Age | Commit message (Collapse) | Author | Files | Lines |
|
http://git.gnome.org/browse/gnome-settings-daemon/commit/?id=b9b785508085d77818c76ab4aed8ee574dbd1756
|
|
Most of the plugins do not need gnome-desktop's functionality, so there is
no point in the main process linking against it.
http://git.gnome.org/browse/gnome-settings-daemon/commit/?id=ed54ceee46b6a56613c8aeb3d594f6a485400ef9
|
|
a11y-keyboard: gtk_dialog_set_has_separator(), deprecated in GTK+ 2.22
common|mouse: prefix GDK key constants with _KEY (GTK3)
|
|
http://git.gnome.org/browse/gnome-settings-daemon/commit/?id=1c1677dcfd55eb7e1f6a20435af466331ec2a911
|
|
1. Use g_settings_get_mapped instead of our own hack get_gsettings_int_with_default.
2. Don't force hard-coded default when value is 0, as that's used to allow
paranoid scanning on shutdown.
3. Only unref the settings object at the end of _stop routine, not before
do_cleanup is potentially called, since purge_thumbnail_cache uses the settings.
This was also causing asserts when killing msd, as described in (now closed) #23.
These are regressions introduced by the gsettings port of the plugin -- 03478b4d
|
|
gtk_dialog_set_has_separator deprecated in GTK+ 2.21.8
simple gtk_object -> gtk_widget switch for GTK3 forward compat.
|
|
Namely:
1. In caja_is_drawing..(), only get atom for WM_CLASS property if it exists.
2. Cache gsettings draw-background & show-desktop-icons key-values, and only update
when either actually changes, to avoid unnecessary queries to the dconf-service.
3. Avoid conflicts between settings_change_event_cb() & on_bg_handling_changed().
4. Keep track of screen sizes, to ensure size actually changed when signalled, to
avoid double-redraw on an extraneous signal, which can lockup the daemon, because
we draw immediately (w/out g_idle_add). Though we don't fade in re-draws triggered
by size/monitors-changed signals, making all re-draws queued causes issues with
fading, when we do fade (on bg change), and queueing is unnecessary here anyway.
5. Track a redraw in progress, to avoid any possible (and unforseeable) clashes.
|
|
|
|
|
|
instead of g_timeout_add().
daemon: Use g_timeout_add_seconds (30 seconds)
http://git.gnome.org/browse/gnome-settings-daemon/commit/?id=61c8ca9ad8add09830e72d0c10b97f59e80dd0f5
smartcard: Use g_timeout_add_seconds() (2 seconds)
http://git.gnome.org/browse/gnome-settings-daemon/commit/?id=5e85aa9cdf6624effea5a890a2516f11d25a8bf3
|
|
Was used to setup old X legacy fonts, and was disabled by default in bug:
https://bugzilla.gnome.org/show_bug.cgi?id=613604
|
|
had xklavier-related warning everytime m-s-d is started, specifically after
we stopped daemonizing the daemon, in a previous commit.
and the bug revolves around libxklavier issue
https://bugzilla.gnome.org/show_bug.cgi?id=150542
http://git.gnome.org/browse/gnome-settings-daemon/commit/?id=4239cb8a5d73af79d2cdf0f294a33d31fbbb9afe
http://git.gnome.org/browse/gnome-settings-daemon/commit/?id=241c970c1361c4d1b2663af20252ef127055d768
|
|
Updates volume code from gnome-media
http://git.gnome.org/browse/gnome-settings-daemon/commit/?id=5f4b71733713741ec6c48480e3ddc73bf72eb290
referenced bug also says fixes a memory leak
https://bugzilla.gnome.org/show_bug.cgi?id=612024
|
|
And instead, use glib macros (since 2.14)
http://git.gnome.org/browse/gnome-settings-daemon/commit/?id=0dda56c4462e070dabdab68092b6574b5e894181
|
|
for reference: xsettings: Port XSettings sync to GSettings
http://git.gnome.org/browse/gnome-settings-daemon/commit/?id=6d7498114930de358573e97f54880c586c9f9c66
|
|
1. import: Don't choke if there are old plugins laying around
http://git.gnome.org/browse/gnome-settings-daemon/commit/?id=18e57e126ff6a6ab2980c119aebef087227e2a54
2. Alleviate #23 -- no more "g_object_unref: assertion 'G_IS_OBJECT (object)' failed"
when starting m-c-c "Appearance" properties.
|
|
Observe new mate-bg (global) fade option, via do_crossfade_background()
Change dont_draw_background() to a do_, to avoid double-negative like !dont.
|
|
background revamped
|
|
from 24 -> 20, to account for 'nautilus' -> 'caja' change (4 bytes less).
|
|
hopefully to avoid taking over background-drawing when Caja is doing it.
|
|
Because we should handle the background drawing when either:
- "draw-background" is re-enabled, or
- "show-desktop-icons" is disabled.
N.B.
We may still need to consider this issue, which was never resolved in gnome,
because they stopped auto-launching nautilus & removed it from 3.0 sessions:
Nautilus registers with the session before it's ready.
https://bugzilla.gnome.org/show_bug.cgi?id=568588
Would also be wise to import the (recent) changes made in the bug below:
Provide a singleton SessionManager proxy object
https://bugzilla.gnome.org/show_bug.cgi?id=686556
|
|
https://bugzilla.gnome.org/show_bug.cgi?id=564909
|
|
http://git.gnome.org/browse/gnome-settings-daemon/commit/?id=da9bd9769caa0f776edf5636c6ae09ba425d6f6e
|
|
prevents some strange flicker when quickly selecting multiple backgrounds
http://git.gnome.org/browse/gnome-settings-daemon/commit/?id=5cb80995b6899c6ed185797333b709ea9342de42
|
|
http://git.gnome.org/browse/gnome-settings-daemon/commit/?id=34d6831d7bf5fdf869f1b4802d8920c4a90825db
|
|
and use new mate-bg API ("show-desktop-icons")
For reference, see:
Port backround module to GSettings
http://git.gnome.org/browse/gnome-settings-daemon/commit/?id=23b50dfc7078b78a513598060ee22ae46c262534
http://git.gnome.org/browse/gnome-settings-daemon/commit/?id=c9d62bab8acb166aa720fa5d0356657177be6be5
background: always check if Nautilus is running
http://git.gnome.org/browse/gnome-settings-daemon/commit/?id=d22b75db508a792e3786a2011b707e930a0f0355
background: reintroduce GnomeBg "changed" signal
http://git.gnome.org/browse/gnome-settings-daemon/commit/?id=d7a6f707c9e6f924fea7565967d9c5c1ad8b1533
background: Listen to screen changes only after GnomeBg is constructed
http://git.gnome.org/browse/gnome-settings-daemon/commit/?id=d6d12b643aeec91fa3d1ed19104f1c0d1ded7870
|
|
http://git.gnome.org/browse/gnome-settings-daemon/commit/?id=d38bb683005b71f3bf483aafee2f7c36d800a0c7
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
This reverts commit 91f651c6f6f0ee7fdff1d49e36161d4e76e16e02.
|
|
do_multimedia_player_action did nothing more than call
msd_media_player_key_pressed, which was pointless overhead.
|
|
|
|
(https://github.com/mate-desktop/mate-desktop/issues/16)
|
|
(http://imageshack.us/photo/my-images/9/254223656580e542a60a.jpg/)
|
|
|
|
try to find mate-calc, fall back to gcalctool if it is not found.
Signed-off-by: Stefan Tauner <[email protected]>
|
|
dialog is displayed
|
|
|
|
In a lot of situations it is undesirable to have g-s-d change the XRandR
settings, because it overrides X.org customizations, leads to unnecessary mode
switches, or increases boot time.
Add a mateconf key "use_xorg_monitor_settings" to disable
apply_default_boot_configuration(), in which case the XRandR configuration will
not be touched unless there is a global or per-user configuration file.
Bug: https://bugzilla.gnome.org/show_bug.cgi?id=631388
Bug-Ubuntu: https://launchpad.net/bugs/640807
|
|
fallback configurations
Previously, if a stored configuration existed but it could not be applied due
to nonmatching monitors, we would do nothing else - potentially leaving the user
in whatever startup state was used by the X server. Now, in that condition
we simply fall back to trying the system-global configuration or the boot-time
configuration.
Signed-off-by: Federico Mena Quintero <[email protected]>
|
|
In 2.32, we still want to use mate-display-properties.
This reverts commit 03112091d18ad0f9cc66b5f8835aadb1c47bc9ee.
|
|
|
|
|
|
(https://github.com/mate-desktop/mate-panel/issues/8)
|