[GHC] #13083: Constraint solving failure with Coercible + type family + newtype

[GHC]

#13083: Constraint solving failure with Coercible + type family + newtype
-------------------------------------+-------------------------------------
        Reporter:  conal             |                Owner:
            Type:  bug               |               Status:  new
        Priority:  normal            |            Milestone:
       Component:  Compiler (Type    |              Version:  8.0.1
  checker)                           |
      Resolution:                    |             Keywords:  Coercible
Operating System:  MacOS X           |         Architecture:
 Type of failure:  GHC rejects       |  Unknown/Multiple
  valid program                      |            Test Case:
      Blocked By:                    |             Blocking:
 Related Tickets:                    |  Differential Rev(s):
       Wiki Page:                    |
-------------------------------------+-------------------------------------
Changes (by RyanGlScott):

 * cc: goldfire (added)


Comment:

 I believe this is expected behavior. You can only coerce from `V Par1` to
 `Par` if the [https://ghc.haskell.org/trac/ghc/wiki/Roles role] of `V1`'s
 argument is permissive enough (that is, it must have a representational or
 phantom role, not a nominal role). But currently, all arguments to a type
 family (such as `V`) are always assigned nominal roles. This solves
 problems like the
 [https://ghc.haskell.org/trac/ghc/wiki/Roles#Theproblemwewishtosolve
 original motivation for roles], but it does rule out scenarios like yours.

 FWIW, I've been advocating for role inference/the ability to give type
 families role annotations in #8177. I think a type family like:

 {{{#!hs
 type family Id a where
   Id a = a
 }}}

 should definitely be able to have a representational role for its
 argument. Not sure about `V`, though. I'll cc goldfire for his thoughts as
 well.

--
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13083#comment:1>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler