Since arrays are not immutable, it's certainly wise to only consider
equality to be based on java object identity ? At least I remember
Rich mentioning this rationale, which sounds reasonable to me.
2010/1/11 Stuart Halloway :
> clojure.core/= handles strings and seqs, but not arrays:
>
> (apply =
clojure.core/= handles strings and seqs, but not arrays:
(apply = ["A" "A"])
=> true
(apply = (map #(.getBytes %) ["A" "A"]))
=> false
(apply = (map #(seq (.getBytes %)) ["A" "A"]))
=> true
I know how this happens, but not why. Is it a motivated design choice,
or would a patch be welcome?
