platform/upstream/glib.git
15 years agoFix GNetworkAddress skipping addresses when enumerating
Sjoerd Simons [Wed, 2 Sep 2009 16:48:09 +0000 (17:48 +0100)]
Fix GNetworkAddress skipping addresses when enumerating

g_network_address_address_enumerator_next_finish takes the first item of the
address list and moves the pointer to the next one, so we shouldn't do the same
in g_network_address_address_enumerator_next_async function

Fixes bug #593941

15 years agoMake the error const for g_simple_async_result_set_from_error
Richard Hughes [Wed, 2 Sep 2009 18:15:38 +0000 (19:15 +0100)]
Make the error const for g_simple_async_result_set_from_error

15 years agog_socket_send_message() fails due to invalid sendmsg(2) params.
James Hunt [Thu, 10 Sep 2009 16:18:13 +0000 (17:18 +0100)]
g_socket_send_message() fails due to invalid sendmsg(2) params.

g_socket_send_message() and g_socket_send_to() fail with ENOBUFS or
EFAULT due to the fact that if no "address" argument is specified to
g_socket_send_message, when g_socket_send_message() calls sendmsg(2),
the 2nd parameter to sendmsg ("const struct msghdr *msg") contains
uninitialized values. The fix is simple - initialize msg.msg_name to
NULL and msg.msg_msg_namelen to 0.

https://bugzilla.gnome.org/show_bug.cgi?id=594759

15 years agohindi update by Rajesh Ranjan
Rajesh Ranjan [Fri, 11 Sep 2009 10:30:29 +0000 (16:00 +0530)]
hindi update by Rajesh Ranjan

15 years agoUpdated Malayalam Translations
Ani [Fri, 11 Sep 2009 06:48:50 +0000 (12:18 +0530)]
Updated Malayalam Translations

15 years agoUpdated Ukrainian translation
Maxim V. Dziumanenko [Thu, 10 Sep 2009 22:48:56 +0000 (01:48 +0300)]
Updated Ukrainian translation

15 years agoBug 579050 Allow making selected critical and warning messages non-fatal
Christian Dywan [Thu, 10 Sep 2009 14:40:11 +0000 (16:40 +0200)]
Bug 579050 Allow making selected critical and warning messages non-fatal

Implement g_test_log_set_fatal_handler which is a function similar to
g_log_set_default_handler but for use in unit tests where certain
errors have to be ignored because it is not possible to fix or avoid
them otherwise. A unit test is added.

15 years agoUpdated Kannada(kn) translation
Shankar Prasad [Thu, 10 Sep 2009 06:56:13 +0000 (12:26 +0530)]
Updated Kannada(kn) translation

15 years agoUpdated Greek translation.
Kostas Papadimas [Wed, 9 Sep 2009 15:38:12 +0000 (18:38 +0300)]
Updated Greek translation.

15 years agoUpdated Gujarati Translations
Sweta Kothari [Wed, 9 Sep 2009 11:24:43 +0000 (16:54 +0530)]
Updated Gujarati Translations

15 years agoClarify g_ptr_array_set_size() docs re: free_func
Dan Winship [Tue, 8 Sep 2009 15:19:13 +0000 (11:19 -0400)]
Clarify g_ptr_array_set_size() docs re: free_func

15 years agoUpdated Bulgarian translation
Alexander Shopov [Tue, 8 Sep 2009 04:41:28 +0000 (07:41 +0300)]
Updated Bulgarian translation

15 years agoCall element_free_func when shrinking array with g_ptr_array_set_size
Dan Winship [Sun, 6 Sep 2009 17:54:50 +0000 (13:54 -0400)]
Call element_free_func when shrinking array with g_ptr_array_set_size

15 years agoUpdated Marathi Translations
Sandeep Shedmake [Mon, 7 Sep 2009 13:44:52 +0000 (19:14 +0530)]
Updated Marathi Translations

15 years agoAvoid reading uninitialized memory
Alexander Larsson [Mon, 7 Sep 2009 10:50:58 +0000 (12:50 +0200)]
Avoid reading uninitialized memory

If the statfs call fails, don't look at the result.

15 years agoRemove warning in g_simple_async_result_complete
Alexander Larsson [Mon, 7 Sep 2009 08:25:14 +0000 (10:25 +0200)]
Remove warning in g_simple_async_result_complete

This warning hits code that uses GSimpleAsyncResult outside of a
mainloop as a helper object. For instance EggDBus does this.
Since the bugs this warning would fix are pretty easy to spot
and since EggDBus is deployed already we just remove the
"called from outside main loop" warning.

However, we need to keep the "called from wrong context" warning
as that is very helpful when debugging misuse of the new multiple
main context code.

15 years agoAnother dead assignment
Matthias Clasen [Mon, 7 Sep 2009 07:28:35 +0000 (03:28 -0400)]
Another dead assignment

15 years agoRemove a dead initialization
Matthias Clasen [Mon, 7 Sep 2009 07:14:15 +0000 (03:14 -0400)]
Remove a dead initialization

15 years agoRemove a dead increment
Matthias Clasen [Mon, 7 Sep 2009 07:12:06 +0000 (03:12 -0400)]
Remove a dead increment

15 years agoRemove another dead assignment
Matthias Clasen [Mon, 7 Sep 2009 07:10:01 +0000 (03:10 -0400)]
Remove another dead assignment

15 years agoRemmove a dead assignment.
Matthias Clasen [Mon, 7 Sep 2009 07:07:22 +0000 (03:07 -0400)]
Remmove a dead assignment.

Spotted by clang.

15 years agoMove a assignment to the relevant #ifdef branch
Matthias Clasen [Mon, 7 Sep 2009 07:02:58 +0000 (03:02 -0400)]
Move a assignment to the relevant #ifdef branch

Dead code spotted by clang.

15 years agoRemove dead code
Matthias Clasen [Mon, 7 Sep 2009 07:01:02 +0000 (03:01 -0400)]
Remove dead code

15 years agoDon't evaluate an uninitialized value
Matthias Clasen [Mon, 7 Sep 2009 06:50:51 +0000 (02:50 -0400)]
Don't evaluate an uninitialized value

Bug found by clang.

15 years agoUpdated Polish translation
Tomasz Dominikowski [Sun, 6 Sep 2009 12:20:45 +0000 (14:20 +0200)]
Updated Polish translation

15 years agoUpdate Japanese translation
Takayuki KUSANO [Sat, 5 Sep 2009 13:34:58 +0000 (22:34 +0900)]
Update Japanese translation

15 years agoUpdated Spanish translation
Jorge González [Sat, 5 Sep 2009 11:47:41 +0000 (13:47 +0200)]
Updated Spanish translation

15 years agoBump version
Matthias Clasen [Sat, 5 Sep 2009 01:20:09 +0000 (21:20 -0400)]
Bump version

15 years ago2.21.6 2.21.6
Matthias Clasen [Sat, 5 Sep 2009 01:19:02 +0000 (21:19 -0400)]
2.21.6

15 years agoAdd a Since: tag
Matthias Clasen [Sat, 5 Sep 2009 00:41:25 +0000 (20:41 -0400)]
Add a Since: tag

15 years agoUpdates
Matthias Clasen [Sat, 5 Sep 2009 00:26:26 +0000 (20:26 -0400)]
Updates

15 years agoBug 594034 - Use g_mkstemp_full() when creating the replacement file
Benjamin Otte [Thu, 3 Sep 2009 13:39:57 +0000 (15:39 +0200)]
Bug 594034 - Use g_mkstemp_full() when creating the replacement file

Previous code used g_mkstemp(). But when using
G_FILE_CREATE_REPLACE_DESTINATION, no attempt was made to ensure proper
mode and flags of the created temporary file. The visible issue was that
the file was always created with mode 0600 as opposed to using 0666.
(The invisible issue was that O_RDWR was used instead of O_WRONLY.)

15 years agoBug 594034 - Add g_mkstemp_full()
Benjamin Otte [Thu, 3 Sep 2009 13:36:37 +0000 (15:36 +0200)]
Bug 594034 - Add g_mkstemp_full()

This function exposes more variables than g_mkstemp() and therefor
allows more flexibility when creating temporary files.
The intended use is gio's code for g_file_replace() (see next patch)

15 years agoUpdating Estonian translation
Ivar Smolin [Fri, 4 Sep 2009 15:58:00 +0000 (18:58 +0300)]
Updating Estonian translation

15 years agoUpdated German translation
Christian Kirbach [Fri, 4 Sep 2009 13:16:04 +0000 (15:16 +0200)]
Updated German translation

15 years agoUpdated Swedish translation
Daniel Nylander [Fri, 4 Sep 2009 04:31:34 +0000 (06:31 +0200)]
Updated Swedish translation

15 years agoUpdated French translation
Claude Paroz [Thu, 3 Sep 2009 22:07:40 +0000 (00:07 +0200)]
Updated French translation

15 years agoUpdated Brazilian Portuguese translation.
Fábio Nogueira [Wed, 2 Sep 2009 16:28:59 +0000 (12:28 -0400)]
Updated Brazilian Portuguese translation.

15 years agoUpdated Basque language
Inaki Larranaga Murgoitio [Wed, 2 Sep 2009 16:19:27 +0000 (18:19 +0200)]
Updated Basque language

15 years agoHungarian translation updated
Gabor Kelemen [Wed, 2 Sep 2009 16:05:23 +0000 (18:05 +0200)]
Hungarian translation updated

15 years agoUpdated Portuguese translation
Duarte Loreto [Tue, 1 Sep 2009 22:34:41 +0000 (23:34 +0100)]
Updated Portuguese translation

15 years agoUpdated Italian translation
Luca Ferretti [Tue, 1 Sep 2009 21:47:14 +0000 (23:47 +0200)]
Updated Italian translation

15 years agoUse lchmod instead of stat + chown if available
Benjamin Otte [Tue, 1 Sep 2009 19:53:35 +0000 (21:53 +0200)]
Use lchmod instead of stat + chown if available

Fallout of the NOFLOOW_SYMLINKS fix from bug 593406

15 years agoBug 593406 - Permissions set to 777 after copying via Nautilus
Benjamin Otte [Tue, 1 Sep 2009 19:26:08 +0000 (21:26 +0200)]
Bug 593406 - Permissions set to 777 after copying via Nautilus

Only fail to set the permissions when the actual file is a symlink.
The previous fix failed for every file when NOFOLLOW_SYMLINKS was set.

15 years agoOnly do the chmod NOFOLLOW_SYMLINK checks with HAVE_SYMLINK
Benjamin Otte [Tue, 1 Sep 2009 18:36:31 +0000 (20:36 +0200)]
Only do the chmod NOFOLLOW_SYMLINK checks with HAVE_SYMLINK

15 years agog_time_val_from_iso8601: handle timezoneless dates
Dan Winship [Tue, 1 Sep 2009 13:37:48 +0000 (09:37 -0400)]
g_time_val_from_iso8601: handle timezoneless dates

per ISO 8601:2004 4.2.5.2

Based on a patch from Andy Shevchenko
http://bugzilla.gnome.org/show_bug.cgi?id=589491

15 years agoFix gtk-doc syntax
Benjamin Otte [Tue, 1 Sep 2009 10:48:55 +0000 (12:48 +0200)]
Fix gtk-doc syntax

15 years agoBug 593406 - Permissions set to 777 after copying via Nautilus
Benjamin Otte [Tue, 1 Sep 2009 09:54:48 +0000 (11:54 +0200)]
Bug 593406 - Permissions set to 777 after copying via Nautilus

When doing a g_file_copy() with nofollow-symlinks (to copy a link for
example), the later copying of the file attributes copies the source
links 777 attributes to the target's attributes. As chmod affects the
symlink target, this would cause such copies to always set the target to
777 mode.

This patch makes setting the mode with nofollow-symlinks fail with
NOT_SUPPORTED.

