1 Wed Jan 4 13:33:25 2006 Tim Janik <timj@gtk.org>
3 * glib/gslice.c (magazine_cache_pop_magazine): don't reverse chunk
4 order when creating magazines, so we hand out chunks with ascending
7 2006-01-04 Abel Cheung <maddog@linuxhall.org>
9 * configure.in: Added "ml" "zh_HK" to ALL_LINGUAS.
11 2006-01-03 Martyn Russell <martyn@imendio.com>
13 * docs/reference/glib/glib-sections.txt:
15 * glib/gthreadpool.[ch]:
16 - Added new API g_thread_pool_get_idle_time() and
17 g_thread_pool_set_idle_time(). (#324228).
19 * tests/threadpool-test.c:
20 - Updated test case to do thread pool sorting, thread pool with
21 no sorting and a thread pool with idle thread timeouts.
23 2006-01-03 Matthias Clasen <mclasen@redhat.com>
25 * glib/gmain.h: Add new functions here, too.
27 * glib/glib.symbols: Add new functions.
29 2005-12-20 Michael Meeks <michael.meeks@novell.com>
31 * glib/gmain.c (g_main_context_is_owner): new method
32 to determine if the current thread is the owner of the
35 2006-01-02 Matthias Clasen <mclasen@redhat.com>
39 * glib/gstdio.c (g_rmdir): Don't provide g_rmdir() as a macro
40 expanding to rmdir, since rmdir is not declared in a portable
41 system header we can include in gstdio.h. (#325249, Jani Monoses)
43 2006-01-01 Tor Lillqvist <tml@novell.com>
45 * glib/gspawn-win32.c (g_spawn_sync_utf8): Set the GIOChannels for
46 stdout and stderr to unbuffered. Otherwise the giochannel layer
47 will try to read from them regardless whether the
48 g_io_channel_win32_poll() call here has indicated
49 readability or not. (#325310)
51 2005-12-29 Matthias Clasen <mclasen@redhat.com>
53 * glib/gutils.c (glib_check_version): Fix a copy-and-paste error
54 in the docs. (#325273, Declan Naughton)
56 2005-12-27 Manish Singh <yosh@gimp.org>
58 * tests/run-collate-tests.sh: use LC_ALL instead of LC_COLLATE, to
59 make sure we really override things.
61 2005-12-27 Matthias Clasen <mclasen@redhat.com>
63 Fix #316221, Michal Benes, Stanislav Brabec;
65 * configure.in: Fix a strict aliasing problem in
66 g_static_mutex_get_mutex().
67 * glib/gthread.h: ...and in
68 g_static_mutex_get_mutex_impl_shortcut().
70 * glib/gdatasetprivate.h: Add a cast to silence compiler
71 warnings. (#321978, Andrew Paprocki)
73 Partial fix for bug #32937, Bogdan Nicula.
75 * configure.in: Check for malloc.h
76 * glib/gslice.c: Don't include malloc.h unconditionally.
78 2005-12-27 Manish Singh <yosh@gimp.org>
80 * tests/run-collate-tests.sh: export LC_COLLATE so it takes.
81 (#324950, Dan Yefimov)
83 2005-12-26 Matthias Clasen <mclasen@redhat.com>
85 * glib/gslice.c: On Win32, include process.h (#325015, Kazuki
88 2005-12-25 Matthias Clasen <mclasen@redhat.com>
90 * AUTHORS: Update my email
92 * tests/slice-test.c: Fix C99isms. (#324950, Dan Yefimov)
94 2005-12-21 Matthias Clasen <mclasen@redhat.com>
96 * README.in: Add some notes on when 'make check' may fail.
98 2005-12-20 Matthias Clasen <mclasen@redhat.com>
101 * glib/gthreadpool.h:
102 * glib/gthreadpool.c (g_thread_pool_set_sort_function): New function
103 to sort tasks pushed into a threadpool. (#324479, Martyn Russell)
105 * tests/threadpool-test.c: Test this.
107 Tue Dec 20 18:14:14 2005 Tim Janik <timj@imendio.com>
109 * glib/gslice.[hc]: added mem_error() and mem_assert() to test and
110 handle errors without depending on gmessage.c which might not be
111 setup when the error occours.
112 removed G_SLICE_CONFIG_ALWAYS_FREE config option, fixed the code so
113 always freeing can be achieved by adjusting the working set time to
114 0 with G_SLICE_CONFIG_WORKING_SET_MSECS.
115 added G_SLICE_CONFIG_COLOR_INCREMENT to test different color increments
116 (mainly 0 and 1). reduced the minimum block size to 128 bytes, to
117 minimize wastage if small amounts of differently sized structrues are
118 allocated, this does come at a performance cost of roughly 5% though.
119 fixed up block alignment calculation, so it works for varying
120 block sizes. only use strerror() not g_strerror() since the latter
121 depends on working GQuark and GSlice.
122 mem_error(): implemented in terms of fprintf and vfprintf.
124 * tests/slice-color.c: new program to test cache colorization effects.
126 * tests/slice-test.c: trade G_SLICE_CONFIG_ALWAYS_FREE for 0 duration
127 G_SLICE_CONFIG_WORKING_SET_MSECS.
129 2005-12-17 Matthias Clasen <mclasen@redhat.com>
131 * glib/goption.c (parse_short_option): Set an error in all
132 failure cases. (#324332, Tim-Philipp Müller)
134 2005-12-17 Sebastian Wilhelmi <seppi@seppi.de>
136 * glib/gatomic.c: Fix memory barrier position in g_atomic_int_get
137 and g_atomic_pointer_get. Add g_atomic_int_set and
138 g_atomic_pointer_set implementations for the !DEFINE_WITH_MUTEXES &&
139 G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case, as well as defining them
140 as functions (additionally to the macros in the header) for the
141 !G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case.
143 2005-12-16 Matthias Clasen <mclasen@redhat.com>
145 * glib/gmem.c (g_allocator_new): Don't return a pointer to
146 a const struct, since apps expect to be able to modify it.
147 (#324179, J. Ali Harlow)
149 Tue Dec 13 10:13:32 2005 Tim Janik <timj@imendio.com>
151 * glib/gatomic.h: added g_atomic_pointer_set() and g_atomic_int_set()
152 for systems where the initialization of atomic variables requires a
153 write memory barrier.
155 2005-12-09 Matthias Clasen <mclasen@redhat.com>
157 * configure.in: Bump version
159 * === Released 2.9.1 ===
161 2005-12-09 Alexander Larsson <alexl@redhat.com>
165 Use G_GNUC_WARN_UNUSED_RESULT on list functions that return
168 2005-12-08 Matthias Clasen <mclasen@redhat.com>
172 2005-12-07 Martyn Russell <martyn@imendio.com>
174 * glib/gasyncqueue.c:
175 - Call g_queue_insert_sorted() instead of duplicating the code.
176 - Call g_queue_sort() instead of duplicating the code.
177 - Invert sort function results to make sure the same sort function
178 gives the same results across glist, gslist, gqueue and
181 * tests/asyncqueue-test.c:
182 - Updated the sort function to reflect the example in the
183 documentation for gasyncqueue.c.
185 2005-12-07 Martyn Russell <martyn@imendio.com>
187 * docs/reference/glib/glib-sections.txt:
188 * docs/reference/glib/tmpl/linked_lists_double.sgml:
189 * docs/reference/glib/tmpl/linked_lists_single.sgml:
192 - Added g_list_insert_sorted_with_data () and
193 g_slist_insert_sorted_with_data ().
194 - Removed the extra check in g_list_sort() and g_slist_sort() for
195 GCompareDataFunc vs. GCompareFunc.
197 2005-12-07 Tor Lillqvist <tml@novell.com>
199 * glib-gettextize.in: Look up prefix at run-time on Win32,
200 assuming the standard directory structure with glib-gettextize in
203 2005-12-06 Matthias Clasen <mclasen@redhat.com>
205 * glib/gmem.h: Exempt GTK+ from the mem chunk deprecation, since
206 we need GTK+ 2.8 to compile against GLib 2.10.
208 2005-12-05 Manish Singh <yosh@gimp.org>
210 * tests/run-collate-test.sh: set LC_COLLATE, not LANG, to be sure
211 to override any user settings.
213 2005-12-05 Matthias Clasen <mclasen@redhat.com>
215 * configure.in: Revert an accidential commit
217 2005-12-05 Matthias Clasen <mclasen@redhat.com>
219 * glib/gutf8.c: Documentation updates. (#323291, Morten Welinder)
221 * tests/Makefile.am (TESTS_ENVIRONMENT):
222 * tests/gobject/Makefile.am (TESTS_ENVIRONMENT): Set
223 MALLOC_CHECK_ and MALLOC_PERTURB_
225 * tests/run-collate-tests.sh: Run the collation tests explicitly
226 in en_US locale. (#320463)
228 * glib/gmem.h: Really deprecate GMemChunk.
235 * glib/gstrfuncs.c: Add versioned deprecation docs.
237 2005-12-05 Martyn Russell <martyn@imendio.com>
239 * docs/reference/glib/glib-sections.txt:
240 * glib/gasyncqueue.[ch]:
241 - Added support for sorting async queues by with _push_sorted(),
242 _push_sorted_unlocked(), _sort() and _sort_unlocked() (#323047).
245 * tests/asyncqueue-test.c:
246 - Added test case for gasyncqueue.c
248 Mon Dec 5 15:53:20 2005 Tim Janik <timj@imendio.com>
250 * glib/gslice.c: implement chain walking for arbitrary ->next pointer
251 offsets in g_slice_free_chain_with_offset() based on a patch by behdad
252 in bug 323178. moved time consuming logic from g_slice_free() out of
253 the inner loop, so g_slice_free_chain_with_offset() provides a real
254 performance benefit over g_slice_free1() now.
256 * glib/gslice.h: renamed g_slice_free_chain() to
257 g_slice_free_chain_with_offset(). implemented g_slice_free_chain() as
258 a type-safe macro as suggested in bug 323178.
259 simplified the macro implementation of g_slice_free() and implemented
260 it in a type safe manner for all compliers as suggested by Morten
261 Welinder <mortenw@gnome.org>.
266 * glib/glib.symbols: s/g_slice_free_chain/g_slice_free_chain_with_offset/
268 2005-12-05 Matthias Clasen <mclasen@redhat.com>
270 * glib/gasyncqueue.c: Add some docs.
272 * tests/libmoduletestplugin_a.c: Fix compiler warnings.
274 * glib/gatomic.c: In the ia64 implementation, use
275 __sync builtin without _si or _di suffix. (#321229,
276 Stanislav Brabec, patch by Andreas Schwab)
278 2005-12-04 Behdad Esfahbod <behdad@gnome.org>
280 * glib/gslice.h: Remove comma at the end of enum.
282 2005-12-04 Matthias Clasen <mclasen@redhat.com>
284 Handle multiple user names with the same UID better.
285 (#319535, Laszlo Peter)
287 * glib/gutils.c (g_get_any_init_do): When determining user
288 data, first look up $LOGNAME. If the UID doesn't match
289 getuid(), fall back to the current behaviour of looking
290 up the user data based on getuid().
292 2005-12-04 Matthias Clasen <mclasen@redhat.com>
294 * glib/gmacros.h (G_GNUC_WARN_UNUSED_RESULT): Add a macro
295 to make gcc warn if a function result is ignored. (#145466,
296 Arjan van de Ven, Alex Larsson)
298 * glib/gmem.h: Add the new attribute to g_realloc and
301 2005-12-03 Matthias Clasen <mclasen@redhat.com>
304 * glib/gcache.h: Deprecate g_cache_value_foreach. (#322956,
307 * glib/gmappedfile.c: Make mapping of empty files
310 * glib/gfileutils.c: Don't fork a new process just to
311 fix the permissions of the created temp file. (#321318,
312 Alexis S. L. Carvalho)
314 2005-12-02 Matthias Clasen <mclasen@redhat.com>
316 * README.in: Add a note about Solaris threads.
319 * configure.in: Remove support for Solaris threads.
320 (#136971, Sebastian Wilhelmi, patch by Andrew Paprocki)
322 2005-12-02 Matthias Clasen <mclasen@redhat.com>
324 * configure.in: Define G_GUINT64_CONSTANT in analogy to
325 G_GINT64_CONSTANT. (#322568, Andrew Paprocki)
327 2005-12-02 Matthias Clasen <mclasen@redhat.com>
329 * glib/gslice.c: Win32 portability fixes and C99ism removal,
330 pointed out by Kazuki Iwamoto. (#323052)
332 2005-12-02 Matthias Clasen <mclasen@redhat.com>
334 * glib/gatomic.c (g_atomic_int_add): Add a missing volatile for
335 the IA64 implementation.
337 Fri Dec 2 16:18:09 2005 Tim Janik <timj@imendio.com>
339 * glib/gslice.c: some naming and type size fixups.
341 Fri Dec 2 13:08:58 2005 Tim Janik <timj@gtk.org>
343 * glib/gslice.h (g_slice_free): reworked GCC-specific type-safe macro
344 variant into something less verbose, but digestible for gcc-3.4.
346 Fri Dec 2 10:55:07 2005 Tim Janik <timj@imendio.com>
348 * tests/slice-test.c: extended to perform the benchmarking on the old
349 memchunk code if 'O' is selected.
351 * tests/memchunks.c: new file which contains the old GLib mem chunks
352 implementation with prefix old_mem_chunk_.
354 * tests/Makefile.am: added memchunks.c
356 Fri Dec 2 00:16:59 2005 Tim Janik <timj@imendio.com>
358 * glib/gslice.c: improved gettimeofday() timer resolution by maintaining
359 acceptable load distribution of the syscall.
360 (allocator_get_magazine_threshold): reduce minimum/single-thread magazine
361 sizes to approximately page_size/2.5 to avoid excessive startup allocations.
363 2005-12-01 Matthias Clasen <mclasen@redhat.com>
365 * tests/gobject/ifacecheck.c (test_iface_base_init): Don't
366 double-free base interface members.
368 * tests/Makefile.am: Remove duplicate variable.
370 Thu Dec 1 17:32:46 2005 Tim Janik <timj@imendio.com>
372 * glib/gslice.[hc]: new slice allocator implementation.
374 * tests/slice-test.c: added random slice allocation test.
376 * glib/gthread.[hc]: removed newly added private thread mem API.
378 * glib/gthreadinit.h:
381 * glib/gmem.c: divided glib threading initialisation into three phases,
382 initialisation where private keys and messaging are not available (only
383 needed by gmem.c), initialisation without messaging but private keys
384 available (gslice.c, gmessage.c), and full fledged initialisers that
385 server the rest of glib. initialisation functions got renamed to reflect
386 the limitations of their corresponding phases.
388 * glib/gmem.c: removed memchunk code, defer allocations to
391 * glib/gmem.[hc]: removed g_slice_* skeletons.
393 * glib/glib.symbols: added g_slice_* symbols.
395 * configure.in: check for availability of posix_memalign(3), memalign(3)
398 * glib/Makefile.am: added gslice.[hc].
400 2005-12-01 Tor Lillqvist <tml@novell.com>
402 * glib/gstdio.c (g_stat): In the Win32 implementation, strip
403 trailing slash(es) for non-root folders. stat() fails if non-root
404 folders are specified with trailing slashes. It's too much hassle
405 to demand that callers strip such slashes themselves, especially
406 as it is easy to get it wrong and strip the slash of a root
408 (g_rename): On NT-based Windows, use MoveFileEx() with
409 MOVEFILE_REPLACE_EXISTING to better match Unix behaviour.
411 2005-11-28 Matthias Clasen <mclasen@redhat.com>
413 Fix G_STMT_START / G_STMT_END on Solaris. (#321972,
416 * configure.in: Check whether do { } while (0) works.
418 * glib/gmacros.h: Use do { } while (0) for G_STMT_START /
419 G_STMT_END if it works.
421 2005-11-28 Matthias Clasen <mclasen@redhat.com>
423 * glib/gthread.c (g_static_rw_lock_wait, g_static_rw_lock_signal):
424 * glib/gnode.c (g_node_depth_traverse_level):
425 * glib/gmem.c (g_allocator_new):
426 * glib/ghash.c (g_hash_table_unref, g_hash_table_destroy)
427 (g_hash_table_foreach_remove_or_steal): Silence compiler
430 2005-11-27 Matthias Clasen <mclasen@redhat.com>
432 * glib/gunicollate.c (g_utf8_collate_key_for_filename):
433 Don't read beyond len. (#322520, Christian Persch)
435 Wed Nov 23 17:34:01 2005 Tim Janik <timj@imendio.com>
437 * glib/gdataset.c: access datalist flags via atomic pointer access
438 functions, instead of acquiring the dataset lock. this is faster and
439 also matches the atomic pointer readouts in gdatalistprivate.h and
440 direct pointer modifications required by gobject.c.
442 Wed Nov 23 13:35:31 2005 Tim Janik <timj@gtk.org>
444 * glib/gdataset.c: streamlined initialization code somewhat,
445 removed GData node cache.
447 2005-11-22 Matthias Clasen <mclasen@redhat.com>
449 * glib/ghash.c (g_hash_table_ref):
450 (g_hash_table_unref): Mark these as new API.
452 Avoid double locking in g_intern_string (#322133,
455 * glib/gdataset.c (g_quark_from_string_internal): New
456 internal function which factors out common parts of
457 g_quark_from[_static]_string.
458 (g_quark_from_string, g_quark_from_static_string):
459 Use g_quark_from_string_internal.
460 (g_intern_string, g_intern_static_string): Use
461 g_quark_from_string_internal, and only take the
463 (g_quark_new): Don't store the strings shifted by -1
464 in the g_quarks array.
465 (g_quark_to_string): Adapt to the previous change.
467 Tue Nov 22 14:04:26 2005 Tim Janik <timj@imendio.com>
471 g_hash_table_new_full(): create hash tables with a ref count of 1.
472 g_hash_table_ref(): atomically ref_count+=1
473 g_hash_table_unref(): atomically ref_count-=1, destroys hash table
474 when refcount reaches 0.
475 g_hash_table_destroy(): just destroy keys and values, unref by 1.
476 g_hash_table_insert():
477 g_hash_table_replace(): assert ref_count>0.
480 * glib/gatomic.c: added 'volatile' qualifier to all atomic pointer and
483 2005-11-20 Behdad Esfahbod <behdad@gnome.org>
485 * glib/guniprop.c (g_unichar_get_mirror_char): Remove unused
486 variables. (#321984, Andrew Paprocki)
488 2005-11-18 Matthias Clasen <mclasen@redhat.com>
490 * configure.in: Bump version
492 2005-11-17 Matthias Clasen <mclasen@redhat.com>
494 * === Released 2.9.0 ===
496 * glib/glib.symbols: Updates
498 * glib/gutf8.c (g_utf8_offset_to_pointer): Add some performance
499 hints to the docs. (Paolo Borelli)
503 * glib/gthread.c (g_thread_foreach): Mark as new api.
505 * README.in: Updates.
509 * glib/gdate.c (g_date_set_time_t):
510 (g_date_set_time_val): Add functions to set a date from
511 a time_t and from a GTimeVal.
512 (g_date_set_time): Deprecate. (#314067, Roger Leigh)
515 * tests/utf8-pointer.c: Unit tests for g_utf8_pointer_to_offset
516 and g_utf8_offset_to_pointer.
518 * glib/gutf8.c (g_utf8_pointer_to_offset)
519 (g_utf8_offset_to_pointer): Handle negative offsets, and use
520 "stutter stepping" for going backwards. (#320638, Larry
525 * glib/gthread.c: const correctness fixes, found
526 by Arjan van de Ven and gcc.
528 2005-11-16 Behdad Esfahbod <behdad@gnome.org>
530 * docs/reference/glib/tmpl/unicode.sgml: Correct typo on mentioning
531 @G_UNICODE_BREAK_UNKNOWN that should be #G_UNICODE_BREAK_UNKNOWN.
533 2005-11-09 Behdad Esfahbod <behdad@gnome.org>
535 * glib/guniprop.c: Use bit hacks instead when checking a general
536 category value against multiple values.
538 * glib/gutf8.c: Change ISO10646 to Unicode in docs.
540 2005-11-10 Simos Xenitellis <simos@gnome.org>
542 * configure.in: Added tt (Tatar) to ALL_LINGUAS.
544 2005-11-05 Tor Lillqvist <tml@novell.com>
546 * glib/giowin32.c (g_io_win32_set_flags): Don't call g_warning().
547 (#320688) Instead set error and return failure.
549 2005-11-07 Matthias Clasen <mclasen@redhat.com>
551 * glib/gmain.c (g_main_dispatch): Don't call cb_funcs->unref
552 while holding the context lock. (#320886, Andy Wingo)
554 2005-11-07 Billy Biggs <vektor@dumbterm.net>
556 * tests/refcount/Makefile.am: Add a missing reference to libglib
557 in the LDADD for this test.
559 2005-11-07 Matthias Clasen <mclasen@redhat.com>
561 * glib/gmem.c (g_allocator_new): We need to set n_preallocs to a
562 nonzero value, otherwise GTK+ 2.8 breaks when compiled against
563 GLib 2.9. (#320755, Luca Ferretti)
565 2005-11-04 Matthias Clasen <mclasen@redhat.com>
567 * glib/Makefile.am: Apply a patch from Behdad Esfahbod to
568 use a faster lookup table for g_unichar_get_mirror_char().
569 * glib/gmirroringtable.h: The new table.
570 * glib/gunichartables.h: Remove bidi_mirroring_table.
571 * glib/gen-unicode-tables.pl: Don't generate the mirroring
573 * glib/glib-mirroring-tab/*: Sources for the program
574 which generated gmirroringtable.h.
576 * glib/glist.c: Avoid some code duplication.
578 * glib/gscanner.h: Include gdataset.h. (#320322)
580 * glib/gdebug.h: add new GDebugFlag for fatal_criticals
581 * glib/gmessages.c: (_g_debug_init): handle G_DEBUG=fatal_criticals,
582 to help find critical warnings in applications. (#320017,
585 2005-11-02 Tor Lillqvist <tml@novell.com>
587 * glib/glib.symbols: Remove large amount of trailing whitespace
588 from one line. Remove #ifdef G_UNDEFINED from around
589 g_hash_table_new and g_hash_table_new_full symbols, what was the
590 purpose of that? Add the g_slice functions.
592 Tue Nov 1 16:24:20 2005 Tim Janik <timj@imendio.com>
594 * glib/gmem.[hc]: prepared deprecation of GMemChunk and GAllocator.
595 added g_slice_*() API to allocate and cache small bits of memory.
596 an actuall allocator implementation for g_slice_*() is still pending.
598 * glib/gthread.[hc]: changes from a patch by Matthias Clasen.
599 changed GRealThread list to use in-structure *next; fields instead
600 of GSList, in order for thread iteration to not depenend on g_slice_*()
602 _g_thread_mem_private_get():
603 _g_thread_mem_private_set(): added accessors for private memory,
604 needed because the ordinary GPrivate implementation relies on GArray
605 and GSList and therefore indirectly on working g_slice_*() allocations.
608 g_thread_foreach(): new public API function to loop over all existing threads.
622 removed GAllocator and free list usages and accompanying locks.
623 use g_slice_*() API to allocate and cache small bits of memory.
625 * glib/ghook.h: removed GMemChunk field from public API.
628 * glib/glist.h: deprecate allocator API, provide _free1() for consistency.
630 * glib/gnode.h: deprecate allocator API.
632 * glib/gmain.c: reordered GPollRec fields so g_slice_free_chain() can
633 be used for poll rec lists.
635 * glib/grel.c: removed mem chunk usage, and allocated tuples via g_slice_*().
636 g_relation_destroy(): free all tuples from the all_tuples hash table,
637 this effectively maintains the life time track keeping of tuples.
638 g_relation_delete_tuple(): free tuples which are removed from the
639 all_tuples hash table. this fixes a temporary leak that was present
640 in the memchunk code until the destruction of the relation.
642 2005-10-29 Matthias Clasen <mclasen@redhat.com>
644 * tests/convert-test.c: Add some tests for conversions between
645 UTF-8, UCS-4 and UTF-16.
647 * glib/gutf8.c (g_utf8_to_ucs4, g_utf8_to_utf16): Fix handling
648 of len == -1, noticed by Morten Welinder.
650 2005-10-27 Erdal Ronahi <erdal.ronahi@gmail.com>
652 * configure.in: Added ku (Kurdish) to ALL_LINGUAS
654 2005-10-26 Matthias Clasen <mclasen@redhat.com>
656 * glib/gutf8.c (g_ucs4_to_utf8): Don't set items_read twice
657 in the error case, and add some documentation. (#319806, Morten
660 2005-10-19 Manish Singh <yosh@gimp.org>
662 * configure.in: Use AC_CHECK_FUNCS for _NSGetEnviron, to get the
663 config.h symbol automatically. Fixes bug #313731.
665 2005-10-19 Tor Lillqvist <tml@novell.com>
667 * glib/gwin32.c (g_win32_get_package_installation_directory):
668 Return a g_strdup()ed copy of the value stored in the hash table,
669 so that it can be g_free()d without leaving a dangling pointer in
670 the hash table. (#319232)
672 2005-10-06 Matthias Clasen <mclasen@redhat.com>
674 * glib/gunicollate.c (g_utf8_collate_key_for_filename): Handle
675 all-zero sequences correctly. (#317930, Sebastien Bacher)
677 2005-10-05 Matthias Clasen <mclasen@redhat.com>
679 * glib/gmarkup.c (g_markup_escape_text): Doc addition
681 * Makefile.am: Add Collation tests.
683 * tests/collate/*: Inputs and expected outputs for collation tests.
685 * tests/run-collate-tests.sh: Script to run collation tests.
687 * tests/unicode-collate.c (main): Rework slightly to make
688 it usable in unit tests. Also test g_utf8_collate_key_for_filename().
690 2005-10-01 Behdad Esfahbod <behdad@gnome.org>
692 * docs/reference/glib/tmpl/unicode.sgml:
693 * glib/gen-unicode-tables.pl:
695 * glib/gunichartables.h:
697 * tests/casefold.txt:
698 * tests/casemap.txt: Updated to Unicode 4.1. There are five new
699 GUnicodeBreakType types. That may break some applications, like
702 2005-09-26 Matthias Clasen <mclasen@redhat.com>
704 * glib/gstrfuncs.c (g_ascii_strtoull): Add details to the
705 docs. (#314393, Matthew F. Barnes)
708 * glib/gprintf.h: Remove g_snprintf() and g_vsnprintf(), since
709 they are already declared in glib.h. This doesn't break documented
710 use of gprintf.h, but should probably be pointed out in the
711 release notes for 2.10. (#314232, Behdad Esfahbod)
713 Tue Sep 20 13:16:04 2005 Tim Janik <timj@imendio.com>
715 * glib/gpattern.c (g_pattern_ph_match): applied significant recursion
716 complexity optimization, based on a patch from Matthias Clasen.
718 * tests/patterntest.c: more tests, mostly from matthias.
720 2005-09-20 Matthias Clasen <mclasen@redhat.com>
722 * glib/gqueue.c (g_queue_insert_sorted): Correct the docs.
723 (#316703, Mark Drago)
725 Mon Sep 19 17:23:23 2005 Tim Janik <timj@imendio.com>
727 * glib/gpattern.c: applied a patch from matthias which checks on the
728 upper bound of GPatternSpec length to optimize matches.
731 * tests/patterntest.c: added more match cases.
733 2005-09-16 Tor Lillqvist <tml@novell.com>
735 * glib/gstrfuncs.c (g_ascii_strcasecmp, g_ascii_strncasecmp): Add
736 warning to doc comment that these functions should not be used on
737 encodings like CP932.
739 2005-09-14 Matthias Clasen <mclasen@redhat.com>
741 * tests/keyfile-test.c: Add a test for grup names of length 1.
743 * glib/gkeyfile.c (g_key_file_line_is_group): Accept group names
744 of length 1. (#316309)
746 2005-09-12 Matthias Clasen <mclasen@redhat.com>
748 * glib/gmarkup.c (g_markup_escape_text): Clarify docs.
750 2005-09-11 Sebastian Wilhelmi <seppi@seppi.de>
752 * tests/refcount/Makefile.am (INCLUDES): Link the the refcount
753 tests to the system thread library $(G_THREAD_LIBS). Fixes #313744
756 2005-09-11 Kjartan Maraas <kmaraas@gnome.org>
758 * glib/gmain.c: (g_child_watch_prepare), (g_child_watch_check),
759 (child_watch_helper_thread): Remove some dead code. Closes
762 2005-09-07 Tor Lillqvist <tml@novell.com>
764 * glib/Makefile.am: Create also a console version of the
765 gspawn-win32-helper program, gspawn-win32-helper-console.exe.
766 It's otherwise identical to gspawn-win32-helper.exe, except marked
767 as a console application (linked without the -mwindows option).
769 * glib/gspawn-win32.c (do_spawn_directly, do_spawn_with_pipes):
770 Drop the dont_wait parameter. Its truth value correlated 100% with
771 the NULLness of the exit_status parameter anyway, so it's enough
772 to check whether exit_status is NULL. Invert the sense of the
773 dont_return_handle parameter and rename it to do_return_handle, to
774 make the code easier to read by avoiding double negations.
776 (g_spawn_sync_utf8, g_spawn_async_with_pipes_utf8): Modify calls
777 to do_spawn_with_pipes() accordingly.
779 (do_spawn_with_pipes): If we have a console, use the console
780 version of the helper program, otherwise use the GUI one. This
781 avoids extra console windows opening up in some situations. (In
782 case a console application uses the GUI gspawn-win32-helper.exe to
783 spawn another console application we would get a separate console
784 for the spawned console application).
786 * glib-zip.in: Distribute also gspawn-win32-helper-console.exe.
788 2005-09-05 Matthias Clasen <mclasen@redhat.com>
790 * glib/gmappedfile.c (g_mapped_file_new): Report an error
791 if the file is too large. (#315275, Kjartan Maraas)
793 * glib/gkeyfile.c (g_key_file_load_from_fd): The return value
794 of read() is signed. (#315273, Kjartan Maraas)
796 2005-08-31 Tor Lillqvist <tml@novell.com>
798 * glib/gutils.h: Wrapping atexit() is a bad idea on Windows, where
799 the EXE and each DLL have their own atexit function chains.
801 #define g_atexit as atexit instead. This means it has a
802 better chance of doing what the caller wants. For instance,
803 gtkhtml calls g_atexit() registering a function in gtkhtml
804 itself. This caused a crash when g_atexit() was implemented as a
805 function in the GLib DLL. The gtkhtml DLL was already unloaded by
806 the time the GLib DLL got unloaded.
