Hello,

Just a word to say that I was aware of the problem with MS-DOS at the 
time I released the patch, but that I hadn't a real solution. I traced 
it back to a kind of registers corruption, but then it was too hard to 
follow.

[ Meanwhile, I was (trying to) rewriting the keyboard part of the BIOS 
in order to support AT-style scancodes, extended keys, etc... 
Unfortunately, I went in the assembly code path, and it's not always 
easy to debug... And this BIOS is not very usable right now.
I may clean up the keyboard BIOS and to release it as a patch, if 
someone is interested. Tell me ]

Unfortunately, due to a lack of time, I had to slow down my hacking in 
Plex86 ( and I'm sad with this )... I hope to have some time to finish 
this keyboard hack, and I'll be very happy to see someone (Drew ? Anyone 
else ?) maintain/merge with CVS/etc... this first attempt to have 
international keyboards working in plex86.

Today I see two paths for international keyboard handling :

- The Right One (tm) would be to determine the current (X or SDL) 
keyboard mapping, then 'unmap' the keys to get back to physical keypress 
and send them to the emulated-hardware. I think that's what the Wine 
project is doing to handle keyboards, and I never had any problems with 
Wine, so it must be working really well.
- The Hackish One (tm) is the one I used in my patch : it involves using 
a (quasi-unsupported) method to query the (X/SDL) raw keycodes, but it 
may not be portable across platforms (PC, Mac, etc...) and X servers 
(XFree is ok, but what about others ?)  (and maybe one day plex86 will 
care about being multi-platforms, if not today.)

Best regards,

Ludovic LANGE

Drew Northup wrote:

> Well the proper coding of the right shift, alt, and ctrl keys should include the
> proper release handling of the keys also.  I'll have to take a look at Ludovic's
> code to see what common ground our patches may have.  My main concern was that
> the way that the keyboard code was originally written, it only handles 8-bit
> scancodes.  Many keys are macros, and thus this approach fails.  That is why I
> went and re-did the keys that I did.  For the record, they did seem to work when
> I was done. ;-)
> 
> Philipp M�ller wrote:
> 
>> Hello Drew
>> 
>>> As for the keys that I've messed with, I did the grey keys--and I think
>>> that I
>>> started to re-map the right alt, ctrl, and shift keys as well.  I am also
>>> working on F11, F12, proper prinscreen, and a few others being supported.
>>> (This
>>> isn't as easy as it sounds, since some keys on the 101/104 key keyboad are
>>> actually macros.)
>> 
>> as long as it is only the additional keys, I think it's not necessary to do
>> this asap as I can live with the current way or more precisely with the
>> current patch of Ludovic. You can find his work at
>> http://ludovic.lange.free.fr/PLEX86/.
>> 
>> My point was the missing _release_ of the Shift, Control and Alt keys.
>> But if it's only related to MS DOS, I think I don't want to annoy anybody
>> any longer for this dead OS and use the living OS's ;-)
>> 
>> Thanks anyway
>> 
>> Philipp
>> 
>> --
>> Machen Sie Ihr Hobby zu Geld bei unserem Partner 1&1!
>> http://profiseller.de/info/index.php3?ac=OM.PS.PS003K00596T0409a
>> 
>> --
>> GMX - Die Kommunikationsplattform im Internet.
>> http://www.gmx.net
> 



Reply via email to