During session startup, we have the following chain:
LayoutManager._prepareStartupAnimation()
`- LayoutManager._startupAnimation()
`- LayoutManager._startupAnimationComplete() (a)
| or
`- LayoutManager._startupAnimationGreeter() (b)
| or
`- LayoutManager._startupAnimationSession() (c)
`- Overview.runStartupAnimation() (d)
`- OverviewActor.runStartupAnimation() (e)
`- ControlsManager.runStartupAnimation() (f)
Branch (b) calls LayoutManager._startupAnimationComplete() once the
animation is complete. Branch (c) does the same, except that
ControlsManager.runStartupAnimation() is an async function that awaits
LayoutManager.ensureAllocation().
LayoutManager._prepareStartupAnimation() is an `async` function, and its
caller catches & logs any error it raises.
Previously, ControlsManager.runStartupAnimation() – (f) in the above
diagram – was declared `async`, because it uses `await` internally, but
also accepted a callback, which was called on successful completion of
the function and the animation it triggers.
If an exception is raised during execution of the function, its callback
would never be called, and because the promise it implicitly returns is
discarded, the exception would never be logged either. And, stepping a
few levels up the call stack, the callback that
LayoutManager._startupAnimationSession() (c) provided would never be
called, and so LayoutManager._startupAnimationComplete() would never be
called, meaning the cover pane that prevents input would never be
removed and the session would be unusable.
Remove the callback parameter from ControlsManager.runStartupAnimation()
(f), and instead make it resolve in the case where the callback would
previously have been called.
Remove the callback parameter from OverviewActor.runStartupAnimation()
(e) – it is just a wrapper around ControlsManager.runStartupAnimation().
Adjust Overview.runStartupAnimation() (d) accordingly: rather than
passing a callback to OverviewActor.runStartupAnimation(), await its
return and then proceed. There is a slight behaviour change here:
previously, in the branch where this._syncGrab() is false, the callback
would be called before calling this.hide(), whereas now this.hide() is
called before the async function returns.
Adjust LayoutManager._startupAnimationSession() and its siblings to be
async, rather than each calling _startupAnimationComplete() directly.
Finally, in _prepareStartupAnimation() await _startupAnimation() either
succeeding or failing, and in either case call
_startupAnimationComplete(), removing the cover pane that prevents all
input.
Part-of: <https://gitlab.gnome.org/GNOME/gnome-shell/-/merge_requests/3428>
In commit e69da36095d5093c1c7bec7a9c96c079c0b837f9, Florian Müllner
wrote:
> We currently complete the animation using an onComplete handler,
> which only runs if the corresponding transition was stopped when
> finished.
>
> While it is unexpected that the transition is interrupted, it can
> apparently happen under some circumstances (like VMs with qlx).
> The consequences of that are pretty bad, mainly due to the cover
> pane that prevents input during the animation not getting removed.
>
> Address this by always completing the animation when the transition
> is stopped, regardless of whether it completed or not.
There are effectively four branches of the startup animation:
1. if Meta.is_restart() is true, no animation is run on startup; I
believe this is the X11-only case of restarting the shell
mid-session.
2. if Main.sessionMode.isGreeter is true, just the panel is eased onto
the screen; this is the GDM case.
3. if Main.sessionMode.hasOverview is true, then a whole sequence of
animations are run; this is the normal session case.
4. otherwise, the whole UI zooms in to full size, and from full
transparency to full opacity; this is the Initial Setup case.
The fix above handles cases 2 and 4, but not 3. This patch applies the
same fix to case 3, so that the callback is always called on session
startup even if the transition is interrupted.
Part-of: <https://gitlab.gnome.org/GNOME/gnome-shell/-/merge_requests/3422>
We set the expand-fraction property of the workspace thumbnails on changes
to their should-show property. The should-show property defaults to true, and
the expand-fraction property defaults to 1. The should-show property gets
updated in the constructor of the ThumbnailsBox though, and it might become
false (that is the case when starting up the shell) during construction.
In this case we're not yet listening to "notify::should-show" in the
ControlsManager, and therefore forget to update the expand-fraction
accordingly. Fix that and ensure the expandFraction is always set to the value
we expect after hiding or showing the thumbnails (we call _updateThumbnailsBox()
from _update(), which we call in the constructor).
This should be the proper fix for the problem that 9bf550da8 (which caused
the workspace thumbnails to never show) originally tried to fix.
Fixes: 9bf550da88 ("overviewControls: Set expandFraction for ws thumbnails to 0 when hidden")
Closes: https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/7668
Part-of: <https://gitlab.gnome.org/GNOME/gnome-shell/-/merge_requests/3323>
We multiply the spacing around the workspace thumbnails with the
expandFraction so that it grows with the animation to show/hide them. We
shouldn't factor in the extra spacing to accomodate for the thumbnails in
case they're hidden though, so make sure expandFraction is set to 0 after
the thumbnails got hidden.
This fixes the spacing when a user never uses workspaces (and the workspace
thumbnails are never shown), and also the first transition from
hidden->shown for the thumbnails, which currently starts at 100% scale (ie.
pops into existence).
Part-of: <https://gitlab.gnome.org/GNOME/gnome-shell/-/merge_requests/3278>
With a previous commit we've made the overall dynamic spacing dynamic, but
the spacing is now a bit much when it comes to the space between search and
workspaces minimap, just as workspaces minimap and large workspaces.
The minimap should somewhat "squeeze into the empty space" here, so
reduce the spacings above and below it a bit.
Part-of: <https://gitlab.gnome.org/GNOME/gnome-shell/-/merge_requests/3278>
The naming of the appGridBox is quite confusing, as it actually
describes the box of the workspaces strip, but in the appGrid state of
the overview. Change that name and call it workspacesBox instead, which
seems to make more sense here.
Part-of: <https://gitlab.gnome.org/GNOME/gnome-shell/-/merge_requests/3278>
All the ControlsManagerLayout uses from BoxLayout is the spacing
property, both size requests and allocation are completely custom.
That makes subclassing really questionable, so stop doing that
and just hook up to the `spacing` style property manually.
Part-of: <https://gitlab.gnome.org/GNOME/gnome-shell/-/merge_requests/3104>
Some extensions want to modify the value of the MAX_THUMBNAIL_SCALE
constant. That is no longer possible, as exports are always read-only
from the outside.
Make this possible again by exposing the scale as a property on the
object itself, so extensions can override it.
Part-of: <https://gitlab.gnome.org/GNOME/gnome-shell/-/merge_requests/2939>
Derive the workspaces adjustment used by OverviewControls (and
shared with WorkspacesView) from the main workspaces adjustment
using the new Main.createWorkspacesAdjustment() method.
Part-of: <https://gitlab.gnome.org/GNOME/gnome-shell/-/merge_requests/2881>
Provide a reasonably public way to register and unregister search
providers, without adding too much API.
`Main.overview.searchController` provides access that may be generally
useful, while `SearchController.addProvider()` and
`SearchController.removeProvider()` provide a simple interface for
extensions.
Part-of: <https://gitlab.gnome.org/GNOME/gnome-shell/-/merge_requests/2783>
When we unmap, the child widgets have already been destroyed, so we
shouldn't try to. To detect this, delete the references we keep to them
on destroy, and null-check the hide call.
Part-of: <https://gitlab.gnome.org/GNOME/gnome-shell/-/merge_requests/2623>
The cached work area, which is the work area of the primary monitor,
effectively depends on two properties - the (global) work area and the
primary monitor - and we are only tracking changes to one of them. Also
track monitor changes to also cover the second case.
Part-of: <https://gitlab.gnome.org/GNOME/gnome-shell/-/merge_requests/2623>
It existed to fade out/in `nautilus-desktop` for the overview, but it only
ever worked for X11 sessions (`Meta.WindowType.DESKTOP`) and
`nautilus-desktop` no longer exists anyway.
While I had suggested extending it in the past (!1395), that work was
never finished and since then the DING extension has implemented its
own visibility toggling. There seems little value in keeping the old
fade logic around in gnome-shell. Removing it actually fixes a bug with
DING (https://launchpad.net/bugs/1965072).
Part-of: <https://gitlab.gnome.org/GNOME/gnome-shell/-/merge_requests/2244>
While we don't want to position overview elements according to
the work area, it does apply in the session when the overview
is hidden. That is, we should take it into account for the HIDDEN
state, so that transitions from and to the overview are correct.
Part-of: <https://gitlab.gnome.org/GNOME/gnome-shell/-/merge_requests/2223>
The work area is really the space that is available to application
windows. Applying it to the overview is therefore questionable,
in particular given that
- X11 clients that affect struts aren't shown
- elements added by extensions have other means
to affect the overview layout
Not applying the work area to the overview also makes ignoring
actors' visibility for struts acceptable again: Assuming that
strut actors are only hidden when the monitor is in fullscreen,
freezing the work area instead of updating it for windows that
are fully obscured by a fullscreen windows makes perfect sense.
This reverts commits 81a1e294f8bd161 and 2b074882f44c32862e3.
Part-of: <https://gitlab.gnome.org/GNOME/gnome-shell/-/merge_requests/2223>
Start using the new methods to simplify signal cleanup. For now,
focus on replacing existing cleanups; in most cases this means
signals connected in the constructor and disconnected on destroy,
but also other cases with a similarly defined lifetime (say: from
show to hide).
This doesn't change signal connections that only exist for a short
time (say: once), handlers that are connected on-demand (say: the
first time a particular method is called), or connections that
aren't tracked (read: disconnected) at all.
We will eventually replace the latter with connectObject() as
well - especially from actor subclasses - but the changeset is
already big enough as-is :-)
Part-of: <https://gitlab.gnome.org/GNOME/gnome-shell/-/merge_requests/1953>
Hiding the `WorkspacesDisplay` triggers a reallocation of the
`ControlsManagerLayout` which can fail with the following error:
```
JS ERROR: TypeError: workspace is undefined
_getSpacing@resource:///org/gnome/shell/ui/workspacesView.js:229:13
vfunc_allocate@resource:///org/gnome/shell/ui/workspacesView.js:355:18
vfunc_allocate@resource:///org/gnome/shell/ui/overviewControls.js:200:33
vfunc_hide@resource:///org/gnome/shell/ui/workspacesView.js:1070:38
vfunc_unmap@resource:///org/gnome/shell/ui/overviewControls.js:672:33
hideOverview@resource:///org/gnome/shell/ui/layout.js:312:28
_hideDone@resource:///org/gnome/shell/ui/overview.js:617:32
onComplete@resource:///org/gnome/shell/ui/overview.js:390:37
_makeEaseCallback/<@resource:///org/gnome/shell/ui/environment.js:134:13
_easeActorProperty/<@resource:///org/gnome/shell/ui/environment.js:298:60
```
This can be reproduced by closing the overview with the three-finger
gesture.
Thus propagate the unmap before hiding the `WorkspacesDisplay`.
Part-of: <https://gitlab.gnome.org/GNOME/gnome-shell/-/merge_requests/2088>
We build controls layout using the whole monitor vertical space as
available, however extensions or external apps in X11 may reduce the
workarea size horizontally and the shell should always take care of it.
Given that we're already assuming that the allocation is monitor-based
and that we're adjusting it to the workarea, we can just make it more
explicit by using a workarea box that is used as the allocation area.
As per this, we also apply the same logic of applied to the vertical
dimension to the horizontal one.
Part-of: <https://gitlab.gnome.org/GNOME/gnome-shell/-/merge_requests/1892>
To compute the available height for the layout we're currently using the
panel position, while this works for the current and default setup, the
shell may be configured to use a different workarea, so we should rely on
it to compute the available space, instead of a specific widget.
So get the current monitor index for the current view and use its coordinates
instead.
Part-of: <https://gitlab.gnome.org/GNOME/gnome-shell/-/merge_requests/1892>
When starting a gesture to open the overview while a transition to hide
the overview is running, Overview._shown will be first set to false when
starting the transition and then to true by the gesture before the
`onComplete` callback is called. The `onComplete` callback in this case
is `Overview._hideDone()` which starts a transition to show the overview
again which also emits the `showing` signal. Since the gesture emits a
`showing` signal as well, this results in two consecutive `showing`
signals without a `hiding` signal in between.
This breaks the `searchControler` which adds a key press handler to
start the search on `showing` and removes it on `hiding`. So every time
this happens a new handler that will never be removed is added,
resulting in the first key press being repeated.
Fixes: https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/4004
Related: https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/3819
Part-of: <https://gitlab.gnome.org/GNOME/gnome-shell/-/merge_requests/1828>
When super is pressed again during the overview transition, we shift
up to the app grid. That means that the feature currently doesn't
work when animations are disabled (like in a VM), because there is
no transition in that case.
Address this by adding a time-based fallback in that case, i.e.
shift up when a second super-press occurs within 250ms after the
first one.
https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/4121
Part-of: <https://gitlab.gnome.org/GNOME/gnome-shell/-/merge_requests/1811>
For the primary monitor workspace thumbnail, we must keep the monitor
index in sync with what is currently the primary monitor index,
otherwise we might end up trying to move windows to non-existing
monitors.
For example, if the primary monitor index was 1 when the thumbnail box
was created, but later, the primary monitor index changed to 0, with the
other monitor being turned off, moving a window to one of the workspaces
on the workspace thumbnail, gnome-shell would attempt to move it to the
monitor with the index the primary monitor had in the past, with the
problem being that that monitor no longer exists.
Fix this by listening on the 'monitors-changed' signal on the layout
manager, and update the monitor index of the primary workspace
thumbnails box. Make sure to connect to the signal before creating the
thumbnails box, as the thumbnails box itself will listen to the signal
and recreate its actual thumbnails, and it must do this with the up to
date monitor index.
Closes: https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/4075
Part-of: <https://gitlab.gnome.org/GNOME/gnome-shell/-/merge_requests/1810>
When OverviewControls goes from HIDDEN to APP_GRID, it constantly checks
if AppDisplay needs to be visible or not by checking the current overview
state is bigger than WINDOW_PICKER. Turns out in this case this check is
problematic, because when the current state trespasses WINDOW_PICKER, the
layout manager will have already positioned AppDisplay halfway to its final
position.
Use either the final or the current state, whichever is biggest, when updating
the AppDisplay visibility. It optionally allows passing the overview state
params to _updateAppDisplayVisibility() so that we avoid a few trampolines to
recaltulate the adjustment state.
Part-of: <https://gitlab.gnome.org/GNOME/gnome-shell/-/merge_requests/1805>
Right now we use a ClutterClone ghost of the panel simply to add some
spacing to the top monitor edge.
We can remove the panelGhost by adjusting the allocations of all the
parts of the overview ourselves inside ControlsManagers vfunc_allocate,
this should also work with extensions that move the panel somewhere
else.
This makes the initial relayout of the overview significantly faster,
because we now no longer have to relayout the whole panel in the
process.
Part-of: <https://gitlab.gnome.org/GNOME/gnome-shell/-/merge_requests/1755>
The OverviewControls actor gets allocated a fixed size by its parent,
the OverviewActor, anyway, so it's pretty useless to go through the size
request machinery and add up all the sizes of items in the iconGrid,
coming up with a preferred size that's wrong anyway.
Instead simply return a min and preferred size of 0 in
get_preferred_height/width of ControlsManagerLayout.
Part-of: <https://gitlab.gnome.org/GNOME/gnome-shell/-/merge_requests/1755>
