3 from __future__ import unicode_literals
8 from .common import InfoExtractor
9 from .youtube import YoutubeIE
10 from ..compat import (
12 compat_urllib_parse_unquote,
13 compat_urllib_request,
15 compat_xml_parse_error,
33 from .brightcove import BrightcoveIE
34 from .nbc import NBCSportsVPlayerIE
35 from .ooyala import OoyalaIE
36 from .rutv import RUTVIE
37 from .sportbox import SportBoxEmbedIE
38 from .smotri import SmotriIE
39 from .condenast import CondeNastIE
40 from .udn import UDNEmbedIE
41 from .senateisvp import SenateISVPIE
42 from .bliptv import BlipTVIE
43 from .svt import SVTIE
46 class GenericIE(InfoExtractor):
47 IE_DESC = 'Generic downloader that works on some sites'
51 # Direct link to a video
53 'url': 'http://media.w3.org/2010/05/sintel/trailer.mp4',
54 'md5': '67d406c2bcb6af27fa886f31aa934bbe',
59 'upload_date': '20100513',
62 # Direct link to media delivered compressed (until Accept-Encoding is *)
64 'url': 'http://calimero.tk/muzik/FictionJunction-Parallel_Hearts.flac',
65 'md5': '128c42e68b13950268b648275386fc74',
67 'id': 'FictionJunction-Parallel_Hearts',
69 'title': 'FictionJunction-Parallel_Hearts',
70 'upload_date': '20140522',
72 'expected_warnings': [
73 'URL could be a direct video link, returning it as such.'
76 # Direct download with broken HEAD
78 'url': 'http://ai-radio.org:8000/radio.opus',
85 'skip_download': True, # infinite live stream
87 'expected_warnings': [
88 r'501.*Not Implemented'
91 # Direct link with incorrect MIME type
93 'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
94 'md5': '4ccbebe5f36706d85221f204d7eb5913',
96 'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
97 'id': '5_Lennart_Poettering_-_Systemd',
99 'title': '5_Lennart_Poettering_-_Systemd',
100 'upload_date': '20141120',
102 'expected_warnings': [
103 'URL could be a direct video link, returning it as such.'
108 'url': 'http://phihag.de/2014/youtube-dl/rss2.xml',
110 'id': 'http://phihag.de/2014/youtube-dl/rss2.xml',
111 'title': 'Zero Punctuation',
112 'description': 're:.*groundbreaking video review series.*'
114 'playlist_mincount': 11,
116 # RSS feed with enclosure
118 'url': 'http://podcastfeeds.nbcnews.com/audio/podcast/MSNBC-MADDOW-NETCAST-M4V.xml',
120 'id': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
122 'upload_date': '20150228',
123 'title': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
128 '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',
132 'upload_date': '20130224',
133 'uploader_id': 'TheVerge',
134 'description': 're:^Chris Ziegler takes a look at the\.*',
135 'uploader': 'The Verge',
136 'title': 'First Firefox OS phones side-by-side',
139 'skip_download': False,
143 'url': 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
144 'md5': '85b90ccc9d73b4acd9138d3af4c27f89',
146 'id': '13601338388002',
148 'uploader': 'www.hodiho.fr',
149 'title': 'R\u00e9gis plante sa Jeep',
152 # bandcamp page with custom domain
154 'add_ie': ['Bandcamp'],
155 'url': 'http://bronyrock.com/track/the-pony-mash',
159 'title': 'The Pony Mash',
160 'uploader': 'M_Pallante',
162 'skip': 'There is a limit of 200 free downloads / month for the test song',
164 # embedded brightcove video
165 # it also tests brightcove videos that need to set the 'Referer' in the
168 'add_ie': ['Brightcove'],
169 'url': 'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/',
171 'id': '2765128793001',
173 'title': 'Le cours de bourse : l’analyse technique',
174 'description': 'md5:7e9ad046e968cb2d1114004aba466fd9',
175 'uploader': 'BFM BUSINESS',
178 'skip_download': True,
182 # https://github.com/rg3/youtube-dl/issues/2253
183 'url': 'http://bcove.me/i6nfkrc3',
184 'md5': '0ba9446db037002366bab3b3eb30c88c',
186 'id': '3101154703001',
188 'title': 'Still no power',
189 'uploader': 'thestar.com',
190 '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.',
192 'add_ie': ['Brightcove'],
195 'url': 'http://www.championat.com/video/football/v/87/87499.html',
196 'md5': 'fb973ecf6e4a78a67453647444222983',
198 'id': '3414141473001',
200 'title': 'Видео. Удаление Дзагоева (ЦСКА)',
201 'description': 'Онлайн-трансляция матча ЦСКА - "Волга"',
202 'uploader': 'Championat',
206 # https://github.com/rg3/youtube-dl/issues/3541
207 'add_ie': ['Brightcove'],
208 'url': 'http://www.kijk.nl/sbs6/leermijvrouwenkennen/videos/jqMiXKAYan2S/aflevering-1',
210 'id': '3866516442001',
212 'title': 'Leer mij vrouwen kennen: Aflevering 1',
213 'description': 'Leer mij vrouwen kennen: Aflevering 1',
214 'uploader': 'SBS Broadcasting',
216 'skip': 'Restricted to Netherlands',
218 'skip_download': True, # m3u8 download
223 'url': 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219',
224 'md5': '166dd577b433b4d4ebfee10b0824d8ff',
226 'id': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ',
228 'title': '2cc213299525360.mov', # that's what we get
230 'add_ie': ['Ooyala'],
232 # multiple ooyala embeds on SBN network websites
234 'url': 'http://www.sbnation.com/college-football-recruiting/2015/2/3/7970291/national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
236 'id': 'national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
237 'title': '25 lies you will tell yourself on National Signing Day - SBNation.com',
239 'playlist_mincount': 3,
241 'skip_download': True,
243 'add_ie': ['Ooyala'],
247 'url': 'http://www.tested.com/science/weird/460206-tested-grinding-coffee-2000-frames-second/',
251 'title': 'Tested: Grinding Coffee at 2000 Frames Per Second',
252 'upload_date': '20140225',
253 'description': 'md5:06a40fbf30b220468f1e0957c0f558ff',
254 'uploader': 'Tested',
255 'uploader_id': 'testedcom',
257 # No need to test YoutubeIE here
259 'skip_download': True,
264 'url': 'http://www.theguardian.com/world/2014/mar/11/obama-zach-galifianakis-between-two-ferns',
268 'title': 'Between Two Ferns with Zach Galifianakis: President Barack Obama',
269 'description': 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.',
274 'url': 'http://www.bbc.co.uk/blogs/adamcurtis/posts/BUGGER',
276 'title': 'BBC - Blogs - Adam Curtis - BUGGER',
278 'playlist_mincount': 18,
282 'url': 'http://www.rg.ru/2014/03/15/reg-dfo/anklav-anons.html',
286 'title': 'Охотское море стало целиком российским',
287 'description': 'md5:5ed62483b14663e2a95ebbe115eb8f43',
291 'skip_download': True,
296 'url': 'http://www.vestifinance.ru/articles/25753',
299 'title': 'Вести Экономика ― Прямые трансляции с Форума-выставки "Госзаказ-2013"',
304 'title': 'Госзаказ. День 3',
310 'title': 'Госзаказ. День 2',
316 'title': 'Госзаказ. День 1',
322 'skip_download': True,
327 'url': 'http://en.support.wordpress.com/videos/ted-talks/',
328 'md5': '65fdff94098e4a607385a60c5177c638',
332 'title': 'Hidden miracles of the natural world',
333 'uploader': 'Louie Schwartzberg',
334 'description': 'md5:8145d19d320ff3e52f28401f4c4283b9',
337 # Embeded Ustream video
339 'url': 'http://www.american.edu/spa/pti/nsa-privacy-janus-2014.cfm',
340 'md5': '27b99cdb639c9b12a79bca876a073417',
344 'uploader': 'AU SPA: The NSA and Privacy',
345 'title': 'NSA and Privacy Forum Debate featuring General Hayden and Barton Gellman'
348 # nowvideo embed hidden behind percent encoding
350 'url': 'http://www.waoanime.tv/the-super-dimension-fortress-macross-episode-1/',
351 'md5': '2baf4ddd70f697d94b1c18cf796d5107',
353 'id': '06e53103ca9aa',
355 'title': 'Macross Episode 001 Watch Macross Episode 001 onl',
356 'description': 'No description',
361 'url': 'http://www.tv-replay.fr/redirection/20-03-14/x-enius-arte-10753389.html',
362 'md5': '7653032cbb25bf6c80d80f217055fa43',
364 'id': '048195-004_PLUS7-F',
367 'description': 'md5:d5fdf32ef6613cdbfd516ae658abf168',
368 'upload_date': '20140320',
371 'skip_download': 'Requires rtmpdump'
376 'url': 'http://www.wired.com/2014/04/honda-asimo/',
377 'md5': 'ba0dfe966fa007657bd1443ee672db0f',
379 'id': '53501be369702d3275860000',
381 'title': 'Honda’s New Asimo Robot Is More Human Than Ever',
386 'url': 'http://www.spi0n.com/zap-spi0n-com-n216/',
387 'md5': '441aeeb82eb72c422c7f14ec533999cd',
389 'id': 'k2mm4bCdJ6CQ2i7c8o2',
391 'title': 'Le Zap de Spi0n n°216 - Zapping du Web',
394 'add_ie': ['Dailymotion'],
398 'url': 'http://www.badzine.de/ansicht/datum/2014/06/09/so-funktioniert-die-neue-englische-badminton-liga.html',
402 'title': 'The NBL Auction 2014',
403 'uploader': 'BADMINTON England',
404 'uploader_id': 'BADMINTONEvents',
405 'upload_date': '20140603',
406 'description': 'md5:9ef128a69f1e262a700ed83edb163a73',
408 'add_ie': ['Youtube'],
410 'skip_download': True,
415 'url': 'http://www.gametrailers.com/news-post/76093/north-america-europe-is-getting-that-mario-kart-8-mercedes-dlc-too',
416 'md5': '35727f82f58c76d996fc188f9755b0d5',
418 'id': '0306a69b-8adf-4fb5-aace-75f8e8cbfca9',
421 'description': 'Mario\'s life in the fast lane has never looked so good.',
424 # YouTube embed via <data-embed-url="">
426 'url': 'https://play.google.com/store/apps/details?id=com.gameloft.android.ANMP.GloftA8HM',
430 'title': 'Asphalt 8: Airborne - Update - Welcome to Dubai!',
431 'uploader': 'Gameloft',
432 'uploader_id': 'gameloft',
433 'upload_date': '20140828',
434 'description': 'md5:c80da9ed3d83ae6d1876c834de03e1c4',
437 'skip_download': True,
442 'url': 'http://www.ll.mit.edu/workshops/education/videocourses/antennas/lecture1/video/',
444 'md5': '0c5e352edabf715d762b0ad4e6d9ee67',
446 'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
447 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - video1',
452 'md5': '10e4bb3aaca9fd630e273ff92d9f3c63',
454 'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final_PIP',
455 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - pip',
461 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
466 'url': 'http://www.handjobhub.com/video/busty-blonde-siri-tit-fuck-while-wank-6313.html',
467 'md5': '9d65602bf31c6e20014319c7d07fba27',
469 'id': '5123ea6d5e5a7',
472 'uploader': 'www.handjobhub.com',
473 'title': 'Busty Blonde Siri Tit Fuck While Wank at HandjobHub.com',
476 # Multiple brightcove videos
477 # https://github.com/rg3/youtube-dl/issues/2283
479 'url': 'http://www.newyorker.com/online/blogs/newsdesk/2014/01/always-never-nuclear-command-and-control.html',
481 'id': 'always-never',
482 'title': 'Always / Never - The New Yorker',
486 'extract_flat': False,
487 'skip_download': True,
492 'url': 'http://umpire-empire.com/index.php/topic/58125-laz-decides-no-thats-low/',
493 'md5': '96f09a37e44da40dd083e12d9a683327',
497 'title': 'Ump changes call to ball',
498 'description': 'md5:71c11215384298a172a6dcb4c2e20685',
500 'timestamp': 1401537900,
501 'upload_date': '20140531',
502 'thumbnail': 're:^https?://.*\.jpg$',
507 'url': 'http://education-portal.com/academy/lesson/north-american-exploration-failed-colonies-of-spain-france-england.html#lesson',
508 'md5': '8788b683c777a5cf25621eaf286d0c23',
512 'title': 'md5:51364a8d3d009997ba99656004b5e20d',
514 'filesize': 182808282,
515 'uploader': 'education-portal.com',
519 'url': 'http://thoughtworks.wistia.com/medias/uxjb0lwrcz',
520 'md5': 'baf49c2baa8a7de5f3fc145a8506dcd4',
524 'title': 'Conversation about Hexagonal Rails Part 1 - ThoughtWorks',
526 'uploader': 'thoughtworks.wistia.com',
531 'url': 'http://nakedsecurity.sophos.com/2014/10/29/sscc-171-are-you-sure-that-1234-is-a-bad-password-podcast/',
535 'description': 'md5:ff867d6b555488ad3c52572bb33d432c',
536 'uploader': 'Sophos Security',
537 'title': 'Chet Chat 171 - Oct 29, 2014',
538 'upload_date': '20141029',
543 'url': 'http://www.esa.int/Our_Activities/Space_Science/Rosetta/Philae_comet_touch-down_webcast',
547 'upload_date': '20141112',
548 'title': 'Rosetta #CometLanding webcast HL 10',
553 'url': 'http://discourse.ubuntu.com/t/unity-8-desktop-mode-windows-on-mir/1986',
556 'title': 'Unity 8 desktop-mode windows on Mir! - Ubuntu Discourse',
558 'playlist_mincount': 2,
562 'url': 'http://undergroundwellness.com/podcasts/306-5-steps-to-permanent-gut-healing/',
566 'upload_date': '20141126',
567 'title': 'Jack Tips: 5 Steps to Permanent Gut Healing',
572 'url': 'http://www.abc.net.au/7.30/content/2015/s4164797.htm',
574 'id': '730m_DandD_1901_512k',
576 'uploader': 'www.abc.net.au',
577 'title': 'Game of Thrones with dice - Dungeons and Dragons fantasy role-playing game gets new life - 19/01/2015',
580 # embedded viddler video
582 'url': 'http://deadspin.com/i-cant-stop-watching-john-wall-chop-the-nuggets-with-th-1681801597',
586 'uploader': 'deadspin',
587 'title': 'WALL-TO-GORTAT',
588 'timestamp': 1422285291,
589 'upload_date': '20150126',
591 'add_ie': ['Viddler'],
595 'url': 'http://thedailyshow.cc.com/podcast/episodetwelve',
599 'title': "The Daily Show Podcast without Jon Stewart - Episode 12: Bassem Youssef: Egypt's Jon Stewart",
600 'description': 'md5:601cb790edd05908957dae8aaa866465',
601 'upload_date': '20150220',
606 'url': 'http://media.nationalarchives.gov.uk/index.php/webinar-using-discovery-national-archives-online-catalogue/',
610 'upload_date': '20150212',
611 'uploader': 'The National Archives UK',
612 'description': 'md5:a236581cd2449dd2df4f93412f3f01c6',
613 'uploader_id': 'NationalArchives08',
614 'title': 'Webinar: Using Discovery, The National Archives’ online catalogue',
619 'url': 'http://www.rtlnieuws.nl/nieuws/buitenland/aanslagen-kopenhagen',
620 'playlist_mincount': 5,
622 'id': 'aanslagen-kopenhagen',
623 'title': 'Aanslagen Kopenhagen | RTL Nieuws',
628 'url': 'http://www.skipass.com/news/116090-bon-appetit-s5ep3-baqueira-mi-cor.html',
632 'title': 'EP3S5 - Bon Appétit - Baqueira Mi Corazon !',
637 'url': 'http://www.monumentalnetwork.com/videos/john-carlson-postgame-2-25-15',
641 'upload_date': '20150226',
642 'uploader_id': 'MonumentalSports-Kaltura@perfectsensedigital.com',
644 'title': 'John Carlson Postgame 2/25/15',
647 # Eagle.Platform embed (generic URL)
649 'url': 'http://lenta.ru/news/2015/03/06/navalny/',
653 'title': 'Навальный вышел на свободу',
654 'description': 'md5:d97861ac9ae77377f3f20eaf9d04b4f5',
655 'thumbnail': 're:^https?://.*\.jpg$',
661 # ClipYou (Eagle.Platform) embed (custom URL)
663 'url': 'http://muz-tv.ru/play/7129/',
667 'title': "'O Sole Mio",
668 'thumbnail': 're:^https?://.*\.jpg$',
675 'url': 'http://muz-tv.ru/kinozal/view/7400/',
679 'title': 'Тайны перевала Дятлова • 1 серия 2 часть',
680 'description': 'Документальный сериал-расследование одной из самых жутких тайн ХХ века',
681 'thumbnail': 're:^https?://.*\.jpg$',
688 'url': 'http://www.cinemablend.com/new/First-Joe-Dirt-2-Trailer-Teaser-Stupid-Greatness-70874.html',
692 'title': 'Joe Dirt 2 Beautiful Loser Teaser Trailer',
693 'thumbnail': 're:^https?://.*\.png$',
699 'url': 'http://techcrunch.com/video/facebook-creates-on-this-day-crunch-report/518726732/',
700 'md5': '4c6f127a30736b59b3e2c19234ee2bf7',
704 'title': 'Facebook Creates "On This Day" | Crunch Report',
709 'url': 'http://www.svt.se/sport/ishockey/jagr-tacklar-giroux-under-intervjun',
713 'title': 'Här trycker Jagr till Giroux (under SVT-intervjun)',
718 # Crooks and Liars embed
720 'url': 'http://crooksandliars.com/2015/04/fox-friends-says-protecting-atheists',
724 'title': "Fox & Friends Says Protecting Atheists From Discrimination Is Anti-Christian!",
725 'description': 'md5:e1a46ad1650e3a5ec7196d432799127f',
726 'timestamp': 1428207000,
727 'upload_date': '20150405',
728 'uploader': 'Heather',
731 # Crooks and Liars external embed
733 'url': 'http://theothermccain.com/2010/02/02/video-proves-that-bill-kristol-has-been-watching-glenn-beck/comment-page-1/',
735 'id': 'MTE3MjUtMzQ2MzA',
737 'title': 'md5:5e3662a81a4014d24c250d76d41a08d5',
738 'description': 'md5:9b8e9542d6c3c5de42d6451b7d780cec',
739 'timestamp': 1265032391,
740 'upload_date': '20100201',
741 'uploader': 'Heather',
744 # NBC Sports vplayer embed
746 'url': 'http://www.riderfans.com/forum/showthread.php?121827-Freeman&s=e98fa1ea6dc08e886b1678d35212494a',
748 'id': 'ln7x1qSThw4k',
750 'title': "PFT Live: New leader in the 'new-look' defense",
751 'description': 'md5:65a19b4bbfb3b0c0c5768bed1dfad74e',
756 'url': 'http://www.udn.com/news/story/7314/822787',
757 'md5': 'fd2060e988c326991037b9aff9df21a6',
761 'title': '中一中男師變性 全校師生力挺',
762 'thumbnail': 're:^https?://.*\.jpg$',
767 'url': 'http://www.businessinsider.com/excel-index-match-vlookup-video-how-to-2015-2?IR=T',
769 'id': '50YnY4czr4ms1vJ7yz3xzq0excz_pUMs',
771 'description': 'VIDEO: Index/Match versus VLOOKUP.',
772 'title': 'This is what separates the Excel masters from the wannabes',
776 'skip_download': True,
779 # Contains a SMIL manifest
781 'url': 'http://www.telewebion.com/fa/1263668/%D9%82%D8%B1%D8%B9%D9%87%E2%80%8C%DA%A9%D8%B4%DB%8C-%D9%84%DB%8C%DA%AF-%D9%82%D9%87%D8%B1%D9%85%D8%A7%D9%86%D8%A7%D9%86-%D8%A7%D8%B1%D9%88%D9%BE%D8%A7/%2B-%D9%81%D9%88%D8%AA%D8%A8%D8%A7%D9%84.html',
785 'title': '+ Football: Lottery Champions League Europe',
786 'uploader': 'www.telewebion.com',
790 'skip_download': True,
793 # Brightcove URL in single quotes
795 'url': 'http://www.sportsnet.ca/baseball/mlb/sn-presents-russell-martin-world-citizen/',
796 'md5': '4ae374f1f8b91c889c4b9203c8c752af',
798 'id': '4255764656001',
800 'title': 'SN Presents: Russell Martin, World Citizen',
801 'description': 'To understand why he was the Toronto Blue Jays’ top off-season priority is to appreciate his background and upbringing in Montreal, where he first developed his baseball skills. Written and narrated by Stephen Brunt.',
802 'uploader': 'Rogers Sportsnet',
807 def report_following_redirect(self, new_url):
808 """Report information extraction."""
809 self._downloader.to_screen('[redirect] Following redirect to %s' % new_url)
811 def _extract_rss(self, url, video_id, doc):
812 playlist_title = doc.find('./channel/title').text
813 playlist_desc_el = doc.find('./channel/description')
814 playlist_desc = None if playlist_desc_el is None else playlist_desc_el.text
817 for it in doc.findall('./channel/item'):
818 next_url = xpath_text(it, 'link', fatal=False)
820 enclosure_nodes = it.findall('./enclosure')
821 for e in enclosure_nodes:
822 next_url = e.attrib.get('url')
832 'title': it.find('title').text,
838 'title': playlist_title,
839 'description': playlist_desc,
843 def _extract_camtasia(self, url, video_id, webpage):
844 """ Returns None if no camtasia video can be found. """
846 camtasia_cfg = self._search_regex(
847 r'fo\.addVariable\(\s*"csConfigFile",\s*"([^"]+)"\s*\);',
848 webpage, 'camtasia configuration file', default=None)
849 if camtasia_cfg is None:
852 title = self._html_search_meta('DC.title', webpage, fatal=True)
854 camtasia_url = compat_urlparse.urljoin(url, camtasia_cfg)
855 camtasia_cfg = self._download_xml(
856 camtasia_url, video_id,
857 note='Downloading camtasia configuration',
858 errnote='Failed to download camtasia configuration')
859 fileset_node = camtasia_cfg.find('./playlist/array/fileset')
862 for n in fileset_node.getchildren():
863 url_n = n.find('./uri')
868 'id': os.path.splitext(url_n.text.rpartition('/')[2])[0],
869 'title': '%s - %s' % (title, n.tag),
870 'url': compat_urlparse.urljoin(url, url_n.text),
871 'duration': float_or_none(n.find('./duration').text),
880 def _real_extract(self, url):
881 if url.startswith('//'):
884 'url': self.http_scheme() + url,
887 parsed_url = compat_urlparse.urlparse(url)
888 if not parsed_url.scheme:
889 default_search = self._downloader.params.get('default_search')
890 if default_search is None:
891 default_search = 'fixup_error'
893 if default_search in ('auto', 'auto_warning', 'fixup_error'):
895 self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http')
896 return self.url_result('http://' + url)
897 elif default_search != 'fixup_error':
898 if default_search == 'auto_warning':
899 if re.match(r'^(?:url|URL)$', url):
900 raise ExtractorError(
901 'Invalid URL: %r . Call youtube-dl like this: youtube-dl -v "https://www.youtube.com/watch?v=BaW_jenozKc" ' % url,
904 self._downloader.report_warning(
905 'Falling back to youtube search for %s . Set --default-search "auto" to suppress this warning.' % url)
906 return self.url_result('ytsearch:' + url)
908 if default_search in ('error', 'fixup_error'):
909 raise ExtractorError(
910 '%r is not a valid URL. '
911 'Set --default-search "ytsearch" (or run youtube-dl "ytsearch:%s" ) to search YouTube'
912 % (url, url), expected=True)
914 if ':' not in default_search:
915 default_search += ':'
916 return self.url_result(default_search + url)
918 url, smuggled_data = unsmuggle_url(url)
920 is_intentional = smuggled_data and smuggled_data.get('to_generic')
921 if smuggled_data and 'force_videoid' in smuggled_data:
922 force_videoid = smuggled_data['force_videoid']
923 video_id = force_videoid
925 video_id = compat_urllib_parse_unquote(os.path.splitext(url.rstrip('/').split('/')[-1])[0])
927 self.to_screen('%s: Requesting header' % video_id)
929 head_req = HEADRequest(url)
930 head_response = self._request_webpage(
932 note=False, errnote='Could not send HEAD request to %s' % url,
935 if head_response is not False:
937 new_url = head_response.geturl()
939 self.report_following_redirect(new_url)
941 new_url = smuggle_url(
942 new_url, {'force_videoid': force_videoid})
943 return self.url_result(new_url)
946 if head_response is False:
947 request = compat_urllib_request.Request(url)
948 request.add_header('Accept-Encoding', '*')
949 full_response = self._request_webpage(request, video_id)
950 head_response = full_response
952 # Check for direct link to a video
953 content_type = head_response.headers.get('Content-Type', '')
954 m = re.match(r'^(?P<type>audio|video|application(?=/ogg$))/(?P<format_id>.+)$', content_type)
956 upload_date = unified_strdate(
957 head_response.headers.get('Last-Modified'))
960 'title': compat_urllib_parse_unquote(os.path.splitext(url_basename(url))[0]),
963 'format_id': m.group('format_id'),
965 'vcodec': 'none' if m.group('type') == 'audio' else None
967 'upload_date': upload_date,
970 if not self._downloader.params.get('test', False) and not is_intentional:
971 self._downloader.report_warning('Falling back on generic information extractor.')
973 if not full_response:
974 request = compat_urllib_request.Request(url)
975 # Some webservers may serve compressed content of rather big size (e.g. gzipped flac)
976 # making it impossible to download only chunk of the file (yet we need only 512kB to
977 # test whether it's HTML or not). According to youtube-dl default Accept-Encoding
978 # that will always result in downloading the whole file that is not desirable.
979 # Therefore for extraction pass we have to override Accept-Encoding to any in order
980 # to accept raw bytes and being able to download only a chunk.
981 # It may probably better to solve this by checking Content-Type for application/octet-stream
982 # after HEAD request finishes, but not sure if we can rely on this.
983 request.add_header('Accept-Encoding', '*')
984 full_response = self._request_webpage(request, video_id)
986 # Maybe it's a direct link to a video?
987 # Be careful not to download the whole thing!
988 first_bytes = full_response.read(512)
989 if not is_html(first_bytes):
990 self._downloader.report_warning(
991 'URL could be a direct video link, returning it as such.')
992 upload_date = unified_strdate(
993 head_response.headers.get('Last-Modified'))
996 'title': compat_urllib_parse_unquote(os.path.splitext(url_basename(url))[0]),
999 'upload_date': upload_date,
1002 webpage = self._webpage_read_content(
1003 full_response, url, video_id, prefix=first_bytes)
1005 self.report_extraction(video_id)
1007 # Is it an RSS feed?
1009 doc = parse_xml(webpage)
1010 if doc.tag == 'rss':
1011 return self._extract_rss(url, video_id, doc)
1012 except compat_xml_parse_error:
1015 # Is it a Camtasia project?
1016 camtasia_res = self._extract_camtasia(url, video_id, webpage)
1017 if camtasia_res is not None:
1020 # Sometimes embedded video player is hidden behind percent encoding
1021 # (e.g. https://github.com/rg3/youtube-dl/issues/2448)
1022 # Unescaping the whole page allows to handle those cases in a generic way
1023 webpage = compat_urllib_parse.unquote(webpage)
1025 # it's tempting to parse this further, but you would
1026 # have to take into account all the variations like
1027 # Video Title - Site Name
1028 # Site Name | Video Title
1029 # Video Title - Tagline | Site Name
1030 # and so on and so forth; it's just not practical
1031 video_title = self._html_search_regex(
1032 r'(?s)<title>(.*?)</title>', webpage, 'video title',
1035 # Try to detect age limit automatically
1036 age_limit = self._rta_search(webpage)
1037 # And then there are the jokers who advertise that they use RTA,
1038 # but actually don't.
1039 AGE_LIMIT_MARKERS = [
1040 r'Proudly Labeled <a href="http://www.rtalabel.org/" title="Restricted to Adults">RTA</a>',
1042 if any(re.search(marker, webpage) for marker in AGE_LIMIT_MARKERS):
1045 # video uploader is domain name
1046 video_uploader = self._search_regex(
1047 r'^(?:https?://)?([^/]*)/.*', url, 'video uploader')
1050 def _playlist_from_matches(matches, getter=None, ie=None):
1052 self.url_result(self._proto_relative_url(getter(m) if getter else m), ie)
1054 return self.playlist_result(
1055 urlrs, playlist_id=video_id, playlist_title=video_title)
1057 # Look for BrightCove:
1058 bc_urls = BrightcoveIE._extract_brightcove_urls(webpage)
1060 self.to_screen('Brightcove video detected.')
1063 'url': smuggle_url(bc_url, {'Referer': url}),
1064 'ie_key': 'Brightcove'
1065 } for bc_url in bc_urls]
1068 '_type': 'playlist',
1069 'title': video_title,
1074 # Look for embedded rtl.nl player
1075 matches = re.findall(
1076 r'<iframe[^>]+?src="((?:https?:)?//(?:www\.)?rtl\.nl/system/videoplayer/[^"]+(?:video_)?embed[^"]+)"',
1079 return _playlist_from_matches(matches, ie='RtlNl')
1081 # Look for embedded (iframe) Vimeo player
1083 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//player\.vimeo\.com/video/.+?)\1', webpage)
1085 player_url = unescapeHTML(mobj.group('url'))
1086 surl = smuggle_url(player_url, {'Referer': url})
1087 return self.url_result(surl)
1088 # Look for embedded (swf embed) Vimeo player
1090 r'<embed[^>]+?src="((?:https?:)?//(?:www\.)?vimeo\.com/moogaloop\.swf.+?)"', webpage)
1092 return self.url_result(mobj.group(1))
1094 # Look for embedded YouTube player
1095 matches = re.findall(r'''(?x)
1104 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1108 return _playlist_from_matches(
1109 matches, lambda m: unescapeHTML(m[1]))
1111 # Look for lazyYT YouTube embed
1112 matches = re.findall(
1113 r'class="lazyYT" data-youtube-id="([^"]+)"', webpage)
1115 return _playlist_from_matches(matches, lambda m: unescapeHTML(m))
1117 # Look for embedded Dailymotion player
1118 matches = re.findall(
1119 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/embed/video/.+?)\1', webpage)
1121 return _playlist_from_matches(
1122 matches, lambda m: unescapeHTML(m[1]))
1124 # Look for embedded Dailymotion playlist player (#3822)
1126 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.[a-z]{2,3}/widget/jukebox\?.+?)\1', webpage)
1128 playlists = re.findall(
1129 r'list\[\]=/playlist/([^/]+)/', unescapeHTML(m.group('url')))
1131 return _playlist_from_matches(
1132 playlists, lambda p: '//dailymotion.com/playlist/%s' % p)
1134 # Look for embedded Wistia player
1136 r'<(?:meta[^>]+?content|iframe[^>]+?src)=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage)
1138 embed_url = self._proto_relative_url(
1139 unescapeHTML(match.group('url')))
1141 '_type': 'url_transparent',
1144 'uploader': video_uploader,
1145 'title': video_title,
1149 match = re.search(r'(?:id=["\']wistia_|data-wistia-?id=["\']|Wistia\.embed\(["\'])(?P<id>[^"\']+)', webpage)
1152 '_type': 'url_transparent',
1153 'url': 'http://fast.wistia.net/embed/iframe/{0:}'.format(match.group('id')),
1155 'uploader': video_uploader,
1156 'title': video_title,
1157 'id': match.group('id')
1160 # Look for embedded blip.tv player
1161 bliptv_url = BlipTVIE._extract_url(webpage)
1163 return self.url_result(bliptv_url, 'BlipTV')
1165 # Look for SVT player
1166 svt_url = SVTIE._extract_url(webpage)
1168 return self.url_result(svt_url, 'SVT')
1170 # Look for embedded condenast player
1171 matches = re.findall(
1172 r'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="(https?://player\.cnevids\.com/embed/[^"]+")',
1176 '_type': 'playlist',
1179 'ie_key': 'CondeNast',
1181 } for ma in matches],
1182 'title': video_title,
1186 # Look for Bandcamp pages with custom domain
1187 mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage)
1188 if mobj is not None:
1189 burl = unescapeHTML(mobj.group(1))
1190 # Don't set the extractor because it can be a track url or an album
1191 return self.url_result(burl)
1193 # Look for embedded Vevo player
1195 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:cache\.)?vevo\.com/.+?)\1', webpage)
1196 if mobj is not None:
1197 return self.url_result(mobj.group('url'))
1199 # Look for embedded Viddler player
1201 r'<(?:iframe[^>]+?src|param[^>]+?value)=(["\'])(?P<url>(?:https?:)?//(?:www\.)?viddler\.com/(?:embed|player)/.+?)\1',
1203 if mobj is not None:
1204 return self.url_result(mobj.group('url'))
1206 # Look for NYTimes player
1208 r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//graphics8\.nytimes\.com/bcvideo/[^/]+/iframe/embed\.html.+?)\1>',
1210 if mobj is not None:
1211 return self.url_result(mobj.group('url'))
1213 # Look for Libsyn player
1215 r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//html5-player\.libsyn\.com/embed/.+?)\1', webpage)
1216 if mobj is not None:
1217 return self.url_result(mobj.group('url'))
1219 # Look for Ooyala videos
1220 mobj = (re.search(r'player\.ooyala\.com/[^"?]+\?[^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or
1221 re.search(r'OO\.Player\.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage) or
1222 re.search(r'SBN\.VideoLinkset\.ooyala\([\'"](?P<ec>.{32})[\'"]\)', webpage) or
1223 re.search(r'data-ooyala-video-id\s*=\s*[\'"](?P<ec>.{32})[\'"]', webpage))
1224 if mobj is not None:
1225 return OoyalaIE._build_url_result(mobj.group('ec'))
1227 # Look for multiple Ooyala embeds on SBN network websites
1228 mobj = re.search(r'SBN\.VideoLinkset\.entryGroup\((\[.*?\])', webpage)
1229 if mobj is not None:
1230 embeds = self._parse_json(mobj.group(1), video_id, fatal=False)
1232 return _playlist_from_matches(
1233 embeds, getter=lambda v: OoyalaIE._url_for_embed_code(v['provider_video_id']), ie='Ooyala')
1235 # Look for Aparat videos
1236 mobj = re.search(r'<iframe .*?src="(http://www\.aparat\.com/video/[^"]+)"', webpage)
1237 if mobj is not None:
1238 return self.url_result(mobj.group(1), 'Aparat')
1240 # Look for MPORA videos
1241 mobj = re.search(r'<iframe .*?src="(http://mpora\.(?:com|de)/videos/[^"]+)"', webpage)
1242 if mobj is not None:
1243 return self.url_result(mobj.group(1), 'Mpora')
1245 # Look for embedded NovaMov-based player
1247 r'''(?x)<(?:pagespeed_)?iframe[^>]+?src=(["\'])
1248 (?P<url>http://(?:(?:embed|www)\.)?
1250 nowvideo\.(?:ch|sx|eu|at|ag|co)|
1251 videoweed\.(?:es|com)|
1252 movshare\.(?:net|sx|ag)|
1253 divxstage\.(?:eu|net|ch|co|at|ag))
1254 /embed\.php.+?)\1''', webpage)
1255 if mobj is not None:
1256 return self.url_result(mobj.group('url'))
1258 # Look for embedded Facebook player
1260 r'<iframe[^>]+?src=(["\'])(?P<url>https://www\.facebook\.com/video/embed.+?)\1', webpage)
1261 if mobj is not None:
1262 return self.url_result(mobj.group('url'), 'Facebook')
1264 # Look for embedded VK player
1265 mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://vk\.com/video_ext\.php.+?)\1', webpage)
1266 if mobj is not None:
1267 return self.url_result(mobj.group('url'), 'VK')
1269 # Look for embedded ivi player
1270 mobj = re.search(r'<embed[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?ivi\.ru/video/player.+?)\1', webpage)
1271 if mobj is not None:
1272 return self.url_result(mobj.group('url'), 'Ivi')
1274 # Look for embedded Huffington Post player
1276 r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed\.live\.huffingtonpost\.com/.+?)\1', webpage)
1277 if mobj is not None:
1278 return self.url_result(mobj.group('url'), 'HuffPost')
1281 mobj = re.search(r'class=["\']embedly-card["\'][^>]href=["\'](?P<url>[^"\']+)', webpage)
1282 if mobj is not None:
1283 return self.url_result(mobj.group('url'))
1284 mobj = re.search(r'class=["\']embedly-embed["\'][^>]src=["\'][^"\']*url=(?P<url>[^&]+)', webpage)
1285 if mobj is not None:
1286 return self.url_result(compat_urllib_parse.unquote(mobj.group('url')))
1288 # Look for funnyordie embed
1289 matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage)
1291 return _playlist_from_matches(
1292 matches, getter=unescapeHTML, ie='FunnyOrDie')
1294 # Look for BBC iPlayer embed
1295 matches = re.findall(r'setPlaylist\("(https?://www\.bbc\.co\.uk/iplayer/[^/]+/[\da-z]{8})"\)', webpage)
1297 return _playlist_from_matches(matches, ie='BBCCoUk')
1299 # Look for embedded RUTV player
1300 rutv_url = RUTVIE._extract_url(webpage)
1302 return self.url_result(rutv_url, 'RUTV')
1304 # Look for embedded SportBox player
1305 sportbox_urls = SportBoxEmbedIE._extract_urls(webpage)
1307 return _playlist_from_matches(sportbox_urls, ie='SportBoxEmbed')
1309 # Look for embedded TED player
1311 r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed(?:-ssl)?\.ted\.com/.+?)\1', webpage)
1312 if mobj is not None:
1313 return self.url_result(mobj.group('url'), 'TED')
1315 # Look for embedded Ustream videos
1317 r'<iframe[^>]+?src=(["\'])(?P<url>http://www\.ustream\.tv/embed/.+?)\1', webpage)
1318 if mobj is not None:
1319 return self.url_result(mobj.group('url'), 'Ustream')
1321 # Look for embedded arte.tv player
1323 r'<script [^>]*?src="(?P<url>http://www\.arte\.tv/playerv2/embed[^"]+)"',
1325 if mobj is not None:
1326 return self.url_result(mobj.group('url'), 'ArteTVEmbed')
1328 # Look for embedded smotri.com player
1329 smotri_url = SmotriIE._extract_url(webpage)
1331 return self.url_result(smotri_url, 'Smotri')
1333 # Look for embeded soundcloud player
1335 r'<iframe\s+(?:[a-zA-Z0-9_-]+="[^"]+"\s+)*src="(?P<url>https?://(?:w\.)?soundcloud\.com/player[^"]+)"',
1337 if mobj is not None:
1338 url = unescapeHTML(mobj.group('url'))
1339 return self.url_result(url)
1341 # Look for embedded vulture.com player
1343 r'<iframe src="(?P<url>https?://video\.vulture\.com/[^"]+)"',
1345 if mobj is not None:
1346 url = unescapeHTML(mobj.group('url'))
1347 return self.url_result(url, ie='Vulture')
1349 # Look for embedded mtvservices player
1351 r'<iframe src="(?P<url>https?://media\.mtvnservices\.com/embed/[^"]+)"',
1353 if mobj is not None:
1354 url = unescapeHTML(mobj.group('url'))
1355 return self.url_result(url, ie='MTVServicesEmbedded')
1357 # Look for embedded yahoo player
1359 r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:screen|movies)\.yahoo\.com/.+?\.html\?format=embed)\1',
1361 if mobj is not None:
1362 return self.url_result(mobj.group('url'), 'Yahoo')
1364 # Look for embedded sbs.com.au player
1368 <meta\s+property="og:video"\s+content=|
1371 (["\'])(?P<url>https?://(?:www\.)?sbs\.com\.au/ondemand/video/.+?)\1''',
1373 if mobj is not None:
1374 return self.url_result(mobj.group('url'), 'SBS')
1376 # Look for embedded Cinchcast player
1378 r'<iframe[^>]+?src=(["\'])(?P<url>https?://player\.cinchcast\.com/.+?)\1',
1380 if mobj is not None:
1381 return self.url_result(mobj.group('url'), 'Cinchcast')
1384 r'<iframe[^>]+?src=(["\'])(?P<url>https?://m(?:lb)?\.mlb\.com/shared/video/embed/embed\.html\?.+?)\1',
1388 r'data-video-link=["\'](?P<url>http://m.mlb.com/video/[^"\']+)',
1390 if mobj is not None:
1391 return self.url_result(mobj.group('url'), 'MLB')
1394 r'<iframe[^>]+?src=(["\'])(?P<url>%s)\1' % CondeNastIE.EMBED_URL,
1396 if mobj is not None:
1397 return self.url_result(self._proto_relative_url(mobj.group('url'), scheme='http:'), 'CondeNast')
1400 r'<iframe[^>]+src="(?P<url>https?://new\.livestream\.com/[^"]+/player[^"]+)"',
1402 if mobj is not None:
1403 return self.url_result(mobj.group('url'), 'Livestream')
1405 # Look for Zapiks embed
1407 r'<iframe[^>]+src="(?P<url>https?://(?:www\.)?zapiks\.fr/index\.php\?.+?)"', webpage)
1408 if mobj is not None:
1409 return self.url_result(mobj.group('url'), 'Zapiks')
1411 # Look for Kaltura embeds
1413 r"(?s)kWidget\.(?:thumb)?[Ee]mbed\(\{.*?'wid'\s*:\s*'_?(?P<partner_id>[^']+)',.*?'entry_id'\s*:\s*'(?P<id>[^']+)',", webpage)
