youtube_dl/extractor/metacritic.py

   1 import re
   2 import xml.etree.ElementTree
   3 import operator
   4
   5 from .common import InfoExtractor
   6
   7
   8 class MetacriticIE(InfoExtractor):
   9     _VALID_URL = r'https?://www\.metacritic\.com/.+?/trailers/(?P<id>\d+)'
  10
  11     _TEST = {
  12         u'url': u'http://www.metacritic.com/game/playstation-4/infamous-second-son/trailers/3698222',
  13         u'file': u'3698222.mp4',
  14         u'info_dict': {
  15             u'title': u'inFamous: Second Son - inSide Sucker Punch: Smoke & Mirrors',
  16             u'description': u'Take a peak behind-the-scenes to see how Sucker Punch brings smoke into the universe of inFAMOUS Second Son on the PS4.',
  17             u'duration': 221,
  18         },
  19     }
  20
  21     def _real_extract(self, url):
  22         mobj = re.match(self._VALID_URL, url)
  23         video_id = mobj.group('id')
  24         webpage = self._download_webpage(url, video_id)
  25         # The xml is not well formatted, there are raw '&'
  26         info_xml = self._download_webpage('http://www.metacritic.com/video_data?video=' + video_id,
  27             video_id, u'Downloading info xml').replace('&', '&amp;')
  28         info = xml.etree.ElementTree.fromstring(info_xml.encode('utf-8'))
  29
  30         clip = next(c for c in info.findall('playList/clip') if c.find('id').text == video_id)
  31         formats = []
  32         for videoFile in clip.findall('httpURI/videoFile'):
  33             rate_str = videoFile.find('rate').text
  34             video_url = videoFile.find('filePath').text
  35             formats.append({
  36                 'url': video_url,
  37                 'ext': 'mp4',
  38                 'format_id': rate_str,
  39                 'rate': int(rate_str),
  40             })
  41         formats.sort(key=operator.itemgetter('rate'))
  42
  43         description = self._html_search_regex(r'<b>Description:</b>(.*?)</p>',
  44             webpage, u'description', flags=re.DOTALL)
  45
  46         info = {
  47             'id': video_id,
  48             'title': clip.find('title').text,
  49             'formats': formats,
  50             'description': description,
  51             'duration': int(clip.find('duration').text),
  52         }
  53         # TODO: Remove when #980 has been merged
  54         info.update(formats[-1])
  55         return info