1 2006-03-19 Matthias Clasen <mclasen@redhat.com>
3 * MAINTAINERS: Add this, at the request of the GNOME sysadmin team.
5 2006-03-15 Matthias Clasen <mclasen@redhat.com>
7 * glib/goption.c (g_option_context_parse): Only set the prgname
8 if it hasn't been set before. (#334611, Chong Kai Xiong)
10 2006-03-14 Matthias Clasen <mclasen@redhat.com>
12 * glib/gutils.c (g_parse_debug_string): Don't read past the
13 end of the string. (#334471, Morten Welinder)
15 * tests/testglib.c (test_g_parse_debug_string): Add testss
16 for g_parse_debug_string.
18 * glib/goption.c (parse_short_option): Don't create the
19 option_name twice. (#334519, Chris Wilson)
21 2006-03-13 Anders Carlsson <andersca@luminoth.local>
23 * configure.in: Revert fix for #322476, it breaks module loading
24 since libtool on darwin makes shared modules use .so and shared
25 libraries use .dylib. The fix breaks shared module loading everywhere
28 2006-03-02 Marcus Brinkmann <mb@g10code.de>
30 Implement watches for GIOChannels for write file descriptors on
33 * glib/giowin32.c (GIOWin32Channel): Add a new direction field.
34 (read_thread): Initialize direction.
35 (write_thread): New function.
36 (buffer_write): New function.
37 (g_io_win32_prepare): Handle the G_IO_WIN32_FILE_DESC case for the
39 (g_io_win32_fd_write): Call buffer_write() if there is a writer
41 (g_io_win32_fd_close): Set space_avail_event for writer threads.
42 (g_io_win32_fd_create_watch): Create the writer thread if
43 condition is G_IO_OUT.
44 (g_io_channel_win32_make_pollfd): Likewise here.
46 2006-03-09 Matthias Clasen <mclasen@redhat.com>
48 * Makefile.am: Add ChangeLog.pre-2.8 to EXTRA_DIST.
50 2006-03-08 Tor Lillqvist <tml@novell.com>
52 * glibconfig.h.win32.in: Add G_GUINT64_CONSTANT.
54 2006-03-07 Matthias Clasen <mclasen@redhat.com>
56 * configure.in: Bump version
58 * === Released 2.10.1 ===
62 * glib/gutf8.c (_g_charset_get_aliases): Match the prototype
63 in gconvert.c to fix build problems in NetBSD. (#333651,
64 Julio M. Merino Vidal)
66 2006-03-06 Anders Carlsson <andersca@imendio.com>
69 Make .dylib the shared library suffix on darwin.
70 (#322476, Vladimir Panov)
72 2006-03-06 Anders Carlsson <andersca@imendio.com>
75 Skip checking thread flags on Darwin. Fix suggested by Manish Singh.
76 (#314794, Gregor Riepl)
78 2006-03-01 Tor Lillqvist <tml@novell.com>
80 * glib/gutils.c (g_listenv): Separate implementation on Win32: Use
81 the wide character API on NT-based Windows. Return UTF-8 strings.
83 * glib/glib.symbols: Don't mark g_listenv as PRIVATE, as that
84 meant it wasn't present in the import library. PRIVATE is used
85 only for the backwards-compatibility DLL ABI stability hacks.
87 2006-02-24 Matthias Clasen <mclasen@redhat.com>
89 * configure.in: Bump version
91 * === Released 2.10.0 ===
93 * README.in: Mention the slice allocator and GInitiallyUnowned.
97 * configure.in: Bump version to 2.10.0
99 * tests/utf8-pointer.c: Add tests for g_utf8_strlen().
101 * glib/gutf8.c: Fix boundary cases in g_utf8_strlen().
104 2006-02-23 Kjartan Maraas <kmaraas@gnome.org>
106 * tests/completion-test.c: (main): Plug leaks reported by
108 * tests/convert-test.c: (test_iconv_state), (test_one_half):
110 * tests/patterntest.c: (test_compilation): Same
111 * tests/shell-test.c: (do_argv_test): Same.
112 * tests/unicode-caseconv.c: (main): Same.
113 * tests/uri-test.c: (run_to_uri_tests): Same.
116 2006-02-19 Martyn Russell <martyn@imendio.com>
118 * tests/asyncqueue-test.c:
120 * tests/slist-test.c: Updated to test _sort, _sort_with_data,
121 _insert_sorted and _insert_sorted_with_data API.
123 2006-02-18 Matthias Clasen <mclasen@redhat.com>
125 * tests/gobject/Makefile.am: Add paramspec-test
127 * tests/gobject/paramspec-test.c: Some GParamSpec tests.
129 * tests/gobject/gvalue-test.c: Add more tests.
131 2006-02-17 Kang Jeong-Hee <keizie@gmail.com>
133 * glib/gutf8.c (g_utf8_find_prev_char): Correct documentation typo.
135 2006-02-15 Sebastian Wilhelmi <seppi@seppi.de>
137 * glib/gthreadpool.c: Fix deadlock when signalling the thread
138 which freed a thread pool (#331110, Chris Wilson).
140 Tue Feb 14 17:00:43 2006 Tim Janik <timj@imendio.com>
142 * glib/gslice.c: only define _XOPEN_SOURCE to 600 to get at
143 posix_memalign() in case we actually are going to use it, because
144 we detected a compliant implementation (#328997).
146 * configure.in (enable_included_printf): don't include malloc.h when
147 testing for posix_memalign() funcitonality, since this may break the
148 test on some systems (#328997).
150 2006-02-14 Matthias Clasen <mclasen@redhat.com>
152 * glib/gstrfuncs.c: Improve docs.
154 2006-02-11 Matthias Clasen <mclasen@redhat.com>
156 * configure.in: Bump version
158 2006-02-10 Matthias Clasen <mclasen@redhat.com>
160 * === Released 2.9.6 ===
164 2006-01-31 Behdad Esfahbod <behdad@gnome.org>
166 * docs/reference/glib/tmpl/relations.sgml (g_relation_insert): Specify
167 the type of varargs arguments. (bug #317679)
169 2006-01-31 Matthias Clasen <mclasen@redhat.com>
171 * autogen.sh: Touch README and INSTALL here to pacify
172 automake. (#329124, Kjartan Maraas, Tim Janik)
174 Tue Jan 31 16:45:03 2006 Tim Janik <timj@imendio.com>
176 * glib/gdataset.c: carry out all dtalist pointer accesses atomically,
177 some missing cases where pointed out by Sebastian Wilhelmi.
179 Tue Jan 31 12:33:48 2006 Tim Janik <timj@imendio.com>
181 * configure.in: generate build/win32/vs8/Makefile, to fix build breakage.
183 2006-01-30 David Schleef <ds@schleef.org>
185 * glib/grand.c: Update URL
187 2006-01-29 Sven Herzberg <herzi@gnome-de.org>
189 * docs/reference/gobject/tmpl/param_value_types.sgml: tell how to
190 create container classes which are as flexible as a GValue is
192 2006-01-27 Matthias Clasen <mclasen@redhat.com>
194 * configure.in: Bump version
196 * === Released 2.9.5 ===
198 * glib/glib.symbols: Add g_mem_gc_friendly as an exported
203 2006-01-26 Matthias Clasen <mclasen@redhat.com>
205 * glib/gmem.c (g_mem_init_nomessage): Fix C99isms. (#328705,
208 Wed Jan 25 19:16:57 2006 Tim Janik <timj@imendio.com>
210 * fixed buglets reported by Jens Granseuer in #328254.
212 * configure.in: free the memory allocated in posix_memalign() tests.
214 * glib/gslice.c: spelling fixes.
216 Wed Jan 25 16:39:18 2006 Tim Janik <timj@imendio.com>
218 * glib/gslice.c: honour g_mem_gc_friendly settings when freeing
219 slices, make sure g_mem_gc_friendly is properly initialized.
221 * gmem.[hc]: ensure g_mem_gc_friendly is initialized from G_DEBUG upon
222 the first allocation. applied some branching optimizations.
224 * docs/macros.txt: reflected --enable-gc-friendly change and
225 described ENABLE_GC_FRIENDLY_DEFAULT as well as G_DEBUG=gc-friendly.
227 * configure.in: changed --enable-gc-friendly=yes to define
228 ENABLE_GC_FRIENDLY_DEFAULT.
230 * glib/garray.c: changed ENABLE_GC_FRIENDLY macro #ifdef-s to
231 if (G_UNLIKELY (g_mem_gc_friendly)).
234 * glib/ghash.c: removed ENABLE_GC_FRIENDLY code which is now taken
235 care of by g_slice_free1().
237 * tests/slice-test.c: fixed leaks, reported by Kjartan Maraas.
239 Tue Jan 24 17:49:36 2006 Tim Janik <timj@imendio.com>
241 * glib/gslice.c: only use posix_memalign() if it's known to work,
242 revert to memalign() otherwise.
244 * configure.in: check for broken posix_memalign() implementations
247 2006-01-24 Matthias Clasen <mclasen@redhat.com>
249 * tests/unicode-encoding.c: Use UTF-16LE as target encoding
250 on all little-endian systems. (#143380, Marc Moorcroft)
252 2006-01-23 Matthias Clasen <mclasen@redhat.com>
254 * configure.in: Change the shared libary extension for hpux-ia64
255 to so. (#328253, The Written Word)
257 Mon Jan 23 17:30:33 2006 Tim Janik <timj@imendio.com>
259 * glib/gutils.c (_g_getenv_nomalloc): wiped out all the wonderfull
260 G_OS_WIN32 code i wrote ;-[) after tml told me windows has getenv()
263 Mon Jan 23 16:46:20 2006 Tim Janik <timj@imendio.com>
265 * glib/gslice.c (slice_config_init): initialize GSlice config from
266 G_SLICE environment variable. we support G_SLICE=always-malloc
267 currently, which forces all g_slice_*() allocations to use the system
271 g_parse_debug_string(): added a note about not using g_malloc() here.
272 _g_getenv_nomalloc(): getenv() variant that doesn't use g_malloc or
273 g_slice. contains only guesswork in the WIN32 branch.
275 2006-01-18 Matthias Clasen <mclasen@redhat.com>
279 * === Released 2.9.4 ===
283 2006-01-17 Matthias Clasen <mclasen@redhat.com>
285 * glib/gkeyfile.c (g_key_file_get_groups): Correct a g_new0()
286 call. (#327421, Morten Welinder)
288 2006-01-17 Sebastian Wilhelmi <seppi@seppi.de>
290 * glib/gthreadpool.c: To avoid deadlocks get rid of the settings
291 G_LOCK. Use the unused_thread_queue lock instead. Change
292 g_thread_pool_thread_proxy such that threads only wait on
293 non-exlusive pools for at most a 1/2 second. Do not reorder tasks
294 due to superfluous tasks. Global tasks wait at most for
295 max-idle-time milliseconds. Make sure, that no task is woken up
296 twice for the same event via a wakeup_serial. This fixes #324228.
298 * tests/threadpool-test.c: Adapt test accordingly. Do not pass
299 invalid NULL into the thread pools. This as well fixes #327290.
301 2006-01-16 Matthias Clasen <mclasen@redhat.com>
303 * configure.in: Bump version
305 * === Released 2.9.3 ===
309 * glib/gthread.h (g_static_mutex_get_mutex_impl_shortcut):
310 Rework the strict aliasing fix to not break C++, pointed
311 out by Murray Cumming.
313 * glib/gasyncqueue.c (g_async_queue_push_sorted_unlocked):
314 Signal waiting threads, problem noticed by Christian Kellner.
316 2006-01-16 Matthias Clasen <mclasen@redhat.com>
318 Fix bug 326747, Alberto Ruiz:
320 * glib/gconvert.c (make_valid_utf8): Change this function to
321 replace unknown characters by the Unicode replacement character
322 instead of '?', and don't append "(invalid encoding)".
323 (g_filename_display_name, g_filename_display_basename): Document
324 how to determine if the filename was in an invalid encoding.
326 2006-01-14 Matthias Clasen <mclasen@redhat.com>
328 * glib/gtree.c: Replace the simple recursive implementation
329 by a nonrecursive, threaded implementation by Maurizio
332 2006-01-12 Matthias Clasen <mclasen@redhat.com>
334 * tests/asyncqueue-test.c (main): Fix the
335 assert to not always trigger. (#326558,
338 2006-01-11 Matthias Clasen <mclasen@redhat.com>
340 * tests/convert-test.c: Don't test items_read and
341 items_written in error cases where it is not set.
343 * tests/asyncqueue-test.c: Set error to NULL.
345 2006-01-08 Matthias Clasen <mclasen@redhat.com>
347 * glib/gtree.c: Break some long lines.
349 * tests/tree-test.c: Add more tests.
351 * glib/gslice.c: Include config.h first, pointed out
354 2005-12-20 Sven Herzberg <herzi@gnome-de.org>
356 * docs/reference/gobject/tmpl/gtype.sgml: explain that
357 G_TYPE_INSTANCE_GET_CLASS() does behave different during
360 2006-01-06 Matthias Clasen <mclasen@redhat.com>
362 * glib/gslice.c: Only define _XOPEN_SOURCE if we know
363 that we have posix_memalign(). (#323937, Bogdan Nicula)
365 2006-01-05 Matthias Clasen <mclasen@redhat.com>
367 * configure.in: Bump version
369 * === Released 2.9.2 ===
371 2006-01-05 Hans Breuer <hans@breuer.org>
373 * glib/makefile.msc.in : added gslice.obj
375 2006-01-05 Matthias Clasen <mclasen@redhat.com>
379 * glib/gmain.c: Clarify the documentation of source ids
380 a bit. (#325874, Dan Williams)
382 * configure.in: Fix another strict aliasing problem.
384 * glib/gthreadpool.c: Avoid use of varargs macro. (#325864,
387 2006-01-04 Manish Singh <yosh@gimp.org>
390 * glib/gstdio.h: don't macroized g_access, g_chdir, and g_unlink
391 either, since they have the same issue as g_rmdir. (Related to
394 Wed Jan 4 13:33:25 2006 Tim Janik <timj@gtk.org>
396 * glib/gslice.c (magazine_cache_pop_magazine): don't reverse chunk
397 order when creating magazines, so we hand out chunks with ascending
400 2006-01-04 Abel Cheung <maddog@linuxhall.org>
402 * configure.in: Added "ml" "zh_HK" to ALL_LINGUAS.
404 2006-01-03 Martyn Russell <martyn@imendio.com>
406 * docs/reference/glib/glib-sections.txt:
408 * glib/gthreadpool.[ch]:
409 - Added new API g_thread_pool_get_idle_time() and
410 g_thread_pool_set_idle_time(). (#324228).
412 * tests/threadpool-test.c:
413 - Updated test case to do thread pool sorting, thread pool with
414 no sorting and a thread pool with idle thread timeouts.
416 2006-01-03 Matthias Clasen <mclasen@redhat.com>
418 * glib/gmain.h: Add new functions here, too.
420 * glib/glib.symbols: Add new functions.
422 2005-12-20 Michael Meeks <michael.meeks@novell.com>
424 * glib/gmain.c (g_main_context_is_owner): new method
425 to determine if the current thread is the owner of the
428 2006-01-02 Matthias Clasen <mclasen@redhat.com>
432 * glib/gstdio.c (g_rmdir): Don't provide g_rmdir() as a macro
433 expanding to rmdir, since rmdir is not declared in a portable
434 system header we can include in gstdio.h. (#325249, Jani Monoses)
436 2006-01-01 Tor Lillqvist <tml@novell.com>
438 * glib/gspawn-win32.c (g_spawn_sync_utf8): Set the GIOChannels for
439 stdout and stderr to unbuffered. Otherwise the giochannel layer
440 will try to read from them regardless whether the
441 g_io_channel_win32_poll() call here has indicated
442 readability or not. (#325310)
444 2005-12-29 Matthias Clasen <mclasen@redhat.com>
446 * glib/gutils.c (glib_check_version): Fix a copy-and-paste error
447 in the docs. (#325273, Declan Naughton)
449 2005-12-27 Manish Singh <yosh@gimp.org>
451 * tests/run-collate-tests.sh: use LC_ALL instead of LC_COLLATE, to
452 make sure we really override things.
454 2005-12-27 Matthias Clasen <mclasen@redhat.com>
456 Fix #316221, Michal Benes, Stanislav Brabec;
458 * configure.in: Fix a strict aliasing problem in
459 g_static_mutex_get_mutex().
460 * glib/gthread.h: ...and in
461 g_static_mutex_get_mutex_impl_shortcut().
463 * glib/gdatasetprivate.h: Add a cast to silence compiler
464 warnings. (#321978, Andrew Paprocki)
466 Partial fix for bug #323937, Bogdan Nicula.
468 * configure.in: Check for malloc.h
469 * glib/gslice.c: Don't include malloc.h unconditionally.
471 2005-12-27 Manish Singh <yosh@gimp.org>
473 * tests/run-collate-tests.sh: export LC_COLLATE so it takes.
474 (#324950, Dan Yefimov)
476 2005-12-26 Matthias Clasen <mclasen@redhat.com>
478 * glib/gslice.c: On Win32, include process.h (#325015, Kazuki
481 2005-12-25 Matthias Clasen <mclasen@redhat.com>
483 * AUTHORS: Update my email
485 * tests/slice-test.c: Fix C99isms. (#324950, Dan Yefimov)
487 2005-12-21 Matthias Clasen <mclasen@redhat.com>
489 * README.in: Add some notes on when 'make check' may fail.
491 2005-12-20 Matthias Clasen <mclasen@redhat.com>
494 * glib/gthreadpool.h:
495 * glib/gthreadpool.c (g_thread_pool_set_sort_function): New function
496 to sort tasks pushed into a threadpool. (#324479, Martyn Russell)
498 * tests/threadpool-test.c: Test this.
500 Tue Dec 20 18:14:14 2005 Tim Janik <timj@imendio.com>
502 * glib/gslice.[hc]: added mem_error() and mem_assert() to test and
503 handle errors without depending on gmessage.c which might not be
504 setup when the error occours.
505 removed G_SLICE_CONFIG_ALWAYS_FREE config option, fixed the code so
506 always freeing can be achieved by adjusting the working set time to
507 0 with G_SLICE_CONFIG_WORKING_SET_MSECS.
508 added G_SLICE_CONFIG_COLOR_INCREMENT to test different color increments
509 (mainly 0 and 1). reduced the minimum block size to 128 bytes, to
510 minimize wastage if small amounts of differently sized structrues are
511 allocated, this does come at a performance cost of roughly 5% though.
512 fixed up block alignment calculation, so it works for varying
513 block sizes. only use strerror() not g_strerror() since the latter
514 depends on working GQuark and GSlice.
515 mem_error(): implemented in terms of fprintf and vfprintf.
517 * tests/slice-color.c: new program to test cache colorization effects.
519 * tests/slice-test.c: trade G_SLICE_CONFIG_ALWAYS_FREE for 0 duration
520 G_SLICE_CONFIG_WORKING_SET_MSECS.
522 2005-12-17 Matthias Clasen <mclasen@redhat.com>
524 * glib/goption.c (parse_short_option): Set an error in all
525 failure cases. (#324332, Tim-Philipp Müller)
527 2005-12-17 Sebastian Wilhelmi <seppi@seppi.de>
529 * glib/gatomic.c: Fix memory barrier position in g_atomic_int_get
530 and g_atomic_pointer_get. Add g_atomic_int_set and
531 g_atomic_pointer_set implementations for the !DEFINE_WITH_MUTEXES &&
532 G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case, as well as defining them
533 as functions (additionally to the macros in the header) for the
534 !G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case.
536 2005-12-16 Matthias Clasen <mclasen@redhat.com>
538 * glib/gmem.c (g_allocator_new): Don't return a pointer to
539 a const struct, since apps expect to be able to modify it.
540 (#324179, J. Ali Harlow)
542 Tue Dec 13 10:13:32 2005 Tim Janik <timj@imendio.com>
544 * glib/gatomic.h: added g_atomic_pointer_set() and g_atomic_int_set()
545 for systems where the initialization of atomic variables requires a
546 write memory barrier.
548 2005-12-09 Matthias Clasen <mclasen@redhat.com>
550 * configure.in: Bump version
552 * === Released 2.9.1 ===
554 2005-12-09 Alexander Larsson <alexl@redhat.com>
558 Use G_GNUC_WARN_UNUSED_RESULT on list functions that return
561 2005-12-08 Matthias Clasen <mclasen@redhat.com>
565 2005-12-07 Martyn Russell <martyn@imendio.com>
567 * glib/gasyncqueue.c:
568 - Call g_queue_insert_sorted() instead of duplicating the code.
569 - Call g_queue_sort() instead of duplicating the code.
570 - Invert sort function results to make sure the same sort function
571 gives the same results across glist, gslist, gqueue and
574 * tests/asyncqueue-test.c:
575 - Updated the sort function to reflect the example in the
576 documentation for gasyncqueue.c.
578 2005-12-07 Martyn Russell <martyn@imendio.com>
580 * docs/reference/glib/glib-sections.txt:
581 * docs/reference/glib/tmpl/linked_lists_double.sgml:
582 * docs/reference/glib/tmpl/linked_lists_single.sgml:
585 - Added g_list_insert_sorted_with_data () and
586 g_slist_insert_sorted_with_data ().
587 - Removed the extra check in g_list_sort() and g_slist_sort() for
588 GCompareDataFunc vs. GCompareFunc.
590 2005-12-07 Tor Lillqvist <tml@novell.com>
592 * glib-gettextize.in: Look up prefix at run-time on Win32,
593 assuming the standard directory structure with glib-gettextize in
596 2005-12-06 Matthias Clasen <mclasen@redhat.com>
598 * glib/gmem.h: Exempt GTK+ from the mem chunk deprecation, since
599 we need GTK+ 2.8 to compile against GLib 2.10.
601 2005-12-05 Manish Singh <yosh@gimp.org>
603 * tests/run-collate-test.sh: set LC_COLLATE, not LANG, to be sure
604 to override any user settings.
606 2005-12-05 Matthias Clasen <mclasen@redhat.com>
608 * configure.in: Revert an accidential commit
610 2005-12-05 Matthias Clasen <mclasen@redhat.com>
612 * glib/gutf8.c: Documentation updates. (#323291, Morten Welinder)
614 * tests/Makefile.am (TESTS_ENVIRONMENT):
615 * tests/gobject/Makefile.am (TESTS_ENVIRONMENT): Set
616 MALLOC_CHECK_ and MALLOC_PERTURB_
618 * tests/run-collate-tests.sh: Run the collation tests explicitly
619 in en_US locale. (#320463)
621 * glib/gmem.h: Really deprecate GMemChunk.
628 * glib/gstrfuncs.c: Add versioned deprecation docs.
630 2005-12-05 Martyn Russell <martyn@imendio.com>
632 * docs/reference/glib/glib-sections.txt:
633 * glib/gasyncqueue.[ch]:
634 - Added support for sorting async queues by with _push_sorted(),
635 _push_sorted_unlocked(), _sort() and _sort_unlocked() (#323047).
638 * tests/asyncqueue-test.c:
639 - Added test case for gasyncqueue.c
641 Mon Dec 5 15:53:20 2005 Tim Janik <timj@imendio.com>
643 * glib/gslice.c: implement chain walking for arbitrary ->next pointer
644 offsets in g_slice_free_chain_with_offset() based on a patch by behdad
645 in bug 323178. moved time consuming logic from g_slice_free() out of
646 the inner loop, so g_slice_free_chain_with_offset() provides a real
647 performance benefit over g_slice_free1() now.
649 * glib/gslice.h: renamed g_slice_free_chain() to
650 g_slice_free_chain_with_offset(). implemented g_slice_free_chain() as
651 a type-safe macro as suggested in bug 323178.
652 simplified the macro implementation of g_slice_free() and implemented
653 it in a type safe manner for all compliers as suggested by Morten
654 Welinder <mortenw@gnome.org>.
659 * glib/glib.symbols: s/g_slice_free_chain/g_slice_free_chain_with_offset/
661 2005-12-05 Matthias Clasen <mclasen@redhat.com>
663 * glib/gasyncqueue.c: Add some docs.
665 * tests/libmoduletestplugin_a.c: Fix compiler warnings.
667 * glib/gatomic.c: In the ia64 implementation, use
668 __sync builtin without _si or _di suffix. (#321229,
669 Stanislav Brabec, patch by Andreas Schwab)
671 2005-12-04 Behdad Esfahbod <behdad@gnome.org>
673 * glib/gslice.h: Remove comma at the end of enum.
675 2005-12-04 Matthias Clasen <mclasen@redhat.com>
677 Handle multiple user names with the same UID better.
678 (#319535, Laszlo Peter)
680 * glib/gutils.c (g_get_any_init_do): When determining user
681 data, first look up $LOGNAME. If the UID doesn't match
682 getuid(), fall back to the current behaviour of looking
683 up the user data based on getuid().
685 2005-12-04 Matthias Clasen <mclasen@redhat.com>
687 * glib/gmacros.h (G_GNUC_WARN_UNUSED_RESULT): Add a macro
688 to make gcc warn if a function result is ignored. (#145466,
689 Arjan van de Ven, Alex Larsson)
691 * glib/gmem.h: Add the new attribute to g_realloc and
694 2005-12-03 Matthias Clasen <mclasen@redhat.com>
697 * glib/gcache.h: Deprecate g_cache_value_foreach. (#322956,
700 * glib/gmappedfile.c: Make mapping of empty files
703 * glib/gfileutils.c: Don't fork a new process just to
704 fix the permissions of the created temp file. (#321318,
705 Alexis S. L. Carvalho)
707 2005-12-02 Matthias Clasen <mclasen@redhat.com>
709 * README.in: Add a note about Solaris threads.
712 * configure.in: Remove support for Solaris threads.
713 (#136971, Sebastian Wilhelmi, patch by Andrew Paprocki)
715 2005-12-02 Matthias Clasen <mclasen@redhat.com>
717 * configure.in: Define G_GUINT64_CONSTANT in analogy to
718 G_GINT64_CONSTANT. (#322568, Andrew Paprocki)
720 2005-12-02 Matthias Clasen <mclasen@redhat.com>
722 * glib/gslice.c: Win32 portability fixes and C99ism removal,
723 pointed out by Kazuki Iwamoto. (#323052)
725 2005-12-02 Matthias Clasen <mclasen@redhat.com>
727 * glib/gatomic.c (g_atomic_int_add): Add a missing volatile for
728 the IA64 implementation.
730 Fri Dec 2 16:18:09 2005 Tim Janik <timj@imendio.com>
732 * glib/gslice.c: some naming and type size fixups.
734 Fri Dec 2 13:08:58 2005 Tim Janik <timj@gtk.org>
736 * glib/gslice.h (g_slice_free): reworked GCC-specific type-safe macro
737 variant into something less verbose, but digestible for gcc-3.4.
739 Fri Dec 2 10:55:07 2005 Tim Janik <timj@imendio.com>
741 * tests/slice-test.c: extended to perform the benchmarking on the old
742 memchunk code if 'O' is selected.
744 * tests/memchunks.c: new file which contains the old GLib mem chunks
745 implementation with prefix old_mem_chunk_.
747 * tests/Makefile.am: added memchunks.c
749 Fri Dec 2 00:16:59 2005 Tim Janik <timj@imendio.com>
751 * glib/gslice.c: improved gettimeofday() timer resolution by maintaining
752 acceptable load distribution of the syscall.
753 (allocator_get_magazine_threshold): reduce minimum/single-thread magazine
754 sizes to approximately page_size/2.5 to avoid excessive startup allocations.
756 2005-12-01 Matthias Clasen <mclasen@redhat.com>
758 * tests/gobject/ifacecheck.c (test_iface_base_init): Don't
759 double-free base interface members.
761 * tests/Makefile.am: Remove duplicate variable.
763 Thu Dec 1 17:32:46 2005 Tim Janik <timj@imendio.com>
765 * glib/gslice.[hc]: new slice allocator implementation.
767 * tests/slice-test.c: added random slice allocation test.
769 * glib/gthread.[hc]: removed newly added private thread mem API.
771 * glib/gthreadinit.h:
774 * glib/gmem.c: divided glib threading initialisation into three phases,
775 initialisation where private keys and messaging are not available (only
776 needed by gmem.c), initialisation without messaging but private keys
777 available (gslice.c, gmessage.c), and full fledged initialisers that
778 server the rest of glib. initialisation functions got renamed to reflect
779 the limitations of their corresponding phases.
781 * glib/gmem.c: removed memchunk code, defer allocations to
784 * glib/gmem.[hc]: removed g_slice_* skeletons.
786 * glib/glib.symbols: added g_slice_* symbols.
788 * configure.in: check for availability of posix_memalign(3), memalign(3)
791 * glib/Makefile.am: added gslice.[hc].
793 2005-12-01 Tor Lillqvist <tml@novell.com>
795 * glib/gstdio.c (g_stat): In the Win32 implementation, strip
796 trailing slash(es) for non-root folders. stat() fails if non-root
797 folders are specified with trailing slashes. It's too much hassle
798 to demand that callers strip such slashes themselves, especially
799 as it is easy to get it wrong and strip the slash of a root
801 (g_rename): On NT-based Windows, use MoveFileEx() with
802 MOVEFILE_REPLACE_EXISTING to better match Unix behaviour.
804 2005-11-28 Matthias Clasen <mclasen@redhat.com>
806 Fix G_STMT_START / G_STMT_END on Solaris. (#321972,
809 * configure.in: Check whether do { } while (0) works.
811 * glib/gmacros.h: Use do { } while (0) for G_STMT_START /
812 G_STMT_END if it works.
814 2005-11-28 Matthias Clasen <mclasen@redhat.com>
816 * glib/gthread.c (g_static_rw_lock_wait, g_static_rw_lock_signal):
817 * glib/gnode.c (g_node_depth_traverse_level):
818 * glib/gmem.c (g_allocator_new):
819 * glib/ghash.c (g_hash_table_unref, g_hash_table_destroy)
820 (g_hash_table_foreach_remove_or_steal): Silence compiler
823 2005-11-27 Matthias Clasen <mclasen@redhat.com>
825 * glib/gunicollate.c (g_utf8_collate_key_for_filename):
826 Don't read beyond len. (#322520, Christian Persch)
828 Wed Nov 23 17:34:01 2005 Tim Janik <timj@imendio.com>
830 * glib/gdataset.c: access datalist flags via atomic pointer access
831 functions, instead of acquiring the dataset lock. this is faster and
832 also matches the atomic pointer readouts in gdatalistprivate.h and
833 direct pointer modifications required by gobject.c.
835 Wed Nov 23 13:35:31 2005 Tim Janik <timj@gtk.org>
837 * glib/gdataset.c: streamlined initialization code somewhat,
838 removed GData node cache.
840 2005-11-22 Matthias Clasen <mclasen@redhat.com>
842 * glib/ghash.c (g_hash_table_ref):
843 (g_hash_table_unref): Mark these as new API.
845 Avoid double locking in g_intern_string (#322133,
848 * glib/gdataset.c (g_quark_from_string_internal): New
849 internal function which factors out common parts of
850 g_quark_from[_static]_string.
851 (g_quark_from_string, g_quark_from_static_string):
852 Use g_quark_from_string_internal.
853 (g_intern_string, g_intern_static_string): Use
854 g_quark_from_string_internal, and only take the
856 (g_quark_new): Don't store the strings shifted by -1
857 in the g_quarks array.
858 (g_quark_to_string): Adapt to the previous change.
860 Tue Nov 22 14:04:26 2005 Tim Janik <timj@imendio.com>
864 g_hash_table_new_full(): create hash tables with a ref count of 1.
865 g_hash_table_ref(): atomically ref_count+=1
866 g_hash_table_unref(): atomically ref_count-=1, destroys hash table
867 when refcount reaches 0.
868 g_hash_table_destroy(): just destroy keys and values, unref by 1.
869 g_hash_table_insert():
870 g_hash_table_replace(): assert ref_count>0.
873 * glib/gatomic.c: added 'volatile' qualifier to all atomic pointer and
876 2005-11-20 Behdad Esfahbod <behdad@gnome.org>
878 * glib/guniprop.c (g_unichar_get_mirror_char): Remove unused
879 variables. (#321984, Andrew Paprocki)
881 2005-11-18 Matthias Clasen <mclasen@redhat.com>
883 * configure.in: Bump version
885 2005-11-17 Matthias Clasen <mclasen@redhat.com>
887 * === Released 2.9.0 ===
889 * glib/glib.symbols: Updates
891 * glib/gutf8.c (g_utf8_offset_to_pointer): Add some performance
892 hints to the docs. (Paolo Borelli)
896 * glib/gthread.c (g_thread_foreach): Mark as new api.
898 * README.in: Updates.
902 * glib/gdate.c (g_date_set_time_t):
903 (g_date_set_time_val): Add functions to set a date from
904 a time_t and from a GTimeVal.
905 (g_date_set_time): Deprecate. (#314067, Roger Leigh)
908 * tests/utf8-pointer.c: Unit tests for g_utf8_pointer_to_offset
909 and g_utf8_offset_to_pointer.
911 * glib/gutf8.c (g_utf8_pointer_to_offset)
912 (g_utf8_offset_to_pointer): Handle negative offsets, and use
913 "stutter stepping" for going backwards. (#320638, Larry
918 * glib/gthread.c: const correctness fixes, found
919 by Arjan van de Ven and gcc.
921 2005-11-16 Behdad Esfahbod <behdad@gnome.org>
923 * docs/reference/glib/tmpl/unicode.sgml: Correct typo on mentioning
924 @G_UNICODE_BREAK_UNKNOWN that should be #G_UNICODE_BREAK_UNKNOWN.
926 2005-11-09 Behdad Esfahbod <behdad@gnome.org>
928 * glib/guniprop.c: Use bit hacks instead when checking a general
929 category value against multiple values.
931 * glib/gutf8.c: Change ISO10646 to Unicode in docs.
933 2005-11-10 Simos Xenitellis <simos@gnome.org>
935 * configure.in: Added tt (Tatar) to ALL_LINGUAS.
937 2005-11-05 Tor Lillqvist <tml@novell.com>
939 * glib/giowin32.c (g_io_win32_set_flags): Don't call g_warning().
940 (#320688) Instead set error and return failure.
942 2005-11-07 Matthias Clasen <mclasen@redhat.com>
944 * glib/gmain.c (g_main_dispatch): Don't call cb_funcs->unref
945 while holding the context lock. (#320886, Andy Wingo)
947 2005-11-07 Billy Biggs <vektor@dumbterm.net>
949 * tests/refcount/Makefile.am: Add a missing reference to libglib
950 in the LDADD for this test.
952 2005-11-07 Matthias Clasen <mclasen@redhat.com>
954 * glib/gmem.c (g_allocator_new): We need to set n_preallocs to a
955 nonzero value, otherwise GTK+ 2.8 breaks when compiled against
956 GLib 2.9. (#320755, Luca Ferretti)
958 2005-11-04 Matthias Clasen <mclasen@redhat.com>
960 * glib/Makefile.am: Apply a patch from Behdad Esfahbod to
961 use a faster lookup table for g_unichar_get_mirror_char().
962 * glib/gmirroringtable.h: The new table.
963 * glib/gunichartables.h: Remove bidi_mirroring_table.
964 * glib/gen-unicode-tables.pl: Don't generate the mirroring
966 * glib/glib-mirroring-tab/*: Sources for the program
967 which generated gmirroringtable.h.
969 * glib/glist.c: Avoid some code duplication.
971 * glib/gscanner.h: Include gdataset.h. (#320322)
973 * glib/gdebug.h: add new GDebugFlag for fatal_criticals
974 * glib/gmessages.c: (_g_debug_init): handle G_DEBUG=fatal_criticals,
975 to help find critical warnings in applications. (#320017,
978 2005-11-02 Tor Lillqvist <tml@novell.com>
980 * glib/glib.symbols: Remove large amount of trailing whitespace
981 from one line. Remove #ifdef G_UNDEFINED from around
982 g_hash_table_new and g_hash_table_new_full symbols, what was the
983 purpose of that? Add the g_slice functions.
985 Tue Nov 1 16:24:20 2005 Tim Janik <timj@imendio.com>
987 * glib/gmem.[hc]: prepared deprecation of GMemChunk and GAllocator.
988 added g_slice_*() API to allocate and cache small bits of memory.
989 an actuall allocator implementation for g_slice_*() is still pending.
991 * glib/gthread.[hc]: changes from a patch by Matthias Clasen.
992 changed GRealThread list to use in-structure *next; fields instead
993 of GSList, in order for thread iteration to not depenend on g_slice_*()
995 _g_thread_mem_private_get():
996 _g_thread_mem_private_set(): added accessors for private memory,
997 needed because the ordinary GPrivate implementation relies on GArray
998 and GSList and therefore indirectly on working g_slice_*() allocations.
1000 * glib/gthread.[hc]:
1001 g_thread_foreach(): new public API function to loop over all existing threads.
1015 removed GAllocator and free list usages and accompanying locks.
1016 use g_slice_*() API to allocate and cache small bits of memory.
1018 * glib/ghook.h: removed GMemChunk field from public API.
1021 * glib/glist.h: deprecate allocator API, provide _free1() for consistency.
