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 .ooyala import OoyalaIE
33 from .rutv import RUTVIE
34 from .smotri import SmotriIE
35 from .condenast import CondeNastIE
38 class GenericIE(InfoExtractor):
39 IE_DESC = 'Generic downloader that works on some sites'
44 'url': 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
45 'md5': '85b90ccc9d73b4acd9138d3af4c27f89',
47 'id': '13601338388002',
49 'uploader': 'www.hodiho.fr',
50 'title': 'R\u00e9gis plante sa Jeep',
53 # bandcamp page with custom domain
55 'add_ie': ['Bandcamp'],
56 'url': 'http://bronyrock.com/track/the-pony-mash',
60 'title': 'The Pony Mash',
61 'uploader': 'M_Pallante',
63 'skip': 'There is a limit of 200 free downloads / month for the test song',
65 # embedded brightcove video
66 # it also tests brightcove videos that need to set the 'Referer' in the
69 'add_ie': ['Brightcove'],
70 'url': 'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/',
72 'id': '2765128793001',
74 'title': 'Le cours de bourse : l’analyse technique',
75 'description': 'md5:7e9ad046e968cb2d1114004aba466fd9',
76 'uploader': 'BFM BUSINESS',
79 'skip_download': True,
83 # https://github.com/rg3/youtube-dl/issues/2253
84 'url': 'http://bcove.me/i6nfkrc3',
85 'md5': '0ba9446db037002366bab3b3eb30c88c',
87 'id': '3101154703001',
89 'title': 'Still no power',
90 'uploader': 'thestar.com',
91 '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.',
93 'add_ie': ['Brightcove'],
96 'url': 'http://www.championat.com/video/football/v/87/87499.html',
97 'md5': 'fb973ecf6e4a78a67453647444222983',
99 'id': '3414141473001',
101 'title': 'Видео. Удаление Дзагоева (ЦСКА)',
102 'description': 'Онлайн-трансляция матча ЦСКА - "Волга"',
103 'uploader': 'Championat',
107 # https://github.com/rg3/youtube-dl/issues/3541
108 'add_ie': ['Brightcove'],
109 'url': 'http://www.kijk.nl/sbs6/leermijvrouwenkennen/videos/jqMiXKAYan2S/aflevering-1',
111 'id': '3866516442001',
113 'title': 'Leer mij vrouwen kennen: Aflevering 1',
114 'description': 'Leer mij vrouwen kennen: Aflevering 1',
115 'uploader': 'SBS Broadcasting',
117 'skip': 'Restricted to Netherlands',
119 'skip_download': True, # m3u8 download
122 # Direct link to a video
124 'url': 'http://media.w3.org/2010/05/sintel/trailer.mp4',
125 'md5': '67d406c2bcb6af27fa886f31aa934bbe',
130 'upload_date': '20100513',
135 'url': 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219',
136 'md5': '166dd577b433b4d4ebfee10b0824d8ff',
138 'id': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ',
140 'title': '2cc213299525360.mov', # that's what we get
142 'add_ie': ['Ooyala'],
144 # multiple ooyala embeds on SBN network websites
146 'url': 'http://www.sbnation.com/college-football-recruiting/2015/2/3/7970291/national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
148 'id': 'national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
149 'title': '25 lies you will tell yourself on National Signing Day - SBNation.com',
151 'playlist_mincount': 3,
153 'skip_download': True,
155 'add_ie': ['Ooyala'],
159 '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',
163 'upload_date': '20130224',
164 'uploader_id': 'TheVerge',
165 'description': 're:^Chris Ziegler takes a look at the\.*',
166 'uploader': 'The Verge',
167 'title': 'First Firefox OS phones side-by-side',
170 'skip_download': False,
175 'url': 'http://www.tested.com/science/weird/460206-tested-grinding-coffee-2000-frames-second/',
179 'title': 'Tested: Grinding Coffee at 2000 Frames Per Second',
180 'upload_date': '20140225',
181 'description': 'md5:06a40fbf30b220468f1e0957c0f558ff',
182 'uploader': 'Tested',
183 'uploader_id': 'testedcom',
185 # No need to test YoutubeIE here
187 'skip_download': True,
192 'url': 'http://www.theguardian.com/world/2014/mar/11/obama-zach-galifianakis-between-two-ferns',
196 'title': 'Between Two Ferns with Zach Galifianakis: President Barack Obama',
197 'description': 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.',
202 'url': 'http://www.bbc.co.uk/blogs/adamcurtis/posts/BUGGER',
204 'title': 'BBC - Blogs - Adam Curtis - BUGGER',
206 'playlist_mincount': 18,
210 'url': 'http://www.rg.ru/2014/03/15/reg-dfo/anklav-anons.html',
214 'title': 'Охотское море стало целиком российским',
215 'description': 'md5:5ed62483b14663e2a95ebbe115eb8f43',
219 'skip_download': True,
224 'url': 'http://en.support.wordpress.com/videos/ted-talks/',
225 'md5': '65fdff94098e4a607385a60c5177c638',
229 'title': 'Hidden miracles of the natural world',
230 'uploader': 'Louie Schwartzberg',
231 'description': 'md5:8145d19d320ff3e52f28401f4c4283b9',
234 # Embeded Ustream video
236 'url': 'http://www.american.edu/spa/pti/nsa-privacy-janus-2014.cfm',
237 'md5': '27b99cdb639c9b12a79bca876a073417',
241 'uploader': 'AU SPA: The NSA and Privacy',
242 'title': 'NSA and Privacy Forum Debate featuring General Hayden and Barton Gellman'
245 # nowvideo embed hidden behind percent encoding
247 'url': 'http://www.waoanime.tv/the-super-dimension-fortress-macross-episode-1/',
248 'md5': '2baf4ddd70f697d94b1c18cf796d5107',
250 'id': '06e53103ca9aa',
252 'title': 'Macross Episode 001 Watch Macross Episode 001 onl',
253 'description': 'No description',
258 'url': 'http://www.tv-replay.fr/redirection/20-03-14/x-enius-arte-10753389.html',
259 'md5': '7653032cbb25bf6c80d80f217055fa43',
261 'id': '048195-004_PLUS7-F',
264 'description': 'md5:d5fdf32ef6613cdbfd516ae658abf168',
265 'upload_date': '20140320',
268 'skip_download': 'Requires rtmpdump'
273 'url': 'http://www.wired.com/2014/04/honda-asimo/',
274 'md5': 'ba0dfe966fa007657bd1443ee672db0f',
276 'id': '53501be369702d3275860000',
278 'title': 'Honda’s New Asimo Robot Is More Human Than Ever',
283 'url': 'http://www.spi0n.com/zap-spi0n-com-n216/',
284 'md5': '441aeeb82eb72c422c7f14ec533999cd',
286 'id': 'k2mm4bCdJ6CQ2i7c8o2',
288 'title': 'Le Zap de Spi0n n°216 - Zapping du Web',
291 'add_ie': ['Dailymotion'],
295 'url': 'http://www.badzine.de/ansicht/datum/2014/06/09/so-funktioniert-die-neue-englische-badminton-liga.html',
299 'title': 'The NBL Auction 2014',
300 'uploader': 'BADMINTON England',
301 'uploader_id': 'BADMINTONEvents',
302 'upload_date': '20140603',
303 'description': 'md5:9ef128a69f1e262a700ed83edb163a73',
305 'add_ie': ['Youtube'],
307 'skip_download': True,
312 'url': 'http://www.gametrailers.com/news-post/76093/north-america-europe-is-getting-that-mario-kart-8-mercedes-dlc-too',
313 'md5': '35727f82f58c76d996fc188f9755b0d5',
315 'id': '0306a69b-8adf-4fb5-aace-75f8e8cbfca9',
318 'description': 'Mario\'s life in the fast lane has never looked so good.',
321 # YouTube embed via <data-embed-url="">
323 'url': 'https://play.google.com/store/apps/details?id=com.gameloft.android.ANMP.GloftA8HM',
327 'title': 'Asphalt 8: Airborne - Update - Welcome to Dubai!',
328 'uploader': 'Gameloft',
329 'uploader_id': 'gameloft',
330 'upload_date': '20140828',
331 'description': 'md5:c80da9ed3d83ae6d1876c834de03e1c4',
334 'skip_download': True,
339 'url': 'http://www.ll.mit.edu/workshops/education/videocourses/antennas/lecture1/video/',
341 'md5': '0c5e352edabf715d762b0ad4e6d9ee67',
343 'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
344 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - video1',
349 'md5': '10e4bb3aaca9fd630e273ff92d9f3c63',
351 'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final_PIP',
352 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - pip',
358 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
363 'url': 'http://www.handjobhub.com/video/busty-blonde-siri-tit-fuck-while-wank-6313.html',
364 'md5': '9d65602bf31c6e20014319c7d07fba27',
366 'id': '5123ea6d5e5a7',
369 'uploader': 'www.handjobhub.com',
370 'title': 'Busty Blonde Siri Tit Fuck While Wank at HandjobHub.com',
375 'url': 'http://phihag.de/2014/youtube-dl/rss2.xml',
377 'id': 'http://phihag.de/2014/youtube-dl/rss2.xml',
378 'title': 'Zero Punctuation',
379 'description': 're:.*groundbreaking video review series.*'
381 'playlist_mincount': 11,
383 # Multiple brightcove videos
384 # https://github.com/rg3/youtube-dl/issues/2283
386 'url': 'http://www.newyorker.com/online/blogs/newsdesk/2014/01/always-never-nuclear-command-and-control.html',
388 'id': 'always-never',
389 'title': 'Always / Never - The New Yorker',
393 'extract_flat': False,
394 'skip_download': True,
399 'url': 'http://umpire-empire.com/index.php/topic/58125-laz-decides-no-thats-low/',
400 'md5': '96f09a37e44da40dd083e12d9a683327',
404 'title': 'Ump changes call to ball',
405 'description': 'md5:71c11215384298a172a6dcb4c2e20685',
407 'timestamp': 1401537900,
408 'upload_date': '20140531',
409 'thumbnail': 're:^https?://.*\.jpg$',
414 'url': 'http://education-portal.com/academy/lesson/north-american-exploration-failed-colonies-of-spain-france-england.html#lesson',
415 'md5': '8788b683c777a5cf25621eaf286d0c23',
419 'title': 'md5:51364a8d3d009997ba99656004b5e20d',
421 'filesize': 182808282,
422 'uploader': 'education-portal.com',
426 'url': 'http://thoughtworks.wistia.com/medias/uxjb0lwrcz',
427 'md5': 'baf49c2baa8a7de5f3fc145a8506dcd4',
431 'title': 'Conversation about Hexagonal Rails Part 1 - ThoughtWorks',
433 'uploader': 'thoughtworks.wistia.com',
436 # Direct download with broken HEAD
438 'url': 'http://ai-radio.org:8000/radio.opus',
445 'skip_download': True, # infinite live stream
447 'expected_warnings': [
448 r'501.*Not Implemented'
453 'url': 'http://nakedsecurity.sophos.com/2014/10/29/sscc-171-are-you-sure-that-1234-is-a-bad-password-podcast/',
457 'description': 'md5:ff867d6b555488ad3c52572bb33d432c',
458 'uploader': 'Sophos Security',
459 'title': 'Chet Chat 171 - Oct 29, 2014',
460 'upload_date': '20141029',
465 'url': 'http://www.esa.int/Our_Activities/Space_Science/Rosetta/Philae_comet_touch-down_webcast',
469 'upload_date': '20141112',
470 'title': 'Rosetta #CometLanding webcast HL 10',
475 'url': 'http://discourse.ubuntu.com/t/unity-8-desktop-mode-windows-on-mir/1986',
478 'title': 'Unity 8 desktop-mode windows on Mir! - Ubuntu Discourse',
480 'playlist_mincount': 2,
482 # Direct link with incorrect MIME type
484 'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
485 'md5': '4ccbebe5f36706d85221f204d7eb5913',
487 'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
488 'id': '5_Lennart_Poettering_-_Systemd',
490 'title': '5_Lennart_Poettering_-_Systemd',
491 'upload_date': '20141120',
493 'expected_warnings': [
494 'URL could be a direct video link, returning it as such.'
499 'url': 'http://undergroundwellness.com/podcasts/306-5-steps-to-permanent-gut-healing/',
503 'upload_date': '20141126',
504 'title': 'Jack Tips: 5 Steps to Permanent Gut Healing',
509 'url': 'http://www.abc.net.au/7.30/content/2015/s4164797.htm',
511 'id': '730m_DandD_1901_512k',
513 'uploader': 'www.abc.net.au',
514 'title': 'Game of Thrones with dice - Dungeons and Dragons fantasy role-playing game gets new life - 19/01/2015',
517 # embedded viddler video
519 'url': 'http://deadspin.com/i-cant-stop-watching-john-wall-chop-the-nuggets-with-th-1681801597',
523 'uploader': 'deadspin',
524 'title': 'WALL-TO-GORTAT',
525 'timestamp': 1422285291,
526 'upload_date': '20150126',
528 'add_ie': ['Viddler'],
532 'url': 'http://media.nationalarchives.gov.uk/index.php/webinar-using-discovery-national-archives-online-catalogue/',
536 'upload_date': '20150212',
537 'uploader': 'The National Archives UK',
538 'description': 'md5:a236581cd2449dd2df4f93412f3f01c6',
539 'uploader_id': 'NationalArchives08',
540 'title': 'Webinar: Using Discovery, The National Archives’ online catalogue',
545 'url': 'http://www.rtlnieuws.nl/nieuws/buitenland/aanslagen-kopenhagen',
546 'playlist_mincount': 5,
548 'id': 'aanslagen-kopenhagen',
549 'title': 'Aanslagen Kopenhagen | RTL Nieuws',
554 'url': 'http://www.skipass.com/news/116090-bon-appetit-s5ep3-baqueira-mi-cor.html',
558 'title': 'EP3S5 - Bon Appétit - Baqueira Mi Corazon !',
563 'url': 'http://www.monumentalnetwork.com/videos/john-carlson-postgame-2-25-15',
567 'upload_date': '20150226',
568 'uploader_id': 'MonumentalSports-Kaltura@perfectsensedigital.com',
570 'title': 'John Carlson Postgame 2/25/15',
573 # RSS feed with enclosure
575 'url': 'http://podcastfeeds.nbcnews.com/audio/podcast/MSNBC-MADDOW-NETCAST-M4V.xml',
577 'id': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
579 'upload_date': '20150228',
580 'title': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
585 def report_following_redirect(self, new_url):
586 """Report information extraction."""
587 self._downloader.to_screen('[redirect] Following redirect to %s' % new_url)
589 def _extract_rss(self, url, video_id, doc):
590 playlist_title = doc.find('./channel/title').text
591 playlist_desc_el = doc.find('./channel/description')
592 playlist_desc = None if playlist_desc_el is None else playlist_desc_el.text
595 for it in doc.findall('./channel/item'):
596 next_url = xpath_text(it, 'link', fatal=False)
598 enclosure_nodes = it.findall('./enclosure')
599 for e in enclosure_nodes:
600 next_url = e.attrib.get('url')
610 'title': it.find('title').text,
616 'title': playlist_title,
617 'description': playlist_desc,
621 def _extract_camtasia(self, url, video_id, webpage):
622 """ Returns None if no camtasia video can be found. """
624 camtasia_cfg = self._search_regex(
625 r'fo\.addVariable\(\s*"csConfigFile",\s*"([^"]+)"\s*\);',
626 webpage, 'camtasia configuration file', default=None)
627 if camtasia_cfg is None:
630 title = self._html_search_meta('DC.title', webpage, fatal=True)
632 camtasia_url = compat_urlparse.urljoin(url, camtasia_cfg)
633 camtasia_cfg = self._download_xml(
634 camtasia_url, video_id,
635 note='Downloading camtasia configuration',
636 errnote='Failed to download camtasia configuration')
637 fileset_node = camtasia_cfg.find('./playlist/array/fileset')
640 for n in fileset_node.getchildren():
641 url_n = n.find('./uri')
646 'id': os.path.splitext(url_n.text.rpartition('/')[2])[0],
647 'title': '%s - %s' % (title, n.tag),
648 'url': compat_urlparse.urljoin(url, url_n.text),
649 'duration': float_or_none(n.find('./duration').text),
658 def _real_extract(self, url):
659 if url.startswith('//'):
662 'url': self.http_scheme() + url,
665 parsed_url = compat_urlparse.urlparse(url)
666 if not parsed_url.scheme:
667 default_search = self._downloader.params.get('default_search')
668 if default_search is None:
669 default_search = 'fixup_error'
671 if default_search in ('auto', 'auto_warning', 'fixup_error'):
673 self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http')
674 return self.url_result('http://' + url)
675 elif default_search != 'fixup_error':
676 if default_search == 'auto_warning':
677 if re.match(r'^(?:url|URL)$', url):
678 raise ExtractorError(
679 'Invalid URL: %r . Call youtube-dl like this: youtube-dl -v "https://www.youtube.com/watch?v=BaW_jenozKc" ' % url,
682 self._downloader.report_warning(
683 'Falling back to youtube search for %s . Set --default-search "auto" to suppress this warning.' % url)
684 return self.url_result('ytsearch:' + url)
686 if default_search in ('error', 'fixup_error'):
687 raise ExtractorError(
688 '%r is not a valid URL. '
689 'Set --default-search "ytsearch" (or run youtube-dl "ytsearch:%s" ) to search YouTube'
690 % (url, url), expected=True)
692 if ':' not in default_search:
693 default_search += ':'
694 return self.url_result(default_search + url)
696 url, smuggled_data = unsmuggle_url(url)
698 is_intentional = smuggled_data and smuggled_data.get('to_generic')
699 if smuggled_data and 'force_videoid' in smuggled_data:
700 force_videoid = smuggled_data['force_videoid']
701 video_id = force_videoid
703 video_id = os.path.splitext(url.rstrip('/').split('/')[-1])[0]
705 self.to_screen('%s: Requesting header' % video_id)
707 head_req = HEADRequest(url)
708 head_response = self._request_webpage(
710 note=False, errnote='Could not send HEAD request to %s' % url,
713 if head_response is not False:
715 new_url = head_response.geturl()
717 self.report_following_redirect(new_url)
719 new_url = smuggle_url(
720 new_url, {'force_videoid': force_videoid})
721 return self.url_result(new_url)
724 if head_response is False:
725 full_response = self._request_webpage(url, video_id)
726 head_response = full_response
728 # Check for direct link to a video
729 content_type = head_response.headers.get('Content-Type', '')
730 m = re.match(r'^(?P<type>audio|video|application(?=/ogg$))/(?P<format_id>.+)$', content_type)
732 upload_date = unified_strdate(
733 head_response.headers.get('Last-Modified'))
736 'title': os.path.splitext(url_basename(url))[0],
739 'format_id': m.group('format_id'),
741 'vcodec': 'none' if m.group('type') == 'audio' else None
743 'upload_date': upload_date,
746 if not self._downloader.params.get('test', False) and not is_intentional:
747 self._downloader.report_warning('Falling back on generic information extractor.')
749 if not full_response:
750 full_response = self._request_webpage(url, video_id)
752 # Maybe it's a direct link to a video?
753 # Be careful not to download the whole thing!
754 first_bytes = full_response.read(512)
755 if not is_html(first_bytes):
756 self._downloader.report_warning(
757 'URL could be a direct video link, returning it as such.')
758 upload_date = unified_strdate(
759 head_response.headers.get('Last-Modified'))
762 'title': os.path.splitext(url_basename(url))[0],
765 'upload_date': upload_date,
768 webpage = self._webpage_read_content(
769 full_response, url, video_id, prefix=first_bytes)
771 self.report_extraction(video_id)
775 doc = parse_xml(webpage)
777 return self._extract_rss(url, video_id, doc)
778 except compat_xml_parse_error:
781 # Is it a Camtasia project?
782 camtasia_res = self._extract_camtasia(url, video_id, webpage)
783 if camtasia_res is not None:
786 # Sometimes embedded video player is hidden behind percent encoding
787 # (e.g. https://github.com/rg3/youtube-dl/issues/2448)
788 # Unescaping the whole page allows to handle those cases in a generic way
789 webpage = compat_urllib_parse.unquote(webpage)
791 # it's tempting to parse this further, but you would
792 # have to take into account all the variations like
793 # Video Title - Site Name
794 # Site Name | Video Title
795 # Video Title - Tagline | Site Name
796 # and so on and so forth; it's just not practical
797 video_title = self._html_search_regex(
798 r'(?s)<title>(.*?)</title>', webpage, 'video title',
801 # Try to detect age limit automatically
802 age_limit = self._rta_search(webpage)
803 # And then there are the jokers who advertise that they use RTA,
804 # but actually don't.
805 AGE_LIMIT_MARKERS = [
806 r'Proudly Labeled <a href="http://www.rtalabel.org/" title="Restricted to Adults">RTA</a>',
808 if any(re.search(marker, webpage) for marker in AGE_LIMIT_MARKERS):
811 # video uploader is domain name
812 video_uploader = self._search_regex(
813 r'^(?:https?://)?([^/]*)/.*', url, 'video uploader')
816 def _playlist_from_matches(matches, getter=None, ie=None):
818 self.url_result(self._proto_relative_url(getter(m) if getter else m), ie)
820 return self.playlist_result(
821 urlrs, playlist_id=video_id, playlist_title=video_title)
823 # Look for BrightCove:
824 bc_urls = BrightcoveIE._extract_brightcove_urls(webpage)
826 self.to_screen('Brightcove video detected.')
829 'url': smuggle_url(bc_url, {'Referer': url}),
830 'ie_key': 'Brightcove'
831 } for bc_url in bc_urls]
835 'title': video_title,
840 # Look for embedded rtl.nl player
841 matches = re.findall(
842 r'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="((?:https?:)?//(?:www\.)?rtl\.nl/system/videoplayer/[^"]+video_embed[^"]+)"',
845 return _playlist_from_matches(matches, ie='RtlNl')
847 # Look for embedded (iframe) Vimeo player
849 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//player\.vimeo\.com/video/.+?)\1', webpage)
851 player_url = unescapeHTML(mobj.group('url'))
852 surl = smuggle_url(player_url, {'Referer': url})
853 return self.url_result(surl)
854 # Look for embedded (swf embed) Vimeo player
856 r'<embed[^>]+?src="((?:https?:)?//(?:www\.)?vimeo\.com/moogaloop\.swf.+?)"', webpage)
858 return self.url_result(mobj.group(1))
860 # Look for embedded YouTube player
861 matches = re.findall(r'''(?x)
870 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
874 return _playlist_from_matches(
875 matches, lambda m: unescapeHTML(m[1]))
877 # Look for lazyYT YouTube embed
878 matches = re.findall(
879 r'class="lazyYT" data-youtube-id="([^"]+)"', webpage)
881 return _playlist_from_matches(matches, lambda m: unescapeHTML(m))
883 # Look for embedded Dailymotion player
884 matches = re.findall(
885 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/embed/video/.+?)\1', webpage)
887 return _playlist_from_matches(
888 matches, lambda m: unescapeHTML(m[1]))
890 # Look for embedded Dailymotion playlist player (#3822)
892 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.[a-z]{2,3}/widget/jukebox\?.+?)\1', webpage)
894 playlists = re.findall(
895 r'list\[\]=/playlist/([^/]+)/', unescapeHTML(m.group('url')))
897 return _playlist_from_matches(
898 playlists, lambda p: '//dailymotion.com/playlist/%s' % p)
900 # Look for embedded Wistia player
902 r'<(?:meta[^>]+?content|iframe[^>]+?src)=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage)
904 embed_url = self._proto_relative_url(
905 unescapeHTML(match.group('url')))
907 '_type': 'url_transparent',
910 'uploader': video_uploader,
911 'title': video_title,
915 match = re.search(r'(?:id=["\']wistia_|data-wistia-?id=["\']|Wistia\.embed\(["\'])(?P<id>[^"\']+)', webpage)
918 '_type': 'url_transparent',
919 'url': 'http://fast.wistia.net/embed/iframe/{0:}'.format(match.group('id')),
921 'uploader': video_uploader,
922 'title': video_title,
923 'id': match.group('id')
926 # Look for embedded blip.tv player
927 mobj = re.search(r'<meta\s[^>]*https?://api\.blip\.tv/\w+/redirect/\w+/(\d+)', webpage)
929 return self.url_result('http://blip.tv/a/a-' + mobj.group(1), 'BlipTV')
930 mobj = re.search(r'<(?:iframe|embed|object)\s[^>]*(https?://(?:\w+\.)?blip\.tv/(?:play/|api\.swf#)[a-zA-Z0-9_]+)', webpage)
932 return self.url_result(mobj.group(1), 'BlipTV')
934 # Look for embedded condenast player
935 matches = re.findall(
936 r'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="(https?://player\.cnevids\.com/embed/[^"]+")',
943 'ie_key': 'CondeNast',
945 } for ma in matches],
946 'title': video_title,
950 # Look for Bandcamp pages with custom domain
951 mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage)
953 burl = unescapeHTML(mobj.group(1))
954 # Don't set the extractor because it can be a track url or an album
955 return self.url_result(burl)
957 # Look for embedded Vevo player
959 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:cache\.)?vevo\.com/.+?)\1', webpage)
961 return self.url_result(mobj.group('url'))
963 # Look for embedded Viddler player
965 r'<(?:iframe[^>]+?src|param[^>]+?value)=(["\'])(?P<url>(?:https?:)?//(?:www\.)?viddler\.com/(?:embed|player)/.+?)\1',
968 return self.url_result(mobj.group('url'))
970 # Look for Ooyala videos
971 mobj = (re.search(r'player\.ooyala\.com/[^"?]+\?[^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or
972 re.search(r'OO\.Player\.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage) or
973 re.search(r'SBN\.VideoLinkset\.ooyala\([\'"](?P<ec>.{32})[\'"]\)', webpage))
975 return OoyalaIE._build_url_result(mobj.group('ec'))
977 # Look for multiple Ooyala embeds on SBN network websites
978 mobj = re.search(r'SBN\.VideoLinkset\.entryGroup\((\[.*?\])', webpage)
980 embeds = self._parse_json(mobj.group(1), video_id, fatal=False)
982 return _playlist_from_matches(
983 embeds, getter=lambda v: OoyalaIE._url_for_embed_code(v['provider_video_id']), ie='Ooyala')
985 # Look for Aparat videos
986 mobj = re.search(r'<iframe .*?src="(http://www\.aparat\.com/video/[^"]+)"', webpage)
988 return self.url_result(mobj.group(1), 'Aparat')
990 # Look for MPORA videos
991 mobj = re.search(r'<iframe .*?src="(http://mpora\.(?:com|de)/videos/[^"]+)"', webpage)
993 return self.url_result(mobj.group(1), 'Mpora')
995 # Look for embedded NovaMov-based player
997 r'''(?x)<(?:pagespeed_)?iframe[^>]+?src=(["\'])
998 (?P<url>http://(?:(?:embed|www)\.)?
1000 nowvideo\.(?:ch|sx|eu|at|ag|co)|
1001 videoweed\.(?:es|com)|
1002 movshare\.(?:net|sx|ag)|
1003 divxstage\.(?:eu|net|ch|co|at|ag))
1004 /embed\.php.+?)\1''', webpage)
1005 if mobj is not None:
1006 return self.url_result(mobj.group('url'))
1008 # Look for embedded Facebook player
1010 r'<iframe[^>]+?src=(["\'])(?P<url>https://www\.facebook\.com/video/embed.+?)\1', webpage)
1011 if mobj is not None:
1012 return self.url_result(mobj.group('url'), 'Facebook')
1014 # Look for embedded VK player
1015 mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://vk\.com/video_ext\.php.+?)\1', webpage)
1016 if mobj is not None:
1017 return self.url_result(mobj.group('url'), 'VK')
1019 # Look for embedded ivi player
1020 mobj = re.search(r'<embed[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?ivi\.ru/video/player.+?)\1', webpage)
1021 if mobj is not None:
1022 return self.url_result(mobj.group('url'), 'Ivi')
1024 # Look for embedded Huffington Post player
1026 r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed\.live\.huffingtonpost\.com/.+?)\1', webpage)
1027 if mobj is not None:
1028 return self.url_result(mobj.group('url'), 'HuffPost')
1031 mobj = re.search(r'class=["\']embedly-card["\'][^>]href=["\'](?P<url>[^"\']+)', webpage)
1032 if mobj is not None:
1033 return self.url_result(mobj.group('url'))
1034 mobj = re.search(r'class=["\']embedly-embed["\'][^>]src=["\'][^"\']*url=(?P<url>[^&]+)', webpage)
1035 if mobj is not None:
1036 return self.url_result(compat_urllib_parse.unquote(mobj.group('url')))
1038 # Look for funnyordie embed
1039 matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage)
1041 return _playlist_from_matches(
1042 matches, getter=unescapeHTML, ie='FunnyOrDie')
1044 # Look for BBC iPlayer embed
1045 matches = re.findall(r'setPlaylist\("(https?://www\.bbc\.co\.uk/iplayer/[^/]+/[\da-z]{8})"\)', webpage)
1047 return _playlist_from_matches(matches, ie='BBCCoUk')
1049 # Look for embedded RUTV player
1050 rutv_url = RUTVIE._extract_url(webpage)
1052 return self.url_result(rutv_url, 'RUTV')
1054 # Look for embedded TED player
1056 r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed(?:-ssl)?\.ted\.com/.+?)\1', webpage)
1057 if mobj is not None:
1058 return self.url_result(mobj.group('url'), 'TED')
1060 # Look for embedded Ustream videos
1062 r'<iframe[^>]+?src=(["\'])(?P<url>http://www\.ustream\.tv/embed/.+?)\1', webpage)
1063 if mobj is not None:
1064 return self.url_result(mobj.group('url'), 'Ustream')
1066 # Look for embedded arte.tv player
1068 r'<script [^>]*?src="(?P<url>http://www\.arte\.tv/playerv2/embed[^"]+)"',
1070 if mobj is not None:
1071 return self.url_result(mobj.group('url'), 'ArteTVEmbed')
1073 # Look for embedded smotri.com player
1074 smotri_url = SmotriIE._extract_url(webpage)
1076 return self.url_result(smotri_url, 'Smotri')
1078 # Look for embeded soundcloud player
1080 r'<iframe\s+(?:[a-zA-Z0-9_-]+="[^"]+"\s+)*src="(?P<url>https?://(?:w\.)?soundcloud\.com/player[^"]+)"',
1082 if mobj is not None:
1083 url = unescapeHTML(mobj.group('url'))
1084 return self.url_result(url)
1086 # Look for embedded vulture.com player
1088 r'<iframe src="(?P<url>https?://video\.vulture\.com/[^"]+)"',
1090 if mobj is not None:
1091 url = unescapeHTML(mobj.group('url'))
1092 return self.url_result(url, ie='Vulture')
1094 # Look for embedded mtvservices player
1096 r'<iframe src="(?P<url>https?://media\.mtvnservices\.com/embed/[^"]+)"',
1098 if mobj is not None:
1099 url = unescapeHTML(mobj.group('url'))
1100 return self.url_result(url, ie='MTVServicesEmbedded')
1102 # Look for embedded yahoo player
1104 r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:screen|movies)\.yahoo\.com/.+?\.html\?format=embed)\1',
1106 if mobj is not None:
1107 return self.url_result(mobj.group('url'), 'Yahoo')
1109 # Look for embedded sbs.com.au player
1113 <meta\s+property="og:video"\s+content=|
1116 (["\'])(?P<url>https?://(?:www\.)?sbs\.com\.au/ondemand/video/.+?)\1''',
1118 if mobj is not None:
1119 return self.url_result(mobj.group('url'), 'SBS')
1121 # Look for embedded Cinchcast player
1123 r'<iframe[^>]+?src=(["\'])(?P<url>https?://player\.cinchcast\.com/.+?)\1',
1125 if mobj is not None:
1126 return self.url_result(mobj.group('url'), 'Cinchcast')
1129 r'<iframe[^>]+?src=(["\'])(?P<url>https?://m(?:lb)?\.mlb\.com/shared/video/embed/embed\.html\?.+?)\1',
1131 if mobj is not None:
1132 return self.url_result(mobj.group('url'), 'MLB')
1135 r'<iframe[^>]+?src=(["\'])(?P<url>%s)\1' % CondeNastIE.EMBED_URL,
1137 if mobj is not None:
1138 return self.url_result(self._proto_relative_url(mobj.group('url'), scheme='http:'), 'CondeNast')
1141 r'<iframe[^>]+src="(?P<url>https?://new\.livestream\.com/[^"]+/player[^"]+)"',
1143 if mobj is not None:
1144 return self.url_result(mobj.group('url'), 'Livestream')
1146 # Look for Zapiks embed
1148 r'<iframe[^>]+src="(?P<url>https?://(?:www\.)?zapiks\.fr/index\.php\?.+?)"', webpage)
1149 if mobj is not None:
1150 return self.url_result(mobj.group('url'), 'Zapiks')
1152 # Look for Kaltura embeds
1154 r"(?s)kWidget\.(?:thumb)?[Ee]mbed\(\{.*?'wid'\s*:\s*'_?(?P<partner_id>[^']+)',.*?'entry_id'\s*:\s*'(?P<id>[^']+)',", webpage)
1155 if mobj is not None:
1156 return self.url_result('kaltura:%(partner_id)s:%(id)s' % mobj.groupdict(), 'Kaltura')
1158 def check_video(vurl):
1159 if YoutubeIE.suitable(vurl):
1161 vpath = compat_urlparse.urlparse(vurl).path
1162 vext = determine_ext(vpath)
1163 return '.' in vpath and vext not in ('swf', 'png', 'jpg', 'srt', 'sbv', 'sub', 'vtt', 'ttml')
1165 def filter_video(urls):
1166 return list(filter(check_video, urls))
1168 # Start with something easy: JW Player in SWFObject
1169 found = filter_video(re.findall(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage))
1171 # Look for gorilla-vid style embedding
1172 found = filter_video(re.findall(r'''(?sx)
1176 jwplayer\s*\(\s*["'][^'"]+["']\s*\)\s*\.setup
1179 ['"]?file['"]?\s*:\s*["\'](.*?)["\']''', webpage))
1181 # Broaden the search a little bit
1182 found = filter_video(re.findall(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage))
1184 # Broaden the findall a little bit: JWPlayer JS loader
1185 found = filter_video(re.findall(
1186 r'[^A-Za-z0-9]?file["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage))
1189 found = filter_video(re.findall(r'''(?xs)
1190 flowplayer\("[^"]+",\s*
1192 \s*\{[^}]+? ["']?clip["']?\s*:\s*\{\s*
1193 ["']?url["']?\s*:\s*["']([^"']+)["']
1198 r"cinerama\.embedPlayer\(\s*\'[^']+\',\s*'([^']+)'", webpage)
1200 # Try to find twitter cards info
1201 found = filter_video(re.findall(
1202 r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage))
1204 # We look for Open Graph info:
1205 # We have to match any number spaces between elements, some sites try to align them (eg.: statigr.am)
1206 m_video_type = re.findall(r'<meta.*?property="og:video:type".*?content="video/(.*?)"', webpage)
1207 # We only look in og:video if the MIME type is a video, don't try if it's a Flash player:
1208 if m_video_type is not None:
1209 found = filter_video(re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage))
1212 found = re.findall(r'(?s)<video[^<]*(?:>.*?<source[^>]*)?\s+src=["\'](.*?)["\']', webpage)
1215 r'(?i)<meta\s+(?=(?:[a-z-]+="[^"]+"\s+)*http-equiv="refresh")'
1216 r'(?:[a-z-]+="[^"]+"\s+)*?content="[0-9]{,2};url=\'?([^\'"]+)',
1219 new_url = found.group(1)
1220 self.report_following_redirect(new_url)
1226 raise UnsupportedError(url)
1229 for video_url in found:
1230 video_url = compat_urlparse.urljoin(url, video_url)
1231 video_id = compat_urllib_parse.unquote(os.path.basename(video_url))
1233 # Sometimes, jwplayer extraction will result in a YouTube URL
1234 if YoutubeIE.suitable(video_url):
1235 entries.append(self.url_result(video_url, 'Youtube'))
1238 # here's a fun little line of code for you:
1239 video_id = os.path.splitext(video_id)[0]
1244 'uploader': video_uploader,
1245 'title': video_title,
1246 'age_limit': age_limit,
1249 if len(entries) == 1:
1252 for num, e in enumerate(entries, start=1):
1253 # 'url' results don't have a title
1254 if e.get('title') is not None:
1255 e['title'] = '%s (%d)' % (e['title'], num)
1257 '_type': 'playlist',