1 2005-12-27 Manish Singh <yosh@gimp.org>
3 * tests/run-collate-tests.sh: export LC_COLLATE so it takes.
6 2005-12-26 Matthias Clasen <mclasen@redhat.com>
8 * glib/gslice.c: On Win32, include process.h (#325015, Kazuki
11 2005-12-25 Matthias Clasen <mclasen@redhat.com>
13 * AUTHORS: Update my email
15 * tests/slice-test.c: Fix C99isms. (#324950, Dan Yefimov)
17 2005-12-21 Matthias Clasen <mclasen@redhat.com>
19 * README.in: Add some notes on when 'make check' may fail.
21 2005-12-20 Matthias Clasen <mclasen@redhat.com>
25 * glib/gthreadpool.c (g_thread_pool_set_sort_function): New function
26 to sort tasks pushed into a threadpool. (#324479, Martyn Russell)
28 * tests/threadpool-test.c: Test this.
30 Tue Dec 20 18:14:14 2005 Tim Janik <timj@imendio.com>
32 * glib/gslice.[hc]: added mem_error() and mem_assert() to test and
33 handle errors without depending on gmessage.c which might not be
34 setup when the error occours.
35 removed G_SLICE_CONFIG_ALWAYS_FREE config option, fixed the code so
36 always freeing can be achieved by adjusting the working set time to
37 0 with G_SLICE_CONFIG_WORKING_SET_MSECS.
38 added G_SLICE_CONFIG_COLOR_INCREMENT to test different color increments
39 (mainly 0 and 1). reduced the minimum block size to 128 bytes, to
40 minimize wastage if small amounts of differently sized structrues are
41 allocated, this does come at a performance cost of roughly 5% though.
42 fixed up block alignment calculation, so it works for varying
43 block sizes. only use strerror() not g_strerror() since the latter
44 depends on working GQuark and GSlice.
45 mem_error(): implemented in terms of fprintf and vfprintf.
47 * tests/slice-color.c: new program to test cache colorization effects.
49 * tests/slice-test.c: trade G_SLICE_CONFIG_ALWAYS_FREE for 0 duration
50 G_SLICE_CONFIG_WORKING_SET_MSECS.
52 2005-12-17 Matthias Clasen <mclasen@redhat.com>
54 * glib/goption.c (parse_short_option): Set an error in all
55 failure cases. (#324332, Tim-Philipp Müller)
57 2005-12-17 Sebastian Wilhelmi <seppi@seppi.de>
59 * glib/gatomic.c: Fix memory barrier position in g_atomic_int_get
60 and g_atomic_pointer_get. Add g_atomic_int_set and
61 g_atomic_pointer_set implementations for the !DEFINE_WITH_MUTEXES &&
62 G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case, as well as defining them
63 as functions (additionally to the macros in the header) for the
64 !G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case.
66 2005-12-16 Matthias Clasen <mclasen@redhat.com>
68 * glib/gmem.c (g_allocator_new): Don't return a pointer to
69 a const struct, since apps expect to be able to modify it.
70 (#324179, J. Ali Harlow)
72 Tue Dec 13 10:13:32 2005 Tim Janik <timj@imendio.com>
74 * glib/gatomic.h: added g_atomic_pointer_set() and g_atomic_int_set()
75 for systems where the initialization of atomic variables requires a
78 2005-12-09 Matthias Clasen <mclasen@redhat.com>
80 * configure.in: Bump version
82 * === Released 2.9.1 ===
84 2005-12-09 Alexander Larsson <alexl@redhat.com>
88 Use G_GNUC_WARN_UNUSED_RESULT on list functions that return
91 2005-12-08 Matthias Clasen <mclasen@redhat.com>
95 2005-12-07 Martyn Russell <martyn@imendio.com>
98 - Call g_queue_insert_sorted() instead of duplicating the code.
99 - Call g_queue_sort() instead of duplicating the code.
100 - Invert sort function results to make sure the same sort function
101 gives the same results across glist, gslist, gqueue and
104 * tests/asyncqueue-test.c:
105 - Updated the sort function to reflect the example in the
106 documentation for gasyncqueue.c.
