1 2005-12-05 Martyn Russell <martyn@imendio.com>
3 * docs/reference/glib/glib-sections.txt:
4 * glib/gasyncqueue.[ch]:
5 - Added support for sorting async queues by with _push_sorted(),
6 _push_sorted_unlocked(), _sort() and _sort_unlocked() (#323047).
9 * tests/asyncqueue-test.c:
10 - Added test case for gasyncqueue.c
12 Mon Dec 5 15:53:20 2005 Tim Janik <timj@imendio.com>
14 * glib/gslice.c: implement chain walking for arbitrary ->next pointer
15 offsets in g_slice_free_chain_with_offset() based on a patch by behdad
16 in bug 323178. moved time consuming logic from g_slice_free() out of
17 the inner loop, so g_slice_free_chain_with_offset() provides a real
18 performance benefit over g_slice_free1() now.
20 * glib/gslice.h: renamed g_slice_free_chain() to
21 g_slice_free_chain_with_offset(). implemented g_slice_free_chain() as
22 a type-safe macro as suggested in bug 323178.
23 simplified the macro implementation of g_slice_free() and implemented
24 it in a type safe manner for all compliers as suggested by Morten
25 Welinder <mortenw@gnome.org>.
30 * glib/glib.symbols: s/g_slice_free_chain/g_slice_free_chain_with_offset/
32 2005-12-05 Matthias Clasen <mclasen@redhat.com>
34 * glib/gasyncqueue.c: Add some docs.
36 * tests/libmoduletestplugin_a.c: Fix compiler warnings.
38 * glib/gatomic.c: In the ia64 implementation, use
39 __sync builtin without _si or _di suffix. (#321229,
40 Stanislav Brabec, patch by Andreas Schwab)
42 2005-12-04 Behdad Esfahbod <behdad@gnome.org>
44 * glib/gslice.h: Remove comma at the end of enum.
46 2005-12-04 Matthias Clasen <mclasen@redhat.com>
48 Handle multiple user names with the same UID better.
49 (#319535, Laszlo Peter)
51 * glib/gutils.c (g_get_any_init_do): When determining user
52 data, first look up $LOGNAME. If the UID doesn't match
53 getuid(), fall back to the current behaviour of looking
54 up the user data based on getuid().
56 2005-12-04 Matthias Clasen <mclasen@redhat.com>
58 * glib/gmacros.h (G_GNUC_WARN_UNUSED_RESULT): Add a macro
59 to make gcc warn if a function result is ignored. (#145466,
60 Arjan van de Ven, Alex Larsson)
62 * glib/gmem.h: Add the new attribute to g_realloc and
65 2005-12-03 Matthias Clasen <mclasen@redhat.com>
68 * glib/gcache.h: Deprecate g_cache_value_foreach. (#322956,
71 * glib/gmappedfile.c: Make mapping of empty files
74 * glib/gfileutils.c: Don't fork a new process just to
75 fix the permissions of the created temp file. (#321318,
76 Alexis S. L. Carvalho)
78 2005-12-02 Matthias Clasen <mclasen@redhat.com>
80 * README.in: Add a note about Solaris threads.
83 * configure.in: Remove support for Solaris threads.
84 (#136971, Sebastian Wilhelmi, patch by Andrew Paprocki)
86 2005-12-02 Matthias Clasen <mclasen@redhat.com>
88 * configure.in: Define G_GUINT64_CONSTANT in analogy to
89 G_GINT64_CONSTANT. (#322568, Andrew Paprocki)
91 2005-12-02 Matthias Clasen <mclasen@redhat.com>
93 * glib/gslice.c: Win32 portability fixes and C99ism removal,
94 pointed out by Kazuki Iwamoto. (#323052)
96 2005-12-02 Matthias Clasen <mclasen@redhat.com>
98 * glib/gatomic.c (g_atomic_int_add): Add a missing volatile for
99 the IA64 implementation.
