1 2006-03-15 Matthias Clasen <mclasen@redhat.com>
3 * glib/goption.c (g_option_context_parse): Only set the prgname
4 if it hasn't been set before. (#334611, Chong Kai Xiong)
6 2006-03-14 Matthias Clasen <mclasen@redhat.com>
8 * glib/gutils.c (g_parse_debug_string): Don't read past the
9 end of the string. (#334471, Morten Welinder)
11 * tests/testglib.c (test_g_parse_debug_string): Add testss
12 for g_parse_debug_string.
14 * glib/goption.c (parse_short_option): Don't create the
15 option_name twice. (#334519, Chris Wilson)
17 2006-03-13 Anders Carlsson <andersca@luminoth.local>
19 * configure.in: Revert fix for #322476, it breaks module loading
20 since libtool on darwin makes shared modules use .so and shared
21 libraries use .dylib. The fix breaks shared module loading everywhere
24 2006-03-02 Marcus Brinkmann <mb@g10code.de>
26 Implement watches for GIOChannels for write file descriptors on
29 * glib/giowin32.c (GIOWin32Channel): Add a new direction field.
30 (read_thread): Initialize direction.
31 (write_thread): New function.
32 (buffer_write): New function.
33 (g_io_win32_prepare): Handle the G_IO_WIN32_FILE_DESC case for the
35 (g_io_win32_fd_write): Call buffer_write() if there is a writer
37 (g_io_win32_fd_close): Set space_avail_event for writer threads.
38 (g_io_win32_fd_create_watch): Create the writer thread if
39 condition is G_IO_OUT.
40 (g_io_channel_win32_make_pollfd): Likewise here.
42 2006-03-09 Matthias Clasen <mclasen@redhat.com>
44 * Makefile.am: Add ChangeLog.pre-2.8 to EXTRA_DIST.
46 2006-03-08 Tor Lillqvist <tml@novell.com>
48 * glibconfig.h.win32.in: Add G_GUINT64_CONSTANT.
50 2006-03-07 Matthias Clasen <mclasen@redhat.com>
52 * configure.in: Bump version
54 * === Released 2.10.1 ===
58 * glib/gutf8.c (_g_charset_get_aliases): Match the prototype
59 in gconvert.c to fix build problems in NetBSD. (#333651,
60 Julio M. Merino Vidal)
62 2006-03-06 Anders Carlsson <andersca@imendio.com>
65 Make .dylib the shared library suffix on darwin.
66 (#322476, Vladimir Panov)
68 2006-03-06 Anders Carlsson <andersca@imendio.com>
71 Skip checking thread flags on Darwin. Fix suggested by Manish Singh.
72 (#314794, Gregor Riepl)
74 2006-03-01 Tor Lillqvist <tml@novell.com>
76 * glib/gutils.c (g_listenv): Separate implementation on Win32: Use
77 the wide character API on NT-based Windows. Return UTF-8 strings.
79 * glib/glib.symbols: Don't mark g_listenv as PRIVATE, as that
80 meant it wasn't present in the import library. PRIVATE is used
81 only for the backwards-compatibility DLL ABI stability hacks.
83 2006-02-24 Matthias Clasen <mclasen@redhat.com>
85 * configure.in: Bump version
87 * === Released 2.10.0 ===
89 * README.in: Mention the slice allocator and GInitiallyUnowned.
93 * configure.in: Bump version to 2.10.0
95 * tests/utf8-pointer.c: Add tests for g_utf8_strlen().
97 * glib/gutf8.c: Fix boundary cases in g_utf8_strlen().
100 2006-02-23 Kjartan Maraas <kmaraas@gnome.org>
102 * tests/completion-test.c: (main): Plug leaks reported by
104 * tests/convert-test.c: (test_iconv_state), (test_one_half):
106 * tests/patterntest.c: (test_compilation): Same
107 * tests/shell-test.c: (do_argv_test): Same.
108 * tests/unicode-caseconv.c: (main): Same.
109 * tests/uri-test.c: (run_to_uri_tests): Same.
112 2006-02-19 Martyn Russell <martyn@imendio.com>
114 * tests/asyncqueue-test.c:
116 * tests/slist-test.c: Updated to test _sort, _sort_with_data,
117 _insert_sorted and _insert_sorted_with_data API.
119 2006-02-18 Matthias Clasen <mclasen@redhat.com>
121 * tests/gobject/Makefile.am: Add paramspec-test
123 * tests/gobject/paramspec-test.c: Some GParamSpec tests.
125 * tests/gobject/gvalue-test.c: Add more tests.
127 2006-02-17 Kang Jeong-Hee <keizie@gmail.com>
129 * glib/gutf8.c (g_utf8_find_prev_char): Correct documentation typo.
131 2006-02-15 Sebastian Wilhelmi <seppi@seppi.de>
133 * glib/gthreadpool.c: Fix deadlock when signalling the thread
134 which freed a thread pool (#331110, Chris Wilson).
136 Tue Feb 14 17:00:43 2006 Tim Janik <timj@imendio.com>
138 * glib/gslice.c: only define _XOPEN_SOURCE to 600 to get at
139 posix_memalign() in case we actually are going to use it, because
140 we detected a compliant implementation (#328997).
142 * configure.in (enable_included_printf): don't include malloc.h when
143 testing for posix_memalign() funcitonality, since this may break the
144 test on some systems (#328997).
146 2006-02-14 Matthias Clasen <mclasen@redhat.com>
148 * glib/gstrfuncs.c: Improve docs.
150 2006-02-11 Matthias Clasen <mclasen@redhat.com>
152 * configure.in: Bump version
154 2006-02-10 Matthias Clasen <mclasen@redhat.com>
156 * === Released 2.9.6 ===
160 2006-01-31 Behdad Esfahbod <behdad@gnome.org>
162 * docs/reference/glib/tmpl/relations.sgml (g_relation_insert): Specify
163 the type of varargs arguments. (bug #317679)
165 2006-01-31 Matthias Clasen <mclasen@redhat.com>
167 * autogen.sh: Touch README and INSTALL here to pacify
168 automake. (#329124, Kjartan Maraas, Tim Janik)
170 Tue Jan 31 16:45:03 2006 Tim Janik <timj@imendio.com>
172 * glib/gdataset.c: carry out all dtalist pointer accesses atomically,
173 some missing cases where pointed out by Sebastian Wilhelmi.
175 Tue Jan 31 12:33:48 2006 Tim Janik <timj@imendio.com>
177 * configure.in: generate build/win32/vs8/Makefile, to fix build breakage.
179 2006-01-30 David Schleef <ds@schleef.org>
181 * glib/grand.c: Update URL
183 2006-01-29 Sven Herzberg <herzi@gnome-de.org>
185 * docs/reference/gobject/tmpl/param_value_types.sgml: tell how to
186 create container classes which are as flexible as a GValue is
188 2006-01-27 Matthias Clasen <mclasen@redhat.com>
190 * configure.in: Bump version
192 * === Released 2.9.5 ===
194 * glib/glib.symbols: Add g_mem_gc_friendly as an exported
199 2006-01-26 Matthias Clasen <mclasen@redhat.com>
201 * glib/gmem.c (g_mem_init_nomessage): Fix C99isms. (#328705,
204 Wed Jan 25 19:16:57 2006 Tim Janik <timj@imendio.com>
206 * fixed buglets reported by Jens Granseuer in #328254.
208 * configure.in: free the memory allocated in posix_memalign() tests.
210 * glib/gslice.c: spelling fixes.
212 Wed Jan 25 16:39:18 2006 Tim Janik <timj@imendio.com>
214 * glib/gslice.c: honour g_mem_gc_friendly settings when freeing
215 slices, make sure g_mem_gc_friendly is properly initialized.
217 * gmem.[hc]: ensure g_mem_gc_friendly is initialized from G_DEBUG upon
218 the first allocation. applied some branching optimizations.
220 * docs/macros.txt: reflected --enable-gc-friendly change and
221 described ENABLE_GC_FRIENDLY_DEFAULT as well as G_DEBUG=gc-friendly.
223 * configure.in: changed --enable-gc-friendly=yes to define
224 ENABLE_GC_FRIENDLY_DEFAULT.
226 * glib/garray.c: changed ENABLE_GC_FRIENDLY macro #ifdef-s to
227 if (G_UNLIKELY (g_mem_gc_friendly)).
230 * glib/ghash.c: removed ENABLE_GC_FRIENDLY code which is now taken
231 care of by g_slice_free1().
233 * tests/slice-test.c: fixed leaks, reported by Kjartan Maraas.
235 Tue Jan 24 17:49:36 2006 Tim Janik <timj@imendio.com>
237 * glib/gslice.c: only use posix_memalign() if it's known to work,
238 revert to memalign() otherwise.
240 * configure.in: check for broken posix_memalign() implementations
243 2006-01-24 Matthias Clasen <mclasen@redhat.com>
245 * tests/unicode-encoding.c: Use UTF-16LE as target encoding
246 on all little-endian systems. (#143380, Marc Moorcroft)
248 2006-01-23 Matthias Clasen <mclasen@redhat.com>
250 * configure.in: Change the shared libary extension for hpux-ia64
251 to so. (#328253, The Written Word)
253 Mon Jan 23 17:30:33 2006 Tim Janik <timj@imendio.com>
255 * glib/gutils.c (_g_getenv_nomalloc): wiped out all the wonderfull
256 G_OS_WIN32 code i wrote ;-[) after tml told me windows has getenv()
259 Mon Jan 23 16:46:20 2006 Tim Janik <timj@imendio.com>
261 * glib/gslice.c (slice_config_init): initialize GSlice config from
262 G_SLICE environment variable. we support G_SLICE=always-malloc
263 currently, which forces all g_slice_*() allocations to use the system
267 g_parse_debug_string(): added a note about not using g_malloc() here.
268 _g_getenv_nomalloc(): getenv() variant that doesn't use g_malloc or
269 g_slice. contains only guesswork in the WIN32 branch.
271 2006-01-18 Matthias Clasen <mclasen@redhat.com>
275 * === Released 2.9.4 ===
279 2006-01-17 Matthias Clasen <mclasen@redhat.com>
281 * glib/gkeyfile.c (g_key_file_get_groups): Correct a g_new0()
282 call. (#327421, Morten Welinder)
284 2006-01-17 Sebastian Wilhelmi <seppi@seppi.de>
286 * glib/gthreadpool.c: To avoid deadlocks get rid of the settings
287 G_LOCK. Use the unused_thread_queue lock instead. Change
288 g_thread_pool_thread_proxy such that threads only wait on
289 non-exlusive pools for at most a 1/2 second. Do not reorder tasks
290 due to superfluous tasks. Global tasks wait at most for
291 max-idle-time milliseconds. Make sure, that no task is woken up
292 twice for the same event via a wakeup_serial. This fixes #324228.
294 * tests/threadpool-test.c: Adapt test accordingly. Do not pass
295 invalid NULL into the thread pools. This as well fixes #327290.
297 2006-01-16 Matthias Clasen <mclasen@redhat.com>
299 * configure.in: Bump version
301 * === Released 2.9.3 ===
305 * glib/gthread.h (g_static_mutex_get_mutex_impl_shortcut):
306 Rework the strict aliasing fix to not break C++, pointed
307 out by Murray Cumming.
309 * glib/gasyncqueue.c (g_async_queue_push_sorted_unlocked):
310 Signal waiting threads, problem noticed by Christian Kellner.
312 2006-01-16 Matthias Clasen <mclasen@redhat.com>
314 Fix bug 326747, Alberto Ruiz:
316 * glib/gconvert.c (make_valid_utf8): Change this function to
317 replace unknown characters by the Unicode replacement character
318 instead of '?', and don't append "(invalid encoding)".
319 (g_filename_display_name, g_filename_display_basename): Document
320 how to determine if the filename was in an invalid encoding.
322 2006-01-14 Matthias Clasen <mclasen@redhat.com>
324 * glib/gtree.c: Replace the simple recursive implementation
325 by a nonrecursive, threaded implementation by Maurizio
328 2006-01-12 Matthias Clasen <mclasen@redhat.com>
330 * tests/asyncqueue-test.c (main): Fix the
331 assert to not always trigger. (#326558,
334 2006-01-11 Matthias Clasen <mclasen@redhat.com>
336 * tests/convert-test.c: Don't test items_read and
337 items_written in error cases where it is not set.
339 * tests/asyncqueue-test.c: Set error to NULL.
341 2006-01-08 Matthias Clasen <mclasen@redhat.com>
343 * glib/gtree.c: Break some long lines.
345 * tests/tree-test.c: Add more tests.
347 * glib/gslice.c: Include config.h first, pointed out
350 2005-12-20 Sven Herzberg <herzi@gnome-de.org>
352 * docs/reference/gobject/tmpl/gtype.sgml: explain that
353 G_TYPE_INSTANCE_GET_CLASS() does behave different during
356 2006-01-06 Matthias Clasen <mclasen@redhat.com>
358 * glib/gslice.c: Only define _XOPEN_SOURCE if we know
359 that we have posix_memalign(). (#323937, Bogdan Nicula)
361 2006-01-05 Matthias Clasen <mclasen@redhat.com>
363 * configure.in: Bump version
365 * === Released 2.9.2 ===
367 2006-01-05 Hans Breuer <hans@breuer.org>
369 * glib/makefile.msc.in : added gslice.obj
371 2006-01-05 Matthias Clasen <mclasen@redhat.com>
375 * glib/gmain.c: Clarify the documentation of source ids
376 a bit. (#325874, Dan Williams)
378 * configure.in: Fix another strict aliasing problem.
380 * glib/gthreadpool.c: Avoid use of varargs macro. (#325864,
383 2006-01-04 Manish Singh <yosh@gimp.org>
386 * glib/gstdio.h: don't macroized g_access, g_chdir, and g_unlink
387 either, since they have the same issue as g_rmdir. (Related to
390 Wed Jan 4 13:33:25 2006 Tim Janik <timj@gtk.org>
392 * glib/gslice.c (magazine_cache_pop_magazine): don't reverse chunk
393 order when creating magazines, so we hand out chunks with ascending
396 2006-01-04 Abel Cheung <maddog@linuxhall.org>
398 * configure.in: Added "ml" "zh_HK" to ALL_LINGUAS.
400 2006-01-03 Martyn Russell <martyn@imendio.com>
402 * docs/reference/glib/glib-sections.txt:
404 * glib/gthreadpool.[ch]:
405 - Added new API g_thread_pool_get_idle_time() and
406 g_thread_pool_set_idle_time(). (#324228).
408 * tests/threadpool-test.c:
409 - Updated test case to do thread pool sorting, thread pool with
410 no sorting and a thread pool with idle thread timeouts.
412 2006-01-03 Matthias Clasen <mclasen@redhat.com>
414 * glib/gmain.h: Add new functions here, too.
416 * glib/glib.symbols: Add new functions.
418 2005-12-20 Michael Meeks <michael.meeks@novell.com>
420 * glib/gmain.c (g_main_context_is_owner): new method
421 to determine if the current thread is the owner of the
424 2006-01-02 Matthias Clasen <mclasen@redhat.com>
428 * glib/gstdio.c (g_rmdir): Don't provide g_rmdir() as a macro
429 expanding to rmdir, since rmdir is not declared in a portable
430 system header we can include in gstdio.h. (#325249, Jani Monoses)
432 2006-01-01 Tor Lillqvist <tml@novell.com>
434 * glib/gspawn-win32.c (g_spawn_sync_utf8): Set the GIOChannels for
435 stdout and stderr to unbuffered. Otherwise the giochannel layer
436 will try to read from them regardless whether the
437 g_io_channel_win32_poll() call here has indicated
438 readability or not. (#325310)
440 2005-12-29 Matthias Clasen <mclasen@redhat.com>
442 * glib/gutils.c (glib_check_version): Fix a copy-and-paste error
443 in the docs. (#325273, Declan Naughton)
445 2005-12-27 Manish Singh <yosh@gimp.org>
447 * tests/run-collate-tests.sh: use LC_ALL instead of LC_COLLATE, to
448 make sure we really override things.
450 2005-12-27 Matthias Clasen <mclasen@redhat.com>
452 Fix #316221, Michal Benes, Stanislav Brabec;
454 * configure.in: Fix a strict aliasing problem in
455 g_static_mutex_get_mutex().
456 * glib/gthread.h: ...and in
457 g_static_mutex_get_mutex_impl_shortcut().
459 * glib/gdatasetprivate.h: Add a cast to silence compiler
460 warnings. (#321978, Andrew Paprocki)
462 Partial fix for bug #323937, Bogdan Nicula.
464 * configure.in: Check for malloc.h
465 * glib/gslice.c: Don't include malloc.h unconditionally.
467 2005-12-27 Manish Singh <yosh@gimp.org>
469 * tests/run-collate-tests.sh: export LC_COLLATE so it takes.
470 (#324950, Dan Yefimov)
472 2005-12-26 Matthias Clasen <mclasen@redhat.com>
474 * glib/gslice.c: On Win32, include process.h (#325015, Kazuki
477 2005-12-25 Matthias Clasen <mclasen@redhat.com>
479 * AUTHORS: Update my email
481 * tests/slice-test.c: Fix C99isms. (#324950, Dan Yefimov)
483 2005-12-21 Matthias Clasen <mclasen@redhat.com>
485 * README.in: Add some notes on when 'make check' may fail.
487 2005-12-20 Matthias Clasen <mclasen@redhat.com>
490 * glib/gthreadpool.h:
491 * glib/gthreadpool.c (g_thread_pool_set_sort_function): New function
492 to sort tasks pushed into a threadpool. (#324479, Martyn Russell)
494 * tests/threadpool-test.c: Test this.
496 Tue Dec 20 18:14:14 2005 Tim Janik <timj@imendio.com>
498 * glib/gslice.[hc]: added mem_error() and mem_assert() to test and
499 handle errors without depending on gmessage.c which might not be
500 setup when the error occours.
501 removed G_SLICE_CONFIG_ALWAYS_FREE config option, fixed the code so
502 always freeing can be achieved by adjusting the working set time to
503 0 with G_SLICE_CONFIG_WORKING_SET_MSECS.
504 added G_SLICE_CONFIG_COLOR_INCREMENT to test different color increments
505 (mainly 0 and 1). reduced the minimum block size to 128 bytes, to
506 minimize wastage if small amounts of differently sized structrues are
507 allocated, this does come at a performance cost of roughly 5% though.
508 fixed up block alignment calculation, so it works for varying
509 block sizes. only use strerror() not g_strerror() since the latter
510 depends on working GQuark and GSlice.
511 mem_error(): implemented in terms of fprintf and vfprintf.
513 * tests/slice-color.c: new program to test cache colorization effects.
515 * tests/slice-test.c: trade G_SLICE_CONFIG_ALWAYS_FREE for 0 duration
516 G_SLICE_CONFIG_WORKING_SET_MSECS.
518 2005-12-17 Matthias Clasen <mclasen@redhat.com>
520 * glib/goption.c (parse_short_option): Set an error in all
521 failure cases. (#324332, Tim-Philipp Müller)
523 2005-12-17 Sebastian Wilhelmi <seppi@seppi.de>
525 * glib/gatomic.c: Fix memory barrier position in g_atomic_int_get
526 and g_atomic_pointer_get. Add g_atomic_int_set and
527 g_atomic_pointer_set implementations for the !DEFINE_WITH_MUTEXES &&
528 G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case, as well as defining them
529 as functions (additionally to the macros in the header) for the
530 !G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case.
532 2005-12-16 Matthias Clasen <mclasen@redhat.com>
534 * glib/gmem.c (g_allocator_new): Don't return a pointer to
535 a const struct, since apps expect to be able to modify it.
536 (#324179, J. Ali Harlow)
538 Tue Dec 13 10:13:32 2005 Tim Janik <timj@imendio.com>
540 * glib/gatomic.h: added g_atomic_pointer_set() and g_atomic_int_set()
541 for systems where the initialization of atomic variables requires a
542 write memory barrier.
544 2005-12-09 Matthias Clasen <mclasen@redhat.com>
546 * configure.in: Bump version
548 * === Released 2.9.1 ===
550 2005-12-09 Alexander Larsson <alexl@redhat.com>
554 Use G_GNUC_WARN_UNUSED_RESULT on list functions that return
557 2005-12-08 Matthias Clasen <mclasen@redhat.com>
561 2005-12-07 Martyn Russell <martyn@imendio.com>
563 * glib/gasyncqueue.c:
564 - Call g_queue_insert_sorted() instead of duplicating the code.
565 - Call g_queue_sort() instead of duplicating the code.
566 - Invert sort function results to make sure the same sort function
567 gives the same results across glist, gslist, gqueue and
570 * tests/asyncqueue-test.c:
571 - Updated the sort function to reflect the example in the
572 documentation for gasyncqueue.c.
574 2005-12-07 Martyn Russell <martyn@imendio.com>
576 * docs/reference/glib/glib-sections.txt:
577 * docs/reference/glib/tmpl/linked_lists_double.sgml:
578 * docs/reference/glib/tmpl/linked_lists_single.sgml:
581 - Added g_list_insert_sorted_with_data () and
582 g_slist_insert_sorted_with_data ().
583 - Removed the extra check in g_list_sort() and g_slist_sort() for
584 GCompareDataFunc vs. GCompareFunc.
586 2005-12-07 Tor Lillqvist <tml@novell.com>
588 * glib-gettextize.in: Look up prefix at run-time on Win32,
589 assuming the standard directory structure with glib-gettextize in
592 2005-12-06 Matthias Clasen <mclasen@redhat.com>
594 * glib/gmem.h: Exempt GTK+ from the mem chunk deprecation, since
595 we need GTK+ 2.8 to compile against GLib 2.10.
597 2005-12-05 Manish Singh <yosh@gimp.org>
599 * tests/run-collate-test.sh: set LC_COLLATE, not LANG, to be sure
600 to override any user settings.
602 2005-12-05 Matthias Clasen <mclasen@redhat.com>
604 * configure.in: Revert an accidential commit
606 2005-12-05 Matthias Clasen <mclasen@redhat.com>
608 * glib/gutf8.c: Documentation updates. (#323291, Morten Welinder)
610 * tests/Makefile.am (TESTS_ENVIRONMENT):
611 * tests/gobject/Makefile.am (TESTS_ENVIRONMENT): Set
612 MALLOC_CHECK_ and MALLOC_PERTURB_
614 * tests/run-collate-tests.sh: Run the collation tests explicitly
615 in en_US locale. (#320463)
617 * glib/gmem.h: Really deprecate GMemChunk.
624 * glib/gstrfuncs.c: Add versioned deprecation docs.
626 2005-12-05 Martyn Russell <martyn@imendio.com>
628 * docs/reference/glib/glib-sections.txt:
629 * glib/gasyncqueue.[ch]:
630 - Added support for sorting async queues by with _push_sorted(),
631 _push_sorted_unlocked(), _sort() and _sort_unlocked() (#323047).
634 * tests/asyncqueue-test.c:
635 - Added test case for gasyncqueue.c
637 Mon Dec 5 15:53:20 2005 Tim Janik <timj@imendio.com>
639 * glib/gslice.c: implement chain walking for arbitrary ->next pointer
640 offsets in g_slice_free_chain_with_offset() based on a patch by behdad
641 in bug 323178. moved time consuming logic from g_slice_free() out of
642 the inner loop, so g_slice_free_chain_with_offset() provides a real
643 performance benefit over g_slice_free1() now.
645 * glib/gslice.h: renamed g_slice_free_chain() to
646 g_slice_free_chain_with_offset(). implemented g_slice_free_chain() as
647 a type-safe macro as suggested in bug 323178.
648 simplified the macro implementation of g_slice_free() and implemented
649 it in a type safe manner for all compliers as suggested by Morten
650 Welinder <mortenw@gnome.org>.
655 * glib/glib.symbols: s/g_slice_free_chain/g_slice_free_chain_with_offset/
657 2005-12-05 Matthias Clasen <mclasen@redhat.com>
659 * glib/gasyncqueue.c: Add some docs.
661 * tests/libmoduletestplugin_a.c: Fix compiler warnings.
663 * glib/gatomic.c: In the ia64 implementation, use
664 __sync builtin without _si or _di suffix. (#321229,
665 Stanislav Brabec, patch by Andreas Schwab)
667 2005-12-04 Behdad Esfahbod <behdad@gnome.org>
669 * glib/gslice.h: Remove comma at the end of enum.
671 2005-12-04 Matthias Clasen <mclasen@redhat.com>
673 Handle multiple user names with the same UID better.
674 (#319535, Laszlo Peter)
676 * glib/gutils.c (g_get_any_init_do): When determining user
677 data, first look up $LOGNAME. If the UID doesn't match
678 getuid(), fall back to the current behaviour of looking
679 up the user data based on getuid().
