_ Git - youtube-dl/blob - youtube_dl/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     unsmuggle_url,
  14     ExtractorError,
  15 )
  16
  17
  18 class LimelightBaseIE(InfoExtractor):
  19     _PLAYLIST_SERVICE_URL = 'http://production-ps.lvp.llnw.net/r/PlaylistService/%s/%s/%s'
  20     _API_URL = 'http://api.video.limelight.com/rest/organizations/%s/%s/%s/%s.json'
  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, fatal=fatal, headers=headers)
  73         except ExtractorError as e:
  74             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
  75                 error = self._parse_json(e.cause.read().decode(), item_id)['detail']['contentAccessPermission']
  76                 if error == 'CountryDisabled':
  77                     self.raise_geo_restricted()
  78                 raise ExtractorError(error, expected=True)
  79             raise
  80
  81     def _call_api(self, organization_id, item_id, method):
  82         return self._download_json(
  83             self._API_URL % (organization_id, self._API_PATH, item_id, method),
  84             item_id, 'Downloading API %s JSON' % method)
  85
  86     def _extract(self, item_id, pc_method, mobile_method, meta_method, referer=None):
  87         pc = self._call_playlist_service(item_id, pc_method, referer=referer)
  88         metadata = self._call_api(pc['orgId'], item_id, meta_method)
  89         mobile = self._call_playlist_service(item_id, mobile_method, fatal=False, referer=referer)
  90         return pc, mobile, metadata
  91
  92     def _extract_info(self, streams, mobile_urls, properties):
  93         video_id = properties['media_id']
  94         formats = []
  95         urls = []
  96         for stream in streams:
  97             stream_url = stream.get('url')
  98             if not stream_url or stream.get('drmProtected') or stream_url in urls:
  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_urls:
 158             media_url = mobile_url.get('mobileUrl')
 159             format_id = mobile_url.get('targetMediaPlatform')
 160             if not media_url or format_id in ('Widevine', 'SmoothStreaming') or media_url in urls:
 161                 continue
 162             urls.append(media_url)
 163             ext = determine_ext(media_url)
 164             if ext == 'm3u8':
 165                 formats.extend(self._extract_m3u8_formats(
 166                     media_url, video_id, 'mp4', 'm3u8_native',
 167                     m3u8_id=format_id, fatal=False))
 168             elif ext == 'f4m':
 169                 formats.extend(self._extract_f4m_formats(
 170                     stream_url, video_id, f4m_id=format_id, fatal=False))
 171             else:
 172                 formats.append({
 173                     'url': media_url,
 174                     'format_id': format_id,
 175                     'preference': -1,
 176                     'ext': ext,
 177                 })
 178
 179         self._sort_formats(formats)
 180
 181         title = properties['title']
 182         description = properties.get('description')
 183         timestamp = int_or_none(properties.get('publish_date') or properties.get('create_date'))
 184         duration = float_or_none(properties.get('duration_in_milliseconds'), 1000)
 185         filesize = int_or_none(properties.get('total_storage_in_bytes'))
 186         categories = [properties.get('category')]
 187         tags = properties.get('tags', [])
 188         thumbnails = [{
 189             'url': thumbnail['url'],
 190             'width': int_or_none(thumbnail.get('width')),
 191             'height': int_or_none(thumbnail.get('height')),
 192         } for thumbnail in properties.get('thumbnails', []) if thumbnail.get('url')]
 193
 194         subtitles = {}
 195         for caption in properties.get('captions', []):
 196             lang = caption.get('language_code')
 197             subtitles_url = caption.get('url')
 198             if lang and subtitles_url:
 199                 subtitles.setdefault(lang, []).append({
 200                     'url': subtitles_url,
 201                 })
 202         closed_captions_url = properties.get('closed_captions_url')
 203         if closed_captions_url:
 204             subtitles.setdefault('en', []).append({
 205                 'url': closed_captions_url,
 206                 'ext': 'ttml',
 207             })
 208
 209         return {
 210             'id': video_id,
 211             'title': title,
 212             'description': description,
 213             'formats': formats,
 214             'timestamp': timestamp,
 215             'duration': duration,
 216             'filesize': filesize,
 217             'categories': categories,
 218             'tags': tags,
 219             'thumbnails': thumbnails,
 220             'subtitles': subtitles,
 221         }
 222
 223
 224 class LimelightMediaIE(LimelightBaseIE):
 225     IE_NAME = 'limelight'
 226     _VALID_URL = r'''(?x)
 227                         (?:
 228                             limelight:media:|
 229                             https?://
 230                                 (?:
 231                                     link\.videoplatform\.limelight\.com/media/|
 232                                     assets\.delvenetworks\.com/player/loader\.swf
 233                                 )
 234                                 \?.*?\bmediaId=
 235                         )
 236                         (?P<id>[a-z0-9]{32})
 237                     '''
 238     _TESTS = [{
 239         'url': 'http://link.videoplatform.limelight.com/media/?mediaId=3ffd040b522b4485b6d84effc750cd86',
 240         'info_dict': {
 241             'id': '3ffd040b522b4485b6d84effc750cd86',
 242             'ext': 'mp4',
 243             'title': 'HaP and the HB Prince Trailer',
 244             'description': 'md5:8005b944181778e313d95c1237ddb640',
 245             'thumbnail': r're:^https?://.*\.jpeg$',
 246             'duration': 144.23,
 247             'timestamp': 1244136834,
 248             'upload_date': '20090604',
 249         },
 250         'params': {
 251             # m3u8 download
 252             'skip_download': True,
 253         },
 254     }, {
 255         # video with subtitles
 256         'url': 'limelight:media:a3e00274d4564ec4a9b29b9466432335',
 257         'md5': '2fa3bad9ac321e23860ca23bc2c69e3d',
 258         'info_dict': {
 259             'id': 'a3e00274d4564ec4a9b29b9466432335',
 260             'ext': 'mp4',
 261             'title': '3Play Media Overview Video',
 262             'thumbnail': r're:^https?://.*\.jpeg$',
 263             'duration': 78.101,
 264             'timestamp': 1338929955,
 265             'upload_date': '20120605',
 266             'subtitles': 'mincount:9',
 267         },
 268     }, {
 269         'url': 'https://assets.delvenetworks.com/player/loader.swf?mediaId=8018a574f08d416e95ceaccae4ba0452',
 270         'only_matching': True,
 271     }]
 272     _PLAYLIST_SERVICE_PATH = 'media'
 273     _API_PATH = 'media'
 274
 275     def _real_extract(self, url):
 276         url, smuggled_data = unsmuggle_url(url, {})
 277         video_id = self._match_id(url)
 278         self._initialize_geo_bypass(smuggled_data.get('geo_countries'))
 279
 280         pc, mobile, metadata = self._extract(
 281             video_id, 'getPlaylistByMediaId',
 282             'getMobilePlaylistByMediaId', 'properties',
 283             smuggled_data.get('source_url'))
 284
 285         return self._extract_info(
 286             pc['playlistItems'][0].get('streams', []),
 287             mobile['mediaList'][0].get('mobileUrls', []) if mobile else [],
 288             metadata)
 289
 290
 291 class LimelightChannelIE(LimelightBaseIE):
 292     IE_NAME = 'limelight:channel'
 293     _VALID_URL = r'''(?x)
 294                         (?:
 295                             limelight:channel:|
 296                             https?://
 297                                 (?:
 298                                     link\.videoplatform\.limelight\.com/media/|
 299                                     assets\.delvenetworks\.com/player/loader\.swf
 300                                 )
 301                                 \?.*?\bchannelId=
 302                         )
 303                         (?P<id>[a-z0-9]{32})
 304                     '''
 305     _TESTS = [{
 306         'url': 'http://link.videoplatform.limelight.com/media/?channelId=ab6a524c379342f9b23642917020c082',
 307         'info_dict': {
 308             'id': 'ab6a524c379342f9b23642917020c082',
 309             'title': 'Javascript Sample Code',
 310         },
 311         'playlist_mincount': 3,
 312     }, {
 313         'url': 'http://assets.delvenetworks.com/player/loader.swf?channelId=ab6a524c379342f9b23642917020c082',
 314         'only_matching': True,
 315     }]
 316     _PLAYLIST_SERVICE_PATH = 'channel'
 317     _API_PATH = 'channels'
 318
 319     def _real_extract(self, url):
 320         url, smuggled_data = unsmuggle_url(url, {})
 321         channel_id = self._match_id(url)
 322
 323         pc, mobile, medias = self._extract(
 324             channel_id, 'getPlaylistByChannelId',
 325             'getMobilePlaylistWithNItemsByChannelId?begin=0&count=-1',
 326             'media', smuggled_data.get('source_url'))
 327
 328         entries = [
 329             self._extract_info(
 330                 pc['playlistItems'][i].get('streams', []),
 331                 mobile['mediaList'][i].get('mobileUrls', []) if mobile else [],
 332                 medias['media_list'][i])
 333             for i in range(len(medias['media_list']))]
 334
 335         return self.playlist_result(entries, channel_id, pc['title'])
 336
 337
 338 class LimelightChannelListIE(LimelightBaseIE):
 339     IE_NAME = 'limelight:channel_list'
 340     _VALID_URL = r'''(?x)
 341                         (?:
 342                             limelight:channel_list:|
 343                             https?://
 344                                 (?:
 345                                     link\.videoplatform\.limelight\.com/media/|
 346                                     assets\.delvenetworks\.com/player/loader\.swf
 347                                 )
 348                                 \?.*?\bchannelListId=
 349                         )
 350                         (?P<id>[a-z0-9]{32})
 351                     '''
 352     _TESTS = [{
 353         'url': 'http://link.videoplatform.limelight.com/media/?channelListId=301b117890c4465c8179ede21fd92e2b',
 354         'info_dict': {
 355             'id': '301b117890c4465c8179ede21fd92e2b',
 356             'title': 'Website - Hero Player',
 357         },
 358         'playlist_mincount': 2,
 359     }, {
 360         'url': 'https://assets.delvenetworks.com/player/loader.swf?channelListId=301b117890c4465c8179ede21fd92e2b',
 361         'only_matching': True,
 362     }]
 363     _PLAYLIST_SERVICE_PATH = 'channel_list'
 364
 365     def _real_extract(self, url):
 366         channel_list_id = self._match_id(url)
 367
 368         channel_list = self._call_playlist_service(channel_list_id, 'getMobileChannelListById')
 369
 370         entries = [
 371             self.url_result('limelight:channel:%s' % channel['id'], 'LimelightChannel')
 372             for channel in channel_list['channelList']]
 373
 374         return self.playlist_result(entries, channel_list_id, channel_list['title'])