]> Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/generic.py
Rapsys Git / youtubedl / blob
? search:


a27ec342e905445b6cc640f028897335a908fc9f
[youtubedl] / youtube_dl / extractor / generic.py
1 # encoding: utf-8
2
3 from __future__ import unicode_literals
4
5 import os
6 import re
7
8 from .common import InfoExtractor
9 from .youtube import YoutubeIE
10 from ..utils import (
11 compat_urllib_parse,
12 compat_urlparse,
13 compat_xml_parse_error,
14
15 determine_ext,
16 ExtractorError,
17 float_or_none,
18 HEADRequest,
19 orderedSet,
20 parse_xml,
21 smuggle_url,
22 unescapeHTML,
23 unified_strdate,
24 unsmuggle_url,
25 url_basename,
26 )
27 from .brightcove import BrightcoveIE
28 from .ooyala import OoyalaIE
29 from .rutv import RUTVIE
30 from .smotri import SmotriIE
31 from .condenast import CondeNastIE
32
33
34 class GenericIE(InfoExtractor):
35 IE_DESC = 'Generic downloader that works on some sites'
36 _VALID_URL = r'.*'
37 IE_NAME = 'generic'
38 _TESTS = [
39 {
40 'url': 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
41 'md5': '85b90ccc9d73b4acd9138d3af4c27f89',
42 'info_dict': {
43 'id': '13601338388002',
44 'ext': 'mp4',
45 'uploader': 'www.hodiho.fr',
46 'title': 'R\u00e9gis plante sa Jeep',
47 }
48 },
49 # bandcamp page with custom domain
50 {
51 'add_ie': ['Bandcamp'],
52 'url': 'http://bronyrock.com/track/the-pony-mash',
53 'info_dict': {
54 'id': '3235767654',
55 'ext': 'mp3',
56 'title': 'The Pony Mash',
57 'uploader': 'M_Pallante',
58 },
59 'skip': 'There is a limit of 200 free downloads / month for the test song',
60 },
61 # embedded brightcove video
62 # it also tests brightcove videos that need to set the 'Referer' in the
63 # http requests
64 {
65 'add_ie': ['Brightcove'],
66 'url': 'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/',
67 'info_dict': {
68 'id': '2765128793001',
69 'ext': 'mp4',
70 'title': 'Le cours de bourse : l’analyse technique',
71 'description': 'md5:7e9ad046e968cb2d1114004aba466fd9',
72 'uploader': 'BFM BUSINESS',
73 },
74 'params': {
75 'skip_download': True,
76 },
77 },
78 {
79 # https://github.com/rg3/youtube-dl/issues/2253
80 'url': 'http://bcove.me/i6nfkrc3',
81 'md5': '0ba9446db037002366bab3b3eb30c88c',
82 'info_dict': {
83 'id': '3101154703001',
84 'ext': 'mp4',
85 'title': 'Still no power',
86 'uploader': 'thestar.com',
87 '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 },
89 'add_ie': ['Brightcove'],
90 },
91 {
92 'url': 'http://www.championat.com/video/football/v/87/87499.html',
93 'md5': 'fb973ecf6e4a78a67453647444222983',
94 'info_dict': {
95 'id': '3414141473001',
96 'ext': 'mp4',
97 'title': 'Видео. Удаление Дзагоева (ЦСКА)',
98 'description': 'Онлайн-трансляция матча ЦСКА - "Волга"',
99 'uploader': 'Championat',
100 },
101 },
102 # Direct link to a video
103 {
104 'url': 'http://media.w3.org/2010/05/sintel/trailer.mp4',
105 'md5': '67d406c2bcb6af27fa886f31aa934bbe',
106 'info_dict': {
107 'id': 'trailer',
108 'ext': 'mp4',
109 'title': 'trailer',
110 'upload_date': '20100513',
111 }
112 },
113 # ooyala video
114 {
115 'url': 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219',
116 'md5': '5644c6ca5d5782c1d0d350dad9bd840c',
117 'info_dict': {
118 'id': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ',
119 'ext': 'mp4',
120 'title': '2cc213299525360.mov', # that's what we get
121 },
122 },
123 # google redirect
124 {
125 '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 'info_dict': {
127 'id': 'cmQHVoWB5FY',
128 'ext': 'mp4',
129 'upload_date': '20130224',
130 'uploader_id': 'TheVerge',
131 '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.',
132 'uploader': 'The Verge',
133 'title': 'First Firefox OS phones side-by-side',
134 },
135 'params': {
136 'skip_download': False,
137 }
138 },
139 # embed.ly video
140 {
141 'url': 'http://www.tested.com/science/weird/460206-tested-grinding-coffee-2000-frames-second/',
142 'info_dict': {
143 'id': '9ODmcdjQcHQ',
144 'ext': 'mp4',
145 'title': 'Tested: Grinding Coffee at 2000 Frames Per Second',
146 'upload_date': '20140225',
147 'description': 'md5:06a40fbf30b220468f1e0957c0f558ff',
148 'uploader': 'Tested',
149 'uploader_id': 'testedcom',
150 },
151 # No need to test YoutubeIE here
152 'params': {
153 'skip_download': True,
154 },
155 },
156 # funnyordie embed
157 {
158 'url': 'http://www.theguardian.com/world/2014/mar/11/obama-zach-galifianakis-between-two-ferns',
159 'info_dict': {
160 'id': '18e820ec3f',
161 'ext': 'mp4',
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.',
164 },
165 },
166 # RUTV embed
167 {
168 'url': 'http://www.rg.ru/2014/03/15/reg-dfo/anklav-anons.html',
169 'info_dict': {
170 'id': '776940',
171 'ext': 'mp4',
172 'title': 'Охотское море стало целиком российским',
173 'description': 'md5:5ed62483b14663e2a95ebbe115eb8f43',
174 },
175 'params': {
176 # m3u8 download
177 'skip_download': True,
178 },
179 },
180 # Embedded TED video
181 {
182 'url': 'http://en.support.wordpress.com/videos/ted-talks/',
183 'md5': '65fdff94098e4a607385a60c5177c638',
184 'info_dict': {
185 'id': '1969',
186 'ext': 'mp4',
187 'title': 'Hidden miracles of the natural world',
188 'uploader': 'Louie Schwartzberg',
189 'description': 'md5:8145d19d320ff3e52f28401f4c4283b9',
190 }
191 },
192 # Embeded Ustream video
193 {
194 'url': 'http://www.american.edu/spa/pti/nsa-privacy-janus-2014.cfm',
195 'md5': '27b99cdb639c9b12a79bca876a073417',
196 'info_dict': {
197 'id': '45734260',
198 'ext': 'flv',
199 'uploader': 'AU SPA: The NSA and Privacy',
200 'title': 'NSA and Privacy Forum Debate featuring General Hayden and Barton Gellman'
201 }
202 },
203 # nowvideo embed hidden behind percent encoding
204 {
205 'url': 'http://www.waoanime.tv/the-super-dimension-fortress-macross-episode-1/',
206 'md5': '2baf4ddd70f697d94b1c18cf796d5107',
207 'info_dict': {
208 'id': '06e53103ca9aa',
209 'ext': 'flv',
210 'title': 'Macross Episode 001 Watch Macross Episode 001 onl',
211 'description': 'No description',
212 },
213 },
214 # arte embed
215 {
216 'url': 'http://www.tv-replay.fr/redirection/20-03-14/x-enius-arte-10753389.html',
217 'md5': '7653032cbb25bf6c80d80f217055fa43',
218 'info_dict': {
219 'id': '048195-004_PLUS7-F',
220 'ext': 'flv',
221 'title': 'X:enius',
222 'description': 'md5:d5fdf32ef6613cdbfd516ae658abf168',
223 'upload_date': '20140320',
224 },
225 'params': {
226 'skip_download': 'Requires rtmpdump'
227 }
228 },
229 # Condé Nast embed
230 {
231 'url': 'http://www.wired.com/2014/04/honda-asimo/',
232 'md5': 'ba0dfe966fa007657bd1443ee672db0f',
233 'info_dict': {
234 'id': '53501be369702d3275860000',
235 'ext': 'mp4',
236 'title': 'Honda’s New Asimo Robot Is More Human Than Ever',
237 }
238 },
239 # Dailymotion embed
240 {
241 'url': 'http://www.spi0n.com/zap-spi0n-com-n216/',
242 'md5': '441aeeb82eb72c422c7f14ec533999cd',
243 'info_dict': {
244 'id': 'k2mm4bCdJ6CQ2i7c8o2',
245 'ext': 'mp4',
246 'title': 'Le Zap de Spi0n n°216 - Zapping du Web',
247 'uploader': 'Spi0n',
248 },
249 'add_ie': ['Dailymotion'],
250 },
251 # YouTube embed
252 {
253 'url': 'http://www.badzine.de/ansicht/datum/2014/06/09/so-funktioniert-die-neue-englische-badminton-liga.html',
254 'info_dict': {
255 'id': 'FXRb4ykk4S0',
256 'ext': 'mp4',
257 'title': 'The NBL Auction 2014',
258 'uploader': 'BADMINTON England',
259 'uploader_id': 'BADMINTONEvents',
260 'upload_date': '20140603',
261 'description': 'md5:9ef128a69f1e262a700ed83edb163a73',
262 },
263 'add_ie': ['Youtube'],
264 'params': {
265 'skip_download': True,
266 }
267 },
268 # MTVSercices embed
269 {
270 'url': 'http://www.gametrailers.com/news-post/76093/north-america-europe-is-getting-that-mario-kart-8-mercedes-dlc-too',
271 'md5': '35727f82f58c76d996fc188f9755b0d5',
272 'info_dict': {
273 'id': '0306a69b-8adf-4fb5-aace-75f8e8cbfca9',
274 'ext': 'mp4',
275 'title': 'Review',
276 'description': 'Mario\'s life in the fast lane has never looked so good.',
277 },
278 },
279 # YouTube embed via <data-embed-url="">
280 {
281 'url': 'https://play.google.com/store/apps/details?id=com.gameloft.android.ANMP.GloftA8HM',
282 'info_dict': {
283 'id': '4vAffPZIT44',
284 'ext': 'mp4',
285 'title': 'Asphalt 8: Airborne - Update - Welcome to Dubai!',
286 'uploader': 'Gameloft',
287 'uploader_id': 'gameloft',
288 'upload_date': '20140828',
289 'description': 'md5:c80da9ed3d83ae6d1876c834de03e1c4',
290 },
291 'params': {
292 'skip_download': True,
293 }
294 },
295 # Camtasia studio
296 {
297 'url': 'http://www.ll.mit.edu/workshops/education/videocourses/antennas/lecture1/video/',
298 'playlist': [{
299 'md5': '0c5e352edabf715d762b0ad4e6d9ee67',
300 'info_dict': {
301 'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
302 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - video1',
303 'ext': 'flv',
304 'duration': 2235.90,
305 }
306 }, {
307 'md5': '10e4bb3aaca9fd630e273ff92d9f3c63',
308 'info_dict': {
309 'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final_PIP',
310 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - pip',
311 'ext': 'flv',
312 'duration': 2235.93,
313 }
314 }],
315 'info_dict': {
316 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
317 }
318 },
319 # Flowplayer
320 {
321 'url': 'http://www.handjobhub.com/video/busty-blonde-siri-tit-fuck-while-wank-6313.html',
322 'md5': '9d65602bf31c6e20014319c7d07fba27',
323 'info_dict': {
324 'id': '5123ea6d5e5a7',
325 'ext': 'mp4',
326 'age_limit': 18,
327 'uploader': 'www.handjobhub.com',
328 'title': 'Busty Blonde Siri Tit Fuck While Wank at HandjobHub.com',
329 }
330 },
331 # RSS feed
332 {
333 'url': 'http://phihag.de/2014/youtube-dl/rss2.xml',
334 'info_dict': {
335 'id': 'http://phihag.de/2014/youtube-dl/rss2.xml',
336 'title': 'Zero Punctuation',
337 'description': 're:'
338 },
339 'playlist_mincount': 11,
340 },
341 # Multiple brightcove videos
342 # https://github.com/rg3/youtube-dl/issues/2283
343 {
344 'url': 'http://www.newyorker.com/online/blogs/newsdesk/2014/01/always-never-nuclear-command-and-control.html',
345 'info_dict': {
346 'id': 'always-never',
347 'title': 'Always / Never - The New Yorker',
348 },
349 'playlist_count': 3,
350 'params': {
351 'extract_flat': False,
352 'skip_download': True,
353 }
354 },
355 # MLB embed
356 {
357 'url': 'http://umpire-empire.com/index.php/topic/58125-laz-decides-no-thats-low/',
358 'md5': '96f09a37e44da40dd083e12d9a683327',
359 'info_dict': {
360 'id': '33322633',
361 'ext': 'mp4',
362 'title': 'Ump changes call to ball',
363 'description': 'md5:71c11215384298a172a6dcb4c2e20685',
364 'duration': 48,
365 'timestamp': 1401537900,
366 'upload_date': '20140531',
367 'thumbnail': 're:^https?://.*\.jpg$',
368 },
369 },
370 # Wistia embed
371 {
372 'url': 'http://education-portal.com/academy/lesson/north-american-exploration-failed-colonies-of-spain-france-england.html#lesson',
373 'md5': '8788b683c777a5cf25621eaf286d0c23',
374 'info_dict': {
375 'id': '1cfaf6b7ea',
376 'ext': 'mov',
377 'title': 'md5:51364a8d3d009997ba99656004b5e20d',
378 'duration': 643.0,
379 'filesize': 182808282,
380 'uploader': 'education-portal.com',
381 },
382 },
383 {
384 'url': 'http://thoughtworks.wistia.com/medias/uxjb0lwrcz',
385 'md5': 'baf49c2baa8a7de5f3fc145a8506dcd4',
386 'info_dict': {
387 'id': 'uxjb0lwrcz',
388 'ext': 'mp4',
389 'title': 'Conversation about Hexagonal Rails Part 1 - ThoughtWorks',
390 'duration': 1715.0,
391 'uploader': 'thoughtworks.wistia.com',
392 },
393 },
394 # Direct download with broken HEAD
395 {
396 'url': 'http://ai-radio.org:8000/radio.opus',
397 'info_dict': {
398 'id': 'radio',
399 'ext': 'opus',
400 'title': 'radio',
401 },
402 'params': {
403 'skip_download': True, # infinite live stream
404 },
405 'expected_warnings': [
406 r'501.*Not Implemented'
407 ],
408 },
409 # Soundcloud embed
410 {
411 'url': 'http://nakedsecurity.sophos.com/2014/10/29/sscc-171-are-you-sure-that-1234-is-a-bad-password-podcast/',
412 'info_dict': {
413 'id': '174391317',
414 'ext': 'mp3',
415 'description': 'md5:ff867d6b555488ad3c52572bb33d432c',
416 'uploader': 'Sophos Security',
417 'title': 'Chet Chat 171 - Oct 29, 2014',
418 'upload_date': '20141029',
419 }
420 }
421 ]
422
423 def report_following_redirect(self, new_url):
424 """Report information extraction."""
425 self._downloader.to_screen('[redirect] Following redirect to %s' % new_url)
426
427 def _extract_rss(self, url, video_id, doc):
428 playlist_title = doc.find('./channel/title').text
429 playlist_desc_el = doc.find('./channel/description')
430 playlist_desc = None if playlist_desc_el is None else playlist_desc_el.text
431
432 entries = [{
433 '_type': 'url',
434 'url': e.find('link').text,
435 'title': e.find('title').text,
436 } for e in doc.findall('./channel/item')]
437
438 return {
439 '_type': 'playlist',
440 'id': url,
441 'title': playlist_title,
442 'description': playlist_desc,
443 'entries': entries,
444 }
445
446 def _extract_camtasia(self, url, video_id, webpage):
447 """ Returns None if no camtasia video can be found. """
448
449 camtasia_cfg = self._search_regex(
450 r'fo\.addVariable\(\s*"csConfigFile",\s*"([^"]+)"\s*\);',
451 webpage, 'camtasia configuration file', default=None)
452 if camtasia_cfg is None:
453 return None
454
455 title = self._html_search_meta('DC.title', webpage, fatal=True)
456
457 camtasia_url = compat_urlparse.urljoin(url, camtasia_cfg)
458 camtasia_cfg = self._download_xml(
459 camtasia_url, video_id,
460 note='Downloading camtasia configuration',
461 errnote='Failed to download camtasia configuration')
462 fileset_node = camtasia_cfg.find('./playlist/array/fileset')
463
464 entries = []
465 for n in fileset_node.getchildren():
466 url_n = n.find('./uri')
467 if url_n is None:
468 continue
469
470 entries.append({
471 'id': os.path.splitext(url_n.text.rpartition('/')[2])[0],
472 'title': '%s - %s' % (title, n.tag),
473 'url': compat_urlparse.urljoin(url, url_n.text),
474 'duration': float_or_none(n.find('./duration').text),
475 })
476
477 return {
478 '_type': 'playlist',
479 'entries': entries,
480 'title': title,
481 }
482
483 def _real_extract(self, url):
484 if url.startswith('//'):
485 return {
486 '_type': 'url',
487 'url': self.http_scheme() + url,
488 }
489
490 parsed_url = compat_urlparse.urlparse(url)
491 if not parsed_url.scheme:
492 default_search = self._downloader.params.get('default_search')
493 if default_search is None:
494 default_search = 'fixup_error'
495
496 if default_search in ('auto', 'auto_warning', 'fixup_error'):
497 if '/' in url:
498 self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http')
499 return self.url_result('http://' + url)
500 elif default_search != 'fixup_error':
501 if default_search == 'auto_warning':
502 if re.match(r'^(?:url|URL)$', url):
503 raise ExtractorError(
504 'Invalid URL: %r . Call youtube-dl like this: youtube-dl -v "https://www.youtube.com/watch?v=BaW_jenozKc" ' % url,
505 expected=True)
506 else:
507 self._downloader.report_warning(
508 'Falling back to youtube search for %s . Set --default-search "auto" to suppress this warning.' % url)
509 return self.url_result('ytsearch:' + url)
510
511 if default_search in ('error', 'fixup_error'):
512 raise ExtractorError(
513 ('%r is not a valid URL. '
514 'Set --default-search "ytsearch" (or run youtube-dl "ytsearch:%s" ) to search YouTube'
515 ) % (url, url), expected=True)
516 else:
517 if ':' not in default_search:
518 default_search += ':'
519 return self.url_result(default_search + url)
520
521 url, smuggled_data = unsmuggle_url(url)
522 force_videoid = None
523 is_intentional = smuggled_data and smuggled_data.get('to_generic')
524 if smuggled_data and 'force_videoid' in smuggled_data:
525 force_videoid = smuggled_data['force_videoid']
526 video_id = force_videoid
527 else:
528 video_id = os.path.splitext(url.rstrip('/').split('/')[-1])[0]
529
530 self.to_screen('%s: Requesting header' % video_id)
531
532 head_req = HEADRequest(url)
533 head_response = self._request_webpage(
534 head_req, video_id,
535 note=False, errnote='Could not send HEAD request to %s' % url,
536 fatal=False)
537
538 if head_response is not False:
539 # Check for redirect
540 new_url = head_response.geturl()
541 if url != new_url:
542 self.report_following_redirect(new_url)
543 if force_videoid:
544 new_url = smuggle_url(
545 new_url, {'force_videoid': force_videoid})
546 return self.url_result(new_url)
547
548 full_response = None
549 if head_response is False:
550 full_response = self._request_webpage(url, video_id)
551 head_response = full_response
552
553 # Check for direct link to a video
554 content_type = head_response.headers.get('Content-Type', '')
555 m = re.match(r'^(?P<type>audio|video|application(?=/ogg$))/(?P<format_id>.+)$', content_type)
556 if m:
557 upload_date = unified_strdate(
558 head_response.headers.get('Last-Modified'))
559 return {
560 'id': video_id,
561 'title': os.path.splitext(url_basename(url))[0],
562 'formats': [{
563 'format_id': m.group('format_id'),
564 'url': url,
565 'vcodec': 'none' if m.group('type') == 'audio' else None
566 }],
567 'upload_date': upload_date,
568 }
569
570 if not self._downloader.params.get('test', False) and not is_intentional:
571 self._downloader.report_warning('Falling back on generic information extractor.')
572
573 if full_response:
574 webpage = self._webpage_read_content(full_response, url, video_id)
575 else:
576 webpage = self._download_webpage(url, video_id)
577 self.report_extraction(video_id)
578
579 # Is it an RSS feed?
580 try:
581 doc = parse_xml(webpage)
582 if doc.tag == 'rss':
583 return self._extract_rss(url, video_id, doc)
584 except compat_xml_parse_error:
585 pass
586
587 # Is it a Camtasia project?
588 camtasia_res = self._extract_camtasia(url, video_id, webpage)
589 if camtasia_res is not None:
590 return camtasia_res
591
592 # Sometimes embedded video player is hidden behind percent encoding
593 # (e.g. https://github.com/rg3/youtube-dl/issues/2448)
594 # Unescaping the whole page allows to handle those cases in a generic way
595 webpage = compat_urllib_parse.unquote(webpage)
596
597 # it's tempting to parse this further, but you would
598 # have to take into account all the variations like
599 # Video Title - Site Name
600 # Site Name | Video Title
601 # Video Title - Tagline | Site Name
602 # and so on and so forth; it's just not practical
603 video_title = self._html_search_regex(
604 r'(?s)<title>(.*?)</title>', webpage, 'video title',
605 default='video')
606
607 # Try to detect age limit automatically
608 age_limit = self._rta_search(webpage)
609 # And then there are the jokers who advertise that they use RTA,
610 # but actually don't.
611 AGE_LIMIT_MARKERS = [
612 r'Proudly Labeled <a href="http://www.rtalabel.org/" title="Restricted to Adults">RTA</a>',
613 ]
614 if any(re.search(marker, webpage) for marker in AGE_LIMIT_MARKERS):
615 age_limit = 18
616
617 # video uploader is domain name
618 video_uploader = self._search_regex(
619 r'^(?:https?://)?([^/]*)/.*', url, 'video uploader')
620
621 # Helper method
622 def _playlist_from_matches(matches, getter, ie=None):
623 urlrs = orderedSet(
624 self.url_result(self._proto_relative_url(getter(m)), ie)
625 for m in matches)
626 return self.playlist_result(
627 urlrs, playlist_id=video_id, playlist_title=video_title)
628
629 # Look for BrightCove:
630 bc_urls = BrightcoveIE._extract_brightcove_urls(webpage)
631 if bc_urls:
632 self.to_screen('Brightcove video detected.')
633 entries = [{
634 '_type': 'url',
635 'url': smuggle_url(bc_url, {'Referer': url}),
636 'ie_key': 'Brightcove'
637 } for bc_url in bc_urls]
638
639 return {
640 '_type': 'playlist',
641 'title': video_title,
642 'id': video_id,
643 'entries': entries,
644 }
645
646 # Look for embedded (iframe) Vimeo player
647 mobj = re.search(
648 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//player\.vimeo\.com/video/.+?)\1', webpage)
649 if mobj:
650 player_url = unescapeHTML(mobj.group('url'))
651 surl = smuggle_url(player_url, {'Referer': url})
652 return self.url_result(surl)
653
654 # Look for embedded (swf embed) Vimeo player
655 mobj = re.search(
656 r'<embed[^>]+?src="((?:https?:)?//(?:www\.)?vimeo\.com/moogaloop\.swf.+?)"', webpage)
657 if mobj:
658 return self.url_result(mobj.group(1))
659
660 # Look for embedded YouTube player
661 matches = re.findall(r'''(?x)
662 (?:
663 <iframe[^>]+?src=|
664 data-video-url=|
665 <embed[^>]+?src=|
666 embedSWF\(?:\s*|
667 new\s+SWFObject\(
668 )
669 (["\'])
670 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
671 (?:embed|v|p)/.+?)
