yt_dlp/extractor/projectveritas.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     ExtractorError,
   7     traverse_obj,
   8     unified_strdate,
   9 )
  10
  11
  12 class ProjectVeritasIE(InfoExtractor):
  13     _VALID_URL = r'(?:https?://)(?:www\.)?projectveritas\.com/(?P<type>news|video)/(?P<id>[^/?#]+)'
  14     _TESTS = [{
  15         'url': 'https://www.projectveritas.com/news/exclusive-inside-the-new-york-and-new-jersey-hospitals-battling-coronavirus/',
  16         'info_dict': {
  17             'id': '51910aab-365a-5cf1-88f2-8eb1ca5fd3c6',
  18             'ext': 'mp4',
  19             'title': 'Exclusive: Inside The New York and New Jersey Hospitals Battling Coronavirus',
  20             'upload_date': '20200327',
  21             'thumbnail': 'md5:6076477fe50b03eb8708be9415e18e1c',
  22         }
  23     }, {
  24         'url': 'https://www.projectveritas.com/video/ilhan-omar-connected-ballot-harvester-in-cash-for-ballots-scheme-car-is-full/',
  25         'info_dict': {
  26             'id': 'c5aab304-a56b-54b1-9f0b-03b77bc5f2f6',
  27             'ext': 'mp4',
  28             'title': 'Ilhan Omar connected Ballot Harvester in cash-for-ballots scheme: "Car is full" of absentee ballots',
  29             'upload_date': '20200927',
  30             'thumbnail': 'md5:194b8edf0e2ba64f25500ff4378369a4',
  31         }
  32     }]
  33
  34     def _real_extract(self, url):
  35         id, type = self._match_valid_url(url).group('id', 'type')
  36         api_url = f'https://www.projectveritas.com/page-data/{type}/{id}/page-data.json'
  37         data_json = self._download_json(api_url, id)['result']['data']
  38         main_data = traverse_obj(data_json, 'video', 'post')
  39         video_id = main_data['id']
  40         thumbnail = traverse_obj(main_data, ('image', 'ogImage', 'src'))
  41         mux_asset = traverse_obj(main_data,
  42                                  'muxAsset', ('body', 'json', 'content', ..., 'data', 'target', 'fields', 'muxAsset'),
  43                                  get_all=False, expected_type=dict)
  44         if not mux_asset:
  45             raise ExtractorError('No video on the provided url.', expected=True)
  46         playback_id = traverse_obj(mux_asset, 'playbackId', ('en-US', 'playbackId'))
  47         formats = self._extract_m3u8_formats(f'https://stream.mux.com/{playback_id}.m3u8', video_id)
  48         self._sort_formats(formats)
  49         return {
  50             'id': video_id,
  51             'title': main_data['title'],
  52             'upload_date': unified_strdate(main_data.get('date')),
  53             'thumbnail': thumbnail.replace('//', ''),
  54             'formats': formats,
  55         }