[Wayland-bugs] [Bug 766860] tiled (snapped, half-maximized) windows in Wayland aren't GDK_WINDOW_STATE_TILED

[gtk+ (GNOME Bugzilla)]

https://bugzilla.gnome.org/show_bug.cgi?id=766860

--- Comment #20 from Jonas Ådahl <jadahl at gmail.com> ---
(In reply to Olivier Fourdan from comment #19)
> (In reply to Jonas Ådahl from comment #18)
> > Those changes are not directly related to tiling, they are more about
> > negotiating the "state" of which a surface is drawn. In the wip we have so
> > far introduced a "no drop shadow" state. All these states are supposed to be
> > optional, and guaranteed to be respected when supported.
> 
> I fail to see "no drop shadow"  as a state to be honest, but that's another
> topic.
>

State, mode, .. the major difference from window states is that its negotiated,
not just notified.

> > I think it might make sense to have tiling state part of the "window state"
> > enum, and and probably tiling edges. For example, I suspect a "no_shadow"
> > mode would mean we still have rounded corners, but a "tiling" mode, we'd
> > have sharp corners.
> 
> I reckon possible tiling state(s) discussions should be driven by developers
> of tiling WM/CM, because as far as I am concerned, I reckon "tiled" in
> itself is sufficient, anything more precise (thus as edge tiling) might end
> up being restrictive, we could have tiling WM/CMs who tile windows in random
> places on screen, not just screen edges. But then again, we should discuss
> that on wayland-devel ML for a broader audience.

Shouldn't we just mark the left, top and bottom edges tiled, and the right edge
untiled, meaning drop shadow should only be drawn on the right edge, if we tile
on the left half of the screen, in gnome shell? I.e. by pressing
Ctrl-Super-Left (or whatever the default is if I changed it).

-- 
You are receiving this mail because:
You are on the CC list for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/wayland-bugs/attachments/20160526/f576120a/attachment.html>