1 2006-01-16 Matthias Clasen <mclasen@redhat.com>
3 * glib/gasyncqueue.c (g_async_queue_push_sorted_unlocked):
4 Signal waiting threads, problem noticed by Christian Kellner.
6 2006-01-16 Matthias Clasen <mclasen@redhat.com>
8 Fix bug 326747, Alberto Ruiz:
10 * glib/gconvert.c (make_valid_utf8): Change this function to
11 replace unknown characters by the Unicode replacement character
12 instead of '?', and don't append "(invalid encoding)".
13 (g_filename_display_name, g_filename_display_basename): Document
14 how to determine if the filename was in an invalid encoding.
16 2006-01-14 Matthias Clasen <mclasen@redhat.com>
18 * glib/gtree.c: Replace the simple recursive implementation
19 by a nonrecursive, threaded implementation by Maurizio
22 2006-01-12 Matthias Clasen <mclasen@redhat.com>
24 * tests/asyncqueue-test.c (main): Fix the
25 assert to not always trigger. (#326558,
28 2006-01-11 Matthias Clasen <mclasen@redhat.com>
30 * tests/convert-test.c: Don't test items_read and
31 items_written in error cases where it is not set.
33 * tests/asyncqueue-test.c: Set error to NULL.
35 2006-01-08 Matthias Clasen <mclasen@redhat.com>
37 * glib/gtree.c: Break some long lines.
39 * tests/tree-test.c: Add more tests.
41 * glib/gslice.c: Include config.h first, pointed out
44 2005-12-20 Sven Herzberg <herzi@gnome-de.org>
46 * docs/reference/gobject/tmpl/gtype.sgml: explain that
47 G_TYPE_INSTANCE_GET_CLASS() does behave different during
50 2006-01-06 Matthias Clasen <mclasen@redhat.com>
52 * glib/gslice.c: Only define _XOPEN_SOURCE if we know
53 that we have posix_memalign(). (#323937, Bogdan Nicula)
55 2006-01-05 Matthias Clasen <mclasen@redhat.com>
57 * configure.in: Bump version
59 * === Released 2.9.2 ===
61 2006-01-05 Hans Breuer <hans@breuer.org>
63 * glib/makefile.msc.in : added gslice.obj
65 2006-01-05 Matthias Clasen <mclasen@redhat.com>
69 * glib/gmain.c: Clarify the documentation of source ids
70 a bit. (#325874, Dan Williams)
72 * configure.in: Fix another strict aliasing problem.
74 * glib/gthreadpool.c: Avoid use of varargs macro. (#325864,
77 2006-01-04 Manish Singh <yosh@gimp.org>
80 * glib/gstdio.h: don't macroized g_access, g_chdir, and g_unlink
81 either, since they have the same issue as g_rmdir. (Related to
84 Wed Jan 4 13:33:25 2006 Tim Janik <timj@gtk.org>
86 * glib/gslice.c (magazine_cache_pop_magazine): don't reverse chunk
87 order when creating magazines, so we hand out chunks with ascending
90 2006-01-04 Abel Cheung <maddog@linuxhall.org>
92 * configure.in: Added "ml" "zh_HK" to ALL_LINGUAS.
94 2006-01-03 Martyn Russell <martyn@imendio.com>
96 * docs/reference/glib/glib-sections.txt:
98 * glib/gthreadpool.[ch]:
99 - Added new API g_thread_pool_get_idle_time() and
100 g_thread_pool_set_idle_time(). (#324228).
102 * tests/threadpool-test.c:
103 - Updated test case to do thread pool sorting, thread pool with
104 no sorting and a thread pool with idle thread timeouts.
106 2006-01-03 Matthias Clasen <mclasen@redhat.com>
108 * glib/gmain.h: Add new functions here, too.
110 * glib/glib.symbols: Add new functions.
112 2005-12-20 Michael Meeks <michael.meeks@novell.com>
114 * glib/gmain.c (g_main_context_is_owner): new method
115 to determine if the current thread is the owner of the
118 2006-01-02 Matthias Clasen <mclasen@redhat.com>
122 * glib/gstdio.c (g_rmdir): Don't provide g_rmdir() as a macro
123 expanding to rmdir, since rmdir is not declared in a portable
124 system header we can include in gstdio.h. (#325249, Jani Monoses)
126 2006-01-01 Tor Lillqvist <tml@novell.com>
128 * glib/gspawn-win32.c (g_spawn_sync_utf8): Set the GIOChannels for
129 stdout and stderr to unbuffered. Otherwise the giochannel layer
130 will try to read from them regardless whether the
131 g_io_channel_win32_poll() call here has indicated
132 readability or not. (#325310)
134 2005-12-29 Matthias Clasen <mclasen@redhat.com>
136 * glib/gutils.c (glib_check_version): Fix a copy-and-paste error
137 in the docs. (#325273, Declan Naughton)
139 2005-12-27 Manish Singh <yosh@gimp.org>
141 * tests/run-collate-tests.sh: use LC_ALL instead of LC_COLLATE, to
142 make sure we really override things.
144 2005-12-27 Matthias Clasen <mclasen@redhat.com>
146 Fix #316221, Michal Benes, Stanislav Brabec;
148 * configure.in: Fix a strict aliasing problem in
149 g_static_mutex_get_mutex().
150 * glib/gthread.h: ...and in
151 g_static_mutex_get_mutex_impl_shortcut().
153 * glib/gdatasetprivate.h: Add a cast to silence compiler
154 warnings. (#321978, Andrew Paprocki)
156 Partial fix for bug #323937, Bogdan Nicula.
158 * configure.in: Check for malloc.h
159 * glib/gslice.c: Don't include malloc.h unconditionally.
161 2005-12-27 Manish Singh <yosh@gimp.org>
163 * tests/run-collate-tests.sh: export LC_COLLATE so it takes.
164 (#324950, Dan Yefimov)
166 2005-12-26 Matthias Clasen <mclasen@redhat.com>
168 * glib/gslice.c: On Win32, include process.h (#325015, Kazuki
171 2005-12-25 Matthias Clasen <mclasen@redhat.com>
173 * AUTHORS: Update my email
175 * tests/slice-test.c: Fix C99isms. (#324950, Dan Yefimov)
177 2005-12-21 Matthias Clasen <mclasen@redhat.com>
179 * README.in: Add some notes on when 'make check' may fail.
181 2005-12-20 Matthias Clasen <mclasen@redhat.com>
184 * glib/gthreadpool.h:
185 * glib/gthreadpool.c (g_thread_pool_set_sort_function): New function
186 to sort tasks pushed into a threadpool. (#324479, Martyn Russell)
188 * tests/threadpool-test.c: Test this.
190 Tue Dec 20 18:14:14 2005 Tim Janik <timj@imendio.com>
192 * glib/gslice.[hc]: added mem_error() and mem_assert() to test and
193 handle errors without depending on gmessage.c which might not be
194 setup when the error occours.
195 removed G_SLICE_CONFIG_ALWAYS_FREE config option, fixed the code so
196 always freeing can be achieved by adjusting the working set time to
197 0 with G_SLICE_CONFIG_WORKING_SET_MSECS.
198 added G_SLICE_CONFIG_COLOR_INCREMENT to test different color increments
199 (mainly 0 and 1). reduced the minimum block size to 128 bytes, to
200 minimize wastage if small amounts of differently sized structrues are
201 allocated, this does come at a performance cost of roughly 5% though.
202 fixed up block alignment calculation, so it works for varying
203 block sizes. only use strerror() not g_strerror() since the latter
204 depends on working GQuark and GSlice.
205 mem_error(): implemented in terms of fprintf and vfprintf.
207 * tests/slice-color.c: new program to test cache colorization effects.
209 * tests/slice-test.c: trade G_SLICE_CONFIG_ALWAYS_FREE for 0 duration
210 G_SLICE_CONFIG_WORKING_SET_MSECS.
212 2005-12-17 Matthias Clasen <mclasen@redhat.com>
214 * glib/goption.c (parse_short_option): Set an error in all
215 failure cases. (#324332, Tim-Philipp Müller)
217 2005-12-17 Sebastian Wilhelmi <seppi@seppi.de>
219 * glib/gatomic.c: Fix memory barrier position in g_atomic_int_get
220 and g_atomic_pointer_get. Add g_atomic_int_set and
221 g_atomic_pointer_set implementations for the !DEFINE_WITH_MUTEXES &&
222 G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case, as well as defining them
223 as functions (additionally to the macros in the header) for the
224 !G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case.
226 2005-12-16 Matthias Clasen <mclasen@redhat.com>
228 * glib/gmem.c (g_allocator_new): Don't return a pointer to
229 a const struct, since apps expect to be able to modify it.
230 (#324179, J. Ali Harlow)
232 Tue Dec 13 10:13:32 2005 Tim Janik <timj@imendio.com>
234 * glib/gatomic.h: added g_atomic_pointer_set() and g_atomic_int_set()
235 for systems where the initialization of atomic variables requires a
236 write memory barrier.
238 2005-12-09 Matthias Clasen <mclasen@redhat.com>
240 * configure.in: Bump version
242 * === Released 2.9.1 ===
244 2005-12-09 Alexander Larsson <alexl@redhat.com>
248 Use G_GNUC_WARN_UNUSED_RESULT on list functions that return
251 2005-12-08 Matthias Clasen <mclasen@redhat.com>
255 2005-12-07 Martyn Russell <martyn@imendio.com>
257 * glib/gasyncqueue.c:
258 - Call g_queue_insert_sorted() instead of duplicating the code.
259 - Call g_queue_sort() instead of duplicating the code.
