1 2006-01-08 Matthias Clasen <mclasen@redhat.com>
3 * glib/gtree.c: Break some long lines.
5 * tests/tree-test.c: Add more tests.
7 * glib/gslice.c: Include config.h first, pointed out
10 2005-12-20 Sven Herzberg <herzi@gnome-de.org>
12 * docs/reference/gobject/tmpl/gtype.sgml: explain that
13 G_TYPE_INSTANCE_GET_CLASS() does behave different during
16 2006-01-06 Matthias Clasen <mclasen@redhat.com>
18 * glib/gslice.c: Only define _XOPEN_SOURCE if we know
19 that we have posix_memalign(). (#323937, Bogdan Nicula)
21 2006-01-05 Matthias Clasen <mclasen@redhat.com>
23 * configure.in: Bump version
25 * === Released 2.9.2 ===
27 2006-01-05 Hans Breuer <hans@breuer.org>
29 * glib/makefile.msc.in : added gslice.obj
31 2006-01-05 Matthias Clasen <mclasen@redhat.com>
35 * glib/gmain.c: Clarify the documentation of source ids
36 a bit. (#325874, Dan Williams)
38 * configure.in: Fix another strict aliasing problem.
40 * glib/gthreadpool.c: Avoid use of varargs macro. (#325864,
43 2006-01-04 Manish Singh <yosh@gimp.org>
46 * glib/gstdio.h: don't macroized g_access, g_chdir, and g_unlink
47 either, since they have the same issue as g_rmdir. (Related to
50 Wed Jan 4 13:33:25 2006 Tim Janik <timj@gtk.org>
52 * glib/gslice.c (magazine_cache_pop_magazine): don't reverse chunk
53 order when creating magazines, so we hand out chunks with ascending
56 2006-01-04 Abel Cheung <maddog@linuxhall.org>
58 * configure.in: Added "ml" "zh_HK" to ALL_LINGUAS.
60 2006-01-03 Martyn Russell <martyn@imendio.com>
62 * docs/reference/glib/glib-sections.txt:
64 * glib/gthreadpool.[ch]:
65 - Added new API g_thread_pool_get_idle_time() and
66 g_thread_pool_set_idle_time(). (#324228).
68 * tests/threadpool-test.c:
69 - Updated test case to do thread pool sorting, thread pool with
70 no sorting and a thread pool with idle thread timeouts.
72 2006-01-03 Matthias Clasen <mclasen@redhat.com>
74 * glib/gmain.h: Add new functions here, too.
76 * glib/glib.symbols: Add new functions.
78 2005-12-20 Michael Meeks <michael.meeks@novell.com>
80 * glib/gmain.c (g_main_context_is_owner): new method
81 to determine if the current thread is the owner of the
84 2006-01-02 Matthias Clasen <mclasen@redhat.com>
88 * glib/gstdio.c (g_rmdir): Don't provide g_rmdir() as a macro
89 expanding to rmdir, since rmdir is not declared in a portable
90 system header we can include in gstdio.h. (#325249, Jani Monoses)
92 2006-01-01 Tor Lillqvist <tml@novell.com>
94 * glib/gspawn-win32.c (g_spawn_sync_utf8): Set the GIOChannels for
95 stdout and stderr to unbuffered. Otherwise the giochannel layer
96 will try to read from them regardless whether the
97 g_io_channel_win32_poll() call here has indicated
98 readability or not. (#325310)
100 2005-12-29 Matthias Clasen <mclasen@redhat.com>
102 * glib/gutils.c (glib_check_version): Fix a copy-and-paste error
103 in the docs. (#325273, Declan Naughton)
105 2005-12-27 Manish Singh <yosh@gimp.org>
107 * tests/run-collate-tests.sh: use LC_ALL instead of LC_COLLATE, to
108 make sure we really override things.
110 2005-12-27 Matthias Clasen <mclasen@redhat.com>
112 Fix #316221, Michal Benes, Stanislav Brabec;
114 * configure.in: Fix a strict aliasing problem in
115 g_static_mutex_get_mutex().
116 * glib/gthread.h: ...and in
117 g_static_mutex_get_mutex_impl_shortcut().
119 * glib/gdatasetprivate.h: Add a cast to silence compiler
120 warnings. (#321978, Andrew Paprocki)
122 Partial fix for bug #323937, Bogdan Nicula.
124 * configure.in: Check for malloc.h
125 * glib/gslice.c: Don't include malloc.h unconditionally.
127 2005-12-27 Manish Singh <yosh@gimp.org>
129 * tests/run-collate-tests.sh: export LC_COLLATE so it takes.
130 (#324950, Dan Yefimov)
132 2005-12-26 Matthias Clasen <mclasen@redhat.com>
134 * glib/gslice.c: On Win32, include process.h (#325015, Kazuki
137 2005-12-25 Matthias Clasen <mclasen@redhat.com>
139 * AUTHORS: Update my email
141 * tests/slice-test.c: Fix C99isms. (#324950, Dan Yefimov)
143 2005-12-21 Matthias Clasen <mclasen@redhat.com>
145 * README.in: Add some notes on when 'make check' may fail.
147 2005-12-20 Matthias Clasen <mclasen@redhat.com>
150 * glib/gthreadpool.h:
151 * glib/gthreadpool.c (g_thread_pool_set_sort_function): New function
152 to sort tasks pushed into a threadpool. (#324479, Martyn Russell)
154 * tests/threadpool-test.c: Test this.
156 Tue Dec 20 18:14:14 2005 Tim Janik <timj@imendio.com>
158 * glib/gslice.[hc]: added mem_error() and mem_assert() to test and
159 handle errors without depending on gmessage.c which might not be
160 setup when the error occours.
161 removed G_SLICE_CONFIG_ALWAYS_FREE config option, fixed the code so
162 always freeing can be achieved by adjusting the working set time to
163 0 with G_SLICE_CONFIG_WORKING_SET_MSECS.
164 added G_SLICE_CONFIG_COLOR_INCREMENT to test different color increments
165 (mainly 0 and 1). reduced the minimum block size to 128 bytes, to
166 minimize wastage if small amounts of differently sized structrues are
167 allocated, this does come at a performance cost of roughly 5% though.
168 fixed up block alignment calculation, so it works for varying
169 block sizes. only use strerror() not g_strerror() since the latter
170 depends on working GQuark and GSlice.
171 mem_error(): implemented in terms of fprintf and vfprintf.
173 * tests/slice-color.c: new program to test cache colorization effects.
175 * tests/slice-test.c: trade G_SLICE_CONFIG_ALWAYS_FREE for 0 duration
176 G_SLICE_CONFIG_WORKING_SET_MSECS.
178 2005-12-17 Matthias Clasen <mclasen@redhat.com>
180 * glib/goption.c (parse_short_option): Set an error in all
181 failure cases. (#324332, Tim-Philipp Müller)
183 2005-12-17 Sebastian Wilhelmi <seppi@seppi.de>
185 * glib/gatomic.c: Fix memory barrier position in g_atomic_int_get
186 and g_atomic_pointer_get. Add g_atomic_int_set and
187 g_atomic_pointer_set implementations for the !DEFINE_WITH_MUTEXES &&
188 G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case, as well as defining them
189 as functions (additionally to the macros in the header) for the
190 !G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case.
192 2005-12-16 Matthias Clasen <mclasen@redhat.com>
194 * glib/gmem.c (g_allocator_new): Don't return a pointer to
195 a const struct, since apps expect to be able to modify it.
196 (#324179, J. Ali Harlow)
198 Tue Dec 13 10:13:32 2005 Tim Janik <timj@imendio.com>
200 * glib/gatomic.h: added g_atomic_pointer_set() and g_atomic_int_set()
201 for systems where the initialization of atomic variables requires a
202 write memory barrier.
204 2005-12-09 Matthias Clasen <mclasen@redhat.com>
206 * configure.in: Bump version
208 * === Released 2.9.1 ===
210 2005-12-09 Alexander Larsson <alexl@redhat.com>
214 Use G_GNUC_WARN_UNUSED_RESULT on list functions that return
217 2005-12-08 Matthias Clasen <mclasen@redhat.com>
221 2005-12-07 Martyn Russell <martyn@imendio.com>
223 * glib/gasyncqueue.c:
224 - Call g_queue_insert_sorted() instead of duplicating the code.
225 - Call g_queue_sort() instead of duplicating the code.
226 - Invert sort function results to make sure the same sort function
227 gives the same results across glist, gslist, gqueue and
230 * tests/asyncqueue-test.c:
231 - Updated the sort function to reflect the example in the
232 documentation for gasyncqueue.c.
234 2005-12-07 Martyn Russell <martyn@imendio.com>
236 * docs/reference/glib/glib-sections.txt:
237 * docs/reference/glib/tmpl/linked_lists_double.sgml:
238 * docs/reference/glib/tmpl/linked_lists_single.sgml:
241 - Added g_list_insert_sorted_with_data () and
242 g_slist_insert_sorted_with_data ().
243 - Removed the extra check in g_list_sort() and g_slist_sort() for
244 GCompareDataFunc vs. GCompareFunc.
246 2005-12-07 Tor Lillqvist <tml@novell.com>
248 * glib-gettextize.in: Look up prefix at run-time on Win32,
249 assuming the standard directory structure with glib-gettextize in
252 2005-12-06 Matthias Clasen <mclasen@redhat.com>
254 * glib/gmem.h: Exempt GTK+ from the mem chunk deprecation, since
255 we need GTK+ 2.8 to compile against GLib 2.10.
257 2005-12-05 Manish Singh <yosh@gimp.org>
259 * tests/run-collate-test.sh: set LC_COLLATE, not LANG, to be sure
260 to override any user settings.
262 2005-12-05 Matthias Clasen <mclasen@redhat.com>
264 * configure.in: Revert an accidential commit
266 2005-12-05 Matthias Clasen <mclasen@redhat.com>
268 * glib/gutf8.c: Documentation updates. (#323291, Morten Welinder)
270 * tests/Makefile.am (TESTS_ENVIRONMENT):
271 * tests/gobject/Makefile.am (TESTS_ENVIRONMENT): Set
272 MALLOC_CHECK_ and MALLOC_PERTURB_
274 * tests/run-collate-tests.sh: Run the collation tests explicitly
275 in en_US locale. (#320463)
277 * glib/gmem.h: Really deprecate GMemChunk.
284 * glib/gstrfuncs.c: Add versioned deprecation docs.
286 2005-12-05 Martyn Russell <martyn@imendio.com>
288 * docs/reference/glib/glib-sections.txt:
289 * glib/gasyncqueue.[ch]:
290 - Added support for sorting async queues by with _push_sorted(),
291 _push_sorted_unlocked(), _sort() and _sort_unlocked() (#323047).
294 * tests/asyncqueue-test.c:
295 - Added test case for gasyncqueue.c
297 Mon Dec 5 15:53:20 2005 Tim Janik <timj@imendio.com>
299 * glib/gslice.c: implement chain walking for arbitrary ->next pointer
300 offsets in g_slice_free_chain_with_offset() based on a patch by behdad
301 in bug 323178. moved time consuming logic from g_slice_free() out of
302 the inner loop, so g_slice_free_chain_with_offset() provides a real
303 performance benefit over g_slice_free1() now.
305 * glib/gslice.h: renamed g_slice_free_chain() to
306 g_slice_free_chain_with_offset(). implemented g_slice_free_chain() as
307 a type-safe macro as suggested in bug 323178.
308 simplified the macro implementation of g_slice_free() and implemented
309 it in a type safe manner for all compliers as suggested by Morten
310 Welinder <mortenw@gnome.org>.
315 * glib/glib.symbols: s/g_slice_free_chain/g_slice_free_chain_with_offset/
317 2005-12-05 Matthias Clasen <mclasen@redhat.com>
319 * glib/gasyncqueue.c: Add some docs.
321 * tests/libmoduletestplugin_a.c: Fix compiler warnings.
323 * glib/gatomic.c: In the ia64 implementation, use
324 __sync builtin without _si or _di suffix. (#321229,
325 Stanislav Brabec, patch by Andreas Schwab)
327 2005-12-04 Behdad Esfahbod <behdad@gnome.org>
329 * glib/gslice.h: Remove comma at the end of enum.
331 2005-12-04 Matthias Clasen <mclasen@redhat.com>
333 Handle multiple user names with the same UID better.
334 (#319535, Laszlo Peter)
336 * glib/gutils.c (g_get_any_init_do): When determining user
337 data, first look up $LOGNAME. If the UID doesn't match
338 getuid(), fall back to the current behaviour of looking
339 up the user data based on getuid().
341 2005-12-04 Matthias Clasen <mclasen@redhat.com>
343 * glib/gmacros.h (G_GNUC_WARN_UNUSED_RESULT): Add a macro
344 to make gcc warn if a function result is ignored. (#145466,
345 Arjan van de Ven, Alex Larsson)
347 * glib/gmem.h: Add the new attribute to g_realloc and
350 2005-12-03 Matthias Clasen <mclasen@redhat.com>
353 * glib/gcache.h: Deprecate g_cache_value_foreach. (#322956,
356 * glib/gmappedfile.c: Make mapping of empty files
359 * glib/gfileutils.c: Don't fork a new process just to
360 fix the permissions of the created temp file. (#321318,
361 Alexis S. L. Carvalho)
363 2005-12-02 Matthias Clasen <mclasen@redhat.com>
365 * README.in: Add a note about Solaris threads.
368 * configure.in: Remove support for Solaris threads.
369 (#136971, Sebastian Wilhelmi, patch by Andrew Paprocki)
371 2005-12-02 Matthias Clasen <mclasen@redhat.com>
373 * configure.in: Define G_GUINT64_CONSTANT in analogy to
374 G_GINT64_CONSTANT. (#322568, Andrew Paprocki)
376 2005-12-02 Matthias Clasen <mclasen@redhat.com>
378 * glib/gslice.c: Win32 portability fixes and C99ism removal,
379 pointed out by Kazuki Iwamoto. (#323052)
381 2005-12-02 Matthias Clasen <mclasen@redhat.com>
383 * glib/gatomic.c (g_atomic_int_add): Add a missing volatile for
384 the IA64 implementation.
386 Fri Dec 2 16:18:09 2005 Tim Janik <timj@imendio.com>
388 * glib/gslice.c: some naming and type size fixups.
390 Fri Dec 2 13:08:58 2005 Tim Janik <timj@gtk.org>
392 * glib/gslice.h (g_slice_free): reworked GCC-specific type-safe macro
393 variant into something less verbose, but digestible for gcc-3.4.
395 Fri Dec 2 10:55:07 2005 Tim Janik <timj@imendio.com>
397 * tests/slice-test.c: extended to perform the benchmarking on the old
398 memchunk code if 'O' is selected.
400 * tests/memchunks.c: new file which contains the old GLib mem chunks
401 implementation with prefix old_mem_chunk_.
403 * tests/Makefile.am: added memchunks.c
405 Fri Dec 2 00:16:59 2005 Tim Janik <timj@imendio.com>
407 * glib/gslice.c: improved gettimeofday() timer resolution by maintaining
408 acceptable load distribution of the syscall.
409 (allocator_get_magazine_threshold): reduce minimum/single-thread magazine
410 sizes to approximately page_size/2.5 to avoid excessive startup allocations.
412 2005-12-01 Matthias Clasen <mclasen@redhat.com>
414 * tests/gobject/ifacecheck.c (test_iface_base_init): Don't
415 double-free base interface members.
417 * tests/Makefile.am: Remove duplicate variable.
419 Thu Dec 1 17:32:46 2005 Tim Janik <timj@imendio.com>
421 * glib/gslice.[hc]: new slice allocator implementation.
423 * tests/slice-test.c: added random slice allocation test.
425 * glib/gthread.[hc]: removed newly added private thread mem API.
427 * glib/gthreadinit.h:
430 * glib/gmem.c: divided glib threading initialisation into three phases,
431 initialisation where private keys and messaging are not available (only
432 needed by gmem.c), initialisation without messaging but private keys
433 available (gslice.c, gmessage.c), and full fledged initialisers that
434 server the rest of glib. initialisation functions got renamed to reflect
435 the limitations of their corresponding phases.
437 * glib/gmem.c: removed memchunk code, defer allocations to
440 * glib/gmem.[hc]: removed g_slice_* skeletons.
442 * glib/glib.symbols: added g_slice_* symbols.
444 * configure.in: check for availability of posix_memalign(3), memalign(3)
447 * glib/Makefile.am: added gslice.[hc].
449 2005-12-01 Tor Lillqvist <tml@novell.com>
451 * glib/gstdio.c (g_stat): In the Win32 implementation, strip
452 trailing slash(es) for non-root folders. stat() fails if non-root
453 folders are specified with trailing slashes. It's too much hassle
454 to demand that callers strip such slashes themselves, especially
455 as it is easy to get it wrong and strip the slash of a root
457 (g_rename): On NT-based Windows, use MoveFileEx() with
458 MOVEFILE_REPLACE_EXISTING to better match Unix behaviour.
460 2005-11-28 Matthias Clasen <mclasen@redhat.com>
462 Fix G_STMT_START / G_STMT_END on Solaris. (#321972,
465 * configure.in: Check whether do { } while (0) works.
467 * glib/gmacros.h: Use do { } while (0) for G_STMT_START /
468 G_STMT_END if it works.
470 2005-11-28 Matthias Clasen <mclasen@redhat.com>
472 * glib/gthread.c (g_static_rw_lock_wait, g_static_rw_lock_signal):
473 * glib/gnode.c (g_node_depth_traverse_level):
474 * glib/gmem.c (g_allocator_new):
475 * glib/ghash.c (g_hash_table_unref, g_hash_table_destroy)
476 (g_hash_table_foreach_remove_or_steal): Silence compiler
479 2005-11-27 Matthias Clasen <mclasen@redhat.com>
481 * glib/gunicollate.c (g_utf8_collate_key_for_filename):
482 Don't read beyond len. (#322520, Christian Persch)
484 Wed Nov 23 17:34:01 2005 Tim Janik <timj@imendio.com>
486 * glib/gdataset.c: access datalist flags via atomic pointer access
487 functions, instead of acquiring the dataset lock. this is faster and
488 also matches the atomic pointer readouts in gdatalistprivate.h and
489 direct pointer modifications required by gobject.c.
491 Wed Nov 23 13:35:31 2005 Tim Janik <timj@gtk.org>
493 * glib/gdataset.c: streamlined initialization code somewhat,
494 removed GData node cache.
496 2005-11-22 Matthias Clasen <mclasen@redhat.com>
498 * glib/ghash.c (g_hash_table_ref):
499 (g_hash_table_unref): Mark these as new API.
501 Avoid double locking in g_intern_string (#322133,
504 * glib/gdataset.c (g_quark_from_string_internal): New
505 internal function which factors out common parts of
506 g_quark_from[_static]_string.
507 (g_quark_from_string, g_quark_from_static_string):
508 Use g_quark_from_string_internal.
509 (g_intern_string, g_intern_static_string): Use
510 g_quark_from_string_internal, and only take the
512 (g_quark_new): Don't store the strings shifted by -1
513 in the g_quarks array.
514 (g_quark_to_string): Adapt to the previous change.
516 Tue Nov 22 14:04:26 2005 Tim Janik <timj@imendio.com>
520 g_hash_table_new_full(): create hash tables with a ref count of 1.
521 g_hash_table_ref(): atomically ref_count+=1
522 g_hash_table_unref(): atomically ref_count-=1, destroys hash table
523 when refcount reaches 0.
524 g_hash_table_destroy(): just destroy keys and values, unref by 1.
525 g_hash_table_insert():
526 g_hash_table_replace(): assert ref_count>0.
529 * glib/gatomic.c: added 'volatile' qualifier to all atomic pointer and
532 2005-11-20 Behdad Esfahbod <behdad@gnome.org>
534 * glib/guniprop.c (g_unichar_get_mirror_char): Remove unused
535 variables. (#321984, Andrew Paprocki)
537 2005-11-18 Matthias Clasen <mclasen@redhat.com>
539 * configure.in: Bump version
541 2005-11-17 Matthias Clasen <mclasen@redhat.com>
543 * === Released 2.9.0 ===
545 * glib/glib.symbols: Updates
547 * glib/gutf8.c (g_utf8_offset_to_pointer): Add some performance
548 hints to the docs. (Paolo Borelli)
552 * glib/gthread.c (g_thread_foreach): Mark as new api.
554 * README.in: Updates.
558 * glib/gdate.c (g_date_set_time_t):
559 (g_date_set_time_val): Add functions to set a date from
560 a time_t and from a GTimeVal.
561 (g_date_set_time): Deprecate. (#314067, Roger Leigh)
564 * tests/utf8-pointer.c: Unit tests for g_utf8_pointer_to_offset
565 and g_utf8_offset_to_pointer.
567 * glib/gutf8.c (g_utf8_pointer_to_offset)
568 (g_utf8_offset_to_pointer): Handle negative offsets, and use
569 "stutter stepping" for going backwards. (#320638, Larry
574 * glib/gthread.c: const correctness fixes, found
575 by Arjan van de Ven and gcc.
577 2005-11-16 Behdad Esfahbod <behdad@gnome.org>
579 * docs/reference/glib/tmpl/unicode.sgml: Correct typo on mentioning
580 @G_UNICODE_BREAK_UNKNOWN that should be #G_UNICODE_BREAK_UNKNOWN.
582 2005-11-09 Behdad Esfahbod <behdad@gnome.org>
584 * glib/guniprop.c: Use bit hacks instead when checking a general
585 category value against multiple values.
587 * glib/gutf8.c: Change ISO10646 to Unicode in docs.
589 2005-11-10 Simos Xenitellis <simos@gnome.org>
591 * configure.in: Added tt (Tatar) to ALL_LINGUAS.
593 2005-11-05 Tor Lillqvist <tml@novell.com>
595 * glib/giowin32.c (g_io_win32_set_flags): Don't call g_warning().
596 (#320688) Instead set error and return failure.
598 2005-11-07 Matthias Clasen <mclasen@redhat.com>
600 * glib/gmain.c (g_main_dispatch): Don't call cb_funcs->unref
601 while holding the context lock. (#320886, Andy Wingo)
603 2005-11-07 Billy Biggs <vektor@dumbterm.net>
605 * tests/refcount/Makefile.am: Add a missing reference to libglib
606 in the LDADD for this test.
608 2005-11-07 Matthias Clasen <mclasen@redhat.com>
610 * glib/gmem.c (g_allocator_new): We need to set n_preallocs to a
611 nonzero value, otherwise GTK+ 2.8 breaks when compiled against
612 GLib 2.9. (#320755, Luca Ferretti)
614 2005-11-04 Matthias Clasen <mclasen@redhat.com>
616 * glib/Makefile.am: Apply a patch from Behdad Esfahbod to
617 use a faster lookup table for g_unichar_get_mirror_char().
618 * glib/gmirroringtable.h: The new table.
619 * glib/gunichartables.h: Remove bidi_mirroring_table.
620 * glib/gen-unicode-tables.pl: Don't generate the mirroring
622 * glib/glib-mirroring-tab/*: Sources for the program
623 which generated gmirroringtable.h.
625 * glib/glist.c: Avoid some code duplication.
627 * glib/gscanner.h: Include gdataset.h. (#320322)
629 * glib/gdebug.h: add new GDebugFlag for fatal_criticals
630 * glib/gmessages.c: (_g_debug_init): handle G_DEBUG=fatal_criticals,
631 to help find critical warnings in applications. (#320017,
634 2005-11-02 Tor Lillqvist <tml@novell.com>
636 * glib/glib.symbols: Remove large amount of trailing whitespace
637 from one line. Remove #ifdef G_UNDEFINED from around
638 g_hash_table_new and g_hash_table_new_full symbols, what was the
639 purpose of that? Add the g_slice functions.
641 Tue Nov 1 16:24:20 2005 Tim Janik <timj@imendio.com>
643 * glib/gmem.[hc]: prepared deprecation of GMemChunk and GAllocator.
644 added g_slice_*() API to allocate and cache small bits of memory.
645 an actuall allocator implementation for g_slice_*() is still pending.
647 * glib/gthread.[hc]: changes from a patch by Matthias Clasen.
648 changed GRealThread list to use in-structure *next; fields instead
649 of GSList, in order for thread iteration to not depenend on g_slice_*()
651 _g_thread_mem_private_get():
652 _g_thread_mem_private_set(): added accessors for private memory,
653 needed because the ordinary GPrivate implementation relies on GArray
654 and GSList and therefore indirectly on working g_slice_*() allocations.
657 g_thread_foreach(): new public API function to loop over all existing threads.
671 removed GAllocator and free list usages and accompanying locks.
672 use g_slice_*() API to allocate and cache small bits of memory.
674 * glib/ghook.h: removed GMemChunk field from public API.
677 * glib/glist.h: deprecate allocator API, provide _free1() for consistency.
679 * glib/gnode.h: deprecate allocator API.
681 * glib/gmain.c: reordered GPollRec fields so g_slice_free_chain() can
682 be used for poll rec lists.
684 * glib/grel.c: removed mem chunk usage, and allocated tuples via g_slice_*().
685 g_relation_destroy(): free all tuples from the all_tuples hash table,
686 this effectively maintains the life time track keeping of tuples.
687 g_relation_delete_tuple(): free tuples which are removed from the
688 all_tuples hash table. this fixes a temporary leak that was present
689 in the memchunk code until the destruction of the relation.
691 2005-10-29 Matthias Clasen <mclasen@redhat.com>
693 * tests/convert-test.c: Add some tests for conversions between
694 UTF-8, UCS-4 and UTF-16.
696 * glib/gutf8.c (g_utf8_to_ucs4, g_utf8_to_utf16): Fix handling
697 of len == -1, noticed by Morten Welinder.
699 2005-10-27 Erdal Ronahi <erdal.ronahi@gmail.com>
701 * configure.in: Added ku (Kurdish) to ALL_LINGUAS
703 2005-10-26 Matthias Clasen <mclasen@redhat.com>
705 * glib/gutf8.c (g_ucs4_to_utf8): Don't set items_read twice
706 in the error case, and add some documentation. (#319806, Morten
709 2005-10-19 Manish Singh <yosh@gimp.org>
711 * configure.in: Use AC_CHECK_FUNCS for _NSGetEnviron, to get the
712 config.h symbol automatically. Fixes bug #313731.
714 2005-10-19 Tor Lillqvist <tml@novell.com>
716 * glib/gwin32.c (g_win32_get_package_installation_directory):
717 Return a g_strdup()ed copy of the value stored in the hash table,
718 so that it can be g_free()d without leaving a dangling pointer in
719 the hash table. (#319232)
721 2005-10-06 Matthias Clasen <mclasen@redhat.com>
723 * glib/gunicollate.c (g_utf8_collate_key_for_filename): Handle
724 all-zero sequences correctly. (#317930, Sebastien Bacher)
726 2005-10-05 Matthias Clasen <mclasen@redhat.com>
728 * glib/gmarkup.c (g_markup_escape_text): Doc addition
730 * Makefile.am: Add Collation tests.
732 * tests/collate/*: Inputs and expected outputs for collation tests.
734 * tests/run-collate-tests.sh: Script to run collation tests.
736 * tests/unicode-collate.c (main): Rework slightly to make
737 it usable in unit tests. Also test g_utf8_collate_key_for_filename().
739 2005-10-01 Behdad Esfahbod <behdad@gnome.org>
741 * docs/reference/glib/tmpl/unicode.sgml:
742 * glib/gen-unicode-tables.pl:
744 * glib/gunichartables.h:
746 * tests/casefold.txt:
747 * tests/casemap.txt: Updated to Unicode 4.1. There are five new
748 GUnicodeBreakType types. That may break some applications, like
751 2005-09-26 Matthias Clasen <mclasen@redhat.com>
753 * glib/gstrfuncs.c (g_ascii_strtoull): Add details to the
754 docs. (#314393, Matthew F. Barnes)
757 * glib/gprintf.h: Remove g_snprintf() and g_vsnprintf(), since
758 they are already declared in glib.h. This doesn't break documented
759 use of gprintf.h, but should probably be pointed out in the
760 release notes for 2.10. (#314232, Behdad Esfahbod)
762 Tue Sep 20 13:16:04 2005 Tim Janik <timj@imendio.com>
764 * glib/gpattern.c (g_pattern_ph_match): applied significant recursion
765 complexity optimization, based on a patch from Matthias Clasen.
767 * tests/patterntest.c: more tests, mostly from matthias.
769 2005-09-20 Matthias Clasen <mclasen@redhat.com>
771 * glib/gqueue.c (g_queue_insert_sorted): Correct the docs.
772 (#316703, Mark Drago)
774 Mon Sep 19 17:23:23 2005 Tim Janik <timj@imendio.com>
776 * glib/gpattern.c: applied a patch from matthias which checks on the
777 upper bound of GPatternSpec length to optimize matches.
780 * tests/patterntest.c: added more match cases.
782 2005-09-16 Tor Lillqvist <tml@novell.com>
784 * glib/gstrfuncs.c (g_ascii_strcasecmp, g_ascii_strncasecmp): Add
785 warning to doc comment that these functions should not be used on
786 encodings like CP932.
788 2005-09-14 Matthias Clasen <mclasen@redhat.com>
790 * tests/keyfile-test.c: Add a test for grup names of length 1.
792 * glib/gkeyfile.c (g_key_file_line_is_group): Accept group names
793 of length 1. (#316309)
795 2005-09-12 Matthias Clasen <mclasen@redhat.com>
797 * glib/gmarkup.c (g_markup_escape_text): Clarify docs.
799 2005-09-11 Sebastian Wilhelmi <seppi@seppi.de>
801 * tests/refcount/Makefile.am (INCLUDES): Link the the refcount
802 tests to the system thread library $(G_THREAD_LIBS). Fixes #313744
805 2005-09-11 Kjartan Maraas <kmaraas@gnome.org>
807 * glib/gmain.c: (g_child_watch_prepare), (g_child_watch_check),
808 (child_watch_helper_thread): Remove some dead code. Closes
811 2005-09-07 Tor Lillqvist <tml@novell.com>
813 * glib/Makefile.am: Create also a console version of the
814 gspawn-win32-helper program, gspawn-win32-helper-console.exe.
815 It's otherwise identical to gspawn-win32-helper.exe, except marked
816 as a console application (linked without the -mwindows option).
818 * glib/gspawn-win32.c (do_spawn_directly, do_spawn_with_pipes):
819 Drop the dont_wait parameter. Its truth value correlated 100% with
820 the NULLness of the exit_status parameter anyway, so it's enough
821 to check whether exit_status is NULL. Invert the sense of the
822 dont_return_handle parameter and rename it to do_return_handle, to
823 make the code easier to read by avoiding double negations.
825 (g_spawn_sync_utf8, g_spawn_async_with_pipes_utf8): Modify calls
826 to do_spawn_with_pipes() accordingly.
828 (do_spawn_with_pipes): If we have a console, use the console
829 version of the helper program, otherwise use the GUI one. This
830 avoids extra console windows opening up in some situations. (In
831 case a console application uses the GUI gspawn-win32-helper.exe to
832 spawn another console application we would get a separate console
833 for the spawned console application).
835 * glib-zip.in: Distribute also gspawn-win32-helper-console.exe.
837 2005-09-05 Matthias Clasen <mclasen@redhat.com>
839 * glib/gmappedfile.c (g_mapped_file_new): Report an error
840 if the file is too large. (#315275, Kjartan Maraas)
842 * glib/gkeyfile.c (g_key_file_load_from_fd): The return value
843 of read() is signed. (#315273, Kjartan Maraas)
845 2005-08-31 Tor Lillqvist <tml@novell.com>
847 * glib/gutils.h: Wrapping atexit() is a bad idea on Windows, where
848 the EXE and each DLL have their own atexit function chains.
850 #define g_atexit as atexit instead. This means it has a
851 better chance of doing what the caller wants. For instance,
852 gtkhtml calls g_atexit() registering a function in gtkhtml
853 itself. This caused a crash when g_atexit() was implemented as a
854 function in the GLib DLL. The gtkhtml DLL was already unloaded by
855 the time the GLib DLL got unloaded.
857 * glib/gutils.c: #undef the #define mentioned above, to also get a
858 real g_atexit() into the DLL for backward compatibility. Document
859 the Windows behaviour of g_atexit(), and document the varying ways
860 atexit() can behave in the context of dynamically loaded modules
863 2005-08-31 Matthias Clasen <mclasen@redhat.com>
867 * glib/gdataset.c: Add string interning functions.
869 2005-08-28 Matthias Clasen <mclasen@redhat.com>
871 * glib/giochannel.c: Unify some near-duplicate strings. (#314654,
874 2005-08-26 Matthias Clasen <mclasen@redhat.com>
876 * configure.in: Bump version to 2.9.0
878 2005-08-25 Tor Lillqvist <tml@novell.com>
880 Make also the g_spawn*() functions take parameters in the GLib
881 file name encoding, i.e. UTF-8, on Windows. Has no impact on Unix
882 API or ABI. Like the other GLib API that was earlier changed to
883 use UTF-8 on Windows, the names of the functions that take UTF-8
884 have _utf8 suffixes added by using preprocessor macros in the
885 header file. The old names are kept for functions with the old
886 behaviour, taking parameters in the system codepage, for DLL ABI
889 * glib/gspawn.h: On Win32 add the suffix _utf8 to the names of the
890 g_spawn*() functions.
892 * glib/gspawn-win32.c: Use wide-char API on NT-based
893 Windows. Convert parameters from UTF-8 to wide chars (NT) or
894 system codepage (Win9x) and call the C library _wspawn*() or
895 spawn*() functions respectvely. Add DLL ABI stability versions
896 that take parameters in the system codepage.
898 * glib/gspawn-win32-helper.c: On NT-based Windows use the
899 wide-char versions of argv and envp, and use wide-char API to
900 change directory and spawn the program to run. Remove the verbose
901 debugging output, it was too complex to modify for the wide-char
902 features. (Just add temporary debugging printouts if needed, no
903 need to have them permanently in the source.)
905 * glib/gspawn.c: Corresponding documentation updates.
907 * glib/glib.symbols: Corresponding changes: Mark the ABI stability
908 symbols as PRIVATE, add the new _utf8-suffixed ones.
910 2005-08-24 Stepan Kasal <kasal@ucw.cz>
912 * glib/gtypes.h (G_MININT64): Cast the constant to gint64; it is
913 guint64 otherwise and that can produce warnings about comparison
914 between signed and unsigned.
916 2005-08-23 Matthias Clasen <mclasen@redhat.com>
918 * glib/gutils.c: Fix the crt_externs.h include.
920 2005-08-23 Stepan Kasal <kasal@ucw.cz>
922 * NEWS: Fix spelling of my first name.
924 2005-08-23 Matthias Clasen <mclasen@redhat.com>
928 * === Released 2.8.1 ===
932 2005-08-20 Hans Breuer <hans@breuer.org>
934 * glib/makefile.msc.in : link with ws2_32.lib
936 2005-08-18 Tor Lillqvist <tml@novell.com>
938 * configure.in: Check for <sys/wait.h>
940 * glib/gbacktrace.c: Include <sys/wait.h> on if HAVE_SYS_WAIT_H.
942 2005-08-18 Ross Burton <ross@burtonini.com>
945 Optimise single-character insertions.
950 * tests/string-test.c:
951 Add tests for new optimisation, and fix a leak.
953 2005-08-17 Matthias Clasen <mclasen@redhat.com>
955 * configure.in: Check for crt_externs.h and _NSGetEnviron.
957 * glib/gutils.c: On Darwin, include crt-externs.h and
958 define environ using _NSGetEnviron(). (#313731)
960 2005-08-16 Stepan Kasal <kasal@ucw.cz>
962 * glib/gutils.c (g_get_any_init): Move the body of the big if...
963 (g_get_any_init_do): ... to this new function.
964 (g_get_any_init): Declare as inline.
965 (g_get_any_init_locked): New inline function, does the locking.
966 Make use of these two throughout the code.
968 2005-08-15 Matthias Clasen <mclasen@redhat.com>
970 * glib/gbacktrace.c (g_on_error_stack_trace): Wait for
971 the child process and then simply return. This makes
972 The "S" option work as documented in g_on_error_query().
973 (#313125, Matthew F. Barnes)
975 * glib/gunicode.h: Update the link to Unicode category
976 values. (#313369, Behnam Esfahbod)
978 * glib/gqueue.c (g_queue_find_custom): Clarify docs
979 a little. (#311727, Tristan van Berkom)
981 * glib/abicheck.sh, gobject/abicheck.sh: Make the
982 check work on ia64 too, where some symbols ended up
983 in yet another section.