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