_ Git - youtube-dl/blob - youtube_dl/extractor/dump.py

   1 # encoding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     ExtractorError,
   9 )
  10
  11 class DumpIE(InfoExtractor):
  12     _VALID_URL = r'^https?://(?:www\.)?dump\.com/(?P<id>[a-zA-Z0-9]+)/'
  13
  14     def _real_extract(self, url):
  15         m = re.match(self._VALID_URL, url)
  16         video_id = m.group('id')
  17
  18         # Note: There is an easier-to-parse configuration at
  19         # http://www.aparat.com/video/video/config/videohash/%video_id
  20         # but the URL in there does not work
  21
  22         webpage = self._download_webpage(url, video_id)
  23
  24         try:
  25             video_url = re.findall(r'file","(.+?.flv)"', webpage)[-1]
  26         except IndexError:
  27             raise ExtractorError(u'No video URL found')
  28
  29         thumb = re.findall('<meta property="og:image" content="(.+?)"',webpage)[0]
  30
  31         title = self._search_regex(r'<b>([^"]+)</b>', webpage, u'title')
  32
  33         return {
  34             'id': video_id,
  35             'title': title,
  36             'url': video_url,
  37             'ext': 'flv',
  38             'thumbnail': thumb,
  39         }