1 2006-03-08 Tor Lillqvist <tml@novell.com>
3 * glibconfig.h.win32.in: Add G_GUINT64_CONSTANT.
5 2006-03-07 Matthias Clasen <mclasen@redhat.com>
7 * configure.in: Bump version
9 * === Released 2.10.1 ===
13 * glib/gutf8.c (_g_charset_get_aliases): Match the prototype
14 in gconvert.c to fix build problems in NetBSD. (#333651,
15 Julio M. Merino Vidal)
17 2006-03-06 Anders Carlsson <andersca@imendio.com>
20 Make .dylib the shared library suffix on darwin.
21 (#322476, Vladimir Panov)
23 2006-03-06 Anders Carlsson <andersca@imendio.com>
26 Skip checking thread flags on Darwin. Fix suggested by Manish Singh.
27 (#314794, Gregor Riepl)
29 2006-03-01 Tor Lillqvist <tml@novell.com>
31 * glib/gutils.c (g_listenv): Separate implementation on Win32: Use
32 the wide character API on NT-based Windows. Return UTF-8 strings.
34 * glib/glib.symbols: Don't mark g_listenv as PRIVATE, as that
35 meant it wasn't present in the import library. PRIVATE is used
36 only for the backwards-compatibility DLL ABI stability hacks.
38 2006-02-24 Matthias Clasen <mclasen@redhat.com>
40 * configure.in: Bump version
42 * === Released 2.10.0 ===
44 * README.in: Mention the slice allocator and GInitiallyUnowned.
48 * configure.in: Bump version to 2.10.0
50 * tests/utf8-pointer.c: Add tests for g_utf8_strlen().
52 * glib/gutf8.c: Fix boundary cases in g_utf8_strlen().
55 2006-02-23 Kjartan Maraas <kmaraas@gnome.org>
57 * tests/completion-test.c: (main): Plug leaks reported by
59 * tests/convert-test.c: (test_iconv_state), (test_one_half):
61 * tests/patterntest.c: (test_compilation): Same
62 * tests/shell-test.c: (do_argv_test): Same.
63 * tests/unicode-caseconv.c: (main): Same.
64 * tests/uri-test.c: (run_to_uri_tests): Same.
67 2006-02-19 Martyn Russell <martyn@imendio.com>
69 * tests/asyncqueue-test.c:
71 * tests/slist-test.c: Updated to test _sort, _sort_with_data,
72 _insert_sorted and _insert_sorted_with_data API.
74 2006-02-18 Matthias Clasen <mclasen@redhat.com>
76 * tests/gobject/Makefile.am: Add paramspec-test
78 * tests/gobject/paramspec-test.c: Some GParamSpec tests.
80 * tests/gobject/gvalue-test.c: Add more tests.
82 2006-02-17 Kang Jeong-Hee <keizie@gmail.com>
84 * glib/gutf8.c (g_utf8_find_prev_char): Correct documentation typo.
86 2006-02-15 Sebastian Wilhelmi <seppi@seppi.de>
88 * glib/gthreadpool.c: Fix deadlock when signalling the thread
89 which freed a thread pool (#331110, Chris Wilson).
91 Tue Feb 14 17:00:43 2006 Tim Janik <timj@imendio.com>
93 * glib/gslice.c: only define _XOPEN_SOURCE to 600 to get at
94 posix_memalign() in case we actually are going to use it, because
95 we detected a compliant implementation (#328997).
97 * configure.in (enable_included_printf): don't include malloc.h when
98 testing for posix_memalign() funcitonality, since this may break the
99 test on some systems (#328997).
101 2006-02-14 Matthias Clasen <mclasen@redhat.com>
103 * glib/gstrfuncs.c: Improve docs.
105 2006-02-11 Matthias Clasen <mclasen@redhat.com>
107 * configure.in: Bump version
109 2006-02-10 Matthias Clasen <mclasen@redhat.com>
111 * === Released 2.9.6 ===
115 2006-01-31 Behdad Esfahbod <behdad@gnome.org>
117 * docs/reference/glib/tmpl/relations.sgml (g_relation_insert): Specify
118 the type of varargs arguments. (bug #317679)
120 2006-01-31 Matthias Clasen <mclasen@redhat.com>
122 * autogen.sh: Touch README and INSTALL here to pacify
123 automake. (#329124, Kjartan Maraas, Tim Janik)
125 Tue Jan 31 16:45:03 2006 Tim Janik <timj@imendio.com>
127 * glib/gdataset.c: carry out all dtalist pointer accesses atomically,
128 some missing cases where pointed out by Sebastian Wilhelmi.
130 Tue Jan 31 12:33:48 2006 Tim Janik <timj@imendio.com>
132 * configure.in: generate build/win32/vs8/Makefile, to fix build breakage.
134 2006-01-30 David Schleef <ds@schleef.org>
136 * glib/grand.c: Update URL
138 2006-01-29 Sven Herzberg <herzi@gnome-de.org>
140 * docs/reference/gobject/tmpl/param_value_types.sgml: tell how to
141 create container classes which are as flexible as a GValue is
143 2006-01-27 Matthias Clasen <mclasen@redhat.com>
145 * configure.in: Bump version
147 * === Released 2.9.5 ===
149 * glib/glib.symbols: Add g_mem_gc_friendly as an exported
154 2006-01-26 Matthias Clasen <mclasen@redhat.com>
156 * glib/gmem.c (g_mem_init_nomessage): Fix C99isms. (#328705,
159 Wed Jan 25 19:16:57 2006 Tim Janik <timj@imendio.com>
161 * fixed buglets reported by Jens Granseuer in #328254.
163 * configure.in: free the memory allocated in posix_memalign() tests.
165 * glib/gslice.c: spelling fixes.
167 Wed Jan 25 16:39:18 2006 Tim Janik <timj@imendio.com>
169 * glib/gslice.c: honour g_mem_gc_friendly settings when freeing
170 slices, make sure g_mem_gc_friendly is properly initialized.
172 * gmem.[hc]: ensure g_mem_gc_friendly is initialized from G_DEBUG upon
173 the first allocation. applied some branching optimizations.
175 * docs/macros.txt: reflected --enable-gc-friendly change and
176 described ENABLE_GC_FRIENDLY_DEFAULT as well as G_DEBUG=gc-friendly.
178 * configure.in: changed --enable-gc-friendly=yes to define
179 ENABLE_GC_FRIENDLY_DEFAULT.
181 * glib/garray.c: changed ENABLE_GC_FRIENDLY macro #ifdef-s to
182 if (G_UNLIKELY (g_mem_gc_friendly)).
185 * glib/ghash.c: removed ENABLE_GC_FRIENDLY code which is now taken
186 care of by g_slice_free1().
188 * tests/slice-test.c: fixed leaks, reported by Kjartan Maraas.
190 Tue Jan 24 17:49:36 2006 Tim Janik <timj@imendio.com>
192 * glib/gslice.c: only use posix_memalign() if it's known to work,
193 revert to memalign() otherwise.
195 * configure.in: check for broken posix_memalign() implementations
198 2006-01-24 Matthias Clasen <mclasen@redhat.com>
200 * tests/unicode-encoding.c: Use UTF-16LE as target encoding
201 on all little-endian systems. (#143380, Marc Moorcroft)
203 2006-01-23 Matthias Clasen <mclasen@redhat.com>
205 * configure.in: Change the shared libary extension for hpux-ia64
206 to so. (#328253, The Written Word)
208 Mon Jan 23 17:30:33 2006 Tim Janik <timj@imendio.com>
210 * glib/gutils.c (_g_getenv_nomalloc): wiped out all the wonderfull
211 G_OS_WIN32 code i wrote ;-[) after tml told me windows has getenv()
214 Mon Jan 23 16:46:20 2006 Tim Janik <timj@imendio.com>
216 * glib/gslice.c (slice_config_init): initialize GSlice config from
217 G_SLICE environment variable. we support G_SLICE=always-malloc
218 currently, which forces all g_slice_*() allocations to use the system
222 g_parse_debug_string(): added a note about not using g_malloc() here.
223 _g_getenv_nomalloc(): getenv() variant that doesn't use g_malloc or
224 g_slice. contains only guesswork in the WIN32 branch.
226 2006-01-18 Matthias Clasen <mclasen@redhat.com>
230 * === Released 2.9.4 ===
234 2006-01-17 Matthias Clasen <mclasen@redhat.com>
236 * glib/gkeyfile.c (g_key_file_get_groups): Correct a g_new0()
237 call. (#327421, Morten Welinder)
239 2006-01-17 Sebastian Wilhelmi <seppi@seppi.de>
241 * glib/gthreadpool.c: To avoid deadlocks get rid of the settings
242 G_LOCK. Use the unused_thread_queue lock instead. Change
243 g_thread_pool_thread_proxy such that threads only wait on
244 non-exlusive pools for at most a 1/2 second. Do not reorder tasks
245 due to superfluous tasks. Global tasks wait at most for
246 max-idle-time milliseconds. Make sure, that no task is woken up
247 twice for the same event via a wakeup_serial. This fixes #324228.
249 * tests/threadpool-test.c: Adapt test accordingly. Do not pass
250 invalid NULL into the thread pools. This as well fixes #327290.
252 2006-01-16 Matthias Clasen <mclasen@redhat.com>
254 * configure.in: Bump version
256 * === Released 2.9.3 ===
260 * glib/gthread.h (g_static_mutex_get_mutex_impl_shortcut):
261 Rework the strict aliasing fix to not break C++, pointed
262 out by Murray Cumming.
264 * glib/gasyncqueue.c (g_async_queue_push_sorted_unlocked):
265 Signal waiting threads, problem noticed by Christian Kellner.
267 2006-01-16 Matthias Clasen <mclasen@redhat.com>
269 Fix bug 326747, Alberto Ruiz:
271 * glib/gconvert.c (make_valid_utf8): Change this function to
272 replace unknown characters by the Unicode replacement character
273 instead of '?', and don't append "(invalid encoding)".
274 (g_filename_display_name, g_filename_display_basename): Document
275 how to determine if the filename was in an invalid encoding.
277 2006-01-14 Matthias Clasen <mclasen@redhat.com>
279 * glib/gtree.c: Replace the simple recursive implementation
280 by a nonrecursive, threaded implementation by Maurizio
283 2006-01-12 Matthias Clasen <mclasen@redhat.com>
285 * tests/asyncqueue-test.c (main): Fix the
286 assert to not always trigger. (#326558,
289 2006-01-11 Matthias Clasen <mclasen@redhat.com>
291 * tests/convert-test.c: Don't test items_read and
292 items_written in error cases where it is not set.
294 * tests/asyncqueue-test.c: Set error to NULL.
296 2006-01-08 Matthias Clasen <mclasen@redhat.com>
298 * glib/gtree.c: Break some long lines.
300 * tests/tree-test.c: Add more tests.
302 * glib/gslice.c: Include config.h first, pointed out
305 2005-12-20 Sven Herzberg <herzi@gnome-de.org>
307 * docs/reference/gobject/tmpl/gtype.sgml: explain that
308 G_TYPE_INSTANCE_GET_CLASS() does behave different during
311 2006-01-06 Matthias Clasen <mclasen@redhat.com>
313 * glib/gslice.c: Only define _XOPEN_SOURCE if we know
314 that we have posix_memalign(). (#323937, Bogdan Nicula)
316 2006-01-05 Matthias Clasen <mclasen@redhat.com>
318 * configure.in: Bump version
320 * === Released 2.9.2 ===
322 2006-01-05 Hans Breuer <hans@breuer.org>
324 * glib/makefile.msc.in : added gslice.obj
326 2006-01-05 Matthias Clasen <mclasen@redhat.com>
330 * glib/gmain.c: Clarify the documentation of source ids
331 a bit. (#325874, Dan Williams)
333 * configure.in: Fix another strict aliasing problem.
335 * glib/gthreadpool.c: Avoid use of varargs macro. (#325864,
338 2006-01-04 Manish Singh <yosh@gimp.org>
341 * glib/gstdio.h: don't macroized g_access, g_chdir, and g_unlink
342 either, since they have the same issue as g_rmdir. (Related to
345 Wed Jan 4 13:33:25 2006 Tim Janik <timj@gtk.org>
347 * glib/gslice.c (magazine_cache_pop_magazine): don't reverse chunk
348 order when creating magazines, so we hand out chunks with ascending
351 2006-01-04 Abel Cheung <maddog@linuxhall.org>
353 * configure.in: Added "ml" "zh_HK" to ALL_LINGUAS.
355 2006-01-03 Martyn Russell <martyn@imendio.com>
357 * docs/reference/glib/glib-sections.txt:
359 * glib/gthreadpool.[ch]:
360 - Added new API g_thread_pool_get_idle_time() and
361 g_thread_pool_set_idle_time(). (#324228).
363 * tests/threadpool-test.c:
364 - Updated test case to do thread pool sorting, thread pool with
365 no sorting and a thread pool with idle thread timeouts.
367 2006-01-03 Matthias Clasen <mclasen@redhat.com>
369 * glib/gmain.h: Add new functions here, too.
371 * glib/glib.symbols: Add new functions.
373 2005-12-20 Michael Meeks <michael.meeks@novell.com>
375 * glib/gmain.c (g_main_context_is_owner): new method
376 to determine if the current thread is the owner of the
379 2006-01-02 Matthias Clasen <mclasen@redhat.com>
383 * glib/gstdio.c (g_rmdir): Don't provide g_rmdir() as a macro
384 expanding to rmdir, since rmdir is not declared in a portable
385 system header we can include in gstdio.h. (#325249, Jani Monoses)
387 2006-01-01 Tor Lillqvist <tml@novell.com>
389 * glib/gspawn-win32.c (g_spawn_sync_utf8): Set the GIOChannels for
390 stdout and stderr to unbuffered. Otherwise the giochannel layer
391 will try to read from them regardless whether the
392 g_io_channel_win32_poll() call here has indicated
393 readability or not. (#325310)
395 2005-12-29 Matthias Clasen <mclasen@redhat.com>
397 * glib/gutils.c (glib_check_version): Fix a copy-and-paste error
398 in the docs. (#325273, Declan Naughton)
400 2005-12-27 Manish Singh <yosh@gimp.org>
402 * tests/run-collate-tests.sh: use LC_ALL instead of LC_COLLATE, to
403 make sure we really override things.
405 2005-12-27 Matthias Clasen <mclasen@redhat.com>
407 Fix #316221, Michal Benes, Stanislav Brabec;
409 * configure.in: Fix a strict aliasing problem in
410 g_static_mutex_get_mutex().
411 * glib/gthread.h: ...and in
412 g_static_mutex_get_mutex_impl_shortcut().
414 * glib/gdatasetprivate.h: Add a cast to silence compiler
415 warnings. (#321978, Andrew Paprocki)
417 Partial fix for bug #323937, Bogdan Nicula.
419 * configure.in: Check for malloc.h
420 * glib/gslice.c: Don't include malloc.h unconditionally.
422 2005-12-27 Manish Singh <yosh@gimp.org>
424 * tests/run-collate-tests.sh: export LC_COLLATE so it takes.
425 (#324950, Dan Yefimov)
427 2005-12-26 Matthias Clasen <mclasen@redhat.com>
429 * glib/gslice.c: On Win32, include process.h (#325015, Kazuki
432 2005-12-25 Matthias Clasen <mclasen@redhat.com>
434 * AUTHORS: Update my email
436 * tests/slice-test.c: Fix C99isms. (#324950, Dan Yefimov)
438 2005-12-21 Matthias Clasen <mclasen@redhat.com>
440 * README.in: Add some notes on when 'make check' may fail.
442 2005-12-20 Matthias Clasen <mclasen@redhat.com>
445 * glib/gthreadpool.h:
446 * glib/gthreadpool.c (g_thread_pool_set_sort_function): New function
447 to sort tasks pushed into a threadpool. (#324479, Martyn Russell)
449 * tests/threadpool-test.c: Test this.
451 Tue Dec 20 18:14:14 2005 Tim Janik <timj@imendio.com>
453 * glib/gslice.[hc]: added mem_error() and mem_assert() to test and
454 handle errors without depending on gmessage.c which might not be
455 setup when the error occours.
456 removed G_SLICE_CONFIG_ALWAYS_FREE config option, fixed the code so
457 always freeing can be achieved by adjusting the working set time to
458 0 with G_SLICE_CONFIG_WORKING_SET_MSECS.
459 added G_SLICE_CONFIG_COLOR_INCREMENT to test different color increments
460 (mainly 0 and 1). reduced the minimum block size to 128 bytes, to
461 minimize wastage if small amounts of differently sized structrues are
462 allocated, this does come at a performance cost of roughly 5% though.
463 fixed up block alignment calculation, so it works for varying
464 block sizes. only use strerror() not g_strerror() since the latter
465 depends on working GQuark and GSlice.
466 mem_error(): implemented in terms of fprintf and vfprintf.
468 * tests/slice-color.c: new program to test cache colorization effects.
470 * tests/slice-test.c: trade G_SLICE_CONFIG_ALWAYS_FREE for 0 duration
471 G_SLICE_CONFIG_WORKING_SET_MSECS.
473 2005-12-17 Matthias Clasen <mclasen@redhat.com>
475 * glib/goption.c (parse_short_option): Set an error in all
476 failure cases. (#324332, Tim-Philipp Müller)
478 2005-12-17 Sebastian Wilhelmi <seppi@seppi.de>
480 * glib/gatomic.c: Fix memory barrier position in g_atomic_int_get
481 and g_atomic_pointer_get. Add g_atomic_int_set and
482 g_atomic_pointer_set implementations for the !DEFINE_WITH_MUTEXES &&
483 G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case, as well as defining them
484 as functions (additionally to the macros in the header) for the
485 !G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case.
487 2005-12-16 Matthias Clasen <mclasen@redhat.com>
489 * glib/gmem.c (g_allocator_new): Don't return a pointer to
490 a const struct, since apps expect to be able to modify it.
491 (#324179, J. Ali Harlow)
493 Tue Dec 13 10:13:32 2005 Tim Janik <timj@imendio.com>
495 * glib/gatomic.h: added g_atomic_pointer_set() and g_atomic_int_set()
496 for systems where the initialization of atomic variables requires a
497 write memory barrier.
499 2005-12-09 Matthias Clasen <mclasen@redhat.com>
501 * configure.in: Bump version
503 * === Released 2.9.1 ===
505 2005-12-09 Alexander Larsson <alexl@redhat.com>
509 Use G_GNUC_WARN_UNUSED_RESULT on list functions that return
512 2005-12-08 Matthias Clasen <mclasen@redhat.com>
516 2005-12-07 Martyn Russell <martyn@imendio.com>
518 * glib/gasyncqueue.c:
519 - Call g_queue_insert_sorted() instead of duplicating the code.
520 - Call g_queue_sort() instead of duplicating the code.
521 - Invert sort function results to make sure the same sort function
522 gives the same results across glist, gslist, gqueue and
525 * tests/asyncqueue-test.c:
526 - Updated the sort function to reflect the example in the
527 documentation for gasyncqueue.c.
529 2005-12-07 Martyn Russell <martyn@imendio.com>
531 * docs/reference/glib/glib-sections.txt:
532 * docs/reference/glib/tmpl/linked_lists_double.sgml:
533 * docs/reference/glib/tmpl/linked_lists_single.sgml:
536 - Added g_list_insert_sorted_with_data () and
537 g_slist_insert_sorted_with_data ().
538 - Removed the extra check in g_list_sort() and g_slist_sort() for
539 GCompareDataFunc vs. GCompareFunc.
541 2005-12-07 Tor Lillqvist <tml@novell.com>
543 * glib-gettextize.in: Look up prefix at run-time on Win32,
544 assuming the standard directory structure with glib-gettextize in
547 2005-12-06 Matthias Clasen <mclasen@redhat.com>
549 * glib/gmem.h: Exempt GTK+ from the mem chunk deprecation, since
550 we need GTK+ 2.8 to compile against GLib 2.10.
552 2005-12-05 Manish Singh <yosh@gimp.org>
554 * tests/run-collate-test.sh: set LC_COLLATE, not LANG, to be sure
555 to override any user settings.
557 2005-12-05 Matthias Clasen <mclasen@redhat.com>
559 * configure.in: Revert an accidential commit
561 2005-12-05 Matthias Clasen <mclasen@redhat.com>
563 * glib/gutf8.c: Documentation updates. (#323291, Morten Welinder)
565 * tests/Makefile.am (TESTS_ENVIRONMENT):
566 * tests/gobject/Makefile.am (TESTS_ENVIRONMENT): Set
567 MALLOC_CHECK_ and MALLOC_PERTURB_
569 * tests/run-collate-tests.sh: Run the collation tests explicitly
570 in en_US locale. (#320463)
572 * glib/gmem.h: Really deprecate GMemChunk.
579 * glib/gstrfuncs.c: Add versioned deprecation docs.
581 2005-12-05 Martyn Russell <martyn@imendio.com>
583 * docs/reference/glib/glib-sections.txt:
584 * glib/gasyncqueue.[ch]:
585 - Added support for sorting async queues by with _push_sorted(),
586 _push_sorted_unlocked(), _sort() and _sort_unlocked() (#323047).
589 * tests/asyncqueue-test.c:
590 - Added test case for gasyncqueue.c
592 Mon Dec 5 15:53:20 2005 Tim Janik <timj@imendio.com>
594 * glib/gslice.c: implement chain walking for arbitrary ->next pointer
595 offsets in g_slice_free_chain_with_offset() based on a patch by behdad
596 in bug 323178. moved time consuming logic from g_slice_free() out of
597 the inner loop, so g_slice_free_chain_with_offset() provides a real
598 performance benefit over g_slice_free1() now.
600 * glib/gslice.h: renamed g_slice_free_chain() to
601 g_slice_free_chain_with_offset(). implemented g_slice_free_chain() as
602 a type-safe macro as suggested in bug 323178.
603 simplified the macro implementation of g_slice_free() and implemented
604 it in a type safe manner for all compliers as suggested by Morten
605 Welinder <mortenw@gnome.org>.
610 * glib/glib.symbols: s/g_slice_free_chain/g_slice_free_chain_with_offset/
612 2005-12-05 Matthias Clasen <mclasen@redhat.com>
614 * glib/gasyncqueue.c: Add some docs.
616 * tests/libmoduletestplugin_a.c: Fix compiler warnings.
618 * glib/gatomic.c: In the ia64 implementation, use
619 __sync builtin without _si or _di suffix. (#321229,
620 Stanislav Brabec, patch by Andreas Schwab)
622 2005-12-04 Behdad Esfahbod <behdad@gnome.org>
624 * glib/gslice.h: Remove comma at the end of enum.
626 2005-12-04 Matthias Clasen <mclasen@redhat.com>
628 Handle multiple user names with the same UID better.
629 (#319535, Laszlo Peter)
631 * glib/gutils.c (g_get_any_init_do): When determining user
632 data, first look up $LOGNAME. If the UID doesn't match
633 getuid(), fall back to the current behaviour of looking
634 up the user data based on getuid().
636 2005-12-04 Matthias Clasen <mclasen@redhat.com>
638 * glib/gmacros.h (G_GNUC_WARN_UNUSED_RESULT): Add a macro
639 to make gcc warn if a function result is ignored. (#145466,
640 Arjan van de Ven, Alex Larsson)
642 * glib/gmem.h: Add the new attribute to g_realloc and
645 2005-12-03 Matthias Clasen <mclasen@redhat.com>
648 * glib/gcache.h: Deprecate g_cache_value_foreach. (#322956,
651 * glib/gmappedfile.c: Make mapping of empty files
654 * glib/gfileutils.c: Don't fork a new process just to
655 fix the permissions of the created temp file. (#321318,
656 Alexis S. L. Carvalho)
658 2005-12-02 Matthias Clasen <mclasen@redhat.com>
660 * README.in: Add a note about Solaris threads.
663 * configure.in: Remove support for Solaris threads.
664 (#136971, Sebastian Wilhelmi, patch by Andrew Paprocki)
666 2005-12-02 Matthias Clasen <mclasen@redhat.com>
668 * configure.in: Define G_GUINT64_CONSTANT in analogy to
669 G_GINT64_CONSTANT. (#322568, Andrew Paprocki)
671 2005-12-02 Matthias Clasen <mclasen@redhat.com>
673 * glib/gslice.c: Win32 portability fixes and C99ism removal,
674 pointed out by Kazuki Iwamoto. (#323052)
676 2005-12-02 Matthias Clasen <mclasen@redhat.com>
678 * glib/gatomic.c (g_atomic_int_add): Add a missing volatile for
679 the IA64 implementation.
681 Fri Dec 2 16:18:09 2005 Tim Janik <timj@imendio.com>
683 * glib/gslice.c: some naming and type size fixups.
685 Fri Dec 2 13:08:58 2005 Tim Janik <timj@gtk.org>
687 * glib/gslice.h (g_slice_free): reworked GCC-specific type-safe macro
688 variant into something less verbose, but digestible for gcc-3.4.
690 Fri Dec 2 10:55:07 2005 Tim Janik <timj@imendio.com>
692 * tests/slice-test.c: extended to perform the benchmarking on the old
693 memchunk code if 'O' is selected.
695 * tests/memchunks.c: new file which contains the old GLib mem chunks
696 implementation with prefix old_mem_chunk_.
698 * tests/Makefile.am: added memchunks.c
700 Fri Dec 2 00:16:59 2005 Tim Janik <timj@imendio.com>
702 * glib/gslice.c: improved gettimeofday() timer resolution by maintaining
703 acceptable load distribution of the syscall.
704 (allocator_get_magazine_threshold): reduce minimum/single-thread magazine
705 sizes to approximately page_size/2.5 to avoid excessive startup allocations.
707 2005-12-01 Matthias Clasen <mclasen@redhat.com>
709 * tests/gobject/ifacecheck.c (test_iface_base_init): Don't
710 double-free base interface members.
712 * tests/Makefile.am: Remove duplicate variable.
714 Thu Dec 1 17:32:46 2005 Tim Janik <timj@imendio.com>
716 * glib/gslice.[hc]: new slice allocator implementation.
718 * tests/slice-test.c: added random slice allocation test.
720 * glib/gthread.[hc]: removed newly added private thread mem API.
722 * glib/gthreadinit.h:
725 * glib/gmem.c: divided glib threading initialisation into three phases,
726 initialisation where private keys and messaging are not available (only
727 needed by gmem.c), initialisation without messaging but private keys
728 available (gslice.c, gmessage.c), and full fledged initialisers that
729 server the rest of glib. initialisation functions got renamed to reflect
730 the limitations of their corresponding phases.
732 * glib/gmem.c: removed memchunk code, defer allocations to
735 * glib/gmem.[hc]: removed g_slice_* skeletons.
737 * glib/glib.symbols: added g_slice_* symbols.
739 * configure.in: check for availability of posix_memalign(3), memalign(3)
742 * glib/Makefile.am: added gslice.[hc].
744 2005-12-01 Tor Lillqvist <tml@novell.com>
746 * glib/gstdio.c (g_stat): In the Win32 implementation, strip
747 trailing slash(es) for non-root folders. stat() fails if non-root
748 folders are specified with trailing slashes. It's too much hassle
749 to demand that callers strip such slashes themselves, especially
750 as it is easy to get it wrong and strip the slash of a root
752 (g_rename): On NT-based Windows, use MoveFileEx() with
753 MOVEFILE_REPLACE_EXISTING to better match Unix behaviour.
755 2005-11-28 Matthias Clasen <mclasen@redhat.com>
757 Fix G_STMT_START / G_STMT_END on Solaris. (#321972,
760 * configure.in: Check whether do { } while (0) works.
762 * glib/gmacros.h: Use do { } while (0) for G_STMT_START /
763 G_STMT_END if it works.
765 2005-11-28 Matthias Clasen <mclasen@redhat.com>
767 * glib/gthread.c (g_static_rw_lock_wait, g_static_rw_lock_signal):
768 * glib/gnode.c (g_node_depth_traverse_level):
769 * glib/gmem.c (g_allocator_new):
770 * glib/ghash.c (g_hash_table_unref, g_hash_table_destroy)
771 (g_hash_table_foreach_remove_or_steal): Silence compiler
774 2005-11-27 Matthias Clasen <mclasen@redhat.com>
776 * glib/gunicollate.c (g_utf8_collate_key_for_filename):
777 Don't read beyond len. (#322520, Christian Persch)
779 Wed Nov 23 17:34:01 2005 Tim Janik <timj@imendio.com>
781 * glib/gdataset.c: access datalist flags via atomic pointer access
782 functions, instead of acquiring the dataset lock. this is faster and
783 also matches the atomic pointer readouts in gdatalistprivate.h and
784 direct pointer modifications required by gobject.c.
786 Wed Nov 23 13:35:31 2005 Tim Janik <timj@gtk.org>
788 * glib/gdataset.c: streamlined initialization code somewhat,
789 removed GData node cache.
791 2005-11-22 Matthias Clasen <mclasen@redhat.com>
793 * glib/ghash.c (g_hash_table_ref):
794 (g_hash_table_unref): Mark these as new API.
796 Avoid double locking in g_intern_string (#322133,
799 * glib/gdataset.c (g_quark_from_string_internal): New
800 internal function which factors out common parts of
801 g_quark_from[_static]_string.
802 (g_quark_from_string, g_quark_from_static_string):
803 Use g_quark_from_string_internal.
804 (g_intern_string, g_intern_static_string): Use
805 g_quark_from_string_internal, and only take the
807 (g_quark_new): Don't store the strings shifted by -1
808 in the g_quarks array.
809 (g_quark_to_string): Adapt to the previous change.
811 Tue Nov 22 14:04:26 2005 Tim Janik <timj@imendio.com>
815 g_hash_table_new_full(): create hash tables with a ref count of 1.
816 g_hash_table_ref(): atomically ref_count+=1
817 g_hash_table_unref(): atomically ref_count-=1, destroys hash table
818 when refcount reaches 0.
819 g_hash_table_destroy(): just destroy keys and values, unref by 1.
820 g_hash_table_insert():
821 g_hash_table_replace(): assert ref_count>0.
824 * glib/gatomic.c: added 'volatile' qualifier to all atomic pointer and
827 2005-11-20 Behdad Esfahbod <behdad@gnome.org>
829 * glib/guniprop.c (g_unichar_get_mirror_char): Remove unused
830 variables. (#321984, Andrew Paprocki)
832 2005-11-18 Matthias Clasen <mclasen@redhat.com>
834 * configure.in: Bump version
836 2005-11-17 Matthias Clasen <mclasen@redhat.com>
838 * === Released 2.9.0 ===
840 * glib/glib.symbols: Updates
842 * glib/gutf8.c (g_utf8_offset_to_pointer): Add some performance
843 hints to the docs. (Paolo Borelli)
847 * glib/gthread.c (g_thread_foreach): Mark as new api.
849 * README.in: Updates.
853 * glib/gdate.c (g_date_set_time_t):
854 (g_date_set_time_val): Add functions to set a date from
855 a time_t and from a GTimeVal.
856 (g_date_set_time): Deprecate. (#314067, Roger Leigh)
859 * tests/utf8-pointer.c: Unit tests for g_utf8_pointer_to_offset
860 and g_utf8_offset_to_pointer.
862 * glib/gutf8.c (g_utf8_pointer_to_offset)
863 (g_utf8_offset_to_pointer): Handle negative offsets, and use
864 "stutter stepping" for going backwards. (#320638, Larry
869 * glib/gthread.c: const correctness fixes, found
870 by Arjan van de Ven and gcc.
872 2005-11-16 Behdad Esfahbod <behdad@gnome.org>
874 * docs/reference/glib/tmpl/unicode.sgml: Correct typo on mentioning
875 @G_UNICODE_BREAK_UNKNOWN that should be #G_UNICODE_BREAK_UNKNOWN.
877 2005-11-09 Behdad Esfahbod <behdad@gnome.org>
879 * glib/guniprop.c: Use bit hacks instead when checking a general
880 category value against multiple values.
882 * glib/gutf8.c: Change ISO10646 to Unicode in docs.
884 2005-11-10 Simos Xenitellis <simos@gnome.org>
886 * configure.in: Added tt (Tatar) to ALL_LINGUAS.
888 2005-11-05 Tor Lillqvist <tml@novell.com>
890 * glib/giowin32.c (g_io_win32_set_flags): Don't call g_warning().
891 (#320688) Instead set error and return failure.
893 2005-11-07 Matthias Clasen <mclasen@redhat.com>
895 * glib/gmain.c (g_main_dispatch): Don't call cb_funcs->unref
896 while holding the context lock. (#320886, Andy Wingo)
898 2005-11-07 Billy Biggs <vektor@dumbterm.net>
900 * tests/refcount/Makefile.am: Add a missing reference to libglib
901 in the LDADD for this test.
903 2005-11-07 Matthias Clasen <mclasen@redhat.com>
905 * glib/gmem.c (g_allocator_new): We need to set n_preallocs to a
906 nonzero value, otherwise GTK+ 2.8 breaks when compiled against
907 GLib 2.9. (#320755, Luca Ferretti)
909 2005-11-04 Matthias Clasen <mclasen@redhat.com>
911 * glib/Makefile.am: Apply a patch from Behdad Esfahbod to
912 use a faster lookup table for g_unichar_get_mirror_char().
913 * glib/gmirroringtable.h: The new table.
914 * glib/gunichartables.h: Remove bidi_mirroring_table.
915 * glib/gen-unicode-tables.pl: Don't generate the mirroring
917 * glib/glib-mirroring-tab/*: Sources for the program
918 which generated gmirroringtable.h.
920 * glib/glist.c: Avoid some code duplication.
922 * glib/gscanner.h: Include gdataset.h. (#320322)
924 * glib/gdebug.h: add new GDebugFlag for fatal_criticals
925 * glib/gmessages.c: (_g_debug_init): handle G_DEBUG=fatal_criticals,
926 to help find critical warnings in applications. (#320017,
929 2005-11-02 Tor Lillqvist <tml@novell.com>
931 * glib/glib.symbols: Remove large amount of trailing whitespace
932 from one line. Remove #ifdef G_UNDEFINED from around
933 g_hash_table_new and g_hash_table_new_full symbols, what was the
934 purpose of that? Add the g_slice functions.
936 Tue Nov 1 16:24:20 2005 Tim Janik <timj@imendio.com>
938 * glib/gmem.[hc]: prepared deprecation of GMemChunk and GAllocator.
939 added g_slice_*() API to allocate and cache small bits of memory.
940 an actuall allocator implementation for g_slice_*() is still pending.
942 * glib/gthread.[hc]: changes from a patch by Matthias Clasen.
943 changed GRealThread list to use in-structure *next; fields instead
944 of GSList, in order for thread iteration to not depenend on g_slice_*()
946 _g_thread_mem_private_get():
947 _g_thread_mem_private_set(): added accessors for private memory,
948 needed because the ordinary GPrivate implementation relies on GArray
949 and GSList and therefore indirectly on working g_slice_*() allocations.
952 g_thread_foreach(): new public API function to loop over all existing threads.
966 removed GAllocator and free list usages and accompanying locks.
967 use g_slice_*() API to allocate and cache small bits of memory.
969 * glib/ghook.h: removed GMemChunk field from public API.
972 * glib/glist.h: deprecate allocator API, provide _free1() for consistency.
974 * glib/gnode.h: deprecate allocator API.
976 * glib/gmain.c: reordered GPollRec fields so g_slice_free_chain() can
977 be used for poll rec lists.
979 * glib/grel.c: removed mem chunk usage, and allocated tuples via g_slice_*().
980 g_relation_destroy(): free all tuples from the all_tuples hash table,
981 this effectively maintains the life time track keeping of tuples.
982 g_relation_delete_tuple(): free tuples which are removed from the
983 all_tuples hash table. this fixes a temporary leak that was present
984 in the memchunk code until the destruction of the relation.
986 2005-10-29 Matthias Clasen <mclasen@redhat.com>
988 * tests/convert-test.c: Add some tests for conversions between
989 UTF-8, UCS-4 and UTF-16.
991 * glib/gutf8.c (g_utf8_to_ucs4, g_utf8_to_utf16): Fix handling
992 of len == -1, noticed by Morten Welinder.
994 2005-10-27 Erdal Ronahi <erdal.ronahi@gmail.com>
996 * configure.in: Added ku (Kurdish) to ALL_LINGUAS
998 2005-10-26 Matthias Clasen <mclasen@redhat.com>
1000 * glib/gutf8.c (g_ucs4_to_utf8): Don't set items_read twice
1001 in the error case, and add some documentation. (#319806, Morten
1004 2005-10-19 Manish Singh <yosh@gimp.org>
1006 * configure.in: Use AC_CHECK_FUNCS for _NSGetEnviron, to get the
1007 config.h symbol automatically. Fixes bug #313731.
1009 2005-10-19 Tor Lillqvist <tml@novell.com>
1011 * glib/gwin32.c (g_win32_get_package_installation_directory):
1012 Return a g_strdup()ed copy of the value stored in the hash table,
1013 so that it can be g_free()d without leaving a dangling pointer in
1014 the hash table. (#319232)
1016 2005-10-06 Matthias Clasen <mclasen@redhat.com>
1018 * glib/gunicollate.c (g_utf8_collate_key_for_filename): Handle
1019 all-zero sequences correctly. (#317930, Sebastien Bacher)
1021 2005-10-05 Matthias Clasen <mclasen@redhat.com>
1023 * glib/gmarkup.c (g_markup_escape_text): Doc addition
1025 * Makefile.am: Add Collation tests.
1027 * tests/collate/*: Inputs and expected outputs for collation tests.
1029 * tests/run-collate-tests.sh: Script to run collation tests.
1031 * tests/unicode-collate.c (main): Rework slightly to make
1032 it usable in unit tests. Also test g_utf8_collate_key_for_filename().
1034 2005-10-01 Behdad Esfahbod <behdad@gnome.org>
1036 * docs/reference/glib/tmpl/unicode.sgml:
1037 * glib/gen-unicode-tables.pl:
1039 * glib/gunichartables.h:
1041 * tests/casefold.txt:
1042 * tests/casemap.txt: Updated to Unicode 4.1. There are five new
1043 GUnicodeBreakType types. That may break some applications, like
1046 2005-09-26 Matthias Clasen <mclasen@redhat.com>
1048 * glib/gstrfuncs.c (g_ascii_strtoull): Add details to the
1049 docs. (#314393, Matthew F. Barnes)
1051 * glib/glib.symbols:
1052 * glib/gprintf.h: Remove g_snprintf() and g_vsnprintf(), since
1053 they are already declared in glib.h. This doesn't break documented
1054 use of gprintf.h, but should probably be pointed out in the
1055 release notes for 2.10. (#314232, Behdad Esfahbod)
1057 Tue Sep 20 13:16:04 2005 Tim Janik <timj@imendio.com>
1059 * glib/gpattern.c (g_pattern_ph_match): applied significant recursion
1060 complexity optimization, based on a patch from Matthias Clasen.
1062 * tests/patterntest.c: more tests, mostly from matthias.
1064 2005-09-20 Matthias Clasen <mclasen@redhat.com>
1066 * glib/gqueue.c (g_queue_insert_sorted): Correct the docs.
1067 (#316703, Mark Drago)
1069 Mon Sep 19 17:23:23 2005 Tim Janik <timj@imendio.com>
1071 * glib/gpattern.c: applied a patch from matthias which checks on the
1072 upper bound of GPatternSpec length to optimize matches.
1075 * tests/patterntest.c: added more match cases.
1077 2005-09-16 Tor Lillqvist <tml@novell.com>
1079 * glib/gstrfuncs.c (g_ascii_strcasecmp, g_ascii_strncasecmp): Add
1080 warning to doc comment that these functions should not be used on
1081 encodings like CP932.
1083 2005-09-14 Matthias Clasen <mclasen@redhat.com>
1085 * tests/keyfile-test.c: Add a test for grup names of length 1.
1087 * glib/gkeyfile.c (g_key_file_line_is_group): Accept group names
1088 of length 1. (#316309)
1090 2005-09-12 Matthias Clasen <mclasen@redhat.com>
1092 * glib/gmarkup.c (g_markup_escape_text): Clarify docs.
1094 2005-09-11 Sebastian Wilhelmi <seppi@seppi.de>
1096 * tests/refcount/Makefile.am (INCLUDES): Link the the refcount
1097 tests to the system thread library $(G_THREAD_LIBS). Fixes #313744
1100 2005-09-11 Kjartan Maraas <kmaraas@gnome.org>
1102 * glib/gmain.c: (g_child_watch_prepare), (g_child_watch_check),
1103 (child_watch_helper_thread): Remove some dead code. Closes
1106 2005-09-07 Tor Lillqvist <tml@novell.com>
1108 * glib/Makefile.am: Create also a console version of the
1109 gspawn-win32-helper program, gspawn-win32-helper-console.exe.
1110 It's otherwise identical to gspawn-win32-helper.exe, except marked
1111 as a console application (linked without the -mwindows option).
1113 * glib/gspawn-win32.c (do_spawn_directly, do_spawn_with_pipes):
1114 Drop the dont_wait parameter. Its truth value correlated 100% with
1115 the NULLness of the exit_status parameter anyway, so it's enough
1116 to check whether exit_status is NULL. Invert the sense of the
1117 dont_return_handle parameter and rename it to do_return_handle, to
1118 make the code easier to read by avoiding double negations.
1120 (g_spawn_sync_utf8, g_spawn_async_with_pipes_utf8): Modify calls
1121 to do_spawn_with_pipes() accordingly.
1123 (do_spawn_with_pipes): If we have a console, use the console
1124 version of the helper program, otherwise use the GUI one. This
1125 avoids extra console windows opening up in some situations. (In
1126 case a console application uses the GUI gspawn-win32-helper.exe to
1127 spawn another console application we would get a separate console
1128 for the spawned console application).
1130 * glib-zip.in: Distribute also gspawn-win32-helper-console.exe.
1132 2005-09-05 Matthias Clasen <mclasen@redhat.com>
1134 * glib/gmappedfile.c (g_mapped_file_new): Report an error
1135 if the file is too large. (#315275, Kjartan Maraas)
1137 * glib/gkeyfile.c (g_key_file_load_from_fd): The return value
1138 of read() is signed. (#315273, Kjartan Maraas)
1140 2005-08-31 Tor Lillqvist <tml@novell.com>
1142 * glib/gutils.h: Wrapping atexit() is a bad idea on Windows, where
1143 the EXE and each DLL have their own atexit function chains.
1145 #define g_atexit as atexit instead. This means it has a
1146 better chance of doing what the caller wants. For instance,
1147 gtkhtml calls g_atexit() registering a function in gtkhtml
1148 itself. This caused a crash when g_atexit() was implemented as a
1149 function in the GLib DLL. The gtkhtml DLL was already unloaded by
1150 the time the GLib DLL got unloaded.
1152 * glib/gutils.c: #undef the #define mentioned above, to also get a
1153 real g_atexit() into the DLL for backward compatibility. Document
1154 the Windows behaviour of g_atexit(), and document the varying ways
1155 atexit() can behave in the context of dynamically loaded modules
1158 2005-08-31 Matthias Clasen <mclasen@redhat.com>
1160 * glib/glib.symbols:
1162 * glib/gdataset.c: Add string interning functions.
1164 2005-08-28 Matthias Clasen <mclasen@redhat.com>
1166 * glib/giochannel.c: Unify some near-duplicate strings. (#314654,
1169 2005-08-26 Matthias Clasen <mclasen@redhat.com>
1171 * configure.in: Bump version to 2.9.0
1173 2005-08-25 Tor Lillqvist <tml@novell.com>
1175 Make also the g_spawn*() functions take parameters in the GLib
1176 file name encoding, i.e. UTF-8, on Windows. Has no impact on Unix
1177 API or ABI. Like the other GLib API that was earlier changed to
1178 use UTF-8 on Windows, the names of the functions that take UTF-8
1179 have _utf8 suffixes added by using preprocessor macros in the
1180 header file. The old names are kept for functions with the old
1181 behaviour, taking parameters in the system codepage, for DLL ABI
1184 * glib/gspawn.h: On Win32 add the suffix _utf8 to the names of the
1185 g_spawn*() functions.
1187 * glib/gspawn-win32.c: Use wide-char API on NT-based
1188 Windows. Convert parameters from UTF-8 to wide chars (NT) or
1189 system codepage (Win9x) and call the C library _wspawn*() or
1190 spawn*() functions respectvely. Add DLL ABI stability versions
1191 that take parameters in the system codepage.
1193 * glib/gspawn-win32-helper.c: On NT-based Windows use the
1194 wide-char versions of argv and envp, and use wide-char API to
1195 change directory and spawn the program to run. Remove the verbose
1196 debugging output, it was too complex to modify for the wide-char
1197 features. (Just add temporary debugging printouts if needed, no
1198 need to have them permanently in the source.)
1200 * glib/gspawn.c: Corresponding documentation updates.
1202 * glib/glib.symbols: Corresponding changes: Mark the ABI stability
1203 symbols as PRIVATE, add the new _utf8-suffixed ones.
1205 2005-08-24 Stepan Kasal <kasal@ucw.cz>
1207 * glib/gtypes.h (G_MININT64): Cast the constant to gint64; it is
1208 guint64 otherwise and that can produce warnings about comparison
1209 between signed and unsigned.
1211 2005-08-23 Matthias Clasen <mclasen@redhat.com>
1213 * glib/gutils.c: Fix the crt_externs.h include.
1215 2005-08-23 Stepan Kasal <kasal@ucw.cz>
1217 * NEWS: Fix spelling of my first name.
1219 2005-08-23 Matthias Clasen <mclasen@redhat.com>
1223 * === Released 2.8.1 ===
1227 2005-08-20 Hans Breuer <hans@breuer.org>
1229 * glib/makefile.msc.in : link with ws2_32.lib
1231 2005-08-18 Tor Lillqvist <tml@novell.com>
1233 * configure.in: Check for <sys/wait.h>
1235 * glib/gbacktrace.c: Include <sys/wait.h> on if HAVE_SYS_WAIT_H.
1237 2005-08-18 Ross Burton <ross@burtonini.com>
1240 Optimise single-character insertions.
1245 * tests/string-test.c:
1246 Add tests for new optimisation, and fix a leak.
1248 2005-08-17 Matthias Clasen <mclasen@redhat.com>
1250 * configure.in: Check for crt_externs.h and _NSGetEnviron.
1252 * glib/gutils.c: On Darwin, include crt-externs.h and
1253 define environ using _NSGetEnviron(). (#313731)
1255 2005-08-16 Stepan Kasal <kasal@ucw.cz>
1257 * glib/gutils.c (g_get_any_init): Move the body of the big if...
1258 (g_get_any_init_do): ... to this new function.
1259 (g_get_any_init): Declare as inline.
1260 (g_get_any_init_locked): New inline function, does the locking.
1261 Make use of these two throughout the code.
1263 2005-08-15 Matthias Clasen <mclasen@redhat.com>
1265 * glib/gbacktrace.c (g_on_error_stack_trace): Wait for
1266 the child process and then simply return. This makes
1267 The "S" option work as documented in g_on_error_query().
1268 (#313125, Matthew F. Barnes)
1270 * glib/gunicode.h: Update the link to Unicode category
1271 values. (#313369, Behnam Esfahbod)
1273 * glib/gqueue.c (g_queue_find_custom): Clarify docs
1274 a little. (#311727, Tristan van Berkom)
1276 * glib/abicheck.sh, gobject/abicheck.sh: Make the
1277 check work on ia64 too, where some symbols ended up
1278 in yet another section.