1 2008-11-28 Matthias Clasen <mclasen@redhat.com>
3 Bug 562378 – callback return value not respected for callback option
6 * glib/goption.c (parse_long_option): Return the parse_arg return
7 value even for no-arg callbacks. Patch by Christian Persch
9 * glib/tests/option-context.c: Add a test for a callback which
12 2008-11-23 Christian Persch <chpe@gnome.org>
14 Bug 559413 – g_option_group_set_error_hook docs buglet
16 * glib/goption.c: Doc fix.
18 2008-11-23 Christian Persch <chpe@gnome.org>
20 Bug 560568 – gkeyfile docs buglet
22 * glib/gkeyfile.c: Clarify the docs.
24 2008-11-23 Christian Persch <chpe@gnome.org>
26 Bug 560569 – gkeyfile doesn't use the set list_separator in some cases
28 * glib/gkeyfile.c: (g_key_file_get_locale_string_list),
29 (g_key_file_set_locale_string_list), (g_key_file_set_integer_list),
30 (g_key_file_set_double_list): Use the key file's list separator character,
33 * glib/tests/keyfile.c: (test_lists), (test_reload_idempotency): Test
36 2008-11-21 Christophe Fergeau <teuf@gnome.org>
38 Bug 561212 – GFileReadMoreCallback API doc refers to non-existant function
40 * gio/giotypes.h: fix name of function referred to in
41 GFileReadMoreCallback API documentation
43 2008-11-21 Matthias Clasen <mclasen@redhat.com>
45 * configure.in: Bump gtk-doc dependency to 1.11 for
46 nicer index-generation.
48 2008-11-14 Matthias Clasen <mclasen@redhat.com>
52 2008-11-04 Christian Dywan <christian@imendio.com>
54 Bug 558672 – NULL key lookup using g_hash_table_lookup_extended()
57 Clarify g_hash_table_lookup_extended
59 2008-11-02 Tor Lillqvist <tml@novell.com>
61 * configure.in: Expand also build/win32/vs9/Makefile.
63 2008-10-31 Christian Dywan <christian@imendio.com>
65 Bug 558513 – g_warn_if_fail FIXME in gtestutils
67 * glib/gtestutils.c (g_test_log_send):
68 Turn g_assert into g_warn_if_fail as advised
70 2008-10-31 Grahame Bowland <grahame@angrygoats.net>
72 Bug 558185 – 'parent' variable in g_local_file_get_child_for_display_name()
73 hits g_object_unref(NULL) assertion
76 - remove unused variable. Patch by Matt Johnston <matt@ucc.asn.au>
78 2008-10-29 16:11:14 Tim Janik <timj@gtk.org>
80 * glib/gmacros.h: added G_PASTE() and G_STATIC_ASSERT(), based on
81 patches by Christian Persch and Christian Dywan. Fixes:
82 Bug 558381 - Add support for compile time assertions
84 2008-10-27 Joseph Pingenot <gnomesvn@digitasaru.net>
86 * glib/ghash.c: changed "#GHashTableIterator" to "#GHashTableIter"
87 in the documentation-comments above g_hash_table_foreach_remove()
88 and g_hash_table_foreach_steal() to correctly name and link to
89 GHashTableIter. This affects the GNOME documentation website
90 as well, so further steps are likely necessary beyond this subproject..
92 2008-10-23 Matthias Clasen <mclasen@redhat.com>
94 Bug 557210 – g_compute_checksum_for_* asserts with less than 2 bytes
96 * glib/gchecksum.c (g_compute_checksum_for_data): Accept
97 lengths < 2. Patch by Tommi Komulainen
99 * tests/checksum-test.c: Add a test for this
101 2008-10-23 Matthias Clasen <mclasen@redhat.com>
103 Bug 556921 – gpoll.h breaks hal compilation
105 * glib/gpoll.h: Include gtypes.h. Pointed out by Anis Elleuch
107 2008-10-20 Christian Persch <chpe@gnome.org>
109 Bug 557087 – mem leak in g_content_types_get_registered
111 * gio/gcontenttype.c: Plug a mem leak.
113 2008-10-19 Michael Natterer <mitch@imendio.com>
115 Bug 556186 – gpoll.h breaks gmain.h inclusion
117 * glib/gpoll.h: different fix: disallow direct inclusion
118 unconditionally again but make an exception if included from
119 gmain.h to maintain compatibility.
121 2008-10-16 Matthias Clasen <mclasen@redhat.com>
123 * configure.in: Bump version to 2.19.1
125 * === Released 2.19.0 ===
129 2008-10-14 Christian Persch <chpe@gnome.org>
131 Bug 556186 – gpoll.h breaks gmain.h inclusion
133 * glib/gpoll.h: Only disallow direct inclusion when
134 G_DISABLE_SINGLE_INCLUDES is defined.
136 2008-10-13 Sven Herzberg <sven@imendio.com>
138 Bug 556101 – static mutex yields warnings with g++
139 Reviewed by Tim Janik
141 * configure.in: added an intermediate cast to gpointer
143 2008-10-13 Christian Persch <chpe@gnome.org>
145 Bug 555311 – format not a string literal and no format arguments
148 * glib/gshell.c: Use literal errors where appropriate.
