youtube_dl/extractor/zdf.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     determine_ext,
   6     ExtractorError,
   7 )
   8
   9
  10 class ZDFIE(InfoExtractor):
  11     _VALID_URL = r'^http://www\.zdf\.de\/ZDFmediathek(?P<hash>#)?\/(.*beitrag\/video\/)(?P<video_id>[^/\?]+)(?:\?.*)?'
  12     _MEDIA_STREAM = r'<a href="(?P<video_url>.+(?P<media_type>.streaming).+/zdf/(?P<quality>[^\/]+)/[^"]*)".+class="play".+>'
  13
  14     def _real_extract(self, url):
  15         mobj = re.match(self._VALID_URL, url)
  16         if mobj is None:
  17             raise ExtractorError(u'Invalid URL: %s' % url)
  18         video_id = mobj.group('video_id')
  19
  20         if mobj.group('hash'):
  21             url = url.replace(u'#', u'', 1)
  22
  23         html = self._download_webpage(url, video_id)
  24         streams = [m.groupdict() for m in re.finditer(self._MEDIA_STREAM, html)]
  25         if streams is None:
  26             raise ExtractorError(u'No media url found.')
  27
  28         # s['media_type'] == 'wstreaming' -> use 'Windows Media Player' and mms url
  29         # s['media_type'] == 'hstreaming' -> use 'Quicktime' and rtsp url
  30         # choose first/default media type and highest quality for now
  31         def stream_pref(s):
  32             TYPE_ORDER = ['ostreaming', 'hstreaming', 'wstreaming']
  33             try:
  34                 type_pref = TYPE_ORDER.index(s['media_type'])
  35             except ValueError:
  36                 type_pref = 999
  37
  38             QUALITY_ORDER = ['veryhigh', '300']
  39             try:
  40                 quality_pref = QUALITY_ORDER.index(s['quality'])
  41             except ValueError:
  42                 quality_pref = 999
  43
  44             return (type_pref, quality_pref)
  45
  46         sorted_streams = sorted(streams, key=stream_pref)
  47         if not sorted_streams:
  48             raise ExtractorError(u'No stream found.')
  49         stream = sorted_streams[0]
  50
  51         media_link = self._download_webpage(
  52             stream['video_url'],
  53             video_id,
  54             u'Get stream URL')
  55
  56         #MMS_STREAM = r'href="(?P<video_url>mms://[^"]*)"'
  57         RTSP_STREAM = r'(?P<video_url>rtsp://[^"]*.mp4)'
  58
  59         mobj = re.search(self._MEDIA_STREAM, media_link)
  60         if mobj is None:
  61             mobj = re.search(RTSP_STREAM, media_link)
  62             if mobj is None:
  63                 raise ExtractorError(u'Cannot extract mms:// or rtsp:// URL')
  64         video_url = mobj.group('video_url')
  65
  66         title = self._html_search_regex(
  67             r'<h1(?: class="beitragHeadline")?>(.*?)</h1>',
  68             html, u'title')
  69
  70         return {
  71             'id': video_id,
  72             'url': video_url,
  73             'title': title,
  74             'ext': determine_ext(video_url)
  75         }