regex.clj

;; Growing a little DSL for regexes
;; prompted by http://stackoverflow.com/questions/2553668/how-to-remove-list-of-words-from-strings

(defmulti pattern type)
(defn regex [spec]
  (-> spec pattern java.util.regex.Pattern/compile))

(defmethod pattern String [s]
  (java.util.regex.Pattern/quote s))
(defmethod pattern clojure.lang.IPersistentSet [alts]
  (str "(?:" (->> alts (map regex) (interpose \|) (apply str)) ")"))
(defmethod pattern clojure.lang.IPersistentVector [alts]
  (str "(?:" (->> alts (map regex) (apply str)) ")"))

(comment
user=> (def forbidden-words [":)" "the" "." ","])
user=> (def strings ["the movie list" "this.is.a.string" "haha :)"])
user=> (map #(.replaceAll % (-> forbidden-words set pattern) "") strings)
(" movie list" "thisisastring" "haha ")
)