Re: [GHC] #14045: Data family instances must list all patterns of family, despite documentation's claims to the contrary

31 Jul 2017

      #14045: Data family instances must list all patterns of family, despite
documentation's claims to the contrary
-------------------------------------+-------------------------------------
        Reporter:  RyanGlScott       |                Owner:  (none)
            Type:  bug               |               Status:  new
        Priority:  normal            |            Milestone:
       Component:  Compiler (Type    |              Version:  8.3
  checker)                           |
      Resolution:                    |             Keywords:  TypeFamilies
Operating System:  Unknown/Multiple  |         Architecture:
                                     |  Unknown/Multiple
 Type of failure:  GHC rejects       |            Test Case:  indexed-
  valid program                      |  types/should_compile/T14045,
                                     |  indexed-types/should_fail/T14045a
      Blocked By:                    |             Blocking:
 Related Tickets:  #12369            |  Differential Rev(s):  Phab:D3804
       Wiki Page:                    |
-------------------------------------+-------------------------------------

Comment (by goldfire):

 There isn't a difference.

 {{{#!hs
 data instance Sing (a :: Bool) where ...
 }}}

 and

 {{{#!hs
 data instance Sing :: Bool -> Type where ...
 }}}

 compile to precisely the same internal representation. It's the same
 because the first one gets eta-reduced to look like the second. That's why
 the note says

 {{{
 This instance is equivalent to `data instance Sing (a :: Bool)`
 }}}

 I'm not sure where you get "very nearly equivalent" above.

-- 
Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/14045#comment:12
GHC http://www.haskell.org/ghc/
The Glasgow Haskell Compiler