[MLton] Re: testing MLton's new front end

[Stephen Weeks]

> As to the other examples where there are two topdecs, MLton now
> rejects those as well.
> 
>    val a = ref nil;
>    val _ = a := [1];
> 
>    val a = ref nil;
>    structure X : sig val a : int list ref end =
>    struct
>        val a = a
>    end;
> 
> It doesn't do it exactly according to section G.8.  Rather, it chooses
> unique monotypes for the undetermined types that then fail to unify,
> which I guess is what Poly/ML and SML/NJ do.

Upon further reflection, I went ahead and made MLton accept these,
while issuing a warning about the undetermined types.