1 2006-01-05 Matthias Clasen <mclasen@redhat.com>
3 * configure.in: Bump version
5 * === Released 2.9.2 ===
7 2006-01-05 Hans Breuer <hans@breuer.org>
9 * glib/makefile.msc.in : added gslice.obj
11 2006-01-05 Matthias Clasen <mclasen@redhat.com>
15 * glib/gmain.c: Clarify the documentation of source ids
16 a bit. (#325874, Dan Williams)
18 * configure.in: Fix another strict aliasing problem.
20 * glib/gthreadpool.c: Avoid use of varargs macro. (#325864,
23 2006-01-04 Manish Singh <yosh@gimp.org>
26 * glib/gstdio.h: don't macroized g_access, g_chdir, and g_unlink
27 either, since they have the same issue as g_rmdir. (Related to
30 Wed Jan 4 13:33:25 2006 Tim Janik <timj@gtk.org>
32 * glib/gslice.c (magazine_cache_pop_magazine): don't reverse chunk
33 order when creating magazines, so we hand out chunks with ascending
36 2006-01-04 Abel Cheung <maddog@linuxhall.org>
38 * configure.in: Added "ml" "zh_HK" to ALL_LINGUAS.
40 2006-01-03 Martyn Russell <martyn@imendio.com>
42 * docs/reference/glib/glib-sections.txt:
44 * glib/gthreadpool.[ch]:
45 - Added new API g_thread_pool_get_idle_time() and
46 g_thread_pool_set_idle_time(). (#324228).
48 * tests/threadpool-test.c:
49 - Updated test case to do thread pool sorting, thread pool with
50 no sorting and a thread pool with idle thread timeouts.
52 2006-01-03 Matthias Clasen <mclasen@redhat.com>
54 * glib/gmain.h: Add new functions here, too.
56 * glib/glib.symbols: Add new functions.
58 2005-12-20 Michael Meeks <michael.meeks@novell.com>
60 * glib/gmain.c (g_main_context_is_owner): new method
61 to determine if the current thread is the owner of the
64 2006-01-02 Matthias Clasen <mclasen@redhat.com>
68 * glib/gstdio.c (g_rmdir): Don't provide g_rmdir() as a macro
69 expanding to rmdir, since rmdir is not declared in a portable
70 system header we can include in gstdio.h. (#325249, Jani Monoses)
72 2006-01-01 Tor Lillqvist <tml@novell.com>
74 * glib/gspawn-win32.c (g_spawn_sync_utf8): Set the GIOChannels for
75 stdout and stderr to unbuffered. Otherwise the giochannel layer
76 will try to read from them regardless whether the
77 g_io_channel_win32_poll() call here has indicated
78 readability or not. (#325310)
80 2005-12-29 Matthias Clasen <mclasen@redhat.com>
82 * glib/gutils.c (glib_check_version): Fix a copy-and-paste error
83 in the docs. (#325273, Declan Naughton)
85 2005-12-27 Manish Singh <yosh@gimp.org>
87 * tests/run-collate-tests.sh: use LC_ALL instead of LC_COLLATE, to
88 make sure we really override things.
90 2005-12-27 Matthias Clasen <mclasen@redhat.com>
92 Fix #316221, Michal Benes, Stanislav Brabec;
94 * configure.in: Fix a strict aliasing problem in
95 g_static_mutex_get_mutex().
96 * glib/gthread.h: ...and in
97 g_static_mutex_get_mutex_impl_shortcut().
99 * glib/gdatasetprivate.h: Add a cast to silence compiler
100 warnings. (#321978, Andrew Paprocki)
102 Partial fix for bug #323937, Bogdan Nicula.
104 * configure.in: Check for malloc.h
105 * glib/gslice.c: Don't include malloc.h unconditionally.
107 2005-12-27 Manish Singh <yosh@gimp.org>
109 * tests/run-collate-tests.sh: export LC_COLLATE so it takes.
110 (#324950, Dan Yefimov)
112 2005-12-26 Matthias Clasen <mclasen@redhat.com>
114 * glib/gslice.c: On Win32, include process.h (#325015, Kazuki
117 2005-12-25 Matthias Clasen <mclasen@redhat.com>
119 * AUTHORS: Update my email
121 * tests/slice-test.c: Fix C99isms. (#324950, Dan Yefimov)
123 2005-12-21 Matthias Clasen <mclasen@redhat.com>
125 * README.in: Add some notes on when 'make check' may fail.
127 2005-12-20 Matthias Clasen <mclasen@redhat.com>
130 * glib/gthreadpool.h:
131 * glib/gthreadpool.c (g_thread_pool_set_sort_function): New function
132 to sort tasks pushed into a threadpool. (#324479, Martyn Russell)
134 * tests/threadpool-test.c: Test this.
136 Tue Dec 20 18:14:14 2005 Tim Janik <timj@imendio.com>
138 * glib/gslice.[hc]: added mem_error() and mem_assert() to test and
139 handle errors without depending on gmessage.c which might not be
140 setup when the error occours.
141 removed G_SLICE_CONFIG_ALWAYS_FREE config option, fixed the code so
142 always freeing can be achieved by adjusting the working set time to
143 0 with G_SLICE_CONFIG_WORKING_SET_MSECS.
144 added G_SLICE_CONFIG_COLOR_INCREMENT to test different color increments
145 (mainly 0 and 1). reduced the minimum block size to 128 bytes, to
146 minimize wastage if small amounts of differently sized structrues are
147 allocated, this does come at a performance cost of roughly 5% though.
148 fixed up block alignment calculation, so it works for varying
149 block sizes. only use strerror() not g_strerror() since the latter
150 depends on working GQuark and GSlice.
151 mem_error(): implemented in terms of fprintf and vfprintf.
153 * tests/slice-color.c: new program to test cache colorization effects.
155 * tests/slice-test.c: trade G_SLICE_CONFIG_ALWAYS_FREE for 0 duration
156 G_SLICE_CONFIG_WORKING_SET_MSECS.
158 2005-12-17 Matthias Clasen <mclasen@redhat.com>
160 * glib/goption.c (parse_short_option): Set an error in all
161 failure cases. (#324332, Tim-Philipp Müller)
163 2005-12-17 Sebastian Wilhelmi <seppi@seppi.de>
165 * glib/gatomic.c: Fix memory barrier position in g_atomic_int_get
166 and g_atomic_pointer_get. Add g_atomic_int_set and
167 g_atomic_pointer_set implementations for the !DEFINE_WITH_MUTEXES &&
168 G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case, as well as defining them
169 as functions (additionally to the macros in the header) for the
170 !G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case.
172 2005-12-16 Matthias Clasen <mclasen@redhat.com>
174 * glib/gmem.c (g_allocator_new): Don't return a pointer to
175 a const struct, since apps expect to be able to modify it.
176 (#324179, J. Ali Harlow)
178 Tue Dec 13 10:13:32 2005 Tim Janik <timj@imendio.com>
180 * glib/gatomic.h: added g_atomic_pointer_set() and g_atomic_int_set()
181 for systems where the initialization of atomic variables requires a
182 write memory barrier.
184 2005-12-09 Matthias Clasen <mclasen@redhat.com>
186 * configure.in: Bump version
188 * === Released 2.9.1 ===
190 2005-12-09 Alexander Larsson <alexl@redhat.com>
194 Use G_GNUC_WARN_UNUSED_RESULT on list functions that return
197 2005-12-08 Matthias Clasen <mclasen@redhat.com>
201 2005-12-07 Martyn Russell <martyn@imendio.com>
203 * glib/gasyncqueue.c:
204 - Call g_queue_insert_sorted() instead of duplicating the code.
205 - Call g_queue_sort() instead of duplicating the code.
206 - Invert sort function results to make sure the same sort function
207 gives the same results across glist, gslist, gqueue and
210 * tests/asyncqueue-test.c:
211 - Updated the sort function to reflect the example in the
212 documentation for gasyncqueue.c.
214 2005-12-07 Martyn Russell <martyn@imendio.com>
216 * docs/reference/glib/glib-sections.txt:
217 * docs/reference/glib/tmpl/linked_lists_double.sgml:
218 * docs/reference/glib/tmpl/linked_lists_single.sgml:
221 - Added g_list_insert_sorted_with_data () and
222 g_slist_insert_sorted_with_data ().
223 - Removed the extra check in g_list_sort() and g_slist_sort() for
224 GCompareDataFunc vs. GCompareFunc.
226 2005-12-07 Tor Lillqvist <tml@novell.com>
228 * glib-gettextize.in: Look up prefix at run-time on Win32,
229 assuming the standard directory structure with glib-gettextize in
232 2005-12-06 Matthias Clasen <mclasen@redhat.com>
234 * glib/gmem.h: Exempt GTK+ from the mem chunk deprecation, since
235 we need GTK+ 2.8 to compile against GLib 2.10.
237 2005-12-05 Manish Singh <yosh@gimp.org>
239 * tests/run-collate-test.sh: set LC_COLLATE, not LANG, to be sure
240 to override any user settings.
242 2005-12-05 Matthias Clasen <mclasen@redhat.com>
244 * configure.in: Revert an accidential commit
246 2005-12-05 Matthias Clasen <mclasen@redhat.com>
248 * glib/gutf8.c: Documentation updates. (#323291, Morten Welinder)
250 * tests/Makefile.am (TESTS_ENVIRONMENT):
251 * tests/gobject/Makefile.am (TESTS_ENVIRONMENT): Set
252 MALLOC_CHECK_ and MALLOC_PERTURB_
254 * tests/run-collate-tests.sh: Run the collation tests explicitly
255 in en_US locale. (#320463)
257 * glib/gmem.h: Really deprecate GMemChunk.
264 * glib/gstrfuncs.c: Add versioned deprecation docs.
266 2005-12-05 Martyn Russell <martyn@imendio.com>
268 * docs/reference/glib/glib-sections.txt:
269 * glib/gasyncqueue.[ch]:
270 - Added support for sorting async queues by with _push_sorted(),
271 _push_sorted_unlocked(), _sort() and _sort_unlocked() (#323047).
274 * tests/asyncqueue-test.c:
275 - Added test case for gasyncqueue.c
277 Mon Dec 5 15:53:20 2005 Tim Janik <timj@imendio.com>
279 * glib/gslice.c: implement chain walking for arbitrary ->next pointer
280 offsets in g_slice_free_chain_with_offset() based on a patch by behdad
281 in bug 323178. moved time consuming logic from g_slice_free() out of
282 the inner loop, so g_slice_free_chain_with_offset() provides a real
283 performance benefit over g_slice_free1() now.
285 * glib/gslice.h: renamed g_slice_free_chain() to
286 g_slice_free_chain_with_offset(). implemented g_slice_free_chain() as
287 a type-safe macro as suggested in bug 323178.
288 simplified the macro implementation of g_slice_free() and implemented
289 it in a type safe manner for all compliers as suggested by Morten
290 Welinder <mortenw@gnome.org>.
295 * glib/glib.symbols: s/g_slice_free_chain/g_slice_free_chain_with_offset/
297 2005-12-05 Matthias Clasen <mclasen@redhat.com>
299 * glib/gasyncqueue.c: Add some docs.
301 * tests/libmoduletestplugin_a.c: Fix compiler warnings.
303 * glib/gatomic.c: In the ia64 implementation, use
304 __sync builtin without _si or _di suffix. (#321229,
305 Stanislav Brabec, patch by Andreas Schwab)
307 2005-12-04 Behdad Esfahbod <behdad@gnome.org>
309 * glib/gslice.h: Remove comma at the end of enum.
311 2005-12-04 Matthias Clasen <mclasen@redhat.com>
313 Handle multiple user names with the same UID better.
314 (#319535, Laszlo Peter)
316 * glib/gutils.c (g_get_any_init_do): When determining user
317 data, first look up $LOGNAME. If the UID doesn't match
318 getuid(), fall back to the current behaviour of looking
319 up the user data based on getuid().
321 2005-12-04 Matthias Clasen <mclasen@redhat.com>
323 * glib/gmacros.h (G_GNUC_WARN_UNUSED_RESULT): Add a macro
324 to make gcc warn if a function result is ignored. (#145466,
325 Arjan van de Ven, Alex Larsson)
327 * glib/gmem.h: Add the new attribute to g_realloc and
330 2005-12-03 Matthias Clasen <mclasen@redhat.com>
333 * glib/gcache.h: Deprecate g_cache_value_foreach. (#322956,
336 * glib/gmappedfile.c: Make mapping of empty files
339 * glib/gfileutils.c: Don't fork a new process just to
340 fix the permissions of the created temp file. (#321318,
341 Alexis S. L. Carvalho)
343 2005-12-02 Matthias Clasen <mclasen@redhat.com>
345 * README.in: Add a note about Solaris threads.
348 * configure.in: Remove support for Solaris threads.
349 (#136971, Sebastian Wilhelmi, patch by Andrew Paprocki)
351 2005-12-02 Matthias Clasen <mclasen@redhat.com>
353 * configure.in: Define G_GUINT64_CONSTANT in analogy to
354 G_GINT64_CONSTANT. (#322568, Andrew Paprocki)
356 2005-12-02 Matthias Clasen <mclasen@redhat.com>
358 * glib/gslice.c: Win32 portability fixes and C99ism removal,
359 pointed out by Kazuki Iwamoto. (#323052)
361 2005-12-02 Matthias Clasen <mclasen@redhat.com>
363 * glib/gatomic.c (g_atomic_int_add): Add a missing volatile for
364 the IA64 implementation.
366 Fri Dec 2 16:18:09 2005 Tim Janik <timj@imendio.com>
368 * glib/gslice.c: some naming and type size fixups.
