X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/415fdb62500dca2e22067a05008dfbf87c75b662..2e86127b07f6a63b7ce457331d600d9486b22b44:/youtube_dl/extractor/mtv.py?ds=sidebyside diff --git a/youtube_dl/extractor/mtv.py b/youtube_dl/extractor/mtv.py index e5ca41b..b482d6d 100644 --- a/youtube_dl/extractor/mtv.py +++ b/youtube_dl/extractor/mtv.py @@ -22,6 +22,7 @@ def _media_xml_tag(tag): class MTVServicesInfoExtractor(InfoExtractor): _MOBILE_TEMPLATE = None + @staticmethod def _id_from_uri(uri): return uri.split(':')[-1] @@ -32,9 +33,12 @@ class MTVServicesInfoExtractor(InfoExtractor): m = re.match(r'^rtmpe?://.*?/(?Pgsp\..+?/.*)$', rtmp_video_url) if not m: return rtmp_video_url - base = 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=1+_pxI0=Ripod-h264+_pxL0=undefined+_pxM0=+_pxK=18639+_pxE=mp4/44620/mtvnorigin/' + base = 'http://viacommtvstrmfs.fplive.net/' return base + m.group('finalid') + def _get_feed_url(self, uri): + return self._FEED_URL + def _get_thumbnail_url(self, uri, itemdoc): search_path = '%s/%s' % (_media_xml_tag('group'), _media_xml_tag('thumbnail')) thumb_node = itemdoc.find(search_path) @@ -49,23 +53,23 @@ class MTVServicesInfoExtractor(InfoExtractor): # Otherwise we get a webpage that would execute some javascript req.add_header('Youtubedl-user-agent', 'curl/7') webpage = self._download_webpage(req, mtvn_id, - 'Downloading mobile page') + 'Downloading mobile page') metrics_url = unescapeHTML(self._search_regex(r'.+?)(\?|/|$)' + + _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) @@ -200,15 +240,15 @@ class MTVIE(MTVServicesInfoExtractor): 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); + 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)