1 2005-12-08 Matthias Clasen <mclasen@redhat.com>
5 2005-12-07 Martyn Russell <martyn@imendio.com>
8 - Call g_queue_insert_sorted() instead of duplicating the code.
9 - Call g_queue_sort() instead of duplicating the code.
10 - Invert sort function results to make sure the same sort function
11 gives the same results across glist, gslist, gqueue and
14 * tests/asyncqueue-test.c:
15 - Updated the sort function to reflect the example in the
16 documentation for gasyncqueue.c.
18 2005-12-07 Martyn Russell <martyn@imendio.com>
20 * docs/reference/glib/glib-sections.txt:
21 * docs/reference/glib/tmpl/linked_lists_double.sgml:
22 * docs/reference/glib/tmpl/linked_lists_single.sgml:
25 - Added g_list_insert_sorted_with_data () and
26 g_slist_insert_sorted_with_data ().
27 - Removed the extra check in g_list_sort() and g_slist_sort() for
28 GCompareDataFunc vs. GCompareFunc.
30 2005-12-07 Tor Lillqvist <tml@novell.com>
32 * glib-gettextize.in: Look up prefix at run-time on Win32,
33 assuming the standard directory structure with glib-gettextize in
36 2005-12-06 Matthias Clasen <mclasen@redhat.com>
38 * glib/gmem.h: Exempt GTK+ from the mem chunk deprecation, since
39 we need GTK+ 2.8 to compile against GLib 2.10.
41 2005-12-05 Manish Singh <yosh@gimp.org>
43 * tests/run-collate-test.sh: set LC_COLLATE, not LANG, to be sure
44 to override any user settings.
46 2005-12-05 Matthias Clasen <mclasen@redhat.com>
48 * configure.in: Revert an accidential commit
50 2005-12-05 Matthias Clasen <mclasen@redhat.com>
52 * glib/gutf8.c: Documentation updates. (#323291, Morten Welinder)
54 * tests/Makefile.am (TESTS_ENVIRONMENT):
55 * tests/gobject/Makefile.am (TESTS_ENVIRONMENT): Set
56 MALLOC_CHECK_ and MALLOC_PERTURB_
58 * tests/run-collate-tests.sh: Run the collation tests explicitly
59 in en_US locale. (#320463)
61 * glib/gmem.h: Really deprecate GMemChunk.
68 * glib/gstrfuncs.c: Add versioned deprecation docs.
70 2005-12-05 Martyn Russell <martyn@imendio.com>
72 * docs/reference/glib/glib-sections.txt:
73 * glib/gasyncqueue.[ch]:
74 - Added support for sorting async queues by with _push_sorted(),
75 _push_sorted_unlocked(), _sort() and _sort_unlocked() (#323047).
78 * tests/asyncqueue-test.c:
79 - Added test case for gasyncqueue.c
81 Mon Dec 5 15:53:20 2005 Tim Janik <timj@imendio.com>
83 * glib/gslice.c: implement chain walking for arbitrary ->next pointer
84 offsets in g_slice_free_chain_with_offset() based on a patch by behdad
85 in bug 323178. moved time consuming logic from g_slice_free() out of
86 the inner loop, so g_slice_free_chain_with_offset() provides a real
87 performance benefit over g_slice_free1() now.
89 * glib/gslice.h: renamed g_slice_free_chain() to
90 g_slice_free_chain_with_offset(). implemented g_slice_free_chain() as
91 a type-safe macro as suggested in bug 323178.
92 simplified the macro implementation of g_slice_free() and implemented
93 it in a type safe manner for all compliers as suggested by Morten
94 Welinder <mortenw@gnome.org>.
99 * glib/glib.symbols: s/g_slice_free_chain/g_slice_free_chain_with_offset/
101 2005-12-05 Matthias Clasen <mclasen@redhat.com>
103 * glib/gasyncqueue.c: Add some docs.
105 * tests/libmoduletestplugin_a.c: Fix compiler warnings.
107 * glib/gatomic.c: In the ia64 implementation, use
108 __sync builtin without _si or _di suffix. (#321229,
109 Stanislav Brabec, patch by Andreas Schwab)
111 2005-12-04 Behdad Esfahbod <behdad@gnome.org>
113 * glib/gslice.h: Remove comma at the end of enum.
115 2005-12-04 Matthias Clasen <mclasen@redhat.com>
117 Handle multiple user names with the same UID better.
118 (#319535, Laszlo Peter)
120 * glib/gutils.c (g_get_any_init_do): When determining user
121 data, first look up $LOGNAME. If the UID doesn't match
122 getuid(), fall back to the current behaviour of looking
123 up the user data based on getuid().
125 2005-12-04 Matthias Clasen <mclasen@redhat.com>
127 * glib/gmacros.h (G_GNUC_WARN_UNUSED_RESULT): Add a macro
128 to make gcc warn if a function result is ignored. (#145466,
129 Arjan van de Ven, Alex Larsson)
131 * glib/gmem.h: Add the new attribute to g_realloc and
134 2005-12-03 Matthias Clasen <mclasen@redhat.com>
137 * glib/gcache.h: Deprecate g_cache_value_foreach. (#322956,
140 * glib/gmappedfile.c: Make mapping of empty files
143 * glib/gfileutils.c: Don't fork a new process just to
144 fix the permissions of the created temp file. (#321318,
145 Alexis S. L. Carvalho)
147 2005-12-02 Matthias Clasen <mclasen@redhat.com>
149 * README.in: Add a note about Solaris threads.
152 * configure.in: Remove support for Solaris threads.
153 (#136971, Sebastian Wilhelmi, patch by Andrew Paprocki)
155 2005-12-02 Matthias Clasen <mclasen@redhat.com>
157 * configure.in: Define G_GUINT64_CONSTANT in analogy to
158 G_GINT64_CONSTANT. (#322568, Andrew Paprocki)
160 2005-12-02 Matthias Clasen <mclasen@redhat.com>
162 * glib/gslice.c: Win32 portability fixes and C99ism removal,
163 pointed out by Kazuki Iwamoto. (#323052)
165 2005-12-02 Matthias Clasen <mclasen@redhat.com>
167 * glib/gatomic.c (g_atomic_int_add): Add a missing volatile for
168 the IA64 implementation.
170 Fri Dec 2 16:18:09 2005 Tim Janik <timj@imendio.com>
172 * glib/gslice.c: some naming and type size fixups.
174 Fri Dec 2 13:08:58 2005 Tim Janik <timj@gtk.org>
176 * glib/gslice.h (g_slice_free): reworked GCC-specific type-safe macro
177 variant into something less verbose, but digestible for gcc-3.4.
179 Fri Dec 2 10:55:07 2005 Tim Janik <timj@imendio.com>
181 * tests/slice-test.c: extended to perform the benchmarking on the old
182 memchunk code if 'O' is selected.
184 * tests/memchunks.c: new file which contains the old GLib mem chunks
185 implementation with prefix old_mem_chunk_.
187 * tests/Makefile.am: added memchunks.c
189 Fri Dec 2 00:16:59 2005 Tim Janik <timj@imendio.com>
191 * glib/gslice.c: improved gettimeofday() timer resolution by maintaining
192 acceptable load distribution of the syscall.
193 (allocator_get_magazine_threshold): reduce minimum/single-thread magazine
194 sizes to approximately page_size/2.5 to avoid excessive startup allocations.
196 2005-12-01 Matthias Clasen <mclasen@redhat.com>
198 * tests/gobject/ifacecheck.c (test_iface_base_init): Don't
199 double-free base interface members.
201 * tests/Makefile.am: Remove duplicate variable.
203 Thu Dec 1 17:32:46 2005 Tim Janik <timj@imendio.com>
205 * glib/gslice.[hc]: new slice allocator implementation.
207 * tests/slice-test.c: added random slice allocation test.
209 * glib/gthread.[hc]: removed newly added private thread mem API.
211 * glib/gthreadinit.h:
214 * glib/gmem.c: divided glib threading initialisation into three phases,
215 initialisation where private keys and messaging are not available (only
216 needed by gmem.c), initialisation without messaging but private keys
217 available (gslice.c, gmessage.c), and full fledged initialisers that
218 server the rest of glib. initialisation functions got renamed to reflect
219 the limitations of their corresponding phases.
221 * glib/gmem.c: removed memchunk code, defer allocations to
224 * glib/gmem.[hc]: removed g_slice_* skeletons.
226 * glib/glib.symbols: added g_slice_* symbols.
228 * configure.in: check for availability of posix_memalign(3), memalign(3)
231 * glib/Makefile.am: added gslice.[hc].
233 2005-12-01 Tor Lillqvist <tml@novell.com>
235 * glib/gstdio.c (g_stat): In the Win32 implementation, strip
236 trailing slash(es) for non-root folders. stat() fails if non-root
237 folders are specified with trailing slashes. It's too much hassle
238 to demand that callers strip such slashes themselves, especially
239 as it is easy to get it wrong and strip the slash of a root
241 (g_rename): On NT-based Windows, use MoveFileEx() with
242 MOVEFILE_REPLACE_EXISTING to better match Unix behaviour.
244 2005-11-28 Matthias Clasen <mclasen@redhat.com>
246 Fix G_STMT_START / G_STMT_END on Solaris. (#321972,
249 * configure.in: Check whether do { } while (0) works.
251 * glib/gmacros.h: Use do { } while (0) for G_STMT_START /
252 G_STMT_END if it works.
254 2005-11-28 Matthias Clasen <mclasen@redhat.com>
256 * glib/gthread.c (g_static_rw_lock_wait, g_static_rw_lock_signal):
257 * glib/gnode.c (g_node_depth_traverse_level):
258 * glib/gmem.c (g_allocator_new):
259 * glib/ghash.c (g_hash_table_unref, g_hash_table_destroy)
260 (g_hash_table_foreach_remove_or_steal): Silence compiler
263 2005-11-27 Matthias Clasen <mclasen@redhat.com>
265 * glib/gunicollate.c (g_utf8_collate_key_for_filename):
266 Don't read beyond len. (#322520, Christian Persch)
268 Wed Nov 23 17:34:01 2005 Tim Janik <timj@imendio.com>
270 * glib/gdataset.c: access datalist flags via atomic pointer access
271 functions, instead of acquiring the dataset lock. this is faster and
272 also matches the atomic pointer readouts in gdatalistprivate.h and
273 direct pointer modifications required by gobject.c.
275 Wed Nov 23 13:35:31 2005 Tim Janik <timj@gtk.org>
277 * glib/gdataset.c: streamlined initialization code somewhat,
278 removed GData node cache.
280 2005-11-22 Matthias Clasen <mclasen@redhat.com>
282 * glib/ghash.c (g_hash_table_ref):
283 (g_hash_table_unref): Mark these as new API.
285 Avoid double locking in g_intern_string (#322133,
288 * glib/gdataset.c (g_quark_from_string_internal): New
289 internal function which factors out common parts of
290 g_quark_from[_static]_string.
291 (g_quark_from_string, g_quark_from_static_string):
292 Use g_quark_from_string_internal.
293 (g_intern_string, g_intern_static_string): Use
294 g_quark_from_string_internal, and only take the
296 (g_quark_new): Don't store the strings shifted by -1
297 in the g_quarks array.
298 (g_quark_to_string): Adapt to the previous change.
300 Tue Nov 22 14:04:26 2005 Tim Janik <timj@imendio.com>
304 g_hash_table_new_full(): create hash tables with a ref count of 1.
305 g_hash_table_ref(): atomically ref_count+=1
306 g_hash_table_unref(): atomically ref_count-=1, destroys hash table
307 when refcount reaches 0.
308 g_hash_table_destroy(): just destroy keys and values, unref by 1.
309 g_hash_table_insert():
310 g_hash_table_replace(): assert ref_count>0.
313 * glib/gatomic.c: added 'volatile' qualifier to all atomic pointer and
316 2005-11-20 Behdad Esfahbod <behdad@gnome.org>
318 * glib/guniprop.c (g_unichar_get_mirror_char): Remove unused
319 variables. (#321984, Andrew Paprocki)
321 2005-11-18 Matthias Clasen <mclasen@redhat.com>
323 * configure.in: Bump version
325 2005-11-17 Matthias Clasen <mclasen@redhat.com>
327 * === Released 2.9.0 ===
329 * glib/glib.symbols: Updates
331 * glib/gutf8.c (g_utf8_offset_to_pointer): Add some performance
332 hints to the docs. (Paolo Borelli)
336 * glib/gthread.c (g_thread_foreach): Mark as new api.
338 * README.in: Updates.
342 * glib/gdate.c (g_date_set_time_t):
343 (g_date_set_time_val): Add functions to set a date from
344 a time_t and from a GTimeVal.
345 (g_date_set_time): Deprecate. (#314067, Roger Leigh)
348 * tests/utf8-pointer.c: Unit tests for g_utf8_pointer_to_offset
349 and g_utf8_offset_to_pointer.
351 * glib/gutf8.c (g_utf8_pointer_to_offset)
352 (g_utf8_offset_to_pointer): Handle negative offsets, and use
353 "stutter stepping" for going backwards. (#320638, Larry
358 * glib/gthread.c: const correctness fixes, found
359 by Arjan van de Ven and gcc.
361 2005-11-16 Behdad Esfahbod <behdad@gnome.org>
363 * docs/reference/glib/tmpl/unicode.sgml: Correct typo on mentioning
364 @G_UNICODE_BREAK_UNKNOWN that should be #G_UNICODE_BREAK_UNKNOWN.
366 2005-11-09 Behdad Esfahbod <behdad@gnome.org>
368 * glib/guniprop.c: Use bit hacks instead when checking a general
369 category value against multiple values.
371 * glib/gutf8.c: Change ISO10646 to Unicode in docs.
373 2005-11-10 Simos Xenitellis <simos@gnome.org>
375 * configure.in: Added tt (Tatar) to ALL_LINGUAS.
377 2005-11-05 Tor Lillqvist <tml@novell.com>
379 * glib/giowin32.c (g_io_win32_set_flags): Don't call g_warning().
380 (#320688) Instead set error and return failure.
382 2005-11-07 Matthias Clasen <mclasen@redhat.com>
384 * glib/gmain.c (g_main_dispatch): Don't call cb_funcs->unref
385 while holding the context lock. (#320886, Andy Wingo)
387 2005-11-07 Billy Biggs <vektor@dumbterm.net>
389 * tests/refcount/Makefile.am: Add a missing reference to libglib
390 in the LDADD for this test.
392 2005-11-07 Matthias Clasen <mclasen@redhat.com>
394 * glib/gmem.c (g_allocator_new): We need to set n_preallocs to a
395 nonzero value, otherwise GTK+ 2.8 breaks when compiled against
396 GLib 2.9. (#320755, Luca Ferretti)
398 2005-11-04 Matthias Clasen <mclasen@redhat.com>
400 * glib/Makefile.am: Apply a patch from Behdad Esfahbod to
401 use a faster lookup table for g_unichar_get_mirror_char().
402 * glib/gmirroringtable.h: The new table.
403 * glib/gunichartables.h: Remove bidi_mirroring_table.
404 * glib/gen-unicode-tables.pl: Don't generate the mirroring
406 * glib/glib-mirroring-tab/*: Sources for the program
407 which generated gmirroringtable.h.
409 * glib/glist.c: Avoid some code duplication.
411 * glib/gscanner.h: Include gdataset.h. (#320322)
413 * glib/gdebug.h: add new GDebugFlag for fatal_criticals
414 * glib/gmessages.c: (_g_debug_init): handle G_DEBUG=fatal_criticals,
415 to help find critical warnings in applications. (#320017,
418 2005-11-02 Tor Lillqvist <tml@novell.com>
420 * glib/glib.symbols: Remove large amount of trailing whitespace
421 from one line. Remove #ifdef G_UNDEFINED from around
422 g_hash_table_new and g_hash_table_new_full symbols, what was the
423 purpose of that? Add the g_slice functions.
425 Tue Nov 1 16:24:20 2005 Tim Janik <timj@imendio.com>
427 * glib/gmem.[hc]: prepared deprecation of GMemChunk and GAllocator.
428 added g_slice_*() API to allocate and cache small bits of memory.
429 an actuall allocator implementation for g_slice_*() is still pending.
431 * glib/gthread.[hc]: changes from a patch by Matthias Clasen.
432 changed GRealThread list to use in-structure *next; fields instead
433 of GSList, in order for thread iteration to not depenend on g_slice_*()
435 _g_thread_mem_private_get():
436 _g_thread_mem_private_set(): added accessors for private memory,
437 needed because the ordinary GPrivate implementation relies on GArray
438 and GSList and therefore indirectly on working g_slice_*() allocations.
441 g_thread_foreach(): new public API function to loop over all existing threads.
455 removed GAllocator and free list usages and accompanying locks.
456 use g_slice_*() API to allocate and cache small bits of memory.
458 * glib/ghook.h: removed GMemChunk field from public API.
461 * glib/glist.h: deprecate allocator API, provide _free1() for consistency.
463 * glib/gnode.h: deprecate allocator API.
465 * glib/gmain.c: reordered GPollRec fields so g_slice_free_chain() can
466 be used for poll rec lists.
468 * glib/grel.c: removed mem chunk usage, and allocated tuples via g_slice_*().
469 g_relation_destroy(): free all tuples from the all_tuples hash table,
470 this effectively maintains the life time track keeping of tuples.
471 g_relation_delete_tuple(): free tuples which are removed from the
472 all_tuples hash table. this fixes a temporary leak that was present
473 in the memchunk code until the destruction of the relation.
475 2005-10-29 Matthias Clasen <mclasen@redhat.com>
477 * tests/convert-test.c: Add some tests for conversions between
478 UTF-8, UCS-4 and UTF-16.
480 * glib/gutf8.c (g_utf8_to_ucs4, g_utf8_to_utf16): Fix handling
481 of len == -1, noticed by Morten Welinder.
483 2005-10-27 Erdal Ronahi <erdal.ronahi@gmail.com>
485 * configure.in: Added ku (Kurdish) to ALL_LINGUAS
487 2005-10-26 Matthias Clasen <mclasen@redhat.com>
489 * glib/gutf8.c (g_ucs4_to_utf8): Don't set items_read twice
490 in the error case, and add some documentation. (#319806, Morten
493 2005-10-19 Manish Singh <yosh@gimp.org>
495 * configure.in: Use AC_CHECK_FUNCS for _NSGetEnviron, to get the
496 config.h symbol automatically. Fixes bug #313731.
498 2005-10-19 Tor Lillqvist <tml@novell.com>
500 * glib/gwin32.c (g_win32_get_package_installation_directory):
501 Return a g_strdup()ed copy of the value stored in the hash table,
502 so that it can be g_free()d without leaving a dangling pointer in
503 the hash table. (#319232)
505 2005-10-06 Matthias Clasen <mclasen@redhat.com>
507 * glib/gunicollate.c (g_utf8_collate_key_for_filename): Handle
508 all-zero sequences correctly. (#317930, Sebastien Bacher)
510 2005-10-05 Matthias Clasen <mclasen@redhat.com>
512 * glib/gmarkup.c (g_markup_escape_text): Doc addition
514 * Makefile.am: Add Collation tests.
516 * tests/collate/*: Inputs and expected outputs for collation tests.
518 * tests/run-collate-tests.sh: Script to run collation tests.
520 * tests/unicode-collate.c (main): Rework slightly to make
521 it usable in unit tests. Also test g_utf8_collate_key_for_filename().
523 2005-10-01 Behdad Esfahbod <behdad@gnome.org>
525 * docs/reference/glib/tmpl/unicode.sgml:
526 * glib/gen-unicode-tables.pl:
528 * glib/gunichartables.h:
530 * tests/casefold.txt:
531 * tests/casemap.txt: Updated to Unicode 4.1. There are five new
532 GUnicodeBreakType types. That may break some applications, like
535 2005-09-26 Matthias Clasen <mclasen@redhat.com>
537 * glib/gstrfuncs.c (g_ascii_strtoull): Add details to the
538 docs. (#314393, Matthew F. Barnes)
541 * glib/gprintf.h: Remove g_snprintf() and g_vsnprintf(), since
542 they are already declared in glib.h. This doesn't break documented
543 use of gprintf.h, but should probably be pointed out in the
544 release notes for 2.10. (#314232, Behdad Esfahbod)
546 Tue Sep 20 13:16:04 2005 Tim Janik <timj@imendio.com>
548 * glib/gpattern.c (g_pattern_ph_match): applied significant recursion
549 complexity optimization, based on a patch from Matthias Clasen.
551 * tests/patterntest.c: more tests, mostly from matthias.
