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
47 class GenericIE(InfoExtractor):
48 IE_DESC = 'Generic downloader that works on some sites'
52 # Direct link to a video
54 'url': 'http://media.w3.org/2010/05/sintel/trailer.mp4',
55 'md5': '67d406c2bcb6af27fa886f31aa934bbe',
60 'upload_date': '20100513',
63 # Direct link to media delivered compressed (until Accept-Encoding is *)
65 'url': 'http://calimero.tk/muzik/FictionJunction-Parallel_Hearts.flac',
66 'md5': '128c42e68b13950268b648275386fc74',
68 'id': 'FictionJunction-Parallel_Hearts',
70 'title': 'FictionJunction-Parallel_Hearts',
71 'upload_date': '20140522',
73 'expected_warnings': [
74 'URL could be a direct video link, returning it as such.'
77 # Direct download with broken HEAD
79 'url': 'http://ai-radio.org:8000/radio.opus',
86 'skip_download': True, # infinite live stream
88 'expected_warnings': [
89 r'501.*Not Implemented'
92 # Direct link with incorrect MIME type
94 'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
95 'md5': '4ccbebe5f36706d85221f204d7eb5913',
97 'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
98 'id': '5_Lennart_Poettering_-_Systemd',
100 'title': '5_Lennart_Poettering_-_Systemd',
101 'upload_date': '20141120',
103 'expected_warnings': [
104 'URL could be a direct video link, returning it as such.'
109 'url': 'http://phihag.de/2014/youtube-dl/rss2.xml',
111 'id': 'http://phihag.de/2014/youtube-dl/rss2.xml',
112 'title': 'Zero Punctuation',
113 'description': 're:.*groundbreaking video review series.*'
115 'playlist_mincount': 11,
117 # RSS feed with enclosure
119 'url': 'http://podcastfeeds.nbcnews.com/audio/podcast/MSNBC-MADDOW-NETCAST-M4V.xml',
121 'id': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
123 'upload_date': '20150228',
124 'title': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
129 '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',
133 'upload_date': '20130224',
134 'uploader_id': 'TheVerge',
135 'description': 're:^Chris Ziegler takes a look at the\.*',
136 'uploader': 'The Verge',
137 'title': 'First Firefox OS phones side-by-side',
140 'skip_download': False,
144 'url': 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
145 'md5': '85b90ccc9d73b4acd9138d3af4c27f89',
147 'id': '13601338388002',
149 'uploader': 'www.hodiho.fr',
150 'title': 'R\u00e9gis plante sa Jeep',
153 # bandcamp page with custom domain
155 'add_ie': ['Bandcamp'],
156 'url': 'http://bronyrock.com/track/the-pony-mash',
160 'title': 'The Pony Mash',
161 'uploader': 'M_Pallante',
163 'skip': 'There is a limit of 200 free downloads / month for the test song',
165 # embedded brightcove video
166 # it also tests brightcove videos that need to set the 'Referer' in the
169 'add_ie': ['Brightcove'],
170 'url': 'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/',
172 'id': '2765128793001',
174 'title': 'Le cours de bourse : l’analyse technique',
175 'description': 'md5:7e9ad046e968cb2d1114004aba466fd9',
176 'uploader': 'BFM BUSINESS',
179 'skip_download': True,
183 # https://github.com/rg3/youtube-dl/issues/2253
184 'url': 'http://bcove.me/i6nfkrc3',
185 'md5': '0ba9446db037002366bab3b3eb30c88c',
187 'id': '3101154703001',
189 'title': 'Still no power',
190 'uploader': 'thestar.com',
191 '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.',
193 'add_ie': ['Brightcove'],
196 'url': 'http://www.championat.com/video/football/v/87/87499.html',
197 'md5': 'fb973ecf6e4a78a67453647444222983',
199 'id': '3414141473001',
201 'title': 'Видео. Удаление Дзагоева (ЦСКА)',
202 'description': 'Онлайн-трансляция матча ЦСКА - "Волга"',
203 'uploader': 'Championat',
207 # https://github.com/rg3/youtube-dl/issues/3541
208 'add_ie': ['Brightcove'],
209 'url': 'http://www.kijk.nl/sbs6/leermijvrouwenkennen/videos/jqMiXKAYan2S/aflevering-1',
211 'id': '3866516442001',
213 'title': 'Leer mij vrouwen kennen: Aflevering 1',
214 'description': 'Leer mij vrouwen kennen: Aflevering 1',
215 'uploader': 'SBS Broadcasting',
217 'skip': 'Restricted to Netherlands',
219 'skip_download': True, # m3u8 download
224 'url': 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219',
225 'md5': '166dd577b433b4d4ebfee10b0824d8ff',
227 'id': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ',
229 'title': '2cc213299525360.mov', # that's what we get
231 'add_ie': ['Ooyala'],
233 # multiple ooyala embeds on SBN network websites
235 'url': 'http://www.sbnation.com/college-football-recruiting/2015/2/3/7970291/national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
237 'id': 'national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
238 'title': '25 lies you will tell yourself on National Signing Day - SBNation.com',
240 'playlist_mincount': 3,
242 'skip_download': True,
244 'add_ie': ['Ooyala'],
248 'url': 'http://www.tested.com/science/weird/460206-tested-grinding-coffee-2000-frames-second/',
252 'title': 'Tested: Grinding Coffee at 2000 Frames Per Second',
253 'upload_date': '20140225',
254 'description': 'md5:06a40fbf30b220468f1e0957c0f558ff',
255 'uploader': 'Tested',
256 'uploader_id': 'testedcom',
258 # No need to test YoutubeIE here
260 'skip_download': True,
265 'url': 'http://www.theguardian.com/world/2014/mar/11/obama-zach-galifianakis-between-two-ferns',
269 'title': 'Between Two Ferns with Zach Galifianakis: President Barack Obama',
270 'description': 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.',
275 'url': 'http://www.bbc.co.uk/blogs/adamcurtis/posts/BUGGER',
277 'title': 'BBC - Blogs - Adam Curtis - BUGGER',
279 'playlist_mincount': 18,
283 'url': 'http://www.rg.ru/2014/03/15/reg-dfo/anklav-anons.html',
287 'title': 'Охотское море стало целиком российским',
288 'description': 'md5:5ed62483b14663e2a95ebbe115eb8f43',
292 'skip_download': True,
297 '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/',
301 'title': 'Дошкольное воспитание',
306 'url': 'http://www.vestifinance.ru/articles/25753',
309 'title': 'Вести Экономика ― Прямые трансляции с Форума-выставки "Госзаказ-2013"',
314 'title': 'Госзаказ. День 3',
320 'title': 'Госзаказ. День 2',
326 'title': 'Госзаказ. День 1',
332 'skip_download': True,
337 'url': 'http://en.support.wordpress.com/videos/ted-talks/',
338 'md5': '65fdff94098e4a607385a60c5177c638',
342 'title': 'Hidden miracles of the natural world',
343 'uploader': 'Louie Schwartzberg',
344 'description': 'md5:8145d19d320ff3e52f28401f4c4283b9',
347 # Embeded Ustream video
349 'url': 'http://www.american.edu/spa/pti/nsa-privacy-janus-2014.cfm',
350 'md5': '27b99cdb639c9b12a79bca876a073417',
354 'uploader': 'AU SPA: The NSA and Privacy',
355 'title': 'NSA and Privacy Forum Debate featuring General Hayden and Barton Gellman'
358 # nowvideo embed hidden behind percent encoding
360 'url': 'http://www.waoanime.tv/the-super-dimension-fortress-macross-episode-1/',
361 'md5': '2baf4ddd70f697d94b1c18cf796d5107',
363 'id': '06e53103ca9aa',
365 'title': 'Macross Episode 001 Watch Macross Episode 001 onl',
366 'description': 'No description',
371 'url': 'http://www.tv-replay.fr/redirection/20-03-14/x-enius-arte-10753389.html',
372 'md5': '7653032cbb25bf6c80d80f217055fa43',
374 'id': '048195-004_PLUS7-F',
377 'description': 'md5:d5fdf32ef6613cdbfd516ae658abf168',
378 'upload_date': '20140320',
381 'skip_download': 'Requires rtmpdump'
386 'url': 'http://www.wired.com/2014/04/honda-asimo/',
387 'md5': 'ba0dfe966fa007657bd1443ee672db0f',
389 'id': '53501be369702d3275860000',
391 'title': 'Honda’s New Asimo Robot Is More Human Than Ever',
396 'url': 'http://www.spi0n.com/zap-spi0n-com-n216/',
397 'md5': '441aeeb82eb72c422c7f14ec533999cd',
399 'id': 'k2mm4bCdJ6CQ2i7c8o2',
401 'title': 'Le Zap de Spi0n n°216 - Zapping du Web',
404 'add_ie': ['Dailymotion'],
408 'url': 'http://www.badzine.de/ansicht/datum/2014/06/09/so-funktioniert-die-neue-englische-badminton-liga.html',
412 'title': 'The NBL Auction 2014',
413 'uploader': 'BADMINTON England',
414 'uploader_id': 'BADMINTONEvents',
415 'upload_date': '20140603',
416 'description': 'md5:9ef128a69f1e262a700ed83edb163a73',
418 'add_ie': ['Youtube'],
420 'skip_download': True,
425 'url': 'http://www.gametrailers.com/news-post/76093/north-america-europe-is-getting-that-mario-kart-8-mercedes-dlc-too',
426 'md5': '35727f82f58c76d996fc188f9755b0d5',
428 'id': '0306a69b-8adf-4fb5-aace-75f8e8cbfca9',
431 'description': 'Mario\'s life in the fast lane has never looked so good.',
434 # YouTube embed via <data-embed-url="">
436 'url': 'https://play.google.com/store/apps/details?id=com.gameloft.android.ANMP.GloftA8HM',
440 'title': 'Asphalt 8: Airborne - Update - Welcome to Dubai!',
441 'uploader': 'Gameloft',
442 'uploader_id': 'gameloft',
443 'upload_date': '20140828',
444 'description': 'md5:c80da9ed3d83ae6d1876c834de03e1c4',
447 'skip_download': True,
452 'url': 'http://www.ll.mit.edu/workshops/education/videocourses/antennas/lecture1/video/',
454 'md5': '0c5e352edabf715d762b0ad4e6d9ee67',
456 'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
457 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - video1',
462 'md5': '10e4bb3aaca9fd630e273ff92d9f3c63',
464 'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final_PIP',
465 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - pip',
471 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
476 'url': 'http://www.handjobhub.com/video/busty-blonde-siri-tit-fuck-while-wank-6313.html',
477 'md5': '9d65602bf31c6e20014319c7d07fba27',
479 'id': '5123ea6d5e5a7',
482 'uploader': 'www.handjobhub.com',
483 'title': 'Busty Blonde Siri Tit Fuck While Wank at HandjobHub.com',
486 # Multiple brightcove videos
487 # https://github.com/rg3/youtube-dl/issues/2283
489 'url': 'http://www.newyorker.com/online/blogs/newsdesk/2014/01/always-never-nuclear-command-and-control.html',
491 'id': 'always-never',
492 'title': 'Always / Never - The New Yorker',
496 'extract_flat': False,
497 'skip_download': True,
502 'url': 'http://umpire-empire.com/index.php/topic/58125-laz-decides-no-thats-low/',
503 'md5': '96f09a37e44da40dd083e12d9a683327',
507 'title': 'Ump changes call to ball',
508 'description': 'md5:71c11215384298a172a6dcb4c2e20685',
510 'timestamp': 1401537900,
511 'upload_date': '20140531',
512 'thumbnail': 're:^https?://.*\.jpg$',
517 'url': 'http://education-portal.com/academy/lesson/north-american-exploration-failed-colonies-of-spain-france-england.html#lesson',
518 'md5': '8788b683c777a5cf25621eaf286d0c23',
522 'title': 'md5:51364a8d3d009997ba99656004b5e20d',
524 'filesize': 182808282,
525 'uploader': 'education-portal.com',
529 'url': 'http://thoughtworks.wistia.com/medias/uxjb0lwrcz',
530 'md5': 'baf49c2baa8a7de5f3fc145a8506dcd4',
534 'title': 'Conversation about Hexagonal Rails Part 1 - ThoughtWorks',
536 'uploader': 'thoughtworks.wistia.com',
541 'url': 'http://nakedsecurity.sophos.com/2014/10/29/sscc-171-are-you-sure-that-1234-is-a-bad-password-podcast/',
545 'description': 'md5:ff867d6b555488ad3c52572bb33d432c',
546 'uploader': 'Sophos Security',
547 'title': 'Chet Chat 171 - Oct 29, 2014',
548 'upload_date': '20141029',
553 'url': 'http://www.esa.int/Our_Activities/Space_Science/Rosetta/Philae_comet_touch-down_webcast',
557 'upload_date': '20141112',
558 'title': 'Rosetta #CometLanding webcast HL 10',
563 'url': 'http://discourse.ubuntu.com/t/unity-8-desktop-mode-windows-on-mir/1986',
566 'title': 'Unity 8 desktop-mode windows on Mir! - Ubuntu Discourse',
568 'playlist_mincount': 2,
572 'url': 'http://undergroundwellness.com/podcasts/306-5-steps-to-permanent-gut-healing/',
576 'upload_date': '20141126',
577 'title': 'Jack Tips: 5 Steps to Permanent Gut Healing',
582 'url': 'http://www.abc.net.au/7.30/content/2015/s4164797.htm',
584 'id': '730m_DandD_1901_512k',
586 'uploader': 'www.abc.net.au',
587 'title': 'Game of Thrones with dice - Dungeons and Dragons fantasy role-playing game gets new life - 19/01/2015',
590 # embedded viddler video
592 'url': 'http://deadspin.com/i-cant-stop-watching-john-wall-chop-the-nuggets-with-th-1681801597',
596 'uploader': 'deadspin',
597 'title': 'WALL-TO-GORTAT',
598 'timestamp': 1422285291,
599 'upload_date': '20150126',
601 'add_ie': ['Viddler'],
605 'url': 'http://thedailyshow.cc.com/podcast/episodetwelve',
609 'title': "The Daily Show Podcast without Jon Stewart - Episode 12: Bassem Youssef: Egypt's Jon Stewart",
610 'description': 'md5:601cb790edd05908957dae8aaa866465',
611 'upload_date': '20150220',
616 'url': 'http://media.nationalarchives.gov.uk/index.php/webinar-using-discovery-national-archives-online-catalogue/',
620 'upload_date': '20150212',
621 'uploader': 'The National Archives UK',
622 'description': 'md5:a236581cd2449dd2df4f93412f3f01c6',
623 'uploader_id': 'NationalArchives08',
624 'title': 'Webinar: Using Discovery, The National Archives’ online catalogue',
629 'url': 'http://www.rtlnieuws.nl/nieuws/buitenland/aanslagen-kopenhagen',
630 'playlist_mincount': 5,
632 'id': 'aanslagen-kopenhagen',
633 'title': 'Aanslagen Kopenhagen | RTL Nieuws',
638 'url': 'http://www.skipass.com/news/116090-bon-appetit-s5ep3-baqueira-mi-cor.html',
642 'title': 'EP3S5 - Bon Appétit - Baqueira Mi Corazon !',
647 'url': 'http://www.monumentalnetwork.com/videos/john-carlson-postgame-2-25-15',
651 'upload_date': '20150226',
652 'uploader_id': 'MonumentalSports-Kaltura@perfectsensedigital.com',
654 'title': 'John Carlson Postgame 2/25/15',
657 # Eagle.Platform embed (generic URL)
659 'url': 'http://lenta.ru/news/2015/03/06/navalny/',
663 'title': 'Навальный вышел на свободу',
664 'description': 'md5:d97861ac9ae77377f3f20eaf9d04b4f5',
665 'thumbnail': 're:^https?://.*\.jpg$',
671 # ClipYou (Eagle.Platform) embed (custom URL)
673 'url': 'http://muz-tv.ru/play/7129/',
677 'title': "'O Sole Mio",
678 'thumbnail': 're:^https?://.*\.jpg$',
685 'url': 'http://muz-tv.ru/kinozal/view/7400/',
689 'title': 'Тайны перевала Дятлова • 1 серия 2 часть',
690 'description': 'Документальный сериал-расследование одной из самых жутких тайн ХХ века',
691 'thumbnail': 're:^https?://.*\.jpg$',
698 'url': 'http://www.cinemablend.com/new/First-Joe-Dirt-2-Trailer-Teaser-Stupid-Greatness-70874.html',
702 'title': 'Joe Dirt 2 Beautiful Loser Teaser Trailer',
703 'thumbnail': 're:^https?://.*\.png$',
709 'url': 'http://techcrunch.com/video/facebook-creates-on-this-day-crunch-report/518726732/',
710 'md5': '4c6f127a30736b59b3e2c19234ee2bf7',
714 'title': 'Facebook Creates "On This Day" | Crunch Report',
719 'url': 'http://www.svt.se/sport/ishockey/jagr-tacklar-giroux-under-intervjun',
723 'title': 'Här trycker Jagr till Giroux (under SVT-intervjun)',
728 # Crooks and Liars embed
730 'url': 'http://crooksandliars.com/2015/04/fox-friends-says-protecting-atheists',
734 'title': "Fox & Friends Says Protecting Atheists From Discrimination Is Anti-Christian!",
735 'description': 'md5:e1a46ad1650e3a5ec7196d432799127f',
736 'timestamp': 1428207000,
737 'upload_date': '20150405',
738 'uploader': 'Heather',
741 # Crooks and Liars external embed
743 'url': 'http://theothermccain.com/2010/02/02/video-proves-that-bill-kristol-has-been-watching-glenn-beck/comment-page-1/',
745 'id': 'MTE3MjUtMzQ2MzA',
747 'title': 'md5:5e3662a81a4014d24c250d76d41a08d5',
748 'description': 'md5:9b8e9542d6c3c5de42d6451b7d780cec',
749 'timestamp': 1265032391,
750 'upload_date': '20100201',
751 'uploader': 'Heather',
754 # NBC Sports vplayer embed
756 'url': 'http://www.riderfans.com/forum/showthread.php?121827-Freeman&s=e98fa1ea6dc08e886b1678d35212494a',
758 'id': 'ln7x1qSThw4k',
760 'title': "PFT Live: New leader in the 'new-look' defense",
761 'description': 'md5:65a19b4bbfb3b0c0c5768bed1dfad74e',
766 'url': 'http://www.udn.com/news/story/7314/822787',
767 'md5': 'fd2060e988c326991037b9aff9df21a6',
771 'title': '中一中男師變性 全校師生力挺',
772 'thumbnail': 're:^https?://.*\.jpg$',
777 'url': 'http://www.businessinsider.com/excel-index-match-vlookup-video-how-to-2015-2?IR=T',
779 'id': '50YnY4czr4ms1vJ7yz3xzq0excz_pUMs',
781 'description': 'VIDEO: Index/Match versus VLOOKUP.',
782 'title': 'This is what separates the Excel masters from the wannabes',
786 'skip_download': True,
789 # Contains a SMIL manifest
791 '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',
795 'title': '+ Football: Lottery Champions League Europe',
796 'uploader': 'www.telewebion.com',
800 'skip_download': True,
803 # Brightcove URL in single quotes
805 'url': 'http://www.sportsnet.ca/baseball/mlb/sn-presents-russell-martin-world-citizen/',
806 'md5': '4ae374f1f8b91c889c4b9203c8c752af',
808 'id': '4255764656001',
810 'title': 'SN Presents: Russell Martin, World Citizen',
811 '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.',
812 'uploader': 'Rogers Sportsnet',
817 def report_following_redirect(self, new_url):
818 """Report information extraction."""
819 self._downloader.to_screen('[redirect] Following redirect to %s' % new_url)
821 def _extract_rss(self, url, video_id, doc):
822 playlist_title = doc.find('./channel/title').text
823 playlist_desc_el = doc.find('./channel/description')
824 playlist_desc = None if playlist_desc_el is None else playlist_desc_el.text
827 for it in doc.findall('./channel/item'):
828 next_url = xpath_text(it, 'link', fatal=False)
830 enclosure_nodes = it.findall('./enclosure')
831 for e in enclosure_nodes:
832 next_url = e.attrib.get('url')
842 'title': it.find('title').text,
848 'title': playlist_title,
849 'description': playlist_desc,
853 def _extract_camtasia(self, url, video_id, webpage):
854 """ Returns None if no camtasia video can be found. """
856 camtasia_cfg = self._search_regex(
857 r'fo\.addVariable\(\s*"csConfigFile",\s*"([^"]+)"\s*\);',
858 webpage, 'camtasia configuration file', default=None)
859 if camtasia_cfg is None:
862 title = self._html_search_meta('DC.title', webpage, fatal=True)
864 camtasia_url = compat_urlparse.urljoin(url, camtasia_cfg)
865 camtasia_cfg = self._download_xml(
866 camtasia_url, video_id,
867 note='Downloading camtasia configuration',
868 errnote='Failed to download camtasia configuration')
869 fileset_node = camtasia_cfg.find('./playlist/array/fileset')
872 for n in fileset_node.getchildren():
873 url_n = n.find('./uri')
878 'id': os.path.splitext(url_n.text.rpartition('/')[2])[0],
879 'title': '%s - %s' % (title, n.tag),
880 'url': compat_urlparse.urljoin(url, url_n.text),
881 'duration': float_or_none(n.find('./duration').text),
890 def _real_extract(self, url):
891 if url.startswith('//'):
894 'url': self.http_scheme() + url,
897 parsed_url = compat_urlparse.urlparse(url)
898 if not parsed_url.scheme:
899 default_search = self._downloader.params.get('default_search')
900 if default_search is None:
901 default_search = 'fixup_error'
903 if default_search in ('auto', 'auto_warning', 'fixup_error'):
905 self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http')
906 return self.url_result('http://' + url)
907 elif default_search != 'fixup_error':
908 if default_search == 'auto_warning':
909 if re.match(r'^(?:url|URL)$', url):
910 raise ExtractorError(
911 'Invalid URL: %r . Call youtube-dl like this: youtube-dl -v "https://www.youtube.com/watch?v=BaW_jenozKc" ' % url,
914 self._downloader.report_warning(
915 'Falling back to youtube search for %s . Set --default-search "auto" to suppress this warning.' % url)
916 return self.url_result('ytsearch:' + url)
918 if default_search in ('error', 'fixup_error'):
919 raise ExtractorError(
920 '%r is not a valid URL. '
921 'Set --default-search "ytsearch" (or run youtube-dl "ytsearch:%s" ) to search YouTube'
922 % (url, url), expected=True)
924 if ':' not in default_search:
925 default_search += ':'
926 return self.url_result(default_search + url)
928 url, smuggled_data = unsmuggle_url(url)
930 is_intentional = smuggled_data and smuggled_data.get('to_generic')
931 if smuggled_data and 'force_videoid' in smuggled_data:
932 force_videoid = smuggled_data['force_videoid']
933 video_id = force_videoid
935 video_id = compat_urllib_parse_unquote(os.path.splitext(url.rstrip('/').split('/')[-1])[0])
937 self.to_screen('%s: Requesting header' % video_id)
939 head_req = HEADRequest(url)
940 head_response = self._request_webpage(
942 note=False, errnote='Could not send HEAD request to %s' % url,
945 if head_response is not False:
947 new_url = head_response.geturl()
949 self.report_following_redirect(new_url)
951 new_url = smuggle_url(
952 new_url, {'force_videoid': force_videoid})
953 return self.url_result(new_url)
956 if head_response is False:
957 request = compat_urllib_request.Request(url)
958 request.add_header('Accept-Encoding', '*')
959 full_response = self._request_webpage(request, video_id)
960 head_response = full_response
962 # Check for direct link to a video
963 content_type = head_response.headers.get('Content-Type', '')
964 m = re.match(r'^(?P<type>audio|video|application(?=/ogg$))/(?P<format_id>.+)$', content_type)
966 upload_date = unified_strdate(
967 head_response.headers.get('Last-Modified'))
970 'title': compat_urllib_parse_unquote(os.path.splitext(url_basename(url))[0]),
973 'format_id': m.group('format_id'),
975 'vcodec': 'none' if m.group('type') == 'audio' else None
977 'upload_date': upload_date,
980 if (not self._downloader.params.get('test', False) and
981 not is_intentional and
982 not self._downloader.params.get('force_generic_extractor', False)):
983 self._downloader.report_warning('Falling back on generic information extractor.')
985 if not full_response:
986 request = compat_urllib_request.Request(url)
987 # Some webservers may serve compressed content of rather big size (e.g. gzipped flac)
988 # making it impossible to download only chunk of the file (yet we need only 512kB to
989 # test whether it's HTML or not). According to youtube-dl default Accept-Encoding
990 # that will always result in downloading the whole file that is not desirable.
991 # Therefore for extraction pass we have to override Accept-Encoding to any in order
992 # to accept raw bytes and being able to download only a chunk.
993 # It may probably better to solve this by checking Content-Type for application/octet-stream
994 # after HEAD request finishes, but not sure if we can rely on this.
995 request.add_header('Accept-Encoding', '*')
996 full_response = self._request_webpage(request, video_id)
998 # Maybe it's a direct link to a video?
999 # Be careful not to download the whole thing!
1000 first_bytes = full_response.read(512)
1001 if not is_html(first_bytes):
1002 self._downloader.report_warning(
1003 'URL could be a direct video link, returning it as such.')
1004 upload_date = unified_strdate(
1005 head_response.headers.get('Last-Modified'))
1008 'title': compat_urllib_parse_unquote(os.path.splitext(url_basename(url))[0]),
1011 'upload_date': upload_date,
1014 webpage = self._webpage_read_content(
1015 full_response, url, video_id, prefix=first_bytes)
1017 self.report_extraction(video_id)
1019 # Is it an RSS feed?
1021 doc = parse_xml(webpage)
1022 if doc.tag == 'rss':
1023 return self._extract_rss(url, video_id, doc)
1024 except compat_xml_parse_error:
1027 # Is it a Camtasia project?
1028 camtasia_res = self._extract_camtasia(url, video_id, webpage)
1029 if camtasia_res is not None:
1032 # Sometimes embedded video player is hidden behind percent encoding
1033 # (e.g. https://github.com/rg3/youtube-dl/issues/2448)
1034 # Unescaping the whole page allows to handle those cases in a generic way
1035 webpage = compat_urllib_parse.unquote(webpage)
1037 # it's tempting to parse this further, but you would
1038 # have to take into account all the variations like
1039 # Video Title - Site Name
1040 # Site Name | Video Title
1041 # Video Title - Tagline | Site Name
1042 # and so on and so forth; it's just not practical
1043 video_title = self._html_search_regex(
1044 r'(?s)<title>(.*?)</title>', webpage, 'video title',
1047 # Try to detect age limit automatically
1048 age_limit = self._rta_search(webpage)
1049 # And then there are the jokers who advertise that they use RTA,
1050 # but actually don't.
1051 AGE_LIMIT_MARKERS = [
1052 r'Proudly Labeled <a href="http://www.rtalabel.org/" title="Restricted to Adults">RTA</a>',
1054 if any(re.search(marker, webpage) for marker in AGE_LIMIT_MARKERS):
1057 # video uploader is domain name
1058 video_uploader = self._search_regex(
1059 r'^(?:https?://)?([^/]*)/.*', url, 'video uploader')
1062 def _playlist_from_matches(matches, getter=None, ie=None):
1064 self.url_result(self._proto_relative_url(getter(m) if getter else m), ie)
1066 return self.playlist_result(
1067 urlrs, playlist_id=video_id, playlist_title=video_title)
1069 # Look for BrightCove:
1070 bc_urls = BrightcoveIE._extract_brightcove_urls(webpage)
1072 self.to_screen('Brightcove video detected.')
1075 'url': smuggle_url(bc_url, {'Referer': url}),
1076 'ie_key': 'Brightcove'
1077 } for bc_url in bc_urls]
1080 '_type': 'playlist',
1081 'title': video_title,
1086 # Look for embedded rtl.nl player
1087 matches = re.findall(
1088 r'<iframe[^>]+?src="((?:https?:)?//(?:www\.)?rtl\.nl/system/videoplayer/[^"]+(?:video_)?embed[^"]+)"',
1091 return _playlist_from_matches(matches, ie='RtlNl')
1093 # Look for embedded (iframe) Vimeo player
1095 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//player\.vimeo\.com/video/.+?)\1', webpage)
1097 player_url = unescapeHTML(mobj.group('url'))
1098 surl = smuggle_url(player_url, {'Referer': url})
1099 return self.url_result(surl)
1100 # Look for embedded (swf embed) Vimeo player
1102 r'<embed[^>]+?src="((?:https?:)?//(?:www\.)?vimeo\.com/moogaloop\.swf.+?)"', webpage)
1104 return self.url_result(mobj.group(1))
1106 # Look for embedded YouTube player
1107 matches = re.findall(r'''(?x)
1116 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1120 return _playlist_from_matches(
1121 matches, lambda m: unescapeHTML(m[1]))
1123 # Look for lazyYT YouTube embed
1124 matches = re.findall(
