3 from __future__ import unicode_literals
8 from .common import InfoExtractor
9 from .youtube import YoutubeIE
10 from ..compat import (
13 compat_xml_parse_error,
31 from .brightcove import BrightcoveIE
32 from .nbc import NBCSportsVPlayerIE
33 from .ooyala import OoyalaIE
34 from .rutv import RUTVIE
35 from .smotri import SmotriIE
36 from .condenast import CondeNastIE
37 from .udn import UDNEmbedIE
40 class GenericIE(InfoExtractor):
41 IE_DESC = 'Generic downloader that works on some sites'
46 'url': 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
47 'md5': '85b90ccc9d73b4acd9138d3af4c27f89',
49 'id': '13601338388002',
51 'uploader': 'www.hodiho.fr',
52 'title': 'R\u00e9gis plante sa Jeep',
55 # bandcamp page with custom domain
57 'add_ie': ['Bandcamp'],
58 'url': 'http://bronyrock.com/track/the-pony-mash',
62 'title': 'The Pony Mash',
63 'uploader': 'M_Pallante',
65 'skip': 'There is a limit of 200 free downloads / month for the test song',
67 # embedded brightcove video
68 # it also tests brightcove videos that need to set the 'Referer' in the
71 'add_ie': ['Brightcove'],
72 'url': 'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/',
74 'id': '2765128793001',
76 'title': 'Le cours de bourse : l’analyse technique',
77 'description': 'md5:7e9ad046e968cb2d1114004aba466fd9',
78 'uploader': 'BFM BUSINESS',
81 'skip_download': True,
85 # https://github.com/rg3/youtube-dl/issues/2253
86 'url': 'http://bcove.me/i6nfkrc3',
87 'md5': '0ba9446db037002366bab3b3eb30c88c',
89 'id': '3101154703001',
91 'title': 'Still no power',
92 'uploader': 'thestar.com',
93 '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.',
95 'add_ie': ['Brightcove'],
98 'url': 'http://www.championat.com/video/football/v/87/87499.html',
99 'md5': 'fb973ecf6e4a78a67453647444222983',
101 'id': '3414141473001',
103 'title': 'Видео. Удаление Дзагоева (ЦСКА)',
104 'description': 'Онлайн-трансляция матча ЦСКА - "Волга"',
105 'uploader': 'Championat',
109 # https://github.com/rg3/youtube-dl/issues/3541
110 'add_ie': ['Brightcove'],
111 'url': 'http://www.kijk.nl/sbs6/leermijvrouwenkennen/videos/jqMiXKAYan2S/aflevering-1',
113 'id': '3866516442001',
115 'title': 'Leer mij vrouwen kennen: Aflevering 1',
116 'description': 'Leer mij vrouwen kennen: Aflevering 1',
117 'uploader': 'SBS Broadcasting',
119 'skip': 'Restricted to Netherlands',
121 'skip_download': True, # m3u8 download
124 # Direct link to a video
126 'url': 'http://media.w3.org/2010/05/sintel/trailer.mp4',
127 'md5': '67d406c2bcb6af27fa886f31aa934bbe',
132 'upload_date': '20100513',
137 'url': 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219',
138 'md5': '166dd577b433b4d4ebfee10b0824d8ff',
140 'id': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ',
142 'title': '2cc213299525360.mov', # that's what we get
144 'add_ie': ['Ooyala'],
146 # multiple ooyala embeds on SBN network websites
148 'url': 'http://www.sbnation.com/college-football-recruiting/2015/2/3/7970291/national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
150 'id': 'national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
151 'title': '25 lies you will tell yourself on National Signing Day - SBNation.com',
153 'playlist_mincount': 3,
155 'skip_download': True,
157 'add_ie': ['Ooyala'],
161 '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',
165 'upload_date': '20130224',
166 'uploader_id': 'TheVerge',
167 'description': 're:^Chris Ziegler takes a look at the\.*',
168 'uploader': 'The Verge',
169 'title': 'First Firefox OS phones side-by-side',
172 'skip_download': False,
177 'url': 'http://www.tested.com/science/weird/460206-tested-grinding-coffee-2000-frames-second/',
181 'title': 'Tested: Grinding Coffee at 2000 Frames Per Second',
182 'upload_date': '20140225',
183 'description': 'md5:06a40fbf30b220468f1e0957c0f558ff',
184 'uploader': 'Tested',
185 'uploader_id': 'testedcom',
187 # No need to test YoutubeIE here
189 'skip_download': True,
194 'url': 'http://www.theguardian.com/world/2014/mar/11/obama-zach-galifianakis-between-two-ferns',
198 'title': 'Between Two Ferns with Zach Galifianakis: President Barack Obama',
199 'description': 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.',
204 'url': 'http://www.bbc.co.uk/blogs/adamcurtis/posts/BUGGER',
206 'title': 'BBC - Blogs - Adam Curtis - BUGGER',
208 'playlist_mincount': 18,
212 'url': 'http://www.rg.ru/2014/03/15/reg-dfo/anklav-anons.html',
216 'title': 'Охотское море стало целиком российским',
217 'description': 'md5:5ed62483b14663e2a95ebbe115eb8f43',
221 'skip_download': True,
226 'url': 'http://en.support.wordpress.com/videos/ted-talks/',
227 'md5': '65fdff94098e4a607385a60c5177c638',
231 'title': 'Hidden miracles of the natural world',
232 'uploader': 'Louie Schwartzberg',
233 'description': 'md5:8145d19d320ff3e52f28401f4c4283b9',
236 # Embeded Ustream video
238 'url': 'http://www.american.edu/spa/pti/nsa-privacy-janus-2014.cfm',
239 'md5': '27b99cdb639c9b12a79bca876a073417',
243 'uploader': 'AU SPA: The NSA and Privacy',
244 'title': 'NSA and Privacy Forum Debate featuring General Hayden and Barton Gellman'
247 # nowvideo embed hidden behind percent encoding
249 'url': 'http://www.waoanime.tv/the-super-dimension-fortress-macross-episode-1/',
250 'md5': '2baf4ddd70f697d94b1c18cf796d5107',
252 'id': '06e53103ca9aa',
254 'title': 'Macross Episode 001 Watch Macross Episode 001 onl',
255 'description': 'No description',
260 'url': 'http://www.tv-replay.fr/redirection/20-03-14/x-enius-arte-10753389.html',
261 'md5': '7653032cbb25bf6c80d80f217055fa43',
263 'id': '048195-004_PLUS7-F',
266 'description': 'md5:d5fdf32ef6613cdbfd516ae658abf168',
267 'upload_date': '20140320',
270 'skip_download': 'Requires rtmpdump'
275 'url': 'http://www.wired.com/2014/04/honda-asimo/',
276 'md5': 'ba0dfe966fa007657bd1443ee672db0f',
278 'id': '53501be369702d3275860000',
280 'title': 'Honda’s New Asimo Robot Is More Human Than Ever',
285 'url': 'http://www.spi0n.com/zap-spi0n-com-n216/',
286 'md5': '441aeeb82eb72c422c7f14ec533999cd',
288 'id': 'k2mm4bCdJ6CQ2i7c8o2',
290 'title': 'Le Zap de Spi0n n°216 - Zapping du Web',
293 'add_ie': ['Dailymotion'],
297 'url': 'http://www.badzine.de/ansicht/datum/2014/06/09/so-funktioniert-die-neue-englische-badminton-liga.html',
301 'title': 'The NBL Auction 2014',
302 'uploader': 'BADMINTON England',
303 'uploader_id': 'BADMINTONEvents',
304 'upload_date': '20140603',
305 'description': 'md5:9ef128a69f1e262a700ed83edb163a73',
307 'add_ie': ['Youtube'],
309 'skip_download': True,
314 'url': 'http://www.gametrailers.com/news-post/76093/north-america-europe-is-getting-that-mario-kart-8-mercedes-dlc-too',
315 'md5': '35727f82f58c76d996fc188f9755b0d5',
317 'id': '0306a69b-8adf-4fb5-aace-75f8e8cbfca9',
320 'description': 'Mario\'s life in the fast lane has never looked so good.',
323 # YouTube embed via <data-embed-url="">
325 'url': 'https://play.google.com/store/apps/details?id=com.gameloft.android.ANMP.GloftA8HM',
329 'title': 'Asphalt 8: Airborne - Update - Welcome to Dubai!',
330 'uploader': 'Gameloft',
331 'uploader_id': 'gameloft',
332 'upload_date': '20140828',
333 'description': 'md5:c80da9ed3d83ae6d1876c834de03e1c4',
336 'skip_download': True,
341 'url': 'http://www.ll.mit.edu/workshops/education/videocourses/antennas/lecture1/video/',
343 'md5': '0c5e352edabf715d762b0ad4e6d9ee67',
345 'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
346 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - video1',
351 'md5': '10e4bb3aaca9fd630e273ff92d9f3c63',
353 'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final_PIP',
354 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - pip',
360 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
365 'url': 'http://www.handjobhub.com/video/busty-blonde-siri-tit-fuck-while-wank-6313.html',
366 'md5': '9d65602bf31c6e20014319c7d07fba27',
368 'id': '5123ea6d5e5a7',
371 'uploader': 'www.handjobhub.com',
372 'title': 'Busty Blonde Siri Tit Fuck While Wank at HandjobHub.com',
377 'url': 'http://phihag.de/2014/youtube-dl/rss2.xml',
379 'id': 'http://phihag.de/2014/youtube-dl/rss2.xml',
380 'title': 'Zero Punctuation',
381 'description': 're:.*groundbreaking video review series.*'
383 'playlist_mincount': 11,
385 # Multiple brightcove videos
386 # https://github.com/rg3/youtube-dl/issues/2283
388 'url': 'http://www.newyorker.com/online/blogs/newsdesk/2014/01/always-never-nuclear-command-and-control.html',
390 'id': 'always-never',
391 'title': 'Always / Never - The New Yorker',
395 'extract_flat': False,
396 'skip_download': True,
401 'url': 'http://umpire-empire.com/index.php/topic/58125-laz-decides-no-thats-low/',
402 'md5': '96f09a37e44da40dd083e12d9a683327',
406 'title': 'Ump changes call to ball',
407 'description': 'md5:71c11215384298a172a6dcb4c2e20685',
409 'timestamp': 1401537900,
410 'upload_date': '20140531',
411 'thumbnail': 're:^https?://.*\.jpg$',
416 'url': 'http://education-portal.com/academy/lesson/north-american-exploration-failed-colonies-of-spain-france-england.html#lesson',
417 'md5': '8788b683c777a5cf25621eaf286d0c23',
421 'title': 'md5:51364a8d3d009997ba99656004b5e20d',
423 'filesize': 182808282,
424 'uploader': 'education-portal.com',
428 'url': 'http://thoughtworks.wistia.com/medias/uxjb0lwrcz',
429 'md5': 'baf49c2baa8a7de5f3fc145a8506dcd4',
433 'title': 'Conversation about Hexagonal Rails Part 1 - ThoughtWorks',
435 'uploader': 'thoughtworks.wistia.com',
438 # Direct download with broken HEAD
440 'url': 'http://ai-radio.org:8000/radio.opus',
447 'skip_download': True, # infinite live stream
449 'expected_warnings': [
450 r'501.*Not Implemented'
455 'url': 'http://nakedsecurity.sophos.com/2014/10/29/sscc-171-are-you-sure-that-1234-is-a-bad-password-podcast/',
459 'description': 'md5:ff867d6b555488ad3c52572bb33d432c',
460 'uploader': 'Sophos Security',
461 'title': 'Chet Chat 171 - Oct 29, 2014',
462 'upload_date': '20141029',
467 'url': 'http://www.esa.int/Our_Activities/Space_Science/Rosetta/Philae_comet_touch-down_webcast',
471 'upload_date': '20141112',
472 'title': 'Rosetta #CometLanding webcast HL 10',
477 'url': 'http://discourse.ubuntu.com/t/unity-8-desktop-mode-windows-on-mir/1986',
480 'title': 'Unity 8 desktop-mode windows on Mir! - Ubuntu Discourse',
482 'playlist_mincount': 2,
484 # Direct link with incorrect MIME type
486 'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
487 'md5': '4ccbebe5f36706d85221f204d7eb5913',
489 'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
490 'id': '5_Lennart_Poettering_-_Systemd',
492 'title': '5_Lennart_Poettering_-_Systemd',
493 'upload_date': '20141120',
495 'expected_warnings': [
496 'URL could be a direct video link, returning it as such.'
501 'url': 'http://undergroundwellness.com/podcasts/306-5-steps-to-permanent-gut-healing/',
505 'upload_date': '20141126',
506 'title': 'Jack Tips: 5 Steps to Permanent Gut Healing',
511 'url': 'http://www.abc.net.au/7.30/content/2015/s4164797.htm',
513 'id': '730m_DandD_1901_512k',
515 'uploader': 'www.abc.net.au',
516 'title': 'Game of Thrones with dice - Dungeons and Dragons fantasy role-playing game gets new life - 19/01/2015',
519 # embedded viddler video
521 'url': 'http://deadspin.com/i-cant-stop-watching-john-wall-chop-the-nuggets-with-th-1681801597',
525 'uploader': 'deadspin',
526 'title': 'WALL-TO-GORTAT',
527 'timestamp': 1422285291,
528 'upload_date': '20150126',
530 'add_ie': ['Viddler'],
534 'url': 'http://thedailyshow.cc.com/podcast/episodetwelve',
538 'title': "The Daily Show Podcast without Jon Stewart - Episode 12: Bassem Youssef: Egypt's Jon Stewart",
539 'description': 'md5:601cb790edd05908957dae8aaa866465',
540 'upload_date': '20150220',
545 'url': 'http://media.nationalarchives.gov.uk/index.php/webinar-using-discovery-national-archives-online-catalogue/',
549 'upload_date': '20150212',
550 'uploader': 'The National Archives UK',
551 'description': 'md5:a236581cd2449dd2df4f93412f3f01c6',
552 'uploader_id': 'NationalArchives08',
553 'title': 'Webinar: Using Discovery, The National Archives’ online catalogue',
558 'url': 'http://www.rtlnieuws.nl/nieuws/buitenland/aanslagen-kopenhagen',
559 'playlist_mincount': 5,
561 'id': 'aanslagen-kopenhagen',
562 'title': 'Aanslagen Kopenhagen | RTL Nieuws',
567 'url': 'http://www.skipass.com/news/116090-bon-appetit-s5ep3-baqueira-mi-cor.html',
571 'title': 'EP3S5 - Bon Appétit - Baqueira Mi Corazon !',
576 'url': 'http://www.monumentalnetwork.com/videos/john-carlson-postgame-2-25-15',
580 'upload_date': '20150226',
581 'uploader_id': 'MonumentalSports-Kaltura@perfectsensedigital.com',
583 'title': 'John Carlson Postgame 2/25/15',
586 # Eagle.Platform embed (generic URL)
588 'url': 'http://lenta.ru/news/2015/03/06/navalny/',
592 'title': 'Навальный вышел на свободу',
593 'description': 'md5:d97861ac9ae77377f3f20eaf9d04b4f5',
594 'thumbnail': 're:^https?://.*\.jpg$',
600 # ClipYou (Eagle.Platform) embed (custom URL)
602 'url': 'http://muz-tv.ru/play/7129/',
606 'title': "'O Sole Mio",
607 'thumbnail': 're:^https?://.*\.jpg$',
614 'url': 'http://muz-tv.ru/kinozal/view/7400/',
618 'title': 'Тайны перевала Дятлова • Тайна перевала Дятлова 1 серия 2 часть',
619 'description': 'Документальный сериал-расследование одной из самых жутких тайн ХХ века',
620 'thumbnail': 're:^https?://.*\.jpg$',
627 'url': 'http://techcrunch.com/video/facebook-creates-on-this-day-crunch-report/518726732/',
628 'md5': '4c6f127a30736b59b3e2c19234ee2bf7',
632 'title': 'Facebook Creates "On This Day" | Crunch Report',
635 # RSS feed with enclosure
637 'url': 'http://podcastfeeds.nbcnews.com/audio/podcast/MSNBC-MADDOW-NETCAST-M4V.xml',
639 'id': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
641 'upload_date': '20150228',
642 'title': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
645 # Crooks and Liars embed
647 'url': 'http://crooksandliars.com/2015/04/fox-friends-says-protecting-atheists',
651 'title': "Fox & Friends Says Protecting Atheists From Discrimination Is Anti-Christian!",
652 'description': 'md5:e1a46ad1650e3a5ec7196d432799127f',
653 'timestamp': 1428207000,
654 'upload_date': '20150405',
655 'uploader': 'Heather',
658 # Crooks and Liars external embed
660 'url': 'http://theothermccain.com/2010/02/02/video-proves-that-bill-kristol-has-been-watching-glenn-beck/comment-page-1/',
662 'id': 'MTE3MjUtMzQ2MzA',
664 'title': 'md5:5e3662a81a4014d24c250d76d41a08d5',
665 'description': 'md5:9b8e9542d6c3c5de42d6451b7d780cec',
666 'timestamp': 1265032391,
667 'upload_date': '20100201',
668 'uploader': 'Heather',
671 # NBC Sports vplayer embed
673 'url': 'http://www.riderfans.com/forum/showthread.php?121827-Freeman&s=e98fa1ea6dc08e886b1678d35212494a',
675 'id': 'ln7x1qSThw4k',
677 'title': "PFT Live: New leader in the 'new-look' defense",
678 'description': 'md5:65a19b4bbfb3b0c0c5768bed1dfad74e',
683 'url': 'http://www.udn.com/news/story/7314/822787',
684 'md5': 'de06b4c90b042c128395a88f0384817e',
688 'title': '生物老師男變女 全校挺"做自己"',
689 'thumbnail': 're:^https?://.*\.jpg$',
694 def report_following_redirect(self, new_url):
695 """Report information extraction."""
696 self._downloader.to_screen('[redirect] Following redirect to %s' % new_url)
698 def _extract_rss(self, url, video_id, doc):
699 playlist_title = doc.find('./channel/title').text
700 playlist_desc_el = doc.find('./channel/description')
701 playlist_desc = None if playlist_desc_el is None else playlist_desc_el.text
704 for it in doc.findall('./channel/item'):
705 next_url = xpath_text(it, 'link', fatal=False)
707 enclosure_nodes = it.findall('./enclosure')
708 for e in enclosure_nodes:
709 next_url = e.attrib.get('url')
719 'title': it.find('title').text,
725 'title': playlist_title,
726 'description': playlist_desc,
730 def _extract_camtasia(self, url, video_id, webpage):
731 """ Returns None if no camtasia video can be found. """
733 camtasia_cfg = self._search_regex(
734 r'fo\.addVariable\(\s*"csConfigFile",\s*"([^"]+)"\s*\);',
735 webpage, 'camtasia configuration file', default=None)
736 if camtasia_cfg is None:
739 title = self._html_search_meta('DC.title', webpage, fatal=True)
741 camtasia_url = compat_urlparse.urljoin(url, camtasia_cfg)
742 camtasia_cfg = self._download_xml(
743 camtasia_url, video_id,
744 note='Downloading camtasia configuration',
745 errnote='Failed to download camtasia configuration')
746 fileset_node = camtasia_cfg.find('./playlist/array/fileset')
749 for n in fileset_node.getchildren():
750 url_n = n.find('./uri')
755 'id': os.path.splitext(url_n.text.rpartition('/')[2])[0],
756 'title': '%s - %s' % (title, n.tag),
757 'url': compat_urlparse.urljoin(url, url_n.text),
758 'duration': float_or_none(n.find('./duration').text),
767 def _real_extract(self, url):
768 if url.startswith('//'):
771 'url': self.http_scheme() + url,
774 parsed_url = compat_urlparse.urlparse(url)
775 if not parsed_url.scheme:
776 default_search = self._downloader.params.get('default_search')
777 if default_search is None:
778 default_search = 'fixup_error'
780 if default_search in ('auto', 'auto_warning', 'fixup_error'):
782 self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http')
783 return self.url_result('http://' + url)
784 elif default_search != 'fixup_error':
785 if default_search == 'auto_warning':
786 if re.match(r'^(?:url|URL)$', url):
787 raise ExtractorError(
788 'Invalid URL: %r . Call youtube-dl like this: youtube-dl -v "https://www.youtube.com/watch?v=BaW_jenozKc" ' % url,
791 self._downloader.report_warning(
792 'Falling back to youtube search for %s . Set --default-search "auto" to suppress this warning.' % url)
793 return self.url_result('ytsearch:' + url)
795 if default_search in ('error', 'fixup_error'):
796 raise ExtractorError(
797 '%r is not a valid URL. '
798 'Set --default-search "ytsearch" (or run youtube-dl "ytsearch:%s" ) to search YouTube'
799 % (url, url), expected=True)
801 if ':' not in default_search:
802 default_search += ':'
803 return self.url_result(default_search + url)
805 url, smuggled_data = unsmuggle_url(url)
807 is_intentional = smuggled_data and smuggled_data.get('to_generic')
808 if smuggled_data and 'force_videoid' in smuggled_data:
809 force_videoid = smuggled_data['force_videoid']
810 video_id = force_videoid
812 video_id = os.path.splitext(url.rstrip('/').split('/')[-1])[0]
814 self.to_screen('%s: Requesting header' % video_id)
816 head_req = HEADRequest(url)
817 head_response = self._request_webpage(
819 note=False, errnote='Could not send HEAD request to %s' % url,
822 if head_response is not False:
824 new_url = head_response.geturl()
826 self.report_following_redirect(new_url)
828 new_url = smuggle_url(
829 new_url, {'force_videoid': force_videoid})
830 return self.url_result(new_url)
833 if head_response is False:
834 full_response = self._request_webpage(url, video_id)
835 head_response = full_response
837 # Check for direct link to a video
838 content_type = head_response.headers.get('Content-Type', '')
839 m = re.match(r'^(?P<type>audio|video|application(?=/ogg$))/(?P<format_id>.+)$', content_type)
841 upload_date = unified_strdate(
842 head_response.headers.get('Last-Modified'))
845 'title': os.path.splitext(url_basename(url))[0],
848 'format_id': m.group('format_id'),
850 'vcodec': 'none' if m.group('type') == 'audio' else None
852 'upload_date': upload_date,
855 if not self._downloader.params.get('test', False) and not is_intentional:
856 self._downloader.report_warning('Falling back on generic information extractor.')
858 if not full_response:
859 full_response = self._request_webpage(url, video_id)
861 # Maybe it's a direct link to a video?
862 # Be careful not to download the whole thing!
863 first_bytes = full_response.read(512)
864 if not is_html(first_bytes):
865 self._downloader.report_warning(
866 'URL could be a direct video link, returning it as such.')
867 upload_date = unified_strdate(
868 head_response.headers.get('Last-Modified'))
871 'title': os.path.splitext(url_basename(url))[0],
874 'upload_date': upload_date,
877 webpage = self._webpage_read_content(
878 full_response, url, video_id, prefix=first_bytes)
880 self.report_extraction(video_id)
884 doc = parse_xml(webpage)
886 return self._extract_rss(url, video_id, doc)
887 except compat_xml_parse_error:
890 # Is it a Camtasia project?
891 camtasia_res = self._extract_camtasia(url, video_id, webpage)
892 if camtasia_res is not None:
895 # Sometimes embedded video player is hidden behind percent encoding
896 # (e.g. https://github.com/rg3/youtube-dl/issues/2448)
897 # Unescaping the whole page allows to handle those cases in a generic way
898 webpage = compat_urllib_parse.unquote(webpage)
900 # it's tempting to parse this further, but you would
901 # have to take into account all the variations like
902 # Video Title - Site Name
903 # Site Name | Video Title
904 # Video Title - Tagline | Site Name
905 # and so on and so forth; it's just not practical
906 video_title = self._html_search_regex(
907 r'(?s)<title>(.*?)</title>', webpage, 'video title',
910 # Try to detect age limit automatically
911 age_limit = self._rta_search(webpage)
912 # And then there are the jokers who advertise that they use RTA,
913 # but actually don't.
914 AGE_LIMIT_MARKERS = [
915 r'Proudly Labeled <a href="http://www.rtalabel.org/" title="Restricted to Adults">RTA</a>',
917 if any(re.search(marker, webpage) for marker in AGE_LIMIT_MARKERS):
920 # video uploader is domain name
921 video_uploader = self._search_regex(
922 r'^(?:https?://)?([^/]*)/.*', url, 'video uploader')
925 def _playlist_from_matches(matches, getter=None, ie=None):
927 self.url_result(self._proto_relative_url(getter(m) if getter else m), ie)
929 return self.playlist_result(
930 urlrs, playlist_id=video_id, playlist_title=video_title)
932 # Look for BrightCove:
933 bc_urls = BrightcoveIE._extract_brightcove_urls(webpage)
935 self.to_screen('Brightcove video detected.')
938 'url': smuggle_url(bc_url, {'Referer': url}),
939 'ie_key': 'Brightcove'
940 } for bc_url in bc_urls]
944 'title': video_title,
949 # Look for embedded rtl.nl player
950 matches = re.findall(
951 r'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="((?:https?:)?//(?:www\.)?rtl\.nl/system/videoplayer/[^"]+video_embed[^"]+)"',
954 return _playlist_from_matches(matches, ie='RtlNl')
956 # Look for embedded (iframe) Vimeo player
958 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//player\.vimeo\.com/video/.+?)\1', webpage)
960 player_url = unescapeHTML(mobj.group('url'))
961 surl = smuggle_url(player_url, {'Referer': url})
962 return self.url_result(surl)
963 # Look for embedded (swf embed) Vimeo player
965 r'<embed[^>]+?src="((?:https?:)?//(?:www\.)?vimeo\.com/moogaloop\.swf.+?)"', webpage)
967 return self.url_result(mobj.group(1))
969 # Look for embedded YouTube player
970 matches = re.findall(r'''(?x)
979 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
983 return _playlist_from_matches(
984 matches, lambda m: unescapeHTML(m[1]))
986 # Look for lazyYT YouTube embed
987 matches = re.findall(
988 r'class="lazyYT" data-youtube-id="([^"]+)"', webpage)
990 return _playlist_from_matches(matches, lambda m: unescapeHTML(m))
992 # Look for embedded Dailymotion player
993 matches = re.findall(
994 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/embed/video/.+?)\1', webpage)
996 return _playlist_from_matches(
997 matches, lambda m: unescapeHTML(m[1]))
999 # Look for embedded Dailymotion playlist player (#3822)
1001 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.[a-z]{2,3}/widget/jukebox\?.+?)\1', webpage)
1003 playlists = re.findall(
1004 r'list\[\]=/playlist/([^/]+)/', unescapeHTML(m.group('url')))
1006 return _playlist_from_matches(
1007 playlists, lambda p: '//dailymotion.com/playlist/%s' % p)
1009 # Look for embedded Wistia player
1011 r'<(?:meta[^>]+?content|iframe[^>]+?src)=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage)
1013 embed_url = self._proto_relative_url(
1014 unescapeHTML(match.group('url')))
1016 '_type': 'url_transparent',
1019 'uploader': video_uploader,
1020 'title': video_title,
1024 match = re.search(r'(?:id=["\']wistia_|data-wistia-?id=["\']|Wistia\.embed\(["\'])(?P<id>[^"\']+)', webpage)
1027 '_type': 'url_transparent',
1028 'url': 'http://fast.wistia.net/embed/iframe/{0:}'.format(match.group('id')),
1030 'uploader': video_uploader,
1031 'title': video_title,
1032 'id': match.group('id')
1035 # Look for embedded blip.tv player
1036 mobj = re.search(r'<meta\s[^>]*https?://api\.blip\.tv/\w+/redirect/\w+/(\d+)', webpage)
1038 return self.url_result('http://blip.tv/a/a-' + mobj.group(1), 'BlipTV')
1039 mobj = re.search(r'<(?:iframe|embed|object)\s[^>]*(https?://(?:\w+\.)?blip\.tv/(?:play/|api\.swf#)[a-zA-Z0-9_]+)', webpage)
1041 return self.url_result(mobj.group(1), 'BlipTV')
1043 # Look for embedded condenast player
1044 matches = re.findall(
1045 r'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="(https?://player\.cnevids\.com/embed/[^"]+")',
1049 '_type': 'playlist',
1052 'ie_key': 'CondeNast',
1054 } for ma in matches],
1055 'title': video_title,
1059 # Look for Bandcamp pages with custom domain
1060 mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage)
1061 if mobj is not None:
1062 burl = unescapeHTML(mobj.group(1))
1063 # Don't set the extractor because it can be a track url or an album
1064 return self.url_result(burl)
1066 # Look for embedded Vevo player
1068 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:cache\.)?vevo\.com/.+?)\1', webpage)
1069 if mobj is not None:
1070 return self.url_result(mobj.group('url'))
1072 # Look for embedded Viddler player
1074 r'<(?:iframe[^>]+?src|param[^>]+?value)=(["\'])(?P<url>(?:https?:)?//(?:www\.)?viddler\.com/(?:embed|player)/.+?)\1',
1076 if mobj is not None:
1077 return self.url_result(mobj.group('url'))
1079 # Look for NYTimes player
1081 r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//graphics8\.nytimes\.com/bcvideo/[^/]+/iframe/embed\.html.+?)\1>',
1083 if mobj is not None:
1084 return self.url_result(mobj.group('url'))
1086 # Look for Libsyn player
1088 r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//html5-player\.libsyn\.com/embed/.+?)\1', webpage)
1089 if mobj is not None:
1090 return self.url_result(mobj.group('url'))
1092 # Look for Ooyala videos
1093 mobj = (re.search(r'player\.ooyala\.com/[^"?]+\?[^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or
1094 re.search(r'OO\.Player\.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage) or
1095 re.search(r'SBN\.VideoLinkset\.ooyala\([\'"](?P<ec>.{32})[\'"]\)', webpage))
1096 if mobj is not None:
1097 return OoyalaIE._build_url_result(mobj.group('ec'))
1099 # Look for multiple Ooyala embeds on SBN network websites
1100 mobj = re.search(r'SBN\.VideoLinkset\.entryGroup\((\[.*?\])', webpage)
1101 if mobj is not None:
1102 embeds = self._parse_json(mobj.group(1), video_id, fatal=False)
1104 return _playlist_from_matches(
1105 embeds, getter=lambda v: OoyalaIE._url_for_embed_code(v['provider_video_id']), ie='Ooyala')
1107 # Look for Aparat videos
1108 mobj = re.search(r'<iframe .*?src="(http://www\.aparat\.com/video/[^"]+)"', webpage)
1109 if mobj is not None:
1110 return self.url_result(mobj.group(1), 'Aparat')
1112 # Look for MPORA videos
1113 mobj = re.search(r'<iframe .*?src="(http://mpora\.(?:com|de)/videos/[^"]+)"', webpage)
1114 if mobj is not None:
1115 return self.url_result(mobj.group(1), 'Mpora')
1117 # Look for embedded NovaMov-based player
1119 r'''(?x)<(?:pagespeed_)?iframe[^>]+?src=(["\'])
1120 (?P<url>http://(?:(?:embed|www)\.)?
1122 nowvideo\.(?:ch|sx|eu|at|ag|co)|
1123 videoweed\.(?:es|com)|
1124 movshare\.(?:net|sx|ag)|
1125 divxstage\.(?:eu|net|ch|co|at|ag))
1126 /embed\.php.+?)\1''', webpage)
1127 if mobj is not None:
1128 return self.url_result(mobj.group('url'))
1130 # Look for embedded Facebook player
1132 r'<iframe[^>]+?src=(["\'])(?P<url>https://www\.facebook\.com/video/embed.+?)\1', webpage)
1133 if mobj is not None:
1134 return self.url_result(mobj.group('url'), 'Facebook')
1136 # Look for embedded VK player
1137 mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://vk\.com/video_ext\.php.+?)\1', webpage)
1138 if mobj is not None:
1139 return self.url_result(mobj.group('url'), 'VK')
1141 # Look for embedded ivi player
1142 mobj = re.search(r'<embed[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?ivi\.ru/video/player.+?)\1', webpage)
1143 if mobj is not None:
1144 return self.url_result(mobj.group('url'), 'Ivi')
1146 # Look for embedded Huffington Post player
1148 r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed\.live\.huffingtonpost\.com/.+?)\1', webpage)
1149 if mobj is not None:
1150 return self.url_result(mobj.group('url'), 'HuffPost')
1153 mobj = re.search(r'class=["\']embedly-card["\'][^>]href=["\'](?P<url>[^"\']+)', webpage)
1154 if mobj is not None:
1155 return self.url_result(mobj.group('url'))
1156 mobj = re.search(r'class=["\']embedly-embed["\'][^>]src=["\'][^"\']*url=(?P<url>[^&]+)', webpage)
1157 if mobj is not None:
1158 return self.url_result(compat_urllib_parse.unquote(mobj.group('url')))
1160 # Look for funnyordie embed
1161 matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage)
1163 return _playlist_from_matches(
1164 matches, getter=unescapeHTML, ie='FunnyOrDie')
1166 # Look for BBC iPlayer embed
1167 matches = re.findall(r'setPlaylist\("(https?://www\.bbc\.co\.uk/iplayer/[^/]+/[\da-z]{8})"\)', webpage)
1169 return _playlist_from_matches(matches, ie='BBCCoUk')
1171 # Look for embedded RUTV player
1172 rutv_url = RUTVIE._extract_url(webpage)
1174 return self.url_result(rutv_url, 'RUTV')
1176 # Look for embedded TED player
1178 r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed(?:-ssl)?\.ted\.com/.+?)\1', webpage)
1179 if mobj is not None:
1180 return self.url_result(mobj.group('url'), 'TED')
1182 # Look for embedded Ustream videos
1184 r'<iframe[^>]+?src=(["\'])(?P<url>http://www\.ustream\.tv/embed/.+?)\1', webpage)
1185 if mobj is not None:
1186 return self.url_result(mobj.group('url'), 'Ustream')
1188 # Look for embedded arte.tv player
1190 r'<script [^>]*?src="(?P<url>http://www\.arte\.tv/playerv2/embed[^"]+)"',
1192 if mobj is not None:
1193 return self.url_result(mobj.group('url'), 'ArteTVEmbed')
1195 # Look for embedded smotri.com player
1196 smotri_url = SmotriIE._extract_url(webpage)
1198 return self.url_result(smotri_url, 'Smotri')
1200 # Look for embeded soundcloud player
1202 r'<iframe\s+(?:[a-zA-Z0-9_-]+="[^"]+"\s+)*src="(?P<url>https?://(?:w\.)?soundcloud\.com/player[^"]+)"',
1204 if mobj is not None:
1205 url = unescapeHTML(mobj.group('url'))
1206 return self.url_result(url)
1208 # Look for embedded vulture.com player
1210 r'<iframe src="(?P<url>https?://video\.vulture\.com/[^"]+)"',
1212 if mobj is not None:
1213 url = unescapeHTML(mobj.group('url'))
1214 return self.url_result(url, ie='Vulture')
1216 # Look for embedded mtvservices player
1218 r'<iframe src="(?P<url>https?://media\.mtvnservices\.com/embed/[^"]+)"',
1220 if mobj is not None:
1221 url = unescapeHTML(mobj.group('url'))
1222 return self.url_result(url, ie='MTVServicesEmbedded')
1224 # Look for embedded yahoo player
1226 r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:screen|movies)\.yahoo\.com/.+?\.html\?format=embed)\1',
1228 if mobj is not None:
1229 return self.url_result(mobj.group('url'), 'Yahoo')
1231 # Look for embedded sbs.com.au player
1235 <meta\s+property="og:video"\s+content=|
1238 (["\'])(?P<url>https?://(?:www\.)?sbs\.com\.au/ondemand/video/.+?)\1''',
1240 if mobj is not None:
1241 return self.url_result(mobj.group('url'), 'SBS')
1243 # Look for embedded Cinchcast player
1245 r'<iframe[^>]+?src=(["\'])(?P<url>https?://player\.cinchcast\.com/.+?)\1',
1247 if mobj is not None:
1248 return self.url_result(mobj.group('url'), 'Cinchcast')
1251 r'<iframe[^>]+?src=(["\'])(?P<url>https?://m(?:lb)?\.mlb\.com/shared/video/embed/embed\.html\?.+?)\1',
1253 if mobj is not None:
1254 return self.url_result(mobj.group('url'), 'MLB')
1257 r'<iframe[^>]+?src=(["\'])(?P<url>%s)\1' % CondeNastIE.EMBED_URL,
1259 if mobj is not None:
1260 return self.url_result(self._proto_relative_url(mobj.group('url'), scheme='http:'), 'CondeNast')
1263 r'<iframe[^>]+src="(?P<url>https?://new\.livestream\.com/[^"]+/player[^"]+)"',
1265 if mobj is not None:
1266 return self.url_result(mobj.group('url'), 'Livestream')
1268 # Look for Zapiks embed
1270 r'<iframe[^>]+src="(?P<url>https?://(?:www\.)?zapiks\.fr/index\.php\?.+?)"', webpage)
1271 if mobj is not None:
1272 return self.url_result(mobj.group('url'), 'Zapiks')
1274 # Look for Kaltura embeds
1276 r"(?s)kWidget\.(?:thumb)?[Ee]mbed\(\{.*?'wid'\s*:\s*'_?(?P<partner_id>[^']+)',.*?'entry_id'\s*:\s*'(?P<id>[^']+)',", webpage)
1277 if mobj is not None:
1278 return self.url_result('kaltura:%(partner_id)s:%(id)s' % mobj.groupdict(), 'Kaltura')
1280 # Look for Eagle.Platform embeds
1282 r'<iframe[^>]+src="(?P<url>https?://.+?\.media\.eagleplatform\.com/index/player\?.+?)"', webpage)
1283 if mobj is not None:
1284 return self.url_result(mobj.group('url'), 'EaglePlatform')
1286 # Look for ClipYou (uses Eagle.Platform) embeds
1288 r'<iframe[^>]+src="https?://(?P<host>media\.clipyou\.ru)/index/player\?.*\brecord_id=(?P<id>\d+).*"', webpage)
1289 if mobj is not None:
1290 return self.url_result('eagleplatform:%(host)s:%(id)s' % mobj.groupdict(), 'EaglePlatform')
1292 # Look for Pladform embeds
1294 r'<iframe[^>]+src="(?P<url>https?://out\.pladform\.ru/player\?.+?)"', webpage)
1295 if mobj is not None:
1296 return self.url_result(mobj.group('url'), 'Pladform')
1298 # Look for 5min embeds
1300 r'<meta[^>]+property="og:video"[^>]+content="https?://embed\.5min\.com/(?P<id>[0-9]+)/?', webpage)
1301 if mobj is not None:
1302 return self.url_result('5min:%s' % mobj.group('id'), 'FiveMin')
1304 # Look for Crooks and Liars embeds
1306 r'<(?:iframe[^>]+src|param[^>]+value)=(["\'])(?P<url>(?:https?:)?//embed\.crooksandliars\.com/(?:embed|v)/.+?)\1', webpage)
1307 if mobj is not None:
1308 return self.url_result(mobj.group('url'))
1310 # Look for NBC Sports VPlayer embeds
1311 nbc_sports_url = NBCSportsVPlayerIE._extract_url(webpage)
1313 return self.url_result(nbc_sports_url, 'NBCSportsVPlayer')
1315 # Look for UDN embeds
1317 r'<iframe[^>]+src="(?P<url>%s)"' % UDNEmbedIE._VALID_URL, webpage)
1318 if mobj is not None:
1319 return self.url_result(
1320 compat_urlparse.urljoin(url, mobj.group('url')), 'UDNEmbed')
1322 def check_video(vurl):
1323 if YoutubeIE.suitable(vurl):
1325 vpath = compat_urlparse.urlparse(vurl).path
1326 vext = determine_ext(vpath)
1327 return '.' in vpath and vext not in ('swf', 'png', 'jpg', 'srt', 'sbv', 'sub', 'vtt', 'ttml')
1329 def filter_video(urls):
1330 return list(filter(check_video, urls))
1332 # Start with something easy: JW Player in SWFObject
1333 found = filter_video(re.findall(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage))
1335 # Look for gorilla-vid style embedding
1336 found = filter_video(re.findall(r'''(?sx)
1340 jwplayer\s*\(\s*["'][^'"]+["']\s*\)\s*\.setup
1343 ['"]?file['"]?\s*:\s*["\'](.*?)["\']''', webpage))
1345 # Broaden the search a little bit
1346 found = filter_video(re.findall(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage))
1348 # Broaden the findall a little bit: JWPlayer JS loader
1349 found = filter_video(re.findall(
1350 r'[^A-Za-z0-9]?file["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage))
1353 found = filter_video(re.findall(r'''(?xs)
1354 flowplayer\("[^"]+",\s*
1356 \s*\{[^}]+? ["']?clip["']?\s*:\s*\{\s*
1357 ["']?url["']?\s*:\s*["']([^"']+)["']
1362 r"cinerama\.embedPlayer\(\s*\'[^']+\',\s*'([^']+)'", webpage)
1364 # Try to find twitter cards info
1365 found = filter_video(re.findall(
1366 r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage))
1368 # We look for Open Graph info:
1369 # We have to match any number spaces between elements, some sites try to align them (eg.: statigr.am)
1370 m_video_type = re.findall(r'<meta.*?property="og:video:type".*?content="video/(.*?)"', webpage)
1371 # We only look in og:video if the MIME type is a video, don't try if it's a Flash player:
1372 if m_video_type is not None:
1373 found = filter_video(re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage))
1376 found = re.findall(r'(?s)<video[^<]*(?:>.*?<source[^>]*)?\s+src=["\'](.*?)["\']', webpage)
1378 REDIRECT_REGEX = r'[0-9]{,2};\s*(?:URL|url)=\'?([^\'"]+)'
1380 r'(?i)<meta\s+(?=(?:[a-z-]+="[^"]+"\s+)*http-equiv="refresh")'
1381 r'(?:[a-z-]+="[^"]+"\s+)*?content="%s' % REDIRECT_REGEX,
1384 # Look also in Refresh HTTP header
1385 refresh_header = head_response.headers.get('Refresh')
1387 found = re.search(REDIRECT_REGEX, refresh_header)
1389 new_url = found.group(1)
1390 self.report_following_redirect(new_url)
1396 raise UnsupportedError(url)
1399 for video_url in found:
1400 video_url = compat_urlparse.urljoin(url, video_url)
1401 video_id = compat_urllib_parse.unquote(os.path.basename(video_url))
1403 # Sometimes, jwplayer extraction will result in a YouTube URL
1404 if YoutubeIE.suitable(video_url):
1405 entries.append(self.url_result(video_url, 'Youtube'))
1408 # here's a fun little line of code for you:
1409 video_id = os.path.splitext(video_id)[0]
1414 'uploader': video_uploader,
1415 'title': video_title,
1416 'age_limit': age_limit,
1419 if len(entries) == 1:
1422 for num, e in enumerate(entries, start=1):
1423 # 'url' results don't have a title
1424 if e.get('title') is not None:
1425 e['title'] = '%s (%d)' % (e['title'], num)
1427 '_type': 'playlist',