Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/kaltura.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_urllib_parse
   8 from ..utils import (
   9     ExtractorError,
  10     int_or_none,
  11 )
  12
  13
  14 class KalturaIE(InfoExtractor):
  15     _VALID_URL = r'''(?x)
  16                 (?:
  17                     kaltura:(?P<partner_id_s>\d+):(?P<id_s>[0-9a-z_]+)|
  18                     https?://
  19                         (:?(?:www|cdnapi(?:sec)?)\.)?kaltura\.com/
  20                         (?:
  21                             (?:
  22                                 # flash player
  23                                 index\.php/kwidget/
  24                                 (?:[^/]+/)*?wid/_(?P<partner_id>\d+)/
  25                                 (?:[^/]+/)*?entry_id/(?P<id>[0-9a-z_]+)|
  26                                 # html5 player
  27                                 html5/html5lib/
  28                                 (?:[^/]+/)*?entry_id/(?P<id_html5>[0-9a-z_]+)
  29                                 .*\?.*\bwid=_(?P<partner_id_html5>\d+)
  30                             )
  31                         )
  32                 )
  33                 '''
  34     _API_BASE = 'http://cdnapi.kaltura.com/api_v3/index.php?'
  35     _TESTS = [
  36         {
  37             'url': 'kaltura:269692:1_1jc2y3e4',
  38             'md5': '3adcbdb3dcc02d647539e53f284ba171',
  39             'info_dict': {
  40                 'id': '1_1jc2y3e4',
  41                 'ext': 'mp4',
  42                 'title': 'Track 4',
  43                 'upload_date': '20131219',
  44                 'uploader_id': 'mlundberg@wolfgangsvault.com',
  45                 'description': 'The Allman Brothers Band, 12/16/1981',
  46                 'thumbnail': 're:^https?://.*/thumbnail/.*',
  47                 'timestamp': int,
  48             },
  49         },
  50         {
  51             'url': 'http://www.kaltura.com/index.php/kwidget/cache_st/1300318621/wid/_269692/uiconf_id/3873291/entry_id/1_1jc2y3e4',
  52             'only_matching': True,
  53         },
  54         {
  55             'url': 'https://cdnapisec.kaltura.com/index.php/kwidget/wid/_557781/uiconf_id/22845202/entry_id/1_plr1syf3',
  56             'only_matching': True,
  57         },
  58         {
  59             'url': 'https://cdnapisec.kaltura.com/html5/html5lib/v2.30.2/mwEmbedFrame.php/p/1337/uiconf_id/20540612/entry_id/1_sf5ovm7u?wid=_243342',
  60             'only_matching': True,
  61         }
  62     ]
  63
  64     def _kaltura_api_call(self, video_id, actions, *args, **kwargs):
  65         params = actions[0]
  66         if len(actions) > 1:
  67             for i, a in enumerate(actions[1:], start=1):
  68                 for k, v in a.items():
  69                     params['%d:%s' % (i, k)] = v
  70
  71         query = compat_urllib_parse.urlencode(params)
  72         url = self._API_BASE + query
  73         data = self._download_json(url, video_id, *args, **kwargs)
  74
  75         status = data if len(actions) == 1 else data[0]
  76         if status.get('objectType') == 'KalturaAPIException':
  77             raise ExtractorError(
  78                 '%s said: %s' % (self.IE_NAME, status['message']))
  79
  80         return data
  81
  82     def _get_kaltura_signature(self, video_id, partner_id):
  83         actions = [{
  84             'apiVersion': '3.1',
  85             'expiry': 86400,
  86             'format': 1,
  87             'service': 'session',
  88             'action': 'startWidgetSession',
  89             'widgetId': '_%s' % partner_id,
  90         }]
  91         return self._kaltura_api_call(
  92             video_id, actions, note='Downloading Kaltura signature')['ks']
  93
  94     def _get_video_info(self, video_id, partner_id):
  95         signature = self._get_kaltura_signature(video_id, partner_id)
  96         actions = [
  97             {
  98                 'action': 'null',
  99                 'apiVersion': '3.1.5',
 100                 'clientTag': 'kdp:v3.8.5',
 101                 'format': 1,  # JSON, 2 = XML, 3 = PHP
 102                 'service': 'multirequest',
 103                 'ks': signature,
 104             },
 105             {
 106                 'action': 'get',
 107                 'entryId': video_id,
 108                 'service': 'baseentry',
 109                 'version': '-1',
 110             },
 111             {
 112                 'action': 'getContextData',
 113                 'contextDataParams:objectType': 'KalturaEntryContextDataParams',
 114                 'contextDataParams:referrer': 'http://www.kaltura.com/',
 115                 'contextDataParams:streamerType': 'http',
 116                 'entryId': video_id,
 117                 'service': 'baseentry',
 118             },
 119         ]
 120         return self._kaltura_api_call(
 121             video_id, actions, note='Downloading video info JSON')
 122
 123     def _real_extract(self, url):
 124         mobj = re.match(self._VALID_URL, url)
 125         partner_id = mobj.group('partner_id_s') or mobj.group('partner_id') or mobj.group('partner_id_html5')
 126         entry_id = mobj.group('id_s') or mobj.group('id') or mobj.group('id_html5')
 127
 128         info, source_data = self._get_video_info(entry_id, partner_id)
 129
 130         formats = [{
 131             'format_id': '%(fileExt)s-%(bitrate)s' % f,
 132             'ext': f['fileExt'],
 133             'tbr': f['bitrate'],
 134             'fps': f.get('frameRate'),
 135             'filesize_approx': int_or_none(f.get('size'), invscale=1024),
 136             'container': f.get('containerFormat'),
 137             'vcodec': f.get('videoCodecId'),
 138             'height': f.get('height'),
 139             'width': f.get('width'),
 140             'url': '%s/flavorId/%s' % (info['dataUrl'], f['id']),
 141         } for f in source_data['flavorAssets']]
 142         self._sort_formats(formats)
 143
 144         return {
 145             'id': entry_id,
 146             'title': info['name'],
 147             'formats': formats,
 148             'description': info.get('description'),
 149             'thumbnail': info.get('thumbnailUrl'),
 150             'duration': info.get('duration'),
 151             'timestamp': info.get('createdAt'),
 152             'uploader_id': info.get('userId'),
 153             'view_count': info.get('plays'),
 154         }