]> Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/vk.py
New upstream version 2019.06.08
[youtubedl] / youtube_dl / extractor / vk.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 import collections
5 import re
6
7 from .common import InfoExtractor
8 from ..compat import compat_urlparse
9 from ..utils import (
10 clean_html,
11 ExtractorError,
12 get_element_by_class,
13 int_or_none,
14 orderedSet,
15 remove_start,
16 str_or_none,
17 str_to_int,
18 unescapeHTML,
19 unified_timestamp,
20 url_or_none,
21 urlencode_postdata,
22 )
23 from .dailymotion import DailymotionIE
24 from .pladform import PladformIE
25 from .vimeo import VimeoIE
26 from .youtube import YoutubeIE
27
28
29 class VKBaseIE(InfoExtractor):
30 _NETRC_MACHINE = 'vk'
31
32 def _login(self):
33 username, password = self._get_login_info()
34 if username is None:
35 return
36
37 login_page, url_handle = self._download_webpage_handle(
38 'https://vk.com', None, 'Downloading login page')
39
40 login_form = self._hidden_inputs(login_page)
41
42 login_form.update({
43 'email': username.encode('cp1251'),
44 'pass': password.encode('cp1251'),
45 })
46
47 # vk serves two same remixlhk cookies in Set-Cookie header and expects
48 # first one to be actually set
49 self._apply_first_set_cookie_header(url_handle, 'remixlhk')
50
51 login_page = self._download_webpage(
52 'https://login.vk.com/?act=login', None,
53 note='Logging in',
54 data=urlencode_postdata(login_form))
55
56 if re.search(r'onLoginFailed', login_page):
57 raise ExtractorError(
58 'Unable to login, incorrect username and/or password', expected=True)
59
60 def _real_initialize(self):
61 self._login()
62
63
64 class VKIE(VKBaseIE):
65 IE_NAME = 'vk'
66 IE_DESC = 'VK'
67 _VALID_URL = r'''(?x)
68 https?://
69 (?:
70 (?:
71 (?:(?:m|new)\.)?vk\.com/video_|
72 (?:www\.)?daxab.com/
73 )
74 ext\.php\?(?P<embed_query>.*?\boid=(?P<oid>-?\d+).*?\bid=(?P<id>\d+).*)|
75 (?:
76 (?:(?:m|new)\.)?vk\.com/(?:.+?\?.*?z=)?video|
77 (?:www\.)?daxab.com/embed/
78 )
79 (?P<videoid>-?\d+_\d+)(?:.*\blist=(?P<list_id>[\da-f]+))?
80 )
81 '''
82 _TESTS = [
83 {
84 'url': 'http://vk.com/videos-77521?z=video-77521_162222515%2Fclub77521',
85 'md5': '7babad3b85ea2e91948005b1b8b0cb84',
86 'info_dict': {
87 'id': '-77521_162222515',
88 'ext': 'mp4',
89 'title': 'ProtivoGunz - Хуёвая песня',
90 'uploader': 're:(?:Noize MC|Alexander Ilyashenko).*',
91 'uploader_id': '-77521',
92 'duration': 195,
93 'timestamp': 1329049880,
94 'upload_date': '20120212',
95 },
96 },
97 {
98 'url': 'http://vk.com/video205387401_165548505',
99 'md5': '6c0aeb2e90396ba97035b9cbde548700',
100 'info_dict': {
101 'id': '205387401_165548505',
102 'ext': 'mp4',
103 'title': 'No name',
104 'uploader': 'Tom Cruise',
105 'uploader_id': '205387401',
106 'duration': 9,
107 'timestamp': 1374364108,
108 'upload_date': '20130720',
109 }
110 },
111 {
112 'note': 'Embedded video',
113 'url': 'http://vk.com/video_ext.php?oid=32194266&id=162925554&hash=7d8c2e0d5e05aeaa&hd=1',
114 'md5': 'c7ce8f1f87bec05b3de07fdeafe21a0a',
115 'info_dict': {
116 'id': '32194266_162925554',
117 'ext': 'mp4',
118 'uploader': 'Vladimir Gavrin',
119 'title': 'Lin Dan',
120 'duration': 101,
121 'upload_date': '20120730',
122 'view_count': int,
123 },
124 'skip': 'This video has been removed from public access.',
125 },
126 {
127 # VIDEO NOW REMOVED
128 # please update if you find a video whose URL follows the same pattern
129 'url': 'http://vk.com/video-8871596_164049491',
130 'md5': 'a590bcaf3d543576c9bd162812387666',
131 'note': 'Only available for registered users',
132 'info_dict': {
133 'id': '-8871596_164049491',
134 'ext': 'mp4',
135 'uploader': 'Триллеры',
136 'title': '► Бойцовский клуб / Fight Club 1999 [HD 720]',
137 'duration': 8352,
138 'upload_date': '20121218',
139 'view_count': int,
140 },
141 'skip': 'Requires vk account credentials',
142 },
143 {
144 'url': 'http://vk.com/hd_kino_mania?z=video-43215063_168067957%2F15c66b9b533119788d',
145 'md5': '4d7a5ef8cf114dfa09577e57b2993202',
146 'info_dict': {
147 'id': '-43215063_168067957',
148 'ext': 'mp4',
149 'uploader': 'Киномания - лучшее из мира кино',
150 'title': ' ',
151 'duration': 7291,
152 'upload_date': '20140328',
153 },
154 'skip': 'Requires vk account credentials',
155 },
156 {
157 'url': 'http://m.vk.com/video-43215063_169084319?list=125c627d1aa1cebb83&from=wall-43215063_2566540',
158 'md5': '0c45586baa71b7cb1d0784ee3f4e00a6',
159 'note': 'ivi.ru embed',
160 'info_dict': {
161 'id': '-43215063_169084319',
162 'ext': 'mp4',
163 'title': 'Книга Илая',
164 'duration': 6771,
165 'upload_date': '20140626',
166 'view_count': int,
167 },
168 'skip': 'Only works from Russia',
169 },
170 {
171 # video (removed?) only available with list id
172 'url': 'https://vk.com/video30481095_171201961?list=8764ae2d21f14088d4',
173 'md5': '091287af5402239a1051c37ec7b92913',
174 'info_dict': {
175 'id': '30481095_171201961',
176 'ext': 'mp4',
177 'title': 'ТюменцевВВ_09.07.2015',
178 'uploader': 'Anton Ivanov',
179 'duration': 109,
180 'upload_date': '20150709',
181 'view_count': int,
182 },
183 'skip': 'Removed',
184 },
185 {
186 # youtube embed
187 'url': 'https://vk.com/video276849682_170681728',
188 'info_dict': {
189 'id': 'V3K4mi0SYkc',
190 'ext': 'mp4',
191 'title': "DSWD Awards 'Children's Joy Foundation, Inc.' Certificate of Registration and License to Operate",
192 'description': 'md5:bf9c26cfa4acdfb146362682edd3827a',
193 'duration': 178,
194 'upload_date': '20130116',
195 'uploader': "Children's Joy Foundation Inc.",
196 'uploader_id': 'thecjf',
197 'view_count': int,
198 },
199 },
200 {
201 # dailymotion embed
202 'url': 'https://vk.com/video-37468416_456239855',
203 'info_dict': {
204 'id': 'k3lz2cmXyRuJQSjGHUv',
205 'ext': 'mp4',
206 'title': 'md5:d52606645c20b0ddbb21655adaa4f56f',
207 # TODO: fix test by fixing dailymotion description extraction
208 'description': 'md5:c651358f03c56f1150b555c26d90a0fd',
209 'uploader': 'AniLibria.Tv',
210 'upload_date': '20160914',
211 'uploader_id': 'x1p5vl5',
212 'timestamp': 1473877246,
213 },
214 'params': {
215 'skip_download': True,
216 },
217 },
218 {
219 # video key is extra_data not url\d+
220 'url': 'http://vk.com/video-110305615_171782105',
221 'md5': 'e13fcda136f99764872e739d13fac1d1',
222 'info_dict': {
223 'id': '-110305615_171782105',
224 'ext': 'mp4',
225 'title': 'S-Dance, репетиции к The way show',
226 'uploader': 'THE WAY SHOW | 17 апреля',
227 'uploader_id': '-110305615',
228 'timestamp': 1454859345,
229 'upload_date': '20160207',
230 },
231 'params': {
232 'skip_download': True,
233 },
234 },
235 {
236 # finished live stream, postlive_mp4
237 'url': 'https://vk.com/videos-387766?z=video-387766_456242764%2Fpl_-387766_-2',
238 'info_dict': {
239 'id': '-387766_456242764',
240 'ext': 'mp4',
241 'title': 'ИгроМир 2016 День 1 — Игромания Утром',
242 'uploader': 'Игромания',
243 'duration': 5239,
244 # TODO: use act=show to extract view_count
245 # 'view_count': int,
246 'upload_date': '20160929',
247 'uploader_id': '-387766',
248 'timestamp': 1475137527,
249 },
250 },
251 {
252 # live stream, hls and rtmp links, most likely already finished live
253 # stream by the time you are reading this comment
254 'url': 'https://vk.com/video-140332_456239111',
255 'only_matching': True,
256 },
257 {
258 # removed video, just testing that we match the pattern
259 'url': 'http://vk.com/feed?z=video-43215063_166094326%2Fbb50cacd3177146d7a',
260 'only_matching': True,
261 },
262 {
263 # age restricted video, requires vk account credentials
264 'url': 'https://vk.com/video205387401_164765225',
265 'only_matching': True,
266 },
267 {
268 # pladform embed
269 'url': 'https://vk.com/video-76116461_171554880',
270 'only_matching': True,
271 },
272 {
273 'url': 'http://new.vk.com/video205387401_165548505',
274 'only_matching': True,
275 },
276 {
277 # This video is no longer available, because its author has been blocked.
