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 ..compat import compat_HTTPError
   8 from ..utils import (
   9     determine_ext,
  10     float_or_none,
  11     int_or_none,
  12     unsmuggle_url,
  13     ExtractorError,
  14 )
  15
  16
  17 class LimelightBaseIE(InfoExtractor):
  18     _PLAYLIST_SERVICE_URL = 'http://production-ps.lvp.llnw.net/r/PlaylistService/%s/%s/%s'
  19     _API_URL = 'http://api.video.limelight.com/rest/organizations/%s/%s/%s/%s.json'
  20
  21     def _call_playlist_service(self, item_id, method, fatal=True, referer=None):
  22         headers = {}
  23         if referer:
  24             headers['Referer'] = referer
  25         try:
  26             return self._download_json(
  27                 self._PLAYLIST_SERVICE_URL % (self._PLAYLIST_SERVICE_PATH, item_id, method),
  28                 item_id, 'Downloading PlaylistService %s JSON' % method, fatal=fatal, headers=headers)
  29         except ExtractorError as e:
  30             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
  31                 error = self._parse_json(e.cause.read().decode(), item_id)['detail']['contentAccessPermission']
  32                 if error == 'CountryDisabled':
  33                     self.raise_geo_restricted()
  34                 raise ExtractorError(error, expected=True)
  35             raise
  36
  37     def _call_api(self, organization_id, item_id, method):
  38         return self._download_json(
  39             self._API_URL % (organization_id, self._API_PATH, item_id, method),
  40             item_id, 'Downloading API %s JSON' % method)
  41
  42     def _extract(self, item_id, pc_method, mobile_method, meta_method, referer=None):
  43         pc = self._call_playlist_service(item_id, pc_method, referer=referer)
  44         metadata = self._call_api(pc['orgId'], item_id, meta_method)
  45         mobile = self._call_playlist_service(item_id, mobile_method, fatal=False, referer=referer)
  46         return pc, mobile, metadata
  47
  48     def _extract_info(self, streams, mobile_urls, properties):
  49         video_id = properties['media_id']
  50         formats = []
  51         urls = []
  52         for stream in streams:
  53             stream_url = stream.get('url')
  54             if not stream_url or stream.get('drmProtected') or stream_url in urls:
  55                 continue
  56             urls.append(stream_url)
  57             ext = determine_ext(stream_url)
  58             if ext == 'f4m':
  59                 formats.extend(self._extract_f4m_formats(
  60                     stream_url, video_id, f4m_id='hds', fatal=False))
  61             else:
  62                 fmt = {
  63                     'url': stream_url,
  64                     'abr': float_or_none(stream.get('audioBitRate')),
  65                     'vbr': float_or_none(stream.get('videoBitRate')),
  66                     'fps': float_or_none(stream.get('videoFrameRate')),
  67                     'width': int_or_none(stream.get('videoWidthInPixels')),
  68                     'height': int_or_none(stream.get('videoHeightInPixels')),
  69                     'ext': ext,
  70                 }
  71                 rtmp = re.search(r'^(?P<url>rtmpe?://(?P<host>[^/]+)/(?P<app>.+))/(?P<playpath>mp4:.+)$', stream_url)
  72                 if rtmp:
  73                     format_id = 'rtmp'
  74                     if stream.get('videoBitRate'):
  75                         format_id += '-%d' % int_or_none(stream['videoBitRate'])
  76                     http_format_id = format_id.replace('rtmp', 'http')
  77
  78                     CDN_HOSTS = (
  79                         ('delvenetworks.com', 'cpl.delvenetworks.com'),
  80                         ('video.llnw.net', 's2.content.video.llnw.net'),
  81                     )
  82                     for cdn_host, http_host in CDN_HOSTS:
  83                         if cdn_host not in rtmp.group('host').lower():
  84                             continue
  85                         http_url = 'http://%s/%s' % (http_host, rtmp.group('playpath')[4:])
  86                         urls.append(http_url)
  87                         if self._is_valid_url(http_url, video_id, http_format_id):
  88                             http_fmt = fmt.copy()
  89                             http_fmt.update({
  90                                 'url': http_url,
  91                                 'format_id': http_format_id,
  92                             })
  93                             formats.append(http_fmt)
  94                             break
  95
  96                     fmt.update({
  97                         'url': rtmp.group('url'),
  98                         'play_path': rtmp.group('playpath'),
  99                         'app': rtmp.group('app'),
 100                         'ext': 'flv',
 101                         'format_id': format_id,
 102                     })
 103                 formats.append(fmt)
 104
 105         for mobile_url in mobile_urls:
 106             media_url = mobile_url.get('mobileUrl')
 107             format_id = mobile_url.get('targetMediaPlatform')
 108             if not media_url or format_id in ('Widevine', 'SmoothStreaming') or media_url in urls:
 109                 continue
 110             urls.append(media_url)
 111             ext = determine_ext(media_url)
 112             if ext == 'm3u8':
 113                 formats.extend(self._extract_m3u8_formats(
 114                     media_url, video_id, 'mp4', 'm3u8_native',
 115                     m3u8_id=format_id, fatal=False))
 116             elif ext == 'f4m':
 117                 formats.extend(self._extract_f4m_formats(
 118                     stream_url, video_id, f4m_id=format_id, fatal=False))
 119             else:
 120                 formats.append({
 121                     'url': media_url,
 122                     'format_id': format_id,
 123                     'preference': -1,
 124                     'ext': ext,
 125                 })
 126
 127         self._sort_formats(formats)
 128
 129         title = properties['title']
 130         description = properties.get('description')
 131         timestamp = int_or_none(properties.get('publish_date') or properties.get('create_date'))
 132         duration = float_or_none(properties.get('duration_in_milliseconds'), 1000)
 133         filesize = int_or_none(properties.get('total_storage_in_bytes'))
 134         categories = [properties.get('category')]
 135         tags = properties.get('tags', [])
 136         thumbnails = [{
 137             'url': thumbnail['url'],
 138             'width': int_or_none(thumbnail.get('width')),
 139             'height': int_or_none(thumbnail.get('height')),
 140         } for thumbnail in properties.get('thumbnails', []) if thumbnail.get('url')]
 141
 142         subtitles = {}
 143         for caption in properties.get('captions', []):
 144             lang = caption.get('language_code')
 145             subtitles_url = caption.get('url')
 146             if lang and subtitles_url:
 147                 subtitles.setdefault(lang, []).append({
 148                     'url': subtitles_url,
 149                 })
 150         closed_captions_url = properties.get('closed_captions_url')
 151         if closed_captions_url:
 152             subtitles.setdefault('en', []).append({
 153                 'url': closed_captions_url,
 154                 'ext': 'ttml',
 155             })
 156
 157         return {
 158             'id': video_id,
 159             'title': title,
 160             'description': description,
 161             'formats': formats,
 162             'timestamp': timestamp,
 163             'duration': duration,
 164             'filesize': filesize,
 165             'categories': categories,
 166             'tags': tags,
 167             'thumbnails': thumbnails,
 168             'subtitles': subtitles,
 169         }
 170
 171
 172 class LimelightMediaIE(LimelightBaseIE):
 173     IE_NAME = 'limelight'
 174     _VALID_URL = r'''(?x)
 175                         (?:
 176                             limelight:media:|
 177                             https?://
 178                                 (?:
 179                                     link\.videoplatform\.limelight\.com/media/|
 180                                     assets\.delvenetworks\.com/player/loader\.swf
 181                                 )
 182                                 \?.*?\bmediaId=
 183                         )
 184                         (?P<id>[a-z0-9]{32})
 185                     '''
 186     _TESTS = [{
 187         'url': 'http://link.videoplatform.limelight.com/media/?mediaId=3ffd040b522b4485b6d84effc750cd86',
 188         'info_dict': {
 189             'id': '3ffd040b522b4485b6d84effc750cd86',
 190             'ext': 'mp4',
 191             'title': 'HaP and the HB Prince Trailer',
 192             'description': 'md5:8005b944181778e313d95c1237ddb640',
 193             'thumbnail': r're:^https?://.*\.jpeg$',
 194             'duration': 144.23,
 195             'timestamp': 1244136834,
 196             'upload_date': '20090604',
 197         },
 198         'params': {
 199             # m3u8 download
 200             'skip_download': True,
 201         },
 202     }, {
 203         # video with subtitles
 204         'url': 'limelight:media:a3e00274d4564ec4a9b29b9466432335',
 205         'md5': '2fa3bad9ac321e23860ca23bc2c69e3d',
 206         'info_dict': {
 207             'id': 'a3e00274d4564ec4a9b29b9466432335',
 208             'ext': 'mp4',
 209             'title': '3Play Media Overview Video',
 210             'thumbnail': r're:^https?://.*\.jpeg$',
