1 2006-01-03 Matthias Clasen <mclasen@redhat.com>
3 * glib/glib.symbols: Add new functions.
5 2005-12-20 Michael Meeks <michael.meeks@novell.com>
7 * glib/gmain.c (g_main_context_is_owner): new method
8 to determine if the current thread is the owner of the
11 2006-01-02 Matthias Clasen <mclasen@redhat.com>
15 * glib/gstdio.c (g_rmdir): Don't provide g_rmdir() as a macro
16 expanding to rmdir, since rmdir is not declared in a portable
17 system header we can include in gstdio.h. (#325249, Jani Monoses)
19 2006-01-01 Tor Lillqvist <tml@novell.com>
21 * glib/gspawn-win32.c (g_spawn_sync_utf8): Set the GIOChannels for
22 stdout and stderr to unbuffered. Otherwise the giochannel layer
23 will try to read from them regardless whether the
24 g_io_channel_win32_poll() call here has indicated
25 readability or not. (#325310)
27 2005-12-29 Matthias Clasen <mclasen@redhat.com>
29 * glib/gutils.c (glib_check_version): Fix a copy-and-paste error
30 in the docs. (#325273, Declan Naughton)
32 2005-12-27 Manish Singh <yosh@gimp.org>
34 * tests/run-collate-tests.sh: use LC_ALL instead of LC_COLLATE, to
35 make sure we really override things.
37 2005-12-27 Matthias Clasen <mclasen@redhat.com>
39 Fix #316221, Michal Benes, Stanislav Brabec;
41 * configure.in: Fix a strict aliasing problem in
42 g_static_mutex_get_mutex().
43 * glib/gthread.h: ...and in
44 g_static_mutex_get_mutex_impl_shortcut().
46 * glib/gdatasetprivate.h: Add a cast to silence compiler
47 warnings. (#321978, Andrew Paprocki)
49 Partial fix for bug #32937, Bogdan Nicula.
51 * configure.in: Check for malloc.h
52 * glib/gslice.c: Don't include malloc.h unconditionally.
54 2005-12-27 Manish Singh <yosh@gimp.org>
56 * tests/run-collate-tests.sh: export LC_COLLATE so it takes.
57 (#324950, Dan Yefimov)
59 2005-12-26 Matthias Clasen <mclasen@redhat.com>
61 * glib/gslice.c: On Win32, include process.h (#325015, Kazuki
64 2005-12-25 Matthias Clasen <mclasen@redhat.com>
66 * AUTHORS: Update my email
68 * tests/slice-test.c: Fix C99isms. (#324950, Dan Yefimov)
70 2005-12-21 Matthias Clasen <mclasen@redhat.com>
72 * README.in: Add some notes on when 'make check' may fail.
74 2005-12-20 Matthias Clasen <mclasen@redhat.com>
78 * glib/gthreadpool.c (g_thread_pool_set_sort_function): New function
79 to sort tasks pushed into a threadpool. (#324479, Martyn Russell)
81 * tests/threadpool-test.c: Test this.
83 Tue Dec 20 18:14:14 2005 Tim Janik <timj@imendio.com>
85 * glib/gslice.[hc]: added mem_error() and mem_assert() to test and
86 handle errors without depending on gmessage.c which might not be
87 setup when the error occours.
88 removed G_SLICE_CONFIG_ALWAYS_FREE config option, fixed the code so
89 always freeing can be achieved by adjusting the working set time to
90 0 with G_SLICE_CONFIG_WORKING_SET_MSECS.
91 added G_SLICE_CONFIG_COLOR_INCREMENT to test different color increments
92 (mainly 0 and 1). reduced the minimum block size to 128 bytes, to
93 minimize wastage if small amounts of differently sized structrues are
94 allocated, this does come at a performance cost of roughly 5% though.
95 fixed up block alignment calculation, so it works for varying
96 block sizes. only use strerror() not g_strerror() since the latter
97 depends on working GQuark and GSlice.
98 mem_error(): implemented in terms of fprintf and vfprintf.
100 * tests/slice-color.c: new program to test cache colorization effects.
102 * tests/slice-test.c: trade G_SLICE_CONFIG_ALWAYS_FREE for 0 duration
103 G_SLICE_CONFIG_WORKING_SET_MSECS.
105 2005-12-17 Matthias Clasen <mclasen@redhat.com>
107 * glib/goption.c (parse_short_option): Set an error in all
108 failure cases. (#324332, Tim-Philipp Müller)
110 2005-12-17 Sebastian Wilhelmi <seppi@seppi.de>
112 * glib/gatomic.c: Fix memory barrier position in g_atomic_int_get
113 and g_atomic_pointer_get. Add g_atomic_int_set and
114 g_atomic_pointer_set implementations for the !DEFINE_WITH_MUTEXES &&
115 G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case, as well as defining them
116 as functions (additionally to the macros in the header) for the
117 !G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case.
119 2005-12-16 Matthias Clasen <mclasen@redhat.com>
121 * glib/gmem.c (g_allocator_new): Don't return a pointer to
122 a const struct, since apps expect to be able to modify it.
123 (#324179, J. Ali Harlow)
125 Tue Dec 13 10:13:32 2005 Tim Janik <timj@imendio.com>
127 * glib/gatomic.h: added g_atomic_pointer_set() and g_atomic_int_set()
128 for systems where the initialization of atomic variables requires a
129 write memory barrier.
131 2005-12-09 Matthias Clasen <mclasen@redhat.com>
133 * configure.in: Bump version
135 * === Released 2.9.1 ===
137 2005-12-09 Alexander Larsson <alexl@redhat.com>
141 Use G_GNUC_WARN_UNUSED_RESULT on list functions that return
144 2005-12-08 Matthias Clasen <mclasen@redhat.com>
148 2005-12-07 Martyn Russell <martyn@imendio.com>
150 * glib/gasyncqueue.c:
151 - Call g_queue_insert_sorted() instead of duplicating the code.
152 - Call g_queue_sort() instead of duplicating the code.
153 - Invert sort function results to make sure the same sort function
154 gives the same results across glist, gslist, gqueue and
157 * tests/asyncqueue-test.c:
158 - Updated the sort function to reflect the example in the
159 documentation for gasyncqueue.c.
161 2005-12-07 Martyn Russell <martyn@imendio.com>
163 * docs/reference/glib/glib-sections.txt:
164 * docs/reference/glib/tmpl/linked_lists_double.sgml:
165 * docs/reference/glib/tmpl/linked_lists_single.sgml:
168 - Added g_list_insert_sorted_with_data () and
169 g_slist_insert_sorted_with_data ().
170 - Removed the extra check in g_list_sort() and g_slist_sort() for
171 GCompareDataFunc vs. GCompareFunc.
173 2005-12-07 Tor Lillqvist <tml@novell.com>
175 * glib-gettextize.in: Look up prefix at run-time on Win32,
176 assuming the standard directory structure with glib-gettextize in
179 2005-12-06 Matthias Clasen <mclasen@redhat.com>
181 * glib/gmem.h: Exempt GTK+ from the mem chunk deprecation, since
182 we need GTK+ 2.8 to compile against GLib 2.10.
184 2005-12-05 Manish Singh <yosh@gimp.org>
186 * tests/run-collate-test.sh: set LC_COLLATE, not LANG, to be sure
187 to override any user settings.
189 2005-12-05 Matthias Clasen <mclasen@redhat.com>
191 * configure.in: Revert an accidential commit
193 2005-12-05 Matthias Clasen <mclasen@redhat.com>
195 * glib/gutf8.c: Documentation updates. (#323291, Morten Welinder)
197 * tests/Makefile.am (TESTS_ENVIRONMENT):
198 * tests/gobject/Makefile.am (TESTS_ENVIRONMENT): Set
199 MALLOC_CHECK_ and MALLOC_PERTURB_
201 * tests/run-collate-tests.sh: Run the collation tests explicitly
202 in en_US locale. (#320463)
204 * glib/gmem.h: Really deprecate GMemChunk.
211 * glib/gstrfuncs.c: Add versioned deprecation docs.
213 2005-12-05 Martyn Russell <martyn@imendio.com>
215 * docs/reference/glib/glib-sections.txt:
216 * glib/gasyncqueue.[ch]:
217 - Added support for sorting async queues by with _push_sorted(),
218 _push_sorted_unlocked(), _sort() and _sort_unlocked() (#323047).
221 * tests/asyncqueue-test.c:
222 - Added test case for gasyncqueue.c
224 Mon Dec 5 15:53:20 2005 Tim Janik <timj@imendio.com>
226 * glib/gslice.c: implement chain walking for arbitrary ->next pointer
227 offsets in g_slice_free_chain_with_offset() based on a patch by behdad
228 in bug 323178. moved time consuming logic from g_slice_free() out of
229 the inner loop, so g_slice_free_chain_with_offset() provides a real
230 performance benefit over g_slice_free1() now.