150 2008-10-10 Behdad Esfahbod <behdad@gnome.org>
152 Bug 551355 – [PATCH] Make glib build with libtool 2.2
154 * autogen.sh: Accept libtool 2.2. We are moving towards having
157 2008-10-10 Behdad Esfahbod <behdad@gnome.org>
159 * configure.in: With libtool 2.x, regenerate libtool early.
160 Based on bug 552107, patch by Patryk Zawadzki
162 2008-10-10 Matthias Clasen <mclasen@redhat.com>
164 Bug 554557 – Patch to fix gcc warnings about missing format
169 * glib/grel.c: Fix gcc warnings. Patch by Kjartan Maraas
171 2008-10-10 Matthias Clasen <mclasen@redhat.com>
173 Bug 552861 – glib-2.0.m4 calls system(3) without storing its result
175 * m4macros/glib-2.0.m4: Cosmetic change to make -Werror happy.
176 Patch by Andreas Köhler
178 2008-10-08 Christian Persch <chpe@gnome.org>
180 Bug 555313 – GFileAttribute boxed type get_type function should
181 use usual get_type pattern
183 * gio/gfileattribute.c (g_file_attribute_info_list_get_type): Use
184 g_once_init_enter/leave.
186 2008-10-08 Christian Persch <chpe@gnome.org>
188 Bug 555314 – mem leak in gmarkup
190 * glib/gmarkup.c: (g_markup_parse_context_parse): Plug a mem leak.
192 2008-10-08 Tor Lillqvist <tml@novell.com>
194 Bug 554790 - g_convert() misbehaves with winiconv versions
196 * glib/win_iconv.c (kernel_mbtowc): If converting from ASCII,
197 explicitly check for and reject 8bit chars. MultiByteToWideChar()
198 doesn't, at least not on XP.
200 2008-10-06 Matthias Clasen <mclasen@redhat.com>
202 * glib/gtypes.h: Properly include gmacros.h
204 2008-10-06 Behdad Esfahbod <behdad@gnome.org>
206 Bug 555309 – giochannel breaks on error
207 Patch from Christian Persch
209 * glib/giounix.c (g_io_unix_read), (g_io_unix_write),
210 (g_io_unix_seek), (g_io_unix_close), (g_io_unix_set_flags):
211 Don't shadow err. Oops!
213 2008-10-06 Christophe Fergeau <teuf@gnome.org>
215 Bug 555224 – Improve g_format_size_for_display doc
217 * glib/gfileutils.c: change g_format_size_for_display API doc to
218 explicitly say that the returned string has to be freed. Change
219 spelling of "newly allocated" to "newly-allocated" in g_file_read_link
220 API doc to be more consistent with what is done in that file.
222 2008-10-01 David Zeuthen <davidz@redhat.com>
224 * README.in: Add "Notes about glib 2.20" section detailing the
225 ramifications of the patch from bug #528670.
227 2008-09-30 Behdad Esfahbod <behdad@gnome.org>
229 Bug 554092 – glib doesn't return G_FILE_ERROR_NOENT et al on OS X
231 * glib/giounix.c (g_io_unix_read), (g_io_unix_write),
232 (g_io_unix_seek), (g_io_unix_close), (g_io_unix_set_flags),
233 (g_io_unix_get_flags), (g_io_channel_new_file):
234 Like mclasen says: "well, thats the way errno works...,
235 save it or loose it". Save errno.
237 2008-09-30 Tor Lillqvist <tml@novell.com>
240 * glib/tests/Makefile.am: Bypass gtester related stuff on Windows.
242 2008-09-30 Tor Lillqvist <tml@novell.com>
245 * glib/gnulib/vasnprintf.c: Don't define _GNU_SOURCE on Windows,
246 as _GNU_SOURCE has unintended side effects when compiling against
247 newest mingw headers.
249 2008-09-26 Dan Winship <danw@gnome.org>
251 Bug 553447 – g_assert_no_error()
253 * glib/gtestutils.h (g_assert_no_error, g_assert_error): Macros to
254 assert that a GError is not set, or else is set to a particular
257 * glib/gtestutils.c (g_assertion_message_error): utility for
260 * glib/tests/keyfile.c:
261 * tests/asyncqueue-test.c:
262 * tests/bookmarkfile-test.c:
263 * tests/convert-test.c:
264 * tests/file-test.c: Use g_assert_error/g_assert_no_error
266 2008-09-26 Dan Winship <danw@gnome.org>
268 * glib/gthreadpool.c (wakeup_thread_marker): make this a "const
269 gpointer" rather than a gconstpointer to avoid warnings later
271 * glib/pcre/pcre_ucp_searchfuncs.c:
272 * glib/pcre/pcre_valid_utf8.c: #include "config.h"
274 * glib/tests/printf.c (test_d): fool gcc into not warning about
275 some printf format strings that we know are dubious
277 2008-09-26 Matthias Clasen <mclasen@redhat.com>
279 Bug 553857 – gbacktrace.h requires signal.h
281 * glib/gbacktrace.h: Include signal.h for raise().
