1 2006-03-07 Matthias Clasen <mclasen@redhat.com>
3 * === Released 2.10.1 ===
7 * glib/gutf8.c (_g_charset_get_aliases): Match the prototype
8 in gconvert.c to fix build problems in NetBSD. (#333651,
11 2006-03-06 Anders Carlsson <andersca@imendio.com>
14 Make .dylib the shared library suffix on darwin.
15 (#322476, Vladimir Panov)
17 2006-03-06 Anders Carlsson <andersca@imendio.com>
20 Skip checking thread flags on Darwin. Fix suggested by Manish Singh.
21 (#314794, Gregor Riepl)
23 2006-03-01 Tor Lillqvist <tml@novell.com>
25 * glib/gutils.c (g_listenv): Separate implementation on Win32: Use
26 the wide character API on NT-based Windows. Return UTF-8 strings.
28 * glib/glib.symbols: Don't mark g_listenv as PRIVATE, as that
29 meant it wasn't present in the import library. PRIVATE is used
30 only for the backwards-compatibility DLL ABI stability hacks.
32 2006-02-24 Matthias Clasen <mclasen@redhat.com>
34 * configure.in: Bump version
36 * === Released 2.10.0 ===
38 * README.in: Mention the slice allocator and GInitiallyUnowned.
42 * configure.in: Bump version to 2.10.0
44 * tests/utf8-pointer.c: Add tests for g_utf8_strlen().
46 * glib/gutf8.c: Fix boundary cases in g_utf8_strlen().
49 2006-02-23 Kjartan Maraas <kmaraas@gnome.org>
51 * tests/completion-test.c: (main): Plug leaks reported by
53 * tests/convert-test.c: (test_iconv_state), (test_one_half):
55 * tests/patterntest.c: (test_compilation): Same
56 * tests/shell-test.c: (do_argv_test): Same.
57 * tests/unicode-caseconv.c: (main): Same.
58 * tests/uri-test.c: (run_to_uri_tests): Same.
61 2006-02-19 Martyn Russell <martyn@imendio.com>
63 * tests/asyncqueue-test.c:
65 * tests/slist-test.c: Updated to test _sort, _sort_with_data,
66 _insert_sorted and _insert_sorted_with_data API.
68 2006-02-18 Matthias Clasen <mclasen@redhat.com>
70 * tests/gobject/Makefile.am: Add paramspec-test
72 * tests/gobject/paramspec-test.c: Some GParamSpec tests.
74 * tests/gobject/gvalue-test.c: Add more tests.
76 2006-02-17 Kang Jeong-Hee <keizie@gmail.com>
78 * glib/gutf8.c (g_utf8_find_prev_char): Correct documentation typo.
80 2006-02-15 Sebastian Wilhelmi <seppi@seppi.de>
82 * glib/gthreadpool.c: Fix deadlock when signalling the thread
83 which freed a thread pool (#331110, Chris Wilson).
85 Tue Feb 14 17:00:43 2006 Tim Janik <timj@imendio.com>
87 * glib/gslice.c: only define _XOPEN_SOURCE to 600 to get at
88 posix_memalign() in case we actually are going to use it, because
89 we detected a compliant implementation (#328997).
91 * configure.in (enable_included_printf): don't include malloc.h when
92 testing for posix_memalign() funcitonality, since this may break the
93 test on some systems (#328997).
95 2006-02-14 Matthias Clasen <mclasen@redhat.com>
97 * glib/gstrfuncs.c: Improve docs.
99 2006-02-11 Matthias Clasen <mclasen@redhat.com>
101 * configure.in: Bump version
103 2006-02-10 Matthias Clasen <mclasen@redhat.com>
105 * === Released 2.9.6 ===
109 2006-01-31 Behdad Esfahbod <behdad@gnome.org>
111 * docs/reference/glib/tmpl/relations.sgml (g_relation_insert): Specify
112 the type of varargs arguments. (bug #317679)
114 2006-01-31 Matthias Clasen <mclasen@redhat.com>
116 * autogen.sh: Touch README and INSTALL here to pacify
117 automake. (#329124, Kjartan Maraas, Tim Janik)
119 Tue Jan 31 16:45:03 2006 Tim Janik <timj@imendio.com>
121 * glib/gdataset.c: carry out all dtalist pointer accesses atomically,
122 some missing cases where pointed out by Sebastian Wilhelmi.
124 Tue Jan 31 12:33:48 2006 Tim Janik <timj@imendio.com>
126 * configure.in: generate build/win32/vs8/Makefile, to fix build breakage.
128 2006-01-30 David Schleef <ds@schleef.org>
130 * glib/grand.c: Update URL
132 2006-01-29 Sven Herzberg <herzi@gnome-de.org>
134 * docs/reference/gobject/tmpl/param_value_types.sgml: tell how to
135 create container classes which are as flexible as a GValue is
137 2006-01-27 Matthias Clasen <mclasen@redhat.com>
139 * configure.in: Bump version
141 * === Released 2.9.5 ===
143 * glib/glib.symbols: Add g_mem_gc_friendly as an exported
148 2006-01-26 Matthias Clasen <mclasen@redhat.com>
150 * glib/gmem.c (g_mem_init_nomessage): Fix C99isms. (#328705,
153 Wed Jan 25 19:16:57 2006 Tim Janik <timj@imendio.com>
155 * fixed buglets reported by Jens Granseuer in #328254.
157 * configure.in: free the memory allocated in posix_memalign() tests.
159 * glib/gslice.c: spelling fixes.
161 Wed Jan 25 16:39:18 2006 Tim Janik <timj@imendio.com>
163 * glib/gslice.c: honour g_mem_gc_friendly settings when freeing
164 slices, make sure g_mem_gc_friendly is properly initialized.
166 * gmem.[hc]: ensure g_mem_gc_friendly is initialized from G_DEBUG upon
167 the first allocation. applied some branching optimizations.
169 * docs/macros.txt: reflected --enable-gc-friendly change and
170 described ENABLE_GC_FRIENDLY_DEFAULT as well as G_DEBUG=gc-friendly.
