1. Enhancement to debugging facilities?
I was working on a fairly complex program over the past couple
of days (bad idea when you're down with the flu, but I needed
something to do) and kept on hitting an error that was hell to
track down. It turned out to be fairly trivial, and I easily
fixed it - but it was in a procedure that was called from
several locations, and it was a logic error that caused flow
to be directed into what amounts to my "errorexit" procedure.
When I tried deskchecking the code, I couldn't reproduce the
error. When I finally solved it, I realized that an ex.err
dump would have been helpful.
Would it be possible or reasonable to have a procedure
equivalent to abort() that would, if provided with a non-zero
parameter, create an ex.err file? Or, alternatively, modify
abort() to do so, possibly via a second parameter or setting a
pragma?
=========================================================================
Jeff Zeitlin jeff.zeitlin at execnet.com
---
~ OLXWin 1.00b ~ The "chain" of command is often a noose.
2. Enhancement to debugging facilities?
- Posted by Robert Craig <robert_craig at COMPUSERVE.COM>
Mar 13, 1997
-
Last edited Mar 14, 1997
Jeff Zeitlin writes:
> Would it be possible or reasonable to have a procedure
> equivalent to abort() that would, if provided with a non-zero
> parameter, create an ex.err file? Or, alternatively, modify
> abort() to do so, possibly via a second parameter or setting a
> pragma?
What I usually do is something like:
if something_is_wrong then
? 1/0
end if
This will generate a divide by zero error and
you'll get a full ex.err dump with traceback.
You could put this into your own routine like
the modified "abort" routine in demo\sanity.ex
that overrides the usual abort().
Regards,
Rob Craig
Rapid Deployment Software
