yt_dlp/extractor/ichinanalive.py

   1 from .common import InfoExtractor
   2 from ..utils import ExtractorError, str_or_none, traverse_obj, unified_strdate
   3 from ..compat import compat_str
   4
   5
   6 class IchinanaLiveIE(InfoExtractor):
   7     IE_NAME = '17live'
   8     _VALID_URL = r'https?://(?:www\.)?17\.live/(?:[^/]+/)*(?:live|profile/r)/(?P<id>\d+)'
   9     _TESTS = [{
  10         'url': 'https://17.live/live/3773096',
  11         'info_dict': {
  12             'id': '3773096',
  13             'title': '萠珈☕🤡🍫moka',
  14             'is_live': True,
  15             'uploader': '萠珈☕🤡🍫moka',
  16             'uploader_id': '3773096',
  17             'like_count': 366,
  18             'view_count': 18121,
  19             'timestamp': 1630569012,
  20         },
  21         'skip': 'running as of writing, but may be ended as of testing',
  22     }, {
  23         'note': 'nothing except language differs',
  24         'url': 'https://17.live/ja/live/3773096',
  25         'only_matching': True,
  26     }]
  27
  28     @classmethod
  29     def suitable(cls, url):
  30         return not IchinanaLiveClipIE.suitable(url) and super(IchinanaLiveIE, cls).suitable(url)
  31
  32     def _real_extract(self, url):
  33         video_id = self._match_id(url)
  34         url = 'https://17.live/live/%s' % video_id
  35
  36         enter = self._download_json(
  37             'https://api-dsa.17app.co/api/v1/lives/%s/enter' % video_id, video_id,
  38             headers={'Referer': url}, fatal=False, expected_status=420,
  39             data=b'\0')
  40         if enter and enter.get('message') == 'ended':
  41             raise ExtractorError('This live has ended.', expected=True)
  42
  43         view_data = self._download_json(
  44             'https://api-dsa.17app.co/api/v1/lives/%s' % video_id, video_id,
  45             headers={'Referer': url})
  46
  47         uploader = traverse_obj(
  48             view_data, ('userInfo', 'displayName'), ('userInfo', 'openID'))
  49
  50         video_urls = view_data.get('rtmpUrls')
  51         if not video_urls:
  52             raise ExtractorError('unable to extract live URL information')
  53         formats = []
  54         for (name, value) in video_urls[0].items():
  55             if not isinstance(value, compat_str):
  56                 continue
  57             if not value.startswith('http'):
  58                 continue
  59             quality = -1
  60             if 'web' in name:
  61                 quality -= 1
  62             if 'High' in name:
  63                 quality += 4
  64             if 'Low' in name:
  65                 quality -= 2
  66             formats.append({
  67                 'format_id': name,
  68                 'url': value,
  69                 'quality': quality,
  70                 'http_headers': {'Referer': url},
  71                 'ext': 'flv',
  72                 'vcodec': 'h264',
  73                 'acodec': 'aac',
  74             })
  75
  76         self._sort_formats(formats)
  77
  78         return {
  79             'id': video_id,
  80             'title': uploader or video_id,
  81             'formats': formats,
  82             'is_live': True,
  83             'uploader': uploader,
  84             'uploader_id': video_id,
  85             'like_count': view_data.get('receivedLikeCount'),
  86             'view_count': view_data.get('viewerCount'),
  87             'thumbnail': view_data.get('coverPhoto'),
  88             'description': view_data.get('caption'),
  89             'timestamp': view_data.get('beginTime'),
  90         }
  91
  92
  93 class IchinanaLiveClipIE(InfoExtractor):
  94     IE_NAME = '17live:clip'
  95     _VALID_URL = r'https?://(?:www\.)?17\.live/(?:[^/]+/)*profile/r/(?P<uploader_id>\d+)/clip/(?P<id>[^/]+)'
  96     _TESTS = [{
  97         'url': 'https://17.live/profile/r/1789280/clip/1bHQSK8KUieruFXaCH4A4upCzlN',
  98         'info_dict': {
  99             'id': '1bHQSK8KUieruFXaCH4A4upCzlN',
 100             'title': 'マチコ先生🦋Class💋',
 101             'description': 'マチ戦隊　第一次　バスターコール\n総額200万coin！\n動画制作@うぉーかー🌱Walker🎫',
 102             'uploader_id': '1789280',
 103         },
 104     }, {
 105         'url': 'https://17.live/ja/profile/r/1789280/clip/1bHQSK8KUieruFXaCH4A4upCzlN',
 106         'only_matching': True,
 107     }]
 108
 109     def _real_extract(self, url):
 110         uploader_id, video_id = self._match_valid_url(url).groups()
 111         url = 'https://17.live/profile/r/%s/clip/%s' % (uploader_id, video_id)
 112
 113         view_data = self._download_json(
 114             'https://api-dsa.17app.co/api/v1/clips/%s' % video_id, video_id,
 115             headers={'Referer': url})
 116
 117         uploader = traverse_obj(
 118             view_data, ('userInfo', 'displayName'), ('userInfo', 'name'))
 119
 120         formats = []
 121         if view_data.get('videoURL'):
 122             formats.append({
 123                 'id': 'video',
 124                 'url': view_data['videoURL'],
 125                 'quality': -1,
 126             })
 127         if view_data.get('transcodeURL'):
 128             formats.append({
 129                 'id': 'transcode',
 130                 'url': view_data['transcodeURL'],
 131                 'quality': -1,
 132             })
 133         if view_data.get('srcVideoURL'):
 134             # highest quality
 135             formats.append({
 136                 'id': 'srcVideo',
 137                 'url': view_data['srcVideoURL'],
 138                 'quality': 1,
 139             })
 140
 141         for fmt in formats:
 142             fmt.update({
 143                 'ext': 'mp4',
 144                 'protocol': 'https',
 145                 'vcodec': 'h264',
 146                 'acodec': 'aac',
 147                 'http_headers': {'Referer': url},
 148             })
 149
 150         self._sort_formats(formats)
 151
 152         return {
 153             'id': video_id,
 154             'title': uploader or video_id,
 155             'formats': formats,
 156             'uploader': uploader,
 157             'uploader_id': uploader_id,
 158             'like_count': view_data.get('likeCount'),
 159             'view_count': view_data.get('viewCount'),
 160             'thumbnail': view_data.get('imageURL'),
 161             'duration': view_data.get('duration'),
 162             'description': view_data.get('caption'),
 163             'upload_date': unified_strdate(str_or_none(view_data.get('createdAt'))),
 164         }