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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import json
   5 import re
   6
   7 from .common import InfoExtractor
   8 from .ooyala import OoyalaIE
   9 from ..utils import (
  10     clean_html,
  11     determine_ext,
  12     int_or_none,
  13     str_or_none,
  14     urljoin,
  15 )
  16
  17
  18 class TelecincoIE(InfoExtractor):
  19     IE_DESC = 'telecinco.es, cuatro.com and mediaset.es'
  20     _VALID_URL = r'https?://(?:www\.)?(?:telecinco\.es|cuatro\.com|mediaset\.es)/(?:[^/]+/)+(?P<id>.+?)\.html'
  21
  22     _TESTS = [{
  23         'url': 'http://www.telecinco.es/robinfood/temporada-01/t01xp14/Bacalao-cocochas-pil-pil_0_1876350223.html',
  24         'info_dict': {
  25             'id': '1876350223',
  26             'title': 'Bacalao con kokotxas al pil-pil',
  27             'description': 'md5:1382dacd32dd4592d478cbdca458e5bb',
  28         },
  29         'playlist': [{
  30             'md5': 'adb28c37238b675dad0f042292f209a7',
  31             'info_dict': {
  32                 'id': 'JEA5ijCnF6p5W08A1rNKn7',
  33                 'ext': 'mp4',
  34                 'title': 'Con Martín Berasategui, hacer un bacalao al pil-pil es fácil y divertido',
  35                 'duration': 662,
  36             },
  37         }]
  38     }, {
  39         'url': 'http://www.cuatro.com/deportes/futbol/barcelona/Leo_Messi-Champions-Roma_2_2052780128.html',
  40         'md5': '9468140ebc300fbb8b9d65dc6e5c4b43',
  41         'info_dict': {
  42             'id': 'jn24Od1zGLG4XUZcnUnZB6',
  43             'ext': 'mp4',
  44             'title': '¿Quién es este ex futbolista con el que hablan Leo Messi y Luis Suárez?',
  45             'description': 'md5:a62ecb5f1934fc787107d7b9a2262805',
  46             'duration': 79,
  47         },
  48     }, {
  49         'url': 'http://www.mediaset.es/12meses/campanas/doylacara/conlatratanohaytrato/Ayudame-dar-cara-trata-trato_2_1986630220.html',
  50         'md5': 'ae2dc6b7b50b2392076a51c0f70e01f6',
  51         'info_dict': {
  52             'id': 'aywerkD2Sv1vGNqq9b85Q2',
  53             'ext': 'mp4',
  54             'title': '#DOYLACARA. Con la trata no hay trato',
  55             'description': 'md5:2771356ff7bfad9179c5f5cd954f1477',
  56             'duration': 50,
  57         },
  58     }, {
  59         'url': 'http://www.telecinco.es/informativos/nacional/Pablo_Iglesias-Informativos_Telecinco-entrevista-Pedro_Piqueras_2_1945155182.html',
  60         'only_matching': True,
  61     }, {
  62         'url': 'http://www.telecinco.es/espanasinirmaslejos/Espana-gran-destino-turistico_2_1240605043.html',
  63         'only_matching': True,
  64     }, {
  65         # ooyala video
  66         'url': 'http://www.cuatro.com/chesterinlove/a-carta/chester-chester_in_love-chester_edu_2_2331030022.html',
  67         'only_matching': True,
  68     }]
  69
  70     def _parse_content(self, content, url):
  71         video_id = content['dataMediaId']
  72         if content.get('dataCmsId') == 'ooyala':
  73             return self.url_result(
  74                 'ooyala:%s' % video_id, OoyalaIE.ie_key(), video_id)
  75         config_url = urljoin(url, content['dataConfig'])
  76         config = self._download_json(
  77             config_url, video_id, 'Downloading config JSON')
  78         title = config['info']['title']
  79
  80         def mmc_url(mmc_type):
  81             return re.sub(
  82                 r'/(?:flash|html5)\.json', '/%s.json' % mmc_type,
  83                 config['services']['mmc'])
  84
  85         duration = None
  86         formats = []
  87         for mmc_type in ('flash', 'html5'):
  88             mmc = self._download_json(
  89                 mmc_url(mmc_type), video_id,
  90                 'Downloading %s mmc JSON' % mmc_type, fatal=False)
  91             if not mmc:
  92                 continue
  93             if not duration:
  94                 duration = int_or_none(mmc.get('duration'))
  95             for location in mmc['locations']:
  96                 gat = self._proto_relative_url(location.get('gat'), 'http:')
  97                 gcp = location.get('gcp')
  98                 ogn = location.get('ogn')
  99                 if None in (gat, gcp, ogn):
 100                     continue
 101                 token_data = {
 102                     'gcp': gcp,
 103                     'ogn': ogn,
 104                     'sta': 0,
 105                 }
 106                 media = self._download_json(
 107                     gat, video_id, data=json.dumps(token_data).encode('utf-8'),
 108                     headers={
 109                         'Content-Type': 'application/json;charset=utf-8',
 110                         'Referer': url,
 111                     }, fatal=False) or {}
 112                 stream = media.get('stream') or media.get('file')
 113                 if not stream:
 114                     continue
 115                 ext = determine_ext(stream)
 116                 if ext == 'f4m':
 117                     formats.extend(self._extract_f4m_formats(
 118                         stream + '&hdcore=3.2.0&plugin=aasp-3.2.0.77.18',
 119                         video_id, f4m_id='hds', fatal=False))
 120                 elif ext == 'm3u8':
 121                     formats.extend(self._extract_m3u8_formats(
 122                         stream, video_id, 'mp4', 'm3u8_native',
 123                         m3u8_id='hls', fatal=False))
 124         self._sort_formats(formats)
 125
 126         return {
 127             'id': video_id,
 128             'title': title,
 129             'formats': formats,
 130             'thumbnail': content.get('dataPoster') or config.get('poster', {}).get('imageUrl'),
 131             'duration': duration,
 132         }
 133
 134     def _real_extract(self, url):
 135         display_id = self._match_id(url)
 136         webpage = self._download_webpage(url, display_id)
 137         article = self._parse_json(self._search_regex(
 138             r'window\.\$REACTBASE_STATE\.article\s*=\s*({.+})',
 139             webpage, 'article'), display_id)['article']
 140         title = article.get('title')
 141         description = clean_html(article.get('leadParagraph'))
 142         if article.get('editorialType') != 'VID':
 143             entries = []
 144             for p in article.get('body', []):
 145                 content = p.get('content')
 146                 if p.get('type') != 'video' or not content:
 147                     continue
 148                 entries.append(self._parse_content(content, url))
 149             return self.playlist_result(
 150                 entries, str_or_none(article.get('id')), title, description)
 151         content = article['opening']['content']
 152         info = self._parse_content(content, url)
 153         info.update({
 154             'description': description,
 155         })
 156         return info