1125 r'class="lazyYT" data-youtube-id="([^"]+)"', webpage)
1127 return _playlist_from_matches(matches, lambda m: unescapeHTML(m))
1129 # Look for embedded Dailymotion player
1130 matches = re.findall(
1131 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/embed/video/.+?)\1', webpage)
1133 return _playlist_from_matches(
1134 matches, lambda m: unescapeHTML(m[1]))
1136 # Look for embedded Dailymotion playlist player (#3822)
1138 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.[a-z]{2,3}/widget/jukebox\?.+?)\1', webpage)
1140 playlists = re.findall(
1141 r'list\[\]=/playlist/([^/]+)/', unescapeHTML(m.group('url')))
1143 return _playlist_from_matches(
1144 playlists, lambda p: '//dailymotion.com/playlist/%s' % p)
1146 # Look for embedded Wistia player
1148 r'<(?:meta[^>]+?content|iframe[^>]+?src)=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage)
1150 embed_url = self._proto_relative_url(
1151 unescapeHTML(match.group('url')))
1153 '_type': 'url_transparent',
1156 'uploader': video_uploader,
1157 'title': video_title,
1161 match = re.search(r'(?:id=["\']wistia_|data-wistia-?id=["\']|Wistia\.embed\(["\'])(?P<id>[^"\']+)', webpage)
1164 '_type': 'url_transparent',
1165 'url': 'http://fast.wistia.net/embed/iframe/{0:}'.format(match.group('id')),
1167 'uploader': video_uploader,
1168 'title': video_title,
1169 'id': match.group('id')
1172 # Look for embedded blip.tv player
1173 bliptv_url = BlipTVIE._extract_url(webpage)
1175 return self.url_result(bliptv_url, 'BlipTV')
1177 # Look for SVT player
1178 svt_url = SVTIE._extract_url(webpage)
1180 return self.url_result(svt_url, 'SVT')
1182 # Look for embedded condenast player
1183 matches = re.findall(
1184 r'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="(https?://player\.cnevids\.com/embed/[^"]+")',
1188 '_type': 'playlist',
1191 'ie_key': 'CondeNast',
1193 } for ma in matches],
1194 'title': video_title,
1198 # Look for Bandcamp pages with custom domain
1199 mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage)
1200 if mobj is not None:
1201 burl = unescapeHTML(mobj.group(1))
1202 # Don't set the extractor because it can be a track url or an album
1203 return self.url_result(burl)
1205 # Look for embedded Vevo player
1207 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:cache\.)?vevo\.com/.+?)\1', webpage)
1208 if mobj is not None:
1209 return self.url_result(mobj.group('url'))
1211 # Look for embedded Viddler player
1213 r'<(?:iframe[^>]+?src|param[^>]+?value)=(["\'])(?P<url>(?:https?:)?//(?:www\.)?viddler\.com/(?:embed|player)/.+?)\1',
1215 if mobj is not None:
1216 return self.url_result(mobj.group('url'))
1218 # Look for NYTimes player
1220 r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//graphics8\.nytimes\.com/bcvideo/[^/]+/iframe/embed\.html.+?)\1>',
1222 if mobj is not None:
1223 return self.url_result(mobj.group('url'))
1225 # Look for Libsyn player
1227 r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//html5-player\.libsyn\.com/embed/.+?)\1', webpage)
1228 if mobj is not None:
1229 return self.url_result(mobj.group('url'))
1231 # Look for Ooyala videos
1232 mobj = (re.search(r'player\.ooyala\.com/[^"?]+\?[^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or
1233 re.search(r'OO\.Player\.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage) or
1234 re.search(r'SBN\.VideoLinkset\.ooyala\([\'"](?P<ec>.{32})[\'"]\)', webpage) or
1235 re.search(r'data-ooyala-video-id\s*=\s*[\'"](?P<ec>.{32})[\'"]', webpage))
1236 if mobj is not None:
1237 return OoyalaIE._build_url_result(mobj.group('ec'))
1239 # Look for multiple Ooyala embeds on SBN network websites
1240 mobj = re.search(r'SBN\.VideoLinkset\.entryGroup\((\[.*?\])', webpage)
1241 if mobj is not None:
1242 embeds = self._parse_json(mobj.group(1), video_id, fatal=False)
1244 return _playlist_from_matches(
1245 embeds, getter=lambda v: OoyalaIE._url_for_embed_code(v['provider_video_id']), ie='Ooyala')
1247 # Look for Aparat videos
1248 mobj = re.search(r'<iframe .*?src="(http://www\.aparat\.com/video/[^"]+)"', webpage)
1249 if mobj is not None:
1250 return self.url_result(mobj.group(1), 'Aparat')
1252 # Look for MPORA videos
1253 mobj = re.search(r'<iframe .*?src="(http://mpora\.(?:com|de)/videos/[^"]+)"', webpage)
1254 if mobj is not None:
1255 return self.url_result(mobj.group(1), 'Mpora')
1257 # Look for embedded NovaMov-based player
1259 r'''(?x)<(?:pagespeed_)?iframe[^>]+?src=(["\'])
1260 (?P<url>http://(?:(?:embed|www)\.)?
1262 nowvideo\.(?:ch|sx|eu|at|ag|co)|
1263 videoweed\.(?:es|com)|
1264 movshare\.(?:net|sx|ag)|
1265 divxstage\.(?:eu|net|ch|co|at|ag))
1266 /embed\.php.+?)\1''', webpage)
1267 if mobj is not None:
1268 return self.url_result(mobj.group('url'))
1270 # Look for embedded Facebook player
1272 r'<iframe[^>]+?src=(["\'])(?P<url>https://www\.facebook\.com/video/embed.+?)\1', webpage)
1273 if mobj is not None:
1274 return self.url_result(mobj.group('url'), 'Facebook')
1276 # Look for embedded VK player
1277 mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://vk\.com/video_ext\.php.+?)\1', webpage)
1278 if mobj is not None:
1279 return self.url_result(mobj.group('url'), 'VK')
1281 # Look for embedded ivi player
1282 mobj = re.search(r'<embed[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?ivi\.ru/video/player.+?)\1', webpage)
1283 if mobj is not None:
1284 return self.url_result(mobj.group('url'), 'Ivi')
1286 # Look for embedded Huffington Post player
1288 r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed\.live\.huffingtonpost\.com/.+?)\1', webpage)
1289 if mobj is not None:
1290 return self.url_result(mobj.group('url'), 'HuffPost')
1293 mobj = re.search(r'class=["\']embedly-card["\'][^>]href=["\'](?P<url>[^"\']+)', webpage)
1294 if mobj is not None:
1295 return self.url_result(mobj.group('url'))
1296 mobj = re.search(r'class=["\']embedly-embed["\'][^>]src=["\'][^"\']*url=(?P<url>[^&]+)', webpage)
1297 if mobj is not None:
1298 return self.url_result(compat_urllib_parse.unquote(mobj.group('url')))
1300 # Look for funnyordie embed
1301 matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage)
1303 return _playlist_from_matches(
1304 matches, getter=unescapeHTML, ie='FunnyOrDie')
1306 # Look for BBC iPlayer embed
1307 matches = re.findall(r'setPlaylist\("(https?://www\.bbc\.co\.uk/iplayer/[^/]+/[\da-z]{8})"\)', webpage)
1309 return _playlist_from_matches(matches, ie='BBCCoUk')
1311 # Look for embedded RUTV player
1312 rutv_url = RUTVIE._extract_url(webpage)
1314 return self.url_result(rutv_url, 'RUTV')
1316 # Look for embedded TVC player
1317 tvc_url = TVCIE._extract_url(webpage)
1319 return self.url_result(tvc_url, 'TVC')
1321 # Look for embedded SportBox player
1322 sportbox_urls = SportBoxEmbedIE._extract_urls(webpage)
1324 return _playlist_from_matches(sportbox_urls, ie='SportBoxEmbed')
1326 # Look for embedded Tvigle player
1328 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//cloud\.tvigle\.ru/video/.+?)\1', webpage)
1329 if mobj is not None:
1330 return self.url_result(mobj.group('url'), 'Tvigle')
1332 # Look for embedded TED player
1334 r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed(?:-ssl)?\.ted\.com/.+?)\1', webpage)
1335 if mobj is not None:
1336 return self.url_result(mobj.group('url'), 'TED')
1338 # Look for embedded Ustream videos
1340 r'<iframe[^>]+?src=(["\'])(?P<url>http://www\.ustream\.tv/embed/.+?)\1', webpage)
1341 if mobj is not None:
1342 return self.url_result(mobj.group('url'), 'Ustream')
1344 # Look for embedded arte.tv player
1346 r'<script [^>]*?src="(?P<url>http://www\.arte\.tv/playerv2/embed[^"]+)"',
1348 if mobj is not None:
1349 return self.url_result(mobj.group('url'), 'ArteTVEmbed')
1351 # Look for embedded smotri.com player
1352 smotri_url = SmotriIE._extract_url(webpage)
1354 return self.url_result(smotri_url, 'Smotri')
1356 # Look for embeded soundcloud player
1358 r'<iframe\s+(?:[a-zA-Z0-9_-]+="[^"]+"\s+)*src="(?P<url>https?://(?:w\.)?soundcloud\.com/player[^"]+)"',
1360 if mobj is not None:
1361 url = unescapeHTML(mobj.group('url'))
1362 return self.url_result(url)
1364 # Look for embedded vulture.com player
1366 r'<iframe src="(?P<url>https?://video\.vulture\.com/[^"]+)"',
1368 if mobj is not None:
1369 url = unescapeHTML(mobj.group('url'))
1370 return self.url_result(url, ie='Vulture')
1372 # Look for embedded mtvservices player
1374 r'<iframe src="(?P<url>https?://media\.mtvnservices\.com/embed/[^"]+)"',
1376 if mobj is not None:
1377 url = unescapeHTML(mobj.group('url'))
1378 return self.url_result(url, ie='MTVServicesEmbedded')
1380 # Look for embedded yahoo player
1382 r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:screen|movies)\.yahoo\.com/.+?\.html\?format=embed)\1',
1384 if mobj is not None:
1385 return self.url_result(mobj.group('url'), 'Yahoo')
1387 # Look for embedded sbs.com.au player
1391 <meta\s+property="og:video"\s+content=|
1394 (["\'])(?P<url>https?://(?:www\.)?sbs\.com\.au/ondemand/video/.+?)\1''',
1396 if mobj is not None:
1397 return self.url_result(mobj.group('url'), 'SBS')
1399 # Look for embedded Cinchcast player
1401 r'<iframe[^>]+?src=(["\'])(?P<url>https?://player\.cinchcast\.com/.+?)\1',
1403 if mobj is not None:
1404 return self.url_result(mobj.group('url'), 'Cinchcast')
1407 r'<iframe[^>]+?src=(["\'])(?P<url>https?://m(?:lb)?\.mlb\.com/shared/video/embed/embed\.html\?.+?)\1',
1411 r'data-video-link=["\'](?P<url>http://m.mlb.com/video/[^"\']+)',
1413 if mobj is not None:
1414 return self.url_result(mobj.group('url'), 'MLB')
1417 r'<iframe[^>]+?src=(["\'])(?P<url>%s)\1' % CondeNastIE.EMBED_URL,
1419 if mobj is not None:
1420 return self.url_result(self._proto_relative_url(mobj.group('url'), scheme='http:'), 'CondeNast')
1423 r'<iframe[^>]+src="(?P<url>https?://new\.livestream\.com/[^"]+/player[^"]+)"',
1425 if mobj is not None:
1426 return self.url_result(mobj.group('url'), 'Livestream')
1428 # Look for Zapiks embed
1430 r'<iframe[^>]+src="(?P<url>https?://(?:www\.)?zapiks\.fr/index\.php\?.+?)"', webpage)
1431 if mobj is not None:
1432 return self.url_result(mobj.group('url'), 'Zapiks')
1434 # Look for Kaltura embeds
1436 r"(?s)kWidget\.(?:thumb)?[Ee]mbed\(\{.*?'wid'\s*:\s*'_?(?P<partner_id>[^']+)',.*?'entry_id'\s*:\s*'(?P<id>[^']+)',", webpage)
1437 if mobj is not None:
1438 return self.url_result('kaltura:%(partner_id)s:%(id)s' % mobj.groupdict(), 'Kaltura')
1440 # Look for Eagle.Platform embeds
1442 r'<iframe[^>]+src="(?P<url>https?://.+?\.media\.eagleplatform\.com/index/player\?.+?)"', webpage)
1443 if mobj is not None:
1444 return self.url_result(mobj.group('url'), 'EaglePlatform')
1446 # Look for ClipYou (uses Eagle.Platform) embeds
1448 r'<iframe[^>]+src="https?://(?P<host>media\.clipyou\.ru)/index/player\?.*\brecord_id=(?P<id>\d+).*"', webpage)
1449 if mobj is not None:
1450 return self.url_result('eagleplatform:%(host)s:%(id)s' % mobj.groupdict(), 'EaglePlatform')
1452 # Look for Pladform embeds
1454 r'<iframe[^>]+src="(?P<url>https?://out\.pladform\.ru/player\?.+?)"', webpage)
1455 if mobj is not None:
1456 return self.url_result(mobj.group('url'), 'Pladform')
1458 # Look for Playwire embeds
1460 r'<script[^>]+data-config=(["\'])(?P<url>(?:https?:)?//config\.playwire\.com/.+?)\1', webpage)
1461 if mobj is not None:
1462 return self.url_result(mobj.group('url'))
1464 # Look for 5min embeds
1466 r'<meta[^>]+property="og:video"[^>]+content="https?://embed\.5min\.com/(?P<id>[0-9]+)/?', webpage)
1467 if mobj is not None:
1468 return self.url_result('5min:%s' % mobj.group('id'), 'FiveMin')
1470 # Look for Crooks and Liars embeds
1472 r'<(?:iframe[^>]+src|param[^>]+value)=(["\'])(?P<url>(?:https?:)?//embed\.crooksandliars\.com/(?:embed|v)/.+?)\1', webpage)
1473 if mobj is not None:
1474 return self.url_result(mobj.group('url'))
1476 # Look for NBC Sports VPlayer embeds
1477 nbc_sports_url = NBCSportsVPlayerIE._extract_url(webpage)
1479 return self.url_result(nbc_sports_url, 'NBCSportsVPlayer')
1481 # Look for UDN embeds
1483 r'<iframe[^>]+src="(?P<url>%s)"' % UDNEmbedIE._VALID_URL, webpage)
1484 if mobj is not None:
1485 return self.url_result(
1486 compat_urlparse.urljoin(url, mobj.group('url')), 'UDNEmbed')
1488 # Look for Senate ISVP iframe
1489 senate_isvp_url = SenateISVPIE._search_iframe_url(webpage)
1491 return self.url_result(senate_isvp_url, 'SenateISVP')
1493 def check_video(vurl):
1494 if YoutubeIE.suitable(vurl):
1496 vpath = compat_urlparse.urlparse(vurl).path
1497 vext = determine_ext(vpath)
1498 return '.' in vpath and vext not in ('swf', 'png', 'jpg', 'srt', 'sbv', 'sub', 'vtt', 'ttml')
1500 def filter_video(urls):
1501 return list(filter(check_video, urls))
1503 # Start with something easy: JW Player in SWFObject
1504 found = filter_video(re.findall(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage))
1506 # Look for gorilla-vid style embedding
1507 found = filter_video(re.findall(r'''(?sx)
1511 jwplayer\s*\(\s*["'][^'"]+["']\s*\)\s*\.setup
1514 ['"]?file['"]?\s*:\s*["\'](.*?)["\']''', webpage))
1516 # Broaden the search a little bit
1517 found = filter_video(re.findall(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage))
1519 # Broaden the findall a little bit: JWPlayer JS loader
1520 found = filter_video(re.findall(
1521 r'[^A-Za-z0-9]?file["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage))
1524 found = filter_video(re.findall(r'''(?xs)
1525 flowplayer\("[^"]+",\s*
1527 \s*\{[^}]+? ["']?clip["']?\s*:\s*\{\s*
1528 ["']?url["']?\s*:\s*["']([^"']+)["']
1533 r"cinerama\.embedPlayer\(\s*\'[^']+\',\s*'([^']+)'", webpage)
1535 # Try to find twitter cards info
1536 found = filter_video(re.findall(
1537 r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage))
1539 # We look for Open Graph info:
1540 # We have to match any number spaces between elements, some sites try to align them (eg.: statigr.am)
1541 m_video_type = re.findall(r'<meta.*?property="og:video:type".*?content="video/(.*?)"', webpage)
1542 # We only look in og:video if the MIME type is a video, don't try if it's a Flash player:
1543 if m_video_type is not None:
1544 found = filter_video(re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage))
1547 found = re.findall(r'(?s)<video[^<]*(?:>.*?<source[^>]*)?\s+src=["\'](.*?)["\']', webpage)
1549 REDIRECT_REGEX = r'[0-9]{,2};\s*(?:URL|url)=\'?([^\'"]+)'
1551 r'(?i)<meta\s+(?=(?:[a-z-]+="[^"]+"\s+)*http-equiv="refresh")'
1552 r'(?:[a-z-]+="[^"]+"\s+)*?content="%s' % REDIRECT_REGEX,
1555 # Look also in Refresh HTTP header
1556 refresh_header = head_response.headers.get('Refresh')
1558 found = re.search(REDIRECT_REGEX, refresh_header)
1560 new_url = compat_urlparse.urljoin(url, found.group(1))
1561 self.report_following_redirect(new_url)
1567 raise UnsupportedError(url)
1570 for video_url in found:
1571 video_url = compat_urlparse.urljoin(url, video_url)
1572 video_id = compat_urllib_parse.unquote(os.path.basename(video_url))
1574 # Sometimes, jwplayer extraction will result in a YouTube URL
1575 if YoutubeIE.suitable(video_url):
1576 entries.append(self.url_result(video_url, 'Youtube'))
1579 # here's a fun little line of code for you:
1580 video_id = os.path.splitext(video_id)[0]
1582 if determine_ext(video_url) == 'smil':
1585 'formats': self._extract_smil_formats(video_url, video_id),
1586 'uploader': video_uploader,
1587 'title': video_title,
1588 'age_limit': age_limit,
1594 'uploader': video_uploader,
1595 'title': video_title,
1596 'age_limit': age_limit,
1599 if len(entries) == 1:
1602 for num, e in enumerate(entries, start=1):
1603 # 'url' results don't have a title
1604 if e.get('title') is not None:
1605 e['title'] = '%s (%d)' % (e['title'], num)
1607 '_type': 'playlist',