Nathan Wiger wrote:
> > If =~ allowed "indirect object" notation as -> does, then we could write
> >
> > s $str (pat){rep};
> > and
> > for ( grok %db /Name/$name/g ) {
>
> Yeah, but I'm not sure what those are supposed to do.
Not familiar with indirect object notation?
How about
match $str /pat/;
subst $str /pat/rep/;
Still too confusing?
> They look way too obscure for me.
Not a criterion for Perl, afaik.
> As written I don't see an advantage in the RFC. I think it confuses
> regexps and subs, which are two completely different things.
No, that's not what it confuses. it confuses (intentionally, I feel
obliged to point out) match/subst *operators*, and other funcs.
Who was it that suggested changing the m// operator to the match()
function, and the s/// operator to the subst() function?
RFC139 supports that philosophy, but implemented differently.
> I would drop
> the // overloading and just look at =~, since that's a neat operator.
I suppose I could have proposed them in separate RFC's. But the two
are already married in the regex "binding" operations...
> What if =~ had a very similar effect to what you descibe, only it passed
> the argument as the *last* one
I've seen your RFC on this since you wrote that email.
It's an interesting idea; but the problem with it, imo, is that, while
it seems to jive nicely with perl's (current) set of built-ins, user-
defined subs can (and typically do) have open-ended argument lists.
So, the "last" argument becomes problematic, particularly wrt prototyping.
--
John Porter
We're building the house of the future together.
