2 from __future__ import unicode_literals
6 from .common import InfoExtractor
9 compat_urllib_parse_unquote,
13 get_element_by_attribute,
19 class TelecincoIE(InfoExtractor):
20 IE_DESC = 'telecinco.es and cuatro.es'
21 _VALID_URL = r'https?://www\.(?:telecinco\.es|cuatro\.com)/(?:[^/]+/)+(?P<id>.+?)\.html'
24 'url': 'http://www.telecinco.es/robinfood/temporada-01/t01xp14/Bacalao-cocochas-pil-pil_0_1876350223.html',
25 'md5': '5cbef3ad5ef17bf0d21570332d140729',
27 'id': 'MDSVID20141015_0058',
29 'title': 'Con Martín Berasategui, hacer un bacalao al ...',
33 'url': 'http://www.cuatro.com/deportes/futbol/barcelona/Leo_Messi-Champions-Roma_2_2052780128.html',
34 'md5': '0a5b9f3cc8b074f50a0578f823a12694',
36 'id': 'MDSVID20150916_0128',
38 'title': '¿Quién es este ex futbolista con el que hablan ...',
42 'url': 'http://www.telecinco.es/informativos/nacional/Pablo_Iglesias-Informativos_Telecinco-entrevista-Pedro_Piqueras_2_1945155182.html',
43 'only_matching': True,
45 'url': 'http://www.telecinco.es/espanasinirmaslejos/Espana-gran-destino-turistico_2_1240605043.html',
46 'only_matching': True,
49 def _real_extract(self, url):
50 episode = self._match_id(url)
51 webpage = self._download_webpage(url, episode)
52 embed_data_json = self._search_regex(
53 r'(?s)MSV\.embedData\[.*?\]\s*=\s*({.*?});', webpage, 'embed data',
55 embed_data = json.loads(embed_data_json)
57 domain = embed_data['mediaUrl']
58 if not domain.startswith('http'):
59 # only happens in telecinco.es videos
60 domain = 'http://' + domain
61 info_url = compat_urlparse.urljoin(
63 compat_urllib_parse_unquote(embed_data['flashvars']['host'])
65 info_el = self._download_xml(info_url, episode).find('./video/info')
67 video_link = info_el.find('videoUrl/link').text
68 token_query = compat_urllib_parse.urlencode({'id': video_link})
69 token_info = self._download_json(
70 embed_data['flashvars']['ov_tk'] + '?' + token_query,
72 transform_source=strip_jsonp
74 formats = self._extract_m3u8_formats(
75 token_info['tokenizedUrl'], episode, ext='mp4', entry_protocol='m3u8_native')
78 'id': embed_data['videoId'],
79 'display_id': episode,
80 'title': info_el.find('title').text,
82 'description': get_element_by_attribute('class', 'text', webpage),
83 'thumbnail': info_el.find('thumb').text,
84 'duration': parse_duration(info_el.find('duration').text),