681 2005-12-04 Matthias Clasen <mclasen@redhat.com>
683 * glib/gmacros.h (G_GNUC_WARN_UNUSED_RESULT): Add a macro
684 to make gcc warn if a function result is ignored. (#145466,
685 Arjan van de Ven, Alex Larsson)
687 * glib/gmem.h: Add the new attribute to g_realloc and
690 2005-12-03 Matthias Clasen <mclasen@redhat.com>
693 * glib/gcache.h: Deprecate g_cache_value_foreach. (#322956,
696 * glib/gmappedfile.c: Make mapping of empty files
699 * glib/gfileutils.c: Don't fork a new process just to
700 fix the permissions of the created temp file. (#321318,
701 Alexis S. L. Carvalho)
703 2005-12-02 Matthias Clasen <mclasen@redhat.com>
705 * README.in: Add a note about Solaris threads.
708 * configure.in: Remove support for Solaris threads.
709 (#136971, Sebastian Wilhelmi, patch by Andrew Paprocki)
711 2005-12-02 Matthias Clasen <mclasen@redhat.com>
713 * configure.in: Define G_GUINT64_CONSTANT in analogy to
714 G_GINT64_CONSTANT. (#322568, Andrew Paprocki)
716 2005-12-02 Matthias Clasen <mclasen@redhat.com>
718 * glib/gslice.c: Win32 portability fixes and C99ism removal,
719 pointed out by Kazuki Iwamoto. (#323052)
721 2005-12-02 Matthias Clasen <mclasen@redhat.com>
723 * glib/gatomic.c (g_atomic_int_add): Add a missing volatile for
724 the IA64 implementation.
726 Fri Dec 2 16:18:09 2005 Tim Janik <timj@imendio.com>
728 * glib/gslice.c: some naming and type size fixups.
730 Fri Dec 2 13:08:58 2005 Tim Janik <timj@gtk.org>
732 * glib/gslice.h (g_slice_free): reworked GCC-specific type-safe macro
733 variant into something less verbose, but digestible for gcc-3.4.
735 Fri Dec 2 10:55:07 2005 Tim Janik <timj@imendio.com>
737 * tests/slice-test.c: extended to perform the benchmarking on the old
738 memchunk code if 'O' is selected.
740 * tests/memchunks.c: new file which contains the old GLib mem chunks
741 implementation with prefix old_mem_chunk_.
743 * tests/Makefile.am: added memchunks.c
745 Fri Dec 2 00:16:59 2005 Tim Janik <timj@imendio.com>
747 * glib/gslice.c: improved gettimeofday() timer resolution by maintaining
748 acceptable load distribution of the syscall.
749 (allocator_get_magazine_threshold): reduce minimum/single-thread magazine
750 sizes to approximately page_size/2.5 to avoid excessive startup allocations.
752 2005-12-01 Matthias Clasen <mclasen@redhat.com>
754 * tests/gobject/ifacecheck.c (test_iface_base_init): Don't
755 double-free base interface members.
757 * tests/Makefile.am: Remove duplicate variable.
759 Thu Dec 1 17:32:46 2005 Tim Janik <timj@imendio.com>
761 * glib/gslice.[hc]: new slice allocator implementation.
763 * tests/slice-test.c: added random slice allocation test.
765 * glib/gthread.[hc]: removed newly added private thread mem API.
767 * glib/gthreadinit.h:
770 * glib/gmem.c: divided glib threading initialisation into three phases,
771 initialisation where private keys and messaging are not available (only
772 needed by gmem.c), initialisation without messaging but private keys
773 available (gslice.c, gmessage.c), and full fledged initialisers that
774 server the rest of glib. initialisation functions got renamed to reflect
775 the limitations of their corresponding phases.
777 * glib/gmem.c: removed memchunk code, defer allocations to
780 * glib/gmem.[hc]: removed g_slice_* skeletons.
782 * glib/glib.symbols: added g_slice_* symbols.
784 * configure.in: check for availability of posix_memalign(3), memalign(3)
787 * glib/Makefile.am: added gslice.[hc].
789 2005-12-01 Tor Lillqvist <tml@novell.com>
791 * glib/gstdio.c (g_stat): In the Win32 implementation, strip
792 trailing slash(es) for non-root folders. stat() fails if non-root
793 folders are specified with trailing slashes. It's too much hassle
794 to demand that callers strip such slashes themselves, especially
795 as it is easy to get it wrong and strip the slash of a root
797 (g_rename): On NT-based Windows, use MoveFileEx() with
798 MOVEFILE_REPLACE_EXISTING to better match Unix behaviour.
800 2005-11-28 Matthias Clasen <mclasen@redhat.com>
802 Fix G_STMT_START / G_STMT_END on Solaris. (#321972,
805 * configure.in: Check whether do { } while (0) works.
807 * glib/gmacros.h: Use do { } while (0) for G_STMT_START /
808 G_STMT_END if it works.
810 2005-11-28 Matthias Clasen <mclasen@redhat.com>
812 * glib/gthread.c (g_static_rw_lock_wait, g_static_rw_lock_signal):
813 * glib/gnode.c (g_node_depth_traverse_level):
814 * glib/gmem.c (g_allocator_new):
815 * glib/ghash.c (g_hash_table_unref, g_hash_table_destroy)
816 (g_hash_table_foreach_remove_or_steal): Silence compiler
819 2005-11-27 Matthias Clasen <mclasen@redhat.com>
821 * glib/gunicollate.c (g_utf8_collate_key_for_filename):
822 Don't read beyond len. (#322520, Christian Persch)
824 Wed Nov 23 17:34:01 2005 Tim Janik <timj@imendio.com>
826 * glib/gdataset.c: access datalist flags via atomic pointer access
827 functions, instead of acquiring the dataset lock. this is faster and
828 also matches the atomic pointer readouts in gdatalistprivate.h and
829 direct pointer modifications required by gobject.c.
831 Wed Nov 23 13:35:31 2005 Tim Janik <timj@gtk.org>
833 * glib/gdataset.c: streamlined initialization code somewhat,
834 removed GData node cache.
836 2005-11-22 Matthias Clasen <mclasen@redhat.com>
838 * glib/ghash.c (g_hash_table_ref):
839 (g_hash_table_unref): Mark these as new API.
841 Avoid double locking in g_intern_string (#322133,
844 * glib/gdataset.c (g_quark_from_string_internal): New
845 internal function which factors out common parts of
846 g_quark_from[_static]_string.
847 (g_quark_from_string, g_quark_from_static_string):
848 Use g_quark_from_string_internal.
849 (g_intern_string, g_intern_static_string): Use
850 g_quark_from_string_internal, and only take the
852 (g_quark_new): Don't store the strings shifted by -1
853 in the g_quarks array.
854 (g_quark_to_string): Adapt to the previous change.
856 Tue Nov 22 14:04:26 2005 Tim Janik <timj@imendio.com>
860 g_hash_table_new_full(): create hash tables with a ref count of 1.
861 g_hash_table_ref(): atomically ref_count+=1
862 g_hash_table_unref(): atomically ref_count-=1, destroys hash table
863 when refcount reaches 0.
864 g_hash_table_destroy(): just destroy keys and values, unref by 1.
865 g_hash_table_insert():
866 g_hash_table_replace(): assert ref_count>0.
869 * glib/gatomic.c: added 'volatile' qualifier to all atomic pointer and
872 2005-11-20 Behdad Esfahbod <behdad@gnome.org>
874 * glib/guniprop.c (g_unichar_get_mirror_char): Remove unused
875 variables. (#321984, Andrew Paprocki)
877 2005-11-18 Matthias Clasen <mclasen@redhat.com>
879 * configure.in: Bump version
881 2005-11-17 Matthias Clasen <mclasen@redhat.com>
883 * === Released 2.9.0 ===
885 * glib/glib.symbols: Updates
887 * glib/gutf8.c (g_utf8_offset_to_pointer): Add some performance
888 hints to the docs. (Paolo Borelli)
892 * glib/gthread.c (g_thread_foreach): Mark as new api.
894 * README.in: Updates.
898 * glib/gdate.c (g_date_set_time_t):
899 (g_date_set_time_val): Add functions to set a date from
900 a time_t and from a GTimeVal.
901 (g_date_set_time): Deprecate. (#314067, Roger Leigh)
904 * tests/utf8-pointer.c: Unit tests for g_utf8_pointer_to_offset
905 and g_utf8_offset_to_pointer.
907 * glib/gutf8.c (g_utf8_pointer_to_offset)
908 (g_utf8_offset_to_pointer): Handle negative offsets, and use
909 "stutter stepping" for going backwards. (#320638, Larry
914 * glib/gthread.c: const correctness fixes, found
915 by Arjan van de Ven and gcc.
917 2005-11-16 Behdad Esfahbod <behdad@gnome.org>
919 * docs/reference/glib/tmpl/unicode.sgml: Correct typo on mentioning
920 @G_UNICODE_BREAK_UNKNOWN that should be #G_UNICODE_BREAK_UNKNOWN.
922 2005-11-09 Behdad Esfahbod <behdad@gnome.org>
924 * glib/guniprop.c: Use bit hacks instead when checking a general
925 category value against multiple values.
927 * glib/gutf8.c: Change ISO10646 to Unicode in docs.
929 2005-11-10 Simos Xenitellis <simos@gnome.org>
931 * configure.in: Added tt (Tatar) to ALL_LINGUAS.
933 2005-11-05 Tor Lillqvist <tml@novell.com>
935 * glib/giowin32.c (g_io_win32_set_flags): Don't call g_warning().
936 (#320688) Instead set error and return failure.
938 2005-11-07 Matthias Clasen <mclasen@redhat.com>
940 * glib/gmain.c (g_main_dispatch): Don't call cb_funcs->unref
941 while holding the context lock. (#320886, Andy Wingo)
943 2005-11-07 Billy Biggs <vektor@dumbterm.net>
945 * tests/refcount/Makefile.am: Add a missing reference to libglib
946 in the LDADD for this test.
948 2005-11-07 Matthias Clasen <mclasen@redhat.com>
950 * glib/gmem.c (g_allocator_new): We need to set n_preallocs to a
951 nonzero value, otherwise GTK+ 2.8 breaks when compiled against
952 GLib 2.9. (#320755, Luca Ferretti)
954 2005-11-04 Matthias Clasen <mclasen@redhat.com>
956 * glib/Makefile.am: Apply a patch from Behdad Esfahbod to
957 use a faster lookup table for g_unichar_get_mirror_char().
958 * glib/gmirroringtable.h: The new table.
959 * glib/gunichartables.h: Remove bidi_mirroring_table.
960 * glib/gen-unicode-tables.pl: Don't generate the mirroring
962 * glib/glib-mirroring-tab/*: Sources for the program
963 which generated gmirroringtable.h.
965 * glib/glist.c: Avoid some code duplication.
967 * glib/gscanner.h: Include gdataset.h. (#320322)
969 * glib/gdebug.h: add new GDebugFlag for fatal_criticals
970 * glib/gmessages.c: (_g_debug_init): handle G_DEBUG=fatal_criticals,
971 to help find critical warnings in applications. (#320017,
974 2005-11-02 Tor Lillqvist <tml@novell.com>
976 * glib/glib.symbols: Remove large amount of trailing whitespace
977 from one line. Remove #ifdef G_UNDEFINED from around
978 g_hash_table_new and g_hash_table_new_full symbols, what was the
979 purpose of that? Add the g_slice functions.
981 Tue Nov 1 16:24:20 2005 Tim Janik <timj@imendio.com>
983 * glib/gmem.[hc]: prepared deprecation of GMemChunk and GAllocator.
984 added g_slice_*() API to allocate and cache small bits of memory.
985 an actuall allocator implementation for g_slice_*() is still pending.
987 * glib/gthread.[hc]: changes from a patch by Matthias Clasen.
988 changed GRealThread list to use in-structure *next; fields instead
989 of GSList, in order for thread iteration to not depenend on g_slice_*()
991 _g_thread_mem_private_get():
992 _g_thread_mem_private_set(): added accessors for private memory,
993 needed because the ordinary GPrivate implementation relies on GArray
994 and GSList and therefore indirectly on working g_slice_*() allocations.
997 g_thread_foreach(): new public API function to loop over all existing threads.
1011 removed GAllocator and free list usages and accompanying locks.
1012 use g_slice_*() API to allocate and cache small bits of memory.
1014 * glib/ghook.h: removed GMemChunk field from public API.
1017 * glib/glist.h: deprecate allocator API, provide _free1() for consistency.
1019 * glib/gnode.h: deprecate allocator API.
1021 * glib/gmain.c: reordered GPollRec fields so g_slice_free_chain() can
1022 be used for poll rec lists.
1024 * glib/grel.c: removed mem chunk usage, and allocated tuples via g_slice_*().
1025 g_relation_destroy(): free all tuples from the all_tuples hash table,
1026 this effectively maintains the life time track keeping of tuples.
1027 g_relation_delete_tuple(): free tuples which are removed from the
1028 all_tuples hash table. this fixes a temporary leak that was present
1029 in the memchunk code until the destruction of the relation.
1031 2005-10-29 Matthias Clasen <mclasen@redhat.com>
1033 * tests/convert-test.c: Add some tests for conversions between
1034 UTF-8, UCS-4 and UTF-16.
1036 * glib/gutf8.c (g_utf8_to_ucs4, g_utf8_to_utf16): Fix handling
1037 of len == -1, noticed by Morten Welinder.
1039 2005-10-27 Erdal Ronahi <erdal.ronahi@gmail.com>
1041 * configure.in: Added ku (Kurdish) to ALL_LINGUAS
1043 2005-10-26 Matthias Clasen <mclasen@redhat.com>
1045 * glib/gutf8.c (g_ucs4_to_utf8): Don't set items_read twice
1046 in the error case, and add some documentation. (#319806, Morten
1049 2005-10-19 Manish Singh <yosh@gimp.org>
1051 * configure.in: Use AC_CHECK_FUNCS for _NSGetEnviron, to get the
1052 config.h symbol automatically. Fixes bug #313731.
1054 2005-10-19 Tor Lillqvist <tml@novell.com>
1056 * glib/gwin32.c (g_win32_get_package_installation_directory):
1057 Return a g_strdup()ed copy of the value stored in the hash table,
1058 so that it can be g_free()d without leaving a dangling pointer in
1059 the hash table. (#319232)
1061 2005-10-06 Matthias Clasen <mclasen@redhat.com>
1063 * glib/gunicollate.c (g_utf8_collate_key_for_filename): Handle
1064 all-zero sequences correctly. (#317930, Sebastien Bacher)
1066 2005-10-05 Matthias Clasen <mclasen@redhat.com>
1068 * glib/gmarkup.c (g_markup_escape_text): Doc addition
1070 * Makefile.am: Add Collation tests.
1072 * tests/collate/*: Inputs and expected outputs for collation tests.
1074 * tests/run-collate-tests.sh: Script to run collation tests.
1076 * tests/unicode-collate.c (main): Rework slightly to make
1077 it usable in unit tests. Also test g_utf8_collate_key_for_filename().
1079 2005-10-01 Behdad Esfahbod <behdad@gnome.org>
1081 * docs/reference/glib/tmpl/unicode.sgml:
1082 * glib/gen-unicode-tables.pl:
1084 * glib/gunichartables.h:
1086 * tests/casefold.txt:
1087 * tests/casemap.txt: Updated to Unicode 4.1. There are five new
1088 GUnicodeBreakType types. That may break some applications, like
1091 2005-09-26 Matthias Clasen <mclasen@redhat.com>
1093 * glib/gstrfuncs.c (g_ascii_strtoull): Add details to the
1094 docs. (#314393, Matthew F. Barnes)
1096 * glib/glib.symbols:
1097 * glib/gprintf.h: Remove g_snprintf() and g_vsnprintf(), since
1098 they are already declared in glib.h. This doesn't break documented
1099 use of gprintf.h, but should probably be pointed out in the
1100 release notes for 2.10. (#314232, Behdad Esfahbod)
1102 Tue Sep 20 13:16:04 2005 Tim Janik <timj@imendio.com>
1104 * glib/gpattern.c (g_pattern_ph_match): applied significant recursion
1105 complexity optimization, based on a patch from Matthias Clasen.
1107 * tests/patterntest.c: more tests, mostly from matthias.
1109 2005-09-20 Matthias Clasen <mclasen@redhat.com>
1111 * glib/gqueue.c (g_queue_insert_sorted): Correct the docs.
1112 (#316703, Mark Drago)
1114 Mon Sep 19 17:23:23 2005 Tim Janik <timj@imendio.com>
1116 * glib/gpattern.c: applied a patch from matthias which checks on the
1117 upper bound of GPatternSpec length to optimize matches.
1120 * tests/patterntest.c: added more match cases.
1122 2005-09-16 Tor Lillqvist <tml@novell.com>
1124 * glib/gstrfuncs.c (g_ascii_strcasecmp, g_ascii_strncasecmp): Add
1125 warning to doc comment that these functions should not be used on
1126 encodings like CP932.
1128 2005-09-14 Matthias Clasen <mclasen@redhat.com>
1130 * tests/keyfile-test.c: Add a test for grup names of length 1.
1132 * glib/gkeyfile.c (g_key_file_line_is_group): Accept group names
1133 of length 1. (#316309)
1135 2005-09-12 Matthias Clasen <mclasen@redhat.com>
1137 * glib/gmarkup.c (g_markup_escape_text): Clarify docs.
1139 2005-09-11 Sebastian Wilhelmi <seppi@seppi.de>
1141 * tests/refcount/Makefile.am (INCLUDES): Link the the refcount
1142 tests to the system thread library $(G_THREAD_LIBS). Fixes #313744
1145 2005-09-11 Kjartan Maraas <kmaraas@gnome.org>
1147 * glib/gmain.c: (g_child_watch_prepare), (g_child_watch_check),
1148 (child_watch_helper_thread): Remove some dead code. Closes
1151 2005-09-07 Tor Lillqvist <tml@novell.com>
1153 * glib/Makefile.am: Create also a console version of the
1154 gspawn-win32-helper program, gspawn-win32-helper-console.exe.
1155 It's otherwise identical to gspawn-win32-helper.exe, except marked
1156 as a console application (linked without the -mwindows option).
1158 * glib/gspawn-win32.c (do_spawn_directly, do_spawn_with_pipes):
1159 Drop the dont_wait parameter. Its truth value correlated 100% with
1160 the NULLness of the exit_status parameter anyway, so it's enough
1161 to check whether exit_status is NULL. Invert the sense of the
1162 dont_return_handle parameter and rename it to do_return_handle, to
1163 make the code easier to read by avoiding double negations.
1165 (g_spawn_sync_utf8, g_spawn_async_with_pipes_utf8): Modify calls
1166 to do_spawn_with_pipes() accordingly.
1168 (do_spawn_with_pipes): If we have a console, use the console
1169 version of the helper program, otherwise use the GUI one. This
1170 avoids extra console windows opening up in some situations. (In
1171 case a console application uses the GUI gspawn-win32-helper.exe to
1172 spawn another console application we would get a separate console
1173 for the spawned console application).
1175 * glib-zip.in: Distribute also gspawn-win32-helper-console.exe.
1177 2005-09-05 Matthias Clasen <mclasen@redhat.com>
1179 * glib/gmappedfile.c (g_mapped_file_new): Report an error
1180 if the file is too large. (#315275, Kjartan Maraas)
1182 * glib/gkeyfile.c (g_key_file_load_from_fd): The return value
1183 of read() is signed. (#315273, Kjartan Maraas)
1185 2005-08-31 Tor Lillqvist <tml@novell.com>
1187 * glib/gutils.h: Wrapping atexit() is a bad idea on Windows, where
1188 the EXE and each DLL have their own atexit function chains.
1190 #define g_atexit as atexit instead. This means it has a
1191 better chance of doing what the caller wants. For instance,
1192 gtkhtml calls g_atexit() registering a function in gtkhtml
1193 itself. This caused a crash when g_atexit() was implemented as a
1194 function in the GLib DLL. The gtkhtml DLL was already unloaded by
1195 the time the GLib DLL got unloaded.
1197 * glib/gutils.c: #undef the #define mentioned above, to also get a
1198 real g_atexit() into the DLL for backward compatibility. Document
1199 the Windows behaviour of g_atexit(), and document the varying ways
1200 atexit() can behave in the context of dynamically loaded modules
1203 2005-08-31 Matthias Clasen <mclasen@redhat.com>
1205 * glib/glib.symbols:
1207 * glib/gdataset.c: Add string interning functions.
1209 2005-08-28 Matthias Clasen <mclasen@redhat.com>
1211 * glib/giochannel.c: Unify some near-duplicate strings. (#314654,
1214 2005-08-26 Matthias Clasen <mclasen@redhat.com>
1216 * configure.in: Bump version to 2.9.0
1218 2005-08-25 Tor Lillqvist <tml@novell.com>
1220 Make also the g_spawn*() functions take parameters in the GLib
1221 file name encoding, i.e. UTF-8, on Windows. Has no impact on Unix
1222 API or ABI. Like the other GLib API that was earlier changed to
1223 use UTF-8 on Windows, the names of the functions that take UTF-8
1224 have _utf8 suffixes added by using preprocessor macros in the
1225 header file. The old names are kept for functions with the old
1226 behaviour, taking parameters in the system codepage, for DLL ABI
1229 * glib/gspawn.h: On Win32 add the suffix _utf8 to the names of the
1230 g_spawn*() functions.
1232 * glib/gspawn-win32.c: Use wide-char API on NT-based
1233 Windows. Convert parameters from UTF-8 to wide chars (NT) or
1234 system codepage (Win9x) and call the C library _wspawn*() or
1235 spawn*() functions respectvely. Add DLL ABI stability versions
1236 that take parameters in the system codepage.
1238 * glib/gspawn-win32-helper.c: On NT-based Windows use the
1239 wide-char versions of argv and envp, and use wide-char API to
1240 change directory and spawn the program to run. Remove the verbose
1241 debugging output, it was too complex to modify for the wide-char
1242 features. (Just add temporary debugging printouts if needed, no
1243 need to have them permanently in the source.)
1245 * glib/gspawn.c: Corresponding documentation updates.
1247 * glib/glib.symbols: Corresponding changes: Mark the ABI stability
1248 symbols as PRIVATE, add the new _utf8-suffixed ones.
1250 2005-08-24 Stepan Kasal <kasal@ucw.cz>
1252 * glib/gtypes.h (G_MININT64): Cast the constant to gint64; it is
1253 guint64 otherwise and that can produce warnings about comparison
1254 between signed and unsigned.
1256 2005-08-23 Matthias Clasen <mclasen@redhat.com>
1258 * glib/gutils.c: Fix the crt_externs.h include.
1260 2005-08-23 Stepan Kasal <kasal@ucw.cz>
1262 * NEWS: Fix spelling of my first name.
1264 2005-08-23 Matthias Clasen <mclasen@redhat.com>
1268 * === Released 2.8.1 ===
1272 2005-08-20 Hans Breuer <hans@breuer.org>
1274 * glib/makefile.msc.in : link with ws2_32.lib
1276 2005-08-18 Tor Lillqvist <tml@novell.com>
1278 * configure.in: Check for <sys/wait.h>
1280 * glib/gbacktrace.c: Include <sys/wait.h> on if HAVE_SYS_WAIT_H.
1282 2005-08-18 Ross Burton <ross@burtonini.com>
1285 Optimise single-character insertions.
1290 * tests/string-test.c:
1291 Add tests for new optimisation, and fix a leak.
1293 2005-08-17 Matthias Clasen <mclasen@redhat.com>
1295 * configure.in: Check for crt_externs.h and _NSGetEnviron.
1297 * glib/gutils.c: On Darwin, include crt-externs.h and
1298 define environ using _NSGetEnviron(). (#313731)
1300 2005-08-16 Stepan Kasal <kasal@ucw.cz>
1302 * glib/gutils.c (g_get_any_init): Move the body of the big if...
1303 (g_get_any_init_do): ... to this new function.
1304 (g_get_any_init): Declare as inline.
1305 (g_get_any_init_locked): New inline function, does the locking.
1306 Make use of these two throughout the code.
1308 2005-08-15 Matthias Clasen <mclasen@redhat.com>
1310 * glib/gbacktrace.c (g_on_error_stack_trace): Wait for
1311 the child process and then simply return. This makes
1312 The "S" option work as documented in g_on_error_query().
1313 (#313125, Matthew F. Barnes)
1315 * glib/gunicode.h: Update the link to Unicode category
1316 values. (#313369, Behnam Esfahbod)
1318 * glib/gqueue.c (g_queue_find_custom): Clarify docs
1319 a little. (#311727, Tristan van Berkom)
1321 * glib/abicheck.sh, gobject/abicheck.sh: Make the
1322 check work on ia64 too, where some symbols ended up
1323 in yet another section.