Commit Graph

15239 Commits

Author SHA1 Message Date
Sébastien Wilmet
25990eb2b6 docs: various small fixes
For the GPtrArray example, several variables declared on the same line
is harder to read and to work with (to move, remove or comment a single
variable declaration).
2014-09-13 16:59:31 +02:00
Shankar Prasad
fdc5b8f744 Updated Kannada translation 2014-09-12 12:58:11 +00:00
Chun-wei Fan
d2f3e11907 MSVC Builds: Update "Installation"
This updates the glib-install.[props|vsprops] so that the headers to
"install" for the build are kept up-to-date.  Note that this has been
generated by a script, so that this hopefully means that the header
"installation" can be maintained along with the autotools files for most
uses, to simplify maintenance.
2014-09-12 15:51:49 +08:00
Balázs Úr
0c5ed58342 Updated Hungarian translation 2014-09-11 16:40:14 +00:00
Sebastian Dröge
bebfd422af gutils: Don't use issetugid() on Android
Android had it in older versions but the new 64 bit ABI does not
have it anymore, and some versions of the 32 bit ABI neither.

https://code.google.com/p/android-developer-preview/issues/detail?id=168

https://bugzilla.gnome.org/show_bug.cgi?id=736351
2014-09-11 11:07:48 +03:00
Ryan Lortie
3b8bc8bacf GDesktopAppInfo: avoid inotify on missing dirs
Some desktop file directories, like /usr/local/share/applications may be
missing on some systems.

When we try to inotify on these directories, this will result in a
every-4-seconds poll being setup which is quite bad.

This is an issue that should be fixed in inotify itself but the problem
is much larger there.  For now, we can work around it in GDesktopAppInfo
by refusing to monitor missing directories.

We may get some spurious notifications of changes in the case that
/usr/local/share or /usr/local/share/applications is created without
actually adding desktop files, but spurious changes can already be
reported in other cases, so that's OK.  We won't get (user-visible)
notification for a simple case of a completely unrelated file being
created (however we cannot avoid the wakeup in this case due to how
inotify works).  That's probably pretty theoretical, though, since files
in /usr don't change much and for the home directory we're likely to
have at least ~/.config and ~/.local existing.

https://bugzilla.gnome.org/show_bug.cgi?id=736350
2014-09-09 14:11:38 -04:00
Ryan Lortie
2f55c66c64 apps test: add new "monitor" subcommand
Waits until something modifies a desktop directory, then exits.

https://bugzilla.gnome.org/show_bug.cgi?id=736350
2014-09-09 14:11:38 -04:00
Yuri Myasoedov
9ac7d51a80 Updated Russian translation 2014-09-09 17:04:40 +04:00
Paolo Borelli
8df2cca08a Fix minor mem leak in test case 2014-09-08 08:19:25 +02:00
Paolo Borelli
5bb62d077b GThreadPool: expand g_thread_pool_new docs 2014-09-08 08:19:25 +02:00
Matthias Clasen
aef0e4b7f3 Fix docs to talk about --enable-coverage
The configure options to turn on coverage testing got renamed
at some point, but the docs were not updated. Fix that.

https://bugzilla.gnome.org//show_bug.cgi?id=735915
2014-09-07 22:50:36 -04:00
Rūdolfs Mazurs
281fdd372e Updated Latvian translation 2014-09-07 12:55:20 +03:00
Fran Diéguez
09979125d8 Updated Galician translations 2014-09-04 21:35:07 +02:00
Ryan Lortie
08efbda733 GDesktopAppInfo: fix default app logic
We use "tweaks" structures to track how a particular directory impacts
the list of added, removed and default applications.  We maintain this
set of tweaks for each directory, in a hash table, keyed by unaliased
mime type name, in order to facilitate fast lookups.

A typo in the logic for creating and maintaining the uniqueness of these
structures was causing the default app to be selected incorrectly from
time to time.  Fix that.
2014-09-04 14:48:05 -04:00
Patrick Welche
e763d93456 GCredentials: credentials messages are not correctly supported on NetBSD
https://bugzilla.gnome.org/show_bug.cgi?id=735819
2014-09-04 08:26:27 -04:00
Andika Triwidada
7dd3397820 Updated Indonesian translation 2014-09-03 13:49:48 +00:00
Piotr Drąg
0ba39c1c7a Updated Polish translation 2014-09-03 03:10:27 +02:00
Matthias Clasen
d73886675e 2.41.4 2014-09-02 11:36:32 -04:00
Changwoo Ryu
5fe65ec271 Updated Korean translation 2014-09-02 13:05:58 +09:00
Claude Paroz
62f0d6143a Updated French translation 2014-09-01 09:30:32 +02:00
Christian Kirbach
d4298e5ed1 Updated German translation 2014-08-31 22:00:45 +00:00
Matthias Clasen
a78443a1e4 Don't mark GThread struct as deprecated
Having a definition of struct _GThread inside a deprecation
ifdef confuses gtk-doc into marking it as deprecated. Avoid this.

https://bugzilla.gnome.org//show_bug.cgi?id=735297
2014-08-29 15:12:31 -04:00
Maria Mavridou
325ffb376c Updated Greek translation 2014-08-28 21:25:31 +00:00
Yosef Or Boczko
0a37264fd1 Updated Hebrew translation 2014-08-28 18:11:06 +03:00
Manoj Kumar Giri
f16d7f887e Updated Oriya translation 2014-08-27 13:34:53 +00:00
Patrick Welche
369c8bd566 GUnixMounts: Improve efficiency in polling case
https://bugzilla.gnome.org/show_bug.cgi?id=583330
2014-08-27 15:08:38 +02:00
Thomas Haller
35eaf037bd gmacros.h: add G_GNUC_*_IGNORE_DEPRECATIONS macros for clang
https://bugzilla.gnome.org/show_bug.cgi?id=734126

Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-08-22 20:58:11 +02:00
Philip Withnall
cb320cb5fe gsocketclient: Handle cancellation between CONNECTING and CONNECTED
If a g_socket_client_connect_async() operation is cancelled between the
CONNECTING and CONNECTED events (i.e. while in the
g_socket_connection_connect_async() call), the code in
g_socket_client_connected_callback() would previously unconditionally
loop round and try the next socket address from the address enumerator
(by calling enumerator_next_async()). This would correctly handle the
cancellation and return from the overall task — but not before emitting
a spurious RESOLVING event.

Avoid emitting the spurious RESOLVING event by explicitly handling
cancellation at the beginning of g_socket_client_connected_callback().

https://bugzilla.gnome.org/show_bug.cgi?id=735179
2014-08-22 19:11:26 +01:00
Marek Černocký
607d5a7858 Updated Czech translation 2014-08-22 08:08:51 +02:00
Chao-Hsiung Liao
c74af7369f Updated Traditional Chinese translation(Hong Kong and Taiwan) 2014-08-21 17:06:03 +08:00
Jonas Danielsson
bf9c862504 GApplication: Add g_application_add_main_option
This function adds a single main option entry to be handeled by
GApplication. The option entry has it arg_data field set to NULL
and will be added to the applications packed_options.

The rationale for this is that bindings will be able to add
command line options even when they can't use the un-boxed struct
GOptionEntry.

https://bugzilla.gnome.org/show_bug.cgi?id=727455
2014-08-20 16:02:59 +02:00
Patrick Welche
4f775b7b7f GUnixMounts: Fall back to polling on systems without mtab
This is necessary for many of the BSD family at least.

https://bugzilla.gnome.org/show_bug.cgi?id=583330
2014-08-19 10:48:12 +02:00
ngoswami
d964c7d686 Updated Assamese translation 2014-08-18 15:44:42 +00:00
Rico Tzschichholz
e0e52b60ea docs: Fix broken gtk-doc formatting 2014-08-18 14:08:41 +02:00
Rico Tzschichholz
4c8480e6fa docs: Add index for 2.42 api 2014-08-17 22:54:22 +02:00
Carles Ferrando
96af008bdf [l10n] Updated Catalan (Valencian) translation 2014-08-17 13:19:46 +02:00
Gil Forcada
7585cc49c7 [l10n] Updated Catalan translation 2014-08-17 13:19:42 +02:00
Matthias Clasen
e03fc7a522 Updates for 2.41.3 2014-08-16 09:02:12 -04:00
Christian Kirbach
01fec1f0d5 Updated German translation 2014-08-15 17:58:33 +02:00
Chun-wei Fan
52b5a06dea Visual C++ Builds: Update README.txt's
Update notes about usage of PCRE and for people attempting to build GLib
on a Chinese, Japanese or Korean locale.
2014-08-13 09:55:11 +08:00
Dan Winship
b88fdf0b0f docs: hyphenate "High-level" and "Low-level" 2014-08-12 07:50:20 -04:00
Chun-wei Fan
3d89041220 MSVC Builds: Generate glib-mkenums If Possible
As glib-mkenums would likely be used in the building of the other
components of the stack, such as Cogl, Clutter and
gsettings-desktop-schemas, generate that using a Python script (if Python
can be found) and "install" it.
2014-08-08 17:39:48 +08:00
Edward Hervey
3e2735f468 gtype: Fast-path for g_type_is_a
Do I really need to explain why ?

https://bugzilla.gnome.org/show_bug.cgi?id=732085
2014-08-08 11:13:38 +02:00
Andre Moreira Magalhaes (andrunko)
4b5bdf6ceb Do not crash when checking whether an instance type is of a given fundamental.
Not all instances have a TypeNode associated (e.g. GstEvent), so lets check if node is available
before trying to use it.

This crash can be easily reproduced by creating an event with gst_event_new_eos and using
G_IS_OBJECT on the event instance.

https://bugzilla.gnome.org/show_bug.cgi?id=733982
2014-08-08 11:13:01 +02:00
Matej Urbančič
5053054c62 Updated Slovenian translation 2014-08-06 14:15:40 +02:00
Руслан Ижбулатов
3c6efd4d50 Make sure compiled resources only have / as dirsep
https://bugzilla.gnome.org/show_bug.cgi?id=725511
2014-08-05 21:49:54 +00:00
Martin Srebotnjak
b637714ded Updated Slovenian translation 2014-08-05 21:03:32 +02:00
Inaki Larranaga Murgoitio
10504edfaa Updated Basque language 2014-08-05 14:02:42 +02:00
Руслан Ижбулатов
14f2376c8a Change W32 console detection to look at the window, not stdout
Check for console window being attached to the process, not for stdout
being redirected into a console window.

https://bugzilla.gnome.org/show_bug.cgi?id=733960
2014-08-05 06:44:01 +00:00
Руслан Ижбулатов
4f73487300 Fix tests to compile again - add missing headers to W32, call correct process-id-getting function
https://bugzilla.gnome.org/show_bug.cgi?id=725513
2014-08-02 12:41:14 +00:00