1023 * glib/gnode.h: deprecate allocator API.
1025 * glib/gmain.c: reordered GPollRec fields so g_slice_free_chain() can
1026 be used for poll rec lists.
1028 * glib/grel.c: removed mem chunk usage, and allocated tuples via g_slice_*().
1029 g_relation_destroy(): free all tuples from the all_tuples hash table,
1030 this effectively maintains the life time track keeping of tuples.
1031 g_relation_delete_tuple(): free tuples which are removed from the
1032 all_tuples hash table. this fixes a temporary leak that was present
1033 in the memchunk code until the destruction of the relation.
1035 2005-10-29 Matthias Clasen <mclasen@redhat.com>
1037 * tests/convert-test.c: Add some tests for conversions between
1038 UTF-8, UCS-4 and UTF-16.
1040 * glib/gutf8.c (g_utf8_to_ucs4, g_utf8_to_utf16): Fix handling
1041 of len == -1, noticed by Morten Welinder.
1043 2005-10-27 Erdal Ronahi <erdal.ronahi@gmail.com>
1045 * configure.in: Added ku (Kurdish) to ALL_LINGUAS
1047 2005-10-26 Matthias Clasen <mclasen@redhat.com>
1049 * glib/gutf8.c (g_ucs4_to_utf8): Don't set items_read twice
1050 in the error case, and add some documentation. (#319806, Morten
1053 2005-10-19 Manish Singh <yosh@gimp.org>
1055 * configure.in: Use AC_CHECK_FUNCS for _NSGetEnviron, to get the
1056 config.h symbol automatically. Fixes bug #313731.
1058 2005-10-19 Tor Lillqvist <tml@novell.com>
1060 * glib/gwin32.c (g_win32_get_package_installation_directory):
1061 Return a g_strdup()ed copy of the value stored in the hash table,
1062 so that it can be g_free()d without leaving a dangling pointer in
1063 the hash table. (#319232)
1065 2005-10-06 Matthias Clasen <mclasen@redhat.com>
1067 * glib/gunicollate.c (g_utf8_collate_key_for_filename): Handle
1068 all-zero sequences correctly. (#317930, Sebastien Bacher)
1070 2005-10-05 Matthias Clasen <mclasen@redhat.com>
1072 * glib/gmarkup.c (g_markup_escape_text): Doc addition
1074 * Makefile.am: Add Collation tests.
1076 * tests/collate/*: Inputs and expected outputs for collation tests.
1078 * tests/run-collate-tests.sh: Script to run collation tests.
1080 * tests/unicode-collate.c (main): Rework slightly to make
1081 it usable in unit tests. Also test g_utf8_collate_key_for_filename().
1083 2005-10-01 Behdad Esfahbod <behdad@gnome.org>
1085 * docs/reference/glib/tmpl/unicode.sgml:
1086 * glib/gen-unicode-tables.pl:
1088 * glib/gunichartables.h:
1090 * tests/casefold.txt:
1091 * tests/casemap.txt: Updated to Unicode 4.1. There are five new
1092 GUnicodeBreakType types. That may break some applications, like
1095 2005-09-26 Matthias Clasen <mclasen@redhat.com>
1097 * glib/gstrfuncs.c (g_ascii_strtoull): Add details to the
1098 docs. (#314393, Matthew F. Barnes)
1100 * glib/glib.symbols:
1101 * glib/gprintf.h: Remove g_snprintf() and g_vsnprintf(), since
1102 they are already declared in glib.h. This doesn't break documented
1103 use of gprintf.h, but should probably be pointed out in the
1104 release notes for 2.10. (#314232, Behdad Esfahbod)
1106 Tue Sep 20 13:16:04 2005 Tim Janik <timj@imendio.com>
1108 * glib/gpattern.c (g_pattern_ph_match): applied significant recursion
1109 complexity optimization, based on a patch from Matthias Clasen.
1111 * tests/patterntest.c: more tests, mostly from matthias.
1113 2005-09-20 Matthias Clasen <mclasen@redhat.com>
1115 * glib/gqueue.c (g_queue_insert_sorted): Correct the docs.
1116 (#316703, Mark Drago)
1118 Mon Sep 19 17:23:23 2005 Tim Janik <timj@imendio.com>
1120 * glib/gpattern.c: applied a patch from matthias which checks on the
1121 upper bound of GPatternSpec length to optimize matches.
1124 * tests/patterntest.c: added more match cases.
1126 2005-09-16 Tor Lillqvist <tml@novell.com>
1128 * glib/gstrfuncs.c (g_ascii_strcasecmp, g_ascii_strncasecmp): Add
1129 warning to doc comment that these functions should not be used on
1130 encodings like CP932.
1132 2005-09-14 Matthias Clasen <mclasen@redhat.com>
1134 * tests/keyfile-test.c: Add a test for grup names of length 1.
1136 * glib/gkeyfile.c (g_key_file_line_is_group): Accept group names
1137 of length 1. (#316309)
1139 2005-09-12 Matthias Clasen <mclasen@redhat.com>
1141 * glib/gmarkup.c (g_markup_escape_text): Clarify docs.
1143 2005-09-11 Sebastian Wilhelmi <seppi@seppi.de>
1145 * tests/refcount/Makefile.am (INCLUDES): Link the the refcount
1146 tests to the system thread library $(G_THREAD_LIBS). Fixes #313744
1149 2005-09-11 Kjartan Maraas <kmaraas@gnome.org>
1151 * glib/gmain.c: (g_child_watch_prepare), (g_child_watch_check),
1152 (child_watch_helper_thread): Remove some dead code. Closes
1155 2005-09-07 Tor Lillqvist <tml@novell.com>
1157 * glib/Makefile.am: Create also a console version of the
1158 gspawn-win32-helper program, gspawn-win32-helper-console.exe.
1159 It's otherwise identical to gspawn-win32-helper.exe, except marked
1160 as a console application (linked without the -mwindows option).
1162 * glib/gspawn-win32.c (do_spawn_directly, do_spawn_with_pipes):
1163 Drop the dont_wait parameter. Its truth value correlated 100% with
1164 the NULLness of the exit_status parameter anyway, so it's enough
1165 to check whether exit_status is NULL. Invert the sense of the
1166 dont_return_handle parameter and rename it to do_return_handle, to
1167 make the code easier to read by avoiding double negations.
