1 2005-12-20 Sven Herzberg <herzi@gnome-de.org>
3 * docs/reference/gobject/tmpl/gtype.sgml: explain that
4 G_TYPE_INSTANCE_GET_CLASS() does behave different during
7 2006-01-06 Matthias Clasen <mclasen@redhat.com>
9 * glib/gslice.c: Only define _XOPEN_SOURCE if we know
10 that we have posix_memalign(). (#323937, Bogdan Nicula)
12 2006-01-05 Matthias Clasen <mclasen@redhat.com>
14 * configure.in: Bump version
16 * === Released 2.9.2 ===
18 2006-01-05 Hans Breuer <hans@breuer.org>
20 * glib/makefile.msc.in : added gslice.obj
22 2006-01-05 Matthias Clasen <mclasen@redhat.com>
26 * glib/gmain.c: Clarify the documentation of source ids
27 a bit. (#325874, Dan Williams)
29 * configure.in: Fix another strict aliasing problem.
31 * glib/gthreadpool.c: Avoid use of varargs macro. (#325864,
34 2006-01-04 Manish Singh <yosh@gimp.org>
37 * glib/gstdio.h: don't macroized g_access, g_chdir, and g_unlink
38 either, since they have the same issue as g_rmdir. (Related to
41 Wed Jan 4 13:33:25 2006 Tim Janik <timj@gtk.org>
43 * glib/gslice.c (magazine_cache_pop_magazine): don't reverse chunk
44 order when creating magazines, so we hand out chunks with ascending
47 2006-01-04 Abel Cheung <maddog@linuxhall.org>
49 * configure.in: Added "ml" "zh_HK" to ALL_LINGUAS.
51 2006-01-03 Martyn Russell <martyn@imendio.com>
53 * docs/reference/glib/glib-sections.txt:
55 * glib/gthreadpool.[ch]:
56 - Added new API g_thread_pool_get_idle_time() and
57 g_thread_pool_set_idle_time(). (#324228).
59 * tests/threadpool-test.c:
60 - Updated test case to do thread pool sorting, thread pool with
61 no sorting and a thread pool with idle thread timeouts.
63 2006-01-03 Matthias Clasen <mclasen@redhat.com>
65 * glib/gmain.h: Add new functions here, too.
67 * glib/glib.symbols: Add new functions.
69 2005-12-20 Michael Meeks <michael.meeks@novell.com>
71 * glib/gmain.c (g_main_context_is_owner): new method
72 to determine if the current thread is the owner of the
75 2006-01-02 Matthias Clasen <mclasen@redhat.com>
79 * glib/gstdio.c (g_rmdir): Don't provide g_rmdir() as a macro
80 expanding to rmdir, since rmdir is not declared in a portable
81 system header we can include in gstdio.h. (#325249, Jani Monoses)
83 2006-01-01 Tor Lillqvist <tml@novell.com>
85 * glib/gspawn-win32.c (g_spawn_sync_utf8): Set the GIOChannels for
86 stdout and stderr to unbuffered. Otherwise the giochannel layer
87 will try to read from them regardless whether the
88 g_io_channel_win32_poll() call here has indicated
89 readability or not. (#325310)
91 2005-12-29 Matthias Clasen <mclasen@redhat.com>
93 * glib/gutils.c (glib_check_version): Fix a copy-and-paste error
94 in the docs. (#325273, Declan Naughton)
96 2005-12-27 Manish Singh <yosh@gimp.org>
98 * tests/run-collate-tests.sh: use LC_ALL instead of LC_COLLATE, to
99 make sure we really override things.
101 2005-12-27 Matthias Clasen <mclasen@redhat.com>
103 Fix #316221, Michal Benes, Stanislav Brabec;
105 * configure.in: Fix a strict aliasing problem in
106 g_static_mutex_get_mutex().
107 * glib/gthread.h: ...and in
108 g_static_mutex_get_mutex_impl_shortcut().
110 * glib/gdatasetprivate.h: Add a cast to silence compiler
111 warnings. (#321978, Andrew Paprocki)
113 Partial fix for bug #323937, Bogdan Nicula.
115 * configure.in: Check for malloc.h
116 * glib/gslice.c: Don't include malloc.h unconditionally.
118 2005-12-27 Manish Singh <yosh@gimp.org>
120 * tests/run-collate-tests.sh: export LC_COLLATE so it takes.
121 (#324950, Dan Yefimov)
123 2005-12-26 Matthias Clasen <mclasen@redhat.com>
125 * glib/gslice.c: On Win32, include process.h (#325015, Kazuki
128 2005-12-25 Matthias Clasen <mclasen@redhat.com>
130 * AUTHORS: Update my email
132 * tests/slice-test.c: Fix C99isms. (#324950, Dan Yefimov)
134 2005-12-21 Matthias Clasen <mclasen@redhat.com>
136 * README.in: Add some notes on when 'make check' may fail.
138 2005-12-20 Matthias Clasen <mclasen@redhat.com>
141 * glib/gthreadpool.h:
142 * glib/gthreadpool.c (g_thread_pool_set_sort_function): New function
143 to sort tasks pushed into a threadpool. (#324479, Martyn Russell)
145 * tests/threadpool-test.c: Test this.
147 Tue Dec 20 18:14:14 2005 Tim Janik <timj@imendio.com>
149 * glib/gslice.[hc]: added mem_error() and mem_assert() to test and
150 handle errors without depending on gmessage.c which might not be
151 setup when the error occours.
152 removed G_SLICE_CONFIG_ALWAYS_FREE config option, fixed the code so
153 always freeing can be achieved by adjusting the working set time to
154 0 with G_SLICE_CONFIG_WORKING_SET_MSECS.
155 added G_SLICE_CONFIG_COLOR_INCREMENT to test different color increments
156 (mainly 0 and 1). reduced the minimum block size to 128 bytes, to
157 minimize wastage if small amounts of differently sized structrues are
158 allocated, this does come at a performance cost of roughly 5% though.
159 fixed up block alignment calculation, so it works for varying
160 block sizes. only use strerror() not g_strerror() since the latter
161 depends on working GQuark and GSlice.
162 mem_error(): implemented in terms of fprintf and vfprintf.
164 * tests/slice-color.c: new program to test cache colorization effects.
166 * tests/slice-test.c: trade G_SLICE_CONFIG_ALWAYS_FREE for 0 duration
167 G_SLICE_CONFIG_WORKING_SET_MSECS.
169 2005-12-17 Matthias Clasen <mclasen@redhat.com>
171 * glib/goption.c (parse_short_option): Set an error in all
172 failure cases. (#324332, Tim-Philipp Müller)
174 2005-12-17 Sebastian Wilhelmi <seppi@seppi.de>
176 * glib/gatomic.c: Fix memory barrier position in g_atomic_int_get
177 and g_atomic_pointer_get. Add g_atomic_int_set and
178 g_atomic_pointer_set implementations for the !DEFINE_WITH_MUTEXES &&
179 G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case, as well as defining them
180 as functions (additionally to the macros in the header) for the
181 !G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case.
183 2005-12-16 Matthias Clasen <mclasen@redhat.com>
185 * glib/gmem.c (g_allocator_new): Don't return a pointer to
186 a const struct, since apps expect to be able to modify it.
187 (#324179, J. Ali Harlow)
189 Tue Dec 13 10:13:32 2005 Tim Janik <timj@imendio.com>
191 * glib/gatomic.h: added g_atomic_pointer_set() and g_atomic_int_set()
192 for systems where the initialization of atomic variables requires a
193 write memory barrier.
195 2005-12-09 Matthias Clasen <mclasen@redhat.com>
197 * configure.in: Bump version
199 * === Released 2.9.1 ===
201 2005-12-09 Alexander Larsson <alexl@redhat.com>
205 Use G_GNUC_WARN_UNUSED_RESULT on list functions that return
208 2005-12-08 Matthias Clasen <mclasen@redhat.com>
212 2005-12-07 Martyn Russell <martyn@imendio.com>
214 * glib/gasyncqueue.c:
215 - Call g_queue_insert_sorted() instead of duplicating the code.
216 - Call g_queue_sort() instead of duplicating the code.
217 - Invert sort function results to make sure the same sort function
218 gives the same results across glist, gslist, gqueue and
221 * tests/asyncqueue-test.c:
222 - Updated the sort function to reflect the example in the
223 documentation for gasyncqueue.c.
225 2005-12-07 Martyn Russell <martyn@imendio.com>
227 * docs/reference/glib/glib-sections.txt:
228 * docs/reference/glib/tmpl/linked_lists_double.sgml:
229 * docs/reference/glib/tmpl/linked_lists_single.sgml:
232 - Added g_list_insert_sorted_with_data () and
233 g_slist_insert_sorted_with_data ().
234 - Removed the extra check in g_list_sort() and g_slist_sort() for
235 GCompareDataFunc vs. GCompareFunc.
237 2005-12-07 Tor Lillqvist <tml@novell.com>
239 * glib-gettextize.in: Look up prefix at run-time on Win32,
240 assuming the standard directory structure with glib-gettextize in
243 2005-12-06 Matthias Clasen <mclasen@redhat.com>
245 * glib/gmem.h: Exempt GTK+ from the mem chunk deprecation, since
246 we need GTK+ 2.8 to compile against GLib 2.10.
248 2005-12-05 Manish Singh <yosh@gimp.org>
250 * tests/run-collate-test.sh: set LC_COLLATE, not LANG, to be sure
251 to override any user settings.
253 2005-12-05 Matthias Clasen <mclasen@redhat.com>
255 * configure.in: Revert an accidential commit
257 2005-12-05 Matthias Clasen <mclasen@redhat.com>
259 * glib/gutf8.c: Documentation updates. (#323291, Morten Welinder)
261 * tests/Makefile.am (TESTS_ENVIRONMENT):
262 * tests/gobject/Makefile.am (TESTS_ENVIRONMENT): Set
263 MALLOC_CHECK_ and MALLOC_PERTURB_
265 * tests/run-collate-tests.sh: Run the collation tests explicitly
266 in en_US locale. (#320463)
268 * glib/gmem.h: Really deprecate GMemChunk.
275 * glib/gstrfuncs.c: Add versioned deprecation docs.
277 2005-12-05 Martyn Russell <martyn@imendio.com>
279 * docs/reference/glib/glib-sections.txt:
280 * glib/gasyncqueue.[ch]:
281 - Added support for sorting async queues by with _push_sorted(),
282 _push_sorted_unlocked(), _sort() and _sort_unlocked() (#323047).
285 * tests/asyncqueue-test.c:
286 - Added test case for gasyncqueue.c
288 Mon Dec 5 15:53:20 2005 Tim Janik <timj@imendio.com>
290 * glib/gslice.c: implement chain walking for arbitrary ->next pointer
291 offsets in g_slice_free_chain_with_offset() based on a patch by behdad
292 in bug 323178. moved time consuming logic from g_slice_free() out of
293 the inner loop, so g_slice_free_chain_with_offset() provides a real
294 performance benefit over g_slice_free1() now.
296 * glib/gslice.h: renamed g_slice_free_chain() to
297 g_slice_free_chain_with_offset(). implemented g_slice_free_chain() as
298 a type-safe macro as suggested in bug 323178.
299 simplified the macro implementation of g_slice_free() and implemented
300 it in a type safe manner for all compliers as suggested by Morten
301 Welinder <mortenw@gnome.org>.
306 * glib/glib.symbols: s/g_slice_free_chain/g_slice_free_chain_with_offset/
308 2005-12-05 Matthias Clasen <mclasen@redhat.com>
310 * glib/gasyncqueue.c: Add some docs.
312 * tests/libmoduletestplugin_a.c: Fix compiler warnings.
314 * glib/gatomic.c: In the ia64 implementation, use
315 __sync builtin without _si or _di suffix. (#321229,
316 Stanislav Brabec, patch by Andreas Schwab)
318 2005-12-04 Behdad Esfahbod <behdad@gnome.org>
320 * glib/gslice.h: Remove comma at the end of enum.
322 2005-12-04 Matthias Clasen <mclasen@redhat.com>
324 Handle multiple user names with the same UID better.
325 (#319535, Laszlo Peter)
327 * glib/gutils.c (g_get_any_init_do): When determining user
328 data, first look up $LOGNAME. If the UID doesn't match
329 getuid(), fall back to the current behaviour of looking
330 up the user data based on getuid().
332 2005-12-04 Matthias Clasen <mclasen@redhat.com>
334 * glib/gmacros.h (G_GNUC_WARN_UNUSED_RESULT): Add a macro
335 to make gcc warn if a function result is ignored. (#145466,
336 Arjan van de Ven, Alex Larsson)
338 * glib/gmem.h: Add the new attribute to g_realloc and
341 2005-12-03 Matthias Clasen <mclasen@redhat.com>
344 * glib/gcache.h: Deprecate g_cache_value_foreach. (#322956,
347 * glib/gmappedfile.c: Make mapping of empty files
350 * glib/gfileutils.c: Don't fork a new process just to
351 fix the permissions of the created temp file. (#321318,
352 Alexis S. L. Carvalho)
354 2005-12-02 Matthias Clasen <mclasen@redhat.com>
356 * README.in: Add a note about Solaris threads.
359 * configure.in: Remove support for Solaris threads.
360 (#136971, Sebastian Wilhelmi, patch by Andrew Paprocki)
362 2005-12-02 Matthias Clasen <mclasen@redhat.com>
364 * configure.in: Define G_GUINT64_CONSTANT in analogy to
365 G_GINT64_CONSTANT. (#322568, Andrew Paprocki)
367 2005-12-02 Matthias Clasen <mclasen@redhat.com>
369 * glib/gslice.c: Win32 portability fixes and C99ism removal,
370 pointed out by Kazuki Iwamoto. (#323052)
372 2005-12-02 Matthias Clasen <mclasen@redhat.com>
374 * glib/gatomic.c (g_atomic_int_add): Add a missing volatile for
375 the IA64 implementation.
