[RFC] Reporting of relative device events

[Peter Hutterer]

On Sun, May 25, 2008 at 05:55:10PM +0200, Simon Thum wrote:
> Peter Hutterer wrote:
>> Instead of the current approach, it would be better to differ between
>> DeviceMode and DeviceReportingMode, both independent of each other.
>> Comments?
>> The next question would then be - should this be implemented as a
>> DeviceControl, thus globally. Or should it be implemented on a per-client
>> basis?
> I don't know how much Xi renovation is aniticipated, but why not report 
> both in one (with validity bits of couse)? It avoids per-client processing 
> on the server, and clients can pick what they want (given it's valid).

We could, but that'd require us to switch to generic events for all input
events (and possibly drop XI 1 support).

Currently, clients expect deviceValuator events for coordinates after. We
can't reliably send twice the deviceValuator events w/o expecting XI 1 clients
to break.

Cheers,
  Peter