1 2006-01-08 Matthias Clasen <mclasen@redhat.com>
3 * glib/gslice.c: Include config.h first, pointed out
6 2005-12-20 Sven Herzberg <herzi@gnome-de.org>
8 * docs/reference/gobject/tmpl/gtype.sgml: explain that
9 G_TYPE_INSTANCE_GET_CLASS() does behave different during
12 2006-01-06 Matthias Clasen <mclasen@redhat.com>
14 * glib/gslice.c: Only define _XOPEN_SOURCE if we know
15 that we have posix_memalign(). (#323937, Bogdan Nicula)
17 2006-01-05 Matthias Clasen <mclasen@redhat.com>
19 * configure.in: Bump version
21 * === Released 2.9.2 ===
23 2006-01-05 Hans Breuer <hans@breuer.org>
25 * glib/makefile.msc.in : added gslice.obj
27 2006-01-05 Matthias Clasen <mclasen@redhat.com>
31 * glib/gmain.c: Clarify the documentation of source ids
32 a bit. (#325874, Dan Williams)
34 * configure.in: Fix another strict aliasing problem.
36 * glib/gthreadpool.c: Avoid use of varargs macro. (#325864,
39 2006-01-04 Manish Singh <yosh@gimp.org>
42 * glib/gstdio.h: don't macroized g_access, g_chdir, and g_unlink
43 either, since they have the same issue as g_rmdir. (Related to
46 Wed Jan 4 13:33:25 2006 Tim Janik <timj@gtk.org>
48 * glib/gslice.c (magazine_cache_pop_magazine): don't reverse chunk
49 order when creating magazines, so we hand out chunks with ascending
52 2006-01-04 Abel Cheung <maddog@linuxhall.org>
54 * configure.in: Added "ml" "zh_HK" to ALL_LINGUAS.
56 2006-01-03 Martyn Russell <martyn@imendio.com>
58 * docs/reference/glib/glib-sections.txt:
60 * glib/gthreadpool.[ch]:
61 - Added new API g_thread_pool_get_idle_time() and
62 g_thread_pool_set_idle_time(). (#324228).
64 * tests/threadpool-test.c:
65 - Updated test case to do thread pool sorting, thread pool with
66 no sorting and a thread pool with idle thread timeouts.
68 2006-01-03 Matthias Clasen <mclasen@redhat.com>
70 * glib/gmain.h: Add new functions here, too.
72 * glib/glib.symbols: Add new functions.
74 2005-12-20 Michael Meeks <michael.meeks@novell.com>
76 * glib/gmain.c (g_main_context_is_owner): new method
77 to determine if the current thread is the owner of the
80 2006-01-02 Matthias Clasen <mclasen@redhat.com>
84 * glib/gstdio.c (g_rmdir): Don't provide g_rmdir() as a macro
85 expanding to rmdir, since rmdir is not declared in a portable
86 system header we can include in gstdio.h. (#325249, Jani Monoses)
88 2006-01-01 Tor Lillqvist <tml@novell.com>
90 * glib/gspawn-win32.c (g_spawn_sync_utf8): Set the GIOChannels for
91 stdout and stderr to unbuffered. Otherwise the giochannel layer
92 will try to read from them regardless whether the
93 g_io_channel_win32_poll() call here has indicated
94 readability or not. (#325310)
96 2005-12-29 Matthias Clasen <mclasen@redhat.com>
98 * glib/gutils.c (glib_check_version): Fix a copy-and-paste error
99 in the docs. (#325273, Declan Naughton)
101 2005-12-27 Manish Singh <yosh@gimp.org>
103 * tests/run-collate-tests.sh: use LC_ALL instead of LC_COLLATE, to
104 make sure we really override things.
106 2005-12-27 Matthias Clasen <mclasen@redhat.com>
108 Fix #316221, Michal Benes, Stanislav Brabec;
110 * configure.in: Fix a strict aliasing problem in
111 g_static_mutex_get_mutex().
112 * glib/gthread.h: ...and in
113 g_static_mutex_get_mutex_impl_shortcut().
115 * glib/gdatasetprivate.h: Add a cast to silence compiler
116 warnings. (#321978, Andrew Paprocki)
118 Partial fix for bug #323937, Bogdan Nicula.
120 * configure.in: Check for malloc.h
121 * glib/gslice.c: Don't include malloc.h unconditionally.
123 2005-12-27 Manish Singh <yosh@gimp.org>
125 * tests/run-collate-tests.sh: export LC_COLLATE so it takes.
126 (#324950, Dan Yefimov)
128 2005-12-26 Matthias Clasen <mclasen@redhat.com>
130 * glib/gslice.c: On Win32, include process.h (#325015, Kazuki
133 2005-12-25 Matthias Clasen <mclasen@redhat.com>
135 * AUTHORS: Update my email
137 * tests/slice-test.c: Fix C99isms. (#324950, Dan Yefimov)
139 2005-12-21 Matthias Clasen <mclasen@redhat.com>
141 * README.in: Add some notes on when 'make check' may fail.
143 2005-12-20 Matthias Clasen <mclasen@redhat.com>
146 * glib/gthreadpool.h:
147 * glib/gthreadpool.c (g_thread_pool_set_sort_function): New function
148 to sort tasks pushed into a threadpool. (#324479, Martyn Russell)
150 * tests/threadpool-test.c: Test this.
152 Tue Dec 20 18:14:14 2005 Tim Janik <timj@imendio.com>
154 * glib/gslice.[hc]: added mem_error() and mem_assert() to test and
155 handle errors without depending on gmessage.c which might not be
156 setup when the error occours.
157 removed G_SLICE_CONFIG_ALWAYS_FREE config option, fixed the code so
158 always freeing can be achieved by adjusting the working set time to
159 0 with G_SLICE_CONFIG_WORKING_SET_MSECS.
160 added G_SLICE_CONFIG_COLOR_INCREMENT to test different color increments
161 (mainly 0 and 1). reduced the minimum block size to 128 bytes, to
162 minimize wastage if small amounts of differently sized structrues are
163 allocated, this does come at a performance cost of roughly 5% though.
164 fixed up block alignment calculation, so it works for varying
165 block sizes. only use strerror() not g_strerror() since the latter
166 depends on working GQuark and GSlice.
167 mem_error(): implemented in terms of fprintf and vfprintf.
169 * tests/slice-color.c: new program to test cache colorization effects.
171 * tests/slice-test.c: trade G_SLICE_CONFIG_ALWAYS_FREE for 0 duration
172 G_SLICE_CONFIG_WORKING_SET_MSECS.
174 2005-12-17 Matthias Clasen <mclasen@redhat.com>
176 * glib/goption.c (parse_short_option): Set an error in all
177 failure cases. (#324332, Tim-Philipp Müller)
179 2005-12-17 Sebastian Wilhelmi <seppi@seppi.de>
181 * glib/gatomic.c: Fix memory barrier position in g_atomic_int_get
182 and g_atomic_pointer_get. Add g_atomic_int_set and
183 g_atomic_pointer_set implementations for the !DEFINE_WITH_MUTEXES &&
184 G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case, as well as defining them
185 as functions (additionally to the macros in the header) for the
186 !G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case.
188 2005-12-16 Matthias Clasen <mclasen@redhat.com>
190 * glib/gmem.c (g_allocator_new): Don't return a pointer to
191 a const struct, since apps expect to be able to modify it.
192 (#324179, J. Ali Harlow)
194 Tue Dec 13 10:13:32 2005 Tim Janik <timj@imendio.com>
196 * glib/gatomic.h: added g_atomic_pointer_set() and g_atomic_int_set()
197 for systems where the initialization of atomic variables requires a
198 write memory barrier.
200 2005-12-09 Matthias Clasen <mclasen@redhat.com>
202 * configure.in: Bump version
204 * === Released 2.9.1 ===
206 2005-12-09 Alexander Larsson <alexl@redhat.com>
210 Use G_GNUC_WARN_UNUSED_RESULT on list functions that return
213 2005-12-08 Matthias Clasen <mclasen@redhat.com>
217 2005-12-07 Martyn Russell <martyn@imendio.com>
219 * glib/gasyncqueue.c:
220 - Call g_queue_insert_sorted() instead of duplicating the code.
221 - Call g_queue_sort() instead of duplicating the code.
222 - Invert sort function results to make sure the same sort function
223 gives the same results across glist, gslist, gqueue and
226 * tests/asyncqueue-test.c:
227 - Updated the sort function to reflect the example in the
228 documentation for gasyncqueue.c.
230 2005-12-07 Martyn Russell <martyn@imendio.com>
232 * docs/reference/glib/glib-sections.txt:
233 * docs/reference/glib/tmpl/linked_lists_double.sgml:
234 * docs/reference/glib/tmpl/linked_lists_single.sgml:
237 - Added g_list_insert_sorted_with_data () and
238 g_slist_insert_sorted_with_data ().
239 - Removed the extra check in g_list_sort() and g_slist_sort() for
240 GCompareDataFunc vs. GCompareFunc.
242 2005-12-07 Tor Lillqvist <tml@novell.com>
244 * glib-gettextize.in: Look up prefix at run-time on Win32,
245 assuming the standard directory structure with glib-gettextize in
248 2005-12-06 Matthias Clasen <mclasen@redhat.com>
250 * glib/gmem.h: Exempt GTK+ from the mem chunk deprecation, since
251 we need GTK+ 2.8 to compile against GLib 2.10.
253 2005-12-05 Manish Singh <yosh@gimp.org>
255 * tests/run-collate-test.sh: set LC_COLLATE, not LANG, to be sure
256 to override any user settings.
258 2005-12-05 Matthias Clasen <mclasen@redhat.com>
260 * configure.in: Revert an accidential commit
262 2005-12-05 Matthias Clasen <mclasen@redhat.com>
264 * glib/gutf8.c: Documentation updates. (#323291, Morten Welinder)
266 * tests/Makefile.am (TESTS_ENVIRONMENT):
267 * tests/gobject/Makefile.am (TESTS_ENVIRONMENT): Set
268 MALLOC_CHECK_ and MALLOC_PERTURB_
270 * tests/run-collate-tests.sh: Run the collation tests explicitly
271 in en_US locale. (#320463)
273 * glib/gmem.h: Really deprecate GMemChunk.
280 * glib/gstrfuncs.c: Add versioned deprecation docs.
282 2005-12-05 Martyn Russell <martyn@imendio.com>
284 * docs/reference/glib/glib-sections.txt:
285 * glib/gasyncqueue.[ch]:
286 - Added support for sorting async queues by with _push_sorted(),
287 _push_sorted_unlocked(), _sort() and _sort_unlocked() (#323047).
290 * tests/asyncqueue-test.c:
291 - Added test case for gasyncqueue.c
293 Mon Dec 5 15:53:20 2005 Tim Janik <timj@imendio.com>
295 * glib/gslice.c: implement chain walking for arbitrary ->next pointer
296 offsets in g_slice_free_chain_with_offset() based on a patch by behdad
297 in bug 323178. moved time consuming logic from g_slice_free() out of
298 the inner loop, so g_slice_free_chain_with_offset() provides a real
299 performance benefit over g_slice_free1() now.
301 * glib/gslice.h: renamed g_slice_free_chain() to
302 g_slice_free_chain_with_offset(). implemented g_slice_free_chain() as
303 a type-safe macro as suggested in bug 323178.
304 simplified the macro implementation of g_slice_free() and implemented
305 it in a type safe manner for all compliers as suggested by Morten
306 Welinder <mortenw@gnome.org>.
311 * glib/glib.symbols: s/g_slice_free_chain/g_slice_free_chain_with_offset/
313 2005-12-05 Matthias Clasen <mclasen@redhat.com>
315 * glib/gasyncqueue.c: Add some docs.
317 * tests/libmoduletestplugin_a.c: Fix compiler warnings.
319 * glib/gatomic.c: In the ia64 implementation, use
320 __sync builtin without _si or _di suffix. (#321229,
321 Stanislav Brabec, patch by Andreas Schwab)
323 2005-12-04 Behdad Esfahbod <behdad@gnome.org>
325 * glib/gslice.h: Remove comma at the end of enum.
327 2005-12-04 Matthias Clasen <mclasen@redhat.com>
329 Handle multiple user names with the same UID better.
330 (#319535, Laszlo Peter)
332 * glib/gutils.c (g_get_any_init_do): When determining user
333 data, first look up $LOGNAME. If the UID doesn't match
334 getuid(), fall back to the current behaviour of looking
335 up the user data based on getuid().
337 2005-12-04 Matthias Clasen <mclasen@redhat.com>
339 * glib/gmacros.h (G_GNUC_WARN_UNUSED_RESULT): Add a macro
340 to make gcc warn if a function result is ignored. (#145466,
341 Arjan van de Ven, Alex Larsson)
343 * glib/gmem.h: Add the new attribute to g_realloc and
346 2005-12-03 Matthias Clasen <mclasen@redhat.com>
349 * glib/gcache.h: Deprecate g_cache_value_foreach. (#322956,
352 * glib/gmappedfile.c: Make mapping of empty files
355 * glib/gfileutils.c: Don't fork a new process just to
356 fix the permissions of the created temp file. (#321318,
357 Alexis S. L. Carvalho)
359 2005-12-02 Matthias Clasen <mclasen@redhat.com>
361 * README.in: Add a note about Solaris threads.
364 * configure.in: Remove support for Solaris threads.
365 (#136971, Sebastian Wilhelmi, patch by Andrew Paprocki)
367 2005-12-02 Matthias Clasen <mclasen@redhat.com>
369 * configure.in: Define G_GUINT64_CONSTANT in analogy to
370 G_GINT64_CONSTANT. (#322568, Andrew Paprocki)
372 2005-12-02 Matthias Clasen <mclasen@redhat.com>
374 * glib/gslice.c: Win32 portability fixes and C99ism removal,
375 pointed out by Kazuki Iwamoto. (#323052)
377 2005-12-02 Matthias Clasen <mclasen@redhat.com>
379 * glib/gatomic.c (g_atomic_int_add): Add a missing volatile for
380 the IA64 implementation.
382 Fri Dec 2 16:18:09 2005 Tim Janik <timj@imendio.com>
384 * glib/gslice.c: some naming and type size fixups.
386 Fri Dec 2 13:08:58 2005 Tim Janik <timj@gtk.org>
388 * glib/gslice.h (g_slice_free): reworked GCC-specific type-safe macro
389 variant into something less verbose, but digestible for gcc-3.4.
391 Fri Dec 2 10:55:07 2005 Tim Janik <timj@imendio.com>
393 * tests/slice-test.c: extended to perform the benchmarking on the old
394 memchunk code if 'O' is selected.
396 * tests/memchunks.c: new file which contains the old GLib mem chunks
397 implementation with prefix old_mem_chunk_.
399 * tests/Makefile.am: added memchunks.c
401 Fri Dec 2 00:16:59 2005 Tim Janik <timj@imendio.com>
403 * glib/gslice.c: improved gettimeofday() timer resolution by maintaining
404 acceptable load distribution of the syscall.
405 (allocator_get_magazine_threshold): reduce minimum/single-thread magazine
406 sizes to approximately page_size/2.5 to avoid excessive startup allocations.
408 2005-12-01 Matthias Clasen <mclasen@redhat.com>
410 * tests/gobject/ifacecheck.c (test_iface_base_init): Don't
411 double-free base interface members.
413 * tests/Makefile.am: Remove duplicate variable.
415 Thu Dec 1 17:32:46 2005 Tim Janik <timj@imendio.com>
417 * glib/gslice.[hc]: new slice allocator implementation.
419 * tests/slice-test.c: added random slice allocation test.
421 * glib/gthread.[hc]: removed newly added private thread mem API.
423 * glib/gthreadinit.h:
426 * glib/gmem.c: divided glib threading initialisation into three phases,
427 initialisation where private keys and messaging are not available (only
428 needed by gmem.c), initialisation without messaging but private keys
429 available (gslice.c, gmessage.c), and full fledged initialisers that
430 server the rest of glib. initialisation functions got renamed to reflect
431 the limitations of their corresponding phases.
433 * glib/gmem.c: removed memchunk code, defer allocations to
436 * glib/gmem.[hc]: removed g_slice_* skeletons.
438 * glib/glib.symbols: added g_slice_* symbols.
440 * configure.in: check for availability of posix_memalign(3), memalign(3)
443 * glib/Makefile.am: added gslice.[hc].
445 2005-12-01 Tor Lillqvist <tml@novell.com>
447 * glib/gstdio.c (g_stat): In the Win32 implementation, strip
448 trailing slash(es) for non-root folders. stat() fails if non-root
449 folders are specified with trailing slashes. It's too much hassle
450 to demand that callers strip such slashes themselves, especially
451 as it is easy to get it wrong and strip the slash of a root
453 (g_rename): On NT-based Windows, use MoveFileEx() with
454 MOVEFILE_REPLACE_EXISTING to better match Unix behaviour.
456 2005-11-28 Matthias Clasen <mclasen@redhat.com>
458 Fix G_STMT_START / G_STMT_END on Solaris. (#321972,
461 * configure.in: Check whether do { } while (0) works.
463 * glib/gmacros.h: Use do { } while (0) for G_STMT_START /
464 G_STMT_END if it works.
466 2005-11-28 Matthias Clasen <mclasen@redhat.com>
468 * glib/gthread.c (g_static_rw_lock_wait, g_static_rw_lock_signal):
469 * glib/gnode.c (g_node_depth_traverse_level):
470 * glib/gmem.c (g_allocator_new):
471 * glib/ghash.c (g_hash_table_unref, g_hash_table_destroy)
472 (g_hash_table_foreach_remove_or_steal): Silence compiler
475 2005-11-27 Matthias Clasen <mclasen@redhat.com>
477 * glib/gunicollate.c (g_utf8_collate_key_for_filename):
478 Don't read beyond len. (#322520, Christian Persch)
480 Wed Nov 23 17:34:01 2005 Tim Janik <timj@imendio.com>
482 * glib/gdataset.c: access datalist flags via atomic pointer access
483 functions, instead of acquiring the dataset lock. this is faster and
484 also matches the atomic pointer readouts in gdatalistprivate.h and
485 direct pointer modifications required by gobject.c.
487 Wed Nov 23 13:35:31 2005 Tim Janik <timj@gtk.org>
489 * glib/gdataset.c: streamlined initialization code somewhat,
490 removed GData node cache.
492 2005-11-22 Matthias Clasen <mclasen@redhat.com>
494 * glib/ghash.c (g_hash_table_ref):
495 (g_hash_table_unref): Mark these as new API.
497 Avoid double locking in g_intern_string (#322133,
500 * glib/gdataset.c (g_quark_from_string_internal): New
501 internal function which factors out common parts of
502 g_quark_from[_static]_string.
503 (g_quark_from_string, g_quark_from_static_string):
504 Use g_quark_from_string_internal.
505 (g_intern_string, g_intern_static_string): Use
506 g_quark_from_string_internal, and only take the
508 (g_quark_new): Don't store the strings shifted by -1
509 in the g_quarks array.
510 (g_quark_to_string): Adapt to the previous change.
512 Tue Nov 22 14:04:26 2005 Tim Janik <timj@imendio.com>
516 g_hash_table_new_full(): create hash tables with a ref count of 1.
517 g_hash_table_ref(): atomically ref_count+=1
518 g_hash_table_unref(): atomically ref_count-=1, destroys hash table
519 when refcount reaches 0.
520 g_hash_table_destroy(): just destroy keys and values, unref by 1.
521 g_hash_table_insert():
522 g_hash_table_replace(): assert ref_count>0.
525 * glib/gatomic.c: added 'volatile' qualifier to all atomic pointer and
528 2005-11-20 Behdad Esfahbod <behdad@gnome.org>
530 * glib/guniprop.c (g_unichar_get_mirror_char): Remove unused
531 variables. (#321984, Andrew Paprocki)
533 2005-11-18 Matthias Clasen <mclasen@redhat.com>
535 * configure.in: Bump version
537 2005-11-17 Matthias Clasen <mclasen@redhat.com>
539 * === Released 2.9.0 ===
541 * glib/glib.symbols: Updates
543 * glib/gutf8.c (g_utf8_offset_to_pointer): Add some performance
544 hints to the docs. (Paolo Borelli)
548 * glib/gthread.c (g_thread_foreach): Mark as new api.
550 * README.in: Updates.
554 * glib/gdate.c (g_date_set_time_t):
555 (g_date_set_time_val): Add functions to set a date from
556 a time_t and from a GTimeVal.
557 (g_date_set_time): Deprecate. (#314067, Roger Leigh)
560 * tests/utf8-pointer.c: Unit tests for g_utf8_pointer_to_offset
561 and g_utf8_offset_to_pointer.
563 * glib/gutf8.c (g_utf8_pointer_to_offset)
564 (g_utf8_offset_to_pointer): Handle negative offsets, and use
565 "stutter stepping" for going backwards. (#320638, Larry
570 * glib/gthread.c: const correctness fixes, found
571 by Arjan van de Ven and gcc.
573 2005-11-16 Behdad Esfahbod <behdad@gnome.org>
575 * docs/reference/glib/tmpl/unicode.sgml: Correct typo on mentioning
576 @G_UNICODE_BREAK_UNKNOWN that should be #G_UNICODE_BREAK_UNKNOWN.
578 2005-11-09 Behdad Esfahbod <behdad@gnome.org>
580 * glib/guniprop.c: Use bit hacks instead when checking a general
581 category value against multiple values.
583 * glib/gutf8.c: Change ISO10646 to Unicode in docs.
585 2005-11-10 Simos Xenitellis <simos@gnome.org>
587 * configure.in: Added tt (Tatar) to ALL_LINGUAS.
589 2005-11-05 Tor Lillqvist <tml@novell.com>
591 * glib/giowin32.c (g_io_win32_set_flags): Don't call g_warning().
592 (#320688) Instead set error and return failure.
594 2005-11-07 Matthias Clasen <mclasen@redhat.com>
596 * glib/gmain.c (g_main_dispatch): Don't call cb_funcs->unref
597 while holding the context lock. (#320886, Andy Wingo)
599 2005-11-07 Billy Biggs <vektor@dumbterm.net>
601 * tests/refcount/Makefile.am: Add a missing reference to libglib
602 in the LDADD for this test.
604 2005-11-07 Matthias Clasen <mclasen@redhat.com>
606 * glib/gmem.c (g_allocator_new): We need to set n_preallocs to a
607 nonzero value, otherwise GTK+ 2.8 breaks when compiled against
608 GLib 2.9. (#320755, Luca Ferretti)
610 2005-11-04 Matthias Clasen <mclasen@redhat.com>
612 * glib/Makefile.am: Apply a patch from Behdad Esfahbod to
613 use a faster lookup table for g_unichar_get_mirror_char().
614 * glib/gmirroringtable.h: The new table.
615 * glib/gunichartables.h: Remove bidi_mirroring_table.