282 Pointed out by Sebastien Bacher
284 2008-09-26 Matthias Clasen <mclasen@redhat.com>
286 Bug 553724 – python interpretter path not patched in correctly
288 * glib/Makefile.am: Fix the sed magic to replace python.
290 2008-09-26 Matthias Clasen <mclasen@redhat.com>
292 * glib/gmain.c: Add some more docs.
294 * glib/giochannel.c: Move more docs inline, and improve them
297 2008-09-25 Tor Lillqvist <tml@novell.com>
299 Bug 553820 - gpoll.c: undeclared identifier
302 * glib/gpoll.c: Make the g_poll() function non-static also on
303 Windows. Prefix an underscore to the g_main_poll_debug variable
304 and make it non-static in gmain.c so that it can be used in
305 gpoll.c. Add back missing variable declaration.
307 2008-09-25 Tor Lillqvist <tml@novell.com>
309 * glib/gspawn-win32.c (do_spawn_with_pipes) (do_spawn_directly):
310 Just ignore the child_setup function, never call it. The is no
311 situation in which it could be useful on Windows. Do print a
312 warning, like before.
314 * glib/gspawn.c (g_spawn_async_with_pipes): Corresponding change
317 2008-09-24 Sven Herzberg <sven@imendio.com>
319 Be a little more explcit in the docs. Includes Owen's requested
322 * glib/gmain.c: improved documentation for g_source_attach() and
325 2008-09-23 Michael Natterer <mitch@imendio.com>
327 * glib/glib.h: #include <glib/gpoll.h>
329 * glib/gpoll.h: #error out if gpoll.h is included directly.
331 * glib/gpoll.c: remove trailing whitespace.
333 2008-09-23 Dan Winship <danw@gnome.org>
335 * glib/gpoll.[ch] (g_poll): Move this out of gmain.c and make it part
336 of the public API. (Part of Bug 505361 - gunixinputstream.c assumes
339 2008-09-23 Tor Lillqvist <tml@novell.com>
341 * glib/gmain.c (poll_rest) [Win32]: Fix embarrassing bug: I was
342 passing an incorrect third parameter to memmove(), had forgotten
343 to multiply by the size of the table entry. Just use a for loop
344 instead, clearer. Odd I didn't notice when testing this code.
346 2008-09-19 Hans Petter Jansson <hpj@novell.com>
348 Rewrite most of GHashTable to use open addressing with quadratic
349 probing instead of chaining. This has the potential to reduce memory
350 fragmentation significantly, while being slightly faster due to
351 better locality and no need to call alloc/free functions for nodes.
352 Benchmarks suggest it also uses less memory overall.
354 * glib/ghash.c (prime_mod): Table of suitable primes for
355 initial-probe distribution.
356 (g_hash_table_set_shift): New function.
357 (g_hash_table_find_closest_shift): New function.
358 (g_hash_table_set_shift_from_size): New function.
359 (g_hash_table_lookup_node_for_insertion): New function.
360 (g_hash_table_lookup_node): Rewritten to return node index instead of
361 pointer, use quadratic probe on flat table, and not return insertion
362 data. The latter saves some computation for read-only lookups.
363 (g_hash_table_remove_node): Rewrite to take a pointer directly to the
364 node structure to remove, and clear that. Remove unlinking code.
365 (g_hash_table_remove_all_nodes): Rewrite to not clear nodes
366 individually, but en masse using memset () after potentially calling
368 (iter_remove_or_steal): Use new data structure and algorithm. Vastly
369 simplified - now just a call to g_hash_table_remove_node ().
370 (g_hash_table_resize): New resize code, re-indexing with new prime
371 and cleaning up tombstones.
372 (g_hash_table_maybe_resize): Table may hold 8 buckets minimum, no less
373 than 1/4 load excluding tombstones, and no more than 15/16 load
374 including tombstones. These numbers are the results of a lot of
375 benchmarking with multiple complex applications, and should not be
377 (g_hash_table_iter_next)
378 (g_hash_table_lookup)
379 (g_hash_table_lookup_extended)
380 (g_hash_table_insert_internal)
381 (g_hash_table_remove_internal)
382 (g_hash_table_foreach_remove_or_steal)
383 (g_hash_table_foreach)
385 (g_hash_table_get_keys)
386 (g_hash_table_get_values): Use new data structure and algorithm,
387 fairly trivial changes.
389 2008-09-19 Tor Lillqvist <tml@novell.com>
391 * glib-zip.in: Look for man pages in share/man.
393 * glib/gutils.c (_glib_get_dll_directory)
394 * glib/gspawn-win32.c (do_spawn_with_pipes): Be a bit less
395 restrictive, look for the helper programs in the same folder where
396 the GLib DLL is, not necessarily in a "bin" subfolder of the top
397 GLib installation folder.
399 2008-09-18 Matthias Clasen <mclasen@redhat.com>
401 * configure.in: Bump version to 2.19.0
403 * ChangeLog.pre-2-18: rotate ChangeLog
405 * === branch for 2.18 ===