377 Fri Dec 2 16:18:09 2005 Tim Janik <timj@imendio.com>
379 * glib/gslice.c: some naming and type size fixups.
381 Fri Dec 2 13:08:58 2005 Tim Janik <timj@gtk.org>
383 * glib/gslice.h (g_slice_free): reworked GCC-specific type-safe macro
384 variant into something less verbose, but digestible for gcc-3.4.
386 Fri Dec 2 10:55:07 2005 Tim Janik <timj@imendio.com>
388 * tests/slice-test.c: extended to perform the benchmarking on the old
389 memchunk code if 'O' is selected.
391 * tests/memchunks.c: new file which contains the old GLib mem chunks
392 implementation with prefix old_mem_chunk_.
394 * tests/Makefile.am: added memchunks.c
396 Fri Dec 2 00:16:59 2005 Tim Janik <timj@imendio.com>
398 * glib/gslice.c: improved gettimeofday() timer resolution by maintaining
399 acceptable load distribution of the syscall.
400 (allocator_get_magazine_threshold): reduce minimum/single-thread magazine
401 sizes to approximately page_size/2.5 to avoid excessive startup allocations.
403 2005-12-01 Matthias Clasen <mclasen@redhat.com>
405 * tests/gobject/ifacecheck.c (test_iface_base_init): Don't
406 double-free base interface members.
408 * tests/Makefile.am: Remove duplicate variable.
410 Thu Dec 1 17:32:46 2005 Tim Janik <timj@imendio.com>
412 * glib/gslice.[hc]: new slice allocator implementation.
414 * tests/slice-test.c: added random slice allocation test.
416 * glib/gthread.[hc]: removed newly added private thread mem API.
418 * glib/gthreadinit.h:
421 * glib/gmem.c: divided glib threading initialisation into three phases,
422 initialisation where private keys and messaging are not available (only
423 needed by gmem.c), initialisation without messaging but private keys
424 available (gslice.c, gmessage.c), and full fledged initialisers that
425 server the rest of glib. initialisation functions got renamed to reflect
426 the limitations of their corresponding phases.
428 * glib/gmem.c: removed memchunk code, defer allocations to
431 * glib/gmem.[hc]: removed g_slice_* skeletons.
433 * glib/glib.symbols: added g_slice_* symbols.
435 * configure.in: check for availability of posix_memalign(3), memalign(3)
438 * glib/Makefile.am: added gslice.[hc].
440 2005-12-01 Tor Lillqvist <tml@novell.com>
442 * glib/gstdio.c (g_stat): In the Win32 implementation, strip
443 trailing slash(es) for non-root folders. stat() fails if non-root
444 folders are specified with trailing slashes. It's too much hassle
445 to demand that callers strip such slashes themselves, especially
446 as it is easy to get it wrong and strip the slash of a root
448 (g_rename): On NT-based Windows, use MoveFileEx() with
449 MOVEFILE_REPLACE_EXISTING to better match Unix behaviour.
451 2005-11-28 Matthias Clasen <mclasen@redhat.com>
453 Fix G_STMT_START / G_STMT_END on Solaris. (#321972,
456 * configure.in: Check whether do { } while (0) works.
458 * glib/gmacros.h: Use do { } while (0) for G_STMT_START /
459 G_STMT_END if it works.
461 2005-11-28 Matthias Clasen <mclasen@redhat.com>
463 * glib/gthread.c (g_static_rw_lock_wait, g_static_rw_lock_signal):
464 * glib/gnode.c (g_node_depth_traverse_level):
465 * glib/gmem.c (g_allocator_new):
466 * glib/ghash.c (g_hash_table_unref, g_hash_table_destroy)
467 (g_hash_table_foreach_remove_or_steal): Silence compiler
470 2005-11-27 Matthias Clasen <mclasen@redhat.com>
472 * glib/gunicollate.c (g_utf8_collate_key_for_filename):
473 Don't read beyond len. (#322520, Christian Persch)
475 Wed Nov 23 17:34:01 2005 Tim Janik <timj@imendio.com>
477 * glib/gdataset.c: access datalist flags via atomic pointer access
478 functions, instead of acquiring the dataset lock. this is faster and
479 also matches the atomic pointer readouts in gdatalistprivate.h and
480 direct pointer modifications required by gobject.c.
482 Wed Nov 23 13:35:31 2005 Tim Janik <timj@gtk.org>
484 * glib/gdataset.c: streamlined initialization code somewhat,
485 removed GData node cache.
487 2005-11-22 Matthias Clasen <mclasen@redhat.com>
489 * glib/ghash.c (g_hash_table_ref):
490 (g_hash_table_unref): Mark these as new API.
492 Avoid double locking in g_intern_string (#322133,
495 * glib/gdataset.c (g_quark_from_string_internal): New
496 internal function which factors out common parts of
497 g_quark_from[_static]_string.
498 (g_quark_from_string, g_quark_from_static_string):
499 Use g_quark_from_string_internal.
500 (g_intern_string, g_intern_static_string): Use
501 g_quark_from_string_internal, and only take the
503 (g_quark_new): Don't store the strings shifted by -1
504 in the g_quarks array.
505 (g_quark_to_string): Adapt to the previous change.
507 Tue Nov 22 14:04:26 2005 Tim Janik <timj@imendio.com>
511 g_hash_table_new_full(): create hash tables with a ref count of 1.
512 g_hash_table_ref(): atomically ref_count+=1
513 g_hash_table_unref(): atomically ref_count-=1, destroys hash table
514 when refcount reaches 0.
515 g_hash_table_destroy(): just destroy keys and values, unref by 1.
516 g_hash_table_insert():
517 g_hash_table_replace(): assert ref_count>0.
520 * glib/gatomic.c: added 'volatile' qualifier to all atomic pointer and
523 2005-11-20 Behdad Esfahbod <behdad@gnome.org>
525 * glib/guniprop.c (g_unichar_get_mirror_char): Remove unused
526 variables. (#321984, Andrew Paprocki)
528 2005-11-18 Matthias Clasen <mclasen@redhat.com>
530 * configure.in: Bump version
532 2005-11-17 Matthias Clasen <mclasen@redhat.com>
534 * === Released 2.9.0 ===
536 * glib/glib.symbols: Updates
538 * glib/gutf8.c (g_utf8_offset_to_pointer): Add some performance
539 hints to the docs. (Paolo Borelli)
543 * glib/gthread.c (g_thread_foreach): Mark as new api.
545 * README.in: Updates.
549 * glib/gdate.c (g_date_set_time_t):
550 (g_date_set_time_val): Add functions to set a date from
551 a time_t and from a GTimeVal.
552 (g_date_set_time): Deprecate. (#314067, Roger Leigh)
555 * tests/utf8-pointer.c: Unit tests for g_utf8_pointer_to_offset
556 and g_utf8_offset_to_pointer.
558 * glib/gutf8.c (g_utf8_pointer_to_offset)
559 (g_utf8_offset_to_pointer): Handle negative offsets, and use
560 "stutter stepping" for going backwards. (#320638, Larry
565 * glib/gthread.c: const correctness fixes, found
566 by Arjan van de Ven and gcc.
568 2005-11-16 Behdad Esfahbod <behdad@gnome.org>
570 * docs/reference/glib/tmpl/unicode.sgml: Correct typo on mentioning
571 @G_UNICODE_BREAK_UNKNOWN that should be #G_UNICODE_BREAK_UNKNOWN.
573 2005-11-09 Behdad Esfahbod <behdad@gnome.org>
575 * glib/guniprop.c: Use bit hacks instead when checking a general
576 category value against multiple values.
578 * glib/gutf8.c: Change ISO10646 to Unicode in docs.
580 2005-11-10 Simos Xenitellis <simos@gnome.org>
582 * configure.in: Added tt (Tatar) to ALL_LINGUAS.
584 2005-11-05 Tor Lillqvist <tml@novell.com>
586 * glib/giowin32.c (g_io_win32_set_flags): Don't call g_warning().
587 (#320688) Instead set error and return failure.
589 2005-11-07 Matthias Clasen <mclasen@redhat.com>
591 * glib/gmain.c (g_main_dispatch): Don't call cb_funcs->unref
592 while holding the context lock. (#320886, Andy Wingo)
594 2005-11-07 Billy Biggs <vektor@dumbterm.net>
596 * tests/refcount/Makefile.am: Add a missing reference to libglib
597 in the LDADD for this test.
599 2005-11-07 Matthias Clasen <mclasen@redhat.com>
601 * glib/gmem.c (g_allocator_new): We need to set n_preallocs to a
602 nonzero value, otherwise GTK+ 2.8 breaks when compiled against
603 GLib 2.9. (#320755, Luca Ferretti)
605 2005-11-04 Matthias Clasen <mclasen@redhat.com>
607 * glib/Makefile.am: Apply a patch from Behdad Esfahbod to
608 use a faster lookup table for g_unichar_get_mirror_char().
609 * glib/gmirroringtable.h: The new table.
610 * glib/gunichartables.h: Remove bidi_mirroring_table.
611 * glib/gen-unicode-tables.pl: Don't generate the mirroring
613 * glib/glib-mirroring-tab/*: Sources for the program
614 which generated gmirroringtable.h.
