[GHC] #10009: type inference regression when faking injective type families

[GHC]

#10009: type inference regression when faking injective type families
-------------------------------------+-------------------------------------
        Reporter:  aavogt            |                   Owner:
            Type:  bug               |                  Status:  closed
        Priority:  highest           |               Milestone:  7.12.1
       Component:  Compiler (Type    |                 Version:  7.10.1-rc1
  checker)                           |                Keywords:
      Resolution:  fixed             |            Architecture:
Operating System:  Unknown/Multiple  |  Unknown/Multiple
 Type of failure:  GHC rejects       |               Test Case:
  valid program                      |                Blocking:
      Blocked By:                    |  Differential Revisions:
 Related Tickets:  #10226, #10634    |
-------------------------------------+-------------------------------------

Comment (by Lemming):

 Replying to [comment:27 kosmikus]:
 > Re `hlist`: I'm surprised this ever worked. What I usually do is to
 define the `UnPrime`-equivalent outside of the class (as a type family),
 and `Prime` within. I hadn't considered that there may have been GHC
 versions that would have allowed both to be defined within the class.

 I used that technique several times after I found out that it worked. I
 expected that GHC knows what it does. :-) I am now reverting the affected
 definitions.

--
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10009#comment:41>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler