1 Wed Nov 23 13:35:31 2005 Tim Janik <timj@gtk.org>
3 * glib/gdataset.c: streamlined initialization code somewhat,
4 removed GData node cache.
6 2005-11-22 Matthias Clasen <mclasen@redhat.com>
8 * glib/ghash.c (g_hash_table_ref):
9 (g_hash_table_unref): Mark these as new API.
11 Avoid double locking in g_intern_string (#322133,
14 * glib/gdataset.c (g_quark_from_string_internal): New
15 internal function which factors out common parts of
16 g_quark_from[_static]_string.
17 (g_quark_from_string, g_quark_from_static_string):
18 Use g_quark_from_string_internal.
19 (g_intern_string, g_intern_static_string): Use
20 g_quark_from_string_internal, and only take the
22 (g_quark_new): Don't store the strings shifted by -1
23 in the g_quarks array.
24 (g_quark_to_string): Adapt to the previous change.
26 Tue Nov 22 14:04:26 2005 Tim Janik <timj@imendio.com>
30 g_hash_table_new_full(): create hash tables with a ref count of 1.
31 g_hash_table_ref(): atomically ref_count+=1
32 g_hash_table_unref(): atomically ref_count-=1, destroys hash table
33 when refcount reaches 0.
34 g_hash_table_destroy(): just destroy keys and values, unref by 1.
35 g_hash_table_insert():
36 g_hash_table_replace(): assert ref_count>0.
39 * glib/gatomic.c: added 'volatile' qualifier to all atomic pointer and
42 2005-11-20 Behdad Esfahbod <behdad@gnome.org>
44 * glib/guniprop.c (g_unichar_get_mirror_char): Remove unused
45 variables. (#321984, Andrew Paprocki)
47 2005-11-18 Matthias Clasen <mclasen@redhat.com>
49 * configure.in: Bump version
51 2005-11-17 Matthias Clasen <mclasen@redhat.com>
53 * === Released 2.9.0 ===
55 * glib/glib.symbols: Updates
57 * glib/gutf8.c (g_utf8_offset_to_pointer): Add some performance
58 hints to the docs. (Paolo Borelli)
62 * glib/gthread.c (g_thread_foreach): Mark as new api.
68 * glib/gdate.c (g_date_set_time_t):
69 (g_date_set_time_val): Add functions to set a date from
70 a time_t and from a GTimeVal.
71 (g_date_set_time): Deprecate. (#314067, Roger Leigh)
74 * tests/utf8-pointer.c: Unit tests for g_utf8_pointer_to_offset
75 and g_utf8_offset_to_pointer.
77 * glib/gutf8.c (g_utf8_pointer_to_offset)
78 (g_utf8_offset_to_pointer): Handle negative offsets, and use
79 "stutter stepping" for going backwards. (#320638, Larry
84 * glib/gthread.c: const correctness fixes, found
85 by Arjan van de Ven and gcc.
87 2005-11-16 Behdad Esfahbod <behdad@gnome.org>
89 * docs/reference/glib/tmpl/unicode.sgml: Correct typo on mentioning
90 @G_UNICODE_BREAK_UNKNOWN that should be #G_UNICODE_BREAK_UNKNOWN.
92 2005-11-09 Behdad Esfahbod <behdad@gnome.org>
94 * glib/guniprop.c: Use bit hacks instead when checking a general
95 category value against multiple values.
97 * glib/gutf8.c: Change ISO10646 to Unicode in docs.
99 2005-11-10 Simos Xenitellis <simos@gnome.org>
101 * configure.in: Added tt (Tatar) to ALL_LINGUAS.
103 2005-11-05 Tor Lillqvist <tml@novell.com>
105 * glib/giowin32.c (g_io_win32_set_flags): Don't call g_warning().
106 (#320688) Instead set error and return failure.
108 2005-11-07 Matthias Clasen <mclasen@redhat.com>
110 * glib/gmain.c (g_main_dispatch): Don't call cb_funcs->unref
111 while holding the context lock. (#320886, Andy Wingo)
113 2005-11-07 Billy Biggs <vektor@dumbterm.net>
115 * tests/refcount/Makefile.am: Add a missing reference to libglib
116 in the LDADD for this test.
118 2005-11-07 Matthias Clasen <mclasen@redhat.com>
120 * glib/gmem.c (g_allocator_new): We need to set n_preallocs to a
121 nonzero value, otherwise GTK+ 2.8 breaks when compiled against
122 GLib 2.9. (#320755, Luca Ferretti)
124 2005-11-04 Matthias Clasen <mclasen@redhat.com>
126 * glib/Makefile.am: Apply a patch from Behdad Esfahbod to
127 use a faster lookup table for g_unichar_get_mirror_char().
128 * glib/gmirroringtable.h: The new table.
129 * glib/gunichartables.h: Remove bidi_mirroring_table.
130 * glib/gen-unicode-tables.pl: Don't generate the mirroring
132 * glib/glib-mirroring-tab/*: Sources for the program
133 which generated gmirroringtable.h.
135 * glib/glist.c: Avoid some code duplication.
137 * glib/gscanner.h: Include gdataset.h. (#320322)
139 * glib/gdebug.h: add new GDebugFlag for fatal_criticals
140 * glib/gmessages.c: (_g_debug_init): handle G_DEBUG=fatal_criticals,
141 to help find critical warnings in applications. (#320017,
144 2005-11-02 Tor Lillqvist <tml@novell.com>
146 * glib/glib.symbols: Remove large amount of trailing whitespace
147 from one line. Remove #ifdef G_UNDEFINED from around
148 g_hash_table_new and g_hash_table_new_full symbols, what was the
149 purpose of that? Add the g_slice functions.
151 Tue Nov 1 16:24:20 2005 Tim Janik <timj@imendio.com>
153 * glib/gmem.[hc]: prepared deprecation of GMemChunk and GAllocator.
154 added g_slice_*() API to allocate and cache small bits of memory.
155 an actuall allocator implementation for g_slice_*() is still pending.
157 * glib/gthread.[hc]: changes from a patch by Matthias Clasen.
158 changed GRealThread list to use in-structure *next; fields instead
159 of GSList, in order for thread iteration to not depenend on g_slice_*()
161 _g_thread_mem_private_get():
162 _g_thread_mem_private_set(): added accessors for private memory,
163 needed because the ordinary GPrivate implementation relies on GArray
164 and GSList and therefore indirectly on working g_slice_*() allocations.
167 g_thread_foreach(): new public API function to loop over all existing threads.
181 removed GAllocator and free list usages and accompanying locks.
182 use g_slice_*() API to allocate and cache small bits of memory.
184 * glib/ghook.h: removed GMemChunk field from public API.
187 * glib/glist.h: deprecate allocator API, provide _free1() for consistency.
189 * glib/gnode.h: deprecate allocator API.
191 * glib/gmain.c: reordered GPollRec fields so g_slice_free_chain() can
192 be used for poll rec lists.
194 * glib/grel.c: removed mem chunk usage, and allocated tuples via g_slice_*().
195 g_relation_destroy(): free all tuples from the all_tuples hash table,
196 this effectively maintains the life time track keeping of tuples.
197 g_relation_delete_tuple(): free tuples which are removed from the
198 all_tuples hash table. this fixes a temporary leak that was present
199 in the memchunk code until the destruction of the relation.
201 2005-10-29 Matthias Clasen <mclasen@redhat.com>
203 * tests/convert-test.c: Add some tests for conversions between
204 UTF-8, UCS-4 and UTF-16.
206 * glib/gutf8.c (g_utf8_to_ucs4, g_utf8_to_utf16): Fix handling
207 of len == -1, noticed by Morten Welinder.
209 2005-10-27 Erdal Ronahi <erdal.ronahi@gmail.com>
211 * configure.in: Added ku (Kurdish) to ALL_LINGUAS
213 2005-10-26 Matthias Clasen <mclasen@redhat.com>
215 * glib/gutf8.c (g_ucs4_to_utf8): Don't set items_read twice
216 in the error case, and add some documentation. (#319806, Morten
219 2005-10-19 Manish Singh <yosh@gimp.org>
221 * configure.in: Use AC_CHECK_FUNCS for _NSGetEnviron, to get the
222 config.h symbol automatically. Fixes bug #313731.
224 2005-10-19 Tor Lillqvist <tml@novell.com>
226 * glib/gwin32.c (g_win32_get_package_installation_directory):
227 Return a g_strdup()ed copy of the value stored in the hash table,
228 so that it can be g_free()d without leaving a dangling pointer in
229 the hash table. (#319232)
231 2005-10-06 Matthias Clasen <mclasen@redhat.com>
233 * glib/gunicollate.c (g_utf8_collate_key_for_filename): Handle
234 all-zero sequences correctly. (#317930, Sebastien Bacher)
236 2005-10-05 Matthias Clasen <mclasen@redhat.com>
238 * glib/gmarkup.c (g_markup_escape_text): Doc addition
240 * Makefile.am: Add Collation tests.
242 * tests/collate/*: Inputs and expected outputs for collation tests.
244 * tests/run-collate-tests.sh: Script to run collation tests.
246 * tests/unicode-collate.c (main): Rework slightly to make
247 it usable in unit tests. Also test g_utf8_collate_key_for_filename().
249 2005-10-01 Behdad Esfahbod <behdad@gnome.org>
251 * docs/reference/glib/tmpl/unicode.sgml:
252 * glib/gen-unicode-tables.pl:
254 * glib/gunichartables.h:
256 * tests/casefold.txt:
257 * tests/casemap.txt: Updated to Unicode 4.1. There are five new
258 GUnicodeBreakType types. That may break some applications, like
261 2005-09-26 Matthias Clasen <mclasen@redhat.com>
263 * glib/gstrfuncs.c (g_ascii_strtoull): Add details to the
264 docs. (#314393, Matthew F. Barnes)
267 * glib/gprintf.h: Remove g_snprintf() and g_vsnprintf(), since
268 they are already declared in glib.h. This doesn't break documented
269 use of gprintf.h, but should probably be pointed out in the
270 release notes for 2.10. (#314232, Behdad Esfahbod)
272 Tue Sep 20 13:16:04 2005 Tim Janik <timj@imendio.com>
274 * glib/gpattern.c (g_pattern_ph_match): applied significant recursion
275 complexity optimization, based on a patch from Matthias Clasen.
277 * tests/patterntest.c: more tests, mostly from matthias.
279 2005-09-20 Matthias Clasen <mclasen@redhat.com>
281 * glib/gqueue.c (g_queue_insert_sorted): Correct the docs.
282 (#316703, Mark Drago)
284 Mon Sep 19 17:23:23 2005 Tim Janik <timj@imendio.com>
286 * glib/gpattern.c: applied a patch from matthias which checks on the
287 upper bound of GPatternSpec length to optimize matches.
290 * tests/patterntest.c: added more match cases.
292 2005-09-16 Tor Lillqvist <tml@novell.com>
294 * glib/gstrfuncs.c (g_ascii_strcasecmp, g_ascii_strncasecmp): Add
295 warning to doc comment that these functions should not be used on
296 encodings like CP932.
298 2005-09-14 Matthias Clasen <mclasen@redhat.com>
300 * tests/keyfile-test.c: Add a test for grup names of length 1.
302 * glib/gkeyfile.c (g_key_file_line_is_group): Accept group names
303 of length 1. (#316309)
305 2005-09-12 Matthias Clasen <mclasen@redhat.com>
307 * glib/gmarkup.c (g_markup_escape_text): Clarify docs.
309 2005-09-11 Sebastian Wilhelmi <seppi@seppi.de>
311 * tests/refcount/Makefile.am (INCLUDES): Link the the refcount
312 tests to the system thread library $(G_THREAD_LIBS). Fixes #313744
315 2005-09-11 Kjartan Maraas <kmaraas@gnome.org>
317 * glib/gmain.c: (g_child_watch_prepare), (g_child_watch_check),
318 (child_watch_helper_thread): Remove some dead code. Closes
321 2005-09-07 Tor Lillqvist <tml@novell.com>
323 * glib/Makefile.am: Create also a console version of the
324 gspawn-win32-helper program, gspawn-win32-helper-console.exe.
325 It's otherwise identical to gspawn-win32-helper.exe, except marked
326 as a console application (linked without the -mwindows option).
328 * glib/gspawn-win32.c (do_spawn_directly, do_spawn_with_pipes):
329 Drop the dont_wait parameter. Its truth value correlated 100% with
330 the NULLness of the exit_status parameter anyway, so it's enough
331 to check whether exit_status is NULL. Invert the sense of the
332 dont_return_handle parameter and rename it to do_return_handle, to
333 make the code easier to read by avoiding double negations.
335 (g_spawn_sync_utf8, g_spawn_async_with_pipes_utf8): Modify calls
336 to do_spawn_with_pipes() accordingly.
338 (do_spawn_with_pipes): If we have a console, use the console
339 version of the helper program, otherwise use the GUI one. This
340 avoids extra console windows opening up in some situations. (In
341 case a console application uses the GUI gspawn-win32-helper.exe to
342 spawn another console application we would get a separate console
343 for the spawned console application).
345 * glib-zip.in: Distribute also gspawn-win32-helper-console.exe.
347 2005-09-05 Matthias Clasen <mclasen@redhat.com>
349 * glib/gmappedfile.c (g_mapped_file_new): Report an error
350 if the file is too large. (#315275, Kjartan Maraas)
352 * glib/gkeyfile.c (g_key_file_load_from_fd): The return value
353 of read() is signed. (#315273, Kjartan Maraas)
355 2005-08-31 Tor Lillqvist <tml@novell.com>
357 * glib/gutils.h: Wrapping atexit() is a bad idea on Windows, where
358 the EXE and each DLL have their own atexit function chains.
360 #define g_atexit as atexit instead. This means it has a
361 better chance of doing what the caller wants. For instance,
362 gtkhtml calls g_atexit() registering a function in gtkhtml
363 itself. This caused a crash when g_atexit() was implemented as a
364 function in the GLib DLL. The gtkhtml DLL was already unloaded by
365 the time the GLib DLL got unloaded.
