1 2006-01-05 Matthias Clasen <mclasen@redhat.com>
3 * glib/gthreadpool.c: Avoid use of varargs macro. (#325864,
6 2006-01-04 Manish Singh <yosh@gimp.org>
9 * glib/gstdio.h: don't macroized g_access, g_chdir, and g_unlink
10 either, since they have the same issue as g_rmdir. (Related to
13 Wed Jan 4 13:33:25 2006 Tim Janik <timj@gtk.org>
15 * glib/gslice.c (magazine_cache_pop_magazine): don't reverse chunk
16 order when creating magazines, so we hand out chunks with ascending
19 2006-01-04 Abel Cheung <maddog@linuxhall.org>
21 * configure.in: Added "ml" "zh_HK" to ALL_LINGUAS.
23 2006-01-03 Martyn Russell <martyn@imendio.com>
25 * docs/reference/glib/glib-sections.txt:
27 * glib/gthreadpool.[ch]:
28 - Added new API g_thread_pool_get_idle_time() and
29 g_thread_pool_set_idle_time(). (#324228).
31 * tests/threadpool-test.c:
32 - Updated test case to do thread pool sorting, thread pool with
33 no sorting and a thread pool with idle thread timeouts.
35 2006-01-03 Matthias Clasen <mclasen@redhat.com>
37 * glib/gmain.h: Add new functions here, too.
39 * glib/glib.symbols: Add new functions.
41 2005-12-20 Michael Meeks <michael.meeks@novell.com>
43 * glib/gmain.c (g_main_context_is_owner): new method
44 to determine if the current thread is the owner of the
47 2006-01-02 Matthias Clasen <mclasen@redhat.com>
51 * glib/gstdio.c (g_rmdir): Don't provide g_rmdir() as a macro
52 expanding to rmdir, since rmdir is not declared in a portable
53 system header we can include in gstdio.h. (#325249, Jani Monoses)
55 2006-01-01 Tor Lillqvist <tml@novell.com>
57 * glib/gspawn-win32.c (g_spawn_sync_utf8): Set the GIOChannels for
58 stdout and stderr to unbuffered. Otherwise the giochannel layer
59 will try to read from them regardless whether the
60 g_io_channel_win32_poll() call here has indicated
61 readability or not. (#325310)
63 2005-12-29 Matthias Clasen <mclasen@redhat.com>
65 * glib/gutils.c (glib_check_version): Fix a copy-and-paste error
66 in the docs. (#325273, Declan Naughton)
68 2005-12-27 Manish Singh <yosh@gimp.org>
70 * tests/run-collate-tests.sh: use LC_ALL instead of LC_COLLATE, to
71 make sure we really override things.
73 2005-12-27 Matthias Clasen <mclasen@redhat.com>
75 Fix #316221, Michal Benes, Stanislav Brabec;
77 * configure.in: Fix a strict aliasing problem in
78 g_static_mutex_get_mutex().
79 * glib/gthread.h: ...and in
80 g_static_mutex_get_mutex_impl_shortcut().
82 * glib/gdatasetprivate.h: Add a cast to silence compiler
83 warnings. (#321978, Andrew Paprocki)
85 Partial fix for bug #32937, Bogdan Nicula.
87 * configure.in: Check for malloc.h
88 * glib/gslice.c: Don't include malloc.h unconditionally.
90 2005-12-27 Manish Singh <yosh@gimp.org>
92 * tests/run-collate-tests.sh: export LC_COLLATE so it takes.
93 (#324950, Dan Yefimov)
95 2005-12-26 Matthias Clasen <mclasen@redhat.com>
97 * glib/gslice.c: On Win32, include process.h (#325015, Kazuki
100 2005-12-25 Matthias Clasen <mclasen@redhat.com>
102 * AUTHORS: Update my email
104 * tests/slice-test.c: Fix C99isms. (#324950, Dan Yefimov)
106 2005-12-21 Matthias Clasen <mclasen@redhat.com>
108 * README.in: Add some notes on when 'make check' may fail.
110 2005-12-20 Matthias Clasen <mclasen@redhat.com>
113 * glib/gthreadpool.h:
114 * glib/gthreadpool.c (g_thread_pool_set_sort_function): New function
115 to sort tasks pushed into a threadpool. (#324479, Martyn Russell)
117 * tests/threadpool-test.c: Test this.
119 Tue Dec 20 18:14:14 2005 Tim Janik <timj@imendio.com>
121 * glib/gslice.[hc]: added mem_error() and mem_assert() to test and
122 handle errors without depending on gmessage.c which might not be
123 setup when the error occours.
