_ 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         entries = []
  30         for kind, video_id in re.findall(
  31                 r'LimelightPlayer\.doLoad(Media|Channel|ChannelList)\(["\'](?P<id>[a-z0-9]{32})',
  32                 webpage):
  33             entries.append(cls.url_result(
  34                 smuggle_url(
  35                     'limelight:%s:%s' % (lm[kind], video_id),
  36                     {'source_url': source_url}),
  37                 'Limelight%s' % kind, video_id))
  38         for mobj in re.finditer(
  39                 # As per [1] class attribute should be exactly equal to
  40                 # LimelightEmbeddedPlayerFlash but numerous examples seen
  41                 # that don't exactly match it (e.g. [2]).
  42                 # 1. http://support.3playmedia.com/hc/en-us/articles/227732408-Limelight-Embedding-the-Captions-Plugin-with-the-Limelight-Player-on-Your-Webpage
  43                 # 2. http://www.sedona.com/FacilitatorTraining2017
  44                 r'''(?sx)
  45                     <object[^>]+class=(["\'])(?:(?!\1).)*\bLimelightEmbeddedPlayerFlash\b(?:(?!\1).)*\1[^>]*>.*?
  46                         <param[^>]+
  47                             name=(["\'])flashVars\2[^>]+
  48                             value=(["\'])(?:(?!\3).)*(?P<kind>media|channel(?:List)?)Id=(?P<id>[a-z0-9]{32})
  49                 ''', webpage):
  50             kind, video_id = mobj.group('kind'), mobj.group('id')
  51             entries.append(cls.url_result(
  52                 smuggle_url(
  53                     'limelight:%s:%s' % (kind, video_id),
  54                     {'source_url': source_url}),
  55                 'Limelight%s' % kind.capitalize(), video_id))
  56         return entries
  57
  58     def _call_playlist_service(self, item_id, method, fatal=True, referer=None):
  59         headers = {}
  60         if referer:
  61             headers['Referer'] = referer
  62         try:
  63             return self._download_json(
  64                 self._PLAYLIST_SERVICE_URL % (self._PLAYLIST_SERVICE_PATH, item_id, method),
  65                 item_id, 'Downloading PlaylistService %s JSON' % method, fatal=fatal, headers=headers)
  66         except ExtractorError as e:
  67             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
  68                 error = self._parse_json(e.cause.read().decode(), item_id)['detail']['contentAccessPermission']
  69                 if error == 'CountryDisabled':
  70                     self.raise_geo_restricted()
  71                 raise ExtractorError(error, expected=True)
  72             raise
  73
  74     def _call_api(self, organization_id, item_id, method):
  75         return self._download_json(
  76             self._API_URL % (organization_id, self._API_PATH, item_id, method),
  77             item_id, 'Downloading API %s JSON' % method)
  78
  79     def _extract(self, item_id, pc_method, mobile_method, meta_method, referer=None):
  80         pc = self._call_playlist_service(item_id, pc_method, referer=referer)
  81         metadata = self._call_api(pc['orgId'], item_id, meta_method)
  82         mobile = self._call_playlist_service(item_id, mobile_method, fatal=False, referer=referer)
  83         return pc, mobile, metadata
  84
  85     def _extract_info(self, streams, mobile_urls, properties):
  86         video_id = properties['media_id']
  87         formats = []
  88         urls = []
  89         for stream in streams:
  90             stream_url = stream.get('url')
  91             if not stream_url or stream.get('drmProtected') or stream_url in urls:
  92                 continue
  93             urls.append(stream_url)
  94             ext = determine_ext(stream_url)
  95             if ext == 'f4m':
  96                 formats.extend(self._extract_f4m_formats(
  97                     stream_url, video_id, f4m_id='hds', fatal=False))
  98             else:
  99                 fmt = {
 100                     'url': stream_url,
 101                     'abr': float_or_none(stream.get('audioBitRate')),
 102                     'fps': float_or_none(stream.get('videoFrameRate')),
 103                     'ext': ext,
 104                 }
 105                 width = int_or_none(stream.get('videoWidthInPixels'))
 106                 height = int_or_none(stream.get('videoHeightInPixels'))
 107                 vbr = float_or_none(stream.get('videoBitRate'))
 108                 if width or height or vbr:
 109                     fmt.update({
 110                         'width': width,
 111                         'height': height,
 112                         'vbr': vbr,
 113                     })
 114                 else:
 115                     fmt['vcodec'] = 'none'
 116                 rtmp = re.search(r'^(?P<url>rtmpe?://(?P<host>[^/]+)/(?P<app>.+))/(?P<playpath>mp[34]:.+)$', stream_url)
 117                 if rtmp:
 118                     format_id = 'rtmp'
 119                     if stream.get('videoBitRate'):
 120                         format_id += '-%d' % int_or_none(stream['videoBitRate'])
 121                     http_format_id = format_id.replace('rtmp', 'http')
 122
 123                     CDN_HOSTS = (
 124                         ('delvenetworks.com', 'cpl.delvenetworks.com'),
 125                         ('video.llnw.net', 's2.content.video.llnw.net'),
 126                     )
 127                     for cdn_host, http_host in CDN_HOSTS:
 128                         if cdn_host not in rtmp.group('host').lower():
 129                             continue
 130                         http_url = 'http://%s/%s' % (http_host, rtmp.group('playpath')[4:])
 131                         urls.append(http_url)
 132                         if self._is_valid_url(http_url, video_id, http_format_id):
 133                             http_fmt = fmt.copy()
 134                             http_fmt.update({
 135                                 'url': http_url,
 136                                 'format_id': http_format_id,
 137                             })
 138                             formats.append(http_fmt)
 139                             break
 140
 141                     fmt.update({
 142                         'url': rtmp.group('url'),
 143                         'play_path': rtmp.group('playpath'),
 144                         'app': rtmp.group('app'),
 145                         'ext': 'flv',
 146                         'format_id': format_id,
 147                     })
 148                 formats.append(fmt)
 149
 150         for mobile_url in mobile_urls:
 151             media_url = mobile_url.get('mobileUrl')
 152             format_id = mobile_url.get('targetMediaPlatform')
 153             if not media_url or format_id in ('Widevine', 'SmoothStreaming') or media_url in urls:
 154                 continue
 155             urls.append(media_url)
 156             ext = determine_ext(media_url)
 157             if ext == 'm3u8':
 158                 formats.extend(self._extract_m3u8_formats(
 159                     media_url, video_id, 'mp4', 'm3u8_native',
 160                     m3u8_id=format_id, fatal=False))
 161             elif ext == 'f4m':
 162                 formats.extend(self._extract_f4m_formats(
 163                     stream_url, video_id, f4m_id=format_id, fatal=False))
 164             else:
 165                 formats.append({
 166                     'url': media_url,
 167                     'format_id': format_id,
 168                     'preference': -1,
 169                     'ext': ext,
 170                 })
 171
 172         self._sort_formats(formats)
 173
 174         title = properties['title']
 175         description = properties.get('description')
 176         timestamp = int_or_none(properties.get('publish_date') or properties.get('create_date'))
 177         duration = float_or_none(properties.get('duration_in_milliseconds'), 1000)
 178         filesize = int_or_none(properties.get('total_storage_in_bytes'))
 179         categories = [properties.get('category')]
 180         tags = properties.get('tags', [])
 181         thumbnails = [{
 182             'url': thumbnail['url'],
 183             'width': int_or_none(thumbnail.get('width')),
 184             'height': int_or_none(thumbnail.get('height')),
 185         } for thumbnail in properties.get('thumbnails', []) if thumbnail.get('url')]
 186
 187         subtitles = {}
 188         for caption in properties.get('captions', []):
 189             lang = caption.get('language_code')
 190             subtitles_url = caption.get('url')
 191             if lang and subtitles_url:
 192                 subtitles.setdefault(lang, []).append({
 193                     'url': subtitles_url,
 194                 })
 195         closed_captions_url = properties.get('closed_captions_url')
 196         if closed_captions_url:
 197             subtitles.setdefault('en', []).append({
 198                 'url': closed_captions_url,
 199                 'ext': 'ttml',
 200             })
 201
 202         return {
 203             'id': video_id,
 204             'title': title,
 205             'description': description,
 206             'formats': formats,
 207             'timestamp': timestamp,
 208             'duration': duration,
 209             'filesize': filesize,
 210             'categories': categories,
 211             'tags': tags,
 212             'thumbnails': thumbnails,
 213             'subtitles': subtitles,
 214         }
 215
 216
 217 class LimelightMediaIE(LimelightBaseIE):
 218     IE_NAME = 'limelight'
 219     _VALID_URL = r'''(?x)
 220                         (?:
 221                             limelight:media:|
 222                             https?://
 223                                 (?:
 224                                     link\.videoplatform\.limelight\.com/media/|
 225                                     assets\.delvenetworks\.com/player/loader\.swf
 226                                 )
 227                                 \?.*?\bmediaId=
 228                         )
 229                         (?P<id>[a-z0-9]{32})
 230                     '''
 231     _TESTS = [{
 232         'url': 'http://link.videoplatform.limelight.com/media/?mediaId=3ffd040b522b4485b6d84effc750cd86',
 233         'info_dict': {
 234             'id': '3ffd040b522b4485b6d84effc750cd86',
 235             'ext': 'mp4',
 236             'title': 'HaP and the HB Prince Trailer',
 237             'description': 'md5:8005b944181778e313d95c1237ddb640',
 238             'thumbnail': r're:^https?://.*\.jpeg$',
 239             'duration': 144.23,
 240             'timestamp': 1244136834,
 241             'upload_date': '20090604',
 242         },
 243         'params': {
 244             # m3u8 download
 245             'skip_download': True,
 246         },
 247     }, {
 248         # video with subtitles
 249         'url': 'limelight:media:a3e00274d4564ec4a9b29b9466432335',
 250         'md5': '2fa3bad9ac321e23860ca23bc2c69e3d',
 251         'info_dict': {
 252             'id': 'a3e00274d4564ec4a9b29b9466432335',
 253             'ext': 'mp4',
 254             'title': '3Play Media Overview Video',
 255             'thumbnail': r're:^https?://.*\.jpeg$',
 256             'duration': 78.101,
 257             'timestamp': 1338929955,
 258             'upload_date': '20120605',
 259             'subtitles': 'mincount:9',
 260         },
 261     }, {
 262         'url': 'https://assets.delvenetworks.com/player/loader.swf?mediaId=8018a574f08d416e95ceaccae4ba0452',
 263         'only_matching': True,
 264     }]
 265     _PLAYLIST_SERVICE_PATH = 'media'
 266     _API_PATH = 'media'
 267
 268     def _real_extract(self, url):
 269         url, smuggled_data = unsmuggle_url(url, {})
 270         video_id = self._match_id(url)
 271         self._initialize_geo_bypass(smuggled_data.get('geo_countries'))
 272
 273         pc, mobile, metadata = self._extract(
 274             video_id, 'getPlaylistByMediaId',
 275             'getMobilePlaylistByMediaId', 'properties',
 276             smuggled_data.get('source_url'))
 277
 278         return self._extract_info(
 279             pc['playlistItems'][0].get('streams', []),
 280             mobile['mediaList'][0].get('mobileUrls', []) if mobile else [],
 281             metadata)
 282
 283
 284 class LimelightChannelIE(LimelightBaseIE):
 285     IE_NAME = 'limelight:channel'
 286     _VALID_URL = r'''(?x)
 287                         (?:
 288                             limelight:channel:|
 289                             https?://
 290                                 (?:
 291                                     link\.videoplatform\.limelight\.com/media/|
 292                                     assets\.delvenetworks\.com/player/loader\.swf
 293                                 )
 294                                 \?.*?\bchannelId=
 295                         )
 296                         (?P<id>[a-z0-9]{32})
 297                     '''
 298     _TESTS = [{
 299         'url': 'http://link.videoplatform.limelight.com/media/?channelId=ab6a524c379342f9b23642917020c082',
 300         'info_dict': {
 301             'id': 'ab6a524c379342f9b23642917020c082',
 302             'title': 'Javascript Sample Code',
 303         },
 304         'playlist_mincount': 3,
 305     }, {
 306         'url': 'http://assets.delvenetworks.com/player/loader.swf?channelId=ab6a524c379342f9b23642917020c082',
 307         'only_matching': True,
 308     }]
 309     _PLAYLIST_SERVICE_PATH = 'channel'
 310     _API_PATH = 'channels'
 311
 312     def _real_extract(self, url):
 313         url, smuggled_data = unsmuggle_url(url, {})
 314         channel_id = self._match_id(url)
 315
 316         pc, mobile, medias = self._extract(
 317             channel_id, 'getPlaylistByChannelId',
 318             'getMobilePlaylistWithNItemsByChannelId?begin=0&count=-1',
 319             'media', smuggled_data.get('source_url'))
 320
 321         entries = [
 322             self._extract_info(
 323                 pc['playlistItems'][i].get('streams', []),
 324                 mobile['mediaList'][i].get('mobileUrls', []) if mobile else [],
 325                 medias['media_list'][i])
 326             for i in range(len(medias['media_list']))]
 327
 328         return self.playlist_result(entries, channel_id, pc['title'])
 329
 330
 331 class LimelightChannelListIE(LimelightBaseIE):
 332     IE_NAME = 'limelight:channel_list'
 333     _VALID_URL = r'''(?x)
 334                         (?:
 335                             limelight:channel_list:|
 336                             https?://
 337                                 (?:
 338                                     link\.videoplatform\.limelight\.com/media/|
 339                                     assets\.delvenetworks\.com/player/loader\.swf
 340                                 )
 341                                 \?.*?\bchannelListId=
 342                         )
 343                         (?P<id>[a-z0-9]{32})
 344                     '''
 345     _TESTS = [{
 346         'url': 'http://link.videoplatform.limelight.com/media/?channelListId=301b117890c4465c8179ede21fd92e2b',
 347         'info_dict': {
 348             'id': '301b117890c4465c8179ede21fd92e2b',
 349             'title': 'Website - Hero Player',
 350         },
 351         'playlist_mincount': 2,
 352     }, {
 353         'url': 'https://assets.delvenetworks.com/player/loader.swf?channelListId=301b117890c4465c8179ede21fd92e2b',
 354         'only_matching': True,
 355     }]
 356     _PLAYLIST_SERVICE_PATH = 'channel_list'
 357
 358     def _real_extract(self, url):
 359         channel_list_id = self._match_id(url)
 360
 361         channel_list = self._call_playlist_service(channel_list_id, 'getMobileChannelListById')
 362
 363         entries = [
 364             self.url_result('limelight:channel:%s' % channel['id'], 'LimelightChannel')
 365             for channel in channel_list['channelList']]
 366
 367         return self.playlist_result(entries, channel_list_id, channel_list['title'])