Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!utgpu!water!watnot!watmath!clyde!rutgers!ames!ptsfa!ihnp4!inuxc!pur-ee!uiucdcs!convex!convext!danny
From: danny@convext.UUCP
Newsgroups: comp.sys.amiga
Subject: Manx 3.40a Bug Report w/ Code
Message-ID: <7600031@convext>
Date: Sat, 28-Mar-87 20:28:00 EST
Article-I.D.: convext.7600031
Posted: Sat Mar 28 20:28:00 1987
Date-Received: Fri, 3-Apr-87 06:31:36 EST
Lines: 32
Nf-ID: #N:convext:7600031:000:1097
Nf-From: convext.UUCP!danny    Mar 28 19:28:00 1987


The following code will crash Manx 3.40a giving compiler errors 202 and 203.
This code compiles without error on Manx 3.20a.
I found a patch on a local bulliten board (AmigaSCOPE) and ran it on my compiler
but the bug still exists.  The patch was from Manx and said something about bugs
with VT100 2.6

I would appreciate if somebody would cross-post this to BIX or someplace where
Jim Goodnow (sp?) will see it.

NOTE: the variables are long's.  The program works when the variables are int's.
      Also note, if x[i][j] = x[i][(j + 4) % 5] + dx[i];  is replaced with
      { t=x[i][(j + 4)]; x[i][j]=t; } where t is a long then the code will not
      generate compiler errors.

Dan Wallach
...!ihnp4!convex!danny

-_- _-_ cut -_- _-_ cut -_- _-_ cut -_- _-_ cut -_- _-_ cut -_- _-_ cut -_- _-_

/* This will generate Manx Compiler Errors #202 and #203 */

main ()
{
    long    x[5][5],   /*  <==== notice the long - this code works with int's */
            dx[5],
            i,
            j;
    for (i = 0; i < 5; i++)
	for (j = 0; j < 5; j++)
	    x[i][j] = x[i][(j + 4) % 5] + dx[i];
}