]> Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/generic.py
c7a824c29b15187becb0bb9758ee94348f2309da
[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 ..compat import (
11 compat_urllib_parse,
12 compat_urlparse,
13 compat_xml_parse_error,
14 )
15 from ..utils import (
16 determine_ext,
17 ExtractorError,
18 float_or_none,
19 HEADRequest,
20 orderedSet,
21 parse_xml,
22 smuggle_url,
23 unescapeHTML,
24 unified_strdate,
25 unsmuggle_url,
26 url_basename,
27 )
28 from .brightcove import BrightcoveIE
29 from .ooyala import OoyalaIE
30 from .rutv import RUTVIE
31 from .smotri import SmotriIE
32 from .condenast import CondeNastIE
33
34
35 class GenericIE(InfoExtractor):
36 IE_DESC = 'Generic downloader that works on some sites'
37 _VALID_URL = r'.*'
38 IE_NAME = 'generic'
39 _TESTS = [
40 {
41 'url': 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
42 'md5': '85b90ccc9d73b4acd9138d3af4c27f89',
43 'info_dict': {
44 'id': '13601338388002',
45 'ext': 'mp4',
46 'uploader': 'www.hodiho.fr',
47 'title': 'R\u00e9gis plante sa Jeep',
48 }
49 },
50 # bandcamp page with custom domain
51 {
52 'add_ie': ['Bandcamp'],
53 'url': 'http://bronyrock.com/track/the-pony-mash',
54 'info_dict': {
55 'id': '3235767654',
56 'ext': 'mp3',
57 'title': 'The Pony Mash',
58 'uploader': 'M_Pallante',
59 },
60 'skip': 'There is a limit of 200 free downloads / month for the test song',
61 },
62 # embedded brightcove video
63 # it also tests brightcove videos that need to set the 'Referer' in the
64 # http requests
65 {
66 'add_ie': ['Brightcove'],
67 'url': 'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/',
68 'info_dict': {
69 'id': '2765128793001',
70 'ext': 'mp4',
71 'title': 'Le cours de bourse : l’analyse technique',
72 'description': 'md5:7e9ad046e968cb2d1114004aba466fd9',
73 'uploader': 'BFM BUSINESS',
74 },
75 'params': {
76 'skip_download': True,
77 },
78 },
79 {
80 # https://github.com/rg3/youtube-dl/issues/2253
81 'url': 'http://bcove.me/i6nfkrc3',
82 'md5': '0ba9446db037002366bab3b3eb30c88c',
83 'info_dict': {
84 'id': '3101154703001',
85 'ext': 'mp4',
86 'title': 'Still no power',
87 'uploader': 'thestar.com',
88 '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.',
89 },
90 'add_ie': ['Brightcove'],
91 },
92 {
93 'url': 'http://www.championat.com/video/football/v/87/87499.html',
94 'md5': 'fb973ecf6e4a78a67453647444222983',
95 'info_dict': {
96 'id': '3414141473001',
97 'ext': 'mp4',
98 'title': 'Видео. Удаление Дзагоева (ЦСКА)',
99 'description': 'Онлайн-трансляция матча ЦСКА - "Волга"',
100 'uploader': 'Championat',
101 },
102 },
103 {
104 # https://github.com/rg3/youtube-dl/issues/3541
105 'add_ie': ['Brightcove'],
106 'url': 'http://www.kijk.nl/sbs6/leermijvrouwenkennen/videos/jqMiXKAYan2S/aflevering-1',
107 'info_dict': {
108 'id': '3866516442001',
109 'ext': 'mp4',
110 'title': 'Leer mij vrouwen kennen: Aflevering 1',
111 'description': 'Leer mij vrouwen kennen: Aflevering 1',
112 'uploader': 'SBS Broadcasting',
113 },
114 'skip': 'Restricted to Netherlands',
115 'params': {
116 'skip_download': True, # m3u8 download
117 },
118 },
119 # Direct link to a video
120 {
121 'url': 'http://media.w3.org/2010/05/sintel/trailer.mp4',
122 'md5': '67d406c2bcb6af27fa886f31aa934bbe',
123 'info_dict': {
124 'id': 'trailer',
125 'ext': 'mp4',
126 'title': 'trailer',
127 'upload_date': '20100513',
128 }
129 },
130 # ooyala video
131 {
132 'url': 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219',
133 'md5': '5644c6ca5d5782c1d0d350dad9bd840c',
134 'info_dict': {
135 'id': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ',
136 'ext': 'mp4',
137 'title': '2cc213299525360.mov', # that's what we get
138 },
139 },
140 # google redirect
141 {
142 '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',
143 'info_dict': {
144 'id': 'cmQHVoWB5FY',
145 'ext': 'mp4',
146 'upload_date': '20130224',
147 'uploader_id': 'TheVerge',
148 '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.',
149 'uploader': 'The Verge',
150 'title': 'First Firefox OS phones side-by-side',
151 },
152 'params': {
153 'skip_download': False,
154 }
155 },
156 # embed.ly video
157 {
158 'url': 'http://www.tested.com/science/weird/460206-tested-grinding-coffee-2000-frames-second/',
159 'info_dict': {
160 'id': '9ODmcdjQcHQ',
161 'ext': 'mp4',
162 'title': 'Tested: Grinding Coffee at 2000 Frames Per Second',
163 'upload_date': '20140225',
164 'description': 'md5:06a40fbf30b220468f1e0957c0f558ff',
165 'uploader': 'Tested',
166 'uploader_id': 'testedcom',
167 },
168 # No need to test YoutubeIE here
169 'params': {
170 'skip_download': True,
171 },
172 },
173 # funnyordie embed
174 {
175 'url': 'http://www.theguardian.com/world/2014/mar/11/obama-zach-galifianakis-between-two-ferns',
176 'info_dict': {
177 'id': '18e820ec3f',
178 'ext': 'mp4',
179 'title': 'Between Two Ferns with Zach Galifianakis: President Barack Obama',
180 'description': 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.',
181 },
182 },
183 # RUTV embed
184 {
185 'url': 'http://www.rg.ru/2014/03/15/reg-dfo/anklav-anons.html',
186 'info_dict': {
187 'id': '776940',
188 'ext': 'mp4',
189 'title': 'Охотское море стало целиком российским',
190 'description': 'md5:5ed62483b14663e2a95ebbe115eb8f43',
191 },
192 'params': {
193 # m3u8 download
194 'skip_download': True,
195 },
196 },
197 # Embedded TED video
198 {
199 'url': 'http://en.support.wordpress.com/videos/ted-talks/',
200 'md5': '65fdff94098e4a607385a60c5177c638',
201 'info_dict': {
202 'id': '1969',
203 'ext': 'mp4',
204 'title': 'Hidden miracles of the natural world',
205 'uploader': 'Louie Schwartzberg',
206 'description': 'md5:8145d19d320ff3e52f28401f4c4283b9',
207 }
208 },
209 # Embeded Ustream video
210 {
211 'url': 'http://www.american.edu/spa/pti/nsa-privacy-janus-2014.cfm',
212 'md5': '27b99cdb639c9b12a79bca876a073417',
213 'info_dict': {
214 'id': '45734260',
215 'ext': 'flv',
216 'uploader': 'AU SPA: The NSA and Privacy',
217 'title': 'NSA and Privacy Forum Debate featuring General Hayden and Barton Gellman'
218 }
219 },
220 # nowvideo embed hidden behind percent encoding
221 {
222 'url': 'http://www.waoanime.tv/the-super-dimension-fortress-macross-episode-1/',
223 'md5': '2baf4ddd70f697d94b1c18cf796d5107',
224 'info_dict': {
225 'id': '06e53103ca9aa',
226 'ext': 'flv',
227 'title': 'Macross Episode 001 Watch Macross Episode 001 onl',
228 'description': 'No description',
229 },
230 },
231 # arte embed
232 {
233 'url': 'http://www.tv-replay.fr/redirection/20-03-14/x-enius-arte-10753389.html',
234 'md5': '7653032cbb25bf6c80d80f217055fa43',
235 'info_dict': {
236 'id': '048195-004_PLUS7-F',
237 'ext': 'flv',
238 'title': 'X:enius',
239 'description': 'md5:d5fdf32ef6613cdbfd516ae658abf168',
240 'upload_date': '20140320',
241 },
242 'params': {
243 'skip_download': 'Requires rtmpdump'
244 }
245 },
246 # Condé Nast embed
247 {
248 'url': 'http://www.wired.com/2014/04/honda-asimo/',
249 'md5': 'ba0dfe966fa007657bd1443ee672db0f',
250 'info_dict': {
251 'id': '53501be369702d3275860000',
252 'ext': 'mp4',
253 'title': 'Honda’s New Asimo Robot Is More Human Than Ever',
254 }
255 },
256 # Dailymotion embed
257 {
258 'url': 'http://www.spi0n.com/zap-spi0n-com-n216/',
259 'md5': '441aeeb82eb72c422c7f14ec533999cd',
260 'info_dict': {
261 'id': 'k2mm4bCdJ6CQ2i7c8o2',
262 'ext': 'mp4',
263 'title': 'Le Zap de Spi0n n°216 - Zapping du Web',
264 'uploader': 'Spi0n',
265 },
266 'add_ie': ['Dailymotion'],
267 },
268 # YouTube embed
269 {
270 'url': 'http://www.badzine.de/ansicht/datum/2014/06/09/so-funktioniert-die-neue-englische-badminton-liga.html',
271 'info_dict': {
272 'id': 'FXRb4ykk4S0',
273 'ext': 'mp4',
274 'title': 'The NBL Auction 2014',
275 'uploader': 'BADMINTON England',
276 'uploader_id': 'BADMINTONEvents',
277 'upload_date': '20140603',
278 'description': 'md5:9ef128a69f1e262a700ed83edb163a73',
279 },
280 'add_ie': ['Youtube'],
281 'params': {
282 'skip_download': True,
283 }
284 },
285 # MTVSercices embed
286 {
287 'url': 'http://www.gametrailers.com/news-post/76093/north-america-europe-is-getting-that-mario-kart-8-mercedes-dlc-too',
288 'md5': '35727f82f58c76d996fc188f9755b0d5',
289 'info_dict': {
290 'id': '0306a69b-8adf-4fb5-aace-75f8e8cbfca9',
291 'ext': 'mp4',
292 'title': 'Review',
293 'description': 'Mario\'s life in the fast lane has never looked so good.',
294 },
295 },
296 # YouTube embed via <data-embed-url="">
297 {
298 'url': 'https://play.google.com/store/apps/details?id=com.gameloft.android.ANMP.GloftA8HM',
299 'info_dict': {
300 'id': '4vAffPZIT44',
301 'ext': 'mp4',
302 'title': 'Asphalt 8: Airborne - Update - Welcome to Dubai!',
303 'uploader': 'Gameloft',
304 'uploader_id': 'gameloft',
305 'upload_date': '20140828',
306 'description': 'md5:c80da9ed3d83ae6d1876c834de03e1c4',
307 },
308 'params': {
309 'skip_download': True,
310 }
311 },
312 # Camtasia studio
313 {
314 'url': 'http://www.ll.mit.edu/workshops/education/videocourses/antennas/lecture1/video/',
315 'playlist': [{
316 'md5': '0c5e352edabf715d762b0ad4e6d9ee67',
317 'info_dict': {
318 'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
319 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - video1',
320 'ext': 'flv',
321 'duration': 2235.90,
322 }
323 }, {
324 'md5': '10e4bb3aaca9fd630e273ff92d9f3c63',
325 'info_dict': {
326 'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final_PIP',
327 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - pip',
328 'ext': 'flv',
329 'duration': 2235.93,
330 }
331 }],
332 'info_dict': {
333 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
334 }
335 },
336 # Flowplayer
337 {
338 'url': 'http://www.handjobhub.com/video/busty-blonde-siri-tit-fuck-while-wank-6313.html',
339 'md5': '9d65602bf31c6e20014319c7d07fba27',
340 'info_dict': {
341 'id': '5123ea6d5e5a7',
342 'ext': 'mp4',
343 'age_limit': 18,
344 'uploader': 'www.handjobhub.com',
345 'title': 'Busty Blonde Siri Tit Fuck While Wank at HandjobHub.com',
346 }
347 },
348 # RSS feed
349 {
350 'url': 'http://phihag.de/2014/youtube-dl/rss2.xml',
351 'info_dict': {
352 'id': 'http://phihag.de/2014/youtube-dl/rss2.xml',
353 'title': 'Zero Punctuation',
354 'description': 're:'
355 },
356 'playlist_mincount': 11,
357 },
358 # Multiple brightcove videos
359 # https://github.com/rg3/youtube-dl/issues/2283
360 {
361 'url': 'http://www.newyorker.com/online/blogs/newsdesk/2014/01/always-never-nuclear-command-and-control.html',
362 'info_dict': {
363 'id': 'always-never',
364 'title': 'Always / Never - The New Yorker',
365 },
366 'playlist_count': 3,
367 'params': {
368 'extract_flat': False,
369 'skip_download': True,
370 }
371 },
372 # MLB embed
373 {
374 'url': 'http://umpire-empire.com/index.php/topic/58125-laz-decides-no-thats-low/',
375 'md5': '96f09a37e44da40dd083e12d9a683327',
376 'info_dict': {
377 'id': '33322633',
378 'ext': 'mp4',
379 'title': 'Ump changes call to ball',
380 'description': 'md5:71c11215384298a172a6dcb4c2e20685',
381 'duration': 48,
382 'timestamp': 1401537900,
383 'upload_date': '20140531',
384 'thumbnail': 're:^https?://.*\.jpg$',
385 },
386 },
387 # Wistia embed
388 {
389 'url': 'http://education-portal.com/academy/lesson/north-american-exploration-failed-colonies-of-spain-france-england.html#lesson',
390 'md5': '8788b683c777a5cf25621eaf286d0c23',
391 'info_dict': {
392 'id': '1cfaf6b7ea',
393 'ext': 'mov',
394 'title': 'md5:51364a8d3d009997ba99656004b5e20d',
395 'duration': 643.0,
396 'filesize': 182808282,
397 'uploader': 'education-portal.com',
398 },
399 },
400 {
401 'url': 'http://thoughtworks.wistia.com/medias/uxjb0lwrcz',
402 'md5': 'baf49c2baa8a7de5f3fc145a8506dcd4',
403 'info_dict': {
404 'id': 'uxjb0lwrcz',
405 'ext': 'mp4',
406 'title': 'Conversation about Hexagonal Rails Part 1 - ThoughtWorks',
407 'duration': 1715.0,
408 'uploader': 'thoughtworks.wistia.com',
409 },
410 },
411 # Direct download with broken HEAD
412 {
413 'url': 'http://ai-radio.org:8000/radio.opus',
414 'info_dict': {
415 'id': 'radio',
416 'ext': 'opus',
417 'title': 'radio',
418 },
419 'params': {
420 'skip_download': True, # infinite live stream
421 },
422 'expected_warnings': [
423 r'501.*Not Implemented'
424 ],
425 },
426 # Soundcloud embed
427 {
428 'url': 'http://nakedsecurity.sophos.com/2014/10/29/sscc-171-are-you-sure-that-1234-is-a-bad-password-podcast/',
429 'info_dict': {
430 'id': '174391317',
431 'ext': 'mp3',
432 'description': 'md5:ff867d6b555488ad3c52572bb33d432c',
433 'uploader': 'Sophos Security',
434 'title': 'Chet Chat 171 - Oct 29, 2014',
435 'upload_date': '20141029',
436 }
437 },
438 # Livestream embed
439 {
440 'url': 'http://www.esa.int/Our_Activities/Space_Science/Rosetta/Philae_comet_touch-down_webcast',
441 'info_dict': {
442 'id': '67864563',
443 'ext': 'flv',
444 'upload_date': '20141112',
445 'title': 'Rosetta #CometLanding webcast HL 10',
446 }
447 },
448 ]
449
450 def report_following_redirect(self, new_url):
451 """Report information extraction."""
452 self._downloader.to_screen('[redirect] Following redirect to %s' % new_url)
453
454 def _extract_rss(self, url, video_id, doc):
455 playlist_title = doc.find('./channel/title').text
456 playlist_desc_el = doc.find('./channel/description')
457 playlist_desc = None if playlist_desc_el is None else playlist_desc_el.text
458
459 entries = [{
460 '_type': 'url',
461 'url': e.find('link').text,
462 'title': e.find('title').text,
463 } for e in doc.findall('./channel/item')]
464
465 return {
466 '_type': 'playlist',
467 'id': url,
468 'title': playlist_title,
469 'description': playlist_desc,
470 'entries': entries,
471 }
472
473 def _extract_camtasia(self, url, video_id, webpage):
474 """ Returns None if no camtasia video can be found. """
475
476 camtasia_cfg = self._search_regex(
477 r'fo\.addVariable\(\s*"csConfigFile",\s*"([^"]+)"\s*\);',
478 webpage, 'camtasia configuration file', default=None)
479 if camtasia_cfg is None:
480 return None
481
482 title = self._html_search_meta('DC.title', webpage, fatal=True)
483
484 camtasia_url = compat_urlparse.urljoin(url, camtasia_cfg)
485 camtasia_cfg = self._download_xml(
486 camtasia_url, video_id,
487 note='Downloading camtasia configuration',
488 errnote='Failed to download camtasia configuration')
489 fileset_node = camtasia_cfg.find('./playlist/array/fileset')
490
491 entries = []
492 for n in fileset_node.getchildren():
493 url_n = n.find('./uri')
494 if url_n is None:
495 continue
496
497 entries.append({
498 'id': os.path.splitext(url_n.text.rpartition('/')[2])[0],
499 'title': '%s - %s' % (title, n.tag),
500 'url': compat_urlparse.urljoin(url, url_n.text),
501 'duration': float_or_none(n.find('./duration').text),
502 })
503
504 return {
505 '_type': 'playlist',
506 'entries': entries,
507 'title': title,
508 }
509
510 def _real_extract(self, url):
511 if url.startswith('//'):
512 return {
513 '_type': 'url',
514 'url': self.http_scheme() + url,
515 }
516
517 parsed_url = compat_urlparse.urlparse(url)
518 if not parsed_url.scheme:
519 default_search = self._downloader.params.get('default_search')
520 if default_search is None:
521 default_search = 'fixup_error'
522
523 if default_search in ('auto', 'auto_warning', 'fixup_error'):
524 if '/' in url:
525 self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http')
526 return self.url_result('http://' + url)
527 elif default_search != 'fixup_error':
528 if default_search == 'auto_warning':
529 if re.match(r'^(?:url|URL)$', url):
530 raise ExtractorError(
531 'Invalid URL: %r . Call youtube-dl like this: youtube-dl -v "https://www.youtube.com/watch?v=BaW_jenozKc" ' % url,
532 expected=True)
533 else:
534 self._downloader.report_warning(
535 'Falling back to youtube search for %s . Set --default-search "auto" to suppress this warning.' % url)
536 return self.url_result('ytsearch:' + url)
537
538 if default_search in ('error', 'fixup_error'):
539 raise ExtractorError(
540 ('%r is not a valid URL. '
541 'Set --default-search "ytsearch" (or run youtube-dl "ytsearch:%s" ) to search YouTube'
542 ) % (url, url), expected=True)
543 else:
544 if ':' not in default_search:
545 default_search += ':'
546 return self.url_result(default_search + url)
547
548 url, smuggled_data = unsmuggle_url(url)
549 force_videoid = None
550 is_intentional = smuggled_data and smuggled_data.get('to_generic')
551 if smuggled_data and 'force_videoid' in smuggled_data:
552 force_videoid = smuggled_data['force_videoid']
553 video_id = force_videoid
554 else:
555 video_id = os.path.splitext(url.rstrip('/').split('/')[-1])[0]
556
557 self.to_screen('%s: Requesting header' % video_id)
558
559 head_req = HEADRequest(url)
560 head_response = self._request_webpage(
561 head_req, video_id,
562 note=False, errnote='Could not send HEAD request to %s' % url,
563 fatal=False)
564
565 if head_response is not False:
566 # Check for redirect
567 new_url = head_response.geturl()
568 if url != new_url:
569 self.report_following_redirect(new_url)
570 if force_videoid:
571 new_url = smuggle_url(
572 new_url, {'force_videoid': force_videoid})
573 return self.url_result(new_url)
574
575 full_response = None
576 if head_response is False:
577 full_response = self._request_webpage(url, video_id)
578 head_response = full_response
579
580 # Check for direct link to a video
581 content_type = head_response.headers.get('Content-Type', '')
582 m = re.match(r'^(?P<type>audio|video|application(?=/ogg$))/(?P<format_id>.+)$', content_type)
583 if m:
584 upload_date = unified_strdate(
585 head_response.headers.get('Last-Modified'))
586 return {
587 'id': video_id,
588 'title': os.path.splitext(url_basename(url))[0],
589 'direct': True,
590 'formats': [{
591 'format_id': m.group('format_id'),
592 'url': url,
593 'vcodec': 'none' if m.group('type') == 'audio' else None
594 }],
595 'upload_date': upload_date,
596 }
597
598 if not self._downloader.params.get('test', False) and not is_intentional:
599 self._downloader.report_warning('Falling back on generic information extractor.')
600
601 if full_response:
602 webpage = self._webpage_read_content(full_response, url, video_id)
603 else:
604 webpage = self._download_webpage(url, video_id)
605 self.report_extraction(video_id)
606
607 # Is it an RSS feed?
608 try:
609 doc = parse_xml(webpage)
610 if doc.tag == 'rss':
611 return self._extract_rss(url, video_id, doc)
612 except compat_xml_parse_error:
613 pass
614
615 # Is it a Camtasia project?
616 camtasia_res = self._extract_camtasia(url, video_id, webpage)
617 if camtasia_res is not None:
618 return camtasia_res
619
620 # Sometimes embedded video player is hidden behind percent encoding
621 # (e.g. https://github.com/rg3/youtube-dl/issues/2448)
622 # Unescaping the whole page allows to handle those cases in a generic way
623 webpage = compat_urllib_parse.unquote(webpage)
624
625 # it's tempting to parse this further, but you would
626 # have to take into account all the variations like
627 # Video Title - Site Name
628 # Site Name | Video Title
629 # Video Title - Tagline | Site Name
630 # and so on and so forth; it's just not practical
631 video_title = self._html_search_regex(
632 r'(?s)<title>(.*?)</title>', webpage, 'video title',
633 default='video')
634
635 # Try to detect age limit automatically
636 age_limit = self._rta_search(webpage)
637 # And then there are the jokers who advertise that they use RTA,
638 # but actually don't.
639 AGE_LIMIT_MARKERS = [
640 r'Proudly Labeled <a href="http://www.rtalabel.org/" title="Restricted to Adults">RTA</a>',
641 ]
642 if any(re.search(marker, webpage) for marker in AGE_LIMIT_MARKERS):
643 age_limit = 18
644
645 # video uploader is domain name
646 video_uploader = self._search_regex(
647 r'^(?:https?://)?([^/]*)/.*', url, 'video uploader')
648
649 # Helper method
650 def _playlist_from_matches(matches, getter, ie=None):
651 urlrs = orderedSet(
652 self.url_result(self._proto_relative_url(getter(m)), ie)
653 for m in matches)
654 return self.playlist_result(
655 urlrs, playlist_id=video_id, playlist_title=video_title)
656
657 # Look for BrightCove:
658 bc_urls = BrightcoveIE._extract_brightcove_urls(webpage)
659 if bc_urls:
660 self.to_screen('Brightcove video detected.')
