yt_dlp/extractor/videoken.py

   1 import base64
   2 import functools
   3 import math
   4 import re
   5 import time
   6 import urllib.parse
   7
   8 from .common import InfoExtractor
   9 from .slideslive import SlidesLiveIE
  10 from ..utils import (
  11     ExtractorError,
  12     InAdvancePagedList,
  13     int_or_none,
  14     traverse_obj,
  15     update_url_query,
  16     url_or_none,
  17 )
  18
  19
  20 class VideoKenBaseIE(InfoExtractor):
  21     _ORGANIZATIONS = {
  22         'videos.icts.res.in': 'icts',
  23         'videos.cncf.io': 'cncf',
  24         'videos.neurips.cc': 'neurips',
  25     }
  26     _BASE_URL_RE = rf'https?://(?P<host>{"|".join(map(re.escape, _ORGANIZATIONS))})/'
  27
  28     _PAGE_SIZE = 12
  29
  30     def _get_org_id_and_api_key(self, org, video_id):
  31         details = self._download_json(
  32             f'https://analytics.videoken.com/api/videolake/{org}/details', video_id,
  33             note='Downloading organization ID and API key', headers={
  34                 'Accept': 'application/json',
  35             })
  36         return details['id'], details['apikey']
  37
  38     def _create_slideslive_url(self, video_url, video_id, referer):
  39         if not video_url and not video_id:
  40             return
  41         elif not video_url or 'embed/sign-in' in video_url:
  42             video_url = f'https://slideslive.com/embed/{video_id.lstrip("slideslive-")}'
  43         if url_or_none(referer):
  44             return update_url_query(video_url, {
  45                 'embed_parent_url': referer,
  46                 'embed_container_origin': f'https://{urllib.parse.urlparse(referer).netloc}',
  47             })
  48         return video_url
  49
  50     def _extract_videos(self, videos, url):
  51         for video in traverse_obj(videos, (('videos', 'results'), ...)):
  52             video_id = traverse_obj(video, 'youtube_id', 'videoid')
  53             if not video_id:
  54                 continue
  55             ie_key = None
  56             if traverse_obj(video, 'type', 'source') == 'youtube':
  57                 video_url = video_id
  58                 ie_key = 'Youtube'
  59             else:
  60                 video_url = traverse_obj(video, 'embed_url', 'embeddableurl')
  61                 if urllib.parse.urlparse(video_url).netloc == 'slideslive.com':
  62                     ie_key = SlidesLiveIE
  63                     video_url = self._create_slideslive_url(video_url, video_id, url)
  64             if not video_url:
  65                 continue
  66             yield self.url_result(video_url, ie_key, video_id)
  67
  68
  69 class VideoKenIE(VideoKenBaseIE):
  70     _VALID_URL = VideoKenBaseIE._BASE_URL_RE + r'(?:(?:topic|category)/[^/#?]+/)?video/(?P<id>[\w-]+)'
  71     _TESTS = [{
  72         # neurips -> videoken -> slideslive
  73         'url': 'https://videos.neurips.cc/video/slideslive-38922815',
  74         'info_dict': {
  75             'id': '38922815',
  76             'ext': 'mp4',
  77             'title': 'Efficient Processing of Deep Neural Network: from Algorithms to Hardware Architectures',
  78             'timestamp': 1630939331,
  79             'upload_date': '20210906',
  80             'thumbnail': r're:^https?://.*\.(?:jpg|png)',
  81             'thumbnails': 'count:330',
  82             'chapters': 'count:329',
  83         },
  84         'params': {
  85             'skip_download': 'm3u8',
  86         },
  87         'expected_warnings': ['Failed to download VideoKen API JSON'],
  88     }, {
  89         # neurips -> videoken -> slideslive -> youtube
  90         'url': 'https://videos.neurips.cc/topic/machine%20learning/video/slideslive-38923348',
  91         'info_dict': {
  92             'id': '2Xa_dt78rJE',
  93             'ext': 'mp4',
  94             'display_id': '38923348',
  95             'title': 'Machine Education',
  96             'description': 'Watch full version of this video at https://slideslive.com/38923348.',
  97             'channel': 'SlidesLive Videos - G2',
  98             'channel_id': 'UCOExahQQ588Da8Nft_Ltb9w',
  99             'channel_url': 'https://www.youtube.com/channel/UCOExahQQ588Da8Nft_Ltb9w',
 100             'uploader': 'SlidesLive Videos - G2',
 101             'uploader_id': 'UCOExahQQ588Da8Nft_Ltb9w',
 102             'uploader_url': 'http://www.youtube.com/channel/UCOExahQQ588Da8Nft_Ltb9w',
 103             'duration': 2504,
 104             'timestamp': 1618922125,
 105             'upload_date': '20200131',
 106             'age_limit': 0,
 107             'channel_follower_count': int,
 108             'view_count': int,
 109             'availability': 'unlisted',
 110             'live_status': 'not_live',
 111             'playable_in_embed': True,
 112             'categories': ['People & Blogs'],
 113             'tags': [],
 114             'thumbnail': r're:^https?://.*\.(?:jpg|webp)',
 115             'thumbnails': 'count:78',
 116             'chapters': 'count:77',
 117         },
 118         'params': {
 119             'skip_download': 'm3u8',
 120         },
 121         'expected_warnings': ['Failed to download VideoKen API JSON'],
 122     }, {
 123         # icts -> videoken -> youtube
 124         'url': 'https://videos.icts.res.in/topic/random%20variable/video/zysIsojYdvc',
 125         'info_dict': {
 126             'id': 'zysIsojYdvc',
 127             'ext': 'mp4',
 128             'title': 'Small-worlds, complex networks and random graphs (Lecture 3)  by Remco van der Hofstad',
 129             'description': 'md5:87433069d79719eeadc1962cc2ace00b',
 130             'channel': 'International Centre for Theoretical Sciences',
 131             'channel_id': 'UCO3xnVTHzB7l-nc8mABUJIQ',
 132             'channel_url': 'https://www.youtube.com/channel/UCO3xnVTHzB7l-nc8mABUJIQ',
 133             'uploader': 'International Centre for Theoretical Sciences',
 134             'uploader_id': 'ICTStalks',
 135             'uploader_url': 'http://www.youtube.com/user/ICTStalks',
 136             'duration': 3372,
 137             'upload_date': '20191004',
 138             'age_limit': 0,
 139             'live_status': 'not_live',
 140             'availability': 'public',
 141             'playable_in_embed': True,
 142             'channel_follower_count': int,
 143             'like_count': int,
 144             'view_count': int,
 145             'categories': ['Science & Technology'],
 146             'tags': [],
 147             'thumbnail': r're:^https?://.*\.(?:jpg|webp)',
 148             'thumbnails': 'count:42',
 149             'chapters': 'count:20',
 150         },
 151         'params': {
 152             'skip_download': 'm3u8',
 153         },
 154     }, {
 155         'url': 'https://videos.cncf.io/category/478/video/IL4nxbmUIX8',
 156         'only_matching': True,
 157     }, {
 158         'url': 'https://videos.cncf.io/topic/kubernetes/video/YAM2d7yTrrI',
 159         'only_matching': True,
 160     }, {
 161         'url': 'https://videos.icts.res.in/video/d7HuP_abpKU',
 162         'only_matching': True,
 163     }]
 164
 165     def _real_extract(self, url):
 166         hostname, video_id = self._match_valid_url(url).group('host', 'id')
 167         org_id, _ = self._get_org_id_and_api_key(self._ORGANIZATIONS[hostname], video_id)
 168         details = self._download_json(
 169             'https://analytics.videoken.com/api/videoinfo_private', video_id, query={
 170                 'videoid': video_id,
 171                 'org_id': org_id,
 172             }, headers={'Accept': 'application/json'}, note='Downloading VideoKen API JSON',
 173             errnote='Failed to download VideoKen API JSON', fatal=False)
 174         if details:
 175             return next(self._extract_videos({'videos': [details]}, url))
 176         # fallback for API error 400 response
 177         elif video_id.startswith('slideslive-'):
 178             return self.url_result(
 179                 self._create_slideslive_url(None, video_id, url), SlidesLiveIE, video_id)
 180         elif re.match(r'^[\w-]{11}$', video_id):
 181             self.url_result(video_id, 'Youtube', video_id)
 182         else:
 183             raise ExtractorError('Unable to extract without VideoKen API response')
 184
 185
 186 class VideoKenPlayerIE(VideoKenBaseIE):
 187     _VALID_URL = r'https?://player\.videoken\.com/embed/slideslive-(?P<id>\d+)'
 188     _TESTS = [{
 189         'url': 'https://player.videoken.com/embed/slideslive-38968434',
 190         'info_dict': {
 191             'id': '38968434',
 192             'ext': 'mp4',
 193             'title': 'Deep Learning with Label Differential Privacy',
 194             'timestamp': 1643377020,
 195             'upload_date': '20220128',
 196             'thumbnail': r're:^https?://.*\.(?:jpg|png)',
 197             'thumbnails': 'count:30',
 198             'chapters': 'count:29',
 199         },
 200         'params': {
 201             'skip_download': 'm3u8',
 202         },
 203     }]
 204
 205     def _real_extract(self, url):
 206         video_id = self._match_id(url)
 207         return self.url_result(
 208             self._create_slideslive_url(None, video_id, url), SlidesLiveIE, video_id)
 209
 210
 211 class VideoKenPlaylistIE(VideoKenBaseIE):
 212     _VALID_URL = VideoKenBaseIE._BASE_URL_RE + r'(?:category/\d+/)?playlist/(?P<id>\d+)'
 213     _TESTS = [{
 214         'url': 'https://videos.icts.res.in/category/1822/playlist/381',
 215         'playlist_mincount': 117,
 216         'info_dict': {
 217             'id': '381',
 218             'title': 'Cosmology - The Next Decade',
 219         },
 220     }]
 221
 222     def _real_extract(self, url):
 223         hostname, playlist_id = self._match_valid_url(url).group('host', 'id')
 224         org_id, _ = self._get_org_id_and_api_key(self._ORGANIZATIONS[hostname], playlist_id)
 225         videos = self._download_json(
 226             f'https://analytics.videoken.com/api/{org_id}/playlistitems/{playlist_id}/',
 227             playlist_id, headers={'Accept': 'application/json'}, note='Downloading API JSON')
 228         return self.playlist_result(self._extract_videos(videos, url), playlist_id, videos.get('title'))
 229
 230
 231 class VideoKenCategoryIE(VideoKenBaseIE):
 232     _VALID_URL = VideoKenBaseIE._BASE_URL_RE + r'category/(?P<id>\d+)/?(?:$|[?#])'
 233     _TESTS = [{
 234         'url': 'https://videos.icts.res.in/category/1822/',
 235         'playlist_mincount': 500,
 236         'info_dict': {
 237             'id': '1822',
 238             'title': 'Programs',
 239         },
 240     }, {
 241         'url': 'https://videos.neurips.cc/category/350/',
 242         'playlist_mincount': 34,
 243         'info_dict': {
 244             'id': '350',
 245             'title': 'NeurIPS 2018',
 246         },
 247     }, {
 248         'url': 'https://videos.cncf.io/category/479/',
 249         'playlist_mincount': 328,
 250         'info_dict': {
 251             'id': '479',
 252             'title': 'KubeCon + CloudNativeCon Europe\'19',
 253         },
 254     }]
 255
 256     def _get_category_page(self, category_id, org_id, page=1, note=None):
 257         return self._download_json(
 258             f'https://analytics.videoken.com/api/videolake/{org_id}/category_videos', category_id,
 259             fatal=False, note=note if note else f'Downloading category page {page}',
 260             query={
 261                 'category_id': category_id,
 262                 'page_number': page,
 263                 'length': self._PAGE_SIZE,
 264             }, headers={'Accept': 'application/json'}) or {}
 265
 266     def _entries(self, category_id, org_id, url, page):
 267         videos = self._get_category_page(category_id, org_id, page + 1)
 268         yield from self._extract_videos(videos, url)
 269
 270     def _real_extract(self, url):
 271         hostname, category_id = self._match_valid_url(url).group('host', 'id')
 272         org_id, _ = self._get_org_id_and_api_key(self._ORGANIZATIONS[hostname], category_id)
 273         category_info = self._get_category_page(category_id, org_id, note='Downloading category info')
 274         category = category_info['category_name']
 275         total_pages = math.ceil(int(category_info['recordsTotal']) / self._PAGE_SIZE)
 276         return self.playlist_result(InAdvancePagedList(
 277             functools.partial(self._entries, category_id, org_id, url),
 278             total_pages, self._PAGE_SIZE), category_id, category)
 279
 280
 281 class VideoKenTopicIE(VideoKenBaseIE):
 282     _VALID_URL = VideoKenBaseIE._BASE_URL_RE + r'topic/(?P<id>[^/#?]+)/?(?:$|[?#])'
 283     _TESTS = [{
 284         'url': 'https://videos.neurips.cc/topic/machine%20learning/',
 285         'playlist_mincount': 500,
 286         'info_dict': {
 287             'id': 'machine_learning',
 288             'title': 'machine learning',
 289         },
 290     }, {
 291         'url': 'https://videos.icts.res.in/topic/gravitational%20waves/',
 292         'playlist_mincount': 77,
 293         'info_dict': {
 294             'id': 'gravitational_waves',
 295             'title': 'gravitational waves'
 296         },
 297     }, {
 298         'url': 'https://videos.cncf.io/topic/prometheus/',
 299         'playlist_mincount': 134,
 300         'info_dict': {
 301             'id': 'prometheus',
 302             'title': 'prometheus',
 303         },
 304     }]
 305
 306     def _get_topic_page(self, topic, org_id, search_id, api_key, page=1, note=None):
 307         return self._download_json(
 308             'https://es.videoken.com/api/v1.0/get_results', topic, fatal=False, query={
 309                 'orgid': org_id,
 310                 'size': self._PAGE_SIZE,
 311                 'query': topic,
 312                 'page': page,
 313                 'sort': 'upload_desc',
 314                 'filter': 'all',
 315                 'token': api_key,
 316                 'is_topic': 'true',
 317                 'category': '',
 318                 'searchid': search_id,
 319             }, headers={'Accept': 'application/json'},
 320             note=note if note else f'Downloading topic page {page}') or {}
 321
 322     def _entries(self, topic, org_id, search_id, api_key, url, page):
 323         videos = self._get_topic_page(topic, org_id, search_id, api_key, page + 1)
 324         yield from self._extract_videos(videos, url)
 325
 326     def _real_extract(self, url):
 327         hostname, topic_id = self._match_valid_url(url).group('host', 'id')
 328         topic = urllib.parse.unquote(topic_id)
 329         topic_id = topic.replace(' ', '_')
 330         org_id, api_key = self._get_org_id_and_api_key(self._ORGANIZATIONS[hostname], topic)
 331         search_id = base64.b64encode(f':{topic}:{int(time.time())}:transient'.encode()).decode()
 332         total_pages = int_or_none(self._get_topic_page(
 333             topic, org_id, search_id, api_key, note='Downloading topic info')['total_no_of_pages'])
 334         return self.playlist_result(InAdvancePagedList(
 335             functools.partial(self._entries, topic, org_id, search_id, api_key, url),
 336             total_pages, self._PAGE_SIZE), topic_id, topic)