1 2005-11-27 Matthias Clasen <mclasen@redhat.com>
3 * glib/gunicollate.c (g_utf8_collate_key_for_filename):
4 Don't read beyond len. (#322520, Christian Persch)
6 Wed Nov 23 17:34:01 2005 Tim Janik <timj@imendio.com>
8 * glib/gdataset.c: access datalist flags via atomic pointer access
9 funcitons, instead of acquiring the dataset lock. this is faster and
10 also matches the atomic pointer readouts in gdatalistprivate.h and
11 direct pointer modifications required by gobject.c.
13 Wed Nov 23 13:35:31 2005 Tim Janik <timj@gtk.org>
15 * glib/gdataset.c: streamlined initialization code somewhat,
16 removed GData node cache.
18 2005-11-22 Matthias Clasen <mclasen@redhat.com>
20 * glib/ghash.c (g_hash_table_ref):
21 (g_hash_table_unref): Mark these as new API.
23 Avoid double locking in g_intern_string (#322133,
26 * glib/gdataset.c (g_quark_from_string_internal): New
27 internal function which factors out common parts of
28 g_quark_from[_static]_string.
29 (g_quark_from_string, g_quark_from_static_string):
30 Use g_quark_from_string_internal.
31 (g_intern_string, g_intern_static_string): Use
32 g_quark_from_string_internal, and only take the
34 (g_quark_new): Don't store the strings shifted by -1
35 in the g_quarks array.
36 (g_quark_to_string): Adapt to the previous change.
38 Tue Nov 22 14:04:26 2005 Tim Janik <timj@imendio.com>
42 g_hash_table_new_full(): create hash tables with a ref count of 1.
43 g_hash_table_ref(): atomically ref_count+=1
44 g_hash_table_unref(): atomically ref_count-=1, destroys hash table
45 when refcount reaches 0.
46 g_hash_table_destroy(): just destroy keys and values, unref by 1.
47 g_hash_table_insert():
48 g_hash_table_replace(): assert ref_count>0.
51 * glib/gatomic.c: added 'volatile' qualifier to all atomic pointer and
54 2005-11-20 Behdad Esfahbod <behdad@gnome.org>
56 * glib/guniprop.c (g_unichar_get_mirror_char): Remove unused
57 variables. (#321984, Andrew Paprocki)
59 2005-11-18 Matthias Clasen <mclasen@redhat.com>
61 * configure.in: Bump version
63 2005-11-17 Matthias Clasen <mclasen@redhat.com>
65 * === Released 2.9.0 ===
67 * glib/glib.symbols: Updates
69 * glib/gutf8.c (g_utf8_offset_to_pointer): Add some performance
70 hints to the docs. (Paolo Borelli)
74 * glib/gthread.c (g_thread_foreach): Mark as new api.
80 * glib/gdate.c (g_date_set_time_t):
81 (g_date_set_time_val): Add functions to set a date from
82 a time_t and from a GTimeVal.
83 (g_date_set_time): Deprecate. (#314067, Roger Leigh)
86 * tests/utf8-pointer.c: Unit tests for g_utf8_pointer_to_offset
87 and g_utf8_offset_to_pointer.
89 * glib/gutf8.c (g_utf8_pointer_to_offset)
90 (g_utf8_offset_to_pointer): Handle negative offsets, and use
91 "stutter stepping" for going backwards. (#320638, Larry
96 * glib/gthread.c: const correctness fixes, found
97 by Arjan van de Ven and gcc.
99 2005-11-16 Behdad Esfahbod <behdad@gnome.org>
101 * docs/reference/glib/tmpl/unicode.sgml: Correct typo on mentioning
102 @G_UNICODE_BREAK_UNKNOWN that should be #G_UNICODE_BREAK_UNKNOWN.
104 2005-11-09 Behdad Esfahbod <behdad@gnome.org>
106 * glib/guniprop.c: Use bit hacks instead when checking a general
107 category value against multiple values.
109 * glib/gutf8.c: Change ISO10646 to Unicode in docs.
111 2005-11-10 Simos Xenitellis <simos@gnome.org>
113 * configure.in: Added tt (Tatar) to ALL_LINGUAS.
115 2005-11-05 Tor Lillqvist <tml@novell.com>
117 * glib/giowin32.c (g_io_win32_set_flags): Don't call g_warning().
118 (#320688) Instead set error and return failure.
120 2005-11-07 Matthias Clasen <mclasen@redhat.com>
122 * glib/gmain.c (g_main_dispatch): Don't call cb_funcs->unref
123 while holding the context lock. (#320886, Andy Wingo)
125 2005-11-07 Billy Biggs <vektor@dumbterm.net>
127 * tests/refcount/Makefile.am: Add a missing reference to libglib
128 in the LDADD for this test.
130 2005-11-07 Matthias Clasen <mclasen@redhat.com>
132 * glib/gmem.c (g_allocator_new): We need to set n_preallocs to a
133 nonzero value, otherwise GTK+ 2.8 breaks when compiled against
134 GLib 2.9. (#320755, Luca Ferretti)
136 2005-11-04 Matthias Clasen <mclasen@redhat.com>
138 * glib/Makefile.am: Apply a patch from Behdad Esfahbod to
139 use a faster lookup table for g_unichar_get_mirror_char().
140 * glib/gmirroringtable.h: The new table.
141 * glib/gunichartables.h: Remove bidi_mirroring_table.
142 * glib/gen-unicode-tables.pl: Don't generate the mirroring
144 * glib/glib-mirroring-tab/*: Sources for the program
145 which generated gmirroringtable.h.
147 * glib/glist.c: Avoid some code duplication.
149 * glib/gscanner.h: Include gdataset.h. (#320322)
151 * glib/gdebug.h: add new GDebugFlag for fatal_criticals
152 * glib/gmessages.c: (_g_debug_init): handle G_DEBUG=fatal_criticals,
153 to help find critical warnings in applications. (#320017,
156 2005-11-02 Tor Lillqvist <tml@novell.com>
158 * glib/glib.symbols: Remove large amount of trailing whitespace
159 from one line. Remove #ifdef G_UNDEFINED from around
160 g_hash_table_new and g_hash_table_new_full symbols, what was the
161 purpose of that? Add the g_slice functions.
163 Tue Nov 1 16:24:20 2005 Tim Janik <timj@imendio.com>
165 * glib/gmem.[hc]: prepared deprecation of GMemChunk and GAllocator.
166 added g_slice_*() API to allocate and cache small bits of memory.
167 an actuall allocator implementation for g_slice_*() is still pending.
169 * glib/gthread.[hc]: changes from a patch by Matthias Clasen.
170 changed GRealThread list to use in-structure *next; fields instead
171 of GSList, in order for thread iteration to not depenend on g_slice_*()
173 _g_thread_mem_private_get():
174 _g_thread_mem_private_set(): added accessors for private memory,
175 needed because the ordinary GPrivate implementation relies on GArray
176 and GSList and therefore indirectly on working g_slice_*() allocations.
179 g_thread_foreach(): new public API function to loop over all existing threads.
193 removed GAllocator and free list usages and accompanying locks.
194 use g_slice_*() API to allocate and cache small bits of memory.
196 * glib/ghook.h: removed GMemChunk field from public API.
199 * glib/glist.h: deprecate allocator API, provide _free1() for consistency.
201 * glib/gnode.h: deprecate allocator API.
203 * glib/gmain.c: reordered GPollRec fields so g_slice_free_chain() can
204 be used for poll rec lists.
206 * glib/grel.c: removed mem chunk usage, and allocated tuples via g_slice_*().
207 g_relation_destroy(): free all tuples from the all_tuples hash table,
208 this effectively maintains the life time track keeping of tuples.
209 g_relation_delete_tuple(): free tuples which are removed from the
210 all_tuples hash table. this fixes a temporary leak that was present
211 in the memchunk code until the destruction of the relation.
213 2005-10-29 Matthias Clasen <mclasen@redhat.com>
215 * tests/convert-test.c: Add some tests for conversions between
216 UTF-8, UCS-4 and UTF-16.
218 * glib/gutf8.c (g_utf8_to_ucs4, g_utf8_to_utf16): Fix handling
219 of len == -1, noticed by Morten Welinder.
221 2005-10-27 Erdal Ronahi <erdal.ronahi@gmail.com>
223 * configure.in: Added ku (Kurdish) to ALL_LINGUAS
225 2005-10-26 Matthias Clasen <mclasen@redhat.com>
227 * glib/gutf8.c (g_ucs4_to_utf8): Don't set items_read twice
228 in the error case, and add some documentation. (#319806, Morten
231 2005-10-19 Manish Singh <yosh@gimp.org>
233 * configure.in: Use AC_CHECK_FUNCS for _NSGetEnviron, to get the
234 config.h symbol automatically. Fixes bug #313731.
236 2005-10-19 Tor Lillqvist <tml@novell.com>
238 * glib/gwin32.c (g_win32_get_package_installation_directory):
239 Return a g_strdup()ed copy of the value stored in the hash table,
240 so that it can be g_free()d without leaving a dangling pointer in
241 the hash table. (#319232)
243 2005-10-06 Matthias Clasen <mclasen@redhat.com>
245 * glib/gunicollate.c (g_utf8_collate_key_for_filename): Handle
246 all-zero sequences correctly. (#317930, Sebastien Bacher)
248 2005-10-05 Matthias Clasen <mclasen@redhat.com>
250 * glib/gmarkup.c (g_markup_escape_text): Doc addition
252 * Makefile.am: Add Collation tests.
254 * tests/collate/*: Inputs and expected outputs for collation tests.
256 * tests/run-collate-tests.sh: Script to run collation tests.
258 * tests/unicode-collate.c (main): Rework slightly to make
259 it usable in unit tests. Also test g_utf8_collate_key_for_filename().
261 2005-10-01 Behdad Esfahbod <behdad@gnome.org>
263 * docs/reference/glib/tmpl/unicode.sgml:
264 * glib/gen-unicode-tables.pl:
266 * glib/gunichartables.h:
268 * tests/casefold.txt:
269 * tests/casemap.txt: Updated to Unicode 4.1. There are five new
270 GUnicodeBreakType types. That may break some applications, like
273 2005-09-26 Matthias Clasen <mclasen@redhat.com>
275 * glib/gstrfuncs.c (g_ascii_strtoull): Add details to the
276 docs. (#314393, Matthew F. Barnes)
279 * glib/gprintf.h: Remove g_snprintf() and g_vsnprintf(), since
280 they are already declared in glib.h. This doesn't break documented
281 use of gprintf.h, but should probably be pointed out in the
282 release notes for 2.10. (#314232, Behdad Esfahbod)
284 Tue Sep 20 13:16:04 2005 Tim Janik <timj@imendio.com>
286 * glib/gpattern.c (g_pattern_ph_match): applied significant recursion
287 complexity optimization, based on a patch from Matthias Clasen.
289 * tests/patterntest.c: more tests, mostly from matthias.
291 2005-09-20 Matthias Clasen <mclasen@redhat.com>
293 * glib/gqueue.c (g_queue_insert_sorted): Correct the docs.
294 (#316703, Mark Drago)
296 Mon Sep 19 17:23:23 2005 Tim Janik <timj@imendio.com>
298 * glib/gpattern.c: applied a patch from matthias which checks on the
299 upper bound of GPatternSpec length to optimize matches.
302 * tests/patterntest.c: added more match cases.
304 2005-09-16 Tor Lillqvist <tml@novell.com>
306 * glib/gstrfuncs.c (g_ascii_strcasecmp, g_ascii_strncasecmp): Add
307 warning to doc comment that these functions should not be used on
308 encodings like CP932.
310 2005-09-14 Matthias Clasen <mclasen@redhat.com>
312 * tests/keyfile-test.c: Add a test for grup names of length 1.
314 * glib/gkeyfile.c (g_key_file_line_is_group): Accept group names
315 of length 1. (#316309)
317 2005-09-12 Matthias Clasen <mclasen@redhat.com>
319 * glib/gmarkup.c (g_markup_escape_text): Clarify docs.
321 2005-09-11 Sebastian Wilhelmi <seppi@seppi.de>
323 * tests/refcount/Makefile.am (INCLUDES): Link the the refcount
324 tests to the system thread library $(G_THREAD_LIBS). Fixes #313744
327 2005-09-11 Kjartan Maraas <kmaraas@gnome.org>
329 * glib/gmain.c: (g_child_watch_prepare), (g_child_watch_check),
330 (child_watch_helper_thread): Remove some dead code. Closes
333 2005-09-07 Tor Lillqvist <tml@novell.com>
335 * glib/Makefile.am: Create also a console version of the
336 gspawn-win32-helper program, gspawn-win32-helper-console.exe.
337 It's otherwise identical to gspawn-win32-helper.exe, except marked
338 as a console application (linked without the -mwindows option).
340 * glib/gspawn-win32.c (do_spawn_directly, do_spawn_with_pipes):
341 Drop the dont_wait parameter. Its truth value correlated 100% with
342 the NULLness of the exit_status parameter anyway, so it's enough
343 to check whether exit_status is NULL. Invert the sense of the
344 dont_return_handle parameter and rename it to do_return_handle, to
345 make the code easier to read by avoiding double negations.
347 (g_spawn_sync_utf8, g_spawn_async_with_pipes_utf8): Modify calls
348 to do_spawn_with_pipes() accordingly.
350 (do_spawn_with_pipes): If we have a console, use the console
351 version of the helper program, otherwise use the GUI one. This
352 avoids extra console windows opening up in some situations. (In
353 case a console application uses the GUI gspawn-win32-helper.exe to
354 spawn another console application we would get a separate console
355 for the spawned console application).
357 * glib-zip.in: Distribute also gspawn-win32-helper-console.exe.
359 2005-09-05 Matthias Clasen <mclasen@redhat.com>
361 * glib/gmappedfile.c (g_mapped_file_new): Report an error
362 if the file is too large. (#315275, Kjartan Maraas)
364 * glib/gkeyfile.c (g_key_file_load_from_fd): The return value
365 of read() is signed. (#315273, Kjartan Maraas)
367 2005-08-31 Tor Lillqvist <tml@novell.com>
369 * glib/gutils.h: Wrapping atexit() is a bad idea on Windows, where
370 the EXE and each DLL have their own atexit function chains.
372 #define g_atexit as atexit instead. This means it has a
373 better chance of doing what the caller wants. For instance,
374 gtkhtml calls g_atexit() registering a function in gtkhtml
375 itself. This caused a crash when g_atexit() was implemented as a
376 function in the GLib DLL. The gtkhtml DLL was already unloaded by
377 the time the GLib DLL got unloaded.
379 * glib/gutils.c: #undef the #define mentioned above, to also get a
380 real g_atexit() into the DLL for backward compatibility. Document
381 the Windows behaviour of g_atexit(), and document the varying ways
382 atexit() can behave in the context of dynamically loaded modules
385 2005-08-31 Matthias Clasen <mclasen@redhat.com>
389 * glib/gdataset.c: Add string interning functions.
