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 def report_download_webpage(self
, video_id
):
373 """Report webpage download."""
374 if not self
._downloader
.params
.get('test', False):
375 self
._downloader
.report_warning('Falling back on generic information extractor.')
376 super(GenericIE
, self
).report_download_webpage(video_id
)
378 def report_following_redirect(self
, new_url
):
379 """Report information extraction."""
380 self
._downloader
.to_screen('[redirect] Following redirect to %s' % new_url
)
382 def _extract_rss(self
, url
, video_id
, doc
):
383 playlist_title
= doc
.find('./channel/title').text
384 playlist_desc_el
= doc
.find('./channel/description')
385 playlist_desc
= None if playlist_desc_el
is None else playlist_desc_el
.text
389 'url': e
.find('link').text
,
390 'title': e
.find('title').text
,
391 } for e
in doc
.findall('./channel/item')]
396 'title': playlist_title
,
397 'description': playlist_desc
,
401 def _extract_camtasia(self
, url
, video_id
, webpage
):
402 """ Returns None if no camtasia video can be found. """
404 camtasia_cfg
= self
._search
_regex
(
405 r
'fo\.addVariable\(\s*"csConfigFile",\s*"([^"]+)"\s*\);',
406 webpage
, 'camtasia configuration file', default
=None)
407 if camtasia_cfg
is None:
410 title
= self
._html
_search
_meta
('DC.title', webpage
, fatal
=True)
412 camtasia_url
= compat_urlparse
.urljoin(url
, camtasia_cfg
)
413 camtasia_cfg
= self
._download
_xml
(
414 camtasia_url
, video_id
,
415 note
='Downloading camtasia configuration',
416 errnote
='Failed to download camtasia configuration')
417 fileset_node
= camtasia_cfg
.find('./playlist/array/fileset')
420 for n
in fileset_node
.getchildren():
421 url_n
= n
.find('./uri')
426 'id': os
.path
.splitext(url_n
.text
.rpartition('/')[2])[0],
427 'title': '%s - %s' % (title
, n
.tag
),
428 'url': compat_urlparse
.urljoin(url
, url_n
.text
),
429 'duration': float_or_none(n
.find('./duration').text
),
438 def _real_extract(self
, url
):
439 if url
.startswith('//'):
442 'url': self
.http_scheme() + url
,
445 parsed_url
= compat_urlparse
.urlparse(url
)
446 if not parsed_url
.scheme
:
447 default_search
= self
._downloader
.params
.get('default_search')
448 if default_search
is None:
449 default_search
= 'fixup_error'
451 if default_search
in ('auto', 'auto_warning', 'fixup_error'):
453 self
._downloader
.report_warning('The url doesn\'t specify the protocol, trying with http')
454 return self
.url_result('http://' + url
)
455 elif default_search
!= 'fixup_error':
456 if default_search
== 'auto_warning':
457 if re
.match(r
'^(?:url|URL)$', url
):
458 raise ExtractorError(
459 'Invalid URL: %r . Call youtube-dl like this: youtube-dl -v "https://www.youtube.com/watch?v=BaW_jenozKc" ' % url
,
462 self
._downloader
.report_warning(
463 'Falling back to youtube search for %s . Set --default-search "auto" to suppress this warning.' % url
)
464 return self
.url_result('ytsearch:' + url
)
466 if default_search
in ('error', 'fixup_error'):
467 raise ExtractorError(
468 ('%r is not a valid URL. '
469 'Set --default-search "ytsearch" (or run youtube-dl "ytsearch:%s" ) to search YouTube'
470 ) % (url
, url
), expected
=True)
472 assert ':' in default_search
473 return self
.url_result(default_search
+ url
)
475 url
, smuggled_data
= unsmuggle_url(url
)
477 if smuggled_data
and 'force_videoid' in smuggled_data
:
478 force_videoid
= smuggled_data
['force_videoid']
479 video_id
= force_videoid
481 video_id
= os
.path
.splitext(url
.rstrip('/').split('/')[-1])[0]
483 self
.to_screen('%s: Requesting header' % video_id
)
485 head_req
= HEADRequest(url
)
486 response
= self
._request
_webpage
(
488 note
=False, errnote
='Could not send HEAD request to %s' % url
,
491 if response
is not False:
493 new_url
= response
.geturl()
495 self
.report_following_redirect(new_url
)
497 new_url
= smuggle_url(
498 new_url
, {'force_videoid': force_videoid}
)
499 return self
.url_result(new_url
)
501 # Check for direct link to a video
502 content_type
= response
.headers
.get('Content-Type', '')
503 m
= re
.match(r
'^(?P<type>audio|video|application(?=/ogg$))/(?P<format_id>.+)$', content_type
)
505 upload_date
= response
.headers
.get('Last-Modified')
507 upload_date
= unified_strdate(upload_date
)
510 'title': os
.path
.splitext(url_basename(url
))[0],
512 'format_id': m
.group('format_id'),
514 'vcodec': 'none' if m
.group('type') == 'audio' else None
516 'upload_date': upload_date
,
520 webpage
= self
._download
_webpage
(url
, video_id
)
522 # since this is the last-resort InfoExtractor, if
523 # this error is thrown, it'll be thrown here
524 raise ExtractorError('Failed to download URL: %s' % url
)
526 self
.report_extraction(video_id
)
530 doc
= parse_xml(webpage
)
532 return self
._extract
_rss
(url
, video_id
, doc
)
533 except compat_xml_parse_error
:
536 # Is it a Camtasia project?
537 camtasia_res
= self
._extract
_camtasia
(url
, video_id
, webpage
)
538 if camtasia_res
is not None:
541 # Sometimes embedded video player is hidden behind percent encoding
542 # (e.g. https://github.com/rg3/youtube-dl/issues/2448)
543 # Unescaping the whole page allows to handle those cases in a generic way
544 webpage
= compat_urllib_parse
.unquote(webpage
)
546 # it's tempting to parse this further, but you would
547 # have to take into account all the variations like
548 # Video Title - Site Name
549 # Site Name | Video Title
550 # Video Title - Tagline | Site Name
551 # and so on and so forth; it's just not practical
552 video_title
= self
._html
_search
_regex
(
553 r
'(?s)<title>(.*?)</title>', webpage
, 'video title',
556 # Try to detect age limit automatically
557 age_limit
= self
._rta
_search
(webpage
)
558 # And then there are the jokers who advertise that they use RTA,
559 # but actually don't.
560 AGE_LIMIT_MARKERS
= [
561 r
'Proudly Labeled <a href="http://www.rtalabel.org/" title="Restricted to Adults">RTA</a>',
563 if any(re
.search(marker
, webpage
) for marker
in AGE_LIMIT_MARKERS
):
566 # video uploader is domain name
567 video_uploader
= self
._search
_regex
(
568 r
'^(?:https?://)?([^/]*)/.*', url
, 'video uploader')
571 def _playlist_from_matches(matches
, getter
, ie
=None):
572 urlrs
= orderedSet(self
.url_result(getter(m
), ie
) for m
in matches
)
573 return self
.playlist_result(
574 urlrs
, playlist_id
=video_id
, playlist_title
=video_title
)
576 # Look for BrightCove:
577 bc_urls
= BrightcoveIE
._extract
_brightcove
_urls
(webpage
)
579 self
.to_screen('Brightcove video detected.')
582 'url': smuggle_url(bc_url
, {'Referer': url}
),
583 'ie_key': 'Brightcove'
584 } for bc_url
in bc_urls
]
588 'title': video_title
,
593 # Look for embedded (iframe) Vimeo player
595 r
'<iframe[^>]+?src=(["\'])(?P
<url
>(?
:https?
:)?
//player\
.vimeo\
.com
/video
/.+?
)\
1', webpage)
597 player_url = unescapeHTML(mobj.group('url
'))
598 surl = smuggle_url(player_url, {'Referer': url})
599 return self.url_result(surl, 'Vimeo
')
601 # Look for embedded (swf embed) Vimeo player
603 r'<embed
[^
>]+?src
="(https?://(?:www\.)?vimeo\.com/moogaloop\.swf.+?)"', webpage)
605 return self.url_result(mobj.group(1), 'Vimeo
')
607 # Look for embedded YouTube player
608 matches = re.findall(r'''(?x)
616 (?P<url>(?:https?:)?//(?:www\.)?youtube\.com/
620 return _playlist_from_matches(
621 matches, lambda m: unescapeHTML(m[1]), ie='Youtube
')
623 # Look for embedded Dailymotion player
624 matches = re.findall(
625 r'<iframe
[^
>]+?src
=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/embed/video/.+?)\1', webpage)
627 return _playlist_from_matches(
628 matches, lambda m: unescapeHTML(m[1]))
630 # Look for embedded Wistia player
632 r'<iframe[^>]+?src=(["\'])(?P
<url
>(?
:https?
:)?
//(?
:fast\
.)?wistia\
.net
/embed
/iframe
/.+?
)\
1', webpage)
635 '_type
': 'url_transparent
',
636 'url
': unescapeHTML(match.group('url
')),
638 'uploader
': video_uploader,
639 'title
': video_title,
643 # Look for embedded blip.tv player
644 mobj = re.search(r'<meta\s
[^
>]*https?
://api\
.blip\
.tv
/\w
+/redirect
/\w
+/(\d
+)', webpage)
646 return self.url_result('http
://blip
.tv
/a
/a
-'+mobj.group(1), 'BlipTV
')
647 mobj = re.search(r'<(?
:iframe|embed|
object)\s
[^
>]*(https?
://(?
:\w
+\
.)?blip\
.tv
/(?
:play
/|api\
.swf
#)[a-zA-Z0-9_]+)', webpage)
649 return self
.url_result(mobj
.group(1), 'BlipTV')
651 # Look for embedded condenast player
652 matches
= re
.findall(
653 r
'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="(https?://player\.cnevids\.com/embed/[^"]+")',
660 'ie_key': 'CondeNast',
662 } for ma
in matches
],
663 'title': video_title
,
667 # Look for Bandcamp pages with custom domain
668 mobj
= re
.search(r
'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage
)
670 burl
= unescapeHTML(mobj
.group(1))
671 # Don't set the extractor because it can be a track url or an album
672 return self
.url_result(burl
)
674 # Look for embedded Vevo player
676 r
'<iframe[^>]+?src=(["\'])(?P
<url
>(?
:https?
:)?
//(?
:cache\
.)?vevo\
.com
/.+?
)\
1', webpage)
678 return self.url_result(mobj.group('url
'))
680 # Look for Ooyala videos
681 mobj = (re.search(r'player
.ooyala
.com
/[^
"?]+\?[^"]*?
(?
:embedCode|ec
)=(?P
<ec
>[^
"&]+)', webpage) or
682 re.search(r'OO.Player.create\([\'"].*?
[\'"],\s*[\'"](?P
<ec
>.{32}
)[\'"]', webpage))
684 return OoyalaIE._build_url_result(mobj.group('ec'))
686 # Look for Aparat videos
687 mobj = re.search(r'<iframe .*?src="(http
://www\
.aparat\
.com
/video
/[^
"]+)"', webpage)
689 return self.url_result(mobj.group(1), 'Aparat
')
691 # Look for MPORA videos
692 mobj = re.search(r'<iframe
.*?src
="(http://mpora\.(?:com|de)/videos/[^"]+)"', webpage)
694 return self.url_result(mobj.group(1), 'Mpora')
696 # Look for embedded NovaMov-based player
698 r'''(?x)<(?:pagespeed_)?iframe[^>]+?src=(["\'])
699 (?P
<url
>http
://(?
:(?
:embed|www
)\
.)?
701 nowvideo\
.(?
:ch|sx|eu|at|ag|co
)|
702 videoweed\
.(?
:es|com
)|
703 movshare\
.(?
:net|sx|ag
)|
704 divxstage\
.(?
:eu|net|ch|co|at|ag
))
705 /embed\
.php
.+?
)\
1''', webpage)
707 return self.url_result(mobj.group('url'))
709 # Look for embedded Facebook player
711 r'<iframe[^>]+?src=(["\'])(?P<url>https://www\.facebook\.com/video/embed.+?)\1', webpage)
713 return self.url_result(mobj.group('url'), 'Facebook')
715 # Look for embedded VK player
716 mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://vk\.com/video_ext\.php.+?)\1', webpage)
718 return self.url_result(mobj.group('url'), 'VK')
720 # Look for embedded ivi player
721 mobj = re.search(r'<embed[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?ivi\.ru/video/player.+?)\1', webpage)
723 return self.url_result(mobj.group('url'), 'Ivi')
725 # Look for embedded Huffington Post player
727 r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed\.live\.huffingtonpost\.com/.+?)\1', webpage)
729 return self.url_result(mobj.group('url'), 'HuffPost')
732 mobj = re.search(r'class=["\']embedly-card["\'][^>]href=["\'](?P<url>[^"\']+)', webpage)
734 return self.url_result(mobj.group('url'))
735 mobj = re.search(r'class=["\']embedly-embed["\'][^>]src=["\'][^"\']*url=(?P<url>[^&]+)', webpage)
737 return self.url_result(compat_urllib_parse.unquote(mobj.group('url')))
739 # Look for funnyordie embed
740 matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage)
742 return _playlist_from_matches(
743 matches, getter=unescapeHTML, ie='FunnyOrDie')
745 # Look for embedded RUTV player
746 rutv_url = RUTVIE._extract_url(webpage)
748 return self.url_result(rutv_url, 'RUTV')
750 # Look for embedded TED player
752 r'<iframe[^>]+?src=(["\'])(?P<url>http://embed\.ted\.com/.+?)\1', webpage)
754 return self.url_result(mobj.group('url'), 'TED')
756 # Look for embedded Ustream videos
758 r'<iframe[^>]+?src=(["\'])(?P<url>http://www\.ustream\.tv/embed/.+?)\1', webpage)
760 return self.url_result(mobj.group('url'), 'Ustream')
762 # Look for embedded arte.tv player
764 r'<script [^>]*?src="(?P<url>http://www\.arte\.tv/playerv2/embed[^"]+)"',
767 return self.url_result(mobj.group('url'), 'ArteTVEmbed')
769 # Look for embedded smotri.com player
770 smotri_url = SmotriIE._extract_url(webpage)
772 return self.url_result(smotri_url, 'Smotri')
774 # Look for embeded soundcloud player
776 r'<iframe src="(?P<url>https?://(?:w\.)?soundcloud\.com/player[^"]+)"',
779 url = unescapeHTML(mobj.group('url'))
780 return self.url_result(url)
782 # Look for embedded vulture.com player
784 r'<iframe src="(?P<url>https?://video\.vulture\.com/[^"]+)"',
787 url = unescapeHTML(mobj.group('url'))
788 return self.url_result(url, ie='Vulture')
790 # Look for embedded mtvservices player
792 r'<iframe src="(?P<url>https?://media\.mtvnservices\.com/embed/[^"]+)"',
795 url = unescapeHTML(mobj.group('url'))
796 return self.url_result(url, ie='MTVServicesEmbedded')
798 # Look for embedded yahoo player
800 r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:screen|movies)\.yahoo\.com/.+?\.html\?format=embed)\1',
803 return self.url_result(mobj.group('url'), 'Yahoo')
805 # Look for embedded sbs.com.au player
807 r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:www\.)sbs\.com\.au/ondemand/video/single/.+?)\1',
810 return self.url_result(mobj.group('url'), 'SBS')
812 # Start with something easy: JW Player in SWFObject
813 found = re.findall(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage)
815 # Look for gorilla-vid style embedding
816 found = re.findall(r'''(?sx
)
820 jwplayer\s
*\
(\s
*["'][^'"]+["']\s*\)\s*\.setup
822 .*?file\s*:\s*["\'](.*?
)["\']''', webpage)
824 # Broaden the search a little bit
825 found = re.findall(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage)
827 # Broaden the findall a little bit: JWPlayer JS loader
828 found = re.findall(r'[^A
-Za
-z0
-9]?
file["\']?:\s*["\'](http(?
![^
\'"]+\.[0-9]+[\'"])[^
\'"]+)["\']', webpage)
831 found = re.findall(r'''(?xs)
832 flowplayer\("[^"]+",\s*
834 \s*{[^}]+? ["']?clip
["']?\s*:\s*\{\s*
835 ["']?url["']?\s
*:\s
*["']([^"']+)["']
838 # Try to find twitter cards info
839 found = re.findall(r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage)
841 # We look for Open Graph info:
842 # We have to match any number spaces between elements, some sites try to align them (eg.: statigr.am)
843 m_video_type = re.findall(r'<meta.*?property="og:video:type".*?content="video/(.*?)"', webpage)
844 # We only look in og:video if the MIME type is a video, don't try if it's a Flash player:
845 if m_video_type is not None:
846 def check_video(vurl):
847 vpath = compat_urlparse.urlparse(vurl).path
848 vext = determine_ext(vpath)
849 return '.' in vpath and vext not in ('swf', 'png', 'jpg')
852 re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage)))
855 found = re.findall(r'(?s)<video[^<]*(?:>.*?<source[^>]+)? src="([^"]+)"', webpage)
858 r'(?i)<meta\s+(?=(?:[a-z-]+="[^"]+"\s+)*http-equiv="refresh")'
859 r'(?:[a-z-]+="[^"]+"\s+)*?content="[0-9]{,2};url=\'([^\']+)\'"',
862 new_url = found.group(1)
863 self.report_following_redirect(new_url)
869 raise ExtractorError('Unsupported URL: %s' % url)
872 for video_url in found:
873 video_url = compat_urlparse.urljoin(url, video_url)
874 video_id = compat_urllib_parse.unquote(os.path.basename(video_url))
876 # Sometimes, jwplayer extraction will result in a YouTube URL
877 if YoutubeIE.suitable(video_url):
878 entries.append(self.url_result(video_url, 'Youtube'))
881 # here's a fun little line of code for you:
882 video_id = os.path.splitext(video_id)[0]
887 'uploader': video_uploader,
888 'title': video_title,
889 'age_limit': age_limit,
892 if len(entries) == 1:
895 for num, e in enumerate(entries, start=1):
896 e['title'] = '%s (%d)' % (e['title'], num)