367 * glib/gutils.c: #undef the #define mentioned above, to also get a
368 real g_atexit() into the DLL for backward compatibility. Document
369 the Windows behaviour of g_atexit(), and document the varying ways
370 atexit() can behave in the context of dynamically loaded modules
373 2005-08-31 Matthias Clasen <mclasen@redhat.com>
377 * glib/gdataset.c: Add string interning functions.
379 2005-08-28 Matthias Clasen <mclasen@redhat.com>
381 * glib/giochannel.c: Unify some near-duplicate strings. (#314654,
384 2005-08-26 Matthias Clasen <mclasen@redhat.com>
386 * configure.in: Bump version to 2.9.0
388 2005-08-25 Tor Lillqvist <tml@novell.com>
390 Make also the g_spawn*() functions take parameters in the GLib
391 file name encoding, i.e. UTF-8, on Windows. Has no impact on Unix
392 API or ABI. Like the other GLib API that was earlier changed to
393 use UTF-8 on Windows, the names of the functions that take UTF-8
394 have _utf8 suffixes added by using preprocessor macros in the
395 header file. The old names are kept for functions with the old
396 behaviour, taking parameters in the system codepage, for DLL ABI
399 * glib/gspawn.h: On Win32 add the suffix _utf8 to the names of the
400 g_spawn*() functions.
402 * glib/gspawn-win32.c: Use wide-char API on NT-based
403 Windows. Convert parameters from UTF-8 to wide chars (NT) or
404 system codepage (Win9x) and call the C library _wspawn*() or
405 spawn*() functions respectvely. Add DLL ABI stability versions
406 that take parameters in the system codepage.
408 * glib/gspawn-win32-helper.c: On NT-based Windows use the
409 wide-char versions of argv and envp, and use wide-char API to
410 change directory and spawn the program to run. Remove the verbose
411 debugging output, it was too complex to modify for the wide-char
412 features. (Just add temporary debugging printouts if needed, no
413 need to have them permanently in the source.)
415 * glib/gspawn.c: Corresponding documentation updates.
417 * glib/glib.symbols: Corresponding changes: Mark the ABI stability
418 symbols as PRIVATE, add the new _utf8-suffixed ones.
420 2005-08-24 Stepan Kasal <kasal@ucw.cz>
422 * glib/gtypes.h (G_MININT64): Cast the constant to gint64; it is
423 guint64 otherwise and that can produce warnings about comparison
424 between signed and unsigned.
426 2005-08-23 Matthias Clasen <mclasen@redhat.com>
428 * glib/gutils.c: Fix the crt_externs.h include.
430 2005-08-23 Stepan Kasal <kasal@ucw.cz>
432 * NEWS: Fix spelling of my first name.
434 2005-08-23 Matthias Clasen <mclasen@redhat.com>
438 * === Released 2.8.1 ===
442 2005-08-20 Hans Breuer <hans@breuer.org>
444 * glib/makefile.msc.in : link with ws2_32.lib
446 2005-08-18 Tor Lillqvist <tml@novell.com>
448 * configure.in: Check for <sys/wait.h>
450 * glib/gbacktrace.c: Include <sys/wait.h> on if HAVE_SYS_WAIT_H.
452 2005-08-18 Ross Burton <ross@burtonini.com>
455 Optimise single-character insertions.
460 * tests/string-test.c:
461 Add tests for new optimisation, and fix a leak.
463 2005-08-17 Matthias Clasen <mclasen@redhat.com>
465 * configure.in: Check for crt_externs.h and _NSGetEnviron.
467 * glib/gutils.c: On Darwin, include crt-externs.h and
468 define environ using _NSGetEnviron(). (#313731)
470 2005-08-16 Stepan Kasal <kasal@ucw.cz>
472 * glib/gutils.c (g_get_any_init): Move the body of the big if...
473 (g_get_any_init_do): ... to this new function.
474 (g_get_any_init): Declare as inline.
475 (g_get_any_init_locked): New inline function, does the locking.
476 Make use of these two throughout the code.
478 2005-08-15 Matthias Clasen <mclasen@redhat.com>
480 * glib/gbacktrace.c (g_on_error_stack_trace): Wait for
481 the child process and then simply return. This makes
482 The "S" option work as documented in g_on_error_query().
483 (#313125, Matthew F. Barnes)
485 * glib/gunicode.h: Update the link to Unicode category
486 values. (#313369, Behnam Esfahbod)
488 * glib/gqueue.c (g_queue_find_custom): Clarify docs
489 a little. (#311727, Tristan van Berkom)
491 * glib/abicheck.sh, gobject/abicheck.sh: Make the
492 check work on ia64 too, where some symbols ended up
493 in yet another section.