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