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,
28 from .brightcove import BrightcoveIE
29 from .ooyala import OoyalaIE
30 from .rutv import RUTVIE
31 from .smotri import SmotriIE
32 from .condenast import CondeNastIE
35 class GenericIE(InfoExtractor):
36 IE_DESC = 'Generic downloader that works on some sites'
41 'url': 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
42 'md5': '85b90ccc9d73b4acd9138d3af4c27f89',
44 'id': '13601338388002',
46 'uploader': 'www.hodiho.fr',
47 'title': 'R\u00e9gis plante sa Jeep',
50 # bandcamp page with custom domain
52 'add_ie': ['Bandcamp'],
53 'url': 'http://bronyrock.com/track/the-pony-mash',
57 'title': 'The Pony Mash',
58 'uploader': 'M_Pallante',
60 'skip': 'There is a limit of 200 free downloads / month for the test song',
62 # embedded brightcove video
63 # it also tests brightcove videos that need to set the 'Referer' in the
66 'add_ie': ['Brightcove'],
67 'url': 'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/',
69 'id': '2765128793001',
71 'title': 'Le cours de bourse : l’analyse technique',
72 'description': 'md5:7e9ad046e968cb2d1114004aba466fd9',
73 'uploader': 'BFM BUSINESS',
76 'skip_download': True,
80 # https://github.com/rg3/youtube-dl/issues/2253
81 'url': 'http://bcove.me/i6nfkrc3',
82 'md5': '0ba9446db037002366bab3b3eb30c88c',
84 'id': '3101154703001',
86 'title': 'Still no power',
87 'uploader': 'thestar.com',
88 '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.',
90 'add_ie': ['Brightcove'],
93 'url': 'http://www.championat.com/video/football/v/87/87499.html',
94 'md5': 'fb973ecf6e4a78a67453647444222983',
96 'id': '3414141473001',
98 'title': 'Видео. Удаление Дзагоева (ЦСКА)',
99 'description': 'Онлайн-трансляция матча ЦСКА - "Волга"',
100 'uploader': 'Championat',
104 # https://github.com/rg3/youtube-dl/issues/3541
105 'add_ie': ['Brightcove'],
106 'url': 'http://www.kijk.nl/sbs6/leermijvrouwenkennen/videos/jqMiXKAYan2S/aflevering-1',
108 'id': '3866516442001',
110 'title': 'Leer mij vrouwen kennen: Aflevering 1',
111 'description': 'Leer mij vrouwen kennen: Aflevering 1',
112 'uploader': 'SBS Broadcasting',
114 'skip': 'Restricted to Netherlands',
116 'skip_download': True, # m3u8 download
119 # Direct link to a video
121 'url': 'http://media.w3.org/2010/05/sintel/trailer.mp4',
122 'md5': '67d406c2bcb6af27fa886f31aa934bbe',
127 'upload_date': '20100513',
132 'url': 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219',
133 'md5': '5644c6ca5d5782c1d0d350dad9bd840c',
135 'id': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ',
137 'title': '2cc213299525360.mov', # that's what we get
142 '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',
146 'upload_date': '20130224',
147 'uploader_id': 'TheVerge',
148 'description': 'Chris Ziegler takes a look at the Alcatel OneTouch Fire and the ZTE Open; two of the first Firefox OS handsets to be officially announced.',
149 'uploader': 'The Verge',
150 'title': 'First Firefox OS phones side-by-side',
153 'skip_download': False,
158 'url': 'http://www.tested.com/science/weird/460206-tested-grinding-coffee-2000-frames-second/',
162 'title': 'Tested: Grinding Coffee at 2000 Frames Per Second',
163 'upload_date': '20140225',
164 'description': 'md5:06a40fbf30b220468f1e0957c0f558ff',
165 'uploader': 'Tested',
