from pdfminer.pdfinterp import PDFResourceManager, process_pdf

from pdfminer.converter import TextConverter

from pdfminer.layout import LAParams

import threading, os

class PdfThread(threading.Thread):

    def __init__(self, in_queue, doc_txt_dir):

        threading.Thread.__init__(self)

        self.in_queue = in_queue

        self.doc_txt_dir = doc_txt_dir

    def run(self):

        while True:

            try:

                codec = 'utf-8'

                in_fname = self.in_queue.get()

                rsrc = PDFResourceManager(caching = True)

                base_name = os.path.basename(in_fname)

                out_file = os.path.join(self.doc_txt_dir, base_name[0:base_name.rfind(".")] + ".txt")

                outfp = file(out_file, 'w')

                laparams = LAParams() #加上此参数可保留原pdf中的字符间空格

                device = TextConverter(rsrc, outfp, codec=codec, laparams=laparams)

                fp = file(in_fname, 'rb')

                caching = True

                pagenos = set()

                process_pdf(rsrc, device, fp, pagenos, maxpages=0, password='',caching=caching, check_extractable=True)

                fp.close()

                device.close()

                outfp.close()

                print "have convert pdf file %s to file %s" %(in_fname, out_file)

            finally:

                self.in_queue.task_done()

#TagExtractor