Message-Id: <m0yblF9-000S3GC@inti.gov.ar>
Comments: Authenticated sender is <salvador AT natacha DOT inti DOT gov DOT ar>
From: "Salvador Eduardo Tropea (SET)" <salvador AT inti DOT gov DOT ar>
Organization: INTI
To: sl AT psycode DOT com (sl), djgpp AT delorie DOT com
Date: Tue, 19 May 1998 09:21:11 +0000
MIME-Version: 1.0
Content-type: Multipart/Mixed; boundary=Message-Boundary-15486
Subject: Re: RHIDE 1.4.5
CC: Robert Hoehne <robert DOT hoehne AT gmx DOT net>
Precedence: bulk


--Message-Boundary-15486
Content-type: text/plain; charset=US-ASCII
Content-transfer-encoding: 7BIT
Content-description: Mail message body

Last Week Gili posted a bug report about RHIDE 1.4.5
handling the NumLock key in a wrong way. At first I didn't
understood what he was trying to explain but then Vik
explained it clearlly.
I "fixed" it, even when I don't know if can name it a fix.
The fact is that this IS the BIOS behavior. BIOS reports
the keys in the following way:

Num Lock ON:
Gray 6 => ASCII 6
Shift + Gray 6 => ASCII 0 but the scan to figure out that
is the right key.

Num Lock OFF:
Gray 6 => ASCII 0 and SCAN just like before.
Shift + Gray 6 => ASCII 6 ?!!!!!!!

If course command.com behaves like the current RHIDE 1.4.5
so I don't know if this behavior is a bug, and is this case
the bug is in the BIOS.
Anyways, the new code traps the Shift+Key==Number and remaps
it, I think it will produce the desired effect.
The patched version will be included in the next release of
the editor.
Here is the code you can use to check it:

#include <dpmi.h>

#define REGS __dpmi_regs
#define INTR(nr,r) __dpmi_int(nr,&r)

int main(int argc, char *argv[])
{
 REGS r;

 do
   {
    r.h.ah=0x10;
    INTR(0x16,r);
    printf("ASCII: %d (%c) SCAN: %d\n",r.h.al,r.h.al,r.h.ah);
   }
 while (r.h.al!=27);
 return 0;
}

SET

P.S. The attached file is the patch for TVision code.
------------------------------------ 0 --------------------------------
Visit my home page: http://set-soft.home.ml.org/
or
http://www.geocities.com/SiliconValley/Vista/6552/
Salvador Eduardo Tropea (SET). (Electronics Engineer)
Alternative e-mail: set-soft AT usa DOT net set AT computer DOT org
ICQ: 2951574
Address: Curapaligue 2124, Caseros, 3 de Febrero
Buenos Aires, (1678), ARGENTINA
TE: +(541) 759 0013


--Message-Boundary-15486
Content-type: text/plain; charset=US-ASCII
Content-transfer-encoding: 7BIT
Content-description: Text from file 'dif'

--- rhide-1.3b/tvision/gkey.cc~	Sat Feb 28 17:26:18 1998
+++ rhide-1.3b/tvision/gkey.cc	Fri May 15 20:01:18 1998
@@ -201,6 +201,7 @@
  kbX,kbY,kbZ,kbOpenCurly,kbOr,kbCloseCurly,kbTilde,kbBackSpace
 };
 
+
 unsigned short getshiftstate(void)
 {
  if (TGKey::useBIOS)
@@ -246,6 +247,7 @@
 
  GetRaw();
  // Compose the shift flags:
+
  if (sFlags & 3)
     Abstract|=kbShiftCode;
  if (sFlags & 4)
@@ -312,7 +314,16 @@
      {
       unsigned char val=kbByASCII[rawCode.b.ascii];
       if (val<128)
-         Abstract|=val;
+        {
+         // I think the following is a bug in the BIOS: Numlock OFF then
+         // key pad 6 is right, OK, but if you press shift+6 you get the
+         // ASCII of 6!!!!
+         if (rawCode.b.ascii>='0' && rawCode.b.ascii<='9' &&
+             (Abstract & kbShiftCode))
+            Abstract|=kbWithASCIIE0[rawCode.b.scan];
+         else
+            Abstract|=val;
+        }
       else
         {
          val&=0x7F;

--Message-Boundary-15486--