#4950: Typechecking regression
--------------------------------------+-------------------------------------
Reporter: igloo | Owner:
Type: bug | Status: closed
Priority: highest | Milestone: 7.0.2
Component: Compiler (Type checker) | Version: 7.0.1
Resolution: duplicate | Keywords:
Testcase: | Blockedby:
Difficulty: | Os: Unknown/Multiple
Blocking: | Architecture: Unknown/Multiple
Failure: None/Unknown |
--------------------------------------+-------------------------------------
Changes (by simonpj):
* cc: sjoerd@… (added)
Comment:
PS: the skolem-escape error in `Adjunction` is becuase the module either
needs some type signatures or `NoMonoLocalBinds` (after GADTs).
The problem is here:
{{{
initialPropAdjunction :: (Functor f, Functor g, Category c, Category d,
Dom f ~ d, Cod f ~ c, Dom g ~ c, Cod g ~ d)
=> f -> g -> (forall y. Obj d y -> InitialUniversal y g (f :% y))
-> Adjunction c d f g
initialPropAdjunction f g univ = mkAdjunction f g un coun
where
coun a = initialFactorizer (univ (g % a)) a (g % a)
un a = initialMorphism (univ a)
}}}
Here we need `coun` and `un` to have polymoprphic types, because
`mkAdjunction` has a higher rank type.
So I won't open a new ticket after all. Sjoerd, you might want to update
your code.
Simon
--
Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/4950#comment:4>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler
_______________________________________________
Glasgow-haskell-bugs mailing list
Glasgow-haskell-bugs@haskell.org
http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
