https://bugs.freedesktop.org/show_bug.cgi?id=88857
--- Comment #3 from Pekka Paalanen <[email protected]> --- (In reply to Jonas Ådahl from comment #2) > Ah, right, the position should be considered part of the parents surface, ... > It could maybe be fixed by changing the protocol to no longer require an > explicit parent commit, but just an "parent content applied", I think, but > can we make such changes now? My intention was that parent content updates include the sub-surface's position change, which means that "both ... applied at 8" is the intended behaviour. Weston's implementation works that way, doesn't it? Requests s1::wl_surface.attach and s2::wl_subsurface.set_position are logically part of the same state update for s1. Your example sequence confuses that when the grouping implies that s2::wl_subsurface.set_position would be related to s2::wl_surface requests. In other words, analogous to the Wayland paradigm that a client cannot know the absolute position of its windows, the code driving updates to s2 does not know and does not control where the s2 surface appears. It may not even know it is a sub-surface. The one who positions the s2 surface is the code driving the s1 surface: the same code that drives s1::wl_surface requests drives also s2::wl_subsurface requests. As to which way to correct the things, I suppose that depends on how much would break because of the changes. Naturally I'd like it to be what I intended, because that's the case I have thought out. The other way is unfamiliar to me and would need thinking to see if it can work. -- You are receiving this mail because: You are the assignee for the bug.
_______________________________________________ wayland-bugs mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/wayland-bugs
