The callback would schedule a clutter_actor_destroy() on the first
argument. Unless told otherwise, this is the same actor that is
already being destroyed, so this is a no-op.
Instead, the intent is to make the clone follow up destruction of
the window actor so that it results in correct removal of this
window from the ShellWindowPreviewLayout. Use a swapped connection
to pass the clone actor as the first argument of the signal callback.
Fixes: 04c781674c ("Move WindowPreviewLayout from JS to C")
Closes: https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/6570
Part-of: <https://gitlab.gnome.org/GNOME/gnome-shell/-/merge_requests/2756>
(cherry picked from commit 356bab1121)
Both the subdir and prepare arguments are very specific to
building the extensions-tool subproject stand-alone. In order
to make the script more generic, turn those required arguments
into optional --subdir and --prepare ones.
Part-of: <https://gitlab.gnome.org/GNOME/gnome-shell/-/merge_requests/2712>
(cherry picked from commit ea629cabbf)
We currently remove the directory at the end of the script, but
that code is only reached when all previous operations were
successful.
Address this by first using an absolute directory path in /tmp
instead of a "random" location based on the CWD, then set a trap
to remove it on exit.
Part-of: <https://gitlab.gnome.org/GNOME/gnome-shell/-/merge_requests/2712>
(cherry picked from commit fac05b182c)
The script has four mandatory arguments, and also accepts optional
build options that are passed on to meson. Checking for the number
of arguments *before* filtering out the optional ones means that
`./install-meson-project.sh -Done=1 -Dtwo=2 -Dthree=3 -Dfour=4`
is considered valid, even though not a single required argument
is passed.
Fix this by filtering out the arguments before doing the usage
check. As it is a nice touch to have usage information at the
top of the script, move the message into a usage() function at
the top.
Part-of: <https://gitlab.gnome.org/GNOME/gnome-shell/-/merge_requests/2712>
(cherry picked from commit 919812a851)
If we have multiple desktop ID's that share the same startup-wm class
and none of them is actually matching the desktop-id, then we should
exclude the ones that should not be shown in the current desktop.
This will help preventing cases such as the previous one in case no
desktop file ID would match the startup-wm-class exactly.
Part-of: <https://gitlab.gnome.org/GNOME/gnome-shell/-/merge_requests/2721>
(cherry picked from commit 12bb3a601d)
When fetching the desktop ids into a map of startup-wm classes we meant
to give the ones that match the desktop ID more priority, however
this did not happen because we were always comparing a desktop
file id, including the `.desktop` suffix, with a wm-class that generally
does not include that.
This is the case of gnome-system-monitor, that provides two desktop
files, one of which is OnlyShowIn=KDE but both have the same
StartupWMClass and thus the first parsed is preferred, even though its
desktop-id is gnome-system-monitor-kde.
As per this, remove the .desktop suffix when comparing it with the
startup-wm-class, keeping the old check just in case.
Part-of: <https://gitlab.gnome.org/GNOME/gnome-shell/-/merge_requests/2721>
(cherry picked from commit e7a09946ca)
When a WindowPreview is being destroyed, the class default handler for
the `destroy` signal is responsible for destroying its child actors.
This happens after the emission of the `destroy` signal, i.e. after
`WindowPreview::_onDestroy()` has been run.
The destruction of the WindowPreview's child actors now triggers a
re-pick, but due to WindowPreview having already being marked as
`CLUTTER_IN_DESTRUCTION`, it will not be picked, resulting in a `leave`
event if the cursor was on top of the WindowPreview at the time
`destroy()` was called on it.
So this leads to `WindowPreview::vfunc_leave_event()` being run after
`WindowPreview::_onDestroy()`, which means the idle started by the leave
event handler will not be removed and ends up accessing actors after
they have already been destroyed.
Closes: https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/5512
Closes: https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/6065
Part-of: <https://gitlab.gnome.org/GNOME/gnome-shell/-/merge_requests/2738>
(cherry picked from commit 4bbf6d497d)
Dash labels are children of the main uiGroup and so could be destroyed
before their logical-owner DashItemContainer during shutdown.
This implies that we could destroy them twice. To avoid this, unset them
when destroyed.
This is mostly visible when using dash-to-dock, but it's still
technically possible with upstream code:
** Message: 19:57:49.847: Shutting down GNOME Shell
(gnome-shell:2788214): Gjs-CRITICAL **: 19:57:49.933: Object St.Label
(0x55b33668eab0), has been already disposed — impossible to access it.
This might be caused by the object having been destroyed from C code using
something such as destroy(), dispose(), or remove() vfuncs.
== Stack trace for context 0x55b3345fd3d0 ==
#0 7ffeabd810d0 b /data/GNOME/gnome-shell/js/ui/dash.js:86
#1 55b335b62f88 i /data/GNOME/JHBUILD_HOME/.local/share/gnome-shell/extensions/dash-to-dock@micxgx.gmail.com/docking.js:487
#2 7ffeabd838f0 b self-hosted:1121
#3 55b335b62ec8 i /data/GNOME/gnome-shell/js/ui/layout.js:240
Part-of: <https://gitlab.gnome.org/GNOME/gnome-shell/-/merge_requests/2739>
(cherry picked from commit 8bf06bfc9f)
The left/right navigation between top bar buttons is usually
handled by a key-press handler on the button's menu.
However when a DummyMenu is used, the button itself serves as
fake menu actor and will get grabbed when "opening" the menu.
Due to that grab, the event is not propagated to the stage,
and regular keynav does not work.
To avoid the focus getting stuck in that case, add an explicit
key-press handler that bypasses the grab.
Part-of: <https://gitlab.gnome.org/GNOME/gnome-shell/-/merge_requests/2734>
(cherry picked from commit ecb274cee0)
When the user clears all notifications from the notification menu UI, it's
possible for a queued notification to be destroyed after the currently displayed
notification. The removal of the currently displayed notification is not
processed until the notification menu is closed (due to `this._bannerBlocked`).
By then, it's possible that `_notificationRemoved` has already been overwritten
when `_onNotificationDestroy` is invoked with another (queued) notification.
This eventually results in a notification banner that cannot be removed by the
user as the notification object needed to do so has already been destroyed.
Fix this by only assigning to `_notificationRemoved` if `this._notification ==
notification`.
Part-of: <https://gitlab.gnome.org/GNOME/gnome-shell/-/merge_requests/2736>
(cherry picked from commit cde7d44a28)
Emitting this signal is broken right now: We check for a length of 0 on
this._objects[objectPath], but the
`this._objects[objectPath][interfaceName] = null` we do before the
check doesn't actually remove the key, it only sets the value to null,
leaving the key around and thus the amount of entries in the object doesn't
change.
Fix that by using the delete statement instead, "delete" properly removes
the key and thus affects the amount of entries in the object, making our
length === 0 check effective.
Part-of: <https://gitlab.gnome.org/GNOME/gnome-shell/-/merge_requests/2730>
(cherry picked from commit 8431e7ae51)
When a window is in the background and should not have the cursor on top
of it, its _cursor will be null. By getting the texture through it, we
add this extra check, which was missing before, leading to a cursor
drawn at 0, 0 on windows where it should not have been drawn.
Part-of: <https://gitlab.gnome.org/GNOME/gnome-shell/-/merge_requests/2702>
(cherry picked from commit 237c3e958d)
In GNOME OS, due to a misconfiguration, geoclue was configured without
systemd support. In that configuration, geoclue does not install its
systemd .service file (geoclue.service) but it (incorrectly) includes
the following line in its D-Bus service file:
SystemdService=geoclue.service
As a result, when dbus-daemon tried to activate it at gnome-shell's
request, it would fail with:
Unit geoclue.service not found
Then, GeoclueAgent._onGeoclueVanished() would be called, as the
name_vanished_handler passed to Gio.bus_watch_name(). This is consistent
with Gio.bus_watch_name()'s documentation:
> You are guaranteed that one of the handlers will be invoked after
> calling this function.
But that function assumed that this._managerProxy is defined, leading
to:
JS ERROR: TypeError: this._managerProxy is undefined
_onGeoclueVanished@resource:///org/gnome/shell/ui/status/location.js:163:9
Fix this by checking for nullishness of this._managerProxy. (Strictly
speaking, it's undefined rather than being null, but other code in this
file already uses the vaguer '!= null' test, which considers undefined
to be null.)
Part-of: <https://gitlab.gnome.org/GNOME/gnome-shell/-/merge_requests/2689>
(cherry picked from commit ab0c795e81)
909616b208 seems to have wrapped window actors in a container, so the actor.has_pointer check started failing. Instead, switch to meta_window_has_pointer () which doesn't rely on window actor implementation details.
We check for existence of has_pointer first just in case someone attempts to run gnome-shell 44.1 with mutter 44.0 which does not have the function exported publicly yet.
Part-of: <https://gitlab.gnome.org/GNOME/gnome-shell/-/merge_requests/2710>
(cherry picked from commit 4daea0ccae)
NetworkManager frequently refreshes the list of available access points.
For some reason this often ends up removing some or all access points
only to add them back in a later refresh later. With the exception of
the currently connected access point, which is never removed.
When all access points of a WirelessNetwork have been removed, it gets
destroyed by NMWirelessDeviceItem::_removeAccessPoint(). This however
does not happen for the currently connected network due to the always
present access point. If this network now happens to consist of multiple
access points, the "unused" NMAccessPoints will get removed and added
in these refreshes, without the WirelessNetwork getting destroyed.
Whenever such an unused access point is added, due to the use of signal
tracking this leaks the NMAccessPoint and SignalTracker until the
WirelessNetwork is destroyed.
However when the NMWirelessDeviceItem is destroyed, for example due to
suspending, it stops tracking access point changes, ensuring that the
condition for the WirelessNetwork being destroyed can not occur anymore.
Even with just two access points, such as can be found in 2.4GHz+5GHz
home routers this issue leaks hundreds of NMAccessPoints and
SignalTrackers per day. As well as a small number of WirelessNetworks
which are also kept alive by the SignalTrackers.
To fix this disconnect from the access point when it gets removed and
destroy all remaining networks when the NMWirelessDeviceItem is
destroyed.
(cherry picked from commit feb1c57dde)
Part-of: <https://gitlab.gnome.org/GNOME/gnome-shell/-/merge_requests/2704>
Switching between the app grid and the window picker in the overview via
gestures results in _gestureEnd() getting called with endProgress !== 0
in both cases, which leads to it calling _showDone(). This then
unconditionally changes the state to SHOWN, which in this situation is
already the current state. Since the previous commit this results in a
warning, so check if we are already in the SHOWN state.
Part-of: <https://gitlab.gnome.org/GNOME/gnome-shell/-/merge_requests/2514>
(cherry picked from commit f5d793647b)
