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