from __future__ import unicode_literals import re from .common import InfoExtractor from ..compat import ( compat_str, compat_xpath, ) from ..utils import ( ExtractorError, find_xpath_attr, fix_xml_ampersands, float_or_none, HEADRequest, RegexNotFoundError, sanitized_Request, strip_or_none, timeconvert, unescapeHTML, update_url_query, url_basename, xpath_text, ) def _media_xml_tag(tag): return '{http://search.yahoo.com/mrss/}%s' % tag class MTVServicesInfoExtractor(InfoExtractor): _MOBILE_TEMPLATE = None _LANG = None @staticmethod def _id_from_uri(uri): return uri.split(':')[-1] @staticmethod def _remove_template_parameter(url): # Remove the templates, like &device={device} return re.sub(r'&[^=]*?={.*?}(?=(&|$))', '', url) # This was originally implemented for ComedyCentral, but it also works here @classmethod def _transform_rtmp_url(cls, rtmp_video_url): m = re.match(r'^rtmpe?://.*?/(?Pgsp\..+?/.*)$', rtmp_video_url) if not m: return {'rtmp': rtmp_video_url} base = 'http://viacommtvstrmfs.fplive.net/' return {'http': 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 = sanitized_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.', 'timestamp': 1400126400, 'upload_date': '20140515', }, } @staticmethod def _extract_url(webpage): mobj = re.search( r']+?src=(["\'])(?P(?:https?:)?//media.mtvnservices.com/embed/.+?)\1', webpage) if mobj: return mobj.group('url') def _get_feed_url(self, uri): 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) mgid = mobj.group('mgid') return self._get_videos_info(mgid) class MTVIE(MTVServicesInfoExtractor): _VALID_URL = r'(?x)https?://(?:www\.)?mtv\.com/(video-clips|full-episodes)/(?P[^/?#.]+)' _FEED_URL = 'http://www.mtv.com/feeds/mrss/' _TESTS = [{ 'url': 'http://www.mtv.com/video-clips/vl8qof/unlocking-the-truth-trailer', 'md5': '1edbcdf1e7628e414a8c5dcebca3d32b', 'info_dict': { 'id': '5e14040d-18a4-47c4-a582-43ff602de88e', 'ext': 'mp4', 'title': 'Unlocking The Truth|July 18, 2016|1|101|Trailer', 'description': '"Unlocking the Truth" premieres August 17th at 11/10c.', 'timestamp': 1468846800, 'upload_date': '20160718', }, }, { 'url': 'http://www.mtv.com/full-episodes/94tujl/unlocking-the-truth-gates-of-hell-season-1-ep-101', 'only_matching': True, }] class MTVVideoIE(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', 'md5': '850f3f143316b1e71fa56a4edfd6e0f8', 'info_dict': { 'id': '853555', 'ext': 'mp4', 'title': 'Taylor Swift - "Ours (VH1 Storytellers)"', 'description': 'Album: Taylor Swift performs "Ours" for VH1 Storytellers at Harvey Mudd College.', 'timestamp': 1352610000, 'upload_date': '20121111', }, }, ] 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'(?s)isVevoVideo = true;.*?vevoVideoId = "(.*?)";', webpage) 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 MTVDEIE(MTVServicesInfoExtractor): IE_NAME = 'mtv.de' _VALID_URL = r'https?://(?:www\.)?mtv\.de/(?:artists|shows|news)/(?:[^/]+/)*(?P\d+)-[^/#?]+/*(?:[#?].*)?$' _TESTS = [{ 'url': 'http://www.mtv.de/artists/10571-cro/videos/61131-traum', 'info_dict': { 'id': 'music_video-a50bc5f0b3aa4b3190aa', 'ext': 'flv', 'title': 'MusicVideo_cro-traum', 'description': 'Cro - Traum', }, 'params': { # rtmp download 'skip_download': True, }, 'skip': 'Blocked at Travis CI', }, { # mediagen URL without query (e.g. http://videos.mtvnn.com/mediagen/e865da714c166d18d6f80893195fcb97) 'url': 'http://www.mtv.de/shows/933-teen-mom-2/staffeln/5353/folgen/63565-enthullungen', 'info_dict': { 'id': 'local_playlist-f5ae778b9832cc837189', 'ext': 'flv', 'title': 'Episode_teen-mom-2_shows_season-5_episode-1_full-episode_part1', }, 'params': { # rtmp download 'skip_download': True, }, 'skip': 'Blocked at Travis CI', }, { 'url': 'http://www.mtv.de/news/77491-mtv-movies-spotlight-pixels-teil-3', 'info_dict': { 'id': 'local_playlist-4e760566473c4c8c5344', 'ext': 'mp4', 'title': 'Article_mtv-movies-spotlight-pixels-teil-3_short-clips_part1', 'description': 'MTV Movies Supercut', }, 'params': { # rtmp download 'skip_download': True, }, 'skip': 'Das Video kann zur Zeit nicht abgespielt werden.', }] def _real_extract(self, url): video_id = self._match_id(url) webpage = self._download_webpage(url, video_id) playlist = self._parse_json( self._search_regex( r'window\.pagePlaylist\s*=\s*(\[.+?\]);\n', webpage, 'page playlist'), video_id) def _mrss_url(item): return item['mrss'] + item.get('mrssvars', '') # news pages contain single video in playlist with different id if len(playlist) == 1: return self._get_videos_info_from_url(_mrss_url(playlist[0]), video_id) for item in playlist: item_id = item.get('id') if item_id and compat_str(item_id) == video_id: return self._get_videos_info_from_url(_mrss_url(item), video_id)