108 2005-12-07 Martyn Russell <martyn@imendio.com>
110 * docs/reference/glib/glib-sections.txt:
111 * docs/reference/glib/tmpl/linked_lists_double.sgml:
112 * docs/reference/glib/tmpl/linked_lists_single.sgml:
115 - Added g_list_insert_sorted_with_data () and
116 g_slist_insert_sorted_with_data ().
117 - Removed the extra check in g_list_sort() and g_slist_sort() for
118 GCompareDataFunc vs. GCompareFunc.
120 2005-12-07 Tor Lillqvist <tml@novell.com>
122 * glib-gettextize.in: Look up prefix at run-time on Win32,
123 assuming the standard directory structure with glib-gettextize in
126 2005-12-06 Matthias Clasen <mclasen@redhat.com>
128 * glib/gmem.h: Exempt GTK+ from the mem chunk deprecation, since
129 we need GTK+ 2.8 to compile against GLib 2.10.
131 2005-12-05 Manish Singh <yosh@gimp.org>
133 * tests/run-collate-test.sh: set LC_COLLATE, not LANG, to be sure
134 to override any user settings.
136 2005-12-05 Matthias Clasen <mclasen@redhat.com>
138 * configure.in: Revert an accidential commit
140 2005-12-05 Matthias Clasen <mclasen@redhat.com>
142 * glib/gutf8.c: Documentation updates. (#323291, Morten Welinder)
144 * tests/Makefile.am (TESTS_ENVIRONMENT):
145 * tests/gobject/Makefile.am (TESTS_ENVIRONMENT): Set
146 MALLOC_CHECK_ and MALLOC_PERTURB_
148 * tests/run-collate-tests.sh: Run the collation tests explicitly
149 in en_US locale. (#320463)
151 * glib/gmem.h: Really deprecate GMemChunk.
158 * glib/gstrfuncs.c: Add versioned deprecation docs.
160 2005-12-05 Martyn Russell <martyn@imendio.com>
162 * docs/reference/glib/glib-sections.txt:
163 * glib/gasyncqueue.[ch]:
164 - Added support for sorting async queues by with _push_sorted(),
165 _push_sorted_unlocked(), _sort() and _sort_unlocked() (#323047).
168 * tests/asyncqueue-test.c:
169 - Added test case for gasyncqueue.c
171 Mon Dec 5 15:53:20 2005 Tim Janik <timj@imendio.com>
173 * glib/gslice.c: implement chain walking for arbitrary ->next pointer
174 offsets in g_slice_free_chain_with_offset() based on a patch by behdad
175 in bug 323178. moved time consuming logic from g_slice_free() out of
176 the inner loop, so g_slice_free_chain_with_offset() provides a real
177 performance benefit over g_slice_free1() now.
179 * glib/gslice.h: renamed g_slice_free_chain() to
180 g_slice_free_chain_with_offset(). implemented g_slice_free_chain() as
181 a type-safe macro as suggested in bug 323178.
182 simplified the macro implementation of g_slice_free() and implemented
183 it in a type safe manner for all compliers as suggested by Morten
184 Welinder <mortenw@gnome.org>.
189 * glib/glib.symbols: s/g_slice_free_chain/g_slice_free_chain_with_offset/
191 2005-12-05 Matthias Clasen <mclasen@redhat.com>
193 * glib/gasyncqueue.c: Add some docs.
195 * tests/libmoduletestplugin_a.c: Fix compiler warnings.
197 * glib/gatomic.c: In the ia64 implementation, use
198 __sync builtin without _si or _di suffix. (#321229,
199 Stanislav Brabec, patch by Andreas Schwab)
201 2005-12-04 Behdad Esfahbod <behdad@gnome.org>
203 * glib/gslice.h: Remove comma at the end of enum.
205 2005-12-04 Matthias Clasen <mclasen@redhat.com>
207 Handle multiple user names with the same UID better.
208 (#319535, Laszlo Peter)
210 * glib/gutils.c (g_get_any_init_do): When determining user
211 data, first look up $LOGNAME. If the UID doesn't match
212 getuid(), fall back to the current behaviour of looking
213 up the user data based on getuid().
215 2005-12-04 Matthias Clasen <mclasen@redhat.com>
217 * glib/gmacros.h (G_GNUC_WARN_UNUSED_RESULT): Add a macro
218 to make gcc warn if a function result is ignored. (#145466,
219 Arjan van de Ven, Alex Larsson)
221 * glib/gmem.h: Add the new attribute to g_realloc and
224 2005-12-03 Matthias Clasen <mclasen@redhat.com>
227 * glib/gcache.h: Deprecate g_cache_value_foreach. (#322956,
230 * glib/gmappedfile.c: Make mapping of empty files
233 * glib/gfileutils.c: Don't fork a new process just to
234 fix the permissions of the created temp file. (#321318,
235 Alexis S. L. Carvalho)
237 2005-12-02 Matthias Clasen <mclasen@redhat.com>
239 * README.in: Add a note about Solaris threads.
242 * configure.in: Remove support for Solaris threads.
243 (#136971, Sebastian Wilhelmi, patch by Andrew Paprocki)
245 2005-12-02 Matthias Clasen <mclasen@redhat.com>
247 * configure.in: Define G_GUINT64_CONSTANT in analogy to
248 G_GINT64_CONSTANT. (#322568, Andrew Paprocki)
250 2005-12-02 Matthias Clasen <mclasen@redhat.com>
252 * glib/gslice.c: Win32 portability fixes and C99ism removal,
253 pointed out by Kazuki Iwamoto. (#323052)
255 2005-12-02 Matthias Clasen <mclasen@redhat.com>
257 * glib/gatomic.c (g_atomic_int_add): Add a missing volatile for
258 the IA64 implementation.
260 Fri Dec 2 16:18:09 2005 Tim Janik <timj@imendio.com>
262 * glib/gslice.c: some naming and type size fixups.
264 Fri Dec 2 13:08:58 2005 Tim Janik <timj@gtk.org>
266 * glib/gslice.h (g_slice_free): reworked GCC-specific type-safe macro
267 variant into something less verbose, but digestible for gcc-3.4.
269 Fri Dec 2 10:55:07 2005 Tim Janik <timj@imendio.com>
271 * tests/slice-test.c: extended to perform the benchmarking on the old
272 memchunk code if 'O' is selected.
274 * tests/memchunks.c: new file which contains the old GLib mem chunks
275 implementation with prefix old_mem_chunk_.
277 * tests/Makefile.am: added memchunks.c
279 Fri Dec 2 00:16:59 2005 Tim Janik <timj@imendio.com>
281 * glib/gslice.c: improved gettimeofday() timer resolution by maintaining
282 acceptable load distribution of the syscall.
283 (allocator_get_magazine_threshold): reduce minimum/single-thread magazine
284 sizes to approximately page_size/2.5 to avoid excessive startup allocations.
286 2005-12-01 Matthias Clasen <mclasen@redhat.com>
288 * tests/gobject/ifacecheck.c (test_iface_base_init): Don't
289 double-free base interface members.
291 * tests/Makefile.am: Remove duplicate variable.
293 Thu Dec 1 17:32:46 2005 Tim Janik <timj@imendio.com>
295 * glib/gslice.[hc]: new slice allocator implementation.
297 * tests/slice-test.c: added random slice allocation test.
299 * glib/gthread.[hc]: removed newly added private thread mem API.
301 * glib/gthreadinit.h:
304 * glib/gmem.c: divided glib threading initialisation into three phases,
305 initialisation where private keys and messaging are not available (only
306 needed by gmem.c), initialisation without messaging but private keys
307 available (gslice.c, gmessage.c), and full fledged initialisers that
308 server the rest of glib. initialisation functions got renamed to reflect
309 the limitations of their corresponding phases.
311 * glib/gmem.c: removed memchunk code, defer allocations to
314 * glib/gmem.[hc]: removed g_slice_* skeletons.
316 * glib/glib.symbols: added g_slice_* symbols.
318 * configure.in: check for availability of posix_memalign(3), memalign(3)
321 * glib/Makefile.am: added gslice.[hc].
323 2005-12-01 Tor Lillqvist <tml@novell.com>
325 * glib/gstdio.c (g_stat): In the Win32 implementation, strip
326 trailing slash(es) for non-root folders. stat() fails if non-root
327 folders are specified with trailing slashes. It's too much hassle
328 to demand that callers strip such slashes themselves, especially
329 as it is easy to get it wrong and strip the slash of a root
331 (g_rename): On NT-based Windows, use MoveFileEx() with
332 MOVEFILE_REPLACE_EXISTING to better match Unix behaviour.
334 2005-11-28 Matthias Clasen <mclasen@redhat.com>
336 Fix G_STMT_START / G_STMT_END on Solaris. (#321972,
339 * configure.in: Check whether do { } while (0) works.
341 * glib/gmacros.h: Use do { } while (0) for G_STMT_START /
342 G_STMT_END if it works.
344 2005-11-28 Matthias Clasen <mclasen@redhat.com>
346 * glib/gthread.c (g_static_rw_lock_wait, g_static_rw_lock_signal):
347 * glib/gnode.c (g_node_depth_traverse_level):
348 * glib/gmem.c (g_allocator_new):
349 * glib/ghash.c (g_hash_table_unref, g_hash_table_destroy)
350 (g_hash_table_foreach_remove_or_steal): Silence compiler
353 2005-11-27 Matthias Clasen <mclasen@redhat.com>
355 * glib/gunicollate.c (g_utf8_collate_key_for_filename):
356 Don't read beyond len. (#322520, Christian Persch)
358 Wed Nov 23 17:34:01 2005 Tim Janik <timj@imendio.com>
360 * glib/gdataset.c: access datalist flags via atomic pointer access
361 functions, instead of acquiring the dataset lock. this is faster and
362 also matches the atomic pointer readouts in gdatalistprivate.h and
363 direct pointer modifications required by gobject.c.
365 Wed Nov 23 13:35:31 2005 Tim Janik <timj@gtk.org>
367 * glib/gdataset.c: streamlined initialization code somewhat,
368 removed GData node cache.
370 2005-11-22 Matthias Clasen <mclasen@redhat.com>
372 * glib/ghash.c (g_hash_table_ref):
373 (g_hash_table_unref): Mark these as new API.
375 Avoid double locking in g_intern_string (#322133,
378 * glib/gdataset.c (g_quark_from_string_internal): New
379 internal function which factors out common parts of
380 g_quark_from[_static]_string.
381 (g_quark_from_string, g_quark_from_static_string):
382 Use g_quark_from_string_internal.
383 (g_intern_string, g_intern_static_string): Use
384 g_quark_from_string_internal, and only take the
386 (g_quark_new): Don't store the strings shifted by -1
387 in the g_quarks array.
388 (g_quark_to_string): Adapt to the previous change.
390 Tue Nov 22 14:04:26 2005 Tim Janik <timj@imendio.com>
394 g_hash_table_new_full(): create hash tables with a ref count of 1.
395 g_hash_table_ref(): atomically ref_count+=1
396 g_hash_table_unref(): atomically ref_count-=1, destroys hash table
397 when refcount reaches 0.
398 g_hash_table_destroy(): just destroy keys and values, unref by 1.
399 g_hash_table_insert():
400 g_hash_table_replace(): assert ref_count>0.
403 * glib/gatomic.c: added 'volatile' qualifier to all atomic pointer and
406 2005-11-20 Behdad Esfahbod <behdad@gnome.org>
408 * glib/guniprop.c (g_unichar_get_mirror_char): Remove unused
409 variables. (#321984, Andrew Paprocki)
411 2005-11-18 Matthias Clasen <mclasen@redhat.com>
413 * configure.in: Bump version
415 2005-11-17 Matthias Clasen <mclasen@redhat.com>
417 * === Released 2.9.0 ===
419 * glib/glib.symbols: Updates
421 * glib/gutf8.c (g_utf8_offset_to_pointer): Add some performance
422 hints to the docs. (Paolo Borelli)
426 * glib/gthread.c (g_thread_foreach): Mark as new api.
428 * README.in: Updates.
432 * glib/gdate.c (g_date_set_time_t):
433 (g_date_set_time_val): Add functions to set a date from
434 a time_t and from a GTimeVal.
435 (g_date_set_time): Deprecate. (#314067, Roger Leigh)
438 * tests/utf8-pointer.c: Unit tests for g_utf8_pointer_to_offset
439 and g_utf8_offset_to_pointer.
441 * glib/gutf8.c (g_utf8_pointer_to_offset)
442 (g_utf8_offset_to_pointer): Handle negative offsets, and use
443 "stutter stepping" for going backwards. (#320638, Larry
448 * glib/gthread.c: const correctness fixes, found
449 by Arjan van de Ven and gcc.
451 2005-11-16 Behdad Esfahbod <behdad@gnome.org>
453 * docs/reference/glib/tmpl/unicode.sgml: Correct typo on mentioning
454 @G_UNICODE_BREAK_UNKNOWN that should be #G_UNICODE_BREAK_UNKNOWN.
456 2005-11-09 Behdad Esfahbod <behdad@gnome.org>
458 * glib/guniprop.c: Use bit hacks instead when checking a general
459 category value against multiple values.
461 * glib/gutf8.c: Change ISO10646 to Unicode in docs.
463 2005-11-10 Simos Xenitellis <simos@gnome.org>
465 * configure.in: Added tt (Tatar) to ALL_LINGUAS.
467 2005-11-05 Tor Lillqvist <tml@novell.com>
469 * glib/giowin32.c (g_io_win32_set_flags): Don't call g_warning().
470 (#320688) Instead set error and return failure.
472 2005-11-07 Matthias Clasen <mclasen@redhat.com>
474 * glib/gmain.c (g_main_dispatch): Don't call cb_funcs->unref
475 while holding the context lock. (#320886, Andy Wingo)
477 2005-11-07 Billy Biggs <vektor@dumbterm.net>
479 * tests/refcount/Makefile.am: Add a missing reference to libglib
480 in the LDADD for this test.
482 2005-11-07 Matthias Clasen <mclasen@redhat.com>
484 * glib/gmem.c (g_allocator_new): We need to set n_preallocs to a
485 nonzero value, otherwise GTK+ 2.8 breaks when compiled against
486 GLib 2.9. (#320755, Luca Ferretti)
488 2005-11-04 Matthias Clasen <mclasen@redhat.com>
490 * glib/Makefile.am: Apply a patch from Behdad Esfahbod to
491 use a faster lookup table for g_unichar_get_mirror_char().
492 * glib/gmirroringtable.h: The new table.
493 * glib/gunichartables.h: Remove bidi_mirroring_table.
494 * glib/gen-unicode-tables.pl: Don't generate the mirroring
496 * glib/glib-mirroring-tab/*: Sources for the program
497 which generated gmirroringtable.h.
499 * glib/glist.c: Avoid some code duplication.
501 * glib/gscanner.h: Include gdataset.h. (#320322)
503 * glib/gdebug.h: add new GDebugFlag for fatal_criticals
504 * glib/gmessages.c: (_g_debug_init): handle G_DEBUG=fatal_criticals,
505 to help find critical warnings in applications. (#320017,
508 2005-11-02 Tor Lillqvist <tml@novell.com>
510 * glib/glib.symbols: Remove large amount of trailing whitespace
511 from one line. Remove #ifdef G_UNDEFINED from around
512 g_hash_table_new and g_hash_table_new_full symbols, what was the
513 purpose of that? Add the g_slice functions.
515 Tue Nov 1 16:24:20 2005 Tim Janik <timj@imendio.com>
517 * glib/gmem.[hc]: prepared deprecation of GMemChunk and GAllocator.
518 added g_slice_*() API to allocate and cache small bits of memory.
519 an actuall allocator implementation for g_slice_*() is still pending.
521 * glib/gthread.[hc]: changes from a patch by Matthias Clasen.
522 changed GRealThread list to use in-structure *next; fields instead
523 of GSList, in order for thread iteration to not depenend on g_slice_*()
525 _g_thread_mem_private_get():
526 _g_thread_mem_private_set(): added accessors for private memory,
527 needed because the ordinary GPrivate implementation relies on GArray
528 and GSList and therefore indirectly on working g_slice_*() allocations.
531 g_thread_foreach(): new public API function to loop over all existing threads.
545 removed GAllocator and free list usages and accompanying locks.
