[GHC] #8566: Given kind equalities are discarded

GHC ghc-devs at haskell.org
Fri Jan 10 20:14:25 UTC 2014


#8566: Given kind equalities are discarded
-------------------------------------+------------------------------------
        Reporter:  dreixel           |            Owner:
            Type:  bug               |           Status:  new
        Priority:  normal            |        Milestone:
       Component:  Compiler          |          Version:  7.7
      Resolution:                    |         Keywords:
Operating System:  Unknown/Multiple  |     Architecture:  Unknown/Multiple
 Type of failure:  None/Unknown      |       Difficulty:  Unknown
       Test Case:  polykinds/T8566   |       Blocked By:
        Blocking:                    |  Related Tickets:
-------------------------------------+------------------------------------

Comment (by goldfire):

 I can't find a case where I want an existentially quantified kind variable
 in an equality constraint, so I think that proposal is viable. ''But'', I
 think your comment that "'mentioned in an equality constraint' is not
 obvious" is an understatement. What if a context uses an open type family?
 Then, we can't know whether or not the type family will expand out to an
 equality constraint. We could forbid those, too, I suppose, but I don't
 like where this is going.

 I might prefer emitting a (suppressable) warning in this case, saying that
 the user is pushing the type system beyond its limits, really, and to
 expect the unexpected. Or, we could go the other way and allow such
 declarations only when the user specifies `-XAllowWonkyExistentialKinds`
 or similar.

--
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8566#comment:11>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler


More information about the ghc-tickets mailing list