* tests/write-error-msg: Do not require that a disk full
diagnostic include additional information. In some cases, there
is no valid errno value, so we cannot provide more information.
This was exposed by a patch that coincidentally caused the length
of grep's help output to be precisely 4096 bytes long.
Reported in https://bugs.gnu.org/77800
* tests/mb-non-UTF8-perf-Fw: Ugh. I misread the code and
didn't even test. Given circumstances and the new timing that's
well within the 30-second timeout, I think there's no point
in trying to accommodate systems that are so overburdened they
trigger this failure. Reported by Grisha Levit.
This test would hang on GNU/Hurd because the perl code we use to measure
subsecond duration isn't ported, and that loop would never terminate.
* tests/hash-collision-perf: Detect the always-0 small_ms, and skip the test.
Reported by Bruno Haible in https://bugs.gnu.org/77613
* src/grep.c (grepbuf): Handle this case: echo x|grep -l -m1 .
making it print only the file name, and not the matched line.
(main): Set out_quiet also when exit_on_match (-q) is set, so
"echo x|grep -q -m1 ." no longer prints the matched line.
* tests/max-count-overread: Add those tests, from
https://bugs.gnu.org/68989#21
I misread the email thread and thought there was consensus
for the \d change, but there was wasn’t so revert the change.
Also, document the resulting confusion
somewhat better than it was documented before.
* src/pcresearch.c, tests/pcre-ascii-digits, tests-pcre-utf8-w:
Revert recent \d change, restoring the behavior to that of grep 3.11.
Treating \d differently from Perl was more trouble than it was worth.
* NEWS, doc/grep.texi (grep Programs): Document this.
* src/pcresearch.c (PCRE2_EXTRA_ASCII_BSD):
Remove. All uses removed.
* tests/pcre-ascii-digits: Adjust to this change.
* tests/pcre-utf8-w: Revert to 3.9.
Problem reported by Jan Černohorský (Bug#74159).
* src/grep.c (grepbuf): If exit_on_match, set stdout_errno to
avoid screwups on buggy OSes. Also, ignore errseen since it
cannot be true here.
(main): Do not clear exit_failure if -q is given, as exit status
should be zero only if an input line is selected.
* tests/write-error-msg: Check that -q suppresses diagnostics
of output errors only if a match is found.
To accommodate a new syntax-check rule, ...
Make comments that suggested to write to an old FSF Franklin Street
address refer to https://www.gnu.org/licenses instead:
git grep -l 'if not, write' |xargs \
perl -0777 -pi -e 's{program; if not, write .*? USA\.}{program. If not, see <https://www.gnu.org/licenses/>.}ms'
Problem reported by Grisha Levit <https://bugs.gnu.org/68989>.
* src/grep.c (grep, main): Don’t set done_on_match if -m is used.
* tests/max-count-overread: Add a test case.
Omit Gnulib’s propername module, as it has not been used since my
commit 3c0a36e514237132db711bfef57a74c64592c4e2 dated Thu Dec 20
16:35:55 2018 -0800.
* bootstrap.conf (avoided_gnulib_modules):
Do not avoid mbchar, as it is no longer pulled in by propername.
(gnulib_modules): Remove propername.
* src/Makefile.am (LDADD):
* tests/Makefile.am (LDADD): Remove $(LIBICONV); no longer needed.
* src/grep.c: Do not include propername.h.
* NEWS: document the fixed bug.
* tests/100k-entries: New file, to test for this.
Reported by Vincent Lefevre via Santiago Ruano Rincón in
https://bugs.gnu.org/64773
Fixed by gnulib commit v0.1-6175-gd4d8abb39e.
PCRE2 has a bug when using PCRE2_MATCH_INVALID_UTF: it would
sometimes fail to match patterns using negative classes
like \W and \D.
* NEWS (Bug fixes): Mention it.
* src/pcre2search.c: Restrict impact of the bug.
Do not use the problematic flag with broken versions of PCRE2.
Also, generate locale tables only for single-byte locales,
as the PCRE2 documentation recommends this.
* tests/Makefile.am (TESTS): Add the file name
* tests/pcre-utf8-bug224: New file, to test for this.
* src/grep.c: No need to include pcre2.h.
(main) [HAVE_LIBPCRE]: Call Pprint_version instead of
doing it ourselves.
* src/pcresearch.c (Pprint_version): New function.
It also checks belatedly for buffer overflow, and
says "grep -P uses PCRE2" instead of "Built with PCRE".
* tests/version-pcre: Adjust test to match.
* tests/glibc-infloop: New file.
Based on the command from Koen Claessen
reported in https://bugs.gnu.org/62483
* configure.ac (USE_INCLUDED_REGEX): define.
* tests/Makefile.am (TESTS): Add the file name
* THANKS.in: Add name of reporter.
