gnuvince

(import '(java.net URL)
        '(java.lang StringBuilder)
        '(java.io BufferedReader InputStreamReader)
        '(org.htmlparser Parser)
        '(org.htmlparser.visitors NodeVisitor)
        '(org.htmlparser.tags ImageTag))

(def *comics*
  [{:name "Penny-Arcade"
    :url "http://www.penny-arcade.com/comic/"
    :regex #"images/\d{4}/.+?(?:png|gif|jpg)"
    :prefix "http://www.penny-arcade.com/"
    }
   {:name "We The Robots"
    :url "http://www.wetherobots.com/"
    :regex #"comics/.+?[.](?:jpg|png|gif)"
    }
   {:name "Xkcd"
    :url "http://www.xkcd.com"
    :regex #"comics"
    :type :tooltip-comic
    }
  ])

(defn fetch-url
  "Return the web page as a string."
  [address]
  (let [url (URL. address)]
    (with-open [stream (. url (openStream))]
      (let [buf (BufferedReader. (InputStreamReader. stream))]
        (apply str (line-seq buf))))))

(defmulti fetch-comic :type)

(defmethod fetch-comic :default [comic]
  (let [src (fetch-url (:url comic))
        image (re-find (:regex comic) src)]
    (str (or (:prefix comic) (:url comic))
         image)))

(defmethod fetch-comic :tooltip-comic [comic]
  (let [img-tags (ref [])
        parser (Parser. (:url comic))
        visitor (proxy [NodeVisitor] []
                  (visitTag [tag]
                            (when (and (instance? ImageTag tag)
                                       (re-find (:regex comic)
                                                (.getImageURL tag)))
                              (dosync (alter img-tags conj tag)))))]
    (.visitAllNodesWith parser visitor)
    [(.getImageURL (first @img-tags))
     (.getAttribute (first @img-tags) "title")]))

(doseq [comic *comics*]
  (println (:name comic) ":" (fetch-comic comic)))