Re: Lirc and long keypresses.

[Anders T. <aer...@gm...>]

On 2012-11-15 03:44, Mark DM wrote:
> I see that there seems to be some issues with getting LIRC to
> recognize long keypresses , however this seems to also baffle me as to
> why .  see http://forum.xbmc.org/showthread.php?tid=115023
>
> I have a Creative Labs Sound Blaster Audigy 2 NX . I only use the IR
> functionality.
>
> When I launch irw and hit a key I see this
> 000000000000001f 00 ok RM-1500
> with "ok" corresponding to the key pressed.
>
> Now with irw if I press and hold a key for about 3 seconds I see this
> 000000000000001f 00 ok RM-1500
> 000000000000001f 01 ok RM-1500
> 000000000000001f 02 ok RM-1500
> 000000000000001f 03 ok RM-1500
> 000000000000001f 04 ok RM-1500
> 000000000000001f 05 ok RM-1500
> 000000000000001f 06 ok RM-1500
> 000000000000001f 07 ok RM-1500
> 000000000000001f 08 ok RM-1500
> 000000000000001f 09 ok RM-1500
> 000000000000001f 0a ok RM-1500
> 000000000000001f 0b ok RM-1500
> Clearly the entry before "ok" corresponds to the time the button was pressed.
>
> My question is therefore very simple.....
>
> Since we can now use simply "ok" to identify that the "ok" button was
> pressed, why can we not identify a duration by saying instead of "ok"
> we say , "0b ok" meaning a three second button press would yield a
> specific behavior? I see no issue in making this an option ,
> especially when the button does not conflict with another use. An
> example would be when XBMC is not running and I want to long press
> "ok" or "0b ok" to start XBMC, but that does not mean I want to launch
> XBMC when the remote falls on the floor and that button gets pressed,
> or when the cat steps on it while chasing a fly. In fact it coul;d
> probably be implemented while leaving compatibility with older
> configuration files as only those needing long keypresses would
> require the additional "0b"
>
>
Each non- "00" event generates a "REP" style repeat event when fed into 
the input subsystem. I've configured mouse emulation using this to make 
the mouse accelerate when I press and hold a direction button. X's mouse 
stuff require repeats to be distinguished from a batch of single events. 
I guess you can write an input listener which does something if the 
repeats are 11 or so..

/Anders