[Wayland-bugs] [Bug 88857] subsurface protocol is inconsistent regarding immediate commit vs deferred commit

[bugzilla-daemon at freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=88857

--- Comment #1 from Pekka Paalanen <ppaalanen at gmail.com> ---
Hmm. So the model I had in mind when writing the spec was the following:

A sub-surface's state is all the state with the wl_surface that acts as a
sub-surface. However, the sub-surfaces position and z-order are with the
*parent*, not the sub-surface. The parent dictates its own state (contents),
and a part of those contents is the position and z-order of its immediate
sub-surfaces. So, when new parent state gets applied, as part of that also the
sub-surfaces' positions and z-order get applied.

This is all to keep the parent surface contents in sync with the sub-surface
position and z-order. This should happen regardless of sync or desync mode of
the sub-surface (or the parent, if the parent is itself a sub-surface).

If a change in the parent's content requires a change in the sub-surface's
*size*, then you have to do the synchronization dance with the code driving the
sub wl_surface.

The switch from synchronized to desynchronized sub-surface mode with pending
position or z-order changes is sort of a corner case. I don't think carefully
written clients should ever hit that, so what happens there is not very
interesting IMO. Or am I overlooking something here, perhaps some
sub-sub-surface scenario?

Does this help?

It is quite possible the current wording is a result from lazy thinking.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freedesktop.org/archives/wayland-bugs/attachments/20150202/8b9c9bce/attachment.html>