yt_dlp/extractor/limelight.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..compat import compat_HTTPError
   5 from ..utils import (
   6     determine_ext,
   7     float_or_none,
   8     int_or_none,
   9     smuggle_url,
  10     try_get,
  11     unsmuggle_url,
  12     ExtractorError,
  13 )
  14
  15
  16 class LimelightBaseIE(InfoExtractor):
  17     _PLAYLIST_SERVICE_URL = 'http://production-ps.lvp.llnw.net/r/PlaylistService/%s/%s/%s'
  18
  19     @classmethod
  20     def _extract_embed_urls(cls, url, webpage):
  21         lm = {
  22             'Media': 'media',
  23             'Channel': 'channel',
  24             'ChannelList': 'channel_list',
  25         }
  26
  27         def smuggle(url):
  28             return smuggle_url(url, {'source_url': url})
  29
  30         entries = []
  31         for kind, video_id in re.findall(
  32                 r'LimelightPlayer\.doLoad(Media|Channel|ChannelList)\(["\'](?P<id>[a-z0-9]{32})',
  33                 webpage):
  34             entries.append(cls.url_result(
  35                 smuggle('limelight:%s:%s' % (lm[kind], video_id)),
  36                 'Limelight%s' % kind, video_id))
  37         for mobj in re.finditer(
  38                 # As per [1] class attribute should be exactly equal to
  39                 # LimelightEmbeddedPlayerFlash but numerous examples seen
  40                 # that don't exactly match it (e.g. [2]).
  41                 # 1. http://support.3playmedia.com/hc/en-us/articles/227732408-Limelight-Embedding-the-Captions-Plugin-with-the-Limelight-Player-on-Your-Webpage
  42                 # 2. http://www.sedona.com/FacilitatorTraining2017
  43                 r'''(?sx)
  44                     <object[^>]+class=(["\'])(?:(?!\1).)*\bLimelightEmbeddedPlayerFlash\b(?:(?!\1).)*\1[^>]*>.*?
  45                         <param[^>]+
  46                             name=(["\'])flashVars\2[^>]+
  47                             value=(["\'])(?:(?!\3).)*(?P<kind>media|channel(?:List)?)Id=(?P<id>[a-z0-9]{32})
  48                 ''', webpage):
  49             kind, video_id = mobj.group('kind'), mobj.group('id')
  50             entries.append(cls.url_result(
  51                 smuggle('limelight:%s:%s' % (kind, video_id)),
  52                 'Limelight%s' % kind.capitalize(), video_id))
  53         # http://support.3playmedia.com/hc/en-us/articles/115009517327-Limelight-Embedding-the-Audio-Description-Plugin-with-the-Limelight-Player-on-Your-Web-Page)
  54         for video_id in re.findall(
  55                 r'(?s)LimelightPlayerUtil\.embed\s*\(\s*{.*?\bmediaId["\']\s*:\s*["\'](?P<id>[a-z0-9]{32})',
  56                 webpage):
  57             entries.append(cls.url_result(
  58                 smuggle('limelight:media:%s' % video_id),
  59                 LimelightMediaIE.ie_key(), video_id))
  60         return entries
  61
  62     def _call_playlist_service(self, item_id, method, fatal=True, referer=None):
  63         headers = {}
  64         if referer:
  65             headers['Referer'] = referer
  66         try:
  67             return self._download_json(
  68                 self._PLAYLIST_SERVICE_URL % (self._PLAYLIST_SERVICE_PATH, item_id, method),
  69                 item_id, 'Downloading PlaylistService %s JSON' % method,
  70                 fatal=fatal, headers=headers)
  71         except ExtractorError as e:
  72             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
  73                 error = self._parse_json(e.cause.read().decode(), item_id)['detail']['contentAccessPermission']
  74                 if error == 'CountryDisabled':
  75                     self.raise_geo_restricted()
  76                 raise ExtractorError(error, expected=True)
  77             raise
  78
  79     def _extract(self, item_id, pc_method, mobile_method, referer=None):
  80         pc = self._call_playlist_service(item_id, pc_method, referer=referer)
  81         mobile = self._call_playlist_service(
  82             item_id, mobile_method, fatal=False, referer=referer)
  83         return pc, mobile
  84
  85     def _extract_info(self, pc, mobile, i, referer):