370 Fri Dec 2 13:08:58 2005 Tim Janik <timj@gtk.org>
372 * glib/gslice.h (g_slice_free): reworked GCC-specific type-safe macro
373 variant into something less verbose, but digestible for gcc-3.4.
375 Fri Dec 2 10:55:07 2005 Tim Janik <timj@imendio.com>
377 * tests/slice-test.c: extended to perform the benchmarking on the old
378 memchunk code if 'O' is selected.
380 * tests/memchunks.c: new file which contains the old GLib mem chunks
381 implementation with prefix old_mem_chunk_.
383 * tests/Makefile.am: added memchunks.c
385 Fri Dec 2 00:16:59 2005 Tim Janik <timj@imendio.com>
387 * glib/gslice.c: improved gettimeofday() timer resolution by maintaining
388 acceptable load distribution of the syscall.
389 (allocator_get_magazine_threshold): reduce minimum/single-thread magazine
390 sizes to approximately page_size/2.5 to avoid excessive startup allocations.
392 2005-12-01 Matthias Clasen <mclasen@redhat.com>
394 * tests/gobject/ifacecheck.c (test_iface_base_init): Don't
395 double-free base interface members.
397 * tests/Makefile.am: Remove duplicate variable.
399 Thu Dec 1 17:32:46 2005 Tim Janik <timj@imendio.com>
401 * glib/gslice.[hc]: new slice allocator implementation.
403 * tests/slice-test.c: added random slice allocation test.
405 * glib/gthread.[hc]: removed newly added private thread mem API.
407 * glib/gthreadinit.h:
410 * glib/gmem.c: divided glib threading initialisation into three phases,
411 initialisation where private keys and messaging are not available (only
412 needed by gmem.c), initialisation without messaging but private keys
413 available (gslice.c, gmessage.c), and full fledged initialisers that
414 server the rest of glib. initialisation functions got renamed to reflect
415 the limitations of their corresponding phases.
417 * glib/gmem.c: removed memchunk code, defer allocations to
420 * glib/gmem.[hc]: removed g_slice_* skeletons.
422 * glib/glib.symbols: added g_slice_* symbols.
424 * configure.in: check for availability of posix_memalign(3), memalign(3)
427 * glib/Makefile.am: added gslice.[hc].
429 2005-12-01 Tor Lillqvist <tml@novell.com>
431 * glib/gstdio.c (g_stat): In the Win32 implementation, strip
432 trailing slash(es) for non-root folders. stat() fails if non-root
433 folders are specified with trailing slashes. It's too much hassle
434 to demand that callers strip such slashes themselves, especially
435 as it is easy to get it wrong and strip the slash of a root
437 (g_rename): On NT-based Windows, use MoveFileEx() with
438 MOVEFILE_REPLACE_EXISTING to better match Unix behaviour.
440 2005-11-28 Matthias Clasen <mclasen@redhat.com>
442 Fix G_STMT_START / G_STMT_END on Solaris. (#321972,
445 * configure.in: Check whether do { } while (0) works.
447 * glib/gmacros.h: Use do { } while (0) for G_STMT_START /
448 G_STMT_END if it works.
450 2005-11-28 Matthias Clasen <mclasen@redhat.com>
452 * glib/gthread.c (g_static_rw_lock_wait, g_static_rw_lock_signal):
453 * glib/gnode.c (g_node_depth_traverse_level):
454 * glib/gmem.c (g_allocator_new):
455 * glib/ghash.c (g_hash_table_unref, g_hash_table_destroy)
456 (g_hash_table_foreach_remove_or_steal): Silence compiler
459 2005-11-27 Matthias Clasen <mclasen@redhat.com>
461 * glib/gunicollate.c (g_utf8_collate_key_for_filename):
462 Don't read beyond len. (#322520, Christian Persch)
464 Wed Nov 23 17:34:01 2005 Tim Janik <timj@imendio.com>
466 * glib/gdataset.c: access datalist flags via atomic pointer access
467 functions, instead of acquiring the dataset lock. this is faster and
468 also matches the atomic pointer readouts in gdatalistprivate.h and
469 direct pointer modifications required by gobject.c.
471 Wed Nov 23 13:35:31 2005 Tim Janik <timj@gtk.org>
473 * glib/gdataset.c: streamlined initialization code somewhat,
474 removed GData node cache.
476 2005-11-22 Matthias Clasen <mclasen@redhat.com>
478 * glib/ghash.c (g_hash_table_ref):
479 (g_hash_table_unref): Mark these as new API.
481 Avoid double locking in g_intern_string (#322133,
484 * glib/gdataset.c (g_quark_from_string_internal): New
485 internal function which factors out common parts of
486 g_quark_from[_static]_string.
487 (g_quark_from_string, g_quark_from_static_string):
488 Use g_quark_from_string_internal.
489 (g_intern_string, g_intern_static_string): Use
490 g_quark_from_string_internal, and only take the
492 (g_quark_new): Don't store the strings shifted by -1
493 in the g_quarks array.
494 (g_quark_to_string): Adapt to the previous change.
496 Tue Nov 22 14:04:26 2005 Tim Janik <timj@imendio.com>
500 g_hash_table_new_full(): create hash tables with a ref count of 1.
501 g_hash_table_ref(): atomically ref_count+=1
502 g_hash_table_unref(): atomically ref_count-=1, destroys hash table
503 when refcount reaches 0.
504 g_hash_table_destroy(): just destroy keys and values, unref by 1.
505 g_hash_table_insert():
506 g_hash_table_replace(): assert ref_count>0.
509 * glib/gatomic.c: added 'volatile' qualifier to all atomic pointer and
512 2005-11-20 Behdad Esfahbod <behdad@gnome.org>
514 * glib/guniprop.c (g_unichar_get_mirror_char): Remove unused
515 variables. (#321984, Andrew Paprocki)
517 2005-11-18 Matthias Clasen <mclasen@redhat.com>
519 * configure.in: Bump version
521 2005-11-17 Matthias Clasen <mclasen@redhat.com>
523 * === Released 2.9.0 ===
525 * glib/glib.symbols: Updates
527 * glib/gutf8.c (g_utf8_offset_to_pointer): Add some performance
528 hints to the docs. (Paolo Borelli)
532 * glib/gthread.c (g_thread_foreach): Mark as new api.
534 * README.in: Updates.
538 * glib/gdate.c (g_date_set_time_t):
539 (g_date_set_time_val): Add functions to set a date from
540 a time_t and from a GTimeVal.
541 (g_date_set_time): Deprecate. (#314067, Roger Leigh)
544 * tests/utf8-pointer.c: Unit tests for g_utf8_pointer_to_offset
545 and g_utf8_offset_to_pointer.
547 * glib/gutf8.c (g_utf8_pointer_to_offset)
548 (g_utf8_offset_to_pointer): Handle negative offsets, and use
549 "stutter stepping" for going backwards. (#320638, Larry
554 * glib/gthread.c: const correctness fixes, found
555 by Arjan van de Ven and gcc.
557 2005-11-16 Behdad Esfahbod <behdad@gnome.org>
559 * docs/reference/glib/tmpl/unicode.sgml: Correct typo on mentioning
560 @G_UNICODE_BREAK_UNKNOWN that should be #G_UNICODE_BREAK_UNKNOWN.
562 2005-11-09 Behdad Esfahbod <behdad@gnome.org>
564 * glib/guniprop.c: Use bit hacks instead when checking a general
565 category value against multiple values.
567 * glib/gutf8.c: Change ISO10646 to Unicode in docs.
569 2005-11-10 Simos Xenitellis <simos@gnome.org>
571 * configure.in: Added tt (Tatar) to ALL_LINGUAS.
573 2005-11-05 Tor Lillqvist <tml@novell.com>
575 * glib/giowin32.c (g_io_win32_set_flags): Don't call g_warning().
576 (#320688) Instead set error and return failure.
578 2005-11-07 Matthias Clasen <mclasen@redhat.com>
580 * glib/gmain.c (g_main_dispatch): Don't call cb_funcs->unref
581 while holding the context lock. (#320886, Andy Wingo)
583 2005-11-07 Billy Biggs <vektor@dumbterm.net>
585 * tests/refcount/Makefile.am: Add a missing reference to libglib
586 in the LDADD for this test.
588 2005-11-07 Matthias Clasen <mclasen@redhat.com>
590 * glib/gmem.c (g_allocator_new): We need to set n_preallocs to a
591 nonzero value, otherwise GTK+ 2.8 breaks when compiled against
592 GLib 2.9. (#320755, Luca Ferretti)
594 2005-11-04 Matthias Clasen <mclasen@redhat.com>
596 * glib/Makefile.am: Apply a patch from Behdad Esfahbod to
597 use a faster lookup table for g_unichar_get_mirror_char().
598 * glib/gmirroringtable.h: The new table.
599 * glib/gunichartables.h: Remove bidi_mirroring_table.
600 * glib/gen-unicode-tables.pl: Don't generate the mirroring
602 * glib/glib-mirroring-tab/*: Sources for the program
603 which generated gmirroringtable.h.
605 * glib/glist.c: Avoid some code duplication.
607 * glib/gscanner.h: Include gdataset.h. (#320322)
609 * glib/gdebug.h: add new GDebugFlag for fatal_criticals
610 * glib/gmessages.c: (_g_debug_init): handle G_DEBUG=fatal_criticals,
611 to help find critical warnings in applications. (#320017,
614 2005-11-02 Tor Lillqvist <tml@novell.com>
616 * glib/glib.symbols: Remove large amount of trailing whitespace
617 from one line. Remove #ifdef G_UNDEFINED from around
618 g_hash_table_new and g_hash_table_new_full symbols, what was the
619 purpose of that? Add the g_slice functions.
621 Tue Nov 1 16:24:20 2005 Tim Janik <timj@imendio.com>
623 * glib/gmem.[hc]: prepared deprecation of GMemChunk and GAllocator.
624 added g_slice_*() API to allocate and cache small bits of memory.
625 an actuall allocator implementation for g_slice_*() is still pending.
627 * glib/gthread.[hc]: changes from a patch by Matthias Clasen.
628 changed GRealThread list to use in-structure *next; fields instead
629 of GSList, in order for thread iteration to not depenend on g_slice_*()
631 _g_thread_mem_private_get():
632 _g_thread_mem_private_set(): added accessors for private memory,
633 needed because the ordinary GPrivate implementation relies on GArray
634 and GSList and therefore indirectly on working g_slice_*() allocations.
637 g_thread_foreach(): new public API function to loop over all existing threads.
651 removed GAllocator and free list usages and accompanying locks.
652 use g_slice_*() API to allocate and cache small bits of memory.
654 * glib/ghook.h: removed GMemChunk field from public API.
657 * glib/glist.h: deprecate allocator API, provide _free1() for consistency.
659 * glib/gnode.h: deprecate allocator API.
661 * glib/gmain.c: reordered GPollRec fields so g_slice_free_chain() can
662 be used for poll rec lists.
664 * glib/grel.c: removed mem chunk usage, and allocated tuples via g_slice_*().
665 g_relation_destroy(): free all tuples from the all_tuples hash table,
666 this effectively maintains the life time track keeping of tuples.
667 g_relation_delete_tuple(): free tuples which are removed from the
668 all_tuples hash table. this fixes a temporary leak that was present
669 in the memchunk code until the destruction of the relation.
671 2005-10-29 Matthias Clasen <mclasen@redhat.com>
673 * tests/convert-test.c: Add some tests for conversions between
674 UTF-8, UCS-4 and UTF-16.
676 * glib/gutf8.c (g_utf8_to_ucs4, g_utf8_to_utf16): Fix handling
677 of len == -1, noticed by Morten Welinder.
679 2005-10-27 Erdal Ronahi <erdal.ronahi@gmail.com>
681 * configure.in: Added ku (Kurdish) to ALL_LINGUAS
683 2005-10-26 Matthias Clasen <mclasen@redhat.com>
685 * glib/gutf8.c (g_ucs4_to_utf8): Don't set items_read twice
686 in the error case, and add some documentation. (#319806, Morten
689 2005-10-19 Manish Singh <yosh@gimp.org>
691 * configure.in: Use AC_CHECK_FUNCS for _NSGetEnviron, to get the
692 config.h symbol automatically. Fixes bug #313731.
694 2005-10-19 Tor Lillqvist <tml@novell.com>
696 * glib/gwin32.c (g_win32_get_package_installation_directory):
697 Return a g_strdup()ed copy of the value stored in the hash table,
698 so that it can be g_free()d without leaving a dangling pointer in
699 the hash table. (#319232)
701 2005-10-06 Matthias Clasen <mclasen@redhat.com>
703 * glib/gunicollate.c (g_utf8_collate_key_for_filename): Handle
704 all-zero sequences correctly. (#317930, Sebastien Bacher)
706 2005-10-05 Matthias Clasen <mclasen@redhat.com>
708 * glib/gmarkup.c (g_markup_escape_text): Doc addition
710 * Makefile.am: Add Collation tests.
712 * tests/collate/*: Inputs and expected outputs for collation tests.
714 * tests/run-collate-tests.sh: Script to run collation tests.
716 * tests/unicode-collate.c (main): Rework slightly to make
717 it usable in unit tests. Also test g_utf8_collate_key_for_filename().
719 2005-10-01 Behdad Esfahbod <behdad@gnome.org>
721 * docs/reference/glib/tmpl/unicode.sgml:
722 * glib/gen-unicode-tables.pl:
724 * glib/gunichartables.h:
726 * tests/casefold.txt:
727 * tests/casemap.txt: Updated to Unicode 4.1. There are five new
728 GUnicodeBreakType types. That may break some applications, like
731 2005-09-26 Matthias Clasen <mclasen@redhat.com>
733 * glib/gstrfuncs.c (g_ascii_strtoull): Add details to the
734 docs. (#314393, Matthew F. Barnes)
737 * glib/gprintf.h: Remove g_snprintf() and g_vsnprintf(), since
738 they are already declared in glib.h. This doesn't break documented
739 use of gprintf.h, but should probably be pointed out in the
740 release notes for 2.10. (#314232, Behdad Esfahbod)
742 Tue Sep 20 13:16:04 2005 Tim Janik <timj@imendio.com>
744 * glib/gpattern.c (g_pattern_ph_match): applied significant recursion
745 complexity optimization, based on a patch from Matthias Clasen.
747 * tests/patterntest.c: more tests, mostly from matthias.
749 2005-09-20 Matthias Clasen <mclasen@redhat.com>
751 * glib/gqueue.c (g_queue_insert_sorted): Correct the docs.
752 (#316703, Mark Drago)
754 Mon Sep 19 17:23:23 2005 Tim Janik <timj@imendio.com>
756 * glib/gpattern.c: applied a patch from matthias which checks on the
757 upper bound of GPatternSpec length to optimize matches.
760 * tests/patterntest.c: added more match cases.
762 2005-09-16 Tor Lillqvist <tml@novell.com>
764 * glib/gstrfuncs.c (g_ascii_strcasecmp, g_ascii_strncasecmp): Add
765 warning to doc comment that these functions should not be used on
766 encodings like CP932.
768 2005-09-14 Matthias Clasen <mclasen@redhat.com>
770 * tests/keyfile-test.c: Add a test for grup names of length 1.
772 * glib/gkeyfile.c (g_key_file_line_is_group): Accept group names
773 of length 1. (#316309)
775 2005-09-12 Matthias Clasen <mclasen@redhat.com>
777 * glib/gmarkup.c (g_markup_escape_text): Clarify docs.
779 2005-09-11 Sebastian Wilhelmi <seppi@seppi.de>
781 * tests/refcount/Makefile.am (INCLUDES): Link the the refcount
782 tests to the system thread library $(G_THREAD_LIBS). Fixes #313744
785 2005-09-11 Kjartan Maraas <kmaraas@gnome.org>
787 * glib/gmain.c: (g_child_watch_prepare), (g_child_watch_check),
788 (child_watch_helper_thread): Remove some dead code. Closes
791 2005-09-07 Tor Lillqvist <tml@novell.com>
793 * glib/Makefile.am: Create also a console version of the
794 gspawn-win32-helper program, gspawn-win32-helper-console.exe.
795 It's otherwise identical to gspawn-win32-helper.exe, except marked
796 as a console application (linked without the -mwindows option).
798 * glib/gspawn-win32.c (do_spawn_directly, do_spawn_with_pipes):
799 Drop the dont_wait parameter. Its truth value correlated 100% with
800 the NULLness of the exit_status parameter anyway, so it's enough
801 to check whether exit_status is NULL. Invert the sense of the
802 dont_return_handle parameter and rename it to do_return_handle, to
803 make the code easier to read by avoiding double negations.
805 (g_spawn_sync_utf8, g_spawn_async_with_pipes_utf8): Modify calls
806 to do_spawn_with_pipes() accordingly.
808 (do_spawn_with_pipes): If we have a console, use the console
809 version of the helper program, otherwise use the GUI one. This
810 avoids extra console windows opening up in some situations. (In
811 case a console application uses the GUI gspawn-win32-helper.exe to
812 spawn another console application we would get a separate console
813 for the spawned console application).
815 * glib-zip.in: Distribute also gspawn-win32-helper-console.exe.
817 2005-09-05 Matthias Clasen <mclasen@redhat.com>
819 * glib/gmappedfile.c (g_mapped_file_new): Report an error
820 if the file is too large. (#315275, Kjartan Maraas)
822 * glib/gkeyfile.c (g_key_file_load_from_fd): The return value
823 of read() is signed. (#315273, Kjartan Maraas)
825 2005-08-31 Tor Lillqvist <tml@novell.com>
827 * glib/gutils.h: Wrapping atexit() is a bad idea on Windows, where
828 the EXE and each DLL have their own atexit function chains.
830 #define g_atexit as atexit instead. This means it has a
831 better chance of doing what the caller wants. For instance,
832 gtkhtml calls g_atexit() registering a function in gtkhtml
833 itself. This caused a crash when g_atexit() was implemented as a
834 function in the GLib DLL. The gtkhtml DLL was already unloaded by
835 the time the GLib DLL got unloaded.
