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,
25 from .brightcove import BrightcoveIE
26 from .ooyala import OoyalaIE
27 from .rutv import RUTVIE
28 from .smotri import SmotriIE
31 class GenericIE(InfoExtractor):
32 IE_DESC = 'Generic downloader that works on some sites'
37 'url': 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
38 'file': '13601338388002.mp4',
39 'md5': '6e15c93721d7ec9e9ca3fdbf07982cfd',
41 'uploader': 'www.hodiho.fr',
42 'title': 'R\u00e9gis plante sa Jeep',
45 # bandcamp page with custom domain
47 'add_ie': ['Bandcamp'],
48 'url': 'http://bronyrock.com/track/the-pony-mash',
49 'file': '3235767654.mp3',
51 'title': 'The Pony Mash',
52 'uploader': 'M_Pallante',
54 'skip': 'There is a limit of 200 free downloads / month for the test song',
56 # embedded brightcove video
57 # it also tests brightcove videos that need to set the 'Referer' in the
60 'add_ie': ['Brightcove'],
61 'url': 'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/',
63 'id': '2765128793001',
65 'title': 'Le cours de bourse : l’analyse technique',
66 'description': 'md5:7e9ad046e968cb2d1114004aba466fd9',
67 'uploader': 'BFM BUSINESS',
70 'skip_download': True,
74 # https://github.com/rg3/youtube-dl/issues/2253
75 'url': 'http://bcove.me/i6nfkrc3',
76 'file': '3101154703001.mp4',
77 'md5': '0ba9446db037002366bab3b3eb30c88c',
79 'title': 'Still no power',
80 'uploader': 'thestar.com',
81 '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.',
83 'add_ie': ['Brightcove'],
85 # Direct link to a video
87 'url': 'http://media.w3.org/2010/05/sintel/trailer.mp4',
88 'md5': '67d406c2bcb6af27fa886f31aa934bbe',
93 'upload_date': '20100513',
98 'url': 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219',
99 'md5': '5644c6ca5d5782c1d0d350dad9bd840c',
101 'id': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ',
103 'title': '2cc213299525360.mov', # that's what we get
106 # second style of embedded ooyala videos
108 'url': 'http://www.smh.com.au/tv/business/show/financial-review-sunday/behind-the-scenes-financial-review-sunday--4350201.html',
110 'id': '13djJjYjptA1XpPx8r9kuzPyj3UZH0Uk',
112 'title': 'Behind-the-scenes: Financial Review Sunday ',
113 'description': 'Step inside Channel Nine studios for an exclusive tour of its upcoming financial business show.',
117 'skip_download': True,
122 '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',
126 'upload_date': '20130224',
127 'uploader_id': 'TheVerge',
128 '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.',
129 'uploader': 'The Verge',
130 'title': 'First Firefox OS phones side-by-side',
133 'skip_download': False,
138 'url': 'http://www.tested.com/science/weird/460206-tested-grinding-coffee-2000-frames-second/',
142 'title': 'Tested: Grinding Coffee at 2000 Frames Per Second',
143 'upload_date': '20140225',
144 'description': 'md5:06a40fbf30b220468f1e0957c0f558ff',
145 'uploader': 'Tested',
146 'uploader_id': 'testedcom',
148 # No need to test YoutubeIE here
150 'skip_download': True,
155 'url': 'http://www.theguardian.com/world/2014/mar/11/obama-zach-galifianakis-between-two-ferns',
156 'md5': '7cf780be104d40fea7bae52eed4a470e',
160 'title': 'Between Two Ferns with Zach Galifianakis: President Barack Obama',
161 'description': 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.',
166 'url': 'http://www.rg.ru/2014/03/15/reg-dfo/anklav-anons.html',
170 'title': 'Охотское море стало целиком российским',
171 'description': 'md5:5ed62483b14663e2a95ebbe115eb8f43',
175 'skip_download': True,
180 'url': 'http://en.support.wordpress.com/videos/ted-talks/',
181 'md5': 'deeeabcc1085eb2ba205474e7235a3d5',
185 'title': 'My web playroom',
186 'uploader': 'Ze Frank',
187 'description': 'md5:ddb2a40ecd6b6a147e400e535874947b',
190 # nowvideo embed hidden behind percent encoding
192 'url': 'http://www.waoanime.tv/the-super-dimension-fortress-macross-episode-1/',
193 'md5': '2baf4ddd70f697d94b1c18cf796d5107',
195 'id': '06e53103ca9aa',
197 'title': 'Macross Episode 001 Watch Macross Episode 001 onl',
198 'description': 'No description',
203 'url': 'http://www.tv-replay.fr/redirection/20-03-14/x-enius-arte-10753389.html',
204 'md5': '7653032cbb25bf6c80d80f217055fa43',
206 'id': '048195-004_PLUS7-F',
209 'description': 'md5:d5fdf32ef6613cdbfd516ae658abf168',
210 'upload_date': '20140320',
213 'skip_download': 'Requires rtmpdump'
218 'url': 'http://rbctv.rbc.ru/archive/news/562949990879132.shtml',
219 'md5': 'ec40048448e9284c9a1de77bb188108b',
221 'id': 'v27008541fad',
223 'title': 'Крым и Севастополь вошли в состав России',
224 'description': 'md5:fae01b61f68984c7bd2fa741e11c3175',
226 'upload_date': '20140318',
227 'uploader': 'rbctv_2012_4',
228 'uploader_id': 'rbctv_2012_4',
233 def report_download_webpage(self, video_id):
234 """Report webpage download."""
235 if not self._downloader.params.get('test', False):
236 self._downloader.report_warning('Falling back on generic information extractor.')
237 super(GenericIE, self).report_download_webpage(video_id)
239 def report_following_redirect(self, new_url):
240 """Report information extraction."""
241 self._downloader.to_screen('[redirect] Following redirect to %s' % new_url)
243 def _send_head(self, url):
244 """Check if it is a redirect, like url shorteners, in case return the new url."""
246 class HEADRedirectHandler(compat_urllib_request.HTTPRedirectHandler):
248 Subclass the HTTPRedirectHandler to make it use our
249 HEADRequest also on the redirected URL
251 def redirect_request(self, req, fp, code, msg, headers, newurl):
252 if code in (301, 302, 303, 307):
253 newurl = newurl.replace(' ', '%20')
254 newheaders = dict((k,v) for k,v in req.headers.items()
255 if k.lower() not in ("content-length", "content-type"))
257 # This function was deprecated in python 3.3 and removed in 3.4
258 origin_req_host = req.get_origin_req_host()
259 except AttributeError:
260 origin_req_host = req.origin_req_host
261 return HEADRequest(newurl,
263 origin_req_host=origin_req_host,
266 raise compat_urllib_error.HTTPError(req.get_full_url(), code, msg, headers, fp)
268 class HTTPMethodFallback(compat_urllib_request.BaseHandler):
270 Fallback to GET if HEAD is not allowed (405 HTTP error)
272 def http_error_405(self, req, fp, code, msg, headers):
276 newheaders = dict((k,v) for k,v in req.headers.items()
277 if k.lower() not in ("content-length", "content-type"))
278 return self.parent.open(compat_urllib_request.Request(req.get_full_url(),
280 origin_req_host=req.get_origin_req_host(),
284 opener = compat_urllib_request.OpenerDirector()
285 for handler in [compat_urllib_request.HTTPHandler, compat_urllib_request.HTTPDefaultErrorHandler,
286 HTTPMethodFallback, HEADRedirectHandler,
287 compat_urllib_request.HTTPErrorProcessor, compat_urllib_request.HTTPSHandler]:
288 opener.add_handler(handler())
290 response = opener.open(HEADRequest(url))
292 raise ExtractorError('Invalid URL protocol')
295 def _extract_rss(self, url, video_id, doc):
296 playlist_title = doc.find('./channel/title').text
297 playlist_desc_el = doc.find('./channel/description')
298 playlist_desc = None if playlist_desc_el is None else playlist_desc_el.text
302 'url': e.find('link').text,
303 'title': e.find('title').text,
304 } for e in doc.findall('./channel/item')]
309 'title': playlist_title,
310 'description': playlist_desc,
314 def _real_extract(self, url):
315 parsed_url = compat_urlparse.urlparse(url)
316 if not parsed_url.scheme:
317 default_search = self._downloader.params.get('default_search')
318 if default_search is None:
319 default_search = 'auto'
321 if default_search == 'auto':
323 self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http')
324 return self.url_result('http://' + url)
326 return self.url_result('ytsearch:' + url)
328 assert ':' in default_search
329 return self.url_result(default_search + url)
330 video_id = os.path.splitext(url.rstrip('/').split('/')[-1])[0]
332 self.to_screen('%s: Requesting header' % video_id)
335 response = self._send_head(url)
338 new_url = response.geturl()
340 self.report_following_redirect(new_url)
341 return self.url_result(new_url)
343 # Check for direct link to a video
344 content_type = response.headers.get('Content-Type', '')
345 m = re.match(r'^(?P<type>audio|video|application(?=/ogg$))/(?P<format_id>.+)$', content_type)
347 upload_date = response.headers.get('Last-Modified')
349 upload_date = unified_strdate(upload_date)
352 'title': os.path.splitext(url_basename(url))[0],
354 'format_id': m.group('format_id'),
356 'vcodec': 'none' if m.group('type') == 'audio' else None
358 'upload_date': upload_date,
361 except compat_urllib_error.HTTPError:
362 # This may be a stupid server that doesn't like HEAD, our UA, or so
366 webpage = self._download_webpage(url, video_id)
368 # since this is the last-resort InfoExtractor, if
369 # this error is thrown, it'll be thrown here
370 raise ExtractorError('Failed to download URL: %s' % url)
372 self.report_extraction(video_id)
376 doc = parse_xml(webpage)
378 return self._extract_rss(url, video_id, doc)
379 except compat_xml_parse_error:
382 # Sometimes embedded video player is hidden behind percent encoding
383 # (e.g. https://github.com/rg3/youtube-dl/issues/2448)
384 # Unescaping the whole page allows to handle those cases in a generic way
385 webpage = compat_urllib_parse.unquote(webpage)
387 # it's tempting to parse this further, but you would
388 # have to take into account all the variations like
389 # Video Title - Site Name
390 # Site Name | Video Title
391 # Video Title - Tagline | Site Name
392 # and so on and so forth; it's just not practical
393 video_title = self._html_search_regex(
394 r'(?s)<title>(.*?)</title>', webpage, 'video title',
397 # video uploader is domain name
398 video_uploader = self._search_regex(
399 r'^(?:https?://)?([^/]*)/.*', url, 'video uploader')
401 # Look for BrightCove:
402 bc_urls = BrightcoveIE._extract_brightcove_urls(webpage)
404 self.to_screen('Brightcove video detected.')