1169 (g_spawn_sync_utf8, g_spawn_async_with_pipes_utf8): Modify calls
1170 to do_spawn_with_pipes() accordingly.
1172 (do_spawn_with_pipes): If we have a console, use the console
1173 version of the helper program, otherwise use the GUI one. This
1174 avoids extra console windows opening up in some situations. (In
1175 case a console application uses the GUI gspawn-win32-helper.exe to
1176 spawn another console application we would get a separate console
1177 for the spawned console application).
1179 * glib-zip.in: Distribute also gspawn-win32-helper-console.exe.
1181 2005-09-05 Matthias Clasen <mclasen@redhat.com>
1183 * glib/gmappedfile.c (g_mapped_file_new): Report an error
1184 if the file is too large. (#315275, Kjartan Maraas)
1186 * glib/gkeyfile.c (g_key_file_load_from_fd): The return value
1187 of read() is signed. (#315273, Kjartan Maraas)
1189 2005-08-31 Tor Lillqvist <tml@novell.com>
1191 * glib/gutils.h: Wrapping atexit() is a bad idea on Windows, where
1192 the EXE and each DLL have their own atexit function chains.
1194 #define g_atexit as atexit instead. This means it has a
1195 better chance of doing what the caller wants. For instance,
1196 gtkhtml calls g_atexit() registering a function in gtkhtml
1197 itself. This caused a crash when g_atexit() was implemented as a
1198 function in the GLib DLL. The gtkhtml DLL was already unloaded by
1199 the time the GLib DLL got unloaded.
1201 * glib/gutils.c: #undef the #define mentioned above, to also get a
1202 real g_atexit() into the DLL for backward compatibility. Document
1203 the Windows behaviour of g_atexit(), and document the varying ways
1204 atexit() can behave in the context of dynamically loaded modules
1207 2005-08-31 Matthias Clasen <mclasen@redhat.com>
1209 * glib/glib.symbols:
1211 * glib/gdataset.c: Add string interning functions.
1213 2005-08-28 Matthias Clasen <mclasen@redhat.com>
1215 * glib/giochannel.c: Unify some near-duplicate strings. (#314654,
1218 2005-08-26 Matthias Clasen <mclasen@redhat.com>
1220 * configure.in: Bump version to 2.9.0
1222 2005-08-25 Tor Lillqvist <tml@novell.com>
1224 Make also the g_spawn*() functions take parameters in the GLib
1225 file name encoding, i.e. UTF-8, on Windows. Has no impact on Unix
1226 API or ABI. Like the other GLib API that was earlier changed to
1227 use UTF-8 on Windows, the names of the functions that take UTF-8
1228 have _utf8 suffixes added by using preprocessor macros in the
1229 header file. The old names are kept for functions with the old
1230 behaviour, taking parameters in the system codepage, for DLL ABI
1233 * glib/gspawn.h: On Win32 add the suffix _utf8 to the names of the
1234 g_spawn*() functions.
1236 * glib/gspawn-win32.c: Use wide-char API on NT-based
1237 Windows. Convert parameters from UTF-8 to wide chars (NT) or
1238 system codepage (Win9x) and call the C library _wspawn*() or
1239 spawn*() functions respectvely. Add DLL ABI stability versions
1240 that take parameters in the system codepage.
1242 * glib/gspawn-win32-helper.c: On NT-based Windows use the
1243 wide-char versions of argv and envp, and use wide-char API to
1244 change directory and spawn the program to run. Remove the verbose
1245 debugging output, it was too complex to modify for the wide-char
1246 features. (Just add temporary debugging printouts if needed, no
1247 need to have them permanently in the source.)
1249 * glib/gspawn.c: Corresponding documentation updates.
1251 * glib/glib.symbols: Corresponding changes: Mark the ABI stability
1252 symbols as PRIVATE, add the new _utf8-suffixed ones.
1254 2005-08-24 Stepan Kasal <kasal@ucw.cz>
1256 * glib/gtypes.h (G_MININT64): Cast the constant to gint64; it is
1257 guint64 otherwise and that can produce warnings about comparison
1258 between signed and unsigned.
1260 2005-08-23 Matthias Clasen <mclasen@redhat.com>
1262 * glib/gutils.c: Fix the crt_externs.h include.
1264 2005-08-23 Stepan Kasal <kasal@ucw.cz>
1266 * NEWS: Fix spelling of my first name.
1268 2005-08-23 Matthias Clasen <mclasen@redhat.com>
1272 * === Released 2.8.1 ===
1276 2005-08-20 Hans Breuer <hans@breuer.org>
1278 * glib/makefile.msc.in : link with ws2_32.lib
1280 2005-08-18 Tor Lillqvist <tml@novell.com>
1282 * configure.in: Check for <sys/wait.h>
1284 * glib/gbacktrace.c: Include <sys/wait.h> on if HAVE_SYS_WAIT_H.
1286 2005-08-18 Ross Burton <ross@burtonini.com>
1289 Optimise single-character insertions.
1294 * tests/string-test.c:
1295 Add tests for new optimisation, and fix a leak.
1297 2005-08-17 Matthias Clasen <mclasen@redhat.com>
1299 * configure.in: Check for crt_externs.h and _NSGetEnviron.
1301 * glib/gutils.c: On Darwin, include crt-externs.h and
1302 define environ using _NSGetEnviron(). (#313731)
1304 2005-08-16 Stepan Kasal <kasal@ucw.cz>
1306 * glib/gutils.c (g_get_any_init): Move the body of the big if...
1307 (g_get_any_init_do): ... to this new function.
1308 (g_get_any_init): Declare as inline.
1309 (g_get_any_init_locked): New inline function, does the locking.
1310 Make use of these two throughout the code.
1312 2005-08-15 Matthias Clasen <mclasen@redhat.com>
1314 * glib/gbacktrace.c (g_on_error_stack_trace): Wait for
1315 the child process and then simply return. This makes
1316 The "S" option work as documented in g_on_error_query().
1317 (#313125, Matthew F. Barnes)
1319 * glib/gunicode.h: Update the link to Unicode category
1320 values. (#313369, Behnam Esfahbod)
1322 * glib/gqueue.c (g_queue_find_custom): Clarify docs
1323 a little. (#311727, Tristan van Berkom)
1325 * glib/abicheck.sh, gobject/abicheck.sh: Make the
1326 check work on ia64 too, where some symbols ended up
1327 in yet another section.