Re: RFC 49 (v1) Objects should have builtin string SCALA

[Damian Conway]

   > >STRINGIFY would have my vote. "It's a string!!!". A string is a very
   > >specific subtype of scalar.
   > 
   > How about TO_STRING?  Little less geeky.

AS_STRING. It doesn't convert, it translate.

Or just STRING. It's a verb to, you know ;-)


   > How would this play with overload.pm?  What if that also specifies a 
   > stringify routine?  Which one should win?

Operator overloading is scheduled to be revamped.
It might well be that, rather than specifying q{""}, q{0+}, and q{bool},
we have:

        package MyClass;

        sub STRING  {...}
        sub NUMBER  {...}
        sub BOOLEAN {...}

Hmmmm. Maybe even:

        sub op+     {...}
        sub op*     {...}
        sub op<=>   {...}

        #etc.

Shades of C++! ;-)


Damian

PS: This would tie in very nicely with my forthcoming multimethods proposal.