172 * configure.in: changed --enable-gc-friendly=yes to define
173 ENABLE_GC_FRIENDLY_DEFAULT.
175 * glib/garray.c: changed ENABLE_GC_FRIENDLY macro #ifdef-s to
176 if (G_UNLIKELY (g_mem_gc_friendly)).
179 * glib/ghash.c: removed ENABLE_GC_FRIENDLY code which is now taken
180 care of by g_slice_free1().
182 * tests/slice-test.c: fixed leaks, reported by Kjartan Maraas.
184 Tue Jan 24 17:49:36 2006 Tim Janik <timj@imendio.com>
186 * glib/gslice.c: only use posix_memalign() if it's known to work,
187 revert to memalign() otherwise.
189 * configure.in: check for broken posix_memalign() implementations
192 2006-01-24 Matthias Clasen <mclasen@redhat.com>
194 * tests/unicode-encoding.c: Use UTF-16LE as target encoding
195 on all little-endian systems. (#143380, Marc Moorcroft)
197 2006-01-23 Matthias Clasen <mclasen@redhat.com>
199 * configure.in: Change the shared libary extension for hpux-ia64
200 to so. (#328253, The Written Word)
202 Mon Jan 23 17:30:33 2006 Tim Janik <timj@imendio.com>
204 * glib/gutils.c (_g_getenv_nomalloc): wiped out all the wonderfull
205 G_OS_WIN32 code i wrote ;-[) after tml told me windows has getenv()
208 Mon Jan 23 16:46:20 2006 Tim Janik <timj@imendio.com>
210 * glib/gslice.c (slice_config_init): initialize GSlice config from
211 G_SLICE environment variable. we support G_SLICE=always-malloc
212 currently, which forces all g_slice_*() allocations to use the system
216 g_parse_debug_string(): added a note about not using g_malloc() here.
217 _g_getenv_nomalloc(): getenv() variant that doesn't use g_malloc or
218 g_slice. contains only guesswork in the WIN32 branch.
220 2006-01-18 Matthias Clasen <mclasen@redhat.com>
224 * === Released 2.9.4 ===
228 2006-01-17 Matthias Clasen <mclasen@redhat.com>
230 * glib/gkeyfile.c (g_key_file_get_groups): Correct a g_new0()
231 call. (#327421, Morten Welinder)
233 2006-01-17 Sebastian Wilhelmi <seppi@seppi.de>
235 * glib/gthreadpool.c: To avoid deadlocks get rid of the settings
236 G_LOCK. Use the unused_thread_queue lock instead. Change
237 g_thread_pool_thread_proxy such that threads only wait on
238 non-exlusive pools for at most a 1/2 second. Do not reorder tasks
239 due to superfluous tasks. Global tasks wait at most for
240 max-idle-time milliseconds. Make sure, that no task is woken up
241 twice for the same event via a wakeup_serial. This fixes #324228.
243 * tests/threadpool-test.c: Adapt test accordingly. Do not pass
244 invalid NULL into the thread pools. This as well fixes #327290.
246 2006-01-16 Matthias Clasen <mclasen@redhat.com>
248 * configure.in: Bump version
250 * === Released 2.9.3 ===
254 * glib/gthread.h (g_static_mutex_get_mutex_impl_shortcut):
255 Rework the strict aliasing fix to not break C++, pointed
256 out by Murray Cumming.
258 * glib/gasyncqueue.c (g_async_queue_push_sorted_unlocked):
259 Signal waiting threads, problem noticed by Christian Kellner.
261 2006-01-16 Matthias Clasen <mclasen@redhat.com>
263 Fix bug 326747, Alberto Ruiz:
265 * glib/gconvert.c (make_valid_utf8): Change this function to
266 replace unknown characters by the Unicode replacement character
267 instead of '?', and don't append "(invalid encoding)".
268 (g_filename_display_name, g_filename_display_basename): Document
269 how to determine if the filename was in an invalid encoding.
271 2006-01-14 Matthias Clasen <mclasen@redhat.com>
273 * glib/gtree.c: Replace the simple recursive implementation
274 by a nonrecursive, threaded implementation by Maurizio
277 2006-01-12 Matthias Clasen <mclasen@redhat.com>
279 * tests/asyncqueue-test.c (main): Fix the
280 assert to not always trigger. (#326558,
283 2006-01-11 Matthias Clasen <mclasen@redhat.com>
285 * tests/convert-test.c: Don't test items_read and
286 items_written in error cases where it is not set.
288 * tests/asyncqueue-test.c: Set error to NULL.
290 2006-01-08 Matthias Clasen <mclasen@redhat.com>
292 * glib/gtree.c: Break some long lines.
294 * tests/tree-test.c: Add more tests.
296 * glib/gslice.c: Include config.h first, pointed out
299 2005-12-20 Sven Herzberg <herzi@gnome-de.org>
301 * docs/reference/gobject/tmpl/gtype.sgml: explain that
302 G_TYPE_INSTANCE_GET_CLASS() does behave different during
305 2006-01-06 Matthias Clasen <mclasen@redhat.com>
307 * glib/gslice.c: Only define _XOPEN_SOURCE if we know
308 that we have posix_memalign(). (#323937, Bogdan Nicula)
310 2006-01-05 Matthias Clasen <mclasen@redhat.com>
312 * configure.in: Bump version
314 * === Released 2.9.2 ===
316 2006-01-05 Hans Breuer <hans@breuer.org>
318 * glib/makefile.msc.in : added gslice.obj
320 2006-01-05 Matthias Clasen <mclasen@redhat.com>
324 * glib/gmain.c: Clarify the documentation of source ids
325 a bit. (#325874, Dan Williams)
327 * configure.in: Fix another strict aliasing problem.
329 * glib/gthreadpool.c: Avoid use of varargs macro. (#325864,
332 2006-01-04 Manish Singh <yosh@gimp.org>
335 * glib/gstdio.h: don't macroized g_access, g_chdir, and g_unlink
336 either, since they have the same issue as g_rmdir. (Related to
339 Wed Jan 4 13:33:25 2006 Tim Janik <timj@gtk.org>
341 * glib/gslice.c (magazine_cache_pop_magazine): don't reverse chunk
342 order when creating magazines, so we hand out chunks with ascending
345 2006-01-04 Abel Cheung <maddog@linuxhall.org>
347 * configure.in: Added "ml" "zh_HK" to ALL_LINGUAS.
349 2006-01-03 Martyn Russell <martyn@imendio.com>
351 * docs/reference/glib/glib-sections.txt:
353 * glib/gthreadpool.[ch]:
354 - Added new API g_thread_pool_get_idle_time() and
355 g_thread_pool_set_idle_time(). (#324228).
357 * tests/threadpool-test.c:
358 - Updated test case to do thread pool sorting, thread pool with
359 no sorting and a thread pool with idle thread timeouts.
361 2006-01-03 Matthias Clasen <mclasen@redhat.com>
363 * glib/gmain.h: Add new functions here, too.
365 * glib/glib.symbols: Add new functions.
367 2005-12-20 Michael Meeks <michael.meeks@novell.com>
369 * glib/gmain.c (g_main_context_is_owner): new method
370 to determine if the current thread is the owner of the
373 2006-01-02 Matthias Clasen <mclasen@redhat.com>
377 * glib/gstdio.c (g_rmdir): Don't provide g_rmdir() as a macro
378 expanding to rmdir, since rmdir is not declared in a portable
379 system header we can include in gstdio.h. (#325249, Jani Monoses)
381 2006-01-01 Tor Lillqvist <tml@novell.com>
383 * glib/gspawn-win32.c (g_spawn_sync_utf8): Set the GIOChannels for
384 stdout and stderr to unbuffered. Otherwise the giochannel layer
385 will try to read from them regardless whether the
386 g_io_channel_win32_poll() call here has indicated
387 readability or not. (#325310)
389 2005-12-29 Matthias Clasen <mclasen@redhat.com>
391 * glib/gutils.c (glib_check_version): Fix a copy-and-paste error
392 in the docs. (#325273, Declan Naughton)
394 2005-12-27 Manish Singh <yosh@gimp.org>
396 * tests/run-collate-tests.sh: use LC_ALL instead of LC_COLLATE, to
397 make sure we really override things.
399 2005-12-27 Matthias Clasen <mclasen@redhat.com>
401 Fix #316221, Michal Benes, Stanislav Brabec;
403 * configure.in: Fix a strict aliasing problem in
404 g_static_mutex_get_mutex().
405 * glib/gthread.h: ...and in
406 g_static_mutex_get_mutex_impl_shortcut().
408 * glib/gdatasetprivate.h: Add a cast to silence compiler
409 warnings. (#321978, Andrew Paprocki)
411 Partial fix for bug #323937, Bogdan Nicula.
413 * configure.in: Check for malloc.h
414 * glib/gslice.c: Don't include malloc.h unconditionally.
416 2005-12-27 Manish Singh <yosh@gimp.org>
418 * tests/run-collate-tests.sh: export LC_COLLATE so it takes.
419 (#324950, Dan Yefimov)
421 2005-12-26 Matthias Clasen <mclasen@redhat.com>
423 * glib/gslice.c: On Win32, include process.h (#325015, Kazuki
426 2005-12-25 Matthias Clasen <mclasen@redhat.com>
428 * AUTHORS: Update my email
430 * tests/slice-test.c: Fix C99isms. (#324950, Dan Yefimov)
432 2005-12-21 Matthias Clasen <mclasen@redhat.com>
434 * README.in: Add some notes on when 'make check' may fail.
436 2005-12-20 Matthias Clasen <mclasen@redhat.com>
439 * glib/gthreadpool.h:
440 * glib/gthreadpool.c (g_thread_pool_set_sort_function): New function
441 to sort tasks pushed into a threadpool. (#324479, Martyn Russell)
443 * tests/threadpool-test.c: Test this.
445 Tue Dec 20 18:14:14 2005 Tim Janik <timj@imendio.com>
447 * glib/gslice.[hc]: added mem_error() and mem_assert() to test and
448 handle errors without depending on gmessage.c which might not be
449 setup when the error occours.
450 removed G_SLICE_CONFIG_ALWAYS_FREE config option, fixed the code so
451 always freeing can be achieved by adjusting the working set time to
452 0 with G_SLICE_CONFIG_WORKING_SET_MSECS.
453 added G_SLICE_CONFIG_COLOR_INCREMENT to test different color increments
454 (mainly 0 and 1). reduced the minimum block size to 128 bytes, to
455 minimize wastage if small amounts of differently sized structrues are
456 allocated, this does come at a performance cost of roughly 5% though.
457 fixed up block alignment calculation, so it works for varying
458 block sizes. only use strerror() not g_strerror() since the latter
459 depends on working GQuark and GSlice.
460 mem_error(): implemented in terms of fprintf and vfprintf.
462 * tests/slice-color.c: new program to test cache colorization effects.
464 * tests/slice-test.c: trade G_SLICE_CONFIG_ALWAYS_FREE for 0 duration
465 G_SLICE_CONFIG_WORKING_SET_MSECS.
467 2005-12-17 Matthias Clasen <mclasen@redhat.com>
469 * glib/goption.c (parse_short_option): Set an error in all
470 failure cases. (#324332, Tim-Philipp Müller)
472 2005-12-17 Sebastian Wilhelmi <seppi@seppi.de>
474 * glib/gatomic.c: Fix memory barrier position in g_atomic_int_get
475 and g_atomic_pointer_get. Add g_atomic_int_set and
476 g_atomic_pointer_set implementations for the !DEFINE_WITH_MUTEXES &&
477 G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case, as well as defining them
478 as functions (additionally to the macros in the header) for the
479 !G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case.
481 2005-12-16 Matthias Clasen <mclasen@redhat.com>
483 * glib/gmem.c (g_allocator_new): Don't return a pointer to
484 a const struct, since apps expect to be able to modify it.
485 (#324179, J. Ali Harlow)
487 Tue Dec 13 10:13:32 2005 Tim Janik <timj@imendio.com>
489 * glib/gatomic.h: added g_atomic_pointer_set() and g_atomic_int_set()
490 for systems where the initialization of atomic variables requires a
491 write memory barrier.
493 2005-12-09 Matthias Clasen <mclasen@redhat.com>
495 * configure.in: Bump version
497 * === Released 2.9.1 ===
499 2005-12-09 Alexander Larsson <alexl@redhat.com>
503 Use G_GNUC_WARN_UNUSED_RESULT on list functions that return
506 2005-12-08 Matthias Clasen <mclasen@redhat.com>
510 2005-12-07 Martyn Russell <martyn@imendio.com>
512 * glib/gasyncqueue.c:
513 - Call g_queue_insert_sorted() instead of duplicating the code.
514 - Call g_queue_sort() instead of duplicating the code.
515 - Invert sort function results to make sure the same sort function
516 gives the same results across glist, gslist, gqueue and
519 * tests/asyncqueue-test.c:
520 - Updated the sort function to reflect the example in the
521 documentation for gasyncqueue.c.
523 2005-12-07 Martyn Russell <martyn@imendio.com>
525 * docs/reference/glib/glib-sections.txt:
526 * docs/reference/glib/tmpl/linked_lists_double.sgml:
527 * docs/reference/glib/tmpl/linked_lists_single.sgml:
530 - Added g_list_insert_sorted_with_data () and
531 g_slist_insert_sorted_with_data ().
532 - Removed the extra check in g_list_sort() and g_slist_sort() for
533 GCompareDataFunc vs. GCompareFunc.
535 2005-12-07 Tor Lillqvist <tml@novell.com>
537 * glib-gettextize.in: Look up prefix at run-time on Win32,
538 assuming the standard directory structure with glib-gettextize in
541 2005-12-06 Matthias Clasen <mclasen@redhat.com>
543 * glib/gmem.h: Exempt GTK+ from the mem chunk deprecation, since
544 we need GTK+ 2.8 to compile against GLib 2.10.
546 2005-12-05 Manish Singh <yosh@gimp.org>
548 * tests/run-collate-test.sh: set LC_COLLATE, not LANG, to be sure
549 to override any user settings.
551 2005-12-05 Matthias Clasen <mclasen@redhat.com>
553 * configure.in: Revert an accidential commit
555 2005-12-05 Matthias Clasen <mclasen@redhat.com>
557 * glib/gutf8.c: Documentation updates. (#323291, Morten Welinder)
559 * tests/Makefile.am (TESTS_ENVIRONMENT):
560 * tests/gobject/Makefile.am (TESTS_ENVIRONMENT): Set
561 MALLOC_CHECK_ and MALLOC_PERTURB_
563 * tests/run-collate-tests.sh: Run the collation tests explicitly
564 in en_US locale. (#320463)
566 * glib/gmem.h: Really deprecate GMemChunk.
573 * glib/gstrfuncs.c: Add versioned deprecation docs.
575 2005-12-05 Martyn Russell <martyn@imendio.com>
577 * docs/reference/glib/glib-sections.txt:
578 * glib/gasyncqueue.[ch]:
579 - Added support for sorting async queues by with _push_sorted(),
580 _push_sorted_unlocked(), _sort() and _sort_unlocked() (#323047).
583 * tests/asyncqueue-test.c:
584 - Added test case for gasyncqueue.c
586 Mon Dec 5 15:53:20 2005 Tim Janik <timj@imendio.com>
588 * glib/gslice.c: implement chain walking for arbitrary ->next pointer
589 offsets in g_slice_free_chain_with_offset() based on a patch by behdad
590 in bug 323178. moved time consuming logic from g_slice_free() out of
591 the inner loop, so g_slice_free_chain_with_offset() provides a real
592 performance benefit over g_slice_free1() now.
594 * glib/gslice.h: renamed g_slice_free_chain() to
595 g_slice_free_chain_with_offset(). implemented g_slice_free_chain() as
596 a type-safe macro as suggested in bug 323178.
597 simplified the macro implementation of g_slice_free() and implemented
598 it in a type safe manner for all compliers as suggested by Morten
599 Welinder <mortenw@gnome.org>.
604 * glib/glib.symbols: s/g_slice_free_chain/g_slice_free_chain_with_offset/
606 2005-12-05 Matthias Clasen <mclasen@redhat.com>
608 * glib/gasyncqueue.c: Add some docs.
610 * tests/libmoduletestplugin_a.c: Fix compiler warnings.
612 * glib/gatomic.c: In the ia64 implementation, use
613 __sync builtin without _si or _di suffix. (#321229,
614 Stanislav Brabec, patch by Andreas Schwab)
616 2005-12-04 Behdad Esfahbod <behdad@gnome.org>
618 * glib/gslice.h: Remove comma at the end of enum.
620 2005-12-04 Matthias Clasen <mclasen@redhat.com>
622 Handle multiple user names with the same UID better.
623 (#319535, Laszlo Peter)
625 * glib/gutils.c (g_get_any_init_do): When determining user
626 data, first look up $LOGNAME. If the UID doesn't match
627 getuid(), fall back to the current behaviour of looking
628 up the user data based on getuid().
630 2005-12-04 Matthias Clasen <mclasen@redhat.com>
632 * glib/gmacros.h (G_GNUC_WARN_UNUSED_RESULT): Add a macro
633 to make gcc warn if a function result is ignored. (#145466,
634 Arjan van de Ven, Alex Larsson)
636 * glib/gmem.h: Add the new attribute to g_realloc and
639 2005-12-03 Matthias Clasen <mclasen@redhat.com>
642 * glib/gcache.h: Deprecate g_cache_value_foreach. (#322956,
645 * glib/gmappedfile.c: Make mapping of empty files
648 * glib/gfileutils.c: Don't fork a new process just to
649 fix the permissions of the created temp file. (#321318,
650 Alexis S. L. Carvalho)
652 2005-12-02 Matthias Clasen <mclasen@redhat.com>
654 * README.in: Add a note about Solaris threads.
657 * configure.in: Remove support for Solaris threads.
658 (#136971, Sebastian Wilhelmi, patch by Andrew Paprocki)
660 2005-12-02 Matthias Clasen <mclasen@redhat.com>
662 * configure.in: Define G_GUINT64_CONSTANT in analogy to
663 G_GINT64_CONSTANT. (#322568, Andrew Paprocki)
665 2005-12-02 Matthias Clasen <mclasen@redhat.com>
667 * glib/gslice.c: Win32 portability fixes and C99ism removal,
668 pointed out by Kazuki Iwamoto. (#323052)
670 2005-12-02 Matthias Clasen <mclasen@redhat.com>
672 * glib/gatomic.c (g_atomic_int_add): Add a missing volatile for
673 the IA64 implementation.
675 Fri Dec 2 16:18:09 2005 Tim Janik <timj@imendio.com>
677 * glib/gslice.c: some naming and type size fixups.
679 Fri Dec 2 13:08:58 2005 Tim Janik <timj@gtk.org>
681 * glib/gslice.h (g_slice_free): reworked GCC-specific type-safe macro
682 variant into something less verbose, but digestible for gcc-3.4.
684 Fri Dec 2 10:55:07 2005 Tim Janik <timj@imendio.com>
686 * tests/slice-test.c: extended to perform the benchmarking on the old
687 memchunk code if 'O' is selected.
689 * tests/memchunks.c: new file which contains the old GLib mem chunks
690 implementation with prefix old_mem_chunk_.
692 * tests/Makefile.am: added memchunks.c
694 Fri Dec 2 00:16:59 2005 Tim Janik <timj@imendio.com>
696 * glib/gslice.c: improved gettimeofday() timer resolution by maintaining
697 acceptable load distribution of the syscall.
698 (allocator_get_magazine_threshold): reduce minimum/single-thread magazine
699 sizes to approximately page_size/2.5 to avoid excessive startup allocations.
701 2005-12-01 Matthias Clasen <mclasen@redhat.com>
703 * tests/gobject/ifacecheck.c (test_iface_base_init): Don't
704 double-free base interface members.
706 * tests/Makefile.am: Remove duplicate variable.
708 Thu Dec 1 17:32:46 2005 Tim Janik <timj@imendio.com>
710 * glib/gslice.[hc]: new slice allocator implementation.
712 * tests/slice-test.c: added random slice allocation test.
714 * glib/gthread.[hc]: removed newly added private thread mem API.
716 * glib/gthreadinit.h:
719 * glib/gmem.c: divided glib threading initialisation into three phases,
720 initialisation where private keys and messaging are not available (only
721 needed by gmem.c), initialisation without messaging but private keys
722 available (gslice.c, gmessage.c), and full fledged initialisers that
723 server the rest of glib. initialisation functions got renamed to reflect
724 the limitations of their corresponding phases.
726 * glib/gmem.c: removed memchunk code, defer allocations to
729 * glib/gmem.[hc]: removed g_slice_* skeletons.
731 * glib/glib.symbols: added g_slice_* symbols.
733 * configure.in: check for availability of posix_memalign(3), memalign(3)
736 * glib/Makefile.am: added gslice.[hc].
738 2005-12-01 Tor Lillqvist <tml@novell.com>
740 * glib/gstdio.c (g_stat): In the Win32 implementation, strip
741 trailing slash(es) for non-root folders. stat() fails if non-root
742 folders are specified with trailing slashes. It's too much hassle
743 to demand that callers strip such slashes themselves, especially
744 as it is easy to get it wrong and strip the slash of a root
746 (g_rename): On NT-based Windows, use MoveFileEx() with
747 MOVEFILE_REPLACE_EXISTING to better match Unix behaviour.
749 2005-11-28 Matthias Clasen <mclasen@redhat.com>
751 Fix G_STMT_START / G_STMT_END on Solaris. (#321972,
754 * configure.in: Check whether do { } while (0) works.
756 * glib/gmacros.h: Use do { } while (0) for G_STMT_START /
757 G_STMT_END if it works.
759 2005-11-28 Matthias Clasen <mclasen@redhat.com>
761 * glib/gthread.c (g_static_rw_lock_wait, g_static_rw_lock_signal):
762 * glib/gnode.c (g_node_depth_traverse_level):
763 * glib/gmem.c (g_allocator_new):
764 * glib/ghash.c (g_hash_table_unref, g_hash_table_destroy)
765 (g_hash_table_foreach_remove_or_steal): Silence compiler
768 2005-11-27 Matthias Clasen <mclasen@redhat.com>
770 * glib/gunicollate.c (g_utf8_collate_key_for_filename):
771 Don't read beyond len. (#322520, Christian Persch)
773 Wed Nov 23 17:34:01 2005 Tim Janik <timj@imendio.com>
775 * glib/gdataset.c: access datalist flags via atomic pointer access
776 functions, instead of acquiring the dataset lock. this is faster and
777 also matches the atomic pointer readouts in gdatalistprivate.h and
778 direct pointer modifications required by gobject.c.
780 Wed Nov 23 13:35:31 2005 Tim Janik <timj@gtk.org>
782 * glib/gdataset.c: streamlined initialization code somewhat,
783 removed GData node cache.
785 2005-11-22 Matthias Clasen <mclasen@redhat.com>
787 * glib/ghash.c (g_hash_table_ref):
788 (g_hash_table_unref): Mark these as new API.
790 Avoid double locking in g_intern_string (#322133,
793 * glib/gdataset.c (g_quark_from_string_internal): New
794 internal function which factors out common parts of
795 g_quark_from[_static]_string.
796 (g_quark_from_string, g_quark_from_static_string):
797 Use g_quark_from_string_internal.
798 (g_intern_string, g_intern_static_string): Use
799 g_quark_from_string_internal, and only take the
801 (g_quark_new): Don't store the strings shifted by -1
802 in the g_quarks array.
803 (g_quark_to_string): Adapt to the previous change.
805 Tue Nov 22 14:04:26 2005 Tim Janik <timj@imendio.com>
809 g_hash_table_new_full(): create hash tables with a ref count of 1.
810 g_hash_table_ref(): atomically ref_count+=1
811 g_hash_table_unref(): atomically ref_count-=1, destroys hash table
812 when refcount reaches 0.
813 g_hash_table_destroy(): just destroy keys and values, unref by 1.
814 g_hash_table_insert():
815 g_hash_table_replace(): assert ref_count>0.
818 * glib/gatomic.c: added 'volatile' qualifier to all atomic pointer and
821 2005-11-20 Behdad Esfahbod <behdad@gnome.org>
823 * glib/guniprop.c (g_unichar_get_mirror_char): Remove unused
824 variables. (#321984, Andrew Paprocki)
826 2005-11-18 Matthias Clasen <mclasen@redhat.com>
828 * configure.in: Bump version
830 2005-11-17 Matthias Clasen <mclasen@redhat.com>
832 * === Released 2.9.0 ===
834 * glib/glib.symbols: Updates
836 * glib/gutf8.c (g_utf8_offset_to_pointer): Add some performance
837 hints to the docs. (Paolo Borelli)
841 * glib/gthread.c (g_thread_foreach): Mark as new api.
843 * README.in: Updates.
847 * glib/gdate.c (g_date_set_time_t):
848 (g_date_set_time_val): Add functions to set a date from
849 a time_t and from a GTimeVal.
850 (g_date_set_time): Deprecate. (#314067, Roger Leigh)
853 * tests/utf8-pointer.c: Unit tests for g_utf8_pointer_to_offset
854 and g_utf8_offset_to_pointer.
856 * glib/gutf8.c (g_utf8_pointer_to_offset)
857 (g_utf8_offset_to_pointer): Handle negative offsets, and use
858 "stutter stepping" for going backwards. (#320638, Larry
863 * glib/gthread.c: const correctness fixes, found
864 by Arjan van de Ven and gcc.
866 2005-11-16 Behdad Esfahbod <behdad@gnome.org>
868 * docs/reference/glib/tmpl/unicode.sgml: Correct typo on mentioning
869 @G_UNICODE_BREAK_UNKNOWN that should be #G_UNICODE_BREAK_UNKNOWN.
871 2005-11-09 Behdad Esfahbod <behdad@gnome.org>
873 * glib/guniprop.c: Use bit hacks instead when checking a general
874 category value against multiple values.
876 * glib/gutf8.c: Change ISO10646 to Unicode in docs.
878 2005-11-10 Simos Xenitellis <simos@gnome.org>
880 * configure.in: Added tt (Tatar) to ALL_LINGUAS.
882 2005-11-05 Tor Lillqvist <tml@novell.com>
884 * glib/giowin32.c (g_io_win32_set_flags): Don't call g_warning().
885 (#320688) Instead set error and return failure.
887 2005-11-07 Matthias Clasen <mclasen@redhat.com>
889 * glib/gmain.c (g_main_dispatch): Don't call cb_funcs->unref
890 while holding the context lock. (#320886, Andy Wingo)
892 2005-11-07 Billy Biggs <vektor@dumbterm.net>
894 * tests/refcount/Makefile.am: Add a missing reference to libglib
895 in the LDADD for this test.
897 2005-11-07 Matthias Clasen <mclasen@redhat.com>
899 * glib/gmem.c (g_allocator_new): We need to set n_preallocs to a
900 nonzero value, otherwise GTK+ 2.8 breaks when compiled against
901 GLib 2.9. (#320755, Luca Ferretti)
903 2005-11-04 Matthias Clasen <mclasen@redhat.com>
905 * glib/Makefile.am: Apply a patch from Behdad Esfahbod to
906 use a faster lookup table for g_unichar_get_mirror_char().
907 * glib/gmirroringtable.h: The new table.
908 * glib/gunichartables.h: Remove bidi_mirroring_table.
909 * glib/gen-unicode-tables.pl: Don't generate the mirroring
911 * glib/glib-mirroring-tab/*: Sources for the program
912 which generated gmirroringtable.h.
914 * glib/glist.c: Avoid some code duplication.
916 * glib/gscanner.h: Include gdataset.h. (#320322)
918 * glib/gdebug.h: add new GDebugFlag for fatal_criticals
919 * glib/gmessages.c: (_g_debug_init): handle G_DEBUG=fatal_criticals,
920 to help find critical warnings in applications. (#320017,
923 2005-11-02 Tor Lillqvist <tml@novell.com>
925 * glib/glib.symbols: Remove large amount of trailing whitespace
926 from one line. Remove #ifdef G_UNDEFINED from around
927 g_hash_table_new and g_hash_table_new_full symbols, what was the
928 purpose of that? Add the g_slice functions.
930 Tue Nov 1 16:24:20 2005 Tim Janik <timj@imendio.com>
932 * glib/gmem.[hc]: prepared deprecation of GMemChunk and GAllocator.
933 added g_slice_*() API to allocate and cache small bits of memory.
934 an actuall allocator implementation for g_slice_*() is still pending.
936 * glib/gthread.[hc]: changes from a patch by Matthias Clasen.
937 changed GRealThread list to use in-structure *next; fields instead
938 of GSList, in order for thread iteration to not depenend on g_slice_*()
940 _g_thread_mem_private_get():
941 _g_thread_mem_private_set(): added accessors for private memory,
942 needed because the ordinary GPrivate implementation relies on GArray
943 and GSList and therefore indirectly on working g_slice_*() allocations.
946 g_thread_foreach(): new public API function to loop over all existing threads.
960 removed GAllocator and free list usages and accompanying locks.
961 use g_slice_*() API to allocate and cache small bits of memory.
963 * glib/ghook.h: removed GMemChunk field from public API.
966 * glib/glist.h: deprecate allocator API, provide _free1() for consistency.
968 * glib/gnode.h: deprecate allocator API.
970 * glib/gmain.c: reordered GPollRec fields so g_slice_free_chain() can
971 be used for poll rec lists.
973 * glib/grel.c: removed mem chunk usage, and allocated tuples via g_slice_*().
974 g_relation_destroy(): free all tuples from the all_tuples hash table,
975 this effectively maintains the life time track keeping of tuples.
976 g_relation_delete_tuple(): free tuples which are removed from the
977 all_tuples hash table. this fixes a temporary leak that was present
978 in the memchunk code until the destruction of the relation.
980 2005-10-29 Matthias Clasen <mclasen@redhat.com>
982 * tests/convert-test.c: Add some tests for conversions between
983 UTF-8, UCS-4 and UTF-16.
985 * glib/gutf8.c (g_utf8_to_ucs4, g_utf8_to_utf16): Fix handling
986 of len == -1, noticed by Morten Welinder.
988 2005-10-27 Erdal Ronahi <erdal.ronahi@gmail.com>
990 * configure.in: Added ku (Kurdish) to ALL_LINGUAS
992 2005-10-26 Matthias Clasen <mclasen@redhat.com>
994 * glib/gutf8.c (g_ucs4_to_utf8): Don't set items_read twice
995 in the error case, and add some documentation. (#319806, Morten
998 2005-10-19 Manish Singh <yosh@gimp.org>
1000 * configure.in: Use AC_CHECK_FUNCS for _NSGetEnviron, to get the
1001 config.h symbol automatically. Fixes bug #313731.
1003 2005-10-19 Tor Lillqvist <tml@novell.com>
1005 * glib/gwin32.c (g_win32_get_package_installation_directory):
1006 Return a g_strdup()ed copy of the value stored in the hash table,
1007 so that it can be g_free()d without leaving a dangling pointer in
1008 the hash table. (#319232)
1010 2005-10-06 Matthias Clasen <mclasen@redhat.com>
1012 * glib/gunicollate.c (g_utf8_collate_key_for_filename): Handle
1013 all-zero sequences correctly. (#317930, Sebastien Bacher)
1015 2005-10-05 Matthias Clasen <mclasen@redhat.com>
1017 * glib/gmarkup.c (g_markup_escape_text): Doc addition
1019 * Makefile.am: Add Collation tests.
1021 * tests/collate/*: Inputs and expected outputs for collation tests.
1023 * tests/run-collate-tests.sh: Script to run collation tests.
1025 * tests/unicode-collate.c (main): Rework slightly to make
1026 it usable in unit tests. Also test g_utf8_collate_key_for_filename().
1028 2005-10-01 Behdad Esfahbod <behdad@gnome.org>
1030 * docs/reference/glib/tmpl/unicode.sgml:
1031 * glib/gen-unicode-tables.pl:
1033 * glib/gunichartables.h:
1035 * tests/casefold.txt:
1036 * tests/casemap.txt: Updated to Unicode 4.1. There are five new
1037 GUnicodeBreakType types. That may break some applications, like
1040 2005-09-26 Matthias Clasen <mclasen@redhat.com>
1042 * glib/gstrfuncs.c (g_ascii_strtoull): Add details to the
1043 docs. (#314393, Matthew F. Barnes)
1045 * glib/glib.symbols:
1046 * glib/gprintf.h: Remove g_snprintf() and g_vsnprintf(), since
1047 they are already declared in glib.h. This doesn't break documented
1048 use of gprintf.h, but should probably be pointed out in the
1049 release notes for 2.10. (#314232, Behdad Esfahbod)
1051 Tue Sep 20 13:16:04 2005 Tim Janik <timj@imendio.com>
1053 * glib/gpattern.c (g_pattern_ph_match): applied significant recursion
1054 complexity optimization, based on a patch from Matthias Clasen.
1056 * tests/patterntest.c: more tests, mostly from matthias.
1058 2005-09-20 Matthias Clasen <mclasen@redhat.com>
1060 * glib/gqueue.c (g_queue_insert_sorted): Correct the docs.
1061 (#316703, Mark Drago)
1063 Mon Sep 19 17:23:23 2005 Tim Janik <timj@imendio.com>
1065 * glib/gpattern.c: applied a patch from matthias which checks on the
1066 upper bound of GPatternSpec length to optimize matches.
1069 * tests/patterntest.c: added more match cases.
1071 2005-09-16 Tor Lillqvist <tml@novell.com>
1073 * glib/gstrfuncs.c (g_ascii_strcasecmp, g_ascii_strncasecmp): Add
1074 warning to doc comment that these functions should not be used on
1075 encodings like CP932.
1077 2005-09-14 Matthias Clasen <mclasen@redhat.com>
1079 * tests/keyfile-test.c: Add a test for grup names of length 1.
1081 * glib/gkeyfile.c (g_key_file_line_is_group): Accept group names
1082 of length 1. (#316309)
1084 2005-09-12 Matthias Clasen <mclasen@redhat.com>
1086 * glib/gmarkup.c (g_markup_escape_text): Clarify docs.
1088 2005-09-11 Sebastian Wilhelmi <seppi@seppi.de>
1090 * tests/refcount/Makefile.am (INCLUDES): Link the the refcount
1091 tests to the system thread library $(G_THREAD_LIBS). Fixes #313744
1094 2005-09-11 Kjartan Maraas <kmaraas@gnome.org>
1096 * glib/gmain.c: (g_child_watch_prepare), (g_child_watch_check),
1097 (child_watch_helper_thread): Remove some dead code. Closes
1100 2005-09-07 Tor Lillqvist <tml@novell.com>
1102 * glib/Makefile.am: Create also a console version of the
1103 gspawn-win32-helper program, gspawn-win32-helper-console.exe.
1104 It's otherwise identical to gspawn-win32-helper.exe, except marked
1105 as a console application (linked without the -mwindows option).
1107 * glib/gspawn-win32.c (do_spawn_directly, do_spawn_with_pipes):
1108 Drop the dont_wait parameter. Its truth value correlated 100% with
1109 the NULLness of the exit_status parameter anyway, so it's enough
1110 to check whether exit_status is NULL. Invert the sense of the
1111 dont_return_handle parameter and rename it to do_return_handle, to
1112 make the code easier to read by avoiding double negations.
1114 (g_spawn_sync_utf8, g_spawn_async_with_pipes_utf8): Modify calls
1115 to do_spawn_with_pipes() accordingly.
1117 (do_spawn_with_pipes): If we have a console, use the console
1118 version of the helper program, otherwise use the GUI one. This
1119 avoids extra console windows opening up in some situations. (In
1120 case a console application uses the GUI gspawn-win32-helper.exe to
1121 spawn another console application we would get a separate console
1122 for the spawned console application).
1124 * glib-zip.in: Distribute also gspawn-win32-helper-console.exe.
1126 2005-09-05 Matthias Clasen <mclasen@redhat.com>
1128 * glib/gmappedfile.c (g_mapped_file_new): Report an error
1129 if the file is too large. (#315275, Kjartan Maraas)
1131 * glib/gkeyfile.c (g_key_file_load_from_fd): The return value
1132 of read() is signed. (#315273, Kjartan Maraas)
1134 2005-08-31 Tor Lillqvist <tml@novell.com>
1136 * glib/gutils.h: Wrapping atexit() is a bad idea on Windows, where
1137 the EXE and each DLL have their own atexit function chains.
1139 #define g_atexit as atexit instead. This means it has a
1140 better chance of doing what the caller wants. For instance,
1141 gtkhtml calls g_atexit() registering a function in gtkhtml
1142 itself. This caused a crash when g_atexit() was implemented as a
1143 function in the GLib DLL. The gtkhtml DLL was already unloaded by
1144 the time the GLib DLL got unloaded.
1146 * glib/gutils.c: #undef the #define mentioned above, to also get a
1147 real g_atexit() into the DLL for backward compatibility. Document
1148 the Windows behaviour of g_atexit(), and document the varying ways
1149 atexit() can behave in the context of dynamically loaded modules
1152 2005-08-31 Matthias Clasen <mclasen@redhat.com>
1154 * glib/glib.symbols:
1156 * glib/gdataset.c: Add string interning functions.
1158 2005-08-28 Matthias Clasen <mclasen@redhat.com>
1160 * glib/giochannel.c: Unify some near-duplicate strings. (#314654,
1163 2005-08-26 Matthias Clasen <mclasen@redhat.com>
1165 * configure.in: Bump version to 2.9.0
1167 2005-08-25 Tor Lillqvist <tml@novell.com>
1169 Make also the g_spawn*() functions take parameters in the GLib
1170 file name encoding, i.e. UTF-8, on Windows. Has no impact on Unix
1171 API or ABI. Like the other GLib API that was earlier changed to
1172 use UTF-8 on Windows, the names of the functions that take UTF-8
1173 have _utf8 suffixes added by using preprocessor macros in the
1174 header file. The old names are kept for functions with the old
1175 behaviour, taking parameters in the system codepage, for DLL ABI
1178 * glib/gspawn.h: On Win32 add the suffix _utf8 to the names of the
1179 g_spawn*() functions.
1181 * glib/gspawn-win32.c: Use wide-char API on NT-based
1182 Windows. Convert parameters from UTF-8 to wide chars (NT) or
1183 system codepage (Win9x) and call the C library _wspawn*() or
1184 spawn*() functions respectvely. Add DLL ABI stability versions
1185 that take parameters in the system codepage.
1187 * glib/gspawn-win32-helper.c: On NT-based Windows use the
1188 wide-char versions of argv and envp, and use wide-char API to
1189 change directory and spawn the program to run. Remove the verbose
1190 debugging output, it was too complex to modify for the wide-char
1191 features. (Just add temporary debugging printouts if needed, no
1192 need to have them permanently in the source.)
1194 * glib/gspawn.c: Corresponding documentation updates.
1196 * glib/glib.symbols: Corresponding changes: Mark the ABI stability
1197 symbols as PRIVATE, add the new _utf8-suffixed ones.
1199 2005-08-24 Stepan Kasal <kasal@ucw.cz>
1201 * glib/gtypes.h (G_MININT64): Cast the constant to gint64; it is
1202 guint64 otherwise and that can produce warnings about comparison
1203 between signed and unsigned.
1205 2005-08-23 Matthias Clasen <mclasen@redhat.com>
1207 * glib/gutils.c: Fix the crt_externs.h include.
1209 2005-08-23 Stepan Kasal <kasal@ucw.cz>
1211 * NEWS: Fix spelling of my first name.
1213 2005-08-23 Matthias Clasen <mclasen@redhat.com>
1217 * === Released 2.8.1 ===
1221 2005-08-20 Hans Breuer <hans@breuer.org>
1223 * glib/makefile.msc.in : link with ws2_32.lib
1225 2005-08-18 Tor Lillqvist <tml@novell.com>
1227 * configure.in: Check for <sys/wait.h>
1229 * glib/gbacktrace.c: Include <sys/wait.h> on if HAVE_SYS_WAIT_H.
1231 2005-08-18 Ross Burton <ross@burtonini.com>
1234 Optimise single-character insertions.
1239 * tests/string-test.c:
1240 Add tests for new optimisation, and fix a leak.
1242 2005-08-17 Matthias Clasen <mclasen@redhat.com>
1244 * configure.in: Check for crt_externs.h and _NSGetEnviron.
1246 * glib/gutils.c: On Darwin, include crt-externs.h and
1247 define environ using _NSGetEnviron(). (#313731)
1249 2005-08-16 Stepan Kasal <kasal@ucw.cz>
1251 * glib/gutils.c (g_get_any_init): Move the body of the big if...
1252 (g_get_any_init_do): ... to this new function.
1253 (g_get_any_init): Declare as inline.
1254 (g_get_any_init_locked): New inline function, does the locking.
1255 Make use of these two throughout the code.
1257 2005-08-15 Matthias Clasen <mclasen@redhat.com>
1259 * glib/gbacktrace.c (g_on_error_stack_trace): Wait for
1260 the child process and then simply return. This makes
1261 The "S" option work as documented in g_on_error_query().
1262 (#313125, Matthew F. Barnes)
1264 * glib/gunicode.h: Update the link to Unicode category
1265 values. (#313369, Behnam Esfahbod)
1267 * glib/gqueue.c (g_queue_find_custom): Clarify docs
1268 a little. (#311727, Tristan van Berkom)
1270 * glib/abicheck.sh, gobject/abicheck.sh: Make the
1271 check work on ia64 too, where some symbols ended up
1272 in yet another section.