3 from __future__ import unicode_literals
8 from .common import InfoExtractor
9 from .youtube import YoutubeIE
10 from ..compat import (
13 compat_xml_parse_error,
32 from .brightcove import BrightcoveIE
33 from .nbc import NBCSportsVPlayerIE
34 from .ooyala import OoyalaIE
35 from .rutv import RUTVIE
36 from .smotri import SmotriIE
37 from .condenast import CondeNastIE
38 from .udn import UDNEmbedIE
41 class GenericIE(InfoExtractor):
42 IE_DESC = 'Generic downloader that works on some sites'
47 'url': 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
48 'md5': '85b90ccc9d73b4acd9138d3af4c27f89',
50 'id': '13601338388002',
52 'uploader': 'www.hodiho.fr',
53 'title': 'R\u00e9gis plante sa Jeep',
56 # bandcamp page with custom domain
58 'add_ie': ['Bandcamp'],
59 'url': 'http://bronyrock.com/track/the-pony-mash',
63 'title': 'The Pony Mash',
64 'uploader': 'M_Pallante',
66 'skip': 'There is a limit of 200 free downloads / month for the test song',
68 # embedded brightcove video
69 # it also tests brightcove videos that need to set the 'Referer' in the
72 'add_ie': ['Brightcove'],
73 'url': 'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/',
75 'id': '2765128793001',
77 'title': 'Le cours de bourse : l’analyse technique',
78 'description': 'md5:7e9ad046e968cb2d1114004aba466fd9',
79 'uploader': 'BFM BUSINESS',
82 'skip_download': True,
86 # https://github.com/rg3/youtube-dl/issues/2253
87 'url': 'http://bcove.me/i6nfkrc3',
88 'md5': '0ba9446db037002366bab3b3eb30c88c',
90 'id': '3101154703001',
92 'title': 'Still no power',
93 'uploader': 'thestar.com',
94 'description': 'Mississauga resident David Farmer is still out of power as a result of the ice storm a month ago. To keep the house warm, Farmer cuts wood from his property for a wood burning stove downstairs.',
96 'add_ie': ['Brightcove'],
99 'url': 'http://www.championat.com/video/football/v/87/87499.html',
100 'md5': 'fb973ecf6e4a78a67453647444222983',
102 'id': '3414141473001',
104 'title': 'Видео. Удаление Дзагоева (ЦСКА)',
105 'description': 'Онлайн-трансляция матча ЦСКА - "Волга"',
106 'uploader': 'Championat',
110 # https://github.com/rg3/youtube-dl/issues/3541
111 'add_ie': ['Brightcove'],
112 'url': 'http://www.kijk.nl/sbs6/leermijvrouwenkennen/videos/jqMiXKAYan2S/aflevering-1',
114 'id': '3866516442001',
116 'title': 'Leer mij vrouwen kennen: Aflevering 1',
117 'description': 'Leer mij vrouwen kennen: Aflevering 1',
118 'uploader': 'SBS Broadcasting',
120 'skip': 'Restricted to Netherlands',
122 'skip_download': True, # m3u8 download
125 # Direct link to a video
127 'url': 'http://media.w3.org/2010/05/sintel/trailer.mp4',
128 'md5': '67d406c2bcb6af27fa886f31aa934bbe',
133 'upload_date': '20100513',
138 'url': 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219',
139 'md5': '166dd577b433b4d4ebfee10b0824d8ff',
141 'id': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ',
143 'title': '2cc213299525360.mov', # that's what we get
145 'add_ie': ['Ooyala'],
147 # multiple ooyala embeds on SBN network websites
149 'url': 'http://www.sbnation.com/college-football-recruiting/2015/2/3/7970291/national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
151 'id': 'national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
152 'title': '25 lies you will tell yourself on National Signing Day - SBNation.com',
154 'playlist_mincount': 3,
156 'skip_download': True,
158 'add_ie': ['Ooyala'],
162 'url': 'http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&ved=0CCUQtwIwAA&url=http%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DcmQHVoWB5FY&ei=F-sNU-LLCaXk4QT52ICQBQ&usg=AFQjCNEw4hL29zgOohLXvpJ-Bdh2bils1Q&bvm=bv.61965928,d.bGE',
166 'upload_date': '20130224',
167 'uploader_id': 'TheVerge',
168 'description': 're:^Chris Ziegler takes a look at the\.*',
169 'uploader': 'The Verge',
170 'title': 'First Firefox OS phones side-by-side',
173 'skip_download': False,
178 'url': 'http://www.tested.com/science/weird/460206-tested-grinding-coffee-2000-frames-second/',
182 'title': 'Tested: Grinding Coffee at 2000 Frames Per Second',
183 'upload_date': '20140225',
184 'description': 'md5:06a40fbf30b220468f1e0957c0f558ff',
185 'uploader': 'Tested',
186 'uploader_id': 'testedcom',
188 # No need to test YoutubeIE here
190 'skip_download': True,
195 'url': 'http://www.theguardian.com/world/2014/mar/11/obama-zach-galifianakis-between-two-ferns',
199 'title': 'Between Two Ferns with Zach Galifianakis: President Barack Obama',
200 'description': 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.',
205 'url': 'http://www.bbc.co.uk/blogs/adamcurtis/posts/BUGGER',
207 'title': 'BBC - Blogs - Adam Curtis - BUGGER',
209 'playlist_mincount': 18,
213 'url': 'http://www.rg.ru/2014/03/15/reg-dfo/anklav-anons.html',
217 'title': 'Охотское море стало целиком российским',
218 'description': 'md5:5ed62483b14663e2a95ebbe115eb8f43',
222 'skip_download': True,
227 'url': 'http://en.support.wordpress.com/videos/ted-talks/',
228 'md5': '65fdff94098e4a607385a60c5177c638',
232 'title': 'Hidden miracles of the natural world',
233 'uploader': 'Louie Schwartzberg',
234 'description': 'md5:8145d19d320ff3e52f28401f4c4283b9',
237 # Embeded Ustream video
239 'url': 'http://www.american.edu/spa/pti/nsa-privacy-janus-2014.cfm',
240 'md5': '27b99cdb639c9b12a79bca876a073417',
244 'uploader': 'AU SPA: The NSA and Privacy',
245 'title': 'NSA and Privacy Forum Debate featuring General Hayden and Barton Gellman'
248 # nowvideo embed hidden behind percent encoding
250 'url': 'http://www.waoanime.tv/the-super-dimension-fortress-macross-episode-1/',
251 'md5': '2baf4ddd70f697d94b1c18cf796d5107',
253 'id': '06e53103ca9aa',
255 'title': 'Macross Episode 001 Watch Macross Episode 001 onl',
256 'description': 'No description',
261 'url': 'http://www.tv-replay.fr/redirection/20-03-14/x-enius-arte-10753389.html',
262 'md5': '7653032cbb25bf6c80d80f217055fa43',
264 'id': '048195-004_PLUS7-F',
267 'description': 'md5:d5fdf32ef6613cdbfd516ae658abf168',
268 'upload_date': '20140320',
271 'skip_download': 'Requires rtmpdump'
276 'url': 'http://www.wired.com/2014/04/honda-asimo/',
277 'md5': 'ba0dfe966fa007657bd1443ee672db0f',
279 'id': '53501be369702d3275860000',
281 'title': 'Honda’s New Asimo Robot Is More Human Than Ever',
286 'url': 'http://www.spi0n.com/zap-spi0n-com-n216/',
287 'md5': '441aeeb82eb72c422c7f14ec533999cd',
289 'id': 'k2mm4bCdJ6CQ2i7c8o2',
291 'title': 'Le Zap de Spi0n n°216 - Zapping du Web',
294 'add_ie': ['Dailymotion'],
298 'url': 'http://www.badzine.de/ansicht/datum/2014/06/09/so-funktioniert-die-neue-englische-badminton-liga.html',
302 'title': 'The NBL Auction 2014',
303 'uploader': 'BADMINTON England',
304 'uploader_id': 'BADMINTONEvents',
305 'upload_date': '20140603',
306 'description': 'md5:9ef128a69f1e262a700ed83edb163a73',
308 'add_ie': ['Youtube'],
310 'skip_download': True,
315 'url': 'http://www.gametrailers.com/news-post/76093/north-america-europe-is-getting-that-mario-kart-8-mercedes-dlc-too',
316 'md5': '35727f82f58c76d996fc188f9755b0d5',
318 'id': '0306a69b-8adf-4fb5-aace-75f8e8cbfca9',
321 'description': 'Mario\'s life in the fast lane has never looked so good.',
324 # YouTube embed via <data-embed-url="">
326 'url': 'https://play.google.com/store/apps/details?id=com.gameloft.android.ANMP.GloftA8HM',
330 'title': 'Asphalt 8: Airborne - Update - Welcome to Dubai!',
331 'uploader': 'Gameloft',
332 'uploader_id': 'gameloft',
333 'upload_date': '20140828',
334 'description': 'md5:c80da9ed3d83ae6d1876c834de03e1c4',
337 'skip_download': True,
342 'url': 'http://www.ll.mit.edu/workshops/education/videocourses/antennas/lecture1/video/',
344 'md5': '0c5e352edabf715d762b0ad4e6d9ee67',
346 'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
347 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - video1',
352 'md5': '10e4bb3aaca9fd630e273ff92d9f3c63',
354 'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final_PIP',
355 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - pip',
361 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
366 'url': 'http://www.handjobhub.com/video/busty-blonde-siri-tit-fuck-while-wank-6313.html',
367 'md5': '9d65602bf31c6e20014319c7d07fba27',
369 'id': '5123ea6d5e5a7',
372 'uploader': 'www.handjobhub.com',
373 'title': 'Busty Blonde Siri Tit Fuck While Wank at HandjobHub.com',
378 'url': 'http://phihag.de/2014/youtube-dl/rss2.xml',
380 'id': 'http://phihag.de/2014/youtube-dl/rss2.xml',
381 'title': 'Zero Punctuation',
382 'description': 're:.*groundbreaking video review series.*'
384 'playlist_mincount': 11,
386 # Multiple brightcove videos
387 # https://github.com/rg3/youtube-dl/issues/2283
389 'url': 'http://www.newyorker.com/online/blogs/newsdesk/2014/01/always-never-nuclear-command-and-control.html',
391 'id': 'always-never',
392 'title': 'Always / Never - The New Yorker',
396 'extract_flat': False,
397 'skip_download': True,
402 'url': 'http://umpire-empire.com/index.php/topic/58125-laz-decides-no-thats-low/',
403 'md5': '96f09a37e44da40dd083e12d9a683327',
407 'title': 'Ump changes call to ball',
408 'description': 'md5:71c11215384298a172a6dcb4c2e20685',
410 'timestamp': 1401537900,
411 'upload_date': '20140531',
412 'thumbnail': 're:^https?://.*\.jpg$',
417 'url': 'http://education-portal.com/academy/lesson/north-american-exploration-failed-colonies-of-spain-france-england.html#lesson',
418 'md5': '8788b683c777a5cf25621eaf286d0c23',
422 'title': 'md5:51364a8d3d009997ba99656004b5e20d',
424 'filesize': 182808282,
425 'uploader': 'education-portal.com',
429 'url': 'http://thoughtworks.wistia.com/medias/uxjb0lwrcz',
430 'md5': 'baf49c2baa8a7de5f3fc145a8506dcd4',
434 'title': 'Conversation about Hexagonal Rails Part 1 - ThoughtWorks',
436 'uploader': 'thoughtworks.wistia.com',
439 # Direct download with broken HEAD
441 'url': 'http://ai-radio.org:8000/radio.opus',
448 'skip_download': True, # infinite live stream
450 'expected_warnings': [
451 r'501.*Not Implemented'
456 'url': 'http://nakedsecurity.sophos.com/2014/10/29/sscc-171-are-you-sure-that-1234-is-a-bad-password-podcast/',
460 'description': 'md5:ff867d6b555488ad3c52572bb33d432c',
461 'uploader': 'Sophos Security',
462 'title': 'Chet Chat 171 - Oct 29, 2014',
463 'upload_date': '20141029',
468 'url': 'http://www.esa.int/Our_Activities/Space_Science/Rosetta/Philae_comet_touch-down_webcast',
472 'upload_date': '20141112',
473 'title': 'Rosetta #CometLanding webcast HL 10',
478 'url': 'http://discourse.ubuntu.com/t/unity-8-desktop-mode-windows-on-mir/1986',
481 'title': 'Unity 8 desktop-mode windows on Mir! - Ubuntu Discourse',
483 'playlist_mincount': 2,
485 # Direct link with incorrect MIME type
487 'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
488 'md5': '4ccbebe5f36706d85221f204d7eb5913',
490 'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
491 'id': '5_Lennart_Poettering_-_Systemd',
493 'title': '5_Lennart_Poettering_-_Systemd',
494 'upload_date': '20141120',
496 'expected_warnings': [
497 'URL could be a direct video link, returning it as such.'
502 'url': 'http://undergroundwellness.com/podcasts/306-5-steps-to-permanent-gut-healing/',
506 'upload_date': '20141126',
507 'title': 'Jack Tips: 5 Steps to Permanent Gut Healing',
512 'url': 'http://www.abc.net.au/7.30/content/2015/s4164797.htm',
514 'id': '730m_DandD_1901_512k',
516 'uploader': 'www.abc.net.au',
517 'title': 'Game of Thrones with dice - Dungeons and Dragons fantasy role-playing game gets new life - 19/01/2015',
520 # embedded viddler video
522 'url': 'http://deadspin.com/i-cant-stop-watching-john-wall-chop-the-nuggets-with-th-1681801597',
526 'uploader': 'deadspin',
527 'title': 'WALL-TO-GORTAT',
528 'timestamp': 1422285291,
529 'upload_date': '20150126',
531 'add_ie': ['Viddler'],
535 'url': 'http://thedailyshow.cc.com/podcast/episodetwelve',
539 'title': "The Daily Show Podcast without Jon Stewart - Episode 12: Bassem Youssef: Egypt's Jon Stewart",
540 'description': 'md5:601cb790edd05908957dae8aaa866465',
541 'upload_date': '20150220',
546 'url': 'http://media.nationalarchives.gov.uk/index.php/webinar-using-discovery-national-archives-online-catalogue/',
550 'upload_date': '20150212',
551 'uploader': 'The National Archives UK',
552 'description': 'md5:a236581cd2449dd2df4f93412f3f01c6',
553 'uploader_id': 'NationalArchives08',
554 'title': 'Webinar: Using Discovery, The National Archives’ online catalogue',
559 'url': 'http://www.rtlnieuws.nl/nieuws/buitenland/aanslagen-kopenhagen',
560 'playlist_mincount': 5,
562 'id': 'aanslagen-kopenhagen',
563 'title': 'Aanslagen Kopenhagen | RTL Nieuws',
568 'url': 'http://www.skipass.com/news/116090-bon-appetit-s5ep3-baqueira-mi-cor.html',
572 'title': 'EP3S5 - Bon Appétit - Baqueira Mi Corazon !',
577 'url': 'http://www.monumentalnetwork.com/videos/john-carlson-postgame-2-25-15',
581 'upload_date': '20150226',
582 'uploader_id': 'MonumentalSports-Kaltura@perfectsensedigital.com',
584 'title': 'John Carlson Postgame 2/25/15',
587 # Eagle.Platform embed (generic URL)
589 'url': 'http://lenta.ru/news/2015/03/06/navalny/',
593 'title': 'Навальный вышел на свободу',
594 'description': 'md5:d97861ac9ae77377f3f20eaf9d04b4f5',
595 'thumbnail': 're:^https?://.*\.jpg$',
601 # ClipYou (Eagle.Platform) embed (custom URL)
603 'url': 'http://muz-tv.ru/play/7129/',
607 'title': "'O Sole Mio",
608 'thumbnail': 're:^https?://.*\.jpg$',
615 'url': 'http://muz-tv.ru/kinozal/view/7400/',
619 'title': 'Тайны перевала Дятлова • Тайна перевала Дятлова 1 серия 2 часть',
620 'description': 'Документальный сериал-расследование одной из самых жутких тайн ХХ века',
621 'thumbnail': 're:^https?://.*\.jpg$',
628 'url': 'http://techcrunch.com/video/facebook-creates-on-this-day-crunch-report/518726732/',
629 'md5': '4c6f127a30736b59b3e2c19234ee2bf7',
633 'title': 'Facebook Creates "On This Day" | Crunch Report',
636 # RSS feed with enclosure
638 'url': 'http://podcastfeeds.nbcnews.com/audio/podcast/MSNBC-MADDOW-NETCAST-M4V.xml',
640 'id': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
642 'upload_date': '20150228',
643 'title': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
646 # NBC Sports vplayer embed
648 'url': 'http://www.riderfans.com/forum/showthread.php?121827-Freeman&s=e98fa1ea6dc08e886b1678d35212494a',
650 'id': 'ln7x1qSThw4k',
652 'title': "PFT Live: New leader in the 'new-look' defense",
653 'description': 'md5:65a19b4bbfb3b0c0c5768bed1dfad74e',
658 'url': 'http://www.udn.com/news/story/7314/822787',
659 'md5': 'de06b4c90b042c128395a88f0384817e',
663 'title': '生物老師男變女 全校挺"做自己"',
664 'thumbnail': 're:^https?://.*\.jpg$',
669 def report_following_redirect(self, new_url):
670 """Report information extraction."""
671 self._downloader.to_screen('[redirect] Following redirect to %s' % new_url)
673 def _extract_rss(self, url, video_id, doc):
674 playlist_title = doc.find('./channel/title').text
675 playlist_desc_el = doc.find('./channel/description')
676 playlist_desc = None if playlist_desc_el is None else playlist_desc_el.text
679 for it in doc.findall('./channel/item'):
680 next_url = xpath_text(it, 'link', fatal=False)
682 enclosure_nodes = it.findall('./enclosure')
683 for e in enclosure_nodes:
684 next_url = e.attrib.get('url')
694 'title': it.find('title').text,
700 'title': playlist_title,
701 'description': playlist_desc,
705 def _extract_camtasia(self, url, video_id, webpage):
706 """ Returns None if no camtasia video can be found. """
708 camtasia_cfg = self._search_regex(
709 r'fo\.addVariable\(\s*"csConfigFile",\s*"([^"]+)"\s*\);',
710 webpage, 'camtasia configuration file', default=None)
711 if camtasia_cfg is None:
714 title = self._html_search_meta('DC.title', webpage, fatal=True)
716 camtasia_url = compat_urlparse.urljoin(url, camtasia_cfg)
717 camtasia_cfg = self._download_xml(
718 camtasia_url, video_id,
719 note='Downloading camtasia configuration',
720 errnote='Failed to download camtasia configuration')
721 fileset_node = camtasia_cfg.find('./playlist/array/fileset')
724 for n in fileset_node.getchildren():
725 url_n = n.find('./uri')
730 'id': os.path.splitext(url_n.text.rpartition('/')[2])[0],
731 'title': '%s - %s' % (title, n.tag),
732 'url': compat_urlparse.urljoin(url, url_n.text),
733 'duration': float_or_none(n.find('./duration').text),
742 def _real_extract(self, url):
743 if url.startswith('//'):
746 'url': self.http_scheme() + url,
749 parsed_url = compat_urlparse.urlparse(url)
750 if not parsed_url.scheme:
751 default_search = self._downloader.params.get('default_search')
752 if default_search is None:
753 default_search = 'fixup_error'
755 if default_search in ('auto', 'auto_warning', 'fixup_error'):
757 self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http')
758 return self.url_result('http://' + url)
759 elif default_search != 'fixup_error':
760 if default_search == 'auto_warning':
761 if re.match(r'^(?:url|URL)$', url):
762 raise ExtractorError(
763 'Invalid URL: %r . Call youtube-dl like this: youtube-dl -v "https://www.youtube.com/watch?v=BaW_jenozKc" ' % url,
766 self._downloader.report_warning(
767 'Falling back to youtube search for %s . Set --default-search "auto" to suppress this warning.' % url)
768 return self.url_result('ytsearch:' + url)
770 if default_search in ('error', 'fixup_error'):
771 raise ExtractorError(
772 '%r is not a valid URL. '
773 'Set --default-search "ytsearch" (or run youtube-dl "ytsearch:%s" ) to search YouTube'
774 % (url, url), expected=True)
776 if ':' not in default_search:
777 default_search += ':'
778 return self.url_result(default_search + url)
780 url, smuggled_data = unsmuggle_url(url)
782 is_intentional = smuggled_data and smuggled_data.get('to_generic')
783 if smuggled_data and 'force_videoid' in smuggled_data:
784 force_videoid = smuggled_data['force_videoid']
785 video_id = force_videoid
787 video_id = os.path.splitext(url.rstrip('/').split('/')[-1])[0]
789 self.to_screen('%s: Requesting header' % video_id)
791 head_req = HEADRequest(url)
792 head_response = self._request_webpage(
794 note=False, errnote='Could not send HEAD request to %s' % url,
797 if head_response is not False:
799 new_url = head_response.geturl()
801 self.report_following_redirect(new_url)
803 new_url = smuggle_url(
804 new_url, {'force_videoid': force_videoid})
805 return self.url_result(new_url)
808 if head_response is False:
809 full_response = self._request_webpage(url, video_id)
810 head_response = full_response
812 # Check for direct link to a video
813 content_type = head_response.headers.get('Content-Type', '')
814 m = re.match(r'^(?P<type>audio|video|application(?=/ogg$))/(?P<format_id>.+)$', content_type)
816 upload_date = unified_strdate(
817 head_response.headers.get('Last-Modified'))
820 'title': os.path.splitext(url_basename(url))[0],
823 'format_id': m.group('format_id'),
825 'vcodec': 'none' if m.group('type') == 'audio' else None
827 'upload_date': upload_date,
830 if not self._downloader.params.get('test', False) and not is_intentional:
831 self._downloader.report_warning('Falling back on generic information extractor.')
833 if not full_response:
834 full_response = self._request_webpage(url, video_id)
836 # Maybe it's a direct link to a video?
837 # Be careful not to download the whole thing!
838 first_bytes = full_response.read(512)
839 if not is_html(first_bytes):
840 self._downloader.report_warning(
841 'URL could be a direct video link, returning it as such.')
842 upload_date = unified_strdate(
843 head_response.headers.get('Last-Modified'))
846 'title': os.path.splitext(url_basename(url))[0],
849 'upload_date': upload_date,
852 webpage = self._webpage_read_content(
853 full_response, url, video_id, prefix=first_bytes)
855 self.report_extraction(video_id)
859 doc = parse_xml(webpage)
861 return self._extract_rss(url, video_id, doc)
862 except compat_xml_parse_error:
865 # Is it a Camtasia project?
866 camtasia_res = self._extract_camtasia(url, video_id, webpage)
867 if camtasia_res is not None:
870 # Sometimes embedded video player is hidden behind percent encoding
871 # (e.g. https://github.com/rg3/youtube-dl/issues/2448)
872 # Unescaping the whole page allows to handle those cases in a generic way
873 webpage = compat_urllib_parse.unquote(webpage)
875 # it's tempting to parse this further, but you would
876 # have to take into account all the variations like
877 # Video Title - Site Name
878 # Site Name | Video Title
879 # Video Title - Tagline | Site Name
880 # and so on and so forth; it's just not practical
881 video_title = self._html_search_regex(
882 r'(?s)<title>(.*?)</title>', webpage, 'video title',
885 # Try to detect age limit automatically
886 age_limit = self._rta_search(webpage)
887 # And then there are the jokers who advertise that they use RTA,
888 # but actually don't.
889 AGE_LIMIT_MARKERS = [
890 r'Proudly Labeled <a href="http://www.rtalabel.org/" title="Restricted to Adults">RTA</a>',
892 if any(re.search(marker, webpage) for marker in AGE_LIMIT_MARKERS):
895 # video uploader is domain name
896 video_uploader = self._search_regex(
897 r'^(?:https?://)?([^/]*)/.*', url, 'video uploader')
900 def _playlist_from_matches(matches, getter=None, ie=None):
902 self.url_result(self._proto_relative_url(getter(m) if getter else m), ie)
904 return self.playlist_result(
905 urlrs, playlist_id=video_id, playlist_title=video_title)
907 # Look for BrightCove:
908 bc_urls = BrightcoveIE._extract_brightcove_urls(webpage)
910 self.to_screen('Brightcove video detected.')
913 'url': smuggle_url(bc_url, {'Referer': url}),
914 'ie_key': 'Brightcove'
915 } for bc_url in bc_urls]
919 'title': video_title,
924 # Look for embedded rtl.nl player
925 matches = re.findall(
926 r'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="((?:https?:)?//(?:www\.)?rtl\.nl/system/videoplayer/[^"]+video_embed[^"]+)"',
929 return _playlist_from_matches(matches, ie='RtlNl')
931 # Look for embedded (iframe) Vimeo player
933 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//player\.vimeo\.com/video/.+?)\1', webpage)
935 player_url = unescapeHTML(mobj.group('url'))
936 surl = smuggle_url(player_url, {'Referer': url})
937 return self.url_result(surl)
938 # Look for embedded (swf embed) Vimeo player
940 r'<embed[^>]+?src="((?:https?:)?//(?:www\.)?vimeo\.com/moogaloop\.swf.+?)"', webpage)
942 return self.url_result(mobj.group(1))
944 # Look for embedded YouTube player
945 matches = re.findall(r'''(?x)
954 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
958 return _playlist_from_matches(
959 matches, lambda m: unescapeHTML(m[1]))
961 # Look for lazyYT YouTube embed
962 matches = re.findall(
963 r'class="lazyYT" data-youtube-id="([^"]+)"', webpage)
965 return _playlist_from_matches(matches, lambda m: unescapeHTML(m))
967 # Look for embedded Dailymotion player
968 matches = re.findall(
969 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/embed/video/.+?)\1', webpage)
971 return _playlist_from_matches(
972 matches, lambda m: unescapeHTML(m[1]))
974 # Look for embedded Dailymotion playlist player (#3822)
976 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.[a-z]{2,3}/widget/jukebox\?.+?)\1', webpage)
978 playlists = re.findall(
979 r'list\[\]=/playlist/([^/]+)/', unescapeHTML(m.group('url')))
981 return _playlist_from_matches(
982 playlists, lambda p: '//dailymotion.com/playlist/%s' % p)
984 # Look for embedded Wistia player
986 r'<(?:meta[^>]+?content|iframe[^>]+?src)=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage)
988 embed_url = self._proto_relative_url(
989 unescapeHTML(match.group('url')))
991 '_type': 'url_transparent',
994 'uploader': video_uploader,
995 'title': video_title,
999 match = re.search(r'(?:id=["\']wistia_|data-wistia-?id=["\']|Wistia\.embed\(["\'])(?P<id>[^"\']+)', webpage)
1002 '_type': 'url_transparent',
1003 'url': 'http://fast.wistia.net/embed/iframe/{0:}'.format(match.group('id')),
1005 'uploader': video_uploader,
1006 'title': video_title,
1007 'id': match.group('id')
1010 # Look for embedded blip.tv player
1011 mobj = re.search(r'<meta\s[^>]*https?://api\.blip\.tv/\w+/redirect/\w+/(\d+)', webpage)
1013 return self.url_result('http://blip.tv/a/a-' + mobj.group(1), 'BlipTV')
1014 mobj = re.search(r'<(?:iframe|embed|object)\s[^>]*(https?://(?:\w+\.)?blip\.tv/(?:play/|api\.swf#)[a-zA-Z0-9_]+)', webpage)
1016 return self.url_result(mobj.group(1), 'BlipTV')
1018 # Look for embedded condenast player
1019 matches = re.findall(
1020 r'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="(https?://player\.cnevids\.com/embed/[^"]+")',
1024 '_type': 'playlist',
1027 'ie_key': 'CondeNast',
1029 } for ma in matches],
1030 'title': video_title,
1034 # Look for Bandcamp pages with custom domain
1035 mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage)
1036 if mobj is not None:
1037 burl = unescapeHTML(mobj.group(1))
1038 # Don't set the extractor because it can be a track url or an album
1039 return self.url_result(burl)
1041 # Look for embedded Vevo player
1043 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:cache\.)?vevo\.com/.+?)\1', webpage)
1044 if mobj is not None:
1045 return self.url_result(mobj.group('url'))
1047 # Look for embedded Viddler player
1049 r'<(?:iframe[^>]+?src|param[^>]+?value)=(["\'])(?P<url>(?:https?:)?//(?:www\.)?viddler\.com/(?:embed|player)/.+?)\1',
1051 if mobj is not None:
1052 return self.url_result(mobj.group('url'))
1054 # Look for NYTimes player
1056 r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//graphics8\.nytimes\.com/bcvideo/[^/]+/iframe/embed\.html.+?)\1>',
1058 if mobj is not None:
1059 return self.url_result(mobj.group('url'))
1061 # Look for Libsyn player
1063 r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//html5-player\.libsyn\.com/embed/.+?)\1', webpage)
1064 if mobj is not None:
1065 return self.url_result(mobj.group('url'))
1067 # Look for Ooyala videos
1068 mobj = (re.search(r'player\.ooyala\.com/[^"?]+\?[^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or
1069 re.search(r'OO\.Player\.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage) or
1070 re.search(r'SBN\.VideoLinkset\.ooyala\([\'"](?P<ec>.{32})[\'"]\)', webpage))
1071 if mobj is not None:
1072 return OoyalaIE._build_url_result(mobj.group('ec'))
1074 # Look for multiple Ooyala embeds on SBN network websites
1075 mobj = re.search(r'SBN\.VideoLinkset\.entryGroup\((\[.*?\])', webpage)
1076 if mobj is not None:
1077 embeds = self._parse_json(mobj.group(1), video_id, fatal=False)
1079 return _playlist_from_matches(
1080 embeds, getter=lambda v: OoyalaIE._url_for_embed_code(v['provider_video_id']), ie='Ooyala')
1082 # Look for Aparat videos
1083 mobj = re.search(r'<iframe .*?src="(http://www\.aparat\.com/video/[^"]+)"', webpage)
1084 if mobj is not None:
1085 return self.url_result(mobj.group(1), 'Aparat')
1087 # Look for MPORA videos
1088 mobj = re.search(r'<iframe .*?src="(http://mpora\.(?:com|de)/videos/[^"]+)"', webpage)
1089 if mobj is not None:
1090 return self.url_result(mobj.group(1), 'Mpora')
1092 # Look for embedded NovaMov-based player
1094 r'''(?x)<(?:pagespeed_)?iframe[^>]+?src=(["\'])
1095 (?P<url>http://(?:(?:embed|www)\.)?
1097 nowvideo\.(?:ch|sx|eu|at|ag|co)|
1098 videoweed\.(?:es|com)|
1099 movshare\.(?:net|sx|ag)|
1100 divxstage\.(?:eu|net|ch|co|at|ag))
1101 /embed\.php.+?)\1''', webpage)
1102 if mobj is not None:
1103 return self.url_result(mobj.group('url'))
1105 # Look for embedded Facebook player
1107 r'<iframe[^>]+?src=(["\'])(?P<url>https://www\.facebook\.com/video/embed.+?)\1', webpage)
1108 if mobj is not None:
1109 return self.url_result(mobj.group('url'), 'Facebook')
1111 # Look for embedded VK player
1112 mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://vk\.com/video_ext\.php.+?)\1', webpage)
1113 if mobj is not None:
1114 return self.url_result(mobj.group('url'), 'VK')
1116 # Look for embedded ivi player
1117 mobj = re.search(r'<embed[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?ivi\.ru/video/player.+?)\1', webpage)
1118 if mobj is not None:
1119 return self.url_result(mobj.group('url'), 'Ivi')
1121 # Look for embedded Huffington Post player
1123 r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed\.live\.huffingtonpost\.com/.+?)\1', webpage)
1124 if mobj is not None:
1125 return self.url_result(mobj.group('url'), 'HuffPost')
1128 mobj = re.search(r'class=["\']embedly-card["\'][^>]href=["\'](?P<url>[^"\']+)', webpage)
1129 if mobj is not None:
1130 return self.url_result(mobj.group('url'))
1131 mobj = re.search(r'class=["\']embedly-embed["\'][^>]src=["\'][^"\']*url=(?P<url>[^&]+)', webpage)
1132 if mobj is not None:
1133 return self.url_result(compat_urllib_parse.unquote(mobj.group('url')))
1135 # Look for funnyordie embed
1136 matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage)
1138 return _playlist_from_matches(
1139 matches, getter=unescapeHTML, ie='FunnyOrDie')
1141 # Look for BBC iPlayer embed
1142 matches = re.findall(r'setPlaylist\("(https?://www\.bbc\.co\.uk/iplayer/[^/]+/[\da-z]{8})"\)', webpage)
1144 return _playlist_from_matches(matches, ie='BBCCoUk')
1146 # Look for embedded RUTV player
1147 rutv_url = RUTVIE._extract_url(webpage)
1149 return self.url_result(rutv_url, 'RUTV')
1151 # Look for embedded TED player
1153 r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed(?:-ssl)?\.ted\.com/.+?)\1', webpage)
1154 if mobj is not None:
1155 return self.url_result(mobj.group('url'), 'TED')
1157 # Look for embedded Ustream videos
1159 r'<iframe[^>]+?src=(["\'])(?P<url>http://www\.ustream\.tv/embed/.+?)\1', webpage)
1160 if mobj is not None:
1161 return self.url_result(mobj.group('url'), 'Ustream')
1163 # Look for embedded arte.tv player
1165 r'<script [^>]*?src="(?P<url>http://www\.arte\.tv/playerv2/embed[^"]+)"',
1167 if mobj is not None:
1168 return self.url_result(mobj.group('url'), 'ArteTVEmbed')
1170 # Look for embedded smotri.com player
1171 smotri_url = SmotriIE._extract_url(webpage)
1173 return self.url_result(smotri_url, 'Smotri')
1175 # Look for embeded soundcloud player
1177 r'<iframe\s+(?:[a-zA-Z0-9_-]+="[^"]+"\s+)*src="(?P<url>https?://(?:w\.)?soundcloud\.com/player[^"]+)"',
1179 if mobj is not None:
1180 url = unescapeHTML(mobj.group('url'))
1181 return self.url_result(url)
1183 # Look for embedded vulture.com player
1185 r'<iframe src="(?P<url>https?://video\.vulture\.com/[^"]+)"',
1187 if mobj is not None:
1188 url = unescapeHTML(mobj.group('url'))
1189 return self.url_result(url, ie='Vulture')
1191 # Look for embedded mtvservices player
1193 r'<iframe src="(?P<url>https?://media\.mtvnservices\.com/embed/[^"]+)"',
1195 if mobj is not None:
1196 url = unescapeHTML(mobj.group('url'))
1197 return self.url_result(url, ie='MTVServicesEmbedded')
1199 # Look for embedded yahoo player
1201 r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:screen|movies)\.yahoo\.com/.+?\.html\?format=embed)\1',
1203 if mobj is not None:
1204 return self.url_result(mobj.group('url'), 'Yahoo')
1206 # Look for embedded sbs.com.au player
1210 <meta\s+property="og:video"\s+content=|
1213 (["\'])(?P<url>https?://(?:www\.)?sbs\.com\.au/ondemand/video/.+?)\1''',
1215 if mobj is not None:
1216 return self.url_result(mobj.group('url'), 'SBS')
1218 # Look for embedded Cinchcast player
1220 r'<iframe[^>]+?src=(["\'])(?P<url>https?://player\.cinchcast\.com/.+?)\1',
1222 if mobj is not None:
1223 return self.url_result(mobj.group('url'), 'Cinchcast')
1226 r'<iframe[^>]+?src=(["\'])(?P<url>https?://m(?:lb)?\.mlb\.com/shared/video/embed/embed\.html\?.+?)\1',
1228 if mobj is not None:
1229 return self.url_result(mobj.group('url'), 'MLB')
1232 r'<iframe[^>]+?src=(["\'])(?P<url>%s)\1' % CondeNastIE.EMBED_URL,
1234 if mobj is not None:
1235 return self.url_result(self._proto_relative_url(mobj.group('url'), scheme='http:'), 'CondeNast')
1238 r'<iframe[^>]+src="(?P<url>https?://new\.livestream\.com/[^"]+/player[^"]+)"',
1240 if mobj is not None:
1241 return self.url_result(mobj.group('url'), 'Livestream')
1243 # Look for Zapiks embed
