I intended to fix this quite some time ago but didn't get around to it.
I don't think there's any good reason why we need to un-fullscreen a
view when its output is disconnected. We can handle it the same as a
maximized view, and move it to a new output (remaining fullscreen) or,
if all outputs are disconnected, just leave it as-is.
This is helpful for a media-center use-case, where you have just one
view (e.g. Kodi) fullscreen all the time, but the TV might appear to be
disconnected if you switch it to a different source.
Tested with a couple different scenarios:
1. Single output disconnected and re-connected: view stayed fullscreen.
2. Secondary output disconnected: view stayed fullscreen but moved to
the primary output, and the layer-shell panel on that output was
hidden as expected. When the secondary output was re-connected, the
view was moved back (still fullscreen) and the panel on the primary
appeared again.
Fixes: #864
Adds following settings:
<placement>
<policy>cascade</policy>
<cascadeOffset x="40" y="30" />
</placement>
"Cascade" policy places a new window at the center of the screen like
"center" policy, but possibly shifts its position to bottom-right so the
new window doesn't cover existing windows.
The algorithm is copied from KWin's implementation:
df9f8f8346/src/placement.cpp (L589)
Also added some helper functions to manipulate `wlr_box`.
Currently, initially maximized (or fullscreen) xdg-shell views exhibit
one of two issues:
- some (e.g. GTK and Qt apps) paint an initial frame un-maximized
(before the "map" event) and only maximize in a later commit
- others (e.g. foot) maximize immediately without flicker, but never
store a valid natural size, so we end up using a fallback (640x480)
Under KWin, neither of these issues occur, so I looked into what labwc
is doing wrong. It seems that:
- wlroots internally sends an initial configure event with a size of
0x0 to all xdg-shell views. This requests the client to set its own
preferred (a.k.a. natural) size.
- For an initially maximized/fullscreen view, the initial configure
event should contain the maximized/fullscreen size rather than 0x0.
In labwc, this means we have to call wlr_xdg_toplevel_set_size()
earlier, i.e. from the new_surface event. Tracing with WAYLAND_DEBUG
shows that the initial configure event now has the correct geometry,
matching KWin behavior. With this change, GTK and Qt apps no longer
paint an incorrect un-maximized frame.
- However, this means that all xdg-shell views now suffer from the same
issue as foot, where we never receive a commit with the un-maximized
(natural) geometry. The correct way to get the natural geometry seems
to be to wait until we want to un-maximize, and send a configure
event of 0x0 at that point.
Sending a configure event of 0x0 when un-maximizing is a bit annoying as
it breaks some assumptions in labwc code. In particular:
- view->natural_geometry may now be unknown (0x0), requiring various
wlr_box_empty() checks sprinkled around. I added these in all the
obvious places, but there could be some code paths that I missed.
- Positioning the newly un-maximized view within view_maximize() no
longer works since we don't know the natural size. Instead we have to
run the positioning logic from the surface commit handler. This
results in some extra complexity, especially for interactive move.
See the new do_late_positioning() function in xdg.c.
Some TODOs/FIXMEs (non-blocking in my opinion):
- The view_wants_decorations() check is now duplicated in both the
new_surface and map event handlers. I'm not sure if this is necessary
but it seemed like the safest approach for now. More testing would be
nice, particularly with various combinations of config and client SSD
preferences.
- Aside from the interactive move case, the "late positioning" logic
always centers the view when un-maximizing, and does not invoke any
of the smart placement logic. If we want to invoke smart placement
here, I'd appreciate someone with more knowledge of that code to take
a look and figure out how to do that correctly.
Thonny (Python IDE made with Tk) may set the window geometry to 1x1 and
maximizes the window before mapping. This set `view->natural_geometry`
to 1x1, so labwc tried to restore the window geometry to it on
unmaximize, causing validation errors in `ssd_update_geometry()` as its
width and height are smaller than `LAB_MIN_VIEW_{WIDTH,HEIGHT}`.
This commit fixes it by not allowing geometries smaller than
`LAB_MIN_VIEW_{WIDTH,HEIGHT}` in `view->natural_geometry`.
Before this commit, `top` layers were hidden whenever there is a
fullscreen window in the corresponding output.
With this commit, `top` layers are hidden only when there is a fullscreen
window without other windows above it in the corresponding output.
A caveat is that `bottom` layer is still always hidden by a fullscreen
window.
Add custom field with subset of printf style formatting
to replace the original field formats.
Example:
<windowSwitcher preview="no" outlines="no" allWorkspaces="yes">
<fields>
<field content="custom" format="foobar %b %3s %-10o %-20W %-10i%t" width="100%" />
</fields>
</windowSwitcher>
Mono space font recommended. May need OSD width adjusted
Co-authored-by: @Consolatis (based on work done by them)
When growing or shrinking a view by snapping to an edge, a client may
ignore the requested size and instead keep its original size or
substitute a different (possibly constrained) size. In this case, the
view may not actually contact the snapped edge, and a subsequent snap
attempt will just keep re-trying (and failing) to contact the same ege.
To mitigate this, remember the last-snapped view, snapping direction and
offset of the snapping edge in snap.c; when re-attempting a snap for the
same view in the same direction, ignore the edge that was last "hit", to
allow snapping to progress beyond the problematic edge.
...and unify region overlay and snap-to-edge overlay into overlay.c.
Snap-to-edge overlay is delayed for 500ms to prevent flickering when
the view is dragged from an output to another (demo in discussion labwc#1613).
This also fixes a bug that region overlay is not shown when a modifier
key is re-pressed.
This is undesired when a focusable popup (e.g. applications menu) of an
unfocusable view (e.g. XWayland panel) is closed.
This reverts commit f6e3527767.
Support `wrap` in view_get_adjacent_output(). This means that when seeking
an adjacent output in a particular direction from an output that is
already furthest in that direction within the layout, rather than
returning NULL, wrap around from the leftmost to the rightmost, or topmost
to the bottommost and vice versa.
Example usage:
<action name="MoveToOutput" direction="right" wrap="yes" />
Wrap is disabled by default to keep the user interface consistent.
...to edges actually encountered by motion during interactive moves and
resizes.
In addition, ignore edge resistance and attraction for minimized views.
...in view_move_to_edge. If a view is fully within the usable area of
its original output, it should be fully within the usable area (if
possible) on its new output.
...on view destruction because focus_change_notify() in seat.c changes
server->active_view so the logic introduced by 6c6e406 (which checks if
view =! active_view) is no longer right.
The glitches described in the commit below have not come back.
6c6e406507Fixes#1393
