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