At 07:06 05/12/1999 , Douglas Handy wrote:
>An EVAL will always trigger an error on overflow.  This is intentional
>and a good thing -- but it is not compatible with math opcodes.  For
>example, the infamous (and hideous) date conversion MULT variants
>which intentionally cause overflow do not work with EVAL.  (Actually,
>I think the compiler should watch for these and refuse to compile them
>-- even with MULT! <g>)

True. I hadn't considered compatibility with the overflow behavior of the
fixed format operations. That doesn't change my NSHO though. I maintain
that if you insist on using techniques that require that behavior, you
should be forced to use the fixed columnar format, and key your programs on
punched cards too. No sense using up valuable terminal time when you're
just coding.

Pete Hall
| This is the Midrange System Mailing List!
| To submit a new message, send your mail to
| To subscribe to this list send email to
| To unsubscribe from this list send email to
| Questions should be directed to the list owner/operator:

This thread ...


Return to Archive home page | Return to MIDRANGE.COM home page