Thanks for the information Konrad. I haven't had a chance to take a close look at monads, but you've bumped it much higher on my list of things to look at.
I was going to attempt to implement the data structure in clojure initially, but I wanted to make it as close a cousin to the other data structures as possible. I also wanted to easily integrate the use of the data structure with other math libraries, and lastly I wanted to be sure that I was taking advantage of all the speed java has to offer. I'm going to do some refactoring and rewriting soon, so the feedback is very welcome. -Adler On May 18, 5:54 am, Konrad Hinsen <konrad.hin...@laposte.net> wrote: > On May 18, 2009, at 11:21, Anand Patil wrote: > > > Huh, sounds like just the thing. Security is going to be especially > > difficult in Clojure, though. For example, say I stuck the array > > into a ref from within the monad, then carried on overwriting it. A > > consumer who gets the array out of the ref from another thread > > might get it while I'm still mutating it. > > That is not possible: the monad provides operations on the array, but > all references to the array are implicit. There is no way to get the > reference out of the monad and pass it to a function or assign it to > a var. Well, *almost* no way: as always in Clojure, you can reverse- > engineer the implementation to figure out how the reference is > stored, and then access it directly. But that is not likely to happen > by mistake. > > Here is an example of how this might look, using 1D Java arrays for > simplicity: > > (def an-array > (initalize-array Integer/TYPE 10 > [_ (set-element 0 1) > x (get-element 0) > _ (set-element 1 (inc x)])) > > This would create an integer array of length 10 (make-array Integer/ > TYPE 10) and then execute the sequence of monadic operations on it. > The available operations are (get-element i) for getting an element > and (set element i v) for changing an element's value. As you can > see, these operations do not have an explicit reference to the array. > > Using the standard operations on monads, one could then define more > complex array initializers in terms of get-element and set-element. > None of them would ever contain an explicit array reference, and none > of them could be applied to an already existing array, except by > using "forbidden knowledge". > > BTW, the above expression would expand into > > (def an-array > (let [a make-array Integer/TYPE 10] > init (domonad state-m > [_ (set-element 0 1) > x (get-element 0) > _ (set-element 1 (inc x)] > nil)] > (second (init a)))) > > but that is already forbidden knowledge, as it shows how to inject an > array into a function that modifies it. The monad could be made > private for additional protection. > > Konrad. --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en -~----------~----~----~----~------~----~------~--~---
