[GHC] #15147: Type checker plugin receives Wanteds that are not completely unflattened

GHC ghc-devs at haskell.org
Tue Jun 12 11:57:51 UTC 2018


#15147: Type checker plugin receives Wanteds that are not completely unflattened
-------------------------------------+-------------------------------------
        Reporter:  nfrisby           |                Owner:  (none)
            Type:  bug               |               Status:  new
        Priority:  normal            |            Milestone:  8.6.1
       Component:  Compiler (Type    |              Version:  8.4.1
  checker)                           |             Keywords:
      Resolution:                    |  TypeCheckerPlugins
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 simonpj):

 To me it seems odd that we unflatten the Wanted `CFunEqCans` but not the
 Given ones.  Indeed, it's not entirely clear how you can decide if a type-
 function call is in which class.

 '''So perhaps it'd be more consistent to do no un-flattening
 whatsoever?'''  (The alternative, of unflattening everything, is not easy
 to implement.)  Since the plugins have to deal with flattened Givens, it's
 probably no harder to deal with unflattened Wanteds too.

 I'd be interested in opinions.

 Meanwhile, would you like to suggest concrete alternative wording for the
 user manual?   I can fill in the blanks if you have any.

-- 
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/15147#comment:16>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler


More information about the ghc-tickets mailing list