X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/bc5059672e222ec4df4fe3983c399c0d0fca84d5..a5a28b1eabe0533f26d23c4157c97144215b22b7:/youtube_dl/extractor/mtv.py?ds=sidebyside diff --git a/youtube_dl/extractor/mtv.py b/youtube_dl/extractor/mtv.py index 6b3feb5..7a3b57a 100644 --- a/youtube_dl/extractor/mtv.py +++ b/youtube_dl/extractor/mtv.py @@ -1,119 +1,404 @@ +from __future__ import unicode_literals + import re -import xml.etree.ElementTree from .common import InfoExtractor +from ..compat import ( + compat_str, + compat_xpath, +) from ..utils import ( - compat_urllib_parse, ExtractorError, + find_xpath_attr, + fix_xml_ampersands, + float_or_none, + HEADRequest, + RegexNotFoundError, + sanitized_Request, + strip_or_none, + timeconvert, + try_get, + 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] - # 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\..+?/.*)$', rtmp_video_url) - if not m: - return rtmp_video_url - base = 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=1+_pxI0=Ripod-h264+_pxL0=undefined+_pxM0=+_pxK=18639+_pxE=mp4/44620/mtvnorigin/' - return base + m.group('finalid') + def _remove_template_parameter(url): + # Remove the templates, like &device={device} + return re.sub(r'&[^=]*?={.*?}(?=(&|$))', '', url) + + 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'] + return thumb_node.get('url') or thumb_node.text or None + + 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'.+?_lc_promo.*?)\1', webpage, + 'data zone', default=data_zone, group='zone') + + feed_url = try_get( + triforce_feed, lambda x: x['manifest']['zones'][data_zone]['feed'], + compat_str) + if not feed_url: + return + + feed = self._download_json(feed_url, video_id, fatal=False) + if not feed: + return + + return try_get(feed, lambda x: x['result']['data']['id'], compat_str) + + def _extract_mgid(self, webpage): + 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 = None + + if mgid is None or ':' not in mgid: + mgid = self._search_regex( + [r'data-mgid="(.*?)"', r'swfobject\.embedSWF\(".*?(mgid:.*?)"'], + webpage, 'mgid', default=None) + + if not mgid: + sm4_embed = self._html_search_meta( + 'sm4:video:embed', webpage, 'sm4 embed', default='') + mgid = self._search_regex( + r'embed/(mgid:.+?)["\'&?/]', sm4_embed, 'mgid', default=None) + + if not mgid: + mgid = self._extract_triforce_mgid(webpage) + + return mgid + + def _real_extract(self, url): + title = url_basename(url) + webpage = self._download_webpage(url, title) + mgid = self._extract_mgid(webpage) + videos_info = self._get_videos_info(mgid) + return videos_info + + +class MTVServicesEmbeddedIE(MTVServicesInfoExtractor): + IE_NAME = 'mtvservices:embedded' + _VALID_URL = r'https?://media\.mtvnservices\.com/embed/(?P.+?)(\?|/|$)' + + _TEST = { + # From http://www.thewrap.com/peter-dinklage-sums-up-game-of-thrones-in-45-seconds-video/ + 'url': 'http://media.mtvnservices.com/embed/mgid:uma:video:mtv.com:1043906/cp~vid%3D1043906%26uri%3Dmgid%3Auma%3Avideo%3Amtv.com%3A1043906', + 'md5': 'cb349b21a7897164cede95bd7bf3fbb9', + 'info_dict': { + 'id': '1043906', + 'ext': 'mp4', + 'title': 'Peter Dinklage Sums Up \'Game Of Thrones\' In 45 Seconds', + 'description': '"Sexy sexy sexy, stabby stabby stabby, beautiful language," says Peter Dinklage as he tries summarizing "Game of Thrones" in under a minute.', + 'timestamp': 1400126400, + 'upload_date': '20140515', + }, + } + + @staticmethod + def _extract_url(webpage): + mobj = re.search( + r']+?src=(["\'])(?P(?:https?:)?//media.mtvnservices.com/embed/.+?)\1', webpage) + if mobj: + return mobj.group('url') + + def _get_feed_url(self, uri): video_id = self._id_from_uri(uri) - data = compat_urllib_parse.urlencode({'uri': uri}) - idoc = self._download_xml(self._FEED_URL +'?' + data, video_id, - u'Downloading info') - return [self._get_video_info(item) for item in idoc.findall('.//item')] + config = self._download_json( + 'http://media.mtvnservices.com/pmt/e1/access/index.html?uri=%s&configtype=edge' % uri, video_id) + return self._remove_template_parameter(config['feedWithQueryParams']) + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + mgid = mobj.group('mgid') + return self._get_videos_info(mgid) class MTVIE(MTVServicesInfoExtractor): - _VALID_URL = r'^https?://(?:www\.)?mtv\.com/videos/.+?/(?P[0-9]+)/[^/]+$' + IE_NAME = 'mtv' + _VALID_URL = r'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, + }, { + 'url': 'http://www.mtv.com/episodes/g8xu7q/teen-mom-2-breaking-the-wall-season-7-ep-713', + 'only_matching': True, + }] + + +class MTV81IE(InfoExtractor): + IE_NAME = 'mtv81' + _VALID_URL = r'https?://(?:www\.)?mtv81\.com/videos/(?P[^/?#.]+)' + + _TEST = { + 'url': 'http://www.mtv81.com/videos/artist-to-watch/the-godfather-of-japanese-hip-hop-segment-1/', + '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', + }, + } + + def _extract_mgid(self, webpage): + return self._search_regex( + r'getTheVideo\((["\'])(?Pmgid:.+?)\1', webpage, + 'mgid', group='id') + + def _real_extract(self, url): + video_id = self._match_id(url) + webpage = self._download_webpage(url, video_id) + mgid = self._extract_mgid(webpage) + return self.url_result('http://media.mtvnservices.com/embed/%s' % mgid) + + +class MTVVideoIE(MTVServicesInfoExtractor): + IE_NAME = 'mtv:video' + _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 = [ { - u'url': u'http://www.mtv.com/videos/misc/853555/ours-vh1-storytellers.jhtml', - u'file': u'853555.mp4', - u'md5': u'850f3f143316b1e71fa56a4edfd6e0f8', - u'info_dict': { - u'title': u'Taylor Swift - "Ours (VH1 Storytellers)"', - u'description': u'Album: Taylor Swift performs "Ours" for VH1 Storytellers at Harvey Mudd College.', + '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', }, }, - { - u'add_ie': ['Vevo'], - u'url': u'http://www.mtv.com/videos/taylor-swift/916187/everything-has-changed-ft-ed-sheeran.jhtml', - u'file': u'USCJY1331283.mp4', - u'md5': u'73b4e7fcadd88929292fe52c3ced8caf', - u'info_dict': { - u'title': u'Everything Has Changed', - u'upload_date': u'20130606', - u'uploader': u'Taylor Swift', - }, - u'skip': u'VEVO is only available in some countries', - }, ] def _get_thumbnail_url(self, uri, itemdoc): @@ -122,16 +407,84 @@ class MTVIE(MTVServicesInfoExtractor): 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) - # 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(u'Vevo video detected: %s' % vevo_id) - return self.url_result('vevo:%s' % vevo_id, ie='Vevo') + playlist = self._parse_json( + self._search_regex( + r'window\.pagePlaylist\s*=\s*(\[.+?\]);\n', webpage, 'page playlist'), + video_id) - uri = self._html_search_regex(r'/uri/(.*?)\?', webpage, u'uri') - return self._get_videos_info(uri) + 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)