232 * glib/gslice.h: renamed g_slice_free_chain() to
233 g_slice_free_chain_with_offset(). implemented g_slice_free_chain() as
234 a type-safe macro as suggested in bug 323178.
235 simplified the macro implementation of g_slice_free() and implemented
236 it in a type safe manner for all compliers as suggested by Morten
237 Welinder <mortenw@gnome.org>.
242 * glib/glib.symbols: s/g_slice_free_chain/g_slice_free_chain_with_offset/
244 2005-12-05 Matthias Clasen <mclasen@redhat.com>
246 * glib/gasyncqueue.c: Add some docs.
248 * tests/libmoduletestplugin_a.c: Fix compiler warnings.
250 * glib/gatomic.c: In the ia64 implementation, use
251 __sync builtin without _si or _di suffix. (#321229,
252 Stanislav Brabec, patch by Andreas Schwab)
254 2005-12-04 Behdad Esfahbod <behdad@gnome.org>
256 * glib/gslice.h: Remove comma at the end of enum.
258 2005-12-04 Matthias Clasen <mclasen@redhat.com>
260 Handle multiple user names with the same UID better.
261 (#319535, Laszlo Peter)
263 * glib/gutils.c (g_get_any_init_do): When determining user
264 data, first look up $LOGNAME. If the UID doesn't match
265 getuid(), fall back to the current behaviour of looking
266 up the user data based on getuid().
268 2005-12-04 Matthias Clasen <mclasen@redhat.com>
270 * glib/gmacros.h (G_GNUC_WARN_UNUSED_RESULT): Add a macro
271 to make gcc warn if a function result is ignored. (#145466,
272 Arjan van de Ven, Alex Larsson)
274 * glib/gmem.h: Add the new attribute to g_realloc and
277 2005-12-03 Matthias Clasen <mclasen@redhat.com>
280 * glib/gcache.h: Deprecate g_cache_value_foreach. (#322956,
283 * glib/gmappedfile.c: Make mapping of empty files
286 * glib/gfileutils.c: Don't fork a new process just to
287 fix the permissions of the created temp file. (#321318,
288 Alexis S. L. Carvalho)
290 2005-12-02 Matthias Clasen <mclasen@redhat.com>
292 * README.in: Add a note about Solaris threads.
295 * configure.in: Remove support for Solaris threads.
296 (#136971, Sebastian Wilhelmi, patch by Andrew Paprocki)
298 2005-12-02 Matthias Clasen <mclasen@redhat.com>
300 * configure.in: Define G_GUINT64_CONSTANT in analogy to
301 G_GINT64_CONSTANT. (#322568, Andrew Paprocki)
303 2005-12-02 Matthias Clasen <mclasen@redhat.com>
305 * glib/gslice.c: Win32 portability fixes and C99ism removal,
306 pointed out by Kazuki Iwamoto. (#323052)
308 2005-12-02 Matthias Clasen <mclasen@redhat.com>
310 * glib/gatomic.c (g_atomic_int_add): Add a missing volatile for
311 the IA64 implementation.
313 Fri Dec 2 16:18:09 2005 Tim Janik <timj@imendio.com>
315 * glib/gslice.c: some naming and type size fixups.
317 Fri Dec 2 13:08:58 2005 Tim Janik <timj@gtk.org>
319 * glib/gslice.h (g_slice_free): reworked GCC-specific type-safe macro
320 variant into something less verbose, but digestible for gcc-3.4.
322 Fri Dec 2 10:55:07 2005 Tim Janik <timj@imendio.com>
324 * tests/slice-test.c: extended to perform the benchmarking on the old
325 memchunk code if 'O' is selected.
327 * tests/memchunks.c: new file which contains the old GLib mem chunks
328 implementation with prefix old_mem_chunk_.
330 * tests/Makefile.am: added memchunks.c
332 Fri Dec 2 00:16:59 2005 Tim Janik <timj@imendio.com>
334 * glib/gslice.c: improved gettimeofday() timer resolution by maintaining
335 acceptable load distribution of the syscall.
336 (allocator_get_magazine_threshold): reduce minimum/single-thread magazine
337 sizes to approximately page_size/2.5 to avoid excessive startup allocations.
339 2005-12-01 Matthias Clasen <mclasen@redhat.com>
341 * tests/gobject/ifacecheck.c (test_iface_base_init): Don't
342 double-free base interface members.
344 * tests/Makefile.am: Remove duplicate variable.
346 Thu Dec 1 17:32:46 2005 Tim Janik <timj@imendio.com>
348 * glib/gslice.[hc]: new slice allocator implementation.
350 * tests/slice-test.c: added random slice allocation test.
352 * glib/gthread.[hc]: removed newly added private thread mem API.
354 * glib/gthreadinit.h:
357 * glib/gmem.c: divided glib threading initialisation into three phases,
358 initialisation where private keys and messaging are not available (only
359 needed by gmem.c), initialisation without messaging but private keys
360 available (gslice.c, gmessage.c), and full fledged initialisers that
361 server the rest of glib. initialisation functions got renamed to reflect
362 the limitations of their corresponding phases.
364 * glib/gmem.c: removed memchunk code, defer allocations to
367 * glib/gmem.[hc]: removed g_slice_* skeletons.
369 * glib/glib.symbols: added g_slice_* symbols.
371 * configure.in: check for availability of posix_memalign(3), memalign(3)
374 * glib/Makefile.am: added gslice.[hc].
376 2005-12-01 Tor Lillqvist <tml@novell.com>
378 * glib/gstdio.c (g_stat): In the Win32 implementation, strip
379 trailing slash(es) for non-root folders. stat() fails if non-root
380 folders are specified with trailing slashes. It's too much hassle
381 to demand that callers strip such slashes themselves, especially
382 as it is easy to get it wrong and strip the slash of a root
384 (g_rename): On NT-based Windows, use MoveFileEx() with
385 MOVEFILE_REPLACE_EXISTING to better match Unix behaviour.
387 2005-11-28 Matthias Clasen <mclasen@redhat.com>
389 Fix G_STMT_START / G_STMT_END on Solaris. (#321972,
392 * configure.in: Check whether do { } while (0) works.
394 * glib/gmacros.h: Use do { } while (0) for G_STMT_START /
395 G_STMT_END if it works.
397 2005-11-28 Matthias Clasen <mclasen@redhat.com>
399 * glib/gthread.c (g_static_rw_lock_wait, g_static_rw_lock_signal):
400 * glib/gnode.c (g_node_depth_traverse_level):
401 * glib/gmem.c (g_allocator_new):
402 * glib/ghash.c (g_hash_table_unref, g_hash_table_destroy)
403 (g_hash_table_foreach_remove_or_steal): Silence compiler
406 2005-11-27 Matthias Clasen <mclasen@redhat.com>
408 * glib/gunicollate.c (g_utf8_collate_key_for_filename):
409 Don't read beyond len. (#322520, Christian Persch)
411 Wed Nov 23 17:34:01 2005 Tim Janik <timj@imendio.com>
413 * glib/gdataset.c: access datalist flags via atomic pointer access
414 functions, instead of acquiring the dataset lock. this is faster and
415 also matches the atomic pointer readouts in gdatalistprivate.h and
416 direct pointer modifications required by gobject.c.
418 Wed Nov 23 13:35:31 2005 Tim Janik <timj@gtk.org>
420 * glib/gdataset.c: streamlined initialization code somewhat,
421 removed GData node cache.
423 2005-11-22 Matthias Clasen <mclasen@redhat.com>
425 * glib/ghash.c (g_hash_table_ref):
426 (g_hash_table_unref): Mark these as new API.
428 Avoid double locking in g_intern_string (#322133,
431 * glib/gdataset.c (g_quark_from_string_internal): New
432 internal function which factors out common parts of
433 g_quark_from[_static]_string.
434 (g_quark_from_string, g_quark_from_static_string):
435 Use g_quark_from_string_internal.
436 (g_intern_string, g_intern_static_string): Use
437 g_quark_from_string_internal, and only take the
439 (g_quark_new): Don't store the strings shifted by -1
440 in the g_quarks array.
441 (g_quark_to_string): Adapt to the previous change.
443 Tue Nov 22 14:04:26 2005 Tim Janik <timj@imendio.com>
447 g_hash_table_new_full(): create hash tables with a ref count of 1.
448 g_hash_table_ref(): atomically ref_count+=1
449 g_hash_table_unref(): atomically ref_count-=1, destroys hash table
450 when refcount reaches 0.
451 g_hash_table_destroy(): just destroy keys and values, unref by 1.
452 g_hash_table_insert():
453 g_hash_table_replace(): assert ref_count>0.
456 * glib/gatomic.c: added 'volatile' qualifier to all atomic pointer and
459 2005-11-20 Behdad Esfahbod <behdad@gnome.org>
461 * glib/guniprop.c (g_unichar_get_mirror_char): Remove unused
462 variables. (#321984, Andrew Paprocki)
464 2005-11-18 Matthias Clasen <mclasen@redhat.com>
466 * configure.in: Bump version
468 2005-11-17 Matthias Clasen <mclasen@redhat.com>
470 * === Released 2.9.0 ===
472 * glib/glib.symbols: Updates
474 * glib/gutf8.c (g_utf8_offset_to_pointer): Add some performance
475 hints to the docs. (Paolo Borelli)
479 * glib/gthread.c (g_thread_foreach): Mark as new api.
481 * README.in: Updates.
485 * glib/gdate.c (g_date_set_time_t):
486 (g_date_set_time_val): Add functions to set a date from
487 a time_t and from a GTimeVal.
488 (g_date_set_time): Deprecate. (#314067, Roger Leigh)
491 * tests/utf8-pointer.c: Unit tests for g_utf8_pointer_to_offset
492 and g_utf8_offset_to_pointer.
494 * glib/gutf8.c (g_utf8_pointer_to_offset)
495 (g_utf8_offset_to_pointer): Handle negative offsets, and use
496 "stutter stepping" for going backwards. (#320638, Larry
501 * glib/gthread.c: const correctness fixes, found
502 by Arjan van de Ven and gcc.
504 2005-11-16 Behdad Esfahbod <behdad@gnome.org>
506 * docs/reference/glib/tmpl/unicode.sgml: Correct typo on mentioning
507 @G_UNICODE_BREAK_UNKNOWN that should be #G_UNICODE_BREAK_UNKNOWN.
509 2005-11-09 Behdad Esfahbod <behdad@gnome.org>
511 * glib/guniprop.c: Use bit hacks instead when checking a general
512 category value against multiple values.
514 * glib/gutf8.c: Change ISO10646 to Unicode in docs.
516 2005-11-10 Simos Xenitellis <simos@gnome.org>
518 * configure.in: Added tt (Tatar) to ALL_LINGUAS.
520 2005-11-05 Tor Lillqvist <tml@novell.com>
522 * glib/giowin32.c (g_io_win32_set_flags): Don't call g_warning().
523 (#320688) Instead set error and return failure.
525 2005-11-07 Matthias Clasen <mclasen@redhat.com>
527 * glib/gmain.c (g_main_dispatch): Don't call cb_funcs->unref
528 while holding the context lock. (#320886, Andy Wingo)
530 2005-11-07 Billy Biggs <vektor@dumbterm.net>
532 * tests/refcount/Makefile.am: Add a missing reference to libglib
533 in the LDADD for this test.
535 2005-11-07 Matthias Clasen <mclasen@redhat.com>
537 * glib/gmem.c (g_allocator_new): We need to set n_preallocs to a
538 nonzero value, otherwise GTK+ 2.8 breaks when compiled against
539 GLib 2.9. (#320755, Luca Ferretti)
541 2005-11-04 Matthias Clasen <mclasen@redhat.com>
543 * glib/Makefile.am: Apply a patch from Behdad Esfahbod to
544 use a faster lookup table for g_unichar_get_mirror_char().
545 * glib/gmirroringtable.h: The new table.
546 * glib/gunichartables.h: Remove bidi_mirroring_table.
547 * glib/gen-unicode-tables.pl: Don't generate the mirroring
549 * glib/glib-mirroring-tab/*: Sources for the program
550 which generated gmirroringtable.h.
552 * glib/glist.c: Avoid some code duplication.
554 * glib/gscanner.h: Include gdataset.h. (#320322)
556 * glib/gdebug.h: add new GDebugFlag for fatal_criticals
557 * glib/gmessages.c: (_g_debug_init): handle G_DEBUG=fatal_criticals,
558 to help find critical warnings in applications. (#320017,
561 2005-11-02 Tor Lillqvist <tml@novell.com>
563 * glib/glib.symbols: Remove large amount of trailing whitespace
564 from one line. Remove #ifdef G_UNDEFINED from around
565 g_hash_table_new and g_hash_table_new_full symbols, what was the
566 purpose of that? Add the g_slice functions.
568 Tue Nov 1 16:24:20 2005 Tim Janik <timj@imendio.com>
570 * glib/gmem.[hc]: prepared deprecation of GMemChunk and GAllocator.
571 added g_slice_*() API to allocate and cache small bits of memory.
572 an actuall allocator implementation for g_slice_*() is still pending.
574 * glib/gthread.[hc]: changes from a patch by Matthias Clasen.
575 changed GRealThread list to use in-structure *next; fields instead
576 of GSList, in order for thread iteration to not depenend on g_slice_*()
578 _g_thread_mem_private_get():
579 _g_thread_mem_private_set(): added accessors for private memory,
580 needed because the ordinary GPrivate implementation relies on GArray
581 and GSList and therefore indirectly on working g_slice_*() allocations.
584 g_thread_foreach(): new public API function to loop over all existing threads.
598 removed GAllocator and free list usages and accompanying locks.
599 use g_slice_*() API to allocate and cache small bits of memory.
601 * glib/ghook.h: removed GMemChunk field from public API.
604 * glib/glist.h: deprecate allocator API, provide _free1() for consistency.
606 * glib/gnode.h: deprecate allocator API.
608 * glib/gmain.c: reordered GPollRec fields so g_slice_free_chain() can
609 be used for poll rec lists.
611 * glib/grel.c: removed mem chunk usage, and allocated tuples via g_slice_*().
612 g_relation_destroy(): free all tuples from the all_tuples hash table,
613 this effectively maintains the life time track keeping of tuples.
614 g_relation_delete_tuple(): free tuples which are removed from the
615 all_tuples hash table. this fixes a temporary leak that was present
616 in the memchunk code until the destruction of the relation.
618 2005-10-29 Matthias Clasen <mclasen@redhat.com>
620 * tests/convert-test.c: Add some tests for conversions between
621 UTF-8, UCS-4 and UTF-16.
623 * glib/gutf8.c (g_utf8_to_ucs4, g_utf8_to_utf16): Fix handling
624 of len == -1, noticed by Morten Welinder.
626 2005-10-27 Erdal Ronahi <erdal.ronahi@gmail.com>
628 * configure.in: Added ku (Kurdish) to ALL_LINGUAS
630 2005-10-26 Matthias Clasen <mclasen@redhat.com>
632 * glib/gutf8.c (g_ucs4_to_utf8): Don't set items_read twice
633 in the error case, and add some documentation. (#319806, Morten
636 2005-10-19 Manish Singh <yosh@gimp.org>
638 * configure.in: Use AC_CHECK_FUNCS for _NSGetEnviron, to get the
639 config.h symbol automatically. Fixes bug #313731.
641 2005-10-19 Tor Lillqvist <tml@novell.com>
643 * glib/gwin32.c (g_win32_get_package_installation_directory):
644 Return a g_strdup()ed copy of the value stored in the hash table,
645 so that it can be g_free()d without leaving a dangling pointer in
646 the hash table. (#319232)
648 2005-10-06 Matthias Clasen <mclasen@redhat.com>
650 * glib/gunicollate.c (g_utf8_collate_key_for_filename): Handle
651 all-zero sequences correctly. (#317930, Sebastien Bacher)
653 2005-10-05 Matthias Clasen <mclasen@redhat.com>
655 * glib/gmarkup.c (g_markup_escape_text): Doc addition
657 * Makefile.am: Add Collation tests.
659 * tests/collate/*: Inputs and expected outputs for collation tests.
661 * tests/run-collate-tests.sh: Script to run collation tests.
663 * tests/unicode-collate.c (main): Rework slightly to make
664 it usable in unit tests. Also test g_utf8_collate_key_for_filename().
666 2005-10-01 Behdad Esfahbod <behdad@gnome.org>
668 * docs/reference/glib/tmpl/unicode.sgml:
669 * glib/gen-unicode-tables.pl:
671 * glib/gunichartables.h:
673 * tests/casefold.txt:
674 * tests/casemap.txt: Updated to Unicode 4.1. There are five new
675 GUnicodeBreakType types. That may break some applications, like
678 2005-09-26 Matthias Clasen <mclasen@redhat.com>
680 * glib/gstrfuncs.c (g_ascii_strtoull): Add details to the
681 docs. (#314393, Matthew F. Barnes)
684 * glib/gprintf.h: Remove g_snprintf() and g_vsnprintf(), since
685 they are already declared in glib.h. This doesn't break documented
686 use of gprintf.h, but should probably be pointed out in the
687 release notes for 2.10. (#314232, Behdad Esfahbod)
689 Tue Sep 20 13:16:04 2005 Tim Janik <timj@imendio.com>
691 * glib/gpattern.c (g_pattern_ph_match): applied significant recursion
692 complexity optimization, based on a patch from Matthias Clasen.
694 * tests/patterntest.c: more tests, mostly from matthias.
696 2005-09-20 Matthias Clasen <mclasen@redhat.com>
698 * glib/gqueue.c (g_queue_insert_sorted): Correct the docs.
699 (#316703, Mark Drago)
701 Mon Sep 19 17:23:23 2005 Tim Janik <timj@imendio.com>
703 * glib/gpattern.c: applied a patch from matthias which checks on the
704 upper bound of GPatternSpec length to optimize matches.
707 * tests/patterntest.c: added more match cases.
709 2005-09-16 Tor Lillqvist <tml@novell.com>
711 * glib/gstrfuncs.c (g_ascii_strcasecmp, g_ascii_strncasecmp): Add
712 warning to doc comment that these functions should not be used on
713 encodings like CP932.
715 2005-09-14 Matthias Clasen <mclasen@redhat.com>
717 * tests/keyfile-test.c: Add a test for grup names of length 1.
719 * glib/gkeyfile.c (g_key_file_line_is_group): Accept group names
720 of length 1. (#316309)
722 2005-09-12 Matthias Clasen <mclasen@redhat.com>
724 * glib/gmarkup.c (g_markup_escape_text): Clarify docs.
726 2005-09-11 Sebastian Wilhelmi <seppi@seppi.de>
728 * tests/refcount/Makefile.am (INCLUDES): Link the the refcount
729 tests to the system thread library $(G_THREAD_LIBS). Fixes #313744
732 2005-09-11 Kjartan Maraas <kmaraas@gnome.org>
734 * glib/gmain.c: (g_child_watch_prepare), (g_child_watch_check),
735 (child_watch_helper_thread): Remove some dead code. Closes
738 2005-09-07 Tor Lillqvist <tml@novell.com>
740 * glib/Makefile.am: Create also a console version of the
741 gspawn-win32-helper program, gspawn-win32-helper-console.exe.
742 It's otherwise identical to gspawn-win32-helper.exe, except marked
743 as a console application (linked without the -mwindows option).
745 * glib/gspawn-win32.c (do_spawn_directly, do_spawn_with_pipes):
746 Drop the dont_wait parameter. Its truth value correlated 100% with
747 the NULLness of the exit_status parameter anyway, so it's enough
748 to check whether exit_status is NULL. Invert the sense of the
749 dont_return_handle parameter and rename it to do_return_handle, to
750 make the code easier to read by avoiding double negations.
752 (g_spawn_sync_utf8, g_spawn_async_with_pipes_utf8): Modify calls
753 to do_spawn_with_pipes() accordingly.
755 (do_spawn_with_pipes): If we have a console, use the console
756 version of the helper program, otherwise use the GUI one. This
757 avoids extra console windows opening up in some situations. (In
758 case a console application uses the GUI gspawn-win32-helper.exe to
759 spawn another console application we would get a separate console
760 for the spawned console application).
762 * glib-zip.in: Distribute also gspawn-win32-helper-console.exe.
764 2005-09-05 Matthias Clasen <mclasen@redhat.com>
766 * glib/gmappedfile.c (g_mapped_file_new): Report an error
767 if the file is too large. (#315275, Kjartan Maraas)
769 * glib/gkeyfile.c (g_key_file_load_from_fd): The return value
770 of read() is signed. (#315273, Kjartan Maraas)
772 2005-08-31 Tor Lillqvist <tml@novell.com>
774 * glib/gutils.h: Wrapping atexit() is a bad idea on Windows, where
775 the EXE and each DLL have their own atexit function chains.
777 #define g_atexit as atexit instead. This means it has a
778 better chance of doing what the caller wants. For instance,
779 gtkhtml calls g_atexit() registering a function in gtkhtml
780 itself. This caused a crash when g_atexit() was implemented as a
781 function in the GLib DLL. The gtkhtml DLL was already unloaded by
782 the time the GLib DLL got unloaded.
784 * glib/gutils.c: #undef the #define mentioned above, to also get a
785 real g_atexit() into the DLL for backward compatibility. Document
786 the Windows behaviour of g_atexit(), and document the varying ways
787 atexit() can behave in the context of dynamically loaded modules
790 2005-08-31 Matthias Clasen <mclasen@redhat.com>
794 * glib/gdataset.c: Add string interning functions.
796 2005-08-28 Matthias Clasen <mclasen@redhat.com>
798 * glib/giochannel.c: Unify some near-duplicate strings. (#314654,
801 2005-08-26 Matthias Clasen <mclasen@redhat.com>
803 * configure.in: Bump version to 2.9.0
805 2005-08-25 Tor Lillqvist <tml@novell.com>
807 Make also the g_spawn*() functions take parameters in the GLib
808 file name encoding, i.e. UTF-8, on Windows. Has no impact on Unix
809 API or ABI. Like the other GLib API that was earlier changed to
810 use UTF-8 on Windows, the names of the functions that take UTF-8
811 have _utf8 suffixes added by using preprocessor macros in the
812 header file. The old names are kept for functions with the old
813 behaviour, taking parameters in the system codepage, for DLL ABI
816 * glib/gspawn.h: On Win32 add the suffix _utf8 to the names of the
817 g_spawn*() functions.
819 * glib/gspawn-win32.c: Use wide-char API on NT-based
820 Windows. Convert parameters from UTF-8 to wide chars (NT) or
821 system codepage (Win9x) and call the C library _wspawn*() or
822 spawn*() functions respectvely. Add DLL ABI stability versions
823 that take parameters in the system codepage.
825 * glib/gspawn-win32-helper.c: On NT-based Windows use the
826 wide-char versions of argv and envp, and use wide-char API to
827 change directory and spawn the program to run. Remove the verbose
828 debugging output, it was too complex to modify for the wide-char
829 features. (Just add temporary debugging printouts if needed, no
830 need to have them permanently in the source.)
832 * glib/gspawn.c: Corresponding documentation updates.
834 * glib/glib.symbols: Corresponding changes: Mark the ABI stability
835 symbols as PRIVATE, add the new _utf8-suffixed ones.
837 2005-08-24 Stepan Kasal <kasal@ucw.cz>
839 * glib/gtypes.h (G_MININT64): Cast the constant to gint64; it is
840 guint64 otherwise and that can produce warnings about comparison
841 between signed and unsigned.
843 2005-08-23 Matthias Clasen <mclasen@redhat.com>
845 * glib/gutils.c: Fix the crt_externs.h include.
847 2005-08-23 Stepan Kasal <kasal@ucw.cz>
849 * NEWS: Fix spelling of my first name.
851 2005-08-23 Matthias Clasen <mclasen@redhat.com>
855 * === Released 2.8.1 ===
859 2005-08-20 Hans Breuer <hans@breuer.org>
861 * glib/makefile.msc.in : link with ws2_32.lib
863 2005-08-18 Tor Lillqvist <tml@novell.com>
865 * configure.in: Check for <sys/wait.h>
867 * glib/gbacktrace.c: Include <sys/wait.h> on if HAVE_SYS_WAIT_H.
869 2005-08-18 Ross Burton <ross@burtonini.com>
872 Optimise single-character insertions.
877 * tests/string-test.c:
878 Add tests for new optimisation, and fix a leak.
880 2005-08-17 Matthias Clasen <mclasen@redhat.com>
882 * configure.in: Check for crt_externs.h and _NSGetEnviron.
884 * glib/gutils.c: On Darwin, include crt-externs.h and
885 define environ using _NSGetEnviron(). (#313731)
887 2005-08-16 Stepan Kasal <kasal@ucw.cz>
889 * glib/gutils.c (g_get_any_init): Move the body of the big if...
890 (g_get_any_init_do): ... to this new function.
891 (g_get_any_init): Declare as inline.
892 (g_get_any_init_locked): New inline function, does the locking.
893 Make use of these two throughout the code.
895 2005-08-15 Matthias Clasen <mclasen@redhat.com>
897 * glib/gbacktrace.c (g_on_error_stack_trace): Wait for
898 the child process and then simply return. This makes
899 The "S" option work as documented in g_on_error_query().
900 (#313125, Matthew F. Barnes)
902 * glib/gunicode.h: Update the link to Unicode category
903 values. (#313369, Behnam Esfahbod)
905 * glib/gqueue.c (g_queue_find_custom): Clarify docs
906 a little. (#311727, Tristan van Berkom)
908 * glib/abicheck.sh, gobject/abicheck.sh: Make the
909 check work on ia64 too, where some symbols ended up
910 in yet another section.