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 'add_ie': ['Brightcove'],
105 'url': 'http://www.kijk.nl/sbs6/leermijvrouwenkennen/videos/jqMiXKAYan2S/aflevering-1',
107 'id': '3866516442001',
109 'title': 'Leer mij vrouwen kennen: Aflevering 1',
110 'description': 'Leer mij vrouwen kennen: Aflevering 1',
111 'uploader': 'SBS Broadcasting',
114 'skip_download': True,
117 # Direct link to a video
119 'url': 'http://media.w3.org/2010/05/sintel/trailer.mp4',
120 'md5': '67d406c2bcb6af27fa886f31aa934bbe',
125 'upload_date': '20100513',
130 'url': 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219',
131 'md5': '5644c6ca5d5782c1d0d350dad9bd840c',
133 'id': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ',
135 'title': '2cc213299525360.mov', # that's what we get
140 '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',
144 'upload_date': '20130224',
145 'uploader_id': 'TheVerge',
146 '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.',
147 'uploader': 'The Verge',
148 'title': 'First Firefox OS phones side-by-side',
151 'skip_download': False,
156 'url': 'http://www.tested.com/science/weird/460206-tested-grinding-coffee-2000-frames-second/',
160 'title': 'Tested: Grinding Coffee at 2000 Frames Per Second',
161 'upload_date': '20140225',
162 'description': 'md5:06a40fbf30b220468f1e0957c0f558ff',
163 'uploader': 'Tested',
164 'uploader_id': 'testedcom',
166 # No need to test YoutubeIE here
168 'skip_download': True,
173 'url': 'http://www.theguardian.com/world/2014/mar/11/obama-zach-galifianakis-between-two-ferns',
177 'title': 'Between Two Ferns with Zach Galifianakis: President Barack Obama',
178 'description': 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.',
183 'url': 'http://www.rg.ru/2014/03/15/reg-dfo/anklav-anons.html',
187 'title': 'Охотское море стало целиком российским',
188 'description': 'md5:5ed62483b14663e2a95ebbe115eb8f43',
192 'skip_download': True,
197 'url': 'http://en.support.wordpress.com/videos/ted-talks/',
198 'md5': '65fdff94098e4a607385a60c5177c638',
202 'title': 'Hidden miracles of the natural world',
203 'uploader': 'Louie Schwartzberg',
204 'description': 'md5:8145d19d320ff3e52f28401f4c4283b9',
207 # Embeded Ustream video
209 'url': 'http://www.american.edu/spa/pti/nsa-privacy-janus-2014.cfm',
210 'md5': '27b99cdb639c9b12a79bca876a073417',
214 'uploader': 'AU SPA: The NSA and Privacy',
215 'title': 'NSA and Privacy Forum Debate featuring General Hayden and Barton Gellman'
218 # nowvideo embed hidden behind percent encoding
220 'url': 'http://www.waoanime.tv/the-super-dimension-fortress-macross-episode-1/',
221 'md5': '2baf4ddd70f697d94b1c18cf796d5107',
223 'id': '06e53103ca9aa',
225 'title': 'Macross Episode 001 Watch Macross Episode 001 onl',
226 'description': 'No description',
231 'url': 'http://www.tv-replay.fr/redirection/20-03-14/x-enius-arte-10753389.html',
232 'md5': '7653032cbb25bf6c80d80f217055fa43',
234 'id': '048195-004_PLUS7-F',
237 'description': 'md5:d5fdf32ef6613cdbfd516ae658abf168',
238 'upload_date': '20140320',
241 'skip_download': 'Requires rtmpdump'
246 'url': 'http://www.wired.com/2014/04/honda-asimo/',
247 'md5': 'ba0dfe966fa007657bd1443ee672db0f',
249 'id': '53501be369702d3275860000',
251 'title': 'Honda’s New Asimo Robot Is More Human Than Ever',
256 'url': 'http://www.spi0n.com/zap-spi0n-com-n216/',
257 'md5': '441aeeb82eb72c422c7f14ec533999cd',
259 'id': 'k2mm4bCdJ6CQ2i7c8o2',
261 'title': 'Le Zap de Spi0n n°216 - Zapping du Web',
264 'add_ie': ['Dailymotion'],
268 'url': 'http://www.badzine.de/ansicht/datum/2014/06/09/so-funktioniert-die-neue-englische-badminton-liga.html',
272 'title': 'The NBL Auction 2014',
