yt_dlp/extractor/jixie.py

   1 from .common import InfoExtractor
   2 from ..utils import clean_html, float_or_none, traverse_obj, try_call
   3
   4
   5 class JixieBaseIE(InfoExtractor):
   6     """
   7     API Reference:
   8         https://jixie.atlassian.net/servicedesk/customer/portal/2/article/1339654214?src=-1456335525,
   9         https://scripts.jixie.media/jxvideo.3.1.min.js
  10     """
  11
  12     def _extract_data_from_jixie_id(self, display_id, video_id, webpage):
  13         json_data = self._download_json(
  14             'https://apidam.jixie.io/api/public/stream', display_id,
  15             query={'metadata': 'full', 'video_id': video_id})['data']
  16
  17         formats, subtitles = [], {}
  18         for stream in json_data['streams']:
  19             if stream.get('type') == 'HLS':
  20                 fmt, sub = self._extract_m3u8_formats_and_subtitles(stream.get('url'), display_id, ext='mp4')
  21                 if json_data.get('drm'):
  22                     for f in fmt:
  23                         f['has_drm'] = True
  24                 formats.extend(fmt)
  25                 self._merge_subtitles(sub, target=subtitles)
  26             else:
  27                 formats.append({
  28                     'url': stream.get('url'),
  29                     'width': stream.get('width'),
  30                     'height': stream.get('height'),
  31                     'ext': 'mp4',
  32                 })
  33
  34         return {
  35             'id': video_id,
  36             'display_id': display_id,
  37             'formats': formats,
  38             'subtitles': subtitles,
  39             'title': json_data.get('title') or self._html_search_meta(['og:title', 'twitter:title'], webpage),
  40             'description': (clean_html(traverse_obj(json_data, ('metadata', 'description')))
  41                             or self._html_search_meta(['description', 'og:description', 'twitter:description'], webpage)),
  42             'thumbnails': traverse_obj(json_data, ('metadata', 'thumbnails')),
  43             'duration': float_or_none(traverse_obj(json_data, ('metadata', 'duration'))),
  44             'tags': try_call(lambda: (json_data['metadata']['keywords'] or None).split(',')),
  45             'categories': try_call(lambda: (json_data['metadata']['categories'] or None).split(',')),
  46             'uploader_id': json_data.get('owner_id'),
  47         }