View Bazaar branches
Get this repository:
git clone https://git.launchpad.net/mutter

Branches

Name Last Modified Last Commit
master 2017-10-20 08:25:31 UTC 13 hours ago
Revert "MetaWindowActor: don't start any effect when no compositor is available"

Author: Marco Trevisan (Treviño)
Author Date: 2017-10-20 08:25:31 UTC

Revert "MetaWindowActor: don't start any effect when no compositor is available"

This reverts commit 35fcf4a4ae5e4c7259111def96aa19a2d5081fc4.

gnome-3-26 2017-10-20 00:55:49 UTC 20 hours ago
Revert "ClutterActor: Optimize away idempotent scale/position updates"

Author: Michael Catanzaro
Author Date: 2017-10-20 00:53:13 UTC

Revert "ClutterActor: Optimize away idempotent scale/position updates"

This reverts commit 5cb5baa7d42712f9e0db8dfaa0485e2d833e4d83.

There's a cascade of regressions that needs to be sorted out before
relanding this:

https://bugzilla.gnome.org/show_bug.cgi?id=784314

wip/fmuellner/contraint-tiling 2017-10-03 21:16:44 UTC 2017-10-03
x11: Add support for _GTK_EDGE_CONSTRAINTS atom

Author: Georges Basile Stavracas Neto
Author Date: 2017-08-18 23:21:37 UTC

x11: Add support for _GTK_EDGE_CONSTRAINTS atom

To keep feature parity with the Wayland backend, and
to improve the overall tiling experience with GTK apps,
add the _GTK_EDGE_CONSTRAINTS X11 atom and update it
when necessary.

https://bugzilla.gnome.org/show_bug.cgi?id=751857

gnome-3-24 2017-09-28 13:59:45 UTC 2017-09-28
wayland/surface: Send buffer release events immediately

Author: Jonas Ådahl
Author Date: 2017-09-26 15:28:44 UTC

wayland/surface: Send buffer release events immediately

Just queuing the buffer release event is prone to starvation, so send
the release immediately. See
https://lists.freedesktop.org/archives/wayland-devel/2017-September/035147.html
for more detailed explanation.

https://bugzilla.gnome.org/show_bug.cgi?id=788197

wip/xwayland-on-demand 2017-08-28 02:50:44 UTC 2017-08-28
Fix cursor size update on non-wayland

Author: Armin Krezović
Author Date: 2017-08-27 19:04:14 UTC

Fix cursor size update on non-wayland

wip/rstrode/finish-to-boost 2017-08-18 16:50:17 UTC 2017-08-18
cogl: glFinish before swap buffers for speed boost

Author: Ray Strode
Author Date: 2017-08-18 16:50:17 UTC

cogl: glFinish before swap buffers for speed boost

The intel driver will automatically boost the clockspeed
of the GPU when the client calls glFinish and there's
pending drawing.

This commit makes sure to do that call explicitly before
swap buffers to help prevent animation stutter.

gnome-3-22 2017-08-12 13:35:32 UTC 2017-08-12
Update Indonesian translation

Author: Andika Triwidada
Author Date: 2017-08-12 13:35:32 UTC

Update Indonesian translation

tintou/wip/vapi 2017-06-30 15:02:19 UTC 2017-06-30
Enable vala .vapi generation, create the meta/meta.h umbrella file

Author: Corentin Noël
Author Date: 2017-06-30 15:02:19 UTC

Enable vala .vapi generation, create the meta/meta.h umbrella file

wip/carlosg/performance-improvements 2017-05-22 15:43:10 UTC 2017-05-22
wayland: Use weak ref to keep track of buffers

Author: Carlos Garnacho
Author Date: 2017-05-06 20:46:06 UTC

wayland: Use weak ref to keep track of buffers

Instead of using a one-shot signal meant to happen then the buffer
is being destroyed.

https://bugzilla.gnome.org/show_bug.cgi?id=782344

wip/jstpierre/cursor-position-tracker 2017-02-23 19:06:41 UTC 2017-02-23
backends: Add a method for monitoring cursor position

Author: Jasper St. Pierre
Author Date: 2017-02-23 19:06:41 UTC

backends: Add a method for monitoring cursor position

gnome-3-20 2016-10-29 18:25:41 UTC 2016-10-29
Update zh_CN translation

Author: YunQiang Su
Author Date: 2016-10-29 18:25:41 UTC

Update zh_CN translation

gnome-3-18 2016-10-29 18:19:33 UTC 2016-10-29
Update zh_CN translation

Author: liushuyu
Author Date: 2016-10-29 18:19:33 UTC

Update zh_CN translation

wip/fmuellner/gtk-shadows 2016-10-28 15:10:57 UTC 2016-10-28
theme: Update style classes to match CSD decorations

Author: Florian Müllner
Author Date: 2016-10-28 00:56:28 UTC

theme: Update style classes to match CSD decorations

We currently use the "ssd" style class instead of "csd", so themes
can avoid adding a window shadow to windows that already got a shadow
from the compositor. Now that we no longer add a shadow to regular
windows, we should ask for the appropriate style.

https://bugzilla.gnome.org/show_bug.cgi?id=744667

gnome-3-4 2016-10-11 18:57:23 UTC 2016-10-11
Update Occitan translation

Author: Cédric VALMARY (Tot en òc)
Author Date: 2016-10-11 18:57:23 UTC

Update Occitan translation

gnome-3-6 2016-10-11 18:54:45 UTC 2016-10-11
Update Occitan translation

Author: Cédric VALMARY (Tot en òc)
Author Date: 2016-10-11 18:54:45 UTC

Update Occitan translation

gnome-3-8 2016-10-11 18:53:55 UTC 2016-10-11
Update Occitan translation

Author: Cédric VALMARY (Tot en òc)
Author Date: 2016-10-11 18:53:55 UTC

Update Occitan translation

gnome-3-10 2016-10-11 18:50:33 UTC 2016-10-11
Update Occitan translation

Author: Cédric VALMARY (Tot en òc)
Author Date: 2016-10-11 18:50:33 UTC

Update Occitan translation

gnome-3-12 2016-10-11 18:49:06 UTC 2016-10-11
Update Occitan translation

Author: Cédric VALMARY (Tot en òc)
Author Date: 2016-10-11 18:49:06 UTC

Update Occitan translation

gnome-3-2 2016-10-11 18:10:36 UTC 2016-10-11
Update Occitan translation

Author: Cédric VALMARY (Tot en òc)
Author Date: 2016-10-11 18:10:36 UTC

Update Occitan translation

gnome-3-0 2016-10-11 18:06:03 UTC 2016-10-11
Update Occitan translation

Author: Cédric VALMARY (Tot en òc)
Author Date: 2016-10-11 18:06:03 UTC

Update Occitan translation

gnome-3-14 2016-10-11 17:53:17 UTC 2016-10-11
Update Occitan translation

Author: Cédric VALMARY (Tot en òc)
Author Date: 2016-10-11 17:53:17 UTC

Update Occitan translation

gnome-3-16 2016-10-11 17:50:56 UTC 2016-10-11
Update Occitan translation

Author: Cédric VALMARY (Tot en òc)
Author Date: 2016-10-11 17:50:56 UTC

Update Occitan translation

wip/garnacho/wayland-emulated-output-transform 2016-08-23 17:24:35 UTC 2016-08-23
backends/native: Expose all transform modes in CRTCs

Author: Carlos Garnacho
Author Date: 2016-08-01 01:51:36 UTC

backends/native: Expose all transform modes in CRTCs

We can only honor this properly in the MUTTER_STAGE_VIEWS=1 case. When using
the legacy view, software implemented transforms are only exposed if there is
only one output, as we can only transform the entire stage there.

https://bugzilla.gnome.org/show_bug.cgi?id=745079

wip/garnacho/virtual-input-device 2016-07-26 16:45:26 UTC 2016-07-26
backends: Prepare for virtual devices

Author: Carlos Garnacho
Author Date: 2016-07-11 16:23:09 UTC

backends: Prepare for virtual devices

Those have no backing libinput_device, and configuration does not
apply to those.

https://bugzilla.gnome.org/show_bug.cgi?id=765009

wip/tablet-protocol-v2 2016-07-20 17:27:24 UTC 2016-07-20
wayland: Emit wp_tablet_pad_group.buttons after focus changes

Author: Carlos Garnacho
Author Date: 2016-06-29 16:39:31 UTC

wayland: Emit wp_tablet_pad_group.buttons after focus changes

The buttons grabbed by the compositor might have changed in between,
so just broadcast the button array again.

wip/tablet-protocol 2016-05-02 11:49:13 UTC 2016-05-02
backends: Allow multiple "SW" cursor overlays on the stage

Author: Carlos Garnacho
Author Date: 2016-05-02 11:49:13 UTC

backends: Allow multiple "SW" cursor overlays on the stage

All the upper layers are prepared for multiple onscreen cursors, but
this. All MetaCursorRenderers created would poke the same internal
MetaOverlay in the stage.

This will lead to multiple cursor renderers resorting to the "SW"
rendering paths (as it can be seen with tablet support) to reuse the
same overlay, thus leading to flickering when a different
MetaCursorRenderer takes over the overlay.

Fix this by allowing per-cursor-renderer overlays, their lifetime
is attached to the cursor renderer, so is expected to be tear down
if the relevant device (eg. tablet) disappears.

wip/rtcm/merge 2016-04-22 16:48:00 UTC 2016-04-22
Fix the merged build

Author: Rui Matos
Author Date: 2016-04-08 17:58:20 UTC

Fix the merged build

wip/tiling-2 2016-04-04 21:51:56 UTC 2016-04-04
frame: Don't allow resizing of edges that are constrained

Author: Jasper St. Pierre
Author Date: 2015-07-01 05:14:03 UTC

frame: Don't allow resizing of edges that are constrained

https://bugzilla.gnome.org/show_bug.cgi?id=751857

wip/primary-selection 2016-02-20 00:37:01 UTC 2016-02-20
WIP: Implement the primary selection protocol

Author: Carlos Garnacho
Author Date: 2016-02-03 17:39:58 UTC

WIP: Implement the primary selection protocol

wip/garnacho/wayland-startup-notification 2016-02-17 13:47:54 UTC 2016-02-17
wayland: Implement gtk-shell v3

Author: Carlos Garnacho
Author Date: 2016-02-06 14:04:25 UTC

wayland: Implement gtk-shell v3

Implement the gtk_shell.set_startup_id request, so that the ID is
removed from the sequences list, and feedback updated accordingly.

halfline/wip/kms-cleanups 2016-01-08 21:18:23 UTC 2016-01-08
meta-monitor-manager-kms: get drm fd straight from native renderer

Author: Ray Strode
Author Date: 2016-01-08 19:49:08 UTC

meta-monitor-manager-kms: get drm fd straight from native renderer

This commit changes the MetaMonitorManagerKms class to fetch
the drm fd directly from native renderer instead of roundtripping
through cogl.

wip/garnacho/wayland-extra-scroll-events 2015-10-23 14:21:25 UTC 2015-10-23
wayland: Implement wl_pointer.axis_source/axis_stop/axis_frame emission

Author: Carlos Garnacho
Author Date: 2015-09-08 12:32:50 UTC

wayland: Implement wl_pointer.axis_source/axis_stop/axis_frame emission

As per the spec. wl_pointer.axis_source determines the current source of
scroll events, wl_pointer.axis_stop determines when there's no further
scroll events on any axis (Which Clutter sends as dx/dy=0 events).
wl_pointer.axis_frame marks the end of a series of axis_* events.

v2: Updated to v4 of the protocol spec by
    Peter Hutterer <peter.hutterer@who-t.net>
v3: Update to use Clutter API

wip/halfline/black-stage-before-dpms-off 2015-10-17 19:04:51 UTC 2015-10-17
xrandr: set stage to black before dpms off

Author: Ray Strode
Author Date: 2015-10-17 19:01:17 UTC

xrandr: set stage to black before dpms off

redraws aren't processed when dpms has been turned
off, so we need to set the stage to black before
the screen goes dark, so that when it comes back
it's not showing stale contents.

wip/fullscreen 2015-10-12 19:40:24 UTC 2015-10-12
compositor: add hooks for fullscreen and unfullscreen animations

Author: Cosimo Cecchi
Author Date: 2013-09-01 20:04:19 UTC

compositor: add hooks for fullscreen and unfullscreen animations

https://bugzilla.gnome.org/show_bug.cgi?id=707248

wip/dnd-actions 2015-09-28 14:31:24 UTC 2015-09-28
wayland: Implement DnD actions as per wl_data_device v3 changes

Author: Carlos Garnacho
Author Date: 2015-04-07 14:05:46 UTC

wayland: Implement DnD actions as per wl_data_device v3 changes

We now additionally send:
  - wl_data_offer.source_actions
  - wl_data_source.action
  - wl_data_offer.action
  - wl_data_source.drop_performed
  - wl_data_source.drag_finished

The protocol changes allow for compositors to implement different policies
when chosing the action, mutter uses this to reimplement the same behavior
that GTK+ traditionally had:

  - Alt/Control/Shift modifiers change the chosen action to
    ask/copy/move respectively
  - Drags with middle button start out as "ask" by default

As mutter now also grabs the keyboard and unsets the window focus for these
purposes, the window focus is restored after the drag operation has
finished.

wip/garnacho/touchpad-gestures 2015-08-10 14:48:31 UTC 2015-08-10
wayland: Implement wl_pointer_gestures

Author: Carlos Garnacho
Author Date: 2015-07-22 14:50:20 UTC

wayland: Implement wl_pointer_gestures

The global wl_pointer_gestures object is now created, effectively
bridging pinch/swipe gestures with clients, so they're now
accessible to clients implementing the protocol.

wip/viewport 2014-11-05 18:14:35 UTC 2014-11-05
wayland: Implement support for wl_viewport

Author: Jasper St. Pierre
Author Date: 2014-08-26 19:49:05 UTC

wayland: Implement support for wl_viewport

wip/dnd-surface2 2014-09-30 15:34:32 UTC 2014-09-30
wayland: Update focus window during DnD motion

Author: Carlos Garnacho
Author Date: 2014-09-30 15:15:48 UTC

wayland: Update focus window during DnD motion

But don't emit crossing events, since the DnD grab is in effect.

https://bugzilla.gnome.org/show_bug.cgi?id=728030

wip/dnd-surface 2014-09-01 21:34:33 UTC 2014-09-01
dnd 2

Author: Jasper St. Pierre
Author Date: 2014-08-21 21:50:46 UTC

dnd 2

wip/background-rework 2014-08-28 20:33:51 UTC 2014-08-28
Rewrite background code

Author: Owen Taylor
Author Date: 2014-08-10 15:26:14 UTC

Rewrite background code

The old requirement that multiple MetaBackgroundActor objects be
layered on top of each to produce blended backgrounds resulted in
extremely inefficient drawing since the entire framebuffer had
to be read and written multiple times.

 * Replace the MetaBackground ClutterContent with a plain GObject
   that serves to hold the background parameters and prerender
   textures to be used to draw the background. It handles
   colors, gradients, and blended images, but does not handle
   vignetting

 * Add vignetting to MetaBackgroundActor directly.

 * Add MetaBackgroundImage and MetaBackgroundImageCache to allow
   multiple MetaBackground objects to share the same images

By removing the usage of ClutterContent, the following optimizations
were easy to add:

 Blending is turned off when the actor is fully opaque
 Nearest-neighbour filtering is used when drawing 1:1

The GLSL vignette code is slightly improved to use a vertex shader
snippet for computing the texture coordinate => position in actor
mapping.

https://bugzilla.gnome.org/show_bug.cgi?id=735637

wip/quadbuffer-stereo 2014-07-09 03:09:12 UTC 2014-07-09
Switch to using ClutterStage::after-paint

Author: Owen Taylor
Author Date: 2014-06-27 18:33:01 UTC

Switch to using ClutterStage::after-paint

The experimental API clutter_stage_set_paint_callback() is replaced
with an ::after-paint signal as of Clutter 1.19.5.

wip/gestures 2014-06-25 14:38:29 UTC 2014-06-25
compositor: emit grab-op-begin/end signals on plugin grabs

Author: Carlos Garnacho
Author Date: 2014-06-25 14:38:29 UTC

compositor: emit grab-op-begin/end signals on plugin grabs

This makes it possible to track there grabs triggered this way, in
addition to the ones handled by mutter.

wip/xtoglsync 2014-06-04 08:52:56 UTC 2014-06-04
debug output

Author: Rui Matos
Author Date: 2014-06-04 08:52:56 UTC

debug output

wip/xtoglsync-3-10 2014-05-14 12:48:10 UTC 2014-05-14
MetaSyncRing: disable after a number of reboot attempts

Author: Rui Matos
Author Date: 2014-05-13 10:47:31 UTC

MetaSyncRing: disable after a number of reboot attempts

If we have had to reboot this number of times, something is definitely
wrong and we're likely to just make things worse by continuing to try,
so let's err on the side of caution and disable ourselves.

wayland 2014-04-15 20:46:38 UTC 2014-04-15
Bump version to 3.12.1

Author: Florian Müllner
Author Date: 2014-04-15 20:00:40 UTC

Bump version to 3.12.1

Update NEWS.

wip/split-cursor-tracker 2014-04-07 17:23:35 UTC 2014-04-07
cursor-tracker: Properly handle invisible cursors

Author: Rui Matos
Author Date: 2014-04-07 16:40:41 UTC

cursor-tracker: Properly handle invisible cursors

We might be asked to not display the cursor ourselves but we should
still emit cursor-changed signals and hand out the correct sprite and
hot point information when asked so that consumers can still draw an
updated cursor themselves.

This fixes the cursor not showing up in gnome-shell's zoom mode.

wip/ui-on-own-connection 2014-04-06 21:43:43 UTC 2014-04-06
ff

Author: Jasper St. Pierre
Author Date: 2014-04-06 20:16:21 UTC

ff

wip/icons 2014-03-19 01:27:31 UTC 2014-03-19
icons

Author: Jasper St. Pierre
Author Date: 2012-12-03 19:52:36 UTC

icons

wip/compositor-menu 2014-03-14 14:22:13 UTC 2014-03-14
wayland: Add support for show_window_menu

Author: Jasper St. Pierre
Author Date: 2014-03-13 20:54:04 UTC

wayland: Add support for show_window_menu

wip/login1 2014-03-11 22:01:29 UTC 2014-03-11
hack to get things working for demo

Author: Jasper St. Pierre
Author Date: 2014-03-11 21:38:03 UTC

hack to get things working for demo

have to figure out something better to put here

wip/resize-2 2014-03-06 18:07:02 UTC 2014-03-06
wip/resize

Author: Giovanni Campagna
Author Date: 2014-03-06 00:30:54 UTC

wip/resize

doesn't work...

wip/resize 2014-03-05 22:52:57 UTC 2014-03-05
resize

Author: Jasper St. Pierre
Author Date: 2014-02-19 21:24:38 UTC

resize

wip/surface-two 2014-02-25 00:16:13 UTC 2014-02-25
Fix input and bounding shapes

Author: Giovanni Campagna
Author Date: 2014-02-25 00:06:26 UTC

Fix input and bounding shapes

1) We need to select for shape events
2) For decorated windows, we don't want to apply any input
   shape, because the frame is always rectangular and eats
   all the input
3) For undecorated windows, the "has input shape" check is
   wrong if the window has a bounding shape but not an input
   shape

wip/no-reparent 2014-02-20 22:30:37 UTC 2014-02-20
use gtk+ theme impl

Author: Jasper St. Pierre
Author Date: 2013-04-23 16:53:24 UTC

use gtk+ theme impl

gnome-3-10-wayland 2014-02-19 18:52:25 UTC 2014-02-19
Bump version to 3.10.4

Author: Florian Müllner
Author Date: 2014-02-19 18:52:25 UTC

Bump version to 3.10.4

Update NEWS.

wip/surface-content 2014-02-19 15:24:38 UTC 2014-02-19
window-actor: Split into two subclasses of MetaSurfaceActor

Author: Jasper St. Pierre
Author Date: 2014-02-01 22:21:11 UTC

window-actor: Split into two subclasses of MetaSurfaceActor

The rendering logic before was somewhat complex. We had three independent
cases to take into account when doing rendering:

  * X11 compositor. In this case, we're a traditional X11 compositor,
    not a Wayland compositor. We use XCompositeNameWindowPixmap to get
    the backing pixmap for the window, and deal with the COMPOSITE
    extension messiness.

    In this case, meta_is_wayland_compositor() is FALSE.

  * Wayland clients. In this case, we're a Wayland compositor managing
    Wayland surfaces. The rendering for this is fairly straightforward,
    as Cogl handles most of the complexity with EGL and SHM buffers...
    Wayland clients give us the input and opaque regions through
    wl_surface.

    In this case, meta_is_wayland_compositor() is TRUE and
    priv->window->client_type == META_WINDOW_CLIENT_TYPE_WAYLAND.

  * XWayland clients. In this case, we're a Wayland compositor, like
    above, and XWayland hands us Wayland surfaces. XWayland handles
    the COMPOSITE extension messiness for us, and hands us a buffer
    like any other Wayland client. We have to fetch the input and
    opaque regions from the X11 window ourselves.

    In this case, meta_is_wayland_compositor() is TRUE and
    priv->window->client_type == META_WINDOW_CLIENT_TYPE_X11.

We now split the rendering logic into two subclasses, which are:

  * MetaSurfaceActorX11, which handles the X11 compositor case, in that
    it uses XCompositeNameWindowPixmap to get the backing pixmap, and
    deal with all the COMPOSITE extension messiness.

  * MetaSurfaceActorWayland, which handles the Wayland compositor case
    for both native Wayland clients and XWayland clients. XWayland handles
    COMPOSITE for us, and handles pushing a surface over through the
    xf86-video-wayland DDX.

Frame sync is still in MetaWindowActor, as it needs to work for both the
X11 compositor and XWayland client cases. When Wayland's video display
protocol lands, this will need to be significantly overhauled, as it would
have to work for any wl_surface, including subsurfaces, so we would need
surface-level discretion.

https://bugzilla.gnome.org/show_bug.cgi?id=720631

wip/xdg-shell-props 2014-02-18 23:57:12 UTC 2014-02-18
xdg-shell: Update for new state change mechanism

Author: Jasper St. Pierre
Author Date: 2014-02-09 16:45:09 UTC

xdg-shell: Update for new state change mechanism

We're still not properly going through the request system. This
will require a dense investigation of the code, but it will happen
soon...

wip/subsurfaces 2014-01-16 01:05:31 UTC 2014-01-16
double-buffer place_above/place_below

Author: Jasper St. Pierre
Author Date: 2013-11-26 01:26:00 UTC

double-buffer place_above/place_below

wip/csd-flicker 2014-01-10 19:07:45 UTC 2014-01-10
window: Assume CSD frame extents are 0 in maximized and tiling modes

Author: Jasper St. Pierre
Author Date: 2014-01-10 18:25:54 UTC

window: Assume CSD frame extents are 0 in maximized and tiling modes

When the user maximized a CSD window, this is the order of operations
that happens:

  1. The CSD application sends us a _NET_WM_STATE client message telling
     us to add the _NET_WM_MAXIMIZED_VERT/HORZ state.

  2. We respond by setting _NET_WM_STATE on the window and resizing the
     window. Windows are positioned according to the visible borders of
     the window, so since _GTK_FRAME_EXTENTS has not been updated yet to
     reflect the border/padding of the window state, the borders are wrong.

