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