request for reviews for my first patch -- ticket 7401

[Edward Z. Yang]

Hello Omer,

It is almost certainly the case that the implementation of these
functions should not be the constant return True function.  Furthermore,
the implementation should probably be done using:

    http://ghc.haskell.org/trac/ghc/ticket/2431

Edward

Excerpts from Ömer Sinan Ağacan's message of Fri Aug 09 08:06:19 -0700 2013:
> Hi all,
> 
> I just started GHC development made a commit for ticket 7401:
> http://ghc.haskell.org/trac/ghc/ticket/7401
> 
> My patch is here:
> https://github.com/osa1/ghc/commit/3ec257ff48372de30df59cd8854ce28954c9db95
> 
> 
> `make test` succeeds. My test case for this patch is something like this:
> 
> 
>     data D deriving Eq
> 
>     main :: IO ()
>     main = print ((undefined :: D) == (undefined :: D))
> 
> 
> example:
> 
> 
> ➜  haskell  ./ghc/inplace/bin/ghc-stage2 --interactive derive.hs
> GHCi, version 7.7.20130806: http://www.haskell.org/ghc/  :? for help
> Loading package ghc-prim ... linking ... done.
> Loading package integer-gmp ... linking ... done.
> Loading package base ... linking ... done.
> [1 of 1] Compiling Main             ( derive.hs, interpreted )
> Ok, modules loaded: Main.
> ghci> main
> True
> it :: ()
> 
> 
> This behavior(returning True) is consistent with standalone deriving version:
> 
> 
>     ➜  haskell  cat derive_standalone.hs
>     {-# LANGUAGE StandaloneDeriving #-}
> 
>     data D
> 
>     deriving instance Eq D
> 
>     main :: IO ()
>     main = print ((undefined :: D) == (undefined :: D))
> 
>     ➜  haskell  ./ghc/inplace/bin/ghc-stage2 --interactive derive_standalone.hs
>     GHCi, version 7.7.20130806: http://www.haskell.org/ghc/  :? for help
>     Loading package ghc-prim ... linking ... done.
>     Loading package integer-gmp ... linking ... done.
>     Loading package base ... linking ... done.
>     [1 of 1] Compiling Main             ( derive_standalone.hs, interpreted )
>     Ok, modules loaded: Main.
>     ghci> main
>     True
>     it :: ()
>     ghci>
>     Leaving GHCi.
> 
> 
> 
> However, if you want (==) implementation in this case to be `error
> "Void (==)`, I think I can also do that by first fixing the code
> generated by StandaloneDeriving extension and then fixing my current
> patch.
> 
> 
> 
> Any comments and reviews would be appreciated!
> 
> 
> ---
> Ömer Sinan Ağacan
> http://osa1.net
>