multitouch

[Peter Hutterer]

On Mon, Mar 01, 2010 at 05:55:50PM +0200, Daniel Stone wrote:
> Hi,
> 
> On Mon, Mar 01, 2010 at 04:36:05PM +0100, Bradley T. Hughes wrote:
> > On 03/01/2010 04:26 PM, ext Matthew Ayres wrote:
> >> To me this sounds almost to be saying that touch events should be
> >> handled  no differently than mouse events, but that doesn't seem right. A mouse is
> >> always present, it always has a position. A touch-sensitive slave/physical
> >> device may always be attached, but unless something is touching it, isn't it
> >> essentially absent?
> >
> > Yes, essentially. This could be where the sub-device idea comes in. The  
> > physical device is there, but here are no active points generating 
> > events. As it stands today, the last touch leaves the X pointer in the 
> > location of the last touch, and that can generate Enter/Leave events 
> > should the window structure under the pointer changes.
> 
> Er, Xi has (had?) ProximityIn/ProximityOut events which told you exactly
> this ...

XI2 doesn't have proximity events per se, based on the argument that if a
device supports distance it might as well set up an axis to do so - even if
that axis is just set up with a range of [0..1]. a device that supports true
distance detection can do finer-grained resolutions ("your device is 2.5 cm
from the surface").

Cheers,
  Peter