yt_dlp/extractor/telemundo.py

   1 from .common import InfoExtractor
   2 from ..utils import (
   3     try_get,
   4     unified_timestamp,
   5     HEADRequest,
   6 )
   7
   8
   9 class TelemundoIE(InfoExtractor):
  10
  11     _VALID_URL = r'https?:\/\/(?:www\.)?telemundo\.com\/.+?video\/[^\/]+(?P<id>tmvo\d{7})'
  12     _TESTS = [{
  13         'url': 'https://www.telemundo.com/noticias/noticias-telemundo-en-la-noche/empleo/video/esta-aplicacion-gratuita-esta-ayudando-los-latinos-encontrar-trabajo-en-estados-unidos-tmvo9829325',
  14         'info_dict': {
  15             'id': 'tmvo9829325',
  16             'timestamp': 1621396800,
  17             'title': 'Esta aplicación gratuita está ayudando a los latinos a encontrar trabajo en Estados Unidos',
  18             'uploader': 'Telemundo',
  19             'uploader_id': 'NBCU_Telemundo',
  20             'ext': 'mp4',
  21             'upload_date': '20210519',
  22         },
  23         'params': {
  24             'skip_download': True,
  25         }
  26     }, {
  27         'url': 'https://www.telemundo.com/shows/al-rojo-vivo/empleo/video/personajes-de-times-square-piden-que-la-ciudad-de-nueva-york-los-deje-volver-trabajar-tmvo9816272',
  28         'only_matching': True,
  29     }]
  30
  31     def _real_extract(self, url):
  32         video_id = self._match_id(url)
  33         webpage = self._download_webpage(url, video_id)
  34         metadata = self._search_nextjs_data(webpage, video_id)
  35         redirect_url = try_get(
  36             metadata,
  37             lambda x: x['props']['initialState']['video']['associatedPlaylists'][0]['videos'][0]['videoAssets'][0]['publicUrl'])
  38
  39         m3u8_url = self._request_webpage(HEADRequest(
  40             redirect_url + '?format=redirect&manifest=m3u&format=redirect&Tracking=true&Embedded=true&formats=MPEG4'),
  41             video_id, 'Processing m3u8').geturl()
  42         formats = self._extract_m3u8_formats(m3u8_url, video_id, 'mp4')
  43         date = unified_timestamp(try_get(
  44             metadata, lambda x: x['props']['initialState']['video']['associatedPlaylists'][0]['videos'][0]['datePublished'].split(' ', 1)[1]))
  45         return {
  46             'url': url,
  47             'id': video_id,
  48             'title': self._search_regex(r'<h1[^>]+>([^<]+)', webpage, 'title', fatal=False),
  49             'formats': formats,
  50             'timestamp': date,
  51             'uploader': 'Telemundo',
  52             'uploader_id': self._search_regex(r'https?:\/\/(?:[^/]+\/){3}video\/(?P<id>[^\/]+)', m3u8_url, 'Akamai account', fatal=False)
  53         }