yt_dlp/extractor/mirrativ.py

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