]>
jfr.im git - yt-dlp.git/blob - youtube_dl/extractor/generic.py
3 from __future__
import unicode_literals
8 from . common
import InfoExtractor
9 from . youtube
import YoutubeIE
13 compat_urllib_request
,
15 compat_xml_parse_error
,
26 from . brightcove
import BrightcoveIE
27 from . ooyala
import OoyalaIE
28 from . rutv
import RUTVIE
29 from . smotri
import SmotriIE
32 class GenericIE ( InfoExtractor
):
33 IE_DESC
= 'Generic downloader that works on some sites'
38 'url' : 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html' ,
39 'md5' : '85b90ccc9d73b4acd9138d3af4c27f89' ,
41 'id' : '13601338388002' ,
43 'uploader' : 'www.hodiho.fr' ,
44 'title' : 'R\u00e9gis plante sa Jeep' ,
47 # bandcamp page with custom domain
49 'add_ie' : [ 'Bandcamp' ],
50 'url' : 'http://bronyrock.com/track/the-pony-mash' ,
54 'title' : 'The Pony Mash' ,
55 'uploader' : 'M_Pallante' ,
57 'skip' : 'There is a limit of 200 free downloads / month for the test song' ,
59 # embedded brightcove video
60 # it also tests brightcove videos that need to set the 'Referer' in the
63 'add_ie' : [ 'Brightcove' ],
64 'url' : 'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/' ,
66 'id' : '2765128793001' ,
68 'title' : 'Le cours de bourse : l’analyse technique' ,
69 'description' : 'md5:7e9ad046e968cb2d1114004aba466fd9' ,
70 'uploader' : 'BFM BUSINESS' ,
73 'skip_download' : True ,
77 # https://github.com/rg3/youtube-dl/issues/2253
78 'url' : 'http://bcove.me/i6nfkrc3' ,
79 'md5' : '0ba9446db037002366bab3b3eb30c88c' ,
81 'id' : '3101154703001' ,
83 'title' : 'Still no power' ,
84 'uploader' : 'thestar.com' ,
85 '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.' ,
87 'add_ie' : [ 'Brightcove' ],
90 'url' : 'http://www.championat.com/video/football/v/87/87499.html' ,
91 'md5' : 'fb973ecf6e4a78a67453647444222983' ,
93 'id' : '3414141473001' ,
95 'title' : 'Видео. Удаление Дзагоева (ЦСКА)' ,
96 'description' : 'Онлайн-трансляция матча ЦСКА - "Волга"' ,
97 'uploader' : 'Championat' ,
100 # Direct link to a video
102 'url' : 'http://media.w3.org/2010/05/sintel/trailer.mp4' ,
103 'md5' : '67d406c2bcb6af27fa886f31aa934bbe' ,
108 'upload_date' : '20100513' ,
113 'url' : 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219' ,
114 'md5' : '5644c6ca5d5782c1d0d350dad9bd840c' ,
116 'id' : 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ' ,
118 'title' : '2cc213299525360.mov' , # that's what we get
123 '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' ,
127 'upload_date' : '20130224' ,
128 'uploader_id' : 'TheVerge' ,
129 '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.' ,
130 'uploader' : 'The Verge' ,
131 'title' : 'First Firefox OS phones side-by-side' ,
134 'skip_download' : False ,
139 'url' : 'http://www.tested.com/science/weird/460206-tested-grinding-coffee-2000-frames-second/' ,
143 'title' : 'Tested: Grinding Coffee at 2000 Frames Per Second' ,
144 'upload_date' : '20140225' ,
145 'description' : 'md5:06a40fbf30b220468f1e0957c0f558ff' ,
146 'uploader' : 'Tested' ,
147 'uploader_id' : 'testedcom' ,
149 # No need to test YoutubeIE here
151 'skip_download' : True ,
156 'url' : 'http://www.theguardian.com/world/2014/mar/11/obama-zach-galifianakis-between-two-ferns' ,
157 'md5' : '7cf780be104d40fea7bae52eed4a470e' ,
161 'title' : 'Between Two Ferns with Zach Galifianakis: President Barack Obama' ,
162 'description' : 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.' ,
167 'url' : 'http://www.rg.ru/2014/03/15/reg-dfo/anklav-anons.html' ,
171 'title' : 'Охотское море стало целиком российским' ,
172 'description' : 'md5:5ed62483b14663e2a95ebbe115eb8f43' ,
176 'skip_download' : True ,
181 'url' : 'http://en.support.wordpress.com/videos/ted-talks/' ,
182 'md5' : 'deeeabcc1085eb2ba205474e7235a3d5' ,
186 'title' : 'My web playroom' ,
187 'uploader' : 'Ze Frank' ,
188 'description' : 'md5:ddb2a40ecd6b6a147e400e535874947b' ,
191 # Embeded Ustream video
193 'url' : 'http://www.american.edu/spa/pti/nsa-privacy-janus-2014.cfm' ,
194 'md5' : '27b99cdb639c9b12a79bca876a073417' ,
198 'uploader' : 'AU SPA: The NSA and Privacy' ,
199 'title' : 'NSA and Privacy Forum Debate featuring General Hayden and Barton Gellman'
202 # nowvideo embed hidden behind percent encoding
204 'url' : 'http://www.waoanime.tv/the-super-dimension-fortress-macross-episode-1/' ,
205 'md5' : '2baf4ddd70f697d94b1c18cf796d5107' ,
207 'id' : '06e53103ca9aa' ,
209 'title' : 'Macross Episode 001 Watch Macross Episode 001 onl' ,
210 'description' : 'No description' ,
215 'url' : 'http://www.tv-replay.fr/redirection/20-03-14/x-enius-arte-10753389.html' ,
216 'md5' : '7653032cbb25bf6c80d80f217055fa43' ,
218 'id' : '048195-004_PLUS7-F' ,
221 'description' : 'md5:d5fdf32ef6613cdbfd516ae658abf168' ,
222 'upload_date' : '20140320' ,
225 'skip_download' : 'Requires rtmpdump'
230 'url' : 'http://rbctv.rbc.ru/archive/news/562949990879132.shtml' ,
231 'md5' : 'ec40048448e9284c9a1de77bb188108b' ,
233 'id' : 'v27008541fad' ,
235 'title' : 'Крым и Севастополь вошли в состав России' ,
236 'description' : 'md5:fae01b61f68984c7bd2fa741e11c3175' ,
238 'upload_date' : '20140318' ,
239 'uploader' : 'rbctv_2012_4' ,
240 'uploader_id' : 'rbctv_2012_4' ,
245 'url' : 'http://www.wired.com/2014/04/honda-asimo/' ,
246 'md5' : 'ba0dfe966fa007657bd1443ee672db0f' ,
248 'id' : '53501be369702d3275860000' ,
250 'title' : 'Honda’s New Asimo Robot Is More Human Than Ever' ,
255 'url' : 'http://www.spi0n.com/zap-spi0n-com-n216/' ,
256 'md5' : '441aeeb82eb72c422c7f14ec533999cd' ,
258 'id' : 'k2mm4bCdJ6CQ2i7c8o2' ,
260 'title' : 'Le Zap de Spi0n n°216 - Zapping du Web' ,
263 'add_ie' : [ 'Dailymotion' ],
267 'url' : 'http://www.badzine.de/ansicht/datum/2014/06/09/so-funktioniert-die-neue-englische-badminton-liga.html' ,
271 'title' : 'The NBL Auction 2014' ,
272 'uploader' : 'BADMINTON England' ,
273 'uploader_id' : 'BADMINTONEvents' ,
274 'upload_date' : '20140603' ,
275 'description' : 'md5:9ef128a69f1e262a700ed83edb163a73' ,
277 'add_ie' : [ 'Youtube' ],
279 'skip_download' : True ,
284 'url' : 'http://www.gametrailers.com/news-post/76093/north-america-europe-is-getting-that-mario-kart-8-mercedes-dlc-too' ,
285 'md5' : '35727f82f58c76d996fc188f9755b0d5' ,
287 'id' : '0306a69b-8adf-4fb5-aace-75f8e8cbfca9' ,
290 'description' : 'Mario \' s life in the fast lane has never looked so good.' ,
293 # YouTube embed via <data-embed-url="">
295 'url' : 'https://play.google.com/store/apps/details?id=com.gameloft.android.ANMP.GloftA8HM' ,
299 'title' : 'Asphalt 8: Airborne - Launch Trailer' ,
300 'uploader' : 'Gameloft' ,
301 'uploader_id' : 'gameloft' ,
302 'upload_date' : '20130821' ,
303 'description' : 'md5:87bd95f13d8be3e7da87a5f2c443106a' ,
306 'skip_download' : True ,
311 def report_download_webpage ( self
, video_id
):
312 """Report webpage download."""
313 if not self
._ downloader
. params
. get ( 'test' , False ):
314 self
._ downloader
. report_warning ( 'Falling back on generic information extractor.' )
315 super ( GenericIE
, self
). report_download_webpage ( video_id
)
317 def report_following_redirect ( self
, new_url
):
318 """Report information extraction."""
319 self
._ downloader
. to_screen ( '[redirect] Following redirect to %s ' % new_url
)
321 def _send_head ( self
, url
):
322 """Check if it is a redirect, like url shorteners, in case return the new url."""
324 class HEADRedirectHandler ( compat_urllib_request
. HTTPRedirectHandler
):
326 Subclass the HTTPRedirectHandler to make it use our
327 HEADRequest also on the redirected URL
329 def redirect_request ( self
, req
, fp
, code
, msg
, headers
, newurl
):
330 if code
in ( 301 , 302 , 303 , 307 ):
331 newurl
= newurl
. replace ( ' ' , '%20' )
332 newheaders
= dict (( k
, v
) for k
, v
in req
. headers
. items ()
333 if k
. lower () not in ( "content-length" , "content-type" ))
335 # This function was deprecated in python 3.3 and removed in 3.4
336 origin_req_host
= req
. get_origin_req_host ()
337 except AttributeError :
338 origin_req_host
= req
. origin_req_host
339 return HEADRequest ( newurl
,
341 origin_req_host
= origin_req_host
,
344 raise compat_urllib_error
. HTTPError ( req
. get_full_url (), code
, msg
, headers
, fp
)
346 class HTTPMethodFallback ( compat_urllib_request
. BaseHandler
):
348 Fallback to GET if HEAD is not allowed (405 HTTP error)
350 def http_error_405 ( self
, req
, fp
, code
, msg
, headers
):
354 newheaders
= dict (( k
, v
) for k
, v
in req
. headers
. items ()
355 if k
. lower () not in ( "content-length" , "content-type" ))
356 return self
. parent
. open ( compat_urllib_request
. Request ( req
. get_full_url (),
358 origin_req_host
= req
. get_origin_req_host (),
362 opener
= compat_urllib_request
. OpenerDirector ()
363 for handler
in [ compat_urllib_request
. HTTPHandler
, compat_urllib_request
. HTTPDefaultErrorHandler
,
364 HTTPMethodFallback
, HEADRedirectHandler
,
365 compat_urllib_request
. HTTPErrorProcessor
, compat_urllib_request
. HTTPSHandler
]:
366 opener
. add_handler ( handler ())
368 response
= opener
. open ( HEADRequest ( url
))
370 raise ExtractorError ( 'Invalid URL protocol' )
373 def _extract_rss ( self
, url
, video_id
, doc
):
374 playlist_title
= doc
. find ( './channel/title' ). text
375 playlist_desc_el
= doc
. find ( './channel/description' )
376 playlist_desc
= None if playlist_desc_el
is None else playlist_desc_el
. text
380 'url' : e
. find ( 'link' ). text
,
381 'title' : e
. find ( 'title' ). text
,
382 } for e
in doc
. findall ( './channel/item' )]
387 'title' : playlist_title
,
388 'description' : playlist_desc
,
392 def _real_extract ( self
, url
):
393 if url
. startswith ( '//' ):
396 'url' : self
. http_scheme () + url
,
399 parsed_url
= compat_urlparse
. urlparse ( url
)
400 if not parsed_url
. scheme
:
401 default_search
= self
._ downloader
. params
. get ( 'default_search' )
402 if default_search
is None :
403 default_search
= 'fixup_error'
405 if default_search
in ( 'auto' , 'auto_warning' , 'fixup_error' ):
407 self
._ downloader
. report_warning ( 'The url doesn \' t specify the protocol, trying with http' )
408 return self
. url_result ( 'http://' + url
)
409 elif default_search
!= 'fixup_error' :
410 if default_search
== 'auto_warning' :
411 if re
. match ( r
'^(?:url|URL)$' , url
):
