Hi Bernie,
You ask why Haskell infers kinds for datatypes in dependency order.
As you point out, if Haskell tried instead to infer kinds for all of
the datatypes in a program at the same time, then it would sometimes
accept programs that are currently rejected. For example:
|data C x = Foo
Hi all,
I'm in the middle of writing kind inference for haskell, as part of a
type checker/inferer. After reading the section in the haskell report about
kind inference (section 4.6), I began to wonder why kind inference must
be done in dependency order.
Some friends and I came up with the foll
On 31-Aug-2000, William Lee Irwin III <[EMAIL PROTECTED]> wrote:
> {hugs} Example> :type cmethod . fromNat $ 1
> {hugs} cmethod . fromNat $ 1 :: (C a, Q a) => a
>
> This is the expected typing, which I expect to be valid.
>
> But the inferencer chokes on an actual binding like the following:
> x
William Lee Irwin III wrote:
Example> let x = cmethod . fromNat $ 1 in 0
ERROR: Unresolved overloading
*** Type : (Q a, P a) => Integer
*** Expression : let {...} in 0
The type of x is t ( for some t in Num)
OK. What is t? It's unspecified. You don't actually need it. Too bad.
Type infe
> module Example where
I'm in a small quandary regarding some properties of instances I don't
fully understand.
What we have are three interrelated classes:
> class P t where
> pmethod :: t -> t
> class Q t where
> qmethod :: t -> t
> fromNat :: Integer -> t
> class C t where
>
