]> jfr.im git - yt-dlp.git/blob - youtube_dl/extractor/vk.py
1072550f1945344176b16cec11b2441e462ac926
[yt-dlp.git] / 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 compat_urlparse
10 from ..utils import (
11 clean_html,
12 ExtractorError,
13 get_element_by_class,
14 int_or_none,
15 orderedSet,
16 remove_start,
17 str_or_none,
18 str_to_int,
19 unescapeHTML,
20 unified_timestamp,
21 url_or_none,
22 urlencode_postdata,
23 )
24 from .dailymotion import DailymotionIE
25 from .pladform import PladformIE
26 from .vimeo import VimeoIE
27 from .youtube import YoutubeIE
28
29
30 class VKBaseIE(InfoExtractor):
31 _NETRC_MACHINE = 'vk'
32
33 def _login(self):
34 username, password = self._get_login_info()
35 if username is None:
36 return
37
38 login_page, url_handle = self._download_webpage_handle(
39 'https://vk.com', None, 'Downloading login page')
40
41 login_form = self._hidden_inputs(login_page)
42
43 login_form.update({
44 'email': username.encode('cp1251'),
45 'pass': password.encode('cp1251'),
46 })
47
48 # https://new.vk.com/ serves two same remixlhk cookies in Set-Cookie header
49 # and expects the first one to be set rather than second (see
50 # https://github.com/ytdl-org/youtube-dl/issues/9841#issuecomment-227871201).
51 # As of RFC6265 the newer one cookie should be set into cookie store
52 # what actually happens.
53 # We will workaround this VK issue by resetting the remixlhk cookie to
54 # the first one manually.
55 for header, cookies in url_handle.headers.items():
56 if header.lower() != 'set-cookie':
57 continue
58 if sys.version_info[0] >= 3:
59 cookies = cookies.encode('iso-8859-1')
60 cookies = cookies.decode('utf-8')
61 remixlhk = re.search(r'remixlhk=(.+?);.*?\bdomain=(.+?)(?:[,;]|$)', cookies)
62 if remixlhk:
63 value, domain = remixlhk.groups()
64 self._set_cookie(domain, 'remixlhk', value)
65 break
66
67 login_page = self._download_webpage(
68 'https://login.vk.com/?act=login', None,
69 note='Logging in',
70 data=urlencode_postdata(login_form))
71
72 if re.search(r'onLoginFailed', login_page):
73 raise ExtractorError(
74 'Unable to login, incorrect username and/or password', expected=True)
75
76 def _real_initialize(self):
77 self._login()
78
79
80 class VKIE(VKBaseIE):
81 IE_NAME = 'vk'
82 IE_DESC = 'VK'
83 _VALID_URL = r'''(?x)
84 https?://
85 (?:
86 (?:
87 (?:(?:m|new)\.)?vk\.com/video_|
88 (?:www\.)?daxab.com/
89 )
90 ext\.php\?(?P<embed_query>.*?\boid=(?P<oid>-?\d+).*?\bid=(?P<id>\d+).*)|
91 (?:
92 (?:(?:m|new)\.)?vk\.com/(?:.+?\?.*?z=)?video|
93 (?:www\.)?daxab.com/embed/
94 )
95 (?P<videoid>-?\d+_\d+)(?:.*\blist=(?P<list_id>[\da-f]+))?
96 )
97 '''
98 _TESTS = [
99 {
100 'url': 'http://vk.com/videos-77521?z=video-77521_162222515%2Fclub77521',
101 'md5': '7babad3b85ea2e91948005b1b8b0cb84',
102 'info_dict': {
103 'id': '-77521_162222515',
104 'ext': 'mp4',
105 'title': 'ProtivoGunz - Хуёвая песня',
106 'uploader': 're:(?:Noize MC|Alexander Ilyashenko).*',
107 'uploader_id': '-77521',
108 'duration': 195,
109 'timestamp': 1329049880,
110 'upload_date': '20120212',
111 },
112 },
113 {
114 'url': 'http://vk.com/video205387401_165548505',
115 'md5': '6c0aeb2e90396ba97035b9cbde548700',
116 'info_dict': {
117 'id': '205387401_165548505',
118 'ext': 'mp4',
119 'title': 'No name',
120 'uploader': 'Tom Cruise',
121 'uploader_id': '205387401',
122 'duration': 9,
123 'timestamp': 1374364108,
124 'upload_date': '20130720',
125 }
126 },
127 {
128 'note': 'Embedded video',
129 'url': 'http://vk.com/video_ext.php?oid=32194266&id=162925554&hash=7d8c2e0d5e05aeaa&hd=1',
130 'md5': 'c7ce8f1f87bec05b3de07fdeafe21a0a',
131 'info_dict': {
132 'id': '32194266_162925554',
133 'ext': 'mp4',
134 'uploader': 'Vladimir Gavrin',
135 'title': 'Lin Dan',
136 'duration': 101,
137 'upload_date': '20120730',
138 'view_count': int,
139 },
140 'skip': 'This video has been removed from public access.',
141 },
142 {
143 # VIDEO NOW REMOVED
144 # please update if you find a video whose URL follows the same pattern
145 'url': 'http://vk.com/video-8871596_164049491',
146 'md5': 'a590bcaf3d543576c9bd162812387666',
147 'note': 'Only available for registered users',
148 'info_dict': {
149 'id': '-8871596_164049491',
150 'ext': 'mp4',
151 'uploader': 'Триллеры',
152 'title': '► Бойцовский клуб / Fight Club 1999 [HD 720]',
153 'duration': 8352,
154 'upload_date': '20121218',
155 'view_count': int,
156 },
157 'skip': 'Requires vk account credentials',
158 },
159 {
160 'url': 'http://vk.com/hd_kino_mania?z=video-43215063_168067957%2F15c66b9b533119788d',
161 'md5': '4d7a5ef8cf114dfa09577e57b2993202',
162 'info_dict': {
163 'id': '-43215063_168067957',
164 'ext': 'mp4',
165 'uploader': 'Киномания - лучшее из мира кино',
166 'title': ' ',
167 'duration': 7291,
168 'upload_date': '20140328',
169 },
170 'skip': 'Requires vk account credentials',
171 },
172 {
173 'url': 'http://m.vk.com/video-43215063_169084319?list=125c627d1aa1cebb83&from=wall-43215063_2566540',
174 'md5': '0c45586baa71b7cb1d0784ee3f4e00a6',
175 'note': 'ivi.ru embed',
176 'info_dict': {
177 'id': '-43215063_169084319',
178 'ext': 'mp4',
179 'title': 'Книга Илая',
180 'duration': 6771,
181 'upload_date': '20140626',
182 'view_count': int,
183 },
184 'skip': 'Only works from Russia',
185 },
186 {
187 # video (removed?) only available with list id
188 'url': 'https://vk.com/video30481095_171201961?list=8764ae2d21f14088d4',
189 'md5': '091287af5402239a1051c37ec7b92913',
190 'info_dict': {
191 'id': '30481095_171201961',
192 'ext': 'mp4',
193 'title': 'ТюменцевВВ_09.07.2015',
194 'uploader': 'Anton Ivanov',
195 'duration': 109,
196 'upload_date': '20150709',
197 'view_count': int,
198 },
199 'skip': 'Removed',
200 },
201 {
202 # youtube embed
203 'url': 'https://vk.com/video276849682_170681728',
204 'info_dict': {
205 'id': 'V3K4mi0SYkc',
206 'ext': 'mp4',
207 'title': "DSWD Awards 'Children's Joy Foundation, Inc.' Certificate of Registration and License to Operate",
208 'description': 'md5:bf9c26cfa4acdfb146362682edd3827a',
209 'duration': 178,
210 'upload_date': '20130116',
211 'uploader': "Children's Joy Foundation Inc.",
212 'uploader_id': 'thecjf',
213 'view_count': int,
214 },
215 },
216 {
217 # dailymotion embed
218 'url': 'https://vk.com/video-37468416_456239855',
219 'info_dict': {
220 'id': 'k3lz2cmXyRuJQSjGHUv',
221 'ext': 'mp4',
222 'title': 'md5:d52606645c20b0ddbb21655adaa4f56f',
223 # TODO: fix test by fixing dailymotion description extraction
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': '-110305615_171782105',
240 'ext': 'mp4',
241 'title': 'S-Dance, репетиции к The way show',
242 'uploader': 'THE WAY SHOW | 17 апреля',
243 'uploader_id': '-110305615',
244 'timestamp': 1454859345,
245 'upload_date': '20160207',
246 },
247 'params': {
248 'skip_download': True,
249 },
250 },
251 {
252 # finished live stream, postlive_mp4
253 'url': 'https://vk.com/videos-387766?z=video-387766_456242764%2Fpl_-387766_-2',
254 'info_dict': {
255 'id': '-387766_456242764',
256 'ext': 'mp4',
257 'title': 'ИгроМир 2016 День 1 — Игромания Утром',
258 'uploader': 'Игромания',
259 'duration': 5239,
260 # TODO: use act=show to extract view_count
261 # 'view_count': int,
262 'upload_date': '20160929',
263 'uploader_id': '-387766',
264 'timestamp': 1475137527,
265 },
266 },
267 {
268 # live stream, hls and rtmp links, most likely already finished live
269 # stream by the time you are reading this comment
270 'url': 'https://vk.com/video-140332_456239111',
271 'only_matching': True,
272 },
273 {
274 # removed video, just testing that we match the pattern
275 'url': 'http://vk.com/feed?z=video-43215063_166094326%2Fbb50cacd3177146d7a',
276 'only_matching': True,
277 },
278 {
279 # age restricted video, requires vk account credentials
280 'url': 'https://vk.com/video205387401_164765225',
281 'only_matching': True,
282 },
283 {
284 # pladform embed
285 'url': 'https://vk.com/video-76116461_171554880',
286 'only_matching': True,
287 },
288 {
289 'url': 'http://new.vk.com/video205387401_165548505',
290 'only_matching': True,
291 },
292 {
293 # This video is no longer available, because its author has been blocked.
294 'url': 'https://vk.com/video-10639516_456240611',
295 'only_matching': True,
296 },
297 {
298 # The video is not available in your region.
299 'url': 'https://vk.com/video-51812607_171445436',
300 'only_matching': True,
301 }]
302
303 def _real_extract(self, url):
304 mobj = re.match(self._VALID_URL, url)
305 video_id = mobj.group('videoid')
306
307 if video_id:
308 info_url = 'https://vk.com/al_video.php?act=show_inline&al=1&video=' + video_id
309 # Some videos (removed?) can only be downloaded with list id specified
310 list_id = mobj.group('list_id')
311 if list_id:
312 info_url += '&list=%s' % list_id
313 else:
314 info_url = 'http://vk.com/video_ext.php?' + mobj.group('embed_query')
315 video_id = '%s_%s' % (mobj.group('oid'), mobj.group('id'))
316
317 info_page = self._download_webpage(info_url, video_id)
318
319 error_message = self._html_search_regex(
320 [r'(?s)<!><div[^>]+class="video_layer_message"[^>]*>(.+?)</div>',
321 r'(?s)<div[^>]+id="video_ext_msg"[^>]*>(.+?)</div>'],
322 info_page, 'error message', default=None)
323 if error_message:
324 raise ExtractorError(error_message, expected=True)
325
326 if re.search(r'<!>/login\.php\?.*\bact=security_check', info_page):
327 raise ExtractorError(
328 'You are trying to log in from an unusual location. You should confirm ownership at vk.com to log in with this IP.',
329 expected=True)
330
331 ERROR_COPYRIGHT = 'Video %s has been removed from public access due to rightholder complaint.'
332
333 ERRORS = {
334 r'>Видеозапись .*? была изъята из публичного доступа в связи с обращением правообладателя.<':
335 ERROR_COPYRIGHT,
336
337 r'>The video .*? was removed from public access by request of the copyright holder.<':
338 ERROR_COPYRIGHT,
339
340 r'<!>Please log in or <':
341 'Video %s is only available for registered users, '
342 'use --username and --password options to provide account credentials.',
343
344 r'<!>Unknown error':
345 'Video %s does not exist.',
346
347 r'<!>Видео временно недоступно':
348 'Video %s is temporarily unavailable.',
349
350 r'<!>Access denied':
351 'Access denied to video %s.',
352
353 r'<!>Видеозапись недоступна, так как её автор был заблокирован.':
354 'Video %s is no longer available, because its author has been blocked.',
355
356 r'<!>This video is no longer available, because its author has been blocked.':
357 'Video %s is no longer available, because its author has been blocked.',
358
359 r'<!>This video is no longer available, because it has been deleted.':
360 'Video %s is no longer available, because it has been deleted.',
361
362 r'<!>The video .+? is not available in your region.':
363 'Video %s is not available in your region.',
364 }
365
366 for error_re, error_msg in ERRORS.items():
367 if re.search(error_re, info_page):
368 raise ExtractorError(error_msg % video_id, expected=True)
369
370 youtube_url = YoutubeIE._extract_url(info_page)
371 if youtube_url:
372 return self.url_result(youtube_url, ie=YoutubeIE.ie_key())
373
374 vimeo_url = VimeoIE._extract_url(url, info_page)
375 if vimeo_url is not None:
376 return self.url_result(vimeo_url)
377
378 pladform_url = PladformIE._extract_url(info_page)
379 if pladform_url:
380 return self.url_result(pladform_url)
381
382 m_rutube = re.search(
383 r'\ssrc="((?:https?:)?//rutube\.ru\\?/(?:video|play)\\?/embed(?:.*?))\\?"', info_page)
384 if m_rutube is not None:
385 rutube_url = self._proto_relative_url(
386 m_rutube.group(1).replace('\\', ''))
387 return self.url_result(rutube_url)
388
389 dailymotion_urls = DailymotionIE._extract_urls(info_page)
390 if dailymotion_urls:
391 return self.url_result(dailymotion_urls[0], DailymotionIE.ie_key())
392
393 m_opts = re.search(r'(?s)var\s+opts\s*=\s*({.+?});', info_page)
394 if m_opts:
395 m_opts_url = re.search(r"url\s*:\s*'((?!/\b)[^']+)", m_opts.group(1))
396 if m_opts_url:
397 opts_url = m_opts_url.group(1)
398 if opts_url.startswith('//'):
399 opts_url = 'http:' + opts_url
400 return self.url_result(opts_url)
401
402 # vars does not look to be served anymore since 24.10.2016
403 data = self._parse_json(
404 self._search_regex(
405 r'var\s+vars\s*=\s*({.+?});', info_page, 'vars', default='{}'),
406 video_id, fatal=False)
407
408 # <!json> is served instead
409 if not data:
410 data = self._parse_json(
411 self._search_regex(
412 [r'<!json>\s*({.+?})\s*<!>', r'<!json>\s*({.+})'],
413 info_page, 'json', default='{}'),
414 video_id)
415 if data:
416 data = data['player']['params'][0]
417
418 if not data:
419 data = self._parse_json(
420 self._search_regex(
421 r'var\s+playerParams\s*=\s*({.+?})\s*;\s*\n', info_page,
422 'player params'),
423 video_id)['params'][0]
424
425 title = unescapeHTML(data['md_title'])
426
427 # 2 = live
428 # 3 = post live (finished live)
429 is_live = data.get('live') == 2
430 if is_live:
431 title = self._live_title(title)
432
433 timestamp = unified_timestamp(self._html_search_regex(
434 r'class=["\']mv_info_date[^>]+>([^<]+)(?:<|from)', info_page,
435 'upload date', default=None)) or int_or_none(data.get('date'))
436
437 view_count = str_to_int(self._search_regex(
438 r'class=["\']mv_views_count[^>]+>\s*([\d,.]+)',
439 info_page, 'view count', default=None))
440
441 formats = []
442 for format_id, format_url in data.items():
443 format_url = url_or_none(format_url)
444 if not format_url or not format_url.startswith(('http', '//', 'rtmp')):
445 continue
446 if (format_id.startswith(('url', 'cache')) or
447 format_id in ('extra_data', 'live_mp4', 'postlive_mp4')):
448 height = int_or_none(self._search_regex(
449 r'^(?:url|cache)(\d+)', format_id, 'height', default=None))
450 formats.append({
451 'format_id': format_id,
452 'url': format_url,
453 'height': height,
454 })
455 elif format_id == 'hls':
456 formats.extend(self._extract_m3u8_formats(
457 format_url, video_id, 'mp4', 'm3u8_native',
458 m3u8_id=format_id, fatal=False, live=is_live))
459 elif format_id == 'rtmp':
460 formats.append({
461 'format_id': format_id,
462 'url': format_url,
463 'ext': 'flv',
464 })
465 self._sort_formats(formats)
466
467 return {
468 'id': video_id,
469 'formats': formats,
470 'title': title,
471 'thumbnail': data.get('jpg'),
472 'uploader': data.get('md_author'),
473 'uploader_id': str_or_none(data.get('author_id')),
474 'duration': data.get('duration'),
475 'timestamp': timestamp,
476 'view_count': view_count,
477 'like_count': int_or_none(data.get('liked')),
478 'dislike_count': int_or_none(data.get('nolikes')),
479 'is_live': is_live,
480 }
481
482
483 class VKUserVideosIE(VKBaseIE):
484 IE_NAME = 'vk:uservideos'
485 IE_DESC = "VK - User's Videos"
486 _VALID_URL = r'https?://(?:(?:m|new)\.)?vk\.com/videos(?P<id>-?[0-9]+)(?!\?.*\bz=video)(?:[/?#&]|$)'
487 _TEMPLATE_URL = 'https://vk.com/videos'
488 _TESTS = [{
489 'url': 'http://vk.com/videos205387401',
490 'info_dict': {
491 'id': '205387401',
492 'title': "Tom Cruise's Videos",
493 },
494 'playlist_mincount': 4,
495 }, {
496 'url': 'http://vk.com/videos-77521',
497 'only_matching': True,
498 }, {
499 'url': 'http://vk.com/videos-97664626?section=all',
500 'only_matching': True,
501 }, {
502 'url': 'http://m.vk.com/videos205387401',
503 'only_matching': True,
504 }, {
505 'url': 'http://new.vk.com/videos205387401',
506 'only_matching': True,
507 }]
508
509 def _real_extract(self, url):
510 page_id = self._match_id(url)
511
512 webpage = self._download_webpage(url, page_id)
513
514 entries = [
515 self.url_result(
516 'http://vk.com/video' + video_id, 'VK', video_id=video_id)
517 for video_id in orderedSet(re.findall(r'href="/video(-?[0-9_]+)"', webpage))]
518
519 title = unescapeHTML(self._search_regex(
520 r'<title>\s*([^<]+?)\s+\|\s+\d+\s+videos',
521 webpage, 'title', default=page_id))
522
523 return self.playlist_result(entries, page_id, title)
524
525
526 class VKWallPostIE(VKBaseIE):
527 IE_NAME = 'vk:wallpost'
528 _VALID_URL = r'https?://(?:(?:(?:(?:m|new)\.)?vk\.com/(?:[^?]+\?.*\bw=)?wall(?P<id>-?\d+_\d+)))'
529 _TESTS = [{
530 # public page URL, audio playlist
531 'url': 'https://vk.com/bs.official?w=wall-23538238_35',
532 'info_dict': {
533 'id': '23538238_35',
534 'title': 'Black Shadow - Wall post 23538238_35',
535 'description': 'md5:3f84b9c4f9ef499731cf1ced9998cc0c',
536 },
537 'playlist': [{
538 'md5': '5ba93864ec5b85f7ce19a9af4af080f6',
539 'info_dict': {
540 'id': '135220665_111806521',
541 'ext': 'mp3',
542 'title': 'Black Shadow - Слепое Верование',
543 'duration': 370,
544 'uploader': 'Black Shadow',
545 'artist': 'Black Shadow',
546 'track': 'Слепое Верование',
547 },
548 }, {
549 'md5': '4cc7e804579122b17ea95af7834c9233',
550 'info_dict': {
551 'id': '135220665_111802303',
552 'ext': 'mp3',
553 'title': 'Black Shadow - Война - Негасимое Бездны Пламя!',
554 'duration': 423,
555 'uploader': 'Black Shadow',
556 'artist': 'Black Shadow',
557 'track': 'Война - Негасимое Бездны Пламя!',
558 },
559 'params': {
560 'skip_download': True,
561 },
562 }],
563 'params': {
564 'usenetrc': True,
565 },
566 'skip': 'Requires vk account credentials',
567 }, {
568 # single YouTube embed, no leading -
569 'url': 'https://vk.com/wall85155021_6319',
570 'info_dict': {
571 'id': '85155021_6319',
572 'title': 'Sergey Gorbunov - Wall post 85155021_6319',
573 },
574 'playlist_count': 1,
575 'params': {
576 'usenetrc': True,
577 },
578 'skip': 'Requires vk account credentials',
579 }, {
580 # wall page URL
581 'url': 'https://vk.com/wall-23538238_35',
582 'only_matching': True,
583 }, {
584 # mobile wall page URL
585 'url': 'https://m.vk.com/wall-23538238_35',
586 'only_matching': True,
587 }]
588
589 def _real_extract(self, url):
590 post_id = self._match_id(url)
591
592 wall_url = 'https://vk.com/wall%s' % post_id
593
594 post_id = remove_start(post_id, '-')
595
596 webpage = self._download_webpage(wall_url, post_id)
597
598 error = self._html_search_regex(
599 r'>Error</div>\s*<div[^>]+class=["\']body["\'][^>]*>([^<]+)',
600 webpage, 'error', default=None)
601 if error:
602 raise ExtractorError('VK said: %s' % error, expected=True)
603
604 description = clean_html(get_element_by_class('wall_post_text', webpage))
605 uploader = clean_html(get_element_by_class('author', webpage))
606 thumbnail = self._og_search_thumbnail(webpage)
607
608 entries = []
609
610 audio_ids = re.findall(r'data-full-id=["\'](\d+_\d+)', webpage)
611 if audio_ids:
612 al_audio = self._download_webpage(
613 'https://vk.com/al_audio.php', post_id,
614 note='Downloading audio info', fatal=False,
615 data=urlencode_postdata({
616 'act': 'reload_audio',
617 'al': '1',
618 'ids': ','.join(audio_ids)
619 }))
620 if al_audio:
621 Audio = collections.namedtuple(
622 'Audio', ['id', 'user_id', 'url', 'track', 'artist', 'duration'])
623 audios = self._parse_json(
624 self._search_regex(
625 r'<!json>(.+?)<!>', al_audio, 'audios', default='[]'),
626 post_id, fatal=False, transform_source=unescapeHTML)
627 if isinstance(audios, list):
628 for audio in audios:
629 a = Audio._make(audio[:6])
630 entries.append({
631 'id': '%s_%s' % (a.user_id, a.id),
632 'url': a.url,
633 'title': '%s - %s' % (a.artist, a.track) if a.artist and a.track else a.id,
634 'thumbnail': thumbnail,
635 'duration': a.duration,
636 'uploader': uploader,
637 'artist': a.artist,
638 'track': a.track,
639 })
640
641 for video in re.finditer(
642 r'<a[^>]+href=(["\'])(?P<url>/video(?:-?[\d_]+).*?)\1', webpage):
643 entries.append(self.url_result(
644 compat_urlparse.urljoin(url, video.group('url')), VKIE.ie_key()))
645
646 title = 'Wall post %s' % post_id
647
648 return self.playlist_result(
649 orderedSet(entries), post_id,
650 '%s - %s' % (uploader, title) if uploader else title,
651 description)