Re: [gtk-vnc-devel] Mouse pointer patch, take 2

[Rob Stoddard <rstoddard telanetix com>]

I would personally make it an option to use the real cursor, a fake 
cursor, or no cursor at all.  Best to support all options IMHO.

Rob Stoddard



Daniel P. Berrange wrote:
> On Thu, Apr 10, 2008 at 11:15:20PM -0500, Anthony Liguori wrote:
>   
> > Rob Stoddard wrote:
> >
> > In general, I'm not a big fan of this approach.  You're achieving a 
> > couple notable things.  I do think we should scale the cursor image but 
> > I think we should do it by pre-scaling an image and setting that as the 
> > cursor.  GTK will handle drawing the cursor for us if necessary.  I'm 
> > not really concerned about having hardware accelerated cursor.
> >     
>
> Yes, I agree that we should use a real cursor rather than drawing it 
> ourselves.
>
>   
> > In terms of supporting the warping encoding, I think we need to make it 
> > optional and disabled by default.  The problem is that a malicious (or 
> > broken) server can trap your cursor by constantly warping.  You can 
> > disable the richcursor encoding without disabling tight which should 
> > address your use-case FWIW.
> >     
>
> If we're going to warp, then to take account of the possiblity of running
> multiple VNC clients, we'll need to have it only apply the warp when the
> VNC widget has input focus. That should let multiple clients co-exist.
> We might also want to consider a flag to indicate whether it should always
> be allowed to warp, or whether to only allow the warp when the pointer
> grab is active. In general I agree this should be disabled by default
> too.
>
> Dan.
>