 211             'duration': 78.101,
 212             'timestamp': 1338929955,
 213             'upload_date': '20120605',
 214             'subtitles': 'mincount:9',
 215         },
 216     }, {
 217         'url': 'https://assets.delvenetworks.com/player/loader.swf?mediaId=8018a574f08d416e95ceaccae4ba0452',
 218         'only_matching': True,
 219     }]
 220     _PLAYLIST_SERVICE_PATH = 'media'
 221     _API_PATH = 'media'
 222
 223     def _real_extract(self, url):
 224         url, smuggled_data = unsmuggle_url(url, {})
 225         video_id = self._match_id(url)
 226         self._initialize_geo_bypass(smuggled_data.get('geo_countries'))
 227
 228         pc, mobile, metadata = self._extract(
 229             video_id, 'getPlaylistByMediaId',
 230             'getMobilePlaylistByMediaId', 'properties',
 231             smuggled_data.get('source_url'))
 232
 233         return self._extract_info(
 234             pc['playlistItems'][0].get('streams', []),
 235             mobile['mediaList'][0].get('mobileUrls', []) if mobile else [],
 236             metadata)
 237
 238
 239 class LimelightChannelIE(LimelightBaseIE):
 240     IE_NAME = 'limelight:channel'
 241     _VALID_URL = r'''(?x)
 242                         (?:
 243                             limelight:channel:|
 244                             https?://
 245                                 (?:
 246                                     link\.videoplatform\.limelight\.com/media/|
 247                                     assets\.delvenetworks\.com/player/loader\.swf
 248                                 )
 249                                 \?.*?\bchannelId=
 250                         )
 251                         (?P<id>[a-z0-9]{32})
 252                     '''
 253     _TESTS = [{
 254         'url': 'http://link.videoplatform.limelight.com/media/?channelId=ab6a524c379342f9b23642917020c082',
 255         'info_dict': {
 256             'id': 'ab6a524c379342f9b23642917020c082',
 257             'title': 'Javascript Sample Code',
 258         },
 259         'playlist_mincount': 3,
 260     }, {
 261         'url': 'http://assets.delvenetworks.com/player/loader.swf?channelId=ab6a524c379342f9b23642917020c082',
 262         'only_matching': True,
 263     }]
 264     _PLAYLIST_SERVICE_PATH = 'channel'
 265     _API_PATH = 'channels'
 266
 267     def _real_extract(self, url):
 268         url, smuggled_data = unsmuggle_url(url, {})
 269         channel_id = self._match_id(url)
 270
 271         pc, mobile, medias = self._extract(
 272             channel_id, 'getPlaylistByChannelId',
 273             'getMobilePlaylistWithNItemsByChannelId?begin=0&count=-1',
 274             'media', smuggled_data.get('source_url'))
 275
 276         entries = [
 277             self._extract_info(
 278                 pc['playlistItems'][i].get('streams', []),
 279                 mobile['mediaList'][i].get('mobileUrls', []) if mobile else [],
 280                 medias['media_list'][i])
 281             for i in range(len(medias['media_list']))]
 282
 283         return self.playlist_result(entries, channel_id, pc['title'])
 284
 285
 286 class LimelightChannelListIE(LimelightBaseIE):
 287     IE_NAME = 'limelight:channel_list'
 288     _VALID_URL = r'''(?x)
 289                         (?:
 290                             limelight:channel_list:|
 291                             https?://
 292                                 (?:
 293                                     link\.videoplatform\.limelight\.com/media/|
 294                                     assets\.delvenetworks\.com/player/loader\.swf
 295                                 )
 296                                 \?.*?\bchannelListId=
 297                         )
 298                         (?P<id>[a-z0-9]{32})
 299                     '''
 300     _TESTS = [{
 301         'url': 'http://link.videoplatform.limelight.com/media/?channelListId=301b117890c4465c8179ede21fd92e2b',
 302         'info_dict': {
 303             'id': '301b117890c4465c8179ede21fd92e2b',
 304             'title': 'Website - Hero Player',
 305         },
 306         'playlist_mincount': 2,
 307     }, {
 308         'url': 'https://assets.delvenetworks.com/player/loader.swf?channelListId=301b117890c4465c8179ede21fd92e2b',
 309         'only_matching': True,
 310     }]
 311     _PLAYLIST_SERVICE_PATH = 'channel_list'
 312
 313     def _real_extract(self, url):
 314         channel_list_id = self._match_id(url)
 315
 316         channel_list = self._call_playlist_service(channel_list_id, 'getMobileChannelListById')
 317
 318         entries = [
 319             self.url_result('limelight:channel:%s' % channel['id'], 'LimelightChannel')
 320             for channel in channel_list['channelList']]
 321
 322         return self.playlist_result(entries, channel_list_id, channel_list['title'])