#14860: QuantifiedConstraints: Can't quantify constraint involving type family -------------------------------------+------------------------------------- Reporter: RyanGlScott | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Compiler (Type | Version: 8.5 checker) | Keywords: Resolution: | QuantifiedConstraints wipT2893 Operating System: Unknown/Multiple | Architecture: Type of failure: GHC rejects | Unknown/Multiple valid program | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Rev(s): Wiki Page: | -------------------------------------+------------------------------------- Comment (by goldfire): I'm struggling with this one, worried about order-dependence. Suppose `Apply f Int` is `Bool`. Now suppose we are given `forall a. Show (Apply f a)` and want to prove `Show (Apply f Int)`. In considering the wanted `Show (Apply f Int)`, we have two choices of how to proceed: 1. Use the given and succeed. 2. Reduce `Apply f Int` to `Bool` and fail. The problem is that it seems challenging to guarantee that we don't do (2). Simon's answer appears to be to simply disallow the constraint in the first place. -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/14860#comment:6 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler