from __future__ import unicode_literals import re from .common import InfoExtractor class IGNIE(InfoExtractor): """ Extractor for some of the IGN sites, like www.ign.com, es.ign.com de.ign.com. Some videos of it.ign.com are also supported """ _VALID_URL = r'https?://.+?\.ign\.com/(?Pvideos|show_videos|articles|(?:[^/]*/feature))(/.+)?/(?P.+)' IE_NAME = 'ign.com' _CONFIG_URL_TEMPLATE = 'http://www.ign.com/videos/configs/id/%s.config' _DESCRIPTION_RE = [ r'(.+?)', r'id="my_show_video">.*?

(.*?)

', r']*id="(.+?)"', ] return self._search_regex(res_id, webpage, 'video id') def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) name_or_id = mobj.group('name_or_id') page_type = mobj.group('type') webpage = self._download_webpage(url, name_or_id) if page_type != 'video': multiple_urls = re.findall( ']*value="[^"]*?url=(https?://www\.ign\.com/videos/.*?)["&]', webpage) if multiple_urls: return [self.url_result(u, ie='IGN') for u in multiple_urls] video_id = self._find_video_id(webpage) result = self._get_video_info(video_id) description = self._html_search_regex(self._DESCRIPTION_RE, webpage, 'video description', flags=re.DOTALL) result['description'] = description return result def _get_video_info(self, video_id): config_url = self._CONFIG_URL_TEMPLATE % video_id config = self._download_json(config_url, video_id) media = config['playlist']['media'] return { 'id': media['metadata']['videoId'], 'url': media['url'], 'title': media['metadata']['title'], 'thumbnail': media['poster'][0]['url'].replace('{size}', 'grande'), } class OneUPIE(IGNIE): _VALID_URL = r'https?://gamevideos\.1up\.com/(?Pvideo)/id/(?P.+)' IE_NAME = '1up.com' _DESCRIPTION_RE = r'
(.+?)
' _TESTS = [{ 'url': 'http://gamevideos.1up.com/video/id/34976', 'md5': '68a54ce4ebc772e4b71e3123d413163d', 'info_dict': { 'id': '34976', 'ext': 'mp4', 'title': 'Sniper Elite V2 - Trailer', 'description': 'md5:5d289b722f5a6d940ca3136e9dae89cf', } }] def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) result = super(OneUPIE, self)._real_extract(url) result['id'] = mobj.group('name_or_id') return result