407 'url': smuggle_url(bc_url, {'Referer': url}),
408 'ie_key': 'Brightcove'
409 } for bc_url in bc_urls]
413 'title': video_title,
418 # Look for embedded (iframe) Vimeo player
420 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//player\.vimeo\.com/video/.+?)\1', webpage)
422 player_url = unescapeHTML(mobj.group('url'))
423 surl = smuggle_url(player_url, {'Referer': url})
424 return self.url_result(surl, 'Vimeo')
426 # Look for embedded (swf embed) Vimeo player
428 r'<embed[^>]+?src="(https?://(?:www\.)?vimeo\.com/moogaloop\.swf.+?)"', webpage)
430 return self.url_result(mobj.group(1), 'Vimeo')
432 # Look for embedded YouTube player
433 matches = re.findall(r'''(?x)
434 (?:<iframe[^>]+?src=|embedSWF\(\s*)
435 (["\'])(?P<url>(?:https?:)?//(?:www\.)?youtube\.com/
439 urlrs = [self.url_result(unescapeHTML(tuppl[1]), 'Youtube')
440 for tuppl in matches]
441 return self.playlist_result(
442 urlrs, playlist_id=video_id, playlist_title=video_title)
444 # Look for embedded Dailymotion player
445 matches = re.findall(
446 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/embed/video/.+?)\1', webpage)
448 urlrs = [self.url_result(unescapeHTML(tuppl[1]), 'Dailymotion')
449 for tuppl in matches]
450 return self.playlist_result(
451 urlrs, playlist_id=video_id, playlist_title=video_title)
453 # Look for embedded Wistia player
455 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage)
458 '_type': 'url_transparent',
459 'url': unescapeHTML(match.group('url')),
461 'uploader': video_uploader,
462 'title': video_title,
466 # Look for embedded blip.tv player
467 mobj = re.search(r'<meta\s[^>]*https?://api\.blip\.tv/\w+/redirect/\w+/(\d+)', webpage)
469 return self.url_result('http://blip.tv/a/a-'+mobj.group(1), 'BlipTV')
470 mobj = re.search(r'<(?:iframe|embed|object)\s[^>]*(https?://(?:\w+\.)?blip\.tv/(?:play/|api\.swf#)[a-zA-Z0-9]+)', webpage)
472 return self.url_result(mobj.group(1), 'BlipTV')
474 # Look for Bandcamp pages with custom domain
475 mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage)
477 burl = unescapeHTML(mobj.group(1))
478 # Don't set the extractor because it can be a track url or an album
479 return self.url_result(burl)
481 # Look for embedded Vevo player
483 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:cache\.)?vevo\.com/.+?)\1', webpage)
485 return self.url_result(mobj.group('url'))
487 # Look for Ooyala videos
488 mobj = (re.search(r'player.ooyala.com/[^"?]+\?[^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or
489 re.search(r'OO.Player.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage))
491 return OoyalaIE._build_url_result(mobj.group('ec'))
493 # Look for Aparat videos
494 mobj = re.search(r'<iframe src="(http://www\.aparat\.com/video/[^"]+)"', webpage)
496 return self.url_result(mobj.group(1), 'Aparat')
498 # Look for MPORA videos
499 mobj = re.search(r'<iframe .*?src="(http://mpora\.(?:com|de)/videos/[^"]+)"', webpage)
501 return self.url_result(mobj.group(1), 'Mpora')
503 # Look for embedded NovaMov player
505 r'<iframe[^>]+?src=(["\'])(?P<url>http://(?:(?:embed|www)\.)?novamov\.com/embed\.php.+?)\1', webpage)
507 return self.url_result(mobj.group('url'), 'NovaMov')
509 # Look for embedded NowVideo player
511 r'<iframe[^>]+?src=(["\'])(?P<url>http://(?:(?:embed|www)\.)?nowvideo\.(?:ch|sx|eu)/embed\.php.+?)\1', webpage)
513 return self.url_result(mobj.group('url'), 'NowVideo')
515 # Look for embedded Facebook player
517 r'<iframe[^>]+?src=(["\'])(?P<url>https://www\.facebook\.com/video/embed.+?)\1', webpage)
519 return self.url_result(mobj.group('url'), 'Facebook')
521 # Look for embedded VK player
522 mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://vk\.com/video_ext\.php.+?)\1', webpage)
524 return self.url_result(mobj.group('url'), 'VK')
526 # Look for embedded Huffington Post player
528 r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed\.live\.huffingtonpost\.com/.+?)\1', webpage)
530 return self.url_result(mobj.group('url'), 'HuffPost')
533 mobj = re.search(r'class=["\']embedly-card["\'][^>]href=["\'](?P<url>[^"\']+)', webpage)
535 return self.url_result(mobj.group('url'))
536 mobj = re.search(r'class=["\']embedly-embed["\'][^>]src=["\'][^"\']*url=(?P<url>[^&]+)', webpage)
538 return self.url_result(compat_urllib_parse.unquote(mobj.group('url')))
540 # Look for funnyordie embed
541 matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage)
543 urlrs = [self.url_result(unescapeHTML(eurl), 'FunnyOrDie')
545 return self.playlist_result(
546 urlrs, playlist_id=video_id, playlist_title=video_title)
548 # Look for embedded RUTV player
549 rutv_url = RUTVIE._extract_url(webpage)
551 return self.url_result(rutv_url, 'RUTV')
553 # Look for embedded TED player
555 r'<iframe[^>]+?src=(["\'])(?P<url>http://embed\.ted\.com/.+?)\1', webpage)
557 return self.url_result(mobj.group('url'), 'TED')
559 # Look for embedded arte.tv player
561 r'<script [^>]*?src="(?P<url>http://www\.arte\.tv/playerv2/embed[^"]+)"',
564 return self.url_result(mobj.group('url'), 'ArteTVEmbed')
566 # Look for embedded smotri.com player
567 smotri_url = SmotriIE._extract_url(webpage)
569 return self.url_result(smotri_url, 'Smotri')
571 # Start with something easy: JW Player in SWFObject
572 mobj = re.search(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage)
574 # Look for gorilla-vid style embedding
575 mobj = re.search(r'(?s)(?:jw_plugins|JWPlayerOptions).*?file\s*:\s*["\'](.*?)["\']', webpage)
577 # Broaden the search a little bit
578 mobj = re.search(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage)
580 # Broaden the search a little bit: JWPlayer JS loader
581 mobj = re.search(r'[^A-Za-z0-9]?file["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage)
584 # Try to find twitter cards info
585 mobj = re.search(r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage)
587 # We look for Open Graph info:
588 # We have to match any number spaces between elements, some sites try to align them (eg.: statigr.am)
589 m_video_type = re.search(r'<meta.*?property="og:video:type".*?content="video/(.*?)"', webpage)
590 # We only look in og:video if the MIME type is a video, don't try if it's a Flash player:
591 if m_video_type is not None:
592 mobj = re.search(r'<meta.*?property="og:video".*?content="(.*?)"', webpage)
595 mobj = re.search(r'<video[^<]*(?:>.*?<source.*?)? src="([^"]+)"', webpage, flags=re.DOTALL)
598 r'(?i)<meta\s+(?=(?:[a-z-]+="[^"]+"\s+)*http-equiv="refresh")'
599 r'(?:[a-z-]+="[^"]+"\s+)*?content="[0-9]{,2};url=\'([^\']+)\'"',
602 new_url = mobj.group(1)
603 self.report_following_redirect(new_url)
609 raise ExtractorError('Unsupported URL: %s' % url)
611 # It's possible that one of the regexes
612 # matched, but returned an empty group:
613 if mobj.group(1) is None:
614 raise ExtractorError('Did not find a valid video URL at %s' % url)
616 video_url = mobj.group(1)
617 video_url = compat_urlparse.urljoin(url, video_url)
618 video_id = compat_urllib_parse.unquote(os.path.basename(video_url))
620 # Sometimes, jwplayer extraction will result in a YouTube URL
621 if YoutubeIE.suitable(video_url):
622 return self.url_result(video_url, 'Youtube')
624 # here's a fun little line of code for you:
625 video_id = os.path.splitext(video_id)[0]
630 'uploader': video_uploader,
631 'title': video_title,