1 2005-11-17 Matthias Clasen <mclasen@redhat.com>
4 * tests/utf8-pointer.c: Unit tests for g_utf8_pointer_to_offset
5 and g_utf8_offset_to_pointer.
7 * glib/gutf8.c (g_utf8_pointer_to_offset)
8 (g_utf8_offset_to_pointer): Handle negative offsets, and use
9 "stutter stepping" for going backwards. (#320638, Larry
14 * glib/gthread.c: const correctness fixes, found
15 by Arjan van de Ven and gcc.
17 2005-11-16 Behdad Esfahbod <behdad@gnome.org>
19 * docs/reference/glib/tmpl/unicode.sgml: Correct typo on mentioning
20 @G_UNICODE_BREAK_UNKNOWN that should be #G_UNICODE_BREAK_UNKNOWN.
22 2005-11-09 Behdad Esfahbod <behdad@gnome.org>
24 * glib/guniprop.c: Use bit hacks instead when checking a general
25 category value against multiple values.
27 * glib/gutf8.c: Change ISO10646 to Unicode in docs.
29 2005-11-10 Simos Xenitellis <simos@gnome.org>
31 * configure.in: Added tt (Tatar) to ALL_LINGUAS.
33 2005-11-05 Tor Lillqvist <tml@novell.com>
35 * glib/giowin32.c (g_io_win32_set_flags): Don't call g_warning().
36 (#320688) Instead set error and return failure.
38 2005-11-07 Matthias Clasen <mclasen@redhat.com>
40 * glib/gmain.c (g_main_dispatch): Don't call cb_funcs->unref
41 while holding the context lock. (#320886, Andy Wingo)
43 2005-11-07 Billy Biggs <vektor@dumbterm.net>
45 * tests/refcount/Makefile.am: Add a missing reference to libglib
46 in the LDADD for this test.
48 2005-11-07 Matthias Clasen <mclasen@redhat.com>
50 * glib/gmem.c (g_allocator_new): We need to set n_preallocs to a
51 nonzero value, otherwise GTK+ 2.8 breaks when compiled against
52 GLib 2.9. (#320755, Luca Ferretti)
54 2005-11-04 Matthias Clasen <mclasen@redhat.com>
56 * glib/Makefile.am: Apply a patch from Behdad Esfahbod to
57 use a faster lookup table for g_unichar_get_mirror_char().
58 * glib/gmirroringtable.h: The new table.
59 * glib/gunichartables.h: Remove bidi_mirroring_table.
60 * glib/gen-unicode-tables.pl: Don't generate the mirroring
62 * glib/glib-mirroring-tab/*: Sources for the program
63 which generated gmirroringtable.h.
65 * glib/glist.c: Avoid some code duplication.
67 * glib/gscanner.h: Include gdataset.h. (#320322)
69 * glib/gdebug.h: add new GDebugFlag for fatal_criticals
70 * glib/gmessages.c: (_g_debug_init): handle G_DEBUG=fatal_criticals,
71 to help find critical warnings in applications. (#320017,
74 2005-11-02 Tor Lillqvist <tml@novell.com>
76 * glib/glib.symbols: Remove large amount of trailing whitespace
77 from one line. Remove #ifdef G_UNDEFINED from around
78 g_hash_table_new and g_hash_table_new_full symbols, what was the
79 purpose of that? Add the g_slice functions.
81 Tue Nov 1 16:24:20 2005 Tim Janik <timj@imendio.com>
83 * glib/gmem.[hc]: prepared deprecation of GMemChunk and GAllocator.
84 added g_slice_*() API to allocate and cache small bits of memory.
85 an actuall allocator implementation for g_slice_*() is still pending.
87 * glib/gthread.[hc]: changes from a patch by Matthias Clasen.
88 changed GRealThread list to use in-structure *next; fields instead
89 of GSList, in order for thread iteration to not depenend on g_slice_*()
91 _g_thread_mem_private_get():
92 _g_thread_mem_private_set(): added accessors for private memory,
93 needed because the ordinary GPrivate implementation relies on GArray
94 and GSList and therefore indirectly on working g_slice_*() allocations.
97 g_thread_foreach(): new public API function to loop over all existing threads.
111 removed GAllocator and free list usages and accompanying locks.
112 use g_slice_*() API to allocate and cache small bits of memory.
114 * glib/ghook.h: removed GMemChunk field from public API.
117 * glib/glist.h: deprecate allocator API, provide _free1() for consistency.
119 * glib/gnode.h: deprecate allocator API.
121 * glib/gmain.c: reordered GPollRec fields so g_slice_free_chain() can
122 be used for poll rec lists.
124 * glib/grel.c: removed mem chunk usage, and allocated tuples via g_slice_*().
125 g_relation_destroy(): free all tuples from the all_tuples hash table,
126 this effectively maintains the life time track keeping of tuples.
127 g_relation_delete_tuple(): free tuples which are removed from the
128 all_tuples hash table. this fixes a temporary leak that was present
129 in the memchunk code until the destruction of the relation.
131 2005-10-29 Matthias Clasen <mclasen@redhat.com>
133 * tests/convert-test.c: Add some tests for conversions between
134 UTF-8, UCS-4 and UTF-16.
136 * glib/gutf8.c (g_utf8_to_ucs4, g_utf8_to_utf16): Fix handling
137 of len == -1, noticed by Morten Welinder.
139 2005-10-27 Erdal Ronahi <erdal.ronahi@gmail.com>
141 * configure.in: Added ku (Kurdish) to ALL_LINGUAS
143 2005-10-26 Matthias Clasen <mclasen@redhat.com>
145 * glib/gutf8.c (g_ucs4_to_utf8): Don't set items_read twice
146 in the error case, and add some documentation. (#319806, Morten
149 2005-10-19 Manish Singh <yosh@gimp.org>
151 * configure.in: Use AC_CHECK_FUNCS for _NSGetEnviron, to get the
152 config.h symbol automatically. Fixes bug #313731.
154 2005-10-19 Tor Lillqvist <tml@novell.com>
156 * glib/gwin32.c (g_win32_get_package_installation_directory):
157 Return a g_strdup()ed copy of the value stored in the hash table,
158 so that it can be g_free()d without leaving a dangling pointer in
159 the hash table. (#319232)
161 2005-10-06 Matthias Clasen <mclasen@redhat.com>
163 * glib/gunicollate.c (g_utf8_collate_key_for_filename): Handle
164 all-zero sequences correctly. (#317930, Sebastien Bacher)
166 2005-10-05 Matthias Clasen <mclasen@redhat.com>
168 * glib/gmarkup.c (g_markup_escape_text): Doc addition
170 * Makefile.am: Add Collation tests.
172 * tests/collate/*: Inputs and expected outputs for collation tests.
174 * tests/run-collate-tests.sh: Script to run collation tests.
176 * tests/unicode-collate.c (main): Rework slightly to make
177 it usable in unit tests. Also test g_utf8_collate_key_for_filename().
179 2005-10-01 Behdad Esfahbod <behdad@gnome.org>
181 * docs/reference/glib/tmpl/unicode.sgml:
182 * glib/gen-unicode-tables.pl:
184 * glib/gunichartables.h:
186 * tests/casefold.txt:
187 * tests/casemap.txt: Updated to Unicode 4.1. There are five new
188 GUnicodeBreakType types. That may break some applications, like
191 2005-09-26 Matthias Clasen <mclasen@redhat.com>
193 * glib/gstrfuncs.c (g_ascii_strtoull): Add details to the
194 docs. (#314393, Matthew F. Barnes)
197 * glib/gprintf.h: Remove g_snprintf() and g_vsnprintf(), since
198 they are already declared in glib.h. This doesn't break documented
199 use of gprintf.h, but should probably be pointed out in the
200 release notes for 2.10. (#314232, Behdad Esfahbod)
202 Tue Sep 20 13:16:04 2005 Tim Janik <timj@imendio.com>
204 * glib/gpattern.c (g_pattern_ph_match): applied significant recursion
205 complexity optimization, based on a patch from Matthias Clasen.
207 * tests/patterntest.c: more tests, mostly from matthias.
209 2005-09-20 Matthias Clasen <mclasen@redhat.com>
211 * glib/gqueue.c (g_queue_insert_sorted): Correct the docs.
