]> Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/telecinco.py
Imported Upstream version 2016.06.25
[youtubedl] / youtube_dl / extractor / telecinco.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 import json
5
6 from .common import InfoExtractor
7 from ..compat import (
8 compat_urllib_parse_unquote,
9 compat_urllib_parse_urlencode,
10 compat_urlparse,
11 )
12 from ..utils import (
13 get_element_by_attribute,
14 parse_duration,
15 strip_jsonp,
16 )
17
18
19 class TelecincoIE(InfoExtractor):
20 IE_DESC = 'telecinco.es, cuatro.com and mediaset.es'
21 _VALID_URL = r'https?://www\.(?:telecinco\.es|cuatro\.com|mediaset\.es)/(?:[^/]+/)+(?P<id>.+?)\.html'
22
23 _TESTS = [{
24 'url': 'http://www.telecinco.es/robinfood/temporada-01/t01xp14/Bacalao-cocochas-pil-pil_0_1876350223.html',
25 'md5': '5cbef3ad5ef17bf0d21570332d140729',
26 'info_dict': {
27 'id': 'MDSVID20141015_0058',
28 'ext': 'mp4',
29 'title': 'Con Martín Berasategui, hacer un bacalao al ...',
30 'duration': 662,
31 },
32 }, {
33 'url': 'http://www.cuatro.com/deportes/futbol/barcelona/Leo_Messi-Champions-Roma_2_2052780128.html',
34 'md5': '0a5b9f3cc8b074f50a0578f823a12694',
35 'info_dict': {
36 'id': 'MDSVID20150916_0128',
37 'ext': 'mp4',
38 'title': '¿Quién es este ex futbolista con el que hablan ...',
39 'duration': 79,
40 },
41 }, {
42 'url': 'http://www.mediaset.es/12meses/campanas/doylacara/conlatratanohaytrato/Ayudame-dar-cara-trata-trato_2_1986630220.html',
43 'md5': 'ad1bfaaba922dd4a295724b05b68f86a',
44 'info_dict': {
45 'id': 'MDSVID20150513_0220',
46 'ext': 'mp4',
47 'title': '#DOYLACARA. Con la trata no hay trato',
48 'duration': 50,
49 },
50 }, {
51 'url': 'http://www.telecinco.es/informativos/nacional/Pablo_Iglesias-Informativos_Telecinco-entrevista-Pedro_Piqueras_2_1945155182.html',
52 'only_matching': True,
53 }, {
54 'url': 'http://www.telecinco.es/espanasinirmaslejos/Espana-gran-destino-turistico_2_1240605043.html',
55 'only_matching': True,
56 }]
57
58 def _real_extract(self, url):
59 episode = self._match_id(url)
60 webpage = self._download_webpage(url, episode)
61 embed_data_json = self._search_regex(
62 r'(?s)MSV\.embedData\[.*?\]\s*=\s*({.*?});', webpage, 'embed data',
63 ).replace('\'', '"')
64 embed_data = json.loads(embed_data_json)
65
66 domain = embed_data['mediaUrl']
67 if not domain.startswith('http'):
68 # only happens in telecinco.es videos
69 domain = 'http://' + domain
70 info_url = compat_urlparse.urljoin(
71 domain,
72 compat_urllib_parse_unquote(embed_data['flashvars']['host'])
73 )
74 info_el = self._download_xml(info_url, episode).find('./video/info')
75
76 video_link = info_el.find('videoUrl/link').text
77 token_query = compat_urllib_parse_urlencode({'id': video_link})
78 token_info = self._download_json(
79 embed_data['flashvars']['ov_tk'] + '?' + token_query,
80 episode,
81 transform_source=strip_jsonp
82 )
83 formats = self._extract_m3u8_formats(
84 token_info['tokenizedUrl'], episode, ext='mp4', entry_protocol='m3u8_native')
85 self._sort_formats(formats)
86
87 return {
88 'id': embed_data['videoId'],
89 'display_id': episode,
90 'title': info_el.find('title').text,
91 'formats': formats,
92 'description': get_element_by_attribute('class', 'text', webpage),
93 'thumbnail': info_el.find('thumb').text,
94 'duration': parse_duration(info_el.find('duration').text),
95 }