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,
30 from .brightcove import BrightcoveIE
31 from .ooyala import OoyalaIE
32 from .rutv import RUTVIE
33 from .smotri import SmotriIE
34 from .condenast import CondeNastIE
37 class GenericIE(InfoExtractor):
38 IE_DESC = 'Generic downloader that works on some sites'
43 'url': 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
44 'md5': '85b90ccc9d73b4acd9138d3af4c27f89',
46 'id': '13601338388002',
48 'uploader': 'www.hodiho.fr',
49 'title': 'R\u00e9gis plante sa Jeep',
52 # bandcamp page with custom domain
54 'add_ie': ['Bandcamp'],
55 'url': 'http://bronyrock.com/track/the-pony-mash',
59 'title': 'The Pony Mash',
60 'uploader': 'M_Pallante',
62 'skip': 'There is a limit of 200 free downloads / month for the test song',
64 # embedded brightcove video
65 # it also tests brightcove videos that need to set the 'Referer' in the
68 'add_ie': ['Brightcove'],
69 'url': 'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/',
71 'id': '2765128793001',
73 'title': 'Le cours de bourse : l’analyse technique',
74 'description': 'md5:7e9ad046e968cb2d1114004aba466fd9',
75 'uploader': 'BFM BUSINESS',
78 'skip_download': True,
82 # https://github.com/rg3/youtube-dl/issues/2253
83 'url': 'http://bcove.me/i6nfkrc3',
84 'md5': '0ba9446db037002366bab3b3eb30c88c',
86 'id': '3101154703001',
88 'title': 'Still no power',
89 'uploader': 'thestar.com',
90 '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.',
92 'add_ie': ['Brightcove'],
95 'url': 'http://www.championat.com/video/football/v/87/87499.html',
96 'md5': 'fb973ecf6e4a78a67453647444222983',
98 'id': '3414141473001',
100 'title': 'Видео. Удаление Дзагоева (ЦСКА)',
101 'description': 'Онлайн-трансляция матча ЦСКА - "Волга"',
102 'uploader': 'Championat',
106 # https://github.com/rg3/youtube-dl/issues/3541
107 'add_ie': ['Brightcove'],
108 'url': 'http://www.kijk.nl/sbs6/leermijvrouwenkennen/videos/jqMiXKAYan2S/aflevering-1',
110 'id': '3866516442001',
112 'title': 'Leer mij vrouwen kennen: Aflevering 1',
113 'description': 'Leer mij vrouwen kennen: Aflevering 1',
114 'uploader': 'SBS Broadcasting',
116 'skip': 'Restricted to Netherlands',
118 'skip_download': True, # m3u8 download
121 # Direct link to a video
123 'url': 'http://media.w3.org/2010/05/sintel/trailer.mp4',
124 'md5': '67d406c2bcb6af27fa886f31aa934bbe',
129 'upload_date': '20100513',
134 'url': 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219',
135 'md5': '166dd577b433b4d4ebfee10b0824d8ff',
137 'id': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ',
139 'title': '2cc213299525360.mov', # that's what we get
141 'add_ie': ['Ooyala'],
143 # multiple ooyala embeds on SBN network websites
145 'url': 'http://www.sbnation.com/college-football-recruiting/2015/2/3/7970291/national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
147 'id': 'national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
148 'title': '25 lies you will tell yourself on National Signing Day - SBNation.com',
150 'playlist_mincount': 3,
152 'skip_download': True,
154 'add_ie': ['Ooyala'],
158 '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',
162 'upload_date': '20130224',
163 'uploader_id': 'TheVerge',
164 'description': 're:^Chris Ziegler takes a look at the\.*',
165 'uploader': 'The Verge',
166 'title': 'First Firefox OS phones side-by-side',
169 'skip_download': False,
174 'url': 'http://www.tested.com/science/weird/460206-tested-grinding-coffee-2000-frames-second/',
178 'title': 'Tested: Grinding Coffee at 2000 Frames Per Second',
179 'upload_date': '20140225',
180 'description': 'md5:06a40fbf30b220468f1e0957c0f558ff',
181 'uploader': 'Tested',
182 'uploader_id': 'testedcom',
184 # No need to test YoutubeIE here
186 'skip_download': True,
191 'url': 'http://www.theguardian.com/world/2014/mar/11/obama-zach-galifianakis-between-two-ferns',
195 'title': 'Between Two Ferns with Zach Galifianakis: President Barack Obama',
196 'description': 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.',
201 'url': 'http://www.bbc.co.uk/blogs/adamcurtis/posts/BUGGER',
203 'title': 'BBC - Blogs - Adam Curtis - BUGGER',
205 'playlist_mincount': 18,
209 'url': 'http://www.rg.ru/2014/03/15/reg-dfo/anklav-anons.html',
213 'title': 'Охотское море стало целиком российским',
214 'description': 'md5:5ed62483b14663e2a95ebbe115eb8f43',
218 'skip_download': True,
223 'url': 'http://en.support.wordpress.com/videos/ted-talks/',
224 'md5': '65fdff94098e4a607385a60c5177c638',
228 'title': 'Hidden miracles of the natural world',
229 'uploader': 'Louie Schwartzberg',
230 'description': 'md5:8145d19d320ff3e52f28401f4c4283b9',
233 # Embeded Ustream video
235 'url': 'http://www.american.edu/spa/pti/nsa-privacy-janus-2014.cfm',
236 'md5': '27b99cdb639c9b12a79bca876a073417',
240 'uploader': 'AU SPA: The NSA and Privacy',
241 'title': 'NSA and Privacy Forum Debate featuring General Hayden and Barton Gellman'
244 # nowvideo embed hidden behind percent encoding
246 'url': 'http://www.waoanime.tv/the-super-dimension-fortress-macross-episode-1/',
247 'md5': '2baf4ddd70f697d94b1c18cf796d5107',
249 'id': '06e53103ca9aa',
251 'title': 'Macross Episode 001 Watch Macross Episode 001 onl',
252 'description': 'No description',
257 'url': 'http://www.tv-replay.fr/redirection/20-03-14/x-enius-arte-10753389.html',
258 'md5': '7653032cbb25bf6c80d80f217055fa43',
260 'id': '048195-004_PLUS7-F',
263 'description': 'md5:d5fdf32ef6613cdbfd516ae658abf168',
264 'upload_date': '20140320',
267 'skip_download': 'Requires rtmpdump'
272 'url': 'http://www.wired.com/2014/04/honda-asimo/',
273 'md5': 'ba0dfe966fa007657bd1443ee672db0f',
275 'id': '53501be369702d3275860000',
277 'title': 'Honda’s New Asimo Robot Is More Human Than Ever',
282 'url': 'http://www.spi0n.com/zap-spi0n-com-n216/',
283 'md5': '441aeeb82eb72c422c7f14ec533999cd',
285 'id': 'k2mm4bCdJ6CQ2i7c8o2',
287 'title': 'Le Zap de Spi0n n°216 - Zapping du Web',
290 'add_ie': ['Dailymotion'],
294 'url': 'http://www.badzine.de/ansicht/datum/2014/06/09/so-funktioniert-die-neue-englische-badminton-liga.html',
298 'title': 'The NBL Auction 2014',
299 'uploader': 'BADMINTON England',
300 'uploader_id': 'BADMINTONEvents',
301 'upload_date': '20140603',
302 'description': 'md5:9ef128a69f1e262a700ed83edb163a73',
304 'add_ie': ['Youtube'],
306 'skip_download': True,
311 'url': 'http://www.gametrailers.com/news-post/76093/north-america-europe-is-getting-that-mario-kart-8-mercedes-dlc-too',
312 'md5': '35727f82f58c76d996fc188f9755b0d5',
314 'id': '0306a69b-8adf-4fb5-aace-75f8e8cbfca9',
317 'description': 'Mario\'s life in the fast lane has never looked so good.',
320 # YouTube embed via <data-embed-url="">
322 'url': 'https://play.google.com/store/apps/details?id=com.gameloft.android.ANMP.GloftA8HM',
326 'title': 'Asphalt 8: Airborne - Update - Welcome to Dubai!',
327 'uploader': 'Gameloft',
328 'uploader_id': 'gameloft',
329 'upload_date': '20140828',
330 'description': 'md5:c80da9ed3d83ae6d1876c834de03e1c4',
333 'skip_download': True,
338 'url': 'http://www.ll.mit.edu/workshops/education/videocourses/antennas/lecture1/video/',
340 'md5': '0c5e352edabf715d762b0ad4e6d9ee67',
342 'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
343 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - video1',
348 'md5': '10e4bb3aaca9fd630e273ff92d9f3c63',
350 'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final_PIP',
351 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - pip',
357 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
362 'url': 'http://www.handjobhub.com/video/busty-blonde-siri-tit-fuck-while-wank-6313.html',
363 'md5': '9d65602bf31c6e20014319c7d07fba27',
365 'id': '5123ea6d5e5a7',
368 'uploader': 'www.handjobhub.com',
369 'title': 'Busty Blonde Siri Tit Fuck While Wank at HandjobHub.com',
374 'url': 'http://phihag.de/2014/youtube-dl/rss2.xml',
376 'id': 'http://phihag.de/2014/youtube-dl/rss2.xml',
377 'title': 'Zero Punctuation',
378 'description': 're:.*groundbreaking video review series.*'
380 'playlist_mincount': 11,
382 # Multiple brightcove videos
383 # https://github.com/rg3/youtube-dl/issues/2283
385 'url': 'http://www.newyorker.com/online/blogs/newsdesk/2014/01/always-never-nuclear-command-and-control.html',
387 'id': 'always-never',
388 'title': 'Always / Never - The New Yorker',
392 'extract_flat': False,
393 'skip_download': True,
398 'url': 'http://umpire-empire.com/index.php/topic/58125-laz-decides-no-thats-low/',
399 'md5': '96f09a37e44da40dd083e12d9a683327',
403 'title': 'Ump changes call to ball',
404 'description': 'md5:71c11215384298a172a6dcb4c2e20685',
406 'timestamp': 1401537900,
407 'upload_date': '20140531',
408 'thumbnail': 're:^https?://.*\.jpg$',
413 'url': 'http://education-portal.com/academy/lesson/north-american-exploration-failed-colonies-of-spain-france-england.html#lesson',
414 'md5': '8788b683c777a5cf25621eaf286d0c23',
418 'title': 'md5:51364a8d3d009997ba99656004b5e20d',
420 'filesize': 182808282,
421 'uploader': 'education-portal.com',
425 'url': 'http://thoughtworks.wistia.com/medias/uxjb0lwrcz',
426 'md5': 'baf49c2baa8a7de5f3fc145a8506dcd4',
430 'title': 'Conversation about Hexagonal Rails Part 1 - ThoughtWorks',
432 'uploader': 'thoughtworks.wistia.com',
435 # Direct download with broken HEAD
437 'url': 'http://ai-radio.org:8000/radio.opus',
444 'skip_download': True, # infinite live stream
446 'expected_warnings': [
447 r'501.*Not Implemented'
452 'url': 'http://nakedsecurity.sophos.com/2014/10/29/sscc-171-are-you-sure-that-1234-is-a-bad-password-podcast/',
456 'description': 'md5:ff867d6b555488ad3c52572bb33d432c',
457 'uploader': 'Sophos Security',
458 'title': 'Chet Chat 171 - Oct 29, 2014',
459 'upload_date': '20141029',
464 'url': 'http://www.esa.int/Our_Activities/Space_Science/Rosetta/Philae_comet_touch-down_webcast',
468 'upload_date': '20141112',
469 'title': 'Rosetta #CometLanding webcast HL 10',
474 'url': 'http://discourse.ubuntu.com/t/unity-8-desktop-mode-windows-on-mir/1986',
476 'title': 'Unity 8 desktop-mode windows on Mir! - Ubuntu Discourse',
478 'playlist_mincount': 2,
480 # Direct link with incorrect MIME type
482 'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
483 'md5': '4ccbebe5f36706d85221f204d7eb5913',
485 'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
486 'id': '5_Lennart_Poettering_-_Systemd',
488 'title': '5_Lennart_Poettering_-_Systemd',
489 'upload_date': '20141120',
491 'expected_warnings': [
492 'URL could be a direct video link, returning it as such.'
497 'url': 'http://undergroundwellness.com/podcasts/306-5-steps-to-permanent-gut-healing/',
501 'upload_date': '20141126',
502 'title': 'Jack Tips: 5 Steps to Permanent Gut Healing',
507 'url': 'http://www.abc.net.au/7.30/content/2015/s4164797.htm',
509 'id': '730m_DandD_1901_512k',
511 'uploader': 'www.abc.net.au',
512 'title': 'Game of Thrones with dice - Dungeons and Dragons fantasy role-playing game gets new life - 19/01/2015',
515 # embedded viddler video
517 'url': 'http://deadspin.com/i-cant-stop-watching-john-wall-chop-the-nuggets-with-th-1681801597',
521 'uploader': 'deadspin',
522 'title': 'WALL-TO-GORTAT',
523 'timestamp': 1422285291,
524 'upload_date': '20150126',
526 'add_ie': ['Viddler'],
530 'url': 'http://media.nationalarchives.gov.uk/index.php/webinar-using-discovery-national-archives-online-catalogue/',
534 'upload_date': '20150204',
535 'uploader': 'The National Archives UK',
536 'description': 'md5:a236581cd2449dd2df4f93412f3f01c6',
537 'uploader_id': 'NationalArchives08',
538 'title': 'Webinar: Using Discovery, The National Archives’ online catalogue',
543 'url': 'http://www.rtlnieuws.nl/nieuws/buitenland/aanslagen-kopenhagen',
544 'playlist_mincount': 5,
546 'id': 'aanslagen-kopenhagen',
547 'title': 'Aanslagen Kopenhagen | RTL Nieuws',
552 def report_following_redirect(self, new_url):
553 """Report information extraction."""
554 self._downloader.to_screen('[redirect] Following redirect to %s' % new_url)
556 def _extract_rss(self, url, video_id, doc):
557 playlist_title = doc.find('./channel/title').text
558 playlist_desc_el = doc.find('./channel/description')
559 playlist_desc = None if playlist_desc_el is None else playlist_desc_el.text
563 'url': e.find('link').text,
564 'title': e.find('title').text,
565 } for e in doc.findall('./channel/item')]
570 'title': playlist_title,
571 'description': playlist_desc,
575 def _extract_camtasia(self, url, video_id, webpage):
576 """ Returns None if no camtasia video can be found. """
578 camtasia_cfg = self._search_regex(
579 r'fo\.addVariable\(\s*"csConfigFile",\s*"([^"]+)"\s*\);',
580 webpage, 'camtasia configuration file', default=None)
581 if camtasia_cfg is None:
584 title = self._html_search_meta('DC.title', webpage, fatal=True)
586 camtasia_url = compat_urlparse.urljoin(url, camtasia_cfg)
587 camtasia_cfg = self._download_xml(
588 camtasia_url, video_id,
589 note='Downloading camtasia configuration',
590 errnote='Failed to download camtasia configuration')
591 fileset_node = camtasia_cfg.find('./playlist/array/fileset')
594 for n in fileset_node.getchildren():
595 url_n = n.find('./uri')
600 'id': os.path.splitext(url_n.text.rpartition('/')[2])[0],
601 'title': '%s - %s' % (title, n.tag),
602 'url': compat_urlparse.urljoin(url, url_n.text),
603 'duration': float_or_none(n.find('./duration').text),
612 def _real_extract(self, url):
613 if url.startswith('//'):
616 'url': self.http_scheme() + url,
619 parsed_url = compat_urlparse.urlparse(url)
620 if not parsed_url.scheme:
621 default_search = self._downloader.params.get('default_search')
622 if default_search is None:
623 default_search = 'fixup_error'
625 if default_search in ('auto', 'auto_warning', 'fixup_error'):
627 self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http')
628 return self.url_result('http://' + url)
629 elif default_search != 'fixup_error':
630 if default_search == 'auto_warning':
631 if re.match(r'^(?:url|URL)$', url):
632 raise ExtractorError(
633 'Invalid URL: %r . Call youtube-dl like this: youtube-dl -v "https://www.youtube.com/watch?v=BaW_jenozKc" ' % url,
636 self._downloader.report_warning(
637 'Falling back to youtube search for %s . Set --default-search "auto" to suppress this warning.' % url)
638 return self.url_result('ytsearch:' + url)
640 if default_search in ('error', 'fixup_error'):
641 raise ExtractorError(
642 '%r is not a valid URL. '
643 'Set --default-search "ytsearch" (or run youtube-dl "ytsearch:%s" ) to search YouTube'
644 % (url, url), expected=True)
646 if ':' not in default_search:
647 default_search += ':'
648 return self.url_result(default_search + url)
650 url, smuggled_data = unsmuggle_url(url)
652 is_intentional = smuggled_data and smuggled_data.get('to_generic')
653 if smuggled_data and 'force_videoid' in smuggled_data:
654 force_videoid = smuggled_data['force_videoid']
655 video_id = force_videoid
657 video_id = os.path.splitext(url.rstrip('/').split('/')[-1])[0]
659 self.to_screen('%s: Requesting header' % video_id)
661 head_req = HEADRequest(url)
662 head_response = self._request_webpage(
664 note=False, errnote='Could not send HEAD request to %s' % url,
667 if head_response is not False:
669 new_url = head_response.geturl()
671 self.report_following_redirect(new_url)
673 new_url = smuggle_url(
674 new_url, {'force_videoid': force_videoid})
675 return self.url_result(new_url)
678 if head_response is False:
679 full_response = self._request_webpage(url, video_id)
680 head_response = full_response
682 # Check for direct link to a video
683 content_type = head_response.headers.get('Content-Type', '')
684 m = re.match(r'^(?P<type>audio|video|application(?=/ogg$))/(?P<format_id>.+)$', content_type)
686 upload_date = unified_strdate(
687 head_response.headers.get('Last-Modified'))
690 'title': os.path.splitext(url_basename(url))[0],
693 'format_id': m.group('format_id'),
695 'vcodec': 'none' if m.group('type') == 'audio' else None
697 'upload_date': upload_date,
700 if not self._downloader.params.get('test', False) and not is_intentional:
701 self._downloader.report_warning('Falling back on generic information extractor.')
703 if not full_response:
704 full_response = self._request_webpage(url, video_id)
706 # Maybe it's a direct link to a video?
707 # Be careful not to download the whole thing!
708 first_bytes = full_response.read(512)
709 if not is_html(first_bytes):
710 self._downloader.report_warning(
711 'URL could be a direct video link, returning it as such.')
712 upload_date = unified_strdate(
713 head_response.headers.get('Last-Modified'))
716 'title': os.path.splitext(url_basename(url))[0],
719 'upload_date': upload_date,
722 webpage = self._webpage_read_content(
723 full_response, url, video_id, prefix=first_bytes)
725 self.report_extraction(video_id)
729 doc = parse_xml(webpage)
731 return self._extract_rss(url, video_id, doc)
732 except compat_xml_parse_error:
735 # Is it a Camtasia project?
736 camtasia_res = self._extract_camtasia(url, video_id, webpage)
737 if camtasia_res is not None:
740 # Sometimes embedded video player is hidden behind percent encoding
741 # (e.g. https://github.com/rg3/youtube-dl/issues/2448)
742 # Unescaping the whole page allows to handle those cases in a generic way
743 webpage = compat_urllib_parse.unquote(webpage)
745 # it's tempting to parse this further, but you would
746 # have to take into account all the variations like
747 # Video Title - Site Name
748 # Site Name | Video Title
749 # Video Title - Tagline | Site Name
750 # and so on and so forth; it's just not practical
751 video_title = self._html_search_regex(
752 r'(?s)<title>(.*?)</title>', webpage, 'video title',
755 # Try to detect age limit automatically
756 age_limit = self._rta_search(webpage)
757 # And then there are the jokers who advertise that they use RTA,
758 # but actually don't.
759 AGE_LIMIT_MARKERS = [
760 r'Proudly Labeled <a href="http://www.rtalabel.org/" title="Restricted to Adults">RTA</a>',
762 if any(re.search(marker, webpage) for marker in AGE_LIMIT_MARKERS):
765 # video uploader is domain name
766 video_uploader = self._search_regex(
767 r'^(?:https?://)?([^/]*)/.*', url, 'video uploader')
770 def _playlist_from_matches(matches, getter=None, ie=None):
772 self.url_result(self._proto_relative_url(getter(m) if getter else m), ie)
774 return self.playlist_result(
775 urlrs, playlist_id=video_id, playlist_title=video_title)
777 # Look for BrightCove:
778 bc_urls = BrightcoveIE._extract_brightcove_urls(webpage)
780 self.to_screen('Brightcove video detected.')
783 'url': smuggle_url(bc_url, {'Referer': url}),
784 'ie_key': 'Brightcove'
785 } for bc_url in bc_urls]
789 'title': video_title,
794 # Look for embedded rtl.nl player
795 matches = re.findall(
796 r'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="((?:https?:)?//(?:www\.)?rtl\.nl/system/videoplayer/[^"]+video_embed[^"]+)"',
799 return _playlist_from_matches(matches, ie='RtlNl')
801 # Look for embedded (iframe) Vimeo player
803 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//player\.vimeo\.com/video/.+?)\1', webpage)
805 player_url = unescapeHTML(mobj.group('url'))
806 surl = smuggle_url(player_url, {'Referer': url})
807 return self.url_result(surl)
808 # Look for embedded (swf embed) Vimeo player
810 r'<embed[^>]+?src="((?:https?:)?//(?:www\.)?vimeo\.com/moogaloop\.swf.+?)"', webpage)
812 return self.url_result(mobj.group(1))
814 # Look for embedded YouTube player
815 matches = re.findall(r'''(?x)
824 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
828 return _playlist_from_matches(
829 matches, lambda m: unescapeHTML(m[1]))
831 # Look for lazyYT YouTube embed
832 matches = re.findall(
833 r'class="lazyYT" data-youtube-id="([^"]+)"', webpage)
835 return _playlist_from_matches(matches, lambda m: unescapeHTML(m))
837 # Look for embedded Dailymotion player
838 matches = re.findall(
839 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/embed/video/.+?)\1', webpage)
841 return _playlist_from_matches(
842 matches, lambda m: unescapeHTML(m[1]))
844 # Look for embedded Dailymotion playlist player (#3822)
846 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.[a-z]{2,3}/widget/jukebox\?.+?)\1', webpage)
848 playlists = re.findall(
849 r'list\[\]=/playlist/([^/]+)/', unescapeHTML(m.group('url')))
851 return _playlist_from_matches(
852 playlists, lambda p: '//dailymotion.com/playlist/%s' % p)
854 # Look for embedded Wistia player
856 r'<(?:meta[^>]+?content|iframe[^>]+?src)=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage)
858 embed_url = self._proto_relative_url(
859 unescapeHTML(match.group('url')))
861 '_type': 'url_transparent',
864 'uploader': video_uploader,
865 'title': video_title,
869 match = re.search(r'(?:id=["\']wistia_|data-wistia-?id=["\']|Wistia\.embed\(["\'])(?P<id>[^"\']+)', webpage)
872 '_type': 'url_transparent',
873 'url': 'http://fast.wistia.net/embed/iframe/{0:}'.format(match.group('id')),
875 'uploader': video_uploader,
876 'title': video_title,
877 'id': match.group('id')
880 # Look for embedded blip.tv player
881 mobj = re.search(r'<meta\s[^>]*https?://api\.blip\.tv/\w+/redirect/\w+/(\d+)', webpage)
883 return self.url_result('http://blip.tv/a/a-' + mobj.group(1), 'BlipTV')
884 mobj = re.search(r'<(?:iframe|embed|object)\s[^>]*(https?://(?:\w+\.)?blip\.tv/(?:play/|api\.swf#)[a-zA-Z0-9_]+)', webpage)
886 return self.url_result(mobj.group(1), 'BlipTV')
888 # Look for embedded condenast player
889 matches = re.findall(
890 r'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="(https?://player\.cnevids\.com/embed/[^"]+")',
897 'ie_key': 'CondeNast',
899 } for ma in matches],
900 'title': video_title,
904 # Look for Bandcamp pages with custom domain
905 mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage)
907 burl = unescapeHTML(mobj.group(1))
908 # Don't set the extractor because it can be a track url or an album
909 return self.url_result(burl)
911 # Look for embedded Vevo player
913 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:cache\.)?vevo\.com/.+?)\1', webpage)
915 return self.url_result(mobj.group('url'))
917 # Look for embedded Viddler player
919 r'<(?:iframe[^>]+?src|param[^>]+?value)=(["\'])(?P<url>(?:https?:)?//(?:www\.)?viddler\.com/(?:embed|player)/.+?)\1',
922 return self.url_result(mobj.group('url'))
924 # Look for Ooyala videos
925 mobj = (re.search(r'player\.ooyala\.com/[^"?]+\?[^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or
926 re.search(r'OO\.Player\.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage) or
927 re.search(r'SBN\.VideoLinkset\.ooyala\([\'"](?P<ec>.{32})[\'"]\)', webpage))
929 return OoyalaIE._build_url_result(mobj.group('ec'))
931 # Look for multiple Ooyala embeds on SBN network websites
932 mobj = re.search(r'SBN\.VideoLinkset\.entryGroup\((\[.*?\])', webpage)
934 embeds = self._parse_json(mobj.group(1), video_id, fatal=False)
936 return _playlist_from_matches(
937 embeds, getter=lambda v: OoyalaIE._url_for_embed_code(v['provider_video_id']), ie='Ooyala')
939 # Look for Aparat videos
940 mobj = re.search(r'<iframe .*?src="(http://www\.aparat\.com/video/[^"]+)"', webpage)
942 return self.url_result(mobj.group(1), 'Aparat')
944 # Look for MPORA videos
945 mobj = re.search(r'<iframe .*?src="(http://mpora\.(?:com|de)/videos/[^"]+)"', webpage)
947 return self.url_result(mobj.group(1), 'Mpora')
949 # Look for embedded NovaMov-based player
951 r'''(?x)<(?:pagespeed_)?iframe[^>]+?src=(["\'])
952 (?P<url>http://(?:(?:embed|www)\.)?
