distinct-by clojure transducer

distinct-by.clj

(defn distinct-by
  "Returns a lazy sequence of the elements of coll, removing duplicates of (f item).
  Returns a stateful transducer when no collection is provided."
  {:added "1.0"}
  ([f]
   (fn [rf]
     (let [seen (volatile! #{})]
       (fn
         ([] (rf))
         ([result] result)
         ([result input]
          (let [value (f input)]
            (if (contains? @seen value)
              result
              (do (vswap! seen conj value)
                  (rf result input))))))))
  ([f coll]
   (let [step (fn step [xs seen]
                (lazy-seq
                 (when-let [s (seq xs)]
                   (let [h (first s)
                         t (rest s)
                         value (f h)]
                     (if (contains? seen value)
                       (recur t seen)
                       (cons h (step t (conj seen value)))))))]
     (step coll #{}))))

The excellent medley library has a correct (distinct-by f) transducer:

https://github.com/weavejester/medley/

Here's simple version for anyone else who comes across this page.

(defn uniq-by
  ""
  [f items]
  (:uniq (reduce (fn [acc m]
                      (let [k (f m) ]
                        (if (get-in acc [:seen k])
                          acc
                          (-> acc
                              (update :seen assoc k true)
                              (update :uniq conj m)))))
                    {:seen {} :uniq []}
                    items)))

(comment
  (uniq-by :name [{:name "Bob"} {:name "Jane"} {:name "Bob"}])
  )