124 removed G_SLICE_CONFIG_ALWAYS_FREE config option, fixed the code so
125 always freeing can be achieved by adjusting the working set time to
126 0 with G_SLICE_CONFIG_WORKING_SET_MSECS.
127 added G_SLICE_CONFIG_COLOR_INCREMENT to test different color increments
128 (mainly 0 and 1). reduced the minimum block size to 128 bytes, to
129 minimize wastage if small amounts of differently sized structrues are
130 allocated, this does come at a performance cost of roughly 5% though.
131 fixed up block alignment calculation, so it works for varying
132 block sizes. only use strerror() not g_strerror() since the latter
133 depends on working GQuark and GSlice.
134 mem_error(): implemented in terms of fprintf and vfprintf.
136 * tests/slice-color.c: new program to test cache colorization effects.
138 * tests/slice-test.c: trade G_SLICE_CONFIG_ALWAYS_FREE for 0 duration
139 G_SLICE_CONFIG_WORKING_SET_MSECS.
141 2005-12-17 Matthias Clasen <mclasen@redhat.com>
143 * glib/goption.c (parse_short_option): Set an error in all
144 failure cases. (#324332, Tim-Philipp Müller)
146 2005-12-17 Sebastian Wilhelmi <seppi@seppi.de>
148 * glib/gatomic.c: Fix memory barrier position in g_atomic_int_get
149 and g_atomic_pointer_get. Add g_atomic_int_set and
150 g_atomic_pointer_set implementations for the !DEFINE_WITH_MUTEXES &&
151 G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case, as well as defining them
152 as functions (additionally to the macros in the header) for the
153 !G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case.
155 2005-12-16 Matthias Clasen <mclasen@redhat.com>
157 * glib/gmem.c (g_allocator_new): Don't return a pointer to
158 a const struct, since apps expect to be able to modify it.
159 (#324179, J. Ali Harlow)
161 Tue Dec 13 10:13:32 2005 Tim Janik <timj@imendio.com>
163 * glib/gatomic.h: added g_atomic_pointer_set() and g_atomic_int_set()
164 for systems where the initialization of atomic variables requires a
165 write memory barrier.
167 2005-12-09 Matthias Clasen <mclasen@redhat.com>
169 * configure.in: Bump version
171 * === Released 2.9.1 ===
173 2005-12-09 Alexander Larsson <alexl@redhat.com>
177 Use G_GNUC_WARN_UNUSED_RESULT on list functions that return
180 2005-12-08 Matthias Clasen <mclasen@redhat.com>
184 2005-12-07 Martyn Russell <martyn@imendio.com>
186 * glib/gasyncqueue.c:
187 - Call g_queue_insert_sorted() instead of duplicating the code.
188 - Call g_queue_sort() instead of duplicating the code.
189 - Invert sort function results to make sure the same sort function
190 gives the same results across glist, gslist, gqueue and
193 * tests/asyncqueue-test.c:
194 - Updated the sort function to reflect the example in the
195 documentation for gasyncqueue.c.
197 2005-12-07 Martyn Russell <martyn@imendio.com>
199 * docs/reference/glib/glib-sections.txt:
200 * docs/reference/glib/tmpl/linked_lists_double.sgml:
201 * docs/reference/glib/tmpl/linked_lists_single.sgml:
204 - Added g_list_insert_sorted_with_data () and
205 g_slist_insert_sorted_with_data ().
206 - Removed the extra check in g_list_sort() and g_slist_sort() for
207 GCompareDataFunc vs. GCompareFunc.
209 2005-12-07 Tor Lillqvist <tml@novell.com>
211 * glib-gettextize.in: Look up prefix at run-time on Win32,
212 assuming the standard directory structure with glib-gettextize in
215 2005-12-06 Matthias Clasen <mclasen@redhat.com>
217 * glib/gmem.h: Exempt GTK+ from the mem chunk deprecation, since
218 we need GTK+ 2.8 to compile against GLib 2.10.
220 2005-12-05 Manish Singh <yosh@gimp.org>
222 * tests/run-collate-test.sh: set LC_COLLATE, not LANG, to be sure
223 to override any user settings.
225 2005-12-05 Matthias Clasen <mclasen@redhat.com>
227 * configure.in: Revert an accidential commit
229 2005-12-05 Matthias Clasen <mclasen@redhat.com>
231 * glib/gutf8.c: Documentation updates. (#323291, Morten Welinder)
233 * tests/Makefile.am (TESTS_ENVIRONMENT):
234 * tests/gobject/Makefile.am (TESTS_ENVIRONMENT): Set
235 MALLOC_CHECK_ and MALLOC_PERTURB_
237 * tests/run-collate-tests.sh: Run the collation tests explicitly
238 in en_US locale. (#320463)
240 * glib/gmem.h: Really deprecate GMemChunk.
247 * glib/gstrfuncs.c: Add versioned deprecation docs.
249 2005-12-05 Martyn Russell <martyn@imendio.com>
251 * docs/reference/glib/glib-sections.txt:
252 * glib/gasyncqueue.[ch]:
253 - Added support for sorting async queues by with _push_sorted(),
254 _push_sorted_unlocked(), _sort() and _sort_unlocked() (#323047).
257 * tests/asyncqueue-test.c:
258 - Added test case for gasyncqueue.c
260 Mon Dec 5 15:53:20 2005 Tim Janik <timj@imendio.com>
262 * glib/gslice.c: implement chain walking for arbitrary ->next pointer
263 offsets in g_slice_free_chain_with_offset() based on a patch by behdad
264 in bug 323178. moved time consuming logic from g_slice_free() out of
265 the inner loop, so g_slice_free_chain_with_offset() provides a real
266 performance benefit over g_slice_free1() now.
268 * glib/gslice.h: renamed g_slice_free_chain() to
269 g_slice_free_chain_with_offset(). implemented g_slice_free_chain() as
270 a type-safe macro as suggested in bug 323178.
271 simplified the macro implementation of g_slice_free() and implemented
272 it in a type safe manner for all compliers as suggested by Morten
273 Welinder <mortenw@gnome.org>.
278 * glib/glib.symbols: s/g_slice_free_chain/g_slice_free_chain_with_offset/
280 2005-12-05 Matthias Clasen <mclasen@redhat.com>
282 * glib/gasyncqueue.c: Add some docs.
284 * tests/libmoduletestplugin_a.c: Fix compiler warnings.
286 * glib/gatomic.c: In the ia64 implementation, use
287 __sync builtin without _si or _di suffix. (#321229,
288 Stanislav Brabec, patch by Andreas Schwab)
290 2005-12-04 Behdad Esfahbod <behdad@gnome.org>
292 * glib/gslice.h: Remove comma at the end of enum.
294 2005-12-04 Matthias Clasen <mclasen@redhat.com>
296 Handle multiple user names with the same UID better.
297 (#319535, Laszlo Peter)
299 * glib/gutils.c (g_get_any_init_do): When determining user
300 data, first look up $LOGNAME. If the UID doesn't match
301 getuid(), fall back to the current behaviour of looking
302 up the user data based on getuid().
304 2005-12-04 Matthias Clasen <mclasen@redhat.com>
306 * glib/gmacros.h (G_GNUC_WARN_UNUSED_RESULT): Add a macro
307 to make gcc warn if a function result is ignored. (#145466,
308 Arjan van de Ven, Alex Larsson)
310 * glib/gmem.h: Add the new attribute to g_realloc and
313 2005-12-03 Matthias Clasen <mclasen@redhat.com>
316 * glib/gcache.h: Deprecate g_cache_value_foreach. (#322956,
319 * glib/gmappedfile.c: Make mapping of empty files
322 * glib/gfileutils.c: Don't fork a new process just to
323 fix the permissions of the created temp file. (#321318,
324 Alexis S. L. Carvalho)
326 2005-12-02 Matthias Clasen <mclasen@redhat.com>
328 * README.in: Add a note about Solaris threads.
331 * configure.in: Remove support for Solaris threads.
332 (#136971, Sebastian Wilhelmi, patch by Andrew Paprocki)
334 2005-12-02 Matthias Clasen <mclasen@redhat.com>
336 * configure.in: Define G_GUINT64_CONSTANT in analogy to
337 G_GINT64_CONSTANT. (#322568, Andrew Paprocki)
339 2005-12-02 Matthias Clasen <mclasen@redhat.com>
341 * glib/gslice.c: Win32 portability fixes and C99ism removal,
342 pointed out by Kazuki Iwamoto. (#323052)
344 2005-12-02 Matthias Clasen <mclasen@redhat.com>
346 * glib/gatomic.c (g_atomic_int_add): Add a missing volatile for
347 the IA64 implementation.
349 Fri Dec 2 16:18:09 2005 Tim Janik <timj@imendio.com>
351 * glib/gslice.c: some naming and type size fixups.
353 Fri Dec 2 13:08:58 2005 Tim Janik <timj@gtk.org>
355 * glib/gslice.h (g_slice_free): reworked GCC-specific type-safe macro
356 variant into something less verbose, but digestible for gcc-3.4.
358 Fri Dec 2 10:55:07 2005 Tim Janik <timj@imendio.com>
360 * tests/slice-test.c: extended to perform the benchmarking on the old
361 memchunk code if 'O' is selected.
363 * tests/memchunks.c: new file which contains the old GLib mem chunks
364 implementation with prefix old_mem_chunk_.
366 * tests/Makefile.am: added memchunks.c
368 Fri Dec 2 00:16:59 2005 Tim Janik <timj@imendio.com>
370 * glib/gslice.c: improved gettimeofday() timer resolution by maintaining
371 acceptable load distribution of the syscall.
372 (allocator_get_magazine_threshold): reduce minimum/single-thread magazine
373 sizes to approximately page_size/2.5 to avoid excessive startup allocations.
375 2005-12-01 Matthias Clasen <mclasen@redhat.com>
377 * tests/gobject/ifacecheck.c (test_iface_base_init): Don't
378 double-free base interface members.
380 * tests/Makefile.am: Remove duplicate variable.
382 Thu Dec 1 17:32:46 2005 Tim Janik <timj@imendio.com>
384 * glib/gslice.[hc]: new slice allocator implementation.
386 * tests/slice-test.c: added random slice allocation test.
388 * glib/gthread.[hc]: removed newly added private thread mem API.
390 * glib/gthreadinit.h:
393 * glib/gmem.c: divided glib threading initialisation into three phases,
394 initialisation where private keys and messaging are not available (only
395 needed by gmem.c), initialisation without messaging but private keys
396 available (gslice.c, gmessage.c), and full fledged initialisers that
397 server the rest of glib. initialisation functions got renamed to reflect
398 the limitations of their corresponding phases.
400 * glib/gmem.c: removed memchunk code, defer allocations to
403 * glib/gmem.[hc]: removed g_slice_* skeletons.
405 * glib/glib.symbols: added g_slice_* symbols.
407 * configure.in: check for availability of posix_memalign(3), memalign(3)
410 * glib/Makefile.am: added gslice.[hc].
412 2005-12-01 Tor Lillqvist <tml@novell.com>
414 * glib/gstdio.c (g_stat): In the Win32 implementation, strip
415 trailing slash(es) for non-root folders. stat() fails if non-root
416 folders are specified with trailing slashes. It's too much hassle
417 to demand that callers strip such slashes themselves, especially
418 as it is easy to get it wrong and strip the slash of a root
420 (g_rename): On NT-based Windows, use MoveFileEx() with
421 MOVEFILE_REPLACE_EXISTING to better match Unix behaviour.
423 2005-11-28 Matthias Clasen <mclasen@redhat.com>
425 Fix G_STMT_START / G_STMT_END on Solaris. (#321972,
428 * configure.in: Check whether do { } while (0) works.
430 * glib/gmacros.h: Use do { } while (0) for G_STMT_START /
431 G_STMT_END if it works.
433 2005-11-28 Matthias Clasen <mclasen@redhat.com>
435 * glib/gthread.c (g_static_rw_lock_wait, g_static_rw_lock_signal):
436 * glib/gnode.c (g_node_depth_traverse_level):
437 * glib/gmem.c (g_allocator_new):
438 * glib/ghash.c (g_hash_table_unref, g_hash_table_destroy)
439 (g_hash_table_foreach_remove_or_steal): Silence compiler
442 2005-11-27 Matthias Clasen <mclasen@redhat.com>
444 * glib/gunicollate.c (g_utf8_collate_key_for_filename):
445 Don't read beyond len. (#322520, Christian Persch)
447 Wed Nov 23 17:34:01 2005 Tim Janik <timj@imendio.com>
449 * glib/gdataset.c: access datalist flags via atomic pointer access
450 functions, instead of acquiring the dataset lock. this is faster and
451 also matches the atomic pointer readouts in gdatalistprivate.h and
452 direct pointer modifications required by gobject.c.
454 Wed Nov 23 13:35:31 2005 Tim Janik <timj@gtk.org>
456 * glib/gdataset.c: streamlined initialization code somewhat,
457 removed GData node cache.
459 2005-11-22 Matthias Clasen <mclasen@redhat.com>
461 * glib/ghash.c (g_hash_table_ref):
462 (g_hash_table_unref): Mark these as new API.
464 Avoid double locking in g_intern_string (#322133,
467 * glib/gdataset.c (g_quark_from_string_internal): New
468 internal function which factors out common parts of
469 g_quark_from[_static]_string.
470 (g_quark_from_string, g_quark_from_static_string):
471 Use g_quark_from_string_internal.
472 (g_intern_string, g_intern_static_string): Use
473 g_quark_from_string_internal, and only take the
475 (g_quark_new): Don't store the strings shifted by -1
476 in the g_quarks array.
477 (g_quark_to_string): Adapt to the previous change.
479 Tue Nov 22 14:04:26 2005 Tim Janik <timj@imendio.com>
483 g_hash_table_new_full(): create hash tables with a ref count of 1.
484 g_hash_table_ref(): atomically ref_count+=1
485 g_hash_table_unref(): atomically ref_count-=1, destroys hash table
486 when refcount reaches 0.
487 g_hash_table_destroy(): just destroy keys and values, unref by 1.
488 g_hash_table_insert():
489 g_hash_table_replace(): assert ref_count>0.
492 * glib/gatomic.c: added 'volatile' qualifier to all atomic pointer and
495 2005-11-20 Behdad Esfahbod <behdad@gnome.org>
497 * glib/guniprop.c (g_unichar_get_mirror_char): Remove unused
498 variables. (#321984, Andrew Paprocki)
500 2005-11-18 Matthias Clasen <mclasen@redhat.com>
502 * configure.in: Bump version
504 2005-11-17 Matthias Clasen <mclasen@redhat.com>
506 * === Released 2.9.0 ===
508 * glib/glib.symbols: Updates
510 * glib/gutf8.c (g_utf8_offset_to_pointer): Add some performance
511 hints to the docs. (Paolo Borelli)
515 * glib/gthread.c (g_thread_foreach): Mark as new api.
517 * README.in: Updates.
521 * glib/gdate.c (g_date_set_time_t):
522 (g_date_set_time_val): Add functions to set a date from
523 a time_t and from a GTimeVal.
524 (g_date_set_time): Deprecate. (#314067, Roger Leigh)
527 * tests/utf8-pointer.c: Unit tests for g_utf8_pointer_to_offset
528 and g_utf8_offset_to_pointer.
530 * glib/gutf8.c (g_utf8_pointer_to_offset)
531 (g_utf8_offset_to_pointer): Handle negative offsets, and use
532 "stutter stepping" for going backwards. (#320638, Larry
537 * glib/gthread.c: const correctness fixes, found
538 by Arjan van de Ven and gcc.
540 2005-11-16 Behdad Esfahbod <behdad@gnome.org>
542 * docs/reference/glib/tmpl/unicode.sgml: Correct typo on mentioning
543 @G_UNICODE_BREAK_UNKNOWN that should be #G_UNICODE_BREAK_UNKNOWN.
545 2005-11-09 Behdad Esfahbod <behdad@gnome.org>
547 * glib/guniprop.c: Use bit hacks instead when checking a general
548 category value against multiple values.
550 * glib/gutf8.c: Change ISO10646 to Unicode in docs.
552 2005-11-10 Simos Xenitellis <simos@gnome.org>
554 * configure.in: Added tt (Tatar) to ALL_LINGUAS.
556 2005-11-05 Tor Lillqvist <tml@novell.com>
558 * glib/giowin32.c (g_io_win32_set_flags): Don't call g_warning().
559 (#320688) Instead set error and return failure.
561 2005-11-07 Matthias Clasen <mclasen@redhat.com>
563 * glib/gmain.c (g_main_dispatch): Don't call cb_funcs->unref
564 while holding the context lock. (#320886, Andy Wingo)
566 2005-11-07 Billy Biggs <vektor@dumbterm.net>
568 * tests/refcount/Makefile.am: Add a missing reference to libglib
569 in the LDADD for this test.
571 2005-11-07 Matthias Clasen <mclasen@redhat.com>
573 * glib/gmem.c (g_allocator_new): We need to set n_preallocs to a
574 nonzero value, otherwise GTK+ 2.8 breaks when compiled against
575 GLib 2.9. (#320755, Luca Ferretti)
577 2005-11-04 Matthias Clasen <mclasen@redhat.com>
579 * glib/Makefile.am: Apply a patch from Behdad Esfahbod to
580 use a faster lookup table for g_unichar_get_mirror_char().
581 * glib/gmirroringtable.h: The new table.
582 * glib/gunichartables.h: Remove bidi_mirroring_table.
583 * glib/gen-unicode-tables.pl: Don't generate the mirroring
585 * glib/glib-mirroring-tab/*: Sources for the program
586 which generated gmirroringtable.h.
588 * glib/glist.c: Avoid some code duplication.
