1 2006-03-23 Matthias Clasen <mclasen@redhat.com>
3 Retroactively branch for 2.10 before the last commit.
5 * glib/gtimer.c: Change to "Since 2.12".
7 2006-03-23 Emmanuele Bassi <ebassi@cvs.gnome.org>
9 * configure.in: Check for timegm.
14 * docs/reference/glib/glib-sections.txt: Added g_time_val_to_iso8601
15 and g_time_val_from_iso8601, to convert a GTimeVal to and from an
16 ISO 8601 encoded date.
18 * tests/testglib.c: Added test cases for g_time_val_to_iso8601()
19 and g_time_val_from_iso8601() functions.
21 2006-03-20 Vladimer Sichinava <vlsichinava@gmail.com>
23 * configure.in: Added "ka" (Georgian) to ALL_LINGUAS
25 2006-03-20 Matthias Clasen <mclasen@redhat.com>
27 * glib/gmem.c (profiler_log): use standard_calloc to allocate
28 the profile_data. (#335209, Chris Wilson)
30 * glib/gmain.c (g_main_context_unref): Avoid a deadlock.
31 (#335207, Chris Wilson)
33 Minor optimizations (#335216, Chris Wilson):
35 * glib/gasyncqueue.c (g_async_queue_pop_intern_unlocked): Use
36 g_queue_peek_tail_link instead of g_queue_peek_tail.
39 * glib/gslist.c: Avoid some memset calls.
41 2006-03-19 Matthias Clasen <mclasen@redhat.com>
43 * MAINTAINERS: Add this, at the request of the GNOME sysadmin team.
45 2006-03-15 Matthias Clasen <mclasen@redhat.com>
47 * glib/goption.c (g_option_context_parse): Only set the prgname
48 if it hasn't been set before. (#334611, Chong Kai Xiong)
50 2006-03-14 Matthias Clasen <mclasen@redhat.com>
52 * glib/gutils.c (g_parse_debug_string): Don't read past the
53 end of the string. (#334471, Morten Welinder)
55 * tests/testglib.c (test_g_parse_debug_string): Add testss
56 for g_parse_debug_string.
58 * glib/goption.c (parse_short_option): Don't create the
59 option_name twice. (#334519, Chris Wilson)
61 2006-03-13 Anders Carlsson <andersca@luminoth.local>
63 * configure.in: Revert fix for #322476, it breaks module loading
64 since libtool on darwin makes shared modules use .so and shared
65 libraries use .dylib. The fix breaks shared module loading everywhere
68 2006-03-02 Marcus Brinkmann <mb@g10code.de>
70 Implement watches for GIOChannels for write file descriptors on
73 * glib/giowin32.c (GIOWin32Channel): Add a new direction field.
74 (read_thread): Initialize direction.
75 (write_thread): New function.
76 (buffer_write): New function.
77 (g_io_win32_prepare): Handle the G_IO_WIN32_FILE_DESC case for the
79 (g_io_win32_fd_write): Call buffer_write() if there is a writer
81 (g_io_win32_fd_close): Set space_avail_event for writer threads.
82 (g_io_win32_fd_create_watch): Create the writer thread if
83 condition is G_IO_OUT.
84 (g_io_channel_win32_make_pollfd): Likewise here.
86 2006-03-09 Matthias Clasen <mclasen@redhat.com>
88 * Makefile.am: Add ChangeLog.pre-2.8 to EXTRA_DIST.
90 2006-03-08 Tor Lillqvist <tml@novell.com>
92 * glibconfig.h.win32.in: Add G_GUINT64_CONSTANT.
94 2006-03-07 Matthias Clasen <mclasen@redhat.com>
96 * configure.in: Bump version
98 * === Released 2.10.1 ===
102 * glib/gutf8.c (_g_charset_get_aliases): Match the prototype
103 in gconvert.c to fix build problems in NetBSD. (#333651,
104 Julio M. Merino Vidal)
106 2006-03-06 Anders Carlsson <andersca@imendio.com>
109 Make .dylib the shared library suffix on darwin.
110 (#322476, Vladimir Panov)
112 2006-03-06 Anders Carlsson <andersca@imendio.com>
115 Skip checking thread flags on Darwin. Fix suggested by Manish Singh.
116 (#314794, Gregor Riepl)
118 2006-03-01 Tor Lillqvist <tml@novell.com>
120 * glib/gutils.c (g_listenv): Separate implementation on Win32: Use
121 the wide character API on NT-based Windows. Return UTF-8 strings.
123 * glib/glib.symbols: Don't mark g_listenv as PRIVATE, as that
124 meant it wasn't present in the import library. PRIVATE is used
125 only for the backwards-compatibility DLL ABI stability hacks.
127 2006-02-24 Matthias Clasen <mclasen@redhat.com>
129 * configure.in: Bump version
131 * === Released 2.10.0 ===
133 * README.in: Mention the slice allocator and GInitiallyUnowned.
137 * configure.in: Bump version to 2.10.0
139 * tests/utf8-pointer.c: Add tests for g_utf8_strlen().
141 * glib/gutf8.c: Fix boundary cases in g_utf8_strlen().
144 2006-02-23 Kjartan Maraas <kmaraas@gnome.org>
146 * tests/completion-test.c: (main): Plug leaks reported by
148 * tests/convert-test.c: (test_iconv_state), (test_one_half):
150 * tests/patterntest.c: (test_compilation): Same
151 * tests/shell-test.c: (do_argv_test): Same.
152 * tests/unicode-caseconv.c: (main): Same.
153 * tests/uri-test.c: (run_to_uri_tests): Same.
156 2006-02-19 Martyn Russell <martyn@imendio.com>
158 * tests/asyncqueue-test.c:
160 * tests/slist-test.c: Updated to test _sort, _sort_with_data,
161 _insert_sorted and _insert_sorted_with_data API.
163 2006-02-18 Matthias Clasen <mclasen@redhat.com>
165 * tests/gobject/Makefile.am: Add paramspec-test
167 * tests/gobject/paramspec-test.c: Some GParamSpec tests.
169 * tests/gobject/gvalue-test.c: Add more tests.
171 2006-02-17 Kang Jeong-Hee <keizie@gmail.com>
173 * glib/gutf8.c (g_utf8_find_prev_char): Correct documentation typo.
175 2006-02-15 Sebastian Wilhelmi <seppi@seppi.de>
177 * glib/gthreadpool.c: Fix deadlock when signalling the thread
178 which freed a thread pool (#331110, Chris Wilson).
180 Tue Feb 14 17:00:43 2006 Tim Janik <timj@imendio.com>
182 * glib/gslice.c: only define _XOPEN_SOURCE to 600 to get at
183 posix_memalign() in case we actually are going to use it, because
184 we detected a compliant implementation (#328997).
186 * configure.in (enable_included_printf): don't include malloc.h when
187 testing for posix_memalign() funcitonality, since this may break the
188 test on some systems (#328997).
190 2006-02-14 Matthias Clasen <mclasen@redhat.com>
192 * glib/gstrfuncs.c: Improve docs.
194 2006-02-11 Matthias Clasen <mclasen@redhat.com>
196 * configure.in: Bump version
198 2006-02-10 Matthias Clasen <mclasen@redhat.com>
200 * === Released 2.9.6 ===
204 2006-01-31 Behdad Esfahbod <behdad@gnome.org>
206 * docs/reference/glib/tmpl/relations.sgml (g_relation_insert): Specify
207 the type of varargs arguments. (bug #317679)
209 2006-01-31 Matthias Clasen <mclasen@redhat.com>
211 * autogen.sh: Touch README and INSTALL here to pacify
212 automake. (#329124, Kjartan Maraas, Tim Janik)
214 Tue Jan 31 16:45:03 2006 Tim Janik <timj@imendio.com>
216 * glib/gdataset.c: carry out all dtalist pointer accesses atomically,
217 some missing cases where pointed out by Sebastian Wilhelmi.
219 Tue Jan 31 12:33:48 2006 Tim Janik <timj@imendio.com>
221 * configure.in: generate build/win32/vs8/Makefile, to fix build breakage.
223 2006-01-30 David Schleef <ds@schleef.org>
225 * glib/grand.c: Update URL
227 2006-01-29 Sven Herzberg <herzi@gnome-de.org>
229 * docs/reference/gobject/tmpl/param_value_types.sgml: tell how to
230 create container classes which are as flexible as a GValue is
232 2006-01-27 Matthias Clasen <mclasen@redhat.com>
234 * configure.in: Bump version
236 * === Released 2.9.5 ===
238 * glib/glib.symbols: Add g_mem_gc_friendly as an exported
243 2006-01-26 Matthias Clasen <mclasen@redhat.com>
245 * glib/gmem.c (g_mem_init_nomessage): Fix C99isms. (#328705,
248 Wed Jan 25 19:16:57 2006 Tim Janik <timj@imendio.com>
250 * fixed buglets reported by Jens Granseuer in #328254.
252 * configure.in: free the memory allocated in posix_memalign() tests.
254 * glib/gslice.c: spelling fixes.
256 Wed Jan 25 16:39:18 2006 Tim Janik <timj@imendio.com>
258 * glib/gslice.c: honour g_mem_gc_friendly settings when freeing
259 slices, make sure g_mem_gc_friendly is properly initialized.
261 * gmem.[hc]: ensure g_mem_gc_friendly is initialized from G_DEBUG upon
262 the first allocation. applied some branching optimizations.
264 * docs/macros.txt: reflected --enable-gc-friendly change and
265 described ENABLE_GC_FRIENDLY_DEFAULT as well as G_DEBUG=gc-friendly.
267 * configure.in: changed --enable-gc-friendly=yes to define
268 ENABLE_GC_FRIENDLY_DEFAULT.
270 * glib/garray.c: changed ENABLE_GC_FRIENDLY macro #ifdef-s to
271 if (G_UNLIKELY (g_mem_gc_friendly)).
274 * glib/ghash.c: removed ENABLE_GC_FRIENDLY code which is now taken
275 care of by g_slice_free1().
277 * tests/slice-test.c: fixed leaks, reported by Kjartan Maraas.
279 Tue Jan 24 17:49:36 2006 Tim Janik <timj@imendio.com>
281 * glib/gslice.c: only use posix_memalign() if it's known to work,
282 revert to memalign() otherwise.
284 * configure.in: check for broken posix_memalign() implementations
287 2006-01-24 Matthias Clasen <mclasen@redhat.com>
289 * tests/unicode-encoding.c: Use UTF-16LE as target encoding
290 on all little-endian systems. (#143380, Marc Moorcroft)
292 2006-01-23 Matthias Clasen <mclasen@redhat.com>
294 * configure.in: Change the shared libary extension for hpux-ia64
295 to so. (#328253, The Written Word)
297 Mon Jan 23 17:30:33 2006 Tim Janik <timj@imendio.com>
299 * glib/gutils.c (_g_getenv_nomalloc): wiped out all the wonderfull
300 G_OS_WIN32 code i wrote ;-[) after tml told me windows has getenv()
303 Mon Jan 23 16:46:20 2006 Tim Janik <timj@imendio.com>
305 * glib/gslice.c (slice_config_init): initialize GSlice config from
306 G_SLICE environment variable. we support G_SLICE=always-malloc
307 currently, which forces all g_slice_*() allocations to use the system
311 g_parse_debug_string(): added a note about not using g_malloc() here.
312 _g_getenv_nomalloc(): getenv() variant that doesn't use g_malloc or
313 g_slice. contains only guesswork in the WIN32 branch.
315 2006-01-18 Matthias Clasen <mclasen@redhat.com>
319 * === Released 2.9.4 ===
323 2006-01-17 Matthias Clasen <mclasen@redhat.com>
325 * glib/gkeyfile.c (g_key_file_get_groups): Correct a g_new0()
326 call. (#327421, Morten Welinder)
328 2006-01-17 Sebastian Wilhelmi <seppi@seppi.de>
330 * glib/gthreadpool.c: To avoid deadlocks get rid of the settings
331 G_LOCK. Use the unused_thread_queue lock instead. Change
332 g_thread_pool_thread_proxy such that threads only wait on
333 non-exlusive pools for at most a 1/2 second. Do not reorder tasks
334 due to superfluous tasks. Global tasks wait at most for
335 max-idle-time milliseconds. Make sure, that no task is woken up
336 twice for the same event via a wakeup_serial. This fixes #324228.
338 * tests/threadpool-test.c: Adapt test accordingly. Do not pass
339 invalid NULL into the thread pools. This as well fixes #327290.
341 2006-01-16 Matthias Clasen <mclasen@redhat.com>
343 * configure.in: Bump version
345 * === Released 2.9.3 ===
349 * glib/gthread.h (g_static_mutex_get_mutex_impl_shortcut):
350 Rework the strict aliasing fix to not break C++, pointed
351 out by Murray Cumming.
353 * glib/gasyncqueue.c (g_async_queue_push_sorted_unlocked):
354 Signal waiting threads, problem noticed by Christian Kellner.
356 2006-01-16 Matthias Clasen <mclasen@redhat.com>
358 Fix bug 326747, Alberto Ruiz:
360 * glib/gconvert.c (make_valid_utf8): Change this function to
361 replace unknown characters by the Unicode replacement character
362 instead of '?', and don't append "(invalid encoding)".
363 (g_filename_display_name, g_filename_display_basename): Document
364 how to determine if the filename was in an invalid encoding.
366 2006-01-14 Matthias Clasen <mclasen@redhat.com>
368 * glib/gtree.c: Replace the simple recursive implementation
369 by a nonrecursive, threaded implementation by Maurizio
372 2006-01-12 Matthias Clasen <mclasen@redhat.com>
374 * tests/asyncqueue-test.c (main): Fix the
375 assert to not always trigger. (#326558,
378 2006-01-11 Matthias Clasen <mclasen@redhat.com>
380 * tests/convert-test.c: Don't test items_read and
381 items_written in error cases where it is not set.
383 * tests/asyncqueue-test.c: Set error to NULL.
385 2006-01-08 Matthias Clasen <mclasen@redhat.com>
387 * glib/gtree.c: Break some long lines.
389 * tests/tree-test.c: Add more tests.
391 * glib/gslice.c: Include config.h first, pointed out
394 2005-12-20 Sven Herzberg <herzi@gnome-de.org>
396 * docs/reference/gobject/tmpl/gtype.sgml: explain that
397 G_TYPE_INSTANCE_GET_CLASS() does behave different during
400 2006-01-06 Matthias Clasen <mclasen@redhat.com>
402 * glib/gslice.c: Only define _XOPEN_SOURCE if we know
403 that we have posix_memalign(). (#323937, Bogdan Nicula)
405 2006-01-05 Matthias Clasen <mclasen@redhat.com>
407 * configure.in: Bump version
409 * === Released 2.9.2 ===
411 2006-01-05 Hans Breuer <hans@breuer.org>
413 * glib/makefile.msc.in : added gslice.obj
415 2006-01-05 Matthias Clasen <mclasen@redhat.com>
419 * glib/gmain.c: Clarify the documentation of source ids
420 a bit. (#325874, Dan Williams)
422 * configure.in: Fix another strict aliasing problem.
424 * glib/gthreadpool.c: Avoid use of varargs macro. (#325864,
427 2006-01-04 Manish Singh <yosh@gimp.org>
430 * glib/gstdio.h: don't macroized g_access, g_chdir, and g_unlink
431 either, since they have the same issue as g_rmdir. (Related to
434 Wed Jan 4 13:33:25 2006 Tim Janik <timj@gtk.org>
436 * glib/gslice.c (magazine_cache_pop_magazine): don't reverse chunk
437 order when creating magazines, so we hand out chunks with ascending
440 2006-01-04 Abel Cheung <maddog@linuxhall.org>
442 * configure.in: Added "ml" "zh_HK" to ALL_LINGUAS.
444 2006-01-03 Martyn Russell <martyn@imendio.com>
446 * docs/reference/glib/glib-sections.txt:
448 * glib/gthreadpool.[ch]:
449 - Added new API g_thread_pool_get_idle_time() and
450 g_thread_pool_set_idle_time(). (#324228).
452 * tests/threadpool-test.c:
453 - Updated test case to do thread pool sorting, thread pool with
454 no sorting and a thread pool with idle thread timeouts.
456 2006-01-03 Matthias Clasen <mclasen@redhat.com>
458 * glib/gmain.h: Add new functions here, too.
460 * glib/glib.symbols: Add new functions.
462 2005-12-20 Michael Meeks <michael.meeks@novell.com>
464 * glib/gmain.c (g_main_context_is_owner): new method
465 to determine if the current thread is the owner of the
468 2006-01-02 Matthias Clasen <mclasen@redhat.com>
472 * glib/gstdio.c (g_rmdir): Don't provide g_rmdir() as a macro
473 expanding to rmdir, since rmdir is not declared in a portable
474 system header we can include in gstdio.h. (#325249, Jani Monoses)
476 2006-01-01 Tor Lillqvist <tml@novell.com>
478 * glib/gspawn-win32.c (g_spawn_sync_utf8): Set the GIOChannels for
479 stdout and stderr to unbuffered. Otherwise the giochannel layer
480 will try to read from them regardless whether the
481 g_io_channel_win32_poll() call here has indicated
482 readability or not. (#325310)
484 2005-12-29 Matthias Clasen <mclasen@redhat.com>
486 * glib/gutils.c (glib_check_version): Fix a copy-and-paste error
487 in the docs. (#325273, Declan Naughton)
489 2005-12-27 Manish Singh <yosh@gimp.org>
491 * tests/run-collate-tests.sh: use LC_ALL instead of LC_COLLATE, to
492 make sure we really override things.
494 2005-12-27 Matthias Clasen <mclasen@redhat.com>
496 Fix #316221, Michal Benes, Stanislav Brabec;
498 * configure.in: Fix a strict aliasing problem in
499 g_static_mutex_get_mutex().
500 * glib/gthread.h: ...and in
501 g_static_mutex_get_mutex_impl_shortcut().
503 * glib/gdatasetprivate.h: Add a cast to silence compiler
504 warnings. (#321978, Andrew Paprocki)
506 Partial fix for bug #323937, Bogdan Nicula.
508 * configure.in: Check for malloc.h
509 * glib/gslice.c: Don't include malloc.h unconditionally.
511 2005-12-27 Manish Singh <yosh@gimp.org>
513 * tests/run-collate-tests.sh: export LC_COLLATE so it takes.
514 (#324950, Dan Yefimov)
516 2005-12-26 Matthias Clasen <mclasen@redhat.com>
518 * glib/gslice.c: On Win32, include process.h (#325015, Kazuki
521 2005-12-25 Matthias Clasen <mclasen@redhat.com>
523 * AUTHORS: Update my email
525 * tests/slice-test.c: Fix C99isms. (#324950, Dan Yefimov)
527 2005-12-21 Matthias Clasen <mclasen@redhat.com>
529 * README.in: Add some notes on when 'make check' may fail.
531 2005-12-20 Matthias Clasen <mclasen@redhat.com>
534 * glib/gthreadpool.h:
535 * glib/gthreadpool.c (g_thread_pool_set_sort_function): New function
536 to sort tasks pushed into a threadpool. (#324479, Martyn Russell)
538 * tests/threadpool-test.c: Test this.
540 Tue Dec 20 18:14:14 2005 Tim Janik <timj@imendio.com>
542 * glib/gslice.[hc]: added mem_error() and mem_assert() to test and
543 handle errors without depending on gmessage.c which might not be
544 setup when the error occours.
545 removed G_SLICE_CONFIG_ALWAYS_FREE config option, fixed the code so
546 always freeing can be achieved by adjusting the working set time to
547 0 with G_SLICE_CONFIG_WORKING_SET_MSECS.
548 added G_SLICE_CONFIG_COLOR_INCREMENT to test different color increments
549 (mainly 0 and 1). reduced the minimum block size to 128 bytes, to
550 minimize wastage if small amounts of differently sized structrues are
551 allocated, this does come at a performance cost of roughly 5% though.
552 fixed up block alignment calculation, so it works for varying
553 block sizes. only use strerror() not g_strerror() since the latter
554 depends on working GQuark and GSlice.
555 mem_error(): implemented in terms of fprintf and vfprintf.
557 * tests/slice-color.c: new program to test cache colorization effects.
559 * tests/slice-test.c: trade G_SLICE_CONFIG_ALWAYS_FREE for 0 duration
560 G_SLICE_CONFIG_WORKING_SET_MSECS.
562 2005-12-17 Matthias Clasen <mclasen@redhat.com>
564 * glib/goption.c (parse_short_option): Set an error in all
565 failure cases. (#324332, Tim-Philipp Müller)
567 2005-12-17 Sebastian Wilhelmi <seppi@seppi.de>
569 * glib/gatomic.c: Fix memory barrier position in g_atomic_int_get
570 and g_atomic_pointer_get. Add g_atomic_int_set and
571 g_atomic_pointer_set implementations for the !DEFINE_WITH_MUTEXES &&
572 G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case, as well as defining them
573 as functions (additionally to the macros in the header) for the
574 !G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case.
576 2005-12-16 Matthias Clasen <mclasen@redhat.com>
578 * glib/gmem.c (g_allocator_new): Don't return a pointer to
579 a const struct, since apps expect to be able to modify it.
580 (#324179, J. Ali Harlow)
582 Tue Dec 13 10:13:32 2005 Tim Janik <timj@imendio.com>
584 * glib/gatomic.h: added g_atomic_pointer_set() and g_atomic_int_set()
585 for systems where the initialization of atomic variables requires a
586 write memory barrier.
588 2005-12-09 Matthias Clasen <mclasen@redhat.com>
590 * configure.in: Bump version
592 * === Released 2.9.1 ===
594 2005-12-09 Alexander Larsson <alexl@redhat.com>
598 Use G_GNUC_WARN_UNUSED_RESULT on list functions that return
601 2005-12-08 Matthias Clasen <mclasen@redhat.com>
605 2005-12-07 Martyn Russell <martyn@imendio.com>
607 * glib/gasyncqueue.c:
608 - Call g_queue_insert_sorted() instead of duplicating the code.
609 - Call g_queue_sort() instead of duplicating the code.
610 - Invert sort function results to make sure the same sort function
611 gives the same results across glist, gslist, gqueue and
614 * tests/asyncqueue-test.c:
615 - Updated the sort function to reflect the example in the
616 documentation for gasyncqueue.c.
618 2005-12-07 Martyn Russell <martyn@imendio.com>
620 * docs/reference/glib/glib-sections.txt:
621 * docs/reference/glib/tmpl/linked_lists_double.sgml:
622 * docs/reference/glib/tmpl/linked_lists_single.sgml:
625 - Added g_list_insert_sorted_with_data () and
626 g_slist_insert_sorted_with_data ().
627 - Removed the extra check in g_list_sort() and g_slist_sort() for
628 GCompareDataFunc vs. GCompareFunc.
630 2005-12-07 Tor Lillqvist <tml@novell.com>
632 * glib-gettextize.in: Look up prefix at run-time on Win32,
633 assuming the standard directory structure with glib-gettextize in
636 2005-12-06 Matthias Clasen <mclasen@redhat.com>
638 * glib/gmem.h: Exempt GTK+ from the mem chunk deprecation, since
639 we need GTK+ 2.8 to compile against GLib 2.10.
641 2005-12-05 Manish Singh <yosh@gimp.org>
643 * tests/run-collate-test.sh: set LC_COLLATE, not LANG, to be sure
644 to override any user settings.
646 2005-12-05 Matthias Clasen <mclasen@redhat.com>
648 * configure.in: Revert an accidential commit
650 2005-12-05 Matthias Clasen <mclasen@redhat.com>
652 * glib/gutf8.c: Documentation updates. (#323291, Morten Welinder)
654 * tests/Makefile.am (TESTS_ENVIRONMENT):
655 * tests/gobject/Makefile.am (TESTS_ENVIRONMENT): Set
656 MALLOC_CHECK_ and MALLOC_PERTURB_
658 * tests/run-collate-tests.sh: Run the collation tests explicitly
659 in en_US locale. (#320463)
661 * glib/gmem.h: Really deprecate GMemChunk.
668 * glib/gstrfuncs.c: Add versioned deprecation docs.
670 2005-12-05 Martyn Russell <martyn@imendio.com>
672 * docs/reference/glib/glib-sections.txt:
673 * glib/gasyncqueue.[ch]:
674 - Added support for sorting async queues by with _push_sorted(),
675 _push_sorted_unlocked(), _sort() and _sort_unlocked() (#323047).
678 * tests/asyncqueue-test.c:
679 - Added test case for gasyncqueue.c
681 Mon Dec 5 15:53:20 2005 Tim Janik <timj@imendio.com>
683 * glib/gslice.c: implement chain walking for arbitrary ->next pointer
684 offsets in g_slice_free_chain_with_offset() based on a patch by behdad
685 in bug 323178. moved time consuming logic from g_slice_free() out of
686 the inner loop, so g_slice_free_chain_with_offset() provides a real
687 performance benefit over g_slice_free1() now.
689 * glib/gslice.h: renamed g_slice_free_chain() to
690 g_slice_free_chain_with_offset(). implemented g_slice_free_chain() as
691 a type-safe macro as suggested in bug 323178.
692 simplified the macro implementation of g_slice_free() and implemented
693 it in a type safe manner for all compliers as suggested by Morten
694 Welinder <mortenw@gnome.org>.
699 * glib/glib.symbols: s/g_slice_free_chain/g_slice_free_chain_with_offset/
701 2005-12-05 Matthias Clasen <mclasen@redhat.com>
703 * glib/gasyncqueue.c: Add some docs.
705 * tests/libmoduletestplugin_a.c: Fix compiler warnings.
707 * glib/gatomic.c: In the ia64 implementation, use
708 __sync builtin without _si or _di suffix. (#321229,
709 Stanislav Brabec, patch by Andreas Schwab)
711 2005-12-04 Behdad Esfahbod <behdad@gnome.org>
713 * glib/gslice.h: Remove comma at the end of enum.
715 2005-12-04 Matthias Clasen <mclasen@redhat.com>
717 Handle multiple user names with the same UID better.
718 (#319535, Laszlo Peter)
720 * glib/gutils.c (g_get_any_init_do): When determining user
721 data, first look up $LOGNAME. If the UID doesn't match
722 getuid(), fall back to the current behaviour of looking
723 up the user data based on getuid().
725 2005-12-04 Matthias Clasen <mclasen@redhat.com>
727 * glib/gmacros.h (G_GNUC_WARN_UNUSED_RESULT): Add a macro
728 to make gcc warn if a function result is ignored. (#145466,
729 Arjan van de Ven, Alex Larsson)
731 * glib/gmem.h: Add the new attribute to g_realloc and
734 2005-12-03 Matthias Clasen <mclasen@redhat.com>
737 * glib/gcache.h: Deprecate g_cache_value_foreach. (#322956,
740 * glib/gmappedfile.c: Make mapping of empty files
743 * glib/gfileutils.c: Don't fork a new process just to
744 fix the permissions of the created temp file. (#321318,
745 Alexis S. L. Carvalho)
747 2005-12-02 Matthias Clasen <mclasen@redhat.com>
749 * README.in: Add a note about Solaris threads.
752 * configure.in: Remove support for Solaris threads.
753 (#136971, Sebastian Wilhelmi, patch by Andrew Paprocki)
755 2005-12-02 Matthias Clasen <mclasen@redhat.com>
757 * configure.in: Define G_GUINT64_CONSTANT in analogy to
758 G_GINT64_CONSTANT. (#322568, Andrew Paprocki)
760 2005-12-02 Matthias Clasen <mclasen@redhat.com>
762 * glib/gslice.c: Win32 portability fixes and C99ism removal,
763 pointed out by Kazuki Iwamoto. (#323052)
765 2005-12-02 Matthias Clasen <mclasen@redhat.com>
767 * glib/gatomic.c (g_atomic_int_add): Add a missing volatile for
768 the IA64 implementation.
770 Fri Dec 2 16:18:09 2005 Tim Janik <timj@imendio.com>
772 * glib/gslice.c: some naming and type size fixups.
774 Fri Dec 2 13:08:58 2005 Tim Janik <timj@gtk.org>
776 * glib/gslice.h (g_slice_free): reworked GCC-specific type-safe macro
777 variant into something less verbose, but digestible for gcc-3.4.
779 Fri Dec 2 10:55:07 2005 Tim Janik <timj@imendio.com>
781 * tests/slice-test.c: extended to perform the benchmarking on the old
782 memchunk code if 'O' is selected.
784 * tests/memchunks.c: new file which contains the old GLib mem chunks
785 implementation with prefix old_mem_chunk_.
787 * tests/Makefile.am: added memchunks.c
789 Fri Dec 2 00:16:59 2005 Tim Janik <timj@imendio.com>
791 * glib/gslice.c: improved gettimeofday() timer resolution by maintaining
792 acceptable load distribution of the syscall.
793 (allocator_get_magazine_threshold): reduce minimum/single-thread magazine
794 sizes to approximately page_size/2.5 to avoid excessive startup allocations.
796 2005-12-01 Matthias Clasen <mclasen@redhat.com>
798 * tests/gobject/ifacecheck.c (test_iface_base_init): Don't
799 double-free base interface members.
801 * tests/Makefile.am: Remove duplicate variable.
803 Thu Dec 1 17:32:46 2005 Tim Janik <timj@imendio.com>
805 * glib/gslice.[hc]: new slice allocator implementation.
807 * tests/slice-test.c: added random slice allocation test.
809 * glib/gthread.[hc]: removed newly added private thread mem API.
811 * glib/gthreadinit.h:
814 * glib/gmem.c: divided glib threading initialisation into three phases,
815 initialisation where private keys and messaging are not available (only
816 needed by gmem.c), initialisation without messaging but private keys
817 available (gslice.c, gmessage.c), and full fledged initialisers that
818 server the rest of glib. initialisation functions got renamed to reflect
819 the limitations of their corresponding phases.
821 * glib/gmem.c: removed memchunk code, defer allocations to
824 * glib/gmem.[hc]: removed g_slice_* skeletons.
826 * glib/glib.symbols: added g_slice_* symbols.
828 * configure.in: check for availability of posix_memalign(3), memalign(3)
831 * glib/Makefile.am: added gslice.[hc].
833 2005-12-01 Tor Lillqvist <tml@novell.com>
835 * glib/gstdio.c (g_stat): In the Win32 implementation, strip
836 trailing slash(es) for non-root folders. stat() fails if non-root
837 folders are specified with trailing slashes. It's too much hassle
838 to demand that callers strip such slashes themselves, especially
839 as it is easy to get it wrong and strip the slash of a root
841 (g_rename): On NT-based Windows, use MoveFileEx() with
842 MOVEFILE_REPLACE_EXISTING to better match Unix behaviour.
844 2005-11-28 Matthias Clasen <mclasen@redhat.com>
846 Fix G_STMT_START / G_STMT_END on Solaris. (#321972,
849 * configure.in: Check whether do { } while (0) works.
851 * glib/gmacros.h: Use do { } while (0) for G_STMT_START /
852 G_STMT_END if it works.
854 2005-11-28 Matthias Clasen <mclasen@redhat.com>
856 * glib/gthread.c (g_static_rw_lock_wait, g_static_rw_lock_signal):
857 * glib/gnode.c (g_node_depth_traverse_level):
858 * glib/gmem.c (g_allocator_new):
859 * glib/ghash.c (g_hash_table_unref, g_hash_table_destroy)
860 (g_hash_table_foreach_remove_or_steal): Silence compiler
863 2005-11-27 Matthias Clasen <mclasen@redhat.com>
865 * glib/gunicollate.c (g_utf8_collate_key_for_filename):
866 Don't read beyond len. (#322520, Christian Persch)
868 Wed Nov 23 17:34:01 2005 Tim Janik <timj@imendio.com>
870 * glib/gdataset.c: access datalist flags via atomic pointer access
871 functions, instead of acquiring the dataset lock. this is faster and
872 also matches the atomic pointer readouts in gdatalistprivate.h and
873 direct pointer modifications required by gobject.c.
875 Wed Nov 23 13:35:31 2005 Tim Janik <timj@gtk.org>
877 * glib/gdataset.c: streamlined initialization code somewhat,
878 removed GData node cache.
880 2005-11-22 Matthias Clasen <mclasen@redhat.com>
882 * glib/ghash.c (g_hash_table_ref):
883 (g_hash_table_unref): Mark these as new API.
885 Avoid double locking in g_intern_string (#322133,
888 * glib/gdataset.c (g_quark_from_string_internal): New
889 internal function which factors out common parts of
890 g_quark_from[_static]_string.
891 (g_quark_from_string, g_quark_from_static_string):
892 Use g_quark_from_string_internal.
893 (g_intern_string, g_intern_static_string): Use
894 g_quark_from_string_internal, and only take the
896 (g_quark_new): Don't store the strings shifted by -1
897 in the g_quarks array.
898 (g_quark_to_string): Adapt to the previous change.
900 Tue Nov 22 14:04:26 2005 Tim Janik <timj@imendio.com>
904 g_hash_table_new_full(): create hash tables with a ref count of 1.
905 g_hash_table_ref(): atomically ref_count+=1
906 g_hash_table_unref(): atomically ref_count-=1, destroys hash table
907 when refcount reaches 0.
908 g_hash_table_destroy(): just destroy keys and values, unref by 1.
909 g_hash_table_insert():
910 g_hash_table_replace(): assert ref_count>0.
913 * glib/gatomic.c: added 'volatile' qualifier to all atomic pointer and
916 2005-11-20 Behdad Esfahbod <behdad@gnome.org>
918 * glib/guniprop.c (g_unichar_get_mirror_char): Remove unused
919 variables. (#321984, Andrew Paprocki)
921 2005-11-18 Matthias Clasen <mclasen@redhat.com>
923 * configure.in: Bump version
925 2005-11-17 Matthias Clasen <mclasen@redhat.com>
927 * === Released 2.9.0 ===
929 * glib/glib.symbols: Updates
931 * glib/gutf8.c (g_utf8_offset_to_pointer): Add some performance
932 hints to the docs. (Paolo Borelli)
936 * glib/gthread.c (g_thread_foreach): Mark as new api.
938 * README.in: Updates.
942 * glib/gdate.c (g_date_set_time_t):
943 (g_date_set_time_val): Add functions to set a date from
944 a time_t and from a GTimeVal.
945 (g_date_set_time): Deprecate. (#314067, Roger Leigh)
948 * tests/utf8-pointer.c: Unit tests for g_utf8_pointer_to_offset
949 and g_utf8_offset_to_pointer.
951 * glib/gutf8.c (g_utf8_pointer_to_offset)
952 (g_utf8_offset_to_pointer): Handle negative offsets, and use
953 "stutter stepping" for going backwards. (#320638, Larry
958 * glib/gthread.c: const correctness fixes, found
959 by Arjan van de Ven and gcc.
961 2005-11-16 Behdad Esfahbod <behdad@gnome.org>
963 * docs/reference/glib/tmpl/unicode.sgml: Correct typo on mentioning
964 @G_UNICODE_BREAK_UNKNOWN that should be #G_UNICODE_BREAK_UNKNOWN.
966 2005-11-09 Behdad Esfahbod <behdad@gnome.org>
968 * glib/guniprop.c: Use bit hacks instead when checking a general
969 category value against multiple values.
971 * glib/gutf8.c: Change ISO10646 to Unicode in docs.
973 2005-11-10 Simos Xenitellis <simos@gnome.org>
975 * configure.in: Added tt (Tatar) to ALL_LINGUAS.
977 2005-11-05 Tor Lillqvist <tml@novell.com>
979 * glib/giowin32.c (g_io_win32_set_flags): Don't call g_warning().
980 (#320688) Instead set error and return failure.
982 2005-11-07 Matthias Clasen <mclasen@redhat.com>
984 * glib/gmain.c (g_main_dispatch): Don't call cb_funcs->unref
985 while holding the context lock. (#320886, Andy Wingo)
987 2005-11-07 Billy Biggs <vektor@dumbterm.net>
989 * tests/refcount/Makefile.am: Add a missing reference to libglib
990 in the LDADD for this test.
992 2005-11-07 Matthias Clasen <mclasen@redhat.com>
994 * glib/gmem.c (g_allocator_new): We need to set n_preallocs to a
995 nonzero value, otherwise GTK+ 2.8 breaks when compiled against
996 GLib 2.9. (#320755, Luca Ferretti)
998 2005-11-04 Matthias Clasen <mclasen@redhat.com>
1000 * glib/Makefile.am: Apply a patch from Behdad Esfahbod to
1001 use a faster lookup table for g_unichar_get_mirror_char().
1002 * glib/gmirroringtable.h: The new table.
1003 * glib/gunichartables.h: Remove bidi_mirroring_table.
1004 * glib/gen-unicode-tables.pl: Don't generate the mirroring
1006 * glib/glib-mirroring-tab/*: Sources for the program
1007 which generated gmirroringtable.h.
1009 * glib/glist.c: Avoid some code duplication.
1011 * glib/gscanner.h: Include gdataset.h. (#320322)
1013 * glib/gdebug.h: add new GDebugFlag for fatal_criticals
1014 * glib/gmessages.c: (_g_debug_init): handle G_DEBUG=fatal_criticals,
1015 to help find critical warnings in applications. (#320017,
1018 2005-11-02 Tor Lillqvist <tml@novell.com>
1020 * glib/glib.symbols: Remove large amount of trailing whitespace
1021 from one line. Remove #ifdef G_UNDEFINED from around
1022 g_hash_table_new and g_hash_table_new_full symbols, what was the
1023 purpose of that? Add the g_slice functions.
1025 Tue Nov 1 16:24:20 2005 Tim Janik <timj@imendio.com>
1027 * glib/gmem.[hc]: prepared deprecation of GMemChunk and GAllocator.
1028 added g_slice_*() API to allocate and cache small bits of memory.
1029 an actuall allocator implementation for g_slice_*() is still pending.
1031 * glib/gthread.[hc]: changes from a patch by Matthias Clasen.
1032 changed GRealThread list to use in-structure *next; fields instead
1033 of GSList, in order for thread iteration to not depenend on g_slice_*()
1035 _g_thread_mem_private_get():
1036 _g_thread_mem_private_set(): added accessors for private memory,
1037 needed because the ordinary GPrivate implementation relies on GArray
1038 and GSList and therefore indirectly on working g_slice_*() allocations.
1040 * glib/gthread.[hc]:
1041 g_thread_foreach(): new public API function to loop over all existing threads.
1055 removed GAllocator and free list usages and accompanying locks.
1056 use g_slice_*() API to allocate and cache small bits of memory.
1058 * glib/ghook.h: removed GMemChunk field from public API.
1061 * glib/glist.h: deprecate allocator API, provide _free1() for consistency.
1063 * glib/gnode.h: deprecate allocator API.
1065 * glib/gmain.c: reordered GPollRec fields so g_slice_free_chain() can
1066 be used for poll rec lists.
1068 * glib/grel.c: removed mem chunk usage, and allocated tuples via g_slice_*().
1069 g_relation_destroy(): free all tuples from the all_tuples hash table,
1070 this effectively maintains the life time track keeping of tuples.
1071 g_relation_delete_tuple(): free tuples which are removed from the
1072 all_tuples hash table. this fixes a temporary leak that was present
1073 in the memchunk code until the destruction of the relation.
1075 2005-10-29 Matthias Clasen <mclasen@redhat.com>
1077 * tests/convert-test.c: Add some tests for conversions between
1078 UTF-8, UCS-4 and UTF-16.
1080 * glib/gutf8.c (g_utf8_to_ucs4, g_utf8_to_utf16): Fix handling
1081 of len == -1, noticed by Morten Welinder.
1083 2005-10-27 Erdal Ronahi <erdal.ronahi@gmail.com>
1085 * configure.in: Added ku (Kurdish) to ALL_LINGUAS
1087 2005-10-26 Matthias Clasen <mclasen@redhat.com>
1089 * glib/gutf8.c (g_ucs4_to_utf8): Don't set items_read twice
1090 in the error case, and add some documentation. (#319806, Morten
1093 2005-10-19 Manish Singh <yosh@gimp.org>
1095 * configure.in: Use AC_CHECK_FUNCS for _NSGetEnviron, to get the
1096 config.h symbol automatically. Fixes bug #313731.
1098 2005-10-19 Tor Lillqvist <tml@novell.com>
1100 * glib/gwin32.c (g_win32_get_package_installation_directory):
1101 Return a g_strdup()ed copy of the value stored in the hash table,
1102 so that it can be g_free()d without leaving a dangling pointer in
1103 the hash table. (#319232)
1105 2005-10-06 Matthias Clasen <mclasen@redhat.com>
1107 * glib/gunicollate.c (g_utf8_collate_key_for_filename): Handle
1108 all-zero sequences correctly. (#317930, Sebastien Bacher)
1110 2005-10-05 Matthias Clasen <mclasen@redhat.com>
1112 * glib/gmarkup.c (g_markup_escape_text): Doc addition
1114 * Makefile.am: Add Collation tests.
1116 * tests/collate/*: Inputs and expected outputs for collation tests.
1118 * tests/run-collate-tests.sh: Script to run collation tests.
1120 * tests/unicode-collate.c (main): Rework slightly to make
1121 it usable in unit tests. Also test g_utf8_collate_key_for_filename().
1123 2005-10-01 Behdad Esfahbod <behdad@gnome.org>
1125 * docs/reference/glib/tmpl/unicode.sgml:
1126 * glib/gen-unicode-tables.pl:
1128 * glib/gunichartables.h:
1130 * tests/casefold.txt:
1131 * tests/casemap.txt: Updated to Unicode 4.1. There are five new
1132 GUnicodeBreakType types. That may break some applications, like
1135 2005-09-26 Matthias Clasen <mclasen@redhat.com>
1137 * glib/gstrfuncs.c (g_ascii_strtoull): Add details to the
1138 docs. (#314393, Matthew F. Barnes)
1140 * glib/glib.symbols:
1141 * glib/gprintf.h: Remove g_snprintf() and g_vsnprintf(), since
1142 they are already declared in glib.h. This doesn't break documented
1143 use of gprintf.h, but should probably be pointed out in the
1144 release notes for 2.10. (#314232, Behdad Esfahbod)
1146 Tue Sep 20 13:16:04 2005 Tim Janik <timj@imendio.com>
1148 * glib/gpattern.c (g_pattern_ph_match): applied significant recursion
1149 complexity optimization, based on a patch from Matthias Clasen.
1151 * tests/patterntest.c: more tests, mostly from matthias.
1153 2005-09-20 Matthias Clasen <mclasen@redhat.com>
1155 * glib/gqueue.c (g_queue_insert_sorted): Correct the docs.
1156 (#316703, Mark Drago)
1158 Mon Sep 19 17:23:23 2005 Tim Janik <timj@imendio.com>
1160 * glib/gpattern.c: applied a patch from matthias which checks on the
1161 upper bound of GPatternSpec length to optimize matches.
1164 * tests/patterntest.c: added more match cases.
1166 2005-09-16 Tor Lillqvist <tml@novell.com>
1168 * glib/gstrfuncs.c (g_ascii_strcasecmp, g_ascii_strncasecmp): Add
1169 warning to doc comment that these functions should not be used on
1170 encodings like CP932.
1172 2005-09-14 Matthias Clasen <mclasen@redhat.com>
1174 * tests/keyfile-test.c: Add a test for grup names of length 1.
1176 * glib/gkeyfile.c (g_key_file_line_is_group): Accept group names
1177 of length 1. (#316309)
1179 2005-09-12 Matthias Clasen <mclasen@redhat.com>
1181 * glib/gmarkup.c (g_markup_escape_text): Clarify docs.
1183 2005-09-11 Sebastian Wilhelmi <seppi@seppi.de>
1185 * tests/refcount/Makefile.am (INCLUDES): Link the the refcount
1186 tests to the system thread library $(G_THREAD_LIBS). Fixes #313744
1189 2005-09-11 Kjartan Maraas <kmaraas@gnome.org>
1191 * glib/gmain.c: (g_child_watch_prepare), (g_child_watch_check),
1192 (child_watch_helper_thread): Remove some dead code. Closes
1195 2005-09-07 Tor Lillqvist <tml@novell.com>
1197 * glib/Makefile.am: Create also a console version of the
1198 gspawn-win32-helper program, gspawn-win32-helper-console.exe.
1199 It's otherwise identical to gspawn-win32-helper.exe, except marked
1200 as a console application (linked without the -mwindows option).
1202 * glib/gspawn-win32.c (do_spawn_directly, do_spawn_with_pipes):
1203 Drop the dont_wait parameter. Its truth value correlated 100% with
1204 the NULLness of the exit_status parameter anyway, so it's enough
1205 to check whether exit_status is NULL. Invert the sense of the
1206 dont_return_handle parameter and rename it to do_return_handle, to
1207 make the code easier to read by avoiding double negations.
