yt_dlp/extractor/limelight.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..compat import compat_HTTPError
   8 from ..utils import (
   9     determine_ext,
  10     float_or_none,
  11     int_or_none,
  12     smuggle_url,
  13     try_get,
  14     unsmuggle_url,
  15     ExtractorError,
  16 )
  17
  18
  19 class LimelightBaseIE(InfoExtractor):
  20     _PLAYLIST_SERVICE_URL = 'http://production-ps.lvp.llnw.net/r/PlaylistService/%s/%s/%s'
  21
  22     @classmethod
  23     def _extract_urls(cls, webpage, source_url):
  24         lm = {
  25             'Media': 'media',
  26             'Channel': 'channel',
  27             'ChannelList': 'channel_list',
  28         }
  29
  30         def smuggle(url):
  31             return smuggle_url(url, {'source_url': source_url})
  32
  33         entries = []
  34         for kind, video_id in re.findall(
  35                 r'LimelightPlayer\.doLoad(Media|Channel|ChannelList)\(["\'](?P<id>[a-z0-9]{32})',
  36                 webpage):
  37             entries.append(cls.url_result(
  38                 smuggle('limelight:%s:%s' % (lm[kind], video_id)),
  39                 'Limelight%s' % kind, video_id))
  40         for mobj in re.finditer(
  41                 # As per [1] class attribute should be exactly equal to
  42                 # LimelightEmbeddedPlayerFlash but numerous examples seen
  43                 # that don't exactly match it (e.g. [2]).
  44                 # 1. http://support.3playmedia.com/hc/en-us/articles/227732408-Limelight-Embedding-the-Captions-Plugin-with-the-Limelight-Player-on-Your-Webpage
  45                 # 2. http://www.sedona.com/FacilitatorTraining2017
  46                 r'''(?sx)
  47                     <object[^>]+class=(["\'])(?:(?!\1).)*\bLimelightEmbeddedPlayerFlash\b(?:(?!\1).)*\1[^>]*>.*?
  48                         <param[^>]+
  49                             name=(["\'])flashVars\2[^>]+
  50                             value=(["\'])(?:(?!\3).)*(?P<kind>media|channel(?:List)?)Id=(?P<id>[a-z0-9]{32})
  51                 ''', webpage):
  52             kind, video_id = mobj.group('kind'), mobj.group('id')
  53             entries.append(cls.url_result(
  54                 smuggle('limelight:%s:%s' % (kind, video_id)),
  55                 'Limelight%s' % kind.capitalize(), video_id))
  56         # http://support.3playmedia.com/hc/en-us/articles/115009517327-Limelight-Embedding-the-Audio-Description-Plugin-with-the-Limelight-Player-on-Your-Web-Page)
  57         for video_id in re.findall(
  58                 r'(?s)LimelightPlayerUtil\.embed\s*\(\s*{.*?\bmediaId["\']\s*:\s*["\'](?P<id>[a-z0-9]{32})',
  59                 webpage):
  60             entries.append(cls.url_result(
  61                 smuggle('limelight:media:%s' % video_id),
  62                 LimelightMediaIE.ie_key(), video_id))
  63         return entries
  64
  65     def _call_playlist_service(self, item_id, method, fatal=True, referer=None):
  66         headers = {}
  67         if referer:
  68             headers['Referer'] = referer
  69         try:
  70             return self._download_json(
  71                 self._PLAYLIST_SERVICE_URL % (self._PLAYLIST_SERVICE_PATH, item_id, method),
  72                 item_id, 'Downloading PlaylistService %s JSON' % method,
  73                 fatal=fatal, headers=headers)
  74         except ExtractorError as e:
  75             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
  76                 error = self._parse_json(e.cause.read().decode(), item_id)['detail']['contentAccessPermission']
  77                 if error == 'CountryDisabled':
  78                     self.raise_geo_restricted()
  79                 raise ExtractorError(error, expected=True)
  80             raise
  81
  82     def _extract(self, item_id, pc_method, mobile_method, referer=None):
  83         pc = self._call_playlist_service(item_id, pc_method, referer=referer)
  84         mobile = self._call_playlist_service(
  85             item_id, mobile_method, fatal=False, referer=referer)
  86         return pc, mobile
  87
  88     def _extract_info(self, pc, mobile, i, referer):
  89         get_item = lambda x, y: try_get(x, lambda x: x[y][i], dict) or {}
  90         pc_item = get_item(pc, 'playlistItems')
  91         mobile_item = get_item(mobile, 'mediaList')
  92         video_id = pc_item.get('mediaId') or mobile_item['mediaId']
  93         title = pc_item.get('title') or mobile_item['title']
  94
  95         formats = []
  96         urls = []
  97         for stream in pc_item.get('streams', []):
  98             stream_url = stream.get('url')
  99             if not stream_url or stream_url in urls:
 100                 continue
 101             if not self.get_param('allow_unplayable_formats') and stream.get('drmProtected'):
 102                 continue
 103             urls.append(stream_url)
 104             ext = determine_ext(stream_url)
 105             if ext == 'f4m':
 106                 formats.extend(self._extract_f4m_formats(
 107                     stream_url, video_id, f4m_id='hds', fatal=False))
 108             else:
 109                 fmt = {
 110                     'url': stream_url,
 111                     'abr': float_or_none(stream.get('audioBitRate')),
 112                     'fps': float_or_none(stream.get('videoFrameRate')),
 113                     'ext': ext,
 114                 }
 115                 width = int_or_none(stream.get('videoWidthInPixels'))
 116                 height = int_or_none(stream.get('videoHeightInPixels'))
 117                 vbr = float_or_none(stream.get('videoBitRate'))
 118                 if width or height or vbr:
 119                     fmt.update({
 120                         'width': width,
 121                         'height': height,
 122                         'vbr': vbr,
 123                     })
 124                 else:
 125                     fmt['vcodec'] = 'none'
 126                 rtmp = re.search(r'^(?P<url>rtmpe?://(?P<host>[^/]+)/(?P<app>.+))/(?P<playpath>mp[34]:.+)$', stream_url)
 127                 if rtmp:
 128                     format_id = 'rtmp'
 129                     if stream.get('videoBitRate'):
 130                         format_id += '-%d' % int_or_none(stream['videoBitRate'])
 131                     http_format_id = format_id.replace('rtmp', 'http')
 132
 133                     CDN_HOSTS = (
 134                         ('delvenetworks.com', 'cpl.delvenetworks.com'),
 135                         ('video.llnw.net', 's2.content.video.llnw.net'),
 136                     )
 137                     for cdn_host, http_host in CDN_HOSTS:
 138                         if cdn_host not in rtmp.group('host').lower():
 139                             continue
 140                         http_url = 'http://%s/%s' % (http_host, rtmp.group('playpath')[4:])
 141                         urls.append(http_url)
 142                         if self._is_valid_url(http_url, video_id, http_format_id):
 143                             http_fmt = fmt.copy()
 144                             http_fmt.update({
 145                                 'url': http_url,
 146                                 'format_id': http_format_id,
 147                             })
 148                             formats.append(http_fmt)
 149                             break
 150
 151                     fmt.update({
 152                         'url': rtmp.group('url'),
 153                         'play_path': rtmp.group('playpath'),
 154                         'app': rtmp.group('app'),
 155                         'ext': 'flv',
 156                         'format_id': format_id,
 157                     })
 158                 formats.append(fmt)
 159
 160         for mobile_url in mobile_item.get('mobileUrls', []):
 161             media_url = mobile_url.get('mobileUrl')
 162             format_id = mobile_url.get('targetMediaPlatform')
 163             if not media_url or media_url in urls:
 164                 continue
 165             if (format_id in ('Widevine', 'SmoothStreaming')
 166                     and not self.get_param('allow_unplayable_formats', False)):
 167                 continue
 168             urls.append(media_url)
 169             ext = determine_ext(media_url)
 170             if ext == 'm3u8':
 171                 formats.extend(self._extract_m3u8_formats(
 172                     media_url, video_id, 'mp4', 'm3u8_native',
 173                     m3u8_id=format_id, fatal=False))
 174             elif ext == 'f4m':
 175                 formats.extend(self._extract_f4m_formats(
 176                     stream_url, video_id, f4m_id=format_id, fatal=False))
 177             else:
 178                 formats.append({
 179                     'url': media_url,
 180                     'format_id': format_id,
 181                     'quality': -10,
 182                     'ext': ext,
 183                 })
 184
 185         self._sort_formats(formats)
 186
 187         subtitles = {}
 188         for flag in mobile_item.get('flags'):
 189             if flag == 'ClosedCaptions':
 190                 closed_captions = self._call_playlist_service(
 191                     video_id, 'getClosedCaptionsDetailsByMediaId',
 192                     False, referer) or []
 193                 for cc in closed_captions:
 194                     cc_url = cc.get('webvttFileUrl')
 195                     if not cc_url:
 196                         continue
 197                     lang = cc.get('languageCode') or self._search_regex(r'/[a-z]{2}\.vtt', cc_url, 'lang', default='en')
 198                     subtitles.setdefault(lang, []).append({
 199                         'url': cc_url,
 200                     })
 201                 break
 202
 203         get_meta = lambda x: pc_item.get(x) or mobile_item.get(x)
 204
 205         return {
 206             'id': video_id,
 207             'title': title,
 208             'description': get_meta('description'),
 209             'formats': formats,
 210             'duration': float_or_none(get_meta('durationInMilliseconds'), 1000),
 211             'thumbnail': get_meta('previewImageUrl') or get_meta('thumbnailImageUrl'),
 212             'subtitles': subtitles,
 213         }
 214
 215
 216 class LimelightMediaIE(LimelightBaseIE):
 217     IE_NAME = 'limelight'
 218     _VALID_URL = r'''(?x)
 219                         (?:
 220                             limelight:media:|
 221                             https?://
 222                                 (?:
 223                                     link\.videoplatform\.limelight\.com/media/|
 224                                     assets\.delvenetworks\.com/player/loader\.swf
 225                                 )
 226                                 \?.*?\bmediaId=
 227                         )
 228                         (?P<id>[a-z0-9]{32})
 229                     '''
 230     _TESTS = [{
 231         'url': 'http://link.videoplatform.limelight.com/media/?mediaId=3ffd040b522b4485b6d84effc750cd86',
 232         'info_dict': {
 233             'id': '3ffd040b522b4485b6d84effc750cd86',
 234             'ext': 'mp4',
 235             'title': 'HaP and the HB Prince Trailer',
 236             'description': 'md5:8005b944181778e313d95c1237ddb640',
 237             'thumbnail': r're:^https?://.*\.jpeg$',
 238             'duration': 144.23,
 239         },
 240         'params': {
 241             # m3u8 download
 242             'skip_download': True,
 243         },
 244     }, {
 245         # video with subtitles
 246         'url': 'limelight:media:a3e00274d4564ec4a9b29b9466432335',
 247         'md5': '2fa3bad9ac321e23860ca23bc2c69e3d',
 248         'info_dict': {
 249             'id': 'a3e00274d4564ec4a9b29b9466432335',
 250             'ext': 'mp4',
 251             'title': '3Play Media Overview Video',
 252             'thumbnail': r're:^https?://.*\.jpeg$',
 253             'duration': 78.101,
 254             # TODO: extract all languages that were accessible via API
 255             # 'subtitles': 'mincount:9',
 256             'subtitles': 'mincount:1',
 257         },
 258     }, {
 259         'url': 'https://assets.delvenetworks.com/player/loader.swf?mediaId=8018a574f08d416e95ceaccae4ba0452',
 260         'only_matching': True,
 261     }]
 262     _PLAYLIST_SERVICE_PATH = 'media'
 263
 264     def _real_extract(self, url):
 265         url, smuggled_data = unsmuggle_url(url, {})
 266         video_id = self._match_id(url)
 267         source_url = smuggled_data.get('source_url')
 268         self._initialize_geo_bypass({
 269             'countries': smuggled_data.get('geo_countries'),
 270         })
 271
 272         pc, mobile = self._extract(
 273             video_id, 'getPlaylistByMediaId',
 274             'getMobilePlaylistByMediaId', source_url)
 275
 276         return self._extract_info(pc, mobile, 0, source_url)
 277
 278
 279 class LimelightChannelIE(LimelightBaseIE):
 280     IE_NAME = 'limelight:channel'
 281     _VALID_URL = r'''(?x)
 282                         (?:
 283                             limelight:channel:|
 284                             https?://
 285                                 (?:
 286                                     link\.videoplatform\.limelight\.com/media/|
 287                                     assets\.delvenetworks\.com/player/loader\.swf
 288                                 )
 289                                 \?.*?\bchannelId=
 290                         )
 291                         (?P<id>[a-z0-9]{32})
 292                     '''
 293     _TESTS = [{
 294         'url': 'http://link.videoplatform.limelight.com/media/?channelId=ab6a524c379342f9b23642917020c082',
 295         'info_dict': {
 296             'id': 'ab6a524c379342f9b23642917020c082',
 297             'title': 'Javascript Sample Code',
 298             'description': 'Javascript Sample Code - http://www.delvenetworks.com/sample-code/playerCode-demo.html',
 299         },
 300         'playlist_mincount': 3,
 301     }, {
 302         'url': 'http://assets.delvenetworks.com/player/loader.swf?channelId=ab6a524c379342f9b23642917020c082',
 303         'only_matching': True,
 304     }]
 305     _PLAYLIST_SERVICE_PATH = 'channel'
 306
 307     def _real_extract(self, url):
 308         url, smuggled_data = unsmuggle_url(url, {})
 309         channel_id = self._match_id(url)
 310         source_url = smuggled_data.get('source_url')
 311
 312         pc, mobile = self._extract(
 313             channel_id, 'getPlaylistByChannelId',
 314             'getMobilePlaylistWithNItemsByChannelId?begin=0&count=-1',
 315             source_url)
 316
 317         entries = [
 318             self._extract_info(pc, mobile, i, source_url)
 319             for i in range(len(pc['playlistItems']))]
 320
 321         return self.playlist_result(
 322             entries, channel_id, pc.get('title'), mobile.get('description'))
 323
 324
 325 class LimelightChannelListIE(LimelightBaseIE):
 326     IE_NAME = 'limelight:channel_list'
 327     _VALID_URL = r'''(?x)
 328                         (?:
 329                             limelight:channel_list:|
 330                             https?://
 331                                 (?:
 332                                     link\.videoplatform\.limelight\.com/media/|
 333                                     assets\.delvenetworks\.com/player/loader\.swf
 334                                 )
 335                                 \?.*?\bchannelListId=
 336                         )
 337                         (?P<id>[a-z0-9]{32})
 338                     '''
 339     _TESTS = [{
 340         'url': 'http://link.videoplatform.limelight.com/media/?channelListId=301b117890c4465c8179ede21fd92e2b',
 341         'info_dict': {
 342             'id': '301b117890c4465c8179ede21fd92e2b',
 343             'title': 'Website - Hero Player',
 344         },
 345         'playlist_mincount': 2,
 346     }, {
 347         'url': 'https://assets.delvenetworks.com/player/loader.swf?channelListId=301b117890c4465c8179ede21fd92e2b',
 348         'only_matching': True,
 349     }]
 350     _PLAYLIST_SERVICE_PATH = 'channel_list'
 351
 352     def _real_extract(self, url):
 353         channel_list_id = self._match_id(url)
 354
 355         channel_list = self._call_playlist_service(
 356             channel_list_id, 'getMobileChannelListById')
 357
 358         entries = [
 359             self.url_result('limelight:channel:%s' % channel['id'], 'LimelightChannel')
 360             for channel in channel_list['channelList']]
 361
 362         return self.playlist_result(
 363             entries, channel_list_id, channel_list['title'])