[GHC] #12949: Pattern coverage mistake

[GHC]

#12949: Pattern coverage mistake
-------------------------------------+-------------------------------------
        Reporter:  dfeuer            |                Owner:  gkaracha
            Type:  bug               |               Status:  new
        Priority:  normal            |            Milestone:  8.2.1
       Component:  Compiler          |              Version:  8.0.1
      Resolution:                    |             Keywords:
                                     |  PatternMatchWarnings
Operating System:  Unknown/Multiple  |         Architecture:
 Type of failure:  Incorrect         |  Unknown/Multiple
  error/warning at compile-time      |            Test Case:
      Blocked By:                    |             Blocking:
 Related Tickets:                    |  Differential Rev(s):
       Wiki Page:                    |
-------------------------------------+-------------------------------------

Comment (by gkaracha):

 Replying to [comment:9 simonpj]:
 > > Indeed you are right, this is not a bug in the type constraints that
 the checker generates but the term constraints. I expected the knownBools
 to be already desugared to (knownBool $dKnownBool_a) and (knownBool
 $dKnownBool_b) since they are overloaded but I was mistaken.
 >
 > I'm going to need more guidance before I can help.  Can you give
 instructions to reproduce the bit you think is unexpected?
 >
 > I compiled the program (adding a definition for `Dict` which you
 omitted), with `-ddump-ds` and got
 > {{{
 > ...
 >         ds_d3Xn = knownBool @ a $dKnownBool_a3Ux } in
 > ...
 >             ds_d3Xo = knownBool @ b $dKnownBool_a3UL } in
 >
 > }}}
 > which is what you expect.  This is post-desugaring.  Before desugaring
 there'll be a `HsWrapper` involved.

 Ah, great, thanks! This may be sufficient: I think that `translatePat` in
 `deSugar/Check.hs` just needs to look at the wrapper in the `CoPat` case
 then. It already inspects it for dropping trivial coercions, as an
 optimization. I'll look into it and ping you if anything else gets in the
 way (I hope not).

--
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/12949#comment:10>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler