GrabHelper already takes care of putting the key focus back on the
widget that had it when the grab was established.
ShellEntry's close() has recently become harmful since it's now called
from the menu's destroy() method and that is called in the entry's
destroy handler which means that this._entry might no longer be valid
at close() time.
https://bugzilla.gnome.org/show_bug.cgi?id=697560
The actor is the same as the drag actor in this case, but we tend to
set properties on the drag actor and get them from the actor elsewhere
in this codepath. Make this consistent.
https://bugzilla.gnome.org/show_bug.cgi?id=697504
It turns out that we never destroyed modal dialogs when closing
them, causing them to still linger in the scene graph even when
there were no references to them in the JS. The one case where
we don't want to destroy modal dialogs after being closed is
endSessionDialog, so provide a parameter that allows classes
to override this behavior.
https://bugzilla.gnome.org/show_bug.cgi?id=697295
We chain up on _finish() to drop the grab and destroy the switcher
popup but we should activate the target window first because dropping
the grab results in the previously focused windows getting a focus in
event immediately followed by the focus out event from the target
window activation which we can easily avoid.
https://bugzilla.gnome.org/show_bug.cgi?id=696259
This ensures that when we have windows that are already visible,
like desktop icons, they don't fly across the screen from what
seems to be hyperspace to get into view.
https://bugzilla.gnome.org/show_bug.cgi?id=696323
if a slideshow file has a really large duration we'll currently
throw an exception.
This bug is aggravated by the fact that some versions of
gnome-desktop use UINT_MAX as a sentinel value to mean,
"don't ever update slide".
This commit treats durations that would overflow as infinitely
long.
https://bugzilla.gnome.org/show_bug.cgi?id=696157
If a background gets requested from the cache while it's still
being loaded from an earlier call, then there will be two concurrent
loads of the same file.
That concurrency is mitigates the effectiveness of the cache and
also causes leaks.
This commit consolidates file loads so that concurrency doesn't
happen.
https://bugzilla.gnome.org/show_bug.cgi?id=696157
Right now we only destroy the bgManager object when the
workspaceThumbnail is explicitly destroy with its destroy()
method.
This commit makes sure bgManager gets destroyed when the
workspaceThumbnail actor is destroyed without calling
destroy().
https://bugzilla.gnome.org/show_bug.cgi?id=696157
BackgroundManager connects to the changed signal in
the backgrounds it manages.
The signal ids for the changed signal connectionss are stored
as state on the background manager object.
If the background being managed changes while the manager
is still loading the old background, then the signal id
variable can get out of sync with the background object being
managed.
This commit ties the signal id to the background objects themselves,
so there is no opportunity for them to desynchronize.
https://bugzilla.gnome.org/show_bug.cgi?id=696157
Normally backgrounds get evicted from the cache when their actor is
destroyed. If the actor changes content without destroying itself,
though, we should evict the old content from the cache, too.
https://bugzilla.gnome.org/show_bug.cgi?id=696157
We currently only add the first instance of a background
to the cache. This means if the actor associated with that
background is destroyed, the content will be evicted and
it will need to get reloaded, even if it's already loaded
on another actor.
This commit ensures every content gets added to the cache.
https://bugzilla.gnome.org/show_bug.cgi?id=696157
In the (no barriers) fallback case handleDragOver has somehow
ended up being turned into a nop and thus breaking xdnd
overview opening.
Fix that by calling _toggleOverview() when a xdnd source triggers
it.
https://bugzilla.gnome.org/show_bug.cgi?id=696447
When commit 724a2bd7 changed the way to determine the default
calendar app, it dropped all special handling of evolution.
Unfortunately we still need it to not end up with the default
mail component, so add it back.
https://bugzilla.gnome.org/show_bug.cgi?id=696432
Currently, if a button-press event results in releasing the last modal
grab (e.g. clicks outside the grabbed actors), we don't consider the
event handled and allow its emission to continue. If we consider
dismissing a grab as an action of its own, any additional action
triggered by the same event becomes an unexpected side effect.
Tweak the capture handler accordingly.
https://bugzilla.gnome.org/show_bug.cgi?id=696422
There is no reason why there shouldn't be a battery icon in initial-setup,
knowing how much battery you have left is useful in during setup as well.
This also fixes an exception in the lock screen caused by the combined icon
not finding the battery one.
https://bugzilla.gnome.org/show_bug.cgi?id=696212
There's large performance issues with both the blur and desaturation
that make the screen shield hard to use on slower computers, and this
has always been a temporary stopgap until the user can pick a different
image for the lock screen.
https://bugzilla.gnome.org/show_bug.cgi?id=696322
Right now, if multiple messages come in, they just sort of
clobber each other.
This commit sets up a message queue, and introduces pauses
long enough for the user to hopefully be able to read those
messages.
https://bugzilla.gnome.org/show_bug.cgi?id=694688
GDM sends a reset signal after verification succeeds
so that a user-switched login screen can prepare for
the next time it's going to be used.
The unlock screen treats resets as failures, though.
This means, on success, we're emitting "failed" and
clearing any last second messages.
This commit changes the unlock code to ignore resets from
GDM after successful verification.
https://bugzilla.gnome.org/show_bug.cgi?id=694688
Right, the common code between the login screen and
the unlock screen handles clearing the user verifier
when GDM sends a reset.
We don't actually always want to clear the messages on
reset in the unlock case, though, so doing it implicitly
is problematic.
This commit moves the clear() call from the common code
to the specific reset handlers.
https://bugzilla.gnome.org/show_bug.cgi?id=694688
Since we drop our event handler, we won't see ever the release event
from the button press, so unset the flag so it doesn't muck with the
next time somebody takes a grab.
https://bugzilla.gnome.org/show_bug.cgi?id=696102
Getting fullscreen window tracking right in GNOME Shell turned out to
be very hard, because it depended on details both how Mutter handled
fullscreen windows and the exact timing of that. Fullscreen tracking
and auto-minimization of fullscreen windows that lose their fullscreen
status has thus been implemented in Mutter: use that.
https://bugzilla.gnome.org/show_bug.cgi?id=649748
Previously when a client requests that a window should be docked the
shell would reparent the socket window onto the stage's window and
then use ClutterX11TexturePixmap to get a texture to represent the
window. This will not work if Clutter is no longer using the X11
winsys for example if it becomes its own display server. Instead this
patch leaves the socket window as a child of the root window and lets
mutter create a MetaWindow out of it. If Mutter is acting as a display
server then this mechanism will still work via the headless x server.
The ShellGtkEmbed instance now registers for notification of the
‘window-created’ signal of the display so that it can find the
MetaWindow that gets created to represent the socket window. When this
window is found it is prevented from being displayed on the screen by
setting the actor's opacity to 0. An input shape is then set on the
window to prevent it receiving any input.
Instead of being a subclass of ClutterX11TexturePixmap, ShellGtkEmbed
is now a subclass of ClutterClone. When the MetaWindow is found for
the socket window the clone's source is set to the invisible actor for
the window so it can be displayed in the panel as before.
The ShellEmbeddedWindow no longer needs to know what the stage is
because it no longer reparents the socket window. Therefore the
ShellTrayManager doesn't need to know the stage either so
shell_tray_manager_manage_stage has been replaced with just
shell_tray_manager_manage_screen.
https://bugzilla.gnome.org/show_bug.cgi?id=693438
To create a new workspace by dropping on the placeholder, we move all
the windows down one workspace and then wait for _checkWorkspaces() to
automatically create the empty one at the end.
This means that, from the implementation POV, the new workspace is not
the one created by DND but it's the last one, and this detail was exposed
in the UI because the animation was applied on that one.
Fix that by starting the animation manually from the DND code, and then
blocking the animation from happening when the new workspace is created
with a flag.
https://bugzilla.gnome.org/show_bug.cgi?id=685285
DND of windows has a lot of side effects, including the possibility of
current workspace disappering from under our feet. We need to account
for that when trying to activate it.
https://bugzilla.gnome.org/show_bug.cgi?id=685285
This happens in the case of Workspace/WorkspaceThumbnail: they call
meta_window_change_workspace_by_index(), which fires window-removed
on the old workspace, thus destroying the window clone.
https://bugzilla.gnome.org/show_bug.cgi?id=685285
Windows can be restacked at any time, including when the stackAbove
property of the window clone is being dragged, and thus parented to
the uiGroup. To do stacking properly, we need to skip it for the duration
of the drag, and sync it again at the end (which is already done by
mutter because of the workspace change)
https://bugzilla.gnome.org/show_bug.cgi?id=685285
ClutterBinLayout uses the maximum width/height of all children in
size requests and positions children according to the expand/align
properties. This means that the vertical position of folder popups
is not considered in size requests, so if the main view is smaller
than a folder popup's height and offset, the popup will be truncated
and/or mispositioned. Fix those cases by using a custom LayoutManager
that behaves like ClutterFixedLayout for height requests and like
ClutterBinLayout otherwise.
https://bugzilla.gnome.org/show_bug.cgi?id=694371
We need to look at the duration of the current step of the slideshow
to determine when to next queue an event, rather than the full slideshow
duration.
https://bugzilla.gnome.org/show_bug.cgi?id=695882
Due to weird and strange JS scoping semantics, if we are in a
callback, "i" won't be captured and when the callback is called,
we'll have the wrong index, causing addImage to be called instead
of updateImage.
https://bugzilla.gnome.org/show_bug.cgi?id=695882
The pattern is underneath, so we need to add one. It turns out
that ClutterGroup doesn't care about that, but we need to do this
to remove deprecations in mutter.
https://bugzilla.gnome.org/show_bug.cgi?id=695882
To make sure that the screen shield is shown before suspending, we
take a logind inhibitor and release it when the screen shield is
shown. As the screen shield is not only shown on suspend, we can end
up releasing the inhibitor independently from suspending (lock, idle),
in which case the screen might not be locked when we do suspend.
To fix, only release the inhibitor after showing the screen shield
when we are about to suspend.
https://bugzilla.gnome.org/show_bug.cgi?id=693708
Since commit 7cdb75e7ce, initializing UI is deferred until the session
mode has been loaded. However DBus is still initialized immediately,
which means that for DBus methods that access properties in Main, there
is now a window between the method being exposed on the bus and the
method being ready to be called. At least g-s-d grabbing global keybindings
is likely to fall in this window on session startup, and almost guaranteed
when regrabbing bindings after a shell restart.
To fix, defer initializing DBus as well.
https://bugzilla.gnome.org/show_bug.cgi?id=694321
When we switch into an invalid input source we hide the panel
indicator and return early but we were not hiding the previously
active source label and its menu item dot and thus when switching
again to a third input source we would end up showing 2 overlapping
labels in the panel and the menu would have 2 entries with a dot.
https://bugzilla.gnome.org/show_bug.cgi?id=695841
It's possible in some corner cases for the status of the topwindow
to change and make it not fullscreen without ::restacked being
changed. One way that it could happen with the old code was if the
layer of the top window changed from NORMAL to FULLSCREEN.
Change the logic not to look at the layer, which is a function of
Mutter's *intended* stacking, rather than the *actual* stacking,
which is what ::restacked gives you. Instead, look at the top
portion of the stack, down to the first non-override-redirect
window, and see if their are any monitor-sized windows there.
Connect to changes on the top portion of the stack, so we know
if conditions change.
https://bugzilla.gnome.org/show_bug.cgi?id=649748
We always add external DOCK windows to the ctrl-alt-tab switcher,
e.g. separate dock applications or nautilus' desktop windows.
Since commit 1f46a0dc26, all items in the switcher are expected
to set a proxy parameter, but the aforementioned code was not
updated accordingly.
https://bugzilla.gnome.org/show_bug.cgi?id=695395
* Prompters have state, and cancelling an already prompter will
cause prompts that are in progress to fail.
* In addition allow replacement of our shell prompter for debugging
purposes.
https://bugzilla.gnome.org/show_bug.cgi?id=695485
notify::* doesn't guarantee that the value has changed, only that it
may have been. We need to ensure that we track the old value to make sure
we don't do things like overwrite timeouts if they already exist.
https://bugzilla.gnome.org/show_bug.cgi?id=695659
Now that the tray is modal, the summary is tied to the tray,
and we don't need to have separate states for the tray and
summary. This also removes the nearly invisible opacity tween
on the summary items when opening the message tray.
https://bugzilla.gnome.org/show_bug.cgi?id=695659
The only way that locking happens is with when the summary box
pointer is active. As it can only happen if the summary state
is active, it's impossible for a notification to be expired,
or the summary to be hidden while it's showing.
https://bugzilla.gnome.org/show_bug.cgi?id=695659
It makes more sense to use the monitor the tray is on, rather than the
primary monitor. This also matches us with whether we can open the tray
from a barrier/dwell or not.
https://bugzilla.gnome.org/show_bug.cgi?id=695659
We may release the focus grab at any time, so it's not guaranteed
we'll be in event processing. In particular, hovering over and out
of a notification will cause this to happen, as the notification
is hidden on a timeout.
https://bugzilla.gnome.org/show_bug.cgi?id=695659
For the same reasons that we disable the tray barrier, we should
disable hot corners as well -- when users have a full-screen game
open, we shouldn't allow overview activation by the hotcorner.
https://bugzilla.gnome.org/show_bug.cgi?id=694997
gnome-panel is going away in 3.8, so we can't rely on it to provide our
friendly and reliable companion. But no regret, because we can ship it
ourselves, and at the same time remove some unnecessary configuration.
https://bugzilla.gnome.org/show_bug.cgi?id=695526
Launching the run dialog to open the looking glass or something
like that shouldn't install a bunch of file monitors that monitor
every IO change to the home and system directories.
Instead, simply scan all the paths when trying to complete.
https://bugzilla.gnome.org/show_bug.cgi?id=695338
With fallback mode gone, we can no longer rely on gnome-screensaver
being installed. Rather than handling three different cases (GDM,
gnome-screensaver, no lock), disable the lock functionality when
not running under GDM.
https://bugzilla.gnome.org/show_bug.cgi?id=693403
With fallback mode dropped, we can no longer rely on gnome-screensaver
to be installed, so we'll have cases where we are unable to lock the
screen. The user menu should not show the "Lock" item in this case,
but as UnlockDialog includes UserMenu, we cannot use the existing check
without creating a circular dependency; move the function to a more
generic place to fix.
https://bugzilla.gnome.org/show_bug.cgi?id=693403
We currently resync the stacking order of the two key frames
every iteration of the animation. This is costly and unnecessary.
This commit ensures they're stacked properly up front and doesn't
touch them after that.
https://bugzilla.gnome.org/show_bug.cgi?id=694993
Rather than expose a dizzying array of methods related to managing
state that require infecting every user of the overview methods, try
to do the sensible and smart thing internally. Now, the overview
itself tracks when XDND drags start, and simply calling show, hide or
toggle while an XDnD drag is in effect will show the overview, and
will only take the grab until after the XDND drag ends.
https://bugzilla.gnome.org/show_bug.cgi?id=663661
We currently call the session updated handler as soon as
the session modes are read. This handler sets up keybindings
for leaving the overview (if a user session) and shows the
login dialog (if a gdm session).
We can't do the latter until the stage is mapped because it
takes a grab, and we don't need to do the former until the
user goes into the overview.
This commit defers processing session updates until the
the layout manager says start up is prepared.
It fixes a race condition at login screen startup now that
we don't show the stage right away.
https://bugzilla.gnome.org/show_bug.cgi?id=694321
This cleans up the code considerably, and makes it so that
one path creates all hot corners for all monitors. Why this
wasn't done originally, I have no clue...
The one complication is debouncing if the button and hot corner
are triggered in rapid succession, so we just move this tracking
to the overview.
https://bugzilla.gnome.org/show_bug.cgi?id=663661
For the HotCorner, we want to have different logic for tossing out
specific events based on the grabbed state, etc. so make us have
to pass in an event filter callback.
For the hot corner case, we want to have the pressure apply both in
and outside of the overview, so we need to move this to the user. At
the same time, use keybinding mode math that's more like what's used
in filterKeybinding.
While it may seem like an abuse of the KeyBindingMode API, it may
become more reasonable if one thinks of the pressure barrier as a
binding of sorts, just applied to the mouse. If a ButtonBinding API
was added to mutter, I think we'd use the existing KeyBindingMode
infastructure there as well.
Ensure that the pointer leaves the barrier before we trigger again.
For the message tray case, this doesn't matter much, as the trigger
won't have any effect after the grab is taken, but in the overview
HotCorner case, this ensures that we don't trigger the overview
transition many times simply by holding pressure against the hot
corner, which is easy to do accidentally.
Instead of sometimes having an event source and sometimes not, use
the empty event source when the session mode says the calendar is
disabled. This way, the code can assume an event source object and
avoid checks.
https://bugzilla.gnome.org/show_bug.cgi?id=641383
Check if the event source is currently doing an async call, and prevent
UI updates in that case. This avoids a flash of "No updates" when switching
months.
https://bugzilla.gnome.org/show_bug.cgi?id=641383
SwitcherPopup relies on being able to pushModal(), setting the stage
input mode to FULLSCREEN, and then doing regular event processing on
the actor it adds to uiGroup. But MessageTray uses GrabHelper which
sets up a 'captured-event' handler on the stage and thus gets all
events itself.
This, of course, breaks the switcher if it's brought up in the message
tray so, for now, we'll just prevent it from being used there.
https://bugzilla.gnome.org/show_bug.cgi?id=693907
Toggling the overview during the startup animation reportedly
causes stuck grab and other odd behavior.
There's no reason to handle toggling the overview during this
time anyway.
This commit defers that handling until after startup.
https://bugzilla.gnome.org/show_bug.cgi?id=694837
Curently it is possible to copy the content of password entries,
and paste it elsewhere in clear text. This is undesirable, so
follow GTK+'s behavior and disable the copy action for password
entries.
https://bugzilla.gnome.org/show_bug.cgi?id=695104
We always leave the workspace switcher zoomed out when we can assume
that the user is actually making use of workspaces. For the default
dynamic workspace behavior, we make this assumption when more than
two workspaces are in use (e.g. at least two workspaces contain windows
plus an empty one at the end). However this test does not make sense
when using static workspaces - in that case, not using workspaces
would be indicated by a workspace number of 1 (in which case the
entire switcher is hidden completely), so add a check for dynamic
workspaces to the condition.
https://bugzilla.gnome.org/show_bug.cgi?id=695126
Currently both the app switcher and the thumbnail list divide items
first into two groups (based on whether the item is located on the
current workspace or not), and then sort each group individually
by MRU.
The resulting behavior is often confusing, e.g. when using alt-tab
a second time does not switch back to the original window when the first
invocation involved a workspace switch and the workspace contains
windows of more than one application.
Instead, make the behavior more predictable by sorting both lists
strictly by MRU.
https://bugzilla.gnome.org/show_bug.cgi?id=661156
Shell modal dialogs can take their action on a certain key's
key-release-event. For example on <enter> the affirmative action is
usually run.
Make sure that the key was also pressed on the dialog and we're not
seeing a spurious key-release-event from a key that was pressed before
the dialog was displayed.
https://bugzilla.gnome.org/show_bug.cgi?id=692937
It is possible that a session survives after closing, if any processes
(usually PulseAudio and GConf) are still alive at the end. In that case,
ignore it, as the user is already logged out and there is nothing to lose.
https://bugzilla.gnome.org/show_bug.cgi?id=695002
