[GHC] #8827: Inferring Safe mode with GeneralizedNewtypeDeriving is wrong

[GHC]

#8827: Inferring Safe mode with GeneralizedNewtypeDeriving is wrong
-------------------------------------+-------------------------------------
        Reporter:  goldfire          |                   Owner:
            Type:  bug               |                  Status:  new
        Priority:  normal            |               Milestone:  7.12.1
       Component:  Compiler          |                 Version:  7.9
      Resolution:                    |                Keywords:
Operating System:  Unknown/Multiple  |            Architecture:
                                     |  Unknown/Multiple
 Type of failure:  None/Unknown      |               Test Case:
      Blocked By:                    |                Blocking:
 Related Tickets:  #8226, #8745      |  Differential Revisions:
-------------------------------------+-------------------------------------

Comment (by goldfire):

 You're right about the annoyance of a nominal default, and that it's worse
 than with classes. Bah.

 And, we may be scared about an efficiency issue that does not arise much
 in practice, especially with a user-controllable cutoff mechanism. I don't
 think adding the constructors to interface files helps much, because a
 client would still need to import the constructors from an exporting
 module.

 Simon PJ has had strong opinions here in the past, and I'm curious to see
 what he thinks.

--
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8827#comment:46>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler