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
Implement a basic OSD popup that shows an icon and optionally a label
and a fill level. It is based on the existing OSD implementation in
gnome-settings-daemon, which it will replace.
https://bugzilla.gnome.org/show_bug.cgi?id=613543
Right now we take a still frame of the desktop before showing the
start up animation. This gives us an animation over what was
there before startup.
That's not actually desirable when restarting the shell. We don't
want to animate over undecorated windows, we really want to animate
over the noise texture.
This commit drops the still frames in favor of the noise texture.
https://bugzilla.gnome.org/show_bug.cgi?id=694326
Previously, the overview BoxLayout was sized and positioned explicitly
using the primary monitor coordinates, as its parent was at 0,0 and
with a fixed layout manager. Now we use a bin layout, so we need to
position and size the stack actor, and set the overview boxlayout to
expand.
Take also the occasion to use a MonitorConstraint instead of handling
position and size manually.
https://bugzilla.gnome.org/show_bug.cgi?id=694969
We used to clip the overview group to prevent the dash from sliding into
neighbor monitors, but now it moved to the groupStack, so we must move
the clip too.
https://bugzilla.gnome.org/show_bug.cgi?id=694970
This filters out resident and transient notifications in the normal
case, but just returns the number of unread messages for the Telepathy
implementation.
https://bugzilla.gnome.org/show_bug.cgi?id=687787
We generally want view content centered, in particular where the
view itself is symmetrical. So move the dash to a separate layer
and use a placeholder to account for its size when showing the
window picker, which is the only view where it doesn't make sense
to center the content.
https://bugzilla.gnome.org/show_bug.cgi?id=694261
Commit 43ed66cf26 changed the toplevel overview actor, so it
makes sense to change the show/hide transitions to use that instead
of the existing group, to avoid elements layered on top of the
group being excluded from the transitions.
Windows can sometimes be focused, or appear to be focused, without being
at the exact top of the stack, for example in the case of override-redirect
windows, or with attached modal dialogs. In that case, we should not
try to minimize them (as it creates a loop that makes it impossible to restore
the window)
https://bugzilla.gnome.org/show_bug.cgi?id=694905
Mutter now makes session registration an explicit required
step. This is so we can tell the session manager when
we're ready to move on to the next phase.
This commit calls the new Meta.register_with_session() api after we're
initialized.
https://bugzilla.gnome.org/show_bug.cgi?id=694876
"description" is documented as a valid field for search result metas,
and ListSearchResults implements it, so pass it down to be used.
Also, don't wrap the description in quotes, so that the search provider
can decide if it is an excerpt from the searched text or something else.
And to that extent, set use_markup to true, so that terms can be
highlighted.
https://bugzilla.gnome.org/show_bug.cgi?id=694906
This can happen if you open two or three terminal windows, and then
open the overview -- they're not centered. The issue is that because
of the WINDOW_CLONE_MAXIMUM_SCALE clamping, the scale that is being
laid out is different from the scale that the layout was calculated
for.
Implement and document a hack-ish solution which simply keeps the
scale for the layout as originally calculated, but centers the
windows inside the cell.
https://bugzilla.gnome.org/show_bug.cgi?id=694902
Multiplication is linear, so we can split this out as a separate
component. This will make it easier to think of it as an additional
per-window scaling factor, rather than tweaking the scale a bit,
which is more correct to the model.
https://bugzilla.gnome.org/show_bug.cgi?id=694902
While we won't tear down the entire strategy infrastructure, we want to
rework some layout code in the future, so just tear this piece out for
now.
https://bugzilla.gnome.org/show_bug.cgi?id=694902
gnome-shell-extension-prefs supports opening a specific extension's
preferences directly from the command line by passing the UUID.
However this broke when extension loading was changed to be processed
asynchronously, as no extension has been loaded when the command
line argument is processed. Fix by deferring opening the extension's
preferences until all extensions have been loaded.
https://bugzilla.gnome.org/show_bug.cgi?id=694858
Some of the conditions for showing the user list were not
properly inverted, causing the session list to be hidden
when it shouldn't be and shown when it shouldn't be.
https://bugzilla.gnome.org/show_bug.cgi?id=694784
As a special-case to the "cap event" rules, this allows a heavy swipe
from top to bottom to allow triggering the tray without having to push
into it.
https://bugzilla.gnome.org/show_bug.cgi?id=694467
When pressing against the bottom of the screen, we shouldn't
really take more than 15px from each event, to prevent spruious
mouse movements from opening the barrier.
https://bugzilla.gnome.org/show_bug.cgi?id=694467
gnome-settings-daemon will be changed to override the XSetting in
the case where we're on a remote display rather than overwriting a
user setting, so we need to look at the XSetting here.
https://bugzilla.gnome.org/show_bug.cgi?id=694320
Previously, we would create one StBin per monitor, but each was positioned
at 0,0 and sized as the screen, so they would overlap and draw the box shadows
on top of the other backgrounds.
Instead, we need to size appropriately the bin, and then we need to position
the actual MetaBacgroundActor at 0,0, so add a flag to BackgroundManager
for this.
Also, get rid of MetaBackgroundGroup, they do nothing because the screenshield
is not a descendant of the MetaWindowGroup and because the widget in between
blocks the propagation of the visible region. At the same time, use a
widget, not a bin, because StBin requires you to set .child, not call add_child().
https://bugzilla.gnome.org/show_bug.cgi?id=694394
We recently started to trim leading and trailing whitespace from
the search string, and not to trigger a search when the resulting
string was empty. However we still allow whitespace to trigger
type-ahead-find, so that the key focus is moved briefly to the
search entry and back to the stage, resulting in a disruptive
flickering of the entry.
Fix this by excluding whitespace from triggering type-ahead-find.
https://bugzilla.gnome.org/show_bug.cgi?id=694475
xdndHandler used to wait for the stage to show up before calling
global.init_xdnd() . Since commit 65303d027a xdndHandler is initalized
later so the stage might be already visible at this point causing the
show signal never to be emitted and thus global.init_xdnd will never
be called.
Fix that by checking by calling global.init_xdnd without waiting for
the stage's show signal to be emitted, as the stage is guaranted to be
visible at this point.
The key was removed from gsettings-desktop-schemas, GNOME should
always draw a background: the key existed to support xsetroot, but
we no longer read the _XROOTPMAP_ID.
https://bugzilla.gnome.org/show_bug.cgi?id=694463
If the AllView is scrolled, the vertical scrollbar will take away
some horizontal space on one side, resulting in the content ending
up slightly off-center.
Account for this by using overlay-scrollbars instead.
https://bugzilla.gnome.org/show_bug.cgi?id=694261
The frequent view should not be scrolled, but to work around the
icon grid overflowing, we used a (non-scrolling) scroll view anyway.
Now that IconGrid:fillParent allows us to avoid overflow, we can
remove this hack.
https://bugzilla.gnome.org/show_bug.cgi?id=694256
We sometimes map the stage before we've loaded a background on it
because of a race asynchronously loading the session mode.
This manifests as the startup animating starting over a white
background.
This commit defers showing the stage until after the still frames
are loaded.
https://bugzilla.gnome.org/show_bug.cgi?id=694321
It is useful at times to perform several actions that would usually
close the overview (for instance launching an application) at once.
Currently we allow this by dragging items to a workspace rather than
just clicking it, but it's an odd metaphor with its own set of
problems.
Introduce an alternative approach (inspired by file selection in
file managers) by keeping the overview open if a Control key is
held down.
https://bugzilla.gnome.org/show_bug.cgi?id=686984
When the user clicks on "View Source" or "Web Page" in the "Extensions" tab of
looking glass, the callback _onViewSource() or _onWebPage() is called and they
try to close looking glass: this._lookingGlass.close();
But it does not work and generate the exception "this._lookingGlass is
undefined". This patch fixes that.
https://bugzilla.gnome.org/show_bug.cgi?id=693814
animateIn and animateOut should not reset the scale, otherwise
extra animate calls (which are possible because the diff algorithm
in _redisplay is not optimal) cause unneeded movement.
Therefore, create new items hidden, and have the creator call
animateIn or set the scale/opacity properties manually.
adjustIconSize must be changed to always set the icon size temporarily,
otherwise the first time it computes the icon size with 0 scale.
https://bugzilla.gnome.org/show_bug.cgi?id=690643
The DashActor will known to allocate the show apps button only if the
icon size is (temporarily) too big for the containing box, therefore
it should request just that as the minimum size.
This solves a glitch that happened when removing a favorite and at the
same time causing the dash to expand.
https://bugzilla.gnome.org/show_bug.cgi?id=690643
Using the scroll wheel in the window picker should switch workspaces.
As the picker doesn't have a visible boundary though, it makes sense
to accept scroll-event for the entire overview area. Rather than
making the overview's main actor public, expose scroll-events via
a signal.
https://bugzilla.gnome.org/show_bug.cgi?id=686639
Using the scroll wheel to switch workspaces makes sense, but it is
currently only supported on the workspace switcher, as it conflicts
with window zooming in the picker.
As changing workspaces is far more useful than the zoom feature,
remove the latter in order to "free" the scroll wheel for workspace
switching.
https://bugzilla.gnome.org/show_bug.cgi?id=686639
Right now if a user logs out, they are thrown to the login screen
imediately, without even seeing the dialog close.
This commit fades the dialog out before processing the logout.
https://bugzilla.gnome.org/show_bug.cgi?id=694296
Opening and closing a modal dialog is not instant. There
is a transition animation involved. When the dialog is
finished opening, it currently emits the "opened" signal.
When it's finished closing, however, it doesn't emit a
"closed" signal. This means, there's not a good way to
know when the dialog finishes closing.
This commit adds the "closed" signal.
https://bugzilla.gnome.org/show_bug.cgi?id=694296
Modal dialogs take an optional timestamp in their close method.
If the timestamp is not passed in, then global.get_current_time()
is used.
Some callers of the close method pass in global.get_current_time()
unnecessarly (since it's the default).
This commit drops the argument for those cases.
https://bugzilla.gnome.org/show_bug.cgi?id=694296
And reverse menu order in order to avoid it reading "Change Background… Settings"
"System Settings" shouldn't be used because it's not used to mean the control-center in any other place.
https://bugzilla.gnome.org/show_bug.cgi?id=694290
We disable the entry after a user answers a question while we
process it, but we don't reactivate it later if asked another
one.
This commit makes sure the entry is always reactive when we
are waiting for an answer from the user.
https://bugzilla.gnome.org/show_bug.cgi?id=691806
When a swipe scroll completes, we scroll the active workspace into
view. This works fine if the gesture is completed properly by
emitting 'gesture-end', but not when Clutter considers the gesture
cancelled - in that case, the view remains stuck in an intermediate
position. To fix, treat 'gesture-cancel' the same as 'gesture-end'.
https://bugzilla.gnome.org/show_bug.cgi?id=689394
This means that windows will be positioned correctly with respect
to the panel when the shell starts up, and there won't be adjusting
after the session animation zooms in entirely.
https://bugzilla.gnome.org/show_bug.cgi?id=694227
Due to a bad rebase causing freezeUpdateCount to never get initialized,
these functions effectively did nothing. Since we're going to go to a
different mechanism for freezing region updates, let's just tear these
out now instead of fixing them before tearing them out.
https://bugzilla.gnome.org/show_bug.cgi?id=694227
PopupMenuManager eats the next release event, which would otherwise close
the menu, so we need to tell dnd that we're handling it, and no drag should
be started, so it can ungrab the pointer and restore state.
https://bugzilla.gnome.org/show_bug.cgi?id=694276
The arrow origin must be in the coordinate space of the box pointer,
but it cannot be calculated when the box pointer is shown, because it's
not allocated yet, so pass the source to the boxpointer and adjust
the arrow at the next paint.
https://bugzilla.gnome.org/show_bug.cgi?id=694264
St.Bin() really only expects one child at a time, and the
BackgroundManager will add two. This can cause assertion
failures when destroying one of the background actors.
https://bugzilla.gnome.org/show_bug.cgi?id=694227
BackgroundMenu must ensure the actor it attaches to is reactive, and
the layout manager must create a background menu for the first background
too.
https://bugzilla.gnome.org/show_bug.cgi?id=694227
Right now we arrange the window thumbnails in a grid if there are more than two
rows of them. This was originally intended to reduce the amount of noise in the
thumbnail arrangement. It also made sense when the thumbnails were of a similar
size.
Nowadays we reflect the size of windows in the size of the thumbnail. This
leads to huge amounts of space between some windows when they are grid aligned.
Removing the grid alignment would also lead to larger thumbnails.
https://bugzilla.gnome.org/show_bug.cgi?id=694210
Right now we hide the window group if we're a greeter.
We did this to avoid showing the background. These days
we don't add the background in the first place, so we
can drop this code.
https://bugzilla.gnome.org/show_bug.cgi?id=682429
gnome-desktop's background drawing code supports an
XML format for presenting backgrounds based on time of day,
monitor geometry, etc. Now that we don't use gnome-desktop for drawing the
background, we need to implement that support ourselves to maintain
feature parity.
This commit implements that.
https://bugzilla.gnome.org/show_bug.cgi?id=682429
This commit updates the code to use mutter's new background
api, and changes the shell's startup animation to be closer
to the mockups.
Based on initial work by Giovanni Campagna
https://bugzilla.gnome.org/show_bug.cgi?id=682429
The user list and the "Not Listed?" button get shown and hidden at the
same time, so we can simplify the code by putting them in a new
subcontainer.
This commit creates a userSelectionBox container that both actors get
put in, and changes all the code that shows and hides these actors to
show and hide userSelectionBox instead.
https://bugzilla.gnome.org/show_bug.cgi?id=694062
The sessionList and the prompt hint are all really
part of the prompt, so we should have the code that
hides those things in hidePrompt instead of in
showUserList.
This commit does that.
https://bugzilla.gnome.org/show_bug.cgi?id=694062
Right now, when a user item is clicked we remove all other users from
the list and position the item in the appropriate place on screen.
Ultimately, we're going to want to crossfade from the fully populated
list to the user prompt. Since we're going to need to show the user
avatar in two different positions we can't simply move it.
This commit leaves the user item for the user list, and instead shows
a UserWidget actor during user verification, in the same way the
unlock dialog shows a UserWidget actor during reauthentication.
https://bugzilla.gnome.org/show_bug.cgi?id=694062
The user widget is the username and avatar shown on
the unlock dialog.
The login dialog has something very similar.
This commit separates the user widget out to its own
file, so we can use it from the login dialog in a
later commit.
https://bugzilla.gnome.org/show_bug.cgi?id=694062
Right now we very abruptly kill the login screen
and start the users session without any transition
out.
This commit introduces a fade out of the dialog and
panels.
https://bugzilla.gnome.org/show_bug.cgi?id=694062
The latest mockups don't animate between states by
resizing actors. Instead, crossfades are employed.
This commit strips out many of the existing animations
as a first step toward implementing the new ones.
https://bugzilla.gnome.org/show_bug.cgi?id=694062
According to the design mockups, the app picker should follow the
recent view pattern as used by applications, where the user is first
offered a subset of applications he/she is likely to start, and only
then allow switching to the full set of installed applications.
So implement the ability to manage several views in AppDisplay and add
FrequentView as additional view, which uses the existing ShellAppUsage
to display a list of frequently used applications.
https://bugzilla.gnome.org/show_bug.cgi?id=694192
While a group is open, we want to block events on items underneath,
but still allow interaction with components outside the app display
as well as scrolling of the view as a whole.
https://bugzilla.gnome.org/show_bug.cgi?id=694192
App folders are intended for grouping some applications, not to
assign a category to every single application, so we will only
create folders for a selected subset of the existing categories.
Software/Alacarte will eventually allow to create/modify those
folders, so store the setting in GSettings so that it can be
shared with those applications.
https://bugzilla.gnome.org/show_bug.cgi?id=694192
At the moment when loading the applications, each app is inserted
at its correct (alphabetical) position. Avoid this overhead by
loading all apps first, then sort them once and fill the grid with
the sorted actors.
https://bugzilla.gnome.org/show_bug.cgi?id=694192
AllView will be updated to allow app folders in addition to app
launchers. Folders will pop up in the existing view, so add an
additional stack widget into the hierarchy which popups will use
as parent.
https://bugzilla.gnome.org/show_bug.cgi?id=694192
We are going to introduce app folders, which will also present an
alphabetical list of applications, but use a different UI. In
preparation for that, split out the item logic into an abstract
base class.
https://bugzilla.gnome.org/show_bug.cgi?id=694192
With categories removed, the separation between AllAppDisplay and
ViewByCategories no longer makes sense. Also use this opportunity
to rename the outdated AllAppDisplay to AppDisplay; it will
eventually be used to manage different views.
https://bugzilla.gnome.org/show_bug.cgi?id=694192
AppIcon is just a tiny wrapper around BaseIcon, which does not add
anything over using BaseIcon directly, so merge its code with
AppWellIcon. As the concept of the "app well" has not been used
since well before 3.0, use AppIcon as name for the merged class.
https://bugzilla.gnome.org/show_bug.cgi?id=694192
Since commit 1ae0fadbf4 we
no longer start a PAM conversation as soon as the user clicks
the "Not Listed?" button. Instead we defer starting the PAM
conversation until the user types their username.
Unfortunately, the cancel button resets the dialog back to the
user list indrectly by cancelling the current PAM conversation.
This means if the user hasn't yet entered thier username then
the cancel button doesn't work.
This commit performs a direct dialog reset in the case the
PAM conversation hasn't been started yet.
https://bugzilla.gnome.org/show_bug.cgi?id=693756
Doing it before the grab messes up the grab helper's tracking
of where the focus was before, leading to it not tracking the
saved focus correctly, meaning that when we pop the menu back
down with escape, it doesn't restore focus correctly.
https://bugzilla.gnome.org/show_bug.cgi?id=694123
Starting the startup animation when we don't have that much IO
makes it a lot more visible.
Based on a patch by Giovanni Campagna <gcampagna@src.gnome.org>
https://bugzilla.gnome.org/show_bug.cgi?id=682429
commit 92083eaf76 made
session mode loading an asynchronous operation.
Aspects of the session mode aren't known immediately at
start up. For instance, sessionMode.isGreeter returns
false for greeter sessions until the asynchronous
operation completes.
This commit defers start up processing until the session
mode is fully known.
https://bugzilla.gnome.org/show_bug.cgi?id=682429
The complex overview transition code means that there's no easy
way to handle with this right now. Blocking the message tray
while the overview is animating seems like the correct thing to do.
https://bugzilla.gnome.org/show_bug.cgi?id=694038
The activities button may come and go at any moment now that we
have a dynamic panel. We need to re-check the activities button
whenever the panel is updated.
https://bugzilla.gnome.org/show_bug.cgi?id=694038
Making them not fully opaque just makes them harder to see and there is no reason why the user should care whether the window is minimized or not when
switching to a window display them like any other windows.
https://bugzilla.gnome.org/show_bug.cgi?id=693991
Alt-Tab away from a monitor sized on the primary monitor results into the top
panel being displayed on top of the window which looks very bad.
So just hide those windows by minimizing them.
The icon geometry animation does not really make sense for fullscreen windows
so just fade them out.
https://bugzilla.gnome.org/show_bug.cgi?id=693991
All the complexity with a custom actor and a generic container was
just to add some padding below the overview controls. Remove that,
and use CSS instead.
https://bugzilla.gnome.org/show_bug.cgi?id=694100
Since GNOME 3.6, switching XKB layouts changes the group
configuration. This patch tries to track group configuration changes
and reconstruct UI as needed. See also caribou bug#694011.
https://bugzilla.gnome.org/show_bug.cgi?id=681735
