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:
  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'(?:limelight:media:|http://link\.videoplatform\.limelight\.com/media/\??\bmediaId=)(?P<id>[a-z0-9]{32})'
 127     _TESTS = [{
 128         'url': 'http://link.videoplatform.limelight.com/media/?mediaId=3ffd040b522b4485b6d84effc750cd86',
 129         'info_dict': {
 130             'id': '3ffd040b522b4485b6d84effc750cd86',
 131             'ext': 'flv',
 132             'title': 'HaP and the HB Prince Trailer',
 133             'description': 'md5:8005b944181778e313d95c1237ddb640',
 134             'thumbnail': 're:^https?://.*\.jpeg$',
 135             'duration': 144.23,
 136             'timestamp': 1244136834,
 137             'upload_date': '20090604',
 138         },
 139         'params': {
 140             # rtmp download
 141             'skip_download': True,
 142         },
 143     }, {
 144         # video with subtitles
 145         'url': 'limelight:media:a3e00274d4564ec4a9b29b9466432335',
 146         'info_dict': {
 147             'id': 'a3e00274d4564ec4a9b29b9466432335',
 148             'ext': 'flv',
 149             'title': '3Play Media Overview Video',
 150             'description': '',
 151             'thumbnail': 're:^https?://.*\.jpeg$',
 152             'duration': 78.101,
 153             'timestamp': 1338929955,
 154             'upload_date': '20120605',
 155             'subtitles': 'mincount:9',
 156         },
 157         'params': {
 158             # rtmp download
 159             'skip_download': True,
 160         },
 161     }]
 162     _PLAYLIST_SERVICE_PATH = 'media'
 163     _API_PATH = 'media'
 164
 165     def _real_extract(self, url):
 166         video_id = self._match_id(url)
 167
 168         pc, mobile, metadata = self._extract(
 169             video_id, 'getPlaylistByMediaId', 'getMobilePlaylistByMediaId', 'properties')
 170
 171         return self._extract_info(
 172             pc['playlistItems'][0].get('streams', []),
 173             mobile['mediaList'][0].get('mobileUrls', []) if mobile else [],
 174             metadata)
 175
 176
 177 class LimelightChannelIE(LimelightBaseIE):
 178     IE_NAME = 'limelight:channel'
 179     _VALID_URL = r'(?:limelight:channel:|http://link\.videoplatform\.limelight\.com/media/\??\bchannelId=)(?P<id>[a-z0-9]{32})'
 180     _TEST = {
 181         'url': 'http://link.videoplatform.limelight.com/media/?channelId=ab6a524c379342f9b23642917020c082',
 182         'info_dict': {
 183             'id': 'ab6a524c379342f9b23642917020c082',
 184             'title': 'Javascript Sample Code',
 185         },
 186         'playlist_mincount': 3,
 187     }
 188     _PLAYLIST_SERVICE_PATH = 'channel'
 189     _API_PATH = 'channels'
 190
 191     def _real_extract(self, url):
 192         channel_id = self._match_id(url)
 193
 194         pc, mobile, medias = self._extract(
 195             channel_id, 'getPlaylistByChannelId',
 196             'getMobilePlaylistWithNItemsByChannelId?begin=0&count=-1', 'media')
 197
 198         entries = [
 199             self._extract_info(
 200                 pc['playlistItems'][i].get('streams', []),
 201                 mobile['mediaList'][i].get('mobileUrls', []) if mobile else [],
 202                 medias['media_list'][i])
 203             for i in range(len(medias['media_list']))]
 204
 205         return self.playlist_result(entries, channel_id, pc['title'])
 206
 207
 208 class LimelightChannelListIE(LimelightBaseIE):
 209     IE_NAME = 'limelight:channel_list'
 210     _VALID_URL = r'(?:limelight:channel_list:|http://link\.videoplatform\.limelight\.com/media/\?.*?\bchannelListId=)(?P<id>[a-z0-9]{32})'
 211     _TEST = {
 212         'url': 'http://link.videoplatform.limelight.com/media/?channelListId=301b117890c4465c8179ede21fd92e2b',
 213         'info_dict': {
 214             'id': '301b117890c4465c8179ede21fd92e2b',
 215             'title': 'Website - Hero Player',
 216         },
 217         'playlist_mincount': 2,
 218     }
 219     _PLAYLIST_SERVICE_PATH = 'channel_list'
 220
 221     def _real_extract(self, url):
 222         channel_list_id = self._match_id(url)
 223
 224         channel_list = self._call_playlist_service(channel_list_id, 'getMobileChannelListById')
 225
 226         entries = [
 227             self.url_result('limelight:channel:%s' % channel['id'], 'LimelightChannel')
 228             for channel in channel_list['channelList']]
 229
 230         return self.playlist_result(entries, channel_list_id, channel_list['title'])