platform/upstream/glib.git
12 years agoFix malformed GTK-Doc comment blocks: mutliline annotations are invalid.
Dieter Verfaillie [Mon, 2 Apr 2012 15:02:15 +0000 (17:02 +0200)]
Fix malformed GTK-Doc comment blocks: mutliline annotations are invalid.

Found these thanks to improved gobject-introspection GTK-Doc
comment block/annotation parser from:
https://bugzilla.gnome.org/show_bug.cgi?id=672254

https://bugzilla.gnome.org/show_bug.cgi?id=673385

12 years agoFix malformed GTK-Doc comment blocks: unmark non GTK-Doc comment block.
Dieter Verfaillie [Mon, 2 Apr 2012 15:01:32 +0000 (17:01 +0200)]
Fix malformed GTK-Doc comment blocks: unmark non GTK-Doc comment block.

Found these thanks to improved gobject-introspection GTK-Doc
comment block/annotation parser from:
https://bugzilla.gnome.org/show_bug.cgi?id=672254

https://bugzilla.gnome.org/show_bug.cgi?id=673385

12 years agoFix malformed GTK-Doc comment blocks: add missing colons.
Dieter Verfaillie [Mon, 2 Apr 2012 15:16:35 +0000 (17:16 +0200)]
Fix malformed GTK-Doc comment blocks: add missing colons.

Found these thanks to improved gobject-introspection GTK-Doc
comment block/annotation parser from:
https://bugzilla.gnome.org/show_bug.cgi?id=672254

https://bugzilla.gnome.org/show_bug.cgi?id=673385

12 years agoUpdated Belarusian translation.
Ihar Hrachyshka [Thu, 5 Apr 2012 12:25:42 +0000 (15:25 +0300)]
Updated Belarusian translation.

12 years agogobject/tests/signals.c: Remove C99ism
Chun-wei Fan [Thu, 5 Apr 2012 07:46:34 +0000 (15:46 +0800)]
gobject/tests/signals.c: Remove C99ism

Declare variables in start of block.

12 years agoUpdate GLib Visual C++ projects
Chun-wei Fan [Thu, 5 Apr 2012 07:45:38 +0000 (15:45 +0800)]
Update GLib Visual C++ projects

Some of the bundled PCRE source files are removed as they are not needed
anymore (please see commit e7c0f10e), so don't include them in the projects

12 years agoFix some gtk-doc stuff
Dan Winship [Wed, 4 Apr 2012 19:22:15 +0000 (15:22 -0400)]
Fix some gtk-doc stuff

12 years agoSwitch to using a mainloop when waiting for connection close
Matthias Clasen [Wed, 4 Apr 2012 18:56:17 +0000 (14:56 -0400)]
Switch to using a mainloop when waiting for connection close

GDBus heavily relies on idles for some of its cleanup operations,
and not running a mainloop leads to things not getting cleaned
up properly, which in turn leads to test failures, since the
session bus singleton does not get removed.

12 years agoImprove GApplication tests
Matthias Clasen [Wed, 4 Apr 2012 18:39:17 +0000 (14:39 -0400)]
Improve GApplication tests

Make GApplication tests manage their dbus connections.

12 years agoMake basic-application test more robust
Matthias Clasen [Wed, 4 Apr 2012 10:38:36 +0000 (06:38 -0400)]
Make basic-application test more robust

This program is only used indirectly from gapplication.c in
tests, but that is no reason to let it segfault when it is
run from the commandline without arguments.

12 years agoGActionMap: correct critical warnings
Matthias Clasen [Wed, 4 Apr 2012 10:36:45 +0000 (06:36 -0400)]
GActionMap: correct critical warnings

These warnings were copied from GSimpleActionGroup, but the function
names were not updated. Correct that.

12 years agoAdd tests for action-related gapplication api
Matthias Clasen [Wed, 4 Apr 2012 01:52:45 +0000 (21:52 -0400)]
Add tests for action-related gapplication api

12 years agoAdd a test for g_application_quit
Matthias Clasen [Wed, 4 Apr 2012 01:39:06 +0000 (21:39 -0400)]
Add a  test for g_application_quit

12 years agoImprove keyfile test coverage
Matthias Clasen [Tue, 3 Apr 2012 05:43:45 +0000 (01:43 -0400)]
Improve keyfile test coverage

12 years agoRemove a bit of unused code
Matthias Clasen [Tue, 3 Apr 2012 05:42:21 +0000 (01:42 -0400)]
Remove a bit of unused code

_g_utf8_strwidth is always called with max == -1, so remove the
parameter and update all callers.

12 years agoImprove test coverage
Matthias Clasen [Tue, 3 Apr 2012 05:09:39 +0000 (01:09 -0400)]
Improve test coverage

Add tests for all the suffixes in g_format_size.

12 years agotests/gdatetime: fix new_from_unix test
Dan Winship [Wed, 4 Apr 2012 18:38:18 +0000 (14:38 -0400)]
tests/gdatetime: fix new_from_unix test

When creating a struct tm for "1990-01-01T00:00:00" to pass to
mktime(), we have to set tm_isdst to -1; leaving it set to 0 will
result in the wrong time being generated when run in a timezone where
January 1 would normally be tm_isdst==1 (ie, in southern hemisphere
DST-observing countries, like Australia).

https://bugzilla.gnome.org/show_bug.cgi?id=670254

12 years agoUpdated Spanish translation
Daniel Mustieles [Wed, 4 Apr 2012 15:18:59 +0000 (17:18 +0200)]
Updated Spanish translation

12 years agoUpdated Hungarian translation
Gabor Kelemen [Wed, 4 Apr 2012 15:02:04 +0000 (17:02 +0200)]
Updated Hungarian translation

12 years agoFix warnings caused by an ifdeffed-out test case
Dan Winship [Wed, 4 Apr 2012 14:47:04 +0000 (10:47 -0400)]
Fix warnings caused by an ifdeffed-out test case

12 years agoUpdated Slovenian translation
Matej Urbančič [Wed, 4 Apr 2012 11:42:44 +0000 (13:42 +0200)]
Updated Slovenian translation

12 years agoUse 'if test' for checking auto* variables' values
Rodrigo Moya [Wed, 4 Apr 2012 09:55:40 +0000 (11:55 +0200)]
Use 'if test' for checking auto* variables' values

12 years agoAlways include G_SOCKET_FAMILY_UNIX value in GSocketFamily
Rodrigo Moya [Wed, 4 Apr 2012 09:39:03 +0000 (11:39 +0200)]
Always include G_SOCKET_FAMILY_UNIX value in GSocketFamily

This is needed because glib-mkenums doesn't handle #ifdef values in
enums, and so it needs to have all values always defined in the enum.
When not available, define the missing values to a negative value.

12 years agouse GLIB_DEPRECATED_IN_2_32_FOR to deprecate g_value_{g,s}et_schar
Guillaume Desmottes [Tue, 3 Apr 2012 14:37:22 +0000 (16:37 +0200)]
use GLIB_DEPRECATED_IN_2_32_FOR to deprecate g_value_{g,s}et_schar

https://bugzilla.gnome.org/show_bug.cgi?id=673439

12 years agog_input_stream_read[_finish]: document returning 0 on EOF
Will Thompson [Fri, 30 Mar 2012 13:44:05 +0000 (14:44 +0100)]
g_input_stream_read[_finish]: document returning 0 on EOF

g_input_stream_read() does state that it returns 0 on end of file, but
not in the Returns: line, so it's easy to miss on a quick skim-read.

g_input_stream_read_async() documents that g_input_stream_read_finish()
returns 0 on end of file, but g_input_stream_read_finish() itself does
not.

https://bugzilla.gnome.org/show_bug.cgi?id=673174

12 years agoRework the libelf configure checks one more time
Matthias Clasen [Mon, 2 Apr 2012 12:53:20 +0000 (08:53 -0400)]
Rework the libelf configure checks one more time

It seems that there is quite a bit of variation out there, in
terms of libelf versions and API. Make the checks more thorough,
by not only checking for elf_begin, but also some of the shdr function
that we need. Also, explicitly check for libelf.h.

This should address bug 673132 and 673253.

12 years agoUse new hashset api in the hash set tests
Matthias Clasen [Mon, 2 Apr 2012 12:00:48 +0000 (08:00 -0400)]
Use new hashset api in the hash set tests

12 years agoDispose test object
Matthias Clasen [Mon, 2 Apr 2012 11:59:55 +0000 (07:59 -0400)]
Dispose test object

While not very important, it means one less untested function
in the coverage report.

12 years agoDrop g_thread_init from tests
Matthias Clasen [Mon, 2 Apr 2012 11:59:23 +0000 (07:59 -0400)]
Drop g_thread_init from tests

This is not needed anymore.

12 years agoImprove GBusObjectManager test coverage
Matthias Clasen [Mon, 2 Apr 2012 11:37:55 +0000 (07:37 -0400)]
Improve GBusObjectManager test coverage

12 years agodocs: Fix typo in glib-compile-schemas man page
David King [Mon, 2 Apr 2012 09:22:56 +0000 (10:22 +0100)]
docs: Fix typo in glib-compile-schemas man page

12 years agoImprove boxed test coverage
Matthias Clasen [Mon, 2 Apr 2012 03:04:59 +0000 (23:04 -0400)]
Improve boxed test coverage

This commit add basic tests for all GLib types that are registered
as boxed types.

12 years agoDrop some unused PCRE sources
Matthias Clasen [Mon, 2 Apr 2012 02:04:08 +0000 (22:04 -0400)]
Drop some unused PCRE sources

There is no point in compiling these, since we are not calling
any of the functions in them.

12 years agoUpdated Swedish translation
Daniel Nylander [Sun, 1 Apr 2012 11:11:33 +0000 (13:11 +0200)]
Updated Swedish translation

12 years agogthread-posix: fix typo in error message
Antoine Jacoutot [Sat, 31 Mar 2012 20:48:42 +0000 (22:48 +0200)]
gthread-posix: fix typo in error message

12 years agogresource: libelf, try pkg-config first then fall-back to AC_CHECK_LIB
Antoine Jacoutot [Sat, 31 Mar 2012 18:20:19 +0000 (20:20 +0200)]
gresource: libelf, try pkg-config first then fall-back to AC_CHECK_LIB

On some systems gelf.h may not be stored under the top level include
directory in which case we need to add the correct include paths in
cflags by using pkg-config(1).

12 years agoAdd missing allow-none annotations for function parameters.
Robert Ancell [Sat, 24 Mar 2012 12:58:45 +0000 (23:58 +1100)]
Add missing allow-none annotations for function parameters.

Found using:
find . -name '*.c' | xargs grep 'or %NULL' | grep ' \* @' | grep -v '@error' | grep -v allow-none

12 years agoFix a link
Matthias Clasen [Fri, 30 Mar 2012 21:34:37 +0000 (17:34 -0400)]
Fix a link

As pointed out in bug 673139, the link to the mailing list
information page was outdated.

12 years agoUpdated Odia Translation
ManojKumar Giri [Fri, 30 Mar 2012 12:38:23 +0000 (18:08 +0530)]
Updated Odia Translation

12 years agoresources: compiler: Respect absolute paths
Christian Persch [Thu, 22 Mar 2012 13:01:40 +0000 (14:01 +0100)]
resources: compiler: Respect absolute paths

When a <file> already has an absolute path, use it directly instead of trying to
locate it in the --sourcedir directories.

https://bugzilla.gnome.org/show_bug.cgi?id=672541

12 years agoconfigure: Fix alignment tests when cross compiling
Kalev Lember [Wed, 28 Mar 2012 09:42:49 +0000 (12:42 +0300)]
configure: Fix alignment tests when cross compiling

When inserting custom code to AC_CHECK_ALIGNOF, make sure to not replace
the default includes, but instead append to them.

This fixes ALIGNOF_GUINT32 and ALIGNOF_GUINT64 that were both 0 when cross
compiling. The third 'unsigned long' test wasn't affected because the
AC_CHECK_ALIGNOF call didn't specify the optional 2nd parameter.

12 years agoFix gspawn-win32-helper Visual C++ projects
Chun-wei Fan [Wed, 28 Mar 2012 07:12:04 +0000 (15:12 +0800)]
Fix gspawn-win32-helper Visual C++ projects

The name of the executable for the Release|Win32 and Release|x64 configs
were mixed up.  Fix that.  DOH! for not noting this earlier.

12 years agoFix VS property sheets
Chun-wei Fan [Wed, 28 Mar 2012 06:57:12 +0000 (14:57 +0800)]
Fix VS property sheets

We need to accomodate for gspawn-win64-helper-console.exe for the "install"
phase too.

12 years agoUpdated Spanish translation
Daniel Mustieles [Tue, 27 Mar 2012 19:24:37 +0000 (21:24 +0200)]
Updated Spanish translation

12 years agoGFileMonitor: fix docs bug
Dan Winship [Tue, 27 Mar 2012 16:45:27 +0000 (12:45 -0400)]
GFileMonitor: fix docs bug

12 years agoUpdated Marathi Translations
Sandeep Sheshrao Shedmake [Tue, 27 Mar 2012 09:34:14 +0000 (15:04 +0530)]
Updated Marathi Translations

12 years agoall: s/availible/available/
Bastien Nocera [Tue, 27 Mar 2012 09:01:00 +0000 (11:01 +0200)]
all: s/availible/available/

12 years agohindi translation by Chandan Kumar
Rajesh Ranjan [Tue, 27 Mar 2012 06:52:39 +0000 (12:22 +0530)]
hindi translation by Chandan Kumar

12 years agoUpdated Marathi Translations
Sandeep Sheshrao Shedmake [Mon, 26 Mar 2012 14:18:20 +0000 (19:48 +0530)]
Updated Marathi Translations

12 years agoUpdated Odia Translation
ManojKumar Giri [Mon, 26 Mar 2012 13:12:41 +0000 (18:42 +0530)]
Updated Odia Translation

12 years agoUpdated Czech translation
Marek Černocký [Mon, 26 Mar 2012 12:49:18 +0000 (14:49 +0200)]
Updated Czech translation

12 years agoUpdated Latvian translation.
Rudolfs Mazurs [Sun, 25 Mar 2012 12:05:26 +0000 (15:05 +0300)]
Updated Latvian translation.

12 years ago[l10n] Updated Italian translation
Luca Ferretti [Sun, 25 Mar 2012 11:14:47 +0000 (13:14 +0200)]
[l10n] Updated Italian translation

12 years agoUpdated Kannada Translation
Shankar Prasad [Sun, 25 Mar 2012 07:56:17 +0000 (13:26 +0530)]
Updated Kannada Translation

12 years agoUpdated Kannada Translation
Shankar Prasad [Sun, 25 Mar 2012 07:54:31 +0000 (13:24 +0530)]
Updated Kannada Translation

12 years agoUpdated Kannada Translation
Shankar Prasad [Sun, 25 Mar 2012 07:33:57 +0000 (13:03 +0530)]
Updated Kannada Translation

12 years agoUpdated Kannada Translation
Shankar Prasad [Sun, 25 Mar 2012 07:12:30 +0000 (12:42 +0530)]
Updated Kannada Translation

12 years agoAdd a forgotten include guard
Matthias Clasen [Sun, 25 Mar 2012 03:00:15 +0000 (23:00 -0400)]
Add a forgotten include guard

gbytes.h was missing the single-include guard.

12 years agoUpdated Kannada Translation
Shankar Prasad [Sat, 24 Mar 2012 19:03:10 +0000 (00:33 +0530)]
Updated Kannada Translation

12 years agoUpdated Swedish translation
Daniel Nylander [Sat, 24 Mar 2012 15:58:42 +0000 (16:58 +0100)]
Updated Swedish translation

12 years agoBump version
Matthias Clasen [Sat, 24 Mar 2012 15:31:01 +0000 (11:31 -0400)]
Bump version

12 years ago2.32.0 2.32.0
Matthias Clasen [Sat, 24 Mar 2012 15:28:35 +0000 (11:28 -0400)]
2.32.0

12 years agoRelease notes tweaks
Matthias Clasen [Sat, 24 Mar 2012 14:33:38 +0000 (10:33 -0400)]
Release notes tweaks

Update download locations, mailing list instructions, and shuffle
the version-specific notes to be at the end.

12 years agoMake the `include' glib test case build on FreeBSD.
Raphael Kubo da Costa [Mon, 19 Mar 2012 17:20:08 +0000 (14:20 -0300)]
Make the `include' glib test case build on FreeBSD.

Defining _POSIX_C_SOURCE to 0 will make time.h not create the clockid_t
typedef used by some functions in pthread.h.

The right approach here is to set it to 199309L, which creates the
typedef on FreeBSD and doesn't set __USE_UNIX98 or __USE_XOPEN2K on
glibc, which is what the test is actually testing.

https://bugzilla.gnome.org/show_bug.cgi?id=672406

12 years agoUpdated Telugu Translations
Krishnababu Krothapalli [Thu, 22 Mar 2012 14:26:06 +0000 (19:56 +0530)]
Updated Telugu Translations

12 years agoglib/gqsort.c: Fix C99ism/GCCism
Chun-wei Fan [Tue, 20 Mar 2012 05:19:11 +0000 (13:19 +0800)]
glib/gqsort.c: Fix C99ism/GCCism

-There were a number of variables that were declared in the middle of
 the block, so move these declarations to the start of the block
-There was a use of mempcpy, but it is a GCC extension, so use memcpy since
 we didn't care about the return value of the call to mempcpy.

https://bugzilla.gnome.org/show_bug.cgi?id=672095

12 years agoFix the tests on win32
Alexander Larsson [Tue, 20 Mar 2012 09:20:18 +0000 (10:20 +0100)]
Fix the tests on win32

There was some unix-specific pathname handling in the test utils
which broke the css tests on win32.

12 years agoUse g_unsetenv, not unsetenv
Alexander Larsson [Tue, 20 Mar 2012 09:19:42 +0000 (10:19 +0100)]
Use g_unsetenv, not unsetenv

This fixed the win32 build

12 years agoOnly build actions test on unix
Alexander Larsson [Tue, 20 Mar 2012 09:18:32 +0000 (10:18 +0100)]
Only build actions test on unix

It depends on gdbus-sessionbus.c which only builds on unix.

12 years agoGDBusConnection: Avoid dereferencing vtable for unregistered objects
Marco Trevisan (Treviño) [Mon, 19 Mar 2012 19:13:24 +0000 (20:13 +0100)]
GDBusConnection: Avoid dereferencing vtable for unregistered objects

https://bugzilla.gnome.org/show_bug.cgi?id=671988

Signed-off-by: David Zeuthen <davidz@redhat.com>
12 years agobump version
Matthias Clasen [Mon, 19 Mar 2012 20:59:54 +0000 (16:59 -0400)]
bump version

12 years agoFix distcheck 2.31.22
Matthias Clasen [Mon, 19 Mar 2012 20:58:48 +0000 (16:58 -0400)]
Fix distcheck

automake decided to complain about some more leftovers.

12 years ago2.31.22
Matthias Clasen [Mon, 19 Mar 2012 18:27:20 +0000 (14:27 -0400)]
2.31.22

12 years agoUpdates
Matthias Clasen [Mon, 19 Mar 2012 17:43:31 +0000 (13:43 -0400)]
Updates

12 years agoDon't return from a void function
Matthias Clasen [Mon, 19 Mar 2012 17:28:14 +0000 (13:28 -0400)]
Don't return from a void function

12 years agoFix a C99ism
Matthias Clasen [Mon, 19 Mar 2012 17:26:19 +0000 (13:26 -0400)]
Fix a C99ism

12 years agoSilence a compiler warning
Matthias Clasen [Mon, 19 Mar 2012 17:22:04 +0000 (13:22 -0400)]
Silence a compiler warning

12 years agoGDBus: set no-reply flag on calls with no callback
Ryan Lortie [Fri, 16 Mar 2012 17:32:38 +0000 (13:32 -0400)]
GDBus: set no-reply flag on calls with no callback

If g_dbus_connection_call() or g_dbus_proxy_call() are given a NULL
callback then set the no-reply flag on the outgoing D-Bus message.

https://bugzilla.gnome.org/show_bug.cgi?id=672239

12 years agoGDBusProxy: report an error instead of leaking it
Ryan Lortie [Fri, 16 Mar 2012 17:15:46 +0000 (13:15 -0400)]
GDBusProxy: report an error instead of leaking it

GDBusProxy sets an error on a GSimpleAsyncResult and then returns
without dispatching the result for completion (and leaks the result in
the process).  Fix that.

Also add a testcase.  Unfortunately, adding the testcase uncovered
bug #672248.  We can work around that by reordering the tests.

https://bugzilla.gnome.org/show_bug.cgi?id=672249

12 years agoUpdate config.h.win32(.in)
Chun-wei Fan [Mon, 19 Mar 2012 08:02:37 +0000 (16:02 +0800)]
Update config.h.win32(.in)

Make it more like the one that is generated by autotools.

It is true that Visual C++ has sig_atomic_t, at least for Visual C++ 2008
and later, but this is currently only used for UNIX builds of GLib, as a
point of note here.

12 years agoUpdate glibconfig.h.win32(.in)
Chun-wei Fan [Mon, 19 Mar 2012 07:55:20 +0000 (15:55 +0800)]
Update glibconfig.h.win32(.in)

Added the definitions below, as these #defines are needed for gqsort.c

#define ALIGNOF_GUINT32 4
#define ALIGNOF_GUINT64 8
#define ALIGNOF_UNSIGNED_LONG 4

12 years agoUpdated French translation
Bruno Brouard [Sun, 18 Mar 2012 09:34:15 +0000 (10:34 +0100)]
Updated French translation

12 years agoMake inf testcase more forgiving
Matthias Clasen [Sun, 18 Mar 2012 03:43:52 +0000 (23:43 -0400)]
Make inf testcase more forgiving

printf may produce "infinity" instead of "inf".

https://bugzilla.gnome.org/show_bug.cgi?id=668973

12 years agoUse __builtin_bswap* for GUINT*_SWAP_LE_BE if building with GCC
Sebastian Dröge [Sun, 18 Mar 2012 03:28:48 +0000 (23:28 -0400)]
Use __builtin_bswap* for GUINT*_SWAP_LE_BE if building with GCC

https://bugzilla.gnome.org/show_bug.cgi?id=531901

12 years agoFix out of tree build for MinGW/MSYS
Dieter Verfaillie [Sun, 18 Mar 2012 03:10:01 +0000 (23:10 -0400)]
Fix out of tree build for MinGW/MSYS

When building with MinGW/MSYS with srcdir != builddir the build fails:
- to locate the generated .def files
- creating libglib-gdb.py
- creating libgobject-gdb.py

Solved this by explicitly instructing these files to be generated
in $(builddir)/...

https://bugzilla.gnome.org/show_bug.cgi?id=653167

12 years agoUpdated Portuguese translation
Duarte Loreto [Sun, 18 Mar 2012 00:45:43 +0000 (00:45 +0000)]
Updated Portuguese translation

12 years agoUpdated Lithuanian translation
Aurimas Černius [Sat, 17 Mar 2012 16:49:25 +0000 (18:49 +0200)]
Updated Lithuanian translation

12 years agoman pages: document the order of schema overrides
Ryan Lortie [Sat, 17 Mar 2012 16:39:23 +0000 (12:39 -0400)]
man pages: document the order of schema overrides

GSettings overrides are processed in such a way that
alphabetically-later files have precedence over earlier files (eg: 20_
will beat 10_).  Document that fact.

12 years agoUpdated Russian translation
Yuri Kozlov [Sat, 17 Mar 2012 08:57:44 +0000 (12:57 +0400)]
Updated Russian translation

12 years agogmain: Use sig_atomic_t for list of pending Unix signals
Colin Walters [Tue, 13 Mar 2012 18:49:04 +0000 (14:49 -0400)]
gmain: Use sig_atomic_t for list of pending Unix signals

Pointed out by: Simon McVittie <simon.mcvittie@collabora.co.uk>

https://bugzilla.gnome.org/show_bug.cgi?id=671997

12 years agogio: Only show mounts in /run/media/$USER
Tomas Bzatek [Fri, 16 Mar 2012 17:54:09 +0000 (18:54 +0100)]
gio: Only show mounts in /run/media/$USER

And reflect the recent changes + sync with gvfs+udisks2 rules.

https://bugzilla.gnome.org/show_bug.cgi?id=669797

12 years agoContinue searching in PATH on ENODEV and ETIMEDOUT
David Benjamin [Thu, 15 Mar 2012 23:16:02 +0000 (19:16 -0400)]
Continue searching in PATH on ENODEV and ETIMEDOUT

Synchronize the error-handling logic from glibc. Users with networked
filesystems in their PATH may hit ETIMEDOUT.

https://bugzilla.gnome.org/show_bug.cgi?id=672201

12 years agogtype: Note idempotency of g_type_init()
Colin Walters [Thu, 15 Mar 2012 21:59:01 +0000 (17:59 -0400)]
gtype: Note idempotency of g_type_init()

12 years agoRemove now unused qsort_r checks
Alexander Larsson [Fri, 16 Mar 2012 11:51:33 +0000 (12:51 +0100)]
Remove now unused qsort_r checks

https://bugzilla.gnome.org/show_bug.cgi?id=672095

12 years agoMake g_array_sort* methods use a stable sort
Alexander Larsson [Wed, 14 Mar 2012 20:17:47 +0000 (21:17 +0100)]
Make g_array_sort* methods use a stable sort

Also, remove previous comments about sort stability in g_array_sort docs,
as the method that was explained does not work. Adds a new comment
about this.

https://bugzilla.gnome.org/show_bug.cgi?id=672095

12 years agoMake g_qsort_with_data stable, based on glibc msort
Alexander Larsson [Wed, 14 Mar 2012 20:17:23 +0000 (21:17 +0100)]
Make g_qsort_with_data stable, based on glibc msort

We need a stable sort, and we might as well always use it rather
than have multiple sort versions. This picks up the glibc
merge sort implementation which it uses by default for qsort,
except we don't fall back to non-stable quicksort in some cases
like glibc

https://bugzilla.gnome.org/show_bug.cgi?id=672095

12 years agoUpdated Hungarian translation
Gabor Kelemen [Fri, 16 Mar 2012 10:12:56 +0000 (11:12 +0100)]
Updated Hungarian translation

12 years ago[l10n]Updated Catalan (Valencian) translation
Carles Ferrando [Thu, 15 Mar 2012 22:33:50 +0000 (23:33 +0100)]
[l10n]Updated Catalan (Valencian) translation

12 years ago[l10n]Updated Catalan translation
Gil Forcada [Thu, 15 Mar 2012 22:33:43 +0000 (23:33 +0100)]
[l10n]Updated Catalan translation

12 years agoGDBusConnection: Don't store GCancellable in qdata
David Zeuthen [Thu, 15 Mar 2012 18:04:39 +0000 (14:04 -0400)]
GDBusConnection: Don't store GCancellable in qdata

Now that we're using g_simple_async_result_set_check_cancellable() we
no longer need this terrible hack of carrying the GCancellable on the
GSimpleAsyncResult using qdata. See bug 672013 for more details.

https://bugzilla.gnome.org/show_bug.cgi?id=672013

Signed-off-by: David Zeuthen <davidz@redhat.com>