[GHC] #15681: Take exhaustiveness checking into consideration when using MonadFailDesugaring

[GHC]

#15681: Take exhaustiveness checking into consideration when using
MonadFailDesugaring
-------------------------------------+-------------------------------------
        Reporter:  chshersh          |                Owner:  (none)
            Type:  bug               |               Status:  new
        Priority:  normal            |            Milestone:  8.8.1
       Component:  Compiler          |              Version:  8.6.1
      Resolution:                    |             Keywords:  pattern-
                                     |  matching,monadfail,desugaring,PatternSynonyms,PatternMatchWarnings
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):

 The real problem here is that we are trying to ''infer'' whether any of
 the patterns can fail -- if so, generate a `MonadFail` constraint, but not
 if not.  But pattern-match-coverage is a tricky thing, as the examples
 demonstrate.  And making the generation of a constraint depend on how
 another set of constraints is solved is pretty thin ice.

 I wish it were possible to ''specify unambiguously'' whether you want
 `Applicative` or `Monad` or `MonadFail`.  Something like `do{Monad}` or
 `do{Applicative}` or `do{MonadFail}`.  That would be a lot clearer!

-- 
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/15681#comment:5>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler