Colin Walters [Fri, 3 Jun 2011 18:53:27 +0000 (14:53 -0400)]
Generate $module-public-headers.txt file, feed it to gtk-doc
Rather than having the gtk-doc build machinery have a list of header
files to exclude, change the GLib build to dump a list of public
header files generated from the maintained Makefile.am files for
each of glib/, gobject/, gio/.
Also, for glib, always install glib-unix.h, even on non-Unix
platforms, for the same reason we install gwin32.h even on Unix.
https://bugzilla.gnome.org/show_bug.cgi?id=651745
Colin Walters [Tue, 7 Jun 2011 18:17:11 +0000 (14:17 -0400)]
gdbus-codegen: Clean config.pyc to make distcheck happy
Colin Walters [Tue, 7 Jun 2011 17:02:41 +0000 (13:02 -0400)]
g_get_user_data_dir(): Don't warn if XDG_RUNTIME_DIR isn't set
We have a meaningful fallback, and warning causes the tests to break.
Kjartan Maraas [Tue, 7 Jun 2011 08:02:40 +0000 (10:02 +0200)]
Updated Norwegian bokmål translation
Marek Černocký [Tue, 7 Jun 2011 06:08:16 +0000 (08:08 +0200)]
Updated Czech translation
Matthias Clasen [Tue, 7 Jun 2011 03:24:36 +0000 (23:24 -0400)]
Add a testcase for the previous fix
Matthias Clasen [Tue, 7 Jun 2011 03:23:29 +0000 (23:23 -0400)]
GDataList: don't hold the bitlock over callbacks
g_datalist_id_clear_i was dropping the dataset lock around
the destroy notifies, but kept the bitlock. This was causing
deadlocks when finalizing widgets.
Chun-wei Fan [Tue, 7 Jun 2011 02:58:54 +0000 (10:58 +0800)]
Update VS 2008/2010 property sheets
"Install" the new gobject/glib-types.h as well... it is needed.
Chun-wei Fan [Tue, 7 Jun 2011 02:49:29 +0000 (10:49 +0800)]
Update gsettings utilities in GIO
Avoid C99-style variable declaration
Chun-wei Fan [Tue, 7 Jun 2011 02:32:47 +0000 (10:32 +0800)]
Update config.h.win32.in
-Make contents more like the current config.h(.in)
-vsnprintf is included in VS 2008+
Chun-wei Fan [Tue, 7 Jun 2011 00:55:31 +0000 (08:55 +0800)]
Update the year in the *.rc.in files
Sorry this change came a bit late...
Chun-wei Fan [Tue, 7 Jun 2011 00:25:07 +0000 (08:25 +0800)]
Clean up gobject/gvaluetransform.c a bit
The check for _MSC_VER here is not necessary here because:
-One cannot compile GLib out-of-the-box with VS 2003 (let alone VS6
or earlier) since GLib 2.22.4 or so.
-This code compiles fine with the currently-supported VS versions
(2008/2010) without the error mentioned in the comments.
This will close Bug 652002.
Chun-wei Fan [Tue, 7 Jun 2011 00:20:15 +0000 (08:20 +0800)]
Fix gatomic.c compilation on Windows (MSVC)
-windows.h must be included for the functions[1] to be found during
compilation
-For Win32(x86), the InterlockedAnd, InterlockedOr and InterlockedXor
functions need to be preceded with an underscore. This is mandated
by the "For the x86 architecture..." clause under "Remarks"[2]
-This will close Bug 652000
[1]:http://msdn.microsoft.com/en-us/library/ms684122(v=vs.85).aspx
[2]:http://msdn.microsoft.com/en-us/library/ms683516(v=vs.85).aspx
http://msdn.microsoft.com/en-us/library/ms683626(v=vs.85).aspx
http://msdn.microsoft.com/en-us/library/ms684021(v=vs.85).aspx
Colin Walters [Mon, 6 Jun 2011 21:14:45 +0000 (17:14 -0400)]
gdbus codegen: Ensure PYTHONPATH is set correctly for srcdir != builddir
We need to pick up the Python files from the source directory, so set
PYTHONPATH.
Chun-wei Fan [Mon, 6 Jun 2011 16:37:38 +0000 (00:37 +0800)]
Update GLib VS9/10 projects
There is no longer need to distinguish between gatomic.c and gatomic-gcc.c.
This will avoid gatomic.c from being included in the projects twice.
Dan Winship [Mon, 6 Jun 2011 14:22:40 +0000 (10:22 -0400)]
Fix some socket docs issues, remove unused typedefs
pointed out by Matthias
Colin Walters [Fri, 3 Jun 2011 17:18:44 +0000 (13:18 -0400)]
GBuffer: Suffix header with private.h
This makes it clearer it's not public API yet.
https://bugzilla.gnome.org/show_bug.cgi?id=651745
Matthias Clasen [Mon, 6 Jun 2011 05:14:23 +0000 (01:14 -0400)]
GMappedFile: improve test coverage
Matthias Clasen [Mon, 6 Jun 2011 05:14:00 +0000 (01:14 -0400)]
GKeyFile: improve test coverage
Matthias Clasen [Mon, 6 Jun 2011 05:13:46 +0000 (01:13 -0400)]
GConvert: improve test coverage
Matthias Clasen [Mon, 6 Jun 2011 05:13:26 +0000 (01:13 -0400)]
Improve GChecksum test coverage
Matthias Clasen [Mon, 6 Jun 2011 05:04:19 +0000 (01:04 -0400)]
GMappedFile: remove nonatomic accesses to ref_count
Matthias Clasen [Mon, 6 Jun 2011 04:31:52 +0000 (00:31 -0400)]
Improve test coverage of gtimezone.c
Matthias Clasen [Mon, 6 Jun 2011 04:31:40 +0000 (00:31 -0400)]
Improve test coverage of GVariant
Matthias Clasen [Mon, 6 Jun 2011 04:31:24 +0000 (00:31 -0400)]
Improve test coverage of glist.c
Matthias Clasen [Mon, 6 Jun 2011 04:31:06 +0000 (00:31 -0400)]
Improve test coverage of glib-unix.c
Matthias Clasen [Mon, 6 Jun 2011 04:30:50 +0000 (00:30 -0400)]
Improve test coverage of gurifuncs.c
Matthias Clasen [Mon, 6 Jun 2011 04:30:30 +0000 (00:30 -0400)]
Test runtime dir getter
Matthias Clasen [Mon, 6 Jun 2011 04:30:02 +0000 (00:30 -0400)]
Remove deprecated atomic functions from tests
Matthias Clasen [Mon, 6 Jun 2011 04:28:38 +0000 (00:28 -0400)]
Improve test coverage
Duplicate the atomic tests for the functions, and
test the new atomic operations as well.
Matthias Clasen [Mon, 6 Jun 2011 02:20:18 +0000 (22:20 -0400)]
No need to have two variants of errno-to-GError
Matthias Clasen [Sun, 5 Jun 2011 23:07:06 +0000 (19:07 -0400)]
Fix a typo
And reword some docs.
Fran Diéguez [Sun, 5 Jun 2011 21:30:42 +0000 (23:30 +0200)]
Updated Galician translations
Kalev Lember [Sun, 5 Jun 2011 11:34:33 +0000 (14:34 +0300)]
configure.ac: Check for qsort_r function before running the test program
When cross compiling, test programs cannot be run. In order to make it
easier to cross compile for targets that do not have qsort_r(), check
for the function before trying to run the test program.
This avoid having to populate cache with glib_cv_have_qsort_r=no with
e.g. mingw cross compiler.
https://bugzilla.gnome.org/show_bug.cgi?id=651920
Kalev Lember [Sun, 5 Jun 2011 11:26:12 +0000 (14:26 +0300)]
configure.ac: Print qsort_r result only once
AC_CACHE_CHECK that was introduced in
70a19815 prints the result, so the
following AC_MSG_RESULT is no longer needed.
Jorge González [Sun, 5 Jun 2011 14:59:31 +0000 (16:59 +0200)]
Updated Spanish translation
Matthias Clasen [Sun, 5 Jun 2011 03:43:05 +0000 (23:43 -0400)]
Use g_param_spec_ref_sink instead of separate ref, sink
Matthias Clasen [Sun, 5 Jun 2011 03:42:07 +0000 (23:42 -0400)]
GParamSpec: avoid unnecessary double initialization of qdata
Matthias Clasen [Sun, 5 Jun 2011 03:13:55 +0000 (23:13 -0400)]
No necessary to init qdata atomically
David Zeuthen [Sun, 5 Jun 2011 01:37:37 +0000 (21:37 -0400)]
GDBusObject: Remove comment non-existant virtual functions
These were nuked a while ago.
Signed-off-by: David Zeuthen <davidz@redhat.com>
Matthias Clasen [Sun, 5 Jun 2011 00:06:44 +0000 (20:06 -0400)]
bump version
Matthias Clasen [Sat, 4 Jun 2011 22:48:54 +0000 (18:48 -0400)]
2.29.6
Matthias Clasen [Sat, 4 Jun 2011 23:05:19 +0000 (19:05 -0400)]
Fix doc syntax
Matthias Clasen [Sat, 4 Jun 2011 22:44:44 +0000 (18:44 -0400)]
Fix links in gio docs
Yaron Shahrabani [Sat, 4 Jun 2011 22:05:03 +0000 (01:05 +0300)]
Updated Hebrew translation.
Matthias Clasen [Sat, 4 Jun 2011 18:40:23 +0000 (14:40 -0400)]
Fix parameter documentation for many functions
Use the opt-out mechanism introduced in gtk-doc 1.16 to work
around problems with the _utf8 renaming games that the win32
port is playing in our headers.
https://bugzilla.gnome.org/show_bug.cgi?id=638449
Matthias Clasen [Sat, 4 Jun 2011 15:57:10 +0000 (11:57 -0400)]
Fix doc typos
Now with fewer broken links...
Matthias Clasen [Sat, 4 Jun 2011 15:46:10 +0000 (11:46 -0400)]
configure.in: Use tar-ustar to avoid name length limitations
Also use dist-bzip2.
David Zeuthen [Sat, 4 Jun 2011 16:43:00 +0000 (12:43 -0400)]
Fix up gdbus invocation in example
Signed-off-by: David Zeuthen <davidz@redhat.com>
David Zeuthen [Sat, 4 Jun 2011 16:33:39 +0000 (12:33 -0400)]
gdbus: Add a way to emit a signal
See https://bugs.freedesktop.org/show_bug.cgi?id=37890#c6 where it was
discovered that dbus-send(1) actually doesn't work (either libdbus-1's
flush implementation or dbus-send(1)'s usage of it is broken) so it's
useful to have here.
Signed-off-by: David Zeuthen <davidz@redhat.com>
David Zeuthen [Sat, 4 Jun 2011 15:52:08 +0000 (11:52 -0400)]
Fix typos in "Migrating to GDBus" docs
It's Skeleton now, not Stub.
Signed-off-by: David Zeuthen <davidz@redhat.com>
Matthias Clasen [Sat, 4 Jun 2011 14:46:38 +0000 (10:46 -0400)]
Distcheck fixes
Fix VPATH builds.
Matthias Clasen [Sat, 4 Jun 2011 14:45:30 +0000 (10:45 -0400)]
Distcheck fixes
Accidentally committed debug leftover.
Matthias Clasen [Sat, 4 Jun 2011 12:13:48 +0000 (08:13 -0400)]
Update NEWS
Yaron Shahrabani [Sat, 4 Jun 2011 10:44:19 +0000 (13:44 +0300)]
Updated Hebrew translation.
Paolo Bonzini [Tue, 25 Jan 2011 10:31:41 +0000 (11:31 +0100)]
avoid quadratic behavior of GMainLoop when all fd's have the same priority
https://bugzilla.gnome.org/show_bug.cgi?id=640518
Matthias Clasen [Sat, 4 Jun 2011 03:38:46 +0000 (23:38 -0400)]
Add a test for the previous fix
Matthias Clasen [Sat, 4 Jun 2011 03:37:23 +0000 (23:37 -0400)]
GRand: avoid overflow
Otherwise, g_random_double_range (-G_MAXDOUBLE, G_MAXDOUBLE)
doesn't work as expected.
https://bugzilla.gnome.org/show_bug.cgi?id=502560
Matthias Clasen [Sat, 4 Jun 2011 02:45:51 +0000 (22:45 -0400)]
g_mkdir_with_parents: cope with dirs popping into existence
Don't bail out if a directory suddenly turns out to exist
after all. Proposed in bug 612729.
Mikhail Zabaluev [Wed, 17 Mar 2010 00:34:51 +0000 (02:34 +0200)]
Made g_utf8_to_ucs4_fast() even faster
https://bugzilla.gnome.org/show_bug.cgi?id=619435
Mikhail Zabaluev [Mon, 15 Mar 2010 01:37:11 +0000 (03:37 +0200)]
Added performance tests for UTF-8 decoding functions
https://bugzilla.gnome.org/show_bug.cgi?id=619418
Matthew Barnes [Sat, 16 Oct 2010 11:40:44 +0000 (07:40 -0400)]
Have g_queue_remove() return a boolean
g_queue_remove() should return a boolean so callers can verify that an
element was found and removed, as in the following example:
if (g_queue_remove (queue, referenced_object))
g_object_unref (referenced_object);
Similarly, g_queue_remove_all() should return the number of elements
found and removed.
https://bugzilla.gnome.org/show_bug.cgi?id=632294
Wim Taymans [Wed, 16 Sep 2009 15:39:48 +0000 (17:39 +0200)]
gthread: avoid locking in _get_mutex_impl
When getting the mutex implementation of a static mutex, avoid taking the global
lock every time but only take the lock when there was no mutex and we need to
create one.
https://bugzilla.gnome.org/show_bug.cgi?id=599954
Alexander Larsson [Wed, 25 May 2011 09:56:51 +0000 (11:56 +0200)]
Use g_atomic_pointer_or/and to set datalist flags
https://bugzilla.gnome.org/show_bug.cgi?id=650458
Alexander Larsson [Thu, 19 May 2011 19:55:16 +0000 (21:55 +0200)]
Make g_object_get_data use the new faster g_datalist_get_data
https://bugzilla.gnome.org/show_bug.cgi?id=650458
Alexander Larsson [Thu, 19 May 2011 19:51:49 +0000 (21:51 +0200)]
Make g_datalist_get_data not look up the quark
Instead of converting the string to a quark and comparing quarks we
use the new lockless g_quark_to_string and just compare the quarks
in the datalist with the given string.
This means we avoid the global lock for string to quark. Additionally
most of the time the data list will be quite short, so the cost of
doing the sting comparisons is likely similar to that of the quark
hashtable lookup (which does at least one string comparison for a
successfull lookup).
https://bugzilla.gnome.org/show_bug.cgi?id=650458
Alexander Larsson [Thu, 19 May 2011 19:30:38 +0000 (21:30 +0200)]
Make quark to string lockless
We do this by assigning to g_quarks atomically and leaking it when
replacing it atomically. Then its safe to consume the array
on the reader side (atomically).
Also, since we're leaking quarks on growing, bump the block size
so that we're not leaking as much. gtk3-demo allocates > 1500 quarks,
and gnome apps > 3000. I'm setting the block to 2048 which means no
leaks for small gtk3 apps and just one leak for gnome apps.
https://bugzilla.gnome.org/show_bug.cgi?id=650458
Alexander Larsson [Thu, 19 May 2011 12:48:50 +0000 (14:48 +0200)]
Make g_datalist not use a global lock and perform better
This implementation uses a per-list bitlock for user data, and a
simple array rather than a linked list which uses less memory and less
allocations. It also gets better cache behaviour since related things
are stored close to each other.
https://bugzilla.gnome.org/show_bug.cgi?id=650458
Matthias Clasen [Sat, 4 Jun 2011 00:54:32 +0000 (20:54 -0400)]
Add pointer bitlocks to the docs
Ryan Lortie [Wed, 25 May 2011 09:08:20 +0000 (11:08 +0200)]
Implement pointer sized bitlocks
Based on a patch from Alexander Larsson.
https://bugzilla.gnome.org/show_bug.cgi?id=651467
Ryan Lortie [Fri, 3 Jun 2011 18:29:41 +0000 (20:29 +0200)]
bitlock: hand-code assembly version for x86
The __sync_fetch_and_or() operation on x86 is a bit suboptimal when the
result isn't ignored. Normally we could use the 'lock or' assembly
instruction to accomplish this, but this instruction discards the
previous value.
In order to work around this issue, GCC is forced to emit a
compare-and-exchange loop.
We can easily use the 'lock bts' instruction, though. It can't be used
in the general case for __sync_fetch_and_or() but it works great for our
case (test and set a single bit).
I filed a bug against GCC[1] to get this exposed as a new intrinsic (or
have the optimiser detect the case) but until then we'll hand-code it on
x86 and amd64.
The uncontended case sees a 31% improvement on my test machine.
[1] http://gcc.gnu.org/bugzilla/show_bug.cgi?id=49244
https://bugzilla.gnome.org/show_bug.cgi?id=651467
Giovanni Campagna [Wed, 1 Jun 2011 14:45:43 +0000 (16:45 +0200)]
gvariant: fix introspection annotations
g_variant_get_strv and g_variant_get_bytestring return arrays that
are null terminated and have an explicit length. Since gjs doesn't
support (out) arrays with length, mark them also null-terminated
(but leave the length annotation, so pygobject can remove the argument)
https://bugzilla.gnome.org/show_bug.cgi?id=646635
David Zeuthen [Fri, 3 Jun 2011 18:45:29 +0000 (14:45 -0400)]
Fix include path for gdbus-codegen example XML
Signed-off-by: David Zeuthen <davidz@redhat.com>
David Zeuthen [Fri, 3 Jun 2011 18:39:04 +0000 (14:39 -0400)]
Fix docs for D-Bus introspection data structures
Mark structs as boxed types and use /*< public >*/ so the struct
members are properly shown.
Signed-off-by: David Zeuthen <davidz@redhat.com>
David Zeuthen [Fri, 3 Jun 2011 17:22:31 +0000 (13:22 -0400)]
Move gdbus-codegen example code and docs into separate directories
This avoids the generated types (e.g. ExampleAnimal, ExampleCat,
ExampleObject and ExampleObjectManagerClient) being referenced in the
core gio docs. This was requested by Matthias.
Signed-off-by: David Zeuthen <davidz@redhat.com>
Colin Walters [Thu, 2 Jun 2011 21:14:58 +0000 (17:14 -0400)]
gobject: Use _ prefixing in private headers over G_GNUC_INTERNAL in public headers
This helps out gtk-doc and g-ir-scanner, and also makes much clearer
what's private versus public.
https://bugzilla.gnome.org/show_bug.cgi?id=651745
Colin Walters [Thu, 2 Jun 2011 15:53:25 +0000 (11:53 -0400)]
spawn-multithreaded: New test case
We didn't have any coverage of threads spawning processes, which we
should definitely support.
https://bugzilla.gnome.org/show_bug.cgi?id=651725
Colin Walters [Thu, 2 Jun 2011 13:10:36 +0000 (09:10 -0400)]
gmain: Consolidate UNIX signal init state handling
For a future signalfd() patch, it will be easier to handle if
we don't separate initialization from watching for a particular
signal.
https://bugzilla.gnome.org/show_bug.cgi?id=651725
Colin Walters [Thu, 2 Jun 2011 12:56:25 +0000 (08:56 -0400)]
gmain: Clean up SIGCHLD handling
Unify it more with the rest of the signal handling code. There's
no reason not to specify SA_RESTART and SA_NOCLDSTOP for flags
always, so just do it.
Remove unnecessary initialization, and push the internal API
towards just ensure_unix_signal_handler_installed_unlocked().
https://bugzilla.gnome.org/show_bug.cgi?id=651725
Hanro [Thu, 2 Jun 2011 19:44:19 +0000 (15:44 -0400)]
gkeyfile: Only test file descriptors against -1
Windows CE at least will return other (valid) negative values.
