Re: Lots of _NET_WM_STATE_FULLSCREEN issues ...

[Billy Biggs <vektor dumbterm net>]

Havoc Pennington (hp redhat com):

> On Wed, May 28, 2003 at 10:08:40AM +0200, Iban HATCHONDO wrote: 
> > What about if we decide to say something like:
> > when receiving a wm_state_fullscreen client message
> > the wm is responsible for:
> > If the vidmod extension is present:
> >  - get the current modline (XF86VidModeGetModeLine)
> >  - get viewport position (XF86VidModeGetViewPort)
> > "the application should feet the entire screen" means 
> > it should have the sizes of the current modline and be
> > place according to the viewport position.
> 
> That's a pretty easy short-term solution; just refetch the vidmode
> when moving apps into fullscreen state.

  But what about when the mouse hits the edge of the window, the X
server will pan?

  We can't lock the viewport since that I think breaks Xinerama.  We
have an outstanding bug about this in our app:

  http://sourceforge.net/tracker/index.php?func=detail&aid=691892&group_id=64301&atid=506987

  Basically, users of my app like to fullscreen on one head while
working on the other head, and if their ideal fullscreen resolution
isn't their desktop resolution on that head, we have a problem, since as
they move their mouse over to the other head, the virtual desktop pans.


  Aside: I seriously thought that _NET_WM_STATE_FULLSCREEN meant that my
app didn't need to worry about all of our VidMode code. :(

  -Billy