]> Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/vk.py
New upstream version 2018.04.25
[youtubedl] / youtube_dl / extractor / vk.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 import collections
5 import re
6 import sys
7
8 from .common import InfoExtractor
9 from ..compat import (
10 compat_str,
11 compat_urlparse,
12 )
13 from ..utils import (
14 clean_html,
15 ExtractorError,
16 get_element_by_class,
17 int_or_none,
18 orderedSet,
19 remove_start,
20 str_to_int,
21 unescapeHTML,
22 unified_timestamp,
23 urlencode_postdata,
24 )
25 from .dailymotion import DailymotionIE
26 from .pladform import PladformIE
27 from .vimeo import VimeoIE
28 from .youtube import YoutubeIE
29
30
31 class VKBaseIE(InfoExtractor):
32 _NETRC_MACHINE = 'vk'
33
34 def _login(self):
35 (username, password) = self._get_login_info()
36 if username is None:
37 return
38
39 login_page, url_handle = self._download_webpage_handle(
40 'https://vk.com', None, 'Downloading login page')
41
42 login_form = self._hidden_inputs(login_page)
43
44 login_form.update({
45 'email': username.encode('cp1251'),
46 'pass': password.encode('cp1251'),
47 })
48
49 # https://new.vk.com/ serves two same remixlhk cookies in Set-Cookie header
50 # and expects the first one to be set rather than second (see
51 # https://github.com/rg3/youtube-dl/issues/9841#issuecomment-227871201).
52 # As of RFC6265 the newer one cookie should be set into cookie store
53 # what actually happens.
54 # We will workaround this VK issue by resetting the remixlhk cookie to
55 # the first one manually.
56 for header, cookies in url_handle.headers.items():
57 if header.lower() != 'set-cookie':
58 continue
59 if sys.version_info[0] >= 3:
60 cookies = cookies.encode('iso-8859-1')
61 cookies = cookies.decode('utf-8')
62 remixlhk = re.search(r'remixlhk=(.+?);.*?\bdomain=(.+?)(?:[,;]|$)', cookies)
63 if remixlhk:
64 value, domain = remixlhk.groups()
65 self._set_cookie(domain, 'remixlhk', value)
66 break
67
68 login_page = self._download_webpage(
69 'https://login.vk.com/?act=login', None,
70 note='Logging in',
71 data=urlencode_postdata(login_form))
72
73 if re.search(r'onLoginFailed', login_page):
74 raise ExtractorError(
75 'Unable to login, incorrect username and/or password', expected=True)
76
77 def _real_initialize(self):
78 self._login()
79
80
81 class VKIE(VKBaseIE):
82 IE_NAME = 'vk'
83 IE_DESC = 'VK'
84 _VALID_URL = r'''(?x)
85 https?://
86 (?:
87 (?:
88 (?:(?:m|new)\.)?vk\.com/video_|
89 (?:www\.)?daxab.com/
90 )
91 ext\.php\?(?P<embed_query>.*?\boid=(?P<oid>-?\d+).*?\bid=(?P<id>\d+).*)|
92 (?:
93 (?:(?:m|new)\.)?vk\.com/(?:.+?\?.*?z=)?video|
94 (?:www\.)?daxab.com/embed/
95 )
96 (?P<videoid>-?\d+_\d+)(?:.*\blist=(?P<list_id>[\da-f]+))?
97 )
98 '''
99 _TESTS = [
100 {
101 'url': 'http://vk.com/videos-77521?z=video-77521_162222515%2Fclub77521',
102 'md5': '7babad3b85ea2e91948005b1b8b0cb84',
103 'info_dict': {
104 'id': '162222515',
105 'ext': 'mp4',
106 'title': 'ProtivoGunz - Хуёвая песня',
107 'uploader': 're:(?:Noize MC|Alexander Ilyashenko).*',
108 'duration': 195,
109 'timestamp': 1329060660,
110 'upload_date': '20120212',
111 'view_count': int,
112 },
113 },
114 {
115 'url': 'http://vk.com/video205387401_165548505',
116 'md5': '6c0aeb2e90396ba97035b9cbde548700',
117 'info_dict': {
118 'id': '165548505',
119 'ext': 'mp4',
120 'uploader': 'Tom Cruise',
121 'title': 'No name',
122 'duration': 9,
123 'timestamp': 1374374880,
124 'upload_date': '20130721',
125 'view_count': int,
126 }
127 },
128 {
129 'note': 'Embedded video',
130 'url': 'http://vk.com/video_ext.php?oid=32194266&id=162925554&hash=7d8c2e0d5e05aeaa&hd=1',
131 'md5': 'c7ce8f1f87bec05b3de07fdeafe21a0a',
132 'info_dict': {
133 'id': '162925554',
134 'ext': 'mp4',
135 'uploader': 'Vladimir Gavrin',
136 'title': 'Lin Dan',
137 'duration': 101,
138 'upload_date': '20120730',
139 'view_count': int,
140 },
141 'skip': 'This video has been removed from public access.',
142 },
143 {
144 # VIDEO NOW REMOVED
145 # please update if you find a video whose URL follows the same pattern
146 'url': 'http://vk.com/video-8871596_164049491',
147 'md5': 'a590bcaf3d543576c9bd162812387666',
148 'note': 'Only available for registered users',
149 'info_dict': {
150 'id': '164049491',
151 'ext': 'mp4',
152 'uploader': 'Триллеры',
153 'title': '► Бойцовский клуб / Fight Club 1999 [HD 720]',
154 'duration': 8352,
155 'upload_date': '20121218',
156 'view_count': int,
157 },
158 'skip': 'Requires vk account credentials',
159 },
160 {
161 'url': 'http://vk.com/hd_kino_mania?z=video-43215063_168067957%2F15c66b9b533119788d',
162 'md5': '4d7a5ef8cf114dfa09577e57b2993202',
163 'info_dict': {
164 'id': '168067957',
165 'ext': 'mp4',
166 'uploader': 'Киномания - лучшее из мира кино',
167 'title': ' ',
168 'duration': 7291,
169 'upload_date': '20140328',
170 },
171 'skip': 'Requires vk account credentials',
172 },
173 {
174 'url': 'http://m.vk.com/video-43215063_169084319?list=125c627d1aa1cebb83&from=wall-43215063_2566540',
175 'md5': '0c45586baa71b7cb1d0784ee3f4e00a6',
176 'note': 'ivi.ru embed',
177 'info_dict': {
178 'id': '60690',
179 'ext': 'mp4',
180 'title': 'Книга Илая',
181 'duration': 6771,
182 'upload_date': '20140626',
183 'view_count': int,
184 },
185 'skip': 'Only works from Russia',
186 },
187 {
188 # video (removed?) only available with list id
189 'url': 'https://vk.com/video30481095_171201961?list=8764ae2d21f14088d4',
190 'md5': '091287af5402239a1051c37ec7b92913',
191 'info_dict': {
192 'id': '171201961',
193 'ext': 'mp4',
194 'title': 'ТюменцевВВ_09.07.2015',
195 'uploader': 'Anton Ivanov',
196 'duration': 109,
197 'upload_date': '20150709',
198 'view_count': int,
199 },
200 'skip': 'Removed',
201 },
202 {
203 # youtube embed
204 'url': 'https://vk.com/video276849682_170681728',
205 'info_dict': {
206 'id': 'V3K4mi0SYkc',
207 'ext': 'webm',
208 'title': "DSWD Awards 'Children's Joy Foundation, Inc.' Certificate of Registration and License to Operate",
209 'description': 'md5:d9903938abdc74c738af77f527ca0596',
210 'duration': 178,
211 'upload_date': '20130116',
212 'uploader': "Children's Joy Foundation",
