1 2005-11-04 Matthias Clasen <mclasen@redhat.com>
3 * glib/gscanner.h: Include gdataset.h. (#320322)
5 * glib/gdebug.h: add new GDebugFlag for fatal_criticals
6 * glib/gmessages.c: (_g_debug_init): handle G_DEBUG=fatal_criticals,
7 to help find critical warnings in applications. (#320017,
10 2005-11-02 Tor Lillqvist <tml@novell.com>
12 * glib/glib.symbols: Remove large amount of trailing whitespace
13 from one line. Remove #ifdef G_UNDEFINED from around
14 g_hash_table_new and g_hash_table_new_full symbols, what was the
15 purpose of that? Add the g_slice functions.
17 Tue Nov 1 16:24:20 2005 Tim Janik <timj@imendio.com>
19 * glib/gmem.[hc]: prepared deprecation of GMemChunk and GAllocator.
20 added g_slice_*() API to allocate and cache small bits of memory.
21 an actuall allocator implementation for g_slice_*() is still pending.
23 * glib/gthread.[hc]: changes from a patch by Matthias Clasen.
24 changed GRealThread list to use in-structure *next; fields instead
25 of GSList, in order for thread iteration to not depenend on g_slice_*()
27 _g_thread_mem_private_get():
28 _g_thread_mem_private_set(): added accessors for private memory,
29 needed because the ordinary GPrivate implementation relies on GArray
30 and GSList and therefore indirectly on working g_slice_*() allocations.
33 g_thread_foreach(): new public API function to loop over all existing threads.
47 removed GAllocator and free list usages and accompanying locks.
48 use g_slice_*() API to allocate and cache small bits of memory.
50 * glib/ghook.h: removed GMemChunk field from public API.
53 * glib/glist.h: deprecate allocator API, provide _free1() for consistency.
55 * glib/gnode.h: deprecate allocator API.
57 * glib/gmain.c: reordered GPollRec fields so g_slice_free_chain() can
58 be used for poll rec lists.
60 * glib/grel.c: removed mem chunk usage, and allocated tuples via g_slice_*().
61 g_relation_destroy(): free all tuples from the all_tuples hash table,
62 this effectively maintains the life time track keeping of tuples.
63 g_relation_delete_tuple(): free tuples which are removed from the
64 all_tuples hash table. this fixes a temporary leak that was present
65 in the memchunk code until the destruction of the relation.
67 2005-10-29 Matthias Clasen <mclasen@redhat.com>
69 * tests/convert-test.c: Add some tests for conversions between
70 UTF-8, UCS-4 and UTF-16.
72 * glib/gutf8.c (g_utf8_to_ucs4, g_utf8_to_utf16): Fix handling
73 of len == -1, noticed by Morten Welinder.
75 2005-10-27 Erdal Ronahi <erdal.ronahi@gmail.com>
77 * configure.in: Added ku (Kurdish) to ALL_LINGUAS
79 2005-10-26 Matthias Clasen <mclasen@redhat.com>
81 * glib/gutf8.c (g_ucs4_to_utf8): Don't set items_read twice
82 in the error case, and add some documentation. (#319806, Morten
85 2005-10-19 Manish Singh <yosh@gimp.org>
87 * configure.in: Use AC_CHECK_FUNCS for _NSGetEnviron, to get the
88 config.h symbol automatically. Fixes bug #313731.
90 2005-10-19 Tor Lillqvist <tml@novell.com>
92 * glib/gwin32.c (g_win32_get_package_installation_directory):
93 Return a g_strdup()ed copy of the value stored in the hash table,
94 so that it can be g_free()d without leaving a dangling pointer in
95 the hash table. (#319232)
97 2005-10-06 Matthias Clasen <mclasen@redhat.com>
99 * glib/gunicollate.c (g_utf8_collate_key_for_filename): Handle
100 all-zero sequences correctly. (#317930, Sebastien Bacher)
102 2005-10-05 Matthias Clasen <mclasen@redhat.com>
104 * glib/gmarkup.c (g_markup_escape_text): Doc addition
106 * Makefile.am: Add Collation tests.
108 * tests/collate/*: Inputs and expected outputs for collation tests.
110 * tests/run-collate-tests.sh: Script to run collation tests.
112 * tests/unicode-collate.c (main): Rework slightly to make
113 it usable in unit tests. Also test g_utf8_collate_key_for_filename().
115 2005-10-01 Behdad Esfahbod <behdad@gnome.org>
117 * docs/reference/glib/tmpl/unicode.sgml:
118 * glib/gen-unicode-tables.pl:
120 * glib/gunichartables.h:
122 * tests/casefold.txt:
123 * tests/casemap.txt: Updated to Unicode 4.1. There are five new
124 GUnicodeBreakType types. That may break some applications, like
127 2005-09-26 Matthias Clasen <mclasen@redhat.com>
129 * glib/gstrfuncs.c (g_ascii_strtoull): Add details to the
130 docs. (#314393, Matthew F. Barnes)
133 * glib/gprintf.h: Remove g_snprintf() and g_vsnprintf(), since
134 they are already declared in glib.h. This doesn't break documented
135 use of gprintf.h, but should probably be pointed out in the
136 release notes for 2.10. (#314232, Behdad Esfahbod)
138 Tue Sep 20 13:16:04 2005 Tim Janik <timj@imendio.com>
140 * glib/gpattern.c (g_pattern_ph_match): applied significant recursion
141 complexity optimization, based on a patch from Matthias Clasen.
143 * tests/patterntest.c: more tests, mostly from matthias.
145 2005-09-20 Matthias Clasen <mclasen@redhat.com>
147 * glib/gqueue.c (g_queue_insert_sorted): Correct the docs.
148 (#316703, Mark Drago)
150 Mon Sep 19 17:23:23 2005 Tim Janik <timj@imendio.com>
152 * glib/gpattern.c: applied a patch from matthias which checks on the
153 upper bound of GPatternSpec length to optimize matches.
156 * tests/patterntest.c: added more match cases.
158 2005-09-16 Tor Lillqvist <tml@novell.com>
160 * glib/gstrfuncs.c (g_ascii_strcasecmp, g_ascii_strncasecmp): Add
161 warning to doc comment that these functions should not be used on
162 encodings like CP932.
164 2005-09-14 Matthias Clasen <mclasen@redhat.com>
166 * tests/keyfile-test.c: Add a test for grup names of length 1.
168 * glib/gkeyfile.c (g_key_file_line_is_group): Accept group names
169 of length 1. (#316309)
171 2005-09-12 Matthias Clasen <mclasen@redhat.com>
173 * glib/gmarkup.c (g_markup_escape_text): Clarify docs.
175 2005-09-11 Sebastian Wilhelmi <seppi@seppi.de>
177 * tests/refcount/Makefile.am (INCLUDES): Link the the refcount
178 tests to the system thread library $(G_THREAD_LIBS). Fixes #313744
181 2005-09-11 Kjartan Maraas <kmaraas@gnome.org>
183 * glib/gmain.c: (g_child_watch_prepare), (g_child_watch_check),
184 (child_watch_helper_thread): Remove some dead code. Closes
187 2005-09-07 Tor Lillqvist <tml@novell.com>
189 * glib/Makefile.am: Create also a console version of the
190 gspawn-win32-helper program, gspawn-win32-helper-console.exe.
191 It's otherwise identical to gspawn-win32-helper.exe, except marked
192 as a console application (linked without the -mwindows option).
194 * glib/gspawn-win32.c (do_spawn_directly, do_spawn_with_pipes):
195 Drop the dont_wait parameter. Its truth value correlated 100% with
196 the NULLness of the exit_status parameter anyway, so it's enough
197 to check whether exit_status is NULL. Invert the sense of the
198 dont_return_handle parameter and rename it to do_return_handle, to
199 make the code easier to read by avoiding double negations.
201 (g_spawn_sync_utf8, g_spawn_async_with_pipes_utf8): Modify calls
202 to do_spawn_with_pipes() accordingly.
204 (do_spawn_with_pipes): If we have a console, use the console
205 version of the helper program, otherwise use the GUI one. This
206 avoids extra console windows opening up in some situations. (In
207 case a console application uses the GUI gspawn-win32-helper.exe to
208 spawn another console application we would get a separate console
209 for the spawned console application).
211 * glib-zip.in: Distribute also gspawn-win32-helper-console.exe.
213 2005-09-05 Matthias Clasen <mclasen@redhat.com>
215 * glib/gmappedfile.c (g_mapped_file_new): Report an error
216 if the file is too large. (#315275, Kjartan Maraas)
218 * glib/gkeyfile.c (g_key_file_load_from_fd): The return value
219 of read() is signed. (#315273, Kjartan Maraas)
221 2005-08-31 Tor Lillqvist <tml@novell.com>
223 * glib/gutils.h: Wrapping atexit() is a bad idea on Windows, where
224 the EXE and each DLL have their own atexit function chains.
226 #define g_atexit as atexit instead. This means it has a
227 better chance of doing what the caller wants. For instance,
228 gtkhtml calls g_atexit() registering a function in gtkhtml
229 itself. This caused a crash when g_atexit() was implemented as a
230 function in the GLib DLL. The gtkhtml DLL was already unloaded by
231 the time the GLib DLL got unloaded.
