List is a monad because it has an associated way of sequencing computations: namely, "collect results of the second computation invoked on all results of the first computation". That's not because List is a data structure (similarly to Heap), it's because it is associated with the computational abstraction of computations with multiple results. The heap data structure does not have an associated computational abstraction, at least not that one I'm aware of, and definitely not one that would make any sense in the context of heapsort. So it doesn't make sense to pretend that it does.
On Wed, Jul 11, 2012 at 5:00 PM, Qi Qi <qiqi...@gmail.com> wrote: > List [] is a monad, why not for heap data. Heap data could be an instance of > Monad too. > I have the heapsort function, and just wanted to rewrite a verbose version > of it by using liftIO. > But I would look into Debug.Trace. Thanks for your hint. > > Qi > > > On Wednesday, July 11, 2012 5:28:17 PM UTC-5, Eugene Kirpichov wrote: >> >> Use Debug.Trace. >> It does not make sense to declare that heap is a monad, as a monad is an >> abstraction of sequencing computations, and a heap is not an abstraction of >> sequencing computations at all. You don't make your String class implement >> the "rendering engine" interface just because you want to use it in a >> computer game program, equally you dont pretend that a heap is a way of >> sequencing computations just because you want to sequence computations >> related to heaps. >> >> The actual computation in your case is the heapsort function, not the >> heap. If you absolutely must use IO, add IO to the functions type. >> >> 11.07.2012, в 15:19, Qi Qi <qiqi...@gmail.com> написал(а): >> >> > Hi, >> > >> > I was wondering about creating an instance of MonadIO for a heap data. >> > Any hints? >> > >> > data Heap a = E | T Int a (Heap a) (Heap a) >> > deriving (Eq, Ord, Read, Show) >> > >> > The reason is that I want to use liftIO during a heapsort to print out >> > intermediate results. >> > >> > Thanks. >> > >> > Qi Qi >> > >> > >> > >> > _______________________________________________ >> > Haskell-Cafe mailing list >> > Haskell-Cafe@haskell.org >> > http://www.haskell.org/mailman/listinfo/haskell-cafe >> >> _______________________________________________ >> Haskell-Cafe mailing list >> Haskell-Cafe@haskell.org >> http://www.haskell.org/mailman/listinfo/haskell-cafe > > > _______________________________________________ > Haskell-Cafe mailing list > Haskell-Cafe@haskell.org > http://www.haskell.org/mailman/listinfo/haskell-cafe > -- Eugene Kirpichov http://www.linkedin.com/in/eugenekirpichov _______________________________________________ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