616 * glib/glist.c: Avoid some code duplication.
618 * glib/gscanner.h: Include gdataset.h. (#320322)
620 * glib/gdebug.h: add new GDebugFlag for fatal_criticals
621 * glib/gmessages.c: (_g_debug_init): handle G_DEBUG=fatal_criticals,
622 to help find critical warnings in applications. (#320017,
625 2005-11-02 Tor Lillqvist <tml@novell.com>
627 * glib/glib.symbols: Remove large amount of trailing whitespace
628 from one line. Remove #ifdef G_UNDEFINED from around
629 g_hash_table_new and g_hash_table_new_full symbols, what was the
630 purpose of that? Add the g_slice functions.
632 Tue Nov 1 16:24:20 2005 Tim Janik <timj@imendio.com>
634 * glib/gmem.[hc]: prepared deprecation of GMemChunk and GAllocator.
635 added g_slice_*() API to allocate and cache small bits of memory.
636 an actuall allocator implementation for g_slice_*() is still pending.
638 * glib/gthread.[hc]: changes from a patch by Matthias Clasen.
639 changed GRealThread list to use in-structure *next; fields instead
640 of GSList, in order for thread iteration to not depenend on g_slice_*()
642 _g_thread_mem_private_get():
643 _g_thread_mem_private_set(): added accessors for private memory,
644 needed because the ordinary GPrivate implementation relies on GArray
645 and GSList and therefore indirectly on working g_slice_*() allocations.
648 g_thread_foreach(): new public API function to loop over all existing threads.
662 removed GAllocator and free list usages and accompanying locks.
663 use g_slice_*() API to allocate and cache small bits of memory.
665 * glib/ghook.h: removed GMemChunk field from public API.
668 * glib/glist.h: deprecate allocator API, provide _free1() for consistency.
670 * glib/gnode.h: deprecate allocator API.
672 * glib/gmain.c: reordered GPollRec fields so g_slice_free_chain() can
673 be used for poll rec lists.
675 * glib/grel.c: removed mem chunk usage, and allocated tuples via g_slice_*().
676 g_relation_destroy(): free all tuples from the all_tuples hash table,
677 this effectively maintains the life time track keeping of tuples.
678 g_relation_delete_tuple(): free tuples which are removed from the
679 all_tuples hash table. this fixes a temporary leak that was present
680 in the memchunk code until the destruction of the relation.
682 2005-10-29 Matthias Clasen <mclasen@redhat.com>
684 * tests/convert-test.c: Add some tests for conversions between
685 UTF-8, UCS-4 and UTF-16.
687 * glib/gutf8.c (g_utf8_to_ucs4, g_utf8_to_utf16): Fix handling
688 of len == -1, noticed by Morten Welinder.
690 2005-10-27 Erdal Ronahi <erdal.ronahi@gmail.com>
692 * configure.in: Added ku (Kurdish) to ALL_LINGUAS
694 2005-10-26 Matthias Clasen <mclasen@redhat.com>
696 * glib/gutf8.c (g_ucs4_to_utf8): Don't set items_read twice
697 in the error case, and add some documentation. (#319806, Morten
700 2005-10-19 Manish Singh <yosh@gimp.org>
702 * configure.in: Use AC_CHECK_FUNCS for _NSGetEnviron, to get the
703 config.h symbol automatically. Fixes bug #313731.
705 2005-10-19 Tor Lillqvist <tml@novell.com>
707 * glib/gwin32.c (g_win32_get_package_installation_directory):
708 Return a g_strdup()ed copy of the value stored in the hash table,
709 so that it can be g_free()d without leaving a dangling pointer in
710 the hash table. (#319232)
712 2005-10-06 Matthias Clasen <mclasen@redhat.com>
714 * glib/gunicollate.c (g_utf8_collate_key_for_filename): Handle
715 all-zero sequences correctly. (#317930, Sebastien Bacher)
717 2005-10-05 Matthias Clasen <mclasen@redhat.com>
719 * glib/gmarkup.c (g_markup_escape_text): Doc addition
721 * Makefile.am: Add Collation tests.
723 * tests/collate/*: Inputs and expected outputs for collation tests.
725 * tests/run-collate-tests.sh: Script to run collation tests.
727 * tests/unicode-collate.c (main): Rework slightly to make
728 it usable in unit tests. Also test g_utf8_collate_key_for_filename().
730 2005-10-01 Behdad Esfahbod <behdad@gnome.org>
732 * docs/reference/glib/tmpl/unicode.sgml:
733 * glib/gen-unicode-tables.pl:
735 * glib/gunichartables.h:
737 * tests/casefold.txt:
738 * tests/casemap.txt: Updated to Unicode 4.1. There are five new
739 GUnicodeBreakType types. That may break some applications, like
742 2005-09-26 Matthias Clasen <mclasen@redhat.com>
744 * glib/gstrfuncs.c (g_ascii_strtoull): Add details to the
745 docs. (#314393, Matthew F. Barnes)
748 * glib/gprintf.h: Remove g_snprintf() and g_vsnprintf(), since
749 they are already declared in glib.h. This doesn't break documented
750 use of gprintf.h, but should probably be pointed out in the
751 release notes for 2.10. (#314232, Behdad Esfahbod)
753 Tue Sep 20 13:16:04 2005 Tim Janik <timj@imendio.com>
755 * glib/gpattern.c (g_pattern_ph_match): applied significant recursion
756 complexity optimization, based on a patch from Matthias Clasen.
758 * tests/patterntest.c: more tests, mostly from matthias.
760 2005-09-20 Matthias Clasen <mclasen@redhat.com>
762 * glib/gqueue.c (g_queue_insert_sorted): Correct the docs.
763 (#316703, Mark Drago)
765 Mon Sep 19 17:23:23 2005 Tim Janik <timj@imendio.com>
767 * glib/gpattern.c: applied a patch from matthias which checks on the
768 upper bound of GPatternSpec length to optimize matches.
771 * tests/patterntest.c: added more match cases.
773 2005-09-16 Tor Lillqvist <tml@novell.com>
775 * glib/gstrfuncs.c (g_ascii_strcasecmp, g_ascii_strncasecmp): Add
776 warning to doc comment that these functions should not be used on
777 encodings like CP932.
779 2005-09-14 Matthias Clasen <mclasen@redhat.com>
781 * tests/keyfile-test.c: Add a test for grup names of length 1.
783 * glib/gkeyfile.c (g_key_file_line_is_group): Accept group names
784 of length 1. (#316309)
786 2005-09-12 Matthias Clasen <mclasen@redhat.com>
788 * glib/gmarkup.c (g_markup_escape_text): Clarify docs.
790 2005-09-11 Sebastian Wilhelmi <seppi@seppi.de>
792 * tests/refcount/Makefile.am (INCLUDES): Link the the refcount
793 tests to the system thread library $(G_THREAD_LIBS). Fixes #313744
796 2005-09-11 Kjartan Maraas <kmaraas@gnome.org>
798 * glib/gmain.c: (g_child_watch_prepare), (g_child_watch_check),
799 (child_watch_helper_thread): Remove some dead code. Closes
802 2005-09-07 Tor Lillqvist <tml@novell.com>
804 * glib/Makefile.am: Create also a console version of the
805 gspawn-win32-helper program, gspawn-win32-helper-console.exe.
806 It's otherwise identical to gspawn-win32-helper.exe, except marked
807 as a console application (linked without the -mwindows option).
809 * glib/gspawn-win32.c (do_spawn_directly, do_spawn_with_pipes):
810 Drop the dont_wait parameter. Its truth value correlated 100% with
811 the NULLness of the exit_status parameter anyway, so it's enough
812 to check whether exit_status is NULL. Invert the sense of the
813 dont_return_handle parameter and rename it to do_return_handle, to
814 make the code easier to read by avoiding double negations.
816 (g_spawn_sync_utf8, g_spawn_async_with_pipes_utf8): Modify calls
817 to do_spawn_with_pipes() accordingly.
819 (do_spawn_with_pipes): If we have a console, use the console
820 version of the helper program, otherwise use the GUI one. This
821 avoids extra console windows opening up in some situations. (In
822 case a console application uses the GUI gspawn-win32-helper.exe to
823 spawn another console application we would get a separate console
824 for the spawned console application).
826 * glib-zip.in: Distribute also gspawn-win32-helper-console.exe.
828 2005-09-05 Matthias Clasen <mclasen@redhat.com>
830 * glib/gmappedfile.c (g_mapped_file_new): Report an error
831 if the file is too large. (#315275, Kjartan Maraas)
833 * glib/gkeyfile.c (g_key_file_load_from_fd): The return value
834 of read() is signed. (#315273, Kjartan Maraas)
836 2005-08-31 Tor Lillqvist <tml@novell.com>
838 * glib/gutils.h: Wrapping atexit() is a bad idea on Windows, where
839 the EXE and each DLL have their own atexit function chains.
841 #define g_atexit as atexit instead. This means it has a
842 better chance of doing what the caller wants. For instance,
843 gtkhtml calls g_atexit() registering a function in gtkhtml
844 itself. This caused a crash when g_atexit() was implemented as a
845 function in the GLib DLL. The gtkhtml DLL was already unloaded by
846 the time the GLib DLL got unloaded.
848 * glib/gutils.c: #undef the #define mentioned above, to also get a
849 real g_atexit() into the DLL for backward compatibility. Document
850 the Windows behaviour of g_atexit(), and document the varying ways
851 atexit() can behave in the context of dynamically loaded modules
854 2005-08-31 Matthias Clasen <mclasen@redhat.com>
858 * glib/gdataset.c: Add string interning functions.
860 2005-08-28 Matthias Clasen <mclasen@redhat.com>
862 * glib/giochannel.c: Unify some near-duplicate strings. (#314654,
865 2005-08-26 Matthias Clasen <mclasen@redhat.com>
867 * configure.in: Bump version to 2.9.0
869 2005-08-25 Tor Lillqvist <tml@novell.com>
871 Make also the g_spawn*() functions take parameters in the GLib
872 file name encoding, i.e. UTF-8, on Windows. Has no impact on Unix
873 API or ABI. Like the other GLib API that was earlier changed to
874 use UTF-8 on Windows, the names of the functions that take UTF-8
875 have _utf8 suffixes added by using preprocessor macros in the
876 header file. The old names are kept for functions with the old
877 behaviour, taking parameters in the system codepage, for DLL ABI
880 * glib/gspawn.h: On Win32 add the suffix _utf8 to the names of the
881 g_spawn*() functions.
883 * glib/gspawn-win32.c: Use wide-char API on NT-based
884 Windows. Convert parameters from UTF-8 to wide chars (NT) or
885 system codepage (Win9x) and call the C library _wspawn*() or
886 spawn*() functions respectvely. Add DLL ABI stability versions
887 that take parameters in the system codepage.
889 * glib/gspawn-win32-helper.c: On NT-based Windows use the
890 wide-char versions of argv and envp, and use wide-char API to
891 change directory and spawn the program to run. Remove the verbose
892 debugging output, it was too complex to modify for the wide-char
893 features. (Just add temporary debugging printouts if needed, no
894 need to have them permanently in the source.)
896 * glib/gspawn.c: Corresponding documentation updates.
898 * glib/glib.symbols: Corresponding changes: Mark the ABI stability
899 symbols as PRIVATE, add the new _utf8-suffixed ones.
901 2005-08-24 Stepan Kasal <kasal@ucw.cz>
903 * glib/gtypes.h (G_MININT64): Cast the constant to gint64; it is
904 guint64 otherwise and that can produce warnings about comparison
905 between signed and unsigned.
907 2005-08-23 Matthias Clasen <mclasen@redhat.com>
909 * glib/gutils.c: Fix the crt_externs.h include.
911 2005-08-23 Stepan Kasal <kasal@ucw.cz>
913 * NEWS: Fix spelling of my first name.
915 2005-08-23 Matthias Clasen <mclasen@redhat.com>
919 * === Released 2.8.1 ===
923 2005-08-20 Hans Breuer <hans@breuer.org>
925 * glib/makefile.msc.in : link with ws2_32.lib
927 2005-08-18 Tor Lillqvist <tml@novell.com>
929 * configure.in: Check for <sys/wait.h>
931 * glib/gbacktrace.c: Include <sys/wait.h> on if HAVE_SYS_WAIT_H.
933 2005-08-18 Ross Burton <ross@burtonini.com>
936 Optimise single-character insertions.
941 * tests/string-test.c:
942 Add tests for new optimisation, and fix a leak.
944 2005-08-17 Matthias Clasen <mclasen@redhat.com>
946 * configure.in: Check for crt_externs.h and _NSGetEnviron.
948 * glib/gutils.c: On Darwin, include crt-externs.h and
949 define environ using _NSGetEnviron(). (#313731)
951 2005-08-16 Stepan Kasal <kasal@ucw.cz>
953 * glib/gutils.c (g_get_any_init): Move the body of the big if...
954 (g_get_any_init_do): ... to this new function.
955 (g_get_any_init): Declare as inline.
956 (g_get_any_init_locked): New inline function, does the locking.
957 Make use of these two throughout the code.
959 2005-08-15 Matthias Clasen <mclasen@redhat.com>
961 * glib/gbacktrace.c (g_on_error_stack_trace): Wait for
962 the child process and then simply return. This makes
963 The "S" option work as documented in g_on_error_query().
964 (#313125, Matthew F. Barnes)
966 * glib/gunicode.h: Update the link to Unicode category
967 values. (#313369, Behnam Esfahbod)
969 * glib/gqueue.c (g_queue_find_custom): Clarify docs
970 a little. (#311727, Tristan van Berkom)
972 * glib/abicheck.sh, gobject/abicheck.sh: Make the
973 check work on ia64 too, where some symbols ended up
974 in yet another section.