3 from __future__ import unicode_literals
8 from .common import InfoExtractor
9 from .youtube import YoutubeIE
10 from ..compat import (
13 compat_xml_parse_error,
30 from .brightcove import BrightcoveIE
31 from .ooyala import OoyalaIE
32 from .rutv import RUTVIE
33 from .smotri import SmotriIE
34 from .condenast import CondeNastIE
37 class GenericIE(InfoExtractor):
38 IE_DESC = 'Generic downloader that works on some sites'
43 'url': 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
44 'md5': '85b90ccc9d73b4acd9138d3af4c27f89',
46 'id': '13601338388002',
48 'uploader': 'www.hodiho.fr',
49 'title': 'R\u00e9gis plante sa Jeep',
52 # bandcamp page with custom domain
54 'add_ie': ['Bandcamp'],
55 'url': 'http://bronyrock.com/track/the-pony-mash',
59 'title': 'The Pony Mash',
60 'uploader': 'M_Pallante',
62 'skip': 'There is a limit of 200 free downloads / month for the test song',
64 # embedded brightcove video
65 # it also tests brightcove videos that need to set the 'Referer' in the
68 'add_ie': ['Brightcove'],
69 'url': 'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/',
71 'id': '2765128793001',
73 'title': 'Le cours de bourse : l’analyse technique',
74 'description': 'md5:7e9ad046e968cb2d1114004aba466fd9',
75 'uploader': 'BFM BUSINESS',
78 'skip_download': True,
82 # https://github.com/rg3/youtube-dl/issues/2253
83 'url': 'http://bcove.me/i6nfkrc3',
84 'md5': '0ba9446db037002366bab3b3eb30c88c',
86 'id': '3101154703001',
88 'title': 'Still no power',
89 'uploader': 'thestar.com',
90 'description': 'Mississauga resident David Farmer is still out of power as a result of the ice storm a month ago. To keep the house warm, Farmer cuts wood from his property for a wood burning stove downstairs.',
92 'add_ie': ['Brightcove'],
95 'url': 'http://www.championat.com/video/football/v/87/87499.html',
96 'md5': 'fb973ecf6e4a78a67453647444222983',
98 'id': '3414141473001',
100 'title': 'Видео. Удаление Дзагоева (ЦСКА)',
101 'description': 'Онлайн-трансляция матча ЦСКА - "Волга"',
102 'uploader': 'Championat',
106 # https://github.com/rg3/youtube-dl/issues/3541
107 'add_ie': ['Brightcove'],
108 'url': 'http://www.kijk.nl/sbs6/leermijvrouwenkennen/videos/jqMiXKAYan2S/aflevering-1',
110 'id': '3866516442001',
112 'title': 'Leer mij vrouwen kennen: Aflevering 1',
113 'description': 'Leer mij vrouwen kennen: Aflevering 1',
114 'uploader': 'SBS Broadcasting',
116 'skip': 'Restricted to Netherlands',
118 'skip_download': True, # m3u8 download
121 # Direct link to a video
123 'url': 'http://media.w3.org/2010/05/sintel/trailer.mp4',
124 'md5': '67d406c2bcb6af27fa886f31aa934bbe',
129 'upload_date': '20100513',
134 'url': 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219',
135 'md5': '166dd577b433b4d4ebfee10b0824d8ff',
137 'id': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ',
139 'title': '2cc213299525360.mov', # that's what we get
141 'add_ie': ['Ooyala'],
145 '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',
149 'upload_date': '20130224',
150 'uploader_id': 'TheVerge',
151 'description': 're:^Chris Ziegler takes a look at the\.*',
152 'uploader': 'The Verge',
153 'title': 'First Firefox OS phones side-by-side',
156 'skip_download': False,
161 'url': 'http://www.tested.com/science/weird/460206-tested-grinding-coffee-2000-frames-second/',
165 'title': 'Tested: Grinding Coffee at 2000 Frames Per Second',
166 'upload_date': '20140225',
167 'description': 'md5:06a40fbf30b220468f1e0957c0f558ff',
168 'uploader': 'Tested',
169 'uploader_id': 'testedcom',
171 # No need to test YoutubeIE here
173 'skip_download': True,
178 'url': 'http://www.theguardian.com/world/2014/mar/11/obama-zach-galifianakis-between-two-ferns',
182 'title': 'Between Two Ferns with Zach Galifianakis: President Barack Obama',
183 'description': 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.',
188 'url': 'http://www.bbc.co.uk/blogs/adamcurtis/posts/BUGGER',
190 'title': 'BBC - Blogs - Adam Curtis - BUGGER',
192 'playlist_mincount': 18,
196 'url': 'http://www.rg.ru/2014/03/15/reg-dfo/anklav-anons.html',
200 'title': 'Охотское море стало целиком российским',
201 'description': 'md5:5ed62483b14663e2a95ebbe115eb8f43',
205 'skip_download': True,
210 'url': 'http://en.support.wordpress.com/videos/ted-talks/',
211 'md5': '65fdff94098e4a607385a60c5177c638',
215 'title': 'Hidden miracles of the natural world',
216 'uploader': 'Louie Schwartzberg',
217 'description': 'md5:8145d19d320ff3e52f28401f4c4283b9',
220 # Embeded Ustream video
222 'url': 'http://www.american.edu/spa/pti/nsa-privacy-janus-2014.cfm',
223 'md5': '27b99cdb639c9b12a79bca876a073417',
227 'uploader': 'AU SPA: The NSA and Privacy',
228 'title': 'NSA and Privacy Forum Debate featuring General Hayden and Barton Gellman'
231 # nowvideo embed hidden behind percent encoding
233 'url': 'http://www.waoanime.tv/the-super-dimension-fortress-macross-episode-1/',
234 'md5': '2baf4ddd70f697d94b1c18cf796d5107',
236 'id': '06e53103ca9aa',
238 'title': 'Macross Episode 001 Watch Macross Episode 001 onl',
239 'description': 'No description',
244 'url': 'http://www.tv-replay.fr/redirection/20-03-14/x-enius-arte-10753389.html',
245 'md5': '7653032cbb25bf6c80d80f217055fa43',
247 'id': '048195-004_PLUS7-F',
250 'description': 'md5:d5fdf32ef6613cdbfd516ae658abf168',
251 'upload_date': '20140320',
254 'skip_download': 'Requires rtmpdump'
259 'url': 'http://www.wired.com/2014/04/honda-asimo/',
260 'md5': 'ba0dfe966fa007657bd1443ee672db0f',
262 'id': '53501be369702d3275860000',
264 'title': 'Honda’s New Asimo Robot Is More Human Than Ever',
269 'url': 'http://www.spi0n.com/zap-spi0n-com-n216/',
270 'md5': '441aeeb82eb72c422c7f14ec533999cd',
272 'id': 'k2mm4bCdJ6CQ2i7c8o2',
274 'title': 'Le Zap de Spi0n n°216 - Zapping du Web',
277 'add_ie': ['Dailymotion'],
281 'url': 'http://www.badzine.de/ansicht/datum/2014/06/09/so-funktioniert-die-neue-englische-badminton-liga.html',
285 'title': 'The NBL Auction 2014',
286 'uploader': 'BADMINTON England',
287 'uploader_id': 'BADMINTONEvents',
288 'upload_date': '20140603',
289 'description': 'md5:9ef128a69f1e262a700ed83edb163a73',
291 'add_ie': ['Youtube'],
293 'skip_download': True,
298 'url': 'http://www.gametrailers.com/news-post/76093/north-america-europe-is-getting-that-mario-kart-8-mercedes-dlc-too',
299 'md5': '35727f82f58c76d996fc188f9755b0d5',
301 'id': '0306a69b-8adf-4fb5-aace-75f8e8cbfca9',
304 'description': 'Mario\'s life in the fast lane has never looked so good.',
307 # YouTube embed via <data-embed-url="">
309 'url': 'https://play.google.com/store/apps/details?id=com.gameloft.android.ANMP.GloftA8HM',
313 'title': 'Asphalt 8: Airborne - Update - Welcome to Dubai!',
314 'uploader': 'Gameloft',
315 'uploader_id': 'gameloft',
316 'upload_date': '20140828',
317 'description': 'md5:c80da9ed3d83ae6d1876c834de03e1c4',
320 'skip_download': True,
325 'url': 'http://www.ll.mit.edu/workshops/education/videocourses/antennas/lecture1/video/',
327 'md5': '0c5e352edabf715d762b0ad4e6d9ee67',
329 'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
330 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - video1',
335 'md5': '10e4bb3aaca9fd630e273ff92d9f3c63',
337 'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final_PIP',
338 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - pip',
344 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
349 'url': 'http://www.handjobhub.com/video/busty-blonde-siri-tit-fuck-while-wank-6313.html',
350 'md5': '9d65602bf31c6e20014319c7d07fba27',
352 'id': '5123ea6d5e5a7',
355 'uploader': 'www.handjobhub.com',
356 'title': 'Busty Blonde Siri Tit Fuck While Wank at HandjobHub.com',
361 'url': 'http://phihag.de/2014/youtube-dl/rss2.xml',
363 'id': 'http://phihag.de/2014/youtube-dl/rss2.xml',
364 'title': 'Zero Punctuation',
367 'playlist_mincount': 11,
369 # Multiple brightcove videos
370 # https://github.com/rg3/youtube-dl/issues/2283
372 'url': 'http://www.newyorker.com/online/blogs/newsdesk/2014/01/always-never-nuclear-command-and-control.html',
374 'id': 'always-never',
375 'title': 'Always / Never - The New Yorker',
379 'extract_flat': False,
380 'skip_download': True,
385 'url': 'http://umpire-empire.com/index.php/topic/58125-laz-decides-no-thats-low/',
386 'md5': '96f09a37e44da40dd083e12d9a683327',
390 'title': 'Ump changes call to ball',
391 'description': 'md5:71c11215384298a172a6dcb4c2e20685',
393 'timestamp': 1401537900,
394 'upload_date': '20140531',
395 'thumbnail': 're:^https?://.*\.jpg$',
400 'url': 'http://education-portal.com/academy/lesson/north-american-exploration-failed-colonies-of-spain-france-england.html#lesson',
401 'md5': '8788b683c777a5cf25621eaf286d0c23',
405 'title': 'md5:51364a8d3d009997ba99656004b5e20d',
407 'filesize': 182808282,
408 'uploader': 'education-portal.com',
412 'url': 'http://thoughtworks.wistia.com/medias/uxjb0lwrcz',
413 'md5': 'baf49c2baa8a7de5f3fc145a8506dcd4',
417 'title': 'Conversation about Hexagonal Rails Part 1 - ThoughtWorks',
419 'uploader': 'thoughtworks.wistia.com',
422 # Direct download with broken HEAD
424 'url': 'http://ai-radio.org:8000/radio.opus',
431 'skip_download': True, # infinite live stream
433 'expected_warnings': [
434 r'501.*Not Implemented'
439 'url': 'http://nakedsecurity.sophos.com/2014/10/29/sscc-171-are-you-sure-that-1234-is-a-bad-password-podcast/',
443 'description': 'md5:ff867d6b555488ad3c52572bb33d432c',
444 'uploader': 'Sophos Security',
445 'title': 'Chet Chat 171 - Oct 29, 2014',
446 'upload_date': '20141029',
451 'url': 'http://www.esa.int/Our_Activities/Space_Science/Rosetta/Philae_comet_touch-down_webcast',
455 'upload_date': '20141112',
456 'title': 'Rosetta #CometLanding webcast HL 10',
461 'url': 'http://discourse.ubuntu.com/t/unity-8-desktop-mode-windows-on-mir/1986',
463 'title': 'Unity 8 desktop-mode windows on Mir! - Ubuntu Discourse',
465 'playlist_mincount': 2,
467 # Direct link with incorrect MIME type
469 'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
470 'md5': '4ccbebe5f36706d85221f204d7eb5913',
472 'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
473 'id': '5_Lennart_Poettering_-_Systemd',
475 'title': '5_Lennart_Poettering_-_Systemd',
476 'upload_date': '20141120',
478 'expected_warnings': [
479 'URL could be a direct video link, returning it as such.'
484 'url': 'http://undergroundwellness.com/podcasts/306-5-steps-to-permanent-gut-healing/',
488 'upload_date': '20141126',
489 'title': 'Jack Tips: 5 Steps to Permanent Gut Healing',
494 def report_following_redirect(self, new_url):
495 """Report information extraction."""
496 self._downloader.to_screen('[redirect] Following redirect to %s' % new_url)
498 def _extract_rss(self, url, video_id, doc):
499 playlist_title = doc.find('./channel/title').text
500 playlist_desc_el = doc.find('./channel/description')
501 playlist_desc = None if playlist_desc_el is None else playlist_desc_el.text
505 'url': e.find('link').text,
506 'title': e.find('title').text,
507 } for e in doc.findall('./channel/item')]
512 'title': playlist_title,
513 'description': playlist_desc,
517 def _extract_camtasia(self, url, video_id, webpage):
518 """ Returns None if no camtasia video can be found. """
520 camtasia_cfg = self._search_regex(
521 r'fo\.addVariable\(\s*"csConfigFile",\s*"([^"]+)"\s*\);',
522 webpage, 'camtasia configuration file', default=None)
523 if camtasia_cfg is None:
526 title = self._html_search_meta('DC.title', webpage, fatal=True)
528 camtasia_url = compat_urlparse.urljoin(url, camtasia_cfg)
529 camtasia_cfg = self._download_xml(
530 camtasia_url, video_id,
531 note='Downloading camtasia configuration',
532 errnote='Failed to download camtasia configuration')
533 fileset_node = camtasia_cfg.find('./playlist/array/fileset')
536 for n in fileset_node.getchildren():
537 url_n = n.find('./uri')
542 'id': os.path.splitext(url_n.text.rpartition('/')[2])[0],
543 'title': '%s - %s' % (title, n.tag),
544 'url': compat_urlparse.urljoin(url, url_n.text),
545 'duration': float_or_none(n.find('./duration').text),
554 def _real_extract(self, url):
555 if url.startswith('//'):
558 'url': self.http_scheme() + url,
561 parsed_url = compat_urlparse.urlparse(url)
562 if not parsed_url.scheme:
563 default_search = self._downloader.params.get('default_search')
564 if default_search is None:
565 default_search = 'fixup_error'
567 if default_search in ('auto', 'auto_warning', 'fixup_error'):
569 self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http')
570 return self.url_result('http://' + url)
571 elif default_search != 'fixup_error':
572 if default_search == 'auto_warning':
573 if re.match(r'^(?:url|URL)$', url):
574 raise ExtractorError(
575 'Invalid URL: %r . Call youtube-dl like this: youtube-dl -v "https://www.youtube.com/watch?v=BaW_jenozKc" ' % url,
578 self._downloader.report_warning(
579 'Falling back to youtube search for %s . Set --default-search "auto" to suppress this warning.' % url)
580 return self.url_result('ytsearch:' + url)
582 if default_search in ('error', 'fixup_error'):
583 raise ExtractorError(
584 '%r is not a valid URL. '
585 'Set --default-search "ytsearch" (or run youtube-dl "ytsearch:%s" ) to search YouTube'
586 % (url, url), expected=True)
588 if ':' not in default_search:
589 default_search += ':'
590 return self.url_result(default_search + url)
592 url, smuggled_data = unsmuggle_url(url)
594 is_intentional = smuggled_data and smuggled_data.get('to_generic')
595 if smuggled_data and 'force_videoid' in smuggled_data:
596 force_videoid = smuggled_data['force_videoid']
597 video_id = force_videoid
599 video_id = os.path.splitext(url.rstrip('/').split('/')[-1])[0]
601 self.to_screen('%s: Requesting header' % video_id)
603 head_req = HEADRequest(url)
604 head_response = self._request_webpage(
606 note=False, errnote='Could not send HEAD request to %s' % url,
609 if head_response is not False:
611 new_url = head_response.geturl()
613 self.report_following_redirect(new_url)
615 new_url = smuggle_url(
616 new_url, {'force_videoid': force_videoid})
617 return self.url_result(new_url)
620 if head_response is False:
621 full_response = self._request_webpage(url, video_id)
622 head_response = full_response
624 # Check for direct link to a video
625 content_type = head_response.headers.get('Content-Type', '')
626 m = re.match(r'^(?P<type>audio|video|application(?=/ogg$))/(?P<format_id>.+)$', content_type)
628 upload_date = unified_strdate(
629 head_response.headers.get('Last-Modified'))
632 'title': os.path.splitext(url_basename(url))[0],
635 'format_id': m.group('format_id'),
637 'vcodec': 'none' if m.group('type') == 'audio' else None
639 'upload_date': upload_date,
642 if not self._downloader.params.get('test', False) and not is_intentional:
643 self._downloader.report_warning('Falling back on generic information extractor.')
645 if not full_response:
646 full_response = self._request_webpage(url, video_id)
648 # Maybe it's a direct link to a video?
649 # Be careful not to download the whole thing!
650 first_bytes = full_response.read(512)
651 if not is_html(first_bytes):
652 self._downloader.report_warning(
653 'URL could be a direct video link, returning it as such.')
654 upload_date = unified_strdate(
655 head_response.headers.get('Last-Modified'))
658 'title': os.path.splitext(url_basename(url))[0],
661 'upload_date': upload_date,
664 webpage = self._webpage_read_content(
665 full_response, url, video_id, prefix=first_bytes)
667 self.report_extraction(video_id)
671 doc = parse_xml(webpage)
673 return self._extract_rss(url, video_id, doc)
674 except compat_xml_parse_error:
677 # Is it a Camtasia project?
678 camtasia_res = self._extract_camtasia(url, video_id, webpage)
679 if camtasia_res is not None:
682 # Sometimes embedded video player is hidden behind percent encoding
683 # (e.g. https://github.com/rg3/youtube-dl/issues/2448)
684 # Unescaping the whole page allows to handle those cases in a generic way
685 webpage = compat_urllib_parse.unquote(webpage)
687 # it's tempting to parse this further, but you would
688 # have to take into account all the variations like
689 # Video Title - Site Name
690 # Site Name | Video Title
691 # Video Title - Tagline | Site Name
692 # and so on and so forth; it's just not practical
693 video_title = self._html_search_regex(
694 r'(?s)<title>(.*?)</title>', webpage, 'video title',
697 # Try to detect age limit automatically
698 age_limit = self._rta_search(webpage)
699 # And then there are the jokers who advertise that they use RTA,
700 # but actually don't.
701 AGE_LIMIT_MARKERS = [
702 r'Proudly Labeled <a href="http://www.rtalabel.org/" title="Restricted to Adults">RTA</a>',
704 if any(re.search(marker, webpage) for marker in AGE_LIMIT_MARKERS):
707 # video uploader is domain name
708 video_uploader = self._search_regex(
709 r'^(?:https?://)?([^/]*)/.*', url, 'video uploader')
712 def _playlist_from_matches(matches, getter=None, ie=None):
714 self.url_result(self._proto_relative_url(getter(m) if getter else m), ie)
716 return self.playlist_result(
717 urlrs, playlist_id=video_id, playlist_title=video_title)
719 # Look for BrightCove:
720 bc_urls = BrightcoveIE._extract_brightcove_urls(webpage)
722 self.to_screen('Brightcove video detected.')
725 'url': smuggle_url(bc_url, {'Referer': url}),
726 'ie_key': 'Brightcove'
727 } for bc_url in bc_urls]
731 'title': video_title,
736 # Look for embedded (iframe) Vimeo player
738 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//player\.vimeo\.com/video/.+?)\1', webpage)
740 player_url = unescapeHTML(mobj.group('url'))
741 surl = smuggle_url(player_url, {'Referer': url})
742 return self.url_result(surl)
744 # Look for embedded (swf embed) Vimeo player
746 r'<embed[^>]+?src="((?:https?:)?//(?:www\.)?vimeo\.com/moogaloop\.swf.+?)"', webpage)
748 return self.url_result(mobj.group(1))
750 # Look for embedded YouTube player
751 matches = re.findall(r'''(?x)
760 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
764 return _playlist_from_matches(
765 matches, lambda m: unescapeHTML(m[1]))
767 # Look for lazyYT YouTube embed
768 matches = re.findall(
769 r'class="lazyYT" data-youtube-id="([^"]+)"', webpage)
771 return _playlist_from_matches(matches, lambda m: unescapeHTML(m))
773 # Look for embedded Dailymotion player
774 matches = re.findall(
775 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/embed/video/.+?)\1', webpage)
777 return _playlist_from_matches(
778 matches, lambda m: unescapeHTML(m[1]))
780 # Look for embedded Dailymotion playlist player (#3822)
782 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.[a-z]{2,3}/widget/jukebox\?.+?)\1', webpage)
784 playlists = re.findall(
785 r'list\[\]=/playlist/([^/]+)/', unescapeHTML(m.group('url')))
787 return _playlist_from_matches(
788 playlists, lambda p: '//dailymotion.com/playlist/%s' % p)
790 # Look for embedded Wistia player
792 r'<(?:meta[^>]+?content|iframe[^>]+?src)=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage)
794 embed_url = self._proto_relative_url(
795 unescapeHTML(match.group('url')))
797 '_type': 'url_transparent',
800 'uploader': video_uploader,
801 'title': video_title,
805 match = re.search(r'(?:id=["\']wistia_|data-wistia-?id=["\']|Wistia\.embed\(["\'])(?P<id>[^"\']+)', webpage)
808 '_type': 'url_transparent',
809 'url': 'http://fast.wistia.net/embed/iframe/{0:}'.format(match.group('id')),
811 'uploader': video_uploader,
812 'title': video_title,
813 'id': match.group('id')
816 # Look for embedded blip.tv player
817 mobj = re.search(r'<meta\s[^>]*https?://api\.blip\.tv/\w+/redirect/\w+/(\d+)', webpage)
819 return self.url_result('http://blip.tv/a/a-' + mobj.group(1), 'BlipTV')
820 mobj = re.search(r'<(?:iframe|embed|object)\s[^>]*(https?://(?:\w+\.)?blip\.tv/(?:play/|api\.swf#)[a-zA-Z0-9_]+)', webpage)
822 return self.url_result(mobj.group(1), 'BlipTV')
824 # Look for embedded condenast player
825 matches = re.findall(
826 r'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="(https?://player\.cnevids\.com/embed/[^"]+")',
833 'ie_key': 'CondeNast',
835 } for ma in matches],
836 'title': video_title,
840 # Look for Bandcamp pages with custom domain
841 mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage)
843 burl = unescapeHTML(mobj.group(1))
844 # Don't set the extractor because it can be a track url or an album
845 return self.url_result(burl)
847 # Look for embedded Vevo player
849 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:cache\.)?vevo\.com/.+?)\1', webpage)
851 return self.url_result(mobj.group('url'))
853 # Look for Ooyala videos
854 mobj = (re.search(r'player.ooyala.com/[^"?]+\?[^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or
855 re.search(r'OO.Player.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage))
857 return OoyalaIE._build_url_result(mobj.group('ec'))
859 # Look for Aparat videos
860 mobj = re.search(r'<iframe .*?src="(http://www\.aparat\.com/video/[^"]+)"', webpage)
862 return self.url_result(mobj.group(1), 'Aparat')
864 # Look for MPORA videos
865 mobj = re.search(r'<iframe .*?src="(http://mpora\.(?:com|de)/videos/[^"]+)"', webpage)
867 return self.url_result(mobj.group(1), 'Mpora')
869 # Look for embedded NovaMov-based player
871 r'''(?x)<(?:pagespeed_)?iframe[^>]+?src=(["\'])
872 (?P<url>http://(?:(?:embed|www)\.)?
