yt_dlp/extractor/slideslive.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     bool_or_none,
   7     smuggle_url,
   8     try_get,
   9     url_or_none,
  10 )
  11
  12
  13 class SlidesLiveIE(InfoExtractor):
  14     _VALID_URL = r'https?://slideslive\.com/(?P<id>[0-9]+)'
  15     _TESTS = [{
  16         # video_service_name = YOUTUBE
  17         'url': 'https://slideslive.com/38902413/gcc-ia16-backend',
  18         'md5': 'b29fcd6c6952d0c79c5079b0e7a07e6f',
  19         'info_dict': {
  20             'id': 'LMtgR8ba0b0',
  21             'ext': 'mp4',
  22             'title': 'GCC IA16 backend',
  23             'description': 'Watch full version of this video at https://slideslive.com/38902413.',
  24             'uploader': 'SlidesLive Videos - A',
  25             'uploader_id': 'UC62SdArr41t_-_fX40QCLRw',
  26             'timestamp': 1597615266,
  27             'upload_date': '20170925',
  28         }
  29     }, {
  30         # video_service_name = yoda
  31         'url': 'https://slideslive.com/38935785',
  32         'md5': '575cd7a6c0acc6e28422fe76dd4bcb1a',
  33         'info_dict': {
  34             'id': 'RMraDYN5ozA_',
  35             'ext': 'mp4',
  36             'title': 'Offline Reinforcement Learning: From Algorithms to Practical Challenges',
  37         },
  38         'params': {
  39             'format': 'bestvideo',
  40         },
  41     }, {
  42         # video_service_name = youtube
  43         'url': 'https://slideslive.com/38903721/magic-a-scientific-resurrection-of-an-esoteric-legend',
  44         'only_matching': True,
  45     }, {
  46         # video_service_name = url
  47         'url': 'https://slideslive.com/38922070/learning-transferable-skills-1',
  48         'only_matching': True,
  49     }, {
  50         # video_service_name = vimeo
  51         'url': 'https://slideslive.com/38921896/retrospectives-a-venue-for-selfreflection-in-ml-research-3',
  52         'only_matching': True,
  53     }]
  54
  55     def _real_extract(self, url):
  56         video_id = self._match_id(url)
  57         video_data = self._download_json(
  58             'https://ben.slideslive.com/player/' + video_id, video_id)
  59         service_name = video_data['video_service_name'].lower()
  60         assert service_name in ('url', 'yoda', 'vimeo', 'youtube')
  61         service_id = video_data['video_service_id']
  62         subtitles = {}
  63         for sub in try_get(video_data, lambda x: x['subtitles'], list) or []:
  64             if not isinstance(sub, dict):
  65                 continue
  66             webvtt_url = url_or_none(sub.get('webvtt_url'))
  67             if not webvtt_url:
  68                 continue
  69             lang = sub.get('language') or 'en'
  70             subtitles.setdefault(lang, []).append({
  71                 'url': webvtt_url,
  72             })
  73         info = {
  74             'id': video_id,
  75             'thumbnail': video_data.get('thumbnail'),
  76             'is_live': bool_or_none(video_data.get('is_live')),
  77             'subtitles': subtitles,
  78         }
  79         if service_name in ('url', 'yoda'):
  80             info['title'] = video_data['title']
  81             if service_name == 'url':
  82                 info['url'] = service_id
  83             else:
  84                 formats = []
  85                 _MANIFEST_PATTERN = 'https://01.cdn.yoda.slideslive.com/%s/master.%s'
  86                 # use `m3u8` entry_protocol until EXT-X-MAP is properly supported by `m3u8_native` entry_protocol
  87                 formats.extend(self._extract_m3u8_formats(
  88                     _MANIFEST_PATTERN % (service_id, 'm3u8'),
  89                     service_id, 'mp4', m3u8_id='hls', fatal=False))
  90                 formats.extend(self._extract_mpd_formats(
  91                     _MANIFEST_PATTERN % (service_id, 'mpd'), service_id,
  92                     mpd_id='dash', fatal=False))
  93                 self._sort_formats(formats)
  94                 info.update({
  95                     'id': service_id,
  96                     'formats': formats,
  97                 })
  98         else:
  99             info.update({
 100                 '_type': 'url_transparent',
 101                 'url': service_id,
 102                 'ie_key': service_name.capitalize(),
 103                 'title': video_data.get('title'),
 104             })
 105             if service_name == 'vimeo':
 106                 info['url'] = smuggle_url(
 107                     'https://player.vimeo.com/video/' + service_id,
 108                     {'http_headers': {'Referer': url}})
 109         return info