On 4 July 2010 00:24, Yves Parès <limestr...@gmail.com> wrote: > then it works, so obviously, the trouble is about pattern matching. > What was I doing wrong?
This seems to be in violation of the Haskell Report. See e.g. section 3.3 (http://www.haskell.org/onlinereport/exps.html): """ The following identity holds: \ p1 ... pn -> e = \ x1 ... xn -> case (x1, ..., xn) of (p1, ..., pn) -> e where the xi are new identifiers. """ But: runSomeMonad4, runSomeMonad5 :: (forall s. SomeMonad s a) -> a runSomeMonad4 = \x -> case x of SomeMonad x -> runIdentity x runSomeMonad5 = \(SomeMonad x) -> runIdentity x runSomeMonad4 is accepted and runSomeMonad5 is rejected. Of course, the Report doesn't make provision for rank-N types, so this is not *strictly* speaking a violation. But it does seem against the *spirit* of the report, if nothing else! Cheers, Max _______________________________________________ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