Colin Walters [Wed, 1 Jun 2011 19:11:02 +0000 (15:11 -0400)]
gdbus: Avoid blocking on worker thread in connection initialization
I can't see a reason to spin until the worker thread runs, so don't.
This avoids ugly sched_yield() calls that show up in strace and
annoy me; the code is cleaner now too.
We now grab the types needed for the WebKit workaround in the
thread creation area, but only release them when the thread itself
exits.
https://bugzilla.gnome.org/show_bug.cgi?id=651650
Matthias Clasen [Tue, 31 May 2011 23:23:14 +0000 (19:23 -0400)]
Expand content type docs slighly
Matthias Clasen [Tue, 31 May 2011 18:53:30 +0000 (14:53 -0400)]
Add a note about atomic changes
Colin Walters [Tue, 31 May 2011 17:36:16 +0000 (13:36 -0400)]
Add a boxed type for GVariantBuilder, mark GVariantIter as (skip)
Based on a patch by Giovanni Campagna <gcampagna@src.gnome.org>
From discussion, GVariantIter is not useful for bindings, but
GVariantBuilder may be.
https://bugzilla.gnome.org/show_bug.cgi?id=646635
Colin Walters [Tue, 31 May 2011 17:25:09 +0000 (13:25 -0400)]
gutils.c: Annotation fixes
Colin Walters [Tue, 31 May 2011 17:10:12 +0000 (13:10 -0400)]
g_shell_parse_argv: Annotate
Colin Walters [Tue, 31 May 2011 16:00:41 +0000 (12:00 -0400)]
gfileutils: Annotations for g_file_{get,set}_contents
These are actually byte arrays.
Colin Walters [Tue, 31 May 2011 14:35:38 +0000 (10:35 -0400)]
Consistently use G_GNUC_EXTENSION instead of __extension__
g-ir-scanner doesn't like __extension__, and while I can add it, since
we have this macro, we should be using it consistently.
muralis [Wed, 25 May 2011 08:12:59 +0000 (13:42 +0530)]
gthreadedresolver: fix hang on g_thread_pool_push() failure
In resolve_sync function in gthreadedresolver.c, if g_thread_pool_push
fails due to thread creation failure, we are just simply appending the
data to the queue of work to do. After the failure, we might wait
indefinitely in g_cond_wait. In case of g_thread_pool_push failure,
propagate the error so that this function does not blocks forever in
case of failure.
https://bugzilla.gnome.org/show_bug.cgi?id=651034
Carles Ferrando [Sun, 29 May 2011 12:04:16 +0000 (14:04 +0200)]
[l10n]Updated Catalan (Valencian) translation
Jorge González [Sun, 29 May 2011 07:47:30 +0000 (09:47 +0200)]
Updated Spanish translation
Matthias Clasen [Sun, 29 May 2011 04:05:07 +0000 (00:05 -0400)]
Documentation fixups
Matthias Clasen [Sun, 29 May 2011 03:27:24 +0000 (23:27 -0400)]
Clean up includes
Matthias Clasen [Sun, 29 May 2011 03:15:07 +0000 (23:15 -0400)]
Reduce excessive whitespace
Matthias Clasen [Sun, 29 May 2011 02:49:59 +0000 (22:49 -0400)]
Remove redundant comments
The comments in gthreadpool.h were duplicating the
doc comments in the sources.
Matthias Clasen [Sun, 29 May 2011 02:33:37 +0000 (22:33 -0400)]
Volatile not necessary after all
I was temporarily confused.
Matthias Clasen [Sun, 29 May 2011 02:03:05 +0000 (22:03 -0400)]
GTree: Declare ref_count as volatile
Matthias Clasen [Sun, 29 May 2011 02:02:13 +0000 (22:02 -0400)]
GAsyncQueue: Make ref_count a volatile gint
Matthias Clasen [Sun, 29 May 2011 01:43:45 +0000 (21:43 -0400)]
GAsyncQueue: Remove excessive atomic operations
Almost every function was double-checking the ref count,
unnecessarily.
Matthias Clasen [Sun, 29 May 2011 01:41:08 +0000 (21:41 -0400)]
GArray: Avoid unncessary atomic accesses to refcounts
Not really necessary to double-check the ref-count.