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 .tvc import TVCIE
38 from .sportbox import SportBoxEmbedIE
39 from .smotri import SmotriIE
40 from .condenast import CondeNastIE
41 from .udn import UDNEmbedIE
42 from .senateisvp import SenateISVPIE
43 from .bliptv import BlipTVIE
44 from .svt import SVTIE
45 from .pornhub import PornHubIE
48 class GenericIE(InfoExtractor):
49 IE_DESC = 'Generic downloader that works on some sites'
53 # Direct link to a video
55 'url': 'http://media.w3.org/2010/05/sintel/trailer.mp4',
56 'md5': '67d406c2bcb6af27fa886f31aa934bbe',
61 'upload_date': '20100513',
64 # Direct link to media delivered compressed (until Accept-Encoding is *)
66 'url': 'http://calimero.tk/muzik/FictionJunction-Parallel_Hearts.flac',
67 'md5': '128c42e68b13950268b648275386fc74',
69 'id': 'FictionJunction-Parallel_Hearts',
71 'title': 'FictionJunction-Parallel_Hearts',
72 'upload_date': '20140522',
74 'expected_warnings': [
75 'URL could be a direct video link, returning it as such.'
78 # Direct download with broken HEAD
80 'url': 'http://ai-radio.org:8000/radio.opus',
87 'skip_download': True, # infinite live stream
89 'expected_warnings': [
90 r'501.*Not Implemented'
93 # Direct link with incorrect MIME type
95 'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
96 'md5': '4ccbebe5f36706d85221f204d7eb5913',
98 'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
99 'id': '5_Lennart_Poettering_-_Systemd',
101 'title': '5_Lennart_Poettering_-_Systemd',
102 'upload_date': '20141120',
104 'expected_warnings': [
105 'URL could be a direct video link, returning it as such.'
110 'url': 'http://phihag.de/2014/youtube-dl/rss2.xml',
112 'id': 'http://phihag.de/2014/youtube-dl/rss2.xml',
113 'title': 'Zero Punctuation',
114 'description': 're:.*groundbreaking video review series.*'
116 'playlist_mincount': 11,
118 # RSS feed with enclosure
120 'url': 'http://podcastfeeds.nbcnews.com/audio/podcast/MSNBC-MADDOW-NETCAST-M4V.xml',
122 'id': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
124 'upload_date': '20150228',
125 'title': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
130 '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',
134 'upload_date': '20130224',
135 'uploader_id': 'TheVerge',
136 'description': 're:^Chris Ziegler takes a look at the\.*',
137 'uploader': 'The Verge',
138 'title': 'First Firefox OS phones side-by-side',
141 'skip_download': False,
145 'url': 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
146 'md5': '85b90ccc9d73b4acd9138d3af4c27f89',
148 'id': '13601338388002',
150 'uploader': 'www.hodiho.fr',
151 'title': 'R\u00e9gis plante sa Jeep',
154 # bandcamp page with custom domain
156 'add_ie': ['Bandcamp'],
157 'url': 'http://bronyrock.com/track/the-pony-mash',
161 'title': 'The Pony Mash',
162 'uploader': 'M_Pallante',
164 'skip': 'There is a limit of 200 free downloads / month for the test song',
166 # embedded brightcove video
167 # it also tests brightcove videos that need to set the 'Referer' in the
170 'add_ie': ['Brightcove'],
171 'url': 'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/',
173 'id': '2765128793001',
175 'title': 'Le cours de bourse : l’analyse technique',
176 'description': 'md5:7e9ad046e968cb2d1114004aba466fd9',
177 'uploader': 'BFM BUSINESS',
180 'skip_download': True,
184 # https://github.com/rg3/youtube-dl/issues/2253
185 'url': 'http://bcove.me/i6nfkrc3',
186 'md5': '0ba9446db037002366bab3b3eb30c88c',
188 'id': '3101154703001',
190 'title': 'Still no power',
191 'uploader': 'thestar.com',
192 '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.',
194 'add_ie': ['Brightcove'],
197 'url': 'http://www.championat.com/video/football/v/87/87499.html',
198 'md5': 'fb973ecf6e4a78a67453647444222983',
200 'id': '3414141473001',
202 'title': 'Видео. Удаление Дзагоева (ЦСКА)',
203 'description': 'Онлайн-трансляция матча ЦСКА - "Волга"',
204 'uploader': 'Championat',
208 # https://github.com/rg3/youtube-dl/issues/3541
209 'add_ie': ['Brightcove'],
210 'url': 'http://www.kijk.nl/sbs6/leermijvrouwenkennen/videos/jqMiXKAYan2S/aflevering-1',
212 'id': '3866516442001',
214 'title': 'Leer mij vrouwen kennen: Aflevering 1',
215 'description': 'Leer mij vrouwen kennen: Aflevering 1',
216 'uploader': 'SBS Broadcasting',
218 'skip': 'Restricted to Netherlands',
220 'skip_download': True, # m3u8 download
225 'url': 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219',
226 'md5': '166dd577b433b4d4ebfee10b0824d8ff',
228 'id': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ',
230 'title': '2cc213299525360.mov', # that's what we get
232 'add_ie': ['Ooyala'],
234 # multiple ooyala embeds on SBN network websites
236 'url': 'http://www.sbnation.com/college-football-recruiting/2015/2/3/7970291/national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
238 'id': 'national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
239 'title': '25 lies you will tell yourself on National Signing Day - SBNation.com',
241 'playlist_mincount': 3,
243 'skip_download': True,
245 'add_ie': ['Ooyala'],
249 'url': 'http://www.tested.com/science/weird/460206-tested-grinding-coffee-2000-frames-second/',
253 'title': 'Tested: Grinding Coffee at 2000 Frames Per Second',
254 'upload_date': '20140225',
255 'description': 'md5:06a40fbf30b220468f1e0957c0f558ff',
256 'uploader': 'Tested',
257 'uploader_id': 'testedcom',
259 # No need to test YoutubeIE here
261 'skip_download': True,
266 'url': 'http://www.theguardian.com/world/2014/mar/11/obama-zach-galifianakis-between-two-ferns',
270 'title': 'Between Two Ferns with Zach Galifianakis: President Barack Obama',
271 'description': 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.',
276 'url': 'http://www.bbc.co.uk/blogs/adamcurtis/posts/BUGGER',
278 'title': 'BBC - Blogs - Adam Curtis - BUGGER',
280 'playlist_mincount': 18,
284 'url': 'http://www.rg.ru/2014/03/15/reg-dfo/anklav-anons.html',
288 'title': 'Охотское море стало целиком российским',
289 'description': 'md5:5ed62483b14663e2a95ebbe115eb8f43',
293 'skip_download': True,
298 'url': 'http://sch1298sz.mskobr.ru/dou_edu/karamel_ki/filial_galleries/video/iframe_src_http_tvc_ru_video_iframe_id_55304_isplay_false_acc_video_id_channel_brand_id_11_show_episodes_episode_id_32307_frameb/',
302 'title': 'Дошкольное воспитание',
307 'url': 'http://www.vestifinance.ru/articles/25753',
310 'title': 'Вести Экономика ― Прямые трансляции с Форума-выставки "Госзаказ-2013"',
315 'title': 'Госзаказ. День 3',
321 'title': 'Госзаказ. День 2',
327 'title': 'Госзаказ. День 1',
333 'skip_download': True,
338 'url': 'http://en.support.wordpress.com/videos/ted-talks/',
339 'md5': '65fdff94098e4a607385a60c5177c638',
343 'title': 'Hidden miracles of the natural world',
344 'uploader': 'Louie Schwartzberg',
345 'description': 'md5:8145d19d320ff3e52f28401f4c4283b9',
348 # Embeded Ustream video
350 'url': 'http://www.american.edu/spa/pti/nsa-privacy-janus-2014.cfm',
351 'md5': '27b99cdb639c9b12a79bca876a073417',
355 'uploader': 'AU SPA: The NSA and Privacy',
356 'title': 'NSA and Privacy Forum Debate featuring General Hayden and Barton Gellman'
359 # nowvideo embed hidden behind percent encoding
361 'url': 'http://www.waoanime.tv/the-super-dimension-fortress-macross-episode-1/',
362 'md5': '2baf4ddd70f697d94b1c18cf796d5107',
364 'id': '06e53103ca9aa',
366 'title': 'Macross Episode 001 Watch Macross Episode 001 onl',
367 'description': 'No description',
372 'url': 'http://www.tv-replay.fr/redirection/20-03-14/x-enius-arte-10753389.html',
373 'md5': '7653032cbb25bf6c80d80f217055fa43',
375 'id': '048195-004_PLUS7-F',
378 'description': 'md5:d5fdf32ef6613cdbfd516ae658abf168',
379 'upload_date': '20140320',
382 'skip_download': 'Requires rtmpdump'
387 'url': 'http://www.wired.com/2014/04/honda-asimo/',
388 'md5': 'ba0dfe966fa007657bd1443ee672db0f',
390 'id': '53501be369702d3275860000',
392 'title': 'Honda’s New Asimo Robot Is More Human Than Ever',
397 'url': 'http://www.spi0n.com/zap-spi0n-com-n216/',
398 'md5': '441aeeb82eb72c422c7f14ec533999cd',
400 'id': 'k2mm4bCdJ6CQ2i7c8o2',
402 'title': 'Le Zap de Spi0n n°216 - Zapping du Web',
405 'add_ie': ['Dailymotion'],
409 'url': 'http://www.badzine.de/ansicht/datum/2014/06/09/so-funktioniert-die-neue-englische-badminton-liga.html',
413 'title': 'The NBL Auction 2014',
414 'uploader': 'BADMINTON England',
415 'uploader_id': 'BADMINTONEvents',
416 'upload_date': '20140603',
417 'description': 'md5:9ef128a69f1e262a700ed83edb163a73',
419 'add_ie': ['Youtube'],
421 'skip_download': True,
426 'url': 'http://www.gametrailers.com/news-post/76093/north-america-europe-is-getting-that-mario-kart-8-mercedes-dlc-too',
427 'md5': '35727f82f58c76d996fc188f9755b0d5',
429 'id': '0306a69b-8adf-4fb5-aace-75f8e8cbfca9',
432 'description': 'Mario\'s life in the fast lane has never looked so good.',
435 # YouTube embed via <data-embed-url="">
437 'url': 'https://play.google.com/store/apps/details?id=com.gameloft.android.ANMP.GloftA8HM',
441 'title': 'Asphalt 8: Airborne - Update - Welcome to Dubai!',
442 'uploader': 'Gameloft',
443 'uploader_id': 'gameloft',
444 'upload_date': '20140828',
445 'description': 'md5:c80da9ed3d83ae6d1876c834de03e1c4',
448 'skip_download': True,
453 'url': 'http://www.ll.mit.edu/workshops/education/videocourses/antennas/lecture1/video/',
455 'md5': '0c5e352edabf715d762b0ad4e6d9ee67',
457 'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
458 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - video1',
463 'md5': '10e4bb3aaca9fd630e273ff92d9f3c63',
465 'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final_PIP',
466 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - pip',
472 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
477 'url': 'http://www.handjobhub.com/video/busty-blonde-siri-tit-fuck-while-wank-6313.html',
478 'md5': '9d65602bf31c6e20014319c7d07fba27',
480 'id': '5123ea6d5e5a7',
483 'uploader': 'www.handjobhub.com',
484 'title': 'Busty Blonde Siri Tit Fuck While Wank at HandjobHub.com',
487 # Multiple brightcove videos
488 # https://github.com/rg3/youtube-dl/issues/2283
490 'url': 'http://www.newyorker.com/online/blogs/newsdesk/2014/01/always-never-nuclear-command-and-control.html',
492 'id': 'always-never',
493 'title': 'Always / Never - The New Yorker',
497 'extract_flat': False,
498 'skip_download': True,
503 'url': 'http://umpire-empire.com/index.php/topic/58125-laz-decides-no-thats-low/',
504 'md5': '96f09a37e44da40dd083e12d9a683327',
508 'title': 'Ump changes call to ball',
509 'description': 'md5:71c11215384298a172a6dcb4c2e20685',
511 'timestamp': 1401537900,
512 'upload_date': '20140531',
513 'thumbnail': 're:^https?://.*\.jpg$',
518 'url': 'http://education-portal.com/academy/lesson/north-american-exploration-failed-colonies-of-spain-france-england.html#lesson',
519 'md5': '8788b683c777a5cf25621eaf286d0c23',
523 'title': 'md5:51364a8d3d009997ba99656004b5e20d',
525 'filesize': 182808282,
526 'uploader': 'education-portal.com',
530 'url': 'http://thoughtworks.wistia.com/medias/uxjb0lwrcz',
531 'md5': 'baf49c2baa8a7de5f3fc145a8506dcd4',
535 'title': 'Conversation about Hexagonal Rails Part 1 - ThoughtWorks',
537 'uploader': 'thoughtworks.wistia.com',
542 'url': 'http://nakedsecurity.sophos.com/2014/10/29/sscc-171-are-you-sure-that-1234-is-a-bad-password-podcast/',
546 'description': 'md5:ff867d6b555488ad3c52572bb33d432c',
547 'uploader': 'Sophos Security',
548 'title': 'Chet Chat 171 - Oct 29, 2014',
549 'upload_date': '20141029',
554 'url': 'http://www.esa.int/Our_Activities/Space_Science/Rosetta/Philae_comet_touch-down_webcast',
558 'upload_date': '20141112',
559 'title': 'Rosetta #CometLanding webcast HL 10',
564 'url': 'http://discourse.ubuntu.com/t/unity-8-desktop-mode-windows-on-mir/1986',
567 'title': 'Unity 8 desktop-mode windows on Mir! - Ubuntu Discourse',
569 'playlist_mincount': 2,
573 'url': 'http://undergroundwellness.com/podcasts/306-5-steps-to-permanent-gut-healing/',
577 'upload_date': '20141126',
578 'title': 'Jack Tips: 5 Steps to Permanent Gut Healing',
583 'url': 'http://www.abc.net.au/7.30/content/2015/s4164797.htm',
585 'id': '730m_DandD_1901_512k',
587 'uploader': 'www.abc.net.au',
588 'title': 'Game of Thrones with dice - Dungeons and Dragons fantasy role-playing game gets new life - 19/01/2015',
591 # embedded viddler video
593 'url': 'http://deadspin.com/i-cant-stop-watching-john-wall-chop-the-nuggets-with-th-1681801597',
597 'uploader': 'deadspin',
598 'title': 'WALL-TO-GORTAT',
599 'timestamp': 1422285291,
600 'upload_date': '20150126',
602 'add_ie': ['Viddler'],
606 'url': 'http://thedailyshow.cc.com/podcast/episodetwelve',
610 'title': "The Daily Show Podcast without Jon Stewart - Episode 12: Bassem Youssef: Egypt's Jon Stewart",
611 'description': 'md5:601cb790edd05908957dae8aaa866465',
612 'upload_date': '20150220',
617 'url': 'http://media.nationalarchives.gov.uk/index.php/webinar-using-discovery-national-archives-online-catalogue/',
621 'upload_date': '20150212',
622 'uploader': 'The National Archives UK',
623 'description': 'md5:a236581cd2449dd2df4f93412f3f01c6',
624 'uploader_id': 'NationalArchives08',
625 'title': 'Webinar: Using Discovery, The National Archives’ online catalogue',
630 'url': 'http://www.rtlnieuws.nl/nieuws/buitenland/aanslagen-kopenhagen',
631 'playlist_mincount': 5,
633 'id': 'aanslagen-kopenhagen',
634 'title': 'Aanslagen Kopenhagen | RTL Nieuws',
639 'url': 'http://www.skipass.com/news/116090-bon-appetit-s5ep3-baqueira-mi-cor.html',
643 'title': 'EP3S5 - Bon Appétit - Baqueira Mi Corazon !',
648 'url': 'http://www.monumentalnetwork.com/videos/john-carlson-postgame-2-25-15',
652 'upload_date': '20150226',
653 'uploader_id': 'MonumentalSports-Kaltura@perfectsensedigital.com',
655 'title': 'John Carlson Postgame 2/25/15',
658 # Eagle.Platform embed (generic URL)
660 'url': 'http://lenta.ru/news/2015/03/06/navalny/',
664 'title': 'Навальный вышел на свободу',
665 'description': 'md5:d97861ac9ae77377f3f20eaf9d04b4f5',
666 'thumbnail': 're:^https?://.*\.jpg$',
672 # ClipYou (Eagle.Platform) embed (custom URL)
674 'url': 'http://muz-tv.ru/play/7129/',
678 'title': "'O Sole Mio",
679 'thumbnail': 're:^https?://.*\.jpg$',
686 'url': 'http://muz-tv.ru/kinozal/view/7400/',
690 'title': 'Тайны перевала Дятлова • 1 серия 2 часть',
691 'description': 'Документальный сериал-расследование одной из самых жутких тайн ХХ века',
692 'thumbnail': 're:^https?://.*\.jpg$',
699 'url': 'http://www.cinemablend.com/new/First-Joe-Dirt-2-Trailer-Teaser-Stupid-Greatness-70874.html',
703 'title': 'Joe Dirt 2 Beautiful Loser Teaser Trailer',
704 'thumbnail': 're:^https?://.*\.png$',
710 'url': 'http://techcrunch.com/video/facebook-creates-on-this-day-crunch-report/518726732/',
711 'md5': '4c6f127a30736b59b3e2c19234ee2bf7',
715 'title': 'Facebook Creates "On This Day" | Crunch Report',
720 'url': 'http://www.svt.se/sport/ishockey/jagr-tacklar-giroux-under-intervjun',
724 'title': 'Här trycker Jagr till Giroux (under SVT-intervjun)',
729 # Crooks and Liars embed
731 'url': 'http://crooksandliars.com/2015/04/fox-friends-says-protecting-atheists',
735 'title': "Fox & Friends Says Protecting Atheists From Discrimination Is Anti-Christian!",
736 'description': 'md5:e1a46ad1650e3a5ec7196d432799127f',
737 'timestamp': 1428207000,
738 'upload_date': '20150405',
739 'uploader': 'Heather',
742 # Crooks and Liars external embed
744 'url': 'http://theothermccain.com/2010/02/02/video-proves-that-bill-kristol-has-been-watching-glenn-beck/comment-page-1/',
746 'id': 'MTE3MjUtMzQ2MzA',
748 'title': 'md5:5e3662a81a4014d24c250d76d41a08d5',
749 'description': 'md5:9b8e9542d6c3c5de42d6451b7d780cec',
750 'timestamp': 1265032391,
751 'upload_date': '20100201',
752 'uploader': 'Heather',
755 # NBC Sports vplayer embed
757 'url': 'http://www.riderfans.com/forum/showthread.php?121827-Freeman&s=e98fa1ea6dc08e886b1678d35212494a',
759 'id': 'ln7x1qSThw4k',
761 'title': "PFT Live: New leader in the 'new-look' defense",
762 'description': 'md5:65a19b4bbfb3b0c0c5768bed1dfad74e',
767 'url': 'http://www.udn.com/news/story/7314/822787',
768 'md5': 'fd2060e988c326991037b9aff9df21a6',
772 'title': '中一中男師變性 全校師生力挺',
773 'thumbnail': 're:^https?://.*\.jpg$',
778 'url': 'http://www.businessinsider.com/excel-index-match-vlookup-video-how-to-2015-2?IR=T',
780 'id': '50YnY4czr4ms1vJ7yz3xzq0excz_pUMs',
782 'description': 'VIDEO: Index/Match versus VLOOKUP.',
783 'title': 'This is what separates the Excel masters from the wannabes',
787 'skip_download': True,
790 # Contains a SMIL manifest
792 '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',
796 'title': '+ Football: Lottery Champions League Europe',
797 'uploader': 'www.telewebion.com',
801 'skip_download': True,
804 # Brightcove URL in single quotes
806 'url': 'http://www.sportsnet.ca/baseball/mlb/sn-presents-russell-martin-world-citizen/',
807 'md5': '4ae374f1f8b91c889c4b9203c8c752af',
809 'id': '4255764656001',
811 'title': 'SN Presents: Russell Martin, World Citizen',
812 '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.',
813 'uploader': 'Rogers Sportsnet',
818 def report_following_redirect(self, new_url):
819 """Report information extraction."""
820 self._downloader.to_screen('[redirect] Following redirect to %s' % new_url)
822 def _extract_rss(self, url, video_id, doc):
823 playlist_title = doc.find('./channel/title').text
824 playlist_desc_el = doc.find('./channel/description')
825 playlist_desc = None if playlist_desc_el is None else playlist_desc_el.text
828 for it in doc.findall('./channel/item'):
829 next_url = xpath_text(it, 'link', fatal=False)
831 enclosure_nodes = it.findall('./enclosure')
832 for e in enclosure_nodes:
833 next_url = e.attrib.get('url')
843 'title': it.find('title').text,
849 'title': playlist_title,
850 'description': playlist_desc,
854 def _extract_camtasia(self, url, video_id, webpage):
855 """ Returns None if no camtasia video can be found. """
857 camtasia_cfg = self._search_regex(
858 r'fo\.addVariable\(\s*"csConfigFile",\s*"([^"]+)"\s*\);',
859 webpage, 'camtasia configuration file', default=None)
860 if camtasia_cfg is None:
863 title = self._html_search_meta('DC.title', webpage, fatal=True)
865 camtasia_url = compat_urlparse.urljoin(url, camtasia_cfg)
866 camtasia_cfg = self._download_xml(
867 camtasia_url, video_id,
868 note='Downloading camtasia configuration',
869 errnote='Failed to download camtasia configuration')
870 fileset_node = camtasia_cfg.find('./playlist/array/fileset')
873 for n in fileset_node.getchildren():
874 url_n = n.find('./uri')
879 'id': os.path.splitext(url_n.text.rpartition('/')[2])[0],
880 'title': '%s - %s' % (title, n.tag),
881 'url': compat_urlparse.urljoin(url, url_n.text),
882 'duration': float_or_none(n.find('./duration').text),
891 def _real_extract(self, url):
892 if url.startswith('//'):
895 'url': self.http_scheme() + url,
898 parsed_url = compat_urlparse.urlparse(url)
899 if not parsed_url.scheme:
900 default_search = self._downloader.params.get('default_search')
901 if default_search is None:
902 default_search = 'fixup_error'
904 if default_search in ('auto', 'auto_warning', 'fixup_error'):
906 self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http')
907 return self.url_result('http://' + url)
908 elif default_search != 'fixup_error':
909 if default_search == 'auto_warning':
910 if re.match(r'^(?:url|URL)$', url):
911 raise ExtractorError(
912 'Invalid URL: %r . Call youtube-dl like this: youtube-dl -v "https://www.youtube.com/watch?v=BaW_jenozKc" ' % url,
915 self._downloader.report_warning(
916 'Falling back to youtube search for %s . Set --default-search "auto" to suppress this warning.' % url)
917 return self.url_result('ytsearch:' + url)
919 if default_search in ('error', 'fixup_error'):
920 raise ExtractorError(
921 '%r is not a valid URL. '
922 'Set --default-search "ytsearch" (or run youtube-dl "ytsearch:%s" ) to search YouTube'
923 % (url, url), expected=True)
925 if ':' not in default_search:
926 default_search += ':'
927 return self.url_result(default_search + url)
929 url, smuggled_data = unsmuggle_url(url)
931 is_intentional = smuggled_data and smuggled_data.get('to_generic')
932 if smuggled_data and 'force_videoid' in smuggled_data:
933 force_videoid = smuggled_data['force_videoid']
934 video_id = force_videoid
936 video_id = compat_urllib_parse_unquote(os.path.splitext(url.rstrip('/').split('/')[-1])[0])
938 self.to_screen('%s: Requesting header' % video_id)
940 head_req = HEADRequest(url)
941 head_response = self._request_webpage(
943 note=False, errnote='Could not send HEAD request to %s' % url,
946 if head_response is not False:
948 new_url = head_response.geturl()
950 self.report_following_redirect(new_url)
952 new_url = smuggle_url(
953 new_url, {'force_videoid': force_videoid})
954 return self.url_result(new_url)
957 if head_response is False:
958 request = compat_urllib_request.Request(url)
959 request.add_header('Accept-Encoding', '*')
960 full_response = self._request_webpage(request, video_id)
961 head_response = full_response
963 # Check for direct link to a video
964 content_type = head_response.headers.get('Content-Type', '')
965 m = re.match(r'^(?P<type>audio|video|application(?=/ogg$))/(?P<format_id>.+)$', content_type)
967 upload_date = unified_strdate(
968 head_response.headers.get('Last-Modified'))
971 'title': compat_urllib_parse_unquote(os.path.splitext(url_basename(url))[0]),
974 'format_id': m.group('format_id'),
976 'vcodec': 'none' if m.group('type') == 'audio' else None
978 'upload_date': upload_date,
981 if not self._downloader.params.get('test', False) and not is_intentional:
982 self._downloader.report_warning('Falling back on generic information extractor.')
984 if not full_response:
985 request = compat_urllib_request.Request(url)
986 # Some webservers may serve compressed content of rather big size (e.g. gzipped flac)
987 # making it impossible to download only chunk of the file (yet we need only 512kB to
988 # test whether it's HTML or not). According to youtube-dl default Accept-Encoding
989 # that will always result in downloading the whole file that is not desirable.
990 # Therefore for extraction pass we have to override Accept-Encoding to any in order
991 # to accept raw bytes and being able to download only a chunk.
992 # It may probably better to solve this by checking Content-Type for application/octet-stream
993 # after HEAD request finishes, but not sure if we can rely on this.
994 request.add_header('Accept-Encoding', '*')
995 full_response = self._request_webpage(request, video_id)
997 # Maybe it's a direct link to a video?
998 # Be careful not to download the whole thing!
999 first_bytes = full_response.read(512)
1000 if not is_html(first_bytes):
1001 self._downloader.report_warning(
1002 'URL could be a direct video link, returning it as such.')
1003 upload_date = unified_strdate(
1004 head_response.headers.get('Last-Modified'))
1007 'title': compat_urllib_parse_unquote(os.path.splitext(url_basename(url))[0]),
1010 'upload_date': upload_date,
1013 webpage = self._webpage_read_content(
1014 full_response, url, video_id, prefix=first_bytes)
1016 self.report_extraction(video_id)
1018 # Is it an RSS feed?
1020 doc = parse_xml(webpage)
1021 if doc.tag == 'rss':
1022 return self._extract_rss(url, video_id, doc)
1023 except compat_xml_parse_error:
1026 # Is it a Camtasia project?
1027 camtasia_res = self._extract_camtasia(url, video_id, webpage)
1028 if camtasia_res is not None:
1031 # Sometimes embedded video player is hidden behind percent encoding
1032 # (e.g. https://github.com/rg3/youtube-dl/issues/2448)
1033 # Unescaping the whole page allows to handle those cases in a generic way
1034 webpage = compat_urllib_parse.unquote(webpage)
1036 # it's tempting to parse this further, but you would
1037 # have to take into account all the variations like
1038 # Video Title - Site Name
1039 # Site Name | Video Title
1040 # Video Title - Tagline | Site Name
1041 # and so on and so forth; it's just not practical
1042 video_title = self._html_search_regex(
1043 r'(?s)<title>(.*?)</title>', webpage, 'video title',
1046 # Try to detect age limit automatically
1047 age_limit = self._rta_search(webpage)
1048 # And then there are the jokers who advertise that they use RTA,
1049 # but actually don't.
1050 AGE_LIMIT_MARKERS = [
1051 r'Proudly Labeled <a href="http://www.rtalabel.org/" title="Restricted to Adults">RTA</a>',
1053 if any(re.search(marker, webpage) for marker in AGE_LIMIT_MARKERS):
1056 # video uploader is domain name
1057 video_uploader = self._search_regex(
1058 r'^(?:https?://)?([^/]*)/.*', url, 'video uploader')
1061 def _playlist_from_matches(matches, getter=None, ie=None):
1063 self.url_result(self._proto_relative_url(getter(m) if getter else m), ie)
1065 return self.playlist_result(
1066 urlrs, playlist_id=video_id, playlist_title=video_title)
1068 # Look for BrightCove:
1069 bc_urls = BrightcoveIE._extract_brightcove_urls(webpage)
1071 self.to_screen('Brightcove video detected.')
1074 'url': smuggle_url(bc_url, {'Referer': url}),
1075 'ie_key': 'Brightcove'
1076 } for bc_url in bc_urls]
1079 '_type': 'playlist',
1080 'title': video_title,
1085 # Look for embedded rtl.nl player
1086 matches = re.findall(
1087 r'<iframe[^>]+?src="((?:https?:)?//(?:www\.)?rtl\.nl/system/videoplayer/[^"]+(?:video_)?embed[^"]+)"',
1090 return _playlist_from_matches(matches, ie='RtlNl')
1092 # Look for embedded (iframe) Vimeo player
1094 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//player\.vimeo\.com/video/.+?)\1', webpage)
1096 player_url = unescapeHTML(mobj.group('url'))
1097 surl = smuggle_url(player_url, {'Referer': url})
1098 return self.url_result(surl)
1099 # Look for embedded (swf embed) Vimeo player
1101 r'<embed[^>]+?src="((?:https?:)?//(?:www\.)?vimeo\.com/moogaloop\.swf.+?)"', webpage)
1103 return self.url_result(mobj.group(1))
1105 # Look for embedded YouTube player
1106 matches = re.findall(r'''(?x)
1115 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1119 return _playlist_from_matches(
1120 matches, lambda m: unescapeHTML(m[1]))
1122 # Look for lazyYT YouTube embed
1123 matches = re.findall(
1124 r'class="lazyYT" data-youtube-id="([^"]+)"', webpage)
1126 return _playlist_from_matches(matches, lambda m: unescapeHTML(m))
1128 # Look for embedded Dailymotion player
1129 matches = re.findall(
1130 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/embed/video/.+?)\1', webpage)
1132 return _playlist_from_matches(
1133 matches, lambda m: unescapeHTML(m[1]))
1135 # Look for embedded Dailymotion playlist player (#3822)
1137 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.[a-z]{2,3}/widget/jukebox\?.+?)\1', webpage)
1139 playlists = re.findall(
1140 r'list\[\]=/playlist/([^/]+)/', unescapeHTML(m.group('url')))
1142 return _playlist_from_matches(
1143 playlists, lambda p: '//dailymotion.com/playlist/%s' % p)
1145 # Look for embedded Wistia player
1147 r'<(?:meta[^>]+?content|iframe[^>]+?src)=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage)
1149 embed_url = self._proto_relative_url(
1150 unescapeHTML(match.group('url')))
1152 '_type': 'url_transparent',
1155 'uploader': video_uploader,
1156 'title': video_title,
1160 match = re.search(r'(?:id=["\']wistia_|data-wistia-?id=["\']|Wistia\.embed\(["\'])(?P<id>[^"\']+)', webpage)
1163 '_type': 'url_transparent',
1164 'url': 'http://fast.wistia.net/embed/iframe/{0:}'.format(match.group('id')),
1166 'uploader': video_uploader,
1167 'title': video_title,
1168 'id': match.group('id')
1171 # Look for embedded blip.tv player
1172 bliptv_url = BlipTVIE._extract_url(webpage)
1174 return self.url_result(bliptv_url, 'BlipTV')
1176 # Look for SVT player
1177 svt_url = SVTIE._extract_url(webpage)
1179 return self.url_result(svt_url, 'SVT')
1181 # Look for embedded condenast player
1182 matches = re.findall(
1183 r'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="(https?://player\.cnevids\.com/embed/[^"]+")',
1187 '_type': 'playlist',
1190 'ie_key': 'CondeNast',
1192 } for ma in matches],
1193 'title': video_title,
1197 # Look for Bandcamp pages with custom domain
1198 mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage)
1199 if mobj is not None:
1200 burl = unescapeHTML(mobj.group(1))
1201 # Don't set the extractor because it can be a track url or an album
1202 return self.url_result(burl)
1204 # Look for embedded Vevo player
1206 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:cache\.)?vevo\.com/.+?)\1', webpage)
1207 if mobj is not None:
1208 return self.url_result(mobj.group('url'))
1210 # Look for embedded Viddler player
1212 r'<(?:iframe[^>]+?src|param[^>]+?value)=(["\'])(?P<url>(?:https?:)?//(?:www\.)?viddler\.com/(?:embed|player)/.+?)\1',
1214 if mobj is not None:
1215 return self.url_result(mobj.group('url'))
1217 # Look for NYTimes player
1219 r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//graphics8\.nytimes\.com/bcvideo/[^/]+/iframe/embed\.html.+?)\1>',
1221 if mobj is not None:
1222 return self.url_result(mobj.group('url'))
1224 # Look for Libsyn player
1226 r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//html5-player\.libsyn\.com/embed/.+?)\1', webpage)
1227 if mobj is not None:
1228 return self.url_result(mobj.group('url'))
1230 # Look for Ooyala videos
1231 mobj = (re.search(r'player\.ooyala\.com/[^"?]+\?[^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or
1232 re.search(r'OO\.Player\.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage) or
1233 re.search(r'SBN\.VideoLinkset\.ooyala\([\'"](?P<ec>.{32})[\'"]\)', webpage) or
1234 re.search(r'data-ooyala-video-id\s*=\s*[\'"](?P<ec>.{32})[\'"]', webpage))
1235 if mobj is not None:
1236 return OoyalaIE._build_url_result(mobj.group('ec'))
1238 # Look for multiple Ooyala embeds on SBN network websites
1239 mobj = re.search(r'SBN\.VideoLinkset\.entryGroup\((\[.*?\])', webpage)
1240 if mobj is not None:
1241 embeds = self._parse_json(mobj.group(1), video_id, fatal=False)
1243 return _playlist_from_matches(
1244 embeds, getter=lambda v: OoyalaIE._url_for_embed_code(v['provider_video_id']), ie='Ooyala')
1246 # Look for Aparat videos
1247 mobj = re.search(r'<iframe .*?src="(http://www\.aparat\.com/video/[^"]+)"', webpage)
1248 if mobj is not None:
1249 return self.url_result(mobj.group(1), 'Aparat')
1251 # Look for MPORA videos
1252 mobj = re.search(r'<iframe .*?src="(http://mpora\.(?:com|de)/videos/[^"]+)"', webpage)
1253 if mobj is not None:
1254 return self.url_result(mobj.group(1), 'Mpora')
1256 # Look for embedded NovaMov-based player
1258 r'''(?x)<(?:pagespeed_)?iframe[^>]+?src=(["\'])
1259 (?P<url>http://(?:(?:embed|www)\.)?
1261 nowvideo\.(?:ch|sx|eu|at|ag|co)|
1262 videoweed\.(?:es|com)|
1263 movshare\.(?:net|sx|ag)|
1264 divxstage\.(?:eu|net|ch|co|at|ag))
1265 /embed\.php.+?)\1''', webpage)
1266 if mobj is not None:
1267 return self.url_result(mobj.group('url'))
1269 # Look for embedded Facebook player
1271 r'<iframe[^>]+?src=(["\'])(?P<url>https://www\.facebook\.com/video/embed.+?)\1', webpage)
1272 if mobj is not None:
1273 return self.url_result(mobj.group('url'), 'Facebook')
1275 # Look for embedded VK player
1276 mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://vk\.com/video_ext\.php.+?)\1', webpage)
1277 if mobj is not None:
1278 return self.url_result(mobj.group('url'), 'VK')
1280 # Look for embedded ivi player
1281 mobj = re.search(r'<embed[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?ivi\.ru/video/player.+?)\1', webpage)
1282 if mobj is not None:
1283 return self.url_result(mobj.group('url'), 'Ivi')
1285 # Look for embedded Huffington Post player
1287 r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed\.live\.huffingtonpost\.com/.+?)\1', webpage)
1288 if mobj is not None:
1289 return self.url_result(mobj.group('url'), 'HuffPost')
1292 mobj = re.search(r'class=["\']embedly-card["\'][^>]href=["\'](?P<url>[^"\']+)', webpage)
1293 if mobj is not None:
1294 return self.url_result(mobj.group('url'))
1295 mobj = re.search(r'class=["\']embedly-embed["\'][^>]src=["\'][^"\']*url=(?P<url>[^&]+)', webpage)
1296 if mobj is not None:
1297 return self.url_result(compat_urllib_parse.unquote(mobj.group('url')))
1299 # Look for funnyordie embed
1300 matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage)
1302 return _playlist_from_matches(
1303 matches, getter=unescapeHTML, ie='FunnyOrDie')
1305 # Look for BBC iPlayer embed
1306 matches = re.findall(r'setPlaylist\("(https?://www\.bbc\.co\.uk/iplayer/[^/]+/[\da-z]{8})"\)', webpage)
1308 return _playlist_from_matches(matches, ie='BBCCoUk')
1310 # Look for embedded RUTV player
1311 rutv_url = RUTVIE._extract_url(webpage)
1313 return self.url_result(rutv_url, 'RUTV')
1315 # Look for embedded TVC player
1316 tvc_url = TVCIE._extract_url(webpage)
1318 return self.url_result(tvc_url, 'TVC')
1320 # Look for embedded SportBox player
1321 sportbox_urls = SportBoxEmbedIE._extract_urls(webpage)
1323 return _playlist_from_matches(sportbox_urls, ie='SportBoxEmbed')
1325 # Look for embedded PornHub player
1326 pornhub_url = PornHubIE._extract_url(webpage)
1328 return self.url_result(pornhub_url, 'PornHub')
1330 # Look for embedded Tvigle player
1332 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//cloud\.tvigle\.ru/video/.+?)\1', webpage)
1333 if mobj is not None:
1334 return self.url_result(mobj.group('url'), 'Tvigle')
1336 # Look for embedded TED player
1338 r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed(?:-ssl)?\.ted\.com/.+?)\1', webpage)
1339 if mobj is not None:
1340 return self.url_result(mobj.group('url'), 'TED')
1342 # Look for embedded Ustream videos
1344 r'<iframe[^>]+?src=(["\'])(?P<url>http://www\.ustream\.tv/embed/.+?)\1', webpage)
1345 if mobj is not None:
1346 return self.url_result(mobj.group('url'), 'Ustream')
1348 # Look for embedded arte.tv player
1350 r'<script [^>]*?src="(?P<url>http://www\.arte\.tv/playerv2/embed[^"]+)"',
1352 if mobj is not None:
1353 return self.url_result(mobj.group('url'), 'ArteTVEmbed')
1355 # Look for embedded smotri.com player
1356 smotri_url = SmotriIE._extract_url(webpage)
1358 return self.url_result(smotri_url, 'Smotri')
1360 # Look for embeded soundcloud player
1362 r'<iframe\s+(?:[a-zA-Z0-9_-]+="[^"]+"\s+)*src="(?P<url>https?://(?:w\.)?soundcloud\.com/player[^"]+)"',
1364 if mobj is not None:
1365 url = unescapeHTML(mobj.group('url'))
1366 return self.url_result(url)
1368 # Look for embedded vulture.com player
1370 r'<iframe src="(?P<url>https?://video\.vulture\.com/[^"]+)"',
1372 if mobj is not None:
1373 url = unescapeHTML(mobj.group('url'))
1374 return self.url_result(url, ie='Vulture')
1376 # Look for embedded mtvservices player
1378 r'<iframe src="(?P<url>https?://media\.mtvnservices\.com/embed/[^"]+)"',
1380 if mobj is not None:
1381 url = unescapeHTML(mobj.group('url'))
1382 return self.url_result(url, ie='MTVServicesEmbedded')
1384 # Look for embedded yahoo player
1386 r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:screen|movies)\.yahoo\.com/.+?\.html\?format=embed)\1',
1388 if mobj is not None:
1389 return self.url_result(mobj.group('url'), 'Yahoo')
1391 # Look for embedded sbs.com.au player
1395 <meta\s+property="og:video"\s+content=|
1398 (["\'])(?P<url>https?://(?:www\.)?sbs\.com\.au/ondemand/video/.+?)\1''',
1400 if mobj is not None:
1401 return self.url_result(mobj.group('url'), 'SBS')
1403 # Look for embedded Cinchcast player
1405 r'<iframe[^>]+?src=(["\'])(?P<url>https?://player\.cinchcast\.com/.+?)\1',
1407 if mobj is not None:
1408 return self.url_result(mobj.group('url'), 'Cinchcast')
1411 r'<iframe[^>]+?src=(["\'])(?P<url>https?://m(?:lb)?\.mlb\.com/shared/video/embed/embed\.html\?.+?)\1',
1415 r'data-video-link=["\'](?P<url>http://m.mlb.com/video/[^"\']+)',
1417 if mobj is not None:
1418 return self.url_result(mobj.group('url'), 'MLB')
1421 r'<iframe[^>]+?src=(["\'])(?P<url>%s)\1' % CondeNastIE.EMBED_URL,
1423 if mobj is not None:
1424 return self.url_result(self._proto_relative_url(mobj.group('url'), scheme='http:'), 'CondeNast')
1427 r'<iframe[^>]+src="(?P<url>https?://new\.livestream\.com/[^"]+/player[^"]+)"',
1429 if mobj is not None:
1430 return self.url_result(mobj.group('url'), 'Livestream')
1432 # Look for Zapiks embed
1434 r'<iframe[^>]+src="(?P<url>https?://(?:www\.)?zapiks\.fr/index\.php\?.+?)"', webpage)
1435 if mobj is not None:
1436 return self.url_result(mobj.group('url'), 'Zapiks')
1438 # Look for Kaltura embeds
1440 r"(?s)kWidget\.(?:thumb)?[Ee]mbed\(\{.*?'wid'\s*:\s*'_?(?P<partner_id>[^']+)',.*?'entry_id'\s*:\s*'(?P<id>[^']+)',", webpage)
1441 if mobj is not None:
1442 return self.url_result('kaltura:%(partner_id)s:%(id)s' % mobj.groupdict(), 'Kaltura')
1444 # Look for Eagle.Platform embeds
1446 r'<iframe[^>]+src="(?P<url>https?://.+?\.media\.eagleplatform\.com/index/player\?.+?)"', webpage)
1447 if mobj is not None:
1448 return self.url_result(mobj.group('url'), 'EaglePlatform')
1450 # Look for ClipYou (uses Eagle.Platform) embeds
1452 r'<iframe[^>]+src="https?://(?P<host>media\.clipyou\.ru)/index/player\?.*\brecord_id=(?P<id>\d+).*"', webpage)
1453 if mobj is not None:
1454 return self.url_result('eagleplatform:%(host)s:%(id)s' % mobj.groupdict(), 'EaglePlatform')
1456 # Look for Pladform embeds
1458 r'<iframe[^>]+src="(?P<url>https?://out\.pladform\.ru/player\?.+?)"', webpage)
1459 if mobj is not None:
1460 return self.url_result(mobj.group('url'), 'Pladform')
1462 # Look for Playwire embeds
1464 r'<script[^>]+data-config=(["\'])(?P<url>(?:https?:)?//config\.playwire\.com/.+?)\1', webpage)
1465 if mobj is not None:
1466 return self.url_result(mobj.group('url'))
1468 # Look for 5min embeds
1470 r'<meta[^>]+property="og:video"[^>]+content="https?://embed\.5min\.com/(?P<id>[0-9]+)/?', webpage)
1471 if mobj is not None:
1472 return self.url_result('5min:%s' % mobj.group('id'), 'FiveMin')
1474 # Look for Crooks and Liars embeds
1476 r'<(?:iframe[^>]+src|param[^>]+value)=(["\'])(?P<url>(?:https?:)?//embed\.crooksandliars\.com/(?:embed|v)/.+?)\1', webpage)
1477 if mobj is not None:
1478 return self.url_result(mobj.group('url'))
1480 # Look for NBC Sports VPlayer embeds
1481 nbc_sports_url = NBCSportsVPlayerIE._extract_url(webpage)
1483 return self.url_result(nbc_sports_url, 'NBCSportsVPlayer')
1485 # Look for UDN embeds
1487 r'<iframe[^>]+src="(?P<url>%s)"' % UDNEmbedIE._VALID_URL, webpage)
1488 if mobj is not None:
1489 return self.url_result(
1490 compat_urlparse.urljoin(url, mobj.group('url')), 'UDNEmbed')
1492 # Look for Senate ISVP iframe
1493 senate_isvp_url = SenateISVPIE._search_iframe_url(webpage)
1495 return self.url_result(senate_isvp_url, 'SenateISVP')
1497 def check_video(vurl):
1498 if YoutubeIE.suitable(vurl):
1500 vpath = compat_urlparse.urlparse(vurl).path
1501 vext = determine_ext(vpath)
1502 return '.' in vpath and vext not in ('swf', 'png', 'jpg', 'srt', 'sbv', 'sub', 'vtt', 'ttml')
1504 def filter_video(urls):
1505 return list(filter(check_video, urls))
1507 # Start with something easy: JW Player in SWFObject
1508 found = filter_video(re.findall(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage))
1510 # Look for gorilla-vid style embedding
1511 found = filter_video(re.findall(r'''(?sx)
1515 jwplayer\s*\(\s*["'][^'"]+["']\s*\)\s*\.setup
1518 ['"]?file['"]?\s*:\s*["\'](.*?)["\']''', webpage))
1520 # Broaden the search a little bit
1521 found = filter_video(re.findall(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage))
1523 # Broaden the findall a little bit: JWPlayer JS loader
1524 found = filter_video(re.findall(
1525 r'[^A-Za-z0-9]?file["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage))
1528 found = filter_video(re.findall(r'''(?xs)
1529 flowplayer\("[^"]+",\s*
1531 \s*\{[^}]+? ["']?clip["']?\s*:\s*\{\s*
1532 ["']?url["']?\s*:\s*["']([^"']+)["']
1537 r"cinerama\.embedPlayer\(\s*\'[^']+\',\s*'([^']+)'", webpage)
1539 # Try to find twitter cards info
1540 found = filter_video(re.findall(
1541 r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage))
1543 # We look for Open Graph info:
1544 # We have to match any number spaces between elements, some sites try to align them (eg.: statigr.am)
1545 m_video_type = re.findall(r'<meta.*?property="og:video:type".*?content="video/(.*?)"', webpage)
1546 # We only look in og:video if the MIME type is a video, don't try if it's a Flash player:
1547 if m_video_type is not None:
1548 found = filter_video(re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage))
1551 found = re.findall(r'(?s)<video[^<]*(?:>.*?<source[^>]*)?\s+src=["\'](.*?)["\']', webpage)
1553 REDIRECT_REGEX = r'[0-9]{,2};\s*(?:URL|url)=\'?([^\'"]+)'
1555 r'(?i)<meta\s+(?=(?:[a-z-]+="[^"]+"\s+)*http-equiv="refresh")'
1556 r'(?:[a-z-]+="[^"]+"\s+)*?content="%s' % REDIRECT_REGEX,
1559 # Look also in Refresh HTTP header
1560 refresh_header = head_response.headers.get('Refresh')
1562 found = re.search(REDIRECT_REGEX, refresh_header)
1564 new_url = compat_urlparse.urljoin(url, found.group(1))
1565 self.report_following_redirect(new_url)
1571 raise UnsupportedError(url)
1574 for video_url in found:
1575 video_url = compat_urlparse.urljoin(url, video_url)
1576 video_id = compat_urllib_parse.unquote(os.path.basename(video_url))
1578 # Sometimes, jwplayer extraction will result in a YouTube URL
1579 if YoutubeIE.suitable(video_url):
1580 entries.append(self.url_result(video_url, 'Youtube'))
1583 # here's a fun little line of code for you:
1584 video_id = os.path.splitext(video_id)[0]
1586 if determine_ext(video_url) == 'smil':
1589 'formats': self._extract_smil_formats(video_url, video_id),
1590 'uploader': video_uploader,
1591 'title': video_title,
1592 'age_limit': age_limit,
1598 'uploader': video_uploader,
1599 'title': video_title,
1600 'age_limit': age_limit,
1603 if len(entries) == 1:
1606 for num, e in enumerate(entries, start=1):
1607 # 'url' results don't have a title
1608 if e.get('title') is not None:
1609 e['title'] = '%s (%d)' % (e['title'], num)
1611 '_type': 'playlist',