1 2006-01-03 Martyn Russell <martyn@imendio.com>
3 * docs/reference/glib/glib-sections.txt:
5 * glib/gthreadpool.[ch]:
6 - Added new API g_thread_pool_get_idle_time() and
7 g_thread_pool_set_idle_time(). (#324228).
9 * tests/threadpool-test.c:
10 - Updated test case to do thread pool sorting, thread pool with
11 no sorting and a thread pool with idle thread timeouts.
13 2006-01-03 Matthias Clasen <mclasen@redhat.com>
15 * glib/gmain.h: Add new functions here, too.
17 * glib/glib.symbols: Add new functions.
19 2005-12-20 Michael Meeks <michael.meeks@novell.com>
21 * glib/gmain.c (g_main_context_is_owner): new method
22 to determine if the current thread is the owner of the
25 2006-01-02 Matthias Clasen <mclasen@redhat.com>
29 * glib/gstdio.c (g_rmdir): Don't provide g_rmdir() as a macro
30 expanding to rmdir, since rmdir is not declared in a portable
31 system header we can include in gstdio.h. (#325249, Jani Monoses)
33 2006-01-01 Tor Lillqvist <tml@novell.com>
35 * glib/gspawn-win32.c (g_spawn_sync_utf8): Set the GIOChannels for
36 stdout and stderr to unbuffered. Otherwise the giochannel layer
37 will try to read from them regardless whether the
38 g_io_channel_win32_poll() call here has indicated
39 readability or not. (#325310)
41 2005-12-29 Matthias Clasen <mclasen@redhat.com>
43 * glib/gutils.c (glib_check_version): Fix a copy-and-paste error
44 in the docs. (#325273, Declan Naughton)
46 2005-12-27 Manish Singh <yosh@gimp.org>
48 * tests/run-collate-tests.sh: use LC_ALL instead of LC_COLLATE, to
49 make sure we really override things.
51 2005-12-27 Matthias Clasen <mclasen@redhat.com>
53 Fix #316221, Michal Benes, Stanislav Brabec;
55 * configure.in: Fix a strict aliasing problem in
56 g_static_mutex_get_mutex().
57 * glib/gthread.h: ...and in
58 g_static_mutex_get_mutex_impl_shortcut().
60 * glib/gdatasetprivate.h: Add a cast to silence compiler
61 warnings. (#321978, Andrew Paprocki)
63 Partial fix for bug #32937, Bogdan Nicula.
65 * configure.in: Check for malloc.h
66 * glib/gslice.c: Don't include malloc.h unconditionally.
68 2005-12-27 Manish Singh <yosh@gimp.org>
70 * tests/run-collate-tests.sh: export LC_COLLATE so it takes.
71 (#324950, Dan Yefimov)
73 2005-12-26 Matthias Clasen <mclasen@redhat.com>
75 * glib/gslice.c: On Win32, include process.h (#325015, Kazuki
78 2005-12-25 Matthias Clasen <mclasen@redhat.com>
80 * AUTHORS: Update my email
82 * tests/slice-test.c: Fix C99isms. (#324950, Dan Yefimov)
84 2005-12-21 Matthias Clasen <mclasen@redhat.com>
86 * README.in: Add some notes on when 'make check' may fail.
88 2005-12-20 Matthias Clasen <mclasen@redhat.com>
92 * glib/gthreadpool.c (g_thread_pool_set_sort_function): New function
93 to sort tasks pushed into a threadpool. (#324479, Martyn Russell)
95 * tests/threadpool-test.c: Test this.
97 Tue Dec 20 18:14:14 2005 Tim Janik <timj@imendio.com>
99 * glib/gslice.[hc]: added mem_error() and mem_assert() to test and
100 handle errors without depending on gmessage.c which might not be
101 setup when the error occours.
102 removed G_SLICE_CONFIG_ALWAYS_FREE config option, fixed the code so
103 always freeing can be achieved by adjusting the working set time to
104 0 with G_SLICE_CONFIG_WORKING_SET_MSECS.
105 added G_SLICE_CONFIG_COLOR_INCREMENT to test different color increments
106 (mainly 0 and 1). reduced the minimum block size to 128 bytes, to
107 minimize wastage if small amounts of differently sized structrues are
108 allocated, this does come at a performance cost of roughly 5% though.
109 fixed up block alignment calculation, so it works for varying
110 block sizes. only use strerror() not g_strerror() since the latter
111 depends on working GQuark and GSlice.
112 mem_error(): implemented in terms of fprintf and vfprintf.
114 * tests/slice-color.c: new program to test cache colorization effects.
116 * tests/slice-test.c: trade G_SLICE_CONFIG_ALWAYS_FREE for 0 duration
117 G_SLICE_CONFIG_WORKING_SET_MSECS.
119 2005-12-17 Matthias Clasen <mclasen@redhat.com>
121 * glib/goption.c (parse_short_option): Set an error in all
122 failure cases. (#324332, Tim-Philipp Müller)
124 2005-12-17 Sebastian Wilhelmi <seppi@seppi.de>
126 * glib/gatomic.c: Fix memory barrier position in g_atomic_int_get
127 and g_atomic_pointer_get. Add g_atomic_int_set and
128 g_atomic_pointer_set implementations for the !DEFINE_WITH_MUTEXES &&
129 G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case, as well as defining them
130 as functions (additionally to the macros in the header) for the
131 !G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case.
133 2005-12-16 Matthias Clasen <mclasen@redhat.com>
135 * glib/gmem.c (g_allocator_new): Don't return a pointer to
136 a const struct, since apps expect to be able to modify it.
137 (#324179, J. Ali Harlow)
139 Tue Dec 13 10:13:32 2005 Tim Janik <timj@imendio.com>
141 * glib/gatomic.h: added g_atomic_pointer_set() and g_atomic_int_set()
142 for systems where the initialization of atomic variables requires a
143 write memory barrier.
145 2005-12-09 Matthias Clasen <mclasen@redhat.com>
147 * configure.in: Bump version
149 * === Released 2.9.1 ===
151 2005-12-09 Alexander Larsson <alexl@redhat.com>
155 Use G_GNUC_WARN_UNUSED_RESULT on list functions that return
158 2005-12-08 Matthias Clasen <mclasen@redhat.com>
162 2005-12-07 Martyn Russell <martyn@imendio.com>
164 * glib/gasyncqueue.c:
165 - Call g_queue_insert_sorted() instead of duplicating the code.
166 - Call g_queue_sort() instead of duplicating the code.
167 - Invert sort function results to make sure the same sort function
168 gives the same results across glist, gslist, gqueue and
171 * tests/asyncqueue-test.c:
172 - Updated the sort function to reflect the example in the
173 documentation for gasyncqueue.c.
175 2005-12-07 Martyn Russell <martyn@imendio.com>
177 * docs/reference/glib/glib-sections.txt:
178 * docs/reference/glib/tmpl/linked_lists_double.sgml:
179 * docs/reference/glib/tmpl/linked_lists_single.sgml:
182 - Added g_list_insert_sorted_with_data () and
183 g_slist_insert_sorted_with_data ().
184 - Removed the extra check in g_list_sort() and g_slist_sort() for
185 GCompareDataFunc vs. GCompareFunc.
187 2005-12-07 Tor Lillqvist <tml@novell.com>
189 * glib-gettextize.in: Look up prefix at run-time on Win32,
190 assuming the standard directory structure with glib-gettextize in
193 2005-12-06 Matthias Clasen <mclasen@redhat.com>
195 * glib/gmem.h: Exempt GTK+ from the mem chunk deprecation, since
196 we need GTK+ 2.8 to compile against GLib 2.10.
198 2005-12-05 Manish Singh <yosh@gimp.org>
200 * tests/run-collate-test.sh: set LC_COLLATE, not LANG, to be sure
201 to override any user settings.
203 2005-12-05 Matthias Clasen <mclasen@redhat.com>
205 * configure.in: Revert an accidential commit
207 2005-12-05 Matthias Clasen <mclasen@redhat.com>
209 * glib/gutf8.c: Documentation updates. (#323291, Morten Welinder)
211 * tests/Makefile.am (TESTS_ENVIRONMENT):
212 * tests/gobject/Makefile.am (TESTS_ENVIRONMENT): Set
213 MALLOC_CHECK_ and MALLOC_PERTURB_
215 * tests/run-collate-tests.sh: Run the collation tests explicitly
216 in en_US locale. (#320463)
218 * glib/gmem.h: Really deprecate GMemChunk.
225 * glib/gstrfuncs.c: Add versioned deprecation docs.
227 2005-12-05 Martyn Russell <martyn@imendio.com>
229 * docs/reference/glib/glib-sections.txt:
230 * glib/gasyncqueue.[ch]:
231 - Added support for sorting async queues by with _push_sorted(),
232 _push_sorted_unlocked(), _sort() and _sort_unlocked() (#323047).
235 * tests/asyncqueue-test.c:
236 - Added test case for gasyncqueue.c
238 Mon Dec 5 15:53:20 2005 Tim Janik <timj@imendio.com>
240 * glib/gslice.c: implement chain walking for arbitrary ->next pointer
241 offsets in g_slice_free_chain_with_offset() based on a patch by behdad
242 in bug 323178. moved time consuming logic from g_slice_free() out of
243 the inner loop, so g_slice_free_chain_with_offset() provides a real
244 performance benefit over g_slice_free1() now.
246 * glib/gslice.h: renamed g_slice_free_chain() to
247 g_slice_free_chain_with_offset(). implemented g_slice_free_chain() as
248 a type-safe macro as suggested in bug 323178.
249 simplified the macro implementation of g_slice_free() and implemented
250 it in a type safe manner for all compliers as suggested by Morten
251 Welinder <mortenw@gnome.org>.
256 * glib/glib.symbols: s/g_slice_free_chain/g_slice_free_chain_with_offset/
258 2005-12-05 Matthias Clasen <mclasen@redhat.com>
260 * glib/gasyncqueue.c: Add some docs.
262 * tests/libmoduletestplugin_a.c: Fix compiler warnings.
264 * glib/gatomic.c: In the ia64 implementation, use
265 __sync builtin without _si or _di suffix. (#321229,
266 Stanislav Brabec, patch by Andreas Schwab)
268 2005-12-04 Behdad Esfahbod <behdad@gnome.org>
270 * glib/gslice.h: Remove comma at the end of enum.
272 2005-12-04 Matthias Clasen <mclasen@redhat.com>
274 Handle multiple user names with the same UID better.
275 (#319535, Laszlo Peter)
277 * glib/gutils.c (g_get_any_init_do): When determining user
278 data, first look up $LOGNAME. If the UID doesn't match
279 getuid(), fall back to the current behaviour of looking
280 up the user data based on getuid().
282 2005-12-04 Matthias Clasen <mclasen@redhat.com>
284 * glib/gmacros.h (G_GNUC_WARN_UNUSED_RESULT): Add a macro
285 to make gcc warn if a function result is ignored. (#145466,
286 Arjan van de Ven, Alex Larsson)
288 * glib/gmem.h: Add the new attribute to g_realloc and
291 2005-12-03 Matthias Clasen <mclasen@redhat.com>
294 * glib/gcache.h: Deprecate g_cache_value_foreach. (#322956,
297 * glib/gmappedfile.c: Make mapping of empty files
300 * glib/gfileutils.c: Don't fork a new process just to
301 fix the permissions of the created temp file. (#321318,
302 Alexis S. L. Carvalho)
304 2005-12-02 Matthias Clasen <mclasen@redhat.com>
306 * README.in: Add a note about Solaris threads.
309 * configure.in: Remove support for Solaris threads.
310 (#136971, Sebastian Wilhelmi, patch by Andrew Paprocki)
312 2005-12-02 Matthias Clasen <mclasen@redhat.com>
314 * configure.in: Define G_GUINT64_CONSTANT in analogy to
315 G_GINT64_CONSTANT. (#322568, Andrew Paprocki)
317 2005-12-02 Matthias Clasen <mclasen@redhat.com>
319 * glib/gslice.c: Win32 portability fixes and C99ism removal,
320 pointed out by Kazuki Iwamoto. (#323052)
322 2005-12-02 Matthias Clasen <mclasen@redhat.com>
324 * glib/gatomic.c (g_atomic_int_add): Add a missing volatile for
325 the IA64 implementation.
327 Fri Dec 2 16:18:09 2005 Tim Janik <timj@imendio.com>
329 * glib/gslice.c: some naming and type size fixups.
331 Fri Dec 2 13:08:58 2005 Tim Janik <timj@gtk.org>
333 * glib/gslice.h (g_slice_free): reworked GCC-specific type-safe macro
334 variant into something less verbose, but digestible for gcc-3.4.
336 Fri Dec 2 10:55:07 2005 Tim Janik <timj@imendio.com>
338 * tests/slice-test.c: extended to perform the benchmarking on the old
339 memchunk code if 'O' is selected.
341 * tests/memchunks.c: new file which contains the old GLib mem chunks
342 implementation with prefix old_mem_chunk_.
344 * tests/Makefile.am: added memchunks.c
346 Fri Dec 2 00:16:59 2005 Tim Janik <timj@imendio.com>
348 * glib/gslice.c: improved gettimeofday() timer resolution by maintaining
349 acceptable load distribution of the syscall.
350 (allocator_get_magazine_threshold): reduce minimum/single-thread magazine
351 sizes to approximately page_size/2.5 to avoid excessive startup allocations.
353 2005-12-01 Matthias Clasen <mclasen@redhat.com>
355 * tests/gobject/ifacecheck.c (test_iface_base_init): Don't
356 double-free base interface members.
358 * tests/Makefile.am: Remove duplicate variable.
360 Thu Dec 1 17:32:46 2005 Tim Janik <timj@imendio.com>
362 * glib/gslice.[hc]: new slice allocator implementation.
364 * tests/slice-test.c: added random slice allocation test.
366 * glib/gthread.[hc]: removed newly added private thread mem API.
368 * glib/gthreadinit.h:
371 * glib/gmem.c: divided glib threading initialisation into three phases,
372 initialisation where private keys and messaging are not available (only
373 needed by gmem.c), initialisation without messaging but private keys
374 available (gslice.c, gmessage.c), and full fledged initialisers that
375 server the rest of glib. initialisation functions got renamed to reflect
376 the limitations of their corresponding phases.
378 * glib/gmem.c: removed memchunk code, defer allocations to
381 * glib/gmem.[hc]: removed g_slice_* skeletons.
383 * glib/glib.symbols: added g_slice_* symbols.
385 * configure.in: check for availability of posix_memalign(3), memalign(3)
388 * glib/Makefile.am: added gslice.[hc].
390 2005-12-01 Tor Lillqvist <tml@novell.com>
392 * glib/gstdio.c (g_stat): In the Win32 implementation, strip
393 trailing slash(es) for non-root folders. stat() fails if non-root
394 folders are specified with trailing slashes. It's too much hassle
395 to demand that callers strip such slashes themselves, especially
396 as it is easy to get it wrong and strip the slash of a root
398 (g_rename): On NT-based Windows, use MoveFileEx() with
399 MOVEFILE_REPLACE_EXISTING to better match Unix behaviour.
401 2005-11-28 Matthias Clasen <mclasen@redhat.com>
403 Fix G_STMT_START / G_STMT_END on Solaris. (#321972,
406 * configure.in: Check whether do { } while (0) works.
408 * glib/gmacros.h: Use do { } while (0) for G_STMT_START /
409 G_STMT_END if it works.
411 2005-11-28 Matthias Clasen <mclasen@redhat.com>
413 * glib/gthread.c (g_static_rw_lock_wait, g_static_rw_lock_signal):
414 * glib/gnode.c (g_node_depth_traverse_level):
415 * glib/gmem.c (g_allocator_new):
416 * glib/ghash.c (g_hash_table_unref, g_hash_table_destroy)
417 (g_hash_table_foreach_remove_or_steal): Silence compiler
420 2005-11-27 Matthias Clasen <mclasen@redhat.com>
422 * glib/gunicollate.c (g_utf8_collate_key_for_filename):
423 Don't read beyond len. (#322520, Christian Persch)
425 Wed Nov 23 17:34:01 2005 Tim Janik <timj@imendio.com>
427 * glib/gdataset.c: access datalist flags via atomic pointer access
428 functions, instead of acquiring the dataset lock. this is faster and
429 also matches the atomic pointer readouts in gdatalistprivate.h and
430 direct pointer modifications required by gobject.c.
432 Wed Nov 23 13:35:31 2005 Tim Janik <timj@gtk.org>
434 * glib/gdataset.c: streamlined initialization code somewhat,
435 removed GData node cache.
437 2005-11-22 Matthias Clasen <mclasen@redhat.com>
439 * glib/ghash.c (g_hash_table_ref):
440 (g_hash_table_unref): Mark these as new API.
442 Avoid double locking in g_intern_string (#322133,
445 * glib/gdataset.c (g_quark_from_string_internal): New
446 internal function which factors out common parts of
447 g_quark_from[_static]_string.
448 (g_quark_from_string, g_quark_from_static_string):
449 Use g_quark_from_string_internal.
450 (g_intern_string, g_intern_static_string): Use
451 g_quark_from_string_internal, and only take the
453 (g_quark_new): Don't store the strings shifted by -1
454 in the g_quarks array.
455 (g_quark_to_string): Adapt to the previous change.
457 Tue Nov 22 14:04:26 2005 Tim Janik <timj@imendio.com>
461 g_hash_table_new_full(): create hash tables with a ref count of 1.
