Path: utzoo!utgpu!news-server.csri.toronto.edu!mailrus!wuarchive!zaphod.mps.ohio-state.edu!maverick.ksu.ksu.edu!uafhp!actsn!david
From: david@actsn.fay.ar.us (David Summers)
Newsgroups: comp.unix.xenix.sco
Subject: CURSES! Foiled again! (inch routine HELP!!!)
Keywords: Tandy SCO bug cc compiling
Message-ID: <579@actsn.fay.ar.us>
Date: 9 Sep 90 03:08:25 GMT
Organization: All Church Technical Service, Fayetteville, AR
Lines: 33

I need help!  I have been using the Tandy 286 SYSV compiler for the last several
years and I just upgraded to the SCO 386 2.3.2 system with the SDS 2.3.1b + 
xnx244 SLS.  Now, when I try to compile a curses routine that uses the 'inch'
macro, I get a segmentation error and the inevitable core dump.  This routine
worked FINE NO PROBLEM on the Tandy 286 SYSV compiler and it even works fine
with the the SCO compiler in 386 mode but it does NOT work for the SCO compiler
in 286 mode!  What shall I do!???? I traced it down to what looks like an 
incorrect segment reference while computing the address in a two-dimensional
array pointed to by a pointer.  Here is the offending code:

for ( pos = 0; pos < length; pos++ )
   {
   char c;
   move( y, x + pos );
   c = inch();		/* <-----  It bombs here. */
   bufp[pos] = c;
   }

The 'inch' routine is actually a macro that references another macro that
should use the stdscr structure to point to the actual data that is on the
screen.  This is part of my input routine, and I am copying a portion of the
screen into a buffer.

This has worked for 3 years with no problem and now it bombs!  
If anyone can help then I would greatly appreciate it.

   - David Summers
   (david@actsn.fay.ar.us)

-- 
I'm sick and tired of this machine, I wish that I could sell it.
It never does just what I want, but only what I tell it!
- David Summers (david@actsn.fay.ar.us)