260 - Invert sort function results to make sure the same sort function
261 gives the same results across glist, gslist, gqueue and
264 * tests/asyncqueue-test.c:
265 - Updated the sort function to reflect the example in the
266 documentation for gasyncqueue.c.
268 2005-12-07 Martyn Russell <martyn@imendio.com>
270 * docs/reference/glib/glib-sections.txt:
271 * docs/reference/glib/tmpl/linked_lists_double.sgml:
272 * docs/reference/glib/tmpl/linked_lists_single.sgml:
275 - Added g_list_insert_sorted_with_data () and
276 g_slist_insert_sorted_with_data ().
277 - Removed the extra check in g_list_sort() and g_slist_sort() for
278 GCompareDataFunc vs. GCompareFunc.
280 2005-12-07 Tor Lillqvist <tml@novell.com>
282 * glib-gettextize.in: Look up prefix at run-time on Win32,
283 assuming the standard directory structure with glib-gettextize in
286 2005-12-06 Matthias Clasen <mclasen@redhat.com>
288 * glib/gmem.h: Exempt GTK+ from the mem chunk deprecation, since
289 we need GTK+ 2.8 to compile against GLib 2.10.
291 2005-12-05 Manish Singh <yosh@gimp.org>
293 * tests/run-collate-test.sh: set LC_COLLATE, not LANG, to be sure
294 to override any user settings.
296 2005-12-05 Matthias Clasen <mclasen@redhat.com>
298 * configure.in: Revert an accidential commit
300 2005-12-05 Matthias Clasen <mclasen@redhat.com>
302 * glib/gutf8.c: Documentation updates. (#323291, Morten Welinder)
304 * tests/Makefile.am (TESTS_ENVIRONMENT):
305 * tests/gobject/Makefile.am (TESTS_ENVIRONMENT): Set
306 MALLOC_CHECK_ and MALLOC_PERTURB_
308 * tests/run-collate-tests.sh: Run the collation tests explicitly
309 in en_US locale. (#320463)
311 * glib/gmem.h: Really deprecate GMemChunk.
318 * glib/gstrfuncs.c: Add versioned deprecation docs.
320 2005-12-05 Martyn Russell <martyn@imendio.com>
322 * docs/reference/glib/glib-sections.txt:
323 * glib/gasyncqueue.[ch]:
324 - Added support for sorting async queues by with _push_sorted(),
325 _push_sorted_unlocked(), _sort() and _sort_unlocked() (#323047).
328 * tests/asyncqueue-test.c:
329 - Added test case for gasyncqueue.c
331 Mon Dec 5 15:53:20 2005 Tim Janik <timj@imendio.com>
333 * glib/gslice.c: implement chain walking for arbitrary ->next pointer
334 offsets in g_slice_free_chain_with_offset() based on a patch by behdad
335 in bug 323178. moved time consuming logic from g_slice_free() out of
336 the inner loop, so g_slice_free_chain_with_offset() provides a real
337 performance benefit over g_slice_free1() now.
339 * glib/gslice.h: renamed g_slice_free_chain() to
340 g_slice_free_chain_with_offset(). implemented g_slice_free_chain() as
341 a type-safe macro as suggested in bug 323178.
342 simplified the macro implementation of g_slice_free() and implemented
343 it in a type safe manner for all compliers as suggested by Morten
344 Welinder <mortenw@gnome.org>.
349 * glib/glib.symbols: s/g_slice_free_chain/g_slice_free_chain_with_offset/
351 2005-12-05 Matthias Clasen <mclasen@redhat.com>
353 * glib/gasyncqueue.c: Add some docs.
355 * tests/libmoduletestplugin_a.c: Fix compiler warnings.
357 * glib/gatomic.c: In the ia64 implementation, use
358 __sync builtin without _si or _di suffix. (#321229,
359 Stanislav Brabec, patch by Andreas Schwab)
361 2005-12-04 Behdad Esfahbod <behdad@gnome.org>
363 * glib/gslice.h: Remove comma at the end of enum.
365 2005-12-04 Matthias Clasen <mclasen@redhat.com>
367 Handle multiple user names with the same UID better.
368 (#319535, Laszlo Peter)
370 * glib/gutils.c (g_get_any_init_do): When determining user
371 data, first look up $LOGNAME. If the UID doesn't match
372 getuid(), fall back to the current behaviour of looking
373 up the user data based on getuid().
375 2005-12-04 Matthias Clasen <mclasen@redhat.com>
377 * glib/gmacros.h (G_GNUC_WARN_UNUSED_RESULT): Add a macro
378 to make gcc warn if a function result is ignored. (#145466,
379 Arjan van de Ven, Alex Larsson)
381 * glib/gmem.h: Add the new attribute to g_realloc and
384 2005-12-03 Matthias Clasen <mclasen@redhat.com>
387 * glib/gcache.h: Deprecate g_cache_value_foreach. (#322956,
390 * glib/gmappedfile.c: Make mapping of empty files
393 * glib/gfileutils.c: Don't fork a new process just to
394 fix the permissions of the created temp file. (#321318,
395 Alexis S. L. Carvalho)
397 2005-12-02 Matthias Clasen <mclasen@redhat.com>
399 * README.in: Add a note about Solaris threads.
402 * configure.in: Remove support for Solaris threads.
403 (#136971, Sebastian Wilhelmi, patch by Andrew Paprocki)
405 2005-12-02 Matthias Clasen <mclasen@redhat.com>
407 * configure.in: Define G_GUINT64_CONSTANT in analogy to
408 G_GINT64_CONSTANT. (#322568, Andrew Paprocki)
410 2005-12-02 Matthias Clasen <mclasen@redhat.com>
412 * glib/gslice.c: Win32 portability fixes and C99ism removal,
413 pointed out by Kazuki Iwamoto. (#323052)
415 2005-12-02 Matthias Clasen <mclasen@redhat.com>
417 * glib/gatomic.c (g_atomic_int_add): Add a missing volatile for
418 the IA64 implementation.
420 Fri Dec 2 16:18:09 2005 Tim Janik <timj@imendio.com>
422 * glib/gslice.c: some naming and type size fixups.
424 Fri Dec 2 13:08:58 2005 Tim Janik <timj@gtk.org>
426 * glib/gslice.h (g_slice_free): reworked GCC-specific type-safe macro
427 variant into something less verbose, but digestible for gcc-3.4.
429 Fri Dec 2 10:55:07 2005 Tim Janik <timj@imendio.com>
431 * tests/slice-test.c: extended to perform the benchmarking on the old
432 memchunk code if 'O' is selected.
434 * tests/memchunks.c: new file which contains the old GLib mem chunks
435 implementation with prefix old_mem_chunk_.
437 * tests/Makefile.am: added memchunks.c
439 Fri Dec 2 00:16:59 2005 Tim Janik <timj@imendio.com>
441 * glib/gslice.c: improved gettimeofday() timer resolution by maintaining
442 acceptable load distribution of the syscall.
443 (allocator_get_magazine_threshold): reduce minimum/single-thread magazine
444 sizes to approximately page_size/2.5 to avoid excessive startup allocations.
446 2005-12-01 Matthias Clasen <mclasen@redhat.com>
448 * tests/gobject/ifacecheck.c (test_iface_base_init): Don't
449 double-free base interface members.
451 * tests/Makefile.am: Remove duplicate variable.
453 Thu Dec 1 17:32:46 2005 Tim Janik <timj@imendio.com>
455 * glib/gslice.[hc]: new slice allocator implementation.
457 * tests/slice-test.c: added random slice allocation test.
459 * glib/gthread.[hc]: removed newly added private thread mem API.
461 * glib/gthreadinit.h:
464 * glib/gmem.c: divided glib threading initialisation into three phases,
465 initialisation where private keys and messaging are not available (only
466 needed by gmem.c), initialisation without messaging but private keys
467 available (gslice.c, gmessage.c), and full fledged initialisers that
468 server the rest of glib. initialisation functions got renamed to reflect
469 the limitations of their corresponding phases.
471 * glib/gmem.c: removed memchunk code, defer allocations to
474 * glib/gmem.[hc]: removed g_slice_* skeletons.
476 * glib/glib.symbols: added g_slice_* symbols.
478 * configure.in: check for availability of posix_memalign(3), memalign(3)
481 * glib/Makefile.am: added gslice.[hc].
483 2005-12-01 Tor Lillqvist <tml@novell.com>
485 * glib/gstdio.c (g_stat): In the Win32 implementation, strip
486 trailing slash(es) for non-root folders. stat() fails if non-root
487 folders are specified with trailing slashes. It's too much hassle
488 to demand that callers strip such slashes themselves, especially
489 as it is easy to get it wrong and strip the slash of a root
491 (g_rename): On NT-based Windows, use MoveFileEx() with
492 MOVEFILE_REPLACE_EXISTING to better match Unix behaviour.
494 2005-11-28 Matthias Clasen <mclasen@redhat.com>
496 Fix G_STMT_START / G_STMT_END on Solaris. (#321972,
499 * configure.in: Check whether do { } while (0) works.
501 * glib/gmacros.h: Use do { } while (0) for G_STMT_START /
502 G_STMT_END if it works.
504 2005-11-28 Matthias Clasen <mclasen@redhat.com>
506 * glib/gthread.c (g_static_rw_lock_wait, g_static_rw_lock_signal):
507 * glib/gnode.c (g_node_depth_traverse_level):
508 * glib/gmem.c (g_allocator_new):
509 * glib/ghash.c (g_hash_table_unref, g_hash_table_destroy)
510 (g_hash_table_foreach_remove_or_steal): Silence compiler
513 2005-11-27 Matthias Clasen <mclasen@redhat.com>
515 * glib/gunicollate.c (g_utf8_collate_key_for_filename):
516 Don't read beyond len. (#322520, Christian Persch)
518 Wed Nov 23 17:34:01 2005 Tim Janik <timj@imendio.com>
520 * glib/gdataset.c: access datalist flags via atomic pointer access
521 functions, instead of acquiring the dataset lock. this is faster and
522 also matches the atomic pointer readouts in gdatalistprivate.h and
523 direct pointer modifications required by gobject.c.
525 Wed Nov 23 13:35:31 2005 Tim Janik <timj@gtk.org>
527 * glib/gdataset.c: streamlined initialization code somewhat,
528 removed GData node cache.
530 2005-11-22 Matthias Clasen <mclasen@redhat.com>
532 * glib/ghash.c (g_hash_table_ref):
533 (g_hash_table_unref): Mark these as new API.
535 Avoid double locking in g_intern_string (#322133,
538 * glib/gdataset.c (g_quark_from_string_internal): New
539 internal function which factors out common parts of
540 g_quark_from[_static]_string.
541 (g_quark_from_string, g_quark_from_static_string):
542 Use g_quark_from_string_internal.
543 (g_intern_string, g_intern_static_string): Use
544 g_quark_from_string_internal, and only take the
546 (g_quark_new): Don't store the strings shifted by -1
547 in the g_quarks array.
548 (g_quark_to_string): Adapt to the previous change.
550 Tue Nov 22 14:04:26 2005 Tim Janik <timj@imendio.com>
554 g_hash_table_new_full(): create hash tables with a ref count of 1.
555 g_hash_table_ref(): atomically ref_count+=1
556 g_hash_table_unref(): atomically ref_count-=1, destroys hash table
557 when refcount reaches 0.
558 g_hash_table_destroy(): just destroy keys and values, unref by 1.
559 g_hash_table_insert():
560 g_hash_table_replace(): assert ref_count>0.
563 * glib/gatomic.c: added 'volatile' qualifier to all atomic pointer and
566 2005-11-20 Behdad Esfahbod <behdad@gnome.org>
568 * glib/guniprop.c (g_unichar_get_mirror_char): Remove unused
569 variables. (#321984, Andrew Paprocki)
571 2005-11-18 Matthias Clasen <mclasen@redhat.com>
573 * configure.in: Bump version
575 2005-11-17 Matthias Clasen <mclasen@redhat.com>
577 * === Released 2.9.0 ===
579 * glib/glib.symbols: Updates
581 * glib/gutf8.c (g_utf8_offset_to_pointer): Add some performance
582 hints to the docs. (Paolo Borelli)
586 * glib/gthread.c (g_thread_foreach): Mark as new api.
588 * README.in: Updates.
592 * glib/gdate.c (g_date_set_time_t):
593 (g_date_set_time_val): Add functions to set a date from
594 a time_t and from a GTimeVal.
595 (g_date_set_time): Deprecate. (#314067, Roger Leigh)
598 * tests/utf8-pointer.c: Unit tests for g_utf8_pointer_to_offset
599 and g_utf8_offset_to_pointer.
601 * glib/gutf8.c (g_utf8_pointer_to_offset)
602 (g_utf8_offset_to_pointer): Handle negative offsets, and use
603 "stutter stepping" for going backwards. (#320638, Larry
608 * glib/gthread.c: const correctness fixes, found
609 by Arjan van de Ven and gcc.
611 2005-11-16 Behdad Esfahbod <behdad@gnome.org>
613 * docs/reference/glib/tmpl/unicode.sgml: Correct typo on mentioning
614 @G_UNICODE_BREAK_UNKNOWN that should be #G_UNICODE_BREAK_UNKNOWN.
616 2005-11-09 Behdad Esfahbod <behdad@gnome.org>
618 * glib/guniprop.c: Use bit hacks instead when checking a general
619 category value against multiple values.
621 * glib/gutf8.c: Change ISO10646 to Unicode in docs.
623 2005-11-10 Simos Xenitellis <simos@gnome.org>
625 * configure.in: Added tt (Tatar) to ALL_LINGUAS.
627 2005-11-05 Tor Lillqvist <tml@novell.com>
629 * glib/giowin32.c (g_io_win32_set_flags): Don't call g_warning().
630 (#320688) Instead set error and return failure.
632 2005-11-07 Matthias Clasen <mclasen@redhat.com>
634 * glib/gmain.c (g_main_dispatch): Don't call cb_funcs->unref
635 while holding the context lock. (#320886, Andy Wingo)
637 2005-11-07 Billy Biggs <vektor@dumbterm.net>
639 * tests/refcount/Makefile.am: Add a missing reference to libglib
640 in the LDADD for this test.
642 2005-11-07 Matthias Clasen <mclasen@redhat.com>
644 * glib/gmem.c (g_allocator_new): We need to set n_preallocs to a
645 nonzero value, otherwise GTK+ 2.8 breaks when compiled against
646 GLib 2.9. (#320755, Luca Ferretti)
648 2005-11-04 Matthias Clasen <mclasen@redhat.com>
650 * glib/Makefile.am: Apply a patch from Behdad Esfahbod to
651 use a faster lookup table for g_unichar_get_mirror_char().
652 * glib/gmirroringtable.h: The new table.
653 * glib/gunichartables.h: Remove bidi_mirroring_table.
654 * glib/gen-unicode-tables.pl: Don't generate the mirroring
656 * glib/glib-mirroring-tab/*: Sources for the program
657 which generated gmirroringtable.h.
659 * glib/glist.c: Avoid some code duplication.
661 * glib/gscanner.h: Include gdataset.h. (#320322)
663 * glib/gdebug.h: add new GDebugFlag for fatal_criticals
664 * glib/gmessages.c: (_g_debug_init): handle G_DEBUG=fatal_criticals,
665 to help find critical warnings in applications. (#320017,
668 2005-11-02 Tor Lillqvist <tml@novell.com>
670 * glib/glib.symbols: Remove large amount of trailing whitespace
671 from one line. Remove #ifdef G_UNDEFINED from around
672 g_hash_table_new and g_hash_table_new_full symbols, what was the
673 purpose of that? Add the g_slice functions.
675 Tue Nov 1 16:24:20 2005 Tim Janik <timj@imendio.com>
677 * glib/gmem.[hc]: prepared deprecation of GMemChunk and GAllocator.
678 added g_slice_*() API to allocate and cache small bits of memory.
679 an actuall allocator implementation for g_slice_*() is still pending.
681 * glib/gthread.[hc]: changes from a patch by Matthias Clasen.
682 changed GRealThread list to use in-structure *next; fields instead
683 of GSList, in order for thread iteration to not depenend on g_slice_*()
685 _g_thread_mem_private_get():
686 _g_thread_mem_private_set(): added accessors for private memory,
687 needed because the ordinary GPrivate implementation relies on GArray
688 and GSList and therefore indirectly on working g_slice_*() allocations.
691 g_thread_foreach(): new public API function to loop over all existing threads.
705 removed GAllocator and free list usages and accompanying locks.
706 use g_slice_*() API to allocate and cache small bits of memory.
708 * glib/ghook.h: removed GMemChunk field from public API.
711 * glib/glist.h: deprecate allocator API, provide _free1() for consistency.
713 * glib/gnode.h: deprecate allocator API.
715 * glib/gmain.c: reordered GPollRec fields so g_slice_free_chain() can
716 be used for poll rec lists.
718 * glib/grel.c: removed mem chunk usage, and allocated tuples via g_slice_*().
719 g_relation_destroy(): free all tuples from the all_tuples hash table,
720 this effectively maintains the life time track keeping of tuples.
721 g_relation_delete_tuple(): free tuples which are removed from the
722 all_tuples hash table. this fixes a temporary leak that was present
723 in the memchunk code until the destruction of the relation.
725 2005-10-29 Matthias Clasen <mclasen@redhat.com>
727 * tests/convert-test.c: Add some tests for conversions between
728 UTF-8, UCS-4 and UTF-16.
730 * glib/gutf8.c (g_utf8_to_ucs4, g_utf8_to_utf16): Fix handling
731 of len == -1, noticed by Morten Welinder.
733 2005-10-27 Erdal Ronahi <erdal.ronahi@gmail.com>
735 * configure.in: Added ku (Kurdish) to ALL_LINGUAS
737 2005-10-26 Matthias Clasen <mclasen@redhat.com>
739 * glib/gutf8.c (g_ucs4_to_utf8): Don't set items_read twice
740 in the error case, and add some documentation. (#319806, Morten
743 2005-10-19 Manish Singh <yosh@gimp.org>
745 * configure.in: Use AC_CHECK_FUNCS for _NSGetEnviron, to get the
746 config.h symbol automatically. Fixes bug #313731.
748 2005-10-19 Tor Lillqvist <tml@novell.com>
750 * glib/gwin32.c (g_win32_get_package_installation_directory):
751 Return a g_strdup()ed copy of the value stored in the hash table,
752 so that it can be g_free()d without leaving a dangling pointer in
753 the hash table. (#319232)
755 2005-10-06 Matthias Clasen <mclasen@redhat.com>
757 * glib/gunicollate.c (g_utf8_collate_key_for_filename): Handle
758 all-zero sequences correctly. (#317930, Sebastien Bacher)
760 2005-10-05 Matthias Clasen <mclasen@redhat.com>
762 * glib/gmarkup.c (g_markup_escape_text): Doc addition
764 * Makefile.am: Add Collation tests.
766 * tests/collate/*: Inputs and expected outputs for collation tests.
768 * tests/run-collate-tests.sh: Script to run collation tests.
770 * tests/unicode-collate.c (main): Rework slightly to make
771 it usable in unit tests. Also test g_utf8_collate_key_for_filename().
773 2005-10-01 Behdad Esfahbod <behdad@gnome.org>
775 * docs/reference/glib/tmpl/unicode.sgml:
776 * glib/gen-unicode-tables.pl:
778 * glib/gunichartables.h:
780 * tests/casefold.txt:
781 * tests/casemap.txt: Updated to Unicode 4.1. There are five new
782 GUnicodeBreakType types. That may break some applications, like
785 2005-09-26 Matthias Clasen <mclasen@redhat.com>
787 * glib/gstrfuncs.c (g_ascii_strtoull): Add details to the
788 docs. (#314393, Matthew F. Barnes)
791 * glib/gprintf.h: Remove g_snprintf() and g_vsnprintf(), since
792 they are already declared in glib.h. This doesn't break documented
793 use of gprintf.h, but should probably be pointed out in the
794 release notes for 2.10. (#314232, Behdad Esfahbod)
796 Tue Sep 20 13:16:04 2005 Tim Janik <timj@imendio.com>
798 * glib/gpattern.c (g_pattern_ph_match): applied significant recursion
799 complexity optimization, based on a patch from Matthias Clasen.
801 * tests/patterntest.c: more tests, mostly from matthias.
803 2005-09-20 Matthias Clasen <mclasen@redhat.com>
805 * glib/gqueue.c (g_queue_insert_sorted): Correct the docs.
806 (#316703, Mark Drago)
808 Mon Sep 19 17:23:23 2005 Tim Janik <timj@imendio.com>
810 * glib/gpattern.c: applied a patch from matthias which checks on the
811 upper bound of GPatternSpec length to optimize matches.
814 * tests/patterntest.c: added more match cases.
816 2005-09-16 Tor Lillqvist <tml@novell.com>
818 * glib/gstrfuncs.c (g_ascii_strcasecmp, g_ascii_strncasecmp): Add
819 warning to doc comment that these functions should not be used on
820 encodings like CP932.
822 2005-09-14 Matthias Clasen <mclasen@redhat.com>
824 * tests/keyfile-test.c: Add a test for grup names of length 1.
826 * glib/gkeyfile.c (g_key_file_line_is_group): Accept group names
827 of length 1. (#316309)
829 2005-09-12 Matthias Clasen <mclasen@redhat.com>
831 * glib/gmarkup.c (g_markup_escape_text): Clarify docs.
833 2005-09-11 Sebastian Wilhelmi <seppi@seppi.de>
835 * tests/refcount/Makefile.am (INCLUDES): Link the the refcount
836 tests to the system thread library $(G_THREAD_LIBS). Fixes #313744
839 2005-09-11 Kjartan Maraas <kmaraas@gnome.org>
841 * glib/gmain.c: (g_child_watch_prepare), (g_child_watch_check),
842 (child_watch_helper_thread): Remove some dead code. Closes
845 2005-09-07 Tor Lillqvist <tml@novell.com>
847 * glib/Makefile.am: Create also a console version of the
848 gspawn-win32-helper program, gspawn-win32-helper-console.exe.
849 It's otherwise identical to gspawn-win32-helper.exe, except marked
850 as a console application (linked without the -mwindows option).
852 * glib/gspawn-win32.c (do_spawn_directly, do_spawn_with_pipes):
853 Drop the dont_wait parameter. Its truth value correlated 100% with
854 the NULLness of the exit_status parameter anyway, so it's enough
855 to check whether exit_status is NULL. Invert the sense of the
856 dont_return_handle parameter and rename it to do_return_handle, to
857 make the code easier to read by avoiding double negations.
859 (g_spawn_sync_utf8, g_spawn_async_with_pipes_utf8): Modify calls
860 to do_spawn_with_pipes() accordingly.
862 (do_spawn_with_pipes): If we have a console, use the console
863 version of the helper program, otherwise use the GUI one. This
864 avoids extra console windows opening up in some situations. (In
865 case a console application uses the GUI gspawn-win32-helper.exe to
866 spawn another console application we would get a separate console
867 for the spawned console application).
869 * glib-zip.in: Distribute also gspawn-win32-helper-console.exe.
871 2005-09-05 Matthias Clasen <mclasen@redhat.com>
873 * glib/gmappedfile.c (g_mapped_file_new): Report an error
874 if the file is too large. (#315275, Kjartan Maraas)
876 * glib/gkeyfile.c (g_key_file_load_from_fd): The return value
877 of read() is signed. (#315273, Kjartan Maraas)
879 2005-08-31 Tor Lillqvist <tml@novell.com>
881 * glib/gutils.h: Wrapping atexit() is a bad idea on Windows, where
882 the EXE and each DLL have their own atexit function chains.
884 #define g_atexit as atexit instead. This means it has a
885 better chance of doing what the caller wants. For instance,
886 gtkhtml calls g_atexit() registering a function in gtkhtml
887 itself. This caused a crash when g_atexit() was implemented as a
888 function in the GLib DLL. The gtkhtml DLL was already unloaded by
889 the time the GLib DLL got unloaded.
891 * glib/gutils.c: #undef the #define mentioned above, to also get a
892 real g_atexit() into the DLL for backward compatibility. Document
893 the Windows behaviour of g_atexit(), and document the varying ways
894 atexit() can behave in the context of dynamically loaded modules
897 2005-08-31 Matthias Clasen <mclasen@redhat.com>
901 * glib/gdataset.c: Add string interning functions.
903 2005-08-28 Matthias Clasen <mclasen@redhat.com>
905 * glib/giochannel.c: Unify some near-duplicate strings. (#314654,
908 2005-08-26 Matthias Clasen <mclasen@redhat.com>
910 * configure.in: Bump version to 2.9.0
912 2005-08-25 Tor Lillqvist <tml@novell.com>
914 Make also the g_spawn*() functions take parameters in the GLib
915 file name encoding, i.e. UTF-8, on Windows. Has no impact on Unix
916 API or ABI. Like the other GLib API that was earlier changed to
917 use UTF-8 on Windows, the names of the functions that take UTF-8
918 have _utf8 suffixes added by using preprocessor macros in the
919 header file. The old names are kept for functions with the old
920 behaviour, taking parameters in the system codepage, for DLL ABI
923 * glib/gspawn.h: On Win32 add the suffix _utf8 to the names of the
924 g_spawn*() functions.
926 * glib/gspawn-win32.c: Use wide-char API on NT-based
927 Windows. Convert parameters from UTF-8 to wide chars (NT) or
928 system codepage (Win9x) and call the C library _wspawn*() or
929 spawn*() functions respectvely. Add DLL ABI stability versions
930 that take parameters in the system codepage.
932 * glib/gspawn-win32-helper.c: On NT-based Windows use the
933 wide-char versions of argv and envp, and use wide-char API to
934 change directory and spawn the program to run. Remove the verbose
935 debugging output, it was too complex to modify for the wide-char
936 features. (Just add temporary debugging printouts if needed, no
937 need to have them permanently in the source.)
939 * glib/gspawn.c: Corresponding documentation updates.
941 * glib/glib.symbols: Corresponding changes: Mark the ABI stability
942 symbols as PRIVATE, add the new _utf8-suffixed ones.
944 2005-08-24 Stepan Kasal <kasal@ucw.cz>
946 * glib/gtypes.h (G_MININT64): Cast the constant to gint64; it is
947 guint64 otherwise and that can produce warnings about comparison
948 between signed and unsigned.
950 2005-08-23 Matthias Clasen <mclasen@redhat.com>
952 * glib/gutils.c: Fix the crt_externs.h include.
954 2005-08-23 Stepan Kasal <kasal@ucw.cz>
956 * NEWS: Fix spelling of my first name.
958 2005-08-23 Matthias Clasen <mclasen@redhat.com>
962 * === Released 2.8.1 ===
966 2005-08-20 Hans Breuer <hans@breuer.org>
968 * glib/makefile.msc.in : link with ws2_32.lib
970 2005-08-18 Tor Lillqvist <tml@novell.com>
972 * configure.in: Check for <sys/wait.h>
974 * glib/gbacktrace.c: Include <sys/wait.h> on if HAVE_SYS_WAIT_H.
976 2005-08-18 Ross Burton <ross@burtonini.com>
979 Optimise single-character insertions.
984 * tests/string-test.c:
985 Add tests for new optimisation, and fix a leak.
987 2005-08-17 Matthias Clasen <mclasen@redhat.com>
989 * configure.in: Check for crt_externs.h and _NSGetEnviron.
991 * glib/gutils.c: On Darwin, include crt-externs.h and
992 define environ using _NSGetEnviron(). (#313731)
994 2005-08-16 Stepan Kasal <kasal@ucw.cz>
996 * glib/gutils.c (g_get_any_init): Move the body of the big if...
997 (g_get_any_init_do): ... to this new function.
998 (g_get_any_init): Declare as inline.
999 (g_get_any_init_locked): New inline function, does the locking.
1000 Make use of these two throughout the code.
1002 2005-08-15 Matthias Clasen <mclasen@redhat.com>
1004 * glib/gbacktrace.c (g_on_error_stack_trace): Wait for
1005 the child process and then simply return. This makes
1006 The "S" option work as documented in g_on_error_query().
1007 (#313125, Matthew F. Barnes)
1009 * glib/gunicode.h: Update the link to Unicode category
1010 values. (#313369, Behnam Esfahbod)
1012 * glib/gqueue.c (g_queue_find_custom): Clarify docs
1013 a little. (#311727, Tristan van Berkom)
1015 * glib/abicheck.sh, gobject/abicheck.sh: Make the
1016 check work on ia64 too, where some symbols ended up
1017 in yet another section.