1 2006-01-02 Matthias Clasen <mclasen@redhat.com>
5 * glib/gstdio.c (g_rmdir): Don't provide g_rmdir() as a macro
6 expanding to rmdir, since rmdir is not declared in a portable
7 system header we can include in gstdio.h. (#325249, Jani Monoses)
9 2006-01-01 Tor Lillqvist <tml@novell.com>
11 * glib/gspawn-win32.c (g_spawn_sync_utf8): Set the GIOChannels for
12 stdout and stderr to unbuffered. Otherwise the giochannel layer
13 will try to read from them regardless whether the
14 g_io_channel_win32_poll() call here has indicated
15 readability or not. (#325310)
17 2005-12-29 Matthias Clasen <mclasen@redhat.com>
19 * glib/gutils.c (glib_check_version): Fix a copy-and-paste error
20 in the docs. (#325273, Declan Naughton)
22 2005-12-27 Manish Singh <yosh@gimp.org>
24 * tests/run-collate-tests.sh: use LC_ALL instead of LC_COLLATE, to
25 make sure we really override things.
27 2005-12-27 Matthias Clasen <mclasen@redhat.com>
29 Fix #316221, Michal Benes, Stanislav Brabec;
31 * configure.in: Fix a strict aliasing problem in
32 g_static_mutex_get_mutex().
33 * glib/gthread.h: ...and in
34 g_static_mutex_get_mutex_impl_shortcut().
36 * glib/gdatasetprivate.h: Add a cast to silence compiler
37 warnings. (#321978, Andrew Paprocki)
39 Partial fix for bug #32937, Bogdan Nicula.
41 * configure.in: Check for malloc.h
42 * glib/gslice.c: Don't include malloc.h unconditionally.
44 2005-12-27 Manish Singh <yosh@gimp.org>
46 * tests/run-collate-tests.sh: export LC_COLLATE so it takes.
47 (#324950, Dan Yefimov)
49 2005-12-26 Matthias Clasen <mclasen@redhat.com>
51 * glib/gslice.c: On Win32, include process.h (#325015, Kazuki
54 2005-12-25 Matthias Clasen <mclasen@redhat.com>
56 * AUTHORS: Update my email
58 * tests/slice-test.c: Fix C99isms. (#324950, Dan Yefimov)
60 2005-12-21 Matthias Clasen <mclasen@redhat.com>
62 * README.in: Add some notes on when 'make check' may fail.
64 2005-12-20 Matthias Clasen <mclasen@redhat.com>
68 * glib/gthreadpool.c (g_thread_pool_set_sort_function): New function
69 to sort tasks pushed into a threadpool. (#324479, Martyn Russell)
71 * tests/threadpool-test.c: Test this.
73 Tue Dec 20 18:14:14 2005 Tim Janik <timj@imendio.com>
75 * glib/gslice.[hc]: added mem_error() and mem_assert() to test and
76 handle errors without depending on gmessage.c which might not be
77 setup when the error occours.
78 removed G_SLICE_CONFIG_ALWAYS_FREE config option, fixed the code so
79 always freeing can be achieved by adjusting the working set time to
80 0 with G_SLICE_CONFIG_WORKING_SET_MSECS.
81 added G_SLICE_CONFIG_COLOR_INCREMENT to test different color increments
82 (mainly 0 and 1). reduced the minimum block size to 128 bytes, to
83 minimize wastage if small amounts of differently sized structrues are
84 allocated, this does come at a performance cost of roughly 5% though.
85 fixed up block alignment calculation, so it works for varying
86 block sizes. only use strerror() not g_strerror() since the latter
87 depends on working GQuark and GSlice.
88 mem_error(): implemented in terms of fprintf and vfprintf.
90 * tests/slice-color.c: new program to test cache colorization effects.
92 * tests/slice-test.c: trade G_SLICE_CONFIG_ALWAYS_FREE for 0 duration
93 G_SLICE_CONFIG_WORKING_SET_MSECS.
95 2005-12-17 Matthias Clasen <mclasen@redhat.com>
97 * glib/goption.c (parse_short_option): Set an error in all
98 failure cases. (#324332, Tim-Philipp Müller)
100 2005-12-17 Sebastian Wilhelmi <seppi@seppi.de>
102 * glib/gatomic.c: Fix memory barrier position in g_atomic_int_get
103 and g_atomic_pointer_get. Add g_atomic_int_set and
104 g_atomic_pointer_set implementations for the !DEFINE_WITH_MUTEXES &&
105 G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case, as well as defining them
106 as functions (additionally to the macros in the header) for the
107 !G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case.
109 2005-12-16 Matthias Clasen <mclasen@redhat.com>
111 * glib/gmem.c (g_allocator_new): Don't return a pointer to
112 a const struct, since apps expect to be able to modify it.
113 (#324179, J. Ali Harlow)
115 Tue Dec 13 10:13:32 2005 Tim Janik <timj@imendio.com>
117 * glib/gatomic.h: added g_atomic_pointer_set() and g_atomic_int_set()
118 for systems where the initialization of atomic variables requires a
119 write memory barrier.
121 2005-12-09 Matthias Clasen <mclasen@redhat.com>
123 * configure.in: Bump version
125 * === Released 2.9.1 ===
127 2005-12-09 Alexander Larsson <alexl@redhat.com>
131 Use G_GNUC_WARN_UNUSED_RESULT on list functions that return
134 2005-12-08 Matthias Clasen <mclasen@redhat.com>
138 2005-12-07 Martyn Russell <martyn@imendio.com>
140 * glib/gasyncqueue.c:
141 - Call g_queue_insert_sorted() instead of duplicating the code.
142 - Call g_queue_sort() instead of duplicating the code.
143 - Invert sort function results to make sure the same sort function
144 gives the same results across glist, gslist, gqueue and
147 * tests/asyncqueue-test.c:
148 - Updated the sort function to reflect the example in the
149 documentation for gasyncqueue.c.
151 2005-12-07 Martyn Russell <martyn@imendio.com>
153 * docs/reference/glib/glib-sections.txt:
154 * docs/reference/glib/tmpl/linked_lists_double.sgml:
155 * docs/reference/glib/tmpl/linked_lists_single.sgml:
158 - Added g_list_insert_sorted_with_data () and
159 g_slist_insert_sorted_with_data ().
160 - Removed the extra check in g_list_sort() and g_slist_sort() for
161 GCompareDataFunc vs. GCompareFunc.
163 2005-12-07 Tor Lillqvist <tml@novell.com>
165 * glib-gettextize.in: Look up prefix at run-time on Win32,
166 assuming the standard directory structure with glib-gettextize in
169 2005-12-06 Matthias Clasen <mclasen@redhat.com>
171 * glib/gmem.h: Exempt GTK+ from the mem chunk deprecation, since
172 we need GTK+ 2.8 to compile against GLib 2.10.
174 2005-12-05 Manish Singh <yosh@gimp.org>
176 * tests/run-collate-test.sh: set LC_COLLATE, not LANG, to be sure
177 to override any user settings.
179 2005-12-05 Matthias Clasen <mclasen@redhat.com>
181 * configure.in: Revert an accidential commit
183 2005-12-05 Matthias Clasen <mclasen@redhat.com>
185 * glib/gutf8.c: Documentation updates. (#323291, Morten Welinder)
187 * tests/Makefile.am (TESTS_ENVIRONMENT):
188 * tests/gobject/Makefile.am (TESTS_ENVIRONMENT): Set
189 MALLOC_CHECK_ and MALLOC_PERTURB_
191 * tests/run-collate-tests.sh: Run the collation tests explicitly
192 in en_US locale. (#320463)
194 * glib/gmem.h: Really deprecate GMemChunk.
201 * glib/gstrfuncs.c: Add versioned deprecation docs.
203 2005-12-05 Martyn Russell <martyn@imendio.com>
205 * docs/reference/glib/glib-sections.txt:
206 * glib/gasyncqueue.[ch]:
207 - Added support for sorting async queues by with _push_sorted(),
208 _push_sorted_unlocked(), _sort() and _sort_unlocked() (#323047).
211 * tests/asyncqueue-test.c:
212 - Added test case for gasyncqueue.c
214 Mon Dec 5 15:53:20 2005 Tim Janik <timj@imendio.com>
216 * glib/gslice.c: implement chain walking for arbitrary ->next pointer
217 offsets in g_slice_free_chain_with_offset() based on a patch by behdad
218 in bug 323178. moved time consuming logic from g_slice_free() out of
219 the inner loop, so g_slice_free_chain_with_offset() provides a real
220 performance benefit over g_slice_free1() now.
222 * glib/gslice.h: renamed g_slice_free_chain() to
223 g_slice_free_chain_with_offset(). implemented g_slice_free_chain() as
224 a type-safe macro as suggested in bug 323178.
225 simplified the macro implementation of g_slice_free() and implemented
226 it in a type safe manner for all compliers as suggested by Morten
227 Welinder <mortenw@gnome.org>.
232 * glib/glib.symbols: s/g_slice_free_chain/g_slice_free_chain_with_offset/
234 2005-12-05 Matthias Clasen <mclasen@redhat.com>
236 * glib/gasyncqueue.c: Add some docs.
238 * tests/libmoduletestplugin_a.c: Fix compiler warnings.
240 * glib/gatomic.c: In the ia64 implementation, use
241 __sync builtin without _si or _di suffix. (#321229,
242 Stanislav Brabec, patch by Andreas Schwab)
244 2005-12-04 Behdad Esfahbod <behdad@gnome.org>
246 * glib/gslice.h: Remove comma at the end of enum.
248 2005-12-04 Matthias Clasen <mclasen@redhat.com>
250 Handle multiple user names with the same UID better.
251 (#319535, Laszlo Peter)
253 * glib/gutils.c (g_get_any_init_do): When determining user
254 data, first look up $LOGNAME. If the UID doesn't match
255 getuid(), fall back to the current behaviour of looking
256 up the user data based on getuid().
258 2005-12-04 Matthias Clasen <mclasen@redhat.com>
260 * glib/gmacros.h (G_GNUC_WARN_UNUSED_RESULT): Add a macro
261 to make gcc warn if a function result is ignored. (#145466,
262 Arjan van de Ven, Alex Larsson)
264 * glib/gmem.h: Add the new attribute to g_realloc and
267 2005-12-03 Matthias Clasen <mclasen@redhat.com>
270 * glib/gcache.h: Deprecate g_cache_value_foreach. (#322956,
273 * glib/gmappedfile.c: Make mapping of empty files
276 * glib/gfileutils.c: Don't fork a new process just to
277 fix the permissions of the created temp file. (#321318,
278 Alexis S. L. Carvalho)
280 2005-12-02 Matthias Clasen <mclasen@redhat.com>
282 * README.in: Add a note about Solaris threads.
285 * configure.in: Remove support for Solaris threads.
286 (#136971, Sebastian Wilhelmi, patch by Andrew Paprocki)
288 2005-12-02 Matthias Clasen <mclasen@redhat.com>
290 * configure.in: Define G_GUINT64_CONSTANT in analogy to
291 G_GINT64_CONSTANT. (#322568, Andrew Paprocki)
293 2005-12-02 Matthias Clasen <mclasen@redhat.com>
295 * glib/gslice.c: Win32 portability fixes and C99ism removal,
296 pointed out by Kazuki Iwamoto. (#323052)
298 2005-12-02 Matthias Clasen <mclasen@redhat.com>
300 * glib/gatomic.c (g_atomic_int_add): Add a missing volatile for
301 the IA64 implementation.
303 Fri Dec 2 16:18:09 2005 Tim Janik <timj@imendio.com>
305 * glib/gslice.c: some naming and type size fixups.
307 Fri Dec 2 13:08:58 2005 Tim Janik <timj@gtk.org>
309 * glib/gslice.h (g_slice_free): reworked GCC-specific type-safe macro
310 variant into something less verbose, but digestible for gcc-3.4.
312 Fri Dec 2 10:55:07 2005 Tim Janik <timj@imendio.com>
314 * tests/slice-test.c: extended to perform the benchmarking on the old
315 memchunk code if 'O' is selected.
317 * tests/memchunks.c: new file which contains the old GLib mem chunks
318 implementation with prefix old_mem_chunk_.
320 * tests/Makefile.am: added memchunks.c
322 Fri Dec 2 00:16:59 2005 Tim Janik <timj@imendio.com>
324 * glib/gslice.c: improved gettimeofday() timer resolution by maintaining
325 acceptable load distribution of the syscall.
326 (allocator_get_magazine_threshold): reduce minimum/single-thread magazine
327 sizes to approximately page_size/2.5 to avoid excessive startup allocations.
329 2005-12-01 Matthias Clasen <mclasen@redhat.com>
331 * tests/gobject/ifacecheck.c (test_iface_base_init): Don't
332 double-free base interface members.
334 * tests/Makefile.am: Remove duplicate variable.
336 Thu Dec 1 17:32:46 2005 Tim Janik <timj@imendio.com>
338 * glib/gslice.[hc]: new slice allocator implementation.
340 * tests/slice-test.c: added random slice allocation test.
342 * glib/gthread.[hc]: removed newly added private thread mem API.
344 * glib/gthreadinit.h:
347 * glib/gmem.c: divided glib threading initialisation into three phases,
348 initialisation where private keys and messaging are not available (only
349 needed by gmem.c), initialisation without messaging but private keys
350 available (gslice.c, gmessage.c), and full fledged initialisers that
351 server the rest of glib. initialisation functions got renamed to reflect
352 the limitations of their corresponding phases.
354 * glib/gmem.c: removed memchunk code, defer allocations to
357 * glib/gmem.[hc]: removed g_slice_* skeletons.
359 * glib/glib.symbols: added g_slice_* symbols.
361 * configure.in: check for availability of posix_memalign(3), memalign(3)
364 * glib/Makefile.am: added gslice.[hc].
366 2005-12-01 Tor Lillqvist <tml@novell.com>
368 * glib/gstdio.c (g_stat): In the Win32 implementation, strip
369 trailing slash(es) for non-root folders. stat() fails if non-root
370 folders are specified with trailing slashes. It's too much hassle
371 to demand that callers strip such slashes themselves, especially
372 as it is easy to get it wrong and strip the slash of a root
374 (g_rename): On NT-based Windows, use MoveFileEx() with
375 MOVEFILE_REPLACE_EXISTING to better match Unix behaviour.
377 2005-11-28 Matthias Clasen <mclasen@redhat.com>
379 Fix G_STMT_START / G_STMT_END on Solaris. (#321972,
382 * configure.in: Check whether do { } while (0) works.
384 * glib/gmacros.h: Use do { } while (0) for G_STMT_START /
385 G_STMT_END if it works.
387 2005-11-28 Matthias Clasen <mclasen@redhat.com>
389 * glib/gthread.c (g_static_rw_lock_wait, g_static_rw_lock_signal):
390 * glib/gnode.c (g_node_depth_traverse_level):
391 * glib/gmem.c (g_allocator_new):
392 * glib/ghash.c (g_hash_table_unref, g_hash_table_destroy)
393 (g_hash_table_foreach_remove_or_steal): Silence compiler
396 2005-11-27 Matthias Clasen <mclasen@redhat.com>
398 * glib/gunicollate.c (g_utf8_collate_key_for_filename):
399 Don't read beyond len. (#322520, Christian Persch)
401 Wed Nov 23 17:34:01 2005 Tim Janik <timj@imendio.com>
403 * glib/gdataset.c: access datalist flags via atomic pointer access
404 functions, instead of acquiring the dataset lock. this is faster and
405 also matches the atomic pointer readouts in gdatalistprivate.h and
406 direct pointer modifications required by gobject.c.
408 Wed Nov 23 13:35:31 2005 Tim Janik <timj@gtk.org>
410 * glib/gdataset.c: streamlined initialization code somewhat,
411 removed GData node cache.
413 2005-11-22 Matthias Clasen <mclasen@redhat.com>
415 * glib/ghash.c (g_hash_table_ref):
416 (g_hash_table_unref): Mark these as new API.
418 Avoid double locking in g_intern_string (#322133,
421 * glib/gdataset.c (g_quark_from_string_internal): New
422 internal function which factors out common parts of
423 g_quark_from[_static]_string.
424 (g_quark_from_string, g_quark_from_static_string):
425 Use g_quark_from_string_internal.
426 (g_intern_string, g_intern_static_string): Use
427 g_quark_from_string_internal, and only take the
429 (g_quark_new): Don't store the strings shifted by -1
430 in the g_quarks array.
431 (g_quark_to_string): Adapt to the previous change.
433 Tue Nov 22 14:04:26 2005 Tim Janik <timj@imendio.com>
437 g_hash_table_new_full(): create hash tables with a ref count of 1.
438 g_hash_table_ref(): atomically ref_count+=1
439 g_hash_table_unref(): atomically ref_count-=1, destroys hash table
440 when refcount reaches 0.
441 g_hash_table_destroy(): just destroy keys and values, unref by 1.
442 g_hash_table_insert():
443 g_hash_table_replace(): assert ref_count>0.
446 * glib/gatomic.c: added 'volatile' qualifier to all atomic pointer and
449 2005-11-20 Behdad Esfahbod <behdad@gnome.org>
451 * glib/guniprop.c (g_unichar_get_mirror_char): Remove unused
452 variables. (#321984, Andrew Paprocki)
454 2005-11-18 Matthias Clasen <mclasen@redhat.com>
456 * configure.in: Bump version
458 2005-11-17 Matthias Clasen <mclasen@redhat.com>
460 * === Released 2.9.0 ===
462 * glib/glib.symbols: Updates
464 * glib/gutf8.c (g_utf8_offset_to_pointer): Add some performance
465 hints to the docs. (Paolo Borelli)
469 * glib/gthread.c (g_thread_foreach): Mark as new api.
471 * README.in: Updates.
475 * glib/gdate.c (g_date_set_time_t):
476 (g_date_set_time_val): Add functions to set a date from
477 a time_t and from a GTimeVal.
478 (g_date_set_time): Deprecate. (#314067, Roger Leigh)
481 * tests/utf8-pointer.c: Unit tests for g_utf8_pointer_to_offset
482 and g_utf8_offset_to_pointer.
484 * glib/gutf8.c (g_utf8_pointer_to_offset)
485 (g_utf8_offset_to_pointer): Handle negative offsets, and use
486 "stutter stepping" for going backwards. (#320638, Larry
491 * glib/gthread.c: const correctness fixes, found
492 by Arjan van de Ven and gcc.
494 2005-11-16 Behdad Esfahbod <behdad@gnome.org>
496 * docs/reference/glib/tmpl/unicode.sgml: Correct typo on mentioning
497 @G_UNICODE_BREAK_UNKNOWN that should be #G_UNICODE_BREAK_UNKNOWN.
499 2005-11-09 Behdad Esfahbod <behdad@gnome.org>
501 * glib/guniprop.c: Use bit hacks instead when checking a general
502 category value against multiple values.
504 * glib/gutf8.c: Change ISO10646 to Unicode in docs.
506 2005-11-10 Simos Xenitellis <simos@gnome.org>
508 * configure.in: Added tt (Tatar) to ALL_LINGUAS.
510 2005-11-05 Tor Lillqvist <tml@novell.com>
512 * glib/giowin32.c (g_io_win32_set_flags): Don't call g_warning().
513 (#320688) Instead set error and return failure.
515 2005-11-07 Matthias Clasen <mclasen@redhat.com>
517 * glib/gmain.c (g_main_dispatch): Don't call cb_funcs->unref
518 while holding the context lock. (#320886, Andy Wingo)
520 2005-11-07 Billy Biggs <vektor@dumbterm.net>
522 * tests/refcount/Makefile.am: Add a missing reference to libglib
523 in the LDADD for this test.
525 2005-11-07 Matthias Clasen <mclasen@redhat.com>
527 * glib/gmem.c (g_allocator_new): We need to set n_preallocs to a
528 nonzero value, otherwise GTK+ 2.8 breaks when compiled against
529 GLib 2.9. (#320755, Luca Ferretti)
531 2005-11-04 Matthias Clasen <mclasen@redhat.com>
533 * glib/Makefile.am: Apply a patch from Behdad Esfahbod to
534 use a faster lookup table for g_unichar_get_mirror_char().
535 * glib/gmirroringtable.h: The new table.
536 * glib/gunichartables.h: Remove bidi_mirroring_table.
537 * glib/gen-unicode-tables.pl: Don't generate the mirroring
539 * glib/glib-mirroring-tab/*: Sources for the program
540 which generated gmirroringtable.h.
542 * glib/glist.c: Avoid some code duplication.
544 * glib/gscanner.h: Include gdataset.h. (#320322)
546 * glib/gdebug.h: add new GDebugFlag for fatal_criticals
547 * glib/gmessages.c: (_g_debug_init): handle G_DEBUG=fatal_criticals,
548 to help find critical warnings in applications. (#320017,
551 2005-11-02 Tor Lillqvist <tml@novell.com>
553 * glib/glib.symbols: Remove large amount of trailing whitespace
554 from one line. Remove #ifdef G_UNDEFINED from around
555 g_hash_table_new and g_hash_table_new_full symbols, what was the
556 purpose of that? Add the g_slice functions.
558 Tue Nov 1 16:24:20 2005 Tim Janik <timj@imendio.com>
560 * glib/gmem.[hc]: prepared deprecation of GMemChunk and GAllocator.
561 added g_slice_*() API to allocate and cache small bits of memory.
562 an actuall allocator implementation for g_slice_*() is still pending.
564 * glib/gthread.[hc]: changes from a patch by Matthias Clasen.
565 changed GRealThread list to use in-structure *next; fields instead
566 of GSList, in order for thread iteration to not depenend on g_slice_*()
568 _g_thread_mem_private_get():
569 _g_thread_mem_private_set(): added accessors for private memory,
570 needed because the ordinary GPrivate implementation relies on GArray
571 and GSList and therefore indirectly on working g_slice_*() allocations.
574 g_thread_foreach(): new public API function to loop over all existing threads.
588 removed GAllocator and free list usages and accompanying locks.
589 use g_slice_*() API to allocate and cache small bits of memory.
591 * glib/ghook.h: removed GMemChunk field from public API.
594 * glib/glist.h: deprecate allocator API, provide _free1() for consistency.
596 * glib/gnode.h: deprecate allocator API.
598 * glib/gmain.c: reordered GPollRec fields so g_slice_free_chain() can
599 be used for poll rec lists.
601 * glib/grel.c: removed mem chunk usage, and allocated tuples via g_slice_*().
602 g_relation_destroy(): free all tuples from the all_tuples hash table,
603 this effectively maintains the life time track keeping of tuples.
604 g_relation_delete_tuple(): free tuples which are removed from the
605 all_tuples hash table. this fixes a temporary leak that was present
606 in the memchunk code until the destruction of the relation.
608 2005-10-29 Matthias Clasen <mclasen@redhat.com>
610 * tests/convert-test.c: Add some tests for conversions between
611 UTF-8, UCS-4 and UTF-16.
613 * glib/gutf8.c (g_utf8_to_ucs4, g_utf8_to_utf16): Fix handling
614 of len == -1, noticed by Morten Welinder.
616 2005-10-27 Erdal Ronahi <erdal.ronahi@gmail.com>
618 * configure.in: Added ku (Kurdish) to ALL_LINGUAS
620 2005-10-26 Matthias Clasen <mclasen@redhat.com>
622 * glib/gutf8.c (g_ucs4_to_utf8): Don't set items_read twice
623 in the error case, and add some documentation. (#319806, Morten
626 2005-10-19 Manish Singh <yosh@gimp.org>
628 * configure.in: Use AC_CHECK_FUNCS for _NSGetEnviron, to get the
629 config.h symbol automatically. Fixes bug #313731.
631 2005-10-19 Tor Lillqvist <tml@novell.com>
633 * glib/gwin32.c (g_win32_get_package_installation_directory):
634 Return a g_strdup()ed copy of the value stored in the hash table,
635 so that it can be g_free()d without leaving a dangling pointer in
636 the hash table. (#319232)
638 2005-10-06 Matthias Clasen <mclasen@redhat.com>
640 * glib/gunicollate.c (g_utf8_collate_key_for_filename): Handle
641 all-zero sequences correctly. (#317930, Sebastien Bacher)
643 2005-10-05 Matthias Clasen <mclasen@redhat.com>
645 * glib/gmarkup.c (g_markup_escape_text): Doc addition
647 * Makefile.am: Add Collation tests.
649 * tests/collate/*: Inputs and expected outputs for collation tests.
651 * tests/run-collate-tests.sh: Script to run collation tests.
653 * tests/unicode-collate.c (main): Rework slightly to make
654 it usable in unit tests. Also test g_utf8_collate_key_for_filename().
656 2005-10-01 Behdad Esfahbod <behdad@gnome.org>
658 * docs/reference/glib/tmpl/unicode.sgml:
659 * glib/gen-unicode-tables.pl:
661 * glib/gunichartables.h:
663 * tests/casefold.txt:
664 * tests/casemap.txt: Updated to Unicode 4.1. There are five new
665 GUnicodeBreakType types. That may break some applications, like
668 2005-09-26 Matthias Clasen <mclasen@redhat.com>
670 * glib/gstrfuncs.c (g_ascii_strtoull): Add details to the
671 docs. (#314393, Matthew F. Barnes)
674 * glib/gprintf.h: Remove g_snprintf() and g_vsnprintf(), since
675 they are already declared in glib.h. This doesn't break documented
676 use of gprintf.h, but should probably be pointed out in the
677 release notes for 2.10. (#314232, Behdad Esfahbod)
679 Tue Sep 20 13:16:04 2005 Tim Janik <timj@imendio.com>
681 * glib/gpattern.c (g_pattern_ph_match): applied significant recursion
682 complexity optimization, based on a patch from Matthias Clasen.
