youtube_dl/extractor/muzu.py

   1 import re
   2 import json
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     compat_urllib_parse,
   7     determine_ext,
   8 )
   9
  10
  11 class MuzuTVIE(InfoExtractor):
  12     _VALID_URL = r'https?://www.muzu.tv/(.+?)/(.+?)/(?P<id>\d+)'
  13     IE_NAME = u'muzu.tv'
  14
  15     _TEST = {
  16         u'url': u'http://www.muzu.tv/defected/marcashken-featuring-sos-cat-walk-original-mix-music-video/1981454/',
  17         u'file': u'1981454.mp4',
  18         u'md5': u'98f8b2c7bc50578d6a0364fff2bfb000',
  19         u'info_dict': {
  20             u'title': u'Cat Walk (Original Mix)',
  21             u'description': u'md5:90e868994de201b2570e4e5854e19420',
  22             u'uploader': u'MarcAshken featuring SOS',
  23         },
  24     }
  25
  26     def _real_extract(self, url):
  27         mobj = re.match(self._VALID_URL, url)
  28         video_id = mobj.group('id')
  29
  30         info_data = compat_urllib_parse.urlencode({'format': 'json',
  31                                                    'url': url,
  32                                                    })
  33         video_info_page = self._download_webpage('http://www.muzu.tv/api/oembed/?%s' % info_data,
  34                                                  video_id, u'Downloading video info')
  35         info = json.loads(video_info_page)
  36
  37         player_info_page = self._download_webpage('http://player.muzu.tv/player/playerInit?ai=%s' % video_id,
  38                                                   video_id, u'Downloading player info')
  39         video_info = json.loads(player_info_page)['videos'][0]
  40         for quality in ['1080' , '720', '480', '360']:
  41             if video_info.get('v%s' % quality):
  42                 break
  43
  44         data = compat_urllib_parse.urlencode({'ai': video_id,
  45                                               # Even if each time you watch a video the hash changes,
  46                                               # it seems to work for different videos, and it will work
  47                                               # even if you use any non empty string as a hash
  48                                               'viewhash': 'VBNff6djeV4HV5TRPW5kOHub2k',
  49                                               'device': 'web',
  50                                               'qv': quality,
  51                                               })
  52         video_url_page = self._download_webpage('http://player.muzu.tv/player/requestVideo?%s' % data,
  53                                                 video_id, u'Downloading video url')
  54         video_url_info = json.loads(video_url_page)
  55         video_url = video_url_info['url']
  56
  57         return {'id': video_id,
  58                 'title': info['title'],
  59                 'url': video_url,
  60                 'ext': determine_ext(video_url),
  61                 'thumbnail': info['thumbnail_url'],
  62                 'description': info['description'],
  63                 'uploader': info['author_name'],
  64                 }