[Haskell-cafe] type variable question

[Nicolas Frisby]

I'm developing a typesafe record library (akin to HList but starting
with key/val pairs and enforcing uniqueness and sorted order of keys).
I'm having a GHC problem I've had with other projects and seen
comments regarding it in other people's code (HList for instance: "GHC
doesn't like it's own type" to paraphrase).

I have a function |update|

> --update :: forall sel val rec val' rec' .
> --              SetField sel val rec val' rec' => sel -> (val -> val') -> rec -> rec'
> update sel f rec = rec'
>       where --val :: val
>              val = value sel rec
>              --val' :: val'
>              val' = f val
>              --rec' :: rec'
>              rec' = set sel val' rec

The commented out signature is the signature that GHC infers for the
function. When I uncomment that signature, it will no longer type
check. Why does this happen? Even with the forall and the explicit
signatures in the where clause, it chokes.

I don't plan on sharing the code for |SetField| because it's that wild
looking type class hackery, and the problem isn't specific to this
example--I remember having it happen in far simpler cases which I
can't seem to find right now...

Is there a canonical example example that exhibits this behavior? Or a
ticket for it already? I'd like to understand what's happening.

Thanks for your time,
Nick