1 2005-12-20 Matthias Clasen <mclasen@redhat.com>
5 * glib/gthreadpool.c (g_thread_pool_set_sort_function): New function
6 to sort tasks pushed into a threadpool. (#324479, Martyn Russell)
8 * tests/threadpool-test.c: Test this.
10 Tue Dec 20 18:14:14 2005 Tim Janik <timj@imendio.com>
12 * glib/gslice.[hc]: added mem_error() and mem_assert() to test and
13 handle errors without depending on gmessage.c which might not be
14 setup when the error occours.
15 removed G_SLICE_CONFIG_ALWAYS_FREE config option, fixed the code so
16 always freeing can be achieved by adjusting the working set time to
17 0 with G_SLICE_CONFIG_WORKING_SET_MSECS.
18 added G_SLICE_CONFIG_COLOR_INCREMENT to test different color increments
19 (mainly 0 and 1). reduced the minimum block size to 128 bytes, to
20 minimize wastage if small amounts of differently sized structrues are
21 allocated, this does come at a performance cost of roughly 5% though.
22 fixed up block alignment calculation, so it works for varying
23 block sizes. only use strerror() not g_strerror() since the latter
24 depends on working GQuark and GSlice.
25 mem_error(): implemented in terms of fprintf and vfprintf.
27 * tests/slice-color.c: new program to test cache colorization effects.
29 * tests/slice-test.c: trade G_SLICE_CONFIG_ALWAYS_FREE for 0 duration
30 G_SLICE_CONFIG_WORKING_SET_MSECS.
32 2005-12-17 Matthias Clasen <mclasen@redhat.com>
34 * glib/goption.c (parse_short_option): Set an error in all
35 failure cases. (#324332, Tim-Philipp Müller)
37 2005-12-17 Sebastian Wilhelmi <seppi@seppi.de>
39 * glib/gatomic.c: Fix memory barrier position in g_atomic_int_get
40 and g_atomic_pointer_get. Add g_atomic_int_set and
41 g_atomic_pointer_set implementations for the !DEFINE_WITH_MUTEXES &&
42 G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case, as well as defining them
43 as functions (additionally to the macros in the header) for the
44 !G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case.
46 2005-12-16 Matthias Clasen <mclasen@redhat.com>
48 * glib/gmem.c (g_allocator_new): Don't return a pointer to
49 a const struct, since apps expect to be able to modify it.
50 (#324179, J. Ali Harlow)
52 Tue Dec 13 10:13:32 2005 Tim Janik <timj@imendio.com>
54 * glib/gatomic.h: added g_atomic_pointer_set() and g_atomic_int_set()
55 for systems where the initialization of atomic variables requires a
58 2005-12-09 Matthias Clasen <mclasen@redhat.com>
60 * configure.in: Bump version
62 * === Released 2.9.1 ===
64 2005-12-09 Alexander Larsson <alexl@redhat.com>
68 Use G_GNUC_WARN_UNUSED_RESULT on list functions that return
71 2005-12-08 Matthias Clasen <mclasen@redhat.com>
75 2005-12-07 Martyn Russell <martyn@imendio.com>
78 - Call g_queue_insert_sorted() instead of duplicating the code.
79 - Call g_queue_sort() instead of duplicating the code.
80 - Invert sort function results to make sure the same sort function
81 gives the same results across glist, gslist, gqueue and
84 * tests/asyncqueue-test.c:
85 - Updated the sort function to reflect the example in the
86 documentation for gasyncqueue.c.
88 2005-12-07 Martyn Russell <martyn@imendio.com>
90 * docs/reference/glib/glib-sections.txt:
91 * docs/reference/glib/tmpl/linked_lists_double.sgml:
92 * docs/reference/glib/tmpl/linked_lists_single.sgml:
95 - Added g_list_insert_sorted_with_data () and
96 g_slist_insert_sorted_with_data ().
97 - Removed the extra check in g_list_sort() and g_slist_sort() for
98 GCompareDataFunc vs. GCompareFunc.
100 2005-12-07 Tor Lillqvist <tml@novell.com>
102 * glib-gettextize.in: Look up prefix at run-time on Win32,
103 assuming the standard directory structure with glib-gettextize in
106 2005-12-06 Matthias Clasen <mclasen@redhat.com>
108 * glib/gmem.h: Exempt GTK+ from the mem chunk deprecation, since
109 we need GTK+ 2.8 to compile against GLib 2.10.
111 2005-12-05 Manish Singh <yosh@gimp.org>
113 * tests/run-collate-test.sh: set LC_COLLATE, not LANG, to be sure
114 to override any user settings.
116 2005-12-05 Matthias Clasen <mclasen@redhat.com>
118 * configure.in: Revert an accidential commit
120 2005-12-05 Matthias Clasen <mclasen@redhat.com>
122 * glib/gutf8.c: Documentation updates. (#323291, Morten Welinder)
124 * tests/Makefile.am (TESTS_ENVIRONMENT):
125 * tests/gobject/Makefile.am (TESTS_ENVIRONMENT): Set
126 MALLOC_CHECK_ and MALLOC_PERTURB_
128 * tests/run-collate-tests.sh: Run the collation tests explicitly
129 in en_US locale. (#320463)
131 * glib/gmem.h: Really deprecate GMemChunk.
138 * glib/gstrfuncs.c: Add versioned deprecation docs.
140 2005-12-05 Martyn Russell <martyn@imendio.com>
142 * docs/reference/glib/glib-sections.txt:
143 * glib/gasyncqueue.[ch]:
144 - Added support for sorting async queues by with _push_sorted(),
145 _push_sorted_unlocked(), _sort() and _sort_unlocked() (#323047).
148 * tests/asyncqueue-test.c:
149 - Added test case for gasyncqueue.c
151 Mon Dec 5 15:53:20 2005 Tim Janik <timj@imendio.com>
153 * glib/gslice.c: implement chain walking for arbitrary ->next pointer
154 offsets in g_slice_free_chain_with_offset() based on a patch by behdad
155 in bug 323178. moved time consuming logic from g_slice_free() out of
156 the inner loop, so g_slice_free_chain_with_offset() provides a real
157 performance benefit over g_slice_free1() now.
159 * glib/gslice.h: renamed g_slice_free_chain() to
160 g_slice_free_chain_with_offset(). implemented g_slice_free_chain() as
161 a type-safe macro as suggested in bug 323178.
162 simplified the macro implementation of g_slice_free() and implemented
163 it in a type safe manner for all compliers as suggested by Morten
164 Welinder <mortenw@gnome.org>.
169 * glib/glib.symbols: s/g_slice_free_chain/g_slice_free_chain_with_offset/
171 2005-12-05 Matthias Clasen <mclasen@redhat.com>
173 * glib/gasyncqueue.c: Add some docs.
175 * tests/libmoduletestplugin_a.c: Fix compiler warnings.
177 * glib/gatomic.c: In the ia64 implementation, use
178 __sync builtin without _si or _di suffix. (#321229,
179 Stanislav Brabec, patch by Andreas Schwab)
181 2005-12-04 Behdad Esfahbod <behdad@gnome.org>
183 * glib/gslice.h: Remove comma at the end of enum.
185 2005-12-04 Matthias Clasen <mclasen@redhat.com>
187 Handle multiple user names with the same UID better.
188 (#319535, Laszlo Peter)
190 * glib/gutils.c (g_get_any_init_do): When determining user
191 data, first look up $LOGNAME. If the UID doesn't match
192 getuid(), fall back to the current behaviour of looking
193 up the user data based on getuid().
195 2005-12-04 Matthias Clasen <mclasen@redhat.com>
197 * glib/gmacros.h (G_GNUC_WARN_UNUSED_RESULT): Add a macro
198 to make gcc warn if a function result is ignored. (#145466,
199 Arjan van de Ven, Alex Larsson)
201 * glib/gmem.h: Add the new attribute to g_realloc and
204 2005-12-03 Matthias Clasen <mclasen@redhat.com>
207 * glib/gcache.h: Deprecate g_cache_value_foreach. (#322956,
210 * glib/gmappedfile.c: Make mapping of empty files
213 * glib/gfileutils.c: Don't fork a new process just to
214 fix the permissions of the created temp file. (#321318,
215 Alexis S. L. Carvalho)
217 2005-12-02 Matthias Clasen <mclasen@redhat.com>
219 * README.in: Add a note about Solaris threads.