278 'url': 'https://vk.com/video-10639516_456240611',
279 'only_matching': True,
280 },
281 {
282 # The video is not available in your region.
283 'url': 'https://vk.com/video-51812607_171445436',
284 'only_matching': True,
285 }]
286
287 def _real_extract(self, url):
288 mobj = re.match(self._VALID_URL, url)
289 video_id = mobj.group('videoid')
290
291 if video_id:
292 info_url = 'https://vk.com/al_video.php?act=show_inline&al=1&video=' + video_id
293 # Some videos (removed?) can only be downloaded with list id specified
294 list_id = mobj.group('list_id')
295 if list_id:
296 info_url += '&list=%s' % list_id
297 else:
298 info_url = 'http://vk.com/video_ext.php?' + mobj.group('embed_query')
299 video_id = '%s_%s' % (mobj.group('oid'), mobj.group('id'))
300
301 info_page = self._download_webpage(info_url, video_id)
302
303 error_message = self._html_search_regex(
304 [r'(?s)<!><div[^>]+class="video_layer_message"[^>]*>(.+?)</div>',
305 r'(?s)<div[^>]+id="video_ext_msg"[^>]*>(.+?)</div>'],
306 info_page, 'error message', default=None)
307 if error_message:
308 raise ExtractorError(error_message, expected=True)
309
310 if re.search(r'<!>/login\.php\?.*\bact=security_check', info_page):
311 raise ExtractorError(
312 'You are trying to log in from an unusual location. You should confirm ownership at vk.com to log in with this IP.',
313 expected=True)
314
315 ERROR_COPYRIGHT = 'Video %s has been removed from public access due to rightholder complaint.'
316
317 ERRORS = {
318 r'>Видеозапись .*? была изъята из публичного доступа в связи с обращением правообладателя.<':
319 ERROR_COPYRIGHT,
320
321 r'>The video .*? was removed from public access by request of the copyright holder.<':
322 ERROR_COPYRIGHT,
323
324 r'<!>Please log in or <':
325 'Video %s is only available for registered users, '
326 'use --username and --password options to provide account credentials.',
327
328 r'<!>Unknown error':
329 'Video %s does not exist.',
330
331 r'<!>Видео временно недоступно':
332 'Video %s is temporarily unavailable.',
333
334 r'<!>Access denied':
335 'Access denied to video %s.',
336
337 r'<!>Видеозапись недоступна, так как её автор был заблокирован.':
338 'Video %s is no longer available, because its author has been blocked.',
339
340 r'<!>This video is no longer available, because its author has been blocked.':
341 'Video %s is no longer available, because its author has been blocked.',
342
343 r'<!>This video is no longer available, because it has been deleted.':
344 'Video %s is no longer available, because it has been deleted.',
345
346 r'<!>The video .+? is not available in your region.':
347 'Video %s is not available in your region.',
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*<!>', r'<!json>\s*({.+})'],
397 info_page, 'json', default='{}'),
398 video_id)
399 if data:
400 data = data['player']['params'][0]
401
402 if not data:
403 data = self._parse_json(
404 self._search_regex(
405 r'var\s+playerParams\s*=\s*({.+?})\s*;\s*\n', info_page,
406 'player params'),
407 video_id)['params'][0]
408
409 title = unescapeHTML(data['md_title'])
410
411 # 2 = live
412 # 3 = post live (finished live)
413 is_live = data.get('live') == 2
414 if is_live:
415 title = self._live_title(title)
416
417 timestamp = unified_timestamp(self._html_search_regex(
418 r'class=["\']mv_info_date[^>]+>([^<]+)(?:<|from)', info_page,
