Re: LYNX-DEV Re: msg00160.html

[Doug Kaufman]

On Sun, 15 Feb 1998, Wayne Buttles wrote:

> On Sat, 14 Feb 1998, Doug Kaufman wrote:
> 
> [on RAWDOSKEYHACK]
> 
> > Since this code wasn't written by me, I may be misinterpreting how it
> The code inside RAWDOSKEYHACK was originally inserted by myself to allow
> for the use of the numpad elements.  I don't remember why those keys
> didn't work in the first place, I just knew there was an easy, cheezy way
> to fix it. 
> 
> Later I heard from screen reader people and non-us keyboard people who
> said that it screwed them up.  I added RAWDOSKEYHACK so people could turn
> it off.  

This may be off base, but it appears that the problem may be similar to
the PDCurses handling of colors. PDCurses assigns numerical tokens to
non ASCII keys in curses.h. Some of these keys are assigned numerical
tokens by LYKeymap.c (0x100 through 0x10E). These are different
from what is in curses.h (e.g. F1 = 0x109 in PDCurses, but 0x108 in
LYKeymap.c. Would it make sense to have an alternate array in LYKeymap.c
for the PDCurses users (or alternately, some type of mapping as was
done for colors), so that a remapping in LYStrings.c would no longer be
necessary? I haven't tried any of this, so I can't be sure if this is
what is causing the problem at all.

PDCurses picks up the key scanned by looking at scan make codes
generated by the keypress in the kptab array in PDCKBD.c.  If these are
different in different keyboards (and code has a warning about this),
the fix for different international keyboards may require either a patch
to the array of make codes or a remapping of the keys in curses.h or
LYKeymap.c to reflect which keys PDCurses  thought were being pressed.
                                 Doug

__
Doug Kaufman
Internet: address@hidden (preferred)
          address@hidden