412 raise ExtractorError (
413 'Invalid URL: %r . Call youtube-dl like this: youtube-dl -v "https://www.youtube.com/watch?v=BaW_jenozKc" ' % url
,
416 self
._ downloader
. report_warning (
417 'Falling back to youtube search for %s . Set --default-search "auto" to suppress this warning.' % url
)
418 return self
. url_result ( 'ytsearch:' + url
)
420 if default_search
in ( 'error' , 'fixup_error' ):
421 raise ExtractorError (
422 ( ' %r is not a valid URL. '
423 'Set --default-search "ytsearch" (or run youtube-dl "ytsearch: %s " ) to search YouTube'
424 ) % ( url
, url
), expected
= True )
426 assert ':' in default_search
427 return self
. url_result ( default_search
+ url
)
428 video_id
= os
. path
. splitext ( url
. rstrip ( '/' ). split ( '/' )[- 1 ])[ 0 ]
430 self
. to_screen ( ' %s : Requesting header' % video_id
)
433 response
= self
._ send
_ head
( url
)
436 new_url
= response
. geturl ()
438 self
. report_following_redirect ( new_url
)
439 return self
. url_result ( new_url
)
441 # Check for direct link to a video
442 content_type
= response
. headers
. get ( 'Content-Type' , '' )
443 m
= re
. match ( r
'^(?P<type>audio|video|application(?=/ogg$))/(?P<format_id>.+)$' , content_type
)
445 upload_date
= response
. headers
. get ( 'Last-Modified' )
447 upload_date
= unified_strdate ( upload_date
)
450 'title' : os
. path
. splitext ( url_basename ( url
))[ 0 ],
452 'format_id' : m
. group ( 'format_id' ),
454 'vcodec' : 'none' if m
. group ( 'type' ) == 'audio' else None
456 'upload_date' : upload_date
,
459 except compat_urllib_error
. HTTPError
:
460 # This may be a stupid server that doesn't like HEAD, our UA, or so
464 webpage
= self
._ download
_ webpage
( url
, video_id
)
466 # since this is the last-resort InfoExtractor, if
467 # this error is thrown, it'll be thrown here
468 raise ExtractorError ( 'Failed to download URL: %s ' % url
)
470 self
. report_extraction ( video_id
)
474 doc
= parse_xml ( webpage
)
476 return self
._ extract
_ rss
( url
, video_id
, doc
)
477 except compat_xml_parse_error
:
480 # Sometimes embedded video player is hidden behind percent encoding
481 # (e.g. https://github.com/rg3/youtube-dl/issues/2448)
482 # Unescaping the whole page allows to handle those cases in a generic way
483 webpage
= compat_urllib_parse
. unquote ( webpage
)
485 # it's tempting to parse this further, but you would
486 # have to take into account all the variations like
487 # Video Title - Site Name
488 # Site Name | Video Title
489 # Video Title - Tagline | Site Name
490 # and so on and so forth; it's just not practical
491 video_title
= self
._ html
_ search
_ regex
(
492 r
'(?s)<title>(.*?)</title>' , webpage
, 'video title' ,
495 # video uploader is domain name
496 video_uploader
= self
._ search
_ regex
(
497 r
'^(?:https?://)?([^/]*)/.*' , url
, 'video uploader' )
500 def _playlist_from_matches ( matches
, getter
, ie
= None ):
501 urlrs
= orderedSet ( self
. url_result ( getter ( m
), ie
) for m
in matches
)
502 return self
. playlist_result (
503 urlrs
, playlist_id
= video_id
, playlist_title
= video_title
)
505 # Look for BrightCove:
506 bc_urls
= BrightcoveIE
._ extract
_ brightcove
_u rls
( webpage
)
508 self
. to_screen ( 'Brightcove video detected.' )
511 'url' : smuggle_url ( bc_url
, {'Referer': url}
),
512 'ie_key' : 'Brightcove'
513 } for bc_url
in bc_urls
]
517 'title' : video_title
,
522 # Look for embedded (iframe) Vimeo player
524 r
'<iframe[^>]+?src=(["\' ])( ?P
< url
>( ?
: https?
:) ?
// player\
. vimeo\
. com
/ video
/.+ ?
) \
1 ', webpage)
526 player_url = unescapeHTML(mobj.group(' url
'))
527 surl = smuggle_url(player_url, {'Referer': url} )
528 return self.url_result(surl, ' Vimeo
')
530 # Look for embedded (swf embed) Vimeo player
532 r' < embed
[ ^
>]+ ?src
= "(https?://(?:www\.)?vimeo\.com/moogaloop\.swf.+?)" ', webpage)
534 return self.url_result(mobj.group(1), ' Vimeo
')
536 # Look for embedded YouTube player
537 matches = re.findall(r'''(?x)
545 (?P<url>(?:https?:)?//(?:www\.)?youtube\.com/
549 return _playlist_from_matches(
550 matches, lambda m: unescapeHTML(m[1]), ie=' Youtube
')
552 # Look for embedded Dailymotion player
553 matches = re.findall(
554 r' < iframe
[ ^
>]+ ?src
=([ " \' ])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/embed/video/.+?)\1', webpage)
556 return _playlist_from_matches(
557 matches, lambda m: unescapeHTML(m[1]))
559 # Look for embedded Wistia player
561 r'<iframe[^>]+?src=([" \' ])( ?P
< url
>( ?
: https?
:) ?
//( ?
: fast\
.) ?wistia\
. net
/ embed
/ iframe
/.+ ?
) \
1 ', webpage)
564 ' _type
': ' url_transparent
',
565 ' url
': unescapeHTML(match.group(' url
')),
567 ' uploader
': video_uploader,
568 ' title
': video_title,
572 # Look for embedded blip.tv player
573 mobj = re.search(r' < meta\s
[ ^
>]* https?
:// api\
. blip\
. tv
/ \w
+/ redirect
/ \w
+/( \d
+) ', webpage)
575 return self.url_result(' http
:// blip
. tv
/ a
/ a
- '+mobj.group(1), ' BlipTV
')
576 mobj = re.search(r' <( ?
: iframe|embed|
object ) \s
[ ^
>]*( https?
://( ?
: \w
+ \
.) ?blip\
. tv
/( ?
: play
/ |api\
. swf
#)[a-zA-Z0-9]+)', webpage)
578 return self
. url_result ( mobj
. group ( 1 ), 'BlipTV' )
580 # Look for embedded condenast player
581 matches
= re
. findall (
582 r
'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="(https?://player\.cnevids\.com/embed/[^"]+")' ,
589 'ie_key' : 'CondeNast' ,
591 } for ma
in matches
],
592 'title' : video_title
,
596 # Look for Bandcamp pages with custom domain
597 mobj
= re
. search ( r
'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"' , webpage
)
599 burl
= unescapeHTML ( mobj
. group ( 1 ))
600 # Don't set the extractor because it can be a track url or an album
601 return self
. url_result ( burl
)
603 # Look for embedded Vevo player
605 r
'<iframe[^>]+?src=(["\' ])( ?P
< url
>( ?
: https?
:) ?
//( ?
: cache\
.) ?vevo\
. com
/.+ ?
) \
1 ', webpage)
607 return self.url_result(mobj.group(' url
'))
609 # Look for Ooyala videos
610 mobj = (re.search(r' player
. ooyala
. com
/[ ^
"?]+ \? [^" ]* ?
( ?
: embedCode|ec
)=( ?P
< ec
>[ ^
"&]+)', webpage) or
611 re.search(r'OO.Player.create\([ \' " ].* ?
[ \' "],\s*[ \' " ]( ?P
< ec
>. {32}
)[ \' "]', webpage))
613 return OoyalaIE._build_url_result(mobj.group('ec'))
615 # Look for Aparat videos
616 mobj = re.search(r'<iframe .*?src=" ( http
:// www\
. aparat\
. com
/ video
/[ ^
"]+)" ', webpage)
618 return self.url_result(mobj.group(1), ' Aparat
')
620 # Look for MPORA videos
621 mobj = re.search(r' < iframe
.* ?src
= "(http://mpora\.(?:com|de)/videos/[^" ]+) "', webpage)
623 return self.url_result(mobj.group(1), 'Mpora')
625 # Look for embedded NovaMov-based player
627 r'''(?x)<(?:pagespeed_)?iframe[^>]+?src=([" \' ])
628 ( ?P
< url
> http
://( ?
:( ?
: embed|www
) \
.) ?
630 nowvideo\
.( ?
: ch|sx|eu|at|ag|co
) |
631 videoweed\
.( ?
: es|com
) |
632 movshare\
.( ?
: net|sx|ag
) |
633 divxstage\
.( ?
: eu|net|ch|co|at|ag
))
634 / embed\
. php
.+ ?
) \
1 ''', webpage)
636 return self.url_result(mobj.group('url'))
638 # Look for embedded Facebook player
640 r'<iframe[^>]+?src=([" \' ])(?P<url>https://www\.facebook\.com/video/embed.+?)\1', webpage)
642 return self.url_result(mobj.group('url'), 'Facebook')
644 # Look for embedded VK player
645 mobj = re.search(r'<iframe[^>]+?src=([" \' ])(?P<url>https?://vk\.com/video_ext\.php.+?)\1', webpage)
647 return self.url_result(mobj.group('url'), 'VK')
649 # Look for embedded ivi player
650 mobj = re.search(r'<embed[^>]+?src=([" \' ])(?P<url>https?://(?:www\.)?ivi\.ru/video/player.+?)\1', webpage)
652 return self.url_result(mobj.group('url'), 'Ivi')
654 # Look for embedded Huffington Post player
656 r'<iframe[^>]+?src=([" \' ])(?P<url>https?://embed\.live\.huffingtonpost\.com/.+?)\1', webpage)
658 return self.url_result(mobj.group('url'), 'HuffPost')
661 mobj = re.search(r'class=[" \' ]embedly-card[" \' ][^>]href=[" \' ](?P<url>[^" \' ]+)', webpage)
663 return self.url_result(mobj.group('url'))
664 mobj = re.search(r'class=[" \' ]embedly-embed[" \' ][^>]src=[" \' ][^" \' ]*url=(?P<url>[^&]+)', webpage)
666 return self.url_result(compat_urllib_parse.unquote(mobj.group('url')))
668 # Look for funnyordie embed
669 matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage)
671 return _playlist_from_matches(
672 matches, getter=unescapeHTML, ie='FunnyOrDie')
674 # Look for embedded RUTV player
675 rutv_url = RUTVIE._extract_url(webpage)
677 return self.url_result(rutv_url, 'RUTV')
679 # Look for embedded TED player
681 r'<iframe[^>]+?src=([" \' ])(?P<url>http://embed\.ted\.com/.+?)\1', webpage)
683 return self.url_result(mobj.group('url'), 'TED')
685 # Look for embedded Ustream videos
687 r'<iframe[^>]+?src=([" \' ])(?P<url>http://www\.ustream\.tv/embed/.+?)\1', webpage)
689 return self.url_result(mobj.group('url'), 'Ustream')
691 # Look for embedded arte.tv player
693 r'<script [^>]*?src="(?P<url>http://www\.arte\.tv/playerv2/embed[^"]+)"',
696 return self.url_result(mobj.group('url'), 'ArteTVEmbed')
698 # Look for embedded smotri.com player
699 smotri_url = SmotriIE._extract_url(webpage)
701 return self.url_result(smotri_url, 'Smotri')
703 # Look for embeded soundcloud player
705 r'<iframe src="(?P<url>https?://(?:w\.)?soundcloud\.com/player[^"]+)"',
708 url = unescapeHTML(mobj.group('url'))
709 return self.url_result(url)
711 # Look for embedded vulture.com player
713 r'<iframe src="(?P<url>https?://video\.vulture\.com/[^"]+)"',
716 url = unescapeHTML(mobj.group('url'))
717 return self.url_result(url, ie='Vulture')
719 # Look for embedded mtvservices player
721 r'<iframe src="(?P<url>https?://media\.mtvnservices\.com/embed/[^"]+)"',
724 url = unescapeHTML(mobj.group('url'))
725 return self.url_result(url, ie='MTVServicesEmbedded')
727 # Look for embedded yahoo player
729 r'<iframe[^>]+?src=([" \' ])(?P<url>https?://(?:screen|movies)\.yahoo\.com/.+?\.html \? format=embed)\1',
732 return self.url_result(mobj.group('url'), 'Yahoo')
734 # Look for embedded sbs.com.au player
736 r'<iframe[^>]+?src=([" \' ])(?P<url>https?://(?:www\.)sbs\.com\.au/ondemand/video/single/.+?)\1',
739 return self.url_result(mobj.group('url'), 'SBS')
741 # Start with something easy: JW Player in SWFObject
742 found = re.findall(r'flashvars: [ \' "](?:.*&)?file=(http[^ \' "&]*)', webpage)
744 # Look for gorilla-vid style embedding
745 found = re.findall(r''' ( ?sx
)
749 jwplayer\s
* \
( \s
*[ "'][^'" ]+[ "']\s*\)\s*\.setup
751 .*?file\s*:\s*[" \' ](.* ?
)[ " \' ]''', webpage)
753 # Broaden the search a little bit
754 found = re.findall(r'[^A-Za-z0-9]?(?:file|source)=(http[^ \' " &]*) ', webpage)
756 # Broaden the findall a little bit: JWPlayer JS loader
757 found = re.findall(r' [ ^A
- Za
- z0
- 9 ] ?
file [ " \' ]?:\s*[" \' ]( http ( ?
![ ^
\' "]+\.[0-9]+[ \' " ])[ ^
\' "]+)[" \' ] ', webpage)
759 # Try to find twitter cards info
760 found = re.findall(r' < meta ( ?
: property |name
)= "twitter:player:stream" ( ?
: content|value
)= "(.+?)" ', webpage)
762 # We look for Open Graph info:
763 # We have to match any number spaces between elements, some sites try to align them (eg.: statigr.am)
764 m_video_type = re.findall(r' < meta
.* ?
property = "og:video:type" .* ?content
= "video/(.*?)" ', webpage)
765 # We only look in og:video if the MIME type is a video, don' t
try if it
's a Flash player:
766 if m_video_type is not None:
767 found = re.findall(r' < meta
.* ?
property = "og:video" .* ?content
= "(.*?)" ', webpage)
770 found = re.findall(r' ( ?s
)< video
[ ^
<]*( ?
:>.* ?
< source
.* ?
) ? src
= "([^" ]+) "', webpage)
773 r'(?i)<meta\s+(?=(?:[a-z-]+=" [ ^
"]+" \s
+)* http
- equiv
= "refresh" ) '
774 r' ( ?
:[ a
- z
-]+= "[^" ]+ "\s+)*?content=" [ 0 - 9 ] {,2}
; url
= \' ([ ^
\' ]+) \' "',
777 new_url = found.group(1)
778 self.report_following_redirect(new_url)
784 raise ExtractorError('Unsupported URL: %s ' % url)
787 for video_url in found:
788 video_url = compat_urlparse.urljoin(url, video_url)
789 video_id = compat_urllib_parse.unquote(os.path.basename(video_url))
791 # Sometimes, jwplayer extraction will result in a YouTube URL
792 if YoutubeIE.suitable(video_url):
793 entries.append(self.url_result(video_url, 'Youtube'))
796 # here's a fun little line of code for you:
797 video_id = os.path.splitext(video_id)[0]
802 'uploader': video_uploader,
803 'title': video_title,
806 if len(entries) == 1:
809 for num, e in enumerate(entries, start=1):
810 e['title'] = ' %s ( %d )' % (e['title'], num)