1 2006-01-05 Matthias Clasen <mclasen@redhat.com>
3 * glib/gmain.c: Clarify the documentation of source ids
4 a bit. (#325874, Dan Williams)
6 * configure.in: Fix another strict aliasing problem.
8 * glib/gthreadpool.c: Avoid use of varargs macro. (#325864,
11 2006-01-04 Manish Singh <yosh@gimp.org>
14 * glib/gstdio.h: don't macroized g_access, g_chdir, and g_unlink
15 either, since they have the same issue as g_rmdir. (Related to
18 Wed Jan 4 13:33:25 2006 Tim Janik <timj@gtk.org>
20 * glib/gslice.c (magazine_cache_pop_magazine): don't reverse chunk
21 order when creating magazines, so we hand out chunks with ascending
24 2006-01-04 Abel Cheung <maddog@linuxhall.org>
26 * configure.in: Added "ml" "zh_HK" to ALL_LINGUAS.
28 2006-01-03 Martyn Russell <martyn@imendio.com>
30 * docs/reference/glib/glib-sections.txt:
32 * glib/gthreadpool.[ch]:
33 - Added new API g_thread_pool_get_idle_time() and
34 g_thread_pool_set_idle_time(). (#324228).
36 * tests/threadpool-test.c:
37 - Updated test case to do thread pool sorting, thread pool with
38 no sorting and a thread pool with idle thread timeouts.
40 2006-01-03 Matthias Clasen <mclasen@redhat.com>
42 * glib/gmain.h: Add new functions here, too.
44 * glib/glib.symbols: Add new functions.
46 2005-12-20 Michael Meeks <michael.meeks@novell.com>
48 * glib/gmain.c (g_main_context_is_owner): new method
49 to determine if the current thread is the owner of the
52 2006-01-02 Matthias Clasen <mclasen@redhat.com>
56 * glib/gstdio.c (g_rmdir): Don't provide g_rmdir() as a macro
57 expanding to rmdir, since rmdir is not declared in a portable
58 system header we can include in gstdio.h. (#325249, Jani Monoses)
60 2006-01-01 Tor Lillqvist <tml@novell.com>
62 * glib/gspawn-win32.c (g_spawn_sync_utf8): Set the GIOChannels for
63 stdout and stderr to unbuffered. Otherwise the giochannel layer
64 will try to read from them regardless whether the
65 g_io_channel_win32_poll() call here has indicated
66 readability or not. (#325310)
68 2005-12-29 Matthias Clasen <mclasen@redhat.com>
70 * glib/gutils.c (glib_check_version): Fix a copy-and-paste error
71 in the docs. (#325273, Declan Naughton)
73 2005-12-27 Manish Singh <yosh@gimp.org>
75 * tests/run-collate-tests.sh: use LC_ALL instead of LC_COLLATE, to
76 make sure we really override things.
78 2005-12-27 Matthias Clasen <mclasen@redhat.com>
80 Fix #316221, Michal Benes, Stanislav Brabec;
82 * configure.in: Fix a strict aliasing problem in
83 g_static_mutex_get_mutex().
84 * glib/gthread.h: ...and in
85 g_static_mutex_get_mutex_impl_shortcut().
87 * glib/gdatasetprivate.h: Add a cast to silence compiler
88 warnings. (#321978, Andrew Paprocki)
90 Partial fix for bug #32937, Bogdan Nicula.
92 * configure.in: Check for malloc.h
93 * glib/gslice.c: Don't include malloc.h unconditionally.
95 2005-12-27 Manish Singh <yosh@gimp.org>
97 * tests/run-collate-tests.sh: export LC_COLLATE so it takes.
98 (#324950, Dan Yefimov)
100 2005-12-26 Matthias Clasen <mclasen@redhat.com>
102 * glib/gslice.c: On Win32, include process.h (#325015, Kazuki
105 2005-12-25 Matthias Clasen <mclasen@redhat.com>
107 * AUTHORS: Update my email
109 * tests/slice-test.c: Fix C99isms. (#324950, Dan Yefimov)
111 2005-12-21 Matthias Clasen <mclasen@redhat.com>
113 * README.in: Add some notes on when 'make check' may fail.
115 2005-12-20 Matthias Clasen <mclasen@redhat.com>
118 * glib/gthreadpool.h:
119 * glib/gthreadpool.c (g_thread_pool_set_sort_function): New function
120 to sort tasks pushed into a threadpool. (#324479, Martyn Russell)
122 * tests/threadpool-test.c: Test this.
