1 2006-01-03 Matthias Clasen <mclasen@redhat.com>
3 * glib/gmain.h: Add new functions here, too.
5 * glib/glib.symbols: Add new functions.
7 2005-12-20 Michael Meeks <michael.meeks@novell.com>
9 * glib/gmain.c (g_main_context_is_owner): new method
10 to determine if the current thread is the owner of the
13 2006-01-02 Matthias Clasen <mclasen@redhat.com>
17 * glib/gstdio.c (g_rmdir): Don't provide g_rmdir() as a macro
18 expanding to rmdir, since rmdir is not declared in a portable
19 system header we can include in gstdio.h. (#325249, Jani Monoses)
21 2006-01-01 Tor Lillqvist <tml@novell.com>
23 * glib/gspawn-win32.c (g_spawn_sync_utf8): Set the GIOChannels for
24 stdout and stderr to unbuffered. Otherwise the giochannel layer
25 will try to read from them regardless whether the
26 g_io_channel_win32_poll() call here has indicated
27 readability or not. (#325310)
29 2005-12-29 Matthias Clasen <mclasen@redhat.com>
31 * glib/gutils.c (glib_check_version): Fix a copy-and-paste error
32 in the docs. (#325273, Declan Naughton)
34 2005-12-27 Manish Singh <yosh@gimp.org>
36 * tests/run-collate-tests.sh: use LC_ALL instead of LC_COLLATE, to
37 make sure we really override things.
39 2005-12-27 Matthias Clasen <mclasen@redhat.com>
41 Fix #316221, Michal Benes, Stanislav Brabec;
43 * configure.in: Fix a strict aliasing problem in
44 g_static_mutex_get_mutex().
45 * glib/gthread.h: ...and in
46 g_static_mutex_get_mutex_impl_shortcut().
48 * glib/gdatasetprivate.h: Add a cast to silence compiler
49 warnings. (#321978, Andrew Paprocki)
51 Partial fix for bug #32937, Bogdan Nicula.
53 * configure.in: Check for malloc.h
54 * glib/gslice.c: Don't include malloc.h unconditionally.
56 2005-12-27 Manish Singh <yosh@gimp.org>
58 * tests/run-collate-tests.sh: export LC_COLLATE so it takes.
59 (#324950, Dan Yefimov)
61 2005-12-26 Matthias Clasen <mclasen@redhat.com>
63 * glib/gslice.c: On Win32, include process.h (#325015, Kazuki
66 2005-12-25 Matthias Clasen <mclasen@redhat.com>
68 * AUTHORS: Update my email
70 * tests/slice-test.c: Fix C99isms. (#324950, Dan Yefimov)
72 2005-12-21 Matthias Clasen <mclasen@redhat.com>
74 * README.in: Add some notes on when 'make check' may fail.
76 2005-12-20 Matthias Clasen <mclasen@redhat.com>
80 * glib/gthreadpool.c (g_thread_pool_set_sort_function): New function
81 to sort tasks pushed into a threadpool. (#324479, Martyn Russell)
83 * tests/threadpool-test.c: Test this.
85 Tue Dec 20 18:14:14 2005 Tim Janik <timj@imendio.com>
87 * glib/gslice.[hc]: added mem_error() and mem_assert() to test and
88 handle errors without depending on gmessage.c which might not be
89 setup when the error occours.
90 removed G_SLICE_CONFIG_ALWAYS_FREE config option, fixed the code so
91 always freeing can be achieved by adjusting the working set time to
92 0 with G_SLICE_CONFIG_WORKING_SET_MSECS.
93 added G_SLICE_CONFIG_COLOR_INCREMENT to test different color increments
94 (mainly 0 and 1). reduced the minimum block size to 128 bytes, to
95 minimize wastage if small amounts of differently sized structrues are
96 allocated, this does come at a performance cost of roughly 5% though.
97 fixed up block alignment calculation, so it works for varying
98 block sizes. only use strerror() not g_strerror() since the latter
99 depends on working GQuark and GSlice.
100 mem_error(): implemented in terms of fprintf and vfprintf.
102 * tests/slice-color.c: new program to test cache colorization effects.
