_ 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 ..utils import (
   8     determine_ext,
   9     float_or_none,
  10     int_or_none,
  11     unsmuggle_url,
  12 )
  13
  14
  15 class LimelightBaseIE(InfoExtractor):
  16     _PLAYLIST_SERVICE_URL = 'http://production-ps.lvp.llnw.net/r/PlaylistService/%s/%s/%s'
  17     _API_URL = 'http://api.video.limelight.com/rest/organizations/%s/%s/%s/%s.json'
  18
  19     def _call_playlist_service(self, item_id, method, fatal=True, referer=None):
  20         headers = {}
  21         if referer:
  22             headers['Referer'] = referer
  23         return self._download_json(
  24             self._PLAYLIST_SERVICE_URL % (self._PLAYLIST_SERVICE_PATH, item_id, method),
  25             item_id, 'Downloading PlaylistService %s JSON' % method, fatal=fatal, headers=headers)
  26
  27     def _call_api(self, organization_id, item_id, method):
  28         return self._download_json(
  29             self._API_URL % (organization_id, self._API_PATH, item_id, method),
  30             item_id, 'Downloading API %s JSON' % method)
  31
  32     def _extract(self, item_id, pc_method, mobile_method, meta_method, referer=None):
  33         pc = self._call_playlist_service(item_id, pc_method, referer=referer)
  34         metadata = self._call_api(pc['orgId'], item_id, meta_method)
  35         mobile = self._call_playlist_service(item_id, mobile_method, fatal=False, referer=referer)
  36         return pc, mobile, metadata
  37
  38     def _extract_info(self, streams, mobile_urls, properties):
  39         video_id = properties['media_id']
  40         formats = []
  41         urls = []
  42         for stream in streams:
  43             stream_url = stream.get('url')
  44             if not stream_url or stream.get('drmProtected') or stream_url in urls:
  45                 continue
  46             urls.append(stream_url)
  47             ext = determine_ext(stream_url)
  48             if ext == 'f4m':
  49                 formats.extend(self._extract_f4m_formats(
  50                     stream_url, video_id, f4m_id='hds', fatal=False))
  51             else:
  52                 fmt = {
  53                     'url': stream_url,
  54                     'abr': float_or_none(stream.get('audioBitRate')),
  55                     'vbr': float_or_none(stream.get('videoBitRate')),
  56                     'fps': float_or_none(stream.get('videoFrameRate')),
  57                     'width': int_or_none(stream.get('videoWidthInPixels')),
  58                     'height': int_or_none(stream.get('videoHeightInPixels')),
  59                     'ext': ext,
  60                 }
  61                 rtmp = re.search(r'^(?P<url>rtmpe?://(?P<host>[^/]+)/(?P<app>.+))/(?P<playpath>mp4:.+)$', stream_url)
  62                 if rtmp:
  63                     format_id = 'rtmp'
  64                     if stream.get('videoBitRate'):
  65                         format_id += '-%d' % int_or_none(stream['videoBitRate'])
  66                     http_format_id = format_id.replace('rtmp', 'http')
  67
  68                     CDN_HOSTS = (
  69                         ('delvenetworks.com', 'cpl.delvenetworks.com'),
  70                         ('video.llnw.net', 's2.content.video.llnw.net'),
  71                     )
  72                     for cdn_host, http_host in CDN_HOSTS:
  73                         if cdn_host not in rtmp.group('host').lower():
  74                             continue
  75                         http_url = 'http://%s/%s' % (http_host, rtmp.group('playpath')[4:])
  76                         urls.append(http_url)
  77                         if self._is_valid_url(http_url, video_id, http_format_id):
  78                             http_fmt = fmt.copy()
  79                             http_fmt.update({
  80                                 'url': http_url,
  81                                 'format_id': http_format_id,
  82                             })
  83                             formats.append(http_fmt)
  84                             break
  85
  86                     fmt.update({
  87                         'url': rtmp.group('url'),
  88                         'play_path': rtmp.group('playpath'),
  89                         'app': rtmp.group('app'),
  90                         'ext': 'flv',
  91                         'format_id': format_id,
  92                     })
  93                 formats.append(fmt)
  94
  95         for mobile_url in mobile_urls:
  96             media_url = mobile_url.get('mobileUrl')
  97             format_id = mobile_url.get('targetMediaPlatform')
  98             if not media_url or format_id in ('Widevine', 'SmoothStreaming') or media_url in urls:
  99                 continue
 100             urls.append(media_url)
 101             ext = determine_ext(media_url)
 102             if ext == 'm3u8':
 103                 formats.extend(self._extract_m3u8_formats(
 104                     media_url, video_id, 'mp4', 'm3u8_native',
 105                     m3u8_id=format_id, fatal=False))
 106             elif ext == 'f4m':
 107                 formats.extend(self._extract_f4m_formats(
 108                     stream_url, video_id, f4m_id=format_id, fatal=False))
 109             else:
 110                 formats.append({
 111                     'url': media_url,
 112                     'format_id': format_id,
 113                     'preference': -1,
 114                     'ext': ext,
 115                 })
 116
 117         self._sort_formats(formats)
 118
 119         title = properties['title']
 120         description = properties.get('description')
 121         timestamp = int_or_none(properties.get('publish_date') or properties.get('create_date'))
 122         duration = float_or_none(properties.get('duration_in_milliseconds'), 1000)
 123         filesize = int_or_none(properties.get('total_storage_in_bytes'))
 124         categories = [properties.get('category')]
 125         tags = properties.get('tags', [])
 126         thumbnails = [{
 127             'url': thumbnail['url'],
 128             'width': int_or_none(thumbnail.get('width')),
 129             'height': int_or_none(thumbnail.get('height')),
 130         } for thumbnail in properties.get('thumbnails', []) if thumbnail.get('url')]
 131
 132         subtitles = {}
 133         for caption in properties.get('captions', []):
 134             lang = caption.get('language_code')
 135             subtitles_url = caption.get('url')
 136             if lang and subtitles_url:
 137                 subtitles.setdefault(lang, []).append({
 138                     'url': subtitles_url,
 139                 })
 140         closed_captions_url = properties.get('closed_captions_url')
 141         if closed_captions_url:
 142             subtitles.setdefault('en', []).append({
 143                 'url': closed_captions_url,
 144                 'ext': 'ttml',
 145             })
 146
 147         return {
 148             'id': video_id,
 149             'title': title,
 150             'description': description,
 151             'formats': formats,
 152             'timestamp': timestamp,
 153             'duration': duration,
 154             'filesize': filesize,
 155             'categories': categories,
 156             'tags': tags,
 157             'thumbnails': thumbnails,
 158             'subtitles': subtitles,
 159         }
 160
 161
 162 class LimelightMediaIE(LimelightBaseIE):
 163     IE_NAME = 'limelight'
 164     _VALID_URL = r'''(?x)
 165                         (?:
 166                             limelight:media:|
 167                             https?://
 168                                 (?:
 169                                     link\.videoplatform\.limelight\.com/media/|
 170                                     assets\.delvenetworks\.com/player/loader\.swf
 171                                 )
 172                                 \?.*?\bmediaId=
 173                         )
 174                         (?P<id>[a-z0-9]{32})
 175                     '''
 176     _TESTS = [{
 177         'url': 'http://link.videoplatform.limelight.com/media/?mediaId=3ffd040b522b4485b6d84effc750cd86',
 178         'info_dict': {
 179             'id': '3ffd040b522b4485b6d84effc750cd86',
 180             'ext': 'mp4',
 181             'title': 'HaP and the HB Prince Trailer',
 182             'description': 'md5:8005b944181778e313d95c1237ddb640',
 183             'thumbnail': r're:^https?://.*\.jpeg$',
 184             'duration': 144.23,
 185             'timestamp': 1244136834,
 186             'upload_date': '20090604',
 187         },
 188         'params': {
 189             # m3u8 download
 190             'skip_download': True,
 191         },
 192     }, {
 193         # video with subtitles
 194         'url': 'limelight:media:a3e00274d4564ec4a9b29b9466432335',
 195         'md5': '2fa3bad9ac321e23860ca23bc2c69e3d',
 196         'info_dict': {
 197             'id': 'a3e00274d4564ec4a9b29b9466432335',
 198             'ext': 'mp4',
 199             'title': '3Play Media Overview Video',
 200             'thumbnail': r're:^https?://.*\.jpeg$',
 201             'duration': 78.101,
 202             'timestamp': 1338929955,
 203             'upload_date': '20120605',
 204             'subtitles': 'mincount:9',
 205         },
 206     }, {
 207         'url': 'https://assets.delvenetworks.com/player/loader.swf?mediaId=8018a574f08d416e95ceaccae4ba0452',
 208         'only_matching': True,
 209     }]
 210     _PLAYLIST_SERVICE_PATH = 'media'
 211     _API_PATH = 'media'
 212
 213     def _real_extract(self, url):
 214         url, smuggled_data = unsmuggle_url(url, {})
 215         video_id = self._match_id(url)
 216
 217         pc, mobile, metadata = self._extract(
 218             video_id, 'getPlaylistByMediaId',
 219             'getMobilePlaylistByMediaId', 'properties',
 220             smuggled_data.get('source_url'))
 221
 222         return self._extract_info(
 223             pc['playlistItems'][0].get('streams', []),
 224             mobile['mediaList'][0].get('mobileUrls', []) if mobile else [],
 225             metadata)
 226
 227
 228 class LimelightChannelIE(LimelightBaseIE):
 229     IE_NAME = 'limelight:channel'
 230     _VALID_URL = r'''(?x)
 231                         (?:
 232                             limelight:channel:|
 233                             https?://
 234                                 (?:
 235                                     link\.videoplatform\.limelight\.com/media/|
 236                                     assets\.delvenetworks\.com/player/loader\.swf
 237                                 )
 238                                 \?.*?\bchannelId=
 239                         )
 240                         (?P<id>[a-z0-9]{32})
 241                     '''
 242     _TESTS = [{
 243         'url': 'http://link.videoplatform.limelight.com/media/?channelId=ab6a524c379342f9b23642917020c082',
 244         'info_dict': {
 245             'id': 'ab6a524c379342f9b23642917020c082',
 246             'title': 'Javascript Sample Code',
 247         },
 248         'playlist_mincount': 3,
 249     }, {
 250         'url': 'http://assets.delvenetworks.com/player/loader.swf?channelId=ab6a524c379342f9b23642917020c082',
 251         'only_matching': True,
 252     }]
 253     _PLAYLIST_SERVICE_PATH = 'channel'
 254     _API_PATH = 'channels'
 255
 256     def _real_extract(self, url):
 257         url, smuggled_data = unsmuggle_url(url, {})
 258         channel_id = self._match_id(url)
 259
 260         pc, mobile, medias = self._extract(
 261             channel_id, 'getPlaylistByChannelId',
 262             'getMobilePlaylistWithNItemsByChannelId?begin=0&count=-1',
 263             'media', smuggled_data.get('source_url'))
 264
 265         entries = [
 266             self._extract_info(
 267                 pc['playlistItems'][i].get('streams', []),
 268                 mobile['mediaList'][i].get('mobileUrls', []) if mobile else [],
 269                 medias['media_list'][i])
 270             for i in range(len(medias['media_list']))]
 271
 272         return self.playlist_result(entries, channel_id, pc['title'])
 273
 274
 275 class LimelightChannelListIE(LimelightBaseIE):
 276     IE_NAME = 'limelight:channel_list'
 277     _VALID_URL = r'''(?x)
 278                         (?:
 279                             limelight:channel_list:|
 280                             https?://
 281                                 (?:
 282                                     link\.videoplatform\.limelight\.com/media/|
 283                                     assets\.delvenetworks\.com/player/loader\.swf
 284                                 )
 285                                 \?.*?\bchannelListId=
 286                         )
 287                         (?P<id>[a-z0-9]{32})
 288                     '''
 289     _TESTS = [{
 290         'url': 'http://link.videoplatform.limelight.com/media/?channelListId=301b117890c4465c8179ede21fd92e2b',
 291         'info_dict': {
 292             'id': '301b117890c4465c8179ede21fd92e2b',
 293             'title': 'Website - Hero Player',
 294         },
 295         'playlist_mincount': 2,
 296     }, {
 297         'url': 'https://assets.delvenetworks.com/player/loader.swf?channelListId=301b117890c4465c8179ede21fd92e2b',
 298         'only_matching': True,
 299     }]
 300     _PLAYLIST_SERVICE_PATH = 'channel_list'
 301
 302     def _real_extract(self, url):
 303         channel_list_id = self._match_id(url)
 304
 305         channel_list = self._call_playlist_service(channel_list_id, 'getMobileChannelListById')
 306
 307         entries = [
 308             self.url_result('limelight:channel:%s' % channel['id'], 'LimelightChannel')
 309             for channel in channel_list['channelList']]
 310
 311         return self.playlist_result(entries, channel_list_id, channel_list['title'])