{-# LANGUAGE RankNTypes #-}
{-
From: Ralf Hinze
Date: Fri, 15 Aug 1997 15:20:51 +0200 (MET DST)
I *suppose* that there is a bug in GHC's type checker. The following
program, which I think is ill-typed, passes silently the type checker.
Needless to say that it uses some of GHC's arcane type extensions.
-}
module ShouldFail where
data ContT m a = KContT (forall res. (a -> m res) -> m res)
unKContT (KContT x) = x
callcc :: ((a -> ContT m b) -> ContT m a) -> ContT m a
callcc f = KContT (\cont -> unKContT (f (\a -> KContT (\cont' -> cont a))) cont)
{-
`ContT' is a continuation monad transformer. Note that we locally
qualify over the result type `res' (sometimes called answer or
output). IMHO this make it impossible to define control constructs
like `callcc'. Let's have a closer look: the code of `callcc' contains
the subexpression `KContT (\cont' -> cont a)'. To be well-typed the
argument of `KContT' must have the type `(All res) => (a -> m res) -> m
res'. Quantification is not possible, however, since the type variable
in `cont's type cannot be forall'd, since it also appears at an outer
level. Right? Or wrong?
-}