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