222 * configure.in: Remove support for Solaris threads.
223 (#136971, Sebastian Wilhelmi, patch by Andrew Paprocki)
225 2005-12-02 Matthias Clasen <mclasen@redhat.com>
227 * configure.in: Define G_GUINT64_CONSTANT in analogy to
228 G_GINT64_CONSTANT. (#322568, Andrew Paprocki)
230 2005-12-02 Matthias Clasen <mclasen@redhat.com>
232 * glib/gslice.c: Win32 portability fixes and C99ism removal,
233 pointed out by Kazuki Iwamoto. (#323052)
235 2005-12-02 Matthias Clasen <mclasen@redhat.com>
237 * glib/gatomic.c (g_atomic_int_add): Add a missing volatile for
238 the IA64 implementation.
240 Fri Dec 2 16:18:09 2005 Tim Janik <timj@imendio.com>
242 * glib/gslice.c: some naming and type size fixups.
244 Fri Dec 2 13:08:58 2005 Tim Janik <timj@gtk.org>
246 * glib/gslice.h (g_slice_free): reworked GCC-specific type-safe macro
247 variant into something less verbose, but digestible for gcc-3.4.
249 Fri Dec 2 10:55:07 2005 Tim Janik <timj@imendio.com>
251 * tests/slice-test.c: extended to perform the benchmarking on the old
252 memchunk code if 'O' is selected.
254 * tests/memchunks.c: new file which contains the old GLib mem chunks
255 implementation with prefix old_mem_chunk_.
257 * tests/Makefile.am: added memchunks.c
259 Fri Dec 2 00:16:59 2005 Tim Janik <timj@imendio.com>
261 * glib/gslice.c: improved gettimeofday() timer resolution by maintaining
262 acceptable load distribution of the syscall.
263 (allocator_get_magazine_threshold): reduce minimum/single-thread magazine
264 sizes to approximately page_size/2.5 to avoid excessive startup allocations.
266 2005-12-01 Matthias Clasen <mclasen@redhat.com>
268 * tests/gobject/ifacecheck.c (test_iface_base_init): Don't
269 double-free base interface members.
271 * tests/Makefile.am: Remove duplicate variable.
273 Thu Dec 1 17:32:46 2005 Tim Janik <timj@imendio.com>
275 * glib/gslice.[hc]: new slice allocator implementation.
277 * tests/slice-test.c: added random slice allocation test.
279 * glib/gthread.[hc]: removed newly added private thread mem API.
281 * glib/gthreadinit.h:
284 * glib/gmem.c: divided glib threading initialisation into three phases,
285 initialisation where private keys and messaging are not available (only
286 needed by gmem.c), initialisation without messaging but private keys
287 available (gslice.c, gmessage.c), and full fledged initialisers that
288 server the rest of glib. initialisation functions got renamed to reflect
289 the limitations of their corresponding phases.
291 * glib/gmem.c: removed memchunk code, defer allocations to
294 * glib/gmem.[hc]: removed g_slice_* skeletons.
296 * glib/glib.symbols: added g_slice_* symbols.
298 * configure.in: check for availability of posix_memalign(3), memalign(3)
301 * glib/Makefile.am: added gslice.[hc].
303 2005-12-01 Tor Lillqvist <tml@novell.com>
305 * glib/gstdio.c (g_stat): In the Win32 implementation, strip
306 trailing slash(es) for non-root folders. stat() fails if non-root
307 folders are specified with trailing slashes. It's too much hassle
308 to demand that callers strip such slashes themselves, especially
309 as it is easy to get it wrong and strip the slash of a root
311 (g_rename): On NT-based Windows, use MoveFileEx() with
312 MOVEFILE_REPLACE_EXISTING to better match Unix behaviour.
314 2005-11-28 Matthias Clasen <mclasen@redhat.com>
316 Fix G_STMT_START / G_STMT_END on Solaris. (#321972,
319 * configure.in: Check whether do { } while (0) works.
321 * glib/gmacros.h: Use do { } while (0) for G_STMT_START /
322 G_STMT_END if it works.
324 2005-11-28 Matthias Clasen <mclasen@redhat.com>
326 * glib/gthread.c (g_static_rw_lock_wait, g_static_rw_lock_signal):
327 * glib/gnode.c (g_node_depth_traverse_level):
328 * glib/gmem.c (g_allocator_new):
329 * glib/ghash.c (g_hash_table_unref, g_hash_table_destroy)
330 (g_hash_table_foreach_remove_or_steal): Silence compiler
333 2005-11-27 Matthias Clasen <mclasen@redhat.com>
335 * glib/gunicollate.c (g_utf8_collate_key_for_filename):
336 Don't read beyond len. (#322520, Christian Persch)
338 Wed Nov 23 17:34:01 2005 Tim Janik <timj@imendio.com>
340 * glib/gdataset.c: access datalist flags via atomic pointer access
341 functions, instead of acquiring the dataset lock. this is faster and
342 also matches the atomic pointer readouts in gdatalistprivate.h and
343 direct pointer modifications required by gobject.c.
345 Wed Nov 23 13:35:31 2005 Tim Janik <timj@gtk.org>
347 * glib/gdataset.c: streamlined initialization code somewhat,
348 removed GData node cache.
350 2005-11-22 Matthias Clasen <mclasen@redhat.com>
352 * glib/ghash.c (g_hash_table_ref):
353 (g_hash_table_unref): Mark these as new API.
355 Avoid double locking in g_intern_string (#322133,
358 * glib/gdataset.c (g_quark_from_string_internal): New
359 internal function which factors out common parts of
360 g_quark_from[_static]_string.
361 (g_quark_from_string, g_quark_from_static_string):
362 Use g_quark_from_string_internal.
363 (g_intern_string, g_intern_static_string): Use
364 g_quark_from_string_internal, and only take the
366 (g_quark_new): Don't store the strings shifted by -1
367 in the g_quarks array.
368 (g_quark_to_string): Adapt to the previous change.
370 Tue Nov 22 14:04:26 2005 Tim Janik <timj@imendio.com>
374 g_hash_table_new_full(): create hash tables with a ref count of 1.
375 g_hash_table_ref(): atomically ref_count+=1
376 g_hash_table_unref(): atomically ref_count-=1, destroys hash table
377 when refcount reaches 0.
378 g_hash_table_destroy(): just destroy keys and values, unref by 1.
379 g_hash_table_insert():
380 g_hash_table_replace(): assert ref_count>0.
383 * glib/gatomic.c: added 'volatile' qualifier to all atomic pointer and
386 2005-11-20 Behdad Esfahbod <behdad@gnome.org>
388 * glib/guniprop.c (g_unichar_get_mirror_char): Remove unused
389 variables. (#321984, Andrew Paprocki)
391 2005-11-18 Matthias Clasen <mclasen@redhat.com>
393 * configure.in: Bump version
395 2005-11-17 Matthias Clasen <mclasen@redhat.com>
397 * === Released 2.9.0 ===
399 * glib/glib.symbols: Updates
401 * glib/gutf8.c (g_utf8_offset_to_pointer): Add some performance
402 hints to the docs. (Paolo Borelli)
406 * glib/gthread.c (g_thread_foreach): Mark as new api.
408 * README.in: Updates.
412 * glib/gdate.c (g_date_set_time_t):
413 (g_date_set_time_val): Add functions to set a date from
414 a time_t and from a GTimeVal.
415 (g_date_set_time): Deprecate. (#314067, Roger Leigh)
418 * tests/utf8-pointer.c: Unit tests for g_utf8_pointer_to_offset
419 and g_utf8_offset_to_pointer.
421 * glib/gutf8.c (g_utf8_pointer_to_offset)
422 (g_utf8_offset_to_pointer): Handle negative offsets, and use
423 "stutter stepping" for going backwards. (#320638, Larry
428 * glib/gthread.c: const correctness fixes, found
429 by Arjan van de Ven and gcc.
431 2005-11-16 Behdad Esfahbod <behdad@gnome.org>
433 * docs/reference/glib/tmpl/unicode.sgml: Correct typo on mentioning
434 @G_UNICODE_BREAK_UNKNOWN that should be #G_UNICODE_BREAK_UNKNOWN.
436 2005-11-09 Behdad Esfahbod <behdad@gnome.org>
438 * glib/guniprop.c: Use bit hacks instead when checking a general
439 category value against multiple values.
441 * glib/gutf8.c: Change ISO10646 to Unicode in docs.
443 2005-11-10 Simos Xenitellis <simos@gnome.org>
445 * configure.in: Added tt (Tatar) to ALL_LINGUAS.
447 2005-11-05 Tor Lillqvist <tml@novell.com>
449 * glib/giowin32.c (g_io_win32_set_flags): Don't call g_warning().
450 (#320688) Instead set error and return failure.
452 2005-11-07 Matthias Clasen <mclasen@redhat.com>
454 * glib/gmain.c (g_main_dispatch): Don't call cb_funcs->unref
455 while holding the context lock. (#320886, Andy Wingo)
457 2005-11-07 Billy Biggs <vektor@dumbterm.net>
459 * tests/refcount/Makefile.am: Add a missing reference to libglib
460 in the LDADD for this test.
462 2005-11-07 Matthias Clasen <mclasen@redhat.com>
464 * glib/gmem.c (g_allocator_new): We need to set n_preallocs to a
465 nonzero value, otherwise GTK+ 2.8 breaks when compiled against
466 GLib 2.9. (#320755, Luca Ferretti)
468 2005-11-04 Matthias Clasen <mclasen@redhat.com>
470 * glib/Makefile.am: Apply a patch from Behdad Esfahbod to
471 use a faster lookup table for g_unichar_get_mirror_char().
472 * glib/gmirroringtable.h: The new table.
473 * glib/gunichartables.h: Remove bidi_mirroring_table.
474 * glib/gen-unicode-tables.pl: Don't generate the mirroring
476 * glib/glib-mirroring-tab/*: Sources for the program
477 which generated gmirroringtable.h.
479 * glib/glist.c: Avoid some code duplication.
481 * glib/gscanner.h: Include gdataset.h. (#320322)
483 * glib/gdebug.h: add new GDebugFlag for fatal_criticals
484 * glib/gmessages.c: (_g_debug_init): handle G_DEBUG=fatal_criticals,
485 to help find critical warnings in applications. (#320017,
488 2005-11-02 Tor Lillqvist <tml@novell.com>
490 * glib/glib.symbols: Remove large amount of trailing whitespace
491 from one line. Remove #ifdef G_UNDEFINED from around
492 g_hash_table_new and g_hash_table_new_full symbols, what was the
493 purpose of that? Add the g_slice functions.
495 Tue Nov 1 16:24:20 2005 Tim Janik <timj@imendio.com>
497 * glib/gmem.[hc]: prepared deprecation of GMemChunk and GAllocator.
498 added g_slice_*() API to allocate and cache small bits of memory.
499 an actuall allocator implementation for g_slice_*() is still pending.
501 * glib/gthread.[hc]: changes from a patch by Matthias Clasen.
502 changed GRealThread list to use in-structure *next; fields instead
503 of GSList, in order for thread iteration to not depenend on g_slice_*()
505 _g_thread_mem_private_get():
506 _g_thread_mem_private_set(): added accessors for private memory,
507 needed because the ordinary GPrivate implementation relies on GArray
508 and GSList and therefore indirectly on working g_slice_*() allocations.
511 g_thread_foreach(): new public API function to loop over all existing threads.
525 removed GAllocator and free list usages and accompanying locks.
526 use g_slice_*() API to allocate and cache small bits of memory.
528 * glib/ghook.h: removed GMemChunk field from public API.
531 * glib/glist.h: deprecate allocator API, provide _free1() for consistency.
533 * glib/gnode.h: deprecate allocator API.
535 * glib/gmain.c: reordered GPollRec fields so g_slice_free_chain() can
536 be used for poll rec lists.
538 * glib/grel.c: removed mem chunk usage, and allocated tuples via g_slice_*().
539 g_relation_destroy(): free all tuples from the all_tuples hash table,
540 this effectively maintains the life time track keeping of tuples.
541 g_relation_delete_tuple(): free tuples which are removed from the
542 all_tuples hash table. this fixes a temporary leak that was present
543 in the memchunk code until the destruction of the relation.
545 2005-10-29 Matthias Clasen <mclasen@redhat.com>
547 * tests/convert-test.c: Add some tests for conversions between
548 UTF-8, UCS-4 and UTF-16.
550 * glib/gutf8.c (g_utf8_to_ucs4, g_utf8_to_utf16): Fix handling
551 of len == -1, noticed by Morten Welinder.
553 2005-10-27 Erdal Ronahi <erdal.ronahi@gmail.com>
555 * configure.in: Added ku (Kurdish) to ALL_LINGUAS
557 2005-10-26 Matthias Clasen <mclasen@redhat.com>
559 * glib/gutf8.c (g_ucs4_to_utf8): Don't set items_read twice
560 in the error case, and add some documentation. (#319806, Morten
563 2005-10-19 Manish Singh <yosh@gimp.org>
565 * configure.in: Use AC_CHECK_FUNCS for _NSGetEnviron, to get the
566 config.h symbol automatically. Fixes bug #313731.
568 2005-10-19 Tor Lillqvist <tml@novell.com>
570 * glib/gwin32.c (g_win32_get_package_installation_directory):
571 Return a g_strdup()ed copy of the value stored in the hash table,
572 so that it can be g_free()d without leaving a dangling pointer in
573 the hash table. (#319232)
575 2005-10-06 Matthias Clasen <mclasen@redhat.com>
577 * glib/gunicollate.c (g_utf8_collate_key_for_filename): Handle
578 all-zero sequences correctly. (#317930, Sebastien Bacher)
580 2005-10-05 Matthias Clasen <mclasen@redhat.com>
582 * glib/gmarkup.c (g_markup_escape_text): Doc addition
584 * Makefile.am: Add Collation tests.
586 * tests/collate/*: Inputs and expected outputs for collation tests.
588 * tests/run-collate-tests.sh: Script to run collation tests.
590 * tests/unicode-collate.c (main): Rework slightly to make
591 it usable in unit tests. Also test g_utf8_collate_key_for_filename().
593 2005-10-01 Behdad Esfahbod <behdad@gnome.org>
595 * docs/reference/glib/tmpl/unicode.sgml:
596 * glib/gen-unicode-tables.pl:
598 * glib/gunichartables.h:
600 * tests/casefold.txt:
601 * tests/casemap.txt: Updated to Unicode 4.1. There are five new
602 GUnicodeBreakType types. That may break some applications, like
605 2005-09-26 Matthias Clasen <mclasen@redhat.com>
607 * glib/gstrfuncs.c (g_ascii_strtoull): Add details to the
608 docs. (#314393, Matthew F. Barnes)
611 * glib/gprintf.h: Remove g_snprintf() and g_vsnprintf(), since
612 they are already declared in glib.h. This doesn't break documented
613 use of gprintf.h, but should probably be pointed out in the
614 release notes for 2.10. (#314232, Behdad Esfahbod)
616 Tue Sep 20 13:16:04 2005 Tim Janik <timj@imendio.com>
618 * glib/gpattern.c (g_pattern_ph_match): applied significant recursion
619 complexity optimization, based on a patch from Matthias Clasen.
621 * tests/patterntest.c: more tests, mostly from matthias.
623 2005-09-20 Matthias Clasen <mclasen@redhat.com>
625 * glib/gqueue.c (g_queue_insert_sorted): Correct the docs.
626 (#316703, Mark Drago)
628 Mon Sep 19 17:23:23 2005 Tim Janik <timj@imendio.com>
630 * glib/gpattern.c: applied a patch from matthias which checks on the
631 upper bound of GPatternSpec length to optimize matches.
634 * tests/patterntest.c: added more match cases.
636 2005-09-16 Tor Lillqvist <tml@novell.com>
638 * glib/gstrfuncs.c (g_ascii_strcasecmp, g_ascii_strncasecmp): Add
639 warning to doc comment that these functions should not be used on
640 encodings like CP932.
642 2005-09-14 Matthias Clasen <mclasen@redhat.com>
644 * tests/keyfile-test.c: Add a test for grup names of length 1.
646 * glib/gkeyfile.c (g_key_file_line_is_group): Accept group names
647 of length 1. (#316309)
649 2005-09-12 Matthias Clasen <mclasen@redhat.com>
651 * glib/gmarkup.c (g_markup_escape_text): Clarify docs.
653 2005-09-11 Sebastian Wilhelmi <seppi@seppi.de>
655 * tests/refcount/Makefile.am (INCLUDES): Link the the refcount
656 tests to the system thread library $(G_THREAD_LIBS). Fixes #313744
659 2005-09-11 Kjartan Maraas <kmaraas@gnome.org>
661 * glib/gmain.c: (g_child_watch_prepare), (g_child_watch_check),
662 (child_watch_helper_thread): Remove some dead code. Closes
665 2005-09-07 Tor Lillqvist <tml@novell.com>
667 * glib/Makefile.am: Create also a console version of the
668 gspawn-win32-helper program, gspawn-win32-helper-console.exe.
669 It's otherwise identical to gspawn-win32-helper.exe, except marked
670 as a console application (linked without the -mwindows option).
672 * glib/gspawn-win32.c (do_spawn_directly, do_spawn_with_pipes):
673 Drop the dont_wait parameter. Its truth value correlated 100% with
674 the NULLness of the exit_status parameter anyway, so it's enough
675 to check whether exit_status is NULL. Invert the sense of the
676 dont_return_handle parameter and rename it to do_return_handle, to
677 make the code easier to read by avoiding double negations.
679 (g_spawn_sync_utf8, g_spawn_async_with_pipes_utf8): Modify calls
680 to do_spawn_with_pipes() accordingly.
682 (do_spawn_with_pipes): If we have a console, use the console
683 version of the helper program, otherwise use the GUI one. This
684 avoids extra console windows opening up in some situations. (In
685 case a console application uses the GUI gspawn-win32-helper.exe to
686 spawn another console application we would get a separate console
687 for the spawned console application).
689 * glib-zip.in: Distribute also gspawn-win32-helper-console.exe.
691 2005-09-05 Matthias Clasen <mclasen@redhat.com>
693 * glib/gmappedfile.c (g_mapped_file_new): Report an error
694 if the file is too large. (#315275, Kjartan Maraas)
696 * glib/gkeyfile.c (g_key_file_load_from_fd): The return value
697 of read() is signed. (#315273, Kjartan Maraas)
699 2005-08-31 Tor Lillqvist <tml@novell.com>
701 * glib/gutils.h: Wrapping atexit() is a bad idea on Windows, where
702 the EXE and each DLL have their own atexit function chains.
704 #define g_atexit as atexit instead. This means it has a
705 better chance of doing what the caller wants. For instance,
706 gtkhtml calls g_atexit() registering a function in gtkhtml
707 itself. This caused a crash when g_atexit() was implemented as a
708 function in the GLib DLL. The gtkhtml DLL was already unloaded by
709 the time the GLib DLL got unloaded.
711 * glib/gutils.c: #undef the #define mentioned above, to also get a
712 real g_atexit() into the DLL for backward compatibility. Document
713 the Windows behaviour of g_atexit(), and document the varying ways
714 atexit() can behave in the context of dynamically loaded modules
717 2005-08-31 Matthias Clasen <mclasen@redhat.com>
721 * glib/gdataset.c: Add string interning functions.
