[GHC] #8560: Derive some generic representation for GADTs (was: undeducable Typeable error with data kinds when deriving Data for GADT in GHC version 7.7.20131122)

[GHC]

#8560: Derive some generic representation for GADTs
-------------------------------------+-------------------------------------
              Reporter:  carter      |            Owner:
                  Type:  bug         |           Status:  new
              Priority:  normal      |        Milestone:  ⊥
             Component:  Compiler    |          Version:  7.7
  (Type checker)                     |         Keywords:
            Resolution:              |     Architecture:  Unknown/Multiple
      Operating System:              |       Difficulty:  Rocket Science
  Unknown/Multiple                   |       Blocked By:
       Type of failure:              |  Related Tickets:
  None/Unknown                       |
             Test Case:              |
              Blocking:              |
Differential Revisions:              |
-------------------------------------+-------------------------------------
Changes (by goldfire):

 * difficulty:  Unknown => Rocket Science
 * milestone:   => ⊥


Old description:

> This may be an artifact of  issue #8128 (see
> https://ghc.haskell.org/trac/ghc/ticket/8128#comment:5), but I'm seeing
> examples involving data kinds where Typeable isn't deducible.  This could
> be an artifact of other problems, BUT since with 7.7 onwards, we have
> baked in polykinded Typeable, things should always "just work™" right?
>
> the error with current head is as follows (and i'm attaching the code +
> current finger print too)

New description:

 This may be an artifact of  issue #8128 (see
 https://ghc.haskell.org/trac/ghc/ticket/8128#comment:5), but I'm seeing
 examples involving data kinds where Typeable isn't deducible.  This could
 be an artifact of other problems, BUT since with 7.7 onwards, we have
 baked in polykinded Typeable, things should always "just work™" right?

 the error with current head is as follows (and i'm attaching the code +
 current finger print too)

 EDIT: The request here is really for some generic representation derivable
 for GADTs. This is a Research Problem, but one that would really help a
 fair amount of ordinary folk. It won't ever work with `Data`, as we now
 know `Data`, because the types aren't up to the task. But, perhaps some
 future generic representation will work.

 The original description remains unchanged above so that readers can make
 sense of the comments.

--

--
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8560#comment:7>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler