[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Mac68k keymaps (was: bf-3.0.7 report: failure)



On Wed, 11 Jul 2001, Christian T. Steigies wrote:

> On Wed, Jul 11, 2001 at 08:09:59PM +0200, Bjoern Brill wrote:
> > 
> > The boot-floppie's "configure the keyboard" dialog offers 6 mac keymaps (2
> > US, 2 german, 2 french), but /etc/keymaps.gz contains only three of
> > these 6; here's a tar -tzf:
> > 
> > mac/mac-us-std.bmap
> > mac/mac-us-ext.bmap
> > mac/mac-fr2-ext.bmap
> This comes up when I look for mac keymaps.sh in the boot-floppies source:
> case "$arch" in
>     m68k)
>         export keymaps="atari/atari-de.kmap.gz
> ...
>         mac/mac-us-std.kmap.gz
>         mac/mac-us-ext.kmap.gz
>         mac/mac-fr2-ext.kmap.gz"
>         ;;
>     powerpc)
>         export keymaps="amiga/amiga-de.kmap.gz
>         mac/mac-fr2-ext.kmap.gz
> ...
>         mac/mac-us-ext.kmap.gz
>         mac/mac-us-std.kmap.gz"
> # Broken in console-data see bts#50588
> #        mac/mac-de-latin1-nodeadkeys.kmap.gz
>         ;;
>  
> > The following may be interesting: loadkeys issues an error with the
> > 'broken' mac-de-latin1-nodeadkeys keymap (the others seem to work). Could
> > it be that bf's keymap -> bmap conversion uses loadkeys or something
> > similar, bombs out at the first error and the 'broken' keymap happens to
> > be the fourth that's processed? 
> Looks more like its hardcoded, but I did not see yet how/where the installer
> offers you six keymaps...
> 
Looks like that's hardcoded as well -- utilities/dbootstrap/kbdconfig.c,
near the top. Note MacPPC (a little bit lower) has #if 0'ed out the three
currently missing keymaps, but the comment there shows no clue why the two
non-broken ones (mac-de2-ext, mac-fr3) are missing.

> We can add them in. If you could fix the broken keymap (see bug), might have
I'll try, but it may take some time. And still more time until a new
console-data is uploaded and crept into testing. If the maintainer
accepts my fix.

> them all? Or should we just add some of them, the working ones?
Depends on the time scale, see above. I'd say take the working ones now.

> This should move to debian-boot...
OK


Björn
--
Bj"orn Brill <brill@fs.math.uni-frankfurt.de>
Frankfurt am Main, Germany




Reply to: