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