1 2006-01-12 Matthias Clasen <mclasen@redhat.com>
3 * tests/asyncqueue-test.c (main): Fix the
4 assert to not always trigger. (#326558,
7 2006-01-11 Matthias Clasen <mclasen@redhat.com>
9 * tests/convert-test.c: Don't test items_read and
10 items_written in error cases where it is not set.
12 * tests/asyncqueue-test.c: Set error to NULL.
14 2006-01-08 Matthias Clasen <mclasen@redhat.com>
16 * glib/gtree.c: Break some long lines.
18 * tests/tree-test.c: Add more tests.
20 * glib/gslice.c: Include config.h first, pointed out
23 2005-12-20 Sven Herzberg <herzi@gnome-de.org>
25 * docs/reference/gobject/tmpl/gtype.sgml: explain that
26 G_TYPE_INSTANCE_GET_CLASS() does behave different during
29 2006-01-06 Matthias Clasen <mclasen@redhat.com>
31 * glib/gslice.c: Only define _XOPEN_SOURCE if we know
32 that we have posix_memalign(). (#323937, Bogdan Nicula)
34 2006-01-05 Matthias Clasen <mclasen@redhat.com>
36 * configure.in: Bump version
38 * === Released 2.9.2 ===
40 2006-01-05 Hans Breuer <hans@breuer.org>
42 * glib/makefile.msc.in : added gslice.obj
44 2006-01-05 Matthias Clasen <mclasen@redhat.com>
48 * glib/gmain.c: Clarify the documentation of source ids
49 a bit. (#325874, Dan Williams)
51 * configure.in: Fix another strict aliasing problem.
53 * glib/gthreadpool.c: Avoid use of varargs macro. (#325864,
56 2006-01-04 Manish Singh <yosh@gimp.org>
59 * glib/gstdio.h: don't macroized g_access, g_chdir, and g_unlink
60 either, since they have the same issue as g_rmdir. (Related to
63 Wed Jan 4 13:33:25 2006 Tim Janik <timj@gtk.org>
65 * glib/gslice.c (magazine_cache_pop_magazine): don't reverse chunk
66 order when creating magazines, so we hand out chunks with ascending
69 2006-01-04 Abel Cheung <maddog@linuxhall.org>
71 * configure.in: Added "ml" "zh_HK" to ALL_LINGUAS.
73 2006-01-03 Martyn Russell <martyn@imendio.com>
75 * docs/reference/glib/glib-sections.txt:
77 * glib/gthreadpool.[ch]:
78 - Added new API g_thread_pool_get_idle_time() and
79 g_thread_pool_set_idle_time(). (#324228).
81 * tests/threadpool-test.c:
82 - Updated test case to do thread pool sorting, thread pool with
83 no sorting and a thread pool with idle thread timeouts.
85 2006-01-03 Matthias Clasen <mclasen@redhat.com>
87 * glib/gmain.h: Add new functions here, too.
89 * glib/glib.symbols: Add new functions.
91 2005-12-20 Michael Meeks <michael.meeks@novell.com>
93 * glib/gmain.c (g_main_context_is_owner): new method
94 to determine if the current thread is the owner of the
97 2006-01-02 Matthias Clasen <mclasen@redhat.com>
101 * glib/gstdio.c (g_rmdir): Don't provide g_rmdir() as a macro
102 expanding to rmdir, since rmdir is not declared in a portable
103 system header we can include in gstdio.h. (#325249, Jani Monoses)
105 2006-01-01 Tor Lillqvist <tml@novell.com>
107 * glib/gspawn-win32.c (g_spawn_sync_utf8): Set the GIOChannels for
108 stdout and stderr to unbuffered. Otherwise the giochannel layer
109 will try to read from them regardless whether the
110 g_io_channel_win32_poll() call here has indicated
111 readability or not. (#325310)
113 2005-12-29 Matthias Clasen <mclasen@redhat.com>
115 * glib/gutils.c (glib_check_version): Fix a copy-and-paste error
116 in the docs. (#325273, Declan Naughton)
118 2005-12-27 Manish Singh <yosh@gimp.org>
120 * tests/run-collate-tests.sh: use LC_ALL instead of LC_COLLATE, to
121 make sure we really override things.
123 2005-12-27 Matthias Clasen <mclasen@redhat.com>
125 Fix #316221, Michal Benes, Stanislav Brabec;
127 * configure.in: Fix a strict aliasing problem in
128 g_static_mutex_get_mutex().
129 * glib/gthread.h: ...and in
130 g_static_mutex_get_mutex_impl_shortcut().
132 * glib/gdatasetprivate.h: Add a cast to silence compiler
133 warnings. (#321978, Andrew Paprocki)
135 Partial fix for bug #323937, Bogdan Nicula.
137 * configure.in: Check for malloc.h
138 * glib/gslice.c: Don't include malloc.h unconditionally.
140 2005-12-27 Manish Singh <yosh@gimp.org>
142 * tests/run-collate-tests.sh: export LC_COLLATE so it takes.
143 (#324950, Dan Yefimov)
145 2005-12-26 Matthias Clasen <mclasen@redhat.com>
147 * glib/gslice.c: On Win32, include process.h (#325015, Kazuki
150 2005-12-25 Matthias Clasen <mclasen@redhat.com>
152 * AUTHORS: Update my email
154 * tests/slice-test.c: Fix C99isms. (#324950, Dan Yefimov)
156 2005-12-21 Matthias Clasen <mclasen@redhat.com>
158 * README.in: Add some notes on when 'make check' may fail.
160 2005-12-20 Matthias Clasen <mclasen@redhat.com>
163 * glib/gthreadpool.h:
164 * glib/gthreadpool.c (g_thread_pool_set_sort_function): New function
165 to sort tasks pushed into a threadpool. (#324479, Martyn Russell)
167 * tests/threadpool-test.c: Test this.
169 Tue Dec 20 18:14:14 2005 Tim Janik <timj@imendio.com>
171 * glib/gslice.[hc]: added mem_error() and mem_assert() to test and
172 handle errors without depending on gmessage.c which might not be
173 setup when the error occours.
174 removed G_SLICE_CONFIG_ALWAYS_FREE config option, fixed the code so
175 always freeing can be achieved by adjusting the working set time to
176 0 with G_SLICE_CONFIG_WORKING_SET_MSECS.
177 added G_SLICE_CONFIG_COLOR_INCREMENT to test different color increments
178 (mainly 0 and 1). reduced the minimum block size to 128 bytes, to
179 minimize wastage if small amounts of differently sized structrues are
180 allocated, this does come at a performance cost of roughly 5% though.
181 fixed up block alignment calculation, so it works for varying
182 block sizes. only use strerror() not g_strerror() since the latter
183 depends on working GQuark and GSlice.
184 mem_error(): implemented in terms of fprintf and vfprintf.
186 * tests/slice-color.c: new program to test cache colorization effects.
188 * tests/slice-test.c: trade G_SLICE_CONFIG_ALWAYS_FREE for 0 duration
189 G_SLICE_CONFIG_WORKING_SET_MSECS.
191 2005-12-17 Matthias Clasen <mclasen@redhat.com>
193 * glib/goption.c (parse_short_option): Set an error in all
194 failure cases. (#324332, Tim-Philipp Müller)
196 2005-12-17 Sebastian Wilhelmi <seppi@seppi.de>
198 * glib/gatomic.c: Fix memory barrier position in g_atomic_int_get
199 and g_atomic_pointer_get. Add g_atomic_int_set and
200 g_atomic_pointer_set implementations for the !DEFINE_WITH_MUTEXES &&
201 G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case, as well as defining them
202 as functions (additionally to the macros in the header) for the
203 !G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case.
205 2005-12-16 Matthias Clasen <mclasen@redhat.com>
207 * glib/gmem.c (g_allocator_new): Don't return a pointer to
208 a const struct, since apps expect to be able to modify it.
209 (#324179, J. Ali Harlow)
211 Tue Dec 13 10:13:32 2005 Tim Janik <timj@imendio.com>
213 * glib/gatomic.h: added g_atomic_pointer_set() and g_atomic_int_set()
214 for systems where the initialization of atomic variables requires a
215 write memory barrier.
217 2005-12-09 Matthias Clasen <mclasen@redhat.com>
219 * configure.in: Bump version
221 * === Released 2.9.1 ===
223 2005-12-09 Alexander Larsson <alexl@redhat.com>
227 Use G_GNUC_WARN_UNUSED_RESULT on list functions that return
230 2005-12-08 Matthias Clasen <mclasen@redhat.com>
234 2005-12-07 Martyn Russell <martyn@imendio.com>
236 * glib/gasyncqueue.c:
237 - Call g_queue_insert_sorted() instead of duplicating the code.
238 - Call g_queue_sort() instead of duplicating the code.
239 - Invert sort function results to make sure the same sort function
240 gives the same results across glist, gslist, gqueue and
243 * tests/asyncqueue-test.c:
244 - Updated the sort function to reflect the example in the
245 documentation for gasyncqueue.c.
247 2005-12-07 Martyn Russell <martyn@imendio.com>
249 * docs/reference/glib/glib-sections.txt:
250 * docs/reference/glib/tmpl/linked_lists_double.sgml:
251 * docs/reference/glib/tmpl/linked_lists_single.sgml:
254 - Added g_list_insert_sorted_with_data () and
255 g_slist_insert_sorted_with_data ().
256 - Removed the extra check in g_list_sort() and g_slist_sort() for
257 GCompareDataFunc vs. GCompareFunc.
259 2005-12-07 Tor Lillqvist <tml@novell.com>
261 * glib-gettextize.in: Look up prefix at run-time on Win32,
262 assuming the standard directory structure with glib-gettextize in
265 2005-12-06 Matthias Clasen <mclasen@redhat.com>
267 * glib/gmem.h: Exempt GTK+ from the mem chunk deprecation, since
268 we need GTK+ 2.8 to compile against GLib 2.10.
270 2005-12-05 Manish Singh <yosh@gimp.org>
272 * tests/run-collate-test.sh: set LC_COLLATE, not LANG, to be sure
273 to override any user settings.
275 2005-12-05 Matthias Clasen <mclasen@redhat.com>
277 * configure.in: Revert an accidential commit
279 2005-12-05 Matthias Clasen <mclasen@redhat.com>
281 * glib/gutf8.c: Documentation updates. (#323291, Morten Welinder)
283 * tests/Makefile.am (TESTS_ENVIRONMENT):
284 * tests/gobject/Makefile.am (TESTS_ENVIRONMENT): Set
285 MALLOC_CHECK_ and MALLOC_PERTURB_
287 * tests/run-collate-tests.sh: Run the collation tests explicitly
288 in en_US locale. (#320463)
290 * glib/gmem.h: Really deprecate GMemChunk.
297 * glib/gstrfuncs.c: Add versioned deprecation docs.
299 2005-12-05 Martyn Russell <martyn@imendio.com>
301 * docs/reference/glib/glib-sections.txt:
302 * glib/gasyncqueue.[ch]:
303 - Added support for sorting async queues by with _push_sorted(),
304 _push_sorted_unlocked(), _sort() and _sort_unlocked() (#323047).
307 * tests/asyncqueue-test.c:
308 - Added test case for gasyncqueue.c
310 Mon Dec 5 15:53:20 2005 Tim Janik <timj@imendio.com>
312 * glib/gslice.c: implement chain walking for arbitrary ->next pointer
313 offsets in g_slice_free_chain_with_offset() based on a patch by behdad
314 in bug 323178. moved time consuming logic from g_slice_free() out of
315 the inner loop, so g_slice_free_chain_with_offset() provides a real
316 performance benefit over g_slice_free1() now.
318 * glib/gslice.h: renamed g_slice_free_chain() to
319 g_slice_free_chain_with_offset(). implemented g_slice_free_chain() as
320 a type-safe macro as suggested in bug 323178.
321 simplified the macro implementation of g_slice_free() and implemented
322 it in a type safe manner for all compliers as suggested by Morten
323 Welinder <mortenw@gnome.org>.
328 * glib/glib.symbols: s/g_slice_free_chain/g_slice_free_chain_with_offset/
330 2005-12-05 Matthias Clasen <mclasen@redhat.com>
332 * glib/gasyncqueue.c: Add some docs.
334 * tests/libmoduletestplugin_a.c: Fix compiler warnings.
336 * glib/gatomic.c: In the ia64 implementation, use
337 __sync builtin without _si or _di suffix. (#321229,
338 Stanislav Brabec, patch by Andreas Schwab)
340 2005-12-04 Behdad Esfahbod <behdad@gnome.org>
342 * glib/gslice.h: Remove comma at the end of enum.
344 2005-12-04 Matthias Clasen <mclasen@redhat.com>
346 Handle multiple user names with the same UID better.
347 (#319535, Laszlo Peter)
349 * glib/gutils.c (g_get_any_init_do): When determining user
350 data, first look up $LOGNAME. If the UID doesn't match
351 getuid(), fall back to the current behaviour of looking
352 up the user data based on getuid().
354 2005-12-04 Matthias Clasen <mclasen@redhat.com>
356 * glib/gmacros.h (G_GNUC_WARN_UNUSED_RESULT): Add a macro
357 to make gcc warn if a function result is ignored. (#145466,
358 Arjan van de Ven, Alex Larsson)
360 * glib/gmem.h: Add the new attribute to g_realloc and
363 2005-12-03 Matthias Clasen <mclasen@redhat.com>
366 * glib/gcache.h: Deprecate g_cache_value_foreach. (#322956,
369 * glib/gmappedfile.c: Make mapping of empty files
372 * glib/gfileutils.c: Don't fork a new process just to
373 fix the permissions of the created temp file. (#321318,
374 Alexis S. L. Carvalho)
376 2005-12-02 Matthias Clasen <mclasen@redhat.com>
378 * README.in: Add a note about Solaris threads.
381 * configure.in: Remove support for Solaris threads.
382 (#136971, Sebastian Wilhelmi, patch by Andrew Paprocki)
384 2005-12-02 Matthias Clasen <mclasen@redhat.com>
386 * configure.in: Define G_GUINT64_CONSTANT in analogy to
387 G_GINT64_CONSTANT. (#322568, Andrew Paprocki)
389 2005-12-02 Matthias Clasen <mclasen@redhat.com>
391 * glib/gslice.c: Win32 portability fixes and C99ism removal,
392 pointed out by Kazuki Iwamoto. (#323052)
394 2005-12-02 Matthias Clasen <mclasen@redhat.com>
396 * glib/gatomic.c (g_atomic_int_add): Add a missing volatile for
397 the IA64 implementation.
399 Fri Dec 2 16:18:09 2005 Tim Janik <timj@imendio.com>
401 * glib/gslice.c: some naming and type size fixups.
403 Fri Dec 2 13:08:58 2005 Tim Janik <timj@gtk.org>
405 * glib/gslice.h (g_slice_free): reworked GCC-specific type-safe macro
406 variant into something less verbose, but digestible for gcc-3.4.
408 Fri Dec 2 10:55:07 2005 Tim Janik <timj@imendio.com>
410 * tests/slice-test.c: extended to perform the benchmarking on the old
411 memchunk code if 'O' is selected.
413 * tests/memchunks.c: new file which contains the old GLib mem chunks
414 implementation with prefix old_mem_chunk_.
416 * tests/Makefile.am: added memchunks.c
418 Fri Dec 2 00:16:59 2005 Tim Janik <timj@imendio.com>
420 * glib/gslice.c: improved gettimeofday() timer resolution by maintaining
421 acceptable load distribution of the syscall.
422 (allocator_get_magazine_threshold): reduce minimum/single-thread magazine
423 sizes to approximately page_size/2.5 to avoid excessive startup allocations.
425 2005-12-01 Matthias Clasen <mclasen@redhat.com>
427 * tests/gobject/ifacecheck.c (test_iface_base_init): Don't
428 double-free base interface members.
430 * tests/Makefile.am: Remove duplicate variable.
432 Thu Dec 1 17:32:46 2005 Tim Janik <timj@imendio.com>
434 * glib/gslice.[hc]: new slice allocator implementation.
436 * tests/slice-test.c: added random slice allocation test.
438 * glib/gthread.[hc]: removed newly added private thread mem API.
440 * glib/gthreadinit.h:
443 * glib/gmem.c: divided glib threading initialisation into three phases,
444 initialisation where private keys and messaging are not available (only
445 needed by gmem.c), initialisation without messaging but private keys
446 available (gslice.c, gmessage.c), and full fledged initialisers that
447 server the rest of glib. initialisation functions got renamed to reflect
448 the limitations of their corresponding phases.
450 * glib/gmem.c: removed memchunk code, defer allocations to
453 * glib/gmem.[hc]: removed g_slice_* skeletons.
455 * glib/glib.symbols: added g_slice_* symbols.
457 * configure.in: check for availability of posix_memalign(3), memalign(3)
460 * glib/Makefile.am: added gslice.[hc].
462 2005-12-01 Tor Lillqvist <tml@novell.com>
464 * glib/gstdio.c (g_stat): In the Win32 implementation, strip
465 trailing slash(es) for non-root folders. stat() fails if non-root
466 folders are specified with trailing slashes. It's too much hassle
467 to demand that callers strip such slashes themselves, especially
468 as it is easy to get it wrong and strip the slash of a root
470 (g_rename): On NT-based Windows, use MoveFileEx() with
471 MOVEFILE_REPLACE_EXISTING to better match Unix behaviour.
473 2005-11-28 Matthias Clasen <mclasen@redhat.com>
475 Fix G_STMT_START / G_STMT_END on Solaris. (#321972,
478 * configure.in: Check whether do { } while (0) works.
480 * glib/gmacros.h: Use do { } while (0) for G_STMT_START /
481 G_STMT_END if it works.
483 2005-11-28 Matthias Clasen <mclasen@redhat.com>
485 * glib/gthread.c (g_static_rw_lock_wait, g_static_rw_lock_signal):
486 * glib/gnode.c (g_node_depth_traverse_level):
487 * glib/gmem.c (g_allocator_new):
488 * glib/ghash.c (g_hash_table_unref, g_hash_table_destroy)
489 (g_hash_table_foreach_remove_or_steal): Silence compiler
492 2005-11-27 Matthias Clasen <mclasen@redhat.com>
494 * glib/gunicollate.c (g_utf8_collate_key_for_filename):
495 Don't read beyond len. (#322520, Christian Persch)
497 Wed Nov 23 17:34:01 2005 Tim Janik <timj@imendio.com>
499 * glib/gdataset.c: access datalist flags via atomic pointer access
500 functions, instead of acquiring the dataset lock. this is faster and
501 also matches the atomic pointer readouts in gdatalistprivate.h and
502 direct pointer modifications required by gobject.c.
504 Wed Nov 23 13:35:31 2005 Tim Janik <timj@gtk.org>
506 * glib/gdataset.c: streamlined initialization code somewhat,
507 removed GData node cache.
509 2005-11-22 Matthias Clasen <mclasen@redhat.com>
511 * glib/ghash.c (g_hash_table_ref):
512 (g_hash_table_unref): Mark these as new API.
514 Avoid double locking in g_intern_string (#322133,
517 * glib/gdataset.c (g_quark_from_string_internal): New
518 internal function which factors out common parts of
519 g_quark_from[_static]_string.
520 (g_quark_from_string, g_quark_from_static_string):
521 Use g_quark_from_string_internal.
522 (g_intern_string, g_intern_static_string): Use
523 g_quark_from_string_internal, and only take the
525 (g_quark_new): Don't store the strings shifted by -1
526 in the g_quarks array.
527 (g_quark_to_string): Adapt to the previous change.
529 Tue Nov 22 14:04:26 2005 Tim Janik <timj@imendio.com>
533 g_hash_table_new_full(): create hash tables with a ref count of 1.
534 g_hash_table_ref(): atomically ref_count+=1
535 g_hash_table_unref(): atomically ref_count-=1, destroys hash table
536 when refcount reaches 0.
537 g_hash_table_destroy(): just destroy keys and values, unref by 1.
538 g_hash_table_insert():
539 g_hash_table_replace(): assert ref_count>0.
542 * glib/gatomic.c: added 'volatile' qualifier to all atomic pointer and
545 2005-11-20 Behdad Esfahbod <behdad@gnome.org>
547 * glib/guniprop.c (g_unichar_get_mirror_char): Remove unused
548 variables. (#321984, Andrew Paprocki)
550 2005-11-18 Matthias Clasen <mclasen@redhat.com>
552 * configure.in: Bump version
554 2005-11-17 Matthias Clasen <mclasen@redhat.com>
556 * === Released 2.9.0 ===
558 * glib/glib.symbols: Updates
560 * glib/gutf8.c (g_utf8_offset_to_pointer): Add some performance
561 hints to the docs. (Paolo Borelli)
565 * glib/gthread.c (g_thread_foreach): Mark as new api.
567 * README.in: Updates.
571 * glib/gdate.c (g_date_set_time_t):
572 (g_date_set_time_val): Add functions to set a date from
573 a time_t and from a GTimeVal.
574 (g_date_set_time): Deprecate. (#314067, Roger Leigh)
577 * tests/utf8-pointer.c: Unit tests for g_utf8_pointer_to_offset
578 and g_utf8_offset_to_pointer.
580 * glib/gutf8.c (g_utf8_pointer_to_offset)
581 (g_utf8_offset_to_pointer): Handle negative offsets, and use
582 "stutter stepping" for going backwards. (#320638, Larry
587 * glib/gthread.c: const correctness fixes, found
588 by Arjan van de Ven and gcc.
590 2005-11-16 Behdad Esfahbod <behdad@gnome.org>
592 * docs/reference/glib/tmpl/unicode.sgml: Correct typo on mentioning
593 @G_UNICODE_BREAK_UNKNOWN that should be #G_UNICODE_BREAK_UNKNOWN.
595 2005-11-09 Behdad Esfahbod <behdad@gnome.org>
597 * glib/guniprop.c: Use bit hacks instead when checking a general
598 category value against multiple values.
600 * glib/gutf8.c: Change ISO10646 to Unicode in docs.
602 2005-11-10 Simos Xenitellis <simos@gnome.org>
604 * configure.in: Added tt (Tatar) to ALL_LINGUAS.
606 2005-11-05 Tor Lillqvist <tml@novell.com>
608 * glib/giowin32.c (g_io_win32_set_flags): Don't call g_warning().
609 (#320688) Instead set error and return failure.
611 2005-11-07 Matthias Clasen <mclasen@redhat.com>
613 * glib/gmain.c (g_main_dispatch): Don't call cb_funcs->unref
614 while holding the context lock. (#320886, Andy Wingo)
616 2005-11-07 Billy Biggs <vektor@dumbterm.net>
618 * tests/refcount/Makefile.am: Add a missing reference to libglib
619 in the LDADD for this test.
621 2005-11-07 Matthias Clasen <mclasen@redhat.com>
623 * glib/gmem.c (g_allocator_new): We need to set n_preallocs to a
624 nonzero value, otherwise GTK+ 2.8 breaks when compiled against
625 GLib 2.9. (#320755, Luca Ferretti)
627 2005-11-04 Matthias Clasen <mclasen@redhat.com>
629 * glib/Makefile.am: Apply a patch from Behdad Esfahbod to
630 use a faster lookup table for g_unichar_get_mirror_char().
631 * glib/gmirroringtable.h: The new table.
632 * glib/gunichartables.h: Remove bidi_mirroring_table.
633 * glib/gen-unicode-tables.pl: Don't generate the mirroring
635 * glib/glib-mirroring-tab/*: Sources for the program
636 which generated gmirroringtable.h.
638 * glib/glist.c: Avoid some code duplication.
640 * glib/gscanner.h: Include gdataset.h. (#320322)
642 * glib/gdebug.h: add new GDebugFlag for fatal_criticals
643 * glib/gmessages.c: (_g_debug_init): handle G_DEBUG=fatal_criticals,
644 to help find critical warnings in applications. (#320017,
647 2005-11-02 Tor Lillqvist <tml@novell.com>
649 * glib/glib.symbols: Remove large amount of trailing whitespace
650 from one line. Remove #ifdef G_UNDEFINED from around
651 g_hash_table_new and g_hash_table_new_full symbols, what was the
652 purpose of that? Add the g_slice functions.
654 Tue Nov 1 16:24:20 2005 Tim Janik <timj@imendio.com>
656 * glib/gmem.[hc]: prepared deprecation of GMemChunk and GAllocator.
657 added g_slice_*() API to allocate and cache small bits of memory.
658 an actuall allocator implementation for g_slice_*() is still pending.
660 * glib/gthread.[hc]: changes from a patch by Matthias Clasen.
661 changed GRealThread list to use in-structure *next; fields instead
662 of GSList, in order for thread iteration to not depenend on g_slice_*()
664 _g_thread_mem_private_get():
665 _g_thread_mem_private_set(): added accessors for private memory,
666 needed because the ordinary GPrivate implementation relies on GArray
667 and GSList and therefore indirectly on working g_slice_*() allocations.
670 g_thread_foreach(): new public API function to loop over all existing threads.
684 removed GAllocator and free list usages and accompanying locks.
685 use g_slice_*() API to allocate and cache small bits of memory.
687 * glib/ghook.h: removed GMemChunk field from public API.
690 * glib/glist.h: deprecate allocator API, provide _free1() for consistency.
692 * glib/gnode.h: deprecate allocator API.
694 * glib/gmain.c: reordered GPollRec fields so g_slice_free_chain() can
695 be used for poll rec lists.
697 * glib/grel.c: removed mem chunk usage, and allocated tuples via g_slice_*().
698 g_relation_destroy(): free all tuples from the all_tuples hash table,
699 this effectively maintains the life time track keeping of tuples.
700 g_relation_delete_tuple(): free tuples which are removed from the
701 all_tuples hash table. this fixes a temporary leak that was present
702 in the memchunk code until the destruction of the relation.
704 2005-10-29 Matthias Clasen <mclasen@redhat.com>
706 * tests/convert-test.c: Add some tests for conversions between
707 UTF-8, UCS-4 and UTF-16.
709 * glib/gutf8.c (g_utf8_to_ucs4, g_utf8_to_utf16): Fix handling
710 of len == -1, noticed by Morten Welinder.
712 2005-10-27 Erdal Ronahi <erdal.ronahi@gmail.com>
714 * configure.in: Added ku (Kurdish) to ALL_LINGUAS
716 2005-10-26 Matthias Clasen <mclasen@redhat.com>
718 * glib/gutf8.c (g_ucs4_to_utf8): Don't set items_read twice
719 in the error case, and add some documentation. (#319806, Morten
722 2005-10-19 Manish Singh <yosh@gimp.org>
724 * configure.in: Use AC_CHECK_FUNCS for _NSGetEnviron, to get the
725 config.h symbol automatically. Fixes bug #313731.
727 2005-10-19 Tor Lillqvist <tml@novell.com>
729 * glib/gwin32.c (g_win32_get_package_installation_directory):
730 Return a g_strdup()ed copy of the value stored in the hash table,
731 so that it can be g_free()d without leaving a dangling pointer in
732 the hash table. (#319232)
734 2005-10-06 Matthias Clasen <mclasen@redhat.com>
736 * glib/gunicollate.c (g_utf8_collate_key_for_filename): Handle
737 all-zero sequences correctly. (#317930, Sebastien Bacher)
739 2005-10-05 Matthias Clasen <mclasen@redhat.com>
741 * glib/gmarkup.c (g_markup_escape_text): Doc addition
743 * Makefile.am: Add Collation tests.
745 * tests/collate/*: Inputs and expected outputs for collation tests.
747 * tests/run-collate-tests.sh: Script to run collation tests.
749 * tests/unicode-collate.c (main): Rework slightly to make
750 it usable in unit tests. Also test g_utf8_collate_key_for_filename().
752 2005-10-01 Behdad Esfahbod <behdad@gnome.org>
754 * docs/reference/glib/tmpl/unicode.sgml:
755 * glib/gen-unicode-tables.pl:
757 * glib/gunichartables.h:
759 * tests/casefold.txt:
760 * tests/casemap.txt: Updated to Unicode 4.1. There are five new
761 GUnicodeBreakType types. That may break some applications, like
764 2005-09-26 Matthias Clasen <mclasen@redhat.com>
766 * glib/gstrfuncs.c (g_ascii_strtoull): Add details to the
767 docs. (#314393, Matthew F. Barnes)
770 * glib/gprintf.h: Remove g_snprintf() and g_vsnprintf(), since
771 they are already declared in glib.h. This doesn't break documented
772 use of gprintf.h, but should probably be pointed out in the
773 release notes for 2.10. (#314232, Behdad Esfahbod)
775 Tue Sep 20 13:16:04 2005 Tim Janik <timj@imendio.com>
777 * glib/gpattern.c (g_pattern_ph_match): applied significant recursion
778 complexity optimization, based on a patch from Matthias Clasen.
780 * tests/patterntest.c: more tests, mostly from matthias.
782 2005-09-20 Matthias Clasen <mclasen@redhat.com>
784 * glib/gqueue.c (g_queue_insert_sorted): Correct the docs.
785 (#316703, Mark Drago)
787 Mon Sep 19 17:23:23 2005 Tim Janik <timj@imendio.com>
789 * glib/gpattern.c: applied a patch from matthias which checks on the
790 upper bound of GPatternSpec length to optimize matches.
793 * tests/patterntest.c: added more match cases.
795 2005-09-16 Tor Lillqvist <tml@novell.com>
797 * glib/gstrfuncs.c (g_ascii_strcasecmp, g_ascii_strncasecmp): Add
798 warning to doc comment that these functions should not be used on
799 encodings like CP932.
801 2005-09-14 Matthias Clasen <mclasen@redhat.com>
803 * tests/keyfile-test.c: Add a test for grup names of length 1.
805 * glib/gkeyfile.c (g_key_file_line_is_group): Accept group names
806 of length 1. (#316309)
808 2005-09-12 Matthias Clasen <mclasen@redhat.com>
810 * glib/gmarkup.c (g_markup_escape_text): Clarify docs.
812 2005-09-11 Sebastian Wilhelmi <seppi@seppi.de>
814 * tests/refcount/Makefile.am (INCLUDES): Link the the refcount
815 tests to the system thread library $(G_THREAD_LIBS). Fixes #313744
818 2005-09-11 Kjartan Maraas <kmaraas@gnome.org>
820 * glib/gmain.c: (g_child_watch_prepare), (g_child_watch_check),
821 (child_watch_helper_thread): Remove some dead code. Closes
824 2005-09-07 Tor Lillqvist <tml@novell.com>
826 * glib/Makefile.am: Create also a console version of the
827 gspawn-win32-helper program, gspawn-win32-helper-console.exe.
828 It's otherwise identical to gspawn-win32-helper.exe, except marked
829 as a console application (linked without the -mwindows option).
831 * glib/gspawn-win32.c (do_spawn_directly, do_spawn_with_pipes):
832 Drop the dont_wait parameter. Its truth value correlated 100% with
833 the NULLness of the exit_status parameter anyway, so it's enough
834 to check whether exit_status is NULL. Invert the sense of the
835 dont_return_handle parameter and rename it to do_return_handle, to
836 make the code easier to read by avoiding double negations.
838 (g_spawn_sync_utf8, g_spawn_async_with_pipes_utf8): Modify calls
839 to do_spawn_with_pipes() accordingly.
841 (do_spawn_with_pipes): If we have a console, use the console
842 version of the helper program, otherwise use the GUI one. This
843 avoids extra console windows opening up in some situations. (In
844 case a console application uses the GUI gspawn-win32-helper.exe to
845 spawn another console application we would get a separate console
846 for the spawned console application).
848 * glib-zip.in: Distribute also gspawn-win32-helper-console.exe.
850 2005-09-05 Matthias Clasen <mclasen@redhat.com>
852 * glib/gmappedfile.c (g_mapped_file_new): Report an error
853 if the file is too large. (#315275, Kjartan Maraas)
855 * glib/gkeyfile.c (g_key_file_load_from_fd): The return value
856 of read() is signed. (#315273, Kjartan Maraas)
858 2005-08-31 Tor Lillqvist <tml@novell.com>
860 * glib/gutils.h: Wrapping atexit() is a bad idea on Windows, where
861 the EXE and each DLL have their own atexit function chains.
863 #define g_atexit as atexit instead. This means it has a
864 better chance of doing what the caller wants. For instance,
865 gtkhtml calls g_atexit() registering a function in gtkhtml
866 itself. This caused a crash when g_atexit() was implemented as a
867 function in the GLib DLL. The gtkhtml DLL was already unloaded by
868 the time the GLib DLL got unloaded.
870 * glib/gutils.c: #undef the #define mentioned above, to also get a
871 real g_atexit() into the DLL for backward compatibility. Document
872 the Windows behaviour of g_atexit(), and document the varying ways
873 atexit() can behave in the context of dynamically loaded modules
876 2005-08-31 Matthias Clasen <mclasen@redhat.com>
880 * glib/gdataset.c: Add string interning functions.
882 2005-08-28 Matthias Clasen <mclasen@redhat.com>
884 * glib/giochannel.c: Unify some near-duplicate strings. (#314654,
887 2005-08-26 Matthias Clasen <mclasen@redhat.com>
889 * configure.in: Bump version to 2.9.0
891 2005-08-25 Tor Lillqvist <tml@novell.com>
893 Make also the g_spawn*() functions take parameters in the GLib
894 file name encoding, i.e. UTF-8, on Windows. Has no impact on Unix
895 API or ABI. Like the other GLib API that was earlier changed to
896 use UTF-8 on Windows, the names of the functions that take UTF-8
897 have _utf8 suffixes added by using preprocessor macros in the
898 header file. The old names are kept for functions with the old
899 behaviour, taking parameters in the system codepage, for DLL ABI
902 * glib/gspawn.h: On Win32 add the suffix _utf8 to the names of the
903 g_spawn*() functions.
905 * glib/gspawn-win32.c: Use wide-char API on NT-based
906 Windows. Convert parameters from UTF-8 to wide chars (NT) or
907 system codepage (Win9x) and call the C library _wspawn*() or
908 spawn*() functions respectvely. Add DLL ABI stability versions
909 that take parameters in the system codepage.
911 * glib/gspawn-win32-helper.c: On NT-based Windows use the
912 wide-char versions of argv and envp, and use wide-char API to
913 change directory and spawn the program to run. Remove the verbose
914 debugging output, it was too complex to modify for the wide-char
915 features. (Just add temporary debugging printouts if needed, no
916 need to have them permanently in the source.)
918 * glib/gspawn.c: Corresponding documentation updates.
920 * glib/glib.symbols: Corresponding changes: Mark the ABI stability
921 symbols as PRIVATE, add the new _utf8-suffixed ones.
923 2005-08-24 Stepan Kasal <kasal@ucw.cz>
925 * glib/gtypes.h (G_MININT64): Cast the constant to gint64; it is
926 guint64 otherwise and that can produce warnings about comparison
927 between signed and unsigned.
929 2005-08-23 Matthias Clasen <mclasen@redhat.com>
931 * glib/gutils.c: Fix the crt_externs.h include.
933 2005-08-23 Stepan Kasal <kasal@ucw.cz>
935 * NEWS: Fix spelling of my first name.
937 2005-08-23 Matthias Clasen <mclasen@redhat.com>
941 * === Released 2.8.1 ===
945 2005-08-20 Hans Breuer <hans@breuer.org>
947 * glib/makefile.msc.in : link with ws2_32.lib
949 2005-08-18 Tor Lillqvist <tml@novell.com>
951 * configure.in: Check for <sys/wait.h>
953 * glib/gbacktrace.c: Include <sys/wait.h> on if HAVE_SYS_WAIT_H.
955 2005-08-18 Ross Burton <ross@burtonini.com>
958 Optimise single-character insertions.
963 * tests/string-test.c:
964 Add tests for new optimisation, and fix a leak.
966 2005-08-17 Matthias Clasen <mclasen@redhat.com>
968 * configure.in: Check for crt_externs.h and _NSGetEnviron.
970 * glib/gutils.c: On Darwin, include crt-externs.h and
971 define environ using _NSGetEnviron(). (#313731)
973 2005-08-16 Stepan Kasal <kasal@ucw.cz>
975 * glib/gutils.c (g_get_any_init): Move the body of the big if...
976 (g_get_any_init_do): ... to this new function.
977 (g_get_any_init): Declare as inline.
978 (g_get_any_init_locked): New inline function, does the locking.
979 Make use of these two throughout the code.
981 2005-08-15 Matthias Clasen <mclasen@redhat.com>
983 * glib/gbacktrace.c (g_on_error_stack_trace): Wait for
984 the child process and then simply return. This makes
985 The "S" option work as documented in g_on_error_query().
986 (#313125, Matthew F. Barnes)
988 * glib/gunicode.h: Update the link to Unicode category
989 values. (#313369, Behnam Esfahbod)
991 * glib/gqueue.c (g_queue_find_custom): Clarify docs
992 a little. (#311727, Tristan van Berkom)
994 * glib/abicheck.sh, gobject/abicheck.sh: Make the
995 check work on ia64 too, where some symbols ended up
996 in yet another section.