PCRE is integral to the functioning of grep's -P option, so it is in our
interest to make it easy to see which version of PCRE grep uses.
* src/grep.c [HAVE_LIBPCRE]: Include <pcre2.h>.
[HAVE_LIBPCRE] (main): Print pcre version info.
* tests/version-pcre: New test for this.
* tests/Makefile.am (TESTS): Add the file name.
* NEWS (Changes in behavior): Mention it.
Our prepass-based fixes for the -P \d bug have caused repeated
further bugs. Avoid the need for a prepass, by using PCRE2_UCP
only if PCRE2_EXTRA_ASCII_BSD is also supported. Since the -P \w
bug was present from grep 2.5 through 3.8 it’s OK if we wait a
little longer to fix it.
* NEWS: Mention this.
* src/pcresearch.c (pcre_pattern_expand_backslash_d}: Remove.
Remove its use.
(Pcompile): Use PCRE2_UCP only if PCRE2_EXTRA_ASCII_BSD.
* tests/pcre-ascii-digits, tests/pcre-utf8-w:
Skip tests on older PCRE2 implementations.
* NEWS: Mention \D, too.
* doc/grep.texi: Likewise
* src/pcresearch.c (pcre_pattern_expand_backslash_d): Handle \D.
Also, ifdef-out this new function and its call site when not needed.
* tests/pcre-ascii-digits: Test \D, too.
Tighten one test by using returns_ 1.
Add comments and tests that work only with 10.43 and newer.
Paul Eggert raised the issue of \D in https://bugs.gnu.org/62267#8
Prior to grep-3.9, the PCRE matcher had always treated \d just
like [0-9]. grep-3.9's fix for \w and \b mistakenly relaxed \d
to also match multibyte digits.
* src/grep.c (P_MATCHER_INDEX): Define enum.
(pcre_pattern_expand_backslash_d): New function.
(main): Call it for -P.
* NEWS (Bug fixes): Mention it.
* doc/grep.texi: Document it: with -P, \d matches only ASCII digits.
Provide a PCRE documentation URL and an example of how
to use (?s) with -z.
* tests/pcre-ascii-digits: New test.
* tests/Makefile.am (TESTS): Add that file name.
Reported as https://bugs.gnu.org/62267
It’s obsolete in bleeding-edge Gnulib.
* src/grep.c, tests/get-mb-cur-max.c: Don’t include getprogname.h.
Instead, rely on stdlib.h to declare getprogname.
* tests/hangul-syllable, tests/surrogate-search:
32-bit AIX has WCHAR_MAX == 0xFFFF, and so cannot handle
U+10000 and greater. Skip tests involving such chars
on this platform.
Before this change, if linked with a PCRE library without unicode
any invocations of grep when using a UTF locale will error with:
grep: this version of PCRE2 does not have Unicode support
* src/pcresearch.c: Check whether Unicode was compiled in.
* tests/pcre-utf8-w: Add check to skip test.
* tests/pcre-utf8: Update check.
This fixes a serious bug affecting word-boundary and word-constituent regular
expressions when the desired match involves non-ASCII UTF8 characters.
* src/pcresearch.c: Set PCRE2_UCP together with PCRE2_UTF
* tests/pcre-utf8-w: New file.
* tests/Makefile.am (TESTS): Add it.
* NEWS (Bug fixes): Mention this.
* THANKS.in: Add Gro-Tsen and Karl Petterson.
Reported by Gro-Tsen https://twitter.com/gro_tsen/status/1610972356972875777
via Karl Pettersson in https://github.com/PCRE2Project/pcre2/issues/185
This bug was present from grep-2.5, when --perl-regexp (-P) support was added.
* NEWS: Mention this.
* src/dfasearch.c (GEAcompile): Trim trailing newline from
the last pattern, even if it has back-references and follows
a pattern that lacks back-references.
* tests/backref: Add test for this bug.
* tests/long-pattern-perf: Don't fail due to a syntax error
when one of the subtests exhausts virtual memory. The larger
test (with a 2MiB regexp) needs about 870MiB of virtual memory.
Require that each timing run exit with status 0, else fail with
a framework_failure_. Reported by Bruno Haible in
https://lists.gnu.org/r/grep-devel/2022-07/msg00006.html
* tests/triple-backref: I noticed that our sole XFAIL is still
required, in spite of a glibc comment that bug 11053 is fixed,
so confirmed that it no longer evokes an abort, but still fails
to produce the expected match. I.e., this prints nothing:
echo a | grep -E '(.?)(.?)(.?)\3\2\1' -- it should print its input.
* tests/init.cfg (require_perl_): New function.
* tests/big-hole, tests/hash-collision-perf, tests/long-pattern-perf:
* tests/many-regex-performance, tests/mb-non-UTF8-performance:
Use it.
