1 2005-12-20 Michael Meeks <michael.meeks@novell.com>
3 * glib/gmain.c (g_main_context_is_owner): new method
4 to determine if the current thread is the owner of the
7 2006-01-02 Matthias Clasen <mclasen@redhat.com>
11 * glib/gstdio.c (g_rmdir): Don't provide g_rmdir() as a macro
12 expanding to rmdir, since rmdir is not declared in a portable
13 system header we can include in gstdio.h. (#325249, Jani Monoses)
15 2006-01-01 Tor Lillqvist <tml@novell.com>
17 * glib/gspawn-win32.c (g_spawn_sync_utf8): Set the GIOChannels for
18 stdout and stderr to unbuffered. Otherwise the giochannel layer
19 will try to read from them regardless whether the
20 g_io_channel_win32_poll() call here has indicated
21 readability or not. (#325310)
23 2005-12-29 Matthias Clasen <mclasen@redhat.com>
25 * glib/gutils.c (glib_check_version): Fix a copy-and-paste error
26 in the docs. (#325273, Declan Naughton)
28 2005-12-27 Manish Singh <yosh@gimp.org>
30 * tests/run-collate-tests.sh: use LC_ALL instead of LC_COLLATE, to
31 make sure we really override things.
33 2005-12-27 Matthias Clasen <mclasen@redhat.com>
35 Fix #316221, Michal Benes, Stanislav Brabec;
37 * configure.in: Fix a strict aliasing problem in
38 g_static_mutex_get_mutex().
39 * glib/gthread.h: ...and in
40 g_static_mutex_get_mutex_impl_shortcut().
42 * glib/gdatasetprivate.h: Add a cast to silence compiler
43 warnings. (#321978, Andrew Paprocki)
45 Partial fix for bug #32937, Bogdan Nicula.
47 * configure.in: Check for malloc.h
48 * glib/gslice.c: Don't include malloc.h unconditionally.
50 2005-12-27 Manish Singh <yosh@gimp.org>
52 * tests/run-collate-tests.sh: export LC_COLLATE so it takes.
53 (#324950, Dan Yefimov)
55 2005-12-26 Matthias Clasen <mclasen@redhat.com>
57 * glib/gslice.c: On Win32, include process.h (#325015, Kazuki
60 2005-12-25 Matthias Clasen <mclasen@redhat.com>
62 * AUTHORS: Update my email
64 * tests/slice-test.c: Fix C99isms. (#324950, Dan Yefimov)
66 2005-12-21 Matthias Clasen <mclasen@redhat.com>
68 * README.in: Add some notes on when 'make check' may fail.
70 2005-12-20 Matthias Clasen <mclasen@redhat.com>
74 * glib/gthreadpool.c (g_thread_pool_set_sort_function): New function
75 to sort tasks pushed into a threadpool. (#324479, Martyn Russell)
77 * tests/threadpool-test.c: Test this.
79 Tue Dec 20 18:14:14 2005 Tim Janik <timj@imendio.com>
81 * glib/gslice.[hc]: added mem_error() and mem_assert() to test and
82 handle errors without depending on gmessage.c which might not be
83 setup when the error occours.
84 removed G_SLICE_CONFIG_ALWAYS_FREE config option, fixed the code so
85 always freeing can be achieved by adjusting the working set time to
86 0 with G_SLICE_CONFIG_WORKING_SET_MSECS.
87 added G_SLICE_CONFIG_COLOR_INCREMENT to test different color increments
88 (mainly 0 and 1). reduced the minimum block size to 128 bytes, to
89 minimize wastage if small amounts of differently sized structrues are
90 allocated, this does come at a performance cost of roughly 5% though.
91 fixed up block alignment calculation, so it works for varying
92 block sizes. only use strerror() not g_strerror() since the latter
93 depends on working GQuark and GSlice.
94 mem_error(): implemented in terms of fprintf and vfprintf.
96 * tests/slice-color.c: new program to test cache colorization effects.
98 * tests/slice-test.c: trade G_SLICE_CONFIG_ALWAYS_FREE for 0 duration
99 G_SLICE_CONFIG_WORKING_SET_MSECS.
101 2005-12-17 Matthias Clasen <mclasen@redhat.com>
103 * glib/goption.c (parse_short_option): Set an error in all
104 failure cases. (#324332, Tim-Philipp Müller)
106 2005-12-17 Sebastian Wilhelmi <seppi@seppi.de>
108 * glib/gatomic.c: Fix memory barrier position in g_atomic_int_get
109 and g_atomic_pointer_get. Add g_atomic_int_set and
110 g_atomic_pointer_set implementations for the !DEFINE_WITH_MUTEXES &&
111 G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case, as well as defining them
112 as functions (additionally to the macros in the header) for the
113 !G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case.
115 2005-12-16 Matthias Clasen <mclasen@redhat.com>
117 * glib/gmem.c (g_allocator_new): Don't return a pointer to
118 a const struct, since apps expect to be able to modify it.
119 (#324179, J. Ali Harlow)
121 Tue Dec 13 10:13:32 2005 Tim Janik <timj@imendio.com>
123 * glib/gatomic.h: added g_atomic_pointer_set() and g_atomic_int_set()
124 for systems where the initialization of atomic variables requires a
125 write memory barrier.
127 2005-12-09 Matthias Clasen <mclasen@redhat.com>
129 * configure.in: Bump version
131 * === Released 2.9.1 ===
133 2005-12-09 Alexander Larsson <alexl@redhat.com>
137 Use G_GNUC_WARN_UNUSED_RESULT on list functions that return
140 2005-12-08 Matthias Clasen <mclasen@redhat.com>
144 2005-12-07 Martyn Russell <martyn@imendio.com>
146 * glib/gasyncqueue.c:
147 - Call g_queue_insert_sorted() instead of duplicating the code.
148 - Call g_queue_sort() instead of duplicating the code.
149 - Invert sort function results to make sure the same sort function
150 gives the same results across glist, gslist, gqueue and
153 * tests/asyncqueue-test.c:
154 - Updated the sort function to reflect the example in the
155 documentation for gasyncqueue.c.
157 2005-12-07 Martyn Russell <martyn@imendio.com>
159 * docs/reference/glib/glib-sections.txt:
160 * docs/reference/glib/tmpl/linked_lists_double.sgml:
161 * docs/reference/glib/tmpl/linked_lists_single.sgml:
164 - Added g_list_insert_sorted_with_data () and
165 g_slist_insert_sorted_with_data ().
166 - Removed the extra check in g_list_sort() and g_slist_sort() for
167 GCompareDataFunc vs. GCompareFunc.
169 2005-12-07 Tor Lillqvist <tml@novell.com>
171 * glib-gettextize.in: Look up prefix at run-time on Win32,
172 assuming the standard directory structure with glib-gettextize in
175 2005-12-06 Matthias Clasen <mclasen@redhat.com>
177 * glib/gmem.h: Exempt GTK+ from the mem chunk deprecation, since
178 we need GTK+ 2.8 to compile against GLib 2.10.
180 2005-12-05 Manish Singh <yosh@gimp.org>
182 * tests/run-collate-test.sh: set LC_COLLATE, not LANG, to be sure
183 to override any user settings.
185 2005-12-05 Matthias Clasen <mclasen@redhat.com>
187 * configure.in: Revert an accidential commit
189 2005-12-05 Matthias Clasen <mclasen@redhat.com>
191 * glib/gutf8.c: Documentation updates. (#323291, Morten Welinder)
193 * tests/Makefile.am (TESTS_ENVIRONMENT):
194 * tests/gobject/Makefile.am (TESTS_ENVIRONMENT): Set
195 MALLOC_CHECK_ and MALLOC_PERTURB_
197 * tests/run-collate-tests.sh: Run the collation tests explicitly
198 in en_US locale. (#320463)
200 * glib/gmem.h: Really deprecate GMemChunk.
207 * glib/gstrfuncs.c: Add versioned deprecation docs.
209 2005-12-05 Martyn Russell <martyn@imendio.com>
211 * docs/reference/glib/glib-sections.txt:
212 * glib/gasyncqueue.[ch]:
213 - Added support for sorting async queues by with _push_sorted(),
214 _push_sorted_unlocked(), _sort() and _sort_unlocked() (#323047).
217 * tests/asyncqueue-test.c:
218 - Added test case for gasyncqueue.c
220 Mon Dec 5 15:53:20 2005 Tim Janik <timj@imendio.com>
222 * glib/gslice.c: implement chain walking for arbitrary ->next pointer
223 offsets in g_slice_free_chain_with_offset() based on a patch by behdad
224 in bug 323178. moved time consuming logic from g_slice_free() out of
225 the inner loop, so g_slice_free_chain_with_offset() provides a real
226 performance benefit over g_slice_free1() now.
228 * glib/gslice.h: renamed g_slice_free_chain() to
229 g_slice_free_chain_with_offset(). implemented g_slice_free_chain() as
230 a type-safe macro as suggested in bug 323178.
231 simplified the macro implementation of g_slice_free() and implemented
232 it in a type safe manner for all compliers as suggested by Morten
233 Welinder <mortenw@gnome.org>.
238 * glib/glib.symbols: s/g_slice_free_chain/g_slice_free_chain_with_offset/
240 2005-12-05 Matthias Clasen <mclasen@redhat.com>
242 * glib/gasyncqueue.c: Add some docs.
244 * tests/libmoduletestplugin_a.c: Fix compiler warnings.
246 * glib/gatomic.c: In the ia64 implementation, use
247 __sync builtin without _si or _di suffix. (#321229,
248 Stanislav Brabec, patch by Andreas Schwab)
250 2005-12-04 Behdad Esfahbod <behdad@gnome.org>
252 * glib/gslice.h: Remove comma at the end of enum.
254 2005-12-04 Matthias Clasen <mclasen@redhat.com>
256 Handle multiple user names with the same UID better.
257 (#319535, Laszlo Peter)
259 * glib/gutils.c (g_get_any_init_do): When determining user
260 data, first look up $LOGNAME. If the UID doesn't match
261 getuid(), fall back to the current behaviour of looking
262 up the user data based on getuid().
264 2005-12-04 Matthias Clasen <mclasen@redhat.com>
266 * glib/gmacros.h (G_GNUC_WARN_UNUSED_RESULT): Add a macro
267 to make gcc warn if a function result is ignored. (#145466,
268 Arjan van de Ven, Alex Larsson)
270 * glib/gmem.h: Add the new attribute to g_realloc and
273 2005-12-03 Matthias Clasen <mclasen@redhat.com>
276 * glib/gcache.h: Deprecate g_cache_value_foreach. (#322956,
279 * glib/gmappedfile.c: Make mapping of empty files
282 * glib/gfileutils.c: Don't fork a new process just to
283 fix the permissions of the created temp file. (#321318,
284 Alexis S. L. Carvalho)
286 2005-12-02 Matthias Clasen <mclasen@redhat.com>
288 * README.in: Add a note about Solaris threads.
291 * configure.in: Remove support for Solaris threads.
292 (#136971, Sebastian Wilhelmi, patch by Andrew Paprocki)
294 2005-12-02 Matthias Clasen <mclasen@redhat.com>
296 * configure.in: Define G_GUINT64_CONSTANT in analogy to
297 G_GINT64_CONSTANT. (#322568, Andrew Paprocki)
299 2005-12-02 Matthias Clasen <mclasen@redhat.com>
301 * glib/gslice.c: Win32 portability fixes and C99ism removal,
302 pointed out by Kazuki Iwamoto. (#323052)
304 2005-12-02 Matthias Clasen <mclasen@redhat.com>
306 * glib/gatomic.c (g_atomic_int_add): Add a missing volatile for
307 the IA64 implementation.
309 Fri Dec 2 16:18:09 2005 Tim Janik <timj@imendio.com>
311 * glib/gslice.c: some naming and type size fixups.
313 Fri Dec 2 13:08:58 2005 Tim Janik <timj@gtk.org>
315 * glib/gslice.h (g_slice_free): reworked GCC-specific type-safe macro
316 variant into something less verbose, but digestible for gcc-3.4.
318 Fri Dec 2 10:55:07 2005 Tim Janik <timj@imendio.com>
320 * tests/slice-test.c: extended to perform the benchmarking on the old
321 memchunk code if 'O' is selected.
323 * tests/memchunks.c: new file which contains the old GLib mem chunks
324 implementation with prefix old_mem_chunk_.
326 * tests/Makefile.am: added memchunks.c
328 Fri Dec 2 00:16:59 2005 Tim Janik <timj@imendio.com>
330 * glib/gslice.c: improved gettimeofday() timer resolution by maintaining
331 acceptable load distribution of the syscall.
332 (allocator_get_magazine_threshold): reduce minimum/single-thread magazine
333 sizes to approximately page_size/2.5 to avoid excessive startup allocations.
335 2005-12-01 Matthias Clasen <mclasen@redhat.com>
337 * tests/gobject/ifacecheck.c (test_iface_base_init): Don't
338 double-free base interface members.
340 * tests/Makefile.am: Remove duplicate variable.
342 Thu Dec 1 17:32:46 2005 Tim Janik <timj@imendio.com>
344 * glib/gslice.[hc]: new slice allocator implementation.
346 * tests/slice-test.c: added random slice allocation test.
348 * glib/gthread.[hc]: removed newly added private thread mem API.
350 * glib/gthreadinit.h:
353 * glib/gmem.c: divided glib threading initialisation into three phases,
354 initialisation where private keys and messaging are not available (only
355 needed by gmem.c), initialisation without messaging but private keys
356 available (gslice.c, gmessage.c), and full fledged initialisers that
357 server the rest of glib. initialisation functions got renamed to reflect
358 the limitations of their corresponding phases.
360 * glib/gmem.c: removed memchunk code, defer allocations to
363 * glib/gmem.[hc]: removed g_slice_* skeletons.
365 * glib/glib.symbols: added g_slice_* symbols.
367 * configure.in: check for availability of posix_memalign(3), memalign(3)
370 * glib/Makefile.am: added gslice.[hc].
372 2005-12-01 Tor Lillqvist <tml@novell.com>
374 * glib/gstdio.c (g_stat): In the Win32 implementation, strip
375 trailing slash(es) for non-root folders. stat() fails if non-root
376 folders are specified with trailing slashes. It's too much hassle
377 to demand that callers strip such slashes themselves, especially
378 as it is easy to get it wrong and strip the slash of a root
380 (g_rename): On NT-based Windows, use MoveFileEx() with
381 MOVEFILE_REPLACE_EXISTING to better match Unix behaviour.
383 2005-11-28 Matthias Clasen <mclasen@redhat.com>
385 Fix G_STMT_START / G_STMT_END on Solaris. (#321972,
388 * configure.in: Check whether do { } while (0) works.
390 * glib/gmacros.h: Use do { } while (0) for G_STMT_START /
391 G_STMT_END if it works.
393 2005-11-28 Matthias Clasen <mclasen@redhat.com>
395 * glib/gthread.c (g_static_rw_lock_wait, g_static_rw_lock_signal):
396 * glib/gnode.c (g_node_depth_traverse_level):
397 * glib/gmem.c (g_allocator_new):
398 * glib/ghash.c (g_hash_table_unref, g_hash_table_destroy)
399 (g_hash_table_foreach_remove_or_steal): Silence compiler
402 2005-11-27 Matthias Clasen <mclasen@redhat.com>
404 * glib/gunicollate.c (g_utf8_collate_key_for_filename):
405 Don't read beyond len. (#322520, Christian Persch)
407 Wed Nov 23 17:34:01 2005 Tim Janik <timj@imendio.com>
409 * glib/gdataset.c: access datalist flags via atomic pointer access
410 functions, instead of acquiring the dataset lock. this is faster and
411 also matches the atomic pointer readouts in gdatalistprivate.h and
412 direct pointer modifications required by gobject.c.
414 Wed Nov 23 13:35:31 2005 Tim Janik <timj@gtk.org>
416 * glib/gdataset.c: streamlined initialization code somewhat,
417 removed GData node cache.
419 2005-11-22 Matthias Clasen <mclasen@redhat.com>
421 * glib/ghash.c (g_hash_table_ref):
422 (g_hash_table_unref): Mark these as new API.
424 Avoid double locking in g_intern_string (#322133,
427 * glib/gdataset.c (g_quark_from_string_internal): New
428 internal function which factors out common parts of
429 g_quark_from[_static]_string.
430 (g_quark_from_string, g_quark_from_static_string):
431 Use g_quark_from_string_internal.
432 (g_intern_string, g_intern_static_string): Use
433 g_quark_from_string_internal, and only take the
435 (g_quark_new): Don't store the strings shifted by -1
436 in the g_quarks array.
437 (g_quark_to_string): Adapt to the previous change.
439 Tue Nov 22 14:04:26 2005 Tim Janik <timj@imendio.com>
443 g_hash_table_new_full(): create hash tables with a ref count of 1.
444 g_hash_table_ref(): atomically ref_count+=1
445 g_hash_table_unref(): atomically ref_count-=1, destroys hash table
446 when refcount reaches 0.
447 g_hash_table_destroy(): just destroy keys and values, unref by 1.
448 g_hash_table_insert():
449 g_hash_table_replace(): assert ref_count>0.
452 * glib/gatomic.c: added 'volatile' qualifier to all atomic pointer and
455 2005-11-20 Behdad Esfahbod <behdad@gnome.org>
457 * glib/guniprop.c (g_unichar_get_mirror_char): Remove unused
458 variables. (#321984, Andrew Paprocki)
460 2005-11-18 Matthias Clasen <mclasen@redhat.com>
462 * configure.in: Bump version
464 2005-11-17 Matthias Clasen <mclasen@redhat.com>
466 * === Released 2.9.0 ===
468 * glib/glib.symbols: Updates
470 * glib/gutf8.c (g_utf8_offset_to_pointer): Add some performance
471 hints to the docs. (Paolo Borelli)
475 * glib/gthread.c (g_thread_foreach): Mark as new api.
477 * README.in: Updates.
481 * glib/gdate.c (g_date_set_time_t):
482 (g_date_set_time_val): Add functions to set a date from
483 a time_t and from a GTimeVal.
484 (g_date_set_time): Deprecate. (#314067, Roger Leigh)
487 * tests/utf8-pointer.c: Unit tests for g_utf8_pointer_to_offset
488 and g_utf8_offset_to_pointer.
490 * glib/gutf8.c (g_utf8_pointer_to_offset)
491 (g_utf8_offset_to_pointer): Handle negative offsets, and use
492 "stutter stepping" for going backwards. (#320638, Larry
497 * glib/gthread.c: const correctness fixes, found
498 by Arjan van de Ven and gcc.
500 2005-11-16 Behdad Esfahbod <behdad@gnome.org>
502 * docs/reference/glib/tmpl/unicode.sgml: Correct typo on mentioning
503 @G_UNICODE_BREAK_UNKNOWN that should be #G_UNICODE_BREAK_UNKNOWN.
505 2005-11-09 Behdad Esfahbod <behdad@gnome.org>
507 * glib/guniprop.c: Use bit hacks instead when checking a general
508 category value against multiple values.
510 * glib/gutf8.c: Change ISO10646 to Unicode in docs.
512 2005-11-10 Simos Xenitellis <simos@gnome.org>
514 * configure.in: Added tt (Tatar) to ALL_LINGUAS.
516 2005-11-05 Tor Lillqvist <tml@novell.com>
518 * glib/giowin32.c (g_io_win32_set_flags): Don't call g_warning().
519 (#320688) Instead set error and return failure.
521 2005-11-07 Matthias Clasen <mclasen@redhat.com>
523 * glib/gmain.c (g_main_dispatch): Don't call cb_funcs->unref
524 while holding the context lock. (#320886, Andy Wingo)
526 2005-11-07 Billy Biggs <vektor@dumbterm.net>
528 * tests/refcount/Makefile.am: Add a missing reference to libglib
529 in the LDADD for this test.
531 2005-11-07 Matthias Clasen <mclasen@redhat.com>
533 * glib/gmem.c (g_allocator_new): We need to set n_preallocs to a
534 nonzero value, otherwise GTK+ 2.8 breaks when compiled against
535 GLib 2.9. (#320755, Luca Ferretti)
537 2005-11-04 Matthias Clasen <mclasen@redhat.com>
539 * glib/Makefile.am: Apply a patch from Behdad Esfahbod to
540 use a faster lookup table for g_unichar_get_mirror_char().
541 * glib/gmirroringtable.h: The new table.
542 * glib/gunichartables.h: Remove bidi_mirroring_table.
543 * glib/gen-unicode-tables.pl: Don't generate the mirroring
545 * glib/glib-mirroring-tab/*: Sources for the program
546 which generated gmirroringtable.h.
548 * glib/glist.c: Avoid some code duplication.
550 * glib/gscanner.h: Include gdataset.h. (#320322)
552 * glib/gdebug.h: add new GDebugFlag for fatal_criticals
553 * glib/gmessages.c: (_g_debug_init): handle G_DEBUG=fatal_criticals,
554 to help find critical warnings in applications. (#320017,
557 2005-11-02 Tor Lillqvist <tml@novell.com>
559 * glib/glib.symbols: Remove large amount of trailing whitespace
560 from one line. Remove #ifdef G_UNDEFINED from around
561 g_hash_table_new and g_hash_table_new_full symbols, what was the
562 purpose of that? Add the g_slice functions.
564 Tue Nov 1 16:24:20 2005 Tim Janik <timj@imendio.com>
566 * glib/gmem.[hc]: prepared deprecation of GMemChunk and GAllocator.
567 added g_slice_*() API to allocate and cache small bits of memory.
568 an actuall allocator implementation for g_slice_*() is still pending.
570 * glib/gthread.[hc]: changes from a patch by Matthias Clasen.
571 changed GRealThread list to use in-structure *next; fields instead
572 of GSList, in order for thread iteration to not depenend on g_slice_*()
574 _g_thread_mem_private_get():
575 _g_thread_mem_private_set(): added accessors for private memory,
576 needed because the ordinary GPrivate implementation relies on GArray
577 and GSList and therefore indirectly on working g_slice_*() allocations.
580 g_thread_foreach(): new public API function to loop over all existing threads.
594 removed GAllocator and free list usages and accompanying locks.
595 use g_slice_*() API to allocate and cache small bits of memory.
597 * glib/ghook.h: removed GMemChunk field from public API.
600 * glib/glist.h: deprecate allocator API, provide _free1() for consistency.
602 * glib/gnode.h: deprecate allocator API.
604 * glib/gmain.c: reordered GPollRec fields so g_slice_free_chain() can
605 be used for poll rec lists.
607 * glib/grel.c: removed mem chunk usage, and allocated tuples via g_slice_*().
608 g_relation_destroy(): free all tuples from the all_tuples hash table,
609 this effectively maintains the life time track keeping of tuples.
610 g_relation_delete_tuple(): free tuples which are removed from the
611 all_tuples hash table. this fixes a temporary leak that was present
612 in the memchunk code until the destruction of the relation.
614 2005-10-29 Matthias Clasen <mclasen@redhat.com>
616 * tests/convert-test.c: Add some tests for conversions between
617 UTF-8, UCS-4 and UTF-16.
619 * glib/gutf8.c (g_utf8_to_ucs4, g_utf8_to_utf16): Fix handling
620 of len == -1, noticed by Morten Welinder.
622 2005-10-27 Erdal Ronahi <erdal.ronahi@gmail.com>
624 * configure.in: Added ku (Kurdish) to ALL_LINGUAS
626 2005-10-26 Matthias Clasen <mclasen@redhat.com>
628 * glib/gutf8.c (g_ucs4_to_utf8): Don't set items_read twice
629 in the error case, and add some documentation. (#319806, Morten
632 2005-10-19 Manish Singh <yosh@gimp.org>
634 * configure.in: Use AC_CHECK_FUNCS for _NSGetEnviron, to get the
635 config.h symbol automatically. Fixes bug #313731.
637 2005-10-19 Tor Lillqvist <tml@novell.com>
639 * glib/gwin32.c (g_win32_get_package_installation_directory):
640 Return a g_strdup()ed copy of the value stored in the hash table,
641 so that it can be g_free()d without leaving a dangling pointer in
642 the hash table. (#319232)
644 2005-10-06 Matthias Clasen <mclasen@redhat.com>
646 * glib/gunicollate.c (g_utf8_collate_key_for_filename): Handle
647 all-zero sequences correctly. (#317930, Sebastien Bacher)
649 2005-10-05 Matthias Clasen <mclasen@redhat.com>
651 * glib/gmarkup.c (g_markup_escape_text): Doc addition
653 * Makefile.am: Add Collation tests.
655 * tests/collate/*: Inputs and expected outputs for collation tests.
657 * tests/run-collate-tests.sh: Script to run collation tests.
659 * tests/unicode-collate.c (main): Rework slightly to make
660 it usable in unit tests. Also test g_utf8_collate_key_for_filename().
662 2005-10-01 Behdad Esfahbod <behdad@gnome.org>
664 * docs/reference/glib/tmpl/unicode.sgml:
665 * glib/gen-unicode-tables.pl:
667 * glib/gunichartables.h:
669 * tests/casefold.txt:
670 * tests/casemap.txt: Updated to Unicode 4.1. There are five new
671 GUnicodeBreakType types. That may break some applications, like
674 2005-09-26 Matthias Clasen <mclasen@redhat.com>
676 * glib/gstrfuncs.c (g_ascii_strtoull): Add details to the
677 docs. (#314393, Matthew F. Barnes)
680 * glib/gprintf.h: Remove g_snprintf() and g_vsnprintf(), since
681 they are already declared in glib.h. This doesn't break documented
682 use of gprintf.h, but should probably be pointed out in the
683 release notes for 2.10. (#314232, Behdad Esfahbod)
685 Tue Sep 20 13:16:04 2005 Tim Janik <timj@imendio.com>
687 * glib/gpattern.c (g_pattern_ph_match): applied significant recursion
688 complexity optimization, based on a patch from Matthias Clasen.
690 * tests/patterntest.c: more tests, mostly from matthias.
692 2005-09-20 Matthias Clasen <mclasen@redhat.com>
694 * glib/gqueue.c (g_queue_insert_sorted): Correct the docs.
695 (#316703, Mark Drago)
697 Mon Sep 19 17:23:23 2005 Tim Janik <timj@imendio.com>
699 * glib/gpattern.c: applied a patch from matthias which checks on the
700 upper bound of GPatternSpec length to optimize matches.
703 * tests/patterntest.c: added more match cases.
705 2005-09-16 Tor Lillqvist <tml@novell.com>
707 * glib/gstrfuncs.c (g_ascii_strcasecmp, g_ascii_strncasecmp): Add
708 warning to doc comment that these functions should not be used on
709 encodings like CP932.
711 2005-09-14 Matthias Clasen <mclasen@redhat.com>
713 * tests/keyfile-test.c: Add a test for grup names of length 1.
715 * glib/gkeyfile.c (g_key_file_line_is_group): Accept group names
716 of length 1. (#316309)
718 2005-09-12 Matthias Clasen <mclasen@redhat.com>
720 * glib/gmarkup.c (g_markup_escape_text): Clarify docs.
722 2005-09-11 Sebastian Wilhelmi <seppi@seppi.de>
724 * tests/refcount/Makefile.am (INCLUDES): Link the the refcount
725 tests to the system thread library $(G_THREAD_LIBS). Fixes #313744
728 2005-09-11 Kjartan Maraas <kmaraas@gnome.org>
730 * glib/gmain.c: (g_child_watch_prepare), (g_child_watch_check),
731 (child_watch_helper_thread): Remove some dead code. Closes
734 2005-09-07 Tor Lillqvist <tml@novell.com>
736 * glib/Makefile.am: Create also a console version of the
737 gspawn-win32-helper program, gspawn-win32-helper-console.exe.
738 It's otherwise identical to gspawn-win32-helper.exe, except marked
739 as a console application (linked without the -mwindows option).
741 * glib/gspawn-win32.c (do_spawn_directly, do_spawn_with_pipes):
742 Drop the dont_wait parameter. Its truth value correlated 100% with
743 the NULLness of the exit_status parameter anyway, so it's enough
744 to check whether exit_status is NULL. Invert the sense of the
745 dont_return_handle parameter and rename it to do_return_handle, to
746 make the code easier to read by avoiding double negations.
748 (g_spawn_sync_utf8, g_spawn_async_with_pipes_utf8): Modify calls
749 to do_spawn_with_pipes() accordingly.
751 (do_spawn_with_pipes): If we have a console, use the console
752 version of the helper program, otherwise use the GUI one. This
753 avoids extra console windows opening up in some situations. (In
754 case a console application uses the GUI gspawn-win32-helper.exe to
755 spawn another console application we would get a separate console
756 for the spawned console application).
758 * glib-zip.in: Distribute also gspawn-win32-helper-console.exe.
760 2005-09-05 Matthias Clasen <mclasen@redhat.com>
762 * glib/gmappedfile.c (g_mapped_file_new): Report an error
763 if the file is too large. (#315275, Kjartan Maraas)
765 * glib/gkeyfile.c (g_key_file_load_from_fd): The return value
766 of read() is signed. (#315273, Kjartan Maraas)
768 2005-08-31 Tor Lillqvist <tml@novell.com>
770 * glib/gutils.h: Wrapping atexit() is a bad idea on Windows, where
771 the EXE and each DLL have their own atexit function chains.
773 #define g_atexit as atexit instead. This means it has a
774 better chance of doing what the caller wants. For instance,
775 gtkhtml calls g_atexit() registering a function in gtkhtml
776 itself. This caused a crash when g_atexit() was implemented as a
777 function in the GLib DLL. The gtkhtml DLL was already unloaded by
778 the time the GLib DLL got unloaded.
780 * glib/gutils.c: #undef the #define mentioned above, to also get a
781 real g_atexit() into the DLL for backward compatibility. Document
782 the Windows behaviour of g_atexit(), and document the varying ways
783 atexit() can behave in the context of dynamically loaded modules
786 2005-08-31 Matthias Clasen <mclasen@redhat.com>
790 * glib/gdataset.c: Add string interning functions.