954 nowvideo\.(?:ch|sx|eu|at|ag|co)|
955 videoweed\.(?:es|com)|
956 movshare\.(?:net|sx|ag)|
957 divxstage\.(?:eu|net|ch|co|at|ag))
958 /embed\.php.+?)\1''', webpage)
960 return self.url_result(mobj.group('url'))
962 # Look for embedded Facebook player
964 r'<iframe[^>]+?src=(["\'])(?P<url>https://www\.facebook\.com/video/embed.+?)\1', webpage)
966 return self.url_result(mobj.group('url'), 'Facebook')
968 # Look for embedded VK player
969 mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://vk\.com/video_ext\.php.+?)\1', webpage)
971 return self.url_result(mobj.group('url'), 'VK')
973 # Look for embedded ivi player
974 mobj = re.search(r'<embed[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?ivi\.ru/video/player.+?)\1', webpage)
976 return self.url_result(mobj.group('url'), 'Ivi')
978 # Look for embedded Huffington Post player
980 r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed\.live\.huffingtonpost\.com/.+?)\1', webpage)
982 return self.url_result(mobj.group('url'), 'HuffPost')
985 mobj = re.search(r'class=["\']embedly-card["\'][^>]href=["\'](?P<url>[^"\']+)', webpage)
987 return self.url_result(mobj.group('url'))
988 mobj = re.search(r'class=["\']embedly-embed["\'][^>]src=["\'][^"\']*url=(?P<url>[^&]+)', webpage)
990 return self.url_result(compat_urllib_parse.unquote(mobj.group('url')))
992 # Look for funnyordie embed
993 matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage)
995 return _playlist_from_matches(
996 matches, getter=unescapeHTML, ie='FunnyOrDie')
998 # Look for BBC iPlayer embed
999 matches = re.findall(r'setPlaylist\("(https?://www\.bbc\.co\.uk/iplayer/[^/]+/[\da-z]{8})"\)', webpage)
1001 return _playlist_from_matches(matches, ie='BBCCoUk')
1003 # Look for embedded RUTV player
1004 rutv_url = RUTVIE._extract_url(webpage)
1006 return self.url_result(rutv_url, 'RUTV')
1008 # Look for embedded TED player
1010 r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed(?:-ssl)?\.ted\.com/.+?)\1', webpage)
1011 if mobj is not None:
1012 return self.url_result(mobj.group('url'), 'TED')
1014 # Look for embedded Ustream videos
1016 r'<iframe[^>]+?src=(["\'])(?P<url>http://www\.ustream\.tv/embed/.+?)\1', webpage)
1017 if mobj is not None:
1018 return self.url_result(mobj.group('url'), 'Ustream')
1020 # Look for embedded arte.tv player
1022 r'<script [^>]*?src="(?P<url>http://www\.arte\.tv/playerv2/embed[^"]+)"',
1024 if mobj is not None:
1025 return self.url_result(mobj.group('url'), 'ArteTVEmbed')
1027 # Look for embedded smotri.com player
1028 smotri_url = SmotriIE._extract_url(webpage)
1030 return self.url_result(smotri_url, 'Smotri')
1032 # Look for embeded soundcloud player
1034 r'<iframe\s+(?:[a-zA-Z0-9_-]+="[^"]+"\s+)*src="(?P<url>https?://(?:w\.)?soundcloud\.com/player[^"]+)"',
1036 if mobj is not None:
1037 url = unescapeHTML(mobj.group('url'))
1038 return self.url_result(url)
1040 # Look for embedded vulture.com player
1042 r'<iframe src="(?P<url>https?://video\.vulture\.com/[^"]+)"',
1044 if mobj is not None:
1045 url = unescapeHTML(mobj.group('url'))
1046 return self.url_result(url, ie='Vulture')
1048 # Look for embedded mtvservices player
1050 r'<iframe src="(?P<url>https?://media\.mtvnservices\.com/embed/[^"]+)"',
1052 if mobj is not None:
1053 url = unescapeHTML(mobj.group('url'))
1054 return self.url_result(url, ie='MTVServicesEmbedded')
1056 # Look for embedded yahoo player
1058 r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:screen|movies)\.yahoo\.com/.+?\.html\?format=embed)\1',
1060 if mobj is not None:
1061 return self.url_result(mobj.group('url'), 'Yahoo')
1063 # Look for embedded sbs.com.au player
1067 <meta\s+property="og:video"\s+content=|
1070 (["\'])(?P<url>https?://(?:www\.)?sbs\.com\.au/ondemand/video/.+?)\1''',
1072 if mobj is not None:
1073 return self.url_result(mobj.group('url'), 'SBS')
1075 # Look for embedded Cinchcast player
1077 r'<iframe[^>]+?src=(["\'])(?P<url>https?://player\.cinchcast\.com/.+?)\1',
1079 if mobj is not None:
1080 return self.url_result(mobj.group('url'), 'Cinchcast')
1083 r'<iframe[^>]+?src=(["\'])(?P<url>https?://m(?:lb)?\.mlb\.com/shared/video/embed/embed\.html\?.+?)\1',
1085 if mobj is not None:
1086 return self.url_result(mobj.group('url'), 'MLB')
1089 r'<iframe[^>]+?src=(["\'])(?P<url>%s)\1' % CondeNastIE.EMBED_URL,
1091 if mobj is not None:
1092 return self.url_result(self._proto_relative_url(mobj.group('url'), scheme='http:'), 'CondeNast')
1095 r'<iframe[^>]+src="(?P<url>https?://new\.livestream\.com/[^"]+/player[^"]+)"',
1097 if mobj is not None:
1098 return self.url_result(mobj.group('url'), 'Livestream')
1100 def check_video(vurl):
1101 if YoutubeIE.suitable(vurl):
1103 vpath = compat_urlparse.urlparse(vurl).path
1104 vext = determine_ext(vpath)
1105 return '.' in vpath and vext not in ('swf', 'png', 'jpg', 'srt', 'sbv', 'sub', 'vtt', 'ttml')
1107 def filter_video(urls):
1108 return list(filter(check_video, urls))
1110 # Start with something easy: JW Player in SWFObject
1111 found = filter_video(re.findall(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage))
1113 # Look for gorilla-vid style embedding
1114 found = filter_video(re.findall(r'''(?sx)
1118 jwplayer\s*\(\s*["'][^'"]+["']\s*\)\s*\.setup
1121 ['"]?file['"]?\s*:\s*["\'](.*?)["\']''', webpage))
1123 # Broaden the search a little bit
1124 found = filter_video(re.findall(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage))
1126 # Broaden the findall a little bit: JWPlayer JS loader
1127 found = filter_video(re.findall(
1128 r'[^A-Za-z0-9]?file["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage))
1131 found = filter_video(re.findall(r'''(?xs)
1132 flowplayer\("[^"]+",\s*
1134 \s*\{[^}]+? ["']?clip["']?\s*:\s*\{\s*
1135 ["']?url["']?\s*:\s*["']([^"']+)["']
1140 r"cinerama\.embedPlayer\(\s*\'[^']+\',\s*'([^']+)'", webpage)
1142 # Try to find twitter cards info
1143 found = filter_video(re.findall(
1144 r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage))
1146 # We look for Open Graph info:
1147 # We have to match any number spaces between elements, some sites try to align them (eg.: statigr.am)
1148 m_video_type = re.findall(r'<meta.*?property="og:video:type".*?content="video/(.*?)"', webpage)
1149 # We only look in og:video if the MIME type is a video, don't try if it's a Flash player:
1150 if m_video_type is not None:
1151 found = filter_video(re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage))
1154 found = re.findall(r'(?s)<video[^<]*(?:>.*?<source[^>]*)?\s+src=["\'](.*?)["\']', webpage)
1157 r'(?i)<meta\s+(?=(?:[a-z-]+="[^"]+"\s+)*http-equiv="refresh")'
1158 r'(?:[a-z-]+="[^"]+"\s+)*?content="[0-9]{,2};url=\'?([^\'"]+)',
1161 new_url = found.group(1)
1162 self.report_following_redirect(new_url)
1168 raise UnsupportedError(url)
1171 for video_url in found:
1172 video_url = compat_urlparse.urljoin(url, video_url)
1173 video_id = compat_urllib_parse.unquote(os.path.basename(video_url))
1175 # Sometimes, jwplayer extraction will result in a YouTube URL
1176 if YoutubeIE.suitable(video_url):
1177 entries.append(self.url_result(video_url, 'Youtube'))
1180 # here's a fun little line of code for you:
1181 video_id = os.path.splitext(video_id)[0]
1186 'uploader': video_uploader,
1187 'title': video_title,
1188 'age_limit': age_limit,
1191 if len(entries) == 1:
1194 for num, e in enumerate(entries, start=1):
1195 e['title'] = '%s (%d)' % (e['title'], num)
1197 '_type': 'playlist',