When the command-line parser was reworked, a mutually exclusive group
was deleted and check for conflicting options was eliminated as well.
The commit implements the lost feature in a more elegant way than it
has been done before, allowing more flexibility.
In fact, access to the gnulib repository through CVS has been disabled,
or more precisely, got broken and was never restored; see:
<http://lists.gnu.org/archive/html/bug-gnulib/2013-05/msg00008.html>
gnulib-tool: fix incompatibility with autopoint 0.18.2
* gnulib-tool: Don't indent AM_GNU_GETTEXT_VERSION line.
Problem reported by Tom G. Christensen in
<http://lists.gnu.org/archive/html/bug-gnulib/2013-01/msg00053.html>.
Bruno Haible [Fri, 8 Sep 2017 08:02:47 +0000 (10:02 +0200)]
stddef: Avoid conflict with system-defined max_align_t.
The configure-determined HAVE_MAX_ALIGN_T may not always be accurate.
Reported by Werner Lemberg <wl@gnu.org> in
<https://lists.gnu.org/archive/html/bug-gnulib/2017-08/msg00185.html>.
* lib/stddef.in.h (rpl_max_align_t): Renamed from max_align_t.
(max_align_t): Define as a macro.
(GNULIB_defined_max_align_t): New macro. Guards against multiple
definitions of rpl_max_align_t in different copies of gnulib-generated
<stddef.h>.
Paul Eggert [Wed, 6 Sep 2017 06:32:59 +0000 (23:32 -0700)]
libc-config: port to MSVC
Problems reported by Gisle Vanem in:
http://lists.gnu.org/archive/html/bug-gnulib/2017-09/msg00016.html
* lib/libc-config.h (__inline): Don't define if HAVE___INLINE.
(libc_hidden_proto): Stick to Standard C syntax for varargs macro.
* m4/__inline.m4: New file.
* modules/libc-config (Files): Add it.
(Depends-on): Use it.
Paul Eggert [Wed, 6 Sep 2017 01:58:50 +0000 (18:58 -0700)]
glob: fix for use in glibc
Problem reported by Adhemerval Zanella in:
https://sourceware.org/ml/libc-alpha/2017-09/msg00213.html
* lib/glob.c (DT_UNKNOWN, DT_DIR, DT_LINK):
Do not redefine if _LIBC.
Paul Eggert [Sat, 2 Sep 2017 22:39:16 +0000 (15:39 -0700)]
glob: fix bugs with long login names
Problem reported by Adhemerval Zanella in:
https://sourceware.org/ml/libc-alpha/2017-08/msg00455.html
* lib/glob.c (GET_LOGIN_NAME_MAX): Remove.
(glob): Use the same scratch buffer for both getlogin_r and
getpwnam_r. Don’t require preallocation of the login name. This
simplifies storage allocation, and corrects the handling of
long login names.
Paul Eggert [Fri, 1 Sep 2017 22:41:45 +0000 (15:41 -0700)]
glob: use scratch_buffer instead of extend_alloca
Much of the lib/glob.c part of this patch comes from a glibc patch
proposed by Adhemerval Zanella in:
https://sourceware.org/ml/libc-alpha/2017-08/msg00456.html
* lib/glob.c: Do not include <config.h>, since <libc-config.h>,
included via glob.h, does this for us now.
(__set_errno): Remove, as libc-config does this for us now.
Include <scratch_buffer.h>.
(GETPW_R_SIZE_MAX): Remove.
(glob): Use struct scratch_buffer instead of extend_alloca.
* lib/glob.in.h: Include libc-config.h rather than
including <sys/cdefs.h> conditionally.
(__BEGIN_DECLS, __END_DECLS, __THROW, __THROWNL, attribute_hidden)
(__glibc_unlikely, __restrict, weak_alias):
Remove, as libc-config does this for us now.
* m4/glob.m4 (gl_PREREQ_GLOB):
Remove sys/cdefs.h tests; no longer needed.
* modules/glob (Depends-on): Add libc-config, scratch_buffer.
(glob.h): Do not replace HAVE_SYS_CDEFS_H.
Paul Eggert [Fri, 1 Sep 2017 21:42:53 +0000 (14:42 -0700)]
scratch_buffer: new module
* lib/scratch_buffer.h, lib/scratch_buffer_grow.c:
* lib/scratch_buffer_grow_preserve.c:
* lib/scratch_buffer_set_array_size.c:
New files, copied from glibc with very minor changes that can be
copied back.
* modules/scratch_buffer: New file.
Paul Eggert [Fri, 1 Sep 2017 21:40:58 +0000 (14:40 -0700)]
libc-config: new module
* MODULES.html.sh: Add libc-config.
* lib/cdefs.h: New file, copied from the GNU C Library with very
minor changes that can be copied back.
* lib/libc-config.h, modules/libc-config: New files.
Paul Eggert [Thu, 31 Aug 2017 21:34:25 +0000 (14:34 -0700)]
glob: match dangling symlinks
This fixes a bug I inadvertently introduced to Gnulib when I
merged glibc glob back into gnulib on 2007-10-16. This fix is
inspired by a patch proposed for glibc by Adhemerval Zanella in:
https://sourceware.org/ml/libc-alpha/2017-08/msg00446.html
* doc/posix-functions/glob.texi: Update list of affected platforms.
* lib/glob.c (__lstat64): New macro.
(is_dir): New function.
(glob, glob_in_dir): Match symlinks even if they are dangling.
(link_stat, link_exists_p): Remove. All uses removed.
* lib/glob.in.h (__attribute_noinline__): Remove; no longer used.
* m4/glob.m4 (gl_PREREQ_GLOB): Do not check for fstatat.
* modules/glob (Depends-on): Remove dirfd.
* modules/glob-tests (Depends-on): Add symlink.
* tests/test-glob.c: Include errno.h, unistd.h.
(BASE): New macro.
(main): Test dangling symlinks, if symlinks are supported.
Paul Eggert [Thu, 31 Aug 2017 21:34:24 +0000 (14:34 -0700)]
glob: simplify symlink detection
* lib/glob.c (dirent_type): New type. Use uint_fast8_t not
uint8_t, as C99 does not require uint8_t.
(DT_UNKNOWN, DT_DIR, DT_LNK) [!HAVE_STRUCT_DIRENT_D_TYPE]:
New macros.
(struct readdir_result): Use dirent_type. Do not define skip_entry
unless it is needed; this saves a byte on platforms lacking d_ino.
(readdir_result_type, readdir_result_skip_entry):
New functions, replacing ...
(readdir_result_might_be_symlink, readdir_result_might_be_dir):
... these functions, which were removed. This makes the callers
easier to read. All callers changed.
(D_INO_TO_RESULT): Now empty if there is no d_ino.
Paul Eggert [Fri, 25 Aug 2017 06:51:07 +0000 (23:51 -0700)]
glob: try to port recent changes to MS-Windows
Problem reported by Bruno Haible in:
http://lists.gnu.org/archive/html/bug-gnulib/2017-08/msg00170.html
* lib/glob.c (__glob_pattern_p) [!_LIBC]: Move from here ...
* lib/glob.in.h (__glob_pattern_p): ... to here.
Eric Blake [Fri, 25 Aug 2017 01:28:52 +0000 (20:28 -0500)]
warnings: fix compilation with old autoconf
Autoconf older than 2.63b (such as what ships on CentOS 6) had
a bug that any AC_DEFUN'd macro name that includes shell meta-
characters causes failure due to missing shell quoting during
aclocal's use of autom4te. We can work around the problem by
using m4_defun instead (same semantics in autom4te, but no
longer traced by aclocal, so no longer tickles the shell
quoting bug).
* m4/warnings.m4 (gl_UNKNOWN_WARNINGS_ARE_ERRORS(C))
(gl_UNKNOWN_WARNINGS_ARE_ERRORS(C++)): Use m4_defun rather than
AC_DEFUN.
* m4/manywarnings.m4 (gl_MANYWARN_ALL_GCC(C))
(gl_MANYWARN_ALL_GCC(C++)): Likewise.
Reported-by: Erik Skultety <eskultet@redhat.com> Signed-off-by: Eric Blake <eblake@redhat.com>
Paul Eggert [Wed, 23 Aug 2017 17:45:42 +0000 (10:45 -0700)]
glob: merge from glibc with Zanella glob changes
Merge glob from glibc, with changes for glob proposed
by Adhemerval Zanella in the thread starting here:
https://sourceware.org/ml/libc-alpha/2017-08/msg01079.html
plus some fixes for this merge.
* lib/flexmember.h: Change license wording to something that
works unchanged in Glibc, since this code might be used in
Glibc and this will minimize 'diff' output.
* lib/glob_internal.h, lib/glob_pattern_p.c, lib/globfree.c:
New files, ported from glibc.
* lib/glob_pattern_p.c, lib/globfree.c, lib/glob.c [!_LIBC]:
Include <config.h> first.
* lib/glob-libc.h (_Restrict_): Remove. All uses replaced
with __restrict.
(__size_t): Remove. All uses replaced by size_t.
(size_t): Define by defining __need_size_t and including <stddef.h>.
This should work even in non-glibc platforms, where any name
pollution is OK.
Use __USE_MISC instead of __USE_BSD || __USE_GNU.
(struct stat64): Don’t worry about __GLOB_GNULIB.
(glob, globfree, glob_pattern_p): Remove macros for
__USE_FILE_OFFSET64 && __GNUC__ < 2 && !defined __GLOB_GNULIB
case. Remove _GL_ARG_NONNULL as GNU behavior is to accept NULL
but set errno.
* lib/glob.c (_GL_ARG_NONNULL) [!_LIBC]: Remove. All uses
removed since the glibc behavior works on null pointers.
Do not include stdio.h; old SunOS is irrelevant now.
Do not worry about GLOB_ONLY_P as we now mimic glibc here.
Include glob_internal.h.
(D_INO_TO_RESULT): Depend on (_LIBC || D_INO_IN_DIRENT), not
((POSIX || WINDOWS32) && !__GNU_LIBRARY__). The latter probably
worked only coincidentally.
(attribute_hidden, __attribute_noinline__, __glibc_unlikely):
Remove macros; now done in glob.in.h.
(size_add_wrapv): Do not use __builtin_add_overflow if __ICC.
(glob): Properly initialize glob structure with
GLOB_BRACE|GLOB_DOOFFS (bug 20707).
Remove old code using SHELL since Bash no longer
uses this.
(glob, prefix_array): Separate MS code better.
(glob, glob_in_dir): Use C99 decls before statements when glibc
does.
(glob_in_dir): Remove old Amiga and VMS code.
(globfree, __glob_pattern_type, __glob_pattern_p): Move to
separate files.
* lib/glob.in.h (attribute_hidden, __attribute_noinline__)
(__glibc_unlikely):
Move here from glob.c.
(__restrict): New macro here, replacing the _Restrict_ in glob.c.
(weak_alias): New macro.
(__size_t): Remove. All uses replaced by size_t.
* modules/d-ino (License): Now LGPLv2+, for compatibility with glob.
* modules/flexmember (License): Now LGPLv2+, which it should have
been anyway since flexmember.h is not unlimited-license.
* modules/glob (Files): Add +lib/glob_internal.h,
lib/glob_pattern_p.c, lib/globfree.c.
(Depends-on): Remove snippet/arg-nonnull.
Paul Eggert [Tue, 22 Aug 2017 17:13:50 +0000 (10:13 -0700)]
glob: port to clang's Undefined Sanitizer
Problem reported by Tim Rühsen in:
http://lists.gnu.org/archive/html/bug-gnulib/2017-08/msg00144.html
* lib/glob.c (FLEXIBLE_ARRAY_MEMBER) [_LIBC]: Define to empty.
