yt_dlp/extractor/airtv.py

   1 from .common import InfoExtractor
   2 from .youtube import YoutubeIE
   3 from ..utils import (
   4     determine_ext,
   5     int_or_none,
   6     mimetype2ext,
   7     parse_iso8601,
   8     traverse_obj,
   9 )
  10
  11
  12 class AirTVIE(InfoExtractor):
  13     _VALID_URL = r'https?://www\.air\.tv/watch\?v=(?P<id>\w+)'
  14     _TESTS = [{
  15         # without youtube_id
  16         'url': 'https://www.air.tv/watch?v=W87jcWleSn2hXZN47zJZsQ',
  17         'info_dict': {
  18             'id': 'W87jcWleSn2hXZN47zJZsQ',
  19             'ext': 'mp4',
  20             'release_date': '20221003',
  21             'release_timestamp': 1664792603,
  22             'channel_id': 'vgfManQlRQKgoFQ8i8peFQ',
  23             'title': 'md5:c12d49ed367c3dadaa67659aff43494c',
  24             'upload_date': '20221003',
  25             'duration': 151,
  26             'view_count': int,
  27             'thumbnail': 'https://cdn-sp-gcs.air.tv/videos/W/8/W87jcWleSn2hXZN47zJZsQ/b13fc56464f47d9d62a36d110b9b5a72-4096x2160_9.jpg',
  28             'timestamp': 1664792603,
  29         },
  30     }, {
  31         # with youtube_id
  32         'url': 'https://www.air.tv/watch?v=sv57EC8tRXG6h8dNXFUU1Q',
  33         'info_dict': {
  34             'id': '2ZTqmpee-bQ',
  35             'ext': 'mp4',
  36             'comment_count': int,
  37             'tags': 'count:11',
  38             'channel_follower_count': int,
  39             'like_count': int,
  40             'uploader': 'Newsflare',
  41             'thumbnail': 'https://i.ytimg.com/vi_webp/2ZTqmpee-bQ/maxresdefault.webp',
  42             'availability': 'public',
  43             'title': 'Geese Chase Alligator Across Golf Course',
  44             'uploader_id': 'NewsflareBreaking',
  45             'channel_url': 'https://www.youtube.com/channel/UCzSSoloGEz10HALUAbYhngQ',
  46             'description': 'md5:99b21d9cea59330149efbd9706e208f5',
  47             'age_limit': 0,
  48             'channel_id': 'UCzSSoloGEz10HALUAbYhngQ',
  49             'uploader_url': 'http://www.youtube.com/user/NewsflareBreaking',
  50             'view_count': int,
  51             'categories': ['News & Politics'],
  52             'live_status': 'not_live',
  53             'playable_in_embed': True,
  54             'channel': 'Newsflare',
  55             'duration': 37,
  56             'upload_date': '20180511',
  57         },
  58     }]
  59
  60     def _get_formats_and_subtitle(self, json_data, video_id):
  61         formats, subtitles = [], {}
  62         for source in traverse_obj(json_data, 'sources', 'sources_desktop', ...):
  63             ext = determine_ext(source.get('src'), mimetype2ext(source.get('type')))
  64             if ext == 'm3u8':
  65                 fmts, subs = self._extract_m3u8_formats_and_subtitles(source.get('src'), video_id)
  66                 formats.extend(fmts)
  67                 self._merge_subtitles(subs, target=subtitles)
  68             else:
  69                 formats.append({'url': source.get('src'), 'ext': ext})
  70         return formats, subtitles
  71
  72     def _real_extract(self, url):
  73         display_id = self._match_id(url)
  74         webpage = self._download_webpage(url, display_id)
  75
  76         nextjs_json = self._search_nextjs_data(webpage, display_id)['props']['pageProps']['initialState']['videos'][display_id]
  77         if nextjs_json.get('youtube_id'):
  78             return self.url_result(
  79                 f'https://www.youtube.com/watch?v={nextjs_json.get("youtube_id")}', YoutubeIE)
  80
  81         formats, subtitles = self._get_formats_and_subtitle(nextjs_json, display_id)
  82         return {
  83             'id': display_id,
  84             'title': nextjs_json.get('title') or self._html_search_meta('og:title', webpage),
  85             'formats': formats,
  86             'subtitles': subtitles,
  87             'description': nextjs_json.get('description') or None,
  88             'duration': int_or_none(nextjs_json.get('duration')),
  89             'thumbnails': [
  90                 {'url': thumbnail}
  91                 for thumbnail in traverse_obj(nextjs_json, ('default_thumbnails', ...))],
  92             'channel_id': traverse_obj(nextjs_json, 'channel', 'channel_slug'),
  93             'timestamp': parse_iso8601(nextjs_json.get('created')),
  94             'release_timestamp': parse_iso8601(nextjs_json.get('published')),
  95             'view_count': int_or_none(nextjs_json.get('views')),
  96         }