1245 r'<iframe[^>]+src="(?P<url>https?://(?:www\.)?zapiks\.fr/index\.php\?.+?)"', webpage)
1246 if mobj is not None:
1247 return self.url_result(mobj.group('url'), 'Zapiks')
1249 # Look for Kaltura embeds
1251 r"(?s)kWidget\.(?:thumb)?[Ee]mbed\(\{.*?'wid'\s*:\s*'_?(?P<partner_id>[^']+)',.*?'entry_id'\s*:\s*'(?P<id>[^']+)',", webpage)
1252 if mobj is not None:
1253 return self.url_result('kaltura:%(partner_id)s:%(id)s' % mobj.groupdict(), 'Kaltura')
1255 # Look for Eagle.Platform embeds
1257 r'<iframe[^>]+src="(?P<url>https?://.+?\.media\.eagleplatform\.com/index/player\?.+?)"', webpage)
1258 if mobj is not None:
1259 return self.url_result(mobj.group('url'), 'EaglePlatform')
1261 # Look for ClipYou (uses Eagle.Platform) embeds
1263 r'<iframe[^>]+src="https?://(?P<host>media\.clipyou\.ru)/index/player\?.*\brecord_id=(?P<id>\d+).*"', webpage)
1264 if mobj is not None:
1265 return self.url_result('eagleplatform:%(host)s:%(id)s' % mobj.groupdict(), 'EaglePlatform')
1267 # Look for Pladform embeds
1269 r'<iframe[^>]+src="(?P<url>https?://out\.pladform\.ru/player\?.+?)"', webpage)
1270 if mobj is not None:
1271 return self.url_result(mobj.group('url'), 'Pladform')
1273 # Look for 5min embeds
1275 r'<meta[^>]+property="og:video"[^>]+content="https?://embed\.5min\.com/(?P<id>[0-9]+)/?', webpage)
1276 if mobj is not None:
1277 return self.url_result('5min:%s' % mobj.group('id'), 'FiveMin')
1279 # Look for NBC Sports VPlayer embeds
1280 nbc_sports_url = NBCSportsVPlayerIE._extract_url(webpage)
1282 return self.url_result(nbc_sports_url, 'NBCSportsVPlayer')
1284 # Look for UDN embeds
1286 r'<iframe[^>]+src="(?P<url>%s)"' % UDNEmbedIE._VALID_URL, webpage)
1287 if mobj is not None:
1288 return self.url_result(
1289 url_infer_protocol(url, mobj.group('url')), 'UDNEmbed')
1291 def check_video(vurl):
1292 if YoutubeIE.suitable(vurl):
1294 vpath = compat_urlparse.urlparse(vurl).path
1295 vext = determine_ext(vpath)
1296 return '.' in vpath and vext not in ('swf', 'png', 'jpg', 'srt', 'sbv', 'sub', 'vtt', 'ttml')
1298 def filter_video(urls):
1299 return list(filter(check_video, urls))
1301 # Start with something easy: JW Player in SWFObject
1302 found = filter_video(re.findall(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage))
1304 # Look for gorilla-vid style embedding
1305 found = filter_video(re.findall(r'''(?sx)
1309 jwplayer\s*\(\s*["'][^'"]+["']\s*\)\s*\.setup
1312 ['"]?file['"]?\s*:\s*["\'](.*?)["\']''', webpage))
1314 # Broaden the search a little bit
1315 found = filter_video(re.findall(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage))
1317 # Broaden the findall a little bit: JWPlayer JS loader
1318 found = filter_video(re.findall(
1319 r'[^A-Za-z0-9]?file["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage))
1322 found = filter_video(re.findall(r'''(?xs)
1323 flowplayer\("[^"]+",\s*
1325 \s*\{[^}]+? ["']?clip["']?\s*:\s*\{\s*
1326 ["']?url["']?\s*:\s*["']([^"']+)["']
1331 r"cinerama\.embedPlayer\(\s*\'[^']+\',\s*'([^']+)'", webpage)
1333 # Try to find twitter cards info
1334 found = filter_video(re.findall(
1335 r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage))
1337 # We look for Open Graph info:
1338 # We have to match any number spaces between elements, some sites try to align them (eg.: statigr.am)
1339 m_video_type = re.findall(r'<meta.*?property="og:video:type".*?content="video/(.*?)"', webpage)
1340 # We only look in og:video if the MIME type is a video, don't try if it's a Flash player:
1341 if m_video_type is not None:
1342 found = filter_video(re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage))
1345 found = re.findall(r'(?s)<video[^<]*(?:>.*?<source[^>]*)?\s+src=["\'](.*?)["\']', webpage)
1347 REDIRECT_REGEX = r'[0-9]{,2};\s*(?:URL|url)=\'?([^\'"]+)'
1349 r'(?i)<meta\s+(?=(?:[a-z-]+="[^"]+"\s+)*http-equiv="refresh")'
1350 r'(?:[a-z-]+="[^"]+"\s+)*?content="%s' % REDIRECT_REGEX,
1353 # Look also in Refresh HTTP header
1354 refresh_header = head_response.headers.get('Refresh')
1356 found = re.search(REDIRECT_REGEX, refresh_header)
1358 new_url = found.group(1)
1359 self.report_following_redirect(new_url)
1365 raise UnsupportedError(url)
1368 for video_url in found:
1369 video_url = compat_urlparse.urljoin(url, video_url)
1370 video_id = compat_urllib_parse.unquote(os.path.basename(video_url))
1372 # Sometimes, jwplayer extraction will result in a YouTube URL
1373 if YoutubeIE.suitable(video_url):
1374 entries.append(self.url_result(video_url, 'Youtube'))
1377 # here's a fun little line of code for you:
1378 video_id = os.path.splitext(video_id)[0]
1383 'uploader': video_uploader,
1384 'title': video_title,
1385 'age_limit': age_limit,
1388 if len(entries) == 1:
1391 for num, e in enumerate(entries, start=1):
1392 # 'url' results don't have a title
1393 if e.get('title') is not None:
1394 e['title'] = '%s (%d)' % (e['title'], num)
1396 '_type': 'playlist',