[GHC] #11715: Constraint vs *

[GHC]

#11715: Constraint vs *
-------------------------------------+-------------------------------------
        Reporter:  bgamari           |                Owner:
            Type:  bug               |               Status:  new
        Priority:  high              |            Milestone:  8.0.1
       Component:  Compiler (Type    |              Version:  8.0.1-rc1
  checker)                           |
      Resolution:                    |             Keywords:
Operating System:  Unknown/Multiple  |         Architecture:
                                     |  Unknown/Multiple
 Type of failure:  None/Unknown      |            Test Case:
      Blocked By:                    |             Blocking:
 Related Tickets:                    |  Differential Rev(s):
       Wiki Page:                    |
-------------------------------------+-------------------------------------

Comment (by Iceland_jack):

 Replying to [comment:9 simonpj]:

 > * If we were to silence `checkValidType` (with a language extension), it
 would also allow functions like
 > {{{
 > reify :: c => c
 > with  :: c -> (c => r) -> r
 > }}}

 This was one of my use cases for #11441, this would be an interesting
 solution and may supersede that part of the proposal. Would this be
 enabled for each module or as a pragma:

 {{{#!hs
 {-# ALLOWINVALIDTYPE f #-}
 f :: Int => Int
 f = undefined
 }}}

 Heck, it could go as far as *require* an ill-kinded type — `{-#
 INVALIDTYPE f #-}` — such that this would not compile:

 {{{#!hs
 {-# INVALIDTYPE f #-}
 f :: Int -> Int
 f = undefined
 }}}

 What would be the use of that? It's machine checkable and who knows what
 changed during refactoring, a reader won't have to guess if there is some
 funny business going on.

--
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11715#comment:11>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler