From: Aivils S. <Aiv...@un...> - 2004-01-05 10:09:15
|
>I just switched my computer running backstreet ruby on a 2.4.22 kernel to two >(nearly) identical Logitech USB keyboards. These provide access to their >extended keys through a second logical USB keyboard device. So far I did not >find a way to join two logical keyboard devices on a single VT to be able to >grab the scancodes of all keys. Did I miss something or is it still not >possible with (b)ruby? >If it's not yet implemented, how difficult would it be to add this >functionality? Would I have a chance to do it in "finite" time if I have no >previous experience with Linux's (b)ruby input layer? Unfortunately nobody done with these keys. Typicaly USB multimedia keyboards use secondary USB interface especially for multimedia keys. You can parse /proc/bus/input/devices and make sure by Yourself. One keyboard ware have two physicaly interfaces, each have /dev/input/eventXX device. /dev/input/eventXX is done. You may download CVSed evtest.c and test Your multimedia keys. I don't know any userland util , which read /dev/input/eventX and do user configurable orders. I don't know any userland util , which read /dev/vc/XX multimedia keypress and do some important. Under ruby 2nd USB keyboard interface have another VT and You do not recieve keypress from multimedia keys. Typicaly mutimedia keypresses are ignored by keyboard driver. TTY may recieve multimedia keypresses as 4 - 6 keycode series. That series already decoded by /dev/input/eventXX layer. If anybody has a developer power, then is possible merge evtest.c and http://xkeymouse.sourceforge.net/ utils and so catch multimedia keys. AFAIK eventXX layer works indenticaly with bunch of keyboards fitted with multimedia keys (Microsoft, Logitech and so on). Aivils Stoss |
From: Aivils S. <Aiv...@un...> - 2004-01-07 08:41:13
|
>> IIRC some months ago Aivils added : >> echo "+$PHYS" > "/proc/bus/console/$VT/keyboard" >> to join additional keyboard interfaces, so you'll basicly have to >> use a modified input.agent to map the xxxx..../input1 (multimedia key >> interface) to > >Ah cool, that sounds just as what I was looking for. It did not work directly >with the (somewhat older) version of bruby I'm currently using but thats now >just another reason to step from the buggy 2.4.22 kernel to 2.4.24. :-) Can You drop a note to me, how to decode <0x00> <0x01> <0x14> <0x00> sample keypress sequence ? Aivils Stoss |
From: Svetoslav S. <sv...@gm...> - 2004-01-07 10:02:29
|
> > >> IIRC some months ago Aivils added : > >> echo "+$PHYS" > "/proc/bus/console/$VT/keyboard" > >> to join additional keyboard interfaces, so you'll basicly have to > >> use a modified input.agent to map the xxxx..../input1 (multimedia key > >> interface) to > > > >Ah cool, that sounds just as what I was looking for. It did not work > directly > >with the (somewhat older) version of bruby I'm currently using but thats > now > >just another reason to step from the buggy 2.4.22 kernel to 2.4.24. :-) > > Can You drop a note to me, how to decode <0x00> <0x01> <0x14> <0x00> > sample keypress sequence ? doesn't "xev" give you the keycode ? i also found "xbindkeys" which seems definately nice, but i'm not sure how well does it play with multi-user environment http://hocwp.free.fr/xbindkeys/xbindkeys.html svetljo -- +++ GMX - die erste Adresse für Mail, Message, More +++ Neu: Preissenkung für MMS und FreeMMS! http://www.gmx.net |
From: Svetoslav S. <sv...@gm...> - 2004-01-06 09:08:33
|
> > >I just switched my computer running backstreet ruby on a 2.4.22 kernel to > two > >(nearly) identical Logitech USB keyboards. These provide access to their > >extended keys through a second logical USB keyboard device. So far I did > not > >find a way to join two logical keyboard devices on a single VT to be able > to > >grab the scancodes of all keys. Did I miss something or is it still not > >possible with (b)ruby? > >If it's not yet implemented, how difficult would it be to add this > >functionality? Would I have a chance to do it in "finite" time if I have > no > >previous experience with Linux's (b)ruby input layer? > > Unfortunately nobody done with these keys. Typicaly USB multimedia > keyboards > use secondary USB interface especially for multimedia keys. You can parse > /proc/bus/input/devices and make sure by Yourself. One keyboard ware have > two physicaly interfaces, each have /dev/input/eventXX device. IIRC some months ago Aivils added : echo "+$PHYS" > "/proc/bus/console/$VT/keyboard" to join additional keyboard interfaces, so you'll basicly have to use a modified input.agent to map the xxxx..../input1 (multimedia key interface) to the primary keyboard interfaces xxxx...../input0 and then XFree will get all key events, you only have to create your custom key map to map the additional keys to XFree events(XF86_WWW,XF86_MAIL,....) and then configure XFree/WindowManager to use them, ie start evolution on pressing XF86_MAIL, Mozila on XF86_WWW, ... you may take a look at : http://tldp.org/HOWTO/XFree-Local-multi-user-HOWTO/app_ruby_init_input_agent.html > /dev/input/eventXX is done. You may download CVSed evtest.c and test > Your multimedia keys. > > I don't know any userland util , which read /dev/input/eventX and do user > configurable orders. > I don't know any userland util , which read /dev/vc/XX multimedia keypress > and do some important. > Under ruby 2nd USB keyboard interface have another VT and You do not > recieve keypress from multimedia keys. Typicaly mutimedia keypresses are > ignored by keyboard driver. TTY may recieve multimedia keypresses as 4 - 6 > keycode series. That series already decoded by /dev/input/eventXX layer. > > If anybody has a developer power, then is possible merge evtest.c and > http://xkeymouse.sourceforge.net/ utils and so catch multimedia keys. > > AFAIK eventXX layer works indenticaly with bunch of keyboards fitted with > multimedia keys (Microsoft, Logitech and so on). svetljo -- +++ GMX - die erste Adresse für Mail, Message, More +++ Neu: Preissenkung für MMS und FreeMMS! http://www.gmx.net |
From: Gunter O. <G.O...@po...> - 2004-01-06 22:31:51
|
=2D----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Am Dienstag, 6. Januar 2004 10:08 schrieben Sie: > IIRC some months ago Aivils added : > echo "+$PHYS" > "/proc/bus/console/$VT/keyboard" > to join additional keyboard interfaces, so you'll basicly have to > use a modified input.agent to map the xxxx..../input1 (multimedia key > interface) to Ah cool, that sounds just as what I was looking for. It did not work direct= ly=20 with the (somewhat older) version of bruby I'm currently using but thats no= w=20 just another reason to step from the buggy 2.4.22 kernel to 2.4.24. :-) Greetings, Gunter Ohrner =2D --=20 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + PGP-verschl=FCsselte Mails bevorzugt! + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ It was reckoned to be very healthy there. Very few germs were able to=20 survive. (Taken from the description of an Ankh-Morporkian quarter near=20 the river Ankh...) (Terry Pratchett * Wyrd Sisters - A Discworld Novel) =2D----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (GNU/Linux) iD8DBQE/+zdF0ORHvREo8l8RAgI9AJ9Eu8IHACfbrQsBAExJLwNd7WiHzACffZK4 IVM2OwkmMfs6iGGxc4m9UJ0=3D =3Dt8Y6 =2D----END PGP SIGNATURE----- |