Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/ard.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     determine_ext,
   9     ExtractorError,
  10 )
  11
  12
  13 class ARDIE(InfoExtractor):
  14     _VALID_URL = r'^https?://(?:(?:www\.)?ardmediathek\.de|mediathek\.daserste\.de)/(?:.*/)(?P<video_id>[^/\?]+)(?:\?.*)?'
  15
  16     _TEST = {
  17         'url': 'http://www.ardmediathek.de/das-erste/guenther-jauch/edward-snowden-im-interview-held-oder-verraeter?documentId=19288786',
  18         'file': '19288786.mp4',
  19         'md5': '515bf47ce209fb3f5a61b7aad364634c',
  20         'info_dict': {
  21             'title': 'Edward Snowden im Interview - Held oder Verräter?',
  22             'description': 'Edward Snowden hat alles aufs Spiel gesetzt, um die weltweite \xdcberwachung durch die Geheimdienste zu enttarnen. Nun stellt sich der ehemalige NSA-Mitarbeiter erstmals weltweit in einem TV-Interview den Fragen eines NDR-Journalisten. Die Sendung vom Sonntagabend.',
  23             'thumbnail': 'http://www.ardmediathek.de/ard/servlet/contentblob/19/28/87/90/19288790/bild/2250037',
  24         },
  25         'skip': 'Blocked outside of Germany',
  26     }
  27
  28     def _real_extract(self, url):
  29         # determine video id from url
  30         m = re.match(self._VALID_URL, url)
  31
  32         numid = re.search(r'documentId=([0-9]+)', url)
  33         if numid:
  34             video_id = numid.group(1)
  35         else:
  36             video_id = m.group('video_id')
  37
  38         webpage = self._download_webpage(url, video_id)
  39
  40         title = self._html_search_regex(
  41             r'<h1(?:\s+class="boxTopHeadline")?>(.*?)</h1>', webpage, 'title')
  42         description = self._html_search_meta(
  43             'dcterms.abstract', webpage, 'description')
  44         thumbnail = self._og_search_thumbnail(webpage)
  45
  46         streams = [
  47             mo.groupdict()
  48             for mo in re.finditer(
  49                 r'mediaCollection\.addMediaStream\((?P<media_type>\d+), (?P<quality>\d+), "(?P<rtmp_url>[^"]*)", "(?P<video_url>[^"]*)", "[^"]*"\)', webpage)]
  50         if not streams:
  51             if '"fsk"' in webpage:
  52                 raise ExtractorError('This video is only available after 20:00')
  53
  54         formats = []
  55         for s in streams:
  56             format = {
  57                 'quality': int(s['quality']),
  58             }
  59             if s.get('rtmp_url'):
  60                 format['protocol'] = 'rtmp'
  61                 format['url'] = s['rtmp_url']
  62                 format['playpath'] = s['video_url']
  63             else:
  64                 format['url'] = s['video_url']
  65
  66             quality_name = self._search_regex(
  67                 r'[,.]([a-zA-Z0-9_-]+),?\.mp4', format['url'],
  68                 'quality name', default='NA')
  69             format['format_id'] = '%s-%s-%s-%s' % (
  70                 determine_ext(format['url']), quality_name, s['media_type'],
  71                 s['quality'])
  72
  73             formats.append(format)
  74
  75         self._sort_formats(formats)
  76
  77         return {
  78             'id': video_id,
  79             'title': title,
  80             'description': description,
  81             'formats': formats,
  82             'thumbnail': thumbnail,
  83         }