On Tue, Jan 4, 2011 at 1:22 AM, Isaac Dupree
You also added defaults for most of the Monad methods, though they're obvious and I approve. The (>>) default might have worse performance than the previous default though? (>>) is used in do-notation desugaring, and yet many (most?) Monad instance writers do not explicitly define it, so its default makes some difference. Does anyone know how to test?
Has there been any performance testing of this change? One way to test it would be test the change on some performance sensitive code that uses monads. vector-algorithms [1] (which uses the PrimMonad) comes to mind as a good candidate. Some monadic parser packages, like binary [2] or attoparsec [3], would be interesting to include in the performance test too. Johan 1. http://hackage.haskell.org/package/vector-algorithms 2. http://hackage.haskell.org/package/binary 3. http://hackage.haskell.org/package/attoparsec