233 * glib/gutils.c: #undef the #define mentioned above, to also get a
234 real g_atexit() into the DLL for backward compatibility. Document
235 the Windows behaviour of g_atexit(), and document the varying ways
236 atexit() can behave in the context of dynamically loaded modules
239 2005-08-31 Matthias Clasen <mclasen@redhat.com>
243 * glib/gdataset.c: Add string interning functions.
245 2005-08-28 Matthias Clasen <mclasen@redhat.com>
247 * glib/giochannel.c: Unify some near-duplicate strings. (#314654,
250 2005-08-26 Matthias Clasen <mclasen@redhat.com>
252 * configure.in: Bump version to 2.9.0
254 2005-08-25 Tor Lillqvist <tml@novell.com>
256 Make also the g_spawn*() functions take parameters in the GLib
257 file name encoding, i.e. UTF-8, on Windows. Has no impact on Unix
258 API or ABI. Like the other GLib API that was earlier changed to
259 use UTF-8 on Windows, the names of the functions that take UTF-8
260 have _utf8 suffixes added by using preprocessor macros in the
261 header file. The old names are kept for functions with the old
262 behaviour, taking parameters in the system codepage, for DLL ABI
265 * glib/gspawn.h: On Win32 add the suffix _utf8 to the names of the
266 g_spawn*() functions.
268 * glib/gspawn-win32.c: Use wide-char API on NT-based
269 Windows. Convert parameters from UTF-8 to wide chars (NT) or
270 system codepage (Win9x) and call the C library _wspawn*() or
271 spawn*() functions respectvely. Add DLL ABI stability versions
272 that take parameters in the system codepage.
274 * glib/gspawn-win32-helper.c: On NT-based Windows use the
275 wide-char versions of argv and envp, and use wide-char API to
276 change directory and spawn the program to run. Remove the verbose
277 debugging output, it was too complex to modify for the wide-char
278 features. (Just add temporary debugging printouts if needed, no
279 need to have them permanently in the source.)
281 * glib/gspawn.c: Corresponding documentation updates.
283 * glib/glib.symbols: Corresponding changes: Mark the ABI stability
284 symbols as PRIVATE, add the new _utf8-suffixed ones.
286 2005-08-24 Stepan Kasal <kasal@ucw.cz>
288 * glib/gtypes.h (G_MININT64): Cast the constant to gint64; it is
289 guint64 otherwise and that can produce warnings about comparison
290 between signed and unsigned.
292 2005-08-23 Matthias Clasen <mclasen@redhat.com>
294 * glib/gutils.c: Fix the crt_externs.h include.
296 2005-08-23 Stepan Kasal <kasal@ucw.cz>
298 * NEWS: Fix spelling of my first name.
300 2005-08-23 Matthias Clasen <mclasen@redhat.com>
304 * === Released 2.8.1 ===
308 2005-08-20 Hans Breuer <hans@breuer.org>
310 * glib/makefile.msc.in : link with ws2_32.lib
312 2005-08-18 Tor Lillqvist <tml@novell.com>
314 * configure.in: Check for <sys/wait.h>
316 * glib/gbacktrace.c: Include <sys/wait.h> on if HAVE_SYS_WAIT_H.
318 2005-08-18 Ross Burton <ross@burtonini.com>
321 Optimise single-character insertions.
326 * tests/string-test.c:
327 Add tests for new optimisation, and fix a leak.
329 2005-08-17 Matthias Clasen <mclasen@redhat.com>
331 * configure.in: Check for crt_externs.h and _NSGetEnviron.
333 * glib/gutils.c: On Darwin, include crt-externs.h and
334 define environ using _NSGetEnviron(). (#313731)
336 2005-08-16 Stepan Kasal <kasal@ucw.cz>
338 * glib/gutils.c (g_get_any_init): Move the body of the big if...
339 (g_get_any_init_do): ... to this new function.
340 (g_get_any_init): Declare as inline.
341 (g_get_any_init_locked): New inline function, does the locking.
342 Make use of these two throughout the code.
344 2005-08-15 Matthias Clasen <mclasen@redhat.com>
346 * glib/gbacktrace.c (g_on_error_stack_trace): Wait for
347 the child process and then simply return. This makes
348 The "S" option work as documented in g_on_error_query().
349 (#313125, Matthew F. Barnes)
351 * glib/gunicode.h: Update the link to Unicode category
352 values. (#313369, Behnam Esfahbod)
354 * glib/gqueue.c (g_queue_find_custom): Clarify docs
355 a little. (#311727, Tristan van Berkom)
357 * glib/abicheck.sh, gobject/abicheck.sh: Make the
358 check work on ia64 too, where some symbols ended up
359 in yet another section.