104 * tests/slice-test.c: trade G_SLICE_CONFIG_ALWAYS_FREE for 0 duration
105 G_SLICE_CONFIG_WORKING_SET_MSECS.
107 2005-12-17 Matthias Clasen <mclasen@redhat.com>
109 * glib/goption.c (parse_short_option): Set an error in all
110 failure cases. (#324332, Tim-Philipp Müller)
112 2005-12-17 Sebastian Wilhelmi <seppi@seppi.de>
114 * glib/gatomic.c: Fix memory barrier position in g_atomic_int_get
115 and g_atomic_pointer_get. Add g_atomic_int_set and
116 g_atomic_pointer_set implementations for the !DEFINE_WITH_MUTEXES &&
117 G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case, as well as defining them
118 as functions (additionally to the macros in the header) for the
119 !G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case.
121 2005-12-16 Matthias Clasen <mclasen@redhat.com>
123 * glib/gmem.c (g_allocator_new): Don't return a pointer to
124 a const struct, since apps expect to be able to modify it.
125 (#324179, J. Ali Harlow)
127 Tue Dec 13 10:13:32 2005 Tim Janik <timj@imendio.com>
129 * glib/gatomic.h: added g_atomic_pointer_set() and g_atomic_int_set()
130 for systems where the initialization of atomic variables requires a
131 write memory barrier.
133 2005-12-09 Matthias Clasen <mclasen@redhat.com>
135 * configure.in: Bump version
137 * === Released 2.9.1 ===
139 2005-12-09 Alexander Larsson <alexl@redhat.com>
143 Use G_GNUC_WARN_UNUSED_RESULT on list functions that return
146 2005-12-08 Matthias Clasen <mclasen@redhat.com>
150 2005-12-07 Martyn Russell <martyn@imendio.com>
152 * glib/gasyncqueue.c:
153 - Call g_queue_insert_sorted() instead of duplicating the code.
154 - Call g_queue_sort() instead of duplicating the code.
155 - Invert sort function results to make sure the same sort function
156 gives the same results across glist, gslist, gqueue and
159 * tests/asyncqueue-test.c:
160 - Updated the sort function to reflect the example in the
161 documentation for gasyncqueue.c.
163 2005-12-07 Martyn Russell <martyn@imendio.com>
165 * docs/reference/glib/glib-sections.txt:
166 * docs/reference/glib/tmpl/linked_lists_double.sgml:
167 * docs/reference/glib/tmpl/linked_lists_single.sgml:
170 - Added g_list_insert_sorted_with_data () and
171 g_slist_insert_sorted_with_data ().
172 - Removed the extra check in g_list_sort() and g_slist_sort() for
173 GCompareDataFunc vs. GCompareFunc.
175 2005-12-07 Tor Lillqvist <tml@novell.com>
177 * glib-gettextize.in: Look up prefix at run-time on Win32,
178 assuming the standard directory structure with glib-gettextize in
181 2005-12-06 Matthias Clasen <mclasen@redhat.com>
183 * glib/gmem.h: Exempt GTK+ from the mem chunk deprecation, since
184 we need GTK+ 2.8 to compile against GLib 2.10.
186 2005-12-05 Manish Singh <yosh@gimp.org>
188 * tests/run-collate-test.sh: set LC_COLLATE, not LANG, to be sure
189 to override any user settings.
191 2005-12-05 Matthias Clasen <mclasen@redhat.com>
193 * configure.in: Revert an accidential commit
195 2005-12-05 Matthias Clasen <mclasen@redhat.com>
197 * glib/gutf8.c: Documentation updates. (#323291, Morten Welinder)
199 * tests/Makefile.am (TESTS_ENVIRONMENT):
200 * tests/gobject/Makefile.am (TESTS_ENVIRONMENT): Set
201 MALLOC_CHECK_ and MALLOC_PERTURB_
203 * tests/run-collate-tests.sh: Run the collation tests explicitly
204 in en_US locale. (#320463)
206 * glib/gmem.h: Really deprecate GMemChunk.
213 * glib/gstrfuncs.c: Add versioned deprecation docs.
215 2005-12-05 Martyn Russell <martyn@imendio.com>
217 * docs/reference/glib/glib-sections.txt:
218 * glib/gasyncqueue.[ch]:
219 - Added support for sorting async queues by with _push_sorted(),
220 _push_sorted_unlocked(), _sort() and _sort_unlocked() (#323047).
223 * tests/asyncqueue-test.c:
224 - Added test case for gasyncqueue.c
226 Mon Dec 5 15:53:20 2005 Tim Janik <timj@imendio.com>
228 * glib/gslice.c: implement chain walking for arbitrary ->next pointer
229 offsets in g_slice_free_chain_with_offset() based on a patch by behdad
230 in bug 323178. moved time consuming logic from g_slice_free() out of
231 the inner loop, so g_slice_free_chain_with_offset() provides a real
232 performance benefit over g_slice_free1() now.
234 * glib/gslice.h: renamed g_slice_free_chain() to
235 g_slice_free_chain_with_offset(). implemented g_slice_free_chain() as
236 a type-safe macro as suggested in bug 323178.
237 simplified the macro implementation of g_slice_free() and implemented
238 it in a type safe manner for all compliers as suggested by Morten
239 Welinder <mortenw@gnome.org>.
244 * glib/glib.symbols: s/g_slice_free_chain/g_slice_free_chain_with_offset/
246 2005-12-05 Matthias Clasen <mclasen@redhat.com>
248 * glib/gasyncqueue.c: Add some docs.
250 * tests/libmoduletestplugin_a.c: Fix compiler warnings.
252 * glib/gatomic.c: In the ia64 implementation, use
253 __sync builtin without _si or _di suffix. (#321229,
254 Stanislav Brabec, patch by Andreas Schwab)
256 2005-12-04 Behdad Esfahbod <behdad@gnome.org>
258 * glib/gslice.h: Remove comma at the end of enum.
260 2005-12-04 Matthias Clasen <mclasen@redhat.com>
262 Handle multiple user names with the same UID better.
263 (#319535, Laszlo Peter)
265 * glib/gutils.c (g_get_any_init_do): When determining user
266 data, first look up $LOGNAME. If the UID doesn't match
267 getuid(), fall back to the current behaviour of looking
268 up the user data based on getuid().
270 2005-12-04 Matthias Clasen <mclasen@redhat.com>
272 * glib/gmacros.h (G_GNUC_WARN_UNUSED_RESULT): Add a macro
273 to make gcc warn if a function result is ignored. (#145466,
274 Arjan van de Ven, Alex Larsson)
276 * glib/gmem.h: Add the new attribute to g_realloc and
279 2005-12-03 Matthias Clasen <mclasen@redhat.com>
282 * glib/gcache.h: Deprecate g_cache_value_foreach. (#322956,
285 * glib/gmappedfile.c: Make mapping of empty files
288 * glib/gfileutils.c: Don't fork a new process just to
289 fix the permissions of the created temp file. (#321318,
290 Alexis S. L. Carvalho)
292 2005-12-02 Matthias Clasen <mclasen@redhat.com>
294 * README.in: Add a note about Solaris threads.
297 * configure.in: Remove support for Solaris threads.
298 (#136971, Sebastian Wilhelmi, patch by Andrew Paprocki)
300 2005-12-02 Matthias Clasen <mclasen@redhat.com>
302 * configure.in: Define G_GUINT64_CONSTANT in analogy to
303 G_GINT64_CONSTANT. (#322568, Andrew Paprocki)
305 2005-12-02 Matthias Clasen <mclasen@redhat.com>
307 * glib/gslice.c: Win32 portability fixes and C99ism removal,
308 pointed out by Kazuki Iwamoto. (#323052)
310 2005-12-02 Matthias Clasen <mclasen@redhat.com>
312 * glib/gatomic.c (g_atomic_int_add): Add a missing volatile for
313 the IA64 implementation.
315 Fri Dec 2 16:18:09 2005 Tim Janik <timj@imendio.com>
317 * glib/gslice.c: some naming and type size fixups.
319 Fri Dec 2 13:08:58 2005 Tim Janik <timj@gtk.org>
321 * glib/gslice.h (g_slice_free): reworked GCC-specific type-safe macro
322 variant into something less verbose, but digestible for gcc-3.4.
324 Fri Dec 2 10:55:07 2005 Tim Janik <timj@imendio.com>
326 * tests/slice-test.c: extended to perform the benchmarking on the old
327 memchunk code if 'O' is selected.
329 * tests/memchunks.c: new file which contains the old GLib mem chunks
330 implementation with prefix old_mem_chunk_.
332 * tests/Makefile.am: added memchunks.c
334 Fri Dec 2 00:16:59 2005 Tim Janik <timj@imendio.com>
336 * glib/gslice.c: improved gettimeofday() timer resolution by maintaining
337 acceptable load distribution of the syscall.
338 (allocator_get_magazine_threshold): reduce minimum/single-thread magazine
339 sizes to approximately page_size/2.5 to avoid excessive startup allocations.
341 2005-12-01 Matthias Clasen <mclasen@redhat.com>
343 * tests/gobject/ifacecheck.c (test_iface_base_init): Don't
344 double-free base interface members.
346 * tests/Makefile.am: Remove duplicate variable.
348 Thu Dec 1 17:32:46 2005 Tim Janik <timj@imendio.com>
350 * glib/gslice.[hc]: new slice allocator implementation.
352 * tests/slice-test.c: added random slice allocation test.
354 * glib/gthread.[hc]: removed newly added private thread mem API.
356 * glib/gthreadinit.h:
359 * glib/gmem.c: divided glib threading initialisation into three phases,
360 initialisation where private keys and messaging are not available (only
361 needed by gmem.c), initialisation without messaging but private keys
362 available (gslice.c, gmessage.c), and full fledged initialisers that
363 server the rest of glib. initialisation functions got renamed to reflect
364 the limitations of their corresponding phases.
366 * glib/gmem.c: removed memchunk code, defer allocations to
369 * glib/gmem.[hc]: removed g_slice_* skeletons.
371 * glib/glib.symbols: added g_slice_* symbols.
373 * configure.in: check for availability of posix_memalign(3), memalign(3)
376 * glib/Makefile.am: added gslice.[hc].
378 2005-12-01 Tor Lillqvist <tml@novell.com>
380 * glib/gstdio.c (g_stat): In the Win32 implementation, strip
381 trailing slash(es) for non-root folders. stat() fails if non-root
382 folders are specified with trailing slashes. It's too much hassle
383 to demand that callers strip such slashes themselves, especially
384 as it is easy to get it wrong and strip the slash of a root
386 (g_rename): On NT-based Windows, use MoveFileEx() with
387 MOVEFILE_REPLACE_EXISTING to better match Unix behaviour.
389 2005-11-28 Matthias Clasen <mclasen@redhat.com>
391 Fix G_STMT_START / G_STMT_END on Solaris. (#321972,
394 * configure.in: Check whether do { } while (0) works.
396 * glib/gmacros.h: Use do { } while (0) for G_STMT_START /
397 G_STMT_END if it works.
399 2005-11-28 Matthias Clasen <mclasen@redhat.com>
401 * glib/gthread.c (g_static_rw_lock_wait, g_static_rw_lock_signal):
402 * glib/gnode.c (g_node_depth_traverse_level):
403 * glib/gmem.c (g_allocator_new):
404 * glib/ghash.c (g_hash_table_unref, g_hash_table_destroy)
405 (g_hash_table_foreach_remove_or_steal): Silence compiler
408 2005-11-27 Matthias Clasen <mclasen@redhat.com>
410 * glib/gunicollate.c (g_utf8_collate_key_for_filename):
411 Don't read beyond len. (#322520, Christian Persch)
413 Wed Nov 23 17:34:01 2005 Tim Janik <timj@imendio.com>
415 * glib/gdataset.c: access datalist flags via atomic pointer access
416 functions, instead of acquiring the dataset lock. this is faster and
417 also matches the atomic pointer readouts in gdatalistprivate.h and
418 direct pointer modifications required by gobject.c.
420 Wed Nov 23 13:35:31 2005 Tim Janik <timj@gtk.org>
422 * glib/gdataset.c: streamlined initialization code somewhat,
423 removed GData node cache.
425 2005-11-22 Matthias Clasen <mclasen@redhat.com>
427 * glib/ghash.c (g_hash_table_ref):
428 (g_hash_table_unref): Mark these as new API.
430 Avoid double locking in g_intern_string (#322133,
433 * glib/gdataset.c (g_quark_from_string_internal): New
434 internal function which factors out common parts of
435 g_quark_from[_static]_string.
436 (g_quark_from_string, g_quark_from_static_string):
437 Use g_quark_from_string_internal.
438 (g_intern_string, g_intern_static_string): Use
439 g_quark_from_string_internal, and only take the
441 (g_quark_new): Don't store the strings shifted by -1
442 in the g_quarks array.
443 (g_quark_to_string): Adapt to the previous change.
445 Tue Nov 22 14:04:26 2005 Tim Janik <timj@imendio.com>
449 g_hash_table_new_full(): create hash tables with a ref count of 1.
450 g_hash_table_ref(): atomically ref_count+=1
451 g_hash_table_unref(): atomically ref_count-=1, destroys hash table
452 when refcount reaches 0.
453 g_hash_table_destroy(): just destroy keys and values, unref by 1.
454 g_hash_table_insert():
455 g_hash_table_replace(): assert ref_count>0.
458 * glib/gatomic.c: added 'volatile' qualifier to all atomic pointer and
461 2005-11-20 Behdad Esfahbod <behdad@gnome.org>
463 * glib/guniprop.c (g_unichar_get_mirror_char): Remove unused
464 variables. (#321984, Andrew Paprocki)
466 2005-11-18 Matthias Clasen <mclasen@redhat.com>
468 * configure.in: Bump version
470 2005-11-17 Matthias Clasen <mclasen@redhat.com>
472 * === Released 2.9.0 ===
474 * glib/glib.symbols: Updates
476 * glib/gutf8.c (g_utf8_offset_to_pointer): Add some performance
477 hints to the docs. (Paolo Borelli)
481 * glib/gthread.c (g_thread_foreach): Mark as new api.
483 * README.in: Updates.
487 * glib/gdate.c (g_date_set_time_t):
488 (g_date_set_time_val): Add functions to set a date from
489 a time_t and from a GTimeVal.
490 (g_date_set_time): Deprecate. (#314067, Roger Leigh)
493 * tests/utf8-pointer.c: Unit tests for g_utf8_pointer_to_offset
494 and g_utf8_offset_to_pointer.
496 * glib/gutf8.c (g_utf8_pointer_to_offset)
497 (g_utf8_offset_to_pointer): Handle negative offsets, and use
498 "stutter stepping" for going backwards. (#320638, Larry
503 * glib/gthread.c: const correctness fixes, found
504 by Arjan van de Ven and gcc.
506 2005-11-16 Behdad Esfahbod <behdad@gnome.org>
508 * docs/reference/glib/tmpl/unicode.sgml: Correct typo on mentioning
509 @G_UNICODE_BREAK_UNKNOWN that should be #G_UNICODE_BREAK_UNKNOWN.
511 2005-11-09 Behdad Esfahbod <behdad@gnome.org>
513 * glib/guniprop.c: Use bit hacks instead when checking a general
514 category value against multiple values.
516 * glib/gutf8.c: Change ISO10646 to Unicode in docs.
518 2005-11-10 Simos Xenitellis <simos@gnome.org>
520 * configure.in: Added tt (Tatar) to ALL_LINGUAS.
522 2005-11-05 Tor Lillqvist <tml@novell.com>
524 * glib/giowin32.c (g_io_win32_set_flags): Don't call g_warning().
525 (#320688) Instead set error and return failure.
527 2005-11-07 Matthias Clasen <mclasen@redhat.com>
529 * glib/gmain.c (g_main_dispatch): Don't call cb_funcs->unref
530 while holding the context lock. (#320886, Andy Wingo)
532 2005-11-07 Billy Biggs <vektor@dumbterm.net>
534 * tests/refcount/Makefile.am: Add a missing reference to libglib
535 in the LDADD for this test.
537 2005-11-07 Matthias Clasen <mclasen@redhat.com>
539 * glib/gmem.c (g_allocator_new): We need to set n_preallocs to a
540 nonzero value, otherwise GTK+ 2.8 breaks when compiled against
541 GLib 2.9. (#320755, Luca Ferretti)
543 2005-11-04 Matthias Clasen <mclasen@redhat.com>
545 * glib/Makefile.am: Apply a patch from Behdad Esfahbod to
546 use a faster lookup table for g_unichar_get_mirror_char().
547 * glib/gmirroringtable.h: The new table.
548 * glib/gunichartables.h: Remove bidi_mirroring_table.
549 * glib/gen-unicode-tables.pl: Don't generate the mirroring
551 * glib/glib-mirroring-tab/*: Sources for the program
552 which generated gmirroringtable.h.
554 * glib/glist.c: Avoid some code duplication.
556 * glib/gscanner.h: Include gdataset.h. (#320322)
558 * glib/gdebug.h: add new GDebugFlag for fatal_criticals
559 * glib/gmessages.c: (_g_debug_init): handle G_DEBUG=fatal_criticals,
560 to help find critical warnings in applications. (#320017,
563 2005-11-02 Tor Lillqvist <tml@novell.com>
565 * glib/glib.symbols: Remove large amount of trailing whitespace
566 from one line. Remove #ifdef G_UNDEFINED from around
567 g_hash_table_new and g_hash_table_new_full symbols, what was the
568 purpose of that? Add the g_slice functions.
570 Tue Nov 1 16:24:20 2005 Tim Janik <timj@imendio.com>
572 * glib/gmem.[hc]: prepared deprecation of GMemChunk and GAllocator.
573 added g_slice_*() API to allocate and cache small bits of memory.
574 an actuall allocator implementation for g_slice_*() is still pending.
576 * glib/gthread.[hc]: changes from a patch by Matthias Clasen.
577 changed GRealThread list to use in-structure *next; fields instead
578 of GSList, in order for thread iteration to not depenend on g_slice_*()
580 _g_thread_mem_private_get():
581 _g_thread_mem_private_set(): added accessors for private memory,
582 needed because the ordinary GPrivate implementation relies on GArray
583 and GSList and therefore indirectly on working g_slice_*() allocations.
586 g_thread_foreach(): new public API function to loop over all existing threads.
600 removed GAllocator and free list usages and accompanying locks.
601 use g_slice_*() API to allocate and cache small bits of memory.
603 * glib/ghook.h: removed GMemChunk field from public API.
606 * glib/glist.h: deprecate allocator API, provide _free1() for consistency.
608 * glib/gnode.h: deprecate allocator API.
610 * glib/gmain.c: reordered GPollRec fields so g_slice_free_chain() can
611 be used for poll rec lists.
613 * glib/grel.c: removed mem chunk usage, and allocated tuples via g_slice_*().
614 g_relation_destroy(): free all tuples from the all_tuples hash table,
615 this effectively maintains the life time track keeping of tuples.
616 g_relation_delete_tuple(): free tuples which are removed from the
617 all_tuples hash table. this fixes a temporary leak that was present
618 in the memchunk code until the destruction of the relation.
620 2005-10-29 Matthias Clasen <mclasen@redhat.com>
622 * tests/convert-test.c: Add some tests for conversions between
623 UTF-8, UCS-4 and UTF-16.
625 * glib/gutf8.c (g_utf8_to_ucs4, g_utf8_to_utf16): Fix handling
626 of len == -1, noticed by Morten Welinder.
628 2005-10-27 Erdal Ronahi <erdal.ronahi@gmail.com>
630 * configure.in: Added ku (Kurdish) to ALL_LINGUAS
632 2005-10-26 Matthias Clasen <mclasen@redhat.com>
634 * glib/gutf8.c (g_ucs4_to_utf8): Don't set items_read twice
635 in the error case, and add some documentation. (#319806, Morten
638 2005-10-19 Manish Singh <yosh@gimp.org>
640 * configure.in: Use AC_CHECK_FUNCS for _NSGetEnviron, to get the
641 config.h symbol automatically. Fixes bug #313731.
643 2005-10-19 Tor Lillqvist <tml@novell.com>
645 * glib/gwin32.c (g_win32_get_package_installation_directory):
646 Return a g_strdup()ed copy of the value stored in the hash table,
647 so that it can be g_free()d without leaving a dangling pointer in
648 the hash table. (#319232)
650 2005-10-06 Matthias Clasen <mclasen@redhat.com>
652 * glib/gunicollate.c (g_utf8_collate_key_for_filename): Handle
653 all-zero sequences correctly. (#317930, Sebastien Bacher)
655 2005-10-05 Matthias Clasen <mclasen@redhat.com>
657 * glib/gmarkup.c (g_markup_escape_text): Doc addition
659 * Makefile.am: Add Collation tests.
661 * tests/collate/*: Inputs and expected outputs for collation tests.
663 * tests/run-collate-tests.sh: Script to run collation tests.
665 * tests/unicode-collate.c (main): Rework slightly to make
666 it usable in unit tests. Also test g_utf8_collate_key_for_filename().
668 2005-10-01 Behdad Esfahbod <behdad@gnome.org>
670 * docs/reference/glib/tmpl/unicode.sgml:
671 * glib/gen-unicode-tables.pl:
673 * glib/gunichartables.h:
675 * tests/casefold.txt:
676 * tests/casemap.txt: Updated to Unicode 4.1. There are five new
677 GUnicodeBreakType types. That may break some applications, like
680 2005-09-26 Matthias Clasen <mclasen@redhat.com>
682 * glib/gstrfuncs.c (g_ascii_strtoull): Add details to the
683 docs. (#314393, Matthew F. Barnes)
686 * glib/gprintf.h: Remove g_snprintf() and g_vsnprintf(), since
687 they are already declared in glib.h. This doesn't break documented
688 use of gprintf.h, but should probably be pointed out in the
689 release notes for 2.10. (#314232, Behdad Esfahbod)
691 Tue Sep 20 13:16:04 2005 Tim Janik <timj@imendio.com>
693 * glib/gpattern.c (g_pattern_ph_match): applied significant recursion
694 complexity optimization, based on a patch from Matthias Clasen.
