1 Mon Dec 5 15:53:20 2005 Tim Janik <timj@imendio.com>
3 * glib/gslice.c: implement chain walking for arbitrary ->next pointer
4 offsets in g_slice_free_chain_with_offset() based on a patch by behdad
5 in bug 323178. moved time consuming logic from g_slice_free() out of
6 the inner loop, so g_slice_free_chain_with_offset() provides a real
7 performance benefit over g_slice_free1() now.
9 * glib/gslice.h: renamed g_slice_free_chain() to
10 g_slice_free_chain_with_offset(). implemented g_slice_free_chain() as
11 a type-safe macro as suggested in bug 323178.
12 simplified the macro implementation of g_slice_free() and implemented
13 it in a type safe manner for all compliers as suggested by Morten
14 Welinder <mortenw@gnome.org>.
19 * glib/glib.symbols: s/g_slice_free_chain/g_slice_free_chain_with_offset/
21 2005-12-05 Matthias Clasen <mclasen@redhat.com>
23 * glib/gasyncqueue.c: Add some docs.
25 * tests/libmoduletestplugin_a.c: Fix compiler warnings.
27 * glib/gatomic.c: In the ia64 implementation, use
28 __sync builtin without _si or _di suffix. (#321229,
29 Stanislav Brabec, patch by Andreas Schwab)
31 2005-12-04 Behdad Esfahbod <behdad@gnome.org>
33 * glib/gslice.h: Remove comma at the end of enum.
35 2005-12-04 Matthias Clasen <mclasen@redhat.com>
37 Handle multiple user names with the same UID better.
38 (#319535, Laszlo Peter)
40 * glib/gutils.c (g_get_any_init_do): When determining user
41 data, first look up $LOGNAME. If the UID doesn't match
42 getuid(), fall back to the current behaviour of looking
43 up the user data based on getuid().
45 2005-12-04 Matthias Clasen <mclasen@redhat.com>
47 * glib/gmacros.h (G_GNUC_WARN_UNUSED_RESULT): Add a macro
48 to make gcc warn if a function result is ignored. (#145466,
49 Arjan van de Ven, Alex Larsson)
51 * glib/gmem.h: Add the new attribute to g_realloc and
54 2005-12-03 Matthias Clasen <mclasen@redhat.com>
57 * glib/gcache.h: Deprecate g_cache_value_foreach. (#322956,
60 * glib/gmappedfile.c: Make mapping of empty files
63 * glib/gfileutils.c: Don't fork a new process just to
64 fix the permissions of the created temp file. (#321318,
65 Alexis S. L. Carvalho)
67 2005-12-02 Matthias Clasen <mclasen@redhat.com>
69 * README.in: Add a note about Solaris threads.
72 * configure.in: Remove support for Solaris threads.
73 (#136971, Sebastian Wilhelmi, patch by Andrew Paprocki)
75 2005-12-02 Matthias Clasen <mclasen@redhat.com>
77 * configure.in: Define G_GUINT64_CONSTANT in analogy to
78 G_GINT64_CONSTANT. (#322568, Andrew Paprocki)
80 2005-12-02 Matthias Clasen <mclasen@redhat.com>
82 * glib/gslice.c: Win32 portability fixes and C99ism removal,
83 pointed out by Kazuki Iwamoto. (#323052)
85 2005-12-02 Matthias Clasen <mclasen@redhat.com>
87 * glib/gatomic.c (g_atomic_int_add): Add a missing volatile for
88 the IA64 implementation.
90 Fri Dec 2 16:18:09 2005 Tim Janik <timj@imendio.com>
92 * glib/gslice.c: some naming and type size fixups.
94 Fri Dec 2 13:08:58 2005 Tim Janik <timj@gtk.org>
96 * glib/gslice.h (g_slice_free): reworked GCC-specific type-safe macro
97 variant into something less verbose, but digestible for gcc-3.4.
99 Fri Dec 2 10:55:07 2005 Tim Janik <timj@imendio.com>
101 * tests/slice-test.c: extended to perform the benchmarking on the old
102 memchunk code if 'O' is selected.
104 * tests/memchunks.c: new file which contains the old GLib mem chunks
105 implementation with prefix old_mem_chunk_.
107 * tests/Makefile.am: added memchunks.c
109 Fri Dec 2 00:16:59 2005 Tim Janik <timj@imendio.com>
111 * glib/gslice.c: improved gettimeofday() timer resolution by maintaining
112 acceptable load distribution of the syscall.
113 (allocator_get_magazine_threshold): reduce minimum/single-thread magazine
114 sizes to approximately page_size/2.5 to avoid excessive startup allocations.
116 2005-12-01 Matthias Clasen <mclasen@redhat.com>
118 * tests/gobject/ifacecheck.c (test_iface_base_init): Don't
119 double-free base interface members.
121 * tests/Makefile.am: Remove duplicate variable.
123 Thu Dec 1 17:32:46 2005 Tim Janik <timj@imendio.com>
125 * glib/gslice.[hc]: new slice allocator implementation.
127 * tests/slice-test.c: added random slice allocation test.
129 * glib/gthread.[hc]: removed newly added private thread mem API.
131 * glib/gthreadinit.h:
134 * glib/gmem.c: divided glib threading initialisation into three phases,
135 initialisation where private keys and messaging are not available (only
136 needed by gmem.c), initialisation without messaging but private keys
137 available (gslice.c, gmessage.c), and full fledged initialisers that
138 server the rest of glib. initialisation functions got renamed to reflect
139 the limitations of their corresponding phases.
141 * glib/gmem.c: removed memchunk code, defer allocations to
144 * glib/gmem.[hc]: removed g_slice_* skeletons.
146 * glib/glib.symbols: added g_slice_* symbols.
148 * configure.in: check for availability of posix_memalign(3), memalign(3)
151 * glib/Makefile.am: added gslice.[hc].
153 2005-12-01 Tor Lillqvist <tml@novell.com>
155 * glib/gstdio.c (g_stat): In the Win32 implementation, strip
156 trailing slash(es) for non-root folders. stat() fails if non-root
157 folders are specified with trailing slashes. It's too much hassle
158 to demand that callers strip such slashes themselves, especially
159 as it is easy to get it wrong and strip the slash of a root
161 (g_rename): On NT-based Windows, use MoveFileEx() with
162 MOVEFILE_REPLACE_EXISTING to better match Unix behaviour.
164 2005-11-28 Matthias Clasen <mclasen@redhat.com>
166 Fix G_STMT_START / G_STMT_END on Solaris. (#321972,
169 * configure.in: Check whether do { } while (0) works.
171 * glib/gmacros.h: Use do { } while (0) for G_STMT_START /
172 G_STMT_END if it works.
174 2005-11-28 Matthias Clasen <mclasen@redhat.com>
176 * glib/gthread.c (g_static_rw_lock_wait, g_static_rw_lock_signal):
177 * glib/gnode.c (g_node_depth_traverse_level):
178 * glib/gmem.c (g_allocator_new):
179 * glib/ghash.c (g_hash_table_unref, g_hash_table_destroy)
180 (g_hash_table_foreach_remove_or_steal): Silence compiler
183 2005-11-27 Matthias Clasen <mclasen@redhat.com>
185 * glib/gunicollate.c (g_utf8_collate_key_for_filename):
186 Don't read beyond len. (#322520, Christian Persch)
188 Wed Nov 23 17:34:01 2005 Tim Janik <timj@imendio.com>
190 * glib/gdataset.c: access datalist flags via atomic pointer access
191 functions, instead of acquiring the dataset lock. this is faster and
192 also matches the atomic pointer readouts in gdatalistprivate.h and
193 direct pointer modifications required by gobject.c.
195 Wed Nov 23 13:35:31 2005 Tim Janik <timj@gtk.org>
197 * glib/gdataset.c: streamlined initialization code somewhat,
198 removed GData node cache.
200 2005-11-22 Matthias Clasen <mclasen@redhat.com>
202 * glib/ghash.c (g_hash_table_ref):
203 (g_hash_table_unref): Mark these as new API.
205 Avoid double locking in g_intern_string (#322133,
208 * glib/gdataset.c (g_quark_from_string_internal): New
209 internal function which factors out common parts of
210 g_quark_from[_static]_string.
211 (g_quark_from_string, g_quark_from_static_string):
212 Use g_quark_from_string_internal.
213 (g_intern_string, g_intern_static_string): Use
214 g_quark_from_string_internal, and only take the
216 (g_quark_new): Don't store the strings shifted by -1
217 in the g_quarks array.
218 (g_quark_to_string): Adapt to the previous change.
220 Tue Nov 22 14:04:26 2005 Tim Janik <timj@imendio.com>
224 g_hash_table_new_full(): create hash tables with a ref count of 1.
225 g_hash_table_ref(): atomically ref_count+=1
226 g_hash_table_unref(): atomically ref_count-=1, destroys hash table
227 when refcount reaches 0.
228 g_hash_table_destroy(): just destroy keys and values, unref by 1.
229 g_hash_table_insert():
230 g_hash_table_replace(): assert ref_count>0.
233 * glib/gatomic.c: added 'volatile' qualifier to all atomic pointer and
236 2005-11-20 Behdad Esfahbod <behdad@gnome.org>
238 * glib/guniprop.c (g_unichar_get_mirror_char): Remove unused
239 variables. (#321984, Andrew Paprocki)
241 2005-11-18 Matthias Clasen <mclasen@redhat.com>
243 * configure.in: Bump version
245 2005-11-17 Matthias Clasen <mclasen@redhat.com>
247 * === Released 2.9.0 ===
249 * glib/glib.symbols: Updates
251 * glib/gutf8.c (g_utf8_offset_to_pointer): Add some performance
252 hints to the docs. (Paolo Borelli)
256 * glib/gthread.c (g_thread_foreach): Mark as new api.
258 * README.in: Updates.
262 * glib/gdate.c (g_date_set_time_t):
263 (g_date_set_time_val): Add functions to set a date from
264 a time_t and from a GTimeVal.
265 (g_date_set_time): Deprecate. (#314067, Roger Leigh)
268 * tests/utf8-pointer.c: Unit tests for g_utf8_pointer_to_offset
269 and g_utf8_offset_to_pointer.
271 * glib/gutf8.c (g_utf8_pointer_to_offset)
272 (g_utf8_offset_to_pointer): Handle negative offsets, and use
273 "stutter stepping" for going backwards. (#320638, Larry
278 * glib/gthread.c: const correctness fixes, found
279 by Arjan van de Ven and gcc.
281 2005-11-16 Behdad Esfahbod <behdad@gnome.org>
283 * docs/reference/glib/tmpl/unicode.sgml: Correct typo on mentioning
284 @G_UNICODE_BREAK_UNKNOWN that should be #G_UNICODE_BREAK_UNKNOWN.
286 2005-11-09 Behdad Esfahbod <behdad@gnome.org>
288 * glib/guniprop.c: Use bit hacks instead when checking a general
289 category value against multiple values.
291 * glib/gutf8.c: Change ISO10646 to Unicode in docs.
293 2005-11-10 Simos Xenitellis <simos@gnome.org>
295 * configure.in: Added tt (Tatar) to ALL_LINGUAS.
297 2005-11-05 Tor Lillqvist <tml@novell.com>
299 * glib/giowin32.c (g_io_win32_set_flags): Don't call g_warning().
300 (#320688) Instead set error and return failure.
302 2005-11-07 Matthias Clasen <mclasen@redhat.com>
304 * glib/gmain.c (g_main_dispatch): Don't call cb_funcs->unref
305 while holding the context lock. (#320886, Andy Wingo)
307 2005-11-07 Billy Biggs <vektor@dumbterm.net>
309 * tests/refcount/Makefile.am: Add a missing reference to libglib
310 in the LDADD for this test.
312 2005-11-07 Matthias Clasen <mclasen@redhat.com>
314 * glib/gmem.c (g_allocator_new): We need to set n_preallocs to a
315 nonzero value, otherwise GTK+ 2.8 breaks when compiled against
316 GLib 2.9. (#320755, Luca Ferretti)
318 2005-11-04 Matthias Clasen <mclasen@redhat.com>
320 * glib/Makefile.am: Apply a patch from Behdad Esfahbod to
321 use a faster lookup table for g_unichar_get_mirror_char().
322 * glib/gmirroringtable.h: The new table.
323 * glib/gunichartables.h: Remove bidi_mirroring_table.
324 * glib/gen-unicode-tables.pl: Don't generate the mirroring
326 * glib/glib-mirroring-tab/*: Sources for the program
327 which generated gmirroringtable.h.
329 * glib/glist.c: Avoid some code duplication.
331 * glib/gscanner.h: Include gdataset.h. (#320322)
333 * glib/gdebug.h: add new GDebugFlag for fatal_criticals
334 * glib/gmessages.c: (_g_debug_init): handle G_DEBUG=fatal_criticals,
335 to help find critical warnings in applications. (#320017,
338 2005-11-02 Tor Lillqvist <tml@novell.com>
340 * glib/glib.symbols: Remove large amount of trailing whitespace
341 from one line. Remove #ifdef G_UNDEFINED from around
342 g_hash_table_new and g_hash_table_new_full symbols, what was the
343 purpose of that? Add the g_slice functions.
345 Tue Nov 1 16:24:20 2005 Tim Janik <timj@imendio.com>
347 * glib/gmem.[hc]: prepared deprecation of GMemChunk and GAllocator.
348 added g_slice_*() API to allocate and cache small bits of memory.
349 an actuall allocator implementation for g_slice_*() is still pending.
351 * glib/gthread.[hc]: changes from a patch by Matthias Clasen.
352 changed GRealThread list to use in-structure *next; fields instead
353 of GSList, in order for thread iteration to not depenend on g_slice_*()
355 _g_thread_mem_private_get():
356 _g_thread_mem_private_set(): added accessors for private memory,
357 needed because the ordinary GPrivate implementation relies on GArray
358 and GSList and therefore indirectly on working g_slice_*() allocations.
361 g_thread_foreach(): new public API function to loop over all existing threads.
375 removed GAllocator and free list usages and accompanying locks.
376 use g_slice_*() API to allocate and cache small bits of memory.
378 * glib/ghook.h: removed GMemChunk field from public API.
381 * glib/glist.h: deprecate allocator API, provide _free1() for consistency.
383 * glib/gnode.h: deprecate allocator API.
385 * glib/gmain.c: reordered GPollRec fields so g_slice_free_chain() can
386 be used for poll rec lists.
388 * glib/grel.c: removed mem chunk usage, and allocated tuples via g_slice_*().
389 g_relation_destroy(): free all tuples from the all_tuples hash table,
390 this effectively maintains the life time track keeping of tuples.
391 g_relation_delete_tuple(): free tuples which are removed from the
392 all_tuples hash table. this fixes a temporary leak that was present
393 in the memchunk code until the destruction of the relation.
395 2005-10-29 Matthias Clasen <mclasen@redhat.com>
397 * tests/convert-test.c: Add some tests for conversions between
398 UTF-8, UCS-4 and UTF-16.
400 * glib/gutf8.c (g_utf8_to_ucs4, g_utf8_to_utf16): Fix handling
401 of len == -1, noticed by Morten Welinder.
403 2005-10-27 Erdal Ronahi <erdal.ronahi@gmail.com>
405 * configure.in: Added ku (Kurdish) to ALL_LINGUAS
407 2005-10-26 Matthias Clasen <mclasen@redhat.com>
409 * glib/gutf8.c (g_ucs4_to_utf8): Don't set items_read twice
410 in the error case, and add some documentation. (#319806, Morten
413 2005-10-19 Manish Singh <yosh@gimp.org>
415 * configure.in: Use AC_CHECK_FUNCS for _NSGetEnviron, to get the
416 config.h symbol automatically. Fixes bug #313731.
418 2005-10-19 Tor Lillqvist <tml@novell.com>
420 * glib/gwin32.c (g_win32_get_package_installation_directory):
421 Return a g_strdup()ed copy of the value stored in the hash table,
422 so that it can be g_free()d without leaving a dangling pointer in
423 the hash table. (#319232)
425 2005-10-06 Matthias Clasen <mclasen@redhat.com>
427 * glib/gunicollate.c (g_utf8_collate_key_for_filename): Handle
428 all-zero sequences correctly. (#317930, Sebastien Bacher)
430 2005-10-05 Matthias Clasen <mclasen@redhat.com>
432 * glib/gmarkup.c (g_markup_escape_text): Doc addition
434 * Makefile.am: Add Collation tests.
436 * tests/collate/*: Inputs and expected outputs for collation tests.
438 * tests/run-collate-tests.sh: Script to run collation tests.
440 * tests/unicode-collate.c (main): Rework slightly to make
441 it usable in unit tests. Also test g_utf8_collate_key_for_filename().
443 2005-10-01 Behdad Esfahbod <behdad@gnome.org>
445 * docs/reference/glib/tmpl/unicode.sgml:
446 * glib/gen-unicode-tables.pl:
448 * glib/gunichartables.h:
450 * tests/casefold.txt:
451 * tests/casemap.txt: Updated to Unicode 4.1. There are five new
452 GUnicodeBreakType types. That may break some applications, like
455 2005-09-26 Matthias Clasen <mclasen@redhat.com>
457 * glib/gstrfuncs.c (g_ascii_strtoull): Add details to the
458 docs. (#314393, Matthew F. Barnes)
461 * glib/gprintf.h: Remove g_snprintf() and g_vsnprintf(), since
462 they are already declared in glib.h. This doesn't break documented
463 use of gprintf.h, but should probably be pointed out in the
464 release notes for 2.10. (#314232, Behdad Esfahbod)
466 Tue Sep 20 13:16:04 2005 Tim Janik <timj@imendio.com>
468 * glib/gpattern.c (g_pattern_ph_match): applied significant recursion
469 complexity optimization, based on a patch from Matthias Clasen.