101 Fri Dec 2 16:18:09 2005 Tim Janik <timj@imendio.com>
103 * glib/gslice.c: some naming and type size fixups.
105 Fri Dec 2 13:08:58 2005 Tim Janik <timj@gtk.org>
107 * glib/gslice.h (g_slice_free): reworked GCC-specific type-safe macro
108 variant into something less verbose, but digestible for gcc-3.4.
110 Fri Dec 2 10:55:07 2005 Tim Janik <timj@imendio.com>
112 * tests/slice-test.c: extended to perform the benchmarking on the old
113 memchunk code if 'O' is selected.
115 * tests/memchunks.c: new file which contains the old GLib mem chunks
116 implementation with prefix old_mem_chunk_.
118 * tests/Makefile.am: added memchunks.c
120 Fri Dec 2 00:16:59 2005 Tim Janik <timj@imendio.com>
122 * glib/gslice.c: improved gettimeofday() timer resolution by maintaining
123 acceptable load distribution of the syscall.
124 (allocator_get_magazine_threshold): reduce minimum/single-thread magazine
125 sizes to approximately page_size/2.5 to avoid excessive startup allocations.
127 2005-12-01 Matthias Clasen <mclasen@redhat.com>
129 * tests/gobject/ifacecheck.c (test_iface_base_init): Don't
130 double-free base interface members.
132 * tests/Makefile.am: Remove duplicate variable.
134 Thu Dec 1 17:32:46 2005 Tim Janik <timj@imendio.com>
136 * glib/gslice.[hc]: new slice allocator implementation.
138 * tests/slice-test.c: added random slice allocation test.
140 * glib/gthread.[hc]: removed newly added private thread mem API.
142 * glib/gthreadinit.h:
145 * glib/gmem.c: divided glib threading initialisation into three phases,
146 initialisation where private keys and messaging are not available (only
147 needed by gmem.c), initialisation without messaging but private keys
148 available (gslice.c, gmessage.c), and full fledged initialisers that
149 server the rest of glib. initialisation functions got renamed to reflect
150 the limitations of their corresponding phases.
152 * glib/gmem.c: removed memchunk code, defer allocations to
155 * glib/gmem.[hc]: removed g_slice_* skeletons.
157 * glib/glib.symbols: added g_slice_* symbols.
159 * configure.in: check for availability of posix_memalign(3), memalign(3)
162 * glib/Makefile.am: added gslice.[hc].
164 2005-12-01 Tor Lillqvist <tml@novell.com>
166 * glib/gstdio.c (g_stat): In the Win32 implementation, strip
167 trailing slash(es) for non-root folders. stat() fails if non-root
168 folders are specified with trailing slashes. It's too much hassle
169 to demand that callers strip such slashes themselves, especially
170 as it is easy to get it wrong and strip the slash of a root
172 (g_rename): On NT-based Windows, use MoveFileEx() with
173 MOVEFILE_REPLACE_EXISTING to better match Unix behaviour.
175 2005-11-28 Matthias Clasen <mclasen@redhat.com>
177 Fix G_STMT_START / G_STMT_END on Solaris. (#321972,
180 * configure.in: Check whether do { } while (0) works.
182 * glib/gmacros.h: Use do { } while (0) for G_STMT_START /
183 G_STMT_END if it works.
185 2005-11-28 Matthias Clasen <mclasen@redhat.com>
187 * glib/gthread.c (g_static_rw_lock_wait, g_static_rw_lock_signal):
188 * glib/gnode.c (g_node_depth_traverse_level):
189 * glib/gmem.c (g_allocator_new):
190 * glib/ghash.c (g_hash_table_unref, g_hash_table_destroy)
191 (g_hash_table_foreach_remove_or_steal): Silence compiler
194 2005-11-27 Matthias Clasen <mclasen@redhat.com>
196 * glib/gunicollate.c (g_utf8_collate_key_for_filename):
197 Don't read beyond len. (#322520, Christian Persch)
199 Wed Nov 23 17:34:01 2005 Tim Janik <timj@imendio.com>
201 * glib/gdataset.c: access datalist flags via atomic pointer access
202 functions, instead of acquiring the dataset lock. this is faster and
203 also matches the atomic pointer readouts in gdatalistprivate.h and
204 direct pointer modifications required by gobject.c.
206 Wed Nov 23 13:35:31 2005 Tim Janik <timj@gtk.org>
208 * glib/gdataset.c: streamlined initialization code somewhat,
209 removed GData node cache.
211 2005-11-22 Matthias Clasen <mclasen@redhat.com>
213 * glib/ghash.c (g_hash_table_ref):
214 (g_hash_table_unref): Mark these as new API.
216 Avoid double locking in g_intern_string (#322133,
219 * glib/gdataset.c (g_quark_from_string_internal): New
220 internal function which factors out common parts of
221 g_quark_from[_static]_string.
222 (g_quark_from_string, g_quark_from_static_string):
223 Use g_quark_from_string_internal.
224 (g_intern_string, g_intern_static_string): Use
225 g_quark_from_string_internal, and only take the
227 (g_quark_new): Don't store the strings shifted by -1
228 in the g_quarks array.
229 (g_quark_to_string): Adapt to the previous change.
231 Tue Nov 22 14:04:26 2005 Tim Janik <timj@imendio.com>
235 g_hash_table_new_full(): create hash tables with a ref count of 1.
236 g_hash_table_ref(): atomically ref_count+=1
237 g_hash_table_unref(): atomically ref_count-=1, destroys hash table
238 when refcount reaches 0.
239 g_hash_table_destroy(): just destroy keys and values, unref by 1.
240 g_hash_table_insert():
241 g_hash_table_replace(): assert ref_count>0.
244 * glib/gatomic.c: added 'volatile' qualifier to all atomic pointer and
247 2005-11-20 Behdad Esfahbod <behdad@gnome.org>
249 * glib/guniprop.c (g_unichar_get_mirror_char): Remove unused
250 variables. (#321984, Andrew Paprocki)
252 2005-11-18 Matthias Clasen <mclasen@redhat.com>
254 * configure.in: Bump version
256 2005-11-17 Matthias Clasen <mclasen@redhat.com>
258 * === Released 2.9.0 ===
260 * glib/glib.symbols: Updates
262 * glib/gutf8.c (g_utf8_offset_to_pointer): Add some performance
263 hints to the docs. (Paolo Borelli)
267 * glib/gthread.c (g_thread_foreach): Mark as new api.
269 * README.in: Updates.
273 * glib/gdate.c (g_date_set_time_t):
274 (g_date_set_time_val): Add functions to set a date from
275 a time_t and from a GTimeVal.
276 (g_date_set_time): Deprecate. (#314067, Roger Leigh)
279 * tests/utf8-pointer.c: Unit tests for g_utf8_pointer_to_offset
280 and g_utf8_offset_to_pointer.
282 * glib/gutf8.c (g_utf8_pointer_to_offset)
283 (g_utf8_offset_to_pointer): Handle negative offsets, and use
284 "stutter stepping" for going backwards. (#320638, Larry
289 * glib/gthread.c: const correctness fixes, found
290 by Arjan van de Ven and gcc.
292 2005-11-16 Behdad Esfahbod <behdad@gnome.org>
294 * docs/reference/glib/tmpl/unicode.sgml: Correct typo on mentioning
295 @G_UNICODE_BREAK_UNKNOWN that should be #G_UNICODE_BREAK_UNKNOWN.
297 2005-11-09 Behdad Esfahbod <behdad@gnome.org>
299 * glib/guniprop.c: Use bit hacks instead when checking a general
300 category value against multiple values.
302 * glib/gutf8.c: Change ISO10646 to Unicode in docs.
304 2005-11-10 Simos Xenitellis <simos@gnome.org>
306 * configure.in: Added tt (Tatar) to ALL_LINGUAS.
308 2005-11-05 Tor Lillqvist <tml@novell.com>
310 * glib/giowin32.c (g_io_win32_set_flags): Don't call g_warning().
311 (#320688) Instead set error and return failure.
313 2005-11-07 Matthias Clasen <mclasen@redhat.com>
315 * glib/gmain.c (g_main_dispatch): Don't call cb_funcs->unref
316 while holding the context lock. (#320886, Andy Wingo)
318 2005-11-07 Billy Biggs <vektor@dumbterm.net>
320 * tests/refcount/Makefile.am: Add a missing reference to libglib
321 in the LDADD for this test.
323 2005-11-07 Matthias Clasen <mclasen@redhat.com>
325 * glib/gmem.c (g_allocator_new): We need to set n_preallocs to a
326 nonzero value, otherwise GTK+ 2.8 breaks when compiled against
327 GLib 2.9. (#320755, Luca Ferretti)
329 2005-11-04 Matthias Clasen <mclasen@redhat.com>
331 * glib/Makefile.am: Apply a patch from Behdad Esfahbod to
332 use a faster lookup table for g_unichar_get_mirror_char().
333 * glib/gmirroringtable.h: The new table.
334 * glib/gunichartables.h: Remove bidi_mirroring_table.
335 * glib/gen-unicode-tables.pl: Don't generate the mirroring
337 * glib/glib-mirroring-tab/*: Sources for the program
338 which generated gmirroringtable.h.
340 * glib/glist.c: Avoid some code duplication.
342 * glib/gscanner.h: Include gdataset.h. (#320322)
344 * glib/gdebug.h: add new GDebugFlag for fatal_criticals
345 * glib/gmessages.c: (_g_debug_init): handle G_DEBUG=fatal_criticals,
346 to help find critical warnings in applications. (#320017,
349 2005-11-02 Tor Lillqvist <tml@novell.com>
351 * glib/glib.symbols: Remove large amount of trailing whitespace
352 from one line. Remove #ifdef G_UNDEFINED from around
353 g_hash_table_new and g_hash_table_new_full symbols, what was the
354 purpose of that? Add the g_slice functions.
356 Tue Nov 1 16:24:20 2005 Tim Janik <timj@imendio.com>
358 * glib/gmem.[hc]: prepared deprecation of GMemChunk and GAllocator.
359 added g_slice_*() API to allocate and cache small bits of memory.
360 an actuall allocator implementation for g_slice_*() is still pending.
362 * glib/gthread.[hc]: changes from a patch by Matthias Clasen.
363 changed GRealThread list to use in-structure *next; fields instead
364 of GSList, in order for thread iteration to not depenend on g_slice_*()
366 _g_thread_mem_private_get():
367 _g_thread_mem_private_set(): added accessors for private memory,
368 needed because the ordinary GPrivate implementation relies on GArray
369 and GSList and therefore indirectly on working g_slice_*() allocations.
372 g_thread_foreach(): new public API function to loop over all existing threads.
386 removed GAllocator and free list usages and accompanying locks.
387 use g_slice_*() API to allocate and cache small bits of memory.
389 * glib/ghook.h: removed GMemChunk field from public API.
392 * glib/glist.h: deprecate allocator API, provide _free1() for consistency.
394 * glib/gnode.h: deprecate allocator API.
396 * glib/gmain.c: reordered GPollRec fields so g_slice_free_chain() can
397 be used for poll rec lists.
399 * glib/grel.c: removed mem chunk usage, and allocated tuples via g_slice_*().
400 g_relation_destroy(): free all tuples from the all_tuples hash table,
401 this effectively maintains the life time track keeping of tuples.
402 g_relation_delete_tuple(): free tuples which are removed from the
403 all_tuples hash table. this fixes a temporary leak that was present
404 in the memchunk code until the destruction of the relation.
406 2005-10-29 Matthias Clasen <mclasen@redhat.com>
408 * tests/convert-test.c: Add some tests for conversions between
409 UTF-8, UCS-4 and UTF-16.
411 * glib/gutf8.c (g_utf8_to_ucs4, g_utf8_to_utf16): Fix handling
412 of len == -1, noticed by Morten Welinder.
414 2005-10-27 Erdal Ronahi <erdal.ronahi@gmail.com>
416 * configure.in: Added ku (Kurdish) to ALL_LINGUAS
418 2005-10-26 Matthias Clasen <mclasen@redhat.com>
420 * glib/gutf8.c (g_ucs4_to_utf8): Don't set items_read twice
421 in the error case, and add some documentation. (#319806, Morten
424 2005-10-19 Manish Singh <yosh@gimp.org>
426 * configure.in: Use AC_CHECK_FUNCS for _NSGetEnviron, to get the
427 config.h symbol automatically. Fixes bug #313731.
429 2005-10-19 Tor Lillqvist <tml@novell.com>
431 * glib/gwin32.c (g_win32_get_package_installation_directory):
432 Return a g_strdup()ed copy of the value stored in the hash table,
433 so that it can be g_free()d without leaving a dangling pointer in
434 the hash table. (#319232)
436 2005-10-06 Matthias Clasen <mclasen@redhat.com>
438 * glib/gunicollate.c (g_utf8_collate_key_for_filename): Handle
439 all-zero sequences correctly. (#317930, Sebastien Bacher)
441 2005-10-05 Matthias Clasen <mclasen@redhat.com>
443 * glib/gmarkup.c (g_markup_escape_text): Doc addition
445 * Makefile.am: Add Collation tests.
447 * tests/collate/*: Inputs and expected outputs for collation tests.
449 * tests/run-collate-tests.sh: Script to run collation tests.
451 * tests/unicode-collate.c (main): Rework slightly to make
452 it usable in unit tests. Also test g_utf8_collate_key_for_filename().
454 2005-10-01 Behdad Esfahbod <behdad@gnome.org>
456 * docs/reference/glib/tmpl/unicode.sgml:
457 * glib/gen-unicode-tables.pl:
459 * glib/gunichartables.h:
461 * tests/casefold.txt:
462 * tests/casemap.txt: Updated to Unicode 4.1. There are five new
463 GUnicodeBreakType types. That may break some applications, like
466 2005-09-26 Matthias Clasen <mclasen@redhat.com>
468 * glib/gstrfuncs.c (g_ascii_strtoull): Add details to the
469 docs. (#314393, Matthew F. Barnes)
472 * glib/gprintf.h: Remove g_snprintf() and g_vsnprintf(), since
473 they are already declared in glib.h. This doesn't break documented
474 use of gprintf.h, but should probably be pointed out in the
475 release notes for 2.10. (#314232, Behdad Esfahbod)
477 Tue Sep 20 13:16:04 2005 Tim Janik <timj@imendio.com>
479 * glib/gpattern.c (g_pattern_ph_match): applied significant recursion
480 complexity optimization, based on a patch from Matthias Clasen.
482 * tests/patterntest.c: more tests, mostly from matthias.
484 2005-09-20 Matthias Clasen <mclasen@redhat.com>
486 * glib/gqueue.c (g_queue_insert_sorted): Correct the docs.
487 (#316703, Mark Drago)
489 Mon Sep 19 17:23:23 2005 Tim Janik <timj@imendio.com>
491 * glib/gpattern.c: applied a patch from matthias which checks on the
492 upper bound of GPatternSpec length to optimize matches.
495 * tests/patterntest.c: added more match cases.
497 2005-09-16 Tor Lillqvist <tml@novell.com>
499 * glib/gstrfuncs.c (g_ascii_strcasecmp, g_ascii_strncasecmp): Add
500 warning to doc comment that these functions should not be used on
501 encodings like CP932.
503 2005-09-14 Matthias Clasen <mclasen@redhat.com>
505 * tests/keyfile-test.c: Add a test for grup names of length 1.
507 * glib/gkeyfile.c (g_key_file_line_is_group): Accept group names
508 of length 1. (#316309)
510 2005-09-12 Matthias Clasen <mclasen@redhat.com>
512 * glib/gmarkup.c (g_markup_escape_text): Clarify docs.
514 2005-09-11 Sebastian Wilhelmi <seppi@seppi.de>
516 * tests/refcount/Makefile.am (INCLUDES): Link the the refcount
517 tests to the system thread library $(G_THREAD_LIBS). Fixes #313744
520 2005-09-11 Kjartan Maraas <kmaraas@gnome.org>
522 * glib/gmain.c: (g_child_watch_prepare), (g_child_watch_check),
523 (child_watch_helper_thread): Remove some dead code. Closes
526 2005-09-07 Tor Lillqvist <tml@novell.com>
528 * glib/Makefile.am: Create also a console version of the
529 gspawn-win32-helper program, gspawn-win32-helper-console.exe.
530 It's otherwise identical to gspawn-win32-helper.exe, except marked
531 as a console application (linked without the -mwindows option).
533 * glib/gspawn-win32.c (do_spawn_directly, do_spawn_with_pipes):
534 Drop the dont_wait parameter. Its truth value correlated 100% with
535 the NULLness of the exit_status parameter anyway, so it's enough
536 to check whether exit_status is NULL. Invert the sense of the
537 dont_return_handle parameter and rename it to do_return_handle, to
538 make the code easier to read by avoiding double negations.
540 (g_spawn_sync_utf8, g_spawn_async_with_pipes_utf8): Modify calls
541 to do_spawn_with_pipes() accordingly.
543 (do_spawn_with_pipes): If we have a console, use the console
544 version of the helper program, otherwise use the GUI one. This
545 avoids extra console windows opening up in some situations. (In
546 case a console application uses the GUI gspawn-win32-helper.exe to
547 spawn another console application we would get a separate console
548 for the spawned console application).
550 * glib-zip.in: Distribute also gspawn-win32-helper-console.exe.
552 2005-09-05 Matthias Clasen <mclasen@redhat.com>
554 * glib/gmappedfile.c (g_mapped_file_new): Report an error
555 if the file is too large. (#315275, Kjartan Maraas)
557 * glib/gkeyfile.c (g_key_file_load_from_fd): The return value
558 of read() is signed. (#315273, Kjartan Maraas)
560 2005-08-31 Tor Lillqvist <tml@novell.com>
562 * glib/gutils.h: Wrapping atexit() is a bad idea on Windows, where
563 the EXE and each DLL have their own atexit function chains.
565 #define g_atexit as atexit instead. This means it has a
566 better chance of doing what the caller wants. For instance,
567 gtkhtml calls g_atexit() registering a function in gtkhtml
568 itself. This caused a crash when g_atexit() was implemented as a
569 function in the GLib DLL. The gtkhtml DLL was already unloaded by
570 the time the GLib DLL got unloaded.
572 * glib/gutils.c: #undef the #define mentioned above, to also get a
573 real g_atexit() into the DLL for backward compatibility. Document
574 the Windows behaviour of g_atexit(), and document the varying ways
575 atexit() can behave in the context of dynamically loaded modules
578 2005-08-31 Matthias Clasen <mclasen@redhat.com>
582 * glib/gdataset.c: Add string interning functions.
584 2005-08-28 Matthias Clasen <mclasen@redhat.com>
586 * glib/giochannel.c: Unify some near-duplicate strings. (#314654,
589 2005-08-26 Matthias Clasen <mclasen@redhat.com>
591 * configure.in: Bump version to 2.9.0
593 2005-08-25 Tor Lillqvist <tml@novell.com>
595 Make also the g_spawn*() functions take parameters in the GLib
596 file name encoding, i.e. UTF-8, on Windows. Has no impact on Unix
597 API or ABI. Like the other GLib API that was earlier changed to
598 use UTF-8 on Windows, the names of the functions that take UTF-8
599 have _utf8 suffixes added by using preprocessor macros in the
600 header file. The old names are kept for functions with the old
601 behaviour, taking parameters in the system codepage, for DLL ABI
604 * glib/gspawn.h: On Win32 add the suffix _utf8 to the names of the
605 g_spawn*() functions.
607 * glib/gspawn-win32.c: Use wide-char API on NT-based
608 Windows. Convert parameters from UTF-8 to wide chars (NT) or
609 system codepage (Win9x) and call the C library _wspawn*() or
610 spawn*() functions respectvely. Add DLL ABI stability versions
611 that take parameters in the system codepage.
613 * glib/gspawn-win32-helper.c: On NT-based Windows use the
614 wide-char versions of argv and envp, and use wide-char API to
615 change directory and spawn the program to run. Remove the verbose
616 debugging output, it was too complex to modify for the wide-char
617 features. (Just add temporary debugging printouts if needed, no
618 need to have them permanently in the source.)
620 * glib/gspawn.c: Corresponding documentation updates.
622 * glib/glib.symbols: Corresponding changes: Mark the ABI stability
623 symbols as PRIVATE, add the new _utf8-suffixed ones.
625 2005-08-24 Stepan Kasal <kasal@ucw.cz>
627 * glib/gtypes.h (G_MININT64): Cast the constant to gint64; it is
628 guint64 otherwise and that can produce warnings about comparison
629 between signed and unsigned.
631 2005-08-23 Matthias Clasen <mclasen@redhat.com>
633 * glib/gutils.c: Fix the crt_externs.h include.
635 2005-08-23 Stepan Kasal <kasal@ucw.cz>
637 * NEWS: Fix spelling of my first name.
639 2005-08-23 Matthias Clasen <mclasen@redhat.com>
643 * === Released 2.8.1 ===
647 2005-08-20 Hans Breuer <hans@breuer.org>
649 * glib/makefile.msc.in : link with ws2_32.lib
651 2005-08-18 Tor Lillqvist <tml@novell.com>
653 * configure.in: Check for <sys/wait.h>
655 * glib/gbacktrace.c: Include <sys/wait.h> on if HAVE_SYS_WAIT_H.
657 2005-08-18 Ross Burton <ross@burtonini.com>
660 Optimise single-character insertions.
665 * tests/string-test.c:
666 Add tests for new optimisation, and fix a leak.
668 2005-08-17 Matthias Clasen <mclasen@redhat.com>
670 * configure.in: Check for crt_externs.h and _NSGetEnviron.
672 * glib/gutils.c: On Darwin, include crt-externs.h and
673 define environ using _NSGetEnviron(). (#313731)
675 2005-08-16 Stepan Kasal <kasal@ucw.cz>
677 * glib/gutils.c (g_get_any_init): Move the body of the big if...
678 (g_get_any_init_do): ... to this new function.
679 (g_get_any_init): Declare as inline.
680 (g_get_any_init_locked): New inline function, does the locking.
681 Make use of these two throughout the code.
683 2005-08-15 Matthias Clasen <mclasen@redhat.com>
685 * glib/gbacktrace.c (g_on_error_stack_trace): Wait for
686 the child process and then simply return. This makes
687 The "S" option work as documented in g_on_error_query().
688 (#313125, Matthew F. Barnes)
690 * glib/gunicode.h: Update the link to Unicode category
691 values. (#313369, Behnam Esfahbod)
693 * glib/gqueue.c (g_queue_find_custom): Clarify docs
694 a little. (#311727, Tristan van Berkom)
696 * glib/abicheck.sh, gobject/abicheck.sh: Make the
697 check work on ia64 too, where some symbols ended up
698 in yet another section.