On Dec 19, 5:14 pm, Michael Snoyman <mich...@snoyman.com> wrote: > > Well, here's one way of doing it. You have lots of choices here; these are > the decisions I made in implementing the code: > > * myFuncB no longer lives in the IO monad. I wasn't sure if you specifically > wanted that, but now it can work with *any* instance of Failure. > * Since I assumed you ultimately wanted it to land in the IO monad, I > defined Exception instances. However, if you were dealing with a different > Failure instance (like [] or Maybe), these would be unncesary. > * I also assume that what you meant by "your code" and "their code" is that > you can modify your own code, but not theirs. > > If you show me what the real code is you're working on, I'd be happy to more > fully develop a better solution with you. Anyway, here's the code.
Thank you very much. This helped a lot. I wasn't aware of many of the details you showed me now, such as deriving from Typeable. The assumptions you have made w.r.t to what I want to write all make sense and I guess I can experiment with Control.Failure a little better now. Thanks again, nt _______________________________________________ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe