1 2006-01-11 Matthias Clasen <mclasen@redhat.com>
3 * tests/convert-test.c: Don't test items_read and
4 items_written in error cases where it is not set.
6 * tests/asyncqueue-test.c: Set error to NULL.
8 2006-01-08 Matthias Clasen <mclasen@redhat.com>
10 * glib/gtree.c: Break some long lines.
12 * tests/tree-test.c: Add more tests.
14 * glib/gslice.c: Include config.h first, pointed out
17 2005-12-20 Sven Herzberg <herzi@gnome-de.org>
19 * docs/reference/gobject/tmpl/gtype.sgml: explain that
20 G_TYPE_INSTANCE_GET_CLASS() does behave different during
23 2006-01-06 Matthias Clasen <mclasen@redhat.com>
25 * glib/gslice.c: Only define _XOPEN_SOURCE if we know
26 that we have posix_memalign(). (#323937, Bogdan Nicula)
28 2006-01-05 Matthias Clasen <mclasen@redhat.com>
30 * configure.in: Bump version
32 * === Released 2.9.2 ===
34 2006-01-05 Hans Breuer <hans@breuer.org>
36 * glib/makefile.msc.in : added gslice.obj
38 2006-01-05 Matthias Clasen <mclasen@redhat.com>
42 * glib/gmain.c: Clarify the documentation of source ids
43 a bit. (#325874, Dan Williams)
45 * configure.in: Fix another strict aliasing problem.
47 * glib/gthreadpool.c: Avoid use of varargs macro. (#325864,
50 2006-01-04 Manish Singh <yosh@gimp.org>
53 * glib/gstdio.h: don't macroized g_access, g_chdir, and g_unlink
54 either, since they have the same issue as g_rmdir. (Related to
57 Wed Jan 4 13:33:25 2006 Tim Janik <timj@gtk.org>
59 * glib/gslice.c (magazine_cache_pop_magazine): don't reverse chunk
60 order when creating magazines, so we hand out chunks with ascending
63 2006-01-04 Abel Cheung <maddog@linuxhall.org>
65 * configure.in: Added "ml" "zh_HK" to ALL_LINGUAS.
67 2006-01-03 Martyn Russell <martyn@imendio.com>
69 * docs/reference/glib/glib-sections.txt:
71 * glib/gthreadpool.[ch]:
72 - Added new API g_thread_pool_get_idle_time() and
73 g_thread_pool_set_idle_time(). (#324228).
75 * tests/threadpool-test.c:
76 - Updated test case to do thread pool sorting, thread pool with
77 no sorting and a thread pool with idle thread timeouts.
79 2006-01-03 Matthias Clasen <mclasen@redhat.com>
81 * glib/gmain.h: Add new functions here, too.
83 * glib/glib.symbols: Add new functions.
85 2005-12-20 Michael Meeks <michael.meeks@novell.com>
87 * glib/gmain.c (g_main_context_is_owner): new method
88 to determine if the current thread is the owner of the
91 2006-01-02 Matthias Clasen <mclasen@redhat.com>
95 * glib/gstdio.c (g_rmdir): Don't provide g_rmdir() as a macro
96 expanding to rmdir, since rmdir is not declared in a portable
97 system header we can include in gstdio.h. (#325249, Jani Monoses)
99 2006-01-01 Tor Lillqvist <tml@novell.com>
101 * glib/gspawn-win32.c (g_spawn_sync_utf8): Set the GIOChannels for
102 stdout and stderr to unbuffered. Otherwise the giochannel layer
103 will try to read from them regardless whether the
104 g_io_channel_win32_poll() call here has indicated
105 readability or not. (#325310)
107 2005-12-29 Matthias Clasen <mclasen@redhat.com>
109 * glib/gutils.c (glib_check_version): Fix a copy-and-paste error
110 in the docs. (#325273, Declan Naughton)
112 2005-12-27 Manish Singh <yosh@gimp.org>
114 * tests/run-collate-tests.sh: use LC_ALL instead of LC_COLLATE, to
115 make sure we really override things.
117 2005-12-27 Matthias Clasen <mclasen@redhat.com>
119 Fix #316221, Michal Benes, Stanislav Brabec;
121 * configure.in: Fix a strict aliasing problem in
122 g_static_mutex_get_mutex().
123 * glib/gthread.h: ...and in
124 g_static_mutex_get_mutex_impl_shortcut().
126 * glib/gdatasetprivate.h: Add a cast to silence compiler
127 warnings. (#321978, Andrew Paprocki)
129 Partial fix for bug #323937, Bogdan Nicula.
131 * configure.in: Check for malloc.h
132 * glib/gslice.c: Don't include malloc.h unconditionally.
134 2005-12-27 Manish Singh <yosh@gimp.org>
136 * tests/run-collate-tests.sh: export LC_COLLATE so it takes.
137 (#324950, Dan Yefimov)
139 2005-12-26 Matthias Clasen <mclasen@redhat.com>
141 * glib/gslice.c: On Win32, include process.h (#325015, Kazuki
144 2005-12-25 Matthias Clasen <mclasen@redhat.com>
146 * AUTHORS: Update my email
148 * tests/slice-test.c: Fix C99isms. (#324950, Dan Yefimov)
150 2005-12-21 Matthias Clasen <mclasen@redhat.com>
152 * README.in: Add some notes on when 'make check' may fail.
154 2005-12-20 Matthias Clasen <mclasen@redhat.com>
157 * glib/gthreadpool.h:
158 * glib/gthreadpool.c (g_thread_pool_set_sort_function): New function
159 to sort tasks pushed into a threadpool. (#324479, Martyn Russell)
161 * tests/threadpool-test.c: Test this.
163 Tue Dec 20 18:14:14 2005 Tim Janik <timj@imendio.com>
165 * glib/gslice.[hc]: added mem_error() and mem_assert() to test and
166 handle errors without depending on gmessage.c which might not be
167 setup when the error occours.
168 removed G_SLICE_CONFIG_ALWAYS_FREE config option, fixed the code so
169 always freeing can be achieved by adjusting the working set time to
170 0 with G_SLICE_CONFIG_WORKING_SET_MSECS.
171 added G_SLICE_CONFIG_COLOR_INCREMENT to test different color increments
172 (mainly 0 and 1). reduced the minimum block size to 128 bytes, to
173 minimize wastage if small amounts of differently sized structrues are
174 allocated, this does come at a performance cost of roughly 5% though.
175 fixed up block alignment calculation, so it works for varying
176 block sizes. only use strerror() not g_strerror() since the latter
177 depends on working GQuark and GSlice.
178 mem_error(): implemented in terms of fprintf and vfprintf.
180 * tests/slice-color.c: new program to test cache colorization effects.
182 * tests/slice-test.c: trade G_SLICE_CONFIG_ALWAYS_FREE for 0 duration
183 G_SLICE_CONFIG_WORKING_SET_MSECS.
185 2005-12-17 Matthias Clasen <mclasen@redhat.com>
187 * glib/goption.c (parse_short_option): Set an error in all
188 failure cases. (#324332, Tim-Philipp Müller)
190 2005-12-17 Sebastian Wilhelmi <seppi@seppi.de>
192 * glib/gatomic.c: Fix memory barrier position in g_atomic_int_get
193 and g_atomic_pointer_get. Add g_atomic_int_set and
194 g_atomic_pointer_set implementations for the !DEFINE_WITH_MUTEXES &&
195 G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case, as well as defining them
196 as functions (additionally to the macros in the header) for the
197 !G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case.
199 2005-12-16 Matthias Clasen <mclasen@redhat.com>
201 * glib/gmem.c (g_allocator_new): Don't return a pointer to
202 a const struct, since apps expect to be able to modify it.
203 (#324179, J. Ali Harlow)
205 Tue Dec 13 10:13:32 2005 Tim Janik <timj@imendio.com>
207 * glib/gatomic.h: added g_atomic_pointer_set() and g_atomic_int_set()
208 for systems where the initialization of atomic variables requires a
209 write memory barrier.
211 2005-12-09 Matthias Clasen <mclasen@redhat.com>
213 * configure.in: Bump version
215 * === Released 2.9.1 ===
217 2005-12-09 Alexander Larsson <alexl@redhat.com>
221 Use G_GNUC_WARN_UNUSED_RESULT on list functions that return
224 2005-12-08 Matthias Clasen <mclasen@redhat.com>
228 2005-12-07 Martyn Russell <martyn@imendio.com>
230 * glib/gasyncqueue.c:
231 - Call g_queue_insert_sorted() instead of duplicating the code.
232 - Call g_queue_sort() instead of duplicating the code.
233 - Invert sort function results to make sure the same sort function
234 gives the same results across glist, gslist, gqueue and
237 * tests/asyncqueue-test.c:
238 - Updated the sort function to reflect the example in the
239 documentation for gasyncqueue.c.
241 2005-12-07 Martyn Russell <martyn@imendio.com>
243 * docs/reference/glib/glib-sections.txt:
244 * docs/reference/glib/tmpl/linked_lists_double.sgml:
245 * docs/reference/glib/tmpl/linked_lists_single.sgml:
248 - Added g_list_insert_sorted_with_data () and
249 g_slist_insert_sorted_with_data ().
250 - Removed the extra check in g_list_sort() and g_slist_sort() for
251 GCompareDataFunc vs. GCompareFunc.
253 2005-12-07 Tor Lillqvist <tml@novell.com>
255 * glib-gettextize.in: Look up prefix at run-time on Win32,
256 assuming the standard directory structure with glib-gettextize in
259 2005-12-06 Matthias Clasen <mclasen@redhat.com>
261 * glib/gmem.h: Exempt GTK+ from the mem chunk deprecation, since
262 we need GTK+ 2.8 to compile against GLib 2.10.
264 2005-12-05 Manish Singh <yosh@gimp.org>
266 * tests/run-collate-test.sh: set LC_COLLATE, not LANG, to be sure
267 to override any user settings.
269 2005-12-05 Matthias Clasen <mclasen@redhat.com>
271 * configure.in: Revert an accidential commit
273 2005-12-05 Matthias Clasen <mclasen@redhat.com>
275 * glib/gutf8.c: Documentation updates. (#323291, Morten Welinder)
277 * tests/Makefile.am (TESTS_ENVIRONMENT):
278 * tests/gobject/Makefile.am (TESTS_ENVIRONMENT): Set
279 MALLOC_CHECK_ and MALLOC_PERTURB_
281 * tests/run-collate-tests.sh: Run the collation tests explicitly
282 in en_US locale. (#320463)
284 * glib/gmem.h: Really deprecate GMemChunk.
291 * glib/gstrfuncs.c: Add versioned deprecation docs.
293 2005-12-05 Martyn Russell <martyn@imendio.com>
295 * docs/reference/glib/glib-sections.txt:
296 * glib/gasyncqueue.[ch]:
297 - Added support for sorting async queues by with _push_sorted(),
298 _push_sorted_unlocked(), _sort() and _sort_unlocked() (#323047).
301 * tests/asyncqueue-test.c:
302 - Added test case for gasyncqueue.c
304 Mon Dec 5 15:53:20 2005 Tim Janik <timj@imendio.com>
306 * glib/gslice.c: implement chain walking for arbitrary ->next pointer
307 offsets in g_slice_free_chain_with_offset() based on a patch by behdad
308 in bug 323178. moved time consuming logic from g_slice_free() out of
309 the inner loop, so g_slice_free_chain_with_offset() provides a real
310 performance benefit over g_slice_free1() now.
312 * glib/gslice.h: renamed g_slice_free_chain() to
313 g_slice_free_chain_with_offset(). implemented g_slice_free_chain() as
314 a type-safe macro as suggested in bug 323178.
315 simplified the macro implementation of g_slice_free() and implemented
316 it in a type safe manner for all compliers as suggested by Morten
317 Welinder <mortenw@gnome.org>.
322 * glib/glib.symbols: s/g_slice_free_chain/g_slice_free_chain_with_offset/
324 2005-12-05 Matthias Clasen <mclasen@redhat.com>
326 * glib/gasyncqueue.c: Add some docs.
328 * tests/libmoduletestplugin_a.c: Fix compiler warnings.
330 * glib/gatomic.c: In the ia64 implementation, use
331 __sync builtin without _si or _di suffix. (#321229,
332 Stanislav Brabec, patch by Andreas Schwab)
334 2005-12-04 Behdad Esfahbod <behdad@gnome.org>
336 * glib/gslice.h: Remove comma at the end of enum.
338 2005-12-04 Matthias Clasen <mclasen@redhat.com>
340 Handle multiple user names with the same UID better.
341 (#319535, Laszlo Peter)
343 * glib/gutils.c (g_get_any_init_do): When determining user
344 data, first look up $LOGNAME. If the UID doesn't match
345 getuid(), fall back to the current behaviour of looking
346 up the user data based on getuid().
348 2005-12-04 Matthias Clasen <mclasen@redhat.com>
350 * glib/gmacros.h (G_GNUC_WARN_UNUSED_RESULT): Add a macro
351 to make gcc warn if a function result is ignored. (#145466,
352 Arjan van de Ven, Alex Larsson)
354 * glib/gmem.h: Add the new attribute to g_realloc and
357 2005-12-03 Matthias Clasen <mclasen@redhat.com>
360 * glib/gcache.h: Deprecate g_cache_value_foreach. (#322956,
363 * glib/gmappedfile.c: Make mapping of empty files
366 * glib/gfileutils.c: Don't fork a new process just to
367 fix the permissions of the created temp file. (#321318,
368 Alexis S. L. Carvalho)
370 2005-12-02 Matthias Clasen <mclasen@redhat.com>
372 * README.in: Add a note about Solaris threads.
375 * configure.in: Remove support for Solaris threads.
376 (#136971, Sebastian Wilhelmi, patch by Andrew Paprocki)
378 2005-12-02 Matthias Clasen <mclasen@redhat.com>
380 * configure.in: Define G_GUINT64_CONSTANT in analogy to
381 G_GINT64_CONSTANT. (#322568, Andrew Paprocki)
383 2005-12-02 Matthias Clasen <mclasen@redhat.com>
385 * glib/gslice.c: Win32 portability fixes and C99ism removal,
386 pointed out by Kazuki Iwamoto. (#323052)
388 2005-12-02 Matthias Clasen <mclasen@redhat.com>
390 * glib/gatomic.c (g_atomic_int_add): Add a missing volatile for
391 the IA64 implementation.
393 Fri Dec 2 16:18:09 2005 Tim Janik <timj@imendio.com>
395 * glib/gslice.c: some naming and type size fixups.
397 Fri Dec 2 13:08:58 2005 Tim Janik <timj@gtk.org>
399 * glib/gslice.h (g_slice_free): reworked GCC-specific type-safe macro
400 variant into something less verbose, but digestible for gcc-3.4.
402 Fri Dec 2 10:55:07 2005 Tim Janik <timj@imendio.com>
404 * tests/slice-test.c: extended to perform the benchmarking on the old
405 memchunk code if 'O' is selected.
407 * tests/memchunks.c: new file which contains the old GLib mem chunks
408 implementation with prefix old_mem_chunk_.
410 * tests/Makefile.am: added memchunks.c
412 Fri Dec 2 00:16:59 2005 Tim Janik <timj@imendio.com>
414 * glib/gslice.c: improved gettimeofday() timer resolution by maintaining
415 acceptable load distribution of the syscall.
416 (allocator_get_magazine_threshold): reduce minimum/single-thread magazine
417 sizes to approximately page_size/2.5 to avoid excessive startup allocations.
419 2005-12-01 Matthias Clasen <mclasen@redhat.com>
421 * tests/gobject/ifacecheck.c (test_iface_base_init): Don't
422 double-free base interface members.
424 * tests/Makefile.am: Remove duplicate variable.
426 Thu Dec 1 17:32:46 2005 Tim Janik <timj@imendio.com>
428 * glib/gslice.[hc]: new slice allocator implementation.
430 * tests/slice-test.c: added random slice allocation test.
432 * glib/gthread.[hc]: removed newly added private thread mem API.
434 * glib/gthreadinit.h:
437 * glib/gmem.c: divided glib threading initialisation into three phases,
438 initialisation where private keys and messaging are not available (only
439 needed by gmem.c), initialisation without messaging but private keys
440 available (gslice.c, gmessage.c), and full fledged initialisers that
441 server the rest of glib. initialisation functions got renamed to reflect
442 the limitations of their corresponding phases.
444 * glib/gmem.c: removed memchunk code, defer allocations to
447 * glib/gmem.[hc]: removed g_slice_* skeletons.
449 * glib/glib.symbols: added g_slice_* symbols.
451 * configure.in: check for availability of posix_memalign(3), memalign(3)
454 * glib/Makefile.am: added gslice.[hc].
456 2005-12-01 Tor Lillqvist <tml@novell.com>
458 * glib/gstdio.c (g_stat): In the Win32 implementation, strip
459 trailing slash(es) for non-root folders. stat() fails if non-root
460 folders are specified with trailing slashes. It's too much hassle
461 to demand that callers strip such slashes themselves, especially
462 as it is easy to get it wrong and strip the slash of a root
464 (g_rename): On NT-based Windows, use MoveFileEx() with
465 MOVEFILE_REPLACE_EXISTING to better match Unix behaviour.
467 2005-11-28 Matthias Clasen <mclasen@redhat.com>
469 Fix G_STMT_START / G_STMT_END on Solaris. (#321972,
472 * configure.in: Check whether do { } while (0) works.
474 * glib/gmacros.h: Use do { } while (0) for G_STMT_START /
475 G_STMT_END if it works.
477 2005-11-28 Matthias Clasen <mclasen@redhat.com>
479 * glib/gthread.c (g_static_rw_lock_wait, g_static_rw_lock_signal):
480 * glib/gnode.c (g_node_depth_traverse_level):
481 * glib/gmem.c (g_allocator_new):
482 * glib/ghash.c (g_hash_table_unref, g_hash_table_destroy)
483 (g_hash_table_foreach_remove_or_steal): Silence compiler
486 2005-11-27 Matthias Clasen <mclasen@redhat.com>
488 * glib/gunicollate.c (g_utf8_collate_key_for_filename):
489 Don't read beyond len. (#322520, Christian Persch)
491 Wed Nov 23 17:34:01 2005 Tim Janik <timj@imendio.com>
493 * glib/gdataset.c: access datalist flags via atomic pointer access
494 functions, instead of acquiring the dataset lock. this is faster and
495 also matches the atomic pointer readouts in gdatalistprivate.h and
496 direct pointer modifications required by gobject.c.
498 Wed Nov 23 13:35:31 2005 Tim Janik <timj@gtk.org>
500 * glib/gdataset.c: streamlined initialization code somewhat,
501 removed GData node cache.
503 2005-11-22 Matthias Clasen <mclasen@redhat.com>
505 * glib/ghash.c (g_hash_table_ref):
506 (g_hash_table_unref): Mark these as new API.
508 Avoid double locking in g_intern_string (#322133,
511 * glib/gdataset.c (g_quark_from_string_internal): New
512 internal function which factors out common parts of
513 g_quark_from[_static]_string.
514 (g_quark_from_string, g_quark_from_static_string):
515 Use g_quark_from_string_internal.
516 (g_intern_string, g_intern_static_string): Use
517 g_quark_from_string_internal, and only take the
519 (g_quark_new): Don't store the strings shifted by -1
520 in the g_quarks array.
521 (g_quark_to_string): Adapt to the previous change.
523 Tue Nov 22 14:04:26 2005 Tim Janik <timj@imendio.com>
527 g_hash_table_new_full(): create hash tables with a ref count of 1.
528 g_hash_table_ref(): atomically ref_count+=1
529 g_hash_table_unref(): atomically ref_count-=1, destroys hash table
530 when refcount reaches 0.
531 g_hash_table_destroy(): just destroy keys and values, unref by 1.
532 g_hash_table_insert():
533 g_hash_table_replace(): assert ref_count>0.
536 * glib/gatomic.c: added 'volatile' qualifier to all atomic pointer and
539 2005-11-20 Behdad Esfahbod <behdad@gnome.org>
541 * glib/guniprop.c (g_unichar_get_mirror_char): Remove unused
542 variables. (#321984, Andrew Paprocki)
544 2005-11-18 Matthias Clasen <mclasen@redhat.com>
546 * configure.in: Bump version
548 2005-11-17 Matthias Clasen <mclasen@redhat.com>
550 * === Released 2.9.0 ===
552 * glib/glib.symbols: Updates
554 * glib/gutf8.c (g_utf8_offset_to_pointer): Add some performance
555 hints to the docs. (Paolo Borelli)
559 * glib/gthread.c (g_thread_foreach): Mark as new api.
561 * README.in: Updates.
565 * glib/gdate.c (g_date_set_time_t):
566 (g_date_set_time_val): Add functions to set a date from
567 a time_t and from a GTimeVal.
568 (g_date_set_time): Deprecate. (#314067, Roger Leigh)
571 * tests/utf8-pointer.c: Unit tests for g_utf8_pointer_to_offset
572 and g_utf8_offset_to_pointer.
574 * glib/gutf8.c (g_utf8_pointer_to_offset)
575 (g_utf8_offset_to_pointer): Handle negative offsets, and use
576 "stutter stepping" for going backwards. (#320638, Larry
581 * glib/gthread.c: const correctness fixes, found
582 by Arjan van de Ven and gcc.
584 2005-11-16 Behdad Esfahbod <behdad@gnome.org>
586 * docs/reference/glib/tmpl/unicode.sgml: Correct typo on mentioning
587 @G_UNICODE_BREAK_UNKNOWN that should be #G_UNICODE_BREAK_UNKNOWN.
589 2005-11-09 Behdad Esfahbod <behdad@gnome.org>
591 * glib/guniprop.c: Use bit hacks instead when checking a general
592 category value against multiple values.
594 * glib/gutf8.c: Change ISO10646 to Unicode in docs.
596 2005-11-10 Simos Xenitellis <simos@gnome.org>
598 * configure.in: Added tt (Tatar) to ALL_LINGUAS.
600 2005-11-05 Tor Lillqvist <tml@novell.com>
602 * glib/giowin32.c (g_io_win32_set_flags): Don't call g_warning().
603 (#320688) Instead set error and return failure.
605 2005-11-07 Matthias Clasen <mclasen@redhat.com>
607 * glib/gmain.c (g_main_dispatch): Don't call cb_funcs->unref
608 while holding the context lock. (#320886, Andy Wingo)
610 2005-11-07 Billy Biggs <vektor@dumbterm.net>
612 * tests/refcount/Makefile.am: Add a missing reference to libglib
613 in the LDADD for this test.
615 2005-11-07 Matthias Clasen <mclasen@redhat.com>
617 * glib/gmem.c (g_allocator_new): We need to set n_preallocs to a
618 nonzero value, otherwise GTK+ 2.8 breaks when compiled against
619 GLib 2.9. (#320755, Luca Ferretti)
621 2005-11-04 Matthias Clasen <mclasen@redhat.com>
623 * glib/Makefile.am: Apply a patch from Behdad Esfahbod to
624 use a faster lookup table for g_unichar_get_mirror_char().
625 * glib/gmirroringtable.h: The new table.
626 * glib/gunichartables.h: Remove bidi_mirroring_table.
627 * glib/gen-unicode-tables.pl: Don't generate the mirroring
629 * glib/glib-mirroring-tab/*: Sources for the program
630 which generated gmirroringtable.h.
632 * glib/glist.c: Avoid some code duplication.
634 * glib/gscanner.h: Include gdataset.h. (#320322)
636 * glib/gdebug.h: add new GDebugFlag for fatal_criticals
637 * glib/gmessages.c: (_g_debug_init): handle G_DEBUG=fatal_criticals,
638 to help find critical warnings in applications. (#320017,
641 2005-11-02 Tor Lillqvist <tml@novell.com>
643 * glib/glib.symbols: Remove large amount of trailing whitespace
644 from one line. Remove #ifdef G_UNDEFINED from around
645 g_hash_table_new and g_hash_table_new_full symbols, what was the
646 purpose of that? Add the g_slice functions.
648 Tue Nov 1 16:24:20 2005 Tim Janik <timj@imendio.com>
650 * glib/gmem.[hc]: prepared deprecation of GMemChunk and GAllocator.
651 added g_slice_*() API to allocate and cache small bits of memory.
652 an actuall allocator implementation for g_slice_*() is still pending.
654 * glib/gthread.[hc]: changes from a patch by Matthias Clasen.
655 changed GRealThread list to use in-structure *next; fields instead
656 of GSList, in order for thread iteration to not depenend on g_slice_*()
658 _g_thread_mem_private_get():
659 _g_thread_mem_private_set(): added accessors for private memory,
660 needed because the ordinary GPrivate implementation relies on GArray
661 and GSList and therefore indirectly on working g_slice_*() allocations.
664 g_thread_foreach(): new public API function to loop over all existing threads.
678 removed GAllocator and free list usages and accompanying locks.
679 use g_slice_*() API to allocate and cache small bits of memory.
681 * glib/ghook.h: removed GMemChunk field from public API.
684 * glib/glist.h: deprecate allocator API, provide _free1() for consistency.
686 * glib/gnode.h: deprecate allocator API.
688 * glib/gmain.c: reordered GPollRec fields so g_slice_free_chain() can
689 be used for poll rec lists.
691 * glib/grel.c: removed mem chunk usage, and allocated tuples via g_slice_*().
692 g_relation_destroy(): free all tuples from the all_tuples hash table,
693 this effectively maintains the life time track keeping of tuples.
694 g_relation_delete_tuple(): free tuples which are removed from the
695 all_tuples hash table. this fixes a temporary leak that was present
696 in the memchunk code until the destruction of the relation.
698 2005-10-29 Matthias Clasen <mclasen@redhat.com>
700 * tests/convert-test.c: Add some tests for conversions between
701 UTF-8, UCS-4 and UTF-16.
703 * glib/gutf8.c (g_utf8_to_ucs4, g_utf8_to_utf16): Fix handling
704 of len == -1, noticed by Morten Welinder.
706 2005-10-27 Erdal Ronahi <erdal.ronahi@gmail.com>
708 * configure.in: Added ku (Kurdish) to ALL_LINGUAS
710 2005-10-26 Matthias Clasen <mclasen@redhat.com>
712 * glib/gutf8.c (g_ucs4_to_utf8): Don't set items_read twice
713 in the error case, and add some documentation. (#319806, Morten
716 2005-10-19 Manish Singh <yosh@gimp.org>
718 * configure.in: Use AC_CHECK_FUNCS for _NSGetEnviron, to get the
719 config.h symbol automatically. Fixes bug #313731.
721 2005-10-19 Tor Lillqvist <tml@novell.com>
723 * glib/gwin32.c (g_win32_get_package_installation_directory):
724 Return a g_strdup()ed copy of the value stored in the hash table,
725 so that it can be g_free()d without leaving a dangling pointer in
726 the hash table. (#319232)
728 2005-10-06 Matthias Clasen <mclasen@redhat.com>
730 * glib/gunicollate.c (g_utf8_collate_key_for_filename): Handle
731 all-zero sequences correctly. (#317930, Sebastien Bacher)
733 2005-10-05 Matthias Clasen <mclasen@redhat.com>
735 * glib/gmarkup.c (g_markup_escape_text): Doc addition
737 * Makefile.am: Add Collation tests.
739 * tests/collate/*: Inputs and expected outputs for collation tests.
741 * tests/run-collate-tests.sh: Script to run collation tests.
743 * tests/unicode-collate.c (main): Rework slightly to make
744 it usable in unit tests. Also test g_utf8_collate_key_for_filename().
746 2005-10-01 Behdad Esfahbod <behdad@gnome.org>
748 * docs/reference/glib/tmpl/unicode.sgml:
749 * glib/gen-unicode-tables.pl:
751 * glib/gunichartables.h:
753 * tests/casefold.txt:
754 * tests/casemap.txt: Updated to Unicode 4.1. There are five new
755 GUnicodeBreakType types. That may break some applications, like
758 2005-09-26 Matthias Clasen <mclasen@redhat.com>
760 * glib/gstrfuncs.c (g_ascii_strtoull): Add details to the
761 docs. (#314393, Matthew F. Barnes)
764 * glib/gprintf.h: Remove g_snprintf() and g_vsnprintf(), since
765 they are already declared in glib.h. This doesn't break documented
766 use of gprintf.h, but should probably be pointed out in the
767 release notes for 2.10. (#314232, Behdad Esfahbod)
769 Tue Sep 20 13:16:04 2005 Tim Janik <timj@imendio.com>
771 * glib/gpattern.c (g_pattern_ph_match): applied significant recursion
772 complexity optimization, based on a patch from Matthias Clasen.
774 * tests/patterntest.c: more tests, mostly from matthias.
776 2005-09-20 Matthias Clasen <mclasen@redhat.com>
778 * glib/gqueue.c (g_queue_insert_sorted): Correct the docs.
779 (#316703, Mark Drago)
781 Mon Sep 19 17:23:23 2005 Tim Janik <timj@imendio.com>
783 * glib/gpattern.c: applied a patch from matthias which checks on the
784 upper bound of GPatternSpec length to optimize matches.
787 * tests/patterntest.c: added more match cases.
789 2005-09-16 Tor Lillqvist <tml@novell.com>
791 * glib/gstrfuncs.c (g_ascii_strcasecmp, g_ascii_strncasecmp): Add
792 warning to doc comment that these functions should not be used on
793 encodings like CP932.
795 2005-09-14 Matthias Clasen <mclasen@redhat.com>
797 * tests/keyfile-test.c: Add a test for grup names of length 1.
799 * glib/gkeyfile.c (g_key_file_line_is_group): Accept group names
800 of length 1. (#316309)
802 2005-09-12 Matthias Clasen <mclasen@redhat.com>
804 * glib/gmarkup.c (g_markup_escape_text): Clarify docs.
806 2005-09-11 Sebastian Wilhelmi <seppi@seppi.de>
808 * tests/refcount/Makefile.am (INCLUDES): Link the the refcount
809 tests to the system thread library $(G_THREAD_LIBS). Fixes #313744
812 2005-09-11 Kjartan Maraas <kmaraas@gnome.org>
814 * glib/gmain.c: (g_child_watch_prepare), (g_child_watch_check),
815 (child_watch_helper_thread): Remove some dead code. Closes
818 2005-09-07 Tor Lillqvist <tml@novell.com>
820 * glib/Makefile.am: Create also a console version of the
821 gspawn-win32-helper program, gspawn-win32-helper-console.exe.
822 It's otherwise identical to gspawn-win32-helper.exe, except marked
823 as a console application (linked without the -mwindows option).
825 * glib/gspawn-win32.c (do_spawn_directly, do_spawn_with_pipes):
826 Drop the dont_wait parameter. Its truth value correlated 100% with
827 the NULLness of the exit_status parameter anyway, so it's enough
828 to check whether exit_status is NULL. Invert the sense of the
829 dont_return_handle parameter and rename it to do_return_handle, to
830 make the code easier to read by avoiding double negations.
832 (g_spawn_sync_utf8, g_spawn_async_with_pipes_utf8): Modify calls
833 to do_spawn_with_pipes() accordingly.
835 (do_spawn_with_pipes): If we have a console, use the console
836 version of the helper program, otherwise use the GUI one. This
837 avoids extra console windows opening up in some situations. (In
838 case a console application uses the GUI gspawn-win32-helper.exe to
839 spawn another console application we would get a separate console
840 for the spawned console application).
842 * glib-zip.in: Distribute also gspawn-win32-helper-console.exe.
844 2005-09-05 Matthias Clasen <mclasen@redhat.com>
846 * glib/gmappedfile.c (g_mapped_file_new): Report an error
847 if the file is too large. (#315275, Kjartan Maraas)
849 * glib/gkeyfile.c (g_key_file_load_from_fd): The return value
850 of read() is signed. (#315273, Kjartan Maraas)
852 2005-08-31 Tor Lillqvist <tml@novell.com>
854 * glib/gutils.h: Wrapping atexit() is a bad idea on Windows, where
855 the EXE and each DLL have their own atexit function chains.
857 #define g_atexit as atexit instead. This means it has a
858 better chance of doing what the caller wants. For instance,
859 gtkhtml calls g_atexit() registering a function in gtkhtml
860 itself. This caused a crash when g_atexit() was implemented as a
861 function in the GLib DLL. The gtkhtml DLL was already unloaded by
862 the time the GLib DLL got unloaded.
864 * glib/gutils.c: #undef the #define mentioned above, to also get a
865 real g_atexit() into the DLL for backward compatibility. Document
866 the Windows behaviour of g_atexit(), and document the varying ways
867 atexit() can behave in the context of dynamically loaded modules
870 2005-08-31 Matthias Clasen <mclasen@redhat.com>
874 * glib/gdataset.c: Add string interning functions.
876 2005-08-28 Matthias Clasen <mclasen@redhat.com>
878 * glib/giochannel.c: Unify some near-duplicate strings. (#314654,
881 2005-08-26 Matthias Clasen <mclasen@redhat.com>
883 * configure.in: Bump version to 2.9.0
885 2005-08-25 Tor Lillqvist <tml@novell.com>
887 Make also the g_spawn*() functions take parameters in the GLib
888 file name encoding, i.e. UTF-8, on Windows. Has no impact on Unix
889 API or ABI. Like the other GLib API that was earlier changed to
890 use UTF-8 on Windows, the names of the functions that take UTF-8
891 have _utf8 suffixes added by using preprocessor macros in the
892 header file. The old names are kept for functions with the old
893 behaviour, taking parameters in the system codepage, for DLL ABI
896 * glib/gspawn.h: On Win32 add the suffix _utf8 to the names of the
897 g_spawn*() functions.
899 * glib/gspawn-win32.c: Use wide-char API on NT-based
900 Windows. Convert parameters from UTF-8 to wide chars (NT) or
901 system codepage (Win9x) and call the C library _wspawn*() or
902 spawn*() functions respectvely. Add DLL ABI stability versions
903 that take parameters in the system codepage.
905 * glib/gspawn-win32-helper.c: On NT-based Windows use the
906 wide-char versions of argv and envp, and use wide-char API to
907 change directory and spawn the program to run. Remove the verbose
908 debugging output, it was too complex to modify for the wide-char
909 features. (Just add temporary debugging printouts if needed, no
910 need to have them permanently in the source.)
912 * glib/gspawn.c: Corresponding documentation updates.
914 * glib/glib.symbols: Corresponding changes: Mark the ABI stability
915 symbols as PRIVATE, add the new _utf8-suffixed ones.
917 2005-08-24 Stepan Kasal <kasal@ucw.cz>
919 * glib/gtypes.h (G_MININT64): Cast the constant to gint64; it is
920 guint64 otherwise and that can produce warnings about comparison
921 between signed and unsigned.
923 2005-08-23 Matthias Clasen <mclasen@redhat.com>
925 * glib/gutils.c: Fix the crt_externs.h include.
927 2005-08-23 Stepan Kasal <kasal@ucw.cz>
929 * NEWS: Fix spelling of my first name.
931 2005-08-23 Matthias Clasen <mclasen@redhat.com>
935 * === Released 2.8.1 ===
939 2005-08-20 Hans Breuer <hans@breuer.org>
941 * glib/makefile.msc.in : link with ws2_32.lib
943 2005-08-18 Tor Lillqvist <tml@novell.com>
945 * configure.in: Check for <sys/wait.h>
947 * glib/gbacktrace.c: Include <sys/wait.h> on if HAVE_SYS_WAIT_H.
949 2005-08-18 Ross Burton <ross@burtonini.com>
952 Optimise single-character insertions.
957 * tests/string-test.c:
958 Add tests for new optimisation, and fix a leak.
960 2005-08-17 Matthias Clasen <mclasen@redhat.com>
962 * configure.in: Check for crt_externs.h and _NSGetEnviron.
964 * glib/gutils.c: On Darwin, include crt-externs.h and
965 define environ using _NSGetEnviron(). (#313731)
967 2005-08-16 Stepan Kasal <kasal@ucw.cz>
969 * glib/gutils.c (g_get_any_init): Move the body of the big if...
970 (g_get_any_init_do): ... to this new function.
971 (g_get_any_init): Declare as inline.
972 (g_get_any_init_locked): New inline function, does the locking.
973 Make use of these two throughout the code.
975 2005-08-15 Matthias Clasen <mclasen@redhat.com>
977 * glib/gbacktrace.c (g_on_error_stack_trace): Wait for
978 the child process and then simply return. This makes
979 The "S" option work as documented in g_on_error_query().
980 (#313125, Matthew F. Barnes)
982 * glib/gunicode.h: Update the link to Unicode category
983 values. (#313369, Behnam Esfahbod)
985 * glib/gqueue.c (g_queue_find_custom): Clarify docs
986 a little. (#311727, Tristan van Berkom)
988 * glib/abicheck.sh, gobject/abicheck.sh: Make the
989 check work on ia64 too, where some symbols ended up
990 in yet another section.