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