html scraping test /clojure

htmlscrapetest.clj

(ns htmlscrapetest
  (:require [clojure.xml :as xml]
            [clojure.zip :as zip]
            [clojure.contrib.zip-filter :as z]
            [clojure.contrib.zip-filter.xml :as zf])
  (:use [clojure.java.io :only [as-file]])
  
  ; http://htmlcleaner.sourceforge.net/
  (:import [org.htmlcleaner CleanerProperties CompactXmlSerializer HtmlCleaner])
  )

(defn mytest [zipper]
  (map #(zipmap '(:thumb_url :illust_id :title) %)
       (map conj
            (map #(vector
                    (java.net.URL. %)
                    (second (re-find (re-matcher #"/(\d+)_s\.(?:jpg|gif|png)" %))))
                 (zf/xml-> zipper z/descendants :ul :li :a :img (zf/attr :src)))
            (zf/xml-> zipper z/descendants :ul :li :a :img (zf/attr :alt))
            )))

(defn parse-html [f]
  (let [cleaner (HtmlCleaner.)]
    (doto (.getProperties cleaner)
      (.setOmitComments true)
      (.setPruneTags "script, style"))
    (when-let [node (.clean cleaner (as-file f) "utf-8")]
      (zip/xml-zip
        (xml/parse
          (org.xml.sax.InputSource.
            (java.io.StringReader. 
              (.getAsString (CompactXmlSerializer. (.getProperties cleaner)) node "utf-8"))))))
  ))