g1 x y z = if x>y then show x ++ show z else g2 y x
g2 :: (Show a, Ord a) => a -> a -> String g2 | False = \p q -> g1 q p () | otherwise = \p q -> g1 q p 'a' where x = True
It appears to me that GHC is justified. According to 4.5.1 and 4.5.2, g1 by itself constitutes a declaration group. It is considered by itself and is generalized prior to combining it with g2.
Great, now I'm even more confused. 4.5.1 says:
A binding b1 depends on a binding b2 in the same list of declarations if either
1. b1 contains a free identifier that has no type signature and is bound by b2, or
2. b1 depends on a binding that depends on b2.
A declaration group is a minimal set of mutually dependent bindings. Hindley-Milner type inference is applied to each declaration group in dependency order.
So here the first binding (of g1) contains the free identifier g2, which is bound by the second binding. Conversely, the second binding contains g1 free. So the two bindings are mutually dependent, no?
No, the binding of g1 doesn't depend on the binding of g2, because g2 has a type signature (clause 1). The type of g1 is inferred using the declared type of g2. Then that type is used in inferring a type for g2, which will be compared with its declared signature.