273 'uploader': 'BADMINTON England',
274 'uploader_id': 'BADMINTONEvents',
275 'upload_date': '20140603',
276 'description': 'md5:9ef128a69f1e262a700ed83edb163a73',
278 'add_ie': ['Youtube'],
280 'skip_download': True,
285 'url': 'http://www.gametrailers.com/news-post/76093/north-america-europe-is-getting-that-mario-kart-8-mercedes-dlc-too',
286 'md5': '35727f82f58c76d996fc188f9755b0d5',
288 'id': '0306a69b-8adf-4fb5-aace-75f8e8cbfca9',
291 'description': 'Mario\'s life in the fast lane has never looked so good.',
294 # YouTube embed via <data-embed-url="">
296 'url': 'https://play.google.com/store/apps/details?id=com.gameloft.android.ANMP.GloftA8HM',
300 'title': 'Asphalt 8: Airborne - Update - Welcome to Dubai!',
301 'uploader': 'Gameloft',
302 'uploader_id': 'gameloft',
303 'upload_date': '20140828',
304 'description': 'md5:c80da9ed3d83ae6d1876c834de03e1c4',
307 'skip_download': True,
312 'url': 'http://www.ll.mit.edu/workshops/education/videocourses/antennas/lecture1/video/',
314 'md5': '0c5e352edabf715d762b0ad4e6d9ee67',
316 'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
317 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - video1',
322 'md5': '10e4bb3aaca9fd630e273ff92d9f3c63',
324 'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final_PIP',
325 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - pip',
331 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
336 'url': 'http://www.handjobhub.com/video/busty-blonde-siri-tit-fuck-while-wank-6313.html',
337 'md5': '9d65602bf31c6e20014319c7d07fba27',
339 'id': '5123ea6d5e5a7',
342 'uploader': 'www.handjobhub.com',
343 'title': 'Busty Blonde Siri Tit Fuck While Wank at HandjobHub.com',
348 'url': 'http://phihag.de/2014/youtube-dl/rss2.xml',
350 'id': 'http://phihag.de/2014/youtube-dl/rss2.xml',
351 'title': 'Zero Punctuation',
354 'playlist_mincount': 11,
356 # Multiple brightcove videos
357 # https://github.com/rg3/youtube-dl/issues/2283
359 'url': 'http://www.newyorker.com/online/blogs/newsdesk/2014/01/always-never-nuclear-command-and-control.html',
361 'id': 'always-never',
362 'title': 'Always / Never - The New Yorker',
366 'extract_flat': False,
367 'skip_download': True,
372 'url': 'http://umpire-empire.com/index.php/topic/58125-laz-decides-no-thats-low/',
373 'md5': '96f09a37e44da40dd083e12d9a683327',
377 'title': 'Ump changes call to ball',
378 'description': 'md5:71c11215384298a172a6dcb4c2e20685',
380 'timestamp': 1401537900,
381 'upload_date': '20140531',
382 'thumbnail': 're:^https?://.*\.jpg$',
387 'url': 'http://education-portal.com/academy/lesson/north-american-exploration-failed-colonies-of-spain-france-england.html#lesson',
388 'md5': '8788b683c777a5cf25621eaf286d0c23',
392 'title': 'md5:51364a8d3d009997ba99656004b5e20d',
394 'filesize': 182808282,
395 'uploader': 'education-portal.com',
399 'url': 'http://thoughtworks.wistia.com/medias/uxjb0lwrcz',
400 'md5': 'baf49c2baa8a7de5f3fc145a8506dcd4',
404 'title': 'Conversation about Hexagonal Rails Part 1 - ThoughtWorks',
406 'uploader': 'thoughtworks.wistia.com',
409 # Direct download with broken HEAD
411 'url': 'http://ai-radio.org:8000/radio.opus',
418 'skip_download': True, # infinite live stream
420 'expected_warnings': [
421 r'501.*Not Implemented'
426 'url': 'http://nakedsecurity.sophos.com/2014/10/29/sscc-171-are-you-sure-that-1234-is-a-bad-password-podcast/',
430 'description': 'md5:ff867d6b555488ad3c52572bb33d432c',
431 'uploader': 'Sophos Security',
432 'title': 'Chet Chat 171 - Oct 29, 2014',
433 'upload_date': '20141029',
438 def report_following_redirect(self, new_url):
439 """Report information extraction."""
440 self._downloader.to_screen('[redirect] Following redirect to %s' % new_url)
442 def _extract_rss(self, url, video_id, doc):
443 playlist_title = doc.find('./channel/title').text
444 playlist_desc_el = doc.find('./channel/description')
445 playlist_desc = None if playlist_desc_el is None else playlist_desc_el.text
449 'url': e.find('link').text,
450 'title': e.find('title').text,
451 } for e in doc.findall('./channel/item')]
456 'title': playlist_title,
457 'description': playlist_desc,
461 def _extract_camtasia(self, url, video_id, webpage):
462 """ Returns None if no camtasia video can be found. """
464 camtasia_cfg = self._search_regex(
465 r'fo\.addVariable\(\s*"csConfigFile",\s*"([^"]+)"\s*\);',
466 webpage, 'camtasia configuration file', default=None)
467 if camtasia_cfg is None:
470 title = self._html_search_meta('DC.title', webpage, fatal=True)
472 camtasia_url = compat_urlparse.urljoin(url, camtasia_cfg)
473 camtasia_cfg = self._download_xml(
474 camtasia_url, video_id,
475 note='Downloading camtasia configuration',
476 errnote='Failed to download camtasia configuration')
477 fileset_node = camtasia_cfg.find('./playlist/array/fileset')
480 for n in fileset_node.getchildren():
481 url_n = n.find('./uri')
486 'id': os.path.splitext(url_n.text.rpartition('/')[2])[0],
487 'title': '%s - %s' % (title, n.tag),
488 'url': compat_urlparse.urljoin(url, url_n.text),
489 'duration': float_or_none(n.find('./duration').text),
498 def _real_extract(self, url):
499 if url.startswith('//'):
502 'url': self.http_scheme() + url,
505 parsed_url = compat_urlparse.urlparse(url)
506 if not parsed_url.scheme:
507 default_search = self._downloader.params.get('default_search')
508 if default_search is None:
509 default_search = 'fixup_error'
511 if default_search in ('auto', 'auto_warning', 'fixup_error'):
513 self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http')
514 return self.url_result('http://' + url)
515 elif default_search != 'fixup_error':
516 if default_search == 'auto_warning':
517 if re.match(r'^(?:url|URL)$', url):
518 raise ExtractorError(
519 'Invalid URL: %r . Call youtube-dl like this: youtube-dl -v "https://www.youtube.com/watch?v=BaW_jenozKc" ' % url,
522 self._downloader.report_warning(
523 'Falling back to youtube search for %s . Set --default-search "auto" to suppress this warning.' % url)
524 return self.url_result('ytsearch:' + url)
526 if default_search in ('error', 'fixup_error'):
527 raise ExtractorError(
528 ('%r is not a valid URL. '
529 'Set --default-search "ytsearch" (or run youtube-dl "ytsearch:%s" ) to search YouTube'
530 ) % (url, url), expected=True)
532 if ':' not in default_search:
533 default_search += ':'
534 return self.url_result(default_search + url)
536 url, smuggled_data = unsmuggle_url(url)
538 is_intentional = smuggled_data and smuggled_data.get('to_generic')
539 if smuggled_data and 'force_videoid' in smuggled_data:
540 force_videoid = smuggled_data['force_videoid']
541 video_id = force_videoid
543 video_id = os.path.splitext(url.rstrip('/').split('/')[-1])[0]
545 self.to_screen('%s: Requesting header' % video_id)
547 head_req = HEADRequest(url)
548 head_response = self._request_webpage(
550 note=False, errnote='Could not send HEAD request to %s' % url,
553 if head_response is not False:
555 new_url = head_response.geturl()
557 self.report_following_redirect(new_url)
559 new_url = smuggle_url(
560 new_url, {'force_videoid': force_videoid})
561 return self.url_result(new_url)
564 if head_response is False:
565 full_response = self._request_webpage(url, video_id)
566 head_response = full_response
568 # Check for direct link to a video
569 content_type = head_response.headers.get('Content-Type', '')
570 m = re.match(r'^(?P<type>audio|video|application(?=/ogg$))/(?P<format_id>.+)$', content_type)
572 upload_date = unified_strdate(
573 head_response.headers.get('Last-Modified'))
576 'title': os.path.splitext(url_basename(url))[0],
578 'format_id': m.group('format_id'),
580 'vcodec': 'none' if m.group('type') == 'audio' else None
582 'upload_date': upload_date,
585 if not self._downloader.params.get('test', False) and not is_intentional:
586 self._downloader.report_warning('Falling back on generic information extractor.')
589 webpage = self._webpage_read_content(full_response, url, video_id)
591 webpage = self._download_webpage(url, video_id)
592 self.report_extraction(video_id)
596 doc = parse_xml(webpage)
598 return self._extract_rss(url, video_id, doc)
599 except compat_xml_parse_error:
602 # Is it a Camtasia project?
603 camtasia_res = self._extract_camtasia(url, video_id, webpage)
604 if camtasia_res is not None:
607 # Sometimes embedded video player is hidden behind percent encoding
608 # (e.g. https://github.com/rg3/youtube-dl/issues/2448)
609 # Unescaping the whole page allows to handle those cases in a generic way
610 webpage = compat_urllib_parse.unquote(webpage)
612 # it's tempting to parse this further, but you would
613 # have to take into account all the variations like
614 # Video Title - Site Name
615 # Site Name | Video Title
616 # Video Title - Tagline | Site Name
617 # and so on and so forth; it's just not practical
618 video_title = self._html_search_regex(
619 r'(?s)<title>(.*?)</title>', webpage, 'video title',
622 # Try to detect age limit automatically
623 age_limit = self._rta_search(webpage)
624 # And then there are the jokers who advertise that they use RTA,
625 # but actually don't.
626 AGE_LIMIT_MARKERS = [
627 r'Proudly Labeled <a href="http://www.rtalabel.org/" title="Restricted to Adults">RTA</a>',
629 if any(re.search(marker, webpage) for marker in AGE_LIMIT_MARKERS):
632 # video uploader is domain name
633 video_uploader = self._search_regex(
634 r'^(?:https?://)?([^/]*)/.*', url, 'video uploader')
637 def _playlist_from_matches(matches, getter, ie=None):
639 self.url_result(self._proto_relative_url(getter(m)), ie)
641 return self.playlist_result(
642 urlrs, playlist_id=video_id, playlist_title=video_title)
644 # Look for BrightCove:
645 bc_urls = BrightcoveIE._extract_brightcove_urls(webpage)
647 self.to_screen('Brightcove video detected.')
650 'url': smuggle_url(bc_url, {'Referer': url}),
651 'ie_key': 'Brightcove'
652 } for bc_url in bc_urls]
656 'title': video_title,
661 # Look for embedded (iframe) Vimeo player
663 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//player\.vimeo\.com/video/.+?)\1', webpage)
665 player_url = unescapeHTML(mobj.group('url'))
666 surl = smuggle_url(player_url, {'Referer': url})
667 return self.url_result(surl)
669 # Look for embedded (swf embed) Vimeo player
671 r'<embed[^>]+?src="((?:https?:)?//(?:www\.)?vimeo\.com/moogaloop\.swf.+?)"', webpage)
673 return self.url_result(mobj.group(1))
675 # Look for embedded YouTube player
676 matches = re.findall(r'''(?x)
685 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
689 return _playlist_from_matches(
690 matches, lambda m: unescapeHTML(m[1]))
692 # Look for embedded Dailymotion player
693 matches = re.findall(
694 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/embed/video/.+?)\1', webpage)
696 return _playlist_from_matches(
697 matches, lambda m: unescapeHTML(m[1]))
699 # Look for embedded Dailymotion playlist player (#3822)
701 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.[a-z]{2,3}/widget/jukebox\?.+?)\1', webpage)
703 playlists = re.findall(
704 r'list\[\]=/playlist/([^/]+)/', unescapeHTML(m.group('url')))
706 return _playlist_from_matches(
707 playlists, lambda p: '//dailymotion.com/playlist/%s' % p)
709 # Look for embedded Wistia player
711 r'<(?:meta[^>]+?content|iframe[^>]+?src)=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage)
713 embed_url = self._proto_relative_url(
714 unescapeHTML(match.group('url')))
716 '_type': 'url_transparent',
719 'uploader': video_uploader,
720 'title': video_title,
724 match = re.search(r'(?:id=["\']wistia_|data-wistia-?id=["\']|Wistia\.embed\(["\'])(?P<id>[^"\']+)', webpage)
727 '_type': 'url_transparent',
728 'url': 'http://fast.wistia.net/embed/iframe/{0:}'.format(match.group('id')),
730 'uploader': video_uploader,
731 'title': video_title,
732 'id': match.group('id')
735 # Look for embedded blip.tv player
736 mobj = re.search(r'<meta\s[^>]*https?://api\.blip\.tv/\w+/redirect/\w+/(\d+)', webpage)
738 return self.url_result('http://blip.tv/a/a-'+mobj.group(1), 'BlipTV')
739 mobj = re.search(r'<(?:iframe|embed|object)\s[^>]*(https?://(?:\w+\.)?blip\.tv/(?:play/|api\.swf#)[a-zA-Z0-9_]+)', webpage)
741 return self.url_result(mobj.group(1), 'BlipTV')
743 # Look for embedded condenast player
744 matches = re.findall(
745 r'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="(https?://player\.cnevids\.com/embed/[^"]+")',
752 'ie_key': 'CondeNast',
754 } for ma in matches],
755 'title': video_title,
759 # Look for Bandcamp pages with custom domain
760 mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage)
762 burl = unescapeHTML(mobj.group(1))
763 # Don't set the extractor because it can be a track url or an album
764 return self.url_result(burl)
766 # Look for embedded Vevo player
768 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:cache\.)?vevo\.com/.+?)\1', webpage)
770 return self.url_result(mobj.group('url'))
772 # Look for Ooyala videos
773 mobj = (re.search(r'player.ooyala.com/[^"?]+\?[^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or
774 re.search(r'OO.Player.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage))
776 return OoyalaIE._build_url_result(mobj.group('ec'))
778 # Look for Aparat videos
779 mobj = re.search(r'<iframe .*?src="(http://www\.aparat\.com/video/[^"]+)"', webpage)
781 return self.url_result(mobj.group(1), 'Aparat')
783 # Look for MPORA videos
784 mobj = re.search(r'<iframe .*?src="(http://mpora\.(?:com|de)/videos/[^"]+)"', webpage)
786 return self.url_result(mobj.group(1), 'Mpora')
788 # Look for embedded NovaMov-based player
790 r'''(?x)<(?:pagespeed_)?iframe[^>]+?src=(["\'])
791 (?P<url>http://(?:(?:embed|www)\.)?
793 nowvideo\.(?:ch|sx|eu|at|ag|co)|
794 videoweed\.(?:es|com)|
795 movshare\.(?:net|sx|ag)|
796 divxstage\.(?:eu|net|ch|co|at|ag))
797 /embed\.php.+?)\1''', webpage)
799 return self.url_result(mobj.group('url'))
801 # Look for embedded Facebook player
803 r'<iframe[^>]+?src=(["\'])(?P<url>https://www\.facebook\.com/video/embed.+?)\1', webpage)
805 return self.url_result(mobj.group('url'), 'Facebook')
807 # Look for embedded VK player
808 mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://vk\.com/video_ext\.php.+?)\1', webpage)
810 return self.url_result(mobj.group('url'), 'VK')
812 # Look for embedded ivi player
813 mobj = re.search(r'<embed[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?ivi\.ru/video/player.+?)\1', webpage)
815 return self.url_result(mobj.group('url'), 'Ivi')
817 # Look for embedded Huffington Post player
819 r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed\.live\.huffingtonpost\.com/.+?)\1', webpage)
821 return self.url_result(mobj.group('url'), 'HuffPost')
824 mobj = re.search(r'class=["\']embedly-card["\'][^>]href=["\'](?P<url>[^"\']+)', webpage)
826 return self.url_result(mobj.group('url'))
827 mobj = re.search(r'class=["\']embedly-embed["\'][^>]src=["\'][^"\']*url=(?P<url>[^&]+)', webpage)
829 return self.url_result(compat_urllib_parse.unquote(mobj.group('url')))
831 # Look for funnyordie embed
832 matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage)
834 return _playlist_from_matches(
835 matches, getter=unescapeHTML, ie='FunnyOrDie')
837 # Look for embedded RUTV player
838 rutv_url = RUTVIE._extract_url(webpage)
840 return self.url_result(rutv_url, 'RUTV')
842 # Look for embedded TED player
844 r'<iframe[^>]+?src=(["\'])(?P<url>http://embed\.ted\.com/.+?)\1', webpage)
846 return self.url_result(mobj.group('url'), 'TED')
848 # Look for embedded Ustream videos
850 r'<iframe[^>]+?src=(["\'])(?P<url>http://www\.ustream\.tv/embed/.+?)\1', webpage)
852 return self.url_result(mobj.group('url'), 'Ustream')
854 # Look for embedded arte.tv player
856 r'<script [^>]*?src="(?P<url>http://www\.arte\.tv/playerv2/embed[^"]+)"',
859 return self.url_result(mobj.group('url'), 'ArteTVEmbed')
861 # Look for embedded smotri.com player
862 smotri_url = SmotriIE._extract_url(webpage)
864 return self.url_result(smotri_url, 'Smotri')
866 # Look for embeded soundcloud player
868 r'<iframe\s+(?:[a-zA-Z0-9_-]+="[^"]+"\s+)*src="(?P<url>https?://(?:w\.)?soundcloud\.com/player[^"]+)"',
871 url = unescapeHTML(mobj.group('url'))
872 return self.url_result(url)
874 # Look for embedded vulture.com player
876 r'<iframe src="(?P<url>https?://video\.vulture\.com/[^"]+)"',
879 url = unescapeHTML(mobj.group('url'))
880 return self.url_result(url, ie='Vulture')
882 # Look for embedded mtvservices player
884 r'<iframe src="(?P<url>https?://media\.mtvnservices\.com/embed/[^"]+)"',
887 url = unescapeHTML(mobj.group('url'))
888 return self.url_result(url, ie='MTVServicesEmbedded')
890 # Look for embedded yahoo player
892 r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:screen|movies)\.yahoo\.com/.+?\.html\?format=embed)\1',
895 return self.url_result(mobj.group('url'), 'Yahoo')
897 # Look for embedded sbs.com.au player
899 r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:www\.)sbs\.com\.au/ondemand/video/single/.+?)\1',
902 return self.url_result(mobj.group('url'), 'SBS')
905 r'<iframe[^>]+?src=(["\'])(?P<url>https?://m(?:lb)?\.mlb\.com/shared/video/embed/embed\.html\?.+?)\1',
908 return self.url_result(mobj.group('url'), 'MLB')
911 r'<iframe[^>]+?src=(["\'])(?P<url>%s)\1' % CondeNastIE.EMBED_URL,
914 return self.url_result(self._proto_relative_url(mobj.group('url'), scheme='http:'), 'CondeNast')
916 def check_video(vurl):
917 vpath = compat_urlparse.urlparse(vurl).path
918 vext = determine_ext(vpath)
919 return '.' in vpath and vext not in ('swf', 'png', 'jpg', 'srt', 'sbv', 'sub', 'vtt', 'ttml')
921 def filter_video(urls):
922 return list(filter(check_video, urls))
924 # Start with something easy: JW Player in SWFObject
925 found = filter_video(re.findall(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage))
927 # Look for gorilla-vid style embedding
928 found = filter_video(re.findall(r'''(?sx)
932 jwplayer\s*\(\s*["'][^'"]+["']\s*\)\s*\.setup
934 .*?file\s*:\s*["\'](.*?)["\']''', webpage))
936 # Broaden the search a little bit
937 found = filter_video(re.findall(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage))
939 # Broaden the findall a little bit: JWPlayer JS loader
940 found = filter_video(re.findall(
941 r'[^A-Za-z0-9]?file["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage))
944 found = filter_video(re.findall(r'''(?xs)
945 flowplayer\("[^"]+",\s*
947 \s*{[^}]+? ["']?clip["']?\s*:\s*\{\s*
948 ["']?url["']?\s*:\s*["']([^"']+)["']
951 # Try to find twitter cards info
952 found = filter_video(re.findall(
953 r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage))
955 # We look for Open Graph info:
956 # We have to match any number spaces between elements, some sites try to align them (eg.: statigr.am)
957 m_video_type = re.findall(r'<meta.*?property="og:video:type".*?content="video/(.*?)"', webpage)
958 # We only look in og:video if the MIME type is a video, don't try if it's a Flash player:
959 if m_video_type is not None:
960 found = filter_video(re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage))
963 found = re.findall(r'(?s)<video[^<]*(?:>.*?<source[^>]*)?\s+src="([^"]+)"', webpage)
966 r'(?i)<meta\s+(?=(?:[a-z-]+="[^"]+"\s+)*http-equiv="refresh")'
967 r'(?:[a-z-]+="[^"]+"\s+)*?content="[0-9]{,2};url=\'?([^\'"]+)',
970 new_url = found.group(1)
971 self.report_following_redirect(new_url)
977 raise ExtractorError('Unsupported URL: %s' % url)
980 for video_url in found:
981 video_url = compat_urlparse.urljoin(url, video_url)
982 video_id = compat_urllib_parse.unquote(os.path.basename(video_url))
984 # Sometimes, jwplayer extraction will result in a YouTube URL
985 if YoutubeIE.suitable(video_url):
986 entries.append(self.url_result(video_url, 'Youtube'))
989 # here's a fun little line of code for you:
990 video_id = os.path.splitext(video_id)[0]
995 'uploader': video_uploader,
996 'title': video_title,
997 'age_limit': age_limit,
1000 if len(entries) == 1:
1003 for num, e in enumerate(entries, start=1):
1004 e['title'] = '%s (%d)' % (e['title'], num)
1006 '_type': 'playlist',