1 Tue Dec 13 10:13:32 2005 Tim Janik <timj@imendio.com>
3 * glib/gatomic.h: added g_atomic_pointer_set() and g_atomic_int_set()
4 for systems where the initialization of atomic variables requires a
7 2005-12-09 Matthias Clasen <mclasen@redhat.com>
9 * configure.in: Bump version
11 * === Released 2.9.1 ===
13 2005-12-09 Alexander Larsson <alexl@redhat.com>
17 Use G_GNUC_WARN_UNUSED_RESULT on list functions that return
20 2005-12-08 Matthias Clasen <mclasen@redhat.com>
24 2005-12-07 Martyn Russell <martyn@imendio.com>
27 - Call g_queue_insert_sorted() instead of duplicating the code.
28 - Call g_queue_sort() instead of duplicating the code.
29 - Invert sort function results to make sure the same sort function
30 gives the same results across glist, gslist, gqueue and
33 * tests/asyncqueue-test.c:
34 - Updated the sort function to reflect the example in the
35 documentation for gasyncqueue.c.
37 2005-12-07 Martyn Russell <martyn@imendio.com>
39 * docs/reference/glib/glib-sections.txt:
40 * docs/reference/glib/tmpl/linked_lists_double.sgml:
41 * docs/reference/glib/tmpl/linked_lists_single.sgml:
44 - Added g_list_insert_sorted_with_data () and
45 g_slist_insert_sorted_with_data ().
46 - Removed the extra check in g_list_sort() and g_slist_sort() for
47 GCompareDataFunc vs. GCompareFunc.
49 2005-12-07 Tor Lillqvist <tml@novell.com>
51 * glib-gettextize.in: Look up prefix at run-time on Win32,
52 assuming the standard directory structure with glib-gettextize in
55 2005-12-06 Matthias Clasen <mclasen@redhat.com>
57 * glib/gmem.h: Exempt GTK+ from the mem chunk deprecation, since
58 we need GTK+ 2.8 to compile against GLib 2.10.
60 2005-12-05 Manish Singh <yosh@gimp.org>
62 * tests/run-collate-test.sh: set LC_COLLATE, not LANG, to be sure
63 to override any user settings.
65 2005-12-05 Matthias Clasen <mclasen@redhat.com>
67 * configure.in: Revert an accidential commit
69 2005-12-05 Matthias Clasen <mclasen@redhat.com>
71 * glib/gutf8.c: Documentation updates. (#323291, Morten Welinder)
73 * tests/Makefile.am (TESTS_ENVIRONMENT):
74 * tests/gobject/Makefile.am (TESTS_ENVIRONMENT): Set
75 MALLOC_CHECK_ and MALLOC_PERTURB_
77 * tests/run-collate-tests.sh: Run the collation tests explicitly
78 in en_US locale. (#320463)
80 * glib/gmem.h: Really deprecate GMemChunk.
87 * glib/gstrfuncs.c: Add versioned deprecation docs.
89 2005-12-05 Martyn Russell <martyn@imendio.com>
91 * docs/reference/glib/glib-sections.txt:
92 * glib/gasyncqueue.[ch]:
93 - Added support for sorting async queues by with _push_sorted(),
94 _push_sorted_unlocked(), _sort() and _sort_unlocked() (#323047).
97 * tests/asyncqueue-test.c:
98 - Added test case for gasyncqueue.c
100 Mon Dec 5 15:53:20 2005 Tim Janik <timj@imendio.com>
102 * glib/gslice.c: implement chain walking for arbitrary ->next pointer
103 offsets in g_slice_free_chain_with_offset() based on a patch by behdad
104 in bug 323178. moved time consuming logic from g_slice_free() out of
105 the inner loop, so g_slice_free_chain_with_offset() provides a real
106 performance benefit over g_slice_free1() now.
108 * glib/gslice.h: renamed g_slice_free_chain() to
109 g_slice_free_chain_with_offset(). implemented g_slice_free_chain() as
110 a type-safe macro as suggested in bug 323178.
111 simplified the macro implementation of g_slice_free() and implemented
112 it in a type safe manner for all compliers as suggested by Morten
113 Welinder <mortenw@gnome.org>.
118 * glib/glib.symbols: s/g_slice_free_chain/g_slice_free_chain_with_offset/
120 2005-12-05 Matthias Clasen <mclasen@redhat.com>
122 * glib/gasyncqueue.c: Add some docs.
124 * tests/libmoduletestplugin_a.c: Fix compiler warnings.
126 * glib/gatomic.c: In the ia64 implementation, use
127 __sync builtin without _si or _di suffix. (#321229,
128 Stanislav Brabec, patch by Andreas Schwab)
130 2005-12-04 Behdad Esfahbod <behdad@gnome.org>
132 * glib/gslice.h: Remove comma at the end of enum.
134 2005-12-04 Matthias Clasen <mclasen@redhat.com>
136 Handle multiple user names with the same UID better.
137 (#319535, Laszlo Peter)
139 * glib/gutils.c (g_get_any_init_do): When determining user
140 data, first look up $LOGNAME. If the UID doesn't match
141 getuid(), fall back to the current behaviour of looking
142 up the user data based on getuid().
144 2005-12-04 Matthias Clasen <mclasen@redhat.com>
146 * glib/gmacros.h (G_GNUC_WARN_UNUSED_RESULT): Add a macro
147 to make gcc warn if a function result is ignored. (#145466,
148 Arjan van de Ven, Alex Larsson)
150 * glib/gmem.h: Add the new attribute to g_realloc and
153 2005-12-03 Matthias Clasen <mclasen@redhat.com>
156 * glib/gcache.h: Deprecate g_cache_value_foreach. (#322956,
159 * glib/gmappedfile.c: Make mapping of empty files
162 * glib/gfileutils.c: Don't fork a new process just to
163 fix the permissions of the created temp file. (#321318,
164 Alexis S. L. Carvalho)
166 2005-12-02 Matthias Clasen <mclasen@redhat.com>
168 * README.in: Add a note about Solaris threads.
171 * configure.in: Remove support for Solaris threads.
172 (#136971, Sebastian Wilhelmi, patch by Andrew Paprocki)
174 2005-12-02 Matthias Clasen <mclasen@redhat.com>
176 * configure.in: Define G_GUINT64_CONSTANT in analogy to
177 G_GINT64_CONSTANT. (#322568, Andrew Paprocki)
179 2005-12-02 Matthias Clasen <mclasen@redhat.com>
181 * glib/gslice.c: Win32 portability fixes and C99ism removal,
182 pointed out by Kazuki Iwamoto. (#323052)
184 2005-12-02 Matthias Clasen <mclasen@redhat.com>
186 * glib/gatomic.c (g_atomic_int_add): Add a missing volatile for
187 the IA64 implementation.
189 Fri Dec 2 16:18:09 2005 Tim Janik <timj@imendio.com>
191 * glib/gslice.c: some naming and type size fixups.
193 Fri Dec 2 13:08:58 2005 Tim Janik <timj@gtk.org>
195 * glib/gslice.h (g_slice_free): reworked GCC-specific type-safe macro
196 variant into something less verbose, but digestible for gcc-3.4.
198 Fri Dec 2 10:55:07 2005 Tim Janik <timj@imendio.com>
200 * tests/slice-test.c: extended to perform the benchmarking on the old
201 memchunk code if 'O' is selected.
203 * tests/memchunks.c: new file which contains the old GLib mem chunks
204 implementation with prefix old_mem_chunk_.
206 * tests/Makefile.am: added memchunks.c
208 Fri Dec 2 00:16:59 2005 Tim Janik <timj@imendio.com>
210 * glib/gslice.c: improved gettimeofday() timer resolution by maintaining
211 acceptable load distribution of the syscall.
212 (allocator_get_magazine_threshold): reduce minimum/single-thread magazine
213 sizes to approximately page_size/2.5 to avoid excessive startup allocations.
215 2005-12-01 Matthias Clasen <mclasen@redhat.com>
217 * tests/gobject/ifacecheck.c (test_iface_base_init): Don't
218 double-free base interface members.
220 * tests/Makefile.am: Remove duplicate variable.
222 Thu Dec 1 17:32:46 2005 Tim Janik <timj@imendio.com>
224 * glib/gslice.[hc]: new slice allocator implementation.
226 * tests/slice-test.c: added random slice allocation test.
228 * glib/gthread.[hc]: removed newly added private thread mem API.
230 * glib/gthreadinit.h:
233 * glib/gmem.c: divided glib threading initialisation into three phases,
234 initialisation where private keys and messaging are not available (only
235 needed by gmem.c), initialisation without messaging but private keys
236 available (gslice.c, gmessage.c), and full fledged initialisers that
237 server the rest of glib. initialisation functions got renamed to reflect
238 the limitations of their corresponding phases.
240 * glib/gmem.c: removed memchunk code, defer allocations to
243 * glib/gmem.[hc]: removed g_slice_* skeletons.
245 * glib/glib.symbols: added g_slice_* symbols.
247 * configure.in: check for availability of posix_memalign(3), memalign(3)
250 * glib/Makefile.am: added gslice.[hc].
252 2005-12-01 Tor Lillqvist <tml@novell.com>
254 * glib/gstdio.c (g_stat): In the Win32 implementation, strip
255 trailing slash(es) for non-root folders. stat() fails if non-root
256 folders are specified with trailing slashes. It's too much hassle
257 to demand that callers strip such slashes themselves, especially
258 as it is easy to get it wrong and strip the slash of a root
260 (g_rename): On NT-based Windows, use MoveFileEx() with
261 MOVEFILE_REPLACE_EXISTING to better match Unix behaviour.
263 2005-11-28 Matthias Clasen <mclasen@redhat.com>
265 Fix G_STMT_START / G_STMT_END on Solaris. (#321972,
268 * configure.in: Check whether do { } while (0) works.
270 * glib/gmacros.h: Use do { } while (0) for G_STMT_START /
271 G_STMT_END if it works.
273 2005-11-28 Matthias Clasen <mclasen@redhat.com>
275 * glib/gthread.c (g_static_rw_lock_wait, g_static_rw_lock_signal):
276 * glib/gnode.c (g_node_depth_traverse_level):
277 * glib/gmem.c (g_allocator_new):
278 * glib/ghash.c (g_hash_table_unref, g_hash_table_destroy)
279 (g_hash_table_foreach_remove_or_steal): Silence compiler
282 2005-11-27 Matthias Clasen <mclasen@redhat.com>
284 * glib/gunicollate.c (g_utf8_collate_key_for_filename):
285 Don't read beyond len. (#322520, Christian Persch)
287 Wed Nov 23 17:34:01 2005 Tim Janik <timj@imendio.com>
289 * glib/gdataset.c: access datalist flags via atomic pointer access
290 functions, instead of acquiring the dataset lock. this is faster and
291 also matches the atomic pointer readouts in gdatalistprivate.h and
292 direct pointer modifications required by gobject.c.
294 Wed Nov 23 13:35:31 2005 Tim Janik <timj@gtk.org>
296 * glib/gdataset.c: streamlined initialization code somewhat,
297 removed GData node cache.
299 2005-11-22 Matthias Clasen <mclasen@redhat.com>
301 * glib/ghash.c (g_hash_table_ref):
302 (g_hash_table_unref): Mark these as new API.
304 Avoid double locking in g_intern_string (#322133,
307 * glib/gdataset.c (g_quark_from_string_internal): New
308 internal function which factors out common parts of
309 g_quark_from[_static]_string.
310 (g_quark_from_string, g_quark_from_static_string):
311 Use g_quark_from_string_internal.
312 (g_intern_string, g_intern_static_string): Use
313 g_quark_from_string_internal, and only take the
315 (g_quark_new): Don't store the strings shifted by -1
316 in the g_quarks array.
317 (g_quark_to_string): Adapt to the previous change.
319 Tue Nov 22 14:04:26 2005 Tim Janik <timj@imendio.com>
323 g_hash_table_new_full(): create hash tables with a ref count of 1.
324 g_hash_table_ref(): atomically ref_count+=1
325 g_hash_table_unref(): atomically ref_count-=1, destroys hash table
326 when refcount reaches 0.
327 g_hash_table_destroy(): just destroy keys and values, unref by 1.
328 g_hash_table_insert():
329 g_hash_table_replace(): assert ref_count>0.
332 * glib/gatomic.c: added 'volatile' qualifier to all atomic pointer and
335 2005-11-20 Behdad Esfahbod <behdad@gnome.org>
337 * glib/guniprop.c (g_unichar_get_mirror_char): Remove unused
338 variables. (#321984, Andrew Paprocki)
340 2005-11-18 Matthias Clasen <mclasen@redhat.com>
342 * configure.in: Bump version
344 2005-11-17 Matthias Clasen <mclasen@redhat.com>
346 * === Released 2.9.0 ===
348 * glib/glib.symbols: Updates
350 * glib/gutf8.c (g_utf8_offset_to_pointer): Add some performance
351 hints to the docs. (Paolo Borelli)
355 * glib/gthread.c (g_thread_foreach): Mark as new api.
357 * README.in: Updates.
361 * glib/gdate.c (g_date_set_time_t):
362 (g_date_set_time_val): Add functions to set a date from
363 a time_t and from a GTimeVal.
364 (g_date_set_time): Deprecate. (#314067, Roger Leigh)
367 * tests/utf8-pointer.c: Unit tests for g_utf8_pointer_to_offset
368 and g_utf8_offset_to_pointer.
370 * glib/gutf8.c (g_utf8_pointer_to_offset)
371 (g_utf8_offset_to_pointer): Handle negative offsets, and use
372 "stutter stepping" for going backwards. (#320638, Larry
377 * glib/gthread.c: const correctness fixes, found
378 by Arjan van de Ven and gcc.
380 2005-11-16 Behdad Esfahbod <behdad@gnome.org>
382 * docs/reference/glib/tmpl/unicode.sgml: Correct typo on mentioning
383 @G_UNICODE_BREAK_UNKNOWN that should be #G_UNICODE_BREAK_UNKNOWN.
385 2005-11-09 Behdad Esfahbod <behdad@gnome.org>
387 * glib/guniprop.c: Use bit hacks instead when checking a general
388 category value against multiple values.
390 * glib/gutf8.c: Change ISO10646 to Unicode in docs.
392 2005-11-10 Simos Xenitellis <simos@gnome.org>
394 * configure.in: Added tt (Tatar) to ALL_LINGUAS.
396 2005-11-05 Tor Lillqvist <tml@novell.com>
398 * glib/giowin32.c (g_io_win32_set_flags): Don't call g_warning().
399 (#320688) Instead set error and return failure.
401 2005-11-07 Matthias Clasen <mclasen@redhat.com>
403 * glib/gmain.c (g_main_dispatch): Don't call cb_funcs->unref
404 while holding the context lock. (#320886, Andy Wingo)
406 2005-11-07 Billy Biggs <vektor@dumbterm.net>
408 * tests/refcount/Makefile.am: Add a missing reference to libglib
409 in the LDADD for this test.
411 2005-11-07 Matthias Clasen <mclasen@redhat.com>
413 * glib/gmem.c (g_allocator_new): We need to set n_preallocs to a
414 nonzero value, otherwise GTK+ 2.8 breaks when compiled against
415 GLib 2.9. (#320755, Luca Ferretti)
417 2005-11-04 Matthias Clasen <mclasen@redhat.com>
419 * glib/Makefile.am: Apply a patch from Behdad Esfahbod to
420 use a faster lookup table for g_unichar_get_mirror_char().
421 * glib/gmirroringtable.h: The new table.
422 * glib/gunichartables.h: Remove bidi_mirroring_table.
423 * glib/gen-unicode-tables.pl: Don't generate the mirroring
425 * glib/glib-mirroring-tab/*: Sources for the program
426 which generated gmirroringtable.h.
428 * glib/glist.c: Avoid some code duplication.
430 * glib/gscanner.h: Include gdataset.h. (#320322)
432 * glib/gdebug.h: add new GDebugFlag for fatal_criticals
433 * glib/gmessages.c: (_g_debug_init): handle G_DEBUG=fatal_criticals,
434 to help find critical warnings in applications. (#320017,
437 2005-11-02 Tor Lillqvist <tml@novell.com>
439 * glib/glib.symbols: Remove large amount of trailing whitespace
440 from one line. Remove #ifdef G_UNDEFINED from around
441 g_hash_table_new and g_hash_table_new_full symbols, what was the
442 purpose of that? Add the g_slice functions.
444 Tue Nov 1 16:24:20 2005 Tim Janik <timj@imendio.com>
446 * glib/gmem.[hc]: prepared deprecation of GMemChunk and GAllocator.
447 added g_slice_*() API to allocate and cache small bits of memory.
448 an actuall allocator implementation for g_slice_*() is still pending.
450 * glib/gthread.[hc]: changes from a patch by Matthias Clasen.
451 changed GRealThread list to use in-structure *next; fields instead
452 of GSList, in order for thread iteration to not depenend on g_slice_*()
454 _g_thread_mem_private_get():
455 _g_thread_mem_private_set(): added accessors for private memory,
456 needed because the ordinary GPrivate implementation relies on GArray
457 and GSList and therefore indirectly on working g_slice_*() allocations.
460 g_thread_foreach(): new public API function to loop over all existing threads.
474 removed GAllocator and free list usages and accompanying locks.
475 use g_slice_*() API to allocate and cache small bits of memory.
477 * glib/ghook.h: removed GMemChunk field from public API.
480 * glib/glist.h: deprecate allocator API, provide _free1() for consistency.
482 * glib/gnode.h: deprecate allocator API.
484 * glib/gmain.c: reordered GPollRec fields so g_slice_free_chain() can
485 be used for poll rec lists.
487 * glib/grel.c: removed mem chunk usage, and allocated tuples via g_slice_*().
488 g_relation_destroy(): free all tuples from the all_tuples hash table,
489 this effectively maintains the life time track keeping of tuples.
490 g_relation_delete_tuple(): free tuples which are removed from the
491 all_tuples hash table. this fixes a temporary leak that was present
492 in the memchunk code until the destruction of the relation.
494 2005-10-29 Matthias Clasen <mclasen@redhat.com>
496 * tests/convert-test.c: Add some tests for conversions between
497 UTF-8, UCS-4 and UTF-16.
499 * glib/gutf8.c (g_utf8_to_ucs4, g_utf8_to_utf16): Fix handling
500 of len == -1, noticed by Morten Welinder.
502 2005-10-27 Erdal Ronahi <erdal.ronahi@gmail.com>
504 * configure.in: Added ku (Kurdish) to ALL_LINGUAS
506 2005-10-26 Matthias Clasen <mclasen@redhat.com>
508 * glib/gutf8.c (g_ucs4_to_utf8): Don't set items_read twice
509 in the error case, and add some documentation. (#319806, Morten
512 2005-10-19 Manish Singh <yosh@gimp.org>
514 * configure.in: Use AC_CHECK_FUNCS for _NSGetEnviron, to get the
515 config.h symbol automatically. Fixes bug #313731.
517 2005-10-19 Tor Lillqvist <tml@novell.com>
519 * glib/gwin32.c (g_win32_get_package_installation_directory):
520 Return a g_strdup()ed copy of the value stored in the hash table,
521 so that it can be g_free()d without leaving a dangling pointer in
522 the hash table. (#319232)
524 2005-10-06 Matthias Clasen <mclasen@redhat.com>
526 * glib/gunicollate.c (g_utf8_collate_key_for_filename): Handle
527 all-zero sequences correctly. (#317930, Sebastien Bacher)
529 2005-10-05 Matthias Clasen <mclasen@redhat.com>
531 * glib/gmarkup.c (g_markup_escape_text): Doc addition
533 * Makefile.am: Add Collation tests.
535 * tests/collate/*: Inputs and expected outputs for collation tests.
537 * tests/run-collate-tests.sh: Script to run collation tests.
539 * tests/unicode-collate.c (main): Rework slightly to make
540 it usable in unit tests. Also test g_utf8_collate_key_for_filename().
542 2005-10-01 Behdad Esfahbod <behdad@gnome.org>
544 * docs/reference/glib/tmpl/unicode.sgml:
545 * glib/gen-unicode-tables.pl:
547 * glib/gunichartables.h:
549 * tests/casefold.txt:
550 * tests/casemap.txt: Updated to Unicode 4.1. There are five new
551 GUnicodeBreakType types. That may break some applications, like
554 2005-09-26 Matthias Clasen <mclasen@redhat.com>
556 * glib/gstrfuncs.c (g_ascii_strtoull): Add details to the
557 docs. (#314393, Matthew F. Barnes)
560 * glib/gprintf.h: Remove g_snprintf() and g_vsnprintf(), since
561 they are already declared in glib.h. This doesn't break documented
562 use of gprintf.h, but should probably be pointed out in the
563 release notes for 2.10. (#314232, Behdad Esfahbod)
565 Tue Sep 20 13:16:04 2005 Tim Janik <timj@imendio.com>
567 * glib/gpattern.c (g_pattern_ph_match): applied significant recursion
568 complexity optimization, based on a patch from Matthias Clasen.
570 * tests/patterntest.c: more tests, mostly from matthias.
572 2005-09-20 Matthias Clasen <mclasen@redhat.com>
574 * glib/gqueue.c (g_queue_insert_sorted): Correct the docs.
575 (#316703, Mark Drago)
577 Mon Sep 19 17:23:23 2005 Tim Janik <timj@imendio.com>
579 * glib/gpattern.c: applied a patch from matthias which checks on the
580 upper bound of GPatternSpec length to optimize matches.
583 * tests/patterntest.c: added more match cases.
585 2005-09-16 Tor Lillqvist <tml@novell.com>
587 * glib/gstrfuncs.c (g_ascii_strcasecmp, g_ascii_strncasecmp): Add
588 warning to doc comment that these functions should not be used on
589 encodings like CP932.
591 2005-09-14 Matthias Clasen <mclasen@redhat.com>
593 * tests/keyfile-test.c: Add a test for grup names of length 1.
595 * glib/gkeyfile.c (g_key_file_line_is_group): Accept group names
596 of length 1. (#316309)
598 2005-09-12 Matthias Clasen <mclasen@redhat.com>
600 * glib/gmarkup.c (g_markup_escape_text): Clarify docs.
602 2005-09-11 Sebastian Wilhelmi <seppi@seppi.de>
604 * tests/refcount/Makefile.am (INCLUDES): Link the the refcount
605 tests to the system thread library $(G_THREAD_LIBS). Fixes #313744
608 2005-09-11 Kjartan Maraas <kmaraas@gnome.org>
610 * glib/gmain.c: (g_child_watch_prepare), (g_child_watch_check),
611 (child_watch_helper_thread): Remove some dead code. Closes
614 2005-09-07 Tor Lillqvist <tml@novell.com>
616 * glib/Makefile.am: Create also a console version of the
617 gspawn-win32-helper program, gspawn-win32-helper-console.exe.
618 It's otherwise identical to gspawn-win32-helper.exe, except marked
619 as a console application (linked without the -mwindows option).
621 * glib/gspawn-win32.c (do_spawn_directly, do_spawn_with_pipes):
622 Drop the dont_wait parameter. Its truth value correlated 100% with
623 the NULLness of the exit_status parameter anyway, so it's enough
624 to check whether exit_status is NULL. Invert the sense of the
625 dont_return_handle parameter and rename it to do_return_handle, to
626 make the code easier to read by avoiding double negations.
628 (g_spawn_sync_utf8, g_spawn_async_with_pipes_utf8): Modify calls
629 to do_spawn_with_pipes() accordingly.
631 (do_spawn_with_pipes): If we have a console, use the console
632 version of the helper program, otherwise use the GUI one. This
633 avoids extra console windows opening up in some situations. (In
634 case a console application uses the GUI gspawn-win32-helper.exe to
635 spawn another console application we would get a separate console
636 for the spawned console application).
638 * glib-zip.in: Distribute also gspawn-win32-helper-console.exe.
640 2005-09-05 Matthias Clasen <mclasen@redhat.com>
642 * glib/gmappedfile.c (g_mapped_file_new): Report an error
643 if the file is too large. (#315275, Kjartan Maraas)
645 * glib/gkeyfile.c (g_key_file_load_from_fd): The return value
646 of read() is signed. (#315273, Kjartan Maraas)
648 2005-08-31 Tor Lillqvist <tml@novell.com>
650 * glib/gutils.h: Wrapping atexit() is a bad idea on Windows, where
651 the EXE and each DLL have their own atexit function chains.
653 #define g_atexit as atexit instead. This means it has a
654 better chance of doing what the caller wants. For instance,
655 gtkhtml calls g_atexit() registering a function in gtkhtml
656 itself. This caused a crash when g_atexit() was implemented as a
657 function in the GLib DLL. The gtkhtml DLL was already unloaded by
658 the time the GLib DLL got unloaded.
660 * glib/gutils.c: #undef the #define mentioned above, to also get a
661 real g_atexit() into the DLL for backward compatibility. Document
662 the Windows behaviour of g_atexit(), and document the varying ways
663 atexit() can behave in the context of dynamically loaded modules
666 2005-08-31 Matthias Clasen <mclasen@redhat.com>
670 * glib/gdataset.c: Add string interning functions.
