yt_dlp/extractor/telecinco.py

   1 import json
   2 import re
   3
   4 from .common import InfoExtractor
   5 from ..networking.exceptions import HTTPError
   6 from ..utils import (
   7     ExtractorError,
   8     clean_html,
   9     int_or_none,
  10     join_nonempty,
  11     str_or_none,
  12     traverse_obj,
  13     update_url,
  14     url_or_none,
  15 )
  16
  17
  18 class TelecincoBaseIE(InfoExtractor):
  19     def _parse_content(self, content, url):
  20         video_id = content['dataMediaId']
  21         config = self._download_json(
  22             content['dataConfig'], video_id, 'Downloading config JSON')
  23         services = config['services']
  24         caronte = self._download_json(services['caronte'], video_id)
  25         if traverse_obj(caronte, ('dls', 0, 'drm', {bool})):
  26             self.report_drm(video_id)
  27
  28         stream = caronte['dls'][0]['stream']
  29         headers = {
  30             'Referer': url,
  31             'Origin': re.match(r'https?://[^/]+', url).group(0),
  32         }
  33         geo_headers = {**headers, **self.geo_verification_headers()}
  34
  35         try:
  36             cdn = self._download_json(
  37                 caronte['cerbero'], video_id, data=json.dumps({
  38                     'bbx': caronte['bbx'],
  39                     'gbx': self._download_json(services['gbx'], video_id)['gbx'],
  40                 }).encode(), headers={
  41                     'Content-Type': 'application/json',
  42                     **geo_headers,
  43                 })['tokens']['1']['cdn']
  44         except ExtractorError as error:
  45             if isinstance(error.cause, HTTPError) and error.cause.status == 403:
  46                 error_code = traverse_obj(
  47                     self._webpage_read_content(error.cause.response, caronte['cerbero'], video_id, fatal=False),
  48                     ({json.loads}, 'code', {int}))
  49                 if error_code == 4038:
  50                     self.raise_geo_restricted(countries=['ES'])
  51             raise
  52
  53         formats = self._extract_m3u8_formats(
  54             update_url(stream, query=cdn), video_id, 'mp4', m3u8_id='hls', headers=geo_headers)
  55
  56         return {
  57             'id': video_id,
  58             'title': traverse_obj(config, ('info', 'title', {str})),
  59             'formats': formats,
  60             'thumbnail': (traverse_obj(content, ('dataPoster', {url_or_none}))
  61                           or traverse_obj(config, 'poster', 'imageUrl', expected_type=url_or_none)),
  62             'duration': traverse_obj(content, ('dataDuration', {int_or_none})),
  63             'http_headers': headers,
  64         }
  65
  66
  67 class TelecincoIE(TelecincoBaseIE):
  68     IE_DESC = 'telecinco.es, cuatro.com and mediaset.es'
  69     _VALID_URL = r'https?://(?:www\.)?(?:telecinco\.es|cuatro\.com|mediaset\.es)/(?:[^/]+/)+(?P<id>.+?)\.html'
  70
  71     _TESTS = [{
  72         'url': 'http://www.telecinco.es/robinfood/temporada-01/t01xp14/Bacalao-cocochas-pil-pil_0_1876350223.html',
  73         'info_dict': {
  74             'id': '1876350223',
  75             'title': 'Bacalao con kokotxas al pil-pil',
  76             'description': 'md5:716caf5601e25c3c5ab6605b1ae71529',
  77         },
  78         'playlist': [{
  79             'md5': '7ee56d665cfd241c0e6d80fd175068b0',
  80             'info_dict': {
  81                 'id': 'JEA5ijCnF6p5W08A1rNKn7',
  82                 'ext': 'mp4',
  83                 'title': 'Con Martín Berasategui, hacer un bacalao al pil-pil es fácil y divertido',
  84                 'duration': 662,
  85             },
  86         }],
  87         'skip': 'HTTP Error 410 Gone',
  88     }, {
  89         'url': 'http://www.cuatro.com/deportes/futbol/barcelona/Leo_Messi-Champions-Roma_2_2052780128.html',
  90         'md5': 'c86fe0d99e3bdb46b7950d38bf6ef12a',