661 entries = [{
662 '_type': 'url',
663 'url': smuggle_url(bc_url, {'Referer': url}),
664 'ie_key': 'Brightcove'
665 } for bc_url in bc_urls]
666
667 return {
668 '_type': 'playlist',
669 'title': video_title,
670 'id': video_id,
671 'entries': entries,
672 }
673
674 # Look for embedded (iframe) Vimeo player
675 mobj = re.search(
676 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//player\.vimeo\.com/video/.+?)\1', webpage)
677 if mobj:
678 player_url = unescapeHTML(mobj.group('url'))
679 surl = smuggle_url(player_url, {'Referer': url})
680 return self.url_result(surl)
681
682 # Look for embedded (swf embed) Vimeo player
683 mobj = re.search(
684 r'<embed[^>]+?src="((?:https?:)?//(?:www\.)?vimeo\.com/moogaloop\.swf.+?)"', webpage)
685 if mobj:
686 return self.url_result(mobj.group(1))
687
688 # Look for embedded YouTube player
689 matches = re.findall(r'''(?x)
690 (?:
691 <iframe[^>]+?src=|
692 data-video-url=|
693 <embed[^>]+?src=|
694 embedSWF\(?:\s*|
695 new\s+SWFObject\(
696 )
697 (["\'])
698 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
699 (?:embed|v|p)/.+?)
700 \1''', webpage)
701 if matches:
702 return _playlist_from_matches(
703 matches, lambda m: unescapeHTML(m[1]))
704
705 # Look for embedded Dailymotion player
706 matches = re.findall(
707 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/embed/video/.+?)\1', webpage)
708 if matches:
709 return _playlist_from_matches(
710 matches, lambda m: unescapeHTML(m[1]))
711
712 # Look for embedded Dailymotion playlist player (#3822)
713 m = re.search(
714 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.[a-z]{2,3}/widget/jukebox\?.+?)\1', webpage)
715 if m:
716 playlists = re.findall(
717 r'list\[\]=/playlist/([^/]+)/', unescapeHTML(m.group('url')))
718 if playlists:
719 return _playlist_from_matches(
720 playlists, lambda p: '//dailymotion.com/playlist/%s' % p)
721
722 # Look for embedded Wistia player
723 match = re.search(
724 r'<(?:meta[^>]+?content|iframe[^>]+?src)=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage)
725 if match:
726 embed_url = self._proto_relative_url(
727 unescapeHTML(match.group('url')))
728 return {
729 '_type': 'url_transparent',
730 'url': embed_url,
731 'ie_key': 'Wistia',
732 'uploader': video_uploader,
733 'title': video_title,
734 'id': video_id,
735 }
736
737 match = re.search(r'(?:id=["\']wistia_|data-wistia-?id=["\']|Wistia\.embed\(["\'])(?P<id>[^"\']+)', webpage)
738 if match:
739 return {
740 '_type': 'url_transparent',
741 'url': 'http://fast.wistia.net/embed/iframe/{0:}'.format(match.group('id')),
742 'ie_key': 'Wistia',
743 'uploader': video_uploader,
744 'title': video_title,
745 'id': match.group('id')
746 }
747
748 # Look for embedded blip.tv player
749 mobj = re.search(r'<meta\s[^>]*https?://api\.blip\.tv/\w+/redirect/\w+/(\d+)', webpage)
750 if mobj:
751 return self.url_result('http://blip.tv/a/a-'+mobj.group(1), 'BlipTV')
752 mobj = re.search(r'<(?:iframe|embed|object)\s[^>]*(https?://(?:\w+\.)?blip\.tv/(?:play/|api\.swf#)[a-zA-Z0-9_]+)', webpage)
753 if mobj:
754 return self.url_result(mobj.group(1), 'BlipTV')
755
756 # Look for embedded condenast player
757 matches = re.findall(
758 r'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="(https?://player\.cnevids\.com/embed/[^"]+")',
759 webpage)
760 if matches:
761 return {
762 '_type': 'playlist',
763 'entries': [{
764 '_type': 'url',
765 'ie_key': 'CondeNast',
766 'url': ma,
767 } for ma in matches],
768 'title': video_title,
769 'id': video_id,
770 }
771
772 # Look for Bandcamp pages with custom domain
773 mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage)
774 if mobj is not None:
775 burl = unescapeHTML(mobj.group(1))
776 # Don't set the extractor because it can be a track url or an album
777 return self.url_result(burl)
778
779 # Look for embedded Vevo player
780 mobj = re.search(
781 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:cache\.)?vevo\.com/.+?)\1', webpage)
782 if mobj is not None:
783 return self.url_result(mobj.group('url'))
784
785 # Look for Ooyala videos
786 mobj = (re.search(r'player.ooyala.com/[^"?]+\?[^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or
787 re.search(r'OO.Player.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage))
788 if mobj is not None:
789 return OoyalaIE._build_url_result(mobj.group('ec'))
790
791 # Look for Aparat videos
792 mobj = re.search(r'<iframe .*?src="(http://www\.aparat\.com/video/[^"]+)"', webpage)
793 if mobj is not None:
794 return self.url_result(mobj.group(1), 'Aparat')
795
796 # Look for MPORA videos
797 mobj = re.search(r'<iframe .*?src="(http://mpora\.(?:com|de)/videos/[^"]+)"', webpage)
798 if mobj is not None:
799 return self.url_result(mobj.group(1), 'Mpora')
800
801 # Look for embedded NovaMov-based player
802 mobj = re.search(
803 r'''(?x)<(?:pagespeed_)?iframe[^>]+?src=(["\'])
804 (?P<url>http://(?:(?:embed|www)\.)?
805 (?:novamov\.com|
806 nowvideo\.(?:ch|sx|eu|at|ag|co)|
807 videoweed\.(?:es|com)|
808 movshare\.(?:net|sx|ag)|
809 divxstage\.(?:eu|net|ch|co|at|ag))
810 /embed\.php.+?)\1''', webpage)
811 if mobj is not None:
812 return self.url_result(mobj.group('url'))
813
814 # Look for embedded Facebook player
815 mobj = re.search(
816 r'<iframe[^>]+?src=(["\'])(?P<url>https://www\.facebook\.com/video/embed.+?)\1', webpage)
817 if mobj is not None:
818 return self.url_result(mobj.group('url'), 'Facebook')
819
820 # Look for embedded VK player
821 mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://vk\.com/video_ext\.php.+?)\1', webpage)
822 if mobj is not None:
823 return self.url_result(mobj.group('url'), 'VK')
824
825 # Look for embedded ivi player
826 mobj = re.search(r'<embed[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?ivi\.ru/video/player.+?)\1', webpage)
827 if mobj is not None:
828 return self.url_result(mobj.group('url'), 'Ivi')
829
830 # Look for embedded Huffington Post player
831 mobj = re.search(
832 r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed\.live\.huffingtonpost\.com/.+?)\1', webpage)
833 if mobj is not None:
834 return self.url_result(mobj.group('url'), 'HuffPost')
835
836 # Look for embed.ly
837 mobj = re.search(r'class=["\']embedly-card["\'][^>]href=["\'](?P<url>[^"\']+)', webpage)
838 if mobj is not None:
839 return self.url_result(mobj.group('url'))
840 mobj = re.search(r'class=["\']embedly-embed["\'][^>]src=["\'][^"\']*url=(?P<url>[^&]+)', webpage)
841 if mobj is not None:
842 return self.url_result(compat_urllib_parse.unquote(mobj.group('url')))
843
844 # Look for funnyordie embed
845 matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage)
846 if matches:
847 return _playlist_from_matches(
848 matches, getter=unescapeHTML, ie='FunnyOrDie')
849
850 # Look for embedded RUTV player
851 rutv_url = RUTVIE._extract_url(webpage)
852 if rutv_url:
853 return self.url_result(rutv_url, 'RUTV')
854
855 # Look for embedded TED player
856 mobj = re.search(
857 r'<iframe[^>]+?src=(["\'])(?P<url>http://embed\.ted\.com/.+?)\1', webpage)
858 if mobj is not None:
859 return self.url_result(mobj.group('url'), 'TED')
860
861 # Look for embedded Ustream videos
862 mobj = re.search(
863 r'<iframe[^>]+?src=(["\'])(?P<url>http://www\.ustream\.tv/embed/.+?)\1', webpage)
864 if mobj is not None:
865 return self.url_result(mobj.group('url'), 'Ustream')
866
867 # Look for embedded arte.tv player
868 mobj = re.search(
869 r'<script [^>]*?src="(?P<url>http://www\.arte\.tv/playerv2/embed[^"]+)"',
870 webpage)
871 if mobj is not None:
872 return self.url_result(mobj.group('url'), 'ArteTVEmbed')
873
874 # Look for embedded smotri.com player
875 smotri_url = SmotriIE._extract_url(webpage)
876 if smotri_url:
877 return self.url_result(smotri_url, 'Smotri')
878
879 # Look for embeded soundcloud player
880 mobj = re.search(
881 r'<iframe\s+(?:[a-zA-Z0-9_-]+="[^"]+"\s+)*src="(?P<url>https?://(?:w\.)?soundcloud\.com/player[^"]+)"',
882 webpage)
883 if mobj is not None:
884 url = unescapeHTML(mobj.group('url'))
885 return self.url_result(url)
886
887 # Look for embedded vulture.com player
888 mobj = re.search(
889 r'<iframe src="(?P<url>https?://video\.vulture\.com/[^"]+)"',
890 webpage)
891 if mobj is not None:
892 url = unescapeHTML(mobj.group('url'))
893 return self.url_result(url, ie='Vulture')
894
895 # Look for embedded mtvservices player
896 mobj = re.search(
897 r'<iframe src="(?P<url>https?://media\.mtvnservices\.com/embed/[^"]+)"',
898 webpage)
899 if mobj is not None:
900 url = unescapeHTML(mobj.group('url'))
901 return self.url_result(url, ie='MTVServicesEmbedded')
902
903 # Look for embedded yahoo player
904 mobj = re.search(
905 r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:screen|movies)\.yahoo\.com/.+?\.html\?format=embed)\1',
906 webpage)
907 if mobj is not None:
908 return self.url_result(mobj.group('url'), 'Yahoo')
909
910 # Look for embedded sbs.com.au player
911 mobj = re.search(
912 r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:www\.)sbs\.com\.au/ondemand/video/single/.+?)\1',
913 webpage)
914 if mobj is not None:
915 return self.url_result(mobj.group('url'), 'SBS')
916
917 mobj = re.search(
918 r'<iframe[^>]+?src=(["\'])(?P<url>https?://m(?:lb)?\.mlb\.com/shared/video/embed/embed\.html\?.+?)\1',
919 webpage)
920 if mobj is not None:
921 return self.url_result(mobj.group('url'), 'MLB')
922
923 mobj = re.search(
924 r'<iframe[^>]+?src=(["\'])(?P<url>%s)\1' % CondeNastIE.EMBED_URL,
925 webpage)
926 if mobj is not None:
927 return self.url_result(self._proto_relative_url(mobj.group('url'), scheme='http:'), 'CondeNast')
928
929 mobj = re.search(
930 r'<iframe[^>]+src="(?P<url>https?://new\.livestream\.com/[^"]+/player[^"]+)"',
931 webpage)
932 if mobj is not None:
933 return self.url_result(mobj.group('url'), 'Livestream')
934
935 def check_video(vurl):
936 vpath = compat_urlparse.urlparse(vurl).path
937 vext = determine_ext(vpath)
938 return '.' in vpath and vext not in ('swf', 'png', 'jpg', 'srt', 'sbv', 'sub', 'vtt', 'ttml')
939
940 def filter_video(urls):
941 return list(filter(check_video, urls))
942
943 # Start with something easy: JW Player in SWFObject
944 found = filter_video(re.findall(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage))
945 if not found:
946 # Look for gorilla-vid style embedding
947 found = filter_video(re.findall(r'''(?sx)
948 (?:
949 jw_plugins|
950 JWPlayerOptions|
951 jwplayer\s*\(\s*["'][^'"]+["']\s*\)\s*\.setup
952 )
953 .*?file\s*:\s*["\'](.*?)["\']''', webpage))
954 if not found:
955 # Broaden the search a little bit
956 found = filter_video(re.findall(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage))
957 if not found:
958 # Broaden the findall a little bit: JWPlayer JS loader
959 found = filter_video(re.findall(
960 r'[^A-Za-z0-9]?file["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage))
961 if not found:
962 # Flow player
963 found = filter_video(re.findall(r'''(?xs)
964 flowplayer\("[^"]+",\s*
965 \{[^}]+?\}\s*,
966 \s*{[^}]+? ["']?clip["']?\s*:\s*\{\s*
967 ["']?url["']?\s*:\s*["']([^"']+)["']
968 ''', webpage))
969 if not found:
970 # Try to find twitter cards info
971 found = filter_video(re.findall(
972 r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage))
973 if not found:
974 # We look for Open Graph info:
975 # We have to match any number spaces between elements, some sites try to align them (eg.: statigr.am)
976 m_video_type = re.findall(r'<meta.*?property="og:video:type".*?content="video/(.*?)"', webpage)
977 # We only look in og:video if the MIME type is a video, don't try if it's a Flash player:
978 if m_video_type is not None:
979 found = filter_video(re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage))
980 if not found:
981 # HTML5 video
982 found = re.findall(r'(?s)<video[^<]*(?:>.*?<source[^>]*)?\s+src=["\'](.*?)["\']', webpage)
983 if not found:
984 found = re.search(
985 r'(?i)<meta\s+(?=(?:[a-z-]+="[^"]+"\s+)*http-equiv="refresh")'
986 r'(?:[a-z-]+="[^"]+"\s+)*?content="[0-9]{,2};url=\'?([^\'"]+)',
987 webpage)
988 if found:
989 new_url = found.group(1)
990 self.report_following_redirect(new_url)
991 return {
992 '_type': 'url',
993 'url': new_url,
994 }
995 if not found:
996 raise ExtractorError('Unsupported URL: %s' % url)
997
998 entries = []
999 for video_url in found:
1000 video_url = compat_urlparse.urljoin(url, video_url)
1001 video_id = compat_urllib_parse.unquote(os.path.basename(video_url))
1002
1003 # Sometimes, jwplayer extraction will result in a YouTube URL
1004 if YoutubeIE.suitable(video_url):
1005 entries.append(self.url_result(video_url, 'Youtube'))
1006 continue
1007
1008 # here's a fun little line of code for you:
1009 video_id = os.path.splitext(video_id)[0]
1010
1011 entries.append({
1012 'id': video_id,
1013 'url': video_url,
1014 'uploader': video_uploader,
1015 'title': video_title,
1016 'age_limit': age_limit,
1017 })
1018
1019 if len(entries) == 1:
1020 return entries[0]
1021 else:
1022 for num, e in enumerate(entries, start=1):
1023 e['title'] = '%s (%d)' % (e['title'], num)
1024 return {
1025 '_type': 'playlist',
1026 'entries': entries,
1027 }
1028