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
46 from .vimeo import VimeoIE
49 class GenericIE(InfoExtractor):
50 IE_DESC = 'Generic downloader that works on some sites'
54 # Direct link to a video
56 'url': 'http://media.w3.org/2010/05/sintel/trailer.mp4',
57 'md5': '67d406c2bcb6af27fa886f31aa934bbe',
62 'upload_date': '20100513',
65 # Direct link to media delivered compressed (until Accept-Encoding is *)
67 'url': 'http://calimero.tk/muzik/FictionJunction-Parallel_Hearts.flac',
68 'md5': '128c42e68b13950268b648275386fc74',
70 'id': 'FictionJunction-Parallel_Hearts',
72 'title': 'FictionJunction-Parallel_Hearts',
73 'upload_date': '20140522',
75 'expected_warnings': [
76 'URL could be a direct video link, returning it as such.'
79 # Direct download with broken HEAD
81 'url': 'http://ai-radio.org:8000/radio.opus',
88 'skip_download': True, # infinite live stream
90 'expected_warnings': [
91 r'501.*Not Implemented'
94 # Direct link with incorrect MIME type
96 'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
97 'md5': '4ccbebe5f36706d85221f204d7eb5913',
99 'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
100 'id': '5_Lennart_Poettering_-_Systemd',
102 'title': '5_Lennart_Poettering_-_Systemd',
103 'upload_date': '20141120',
105 'expected_warnings': [
106 'URL could be a direct video link, returning it as such.'
111 'url': 'http://phihag.de/2014/youtube-dl/rss2.xml',
113 'id': 'http://phihag.de/2014/youtube-dl/rss2.xml',
114 'title': 'Zero Punctuation',
115 'description': 're:.*groundbreaking video review series.*'
117 'playlist_mincount': 11,
119 # RSS feed with enclosure
121 'url': 'http://podcastfeeds.nbcnews.com/audio/podcast/MSNBC-MADDOW-NETCAST-M4V.xml',
123 'id': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
125 'upload_date': '20150228',
126 'title': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
131 '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',
135 'upload_date': '20130224',
136 'uploader_id': 'TheVerge',
137 'description': 're:^Chris Ziegler takes a look at the\.*',
138 'uploader': 'The Verge',
139 'title': 'First Firefox OS phones side-by-side',
142 'skip_download': False,
146 'url': 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
147 'md5': '85b90ccc9d73b4acd9138d3af4c27f89',
149 'id': '13601338388002',
151 'uploader': 'www.hodiho.fr',
152 'title': 'R\u00e9gis plante sa Jeep',
155 # bandcamp page with custom domain
157 'add_ie': ['Bandcamp'],
158 'url': 'http://bronyrock.com/track/the-pony-mash',
162 'title': 'The Pony Mash',
163 'uploader': 'M_Pallante',
165 'skip': 'There is a limit of 200 free downloads / month for the test song',
167 # embedded brightcove video
168 # it also tests brightcove videos that need to set the 'Referer' in the
171 'add_ie': ['Brightcove'],
172 'url': 'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/',
174 'id': '2765128793001',
176 'title': 'Le cours de bourse : l’analyse technique',
177 'description': 'md5:7e9ad046e968cb2d1114004aba466fd9',
178 'uploader': 'BFM BUSINESS',
181 'skip_download': True,
185 # https://github.com/rg3/youtube-dl/issues/2253
186 'url': 'http://bcove.me/i6nfkrc3',
187 'md5': '0ba9446db037002366bab3b3eb30c88c',
189 'id': '3101154703001',
191 'title': 'Still no power',
192 'uploader': 'thestar.com',
193 '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.',
195 'add_ie': ['Brightcove'],
198 'url': 'http://www.championat.com/video/football/v/87/87499.html',
199 'md5': 'fb973ecf6e4a78a67453647444222983',
201 'id': '3414141473001',
203 'title': 'Видео. Удаление Дзагоева (ЦСКА)',
204 'description': 'Онлайн-трансляция матча ЦСКА - "Волга"',
205 'uploader': 'Championat',
209 # https://github.com/rg3/youtube-dl/issues/3541
210 'add_ie': ['Brightcove'],
211 'url': 'http://www.kijk.nl/sbs6/leermijvrouwenkennen/videos/jqMiXKAYan2S/aflevering-1',
213 'id': '3866516442001',
215 'title': 'Leer mij vrouwen kennen: Aflevering 1',
216 'description': 'Leer mij vrouwen kennen: Aflevering 1',
217 'uploader': 'SBS Broadcasting',
219 'skip': 'Restricted to Netherlands',
221 'skip_download': True, # m3u8 download
226 'url': 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219',
227 'md5': '166dd577b433b4d4ebfee10b0824d8ff',
229 'id': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ',
231 'title': '2cc213299525360.mov', # that's what we get
233 'add_ie': ['Ooyala'],
235 # multiple ooyala embeds on SBN network websites
237 'url': 'http://www.sbnation.com/college-football-recruiting/2015/2/3/7970291/national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
239 'id': 'national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
240 'title': '25 lies you will tell yourself on National Signing Day - SBNation.com',
242 'playlist_mincount': 3,
244 'skip_download': True,
246 'add_ie': ['Ooyala'],
250 'url': 'http://www.tested.com/science/weird/460206-tested-grinding-coffee-2000-frames-second/',
254 'title': 'Tested: Grinding Coffee at 2000 Frames Per Second',
255 'upload_date': '20140225',
256 'description': 'md5:06a40fbf30b220468f1e0957c0f558ff',
257 'uploader': 'Tested',
258 'uploader_id': 'testedcom',
260 # No need to test YoutubeIE here
262 'skip_download': True,
267 'url': 'http://www.theguardian.com/world/2014/mar/11/obama-zach-galifianakis-between-two-ferns',
271 'title': 'Between Two Ferns with Zach Galifianakis: President Barack Obama',
272 'description': 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.',
277 'url': 'http://www.bbc.co.uk/blogs/adamcurtis/posts/BUGGER',
279 'title': 'BBC - Blogs - Adam Curtis - BUGGER',
281 'playlist_mincount': 18,
285 'url': 'http://www.rg.ru/2014/03/15/reg-dfo/anklav-anons.html',
289 'title': 'Охотское море стало целиком российским',
290 'description': 'md5:5ed62483b14663e2a95ebbe115eb8f43',
294 'skip_download': True,
299 '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/',
303 'title': 'Дошкольное воспитание',
308 'url': 'http://www.vestifinance.ru/articles/25753',
311 'title': 'Вести Экономика ― Прямые трансляции с Форума-выставки "Госзаказ-2013"',
316 'title': 'Госзаказ. День 3',
322 'title': 'Госзаказ. День 2',
328 'title': 'Госзаказ. День 1',
334 'skip_download': True,
339 'url': 'http://en.support.wordpress.com/videos/ted-talks/',
340 'md5': '65fdff94098e4a607385a60c5177c638',
344 'title': 'Hidden miracles of the natural world',
345 'uploader': 'Louie Schwartzberg',
346 'description': 'md5:8145d19d320ff3e52f28401f4c4283b9',
349 # Embeded Ustream video
351 'url': 'http://www.american.edu/spa/pti/nsa-privacy-janus-2014.cfm',
352 'md5': '27b99cdb639c9b12a79bca876a073417',
356 'uploader': 'AU SPA: The NSA and Privacy',
357 'title': 'NSA and Privacy Forum Debate featuring General Hayden and Barton Gellman'
360 # nowvideo embed hidden behind percent encoding
362 'url': 'http://www.waoanime.tv/the-super-dimension-fortress-macross-episode-1/',
363 'md5': '2baf4ddd70f697d94b1c18cf796d5107',
365 'id': '06e53103ca9aa',
367 'title': 'Macross Episode 001 Watch Macross Episode 001 onl',
368 'description': 'No description',
373 'url': 'http://www.tv-replay.fr/redirection/20-03-14/x-enius-arte-10753389.html',
374 'md5': '7653032cbb25bf6c80d80f217055fa43',
376 'id': '048195-004_PLUS7-F',
379 'description': 'md5:d5fdf32ef6613cdbfd516ae658abf168',
380 'upload_date': '20140320',
383 'skip_download': 'Requires rtmpdump'
388 'url': 'http://www.wired.com/2014/04/honda-asimo/',
389 'md5': 'ba0dfe966fa007657bd1443ee672db0f',
391 'id': '53501be369702d3275860000',
393 'title': 'Honda’s New Asimo Robot Is More Human Than Ever',
398 'url': 'http://www.spi0n.com/zap-spi0n-com-n216/',
399 'md5': '441aeeb82eb72c422c7f14ec533999cd',
401 'id': 'k2mm4bCdJ6CQ2i7c8o2',
403 'title': 'Le Zap de Spi0n n°216 - Zapping du Web',
406 'add_ie': ['Dailymotion'],
410 'url': 'http://www.badzine.de/ansicht/datum/2014/06/09/so-funktioniert-die-neue-englische-badminton-liga.html',
414 'title': 'The NBL Auction 2014',
415 'uploader': 'BADMINTON England',
416 'uploader_id': 'BADMINTONEvents',
417 'upload_date': '20140603',
418 'description': 'md5:9ef128a69f1e262a700ed83edb163a73',
420 'add_ie': ['Youtube'],
422 'skip_download': True,
427 'url': 'http://www.gametrailers.com/news-post/76093/north-america-europe-is-getting-that-mario-kart-8-mercedes-dlc-too',
428 'md5': '35727f82f58c76d996fc188f9755b0d5',
430 'id': '0306a69b-8adf-4fb5-aace-75f8e8cbfca9',
433 'description': 'Mario\'s life in the fast lane has never looked so good.',
436 # YouTube embed via <data-embed-url="">
438 'url': 'https://play.google.com/store/apps/details?id=com.gameloft.android.ANMP.GloftA8HM',
442 'title': 'Asphalt 8: Airborne - Update - Welcome to Dubai!',
443 'uploader': 'Gameloft',
444 'uploader_id': 'gameloft',
445 'upload_date': '20140828',
446 'description': 'md5:c80da9ed3d83ae6d1876c834de03e1c4',
449 'skip_download': True,
454 'url': 'http://www.ll.mit.edu/workshops/education/videocourses/antennas/lecture1/video/',
456 'md5': '0c5e352edabf715d762b0ad4e6d9ee67',
458 'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
459 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - video1',
464 'md5': '10e4bb3aaca9fd630e273ff92d9f3c63',
466 'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final_PIP',
467 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - pip',
473 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
478 'url': 'http://www.handjobhub.com/video/busty-blonde-siri-tit-fuck-while-wank-6313.html',
