1 2005-12-25 Matthias Clasen <mclasen@redhat.com>
3 * tests/slice-test.c: Fix C99isms. (#324950, Dan Yefimov)
5 2005-12-21 Matthias Clasen <mclasen@redhat.com>
7 * README.in: Add some notes on when 'make check' may fail.
9 2005-12-20 Matthias Clasen <mclasen@redhat.com>
13 * glib/gthreadpool.c (g_thread_pool_set_sort_function): New function
14 to sort tasks pushed into a threadpool. (#324479, Martyn Russell)
16 * tests/threadpool-test.c: Test this.
18 Tue Dec 20 18:14:14 2005 Tim Janik <timj@imendio.com>
20 * glib/gslice.[hc]: added mem_error() and mem_assert() to test and
21 handle errors without depending on gmessage.c which might not be
22 setup when the error occours.
23 removed G_SLICE_CONFIG_ALWAYS_FREE config option, fixed the code so
24 always freeing can be achieved by adjusting the working set time to
25 0 with G_SLICE_CONFIG_WORKING_SET_MSECS.
26 added G_SLICE_CONFIG_COLOR_INCREMENT to test different color increments
27 (mainly 0 and 1). reduced the minimum block size to 128 bytes, to
28 minimize wastage if small amounts of differently sized structrues are
29 allocated, this does come at a performance cost of roughly 5% though.
30 fixed up block alignment calculation, so it works for varying
31 block sizes. only use strerror() not g_strerror() since the latter
32 depends on working GQuark and GSlice.
33 mem_error(): implemented in terms of fprintf and vfprintf.
35 * tests/slice-color.c: new program to test cache colorization effects.
37 * tests/slice-test.c: trade G_SLICE_CONFIG_ALWAYS_FREE for 0 duration
38 G_SLICE_CONFIG_WORKING_SET_MSECS.
40 2005-12-17 Matthias Clasen <mclasen@redhat.com>
42 * glib/goption.c (parse_short_option): Set an error in all
43 failure cases. (#324332, Tim-Philipp Müller)
45 2005-12-17 Sebastian Wilhelmi <seppi@seppi.de>
47 * glib/gatomic.c: Fix memory barrier position in g_atomic_int_get
48 and g_atomic_pointer_get. Add g_atomic_int_set and
49 g_atomic_pointer_set implementations for the !DEFINE_WITH_MUTEXES &&
50 G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case, as well as defining them
51 as functions (additionally to the macros in the header) for the
52 !G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case.
54 2005-12-16 Matthias Clasen <mclasen@redhat.com>
56 * glib/gmem.c (g_allocator_new): Don't return a pointer to
57 a const struct, since apps expect to be able to modify it.
58 (#324179, J. Ali Harlow)
60 Tue Dec 13 10:13:32 2005 Tim Janik <timj@imendio.com>
62 * glib/gatomic.h: added g_atomic_pointer_set() and g_atomic_int_set()
63 for systems where the initialization of atomic variables requires a
66 2005-12-09 Matthias Clasen <mclasen@redhat.com>
68 * configure.in: Bump version
70 * === Released 2.9.1 ===
72 2005-12-09 Alexander Larsson <alexl@redhat.com>
76 Use G_GNUC_WARN_UNUSED_RESULT on list functions that return
79 2005-12-08 Matthias Clasen <mclasen@redhat.com>
83 2005-12-07 Martyn Russell <martyn@imendio.com>
86 - Call g_queue_insert_sorted() instead of duplicating the code.
87 - Call g_queue_sort() instead of duplicating the code.
88 - Invert sort function results to make sure the same sort function
89 gives the same results across glist, gslist, gqueue and
92 * tests/asyncqueue-test.c:
93 - Updated the sort function to reflect the example in the
94 documentation for gasyncqueue.c.
96 2005-12-07 Martyn Russell <martyn@imendio.com>
98 * docs/reference/glib/glib-sections.txt:
99 * docs/reference/glib/tmpl/linked_lists_double.sgml:
100 * docs/reference/glib/tmpl/linked_lists_single.sgml:
103 - Added g_list_insert_sorted_with_data () and
104 g_slist_insert_sorted_with_data ().
105 - Removed the extra check in g_list_sort() and g_slist_sort() for
106 GCompareDataFunc vs. GCompareFunc.
108 2005-12-07 Tor Lillqvist <tml@novell.com>
110 * glib-gettextize.in: Look up prefix at run-time on Win32,
111 assuming the standard directory structure with glib-gettextize in
114 2005-12-06 Matthias Clasen <mclasen@redhat.com>
116 * glib/gmem.h: Exempt GTK+ from the mem chunk deprecation, since
117 we need GTK+ 2.8 to compile against GLib 2.10.
119 2005-12-05 Manish Singh <yosh@gimp.org>
121 * tests/run-collate-test.sh: set LC_COLLATE, not LANG, to be sure
122 to override any user settings.
124 2005-12-05 Matthias Clasen <mclasen@redhat.com>
126 * configure.in: Revert an accidential commit
128 2005-12-05 Matthias Clasen <mclasen@redhat.com>
130 * glib/gutf8.c: Documentation updates. (#323291, Morten Welinder)
132 * tests/Makefile.am (TESTS_ENVIRONMENT):
133 * tests/gobject/Makefile.am (TESTS_ENVIRONMENT): Set
134 MALLOC_CHECK_ and MALLOC_PERTURB_
136 * tests/run-collate-tests.sh: Run the collation tests explicitly
137 in en_US locale. (#320463)
139 * glib/gmem.h: Really deprecate GMemChunk.
146 * glib/gstrfuncs.c: Add versioned deprecation docs.
148 2005-12-05 Martyn Russell <martyn@imendio.com>
150 * docs/reference/glib/glib-sections.txt:
151 * glib/gasyncqueue.[ch]:
152 - Added support for sorting async queues by with _push_sorted(),
153 _push_sorted_unlocked(), _sort() and _sort_unlocked() (#323047).
156 * tests/asyncqueue-test.c:
157 - Added test case for gasyncqueue.c
159 Mon Dec 5 15:53:20 2005 Tim Janik <timj@imendio.com>
161 * glib/gslice.c: implement chain walking for arbitrary ->next pointer
162 offsets in g_slice_free_chain_with_offset() based on a patch by behdad
163 in bug 323178. moved time consuming logic from g_slice_free() out of
164 the inner loop, so g_slice_free_chain_with_offset() provides a real
165 performance benefit over g_slice_free1() now.
167 * glib/gslice.h: renamed g_slice_free_chain() to
168 g_slice_free_chain_with_offset(). implemented g_slice_free_chain() as
169 a type-safe macro as suggested in bug 323178.
170 simplified the macro implementation of g_slice_free() and implemented
171 it in a type safe manner for all compliers as suggested by Morten
172 Welinder <mortenw@gnome.org>.
177 * glib/glib.symbols: s/g_slice_free_chain/g_slice_free_chain_with_offset/
179 2005-12-05 Matthias Clasen <mclasen@redhat.com>
181 * glib/gasyncqueue.c: Add some docs.
183 * tests/libmoduletestplugin_a.c: Fix compiler warnings.
185 * glib/gatomic.c: In the ia64 implementation, use
186 __sync builtin without _si or _di suffix. (#321229,
187 Stanislav Brabec, patch by Andreas Schwab)
189 2005-12-04 Behdad Esfahbod <behdad@gnome.org>
191 * glib/gslice.h: Remove comma at the end of enum.
193 2005-12-04 Matthias Clasen <mclasen@redhat.com>
195 Handle multiple user names with the same UID better.
196 (#319535, Laszlo Peter)
198 * glib/gutils.c (g_get_any_init_do): When determining user
199 data, first look up $LOGNAME. If the UID doesn't match
200 getuid(), fall back to the current behaviour of looking
201 up the user data based on getuid().
203 2005-12-04 Matthias Clasen <mclasen@redhat.com>
205 * glib/gmacros.h (G_GNUC_WARN_UNUSED_RESULT): Add a macro
206 to make gcc warn if a function result is ignored. (#145466,
207 Arjan van de Ven, Alex Larsson)
209 * glib/gmem.h: Add the new attribute to g_realloc and
212 2005-12-03 Matthias Clasen <mclasen@redhat.com>
215 * glib/gcache.h: Deprecate g_cache_value_foreach. (#322956,
218 * glib/gmappedfile.c: Make mapping of empty files
221 * glib/gfileutils.c: Don't fork a new process just to
222 fix the permissions of the created temp file. (#321318,
223 Alexis S. L. Carvalho)
225 2005-12-02 Matthias Clasen <mclasen@redhat.com>
227 * README.in: Add a note about Solaris threads.
230 * configure.in: Remove support for Solaris threads.
231 (#136971, Sebastian Wilhelmi, patch by Andrew Paprocki)
233 2005-12-02 Matthias Clasen <mclasen@redhat.com>
235 * configure.in: Define G_GUINT64_CONSTANT in analogy to
236 G_GINT64_CONSTANT. (#322568, Andrew Paprocki)
238 2005-12-02 Matthias Clasen <mclasen@redhat.com>
240 * glib/gslice.c: Win32 portability fixes and C99ism removal,
241 pointed out by Kazuki Iwamoto. (#323052)
243 2005-12-02 Matthias Clasen <mclasen@redhat.com>
245 * glib/gatomic.c (g_atomic_int_add): Add a missing volatile for
246 the IA64 implementation.
248 Fri Dec 2 16:18:09 2005 Tim Janik <timj@imendio.com>
250 * glib/gslice.c: some naming and type size fixups.
252 Fri Dec 2 13:08:58 2005 Tim Janik <timj@gtk.org>
254 * glib/gslice.h (g_slice_free): reworked GCC-specific type-safe macro
255 variant into something less verbose, but digestible for gcc-3.4.
257 Fri Dec 2 10:55:07 2005 Tim Janik <timj@imendio.com>
259 * tests/slice-test.c: extended to perform the benchmarking on the old
260 memchunk code if 'O' is selected.
262 * tests/memchunks.c: new file which contains the old GLib mem chunks
263 implementation with prefix old_mem_chunk_.
265 * tests/Makefile.am: added memchunks.c
267 Fri Dec 2 00:16:59 2005 Tim Janik <timj@imendio.com>
269 * glib/gslice.c: improved gettimeofday() timer resolution by maintaining
270 acceptable load distribution of the syscall.
271 (allocator_get_magazine_threshold): reduce minimum/single-thread magazine
272 sizes to approximately page_size/2.5 to avoid excessive startup allocations.
274 2005-12-01 Matthias Clasen <mclasen@redhat.com>
276 * tests/gobject/ifacecheck.c (test_iface_base_init): Don't
277 double-free base interface members.
279 * tests/Makefile.am: Remove duplicate variable.
281 Thu Dec 1 17:32:46 2005 Tim Janik <timj@imendio.com>
283 * glib/gslice.[hc]: new slice allocator implementation.
285 * tests/slice-test.c: added random slice allocation test.
287 * glib/gthread.[hc]: removed newly added private thread mem API.
289 * glib/gthreadinit.h:
292 * glib/gmem.c: divided glib threading initialisation into three phases,
293 initialisation where private keys and messaging are not available (only
294 needed by gmem.c), initialisation without messaging but private keys
295 available (gslice.c, gmessage.c), and full fledged initialisers that
296 server the rest of glib. initialisation functions got renamed to reflect
297 the limitations of their corresponding phases.
299 * glib/gmem.c: removed memchunk code, defer allocations to
302 * glib/gmem.[hc]: removed g_slice_* skeletons.
304 * glib/glib.symbols: added g_slice_* symbols.
306 * configure.in: check for availability of posix_memalign(3), memalign(3)
309 * glib/Makefile.am: added gslice.[hc].
311 2005-12-01 Tor Lillqvist <tml@novell.com>
313 * glib/gstdio.c (g_stat): In the Win32 implementation, strip
314 trailing slash(es) for non-root folders. stat() fails if non-root
315 folders are specified with trailing slashes. It's too much hassle
316 to demand that callers strip such slashes themselves, especially
317 as it is easy to get it wrong and strip the slash of a root
319 (g_rename): On NT-based Windows, use MoveFileEx() with
320 MOVEFILE_REPLACE_EXISTING to better match Unix behaviour.
322 2005-11-28 Matthias Clasen <mclasen@redhat.com>
324 Fix G_STMT_START / G_STMT_END on Solaris. (#321972,
327 * configure.in: Check whether do { } while (0) works.
329 * glib/gmacros.h: Use do { } while (0) for G_STMT_START /
330 G_STMT_END if it works.
332 2005-11-28 Matthias Clasen <mclasen@redhat.com>
334 * glib/gthread.c (g_static_rw_lock_wait, g_static_rw_lock_signal):
335 * glib/gnode.c (g_node_depth_traverse_level):
336 * glib/gmem.c (g_allocator_new):
337 * glib/ghash.c (g_hash_table_unref, g_hash_table_destroy)
338 (g_hash_table_foreach_remove_or_steal): Silence compiler
341 2005-11-27 Matthias Clasen <mclasen@redhat.com>
343 * glib/gunicollate.c (g_utf8_collate_key_for_filename):
344 Don't read beyond len. (#322520, Christian Persch)
346 Wed Nov 23 17:34:01 2005 Tim Janik <timj@imendio.com>
348 * glib/gdataset.c: access datalist flags via atomic pointer access
349 functions, instead of acquiring the dataset lock. this is faster and
350 also matches the atomic pointer readouts in gdatalistprivate.h and
351 direct pointer modifications required by gobject.c.
353 Wed Nov 23 13:35:31 2005 Tim Janik <timj@gtk.org>
355 * glib/gdataset.c: streamlined initialization code somewhat,
356 removed GData node cache.
358 2005-11-22 Matthias Clasen <mclasen@redhat.com>
360 * glib/ghash.c (g_hash_table_ref):
361 (g_hash_table_unref): Mark these as new API.
363 Avoid double locking in g_intern_string (#322133,
366 * glib/gdataset.c (g_quark_from_string_internal): New
367 internal function which factors out common parts of
368 g_quark_from[_static]_string.
369 (g_quark_from_string, g_quark_from_static_string):
370 Use g_quark_from_string_internal.
371 (g_intern_string, g_intern_static_string): Use
372 g_quark_from_string_internal, and only take the
374 (g_quark_new): Don't store the strings shifted by -1
375 in the g_quarks array.
376 (g_quark_to_string): Adapt to the previous change.
378 Tue Nov 22 14:04:26 2005 Tim Janik <timj@imendio.com>
382 g_hash_table_new_full(): create hash tables with a ref count of 1.
383 g_hash_table_ref(): atomically ref_count+=1
384 g_hash_table_unref(): atomically ref_count-=1, destroys hash table
385 when refcount reaches 0.
386 g_hash_table_destroy(): just destroy keys and values, unref by 1.
387 g_hash_table_insert():
388 g_hash_table_replace(): assert ref_count>0.
391 * glib/gatomic.c: added 'volatile' qualifier to all atomic pointer and
394 2005-11-20 Behdad Esfahbod <behdad@gnome.org>
396 * glib/guniprop.c (g_unichar_get_mirror_char): Remove unused
397 variables. (#321984, Andrew Paprocki)
399 2005-11-18 Matthias Clasen <mclasen@redhat.com>
401 * configure.in: Bump version
403 2005-11-17 Matthias Clasen <mclasen@redhat.com>
405 * === Released 2.9.0 ===
407 * glib/glib.symbols: Updates
409 * glib/gutf8.c (g_utf8_offset_to_pointer): Add some performance
410 hints to the docs. (Paolo Borelli)
414 * glib/gthread.c (g_thread_foreach): Mark as new api.
416 * README.in: Updates.
420 * glib/gdate.c (g_date_set_time_t):
421 (g_date_set_time_val): Add functions to set a date from
422 a time_t and from a GTimeVal.
423 (g_date_set_time): Deprecate. (#314067, Roger Leigh)
426 * tests/utf8-pointer.c: Unit tests for g_utf8_pointer_to_offset
427 and g_utf8_offset_to_pointer.
429 * glib/gutf8.c (g_utf8_pointer_to_offset)
430 (g_utf8_offset_to_pointer): Handle negative offsets, and use
431 "stutter stepping" for going backwards. (#320638, Larry
436 * glib/gthread.c: const correctness fixes, found
437 by Arjan van de Ven and gcc.
439 2005-11-16 Behdad Esfahbod <behdad@gnome.org>
441 * docs/reference/glib/tmpl/unicode.sgml: Correct typo on mentioning
442 @G_UNICODE_BREAK_UNKNOWN that should be #G_UNICODE_BREAK_UNKNOWN.
444 2005-11-09 Behdad Esfahbod <behdad@gnome.org>
446 * glib/guniprop.c: Use bit hacks instead when checking a general
447 category value against multiple values.
449 * glib/gutf8.c: Change ISO10646 to Unicode in docs.
451 2005-11-10 Simos Xenitellis <simos@gnome.org>
453 * configure.in: Added tt (Tatar) to ALL_LINGUAS.
455 2005-11-05 Tor Lillqvist <tml@novell.com>
457 * glib/giowin32.c (g_io_win32_set_flags): Don't call g_warning().
458 (#320688) Instead set error and return failure.
460 2005-11-07 Matthias Clasen <mclasen@redhat.com>
462 * glib/gmain.c (g_main_dispatch): Don't call cb_funcs->unref
463 while holding the context lock. (#320886, Andy Wingo)
465 2005-11-07 Billy Biggs <vektor@dumbterm.net>
467 * tests/refcount/Makefile.am: Add a missing reference to libglib
468 in the LDADD for this test.
470 2005-11-07 Matthias Clasen <mclasen@redhat.com>
472 * glib/gmem.c (g_allocator_new): We need to set n_preallocs to a
473 nonzero value, otherwise GTK+ 2.8 breaks when compiled against
474 GLib 2.9. (#320755, Luca Ferretti)
476 2005-11-04 Matthias Clasen <mclasen@redhat.com>
478 * glib/Makefile.am: Apply a patch from Behdad Esfahbod to
479 use a faster lookup table for g_unichar_get_mirror_char().
480 * glib/gmirroringtable.h: The new table.
481 * glib/gunichartables.h: Remove bidi_mirroring_table.
482 * glib/gen-unicode-tables.pl: Don't generate the mirroring
484 * glib/glib-mirroring-tab/*: Sources for the program
485 which generated gmirroringtable.h.
487 * glib/glist.c: Avoid some code duplication.
489 * glib/gscanner.h: Include gdataset.h. (#320322)
491 * glib/gdebug.h: add new GDebugFlag for fatal_criticals
492 * glib/gmessages.c: (_g_debug_init): handle G_DEBUG=fatal_criticals,
493 to help find critical warnings in applications. (#320017,
496 2005-11-02 Tor Lillqvist <tml@novell.com>
498 * glib/glib.symbols: Remove large amount of trailing whitespace
499 from one line. Remove #ifdef G_UNDEFINED from around
500 g_hash_table_new and g_hash_table_new_full symbols, what was the
501 purpose of that? Add the g_slice functions.
503 Tue Nov 1 16:24:20 2005 Tim Janik <timj@imendio.com>
505 * glib/gmem.[hc]: prepared deprecation of GMemChunk and GAllocator.
506 added g_slice_*() API to allocate and cache small bits of memory.
507 an actuall allocator implementation for g_slice_*() is still pending.
509 * glib/gthread.[hc]: changes from a patch by Matthias Clasen.
510 changed GRealThread list to use in-structure *next; fields instead
511 of GSList, in order for thread iteration to not depenend on g_slice_*()
513 _g_thread_mem_private_get():
514 _g_thread_mem_private_set(): added accessors for private memory,
515 needed because the ordinary GPrivate implementation relies on GArray
516 and GSList and therefore indirectly on working g_slice_*() allocations.
519 g_thread_foreach(): new public API function to loop over all existing threads.
533 removed GAllocator and free list usages and accompanying locks.
534 use g_slice_*() API to allocate and cache small bits of memory.
536 * glib/ghook.h: removed GMemChunk field from public API.
539 * glib/glist.h: deprecate allocator API, provide _free1() for consistency.
541 * glib/gnode.h: deprecate allocator API.
543 * glib/gmain.c: reordered GPollRec fields so g_slice_free_chain() can
544 be used for poll rec lists.
546 * glib/grel.c: removed mem chunk usage, and allocated tuples via g_slice_*().
547 g_relation_destroy(): free all tuples from the all_tuples hash table,
548 this effectively maintains the life time track keeping of tuples.
549 g_relation_delete_tuple(): free tuples which are removed from the
550 all_tuples hash table. this fixes a temporary leak that was present
551 in the memchunk code until the destruction of the relation.
553 2005-10-29 Matthias Clasen <mclasen@redhat.com>
555 * tests/convert-test.c: Add some tests for conversions between
556 UTF-8, UCS-4 and UTF-16.
558 * glib/gutf8.c (g_utf8_to_ucs4, g_utf8_to_utf16): Fix handling
559 of len == -1, noticed by Morten Welinder.
561 2005-10-27 Erdal Ronahi <erdal.ronahi@gmail.com>
563 * configure.in: Added ku (Kurdish) to ALL_LINGUAS
565 2005-10-26 Matthias Clasen <mclasen@redhat.com>
567 * glib/gutf8.c (g_ucs4_to_utf8): Don't set items_read twice
568 in the error case, and add some documentation. (#319806, Morten
571 2005-10-19 Manish Singh <yosh@gimp.org>
573 * configure.in: Use AC_CHECK_FUNCS for _NSGetEnviron, to get the
574 config.h symbol automatically. Fixes bug #313731.
576 2005-10-19 Tor Lillqvist <tml@novell.com>
578 * glib/gwin32.c (g_win32_get_package_installation_directory):
579 Return a g_strdup()ed copy of the value stored in the hash table,
580 so that it can be g_free()d without leaving a dangling pointer in
581 the hash table. (#319232)
583 2005-10-06 Matthias Clasen <mclasen@redhat.com>
585 * glib/gunicollate.c (g_utf8_collate_key_for_filename): Handle
586 all-zero sequences correctly. (#317930, Sebastien Bacher)
588 2005-10-05 Matthias Clasen <mclasen@redhat.com>
590 * glib/gmarkup.c (g_markup_escape_text): Doc addition
592 * Makefile.am: Add Collation tests.
594 * tests/collate/*: Inputs and expected outputs for collation tests.
596 * tests/run-collate-tests.sh: Script to run collation tests.
598 * tests/unicode-collate.c (main): Rework slightly to make
599 it usable in unit tests. Also test g_utf8_collate_key_for_filename().
601 2005-10-01 Behdad Esfahbod <behdad@gnome.org>
603 * docs/reference/glib/tmpl/unicode.sgml:
604 * glib/gen-unicode-tables.pl:
606 * glib/gunichartables.h:
608 * tests/casefold.txt:
609 * tests/casemap.txt: Updated to Unicode 4.1. There are five new
610 GUnicodeBreakType types. That may break some applications, like
613 2005-09-26 Matthias Clasen <mclasen@redhat.com>
615 * glib/gstrfuncs.c (g_ascii_strtoull): Add details to the
616 docs. (#314393, Matthew F. Barnes)
619 * glib/gprintf.h: Remove g_snprintf() and g_vsnprintf(), since
620 they are already declared in glib.h. This doesn't break documented
621 use of gprintf.h, but should probably be pointed out in the
622 release notes for 2.10. (#314232, Behdad Esfahbod)
624 Tue Sep 20 13:16:04 2005 Tim Janik <timj@imendio.com>
626 * glib/gpattern.c (g_pattern_ph_match): applied significant recursion
627 complexity optimization, based on a patch from Matthias Clasen.
629 * tests/patterntest.c: more tests, mostly from matthias.
631 2005-09-20 Matthias Clasen <mclasen@redhat.com>
633 * glib/gqueue.c (g_queue_insert_sorted): Correct the docs.
634 (#316703, Mark Drago)
636 Mon Sep 19 17:23:23 2005 Tim Janik <timj@imendio.com>
638 * glib/gpattern.c: applied a patch from matthias which checks on the
639 upper bound of GPatternSpec length to optimize matches.
642 * tests/patterntest.c: added more match cases.
644 2005-09-16 Tor Lillqvist <tml@novell.com>
646 * glib/gstrfuncs.c (g_ascii_strcasecmp, g_ascii_strncasecmp): Add
647 warning to doc comment that these functions should not be used on
648 encodings like CP932.
650 2005-09-14 Matthias Clasen <mclasen@redhat.com>
652 * tests/keyfile-test.c: Add a test for grup names of length 1.
654 * glib/gkeyfile.c (g_key_file_line_is_group): Accept group names
655 of length 1. (#316309)
657 2005-09-12 Matthias Clasen <mclasen@redhat.com>
659 * glib/gmarkup.c (g_markup_escape_text): Clarify docs.
661 2005-09-11 Sebastian Wilhelmi <seppi@seppi.de>
663 * tests/refcount/Makefile.am (INCLUDES): Link the the refcount
664 tests to the system thread library $(G_THREAD_LIBS). Fixes #313744
667 2005-09-11 Kjartan Maraas <kmaraas@gnome.org>
669 * glib/gmain.c: (g_child_watch_prepare), (g_child_watch_check),
670 (child_watch_helper_thread): Remove some dead code. Closes
673 2005-09-07 Tor Lillqvist <tml@novell.com>
675 * glib/Makefile.am: Create also a console version of the
676 gspawn-win32-helper program, gspawn-win32-helper-console.exe.
677 It's otherwise identical to gspawn-win32-helper.exe, except marked
678 as a console application (linked without the -mwindows option).
680 * glib/gspawn-win32.c (do_spawn_directly, do_spawn_with_pipes):
681 Drop the dont_wait parameter. Its truth value correlated 100% with
682 the NULLness of the exit_status parameter anyway, so it's enough
683 to check whether exit_status is NULL. Invert the sense of the
684 dont_return_handle parameter and rename it to do_return_handle, to
685 make the code easier to read by avoiding double negations.
687 (g_spawn_sync_utf8, g_spawn_async_with_pipes_utf8): Modify calls
688 to do_spawn_with_pipes() accordingly.
690 (do_spawn_with_pipes): If we have a console, use the console
691 version of the helper program, otherwise use the GUI one. This
692 avoids extra console windows opening up in some situations. (In
693 case a console application uses the GUI gspawn-win32-helper.exe to
694 spawn another console application we would get a separate console
695 for the spawned console application).
697 * glib-zip.in: Distribute also gspawn-win32-helper-console.exe.
699 2005-09-05 Matthias Clasen <mclasen@redhat.com>
701 * glib/gmappedfile.c (g_mapped_file_new): Report an error
702 if the file is too large. (#315275, Kjartan Maraas)
704 * glib/gkeyfile.c (g_key_file_load_from_fd): The return value
705 of read() is signed. (#315273, Kjartan Maraas)
707 2005-08-31 Tor Lillqvist <tml@novell.com>
709 * glib/gutils.h: Wrapping atexit() is a bad idea on Windows, where
710 the EXE and each DLL have their own atexit function chains.
712 #define g_atexit as atexit instead. This means it has a
713 better chance of doing what the caller wants. For instance,
714 gtkhtml calls g_atexit() registering a function in gtkhtml
715 itself. This caused a crash when g_atexit() was implemented as a
716 function in the GLib DLL. The gtkhtml DLL was already unloaded by
717 the time the GLib DLL got unloaded.
719 * glib/gutils.c: #undef the #define mentioned above, to also get a
720 real g_atexit() into the DLL for backward compatibility. Document
721 the Windows behaviour of g_atexit(), and document the varying ways
722 atexit() can behave in the context of dynamically loaded modules
725 2005-08-31 Matthias Clasen <mclasen@redhat.com>
729 * glib/gdataset.c: Add string interning functions.
731 2005-08-28 Matthias Clasen <mclasen@redhat.com>
733 * glib/giochannel.c: Unify some near-duplicate strings. (#314654,
736 2005-08-26 Matthias Clasen <mclasen@redhat.com>
738 * configure.in: Bump version to 2.9.0
740 2005-08-25 Tor Lillqvist <tml@novell.com>
742 Make also the g_spawn*() functions take parameters in the GLib
743 file name encoding, i.e. UTF-8, on Windows. Has no impact on Unix
744 API or ABI. Like the other GLib API that was earlier changed to
745 use UTF-8 on Windows, the names of the functions that take UTF-8
746 have _utf8 suffixes added by using preprocessor macros in the
747 header file. The old names are kept for functions with the old
748 behaviour, taking parameters in the system codepage, for DLL ABI
751 * glib/gspawn.h: On Win32 add the suffix _utf8 to the names of the
752 g_spawn*() functions.
754 * glib/gspawn-win32.c: Use wide-char API on NT-based
755 Windows. Convert parameters from UTF-8 to wide chars (NT) or
756 system codepage (Win9x) and call the C library _wspawn*() or
757 spawn*() functions respectvely. Add DLL ABI stability versions
758 that take parameters in the system codepage.
760 * glib/gspawn-win32-helper.c: On NT-based Windows use the
761 wide-char versions of argv and envp, and use wide-char API to
762 change directory and spawn the program to run. Remove the verbose
763 debugging output, it was too complex to modify for the wide-char
764 features. (Just add temporary debugging printouts if needed, no
765 need to have them permanently in the source.)
767 * glib/gspawn.c: Corresponding documentation updates.
769 * glib/glib.symbols: Corresponding changes: Mark the ABI stability
770 symbols as PRIVATE, add the new _utf8-suffixed ones.
772 2005-08-24 Stepan Kasal <kasal@ucw.cz>
774 * glib/gtypes.h (G_MININT64): Cast the constant to gint64; it is
775 guint64 otherwise and that can produce warnings about comparison
776 between signed and unsigned.
778 2005-08-23 Matthias Clasen <mclasen@redhat.com>
780 * glib/gutils.c: Fix the crt_externs.h include.
782 2005-08-23 Stepan Kasal <kasal@ucw.cz>
784 * NEWS: Fix spelling of my first name.
786 2005-08-23 Matthias Clasen <mclasen@redhat.com>
790 * === Released 2.8.1 ===
794 2005-08-20 Hans Breuer <hans@breuer.org>
796 * glib/makefile.msc.in : link with ws2_32.lib
798 2005-08-18 Tor Lillqvist <tml@novell.com>
800 * configure.in: Check for <sys/wait.h>
802 * glib/gbacktrace.c: Include <sys/wait.h> on if HAVE_SYS_WAIT_H.
804 2005-08-18 Ross Burton <ross@burtonini.com>
807 Optimise single-character insertions.
812 * tests/string-test.c:
813 Add tests for new optimisation, and fix a leak.
815 2005-08-17 Matthias Clasen <mclasen@redhat.com>
817 * configure.in: Check for crt_externs.h and _NSGetEnviron.
819 * glib/gutils.c: On Darwin, include crt-externs.h and
820 define environ using _NSGetEnviron(). (#313731)
822 2005-08-16 Stepan Kasal <kasal@ucw.cz>
824 * glib/gutils.c (g_get_any_init): Move the body of the big if...
825 (g_get_any_init_do): ... to this new function.
826 (g_get_any_init): Declare as inline.
827 (g_get_any_init_locked): New inline function, does the locking.
828 Make use of these two throughout the code.
830 2005-08-15 Matthias Clasen <mclasen@redhat.com>
832 * glib/gbacktrace.c (g_on_error_stack_trace): Wait for
833 the child process and then simply return. This makes
834 The "S" option work as documented in g_on_error_query().
835 (#313125, Matthew F. Barnes)
837 * glib/gunicode.h: Update the link to Unicode category
838 values. (#313369, Behnam Esfahbod)
840 * glib/gqueue.c (g_queue_find_custom): Clarify docs
841 a little. (#311727, Tristan van Berkom)
843 * glib/abicheck.sh, gobject/abicheck.sh: Make the
844 check work on ia64 too, where some symbols ended up
845 in yet another section.