#13018: TH-spliced pattern synonym declaration fails to typecheck -------------------------------------+------------------------------------- Reporter: RyanGlScott | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler (Type | Version: 8.0.1 checker) | Keywords: Resolution: | PatternSynonyms 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 simonpj): It's not a regression; it's a bug-fix! See Trac #11224. Here's the note from `TcTySigs`: {{{ Note [The pattern-synonym signature splitting rule] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Given a pattern signature, we must split the kind-generalised variables, and the implicitly-bound variables into universal and existential. The rule is this (see discussion on Trac #11224): The universal tyvars are the ones mentioned in - univ_tvs: the user-specified (forall'd) universals - req_theta - res_ty The existential tyvars are all the rest }}} If you provide an explicit forall, they are all universals; but in this case we wanted 'b' to be an existential. According to these rules you can say {{{ pattern P :: b -> T a or pattern P :: forall a. () => forall b. b -> T a }}} We neglected to put this rule in the user manual: that should be fixed. '''Matthew or Ryan, might you do that?''' The TH problem is that TH is taking a quote with an implicit forall, and turning it into a TH data structure with an explicit forall. When that is converted back into `HsSyn` it has an explicit forall, so we get `pattern P :: forall b a. b -> T a`. So the round trip is not faithful. I wonder if we could just genertate TH syntax with an implicit forall? -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/13018#comment:5 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler