Apart from removing a lot of code, this commit also introduces (thanks
to wlr_scene) fine-grained damage tracking.
The lack of our own rendering code will only be an issue if the need for
custom effects (such as blur) arises; however, that's very unlikely and
it would probably require a lot of changes anyway.
At the moment it isn't used at all and a lot of data is duplicated. This
is a temporary state introduced for nicer diff separation into several
commits.
With the migration to wlr_scene, some objects (desktop, output etc)
will no longer have any reason to care about individual shells and
similar; handling each one on its own would introduce code that is,
from their perspective, over-complicated.
Strata (sg. stratum; this name was chosen to avoid confusion with the
over-used 'layer') represent logical layers of surfaces on the desktop,
such as every layer shell layer, one for xdg shell, another for popups.
This is a preparation for switching to the wlroots scene-graph. There is
no replacement planned for output:redraw() because it will no longer be
necessary (as far as i can tell), while scene-compatible replacements
for the view pre_render/post_render events are planned.
The rendered region size is independent on the output transform and
scale. Until now, an output with scale > 1 would render only its
(1 / scale) part; so with a scale of 2, only the top-left quarter (ie.
the left 1/2 of the top 1/2) would be visible.
Similarly, a portrait output (constructed as landscape, rotated by 90
degrees) would have an empty area at its bottom.
This bug was found while testing the output-management PR (#62).
This is needed in order for wlroots to interpret some values (mostly
coords of absolute input events) correctly. It for example fixes how the
pointer behaves with WLR_WL_OUTPUTS=2.
In order to also map the pointer/output pair when the pointer is created
before the output, a list of pointers has to be managed, which wasn't
needed until now.
The default value for the `check` argument on `run_command` will soon
change from false to true. Set it explicitly to false in order to
silence the warning, and check the exit status ourselves for a nicer
error message.
There is no technical reason not to allow resizing on opposite edges,
effectively getting single-directional interactive move (which can be
combined with another edge for resizing in the other direction).
The view jumped to the cursor position vertically.
While making this change, I also put some consistency into how position
and size are set (first the original ones are considered, and then only
updated as needed). It doesn't change the functionality, but is easier
to understand IMO.
Until now, focusing a different view didn't move pointer focus to it,
even though it was under the cursor. The pointer had to move in order
to switch its focus. Similar situations should be handled after this
commit.