874 nowvideo\.(?:ch|sx|eu|at|ag|co)|
875 videoweed\.(?:es|com)|
876 movshare\.(?:net|sx|ag)|
877 divxstage\.(?:eu|net|ch|co|at|ag))
878 /embed\.php.+?)\1''', webpage)
880 return self.url_result(mobj.group('url'))
882 # Look for embedded Facebook player
884 r'<iframe[^>]+?src=(["\'])(?P<url>https://www\.facebook\.com/video/embed.+?)\1', webpage)
886 return self.url_result(mobj.group('url'), 'Facebook')
888 # Look for embedded VK player
889 mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://vk\.com/video_ext\.php.+?)\1', webpage)
891 return self.url_result(mobj.group('url'), 'VK')
893 # Look for embedded ivi player
894 mobj = re.search(r'<embed[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?ivi\.ru/video/player.+?)\1', webpage)
896 return self.url_result(mobj.group('url'), 'Ivi')
898 # Look for embedded Huffington Post player
900 r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed\.live\.huffingtonpost\.com/.+?)\1', webpage)
902 return self.url_result(mobj.group('url'), 'HuffPost')
905 mobj = re.search(r'class=["\']embedly-card["\'][^>]href=["\'](?P<url>[^"\']+)', webpage)
907 return self.url_result(mobj.group('url'))
908 mobj = re.search(r'class=["\']embedly-embed["\'][^>]src=["\'][^"\']*url=(?P<url>[^&]+)', webpage)
910 return self.url_result(compat_urllib_parse.unquote(mobj.group('url')))
912 # Look for funnyordie embed
913 matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage)
915 return _playlist_from_matches(
916 matches, getter=unescapeHTML, ie='FunnyOrDie')
918 # Look for BBC iPlayer embed
919 matches = re.findall(r'setPlaylist\("(https?://www\.bbc\.co\.uk/iplayer/[^/]+/[\da-z]{8})"\)', webpage)
921 return _playlist_from_matches(matches, ie='BBCCoUk')
923 # Look for embedded RUTV player
924 rutv_url = RUTVIE._extract_url(webpage)
926 return self.url_result(rutv_url, 'RUTV')
928 # Look for embedded TED player
930 r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed(?:-ssl)?\.ted\.com/.+?)\1', webpage)
932 return self.url_result(mobj.group('url'), 'TED')
934 # Look for embedded Ustream videos
936 r'<iframe[^>]+?src=(["\'])(?P<url>http://www\.ustream\.tv/embed/.+?)\1', webpage)
938 return self.url_result(mobj.group('url'), 'Ustream')
940 # Look for embedded arte.tv player
942 r'<script [^>]*?src="(?P<url>http://www\.arte\.tv/playerv2/embed[^"]+)"',
945 return self.url_result(mobj.group('url'), 'ArteTVEmbed')
947 # Look for embedded smotri.com player
948 smotri_url = SmotriIE._extract_url(webpage)
950 return self.url_result(smotri_url, 'Smotri')
952 # Look for embeded soundcloud player
954 r'<iframe\s+(?:[a-zA-Z0-9_-]+="[^"]+"\s+)*src="(?P<url>https?://(?:w\.)?soundcloud\.com/player[^"]+)"',
957 url = unescapeHTML(mobj.group('url'))
958 return self.url_result(url)
960 # Look for embedded vulture.com player
962 r'<iframe src="(?P<url>https?://video\.vulture\.com/[^"]+)"',
965 url = unescapeHTML(mobj.group('url'))
966 return self.url_result(url, ie='Vulture')
968 # Look for embedded mtvservices player
970 r'<iframe src="(?P<url>https?://media\.mtvnservices\.com/embed/[^"]+)"',
973 url = unescapeHTML(mobj.group('url'))
974 return self.url_result(url, ie='MTVServicesEmbedded')
976 # Look for embedded yahoo player
978 r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:screen|movies)\.yahoo\.com/.+?\.html\?format=embed)\1',
981 return self.url_result(mobj.group('url'), 'Yahoo')
983 # Look for embedded sbs.com.au player
985 r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:www\.)sbs\.com\.au/ondemand/video/single/.+?)\1',
988 return self.url_result(mobj.group('url'), 'SBS')
990 # Look for embedded Cinchcast player
992 r'<iframe[^>]+?src=(["\'])(?P<url>https?://player\.cinchcast\.com/.+?)\1',
995 return self.url_result(mobj.group('url'), 'Cinchcast')
998 r'<iframe[^>]+?src=(["\'])(?P<url>https?://m(?:lb)?\.mlb\.com/shared/video/embed/embed\.html\?.+?)\1',
1000 if mobj is not None:
1001 return self.url_result(mobj.group('url'), 'MLB')
1004 r'<iframe[^>]+?src=(["\'])(?P<url>%s)\1' % CondeNastIE.EMBED_URL,
1006 if mobj is not None:
1007 return self.url_result(self._proto_relative_url(mobj.group('url'), scheme='http:'), 'CondeNast')
1010 r'<iframe[^>]+src="(?P<url>https?://new\.livestream\.com/[^"]+/player[^"]+)"',
1012 if mobj is not None:
1013 return self.url_result(mobj.group('url'), 'Livestream')
1015 def check_video(vurl):
1016 vpath = compat_urlparse.urlparse(vurl).path
1017 vext = determine_ext(vpath)
1018 return '.' in vpath and vext not in ('swf', 'png', 'jpg', 'srt', 'sbv', 'sub', 'vtt', 'ttml')
1020 def filter_video(urls):
1021 return list(filter(check_video, urls))
1023 # Start with something easy: JW Player in SWFObject
1024 found = filter_video(re.findall(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage))
1026 # Look for gorilla-vid style embedding
1027 found = filter_video(re.findall(r'''(?sx)
1031 jwplayer\s*\(\s*["'][^'"]+["']\s*\)\s*\.setup
1033 .*?file\s*:\s*["\'](.*?)["\']''', webpage))
1035 # Broaden the search a little bit
1036 found = filter_video(re.findall(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage))
1038 # Broaden the findall a little bit: JWPlayer JS loader
1039 found = filter_video(re.findall(
1040 r'[^A-Za-z0-9]?file["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage))
1043 found = filter_video(re.findall(r'''(?xs)
1044 flowplayer\("[^"]+",\s*
1046 \s*{[^}]+? ["']?clip["']?\s*:\s*\{\s*
1047 ["']?url["']?\s*:\s*["']([^"']+)["']
1050 # Try to find twitter cards info
1051 found = filter_video(re.findall(
1052 r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage))
1054 # We look for Open Graph info:
1055 # We have to match any number spaces between elements, some sites try to align them (eg.: statigr.am)
1056 m_video_type = re.findall(r'<meta.*?property="og:video:type".*?content="video/(.*?)"', webpage)
1057 # We only look in og:video if the MIME type is a video, don't try if it's a Flash player:
1058 if m_video_type is not None:
1059 found = filter_video(re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage))
1062 found = re.findall(r'(?s)<video[^<]*(?:>.*?<source[^>]*)?\s+src=["\'](.*?)["\']', webpage)
1065 r'(?i)<meta\s+(?=(?:[a-z-]+="[^"]+"\s+)*http-equiv="refresh")'
1066 r'(?:[a-z-]+="[^"]+"\s+)*?content="[0-9]{,2};url=\'?([^\'"]+)',
1069 new_url = found.group(1)
1070 self.report_following_redirect(new_url)
1076 raise UnsupportedError(url)
1079 for video_url in found:
1080 video_url = compat_urlparse.urljoin(url, video_url)
1081 video_id = compat_urllib_parse.unquote(os.path.basename(video_url))
1083 # Sometimes, jwplayer extraction will result in a YouTube URL
1084 if YoutubeIE.suitable(video_url):
1085 entries.append(self.url_result(video_url, 'Youtube'))
1088 # here's a fun little line of code for you:
1089 video_id = os.path.splitext(video_id)[0]
1094 'uploader': video_uploader,
1095 'title': video_title,
1096 'age_limit': age_limit,
1099 if len(entries) == 1:
1102 for num, e in enumerate(entries, start=1):
1103 e['title'] = '%s (%d)' % (e['title'], num)
1105 '_type': 'playlist',