1 2006-01-17 Matthias Clasen <mclasen@redhat.com>
3 * glib/gkeyfile.c (g_key_file_get_groups): Correct a g_new0()
4 call. (#327421, Morten Welinder)
6 2006-01-17 Sebastian Wilhelmi <seppi@seppi.de>
8 * glib/gthreadpool.c: To avoid deadlocks get rid of the settings
9 G_LOCK. Use the unused_thread_queue lock instead. Change
10 g_thread_pool_thread_proxy such that threads only wait on
11 non-exlusive pools for at most a 1/2 second. Do not reorder tasks
12 due to superfluous tasks. Global tasks wait at most for
13 max-idle-time milliseconds. Make sure, that no task is woken up
14 twice for the same event via a wakeup_serial. This fixes #324228.
16 * tests/threadpool-test.c: Adapt test accordingly. Do not pass
17 invalid NULL into the thread pools. This as well fixes #327290.
19 2006-01-16 Matthias Clasen <mclasen@redhat.com>
21 * configure.in: Bump version
23 * === Released 2.9.3 ===
27 * glib/gthread.h (g_static_mutex_get_mutex_impl_shortcut):
28 Rework the strict aliasing fix to not break C++, pointed
29 out by Murray Cumming.
31 * glib/gasyncqueue.c (g_async_queue_push_sorted_unlocked):
32 Signal waiting threads, problem noticed by Christian Kellner.
34 2006-01-16 Matthias Clasen <mclasen@redhat.com>
36 Fix bug 326747, Alberto Ruiz:
38 * glib/gconvert.c (make_valid_utf8): Change this function to
39 replace unknown characters by the Unicode replacement character
40 instead of '?', and don't append "(invalid encoding)".
41 (g_filename_display_name, g_filename_display_basename): Document
42 how to determine if the filename was in an invalid encoding.
44 2006-01-14 Matthias Clasen <mclasen@redhat.com>
46 * glib/gtree.c: Replace the simple recursive implementation
47 by a nonrecursive, threaded implementation by Maurizio
50 2006-01-12 Matthias Clasen <mclasen@redhat.com>
52 * tests/asyncqueue-test.c (main): Fix the
53 assert to not always trigger. (#326558,
56 2006-01-11 Matthias Clasen <mclasen@redhat.com>
58 * tests/convert-test.c: Don't test items_read and
59 items_written in error cases where it is not set.
61 * tests/asyncqueue-test.c: Set error to NULL.
63 2006-01-08 Matthias Clasen <mclasen@redhat.com>
65 * glib/gtree.c: Break some long lines.
67 * tests/tree-test.c: Add more tests.
69 * glib/gslice.c: Include config.h first, pointed out
72 2005-12-20 Sven Herzberg <herzi@gnome-de.org>
74 * docs/reference/gobject/tmpl/gtype.sgml: explain that
75 G_TYPE_INSTANCE_GET_CLASS() does behave different during
78 2006-01-06 Matthias Clasen <mclasen@redhat.com>
80 * glib/gslice.c: Only define _XOPEN_SOURCE if we know
81 that we have posix_memalign(). (#323937, Bogdan Nicula)
83 2006-01-05 Matthias Clasen <mclasen@redhat.com>
85 * configure.in: Bump version
87 * === Released 2.9.2 ===
89 2006-01-05 Hans Breuer <hans@breuer.org>
91 * glib/makefile.msc.in : added gslice.obj
93 2006-01-05 Matthias Clasen <mclasen@redhat.com>
97 * glib/gmain.c: Clarify the documentation of source ids
98 a bit. (#325874, Dan Williams)
100 * configure.in: Fix another strict aliasing problem.
102 * glib/gthreadpool.c: Avoid use of varargs macro. (#325864,
105 2006-01-04 Manish Singh <yosh@gimp.org>
108 * glib/gstdio.h: don't macroized g_access, g_chdir, and g_unlink
109 either, since they have the same issue as g_rmdir. (Related to
112 Wed Jan 4 13:33:25 2006 Tim Janik <timj@gtk.org>
114 * glib/gslice.c (magazine_cache_pop_magazine): don't reverse chunk
115 order when creating magazines, so we hand out chunks with ascending
118 2006-01-04 Abel Cheung <maddog@linuxhall.org>
120 * configure.in: Added "ml" "zh_HK" to ALL_LINGUAS.
122 2006-01-03 Martyn Russell <martyn@imendio.com>
124 * docs/reference/glib/glib-sections.txt:
126 * glib/gthreadpool.[ch]:
127 - Added new API g_thread_pool_get_idle_time() and
128 g_thread_pool_set_idle_time(). (#324228).
130 * tests/threadpool-test.c:
131 - Updated test case to do thread pool sorting, thread pool with
132 no sorting and a thread pool with idle thread timeouts.
134 2006-01-03 Matthias Clasen <mclasen@redhat.com>
136 * glib/gmain.h: Add new functions here, too.
138 * glib/glib.symbols: Add new functions.
140 2005-12-20 Michael Meeks <michael.meeks@novell.com>
142 * glib/gmain.c (g_main_context_is_owner): new method
143 to determine if the current thread is the owner of the
146 2006-01-02 Matthias Clasen <mclasen@redhat.com>
150 * glib/gstdio.c (g_rmdir): Don't provide g_rmdir() as a macro
151 expanding to rmdir, since rmdir is not declared in a portable
152 system header we can include in gstdio.h. (#325249, Jani Monoses)
154 2006-01-01 Tor Lillqvist <tml@novell.com>
156 * glib/gspawn-win32.c (g_spawn_sync_utf8): Set the GIOChannels for
157 stdout and stderr to unbuffered. Otherwise the giochannel layer
158 will try to read from them regardless whether the
159 g_io_channel_win32_poll() call here has indicated
160 readability or not. (#325310)
162 2005-12-29 Matthias Clasen <mclasen@redhat.com>
164 * glib/gutils.c (glib_check_version): Fix a copy-and-paste error
165 in the docs. (#325273, Declan Naughton)
167 2005-12-27 Manish Singh <yosh@gimp.org>
169 * tests/run-collate-tests.sh: use LC_ALL instead of LC_COLLATE, to
170 make sure we really override things.
172 2005-12-27 Matthias Clasen <mclasen@redhat.com>
174 Fix #316221, Michal Benes, Stanislav Brabec;
176 * configure.in: Fix a strict aliasing problem in
177 g_static_mutex_get_mutex().
178 * glib/gthread.h: ...and in
179 g_static_mutex_get_mutex_impl_shortcut().
181 * glib/gdatasetprivate.h: Add a cast to silence compiler
182 warnings. (#321978, Andrew Paprocki)
184 Partial fix for bug #323937, Bogdan Nicula.
186 * configure.in: Check for malloc.h
187 * glib/gslice.c: Don't include malloc.h unconditionally.
189 2005-12-27 Manish Singh <yosh@gimp.org>
191 * tests/run-collate-tests.sh: export LC_COLLATE so it takes.
192 (#324950, Dan Yefimov)
194 2005-12-26 Matthias Clasen <mclasen@redhat.com>
196 * glib/gslice.c: On Win32, include process.h (#325015, Kazuki
199 2005-12-25 Matthias Clasen <mclasen@redhat.com>
201 * AUTHORS: Update my email
203 * tests/slice-test.c: Fix C99isms. (#324950, Dan Yefimov)
205 2005-12-21 Matthias Clasen <mclasen@redhat.com>
207 * README.in: Add some notes on when 'make check' may fail.
209 2005-12-20 Matthias Clasen <mclasen@redhat.com>
212 * glib/gthreadpool.h:
213 * glib/gthreadpool.c (g_thread_pool_set_sort_function): New function
214 to sort tasks pushed into a threadpool. (#324479, Martyn Russell)
216 * tests/threadpool-test.c: Test this.
218 Tue Dec 20 18:14:14 2005 Tim Janik <timj@imendio.com>
220 * glib/gslice.[hc]: added mem_error() and mem_assert() to test and
221 handle errors without depending on gmessage.c which might not be
222 setup when the error occours.
223 removed G_SLICE_CONFIG_ALWAYS_FREE config option, fixed the code so
224 always freeing can be achieved by adjusting the working set time to
225 0 with G_SLICE_CONFIG_WORKING_SET_MSECS.
226 added G_SLICE_CONFIG_COLOR_INCREMENT to test different color increments
227 (mainly 0 and 1). reduced the minimum block size to 128 bytes, to
228 minimize wastage if small amounts of differently sized structrues are
229 allocated, this does come at a performance cost of roughly 5% though.
230 fixed up block alignment calculation, so it works for varying
231 block sizes. only use strerror() not g_strerror() since the latter
232 depends on working GQuark and GSlice.
233 mem_error(): implemented in terms of fprintf and vfprintf.
235 * tests/slice-color.c: new program to test cache colorization effects.
237 * tests/slice-test.c: trade G_SLICE_CONFIG_ALWAYS_FREE for 0 duration
238 G_SLICE_CONFIG_WORKING_SET_MSECS.
240 2005-12-17 Matthias Clasen <mclasen@redhat.com>
242 * glib/goption.c (parse_short_option): Set an error in all
243 failure cases. (#324332, Tim-Philipp Müller)
245 2005-12-17 Sebastian Wilhelmi <seppi@seppi.de>
247 * glib/gatomic.c: Fix memory barrier position in g_atomic_int_get
248 and g_atomic_pointer_get. Add g_atomic_int_set and
249 g_atomic_pointer_set implementations for the !DEFINE_WITH_MUTEXES &&
250 G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case, as well as defining them
251 as functions (additionally to the macros in the header) for the
252 !G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case.
254 2005-12-16 Matthias Clasen <mclasen@redhat.com>
256 * glib/gmem.c (g_allocator_new): Don't return a pointer to
257 a const struct, since apps expect to be able to modify it.
258 (#324179, J. Ali Harlow)
260 Tue Dec 13 10:13:32 2005 Tim Janik <timj@imendio.com>
262 * glib/gatomic.h: added g_atomic_pointer_set() and g_atomic_int_set()
263 for systems where the initialization of atomic variables requires a
264 write memory barrier.
266 2005-12-09 Matthias Clasen <mclasen@redhat.com>
268 * configure.in: Bump version
270 * === Released 2.9.1 ===
272 2005-12-09 Alexander Larsson <alexl@redhat.com>
276 Use G_GNUC_WARN_UNUSED_RESULT on list functions that return
279 2005-12-08 Matthias Clasen <mclasen@redhat.com>
283 2005-12-07 Martyn Russell <martyn@imendio.com>
285 * glib/gasyncqueue.c:
286 - Call g_queue_insert_sorted() instead of duplicating the code.
287 - Call g_queue_sort() instead of duplicating the code.
288 - Invert sort function results to make sure the same sort function
289 gives the same results across glist, gslist, gqueue and
292 * tests/asyncqueue-test.c:
293 - Updated the sort function to reflect the example in the
294 documentation for gasyncqueue.c.
296 2005-12-07 Martyn Russell <martyn@imendio.com>
298 * docs/reference/glib/glib-sections.txt:
299 * docs/reference/glib/tmpl/linked_lists_double.sgml:
300 * docs/reference/glib/tmpl/linked_lists_single.sgml:
303 - Added g_list_insert_sorted_with_data () and
304 g_slist_insert_sorted_with_data ().
305 - Removed the extra check in g_list_sort() and g_slist_sort() for
306 GCompareDataFunc vs. GCompareFunc.
308 2005-12-07 Tor Lillqvist <tml@novell.com>
310 * glib-gettextize.in: Look up prefix at run-time on Win32,
311 assuming the standard directory structure with glib-gettextize in
314 2005-12-06 Matthias Clasen <mclasen@redhat.com>
316 * glib/gmem.h: Exempt GTK+ from the mem chunk deprecation, since
317 we need GTK+ 2.8 to compile against GLib 2.10.
319 2005-12-05 Manish Singh <yosh@gimp.org>
321 * tests/run-collate-test.sh: set LC_COLLATE, not LANG, to be sure
322 to override any user settings.
324 2005-12-05 Matthias Clasen <mclasen@redhat.com>
326 * configure.in: Revert an accidential commit
328 2005-12-05 Matthias Clasen <mclasen@redhat.com>
330 * glib/gutf8.c: Documentation updates. (#323291, Morten Welinder)
332 * tests/Makefile.am (TESTS_ENVIRONMENT):
333 * tests/gobject/Makefile.am (TESTS_ENVIRONMENT): Set
334 MALLOC_CHECK_ and MALLOC_PERTURB_
336 * tests/run-collate-tests.sh: Run the collation tests explicitly
337 in en_US locale. (#320463)
339 * glib/gmem.h: Really deprecate GMemChunk.
346 * glib/gstrfuncs.c: Add versioned deprecation docs.
348 2005-12-05 Martyn Russell <martyn@imendio.com>
350 * docs/reference/glib/glib-sections.txt:
351 * glib/gasyncqueue.[ch]:
352 - Added support for sorting async queues by with _push_sorted(),
353 _push_sorted_unlocked(), _sort() and _sort_unlocked() (#323047).
356 * tests/asyncqueue-test.c:
357 - Added test case for gasyncqueue.c
359 Mon Dec 5 15:53:20 2005 Tim Janik <timj@imendio.com>
361 * glib/gslice.c: implement chain walking for arbitrary ->next pointer
362 offsets in g_slice_free_chain_with_offset() based on a patch by behdad
363 in bug 323178. moved time consuming logic from g_slice_free() out of
364 the inner loop, so g_slice_free_chain_with_offset() provides a real
365 performance benefit over g_slice_free1() now.
367 * glib/gslice.h: renamed g_slice_free_chain() to
368 g_slice_free_chain_with_offset(). implemented g_slice_free_chain() as
369 a type-safe macro as suggested in bug 323178.
370 simplified the macro implementation of g_slice_free() and implemented
371 it in a type safe manner for all compliers as suggested by Morten
372 Welinder <mortenw@gnome.org>.
377 * glib/glib.symbols: s/g_slice_free_chain/g_slice_free_chain_with_offset/
379 2005-12-05 Matthias Clasen <mclasen@redhat.com>
381 * glib/gasyncqueue.c: Add some docs.
383 * tests/libmoduletestplugin_a.c: Fix compiler warnings.
385 * glib/gatomic.c: In the ia64 implementation, use
386 __sync builtin without _si or _di suffix. (#321229,
387 Stanislav Brabec, patch by Andreas Schwab)
389 2005-12-04 Behdad Esfahbod <behdad@gnome.org>
391 * glib/gslice.h: Remove comma at the end of enum.
393 2005-12-04 Matthias Clasen <mclasen@redhat.com>
395 Handle multiple user names with the same UID better.
396 (#319535, Laszlo Peter)
398 * glib/gutils.c (g_get_any_init_do): When determining user
399 data, first look up $LOGNAME. If the UID doesn't match
400 getuid(), fall back to the current behaviour of looking
401 up the user data based on getuid().
403 2005-12-04 Matthias Clasen <mclasen@redhat.com>
405 * glib/gmacros.h (G_GNUC_WARN_UNUSED_RESULT): Add a macro
406 to make gcc warn if a function result is ignored. (#145466,
407 Arjan van de Ven, Alex Larsson)
409 * glib/gmem.h: Add the new attribute to g_realloc and
412 2005-12-03 Matthias Clasen <mclasen@redhat.com>
415 * glib/gcache.h: Deprecate g_cache_value_foreach. (#322956,
418 * glib/gmappedfile.c: Make mapping of empty files
421 * glib/gfileutils.c: Don't fork a new process just to
422 fix the permissions of the created temp file. (#321318,
423 Alexis S. L. Carvalho)
425 2005-12-02 Matthias Clasen <mclasen@redhat.com>
427 * README.in: Add a note about Solaris threads.
430 * configure.in: Remove support for Solaris threads.
431 (#136971, Sebastian Wilhelmi, patch by Andrew Paprocki)
433 2005-12-02 Matthias Clasen <mclasen@redhat.com>
435 * configure.in: Define G_GUINT64_CONSTANT in analogy to
436 G_GINT64_CONSTANT. (#322568, Andrew Paprocki)
438 2005-12-02 Matthias Clasen <mclasen@redhat.com>
440 * glib/gslice.c: Win32 portability fixes and C99ism removal,
441 pointed out by Kazuki Iwamoto. (#323052)
443 2005-12-02 Matthias Clasen <mclasen@redhat.com>
445 * glib/gatomic.c (g_atomic_int_add): Add a missing volatile for
446 the IA64 implementation.
448 Fri Dec 2 16:18:09 2005 Tim Janik <timj@imendio.com>
450 * glib/gslice.c: some naming and type size fixups.
452 Fri Dec 2 13:08:58 2005 Tim Janik <timj@gtk.org>
454 * glib/gslice.h (g_slice_free): reworked GCC-specific type-safe macro
455 variant into something less verbose, but digestible for gcc-3.4.
457 Fri Dec 2 10:55:07 2005 Tim Janik <timj@imendio.com>
459 * tests/slice-test.c: extended to perform the benchmarking on the old
460 memchunk code if 'O' is selected.
462 * tests/memchunks.c: new file which contains the old GLib mem chunks
463 implementation with prefix old_mem_chunk_.
465 * tests/Makefile.am: added memchunks.c
467 Fri Dec 2 00:16:59 2005 Tim Janik <timj@imendio.com>
469 * glib/gslice.c: improved gettimeofday() timer resolution by maintaining
470 acceptable load distribution of the syscall.
471 (allocator_get_magazine_threshold): reduce minimum/single-thread magazine
472 sizes to approximately page_size/2.5 to avoid excessive startup allocations.
474 2005-12-01 Matthias Clasen <mclasen@redhat.com>
476 * tests/gobject/ifacecheck.c (test_iface_base_init): Don't
477 double-free base interface members.
479 * tests/Makefile.am: Remove duplicate variable.
481 Thu Dec 1 17:32:46 2005 Tim Janik <timj@imendio.com>
483 * glib/gslice.[hc]: new slice allocator implementation.
485 * tests/slice-test.c: added random slice allocation test.
487 * glib/gthread.[hc]: removed newly added private thread mem API.
489 * glib/gthreadinit.h:
492 * glib/gmem.c: divided glib threading initialisation into three phases,
493 initialisation where private keys and messaging are not available (only
494 needed by gmem.c), initialisation without messaging but private keys
495 available (gslice.c, gmessage.c), and full fledged initialisers that
496 server the rest of glib. initialisation functions got renamed to reflect
497 the limitations of their corresponding phases.
499 * glib/gmem.c: removed memchunk code, defer allocations to
502 * glib/gmem.[hc]: removed g_slice_* skeletons.
504 * glib/glib.symbols: added g_slice_* symbols.
506 * configure.in: check for availability of posix_memalign(3), memalign(3)
509 * glib/Makefile.am: added gslice.[hc].
511 2005-12-01 Tor Lillqvist <tml@novell.com>
513 * glib/gstdio.c (g_stat): In the Win32 implementation, strip
514 trailing slash(es) for non-root folders. stat() fails if non-root
515 folders are specified with trailing slashes. It's too much hassle
516 to demand that callers strip such slashes themselves, especially
517 as it is easy to get it wrong and strip the slash of a root
519 (g_rename): On NT-based Windows, use MoveFileEx() with
520 MOVEFILE_REPLACE_EXISTING to better match Unix behaviour.
522 2005-11-28 Matthias Clasen <mclasen@redhat.com>
524 Fix G_STMT_START / G_STMT_END on Solaris. (#321972,
527 * configure.in: Check whether do { } while (0) works.
529 * glib/gmacros.h: Use do { } while (0) for G_STMT_START /
530 G_STMT_END if it works.
532 2005-11-28 Matthias Clasen <mclasen@redhat.com>
534 * glib/gthread.c (g_static_rw_lock_wait, g_static_rw_lock_signal):
535 * glib/gnode.c (g_node_depth_traverse_level):
536 * glib/gmem.c (g_allocator_new):
537 * glib/ghash.c (g_hash_table_unref, g_hash_table_destroy)
538 (g_hash_table_foreach_remove_or_steal): Silence compiler
541 2005-11-27 Matthias Clasen <mclasen@redhat.com>
543 * glib/gunicollate.c (g_utf8_collate_key_for_filename):
544 Don't read beyond len. (#322520, Christian Persch)
546 Wed Nov 23 17:34:01 2005 Tim Janik <timj@imendio.com>
548 * glib/gdataset.c: access datalist flags via atomic pointer access
549 functions, instead of acquiring the dataset lock. this is faster and
550 also matches the atomic pointer readouts in gdatalistprivate.h and
551 direct pointer modifications required by gobject.c.
553 Wed Nov 23 13:35:31 2005 Tim Janik <timj@gtk.org>
555 * glib/gdataset.c: streamlined initialization code somewhat,
556 removed GData node cache.
558 2005-11-22 Matthias Clasen <mclasen@redhat.com>
560 * glib/ghash.c (g_hash_table_ref):
561 (g_hash_table_unref): Mark these as new API.
563 Avoid double locking in g_intern_string (#322133,
566 * glib/gdataset.c (g_quark_from_string_internal): New
567 internal function which factors out common parts of
568 g_quark_from[_static]_string.
569 (g_quark_from_string, g_quark_from_static_string):
570 Use g_quark_from_string_internal.
571 (g_intern_string, g_intern_static_string): Use
572 g_quark_from_string_internal, and only take the
574 (g_quark_new): Don't store the strings shifted by -1
575 in the g_quarks array.
576 (g_quark_to_string): Adapt to the previous change.
578 Tue Nov 22 14:04:26 2005 Tim Janik <timj@imendio.com>
582 g_hash_table_new_full(): create hash tables with a ref count of 1.
583 g_hash_table_ref(): atomically ref_count+=1
584 g_hash_table_unref(): atomically ref_count-=1, destroys hash table
585 when refcount reaches 0.
586 g_hash_table_destroy(): just destroy keys and values, unref by 1.
587 g_hash_table_insert():
588 g_hash_table_replace(): assert ref_count>0.
591 * glib/gatomic.c: added 'volatile' qualifier to all atomic pointer and
594 2005-11-20 Behdad Esfahbod <behdad@gnome.org>
596 * glib/guniprop.c (g_unichar_get_mirror_char): Remove unused
597 variables. (#321984, Andrew Paprocki)
599 2005-11-18 Matthias Clasen <mclasen@redhat.com>
601 * configure.in: Bump version
603 2005-11-17 Matthias Clasen <mclasen@redhat.com>
605 * === Released 2.9.0 ===
607 * glib/glib.symbols: Updates
609 * glib/gutf8.c (g_utf8_offset_to_pointer): Add some performance
610 hints to the docs. (Paolo Borelli)
614 * glib/gthread.c (g_thread_foreach): Mark as new api.
616 * README.in: Updates.
620 * glib/gdate.c (g_date_set_time_t):
621 (g_date_set_time_val): Add functions to set a date from
622 a time_t and from a GTimeVal.
623 (g_date_set_time): Deprecate. (#314067, Roger Leigh)
626 * tests/utf8-pointer.c: Unit tests for g_utf8_pointer_to_offset
627 and g_utf8_offset_to_pointer.
629 * glib/gutf8.c (g_utf8_pointer_to_offset)
630 (g_utf8_offset_to_pointer): Handle negative offsets, and use
631 "stutter stepping" for going backwards. (#320638, Larry
636 * glib/gthread.c: const correctness fixes, found
637 by Arjan van de Ven and gcc.
639 2005-11-16 Behdad Esfahbod <behdad@gnome.org>
641 * docs/reference/glib/tmpl/unicode.sgml: Correct typo on mentioning
642 @G_UNICODE_BREAK_UNKNOWN that should be #G_UNICODE_BREAK_UNKNOWN.
644 2005-11-09 Behdad Esfahbod <behdad@gnome.org>
646 * glib/guniprop.c: Use bit hacks instead when checking a general
647 category value against multiple values.
649 * glib/gutf8.c: Change ISO10646 to Unicode in docs.
651 2005-11-10 Simos Xenitellis <simos@gnome.org>
653 * configure.in: Added tt (Tatar) to ALL_LINGUAS.
655 2005-11-05 Tor Lillqvist <tml@novell.com>
657 * glib/giowin32.c (g_io_win32_set_flags): Don't call g_warning().
658 (#320688) Instead set error and return failure.
660 2005-11-07 Matthias Clasen <mclasen@redhat.com>
662 * glib/gmain.c (g_main_dispatch): Don't call cb_funcs->unref
663 while holding the context lock. (#320886, Andy Wingo)
665 2005-11-07 Billy Biggs <vektor@dumbterm.net>
667 * tests/refcount/Makefile.am: Add a missing reference to libglib
668 in the LDADD for this test.
670 2005-11-07 Matthias Clasen <mclasen@redhat.com>
672 * glib/gmem.c (g_allocator_new): We need to set n_preallocs to a
673 nonzero value, otherwise GTK+ 2.8 breaks when compiled against
674 GLib 2.9. (#320755, Luca Ferretti)
676 2005-11-04 Matthias Clasen <mclasen@redhat.com>
678 * glib/Makefile.am: Apply a patch from Behdad Esfahbod to
679 use a faster lookup table for g_unichar_get_mirror_char().
680 * glib/gmirroringtable.h: The new table.
681 * glib/gunichartables.h: Remove bidi_mirroring_table.
682 * glib/gen-unicode-tables.pl: Don't generate the mirroring
684 * glib/glib-mirroring-tab/*: Sources for the program
685 which generated gmirroringtable.h.
687 * glib/glist.c: Avoid some code duplication.
689 * glib/gscanner.h: Include gdataset.h. (#320322)
691 * glib/gdebug.h: add new GDebugFlag for fatal_criticals
692 * glib/gmessages.c: (_g_debug_init): handle G_DEBUG=fatal_criticals,
693 to help find critical warnings in applications. (#320017,
696 2005-11-02 Tor Lillqvist <tml@novell.com>
698 * glib/glib.symbols: Remove large amount of trailing whitespace
699 from one line. Remove #ifdef G_UNDEFINED from around
700 g_hash_table_new and g_hash_table_new_full symbols, what was the
701 purpose of that? Add the g_slice functions.
703 Tue Nov 1 16:24:20 2005 Tim Janik <timj@imendio.com>
705 * glib/gmem.[hc]: prepared deprecation of GMemChunk and GAllocator.
706 added g_slice_*() API to allocate and cache small bits of memory.
707 an actuall allocator implementation for g_slice_*() is still pending.
709 * glib/gthread.[hc]: changes from a patch by Matthias Clasen.
710 changed GRealThread list to use in-structure *next; fields instead
711 of GSList, in order for thread iteration to not depenend on g_slice_*()
713 _g_thread_mem_private_get():
714 _g_thread_mem_private_set(): added accessors for private memory,
715 needed because the ordinary GPrivate implementation relies on GArray
716 and GSList and therefore indirectly on working g_slice_*() allocations.
719 g_thread_foreach(): new public API function to loop over all existing threads.
733 removed GAllocator and free list usages and accompanying locks.
734 use g_slice_*() API to allocate and cache small bits of memory.
736 * glib/ghook.h: removed GMemChunk field from public API.
739 * glib/glist.h: deprecate allocator API, provide _free1() for consistency.
741 * glib/gnode.h: deprecate allocator API.
743 * glib/gmain.c: reordered GPollRec fields so g_slice_free_chain() can
744 be used for poll rec lists.
746 * glib/grel.c: removed mem chunk usage, and allocated tuples via g_slice_*().
747 g_relation_destroy(): free all tuples from the all_tuples hash table,
748 this effectively maintains the life time track keeping of tuples.
749 g_relation_delete_tuple(): free tuples which are removed from the
750 all_tuples hash table. this fixes a temporary leak that was present
751 in the memchunk code until the destruction of the relation.
753 2005-10-29 Matthias Clasen <mclasen@redhat.com>
755 * tests/convert-test.c: Add some tests for conversions between
756 UTF-8, UCS-4 and UTF-16.
758 * glib/gutf8.c (g_utf8_to_ucs4, g_utf8_to_utf16): Fix handling
759 of len == -1, noticed by Morten Welinder.
761 2005-10-27 Erdal Ronahi <erdal.ronahi@gmail.com>
763 * configure.in: Added ku (Kurdish) to ALL_LINGUAS
765 2005-10-26 Matthias Clasen <mclasen@redhat.com>
767 * glib/gutf8.c (g_ucs4_to_utf8): Don't set items_read twice
768 in the error case, and add some documentation. (#319806, Morten
771 2005-10-19 Manish Singh <yosh@gimp.org>
773 * configure.in: Use AC_CHECK_FUNCS for _NSGetEnviron, to get the
774 config.h symbol automatically. Fixes bug #313731.
776 2005-10-19 Tor Lillqvist <tml@novell.com>
778 * glib/gwin32.c (g_win32_get_package_installation_directory):
779 Return a g_strdup()ed copy of the value stored in the hash table,
780 so that it can be g_free()d without leaving a dangling pointer in
781 the hash table. (#319232)
783 2005-10-06 Matthias Clasen <mclasen@redhat.com>
785 * glib/gunicollate.c (g_utf8_collate_key_for_filename): Handle
786 all-zero sequences correctly. (#317930, Sebastien Bacher)
788 2005-10-05 Matthias Clasen <mclasen@redhat.com>
790 * glib/gmarkup.c (g_markup_escape_text): Doc addition
792 * Makefile.am: Add Collation tests.
794 * tests/collate/*: Inputs and expected outputs for collation tests.
796 * tests/run-collate-tests.sh: Script to run collation tests.
798 * tests/unicode-collate.c (main): Rework slightly to make
799 it usable in unit tests. Also test g_utf8_collate_key_for_filename().
801 2005-10-01 Behdad Esfahbod <behdad@gnome.org>
803 * docs/reference/glib/tmpl/unicode.sgml:
804 * glib/gen-unicode-tables.pl:
806 * glib/gunichartables.h:
808 * tests/casefold.txt:
809 * tests/casemap.txt: Updated to Unicode 4.1. There are five new
810 GUnicodeBreakType types. That may break some applications, like
813 2005-09-26 Matthias Clasen <mclasen@redhat.com>
815 * glib/gstrfuncs.c (g_ascii_strtoull): Add details to the
816 docs. (#314393, Matthew F. Barnes)
819 * glib/gprintf.h: Remove g_snprintf() and g_vsnprintf(), since
820 they are already declared in glib.h. This doesn't break documented
821 use of gprintf.h, but should probably be pointed out in the
822 release notes for 2.10. (#314232, Behdad Esfahbod)
824 Tue Sep 20 13:16:04 2005 Tim Janik <timj@imendio.com>
826 * glib/gpattern.c (g_pattern_ph_match): applied significant recursion
827 complexity optimization, based on a patch from Matthias Clasen.
829 * tests/patterntest.c: more tests, mostly from matthias.
831 2005-09-20 Matthias Clasen <mclasen@redhat.com>
833 * glib/gqueue.c (g_queue_insert_sorted): Correct the docs.
834 (#316703, Mark Drago)
836 Mon Sep 19 17:23:23 2005 Tim Janik <timj@imendio.com>
838 * glib/gpattern.c: applied a patch from matthias which checks on the
839 upper bound of GPatternSpec length to optimize matches.
842 * tests/patterntest.c: added more match cases.
844 2005-09-16 Tor Lillqvist <tml@novell.com>
846 * glib/gstrfuncs.c (g_ascii_strcasecmp, g_ascii_strncasecmp): Add
847 warning to doc comment that these functions should not be used on
848 encodings like CP932.
850 2005-09-14 Matthias Clasen <mclasen@redhat.com>
852 * tests/keyfile-test.c: Add a test for grup names of length 1.
854 * glib/gkeyfile.c (g_key_file_line_is_group): Accept group names
855 of length 1. (#316309)
857 2005-09-12 Matthias Clasen <mclasen@redhat.com>
859 * glib/gmarkup.c (g_markup_escape_text): Clarify docs.
861 2005-09-11 Sebastian Wilhelmi <seppi@seppi.de>
863 * tests/refcount/Makefile.am (INCLUDES): Link the the refcount
864 tests to the system thread library $(G_THREAD_LIBS). Fixes #313744
867 2005-09-11 Kjartan Maraas <kmaraas@gnome.org>
869 * glib/gmain.c: (g_child_watch_prepare), (g_child_watch_check),
870 (child_watch_helper_thread): Remove some dead code. Closes
873 2005-09-07 Tor Lillqvist <tml@novell.com>
875 * glib/Makefile.am: Create also a console version of the
876 gspawn-win32-helper program, gspawn-win32-helper-console.exe.
877 It's otherwise identical to gspawn-win32-helper.exe, except marked
878 as a console application (linked without the -mwindows option).
880 * glib/gspawn-win32.c (do_spawn_directly, do_spawn_with_pipes):
881 Drop the dont_wait parameter. Its truth value correlated 100% with
882 the NULLness of the exit_status parameter anyway, so it's enough
883 to check whether exit_status is NULL. Invert the sense of the
884 dont_return_handle parameter and rename it to do_return_handle, to
885 make the code easier to read by avoiding double negations.
887 (g_spawn_sync_utf8, g_spawn_async_with_pipes_utf8): Modify calls
888 to do_spawn_with_pipes() accordingly.
890 (do_spawn_with_pipes): If we have a console, use the console
891 version of the helper program, otherwise use the GUI one. This
892 avoids extra console windows opening up in some situations. (In
893 case a console application uses the GUI gspawn-win32-helper.exe to
894 spawn another console application we would get a separate console
895 for the spawned console application).
897 * glib-zip.in: Distribute also gspawn-win32-helper-console.exe.
899 2005-09-05 Matthias Clasen <mclasen@redhat.com>
901 * glib/gmappedfile.c (g_mapped_file_new): Report an error
902 if the file is too large. (#315275, Kjartan Maraas)
904 * glib/gkeyfile.c (g_key_file_load_from_fd): The return value
905 of read() is signed. (#315273, Kjartan Maraas)
907 2005-08-31 Tor Lillqvist <tml@novell.com>
909 * glib/gutils.h: Wrapping atexit() is a bad idea on Windows, where
910 the EXE and each DLL have their own atexit function chains.
912 #define g_atexit as atexit instead. This means it has a
913 better chance of doing what the caller wants. For instance,
914 gtkhtml calls g_atexit() registering a function in gtkhtml
915 itself. This caused a crash when g_atexit() was implemented as a
916 function in the GLib DLL. The gtkhtml DLL was already unloaded by
917 the time the GLib DLL got unloaded.
919 * glib/gutils.c: #undef the #define mentioned above, to also get a
920 real g_atexit() into the DLL for backward compatibility. Document
921 the Windows behaviour of g_atexit(), and document the varying ways
922 atexit() can behave in the context of dynamically loaded modules
925 2005-08-31 Matthias Clasen <mclasen@redhat.com>
929 * glib/gdataset.c: Add string interning functions.
931 2005-08-28 Matthias Clasen <mclasen@redhat.com>
933 * glib/giochannel.c: Unify some near-duplicate strings. (#314654,
936 2005-08-26 Matthias Clasen <mclasen@redhat.com>
938 * configure.in: Bump version to 2.9.0
940 2005-08-25 Tor Lillqvist <tml@novell.com>
942 Make also the g_spawn*() functions take parameters in the GLib
943 file name encoding, i.e. UTF-8, on Windows. Has no impact on Unix
944 API or ABI. Like the other GLib API that was earlier changed to
945 use UTF-8 on Windows, the names of the functions that take UTF-8
946 have _utf8 suffixes added by using preprocessor macros in the
947 header file. The old names are kept for functions with the old
948 behaviour, taking parameters in the system codepage, for DLL ABI
951 * glib/gspawn.h: On Win32 add the suffix _utf8 to the names of the
952 g_spawn*() functions.
954 * glib/gspawn-win32.c: Use wide-char API on NT-based
955 Windows. Convert parameters from UTF-8 to wide chars (NT) or
956 system codepage (Win9x) and call the C library _wspawn*() or
957 spawn*() functions respectvely. Add DLL ABI stability versions
958 that take parameters in the system codepage.
960 * glib/gspawn-win32-helper.c: On NT-based Windows use the
961 wide-char versions of argv and envp, and use wide-char API to
962 change directory and spawn the program to run. Remove the verbose
963 debugging output, it was too complex to modify for the wide-char
964 features. (Just add temporary debugging printouts if needed, no
965 need to have them permanently in the source.)
967 * glib/gspawn.c: Corresponding documentation updates.
969 * glib/glib.symbols: Corresponding changes: Mark the ABI stability
970 symbols as PRIVATE, add the new _utf8-suffixed ones.
972 2005-08-24 Stepan Kasal <kasal@ucw.cz>
974 * glib/gtypes.h (G_MININT64): Cast the constant to gint64; it is
975 guint64 otherwise and that can produce warnings about comparison
976 between signed and unsigned.
978 2005-08-23 Matthias Clasen <mclasen@redhat.com>
980 * glib/gutils.c: Fix the crt_externs.h include.
982 2005-08-23 Stepan Kasal <kasal@ucw.cz>
984 * NEWS: Fix spelling of my first name.
986 2005-08-23 Matthias Clasen <mclasen@redhat.com>
990 * === Released 2.8.1 ===
994 2005-08-20 Hans Breuer <hans@breuer.org>
996 * glib/makefile.msc.in : link with ws2_32.lib
998 2005-08-18 Tor Lillqvist <tml@novell.com>
1000 * configure.in: Check for <sys/wait.h>
1002 * glib/gbacktrace.c: Include <sys/wait.h> on if HAVE_SYS_WAIT_H.
1004 2005-08-18 Ross Burton <ross@burtonini.com>
1007 Optimise single-character insertions.
1012 * tests/string-test.c:
1013 Add tests for new optimisation, and fix a leak.
1015 2005-08-17 Matthias Clasen <mclasen@redhat.com>
1017 * configure.in: Check for crt_externs.h and _NSGetEnviron.
1019 * glib/gutils.c: On Darwin, include crt-externs.h and
1020 define environ using _NSGetEnviron(). (#313731)
1022 2005-08-16 Stepan Kasal <kasal@ucw.cz>
1024 * glib/gutils.c (g_get_any_init): Move the body of the big if...
1025 (g_get_any_init_do): ... to this new function.
1026 (g_get_any_init): Declare as inline.
1027 (g_get_any_init_locked): New inline function, does the locking.
1028 Make use of these two throughout the code.
1030 2005-08-15 Matthias Clasen <mclasen@redhat.com>
1032 * glib/gbacktrace.c (g_on_error_stack_trace): Wait for
1033 the child process and then simply return. This makes
1034 The "S" option work as documented in g_on_error_query().
1035 (#313125, Matthew F. Barnes)
1037 * glib/gunicode.h: Update the link to Unicode category
1038 values. (#313369, Behnam Esfahbod)
1040 * glib/gqueue.c (g_queue_find_custom): Clarify docs
1041 a little. (#311727, Tristan van Berkom)
1043 * glib/abicheck.sh, gobject/abicheck.sh: Make the
1044 check work on ia64 too, where some symbols ended up
1045 in yet another section.