1 2006-03-14 Matthias Clasen <mclasen@redhat.com>
3 * glib/goption.c (parse_short_option): Don't create the
4 option_name twice. (#334519, Chris Wilson)
6 2006-03-13 Anders Carlsson <andersca@luminoth.local>
8 * configure.in: Revert fix for #322476, it breaks module loading
9 since libtool on darwin makes shared modules use .so and shared
10 libraries use .dylib. The fix breaks shared module loading everywhere
13 2006-03-02 Marcus Brinkmann <mb@g10code.de>
15 Implement watches for GIOChannels for write file descriptors on
18 * glib/giowin32.c (GIOWin32Channel): Add a new direction field.
19 (read_thread): Initialize direction.
20 (write_thread): New function.
21 (buffer_write): New function.
22 (g_io_win32_prepare): Handle the G_IO_WIN32_FILE_DESC case for the
24 (g_io_win32_fd_write): Call buffer_write() if there is a writer
26 (g_io_win32_fd_close): Set space_avail_event for writer threads.
27 (g_io_win32_fd_create_watch): Create the writer thread if
28 condition is G_IO_OUT.
29 (g_io_channel_win32_make_pollfd): Likewise here.
31 2006-03-09 Matthias Clasen <mclasen@redhat.com>
33 * Makefile.am: Add ChangeLog.pre-2.8 to EXTRA_DIST.
35 2006-03-08 Tor Lillqvist <tml@novell.com>
37 * glibconfig.h.win32.in: Add G_GUINT64_CONSTANT.
39 2006-03-07 Matthias Clasen <mclasen@redhat.com>
41 * configure.in: Bump version
43 * === Released 2.10.1 ===
47 * glib/gutf8.c (_g_charset_get_aliases): Match the prototype
48 in gconvert.c to fix build problems in NetBSD. (#333651,
49 Julio M. Merino Vidal)
51 2006-03-06 Anders Carlsson <andersca@imendio.com>
54 Make .dylib the shared library suffix on darwin.
55 (#322476, Vladimir Panov)
57 2006-03-06 Anders Carlsson <andersca@imendio.com>
60 Skip checking thread flags on Darwin. Fix suggested by Manish Singh.
61 (#314794, Gregor Riepl)
63 2006-03-01 Tor Lillqvist <tml@novell.com>
65 * glib/gutils.c (g_listenv): Separate implementation on Win32: Use
66 the wide character API on NT-based Windows. Return UTF-8 strings.
68 * glib/glib.symbols: Don't mark g_listenv as PRIVATE, as that
69 meant it wasn't present in the import library. PRIVATE is used
70 only for the backwards-compatibility DLL ABI stability hacks.
72 2006-02-24 Matthias Clasen <mclasen@redhat.com>
74 * configure.in: Bump version
76 * === Released 2.10.0 ===
78 * README.in: Mention the slice allocator and GInitiallyUnowned.
82 * configure.in: Bump version to 2.10.0
84 * tests/utf8-pointer.c: Add tests for g_utf8_strlen().
86 * glib/gutf8.c: Fix boundary cases in g_utf8_strlen().
89 2006-02-23 Kjartan Maraas <kmaraas@gnome.org>
91 * tests/completion-test.c: (main): Plug leaks reported by
93 * tests/convert-test.c: (test_iconv_state), (test_one_half):
95 * tests/patterntest.c: (test_compilation): Same
96 * tests/shell-test.c: (do_argv_test): Same.
97 * tests/unicode-caseconv.c: (main): Same.
98 * tests/uri-test.c: (run_to_uri_tests): Same.
101 2006-02-19 Martyn Russell <martyn@imendio.com>
103 * tests/asyncqueue-test.c:
105 * tests/slist-test.c: Updated to test _sort, _sort_with_data,
106 _insert_sorted and _insert_sorted_with_data API.
108 2006-02-18 Matthias Clasen <mclasen@redhat.com>
110 * tests/gobject/Makefile.am: Add paramspec-test
112 * tests/gobject/paramspec-test.c: Some GParamSpec tests.
114 * tests/gobject/gvalue-test.c: Add more tests.
116 2006-02-17 Kang Jeong-Hee <keizie@gmail.com>
118 * glib/gutf8.c (g_utf8_find_prev_char): Correct documentation typo.
120 2006-02-15 Sebastian Wilhelmi <seppi@seppi.de>
122 * glib/gthreadpool.c: Fix deadlock when signalling the thread
123 which freed a thread pool (#331110, Chris Wilson).
125 Tue Feb 14 17:00:43 2006 Tim Janik <timj@imendio.com>
127 * glib/gslice.c: only define _XOPEN_SOURCE to 600 to get at
128 posix_memalign() in case we actually are going to use it, because
129 we detected a compliant implementation (#328997).
131 * configure.in (enable_included_printf): don't include malloc.h when
132 testing for posix_memalign() funcitonality, since this may break the
133 test on some systems (#328997).
135 2006-02-14 Matthias Clasen <mclasen@redhat.com>
137 * glib/gstrfuncs.c: Improve docs.
139 2006-02-11 Matthias Clasen <mclasen@redhat.com>
141 * configure.in: Bump version
143 2006-02-10 Matthias Clasen <mclasen@redhat.com>
145 * === Released 2.9.6 ===
149 2006-01-31 Behdad Esfahbod <behdad@gnome.org>
151 * docs/reference/glib/tmpl/relations.sgml (g_relation_insert): Specify
152 the type of varargs arguments. (bug #317679)
154 2006-01-31 Matthias Clasen <mclasen@redhat.com>
156 * autogen.sh: Touch README and INSTALL here to pacify
157 automake. (#329124, Kjartan Maraas, Tim Janik)
159 Tue Jan 31 16:45:03 2006 Tim Janik <timj@imendio.com>
161 * glib/gdataset.c: carry out all dtalist pointer accesses atomically,
162 some missing cases where pointed out by Sebastian Wilhelmi.
164 Tue Jan 31 12:33:48 2006 Tim Janik <timj@imendio.com>
166 * configure.in: generate build/win32/vs8/Makefile, to fix build breakage.
168 2006-01-30 David Schleef <ds@schleef.org>
170 * glib/grand.c: Update URL
172 2006-01-29 Sven Herzberg <herzi@gnome-de.org>
174 * docs/reference/gobject/tmpl/param_value_types.sgml: tell how to
175 create container classes which are as flexible as a GValue is
177 2006-01-27 Matthias Clasen <mclasen@redhat.com>
179 * configure.in: Bump version
181 * === Released 2.9.5 ===
183 * glib/glib.symbols: Add g_mem_gc_friendly as an exported
188 2006-01-26 Matthias Clasen <mclasen@redhat.com>
190 * glib/gmem.c (g_mem_init_nomessage): Fix C99isms. (#328705,
193 Wed Jan 25 19:16:57 2006 Tim Janik <timj@imendio.com>
195 * fixed buglets reported by Jens Granseuer in #328254.
197 * configure.in: free the memory allocated in posix_memalign() tests.
199 * glib/gslice.c: spelling fixes.
201 Wed Jan 25 16:39:18 2006 Tim Janik <timj@imendio.com>
203 * glib/gslice.c: honour g_mem_gc_friendly settings when freeing
204 slices, make sure g_mem_gc_friendly is properly initialized.
206 * gmem.[hc]: ensure g_mem_gc_friendly is initialized from G_DEBUG upon
207 the first allocation. applied some branching optimizations.
209 * docs/macros.txt: reflected --enable-gc-friendly change and
210 described ENABLE_GC_FRIENDLY_DEFAULT as well as G_DEBUG=gc-friendly.
212 * configure.in: changed --enable-gc-friendly=yes to define
213 ENABLE_GC_FRIENDLY_DEFAULT.
215 * glib/garray.c: changed ENABLE_GC_FRIENDLY macro #ifdef-s to
216 if (G_UNLIKELY (g_mem_gc_friendly)).
219 * glib/ghash.c: removed ENABLE_GC_FRIENDLY code which is now taken
220 care of by g_slice_free1().
222 * tests/slice-test.c: fixed leaks, reported by Kjartan Maraas.
224 Tue Jan 24 17:49:36 2006 Tim Janik <timj@imendio.com>
226 * glib/gslice.c: only use posix_memalign() if it's known to work,
227 revert to memalign() otherwise.
229 * configure.in: check for broken posix_memalign() implementations
232 2006-01-24 Matthias Clasen <mclasen@redhat.com>
234 * tests/unicode-encoding.c: Use UTF-16LE as target encoding
235 on all little-endian systems. (#143380, Marc Moorcroft)
237 2006-01-23 Matthias Clasen <mclasen@redhat.com>
239 * configure.in: Change the shared libary extension for hpux-ia64
240 to so. (#328253, The Written Word)
242 Mon Jan 23 17:30:33 2006 Tim Janik <timj@imendio.com>
244 * glib/gutils.c (_g_getenv_nomalloc): wiped out all the wonderfull
245 G_OS_WIN32 code i wrote ;-[) after tml told me windows has getenv()
248 Mon Jan 23 16:46:20 2006 Tim Janik <timj@imendio.com>
250 * glib/gslice.c (slice_config_init): initialize GSlice config from
251 G_SLICE environment variable. we support G_SLICE=always-malloc
252 currently, which forces all g_slice_*() allocations to use the system
256 g_parse_debug_string(): added a note about not using g_malloc() here.
257 _g_getenv_nomalloc(): getenv() variant that doesn't use g_malloc or
258 g_slice. contains only guesswork in the WIN32 branch.
260 2006-01-18 Matthias Clasen <mclasen@redhat.com>
264 * === Released 2.9.4 ===
268 2006-01-17 Matthias Clasen <mclasen@redhat.com>
270 * glib/gkeyfile.c (g_key_file_get_groups): Correct a g_new0()
271 call. (#327421, Morten Welinder)
273 2006-01-17 Sebastian Wilhelmi <seppi@seppi.de>
275 * glib/gthreadpool.c: To avoid deadlocks get rid of the settings
276 G_LOCK. Use the unused_thread_queue lock instead. Change
277 g_thread_pool_thread_proxy such that threads only wait on
278 non-exlusive pools for at most a 1/2 second. Do not reorder tasks
279 due to superfluous tasks. Global tasks wait at most for
280 max-idle-time milliseconds. Make sure, that no task is woken up
281 twice for the same event via a wakeup_serial. This fixes #324228.
283 * tests/threadpool-test.c: Adapt test accordingly. Do not pass
284 invalid NULL into the thread pools. This as well fixes #327290.
286 2006-01-16 Matthias Clasen <mclasen@redhat.com>
288 * configure.in: Bump version
290 * === Released 2.9.3 ===
294 * glib/gthread.h (g_static_mutex_get_mutex_impl_shortcut):
295 Rework the strict aliasing fix to not break C++, pointed
296 out by Murray Cumming.
298 * glib/gasyncqueue.c (g_async_queue_push_sorted_unlocked):
299 Signal waiting threads, problem noticed by Christian Kellner.
301 2006-01-16 Matthias Clasen <mclasen@redhat.com>
303 Fix bug 326747, Alberto Ruiz:
305 * glib/gconvert.c (make_valid_utf8): Change this function to
306 replace unknown characters by the Unicode replacement character
307 instead of '?', and don't append "(invalid encoding)".
308 (g_filename_display_name, g_filename_display_basename): Document
309 how to determine if the filename was in an invalid encoding.
311 2006-01-14 Matthias Clasen <mclasen@redhat.com>
313 * glib/gtree.c: Replace the simple recursive implementation
314 by a nonrecursive, threaded implementation by Maurizio
317 2006-01-12 Matthias Clasen <mclasen@redhat.com>
319 * tests/asyncqueue-test.c (main): Fix the
320 assert to not always trigger. (#326558,
323 2006-01-11 Matthias Clasen <mclasen@redhat.com>
325 * tests/convert-test.c: Don't test items_read and
326 items_written in error cases where it is not set.
328 * tests/asyncqueue-test.c: Set error to NULL.
330 2006-01-08 Matthias Clasen <mclasen@redhat.com>
332 * glib/gtree.c: Break some long lines.
334 * tests/tree-test.c: Add more tests.
336 * glib/gslice.c: Include config.h first, pointed out
339 2005-12-20 Sven Herzberg <herzi@gnome-de.org>
341 * docs/reference/gobject/tmpl/gtype.sgml: explain that
342 G_TYPE_INSTANCE_GET_CLASS() does behave different during
345 2006-01-06 Matthias Clasen <mclasen@redhat.com>
347 * glib/gslice.c: Only define _XOPEN_SOURCE if we know
348 that we have posix_memalign(). (#323937, Bogdan Nicula)
350 2006-01-05 Matthias Clasen <mclasen@redhat.com>
352 * configure.in: Bump version
354 * === Released 2.9.2 ===
356 2006-01-05 Hans Breuer <hans@breuer.org>
358 * glib/makefile.msc.in : added gslice.obj
360 2006-01-05 Matthias Clasen <mclasen@redhat.com>
364 * glib/gmain.c: Clarify the documentation of source ids
365 a bit. (#325874, Dan Williams)
367 * configure.in: Fix another strict aliasing problem.
369 * glib/gthreadpool.c: Avoid use of varargs macro. (#325864,
372 2006-01-04 Manish Singh <yosh@gimp.org>
375 * glib/gstdio.h: don't macroized g_access, g_chdir, and g_unlink
376 either, since they have the same issue as g_rmdir. (Related to
379 Wed Jan 4 13:33:25 2006 Tim Janik <timj@gtk.org>
381 * glib/gslice.c (magazine_cache_pop_magazine): don't reverse chunk
382 order when creating magazines, so we hand out chunks with ascending
385 2006-01-04 Abel Cheung <maddog@linuxhall.org>
387 * configure.in: Added "ml" "zh_HK" to ALL_LINGUAS.
389 2006-01-03 Martyn Russell <martyn@imendio.com>
391 * docs/reference/glib/glib-sections.txt:
393 * glib/gthreadpool.[ch]:
394 - Added new API g_thread_pool_get_idle_time() and
395 g_thread_pool_set_idle_time(). (#324228).
397 * tests/threadpool-test.c:
398 - Updated test case to do thread pool sorting, thread pool with
399 no sorting and a thread pool with idle thread timeouts.
401 2006-01-03 Matthias Clasen <mclasen@redhat.com>
403 * glib/gmain.h: Add new functions here, too.
405 * glib/glib.symbols: Add new functions.
407 2005-12-20 Michael Meeks <michael.meeks@novell.com>
409 * glib/gmain.c (g_main_context_is_owner): new method
410 to determine if the current thread is the owner of the
413 2006-01-02 Matthias Clasen <mclasen@redhat.com>
417 * glib/gstdio.c (g_rmdir): Don't provide g_rmdir() as a macro
418 expanding to rmdir, since rmdir is not declared in a portable
419 system header we can include in gstdio.h. (#325249, Jani Monoses)
421 2006-01-01 Tor Lillqvist <tml@novell.com>
423 * glib/gspawn-win32.c (g_spawn_sync_utf8): Set the GIOChannels for
424 stdout and stderr to unbuffered. Otherwise the giochannel layer
425 will try to read from them regardless whether the
426 g_io_channel_win32_poll() call here has indicated
427 readability or not. (#325310)
429 2005-12-29 Matthias Clasen <mclasen@redhat.com>
431 * glib/gutils.c (glib_check_version): Fix a copy-and-paste error
432 in the docs. (#325273, Declan Naughton)
434 2005-12-27 Manish Singh <yosh@gimp.org>
436 * tests/run-collate-tests.sh: use LC_ALL instead of LC_COLLATE, to
437 make sure we really override things.
439 2005-12-27 Matthias Clasen <mclasen@redhat.com>
441 Fix #316221, Michal Benes, Stanislav Brabec;
443 * configure.in: Fix a strict aliasing problem in
444 g_static_mutex_get_mutex().
445 * glib/gthread.h: ...and in
446 g_static_mutex_get_mutex_impl_shortcut().
448 * glib/gdatasetprivate.h: Add a cast to silence compiler
449 warnings. (#321978, Andrew Paprocki)
451 Partial fix for bug #323937, Bogdan Nicula.
453 * configure.in: Check for malloc.h
454 * glib/gslice.c: Don't include malloc.h unconditionally.
456 2005-12-27 Manish Singh <yosh@gimp.org>
458 * tests/run-collate-tests.sh: export LC_COLLATE so it takes.
459 (#324950, Dan Yefimov)
461 2005-12-26 Matthias Clasen <mclasen@redhat.com>
463 * glib/gslice.c: On Win32, include process.h (#325015, Kazuki
466 2005-12-25 Matthias Clasen <mclasen@redhat.com>
468 * AUTHORS: Update my email
470 * tests/slice-test.c: Fix C99isms. (#324950, Dan Yefimov)
472 2005-12-21 Matthias Clasen <mclasen@redhat.com>
474 * README.in: Add some notes on when 'make check' may fail.
476 2005-12-20 Matthias Clasen <mclasen@redhat.com>
479 * glib/gthreadpool.h:
480 * glib/gthreadpool.c (g_thread_pool_set_sort_function): New function
481 to sort tasks pushed into a threadpool. (#324479, Martyn Russell)
483 * tests/threadpool-test.c: Test this.
485 Tue Dec 20 18:14:14 2005 Tim Janik <timj@imendio.com>
487 * glib/gslice.[hc]: added mem_error() and mem_assert() to test and
488 handle errors without depending on gmessage.c which might not be
489 setup when the error occours.
490 removed G_SLICE_CONFIG_ALWAYS_FREE config option, fixed the code so
491 always freeing can be achieved by adjusting the working set time to
492 0 with G_SLICE_CONFIG_WORKING_SET_MSECS.
493 added G_SLICE_CONFIG_COLOR_INCREMENT to test different color increments
494 (mainly 0 and 1). reduced the minimum block size to 128 bytes, to
495 minimize wastage if small amounts of differently sized structrues are
496 allocated, this does come at a performance cost of roughly 5% though.
497 fixed up block alignment calculation, so it works for varying
498 block sizes. only use strerror() not g_strerror() since the latter
499 depends on working GQuark and GSlice.
500 mem_error(): implemented in terms of fprintf and vfprintf.
502 * tests/slice-color.c: new program to test cache colorization effects.