672 2005-08-28 Matthias Clasen <mclasen@redhat.com>
674 * glib/giochannel.c: Unify some near-duplicate strings. (#314654,
677 2005-08-26 Matthias Clasen <mclasen@redhat.com>
679 * configure.in: Bump version to 2.9.0
681 2005-08-25 Tor Lillqvist <tml@novell.com>
683 Make also the g_spawn*() functions take parameters in the GLib
684 file name encoding, i.e. UTF-8, on Windows. Has no impact on Unix
685 API or ABI. Like the other GLib API that was earlier changed to
686 use UTF-8 on Windows, the names of the functions that take UTF-8
687 have _utf8 suffixes added by using preprocessor macros in the
688 header file. The old names are kept for functions with the old
689 behaviour, taking parameters in the system codepage, for DLL ABI
692 * glib/gspawn.h: On Win32 add the suffix _utf8 to the names of the
693 g_spawn*() functions.
695 * glib/gspawn-win32.c: Use wide-char API on NT-based
696 Windows. Convert parameters from UTF-8 to wide chars (NT) or
697 system codepage (Win9x) and call the C library _wspawn*() or
698 spawn*() functions respectvely. Add DLL ABI stability versions
699 that take parameters in the system codepage.
701 * glib/gspawn-win32-helper.c: On NT-based Windows use the
702 wide-char versions of argv and envp, and use wide-char API to
703 change directory and spawn the program to run. Remove the verbose
704 debugging output, it was too complex to modify for the wide-char
705 features. (Just add temporary debugging printouts if needed, no
706 need to have them permanently in the source.)
708 * glib/gspawn.c: Corresponding documentation updates.
710 * glib/glib.symbols: Corresponding changes: Mark the ABI stability
711 symbols as PRIVATE, add the new _utf8-suffixed ones.
713 2005-08-24 Stepan Kasal <kasal@ucw.cz>
715 * glib/gtypes.h (G_MININT64): Cast the constant to gint64; it is
716 guint64 otherwise and that can produce warnings about comparison
717 between signed and unsigned.
719 2005-08-23 Matthias Clasen <mclasen@redhat.com>
721 * glib/gutils.c: Fix the crt_externs.h include.
723 2005-08-23 Stepan Kasal <kasal@ucw.cz>
725 * NEWS: Fix spelling of my first name.
727 2005-08-23 Matthias Clasen <mclasen@redhat.com>
731 * === Released 2.8.1 ===
735 2005-08-20 Hans Breuer <hans@breuer.org>
737 * glib/makefile.msc.in : link with ws2_32.lib
739 2005-08-18 Tor Lillqvist <tml@novell.com>
741 * configure.in: Check for <sys/wait.h>
743 * glib/gbacktrace.c: Include <sys/wait.h> on if HAVE_SYS_WAIT_H.
745 2005-08-18 Ross Burton <ross@burtonini.com>
748 Optimise single-character insertions.
753 * tests/string-test.c:
754 Add tests for new optimisation, and fix a leak.
756 2005-08-17 Matthias Clasen <mclasen@redhat.com>
758 * configure.in: Check for crt_externs.h and _NSGetEnviron.
760 * glib/gutils.c: On Darwin, include crt-externs.h and
761 define environ using _NSGetEnviron(). (#313731)
763 2005-08-16 Stepan Kasal <kasal@ucw.cz>
765 * glib/gutils.c (g_get_any_init): Move the body of the big if...
766 (g_get_any_init_do): ... to this new function.
767 (g_get_any_init): Declare as inline.
768 (g_get_any_init_locked): New inline function, does the locking.
769 Make use of these two throughout the code.
771 2005-08-15 Matthias Clasen <mclasen@redhat.com>
773 * glib/gbacktrace.c (g_on_error_stack_trace): Wait for
774 the child process and then simply return. This makes
775 The "S" option work as documented in g_on_error_query().
776 (#313125, Matthew F. Barnes)
778 * glib/gunicode.h: Update the link to Unicode category
779 values. (#313369, Behnam Esfahbod)
781 * glib/gqueue.c (g_queue_find_custom): Clarify docs
782 a little. (#311727, Tristan van Berkom)
784 * glib/abicheck.sh, gobject/abicheck.sh: Make the
785 check work on ia64 too, where some symbols ended up
786 in yet another section.