from __future__ import unicode_literals import re from .common import InfoExtractor from ..compat import ( compat_urllib_parse, compat_urllib_request, ) from ..utils import ( 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] # This was originally implemented for ComedyCentral, but it also works here @staticmethod def _transform_rtmp_url(rtmp_video_url): m = re.match(r'^rtmpe?://.*?/(?Pgsp\..+?/.*)$', rtmp_video_url) if not m: return rtmp_video_url base = 'http://viacommtvstrmfs.fplive.net/' return base + m.group('finalid') def _get_feed_url(self, uri): return self._FEED_URL def _get_thumbnail_url(self, uri, itemdoc): search_path = '%s/%s' % (_media_xml_tag('group'), _media_xml_tag('thumbnail')) thumb_node = itemdoc.find(search_path) if thumb_node is None: return None else: return thumb_node.attrib['url'] def _extract_mobile_video_formats(self, mtvn_id): webpage_url = self._MOBILE_TEMPLATE % mtvn_id req = compat_urllib_request.Request(webpage_url) # Otherwise we get a webpage that would execute some javascript req.add_header('User-Agent', 'curl/7') webpage = self._download_webpage(req, mtvn_id, 'Downloading mobile page') metrics_url = unescapeHTML(self._search_regex(r'.+?)(\?|/|$)' _TEST = { # From http://www.thewrap.com/peter-dinklage-sums-up-game-of-thrones-in-45-seconds-video/ 'url': 'http://media.mtvnservices.com/embed/mgid:uma:video:mtv.com:1043906/cp~vid%3D1043906%26uri%3Dmgid%3Auma%3Avideo%3Amtv.com%3A1043906', 'md5': 'cb349b21a7897164cede95bd7bf3fbb9', 'info_dict': { 'id': '1043906', 'ext': 'mp4', 'title': 'Peter Dinklage Sums Up \'Game Of Thrones\' In 45 Seconds', 'description': '"Sexy sexy sexy, stabby stabby stabby, beautiful language," says Peter Dinklage as he tries summarizing "Game of Thrones" in under a minute.', }, } def _get_feed_url(self, uri): video_id = self._id_from_uri(uri) site_id = uri.replace(video_id, '') config_url = ('http://media.mtvnservices.com/pmt/e1/players/{0}/' 'context4/context5/config.xml'.format(site_id)) config_doc = self._download_xml(config_url, video_id) feed_node = config_doc.find('.//feed') feed_url = feed_node.text.strip().split('?')[0] return feed_url def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) mgid = mobj.group('mgid') return self._get_videos_info(mgid) class MTVIE(MTVServicesInfoExtractor): _VALID_URL = r'''(?x)^https?:// (?:(?:www\.)?mtv\.com/videos/.+?/(?P[0-9]+)/[^/]+$| m\.mtv\.com/videos/video\.rbml\?.*?id=(?P[^&]+))''' _FEED_URL = 'http://www.mtv.com/player/embed/AS3/rss/' _TESTS = [ { 'url': 'http://www.mtv.com/videos/misc/853555/ours-vh1-storytellers.jhtml', 'file': '853555.mp4', 'md5': '850f3f143316b1e71fa56a4edfd6e0f8', 'info_dict': { 'title': 'Taylor Swift - "Ours (VH1 Storytellers)"', 'description': 'Album: Taylor Swift performs "Ours" for VH1 Storytellers at Harvey Mudd College.', }, }, { 'add_ie': ['Vevo'], 'url': 'http://www.mtv.com/videos/taylor-swift/916187/everything-has-changed-ft-ed-sheeran.jhtml', 'file': 'USCJY1331283.mp4', 'md5': '73b4e7fcadd88929292fe52c3ced8caf', 'info_dict': { 'title': 'Everything Has Changed', 'upload_date': '20130606', 'uploader': 'Taylor Swift', }, 'skip': 'VEVO is only available in some countries', }, ] def _get_thumbnail_url(self, uri, itemdoc): return 'http://mtv.mtvnimages.com/uri/' + uri def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) video_id = mobj.group('videoid') uri = mobj.groupdict().get('mgid') if uri is None: webpage = self._download_webpage(url, video_id) # Some videos come from Vevo.com m_vevo = re.search(r'isVevoVideo = true;.*?vevoVideoId = "(.*?)";', webpage, re.DOTALL) if m_vevo: vevo_id = m_vevo.group(1) self.to_screen('Vevo video detected: %s' % vevo_id) return self.url_result('vevo:%s' % vevo_id, ie='Vevo') 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/'