696 * tests/patterntest.c: more tests, mostly from matthias.
698 2005-09-20 Matthias Clasen <mclasen@redhat.com>
700 * glib/gqueue.c (g_queue_insert_sorted): Correct the docs.
701 (#316703, Mark Drago)
703 Mon Sep 19 17:23:23 2005 Tim Janik <timj@imendio.com>
705 * glib/gpattern.c: applied a patch from matthias which checks on the
706 upper bound of GPatternSpec length to optimize matches.
709 * tests/patterntest.c: added more match cases.
711 2005-09-16 Tor Lillqvist <tml@novell.com>
713 * glib/gstrfuncs.c (g_ascii_strcasecmp, g_ascii_strncasecmp): Add
714 warning to doc comment that these functions should not be used on
715 encodings like CP932.
717 2005-09-14 Matthias Clasen <mclasen@redhat.com>
719 * tests/keyfile-test.c: Add a test for grup names of length 1.
721 * glib/gkeyfile.c (g_key_file_line_is_group): Accept group names
722 of length 1. (#316309)
724 2005-09-12 Matthias Clasen <mclasen@redhat.com>
726 * glib/gmarkup.c (g_markup_escape_text): Clarify docs.
728 2005-09-11 Sebastian Wilhelmi <seppi@seppi.de>
730 * tests/refcount/Makefile.am (INCLUDES): Link the the refcount
731 tests to the system thread library $(G_THREAD_LIBS). Fixes #313744
734 2005-09-11 Kjartan Maraas <kmaraas@gnome.org>
736 * glib/gmain.c: (g_child_watch_prepare), (g_child_watch_check),
737 (child_watch_helper_thread): Remove some dead code. Closes
740 2005-09-07 Tor Lillqvist <tml@novell.com>
742 * glib/Makefile.am: Create also a console version of the
743 gspawn-win32-helper program, gspawn-win32-helper-console.exe.
744 It's otherwise identical to gspawn-win32-helper.exe, except marked
745 as a console application (linked without the -mwindows option).
747 * glib/gspawn-win32.c (do_spawn_directly, do_spawn_with_pipes):
748 Drop the dont_wait parameter. Its truth value correlated 100% with
749 the NULLness of the exit_status parameter anyway, so it's enough
750 to check whether exit_status is NULL. Invert the sense of the
751 dont_return_handle parameter and rename it to do_return_handle, to
752 make the code easier to read by avoiding double negations.
754 (g_spawn_sync_utf8, g_spawn_async_with_pipes_utf8): Modify calls
755 to do_spawn_with_pipes() accordingly.
757 (do_spawn_with_pipes): If we have a console, use the console
758 version of the helper program, otherwise use the GUI one. This
759 avoids extra console windows opening up in some situations. (In
760 case a console application uses the GUI gspawn-win32-helper.exe to
761 spawn another console application we would get a separate console
762 for the spawned console application).
764 * glib-zip.in: Distribute also gspawn-win32-helper-console.exe.
766 2005-09-05 Matthias Clasen <mclasen@redhat.com>
768 * glib/gmappedfile.c (g_mapped_file_new): Report an error
769 if the file is too large. (#315275, Kjartan Maraas)
771 * glib/gkeyfile.c (g_key_file_load_from_fd): The return value
772 of read() is signed. (#315273, Kjartan Maraas)
774 2005-08-31 Tor Lillqvist <tml@novell.com>
776 * glib/gutils.h: Wrapping atexit() is a bad idea on Windows, where
777 the EXE and each DLL have their own atexit function chains.
779 #define g_atexit as atexit instead. This means it has a
780 better chance of doing what the caller wants. For instance,
781 gtkhtml calls g_atexit() registering a function in gtkhtml
782 itself. This caused a crash when g_atexit() was implemented as a
783 function in the GLib DLL. The gtkhtml DLL was already unloaded by
784 the time the GLib DLL got unloaded.
786 * glib/gutils.c: #undef the #define mentioned above, to also get a
787 real g_atexit() into the DLL for backward compatibility. Document
788 the Windows behaviour of g_atexit(), and document the varying ways
789 atexit() can behave in the context of dynamically loaded modules
792 2005-08-31 Matthias Clasen <mclasen@redhat.com>
796 * glib/gdataset.c: Add string interning functions.
798 2005-08-28 Matthias Clasen <mclasen@redhat.com>
800 * glib/giochannel.c: Unify some near-duplicate strings. (#314654,
803 2005-08-26 Matthias Clasen <mclasen@redhat.com>
805 * configure.in: Bump version to 2.9.0
807 2005-08-25 Tor Lillqvist <tml@novell.com>
809 Make also the g_spawn*() functions take parameters in the GLib
810 file name encoding, i.e. UTF-8, on Windows. Has no impact on Unix
811 API or ABI. Like the other GLib API that was earlier changed to
812 use UTF-8 on Windows, the names of the functions that take UTF-8
813 have _utf8 suffixes added by using preprocessor macros in the
814 header file. The old names are kept for functions with the old
815 behaviour, taking parameters in the system codepage, for DLL ABI
818 * glib/gspawn.h: On Win32 add the suffix _utf8 to the names of the
819 g_spawn*() functions.
821 * glib/gspawn-win32.c: Use wide-char API on NT-based
822 Windows. Convert parameters from UTF-8 to wide chars (NT) or
823 system codepage (Win9x) and call the C library _wspawn*() or
824 spawn*() functions respectvely. Add DLL ABI stability versions
825 that take parameters in the system codepage.
827 * glib/gspawn-win32-helper.c: On NT-based Windows use the
828 wide-char versions of argv and envp, and use wide-char API to
829 change directory and spawn the program to run. Remove the verbose
830 debugging output, it was too complex to modify for the wide-char
831 features. (Just add temporary debugging printouts if needed, no
832 need to have them permanently in the source.)
834 * glib/gspawn.c: Corresponding documentation updates.
836 * glib/glib.symbols: Corresponding changes: Mark the ABI stability
837 symbols as PRIVATE, add the new _utf8-suffixed ones.
839 2005-08-24 Stepan Kasal <kasal@ucw.cz>
841 * glib/gtypes.h (G_MININT64): Cast the constant to gint64; it is
842 guint64 otherwise and that can produce warnings about comparison
843 between signed and unsigned.
845 2005-08-23 Matthias Clasen <mclasen@redhat.com>
847 * glib/gutils.c: Fix the crt_externs.h include.
849 2005-08-23 Stepan Kasal <kasal@ucw.cz>
851 * NEWS: Fix spelling of my first name.
853 2005-08-23 Matthias Clasen <mclasen@redhat.com>
857 * === Released 2.8.1 ===
861 2005-08-20 Hans Breuer <hans@breuer.org>
863 * glib/makefile.msc.in : link with ws2_32.lib
865 2005-08-18 Tor Lillqvist <tml@novell.com>
867 * configure.in: Check for <sys/wait.h>
869 * glib/gbacktrace.c: Include <sys/wait.h> on if HAVE_SYS_WAIT_H.
871 2005-08-18 Ross Burton <ross@burtonini.com>
874 Optimise single-character insertions.
879 * tests/string-test.c:
880 Add tests for new optimisation, and fix a leak.
882 2005-08-17 Matthias Clasen <mclasen@redhat.com>
884 * configure.in: Check for crt_externs.h and _NSGetEnviron.
886 * glib/gutils.c: On Darwin, include crt-externs.h and
887 define environ using _NSGetEnviron(). (#313731)
889 2005-08-16 Stepan Kasal <kasal@ucw.cz>
891 * glib/gutils.c (g_get_any_init): Move the body of the big if...
892 (g_get_any_init_do): ... to this new function.
893 (g_get_any_init): Declare as inline.
894 (g_get_any_init_locked): New inline function, does the locking.
895 Make use of these two throughout the code.
897 2005-08-15 Matthias Clasen <mclasen@redhat.com>
899 * glib/gbacktrace.c (g_on_error_stack_trace): Wait for
900 the child process and then simply return. This makes
901 The "S" option work as documented in g_on_error_query().
902 (#313125, Matthew F. Barnes)
904 * glib/gunicode.h: Update the link to Unicode category
905 values. (#313369, Behnam Esfahbod)
907 * glib/gqueue.c (g_queue_find_custom): Clarify docs
908 a little. (#311727, Tristan van Berkom)
910 * glib/abicheck.sh, gobject/abicheck.sh: Make the
911 check work on ia64 too, where some symbols ended up
912 in yet another section.