Re: Overloaded unification

On Jan 26, 8:12=A0pm, "David B. Benson" <dben...@[EMAIL PROTECTED]
> wrote:
>
> In SML @[EMAIL PROTECTED]
 has type
>
> 'a list * 'a list -> 'a list
>
> Consider
>
> overload fun @[EMAIL PROTECTED]
(A,i) =3D Array.sub(A,i)
>
> so this @[EMAIL PROTECTED]
 has type 'a array * int -> 'a
> but with the infix notation, we can now write
>
> [A@[EMAIL PROTECTED]
>
> to provide a (slightly silly) example.
>
> Don't see how to use type cl***** for this ad-hocery.

AFAICS, that's possible with multi-parameter type cl*****. Something
along the lines of:

  class At a b c | a b -> c where
    (@[EMAIL PROTECTED]
) :: a -> b -> c

  instance At (List a) (List a) (List a) where ...
  instance At (Array a b) b a where ...

On the other hand, I don't see why anybody in his sane mind would ever
want to have something that ad-hoc -- it is utterly bad, obfuscated
style. In FPLs, you'd rather invent a new operator symbol (e.g.
Haskell uses !! for indexing).

- Andreas