_ 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     unsmuggle_url,
  13     ExtractorError,
  14 )
  15
  16
  17 class LimelightBaseIE(InfoExtractor):
  18     _PLAYLIST_SERVICE_URL = 'http://production-ps.lvp.llnw.net/r/PlaylistService/%s/%s/%s'
  19     _API_URL = 'http://api.video.limelight.com/rest/organizations/%s/%s/%s/%s.json'
  20
  21     def _call_playlist_service(self, item_id, method, fatal=True, referer=None):
  22         headers = {}
  23         if referer:
  24             headers['Referer'] = referer
  25         try:
  26             return self._download_json(
  27                 self._PLAYLIST_SERVICE_URL % (self._PLAYLIST_SERVICE_PATH, item_id, method),
  28                 item_id, 'Downloading PlaylistService %s JSON' % method, fatal=fatal, headers=headers)
  29         except ExtractorError as e:
  30             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
  31                 error = self._parse_json(e.cause.read().decode(), item_id)['detail']['contentAccessPermission']
  32                 if error == 'CountryDisabled':
  33                     self.raise_geo_restricted()
  34                 raise ExtractorError(error, expected=True)
  35             raise
  36
  37     def _call_api(self, organization_id, item_id, method):
  38         return self._download_json(
  39             self._API_URL % (organization_id, self._API_PATH, item_id, method),
  40             item_id, 'Downloading API %s JSON' % method)
  41
  42     def _extract(self, item_id, pc_method, mobile_method, meta_method, referer=None):
  43         pc = self._call_playlist_service(item_id, pc_method, referer=referer)
  44         metadata = self._call_api(pc['orgId'], item_id, meta_method)
  45         mobile = self._call_playlist_service(item_id, mobile_method, fatal=False, referer=referer)
  46         return pc, mobile, metadata
  47
  48     def _extract_info(self, streams, mobile_urls, properties):
  49         video_id = properties['media_id']
  50         formats = []
  51         urls = []
  52         for stream in streams:
  53             stream_url = stream.get('url')
  54             if not stream_url or stream.get('drmProtected') or stream_url in urls:
  55                 continue
  56             urls.append(stream_url)
  57             ext = determine_ext(stream_url)
  58             if ext == 'f4m':
  59                 formats.extend(self._extract_f4m_formats(
  60                     stream_url, video_id, f4m_id='hds', fatal=False))
  61             else:
  62                 fmt = {
  63                     'url': stream_url,
  64                     'abr': float_or_none(stream.get('audioBitRate')),
  65                     'fps': float_or_none(stream.get('videoFrameRate')),
  66                     'ext': ext,
  67                 }
  68                 width = int_or_none(stream.get('videoWidthInPixels'))
  69                 height = int_or_none(stream.get('videoHeightInPixels'))
  70                 vbr = float_or_none(stream.get('videoBitRate'))
  71                 if width or height or vbr:
  72                     fmt.update({
  73                         'width': width,
  74                         'height': height,
  75                         'vbr': vbr,
  76                     })
  77                 else:
  78                     fmt['vcodec'] = 'none'
  79                 rtmp = re.search(r'^(?P<url>rtmpe?://(?P<host>[^/]+)/(?P<app>.+))/(?P<playpath>mp[34]:.+)$', stream_url)
  80                 if rtmp:
  81                     format_id = 'rtmp'
  82                     if stream.get('videoBitRate'):
  83                         format_id += '-%d' % int_or_none(stream['videoBitRate'])
  84                     http_format_id = format_id.replace('rtmp', 'http')
  85
  86                     CDN_HOSTS = (
  87                         ('delvenetworks.com', 'cpl.delvenetworks.com'),
  88                         ('video.llnw.net', 's2.content.video.llnw.net'),
  89                     )
  90                     for cdn_host, http_host in CDN_HOSTS:
  91                         if cdn_host not in rtmp.group('host').lower():
  92                             continue
  93                         http_url = 'http://%s/%s' % (http_host, rtmp.group('playpath')[4:])
  94                         urls.append(http_url)
  95                         if self._is_valid_url(http_url, video_id, http_format_id):
  96                             http_fmt = fmt.copy()
  97                             http_fmt.update({
  98                                 'url': http_url,
  99                                 'format_id': http_format_id,
 100                             })
 101                             formats.append(http_fmt)
 102                             break
 103
 104                     fmt.update({
 105                         'url': rtmp.group('url'),
 106                         'play_path': rtmp.group('playpath'),
 107                         'app': rtmp.group('app'),
 108                         'ext': 'flv',
 109                         'format_id': format_id,
 110                     })
 111                 formats.append(fmt)
 112
 113         for mobile_url in mobile_urls:
 114             media_url = mobile_url.get('mobileUrl')
 115             format_id = mobile_url.get('targetMediaPlatform')
 116             if not media_url or format_id in ('Widevine', 'SmoothStreaming') or media_url in urls:
 117                 continue
 118             urls.append(media_url)
 119             ext = determine_ext(media_url)
 120             if ext == 'm3u8':
 121                 formats.extend(self._extract_m3u8_formats(
 122                     media_url, video_id, 'mp4', 'm3u8_native',
 123                     m3u8_id=format_id, fatal=False))
 124             elif ext == 'f4m':
 125                 formats.extend(self._extract_f4m_formats(
 126                     stream_url, video_id, f4m_id=format_id, fatal=False))
 127             else:
 128                 formats.append({
 129                     'url': media_url,
 130                     'format_id': format_id,
 131                     'preference': -1,
 132                     'ext': ext,
 133                 })
 134
 135         self._sort_formats(formats)
 136
 137         title = properties['title']
 138         description = properties.get('description')
 139         timestamp = int_or_none(properties.get('publish_date') or properties.get('create_date'))
 140         duration = float_or_none(properties.get('duration_in_milliseconds'), 1000)
 141         filesize = int_or_none(properties.get('total_storage_in_bytes'))
 142         categories = [properties.get('category')]
 143         tags = properties.get('tags', [])
 144         thumbnails = [{
 145             'url': thumbnail['url'],
 146             'width': int_or_none(thumbnail.get('width')),
 147             'height': int_or_none(thumbnail.get('height')),
 148         } for thumbnail in properties.get('thumbnails', []) if thumbnail.get('url')]
 149
 150         subtitles = {}
 151         for caption in properties.get('captions', []):
 152             lang = caption.get('language_code')
 153             subtitles_url = caption.get('url')
 154             if lang and subtitles_url:
 155                 subtitles.setdefault(lang, []).append({
 156                     'url': subtitles_url,
 157                 })
 158         closed_captions_url = properties.get('closed_captions_url')
 159         if closed_captions_url:
 160             subtitles.setdefault('en', []).append({
 161                 'url': closed_captions_url,
 162                 'ext': 'ttml',
 163             })
 164
 165         return {
 166             'id': video_id,
 167             'title': title,
 168             'description': description,
 169             'formats': formats,
 170             'timestamp': timestamp,
 171             'duration': duration,
 172             'filesize': filesize,
 173             'categories': categories,
 174             'tags': tags,
 175             'thumbnails': thumbnails,
 176             'subtitles': subtitles,
 177         }
 178
 179
 180 class LimelightMediaIE(LimelightBaseIE):
 181     IE_NAME = 'limelight'
 182     _VALID_URL = r'''(?x)
 183                         (?:
 184                             limelight:media:|
 185                             https?://
 186                                 (?:
 187                                     link\.videoplatform\.limelight\.com/media/|
 188                                     assets\.delvenetworks\.com/player/loader\.swf
 189                                 )
 190                                 \?.*?\bmediaId=
 191                         )
 192                         (?P<id>[a-z0-9]{32})
 193                     '''
 194     _TESTS = [{
 195         'url': 'http://link.videoplatform.limelight.com/media/?mediaId=3ffd040b522b4485b6d84effc750cd86',
 196         'info_dict': {
 197             'id': '3ffd040b522b4485b6d84effc750cd86',
 198             'ext': 'mp4',
 199             'title': 'HaP and the HB Prince Trailer',
 200             'description': 'md5:8005b944181778e313d95c1237ddb640',
 201             'thumbnail': r're:^https?://.*\.jpeg$',
 202             'duration': 144.23,
 203             'timestamp': 1244136834,
 204             'upload_date': '20090604',
 205         },
 206         'params': {
 207             # m3u8 download
 208             'skip_download': True,
 209         },
 210     }, {
 211         # video with subtitles
 212         'url': 'limelight:media:a3e00274d4564ec4a9b29b9466432335',
 213         'md5': '2fa3bad9ac321e23860ca23bc2c69e3d',
 214         'info_dict': {
 215             'id': 'a3e00274d4564ec4a9b29b9466432335',
 216             'ext': 'mp4',
 217             'title': '3Play Media Overview Video',
 218             'thumbnail': r're:^https?://.*\.jpeg$',
 219             'duration': 78.101,
 220             'timestamp': 1338929955,
 221             'upload_date': '20120605',
 222             'subtitles': 'mincount:9',
 223         },
 224     }, {
 225         'url': 'https://assets.delvenetworks.com/player/loader.swf?mediaId=8018a574f08d416e95ceaccae4ba0452',
 226         'only_matching': True,
 227     }]
 228     _PLAYLIST_SERVICE_PATH = 'media'
 229     _API_PATH = 'media'
 230
 231     def _real_extract(self, url):
 232         url, smuggled_data = unsmuggle_url(url, {})
 233         video_id = self._match_id(url)
 234         self._initialize_geo_bypass(smuggled_data.get('geo_countries'))
 235
 236         pc, mobile, metadata = self._extract(
 237             video_id, 'getPlaylistByMediaId',
 238             'getMobilePlaylistByMediaId', 'properties',
 239             smuggled_data.get('source_url'))
 240
 241         return self._extract_info(
 242             pc['playlistItems'][0].get('streams', []),
 243             mobile['mediaList'][0].get('mobileUrls', []) if mobile else [],
 244             metadata)
 245
 246
 247 class LimelightChannelIE(LimelightBaseIE):
 248     IE_NAME = 'limelight:channel'
 249     _VALID_URL = r'''(?x)
 250                         (?:
 251                             limelight:channel:|
 252                             https?://
 253                                 (?:
 254                                     link\.videoplatform\.limelight\.com/media/|
 255                                     assets\.delvenetworks\.com/player/loader\.swf
 256                                 )
 257                                 \?.*?\bchannelId=
 258                         )
 259                         (?P<id>[a-z0-9]{32})
 260                     '''
 261     _TESTS = [{
 262         'url': 'http://link.videoplatform.limelight.com/media/?channelId=ab6a524c379342f9b23642917020c082',
 263         'info_dict': {
 264             'id': 'ab6a524c379342f9b23642917020c082',
 265             'title': 'Javascript Sample Code',
 266         },
 267         'playlist_mincount': 3,
 268     }, {
 269         'url': 'http://assets.delvenetworks.com/player/loader.swf?channelId=ab6a524c379342f9b23642917020c082',
 270         'only_matching': True,
 271     }]
 272     _PLAYLIST_SERVICE_PATH = 'channel'
 273     _API_PATH = 'channels'
 274
 275     def _real_extract(self, url):
 276         url, smuggled_data = unsmuggle_url(url, {})
 277         channel_id = self._match_id(url)
 278
 279         pc, mobile, medias = self._extract(
 280             channel_id, 'getPlaylistByChannelId',
 281             'getMobilePlaylistWithNItemsByChannelId?begin=0&count=-1',
 282             'media', smuggled_data.get('source_url'))
 283
 284         entries = [
 285             self._extract_info(
 286                 pc['playlistItems'][i].get('streams', []),
 287                 mobile['mediaList'][i].get('mobileUrls', []) if mobile else [],
 288                 medias['media_list'][i])
 289             for i in range(len(medias['media_list']))]
 290
 291         return self.playlist_result(entries, channel_id, pc['title'])
 292
 293
 294 class LimelightChannelListIE(LimelightBaseIE):
 295     IE_NAME = 'limelight:channel_list'
 296     _VALID_URL = r'''(?x)
 297                         (?:
 298                             limelight:channel_list:|
 299                             https?://
 300                                 (?:
 301                                     link\.videoplatform\.limelight\.com/media/|
 302                                     assets\.delvenetworks\.com/player/loader\.swf
 303                                 )
 304                                 \?.*?\bchannelListId=
 305                         )
 306                         (?P<id>[a-z0-9]{32})
 307                     '''
 308     _TESTS = [{
 309         'url': 'http://link.videoplatform.limelight.com/media/?channelListId=301b117890c4465c8179ede21fd92e2b',
 310         'info_dict': {
 311             'id': '301b117890c4465c8179ede21fd92e2b',
 312             'title': 'Website - Hero Player',
 313         },
 314         'playlist_mincount': 2,
 315     }, {
 316         'url': 'https://assets.delvenetworks.com/player/loader.swf?channelListId=301b117890c4465c8179ede21fd92e2b',
 317         'only_matching': True,
 318     }]
 319     _PLAYLIST_SERVICE_PATH = 'channel_list'
 320
 321     def _real_extract(self, url):
 322         channel_list_id = self._match_id(url)
 323
 324         channel_list = self._call_playlist_service(channel_list_id, 'getMobileChannelListById')
 325
 326         entries = [
 327             self.url_result('limelight:channel:%s' % channel['id'], 'LimelightChannel')
 328             for channel in channel_list['channelList']]
 329
 330         return self.playlist_result(entries, channel_list_id, channel_list['title'])