[LINUX] Generic HID Remote Driver for XBMC (Drivers for XBMC on Linux and XBMC Live)

Hi UsagiYojimbo and thanks for your remarks.

In fact the loop line 210 tries to find an event with all 0 codes except the first one, as standard keyboard send this event to notify of a key up event. As you said your remote doesn't send this event but sends 43415400FF77 instead which was not detected.

I've tested your patch and in fact it also works with other remotes since the first event following the key press will be deteced as a key up which I belive is correct. I'll integrate it in the next release.

Could you tell me if the key repetitions are working with this or do you still need a special event to repeat the last key ? I don't think this patch should make key repetition work but who knows

ximon Wrote:Hello, I wonder if anyone can help me, I seem to be stuck fairly early on in the process - I can't seem to blacklist my remote (on xbmclive).

Hi ximon, could you try :

I think it should work in your case.

coldsource Wrote:Could you tell me if the key repetitions are working with this or do you still need a special event to repeat the last key ? I don't think this patch should make key repetition work but who knows

It is not working, nor should it.
As the remote uses select, press, press... events for repetition, and the keyboard uses down, down..., up events.
For this to work, I think something like this could be programmed:

• read from hid device, and identify the input event
• if it is a select event, output a key up and a key down event
• if it is a press event, count it
• if counter is more than 1, output a key down event again

I believe, this should work with other remotes, too.
By the way, I mapped the buttons on the case as well.

coldsource Wrote:Hi ximon, could you try :

*snip*

I think it should work in your case.

Thanks very much, I will try this tonight. I have already prepared and tested my mapping, just needed to be able to blacklist the remote from Xorg…

Thanks again!

coldsource Wrote:You can now affect multiple events to the same code and use wildcards in event definition.

Just tried it out and it works well.

Thanks Coldsource

Shayan Wrote:Is there a keyboard shortcut that maps directly to shutdown?

Tried to map to KEY_POWER?

Edit:

akr884 Wrote:Coldsource, everything works really well including the acceleration (using 2,4,6... for the pixel movements)
However, my remote sends the presses like this
02 00 xx yy

I believe removing that first condition would stop akr88's mouse pad from repeating until another key is pressed, as these contain non-first zero bytes, too...

UsagiYojimbo Wrote:I believe removing that first condition would stop akr88's mouse pad from repeating until another key is pressed, as these contain non-first zero bytes, too...

Keys 02 00 xx yy for akr884 are relative mouse movement events (e.g. move up by 4 pixels) so there is only one event. This bloc was already disabled for mouse movement events. Mouse and keyboard events are treated separately with different mappings (--map and --map-mouse) as their management is not the same on uinput.

Hi coldsource, thank you for last reply (few pages before :).

First of all I should mention that building hid_mapper on Debian Squeeze failed with the following message.


I've checked that my /usr/include/linux/input.h doesn't contain these definitions. Commenting out the corresponding three rows (in the integer array) in keys_definition.cpp solved the problem.


After building hid_mapper I wrote down some key codes for my remote control and made a mapping file, but the following command failed.

I've read through the thread and Diego2002 had the same problem. It seems he managed to solve it with. I've tried the same thing.


Any idea? :) Thanks!

Hi coldsource

I'm having an issue with (I think) key up codes not being recognised for my remote. Learn mode worked fine and I made a map, I can start the driver but as soon as I press a button not much else works, it's as if the remote locks up.

I see that my remote is not sending 0 for a key up code in some cases, is this the issue? If so is there anything that can be done to correct it?

The remote seems to consist of 2 HID devices if that makes any difference.

Here is my key map, the commented code is the key up code for the associated key down code:

coldsource Wrote:Hi ximon, could you try :

Code:

Section "InputClass"
Identifier "Remote blacklist"
MatchUSBID  "1241:e000"
Option "Ignore" "on"
EndSection

I think it should work in your case.

Hmm, no luck - Xorg won't load, boots to the console. Any ideas?

Thanks again for your help!

Edit: The remote is the only input device connected to the system, is it possible that Xorg requires at least one input device in order to start?