X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/467f1716f3b46ffc038848b9b08a833a7811be09..03342304420e5daeb428ffdcc7bbd2bbfecfa61a:/youtube_dl/extractor/gametrailers.py?ds=inline diff --git a/youtube_dl/extractor/gametrailers.py b/youtube_dl/extractor/gametrailers.py index 3cc02d9..c3f031d 100644 --- a/youtube_dl/extractor/gametrailers.py +++ b/youtube_dl/extractor/gametrailers.py @@ -1,36 +1,62 @@ -import re +from __future__ import unicode_literals -from .mtv import MTVIE, _media_xml_tag +from .common import InfoExtractor +from ..utils import ( + int_or_none, + parse_age_limit, + url_basename, +) + + +class GametrailersIE(InfoExtractor): + _VALID_URL = r'http://www\.gametrailers\.com/videos/view/[^/]+/(?P.+)' -class GametrailersIE(MTVIE): - """ - Gametrailers use the same videos system as MTVIE, it just changes the feed - url, where the uri is and the method to get the thumbnails. - """ - _VALID_URL = r'http://www.gametrailers.com/(?Pvideos|reviews|full-episodes)/(?P.*?)/(?P.*)' _TEST = { - u'url': u'http://www.gametrailers.com/videos/zbvr8i/mirror-s-edge-2-e3-2013--debut-trailer', - u'file': u'70e9a5d7-cf25-4a10-9104-6f3e7342ae0d.mp4', - u'md5': u'4c8e67681a0ea7ec241e8c09b3ea8cf7', - u'info_dict': { - u'title': u'E3 2013: Debut Trailer', - u'description': u'Faith is back! Check out the World Premiere trailer for Mirror\'s Edge 2 straight from the EA Press Conference at E3 2013!', + 'url': 'http://www.gametrailers.com/videos/view/gametrailers-com/116437-Just-Cause-3-Review', + 'md5': 'f28c4efa0bdfaf9b760f6507955b6a6a', + 'info_dict': { + 'id': '2983958', + 'ext': 'mp4', + 'display_id': '116437-Just-Cause-3-Review', + 'title': 'Just Cause 3 - Review', + 'description': 'It\'s a lot of fun to shoot at things and then watch them explode in Just Cause 3, but should there be more to the experience than that?', }, } - # Overwrite MTVIE properties we don't want - _TESTS = [] - _FEED_URL = 'http://www.gametrailers.com/feeds/mrss' + def _real_extract(self, url): + display_id = self._match_id(url) + webpage = self._download_webpage(url, display_id) + title = self._html_search_regex( + r'<title>(.+?)\|', webpage, 'title').strip() + embed_url = self._proto_relative_url( + self._search_regex( + r'src=\'(//embed.gametrailers.com/embed/[^\']+)\'', webpage, + 'embed url'), + scheme='http:') + video_id = url_basename(embed_url) + embed_page = self._download_webpage(embed_url, video_id) + embed_vars_json = self._search_regex( + r'(?s)var embedVars = (\{.*?\})\s*</script>', embed_page, + 'embed vars') + info = self._parse_json(embed_vars_json, video_id) - def _get_thumbnail_url(self, uri, itemdoc): - search_path = '%s/%s' % (_media_xml_tag('group'), _media_xml_tag('thumbnail')) - return itemdoc.find(search_path).attrib['url'] + formats = [] + for media in info['media']: + if media['mediaPurpose'] == 'play': + formats.append({ + 'url': media['uri'], + 'height': media['height'], + 'width:': media['width'], + }) + self._sort_formats(formats) - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') - webpage = self._download_webpage(url, video_id) - mgid = self._search_regex([r'data-video="(?P<mgid>mgid:.*?)"', - r'data-contentId=\'(?P<mgid>mgid:.*?)\''], - webpage, u'mgid') - return self._get_videos_info(mgid) + return { + 'id': video_id, + 'display_id': display_id, + 'title': title, + 'formats': formats, + 'thumbnail': info.get('thumbUri'), + 'description': self._og_search_description(webpage), + 'duration': int_or_none(info.get('videoLengthInSeconds')), + 'age_limit': parse_age_limit(info.get('audienceRating')), + }