1 2005-12-17 Matthias Clasen <mclasen@redhat.com>
3 * glib/goption.c (parse_short_option): Set an error in all
4 failure cases. (#324332, Tim-Philipp Müller)
6 2005-12-17 Sebastian Wilhelmi <seppi@seppi.de>
8 * glib/gatomic.c: Fix memory barrier position in g_atomic_int_get
9 and g_atomic_pointer_get. Add g_atomic_int_set and
10 g_atomic_pointer_set implementations for the !DEFINE_WITH_MUTEXES &&
11 G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case, as well as defining them
12 as functions (additionally to the macros in the header) for the
13 !G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case.
15 2005-12-16 Matthias Clasen <mclasen@redhat.com>
17 * glib/gmem.c (g_allocator_new): Don't return a pointer to
18 a const struct, since apps expect to be able to modify it.
19 (#324179, J. Ali Harlow)
21 Tue Dec 13 10:13:32 2005 Tim Janik <timj@imendio.com>
23 * glib/gatomic.h: added g_atomic_pointer_set() and g_atomic_int_set()
24 for systems where the initialization of atomic variables requires a
27 2005-12-09 Matthias Clasen <mclasen@redhat.com>
29 * configure.in: Bump version
31 * === Released 2.9.1 ===
33 2005-12-09 Alexander Larsson <alexl@redhat.com>
37 Use G_GNUC_WARN_UNUSED_RESULT on list functions that return
40 2005-12-08 Matthias Clasen <mclasen@redhat.com>
44 2005-12-07 Martyn Russell <martyn@imendio.com>
47 - Call g_queue_insert_sorted() instead of duplicating the code.
48 - Call g_queue_sort() instead of duplicating the code.
49 - Invert sort function results to make sure the same sort function
50 gives the same results across glist, gslist, gqueue and
53 * tests/asyncqueue-test.c:
54 - Updated the sort function to reflect the example in the
55 documentation for gasyncqueue.c.
57 2005-12-07 Martyn Russell <martyn@imendio.com>
59 * docs/reference/glib/glib-sections.txt:
60 * docs/reference/glib/tmpl/linked_lists_double.sgml:
61 * docs/reference/glib/tmpl/linked_lists_single.sgml:
64 - Added g_list_insert_sorted_with_data () and
65 g_slist_insert_sorted_with_data ().
66 - Removed the extra check in g_list_sort() and g_slist_sort() for
67 GCompareDataFunc vs. GCompareFunc.
69 2005-12-07 Tor Lillqvist <tml@novell.com>
71 * glib-gettextize.in: Look up prefix at run-time on Win32,
72 assuming the standard directory structure with glib-gettextize in
75 2005-12-06 Matthias Clasen <mclasen@redhat.com>
77 * glib/gmem.h: Exempt GTK+ from the mem chunk deprecation, since
78 we need GTK+ 2.8 to compile against GLib 2.10.
80 2005-12-05 Manish Singh <yosh@gimp.org>
82 * tests/run-collate-test.sh: set LC_COLLATE, not LANG, to be sure
83 to override any user settings.
85 2005-12-05 Matthias Clasen <mclasen@redhat.com>
87 * configure.in: Revert an accidential commit
89 2005-12-05 Matthias Clasen <mclasen@redhat.com>
91 * glib/gutf8.c: Documentation updates. (#323291, Morten Welinder)
93 * tests/Makefile.am (TESTS_ENVIRONMENT):
94 * tests/gobject/Makefile.am (TESTS_ENVIRONMENT): Set
95 MALLOC_CHECK_ and MALLOC_PERTURB_
97 * tests/run-collate-tests.sh: Run the collation tests explicitly
98 in en_US locale. (#320463)
100 * glib/gmem.h: Really deprecate GMemChunk.
107 * glib/gstrfuncs.c: Add versioned deprecation docs.
109 2005-12-05 Martyn Russell <martyn@imendio.com>
111 * docs/reference/glib/glib-sections.txt:
112 * glib/gasyncqueue.[ch]:
113 - Added support for sorting async queues by with _push_sorted(),
114 _push_sorted_unlocked(), _sort() and _sort_unlocked() (#323047).
117 * tests/asyncqueue-test.c:
118 - Added test case for gasyncqueue.c
120 Mon Dec 5 15:53:20 2005 Tim Janik <timj@imendio.com>
122 * glib/gslice.c: implement chain walking for arbitrary ->next pointer
123 offsets in g_slice_free_chain_with_offset() based on a patch by behdad
124 in bug 323178. moved time consuming logic from g_slice_free() out of
125 the inner loop, so g_slice_free_chain_with_offset() provides a real
126 performance benefit over g_slice_free1() now.
128 * glib/gslice.h: renamed g_slice_free_chain() to
129 g_slice_free_chain_with_offset(). implemented g_slice_free_chain() as
130 a type-safe macro as suggested in bug 323178.
131 simplified the macro implementation of g_slice_free() and implemented
132 it in a type safe manner for all compliers as suggested by Morten
133 Welinder <mortenw@gnome.org>.
138 * glib/glib.symbols: s/g_slice_free_chain/g_slice_free_chain_with_offset/
140 2005-12-05 Matthias Clasen <mclasen@redhat.com>
142 * glib/gasyncqueue.c: Add some docs.
144 * tests/libmoduletestplugin_a.c: Fix compiler warnings.
146 * glib/gatomic.c: In the ia64 implementation, use
147 __sync builtin without _si or _di suffix. (#321229,
148 Stanislav Brabec, patch by Andreas Schwab)
150 2005-12-04 Behdad Esfahbod <behdad@gnome.org>
152 * glib/gslice.h: Remove comma at the end of enum.
154 2005-12-04 Matthias Clasen <mclasen@redhat.com>
156 Handle multiple user names with the same UID better.
157 (#319535, Laszlo Peter)
159 * glib/gutils.c (g_get_any_init_do): When determining user
160 data, first look up $LOGNAME. If the UID doesn't match
161 getuid(), fall back to the current behaviour of looking
162 up the user data based on getuid().
164 2005-12-04 Matthias Clasen <mclasen@redhat.com>
166 * glib/gmacros.h (G_GNUC_WARN_UNUSED_RESULT): Add a macro
167 to make gcc warn if a function result is ignored. (#145466,
168 Arjan van de Ven, Alex Larsson)
170 * glib/gmem.h: Add the new attribute to g_realloc and
173 2005-12-03 Matthias Clasen <mclasen@redhat.com>
176 * glib/gcache.h: Deprecate g_cache_value_foreach. (#322956,
179 * glib/gmappedfile.c: Make mapping of empty files
182 * glib/gfileutils.c: Don't fork a new process just to
183 fix the permissions of the created temp file. (#321318,
184 Alexis S. L. Carvalho)
186 2005-12-02 Matthias Clasen <mclasen@redhat.com>
188 * README.in: Add a note about Solaris threads.
191 * configure.in: Remove support for Solaris threads.
192 (#136971, Sebastian Wilhelmi, patch by Andrew Paprocki)
194 2005-12-02 Matthias Clasen <mclasen@redhat.com>
196 * configure.in: Define G_GUINT64_CONSTANT in analogy to
197 G_GINT64_CONSTANT. (#322568, Andrew Paprocki)
199 2005-12-02 Matthias Clasen <mclasen@redhat.com>
201 * glib/gslice.c: Win32 portability fixes and C99ism removal,
202 pointed out by Kazuki Iwamoto. (#323052)
204 2005-12-02 Matthias Clasen <mclasen@redhat.com>
206 * glib/gatomic.c (g_atomic_int_add): Add a missing volatile for
207 the IA64 implementation.
209 Fri Dec 2 16:18:09 2005 Tim Janik <timj@imendio.com>
211 * glib/gslice.c: some naming and type size fixups.
213 Fri Dec 2 13:08:58 2005 Tim Janik <timj@gtk.org>
215 * glib/gslice.h (g_slice_free): reworked GCC-specific type-safe macro
216 variant into something less verbose, but digestible for gcc-3.4.
218 Fri Dec 2 10:55:07 2005 Tim Janik <timj@imendio.com>
220 * tests/slice-test.c: extended to perform the benchmarking on the old
221 memchunk code if 'O' is selected.
223 * tests/memchunks.c: new file which contains the old GLib mem chunks
224 implementation with prefix old_mem_chunk_.
226 * tests/Makefile.am: added memchunks.c
228 Fri Dec 2 00:16:59 2005 Tim Janik <timj@imendio.com>
230 * glib/gslice.c: improved gettimeofday() timer resolution by maintaining
231 acceptable load distribution of the syscall.
232 (allocator_get_magazine_threshold): reduce minimum/single-thread magazine
233 sizes to approximately page_size/2.5 to avoid excessive startup allocations.
235 2005-12-01 Matthias Clasen <mclasen@redhat.com>
237 * tests/gobject/ifacecheck.c (test_iface_base_init): Don't
238 double-free base interface members.
240 * tests/Makefile.am: Remove duplicate variable.
242 Thu Dec 1 17:32:46 2005 Tim Janik <timj@imendio.com>
244 * glib/gslice.[hc]: new slice allocator implementation.
246 * tests/slice-test.c: added random slice allocation test.
248 * glib/gthread.[hc]: removed newly added private thread mem API.
250 * glib/gthreadinit.h:
253 * glib/gmem.c: divided glib threading initialisation into three phases,
254 initialisation where private keys and messaging are not available (only
255 needed by gmem.c), initialisation without messaging but private keys
256 available (gslice.c, gmessage.c), and full fledged initialisers that
257 server the rest of glib. initialisation functions got renamed to reflect
258 the limitations of their corresponding phases.
260 * glib/gmem.c: removed memchunk code, defer allocations to
263 * glib/gmem.[hc]: removed g_slice_* skeletons.
265 * glib/glib.symbols: added g_slice_* symbols.
267 * configure.in: check for availability of posix_memalign(3), memalign(3)
270 * glib/Makefile.am: added gslice.[hc].
272 2005-12-01 Tor Lillqvist <tml@novell.com>
274 * glib/gstdio.c (g_stat): In the Win32 implementation, strip
275 trailing slash(es) for non-root folders. stat() fails if non-root
276 folders are specified with trailing slashes. It's too much hassle
277 to demand that callers strip such slashes themselves, especially
278 as it is easy to get it wrong and strip the slash of a root
280 (g_rename): On NT-based Windows, use MoveFileEx() with
281 MOVEFILE_REPLACE_EXISTING to better match Unix behaviour.
283 2005-11-28 Matthias Clasen <mclasen@redhat.com>
285 Fix G_STMT_START / G_STMT_END on Solaris. (#321972,
288 * configure.in: Check whether do { } while (0) works.
290 * glib/gmacros.h: Use do { } while (0) for G_STMT_START /
291 G_STMT_END if it works.
293 2005-11-28 Matthias Clasen <mclasen@redhat.com>
295 * glib/gthread.c (g_static_rw_lock_wait, g_static_rw_lock_signal):
296 * glib/gnode.c (g_node_depth_traverse_level):
297 * glib/gmem.c (g_allocator_new):
298 * glib/ghash.c (g_hash_table_unref, g_hash_table_destroy)
299 (g_hash_table_foreach_remove_or_steal): Silence compiler
302 2005-11-27 Matthias Clasen <mclasen@redhat.com>
304 * glib/gunicollate.c (g_utf8_collate_key_for_filename):
305 Don't read beyond len. (#322520, Christian Persch)
307 Wed Nov 23 17:34:01 2005 Tim Janik <timj@imendio.com>
309 * glib/gdataset.c: access datalist flags via atomic pointer access
310 functions, instead of acquiring the dataset lock. this is faster and
311 also matches the atomic pointer readouts in gdatalistprivate.h and
312 direct pointer modifications required by gobject.c.
314 Wed Nov 23 13:35:31 2005 Tim Janik <timj@gtk.org>
316 * glib/gdataset.c: streamlined initialization code somewhat,
317 removed GData node cache.
319 2005-11-22 Matthias Clasen <mclasen@redhat.com>
321 * glib/ghash.c (g_hash_table_ref):
322 (g_hash_table_unref): Mark these as new API.
324 Avoid double locking in g_intern_string (#322133,
327 * glib/gdataset.c (g_quark_from_string_internal): New
328 internal function which factors out common parts of
329 g_quark_from[_static]_string.
330 (g_quark_from_string, g_quark_from_static_string):
331 Use g_quark_from_string_internal.
332 (g_intern_string, g_intern_static_string): Use
333 g_quark_from_string_internal, and only take the
335 (g_quark_new): Don't store the strings shifted by -1
336 in the g_quarks array.
337 (g_quark_to_string): Adapt to the previous change.
339 Tue Nov 22 14:04:26 2005 Tim Janik <timj@imendio.com>
343 g_hash_table_new_full(): create hash tables with a ref count of 1.
344 g_hash_table_ref(): atomically ref_count+=1
345 g_hash_table_unref(): atomically ref_count-=1, destroys hash table
346 when refcount reaches 0.
347 g_hash_table_destroy(): just destroy keys and values, unref by 1.
348 g_hash_table_insert():
349 g_hash_table_replace(): assert ref_count>0.
352 * glib/gatomic.c: added 'volatile' qualifier to all atomic pointer and
355 2005-11-20 Behdad Esfahbod <behdad@gnome.org>
357 * glib/guniprop.c (g_unichar_get_mirror_char): Remove unused
358 variables. (#321984, Andrew Paprocki)
360 2005-11-18 Matthias Clasen <mclasen@redhat.com>
362 * configure.in: Bump version
364 2005-11-17 Matthias Clasen <mclasen@redhat.com>
366 * === Released 2.9.0 ===
368 * glib/glib.symbols: Updates
370 * glib/gutf8.c (g_utf8_offset_to_pointer): Add some performance
371 hints to the docs. (Paolo Borelli)
375 * glib/gthread.c (g_thread_foreach): Mark as new api.
377 * README.in: Updates.
381 * glib/gdate.c (g_date_set_time_t):
382 (g_date_set_time_val): Add functions to set a date from
383 a time_t and from a GTimeVal.
384 (g_date_set_time): Deprecate. (#314067, Roger Leigh)
387 * tests/utf8-pointer.c: Unit tests for g_utf8_pointer_to_offset
388 and g_utf8_offset_to_pointer.
390 * glib/gutf8.c (g_utf8_pointer_to_offset)
391 (g_utf8_offset_to_pointer): Handle negative offsets, and use
392 "stutter stepping" for going backwards. (#320638, Larry
397 * glib/gthread.c: const correctness fixes, found
398 by Arjan van de Ven and gcc.
400 2005-11-16 Behdad Esfahbod <behdad@gnome.org>
402 * docs/reference/glib/tmpl/unicode.sgml: Correct typo on mentioning
403 @G_UNICODE_BREAK_UNKNOWN that should be #G_UNICODE_BREAK_UNKNOWN.
405 2005-11-09 Behdad Esfahbod <behdad@gnome.org>
407 * glib/guniprop.c: Use bit hacks instead when checking a general
408 category value against multiple values.
410 * glib/gutf8.c: Change ISO10646 to Unicode in docs.
412 2005-11-10 Simos Xenitellis <simos@gnome.org>
414 * configure.in: Added tt (Tatar) to ALL_LINGUAS.
416 2005-11-05 Tor Lillqvist <tml@novell.com>
418 * glib/giowin32.c (g_io_win32_set_flags): Don't call g_warning().
419 (#320688) Instead set error and return failure.
421 2005-11-07 Matthias Clasen <mclasen@redhat.com>
423 * glib/gmain.c (g_main_dispatch): Don't call cb_funcs->unref
424 while holding the context lock. (#320886, Andy Wingo)
426 2005-11-07 Billy Biggs <vektor@dumbterm.net>
428 * tests/refcount/Makefile.am: Add a missing reference to libglib
429 in the LDADD for this test.
431 2005-11-07 Matthias Clasen <mclasen@redhat.com>
433 * glib/gmem.c (g_allocator_new): We need to set n_preallocs to a
434 nonzero value, otherwise GTK+ 2.8 breaks when compiled against
435 GLib 2.9. (#320755, Luca Ferretti)
437 2005-11-04 Matthias Clasen <mclasen@redhat.com>
439 * glib/Makefile.am: Apply a patch from Behdad Esfahbod to
440 use a faster lookup table for g_unichar_get_mirror_char().
441 * glib/gmirroringtable.h: The new table.
442 * glib/gunichartables.h: Remove bidi_mirroring_table.
443 * glib/gen-unicode-tables.pl: Don't generate the mirroring
445 * glib/glib-mirroring-tab/*: Sources for the program
446 which generated gmirroringtable.h.
448 * glib/glist.c: Avoid some code duplication.
450 * glib/gscanner.h: Include gdataset.h. (#320322)
452 * glib/gdebug.h: add new GDebugFlag for fatal_criticals
453 * glib/gmessages.c: (_g_debug_init): handle G_DEBUG=fatal_criticals,
454 to help find critical warnings in applications. (#320017,
457 2005-11-02 Tor Lillqvist <tml@novell.com>
459 * glib/glib.symbols: Remove large amount of trailing whitespace
460 from one line. Remove #ifdef G_UNDEFINED from around
461 g_hash_table_new and g_hash_table_new_full symbols, what was the
462 purpose of that? Add the g_slice functions.
464 Tue Nov 1 16:24:20 2005 Tim Janik <timj@imendio.com>
466 * glib/gmem.[hc]: prepared deprecation of GMemChunk and GAllocator.
467 added g_slice_*() API to allocate and cache small bits of memory.
468 an actuall allocator implementation for g_slice_*() is still pending.
470 * glib/gthread.[hc]: changes from a patch by Matthias Clasen.
471 changed GRealThread list to use in-structure *next; fields instead
472 of GSList, in order for thread iteration to not depenend on g_slice_*()
474 _g_thread_mem_private_get():
475 _g_thread_mem_private_set(): added accessors for private memory,
476 needed because the ordinary GPrivate implementation relies on GArray
477 and GSList and therefore indirectly on working g_slice_*() allocations.
480 g_thread_foreach(): new public API function to loop over all existing threads.
494 removed GAllocator and free list usages and accompanying locks.
495 use g_slice_*() API to allocate and cache small bits of memory.
497 * glib/ghook.h: removed GMemChunk field from public API.
500 * glib/glist.h: deprecate allocator API, provide _free1() for consistency.
502 * glib/gnode.h: deprecate allocator API.
504 * glib/gmain.c: reordered GPollRec fields so g_slice_free_chain() can
505 be used for poll rec lists.
507 * glib/grel.c: removed mem chunk usage, and allocated tuples via g_slice_*().
508 g_relation_destroy(): free all tuples from the all_tuples hash table,
509 this effectively maintains the life time track keeping of tuples.
510 g_relation_delete_tuple(): free tuples which are removed from the
511 all_tuples hash table. this fixes a temporary leak that was present
512 in the memchunk code until the destruction of the relation.
514 2005-10-29 Matthias Clasen <mclasen@redhat.com>
516 * tests/convert-test.c: Add some tests for conversions between
517 UTF-8, UCS-4 and UTF-16.
519 * glib/gutf8.c (g_utf8_to_ucs4, g_utf8_to_utf16): Fix handling
520 of len == -1, noticed by Morten Welinder.
522 2005-10-27 Erdal Ronahi <erdal.ronahi@gmail.com>
524 * configure.in: Added ku (Kurdish) to ALL_LINGUAS
526 2005-10-26 Matthias Clasen <mclasen@redhat.com>
528 * glib/gutf8.c (g_ucs4_to_utf8): Don't set items_read twice
529 in the error case, and add some documentation. (#319806, Morten
532 2005-10-19 Manish Singh <yosh@gimp.org>
534 * configure.in: Use AC_CHECK_FUNCS for _NSGetEnviron, to get the
535 config.h symbol automatically. Fixes bug #313731.
537 2005-10-19 Tor Lillqvist <tml@novell.com>
539 * glib/gwin32.c (g_win32_get_package_installation_directory):
540 Return a g_strdup()ed copy of the value stored in the hash table,
541 so that it can be g_free()d without leaving a dangling pointer in
542 the hash table. (#319232)
544 2005-10-06 Matthias Clasen <mclasen@redhat.com>
546 * glib/gunicollate.c (g_utf8_collate_key_for_filename): Handle
547 all-zero sequences correctly. (#317930, Sebastien Bacher)
549 2005-10-05 Matthias Clasen <mclasen@redhat.com>
551 * glib/gmarkup.c (g_markup_escape_text): Doc addition
553 * Makefile.am: Add Collation tests.
555 * tests/collate/*: Inputs and expected outputs for collation tests.
557 * tests/run-collate-tests.sh: Script to run collation tests.
559 * tests/unicode-collate.c (main): Rework slightly to make
560 it usable in unit tests. Also test g_utf8_collate_key_for_filename().
562 2005-10-01 Behdad Esfahbod <behdad@gnome.org>
564 * docs/reference/glib/tmpl/unicode.sgml:
565 * glib/gen-unicode-tables.pl:
567 * glib/gunichartables.h:
569 * tests/casefold.txt:
570 * tests/casemap.txt: Updated to Unicode 4.1. There are five new
571 GUnicodeBreakType types. That may break some applications, like
574 2005-09-26 Matthias Clasen <mclasen@redhat.com>
576 * glib/gstrfuncs.c (g_ascii_strtoull): Add details to the
577 docs. (#314393, Matthew F. Barnes)
580 * glib/gprintf.h: Remove g_snprintf() and g_vsnprintf(), since
581 they are already declared in glib.h. This doesn't break documented
582 use of gprintf.h, but should probably be pointed out in the
583 release notes for 2.10. (#314232, Behdad Esfahbod)
585 Tue Sep 20 13:16:04 2005 Tim Janik <timj@imendio.com>
587 * glib/gpattern.c (g_pattern_ph_match): applied significant recursion
588 complexity optimization, based on a patch from Matthias Clasen.
590 * tests/patterntest.c: more tests, mostly from matthias.
592 2005-09-20 Matthias Clasen <mclasen@redhat.com>
594 * glib/gqueue.c (g_queue_insert_sorted): Correct the docs.
595 (#316703, Mark Drago)
597 Mon Sep 19 17:23:23 2005 Tim Janik <timj@imendio.com>
599 * glib/gpattern.c: applied a patch from matthias which checks on the
600 upper bound of GPatternSpec length to optimize matches.
603 * tests/patterntest.c: added more match cases.
605 2005-09-16 Tor Lillqvist <tml@novell.com>
607 * glib/gstrfuncs.c (g_ascii_strcasecmp, g_ascii_strncasecmp): Add
608 warning to doc comment that these functions should not be used on
609 encodings like CP932.
611 2005-09-14 Matthias Clasen <mclasen@redhat.com>
613 * tests/keyfile-test.c: Add a test for grup names of length 1.
615 * glib/gkeyfile.c (g_key_file_line_is_group): Accept group names
616 of length 1. (#316309)
618 2005-09-12 Matthias Clasen <mclasen@redhat.com>
620 * glib/gmarkup.c (g_markup_escape_text): Clarify docs.
622 2005-09-11 Sebastian Wilhelmi <seppi@seppi.de>
624 * tests/refcount/Makefile.am (INCLUDES): Link the the refcount
625 tests to the system thread library $(G_THREAD_LIBS). Fixes #313744
628 2005-09-11 Kjartan Maraas <kmaraas@gnome.org>
630 * glib/gmain.c: (g_child_watch_prepare), (g_child_watch_check),
631 (child_watch_helper_thread): Remove some dead code. Closes
634 2005-09-07 Tor Lillqvist <tml@novell.com>
636 * glib/Makefile.am: Create also a console version of the
637 gspawn-win32-helper program, gspawn-win32-helper-console.exe.
638 It's otherwise identical to gspawn-win32-helper.exe, except marked
639 as a console application (linked without the -mwindows option).
641 * glib/gspawn-win32.c (do_spawn_directly, do_spawn_with_pipes):
642 Drop the dont_wait parameter. Its truth value correlated 100% with
643 the NULLness of the exit_status parameter anyway, so it's enough
644 to check whether exit_status is NULL. Invert the sense of the
645 dont_return_handle parameter and rename it to do_return_handle, to
646 make the code easier to read by avoiding double negations.
648 (g_spawn_sync_utf8, g_spawn_async_with_pipes_utf8): Modify calls
649 to do_spawn_with_pipes() accordingly.
651 (do_spawn_with_pipes): If we have a console, use the console
652 version of the helper program, otherwise use the GUI one. This
653 avoids extra console windows opening up in some situations. (In
654 case a console application uses the GUI gspawn-win32-helper.exe to
655 spawn another console application we would get a separate console
656 for the spawned console application).
658 * glib-zip.in: Distribute also gspawn-win32-helper-console.exe.
660 2005-09-05 Matthias Clasen <mclasen@redhat.com>
662 * glib/gmappedfile.c (g_mapped_file_new): Report an error
663 if the file is too large. (#315275, Kjartan Maraas)
665 * glib/gkeyfile.c (g_key_file_load_from_fd): The return value
666 of read() is signed. (#315273, Kjartan Maraas)
668 2005-08-31 Tor Lillqvist <tml@novell.com>
670 * glib/gutils.h: Wrapping atexit() is a bad idea on Windows, where
671 the EXE and each DLL have their own atexit function chains.
673 #define g_atexit as atexit instead. This means it has a
674 better chance of doing what the caller wants. For instance,
675 gtkhtml calls g_atexit() registering a function in gtkhtml
676 itself. This caused a crash when g_atexit() was implemented as a
677 function in the GLib DLL. The gtkhtml DLL was already unloaded by
678 the time the GLib DLL got unloaded.
680 * glib/gutils.c: #undef the #define mentioned above, to also get a
681 real g_atexit() into the DLL for backward compatibility. Document
682 the Windows behaviour of g_atexit(), and document the varying ways
683 atexit() can behave in the context of dynamically loaded modules
686 2005-08-31 Matthias Clasen <mclasen@redhat.com>
690 * glib/gdataset.c: Add string interning functions.
