aboutsummaryrefslogtreecommitdiff
path: root/youtube_dl/extractor
diff options
context:
space:
mode:
authorJaime Marquínez Ferrándiz <jaime.marquinez.ferrandiz@gmail.com>2013-07-14 14:29:15 +0200
committerJaime Marquínez Ferrándiz <jaime.marquinez.ferrandiz@gmail.com>2013-07-14 14:29:15 +0200
commitab2f744b906ad4a896746d1648505e8f5efa815a (patch)
tree3c069a25f2d2e9797e1275d91927611e88cece38 /youtube_dl/extractor
parent300fcad8a663fc0bebde541c00535b931ad1af67 (diff)
GametrailersIE: make it a subclass of MTVIE to reuse most of the extraction process
Diffstat (limited to 'youtube_dl/extractor')
-rw-r--r--youtube_dl/extractor/gametrailers.py65
-rw-r--r--youtube_dl/extractor/mtv.py7
2 files changed, 24 insertions, 48 deletions
diff --git a/youtube_dl/extractor/gametrailers.py b/youtube_dl/extractor/gametrailers.py
index cd438bd2f..3cc02d97e 100644
--- a/youtube_dl/extractor/gametrailers.py
+++ b/youtube_dl/extractor/gametrailers.py
@@ -1,63 +1,36 @@
import re
-import xml.etree.ElementTree
-from .common import InfoExtractor
-from ..utils import (
- compat_urllib_parse,
+from .mtv import MTVIE, _media_xml_tag
- ExtractorError,
-)
-
-class GametrailersIE(InfoExtractor):
+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/(?P<type>videos|reviews|full-episodes)/(?P<id>.*?)/(?P<title>.*)'
_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.flv',
- u'md5': u'c3edbc995ab4081976e16779bd96a878',
+ u'file': u'70e9a5d7-cf25-4a10-9104-6f3e7342ae0d.mp4',
+ u'md5': u'4c8e67681a0ea7ec241e8c09b3ea8cf7',
u'info_dict': {
- u"title": u"E3 2013: Debut Trailer"
+ 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!',
},
- u'skip': u'Requires rtmpdump'
}
+ # Overwrite MTVIE properties we don't want
+ _TESTS = []
+
+ _FEED_URL = 'http://www.gametrailers.com/feeds/mrss'
+
+ 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']
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
- if mobj is None:
- raise ExtractorError(u'Invalid URL: %s' % 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')
-
- data = compat_urllib_parse.urlencode({'uri': mgid, 'acceptMethods': 'fms'})
- info_page = self._download_webpage('http://www.gametrailers.com/feeds/mrss?' + data,
- video_id, u'Downloading video info')
- doc = xml.etree.ElementTree.fromstring(info_page.encode('utf-8'))
- default_thumb = doc.find('./channel/image/url').text
-
- media_namespace = {'media': 'http://search.yahoo.com/mrss/'}
- parts = [{
- 'title': video_doc.find('title').text,
- 'ext': 'flv',
- 'id': video_doc.find('guid').text.rpartition(':')[2],
- # Videos are actually flv not mp4
- 'url': self._get_video_url(video_doc.find('media:group/media:content', media_namespace).attrib['url'], video_id),
- # The thumbnail may not be defined, it would be ''
- 'thumbnail': video_doc.find('media:group/media:thumbnail', media_namespace).attrib['url'] or default_thumb,
- 'description': video_doc.find('description').text,
- } for video_doc in doc.findall('./channel/item')]
- return parts
-
- def _get_video_url(self, mediagen_url, video_id):
- if 'acceptMethods' not in mediagen_url:
- mediagen_url += '&acceptMethods=fms'
- links_webpage = self._download_webpage(mediagen_url,
- video_id, u'Downloading video urls info')
- doc = xml.etree.ElementTree.fromstring(links_webpage)
- urls = list(doc.iter('src'))
- if len(urls) == 0:
- raise ExtractorError(u'Unable to extract video url')
- # They are sorted from worst to best quality
- return urls[-1].text
-
+ return self._get_videos_info(mgid)
diff --git a/youtube_dl/extractor/mtv.py b/youtube_dl/extractor/mtv.py
index 45dbc007d..8f956571d 100644
--- a/youtube_dl/extractor/mtv.py
+++ b/youtube_dl/extractor/mtv.py
@@ -49,7 +49,10 @@ class MTVIE(InfoExtractor):
if not m:
raise ExtractorError(u'Cannot transform RTMP url')
base = 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=1+_pxI0=Ripod-h264+_pxL0=undefined+_pxM0=+_pxK=18639+_pxE=mp4/44620/mtvnorigin/'
- return base + m.group('finalid')
+ return base + m.group('finalid')
+
+ def _get_thumbnail_url(self, uri, itemdoc):
+ return 'http://mtv.mtvnimages.com/uri/' + uri
def _extract_video_url(self, metadataXml):
if '/error_country_block.swf' in metadataXml:
@@ -87,7 +90,7 @@ class MTVIE(InfoExtractor):
description = None
video_info.update({'title': itemdoc.find('title').text,
'id': video_id,
- 'thumbnail': 'http://mtv.mtvnimages.com/uri/' + uri,
+ 'thumbnail': self._get_thumbnail_url(uri, itemdoc),
'description': description,
})
return video_info