]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/gametrailers.py
33e59e82c47f10d37c0004de073757568d25347b
[youtubedl] / 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
13 def _real_extract(self, url):
14 mobj = re.match(self._VALID_URL, url)
15 if mobj is None:
16 raise ExtractorError(u'Invalid URL: %s' % url)
17 video_id = mobj.group('id')
18 video_type = mobj.group('type')
19 webpage = self._download_webpage(url, video_id)
20 if video_type == 'full-episodes':
21 mgid_re = r'data-video="(?P<mgid>mgid:.*?)"'
22 else:
23 mgid_re = r'data-contentId=\'(?P<mgid>mgid:.*?)\''
24 mgid = self._search_regex(mgid_re, webpage, u'mgid')
25 data = compat_urllib_parse.urlencode({'uri': mgid, 'acceptMethods': 'fms'})
26
27 info_page = self._download_webpage('http://www.gametrailers.com/feeds/mrss?' + data,
28 video_id, u'Downloading video info')
29 links_webpage = self._download_webpage('http://www.gametrailers.com/feeds/mediagen/?' + data,
30 video_id, u'Downloading video urls info')
31
32 self.report_extraction(video_id)
33 info_re = r'''<title><!\[CDATA\[(?P<title>.*?)\]\]></title>.*
34 <description><!\[CDATA\[(?P<description>.*?)\]\]></description>.*
35 <image>.*
36 <url>(?P<thumb>.*?)</url>.*
37 </image>'''
38
39 m_info = re.search(info_re, info_page, re.VERBOSE|re.DOTALL)
40 if m_info is None:
41 raise ExtractorError(u'Unable to extract video info')
42 video_title = m_info.group('title')
43 video_description = m_info.group('description')
44 video_thumb = m_info.group('thumb')
45
46 m_urls = list(re.finditer(r'<src>(?P<url>.*)</src>', links_webpage))
47 if m_urls is None or len(m_urls) == 0:
48 raise ExtractorError(u'Unable to extract video url')
49 # They are sorted from worst to best quality
50 video_url = m_urls[-1].group('url')
51
52 return {'url': video_url,
53 'id': video_id,
54 'title': video_title,
55 # Videos are actually flv not mp4
56 'ext': 'flv',
57 'thumbnail': video_thumb,
58 'description': video_description,
59 }