504 * tests/slice-test.c: trade G_SLICE_CONFIG_ALWAYS_FREE for 0 duration
505 G_SLICE_CONFIG_WORKING_SET_MSECS.
507 2005-12-17 Matthias Clasen <mclasen@redhat.com>
509 * glib/goption.c (parse_short_option): Set an error in all
510 failure cases. (#324332, Tim-Philipp Müller)
512 2005-12-17 Sebastian Wilhelmi <seppi@seppi.de>
514 * glib/gatomic.c: Fix memory barrier position in g_atomic_int_get
515 and g_atomic_pointer_get. Add g_atomic_int_set and
516 g_atomic_pointer_set implementations for the !DEFINE_WITH_MUTEXES &&
517 G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case, as well as defining them
518 as functions (additionally to the macros in the header) for the
519 !G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case.
521 2005-12-16 Matthias Clasen <mclasen@redhat.com>
523 * glib/gmem.c (g_allocator_new): Don't return a pointer to
524 a const struct, since apps expect to be able to modify it.
525 (#324179, J. Ali Harlow)
527 Tue Dec 13 10:13:32 2005 Tim Janik <timj@imendio.com>
529 * glib/gatomic.h: added g_atomic_pointer_set() and g_atomic_int_set()
530 for systems where the initialization of atomic variables requires a
531 write memory barrier.
533 2005-12-09 Matthias Clasen <mclasen@redhat.com>
535 * configure.in: Bump version
537 * === Released 2.9.1 ===
539 2005-12-09 Alexander Larsson <alexl@redhat.com>
543 Use G_GNUC_WARN_UNUSED_RESULT on list functions that return
546 2005-12-08 Matthias Clasen <mclasen@redhat.com>
550 2005-12-07 Martyn Russell <martyn@imendio.com>
552 * glib/gasyncqueue.c:
553 - Call g_queue_insert_sorted() instead of duplicating the code.
554 - Call g_queue_sort() instead of duplicating the code.
555 - Invert sort function results to make sure the same sort function
556 gives the same results across glist, gslist, gqueue and
559 * tests/asyncqueue-test.c:
560 - Updated the sort function to reflect the example in the
561 documentation for gasyncqueue.c.
563 2005-12-07 Martyn Russell <martyn@imendio.com>
565 * docs/reference/glib/glib-sections.txt:
566 * docs/reference/glib/tmpl/linked_lists_double.sgml:
567 * docs/reference/glib/tmpl/linked_lists_single.sgml:
570 - Added g_list_insert_sorted_with_data () and
571 g_slist_insert_sorted_with_data ().
572 - Removed the extra check in g_list_sort() and g_slist_sort() for
573 GCompareDataFunc vs. GCompareFunc.
575 2005-12-07 Tor Lillqvist <tml@novell.com>
577 * glib-gettextize.in: Look up prefix at run-time on Win32,
578 assuming the standard directory structure with glib-gettextize in
581 2005-12-06 Matthias Clasen <mclasen@redhat.com>
583 * glib/gmem.h: Exempt GTK+ from the mem chunk deprecation, since
584 we need GTK+ 2.8 to compile against GLib 2.10.
586 2005-12-05 Manish Singh <yosh@gimp.org>
588 * tests/run-collate-test.sh: set LC_COLLATE, not LANG, to be sure
589 to override any user settings.
591 2005-12-05 Matthias Clasen <mclasen@redhat.com>
593 * configure.in: Revert an accidential commit
595 2005-12-05 Matthias Clasen <mclasen@redhat.com>
597 * glib/gutf8.c: Documentation updates. (#323291, Morten Welinder)
599 * tests/Makefile.am (TESTS_ENVIRONMENT):
600 * tests/gobject/Makefile.am (TESTS_ENVIRONMENT): Set
601 MALLOC_CHECK_ and MALLOC_PERTURB_
603 * tests/run-collate-tests.sh: Run the collation tests explicitly
604 in en_US locale. (#320463)
606 * glib/gmem.h: Really deprecate GMemChunk.
613 * glib/gstrfuncs.c: Add versioned deprecation docs.
615 2005-12-05 Martyn Russell <martyn@imendio.com>
617 * docs/reference/glib/glib-sections.txt:
618 * glib/gasyncqueue.[ch]:
619 - Added support for sorting async queues by with _push_sorted(),
620 _push_sorted_unlocked(), _sort() and _sort_unlocked() (#323047).
623 * tests/asyncqueue-test.c:
624 - Added test case for gasyncqueue.c
626 Mon Dec 5 15:53:20 2005 Tim Janik <timj@imendio.com>
628 * glib/gslice.c: implement chain walking for arbitrary ->next pointer
629 offsets in g_slice_free_chain_with_offset() based on a patch by behdad
630 in bug 323178. moved time consuming logic from g_slice_free() out of
631 the inner loop, so g_slice_free_chain_with_offset() provides a real
632 performance benefit over g_slice_free1() now.
634 * glib/gslice.h: renamed g_slice_free_chain() to
635 g_slice_free_chain_with_offset(). implemented g_slice_free_chain() as
636 a type-safe macro as suggested in bug 323178.
637 simplified the macro implementation of g_slice_free() and implemented
638 it in a type safe manner for all compliers as suggested by Morten
639 Welinder <mortenw@gnome.org>.
644 * glib/glib.symbols: s/g_slice_free_chain/g_slice_free_chain_with_offset/
646 2005-12-05 Matthias Clasen <mclasen@redhat.com>
648 * glib/gasyncqueue.c: Add some docs.
650 * tests/libmoduletestplugin_a.c: Fix compiler warnings.
652 * glib/gatomic.c: In the ia64 implementation, use
653 __sync builtin without _si or _di suffix. (#321229,
654 Stanislav Brabec, patch by Andreas Schwab)
656 2005-12-04 Behdad Esfahbod <behdad@gnome.org>
658 * glib/gslice.h: Remove comma at the end of enum.
660 2005-12-04 Matthias Clasen <mclasen@redhat.com>
662 Handle multiple user names with the same UID better.
663 (#319535, Laszlo Peter)
665 * glib/gutils.c (g_get_any_init_do): When determining user
666 data, first look up $LOGNAME. If the UID doesn't match
667 getuid(), fall back to the current behaviour of looking
668 up the user data based on getuid().
670 2005-12-04 Matthias Clasen <mclasen@redhat.com>
672 * glib/gmacros.h (G_GNUC_WARN_UNUSED_RESULT): Add a macro
673 to make gcc warn if a function result is ignored. (#145466,
674 Arjan van de Ven, Alex Larsson)
676 * glib/gmem.h: Add the new attribute to g_realloc and
679 2005-12-03 Matthias Clasen <mclasen@redhat.com>
682 * glib/gcache.h: Deprecate g_cache_value_foreach. (#322956,
685 * glib/gmappedfile.c: Make mapping of empty files
688 * glib/gfileutils.c: Don't fork a new process just to
689 fix the permissions of the created temp file. (#321318,
690 Alexis S. L. Carvalho)
692 2005-12-02 Matthias Clasen <mclasen@redhat.com>
694 * README.in: Add a note about Solaris threads.
697 * configure.in: Remove support for Solaris threads.
698 (#136971, Sebastian Wilhelmi, patch by Andrew Paprocki)
700 2005-12-02 Matthias Clasen <mclasen@redhat.com>
702 * configure.in: Define G_GUINT64_CONSTANT in analogy to
703 G_GINT64_CONSTANT. (#322568, Andrew Paprocki)
705 2005-12-02 Matthias Clasen <mclasen@redhat.com>
707 * glib/gslice.c: Win32 portability fixes and C99ism removal,
708 pointed out by Kazuki Iwamoto. (#323052)
710 2005-12-02 Matthias Clasen <mclasen@redhat.com>
712 * glib/gatomic.c (g_atomic_int_add): Add a missing volatile for
713 the IA64 implementation.
715 Fri Dec 2 16:18:09 2005 Tim Janik <timj@imendio.com>
717 * glib/gslice.c: some naming and type size fixups.
719 Fri Dec 2 13:08:58 2005 Tim Janik <timj@gtk.org>
721 * glib/gslice.h (g_slice_free): reworked GCC-specific type-safe macro
722 variant into something less verbose, but digestible for gcc-3.4.
724 Fri Dec 2 10:55:07 2005 Tim Janik <timj@imendio.com>
726 * tests/slice-test.c: extended to perform the benchmarking on the old
727 memchunk code if 'O' is selected.
729 * tests/memchunks.c: new file which contains the old GLib mem chunks
730 implementation with prefix old_mem_chunk_.
732 * tests/Makefile.am: added memchunks.c
734 Fri Dec 2 00:16:59 2005 Tim Janik <timj@imendio.com>
736 * glib/gslice.c: improved gettimeofday() timer resolution by maintaining
737 acceptable load distribution of the syscall.
738 (allocator_get_magazine_threshold): reduce minimum/single-thread magazine
739 sizes to approximately page_size/2.5 to avoid excessive startup allocations.
741 2005-12-01 Matthias Clasen <mclasen@redhat.com>
743 * tests/gobject/ifacecheck.c (test_iface_base_init): Don't
744 double-free base interface members.
746 * tests/Makefile.am: Remove duplicate variable.
748 Thu Dec 1 17:32:46 2005 Tim Janik <timj@imendio.com>
750 * glib/gslice.[hc]: new slice allocator implementation.
752 * tests/slice-test.c: added random slice allocation test.
754 * glib/gthread.[hc]: removed newly added private thread mem API.
756 * glib/gthreadinit.h:
759 * glib/gmem.c: divided glib threading initialisation into three phases,
760 initialisation where private keys and messaging are not available (only
761 needed by gmem.c), initialisation without messaging but private keys
762 available (gslice.c, gmessage.c), and full fledged initialisers that
763 server the rest of glib. initialisation functions got renamed to reflect
764 the limitations of their corresponding phases.
766 * glib/gmem.c: removed memchunk code, defer allocations to
769 * glib/gmem.[hc]: removed g_slice_* skeletons.
771 * glib/glib.symbols: added g_slice_* symbols.
773 * configure.in: check for availability of posix_memalign(3), memalign(3)
776 * glib/Makefile.am: added gslice.[hc].
778 2005-12-01 Tor Lillqvist <tml@novell.com>
780 * glib/gstdio.c (g_stat): In the Win32 implementation, strip
781 trailing slash(es) for non-root folders. stat() fails if non-root
782 folders are specified with trailing slashes. It's too much hassle
783 to demand that callers strip such slashes themselves, especially
784 as it is easy to get it wrong and strip the slash of a root
786 (g_rename): On NT-based Windows, use MoveFileEx() with
787 MOVEFILE_REPLACE_EXISTING to better match Unix behaviour.
789 2005-11-28 Matthias Clasen <mclasen@redhat.com>
791 Fix G_STMT_START / G_STMT_END on Solaris. (#321972,
794 * configure.in: Check whether do { } while (0) works.
796 * glib/gmacros.h: Use do { } while (0) for G_STMT_START /
797 G_STMT_END if it works.
799 2005-11-28 Matthias Clasen <mclasen@redhat.com>
801 * glib/gthread.c (g_static_rw_lock_wait, g_static_rw_lock_signal):
802 * glib/gnode.c (g_node_depth_traverse_level):
803 * glib/gmem.c (g_allocator_new):
804 * glib/ghash.c (g_hash_table_unref, g_hash_table_destroy)
805 (g_hash_table_foreach_remove_or_steal): Silence compiler
808 2005-11-27 Matthias Clasen <mclasen@redhat.com>
810 * glib/gunicollate.c (g_utf8_collate_key_for_filename):
811 Don't read beyond len. (#322520, Christian Persch)
813 Wed Nov 23 17:34:01 2005 Tim Janik <timj@imendio.com>
815 * glib/gdataset.c: access datalist flags via atomic pointer access
816 functions, instead of acquiring the dataset lock. this is faster and
817 also matches the atomic pointer readouts in gdatalistprivate.h and
818 direct pointer modifications required by gobject.c.
820 Wed Nov 23 13:35:31 2005 Tim Janik <timj@gtk.org>
822 * glib/gdataset.c: streamlined initialization code somewhat,
823 removed GData node cache.
825 2005-11-22 Matthias Clasen <mclasen@redhat.com>
827 * glib/ghash.c (g_hash_table_ref):
828 (g_hash_table_unref): Mark these as new API.
830 Avoid double locking in g_intern_string (#322133,
833 * glib/gdataset.c (g_quark_from_string_internal): New
834 internal function which factors out common parts of
835 g_quark_from[_static]_string.
836 (g_quark_from_string, g_quark_from_static_string):
837 Use g_quark_from_string_internal.
838 (g_intern_string, g_intern_static_string): Use
839 g_quark_from_string_internal, and only take the
841 (g_quark_new): Don't store the strings shifted by -1
842 in the g_quarks array.
843 (g_quark_to_string): Adapt to the previous change.
845 Tue Nov 22 14:04:26 2005 Tim Janik <timj@imendio.com>
849 g_hash_table_new_full(): create hash tables with a ref count of 1.
850 g_hash_table_ref(): atomically ref_count+=1
851 g_hash_table_unref(): atomically ref_count-=1, destroys hash table
852 when refcount reaches 0.
853 g_hash_table_destroy(): just destroy keys and values, unref by 1.
854 g_hash_table_insert():
855 g_hash_table_replace(): assert ref_count>0.
858 * glib/gatomic.c: added 'volatile' qualifier to all atomic pointer and
861 2005-11-20 Behdad Esfahbod <behdad@gnome.org>
863 * glib/guniprop.c (g_unichar_get_mirror_char): Remove unused
864 variables. (#321984, Andrew Paprocki)
866 2005-11-18 Matthias Clasen <mclasen@redhat.com>
868 * configure.in: Bump version
870 2005-11-17 Matthias Clasen <mclasen@redhat.com>
872 * === Released 2.9.0 ===
874 * glib/glib.symbols: Updates
876 * glib/gutf8.c (g_utf8_offset_to_pointer): Add some performance
877 hints to the docs. (Paolo Borelli)
881 * glib/gthread.c (g_thread_foreach): Mark as new api.
883 * README.in: Updates.
887 * glib/gdate.c (g_date_set_time_t):
888 (g_date_set_time_val): Add functions to set a date from
889 a time_t and from a GTimeVal.
890 (g_date_set_time): Deprecate. (#314067, Roger Leigh)
893 * tests/utf8-pointer.c: Unit tests for g_utf8_pointer_to_offset
894 and g_utf8_offset_to_pointer.
896 * glib/gutf8.c (g_utf8_pointer_to_offset)
897 (g_utf8_offset_to_pointer): Handle negative offsets, and use
898 "stutter stepping" for going backwards. (#320638, Larry
903 * glib/gthread.c: const correctness fixes, found
904 by Arjan van de Ven and gcc.
906 2005-11-16 Behdad Esfahbod <behdad@gnome.org>
908 * docs/reference/glib/tmpl/unicode.sgml: Correct typo on mentioning
909 @G_UNICODE_BREAK_UNKNOWN that should be #G_UNICODE_BREAK_UNKNOWN.
911 2005-11-09 Behdad Esfahbod <behdad@gnome.org>
913 * glib/guniprop.c: Use bit hacks instead when checking a general
914 category value against multiple values.
916 * glib/gutf8.c: Change ISO10646 to Unicode in docs.
918 2005-11-10 Simos Xenitellis <simos@gnome.org>
920 * configure.in: Added tt (Tatar) to ALL_LINGUAS.
922 2005-11-05 Tor Lillqvist <tml@novell.com>
924 * glib/giowin32.c (g_io_win32_set_flags): Don't call g_warning().
925 (#320688) Instead set error and return failure.
927 2005-11-07 Matthias Clasen <mclasen@redhat.com>
929 * glib/gmain.c (g_main_dispatch): Don't call cb_funcs->unref
930 while holding the context lock. (#320886, Andy Wingo)
932 2005-11-07 Billy Biggs <vektor@dumbterm.net>
934 * tests/refcount/Makefile.am: Add a missing reference to libglib
935 in the LDADD for this test.
937 2005-11-07 Matthias Clasen <mclasen@redhat.com>
939 * glib/gmem.c (g_allocator_new): We need to set n_preallocs to a
940 nonzero value, otherwise GTK+ 2.8 breaks when compiled against
941 GLib 2.9. (#320755, Luca Ferretti)
943 2005-11-04 Matthias Clasen <mclasen@redhat.com>
945 * glib/Makefile.am: Apply a patch from Behdad Esfahbod to
946 use a faster lookup table for g_unichar_get_mirror_char().
947 * glib/gmirroringtable.h: The new table.
948 * glib/gunichartables.h: Remove bidi_mirroring_table.
949 * glib/gen-unicode-tables.pl: Don't generate the mirroring
951 * glib/glib-mirroring-tab/*: Sources for the program
952 which generated gmirroringtable.h.
954 * glib/glist.c: Avoid some code duplication.
956 * glib/gscanner.h: Include gdataset.h. (#320322)
958 * glib/gdebug.h: add new GDebugFlag for fatal_criticals
959 * glib/gmessages.c: (_g_debug_init): handle G_DEBUG=fatal_criticals,
960 to help find critical warnings in applications. (#320017,
963 2005-11-02 Tor Lillqvist <tml@novell.com>
965 * glib/glib.symbols: Remove large amount of trailing whitespace
966 from one line. Remove #ifdef G_UNDEFINED from around
967 g_hash_table_new and g_hash_table_new_full symbols, what was the
968 purpose of that? Add the g_slice functions.
970 Tue Nov 1 16:24:20 2005 Tim Janik <timj@imendio.com>
972 * glib/gmem.[hc]: prepared deprecation of GMemChunk and GAllocator.
973 added g_slice_*() API to allocate and cache small bits of memory.
974 an actuall allocator implementation for g_slice_*() is still pending.
976 * glib/gthread.[hc]: changes from a patch by Matthias Clasen.
977 changed GRealThread list to use in-structure *next; fields instead
978 of GSList, in order for thread iteration to not depenend on g_slice_*()
980 _g_thread_mem_private_get():
981 _g_thread_mem_private_set(): added accessors for private memory,
982 needed because the ordinary GPrivate implementation relies on GArray
983 and GSList and therefore indirectly on working g_slice_*() allocations.
986 g_thread_foreach(): new public API function to loop over all existing threads.
1000 removed GAllocator and free list usages and accompanying locks.
1001 use g_slice_*() API to allocate and cache small bits of memory.
1003 * glib/ghook.h: removed GMemChunk field from public API.
1006 * glib/glist.h: deprecate allocator API, provide _free1() for consistency.
1008 * glib/gnode.h: deprecate allocator API.
1010 * glib/gmain.c: reordered GPollRec fields so g_slice_free_chain() can
1011 be used for poll rec lists.
1013 * glib/grel.c: removed mem chunk usage, and allocated tuples via g_slice_*().
1014 g_relation_destroy(): free all tuples from the all_tuples hash table,
1015 this effectively maintains the life time track keeping of tuples.
1016 g_relation_delete_tuple(): free tuples which are removed from the
1017 all_tuples hash table. this fixes a temporary leak that was present
1018 in the memchunk code until the destruction of the relation.
1020 2005-10-29 Matthias Clasen <mclasen@redhat.com>
1022 * tests/convert-test.c: Add some tests for conversions between
1023 UTF-8, UCS-4 and UTF-16.
1025 * glib/gutf8.c (g_utf8_to_ucs4, g_utf8_to_utf16): Fix handling
1026 of len == -1, noticed by Morten Welinder.
1028 2005-10-27 Erdal Ronahi <erdal.ronahi@gmail.com>
1030 * configure.in: Added ku (Kurdish) to ALL_LINGUAS
1032 2005-10-26 Matthias Clasen <mclasen@redhat.com>
1034 * glib/gutf8.c (g_ucs4_to_utf8): Don't set items_read twice
1035 in the error case, and add some documentation. (#319806, Morten
1038 2005-10-19 Manish Singh <yosh@gimp.org>
1040 * configure.in: Use AC_CHECK_FUNCS for _NSGetEnviron, to get the
1041 config.h symbol automatically. Fixes bug #313731.
1043 2005-10-19 Tor Lillqvist <tml@novell.com>
1045 * glib/gwin32.c (g_win32_get_package_installation_directory):
1046 Return a g_strdup()ed copy of the value stored in the hash table,
1047 so that it can be g_free()d without leaving a dangling pointer in
1048 the hash table. (#319232)
1050 2005-10-06 Matthias Clasen <mclasen@redhat.com>
1052 * glib/gunicollate.c (g_utf8_collate_key_for_filename): Handle
1053 all-zero sequences correctly. (#317930, Sebastien Bacher)
1055 2005-10-05 Matthias Clasen <mclasen@redhat.com>
1057 * glib/gmarkup.c (g_markup_escape_text): Doc addition
1059 * Makefile.am: Add Collation tests.
1061 * tests/collate/*: Inputs and expected outputs for collation tests.
1063 * tests/run-collate-tests.sh: Script to run collation tests.
1065 * tests/unicode-collate.c (main): Rework slightly to make
1066 it usable in unit tests. Also test g_utf8_collate_key_for_filename().
1068 2005-10-01 Behdad Esfahbod <behdad@gnome.org>
1070 * docs/reference/glib/tmpl/unicode.sgml:
1071 * glib/gen-unicode-tables.pl:
1073 * glib/gunichartables.h:
1075 * tests/casefold.txt:
1076 * tests/casemap.txt: Updated to Unicode 4.1. There are five new
1077 GUnicodeBreakType types. That may break some applications, like
1080 2005-09-26 Matthias Clasen <mclasen@redhat.com>
1082 * glib/gstrfuncs.c (g_ascii_strtoull): Add details to the
1083 docs. (#314393, Matthew F. Barnes)
1085 * glib/glib.symbols:
1086 * glib/gprintf.h: Remove g_snprintf() and g_vsnprintf(), since
1087 they are already declared in glib.h. This doesn't break documented
1088 use of gprintf.h, but should probably be pointed out in the
1089 release notes for 2.10. (#314232, Behdad Esfahbod)
1091 Tue Sep 20 13:16:04 2005 Tim Janik <timj@imendio.com>
1093 * glib/gpattern.c (g_pattern_ph_match): applied significant recursion
1094 complexity optimization, based on a patch from Matthias Clasen.
1096 * tests/patterntest.c: more tests, mostly from matthias.
1098 2005-09-20 Matthias Clasen <mclasen@redhat.com>
1100 * glib/gqueue.c (g_queue_insert_sorted): Correct the docs.
1101 (#316703, Mark Drago)
1103 Mon Sep 19 17:23:23 2005 Tim Janik <timj@imendio.com>
1105 * glib/gpattern.c: applied a patch from matthias which checks on the
1106 upper bound of GPatternSpec length to optimize matches.
1109 * tests/patterntest.c: added more match cases.
1111 2005-09-16 Tor Lillqvist <tml@novell.com>
1113 * glib/gstrfuncs.c (g_ascii_strcasecmp, g_ascii_strncasecmp): Add
1114 warning to doc comment that these functions should not be used on
1115 encodings like CP932.
1117 2005-09-14 Matthias Clasen <mclasen@redhat.com>
1119 * tests/keyfile-test.c: Add a test for grup names of length 1.
1121 * glib/gkeyfile.c (g_key_file_line_is_group): Accept group names
1122 of length 1. (#316309)
1124 2005-09-12 Matthias Clasen <mclasen@redhat.com>
1126 * glib/gmarkup.c (g_markup_escape_text): Clarify docs.
1128 2005-09-11 Sebastian Wilhelmi <seppi@seppi.de>
1130 * tests/refcount/Makefile.am (INCLUDES): Link the the refcount
1131 tests to the system thread library $(G_THREAD_LIBS). Fixes #313744
1134 2005-09-11 Kjartan Maraas <kmaraas@gnome.org>
1136 * glib/gmain.c: (g_child_watch_prepare), (g_child_watch_check),
1137 (child_watch_helper_thread): Remove some dead code. Closes
1140 2005-09-07 Tor Lillqvist <tml@novell.com>
1142 * glib/Makefile.am: Create also a console version of the
1143 gspawn-win32-helper program, gspawn-win32-helper-console.exe.
1144 It's otherwise identical to gspawn-win32-helper.exe, except marked
1145 as a console application (linked without the -mwindows option).
1147 * glib/gspawn-win32.c (do_spawn_directly, do_spawn_with_pipes):
1148 Drop the dont_wait parameter. Its truth value correlated 100% with
1149 the NULLness of the exit_status parameter anyway, so it's enough
1150 to check whether exit_status is NULL. Invert the sense of the
1151 dont_return_handle parameter and rename it to do_return_handle, to
1152 make the code easier to read by avoiding double negations.
1154 (g_spawn_sync_utf8, g_spawn_async_with_pipes_utf8): Modify calls
1155 to do_spawn_with_pipes() accordingly.
1157 (do_spawn_with_pipes): If we have a console, use the console
1158 version of the helper program, otherwise use the GUI one. This
1159 avoids extra console windows opening up in some situations. (In
1160 case a console application uses the GUI gspawn-win32-helper.exe to
1161 spawn another console application we would get a separate console
1162 for the spawned console application).
1164 * glib-zip.in: Distribute also gspawn-win32-helper-console.exe.
1166 2005-09-05 Matthias Clasen <mclasen@redhat.com>
1168 * glib/gmappedfile.c (g_mapped_file_new): Report an error
1169 if the file is too large. (#315275, Kjartan Maraas)
1171 * glib/gkeyfile.c (g_key_file_load_from_fd): The return value
1172 of read() is signed. (#315273, Kjartan Maraas)
1174 2005-08-31 Tor Lillqvist <tml@novell.com>
1176 * glib/gutils.h: Wrapping atexit() is a bad idea on Windows, where
1177 the EXE and each DLL have their own atexit function chains.
1179 #define g_atexit as atexit instead. This means it has a
1180 better chance of doing what the caller wants. For instance,
1181 gtkhtml calls g_atexit() registering a function in gtkhtml
1182 itself. This caused a crash when g_atexit() was implemented as a
1183 function in the GLib DLL. The gtkhtml DLL was already unloaded by
1184 the time the GLib DLL got unloaded.
1186 * glib/gutils.c: #undef the #define mentioned above, to also get a
1187 real g_atexit() into the DLL for backward compatibility. Document
1188 the Windows behaviour of g_atexit(), and document the varying ways
1189 atexit() can behave in the context of dynamically loaded modules
1192 2005-08-31 Matthias Clasen <mclasen@redhat.com>
1194 * glib/glib.symbols:
1196 * glib/gdataset.c: Add string interning functions.
1198 2005-08-28 Matthias Clasen <mclasen@redhat.com>
1200 * glib/giochannel.c: Unify some near-duplicate strings. (#314654,
1203 2005-08-26 Matthias Clasen <mclasen@redhat.com>
1205 * configure.in: Bump version to 2.9.0
1207 2005-08-25 Tor Lillqvist <tml@novell.com>
1209 Make also the g_spawn*() functions take parameters in the GLib
1210 file name encoding, i.e. UTF-8, on Windows. Has no impact on Unix
1211 API or ABI. Like the other GLib API that was earlier changed to
1212 use UTF-8 on Windows, the names of the functions that take UTF-8
1213 have _utf8 suffixes added by using preprocessor macros in the
1214 header file. The old names are kept for functions with the old
1215 behaviour, taking parameters in the system codepage, for DLL ABI
1218 * glib/gspawn.h: On Win32 add the suffix _utf8 to the names of the
1219 g_spawn*() functions.
1221 * glib/gspawn-win32.c: Use wide-char API on NT-based
1222 Windows. Convert parameters from UTF-8 to wide chars (NT) or
1223 system codepage (Win9x) and call the C library _wspawn*() or
1224 spawn*() functions respectvely. Add DLL ABI stability versions
1225 that take parameters in the system codepage.
1227 * glib/gspawn-win32-helper.c: On NT-based Windows use the
1228 wide-char versions of argv and envp, and use wide-char API to
1229 change directory and spawn the program to run. Remove the verbose
1230 debugging output, it was too complex to modify for the wide-char
1231 features. (Just add temporary debugging printouts if needed, no
1232 need to have them permanently in the source.)
1234 * glib/gspawn.c: Corresponding documentation updates.
1236 * glib/glib.symbols: Corresponding changes: Mark the ABI stability
1237 symbols as PRIVATE, add the new _utf8-suffixed ones.
1239 2005-08-24 Stepan Kasal <kasal@ucw.cz>
1241 * glib/gtypes.h (G_MININT64): Cast the constant to gint64; it is
1242 guint64 otherwise and that can produce warnings about comparison
1243 between signed and unsigned.
1245 2005-08-23 Matthias Clasen <mclasen@redhat.com>
1247 * glib/gutils.c: Fix the crt_externs.h include.
1249 2005-08-23 Stepan Kasal <kasal@ucw.cz>
1251 * NEWS: Fix spelling of my first name.
1253 2005-08-23 Matthias Clasen <mclasen@redhat.com>
1257 * === Released 2.8.1 ===
1261 2005-08-20 Hans Breuer <hans@breuer.org>
1263 * glib/makefile.msc.in : link with ws2_32.lib
1265 2005-08-18 Tor Lillqvist <tml@novell.com>
1267 * configure.in: Check for <sys/wait.h>
1269 * glib/gbacktrace.c: Include <sys/wait.h> on if HAVE_SYS_WAIT_H.
1271 2005-08-18 Ross Burton <ross@burtonini.com>
1274 Optimise single-character insertions.
1279 * tests/string-test.c:
1280 Add tests for new optimisation, and fix a leak.
1282 2005-08-17 Matthias Clasen <mclasen@redhat.com>
1284 * configure.in: Check for crt_externs.h and _NSGetEnviron.
1286 * glib/gutils.c: On Darwin, include crt-externs.h and
1287 define environ using _NSGetEnviron(). (#313731)
1289 2005-08-16 Stepan Kasal <kasal@ucw.cz>
1291 * glib/gutils.c (g_get_any_init): Move the body of the big if...
1292 (g_get_any_init_do): ... to this new function.
1293 (g_get_any_init): Declare as inline.
1294 (g_get_any_init_locked): New inline function, does the locking.
1295 Make use of these two throughout the code.
1297 2005-08-15 Matthias Clasen <mclasen@redhat.com>
1299 * glib/gbacktrace.c (g_on_error_stack_trace): Wait for
1300 the child process and then simply return. This makes
1301 The "S" option work as documented in g_on_error_query().
1302 (#313125, Matthew F. Barnes)
1304 * glib/gunicode.h: Update the link to Unicode category
1305 values. (#313369, Behnam Esfahbod)
1307 * glib/gqueue.c (g_queue_find_custom): Clarify docs
1308 a little. (#311727, Tristan van Berkom)
1310 * glib/abicheck.sh, gobject/abicheck.sh: Make the
1311 check work on ia64 too, where some symbols ended up
1312 in yet another section.