147483647,
# upstream bug #2208, in upstream after 10.30
Patch6: pcre2-10.30-Change-pcre2grep-line-number-and-count-variables-to-.patch
# Fix incorrect first matching character when a backreference with zero minimum
# repeat starts a pattern, upstream bug #2209, in upstream after 10.30
Patch7: pcre2-10.30-Fix-incorrect-first-matching-character-when-a-backre.patch
# 1/2 Fix handling \K in an assertion in documentation, upstream bug #2211,
# in upstream after 10.30
Patch8: pcre2-10.30-Update-pcre2demo-to-deal-with-various-K-inside-asser.patch
# 2/2 Fix handling \K in an assertion in documentation, upstream bug #2211,
# upstream bug #2211, in upstream after 10.30
Patch9: pcre2-10.30-Documentation-update.patch
# Fix handling \K in an assertion in pcre2grep tool, upstream bug #2211,
# in upstream after 10.30
Patch10: pcre2-10.30-Fix-K-issues-in-pcre2grep.patch
# 1/3 Fix matching at a first code unit of a new line sequence if
# PCRE2_FIRSTLINE is enabled, in upstream after 10.30
Patch11: pcre2-10.30-FIRSTLINE_documentation-update.patch
# 2/3 Fix matching at a first code unit of a new line sequence if
# PCRE2_FIRSTLINE is enabled, in upstream after 10.30
Patch12: pcre2-10.30-Fix-PCRE2_FIRSTLINE-bug-when-a-pattern-match-starts-.patch
# 3/3 Fix matching at a first code unit of a new line sequence if
# PCRE2_FIRSTLINE is enabled, in upstream after 10.30
Patch13: pcre2-10.30-Previous-FIRSTLINE-patch-was-broken.-Fix-it.patch
BuildRequires: autoconf
BuildRequires: automake
BuildRequires: coreutils
@ -154,19 +122,6 @@ Utilities demonstrating PCRE2 capabilities like pcre2grep or pcre2test.
%prep
%setup -q -n %{name}-%{myversion}
%patch0 -p1
%patch1 -p1
%patch2 -p1
%patch3 -p1
%patch4 -p1
%patch5 -p1
%patch6 -p1
%patch7 -p1
%patch8 -p1
%patch9 -p1
%patch10 -p1
%patch11 -p1
%patch12 -p1
%patch13 -p1
# Because of multilib patch
libtoolize --copy --force
autoreconf -vif
@ -203,7 +158,11 @@ autoreconf -vif
--disable-pcre2grep-libbz2 \
--disable-pcre2grep-libz \
--disable-pcre2test-libedit \
%if %{with pcre2_enables_readline}
--enable-pcre2test-libreadline \
%else
--disable-pcre2test-libreadline \
%endif
--disable-rebuild-chartables \
--enable-shared \
--disable-silent-rules \
@ -270,6 +229,9 @@ make %{?_smp_mflags} check VERBOSE=yes
%{_mandir}/man1/pcre2test.*
%changelog
* Mon Jan 15 2018 Petr Pisar <ppisar@redhat.com> - 10.31-0.1.RC1
- 10.31-RC1 bump
* Fri Jan 12 2018 Petr Pisar <ppisar@redhat.com> - 10.30-5
- Fix handling \K in an assertion in pcre2grep tool and documentation
(upstream bug #2211)

2
sources
View File

@ -1 +1 @@
SHA512 (pcre2-10.30.tar.bz2) = f247a9f917c75920793b9919a45bb1426d126246e7a5d04e39d9407e44b5781f894a90cd3d232b385436b2f22be391335ab782664dd3a28c79058a2fcc74dc3e
SHA512 (pcre2-10.31-RC1.tar.bz2) = 1eef755c775575fdea1f3d90590affb1363265995197c04f5b82b3e0f9e9f79aa7e82689815c3e3a4a7673f76be2b00cf9dc9a399ce008baeb76d08a9aa3b50f