462 g_hash_table_ref(): atomically ref_count+=1
463 g_hash_table_unref(): atomically ref_count-=1, destroys hash table
464 when refcount reaches 0.
465 g_hash_table_destroy(): just destroy keys and values, unref by 1.
466 g_hash_table_insert():
467 g_hash_table_replace(): assert ref_count>0.
470 * glib/gatomic.c: added 'volatile' qualifier to all atomic pointer and
473 2005-11-20 Behdad Esfahbod <behdad@gnome.org>
475 * glib/guniprop.c (g_unichar_get_mirror_char): Remove unused
476 variables. (#321984, Andrew Paprocki)
478 2005-11-18 Matthias Clasen <mclasen@redhat.com>
480 * configure.in: Bump version
482 2005-11-17 Matthias Clasen <mclasen@redhat.com>
484 * === Released 2.9.0 ===
486 * glib/glib.symbols: Updates
488 * glib/gutf8.c (g_utf8_offset_to_pointer): Add some performance
489 hints to the docs. (Paolo Borelli)
493 * glib/gthread.c (g_thread_foreach): Mark as new api.
495 * README.in: Updates.
499 * glib/gdate.c (g_date_set_time_t):
500 (g_date_set_time_val): Add functions to set a date from
501 a time_t and from a GTimeVal.
502 (g_date_set_time): Deprecate. (#314067, Roger Leigh)
505 * tests/utf8-pointer.c: Unit tests for g_utf8_pointer_to_offset
506 and g_utf8_offset_to_pointer.
508 * glib/gutf8.c (g_utf8_pointer_to_offset)
509 (g_utf8_offset_to_pointer): Handle negative offsets, and use
510 "stutter stepping" for going backwards. (#320638, Larry
515 * glib/gthread.c: const correctness fixes, found
516 by Arjan van de Ven and gcc.
518 2005-11-16 Behdad Esfahbod <behdad@gnome.org>
520 * docs/reference/glib/tmpl/unicode.sgml: Correct typo on mentioning
521 @G_UNICODE_BREAK_UNKNOWN that should be #G_UNICODE_BREAK_UNKNOWN.
523 2005-11-09 Behdad Esfahbod <behdad@gnome.org>
525 * glib/guniprop.c: Use bit hacks instead when checking a general
526 category value against multiple values.
528 * glib/gutf8.c: Change ISO10646 to Unicode in docs.
530 2005-11-10 Simos Xenitellis <simos@gnome.org>
532 * configure.in: Added tt (Tatar) to ALL_LINGUAS.
534 2005-11-05 Tor Lillqvist <tml@novell.com>
536 * glib/giowin32.c (g_io_win32_set_flags): Don't call g_warning().
537 (#320688) Instead set error and return failure.
539 2005-11-07 Matthias Clasen <mclasen@redhat.com>
541 * glib/gmain.c (g_main_dispatch): Don't call cb_funcs->unref
542 while holding the context lock. (#320886, Andy Wingo)
544 2005-11-07 Billy Biggs <vektor@dumbterm.net>
546 * tests/refcount/Makefile.am: Add a missing reference to libglib
547 in the LDADD for this test.
549 2005-11-07 Matthias Clasen <mclasen@redhat.com>
551 * glib/gmem.c (g_allocator_new): We need to set n_preallocs to a
552 nonzero value, otherwise GTK+ 2.8 breaks when compiled against
553 GLib 2.9. (#320755, Luca Ferretti)
555 2005-11-04 Matthias Clasen <mclasen@redhat.com>
557 * glib/Makefile.am: Apply a patch from Behdad Esfahbod to
558 use a faster lookup table for g_unichar_get_mirror_char().
559 * glib/gmirroringtable.h: The new table.
560 * glib/gunichartables.h: Remove bidi_mirroring_table.
561 * glib/gen-unicode-tables.pl: Don't generate the mirroring
563 * glib/glib-mirroring-tab/*: Sources for the program
564 which generated gmirroringtable.h.
566 * glib/glist.c: Avoid some code duplication.
568 * glib/gscanner.h: Include gdataset.h. (#320322)
570 * glib/gdebug.h: add new GDebugFlag for fatal_criticals
571 * glib/gmessages.c: (_g_debug_init): handle G_DEBUG=fatal_criticals,
572 to help find critical warnings in applications. (#320017,
575 2005-11-02 Tor Lillqvist <tml@novell.com>
577 * glib/glib.symbols: Remove large amount of trailing whitespace
578 from one line. Remove #ifdef G_UNDEFINED from around
579 g_hash_table_new and g_hash_table_new_full symbols, what was the
580 purpose of that? Add the g_slice functions.
582 Tue Nov 1 16:24:20 2005 Tim Janik <timj@imendio.com>
584 * glib/gmem.[hc]: prepared deprecation of GMemChunk and GAllocator.
585 added g_slice_*() API to allocate and cache small bits of memory.
586 an actuall allocator implementation for g_slice_*() is still pending.
588 * glib/gthread.[hc]: changes from a patch by Matthias Clasen.
589 changed GRealThread list to use in-structure *next; fields instead
590 of GSList, in order for thread iteration to not depenend on g_slice_*()
592 _g_thread_mem_private_get():
593 _g_thread_mem_private_set(): added accessors for private memory,
594 needed because the ordinary GPrivate implementation relies on GArray
595 and GSList and therefore indirectly on working g_slice_*() allocations.
598 g_thread_foreach(): new public API function to loop over all existing threads.
612 removed GAllocator and free list usages and accompanying locks.
613 use g_slice_*() API to allocate and cache small bits of memory.
615 * glib/ghook.h: removed GMemChunk field from public API.
618 * glib/glist.h: deprecate allocator API, provide _free1() for consistency.
620 * glib/gnode.h: deprecate allocator API.
622 * glib/gmain.c: reordered GPollRec fields so g_slice_free_chain() can
623 be used for poll rec lists.
625 * glib/grel.c: removed mem chunk usage, and allocated tuples via g_slice_*().
626 g_relation_destroy(): free all tuples from the all_tuples hash table,
627 this effectively maintains the life time track keeping of tuples.
628 g_relation_delete_tuple(): free tuples which are removed from the
629 all_tuples hash table. this fixes a temporary leak that was present
630 in the memchunk code until the destruction of the relation.
632 2005-10-29 Matthias Clasen <mclasen@redhat.com>
634 * tests/convert-test.c: Add some tests for conversions between
635 UTF-8, UCS-4 and UTF-16.
637 * glib/gutf8.c (g_utf8_to_ucs4, g_utf8_to_utf16): Fix handling
638 of len == -1, noticed by Morten Welinder.
640 2005-10-27 Erdal Ronahi <erdal.ronahi@gmail.com>
642 * configure.in: Added ku (Kurdish) to ALL_LINGUAS
644 2005-10-26 Matthias Clasen <mclasen@redhat.com>
646 * glib/gutf8.c (g_ucs4_to_utf8): Don't set items_read twice
647 in the error case, and add some documentation. (#319806, Morten
650 2005-10-19 Manish Singh <yosh@gimp.org>
652 * configure.in: Use AC_CHECK_FUNCS for _NSGetEnviron, to get the
653 config.h symbol automatically. Fixes bug #313731.
655 2005-10-19 Tor Lillqvist <tml@novell.com>
657 * glib/gwin32.c (g_win32_get_package_installation_directory):
658 Return a g_strdup()ed copy of the value stored in the hash table,
659 so that it can be g_free()d without leaving a dangling pointer in
660 the hash table. (#319232)
662 2005-10-06 Matthias Clasen <mclasen@redhat.com>
664 * glib/gunicollate.c (g_utf8_collate_key_for_filename): Handle
665 all-zero sequences correctly. (#317930, Sebastien Bacher)
667 2005-10-05 Matthias Clasen <mclasen@redhat.com>
669 * glib/gmarkup.c (g_markup_escape_text): Doc addition
671 * Makefile.am: Add Collation tests.
673 * tests/collate/*: Inputs and expected outputs for collation tests.
675 * tests/run-collate-tests.sh: Script to run collation tests.
677 * tests/unicode-collate.c (main): Rework slightly to make
678 it usable in unit tests. Also test g_utf8_collate_key_for_filename().
680 2005-10-01 Behdad Esfahbod <behdad@gnome.org>
682 * docs/reference/glib/tmpl/unicode.sgml:
683 * glib/gen-unicode-tables.pl:
685 * glib/gunichartables.h:
687 * tests/casefold.txt:
688 * tests/casemap.txt: Updated to Unicode 4.1. There are five new
689 GUnicodeBreakType types. That may break some applications, like
692 2005-09-26 Matthias Clasen <mclasen@redhat.com>
694 * glib/gstrfuncs.c (g_ascii_strtoull): Add details to the
695 docs. (#314393, Matthew F. Barnes)
698 * glib/gprintf.h: Remove g_snprintf() and g_vsnprintf(), since
699 they are already declared in glib.h. This doesn't break documented
700 use of gprintf.h, but should probably be pointed out in the
701 release notes for 2.10. (#314232, Behdad Esfahbod)
703 Tue Sep 20 13:16:04 2005 Tim Janik <timj@imendio.com>
705 * glib/gpattern.c (g_pattern_ph_match): applied significant recursion
706 complexity optimization, based on a patch from Matthias Clasen.
708 * tests/patterntest.c: more tests, mostly from matthias.
710 2005-09-20 Matthias Clasen <mclasen@redhat.com>
712 * glib/gqueue.c (g_queue_insert_sorted): Correct the docs.
713 (#316703, Mark Drago)
715 Mon Sep 19 17:23:23 2005 Tim Janik <timj@imendio.com>
717 * glib/gpattern.c: applied a patch from matthias which checks on the
718 upper bound of GPatternSpec length to optimize matches.
721 * tests/patterntest.c: added more match cases.
723 2005-09-16 Tor Lillqvist <tml@novell.com>
725 * glib/gstrfuncs.c (g_ascii_strcasecmp, g_ascii_strncasecmp): Add
726 warning to doc comment that these functions should not be used on
727 encodings like CP932.
729 2005-09-14 Matthias Clasen <mclasen@redhat.com>
731 * tests/keyfile-test.c: Add a test for grup names of length 1.
733 * glib/gkeyfile.c (g_key_file_line_is_group): Accept group names
734 of length 1. (#316309)
736 2005-09-12 Matthias Clasen <mclasen@redhat.com>
738 * glib/gmarkup.c (g_markup_escape_text): Clarify docs.
740 2005-09-11 Sebastian Wilhelmi <seppi@seppi.de>
742 * tests/refcount/Makefile.am (INCLUDES): Link the the refcount
743 tests to the system thread library $(G_THREAD_LIBS). Fixes #313744
746 2005-09-11 Kjartan Maraas <kmaraas@gnome.org>
748 * glib/gmain.c: (g_child_watch_prepare), (g_child_watch_check),
749 (child_watch_helper_thread): Remove some dead code. Closes
752 2005-09-07 Tor Lillqvist <tml@novell.com>
754 * glib/Makefile.am: Create also a console version of the
755 gspawn-win32-helper program, gspawn-win32-helper-console.exe.
756 It's otherwise identical to gspawn-win32-helper.exe, except marked
757 as a console application (linked without the -mwindows option).
759 * glib/gspawn-win32.c (do_spawn_directly, do_spawn_with_pipes):
760 Drop the dont_wait parameter. Its truth value correlated 100% with
761 the NULLness of the exit_status parameter anyway, so it's enough
762 to check whether exit_status is NULL. Invert the sense of the
763 dont_return_handle parameter and rename it to do_return_handle, to
764 make the code easier to read by avoiding double negations.
766 (g_spawn_sync_utf8, g_spawn_async_with_pipes_utf8): Modify calls
767 to do_spawn_with_pipes() accordingly.
769 (do_spawn_with_pipes): If we have a console, use the console
770 version of the helper program, otherwise use the GUI one. This
771 avoids extra console windows opening up in some situations. (In
772 case a console application uses the GUI gspawn-win32-helper.exe to
773 spawn another console application we would get a separate console
774 for the spawned console application).
776 * glib-zip.in: Distribute also gspawn-win32-helper-console.exe.
778 2005-09-05 Matthias Clasen <mclasen@redhat.com>
780 * glib/gmappedfile.c (g_mapped_file_new): Report an error
781 if the file is too large. (#315275, Kjartan Maraas)
783 * glib/gkeyfile.c (g_key_file_load_from_fd): The return value
784 of read() is signed. (#315273, Kjartan Maraas)
786 2005-08-31 Tor Lillqvist <tml@novell.com>
788 * glib/gutils.h: Wrapping atexit() is a bad idea on Windows, where
789 the EXE and each DLL have their own atexit function chains.
791 #define g_atexit as atexit instead. This means it has a
792 better chance of doing what the caller wants. For instance,
793 gtkhtml calls g_atexit() registering a function in gtkhtml
794 itself. This caused a crash when g_atexit() was implemented as a
795 function in the GLib DLL. The gtkhtml DLL was already unloaded by
796 the time the GLib DLL got unloaded.
798 * glib/gutils.c: #undef the #define mentioned above, to also get a
799 real g_atexit() into the DLL for backward compatibility. Document
800 the Windows behaviour of g_atexit(), and document the varying ways
801 atexit() can behave in the context of dynamically loaded modules
804 2005-08-31 Matthias Clasen <mclasen@redhat.com>
808 * glib/gdataset.c: Add string interning functions.
810 2005-08-28 Matthias Clasen <mclasen@redhat.com>
812 * glib/giochannel.c: Unify some near-duplicate strings. (#314654,
815 2005-08-26 Matthias Clasen <mclasen@redhat.com>
817 * configure.in: Bump version to 2.9.0
819 2005-08-25 Tor Lillqvist <tml@novell.com>
821 Make also the g_spawn*() functions take parameters in the GLib
822 file name encoding, i.e. UTF-8, on Windows. Has no impact on Unix
823 API or ABI. Like the other GLib API that was earlier changed to
824 use UTF-8 on Windows, the names of the functions that take UTF-8
825 have _utf8 suffixes added by using preprocessor macros in the
826 header file. The old names are kept for functions with the old
827 behaviour, taking parameters in the system codepage, for DLL ABI
830 * glib/gspawn.h: On Win32 add the suffix _utf8 to the names of the
831 g_spawn*() functions.
833 * glib/gspawn-win32.c: Use wide-char API on NT-based
834 Windows. Convert parameters from UTF-8 to wide chars (NT) or
835 system codepage (Win9x) and call the C library _wspawn*() or
836 spawn*() functions respectvely. Add DLL ABI stability versions
837 that take parameters in the system codepage.
839 * glib/gspawn-win32-helper.c: On NT-based Windows use the
840 wide-char versions of argv and envp, and use wide-char API to
841 change directory and spawn the program to run. Remove the verbose
842 debugging output, it was too complex to modify for the wide-char
843 features. (Just add temporary debugging printouts if needed, no
844 need to have them permanently in the source.)
846 * glib/gspawn.c: Corresponding documentation updates.
848 * glib/glib.symbols: Corresponding changes: Mark the ABI stability
849 symbols as PRIVATE, add the new _utf8-suffixed ones.
851 2005-08-24 Stepan Kasal <kasal@ucw.cz>
853 * glib/gtypes.h (G_MININT64): Cast the constant to gint64; it is
854 guint64 otherwise and that can produce warnings about comparison
855 between signed and unsigned.
857 2005-08-23 Matthias Clasen <mclasen@redhat.com>
859 * glib/gutils.c: Fix the crt_externs.h include.
861 2005-08-23 Stepan Kasal <kasal@ucw.cz>
863 * NEWS: Fix spelling of my first name.
865 2005-08-23 Matthias Clasen <mclasen@redhat.com>
869 * === Released 2.8.1 ===
873 2005-08-20 Hans Breuer <hans@breuer.org>
875 * glib/makefile.msc.in : link with ws2_32.lib
877 2005-08-18 Tor Lillqvist <tml@novell.com>
879 * configure.in: Check for <sys/wait.h>
881 * glib/gbacktrace.c: Include <sys/wait.h> on if HAVE_SYS_WAIT_H.
883 2005-08-18 Ross Burton <ross@burtonini.com>
886 Optimise single-character insertions.
891 * tests/string-test.c:
892 Add tests for new optimisation, and fix a leak.
894 2005-08-17 Matthias Clasen <mclasen@redhat.com>
896 * configure.in: Check for crt_externs.h and _NSGetEnviron.
898 * glib/gutils.c: On Darwin, include crt-externs.h and
899 define environ using _NSGetEnviron(). (#313731)
901 2005-08-16 Stepan Kasal <kasal@ucw.cz>
903 * glib/gutils.c (g_get_any_init): Move the body of the big if...
904 (g_get_any_init_do): ... to this new function.
905 (g_get_any_init): Declare as inline.
906 (g_get_any_init_locked): New inline function, does the locking.
907 Make use of these two throughout the code.
909 2005-08-15 Matthias Clasen <mclasen@redhat.com>
911 * glib/gbacktrace.c (g_on_error_stack_trace): Wait for
912 the child process and then simply return. This makes
913 The "S" option work as documented in g_on_error_query().
914 (#313125, Matthew F. Barnes)
916 * glib/gunicode.h: Update the link to Unicode category
917 values. (#313369, Behnam Esfahbod)
919 * glib/gqueue.c (g_queue_find_custom): Clarify docs
920 a little. (#311727, Tristan van Berkom)
922 * glib/abicheck.sh, gobject/abicheck.sh: Make the
923 check work on ia64 too, where some symbols ended up
924 in yet another section.