479 'md5': '9d65602bf31c6e20014319c7d07fba27',
481 'id': '5123ea6d5e5a7',
484 'uploader': 'www.handjobhub.com',
485 'title': 'Busty Blonde Siri Tit Fuck While Wank at HandjobHub.com',
488 # Multiple brightcove videos
489 # https://github.com/rg3/youtube-dl/issues/2283
491 'url': 'http://www.newyorker.com/online/blogs/newsdesk/2014/01/always-never-nuclear-command-and-control.html',
493 'id': 'always-never',
494 'title': 'Always / Never - The New Yorker',
498 'extract_flat': False,
499 'skip_download': True,
504 'url': 'http://umpire-empire.com/index.php/topic/58125-laz-decides-no-thats-low/',
505 'md5': '96f09a37e44da40dd083e12d9a683327',
509 'title': 'Ump changes call to ball',
510 'description': 'md5:71c11215384298a172a6dcb4c2e20685',
512 'timestamp': 1401537900,
513 'upload_date': '20140531',
514 'thumbnail': 're:^https?://.*\.jpg$',
519 'url': 'http://education-portal.com/academy/lesson/north-american-exploration-failed-colonies-of-spain-france-england.html#lesson',
520 'md5': '8788b683c777a5cf25621eaf286d0c23',
524 'title': 'md5:51364a8d3d009997ba99656004b5e20d',
526 'filesize': 182808282,
527 'uploader': 'education-portal.com',
531 'url': 'http://thoughtworks.wistia.com/medias/uxjb0lwrcz',
532 'md5': 'baf49c2baa8a7de5f3fc145a8506dcd4',
536 'title': 'Conversation about Hexagonal Rails Part 1 - ThoughtWorks',
538 'uploader': 'thoughtworks.wistia.com',
543 'url': 'http://nakedsecurity.sophos.com/2014/10/29/sscc-171-are-you-sure-that-1234-is-a-bad-password-podcast/',
547 'description': 'md5:ff867d6b555488ad3c52572bb33d432c',
548 'uploader': 'Sophos Security',
549 'title': 'Chet Chat 171 - Oct 29, 2014',
550 'upload_date': '20141029',
555 'url': 'http://www.esa.int/Our_Activities/Space_Science/Rosetta/Philae_comet_touch-down_webcast',
559 'upload_date': '20141112',
560 'title': 'Rosetta #CometLanding webcast HL 10',
565 'url': 'http://discourse.ubuntu.com/t/unity-8-desktop-mode-windows-on-mir/1986',
568 'title': 'Unity 8 desktop-mode windows on Mir! - Ubuntu Discourse',
570 'playlist_mincount': 2,
574 'url': 'http://undergroundwellness.com/podcasts/306-5-steps-to-permanent-gut-healing/',
578 'upload_date': '20141126',
579 'title': 'Jack Tips: 5 Steps to Permanent Gut Healing',
584 'url': 'http://www.abc.net.au/7.30/content/2015/s4164797.htm',
586 'id': '730m_DandD_1901_512k',
588 'uploader': 'www.abc.net.au',
589 'title': 'Game of Thrones with dice - Dungeons and Dragons fantasy role-playing game gets new life - 19/01/2015',
592 # embedded viddler video
594 'url': 'http://deadspin.com/i-cant-stop-watching-john-wall-chop-the-nuggets-with-th-1681801597',
598 'uploader': 'deadspin',
599 'title': 'WALL-TO-GORTAT',
600 'timestamp': 1422285291,
601 'upload_date': '20150126',
603 'add_ie': ['Viddler'],
607 'url': 'http://thedailyshow.cc.com/podcast/episodetwelve',
611 'title': "The Daily Show Podcast without Jon Stewart - Episode 12: Bassem Youssef: Egypt's Jon Stewart",
612 'description': 'md5:601cb790edd05908957dae8aaa866465',
613 'upload_date': '20150220',
618 'url': 'http://media.nationalarchives.gov.uk/index.php/webinar-using-discovery-national-archives-online-catalogue/',
622 'upload_date': '20150212',
623 'uploader': 'The National Archives UK',
624 'description': 'md5:a236581cd2449dd2df4f93412f3f01c6',
625 'uploader_id': 'NationalArchives08',
626 'title': 'Webinar: Using Discovery, The National Archives’ online catalogue',
631 'url': 'http://www.rtlnieuws.nl/nieuws/buitenland/aanslagen-kopenhagen',
632 'playlist_mincount': 5,
634 'id': 'aanslagen-kopenhagen',
635 'title': 'Aanslagen Kopenhagen | RTL Nieuws',
640 'url': 'http://www.skipass.com/news/116090-bon-appetit-s5ep3-baqueira-mi-cor.html',
644 'title': 'EP3S5 - Bon Appétit - Baqueira Mi Corazon !',
649 'url': 'http://www.monumentalnetwork.com/videos/john-carlson-postgame-2-25-15',
653 'upload_date': '20150226',
654 'uploader_id': 'MonumentalSports-Kaltura@perfectsensedigital.com',
656 'title': 'John Carlson Postgame 2/25/15',
659 # Eagle.Platform embed (generic URL)
661 'url': 'http://lenta.ru/news/2015/03/06/navalny/',
665 'title': 'Навальный вышел на свободу',
666 'description': 'md5:d97861ac9ae77377f3f20eaf9d04b4f5',
667 'thumbnail': 're:^https?://.*\.jpg$',
673 # ClipYou (Eagle.Platform) embed (custom URL)
675 'url': 'http://muz-tv.ru/play/7129/',
679 'title': "'O Sole Mio",
680 'thumbnail': 're:^https?://.*\.jpg$',
687 'url': 'http://muz-tv.ru/kinozal/view/7400/',
691 'title': 'Тайны перевала Дятлова • 1 серия 2 часть',
692 'description': 'Документальный сериал-расследование одной из самых жутких тайн ХХ века',
693 'thumbnail': 're:^https?://.*\.jpg$',
700 'url': 'http://www.cinemablend.com/new/First-Joe-Dirt-2-Trailer-Teaser-Stupid-Greatness-70874.html',
704 'title': 'Joe Dirt 2 Beautiful Loser Teaser Trailer',
705 'thumbnail': 're:^https?://.*\.png$',
711 'url': 'http://techcrunch.com/video/facebook-creates-on-this-day-crunch-report/518726732/',
712 'md5': '4c6f127a30736b59b3e2c19234ee2bf7',
716 'title': 'Facebook Creates "On This Day" | Crunch Report',
721 'url': 'http://www.svt.se/sport/ishockey/jagr-tacklar-giroux-under-intervjun',
725 'title': 'Här trycker Jagr till Giroux (under SVT-intervjun)',
730 # Crooks and Liars embed
732 'url': 'http://crooksandliars.com/2015/04/fox-friends-says-protecting-atheists',
736 'title': "Fox & Friends Says Protecting Atheists From Discrimination Is Anti-Christian!",
737 'description': 'md5:e1a46ad1650e3a5ec7196d432799127f',
738 'timestamp': 1428207000,
739 'upload_date': '20150405',
740 'uploader': 'Heather',
743 # Crooks and Liars external embed
745 'url': 'http://theothermccain.com/2010/02/02/video-proves-that-bill-kristol-has-been-watching-glenn-beck/comment-page-1/',
747 'id': 'MTE3MjUtMzQ2MzA',
749 'title': 'md5:5e3662a81a4014d24c250d76d41a08d5',
750 'description': 'md5:9b8e9542d6c3c5de42d6451b7d780cec',
751 'timestamp': 1265032391,
752 'upload_date': '20100201',
753 'uploader': 'Heather',
756 # NBC Sports vplayer embed
758 'url': 'http://www.riderfans.com/forum/showthread.php?121827-Freeman&s=e98fa1ea6dc08e886b1678d35212494a',
760 'id': 'ln7x1qSThw4k',
762 'title': "PFT Live: New leader in the 'new-look' defense",
763 'description': 'md5:65a19b4bbfb3b0c0c5768bed1dfad74e',
768 'url': 'http://www.udn.com/news/story/7314/822787',
769 'md5': 'fd2060e988c326991037b9aff9df21a6',
773 'title': '中一中男師變性 全校師生力挺',
774 'thumbnail': 're:^https?://.*\.jpg$',
779 'url': 'http://www.businessinsider.com/excel-index-match-vlookup-video-how-to-2015-2?IR=T',
781 'id': '50YnY4czr4ms1vJ7yz3xzq0excz_pUMs',
783 'description': 'VIDEO: Index/Match versus VLOOKUP.',
784 'title': 'This is what separates the Excel masters from the wannabes',
788 'skip_download': True,
791 # Contains a SMIL manifest
793 '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',
797 'title': '+ Football: Lottery Champions League Europe',
798 'uploader': 'www.telewebion.com',
802 'skip_download': True,
805 # Brightcove URL in single quotes
807 'url': 'http://www.sportsnet.ca/baseball/mlb/sn-presents-russell-martin-world-citizen/',
808 'md5': '4ae374f1f8b91c889c4b9203c8c752af',
810 'id': '4255764656001',
812 'title': 'SN Presents: Russell Martin, World Citizen',
813 '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.',
814 'uploader': 'Rogers Sportsnet',
819 def report_following_redirect(self, new_url):
820 """Report information extraction."""
821 self._downloader.to_screen('[redirect] Following redirect to %s' % new_url)
823 def _extract_rss(self, url, video_id, doc):
824 playlist_title = doc.find('./channel/title').text
825 playlist_desc_el = doc.find('./channel/description')
826 playlist_desc = None if playlist_desc_el is None else playlist_desc_el.text
829 for it in doc.findall('./channel/item'):
830 next_url = xpath_text(it, 'link', fatal=False)
832 enclosure_nodes = it.findall('./enclosure')
833 for e in enclosure_nodes:
834 next_url = e.attrib.get('url')
844 'title': it.find('title').text,
850 'title': playlist_title,
851 'description': playlist_desc,
855 def _extract_camtasia(self, url, video_id, webpage):
856 """ Returns None if no camtasia video can be found. """
858 camtasia_cfg = self._search_regex(
859 r'fo\.addVariable\(\s*"csConfigFile",\s*"([^"]+)"\s*\);',
860 webpage, 'camtasia configuration file', default=None)
861 if camtasia_cfg is None:
864 title = self._html_search_meta('DC.title', webpage, fatal=True)
866 camtasia_url = compat_urlparse.urljoin(url, camtasia_cfg)
867 camtasia_cfg = self._download_xml(
868 camtasia_url, video_id,
869 note='Downloading camtasia configuration',
870 errnote='Failed to download camtasia configuration')
871 fileset_node = camtasia_cfg.find('./playlist/array/fileset')
874 for n in fileset_node.getchildren():
875 url_n = n.find('./uri')
880 'id': os.path.splitext(url_n.text.rpartition('/')[2])[0],
881 'title': '%s - %s' % (title, n.tag),
882 'url': compat_urlparse.urljoin(url, url_n.text),
883 'duration': float_or_none(n.find('./duration').text),
892 def _real_extract(self, url):
893 if url.startswith('//'):
896 'url': self.http_scheme() + url,
899 parsed_url = compat_urlparse.urlparse(url)
900 if not parsed_url.scheme:
901 default_search = self._downloader.params.get('default_search')
902 if default_search is None:
903 default_search = 'fixup_error'
905 if default_search in ('auto', 'auto_warning', 'fixup_error'):