(glob_in_dir): Do not rely on undefined behavior in accessing
struct members beyond their bounds. Use a flexible array member
instead.
Paul Eggert [Mon, 21 Aug 2017 00:03:55 +0000 (17:03 -0700)]
git-version-gen: port to Solaris 10
Problem reported by Dagobert Michelsen in:
http://lists.gnu.org/archive/html/grep-devel/2017-08/msg00002.html
* build-aux/git-version-gen (v_from_git):
Use expr instead of shell substitution.
Paul Eggert [Wed, 16 Aug 2017 15:34:41 +0000 (08:34 -0700)]
rename: port better to NetBSD
* doc/posix-functions/rename.texi (rename): NetBSD 7
does not have the link-count bug.
* m4/rename.m4 (gl_FUNC_RENAME): Don’t consider NetBSD to be
broken merely because rename ("a", "b") removes "a" when the two
names are hard links to the same file.
Bruno Haible [Wed, 16 Aug 2017 08:13:42 +0000 (10:13 +0200)]
rename, renameat: Update doc regarding NetBSD.
* doc/posix-functions/rename.texi: Clarify that when using
-D_XOPEN_SOURCE=500 on NetBSD 7.0, the hard link bug is gone.
* doc/posix-functions/renameat.texi: Be more precise about NetBSD
version.
Paul Eggert [Tue, 15 Aug 2017 23:47:22 +0000 (16:47 -0700)]
futimens: don’t assume struct timespec layout
* m4/futimens.m4 (gl_FUNC_FUTIMENS):
* m4/utimensat.m4 (gl_FUNC_UTIMENSAT):
* tests/test-fdutimensat.c (main):
* tests/test-futimens.h (test_futimens):
* tests/test-lutimens.h (test_lutimens):
* tests/test-utimens.h (test_utimens):
* tests/test-utimensat.c (main):
Don’t assume that struct timespec is a two-member structure in
tv_sec, tv_nsec order. Although this is true on all platforms we
know about, POSIX does not guarantee it.
Paul Eggert [Tue, 15 Aug 2017 22:53:50 +0000 (15:53 -0700)]
rename: document+test NetBSD rename
Test failure reported by Bruno Haible in:
http://lists.gnu.org/archive/html/bug-gnulib/2017-08/msg00104.html
This is an area where NetBSD is better-behaved than POSIX,
so allow the NetBSD behavior in tests.
* doc/posix-functions/rename.texi:
* doc/posix-functions/renameat.texi: Document NetBSD behavior.
* tests/test-rename.h (test_rename): Allow NetBSD behavior.
Bruno Haible [Tue, 15 Aug 2017 19:18:44 +0000 (21:18 +0200)]
duplocale: Work around NetBSD 7.0 bug.
* m4/duplocale.m4 (gl_FUNC_DUPLOCALE): Test against the NetBSD 7.0 bug.
* lib/duplocale.c: Add comment about NetBSD problem.
* doc/posix-functions/duplocale.texi: Mention the NetBSD problem.
Bruno Haible [Tue, 15 Aug 2017 18:23:00 +0000 (20:23 +0200)]
duplocale tests: Verify use with *_l functions.
* modules/duplocale-tests (configure.ac): Test for uselocale and
some *_l functions.
* tests/test-duplocale.c (test_with_uselocale): New function, extracted
from main.
(get_locale_dependent_values_from, test_with_locale_parameter): New
functions.
(main): Test both test_with_uselocale and test_with_locale_parameter.
Paul Eggert [Mon, 14 Aug 2017 20:04:46 +0000 (13:04 -0700)]
open: support O_CLOEXEC
* NEWS, doc/posix-functions/open.texi:
* doc/posix-functions/openat.texi: Document this.
* lib/fcntl.in.h (O_CLOEXEC): Default to a nonzero value.
(GNULIB_defined_O_CLOEXEC): New symbol.
* lib/open.c: Include cloexec.h.
(open): Support O_CLOEXEC.