590 * glib/gscanner.h: Include gdataset.h. (#320322)
592 * glib/gdebug.h: add new GDebugFlag for fatal_criticals
593 * glib/gmessages.c: (_g_debug_init): handle G_DEBUG=fatal_criticals,
594 to help find critical warnings in applications. (#320017,
597 2005-11-02 Tor Lillqvist <tml@novell.com>
599 * glib/glib.symbols: Remove large amount of trailing whitespace
600 from one line. Remove #ifdef G_UNDEFINED from around
601 g_hash_table_new and g_hash_table_new_full symbols, what was the
602 purpose of that? Add the g_slice functions.
604 Tue Nov 1 16:24:20 2005 Tim Janik <timj@imendio.com>
606 * glib/gmem.[hc]: prepared deprecation of GMemChunk and GAllocator.
607 added g_slice_*() API to allocate and cache small bits of memory.
608 an actuall allocator implementation for g_slice_*() is still pending.
610 * glib/gthread.[hc]: changes from a patch by Matthias Clasen.
611 changed GRealThread list to use in-structure *next; fields instead
612 of GSList, in order for thread iteration to not depenend on g_slice_*()
614 _g_thread_mem_private_get():
615 _g_thread_mem_private_set(): added accessors for private memory,
616 needed because the ordinary GPrivate implementation relies on GArray
617 and GSList and therefore indirectly on working g_slice_*() allocations.
620 g_thread_foreach(): new public API function to loop over all existing threads.
634 removed GAllocator and free list usages and accompanying locks.
635 use g_slice_*() API to allocate and cache small bits of memory.
637 * glib/ghook.h: removed GMemChunk field from public API.
640 * glib/glist.h: deprecate allocator API, provide _free1() for consistency.
642 * glib/gnode.h: deprecate allocator API.
644 * glib/gmain.c: reordered GPollRec fields so g_slice_free_chain() can
645 be used for poll rec lists.
647 * glib/grel.c: removed mem chunk usage, and allocated tuples via g_slice_*().
648 g_relation_destroy(): free all tuples from the all_tuples hash table,
649 this effectively maintains the life time track keeping of tuples.
650 g_relation_delete_tuple(): free tuples which are removed from the
651 all_tuples hash table. this fixes a temporary leak that was present
652 in the memchunk code until the destruction of the relation.
654 2005-10-29 Matthias Clasen <mclasen@redhat.com>
656 * tests/convert-test.c: Add some tests for conversions between
657 UTF-8, UCS-4 and UTF-16.
659 * glib/gutf8.c (g_utf8_to_ucs4, g_utf8_to_utf16): Fix handling
660 of len == -1, noticed by Morten Welinder.
662 2005-10-27 Erdal Ronahi <erdal.ronahi@gmail.com>
664 * configure.in: Added ku (Kurdish) to ALL_LINGUAS
666 2005-10-26 Matthias Clasen <mclasen@redhat.com>
668 * glib/gutf8.c (g_ucs4_to_utf8): Don't set items_read twice
669 in the error case, and add some documentation. (#319806, Morten
672 2005-10-19 Manish Singh <yosh@gimp.org>
674 * configure.in: Use AC_CHECK_FUNCS for _NSGetEnviron, to get the
675 config.h symbol automatically. Fixes bug #313731.
677 2005-10-19 Tor Lillqvist <tml@novell.com>
679 * glib/gwin32.c (g_win32_get_package_installation_directory):
680 Return a g_strdup()ed copy of the value stored in the hash table,
681 so that it can be g_free()d without leaving a dangling pointer in
682 the hash table. (#319232)
684 2005-10-06 Matthias Clasen <mclasen@redhat.com>
686 * glib/gunicollate.c (g_utf8_collate_key_for_filename): Handle
687 all-zero sequences correctly. (#317930, Sebastien Bacher)
689 2005-10-05 Matthias Clasen <mclasen@redhat.com>
691 * glib/gmarkup.c (g_markup_escape_text): Doc addition
693 * Makefile.am: Add Collation tests.
695 * tests/collate/*: Inputs and expected outputs for collation tests.
697 * tests/run-collate-tests.sh: Script to run collation tests.
699 * tests/unicode-collate.c (main): Rework slightly to make
700 it usable in unit tests. Also test g_utf8_collate_key_for_filename().
702 2005-10-01 Behdad Esfahbod <behdad@gnome.org>
704 * docs/reference/glib/tmpl/unicode.sgml:
705 * glib/gen-unicode-tables.pl:
707 * glib/gunichartables.h:
709 * tests/casefold.txt:
710 * tests/casemap.txt: Updated to Unicode 4.1. There are five new
711 GUnicodeBreakType types. That may break some applications, like
714 2005-09-26 Matthias Clasen <mclasen@redhat.com>
716 * glib/gstrfuncs.c (g_ascii_strtoull): Add details to the
717 docs. (#314393, Matthew F. Barnes)
720 * glib/gprintf.h: Remove g_snprintf() and g_vsnprintf(), since
721 they are already declared in glib.h. This doesn't break documented
722 use of gprintf.h, but should probably be pointed out in the
723 release notes for 2.10. (#314232, Behdad Esfahbod)
725 Tue Sep 20 13:16:04 2005 Tim Janik <timj@imendio.com>
727 * glib/gpattern.c (g_pattern_ph_match): applied significant recursion
728 complexity optimization, based on a patch from Matthias Clasen.
730 * tests/patterntest.c: more tests, mostly from matthias.
732 2005-09-20 Matthias Clasen <mclasen@redhat.com>
734 * glib/gqueue.c (g_queue_insert_sorted): Correct the docs.
735 (#316703, Mark Drago)
737 Mon Sep 19 17:23:23 2005 Tim Janik <timj@imendio.com>
739 * glib/gpattern.c: applied a patch from matthias which checks on the
740 upper bound of GPatternSpec length to optimize matches.
743 * tests/patterntest.c: added more match cases.
745 2005-09-16 Tor Lillqvist <tml@novell.com>
747 * glib/gstrfuncs.c (g_ascii_strcasecmp, g_ascii_strncasecmp): Add
748 warning to doc comment that these functions should not be used on
749 encodings like CP932.
751 2005-09-14 Matthias Clasen <mclasen@redhat.com>
753 * tests/keyfile-test.c: Add a test for grup names of length 1.
755 * glib/gkeyfile.c (g_key_file_line_is_group): Accept group names
756 of length 1. (#316309)
758 2005-09-12 Matthias Clasen <mclasen@redhat.com>
760 * glib/gmarkup.c (g_markup_escape_text): Clarify docs.
762 2005-09-11 Sebastian Wilhelmi <seppi@seppi.de>
764 * tests/refcount/Makefile.am (INCLUDES): Link the the refcount
765 tests to the system thread library $(G_THREAD_LIBS). Fixes #313744
768 2005-09-11 Kjartan Maraas <kmaraas@gnome.org>
770 * glib/gmain.c: (g_child_watch_prepare), (g_child_watch_check),
771 (child_watch_helper_thread): Remove some dead code. Closes
774 2005-09-07 Tor Lillqvist <tml@novell.com>
776 * glib/Makefile.am: Create also a console version of the
777 gspawn-win32-helper program, gspawn-win32-helper-console.exe.
778 It's otherwise identical to gspawn-win32-helper.exe, except marked
779 as a console application (linked without the -mwindows option).
781 * glib/gspawn-win32.c (do_spawn_directly, do_spawn_with_pipes):
782 Drop the dont_wait parameter. Its truth value correlated 100% with
783 the NULLness of the exit_status parameter anyway, so it's enough
784 to check whether exit_status is NULL. Invert the sense of the
785 dont_return_handle parameter and rename it to do_return_handle, to
786 make the code easier to read by avoiding double negations.
788 (g_spawn_sync_utf8, g_spawn_async_with_pipes_utf8): Modify calls
789 to do_spawn_with_pipes() accordingly.
791 (do_spawn_with_pipes): If we have a console, use the console
792 version of the helper program, otherwise use the GUI one. This
793 avoids extra console windows opening up in some situations. (In
794 case a console application uses the GUI gspawn-win32-helper.exe to
795 spawn another console application we would get a separate console
796 for the spawned console application).
798 * glib-zip.in: Distribute also gspawn-win32-helper-console.exe.
800 2005-09-05 Matthias Clasen <mclasen@redhat.com>
802 * glib/gmappedfile.c (g_mapped_file_new): Report an error
803 if the file is too large. (#315275, Kjartan Maraas)
805 * glib/gkeyfile.c (g_key_file_load_from_fd): The return value
806 of read() is signed. (#315273, Kjartan Maraas)
808 2005-08-31 Tor Lillqvist <tml@novell.com>
810 * glib/gutils.h: Wrapping atexit() is a bad idea on Windows, where
811 the EXE and each DLL have their own atexit function chains.
813 #define g_atexit as atexit instead. This means it has a
814 better chance of doing what the caller wants. For instance,
815 gtkhtml calls g_atexit() registering a function in gtkhtml
816 itself. This caused a crash when g_atexit() was implemented as a
817 function in the GLib DLL. The gtkhtml DLL was already unloaded by
818 the time the GLib DLL got unloaded.
