yt_dlp/extractor/mixch.py

   1 from .common import InfoExtractor
   2 from ..utils import UserNotLive, traverse_obj
   3
   4
   5 class MixchIE(InfoExtractor):
   6     IE_NAME = 'mixch'
   7     _VALID_URL = r'https?://(?:www\.)?mixch\.tv/u/(?P<id>\d+)'
   8
   9     _TESTS = [{
  10         'url': 'https://mixch.tv/u/16236849/live',
  11         'skip': 'don\'t know if this live persists',
  12         'info_dict': {
  13             'id': '16236849',
  14             'title': '24配信シェア⭕️投票🙏💦',
  15             'comment_count': 13145,
  16             'view_count': 28348,
  17             'timestamp': 1636189377,
  18             'uploader': '🦥伊咲👶🏻#フレアワ',
  19             'uploader_id': '16236849',
  20         }
  21     }, {
  22         'url': 'https://mixch.tv/u/16137876/live',
  23         'only_matching': True,
  24     }]
  25
  26     def _real_extract(self, url):
  27         video_id = self._match_id(url)
  28         webpage = self._download_webpage(f'https://mixch.tv/u/{video_id}/live', video_id)
  29
  30         initial_js_state = self._parse_json(self._search_regex(
  31             r'(?m)^\s*window\.__INITIAL_JS_STATE__\s*=\s*(\{.+?\});\s*$', webpage, 'initial JS state'), video_id)
  32         if not initial_js_state.get('liveInfo'):
  33             raise UserNotLive(video_id=video_id)
  34
  35         return {
  36             'id': video_id,
  37             'title': traverse_obj(initial_js_state, ('liveInfo', 'title')),
  38             'comment_count': traverse_obj(initial_js_state, ('liveInfo', 'comments')),
  39             'view_count': traverse_obj(initial_js_state, ('liveInfo', 'visitor')),
  40             'timestamp': traverse_obj(initial_js_state, ('liveInfo', 'created')),
  41             'uploader': traverse_obj(initial_js_state, ('broadcasterInfo', 'name')),
  42             'uploader_id': video_id,
  43             'formats': [{
  44                 'format_id': 'hls',
  45                 'url': (traverse_obj(initial_js_state, ('liveInfo', 'hls'))
  46                         or f'https://d1hd0ww6piyb43.cloudfront.net/hls/torte_{video_id}.m3u8'),
  47                 'ext': 'mp4',
  48                 'protocol': 'm3u8',
  49             }],
  50             'is_live': True,
  51         }
  52
  53
  54 class MixchArchiveIE(InfoExtractor):
  55     IE_NAME = 'mixch:archive'
  56     _VALID_URL = r'https?://(?:www\.)?mixch\.tv/archive/(?P<id>\d+)'
  57
  58     _TESTS = [{
  59         'url': 'https://mixch.tv/archive/421',
  60         'skip': 'paid video, no DRM. expires at Jan 23',
  61         'info_dict': {
  62             'id': '421',
  63             'title': '96NEKO SHOW TIME',
  64         }
  65     }]
  66
  67     def _real_extract(self, url):
  68         video_id = self._match_id(url)
  69         webpage = self._download_webpage(url, video_id)
  70
  71         html5_videos = self._parse_html5_media_entries(
  72             url, webpage.replace('video-js', 'video'), video_id, 'hls')
  73         if not html5_videos:
  74             self.raise_login_required(method='cookies')
  75         infodict = html5_videos[0]
  76         infodict.update({
  77             'id': video_id,
  78             'title': self._html_search_regex(r'class="archive-title">(.+?)</', webpage, 'title')
  79         })
  80
  81         return infodict