* lib/openat.c: Include cloexec.h.
(rpl_openat): Support O_CLOEXEC.
* lib/popen-safer.c: Do not include cloexec.h.
(open_noinherit): Remove.
(popen_safer): Use O_CLOEXEC instead of set_cloexec_flag.
* lib/save-cwd.c: Do not include cloexec.h.
(save_cwd): Use O_CLOEXEC instead of set_cloexec_flag.
* m4/open-cloexec.m4: New file.
* m4/open.m4 (gl_FUNC_OPEN): Require gl_PREPROC_O_CLOEXEC.
Replace 'open' if O_CLOEXEC is not present.
* m4/openat.m4 (gl_FUNC_OPENAT): Require gl_PREPROC_O_CLOEXEC.
Replace 'openat' if O_CLOEXEC is not present.
* modules/freopen (Depends-on): Depend on 'open' if replacing freopen.
* modules/open (Files): Add m4/open-cloexec.m4.
(Depends-on): Depend on cloexec if replacing 'open'.
* modules/openat (Files): Add m4/open-cloexec.m4.
(Depends-on): Depend on cloexec if replacing openat.
* modules/popen-safer (Depends-on): Remove cloexec.
* modules/save-cwd (Depends-on): Remove cloexec, and add
fd-safer-flag and 'open'.
Darshit Shah [Sun, 13 Aug 2017 17:13:33 +0000 (10:13 -0700)]
reallocarray: New module
reallocarray is a new function in glibc 2.26 to safely allocate an array
of memory locations with integer overflow protection.
* MODULES.html.sh: Add reallocarray.
* doc/glibc-functions/reallocarray.texi: Documentation for reallocarray.
* lib/reallocarray.c: New file to implement module reallocarray.
* lib/stdlib.in.h: Add function declarations for reallocarray.
* m4/reallocarray.m4: New file.
* m4/stdlib_h.m4: Declare reallocarray.
* modules/reallocarray: New file.
* modules/reallocarray-test: New file.
* modules/stdlib: Coerce stdlib.h to export reallocarray.
* tests/test-reallocarray.c: New test.
Paul Eggert [Sat, 12 Aug 2017 18:34:50 +0000 (11:34 -0700)]
dirent-safer: fix cloexec race
* lib/opendir-safer.c: Include fcntl.h instead of unistd-safer.h.
(opendir_safer): Use F_DUPFD_CLOEXEC.
* modules/dirent-safer (Depends-on): Add fcntl. Remove unistd-safer.
* tests/test-dirent-safer.c: Do not include unistd-safer.h,
as it is no longer a prerequisite. Use F_DUPFD_CLOEXEC
instead of dup_safer.
Paul Eggert [Sat, 12 Aug 2017 18:20:41 +0000 (11:20 -0700)]
fts: fix cloexec races
* lib/fts.c [!_LIBC]: Do not include dirent--.h, unistd--.h, cloexec.h.
(opendirat, diropen): Use O_CLOEXEC instead of set_cloexec_flag.
(fts_build): Use F_DUPD_CLOEXEC rinstad of set_cloexec_flag.
(fd_ring_check): Set cloexec flag on new file descriptors.
(fts_build, fd_ring_check): While we’re at it, make sure the
resulting file descriptor is not 0, 1, or 2, since that is easy.
* modules/fts (Depends-on): Remove cloexec, dirent-safer, dup,
fcntl-safer, unistd-safer. Add fcntl.
Paul Eggert [Thu, 10 Aug 2017 13:06:06 +0000 (06:06 -0700)]
fts: port recent changes to CentOS 6
Problem reported by Tom G. Christensen in:
http://lists.gnu.org/archive/html/bug-gnulib/2017-08/msg00064.html
* lib/fts.c (fsword): New type.
(struct dev_type, filesystem_type): Use it.
* m4/fts.m4 (gl_FUNC_FTS_CORE): Check for __fsword_t.
Also, check for f_type only if fstatfs and sys/vfs.h work.