On 28/07/2009, at 6:41 AM, John Dorsey wrote:
I'm assuming that name resolution is currently independent of type inference, and will happen before type inference. With the proposal this is no longer true, and in general some partial type inference will have to happen before conflicting unqualified names are resolved.
My worry is that the proposal will require a compliant compiler to interweave name resolution and type inference iteratively.
To my untrained eye it looks complicated and invasive, even without the mutually recursive case. Can anyone shed light on whether this would be a problem for, say, GHC?
My experimental compiler DDC [1] implements TDNR almost exactly as given on the Haskell' wiki. Yes, you have to interweave name resolution with type inference, because there is no way to compute the binding dependency graph/call graph before type inference proper. This is discussed in section 3.5 of my thesis [2] (which is currently under examination). For DDC I used a constraint based inference algorithm to compute the binding dependency graph "on the fly", but I don't know how easy it would be to retrofit this method into GHC. Cheers, Ben. [1] http://www.haskell.org/haskellwiki/DDC [2] http://cs.anu.edu.au/people/Ben.Lippmeier/project/thesis/thesis-lippmeier-su...