1209 (g_spawn_sync_utf8, g_spawn_async_with_pipes_utf8): Modify calls
1210 to do_spawn_with_pipes() accordingly.
1212 (do_spawn_with_pipes): If we have a console, use the console
1213 version of the helper program, otherwise use the GUI one. This
1214 avoids extra console windows opening up in some situations. (In
1215 case a console application uses the GUI gspawn-win32-helper.exe to
1216 spawn another console application we would get a separate console
1217 for the spawned console application).
1219 * glib-zip.in: Distribute also gspawn-win32-helper-console.exe.
1221 2005-09-05 Matthias Clasen <mclasen@redhat.com>
1223 * glib/gmappedfile.c (g_mapped_file_new): Report an error
1224 if the file is too large. (#315275, Kjartan Maraas)
1226 * glib/gkeyfile.c (g_key_file_load_from_fd): The return value
1227 of read() is signed. (#315273, Kjartan Maraas)
1229 2005-08-31 Tor Lillqvist <tml@novell.com>
1231 * glib/gutils.h: Wrapping atexit() is a bad idea on Windows, where
1232 the EXE and each DLL have their own atexit function chains.
1234 #define g_atexit as atexit instead. This means it has a
1235 better chance of doing what the caller wants. For instance,
1236 gtkhtml calls g_atexit() registering a function in gtkhtml
1237 itself. This caused a crash when g_atexit() was implemented as a
1238 function in the GLib DLL. The gtkhtml DLL was already unloaded by
1239 the time the GLib DLL got unloaded.
1241 * glib/gutils.c: #undef the #define mentioned above, to also get a
1242 real g_atexit() into the DLL for backward compatibility. Document
1243 the Windows behaviour of g_atexit(), and document the varying ways
1244 atexit() can behave in the context of dynamically loaded modules
1247 2005-08-31 Matthias Clasen <mclasen@redhat.com>
1249 * glib/glib.symbols:
1251 * glib/gdataset.c: Add string interning functions.
1253 2005-08-28 Matthias Clasen <mclasen@redhat.com>
1255 * glib/giochannel.c: Unify some near-duplicate strings. (#314654,
1258 2005-08-26 Matthias Clasen <mclasen@redhat.com>
1260 * configure.in: Bump version to 2.9.0
1262 2005-08-25 Tor Lillqvist <tml@novell.com>
1264 Make also the g_spawn*() functions take parameters in the GLib
1265 file name encoding, i.e. UTF-8, on Windows. Has no impact on Unix
1266 API or ABI. Like the other GLib API that was earlier changed to
1267 use UTF-8 on Windows, the names of the functions that take UTF-8
1268 have _utf8 suffixes added by using preprocessor macros in the
1269 header file. The old names are kept for functions with the old
1270 behaviour, taking parameters in the system codepage, for DLL ABI
1273 * glib/gspawn.h: On Win32 add the suffix _utf8 to the names of the
1274 g_spawn*() functions.
1276 * glib/gspawn-win32.c: Use wide-char API on NT-based
1277 Windows. Convert parameters from UTF-8 to wide chars (NT) or
1278 system codepage (Win9x) and call the C library _wspawn*() or
1279 spawn*() functions respectvely. Add DLL ABI stability versions
1280 that take parameters in the system codepage.
1282 * glib/gspawn-win32-helper.c: On NT-based Windows use the
1283 wide-char versions of argv and envp, and use wide-char API to
1284 change directory and spawn the program to run. Remove the verbose
1285 debugging output, it was too complex to modify for the wide-char
1286 features. (Just add temporary debugging printouts if needed, no
1287 need to have them permanently in the source.)
1289 * glib/gspawn.c: Corresponding documentation updates.
1291 * glib/glib.symbols: Corresponding changes: Mark the ABI stability
1292 symbols as PRIVATE, add the new _utf8-suffixed ones.
1294 2005-08-24 Stepan Kasal <kasal@ucw.cz>
1296 * glib/gtypes.h (G_MININT64): Cast the constant to gint64; it is
1297 guint64 otherwise and that can produce warnings about comparison
1298 between signed and unsigned.
1300 2005-08-23 Matthias Clasen <mclasen@redhat.com>
1302 * glib/gutils.c: Fix the crt_externs.h include.
1304 2005-08-23 Stepan Kasal <kasal@ucw.cz>
1306 * NEWS: Fix spelling of my first name.
1308 2005-08-23 Matthias Clasen <mclasen@redhat.com>
1312 * === Released 2.8.1 ===
1316 2005-08-20 Hans Breuer <hans@breuer.org>
1318 * glib/makefile.msc.in : link with ws2_32.lib
1320 2005-08-18 Tor Lillqvist <tml@novell.com>
1322 * configure.in: Check for <sys/wait.h>
1324 * glib/gbacktrace.c: Include <sys/wait.h> on if HAVE_SYS_WAIT_H.
1326 2005-08-18 Ross Burton <ross@burtonini.com>
1329 Optimise single-character insertions.
1334 * tests/string-test.c:
1335 Add tests for new optimisation, and fix a leak.
1337 2005-08-17 Matthias Clasen <mclasen@redhat.com>
1339 * configure.in: Check for crt_externs.h and _NSGetEnviron.
1341 * glib/gutils.c: On Darwin, include crt-externs.h and
1342 define environ using _NSGetEnviron(). (#313731)
1344 2005-08-16 Stepan Kasal <kasal@ucw.cz>
1346 * glib/gutils.c (g_get_any_init): Move the body of the big if...
1347 (g_get_any_init_do): ... to this new function.
1348 (g_get_any_init): Declare as inline.
1349 (g_get_any_init_locked): New inline function, does the locking.
1350 Make use of these two throughout the code.
1352 2005-08-15 Matthias Clasen <mclasen@redhat.com>
1354 * glib/gbacktrace.c (g_on_error_stack_trace): Wait for
1355 the child process and then simply return. This makes
1356 The "S" option work as documented in g_on_error_query().
1357 (#313125, Matthew F. Barnes)
1359 * glib/gunicode.h: Update the link to Unicode category
1360 values. (#313369, Behnam Esfahbod)
1362 * glib/gqueue.c (g_queue_find_custom): Clarify docs
1363 a little. (#311727, Tristan van Berkom)
1365 * glib/abicheck.sh, gobject/abicheck.sh: Make the
1366 check work on ia64 too, where some symbols ended up
1367 in yet another section.