exploring-data.clj

(ns duplicates
  (:require [clojure.data.csv :as csv]
            [clojure.java.io :as io]
            [clojure.string :as str]
            [spec-provider.stats :as stats]
            [clojure.pprint :as pp]))

(defn csv-data->maps [csv-data]
  (map zipmap
       (->> (first csv-data)
            (map (comp keyword #(str/replace % #" " "-")))
            repeat)
       (rest csv-data)))

(defn write-dataset-edn! [out-file raw-dataset-map]
  (with-open [w (clojure.java.io/writer out-file)]
    (binding [*out* w]
      (pp/write raw-dataset-map))))

(defn -main []
  (->>
   (csv/read-csv (io/reader "data/file.csv") :separator \;)
   (csv-data->maps)
   (stats/collect)
   (write-dataset-edn! "output/stats.edn")))

deps.edn

{:deps
 {org.clojure/data.csv {:mvn/version "0.1.4"}
  spec-provider {:mvn/version "0.4.11"}}}

Associated article (in French) : https://iteracode.fr/blog/clojure-et-exploration-de-donnees