  91         'info_dict': {
  92             'id': 'jn24Od1zGLG4XUZcnUnZB6',
  93             'ext': 'mp4',
  94             'title': '¿Quién es este ex futbolista con el que hablan Leo Messi y Luis Suárez?',
  95             'description': 'md5:a62ecb5f1934fc787107d7b9a2262805',
  96             'duration': 79,
  97         },
  98         'skip': 'Redirects to main page',
  99     }, {
 100         'url': 'http://www.mediaset.es/12meses/campanas/doylacara/conlatratanohaytrato/Ayudame-dar-cara-trata-trato_2_1986630220.html',
 101         'md5': '5ce057f43f30b634fbaf0f18c71a140a',
 102         'info_dict': {
 103             'id': 'aywerkD2Sv1vGNqq9b85Q2',
 104             'ext': 'mp4',
 105             'title': '#DOYLACARA. Con la trata no hay trato',
 106             'duration': 50,
 107             'thumbnail': 'https://album.mediaset.es/eimg/2017/11/02/1tlQLO5Q3mtKT24f3EaC24.jpg',
 108         },
 109     }, {
 110         # video in opening's content
 111         'url': 'https://www.telecinco.es/vivalavida/fiorella-sobrina-edmundo-arrocet-entrevista_18_2907195140.html',
 112         'info_dict': {
 113             'id': '1691427',
 114             'title': 'La surrealista entrevista a la sobrina de Edmundo Arrocet: "No puedes venir aquí y tomarnos por tontos"',
 115             'description': r're:Fiorella, la sobrina de Edmundo Arrocet, concedió .{727}',
 116         },
 117         'playlist': [{
 118             'md5': 'adb28c37238b675dad0f042292f209a7',
 119             'info_dict': {
 120                 'id': 'TpI2EttSDAReWpJ1o0NVh2',
 121                 'ext': 'mp4',
 122                 'title': 'La surrealista entrevista a la sobrina de Edmundo Arrocet: "No puedes venir aquí y tomarnos por tontos"',
 123                 'duration': 1015,
 124                 'thumbnail': 'https://album.mediaset.es/eimg/2020/02/29/5opaC37lUhKlZ7FoDhiVC.jpg',
 125             },
 126         }],
 127         'params': {
 128             'skip_download': True,
 129         },
 130     }, {
 131         'url': 'http://www.telecinco.es/informativos/nacional/Pablo_Iglesias-Informativos_Telecinco-entrevista-Pedro_Piqueras_2_1945155182.html',
 132         'only_matching': True,
 133     }, {
 134         'url': 'http://www.telecinco.es/espanasinirmaslejos/Espana-gran-destino-turistico_2_1240605043.html',
 135         'only_matching': True,
 136     }, {
 137         'url': 'http://www.cuatro.com/chesterinlove/a-carta/chester-chester_in_love-chester_edu_2_2331030022.html',
 138         'only_matching': True,
 139     }]
 140
 141     def _real_extract(self, url):
 142         display_id = self._match_id(url)
 143         webpage = self._download_webpage(url, display_id)
 144         article = self._search_json(
 145             r'window\.\$REACTBASE_STATE\.article(?:_multisite)?\s*=',
 146             webpage, 'article', display_id)['article']
 147         description = traverse_obj(article, ('leadParagraph', {clean_html}, filter))
 148
 149         if article.get('editorialType') != 'VID':
 150             entries = []
 151
 152             for p in traverse_obj(article, ((('opening', all), 'body'), lambda _, v: v['content'])):
 153                 content = p['content']
 154                 type_ = p.get('type')
 155                 if type_ == 'paragraph' and isinstance(content, str):
 156                     description = join_nonempty(description, content, delim='')
 157                 elif type_ == 'video' and isinstance(content, dict):
 158                     entries.append(self._parse_content(content, url))
 159
 160             return self.playlist_result(
 161                 entries, str_or_none(article.get('id')),
 162                 traverse_obj(article, ('title', {str})), clean_html(description))
 163
 164         info = self._parse_content(article['opening']['content'], url)
 165         info['description'] = description
 166         return info