from __future__ import unicode_literals
import re
-import xml.etree.ElementTree
from .common import InfoExtractor
from ..utils import (
compat_urllib_parse,
+ compat_urllib_request,
ExtractorError,
+ find_xpath_attr,
fix_xml_ampersands,
+ HEADRequest,
+ unescapeHTML,
+ url_basename,
+ RegexNotFoundError,
)
+
def _media_xml_tag(tag):
return '{http://search.yahoo.com/mrss/}%s' % tag
class MTVServicesInfoExtractor(InfoExtractor):
+ _MOBILE_TEMPLATE = None
@staticmethod
def _id_from_uri(uri):
return uri.split(':')[-1]
else:
return thumb_node.attrib['url']
- def _extract_video_formats(self, metadataXml):
- if '/error_country_block.swf' in metadataXml:
- raise ExtractorError('This video is not available from your country.', expected=True)
- mdoc = xml.etree.ElementTree.fromstring(metadataXml.encode('utf-8'))
+ def _extract_mobile_video_formats(self, mtvn_id):
+ webpage_url = self._MOBILE_TEMPLATE % mtvn_id
+ req = compat_urllib_request.Request(webpage_url)
+ # Otherwise we get a webpage that would execute some javascript
+ req.add_header('Youtubedl-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()
+ # 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'}]
+
+ def _extract_video_formats(self, mdoc, mtvn_id):
+ if re.match(r'.*/(error_country_block\.swf|geoblock\.mp4)$', mdoc.find('.//src').text) is not None:
+ if mtvn_id is not None and self._MOBILE_TEMPLATE is not None:
+ self.to_screen('The normal version is not available from your '
+ 'country, trying with the mobile version')
+ return self._extract_mobile_video_formats(mtvn_id)
+ raise ExtractorError('This video is not available from your country.',
+ expected=True)
formats = []
for rendition in mdoc.findall('.//rendition'):
mediagen_url = re.sub(r'&[^=]*?={.*?}(?=(&|$))', '', mediagen_url)
if 'acceptMethods' not in mediagen_url:
mediagen_url += '&acceptMethods=fms'
- mediagen_page = self._download_webpage(mediagen_url, video_id,
- 'Downloading video urls')
+
+ mediagen_doc = self._download_xml(mediagen_url, video_id,
+ 'Downloading video urls')
description_node = itemdoc.find('description')
if description_node is not None:
else:
description = None
+ title_el = None
+ if title_el is None:
+ title_el = find_xpath_attr(
+ itemdoc, './/{http://search.yahoo.com/mrss/}category',
+ 'scheme', 'urn:mtvn:video_title')
+ if title_el is None:
+ title_el = itemdoc.find('.//{http://search.yahoo.com/mrss/}title')
+ if title_el is None:
+ title_el = itemdoc.find('.//title')
+ if title_el.text is None:
+ title_el = None
+
+ title = title_el.text
+ if title is None:
+ raise ExtractorError('Could not find video title')
+ title = title.strip()
+
+ # This a short id that's used in the webpage urls
+ mtvn_id = None
+ mtvn_id_node = find_xpath_attr(itemdoc, './/{http://search.yahoo.com/mrss/}category',
+ 'scheme', 'urn:mtvn:id')
+ if mtvn_id_node is not None:
+ mtvn_id = mtvn_id_node.text
+
return {
- 'title': itemdoc.find('title').text,
- 'formats': self._extract_video_formats(mediagen_page),
+ 'title': title,
+ 'formats': self._extract_video_formats(mediagen_doc, mtvn_id),
'id': video_id,
'thumbnail': self._get_thumbnail_url(uri, itemdoc),
'description': description,
'Downloading info', transform_source=fix_xml_ampersands)
return [self._get_video_info(item) for item in idoc.findall('.//item')]
+ def _real_extract(self, url):
+ title = url_basename(url)
+ webpage = self._download_webpage(url, title)
+ try:
+ # the url can be http://media.mtvnservices.com/fb/{mgid}.swf
+ # or http://media.mtvnservices.com/{mgid}
+ og_url = self._og_search_video_url(webpage)
+ mgid = url_basename(og_url)
+ if mgid.endswith('.swf'):
+ mgid = mgid[:-4]
+ except RegexNotFoundError:
+ mgid = self._search_regex(
+ [r'data-mgid="(.*?)"', r'swfobject.embedSWF\(".*?(mgid:.*?)"'],
+ webpage, u'mgid')
+ return self._get_videos_info(mgid)
+
class MTVIE(MTVServicesInfoExtractor):
_VALID_URL = r'''(?x)^https?://
uri = self._html_search_regex(r'/uri/(.*?)\?', webpage, 'uri')
return self._get_videos_info(uri)
+
+
+class MTVIggyIE(MTVServicesInfoExtractor):
+ IE_NAME = 'mtviggy.com'
+ _VALID_URL = r'https?://www\.mtviggy\.com/videos/.+'
+ _TEST = {
+ 'url': 'http://www.mtviggy.com/videos/arcade-fire-behind-the-scenes-at-the-biggest-music-experiment-yet/',
+ 'info_dict': {
+ 'id': '984696',
+ 'ext': 'mp4',
+ 'title': 'Arcade Fire: Behind the Scenes at the Biggest Music Experiment Yet',
+ }
+ }
+ _FEED_URL = 'http://all.mtvworldverticals.com/feed-xml/'
projects / youtube-dl / blobdiff