692 2005-08-28 Matthias Clasen <mclasen@redhat.com>
694 * glib/giochannel.c: Unify some near-duplicate strings. (#314654,
697 2005-08-26 Matthias Clasen <mclasen@redhat.com>
699 * configure.in: Bump version to 2.9.0
701 2005-08-25 Tor Lillqvist <tml@novell.com>
703 Make also the g_spawn*() functions take parameters in the GLib
704 file name encoding, i.e. UTF-8, on Windows. Has no impact on Unix
705 API or ABI. Like the other GLib API that was earlier changed to
706 use UTF-8 on Windows, the names of the functions that take UTF-8
707 have _utf8 suffixes added by using preprocessor macros in the
708 header file. The old names are kept for functions with the old
709 behaviour, taking parameters in the system codepage, for DLL ABI
712 * glib/gspawn.h: On Win32 add the suffix _utf8 to the names of the
713 g_spawn*() functions.
715 * glib/gspawn-win32.c: Use wide-char API on NT-based
716 Windows. Convert parameters from UTF-8 to wide chars (NT) or
717 system codepage (Win9x) and call the C library _wspawn*() or
718 spawn*() functions respectvely. Add DLL ABI stability versions
719 that take parameters in the system codepage.
721 * glib/gspawn-win32-helper.c: On NT-based Windows use the
722 wide-char versions of argv and envp, and use wide-char API to
723 change directory and spawn the program to run. Remove the verbose
724 debugging output, it was too complex to modify for the wide-char
725 features. (Just add temporary debugging printouts if needed, no
726 need to have them permanently in the source.)
728 * glib/gspawn.c: Corresponding documentation updates.
730 * glib/glib.symbols: Corresponding changes: Mark the ABI stability
731 symbols as PRIVATE, add the new _utf8-suffixed ones.
733 2005-08-24 Stepan Kasal <kasal@ucw.cz>
735 * glib/gtypes.h (G_MININT64): Cast the constant to gint64; it is
736 guint64 otherwise and that can produce warnings about comparison
737 between signed and unsigned.
739 2005-08-23 Matthias Clasen <mclasen@redhat.com>
741 * glib/gutils.c: Fix the crt_externs.h include.
743 2005-08-23 Stepan Kasal <kasal@ucw.cz>
745 * NEWS: Fix spelling of my first name.
747 2005-08-23 Matthias Clasen <mclasen@redhat.com>
751 * === Released 2.8.1 ===
755 2005-08-20 Hans Breuer <hans@breuer.org>
757 * glib/makefile.msc.in : link with ws2_32.lib
759 2005-08-18 Tor Lillqvist <tml@novell.com>
761 * configure.in: Check for <sys/wait.h>
763 * glib/gbacktrace.c: Include <sys/wait.h> on if HAVE_SYS_WAIT_H.
765 2005-08-18 Ross Burton <ross@burtonini.com>
768 Optimise single-character insertions.
773 * tests/string-test.c:
774 Add tests for new optimisation, and fix a leak.
776 2005-08-17 Matthias Clasen <mclasen@redhat.com>
778 * configure.in: Check for crt_externs.h and _NSGetEnviron.
780 * glib/gutils.c: On Darwin, include crt-externs.h and
781 define environ using _NSGetEnviron(). (#313731)
783 2005-08-16 Stepan Kasal <kasal@ucw.cz>
785 * glib/gutils.c (g_get_any_init): Move the body of the big if...
786 (g_get_any_init_do): ... to this new function.
787 (g_get_any_init): Declare as inline.
788 (g_get_any_init_locked): New inline function, does the locking.
789 Make use of these two throughout the code.
791 2005-08-15 Matthias Clasen <mclasen@redhat.com>
793 * glib/gbacktrace.c (g_on_error_stack_trace): Wait for
794 the child process and then simply return. This makes
795 The "S" option work as documented in g_on_error_query().
796 (#313125, Matthew F. Barnes)
798 * glib/gunicode.h: Update the link to Unicode category
799 values. (#313369, Behnam Esfahbod)
801 * glib/gqueue.c (g_queue_find_custom): Clarify docs
802 a little. (#311727, Tristan van Berkom)
804 * glib/abicheck.sh, gobject/abicheck.sh: Make the
805 check work on ia64 too, where some symbols ended up
806 in yet another section.