[Haskell-cafe] type families and type signatures

[Manuel M T Chakravarty]

Lennart Augustsson:
> On Thu, Apr 10, 2008 at 4:20 AM, Manuel M T Chakravarty <chak at cse.unsw.edu.au 
> > wrote:
> Lennart Augustsson:
>
> On Wed, Apr 9, 2008 at 8:53 AM, Martin Sulzmann <martin.sulzmann at gmail.com 
> > wrote:
>
> Lennart, you said
>
> (It's also pretty easy to fix the problem.)
>
> What do you mean? Easy to fix the type checker, or easy to fix the  
> program by inserting annotations
> to guide the type checker?
>
> Martin
>
> I'm referring to the situation where the type inferred by the type  
> checker is illegal for me to put into the program.
> In our example we can fix this in two ways, by making foo' illegal  
> even when it has no signature, or making foo' legal even when it has  
> a signature.
>
> To make it illegal:  If foo' has no type signature, infer a type for  
> foo', insert this type as a signature and type check again.  If this  
> fails, foo' is illegal.
>
> That would be possible, but it means we have to do this for all  
> bindings in a program (also all lets bindings etc).
>
> Of course, but I'd rather the compiler did it than I.  It's not that  
> hard, btw.  If the whole module type checks, insert all signatures  
> and type check again.

Sure.  I did not argue against this.  I think it would be a reasonable  
way forward.

Manuel