yt_dlp/extractor/whowatch.py

   1 from .common import InfoExtractor
   2 from ..compat import compat_str
   3 from ..utils import (
   4     ExtractorError,
   5     int_or_none,
   6     qualities,
   7     try_call,
   8     try_get,
   9 )
  10
  11
  12 class WhoWatchIE(InfoExtractor):
  13     IE_NAME = 'whowatch'
  14     _VALID_URL = r'https?://whowatch\.tv/viewer/(?P<id>\d+)'
  15
  16     _TESTS = [{
  17         'url': 'https://whowatch.tv/viewer/21450171',
  18         'only_matching': True,
  19     }]
  20
  21     def _real_extract(self, url):
  22         video_id = self._match_id(url)
  23         self._download_webpage(url, video_id)
  24         metadata = self._download_json('https://api.whowatch.tv/lives/%s' % video_id, video_id)
  25         live_data = self._download_json('https://api.whowatch.tv/lives/%s/play' % video_id, video_id)
  26
  27         title = try_call(
  28             lambda: live_data['share_info']['live_title'][1:-1],
  29             lambda: metadata['live']['title'],
  30             expected_type=str)
  31
  32         hls_url = live_data.get('hls_url')
  33         if not hls_url:
  34             raise ExtractorError(live_data.get('error_message') or 'The user is offline.', expected=True)
  35
  36         QUALITIES = qualities(['low', 'medium', 'high', 'veryhigh'])
  37         formats = []
  38
  39         for i, fmt in enumerate(live_data.get('streams') or []):
  40             name = fmt.get('quality') or fmt.get('name') or compat_str(i)
  41             hls_url = fmt.get('hls_url')
  42             rtmp_url = fmt.get('rtmp_url')
  43             audio_only = fmt.get('audio_only')
  44             quality = QUALITIES(fmt.get('quality'))
  45
  46             if hls_url:
  47                 hls_fmts = self._extract_m3u8_formats(
  48                     hls_url, video_id, ext='mp4', m3u8_id='hls-%s' % name, quality=quality)
  49                 formats.extend(hls_fmts)
  50             else:
  51                 hls_fmts = []
  52
  53             # RTMP url for audio_only is same as high format, so skip it
  54             if rtmp_url and not audio_only:
  55                 formats.append({
  56                     'url': rtmp_url,
  57                     'format_id': 'rtmp-%s' % name,
  58                     'ext': 'mp4',
  59                     'protocol': 'rtmp_ffmpeg',  # ffmpeg can, while rtmpdump can't
  60                     'vcodec': 'h264',
  61                     'acodec': 'aac',
  62                     'quality': quality,
  63                     'format_note': fmt.get('label'),
  64                     # note: HLS and RTMP have same resolution for now, so it's acceptable
  65                     'width': try_get(hls_fmts, lambda x: x[0]['width'], int),
  66                     'height': try_get(hls_fmts, lambda x: x[0]['height'], int),
  67                 })
  68
  69         # This contains the same formats as the above manifests and is used only as a fallback
  70         formats.extend(self._extract_m3u8_formats(
  71             hls_url, video_id, ext='mp4', m3u8_id='hls'))
  72         self._remove_duplicate_formats(formats)
  73
  74         uploader_url = try_get(metadata, lambda x: x['live']['user']['user_path'], compat_str)
  75         if uploader_url:
  76             uploader_url = 'https://whowatch.tv/profile/%s' % uploader_url
  77         uploader_id = compat_str(try_get(metadata, lambda x: x['live']['user']['id'], int))
  78         uploader = try_get(metadata, lambda x: x['live']['user']['name'], compat_str)
  79         thumbnail = try_get(metadata, lambda x: x['live']['latest_thumbnail_url'], compat_str)
  80         timestamp = int_or_none(try_get(metadata, lambda x: x['live']['started_at'], int), scale=1000)
  81         view_count = try_get(metadata, lambda x: x['live']['total_view_count'], int)
  82         comment_count = try_get(metadata, lambda x: x['live']['comment_count'], int)
  83
  84         return {
  85             'id': video_id,
  86             'title': title,
  87             'uploader_id': uploader_id,
  88             'uploader_url': uploader_url,
  89             'uploader': uploader,
  90             'formats': formats,
  91             'thumbnail': thumbnail,
  92             'timestamp': timestamp,
  93             'view_count': view_count,
  94             'comment_count': comment_count,
  95             'is_live': True,
  96         }