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
33 class GenericIE(InfoExtractor):
34 IE_DESC = 'Generic downloader that works on some sites'
39 'url': 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
40 'md5': '85b90ccc9d73b4acd9138d3af4c27f89',
42 'id': '13601338388002',
44 'uploader': 'www.hodiho.fr',
45 'title': 'R\u00e9gis plante sa Jeep',
48 # bandcamp page with custom domain
50 'add_ie': ['Bandcamp'],
51 'url': 'http://bronyrock.com/track/the-pony-mash',
55 'title': 'The Pony Mash',
56 'uploader': 'M_Pallante',
58 'skip': 'There is a limit of 200 free downloads / month for the test song',
60 # embedded brightcove video
61 # it also tests brightcove videos that need to set the 'Referer' in the
64 'add_ie': ['Brightcove'],
65 'url': 'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/',
67 'id': '2765128793001',
69 'title': 'Le cours de bourse : l’analyse technique',
70 'description': 'md5:7e9ad046e968cb2d1114004aba466fd9',
71 'uploader': 'BFM BUSINESS',
74 'skip_download': True,
78 # https://github.com/rg3/youtube-dl/issues/2253
79 'url': 'http://bcove.me/i6nfkrc3',
80 'md5': '0ba9446db037002366bab3b3eb30c88c',
82 'id': '3101154703001',
84 'title': 'Still no power',
85 'uploader': 'thestar.com',
86 '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.',
88 'add_ie': ['Brightcove'],
91 'url': 'http://www.championat.com/video/football/v/87/87499.html',
92 'md5': 'fb973ecf6e4a78a67453647444222983',
94 'id': '3414141473001',
96 'title': 'Видео. Удаление Дзагоева (ЦСКА)',
97 'description': 'Онлайн-трансляция матча ЦСКА - "Волга"',
98 'uploader': 'Championat',
101 # Direct link to a video
103 'url': 'http://media.w3.org/2010/05/sintel/trailer.mp4',
104 'md5': '67d406c2bcb6af27fa886f31aa934bbe',
109 'upload_date': '20100513',
114 'url': 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219',
115 'md5': '5644c6ca5d5782c1d0d350dad9bd840c',
117 'id': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ',
119 'title': '2cc213299525360.mov', # that's what we get
124 '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',
128 'upload_date': '20130224',
129 'uploader_id': 'TheVerge',
130 '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.',
131 'uploader': 'The Verge',
132 'title': 'First Firefox OS phones side-by-side',
135 'skip_download': False,
140 'url': 'http://www.tested.com/science/weird/460206-tested-grinding-coffee-2000-frames-second/',
144 'title': 'Tested: Grinding Coffee at 2000 Frames Per Second',
145 'upload_date': '20140225',
146 'description': 'md5:06a40fbf30b220468f1e0957c0f558ff',
147 'uploader': 'Tested',
148 'uploader_id': 'testedcom',
150 # No need to test YoutubeIE here
152 'skip_download': True,
157 'url': 'http://www.theguardian.com/world/2014/mar/11/obama-zach-galifianakis-between-two-ferns',
158 'md5': '7cf780be104d40fea7bae52eed4a470e',
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': 'deeeabcc1085eb2ba205474e7235a3d5',
187 'title': 'My web playroom',
188 'uploader': 'Ze Frank',
189 'description': 'md5:ddb2a40ecd6b6a147e400e535874947b',
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://rbctv.rbc.ru/archive/news/562949990879132.shtml',
232 'md5': 'ec40048448e9284c9a1de77bb188108b',
234 'id': 'v27008541fad',
236 'title': 'Крым и Севастополь вошли в состав России',
237 'description': 'md5:fae01b61f68984c7bd2fa741e11c3175',
239 'upload_date': '20140318',
240 'uploader': 'rbctv_2012_4',
241 'uploader_id': 'rbctv_2012_4',
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 - Launch Trailer',
301 'uploader': 'Gameloft',
302 'uploader_id': 'gameloft',
303 'upload_date': '20130821',
304 'description': 'md5:87bd95f13d8be3e7da87a5f2c443106a',
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 Handjob Hub',
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 def report_download_webpage(self, video_id):
388 """Report webpage download."""
389 if not self._downloader.params.get('test', False):
390 self._downloader.report_warning('Falling back on generic information extractor.')
391 super(GenericIE, self).report_download_webpage(video_id)
393 def report_following_redirect(self, new_url):
394 """Report information extraction."""
395 self._downloader.to_screen('[redirect] Following redirect to %s' % new_url)
397 def _extract_rss(self, url, video_id, doc):
398 playlist_title = doc.find('./channel/title').text
399 playlist_desc_el = doc.find('./channel/description')
400 playlist_desc = None if playlist_desc_el is None else playlist_desc_el.text
404 'url': e.find('link').text,
405 'title': e.find('title').text,
406 } for e in doc.findall('./channel/item')]
411 'title': playlist_title,
412 'description': playlist_desc,
416 def _extract_camtasia(self, url, video_id, webpage):
417 """ Returns None if no camtasia video can be found. """
419 camtasia_cfg = self._search_regex(
420 r'fo\.addVariable\(\s*"csConfigFile",\s*"([^"]+)"\s*\);',
421 webpage, 'camtasia configuration file', default=None)
422 if camtasia_cfg is None:
425 title = self._html_search_meta('DC.title', webpage, fatal=True)
427 camtasia_url = compat_urlparse.urljoin(url, camtasia_cfg)
428 camtasia_cfg = self._download_xml(
429 camtasia_url, video_id,
430 note='Downloading camtasia configuration',
431 errnote='Failed to download camtasia configuration')
432 fileset_node = camtasia_cfg.find('./playlist/array/fileset')
435 for n in fileset_node.getchildren():
436 url_n = n.find('./uri')
441 'id': os.path.splitext(url_n.text.rpartition('/')[2])[0],
442 'title': '%s - %s' % (title, n.tag),
443 'url': compat_urlparse.urljoin(url, url_n.text),
444 'duration': float_or_none(n.find('./duration').text),
453 def _real_extract(self, url):
454 if url.startswith('//'):
457 'url': self.http_scheme() + url,
460 parsed_url = compat_urlparse.urlparse(url)
461 if not parsed_url.scheme:
462 default_search = self._downloader.params.get('default_search')
463 if default_search is None:
464 default_search = 'fixup_error'
466 if default_search in ('auto', 'auto_warning', 'fixup_error'):
468 self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http')
469 return self.url_result('http://' + url)
470 elif default_search != 'fixup_error':
471 if default_search == 'auto_warning':
472 if re.match(r'^(?:url|URL)$', url):
473 raise ExtractorError(
474 'Invalid URL: %r . Call youtube-dl like this: youtube-dl -v "https://www.youtube.com/watch?v=BaW_jenozKc" ' % url,
477 self._downloader.report_warning(
478 'Falling back to youtube search for %s . Set --default-search "auto" to suppress this warning.' % url)
479 return self.url_result('ytsearch:' + url)
481 if default_search in ('error', 'fixup_error'):
482 raise ExtractorError(
483 ('%r is not a valid URL. '
484 'Set --default-search "ytsearch" (or run youtube-dl "ytsearch:%s" ) to search YouTube'
485 ) % (url, url), expected=True)
487 assert ':' in default_search
488 return self.url_result(default_search + url)
490 url, smuggled_data = unsmuggle_url(url)
492 if smuggled_data and 'force_videoid' in smuggled_data:
493 force_videoid = smuggled_data['force_videoid']
494 video_id = force_videoid
496 video_id = os.path.splitext(url.rstrip('/').split('/')[-1])[0]
498 self.to_screen('%s: Requesting header' % video_id)
500 head_req = HEADRequest(url)
501 response = self._request_webpage(
503 note=False, errnote='Could not send HEAD request to %s' % url,
506 if response is not False:
508 new_url = response.geturl()
510 self.report_following_redirect(new_url)
512 new_url = smuggle_url(
513 new_url, {'force_videoid': force_videoid})
514 return self.url_result(new_url)
516 # Check for direct link to a video
517 content_type = response.headers.get('Content-Type', '')
518 m = re.match(r'^(?P<type>audio|video|application(?=/ogg$))/(?P<format_id>.+)$', content_type)
520 upload_date = response.headers.get('Last-Modified')
522 upload_date = unified_strdate(upload_date)
525 'title': os.path.splitext(url_basename(url))[0],
527 'format_id': m.group('format_id'),
529 'vcodec': 'none' if m.group('type') == 'audio' else None
531 'upload_date': upload_date,
535 webpage = self._download_webpage(url, video_id)
537 # since this is the last-resort InfoExtractor, if
538 # this error is thrown, it'll be thrown here
539 raise ExtractorError('Failed to download URL: %s' % url)
541 self.report_extraction(video_id)
545 doc = parse_xml(webpage)
547 return self._extract_rss(url, video_id, doc)
548 except compat_xml_parse_error:
551 # Is it a Camtasia project?
552 camtasia_res = self._extract_camtasia(url, video_id, webpage)
553 if camtasia_res is not None:
556 # Sometimes embedded video player is hidden behind percent encoding
557 # (e.g. https://github.com/rg3/youtube-dl/issues/2448)
558 # Unescaping the whole page allows to handle those cases in a generic way
559 webpage = compat_urllib_parse.unquote(webpage)
561 # it's tempting to parse this further, but you would
562 # have to take into account all the variations like
563 # Video Title - Site Name
564 # Site Name | Video Title
565 # Video Title - Tagline | Site Name
566 # and so on and so forth; it's just not practical
567 video_title = self._html_search_regex(
568 r'(?s)<title>(.*?)</title>', webpage, 'video title',
571 # Try to detect age limit automatically
572 age_limit = self._rta_search(webpage)
573 # And then there are the jokers who advertise that they use RTA,
574 # but actually don't.
575 AGE_LIMIT_MARKERS = [
576 r'Proudly Labeled <a href="http://www.rtalabel.org/" title="Restricted to Adults">RTA</a>',
578 if any(re.search(marker, webpage) for marker in AGE_LIMIT_MARKERS):
581 # video uploader is domain name
582 video_uploader = self._search_regex(
583 r'^(?:https?://)?([^/]*)/.*', url, 'video uploader')
586 def _playlist_from_matches(matches, getter, ie=None):
588 self.url_result(self._proto_relative_url(getter(m)), ie)
590 return self.playlist_result(
591 urlrs, playlist_id=video_id, playlist_title=video_title)
593 # Look for BrightCove:
594 bc_urls = BrightcoveIE._extract_brightcove_urls(webpage)
596 self.to_screen('Brightcove video detected.')
599 'url': smuggle_url(bc_url, {'Referer': url}),
600 'ie_key': 'Brightcove'
601 } for bc_url in bc_urls]
605 'title': video_title,
610 # Look for embedded (iframe) Vimeo player
612 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//player\.vimeo\.com/video/.+?)\1', webpage)
614 player_url = unescapeHTML(mobj.group('url'))
615 surl = smuggle_url(player_url, {'Referer': url})
616 return self.url_result(surl, 'Vimeo')
618 # Look for embedded (swf embed) Vimeo player
620 r'<embed[^>]+?src="(https?://(?:www\.)?vimeo\.com/moogaloop\.swf.+?)"', webpage)
622 return self.url_result(mobj.group(1), 'Vimeo')
624 # Look for embedded YouTube player
625 matches = re.findall(r'''(?x)
633 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
637 return _playlist_from_matches(
638 matches, lambda m: unescapeHTML(m[1]))
640 # Look for embedded Dailymotion player
641 matches = re.findall(
642 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/embed/video/.+?)\1', webpage)
644 return _playlist_from_matches(
645 matches, lambda m: unescapeHTML(m[1]))
647 # Look for embedded Wistia player
649 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage)
652 '_type': 'url_transparent',
653 'url': unescapeHTML(match.group('url')),
655 'uploader': video_uploader,
656 'title': video_title,
660 # Look for embedded blip.tv player
661 mobj = re.search(r'<meta\s[^>]*https?://api\.blip\.tv/\w+/redirect/\w+/(\d+)', webpage)
663 return self.url_result('http://blip.tv/a/a-'+mobj.group(1), 'BlipTV')
664 mobj = re.search(r'<(?:iframe|embed|object)\s[^>]*(https?://(?:\w+\.)?blip\.tv/(?:play/|api\.swf#)[a-zA-Z0-9_]+)', webpage)
666 return self.url_result(mobj.group(1), 'BlipTV')
668 # Look for embedded condenast player
669 matches = re.findall(
670 r'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="(https?://player\.cnevids\.com/embed/[^"]+")',
677 'ie_key': 'CondeNast',
679 } for ma in matches],
680 'title': video_title,
684 # Look for Bandcamp pages with custom domain
685 mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage)
687 burl = unescapeHTML(mobj.group(1))
688 # Don't set the extractor because it can be a track url or an album
689 return self.url_result(burl)
691 # Look for embedded Vevo player
693 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:cache\.)?vevo\.com/.+?)\1', webpage)
695 return self.url_result(mobj.group('url'))
697 # Look for Ooyala videos
698 mobj = (re.search(r'player.ooyala.com/[^"?]+\?[^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or
699 re.search(r'OO.Player.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage))
701 return OoyalaIE._build_url_result(mobj.group('ec'))
703 # Look for Aparat videos
704 mobj = re.search(r'<iframe .*?src="(http://www\.aparat\.com/video/[^"]+)"', webpage)
706 return self.url_result(mobj.group(1), 'Aparat')
708 # Look for MPORA videos
709 mobj = re.search(r'<iframe .*?src="(http://mpora\.(?:com|de)/videos/[^"]+)"', webpage)
711 return self.url_result(mobj.group(1), 'Mpora')
713 # Look for embedded NovaMov-based player
715 r'''(?x)<(?:pagespeed_)?iframe[^>]+?src=(["\'])
716 (?P<url>http://(?:(?:embed|www)\.)?
718 nowvideo\.(?:ch|sx|eu|at|ag|co)|
719 videoweed\.(?:es|com)|
720 movshare\.(?:net|sx|ag)|
721 divxstage\.(?:eu|net|ch|co|at|ag))
722 /embed\.php.+?)\1''', webpage)
724 return self.url_result(mobj.group('url'))
726 # Look for embedded Facebook player
728 r'<iframe[^>]+?src=(["\'])(?P<url>https://www\.facebook\.com/video/embed.+?)\1', webpage)
730 return self.url_result(mobj.group('url'), 'Facebook')
732 # Look for embedded VK player
733 mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://vk\.com/video_ext\.php.+?)\1', webpage)
735 return self.url_result(mobj.group('url'), 'VK')
737 # Look for embedded ivi player
738 mobj = re.search(r'<embed[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?ivi\.ru/video/player.+?)\1', webpage)
740 return self.url_result(mobj.group('url'), 'Ivi')
742 # Look for embedded Huffington Post player
744 r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed\.live\.huffingtonpost\.com/.+?)\1', webpage)
746 return self.url_result(mobj.group('url'), 'HuffPost')
749 mobj = re.search(r'class=["\']embedly-card["\'][^>]href=["\'](?P<url>[^"\']+)', webpage)
751 return self.url_result(mobj.group('url'))
752 mobj = re.search(r'class=["\']embedly-embed["\'][^>]src=["\'][^"\']*url=(?P<url>[^&]+)', webpage)
754 return self.url_result(compat_urllib_parse.unquote(mobj.group('url')))
756 # Look for funnyordie embed
757 matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage)
759 return _playlist_from_matches(
760 matches, getter=unescapeHTML, ie='FunnyOrDie')
762 # Look for embedded RUTV player
763 rutv_url = RUTVIE._extract_url(webpage)
765 return self.url_result(rutv_url, 'RUTV')
767 # Look for embedded TED player
769 r'<iframe[^>]+?src=(["\'])(?P<url>http://embed\.ted\.com/.+?)\1', webpage)
771 return self.url_result(mobj.group('url'), 'TED')
773 # Look for embedded Ustream videos
775 r'<iframe[^>]+?src=(["\'])(?P<url>http://www\.ustream\.tv/embed/.+?)\1', webpage)
777 return self.url_result(mobj.group('url'), 'Ustream')
779 # Look for embedded arte.tv player
781 r'<script [^>]*?src="(?P<url>http://www\.arte\.tv/playerv2/embed[^"]+)"',
784 return self.url_result(mobj.group('url'), 'ArteTVEmbed')
786 # Look for embedded smotri.com player
787 smotri_url = SmotriIE._extract_url(webpage)
789 return self.url_result(smotri_url, 'Smotri')
791 # Look for embeded soundcloud player
793 r'<iframe src="(?P<url>https?://(?:w\.)?soundcloud\.com/player[^"]+)"',
796 url = unescapeHTML(mobj.group('url'))
797 return self.url_result(url)
799 # Look for embedded vulture.com player
801 r'<iframe src="(?P<url>https?://video\.vulture\.com/[^"]+)"',
804 url = unescapeHTML(mobj.group('url'))
805 return self.url_result(url, ie='Vulture')
807 # Look for embedded mtvservices player
809 r'<iframe src="(?P<url>https?://media\.mtvnservices\.com/embed/[^"]+)"',
812 url = unescapeHTML(mobj.group('url'))
813 return self.url_result(url, ie='MTVServicesEmbedded')
815 # Look for embedded yahoo player
817 r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:screen|movies)\.yahoo\.com/.+?\.html\?format=embed)\1',
820 return self.url_result(mobj.group('url'), 'Yahoo')
822 # Look for embedded sbs.com.au player
824 r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:www\.)sbs\.com\.au/ondemand/video/single/.+?)\1',
827 return self.url_result(mobj.group('url'), 'SBS')
830 r'<iframe[^>]+?src=(["\'])(?P<url>https?://m\.mlb\.com/shared/video/embed/embed\.html\?.+?)\1',
833 return self.url_result(mobj.group('url'), 'MLB')
835 # Start with something easy: JW Player in SWFObject
836 found = re.findall(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage)
838 # Look for gorilla-vid style embedding
839 found = re.findall(r'''(?sx)
843 jwplayer\s*\(\s*["'][^'"]+["']\s*\)\s*\.setup
845 .*?file\s*:\s*["\'](.*?)["\']''', webpage)
847 # Broaden the search a little bit
848 found = re.findall(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage)
850 # Broaden the findall a little bit: JWPlayer JS loader
851 found = re.findall(r'[^A-Za-z0-9]?file["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage)
854 found = re.findall(r'''(?xs)
855 flowplayer\("[^"]+",\s*
857 \s*{[^}]+? ["']?clip["']?\s*:\s*\{\s*
858 ["']?url["']?\s*:\s*["']([^"']+)["']
861 # Try to find twitter cards info
862 found = re.findall(r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage)
864 # We look for Open Graph info:
865 # We have to match any number spaces between elements, some sites try to align them (eg.: statigr.am)
866 m_video_type = re.findall(r'<meta.*?property="og:video:type".*?content="video/(.*?)"', webpage)
867 # We only look in og:video if the MIME type is a video, don't try if it's a Flash player:
868 if m_video_type is not None:
869 def check_video(vurl):
870 vpath = compat_urlparse.urlparse(vurl).path
871 vext = determine_ext(vpath)
872 return '.' in vpath and vext not in ('swf', 'png', 'jpg')
875 re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage)))
878 found = re.findall(r'(?s)<video[^<]*(?:>.*?<source[^>]+)? src="([^"]+)"', webpage)
881 r'(?i)<meta\s+(?=(?:[a-z-]+="[^"]+"\s+)*http-equiv="refresh")'
882 r'(?:[a-z-]+="[^"]+"\s+)*?content="[0-9]{,2};url=\'?([^\'"]+)',
885 new_url = found.group(1)
886 self.report_following_redirect(new_url)
892 raise ExtractorError('Unsupported URL: %s' % url)
895 for video_url in found:
896 video_url = compat_urlparse.urljoin(url, video_url)
897 video_id = compat_urllib_parse.unquote(os.path.basename(video_url))
899 # Sometimes, jwplayer extraction will result in a YouTube URL
900 if YoutubeIE.suitable(video_url):
901 entries.append(self.url_result(video_url, 'Youtube'))
904 # here's a fun little line of code for you:
905 video_id = os.path.splitext(video_id)[0]
910 'uploader': video_uploader,
911 'title': video_title,
912 'age_limit': age_limit,
915 if len(entries) == 1:
918 for num, e in enumerate(entries, start=1):
919 e['title'] = '%s (%d)' % (e['title'], num)