391 2005-08-28 Matthias Clasen <mclasen@redhat.com>
393 * glib/giochannel.c: Unify some near-duplicate strings. (#314654,
396 2005-08-26 Matthias Clasen <mclasen@redhat.com>
398 * configure.in: Bump version to 2.9.0
400 2005-08-25 Tor Lillqvist <tml@novell.com>
402 Make also the g_spawn*() functions take parameters in the GLib
403 file name encoding, i.e. UTF-8, on Windows. Has no impact on Unix
404 API or ABI. Like the other GLib API that was earlier changed to
405 use UTF-8 on Windows, the names of the functions that take UTF-8
406 have _utf8 suffixes added by using preprocessor macros in the
407 header file. The old names are kept for functions with the old
408 behaviour, taking parameters in the system codepage, for DLL ABI
411 * glib/gspawn.h: On Win32 add the suffix _utf8 to the names of the
412 g_spawn*() functions.
414 * glib/gspawn-win32.c: Use wide-char API on NT-based
415 Windows. Convert parameters from UTF-8 to wide chars (NT) or
416 system codepage (Win9x) and call the C library _wspawn*() or
417 spawn*() functions respectvely. Add DLL ABI stability versions
418 that take parameters in the system codepage.
420 * glib/gspawn-win32-helper.c: On NT-based Windows use the
421 wide-char versions of argv and envp, and use wide-char API to
422 change directory and spawn the program to run. Remove the verbose
423 debugging output, it was too complex to modify for the wide-char
424 features. (Just add temporary debugging printouts if needed, no
425 need to have them permanently in the source.)
427 * glib/gspawn.c: Corresponding documentation updates.
429 * glib/glib.symbols: Corresponding changes: Mark the ABI stability
430 symbols as PRIVATE, add the new _utf8-suffixed ones.
432 2005-08-24 Stepan Kasal <kasal@ucw.cz>
434 * glib/gtypes.h (G_MININT64): Cast the constant to gint64; it is
435 guint64 otherwise and that can produce warnings about comparison
436 between signed and unsigned.
438 2005-08-23 Matthias Clasen <mclasen@redhat.com>
440 * glib/gutils.c: Fix the crt_externs.h include.
442 2005-08-23 Stepan Kasal <kasal@ucw.cz>
444 * NEWS: Fix spelling of my first name.
446 2005-08-23 Matthias Clasen <mclasen@redhat.com>
450 * === Released 2.8.1 ===
454 2005-08-20 Hans Breuer <hans@breuer.org>
456 * glib/makefile.msc.in : link with ws2_32.lib
458 2005-08-18 Tor Lillqvist <tml@novell.com>
460 * configure.in: Check for <sys/wait.h>
462 * glib/gbacktrace.c: Include <sys/wait.h> on if HAVE_SYS_WAIT_H.
464 2005-08-18 Ross Burton <ross@burtonini.com>
467 Optimise single-character insertions.
472 * tests/string-test.c:
473 Add tests for new optimisation, and fix a leak.
475 2005-08-17 Matthias Clasen <mclasen@redhat.com>
477 * configure.in: Check for crt_externs.h and _NSGetEnviron.
479 * glib/gutils.c: On Darwin, include crt-externs.h and
480 define environ using _NSGetEnviron(). (#313731)
482 2005-08-16 Stepan Kasal <kasal@ucw.cz>
484 * glib/gutils.c (g_get_any_init): Move the body of the big if...
485 (g_get_any_init_do): ... to this new function.
486 (g_get_any_init): Declare as inline.
487 (g_get_any_init_locked): New inline function, does the locking.
488 Make use of these two throughout the code.
490 2005-08-15 Matthias Clasen <mclasen@redhat.com>
492 * glib/gbacktrace.c (g_on_error_stack_trace): Wait for
493 the child process and then simply return. This makes
494 The "S" option work as documented in g_on_error_query().
495 (#313125, Matthew F. Barnes)
497 * glib/gunicode.h: Update the link to Unicode category
498 values. (#313369, Behnam Esfahbod)
500 * glib/gqueue.c (g_queue_find_custom): Clarify docs
501 a little. (#311727, Tristan van Berkom)
503 * glib/abicheck.sh, gobject/abicheck.sh: Make the
504 check work on ia64 too, where some symbols ended up
505 in yet another section.