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

   1 from __future__ import unicode_literals
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     xpath_text,
   6     parse_duration,
   7 )
   8
   9
  10 class DHMIE(InfoExtractor):
  11     IE_DESC = 'Filmarchiv - Deutsches Historisches Museum'
  12     _VALID_URL = r'http://www\.dhm\.de/filmarchiv/die-filme/(?P<id>[^/]+)'
  13
  14     _TEST = {
  15         'url': 'http://www.dhm.de/filmarchiv/die-filme/the-marshallplan-at-work-in-west-germany/',
  16         'md5': '11c475f670209bf6acca0b2b7ef51827',
  17         'info_dict': {
  18             'id': 'the-marshallplan-at-work-in-west-germany',
  19             'ext': 'flv',
  20             'title': 'MARSHALL PLAN AT WORK IN WESTERN GERMANY, THE',
  21             'description': 'md5:1fabd480c153f97b07add61c44407c82',
  22             'duration': 660,
  23             'thumbnail': 're:^https?://.*\.jpg$',
  24         }
  25     }
  26
  27     def _real_extract(self, url):
  28         video_id = self._match_id(url)
  29
  30         webpage = self._download_webpage(url, video_id)
  31
  32         playlist_url = self._search_regex(
  33             r"file\s*:\s*'([^']+)'", webpage, 'playlist url')
  34
  35         playlist = self._download_xml(playlist_url, video_id)
  36
  37         track = playlist.find(
  38             './{http://xspf.org/ns/0/}trackList/{http://xspf.org/ns/0/}track')
  39
  40         video_url = xpath_text(
  41             track, './{http://xspf.org/ns/0/}location',
  42             'video url', fatal=True)
  43         thumbnail = xpath_text(
  44             track, './{http://xspf.org/ns/0/}image',
  45             'thumbnail')
  46
  47         title = self._search_regex(
  48             [r'dc:title="([^"]+)"', r'<title> &raquo;([^<]+)</title>'],
  49             webpage, 'title').strip()
  50         description = self._html_search_regex(
  51             r'<p><strong>Description:</strong>(.+?)</p>',
  52             webpage, 'description', fatal=False)
  53         duration = parse_duration(self._search_regex(
  54             r'<em>Length\s*</em>\s*:\s*</strong>([^<]+)',
  55             webpage, 'duration', fatal=False))
  56
  57         return {
  58             'id': video_id,
  59             'url': video_url,
  60             'title': title,
  61             'description': description,
  62             'duration': duration,
  63             'thumbnail': thumbnail,
  64         }