907 self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http')
908 return self.url_result('http://' + url)
909 elif default_search != 'fixup_error':
910 if default_search == 'auto_warning':
911 if re.match(r'^(?:url|URL)$', url):
912 raise ExtractorError(
913 'Invalid URL: %r . Call youtube-dl like this: youtube-dl -v "https://www.youtube.com/watch?v=BaW_jenozKc" ' % url,
916 self._downloader.report_warning(
917 'Falling back to youtube search for %s . Set --default-search "auto" to suppress this warning.' % url)
918 return self.url_result('ytsearch:' + url)
920 if default_search in ('error', 'fixup_error'):
921 raise ExtractorError(
922 '%r is not a valid URL. '
923 'Set --default-search "ytsearch" (or run youtube-dl "ytsearch:%s" ) to search YouTube'
924 % (url, url), expected=True)
926 if ':' not in default_search:
927 default_search += ':'
928 return self.url_result(default_search + url)
930 url, smuggled_data = unsmuggle_url(url)
932 is_intentional = smuggled_data and smuggled_data.get('to_generic')
933 if smuggled_data and 'force_videoid' in smuggled_data:
934 force_videoid = smuggled_data['force_videoid']
935 video_id = force_videoid
937 video_id = compat_urllib_parse_unquote(os.path.splitext(url.rstrip('/').split('/')[-1])[0])
939 self.to_screen('%s: Requesting header' % video_id)
941 head_req = HEADRequest(url)
942 head_response = self._request_webpage(
944 note=False, errnote='Could not send HEAD request to %s' % url,
947 if head_response is not False:
949 new_url = head_response.geturl()
951 self.report_following_redirect(new_url)
953 new_url = smuggle_url(
954 new_url, {'force_videoid': force_videoid})
955 return self.url_result(new_url)
958 if head_response is False:
959 request = compat_urllib_request.Request(url)
960 request.add_header('Accept-Encoding', '*')
961 full_response = self._request_webpage(request, video_id)
962 head_response = full_response
964 # Check for direct link to a video
965 content_type = head_response.headers.get('Content-Type', '')
966 m = re.match(r'^(?P<type>audio|video|application(?=/ogg$))/(?P<format_id>.+)$', content_type)
968 upload_date = unified_strdate(
969 head_response.headers.get('Last-Modified'))
972 'title': compat_urllib_parse_unquote(os.path.splitext(url_basename(url))[0]),
975 'format_id': m.group('format_id'),
977 'vcodec': 'none' if m.group('type') == 'audio' else None
979 'upload_date': upload_date,
982 if not self._downloader.params.get('test', False) and not is_intentional:
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 vimeo_url = VimeoIE._extract_vimeo_url(url, webpage)
1094 if vimeo_url is not None:
1095 return self.url_result(vimeo_url)
1097 # Look for embedded YouTube player
1098 matches = re.findall(r'''(?x)
1107 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1111 return _playlist_from_matches(
1112 matches, lambda m: unescapeHTML(m[1]))
1114 # Look for lazyYT YouTube embed
1115 matches = re.findall(
1116 r'class="lazyYT" data-youtube-id="([^"]+)"', webpage)
1118 return _playlist_from_matches(matches, lambda m: unescapeHTML(m))
1120 # Look for embedded Dailymotion player
1121 matches = re.findall(
1122 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/embed/video/.+?)\1', webpage)
1124 return _playlist_from_matches(
1125 matches, lambda m: unescapeHTML(m[1]))
1127 # Look for embedded Dailymotion playlist player (#3822)
1129 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.[a-z]{2,3}/widget/jukebox\?.+?)\1', webpage)
1131 playlists = re.findall(
1132 r'list\[\]=/playlist/([^/]+)/', unescapeHTML(m.group('url')))
1134 return _playlist_from_matches(
1135 playlists, lambda p: '//dailymotion.com/playlist/%s' % p)
1137 # Look for embedded Wistia player
1139 r'<(?:meta[^>]+?content|iframe[^>]+?src)=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage)
1141 embed_url = self._proto_relative_url(
1142 unescapeHTML(match.group('url')))
1144 '_type': 'url_transparent',
1147 'uploader': video_uploader,
1148 'title': video_title,
1152 match = re.search(r'(?:id=["\']wistia_|data-wistia-?id=["\']|Wistia\.embed\(["\'])(?P<id>[^"\']+)', webpage)
1155 '_type': 'url_transparent',
1156 'url': 'http://fast.wistia.net/embed/iframe/{0:}'.format(match.group('id')),
1158 'uploader': video_uploader,
1159 'title': video_title,
1160 'id': match.group('id')
1163 # Look for embedded blip.tv player
1164 bliptv_url = BlipTVIE._extract_url(webpage)
1166 return self.url_result(bliptv_url, 'BlipTV')
1168 # Look for SVT player
1169 svt_url = SVTIE._extract_url(webpage)
1171 return self.url_result(svt_url, 'SVT')
1173 # Look for embedded condenast player
1174 matches = re.findall(
1175 r'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="(https?://player\.cnevids\.com/embed/[^"]+")',
1179 '_type': 'playlist',
1182 'ie_key': 'CondeNast',
1184 } for ma in matches],
1185 'title': video_title,
1189 # Look for Bandcamp pages with custom domain
1190 mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage)
1191 if mobj is not None:
1192 burl = unescapeHTML(mobj.group(1))
1193 # Don't set the extractor because it can be a track url or an album
1194 return self.url_result(burl)
1196 # Look for embedded Vevo player
1198 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:cache\.)?vevo\.com/.+?)\1', webpage)
1199 if mobj is not None:
1200 return self.url_result(mobj.group('url'))
1202 # Look for embedded Viddler player
1204 r'<(?:iframe[^>]+?src|param[^>]+?value)=(["\'])(?P<url>(?:https?:)?//(?:www\.)?viddler\.com/(?:embed|player)/.+?)\1',
1206 if mobj is not None:
1207 return self.url_result(mobj.group('url'))
1209 # Look for NYTimes player
1211 r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//graphics8\.nytimes\.com/bcvideo/[^/]+/iframe/embed\.html.+?)\1>',
1213 if mobj is not None:
1214 return self.url_result(mobj.group('url'))
1216 # Look for Libsyn player
1218 r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//html5-player\.libsyn\.com/embed/.+?)\1', webpage)
1219 if mobj is not None:
1220 return self.url_result(mobj.group('url'))
1222 # Look for Ooyala videos
1223 mobj = (re.search(r'player\.ooyala\.com/[^"?]+\?[^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or
1224 re.search(r'OO\.Player\.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage) or
1225 re.search(r'SBN\.VideoLinkset\.ooyala\([\'"](?P<ec>.{32})[\'"]\)', webpage) or
1226 re.search(r'data-ooyala-video-id\s*=\s*[\'"](?P<ec>.{32})[\'"]', webpage))
1227 if mobj is not None:
1228 return OoyalaIE._build_url_result(mobj.group('ec'))
1230 # Look for multiple Ooyala embeds on SBN network websites
1231 mobj = re.search(r'SBN\.VideoLinkset\.entryGroup\((\[.*?\])', webpage)
1232 if mobj is not None:
1233 embeds = self._parse_json(mobj.group(1), video_id, fatal=False)
1235 return _playlist_from_matches(
1236 embeds, getter=lambda v: OoyalaIE._url_for_embed_code(v['provider_video_id']), ie='Ooyala')
1238 # Look for Aparat videos
1239 mobj = re.search(r'<iframe .*?src="(http://www\.aparat\.com/video/[^"]+)"', webpage)
1240 if mobj is not None:
1241 return self.url_result(mobj.group(1), 'Aparat')
1243 # Look for MPORA videos
1244 mobj = re.search(r'<iframe .*?src="(http://mpora\.(?:com|de)/videos/[^"]+)"', webpage)
1245 if mobj is not None:
1246 return self.url_result(mobj.group(1), 'Mpora')
1248 # Look for embedded NovaMov-based player
1250 r'''(?x)<(?:pagespeed_)?iframe[^>]+?src=(["\'])
1251 (?P<url>http://(?:(?:embed|www)\.)?
1253 nowvideo\.(?:ch|sx|eu|at|ag|co)|
1254 videoweed\.(?:es|com)|
1255 movshare\.(?:net|sx|ag)|
1256 divxstage\.(?:eu|net|ch|co|at|ag))
1257 /embed\.php.+?)\1''', webpage)
1258 if mobj is not None:
1259 return self.url_result(mobj.group('url'))
1261 # Look for embedded Facebook player
1263 r'<iframe[^>]+?src=(["\'])(?P<url>https://www\.facebook\.com/video/embed.+?)\1', webpage)
1264 if mobj is not None:
1265 return self.url_result(mobj.group('url'), 'Facebook')
1267 # Look for embedded VK player
1268 mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://vk\.com/video_ext\.php.+?)\1', webpage)
1269 if mobj is not None:
1270 return self.url_result(mobj.group('url'), 'VK')
1272 # Look for embedded ivi player
1273 mobj = re.search(r'<embed[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?ivi\.ru/video/player.+?)\1', webpage)
1274 if mobj is not None:
1275 return self.url_result(mobj.group('url'), 'Ivi')
1277 # Look for embedded Huffington Post player
1279 r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed\.live\.huffingtonpost\.com/.+?)\1', webpage)
1280 if mobj is not None:
1281 return self.url_result(mobj.group('url'), 'HuffPost')
1284 mobj = re.search(r'class=["\']embedly-card["\'][^>]href=["\'](?P<url>[^"\']+)', webpage)
1285 if mobj is not None:
1286 return self.url_result(mobj.group('url'))
1287 mobj = re.search(r'class=["\']embedly-embed["\'][^>]src=["\'][^"\']*url=(?P<url>[^&]+)', webpage)
1288 if mobj is not None:
1289 return self.url_result(compat_urllib_parse.unquote(mobj.group('url')))
1291 # Look for funnyordie embed
1292 matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage)
1294 return _playlist_from_matches(
1295 matches, getter=unescapeHTML, ie='FunnyOrDie')
1297 # Look for BBC iPlayer embed
1298 matches = re.findall(r'setPlaylist\("(https?://www\.bbc\.co\.uk/iplayer/[^/]+/[\da-z]{8})"\)', webpage)
1300 return _playlist_from_matches(matches, ie='BBCCoUk')
1302 # Look for embedded RUTV player
1303 rutv_url = RUTVIE._extract_url(webpage)
1305 return self.url_result(rutv_url, 'RUTV')
1307 # Look for embedded TVC player
1308 tvc_url = TVCIE._extract_url(webpage)
1310 return self.url_result(tvc_url, 'TVC')
1312 # Look for embedded SportBox player
1313 sportbox_urls = SportBoxEmbedIE._extract_urls(webpage)
1315 return _playlist_from_matches(sportbox_urls, ie='SportBoxEmbed')
1317 # Look for embedded PornHub player
1318 pornhub_url = PornHubIE._extract_url(webpage)
1320 return self.url_result(pornhub_url, 'PornHub')
1322 # Look for embedded Tvigle player
1324 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//cloud\.tvigle\.ru/video/.+?)\1', webpage)
1325 if mobj is not None:
1326 return self.url_result(mobj.group('url'), 'Tvigle')
1328 # Look for embedded TED player
1330 r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed(?:-ssl)?\.ted\.com/.+?)\1', webpage)
1331 if mobj is not None:
1332 return self.url_result(mobj.group('url'), 'TED')
1334 # Look for embedded Ustream videos
1336 r'<iframe[^>]+?src=(["\'])(?P<url>http://www\.ustream\.tv/embed/.+?)\1', webpage)
1337 if mobj is not None:
1338 return self.url_result(mobj.group('url'), 'Ustream')
1340 # Look for embedded arte.tv player
1342 r'<script [^>]*?src="(?P<url>http://www\.arte\.tv/playerv2/embed[^"]+)"',
1344 if mobj is not None:
1345 return self.url_result(mobj.group('url'), 'ArteTVEmbed')
1347 # Look for embedded smotri.com player
1348 smotri_url = SmotriIE._extract_url(webpage)
1350 return self.url_result(smotri_url, 'Smotri')
1352 # Look for embeded soundcloud player
1354 r'<iframe\s+(?:[a-zA-Z0-9_-]+="[^"]+"\s+)*src="(?P<url>https?://(?:w\.)?soundcloud\.com/player[^"]+)"',
1356 if mobj is not None:
1357 url = unescapeHTML(mobj.group('url'))
1358 return self.url_result(url)
1360 # Look for embedded vulture.com player
1362 r'<iframe src="(?P<url>https?://video\.vulture\.com/[^"]+)"',
1364 if mobj is not None:
1365 url = unescapeHTML(mobj.group('url'))
1366 return self.url_result(url, ie='Vulture')
1368 # Look for embedded mtvservices player
1370 r'<iframe src="(?P<url>https?://media\.mtvnservices\.com/embed/[^"]+)"',
1372 if mobj is not None:
1373 url = unescapeHTML(mobj.group('url'))
1374 return self.url_result(url, ie='MTVServicesEmbedded')
1376 # Look for embedded yahoo player
1378 r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:screen|movies)\.yahoo\.com/.+?\.html\?format=embed)\1',
1380 if mobj is not None:
1381 return self.url_result(mobj.group('url'), 'Yahoo')
1383 # Look for embedded sbs.com.au player
1387 <meta\s+property="og:video"\s+content=|
1390 (["\'])(?P<url>https?://(?:www\.)?sbs\.com\.au/ondemand/video/.+?)\1''',
1392 if mobj is not None:
1393 return self.url_result(mobj.group('url'), 'SBS')
1395 # Look for embedded Cinchcast player
1397 r'<iframe[^>]+?src=(["\'])(?P<url>https?://player\.cinchcast\.com/.+?)\1',
1399 if mobj is not None:
1400 return self.url_result(mobj.group('url'), 'Cinchcast')
1403 r'<iframe[^>]+?src=(["\'])(?P<url>https?://m(?:lb)?\.mlb\.com/shared/video/embed/embed\.html\?.+?)\1',
1407 r'data-video-link=["\'](?P<url>http://m.mlb.com/video/[^"\']+)',
1409 if mobj is not None:
1410 return self.url_result(mobj.group('url'), 'MLB')
1413 r'<iframe[^>]+?src=(["\'])(?P<url>%s)\1' % CondeNastIE.EMBED_URL,
1415 if mobj is not None:
1416 return self.url_result(self._proto_relative_url(mobj.group('url'), scheme='http:'), 'CondeNast')
1419 r'<iframe[^>]+src="(?P<url>https?://new\.livestream\.com/[^"]+/player[^"]+)"',
1421 if mobj is not None:
1422 return self.url_result(mobj.group('url'), 'Livestream')
1424 # Look for Zapiks embed
1426 r'<iframe[^>]+src="(?P<url>https?://(?:www\.)?zapiks\.fr/index\.php\?.+?)"', webpage)
1427 if mobj is not None:
1428 return self.url_result(mobj.group('url'), 'Zapiks')
1430 # Look for Kaltura embeds
1432 r"(?s)kWidget\.(?:thumb)?[Ee]mbed\(\{.*?'wid'\s*:\s*'_?(?P<partner_id>[^']+)',.*?'entry_id'\s*:\s*'(?P<id>[^']+)',", webpage)
1433 if mobj is not None:
1434 return self.url_result('kaltura:%(partner_id)s:%(id)s' % mobj.groupdict(), 'Kaltura')
1436 # Look for Eagle.Platform embeds
1438 r'<iframe[^>]+src="(?P<url>https?://.+?\.media\.eagleplatform\.com/index/player\?.+?)"', webpage)
1439 if mobj is not None:
1440 return self.url_result(mobj.group('url'), 'EaglePlatform')
1442 # Look for ClipYou (uses Eagle.Platform) embeds
1444 r'<iframe[^>]+src="https?://(?P<host>media\.clipyou\.ru)/index/player\?.*\brecord_id=(?P<id>\d+).*"', webpage)
1445 if mobj is not None:
1446 return self.url_result('eagleplatform:%(host)s:%(id)s' % mobj.groupdict(), 'EaglePlatform')
1448 # Look for Pladform embeds
1450 r'<iframe[^>]+src="(?P<url>https?://out\.pladform\.ru/player\?.+?)"', webpage)
1451 if mobj is not None:
1452 return self.url_result(mobj.group('url'), 'Pladform')
1454 # Look for Playwire embeds
1456 r'<script[^>]+data-config=(["\'])(?P<url>(?:https?:)?//config\.playwire\.com/.+?)\1', webpage)
1457 if mobj is not None:
1458 return self.url_result(mobj.group('url'))
1460 # Look for 5min embeds
1462 r'<meta[^>]+property="og:video"[^>]+content="https?://embed\.5min\.com/(?P<id>[0-9]+)/?', webpage)
1463 if mobj is not None:
1464 return self.url_result('5min:%s' % mobj.group('id'), 'FiveMin')
1466 # Look for Crooks and Liars embeds
1468 r'<(?:iframe[^>]+src|param[^>]+value)=(["\'])(?P<url>(?:https?:)?//embed\.crooksandliars\.com/(?:embed|v)/.+?)\1', webpage)
1469 if mobj is not None:
1470 return self.url_result(mobj.group('url'))
1472 # Look for NBC Sports VPlayer embeds
1473 nbc_sports_url = NBCSportsVPlayerIE._extract_url(webpage)
1475 return self.url_result(nbc_sports_url, 'NBCSportsVPlayer')
1477 # Look for UDN embeds
1479 r'<iframe[^>]+src="(?P<url>%s)"' % UDNEmbedIE._VALID_URL, webpage)
1480 if mobj is not None:
1481 return self.url_result(
1482 compat_urlparse.urljoin(url, mobj.group('url')), 'UDNEmbed')
1484 # Look for Senate ISVP iframe
1485 senate_isvp_url = SenateISVPIE._search_iframe_url(webpage)
1487 return self.url_result(senate_isvp_url, 'SenateISVP')
1489 def check_video(vurl):
1490 if YoutubeIE.suitable(vurl):
1492 vpath = compat_urlparse.urlparse(vurl).path
1493 vext = determine_ext(vpath)
1494 return '.' in vpath and vext not in ('swf', 'png', 'jpg', 'srt', 'sbv', 'sub', 'vtt', 'ttml')
1496 def filter_video(urls):
1497 return list(filter(check_video, urls))
1499 # Start with something easy: JW Player in SWFObject
1500 found = filter_video(re.findall(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage))
1502 # Look for gorilla-vid style embedding
1503 found = filter_video(re.findall(r'''(?sx)
1507 jwplayer\s*\(\s*["'][^'"]+["']\s*\)\s*\.setup
1510 ['"]?file['"]?\s*:\s*["\'](.*?)["\']''', webpage))
1512 # Broaden the search a little bit
1513 found = filter_video(re.findall(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage))
1515 # Broaden the findall a little bit: JWPlayer JS loader
1516 found = filter_video(re.findall(
1517 r'[^A-Za-z0-9]?file["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage))
1520 found = filter_video(re.findall(r'''(?xs)
1521 flowplayer\("[^"]+",\s*
1523 \s*\{[^}]+? ["']?clip["']?\s*:\s*\{\s*
1524 ["']?url["']?\s*:\s*["']([^"']+)["']
1529 r"cinerama\.embedPlayer\(\s*\'[^']+\',\s*'([^']+)'", webpage)
1531 # Try to find twitter cards info
1532 found = filter_video(re.findall(
1533 r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage))
1535 # We look for Open Graph info:
1536 # We have to match any number spaces between elements, some sites try to align them (eg.: statigr.am)
1537 m_video_type = re.findall(r'<meta.*?property="og:video:type".*?content="video/(.*?)"', webpage)
1538 # We only look in og:video if the MIME type is a video, don't try if it's a Flash player:
1539 if m_video_type is not None:
1540 found = filter_video(re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage))
1543 found = re.findall(r'(?s)<video[^<]*(?:>.*?<source[^>]*)?\s+src=["\'](.*?)["\']', webpage)
1545 REDIRECT_REGEX = r'[0-9]{,2};\s*(?:URL|url)=\'?([^\'"]+)'
1547 r'(?i)<meta\s+(?=(?:[a-z-]+="[^"]+"\s+)*http-equiv="refresh")'
1548 r'(?:[a-z-]+="[^"]+"\s+)*?content="%s' % REDIRECT_REGEX,
1551 # Look also in Refresh HTTP header
1552 refresh_header = head_response.headers.get('Refresh')
1554 found = re.search(REDIRECT_REGEX, refresh_header)
1556 new_url = compat_urlparse.urljoin(url, found.group(1))
1557 self.report_following_redirect(new_url)
1563 raise UnsupportedError(url)
1566 for video_url in found:
1567 video_url = compat_urlparse.urljoin(url, video_url)
1568 video_id = compat_urllib_parse.unquote(os.path.basename(video_url))
1570 # Sometimes, jwplayer extraction will result in a YouTube URL
1571 if YoutubeIE.suitable(video_url):
1572 entries.append(self.url_result(video_url, 'Youtube'))
1575 # here's a fun little line of code for you:
1576 video_id = os.path.splitext(video_id)[0]
1578 if determine_ext(video_url) == 'smil':
1581 'formats': self._extract_smil_formats(video_url, video_id),
1582 'uploader': video_uploader,
1583 'title': video_title,
1584 'age_limit': age_limit,
1590 'uploader': video_uploader,
1591 'title': video_title,
1592 'age_limit': age_limit,
1595 if len(entries) == 1:
1598 for num, e in enumerate(entries, start=1):
1599 # 'url' results don't have a title
1600 if e.get('title') is not None:
1601 e['title'] = '%s (%d)' % (e['title'], num)
1603 '_type': 'playlist',