yt_dlp/extractor/mirrativ.py

   1 from .common import InfoExtractor
   2 from ..utils import (
   3     ExtractorError,
   4     dict_get,
   5     traverse_obj,
   6     try_get,
   7 )
   8
   9
  10 class MirrativBaseIE(InfoExtractor):
  11     def assert_error(self, response):
  12         error_message = traverse_obj(response, ('status', 'error'))
  13         if error_message:
  14             raise ExtractorError(f'Mirrativ says: {error_message}', expected=True)
  15
  16
  17 class MirrativIE(MirrativBaseIE):
  18     IE_NAME = 'mirrativ'
  19     _VALID_URL = r'https?://(?:www\.)?mirrativ\.com/live/(?P<id>[^/?#&]+)'
  20
  21     TESTS = [{
  22         'url': 'https://mirrativ.com/live/UQomuS7EMgHoxRHjEhNiHw',
  23         'info_dict': {
  24             'id': 'UQomuS7EMgHoxRHjEhNiHw',
  25             'title': 'ねむいぃ、。『参加型』🔰jcが初めてやるCOD✨初見さん大歓迎💗',
  26             'is_live': True,
  27             'description': 'md5:bfcd8f77f2fab24c3c672e5620f3f16e',
  28             'thumbnail': r're:https?://.+',
  29             'uploader': '# あ ち ゅ 。💡',
  30             'uploader_id': '118572165',
  31             'duration': None,
  32             'view_count': 1241,
  33             'release_timestamp': 1646229192,
  34             'timestamp': 1646229167,
  35             'was_live': False,
  36         },
  37         'skip': 'livestream',
  38     }, {
  39         'url': 'https://mirrativ.com/live/POxyuG1KmW2982lqlDTuPw',
  40         'only_matching': True,
  41     }]
  42
  43     def _real_extract(self, url):
  44         video_id = self._match_id(url)
  45         webpage = self._download_webpage(f'https://www.mirrativ.com/live/{video_id}', video_id)
  46         live_response = self._download_json(f'https://www.mirrativ.com/api/live/live?live_id={video_id}', video_id)
  47         self.assert_error(live_response)
  48
  49         hls_url = dict_get(live_response, ('archive_url_hls', 'streaming_url_hls'))
  50         is_live = bool(live_response.get('is_live'))
  51         if not hls_url:
  52             raise ExtractorError('Neither archive nor live is available.', expected=True)
  53
  54         formats = self._extract_m3u8_formats(
  55             hls_url, video_id,
  56             ext='mp4', entry_protocol='m3u8_native',
  57             m3u8_id='hls', live=is_live)
  58
  59         return {
  60             'id': video_id,
  61             'title': self._og_search_title(webpage, default=None) or self._search_regex(
  62                 r'<title>\s*(.+?) - Mirrativ\s*</title>', webpage) or live_response.get('title'),
  63             'is_live': is_live,
  64             'description': live_response.get('description'),
  65             'formats': formats,
  66             'thumbnail': live_response.get('image_url'),
  67             'uploader': traverse_obj(live_response, ('owner', 'name')),
  68             'uploader_id': traverse_obj(live_response, ('owner', 'user_id')),
  69             'duration': try_get(live_response, lambda x: x['ended_at'] - x['started_at']) if not is_live else None,
  70             'view_count': live_response.get('total_viewer_num'),
  71             'release_timestamp': live_response.get('started_at'),
  72             'timestamp': live_response.get('created_at'),
  73             'was_live': bool(live_response.get('is_archive')),
  74         }
  75
  76
  77 class MirrativUserIE(MirrativBaseIE):
  78     IE_NAME = 'mirrativ:user'
  79     _VALID_URL = r'https?://(?:www\.)?mirrativ\.com/user/(?P<id>\d+)'
  80
  81     _TESTS = [{
  82         # Live archive is available up to 3 days
  83         # see: https://helpfeel.com/mirrativ/%E9%8C%B2%E7%94%BB-5e26d3ad7b59ef0017fb49ac (Japanese)
  84         'url': 'https://www.mirrativ.com/user/110943130',
  85         'note': 'multiple archives available',
  86         'only_matching': True,
  87     }]
  88
  89     def _entries(self, user_id):
  90         page = 1
  91         while page is not None:
  92             api_response = self._download_json(
  93                 f'https://www.mirrativ.com/api/live/live_history?user_id={user_id}&page={page}', user_id,
  94                 note=f'Downloading page {page}')
  95             self.assert_error(api_response)
  96             lives = api_response.get('lives')
  97             if not lives:
  98                 break
  99             for live in lives:
 100                 if not live.get('is_archive') and not live.get('is_live'):
 101                     # neither archive nor live is available, so skip it
 102                     # or the service will ban your IP address for a while
 103                     continue
 104                 live_id = live.get('live_id')
 105                 url = f'https://www.mirrativ.com/live/{live_id}'
 106                 yield self.url_result(url, video_id=live_id, video_title=live.get('title'))
 107             page = api_response.get('next_page')
 108
 109     def _real_extract(self, url):
 110         user_id = self._match_id(url)
 111         user_info = self._download_json(
 112             f'https://www.mirrativ.com/api/user/profile?user_id={user_id}', user_id,
 113             note='Downloading user info', fatal=False)
 114         self.assert_error(user_info)
 115
 116         return self.playlist_result(
 117             self._entries(user_id), user_id,
 118             user_info.get('name'), user_info.get('description'))