546 use g_slice_*() API to allocate and cache small bits of memory.
548 * glib/ghook.h: removed GMemChunk field from public API.
551 * glib/glist.h: deprecate allocator API, provide _free1() for consistency.
553 * glib/gnode.h: deprecate allocator API.
555 * glib/gmain.c: reordered GPollRec fields so g_slice_free_chain() can
556 be used for poll rec lists.
558 * glib/grel.c: removed mem chunk usage, and allocated tuples via g_slice_*().
559 g_relation_destroy(): free all tuples from the all_tuples hash table,
560 this effectively maintains the life time track keeping of tuples.
561 g_relation_delete_tuple(): free tuples which are removed from the
562 all_tuples hash table. this fixes a temporary leak that was present
563 in the memchunk code until the destruction of the relation.
565 2005-10-29 Matthias Clasen <mclasen@redhat.com>
567 * tests/convert-test.c: Add some tests for conversions between
568 UTF-8, UCS-4 and UTF-16.
570 * glib/gutf8.c (g_utf8_to_ucs4, g_utf8_to_utf16): Fix handling
571 of len == -1, noticed by Morten Welinder.
573 2005-10-27 Erdal Ronahi <erdal.ronahi@gmail.com>
575 * configure.in: Added ku (Kurdish) to ALL_LINGUAS
577 2005-10-26 Matthias Clasen <mclasen@redhat.com>
579 * glib/gutf8.c (g_ucs4_to_utf8): Don't set items_read twice
580 in the error case, and add some documentation. (#319806, Morten
583 2005-10-19 Manish Singh <yosh@gimp.org>
585 * configure.in: Use AC_CHECK_FUNCS for _NSGetEnviron, to get the
586 config.h symbol automatically. Fixes bug #313731.
588 2005-10-19 Tor Lillqvist <tml@novell.com>
590 * glib/gwin32.c (g_win32_get_package_installation_directory):
591 Return a g_strdup()ed copy of the value stored in the hash table,
592 so that it can be g_free()d without leaving a dangling pointer in
593 the hash table. (#319232)
595 2005-10-06 Matthias Clasen <mclasen@redhat.com>
597 * glib/gunicollate.c (g_utf8_collate_key_for_filename): Handle
598 all-zero sequences correctly. (#317930, Sebastien Bacher)
600 2005-10-05 Matthias Clasen <mclasen@redhat.com>
602 * glib/gmarkup.c (g_markup_escape_text): Doc addition
604 * Makefile.am: Add Collation tests.
606 * tests/collate/*: Inputs and expected outputs for collation tests.
608 * tests/run-collate-tests.sh: Script to run collation tests.
610 * tests/unicode-collate.c (main): Rework slightly to make
611 it usable in unit tests. Also test g_utf8_collate_key_for_filename().
613 2005-10-01 Behdad Esfahbod <behdad@gnome.org>
615 * docs/reference/glib/tmpl/unicode.sgml:
616 * glib/gen-unicode-tables.pl:
618 * glib/gunichartables.h:
620 * tests/casefold.txt:
621 * tests/casemap.txt: Updated to Unicode 4.1. There are five new
622 GUnicodeBreakType types. That may break some applications, like
625 2005-09-26 Matthias Clasen <mclasen@redhat.com>
627 * glib/gstrfuncs.c (g_ascii_strtoull): Add details to the
628 docs. (#314393, Matthew F. Barnes)
631 * glib/gprintf.h: Remove g_snprintf() and g_vsnprintf(), since
632 they are already declared in glib.h. This doesn't break documented
633 use of gprintf.h, but should probably be pointed out in the
634 release notes for 2.10. (#314232, Behdad Esfahbod)
636 Tue Sep 20 13:16:04 2005 Tim Janik <timj@imendio.com>
638 * glib/gpattern.c (g_pattern_ph_match): applied significant recursion
639 complexity optimization, based on a patch from Matthias Clasen.
641 * tests/patterntest.c: more tests, mostly from matthias.
643 2005-09-20 Matthias Clasen <mclasen@redhat.com>
645 * glib/gqueue.c (g_queue_insert_sorted): Correct the docs.
646 (#316703, Mark Drago)
648 Mon Sep 19 17:23:23 2005 Tim Janik <timj@imendio.com>
650 * glib/gpattern.c: applied a patch from matthias which checks on the
651 upper bound of GPatternSpec length to optimize matches.
654 * tests/patterntest.c: added more match cases.
656 2005-09-16 Tor Lillqvist <tml@novell.com>
658 * glib/gstrfuncs.c (g_ascii_strcasecmp, g_ascii_strncasecmp): Add
659 warning to doc comment that these functions should not be used on
660 encodings like CP932.
662 2005-09-14 Matthias Clasen <mclasen@redhat.com>
664 * tests/keyfile-test.c: Add a test for grup names of length 1.
666 * glib/gkeyfile.c (g_key_file_line_is_group): Accept group names
667 of length 1. (#316309)
669 2005-09-12 Matthias Clasen <mclasen@redhat.com>
671 * glib/gmarkup.c (g_markup_escape_text): Clarify docs.
673 2005-09-11 Sebastian Wilhelmi <seppi@seppi.de>
675 * tests/refcount/Makefile.am (INCLUDES): Link the the refcount
676 tests to the system thread library $(G_THREAD_LIBS). Fixes #313744
679 2005-09-11 Kjartan Maraas <kmaraas@gnome.org>
681 * glib/gmain.c: (g_child_watch_prepare), (g_child_watch_check),
682 (child_watch_helper_thread): Remove some dead code. Closes
685 2005-09-07 Tor Lillqvist <tml@novell.com>
687 * glib/Makefile.am: Create also a console version of the
688 gspawn-win32-helper program, gspawn-win32-helper-console.exe.
689 It's otherwise identical to gspawn-win32-helper.exe, except marked
690 as a console application (linked without the -mwindows option).
692 * glib/gspawn-win32.c (do_spawn_directly, do_spawn_with_pipes):
693 Drop the dont_wait parameter. Its truth value correlated 100% with
694 the NULLness of the exit_status parameter anyway, so it's enough
695 to check whether exit_status is NULL. Invert the sense of the
696 dont_return_handle parameter and rename it to do_return_handle, to
697 make the code easier to read by avoiding double negations.
699 (g_spawn_sync_utf8, g_spawn_async_with_pipes_utf8): Modify calls
700 to do_spawn_with_pipes() accordingly.
702 (do_spawn_with_pipes): If we have a console, use the console
703 version of the helper program, otherwise use the GUI one. This
704 avoids extra console windows opening up in some situations. (In
705 case a console application uses the GUI gspawn-win32-helper.exe to
706 spawn another console application we would get a separate console
707 for the spawned console application).
709 * glib-zip.in: Distribute also gspawn-win32-helper-console.exe.
711 2005-09-05 Matthias Clasen <mclasen@redhat.com>
713 * glib/gmappedfile.c (g_mapped_file_new): Report an error
714 if the file is too large. (#315275, Kjartan Maraas)
716 * glib/gkeyfile.c (g_key_file_load_from_fd): The return value
717 of read() is signed. (#315273, Kjartan Maraas)
719 2005-08-31 Tor Lillqvist <tml@novell.com>
721 * glib/gutils.h: Wrapping atexit() is a bad idea on Windows, where
722 the EXE and each DLL have their own atexit function chains.
724 #define g_atexit as atexit instead. This means it has a
725 better chance of doing what the caller wants. For instance,
726 gtkhtml calls g_atexit() registering a function in gtkhtml
727 itself. This caused a crash when g_atexit() was implemented as a
728 function in the GLib DLL. The gtkhtml DLL was already unloaded by
729 the time the GLib DLL got unloaded.
731 * glib/gutils.c: #undef the #define mentioned above, to also get a
732 real g_atexit() into the DLL for backward compatibility. Document
733 the Windows behaviour of g_atexit(), and document the varying ways
734 atexit() can behave in the context of dynamically loaded modules
737 2005-08-31 Matthias Clasen <mclasen@redhat.com>
741 * glib/gdataset.c: Add string interning functions.
743 2005-08-28 Matthias Clasen <mclasen@redhat.com>
745 * glib/giochannel.c: Unify some near-duplicate strings. (#314654,
748 2005-08-26 Matthias Clasen <mclasen@redhat.com>
750 * configure.in: Bump version to 2.9.0
752 2005-08-25 Tor Lillqvist <tml@novell.com>
754 Make also the g_spawn*() functions take parameters in the GLib
755 file name encoding, i.e. UTF-8, on Windows. Has no impact on Unix
756 API or ABI. Like the other GLib API that was earlier changed to
757 use UTF-8 on Windows, the names of the functions that take UTF-8
758 have _utf8 suffixes added by using preprocessor macros in the
759 header file. The old names are kept for functions with the old
760 behaviour, taking parameters in the system codepage, for DLL ABI
763 * glib/gspawn.h: On Win32 add the suffix _utf8 to the names of the
764 g_spawn*() functions.
766 * glib/gspawn-win32.c: Use wide-char API on NT-based
767 Windows. Convert parameters from UTF-8 to wide chars (NT) or
768 system codepage (Win9x) and call the C library _wspawn*() or
769 spawn*() functions respectvely. Add DLL ABI stability versions
770 that take parameters in the system codepage.
772 * glib/gspawn-win32-helper.c: On NT-based Windows use the
773 wide-char versions of argv and envp, and use wide-char API to
774 change directory and spawn the program to run. Remove the verbose
775 debugging output, it was too complex to modify for the wide-char
776 features. (Just add temporary debugging printouts if needed, no
777 need to have them permanently in the source.)
779 * glib/gspawn.c: Corresponding documentation updates.
781 * glib/glib.symbols: Corresponding changes: Mark the ABI stability
782 symbols as PRIVATE, add the new _utf8-suffixed ones.
784 2005-08-24 Stepan Kasal <kasal@ucw.cz>
786 * glib/gtypes.h (G_MININT64): Cast the constant to gint64; it is
787 guint64 otherwise and that can produce warnings about comparison
788 between signed and unsigned.
790 2005-08-23 Matthias Clasen <mclasen@redhat.com>
792 * glib/gutils.c: Fix the crt_externs.h include.
794 2005-08-23 Stepan Kasal <kasal@ucw.cz>
796 * NEWS: Fix spelling of my first name.
798 2005-08-23 Matthias Clasen <mclasen@redhat.com>
802 * === Released 2.8.1 ===
806 2005-08-20 Hans Breuer <hans@breuer.org>
808 * glib/makefile.msc.in : link with ws2_32.lib
810 2005-08-18 Tor Lillqvist <tml@novell.com>
812 * configure.in: Check for <sys/wait.h>
814 * glib/gbacktrace.c: Include <sys/wait.h> on if HAVE_SYS_WAIT_H.
816 2005-08-18 Ross Burton <ross@burtonini.com>
819 Optimise single-character insertions.
824 * tests/string-test.c:
825 Add tests for new optimisation, and fix a leak.
827 2005-08-17 Matthias Clasen <mclasen@redhat.com>
829 * configure.in: Check for crt_externs.h and _NSGetEnviron.
831 * glib/gutils.c: On Darwin, include crt-externs.h and
832 define environ using _NSGetEnviron(). (#313731)
834 2005-08-16 Stepan Kasal <kasal@ucw.cz>
836 * glib/gutils.c (g_get_any_init): Move the body of the big if...
837 (g_get_any_init_do): ... to this new function.
838 (g_get_any_init): Declare as inline.
839 (g_get_any_init_locked): New inline function, does the locking.
840 Make use of these two throughout the code.
842 2005-08-15 Matthias Clasen <mclasen@redhat.com>
844 * glib/gbacktrace.c (g_on_error_stack_trace): Wait for
845 the child process and then simply return. This makes
846 The "S" option work as documented in g_on_error_query().
847 (#313125, Matthew F. Barnes)
849 * glib/gunicode.h: Update the link to Unicode category
850 values. (#313369, Behnam Esfahbod)
852 * glib/gqueue.c (g_queue_find_custom): Clarify docs
853 a little. (#311727, Tristan van Berkom)
855 * glib/abicheck.sh, gobject/abicheck.sh: Make the
856 check work on ia64 too, where some symbols ended up
857 in yet another section.