1 Tue Dec 20 18:14:14 2005 Tim Janik <timj@imendio.com>
3 * glib/gslice.[hc]: added mem_error() and mem_assert() to test and
4 handle errors without depending on gmessage.c which might not be
5 setup when the error occours.
6 removed G_SLICE_CONFIG_ALWAYS_FREE config option, fixed the code so
7 always freeing can be achieved by adjusting the working set time to
8 0 with G_SLICE_CONFIG_WORKING_SET_MSECS.
9 added G_SLICE_CONFIG_COLOR_INCREMENT to test different color increments
10 (mainly 0 and 1). reduced the minimum block size to 128 bytes, to
11 minimize wastage if small amounts of differently sized structrues are
12 allocated, this does come at a performance cost of roughly 5% though.
13 fixed up block alignment calculation, so it works for varying
14 block sizes. only use strerror() not g_strerror() since the latter
15 depends on working GQuark and GSlice.
16 mem_error(): implemented in terms of fprintf and vfprintf.
18 * tests/slice-color.c: new program to test cache colorization effects.
20 * tests/slice-test.c: trade G_SLICE_CONFIG_ALWAYS_FREE for 0 duration
21 G_SLICE_CONFIG_WORKING_SET_MSECS.
23 2005-12-17 Matthias Clasen <mclasen@redhat.com>
25 * glib/goption.c (parse_short_option): Set an error in all
26 failure cases. (#324332, Tim-Philipp Müller)
28 2005-12-17 Sebastian Wilhelmi <seppi@seppi.de>
30 * glib/gatomic.c: Fix memory barrier position in g_atomic_int_get
31 and g_atomic_pointer_get. Add g_atomic_int_set and
32 g_atomic_pointer_set implementations for the !DEFINE_WITH_MUTEXES &&
33 G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case, as well as defining them
34 as functions (additionally to the macros in the header) for the
35 !G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case.
37 2005-12-16 Matthias Clasen <mclasen@redhat.com>
39 * glib/gmem.c (g_allocator_new): Don't return a pointer to
40 a const struct, since apps expect to be able to modify it.
41 (#324179, J. Ali Harlow)
43 Tue Dec 13 10:13:32 2005 Tim Janik <timj@imendio.com>
45 * glib/gatomic.h: added g_atomic_pointer_set() and g_atomic_int_set()
46 for systems where the initialization of atomic variables requires a
49 2005-12-09 Matthias Clasen <mclasen@redhat.com>
51 * configure.in: Bump version
53 * === Released 2.9.1 ===
55 2005-12-09 Alexander Larsson <alexl@redhat.com>
59 Use G_GNUC_WARN_UNUSED_RESULT on list functions that return
62 2005-12-08 Matthias Clasen <mclasen@redhat.com>
66 2005-12-07 Martyn Russell <martyn@imendio.com>
69 - Call g_queue_insert_sorted() instead of duplicating the code.
70 - Call g_queue_sort() instead of duplicating the code.
71 - Invert sort function results to make sure the same sort function
72 gives the same results across glist, gslist, gqueue and
75 * tests/asyncqueue-test.c:
76 - Updated the sort function to reflect the example in the
77 documentation for gasyncqueue.c.
79 2005-12-07 Martyn Russell <martyn@imendio.com>
81 * docs/reference/glib/glib-sections.txt:
82 * docs/reference/glib/tmpl/linked_lists_double.sgml:
83 * docs/reference/glib/tmpl/linked_lists_single.sgml:
86 - Added g_list_insert_sorted_with_data () and
87 g_slist_insert_sorted_with_data ().
88 - Removed the extra check in g_list_sort() and g_slist_sort() for
89 GCompareDataFunc vs. GCompareFunc.
91 2005-12-07 Tor Lillqvist <tml@novell.com>
93 * glib-gettextize.in: Look up prefix at run-time on Win32,
94 assuming the standard directory structure with glib-gettextize in
97 2005-12-06 Matthias Clasen <mclasen@redhat.com>
99 * glib/gmem.h: Exempt GTK+ from the mem chunk deprecation, since
100 we need GTK+ 2.8 to compile against GLib 2.10.
102 2005-12-05 Manish Singh <yosh@gimp.org>
104 * tests/run-collate-test.sh: set LC_COLLATE, not LANG, to be sure
105 to override any user settings.
107 2005-12-05 Matthias Clasen <mclasen@redhat.com>
109 * configure.in: Revert an accidential commit
111 2005-12-05 Matthias Clasen <mclasen@redhat.com>
113 * glib/gutf8.c: Documentation updates. (#323291, Morten Welinder)
115 * tests/Makefile.am (TESTS_ENVIRONMENT):
116 * tests/gobject/Makefile.am (TESTS_ENVIRONMENT): Set
117 MALLOC_CHECK_ and MALLOC_PERTURB_
119 * tests/run-collate-tests.sh: Run the collation tests explicitly
120 in en_US locale. (#320463)
122 * glib/gmem.h: Really deprecate GMemChunk.
129 * glib/gstrfuncs.c: Add versioned deprecation docs.
131 2005-12-05 Martyn Russell <martyn@imendio.com>
133 * docs/reference/glib/glib-sections.txt:
134 * glib/gasyncqueue.[ch]:
135 - Added support for sorting async queues by with _push_sorted(),
136 _push_sorted_unlocked(), _sort() and _sort_unlocked() (#323047).
139 * tests/asyncqueue-test.c:
140 - Added test case for gasyncqueue.c
142 Mon Dec 5 15:53:20 2005 Tim Janik <timj@imendio.com>
144 * glib/gslice.c: implement chain walking for arbitrary ->next pointer
145 offsets in g_slice_free_chain_with_offset() based on a patch by behdad
146 in bug 323178. moved time consuming logic from g_slice_free() out of
147 the inner loop, so g_slice_free_chain_with_offset() provides a real
148 performance benefit over g_slice_free1() now.
150 * glib/gslice.h: renamed g_slice_free_chain() to
151 g_slice_free_chain_with_offset(). implemented g_slice_free_chain() as
152 a type-safe macro as suggested in bug 323178.
153 simplified the macro implementation of g_slice_free() and implemented
154 it in a type safe manner for all compliers as suggested by Morten
155 Welinder <mortenw@gnome.org>.
160 * glib/glib.symbols: s/g_slice_free_chain/g_slice_free_chain_with_offset/
162 2005-12-05 Matthias Clasen <mclasen@redhat.com>
164 * glib/gasyncqueue.c: Add some docs.
166 * tests/libmoduletestplugin_a.c: Fix compiler warnings.
168 * glib/gatomic.c: In the ia64 implementation, use
169 __sync builtin without _si or _di suffix. (#321229,
170 Stanislav Brabec, patch by Andreas Schwab)
172 2005-12-04 Behdad Esfahbod <behdad@gnome.org>
174 * glib/gslice.h: Remove comma at the end of enum.
176 2005-12-04 Matthias Clasen <mclasen@redhat.com>
178 Handle multiple user names with the same UID better.
179 (#319535, Laszlo Peter)
181 * glib/gutils.c (g_get_any_init_do): When determining user
182 data, first look up $LOGNAME. If the UID doesn't match
183 getuid(), fall back to the current behaviour of looking
184 up the user data based on getuid().
186 2005-12-04 Matthias Clasen <mclasen@redhat.com>
188 * glib/gmacros.h (G_GNUC_WARN_UNUSED_RESULT): Add a macro
189 to make gcc warn if a function result is ignored. (#145466,
190 Arjan van de Ven, Alex Larsson)
192 * glib/gmem.h: Add the new attribute to g_realloc and
195 2005-12-03 Matthias Clasen <mclasen@redhat.com>
198 * glib/gcache.h: Deprecate g_cache_value_foreach. (#322956,
201 * glib/gmappedfile.c: Make mapping of empty files
204 * glib/gfileutils.c: Don't fork a new process just to
205 fix the permissions of the created temp file. (#321318,
206 Alexis S. L. Carvalho)
208 2005-12-02 Matthias Clasen <mclasen@redhat.com>
210 * README.in: Add a note about Solaris threads.