213 'uploader_id': 'thecjf',
214 'view_count': int,
215 },
216 },
217 {
218 # dailymotion embed
219 'url': 'https://vk.com/video-37468416_456239855',
220 'info_dict': {
221 'id': 'k3lz2cmXyRuJQSjGHUv',
222 'ext': 'mp4',
223 'title': 'md5:d52606645c20b0ddbb21655adaa4f56f',
224 'description': 'md5:c651358f03c56f1150b555c26d90a0fd',
225 'uploader': 'AniLibria.Tv',
226 'upload_date': '20160914',
227 'uploader_id': 'x1p5vl5',
228 'timestamp': 1473877246,
229 },
230 'params': {
231 'skip_download': True,
232 },
233 },
234 {
235 # video key is extra_data not url\d+
236 'url': 'http://vk.com/video-110305615_171782105',
237 'md5': 'e13fcda136f99764872e739d13fac1d1',
238 'info_dict': {
239 'id': '171782105',
240 'ext': 'mp4',
241 'title': 'S-Dance, репетиции к The way show',
242 'uploader': 'THE WAY SHOW | 17 апреля',
243 'timestamp': 1454870100,
244 'upload_date': '20160207',
245 'view_count': int,
246 },
247 },
248 {
249 # finished live stream, postlive_mp4
250 'url': 'https://vk.com/videos-387766?z=video-387766_456242764%2Fpl_-387766_-2',
251 'md5': '90d22d051fccbbe9becfccc615be6791',
252 'info_dict': {
253 'id': '456242764',
254 'ext': 'mp4',
255 'title': 'ИгроМир 2016 — день 1',
256 'uploader': 'Игромания',
257 'duration': 5239,
258 'view_count': int,
259 },
260 },
261 {
262 # live stream, hls and rtmp links, most likely already finished live
263 # stream by the time you are reading this comment
264 'url': 'https://vk.com/video-140332_456239111',
265 'only_matching': True,
266 },
267 {
268 # removed video, just testing that we match the pattern
269 'url': 'http://vk.com/feed?z=video-43215063_166094326%2Fbb50cacd3177146d7a',
270 'only_matching': True,
271 },
272 {
273 # age restricted video, requires vk account credentials
274 'url': 'https://vk.com/video205387401_164765225',
275 'only_matching': True,
276 },
277 {
278 # pladform embed
279 'url': 'https://vk.com/video-76116461_171554880',
280 'only_matching': True,
281 },
282 {
283 'url': 'http://new.vk.com/video205387401_165548505',
284 'only_matching': True,
285 },
286 {
287 # This video is no longer available, because its author has been blocked.
288 'url': 'https://vk.com/video-10639516_456240611',
289 'only_matching': True,
290 }
291 ]
292
293 def _real_extract(self, url):
294 mobj = re.match(self._VALID_URL, url)
295 video_id = mobj.group('videoid')
296
297 if video_id:
298 info_url = 'https://vk.com/al_video.php?act=show&al=1&module=video&video=%s' % video_id
299 # Some videos (removed?) can only be downloaded with list id specified
300 list_id = mobj.group('list_id')
301 if list_id:
302 info_url += '&list=%s' % list_id
303 else:
304 info_url = 'http://vk.com/video_ext.php?' + mobj.group('embed_query')
305 video_id = '%s_%s' % (mobj.group('oid'), mobj.group('id'))
306
307 info_page = self._download_webpage(info_url, video_id)
308
309 error_message = self._html_search_regex(
310 [r'(?s)<!><div[^>]+class="video_layer_message"[^>]*>(.+?)</div>',
311 r'(?s)<div[^>]+id="video_ext_msg"[^>]*>(.+?)</div>'],
312 info_page, 'error message', default=None)
313 if error_message:
314 raise ExtractorError(error_message, expected=True)
315
316 if re.search(r'<!>/login\.php\?.*\bact=security_check', info_page):
317 raise ExtractorError(
318 'You are trying to log in from an unusual location. You should confirm ownership at vk.com to log in with this IP.',
319 expected=True)
320
321 ERROR_COPYRIGHT = 'Video %s has been removed from public access due to rightholder complaint.'
322
323 ERRORS = {
324 r'>Видеозапись .*? была изъята из публичного доступа в связи с обращением правообладателя.<':
325 ERROR_COPYRIGHT,
326
327 r'>The video .*? was removed from public access by request of the copyright holder.<':
328 ERROR_COPYRIGHT,
329
330 r'<!>Please log in or <':
331 'Video %s is only available for registered users, '
332 'use --username and --password options to provide account credentials.',
333
334 r'<!>Unknown error':
335 'Video %s does not exist.',
336
337 r'<!>Видео временно недоступно':
338 'Video %s is temporarily unavailable.',
339
340 r'<!>Access denied':
341 'Access denied to video %s.',
342
343 r'<!>Видеозапись недоступна, так как её автор был заблокирован.':
344 'Video %s is no longer available, because its author has been blocked.',
345
346 r'<!>This video is no longer available, because its author has been blocked.':
347 'Video %s is no longer available, because its author has been blocked.',
348 }
349
350 for error_re, error_msg in ERRORS.items():
351 if re.search(error_re, info_page):
352 raise ExtractorError(error_msg % video_id, expected=True)
353
354 youtube_url = YoutubeIE._extract_url(info_page)
355 if youtube_url:
356 return self.url_result(youtube_url, ie=YoutubeIE.ie_key())
357
358 vimeo_url = VimeoIE._extract_url(url, info_page)
359 if vimeo_url is not None:
360 return self.url_result(vimeo_url)
361
362 pladform_url = PladformIE._extract_url(info_page)
363 if pladform_url:
364 return self.url_result(pladform_url)
365
366 m_rutube = re.search(
367 r'\ssrc="((?:https?:)?//rutube\.ru\\?/(?:video|play)\\?/embed(?:.*?))\\?"', info_page)
368 if m_rutube is not None:
369 rutube_url = self._proto_relative_url(
370 m_rutube.group(1).replace('\\', ''))
371 return self.url_result(rutube_url)
372
373 dailymotion_urls = DailymotionIE._extract_urls(info_page)
374 if dailymotion_urls:
375 return self.url_result(dailymotion_urls[0], DailymotionIE.ie_key())
376
377 m_opts = re.search(r'(?s)var\s+opts\s*=\s*({.+?});', info_page)
378 if m_opts:
379 m_opts_url = re.search(r"url\s*:\s*'((?!/\b)[^']+)", m_opts.group(1))
380 if m_opts_url:
381 opts_url = m_opts_url.group(1)
382 if opts_url.startswith('//'):
383 opts_url = 'http:' + opts_url
384 return self.url_result(opts_url)
385
386 # vars does not look to be served anymore since 24.10.2016
387 data = self._parse_json(
388 self._search_regex(
389 r'var\s+vars\s*=\s*({.+?});', info_page, 'vars', default='{}'),
390 video_id, fatal=False)
391
392 # <!json> is served instead
393 if not data:
394 data = self._parse_json(
395 self._search_regex(
396 r'<!json>\s*({.+?})\s*<!>', info_page, 'json', default='{}'),
397 video_id)
398 if data:
399 data = data['player']['params'][0]
400
401 if not data:
402 data = self._parse_json(
403 self._search_regex(
404 r'var\s+playerParams\s*=\s*({.+?})\s*;\s*\n', info_page,
405 'player params'),
406 video_id)['params'][0]
407
408 title = unescapeHTML(data['md_title'])
409
410 # 2 = live
411 # 3 = post live (finished live)
412 is_live = data.get('live') == 2
413 if is_live:
414 title = self._live_title(title)
415
416 timestamp = unified_timestamp(self._html_search_regex(
417 r'class=["\']mv_info_date[^>]+>([^<]+)(?:<|from)', info_page,
418 'upload date', fatal=False))
419
420 view_count = str_to_int(self._search_regex(
421 r'class=["\']mv_views_count[^>]+>\s*([\d,.]+)',
422 info_page, 'view count', default=None))
423
424 formats = []
425 for format_id, format_url in data.items():
426 if not isinstance(format_url, compat_str) or not format_url.startswith(('http', '//', 'rtmp')):
427 continue
428 if (format_id.startswith(('url', 'cache')) or
429 format_id in ('extra_data', 'live_mp4', 'postlive_mp4')):
430 height = int_or_none(self._search_regex(
431 r'^(?:url|cache)(\d+)', format_id, 'height', default=None))
432 formats.append({
433 'format_id': format_id,
434 'url': format_url,
435 'height': height,
436 })
437 elif format_id == 'hls':
438 formats.extend(self._extract_m3u8_formats(
439 format_url, video_id, 'mp4', 'm3u8_native',
440 m3u8_id=format_id, fatal=False, live=is_live))
441 elif format_id == 'rtmp':
442 formats.append({
443 'format_id': format_id,
444 'url': format_url,
445 'ext': 'flv',
446 })
447 self._sort_formats(formats)
448
449 return {
450 'id': compat_str(data.get('vid') or video_id),
451 'formats': formats,
452 'title': title,
453 'thumbnail': data.get('jpg'),
454 'uploader': data.get('md_author'),
455 'duration': data.get('duration'),
456 'timestamp': timestamp,
457 'view_count': view_count,
458 'is_live': is_live,
459 }
460
461
462 class VKUserVideosIE(VKBaseIE):
463 IE_NAME = 'vk:uservideos'
464 IE_DESC = "VK - User's Videos"
465 _VALID_URL = r'https?://(?:(?:m|new)\.)?vk\.com/videos(?P<id>-?[0-9]+)(?!\?.*\bz=video)(?:[/?#&]|$)'
466 _TEMPLATE_URL = 'https://vk.com/videos'
467 _TESTS = [{
468 'url': 'http://vk.com/videos205387401',
469 'info_dict': {
470 'id': '205387401',
471 'title': "Tom Cruise's Videos",
472 },
473 'playlist_mincount': 4,
474 }, {
475 'url': 'http://vk.com/videos-77521',
476 'only_matching': True,
477 }, {
478 'url': 'http://vk.com/videos-97664626?section=all',
479 'only_matching': True,
480 }, {
481 'url': 'http://m.vk.com/videos205387401',
482 'only_matching': True,
483 }, {
484 'url': 'http://new.vk.com/videos205387401',
485 'only_matching': True,
486 }]
487
488 def _real_extract(self, url):
489 page_id = self._match_id(url)
490
491 webpage = self._download_webpage(url, page_id)
492
493 entries = [
494 self.url_result(
495 'http://vk.com/video' + video_id, 'VK', video_id=video_id)
496 for video_id in orderedSet(re.findall(r'href="/video(-?[0-9_]+)"', webpage))]
497
498 title = unescapeHTML(self._search_regex(
499 r'<title>\s*([^<]+?)\s+\|\s+\d+\s+videos',
500 webpage, 'title', default=page_id))
501
502 return self.playlist_result(entries, page_id, title)
503
504
505 class VKWallPostIE(VKBaseIE):
506 IE_NAME = 'vk:wallpost'
507 _VALID_URL = r'https?://(?:(?:(?:(?:m|new)\.)?vk\.com/(?:[^?]+\?.*\bw=)?wall(?P<id>-?\d+_\d+)))'
508 _TESTS = [{
509 # public page URL, audio playlist
510 'url': 'https://vk.com/bs.official?w=wall-23538238_35',
511 'info_dict': {
512 'id': '23538238_35',
513 'title': 'Black Shadow - Wall post 23538238_35',
514 'description': 'md5:3f84b9c4f9ef499731cf1ced9998cc0c',
515 },
516 'playlist': [{
517 'md5': '5ba93864ec5b85f7ce19a9af4af080f6',
518 'info_dict': {
519 'id': '135220665_111806521',
520 'ext': 'mp3',
521 'title': 'Black Shadow - Слепое Верование',
522 'duration': 370,
523 'uploader': 'Black Shadow',
524 'artist': 'Black Shadow',
525 'track': 'Слепое Верование',
526 },
527 }, {
528 'md5': '4cc7e804579122b17ea95af7834c9233',
529 'info_dict': {
530 'id': '135220665_111802303',
531 'ext': 'mp3',
532 'title': 'Black Shadow - Война - Негасимое Бездны Пламя!',
533 'duration': 423,
534 'uploader': 'Black Shadow',
535 'artist': 'Black Shadow',
536 'track': 'Война - Негасимое Бездны Пламя!',
537 },
538 'params': {
539 'skip_download': True,
540 },
541 }],
542 'params': {
543 'usenetrc': True,
544 },
545 'skip': 'Requires vk account credentials',
546 }, {
547 # single YouTube embed, no leading -
548 'url': 'https://vk.com/wall85155021_6319',
549 'info_dict': {
550 'id': '85155021_6319',
551 'title': 'Sergey Gorbunov - Wall post 85155021_6319',
552 },
553 'playlist_count': 1,
554 'params': {
555 'usenetrc': True,
556 },
557 'skip': 'Requires vk account credentials',
558 }, {
559 # wall page URL
560 'url': 'https://vk.com/wall-23538238_35',
561 'only_matching': True,
562 }, {
563 # mobile wall page URL
564 'url': 'https://m.vk.com/wall-23538238_35',
565 'only_matching': True,
566 }]
567
568 def _real_extract(self, url):
569 post_id = self._match_id(url)
570
571 wall_url = 'https://vk.com/wall%s' % post_id
572
573 post_id = remove_start(post_id, '-')
574
575 webpage = self._download_webpage(wall_url, post_id)
576
577 error = self._html_search_regex(
578 r'>Error</div>\s*<div[^>]+class=["\']body["\'][^>]*>([^<]+)',
579 webpage, 'error', default=None)
580 if error:
581 raise ExtractorError('VK said: %s' % error, expected=True)
582
583 description = clean_html(get_element_by_class('wall_post_text', webpage))
584 uploader = clean_html(get_element_by_class('author', webpage))
585 thumbnail = self._og_search_thumbnail(webpage)
586
587 entries = []
588
589 audio_ids = re.findall(r'data-full-id=["\'](\d+_\d+)', webpage)
590 if audio_ids:
591 al_audio = self._download_webpage(
592 'https://vk.com/al_audio.php', post_id,
593 note='Downloading audio info', fatal=False,
594 data=urlencode_postdata({
595 'act': 'reload_audio',
596 'al': '1',
597 'ids': ','.join(audio_ids)
598 }))
599 if al_audio:
600 Audio = collections.namedtuple(
601 'Audio', ['id', 'user_id', 'url', 'track', 'artist', 'duration'])
602 audios = self._parse_json(
603 self._search_regex(
604 r'<!json>(.+?)<!>', al_audio, 'audios', default='[]'),
605 post_id, fatal=False, transform_source=unescapeHTML)
606 if isinstance(audios, list):
607 for audio in audios:
608 a = Audio._make(audio[:6])
609 entries.append({
610 'id': '%s_%s' % (a.user_id, a.id),
611 'url': a.url,
612 'title': '%s - %s' % (a.artist, a.track) if a.artist and a.track else a.id,
613 'thumbnail': thumbnail,
614 'duration': a.duration,
615 'uploader': uploader,
616 'artist': a.artist,
617 'track': a.track,
618 })
619
620 for video in re.finditer(
621 r'<a[^>]+href=(["\'])(?P<url>/video(?:-?[\d_]+).*?)\1', webpage):
622 entries.append(self.url_result(
623 compat_urlparse.urljoin(url, video.group('url')), VKIE.ie_key()))
624
625 title = 'Wall post %s' % post_id
626
627 return self.playlist_result(
628 orderedSet(entries), post_id,
629 '%s - %s' % (uploader, title) if uploader else title,
630 description)