yt_dlp/extractor/openrec.py

   1 from .common import InfoExtractor
   2 from ..utils import (
   3     ExtractorError,
   4     get_first,
   5     int_or_none,
   6     traverse_obj,
   7     try_get,
   8     unified_strdate,
   9     unified_timestamp,
  10 )
  11
  12
  13 class OpenRecBaseIE(InfoExtractor):
  14     _M3U8_HEADERS = {'Referer': 'https://www.openrec.tv/'}
  15
  16     def _extract_pagestore(self, webpage, video_id):
  17         return self._parse_json(
  18             self._search_regex(r'(?m)window\.pageStore\s*=\s*(\{.+?\});$', webpage, 'window.pageStore'), video_id)
  19
  20     def _expand_media(self, video_id, media):
  21         for name, m3u8_url in (media or {}).items():
  22             if not m3u8_url:
  23                 continue
  24             yield from self._extract_m3u8_formats(
  25                 m3u8_url, video_id, ext='mp4', m3u8_id=name, headers=self._M3U8_HEADERS)
  26
  27     def _extract_movie(self, webpage, video_id, name, is_live):
  28         window_stores = self._extract_pagestore(webpage, video_id)
  29         movie_stores = [
  30             # extract all three important data (most of data are duplicated each other, but slightly different!)
  31             traverse_obj(window_stores, ('v8', 'state', 'movie'), expected_type=dict),
  32             traverse_obj(window_stores, ('v8', 'movie'), expected_type=dict),
  33             traverse_obj(window_stores, 'movieStore', expected_type=dict),
  34         ]
  35         if not any(movie_stores):
  36             raise ExtractorError(f'Failed to extract {name} info')
  37
  38         formats = list(self._expand_media(video_id, get_first(movie_stores, 'media')))
  39         if not formats:
  40             # archived livestreams or subscriber-only videos
  41             cookies = self._get_cookies('https://www.openrec.tv/')
  42             detail = self._download_json(
  43                 f'https://apiv5.openrec.tv/api/v5/movies/{video_id}/detail', video_id,
  44                 headers={
  45                     'Origin': 'https://www.openrec.tv',
  46                     'Referer': 'https://www.openrec.tv/',
  47                     'access-token': try_get(cookies, lambda x: x.get('access_token').value),
  48                     'uuid': try_get(cookies, lambda x: x.get('uuid').value),
  49                 })
  50             new_media = traverse_obj(detail, ('data', 'items', ..., 'media'), get_all=False)
  51             formats = list(self._expand_media(video_id, new_media))
  52             is_live = False
  53
  54         return {
  55             'id': video_id,
  56             'title': get_first(movie_stores, 'title'),
  57             'description': get_first(movie_stores, 'introduction'),
  58             'thumbnail': get_first(movie_stores, 'thumbnailUrl'),
  59             'formats': formats,
  60             'uploader': get_first(movie_stores, ('channel', 'user', 'name')),
  61             'uploader_id': get_first(movie_stores, ('channel', 'user', 'id')),
  62             'timestamp': int_or_none(get_first(movie_stores, ['publishedAt', 'time']), scale=1000) or unified_timestamp(get_first(movie_stores, 'publishedAt')),
  63             'is_live': is_live,
  64             'http_headers': self._M3U8_HEADERS,
  65         }
  66
  67
  68 class OpenRecIE(OpenRecBaseIE):
  69     IE_NAME = 'openrec'
  70     _VALID_URL = r'https?://(?:www\.)?openrec\.tv/live/(?P<id>[^/]+)'
  71     _TESTS = [{
  72         'url': 'https://www.openrec.tv/live/2p8v31qe4zy',
  73         'only_matching': True,
  74     }, {
  75         'url': 'https://www.openrec.tv/live/wez93eqvjzl',
  76         'only_matching': True,
  77     }]
  78
  79     def _real_extract(self, url):
  80         video_id = self._match_id(url)
  81         webpage = self._download_webpage(f'https://www.openrec.tv/live/{video_id}', video_id)
  82
  83         return self._extract_movie(webpage, video_id, 'live', True)
  84
  85
  86 class OpenRecCaptureIE(OpenRecBaseIE):
  87     IE_NAME = 'openrec:capture'
  88     _VALID_URL = r'https?://(?:www\.)?openrec\.tv/capture/(?P<id>[^/]+)'
  89     _TESTS = [{
  90         'url': 'https://www.openrec.tv/capture/l9nk2x4gn14',
  91         'only_matching': True,
  92     }, {
  93         'url': 'https://www.openrec.tv/capture/mldjr82p7qk',
  94         'info_dict': {
  95             'id': 'mldjr82p7qk',
  96             'title': 'たいじの恥ずかしい英語力',
  97             'uploader': 'たいちゃんねる',
  98             'uploader_id': 'Yaritaiji',
  99             'upload_date': '20210803',
 100         },
 101     }]
 102
 103     def _real_extract(self, url):
 104         video_id = self._match_id(url)
 105         webpage = self._download_webpage(f'https://www.openrec.tv/capture/{video_id}', video_id)
 106
 107         window_stores = self._extract_pagestore(webpage, video_id)
 108         movie_store = window_stores.get('movie')
 109
 110         capture_data = window_stores.get('capture')
 111         if not capture_data:
 112             raise ExtractorError('Cannot extract title')
 113
 114         formats = self._extract_m3u8_formats(
 115             capture_data.get('source'), video_id, ext='mp4', headers=self._M3U8_HEADERS)
 116
 117         return {
 118             'id': video_id,
 119             'title': capture_data.get('title'),
 120             'thumbnail': capture_data.get('thumbnailUrl'),
 121             'formats': formats,
 122             'timestamp': unified_timestamp(traverse_obj(movie_store, 'createdAt', expected_type=str)),
 123             'uploader': traverse_obj(movie_store, ('channel', 'name'), expected_type=str),
 124             'uploader_id': traverse_obj(movie_store, ('channel', 'id'), expected_type=str),
 125             'upload_date': unified_strdate(capture_data.get('createdAt')),
 126             'http_headers': self._M3U8_HEADERS,
 127         }
 128
 129
 130 class OpenRecMovieIE(OpenRecBaseIE):
 131     IE_NAME = 'openrec:movie'
 132     _VALID_URL = r'https?://(?:www\.)?openrec\.tv/movie/(?P<id>[^/]+)'
 133     _TESTS = [{
 134         'url': 'https://www.openrec.tv/movie/nqz5xl5km8v',
 135         'info_dict': {
 136             'id': 'nqz5xl5km8v',
 137             'title': '限定コミュニティ(Discord)参加方法ご説明動画',
 138             'description': 'md5:ebd563e5f5b060cda2f02bf26b14d87f',
 139             'thumbnail': r're:https://.+',
 140             'uploader': 'タイキとカズヒロ',
 141             'uploader_id': 'taiki_to_kazuhiro',
 142             'timestamp': 1638856800,
 143         },
 144     }]
 145
 146     def _real_extract(self, url):
 147         video_id = self._match_id(url)
 148         webpage = self._download_webpage(f'https://www.openrec.tv/movie/{video_id}', video_id)
 149
 150         return self._extract_movie(webpage, video_id, 'movie', False)