837 * glib/gutils.c: #undef the #define mentioned above, to also get a
838 real g_atexit() into the DLL for backward compatibility. Document
839 the Windows behaviour of g_atexit(), and document the varying ways
840 atexit() can behave in the context of dynamically loaded modules
843 2005-08-31 Matthias Clasen <mclasen@redhat.com>
847 * glib/gdataset.c: Add string interning functions.
849 2005-08-28 Matthias Clasen <mclasen@redhat.com>
851 * glib/giochannel.c: Unify some near-duplicate strings. (#314654,
854 2005-08-26 Matthias Clasen <mclasen@redhat.com>
856 * configure.in: Bump version to 2.9.0
858 2005-08-25 Tor Lillqvist <tml@novell.com>
860 Make also the g_spawn*() functions take parameters in the GLib
861 file name encoding, i.e. UTF-8, on Windows. Has no impact on Unix
862 API or ABI. Like the other GLib API that was earlier changed to
863 use UTF-8 on Windows, the names of the functions that take UTF-8
864 have _utf8 suffixes added by using preprocessor macros in the
865 header file. The old names are kept for functions with the old
866 behaviour, taking parameters in the system codepage, for DLL ABI
869 * glib/gspawn.h: On Win32 add the suffix _utf8 to the names of the
870 g_spawn*() functions.
872 * glib/gspawn-win32.c: Use wide-char API on NT-based
873 Windows. Convert parameters from UTF-8 to wide chars (NT) or
874 system codepage (Win9x) and call the C library _wspawn*() or
875 spawn*() functions respectvely. Add DLL ABI stability versions
876 that take parameters in the system codepage.
878 * glib/gspawn-win32-helper.c: On NT-based Windows use the
879 wide-char versions of argv and envp, and use wide-char API to
880 change directory and spawn the program to run. Remove the verbose
881 debugging output, it was too complex to modify for the wide-char
882 features. (Just add temporary debugging printouts if needed, no
883 need to have them permanently in the source.)
885 * glib/gspawn.c: Corresponding documentation updates.
887 * glib/glib.symbols: Corresponding changes: Mark the ABI stability
888 symbols as PRIVATE, add the new _utf8-suffixed ones.
890 2005-08-24 Stepan Kasal <kasal@ucw.cz>
892 * glib/gtypes.h (G_MININT64): Cast the constant to gint64; it is
893 guint64 otherwise and that can produce warnings about comparison
894 between signed and unsigned.
896 2005-08-23 Matthias Clasen <mclasen@redhat.com>
898 * glib/gutils.c: Fix the crt_externs.h include.
900 2005-08-23 Stepan Kasal <kasal@ucw.cz>
902 * NEWS: Fix spelling of my first name.
904 2005-08-23 Matthias Clasen <mclasen@redhat.com>
908 * === Released 2.8.1 ===
912 2005-08-20 Hans Breuer <hans@breuer.org>
914 * glib/makefile.msc.in : link with ws2_32.lib
916 2005-08-18 Tor Lillqvist <tml@novell.com>
918 * configure.in: Check for <sys/wait.h>
920 * glib/gbacktrace.c: Include <sys/wait.h> on if HAVE_SYS_WAIT_H.
922 2005-08-18 Ross Burton <ross@burtonini.com>
925 Optimise single-character insertions.
930 * tests/string-test.c:
931 Add tests for new optimisation, and fix a leak.
933 2005-08-17 Matthias Clasen <mclasen@redhat.com>
935 * configure.in: Check for crt_externs.h and _NSGetEnviron.
937 * glib/gutils.c: On Darwin, include crt-externs.h and
938 define environ using _NSGetEnviron(). (#313731)
940 2005-08-16 Stepan Kasal <kasal@ucw.cz>
942 * glib/gutils.c (g_get_any_init): Move the body of the big if...
943 (g_get_any_init_do): ... to this new function.
944 (g_get_any_init): Declare as inline.
945 (g_get_any_init_locked): New inline function, does the locking.
946 Make use of these two throughout the code.
948 2005-08-15 Matthias Clasen <mclasen@redhat.com>
950 * glib/gbacktrace.c (g_on_error_stack_trace): Wait for
951 the child process and then simply return. This makes
952 The "S" option work as documented in g_on_error_query().
953 (#313125, Matthew F. Barnes)
955 * glib/gunicode.h: Update the link to Unicode category
956 values. (#313369, Behnam Esfahbod)
958 * glib/gqueue.c (g_queue_find_custom): Clarify docs
959 a little. (#311727, Tristan van Berkom)
961 * glib/abicheck.sh, gobject/abicheck.sh: Make the
962 check work on ia64 too, where some symbols ended up
963 in yet another section.