-# coding: utf-8
-from __future__ import unicode_literals
-
import re
+import xml.etree.ElementTree
from .common import InfoExtractor
-from ..compat import (
- compat_str,
- compat_xpath,
-)
+from ..compat import compat_str
+from ..networking import HEADRequest, Request
from ..utils import (
ExtractorError,
+ RegexNotFoundError,
find_xpath_attr,
fix_xml_ampersands,
float_or_none,
- HEADRequest,
int_or_none,
- RegexNotFoundError,
- sanitized_Request,
+ join_nonempty,
strip_or_none,
timeconvert,
try_get,
# 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):
def _extract_mobile_video_formats(self, mtvn_id):
webpage_url = self._MOBILE_TEMPLATE % mtvn_id
- req = sanitized_Request(webpage_url)
+ req = Request(webpage_url)
# Otherwise we get a webpage that would execute some javascript
- req.add_header('User-Agent', 'curl/7')
+ req.headers['User-Agent'] = 'curl/7'
webpage = self._download_webpage(req, mtvn_id,
'Downloading mobile page')
metrics_url = unescapeHTML(self._search_regex(r'<a href="(http://metrics.+?)"', webpage, 'url'))
req = HEADRequest(metrics_url)
response = self._request_webpage(req, mtvn_id, 'Resolving url')
- url = response.geturl()
+ url = response.url
# Transform the url to get the best quality:
url = re.sub(r'.+pxE=mp4', 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=0+_pxK=18639+_pxE=mp4', url, 1)
return [{'url': url, 'ext': 'mp4'}]
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')),
}])
except (KeyError, TypeError):
raise ExtractorError('Invalid rendition field.')
- if formats:
- self._sort_formats(formats)
return formats
def _extract_subtitles(self, mdoc, mtvn_id):
mediagen_doc = self._download_xml(
mediagen_url, video_id, 'Downloading video urls', fatal=False)
- if mediagen_doc is False:
+ if not isinstance(mediagen_doc, xml.etree.ElementTree.Element):
return None
item = mediagen_doc.find('./video/item')
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
if not formats:
return None
- self._sort_formats(formats)
-
return {
'title': title,
'formats': formats,
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)
if info:
entries.append(info)
+ # TODO: should be multi-video
return self.playlist_result(
entries, playlist_title=title, playlist_description=description)
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:
+ 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)
return self._remove_template_parameter(config['feedWithQueryParams'])
def _real_extract(self, url):
- mobj = re.match(self._VALID_URL, url)
+ mobj = self._match_valid_url(url)
mgid = mobj.group('mgid')
return self._get_videos_info(mgid)
return 'http://mtv.mtvnimages.com/uri/' + uri
def _real_extract(self, url):
- mobj = re.match(self._VALID_URL, url)
+ mobj = self._match_valid_url(url)
video_id = mobj.group('videoid')
uri = mobj.groupdict().get('mgid')
if uri is None:
}
-class MTVItaliaProgrammaIE(MTVItaliaIE):
+class MTVItaliaProgrammaIE(MTVItaliaIE): # XXX: Do not subclass from concrete IE
IE_NAME = 'mtv.it:programma'
_VALID_URL = r'https?://(?:www\.)?mtv\.it/(?:programmi|playlist)/(?P<id>[0-9a-z]+)'
_TESTS = [{