]> Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/generic.py
debian/control: Add recommends on aria2 | wget | curl to use external downloaders.
[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 'title': 'Unity 8 desktop-mode windows on Mir! - Ubuntu Discourse',
477 },
478 'playlist_mincount': 2,
479 },
480 # Direct link with incorrect MIME type
481 {
482 'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
483 'md5': '4ccbebe5f36706d85221f204d7eb5913',
484 'info_dict': {
485 'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
486 'id': '5_Lennart_Poettering_-_Systemd',
487 'ext': 'webm',
488 'title': '5_Lennart_Poettering_-_Systemd',
489 'upload_date': '20141120',
490 },
491 'expected_warnings': [
492 'URL could be a direct video link, returning it as such.'
493 ]
494 },
495 # Cinchcast embed
496 {
497 'url': 'http://undergroundwellness.com/podcasts/306-5-steps-to-permanent-gut-healing/',
498 'info_dict': {
499 'id': '7141703',
500 'ext': 'mp3',
501 'upload_date': '20141126',
502 'title': 'Jack Tips: 5 Steps to Permanent Gut Healing',
503 }
504 },
505 # Cinerama player
506 {
507 'url': 'http://www.abc.net.au/7.30/content/2015/s4164797.htm',
508 'info_dict': {
509 'id': '730m_DandD_1901_512k',
510 'ext': 'mp4',
511 'uploader': 'www.abc.net.au',
512 'title': 'Game of Thrones with dice - Dungeons and Dragons fantasy role-playing game gets new life - 19/01/2015',
513 }
514 },
515 # embedded viddler video
516 {
517 'url': 'http://deadspin.com/i-cant-stop-watching-john-wall-chop-the-nuggets-with-th-1681801597',
518 'info_dict': {
519 'id': '4d03aad9',
520 'ext': 'mp4',
521 'uploader': 'deadspin',
522 'title': 'WALL-TO-GORTAT',
523 'timestamp': 1422285291,
524 'upload_date': '20150126',
525 },
526 'add_ie': ['Viddler'],
527 }
528 ]
529
530 def report_following_redirect(self, new_url):
531 """Report information extraction."""
532 self._downloader.to_screen('[redirect] Following redirect to %s' % new_url)
533
534 def _extract_rss(self, url, video_id, doc):
535 playlist_title = doc.find('./channel/title').text
536 playlist_desc_el = doc.find('./channel/description')
537 playlist_desc = None if playlist_desc_el is None else playlist_desc_el.text
538
539 entries = [{
540 '_type': 'url',
541 'url': e.find('link').text,
542 'title': e.find('title').text,
543 } for e in doc.findall('./channel/item')]
544
545 return {
546 '_type': 'playlist',
547 'id': url,
548 'title': playlist_title,
549 'description': playlist_desc,
550 'entries': entries,
551 }
552
553 def _extract_camtasia(self, url, video_id, webpage):
554 """ Returns None if no camtasia video can be found. """
555
556 camtasia_cfg = self._search_regex(
557 r'fo\.addVariable\(\s*"csConfigFile",\s*"([^"]+)"\s*\);',
558 webpage, 'camtasia configuration file', default=None)
559 if camtasia_cfg is None:
560 return None
561
562 title = self._html_search_meta('DC.title', webpage, fatal=True)
563
564 camtasia_url = compat_urlparse.urljoin(url, camtasia_cfg)
565 camtasia_cfg = self._download_xml(
566 camtasia_url, video_id,
567 note='Downloading camtasia configuration',
568 errnote='Failed to download camtasia configuration')
569 fileset_node = camtasia_cfg.find('./playlist/array/fileset')
570
571 entries = []
572 for n in fileset_node.getchildren():
573 url_n = n.find('./uri')
574 if url_n is None:
575 continue
576
577 entries.append({
578 'id': os.path.splitext(url_n.text.rpartition('/')[2])[0],
579 'title': '%s - %s' % (title, n.tag),
580 'url': compat_urlparse.urljoin(url, url_n.text),
581 'duration': float_or_none(n.find('./duration').text),
582 })
583
584 return {
585 '_type': 'playlist',
586 'entries': entries,
587 'title': title,
588 }
589
590 def _real_extract(self, url):
591 if url.startswith('//'):
592 return {
593 '_type': 'url',
594 'url': self.http_scheme() + url,
595 }
596
597 parsed_url = compat_urlparse.urlparse(url)
598 if not parsed_url.scheme:
599 default_search = self._downloader.params.get('default_search')
600 if default_search is None:
601 default_search = 'fixup_error'
602
603 if default_search in ('auto', 'auto_warning', 'fixup_error'):
604 if '/' in url:
605 self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http')
606 return self.url_result('http://' + url)
607 elif default_search != 'fixup_error':
608 if default_search == 'auto_warning':
609 if re.match(r'^(?:url|URL)$', url):
610 raise ExtractorError(
611 'Invalid URL: %r . Call youtube-dl like this: youtube-dl -v "https://www.youtube.com/watch?v=BaW_jenozKc" ' % url,
612 expected=True)
613 else:
614 self._downloader.report_warning(
615 'Falling back to youtube search for %s . Set --default-search "auto" to suppress this warning.' % url)
616 return self.url_result('ytsearch:' + url)
617
618 if default_search in ('error', 'fixup_error'):
619 raise ExtractorError(
620 '%r is not a valid URL. '
621 'Set --default-search "ytsearch" (or run youtube-dl "ytsearch:%s" ) to search YouTube'
622 % (url, url), expected=True)
623 else:
624 if ':' not in default_search:
625 default_search += ':'
626 return self.url_result(default_search + url)
627
628 url, smuggled_data = unsmuggle_url(url)
629 force_videoid = None
630 is_intentional = smuggled_data and smuggled_data.get('to_generic')
631 if smuggled_data and 'force_videoid' in smuggled_data:
632 force_videoid = smuggled_data['force_videoid']
633 video_id = force_videoid
634 else:
635 video_id = os.path.splitext(url.rstrip('/').split('/')[-1])[0]
636
637 self.to_screen('%s: Requesting header' % video_id)
638
639 head_req = HEADRequest(url)
640 head_response = self._request_webpage(
641 head_req, video_id,
642 note=False, errnote='Could not send HEAD request to %s' % url,
643 fatal=False)
644
645 if head_response is not False:
646 # Check for redirect
647 new_url = head_response.geturl()
648 if url != new_url:
649 self.report_following_redirect(new_url)
650 if force_videoid:
651 new_url = smuggle_url(
652 new_url, {'force_videoid': force_videoid})
653 return self.url_result(new_url)
654
655 full_response = None
656 if head_response is False:
657 full_response = self._request_webpage(url, video_id)
658 head_response = full_response
659
660 # Check for direct link to a video
661 content_type = head_response.headers.get('Content-Type', '')
662 m = re.match(r'^(?P<type>audio|video|application(?=/ogg$))/(?P<format_id>.+)$', content_type)
663 if m:
664 upload_date = unified_strdate(
665 head_response.headers.get('Last-Modified'))
666 return {
667 'id': video_id,
668 'title': os.path.splitext(url_basename(url))[0],
669 'direct': True,
670 'formats': [{
671 'format_id': m.group('format_id'),
672 'url': url,
673 'vcodec': 'none' if m.group('type') == 'audio' else None
674 }],
675 'upload_date': upload_date,
676 }
677
678 if not self._downloader.params.get('test', False) and not is_intentional:
679 self._downloader.report_warning('Falling back on generic information extractor.')
680
681 if not full_response:
682 full_response = self._request_webpage(url, video_id)
683
684 # Maybe it's a direct link to a video?
685 # Be careful not to download the whole thing!
686 first_bytes = full_response.read(512)
687 if not is_html(first_bytes):
688 self._downloader.report_warning(
689 'URL could be a direct video link, returning it as such.')
690 upload_date = unified_strdate(
691 head_response.headers.get('Last-Modified'))
692 return {
693 'id': video_id,
694 'title': os.path.splitext(url_basename(url))[0],
695 'direct': True,
696 'url': url,
697 'upload_date': upload_date,
698 }
699
700 webpage = self._webpage_read_content(
701 full_response, url, video_id, prefix=first_bytes)
702
703 self.report_extraction(video_id)
704
705 # Is it an RSS feed?
706 try:
707 doc = parse_xml(webpage)
708 if doc.tag == 'rss':
709 return self._extract_rss(url, video_id, doc)
710 except compat_xml_parse_error:
711 pass
712
713 # Is it a Camtasia project?
714 camtasia_res = self._extract_camtasia(url, video_id, webpage)
715 if camtasia_res is not None:
716 return camtasia_res
717
718 # Sometimes embedded video player is hidden behind percent encoding
719 # (e.g. https://github.com/rg3/youtube-dl/issues/2448)
720 # Unescaping the whole page allows to handle those cases in a generic way
721 webpage = compat_urllib_parse.unquote(webpage)
722
723 # it's tempting to parse this further, but you would
724 # have to take into account all the variations like
725 # Video Title - Site Name
726 # Site Name | Video Title
727 # Video Title - Tagline | Site Name
728 # and so on and so forth; it's just not practical
729 video_title = self._html_search_regex(
730 r'(?s)<title>(.*?)</title>', webpage, 'video title',
731 default='video')
732
733 # Try to detect age limit automatically
734 age_limit = self._rta_search(webpage)
735 # And then there are the jokers who advertise that they use RTA,
736 # but actually don't.
737 AGE_LIMIT_MARKERS = [
738 r'Proudly Labeled <a href="http://www.rtalabel.org/" title="Restricted to Adults">RTA</a>',
739 ]
740 if any(re.search(marker, webpage) for marker in AGE_LIMIT_MARKERS):
741 age_limit = 18
742
743 # video uploader is domain name
744 video_uploader = self._search_regex(
745 r'^(?:https?://)?([^/]*)/.*', url, 'video uploader')
746
747 # Helper method
748 def _playlist_from_matches(matches, getter=None, ie=None):
749 urlrs = orderedSet(
750 self.url_result(self._proto_relative_url(getter(m) if getter else m), ie)
751 for m in matches)
752 return self.playlist_result(
753 urlrs, playlist_id=video_id, playlist_title=video_title)
754
755 # Look for BrightCove:
756 bc_urls = BrightcoveIE._extract_brightcove_urls(webpage)
757 if bc_urls:
758 self.to_screen('Brightcove video detected.')
759 entries = [{
760 '_type': 'url',
761 'url': smuggle_url(bc_url, {'Referer': url}),
762 'ie_key': 'Brightcove'
763 } for bc_url in bc_urls]
764
765 return {
766 '_type': 'playlist',
767 'title': video_title,
768 'id': video_id,
769 'entries': entries,
770 }
771
772 # Look for embedded (iframe) Vimeo player
773 mobj = re.search(
774 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//player\.vimeo\.com/video/.+?)\1', webpage)
775 if mobj:
776 player_url = unescapeHTML(mobj.group('url'))
777 surl = smuggle_url(player_url, {'Referer': url})
778 return self.url_result(surl)
779
780 # Look for embedded (swf embed) Vimeo player
781 mobj = re.search(
782 r'<embed[^>]+?src="((?:https?:)?//(?:www\.)?vimeo\.com/moogaloop\.swf.+?)"', webpage)
783 if mobj:
784 return self.url_result(mobj.group(1))
785
786 # Look for embedded YouTube player
787 matches = re.findall(r'''(?x)
788 (?:
789 <iframe[^>]+?src=|
790 data-video-url=|
791 <embed[^>]+?src=|
792 embedSWF\(?:\s*|
793 new\s+SWFObject\(
794 )
795 (["\'])
796 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
797 (?:embed|v|p)/.+?)
798 \1''', webpage)
799 if matches:
800 return _playlist_from_matches(
801 matches, lambda m: unescapeHTML(m[1]))
802
803 # Look for lazyYT YouTube embed
804 matches = re.findall(
805 r'class="lazyYT" data-youtube-id="([^"]+)"', webpage)
806 if matches:
807 return _playlist_from_matches(matches, lambda m: unescapeHTML(m))
808
809 # Look for embedded Dailymotion player
810 matches = re.findall(
811 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/embed/video/.+?)\1', webpage)
812 if matches:
813 return _playlist_from_matches(
814 matches, lambda m: unescapeHTML(m[1]))
815
816 # Look for embedded Dailymotion playlist player (#3822)
817 m = re.search(
818 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.[a-z]{2,3}/widget/jukebox\?.+?)\1', webpage)
819 if m:
820 playlists = re.findall(
821 r'list\[\]=/playlist/([^/]+)/', unescapeHTML(m.group('url')))
822 if playlists:
823 return _playlist_from_matches(
824 playlists, lambda p: '//dailymotion.com/playlist/%s' % p)
825
826 # Look for embedded Wistia player
827 match = re.search(
828 r'<(?:meta[^>]+?content|iframe[^>]+?src)=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage)
829 if match:
830 embed_url = self._proto_relative_url(
831 unescapeHTML(match.group('url')))
832 return {
833 '_type': 'url_transparent',
834 'url': embed_url,
835 'ie_key': 'Wistia',
836 'uploader': video_uploader,
837 'title': video_title,
838 'id': video_id,
839 }
840
841 match = re.search(r'(?:id=["\']wistia_|data-wistia-?id=["\']|Wistia\.embed\(["\'])(?P<id>[^"\']+)', webpage)
842 if match:
843 return {
844 '_type': 'url_transparent',
845 'url': 'http://fast.wistia.net/embed/iframe/{0:}'.format(match.group('id')),
846 'ie_key': 'Wistia',
847 'uploader': video_uploader,
848 'title': video_title,
849 'id': match.group('id')
850 }
851
852 # Look for embedded blip.tv player
853 mobj = re.search(r'<meta\s[^>]*https?://api\.blip\.tv/\w+/redirect/\w+/(\d+)', webpage)
854 if mobj:
855 return self.url_result('http://blip.tv/a/a-' + mobj.group(1), 'BlipTV')
856 mobj = re.search(r'<(?:iframe|embed|object)\s[^>]*(https?://(?:\w+\.)?blip\.tv/(?:play/|api\.swf#)[a-zA-Z0-9_]+)', webpage)
857 if mobj:
858 return self.url_result(mobj.group(1), 'BlipTV')
859
860 # Look for embedded condenast player
861 matches = re.findall(
862 r'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="(https?://player\.cnevids\.com/embed/[^"]+")',
863 webpage)
864 if matches:
865 return {
866 '_type': 'playlist',
867 'entries': [{
868 '_type': 'url',
869 'ie_key': 'CondeNast',
870 'url': ma,
871 } for ma in matches],
872 'title': video_title,
873 'id': video_id,
874 }
875
876 # Look for Bandcamp pages with custom domain
877 mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage)
878 if mobj is not None:
879 burl = unescapeHTML(mobj.group(1))
880 # Don't set the extractor because it can be a track url or an album
881 return self.url_result(burl)
882
883 # Look for embedded Vevo player
884 mobj = re.search(
885 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:cache\.)?vevo\.com/.+?)\1', webpage)
886 if mobj is not None:
887 return self.url_result(mobj.group('url'))
888
889 # Look for embedded Viddler player
890 mobj = re.search(
891 r'<(?:iframe[^>]+?src|param[^>]+?value)=(["\'])(?P<url>(?:https?:)?//(?:www\.)?viddler\.com/(?:embed|player)/.+?)\1',
892 webpage)
893 if mobj is not None:
894 return self.url_result(mobj.group('url'))
895
896 # Look for Ooyala videos
897 mobj = (re.search(r'player\.ooyala\.com/[^"?]+\?[^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or
898 re.search(r'OO\.Player\.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage) or
899 re.search(r'SBN\.VideoLinkset\.ooyala\([\'"](?P<ec>.{32})[\'"]\)', webpage))
900 if mobj is not None:
901 return OoyalaIE._build_url_result(mobj.group('ec'))
902
903 # Look for multiple Ooyala embeds on SBN network websites
904 mobj = re.search(r'SBN\.VideoLinkset\.entryGroup\((\[.*?\])', webpage)
905 if mobj is not None:
906 embeds = self._parse_json(mobj.group(1), video_id, fatal=False)
907 if embeds:
908 return _playlist_from_matches(
909 embeds, getter=lambda v: OoyalaIE._url_for_embed_code(v['provider_video_id']), ie='Ooyala')
910
911 # Look for Aparat videos
912 mobj = re.search(r'<iframe .*?src="(http://www\.aparat\.com/video/[^"]+)"', webpage)
913 if mobj is not None:
914 return self.url_result(mobj.group(1), 'Aparat')
915
916 # Look for MPORA videos
917 mobj = re.search(r'<iframe .*?src="(http://mpora\.(?:com|de)/videos/[^"]+)"', webpage)
918 if mobj is not None:
919 return self.url_result(mobj.group(1), 'Mpora')
920
921 # Look for embedded NovaMov-based player
922 mobj = re.search(
923 r'''(?x)<(?:pagespeed_)?iframe[^>]+?src=(["\'])
924 (?P<url>http://(?:(?:embed|www)\.)?
925 (?:novamov\.com|
926 nowvideo\.(?:ch|sx|eu|at|ag|co)|
927 videoweed\.(?:es|com)|
928 movshare\.(?:net|sx|ag)|
929 divxstage\.(?:eu|net|ch|co|at|ag))
930 /embed\.php.+?)\1''', webpage)
931 if mobj is not None:
932 return self.url_result(mobj.group('url'))
933
934 # Look for embedded Facebook player
935 mobj = re.search(
936 r'<iframe[^>]+?src=(["\'])(?P<url>https://www\.facebook\.com/video/embed.+?)\1', webpage)
937 if mobj is not None:
938 return self.url_result(mobj.group('url'), 'Facebook')
939
940 # Look for embedded VK player
941 mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://vk\.com/video_ext\.php.+?)\1', webpage)
942 if mobj is not None:
943 return self.url_result(mobj.group('url'), 'VK')
944
945 # Look for embedded ivi player
946 mobj = re.search(r'<embed[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?ivi\.ru/video/player.+?)\1', webpage)
947 if mobj is not None:
948 return self.url_result(mobj.group('url'), 'Ivi')
949
950 # Look for embedded Huffington Post player
951 mobj = re.search(
952 r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed\.live\.huffingtonpost\.com/.+?)\1', webpage)
953 if mobj is not None:
954 return self.url_result(mobj.group('url'), 'HuffPost')
955
956 # Look for embed.ly
957 mobj = re.search(r'class=["\']embedly-card["\'][^>]href=["\'](?P<url>[^"\']+)', webpage)
958 if mobj is not None:
959 return self.url_result(mobj.group('url'))
960 mobj = re.search(r'class=["\']embedly-embed["\'][^>]src=["\'][^"\']*url=(?P<url>[^&]+)', webpage)
961 if mobj is not None:
962 return self.url_result(compat_urllib_parse.unquote(mobj.group('url')))
963
964 # Look for funnyordie embed
965 matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage)
966 if matches:
967 return _playlist_from_matches(
968 matches, getter=unescapeHTML, ie='FunnyOrDie')
969
970 # Look for BBC iPlayer embed
971 matches = re.findall(r'setPlaylist\("(https?://www\.bbc\.co\.uk/iplayer/[^/]+/[\da-z]{8})"\)', webpage)
972 if matches:
973 return _playlist_from_matches(matches, ie='BBCCoUk')
974
975 # Look for embedded RUTV player
976 rutv_url = RUTVIE._extract_url(webpage)
977 if rutv_url:
978 return self.url_result(rutv_url, 'RUTV')
979
980 # Look for embedded TED player
981 mobj = re.search(
982 r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed(?:-ssl)?\.ted\.com/.+?)\1', webpage)
983 if mobj is not None:
984 return self.url_result(mobj.group('url'), 'TED')
985
986 # Look for embedded Ustream videos
987 mobj = re.search(
988 r'<iframe[^>]+?src=(["\'])(?P<url>http://www\.ustream\.tv/embed/.+?)\1', webpage)
989 if mobj is not None:
990 return self.url_result(mobj.group('url'), 'Ustream')
991
992 # Look for embedded arte.tv player
993 mobj = re.search(
994 r'<script [^>]*?src="(?P<url>http://www\.arte\.tv/playerv2/embed[^"]+)"',
995 webpage)
996 if mobj is not None:
997 return self.url_result(mobj.group('url'), 'ArteTVEmbed')
998
999 # Look for embedded smotri.com player
1000 smotri_url = SmotriIE._extract_url(webpage)
1001 if smotri_url:
1002 return self.url_result(smotri_url, 'Smotri')
1003
1004 # Look for embeded soundcloud player
1005 mobj = re.search(
1006 r'<iframe\s+(?:[a-zA-Z0-9_-]+="[^"]+"\s+)*src="(?P<url>https?://(?:w\.)?soundcloud\.com/player[^"]+)"',
1007 webpage)
1008 if mobj is not None:
1009 url = unescapeHTML(mobj.group('url'))
1010 return self.url_result(url)
1011
1012 # Look for embedded vulture.com player
1013 mobj = re.search(
1014 r'<iframe src="(?P<url>https?://video\.vulture\.com/[^"]+)"',
1015 webpage)
1016 if mobj is not None:
1017 url = unescapeHTML(mobj.group('url'))
1018 return self.url_result(url, ie='Vulture')
1019
1020 # Look for embedded mtvservices player
1021 mobj = re.search(
1022 r'<iframe src="(?P<url>https?://media\.mtvnservices\.com/embed/[^"]+)"',
1023 webpage)
1024 if mobj is not None:
1025 url = unescapeHTML(mobj.group('url'))
1026 return self.url_result(url, ie='MTVServicesEmbedded')
1027
1028 # Look for embedded yahoo player
1029 mobj = re.search(
1030 r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:screen|movies)\.yahoo\.com/.+?\.html\?format=embed)\1',
1031 webpage)
1032 if mobj is not None:
1033 return self.url_result(mobj.group('url'), 'Yahoo')
1034
1035 # Look for embedded sbs.com.au player
1036 mobj = re.search(
1037 r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:www\.)sbs\.com\.au/ondemand/video/single/.+?)\1',
1038 webpage)
1039 if mobj is not None:
1040 return self.url_result(mobj.group('url'), 'SBS')
1041
1042 # Look for embedded Cinchcast player
1043 mobj = re.search(
1044 r'<iframe[^>]+?src=(["\'])(?P<url>https?://player\.cinchcast\.com/.+?)\1',
1045 webpage)
1046 if mobj is not None:
1047 return self.url_result(mobj.group('url'), 'Cinchcast')
1048
1049 mobj = re.search(
1050 r'<iframe[^>]+?src=(["\'])(?P<url>https?://m(?:lb)?\.mlb\.com/shared/video/embed/embed\.html\?.+?)\1',
1051 webpage)
1052 if mobj is not None:
1053 return self.url_result(mobj.group('url'), 'MLB')
1054
1055 mobj = re.search(
1056 r'<iframe[^>]+?src=(["\'])(?P<url>%s)\1' % CondeNastIE.EMBED_URL,
1057 webpage)
1058 if mobj is not None:
1059 return self.url_result(self._proto_relative_url(mobj.group('url'), scheme='http:'), 'CondeNast')
1060
1061 mobj = re.search(
1062 r'<iframe[^>]+src="(?P<url>https?://new\.livestream\.com/[^"]+/player[^"]+)"',
1063 webpage)
1064 if mobj is not None:
1065 return self.url_result(mobj.group('url'), 'Livestream')
1066
1067 def check_video(vurl):
1068 vpath = compat_urlparse.urlparse(vurl).path
1069 vext = determine_ext(vpath)
1070 return '.' in vpath and vext not in ('swf', 'png', 'jpg', 'srt', 'sbv', 'sub', 'vtt', 'ttml')
1071
1072 def filter_video(urls):
1073 return list(filter(check_video, urls))
1074
1075 # Start with something easy: JW Player in SWFObject
1076 found = filter_video(re.findall(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage))
1077 if not found:
1078 # Look for gorilla-vid style embedding
1079 found = filter_video(re.findall(r'''(?sx)
1080 (?:
1081 jw_plugins|
1082 JWPlayerOptions|
1083 jwplayer\s*\(\s*["'][^'"]+["']\s*\)\s*\.setup
1084 )
1085 .*?file\s*:\s*["\'](.*?)["\']''', webpage))
1086 if not found:
1087 # Broaden the search a little bit
1088 found = filter_video(re.findall(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage))
1089 if not found:
1090 # Broaden the findall a little bit: JWPlayer JS loader
1091 found = filter_video(re.findall(
1092 r'[^A-Za-z0-9]?file["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage))
1093 if not found:
1094 # Flow player
1095 found = filter_video(re.findall(r'''(?xs)
1096 flowplayer\("[^"]+",\s*
1097 \{[^}]+?\}\s*,
1098 \s*\{[^}]+? ["']?clip["']?\s*:\s*\{\s*
1099 ["']?url["']?\s*:\s*["']([^"']+)["']
1100 ''', webpage))
1101 if not found:
1102 # Cinerama player
1103 found = re.findall(
1104 r"cinerama\.embedPlayer\(\s*\'[^']+\',\s*'([^']+)'", webpage)
1105 if not found:
1106 # Try to find twitter cards info
1107 found = filter_video(re.findall(
1108 r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage))
1109 if not found:
1110 # We look for Open Graph info:
1111 # We have to match any number spaces between elements, some sites try to align them (eg.: statigr.am)
1112 m_video_type = re.findall(r'<meta.*?property="og:video:type".*?content="video/(.*?)"', webpage)
1113 # We only look in og:video if the MIME type is a video, don't try if it's a Flash player:
1114 if m_video_type is not None:
1115 found = filter_video(re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage))
1116 if not found:
1117 # HTML5 video
1118 found = re.findall(r'(?s)<video[^<]*(?:>.*?<source[^>]*)?\s+src=["\'](.*?)["\']', webpage)
1119 if not found:
1120 found = re.search(
1121 r'(?i)<meta\s+(?=(?:[a-z-]+="[^"]+"\s+)*http-equiv="refresh")'
1122 r'(?:[a-z-]+="[^"]+"\s+)*?content="[0-9]{,2};url=\'?([^\'"]+)',
1123 webpage)
1124 if found:
1125 new_url = found.group(1)
1126 self.report_following_redirect(new_url)
1127 return {
1128 '_type': 'url',
1129 'url': new_url,
1130 }
1131 if not found:
1132 raise UnsupportedError(url)
1133
1134 entries = []
1135 for video_url in found:
1136 video_url = compat_urlparse.urljoin(url, video_url)
1137 video_id = compat_urllib_parse.unquote(os.path.basename(video_url))
1138
1139 # Sometimes, jwplayer extraction will result in a YouTube URL
1140 if YoutubeIE.suitable(video_url):
1141 entries.append(self.url_result(video_url, 'Youtube'))
1142 continue
1143
1144 # here's a fun little line of code for you:
1145 video_id = os.path.splitext(video_id)[0]
1146
1147 entries.append({
1148 'id': video_id,
1149 'url': video_url,
1150 'uploader': video_uploader,
1151 'title': video_title,
1152 'age_limit': age_limit,
1153 })
1154
1155 if len(entries) == 1:
1156 return entries[0]
1157 else:
1158 for num, e in enumerate(entries, start=1):
1159 e['title'] = '%s (%d)' % (e['title'], num)
1160 return {
1161 '_type': 'playlist',
1162 'entries': entries,
1163 }