From: nholtz AT ccs DOT carleton DOT ca (Neal Holtz)
Subject: signal() handling
To: djgpp AT sun DOT soe DOT clarkson DOT edu
Date: Mon, 19 Sep 94 11:57:43 EDT

I've used djgpp 1.12m2 to compile RLaB, a 'matlab-like' interactive
matrix manipulation system, with graphics.  It makes a very nice package
for PC's, except for the fact the math errors (like divide-by-zero)
aren't trapped in a useful fashion.  

It really needs that to be able to return to the read in the
read-eval-print loop.

Is there any hope of being able to do something about this?
I don't care, too much, if it is done in a 'non-standard' way,
as long as the mods aren't too extensive or complex.  

In summary, how can I trap math errors such as divide-by-zero or overflow,
issue a vague error message, and return to reading user input?

-- 
Neal Holtz                     http://www.civeng.carleton.ca/~nholtz/Bio.html
Dept. of Civil and Environmental Eng., Carleton University,  Ottawa,  Canada.
Internet: nholtz AT ccs DOT carleton DOT ca   Ph: (613)788-2600x5797  Fax: (613)788-3951