1 2005-12-21 Matthias Clasen <mclasen@redhat.com>
3 * README.in: Add some notes on when 'make check' may fail.
5 2005-12-20 Matthias Clasen <mclasen@redhat.com>
9 * glib/gthreadpool.c (g_thread_pool_set_sort_function): New function
10 to sort tasks pushed into a threadpool. (#324479, Martyn Russell)
12 * tests/threadpool-test.c: Test this.
14 Tue Dec 20 18:14:14 2005 Tim Janik <timj@imendio.com>
16 * glib/gslice.[hc]: added mem_error() and mem_assert() to test and
17 handle errors without depending on gmessage.c which might not be
18 setup when the error occours.
19 removed G_SLICE_CONFIG_ALWAYS_FREE config option, fixed the code so
20 always freeing can be achieved by adjusting the working set time to
21 0 with G_SLICE_CONFIG_WORKING_SET_MSECS.
22 added G_SLICE_CONFIG_COLOR_INCREMENT to test different color increments
23 (mainly 0 and 1). reduced the minimum block size to 128 bytes, to
24 minimize wastage if small amounts of differently sized structrues are
25 allocated, this does come at a performance cost of roughly 5% though.
26 fixed up block alignment calculation, so it works for varying
27 block sizes. only use strerror() not g_strerror() since the latter
28 depends on working GQuark and GSlice.
29 mem_error(): implemented in terms of fprintf and vfprintf.
31 * tests/slice-color.c: new program to test cache colorization effects.
33 * tests/slice-test.c: trade G_SLICE_CONFIG_ALWAYS_FREE for 0 duration
34 G_SLICE_CONFIG_WORKING_SET_MSECS.
36 2005-12-17 Matthias Clasen <mclasen@redhat.com>
38 * glib/goption.c (parse_short_option): Set an error in all
39 failure cases. (#324332, Tim-Philipp Müller)
41 2005-12-17 Sebastian Wilhelmi <seppi@seppi.de>
43 * glib/gatomic.c: Fix memory barrier position in g_atomic_int_get
44 and g_atomic_pointer_get. Add g_atomic_int_set and
45 g_atomic_pointer_set implementations for the !DEFINE_WITH_MUTEXES &&
46 G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case, as well as defining them
47 as functions (additionally to the macros in the header) for the
48 !G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case.
50 2005-12-16 Matthias Clasen <mclasen@redhat.com>
52 * glib/gmem.c (g_allocator_new): Don't return a pointer to
53 a const struct, since apps expect to be able to modify it.
54 (#324179, J. Ali Harlow)
56 Tue Dec 13 10:13:32 2005 Tim Janik <timj@imendio.com>
58 * glib/gatomic.h: added g_atomic_pointer_set() and g_atomic_int_set()
59 for systems where the initialization of atomic variables requires a
62 2005-12-09 Matthias Clasen <mclasen@redhat.com>
64 * configure.in: Bump version
66 * === Released 2.9.1 ===
68 2005-12-09 Alexander Larsson <alexl@redhat.com>
72 Use G_GNUC_WARN_UNUSED_RESULT on list functions that return
75 2005-12-08 Matthias Clasen <mclasen@redhat.com>
79 2005-12-07 Martyn Russell <martyn@imendio.com>
82 - Call g_queue_insert_sorted() instead of duplicating the code.
83 - Call g_queue_sort() instead of duplicating the code.
84 - Invert sort function results to make sure the same sort function
85 gives the same results across glist, gslist, gqueue and
88 * tests/asyncqueue-test.c:
89 - Updated the sort function to reflect the example in the
90 documentation for gasyncqueue.c.
92 2005-12-07 Martyn Russell <martyn@imendio.com>
94 * docs/reference/glib/glib-sections.txt:
95 * docs/reference/glib/tmpl/linked_lists_double.sgml:
96 * docs/reference/glib/tmpl/linked_lists_single.sgml:
99 - Added g_list_insert_sorted_with_data () and
100 g_slist_insert_sorted_with_data ().
101 - Removed the extra check in g_list_sort() and g_slist_sort() for
102 GCompareDataFunc vs. GCompareFunc.
104 2005-12-07 Tor Lillqvist <tml@novell.com>
106 * glib-gettextize.in: Look up prefix at run-time on Win32,
107 assuming the standard directory structure with glib-gettextize in
110 2005-12-06 Matthias Clasen <mclasen@redhat.com>
112 * glib/gmem.h: Exempt GTK+ from the mem chunk deprecation, since
113 we need GTK+ 2.8 to compile against GLib 2.10.
115 2005-12-05 Manish Singh <yosh@gimp.org>
117 * tests/run-collate-test.sh: set LC_COLLATE, not LANG, to be sure
118 to override any user settings.
120 2005-12-05 Matthias Clasen <mclasen@redhat.com>
122 * configure.in: Revert an accidential commit
124 2005-12-05 Matthias Clasen <mclasen@redhat.com>
126 * glib/gutf8.c: Documentation updates. (#323291, Morten Welinder)
128 * tests/Makefile.am (TESTS_ENVIRONMENT):
129 * tests/gobject/Makefile.am (TESTS_ENVIRONMENT): Set
130 MALLOC_CHECK_ and MALLOC_PERTURB_
132 * tests/run-collate-tests.sh: Run the collation tests explicitly
133 in en_US locale. (#320463)
135 * glib/gmem.h: Really deprecate GMemChunk.
142 * glib/gstrfuncs.c: Add versioned deprecation docs.
144 2005-12-05 Martyn Russell <martyn@imendio.com>
146 * docs/reference/glib/glib-sections.txt:
147 * glib/gasyncqueue.[ch]:
148 - Added support for sorting async queues by with _push_sorted(),
149 _push_sorted_unlocked(), _sort() and _sort_unlocked() (#323047).
152 * tests/asyncqueue-test.c:
153 - Added test case for gasyncqueue.c
155 Mon Dec 5 15:53:20 2005 Tim Janik <timj@imendio.com>
157 * glib/gslice.c: implement chain walking for arbitrary ->next pointer
158 offsets in g_slice_free_chain_with_offset() based on a patch by behdad
159 in bug 323178. moved time consuming logic from g_slice_free() out of
160 the inner loop, so g_slice_free_chain_with_offset() provides a real
161 performance benefit over g_slice_free1() now.
163 * glib/gslice.h: renamed g_slice_free_chain() to
164 g_slice_free_chain_with_offset(). implemented g_slice_free_chain() as
165 a type-safe macro as suggested in bug 323178.
166 simplified the macro implementation of g_slice_free() and implemented
167 it in a type safe manner for all compliers as suggested by Morten
168 Welinder <mortenw@gnome.org>.
173 * glib/glib.symbols: s/g_slice_free_chain/g_slice_free_chain_with_offset/
175 2005-12-05 Matthias Clasen <mclasen@redhat.com>
177 * glib/gasyncqueue.c: Add some docs.
179 * tests/libmoduletestplugin_a.c: Fix compiler warnings.
181 * glib/gatomic.c: In the ia64 implementation, use
182 __sync builtin without _si or _di suffix. (#321229,
183 Stanislav Brabec, patch by Andreas Schwab)
185 2005-12-04 Behdad Esfahbod <behdad@gnome.org>
187 * glib/gslice.h: Remove comma at the end of enum.
189 2005-12-04 Matthias Clasen <mclasen@redhat.com>
191 Handle multiple user names with the same UID better.
192 (#319535, Laszlo Peter)
194 * glib/gutils.c (g_get_any_init_do): When determining user
195 data, first look up $LOGNAME. If the UID doesn't match
196 getuid(), fall back to the current behaviour of looking
197 up the user data based on getuid().
199 2005-12-04 Matthias Clasen <mclasen@redhat.com>
201 * glib/gmacros.h (G_GNUC_WARN_UNUSED_RESULT): Add a macro
202 to make gcc warn if a function result is ignored. (#145466,
203 Arjan van de Ven, Alex Larsson)
205 * glib/gmem.h: Add the new attribute to g_realloc and
208 2005-12-03 Matthias Clasen <mclasen@redhat.com>
211 * glib/gcache.h: Deprecate g_cache_value_foreach. (#322956,
214 * glib/gmappedfile.c: Make mapping of empty files
217 * glib/gfileutils.c: Don't fork a new process just to
218 fix the permissions of the created temp file. (#321318,
219 Alexis S. L. Carvalho)
221 2005-12-02 Matthias Clasen <mclasen@redhat.com>
223 * README.in: Add a note about Solaris threads.
226 * configure.in: Remove support for Solaris threads.
227 (#136971, Sebastian Wilhelmi, patch by Andrew Paprocki)
229 2005-12-02 Matthias Clasen <mclasen@redhat.com>
231 * configure.in: Define G_GUINT64_CONSTANT in analogy to
232 G_GINT64_CONSTANT. (#322568, Andrew Paprocki)
234 2005-12-02 Matthias Clasen <mclasen@redhat.com>
236 * glib/gslice.c: Win32 portability fixes and C99ism removal,
237 pointed out by Kazuki Iwamoto. (#323052)
239 2005-12-02 Matthias Clasen <mclasen@redhat.com>
241 * glib/gatomic.c (g_atomic_int_add): Add a missing volatile for
242 the IA64 implementation.
244 Fri Dec 2 16:18:09 2005 Tim Janik <timj@imendio.com>
246 * glib/gslice.c: some naming and type size fixups.
248 Fri Dec 2 13:08:58 2005 Tim Janik <timj@gtk.org>
250 * glib/gslice.h (g_slice_free): reworked GCC-specific type-safe macro
251 variant into something less verbose, but digestible for gcc-3.4.
253 Fri Dec 2 10:55:07 2005 Tim Janik <timj@imendio.com>
255 * tests/slice-test.c: extended to perform the benchmarking on the old
256 memchunk code if 'O' is selected.
258 * tests/memchunks.c: new file which contains the old GLib mem chunks
259 implementation with prefix old_mem_chunk_.
261 * tests/Makefile.am: added memchunks.c
263 Fri Dec 2 00:16:59 2005 Tim Janik <timj@imendio.com>
265 * glib/gslice.c: improved gettimeofday() timer resolution by maintaining
266 acceptable load distribution of the syscall.
267 (allocator_get_magazine_threshold): reduce minimum/single-thread magazine
268 sizes to approximately page_size/2.5 to avoid excessive startup allocations.
270 2005-12-01 Matthias Clasen <mclasen@redhat.com>
272 * tests/gobject/ifacecheck.c (test_iface_base_init): Don't
273 double-free base interface members.
275 * tests/Makefile.am: Remove duplicate variable.
277 Thu Dec 1 17:32:46 2005 Tim Janik <timj@imendio.com>
279 * glib/gslice.[hc]: new slice allocator implementation.
281 * tests/slice-test.c: added random slice allocation test.
283 * glib/gthread.[hc]: removed newly added private thread mem API.
285 * glib/gthreadinit.h:
288 * glib/gmem.c: divided glib threading initialisation into three phases,
289 initialisation where private keys and messaging are not available (only
290 needed by gmem.c), initialisation without messaging but private keys
291 available (gslice.c, gmessage.c), and full fledged initialisers that
292 server the rest of glib. initialisation functions got renamed to reflect
293 the limitations of their corresponding phases.
295 * glib/gmem.c: removed memchunk code, defer allocations to
298 * glib/gmem.[hc]: removed g_slice_* skeletons.
300 * glib/glib.symbols: added g_slice_* symbols.
302 * configure.in: check for availability of posix_memalign(3), memalign(3)
305 * glib/Makefile.am: added gslice.[hc].
307 2005-12-01 Tor Lillqvist <tml@novell.com>
309 * glib/gstdio.c (g_stat): In the Win32 implementation, strip
310 trailing slash(es) for non-root folders. stat() fails if non-root
311 folders are specified with trailing slashes. It's too much hassle
312 to demand that callers strip such slashes themselves, especially
313 as it is easy to get it wrong and strip the slash of a root
315 (g_rename): On NT-based Windows, use MoveFileEx() with
316 MOVEFILE_REPLACE_EXISTING to better match Unix behaviour.
318 2005-11-28 Matthias Clasen <mclasen@redhat.com>
320 Fix G_STMT_START / G_STMT_END on Solaris. (#321972,
323 * configure.in: Check whether do { } while (0) works.
325 * glib/gmacros.h: Use do { } while (0) for G_STMT_START /
326 G_STMT_END if it works.
328 2005-11-28 Matthias Clasen <mclasen@redhat.com>
330 * glib/gthread.c (g_static_rw_lock_wait, g_static_rw_lock_signal):
331 * glib/gnode.c (g_node_depth_traverse_level):
332 * glib/gmem.c (g_allocator_new):
333 * glib/ghash.c (g_hash_table_unref, g_hash_table_destroy)
334 (g_hash_table_foreach_remove_or_steal): Silence compiler
337 2005-11-27 Matthias Clasen <mclasen@redhat.com>
339 * glib/gunicollate.c (g_utf8_collate_key_for_filename):
340 Don't read beyond len. (#322520, Christian Persch)
342 Wed Nov 23 17:34:01 2005 Tim Janik <timj@imendio.com>
344 * glib/gdataset.c: access datalist flags via atomic pointer access
345 functions, instead of acquiring the dataset lock. this is faster and
346 also matches the atomic pointer readouts in gdatalistprivate.h and
347 direct pointer modifications required by gobject.c.
349 Wed Nov 23 13:35:31 2005 Tim Janik <timj@gtk.org>
351 * glib/gdataset.c: streamlined initialization code somewhat,
352 removed GData node cache.
354 2005-11-22 Matthias Clasen <mclasen@redhat.com>
356 * glib/ghash.c (g_hash_table_ref):
357 (g_hash_table_unref): Mark these as new API.
359 Avoid double locking in g_intern_string (#322133,
362 * glib/gdataset.c (g_quark_from_string_internal): New
363 internal function which factors out common parts of
364 g_quark_from[_static]_string.
365 (g_quark_from_string, g_quark_from_static_string):
366 Use g_quark_from_string_internal.
367 (g_intern_string, g_intern_static_string): Use
368 g_quark_from_string_internal, and only take the
370 (g_quark_new): Don't store the strings shifted by -1
371 in the g_quarks array.
372 (g_quark_to_string): Adapt to the previous change.
374 Tue Nov 22 14:04:26 2005 Tim Janik <timj@imendio.com>
378 g_hash_table_new_full(): create hash tables with a ref count of 1.
379 g_hash_table_ref(): atomically ref_count+=1
380 g_hash_table_unref(): atomically ref_count-=1, destroys hash table
381 when refcount reaches 0.
382 g_hash_table_destroy(): just destroy keys and values, unref by 1.
383 g_hash_table_insert():
384 g_hash_table_replace(): assert ref_count>0.
387 * glib/gatomic.c: added 'volatile' qualifier to all atomic pointer and
390 2005-11-20 Behdad Esfahbod <behdad@gnome.org>
392 * glib/guniprop.c (g_unichar_get_mirror_char): Remove unused
393 variables. (#321984, Andrew Paprocki)
395 2005-11-18 Matthias Clasen <mclasen@redhat.com>
397 * configure.in: Bump version
399 2005-11-17 Matthias Clasen <mclasen@redhat.com>
401 * === Released 2.9.0 ===
403 * glib/glib.symbols: Updates
405 * glib/gutf8.c (g_utf8_offset_to_pointer): Add some performance
406 hints to the docs. (Paolo Borelli)
410 * glib/gthread.c (g_thread_foreach): Mark as new api.
412 * README.in: Updates.
416 * glib/gdate.c (g_date_set_time_t):
417 (g_date_set_time_val): Add functions to set a date from
418 a time_t and from a GTimeVal.
419 (g_date_set_time): Deprecate. (#314067, Roger Leigh)
422 * tests/utf8-pointer.c: Unit tests for g_utf8_pointer_to_offset
423 and g_utf8_offset_to_pointer.
425 * glib/gutf8.c (g_utf8_pointer_to_offset)
426 (g_utf8_offset_to_pointer): Handle negative offsets, and use
427 "stutter stepping" for going backwards. (#320638, Larry
432 * glib/gthread.c: const correctness fixes, found
433 by Arjan van de Ven and gcc.
435 2005-11-16 Behdad Esfahbod <behdad@gnome.org>
437 * docs/reference/glib/tmpl/unicode.sgml: Correct typo on mentioning
438 @G_UNICODE_BREAK_UNKNOWN that should be #G_UNICODE_BREAK_UNKNOWN.
440 2005-11-09 Behdad Esfahbod <behdad@gnome.org>
442 * glib/guniprop.c: Use bit hacks instead when checking a general
443 category value against multiple values.
445 * glib/gutf8.c: Change ISO10646 to Unicode in docs.
447 2005-11-10 Simos Xenitellis <simos@gnome.org>
449 * configure.in: Added tt (Tatar) to ALL_LINGUAS.
451 2005-11-05 Tor Lillqvist <tml@novell.com>
453 * glib/giowin32.c (g_io_win32_set_flags): Don't call g_warning().
454 (#320688) Instead set error and return failure.
456 2005-11-07 Matthias Clasen <mclasen@redhat.com>
458 * glib/gmain.c (g_main_dispatch): Don't call cb_funcs->unref
459 while holding the context lock. (#320886, Andy Wingo)
461 2005-11-07 Billy Biggs <vektor@dumbterm.net>
463 * tests/refcount/Makefile.am: Add a missing reference to libglib
464 in the LDADD for this test.
466 2005-11-07 Matthias Clasen <mclasen@redhat.com>
468 * glib/gmem.c (g_allocator_new): We need to set n_preallocs to a
469 nonzero value, otherwise GTK+ 2.8 breaks when compiled against
470 GLib 2.9. (#320755, Luca Ferretti)
472 2005-11-04 Matthias Clasen <mclasen@redhat.com>
474 * glib/Makefile.am: Apply a patch from Behdad Esfahbod to
475 use a faster lookup table for g_unichar_get_mirror_char().
476 * glib/gmirroringtable.h: The new table.
477 * glib/gunichartables.h: Remove bidi_mirroring_table.
478 * glib/gen-unicode-tables.pl: Don't generate the mirroring
480 * glib/glib-mirroring-tab/*: Sources for the program
481 which generated gmirroringtable.h.
483 * glib/glist.c: Avoid some code duplication.
485 * glib/gscanner.h: Include gdataset.h. (#320322)
487 * glib/gdebug.h: add new GDebugFlag for fatal_criticals
488 * glib/gmessages.c: (_g_debug_init): handle G_DEBUG=fatal_criticals,
489 to help find critical warnings in applications. (#320017,
492 2005-11-02 Tor Lillqvist <tml@novell.com>
494 * glib/glib.symbols: Remove large amount of trailing whitespace
495 from one line. Remove #ifdef G_UNDEFINED from around
496 g_hash_table_new and g_hash_table_new_full symbols, what was the
497 purpose of that? Add the g_slice functions.
499 Tue Nov 1 16:24:20 2005 Tim Janik <timj@imendio.com>
501 * glib/gmem.[hc]: prepared deprecation of GMemChunk and GAllocator.
502 added g_slice_*() API to allocate and cache small bits of memory.
503 an actuall allocator implementation for g_slice_*() is still pending.
505 * glib/gthread.[hc]: changes from a patch by Matthias Clasen.
506 changed GRealThread list to use in-structure *next; fields instead
507 of GSList, in order for thread iteration to not depenend on g_slice_*()
509 _g_thread_mem_private_get():
510 _g_thread_mem_private_set(): added accessors for private memory,
511 needed because the ordinary GPrivate implementation relies on GArray
512 and GSList and therefore indirectly on working g_slice_*() allocations.
515 g_thread_foreach(): new public API function to loop over all existing threads.
529 removed GAllocator and free list usages and accompanying locks.
530 use g_slice_*() API to allocate and cache small bits of memory.
532 * glib/ghook.h: removed GMemChunk field from public API.
535 * glib/glist.h: deprecate allocator API, provide _free1() for consistency.
537 * glib/gnode.h: deprecate allocator API.
539 * glib/gmain.c: reordered GPollRec fields so g_slice_free_chain() can
540 be used for poll rec lists.
542 * glib/grel.c: removed mem chunk usage, and allocated tuples via g_slice_*().
543 g_relation_destroy(): free all tuples from the all_tuples hash table,
544 this effectively maintains the life time track keeping of tuples.
545 g_relation_delete_tuple(): free tuples which are removed from the
546 all_tuples hash table. this fixes a temporary leak that was present
547 in the memchunk code until the destruction of the relation.
549 2005-10-29 Matthias Clasen <mclasen@redhat.com>
551 * tests/convert-test.c: Add some tests for conversions between
552 UTF-8, UCS-4 and UTF-16.
554 * glib/gutf8.c (g_utf8_to_ucs4, g_utf8_to_utf16): Fix handling
555 of len == -1, noticed by Morten Welinder.
557 2005-10-27 Erdal Ronahi <erdal.ronahi@gmail.com>
559 * configure.in: Added ku (Kurdish) to ALL_LINGUAS
561 2005-10-26 Matthias Clasen <mclasen@redhat.com>
563 * glib/gutf8.c (g_ucs4_to_utf8): Don't set items_read twice
564 in the error case, and add some documentation. (#319806, Morten
567 2005-10-19 Manish Singh <yosh@gimp.org>
569 * configure.in: Use AC_CHECK_FUNCS for _NSGetEnviron, to get the
570 config.h symbol automatically. Fixes bug #313731.
572 2005-10-19 Tor Lillqvist <tml@novell.com>
574 * glib/gwin32.c (g_win32_get_package_installation_directory):
575 Return a g_strdup()ed copy of the value stored in the hash table,
576 so that it can be g_free()d without leaving a dangling pointer in
577 the hash table. (#319232)
579 2005-10-06 Matthias Clasen <mclasen@redhat.com>
581 * glib/gunicollate.c (g_utf8_collate_key_for_filename): Handle
582 all-zero sequences correctly. (#317930, Sebastien Bacher)
584 2005-10-05 Matthias Clasen <mclasen@redhat.com>
586 * glib/gmarkup.c (g_markup_escape_text): Doc addition
588 * Makefile.am: Add Collation tests.
590 * tests/collate/*: Inputs and expected outputs for collation tests.
592 * tests/run-collate-tests.sh: Script to run collation tests.
594 * tests/unicode-collate.c (main): Rework slightly to make
595 it usable in unit tests. Also test g_utf8_collate_key_for_filename().
597 2005-10-01 Behdad Esfahbod <behdad@gnome.org>
599 * docs/reference/glib/tmpl/unicode.sgml:
600 * glib/gen-unicode-tables.pl:
602 * glib/gunichartables.h:
604 * tests/casefold.txt:
605 * tests/casemap.txt: Updated to Unicode 4.1. There are five new
606 GUnicodeBreakType types. That may break some applications, like
609 2005-09-26 Matthias Clasen <mclasen@redhat.com>
611 * glib/gstrfuncs.c (g_ascii_strtoull): Add details to the
612 docs. (#314393, Matthew F. Barnes)
615 * glib/gprintf.h: Remove g_snprintf() and g_vsnprintf(), since
616 they are already declared in glib.h. This doesn't break documented
617 use of gprintf.h, but should probably be pointed out in the
618 release notes for 2.10. (#314232, Behdad Esfahbod)
620 Tue Sep 20 13:16:04 2005 Tim Janik <timj@imendio.com>
622 * glib/gpattern.c (g_pattern_ph_match): applied significant recursion
623 complexity optimization, based on a patch from Matthias Clasen.
625 * tests/patterntest.c: more tests, mostly from matthias.
627 2005-09-20 Matthias Clasen <mclasen@redhat.com>
629 * glib/gqueue.c (g_queue_insert_sorted): Correct the docs.
630 (#316703, Mark Drago)
632 Mon Sep 19 17:23:23 2005 Tim Janik <timj@imendio.com>
634 * glib/gpattern.c: applied a patch from matthias which checks on the
635 upper bound of GPatternSpec length to optimize matches.
638 * tests/patterntest.c: added more match cases.
640 2005-09-16 Tor Lillqvist <tml@novell.com>
642 * glib/gstrfuncs.c (g_ascii_strcasecmp, g_ascii_strncasecmp): Add
643 warning to doc comment that these functions should not be used on
644 encodings like CP932.
646 2005-09-14 Matthias Clasen <mclasen@redhat.com>
648 * tests/keyfile-test.c: Add a test for grup names of length 1.
650 * glib/gkeyfile.c (g_key_file_line_is_group): Accept group names
651 of length 1. (#316309)
653 2005-09-12 Matthias Clasen <mclasen@redhat.com>
655 * glib/gmarkup.c (g_markup_escape_text): Clarify docs.
657 2005-09-11 Sebastian Wilhelmi <seppi@seppi.de>
659 * tests/refcount/Makefile.am (INCLUDES): Link the the refcount
660 tests to the system thread library $(G_THREAD_LIBS). Fixes #313744
663 2005-09-11 Kjartan Maraas <kmaraas@gnome.org>
665 * glib/gmain.c: (g_child_watch_prepare), (g_child_watch_check),
666 (child_watch_helper_thread): Remove some dead code. Closes
669 2005-09-07 Tor Lillqvist <tml@novell.com>
671 * glib/Makefile.am: Create also a console version of the
672 gspawn-win32-helper program, gspawn-win32-helper-console.exe.
673 It's otherwise identical to gspawn-win32-helper.exe, except marked
674 as a console application (linked without the -mwindows option).
676 * glib/gspawn-win32.c (do_spawn_directly, do_spawn_with_pipes):
677 Drop the dont_wait parameter. Its truth value correlated 100% with
678 the NULLness of the exit_status parameter anyway, so it's enough
679 to check whether exit_status is NULL. Invert the sense of the
680 dont_return_handle parameter and rename it to do_return_handle, to
681 make the code easier to read by avoiding double negations.
683 (g_spawn_sync_utf8, g_spawn_async_with_pipes_utf8): Modify calls
684 to do_spawn_with_pipes() accordingly.
686 (do_spawn_with_pipes): If we have a console, use the console
687 version of the helper program, otherwise use the GUI one. This
688 avoids extra console windows opening up in some situations. (In
689 case a console application uses the GUI gspawn-win32-helper.exe to
690 spawn another console application we would get a separate console
691 for the spawned console application).
693 * glib-zip.in: Distribute also gspawn-win32-helper-console.exe.
695 2005-09-05 Matthias Clasen <mclasen@redhat.com>
697 * glib/gmappedfile.c (g_mapped_file_new): Report an error
698 if the file is too large. (#315275, Kjartan Maraas)
700 * glib/gkeyfile.c (g_key_file_load_from_fd): The return value
701 of read() is signed. (#315273, Kjartan Maraas)
703 2005-08-31 Tor Lillqvist <tml@novell.com>
705 * glib/gutils.h: Wrapping atexit() is a bad idea on Windows, where
706 the EXE and each DLL have their own atexit function chains.
708 #define g_atexit as atexit instead. This means it has a
709 better chance of doing what the caller wants. For instance,
710 gtkhtml calls g_atexit() registering a function in gtkhtml
711 itself. This caused a crash when g_atexit() was implemented as a
712 function in the GLib DLL. The gtkhtml DLL was already unloaded by
713 the time the GLib DLL got unloaded.
715 * glib/gutils.c: #undef the #define mentioned above, to also get a
716 real g_atexit() into the DLL for backward compatibility. Document
717 the Windows behaviour of g_atexit(), and document the varying ways
718 atexit() can behave in the context of dynamically loaded modules
721 2005-08-31 Matthias Clasen <mclasen@redhat.com>
725 * glib/gdataset.c: Add string interning functions.
727 2005-08-28 Matthias Clasen <mclasen@redhat.com>
729 * glib/giochannel.c: Unify some near-duplicate strings. (#314654,
732 2005-08-26 Matthias Clasen <mclasen@redhat.com>
734 * configure.in: Bump version to 2.9.0
736 2005-08-25 Tor Lillqvist <tml@novell.com>
738 Make also the g_spawn*() functions take parameters in the GLib
739 file name encoding, i.e. UTF-8, on Windows. Has no impact on Unix
740 API or ABI. Like the other GLib API that was earlier changed to
741 use UTF-8 on Windows, the names of the functions that take UTF-8
742 have _utf8 suffixes added by using preprocessor macros in the
743 header file. The old names are kept for functions with the old
744 behaviour, taking parameters in the system codepage, for DLL ABI
747 * glib/gspawn.h: On Win32 add the suffix _utf8 to the names of the
748 g_spawn*() functions.
750 * glib/gspawn-win32.c: Use wide-char API on NT-based
751 Windows. Convert parameters from UTF-8 to wide chars (NT) or
752 system codepage (Win9x) and call the C library _wspawn*() or
753 spawn*() functions respectvely. Add DLL ABI stability versions
754 that take parameters in the system codepage.
756 * glib/gspawn-win32-helper.c: On NT-based Windows use the
757 wide-char versions of argv and envp, and use wide-char API to
758 change directory and spawn the program to run. Remove the verbose
759 debugging output, it was too complex to modify for the wide-char
760 features. (Just add temporary debugging printouts if needed, no
761 need to have them permanently in the source.)
763 * glib/gspawn.c: Corresponding documentation updates.
765 * glib/glib.symbols: Corresponding changes: Mark the ABI stability
766 symbols as PRIVATE, add the new _utf8-suffixed ones.
768 2005-08-24 Stepan Kasal <kasal@ucw.cz>
770 * glib/gtypes.h (G_MININT64): Cast the constant to gint64; it is
771 guint64 otherwise and that can produce warnings about comparison
772 between signed and unsigned.
774 2005-08-23 Matthias Clasen <mclasen@redhat.com>
776 * glib/gutils.c: Fix the crt_externs.h include.
778 2005-08-23 Stepan Kasal <kasal@ucw.cz>
780 * NEWS: Fix spelling of my first name.
782 2005-08-23 Matthias Clasen <mclasen@redhat.com>
786 * === Released 2.8.1 ===
790 2005-08-20 Hans Breuer <hans@breuer.org>
792 * glib/makefile.msc.in : link with ws2_32.lib
794 2005-08-18 Tor Lillqvist <tml@novell.com>
796 * configure.in: Check for <sys/wait.h>
798 * glib/gbacktrace.c: Include <sys/wait.h> on if HAVE_SYS_WAIT_H.
800 2005-08-18 Ross Burton <ross@burtonini.com>
803 Optimise single-character insertions.
808 * tests/string-test.c:
809 Add tests for new optimisation, and fix a leak.
811 2005-08-17 Matthias Clasen <mclasen@redhat.com>
813 * configure.in: Check for crt_externs.h and _NSGetEnviron.
815 * glib/gutils.c: On Darwin, include crt-externs.h and
816 define environ using _NSGetEnviron(). (#313731)
818 2005-08-16 Stepan Kasal <kasal@ucw.cz>
820 * glib/gutils.c (g_get_any_init): Move the body of the big if...
821 (g_get_any_init_do): ... to this new function.
822 (g_get_any_init): Declare as inline.
823 (g_get_any_init_locked): New inline function, does the locking.
824 Make use of these two throughout the code.
826 2005-08-15 Matthias Clasen <mclasen@redhat.com>
828 * glib/gbacktrace.c (g_on_error_stack_trace): Wait for
829 the child process and then simply return. This makes
830 The "S" option work as documented in g_on_error_query().
831 (#313125, Matthew F. Barnes)
833 * glib/gunicode.h: Update the link to Unicode category
834 values. (#313369, Behnam Esfahbod)
836 * glib/gqueue.c (g_queue_find_custom): Clarify docs
837 a little. (#311727, Tristan van Berkom)
839 * glib/abicheck.sh, gobject/abicheck.sh: Make the
840 check work on ia64 too, where some symbols ended up
841 in yet another section.