213 * configure.in: Remove support for Solaris threads.
214 (#136971, Sebastian Wilhelmi, patch by Andrew Paprocki)
216 2005-12-02 Matthias Clasen <mclasen@redhat.com>
218 * configure.in: Define G_GUINT64_CONSTANT in analogy to
219 G_GINT64_CONSTANT. (#322568, Andrew Paprocki)
221 2005-12-02 Matthias Clasen <mclasen@redhat.com>
223 * glib/gslice.c: Win32 portability fixes and C99ism removal,
224 pointed out by Kazuki Iwamoto. (#323052)
226 2005-12-02 Matthias Clasen <mclasen@redhat.com>
228 * glib/gatomic.c (g_atomic_int_add): Add a missing volatile for
229 the IA64 implementation.
231 Fri Dec 2 16:18:09 2005 Tim Janik <timj@imendio.com>
233 * glib/gslice.c: some naming and type size fixups.
235 Fri Dec 2 13:08:58 2005 Tim Janik <timj@gtk.org>
237 * glib/gslice.h (g_slice_free): reworked GCC-specific type-safe macro
238 variant into something less verbose, but digestible for gcc-3.4.
240 Fri Dec 2 10:55:07 2005 Tim Janik <timj@imendio.com>
242 * tests/slice-test.c: extended to perform the benchmarking on the old
243 memchunk code if 'O' is selected.
245 * tests/memchunks.c: new file which contains the old GLib mem chunks
246 implementation with prefix old_mem_chunk_.
248 * tests/Makefile.am: added memchunks.c
250 Fri Dec 2 00:16:59 2005 Tim Janik <timj@imendio.com>
252 * glib/gslice.c: improved gettimeofday() timer resolution by maintaining
253 acceptable load distribution of the syscall.
254 (allocator_get_magazine_threshold): reduce minimum/single-thread magazine
255 sizes to approximately page_size/2.5 to avoid excessive startup allocations.
257 2005-12-01 Matthias Clasen <mclasen@redhat.com>
259 * tests/gobject/ifacecheck.c (test_iface_base_init): Don't
260 double-free base interface members.
262 * tests/Makefile.am: Remove duplicate variable.
264 Thu Dec 1 17:32:46 2005 Tim Janik <timj@imendio.com>
266 * glib/gslice.[hc]: new slice allocator implementation.
268 * tests/slice-test.c: added random slice allocation test.
270 * glib/gthread.[hc]: removed newly added private thread mem API.
272 * glib/gthreadinit.h:
275 * glib/gmem.c: divided glib threading initialisation into three phases,
276 initialisation where private keys and messaging are not available (only
277 needed by gmem.c), initialisation without messaging but private keys
278 available (gslice.c, gmessage.c), and full fledged initialisers that
279 server the rest of glib. initialisation functions got renamed to reflect
280 the limitations of their corresponding phases.
282 * glib/gmem.c: removed memchunk code, defer allocations to
285 * glib/gmem.[hc]: removed g_slice_* skeletons.
287 * glib/glib.symbols: added g_slice_* symbols.
289 * configure.in: check for availability of posix_memalign(3), memalign(3)
292 * glib/Makefile.am: added gslice.[hc].
294 2005-12-01 Tor Lillqvist <tml@novell.com>
296 * glib/gstdio.c (g_stat): In the Win32 implementation, strip
297 trailing slash(es) for non-root folders. stat() fails if non-root
298 folders are specified with trailing slashes. It's too much hassle
299 to demand that callers strip such slashes themselves, especially
300 as it is easy to get it wrong and strip the slash of a root
302 (g_rename): On NT-based Windows, use MoveFileEx() with
303 MOVEFILE_REPLACE_EXISTING to better match Unix behaviour.
305 2005-11-28 Matthias Clasen <mclasen@redhat.com>
307 Fix G_STMT_START / G_STMT_END on Solaris. (#321972,
310 * configure.in: Check whether do { } while (0) works.
312 * glib/gmacros.h: Use do { } while (0) for G_STMT_START /
313 G_STMT_END if it works.
315 2005-11-28 Matthias Clasen <mclasen@redhat.com>
317 * glib/gthread.c (g_static_rw_lock_wait, g_static_rw_lock_signal):
318 * glib/gnode.c (g_node_depth_traverse_level):
319 * glib/gmem.c (g_allocator_new):
320 * glib/ghash.c (g_hash_table_unref, g_hash_table_destroy)
321 (g_hash_table_foreach_remove_or_steal): Silence compiler
324 2005-11-27 Matthias Clasen <mclasen@redhat.com>
326 * glib/gunicollate.c (g_utf8_collate_key_for_filename):
327 Don't read beyond len. (#322520, Christian Persch)
329 Wed Nov 23 17:34:01 2005 Tim Janik <timj@imendio.com>
331 * glib/gdataset.c: access datalist flags via atomic pointer access
332 functions, instead of acquiring the dataset lock. this is faster and
333 also matches the atomic pointer readouts in gdatalistprivate.h and
334 direct pointer modifications required by gobject.c.
336 Wed Nov 23 13:35:31 2005 Tim Janik <timj@gtk.org>
338 * glib/gdataset.c: streamlined initialization code somewhat,
339 removed GData node cache.
341 2005-11-22 Matthias Clasen <mclasen@redhat.com>
343 * glib/ghash.c (g_hash_table_ref):
344 (g_hash_table_unref): Mark these as new API.
346 Avoid double locking in g_intern_string (#322133,
349 * glib/gdataset.c (g_quark_from_string_internal): New
350 internal function which factors out common parts of
351 g_quark_from[_static]_string.
352 (g_quark_from_string, g_quark_from_static_string):
353 Use g_quark_from_string_internal.
354 (g_intern_string, g_intern_static_string): Use
355 g_quark_from_string_internal, and only take the
357 (g_quark_new): Don't store the strings shifted by -1
358 in the g_quarks array.
359 (g_quark_to_string): Adapt to the previous change.
361 Tue Nov 22 14:04:26 2005 Tim Janik <timj@imendio.com>
365 g_hash_table_new_full(): create hash tables with a ref count of 1.
366 g_hash_table_ref(): atomically ref_count+=1
367 g_hash_table_unref(): atomically ref_count-=1, destroys hash table
368 when refcount reaches 0.
369 g_hash_table_destroy(): just destroy keys and values, unref by 1.
370 g_hash_table_insert():
371 g_hash_table_replace(): assert ref_count>0.
374 * glib/gatomic.c: added 'volatile' qualifier to all atomic pointer and
377 2005-11-20 Behdad Esfahbod <behdad@gnome.org>
379 * glib/guniprop.c (g_unichar_get_mirror_char): Remove unused
380 variables. (#321984, Andrew Paprocki)
382 2005-11-18 Matthias Clasen <mclasen@redhat.com>
384 * configure.in: Bump version
386 2005-11-17 Matthias Clasen <mclasen@redhat.com>
388 * === Released 2.9.0 ===
390 * glib/glib.symbols: Updates
392 * glib/gutf8.c (g_utf8_offset_to_pointer): Add some performance
393 hints to the docs. (Paolo Borelli)
397 * glib/gthread.c (g_thread_foreach): Mark as new api.
399 * README.in: Updates.
403 * glib/gdate.c (g_date_set_time_t):
404 (g_date_set_time_val): Add functions to set a date from
405 a time_t and from a GTimeVal.
406 (g_date_set_time): Deprecate. (#314067, Roger Leigh)
409 * tests/utf8-pointer.c: Unit tests for g_utf8_pointer_to_offset
410 and g_utf8_offset_to_pointer.
412 * glib/gutf8.c (g_utf8_pointer_to_offset)
413 (g_utf8_offset_to_pointer): Handle negative offsets, and use
414 "stutter stepping" for going backwards. (#320638, Larry
419 * glib/gthread.c: const correctness fixes, found
420 by Arjan van de Ven and gcc.
422 2005-11-16 Behdad Esfahbod <behdad@gnome.org>
424 * docs/reference/glib/tmpl/unicode.sgml: Correct typo on mentioning
425 @G_UNICODE_BREAK_UNKNOWN that should be #G_UNICODE_BREAK_UNKNOWN.
427 2005-11-09 Behdad Esfahbod <behdad@gnome.org>
429 * glib/guniprop.c: Use bit hacks instead when checking a general
430 category value against multiple values.
432 * glib/gutf8.c: Change ISO10646 to Unicode in docs.
434 2005-11-10 Simos Xenitellis <simos@gnome.org>
436 * configure.in: Added tt (Tatar) to ALL_LINGUAS.
438 2005-11-05 Tor Lillqvist <tml@novell.com>
440 * glib/giowin32.c (g_io_win32_set_flags): Don't call g_warning().
441 (#320688) Instead set error and return failure.
443 2005-11-07 Matthias Clasen <mclasen@redhat.com>
445 * glib/gmain.c (g_main_dispatch): Don't call cb_funcs->unref
446 while holding the context lock. (#320886, Andy Wingo)
448 2005-11-07 Billy Biggs <vektor@dumbterm.net>
450 * tests/refcount/Makefile.am: Add a missing reference to libglib
451 in the LDADD for this test.
453 2005-11-07 Matthias Clasen <mclasen@redhat.com>
455 * glib/gmem.c (g_allocator_new): We need to set n_preallocs to a
456 nonzero value, otherwise GTK+ 2.8 breaks when compiled against
457 GLib 2.9. (#320755, Luca Ferretti)
459 2005-11-04 Matthias Clasen <mclasen@redhat.com>
461 * glib/Makefile.am: Apply a patch from Behdad Esfahbod to
462 use a faster lookup table for g_unichar_get_mirror_char().
463 * glib/gmirroringtable.h: The new table.
464 * glib/gunichartables.h: Remove bidi_mirroring_table.
465 * glib/gen-unicode-tables.pl: Don't generate the mirroring
467 * glib/glib-mirroring-tab/*: Sources for the program
468 which generated gmirroringtable.h.
470 * glib/glist.c: Avoid some code duplication.
472 * glib/gscanner.h: Include gdataset.h. (#320322)
474 * glib/gdebug.h: add new GDebugFlag for fatal_criticals
475 * glib/gmessages.c: (_g_debug_init): handle G_DEBUG=fatal_criticals,
476 to help find critical warnings in applications. (#320017,
479 2005-11-02 Tor Lillqvist <tml@novell.com>
481 * glib/glib.symbols: Remove large amount of trailing whitespace
482 from one line. Remove #ifdef G_UNDEFINED from around
483 g_hash_table_new and g_hash_table_new_full symbols, what was the
484 purpose of that? Add the g_slice functions.
486 Tue Nov 1 16:24:20 2005 Tim Janik <timj@imendio.com>
488 * glib/gmem.[hc]: prepared deprecation of GMemChunk and GAllocator.
489 added g_slice_*() API to allocate and cache small bits of memory.
490 an actuall allocator implementation for g_slice_*() is still pending.
492 * glib/gthread.[hc]: changes from a patch by Matthias Clasen.
493 changed GRealThread list to use in-structure *next; fields instead
494 of GSList, in order for thread iteration to not depenend on g_slice_*()
496 _g_thread_mem_private_get():
497 _g_thread_mem_private_set(): added accessors for private memory,
498 needed because the ordinary GPrivate implementation relies on GArray
499 and GSList and therefore indirectly on working g_slice_*() allocations.
502 g_thread_foreach(): new public API function to loop over all existing threads.
516 removed GAllocator and free list usages and accompanying locks.
517 use g_slice_*() API to allocate and cache small bits of memory.
519 * glib/ghook.h: removed GMemChunk field from public API.
522 * glib/glist.h: deprecate allocator API, provide _free1() for consistency.
524 * glib/gnode.h: deprecate allocator API.
526 * glib/gmain.c: reordered GPollRec fields so g_slice_free_chain() can
527 be used for poll rec lists.
529 * glib/grel.c: removed mem chunk usage, and allocated tuples via g_slice_*().
530 g_relation_destroy(): free all tuples from the all_tuples hash table,
531 this effectively maintains the life time track keeping of tuples.
532 g_relation_delete_tuple(): free tuples which are removed from the
533 all_tuples hash table. this fixes a temporary leak that was present
534 in the memchunk code until the destruction of the relation.
536 2005-10-29 Matthias Clasen <mclasen@redhat.com>
538 * tests/convert-test.c: Add some tests for conversions between
539 UTF-8, UCS-4 and UTF-16.
541 * glib/gutf8.c (g_utf8_to_ucs4, g_utf8_to_utf16): Fix handling
542 of len == -1, noticed by Morten Welinder.
544 2005-10-27 Erdal Ronahi <erdal.ronahi@gmail.com>
546 * configure.in: Added ku (Kurdish) to ALL_LINGUAS
548 2005-10-26 Matthias Clasen <mclasen@redhat.com>
550 * glib/gutf8.c (g_ucs4_to_utf8): Don't set items_read twice
551 in the error case, and add some documentation. (#319806, Morten
554 2005-10-19 Manish Singh <yosh@gimp.org>
556 * configure.in: Use AC_CHECK_FUNCS for _NSGetEnviron, to get the
557 config.h symbol automatically. Fixes bug #313731.
559 2005-10-19 Tor Lillqvist <tml@novell.com>
561 * glib/gwin32.c (g_win32_get_package_installation_directory):
562 Return a g_strdup()ed copy of the value stored in the hash table,
563 so that it can be g_free()d without leaving a dangling pointer in
564 the hash table. (#319232)
566 2005-10-06 Matthias Clasen <mclasen@redhat.com>
568 * glib/gunicollate.c (g_utf8_collate_key_for_filename): Handle
569 all-zero sequences correctly. (#317930, Sebastien Bacher)
571 2005-10-05 Matthias Clasen <mclasen@redhat.com>
573 * glib/gmarkup.c (g_markup_escape_text): Doc addition
575 * Makefile.am: Add Collation tests.
577 * tests/collate/*: Inputs and expected outputs for collation tests.
579 * tests/run-collate-tests.sh: Script to run collation tests.
581 * tests/unicode-collate.c (main): Rework slightly to make
582 it usable in unit tests. Also test g_utf8_collate_key_for_filename().
584 2005-10-01 Behdad Esfahbod <behdad@gnome.org>
586 * docs/reference/glib/tmpl/unicode.sgml:
587 * glib/gen-unicode-tables.pl:
589 * glib/gunichartables.h:
591 * tests/casefold.txt:
592 * tests/casemap.txt: Updated to Unicode 4.1. There are five new
593 GUnicodeBreakType types. That may break some applications, like
596 2005-09-26 Matthias Clasen <mclasen@redhat.com>
598 * glib/gstrfuncs.c (g_ascii_strtoull): Add details to the
599 docs. (#314393, Matthew F. Barnes)
602 * glib/gprintf.h: Remove g_snprintf() and g_vsnprintf(), since
603 they are already declared in glib.h. This doesn't break documented
604 use of gprintf.h, but should probably be pointed out in the
605 release notes for 2.10. (#314232, Behdad Esfahbod)
607 Tue Sep 20 13:16:04 2005 Tim Janik <timj@imendio.com>
609 * glib/gpattern.c (g_pattern_ph_match): applied significant recursion
610 complexity optimization, based on a patch from Matthias Clasen.
612 * tests/patterntest.c: more tests, mostly from matthias.
614 2005-09-20 Matthias Clasen <mclasen@redhat.com>
616 * glib/gqueue.c (g_queue_insert_sorted): Correct the docs.
617 (#316703, Mark Drago)
619 Mon Sep 19 17:23:23 2005 Tim Janik <timj@imendio.com>
621 * glib/gpattern.c: applied a patch from matthias which checks on the
622 upper bound of GPatternSpec length to optimize matches.
625 * tests/patterntest.c: added more match cases.
627 2005-09-16 Tor Lillqvist <tml@novell.com>
629 * glib/gstrfuncs.c (g_ascii_strcasecmp, g_ascii_strncasecmp): Add
630 warning to doc comment that these functions should not be used on
631 encodings like CP932.
633 2005-09-14 Matthias Clasen <mclasen@redhat.com>
635 * tests/keyfile-test.c: Add a test for grup names of length 1.
637 * glib/gkeyfile.c (g_key_file_line_is_group): Accept group names
638 of length 1. (#316309)
640 2005-09-12 Matthias Clasen <mclasen@redhat.com>
642 * glib/gmarkup.c (g_markup_escape_text): Clarify docs.
644 2005-09-11 Sebastian Wilhelmi <seppi@seppi.de>
646 * tests/refcount/Makefile.am (INCLUDES): Link the the refcount
647 tests to the system thread library $(G_THREAD_LIBS). Fixes #313744
650 2005-09-11 Kjartan Maraas <kmaraas@gnome.org>
652 * glib/gmain.c: (g_child_watch_prepare), (g_child_watch_check),
653 (child_watch_helper_thread): Remove some dead code. Closes
656 2005-09-07 Tor Lillqvist <tml@novell.com>
658 * glib/Makefile.am: Create also a console version of the
659 gspawn-win32-helper program, gspawn-win32-helper-console.exe.
660 It's otherwise identical to gspawn-win32-helper.exe, except marked
661 as a console application (linked without the -mwindows option).
663 * glib/gspawn-win32.c (do_spawn_directly, do_spawn_with_pipes):
664 Drop the dont_wait parameter. Its truth value correlated 100% with
665 the NULLness of the exit_status parameter anyway, so it's enough
666 to check whether exit_status is NULL. Invert the sense of the
667 dont_return_handle parameter and rename it to do_return_handle, to
668 make the code easier to read by avoiding double negations.
670 (g_spawn_sync_utf8, g_spawn_async_with_pipes_utf8): Modify calls
671 to do_spawn_with_pipes() accordingly.
673 (do_spawn_with_pipes): If we have a console, use the console
674 version of the helper program, otherwise use the GUI one. This
675 avoids extra console windows opening up in some situations. (In
676 case a console application uses the GUI gspawn-win32-helper.exe to
677 spawn another console application we would get a separate console
678 for the spawned console application).
680 * glib-zip.in: Distribute also gspawn-win32-helper-console.exe.
682 2005-09-05 Matthias Clasen <mclasen@redhat.com>
684 * glib/gmappedfile.c (g_mapped_file_new): Report an error
685 if the file is too large. (#315275, Kjartan Maraas)
687 * glib/gkeyfile.c (g_key_file_load_from_fd): The return value
688 of read() is signed. (#315273, Kjartan Maraas)
690 2005-08-31 Tor Lillqvist <tml@novell.com>
692 * glib/gutils.h: Wrapping atexit() is a bad idea on Windows, where
693 the EXE and each DLL have their own atexit function chains.
695 #define g_atexit as atexit instead. This means it has a
696 better chance of doing what the caller wants. For instance,
697 gtkhtml calls g_atexit() registering a function in gtkhtml
698 itself. This caused a crash when g_atexit() was implemented as a
699 function in the GLib DLL. The gtkhtml DLL was already unloaded by
700 the time the GLib DLL got unloaded.
702 * glib/gutils.c: #undef the #define mentioned above, to also get a
703 real g_atexit() into the DLL for backward compatibility. Document
704 the Windows behaviour of g_atexit(), and document the varying ways
705 atexit() can behave in the context of dynamically loaded modules
708 2005-08-31 Matthias Clasen <mclasen@redhat.com>
712 * glib/gdataset.c: Add string interning functions.
714 2005-08-28 Matthias Clasen <mclasen@redhat.com>
716 * glib/giochannel.c: Unify some near-duplicate strings. (#314654,
719 2005-08-26 Matthias Clasen <mclasen@redhat.com>
721 * configure.in: Bump version to 2.9.0
723 2005-08-25 Tor Lillqvist <tml@novell.com>
725 Make also the g_spawn*() functions take parameters in the GLib
726 file name encoding, i.e. UTF-8, on Windows. Has no impact on Unix
727 API or ABI. Like the other GLib API that was earlier changed to
728 use UTF-8 on Windows, the names of the functions that take UTF-8
729 have _utf8 suffixes added by using preprocessor macros in the
730 header file. The old names are kept for functions with the old
731 behaviour, taking parameters in the system codepage, for DLL ABI
734 * glib/gspawn.h: On Win32 add the suffix _utf8 to the names of the
735 g_spawn*() functions.
737 * glib/gspawn-win32.c: Use wide-char API on NT-based
738 Windows. Convert parameters from UTF-8 to wide chars (NT) or
739 system codepage (Win9x) and call the C library _wspawn*() or
740 spawn*() functions respectvely. Add DLL ABI stability versions
741 that take parameters in the system codepage.
743 * glib/gspawn-win32-helper.c: On NT-based Windows use the
744 wide-char versions of argv and envp, and use wide-char API to
745 change directory and spawn the program to run. Remove the verbose
746 debugging output, it was too complex to modify for the wide-char
747 features. (Just add temporary debugging printouts if needed, no
748 need to have them permanently in the source.)
750 * glib/gspawn.c: Corresponding documentation updates.
752 * glib/glib.symbols: Corresponding changes: Mark the ABI stability
753 symbols as PRIVATE, add the new _utf8-suffixed ones.
755 2005-08-24 Stepan Kasal <kasal@ucw.cz>
757 * glib/gtypes.h (G_MININT64): Cast the constant to gint64; it is
758 guint64 otherwise and that can produce warnings about comparison
759 between signed and unsigned.
761 2005-08-23 Matthias Clasen <mclasen@redhat.com>
763 * glib/gutils.c: Fix the crt_externs.h include.
765 2005-08-23 Stepan Kasal <kasal@ucw.cz>
767 * NEWS: Fix spelling of my first name.
769 2005-08-23 Matthias Clasen <mclasen@redhat.com>
773 * === Released 2.8.1 ===
777 2005-08-20 Hans Breuer <hans@breuer.org>
779 * glib/makefile.msc.in : link with ws2_32.lib
781 2005-08-18 Tor Lillqvist <tml@novell.com>
783 * configure.in: Check for <sys/wait.h>
785 * glib/gbacktrace.c: Include <sys/wait.h> on if HAVE_SYS_WAIT_H.
787 2005-08-18 Ross Burton <ross@burtonini.com>
790 Optimise single-character insertions.
795 * tests/string-test.c:
796 Add tests for new optimisation, and fix a leak.
798 2005-08-17 Matthias Clasen <mclasen@redhat.com>
800 * configure.in: Check for crt_externs.h and _NSGetEnviron.
802 * glib/gutils.c: On Darwin, include crt-externs.h and
803 define environ using _NSGetEnviron(). (#313731)
805 2005-08-16 Stepan Kasal <kasal@ucw.cz>
807 * glib/gutils.c (g_get_any_init): Move the body of the big if...
808 (g_get_any_init_do): ... to this new function.
809 (g_get_any_init): Declare as inline.
810 (g_get_any_init_locked): New inline function, does the locking.
811 Make use of these two throughout the code.
813 2005-08-15 Matthias Clasen <mclasen@redhat.com>
815 * glib/gbacktrace.c (g_on_error_stack_trace): Wait for
816 the child process and then simply return. This makes
817 The "S" option work as documented in g_on_error_query().
818 (#313125, Matthew F. Barnes)
820 * glib/gunicode.h: Update the link to Unicode category
821 values. (#313369, Behnam Esfahbod)
823 * glib/gqueue.c (g_queue_find_custom): Clarify docs
824 a little. (#311727, Tristan van Berkom)
826 * glib/abicheck.sh, gobject/abicheck.sh: Make the
827 check work on ia64 too, where some symbols ended up
828 in yet another section.