The aforementioned g_file_copy() will still succeed, because it ignores
errors of the attribute copy.

15 years agoUpdate Norwegian bokmål translation.
Kjartan Maraas [Tue, 1 Sep 2009 08:52:07 +0000 (10:52 +0200)]
Update Norwegian bokmål translation.

15 years agoUpdated Marathi Translations
Sandeep Shedmake [Mon, 31 Aug 2009 15:17:13 +0000 (20:47 +0530)]
Updated Marathi Translations

15 years agoUpdated Marathi Translations
Sandeep Shedmake [Mon, 31 Aug 2009 14:59:16 +0000 (20:29 +0530)]
Updated Marathi Translations

15 years agoUpdated Telugu Translations
krishnababu k [Mon, 31 Aug 2009 14:11:54 +0000 (19:41 +0530)]
Updated Telugu Translations

15 years agoUpdated Kannada(kn) translation
Shankar Prasad [Mon, 31 Aug 2009 08:04:44 +0000 (13:34 +0530)]
Updated Kannada(kn) translation

15 years agoUpdated Brazilian Portuguese mailing list address.
Rodrigo L. M. Flores [Sun, 30 Aug 2009 22:41:47 +0000 (19:41 -0300)]
Updated Brazilian Portuguese mailing list address.

15 years agoUpdated Czech translation
Petr Kovar [Sat, 29 Aug 2009 19:05:19 +0000 (21:05 +0200)]
Updated Czech translation

15 years agoRe-commit 30b8774f7f3f7f329af6b041b6c86bad14717534
Matthias Clasen [Fri, 28 Aug 2009 03:40:16 +0000 (23:40 -0400)]
Re-commit 30b8774f7f3f7f329af6b041b6c86bad14717534

Turns out Alex had agreed to this.

15 years agoUse the saved errno value
Matthias Clasen [Fri, 28 Aug 2009 03:02:08 +0000 (23:02 -0400)]
Use the saved errno value

Bug 591995

15 years agoRead /dev/urandom unbuffered
Matthias Clasen [Fri, 28 Aug 2009 02:59:05 +0000 (22:59 -0400)]
Read /dev/urandom unbuffered

To avoid wasting entropy. Bug 593232.

15 years agoRevert "add a "gsettings-backend" extension point to GIO"
Matthias Clasen [Fri, 28 Aug 2009 02:51:53 +0000 (22:51 -0400)]
Revert "add a "gsettings-backend" extension point to GIO"

This reverts commit 30b8774f7f3f7f329af6b041b6c86bad14717534.

15 years agoadd a "gsettings-backend" extension point to GIO
Ryan Lortie [Thu, 27 Aug 2009 17:45:48 +0000 (13:45 -0400)]
add a "gsettings-backend" extension point to GIO

15 years agoUpdated Italian translation
Luca Ferretti [Wed, 26 Aug 2009 20:04:58 +0000 (22:04 +0200)]
Updated Italian translation

15 years agoUpdated Italian translation
Luca Ferretti [Wed, 26 Aug 2009 13:32:23 +0000 (15:32 +0200)]
Updated Italian translation

15 years agoUpdating Assamese translations
Amitakhya Phukan [Wed, 26 Aug 2009 12:04:17 +0000 (17:34 +0530)]
Updating Assamese translations

15 years agoUpdated Malayalam Translations
Ani [Wed, 26 Aug 2009 09:48:57 +0000 (15:18 +0530)]
Updated Malayalam Translations

15 years agoUpdated Tamil Translations
ifelix [Tue, 25 Aug 2009 08:40:56 +0000 (14:10 +0530)]
Updated Tamil Translations

15 years agoUpdated Oriya Translation
Manoj Kumar Giri [Tue, 25 Aug 2009 06:44:41 +0000 (12:14 +0530)]
Updated Oriya Translation

15 years agoUpdated Turkish translation.
Baris Cicek [Mon, 24 Aug 2009 21:38:34 +0000 (00:38 +0300)]
Updated Turkish translation.

15 years agoFix bump version
Luca Ferretti [Mon, 24 Aug 2009 19:28:11 +0000 (21:28 +0200)]
Fix bump version

15 years agoBump version
Matthias Clasen [Mon, 24 Aug 2009 18:15:13 +0000 (14:15 -0400)]
Bump version

15 years ago2.21.5 2.21.5
Matthias Clasen [Mon, 24 Aug 2009 18:13:49 +0000 (14:13 -0400)]
2.21.5

15 years agoDocumentation fixes
Matthias Clasen [Mon, 24 Aug 2009 17:34:06 +0000 (13:34 -0400)]
Documentation fixes

15 years agoUpdates
Matthias Clasen [Mon, 24 Aug 2009 16:21:08 +0000 (12:21 -0400)]
Updates

15 years agoMinor update to Catalan (Valencian) translation
Carles Ferrando [Mon, 24 Aug 2009 16:38:33 +0000 (18:38 +0200)]
Minor update to Catalan (Valencian) translation

15 years agohindi updated by Rajesh Ranjan
Rajesh Ranjan [Mon, 24 Aug 2009 11:22:25 +0000 (16:52 +0530)]
hindi updated by Rajesh Ranjan

15 years agoUpdated Gujarati Translations
Sweta Kothari [Mon, 24 Aug 2009 08:53:56 +0000 (14:23 +0530)]
Updated Gujarati Translations

15 years agoUpdated Portuguese translation
Duarte Loreto [Sun, 23 Aug 2009 13:59:14 +0000 (14:59 +0100)]
Updated Portuguese translation

15 years agoUpdate Korean translation
Changwoo Ryu [Sat, 22 Aug 2009 15:50:28 +0000 (00:50 +0900)]
Update Korean translation

15 years agoPreserve errno
Christian Persch [Thu, 20 Aug 2009 13:13:43 +0000 (15:13 +0200)]
Preserve errno

When using errno in g_set_error with _(), preserve errno. Bug #592457.

15 years agoUpdate Punjabi after Review
A S Alam [Sat, 22 Aug 2009 00:45:29 +0000 (06:15 +0530)]
Update Punjabi after Review

15 years agoUpdated Finnish translation
Tommi Vainikainen [Fri, 21 Aug 2009 20:43:53 +0000 (23:43 +0300)]
Updated Finnish translation

15 years agoUpdate Japanese translation.
Takayuki KUSANO [Thu, 20 Aug 2009 15:15:12 +0000 (00:15 +0900)]
Update Japanese translation.

15 years agoUpdated Polish translation
Tomasz Dominikowski [Thu, 20 Aug 2009 09:55:03 +0000 (11:55 +0200)]
Updated Polish translation

15 years agoUse MSG_NOSIGNAL in GSocket if it's available
Dan Winship [Wed, 19 Aug 2009 16:12:06 +0000 (12:12 -0400)]
Use MSG_NOSIGNAL in GSocket if it's available

Even though we ignore SIGPIPE, gdb will still stop when the process
receives one, which sometimes confuses people into thinking the app
has crashed (eg, bug 578984, bug 590420), and is annoying anyway. So
use MSG_NOSIGNAL if it's there.

http://bugzilla.gnome.org/show_bug.cgi?id=591378

15 years agoRe-run res_init() when resolv.conf changes
Dan Winship [Wed, 19 Aug 2009 16:07:53 +0000 (12:07 -0400)]
Re-run res_init() when resolv.conf changes

libc caches the contents of resolv.conf, so if it changes (eg, because
the network state changed), we need re-run res_init().

http://bugzilla.gnome.org/show_bug.cgi?id=584246

15 years agoWhen creating a pipe for a cancelled cancellable, write to it
Benjamin Otte [Tue, 11 Aug 2009 16:50:28 +0000 (18:50 +0200)]
When creating a pipe for a cancelled cancellable, write to it

Includes (untested) fix for win32 that calls SetEvent() in that case.

15 years agoUse g_cancellable_release_fd()
Benjamin Otte [Tue, 11 Aug 2009 13:12:20 +0000 (15:12 +0200)]
Use g_cancellable_release_fd()

Part of: Bug 591388 - number of GCancellables available is too limited

15 years agoImplement g_cancellable_release_fd()
Benjamin Otte [Tue, 11 Aug 2009 13:04:43 +0000 (15:04 +0200)]
Implement g_cancellable_release_fd()

Part of: Bug 591388 - number of GCancellables available is too limited

15 years agoAdd g_cancellable_release_fd()
Benjamin Otte [Tue, 11 Aug 2009 12:52:56 +0000 (14:52 +0200)]
Add g_cancellable_release_fd()

This patch only adds the function. The function is a NOP.
See the API documentation for a rationale.

Part of: Bug 591388 - number of GCancellables available is too limited

15 years agoBug 591714 – Figure out failure handling for g_cancellable_make_pollfd()
Benjamin Otte [Thu, 13 Aug 2009 18:19:15 +0000 (20:19 +0200)]
Bug 591714 – Figure out failure handling for g_cancellable_make_pollfd()

Make g_cancellable_make_pollfd() return a gboolean that indicates its error
status. Update the code that calls this function accordingly.

15 years agoUpdated Brazilian Portuguese translation.
Og B. Maciel [Wed, 19 Aug 2009 03:09:12 +0000 (23:09 -0400)]
Updated Brazilian Portuguese translation.

15 years agoImprove documentation for g_error_matches()
Benjamin Otte [Thu, 6 Aug 2009 16:17:32 +0000 (18:17 +0200)]
Improve documentation for g_error_matches()

15 years agoFix multiple returns in gtk-doc comment
Benjamin Otte [Mon, 27 Jul 2009 16:48:11 +0000 (18:48 +0200)]
Fix multiple returns in gtk-doc comment

15 years agog_inet_socket_address_to_native: properly zero out sockaddr_in6
Dan Winship [Mon, 17 Aug 2009 17:20:49 +0000 (13:20 -0400)]
g_inet_socket_address_to_native: properly zero out sockaddr_in6

15 years agoUpdated Danish translation
Ask H. Larsen [Sun, 16 Aug 2009 17:10:50 +0000 (19:10 +0200)]
Updated Danish translation

15 years agoUpdating Estonian translation
Mattias Põldaru [Sat, 15 Aug 2009 05:21:01 +0000 (08:21 +0300)]
Updating Estonian translation

15 years ago[Win32] Avoid superfluous '/' from g_file_resolve_relative_path()
Shixin Zeng [Fri, 14 Aug 2009 20:21:12 +0000 (23:21 +0300)]
[Win32] Avoid superfluous '/' from g_file_resolve_relative_path()

Patch from bug #591532.

15 years agoAvoid gcc warning on Windows
Tor Lillqvist [Fri, 14 Aug 2009 19:59:47 +0000 (22:59 +0300)]
Avoid gcc warning on Windows

Avoid "function declaration isn't a prototype" warnings from gcc 4.4
in the Windows-specific code.

15 years agoUpdate GLIB_CHECK_COMPILE_WARNINGS macro to work with Autoconf 2.64
Frédéric Péters [Fri, 14 Aug 2009 19:00:41 +0000 (21:00 +0200)]
Update GLIB_CHECK_COMPILE_WARNINGS macro to work with Autoconf 2.64

Quoting its NEWS file: "The m4sh macros AS_IF and AS_CASE can now be
used in shell lists.  The responsibility for supplying a trailing
newline now belongs to the call site, but since most users did not add
dnl, this generally results in fewer empty lines in configure."

http://bugzilla.gnome.org/show_bug.cgi?id=591840

15 years agoUpdating Estonian translation
Priit Laes [Fri, 14 Aug 2009 19:09:18 +0000 (22:09 +0300)]
Updating Estonian translation

15 years agoUpdated Telugu Translations
krishnababu k [Fri, 14 Aug 2009 11:26:09 +0000 (16:56 +0530)]
Updated Telugu Translations