Re: explicit signatures and default for integer literals

Ghc can't assume "in this context" - the object file produced by compilation could be linked into code that provides other instances. Abe On 5/31/05, Dinko Tenev wrote:

On 5/31/05, robert dockins wrote:

...

Dinko Tenev wrote:

...

First we observe that, g = new . flip zip [0..], so, without the type specification, it has the general type (New [(a, b1)] b, Num b1, Enum b1) => [a] -> b, as reported by GHC.

Then we infer from

(1) g :: (New [(u, v)] w, Num v, Enum v) => [u] -> w

and

(2) instance New [(a, b)] (Map a b)

that in (New [(u, v)] w), w can only be (Map u v)

This step in the reasoning requires a functional dependency, which you mentioned earlier you were unwilling to supply. Without functional dependencies w can, in fact, be something other than (Map u v).

We need to infer New [(u, v)] w, and the only thing we know so far is New [(a, b)] (Map a b). In this context, what else could we possibly have for w besides (Map u v) ?

Cheers,

D. Tenev _______________________________________________ Glasgow-haskell-users mailing list Glasgow-haskell-users@haskell.org http://www.haskell.org/mailman/listinfo/glasgow-haskell-users