471 * tests/patterntest.c: more tests, mostly from matthias.
473 2005-09-20 Matthias Clasen <mclasen@redhat.com>
475 * glib/gqueue.c (g_queue_insert_sorted): Correct the docs.
476 (#316703, Mark Drago)
478 Mon Sep 19 17:23:23 2005 Tim Janik <timj@imendio.com>
480 * glib/gpattern.c: applied a patch from matthias which checks on the
481 upper bound of GPatternSpec length to optimize matches.
484 * tests/patterntest.c: added more match cases.
486 2005-09-16 Tor Lillqvist <tml@novell.com>
488 * glib/gstrfuncs.c (g_ascii_strcasecmp, g_ascii_strncasecmp): Add
489 warning to doc comment that these functions should not be used on
490 encodings like CP932.
492 2005-09-14 Matthias Clasen <mclasen@redhat.com>
494 * tests/keyfile-test.c: Add a test for grup names of length 1.
496 * glib/gkeyfile.c (g_key_file_line_is_group): Accept group names
497 of length 1. (#316309)
499 2005-09-12 Matthias Clasen <mclasen@redhat.com>
501 * glib/gmarkup.c (g_markup_escape_text): Clarify docs.
503 2005-09-11 Sebastian Wilhelmi <seppi@seppi.de>
505 * tests/refcount/Makefile.am (INCLUDES): Link the the refcount
506 tests to the system thread library $(G_THREAD_LIBS). Fixes #313744
509 2005-09-11 Kjartan Maraas <kmaraas@gnome.org>
511 * glib/gmain.c: (g_child_watch_prepare), (g_child_watch_check),
512 (child_watch_helper_thread): Remove some dead code. Closes
515 2005-09-07 Tor Lillqvist <tml@novell.com>
517 * glib/Makefile.am: Create also a console version of the
518 gspawn-win32-helper program, gspawn-win32-helper-console.exe.
519 It's otherwise identical to gspawn-win32-helper.exe, except marked
520 as a console application (linked without the -mwindows option).
522 * glib/gspawn-win32.c (do_spawn_directly, do_spawn_with_pipes):
523 Drop the dont_wait parameter. Its truth value correlated 100% with
524 the NULLness of the exit_status parameter anyway, so it's enough
525 to check whether exit_status is NULL. Invert the sense of the
526 dont_return_handle parameter and rename it to do_return_handle, to
527 make the code easier to read by avoiding double negations.
529 (g_spawn_sync_utf8, g_spawn_async_with_pipes_utf8): Modify calls
530 to do_spawn_with_pipes() accordingly.
532 (do_spawn_with_pipes): If we have a console, use the console
533 version of the helper program, otherwise use the GUI one. This
534 avoids extra console windows opening up in some situations. (In
535 case a console application uses the GUI gspawn-win32-helper.exe to
536 spawn another console application we would get a separate console
537 for the spawned console application).
539 * glib-zip.in: Distribute also gspawn-win32-helper-console.exe.
541 2005-09-05 Matthias Clasen <mclasen@redhat.com>
543 * glib/gmappedfile.c (g_mapped_file_new): Report an error
544 if the file is too large. (#315275, Kjartan Maraas)
546 * glib/gkeyfile.c (g_key_file_load_from_fd): The return value
547 of read() is signed. (#315273, Kjartan Maraas)
549 2005-08-31 Tor Lillqvist <tml@novell.com>
551 * glib/gutils.h: Wrapping atexit() is a bad idea on Windows, where
552 the EXE and each DLL have their own atexit function chains.
554 #define g_atexit as atexit instead. This means it has a
555 better chance of doing what the caller wants. For instance,
556 gtkhtml calls g_atexit() registering a function in gtkhtml
557 itself. This caused a crash when g_atexit() was implemented as a
558 function in the GLib DLL. The gtkhtml DLL was already unloaded by
559 the time the GLib DLL got unloaded.
561 * glib/gutils.c: #undef the #define mentioned above, to also get a
562 real g_atexit() into the DLL for backward compatibility. Document
563 the Windows behaviour of g_atexit(), and document the varying ways
564 atexit() can behave in the context of dynamically loaded modules
567 2005-08-31 Matthias Clasen <mclasen@redhat.com>
571 * glib/gdataset.c: Add string interning functions.
