youtube_dlc/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 ..utils import (
   9     clean_html,
  10     int_or_none,
  11     str_or_none,
  12     try_get,
  13 )
  14
  15
  16 class TelecincoIE(InfoExtractor):
  17     IE_DESC = 'telecinco.es, cuatro.com and mediaset.es'
  18     _VALID_URL = r'https?://(?:www\.)?(?:telecinco\.es|cuatro\.com|mediaset\.es)/(?:[^/]+/)+(?P<id>.+?)\.html'
  19
  20     _TESTS = [{
  21         'url': 'http://www.telecinco.es/robinfood/temporada-01/t01xp14/Bacalao-cocochas-pil-pil_0_1876350223.html',
  22         'info_dict': {
  23             'id': '1876350223',
  24             'title': 'Bacalao con kokotxas al pil-pil',
  25             'description': 'md5:716caf5601e25c3c5ab6605b1ae71529',
  26         },
  27         'playlist': [{
  28             'md5': '7ee56d665cfd241c0e6d80fd175068b0',
  29             'info_dict': {
  30                 'id': 'JEA5ijCnF6p5W08A1rNKn7',
  31                 'ext': 'mp4',
  32                 'title': 'Con Martín Berasategui, hacer un bacalao al pil-pil es fácil y divertido',
  33                 'duration': 662,
  34             },
  35         }]
  36     }, {
  37         'url': 'http://www.cuatro.com/deportes/futbol/barcelona/Leo_Messi-Champions-Roma_2_2052780128.html',
  38         'md5': 'c86fe0d99e3bdb46b7950d38bf6ef12a',
  39         'info_dict': {
  40             'id': 'jn24Od1zGLG4XUZcnUnZB6',
  41             'ext': 'mp4',
  42             'title': '¿Quién es este ex futbolista con el que hablan Leo Messi y Luis Suárez?',
  43             'description': 'md5:a62ecb5f1934fc787107d7b9a2262805',
  44             'duration': 79,
  45         },
  46     }, {
  47         'url': 'http://www.mediaset.es/12meses/campanas/doylacara/conlatratanohaytrato/Ayudame-dar-cara-trata-trato_2_1986630220.html',
  48         'md5': 'eddb50291df704ce23c74821b995bcac',
  49         'info_dict': {
  50             'id': 'aywerkD2Sv1vGNqq9b85Q2',
  51             'ext': 'mp4',
  52             'title': '#DOYLACARA. Con la trata no hay trato',
  53             'description': 'md5:2771356ff7bfad9179c5f5cd954f1477',
  54             'duration': 50,
  55         },
  56     }, {
  57         # video in opening's content
  58         'url': 'https://www.telecinco.es/vivalavida/fiorella-sobrina-edmundo-arrocet-entrevista_18_2907195140.html',
  59         'info_dict': {
  60             'id': '2907195140',
  61             'title': 'La surrealista entrevista a la sobrina de Edmundo Arrocet: "No puedes venir aquí y tomarnos por tontos"',
  62             'description': 'md5:73f340a7320143d37ab895375b2bf13a',
  63         },
  64         'playlist': [{
  65             'md5': 'adb28c37238b675dad0f042292f209a7',
  66             'info_dict': {
  67                 'id': 'TpI2EttSDAReWpJ1o0NVh2',
  68                 'ext': 'mp4',
  69                 'title': 'La surrealista entrevista a la sobrina de Edmundo Arrocet: "No puedes venir aquí y tomarnos por tontos"',
  70                 'duration': 1015,
  71             },
  72         }],
  73         'params': {
  74             'skip_download': True,
  75         },
  76     }, {
  77         'url': 'http://www.telecinco.es/informativos/nacional/Pablo_Iglesias-Informativos_Telecinco-entrevista-Pedro_Piqueras_2_1945155182.html',
  78         'only_matching': True,
  79     }, {
  80         'url': 'http://www.telecinco.es/espanasinirmaslejos/Espana-gran-destino-turistico_2_1240605043.html',
  81         'only_matching': True,
  82     }, {
  83         # ooyala video
  84         'url': 'http://www.cuatro.com/chesterinlove/a-carta/chester-chester_in_love-chester_edu_2_2331030022.html',
  85         'only_matching': True,
  86     }]
  87
  88     def _parse_content(self, content, url):
  89         video_id = content['dataMediaId']
  90         config = self._download_json(
  91             content['dataConfig'], video_id, 'Downloading config JSON')
  92         title = config['info']['title']
  93         services = config['services']
  94         caronte = self._download_json(services['caronte'], video_id)
  95         stream = caronte['dls'][0]['stream']
  96         headers = self.geo_verification_headers()
  97         headers.update({
  98             'Content-Type': 'application/json;charset=UTF-8',
  99             'Origin': re.match(r'https?://[^/]+', url).group(0),
 100         })
 101         cdn = self._download_json(
 102             caronte['cerbero'], video_id, data=json.dumps({
 103                 'bbx': caronte['bbx'],
 104                 'gbx': self._download_json(services['gbx'], video_id)['gbx'],
 105             }).encode(), headers=headers)['tokens']['1']['cdn']
 106         formats = self._extract_m3u8_formats(
 107             stream + '?' + cdn, video_id, 'mp4', 'm3u8_native', m3u8_id='hls')
 108         self._sort_formats(formats)
 109
 110         return {
 111             'id': video_id,
 112             'title': title,
 113             'formats': formats,
 114             'thumbnail': content.get('dataPoster') or config.get('poster', {}).get('imageUrl'),
 115             'duration': int_or_none(content.get('dataDuration')),
 116         }
 117
 118     def _real_extract(self, url):
 119         display_id = self._match_id(url)
 120         webpage = self._download_webpage(url, display_id)
 121         article = self._parse_json(self._search_regex(
 122             r'window\.\$REACTBASE_STATE\.article(?:_multisite)?\s*=\s*({.+})',
 123             webpage, 'article'), display_id)['article']
 124         title = article.get('title')
 125         description = clean_html(article.get('leadParagraph')) or ''
 126         if article.get('editorialType') != 'VID':
 127             entries = []
 128             body = [article.get('opening')]
 129             body.extend(try_get(article, lambda x: x['body'], list) or [])
 130             for p in body:
 131                 if not isinstance(p, dict):
 132                     continue
 133                 content = p.get('content')
 134                 if not content:
 135                     continue
 136                 type_ = p.get('type')
 137                 if type_ == 'paragraph':
 138                     content_str = str_or_none(content)
 139                     if content_str:
 140                         description += content_str
 141                     continue
 142                 if type_ == 'video' and isinstance(content, dict):
 143                     entries.append(self._parse_content(content, url))
 144             return self.playlist_result(
 145                 entries, str_or_none(article.get('id')), title, description)
 146         content = article['opening']['content']
 147         info = self._parse_content(content, url)
 148         info.update({
 149             'description': description,
 150         })
 151         return info