[GHC] #12087: Inconsistency in GADTs?

[GHC]

#12087: Inconsistency in GADTs?
-------------------------------------+-------------------------------------
        Reporter:  Iceland_jack      |                Owner:  RyanGlScott
            Type:  task              |               Status:  new
        Priority:  normal            |            Milestone:
       Component:  Compiler          |              Version:  8.0.1
      Resolution:                    |             Keywords:  GADTs
Operating System:  Unknown/Multiple  |         Architecture:
                                     |  Unknown/Multiple
 Type of failure:  None/Unknown      |            Test Case:
      Blocked By:                    |             Blocking:
 Related Tickets:  #11540            |  Differential Rev(s):
       Wiki Page:                    |
-------------------------------------+-------------------------------------

Comment (by RyanGlScott):

 Replying to [comment:4 simonpj]:
 > I agree this is possible in principle but it is fiddly in practice.
 What about
 > {{{
 > MkF :: Ord a => a -> Eq a => F a
 > }}}
 > which ought to work too.

 Sure, I completely agree. `tcSplitNestedSigmaTys` can handle that case, so
 `tcSplitNestedLHsSigmaTys` ought to as well.

 > > This isn't exactly what the user wrote, but it's equivalent and
 serviceable.
 >
 > Yes but we'll soon have people asking that `:t MkF` prints out the type
 they wrote.

 We can cross that bridge when we come to it.

 FWIW, GHC doesn't have a solid track record of preserving nested sigma
 types for functions either:

 {{{
 λ> let f :: Ord a => Eq a => a -> Bool; f = undefined
 λ> :type f
 f :: Ord a => a -> Bool
 λ> let g :: Ord a => a -> Eq a => a; g = undefined
 λ> :type g
 g :: Ord a1 => a2 -> a1
 }}}

 So it (personally) wouldn't bother me to have `:type MkF` print out
 something slightly different than what I typed. After all, if I'm using
 nested sigma types in the first place, there's a good chance I know why
 GHC's displayed type is tweaked (but equivalent).

--
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/12087#comment:5>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler