No-seq parsing implementation

altparser.clj

;;exctracted from original function, refined to
;;work with iterables via eduction.
(defn derive-parsers
  [headers parser-fn parser-scan-len data]
  (let [n-cols  (count   headers)]
    (if-not parser-fn
      ;;a bunch of default parsers
      (repeatedly n-cols default-column-parser)
      ;;only sample what we need, don't retain any seq head.
      (->>  data
            (eduction (take parser-scan-len))
            (apply interleave)
            (partition parser-scan-len)
            (map parser-fn headers)))))

;;our reducing function.
;;returns our column parsers (really parse state), uses
;;side effects.
(defn parse-row [^List parsers ^"[Ljava.lang.String;" row]
  (loop [col-idx 0]
    (when (< col-idx n-cols)
      (let [^String row-data (aget row col-idx)
            parser (.get parsers col-idx)]
        (if (and row-data
                 (> (.length row-data) 0)
                 (not (.equalsIgnoreCase "na" row-data)))
          (parse! parser row-data)
          (missing! parser))
        (recur (unchecked-inc col-idx)))))
  parsers)

;;slight modification to original implementation.
;;avoids retaining head of the seq.  uses
;;iterables to generate eduction recipes,
;;and reduce instead of generating intermediate
;;seqs.
(defn csv->columns
  [input & {:keys [header-row? parser-fn
                   parser-scan-len]
            :or {header-row? true
                 parser-scan-len 100}}]
  (let [;;I think we have to create a new iterator each time, so use a function.
        ->rows      (fn [] (raw-row-iterable input))
        initial-row (->> (->rows) (eduction (take 1)) first)
        n-cols      (count initial-row)
        ->data      (fn [] (if header-row?
                        (eduction (drop 1) (->rows))
                        (->rows)))
        header-row  (or (and header-row? initial-row)
                        (vec (map (str "col_" (range n-cols)))))
        ^List column-parsers
        (derive-parsers header-row parser-fn parser-scan-len (->data))]
    (->> (->data)
         (reduce parse-row column-parsers)
         (mapv (fn [init-row-data parser]
                 (assoc (column-data parser)
                        :name init-row-data))
               (if header-row?
                 initial-row
                 (range n-cols))))))