Re: Enabling Mouse/Touch input w/o IR open

[Jarod W. <ja...@wi...>]

On 06/17/2009 09:47 PM, Jarod Wilson wrote:
> On Wednesday 17 June 2009 21:16:16 Rene Harder wrote:
>> Hi Jarod,
>>
>>
>> please check the attached patch, It fixes our false mouse event
>> generation. It works for my device even the scroll emulation is not
>> effected.
>
> I had a similar thought for a fix in my head, looks good. Related to
> this...
>
>> Maybe we should add this for the old imon's as well but I'm
>> not quite sure if they have the same packet length, do they?
>
> ...I finally remembered the clauses just a bit further down in
> imon_incoming_packet(), where there's an explicit len == 5 check
> for the newer devices, while the older devices actually do have a len
> of 8:
>
> Recent iMON:
> if ((len == 5)&&  (buf[0] == 0x01)&&  (buf[4] == 0x00)) {
>
> Older iMON:
> } else if ((len == 8)&&  (buf[0]&  0x40)&&
>             !(buf[1]&  0x01 || buf[1]>>  2&  0x01)) {
>
> Those are for handling dpad presses, converting them into arrow key
> presses when we're in keyboard mode. So yeah, your patch looks to be
> golden. (I meant to pull that ts_input out earlier, but somehow it
> managed to survive...).
>
> I'll throw this into git and then get started on adding everything to
> lirc cvs as well... I think we've got a pretty damned solid thing going
> here now!

Okay, everything is in cvs now. Have at it, folks!

--jarod