_ Git - youtube-dl/blob - youtube_dl/extractor/gametrailers.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     compat_urllib_parse,
   6
   7     ExtractorError,
   8 )
   9
  10 class GametrailersIE(InfoExtractor):
  11     _VALID_URL = r'http://www.gametrailers.com/(?P<type>videos|reviews|full-episodes)/(?P<id>.*?)/(?P<title>.*)'
  12     _TEST = {
  13         u'url': u'http://www.gametrailers.com/videos/zbvr8i/mirror-s-edge-2-e3-2013--debut-trailer',
  14         u'file': u'zbvr8i.flv',
  15         u'md5': u'c3edbc995ab4081976e16779bd96a878',
  16         u'info_dict': {
  17             u"title": u"E3 2013: Debut Trailer"
  18         },
  19         u'skip': u'Requires rtmpdump'
  20     }
  21
  22     def _real_extract(self, url):
  23         mobj = re.match(self._VALID_URL, url)
  24         if mobj is None:
  25             raise ExtractorError(u'Invalid URL: %s' % url)
  26         video_id = mobj.group('id')
  27         video_type = mobj.group('type')
  28         webpage = self._download_webpage(url, video_id)
  29         if video_type == 'full-episodes':
  30             mgid_re = r'data-video="(?P<mgid>mgid:.*?)"'
  31         else:
  32             mgid_re = r'data-contentId=\'(?P<mgid>mgid:.*?)\''
  33         mgid = self._search_regex(mgid_re, webpage, u'mgid')
  34         data = compat_urllib_parse.urlencode({'uri': mgid, 'acceptMethods': 'fms'})
  35
  36         info_page = self._download_webpage('http://www.gametrailers.com/feeds/mrss?' + data,
  37                                            video_id, u'Downloading video info')
  38         links_webpage = self._download_webpage('http://www.gametrailers.com/feeds/mediagen/?' + data,
  39                                                video_id, u'Downloading video urls info')
  40
  41         self.report_extraction(video_id)
  42         info_re = r'''<title><!\[CDATA\[(?P<title>.*?)\]\]></title>.*
  43                       <description><!\[CDATA\[(?P<description>.*?)\]\]></description>.*
  44                       <image>.*
  45                         <url>(?P<thumb>.*?)</url>.*
  46                       </image>'''
  47
  48         m_info = re.search(info_re, info_page, re.VERBOSE|re.DOTALL)
  49         if m_info is None:
  50             raise ExtractorError(u'Unable to extract video info')
  51         video_title = m_info.group('title')
  52         video_description = m_info.group('description')
  53         video_thumb = m_info.group('thumb')
  54
  55         m_urls = list(re.finditer(r'<src>(?P<url>.*)</src>', links_webpage))
  56         if m_urls is None or len(m_urls) == 0:
  57             raise ExtractorError(u'Unable to extract video url')
  58         # They are sorted from worst to best quality
  59         video_url = m_urls[-1].group('url')
  60
  61         return {'url':         video_url,
  62                 'id':          video_id,
  63                 'title':       video_title,
  64                 # Videos are actually flv not mp4
  65                 'ext':         'flv',
  66                 'thumbnail':   video_thumb,
  67                 'description': video_description,
  68                 }