672 \1''', webpage)
673 if matches:
674 return _playlist_from_matches(
675 matches, lambda m: unescapeHTML(m[1]))
676
677 # Look for embedded Dailymotion player
678 matches = re.findall(
679 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/embed/video/.+?)\1', webpage)
680 if matches:
681 return _playlist_from_matches(
682 matches, lambda m: unescapeHTML(m[1]))
683
684 # Look for embedded Dailymotion playlist player (#3822)
685 m = re.search(
686 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.[a-z]{2,3}/widget/jukebox\?.+?)\1', webpage)
687 if m:
688 playlists = re.findall(
689 r'list\[\]=/playlist/([^/]+)/', unescapeHTML(m.group('url')))
690 if playlists:
691 return _playlist_from_matches(
692 playlists, lambda p: '//dailymotion.com/playlist/%s' % p)
693
694 # Look for embedded Wistia player
695 match = re.search(
696 r'<(?:meta[^>]+?content|iframe[^>]+?src)=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage)
697 if match:
698 embed_url = self._proto_relative_url(
699 unescapeHTML(match.group('url')))
700 return {
701 '_type': 'url_transparent',
702 'url': embed_url,
703 'ie_key': 'Wistia',
704 'uploader': video_uploader,
705 'title': video_title,
706 'id': video_id,
707 }
708
709 match = re.search(r'(?:id=["\']wistia_|data-wistia-?id=["\']|Wistia\.embed\(["\'])(?P<id>[^"\']+)', webpage)
710 if match:
711 return {
712 '_type': 'url_transparent',
713 'url': 'http://fast.wistia.net/embed/iframe/{0:}'.format(match.group('id')),
714 'ie_key': 'Wistia',
715 'uploader': video_uploader,
716 'title': video_title,
717 'id': match.group('id')
718 }
719
720 # Look for embedded blip.tv player
721 mobj = re.search(r'<meta\s[^>]*https?://api\.blip\.tv/\w+/redirect/\w+/(\d+)', webpage)
722 if mobj:
723 return self.url_result('http://blip.tv/a/a-'+mobj.group(1), 'BlipTV')
724 mobj = re.search(r'<(?:iframe|embed|object)\s[^>]*(https?://(?:\w+\.)?blip\.tv/(?:play/|api\.swf#)[a-zA-Z0-9_]+)', webpage)
725 if mobj:
726 return self.url_result(mobj.group(1), 'BlipTV')
727
728 # Look for embedded condenast player
729 matches = re.findall(
730 r'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="(https?://player\.cnevids\.com/embed/[^"]+")',
731 webpage)
732 if matches:
733 return {
734 '_type': 'playlist',
735 'entries': [{
736 '_type': 'url',
737 'ie_key': 'CondeNast',
738 'url': ma,
739 } for ma in matches],
740 'title': video_title,
741 'id': video_id,
742 }
743
744 # Look for Bandcamp pages with custom domain
745 mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage)
746 if mobj is not None:
747 burl = unescapeHTML(mobj.group(1))
748 # Don't set the extractor because it can be a track url or an album
749 return self.url_result(burl)
750
751 # Look for embedded Vevo player
752 mobj = re.search(
753 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:cache\.)?vevo\.com/.+?)\1', webpage)
754 if mobj is not None:
755 return self.url_result(mobj.group('url'))
756
757 # Look for Ooyala videos
758 mobj = (re.search(r'player.ooyala.com/[^"?]+\?[^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or
759 re.search(r'OO.Player.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage))
760 if mobj is not None:
761 return OoyalaIE._build_url_result(mobj.group('ec'))
762
763 # Look for Aparat videos
764 mobj = re.search(r'<iframe .*?src="(http://www\.aparat\.com/video/[^"]+)"', webpage)
765 if mobj is not None:
766 return self.url_result(mobj.group(1), 'Aparat')
767
768 # Look for MPORA videos
769 mobj = re.search(r'<iframe .*?src="(http://mpora\.(?:com|de)/videos/[^"]+)"', webpage)
770 if mobj is not None:
771 return self.url_result(mobj.group(1), 'Mpora')
772
773 # Look for embedded NovaMov-based player
774 mobj = re.search(
775 r'''(?x)<(?:pagespeed_)?iframe[^>]+?src=(["\'])
776 (?P<url>http://(?:(?:embed|www)\.)?
777 (?:novamov\.com|
778 nowvideo\.(?:ch|sx|eu|at|ag|co)|
779 videoweed\.(?:es|com)|
780 movshare\.(?:net|sx|ag)|
781 divxstage\.(?:eu|net|ch|co|at|ag))
782 /embed\.php.+?)\1''', webpage)
783 if mobj is not None:
784 return self.url_result(mobj.group('url'))
785
786 # Look for embedded Facebook player
787 mobj = re.search(
788 r'<iframe[^>]+?src=(["\'])(?P<url>https://www\.facebook\.com/video/embed.+?)\1', webpage)
789 if mobj is not None:
790 return self.url_result(mobj.group('url'), 'Facebook')
791
792 # Look for embedded VK player
793 mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://vk\.com/video_ext\.php.+?)\1', webpage)
794 if mobj is not None:
795 return self.url_result(mobj.group('url'), 'VK')
796
797 # Look for embedded ivi player
798 mobj = re.search(r'<embed[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?ivi\.ru/video/player.+?)\1', webpage)
799 if mobj is not None:
800 return self.url_result(mobj.group('url'), 'Ivi')
801
802 # Look for embedded Huffington Post player
803 mobj = re.search(
804 r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed\.live\.huffingtonpost\.com/.+?)\1', webpage)
805 if mobj is not None:
806 return self.url_result(mobj.group('url'), 'HuffPost')
807
808 # Look for embed.ly
809 mobj = re.search(r'class=["\']embedly-card["\'][^>]href=["\'](?P<url>[^"\']+)', webpage)
810 if mobj is not None:
811 return self.url_result(mobj.group('url'))
812 mobj = re.search(r'class=["\']embedly-embed["\'][^>]src=["\'][^"\']*url=(?P<url>[^&]+)', webpage)
813 if mobj is not None:
814 return self.url_result(compat_urllib_parse.unquote(mobj.group('url')))
815
816 # Look for funnyordie embed
817 matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage)
818 if matches:
819 return _playlist_from_matches(
820 matches, getter=unescapeHTML, ie='FunnyOrDie')
821
822 # Look for embedded RUTV player
823 rutv_url = RUTVIE._extract_url(webpage)
824 if rutv_url:
825 return self.url_result(rutv_url, 'RUTV')
826
827 # Look for embedded TED player
828 mobj = re.search(
829 r'<iframe[^>]+?src=(["\'])(?P<url>http://embed\.ted\.com/.+?)\1', webpage)
830 if mobj is not None:
831 return self.url_result(mobj.group('url'), 'TED')
832
833 # Look for embedded Ustream videos
834 mobj = re.search(
835 r'<iframe[^>]+?src=(["\'])(?P<url>http://www\.ustream\.tv/embed/.+?)\1', webpage)
836 if mobj is not None:
837 return self.url_result(mobj.group('url'), 'Ustream')
838
839 # Look for embedded arte.tv player
840 mobj = re.search(
841 r'<script [^>]*?src="(?P<url>http://www\.arte\.tv/playerv2/embed[^"]+)"',
842 webpage)
843 if mobj is not None:
844 return self.url_result(mobj.group('url'), 'ArteTVEmbed')
845
846 # Look for embedded smotri.com player
847 smotri_url = SmotriIE._extract_url(webpage)
848 if smotri_url:
849 return self.url_result(smotri_url, 'Smotri')
850
851 # Look for embeded soundcloud player
852 mobj = re.search(
853 r'<iframe\s+(?:[a-zA-Z0-9_-]+="[^"]+"\s+)*src="(?P<url>https?://(?:w\.)?soundcloud\.com/player[^"]+)"',
854 webpage)
855 if mobj is not None:
856 url = unescapeHTML(mobj.group('url'))
857 return self.url_result(url)
858
859 # Look for embedded vulture.com player
860 mobj = re.search(
861 r'<iframe src="(?P<url>https?://video\.vulture\.com/[^"]+)"',
862 webpage)
863 if mobj is not None:
864 url = unescapeHTML(mobj.group('url'))
865 return self.url_result(url, ie='Vulture')
866
867 # Look for embedded mtvservices player
868 mobj = re.search(
869 r'<iframe src="(?P<url>https?://media\.mtvnservices\.com/embed/[^"]+)"',
870 webpage)
871 if mobj is not None:
872 url = unescapeHTML(mobj.group('url'))
873 return self.url_result(url, ie='MTVServicesEmbedded')
874
875 # Look for embedded yahoo player
876 mobj = re.search(
877 r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:screen|movies)\.yahoo\.com/.+?\.html\?format=embed)\1',
878 webpage)
879 if mobj is not None:
880 return self.url_result(mobj.group('url'), 'Yahoo')
881
882 # Look for embedded sbs.com.au player
883 mobj = re.search(
884 r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:www\.)sbs\.com\.au/ondemand/video/single/.+?)\1',
885 webpage)
886 if mobj is not None:
887 return self.url_result(mobj.group('url'), 'SBS')
888
889 mobj = re.search(
890 r'<iframe[^>]+?src=(["\'])(?P<url>https?://m\.mlb\.com/shared/video/embed/embed\.html\?.+?)\1',
891 webpage)
892 if mobj is not None:
893 return self.url_result(mobj.group('url'), 'MLB')
894
895 mobj = re.search(
896 r'<iframe[^>]+?src=(["\'])(?P<url>%s)\1' % CondeNastIE.EMBED_URL,
897 webpage)
898 if mobj is not None:
899 return self.url_result(self._proto_relative_url(mobj.group('url'), scheme='http:'), 'CondeNast')
900
901 def check_video(vurl):
902 vpath = compat_urlparse.urlparse(vurl).path
903 vext = determine_ext(vpath)
904 return '.' in vpath and vext not in ('swf', 'png', 'jpg', 'srt', 'sbv', 'sub', 'vtt', 'ttml')
905
906 def filter_video(urls):
907 return list(filter(check_video, urls))
908
909 # Start with something easy: JW Player in SWFObject
910 found = filter_video(re.findall(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage))
911 if not found:
912 # Look for gorilla-vid style embedding
913 found = filter_video(re.findall(r'''(?sx)
914 (?:
915 jw_plugins|
916 JWPlayerOptions|
917 jwplayer\s*\(\s*["'][^'"]+["']\s*\)\s*\.setup
918 )
919 .*?file\s*:\s*["\'](.*?)["\']''', webpage))
920 if not found:
921 # Broaden the search a little bit
922 found = filter_video(re.findall(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage))
923 if not found:
924 # Broaden the findall a little bit: JWPlayer JS loader
925 found = filter_video(re.findall(
926 r'[^A-Za-z0-9]?file["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage))
927 if not found:
928 # Flow player
929 found = filter_video(re.findall(r'''(?xs)
930 flowplayer\("[^"]+",\s*
931 \{[^}]+?\}\s*,
932 \s*{[^}]+? ["']?clip["']?\s*:\s*\{\s*
933 ["']?url["']?\s*:\s*["']([^"']+)["']
934 ''', webpage))
935 if not found:
936 # Try to find twitter cards info
937 found = filter_video(re.findall(
938 r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage))
939 if not found:
940 # We look for Open Graph info:
941 # We have to match any number spaces between elements, some sites try to align them (eg.: statigr.am)
942 m_video_type = re.findall(r'<meta.*?property="og:video:type".*?content="video/(.*?)"', webpage)
943 # We only look in og:video if the MIME type is a video, don't try if it's a Flash player:
944 if m_video_type is not None:
945 found = filter_video(re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage))
946 if not found:
947 # HTML5 video
948 found = re.findall(r'(?s)<video[^<]*(?:>.*?<source[^>]*)?\s+src="([^"]+)"', webpage)
949 if not found:
950 found = re.search(
951 r'(?i)<meta\s+(?=(?:[a-z-]+="[^"]+"\s+)*http-equiv="refresh")'
952 r'(?:[a-z-]+="[^"]+"\s+)*?content="[0-9]{,2};url=\'?([^\'"]+)',
953 webpage)
954 if found:
955 new_url = found.group(1)
956 self.report_following_redirect(new_url)
957 return {
958 '_type': 'url',
959 'url': new_url,
960 }
961 if not found:
962 raise ExtractorError('Unsupported URL: %s' % url)
963
964 entries = []
965 for video_url in found:
966 video_url = compat_urlparse.urljoin(url, video_url)
967 video_id = compat_urllib_parse.unquote(os.path.basename(video_url))
968
969 # Sometimes, jwplayer extraction will result in a YouTube URL
970 if YoutubeIE.suitable(video_url):
971 entries.append(self.url_result(video_url, 'Youtube'))
972 continue
973
974 # here's a fun little line of code for you:
975 video_id = os.path.splitext(video_id)[0]
976
977 entries.append({
978 'id': video_id,
979 'url': video_url,
980 'uploader': video_uploader,
981 'title': video_title,
982 'age_limit': age_limit,
983 })
984
985 if len(entries) == 1:
986 return entries[0]
987 else:
988 for num, e in enumerate(entries, start=1):
989 e['title'] = '%s (%d)' % (e['title'], num)
990 return {
991 '_type': 'playlist',
992 'entries': entries,
993 }
994
Backup clone of https://github.com/ytdl-org unavailable since RIAA fucking raid...