684 * tests/patterntest.c: more tests, mostly from matthias.
686 2005-09-20 Matthias Clasen <mclasen@redhat.com>
688 * glib/gqueue.c (g_queue_insert_sorted): Correct the docs.
689 (#316703, Mark Drago)
691 Mon Sep 19 17:23:23 2005 Tim Janik <timj@imendio.com>
693 * glib/gpattern.c: applied a patch from matthias which checks on the
694 upper bound of GPatternSpec length to optimize matches.
697 * tests/patterntest.c: added more match cases.
699 2005-09-16 Tor Lillqvist <tml@novell.com>
701 * glib/gstrfuncs.c (g_ascii_strcasecmp, g_ascii_strncasecmp): Add
702 warning to doc comment that these functions should not be used on
703 encodings like CP932.
705 2005-09-14 Matthias Clasen <mclasen@redhat.com>
707 * tests/keyfile-test.c: Add a test for grup names of length 1.
709 * glib/gkeyfile.c (g_key_file_line_is_group): Accept group names
710 of length 1. (#316309)
712 2005-09-12 Matthias Clasen <mclasen@redhat.com>
714 * glib/gmarkup.c (g_markup_escape_text): Clarify docs.
716 2005-09-11 Sebastian Wilhelmi <seppi@seppi.de>
718 * tests/refcount/Makefile.am (INCLUDES): Link the the refcount
719 tests to the system thread library $(G_THREAD_LIBS). Fixes #313744
722 2005-09-11 Kjartan Maraas <kmaraas@gnome.org>
724 * glib/gmain.c: (g_child_watch_prepare), (g_child_watch_check),
725 (child_watch_helper_thread): Remove some dead code. Closes
728 2005-09-07 Tor Lillqvist <tml@novell.com>
730 * glib/Makefile.am: Create also a console version of the
731 gspawn-win32-helper program, gspawn-win32-helper-console.exe.
732 It's otherwise identical to gspawn-win32-helper.exe, except marked
733 as a console application (linked without the -mwindows option).
735 * glib/gspawn-win32.c (do_spawn_directly, do_spawn_with_pipes):
736 Drop the dont_wait parameter. Its truth value correlated 100% with
737 the NULLness of the exit_status parameter anyway, so it's enough
738 to check whether exit_status is NULL. Invert the sense of the
739 dont_return_handle parameter and rename it to do_return_handle, to
740 make the code easier to read by avoiding double negations.
742 (g_spawn_sync_utf8, g_spawn_async_with_pipes_utf8): Modify calls
743 to do_spawn_with_pipes() accordingly.
745 (do_spawn_with_pipes): If we have a console, use the console
746 version of the helper program, otherwise use the GUI one. This
747 avoids extra console windows opening up in some situations. (In
748 case a console application uses the GUI gspawn-win32-helper.exe to
749 spawn another console application we would get a separate console
750 for the spawned console application).
752 * glib-zip.in: Distribute also gspawn-win32-helper-console.exe.
754 2005-09-05 Matthias Clasen <mclasen@redhat.com>
756 * glib/gmappedfile.c (g_mapped_file_new): Report an error
757 if the file is too large. (#315275, Kjartan Maraas)
759 * glib/gkeyfile.c (g_key_file_load_from_fd): The return value
760 of read() is signed. (#315273, Kjartan Maraas)
762 2005-08-31 Tor Lillqvist <tml@novell.com>
764 * glib/gutils.h: Wrapping atexit() is a bad idea on Windows, where
765 the EXE and each DLL have their own atexit function chains.
767 #define g_atexit as atexit instead. This means it has a
768 better chance of doing what the caller wants. For instance,
769 gtkhtml calls g_atexit() registering a function in gtkhtml
770 itself. This caused a crash when g_atexit() was implemented as a
771 function in the GLib DLL. The gtkhtml DLL was already unloaded by
772 the time the GLib DLL got unloaded.
774 * glib/gutils.c: #undef the #define mentioned above, to also get a
775 real g_atexit() into the DLL for backward compatibility. Document
776 the Windows behaviour of g_atexit(), and document the varying ways
777 atexit() can behave in the context of dynamically loaded modules
780 2005-08-31 Matthias Clasen <mclasen@redhat.com>
784 * glib/gdataset.c: Add string interning functions.
786 2005-08-28 Matthias Clasen <mclasen@redhat.com>
788 * glib/giochannel.c: Unify some near-duplicate strings. (#314654,
791 2005-08-26 Matthias Clasen <mclasen@redhat.com>
793 * configure.in: Bump version to 2.9.0
795 2005-08-25 Tor Lillqvist <tml@novell.com>
797 Make also the g_spawn*() functions take parameters in the GLib
798 file name encoding, i.e. UTF-8, on Windows. Has no impact on Unix
799 API or ABI. Like the other GLib API that was earlier changed to
800 use UTF-8 on Windows, the names of the functions that take UTF-8
801 have _utf8 suffixes added by using preprocessor macros in the
802 header file. The old names are kept for functions with the old
803 behaviour, taking parameters in the system codepage, for DLL ABI
806 * glib/gspawn.h: On Win32 add the suffix _utf8 to the names of the
807 g_spawn*() functions.
809 * glib/gspawn-win32.c: Use wide-char API on NT-based
810 Windows. Convert parameters from UTF-8 to wide chars (NT) or
811 system codepage (Win9x) and call the C library _wspawn*() or
812 spawn*() functions respectvely. Add DLL ABI stability versions
813 that take parameters in the system codepage.
815 * glib/gspawn-win32-helper.c: On NT-based Windows use the
816 wide-char versions of argv and envp, and use wide-char API to
817 change directory and spawn the program to run. Remove the verbose
818 debugging output, it was too complex to modify for the wide-char
819 features. (Just add temporary debugging printouts if needed, no
820 need to have them permanently in the source.)
822 * glib/gspawn.c: Corresponding documentation updates.
824 * glib/glib.symbols: Corresponding changes: Mark the ABI stability
825 symbols as PRIVATE, add the new _utf8-suffixed ones.
827 2005-08-24 Stepan Kasal <kasal@ucw.cz>
829 * glib/gtypes.h (G_MININT64): Cast the constant to gint64; it is
830 guint64 otherwise and that can produce warnings about comparison
831 between signed and unsigned.
833 2005-08-23 Matthias Clasen <mclasen@redhat.com>
835 * glib/gutils.c: Fix the crt_externs.h include.
837 2005-08-23 Stepan Kasal <kasal@ucw.cz>
839 * NEWS: Fix spelling of my first name.
841 2005-08-23 Matthias Clasen <mclasen@redhat.com>
845 * === Released 2.8.1 ===
849 2005-08-20 Hans Breuer <hans@breuer.org>
851 * glib/makefile.msc.in : link with ws2_32.lib
853 2005-08-18 Tor Lillqvist <tml@novell.com>
855 * configure.in: Check for <sys/wait.h>
857 * glib/gbacktrace.c: Include <sys/wait.h> on if HAVE_SYS_WAIT_H.
859 2005-08-18 Ross Burton <ross@burtonini.com>
862 Optimise single-character insertions.
867 * tests/string-test.c:
868 Add tests for new optimisation, and fix a leak.
870 2005-08-17 Matthias Clasen <mclasen@redhat.com>
872 * configure.in: Check for crt_externs.h and _NSGetEnviron.
874 * glib/gutils.c: On Darwin, include crt-externs.h and
875 define environ using _NSGetEnviron(). (#313731)
877 2005-08-16 Stepan Kasal <kasal@ucw.cz>
879 * glib/gutils.c (g_get_any_init): Move the body of the big if...
880 (g_get_any_init_do): ... to this new function.
881 (g_get_any_init): Declare as inline.
882 (g_get_any_init_locked): New inline function, does the locking.
883 Make use of these two throughout the code.
885 2005-08-15 Matthias Clasen <mclasen@redhat.com>
887 * glib/gbacktrace.c (g_on_error_stack_trace): Wait for
888 the child process and then simply return. This makes
889 The "S" option work as documented in g_on_error_query().
890 (#313125, Matthew F. Barnes)
892 * glib/gunicode.h: Update the link to Unicode category
893 values. (#313369, Behnam Esfahbod)
895 * glib/gqueue.c (g_queue_find_custom): Clarify docs
896 a little. (#311727, Tristan van Berkom)
898 * glib/abicheck.sh, gobject/abicheck.sh: Make the
899 check work on ia64 too, where some symbols ended up
900 in yet another section.