419 'upload date', default=None)) or int_or_none(data.get('date'))
420
421 view_count = str_to_int(self._search_regex(
422 r'class=["\']mv_views_count[^>]+>\s*([\d,.]+)',
423 info_page, 'view count', default=None))
424
425 formats = []
426 for format_id, format_url in data.items():
427 format_url = url_or_none(format_url)
428 if not format_url or not format_url.startswith(('http', '//', 'rtmp')):
429 continue
430 if (format_id.startswith(('url', 'cache'))
431 or format_id in ('extra_data', 'live_mp4', 'postlive_mp4')):
432 height = int_or_none(self._search_regex(
433 r'^(?:url|cache)(\d+)', format_id, 'height', default=None))
434 formats.append({
435 'format_id': format_id,
436 'url': format_url,
437 'height': height,
438 })
439 elif format_id == 'hls':
440 formats.extend(self._extract_m3u8_formats(
441 format_url, video_id, 'mp4', 'm3u8_native',
442 m3u8_id=format_id, fatal=False, live=is_live))
443 elif format_id == 'rtmp':
444 formats.append({
445 'format_id': format_id,
446 'url': format_url,
447 'ext': 'flv',
448 })
449 self._sort_formats(formats)
450
451 return {
452 'id': video_id,
453 'formats': formats,
454 'title': title,
455 'thumbnail': data.get('jpg'),
456 'uploader': data.get('md_author'),
457 'uploader_id': str_or_none(data.get('author_id')),
458 'duration': data.get('duration'),
459 'timestamp': timestamp,
460 'view_count': view_count,
461 'like_count': int_or_none(data.get('liked')),
462 'dislike_count': int_or_none(data.get('nolikes')),
463 'is_live': is_live,
464 }
465
466
467 class VKUserVideosIE(VKBaseIE):
468 IE_NAME = 'vk:uservideos'
469 IE_DESC = "VK - User's Videos"
470 _VALID_URL = r'https?://(?:(?:m|new)\.)?vk\.com/videos(?P<id>-?[0-9]+)(?!\?.*\bz=video)(?:[/?#&]|$)'
471 _TEMPLATE_URL = 'https://vk.com/videos'
472 _TESTS = [{
473 'url': 'http://vk.com/videos205387401',
474 'info_dict': {
475 'id': '205387401',
476 'title': "Tom Cruise's Videos",
477 },
478 'playlist_mincount': 4,
479 }, {
480 'url': 'http://vk.com/videos-77521',
481 'only_matching': True,
482 }, {
483 'url': 'http://vk.com/videos-97664626?section=all',
484 'only_matching': True,
485 }, {
486 'url': 'http://m.vk.com/videos205387401',
487 'only_matching': True,
488 }, {
489 'url': 'http://new.vk.com/videos205387401',
490 'only_matching': True,
491 }]
492
493 def _real_extract(self, url):
494 page_id = self._match_id(url)
495
496 webpage = self._download_webpage(url, page_id)
497
498 entries = [
499 self.url_result(
500 'http://vk.com/video' + video_id, 'VK', video_id=video_id)
501 for video_id in orderedSet(re.findall(r'href="/video(-?[0-9_]+)"', webpage))]
502
503 title = unescapeHTML(self._search_regex(
504 r'<title>\s*([^<]+?)\s+\|\s+\d+\s+videos',
505 webpage, 'title', default=page_id))
506
507 return self.playlist_result(entries, page_id, title)
508
509
510 class VKWallPostIE(VKBaseIE):
511 IE_NAME = 'vk:wallpost'
512 _VALID_URL = r'https?://(?:(?:(?:(?:m|new)\.)?vk\.com/(?:[^?]+\?.*\bw=)?wall(?P<id>-?\d+_\d+)))'
513 _TESTS = [{
514 # public page URL, audio playlist
515 'url': 'https://vk.com/bs.official?w=wall-23538238_35',
516 'info_dict': {
517 'id': '23538238_35',
518 'title': 'Black Shadow - Wall post 23538238_35',
519 'description': 'md5:3f84b9c4f9ef499731cf1ced9998cc0c',
520 },
521 'playlist': [{
522 'md5': '5ba93864ec5b85f7ce19a9af4af080f6',
523 'info_dict': {
524 'id': '135220665_111806521',
525 'ext': 'mp3',
526 'title': 'Black Shadow - Слепое Верование',
527 'duration': 370,
528 'uploader': 'Black Shadow',
529 'artist': 'Black Shadow',
530 'track': 'Слепое Верование',
531 },
532 }, {
533 'md5': '4cc7e804579122b17ea95af7834c9233',
534 'info_dict': {
535 'id': '135220665_111802303',
536 'ext': 'mp3',
537 'title': 'Black Shadow - Война - Негасимое Бездны Пламя!',
538 'duration': 423,
539 'uploader': 'Black Shadow',
540 'artist': 'Black Shadow',
541 'track': 'Война - Негасимое Бездны Пламя!',
542 },
543 'params': {
544 'skip_download': True,
545 },
546 }],
547 'params': {
548 'usenetrc': True,
549 },
550 'skip': 'Requires vk account credentials',
551 }, {
552 # single YouTube embed, no leading -
553 'url': 'https://vk.com/wall85155021_6319',
554 'info_dict': {
555 'id': '85155021_6319',
556 'title': 'Sergey Gorbunov - Wall post 85155021_6319',
557 },
558 'playlist_count': 1,
559 'params': {
560 'usenetrc': True,
561 },
562 'skip': 'Requires vk account credentials',
563 }, {
564 # wall page URL
565 'url': 'https://vk.com/wall-23538238_35',
566 'only_matching': True,
567 }, {
568 # mobile wall page URL
569 'url': 'https://m.vk.com/wall-23538238_35',
570 'only_matching': True,
571 }]
572
573 def _real_extract(self, url):
574 post_id = self._match_id(url)
575
576 wall_url = 'https://vk.com/wall%s' % post_id
577
578 post_id = remove_start(post_id, '-')
579
580 webpage = self._download_webpage(wall_url, post_id)
581
582 error = self._html_search_regex(
583 r'>Error</div>\s*<div[^>]+class=["\']body["\'][^>]*>([^<]+)',
584 webpage, 'error', default=None)
585 if error:
586 raise ExtractorError('VK said: %s' % error, expected=True)
587
588 description = clean_html(get_element_by_class('wall_post_text', webpage))
589 uploader = clean_html(get_element_by_class('author', webpage))
590 thumbnail = self._og_search_thumbnail(webpage)
591
592 entries = []
593
594 audio_ids = re.findall(r'data-full-id=["\'](\d+_\d+)', webpage)
595 if audio_ids:
596 al_audio = self._download_webpage(
597 'https://vk.com/al_audio.php', post_id,
598 note='Downloading audio info', fatal=False,
599 data=urlencode_postdata({
600 'act': 'reload_audio',
601 'al': '1',
602 'ids': ','.join(audio_ids)
603 }))
604 if al_audio:
605 Audio = collections.namedtuple(
606 'Audio', ['id', 'user_id', 'url', 'track', 'artist', 'duration'])
607 audios = self._parse_json(
608 self._search_regex(
609 r'<!json>(.+?)<!>', al_audio, 'audios', default='[]'),
610 post_id, fatal=False, transform_source=unescapeHTML)
611 if isinstance(audios, list):
612 for audio in audios:
613 a = Audio._make(audio[:6])
614 entries.append({
615 'id': '%s_%s' % (a.user_id, a.id),
616 'url': a.url,
617 'title': '%s - %s' % (a.artist, a.track) if a.artist and a.track else a.id,
618 'thumbnail': thumbnail,
619 'duration': a.duration,
620 'uploader': uploader,
621 'artist': a.artist,
622 'track': a.track,
623 })
624
625 for video in re.finditer(
626 r'<a[^>]+href=(["\'])(?P<url>/video(?:-?[\d_]+).*?)\1', webpage):
627 entries.append(self.url_result(
628 compat_urlparse.urljoin(url, video.group('url')), VKIE.ie_key()))
629
630 title = 'Wall post %s' % post_id
631
632 return self.playlist_result(
633 orderedSet(entries), post_id,
634 '%s - %s' % (uploader, title) if uploader else title,
635 description)