Fetching PDF text with attributes in Clojure using PDFBox

gistfile1.clj

(ns pdfbox.core
  (:import [org.apache.pdfbox.pdmodel PDDocument]
           [org.apache.pdfbox.util PDFMarkedContentExtractor TextPosition]
           [java.util ArrayList]))

(defn parse-pdf [filename]
  (let [pages (.getAllPages (.getDocumentCatalog (PDDocument/load filename)))
        textpool (ArrayList.)
        extract-text (proxy [PDFMarkedContentExtractor] []
                       (processTextPosition [text]
                         (.add textpool text)))]
    (doseq [page pages]
      (when-let [contents (.getStream (.getContents page))]
        (.processStream extract-text page  (.findResources page) contents)))
    textpool))