212 (#316703, Mark Drago)
214 Mon Sep 19 17:23:23 2005 Tim Janik <timj@imendio.com>
216 * glib/gpattern.c: applied a patch from matthias which checks on the
217 upper bound of GPatternSpec length to optimize matches.
220 * tests/patterntest.c: added more match cases.
222 2005-09-16 Tor Lillqvist <tml@novell.com>
224 * glib/gstrfuncs.c (g_ascii_strcasecmp, g_ascii_strncasecmp): Add
225 warning to doc comment that these functions should not be used on
226 encodings like CP932.
228 2005-09-14 Matthias Clasen <mclasen@redhat.com>
230 * tests/keyfile-test.c: Add a test for grup names of length 1.
232 * glib/gkeyfile.c (g_key_file_line_is_group): Accept group names
233 of length 1. (#316309)
235 2005-09-12 Matthias Clasen <mclasen@redhat.com>
237 * glib/gmarkup.c (g_markup_escape_text): Clarify docs.
239 2005-09-11 Sebastian Wilhelmi <seppi@seppi.de>
241 * tests/refcount/Makefile.am (INCLUDES): Link the the refcount
242 tests to the system thread library $(G_THREAD_LIBS). Fixes #313744
245 2005-09-11 Kjartan Maraas <kmaraas@gnome.org>
247 * glib/gmain.c: (g_child_watch_prepare), (g_child_watch_check),
248 (child_watch_helper_thread): Remove some dead code. Closes
251 2005-09-07 Tor Lillqvist <tml@novell.com>
253 * glib/Makefile.am: Create also a console version of the
254 gspawn-win32-helper program, gspawn-win32-helper-console.exe.
255 It's otherwise identical to gspawn-win32-helper.exe, except marked
256 as a console application (linked without the -mwindows option).
258 * glib/gspawn-win32.c (do_spawn_directly, do_spawn_with_pipes):
259 Drop the dont_wait parameter. Its truth value correlated 100% with
260 the NULLness of the exit_status parameter anyway, so it's enough
261 to check whether exit_status is NULL. Invert the sense of the
262 dont_return_handle parameter and rename it to do_return_handle, to
263 make the code easier to read by avoiding double negations.
265 (g_spawn_sync_utf8, g_spawn_async_with_pipes_utf8): Modify calls
266 to do_spawn_with_pipes() accordingly.
268 (do_spawn_with_pipes): If we have a console, use the console
269 version of the helper program, otherwise use the GUI one. This
270 avoids extra console windows opening up in some situations. (In
271 case a console application uses the GUI gspawn-win32-helper.exe to
272 spawn another console application we would get a separate console
273 for the spawned console application).
275 * glib-zip.in: Distribute also gspawn-win32-helper-console.exe.
277 2005-09-05 Matthias Clasen <mclasen@redhat.com>
279 * glib/gmappedfile.c (g_mapped_file_new): Report an error
280 if the file is too large. (#315275, Kjartan Maraas)
282 * glib/gkeyfile.c (g_key_file_load_from_fd): The return value
283 of read() is signed. (#315273, Kjartan Maraas)
285 2005-08-31 Tor Lillqvist <tml@novell.com>
287 * glib/gutils.h: Wrapping atexit() is a bad idea on Windows, where
288 the EXE and each DLL have their own atexit function chains.
290 #define g_atexit as atexit instead. This means it has a
291 better chance of doing what the caller wants. For instance,
292 gtkhtml calls g_atexit() registering a function in gtkhtml
293 itself. This caused a crash when g_atexit() was implemented as a
294 function in the GLib DLL. The gtkhtml DLL was already unloaded by
295 the time the GLib DLL got unloaded.
297 * glib/gutils.c: #undef the #define mentioned above, to also get a
298 real g_atexit() into the DLL for backward compatibility. Document
299 the Windows behaviour of g_atexit(), and document the varying ways
300 atexit() can behave in the context of dynamically loaded modules
303 2005-08-31 Matthias Clasen <mclasen@redhat.com>
307 * glib/gdataset.c: Add string interning functions.
309 2005-08-28 Matthias Clasen <mclasen@redhat.com>
311 * glib/giochannel.c: Unify some near-duplicate strings. (#314654,
314 2005-08-26 Matthias Clasen <mclasen@redhat.com>
316 * configure.in: Bump version to 2.9.0
318 2005-08-25 Tor Lillqvist <tml@novell.com>
320 Make also the g_spawn*() functions take parameters in the GLib
321 file name encoding, i.e. UTF-8, on Windows. Has no impact on Unix
322 API or ABI. Like the other GLib API that was earlier changed to
323 use UTF-8 on Windows, the names of the functions that take UTF-8
324 have _utf8 suffixes added by using preprocessor macros in the
325 header file. The old names are kept for functions with the old
326 behaviour, taking parameters in the system codepage, for DLL ABI
329 * glib/gspawn.h: On Win32 add the suffix _utf8 to the names of the
330 g_spawn*() functions.
332 * glib/gspawn-win32.c: Use wide-char API on NT-based
333 Windows. Convert parameters from UTF-8 to wide chars (NT) or
334 system codepage (Win9x) and call the C library _wspawn*() or
335 spawn*() functions respectvely. Add DLL ABI stability versions
336 that take parameters in the system codepage.
338 * glib/gspawn-win32-helper.c: On NT-based Windows use the
339 wide-char versions of argv and envp, and use wide-char API to
340 change directory and spawn the program to run. Remove the verbose
341 debugging output, it was too complex to modify for the wide-char
342 features. (Just add temporary debugging printouts if needed, no
343 need to have them permanently in the source.)
345 * glib/gspawn.c: Corresponding documentation updates.
347 * glib/glib.symbols: Corresponding changes: Mark the ABI stability
348 symbols as PRIVATE, add the new _utf8-suffixed ones.
350 2005-08-24 Stepan Kasal <kasal@ucw.cz>
352 * glib/gtypes.h (G_MININT64): Cast the constant to gint64; it is
353 guint64 otherwise and that can produce warnings about comparison
354 between signed and unsigned.
356 2005-08-23 Matthias Clasen <mclasen@redhat.com>
358 * glib/gutils.c: Fix the crt_externs.h include.
360 2005-08-23 Stepan Kasal <kasal@ucw.cz>
362 * NEWS: Fix spelling of my first name.
364 2005-08-23 Matthias Clasen <mclasen@redhat.com>
368 * === Released 2.8.1 ===
372 2005-08-20 Hans Breuer <hans@breuer.org>
374 * glib/makefile.msc.in : link with ws2_32.lib
376 2005-08-18 Tor Lillqvist <tml@novell.com>
378 * configure.in: Check for <sys/wait.h>
380 * glib/gbacktrace.c: Include <sys/wait.h> on if HAVE_SYS_WAIT_H.
382 2005-08-18 Ross Burton <ross@burtonini.com>
385 Optimise single-character insertions.
390 * tests/string-test.c:
391 Add tests for new optimisation, and fix a leak.
393 2005-08-17 Matthias Clasen <mclasen@redhat.com>
395 * configure.in: Check for crt_externs.h and _NSGetEnviron.
397 * glib/gutils.c: On Darwin, include crt-externs.h and
398 define environ using _NSGetEnviron(). (#313731)
400 2005-08-16 Stepan Kasal <kasal@ucw.cz>
402 * glib/gutils.c (g_get_any_init): Move the body of the big if...
403 (g_get_any_init_do): ... to this new function.
404 (g_get_any_init): Declare as inline.
405 (g_get_any_init_locked): New inline function, does the locking.
406 Make use of these two throughout the code.
408 2005-08-15 Matthias Clasen <mclasen@redhat.com>
410 * glib/gbacktrace.c (g_on_error_stack_trace): Wait for
411 the child process and then simply return. This makes
412 The "S" option work as documented in g_on_error_query().
413 (#313125, Matthew F. Barnes)
415 * glib/gunicode.h: Update the link to Unicode category
416 values. (#313369, Behnam Esfahbod)
418 * glib/gqueue.c (g_queue_find_custom): Clarify docs
419 a little. (#311727, Tristan van Berkom)
421 * glib/abicheck.sh, gobject/abicheck.sh: Make the
422 check work on ia64 too, where some symbols ended up
423 in yet another section.