723 2005-08-28 Matthias Clasen <mclasen@redhat.com>
725 * glib/giochannel.c: Unify some near-duplicate strings. (#314654,
728 2005-08-26 Matthias Clasen <mclasen@redhat.com>
730 * configure.in: Bump version to 2.9.0
732 2005-08-25 Tor Lillqvist <tml@novell.com>
734 Make also the g_spawn*() functions take parameters in the GLib
735 file name encoding, i.e. UTF-8, on Windows. Has no impact on Unix
736 API or ABI. Like the other GLib API that was earlier changed to
737 use UTF-8 on Windows, the names of the functions that take UTF-8
738 have _utf8 suffixes added by using preprocessor macros in the
739 header file. The old names are kept for functions with the old
740 behaviour, taking parameters in the system codepage, for DLL ABI
743 * glib/gspawn.h: On Win32 add the suffix _utf8 to the names of the
744 g_spawn*() functions.
746 * glib/gspawn-win32.c: Use wide-char API on NT-based
747 Windows. Convert parameters from UTF-8 to wide chars (NT) or
748 system codepage (Win9x) and call the C library _wspawn*() or
749 spawn*() functions respectvely. Add DLL ABI stability versions
750 that take parameters in the system codepage.
752 * glib/gspawn-win32-helper.c: On NT-based Windows use the
753 wide-char versions of argv and envp, and use wide-char API to
754 change directory and spawn the program to run. Remove the verbose
755 debugging output, it was too complex to modify for the wide-char
756 features. (Just add temporary debugging printouts if needed, no
757 need to have them permanently in the source.)
759 * glib/gspawn.c: Corresponding documentation updates.
761 * glib/glib.symbols: Corresponding changes: Mark the ABI stability
762 symbols as PRIVATE, add the new _utf8-suffixed ones.
764 2005-08-24 Stepan Kasal <kasal@ucw.cz>
766 * glib/gtypes.h (G_MININT64): Cast the constant to gint64; it is
767 guint64 otherwise and that can produce warnings about comparison
768 between signed and unsigned.
770 2005-08-23 Matthias Clasen <mclasen@redhat.com>
772 * glib/gutils.c: Fix the crt_externs.h include.
774 2005-08-23 Stepan Kasal <kasal@ucw.cz>
776 * NEWS: Fix spelling of my first name.
778 2005-08-23 Matthias Clasen <mclasen@redhat.com>
782 * === Released 2.8.1 ===
786 2005-08-20 Hans Breuer <hans@breuer.org>
788 * glib/makefile.msc.in : link with ws2_32.lib
790 2005-08-18 Tor Lillqvist <tml@novell.com>
792 * configure.in: Check for <sys/wait.h>
794 * glib/gbacktrace.c: Include <sys/wait.h> on if HAVE_SYS_WAIT_H.
796 2005-08-18 Ross Burton <ross@burtonini.com>
799 Optimise single-character insertions.
804 * tests/string-test.c:
805 Add tests for new optimisation, and fix a leak.
807 2005-08-17 Matthias Clasen <mclasen@redhat.com>
809 * configure.in: Check for crt_externs.h and _NSGetEnviron.
811 * glib/gutils.c: On Darwin, include crt-externs.h and
812 define environ using _NSGetEnviron(). (#313731)
814 2005-08-16 Stepan Kasal <kasal@ucw.cz>
816 * glib/gutils.c (g_get_any_init): Move the body of the big if...
817 (g_get_any_init_do): ... to this new function.
818 (g_get_any_init): Declare as inline.
819 (g_get_any_init_locked): New inline function, does the locking.
820 Make use of these two throughout the code.
822 2005-08-15 Matthias Clasen <mclasen@redhat.com>
824 * glib/gbacktrace.c (g_on_error_stack_trace): Wait for
825 the child process and then simply return. This makes
826 The "S" option work as documented in g_on_error_query().
827 (#313125, Matthew F. Barnes)
829 * glib/gunicode.h: Update the link to Unicode category
830 values. (#313369, Behnam Esfahbod)
832 * glib/gqueue.c (g_queue_find_custom): Clarify docs
833 a little. (#311727, Tristan van Berkom)
835 * glib/abicheck.sh, gobject/abicheck.sh: Make the
836 check work on ia64 too, where some symbols ended up
837 in yet another section.