  3. GTK+ gets the _NET_WM_STATE event, updates its state, and then sets
     new theme properties, including _GTK_FRAME_EXTENTS.

  4. mutter sees that _GTK_FRAME_EXTENTS has been updated, and repositions
     the window to where it should be.

This means that the user sees the borders "flicker" for a half-second while
the maximization state is in limbo. Trying to make this resize operation
atomic is a foregone conclusion under X11, so to solve this, we're going to
make two changes to our _GTK_FRAME_EXTENTS protocol:

  1. _GTK_FRAME_EXTENTS now reflects the "normal" borders, always. This
     means that _GTK_FRAME_EXTENTS should not change at runtime unless the
     theme changes.

  2. mutter will only respect _GTK_FRAME_EXTENTS in places where the user
     expects to see a shadow. This means that mutter will hardcode the
     extents to 0 when the window is maximized. If the window is tiled-left,
     then all extents will be 0, except for the shadow on the right hand
     side, and vice versa for the tiled-right state.

wip/remove-map 2013-12-09 20:49:50 UTC 2013-12-09
compositor: Remove meta_compositor_window_[un]mapped

Author: Jasper St. Pierre
Author Date: 2013-12-09 20:42:31 UTC

compositor: Remove meta_compositor_window_[un]mapped

We no longer unmap the toplevel windows during normal operation. The
toplevel state is tied to the window's lifetime.

Call meta_compositor_add_window / meta_compositor_remove_window instead...

wip/mult-window-actors 2013-12-06 21:24:01 UTC 2013-12-06
window-actor: Don't do any special initialization in _new

Author: Jasper St. Pierre
Author Date: 2013-12-06 06:00:13 UTC

window-actor: Don't do any special initialization in _new

https://bugzilla.gnome.org/show_bug.cgi?id=678989

wip/culling 2013-11-25 20:08:22 UTC 2013-11-25
window: Separate pointer-focus operations and explicit-focus operations

Author: Jasper St. Pierre
Author Date: 2013-11-22 17:12:49 UTC

window: Separate pointer-focus operations and explicit-focus operations

Some clients, like on-screen keyboards, don't want their windows to be
activated on click, as that would steal focus from the window they're
trying to send events to. They do this by setting the Input Hint in
WM_HINTS to be FALSE, along with not specifying WM_TAKE_FOCUS in their
WM_PROTOCOLS.

However, in this case, when a window tries to get focus in this scenario,
we focus the frame so that a11y and key navigation works properly.
Both policies aren't acceptable -- we can't make OSK steal focus from
windows on click, and we can't make an OSK un-Alt-Tabbable-to.

To solve this, split meta_window_focus() into two different focus policies:

 * meta_window_focus_implicitly() should be called on pointer click or
   focus-follows-mouse or similar cases where we may not want to forcibly
   set focus for clients that don't want it.

 * meta_window_focus_explicitly() should be called by pagers, like
   the gnome-shell overview, or Alt-Tab. In this case, most of the existing
   clients are using meta_window_activate(), so simply adapting that so
   it calls meta_window_focus_explicitly() should be enough.

https://bugzilla.gnome.org/show_bug.cgi?id=715030

wip/wayland-clutter-events-2 2013-11-02 16:53:32 UTC 2013-11-02
wayland: Support alternative focus modes like focus-follows-mouse

Author: Jasper St. Pierre
Author Date: 2013-10-16 04:06:10 UTC

wayland: Support alternative focus modes like focus-follows-mouse

Use the existing code for MetaWindow focus-follows-mouse to support this.

wip/wayland-work 2013-10-03 19:11:13 UTC 2013-10-03
wayland: add support for pointer barriers

Author: Giovanni Campagna
Author Date: 2013-09-06 12:14:31 UTC

wayland: add support for pointer barriers

Use the clutter pointer constrain callback and a lot of copypasted
code from Xorg to implement reactive pointer barriers and pointer
barrier events.

https://bugzilla.gnome.org/show_bug.cgi?id=706655

wip/wayland-clutter-events 2013-09-05 11:19:58 UTC 2013-09-05
wayland: Use an event filter instead of the captured event signal

Author: Neil Roberts
Author Date: 2013-09-05 11:19:58 UTC

wayland: Use an event filter instead of the captured event signal

In order to see all Clutter events, Mutter was previously installing a
signal handler on the ‘captured-event’ signal on the stage and
additionally using a signal emission hook to cope with grabs. This
changes it to use the new clutter_event_add_filter API so that we can
avoid the signal emission hook hack.

wip/new-api 2013-09-05 08:40:12 UTC 2013-09-05
MetaPluginManager: don't require plugins to pass events to clutter

Author: Giovanni Campagna
Author Date: 2013-09-04 16:11:43 UTC

MetaPluginManager: don't require plugins to pass events to clutter

We don't want the shell to know if it's running on ClutterX11 or
not, so we should forward the event ourselves.

https://bugzilla.gnome.org/show_bug.cgi?id=707482

wip/wayland-opaque 2013-08-28 17:23:09 UTC 2013-08-28
shaped-texture: Turn blending off when drawing entirely opaque regions

Author: Jasper St. Pierre
Author Date: 2013-08-27 20:17:34 UTC

shaped-texture: Turn blending off when drawing entirely opaque regions

When drawing entirely opaque regions, we traditionally kept blending on
simply because it made the code more convenient and obvious to handle.
However, this can cause lots of performance issues on GPUs that aren't
too powerful, as they have to readback the buffer underneath.

Keep track of the opaque region set by windows (through _NET_WM_OPAQUE_REGION,
Wayland opaque_region hints, standard RGB32 frame masks or similar), and draw
those rectangles separately through a different path with blending turned off.

https://bugzilla.gnome.org/show_bug.cgi?id=706930

wip/wayland-bare-metal 2013-08-28 09:12:44 UTC 2013-08-28
Add keybindings for switching VTs

Author: Giovanni Campagna
Author Date: 2013-07-16 12:37:49 UTC

Add keybindings for switching VTs

Once mutter is started from weston-launch on its own VT, there is
no way to change VT again (for example to actually start an application),
because the keyboard is put in raw mode.
So introduce some keybindings mimicking the standard X ones (Ctrl+Alt+Fn)
that switch the VT manually when activated.

https://bugzilla.gnome.org/show_bug.cgi?id=705861

wip/tower 2013-08-20 22:10:54 UTC 2013-08-20
remove texture tower

Author: Jasper St. Pierre
Author Date: 2013-08-20 20:45:26 UTC

remove texture tower

wip/display-no-wayland 2013-08-17 14:57:33 UTC 2013-08-17
MonitorXrandr: follow the right order in applying the new configuration

Author: Giovanni Campagna
Author Date: 2013-08-16 17:02:57 UTC

MonitorXrandr: follow the right order in applying the new configuration

First disable CRTCs that should be off in the new configuration,
then resize the framebuffer, then enable the new CRTCs.
If we don't do that, and we're making the screen smaller, X complains
with a BadMatch.

https://bugzilla.gnome.org/show_bug.cgi?id=705670

wip/wayland-display 2013-08-14 16:42:07 UTC 2013-08-14
MetaWaylandSeat: reset pointer focus always when it might be different

Author: Giovanni Campagna
Author Date: 2013-08-14 14:29:09 UTC

MetaWaylandSeat: reset pointer focus always when it might be different

It can happen because of a grab that the current surface changes
but the focus doesn't. Later on, when the grab ends, we would keep
the previous focus until the user goes out of the window and
back in again.

focus-new 2013-05-28 19:07:13 UTC 2013-05-28
compositor: Add an API to query if the stage is focused

Author: Jasper St. Pierre
Author Date: 2013-05-24 21:58:55 UTC

compositor: Add an API to query if the stage is focused

gnome-shell needs to know whether the stage window is focused so
it can synchronize between stage window focus and Clutter key actor
focus. Track all X windows, even those without MetaWindows, when
tracking the focus window, and add a compositor-level API to determine
when the stage is focused.

wip/is-switch-fixes 2013-04-04 10:00:08 UTC 2013-04-04
keybindings: Fix ungrabbing of keys

Author: Rui Matos
Author Date: 2013-03-31 17:06:40 UTC

keybindings: Fix ungrabbing of keys

XUngrabKey() doesn't work for XI2 grabs and XI2 doesn't provide API
with similar functionality. As such, we have to refactor the code a
bit to be able to call XIUngrabKeycode() for each key binding, then
reload keybindings and finally grab the new ones.

https://bugzilla.gnome.org/show_bug.cgi?id=697003

wip/gcampax/background 2013-02-13 21:30:15 UTC 2013-02-13
MetaBackgroundActor: stop waiting synchronously on the first frame

Author: Giovanni Campagna
Author Date: 2013-02-13 21:19:54 UTC

MetaBackgroundActor: stop waiting synchronously on the first frame

We can draw the gradient while we wait for the wallpaper to load.
This also removes the dependency on g_task_wait_sync(), which may
never appear in glib.

wip/frame-synchronization 2013-01-30 20:47:52 UTC 2013-01-30
Consistently use meta_grab_op_is_resizing() for _NET_WM_SYNC_REQUEST

Author: Owen Taylor
Author Date: 2013-01-30 19:49:31 UTC

Consistently use meta_grab_op_is_resizing() for _NET_WM_SYNC_REQUEST

In different places we checked the grab op differently when determing
whether we are using _NET_WM_SYNC_REQUEST. This was somewhat covered
up previously by the fact that we only had a sync alarm when using
_NET_WM_SYNC_REQUEST, but that is no longer the case, so consistently
use meta_grab_op_is_resizing() everywhere.

wip/media-keys 2012-11-10 16:06:29 UTC 2012-11-10
Keybindings: uniquify the name for non-builtin keybindings

Author: Giovanni Campagna
Author Date: 2012-06-23 15:16:37 UTC

Keybindings: uniquify the name for non-builtin keybindings

Multiple settings objects could have the same key, so that alone
is not enough to identify the binding. Add also the pointer value
of the GSettings object.

https://bugzilla.gnome.org/show_bug.cgi?id=613543

wip/multitouchr 2012-11-04 00:00:21 UTC 2012-11-04
window: Update drag hotspot as new touches approach

Author: Carlos Garnacho
Author Date: 2011-10-30 17:05:04 UTC

window: Update drag hotspot as new touches approach

Any number of touches >= 3 starts a drag operation, but the
window jumped around as new touches joined. So recalculate
the touch drag area/hotspot and tiling mode as expected
as soon as touches enter/leave the drag op.

wip/xinput2r 2012-11-03 23:47:37 UTC 2012-11-03
frame: disallow frame operations if the popup is open

Author: Carlos Garnacho
Author Date: 2011-08-01 23:44:10 UTC

frame: disallow frame operations if the popup is open

If some device has a popped up menu, meta_frame_get_flags()
returns 0 to indicate other devices cannot trigger any
actions on the frame.

wip/message-tray 2012-07-18 15:06:43 UTC 2012-07-18
display: Unset grab_frame_action in meta_display_end_grab_op

Author: Debarshi Ray
Author Date: 2012-07-18 15:04:45 UTC

display: Unset grab_frame_action in meta_display_end_grab_op

Otherwise, if the user had moved the windows around using a mouse,
they can not be moved beyond the top edge of the primary monitor when
the tray needs to be brought up, . This is because
constrain_titlebar_visible looks at grab_frame_action and thinks that
the movement is being caused by the user trying to move the windows up
by grabbing the frame.

Fixes: https://bugzilla.gnome.org/677215

wip/cb2eb3 2012-05-07 23:37:09 UTC 2012-05-07
Remove legacy theming API

Author: Jasper St. Pierre
Author Date: 2012-05-05 05:48:14 UTC

Remove legacy theming API

Since the move to GTK+ toplevels, it required removing the preview-widget
and theme-viewer parts of the code. (They didn't work with the new CSS
stuff anyway... shhh..)

wip/gmenu 2011-12-05 17:54:29 UTC 2011-12-05
Add _DBUS_UNIQUE_NAME and _OBJECT_PATH properties

Author: Allison Lortie
Author Date: 2011-12-05 17:52:35 UTC

Add _DBUS_UNIQUE_NAME and _OBJECT_PATH properties

https://bugzilla.gnome.org/show_bug.cgi?id=664851

wip/multitouch 2011-10-31 00:09:36 UTC 2011-10-31
window: Update drag hotspot as new touches approach

Author: Carlos Garnacho
Author Date: 2011-10-30 17:05:04 UTC

window: Update drag hotspot as new touches approach

Any number of touches >= 3 starts a drag operation, but the
window jumped around as new touches joined. So recalculate
the touch drag area/hotspot and tiling mode as expected
as soon as touches enter/leave the drag op.

wip/xinput2 2011-08-02 16:39:33 UTC 2011-08-02
frame: disallow frame operations if the popup is open

Author: Carlos Garnacho
Author Date: 2011-08-01 23:44:10 UTC

frame: disallow frame operations if the popup is open

If some device has a popped up menu, meta_frame_get_flags()
returns 0 to indicate other devices cannot trigger any
actions on the frame.

configurable-shadows 2010-11-13 16:44:47 UTC 2010-11-13
MetaWindowActor: Fix crashes when mapping and unmapping windows

Author: Owen Taylor
Author Date: 2010-11-12 01:05:35 UTC

MetaWindowActor: Fix crashes when mapping and unmapping windows

The assumptions made when getting the size of the window for the
paint volume turned out not to be accurate in all cases -
get_paint_volume() could be called on windows without computed
bounds.

https://bugzilla.gnome.org/show_bug.cgi?id=592382

override-redirect-exclusion 2009-06-27 13:58:45 UTC 2009-06-27
Avoid restacking animating hidden actors

Author: Owen Taylor
Author Date: 2009-06-26 21:05:11 UTC

Avoid restacking animating hidden actors

Since the stack passed to the compositor now accurately reflects
the X stacking order, we need to treat hidden windows (which are
at the bottom of the X stacking order) specially - when the
compositor stacking order is synced, try to keep animating hidden
actors in their old positions in the stack.

http://bugzilla.gnome.org/show_bug.cgi?id=585984

185 of 85 results

Other repositories

Name Last Modified
lp:mutter 11 hours ago
lp:~vanvugt/mutter 2017-10-18
12 of 2 results
You can't create new repositories for Mutter.