Bug in *report with -fp

[Bruce Easton]

Jay R. Ashworth wrote Tuesday, March 13, 2007 2:43 PM:
> > (Bruce Easton wrote;)
> > Why would a reason for forming an fp-executable line with an 
> unresolvable
> > processing name for use with -y, -z, etc. be any better reason than
> > forming an fp-executable line with an unresolvable PROCESSING name for
> > use with -fp?
> 
> Because there's a semantically viable reason for wanting to specify a
> name of a non-existant file for -y and -z: it permits you to force the
> use of no table without having to introduce another command line switch.
> 
> And, because there's an operationally viable reason for running an
> output format with either processing table overridden.
> 
> There is, on the other hand, no sensible reason to say "don't actually
> run a format, just run this processing table ... which, oh, doesn't
> really exist."
> 
> There seems no rational reason to permit that -- and indeed, it's
> demonstrably bad: Mark demonstrated it pretty clearly, I thought.
> 

Jay - see my last post regarding a cron script - yes, there may be 
reasons.  Like I said there, I would not want to design something 
that way but I've seen weirder things. I would say I still don't agree 
for consisteny's sake, but would be fine for a change of behavior 
if -fp was not yet in use.  

Bruce

Bruce Easton
STN, Inc.