]> Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/generic.py
27e2bc3001c27750378cd790763d86b38442ffa7
[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 is_html,
21 orderedSet,
22 parse_xml,
23 smuggle_url,
24 unescapeHTML,
25 unified_strdate,
26 unsmuggle_url,
27 UnsupportedError,
28 url_basename,
29 )
30 from .brightcove import BrightcoveIE
31 from .ooyala import OoyalaIE
32 from .rutv import RUTVIE
33 from .smotri import SmotriIE
34 from .condenast import CondeNastIE
35
36
37 class GenericIE(InfoExtractor):
38 IE_DESC = 'Generic downloader that works on some sites'
39 _VALID_URL = r'.*'
40 IE_NAME = 'generic'
41 _TESTS = [
42 {
43 'url': 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
44 'md5': '85b90ccc9d73b4acd9138d3af4c27f89',
45 'info_dict': {
46 'id': '13601338388002',
47 'ext': 'mp4',
48 'uploader': 'www.hodiho.fr',
49 'title': 'R\u00e9gis plante sa Jeep',
50 }
51 },
52 # bandcamp page with custom domain
53 {
54 'add_ie': ['Bandcamp'],
55 'url': 'http://bronyrock.com/track/the-pony-mash',
56 'info_dict': {
57 'id': '3235767654',
58 'ext': 'mp3',
59 'title': 'The Pony Mash',
60 'uploader': 'M_Pallante',
61 },
62 'skip': 'There is a limit of 200 free downloads / month for the test song',
63 },
64 # embedded brightcove video
65 # it also tests brightcove videos that need to set the 'Referer' in the
66 # http requests
67 {
68 'add_ie': ['Brightcove'],
69 'url': 'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/',
70 'info_dict': {
71 'id': '2765128793001',
72 'ext': 'mp4',
73 'title': 'Le cours de bourse : l’analyse technique',
74 'description': 'md5:7e9ad046e968cb2d1114004aba466fd9',
75 'uploader': 'BFM BUSINESS',
76 },
77 'params': {
78 'skip_download': True,
79 },
80 },
81 {
82 # https://github.com/rg3/youtube-dl/issues/2253
83 'url': 'http://bcove.me/i6nfkrc3',
84 'md5': '0ba9446db037002366bab3b3eb30c88c',
85 'info_dict': {
86 'id': '3101154703001',
87 'ext': 'mp4',
88 'title': 'Still no power',
89 'uploader': 'thestar.com',
90 '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.',
91 },
92 'add_ie': ['Brightcove'],
93 },
94 {
95 'url': 'http://www.championat.com/video/football/v/87/87499.html',
96 'md5': 'fb973ecf6e4a78a67453647444222983',
97 'info_dict': {
98 'id': '3414141473001',
99 'ext': 'mp4',
100 'title': 'Видео. Удаление Дзагоева (ЦСКА)',
101 'description': 'Онлайн-трансляция матча ЦСКА - "Волга"',
102 'uploader': 'Championat',
103 },
104 },
105 {
106 # https://github.com/rg3/youtube-dl/issues/3541
107 'add_ie': ['Brightcove'],
108 'url': 'http://www.kijk.nl/sbs6/leermijvrouwenkennen/videos/jqMiXKAYan2S/aflevering-1',
109 'info_dict': {
110 'id': '3866516442001',
111 'ext': 'mp4',
112 'title': 'Leer mij vrouwen kennen: Aflevering 1',
113 'description': 'Leer mij vrouwen kennen: Aflevering 1',
114 'uploader': 'SBS Broadcasting',
115 },
116 'skip': 'Restricted to Netherlands',
117 'params': {
118 'skip_download': True, # m3u8 download
119 },
120 },
121 # Direct link to a video
122 {
123 'url': 'http://media.w3.org/2010/05/sintel/trailer.mp4',
124 'md5': '67d406c2bcb6af27fa886f31aa934bbe',
125 'info_dict': {
126 'id': 'trailer',
127 'ext': 'mp4',
128 'title': 'trailer',
129 'upload_date': '20100513',
130 }
131 },
132 # ooyala video
133 {
134 'url': 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219',
135 'md5': '166dd577b433b4d4ebfee10b0824d8ff',
136 'info_dict': {
137 'id': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ',
138 'ext': 'mp4',
139 'title': '2cc213299525360.mov', # that's what we get
140 },
141 'add_ie': ['Ooyala'],
142 },
143 # multiple ooyala embeds on SBN network websites
144 {
145 'url': 'http://www.sbnation.com/college-football-recruiting/2015/2/3/7970291/national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
146 'info_dict': {
147 'id': 'national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
148 'title': '25 lies you will tell yourself on National Signing Day - SBNation.com',
149 },
150 'playlist_mincount': 3,
151 'params': {
152 'skip_download': True,
153 },
154 'add_ie': ['Ooyala'],
155 },
156 # google redirect
157 {
158 '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',
159 'info_dict': {
160 'id': 'cmQHVoWB5FY',
161 'ext': 'mp4',
162 'upload_date': '20130224',
163 'uploader_id': 'TheVerge',
164 'description': 're:^Chris Ziegler takes a look at the\.*',
165 'uploader': 'The Verge',
166 'title': 'First Firefox OS phones side-by-side',
167 },
168 'params': {
169 'skip_download': False,
170 }
171 },
172 # embed.ly video
173 {
174 'url': 'http://www.tested.com/science/weird/460206-tested-grinding-coffee-2000-frames-second/',
175 'info_dict': {
176 'id': '9ODmcdjQcHQ',
177 'ext': 'mp4',
178 'title': 'Tested: Grinding Coffee at 2000 Frames Per Second',
179 'upload_date': '20140225',
180 'description': 'md5:06a40fbf30b220468f1e0957c0f558ff',
181 'uploader': 'Tested',
182 'uploader_id': 'testedcom',
183 },
184 # No need to test YoutubeIE here
185 'params': {
186 'skip_download': True,
187 },
188 },
189 # funnyordie embed
190 {
191 'url': 'http://www.theguardian.com/world/2014/mar/11/obama-zach-galifianakis-between-two-ferns',
192 'info_dict': {
193 'id': '18e820ec3f',
194 'ext': 'mp4',
195 'title': 'Between Two Ferns with Zach Galifianakis: President Barack Obama',
196 'description': 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.',
197 },
198 },
199 # BBC iPlayer embeds
200 {
201 'url': 'http://www.bbc.co.uk/blogs/adamcurtis/posts/BUGGER',
202 'info_dict': {
203 'title': 'BBC - Blogs - Adam Curtis - BUGGER',
204 },
205 'playlist_mincount': 18,
206 },
207 # RUTV embed
208 {
209 'url': 'http://www.rg.ru/2014/03/15/reg-dfo/anklav-anons.html',
210 'info_dict': {
211 'id': '776940',
212 'ext': 'mp4',
213 'title': 'Охотское море стало целиком российским',
214 'description': 'md5:5ed62483b14663e2a95ebbe115eb8f43',
215 },
216 'params': {
217 # m3u8 download
218 'skip_download': True,
219 },
220 },
221 # Embedded TED video
222 {
223 'url': 'http://en.support.wordpress.com/videos/ted-talks/',
224 'md5': '65fdff94098e4a607385a60c5177c638',
225 'info_dict': {
226 'id': '1969',
227 'ext': 'mp4',
228 'title': 'Hidden miracles of the natural world',
229 'uploader': 'Louie Schwartzberg',
230 'description': 'md5:8145d19d320ff3e52f28401f4c4283b9',
231 }
232 },
233 # Embeded Ustream video
234 {
235 'url': 'http://www.american.edu/spa/pti/nsa-privacy-janus-2014.cfm',
236 'md5': '27b99cdb639c9b12a79bca876a073417',
237 'info_dict': {
238 'id': '45734260',
239 'ext': 'flv',
240 'uploader': 'AU SPA: The NSA and Privacy',
241 'title': 'NSA and Privacy Forum Debate featuring General Hayden and Barton Gellman'
242 }
243 },
244 # nowvideo embed hidden behind percent encoding
245 {
246 'url': 'http://www.waoanime.tv/the-super-dimension-fortress-macross-episode-1/',
247 'md5': '2baf4ddd70f697d94b1c18cf796d5107',
248 'info_dict': {
249 'id': '06e53103ca9aa',
250 'ext': 'flv',
251 'title': 'Macross Episode 001 Watch Macross Episode 001 onl',
252 'description': 'No description',
253 },
254 },
255 # arte embed
256 {
257 'url': 'http://www.tv-replay.fr/redirection/20-03-14/x-enius-arte-10753389.html',
258 'md5': '7653032cbb25bf6c80d80f217055fa43',
259 'info_dict': {
260 'id': '048195-004_PLUS7-F',
261 'ext': 'flv',
262 'title': 'X:enius',
263 'description': 'md5:d5fdf32ef6613cdbfd516ae658abf168',
264 'upload_date': '20140320',
265 },
266 'params': {
267 'skip_download': 'Requires rtmpdump'
268 }
269 },
270 # Condé Nast embed
271 {
272 'url': 'http://www.wired.com/2014/04/honda-asimo/',
273 'md5': 'ba0dfe966fa007657bd1443ee672db0f',
274 'info_dict': {
275 'id': '53501be369702d3275860000',
276 'ext': 'mp4',
277 'title': 'Honda’s New Asimo Robot Is More Human Than Ever',
278 }
279 },
280 # Dailymotion embed
281 {
282 'url': 'http://www.spi0n.com/zap-spi0n-com-n216/',
283 'md5': '441aeeb82eb72c422c7f14ec533999cd',
284 'info_dict': {
285 'id': 'k2mm4bCdJ6CQ2i7c8o2',
286 'ext': 'mp4',
287 'title': 'Le Zap de Spi0n n°216 - Zapping du Web',
288 'uploader': 'Spi0n',
289 },
290 'add_ie': ['Dailymotion'],
291 },
292 # YouTube embed
293 {
294 'url': 'http://www.badzine.de/ansicht/datum/2014/06/09/so-funktioniert-die-neue-englische-badminton-liga.html',
295 'info_dict': {
296 'id': 'FXRb4ykk4S0',
297 'ext': 'mp4',
298 'title': 'The NBL Auction 2014',
299 'uploader': 'BADMINTON England',
300 'uploader_id': 'BADMINTONEvents',
301 'upload_date': '20140603',
302 'description': 'md5:9ef128a69f1e262a700ed83edb163a73',
303 },
304 'add_ie': ['Youtube'],
305 'params': {
306 'skip_download': True,
307 }
308 },
309 # MTVSercices embed
310 {
311 'url': 'http://www.gametrailers.com/news-post/76093/north-america-europe-is-getting-that-mario-kart-8-mercedes-dlc-too',
312 'md5': '35727f82f58c76d996fc188f9755b0d5',
313 'info_dict': {
314 'id': '0306a69b-8adf-4fb5-aace-75f8e8cbfca9',
315 'ext': 'mp4',
316 'title': 'Review',
317 'description': 'Mario\'s life in the fast lane has never looked so good.',
318 },
319 },
320 # YouTube embed via <data-embed-url="">
321 {
322 'url': 'https://play.google.com/store/apps/details?id=com.gameloft.android.ANMP.GloftA8HM',
323 'info_dict': {
324 'id': '4vAffPZIT44',
325 'ext': 'mp4',
326 'title': 'Asphalt 8: Airborne - Update - Welcome to Dubai!',
327 'uploader': 'Gameloft',
328 'uploader_id': 'gameloft',
329 'upload_date': '20140828',
330 'description': 'md5:c80da9ed3d83ae6d1876c834de03e1c4',
331 },
332 'params': {
333 'skip_download': True,
334 }
335 },
336 # Camtasia studio
337 {
338 'url': 'http://www.ll.mit.edu/workshops/education/videocourses/antennas/lecture1/video/',
339 'playlist': [{
340 'md5': '0c5e352edabf715d762b0ad4e6d9ee67',
341 'info_dict': {
342 'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
343 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - video1',
344 'ext': 'flv',
345 'duration': 2235.90,
346 }
347 }, {
348 'md5': '10e4bb3aaca9fd630e273ff92d9f3c63',
349 'info_dict': {
350 'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final_PIP',
351 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - pip',
352 'ext': 'flv',
353 'duration': 2235.93,
354 }
355 }],
356 'info_dict': {
357 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
358 }
359 },
360 # Flowplayer
361 {
362 'url': 'http://www.handjobhub.com/video/busty-blonde-siri-tit-fuck-while-wank-6313.html',
363 'md5': '9d65602bf31c6e20014319c7d07fba27',
364 'info_dict': {
365 'id': '5123ea6d5e5a7',
366 'ext': 'mp4',
367 'age_limit': 18,
368 'uploader': 'www.handjobhub.com',
369 'title': 'Busty Blonde Siri Tit Fuck While Wank at HandjobHub.com',
370 }
371 },
372 # RSS feed
373 {
374 'url': 'http://phihag.de/2014/youtube-dl/rss2.xml',
375 'info_dict': {
376 'id': 'http://phihag.de/2014/youtube-dl/rss2.xml',
377 'title': 'Zero Punctuation',
378 'description': 're:.*groundbreaking video review series.*'
379 },
380 'playlist_mincount': 11,
381 },
382 # Multiple brightcove videos
383 # https://github.com/rg3/youtube-dl/issues/2283
384 {
385 'url': 'http://www.newyorker.com/online/blogs/newsdesk/2014/01/always-never-nuclear-command-and-control.html',
386 'info_dict': {
387 'id': 'always-never',
388 'title': 'Always / Never - The New Yorker',
389 },
390 'playlist_count': 3,
391 'params': {
392 'extract_flat': False,
393 'skip_download': True,
394 }
395 },
396 # MLB embed
397 {
398 'url': 'http://umpire-empire.com/index.php/topic/58125-laz-decides-no-thats-low/',
399 'md5': '96f09a37e44da40dd083e12d9a683327',
400 'info_dict': {
401 'id': '33322633',
402 'ext': 'mp4',
403 'title': 'Ump changes call to ball',
404 'description': 'md5:71c11215384298a172a6dcb4c2e20685',
405 'duration': 48,
406 'timestamp': 1401537900,
407 'upload_date': '20140531',
408 'thumbnail': 're:^https?://.*\.jpg$',
409 },
410 },
411 # Wistia embed
412 {
413 'url': 'http://education-portal.com/academy/lesson/north-american-exploration-failed-colonies-of-spain-france-england.html#lesson',
414 'md5': '8788b683c777a5cf25621eaf286d0c23',
415 'info_dict': {
416 'id': '1cfaf6b7ea',
417 'ext': 'mov',
418 'title': 'md5:51364a8d3d009997ba99656004b5e20d',
419 'duration': 643.0,
420 'filesize': 182808282,
421 'uploader': 'education-portal.com',
422 },
423 },
424 {
425 'url': 'http://thoughtworks.wistia.com/medias/uxjb0lwrcz',
426 'md5': 'baf49c2baa8a7de5f3fc145a8506dcd4',
427 'info_dict': {
428 'id': 'uxjb0lwrcz',
429 'ext': 'mp4',
430 'title': 'Conversation about Hexagonal Rails Part 1 - ThoughtWorks',
431 'duration': 1715.0,
432 'uploader': 'thoughtworks.wistia.com',
433 },
434 },
435 # Direct download with broken HEAD
436 {
437 'url': 'http://ai-radio.org:8000/radio.opus',
438 'info_dict': {
439 'id': 'radio',
440 'ext': 'opus',
441 'title': 'radio',
442 },
443 'params': {
444 'skip_download': True, # infinite live stream
445 },
446 'expected_warnings': [
447 r'501.*Not Implemented'
448 ],
449 },
450 # Soundcloud embed
451 {
452 'url': 'http://nakedsecurity.sophos.com/2014/10/29/sscc-171-are-you-sure-that-1234-is-a-bad-password-podcast/',
453 'info_dict': {
454 'id': '174391317',
455 'ext': 'mp3',
456 'description': 'md5:ff867d6b555488ad3c52572bb33d432c',
457 'uploader': 'Sophos Security',
458 'title': 'Chet Chat 171 - Oct 29, 2014',
459 'upload_date': '20141029',
460 }
461 },
462 # Livestream embed
463 {
464 'url': 'http://www.esa.int/Our_Activities/Space_Science/Rosetta/Philae_comet_touch-down_webcast',
465 'info_dict': {
466 'id': '67864563',
467 'ext': 'flv',
468 'upload_date': '20141112',
469 'title': 'Rosetta #CometLanding webcast HL 10',
470 }
471 },
472 # LazyYT
473 {
474 'url': 'http://discourse.ubuntu.com/t/unity-8-desktop-mode-windows-on-mir/1986',
475 'info_dict': {
476 'id': '1986',
477 'title': 'Unity 8 desktop-mode windows on Mir! - Ubuntu Discourse',
478 },
479 'playlist_mincount': 2,
480 },
481 # Direct link with incorrect MIME type
482 {
483 'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
484 'md5': '4ccbebe5f36706d85221f204d7eb5913',
485 'info_dict': {
486 'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
487 'id': '5_Lennart_Poettering_-_Systemd',
488 'ext': 'webm',
489 'title': '5_Lennart_Poettering_-_Systemd',
490 'upload_date': '20141120',
491 },
492 'expected_warnings': [
493 'URL could be a direct video link, returning it as such.'
494 ]
495 },
496 # Cinchcast embed
497 {
498 'url': 'http://undergroundwellness.com/podcasts/306-5-steps-to-permanent-gut-healing/',
499 'info_dict': {
500 'id': '7141703',
501 'ext': 'mp3',
502 'upload_date': '20141126',
503 'title': 'Jack Tips: 5 Steps to Permanent Gut Healing',
504 }
505 },
506 # Cinerama player
507 {
508 'url': 'http://www.abc.net.au/7.30/content/2015/s4164797.htm',
509 'info_dict': {
510 'id': '730m_DandD_1901_512k',
511 'ext': 'mp4',
512 'uploader': 'www.abc.net.au',
513 'title': 'Game of Thrones with dice - Dungeons and Dragons fantasy role-playing game gets new life - 19/01/2015',
514 }
515 },
516 # embedded viddler video
517 {
518 'url': 'http://deadspin.com/i-cant-stop-watching-john-wall-chop-the-nuggets-with-th-1681801597',
519 'info_dict': {
520 'id': '4d03aad9',
521 'ext': 'mp4',
522 'uploader': 'deadspin',
523 'title': 'WALL-TO-GORTAT',
524 'timestamp': 1422285291,
525 'upload_date': '20150126',
526 },
527 'add_ie': ['Viddler'],
528 },
529 # jwplayer YouTube
530 {
531 'url': 'http://media.nationalarchives.gov.uk/index.php/webinar-using-discovery-national-archives-online-catalogue/',
532 'info_dict': {
533 'id': 'Mrj4DVp2zeA',
534 'ext': 'mp4',
535 'upload_date': '20150212',
536 'uploader': 'The National Archives UK',
537 'description': 'md5:a236581cd2449dd2df4f93412f3f01c6',
538 'uploader_id': 'NationalArchives08',
539 'title': 'Webinar: Using Discovery, The National Archives’ online catalogue',
540 },
541 },
542 # rtl.nl embed
543 {
544 'url': 'http://www.rtlnieuws.nl/nieuws/buitenland/aanslagen-kopenhagen',
545 'playlist_mincount': 5,
546 'info_dict': {
547 'id': 'aanslagen-kopenhagen',
548 'title': 'Aanslagen Kopenhagen | RTL Nieuws',
549 }
550 },
551 # Zapiks embed
552 {
553 'url': 'http://www.skipass.com/news/116090-bon-appetit-s5ep3-baqueira-mi-cor.html',
554 'info_dict': {
555 'id': '118046',
556 'ext': 'mp4',
557 'title': 'EP3S5 - Bon Appétit - Baqueira Mi Corazon !',
558 }
559 },
560 # Kaltura embed
561 {
562 'url': 'http://www.monumentalnetwork.com/videos/john-carlson-postgame-2-25-15',
563 'info_dict': {
564 'id': '1_eergr3h1',
565 'ext': 'mp4',
566 'upload_date': '20150226',
567 'uploader_id': 'MonumentalSports-Kaltura@perfectsensedigital.com',
568 'timestamp': int,
569 'title': 'John Carlson Postgame 2/25/15',
570 },
571 },
572 ]
573
574 def report_following_redirect(self, new_url):
575 """Report information extraction."""
576 self._downloader.to_screen('[redirect] Following redirect to %s' % new_url)
577
578 def _extract_rss(self, url, video_id, doc):
579 playlist_title = doc.find('./channel/title').text
580 playlist_desc_el = doc.find('./channel/description')
581 playlist_desc = None if playlist_desc_el is None else playlist_desc_el.text
582
583 entries = [{
584 '_type': 'url',
585 'url': e.find('link').text,
586 'title': e.find('title').text,
587 } for e in doc.findall('./channel/item')]
588
589 return {
590 '_type': 'playlist',
591 'id': url,
592 'title': playlist_title,
593 'description': playlist_desc,
594 'entries': entries,
595 }
596
597 def _extract_camtasia(self, url, video_id, webpage):
598 """ Returns None if no camtasia video can be found. """
599
600 camtasia_cfg = self._search_regex(
601 r'fo\.addVariable\(\s*"csConfigFile",\s*"([^"]+)"\s*\);',
602 webpage, 'camtasia configuration file', default=None)
603 if camtasia_cfg is None:
604 return None
605
606 title = self._html_search_meta('DC.title', webpage, fatal=True)
607
608 camtasia_url = compat_urlparse.urljoin(url, camtasia_cfg)
609 camtasia_cfg = self._download_xml(
610 camtasia_url, video_id,
611 note='Downloading camtasia configuration',
612 errnote='Failed to download camtasia configuration')
613 fileset_node = camtasia_cfg.find('./playlist/array/fileset')
614
615 entries = []
616 for n in fileset_node.getchildren():
617 url_n = n.find('./uri')
618 if url_n is None:
619 continue
620
621 entries.append({
622 'id': os.path.splitext(url_n.text.rpartition('/')[2])[0],
623 'title': '%s - %s' % (title, n.tag),
624 'url': compat_urlparse.urljoin(url, url_n.text),
625 'duration': float_or_none(n.find('./duration').text),
626 })
627
628 return {
629 '_type': 'playlist',
630 'entries': entries,
631 'title': title,
632 }
633
634 def _real_extract(self, url):
635 if url.startswith('//'):
636 return {
637 '_type': 'url',
638 'url': self.http_scheme() + url,
639 }
640
641 parsed_url = compat_urlparse.urlparse(url)
642 if not parsed_url.scheme:
643 default_search = self._downloader.params.get('default_search')
644 if default_search is None:
645 default_search = 'fixup_error'
646
647 if default_search in ('auto', 'auto_warning', 'fixup_error'):
648 if '/' in url:
649 self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http')
650 return self.url_result('http://' + url)
651 elif default_search != 'fixup_error':
652 if default_search == 'auto_warning':
653 if re.match(r'^(?:url|URL)$', url):
654 raise ExtractorError(
655 'Invalid URL: %r . Call youtube-dl like this: youtube-dl -v "https://www.youtube.com/watch?v=BaW_jenozKc" ' % url,
656 expected=True)
657 else:
658 self._downloader.report_warning(
659 'Falling back to youtube search for %s . Set --default-search "auto" to suppress this warning.' % url)
660 return self.url_result('ytsearch:' + url)
661
662 if default_search in ('error', 'fixup_error'):
663 raise ExtractorError(
664 '%r is not a valid URL. '
665 'Set --default-search "ytsearch" (or run youtube-dl "ytsearch:%s" ) to search YouTube'
666 % (url, url), expected=True)
667 else:
668 if ':' not in default_search:
669 default_search += ':'
670 return self.url_result(default_search + url)
671
672 url, smuggled_data = unsmuggle_url(url)
673 force_videoid = None
674 is_intentional = smuggled_data and smuggled_data.get('to_generic')
675 if smuggled_data and 'force_videoid' in smuggled_data:
676 force_videoid = smuggled_data['force_videoid']
677 video_id = force_videoid
678 else:
679 video_id = os.path.splitext(url.rstrip('/').split('/')[-1])[0]
680
681 self.to_screen('%s: Requesting header' % video_id)
682
683 head_req = HEADRequest(url)
684 head_response = self._request_webpage(
685 head_req, video_id,
686 note=False, errnote='Could not send HEAD request to %s' % url,
687 fatal=False)
688
689 if head_response is not False:
690 # Check for redirect
691 new_url = head_response.geturl()
692 if url != new_url:
693 self.report_following_redirect(new_url)
694 if force_videoid:
695 new_url = smuggle_url(
696 new_url, {'force_videoid': force_videoid})
697 return self.url_result(new_url)
698
699 full_response = None
700 if head_response is False:
701 full_response = self._request_webpage(url, video_id)
702 head_response = full_response
703
704 # Check for direct link to a video
705 content_type = head_response.headers.get('Content-Type', '')
706 m = re.match(r'^(?P<type>audio|video|application(?=/ogg$))/(?P<format_id>.+)$', content_type)
707 if m:
708 upload_date = unified_strdate(
709 head_response.headers.get('Last-Modified'))
710 return {
711 'id': video_id,
712 'title': os.path.splitext(url_basename(url))[0],
713 'direct': True,
714 'formats': [{
715 'format_id': m.group('format_id'),
716 'url': url,
717 'vcodec': 'none' if m.group('type') == 'audio' else None
718 }],
719 'upload_date': upload_date,
720 }
721
722 if not self._downloader.params.get('test', False) and not is_intentional:
723 self._downloader.report_warning('Falling back on generic information extractor.')
724
725 if not full_response:
726 full_response = self._request_webpage(url, video_id)
727
728 # Maybe it's a direct link to a video?
729 # Be careful not to download the whole thing!
730 first_bytes = full_response.read(512)
731 if not is_html(first_bytes):
732 self._downloader.report_warning(
733 'URL could be a direct video link, returning it as such.')
734 upload_date = unified_strdate(
735 head_response.headers.get('Last-Modified'))
736 return {
737 'id': video_id,
738 'title': os.path.splitext(url_basename(url))[0],
739 'direct': True,
740 'url': url,
741 'upload_date': upload_date,
742 }
743
744 webpage = self._webpage_read_content(
745 full_response, url, video_id, prefix=first_bytes)
746
747 self.report_extraction(video_id)
748
749 # Is it an RSS feed?
750 try:
751 doc = parse_xml(webpage)
752 if doc.tag == 'rss':
753 return self._extract_rss(url, video_id, doc)
754 except compat_xml_parse_error:
755 pass
756
757 # Is it a Camtasia project?
758 camtasia_res = self._extract_camtasia(url, video_id, webpage)
759 if camtasia_res is not None:
760 return camtasia_res
761
762 # Sometimes embedded video player is hidden behind percent encoding
763 # (e.g. https://github.com/rg3/youtube-dl/issues/2448)
764 # Unescaping the whole page allows to handle those cases in a generic way
765 webpage = compat_urllib_parse.unquote(webpage)
766
767 # it's tempting to parse this further, but you would
768 # have to take into account all the variations like
769 # Video Title - Site Name
770 # Site Name | Video Title
771 # Video Title - Tagline | Site Name
772 # and so on and so forth; it's just not practical
773 video_title = self._html_search_regex(
774 r'(?s)<title>(.*?)</title>', webpage, 'video title',
775 default='video')
776
777 # Try to detect age limit automatically
778 age_limit = self._rta_search(webpage)
779 # And then there are the jokers who advertise that they use RTA,
780 # but actually don't.
781 AGE_LIMIT_MARKERS = [
782 r'Proudly Labeled <a href="http://www.rtalabel.org/" title="Restricted to Adults">RTA</a>',
783 ]
784 if any(re.search(marker, webpage) for marker in AGE_LIMIT_MARKERS):
785 age_limit = 18
786
787 # video uploader is domain name
788 video_uploader = self._search_regex(
789 r'^(?:https?://)?([^/]*)/.*', url, 'video uploader')
790
791 # Helper method
792 def _playlist_from_matches(matches, getter=None, ie=None):
793 urlrs = orderedSet(
794 self.url_result(self._proto_relative_url(getter(m) if getter else m), ie)
795 for m in matches)
796 return self.playlist_result(
797 urlrs, playlist_id=video_id, playlist_title=video_title)
798
799 # Look for BrightCove:
800 bc_urls = BrightcoveIE._extract_brightcove_urls(webpage)
801 if bc_urls:
802 self.to_screen('Brightcove video detected.')
803 entries = [{
804 '_type': 'url',
805 'url': smuggle_url(bc_url, {'Referer': url}),
806 'ie_key': 'Brightcove'
807 } for bc_url in bc_urls]
808
809 return {
810 '_type': 'playlist',
811 'title': video_title,
812 'id': video_id,
813 'entries': entries,
814 }
815
816 # Look for embedded rtl.nl player
817 matches = re.findall(
818 r'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="((?:https?:)?//(?:www\.)?rtl\.nl/system/videoplayer/[^"]+video_embed[^"]+)"',
819 webpage)
820 if matches:
821 return _playlist_from_matches(matches, ie='RtlNl')
822
823 # Look for embedded (iframe) Vimeo player
824 mobj = re.search(
825 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//player\.vimeo\.com/video/.+?)\1', webpage)
826 if mobj:
827 player_url = unescapeHTML(mobj.group('url'))
828 surl = smuggle_url(player_url, {'Referer': url})
829 return self.url_result(surl)
830 # Look for embedded (swf embed) Vimeo player
831 mobj = re.search(
832 r'<embed[^>]+?src="((?:https?:)?//(?:www\.)?vimeo\.com/moogaloop\.swf.+?)"', webpage)
833 if mobj:
834 return self.url_result(mobj.group(1))
835
836 # Look for embedded YouTube player
837 matches = re.findall(r'''(?x)
838 (?:
839 <iframe[^>]+?src=|
840 data-video-url=|
841 <embed[^>]+?src=|
842 embedSWF\(?:\s*|
843 new\s+SWFObject\(
844 )
845 (["\'])
846 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
847 (?:embed|v|p)/.+?)
848 \1''', webpage)
849 if matches:
850 return _playlist_from_matches(
851 matches, lambda m: unescapeHTML(m[1]))
852
853 # Look for lazyYT YouTube embed
854 matches = re.findall(
855 r'class="lazyYT" data-youtube-id="([^"]+)"', webpage)
856 if matches:
857 return _playlist_from_matches(matches, lambda m: unescapeHTML(m))
858
859 # Look for embedded Dailymotion player
860 matches = re.findall(
861 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/embed/video/.+?)\1', webpage)
862 if matches:
863 return _playlist_from_matches(
864 matches, lambda m: unescapeHTML(m[1]))
865
866 # Look for embedded Dailymotion playlist player (#3822)
867 m = re.search(
868 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.[a-z]{2,3}/widget/jukebox\?.+?)\1', webpage)
869 if m:
870 playlists = re.findall(
871 r'list\[\]=/playlist/([^/]+)/', unescapeHTML(m.group('url')))
872 if playlists:
873 return _playlist_from_matches(
874 playlists, lambda p: '//dailymotion.com/playlist/%s' % p)
875
876 # Look for embedded Wistia player
877 match = re.search(
878 r'<(?:meta[^>]+?content|iframe[^>]+?src)=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage)
879 if match:
880 embed_url = self._proto_relative_url(
881 unescapeHTML(match.group('url')))
882 return {
883 '_type': 'url_transparent',
884 'url': embed_url,
885 'ie_key': 'Wistia',
886 'uploader': video_uploader,
887 'title': video_title,
888 'id': video_id,
889 }
890
891 match = re.search(r'(?:id=["\']wistia_|data-wistia-?id=["\']|Wistia\.embed\(["\'])(?P<id>[^"\']+)', webpage)
892 if match:
893 return {
894 '_type': 'url_transparent',
895 'url': 'http://fast.wistia.net/embed/iframe/{0:}'.format(match.group('id')),
896 'ie_key': 'Wistia',
897 'uploader': video_uploader,
898 'title': video_title,
899 'id': match.group('id')
900 }
901
902 # Look for embedded blip.tv player
903 mobj = re.search(r'<meta\s[^>]*https?://api\.blip\.tv/\w+/redirect/\w+/(\d+)', webpage)
904 if mobj:
905 return self.url_result('http://blip.tv/a/a-' + mobj.group(1), 'BlipTV')
906 mobj = re.search(r'<(?:iframe|embed|object)\s[^>]*(https?://(?:\w+\.)?blip\.tv/(?:play/|api\.swf#)[a-zA-Z0-9_]+)', webpage)
907 if mobj:
908 return self.url_result(mobj.group(1), 'BlipTV')
909
910 # Look for embedded condenast player
911 matches = re.findall(
912 r'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="(https?://player\.cnevids\.com/embed/[^"]+")',
913 webpage)
914 if matches:
915 return {
916 '_type': 'playlist',
917 'entries': [{
918 '_type': 'url',
919 'ie_key': 'CondeNast',
920 'url': ma,
921 } for ma in matches],
922 'title': video_title,
923 'id': video_id,
924 }
925
926 # Look for Bandcamp pages with custom domain
927 mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage)
928 if mobj is not None:
929 burl = unescapeHTML(mobj.group(1))
930 # Don't set the extractor because it can be a track url or an album
931 return self.url_result(burl)
932
933 # Look for embedded Vevo player
934 mobj = re.search(
935 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:cache\.)?vevo\.com/.+?)\1', webpage)
936 if mobj is not None:
937 return self.url_result(mobj.group('url'))
938
939 # Look for embedded Viddler player
940 mobj = re.search(
941 r'<(?:iframe[^>]+?src|param[^>]+?value)=(["\'])(?P<url>(?:https?:)?//(?:www\.)?viddler\.com/(?:embed|player)/.+?)\1',
942 webpage)
943 if mobj is not None:
944 return self.url_result(mobj.group('url'))
945
946 # Look for Ooyala videos
947 mobj = (re.search(r'player\.ooyala\.com/[^"?]+\?[^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or
948 re.search(r'OO\.Player\.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage) or
949 re.search(r'SBN\.VideoLinkset\.ooyala\([\'"](?P<ec>.{32})[\'"]\)', webpage))
950 if mobj is not None:
951 return OoyalaIE._build_url_result(mobj.group('ec'))
952
953 # Look for multiple Ooyala embeds on SBN network websites
954 mobj = re.search(r'SBN\.VideoLinkset\.entryGroup\((\[.*?\])', webpage)
955 if mobj is not None:
956 embeds = self._parse_json(mobj.group(1), video_id, fatal=False)
957 if embeds:
958 return _playlist_from_matches(
959 embeds, getter=lambda v: OoyalaIE._url_for_embed_code(v['provider_video_id']), ie='Ooyala')
960
961 # Look for Aparat videos
962 mobj = re.search(r'<iframe .*?src="(http://www\.aparat\.com/video/[^"]+)"', webpage)
963 if mobj is not None:
964 return self.url_result(mobj.group(1), 'Aparat')
965
966 # Look for MPORA videos
967 mobj = re.search(r'<iframe .*?src="(http://mpora\.(?:com|de)/videos/[^"]+)"', webpage)
968 if mobj is not None:
969 return self.url_result(mobj.group(1), 'Mpora')
970
971 # Look for embedded NovaMov-based player
972 mobj = re.search(
973 r'''(?x)<(?:pagespeed_)?iframe[^>]+?src=(["\'])
974 (?P<url>http://(?:(?:embed|www)\.)?
975 (?:novamov\.com|
976 nowvideo\.(?:ch|sx|eu|at|ag|co)|
977 videoweed\.(?:es|com)|
978 movshare\.(?:net|sx|ag)|
979 divxstage\.(?:eu|net|ch|co|at|ag))
980 /embed\.php.+?)\1''', webpage)
981 if mobj is not None:
982 return self.url_result(mobj.group('url'))
983
984 # Look for embedded Facebook player
985 mobj = re.search(
986 r'<iframe[^>]+?src=(["\'])(?P<url>https://www\.facebook\.com/video/embed.+?)\1', webpage)
987 if mobj is not None:
988 return self.url_result(mobj.group('url'), 'Facebook')
989
990 # Look for embedded VK player
991 mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://vk\.com/video_ext\.php.+?)\1', webpage)
992 if mobj is not None:
993 return self.url_result(mobj.group('url'), 'VK')
994
995 # Look for embedded ivi player
996 mobj = re.search(r'<embed[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?ivi\.ru/video/player.+?)\1', webpage)
997 if mobj is not None:
998 return self.url_result(mobj.group('url'), 'Ivi')
999
1000 # Look for embedded Huffington Post player
1001 mobj = re.search(
1002 r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed\.live\.huffingtonpost\.com/.+?)\1', webpage)
1003 if mobj is not None:
1004 return self.url_result(mobj.group('url'), 'HuffPost')
1005
1006 # Look for embed.ly
1007 mobj = re.search(r'class=["\']embedly-card["\'][^>]href=["\'](?P<url>[^"\']+)', webpage)
1008 if mobj is not None:
1009 return self.url_result(mobj.group('url'))
1010 mobj = re.search(r'class=["\']embedly-embed["\'][^>]src=["\'][^"\']*url=(?P<url>[^&]+)', webpage)
1011 if mobj is not None:
1012 return self.url_result(compat_urllib_parse.unquote(mobj.group('url')))
1013
1014 # Look for funnyordie embed
1015 matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage)
1016 if matches:
1017 return _playlist_from_matches(
1018 matches, getter=unescapeHTML, ie='FunnyOrDie')
1019
1020 # Look for BBC iPlayer embed
1021 matches = re.findall(r'setPlaylist\("(https?://www\.bbc\.co\.uk/iplayer/[^/]+/[\da-z]{8})"\)', webpage)
1022 if matches:
1023 return _playlist_from_matches(matches, ie='BBCCoUk')
1024
1025 # Look for embedded RUTV player
1026 rutv_url = RUTVIE._extract_url(webpage)
1027 if rutv_url:
1028 return self.url_result(rutv_url, 'RUTV')
1029
1030 # Look for embedded TED player
1031 mobj = re.search(
1032 r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed(?:-ssl)?\.ted\.com/.+?)\1', webpage)
1033 if mobj is not None:
1034 return self.url_result(mobj.group('url'), 'TED')
1035
1036 # Look for embedded Ustream videos
1037 mobj = re.search(
1038 r'<iframe[^>]+?src=(["\'])(?P<url>http://www\.ustream\.tv/embed/.+?)\1', webpage)
1039 if mobj is not None:
1040 return self.url_result(mobj.group('url'), 'Ustream')
1041
1042 # Look for embedded arte.tv player
1043 mobj = re.search(
1044 r'<script [^>]*?src="(?P<url>http://www\.arte\.tv/playerv2/embed[^"]+)"',
1045 webpage)
1046 if mobj is not None:
1047 return self.url_result(mobj.group('url'), 'ArteTVEmbed')
1048
1049 # Look for embedded smotri.com player
1050 smotri_url = SmotriIE._extract_url(webpage)
1051 if smotri_url:
1052 return self.url_result(smotri_url, 'Smotri')
1053
1054 # Look for embeded soundcloud player
1055 mobj = re.search(
1056 r'<iframe\s+(?:[a-zA-Z0-9_-]+="[^"]+"\s+)*src="(?P<url>https?://(?:w\.)?soundcloud\.com/player[^"]+)"',
1057 webpage)
1058 if mobj is not None:
1059 url = unescapeHTML(mobj.group('url'))
1060 return self.url_result(url)
1061
1062 # Look for embedded vulture.com player
1063 mobj = re.search(
1064 r'<iframe src="(?P<url>https?://video\.vulture\.com/[^"]+)"',
1065 webpage)
1066 if mobj is not None:
1067 url = unescapeHTML(mobj.group('url'))
1068 return self.url_result(url, ie='Vulture')
1069
1070 # Look for embedded mtvservices player
1071 mobj = re.search(
1072 r'<iframe src="(?P<url>https?://media\.mtvnservices\.com/embed/[^"]+)"',
1073 webpage)
1074 if mobj is not None:
1075 url = unescapeHTML(mobj.group('url'))
1076 return self.url_result(url, ie='MTVServicesEmbedded')
1077
1078 # Look for embedded yahoo player
1079 mobj = re.search(
1080 r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:screen|movies)\.yahoo\.com/.+?\.html\?format=embed)\1',
1081 webpage)
1082 if mobj is not None:
1083 return self.url_result(mobj.group('url'), 'Yahoo')
1084
1085 # Look for embedded sbs.com.au player
1086 mobj = re.search(
1087 r'''(?x)
1088 (?:
1089 <meta\s+property="og:video"\s+content=|
1090 <iframe[^>]+?src=
1091 )
1092 (["\'])(?P<url>https?://(?:www\.)?sbs\.com\.au/ondemand/video/.+?)\1''',
1093 webpage)
1094 if mobj is not None:
1095 return self.url_result(mobj.group('url'), 'SBS')
1096
1097 # Look for embedded Cinchcast player
1098 mobj = re.search(
1099 r'<iframe[^>]+?src=(["\'])(?P<url>https?://player\.cinchcast\.com/.+?)\1',
1100 webpage)
1101 if mobj is not None:
1102 return self.url_result(mobj.group('url'), 'Cinchcast')
1103
1104 mobj = re.search(
1105 r'<iframe[^>]+?src=(["\'])(?P<url>https?://m(?:lb)?\.mlb\.com/shared/video/embed/embed\.html\?.+?)\1',
1106 webpage)
1107 if mobj is not None:
1108 return self.url_result(mobj.group('url'), 'MLB')
1109
1110 mobj = re.search(
1111 r'<iframe[^>]+?src=(["\'])(?P<url>%s)\1' % CondeNastIE.EMBED_URL,
1112 webpage)
1113 if mobj is not None:
1114 return self.url_result(self._proto_relative_url(mobj.group('url'), scheme='http:'), 'CondeNast')
1115
1116 mobj = re.search(
1117 r'<iframe[^>]+src="(?P<url>https?://new\.livestream\.com/[^"]+/player[^"]+)"',
1118 webpage)
1119 if mobj is not None:
1120 return self.url_result(mobj.group('url'), 'Livestream')
1121
1122 # Look for Zapiks embed
1123 mobj = re.search(
1124 r'<iframe[^>]+src="(?P<url>https?://(?:www\.)?zapiks\.fr/index\.php\?.+?)"', webpage)
1125 if mobj is not None:
1126 return self.url_result(mobj.group('url'), 'Zapiks')
1127
1128 # Look for Kaltura embeds
1129 mobj = re.search(
1130 r"(?s)kWidget\.(?:thumb)?[Ee]mbed\(\{.*?'wid'\s*:\s*'_?(?P<partner_id>[^']+)',.*?'entry_id'\s*:\s*'(?P<id>[^']+)',", webpage)
1131 if mobj is not None:
1132 return self.url_result('kaltura:%(partner_id)s:%(id)s' % mobj.groupdict(), 'Kaltura')
1133
1134 def check_video(vurl):
1135 if YoutubeIE.suitable(vurl):
1136 return True
1137 vpath = compat_urlparse.urlparse(vurl).path
1138 vext = determine_ext(vpath)
1139 return '.' in vpath and vext not in ('swf', 'png', 'jpg', 'srt', 'sbv', 'sub', 'vtt', 'ttml')
1140
1141 def filter_video(urls):
1142 return list(filter(check_video, urls))
1143
1144 # Start with something easy: JW Player in SWFObject
1145 found = filter_video(re.findall(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage))
1146 if not found:
1147 # Look for gorilla-vid style embedding
1148 found = filter_video(re.findall(r'''(?sx)
1149 (?:
1150 jw_plugins|
1151 JWPlayerOptions|
1152 jwplayer\s*\(\s*["'][^'"]+["']\s*\)\s*\.setup
1153 )
1154 .*?
1155 ['"]?file['"]?\s*:\s*["\'](.*?)["\']''', webpage))
1156 if not found:
1157 # Broaden the search a little bit
1158 found = filter_video(re.findall(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage))
1159 if not found:
1160 # Broaden the findall a little bit: JWPlayer JS loader
1161 found = filter_video(re.findall(
1162 r'[^A-Za-z0-9]?file["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage))
1163 if not found:
1164 # Flow player
1165 found = filter_video(re.findall(r'''(?xs)
1166 flowplayer\("[^"]+",\s*
1167 \{[^}]+?\}\s*,
1168 \s*\{[^}]+? ["']?clip["']?\s*:\s*\{\s*
1169 ["']?url["']?\s*:\s*["']([^"']+)["']
1170 ''', webpage))
1171 if not found:
1172 # Cinerama player
1173 found = re.findall(
1174 r"cinerama\.embedPlayer\(\s*\'[^']+\',\s*'([^']+)'", webpage)
1175 if not found:
1176 # Try to find twitter cards info
1177 found = filter_video(re.findall(
1178 r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage))
1179 if not found:
1180 # We look for Open Graph info:
1181 # We have to match any number spaces between elements, some sites try to align them (eg.: statigr.am)
1182 m_video_type = re.findall(r'<meta.*?property="og:video:type".*?content="video/(.*?)"', webpage)
1183 # We only look in og:video if the MIME type is a video, don't try if it's a Flash player:
1184 if m_video_type is not None:
1185 found = filter_video(re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage))
1186 if not found:
1187 # HTML5 video
1188 found = re.findall(r'(?s)<video[^<]*(?:>.*?<source[^>]*)?\s+src=["\'](.*?)["\']', webpage)
1189 if not found:
1190 found = re.search(
1191 r'(?i)<meta\s+(?=(?:[a-z-]+="[^"]+"\s+)*http-equiv="refresh")'
1192 r'(?:[a-z-]+="[^"]+"\s+)*?content="[0-9]{,2};url=\'?([^\'"]+)',
1193 webpage)
1194 if found:
1195 new_url = found.group(1)
1196 self.report_following_redirect(new_url)
1197 return {
1198 '_type': 'url',
1199 'url': new_url,
1200 }
1201 if not found:
1202 raise UnsupportedError(url)
1203
1204 entries = []
1205 for video_url in found:
1206 video_url = compat_urlparse.urljoin(url, video_url)
1207 video_id = compat_urllib_parse.unquote(os.path.basename(video_url))
1208
1209 # Sometimes, jwplayer extraction will result in a YouTube URL
1210 if YoutubeIE.suitable(video_url):
1211 entries.append(self.url_result(video_url, 'Youtube'))
1212 continue
1213
1214 # here's a fun little line of code for you:
1215 video_id = os.path.splitext(video_id)[0]
1216
1217 entries.append({
1218 'id': video_id,
1219 'url': video_url,
1220 'uploader': video_uploader,
1221 'title': video_title,
1222 'age_limit': age_limit,
1223 })
1224
1225 if len(entries) == 1:
1226 return entries[0]
1227 else:
1228 for num, e in enumerate(entries, start=1):
1229 # 'url' results don't have a title
1230 if e.get('title') is not None:
1231 e['title'] = '%s (%d)' % (e['title'], num)
1232 return {
1233 '_type': 'playlist',
1234 'entries': entries,
1235 }