616 * glib/gen-unicode-tables.pl: Don't generate the mirroring
618 * glib/glib-mirroring-tab/*: Sources for the program
619 which generated gmirroringtable.h.
621 * glib/glist.c: Avoid some code duplication.
623 * glib/gscanner.h: Include gdataset.h. (#320322)
625 * glib/gdebug.h: add new GDebugFlag for fatal_criticals
626 * glib/gmessages.c: (_g_debug_init): handle G_DEBUG=fatal_criticals,
627 to help find critical warnings in applications. (#320017,
630 2005-11-02 Tor Lillqvist <tml@novell.com>
632 * glib/glib.symbols: Remove large amount of trailing whitespace
633 from one line. Remove #ifdef G_UNDEFINED from around
634 g_hash_table_new and g_hash_table_new_full symbols, what was the
635 purpose of that? Add the g_slice functions.
637 Tue Nov 1 16:24:20 2005 Tim Janik <timj@imendio.com>
639 * glib/gmem.[hc]: prepared deprecation of GMemChunk and GAllocator.
640 added g_slice_*() API to allocate and cache small bits of memory.
641 an actuall allocator implementation for g_slice_*() is still pending.
643 * glib/gthread.[hc]: changes from a patch by Matthias Clasen.
644 changed GRealThread list to use in-structure *next; fields instead
645 of GSList, in order for thread iteration to not depenend on g_slice_*()
647 _g_thread_mem_private_get():
648 _g_thread_mem_private_set(): added accessors for private memory,
649 needed because the ordinary GPrivate implementation relies on GArray
650 and GSList and therefore indirectly on working g_slice_*() allocations.
653 g_thread_foreach(): new public API function to loop over all existing threads.
667 removed GAllocator and free list usages and accompanying locks.
668 use g_slice_*() API to allocate and cache small bits of memory.
670 * glib/ghook.h: removed GMemChunk field from public API.
673 * glib/glist.h: deprecate allocator API, provide _free1() for consistency.
675 * glib/gnode.h: deprecate allocator API.
677 * glib/gmain.c: reordered GPollRec fields so g_slice_free_chain() can
678 be used for poll rec lists.
680 * glib/grel.c: removed mem chunk usage, and allocated tuples via g_slice_*().
681 g_relation_destroy(): free all tuples from the all_tuples hash table,
682 this effectively maintains the life time track keeping of tuples.
683 g_relation_delete_tuple(): free tuples which are removed from the
684 all_tuples hash table. this fixes a temporary leak that was present
685 in the memchunk code until the destruction of the relation.
687 2005-10-29 Matthias Clasen <mclasen@redhat.com>
689 * tests/convert-test.c: Add some tests for conversions between
690 UTF-8, UCS-4 and UTF-16.
692 * glib/gutf8.c (g_utf8_to_ucs4, g_utf8_to_utf16): Fix handling
693 of len == -1, noticed by Morten Welinder.
695 2005-10-27 Erdal Ronahi <erdal.ronahi@gmail.com>
697 * configure.in: Added ku (Kurdish) to ALL_LINGUAS
699 2005-10-26 Matthias Clasen <mclasen@redhat.com>
701 * glib/gutf8.c (g_ucs4_to_utf8): Don't set items_read twice
702 in the error case, and add some documentation. (#319806, Morten
705 2005-10-19 Manish Singh <yosh@gimp.org>
707 * configure.in: Use AC_CHECK_FUNCS for _NSGetEnviron, to get the
708 config.h symbol automatically. Fixes bug #313731.
710 2005-10-19 Tor Lillqvist <tml@novell.com>
712 * glib/gwin32.c (g_win32_get_package_installation_directory):
713 Return a g_strdup()ed copy of the value stored in the hash table,
714 so that it can be g_free()d without leaving a dangling pointer in
715 the hash table. (#319232)
717 2005-10-06 Matthias Clasen <mclasen@redhat.com>
719 * glib/gunicollate.c (g_utf8_collate_key_for_filename): Handle
720 all-zero sequences correctly. (#317930, Sebastien Bacher)
722 2005-10-05 Matthias Clasen <mclasen@redhat.com>
724 * glib/gmarkup.c (g_markup_escape_text): Doc addition
726 * Makefile.am: Add Collation tests.
728 * tests/collate/*: Inputs and expected outputs for collation tests.
730 * tests/run-collate-tests.sh: Script to run collation tests.
732 * tests/unicode-collate.c (main): Rework slightly to make
733 it usable in unit tests. Also test g_utf8_collate_key_for_filename().
735 2005-10-01 Behdad Esfahbod <behdad@gnome.org>
737 * docs/reference/glib/tmpl/unicode.sgml:
738 * glib/gen-unicode-tables.pl:
740 * glib/gunichartables.h:
742 * tests/casefold.txt:
743 * tests/casemap.txt: Updated to Unicode 4.1. There are five new
744 GUnicodeBreakType types. That may break some applications, like
747 2005-09-26 Matthias Clasen <mclasen@redhat.com>
749 * glib/gstrfuncs.c (g_ascii_strtoull): Add details to the
750 docs. (#314393, Matthew F. Barnes)
753 * glib/gprintf.h: Remove g_snprintf() and g_vsnprintf(), since
754 they are already declared in glib.h. This doesn't break documented
755 use of gprintf.h, but should probably be pointed out in the
756 release notes for 2.10. (#314232, Behdad Esfahbod)
758 Tue Sep 20 13:16:04 2005 Tim Janik <timj@imendio.com>
760 * glib/gpattern.c (g_pattern_ph_match): applied significant recursion
761 complexity optimization, based on a patch from Matthias Clasen.
763 * tests/patterntest.c: more tests, mostly from matthias.
765 2005-09-20 Matthias Clasen <mclasen@redhat.com>
767 * glib/gqueue.c (g_queue_insert_sorted): Correct the docs.
768 (#316703, Mark Drago)
770 Mon Sep 19 17:23:23 2005 Tim Janik <timj@imendio.com>
772 * glib/gpattern.c: applied a patch from matthias which checks on the
773 upper bound of GPatternSpec length to optimize matches.
776 * tests/patterntest.c: added more match cases.
778 2005-09-16 Tor Lillqvist <tml@novell.com>
780 * glib/gstrfuncs.c (g_ascii_strcasecmp, g_ascii_strncasecmp): Add
781 warning to doc comment that these functions should not be used on
782 encodings like CP932.
784 2005-09-14 Matthias Clasen <mclasen@redhat.com>
786 * tests/keyfile-test.c: Add a test for grup names of length 1.
788 * glib/gkeyfile.c (g_key_file_line_is_group): Accept group names
789 of length 1. (#316309)
791 2005-09-12 Matthias Clasen <mclasen@redhat.com>
793 * glib/gmarkup.c (g_markup_escape_text): Clarify docs.
795 2005-09-11 Sebastian Wilhelmi <seppi@seppi.de>
797 * tests/refcount/Makefile.am (INCLUDES): Link the the refcount
798 tests to the system thread library $(G_THREAD_LIBS). Fixes #313744
801 2005-09-11 Kjartan Maraas <kmaraas@gnome.org>
803 * glib/gmain.c: (g_child_watch_prepare), (g_child_watch_check),
804 (child_watch_helper_thread): Remove some dead code. Closes
807 2005-09-07 Tor Lillqvist <tml@novell.com>
809 * glib/Makefile.am: Create also a console version of the
810 gspawn-win32-helper program, gspawn-win32-helper-console.exe.
811 It's otherwise identical to gspawn-win32-helper.exe, except marked
812 as a console application (linked without the -mwindows option).
814 * glib/gspawn-win32.c (do_spawn_directly, do_spawn_with_pipes):
815 Drop the dont_wait parameter. Its truth value correlated 100% with
816 the NULLness of the exit_status parameter anyway, so it's enough
817 to check whether exit_status is NULL. Invert the sense of the
818 dont_return_handle parameter and rename it to do_return_handle, to
819 make the code easier to read by avoiding double negations.
821 (g_spawn_sync_utf8, g_spawn_async_with_pipes_utf8): Modify calls
822 to do_spawn_with_pipes() accordingly.
824 (do_spawn_with_pipes): If we have a console, use the console
825 version of the helper program, otherwise use the GUI one. This
826 avoids extra console windows opening up in some situations. (In
827 case a console application uses the GUI gspawn-win32-helper.exe to
828 spawn another console application we would get a separate console
829 for the spawned console application).
831 * glib-zip.in: Distribute also gspawn-win32-helper-console.exe.
833 2005-09-05 Matthias Clasen <mclasen@redhat.com>
835 * glib/gmappedfile.c (g_mapped_file_new): Report an error
836 if the file is too large. (#315275, Kjartan Maraas)
838 * glib/gkeyfile.c (g_key_file_load_from_fd): The return value
839 of read() is signed. (#315273, Kjartan Maraas)
841 2005-08-31 Tor Lillqvist <tml@novell.com>
843 * glib/gutils.h: Wrapping atexit() is a bad idea on Windows, where
844 the EXE and each DLL have their own atexit function chains.
846 #define g_atexit as atexit instead. This means it has a
847 better chance of doing what the caller wants. For instance,
848 gtkhtml calls g_atexit() registering a function in gtkhtml
849 itself. This caused a crash when g_atexit() was implemented as a
850 function in the GLib DLL. The gtkhtml DLL was already unloaded by
851 the time the GLib DLL got unloaded.
853 * glib/gutils.c: #undef the #define mentioned above, to also get a
854 real g_atexit() into the DLL for backward compatibility. Document
855 the Windows behaviour of g_atexit(), and document the varying ways
856 atexit() can behave in the context of dynamically loaded modules
859 2005-08-31 Matthias Clasen <mclasen@redhat.com>
863 * glib/gdataset.c: Add string interning functions.
865 2005-08-28 Matthias Clasen <mclasen@redhat.com>
867 * glib/giochannel.c: Unify some near-duplicate strings. (#314654,
870 2005-08-26 Matthias Clasen <mclasen@redhat.com>
872 * configure.in: Bump version to 2.9.0
874 2005-08-25 Tor Lillqvist <tml@novell.com>
876 Make also the g_spawn*() functions take parameters in the GLib
877 file name encoding, i.e. UTF-8, on Windows. Has no impact on Unix
878 API or ABI. Like the other GLib API that was earlier changed to
879 use UTF-8 on Windows, the names of the functions that take UTF-8
880 have _utf8 suffixes added by using preprocessor macros in the
881 header file. The old names are kept for functions with the old
882 behaviour, taking parameters in the system codepage, for DLL ABI
885 * glib/gspawn.h: On Win32 add the suffix _utf8 to the names of the
886 g_spawn*() functions.
888 * glib/gspawn-win32.c: Use wide-char API on NT-based
889 Windows. Convert parameters from UTF-8 to wide chars (NT) or
890 system codepage (Win9x) and call the C library _wspawn*() or
891 spawn*() functions respectvely. Add DLL ABI stability versions
892 that take parameters in the system codepage.
894 * glib/gspawn-win32-helper.c: On NT-based Windows use the
895 wide-char versions of argv and envp, and use wide-char API to
896 change directory and spawn the program to run. Remove the verbose
897 debugging output, it was too complex to modify for the wide-char
898 features. (Just add temporary debugging printouts if needed, no
899 need to have them permanently in the source.)
901 * glib/gspawn.c: Corresponding documentation updates.
903 * glib/glib.symbols: Corresponding changes: Mark the ABI stability
904 symbols as PRIVATE, add the new _utf8-suffixed ones.
906 2005-08-24 Stepan Kasal <kasal@ucw.cz>
908 * glib/gtypes.h (G_MININT64): Cast the constant to gint64; it is
909 guint64 otherwise and that can produce warnings about comparison
910 between signed and unsigned.
912 2005-08-23 Matthias Clasen <mclasen@redhat.com>
914 * glib/gutils.c: Fix the crt_externs.h include.
916 2005-08-23 Stepan Kasal <kasal@ucw.cz>
918 * NEWS: Fix spelling of my first name.
920 2005-08-23 Matthias Clasen <mclasen@redhat.com>
924 * === Released 2.8.1 ===
928 2005-08-20 Hans Breuer <hans@breuer.org>
930 * glib/makefile.msc.in : link with ws2_32.lib
932 2005-08-18 Tor Lillqvist <tml@novell.com>
934 * configure.in: Check for <sys/wait.h>
936 * glib/gbacktrace.c: Include <sys/wait.h> on if HAVE_SYS_WAIT_H.
938 2005-08-18 Ross Burton <ross@burtonini.com>
941 Optimise single-character insertions.
946 * tests/string-test.c:
947 Add tests for new optimisation, and fix a leak.
949 2005-08-17 Matthias Clasen <mclasen@redhat.com>
951 * configure.in: Check for crt_externs.h and _NSGetEnviron.
953 * glib/gutils.c: On Darwin, include crt-externs.h and
954 define environ using _NSGetEnviron(). (#313731)
956 2005-08-16 Stepan Kasal <kasal@ucw.cz>
958 * glib/gutils.c (g_get_any_init): Move the body of the big if...
959 (g_get_any_init_do): ... to this new function.
960 (g_get_any_init): Declare as inline.
961 (g_get_any_init_locked): New inline function, does the locking.
962 Make use of these two throughout the code.
964 2005-08-15 Matthias Clasen <mclasen@redhat.com>
966 * glib/gbacktrace.c (g_on_error_stack_trace): Wait for
967 the child process and then simply return. This makes
968 The "S" option work as documented in g_on_error_query().
969 (#313125, Matthew F. Barnes)
971 * glib/gunicode.h: Update the link to Unicode category
972 values. (#313369, Behnam Esfahbod)
974 * glib/gqueue.c (g_queue_find_custom): Clarify docs
975 a little. (#311727, Tristan van Berkom)
977 * glib/abicheck.sh, gobject/abicheck.sh: Make the
978 check work on ia64 too, where some symbols ended up
979 in yet another section.