[GHC] #10451: Constraint tuple regression in HEAD

[GHC]

#10451: Constraint tuple regression in HEAD
-------------------------------------+-------------------------------------
        Reporter:  darchon           |                   Owner:
            Type:  bug               |                  Status:  new
        Priority:  normal            |               Milestone:
       Component:  Compiler          |                 Version:  7.11
      Resolution:                    |                Keywords:
Operating System:  Unknown/Multiple  |  ConstraintKinds
 Type of failure:  GHC rejects       |            Architecture:
  valid program                      |  Unknown/Multiple
      Blocked By:                    |               Test Case:
 Related Tickets:                    |                Blocking:
                                     |  Differential Revisions:
-------------------------------------+-------------------------------------

Comment (by thoughtpolice):

 For the record, I strongly prefer that _if anything_, we only bump the
 limit, and '''not''' decrease it. There's just no reason to.

 It's true that 62 is pretty arbitrary, but that's what it already was -
 lowering it to 8 is going to break some existing programs for presumably
 zero user-facing benefit, which basically amounts to breaking programs
 gratuitously. We should really not do this. The fact at least two people
 run into this seems to be enough support for me!

 If anything, we should just increase it. Why not just bump it up to 64, or
 hell, given Anthony said it's already annoying at the current limit, 128?
 If there's no technical reason otherwise, of course. That's at least a net
 benefit for one user. If this is inconsistent with other deriving
 mechanisms for tuples like `Eq`, we should fix that, or at least document
 it clearly. And if ''that'' is hard, well, we may want to fix it if
 possible.

--
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10451#comment:7>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler