yt_dlp/extractor/telemundo.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     try_get,
   7     unified_timestamp,
   8     HEADRequest,
   9 )
  10
  11
  12 class TelemundoIE(InfoExtractor):
  13
  14     _VALID_URL = r'https?:\/\/(?:www\.)?telemundo\.com\/.+?video\/[^\/]+(?P<id>tmvo\d{7})'
  15     _TESTS = [{
  16         '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',
  17         'info_dict': {
  18             'id': 'tmvo9829325',
  19             'timestamp': 1621396800,
  20             'title': 'Esta aplicación gratuita está ayudando a los latinos a encontrar trabajo en Estados Unidos',
  21             'uploader': 'Telemundo',
  22             'uploader_id': 'NBCU_Telemundo',
  23             'ext': 'mp4',
  24             'upload_date': '20210519',
  25         },
  26         'params': {
  27             'skip_download': True,
  28         }
  29     }, {
  30         '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',
  31         'only_matching': True,
  32     }]
  33
  34     def _real_extract(self, url):
  35         video_id = self._match_id(url)
  36         webpage = self._download_webpage(url, video_id)
  37         metadata = self._parse_json(
  38             self._search_regex(r'<[^>]+id="__NEXT_DATA__"[^>]+>([^<]+)', webpage, 'JSON metadata'), video_id)
  39         redirect_url = try_get(
  40             metadata,
  41             lambda x: x['props']['initialState']['video']['associatedPlaylists'][0]['videos'][0]['videoAssets'][0]['publicUrl'])
  42
  43         m3u8_url = self._request_webpage(HEADRequest(
  44             redirect_url + '?format=redirect&manifest=m3u&format=redirect&Tracking=true&Embedded=true&formats=MPEG4'),
  45             video_id, 'Processing m3u8').geturl()
  46         formats = self._extract_m3u8_formats(m3u8_url, video_id, 'mp4')
  47         self._sort_formats(formats)
  48         date = unified_timestamp(try_get(
  49             metadata, lambda x: x['props']['initialState']['video']['associatedPlaylists'][0]['videos'][0]['datePublished'].split(' ', 1)[1]))
  50         return {
  51             'url': url,
  52             'id': video_id,
  53             'title': self._search_regex(r'<h1[^>]+>([^<]+)', webpage, 'title', fatal=False),
  54             'formats': formats,
  55             'timestamp': date,
  56             'uploader': 'Telemundo',
  57             'uploader_id': self._search_regex(r'https?:\/\/(?:[^/]+\/){3}video\/(?P<id>[^\/]+)', m3u8_url, 'Akamai account', fatal=False)
  58         }