Raphaël G. Git Repositories - youtubedl/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     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     _API_URL = 'http://api.video.limelight.com/rest/organizations/%s/%s/%s/%s.json'
  22
  23     @classmethod
  24     def _extract_urls(cls, webpage, source_url):
  25         lm = {
  26             'Media': 'media',
  27             'Channel': 'channel',
  28             'ChannelList': 'channel_list',
  29         }
  30
  31         def smuggle(url):
  32             return smuggle_url(url, {'source_url': source_url})
  33
  34         entries = []
  35         for kind, video_id in re.findall(
  36                 r'LimelightPlayer\.doLoad(Media|Channel|ChannelList)\(["\'](?P<id>[a-z0-9]{32})',
  37                 webpage):
  38             entries.append(cls.url_result(
  39                 smuggle('limelight:%s:%s' % (lm[kind], video_id)),
  40                 'Limelight%s' % kind, video_id))
  41         for mobj in re.finditer(
  42                 # As per [1] class attribute should be exactly equal to
  43                 # LimelightEmbeddedPlayerFlash but numerous examples seen
  44                 # that don't exactly match it (e.g. [2]).
  45                 # 1. http://support.3playmedia.com/hc/en-us/articles/227732408-Limelight-Embedding-the-Captions-Plugin-with-the-Limelight-Player-on-Your-Webpage
  46                 # 2. http://www.sedona.com/FacilitatorTraining2017
  47                 r'''(?sx)
  48                     <object[^>]+class=(["\'])(?:(?!\1).)*\bLimelightEmbeddedPlayerFlash\b(?:(?!\1).)*\1[^>]*>.*?
  49                         <param[^>]+
  50                             name=(["\'])flashVars\2[^>]+
  51                             value=(["\'])(?:(?!\3).)*(?P<kind>media|channel(?:List)?)Id=(?P<id>[a-z0-9]{32})
  52                 ''', webpage):
  53             kind, video_id = mobj.group('kind'), mobj.group('id')
  54             entries.append(cls.url_result(
  55                 smuggle('limelight:%s:%s' % (kind, video_id)),
  56                 'Limelight%s' % kind.capitalize(), video_id))
  57         # http://support.3playmedia.com/hc/en-us/articles/115009517327-Limelight-Embedding-the-Audio-Description-Plugin-with-the-Limelight-Player-on-Your-Web-Page)
  58         for video_id in re.findall(
  59                 r'(?s)LimelightPlayerUtil\.embed\s*\(\s*{.*?\bmediaId["\']\s*:\s*["\'](?P<id>[a-z0-9]{32})',
  60                 webpage):
  61             entries.append(cls.url_result(
  62                 smuggle('limelight:media:%s' % video_id),
  63                 LimelightMediaIE.ie_key(), video_id))
  64         return entries
  65
  66     def _call_playlist_service(self, item_id, method, fatal=True, referer=None):
  67         headers = {}
  68         if referer:
  69             headers['Referer'] = referer
  70         try:
  71             return self._download_json(
  72                 self._PLAYLIST_SERVICE_URL % (self._PLAYLIST_SERVICE_PATH, item_id, method),
  73                 item_id, 'Downloading PlaylistService %s JSON' % method, 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 _call_api(self, organization_id, item_id, method):
  83         return self._download_json(
  84             self._API_URL % (organization_id, self._API_PATH, item_id, method),
  85             item_id, 'Downloading API %s JSON' % method)
  86
  87     def _extract(self, item_id, pc_method, mobile_method, meta_method, referer=None):
  88         pc = self._call_playlist_service(item_id, pc_method, referer=referer)
  89         metadata = self._call_api(pc['orgId'], item_id, meta_method)
  90         mobile = self._call_playlist_service(item_id, mobile_method, fatal=False, referer=referer)
  91         return pc, mobile, metadata
  92
  93     def _extract_info(self, streams, mobile_urls, properties):
  94         video_id = properties['media_id']
  95         formats = []
  96         urls = []
  97         for stream in streams:
  98             stream_url = stream.get('url')
  99             if not stream_url or stream.get('drmProtected') or stream_url in urls:
 100                 continue
 101             urls.append(stream_url)
 102             ext = determine_ext(stream_url)
 103             if ext == 'f4m':
 104                 formats.extend(self._extract_f4m_formats(
 105                     stream_url, video_id, f4m_id='hds', fatal=False))
 106             else:
 107                 fmt = {
 108                     'url': stream_url,
 109                     'abr': float_or_none(stream.get('audioBitRate')),
 110                     'fps': float_or_none(stream.get('videoFrameRate')),
 111                     'ext': ext,
 112                 }
 113                 width = int_or_none(stream.get('videoWidthInPixels'))
 114                 height = int_or_none(stream.get('videoHeightInPixels'))
 115                 vbr = float_or_none(stream.get('videoBitRate'))
 116                 if width or height or vbr:
 117                     fmt.update({
 118                         'width': width,
 119                         'height': height,
 120                         'vbr': vbr,
 121                     })
 122                 else:
 123                     fmt['vcodec'] = 'none'
 124                 rtmp = re.search(r'^(?P<url>rtmpe?://(?P<host>[^/]+)/(?P<app>.+))/(?P<playpath>mp[34]:.+)$', stream_url)
 125                 if rtmp:
 126                     format_id = 'rtmp'
 127                     if stream.get('videoBitRate'):
 128                         format_id += '-%d' % int_or_none(stream['videoBitRate'])
 129                     http_format_id = format_id.replace('rtmp', 'http')
 130
 131                     CDN_HOSTS = (
 132                         ('delvenetworks.com', 'cpl.delvenetworks.com'),
 133                         ('video.llnw.net', 's2.content.video.llnw.net'),
 134                     )
 135                     for cdn_host, http_host in CDN_HOSTS:
 136                         if cdn_host not in rtmp.group('host').lower():
 137                             continue
 138                         http_url = 'http://%s/%s' % (http_host, rtmp.group('playpath')[4:])
 139                         urls.append(http_url)
 140                         if self._is_valid_url(http_url, video_id, http_format_id):
 141                             http_fmt = fmt.copy()
 142                             http_fmt.update({
 143                                 'url': http_url,
 144                                 'format_id': http_format_id,
 145                             })
 146                             formats.append(http_fmt)
 147                             break
 148
 149                     fmt.update({
 150                         'url': rtmp.group('url'),
 151                         'play_path': rtmp.group('playpath'),
 152                         'app': rtmp.group('app'),
 153                         'ext': 'flv',
 154                         'format_id': format_id,
 155                     })
 156                 formats.append(fmt)
 157
 158         for mobile_url in mobile_urls:
 159             media_url = mobile_url.get('mobileUrl')
 160             format_id = mobile_url.get('targetMediaPlatform')
 161             if not media_url or format_id in ('Widevine', 'SmoothStreaming') or media_url in urls:
 162                 continue
 163             urls.append(media_url)
 164             ext = determine_ext(media_url)
 165             if ext == 'm3u8':
 166                 formats.extend(self._extract_m3u8_formats(
 167                     media_url, video_id, 'mp4', 'm3u8_native',
 168                     m3u8_id=format_id, fatal=False))
 169             elif ext == 'f4m':
 170                 formats.extend(self._extract_f4m_formats(
 171                     stream_url, video_id, f4m_id=format_id, fatal=False))
 172             else:
 173                 formats.append({
 174                     'url': media_url,
 175                     'format_id': format_id,
 176                     'preference': -1,
 177                     'ext': ext,
 178                 })
 179
 180         self._sort_formats(formats)
 181
 182         title = properties['title']
 183         description = properties.get('description')
 184         timestamp = int_or_none(properties.get('publish_date') or properties.get('create_date'))
 185         duration = float_or_none(properties.get('duration_in_milliseconds'), 1000)
 186         filesize = int_or_none(properties.get('total_storage_in_bytes'))
 187         categories = [properties.get('category')]
 188         tags = properties.get('tags', [])
 189         thumbnails = [{
 190             'url': thumbnail['url'],
 191             'width': int_or_none(thumbnail.get('width')),
 192             'height': int_or_none(thumbnail.get('height')),
 193         } for thumbnail in properties.get('thumbnails', []) if thumbnail.get('url')]
 194
 195         subtitles = {}
 196         for caption in properties.get('captions', []):
 197             lang = caption.get('language_code')
 198             subtitles_url = caption.get('url')
 199             if lang and subtitles_url:
 200                 subtitles.setdefault(lang, []).append({
 201                     'url': subtitles_url,
 202                 })
 203         closed_captions_url = properties.get('closed_captions_url')
 204         if closed_captions_url:
 205             subtitles.setdefault('en', []).append({
 206                 'url': closed_captions_url,
 207                 'ext': 'ttml',
 208             })
 209
 210         return {
 211             'id': video_id,
 212             'title': title,
 213             'description': description,
 214             'formats': formats,
 215             'timestamp': timestamp,
 216             'duration': duration,
 217             'filesize': filesize,
 218             'categories': categories,
 219             'tags': tags,
 220             'thumbnails': thumbnails,
 221             'subtitles': subtitles,
 222         }
 223
 224     def _extract_info_helper(self, pc, mobile, i, metadata):
 225         return self._extract_info(
 226             try_get(pc, lambda x: x['playlistItems'][i]['streams'], list) or [],
 227             try_get(mobile, lambda x: x['mediaList'][i]['mobileUrls'], list) or [],
 228             metadata)
 229
 230
 231 class LimelightMediaIE(LimelightBaseIE):
 232     IE_NAME = 'limelight'
 233     _VALID_URL = r'''(?x)
 234                         (?:
 235                             limelight:media:|
 236                             https?://
 237                                 (?:
 238                                     link\.videoplatform\.limelight\.com/media/|
 239                                     assets\.delvenetworks\.com/player/loader\.swf
 240                                 )
 241                                 \?.*?\bmediaId=
 242                         )
 243                         (?P<id>[a-z0-9]{32})
 244                     '''
 245     _TESTS = [{
 246         'url': 'http://link.videoplatform.limelight.com/media/?mediaId=3ffd040b522b4485b6d84effc750cd86',
 247         'info_dict': {
 248             'id': '3ffd040b522b4485b6d84effc750cd86',
 249             'ext': 'mp4',
 250             'title': 'HaP and the HB Prince Trailer',
 251             'description': 'md5:8005b944181778e313d95c1237ddb640',
 252             'thumbnail': r're:^https?://.*\.jpeg$',
 253             'duration': 144.23,
 254             'timestamp': 1244136834,
 255             'upload_date': '20090604',
 256         },
 257         'params': {
 258             # m3u8 download
 259             'skip_download': True,
 260         },
 261     }, {
 262         # video with subtitles
 263         'url': 'limelight:media:a3e00274d4564ec4a9b29b9466432335',
 264         'md5': '2fa3bad9ac321e23860ca23bc2c69e3d',
 265         'info_dict': {
 266             'id': 'a3e00274d4564ec4a9b29b9466432335',
 267             'ext': 'mp4',
 268             'title': '3Play Media Overview Video',
 269             'thumbnail': r're:^https?://.*\.jpeg$',
 270             'duration': 78.101,
 271             'timestamp': 1338929955,
 272             'upload_date': '20120605',
 273             'subtitles': 'mincount:9',
 274         },
 275     }, {
 276         'url': 'https://assets.delvenetworks.com/player/loader.swf?mediaId=8018a574f08d416e95ceaccae4ba0452',
 277         'only_matching': True,
 278     }]
 279     _PLAYLIST_SERVICE_PATH = 'media'
 280     _API_PATH = 'media'
 281
 282     def _real_extract(self, url):
 283         url, smuggled_data = unsmuggle_url(url, {})
 284         video_id = self._match_id(url)
 285         self._initialize_geo_bypass(smuggled_data.get('geo_countries'))
 286
 287         pc, mobile, metadata = self._extract(
 288             video_id, 'getPlaylistByMediaId',
 289             'getMobilePlaylistByMediaId', 'properties',
 290             smuggled_data.get('source_url'))
 291
 292         return self._extract_info_helper(pc, mobile, 0, metadata)
 293
 294
 295 class LimelightChannelIE(LimelightBaseIE):
 296     IE_NAME = 'limelight:channel'
 297     _VALID_URL = r'''(?x)
 298                         (?:
 299                             limelight:channel:|
 300                             https?://
 301                                 (?:
 302                                     link\.videoplatform\.limelight\.com/media/|
 303                                     assets\.delvenetworks\.com/player/loader\.swf
 304                                 )
 305                                 \?.*?\bchannelId=
 306                         )
 307                         (?P<id>[a-z0-9]{32})
 308                     '''
 309     _TESTS = [{
 310         'url': 'http://link.videoplatform.limelight.com/media/?channelId=ab6a524c379342f9b23642917020c082',
 311         'info_dict': {
 312             'id': 'ab6a524c379342f9b23642917020c082',
 313             'title': 'Javascript Sample Code',
 314         },
 315         'playlist_mincount': 3,
 316     }, {
 317         'url': 'http://assets.delvenetworks.com/player/loader.swf?channelId=ab6a524c379342f9b23642917020c082',
 318         'only_matching': True,
 319     }]
 320     _PLAYLIST_SERVICE_PATH = 'channel'
 321     _API_PATH = 'channels'
 322
 323     def _real_extract(self, url):
 324         url, smuggled_data = unsmuggle_url(url, {})
 325         channel_id = self._match_id(url)
 326
 327         pc, mobile, medias = self._extract(
 328             channel_id, 'getPlaylistByChannelId',
 329             'getMobilePlaylistWithNItemsByChannelId?begin=0&count=-1',
 330             'media', smuggled_data.get('source_url'))
 331
 332         entries = [
 333             self._extract_info_helper(pc, mobile, i, medias['media_list'][i])
 334             for i in range(len(medias['media_list']))]
 335
 336         return self.playlist_result(entries, channel_id, pc['title'])
 337
 338
 339 class LimelightChannelListIE(LimelightBaseIE):
 340     IE_NAME = 'limelight:channel_list'
 341     _VALID_URL = r'''(?x)
 342                         (?:
 343                             limelight:channel_list:|
 344                             https?://
 345                                 (?:
 346                                     link\.videoplatform\.limelight\.com/media/|
 347                                     assets\.delvenetworks\.com/player/loader\.swf
 348                                 )
 349                                 \?.*?\bchannelListId=
 350                         )
 351                         (?P<id>[a-z0-9]{32})
 352                     '''
 353     _TESTS = [{
 354         'url': 'http://link.videoplatform.limelight.com/media/?channelListId=301b117890c4465c8179ede21fd92e2b',
 355         'info_dict': {
 356             'id': '301b117890c4465c8179ede21fd92e2b',
 357             'title': 'Website - Hero Player',
 358         },
 359         'playlist_mincount': 2,
 360     }, {
 361         'url': 'https://assets.delvenetworks.com/player/loader.swf?channelListId=301b117890c4465c8179ede21fd92e2b',
 362         'only_matching': True,
 363     }]
 364     _PLAYLIST_SERVICE_PATH = 'channel_list'
 365
 366     def _real_extract(self, url):
 367         channel_list_id = self._match_id(url)
 368
 369         channel_list = self._call_playlist_service(channel_list_id, 'getMobileChannelListById')
 370
 371         entries = [
 372             self.url_result('limelight:channel:%s' % channel['id'], 'LimelightChannel')
 373             for channel in channel_list['channelList']]
 374
 375         return self.playlist_result(entries, channel_list_id, channel_list['title'])