GameSpotIE: support more urls and download videos in the best quality

This commit is contained in:
Jaime Marquínez Ferrándiz 2013-07-09 20:07:52 +02:00
parent c108eb73cc
commit ebdf2af727

View file

@ -4,14 +4,15 @@
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import ( from ..utils import (
unified_strdate, unified_strdate,
compat_urllib_parse,
) )
class GameSpotIE(InfoExtractor): class GameSpotIE(InfoExtractor):
_VALID_URL = r'(?:http://)?(?:www\.)?gamespot\.com/([^/]+)/videos/([^/]+)-([^/d]+)/' _VALID_URL = r'(?:http://)?(?:www\.)?gamespot\.com/.*-(?P<page_id>\d+)/?'
_TEST = { _TEST = {
u"url": u"http://www.gamespot.com/arma-iii/videos/arma-iii-community-guide-sitrep-i-6410818/", u"url": u"http://www.gamespot.com/arma-iii/videos/arma-iii-community-guide-sitrep-i-6410818/",
u"file": u"6410818.mp4", u"file": u"6410818.mp4",
u"md5": u"5569d64ca98db01f0177c934fe8c1e9b", u"md5": u"b2a30deaa8654fcccd43713a6b6a4825",
u"info_dict": { u"info_dict": {
u"title": u"Arma III - Community Guide: SITREP I", u"title": u"Arma III - Community Guide: SITREP I",
u"upload_date": u"20130627", u"upload_date": u"20130627",
@ -21,13 +22,22 @@ class GameSpotIE(InfoExtractor):
def _real_extract(self, url): def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url) mobj = re.match(self._VALID_URL, url)
video_id = mobj.group(3).split("-")[-1] page_id = mobj.group('page_id')
info_url = "http://www.gamespot.com/pages/video_player/xml.php?id="+str(video_id) webpage = self._download_webpage(url, page_id)
video_id = self._html_search_regex([r'"og:video" content=".*?\?id=(\d+)"',
r'http://www\.gamespot\.com/videoembed/(\d+)'],
webpage, 'video id')
data = compat_urllib_parse.urlencode({'id': video_id, 'newplayer': '1'})
info_url = 'http://www.gamespot.com/pages/video_player/xml.php?' + data
info_xml = self._download_webpage(info_url, video_id) info_xml = self._download_webpage(info_url, video_id)
doc = xml.etree.ElementTree.fromstring(info_xml) doc = xml.etree.ElementTree.fromstring(info_xml)
clip_el = doc.find('./playList/clip') clip_el = doc.find('./playList/clip')
video_url = clip_el.find('./URI').text http_urls = [{'url': node.find('filePath').text,
'rate': int(node.find('rate').text)}
for node in clip_el.find('./httpURI')]
best_quality = sorted(http_urls, key=lambda f: f['rate'])[-1]
video_url = best_quality['url']
title = clip_el.find('./title').text title = clip_el.find('./title').text
ext = video_url.rpartition('.')[2] ext = video_url.rpartition('.')[2]
thumbnail_url = clip_el.find('./screenGrabURI').text thumbnail_url = clip_el.find('./screenGrabURI').text