#6159: unexpected GADT kind error
----------------------------------------+-----------------------------------
Reporter: kosmikus | Owner:
Type: bug | Status: closed
Priority: normal | Milestone:
Component: Compiler | Version: 7.5
Resolution: invalid | Keywords:
Os: Unknown/Multiple | Architecture: Unknown/Multiple
Failure: GHC rejects valid program | Difficulty: Unknown
Testcase: | Blockedby:
Blocking: | Related:
----------------------------------------+-----------------------------------
Comment(by simonpj):
Ah yes, that's a good point. The difficult is this. For polymorphic
recursion we need to know when you have supplied a complete kind signature
for a type contructor. When (and only when) you supply a complete kind
signature, you can use polymorphic recursion.
At the value level this is obvious enough: you have supplied a type
signature for `f` if you write `f :: blah`. But at the kind level it's
not so obvious. In your definition for `Y` you have supplied part, but
not all, of a kind signature.
Currently I've said that you have supplied "a complete kind signature"
when you give a "::" in the definition. That is,
* with "::" the kind is ''declared'';
* without "::" it is ''inferred''.
That certainly is a change but it seems one that is easy to explain. I'd
welcome suggestions for what changes I should make elsewhere in the user
manual; where did you look when you were seeking information?
--
Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/6159#comment:3>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler
_______________________________________________
Glasgow-haskell-bugs mailing list
[email protected]
http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