820 * glib/gutils.c: #undef the #define mentioned above, to also get a
821 real g_atexit() into the DLL for backward compatibility. Document
822 the Windows behaviour of g_atexit(), and document the varying ways
823 atexit() can behave in the context of dynamically loaded modules
826 2005-08-31 Matthias Clasen <mclasen@redhat.com>
830 * glib/gdataset.c: Add string interning functions.
832 2005-08-28 Matthias Clasen <mclasen@redhat.com>
834 * glib/giochannel.c: Unify some near-duplicate strings. (#314654,
837 2005-08-26 Matthias Clasen <mclasen@redhat.com>
839 * configure.in: Bump version to 2.9.0
841 2005-08-25 Tor Lillqvist <tml@novell.com>
843 Make also the g_spawn*() functions take parameters in the GLib
844 file name encoding, i.e. UTF-8, on Windows. Has no impact on Unix
845 API or ABI. Like the other GLib API that was earlier changed to
846 use UTF-8 on Windows, the names of the functions that take UTF-8
847 have _utf8 suffixes added by using preprocessor macros in the
848 header file. The old names are kept for functions with the old
849 behaviour, taking parameters in the system codepage, for DLL ABI
852 * glib/gspawn.h: On Win32 add the suffix _utf8 to the names of the
853 g_spawn*() functions.
855 * glib/gspawn-win32.c: Use wide-char API on NT-based
856 Windows. Convert parameters from UTF-8 to wide chars (NT) or
857 system codepage (Win9x) and call the C library _wspawn*() or
858 spawn*() functions respectvely. Add DLL ABI stability versions
859 that take parameters in the system codepage.
861 * glib/gspawn-win32-helper.c: On NT-based Windows use the
862 wide-char versions of argv and envp, and use wide-char API to
863 change directory and spawn the program to run. Remove the verbose
864 debugging output, it was too complex to modify for the wide-char
865 features. (Just add temporary debugging printouts if needed, no
866 need to have them permanently in the source.)
868 * glib/gspawn.c: Corresponding documentation updates.
870 * glib/glib.symbols: Corresponding changes: Mark the ABI stability
871 symbols as PRIVATE, add the new _utf8-suffixed ones.
873 2005-08-24 Stepan Kasal <kasal@ucw.cz>
875 * glib/gtypes.h (G_MININT64): Cast the constant to gint64; it is
876 guint64 otherwise and that can produce warnings about comparison
877 between signed and unsigned.
879 2005-08-23 Matthias Clasen <mclasen@redhat.com>
881 * glib/gutils.c: Fix the crt_externs.h include.
883 2005-08-23 Stepan Kasal <kasal@ucw.cz>
885 * NEWS: Fix spelling of my first name.
887 2005-08-23 Matthias Clasen <mclasen@redhat.com>
891 * === Released 2.8.1 ===
895 2005-08-20 Hans Breuer <hans@breuer.org>
897 * glib/makefile.msc.in : link with ws2_32.lib
899 2005-08-18 Tor Lillqvist <tml@novell.com>
901 * configure.in: Check for <sys/wait.h>
903 * glib/gbacktrace.c: Include <sys/wait.h> on if HAVE_SYS_WAIT_H.
905 2005-08-18 Ross Burton <ross@burtonini.com>
908 Optimise single-character insertions.
913 * tests/string-test.c:
914 Add tests for new optimisation, and fix a leak.
916 2005-08-17 Matthias Clasen <mclasen@redhat.com>
918 * configure.in: Check for crt_externs.h and _NSGetEnviron.
920 * glib/gutils.c: On Darwin, include crt-externs.h and
921 define environ using _NSGetEnviron(). (#313731)
923 2005-08-16 Stepan Kasal <kasal@ucw.cz>
925 * glib/gutils.c (g_get_any_init): Move the body of the big if...
926 (g_get_any_init_do): ... to this new function.
927 (g_get_any_init): Declare as inline.
928 (g_get_any_init_locked): New inline function, does the locking.
929 Make use of these two throughout the code.
931 2005-08-15 Matthias Clasen <mclasen@redhat.com>
933 * glib/gbacktrace.c (g_on_error_stack_trace): Wait for
934 the child process and then simply return. This makes
935 The "S" option work as documented in g_on_error_query().
936 (#313125, Matthew F. Barnes)
938 * glib/gunicode.h: Update the link to Unicode category
939 values. (#313369, Behnam Esfahbod)
941 * glib/gqueue.c (g_queue_find_custom): Clarify docs
942 a little. (#311727, Tristan van Berkom)
944 * glib/abicheck.sh, gobject/abicheck.sh: Make the
945 check work on ia64 too, where some symbols ended up
946 in yet another section.