Hello!
I attach a program which I suspect demonstrates a bug in GHC. The
important lines are:
showType :: forall a . Expr a -> String
showType (Lit _) = show (typeOf (undefined :: a))
test1 = showType (mk :: Expr BOOL) -- Prints "Bool" (wrong?)
test2 = showType (Lit mk :: Expr BOOL) -- Prints "Main.BOOL" (correct)
test1 and test2 give different results, even though showType shouldn't
be able to tell them apart. It seems that the Typeable context packed
with the Lit constructor is wrong in test2.
I had to use two extra classes and newtype deriving to trigger this
behavior. Note that if I change
deriving instance B BOOL
to
instance B BOOL
the result is correct.
Tested with both 6.12.3 and 7.0.1.
Should I report this as a bug?
Thanks!
/ Emil
import Data.Typeable
data Expr a where
Lit :: Typeable a => a -> Expr a
class A a where
mk :: a
class (Typeable a, A a) => B a where
mkExpr :: Expr a
mkExpr = Lit mk
instance B a => A (Expr a) where
mk = mkExpr
instance A Bool where
mk = True
newtype BOOL = BOOL Bool
deriving (Typeable, A)
instance B Bool
deriving instance B BOOL
showType :: forall a . Expr a -> String
showType (Lit _) = show (typeOf (undefined :: a))
test1 = showType (mk :: Expr BOOL) -- Prints "Bool" (wrong?)
test2 = showType (Lit mk :: Expr BOOL) -- Prints "Main.BOOL" (correct)
_______________________________________________
Glasgow-haskell-users mailing list
Glasgow-haskell-users@haskell.org
http://www.haskell.org/mailman/listinfo/glasgow-haskell-users