808 * glib/gutils.c: #undef the #define mentioned above, to also get a
809 real g_atexit() into the DLL for backward compatibility. Document
810 the Windows behaviour of g_atexit(), and document the varying ways
811 atexit() can behave in the context of dynamically loaded modules
814 2005-08-31 Matthias Clasen <mclasen@redhat.com>
818 * glib/gdataset.c: Add string interning functions.
820 2005-08-28 Matthias Clasen <mclasen@redhat.com>
822 * glib/giochannel.c: Unify some near-duplicate strings. (#314654,
825 2005-08-26 Matthias Clasen <mclasen@redhat.com>
827 * configure.in: Bump version to 2.9.0
829 2005-08-25 Tor Lillqvist <tml@novell.com>
831 Make also the g_spawn*() functions take parameters in the GLib
832 file name encoding, i.e. UTF-8, on Windows. Has no impact on Unix
833 API or ABI. Like the other GLib API that was earlier changed to
834 use UTF-8 on Windows, the names of the functions that take UTF-8
835 have _utf8 suffixes added by using preprocessor macros in the
836 header file. The old names are kept for functions with the old
837 behaviour, taking parameters in the system codepage, for DLL ABI
840 * glib/gspawn.h: On Win32 add the suffix _utf8 to the names of the
841 g_spawn*() functions.
843 * glib/gspawn-win32.c: Use wide-char API on NT-based
844 Windows. Convert parameters from UTF-8 to wide chars (NT) or
845 system codepage (Win9x) and call the C library _wspawn*() or
846 spawn*() functions respectvely. Add DLL ABI stability versions
847 that take parameters in the system codepage.
849 * glib/gspawn-win32-helper.c: On NT-based Windows use the
850 wide-char versions of argv and envp, and use wide-char API to
851 change directory and spawn the program to run. Remove the verbose
852 debugging output, it was too complex to modify for the wide-char
853 features. (Just add temporary debugging printouts if needed, no
854 need to have them permanently in the source.)
856 * glib/gspawn.c: Corresponding documentation updates.
858 * glib/glib.symbols: Corresponding changes: Mark the ABI stability
859 symbols as PRIVATE, add the new _utf8-suffixed ones.
861 2005-08-24 Stepan Kasal <kasal@ucw.cz>
863 * glib/gtypes.h (G_MININT64): Cast the constant to gint64; it is
864 guint64 otherwise and that can produce warnings about comparison
865 between signed and unsigned.
867 2005-08-23 Matthias Clasen <mclasen@redhat.com>
869 * glib/gutils.c: Fix the crt_externs.h include.
871 2005-08-23 Stepan Kasal <kasal@ucw.cz>
873 * NEWS: Fix spelling of my first name.
875 2005-08-23 Matthias Clasen <mclasen@redhat.com>
879 * === Released 2.8.1 ===
883 2005-08-20 Hans Breuer <hans@breuer.org>
885 * glib/makefile.msc.in : link with ws2_32.lib
887 2005-08-18 Tor Lillqvist <tml@novell.com>
889 * configure.in: Check for <sys/wait.h>
891 * glib/gbacktrace.c: Include <sys/wait.h> on if HAVE_SYS_WAIT_H.
893 2005-08-18 Ross Burton <ross@burtonini.com>
896 Optimise single-character insertions.
901 * tests/string-test.c:
902 Add tests for new optimisation, and fix a leak.
904 2005-08-17 Matthias Clasen <mclasen@redhat.com>
906 * configure.in: Check for crt_externs.h and _NSGetEnviron.
908 * glib/gutils.c: On Darwin, include crt-externs.h and
909 define environ using _NSGetEnviron(). (#313731)
911 2005-08-16 Stepan Kasal <kasal@ucw.cz>
913 * glib/gutils.c (g_get_any_init): Move the body of the big if...
914 (g_get_any_init_do): ... to this new function.
915 (g_get_any_init): Declare as inline.
916 (g_get_any_init_locked): New inline function, does the locking.
917 Make use of these two throughout the code.
919 2005-08-15 Matthias Clasen <mclasen@redhat.com>
921 * glib/gbacktrace.c (g_on_error_stack_trace): Wait for
922 the child process and then simply return. This makes
923 The "S" option work as documented in g_on_error_query().
924 (#313125, Matthew F. Barnes)
926 * glib/gunicode.h: Update the link to Unicode category
927 values. (#313369, Behnam Esfahbod)
929 * glib/gqueue.c (g_queue_find_custom): Clarify docs
930 a little. (#311727, Tristan van Berkom)
932 * glib/abicheck.sh, gobject/abicheck.sh: Make the
933 check work on ia64 too, where some symbols ended up
934 in yet another section.