1 2005-12-27 Matthias Clasen <mclasen@redhat.com>
3 Partial fix for bug #32937, Bogdan Nicula.
5 * configure.in: Check for malloc.h
6 * glib/gslice.c: Don't include malloc.h unconditionally.
8 2005-12-27 Manish Singh <yosh@gimp.org>
10 * tests/run-collate-tests.sh: export LC_COLLATE so it takes.
11 (#324950, Dan Yefimov)
13 2005-12-26 Matthias Clasen <mclasen@redhat.com>
15 * glib/gslice.c: On Win32, include process.h (#325015, Kazuki
18 2005-12-25 Matthias Clasen <mclasen@redhat.com>
20 * AUTHORS: Update my email
22 * tests/slice-test.c: Fix C99isms. (#324950, Dan Yefimov)
24 2005-12-21 Matthias Clasen <mclasen@redhat.com>
26 * README.in: Add some notes on when 'make check' may fail.
28 2005-12-20 Matthias Clasen <mclasen@redhat.com>
32 * glib/gthreadpool.c (g_thread_pool_set_sort_function): New function
33 to sort tasks pushed into a threadpool. (#324479, Martyn Russell)
35 * tests/threadpool-test.c: Test this.
37 Tue Dec 20 18:14:14 2005 Tim Janik <timj@imendio.com>
39 * glib/gslice.[hc]: added mem_error() and mem_assert() to test and
40 handle errors without depending on gmessage.c which might not be
41 setup when the error occours.
42 removed G_SLICE_CONFIG_ALWAYS_FREE config option, fixed the code so
43 always freeing can be achieved by adjusting the working set time to
44 0 with G_SLICE_CONFIG_WORKING_SET_MSECS.
45 added G_SLICE_CONFIG_COLOR_INCREMENT to test different color increments
46 (mainly 0 and 1). reduced the minimum block size to 128 bytes, to
47 minimize wastage if small amounts of differently sized structrues are
48 allocated, this does come at a performance cost of roughly 5% though.
49 fixed up block alignment calculation, so it works for varying
50 block sizes. only use strerror() not g_strerror() since the latter
51 depends on working GQuark and GSlice.
52 mem_error(): implemented in terms of fprintf and vfprintf.
54 * tests/slice-color.c: new program to test cache colorization effects.
56 * tests/slice-test.c: trade G_SLICE_CONFIG_ALWAYS_FREE for 0 duration
57 G_SLICE_CONFIG_WORKING_SET_MSECS.
59 2005-12-17 Matthias Clasen <mclasen@redhat.com>
61 * glib/goption.c (parse_short_option): Set an error in all
62 failure cases. (#324332, Tim-Philipp Müller)
64 2005-12-17 Sebastian Wilhelmi <seppi@seppi.de>
66 * glib/gatomic.c: Fix memory barrier position in g_atomic_int_get
67 and g_atomic_pointer_get. Add g_atomic_int_set and
68 g_atomic_pointer_set implementations for the !DEFINE_WITH_MUTEXES &&
69 G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case, as well as defining them
70 as functions (additionally to the macros in the header) for the
71 !G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case.
73 2005-12-16 Matthias Clasen <mclasen@redhat.com>
75 * glib/gmem.c (g_allocator_new): Don't return a pointer to
76 a const struct, since apps expect to be able to modify it.
77 (#324179, J. Ali Harlow)
79 Tue Dec 13 10:13:32 2005 Tim Janik <timj@imendio.com>
81 * glib/gatomic.h: added g_atomic_pointer_set() and g_atomic_int_set()
82 for systems where the initialization of atomic variables requires a
85 2005-12-09 Matthias Clasen <mclasen@redhat.com>
87 * configure.in: Bump version
89 * === Released 2.9.1 ===
91 2005-12-09 Alexander Larsson <alexl@redhat.com>
95 Use G_GNUC_WARN_UNUSED_RESULT on list functions that return
98 2005-12-08 Matthias Clasen <mclasen@redhat.com>
102 2005-12-07 Martyn Russell <martyn@imendio.com>
104 * glib/gasyncqueue.c:
105 - Call g_queue_insert_sorted() instead of duplicating the code.
106 - Call g_queue_sort() instead of duplicating the code.
107 - Invert sort function results to make sure the same sort function
108 gives the same results across glist, gslist, gqueue and
111 * tests/asyncqueue-test.c:
112 - Updated the sort function to reflect the example in the
113 documentation for gasyncqueue.c.
115 2005-12-07 Martyn Russell <martyn@imendio.com>
117 * docs/reference/glib/glib-sections.txt:
118 * docs/reference/glib/tmpl/linked_lists_double.sgml:
119 * docs/reference/glib/tmpl/linked_lists_single.sgml:
122 - Added g_list_insert_sorted_with_data () and
123 g_slist_insert_sorted_with_data ().
124 - Removed the extra check in g_list_sort() and g_slist_sort() for
125 GCompareDataFunc vs. GCompareFunc.
127 2005-12-07 Tor Lillqvist <tml@novell.com>
129 * glib-gettextize.in: Look up prefix at run-time on Win32,
130 assuming the standard directory structure with glib-gettextize in
133 2005-12-06 Matthias Clasen <mclasen@redhat.com>
135 * glib/gmem.h: Exempt GTK+ from the mem chunk deprecation, since
136 we need GTK+ 2.8 to compile against GLib 2.10.
138 2005-12-05 Manish Singh <yosh@gimp.org>
140 * tests/run-collate-test.sh: set LC_COLLATE, not LANG, to be sure
141 to override any user settings.
143 2005-12-05 Matthias Clasen <mclasen@redhat.com>
145 * configure.in: Revert an accidential commit
147 2005-12-05 Matthias Clasen <mclasen@redhat.com>
149 * glib/gutf8.c: Documentation updates. (#323291, Morten Welinder)
151 * tests/Makefile.am (TESTS_ENVIRONMENT):
152 * tests/gobject/Makefile.am (TESTS_ENVIRONMENT): Set
153 MALLOC_CHECK_ and MALLOC_PERTURB_
155 * tests/run-collate-tests.sh: Run the collation tests explicitly
156 in en_US locale. (#320463)
158 * glib/gmem.h: Really deprecate GMemChunk.
165 * glib/gstrfuncs.c: Add versioned deprecation docs.
167 2005-12-05 Martyn Russell <martyn@imendio.com>
169 * docs/reference/glib/glib-sections.txt:
170 * glib/gasyncqueue.[ch]:
171 - Added support for sorting async queues by with _push_sorted(),
172 _push_sorted_unlocked(), _sort() and _sort_unlocked() (#323047).
175 * tests/asyncqueue-test.c:
176 - Added test case for gasyncqueue.c
178 Mon Dec 5 15:53:20 2005 Tim Janik <timj@imendio.com>
180 * glib/gslice.c: implement chain walking for arbitrary ->next pointer
181 offsets in g_slice_free_chain_with_offset() based on a patch by behdad
182 in bug 323178. moved time consuming logic from g_slice_free() out of
183 the inner loop, so g_slice_free_chain_with_offset() provides a real
184 performance benefit over g_slice_free1() now.
186 * glib/gslice.h: renamed g_slice_free_chain() to
187 g_slice_free_chain_with_offset(). implemented g_slice_free_chain() as
188 a type-safe macro as suggested in bug 323178.
189 simplified the macro implementation of g_slice_free() and implemented
190 it in a type safe manner for all compliers as suggested by Morten
191 Welinder <mortenw@gnome.org>.
196 * glib/glib.symbols: s/g_slice_free_chain/g_slice_free_chain_with_offset/
198 2005-12-05 Matthias Clasen <mclasen@redhat.com>
200 * glib/gasyncqueue.c: Add some docs.
202 * tests/libmoduletestplugin_a.c: Fix compiler warnings.
204 * glib/gatomic.c: In the ia64 implementation, use
205 __sync builtin without _si or _di suffix. (#321229,
206 Stanislav Brabec, patch by Andreas Schwab)
208 2005-12-04 Behdad Esfahbod <behdad@gnome.org>
210 * glib/gslice.h: Remove comma at the end of enum.
212 2005-12-04 Matthias Clasen <mclasen@redhat.com>
214 Handle multiple user names with the same UID better.
215 (#319535, Laszlo Peter)
217 * glib/gutils.c (g_get_any_init_do): When determining user
218 data, first look up $LOGNAME. If the UID doesn't match
219 getuid(), fall back to the current behaviour of looking
220 up the user data based on getuid().
222 2005-12-04 Matthias Clasen <mclasen@redhat.com>
224 * glib/gmacros.h (G_GNUC_WARN_UNUSED_RESULT): Add a macro
225 to make gcc warn if a function result is ignored. (#145466,
226 Arjan van de Ven, Alex Larsson)
228 * glib/gmem.h: Add the new attribute to g_realloc and
231 2005-12-03 Matthias Clasen <mclasen@redhat.com>
234 * glib/gcache.h: Deprecate g_cache_value_foreach. (#322956,
237 * glib/gmappedfile.c: Make mapping of empty files
240 * glib/gfileutils.c: Don't fork a new process just to
241 fix the permissions of the created temp file. (#321318,
242 Alexis S. L. Carvalho)
244 2005-12-02 Matthias Clasen <mclasen@redhat.com>
246 * README.in: Add a note about Solaris threads.
249 * configure.in: Remove support for Solaris threads.
250 (#136971, Sebastian Wilhelmi, patch by Andrew Paprocki)
252 2005-12-02 Matthias Clasen <mclasen@redhat.com>
254 * configure.in: Define G_GUINT64_CONSTANT in analogy to
255 G_GINT64_CONSTANT. (#322568, Andrew Paprocki)
257 2005-12-02 Matthias Clasen <mclasen@redhat.com>
259 * glib/gslice.c: Win32 portability fixes and C99ism removal,
260 pointed out by Kazuki Iwamoto. (#323052)
262 2005-12-02 Matthias Clasen <mclasen@redhat.com>
264 * glib/gatomic.c (g_atomic_int_add): Add a missing volatile for
265 the IA64 implementation.
267 Fri Dec 2 16:18:09 2005 Tim Janik <timj@imendio.com>
269 * glib/gslice.c: some naming and type size fixups.
271 Fri Dec 2 13:08:58 2005 Tim Janik <timj@gtk.org>
273 * glib/gslice.h (g_slice_free): reworked GCC-specific type-safe macro
274 variant into something less verbose, but digestible for gcc-3.4.
276 Fri Dec 2 10:55:07 2005 Tim Janik <timj@imendio.com>
278 * tests/slice-test.c: extended to perform the benchmarking on the old
279 memchunk code if 'O' is selected.
281 * tests/memchunks.c: new file which contains the old GLib mem chunks
282 implementation with prefix old_mem_chunk_.
284 * tests/Makefile.am: added memchunks.c
286 Fri Dec 2 00:16:59 2005 Tim Janik <timj@imendio.com>
288 * glib/gslice.c: improved gettimeofday() timer resolution by maintaining
289 acceptable load distribution of the syscall.
290 (allocator_get_magazine_threshold): reduce minimum/single-thread magazine
291 sizes to approximately page_size/2.5 to avoid excessive startup allocations.
293 2005-12-01 Matthias Clasen <mclasen@redhat.com>
295 * tests/gobject/ifacecheck.c (test_iface_base_init): Don't
296 double-free base interface members.
298 * tests/Makefile.am: Remove duplicate variable.
300 Thu Dec 1 17:32:46 2005 Tim Janik <timj@imendio.com>
302 * glib/gslice.[hc]: new slice allocator implementation.
304 * tests/slice-test.c: added random slice allocation test.
306 * glib/gthread.[hc]: removed newly added private thread mem API.
308 * glib/gthreadinit.h:
311 * glib/gmem.c: divided glib threading initialisation into three phases,
312 initialisation where private keys and messaging are not available (only
313 needed by gmem.c), initialisation without messaging but private keys
314 available (gslice.c, gmessage.c), and full fledged initialisers that
315 server the rest of glib. initialisation functions got renamed to reflect
316 the limitations of their corresponding phases.
318 * glib/gmem.c: removed memchunk code, defer allocations to
321 * glib/gmem.[hc]: removed g_slice_* skeletons.
323 * glib/glib.symbols: added g_slice_* symbols.
325 * configure.in: check for availability of posix_memalign(3), memalign(3)
328 * glib/Makefile.am: added gslice.[hc].
330 2005-12-01 Tor Lillqvist <tml@novell.com>
332 * glib/gstdio.c (g_stat): In the Win32 implementation, strip
333 trailing slash(es) for non-root folders. stat() fails if non-root
334 folders are specified with trailing slashes. It's too much hassle
335 to demand that callers strip such slashes themselves, especially
336 as it is easy to get it wrong and strip the slash of a root
338 (g_rename): On NT-based Windows, use MoveFileEx() with
339 MOVEFILE_REPLACE_EXISTING to better match Unix behaviour.
341 2005-11-28 Matthias Clasen <mclasen@redhat.com>
343 Fix G_STMT_START / G_STMT_END on Solaris. (#321972,
346 * configure.in: Check whether do { } while (0) works.
348 * glib/gmacros.h: Use do { } while (0) for G_STMT_START /
349 G_STMT_END if it works.
351 2005-11-28 Matthias Clasen <mclasen@redhat.com>
353 * glib/gthread.c (g_static_rw_lock_wait, g_static_rw_lock_signal):
354 * glib/gnode.c (g_node_depth_traverse_level):
355 * glib/gmem.c (g_allocator_new):
356 * glib/ghash.c (g_hash_table_unref, g_hash_table_destroy)
357 (g_hash_table_foreach_remove_or_steal): Silence compiler
360 2005-11-27 Matthias Clasen <mclasen@redhat.com>
362 * glib/gunicollate.c (g_utf8_collate_key_for_filename):
363 Don't read beyond len. (#322520, Christian Persch)
365 Wed Nov 23 17:34:01 2005 Tim Janik <timj@imendio.com>
367 * glib/gdataset.c: access datalist flags via atomic pointer access
368 functions, instead of acquiring the dataset lock. this is faster and
369 also matches the atomic pointer readouts in gdatalistprivate.h and
370 direct pointer modifications required by gobject.c.
372 Wed Nov 23 13:35:31 2005 Tim Janik <timj@gtk.org>
374 * glib/gdataset.c: streamlined initialization code somewhat,
375 removed GData node cache.
377 2005-11-22 Matthias Clasen <mclasen@redhat.com>
379 * glib/ghash.c (g_hash_table_ref):
380 (g_hash_table_unref): Mark these as new API.
382 Avoid double locking in g_intern_string (#322133,
385 * glib/gdataset.c (g_quark_from_string_internal): New
386 internal function which factors out common parts of
387 g_quark_from[_static]_string.
388 (g_quark_from_string, g_quark_from_static_string):
389 Use g_quark_from_string_internal.
390 (g_intern_string, g_intern_static_string): Use
391 g_quark_from_string_internal, and only take the
393 (g_quark_new): Don't store the strings shifted by -1
394 in the g_quarks array.
395 (g_quark_to_string): Adapt to the previous change.
397 Tue Nov 22 14:04:26 2005 Tim Janik <timj@imendio.com>
401 g_hash_table_new_full(): create hash tables with a ref count of 1.
402 g_hash_table_ref(): atomically ref_count+=1
403 g_hash_table_unref(): atomically ref_count-=1, destroys hash table
404 when refcount reaches 0.
405 g_hash_table_destroy(): just destroy keys and values, unref by 1.
406 g_hash_table_insert():
407 g_hash_table_replace(): assert ref_count>0.
410 * glib/gatomic.c: added 'volatile' qualifier to all atomic pointer and
413 2005-11-20 Behdad Esfahbod <behdad@gnome.org>
415 * glib/guniprop.c (g_unichar_get_mirror_char): Remove unused
416 variables. (#321984, Andrew Paprocki)
418 2005-11-18 Matthias Clasen <mclasen@redhat.com>
420 * configure.in: Bump version
422 2005-11-17 Matthias Clasen <mclasen@redhat.com>
424 * === Released 2.9.0 ===
426 * glib/glib.symbols: Updates
428 * glib/gutf8.c (g_utf8_offset_to_pointer): Add some performance
429 hints to the docs. (Paolo Borelli)
433 * glib/gthread.c (g_thread_foreach): Mark as new api.
435 * README.in: Updates.