792 2005-08-28 Matthias Clasen <mclasen@redhat.com>
794 * glib/giochannel.c: Unify some near-duplicate strings. (#314654,
797 2005-08-26 Matthias Clasen <mclasen@redhat.com>
799 * configure.in: Bump version to 2.9.0
801 2005-08-25 Tor Lillqvist <tml@novell.com>
803 Make also the g_spawn*() functions take parameters in the GLib
804 file name encoding, i.e. UTF-8, on Windows. Has no impact on Unix
805 API or ABI. Like the other GLib API that was earlier changed to
806 use UTF-8 on Windows, the names of the functions that take UTF-8
807 have _utf8 suffixes added by using preprocessor macros in the
808 header file. The old names are kept for functions with the old
809 behaviour, taking parameters in the system codepage, for DLL ABI
812 * glib/gspawn.h: On Win32 add the suffix _utf8 to the names of the
813 g_spawn*() functions.
815 * glib/gspawn-win32.c: Use wide-char API on NT-based
816 Windows. Convert parameters from UTF-8 to wide chars (NT) or
817 system codepage (Win9x) and call the C library _wspawn*() or
818 spawn*() functions respectvely. Add DLL ABI stability versions
819 that take parameters in the system codepage.
821 * glib/gspawn-win32-helper.c: On NT-based Windows use the
822 wide-char versions of argv and envp, and use wide-char API to
823 change directory and spawn the program to run. Remove the verbose
824 debugging output, it was too complex to modify for the wide-char
825 features. (Just add temporary debugging printouts if needed, no
826 need to have them permanently in the source.)
828 * glib/gspawn.c: Corresponding documentation updates.
830 * glib/glib.symbols: Corresponding changes: Mark the ABI stability
831 symbols as PRIVATE, add the new _utf8-suffixed ones.
833 2005-08-24 Stepan Kasal <kasal@ucw.cz>
835 * glib/gtypes.h (G_MININT64): Cast the constant to gint64; it is
836 guint64 otherwise and that can produce warnings about comparison
837 between signed and unsigned.
839 2005-08-23 Matthias Clasen <mclasen@redhat.com>
841 * glib/gutils.c: Fix the crt_externs.h include.
843 2005-08-23 Stepan Kasal <kasal@ucw.cz>
845 * NEWS: Fix spelling of my first name.
847 2005-08-23 Matthias Clasen <mclasen@redhat.com>
851 * === Released 2.8.1 ===
855 2005-08-20 Hans Breuer <hans@breuer.org>
857 * glib/makefile.msc.in : link with ws2_32.lib
859 2005-08-18 Tor Lillqvist <tml@novell.com>
861 * configure.in: Check for <sys/wait.h>
863 * glib/gbacktrace.c: Include <sys/wait.h> on if HAVE_SYS_WAIT_H.
865 2005-08-18 Ross Burton <ross@burtonini.com>
868 Optimise single-character insertions.
873 * tests/string-test.c:
874 Add tests for new optimisation, and fix a leak.
876 2005-08-17 Matthias Clasen <mclasen@redhat.com>
878 * configure.in: Check for crt_externs.h and _NSGetEnviron.
880 * glib/gutils.c: On Darwin, include crt-externs.h and
881 define environ using _NSGetEnviron(). (#313731)
883 2005-08-16 Stepan Kasal <kasal@ucw.cz>
885 * glib/gutils.c (g_get_any_init): Move the body of the big if...
886 (g_get_any_init_do): ... to this new function.
887 (g_get_any_init): Declare as inline.
888 (g_get_any_init_locked): New inline function, does the locking.
889 Make use of these two throughout the code.
891 2005-08-15 Matthias Clasen <mclasen@redhat.com>
893 * glib/gbacktrace.c (g_on_error_stack_trace): Wait for
894 the child process and then simply return. This makes
895 The "S" option work as documented in g_on_error_query().
896 (#313125, Matthew F. Barnes)
898 * glib/gunicode.h: Update the link to Unicode category
899 values. (#313369, Behnam Esfahbod)
901 * glib/gqueue.c (g_queue_find_custom): Clarify docs
902 a little. (#311727, Tristan van Berkom)
904 * glib/abicheck.sh, gobject/abicheck.sh: Make the
905 check work on ia64 too, where some symbols ended up
906 in yet another section.