On Thu, 2012-08-02 at 12:01 -0400, Kristian Høgsberg wrote: > We need to loop through the rectangles in the region and make sure at > least one of them is at least boxsize wide and high. Consider a > common configuration: > > +------------+-------------------------------+ > | | | > | LVDS1 | DP1 | > | | | > | | | > +------------+ | > | | > | | > | | > | | > +-------------------------------+ > > If you put a surface here: > > +------------+-------------------------------+ > | | | > | LVDS1 | DP1 | > | | | > | +------++ | > +-----|------+| | > | SURF || | > | || | > +------++ | > | | > +-------------------------------+ > > There will be two rectangles in the region, one really wide and low > and one relaly tall and narrow. While the extent of the region may be > wide and tall, this may only leave a 1-pixel top edge and a 1-pixel > right edge of the window on the screen. So to prevent this, we need > to make sure at least on of the rectangles in the region are at least > boxsize wide and high.
Yes, this should be a problem to be fixed. I woll go through the rectangles inside. I still think your suggestion about the pointer bounding box would be better. +-------------------------------------------+ | | | | | | | | | | | | | | | | | | _______________________________________________ wayland-devel mailing list wayland-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/wayland-devel
