This commit moves the virtual output related functions
into their own file at `src/output-virtual.c` with its
own include file to reduce `include/labwc.h` bit by bit.
Additionally, it removes the need to keep the
`server->headless.pending_output_name` char array around
by temporarily disconnecting the handler when creating a
new virtual output. This allows to set the output name
right in the `output_virtual_add()` call rather than to
store the pending name until the new output event handler
has been called.
It also makes adding a virtual fallback output easier in
a follow-up PR.
PR #1301 did partly revert the first variant because
`lab_wlr_scene_output_commit()` (which uses the `output->pending`
state in contrast to `wlr_scene_output_commit()`) was deemd to
cause 'frozen' frames during video playback (#1273).
We are now back at using `lab_wlr_scene_output_commit()` for
unrelated reasons and the cause of the original issue #1273 was
likely that `wlr_scene_output_send_frame_done()` was only sent
on a successful commit. This was changed to always be sent.
So lets restore the flicker free nested resize as well.
Preliminary fix for #1525.
Based on the protocol we should also revert all
previously correctly committed outputs. #1528
is doing just that but may cause regressions so
we need a short term solution and then deal with
potential issues in #1528 after the release.
Prevent carrying around the wlr_output->data pointer when
destroying an output. The set_gamma handler may be called
during the destruction of a wlr_output after our own
destroy handler has already been called. This patch resets
the wlr_output->data pointer in our destroy handler and
adds a check in the set_gamma handler to verify the output
is actually valid.
Fixes#1270
Reported-by: @Flrian
Account for space taken up by XWayland panels (as indicated by the
_NET_WM_STRUT_PARTIAL property) in the usable_area calculation.
This makes it possible to use labwc in a "transitional" setup, where it
replaces the X11 window manager and compositor, but most other parts of
a existing X11 desktop environment can still be used via XWayland.
(Some remaining drawbacks of such a setup would be the lack of desktop
icons, and native Wayland clients not showing up in X11-based taskbars.)
XWayland clients use the _NET_WORKAREA root window property to determine
how much of the screen is not covered by panels/docks. The property is
used for example by Qt to determine areas of the screen that popup menus
should not overlap (see QScreen::availableVirtualGeometry).
Depends on wlroots MR:
https://gitlab.freedesktop.org/wlroots/wlroots/-/merge_requests/4406
v2: prevent calling wlr_xwayland_set_workareas() too early
v3: fix segfault at exit (server->xwayland == NULL)
Chases: 756ecf8ee9f1e75bc7b8297dc84f97c7d699174b
backend/wayland: use request_state when toplevel is resized
Chases: 3ef68a484243555b020200c6f95246d994932c3f
backend/x11: use request_state when window is resized
Ref: https://gitlab.freedesktop.org/wlroots/wlroots/-/merge_requests/2693
We now delay requested resolution changes by the backend until
the next frame event which causes us to render the new content
on the already enlarged buffer. Before this change, an empty
(black) buffer would have been shown instead before the next
frame event caused a new render of the actual contents.
Keep commiting the new state and then scheduling a frame event
would not help as due to the commit call it would still show an
empty buffer in the meantime.
Just modifying wlr_output->pending wouldn't work either because
wlr_scene_output_commit() *completely* ignores it (and it will
be removed in future wlroots commits). For this reason we move
to wlr_scene_output_build_state() directly because it allows us
to supply the current wlr_output->pending state and thus apply
any resolution change in lockstep with new rendering. Result:
No more flickering in the wayland backend and resizing is again
smooth as butter.
This prevents constant flicker while resizing
when running nested via the wayland backend.
For the X11 backend (can be tested via `WLR_BACKENDS=x11 labwc`),
it is still rather janky but at least doesn't cause endless self-
resizing anymore.
Use the same fix/workaround as in output_update_for_layout_change() to
make sure that the cursor is also visible after (re-)enabling an output
in handle_output_power_manager_set_mode().
Currently, if the output layout changes while the session is locked,
the lock surfaces may end up wrongly positioned, which looks bad and
may reveal some of the user's workspace underneath.
To prevent this, re-align the scene trees and reconfigure the lock
surfaces when the output layout changes.
The wlroots wlr_output_layout provides two different modes of operation:
- automatically
- manually
In automatic mode wlroots reacts to new / removed outputs and resolution
changes and then adjusts the output positions within the layout itself.
The manual mode disables this behavior and thus it is the whole
responsibility of the caller (e.g. us relaying whatever wdisplays /
kanshi provides) to ensure that the layout is somewhat sane. E.g. that
it doesn't have any holes between outputs and there are no overlapping
outputs.
The mode is set for each output individually. To use (and keep using)
the automatic mode, outputs have to be added via the _add_auto() variant
(which we do) and they are not allowed to be moved via _move() (which
we currently do and thus break the automatic layout).
To fix that, this patch compares the user (tool) supplied position to the
automatically calculated position and only if they differ we call _move().
This restores the original approach of naming the argument `layout_changed`
which fits much better than `enforce_view_arrange`. Especially when extending
the function to also handle region updates once merged.
Move the desktop_arrange_all_views() call outside layers_arrange() into
a new function, output_update_usable_area(). The new function currently
does exactly what layers_arrange() used to, but will be expanded in a
later commit.
Add output_update_all_usable_areas(), which is the same as calling
output_update_usable_area() for each output, but only calls
desktop_arrange_all_views() once.
Rebased and slightly modified by @Consolatis
