-# coding: utf-8
-from __future__ import unicode_literals
-
import re
from .common import InfoExtractor
-from ..compat import (
- compat_str,
- compat_xpath,
-)
+from ..compat import compat_str
from ..utils import (
ExtractorError,
find_xpath_attr,
float_or_none,
HEADRequest,
int_or_none,
+ join_nonempty,
RegexNotFoundError,
sanitized_Request,
strip_or_none,
# Remove the templates, like &device={device}
return re.sub(r'&[^=]*?={.*?}(?=(&|$))', '', url)
- def _get_feed_url(self, uri):
+ def _get_feed_url(self, uri, url=None):
return self._FEED_URL
def _get_thumbnail_url(self, uri, itemdoc):
formats.extend([{
'ext': 'flv' if rtmp_video_url.startswith('rtmp') else ext,
'url': rtmp_video_url,
- 'format_id': '-'.join(filter(None, [
+ 'format_id': join_nonempty(
'rtmp' if rtmp_video_url.startswith('rtmp') else None,
- rendition.get('bitrate')])),
+ rendition.get('bitrate')),
'width': int(rendition.get('width')),
'height': int(rendition.get('height')),
}])
itemdoc, './/{http://search.yahoo.com/mrss/}category',
'scheme', 'urn:mtvn:video_title')
if title_el is None:
- title_el = itemdoc.find(compat_xpath('.//{http://search.yahoo.com/mrss/}title'))
+ title_el = itemdoc.find('.//{http://search.yahoo.com/mrss/}title')
if title_el is None:
- title_el = itemdoc.find(compat_xpath('.//title'))
+ title_el = itemdoc.find('.//title')
if title_el.text is None:
title_el = None
data['lang'] = self._LANG
return data
- def _get_videos_info(self, uri, use_hls=True):
+ def _get_videos_info(self, uri, use_hls=True, url=None):
video_id = self._id_from_uri(uri)
- feed_url = self._get_feed_url(uri)
+ feed_url = self._get_feed_url(uri, url)
info_url = update_url_query(feed_url, self._get_feed_query(uri))
return self._get_videos_info_from_url(info_url, video_id, use_hls)
main_container = self._extract_child_with_type(data, 'MainContainer')
ab_testing = self._extract_child_with_type(main_container, 'ABTesting')
video_player = self._extract_child_with_type(ab_testing or main_container, 'VideoPlayer')
- mgid = video_player['props']['media']['video']['config']['uri']
+ if video_player:
+ mgid = try_get(video_player, lambda x: x['props']['media']['video']['config']['uri'])
+ else:
+ flex_wrapper = self._extract_child_with_type(ab_testing or main_container, 'FlexWrapper')
+ auth_suite_wrapper = self._extract_child_with_type(flex_wrapper, 'AuthSuiteWrapper')
+ player = self._extract_child_with_type(auth_suite_wrapper or flex_wrapper, 'Player')
+ if player:
+ mgid = try_get(player, lambda x: x['props']['videoDetail']['mgid'])
if not mgid:
- mgid = self._search_regex(
- r'"media":{"video":{"config":{"uri":"(mgid:.*?)"', webpage, 'mgid', default=None)
+ raise ExtractorError('Could not extract mgid')
return mgid
title = url_basename(url)
webpage = self._download_webpage(url, title)
mgid = self._extract_mgid(webpage)
- videos_info = self._get_videos_info(mgid)
+ videos_info = self._get_videos_info(mgid, url=url)
return videos_info
class MTVServicesEmbeddedIE(MTVServicesInfoExtractor):
IE_NAME = 'mtvservices:embedded'
_VALID_URL = r'https?://media\.mtvnservices\.com/embed/(?P<mgid>.+?)(\?|/|$)'
+ _EMBED_REGEX = [r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//media\.mtvnservices\.com/embed/.+?)\1']
_TEST = {
# From http://www.thewrap.com/peter-dinklage-sums-up-game-of-thrones-in-45-seconds-video/
},
}
- @staticmethod
- def _extract_url(webpage):
- mobj = re.search(
- r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//media\.mtvnservices\.com/embed/.+?)\1', webpage)
- if mobj:
- return mobj.group('url')
-
- def _get_feed_url(self, uri):
+ def _get_feed_url(self, uri, url=None):
video_id = self._id_from_uri(uri)
config = self._download_json(
'http://media.mtvnservices.com/pmt/e1/access/index.html?uri=%s&configtype=edge' % uri, video_id)