1414 if mobj is not None:
1415 return self.url_result('kaltura:%(partner_id)s:%(id)s' % mobj.groupdict(), 'Kaltura')
1417 # Look for Eagle.Platform embeds
1419 r'<iframe[^>]+src="(?P<url>https?://.+?\.media\.eagleplatform\.com/index/player\?.+?)"', webpage)
1420 if mobj is not None:
1421 return self.url_result(mobj.group('url'), 'EaglePlatform')
1423 # Look for ClipYou (uses Eagle.Platform) embeds
1425 r'<iframe[^>]+src="https?://(?P<host>media\.clipyou\.ru)/index/player\?.*\brecord_id=(?P<id>\d+).*"', webpage)
1426 if mobj is not None:
1427 return self.url_result('eagleplatform:%(host)s:%(id)s' % mobj.groupdict(), 'EaglePlatform')
1429 # Look for Pladform embeds
1431 r'<iframe[^>]+src="(?P<url>https?://out\.pladform\.ru/player\?.+?)"', webpage)
1432 if mobj is not None:
1433 return self.url_result(mobj.group('url'), 'Pladform')
1435 # Look for Playwire embeds
1437 r'<script[^>]+data-config=(["\'])(?P<url>(?:https?:)?//config\.playwire\.com/.+?)\1', webpage)
1438 if mobj is not None:
1439 return self.url_result(mobj.group('url'))
1441 # Look for 5min embeds
1443 r'<meta[^>]+property="og:video"[^>]+content="https?://embed\.5min\.com/(?P<id>[0-9]+)/?', webpage)
1444 if mobj is not None:
1445 return self.url_result('5min:%s' % mobj.group('id'), 'FiveMin')
1447 # Look for Crooks and Liars embeds
1449 r'<(?:iframe[^>]+src|param[^>]+value)=(["\'])(?P<url>(?:https?:)?//embed\.crooksandliars\.com/(?:embed|v)/.+?)\1', webpage)
1450 if mobj is not None:
1451 return self.url_result(mobj.group('url'))
1453 # Look for NBC Sports VPlayer embeds
1454 nbc_sports_url = NBCSportsVPlayerIE._extract_url(webpage)
1456 return self.url_result(nbc_sports_url, 'NBCSportsVPlayer')
1458 # Look for UDN embeds
1460 r'<iframe[^>]+src="(?P<url>%s)"' % UDNEmbedIE._VALID_URL, webpage)
1461 if mobj is not None:
1462 return self.url_result(
1463 compat_urlparse.urljoin(url, mobj.group('url')), 'UDNEmbed')
1465 # Look for Senate ISVP iframe
1466 senate_isvp_url = SenateISVPIE._search_iframe_url(webpage)
1468 return self.url_result(senate_isvp_url, 'SenateISVP')
1470 def check_video(vurl):
1471 if YoutubeIE.suitable(vurl):
1473 vpath = compat_urlparse.urlparse(vurl).path
1474 vext = determine_ext(vpath)
1475 return '.' in vpath and vext not in ('swf', 'png', 'jpg', 'srt', 'sbv', 'sub', 'vtt', 'ttml')
1477 def filter_video(urls):
1478 return list(filter(check_video, urls))
1480 # Start with something easy: JW Player in SWFObject
1481 found = filter_video(re.findall(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage))
1483 # Look for gorilla-vid style embedding
1484 found = filter_video(re.findall(r'''(?sx)
1488 jwplayer\s*\(\s*["'][^'"]+["']\s*\)\s*\.setup
1491 ['"]?file['"]?\s*:\s*["\'](.*?)["\']''', webpage))
1493 # Broaden the search a little bit
1494 found = filter_video(re.findall(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage))
1496 # Broaden the findall a little bit: JWPlayer JS loader
1497 found = filter_video(re.findall(
1498 r'[^A-Za-z0-9]?file["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage))
1501 found = filter_video(re.findall(r'''(?xs)
1502 flowplayer\("[^"]+",\s*
1504 \s*\{[^}]+? ["']?clip["']?\s*:\s*\{\s*
1505 ["']?url["']?\s*:\s*["']([^"']+)["']
1510 r"cinerama\.embedPlayer\(\s*\'[^']+\',\s*'([^']+)'", webpage)
1512 # Try to find twitter cards info
1513 found = filter_video(re.findall(
1514 r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage))
1516 # We look for Open Graph info:
1517 # We have to match any number spaces between elements, some sites try to align them (eg.: statigr.am)
1518 m_video_type = re.findall(r'<meta.*?property="og:video:type".*?content="video/(.*?)"', webpage)
1519 # We only look in og:video if the MIME type is a video, don't try if it's a Flash player:
1520 if m_video_type is not None:
1521 found = filter_video(re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage))
1524 found = re.findall(r'(?s)<video[^<]*(?:>.*?<source[^>]*)?\s+src=["\'](.*?)["\']', webpage)
1526 REDIRECT_REGEX = r'[0-9]{,2};\s*(?:URL|url)=\'?([^\'"]+)'
1528 r'(?i)<meta\s+(?=(?:[a-z-]+="[^"]+"\s+)*http-equiv="refresh")'
1529 r'(?:[a-z-]+="[^"]+"\s+)*?content="%s' % REDIRECT_REGEX,
1532 # Look also in Refresh HTTP header
1533 refresh_header = head_response.headers.get('Refresh')
1535 found = re.search(REDIRECT_REGEX, refresh_header)
1537 new_url = compat_urlparse.urljoin(url, found.group(1))
1538 self.report_following_redirect(new_url)
1544 raise UnsupportedError(url)
1547 for video_url in found:
1548 video_url = compat_urlparse.urljoin(url, video_url)
1549 video_id = compat_urllib_parse.unquote(os.path.basename(video_url))
1551 # Sometimes, jwplayer extraction will result in a YouTube URL
1552 if YoutubeIE.suitable(video_url):
1553 entries.append(self.url_result(video_url, 'Youtube'))
1556 # here's a fun little line of code for you:
1557 video_id = os.path.splitext(video_id)[0]
1559 if determine_ext(video_url) == 'smil':
1562 'formats': self._extract_smil_formats(video_url, video_id),
1563 'uploader': video_uploader,
1564 'title': video_title,
1565 'age_limit': age_limit,
1571 'uploader': video_uploader,
1572 'title': video_title,
1573 'age_limit': age_limit,
1576 if len(entries) == 1:
1579 for num, e in enumerate(entries, start=1):
1580 # 'url' results don't have a title
1581 if e.get('title') is not None:
1582 e['title'] = '%s (%d)' % (e['title'], num)
1584 '_type': 'playlist',