On Wed, 21 Dec 2016 14:56:11 +0000 Daniel Stone <dan...@fooishbar.org> wrote:
> Hi Pekka, > > On 21 December 2016 at 14:40, Pekka Paalanen <ppaala...@gmail.com> wrote: > > @@ -2500,6 +2510,17 @@ legacy_fullscreen(struct weston_wm *wm, > > } > > > > static bool > > +weston_wm_window_is_positioned(struct weston_wm_window *window) > > +{ > > + if (window->map_request_x == INT_MIN || > > + window->map_request_y == INT_MIN) > > + weston_log("XWM warning: win %d did not see map request\n", > > + window->id); > > + > > + return window->map_request_x != 0 || window->map_request_y != 0; > > +} > > Er, doesn't this break windows explicitly positioned at the origin? > And I may be wrong, but won't OR windows avoid MapRequest to the WM, > meaning that every OR window mapped would trigger this warning? > > I'd assume this should just be 'return window->map_request_x != > INT_MIN || window->map_request_y != INT_MIN;' ... Hi, if you look at the caller: - this is (supposedly) never called for OR windows - legacy fullscreen windows (manually positioned at 0,0) get detected before this is called (either in xserver_map_shell_surface or after a later patch in MapRequest handler), so this will not be called. You are right that positioning a normal window at 0,0 is not possible. I do not know how to detect a window positioned at 0,0 from a window without a position explicitly set. Any idea? The map_request values are only unset if the MapRequest handler does not run, i.e. for OR windows. Thanks, pq
pgpKKlvBVUcRX.pgp
Description: OpenPGP digital signature
_______________________________________________ wayland-devel mailing list wayland-devel@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/wayland-devel