1 2006-01-05 Matthias Clasen <mclasen@redhat.com>
3 * === Released 2.9.2 ===
5 2006-01-05 Hans Breuer <hans@breuer.org>
7 * glib/makefile.msc.in : added gslice.obj
9 2006-01-05 Matthias Clasen <mclasen@redhat.com>
13 * glib/gmain.c: Clarify the documentation of source ids
14 a bit. (#325874, Dan Williams)
16 * configure.in: Fix another strict aliasing problem.
18 * glib/gthreadpool.c: Avoid use of varargs macro. (#325864,
21 2006-01-04 Manish Singh <yosh@gimp.org>
24 * glib/gstdio.h: don't macroized g_access, g_chdir, and g_unlink
25 either, since they have the same issue as g_rmdir. (Related to
28 Wed Jan 4 13:33:25 2006 Tim Janik <timj@gtk.org>
30 * glib/gslice.c (magazine_cache_pop_magazine): don't reverse chunk
31 order when creating magazines, so we hand out chunks with ascending
34 2006-01-04 Abel Cheung <maddog@linuxhall.org>
36 * configure.in: Added "ml" "zh_HK" to ALL_LINGUAS.
38 2006-01-03 Martyn Russell <martyn@imendio.com>
40 * docs/reference/glib/glib-sections.txt:
42 * glib/gthreadpool.[ch]:
43 - Added new API g_thread_pool_get_idle_time() and
44 g_thread_pool_set_idle_time(). (#324228).
46 * tests/threadpool-test.c:
47 - Updated test case to do thread pool sorting, thread pool with
48 no sorting and a thread pool with idle thread timeouts.
50 2006-01-03 Matthias Clasen <mclasen@redhat.com>
52 * glib/gmain.h: Add new functions here, too.
54 * glib/glib.symbols: Add new functions.
56 2005-12-20 Michael Meeks <michael.meeks@novell.com>
58 * glib/gmain.c (g_main_context_is_owner): new method
59 to determine if the current thread is the owner of the
62 2006-01-02 Matthias Clasen <mclasen@redhat.com>
66 * glib/gstdio.c (g_rmdir): Don't provide g_rmdir() as a macro
67 expanding to rmdir, since rmdir is not declared in a portable
68 system header we can include in gstdio.h. (#325249, Jani Monoses)
70 2006-01-01 Tor Lillqvist <tml@novell.com>
72 * glib/gspawn-win32.c (g_spawn_sync_utf8): Set the GIOChannels for
73 stdout and stderr to unbuffered. Otherwise the giochannel layer
74 will try to read from them regardless whether the
75 g_io_channel_win32_poll() call here has indicated
76 readability or not. (#325310)
78 2005-12-29 Matthias Clasen <mclasen@redhat.com>
80 * glib/gutils.c (glib_check_version): Fix a copy-and-paste error
81 in the docs. (#325273, Declan Naughton)
83 2005-12-27 Manish Singh <yosh@gimp.org>
85 * tests/run-collate-tests.sh: use LC_ALL instead of LC_COLLATE, to
86 make sure we really override things.
88 2005-12-27 Matthias Clasen <mclasen@redhat.com>
90 Fix #316221, Michal Benes, Stanislav Brabec;
92 * configure.in: Fix a strict aliasing problem in
93 g_static_mutex_get_mutex().
94 * glib/gthread.h: ...and in
95 g_static_mutex_get_mutex_impl_shortcut().
97 * glib/gdatasetprivate.h: Add a cast to silence compiler
98 warnings. (#321978, Andrew Paprocki)
100 Partial fix for bug #323937, Bogdan Nicula.
102 * configure.in: Check for malloc.h
103 * glib/gslice.c: Don't include malloc.h unconditionally.
105 2005-12-27 Manish Singh <yosh@gimp.org>
107 * tests/run-collate-tests.sh: export LC_COLLATE so it takes.
108 (#324950, Dan Yefimov)
110 2005-12-26 Matthias Clasen <mclasen@redhat.com>
112 * glib/gslice.c: On Win32, include process.h (#325015, Kazuki
115 2005-12-25 Matthias Clasen <mclasen@redhat.com>
117 * AUTHORS: Update my email
119 * tests/slice-test.c: Fix C99isms. (#324950, Dan Yefimov)
121 2005-12-21 Matthias Clasen <mclasen@redhat.com>
123 * README.in: Add some notes on when 'make check' may fail.
125 2005-12-20 Matthias Clasen <mclasen@redhat.com>
128 * glib/gthreadpool.h:
129 * glib/gthreadpool.c (g_thread_pool_set_sort_function): New function
130 to sort tasks pushed into a threadpool. (#324479, Martyn Russell)
132 * tests/threadpool-test.c: Test this.
134 Tue Dec 20 18:14:14 2005 Tim Janik <timj@imendio.com>
136 * glib/gslice.[hc]: added mem_error() and mem_assert() to test and
137 handle errors without depending on gmessage.c which might not be
138 setup when the error occours.
139 removed G_SLICE_CONFIG_ALWAYS_FREE config option, fixed the code so
140 always freeing can be achieved by adjusting the working set time to
141 0 with G_SLICE_CONFIG_WORKING_SET_MSECS.
142 added G_SLICE_CONFIG_COLOR_INCREMENT to test different color increments
143 (mainly 0 and 1). reduced the minimum block size to 128 bytes, to
144 minimize wastage if small amounts of differently sized structrues are
145 allocated, this does come at a performance cost of roughly 5% though.
146 fixed up block alignment calculation, so it works for varying
147 block sizes. only use strerror() not g_strerror() since the latter
148 depends on working GQuark and GSlice.
149 mem_error(): implemented in terms of fprintf and vfprintf.
151 * tests/slice-color.c: new program to test cache colorization effects.
153 * tests/slice-test.c: trade G_SLICE_CONFIG_ALWAYS_FREE for 0 duration
154 G_SLICE_CONFIG_WORKING_SET_MSECS.
156 2005-12-17 Matthias Clasen <mclasen@redhat.com>
158 * glib/goption.c (parse_short_option): Set an error in all
159 failure cases. (#324332, Tim-Philipp Müller)
161 2005-12-17 Sebastian Wilhelmi <seppi@seppi.de>
163 * glib/gatomic.c: Fix memory barrier position in g_atomic_int_get
164 and g_atomic_pointer_get. Add g_atomic_int_set and
165 g_atomic_pointer_set implementations for the !DEFINE_WITH_MUTEXES &&
166 G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case, as well as defining them
167 as functions (additionally to the macros in the header) for the
168 !G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case.
170 2005-12-16 Matthias Clasen <mclasen@redhat.com>
172 * glib/gmem.c (g_allocator_new): Don't return a pointer to
173 a const struct, since apps expect to be able to modify it.
174 (#324179, J. Ali Harlow)
176 Tue Dec 13 10:13:32 2005 Tim Janik <timj@imendio.com>
178 * glib/gatomic.h: added g_atomic_pointer_set() and g_atomic_int_set()
179 for systems where the initialization of atomic variables requires a
180 write memory barrier.
182 2005-12-09 Matthias Clasen <mclasen@redhat.com>
184 * configure.in: Bump version
186 * === Released 2.9.1 ===
188 2005-12-09 Alexander Larsson <alexl@redhat.com>
192 Use G_GNUC_WARN_UNUSED_RESULT on list functions that return
195 2005-12-08 Matthias Clasen <mclasen@redhat.com>
199 2005-12-07 Martyn Russell <martyn@imendio.com>
201 * glib/gasyncqueue.c:
202 - Call g_queue_insert_sorted() instead of duplicating the code.
203 - Call g_queue_sort() instead of duplicating the code.
204 - Invert sort function results to make sure the same sort function
205 gives the same results across glist, gslist, gqueue and
208 * tests/asyncqueue-test.c:
209 - Updated the sort function to reflect the example in the
210 documentation for gasyncqueue.c.
212 2005-12-07 Martyn Russell <martyn@imendio.com>
214 * docs/reference/glib/glib-sections.txt:
215 * docs/reference/glib/tmpl/linked_lists_double.sgml:
216 * docs/reference/glib/tmpl/linked_lists_single.sgml:
219 - Added g_list_insert_sorted_with_data () and
220 g_slist_insert_sorted_with_data ().
221 - Removed the extra check in g_list_sort() and g_slist_sort() for
222 GCompareDataFunc vs. GCompareFunc.
224 2005-12-07 Tor Lillqvist <tml@novell.com>
226 * glib-gettextize.in: Look up prefix at run-time on Win32,
227 assuming the standard directory structure with glib-gettextize in
230 2005-12-06 Matthias Clasen <mclasen@redhat.com>
232 * glib/gmem.h: Exempt GTK+ from the mem chunk deprecation, since
233 we need GTK+ 2.8 to compile against GLib 2.10.
235 2005-12-05 Manish Singh <yosh@gimp.org>
237 * tests/run-collate-test.sh: set LC_COLLATE, not LANG, to be sure
238 to override any user settings.
240 2005-12-05 Matthias Clasen <mclasen@redhat.com>
242 * configure.in: Revert an accidential commit
244 2005-12-05 Matthias Clasen <mclasen@redhat.com>
246 * glib/gutf8.c: Documentation updates. (#323291, Morten Welinder)
248 * tests/Makefile.am (TESTS_ENVIRONMENT):
249 * tests/gobject/Makefile.am (TESTS_ENVIRONMENT): Set
250 MALLOC_CHECK_ and MALLOC_PERTURB_
252 * tests/run-collate-tests.sh: Run the collation tests explicitly
253 in en_US locale. (#320463)
255 * glib/gmem.h: Really deprecate GMemChunk.
262 * glib/gstrfuncs.c: Add versioned deprecation docs.
264 2005-12-05 Martyn Russell <martyn@imendio.com>
266 * docs/reference/glib/glib-sections.txt:
267 * glib/gasyncqueue.[ch]:
268 - Added support for sorting async queues by with _push_sorted(),
269 _push_sorted_unlocked(), _sort() and _sort_unlocked() (#323047).
272 * tests/asyncqueue-test.c:
273 - Added test case for gasyncqueue.c
275 Mon Dec 5 15:53:20 2005 Tim Janik <timj@imendio.com>
277 * glib/gslice.c: implement chain walking for arbitrary ->next pointer
278 offsets in g_slice_free_chain_with_offset() based on a patch by behdad
279 in bug 323178. moved time consuming logic from g_slice_free() out of
280 the inner loop, so g_slice_free_chain_with_offset() provides a real
281 performance benefit over g_slice_free1() now.
283 * glib/gslice.h: renamed g_slice_free_chain() to
284 g_slice_free_chain_with_offset(). implemented g_slice_free_chain() as
285 a type-safe macro as suggested in bug 323178.
286 simplified the macro implementation of g_slice_free() and implemented
287 it in a type safe manner for all compliers as suggested by Morten
288 Welinder <mortenw@gnome.org>.
293 * glib/glib.symbols: s/g_slice_free_chain/g_slice_free_chain_with_offset/
295 2005-12-05 Matthias Clasen <mclasen@redhat.com>
297 * glib/gasyncqueue.c: Add some docs.
299 * tests/libmoduletestplugin_a.c: Fix compiler warnings.
301 * glib/gatomic.c: In the ia64 implementation, use
302 __sync builtin without _si or _di suffix. (#321229,
303 Stanislav Brabec, patch by Andreas Schwab)
305 2005-12-04 Behdad Esfahbod <behdad@gnome.org>
307 * glib/gslice.h: Remove comma at the end of enum.
309 2005-12-04 Matthias Clasen <mclasen@redhat.com>
311 Handle multiple user names with the same UID better.
312 (#319535, Laszlo Peter)
314 * glib/gutils.c (g_get_any_init_do): When determining user
315 data, first look up $LOGNAME. If the UID doesn't match
316 getuid(), fall back to the current behaviour of looking
317 up the user data based on getuid().
319 2005-12-04 Matthias Clasen <mclasen@redhat.com>
321 * glib/gmacros.h (G_GNUC_WARN_UNUSED_RESULT): Add a macro
322 to make gcc warn if a function result is ignored. (#145466,
323 Arjan van de Ven, Alex Larsson)
325 * glib/gmem.h: Add the new attribute to g_realloc and
328 2005-12-03 Matthias Clasen <mclasen@redhat.com>
331 * glib/gcache.h: Deprecate g_cache_value_foreach. (#322956,
334 * glib/gmappedfile.c: Make mapping of empty files
337 * glib/gfileutils.c: Don't fork a new process just to
338 fix the permissions of the created temp file. (#321318,
339 Alexis S. L. Carvalho)
341 2005-12-02 Matthias Clasen <mclasen@redhat.com>
343 * README.in: Add a note about Solaris threads.
346 * configure.in: Remove support for Solaris threads.
347 (#136971, Sebastian Wilhelmi, patch by Andrew Paprocki)
349 2005-12-02 Matthias Clasen <mclasen@redhat.com>
351 * configure.in: Define G_GUINT64_CONSTANT in analogy to
352 G_GINT64_CONSTANT. (#322568, Andrew Paprocki)
354 2005-12-02 Matthias Clasen <mclasen@redhat.com>
356 * glib/gslice.c: Win32 portability fixes and C99ism removal,
357 pointed out by Kazuki Iwamoto. (#323052)
359 2005-12-02 Matthias Clasen <mclasen@redhat.com>
361 * glib/gatomic.c (g_atomic_int_add): Add a missing volatile for
362 the IA64 implementation.
364 Fri Dec 2 16:18:09 2005 Tim Janik <timj@imendio.com>
366 * glib/gslice.c: some naming and type size fixups.
368 Fri Dec 2 13:08:58 2005 Tim Janik <timj@gtk.org>
370 * glib/gslice.h (g_slice_free): reworked GCC-specific type-safe macro
371 variant into something less verbose, but digestible for gcc-3.4.
373 Fri Dec 2 10:55:07 2005 Tim Janik <timj@imendio.com>
375 * tests/slice-test.c: extended to perform the benchmarking on the old
376 memchunk code if 'O' is selected.
378 * tests/memchunks.c: new file which contains the old GLib mem chunks
379 implementation with prefix old_mem_chunk_.
381 * tests/Makefile.am: added memchunks.c
383 Fri Dec 2 00:16:59 2005 Tim Janik <timj@imendio.com>
385 * glib/gslice.c: improved gettimeofday() timer resolution by maintaining
386 acceptable load distribution of the syscall.
387 (allocator_get_magazine_threshold): reduce minimum/single-thread magazine
388 sizes to approximately page_size/2.5 to avoid excessive startup allocations.
390 2005-12-01 Matthias Clasen <mclasen@redhat.com>
392 * tests/gobject/ifacecheck.c (test_iface_base_init): Don't
393 double-free base interface members.
395 * tests/Makefile.am: Remove duplicate variable.
397 Thu Dec 1 17:32:46 2005 Tim Janik <timj@imendio.com>
399 * glib/gslice.[hc]: new slice allocator implementation.
401 * tests/slice-test.c: added random slice allocation test.
403 * glib/gthread.[hc]: removed newly added private thread mem API.
405 * glib/gthreadinit.h:
408 * glib/gmem.c: divided glib threading initialisation into three phases,
409 initialisation where private keys and messaging are not available (only
410 needed by gmem.c), initialisation without messaging but private keys
411 available (gslice.c, gmessage.c), and full fledged initialisers that
412 server the rest of glib. initialisation functions got renamed to reflect
413 the limitations of their corresponding phases.
415 * glib/gmem.c: removed memchunk code, defer allocations to
418 * glib/gmem.[hc]: removed g_slice_* skeletons.
420 * glib/glib.symbols: added g_slice_* symbols.
422 * configure.in: check for availability of posix_memalign(3), memalign(3)
425 * glib/Makefile.am: added gslice.[hc].
427 2005-12-01 Tor Lillqvist <tml@novell.com>
429 * glib/gstdio.c (g_stat): In the Win32 implementation, strip
430 trailing slash(es) for non-root folders. stat() fails if non-root
431 folders are specified with trailing slashes. It's too much hassle
432 to demand that callers strip such slashes themselves, especially
433 as it is easy to get it wrong and strip the slash of a root
435 (g_rename): On NT-based Windows, use MoveFileEx() with
436 MOVEFILE_REPLACE_EXISTING to better match Unix behaviour.
438 2005-11-28 Matthias Clasen <mclasen@redhat.com>
440 Fix G_STMT_START / G_STMT_END on Solaris. (#321972,
443 * configure.in: Check whether do { } while (0) works.
445 * glib/gmacros.h: Use do { } while (0) for G_STMT_START /
446 G_STMT_END if it works.
448 2005-11-28 Matthias Clasen <mclasen@redhat.com>
450 * glib/gthread.c (g_static_rw_lock_wait, g_static_rw_lock_signal):
451 * glib/gnode.c (g_node_depth_traverse_level):
452 * glib/gmem.c (g_allocator_new):
453 * glib/ghash.c (g_hash_table_unref, g_hash_table_destroy)
454 (g_hash_table_foreach_remove_or_steal): Silence compiler
457 2005-11-27 Matthias Clasen <mclasen@redhat.com>
459 * glib/gunicollate.c (g_utf8_collate_key_for_filename):
460 Don't read beyond len. (#322520, Christian Persch)
462 Wed Nov 23 17:34:01 2005 Tim Janik <timj@imendio.com>
464 * glib/gdataset.c: access datalist flags via atomic pointer access
465 functions, instead of acquiring the dataset lock. this is faster and
466 also matches the atomic pointer readouts in gdatalistprivate.h and
467 direct pointer modifications required by gobject.c.
469 Wed Nov 23 13:35:31 2005 Tim Janik <timj@gtk.org>
471 * glib/gdataset.c: streamlined initialization code somewhat,
472 removed GData node cache.
474 2005-11-22 Matthias Clasen <mclasen@redhat.com>
476 * glib/ghash.c (g_hash_table_ref):
477 (g_hash_table_unref): Mark these as new API.
479 Avoid double locking in g_intern_string (#322133,
482 * glib/gdataset.c (g_quark_from_string_internal): New
483 internal function which factors out common parts of
484 g_quark_from[_static]_string.
485 (g_quark_from_string, g_quark_from_static_string):
486 Use g_quark_from_string_internal.
487 (g_intern_string, g_intern_static_string): Use
488 g_quark_from_string_internal, and only take the
490 (g_quark_new): Don't store the strings shifted by -1
491 in the g_quarks array.
492 (g_quark_to_string): Adapt to the previous change.
494 Tue Nov 22 14:04:26 2005 Tim Janik <timj@imendio.com>
498 g_hash_table_new_full(): create hash tables with a ref count of 1.
499 g_hash_table_ref(): atomically ref_count+=1
500 g_hash_table_unref(): atomically ref_count-=1, destroys hash table
501 when refcount reaches 0.
502 g_hash_table_destroy(): just destroy keys and values, unref by 1.
503 g_hash_table_insert():
504 g_hash_table_replace(): assert ref_count>0.
507 * glib/gatomic.c: added 'volatile' qualifier to all atomic pointer and
510 2005-11-20 Behdad Esfahbod <behdad@gnome.org>
512 * glib/guniprop.c (g_unichar_get_mirror_char): Remove unused
513 variables. (#321984, Andrew Paprocki)
515 2005-11-18 Matthias Clasen <mclasen@redhat.com>
517 * configure.in: Bump version
519 2005-11-17 Matthias Clasen <mclasen@redhat.com>
521 * === Released 2.9.0 ===
523 * glib/glib.symbols: Updates
525 * glib/gutf8.c (g_utf8_offset_to_pointer): Add some performance
526 hints to the docs. (Paolo Borelli)
530 * glib/gthread.c (g_thread_foreach): Mark as new api.
532 * README.in: Updates.
536 * glib/gdate.c (g_date_set_time_t):
537 (g_date_set_time_val): Add functions to set a date from
538 a time_t and from a GTimeVal.
539 (g_date_set_time): Deprecate. (#314067, Roger Leigh)
542 * tests/utf8-pointer.c: Unit tests for g_utf8_pointer_to_offset
543 and g_utf8_offset_to_pointer.
545 * glib/gutf8.c (g_utf8_pointer_to_offset)
546 (g_utf8_offset_to_pointer): Handle negative offsets, and use
547 "stutter stepping" for going backwards. (#320638, Larry
552 * glib/gthread.c: const correctness fixes, found
553 by Arjan van de Ven and gcc.
555 2005-11-16 Behdad Esfahbod <behdad@gnome.org>
557 * docs/reference/glib/tmpl/unicode.sgml: Correct typo on mentioning
558 @G_UNICODE_BREAK_UNKNOWN that should be #G_UNICODE_BREAK_UNKNOWN.
560 2005-11-09 Behdad Esfahbod <behdad@gnome.org>
562 * glib/guniprop.c: Use bit hacks instead when checking a general
563 category value against multiple values.
565 * glib/gutf8.c: Change ISO10646 to Unicode in docs.
567 2005-11-10 Simos Xenitellis <simos@gnome.org>
569 * configure.in: Added tt (Tatar) to ALL_LINGUAS.
571 2005-11-05 Tor Lillqvist <tml@novell.com>
573 * glib/giowin32.c (g_io_win32_set_flags): Don't call g_warning().
574 (#320688) Instead set error and return failure.
576 2005-11-07 Matthias Clasen <mclasen@redhat.com>
578 * glib/gmain.c (g_main_dispatch): Don't call cb_funcs->unref
579 while holding the context lock. (#320886, Andy Wingo)
581 2005-11-07 Billy Biggs <vektor@dumbterm.net>
583 * tests/refcount/Makefile.am: Add a missing reference to libglib
584 in the LDADD for this test.
586 2005-11-07 Matthias Clasen <mclasen@redhat.com>
588 * glib/gmem.c (g_allocator_new): We need to set n_preallocs to a
589 nonzero value, otherwise GTK+ 2.8 breaks when compiled against
590 GLib 2.9. (#320755, Luca Ferretti)
592 2005-11-04 Matthias Clasen <mclasen@redhat.com>
594 * glib/Makefile.am: Apply a patch from Behdad Esfahbod to
595 use a faster lookup table for g_unichar_get_mirror_char().
596 * glib/gmirroringtable.h: The new table.
597 * glib/gunichartables.h: Remove bidi_mirroring_table.
598 * glib/gen-unicode-tables.pl: Don't generate the mirroring
600 * glib/glib-mirroring-tab/*: Sources for the program
601 which generated gmirroringtable.h.
603 * glib/glist.c: Avoid some code duplication.
605 * glib/gscanner.h: Include gdataset.h. (#320322)
607 * glib/gdebug.h: add new GDebugFlag for fatal_criticals
608 * glib/gmessages.c: (_g_debug_init): handle G_DEBUG=fatal_criticals,
609 to help find critical warnings in applications. (#320017,
612 2005-11-02 Tor Lillqvist <tml@novell.com>
614 * glib/glib.symbols: Remove large amount of trailing whitespace
615 from one line. Remove #ifdef G_UNDEFINED from around
616 g_hash_table_new and g_hash_table_new_full symbols, what was the
617 purpose of that? Add the g_slice functions.
619 Tue Nov 1 16:24:20 2005 Tim Janik <timj@imendio.com>
621 * glib/gmem.[hc]: prepared deprecation of GMemChunk and GAllocator.
622 added g_slice_*() API to allocate and cache small bits of memory.
623 an actuall allocator implementation for g_slice_*() is still pending.
625 * glib/gthread.[hc]: changes from a patch by Matthias Clasen.
626 changed GRealThread list to use in-structure *next; fields instead
627 of GSList, in order for thread iteration to not depenend on g_slice_*()
629 _g_thread_mem_private_get():
630 _g_thread_mem_private_set(): added accessors for private memory,
631 needed because the ordinary GPrivate implementation relies on GArray
632 and GSList and therefore indirectly on working g_slice_*() allocations.
635 g_thread_foreach(): new public API function to loop over all existing threads.
649 removed GAllocator and free list usages and accompanying locks.
650 use g_slice_*() API to allocate and cache small bits of memory.
652 * glib/ghook.h: removed GMemChunk field from public API.
655 * glib/glist.h: deprecate allocator API, provide _free1() for consistency.
657 * glib/gnode.h: deprecate allocator API.
659 * glib/gmain.c: reordered GPollRec fields so g_slice_free_chain() can
660 be used for poll rec lists.
662 * glib/grel.c: removed mem chunk usage, and allocated tuples via g_slice_*().
663 g_relation_destroy(): free all tuples from the all_tuples hash table,
664 this effectively maintains the life time track keeping of tuples.
665 g_relation_delete_tuple(): free tuples which are removed from the
666 all_tuples hash table. this fixes a temporary leak that was present
667 in the memchunk code until the destruction of the relation.
669 2005-10-29 Matthias Clasen <mclasen@redhat.com>
671 * tests/convert-test.c: Add some tests for conversions between
672 UTF-8, UCS-4 and UTF-16.
674 * glib/gutf8.c (g_utf8_to_ucs4, g_utf8_to_utf16): Fix handling
675 of len == -1, noticed by Morten Welinder.
677 2005-10-27 Erdal Ronahi <erdal.ronahi@gmail.com>
679 * configure.in: Added ku (Kurdish) to ALL_LINGUAS
681 2005-10-26 Matthias Clasen <mclasen@redhat.com>
683 * glib/gutf8.c (g_ucs4_to_utf8): Don't set items_read twice
684 in the error case, and add some documentation. (#319806, Morten
687 2005-10-19 Manish Singh <yosh@gimp.org>
689 * configure.in: Use AC_CHECK_FUNCS for _NSGetEnviron, to get the
690 config.h symbol automatically. Fixes bug #313731.
692 2005-10-19 Tor Lillqvist <tml@novell.com>
694 * glib/gwin32.c (g_win32_get_package_installation_directory):
695 Return a g_strdup()ed copy of the value stored in the hash table,
696 so that it can be g_free()d without leaving a dangling pointer in
697 the hash table. (#319232)
699 2005-10-06 Matthias Clasen <mclasen@redhat.com>
701 * glib/gunicollate.c (g_utf8_collate_key_for_filename): Handle
702 all-zero sequences correctly. (#317930, Sebastien Bacher)
704 2005-10-05 Matthias Clasen <mclasen@redhat.com>
706 * glib/gmarkup.c (g_markup_escape_text): Doc addition
708 * Makefile.am: Add Collation tests.
710 * tests/collate/*: Inputs and expected outputs for collation tests.
712 * tests/run-collate-tests.sh: Script to run collation tests.
714 * tests/unicode-collate.c (main): Rework slightly to make
715 it usable in unit tests. Also test g_utf8_collate_key_for_filename().
717 2005-10-01 Behdad Esfahbod <behdad@gnome.org>
719 * docs/reference/glib/tmpl/unicode.sgml:
720 * glib/gen-unicode-tables.pl:
722 * glib/gunichartables.h:
724 * tests/casefold.txt:
725 * tests/casemap.txt: Updated to Unicode 4.1. There are five new
726 GUnicodeBreakType types. That may break some applications, like
729 2005-09-26 Matthias Clasen <mclasen@redhat.com>
731 * glib/gstrfuncs.c (g_ascii_strtoull): Add details to the
732 docs. (#314393, Matthew F. Barnes)
735 * glib/gprintf.h: Remove g_snprintf() and g_vsnprintf(), since
736 they are already declared in glib.h. This doesn't break documented
737 use of gprintf.h, but should probably be pointed out in the
738 release notes for 2.10. (#314232, Behdad Esfahbod)
740 Tue Sep 20 13:16:04 2005 Tim Janik <timj@imendio.com>
742 * glib/gpattern.c (g_pattern_ph_match): applied significant recursion
743 complexity optimization, based on a patch from Matthias Clasen.
745 * tests/patterntest.c: more tests, mostly from matthias.
747 2005-09-20 Matthias Clasen <mclasen@redhat.com>
749 * glib/gqueue.c (g_queue_insert_sorted): Correct the docs.
750 (#316703, Mark Drago)
752 Mon Sep 19 17:23:23 2005 Tim Janik <timj@imendio.com>
754 * glib/gpattern.c: applied a patch from matthias which checks on the
755 upper bound of GPatternSpec length to optimize matches.
758 * tests/patterntest.c: added more match cases.
760 2005-09-16 Tor Lillqvist <tml@novell.com>
762 * glib/gstrfuncs.c (g_ascii_strcasecmp, g_ascii_strncasecmp): Add
763 warning to doc comment that these functions should not be used on
764 encodings like CP932.
766 2005-09-14 Matthias Clasen <mclasen@redhat.com>
768 * tests/keyfile-test.c: Add a test for grup names of length 1.
770 * glib/gkeyfile.c (g_key_file_line_is_group): Accept group names
771 of length 1. (#316309)
773 2005-09-12 Matthias Clasen <mclasen@redhat.com>
775 * glib/gmarkup.c (g_markup_escape_text): Clarify docs.
777 2005-09-11 Sebastian Wilhelmi <seppi@seppi.de>
779 * tests/refcount/Makefile.am (INCLUDES): Link the the refcount
780 tests to the system thread library $(G_THREAD_LIBS). Fixes #313744
783 2005-09-11 Kjartan Maraas <kmaraas@gnome.org>
785 * glib/gmain.c: (g_child_watch_prepare), (g_child_watch_check),
786 (child_watch_helper_thread): Remove some dead code. Closes
789 2005-09-07 Tor Lillqvist <tml@novell.com>
791 * glib/Makefile.am: Create also a console version of the
792 gspawn-win32-helper program, gspawn-win32-helper-console.exe.
793 It's otherwise identical to gspawn-win32-helper.exe, except marked
794 as a console application (linked without the -mwindows option).
796 * glib/gspawn-win32.c (do_spawn_directly, do_spawn_with_pipes):
797 Drop the dont_wait parameter. Its truth value correlated 100% with
798 the NULLness of the exit_status parameter anyway, so it's enough
799 to check whether exit_status is NULL. Invert the sense of the
800 dont_return_handle parameter and rename it to do_return_handle, to
801 make the code easier to read by avoiding double negations.
803 (g_spawn_sync_utf8, g_spawn_async_with_pipes_utf8): Modify calls
804 to do_spawn_with_pipes() accordingly.
806 (do_spawn_with_pipes): If we have a console, use the console
807 version of the helper program, otherwise use the GUI one. This
808 avoids extra console windows opening up in some situations. (In
809 case a console application uses the GUI gspawn-win32-helper.exe to
810 spawn another console application we would get a separate console
811 for the spawned console application).
813 * glib-zip.in: Distribute also gspawn-win32-helper-console.exe.
815 2005-09-05 Matthias Clasen <mclasen@redhat.com>
817 * glib/gmappedfile.c (g_mapped_file_new): Report an error
818 if the file is too large. (#315275, Kjartan Maraas)
820 * glib/gkeyfile.c (g_key_file_load_from_fd): The return value
821 of read() is signed. (#315273, Kjartan Maraas)
823 2005-08-31 Tor Lillqvist <tml@novell.com>
825 * glib/gutils.h: Wrapping atexit() is a bad idea on Windows, where
826 the EXE and each DLL have their own atexit function chains.
828 #define g_atexit as atexit instead. This means it has a
829 better chance of doing what the caller wants. For instance,
830 gtkhtml calls g_atexit() registering a function in gtkhtml
831 itself. This caused a crash when g_atexit() was implemented as a
832 function in the GLib DLL. The gtkhtml DLL was already unloaded by
833 the time the GLib DLL got unloaded.
835 * glib/gutils.c: #undef the #define mentioned above, to also get a
836 real g_atexit() into the DLL for backward compatibility. Document
837 the Windows behaviour of g_atexit(), and document the varying ways
838 atexit() can behave in the context of dynamically loaded modules
841 2005-08-31 Matthias Clasen <mclasen@redhat.com>
845 * glib/gdataset.c: Add string interning functions.
847 2005-08-28 Matthias Clasen <mclasen@redhat.com>
849 * glib/giochannel.c: Unify some near-duplicate strings. (#314654,
852 2005-08-26 Matthias Clasen <mclasen@redhat.com>
854 * configure.in: Bump version to 2.9.0
856 2005-08-25 Tor Lillqvist <tml@novell.com>
858 Make also the g_spawn*() functions take parameters in the GLib
859 file name encoding, i.e. UTF-8, on Windows. Has no impact on Unix
860 API or ABI. Like the other GLib API that was earlier changed to
861 use UTF-8 on Windows, the names of the functions that take UTF-8
862 have _utf8 suffixes added by using preprocessor macros in the
863 header file. The old names are kept for functions with the old
864 behaviour, taking parameters in the system codepage, for DLL ABI
867 * glib/gspawn.h: On Win32 add the suffix _utf8 to the names of the
868 g_spawn*() functions.
870 * glib/gspawn-win32.c: Use wide-char API on NT-based
871 Windows. Convert parameters from UTF-8 to wide chars (NT) or
872 system codepage (Win9x) and call the C library _wspawn*() or
873 spawn*() functions respectvely. Add DLL ABI stability versions
874 that take parameters in the system codepage.
876 * glib/gspawn-win32-helper.c: On NT-based Windows use the
877 wide-char versions of argv and envp, and use wide-char API to
878 change directory and spawn the program to run. Remove the verbose
879 debugging output, it was too complex to modify for the wide-char
880 features. (Just add temporary debugging printouts if needed, no
881 need to have them permanently in the source.)
883 * glib/gspawn.c: Corresponding documentation updates.
885 * glib/glib.symbols: Corresponding changes: Mark the ABI stability
886 symbols as PRIVATE, add the new _utf8-suffixed ones.
888 2005-08-24 Stepan Kasal <kasal@ucw.cz>
890 * glib/gtypes.h (G_MININT64): Cast the constant to gint64; it is
891 guint64 otherwise and that can produce warnings about comparison
892 between signed and unsigned.
894 2005-08-23 Matthias Clasen <mclasen@redhat.com>
896 * glib/gutils.c: Fix the crt_externs.h include.
898 2005-08-23 Stepan Kasal <kasal@ucw.cz>
900 * NEWS: Fix spelling of my first name.
902 2005-08-23 Matthias Clasen <mclasen@redhat.com>
906 * === Released 2.8.1 ===
910 2005-08-20 Hans Breuer <hans@breuer.org>
912 * glib/makefile.msc.in : link with ws2_32.lib
914 2005-08-18 Tor Lillqvist <tml@novell.com>
916 * configure.in: Check for <sys/wait.h>
918 * glib/gbacktrace.c: Include <sys/wait.h> on if HAVE_SYS_WAIT_H.
920 2005-08-18 Ross Burton <ross@burtonini.com>
923 Optimise single-character insertions.
928 * tests/string-test.c:
929 Add tests for new optimisation, and fix a leak.
931 2005-08-17 Matthias Clasen <mclasen@redhat.com>
933 * configure.in: Check for crt_externs.h and _NSGetEnviron.
935 * glib/gutils.c: On Darwin, include crt-externs.h and
936 define environ using _NSGetEnviron(). (#313731)
938 2005-08-16 Stepan Kasal <kasal@ucw.cz>
940 * glib/gutils.c (g_get_any_init): Move the body of the big if...
941 (g_get_any_init_do): ... to this new function.
942 (g_get_any_init): Declare as inline.
943 (g_get_any_init_locked): New inline function, does the locking.
944 Make use of these two throughout the code.
946 2005-08-15 Matthias Clasen <mclasen@redhat.com>
948 * glib/gbacktrace.c (g_on_error_stack_trace): Wait for
949 the child process and then simply return. This makes
950 The "S" option work as documented in g_on_error_query().
951 (#313125, Matthew F. Barnes)
953 * glib/gunicode.h: Update the link to Unicode category
954 values. (#313369, Behnam Esfahbod)
956 * glib/gqueue.c (g_queue_find_custom): Clarify docs
957 a little. (#311727, Tristan van Berkom)
959 * glib/abicheck.sh, gobject/abicheck.sh: Make the
960 check work on ia64 too, where some symbols ended up
961 in yet another section.