Carl Friedberg <[EMAIL PROTECTED]> writes:
> {Begin Rant}
> Sorry, but I personally think Set Message/noanything is bad practice. Just
> my 2 cents, but I've found this nailing me more than once. Don't Do That, my
> 2 cents.
> {End Rant}
Okay, $0.02 accepted!
> Now let me think about this rationally. The issue is what comes out during
> the mms test sequence... I would be happier if somehow it worked
> automagically that if you ran through the test harness, no messages, but any
> other way you ran it, you'd get the full message stuff. I think that's what
> will happen, so if that's the case, you can safely ignore my feelings (which
> you will surely do anyway :-).
I think a "set message/no...etc,etc..." in TEST.COM will do this (have to
try it though, haven't yet).
But all the same, having the ability to turn on and off the messages
with a pragma in Perl seems useful to me, whichever way one selects
the default. But then we have the issue of "what do you call the pragma"
as well as "should the pragma be on/off by default"?
So what do we do? Some options:
1. use vmsish 'messages' off by default -> no VMS messages
2. use ??? '?quietude?' off by default -> VMS messages
3. use vmsish 'messages' ON by default (violating Dan's rule)
I don't think the ??? should be "unixish". We're not playing in their
sandbox, we shouldn't use their namespace. But it beats me what pragma
name we could use that is sensible, off by default, and implies getting
the VMSish error messages. Time to hit the thesaurus.
--
Drexel University \V --Chuck Lane
----------------->--------*------------<[EMAIL PROTECTED]
(215) 895-1545 / \ Particle Physics [EMAIL PROTECTED]
FAX: (215) 895-5934 /~~~~~~~~~~~ [EMAIL PROTECTED]
