Hi all,
I was wondering what assumptions does pmap take with regards to the
ordering of the resulting seq? Can I assume it retains the original
ordering or does it just cons whatever future returns earlier? In other
words can I do this and feel safe?
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------
(defn min-max [root? depth dir board]
(if (zero? depth) (score b) ;reached bottom level - evaluate board
(let [mapper (if root? 'pmap 'map) ;if at root-node create futures
team (filter #(= dir (:direction %)) board) ;all the
team-mates (with same direction)
successors (concat (map #(core/getMoves %) team)) ;concat all
potential moves
result (->> successors
(map core/try-move) ;try them all (producing boards)
(mapper (partial min-max false (dec depth) (-
dir))))];pass the resulting boards back to recursion with direction inverted
(get (vec successors)
(.indexOf result (apply max result))))))
(def search (partial min-max true))
----------------------------------------------------------------------------------------------------------------------------------------------------------------------------
It seems to me like a very succinct and idiomatic way to implement the
min-max algorithm, the only problem is this line:
(get (vec successors) (.indexOf result (apply max result)))
It assumes that the initial ordering of moves ('successors') has been
preserved throughout the recursion and so I can use .indexOf to go from
the score back to the move that started it all! Is this the case? I am
suspecting not! any suggestions?
Thanks in advance,
Jim
--
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
Note that posts from new members are moderated - please be patient with your
first post.
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
