1 2006-01-11 Matthias Clasen <mclasen@redhat.com>
3 * tests/asyncqueue-test.c: Set error to NULL.
5 2006-01-08 Matthias Clasen <mclasen@redhat.com>
7 * glib/gtree.c: Break some long lines.
9 * tests/tree-test.c: Add more tests.
11 * glib/gslice.c: Include config.h first, pointed out
14 2005-12-20 Sven Herzberg <herzi@gnome-de.org>
16 * docs/reference/gobject/tmpl/gtype.sgml: explain that
17 G_TYPE_INSTANCE_GET_CLASS() does behave different during
20 2006-01-06 Matthias Clasen <mclasen@redhat.com>
22 * glib/gslice.c: Only define _XOPEN_SOURCE if we know
23 that we have posix_memalign(). (#323937, Bogdan Nicula)
25 2006-01-05 Matthias Clasen <mclasen@redhat.com>
27 * configure.in: Bump version
29 * === Released 2.9.2 ===
31 2006-01-05 Hans Breuer <hans@breuer.org>
33 * glib/makefile.msc.in : added gslice.obj
35 2006-01-05 Matthias Clasen <mclasen@redhat.com>
39 * glib/gmain.c: Clarify the documentation of source ids
40 a bit. (#325874, Dan Williams)
42 * configure.in: Fix another strict aliasing problem.
44 * glib/gthreadpool.c: Avoid use of varargs macro. (#325864,
47 2006-01-04 Manish Singh <yosh@gimp.org>
50 * glib/gstdio.h: don't macroized g_access, g_chdir, and g_unlink
51 either, since they have the same issue as g_rmdir. (Related to
54 Wed Jan 4 13:33:25 2006 Tim Janik <timj@gtk.org>
56 * glib/gslice.c (magazine_cache_pop_magazine): don't reverse chunk
57 order when creating magazines, so we hand out chunks with ascending
60 2006-01-04 Abel Cheung <maddog@linuxhall.org>
62 * configure.in: Added "ml" "zh_HK" to ALL_LINGUAS.
64 2006-01-03 Martyn Russell <martyn@imendio.com>
66 * docs/reference/glib/glib-sections.txt:
68 * glib/gthreadpool.[ch]:
69 - Added new API g_thread_pool_get_idle_time() and
70 g_thread_pool_set_idle_time(). (#324228).
72 * tests/threadpool-test.c:
73 - Updated test case to do thread pool sorting, thread pool with
74 no sorting and a thread pool with idle thread timeouts.
76 2006-01-03 Matthias Clasen <mclasen@redhat.com>
78 * glib/gmain.h: Add new functions here, too.
80 * glib/glib.symbols: Add new functions.
82 2005-12-20 Michael Meeks <michael.meeks@novell.com>
84 * glib/gmain.c (g_main_context_is_owner): new method
85 to determine if the current thread is the owner of the
88 2006-01-02 Matthias Clasen <mclasen@redhat.com>
92 * glib/gstdio.c (g_rmdir): Don't provide g_rmdir() as a macro
93 expanding to rmdir, since rmdir is not declared in a portable
94 system header we can include in gstdio.h. (#325249, Jani Monoses)
96 2006-01-01 Tor Lillqvist <tml@novell.com>
98 * glib/gspawn-win32.c (g_spawn_sync_utf8): Set the GIOChannels for
99 stdout and stderr to unbuffered. Otherwise the giochannel layer
100 will try to read from them regardless whether the
101 g_io_channel_win32_poll() call here has indicated
102 readability or not. (#325310)
104 2005-12-29 Matthias Clasen <mclasen@redhat.com>
106 * glib/gutils.c (glib_check_version): Fix a copy-and-paste error
107 in the docs. (#325273, Declan Naughton)
109 2005-12-27 Manish Singh <yosh@gimp.org>
111 * tests/run-collate-tests.sh: use LC_ALL instead of LC_COLLATE, to
112 make sure we really override things.
114 2005-12-27 Matthias Clasen <mclasen@redhat.com>
116 Fix #316221, Michal Benes, Stanislav Brabec;
118 * configure.in: Fix a strict aliasing problem in
119 g_static_mutex_get_mutex().
120 * glib/gthread.h: ...and in
121 g_static_mutex_get_mutex_impl_shortcut().
123 * glib/gdatasetprivate.h: Add a cast to silence compiler
124 warnings. (#321978, Andrew Paprocki)
126 Partial fix for bug #323937, Bogdan Nicula.
128 * configure.in: Check for malloc.h
129 * glib/gslice.c: Don't include malloc.h unconditionally.
131 2005-12-27 Manish Singh <yosh@gimp.org>
133 * tests/run-collate-tests.sh: export LC_COLLATE so it takes.
134 (#324950, Dan Yefimov)
136 2005-12-26 Matthias Clasen <mclasen@redhat.com>
138 * glib/gslice.c: On Win32, include process.h (#325015, Kazuki
141 2005-12-25 Matthias Clasen <mclasen@redhat.com>
143 * AUTHORS: Update my email
145 * tests/slice-test.c: Fix C99isms. (#324950, Dan Yefimov)
147 2005-12-21 Matthias Clasen <mclasen@redhat.com>
149 * README.in: Add some notes on when 'make check' may fail.
151 2005-12-20 Matthias Clasen <mclasen@redhat.com>
154 * glib/gthreadpool.h:
155 * glib/gthreadpool.c (g_thread_pool_set_sort_function): New function
156 to sort tasks pushed into a threadpool. (#324479, Martyn Russell)
158 * tests/threadpool-test.c: Test this.
160 Tue Dec 20 18:14:14 2005 Tim Janik <timj@imendio.com>
162 * glib/gslice.[hc]: added mem_error() and mem_assert() to test and
163 handle errors without depending on gmessage.c which might not be
164 setup when the error occours.
165 removed G_SLICE_CONFIG_ALWAYS_FREE config option, fixed the code so
166 always freeing can be achieved by adjusting the working set time to
167 0 with G_SLICE_CONFIG_WORKING_SET_MSECS.
168 added G_SLICE_CONFIG_COLOR_INCREMENT to test different color increments
169 (mainly 0 and 1). reduced the minimum block size to 128 bytes, to
170 minimize wastage if small amounts of differently sized structrues are
171 allocated, this does come at a performance cost of roughly 5% though.
172 fixed up block alignment calculation, so it works for varying
173 block sizes. only use strerror() not g_strerror() since the latter
174 depends on working GQuark and GSlice.
175 mem_error(): implemented in terms of fprintf and vfprintf.
177 * tests/slice-color.c: new program to test cache colorization effects.
179 * tests/slice-test.c: trade G_SLICE_CONFIG_ALWAYS_FREE for 0 duration
180 G_SLICE_CONFIG_WORKING_SET_MSECS.
182 2005-12-17 Matthias Clasen <mclasen@redhat.com>
184 * glib/goption.c (parse_short_option): Set an error in all
185 failure cases. (#324332, Tim-Philipp Müller)
187 2005-12-17 Sebastian Wilhelmi <seppi@seppi.de>
189 * glib/gatomic.c: Fix memory barrier position in g_atomic_int_get
190 and g_atomic_pointer_get. Add g_atomic_int_set and
191 g_atomic_pointer_set implementations for the !DEFINE_WITH_MUTEXES &&
192 G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case, as well as defining them
193 as functions (additionally to the macros in the header) for the
194 !G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case.
196 2005-12-16 Matthias Clasen <mclasen@redhat.com>
198 * glib/gmem.c (g_allocator_new): Don't return a pointer to
199 a const struct, since apps expect to be able to modify it.
200 (#324179, J. Ali Harlow)
202 Tue Dec 13 10:13:32 2005 Tim Janik <timj@imendio.com>
204 * glib/gatomic.h: added g_atomic_pointer_set() and g_atomic_int_set()
205 for systems where the initialization of atomic variables requires a
206 write memory barrier.
208 2005-12-09 Matthias Clasen <mclasen@redhat.com>
210 * configure.in: Bump version
212 * === Released 2.9.1 ===
214 2005-12-09 Alexander Larsson <alexl@redhat.com>
218 Use G_GNUC_WARN_UNUSED_RESULT on list functions that return
221 2005-12-08 Matthias Clasen <mclasen@redhat.com>
225 2005-12-07 Martyn Russell <martyn@imendio.com>
227 * glib/gasyncqueue.c:
228 - Call g_queue_insert_sorted() instead of duplicating the code.
229 - Call g_queue_sort() instead of duplicating the code.
230 - Invert sort function results to make sure the same sort function
231 gives the same results across glist, gslist, gqueue and
234 * tests/asyncqueue-test.c:
235 - Updated the sort function to reflect the example in the
236 documentation for gasyncqueue.c.
238 2005-12-07 Martyn Russell <martyn@imendio.com>
240 * docs/reference/glib/glib-sections.txt:
241 * docs/reference/glib/tmpl/linked_lists_double.sgml:
242 * docs/reference/glib/tmpl/linked_lists_single.sgml:
245 - Added g_list_insert_sorted_with_data () and
246 g_slist_insert_sorted_with_data ().
247 - Removed the extra check in g_list_sort() and g_slist_sort() for
248 GCompareDataFunc vs. GCompareFunc.
250 2005-12-07 Tor Lillqvist <tml@novell.com>
252 * glib-gettextize.in: Look up prefix at run-time on Win32,
253 assuming the standard directory structure with glib-gettextize in
256 2005-12-06 Matthias Clasen <mclasen@redhat.com>
258 * glib/gmem.h: Exempt GTK+ from the mem chunk deprecation, since
259 we need GTK+ 2.8 to compile against GLib 2.10.
261 2005-12-05 Manish Singh <yosh@gimp.org>
263 * tests/run-collate-test.sh: set LC_COLLATE, not LANG, to be sure
264 to override any user settings.
266 2005-12-05 Matthias Clasen <mclasen@redhat.com>
268 * configure.in: Revert an accidential commit
270 2005-12-05 Matthias Clasen <mclasen@redhat.com>
272 * glib/gutf8.c: Documentation updates. (#323291, Morten Welinder)
274 * tests/Makefile.am (TESTS_ENVIRONMENT):
275 * tests/gobject/Makefile.am (TESTS_ENVIRONMENT): Set
276 MALLOC_CHECK_ and MALLOC_PERTURB_
278 * tests/run-collate-tests.sh: Run the collation tests explicitly
279 in en_US locale. (#320463)
281 * glib/gmem.h: Really deprecate GMemChunk.
288 * glib/gstrfuncs.c: Add versioned deprecation docs.
290 2005-12-05 Martyn Russell <martyn@imendio.com>
292 * docs/reference/glib/glib-sections.txt:
293 * glib/gasyncqueue.[ch]:
294 - Added support for sorting async queues by with _push_sorted(),
295 _push_sorted_unlocked(), _sort() and _sort_unlocked() (#323047).
298 * tests/asyncqueue-test.c:
299 - Added test case for gasyncqueue.c
301 Mon Dec 5 15:53:20 2005 Tim Janik <timj@imendio.com>
303 * glib/gslice.c: implement chain walking for arbitrary ->next pointer
304 offsets in g_slice_free_chain_with_offset() based on a patch by behdad
305 in bug 323178. moved time consuming logic from g_slice_free() out of
306 the inner loop, so g_slice_free_chain_with_offset() provides a real
307 performance benefit over g_slice_free1() now.
309 * glib/gslice.h: renamed g_slice_free_chain() to
310 g_slice_free_chain_with_offset(). implemented g_slice_free_chain() as
311 a type-safe macro as suggested in bug 323178.
312 simplified the macro implementation of g_slice_free() and implemented
313 it in a type safe manner for all compliers as suggested by Morten
314 Welinder <mortenw@gnome.org>.
319 * glib/glib.symbols: s/g_slice_free_chain/g_slice_free_chain_with_offset/
321 2005-12-05 Matthias Clasen <mclasen@redhat.com>
323 * glib/gasyncqueue.c: Add some docs.
325 * tests/libmoduletestplugin_a.c: Fix compiler warnings.
327 * glib/gatomic.c: In the ia64 implementation, use
328 __sync builtin without _si or _di suffix. (#321229,
329 Stanislav Brabec, patch by Andreas Schwab)
331 2005-12-04 Behdad Esfahbod <behdad@gnome.org>
333 * glib/gslice.h: Remove comma at the end of enum.
335 2005-12-04 Matthias Clasen <mclasen@redhat.com>
337 Handle multiple user names with the same UID better.
338 (#319535, Laszlo Peter)
340 * glib/gutils.c (g_get_any_init_do): When determining user
341 data, first look up $LOGNAME. If the UID doesn't match
342 getuid(), fall back to the current behaviour of looking
343 up the user data based on getuid().
345 2005-12-04 Matthias Clasen <mclasen@redhat.com>
347 * glib/gmacros.h (G_GNUC_WARN_UNUSED_RESULT): Add a macro
348 to make gcc warn if a function result is ignored. (#145466,
349 Arjan van de Ven, Alex Larsson)
351 * glib/gmem.h: Add the new attribute to g_realloc and
354 2005-12-03 Matthias Clasen <mclasen@redhat.com>
357 * glib/gcache.h: Deprecate g_cache_value_foreach. (#322956,
360 * glib/gmappedfile.c: Make mapping of empty files
363 * glib/gfileutils.c: Don't fork a new process just to
364 fix the permissions of the created temp file. (#321318,
365 Alexis S. L. Carvalho)
367 2005-12-02 Matthias Clasen <mclasen@redhat.com>
369 * README.in: Add a note about Solaris threads.
372 * configure.in: Remove support for Solaris threads.
373 (#136971, Sebastian Wilhelmi, patch by Andrew Paprocki)
375 2005-12-02 Matthias Clasen <mclasen@redhat.com>
377 * configure.in: Define G_GUINT64_CONSTANT in analogy to
378 G_GINT64_CONSTANT. (#322568, Andrew Paprocki)
380 2005-12-02 Matthias Clasen <mclasen@redhat.com>
382 * glib/gslice.c: Win32 portability fixes and C99ism removal,
383 pointed out by Kazuki Iwamoto. (#323052)
385 2005-12-02 Matthias Clasen <mclasen@redhat.com>
387 * glib/gatomic.c (g_atomic_int_add): Add a missing volatile for
388 the IA64 implementation.
390 Fri Dec 2 16:18:09 2005 Tim Janik <timj@imendio.com>
392 * glib/gslice.c: some naming and type size fixups.
394 Fri Dec 2 13:08:58 2005 Tim Janik <timj@gtk.org>
396 * glib/gslice.h (g_slice_free): reworked GCC-specific type-safe macro
397 variant into something less verbose, but digestible for gcc-3.4.
399 Fri Dec 2 10:55:07 2005 Tim Janik <timj@imendio.com>
401 * tests/slice-test.c: extended to perform the benchmarking on the old
402 memchunk code if 'O' is selected.
404 * tests/memchunks.c: new file which contains the old GLib mem chunks
405 implementation with prefix old_mem_chunk_.
407 * tests/Makefile.am: added memchunks.c
409 Fri Dec 2 00:16:59 2005 Tim Janik <timj@imendio.com>
411 * glib/gslice.c: improved gettimeofday() timer resolution by maintaining
412 acceptable load distribution of the syscall.
413 (allocator_get_magazine_threshold): reduce minimum/single-thread magazine
414 sizes to approximately page_size/2.5 to avoid excessive startup allocations.
416 2005-12-01 Matthias Clasen <mclasen@redhat.com>
418 * tests/gobject/ifacecheck.c (test_iface_base_init): Don't
419 double-free base interface members.
421 * tests/Makefile.am: Remove duplicate variable.
423 Thu Dec 1 17:32:46 2005 Tim Janik <timj@imendio.com>
425 * glib/gslice.[hc]: new slice allocator implementation.
427 * tests/slice-test.c: added random slice allocation test.
429 * glib/gthread.[hc]: removed newly added private thread mem API.
431 * glib/gthreadinit.h:
434 * glib/gmem.c: divided glib threading initialisation into three phases,
435 initialisation where private keys and messaging are not available (only
436 needed by gmem.c), initialisation without messaging but private keys
437 available (gslice.c, gmessage.c), and full fledged initialisers that
438 server the rest of glib. initialisation functions got renamed to reflect
439 the limitations of their corresponding phases.
441 * glib/gmem.c: removed memchunk code, defer allocations to
444 * glib/gmem.[hc]: removed g_slice_* skeletons.
446 * glib/glib.symbols: added g_slice_* symbols.
448 * configure.in: check for availability of posix_memalign(3), memalign(3)
451 * glib/Makefile.am: added gslice.[hc].
453 2005-12-01 Tor Lillqvist <tml@novell.com>
455 * glib/gstdio.c (g_stat): In the Win32 implementation, strip
456 trailing slash(es) for non-root folders. stat() fails if non-root
457 folders are specified with trailing slashes. It's too much hassle
458 to demand that callers strip such slashes themselves, especially
459 as it is easy to get it wrong and strip the slash of a root
461 (g_rename): On NT-based Windows, use MoveFileEx() with
462 MOVEFILE_REPLACE_EXISTING to better match Unix behaviour.
464 2005-11-28 Matthias Clasen <mclasen@redhat.com>
466 Fix G_STMT_START / G_STMT_END on Solaris. (#321972,
469 * configure.in: Check whether do { } while (0) works.
471 * glib/gmacros.h: Use do { } while (0) for G_STMT_START /
472 G_STMT_END if it works.
474 2005-11-28 Matthias Clasen <mclasen@redhat.com>
476 * glib/gthread.c (g_static_rw_lock_wait, g_static_rw_lock_signal):
477 * glib/gnode.c (g_node_depth_traverse_level):
478 * glib/gmem.c (g_allocator_new):
479 * glib/ghash.c (g_hash_table_unref, g_hash_table_destroy)
480 (g_hash_table_foreach_remove_or_steal): Silence compiler
483 2005-11-27 Matthias Clasen <mclasen@redhat.com>
485 * glib/gunicollate.c (g_utf8_collate_key_for_filename):
486 Don't read beyond len. (#322520, Christian Persch)
488 Wed Nov 23 17:34:01 2005 Tim Janik <timj@imendio.com>
490 * glib/gdataset.c: access datalist flags via atomic pointer access
491 functions, instead of acquiring the dataset lock. this is faster and
492 also matches the atomic pointer readouts in gdatalistprivate.h and
493 direct pointer modifications required by gobject.c.
495 Wed Nov 23 13:35:31 2005 Tim Janik <timj@gtk.org>
497 * glib/gdataset.c: streamlined initialization code somewhat,
498 removed GData node cache.
500 2005-11-22 Matthias Clasen <mclasen@redhat.com>
502 * glib/ghash.c (g_hash_table_ref):
503 (g_hash_table_unref): Mark these as new API.
505 Avoid double locking in g_intern_string (#322133,
508 * glib/gdataset.c (g_quark_from_string_internal): New
509 internal function which factors out common parts of
510 g_quark_from[_static]_string.
511 (g_quark_from_string, g_quark_from_static_string):
512 Use g_quark_from_string_internal.
513 (g_intern_string, g_intern_static_string): Use
514 g_quark_from_string_internal, and only take the
516 (g_quark_new): Don't store the strings shifted by -1
517 in the g_quarks array.
518 (g_quark_to_string): Adapt to the previous change.
520 Tue Nov 22 14:04:26 2005 Tim Janik <timj@imendio.com>
524 g_hash_table_new_full(): create hash tables with a ref count of 1.
525 g_hash_table_ref(): atomically ref_count+=1
526 g_hash_table_unref(): atomically ref_count-=1, destroys hash table
527 when refcount reaches 0.
528 g_hash_table_destroy(): just destroy keys and values, unref by 1.
529 g_hash_table_insert():
530 g_hash_table_replace(): assert ref_count>0.
533 * glib/gatomic.c: added 'volatile' qualifier to all atomic pointer and
536 2005-11-20 Behdad Esfahbod <behdad@gnome.org>
538 * glib/guniprop.c (g_unichar_get_mirror_char): Remove unused
539 variables. (#321984, Andrew Paprocki)
541 2005-11-18 Matthias Clasen <mclasen@redhat.com>
543 * configure.in: Bump version
545 2005-11-17 Matthias Clasen <mclasen@redhat.com>
547 * === Released 2.9.0 ===
549 * glib/glib.symbols: Updates
551 * glib/gutf8.c (g_utf8_offset_to_pointer): Add some performance
552 hints to the docs. (Paolo Borelli)
556 * glib/gthread.c (g_thread_foreach): Mark as new api.
558 * README.in: Updates.
562 * glib/gdate.c (g_date_set_time_t):
563 (g_date_set_time_val): Add functions to set a date from
564 a time_t and from a GTimeVal.
565 (g_date_set_time): Deprecate. (#314067, Roger Leigh)
568 * tests/utf8-pointer.c: Unit tests for g_utf8_pointer_to_offset
569 and g_utf8_offset_to_pointer.
571 * glib/gutf8.c (g_utf8_pointer_to_offset)
572 (g_utf8_offset_to_pointer): Handle negative offsets, and use
573 "stutter stepping" for going backwards. (#320638, Larry
578 * glib/gthread.c: const correctness fixes, found
579 by Arjan van de Ven and gcc.
581 2005-11-16 Behdad Esfahbod <behdad@gnome.org>
583 * docs/reference/glib/tmpl/unicode.sgml: Correct typo on mentioning
584 @G_UNICODE_BREAK_UNKNOWN that should be #G_UNICODE_BREAK_UNKNOWN.
586 2005-11-09 Behdad Esfahbod <behdad@gnome.org>
588 * glib/guniprop.c: Use bit hacks instead when checking a general
589 category value against multiple values.
591 * glib/gutf8.c: Change ISO10646 to Unicode in docs.
593 2005-11-10 Simos Xenitellis <simos@gnome.org>
595 * configure.in: Added tt (Tatar) to ALL_LINGUAS.
597 2005-11-05 Tor Lillqvist <tml@novell.com>
599 * glib/giowin32.c (g_io_win32_set_flags): Don't call g_warning().
600 (#320688) Instead set error and return failure.
602 2005-11-07 Matthias Clasen <mclasen@redhat.com>
604 * glib/gmain.c (g_main_dispatch): Don't call cb_funcs->unref
605 while holding the context lock. (#320886, Andy Wingo)
607 2005-11-07 Billy Biggs <vektor@dumbterm.net>
609 * tests/refcount/Makefile.am: Add a missing reference to libglib
610 in the LDADD for this test.
612 2005-11-07 Matthias Clasen <mclasen@redhat.com>
614 * glib/gmem.c (g_allocator_new): We need to set n_preallocs to a
615 nonzero value, otherwise GTK+ 2.8 breaks when compiled against
616 GLib 2.9. (#320755, Luca Ferretti)
618 2005-11-04 Matthias Clasen <mclasen@redhat.com>
620 * glib/Makefile.am: Apply a patch from Behdad Esfahbod to
621 use a faster lookup table for g_unichar_get_mirror_char().
622 * glib/gmirroringtable.h: The new table.
623 * glib/gunichartables.h: Remove bidi_mirroring_table.
624 * glib/gen-unicode-tables.pl: Don't generate the mirroring
626 * glib/glib-mirroring-tab/*: Sources for the program
627 which generated gmirroringtable.h.
629 * glib/glist.c: Avoid some code duplication.
631 * glib/gscanner.h: Include gdataset.h. (#320322)
633 * glib/gdebug.h: add new GDebugFlag for fatal_criticals
634 * glib/gmessages.c: (_g_debug_init): handle G_DEBUG=fatal_criticals,
635 to help find critical warnings in applications. (#320017,
638 2005-11-02 Tor Lillqvist <tml@novell.com>
640 * glib/glib.symbols: Remove large amount of trailing whitespace
641 from one line. Remove #ifdef G_UNDEFINED from around
642 g_hash_table_new and g_hash_table_new_full symbols, what was the
643 purpose of that? Add the g_slice functions.
645 Tue Nov 1 16:24:20 2005 Tim Janik <timj@imendio.com>
647 * glib/gmem.[hc]: prepared deprecation of GMemChunk and GAllocator.
648 added g_slice_*() API to allocate and cache small bits of memory.
649 an actuall allocator implementation for g_slice_*() is still pending.
651 * glib/gthread.[hc]: changes from a patch by Matthias Clasen.
652 changed GRealThread list to use in-structure *next; fields instead
653 of GSList, in order for thread iteration to not depenend on g_slice_*()
655 _g_thread_mem_private_get():
656 _g_thread_mem_private_set(): added accessors for private memory,
657 needed because the ordinary GPrivate implementation relies on GArray
658 and GSList and therefore indirectly on working g_slice_*() allocations.
661 g_thread_foreach(): new public API function to loop over all existing threads.
675 removed GAllocator and free list usages and accompanying locks.
676 use g_slice_*() API to allocate and cache small bits of memory.
678 * glib/ghook.h: removed GMemChunk field from public API.
681 * glib/glist.h: deprecate allocator API, provide _free1() for consistency.
683 * glib/gnode.h: deprecate allocator API.
685 * glib/gmain.c: reordered GPollRec fields so g_slice_free_chain() can
686 be used for poll rec lists.
688 * glib/grel.c: removed mem chunk usage, and allocated tuples via g_slice_*().
689 g_relation_destroy(): free all tuples from the all_tuples hash table,
690 this effectively maintains the life time track keeping of tuples.
691 g_relation_delete_tuple(): free tuples which are removed from the
692 all_tuples hash table. this fixes a temporary leak that was present
693 in the memchunk code until the destruction of the relation.
695 2005-10-29 Matthias Clasen <mclasen@redhat.com>
697 * tests/convert-test.c: Add some tests for conversions between
698 UTF-8, UCS-4 and UTF-16.
700 * glib/gutf8.c (g_utf8_to_ucs4, g_utf8_to_utf16): Fix handling
701 of len == -1, noticed by Morten Welinder.
703 2005-10-27 Erdal Ronahi <erdal.ronahi@gmail.com>
705 * configure.in: Added ku (Kurdish) to ALL_LINGUAS
707 2005-10-26 Matthias Clasen <mclasen@redhat.com>
709 * glib/gutf8.c (g_ucs4_to_utf8): Don't set items_read twice
710 in the error case, and add some documentation. (#319806, Morten
713 2005-10-19 Manish Singh <yosh@gimp.org>
715 * configure.in: Use AC_CHECK_FUNCS for _NSGetEnviron, to get the
716 config.h symbol automatically. Fixes bug #313731.
718 2005-10-19 Tor Lillqvist <tml@novell.com>
720 * glib/gwin32.c (g_win32_get_package_installation_directory):
721 Return a g_strdup()ed copy of the value stored in the hash table,
722 so that it can be g_free()d without leaving a dangling pointer in
723 the hash table. (#319232)
725 2005-10-06 Matthias Clasen <mclasen@redhat.com>
727 * glib/gunicollate.c (g_utf8_collate_key_for_filename): Handle
728 all-zero sequences correctly. (#317930, Sebastien Bacher)
730 2005-10-05 Matthias Clasen <mclasen@redhat.com>
732 * glib/gmarkup.c (g_markup_escape_text): Doc addition
734 * Makefile.am: Add Collation tests.
736 * tests/collate/*: Inputs and expected outputs for collation tests.
738 * tests/run-collate-tests.sh: Script to run collation tests.
740 * tests/unicode-collate.c (main): Rework slightly to make
741 it usable in unit tests. Also test g_utf8_collate_key_for_filename().
743 2005-10-01 Behdad Esfahbod <behdad@gnome.org>
745 * docs/reference/glib/tmpl/unicode.sgml:
746 * glib/gen-unicode-tables.pl:
748 * glib/gunichartables.h:
750 * tests/casefold.txt:
751 * tests/casemap.txt: Updated to Unicode 4.1. There are five new
752 GUnicodeBreakType types. That may break some applications, like
755 2005-09-26 Matthias Clasen <mclasen@redhat.com>
757 * glib/gstrfuncs.c (g_ascii_strtoull): Add details to the
758 docs. (#314393, Matthew F. Barnes)
761 * glib/gprintf.h: Remove g_snprintf() and g_vsnprintf(), since
762 they are already declared in glib.h. This doesn't break documented
763 use of gprintf.h, but should probably be pointed out in the
764 release notes for 2.10. (#314232, Behdad Esfahbod)
766 Tue Sep 20 13:16:04 2005 Tim Janik <timj@imendio.com>
768 * glib/gpattern.c (g_pattern_ph_match): applied significant recursion
769 complexity optimization, based on a patch from Matthias Clasen.
771 * tests/patterntest.c: more tests, mostly from matthias.
773 2005-09-20 Matthias Clasen <mclasen@redhat.com>
775 * glib/gqueue.c (g_queue_insert_sorted): Correct the docs.
776 (#316703, Mark Drago)
778 Mon Sep 19 17:23:23 2005 Tim Janik <timj@imendio.com>
780 * glib/gpattern.c: applied a patch from matthias which checks on the
781 upper bound of GPatternSpec length to optimize matches.
784 * tests/patterntest.c: added more match cases.
786 2005-09-16 Tor Lillqvist <tml@novell.com>
788 * glib/gstrfuncs.c (g_ascii_strcasecmp, g_ascii_strncasecmp): Add
789 warning to doc comment that these functions should not be used on
790 encodings like CP932.
792 2005-09-14 Matthias Clasen <mclasen@redhat.com>
794 * tests/keyfile-test.c: Add a test for grup names of length 1.
796 * glib/gkeyfile.c (g_key_file_line_is_group): Accept group names
797 of length 1. (#316309)
799 2005-09-12 Matthias Clasen <mclasen@redhat.com>
801 * glib/gmarkup.c (g_markup_escape_text): Clarify docs.
803 2005-09-11 Sebastian Wilhelmi <seppi@seppi.de>
805 * tests/refcount/Makefile.am (INCLUDES): Link the the refcount
806 tests to the system thread library $(G_THREAD_LIBS). Fixes #313744
809 2005-09-11 Kjartan Maraas <kmaraas@gnome.org>
811 * glib/gmain.c: (g_child_watch_prepare), (g_child_watch_check),
812 (child_watch_helper_thread): Remove some dead code. Closes
815 2005-09-07 Tor Lillqvist <tml@novell.com>
817 * glib/Makefile.am: Create also a console version of the
818 gspawn-win32-helper program, gspawn-win32-helper-console.exe.
819 It's otherwise identical to gspawn-win32-helper.exe, except marked
820 as a console application (linked without the -mwindows option).
822 * glib/gspawn-win32.c (do_spawn_directly, do_spawn_with_pipes):
823 Drop the dont_wait parameter. Its truth value correlated 100% with
824 the NULLness of the exit_status parameter anyway, so it's enough
825 to check whether exit_status is NULL. Invert the sense of the
826 dont_return_handle parameter and rename it to do_return_handle, to
827 make the code easier to read by avoiding double negations.
829 (g_spawn_sync_utf8, g_spawn_async_with_pipes_utf8): Modify calls
830 to do_spawn_with_pipes() accordingly.
832 (do_spawn_with_pipes): If we have a console, use the console
833 version of the helper program, otherwise use the GUI one. This
834 avoids extra console windows opening up in some situations. (In
835 case a console application uses the GUI gspawn-win32-helper.exe to
836 spawn another console application we would get a separate console
837 for the spawned console application).
839 * glib-zip.in: Distribute also gspawn-win32-helper-console.exe.
841 2005-09-05 Matthias Clasen <mclasen@redhat.com>
843 * glib/gmappedfile.c (g_mapped_file_new): Report an error
844 if the file is too large. (#315275, Kjartan Maraas)
846 * glib/gkeyfile.c (g_key_file_load_from_fd): The return value
847 of read() is signed. (#315273, Kjartan Maraas)
849 2005-08-31 Tor Lillqvist <tml@novell.com>
851 * glib/gutils.h: Wrapping atexit() is a bad idea on Windows, where
852 the EXE and each DLL have their own atexit function chains.
854 #define g_atexit as atexit instead. This means it has a
855 better chance of doing what the caller wants. For instance,
856 gtkhtml calls g_atexit() registering a function in gtkhtml
857 itself. This caused a crash when g_atexit() was implemented as a
858 function in the GLib DLL. The gtkhtml DLL was already unloaded by
859 the time the GLib DLL got unloaded.
861 * glib/gutils.c: #undef the #define mentioned above, to also get a
862 real g_atexit() into the DLL for backward compatibility. Document
863 the Windows behaviour of g_atexit(), and document the varying ways
864 atexit() can behave in the context of dynamically loaded modules
867 2005-08-31 Matthias Clasen <mclasen@redhat.com>
871 * glib/gdataset.c: Add string interning functions.
873 2005-08-28 Matthias Clasen <mclasen@redhat.com>
875 * glib/giochannel.c: Unify some near-duplicate strings. (#314654,
878 2005-08-26 Matthias Clasen <mclasen@redhat.com>
880 * configure.in: Bump version to 2.9.0
882 2005-08-25 Tor Lillqvist <tml@novell.com>
884 Make also the g_spawn*() functions take parameters in the GLib
885 file name encoding, i.e. UTF-8, on Windows. Has no impact on Unix
886 API or ABI. Like the other GLib API that was earlier changed to
887 use UTF-8 on Windows, the names of the functions that take UTF-8
888 have _utf8 suffixes added by using preprocessor macros in the
889 header file. The old names are kept for functions with the old
890 behaviour, taking parameters in the system codepage, for DLL ABI
893 * glib/gspawn.h: On Win32 add the suffix _utf8 to the names of the
894 g_spawn*() functions.
896 * glib/gspawn-win32.c: Use wide-char API on NT-based
897 Windows. Convert parameters from UTF-8 to wide chars (NT) or
898 system codepage (Win9x) and call the C library _wspawn*() or
899 spawn*() functions respectvely. Add DLL ABI stability versions
900 that take parameters in the system codepage.
902 * glib/gspawn-win32-helper.c: On NT-based Windows use the
903 wide-char versions of argv and envp, and use wide-char API to
904 change directory and spawn the program to run. Remove the verbose
905 debugging output, it was too complex to modify for the wide-char
906 features. (Just add temporary debugging printouts if needed, no
907 need to have them permanently in the source.)
909 * glib/gspawn.c: Corresponding documentation updates.
911 * glib/glib.symbols: Corresponding changes: Mark the ABI stability
912 symbols as PRIVATE, add the new _utf8-suffixed ones.
914 2005-08-24 Stepan Kasal <kasal@ucw.cz>
916 * glib/gtypes.h (G_MININT64): Cast the constant to gint64; it is
917 guint64 otherwise and that can produce warnings about comparison
918 between signed and unsigned.
920 2005-08-23 Matthias Clasen <mclasen@redhat.com>
922 * glib/gutils.c: Fix the crt_externs.h include.
924 2005-08-23 Stepan Kasal <kasal@ucw.cz>
926 * NEWS: Fix spelling of my first name.
928 2005-08-23 Matthias Clasen <mclasen@redhat.com>
932 * === Released 2.8.1 ===
936 2005-08-20 Hans Breuer <hans@breuer.org>
938 * glib/makefile.msc.in : link with ws2_32.lib
940 2005-08-18 Tor Lillqvist <tml@novell.com>
942 * configure.in: Check for <sys/wait.h>
944 * glib/gbacktrace.c: Include <sys/wait.h> on if HAVE_SYS_WAIT_H.
946 2005-08-18 Ross Burton <ross@burtonini.com>
949 Optimise single-character insertions.
954 * tests/string-test.c:
955 Add tests for new optimisation, and fix a leak.
957 2005-08-17 Matthias Clasen <mclasen@redhat.com>
959 * configure.in: Check for crt_externs.h and _NSGetEnviron.
961 * glib/gutils.c: On Darwin, include crt-externs.h and
962 define environ using _NSGetEnviron(). (#313731)
964 2005-08-16 Stepan Kasal <kasal@ucw.cz>
966 * glib/gutils.c (g_get_any_init): Move the body of the big if...
967 (g_get_any_init_do): ... to this new function.
968 (g_get_any_init): Declare as inline.
969 (g_get_any_init_locked): New inline function, does the locking.
970 Make use of these two throughout the code.
972 2005-08-15 Matthias Clasen <mclasen@redhat.com>
974 * glib/gbacktrace.c (g_on_error_stack_trace): Wait for
975 the child process and then simply return. This makes
976 The "S" option work as documented in g_on_error_query().
977 (#313125, Matthew F. Barnes)
979 * glib/gunicode.h: Update the link to Unicode category
980 values. (#313369, Behnam Esfahbod)
982 * glib/gqueue.c (g_queue_find_custom): Clarify docs
983 a little. (#311727, Tristan van Berkom)
985 * glib/abicheck.sh, gobject/abicheck.sh: Make the
986 check work on ia64 too, where some symbols ended up
987 in yet another section.