Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/videofyme.py

   1 import re
   2 import xml.etree.ElementTree
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     find_xpath_attr,
   7     determine_ext,
   8 )
   9
  10 class VideofyMeIE(InfoExtractor):
  11     _VALID_URL = r'https?://(www.videofy.me/.+?|p.videofy.me/v)/(?P<id>\d+)(&|#|$)'
  12     IE_NAME = u'videofy.me'
  13
  14     _TEST = {
  15         u'url': u'http://www.videofy.me/thisisvideofyme/1100701',
  16         u'file':  u'1100701.mp4',
  17         u'md5': u'2046dd5758541d630bfa93e741e2fd79',
  18         u'info_dict': {
  19             u'title': u'This is VideofyMe',
  20             u'description': None,
  21             u'uploader': u'VideofyMe',
  22             u'uploader_id': u'thisisvideofyme',
  23         },
  24
  25     }
  26
  27     def _real_extract(self, url):
  28         mobj = re.match(self._VALID_URL, url)
  29         video_id = mobj.group('id')
  30         config_xml = self._download_webpage('http://sunshine.videofy.me/?videoId=%s' % video_id,
  31                                             video_id)
  32         config = xml.etree.ElementTree.fromstring(config_xml.encode('utf-8'))
  33         video = config.find('video')
  34         sources = video.find('sources')
  35         url_node = find_xpath_attr(sources, 'source', 'id', 'HQ on')
  36         if url_node is None:
  37             url_node = find_xpath_attr(sources, 'source', 'id', 'HQ off')
  38         video_url = url_node.find('url').text
  39
  40         return {'id': video_id,
  41                 'title': video.find('title').text,
  42                 'url': video_url,
  43                 'ext': determine_ext(video_url),
  44                 'thumbnail': video.find('thumb').text,
  45                 'description': video.find('description').text,
  46                 'uploader': config.find('blog/name').text,
  47                 'uploader_id': video.find('identifier').text,
  48                 'view_count': re.search(r'\d+', video.find('views').text).group(),
  49                 }