553 2005-09-20 Matthias Clasen <mclasen@redhat.com>
555 * glib/gqueue.c (g_queue_insert_sorted): Correct the docs.
556 (#316703, Mark Drago)
558 Mon Sep 19 17:23:23 2005 Tim Janik <timj@imendio.com>
560 * glib/gpattern.c: applied a patch from matthias which checks on the
561 upper bound of GPatternSpec length to optimize matches.
564 * tests/patterntest.c: added more match cases.
566 2005-09-16 Tor Lillqvist <tml@novell.com>
568 * glib/gstrfuncs.c (g_ascii_strcasecmp, g_ascii_strncasecmp): Add
569 warning to doc comment that these functions should not be used on
570 encodings like CP932.
572 2005-09-14 Matthias Clasen <mclasen@redhat.com>
574 * tests/keyfile-test.c: Add a test for grup names of length 1.
576 * glib/gkeyfile.c (g_key_file_line_is_group): Accept group names
577 of length 1. (#316309)
579 2005-09-12 Matthias Clasen <mclasen@redhat.com>
581 * glib/gmarkup.c (g_markup_escape_text): Clarify docs.
583 2005-09-11 Sebastian Wilhelmi <seppi@seppi.de>
585 * tests/refcount/Makefile.am (INCLUDES): Link the the refcount
586 tests to the system thread library $(G_THREAD_LIBS). Fixes #313744
589 2005-09-11 Kjartan Maraas <kmaraas@gnome.org>
591 * glib/gmain.c: (g_child_watch_prepare), (g_child_watch_check),
592 (child_watch_helper_thread): Remove some dead code. Closes
595 2005-09-07 Tor Lillqvist <tml@novell.com>
597 * glib/Makefile.am: Create also a console version of the
598 gspawn-win32-helper program, gspawn-win32-helper-console.exe.
599 It's otherwise identical to gspawn-win32-helper.exe, except marked
600 as a console application (linked without the -mwindows option).
602 * glib/gspawn-win32.c (do_spawn_directly, do_spawn_with_pipes):
603 Drop the dont_wait parameter. Its truth value correlated 100% with
604 the NULLness of the exit_status parameter anyway, so it's enough
605 to check whether exit_status is NULL. Invert the sense of the
606 dont_return_handle parameter and rename it to do_return_handle, to
607 make the code easier to read by avoiding double negations.
609 (g_spawn_sync_utf8, g_spawn_async_with_pipes_utf8): Modify calls
610 to do_spawn_with_pipes() accordingly.
612 (do_spawn_with_pipes): If we have a console, use the console
613 version of the helper program, otherwise use the GUI one. This
614 avoids extra console windows opening up in some situations. (In
615 case a console application uses the GUI gspawn-win32-helper.exe to
616 spawn another console application we would get a separate console
617 for the spawned console application).
619 * glib-zip.in: Distribute also gspawn-win32-helper-console.exe.
621 2005-09-05 Matthias Clasen <mclasen@redhat.com>
623 * glib/gmappedfile.c (g_mapped_file_new): Report an error
624 if the file is too large. (#315275, Kjartan Maraas)
626 * glib/gkeyfile.c (g_key_file_load_from_fd): The return value
627 of read() is signed. (#315273, Kjartan Maraas)
629 2005-08-31 Tor Lillqvist <tml@novell.com>
631 * glib/gutils.h: Wrapping atexit() is a bad idea on Windows, where
632 the EXE and each DLL have their own atexit function chains.
634 #define g_atexit as atexit instead. This means it has a
635 better chance of doing what the caller wants. For instance,
636 gtkhtml calls g_atexit() registering a function in gtkhtml
637 itself. This caused a crash when g_atexit() was implemented as a
638 function in the GLib DLL. The gtkhtml DLL was already unloaded by
639 the time the GLib DLL got unloaded.
641 * glib/gutils.c: #undef the #define mentioned above, to also get a
642 real g_atexit() into the DLL for backward compatibility. Document
643 the Windows behaviour of g_atexit(), and document the varying ways
644 atexit() can behave in the context of dynamically loaded modules
647 2005-08-31 Matthias Clasen <mclasen@redhat.com>
651 * glib/gdataset.c: Add string interning functions.
653 2005-08-28 Matthias Clasen <mclasen@redhat.com>
655 * glib/giochannel.c: Unify some near-duplicate strings. (#314654,
658 2005-08-26 Matthias Clasen <mclasen@redhat.com>
660 * configure.in: Bump version to 2.9.0
662 2005-08-25 Tor Lillqvist <tml@novell.com>
664 Make also the g_spawn*() functions take parameters in the GLib
665 file name encoding, i.e. UTF-8, on Windows. Has no impact on Unix
666 API or ABI. Like the other GLib API that was earlier changed to
667 use UTF-8 on Windows, the names of the functions that take UTF-8
668 have _utf8 suffixes added by using preprocessor macros in the
669 header file. The old names are kept for functions with the old
670 behaviour, taking parameters in the system codepage, for DLL ABI
673 * glib/gspawn.h: On Win32 add the suffix _utf8 to the names of the
674 g_spawn*() functions.
676 * glib/gspawn-win32.c: Use wide-char API on NT-based
677 Windows. Convert parameters from UTF-8 to wide chars (NT) or
678 system codepage (Win9x) and call the C library _wspawn*() or
679 spawn*() functions respectvely. Add DLL ABI stability versions
680 that take parameters in the system codepage.
682 * glib/gspawn-win32-helper.c: On NT-based Windows use the
683 wide-char versions of argv and envp, and use wide-char API to
684 change directory and spawn the program to run. Remove the verbose
685 debugging output, it was too complex to modify for the wide-char
686 features. (Just add temporary debugging printouts if needed, no
687 need to have them permanently in the source.)
689 * glib/gspawn.c: Corresponding documentation updates.
691 * glib/glib.symbols: Corresponding changes: Mark the ABI stability
692 symbols as PRIVATE, add the new _utf8-suffixed ones.
694 2005-08-24 Stepan Kasal <kasal@ucw.cz>
696 * glib/gtypes.h (G_MININT64): Cast the constant to gint64; it is
697 guint64 otherwise and that can produce warnings about comparison
698 between signed and unsigned.
700 2005-08-23 Matthias Clasen <mclasen@redhat.com>
702 * glib/gutils.c: Fix the crt_externs.h include.
704 2005-08-23 Stepan Kasal <kasal@ucw.cz>
706 * NEWS: Fix spelling of my first name.
708 2005-08-23 Matthias Clasen <mclasen@redhat.com>
712 * === Released 2.8.1 ===
716 2005-08-20 Hans Breuer <hans@breuer.org>
718 * glib/makefile.msc.in : link with ws2_32.lib
720 2005-08-18 Tor Lillqvist <tml@novell.com>
722 * configure.in: Check for <sys/wait.h>
724 * glib/gbacktrace.c: Include <sys/wait.h> on if HAVE_SYS_WAIT_H.
726 2005-08-18 Ross Burton <ross@burtonini.com>
729 Optimise single-character insertions.
734 * tests/string-test.c:
735 Add tests for new optimisation, and fix a leak.
737 2005-08-17 Matthias Clasen <mclasen@redhat.com>
739 * configure.in: Check for crt_externs.h and _NSGetEnviron.
741 * glib/gutils.c: On Darwin, include crt-externs.h and
742 define environ using _NSGetEnviron(). (#313731)
744 2005-08-16 Stepan Kasal <kasal@ucw.cz>
746 * glib/gutils.c (g_get_any_init): Move the body of the big if...
747 (g_get_any_init_do): ... to this new function.
748 (g_get_any_init): Declare as inline.
749 (g_get_any_init_locked): New inline function, does the locking.
750 Make use of these two throughout the code.
752 2005-08-15 Matthias Clasen <mclasen@redhat.com>
754 * glib/gbacktrace.c (g_on_error_stack_trace): Wait for
755 the child process and then simply return. This makes
756 The "S" option work as documented in g_on_error_query().
757 (#313125, Matthew F. Barnes)
759 * glib/gunicode.h: Update the link to Unicode category
760 values. (#313369, Behnam Esfahbod)
762 * glib/gqueue.c (g_queue_find_custom): Clarify docs
763 a little. (#311727, Tristan van Berkom)
765 * glib/abicheck.sh, gobject/abicheck.sh: Make the
766 check work on ia64 too, where some symbols ended up
767 in yet another section.