#15828: Type family equation foralls allow strange re-quantification of class-bound type variables -------------------------------------+------------------------------------- Reporter: RyanGlScott | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.7 Resolution: | Keywords: TypeFamilies Operating System: Unknown/Multiple | Architecture: | Unknown/Multiple Type of failure: None/Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Rev(s): Wiki Page: | -------------------------------------+------------------------------------- Comment (by mayac): I also agree `C1` should be rejected, though I'm quite baffled by why that is currently not the case. It appears that the two `a`s in the `C1` example truly do have the same name post-`GhcRn`. Shouldn't `bindLHsTyVarBndrs` be coming up with a fresh name for each explicitly quantified type variable anyway? I don't think I understand enough about the way GHC comes up with new names to see where this goes wrong... In response to @goldfire, I would be happy to address these things as they come up! Of course, if someone finds a fix faster than I do they should feel free to go ahead with the change. In this case, I would need some guidance as I expressed above. As for the documentation, how about the following:
When an explicit `forall` is present, all ''type'' variables mentioned which are not already in scope must be bound by the `forall`.
-- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/15828#comment:2 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler