On Fri, 9 Nov 2001, Jeremy wrote:
> On Fri, 9 Nov 2001, Mark Vojkovich wrote:
> > On Fri, 9 Nov 2001, Jeremy wrote:
> > Contrary to some opinion on this list, overlays are not intended
> > to make legacy PseudoColor apps work. Overlays in the X-window
> > system are for layers of windows that don't clip other windows.
>
> I'll accept that. While I knew that was how the various overlays used by Xv
> and friends worked, I perceived the 8+24 Pseudo/True differently for whatever
> reason (I think largely because compatability with old apps was the best
> reason for its existence I could posit).
>
> Also, I suppose I was just concerned with the fact that they had different
> visuals and I didn't expect two windows on any two arbitrary visuals not to
> clip each other. Do they? Is there any way to tell if a non-default visual
> is an overlay (other than assuming anything with a different depth is)?
The SERVER_OVERLAY_VISUALS root window property tells you which
visuals are overlay visuals. It tells you the visualID, the type
of transparency (mask or pixel), the transparency value (of the
mask or pixel) and the layer it is in.
Lookup SERVER_OVERLAY_VISUALS on google for examples.
> > Calling XClearWindow() on the TrueColor window should work just fine.
> > It's not clipped by covering PseudoColor windows and I'm not aware
> > aware of any bugs in that code. Sounds like you're saying it doesn't
> > work anymore. If that's the case, I'll have to take a look at the
> > code again.
>
> To the best of my knowledge, I did precisely that. I created both windows up
> front, with the PseudoColor window a child of the TrueColor window. I then
> called
>
> XStoreColors() on the PseudoColor window
> XSetWindowBackground() on the TrueColor window
> XClearWindow() on the TrueColor window
> XInstallColormap() of the new Colormap
>
> At this point, the new colormap does take effect on the PseudoColor window,
> but the transparent PseudoColor pixels still show the previous background of
> the TrueColor window. However, dragging other toplevel windows over and then
> away again redraws with the new TrueColor background color.
>
> For fun I tried XClearWindow() on the PseudoColor window too, but that didn't
> help, and then I gave up and wrote the prior piece of mail.
>
I'll look into it. Assuming you're doing everything correctly it
could be a bug in the mioverlay code or XAA or maybe the cfb8_32 code.
I'd have thought there would be bigger problems than this if
XClearWindow didn't work though.
Mark.
_______________________________________________
Xpert mailing list
[EMAIL PROTECTED]
http://XFree86.Org/mailman/listinfo/xpert
