On Sunday 21 May 2006 13:29, Matt Diephouse wrote:
> I think chromatic means that he thought :flat _was_ positional. That's
> certainly what I thought.
Yes, exactly. I thought :flat and :slurpy were symmetric.
> Look at the "Flattening argument lists"
> section of S06:
>
> sub foo($x, $y, $z) {...} # expects three scalars
> @onetothree = 1..3; # array stores three scalars
>
> foo(1,2,3); # okay: three args found
> foo(@onetothree); # error: only one arg
> foo([,[EMAIL PROTECTED]); # okay: @onetothree flattened to three
> args
>
> In this example, $x, $y, and $z are positional args. I would expect C<
> foo(1,2,3) > and C< foo([,[EMAIL PROTECTED]) > to be identical calls --
> even wrt multi dispatch. $x, $y, and $z are still positional even if
> foo is called with a flattened array, right?
Yes. The flattening should happen at the point of the call, not at the point
of argument assignment within the call.
> Autoboxing would make this a bit tougher in Parrot, but if ints box to
> Integer and strings to String, then you can write your multis for
> Integer and String (as chromatic did) and be done with it.
I'm not sure why this matters for anything other than :slurpy; use a
primitive-holding PMC array and mark it with :flat.
(For Pheme it doesn't matter at all, as it doesn't use primitive types yet, if
ever.)
-- c
