3 from __future__ import unicode_literals
8 from .common import InfoExtractor
9 from .youtube import YoutubeIE
13 compat_xml_parse_error,
27 from .brightcove import BrightcoveIE
28 from .ooyala import OoyalaIE
29 from .rutv import RUTVIE
30 from .smotri import SmotriIE
31 from .condenast import CondeNastIE
34 class GenericIE(InfoExtractor):
35 IE_DESC = 'Generic downloader that works on some sites'
40 'url': 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
41 'md5': '85b90ccc9d73b4acd9138d3af4c27f89',
43 'id': '13601338388002',
45 'uploader': 'www.hodiho.fr',
46 'title': 'R\u00e9gis plante sa Jeep',
49 # bandcamp page with custom domain
51 'add_ie': ['Bandcamp'],
52 'url': 'http://bronyrock.com/track/the-pony-mash',
56 'title': 'The Pony Mash',
57 'uploader': 'M_Pallante',
59 'skip': 'There is a limit of 200 free downloads / month for the test song',
61 # embedded brightcove video
62 # it also tests brightcove videos that need to set the 'Referer' in the
65 'add_ie': ['Brightcove'],
66 'url': 'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/',
68 'id': '2765128793001',
70 'title': 'Le cours de bourse : l’analyse technique',
71 'description': 'md5:7e9ad046e968cb2d1114004aba466fd9',
72 'uploader': 'BFM BUSINESS',
75 'skip_download': True,
79 # https://github.com/rg3/youtube-dl/issues/2253
80 'url': 'http://bcove.me/i6nfkrc3',
81 'md5': '0ba9446db037002366bab3b3eb30c88c',
83 'id': '3101154703001',
85 'title': 'Still no power',
86 'uploader': 'thestar.com',
87 '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.',
89 'add_ie': ['Brightcove'],
92 'url': 'http://www.championat.com/video/football/v/87/87499.html',
93 'md5': 'fb973ecf6e4a78a67453647444222983',
95 'id': '3414141473001',
97 'title': 'Видео. Удаление Дзагоева (ЦСКА)',
98 'description': 'Онлайн-трансляция матча ЦСКА - "Волга"',
99 'uploader': 'Championat',
102 # Direct link to a video
104 'url': 'http://media.w3.org/2010/05/sintel/trailer.mp4',
105 'md5': '67d406c2bcb6af27fa886f31aa934bbe',
110 'upload_date': '20100513',
115 'url': 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219',
116 'md5': '5644c6ca5d5782c1d0d350dad9bd840c',
118 'id': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ',
120 'title': '2cc213299525360.mov', # that's what we get
125 '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',
129 'upload_date': '20130224',
130 'uploader_id': 'TheVerge',
131 '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.',
132 'uploader': 'The Verge',
133 'title': 'First Firefox OS phones side-by-side',
136 'skip_download': False,
141 'url': 'http://www.tested.com/science/weird/460206-tested-grinding-coffee-2000-frames-second/',
145 'title': 'Tested: Grinding Coffee at 2000 Frames Per Second',
146 'upload_date': '20140225',
147 'description': 'md5:06a40fbf30b220468f1e0957c0f558ff',
148 'uploader': 'Tested',
149 'uploader_id': 'testedcom',
151 # No need to test YoutubeIE here
153 'skip_download': True,
158 'url': 'http://www.theguardian.com/world/2014/mar/11/obama-zach-galifianakis-between-two-ferns',
162 'title': 'Between Two Ferns with Zach Galifianakis: President Barack Obama',
163 'description': 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.',
168 'url': 'http://www.rg.ru/2014/03/15/reg-dfo/anklav-anons.html',
172 'title': 'Охотское море стало целиком российским',
173 'description': 'md5:5ed62483b14663e2a95ebbe115eb8f43',
177 'skip_download': True,
182 'url': 'http://en.support.wordpress.com/videos/ted-talks/',
183 'md5': '65fdff94098e4a607385a60c5177c638',
187 'title': 'Hidden miracles of the natural world',
188 'uploader': 'Louie Schwartzberg',
189 'description': 'md5:8145d19d320ff3e52f28401f4c4283b9',
192 # Embeded Ustream video
194 'url': 'http://www.american.edu/spa/pti/nsa-privacy-janus-2014.cfm',
195 'md5': '27b99cdb639c9b12a79bca876a073417',
199 'uploader': 'AU SPA: The NSA and Privacy',
200 'title': 'NSA and Privacy Forum Debate featuring General Hayden and Barton Gellman'
203 # nowvideo embed hidden behind percent encoding
205 'url': 'http://www.waoanime.tv/the-super-dimension-fortress-macross-episode-1/',
206 'md5': '2baf4ddd70f697d94b1c18cf796d5107',
208 'id': '06e53103ca9aa',
210 'title': 'Macross Episode 001 Watch Macross Episode 001 onl',
211 'description': 'No description',
216 'url': 'http://www.tv-replay.fr/redirection/20-03-14/x-enius-arte-10753389.html',
217 'md5': '7653032cbb25bf6c80d80f217055fa43',
219 'id': '048195-004_PLUS7-F',
222 'description': 'md5:d5fdf32ef6613cdbfd516ae658abf168',
223 'upload_date': '20140320',
226 'skip_download': 'Requires rtmpdump'
231 'url': 'http://www.wired.com/2014/04/honda-asimo/',
232 'md5': 'ba0dfe966fa007657bd1443ee672db0f',
234 'id': '53501be369702d3275860000',
236 'title': 'Honda’s New Asimo Robot Is More Human Than Ever',
241 'url': 'http://www.spi0n.com/zap-spi0n-com-n216/',
242 'md5': '441aeeb82eb72c422c7f14ec533999cd',
244 'id': 'k2mm4bCdJ6CQ2i7c8o2',
246 'title': 'Le Zap de Spi0n n°216 - Zapping du Web',
249 'add_ie': ['Dailymotion'],
253 'url': 'http://www.badzine.de/ansicht/datum/2014/06/09/so-funktioniert-die-neue-englische-badminton-liga.html',
257 'title': 'The NBL Auction 2014',
258 'uploader': 'BADMINTON England',
259 'uploader_id': 'BADMINTONEvents',
260 'upload_date': '20140603',
261 'description': 'md5:9ef128a69f1e262a700ed83edb163a73',
263 'add_ie': ['Youtube'],
265 'skip_download': True,
270 'url': 'http://www.gametrailers.com/news-post/76093/north-america-europe-is-getting-that-mario-kart-8-mercedes-dlc-too',
271 'md5': '35727f82f58c76d996fc188f9755b0d5',
273 'id': '0306a69b-8adf-4fb5-aace-75f8e8cbfca9',
276 'description': 'Mario\'s life in the fast lane has never looked so good.',
279 # YouTube embed via <data-embed-url="">
281 'url': 'https://play.google.com/store/apps/details?id=com.gameloft.android.ANMP.GloftA8HM',
285 'title': 'Asphalt 8: Airborne - Update - Welcome to Dubai!',
286 'uploader': 'Gameloft',
287 'uploader_id': 'gameloft',
288 'upload_date': '20140828',
289 'description': 'md5:c80da9ed3d83ae6d1876c834de03e1c4',
292 'skip_download': True,
297 'url': 'http://www.ll.mit.edu/workshops/education/videocourses/antennas/lecture1/video/',
299 'md5': '0c5e352edabf715d762b0ad4e6d9ee67',
301 'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
302 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - video1',
307 'md5': '10e4bb3aaca9fd630e273ff92d9f3c63',
309 'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final_PIP',
310 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - pip',
316 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
321 'url': 'http://www.handjobhub.com/video/busty-blonde-siri-tit-fuck-while-wank-6313.html',
322 'md5': '9d65602bf31c6e20014319c7d07fba27',
324 'id': '5123ea6d5e5a7',
327 'uploader': 'www.handjobhub.com',
328 'title': 'Busty Blonde Siri Tit Fuck While Wank at Handjob Hub',
333 'url': 'http://phihag.de/2014/youtube-dl/rss2.xml',
335 'id': 'http://phihag.de/2014/youtube-dl/rss2.xml',
336 'title': 'Zero Punctuation',
339 'playlist_mincount': 11,
341 # Multiple brightcove videos
342 # https://github.com/rg3/youtube-dl/issues/2283
344 'url': 'http://www.newyorker.com/online/blogs/newsdesk/2014/01/always-never-nuclear-command-and-control.html',
346 'id': 'always-never',
347 'title': 'Always / Never - The New Yorker',
351 'extract_flat': False,
352 'skip_download': True,
357 'url': 'http://umpire-empire.com/index.php/topic/58125-laz-decides-no-thats-low/',
358 'md5': '96f09a37e44da40dd083e12d9a683327',
362 'title': 'Ump changes call to ball',
363 'description': 'md5:71c11215384298a172a6dcb4c2e20685',
365 'timestamp': 1401537900,
366 'upload_date': '20140531',
367 'thumbnail': 're:^https?://.*\.jpg$',
372 'url': 'http://education-portal.com/academy/lesson/north-american-exploration-failed-colonies-of-spain-france-england.html#lesson',
373 'md5': '8788b683c777a5cf25621eaf286d0c23',
377 'title': 'md5:51364a8d3d009997ba99656004b5e20d',
379 'filesize': 182808282,
380 'uploader': 'education-portal.com',
385 def report_following_redirect(self, new_url):
386 """Report information extraction."""
387 self._downloader.to_screen('[redirect] Following redirect to %s' % new_url)
389 def _extract_rss(self, url, video_id, doc):
390 playlist_title = doc.find('./channel/title').text
391 playlist_desc_el = doc.find('./channel/description')
392 playlist_desc = None if playlist_desc_el is None else playlist_desc_el.text
396 'url': e.find('link').text,
397 'title': e.find('title').text,
398 } for e in doc.findall('./channel/item')]
403 'title': playlist_title,
404 'description': playlist_desc,
408 def _extract_camtasia(self, url, video_id, webpage):
409 """ Returns None if no camtasia video can be found. """
411 camtasia_cfg = self._search_regex(
412 r'fo\.addVariable\(\s*"csConfigFile",\s*"([^"]+)"\s*\);',
413 webpage, 'camtasia configuration file', default=None)
414 if camtasia_cfg is None:
417 title = self._html_search_meta('DC.title', webpage, fatal=True)
419 camtasia_url = compat_urlparse.urljoin(url, camtasia_cfg)
420 camtasia_cfg = self._download_xml(
421 camtasia_url, video_id,
422 note='Downloading camtasia configuration',
423 errnote='Failed to download camtasia configuration')
424 fileset_node = camtasia_cfg.find('./playlist/array/fileset')
427 for n in fileset_node.getchildren():
428 url_n = n.find('./uri')
433 'id': os.path.splitext(url_n.text.rpartition('/')[2])[0],
434 'title': '%s - %s' % (title, n.tag),
435 'url': compat_urlparse.urljoin(url, url_n.text),
436 'duration': float_or_none(n.find('./duration').text),
445 def _real_extract(self, url):
446 if url.startswith('//'):
449 'url': self.http_scheme() + url,
452 parsed_url = compat_urlparse.urlparse(url)
453 if not parsed_url.scheme:
454 default_search = self._downloader.params.get('default_search')
455 if default_search is None:
456 default_search = 'fixup_error'
458 if default_search in ('auto', 'auto_warning', 'fixup_error'):
460 self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http')
461 return self.url_result('http://' + url)
462 elif default_search != 'fixup_error':
463 if default_search == 'auto_warning':
464 if re.match(r'^(?:url|URL)$', url):
465 raise ExtractorError(
466 'Invalid URL: %r . Call youtube-dl like this: youtube-dl -v "https://www.youtube.com/watch?v=BaW_jenozKc" ' % url,
469 self._downloader.report_warning(
470 'Falling back to youtube search for %s . Set --default-search "auto" to suppress this warning.' % url)
471 return self.url_result('ytsearch:' + url)
473 if default_search in ('error', 'fixup_error'):
474 raise ExtractorError(
475 ('%r is not a valid URL. '
476 'Set --default-search "ytsearch" (or run youtube-dl "ytsearch:%s" ) to search YouTube'
477 ) % (url, url), expected=True)
479 assert ':' in default_search
480 return self.url_result(default_search + url)
482 url, smuggled_data = unsmuggle_url(url)
484 is_intentional = smuggled_data and smuggled_data.get('to_generic')
485 if smuggled_data and 'force_videoid' in smuggled_data:
486 force_videoid = smuggled_data['force_videoid']
487 video_id = force_videoid
489 video_id = os.path.splitext(url.rstrip('/').split('/')[-1])[0]
491 self.to_screen('%s: Requesting header' % video_id)
493 head_req = HEADRequest(url)
494 response = self._request_webpage(
496 note=False, errnote='Could not send HEAD request to %s' % url,
499 if response is not False:
501 new_url = response.geturl()
503 self.report_following_redirect(new_url)
505 new_url = smuggle_url(
506 new_url, {'force_videoid': force_videoid})
507 return self.url_result(new_url)
509 # Check for direct link to a video
510 content_type = response.headers.get('Content-Type', '')
511 m = re.match(r'^(?P<type>audio|video|application(?=/ogg$))/(?P<format_id>.+)$', content_type)
513 upload_date = response.headers.get('Last-Modified')
515 upload_date = unified_strdate(upload_date)
518 'title': os.path.splitext(url_basename(url))[0],
520 'format_id': m.group('format_id'),
522 'vcodec': 'none' if m.group('type') == 'audio' else None
524 'upload_date': upload_date,
527 if not self._downloader.params.get('test', False) and not is_intentional:
528 self._downloader.report_warning('Falling back on generic information extractor.')
531 webpage = self._download_webpage(url, video_id)
533 # since this is the last-resort InfoExtractor, if
534 # this error is thrown, it'll be thrown here
535 raise ExtractorError('Failed to download URL: %s' % url)
537 self.report_extraction(video_id)
541 doc = parse_xml(webpage)
543 return self._extract_rss(url, video_id, doc)
544 except compat_xml_parse_error:
547 # Is it a Camtasia project?
548 camtasia_res = self._extract_camtasia(url, video_id, webpage)
549 if camtasia_res is not None:
552 # Sometimes embedded video player is hidden behind percent encoding
553 # (e.g. https://github.com/rg3/youtube-dl/issues/2448)
554 # Unescaping the whole page allows to handle those cases in a generic way
555 webpage = compat_urllib_parse.unquote(webpage)
557 # it's tempting to parse this further, but you would
558 # have to take into account all the variations like
559 # Video Title - Site Name
560 # Site Name | Video Title
561 # Video Title - Tagline | Site Name
562 # and so on and so forth; it's just not practical
563 video_title = self._html_search_regex(
564 r'(?s)<title>(.*?)</title>', webpage, 'video title',
567 # Try to detect age limit automatically
568 age_limit = self._rta_search(webpage)
569 # And then there are the jokers who advertise that they use RTA,
570 # but actually don't.
571 AGE_LIMIT_MARKERS = [
572 r'Proudly Labeled <a href="http://www.rtalabel.org/" title="Restricted to Adults">RTA</a>',
574 if any(re.search(marker, webpage) for marker in AGE_LIMIT_MARKERS):
577 # video uploader is domain name
578 video_uploader = self._search_regex(
579 r'^(?:https?://)?([^/]*)/.*', url, 'video uploader')
582 def _playlist_from_matches(matches, getter, ie=None):
584 self.url_result(self._proto_relative_url(getter(m)), ie)
586 return self.playlist_result(
587 urlrs, playlist_id=video_id, playlist_title=video_title)
589 # Look for BrightCove:
590 bc_urls = BrightcoveIE._extract_brightcove_urls(webpage)
592 self.to_screen('Brightcove video detected.')
595 'url': smuggle_url(bc_url, {'Referer': url}),
596 'ie_key': 'Brightcove'
597 } for bc_url in bc_urls]
601 'title': video_title,
606 # Look for embedded (iframe) Vimeo player
608 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//player\.vimeo\.com/video/.+?)\1', webpage)
610 player_url = unescapeHTML(mobj.group('url'))
611 surl = smuggle_url(player_url, {'Referer': url})
612 return self.url_result(surl)
614 # Look for embedded (swf embed) Vimeo player
616 r'<embed[^>]+?src="((?:https?:)?//(?:www\.)?vimeo\.com/moogaloop\.swf.+?)"', webpage)
618 return self.url_result(mobj.group(1))
620 # Look for embedded YouTube player
621 matches = re.findall(r'''(?x)
629 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
633 return _playlist_from_matches(
634 matches, lambda m: unescapeHTML(m[1]))
636 # Look for embedded Dailymotion player
637 matches = re.findall(
638 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/embed/video/.+?)\1', webpage)
640 return _playlist_from_matches(
641 matches, lambda m: unescapeHTML(m[1]))
643 # Look for embedded Dailymotion playlist player (#3822)
645 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.[a-z]{2,3}/widget/jukebox\?.+?)\1', webpage)
647 playlists = re.findall(
648 r'list\[\]=/playlist/([^/]+)/', unescapeHTML(m.group('url')))
650 return _playlist_from_matches(
651 playlists, lambda p: '//dailymotion.com/playlist/%s' % p)
653 # Look for embedded Wistia player
655 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage)
657 embed_url = self._proto_relative_url(
658 unescapeHTML(match.group('url')))
660 '_type': 'url_transparent',
663 'uploader': video_uploader,
664 'title': video_title,
667 match = re.search(r'(?:id=["\']wistia_|data-wistia-?id=["\']|Wistia\.embed\(["\'])(?P<id>[^"\']+)', webpage)
670 '_type': 'url_transparent',
671 'url': 'http://fast.wistia.net/embed/iframe/{0:}'.format(match.group('id')),
673 'uploader': video_uploader,
674 'title': video_title,
675 'id': match.group('id')
678 # Look for embedded blip.tv player
679 mobj = re.search(r'<meta\s[^>]*https?://api\.blip\.tv/\w+/redirect/\w+/(\d+)', webpage)
681 return self.url_result('http://blip.tv/a/a-'+mobj.group(1), 'BlipTV')
682 mobj = re.search(r'<(?:iframe|embed|object)\s[^>]*(https?://(?:\w+\.)?blip\.tv/(?:play/|api\.swf#)[a-zA-Z0-9_]+)', webpage)
684 return self.url_result(mobj.group(1), 'BlipTV')
686 # Look for embedded condenast player
687 matches = re.findall(
688 r'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="(https?://player\.cnevids\.com/embed/[^"]+")',
695 'ie_key': 'CondeNast',
697 } for ma in matches],
698 'title': video_title,
702 # Look for Bandcamp pages with custom domain
703 mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage)
705 burl = unescapeHTML(mobj.group(1))
706 # Don't set the extractor because it can be a track url or an album
707 return self.url_result(burl)
709 # Look for embedded Vevo player
711 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:cache\.)?vevo\.com/.+?)\1', webpage)
713 return self.url_result(mobj.group('url'))
715 # Look for Ooyala videos
716 mobj = (re.search(r'player.ooyala.com/[^"?]+\?[^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or
717 re.search(r'OO.Player.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage))
719 return OoyalaIE._build_url_result(mobj.group('ec'))
721 # Look for Aparat videos
722 mobj = re.search(r'<iframe .*?src="(http://www\.aparat\.com/video/[^"]+)"', webpage)
724 return self.url_result(mobj.group(1), 'Aparat')
726 # Look for MPORA videos
727 mobj = re.search(r'<iframe .*?src="(http://mpora\.(?:com|de)/videos/[^"]+)"', webpage)
729 return self.url_result(mobj.group(1), 'Mpora')
731 # Look for embedded NovaMov-based player
733 r'''(?x)<(?:pagespeed_)?iframe[^>]+?src=(["\'])
734 (?P<url>http://(?:(?:embed|www)\.)?
736 nowvideo\.(?:ch|sx|eu|at|ag|co)|
737 videoweed\.(?:es|com)|
738 movshare\.(?:net|sx|ag)|
739 divxstage\.(?:eu|net|ch|co|at|ag))
740 /embed\.php.+?)\1''', webpage)
742 return self.url_result(mobj.group('url'))
744 # Look for embedded Facebook player
746 r'<iframe[^>]+?src=(["\'])(?P<url>https://www\.facebook\.com/video/embed.+?)\1', webpage)
748 return self.url_result(mobj.group('url'), 'Facebook')
750 # Look for embedded VK player
751 mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://vk\.com/video_ext\.php.+?)\1', webpage)
753 return self.url_result(mobj.group('url'), 'VK')
755 # Look for embedded ivi player
756 mobj = re.search(r'<embed[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?ivi\.ru/video/player.+?)\1', webpage)
758 return self.url_result(mobj.group('url'), 'Ivi')
760 # Look for embedded Huffington Post player
762 r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed\.live\.huffingtonpost\.com/.+?)\1', webpage)
764 return self.url_result(mobj.group('url'), 'HuffPost')
767 mobj = re.search(r'class=["\']embedly-card["\'][^>]href=["\'](?P<url>[^"\']+)', webpage)
769 return self.url_result(mobj.group('url'))
770 mobj = re.search(r'class=["\']embedly-embed["\'][^>]src=["\'][^"\']*url=(?P<url>[^&]+)', webpage)
772 return self.url_result(compat_urllib_parse.unquote(mobj.group('url')))
774 # Look for funnyordie embed
775 matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage)
777 return _playlist_from_matches(
778 matches, getter=unescapeHTML, ie='FunnyOrDie')
780 # Look for embedded RUTV player
781 rutv_url = RUTVIE._extract_url(webpage)
783 return self.url_result(rutv_url, 'RUTV')
785 # Look for embedded TED player
787 r'<iframe[^>]+?src=(["\'])(?P<url>http://embed\.ted\.com/.+?)\1', webpage)
789 return self.url_result(mobj.group('url'), 'TED')
791 # Look for embedded Ustream videos
793 r'<iframe[^>]+?src=(["\'])(?P<url>http://www\.ustream\.tv/embed/.+?)\1', webpage)
795 return self.url_result(mobj.group('url'), 'Ustream')
797 # Look for embedded arte.tv player
799 r'<script [^>]*?src="(?P<url>http://www\.arte\.tv/playerv2/embed[^"]+)"',
802 return self.url_result(mobj.group('url'), 'ArteTVEmbed')
804 # Look for embedded smotri.com player
805 smotri_url = SmotriIE._extract_url(webpage)
807 return self.url_result(smotri_url, 'Smotri')
809 # Look for embeded soundcloud player
811 r'<iframe src="(?P<url>https?://(?:w\.)?soundcloud\.com/player[^"]+)"',
814 url = unescapeHTML(mobj.group('url'))
815 return self.url_result(url)
817 # Look for embedded vulture.com player
819 r'<iframe src="(?P<url>https?://video\.vulture\.com/[^"]+)"',
822 url = unescapeHTML(mobj.group('url'))
823 return self.url_result(url, ie='Vulture')
825 # Look for embedded mtvservices player
827 r'<iframe src="(?P<url>https?://media\.mtvnservices\.com/embed/[^"]+)"',
830 url = unescapeHTML(mobj.group('url'))
831 return self.url_result(url, ie='MTVServicesEmbedded')
833 # Look for embedded yahoo player
835 r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:screen|movies)\.yahoo\.com/.+?\.html\?format=embed)\1',
838 return self.url_result(mobj.group('url'), 'Yahoo')
840 # Look for embedded sbs.com.au player
842 r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:www\.)sbs\.com\.au/ondemand/video/single/.+?)\1',
845 return self.url_result(mobj.group('url'), 'SBS')
848 r'<iframe[^>]+?src=(["\'])(?P<url>https?://m\.mlb\.com/shared/video/embed/embed\.html\?.+?)\1',
851 return self.url_result(mobj.group('url'), 'MLB')
854 r'<iframe[^>]+?src=(["\'])(?P<url>%s)\1' % CondeNastIE.EMBED_URL,
857 return self.url_result(self._proto_relative_url(mobj.group('url'), scheme='http:'), 'CondeNast')
859 def check_video(vurl):
860 vpath = compat_urlparse.urlparse(vurl).path
861 vext = determine_ext(vpath)
862 return '.' in vpath and vext not in ('swf', 'png', 'jpg', 'srt', 'sbv', 'sub', 'vtt', 'ttml')
864 def filter_video(urls):
865 return list(filter(check_video, urls))
867 # Start with something easy: JW Player in SWFObject
868 found = filter_video(re.findall(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage))
870 # Look for gorilla-vid style embedding
871 found = filter_video(re.findall(r'''(?sx)
875 jwplayer\s*\(\s*["'][^'"]+["']\s*\)\s*\.setup
877 .*?file\s*:\s*["\'](.*?)["\']''', webpage))
879 # Broaden the search a little bit
880 found = filter_video(re.findall(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage))
882 # Broaden the findall a little bit: JWPlayer JS loader
883 found = filter_video(re.findall(
884 r'[^A-Za-z0-9]?file["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage))
887 found = filter_video(re.findall(r'''(?xs)
888 flowplayer\("[^"]+",\s*
890 \s*{[^}]+? ["']?clip["']?\s*:\s*\{\s*
891 ["']?url["']?\s*:\s*["']([^"']+)["']
894 # Try to find twitter cards info
895 found = filter_video(re.findall(
896 r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage))
898 # We look for Open Graph info:
899 # We have to match any number spaces between elements, some sites try to align them (eg.: statigr.am)
900 m_video_type = re.findall(r'<meta.*?property="og:video:type".*?content="video/(.*?)"', webpage)
901 # We only look in og:video if the MIME type is a video, don't try if it's a Flash player:
902 if m_video_type is not None:
903 found = filter_video(re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage))
906 found = re.findall(r'(?s)<video[^<]*(?:>.*?<source[^>]+)? src="([^"]+)"', webpage)
909 r'(?i)<meta\s+(?=(?:[a-z-]+="[^"]+"\s+)*http-equiv="refresh")'
910 r'(?:[a-z-]+="[^"]+"\s+)*?content="[0-9]{,2};url=\'?([^\'"]+)',
913 new_url = found.group(1)
914 self.report_following_redirect(new_url)
920 raise ExtractorError('Unsupported URL: %s' % url)
923 for video_url in found:
924 video_url = compat_urlparse.urljoin(url, video_url)
925 video_id = compat_urllib_parse.unquote(os.path.basename(video_url))
927 # Sometimes, jwplayer extraction will result in a YouTube URL
928 if YoutubeIE.suitable(video_url):
929 entries.append(self.url_result(video_url, 'Youtube'))
932 # here's a fun little line of code for you:
933 video_id = os.path.splitext(video_id)[0]
938 'uploader': video_uploader,
939 'title': video_title,
940 'age_limit': age_limit,
943 if len(entries) == 1:
946 for num, e in enumerate(entries, start=1):
947 e['title'] = '%s (%d)' % (e['title'], num)