Age | Commit message (Collapse) | Author | Files | Lines |
|
|
|
*Build a cssprovider from strings, use a variable to apply the same colors as used on the rectangles shown on each monitor to the matching label backgrounds in the popup menu. Bring back the draw callback to handle it, so necessary data can be passed to mate_rr_labeler_get_rgba_for_output
*Keep disabled item dimming effects off the monitor label icons
*Tie the dynamic cssprovider to the label so it always gets the correct color
|
|
|
|
|
|
|
|
Using black color makes the label unreadable with dark themes
|
|
This change is a super hacky way of supporting the fallback desktop background (without caja) on HiDPI displays. It does this by doing two things: using a new function that scales up the background surface; and by loading settings using an updated function that triggers a redraw.
|
|
This changes how the media-keys manager determines successful
key-bindings from KeyRelease to KeyPress so that it does not conflict
with third-party applications that want to use modifier keys as global
key-bindings. We've already done this change for general keybindings and
marco.
It also cleans up a bunch of deprectation warnings.
|
|
Also fix build warnings on xsettings
|
|
Also set QT_* env variables during init, and restart marco, mate-panel, toggle desktop icons when changing scales
|
|
This change allows for Xft/Gdk settings to be changed dynamically for
the scaling factor and font DPI. It also has an optional auto-detection
algorithm for the most appropriate settings to use for different screen
sizes and resolutions. The auto-detect is re-triggered whenever the
screen size changes.
|
|
|
|
* Support font size and mouse cursor auto-scaling for HiDPI displays
* Fix where DPI is getting scaled
* Remove ceil for scaled DPI
|
|
Fixes build warnings
|
|
this requires https://github.com/mate-desktop/mate-desktop/commit/b133f518f6f2310093fd3b5248049c7048cfb8d4
closes https://github.com/mate-desktop/mate-desktop/issues/296
|
|
format ‘%lld’ expects argument of type ‘long long int’, but argument has type ‘long int’
|
|
passing argument 3 of ‘g_queue_find_custom’ from incompatible pointer type
|
|
The work began with:
https://github.com/mate-desktop/mate-settings-daemon/commit/788babedafaf59d8b7a36cb0ef3b0b1f0e5f732c
This commit reverts:
https://github.com/mate-desktop/mate-settings-daemon/commit/7f887e614812ccb86f5d784e2db58bede8daa712
And it applies an alternative to fix the deprecated functions:
gdk_screen_get_width
gdk_screen_get_height
|
|
Fixes https://bugzilla.redhat.com/show_bug.cgi?id=1517547
|
|
Gdk-CRITICAL **: gdk_window_thaw_toplevel_updates: assertion
'window->update_and_descendants_freeze_count > 0' failed
inspired from:
https://git.xfce.org/apps/xfce4-terminal/commit/?id=eff5e2d
|
|
avoid deprecated:
gdk_screen_get_monitor_geometry
gdk_screen_get_monitor_at_point
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
When setting `KeyPressMask` as a reportable event for all windows, the
keybinding for `Fn+F7` in Xrandr seems to fail. This change adds, rather
than overrides, the `KeyPressMask` as a reportable event for the
corresponding window.
Fixes #180
|
|
* Request keyboard event reports from global keybindings of other windows
This allows mate-settings-daemon to honor global keybindings whenever
a different application has keybindings that seem conflicting, but are
usually not.
For example, an application defining a `Super_L` keybinding
and MATE having defining a shortcut that uses the `<Mod4>` key. They are
the same physical keyboard, but one is a modifier and the other is not.
Without requesting keyboard reports from other windows, the `<Mod4>`
shortcuts are swallowed whenever there is a `Super_L` shortcut
elsewhere.
* Remove KeyReleaseMask as the keybindings_filter only looks at KeyPress
* Guard against BadWindow errors
|
|
|
|
This menu was not noticed back in 2015 when support for custom panel menu themes was added
|
|
|
|
This reduce the width of the dialog
|
|
https://bugzilla.gnome.org/show_bug.cgi?id=642803
taken from:
https://git.gnome.org/browse/gnome-settings-daemon/commit/?id=4546de7
|
|
|
|
Applications might want to establish keyboard grabs on button presses so we
should also ungrab the keyboard when a button press occurs after the locate
pointer key is pressed.
https://bugzilla.gnome.org/show_bug.cgi?id=125618
|
|
|
|
|
|
|
|
and revert "GTK 3.21: Fix background changing, startup delay".
These are no longer of need as Caja reclaimed its ability to draw backgrounds.
|
|
my stupid mistake - it was applied to the first device in the list only
|
|
adapted from:
https://github.com/linuxmint/cinnamon-settings-daemon/commit/4ab4ebab920ee8effb3f83df9318fbd7ad6a371c
|
|
note: threshold setting is ignored as libinput doesn't have it
adapted from:
https://github.com/linuxmint/cinnamon-settings-daemon/commit/1e7d504a5d0003b1c71e8bcacc3476002107d414
|
|
|
|
note: libinput function doesn't have button remapping code,
it's handled in libinput already
adapted from:
https://github.com/linuxmint/cinnamon-settings-daemon/commit/c35f95210f025132e2e551024a01f6102371a0a0
|
|
same as done here:
https://github.com/linuxmint/cinnamon-settings-daemon/commit/c35f95210f025132e2e551024a01f6102371a0a0
|
|
note: libinput does not allow for configuration of the click actions,
the options are just "software buttons" or "clickfinger". If any
clickfinger actions are set, we enable clickfinger, otherwise software
buttons.
adapted from:
https://github.com/linuxmint/cinnamon-settings-daemon/commit/6394ea6c2754c434897cebfc6c331b4ee07ce1b8
|
|
note: libinput does not allow for both edge and twofinger scrolling
to be enabled simultaneously. We prefer twofinger scrolling. The same
goes for horizontal scrolling, it picks the setting for whatever scroll
method we applied.
adapted from:
https://github.com/linuxmint/cinnamon-settings-daemon/commit/82442095a33b1f063f00512d438711ef092e0121
|
|
note: libinput function doesn't use additional arguments
since no button remapping is needed for it
adapted from:
https://github.com/linuxmint/cinnamon-settings-daemon/commit/85510232d53e8ed98697e9a7e3cc33274a531e19
|