_ Git - youtube-dl/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     int_or_none,
   9     determine_ext,
  10 )
  11
  12
  13 class LimeLightBaseIE(InfoExtractor):
  14
  15     def get_playlist_service(self, id, method):
  16         return self._download_json(self.PLAYLIST_SERVICE_URL % (id, method), id)
  17
  18     def get_api(self, orgId, id, method):
  19         return self._download_json(self.API_URL % (orgId, id, method), id)
  20
  21     def process_data(self, mobileUrls, streams, properties):
  22         video_id = properties['media_id']
  23         formats = []
  24
  25         for mobileUrl in mobileUrls:
  26             if '.m3u8' in mobileUrl['mobileUrl']:
  27                 formats.extend(self._extract_m3u8_formats(mobileUrl['mobileUrl'], video_id))
  28             else:
  29                 formats.append({'url': mobileUrl['mobileUrl']})
  30
  31         for stream in streams:
  32             if '.f4m' in stream['url']:
  33                 formats.extend(self._extract_f4m_formats(stream['url'], video_id))
  34             else:
  35                 fmt = {
  36                     'url': stream.get('url'),
  37                     'abr': stream.get('audioBitRate'),
  38                     'vbr': stream.get('videoBitRate'),
  39                     'fps': stream.get('videoFrameRate'),
  40                     'width': stream.get('videoWidthInPixels'),
  41                     'height': stream.get('videoHeightInPixels'),
  42                     'ext': determine_ext(stream.get('url'))
  43                 }
  44                 rtmp = re.search(r'^(?P<url>rtmp://[^/]+/(?P<app>.+))/(?P<playpath>mp4:.+)$', stream['url'])
  45                 if rtmp:
  46                     fmt.update({
  47                         'url': rtmp.group('url'),
  48                         'play_path': rtmp.group('playpath'),
  49                         'app': rtmp.group('app'),
  50                     })
  51                 formats.append(fmt)
  52
  53         self._sort_formats(formats)
  54
  55         title = properties['title']
  56         description = properties.get('description')
  57         timestamp = properties.get('create_date')
  58         duration = int_or_none(properties.get('duration_in_milliseconds'))
  59         filesize = properties.get('total_storage_in_bytes')
  60         categories = [properties.get('category')]
  61         thumbnails = [{
  62             'url': thumbnail.get('url'),
  63             'width': int_or_none(thumbnail.get('width')),
  64             'height': int_or_none(thumbnail.get('height')),
  65         } for thumbnail in properties.get('thumbnails')]
  66         subtitles = {caption.get('language_code'): [{'url': caption.get('url')}] for caption in properties.get('captions')}
  67
  68         return {
  69             'id': video_id,
  70             'title': title,
  71             'description': description,
  72             'formats': formats,
  73             'timestamp': timestamp,
  74             'duration': duration,
  75             'filesize': filesize,
  76             'categories': categories,
  77             'thumbnails': thumbnails,
  78             'subtitles': subtitles,
  79         }
  80
  81
  82 class LimeLightMediaIE(LimeLightBaseIE):
  83     IE_NAME = 'limelight'
  84     _VALID_URL = r'http://link\.videoplatform\.limelight\.com/media/?.*mediaId=(?P<id>[a-z0-9]{32})'
  85     _TEST = {
  86         'url': 'http://link.videoplatform.limelight.com/media/?mediaId=3ffd040b522b4485b6d84effc750cd86',
  87         'md5': '3213605088be599705677ef785db6972',
  88         'info_dict': {
  89             'id': '3ffd040b522b4485b6d84effc750cd86',
  90             'ext': 'mp4',
  91             'title': 'HaP and the HB Prince Trailer',
  92             'description': 'As Harry Potter begins his 6th year at Hogwarts School of Witchcraft and Wizardry, he discovers an old book marked mysteriously "This book is the property of the Half-Blood Prince" and begins to learn more about Lord Voldemort\'s dark past.',
  93             'thumbnail': 're:^https?://.*\.jpeg$',
  94             'duration': 144230,
  95             'timestamp': 1244136834,
  96             "upload_date": "20090604",
  97         }
  98     }
  99     PLAYLIST_SERVICE_URL = 'http://production-ps.lvp.llnw.net/r/PlaylistService/media/%s/%s'
 100     API_URL = 'http://api.video.limelight.com/rest/organizations/%s/media/%s/%s.json'
 101
 102     def _real_extract(self, url):
 103         video_id = self._match_id(url)
 104
 105         mobile_json_data = self.get_playlist_service(video_id, 'getMobilePlaylistByMediaId')
 106         pc_json_data = self.get_playlist_service(video_id, 'getPlaylistByMediaId')
 107         properties = self.get_api(pc_json_data['orgId'], video_id, 'properties')
 108
 109         return self.process_data(mobile_json_data['mediaList'][0]['mobileUrls'], pc_json_data['playlistItems'][0]['streams'], properties)
 110
 111
 112 class LimeLightChannelIE(LimeLightBaseIE):
 113     IE_NAME = 'limelight:channel'
 114     _VALID_URL = r'http://link\.videoplatform\.limelight\.com/media/?.*channelId=(?P<id>[a-z0-9]{32})'
 115     _TEST = {
 116         'url': 'http://link.videoplatform.limelight.com/media/?channelId=ab6a524c379342f9b23642917020c082',
 117         'info_dict': {
 118             'id': 'ab6a524c379342f9b23642917020c082',
 119             'title': 'Javascript Sample Code',
 120         },
 121         'playlist_mincount': 3,
 122     }
 123     PLAYLIST_SERVICE_URL = 'http://production-ps.lvp.llnw.net/r/PlaylistService/channel/%s/%s'
 124     API_URL = 'http://api.video.limelight.com/rest/organizations/%s/channels/%s/%s.json'
 125
 126     def _real_extract(self, url):
 127         channel_id = self._match_id(url)
 128
 129         mobile_json_data = self.get_playlist_service(channel_id, 'getMobilePlaylistWithNItemsByChannelId?begin=0&count=-1')
 130         pc_json_data = self.get_playlist_service(channel_id, 'getPlaylistByChannelId')
 131         medias = self.get_api(pc_json_data['orgId'], channel_id, 'media')
 132
 133         entries = []
 134         for i in range(len(medias['media_list'])):
 135             entries.append(self.process_data(mobile_json_data['mediaList'][i]['mobileUrls'], pc_json_data['playlistItems'][i]['streams'], medias['media_list'][i]))
 136
 137         return {
 138             'id': channel_id,
 139             'title': pc_json_data['title'],
 140             'entries': entries,
 141             '_type': 'playlist',
 142         }
 143
 144
 145 class LimeLightChannelListIE(LimeLightBaseIE):
 146     IE_NAME = 'limelight:channel_list'
 147     _VALID_URL = r'http://link\.videoplatform\.limelight\.com/media/?.*channelListId=(?P<id>[a-z0-9]{32})'
 148     _TEST = {
 149         'url': 'http://link.videoplatform.limelight.com/media/?channelListId=301b117890c4465c8179ede21fd92e2b',
 150         'info_dict': {
 151             'id': '301b117890c4465c8179ede21fd92e2b',
 152             'title': 'Website - Hero Player',
 153         },
 154         'playlist_mincount': 2,
 155     }
 156     PLAYLIST_SERVICE_URL = 'http://production-ps.lvp.llnw.net/r/PlaylistService/channel_list/%s/%s'
 157
 158     def _real_extract(self, url):
 159         channel_list_id = self._match_id(url)
 160
 161         json_data = self.get_playlist_service(channel_list_id, 'getMobileChannelListById')
 162
 163         entries = []
 164         for channel in json_data['channelList']:
 165             entries.append({
 166                 'url': 'http://link.videoplatform.limelight.com/media/?channelId=%s' % channel['id'],
 167                 '_type': 'url',
 168                 'ie_key': 'LimeLightChannel',
 169             })
 170
 171         return {
 172             'id': channel_list_id,
 173             'title': json_data['title'],
 174             'entries': entries,
 175             '_type': 'playlist',
 176         }