yt_dlp/extractor/slideslive.py

   1 from .common import InfoExtractor
   2 from ..utils import (
   3     bool_or_none,
   4     smuggle_url,
   5     try_get,
   6     url_or_none,
   7 )
   8
   9
  10 class SlidesLiveIE(InfoExtractor):
  11     _VALID_URL = r'https?://slideslive\.com/(?P<id>[0-9]+)'
  12     _TESTS = [{
  13         # video_service_name = YOUTUBE
  14         'url': 'https://slideslive.com/38902413/gcc-ia16-backend',
  15         'md5': 'b29fcd6c6952d0c79c5079b0e7a07e6f',
  16         'info_dict': {
  17             'id': 'LMtgR8ba0b0',
  18             'ext': 'mp4',
  19             'title': 'GCC IA16 backend',
  20             'description': 'Watch full version of this video at https://slideslive.com/38902413.',
  21             'uploader': 'SlidesLive Videos - A',
  22             'uploader_id': 'UC62SdArr41t_-_fX40QCLRw',
  23             'timestamp': 1597615266,
  24             'upload_date': '20170925',
  25         }
  26     }, {
  27         # video_service_name = yoda
  28         'url': 'https://slideslive.com/38935785',
  29         'md5': '575cd7a6c0acc6e28422fe76dd4bcb1a',
  30         'info_dict': {
  31             'id': 'RMraDYN5ozA_',
  32             'ext': 'mp4',
  33             'title': 'Offline Reinforcement Learning: From Algorithms to Practical Challenges',
  34         },
  35     }, {
  36         # video_service_name = youtube
  37         'url': 'https://slideslive.com/38903721/magic-a-scientific-resurrection-of-an-esoteric-legend',
  38         'only_matching': True,
  39     }, {
  40         # video_service_name = url
  41         'url': 'https://slideslive.com/38922070/learning-transferable-skills-1',
  42         'only_matching': True,
  43     }, {
  44         # video_service_name = vimeo
  45         'url': 'https://slideslive.com/38921896/retrospectives-a-venue-for-selfreflection-in-ml-research-3',
  46         'only_matching': True,
  47     }]
  48
  49     def _real_extract(self, url):
  50         video_id = self._match_id(url)
  51         video_data = self._download_json(
  52             'https://ben.slideslive.com/player/' + video_id, video_id)
  53         service_name = video_data['video_service_name'].lower()
  54         assert service_name in ('url', 'yoda', 'vimeo', 'youtube')
  55         service_id = video_data['video_service_id']
  56         subtitles = {}
  57         for sub in try_get(video_data, lambda x: x['subtitles'], list) or []:
  58             if not isinstance(sub, dict):
  59                 continue
  60             webvtt_url = url_or_none(sub.get('webvtt_url'))
  61             if not webvtt_url:
  62                 continue
  63             lang = sub.get('language') or 'en'
  64             subtitles.setdefault(lang, []).append({
  65                 'url': webvtt_url,
  66             })
  67         info = {
  68             'id': video_id,
  69             'thumbnail': video_data.get('thumbnail'),
  70             'is_live': bool_or_none(video_data.get('is_live')),
  71             'subtitles': subtitles,
  72         }
  73         if service_name in ('url', 'yoda'):
  74             info['title'] = video_data['title']
  75             if service_name == 'url':
  76                 info['url'] = service_id
  77             else:
  78                 formats = []
  79                 _MANIFEST_PATTERN = 'https://01.cdn.yoda.slideslive.com/%s/master.%s'
  80                 # use `m3u8` entry_protocol until EXT-X-MAP is properly supported by `m3u8_native` entry_protocol
  81                 formats.extend(self._extract_m3u8_formats(
  82                     _MANIFEST_PATTERN % (service_id, 'm3u8'),
  83                     service_id, 'mp4', m3u8_id='hls', fatal=False))
  84                 formats.extend(self._extract_mpd_formats(
  85                     _MANIFEST_PATTERN % (service_id, 'mpd'), service_id,
  86                     mpd_id='dash', fatal=False))
  87                 self._sort_formats(formats)
  88                 info.update({
  89                     'id': service_id,
  90                     'formats': formats,
  91                 })
  92         else:
  93             info.update({
  94                 '_type': 'url_transparent',
  95                 'url': service_id,
  96                 'ie_key': service_name.capitalize(),
  97                 'title': video_data.get('title'),
  98             })
  99             if service_name == 'vimeo':
 100                 info['url'] = smuggle_url(
 101                     'https://player.vimeo.com/video/' + service_id,
 102                     {'http_headers': {'Referer': url}})
 103         return info