124 Tue Dec 20 18:14:14 2005 Tim Janik <timj@imendio.com>
126 * glib/gslice.[hc]: added mem_error() and mem_assert() to test and
127 handle errors without depending on gmessage.c which might not be
128 setup when the error occours.
129 removed G_SLICE_CONFIG_ALWAYS_FREE config option, fixed the code so
130 always freeing can be achieved by adjusting the working set time to
131 0 with G_SLICE_CONFIG_WORKING_SET_MSECS.
132 added G_SLICE_CONFIG_COLOR_INCREMENT to test different color increments
133 (mainly 0 and 1). reduced the minimum block size to 128 bytes, to
134 minimize wastage if small amounts of differently sized structrues are
135 allocated, this does come at a performance cost of roughly 5% though.
136 fixed up block alignment calculation, so it works for varying
137 block sizes. only use strerror() not g_strerror() since the latter
138 depends on working GQuark and GSlice.
139 mem_error(): implemented in terms of fprintf and vfprintf.
141 * tests/slice-color.c: new program to test cache colorization effects.
143 * tests/slice-test.c: trade G_SLICE_CONFIG_ALWAYS_FREE for 0 duration
144 G_SLICE_CONFIG_WORKING_SET_MSECS.
146 2005-12-17 Matthias Clasen <mclasen@redhat.com>
148 * glib/goption.c (parse_short_option): Set an error in all
149 failure cases. (#324332, Tim-Philipp Müller)
151 2005-12-17 Sebastian Wilhelmi <seppi@seppi.de>
153 * glib/gatomic.c: Fix memory barrier position in g_atomic_int_get
154 and g_atomic_pointer_get. Add g_atomic_int_set and
155 g_atomic_pointer_set implementations for the !DEFINE_WITH_MUTEXES &&
156 G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case, as well as defining them
157 as functions (additionally to the macros in the header) for the
158 !G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case.
160 2005-12-16 Matthias Clasen <mclasen@redhat.com>
162 * glib/gmem.c (g_allocator_new): Don't return a pointer to
163 a const struct, since apps expect to be able to modify it.
164 (#324179, J. Ali Harlow)
166 Tue Dec 13 10:13:32 2005 Tim Janik <timj@imendio.com>
168 * glib/gatomic.h: added g_atomic_pointer_set() and g_atomic_int_set()
169 for systems where the initialization of atomic variables requires a
170 write memory barrier.
172 2005-12-09 Matthias Clasen <mclasen@redhat.com>
174 * configure.in: Bump version
176 * === Released 2.9.1 ===
178 2005-12-09 Alexander Larsson <alexl@redhat.com>
182 Use G_GNUC_WARN_UNUSED_RESULT on list functions that return
185 2005-12-08 Matthias Clasen <mclasen@redhat.com>
189 2005-12-07 Martyn Russell <martyn@imendio.com>
191 * glib/gasyncqueue.c:
192 - Call g_queue_insert_sorted() instead of duplicating the code.
193 - Call g_queue_sort() instead of duplicating the code.
194 - Invert sort function results to make sure the same sort function
195 gives the same results across glist, gslist, gqueue and
198 * tests/asyncqueue-test.c:
199 - Updated the sort function to reflect the example in the
200 documentation for gasyncqueue.c.
202 2005-12-07 Martyn Russell <martyn@imendio.com>
204 * docs/reference/glib/glib-sections.txt:
205 * docs/reference/glib/tmpl/linked_lists_double.sgml:
206 * docs/reference/glib/tmpl/linked_lists_single.sgml:
209 - Added g_list_insert_sorted_with_data () and
210 g_slist_insert_sorted_with_data ().
211 - Removed the extra check in g_list_sort() and g_slist_sort() for
212 GCompareDataFunc vs. GCompareFunc.
214 2005-12-07 Tor Lillqvist <tml@novell.com>
216 * glib-gettextize.in: Look up prefix at run-time on Win32,
217 assuming the standard directory structure with glib-gettextize in
220 2005-12-06 Matthias Clasen <mclasen@redhat.com>
222 * glib/gmem.h: Exempt GTK+ from the mem chunk deprecation, since
223 we need GTK+ 2.8 to compile against GLib 2.10.
225 2005-12-05 Manish Singh <yosh@gimp.org>
227 * tests/run-collate-test.sh: set LC_COLLATE, not LANG, to be sure
228 to override any user settings.
230 2005-12-05 Matthias Clasen <mclasen@redhat.com>
232 * configure.in: Revert an accidential commit
234 2005-12-05 Matthias Clasen <mclasen@redhat.com>
236 * glib/gutf8.c: Documentation updates. (#323291, Morten Welinder)
238 * tests/Makefile.am (TESTS_ENVIRONMENT):
239 * tests/gobject/Makefile.am (TESTS_ENVIRONMENT): Set
240 MALLOC_CHECK_ and MALLOC_PERTURB_
242 * tests/run-collate-tests.sh: Run the collation tests explicitly
243 in en_US locale. (#320463)
245 * glib/gmem.h: Really deprecate GMemChunk.
252 * glib/gstrfuncs.c: Add versioned deprecation docs.
254 2005-12-05 Martyn Russell <martyn@imendio.com>
256 * docs/reference/glib/glib-sections.txt:
257 * glib/gasyncqueue.[ch]:
258 - Added support for sorting async queues by with _push_sorted(),
259 _push_sorted_unlocked(), _sort() and _sort_unlocked() (#323047).
262 * tests/asyncqueue-test.c:
263 - Added test case for gasyncqueue.c
265 Mon Dec 5 15:53:20 2005 Tim Janik <timj@imendio.com>
267 * glib/gslice.c: implement chain walking for arbitrary ->next pointer
268 offsets in g_slice_free_chain_with_offset() based on a patch by behdad
269 in bug 323178. moved time consuming logic from g_slice_free() out of
270 the inner loop, so g_slice_free_chain_with_offset() provides a real
271 performance benefit over g_slice_free1() now.
273 * glib/gslice.h: renamed g_slice_free_chain() to
274 g_slice_free_chain_with_offset(). implemented g_slice_free_chain() as
275 a type-safe macro as suggested in bug 323178.
276 simplified the macro implementation of g_slice_free() and implemented
277 it in a type safe manner for all compliers as suggested by Morten
278 Welinder <mortenw@gnome.org>.
283 * glib/glib.symbols: s/g_slice_free_chain/g_slice_free_chain_with_offset/
285 2005-12-05 Matthias Clasen <mclasen@redhat.com>
287 * glib/gasyncqueue.c: Add some docs.
289 * tests/libmoduletestplugin_a.c: Fix compiler warnings.
291 * glib/gatomic.c: In the ia64 implementation, use
292 __sync builtin without _si or _di suffix. (#321229,
293 Stanislav Brabec, patch by Andreas Schwab)
295 2005-12-04 Behdad Esfahbod <behdad@gnome.org>
297 * glib/gslice.h: Remove comma at the end of enum.
299 2005-12-04 Matthias Clasen <mclasen@redhat.com>
301 Handle multiple user names with the same UID better.
302 (#319535, Laszlo Peter)
304 * glib/gutils.c (g_get_any_init_do): When determining user
305 data, first look up $LOGNAME. If the UID doesn't match
306 getuid(), fall back to the current behaviour of looking
307 up the user data based on getuid().
309 2005-12-04 Matthias Clasen <mclasen@redhat.com>
311 * glib/gmacros.h (G_GNUC_WARN_UNUSED_RESULT): Add a macro
312 to make gcc warn if a function result is ignored. (#145466,
313 Arjan van de Ven, Alex Larsson)
315 * glib/gmem.h: Add the new attribute to g_realloc and
318 2005-12-03 Matthias Clasen <mclasen@redhat.com>
321 * glib/gcache.h: Deprecate g_cache_value_foreach. (#322956,
324 * glib/gmappedfile.c: Make mapping of empty files
327 * glib/gfileutils.c: Don't fork a new process just to
328 fix the permissions of the created temp file. (#321318,
329 Alexis S. L. Carvalho)
331 2005-12-02 Matthias Clasen <mclasen@redhat.com>
333 * README.in: Add a note about Solaris threads.
336 * configure.in: Remove support for Solaris threads.
337 (#136971, Sebastian Wilhelmi, patch by Andrew Paprocki)
339 2005-12-02 Matthias Clasen <mclasen@redhat.com>
341 * configure.in: Define G_GUINT64_CONSTANT in analogy to
342 G_GINT64_CONSTANT. (#322568, Andrew Paprocki)
344 2005-12-02 Matthias Clasen <mclasen@redhat.com>
346 * glib/gslice.c: Win32 portability fixes and C99ism removal,
347 pointed out by Kazuki Iwamoto. (#323052)
349 2005-12-02 Matthias Clasen <mclasen@redhat.com>
351 * glib/gatomic.c (g_atomic_int_add): Add a missing volatile for
352 the IA64 implementation.
354 Fri Dec 2 16:18:09 2005 Tim Janik <timj@imendio.com>
356 * glib/gslice.c: some naming and type size fixups.
358 Fri Dec 2 13:08:58 2005 Tim Janik <timj@gtk.org>
360 * glib/gslice.h (g_slice_free): reworked GCC-specific type-safe macro
361 variant into something less verbose, but digestible for gcc-3.4.
363 Fri Dec 2 10:55:07 2005 Tim Janik <timj@imendio.com>
365 * tests/slice-test.c: extended to perform the benchmarking on the old
366 memchunk code if 'O' is selected.
368 * tests/memchunks.c: new file which contains the old GLib mem chunks
369 implementation with prefix old_mem_chunk_.
371 * tests/Makefile.am: added memchunks.c
373 Fri Dec 2 00:16:59 2005 Tim Janik <timj@imendio.com>
375 * glib/gslice.c: improved gettimeofday() timer resolution by maintaining
376 acceptable load distribution of the syscall.
377 (allocator_get_magazine_threshold): reduce minimum/single-thread magazine
378 sizes to approximately page_size/2.5 to avoid excessive startup allocations.
380 2005-12-01 Matthias Clasen <mclasen@redhat.com>
382 * tests/gobject/ifacecheck.c (test_iface_base_init): Don't
383 double-free base interface members.
385 * tests/Makefile.am: Remove duplicate variable.
387 Thu Dec 1 17:32:46 2005 Tim Janik <timj@imendio.com>
389 * glib/gslice.[hc]: new slice allocator implementation.
391 * tests/slice-test.c: added random slice allocation test.
393 * glib/gthread.[hc]: removed newly added private thread mem API.
395 * glib/gthreadinit.h:
398 * glib/gmem.c: divided glib threading initialisation into three phases,
399 initialisation where private keys and messaging are not available (only
400 needed by gmem.c), initialisation without messaging but private keys
401 available (gslice.c, gmessage.c), and full fledged initialisers that
402 server the rest of glib. initialisation functions got renamed to reflect
403 the limitations of their corresponding phases.
405 * glib/gmem.c: removed memchunk code, defer allocations to
408 * glib/gmem.[hc]: removed g_slice_* skeletons.
410 * glib/glib.symbols: added g_slice_* symbols.
412 * configure.in: check for availability of posix_memalign(3), memalign(3)
415 * glib/Makefile.am: added gslice.[hc].
417 2005-12-01 Tor Lillqvist <tml@novell.com>
419 * glib/gstdio.c (g_stat): In the Win32 implementation, strip
420 trailing slash(es) for non-root folders. stat() fails if non-root
421 folders are specified with trailing slashes. It's too much hassle
422 to demand that callers strip such slashes themselves, especially
423 as it is easy to get it wrong and strip the slash of a root
425 (g_rename): On NT-based Windows, use MoveFileEx() with
426 MOVEFILE_REPLACE_EXISTING to better match Unix behaviour.
428 2005-11-28 Matthias Clasen <mclasen@redhat.com>
430 Fix G_STMT_START / G_STMT_END on Solaris. (#321972,
433 * configure.in: Check whether do { } while (0) works.
435 * glib/gmacros.h: Use do { } while (0) for G_STMT_START /
436 G_STMT_END if it works.
438 2005-11-28 Matthias Clasen <mclasen@redhat.com>
440 * glib/gthread.c (g_static_rw_lock_wait, g_static_rw_lock_signal):
441 * glib/gnode.c (g_node_depth_traverse_level):
442 * glib/gmem.c (g_allocator_new):
443 * glib/ghash.c (g_hash_table_unref, g_hash_table_destroy)
444 (g_hash_table_foreach_remove_or_steal): Silence compiler
447 2005-11-27 Matthias Clasen <mclasen@redhat.com>
449 * glib/gunicollate.c (g_utf8_collate_key_for_filename):
450 Don't read beyond len. (#322520, Christian Persch)
452 Wed Nov 23 17:34:01 2005 Tim Janik <timj@imendio.com>
454 * glib/gdataset.c: access datalist flags via atomic pointer access
455 functions, instead of acquiring the dataset lock. this is faster and
456 also matches the atomic pointer readouts in gdatalistprivate.h and
457 direct pointer modifications required by gobject.c.
459 Wed Nov 23 13:35:31 2005 Tim Janik <timj@gtk.org>
461 * glib/gdataset.c: streamlined initialization code somewhat,
462 removed GData node cache.
464 2005-11-22 Matthias Clasen <mclasen@redhat.com>
466 * glib/ghash.c (g_hash_table_ref):
467 (g_hash_table_unref): Mark these as new API.
469 Avoid double locking in g_intern_string (#322133,
472 * glib/gdataset.c (g_quark_from_string_internal): New
473 internal function which factors out common parts of
474 g_quark_from[_static]_string.
475 (g_quark_from_string, g_quark_from_static_string):
476 Use g_quark_from_string_internal.
477 (g_intern_string, g_intern_static_string): Use
478 g_quark_from_string_internal, and only take the
480 (g_quark_new): Don't store the strings shifted by -1
481 in the g_quarks array.
482 (g_quark_to_string): Adapt to the previous change.
484 Tue Nov 22 14:04:26 2005 Tim Janik <timj@imendio.com>
488 g_hash_table_new_full(): create hash tables with a ref count of 1.
489 g_hash_table_ref(): atomically ref_count+=1
490 g_hash_table_unref(): atomically ref_count-=1, destroys hash table
491 when refcount reaches 0.
492 g_hash_table_destroy(): just destroy keys and values, unref by 1.
493 g_hash_table_insert():
494 g_hash_table_replace(): assert ref_count>0.
497 * glib/gatomic.c: added 'volatile' qualifier to all atomic pointer and
500 2005-11-20 Behdad Esfahbod <behdad@gnome.org>
502 * glib/guniprop.c (g_unichar_get_mirror_char): Remove unused
503 variables. (#321984, Andrew Paprocki)
505 2005-11-18 Matthias Clasen <mclasen@redhat.com>
507 * configure.in: Bump version
509 2005-11-17 Matthias Clasen <mclasen@redhat.com>
511 * === Released 2.9.0 ===
513 * glib/glib.symbols: Updates
515 * glib/gutf8.c (g_utf8_offset_to_pointer): Add some performance
516 hints to the docs. (Paolo Borelli)
520 * glib/gthread.c (g_thread_foreach): Mark as new api.
522 * README.in: Updates.
526 * glib/gdate.c (g_date_set_time_t):
527 (g_date_set_time_val): Add functions to set a date from
528 a time_t and from a GTimeVal.
529 (g_date_set_time): Deprecate. (#314067, Roger Leigh)
532 * tests/utf8-pointer.c: Unit tests for g_utf8_pointer_to_offset
533 and g_utf8_offset_to_pointer.
535 * glib/gutf8.c (g_utf8_pointer_to_offset)
536 (g_utf8_offset_to_pointer): Handle negative offsets, and use
537 "stutter stepping" for going backwards. (#320638, Larry
542 * glib/gthread.c: const correctness fixes, found
543 by Arjan van de Ven and gcc.
545 2005-11-16 Behdad Esfahbod <behdad@gnome.org>
547 * docs/reference/glib/tmpl/unicode.sgml: Correct typo on mentioning
548 @G_UNICODE_BREAK_UNKNOWN that should be #G_UNICODE_BREAK_UNKNOWN.
550 2005-11-09 Behdad Esfahbod <behdad@gnome.org>
552 * glib/guniprop.c: Use bit hacks instead when checking a general
553 category value against multiple values.
555 * glib/gutf8.c: Change ISO10646 to Unicode in docs.
557 2005-11-10 Simos Xenitellis <simos@gnome.org>
559 * configure.in: Added tt (Tatar) to ALL_LINGUAS.
561 2005-11-05 Tor Lillqvist <tml@novell.com>
563 * glib/giowin32.c (g_io_win32_set_flags): Don't call g_warning().
564 (#320688) Instead set error and return failure.
566 2005-11-07 Matthias Clasen <mclasen@redhat.com>
568 * glib/gmain.c (g_main_dispatch): Don't call cb_funcs->unref
569 while holding the context lock. (#320886, Andy Wingo)
571 2005-11-07 Billy Biggs <vektor@dumbterm.net>
573 * tests/refcount/Makefile.am: Add a missing reference to libglib
574 in the LDADD for this test.
576 2005-11-07 Matthias Clasen <mclasen@redhat.com>
578 * glib/gmem.c (g_allocator_new): We need to set n_preallocs to a
579 nonzero value, otherwise GTK+ 2.8 breaks when compiled against
580 GLib 2.9. (#320755, Luca Ferretti)
582 2005-11-04 Matthias Clasen <mclasen@redhat.com>
584 * glib/Makefile.am: Apply a patch from Behdad Esfahbod to
585 use a faster lookup table for g_unichar_get_mirror_char().
586 * glib/gmirroringtable.h: The new table.
587 * glib/gunichartables.h: Remove bidi_mirroring_table.
588 * glib/gen-unicode-tables.pl: Don't generate the mirroring
590 * glib/glib-mirroring-tab/*: Sources for the program
591 which generated gmirroringtable.h.
593 * glib/glist.c: Avoid some code duplication.
595 * glib/gscanner.h: Include gdataset.h. (#320322)
597 * glib/gdebug.h: add new GDebugFlag for fatal_criticals
598 * glib/gmessages.c: (_g_debug_init): handle G_DEBUG=fatal_criticals,
599 to help find critical warnings in applications. (#320017,
602 2005-11-02 Tor Lillqvist <tml@novell.com>
604 * glib/glib.symbols: Remove large amount of trailing whitespace
605 from one line. Remove #ifdef G_UNDEFINED from around
606 g_hash_table_new and g_hash_table_new_full symbols, what was the
607 purpose of that? Add the g_slice functions.
609 Tue Nov 1 16:24:20 2005 Tim Janik <timj@imendio.com>
611 * glib/gmem.[hc]: prepared deprecation of GMemChunk and GAllocator.
612 added g_slice_*() API to allocate and cache small bits of memory.
613 an actuall allocator implementation for g_slice_*() is still pending.
615 * glib/gthread.[hc]: changes from a patch by Matthias Clasen.
616 changed GRealThread list to use in-structure *next; fields instead
617 of GSList, in order for thread iteration to not depenend on g_slice_*()
619 _g_thread_mem_private_get():
620 _g_thread_mem_private_set(): added accessors for private memory,
621 needed because the ordinary GPrivate implementation relies on GArray
622 and GSList and therefore indirectly on working g_slice_*() allocations.
625 g_thread_foreach(): new public API function to loop over all existing threads.
639 removed GAllocator and free list usages and accompanying locks.
640 use g_slice_*() API to allocate and cache small bits of memory.
642 * glib/ghook.h: removed GMemChunk field from public API.
645 * glib/glist.h: deprecate allocator API, provide _free1() for consistency.
647 * glib/gnode.h: deprecate allocator API.
649 * glib/gmain.c: reordered GPollRec fields so g_slice_free_chain() can
650 be used for poll rec lists.
652 * glib/grel.c: removed mem chunk usage, and allocated tuples via g_slice_*().
653 g_relation_destroy(): free all tuples from the all_tuples hash table,
654 this effectively maintains the life time track keeping of tuples.
655 g_relation_delete_tuple(): free tuples which are removed from the
656 all_tuples hash table. this fixes a temporary leak that was present
657 in the memchunk code until the destruction of the relation.
659 2005-10-29 Matthias Clasen <mclasen@redhat.com>
661 * tests/convert-test.c: Add some tests for conversions between
662 UTF-8, UCS-4 and UTF-16.
664 * glib/gutf8.c (g_utf8_to_ucs4, g_utf8_to_utf16): Fix handling
665 of len == -1, noticed by Morten Welinder.
667 2005-10-27 Erdal Ronahi <erdal.ronahi@gmail.com>
669 * configure.in: Added ku (Kurdish) to ALL_LINGUAS
671 2005-10-26 Matthias Clasen <mclasen@redhat.com>
673 * glib/gutf8.c (g_ucs4_to_utf8): Don't set items_read twice
674 in the error case, and add some documentation. (#319806, Morten
677 2005-10-19 Manish Singh <yosh@gimp.org>
679 * configure.in: Use AC_CHECK_FUNCS for _NSGetEnviron, to get the
680 config.h symbol automatically. Fixes bug #313731.
682 2005-10-19 Tor Lillqvist <tml@novell.com>
684 * glib/gwin32.c (g_win32_get_package_installation_directory):
685 Return a g_strdup()ed copy of the value stored in the hash table,
686 so that it can be g_free()d without leaving a dangling pointer in
687 the hash table. (#319232)
689 2005-10-06 Matthias Clasen <mclasen@redhat.com>
691 * glib/gunicollate.c (g_utf8_collate_key_for_filename): Handle
692 all-zero sequences correctly. (#317930, Sebastien Bacher)
694 2005-10-05 Matthias Clasen <mclasen@redhat.com>
696 * glib/gmarkup.c (g_markup_escape_text): Doc addition
698 * Makefile.am: Add Collation tests.
700 * tests/collate/*: Inputs and expected outputs for collation tests.
702 * tests/run-collate-tests.sh: Script to run collation tests.
704 * tests/unicode-collate.c (main): Rework slightly to make
705 it usable in unit tests. Also test g_utf8_collate_key_for_filename().
707 2005-10-01 Behdad Esfahbod <behdad@gnome.org>
709 * docs/reference/glib/tmpl/unicode.sgml:
710 * glib/gen-unicode-tables.pl:
712 * glib/gunichartables.h:
714 * tests/casefold.txt:
715 * tests/casemap.txt: Updated to Unicode 4.1. There are five new
716 GUnicodeBreakType types. That may break some applications, like
719 2005-09-26 Matthias Clasen <mclasen@redhat.com>
721 * glib/gstrfuncs.c (g_ascii_strtoull): Add details to the
722 docs. (#314393, Matthew F. Barnes)
725 * glib/gprintf.h: Remove g_snprintf() and g_vsnprintf(), since
726 they are already declared in glib.h. This doesn't break documented
727 use of gprintf.h, but should probably be pointed out in the
728 release notes for 2.10. (#314232, Behdad Esfahbod)
730 Tue Sep 20 13:16:04 2005 Tim Janik <timj@imendio.com>
732 * glib/gpattern.c (g_pattern_ph_match): applied significant recursion
733 complexity optimization, based on a patch from Matthias Clasen.
735 * tests/patterntest.c: more tests, mostly from matthias.
737 2005-09-20 Matthias Clasen <mclasen@redhat.com>
739 * glib/gqueue.c (g_queue_insert_sorted): Correct the docs.
740 (#316703, Mark Drago)
742 Mon Sep 19 17:23:23 2005 Tim Janik <timj@imendio.com>
744 * glib/gpattern.c: applied a patch from matthias which checks on the
745 upper bound of GPatternSpec length to optimize matches.
748 * tests/patterntest.c: added more match cases.
750 2005-09-16 Tor Lillqvist <tml@novell.com>
752 * glib/gstrfuncs.c (g_ascii_strcasecmp, g_ascii_strncasecmp): Add
753 warning to doc comment that these functions should not be used on
754 encodings like CP932.
756 2005-09-14 Matthias Clasen <mclasen@redhat.com>
758 * tests/keyfile-test.c: Add a test for grup names of length 1.
760 * glib/gkeyfile.c (g_key_file_line_is_group): Accept group names
761 of length 1. (#316309)
763 2005-09-12 Matthias Clasen <mclasen@redhat.com>
765 * glib/gmarkup.c (g_markup_escape_text): Clarify docs.
767 2005-09-11 Sebastian Wilhelmi <seppi@seppi.de>
769 * tests/refcount/Makefile.am (INCLUDES): Link the the refcount
770 tests to the system thread library $(G_THREAD_LIBS). Fixes #313744
773 2005-09-11 Kjartan Maraas <kmaraas@gnome.org>
775 * glib/gmain.c: (g_child_watch_prepare), (g_child_watch_check),
776 (child_watch_helper_thread): Remove some dead code. Closes
779 2005-09-07 Tor Lillqvist <tml@novell.com>
781 * glib/Makefile.am: Create also a console version of the
782 gspawn-win32-helper program, gspawn-win32-helper-console.exe.
783 It's otherwise identical to gspawn-win32-helper.exe, except marked
784 as a console application (linked without the -mwindows option).
786 * glib/gspawn-win32.c (do_spawn_directly, do_spawn_with_pipes):
787 Drop the dont_wait parameter. Its truth value correlated 100% with
788 the NULLness of the exit_status parameter anyway, so it's enough
789 to check whether exit_status is NULL. Invert the sense of the
790 dont_return_handle parameter and rename it to do_return_handle, to
791 make the code easier to read by avoiding double negations.
793 (g_spawn_sync_utf8, g_spawn_async_with_pipes_utf8): Modify calls
794 to do_spawn_with_pipes() accordingly.
796 (do_spawn_with_pipes): If we have a console, use the console
797 version of the helper program, otherwise use the GUI one. This
798 avoids extra console windows opening up in some situations. (In
799 case a console application uses the GUI gspawn-win32-helper.exe to
800 spawn another console application we would get a separate console
801 for the spawned console application).
803 * glib-zip.in: Distribute also gspawn-win32-helper-console.exe.
805 2005-09-05 Matthias Clasen <mclasen@redhat.com>
807 * glib/gmappedfile.c (g_mapped_file_new): Report an error
808 if the file is too large. (#315275, Kjartan Maraas)
810 * glib/gkeyfile.c (g_key_file_load_from_fd): The return value
811 of read() is signed. (#315273, Kjartan Maraas)
813 2005-08-31 Tor Lillqvist <tml@novell.com>
815 * glib/gutils.h: Wrapping atexit() is a bad idea on Windows, where
816 the EXE and each DLL have their own atexit function chains.
818 #define g_atexit as atexit instead. This means it has a
819 better chance of doing what the caller wants. For instance,
820 gtkhtml calls g_atexit() registering a function in gtkhtml
821 itself. This caused a crash when g_atexit() was implemented as a
822 function in the GLib DLL. The gtkhtml DLL was already unloaded by
823 the time the GLib DLL got unloaded.
825 * glib/gutils.c: #undef the #define mentioned above, to also get a
826 real g_atexit() into the DLL for backward compatibility. Document
827 the Windows behaviour of g_atexit(), and document the varying ways
828 atexit() can behave in the context of dynamically loaded modules
831 2005-08-31 Matthias Clasen <mclasen@redhat.com>
835 * glib/gdataset.c: Add string interning functions.
837 2005-08-28 Matthias Clasen <mclasen@redhat.com>
839 * glib/giochannel.c: Unify some near-duplicate strings. (#314654,
842 2005-08-26 Matthias Clasen <mclasen@redhat.com>
844 * configure.in: Bump version to 2.9.0
846 2005-08-25 Tor Lillqvist <tml@novell.com>
848 Make also the g_spawn*() functions take parameters in the GLib
849 file name encoding, i.e. UTF-8, on Windows. Has no impact on Unix
850 API or ABI. Like the other GLib API that was earlier changed to
851 use UTF-8 on Windows, the names of the functions that take UTF-8
852 have _utf8 suffixes added by using preprocessor macros in the
853 header file. The old names are kept for functions with the old
854 behaviour, taking parameters in the system codepage, for DLL ABI
857 * glib/gspawn.h: On Win32 add the suffix _utf8 to the names of the
858 g_spawn*() functions.
860 * glib/gspawn-win32.c: Use wide-char API on NT-based
861 Windows. Convert parameters from UTF-8 to wide chars (NT) or
862 system codepage (Win9x) and call the C library _wspawn*() or
863 spawn*() functions respectvely. Add DLL ABI stability versions
864 that take parameters in the system codepage.
866 * glib/gspawn-win32-helper.c: On NT-based Windows use the
867 wide-char versions of argv and envp, and use wide-char API to
868 change directory and spawn the program to run. Remove the verbose
869 debugging output, it was too complex to modify for the wide-char
870 features. (Just add temporary debugging printouts if needed, no
871 need to have them permanently in the source.)
873 * glib/gspawn.c: Corresponding documentation updates.
875 * glib/glib.symbols: Corresponding changes: Mark the ABI stability
876 symbols as PRIVATE, add the new _utf8-suffixed ones.
878 2005-08-24 Stepan Kasal <kasal@ucw.cz>
880 * glib/gtypes.h (G_MININT64): Cast the constant to gint64; it is
881 guint64 otherwise and that can produce warnings about comparison
882 between signed and unsigned.
884 2005-08-23 Matthias Clasen <mclasen@redhat.com>
886 * glib/gutils.c: Fix the crt_externs.h include.
888 2005-08-23 Stepan Kasal <kasal@ucw.cz>
890 * NEWS: Fix spelling of my first name.
892 2005-08-23 Matthias Clasen <mclasen@redhat.com>
896 * === Released 2.8.1 ===
900 2005-08-20 Hans Breuer <hans@breuer.org>
902 * glib/makefile.msc.in : link with ws2_32.lib
904 2005-08-18 Tor Lillqvist <tml@novell.com>
906 * configure.in: Check for <sys/wait.h>
908 * glib/gbacktrace.c: Include <sys/wait.h> on if HAVE_SYS_WAIT_H.
910 2005-08-18 Ross Burton <ross@burtonini.com>
913 Optimise single-character insertions.
918 * tests/string-test.c:
919 Add tests for new optimisation, and fix a leak.
921 2005-08-17 Matthias Clasen <mclasen@redhat.com>
923 * configure.in: Check for crt_externs.h and _NSGetEnviron.
925 * glib/gutils.c: On Darwin, include crt-externs.h and
926 define environ using _NSGetEnviron(). (#313731)
928 2005-08-16 Stepan Kasal <kasal@ucw.cz>
930 * glib/gutils.c (g_get_any_init): Move the body of the big if...
931 (g_get_any_init_do): ... to this new function.
932 (g_get_any_init): Declare as inline.
933 (g_get_any_init_locked): New inline function, does the locking.
934 Make use of these two throughout the code.
936 2005-08-15 Matthias Clasen <mclasen@redhat.com>
938 * glib/gbacktrace.c (g_on_error_stack_trace): Wait for
939 the child process and then simply return. This makes
940 The "S" option work as documented in g_on_error_query().
941 (#313125, Matthew F. Barnes)
943 * glib/gunicode.h: Update the link to Unicode category
944 values. (#313369, Behnam Esfahbod)
946 * glib/gqueue.c (g_queue_find_custom): Clarify docs
947 a little. (#311727, Tristan van Berkom)
949 * glib/abicheck.sh, gobject/abicheck.sh: Make the
950 check work on ia64 too, where some symbols ended up
951 in yet another section.