+from __future__ import unicode_literals
+
import re
import json
class GameSpotIE(InfoExtractor):
_VALID_URL = r'(?:http://)?(?:www\.)?gamespot\.com/.*-(?P<page_id>\d+)/?'
_TEST = {
- u"url": u"http://www.gamespot.com/arma-iii/videos/arma-iii-community-guide-sitrep-i-6410818/",
- u"file": u"gs-2300-6410818.mp4",
- u"md5": u"b2a30deaa8654fcccd43713a6b6a4825",
- u"info_dict": {
- u"title": u"Arma 3 - Community Guide: SITREP I",
- u'description': u'Check out this video where some of the basics of Arma 3 is explained.',
+ 'url': 'http://www.gamespot.com/videos/arma-3-community-guide-sitrep-i/2300-6410818/',
+ 'md5': 'b2a30deaa8654fcccd43713a6b6a4825',
+ 'info_dict': {
+ 'id': 'gs-2300-6410818',
+ 'ext': 'mp4',
+ 'title': 'Arma 3 - Community Guide: SITREP I',
+ 'description': 'Check out this video where some of the basics of Arma 3 is explained.',
}
}
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
- page_id = video_id = mobj.group('page_id')
+ page_id = mobj.group('page_id')
webpage = self._download_webpage(url, page_id)
- data_video_json = self._search_regex(r'data-video=\'(.*?)\'', webpage, u'data video')
+ data_video_json = self._search_regex(r'data-video=["\'](.*?)["\']', webpage, 'data video')
data_video = json.loads(unescapeHTML(data_video_json))
# Transform the manifest url to a link to the mp4 files
f4m_url = data_video['videoStreams']['f4m_stream']
f4m_path = compat_urlparse.urlparse(f4m_url).path
QUALITIES_RE = r'((,\d+)+,?)'
- qualities = self._search_regex(QUALITIES_RE, f4m_path, u'qualities').strip(',').split(',')
+ qualities = self._search_regex(QUALITIES_RE, f4m_path, 'qualities').strip(',').split(',')
http_path = f4m_path[1:].split('/', 1)[1]
http_template = re.sub(QUALITIES_RE, r'%s', http_path)
http_template = http_template.replace('.csmil/manifest.f4m', '')
'format_id': q,
})
- info = {
+ return {
'id': data_video['guid'],
'title': compat_urllib_parse.unquote(data_video['title']),
'formats': formats,
'description': get_meta_content('description', webpage),
'thumbnail': self._og_search_thumbnail(webpage),
}
- # TODO: Remove when #980 has been merged
- info.update(formats[-1])
- return info