yt_dlp/extractor/telecinco.py

   1 import json
   2 import re
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     clean_html,
   7     int_or_none,
   8     str_or_none,
   9     try_get,
  10 )
  11
  12
  13 class TelecincoIE(InfoExtractor):
  14     IE_DESC = 'telecinco.es, cuatro.com and mediaset.es'
  15     _VALID_URL = r'https?://(?:www\.)?(?:telecinco\.es|cuatro\.com|mediaset\.es)/(?:[^/]+/)+(?P<id>.+?)\.html'
  16
  17     _TESTS = [{
  18         'url': 'http://www.telecinco.es/robinfood/temporada-01/t01xp14/Bacalao-cocochas-pil-pil_0_1876350223.html',
  19         'info_dict': {
  20             'id': '1876350223',
  21             'title': 'Bacalao con kokotxas al pil-pil',
  22             'description': 'md5:716caf5601e25c3c5ab6605b1ae71529',
  23         },
  24         'playlist': [{
  25             'md5': '7ee56d665cfd241c0e6d80fd175068b0',
  26             'info_dict': {
  27                 'id': 'JEA5ijCnF6p5W08A1rNKn7',
  28                 'ext': 'mp4',
  29                 'title': 'Con Martín Berasategui, hacer un bacalao al pil-pil es fácil y divertido',
  30                 'duration': 662,
  31             },
  32         }]
  33     }, {
  34         'url': 'http://www.cuatro.com/deportes/futbol/barcelona/Leo_Messi-Champions-Roma_2_2052780128.html',
  35         'md5': 'c86fe0d99e3bdb46b7950d38bf6ef12a',
  36         'info_dict': {
  37             'id': 'jn24Od1zGLG4XUZcnUnZB6',
  38             'ext': 'mp4',
  39             'title': '¿Quién es este ex futbolista con el que hablan Leo Messi y Luis Suárez?',
  40             'description': 'md5:a62ecb5f1934fc787107d7b9a2262805',
  41             'duration': 79,
  42         },
  43     }, {
  44         'url': 'http://www.mediaset.es/12meses/campanas/doylacara/conlatratanohaytrato/Ayudame-dar-cara-trata-trato_2_1986630220.html',
  45         'md5': 'eddb50291df704ce23c74821b995bcac',
  46         'info_dict': {
  47             'id': 'aywerkD2Sv1vGNqq9b85Q2',
  48             'ext': 'mp4',
  49             'title': '#DOYLACARA. Con la trata no hay trato',
  50             'description': 'md5:2771356ff7bfad9179c5f5cd954f1477',
  51             'duration': 50,
  52         },
  53     }, {
  54         # video in opening's content
  55         'url': 'https://www.telecinco.es/vivalavida/fiorella-sobrina-edmundo-arrocet-entrevista_18_2907195140.html',
  56         'info_dict': {
  57             'id': '2907195140',
  58             'title': 'La surrealista entrevista a la sobrina de Edmundo Arrocet: "No puedes venir aquí y tomarnos por tontos"',
  59             'description': 'md5:73f340a7320143d37ab895375b2bf13a',
  60         },
  61         'playlist': [{
  62             'md5': 'adb28c37238b675dad0f042292f209a7',
  63             'info_dict': {
  64                 'id': 'TpI2EttSDAReWpJ1o0NVh2',
  65                 'ext': 'mp4',
  66                 'title': 'La surrealista entrevista a la sobrina de Edmundo Arrocet: "No puedes venir aquí y tomarnos por tontos"',
  67                 'duration': 1015,
  68             },
  69         }],
  70         'params': {
  71             'skip_download': True,
  72         },
  73     }, {
  74         'url': 'http://www.telecinco.es/informativos/nacional/Pablo_Iglesias-Informativos_Telecinco-entrevista-Pedro_Piqueras_2_1945155182.html',
  75         'only_matching': True,
  76     }, {
  77         'url': 'http://www.telecinco.es/espanasinirmaslejos/Espana-gran-destino-turistico_2_1240605043.html',
  78         'only_matching': True,
  79     }, {
  80         'url': 'http://www.cuatro.com/chesterinlove/a-carta/chester-chester_in_love-chester_edu_2_2331030022.html',
  81         'only_matching': True,
  82     }]
  83
  84     def _parse_content(self, content, url):
  85         video_id = content['dataMediaId']
  86         config = self._download_json(
  87             content['dataConfig'], video_id, 'Downloading config JSON')
  88         title = config['info']['title']
  89         services = config['services']
  90         caronte = self._download_json(services['caronte'], video_id)
  91         stream = caronte['dls'][0]['stream']
  92         headers = self.geo_verification_headers()
  93         headers.update({
  94             'Content-Type': 'application/json;charset=UTF-8',
  95             'Origin': re.match(r'https?://[^/]+', url).group(0),
  96         })
  97         cdn = self._download_json(
  98             caronte['cerbero'], video_id, data=json.dumps({
  99                 'bbx': caronte['bbx'],
 100                 'gbx': self._download_json(services['gbx'], video_id)['gbx'],
 101             }).encode(), headers=headers)['tokens']['1']['cdn']
 102         formats = self._extract_m3u8_formats(
 103             stream + '?' + cdn, video_id, 'mp4', 'm3u8_native', m3u8_id='hls')
 104
 105         return {
 106             'id': video_id,
 107             'title': title,
 108             'formats': formats,
 109             'thumbnail': content.get('dataPoster') or config.get('poster', {}).get('imageUrl'),
 110             'duration': int_or_none(content.get('dataDuration')),
 111         }
 112
 113     def _real_extract(self, url):
 114         display_id = self._match_id(url)
 115         webpage = self._download_webpage(url, display_id)
 116         article = self._parse_json(self._search_regex(
 117             r'window\.\$REACTBASE_STATE\.article(?:_multisite)?\s*=\s*({.+})',
 118             webpage, 'article'), display_id)['article']
 119         title = article.get('title')
 120         description = clean_html(article.get('leadParagraph')) or ''
 121         if article.get('editorialType') != 'VID':
 122             entries = []
 123             body = [article.get('opening')]
 124             body.extend(try_get(article, lambda x: x['body'], list) or [])
 125             for p in body:
 126                 if not isinstance(p, dict):
 127                     continue
 128                 content = p.get('content')
 129                 if not content:
 130                     continue
 131                 type_ = p.get('type')
 132                 if type_ == 'paragraph':
 133                     content_str = str_or_none(content)
 134                     if content_str:
 135                         description += content_str
 136                     continue
 137                 if type_ == 'video' and isinstance(content, dict):
 138                     entries.append(self._parse_content(content, url))
 139             return self.playlist_result(
 140                 entries, str_or_none(article.get('id')), title, description)
 141         content = article['opening']['content']
 142         info = self._parse_content(content, url)
 143         info.update({
 144             'description': description,
 145         })
 146         return info