573 2005-08-28 Matthias Clasen <mclasen@redhat.com>
575 * glib/giochannel.c: Unify some near-duplicate strings. (#314654,
578 2005-08-26 Matthias Clasen <mclasen@redhat.com>
580 * configure.in: Bump version to 2.9.0
582 2005-08-25 Tor Lillqvist <tml@novell.com>
584 Make also the g_spawn*() functions take parameters in the GLib
585 file name encoding, i.e. UTF-8, on Windows. Has no impact on Unix
586 API or ABI. Like the other GLib API that was earlier changed to
587 use UTF-8 on Windows, the names of the functions that take UTF-8
588 have _utf8 suffixes added by using preprocessor macros in the
589 header file. The old names are kept for functions with the old
590 behaviour, taking parameters in the system codepage, for DLL ABI
593 * glib/gspawn.h: On Win32 add the suffix _utf8 to the names of the
594 g_spawn*() functions.
596 * glib/gspawn-win32.c: Use wide-char API on NT-based
597 Windows. Convert parameters from UTF-8 to wide chars (NT) or
598 system codepage (Win9x) and call the C library _wspawn*() or
599 spawn*() functions respectvely. Add DLL ABI stability versions
600 that take parameters in the system codepage.
602 * glib/gspawn-win32-helper.c: On NT-based Windows use the
603 wide-char versions of argv and envp, and use wide-char API to
604 change directory and spawn the program to run. Remove the verbose
605 debugging output, it was too complex to modify for the wide-char
606 features. (Just add temporary debugging printouts if needed, no
607 need to have them permanently in the source.)
609 * glib/gspawn.c: Corresponding documentation updates.
611 * glib/glib.symbols: Corresponding changes: Mark the ABI stability
612 symbols as PRIVATE, add the new _utf8-suffixed ones.
614 2005-08-24 Stepan Kasal <kasal@ucw.cz>
616 * glib/gtypes.h (G_MININT64): Cast the constant to gint64; it is
617 guint64 otherwise and that can produce warnings about comparison
618 between signed and unsigned.
620 2005-08-23 Matthias Clasen <mclasen@redhat.com>
622 * glib/gutils.c: Fix the crt_externs.h include.
624 2005-08-23 Stepan Kasal <kasal@ucw.cz>
626 * NEWS: Fix spelling of my first name.
628 2005-08-23 Matthias Clasen <mclasen@redhat.com>
632 * === Released 2.8.1 ===
636 2005-08-20 Hans Breuer <hans@breuer.org>
638 * glib/makefile.msc.in : link with ws2_32.lib
640 2005-08-18 Tor Lillqvist <tml@novell.com>
642 * configure.in: Check for <sys/wait.h>
644 * glib/gbacktrace.c: Include <sys/wait.h> on if HAVE_SYS_WAIT_H.
646 2005-08-18 Ross Burton <ross@burtonini.com>
649 Optimise single-character insertions.
654 * tests/string-test.c:
655 Add tests for new optimisation, and fix a leak.
657 2005-08-17 Matthias Clasen <mclasen@redhat.com>
659 * configure.in: Check for crt_externs.h and _NSGetEnviron.
661 * glib/gutils.c: On Darwin, include crt-externs.h and
662 define environ using _NSGetEnviron(). (#313731)
664 2005-08-16 Stepan Kasal <kasal@ucw.cz>
666 * glib/gutils.c (g_get_any_init): Move the body of the big if...
667 (g_get_any_init_do): ... to this new function.
668 (g_get_any_init): Declare as inline.
669 (g_get_any_init_locked): New inline function, does the locking.
670 Make use of these two throughout the code.
672 2005-08-15 Matthias Clasen <mclasen@redhat.com>
674 * glib/gbacktrace.c (g_on_error_stack_trace): Wait for
675 the child process and then simply return. This makes
676 The "S" option work as documented in g_on_error_query().
677 (#313125, Matthew F. Barnes)
679 * glib/gunicode.h: Update the link to Unicode category
680 values. (#313369, Behnam Esfahbod)
682 * glib/gqueue.c (g_queue_find_custom): Clarify docs
683 a little. (#311727, Tristan van Berkom)
685 * glib/abicheck.sh, gobject/abicheck.sh: Make the
686 check work on ia64 too, where some symbols ended up
687 in yet another section.