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

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