[GHC] #9755: Unhelpful error message when -XScopedTypeVariables is omitted (was: Monomorphism related Ix/Vector error when code is loaded by GHCi/cabal repl)

GHC ghc-devs at haskell.org
Mon Nov 3 12:57:39 UTC 2014


#9755: Unhelpful error message when -XScopedTypeVariables is omitted
-------------------------------------+-------------------------------------
              Reporter:  bitemyapp   |            Owner:
                  Type:  bug         |           Status:  new
              Priority:  normal      |        Milestone:
             Component:  GHCi        |          Version:  7.8.3
            Resolution:              |         Keywords:
      Operating System:              |     Architecture:  Unknown/Multiple
  Unknown/Multiple                   |       Difficulty:  Unknown
       Type of failure:              |       Blocked By:
  None/Unknown                       |  Related Tickets:
             Test Case:              |
              Blocking:              |
Differential Revisions:              |
-------------------------------------+-------------------------------------

Comment (by simonpj):

 I believe that the real problem is that in the definition of
 {{{
 vecIndexIx  :: (Ix ix, Bounded ix) => Vector a -> ix -> a
 vecIndexIx vec ix = vec ! Ix.index (minBound :: ix, maxBound :: ix) ix
 }}}
 you probably think that the uses of `ix` in the RHS mean the same `ix` as
 in the type signature.  To achieve that you need `-XScopedTypeVariables`
 and an explicit `forall`:
 {{{
 vecIndexIx  :: forall a ix. (Ix ix, Bounded ix) => Vector a -> ix -> a
 vecIndexIx vec ix = vec ! Ix.index (minBound :: ix, maxBound :: ix) ix
 }}}
 Now it works fine.  What you wrote is equivalent to
 {{{
 vecIndexIx vec ix = vec ! Ix.index (minBound :: forall ix. ix,
                                     maxBound :: forall ix. ix) ix
 }}}
 where the type signature `blah :: ix` is universally quantified to `blah
 :: forall ix. ix`.

 It's an easy mistake to make.  It might be a good idea if GHC spotted type
 variables that have the same name as one belonging to an enclosing, but
 un-scoped, type signature, and suggested this change.  If someone wanted
 to try that, I could advise.

 Simon

--
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9755#comment:4>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler


More information about the ghc-tickets mailing list