  86         get_item = lambda x, y: try_get(x, lambda x: x[y][i], dict) or {}
  87         pc_item = get_item(pc, 'playlistItems')
  88         mobile_item = get_item(mobile, 'mediaList')
  89         video_id = pc_item.get('mediaId') or mobile_item['mediaId']
  90         title = pc_item.get('title') or mobile_item['title']
  91
  92         formats = []
  93         urls = []
  94         for stream in pc_item.get('streams', []):
  95             stream_url = stream.get('url')
  96             if not stream_url or stream_url in urls:
  97                 continue
  98             if not self.get_param('allow_unplayable_formats') and stream.get('drmProtected'):
  99                 continue
 100             urls.append(stream_url)
 101             ext = determine_ext(stream_url)
 102             if ext == 'f4m':
 103                 formats.extend(self._extract_f4m_formats(
 104                     stream_url, video_id, f4m_id='hds', fatal=False))
 105             else:
 106                 fmt = {
 107                     'url': stream_url,
 108                     'abr': float_or_none(stream.get('audioBitRate')),
 109                     'fps': float_or_none(stream.get('videoFrameRate')),
 110                     'ext': ext,
 111                 }
 112                 width = int_or_none(stream.get('videoWidthInPixels'))
 113                 height = int_or_none(stream.get('videoHeightInPixels'))
 114                 vbr = float_or_none(stream.get('videoBitRate'))
 115                 if width or height or vbr:
 116                     fmt.update({
 117                         'width': width,
 118                         'height': height,
 119                         'vbr': vbr,
 120                     })
 121                 else:
 122                     fmt['vcodec'] = 'none'
 123                 rtmp = re.search(r'^(?P<url>rtmpe?://(?P<host>[^/]+)/(?P<app>.+))/(?P<playpath>mp[34]:.+)$', stream_url)
 124                 if rtmp:
 125                     format_id = 'rtmp'
 126                     if stream.get('videoBitRate'):
 127                         format_id += '-%d' % int_or_none(stream['videoBitRate'])
 128                     http_format_id = format_id.replace('rtmp', 'http')
 129
 130                     CDN_HOSTS = (
 131                         ('delvenetworks.com', 'cpl.delvenetworks.com'),
 132                         ('video.llnw.net', 's2.content.video.llnw.net'),
 133                     )
 134                     for cdn_host, http_host in CDN_HOSTS:
 135                         if cdn_host not in rtmp.group('host').lower():
 136                             continue
 137                         http_url = 'http://%s/%s' % (http_host, rtmp.group('playpath')[4:])
 138                         urls.append(http_url)
 139                         if self._is_valid_url(http_url, video_id, http_format_id):
 140                             http_fmt = fmt.copy()
 141                             http_fmt.update({
 142                                 'url': http_url,
 143                                 'format_id': http_format_id,
 144                             })
 145                             formats.append(http_fmt)
 146                             break
 147
 148                     fmt.update({
 149                         'url': rtmp.group('url'),
 150                         'play_path': rtmp.group('playpath'),
 151                         'app': rtmp.group('app'),
 152                         'ext': 'flv',
 153                         'format_id': format_id,
 154                     })
 155                 formats.append(fmt)
 156
 157         for mobile_url in mobile_item.get('mobileUrls', []):
 158             media_url = mobile_url.get('mobileUrl')
 159             format_id = mobile_url.get('targetMediaPlatform')
 160             if not media_url or media_url in urls:
 161                 continue
 162             if (format_id in ('Widevine', 'SmoothStreaming')
 163                     and not self.get_param('allow_unplayable_formats', False)):
 164                 continue
 165             urls.append(media_url)
 166             ext = determine_ext(media_url)
 167             if ext == 'm3u8':
 168                 formats.extend(self._extract_m3u8_formats(
 169                     media_url, video_id, 'mp4', 'm3u8_native',
 170                     m3u8_id=format_id, fatal=False))
 171             elif ext == 'f4m':
 172                 formats.extend(self._extract_f4m_formats(
 173                     stream_url, video_id, f4m_id=format_id, fatal=False))
 174             else:
 175                 formats.append({
 176                     'url': media_url,
 177                     'format_id': format_id,
 178                     'quality': -10,
 179                     'ext': ext,
 180                 })
 181
 182         self._sort_formats(formats)
 183
 184         subtitles = {}
 185         for flag in mobile_item.get('flags'):
 186             if flag == 'ClosedCaptions':
 187                 closed_captions = self._call_playlist_service(
 188                     video_id, 'getClosedCaptionsDetailsByMediaId',
 189                     False, referer) or []
 190                 for cc in closed_captions:
 191                     cc_url = cc.get('webvttFileUrl')
 192                     if not cc_url:
 193                         continue
 194                     lang = cc.get('languageCode') or self._search_regex(r'/([a-z]{2})\.vtt', cc_url, 'lang', default='en')
 195                     subtitles.setdefault(lang, []).append({
 196                         'url': cc_url,
 197                     })
 198                 break
 199
 200         get_meta = lambda x: pc_item.get(x) or mobile_item.get(x)
 201
 202         return {
 203             'id': video_id,
 204             'title': title,
 205             'description': get_meta('description'),
 206             'formats': formats,
 207             'duration': float_or_none(get_meta('durationInMilliseconds'), 1000),
 208             'thumbnail': get_meta('previewImageUrl') or get_meta('thumbnailImageUrl'),
 209             'subtitles': subtitles,
 210         }
 211
 212
 213 class LimelightMediaIE(LimelightBaseIE):
 214     IE_NAME = 'limelight'
 215     _VALID_URL = r'''(?x)
 216                         (?:
 217                             limelight:media:|
 218                             https?://
 219                                 (?:
 220                                     link\.videoplatform\.limelight\.com/media/|
 221                                     assets\.delvenetworks\.com/player/loader\.swf
 222                                 )
 223                                 \?.*?\bmediaId=
 224                         )
 225                         (?P<id>[a-z0-9]{32})
 226                     '''
 227     _TESTS = [{
 228         'url': 'http://link.videoplatform.limelight.com/media/?mediaId=3ffd040b522b4485b6d84effc750cd86',
 229         'info_dict': {
 230             'id': '3ffd040b522b4485b6d84effc750cd86',
 231             'ext': 'mp4',
 232             'title': 'HaP and the HB Prince Trailer',
 233             'description': 'md5:8005b944181778e313d95c1237ddb640',
 234             'thumbnail': r're:^https?://.*\.jpeg$',
 235             'duration': 144.23,
 236         },
 237         'params': {
 238             # m3u8 download
 239             'skip_download': True,
 240         },
 241     }, {
 242         # video with subtitles
 243         'url': 'limelight:media:a3e00274d4564ec4a9b29b9466432335',
 244         'md5': '2fa3bad9ac321e23860ca23bc2c69e3d',
 245         'info_dict': {
 246             'id': 'a3e00274d4564ec4a9b29b9466432335',
 247             'ext': 'mp4',
 248             'title': '3Play Media Overview Video',
 249             'thumbnail': r're:^https?://.*\.jpeg$',
 250             'duration': 78.101,
 251             # TODO: extract all languages that were accessible via API
 252             # 'subtitles': 'mincount:9',
 253             'subtitles': 'mincount:1',
 254         },
 255     }, {
 256         'url': 'https://assets.delvenetworks.com/player/loader.swf?mediaId=8018a574f08d416e95ceaccae4ba0452',
 257         'only_matching': True,
 258     }]
 259     _PLAYLIST_SERVICE_PATH = 'media'
 260
 261     def _real_extract(self, url):
 262         url, smuggled_data = unsmuggle_url(url, {})
 263         video_id = self._match_id(url)
 264         source_url = smuggled_data.get('source_url')
 265         self._initialize_geo_bypass({
 266             'countries': smuggled_data.get('geo_countries'),
 267         })
 268
 269         pc, mobile = self._extract(
 270             video_id, 'getPlaylistByMediaId',
 271             'getMobilePlaylistByMediaId', source_url)
 272
 273         return self._extract_info(pc, mobile, 0, source_url)
 274
 275
 276 class LimelightChannelIE(LimelightBaseIE):
 277     IE_NAME = 'limelight:channel'
 278     _VALID_URL = r'''(?x)
 279                         (?:
 280                             limelight:channel:|
 281                             https?://
 282                                 (?:
 283                                     link\.videoplatform\.limelight\.com/media/|
 284                                     assets\.delvenetworks\.com/player/loader\.swf
 285                                 )
 286                                 \?.*?\bchannelId=
 287                         )
 288                         (?P<id>[a-z0-9]{32})
 289                     '''
 290     _TESTS = [{
 291         'url': 'http://link.videoplatform.limelight.com/media/?channelId=ab6a524c379342f9b23642917020c082',
 292         'info_dict': {
 293             'id': 'ab6a524c379342f9b23642917020c082',
 294             'title': 'Javascript Sample Code',
 295             'description': 'Javascript Sample Code - http://www.delvenetworks.com/sample-code/playerCode-demo.html',
 296         },
 297         'playlist_mincount': 3,
 298     }, {
 299         'url': 'http://assets.delvenetworks.com/player/loader.swf?channelId=ab6a524c379342f9b23642917020c082',
 300         'only_matching': True,
 301     }]
 302     _PLAYLIST_SERVICE_PATH = 'channel'
 303
 304     def _real_extract(self, url):
 305         url, smuggled_data = unsmuggle_url(url, {})
 306         channel_id = self._match_id(url)
 307         source_url = smuggled_data.get('source_url')
 308
 309         pc, mobile = self._extract(
 310             channel_id, 'getPlaylistByChannelId',
 311             'getMobilePlaylistWithNItemsByChannelId?begin=0&count=-1',
 312             source_url)
 313
 314         entries = [
 315             self._extract_info(pc, mobile, i, source_url)
 316             for i in range(len(pc['playlistItems']))]
 317
 318         return self.playlist_result(
 319             entries, channel_id, pc.get('title'), mobile.get('description'))
 320
 321
 322 class LimelightChannelListIE(LimelightBaseIE):
 323     IE_NAME = 'limelight:channel_list'
 324     _VALID_URL = r'''(?x)
 325                         (?:
 326                             limelight:channel_list:|
 327                             https?://
 328                                 (?:
 329                                     link\.videoplatform\.limelight\.com/media/|
 330                                     assets\.delvenetworks\.com/player/loader\.swf
 331                                 )
 332                                 \?.*?\bchannelListId=
 333                         )
 334                         (?P<id>[a-z0-9]{32})
 335                     '''
 336     _TESTS = [{
 337         'url': 'http://link.videoplatform.limelight.com/media/?channelListId=301b117890c4465c8179ede21fd92e2b',
 338         'info_dict': {
 339             'id': '301b117890c4465c8179ede21fd92e2b',
 340             'title': 'Website - Hero Player',
 341         },
 342         'playlist_mincount': 2,
 343     }, {
 344         'url': 'https://assets.delvenetworks.com/player/loader.swf?channelListId=301b117890c4465c8179ede21fd92e2b',
 345         'only_matching': True,
 346     }]
 347     _PLAYLIST_SERVICE_PATH = 'channel_list'
 348
 349     def _real_extract(self, url):
 350         channel_list_id = self._match_id(url)
 351
 352         channel_list = self._call_playlist_service(
 353             channel_list_id, 'getMobileChannelListById')
 354
 355         entries = [
 356             self.url_result('limelight:channel:%s' % channel['id'], 'LimelightChannel')
 357             for channel in channel_list['channelList']]
 358
 359         return self.playlist_result(
 360             entries, channel_list_id, channel_list['title'])