166 'uploader_id': 'testedcom',
168 # No need to test YoutubeIE here
170 'skip_download': True,
175 'url': 'http://www.theguardian.com/world/2014/mar/11/obama-zach-galifianakis-between-two-ferns',
179 'title': 'Between Two Ferns with Zach Galifianakis: President Barack Obama',
180 'description': 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.',
185 'url': 'http://www.rg.ru/2014/03/15/reg-dfo/anklav-anons.html',
189 'title': 'Охотское море стало целиком российским',
190 'description': 'md5:5ed62483b14663e2a95ebbe115eb8f43',
194 'skip_download': True,
199 'url': 'http://en.support.wordpress.com/videos/ted-talks/',
200 'md5': '65fdff94098e4a607385a60c5177c638',
204 'title': 'Hidden miracles of the natural world',
205 'uploader': 'Louie Schwartzberg',
206 'description': 'md5:8145d19d320ff3e52f28401f4c4283b9',
209 # Embeded Ustream video
211 'url': 'http://www.american.edu/spa/pti/nsa-privacy-janus-2014.cfm',
212 'md5': '27b99cdb639c9b12a79bca876a073417',
216 'uploader': 'AU SPA: The NSA and Privacy',
217 'title': 'NSA and Privacy Forum Debate featuring General Hayden and Barton Gellman'
220 # nowvideo embed hidden behind percent encoding
222 'url': 'http://www.waoanime.tv/the-super-dimension-fortress-macross-episode-1/',
223 'md5': '2baf4ddd70f697d94b1c18cf796d5107',
225 'id': '06e53103ca9aa',
227 'title': 'Macross Episode 001 Watch Macross Episode 001 onl',
228 'description': 'No description',
233 'url': 'http://www.tv-replay.fr/redirection/20-03-14/x-enius-arte-10753389.html',
234 'md5': '7653032cbb25bf6c80d80f217055fa43',
236 'id': '048195-004_PLUS7-F',
239 'description': 'md5:d5fdf32ef6613cdbfd516ae658abf168',
240 'upload_date': '20140320',
243 'skip_download': 'Requires rtmpdump'
248 'url': 'http://www.wired.com/2014/04/honda-asimo/',
249 'md5': 'ba0dfe966fa007657bd1443ee672db0f',
251 'id': '53501be369702d3275860000',
253 'title': 'Honda’s New Asimo Robot Is More Human Than Ever',
258 'url': 'http://www.spi0n.com/zap-spi0n-com-n216/',
259 'md5': '441aeeb82eb72c422c7f14ec533999cd',
261 'id': 'k2mm4bCdJ6CQ2i7c8o2',
263 'title': 'Le Zap de Spi0n n°216 - Zapping du Web',
266 'add_ie': ['Dailymotion'],
270 'url': 'http://www.badzine.de/ansicht/datum/2014/06/09/so-funktioniert-die-neue-englische-badminton-liga.html',
274 'title': 'The NBL Auction 2014',
275 'uploader': 'BADMINTON England',
276 'uploader_id': 'BADMINTONEvents',
277 'upload_date': '20140603',
278 'description': 'md5:9ef128a69f1e262a700ed83edb163a73',
280 'add_ie': ['Youtube'],
282 'skip_download': True,
287 'url': 'http://www.gametrailers.com/news-post/76093/north-america-europe-is-getting-that-mario-kart-8-mercedes-dlc-too',
288 'md5': '35727f82f58c76d996fc188f9755b0d5',
290 'id': '0306a69b-8adf-4fb5-aace-75f8e8cbfca9',
293 'description': 'Mario\'s life in the fast lane has never looked so good.',
296 # YouTube embed via <data-embed-url="">
298 'url': 'https://play.google.com/store/apps/details?id=com.gameloft.android.ANMP.GloftA8HM',
302 'title': 'Asphalt 8: Airborne - Update - Welcome to Dubai!',
303 'uploader': 'Gameloft',
304 'uploader_id': 'gameloft',
305 'upload_date': '20140828',
306 'description': 'md5:c80da9ed3d83ae6d1876c834de03e1c4',
309 'skip_download': True,
314 'url': 'http://www.ll.mit.edu/workshops/education/videocourses/antennas/lecture1/video/',
316 'md5': '0c5e352edabf715d762b0ad4e6d9ee67',
318 'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
319 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - video1',
324 'md5': '10e4bb3aaca9fd630e273ff92d9f3c63',
326 'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final_PIP',
327 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - pip',
333 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
338 'url': 'http://www.handjobhub.com/video/busty-blonde-siri-tit-fuck-while-wank-6313.html',
339 'md5': '9d65602bf31c6e20014319c7d07fba27',
341 'id': '5123ea6d5e5a7',
344 'uploader': 'www.handjobhub.com',
345 'title': 'Busty Blonde Siri Tit Fuck While Wank at HandjobHub.com',
350 'url': 'http://phihag.de/2014/youtube-dl/rss2.xml',
352 'id': 'http://phihag.de/2014/youtube-dl/rss2.xml',
353 'title': 'Zero Punctuation',
356 'playlist_mincount': 11,
358 # Multiple brightcove videos
359 # https://github.com/rg3/youtube-dl/issues/2283
361 'url': 'http://www.newyorker.com/online/blogs/newsdesk/2014/01/always-never-nuclear-command-and-control.html',
363 'id': 'always-never',
364 'title': 'Always / Never - The New Yorker',
368 'extract_flat': False,
369 'skip_download': True,
374 'url': 'http://umpire-empire.com/index.php/topic/58125-laz-decides-no-thats-low/',
375 'md5': '96f09a37e44da40dd083e12d9a683327',
379 'title': 'Ump changes call to ball',
380 'description': 'md5:71c11215384298a172a6dcb4c2e20685',
382 'timestamp': 1401537900,
383 'upload_date': '20140531',
384 'thumbnail': 're:^https?://.*\.jpg$',
389 'url': 'http://education-portal.com/academy/lesson/north-american-exploration-failed-colonies-of-spain-france-england.html#lesson',
390 'md5': '8788b683c777a5cf25621eaf286d0c23',
394 'title': 'md5:51364a8d3d009997ba99656004b5e20d',
396 'filesize': 182808282,
397 'uploader': 'education-portal.com',
401 'url': 'http://thoughtworks.wistia.com/medias/uxjb0lwrcz',
402 'md5': 'baf49c2baa8a7de5f3fc145a8506dcd4',
406 'title': 'Conversation about Hexagonal Rails Part 1 - ThoughtWorks',
408 'uploader': 'thoughtworks.wistia.com',
411 # Direct download with broken HEAD
413 'url': 'http://ai-radio.org:8000/radio.opus',
420 'skip_download': True, # infinite live stream
422 'expected_warnings': [
423 r'501.*Not Implemented'
428 'url': 'http://nakedsecurity.sophos.com/2014/10/29/sscc-171-are-you-sure-that-1234-is-a-bad-password-podcast/',
432 'description': 'md5:ff867d6b555488ad3c52572bb33d432c',
433 'uploader': 'Sophos Security',
434 'title': 'Chet Chat 171 - Oct 29, 2014',
435 'upload_date': '20141029',
440 'url': 'http://www.esa.int/Our_Activities/Space_Science/Rosetta/Philae_comet_touch-down_webcast',
444 'upload_date': '20141112',
445 'title': 'Rosetta #CometLanding webcast HL 10',
450 'url': 'http://discourse.ubuntu.com/t/unity-8-desktop-mode-windows-on-mir/1986',
452 'title': 'Unity 8 desktop-mode windows on Mir! - Ubuntu Discourse',
454 'playlist_mincount': 2,
456 # Direct link with incorrect MIME type
458 'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
459 'md5': '4ccbebe5f36706d85221f204d7eb5913',
461 'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
462 'id': '5_Lennart_Poettering_-_Systemd',
464 'title': '5_Lennart_Poettering_-_Systemd',
465 'upload_date': '20141120',
467 'expected_warnings': [
468 'URL could be a direct video link, returning it as such.'
473 'url': 'http://undergroundwellness.com/podcasts/306-5-steps-to-permanent-gut-healing/',
477 'upload_date': '20141126',
478 'title': 'Jack Tips: 5 Steps to Permanent Gut Healing',
483 def report_following_redirect(self, new_url):
484 """Report information extraction."""
485 self._downloader.to_screen('[redirect] Following redirect to %s' % new_url)
487 def _extract_rss(self, url, video_id, doc):
488 playlist_title = doc.find('./channel/title').text
489 playlist_desc_el = doc.find('./channel/description')
490 playlist_desc = None if playlist_desc_el is None else playlist_desc_el.text
494 'url': e.find('link').text,
495 'title': e.find('title').text,
496 } for e in doc.findall('./channel/item')]
501 'title': playlist_title,
502 'description': playlist_desc,
506 def _extract_camtasia(self, url, video_id, webpage):
507 """ Returns None if no camtasia video can be found. """
509 camtasia_cfg = self._search_regex(
510 r'fo\.addVariable\(\s*"csConfigFile",\s*"([^"]+)"\s*\);',
511 webpage, 'camtasia configuration file', default=None)
512 if camtasia_cfg is None:
515 title = self._html_search_meta('DC.title', webpage, fatal=True)
517 camtasia_url = compat_urlparse.urljoin(url, camtasia_cfg)
518 camtasia_cfg = self._download_xml(
519 camtasia_url, video_id,
520 note='Downloading camtasia configuration',
521 errnote='Failed to download camtasia configuration')
522 fileset_node = camtasia_cfg.find('./playlist/array/fileset')
525 for n in fileset_node.getchildren():
526 url_n = n.find('./uri')
531 'id': os.path.splitext(url_n.text.rpartition('/')[2])[0],
532 'title': '%s - %s' % (title, n.tag),
533 'url': compat_urlparse.urljoin(url, url_n.text),
534 'duration': float_or_none(n.find('./duration').text),
543 def _real_extract(self, url):
544 if url.startswith('//'):
547 'url': self.http_scheme() + url,
550 parsed_url = compat_urlparse.urlparse(url)
551 if not parsed_url.scheme:
552 default_search = self._downloader.params.get('default_search')
553 if default_search is None:
554 default_search = 'fixup_error'
556 if default_search in ('auto', 'auto_warning', 'fixup_error'):
558 self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http')
559 return self.url_result('http://' + url)
560 elif default_search != 'fixup_error':
561 if default_search == 'auto_warning':
562 if re.match(r'^(?:url|URL)$', url):
563 raise ExtractorError(
564 'Invalid URL: %r . Call youtube-dl like this: youtube-dl -v "https://www.youtube.com/watch?v=BaW_jenozKc" ' % url,
567 self._downloader.report_warning(
568 'Falling back to youtube search for %s . Set --default-search "auto" to suppress this warning.' % url)
569 return self.url_result('ytsearch:' + url)
571 if default_search in ('error', 'fixup_error'):
572 raise ExtractorError(
573 '%r is not a valid URL. '
574 'Set --default-search "ytsearch" (or run youtube-dl "ytsearch:%s" ) to search YouTube'
575 % (url, url), expected=True)
577 if ':' not in default_search:
578 default_search += ':'
579 return self.url_result(default_search + url)
581 url, smuggled_data = unsmuggle_url(url)
583 is_intentional = smuggled_data and smuggled_data.get('to_generic')
584 if smuggled_data and 'force_videoid' in smuggled_data:
585 force_videoid = smuggled_data['force_videoid']
586 video_id = force_videoid
588 video_id = os.path.splitext(url.rstrip('/').split('/')[-1])[0]
590 self.to_screen('%s: Requesting header' % video_id)
592 head_req = HEADRequest(url)
593 head_response = self._request_webpage(
595 note=False, errnote='Could not send HEAD request to %s' % url,
598 if head_response is not False:
600 new_url = head_response.geturl()
602 self.report_following_redirect(new_url)
604 new_url = smuggle_url(
605 new_url, {'force_videoid': force_videoid})
606 return self.url_result(new_url)
609 if head_response is False:
610 full_response = self._request_webpage(url, video_id)
611 head_response = full_response
613 # Check for direct link to a video
614 content_type = head_response.headers.get('Content-Type', '')
615 m = re.match(r'^(?P<type>audio|video|application(?=/ogg$))/(?P<format_id>.+)$', content_type)
617 upload_date = unified_strdate(
618 head_response.headers.get('Last-Modified'))
621 'title': os.path.splitext(url_basename(url))[0],
624 'format_id': m.group('format_id'),
626 'vcodec': 'none' if m.group('type') == 'audio' else None
628 'upload_date': upload_date,
631 if not self._downloader.params.get('test', False) and not is_intentional:
632 self._downloader.report_warning('Falling back on generic information extractor.')
634 if not full_response:
635 full_response = self._request_webpage(url, video_id)
637 # Maybe it's a direct link to a video?
638 # Be careful not to download the whole thing!
639 first_bytes = full_response.read(512)
640 if not re.match(r'^\s*<', first_bytes.decode('utf-8', 'replace')):
641 self._downloader.report_warning(
642 'URL could be a direct video link, returning it as such.')
643 upload_date = unified_strdate(
644 head_response.headers.get('Last-Modified'))
647 'title': os.path.splitext(url_basename(url))[0],
650 'upload_date': upload_date,
653 webpage = self._webpage_read_content(
654 full_response, url, video_id, prefix=first_bytes)
656 self.report_extraction(video_id)
660 doc = parse_xml(webpage)
662 return self._extract_rss(url, video_id, doc)
663 except compat_xml_parse_error:
666 # Is it a Camtasia project?
667 camtasia_res = self._extract_camtasia(url, video_id, webpage)
668 if camtasia_res is not None:
671 # Sometimes embedded video player is hidden behind percent encoding
672 # (e.g. https://github.com/rg3/youtube-dl/issues/2448)
673 # Unescaping the whole page allows to handle those cases in a generic way
674 webpage = compat_urllib_parse.unquote(webpage)
676 # it's tempting to parse this further, but you would
677 # have to take into account all the variations like
678 # Video Title - Site Name
679 # Site Name | Video Title
680 # Video Title - Tagline | Site Name
681 # and so on and so forth; it's just not practical
682 video_title = self._html_search_regex(
683 r'(?s)<title>(.*?)</title>', webpage, 'video title',
686 # Try to detect age limit automatically
687 age_limit = self._rta_search(webpage)
688 # And then there are the jokers who advertise that they use RTA,
689 # but actually don't.
690 AGE_LIMIT_MARKERS = [
691 r'Proudly Labeled <a href="http://www.rtalabel.org/" title="Restricted to Adults">RTA</a>',
693 if any(re.search(marker, webpage) for marker in AGE_LIMIT_MARKERS):
696 # video uploader is domain name
697 video_uploader = self._search_regex(
698 r'^(?:https?://)?([^/]*)/.*', url, 'video uploader')
701 def _playlist_from_matches(matches, getter, ie=None):
703 self.url_result(self._proto_relative_url(getter(m)), ie)
705 return self.playlist_result(
706 urlrs, playlist_id=video_id, playlist_title=video_title)
708 # Look for BrightCove:
709 bc_urls = BrightcoveIE._extract_brightcove_urls(webpage)
711 self.to_screen('Brightcove video detected.')
714 'url': smuggle_url(bc_url, {'Referer': url}),
715 'ie_key': 'Brightcove'
716 } for bc_url in bc_urls]
720 'title': video_title,
725 # Look for embedded (iframe) Vimeo player
727 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//player\.vimeo\.com/video/.+?)\1', webpage)
729 player_url = unescapeHTML(mobj.group('url'))
730 surl = smuggle_url(player_url, {'Referer': url})
731 return self.url_result(surl)
733 # Look for embedded (swf embed) Vimeo player
735 r'<embed[^>]+?src="((?:https?:)?//(?:www\.)?vimeo\.com/moogaloop\.swf.+?)"', webpage)
737 return self.url_result(mobj.group(1))
739 # Look for embedded YouTube player
740 matches = re.findall(r'''(?x)
749 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
753 return _playlist_from_matches(
754 matches, lambda m: unescapeHTML(m[1]))
756 # Look for lazyYT YouTube embed
757 matches = re.findall(
758 r'class="lazyYT" data-youtube-id="([^"]+)"', webpage)
760 return _playlist_from_matches(matches, lambda m: unescapeHTML(m))
762 # Look for embedded Dailymotion player
763 matches = re.findall(
764 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/embed/video/.+?)\1', webpage)
766 return _playlist_from_matches(
767 matches, lambda m: unescapeHTML(m[1]))
769 # Look for embedded Dailymotion playlist player (#3822)
771 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.[a-z]{2,3}/widget/jukebox\?.+?)\1', webpage)
773 playlists = re.findall(
774 r'list\[\]=/playlist/([^/]+)/', unescapeHTML(m.group('url')))
776 return _playlist_from_matches(
777 playlists, lambda p: '//dailymotion.com/playlist/%s' % p)
779 # Look for embedded Wistia player
781 r'<(?:meta[^>]+?content|iframe[^>]+?src)=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage)
783 embed_url = self._proto_relative_url(
784 unescapeHTML(match.group('url')))
786 '_type': 'url_transparent',
789 'uploader': video_uploader,
790 'title': video_title,
794 match = re.search(r'(?:id=["\']wistia_|data-wistia-?id=["\']|Wistia\.embed\(["\'])(?P<id>[^"\']+)', webpage)
797 '_type': 'url_transparent',
798 'url': 'http://fast.wistia.net/embed/iframe/{0:}'.format(match.group('id')),
800 'uploader': video_uploader,
801 'title': video_title,
802 'id': match.group('id')
805 # Look for embedded blip.tv player
806 mobj = re.search(r'<meta\s[^>]*https?://api\.blip\.tv/\w+/redirect/\w+/(\d+)', webpage)
808 return self.url_result('http://blip.tv/a/a-' + mobj.group(1), 'BlipTV')
809 mobj = re.search(r'<(?:iframe|embed|object)\s[^>]*(https?://(?:\w+\.)?blip\.tv/(?:play/|api\.swf#)[a-zA-Z0-9_]+)', webpage)
811 return self.url_result(mobj.group(1), 'BlipTV')
813 # Look for embedded condenast player
814 matches = re.findall(
815 r'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="(https?://player\.cnevids\.com/embed/[^"]+")',
822 'ie_key': 'CondeNast',
824 } for ma in matches],
825 'title': video_title,
829 # Look for Bandcamp pages with custom domain
830 mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage)
832 burl = unescapeHTML(mobj.group(1))
833 # Don't set the extractor because it can be a track url or an album
834 return self.url_result(burl)
836 # Look for embedded Vevo player
838 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:cache\.)?vevo\.com/.+?)\1', webpage)
840 return self.url_result(mobj.group('url'))
842 # Look for Ooyala videos
843 mobj = (re.search(r'player.ooyala.com/[^"?]+\?[^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or
844 re.search(r'OO.Player.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage))
846 return OoyalaIE._build_url_result(mobj.group('ec'))
848 # Look for Aparat videos
849 mobj = re.search(r'<iframe .*?src="(http://www\.aparat\.com/video/[^"]+)"', webpage)
851 return self.url_result(mobj.group(1), 'Aparat')
853 # Look for MPORA videos
854 mobj = re.search(r'<iframe .*?src="(http://mpora\.(?:com|de)/videos/[^"]+)"', webpage)
856 return self.url_result(mobj.group(1), 'Mpora')
858 # Look for embedded NovaMov-based player
860 r'''(?x)<(?:pagespeed_)?iframe[^>]+?src=(["\'])
861 (?P<url>http://(?:(?:embed|www)\.)?
863 nowvideo\.(?:ch|sx|eu|at|ag|co)|
864 videoweed\.(?:es|com)|
865 movshare\.(?:net|sx|ag)|
866 divxstage\.(?:eu|net|ch|co|at|ag))
867 /embed\.php.+?)\1''', webpage)
869 return self.url_result(mobj.group('url'))
871 # Look for embedded Facebook player
873 r'<iframe[^>]+?src=(["\'])(?P<url>https://www\.facebook\.com/video/embed.+?)\1', webpage)
875 return self.url_result(mobj.group('url'), 'Facebook')
877 # Look for embedded VK player
878 mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://vk\.com/video_ext\.php.+?)\1', webpage)
880 return self.url_result(mobj.group('url'), 'VK')
882 # Look for embedded ivi player
883 mobj = re.search(r'<embed[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?ivi\.ru/video/player.+?)\1', webpage)
885 return self.url_result(mobj.group('url'), 'Ivi')
887 # Look for embedded Huffington Post player
889 r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed\.live\.huffingtonpost\.com/.+?)\1', webpage)
891 return self.url_result(mobj.group('url'), 'HuffPost')
894 mobj = re.search(r'class=["\']embedly-card["\'][^>]href=["\'](?P<url>[^"\']+)', webpage)
896 return self.url_result(mobj.group('url'))
897 mobj = re.search(r'class=["\']embedly-embed["\'][^>]src=["\'][^"\']*url=(?P<url>[^&]+)', webpage)
899 return self.url_result(compat_urllib_parse.unquote(mobj.group('url')))
901 # Look for funnyordie embed
902 matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage)
904 return _playlist_from_matches(
905 matches, getter=unescapeHTML, ie='FunnyOrDie')
907 # Look for embedded RUTV player
908 rutv_url = RUTVIE._extract_url(webpage)
910 return self.url_result(rutv_url, 'RUTV')
912 # Look for embedded TED player
914 r'<iframe[^>]+?src=(["\'])(?P<url>http://embed\.ted\.com/.+?)\1', webpage)
916 return self.url_result(mobj.group('url'), 'TED')
918 # Look for embedded Ustream videos
920 r'<iframe[^>]+?src=(["\'])(?P<url>http://www\.ustream\.tv/embed/.+?)\1', webpage)
922 return self.url_result(mobj.group('url'), 'Ustream')
924 # Look for embedded arte.tv player
926 r'<script [^>]*?src="(?P<url>http://www\.arte\.tv/playerv2/embed[^"]+)"',
929 return self.url_result(mobj.group('url'), 'ArteTVEmbed')
931 # Look for embedded smotri.com player
932 smotri_url = SmotriIE._extract_url(webpage)
934 return self.url_result(smotri_url, 'Smotri')
936 # Look for embeded soundcloud player
938 r'<iframe\s+(?:[a-zA-Z0-9_-]+="[^"]+"\s+)*src="(?P<url>https?://(?:w\.)?soundcloud\.com/player[^"]+)"',
941 url = unescapeHTML(mobj.group('url'))
942 return self.url_result(url)
944 # Look for embedded vulture.com player
946 r'<iframe src="(?P<url>https?://video\.vulture\.com/[^"]+)"',
949 url = unescapeHTML(mobj.group('url'))
950 return self.url_result(url, ie='Vulture')
952 # Look for embedded mtvservices player
954 r'<iframe src="(?P<url>https?://media\.mtvnservices\.com/embed/[^"]+)"',
957 url = unescapeHTML(mobj.group('url'))
958 return self.url_result(url, ie='MTVServicesEmbedded')
960 # Look for embedded yahoo player
962 r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:screen|movies)\.yahoo\.com/.+?\.html\?format=embed)\1',
965 return self.url_result(mobj.group('url'), 'Yahoo')
967 # Look for embedded sbs.com.au player
969 r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:www\.)sbs\.com\.au/ondemand/video/single/.+?)\1',
972 return self.url_result(mobj.group('url'), 'SBS')
974 # Look for embedded Cinchcast player
976 r'<iframe[^>]+?src=(["\'])(?P<url>https?://player\.cinchcast\.com/.+?)\1',
979 return self.url_result(mobj.group('url'), 'Cinchcast')
982 r'<iframe[^>]+?src=(["\'])(?P<url>https?://m(?:lb)?\.mlb\.com/shared/video/embed/embed\.html\?.+?)\1',
985 return self.url_result(mobj.group('url'), 'MLB')
988 r'<iframe[^>]+?src=(["\'])(?P<url>%s)\1' % CondeNastIE.EMBED_URL,
991 return self.url_result(self._proto_relative_url(mobj.group('url'), scheme='http:'), 'CondeNast')
994 r'<iframe[^>]+src="(?P<url>https?://new\.livestream\.com/[^"]+/player[^"]+)"',
997 return self.url_result(mobj.group('url'), 'Livestream')
999 def check_video(vurl):
1000 vpath = compat_urlparse.urlparse(vurl).path
1001 vext = determine_ext(vpath)
1002 return '.' in vpath and vext not in ('swf', 'png', 'jpg', 'srt', 'sbv', 'sub', 'vtt', 'ttml')
1004 def filter_video(urls):
1005 return list(filter(check_video, urls))
1007 # Start with something easy: JW Player in SWFObject
1008 found = filter_video(re.findall(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage))
1010 # Look for gorilla-vid style embedding
1011 found = filter_video(re.findall(r'''(?sx)
1015 jwplayer\s*\(\s*["'][^'"]+["']\s*\)\s*\.setup
1017 .*?file\s*:\s*["\'](.*?)["\']''', webpage))
1019 # Broaden the search a little bit
1020 found = filter_video(re.findall(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage))
1022 # Broaden the findall a little bit: JWPlayer JS loader
1023 found = filter_video(re.findall(
1024 r'[^A-Za-z0-9]?file["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage))
1027 found = filter_video(re.findall(r'''(?xs)
1028 flowplayer\("[^"]+",\s*
1030 \s*{[^}]+? ["']?clip["']?\s*:\s*\{\s*
1031 ["']?url["']?\s*:\s*["']([^"']+)["']
1034 # Try to find twitter cards info
1035 found = filter_video(re.findall(
1036 r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage))
1038 # We look for Open Graph info:
1039 # We have to match any number spaces between elements, some sites try to align them (eg.: statigr.am)
1040 m_video_type = re.findall(r'<meta.*?property="og:video:type".*?content="video/(.*?)"', webpage)
1041 # We only look in og:video if the MIME type is a video, don't try if it's a Flash player:
1042 if m_video_type is not None:
1043 found = filter_video(re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage))
1046 found = re.findall(r'(?s)<video[^<]*(?:>.*?<source[^>]*)?\s+src=["\'](.*?)["\']', webpage)
1049 r'(?i)<meta\s+(?=(?:[a-z-]+="[^"]+"\s+)*http-equiv="refresh")'
1050 r'(?:[a-z-]+="[^"]+"\s+)*?content="[0-9]{,2};url=\'?([^\'"]+)',
1053 new_url = found.group(1)
1054 self.report_following_redirect(new_url)
1060 raise ExtractorError('Unsupported URL: %s' % url)
1063 for video_url in found:
1064 video_url = compat_urlparse.urljoin(url, video_url)
1065 video_id = compat_urllib_parse.unquote(os.path.basename(video_url))
1067 # Sometimes, jwplayer extraction will result in a YouTube URL
1068 if YoutubeIE.suitable(video_url):
1069 entries.append(self.url_result(video_url, 'Youtube'))
1072 # here's a fun little line of code for you:
1073 video_id = os.path.splitext(video_id)[0]
1078 'uploader': video_uploader,
1079 'title': video_title,
1080 'age_limit': age_limit,
1083 if len(entries) == 1:
1086 for num, e in enumerate(entries, start=1):
1087 e['title'] = '%s (%d)' % (e['title'], num)
1089 '_type': 'playlist',