Tom Christiansen <[EMAIL PROTECTED]> writes:
> >Doesn't that tend to lead us in the direction of pack madness where we
> >end up with yet another 'sub language' within perl. We've already got
> >pack specifiers and regexen and the 'old' prototyping stuff. I'm not
> >arguing *against* these things you understand, I'm just vaguely
> >worried.
>
> As you have observed, nestled lightly within the language of Perl
> there are many sub-languages.
>
> I don't know how to have a rich and powerful enough prototype syntax
> without well, inventing some syntax, you know? :-)
I'm not against inventing syntax. I'm just vaguely worried about
inventing over complicated syntax.
> Now, it may be that [such prototypes] are not desirable. But that's
> another matter.
And I *definitely* think that prototypes are desireable.
> I think it's valuable to be able to say "this function takes any number
> of BLAH", where BLAH might be something like "arrays that will be
> passed by reference". I don't mean any number of hashes, scalars, or
> whatnot. Given that, one must come up with something. I'm sure
> Damian will have it tastefully covered.
Well, Damian is indeed the soul of good taste. Just look at the Latin
and Klingon modules. Hmm... maybe those weren't such good examples.
Text::Autoformat is very tasty though...
--
Piers