439 * glib/gdate.c (g_date_set_time_t):
440 (g_date_set_time_val): Add functions to set a date from
441 a time_t and from a GTimeVal.
442 (g_date_set_time): Deprecate. (#314067, Roger Leigh)
445 * tests/utf8-pointer.c: Unit tests for g_utf8_pointer_to_offset
446 and g_utf8_offset_to_pointer.
448 * glib/gutf8.c (g_utf8_pointer_to_offset)
449 (g_utf8_offset_to_pointer): Handle negative offsets, and use
450 "stutter stepping" for going backwards. (#320638, Larry
455 * glib/gthread.c: const correctness fixes, found
456 by Arjan van de Ven and gcc.
458 2005-11-16 Behdad Esfahbod <behdad@gnome.org>
460 * docs/reference/glib/tmpl/unicode.sgml: Correct typo on mentioning
461 @G_UNICODE_BREAK_UNKNOWN that should be #G_UNICODE_BREAK_UNKNOWN.
463 2005-11-09 Behdad Esfahbod <behdad@gnome.org>
465 * glib/guniprop.c: Use bit hacks instead when checking a general
466 category value against multiple values.
468 * glib/gutf8.c: Change ISO10646 to Unicode in docs.
470 2005-11-10 Simos Xenitellis <simos@gnome.org>
472 * configure.in: Added tt (Tatar) to ALL_LINGUAS.
474 2005-11-05 Tor Lillqvist <tml@novell.com>
476 * glib/giowin32.c (g_io_win32_set_flags): Don't call g_warning().
477 (#320688) Instead set error and return failure.
479 2005-11-07 Matthias Clasen <mclasen@redhat.com>
481 * glib/gmain.c (g_main_dispatch): Don't call cb_funcs->unref
482 while holding the context lock. (#320886, Andy Wingo)
484 2005-11-07 Billy Biggs <vektor@dumbterm.net>
486 * tests/refcount/Makefile.am: Add a missing reference to libglib
487 in the LDADD for this test.
489 2005-11-07 Matthias Clasen <mclasen@redhat.com>
491 * glib/gmem.c (g_allocator_new): We need to set n_preallocs to a
492 nonzero value, otherwise GTK+ 2.8 breaks when compiled against
493 GLib 2.9. (#320755, Luca Ferretti)
495 2005-11-04 Matthias Clasen <mclasen@redhat.com>
497 * glib/Makefile.am: Apply a patch from Behdad Esfahbod to
498 use a faster lookup table for g_unichar_get_mirror_char().
499 * glib/gmirroringtable.h: The new table.
500 * glib/gunichartables.h: Remove bidi_mirroring_table.
501 * glib/gen-unicode-tables.pl: Don't generate the mirroring
503 * glib/glib-mirroring-tab/*: Sources for the program
504 which generated gmirroringtable.h.
506 * glib/glist.c: Avoid some code duplication.
508 * glib/gscanner.h: Include gdataset.h. (#320322)
510 * glib/gdebug.h: add new GDebugFlag for fatal_criticals
511 * glib/gmessages.c: (_g_debug_init): handle G_DEBUG=fatal_criticals,
512 to help find critical warnings in applications. (#320017,
515 2005-11-02 Tor Lillqvist <tml@novell.com>
517 * glib/glib.symbols: Remove large amount of trailing whitespace
518 from one line. Remove #ifdef G_UNDEFINED from around
519 g_hash_table_new and g_hash_table_new_full symbols, what was the
520 purpose of that? Add the g_slice functions.
522 Tue Nov 1 16:24:20 2005 Tim Janik <timj@imendio.com>
524 * glib/gmem.[hc]: prepared deprecation of GMemChunk and GAllocator.
525 added g_slice_*() API to allocate and cache small bits of memory.
526 an actuall allocator implementation for g_slice_*() is still pending.
528 * glib/gthread.[hc]: changes from a patch by Matthias Clasen.
529 changed GRealThread list to use in-structure *next; fields instead
530 of GSList, in order for thread iteration to not depenend on g_slice_*()
532 _g_thread_mem_private_get():
533 _g_thread_mem_private_set(): added accessors for private memory,
534 needed because the ordinary GPrivate implementation relies on GArray
535 and GSList and therefore indirectly on working g_slice_*() allocations.
538 g_thread_foreach(): new public API function to loop over all existing threads.
552 removed GAllocator and free list usages and accompanying locks.
553 use g_slice_*() API to allocate and cache small bits of memory.
555 * glib/ghook.h: removed GMemChunk field from public API.
558 * glib/glist.h: deprecate allocator API, provide _free1() for consistency.
560 * glib/gnode.h: deprecate allocator API.
562 * glib/gmain.c: reordered GPollRec fields so g_slice_free_chain() can
563 be used for poll rec lists.
565 * glib/grel.c: removed mem chunk usage, and allocated tuples via g_slice_*().
566 g_relation_destroy(): free all tuples from the all_tuples hash table,
567 this effectively maintains the life time track keeping of tuples.
568 g_relation_delete_tuple(): free tuples which are removed from the
569 all_tuples hash table. this fixes a temporary leak that was present
570 in the memchunk code until the destruction of the relation.
572 2005-10-29 Matthias Clasen <mclasen@redhat.com>
574 * tests/convert-test.c: Add some tests for conversions between
575 UTF-8, UCS-4 and UTF-16.
577 * glib/gutf8.c (g_utf8_to_ucs4, g_utf8_to_utf16): Fix handling
578 of len == -1, noticed by Morten Welinder.
580 2005-10-27 Erdal Ronahi <erdal.ronahi@gmail.com>
582 * configure.in: Added ku (Kurdish) to ALL_LINGUAS
584 2005-10-26 Matthias Clasen <mclasen@redhat.com>
586 * glib/gutf8.c (g_ucs4_to_utf8): Don't set items_read twice
587 in the error case, and add some documentation. (#319806, Morten
590 2005-10-19 Manish Singh <yosh@gimp.org>
592 * configure.in: Use AC_CHECK_FUNCS for _NSGetEnviron, to get the
593 config.h symbol automatically. Fixes bug #313731.
595 2005-10-19 Tor Lillqvist <tml@novell.com>
597 * glib/gwin32.c (g_win32_get_package_installation_directory):
598 Return a g_strdup()ed copy of the value stored in the hash table,
599 so that it can be g_free()d without leaving a dangling pointer in
600 the hash table. (#319232)
602 2005-10-06 Matthias Clasen <mclasen@redhat.com>
604 * glib/gunicollate.c (g_utf8_collate_key_for_filename): Handle
605 all-zero sequences correctly. (#317930, Sebastien Bacher)
607 2005-10-05 Matthias Clasen <mclasen@redhat.com>
609 * glib/gmarkup.c (g_markup_escape_text): Doc addition
611 * Makefile.am: Add Collation tests.
613 * tests/collate/*: Inputs and expected outputs for collation tests.
615 * tests/run-collate-tests.sh: Script to run collation tests.
617 * tests/unicode-collate.c (main): Rework slightly to make
618 it usable in unit tests. Also test g_utf8_collate_key_for_filename().
620 2005-10-01 Behdad Esfahbod <behdad@gnome.org>
622 * docs/reference/glib/tmpl/unicode.sgml:
623 * glib/gen-unicode-tables.pl:
625 * glib/gunichartables.h:
627 * tests/casefold.txt:
628 * tests/casemap.txt: Updated to Unicode 4.1. There are five new
629 GUnicodeBreakType types. That may break some applications, like
632 2005-09-26 Matthias Clasen <mclasen@redhat.com>
634 * glib/gstrfuncs.c (g_ascii_strtoull): Add details to the
635 docs. (#314393, Matthew F. Barnes)
638 * glib/gprintf.h: Remove g_snprintf() and g_vsnprintf(), since
639 they are already declared in glib.h. This doesn't break documented
640 use of gprintf.h, but should probably be pointed out in the
641 release notes for 2.10. (#314232, Behdad Esfahbod)
643 Tue Sep 20 13:16:04 2005 Tim Janik <timj@imendio.com>
645 * glib/gpattern.c (g_pattern_ph_match): applied significant recursion
646 complexity optimization, based on a patch from Matthias Clasen.
648 * tests/patterntest.c: more tests, mostly from matthias.
650 2005-09-20 Matthias Clasen <mclasen@redhat.com>
652 * glib/gqueue.c (g_queue_insert_sorted): Correct the docs.
653 (#316703, Mark Drago)
655 Mon Sep 19 17:23:23 2005 Tim Janik <timj@imendio.com>
657 * glib/gpattern.c: applied a patch from matthias which checks on the
658 upper bound of GPatternSpec length to optimize matches.
661 * tests/patterntest.c: added more match cases.
663 2005-09-16 Tor Lillqvist <tml@novell.com>
665 * glib/gstrfuncs.c (g_ascii_strcasecmp, g_ascii_strncasecmp): Add
666 warning to doc comment that these functions should not be used on
667 encodings like CP932.
669 2005-09-14 Matthias Clasen <mclasen@redhat.com>
671 * tests/keyfile-test.c: Add a test for grup names of length 1.
673 * glib/gkeyfile.c (g_key_file_line_is_group): Accept group names
674 of length 1. (#316309)
676 2005-09-12 Matthias Clasen <mclasen@redhat.com>
678 * glib/gmarkup.c (g_markup_escape_text): Clarify docs.
680 2005-09-11 Sebastian Wilhelmi <seppi@seppi.de>
682 * tests/refcount/Makefile.am (INCLUDES): Link the the refcount
683 tests to the system thread library $(G_THREAD_LIBS). Fixes #313744
686 2005-09-11 Kjartan Maraas <kmaraas@gnome.org>
688 * glib/gmain.c: (g_child_watch_prepare), (g_child_watch_check),
689 (child_watch_helper_thread): Remove some dead code. Closes
692 2005-09-07 Tor Lillqvist <tml@novell.com>
694 * glib/Makefile.am: Create also a console version of the
695 gspawn-win32-helper program, gspawn-win32-helper-console.exe.
696 It's otherwise identical to gspawn-win32-helper.exe, except marked
697 as a console application (linked without the -mwindows option).
699 * glib/gspawn-win32.c (do_spawn_directly, do_spawn_with_pipes):
700 Drop the dont_wait parameter. Its truth value correlated 100% with
701 the NULLness of the exit_status parameter anyway, so it's enough
702 to check whether exit_status is NULL. Invert the sense of the
703 dont_return_handle parameter and rename it to do_return_handle, to
704 make the code easier to read by avoiding double negations.
706 (g_spawn_sync_utf8, g_spawn_async_with_pipes_utf8): Modify calls
707 to do_spawn_with_pipes() accordingly.
709 (do_spawn_with_pipes): If we have a console, use the console
710 version of the helper program, otherwise use the GUI one. This
711 avoids extra console windows opening up in some situations. (In
712 case a console application uses the GUI gspawn-win32-helper.exe to
713 spawn another console application we would get a separate console
714 for the spawned console application).
716 * glib-zip.in: Distribute also gspawn-win32-helper-console.exe.
718 2005-09-05 Matthias Clasen <mclasen@redhat.com>
720 * glib/gmappedfile.c (g_mapped_file_new): Report an error
721 if the file is too large. (#315275, Kjartan Maraas)
723 * glib/gkeyfile.c (g_key_file_load_from_fd): The return value
724 of read() is signed. (#315273, Kjartan Maraas)
726 2005-08-31 Tor Lillqvist <tml@novell.com>
728 * glib/gutils.h: Wrapping atexit() is a bad idea on Windows, where
729 the EXE and each DLL have their own atexit function chains.
731 #define g_atexit as atexit instead. This means it has a
732 better chance of doing what the caller wants. For instance,
733 gtkhtml calls g_atexit() registering a function in gtkhtml
734 itself. This caused a crash when g_atexit() was implemented as a
735 function in the GLib DLL. The gtkhtml DLL was already unloaded by
736 the time the GLib DLL got unloaded.
738 * glib/gutils.c: #undef the #define mentioned above, to also get a
739 real g_atexit() into the DLL for backward compatibility. Document
740 the Windows behaviour of g_atexit(), and document the varying ways
741 atexit() can behave in the context of dynamically loaded modules
744 2005-08-31 Matthias Clasen <mclasen@redhat.com>
748 * glib/gdataset.c: Add string interning functions.
750 2005-08-28 Matthias Clasen <mclasen@redhat.com>
752 * glib/giochannel.c: Unify some near-duplicate strings. (#314654,
755 2005-08-26 Matthias Clasen <mclasen@redhat.com>
757 * configure.in: Bump version to 2.9.0
759 2005-08-25 Tor Lillqvist <tml@novell.com>
761 Make also the g_spawn*() functions take parameters in the GLib
762 file name encoding, i.e. UTF-8, on Windows. Has no impact on Unix
763 API or ABI. Like the other GLib API that was earlier changed to
764 use UTF-8 on Windows, the names of the functions that take UTF-8
765 have _utf8 suffixes added by using preprocessor macros in the
766 header file. The old names are kept for functions with the old
767 behaviour, taking parameters in the system codepage, for DLL ABI
770 * glib/gspawn.h: On Win32 add the suffix _utf8 to the names of the
771 g_spawn*() functions.
773 * glib/gspawn-win32.c: Use wide-char API on NT-based
774 Windows. Convert parameters from UTF-8 to wide chars (NT) or
775 system codepage (Win9x) and call the C library _wspawn*() or
776 spawn*() functions respectvely. Add DLL ABI stability versions
777 that take parameters in the system codepage.
779 * glib/gspawn-win32-helper.c: On NT-based Windows use the
780 wide-char versions of argv and envp, and use wide-char API to
781 change directory and spawn the program to run. Remove the verbose
782 debugging output, it was too complex to modify for the wide-char
783 features. (Just add temporary debugging printouts if needed, no
784 need to have them permanently in the source.)
786 * glib/gspawn.c: Corresponding documentation updates.
788 * glib/glib.symbols: Corresponding changes: Mark the ABI stability
789 symbols as PRIVATE, add the new _utf8-suffixed ones.
791 2005-08-24 Stepan Kasal <kasal@ucw.cz>
793 * glib/gtypes.h (G_MININT64): Cast the constant to gint64; it is
794 guint64 otherwise and that can produce warnings about comparison
795 between signed and unsigned.
797 2005-08-23 Matthias Clasen <mclasen@redhat.com>
799 * glib/gutils.c: Fix the crt_externs.h include.
801 2005-08-23 Stepan Kasal <kasal@ucw.cz>
803 * NEWS: Fix spelling of my first name.
805 2005-08-23 Matthias Clasen <mclasen@redhat.com>
809 * === Released 2.8.1 ===
813 2005-08-20 Hans Breuer <hans@breuer.org>
815 * glib/makefile.msc.in : link with ws2_32.lib
817 2005-08-18 Tor Lillqvist <tml@novell.com>
819 * configure.in: Check for <sys/wait.h>
821 * glib/gbacktrace.c: Include <sys/wait.h> on if HAVE_SYS_WAIT_H.
823 2005-08-18 Ross Burton <ross@burtonini.com>
826 Optimise single-character insertions.
831 * tests/string-test.c:
832 Add tests for new optimisation, and fix a leak.
834 2005-08-17 Matthias Clasen <mclasen@redhat.com>
836 * configure.in: Check for crt_externs.h and _NSGetEnviron.
838 * glib/gutils.c: On Darwin, include crt-externs.h and
839 define environ using _NSGetEnviron(). (#313731)
841 2005-08-16 Stepan Kasal <kasal@ucw.cz>
843 * glib/gutils.c (g_get_any_init): Move the body of the big if...
844 (g_get_any_init_do): ... to this new function.
845 (g_get_any_init): Declare as inline.
846 (g_get_any_init_locked): New inline function, does the locking.
847 Make use of these two throughout the code.
849 2005-08-15 Matthias Clasen <mclasen@redhat.com>
851 * glib/gbacktrace.c (g_on_error_stack_trace): Wait for
852 the child process and then simply return. This makes
853 The "S" option work as documented in g_on_error_query().
854 (#313125, Matthew F. Barnes)
856 * glib/gunicode.h: Update the link to Unicode category
857 values. (#313369, Behnam Esfahbod)
859 * glib/gqueue.c (g_queue_find_custom): Clarify docs
860 a little. (#311727, Tristan van Berkom)
862 * glib/abicheck.sh, gobject/abicheck.sh: Make the
863 check work on ia64 too, where some symbols ended up
864 in yet another section.