Re: RFC: frame size hints

[Dominik Vogt <dominik vogt gmx de>]

On Tue, Dec 09, 2003 at 06:52:19PM -0500, Thomas Fitzsimmons wrote:
> On Mon, 2003-12-08 at 14:14, Havoc Pennington wrote:
> > On Mon, 2003-12-08 at 09:14, Lubos Lunak wrote:
> > >  Oh well. I guess I can live with it after all. But I'd prefer if the wording 
> > > more stressed the fact that the result will be just an estimate (because it's 
> > > an evil hack).
> > 
> > If you can think of a non-hack solution I'm all for it. I do think this
> > solution can at least in principle be implemented correctly, which is
> > something. It's not _inherently_ only an estimate.
> > 
> 
> OK, here's a patch against the spec.  I guess it's debatable whether the
> _NET_REQUEST_FRAME_EXTENTS wording should be "estimate the frame
> extents" or "calculate the frame extents".  I've left it as "estimate"
> for now.
> 
> Can we leave this up for, say, one week, then check it in if there are
> no objections?
> 
> Thanks for all the comments,
> Tom
> 
> +If the Window Manager receives a _NET_REQUEST_FRAME_EXTENTS message
> +before the window is mapped, it SHOULD estimate the frame extents that
> +the window would be given upon mapping, then set _NET_FRAME_EXTENTS
> +accordingly.

Um, how can the window manager know these dimensions before all of
the window properties have been set?  For example, if the window
does not have a name/class/resource yet?

Ciao

Dominik ^_^  ^_^