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     try_get,
  15     urljoin,
  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         'info_dict': {
  26             'id': '1876350223',
  27             'title': 'Bacalao con kokotxas al pil-pil',
  28             'description': 'md5:716caf5601e25c3c5ab6605b1ae71529',
  29         },
  30         'playlist': [{
  31             'md5': 'adb28c37238b675dad0f042292f209a7',
  32             'info_dict': {
  33                 'id': 'JEA5ijCnF6p5W08A1rNKn7',
  34                 'ext': 'mp4',
  35                 'title': 'Con Martín Berasategui, hacer un bacalao al pil-pil es fácil y divertido',
  36                 'duration': 662,
  37             },
  38         }]
  39     }, {
  40         'url': 'http://www.cuatro.com/deportes/futbol/barcelona/Leo_Messi-Champions-Roma_2_2052780128.html',
  41         'md5': '9468140ebc300fbb8b9d65dc6e5c4b43',
  42         'info_dict': {
  43             'id': 'jn24Od1zGLG4XUZcnUnZB6',
  44             'ext': 'mp4',
  45             'title': '¿Quién es este ex futbolista con el que hablan Leo Messi y Luis Suárez?',
  46             'description': 'md5:a62ecb5f1934fc787107d7b9a2262805',
  47             'duration': 79,
  48         },
  49     }, {
  50         'url': 'http://www.mediaset.es/12meses/campanas/doylacara/conlatratanohaytrato/Ayudame-dar-cara-trata-trato_2_1986630220.html',
  51         'md5': 'ae2dc6b7b50b2392076a51c0f70e01f6',
  52         'info_dict': {
  53             'id': 'aywerkD2Sv1vGNqq9b85Q2',
  54             'ext': 'mp4',
  55             'title': '#DOYLACARA. Con la trata no hay trato',
  56             'description': 'md5:2771356ff7bfad9179c5f5cd954f1477',
  57             'duration': 50,
  58         },
  59     }, {
  60         # video in opening's content
  61         'url': 'https://www.telecinco.es/vivalavida/fiorella-sobrina-edmundo-arrocet-entrevista_18_2907195140.html',
  62         'info_dict': {
  63             'id': '2907195140',
  64             'title': 'La surrealista entrevista a la sobrina de Edmundo Arrocet: "No puedes venir aquí y tomarnos por tontos"',
  65             'description': 'md5:73f340a7320143d37ab895375b2bf13a',
  66         },
  67         'playlist': [{
  68             'md5': 'adb28c37238b675dad0f042292f209a7',
  69             'info_dict': {
  70                 'id': 'TpI2EttSDAReWpJ1o0NVh2',
  71                 'ext': 'mp4',
  72                 'title': 'La surrealista entrevista a la sobrina de Edmundo Arrocet: "No puedes venir aquí y tomarnos por tontos"',
  73                 'duration': 1015,
  74             },
  75         }],
  76         'params': {
  77             'skip_download': True,
  78         },
  79     }, {
  80         'url': 'http://www.telecinco.es/informativos/nacional/Pablo_Iglesias-Informativos_Telecinco-entrevista-Pedro_Piqueras_2_1945155182.html',
  81         'only_matching': True,
  82     }, {
  83         'url': 'http://www.telecinco.es/espanasinirmaslejos/Espana-gran-destino-turistico_2_1240605043.html',
  84         'only_matching': True,
  85     }, {
  86         # ooyala video
  87         'url': 'http://www.cuatro.com/chesterinlove/a-carta/chester-chester_in_love-chester_edu_2_2331030022.html',
  88         'only_matching': True,
  89     }]
  90
  91     def _parse_content(self, content, url):
  92         video_id = content['dataMediaId']
  93         if content.get('dataCmsId') == 'ooyala':
  94             return self.url_result(
  95                 'ooyala:%s' % video_id, OoyalaIE.ie_key(), video_id)
  96         config_url = urljoin(url, content['dataConfig'])
  97         config = self._download_json(
  98             config_url, video_id, 'Downloading config JSON')
  99         title = config['info']['title']
 100
 101         def mmc_url(mmc_type):
 102             return re.sub(
 103                 r'/(?:flash|html5)\.json', '/%s.json' % mmc_type,
 104                 config['services']['mmc'])
 105
 106         duration = None
 107         formats = []
 108         for mmc_type in ('flash', 'html5'):
 109             mmc = self._download_json(
 110                 mmc_url(mmc_type), video_id,
 111                 'Downloading %s mmc JSON' % mmc_type, fatal=False)
 112             if not mmc:
 113                 continue
 114             if not duration:
 115                 duration = int_or_none(mmc.get('duration'))
 116             for location in mmc['locations']:
 117                 gat = self._proto_relative_url(location.get('gat'), 'http:')
 118                 gcp = location.get('gcp')
 119                 ogn = location.get('ogn')
 120                 if None in (gat, gcp, ogn):
 121                     continue
 122                 token_data = {
 123                     'gcp': gcp,
 124                     'ogn': ogn,
 125                     'sta': 0,
 126                 }
 127                 media = self._download_json(
 128                     gat, video_id, data=json.dumps(token_data).encode('utf-8'),
 129                     headers={
 130                         'Content-Type': 'application/json;charset=utf-8',
 131                         'Referer': url,
 132                     }, fatal=False) or {}
 133                 stream = media.get('stream') or media.get('file')
 134                 if not stream:
 135                     continue
 136                 ext = determine_ext(stream)
 137                 if ext == 'f4m':
 138                     formats.extend(self._extract_f4m_formats(
 139                         stream + '&hdcore=3.2.0&plugin=aasp-3.2.0.77.18',
 140                         video_id, f4m_id='hds', fatal=False))
 141                 elif ext == 'm3u8':
 142                     formats.extend(self._extract_m3u8_formats(
 143                         stream, video_id, 'mp4', 'm3u8_native',
 144                         m3u8_id='hls', fatal=False))
 145         self._sort_formats(formats)
 146
 147         return {
 148             'id': video_id,
 149             'title': title,
 150             'formats': formats,
 151             'thumbnail': content.get('dataPoster') or config.get('poster', {}).get('imageUrl'),
 152             'duration': duration,
 153         }
 154
 155     def _real_extract(self, url):
 156         display_id = self._match_id(url)
 157         webpage = self._download_webpage(url, display_id)
 158         article = self._parse_json(self._search_regex(
 159             r'window\.\$REACTBASE_STATE\.article(?:_multisite)?\s*=\s*({.+})',
 160             webpage, 'article'), display_id)['article']
 161         title = article.get('title')
 162         description = clean_html(article.get('leadParagraph')) or ''
 163         if article.get('editorialType') != 'VID':
 164             entries = []
 165             body = [article.get('opening')]
 166             body.extend(try_get(article, lambda x: x['body'], list) or [])
 167             for p in body:
 168                 if not isinstance(p, dict):
 169                     continue
 170                 content = p.get('content')
 171                 if not content:
 172                     continue
 173                 type_ = p.get('type')
 174                 if type_ == 'paragraph':
 175                     content_str = str_or_none(content)
 176                     if content_str:
 177                         description += content_str
 178                     continue
 179                 if type_ == 'video' and isinstance(content, dict):
 180                     entries.append(self._parse_content(content, url))
 181             return self.playlist_result(
 182                 entries, str_or_none(article.get('id')), title, description)
 183         content = article['opening']['content']
 184         info = self._parse_content(content, url)
 185         info.update({
 186             'description': description,
 187         })
 188         return info