]> Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/vk.py
ef8b9bcb7b610c25925ed2725a8bb43d8f3a6d1c
[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_or_none,
21 str_to_int,
22 unescapeHTML,
23 unified_timestamp,
24 url_or_none,
25 urlencode_postdata,
26 )
27 from .dailymotion import DailymotionIE
28 from .pladform import PladformIE
29 from .vimeo import VimeoIE
30 from .youtube import YoutubeIE
31
32
33 class VKBaseIE(InfoExtractor):
34 _NETRC_MACHINE = 'vk'
35
36 def _login(self):
37 username, password = self._get_login_info()
38 if username is None:
39 return
40
41 login_page, url_handle = self._download_webpage_handle(
42 'https://vk.com', None, 'Downloading login page')
43
44 login_form = self._hidden_inputs(login_page)
45
46 login_form.update({
47 'email': username.encode('cp1251'),
48 'pass': password.encode('cp1251'),
49 })
50
51 # https://new.vk.com/ serves two same remixlhk cookies in Set-Cookie header
52 # and expects the first one to be set rather than second (see
53 # https://github.com/rg3/youtube-dl/issues/9841#issuecomment-227871201).
54 # As of RFC6265 the newer one cookie should be set into cookie store
55 # what actually happens.
56 # We will workaround this VK issue by resetting the remixlhk cookie to
57 # the first one manually.
58 for header, cookies in url_handle.headers.items():
59 if header.lower() != 'set-cookie':
60 continue
61 if sys.version_info[0] >= 3:
62 cookies = cookies.encode('iso-8859-1')
63 cookies = cookies.decode('utf-8')
64 remixlhk = re.search(r'remixlhk=(.+?);.*?\bdomain=(.+?)(?:[,;]|$)', cookies)
65 if remixlhk:
66 value, domain = remixlhk.groups()
67 self._set_cookie(domain, 'remixlhk', value)
68 break
69
70 login_page = self._download_webpage(
71 'https://login.vk.com/?act=login', None,
72 note='Logging in',
73 data=urlencode_postdata(login_form))
74
75 if re.search(r'onLoginFailed', login_page):
76 raise ExtractorError(
77 'Unable to login, incorrect username and/or password', expected=True)
78
79 def _real_initialize(self):
80 self._login()
81
82
83 class VKIE(VKBaseIE):
84 IE_NAME = 'vk'
85 IE_DESC = 'VK'
86 _VALID_URL = r'''(?x)
87 https?://
88 (?:
89 (?:
90 (?:(?:m|new)\.)?vk\.com/video_|
91 (?:www\.)?daxab.com/
92 )
93 ext\.php\?(?P<embed_query>.*?\boid=(?P<oid>-?\d+).*?\bid=(?P<id>\d+).*)|
94 (?:
95 (?:(?:m|new)\.)?vk\.com/(?:.+?\?.*?z=)?video|
96 (?:www\.)?daxab.com/embed/
97 )
98 (?P<videoid>-?\d+_\d+)(?:.*\blist=(?P<list_id>[\da-f]+))?
99 )
100 '''
101 _TESTS = [
102 {
103 'url': 'http://vk.com/videos-77521?z=video-77521_162222515%2Fclub77521',
104 'md5': '7babad3b85ea2e91948005b1b8b0cb84',
105 'info_dict': {
106 'id': '162222515',
107 'ext': 'mp4',
108 'title': 'ProtivoGunz - Хуёвая песня',
109 'uploader': 're:(?:Noize MC|Alexander Ilyashenko).*',
110 'uploader_id': '-77521',
111 'duration': 195,
112 'timestamp': 1329049880,
113 'upload_date': '20120212',
114 },
115 },
116 {
117 'url': 'http://vk.com/video205387401_165548505',
118 'md5': '6c0aeb2e90396ba97035b9cbde548700',
119 'info_dict': {
120 'id': '165548505',
121 'ext': 'mp4',
122 'title': 'No name',
123 'uploader': 'Tom Cruise',
124 'uploader_id': '205387401',
125 'duration': 9,
126 'timestamp': 1374364108,
127 'upload_date': '20130720',
128 }
129 },
130 {
131 'note': 'Embedded video',
132 'url': 'http://vk.com/video_ext.php?oid=32194266&id=162925554&hash=7d8c2e0d5e05aeaa&hd=1',
133 'md5': 'c7ce8f1f87bec05b3de07fdeafe21a0a',
134 'info_dict': {
135 'id': '162925554',
136 'ext': 'mp4',
137 'uploader': 'Vladimir Gavrin',
138 'title': 'Lin Dan',
139 'duration': 101,
140 'upload_date': '20120730',
141 'view_count': int,
142 },
143 'skip': 'This video has been removed from public access.',
144 },
145 {
146 # VIDEO NOW REMOVED
147 # please update if you find a video whose URL follows the same pattern
148 'url': 'http://vk.com/video-8871596_164049491',
149 'md5': 'a590bcaf3d543576c9bd162812387666',
150 'note': 'Only available for registered users',
151 'info_dict': {
152 'id': '164049491',
153 'ext': 'mp4',
154 'uploader': 'Триллеры',
155 'title': '► Бойцовский клуб / Fight Club 1999 [HD 720]',
156 'duration': 8352,
157 'upload_date': '20121218',
158 'view_count': int,
159 },
160 'skip': 'Requires vk account credentials',
161 },
162 {
163 'url': 'http://vk.com/hd_kino_mania?z=video-43215063_168067957%2F15c66b9b533119788d',
164 'md5': '4d7a5ef8cf114dfa09577e57b2993202',
165 'info_dict': {
166 'id': '168067957',
167 'ext': 'mp4',
168 'uploader': 'Киномания - лучшее из мира кино',
169 'title': ' ',
170 'duration': 7291,
171 'upload_date': '20140328',
172 },
173 'skip': 'Requires vk account credentials',
174 },
175 {
176 'url': 'http://m.vk.com/video-43215063_169084319?list=125c627d1aa1cebb83&from=wall-43215063_2566540',
177 'md5': '0c45586baa71b7cb1d0784ee3f4e00a6',
178 'note': 'ivi.ru embed',
179 'info_dict': {
180 'id': '60690',
181 'ext': 'mp4',
182 'title': 'Книга Илая',
183 'duration': 6771,
184 'upload_date': '20140626',
185 'view_count': int,
186 },
187 'skip': 'Only works from Russia',
188 },
189 {
190 # video (removed?) only available with list id
191 'url': 'https://vk.com/video30481095_171201961?list=8764ae2d21f14088d4',
192 'md5': '091287af5402239a1051c37ec7b92913',
193 'info_dict': {
194 'id': '171201961',
195 'ext': 'mp4',
196 'title': 'ТюменцевВВ_09.07.2015',
197 'uploader': 'Anton Ivanov',
198 'duration': 109,
199 'upload_date': '20150709',
200 'view_count': int,
201 },
202 'skip': 'Removed',
203 },
204 {
205 # youtube embed
206 'url': 'https://vk.com/video276849682_170681728',
207 'info_dict': {
208 'id': 'V3K4mi0SYkc',
209 'ext': 'webm',
210 'title': "DSWD Awards 'Children's Joy Foundation, Inc.' Certificate of Registration and License to Operate",
211 'description': 'md5:bf9c26cfa4acdfb146362682edd3827a',
212 'duration': 179,
213 'upload_date': '20130116',
214 'uploader': "Children's Joy Foundation Inc.",
215 'uploader_id': 'thecjf',
216 'view_count': int,
217 },
218 },
219 {
220 # dailymotion embed
221 'url': 'https://vk.com/video-37468416_456239855',
222 'info_dict': {
223 'id': 'k3lz2cmXyRuJQSjGHUv',
224 'ext': 'mp4',
225 'title': 'md5:d52606645c20b0ddbb21655adaa4f56f',
226 # TODO: fix test by fixing dailymotion description extraction
227 'description': 'md5:c651358f03c56f1150b555c26d90a0fd',
228 'uploader': 'AniLibria.Tv',
229 'upload_date': '20160914',
230 'uploader_id': 'x1p5vl5',
231 'timestamp': 1473877246,
232 },
233 'params': {
234 'skip_download': True,
235 },
236 },
237 {
238 # video key is extra_data not url\d+
239 'url': 'http://vk.com/video-110305615_171782105',
240 'md5': 'e13fcda136f99764872e739d13fac1d1',
241 'info_dict': {
242 'id': '171782105',
243 'ext': 'mp4',
244 'title': 'S-Dance, репетиции к The way show',
245 'uploader': 'THE WAY SHOW | 17 апреля',
246 'uploader_id': '-110305615',
247 'timestamp': 1454859345,
248 'upload_date': '20160207',
249 },
250 'params': {
251 'skip_download': True,
252 },
253 },
254 {
255 # finished live stream, postlive_mp4
256 'url': 'https://vk.com/videos-387766?z=video-387766_456242764%2Fpl_-387766_-2',
257 'md5': '90d22d051fccbbe9becfccc615be6791',
258 'info_dict': {
259 'id': '456242764',
260 'ext': 'mp4',
261 'title': 'ИгроМир 2016 — день 1',
262 'uploader': 'Игромания',
263 'duration': 5239,
264 'view_count': int,
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
299 def _real_extract(self, url):
300 mobj = re.match(self._VALID_URL, url)
301 video_id = mobj.group('videoid')
302
303 if video_id:
304 info_url = 'https://vk.com/al_video.php?act=show_inline&al=1&video=' + video_id
305 # Some videos (removed?) can only be downloaded with list id specified
306 list_id = mobj.group('list_id')
307 if list_id:
308 info_url += '&list=%s' % list_id
309 else:
310 info_url = 'http://vk.com/video_ext.php?' + mobj.group('embed_query')
311 video_id = '%s_%s' % (mobj.group('oid'), mobj.group('id'))
312
313 info_page = self._download_webpage(info_url, video_id)
314
315 error_message = self._html_search_regex(
316 [r'(?s)<!><div[^>]+class="video_layer_message"[^>]*>(.+?)</div>',
317 r'(?s)<div[^>]+id="video_ext_msg"[^>]*>(.+?)</div>'],
318 info_page, 'error message', default=None)
319 if error_message:
320 raise ExtractorError(error_message, expected=True)
321
322 if re.search(r'<!>/login\.php\?.*\bact=security_check', info_page):
323 raise ExtractorError(
324 'You are trying to log in from an unusual location. You should confirm ownership at vk.com to log in with this IP.',
325 expected=True)
326
327 ERROR_COPYRIGHT = 'Video %s has been removed from public access due to rightholder complaint.'
328
329 ERRORS = {
330 r'>Видеозапись .*? была изъята из публичного доступа в связи с обращением правообладателя.<':
331 ERROR_COPYRIGHT,
332
333 r'>The video .*? was removed from public access by request of the copyright holder.<':
334 ERROR_COPYRIGHT,
335
336 r'<!>Please log in or <':
337 'Video %s is only available for registered users, '
338 'use --username and --password options to provide account credentials.',
339
340 r'<!>Unknown error':
341 'Video %s does not exist.',
342
343 r'<!>Видео временно недоступно':
344 'Video %s is temporarily unavailable.',
345
346 r'<!>Access denied':
347 'Access denied to video %s.',
348
349 r'<!>Видеозапись недоступна, так как её автор был заблокирован.':
350 'Video %s is no longer available, because its author has been blocked.',
351
352 r'<!>This video is no longer available, because its author has been blocked.':
353 'Video %s is no longer available, because its author has been blocked.',
354
355 r'<!>This video is no longer available, because it has been deleted.':
356 'Video %s is no longer available, because it has been deleted.',
357 }
358
359 for error_re, error_msg in ERRORS.items():
360 if re.search(error_re, info_page):
361 raise ExtractorError(error_msg % video_id, expected=True)
362
363 youtube_url = YoutubeIE._extract_url(info_page)
364 if youtube_url:
365 return self.url_result(youtube_url, ie=YoutubeIE.ie_key())
366
367 vimeo_url = VimeoIE._extract_url(url, info_page)
368 if vimeo_url is not None:
369 return self.url_result(vimeo_url)
370
371 pladform_url = PladformIE._extract_url(info_page)
372 if pladform_url:
373 return self.url_result(pladform_url)
374
375 m_rutube = re.search(
376 r'\ssrc="((?:https?:)?//rutube\.ru\\?/(?:video|play)\\?/embed(?:.*?))\\?"', info_page)
377 if m_rutube is not None:
378 rutube_url = self._proto_relative_url(
379 m_rutube.group(1).replace('\\', ''))
380 return self.url_result(rutube_url)
381
382 dailymotion_urls = DailymotionIE._extract_urls(info_page)
383 if dailymotion_urls:
384 return self.url_result(dailymotion_urls[0], DailymotionIE.ie_key())
385
386 m_opts = re.search(r'(?s)var\s+opts\s*=\s*({.+?});', info_page)
387 if m_opts:
388 m_opts_url = re.search(r"url\s*:\s*'((?!/\b)[^']+)", m_opts.group(1))
389 if m_opts_url:
390 opts_url = m_opts_url.group(1)
391 if opts_url.startswith('//'):
392 opts_url = 'http:' + opts_url
393 return self.url_result(opts_url)
394
395 # vars does not look to be served anymore since 24.10.2016
396 data = self._parse_json(
397 self._search_regex(
398 r'var\s+vars\s*=\s*({.+?});', info_page, 'vars', default='{}'),
399 video_id, fatal=False)
400
401 # <!json> is served instead
402 if not data:
403 data = self._parse_json(
404 self._search_regex(
405 [r'<!json>\s*({.+?})\s*<!>', r'<!json>\s*({.+})'],
406 info_page, 'json', default='{}'),
407 video_id)
408 if data:
409 data = data['player']['params'][0]
410
411 if not data:
412 data = self._parse_json(
413 self._search_regex(
414 r'var\s+playerParams\s*=\s*({.+?})\s*;\s*\n', info_page,
415 'player params'),
416 video_id)['params'][0]
417
418 title = unescapeHTML(data['md_title'])
419
420 # 2 = live
421 # 3 = post live (finished live)
422 is_live = data.get('live') == 2
423 if is_live:
424 title = self._live_title(title)
425
426 timestamp = unified_timestamp(self._html_search_regex(
427 r'class=["\']mv_info_date[^>]+>([^<]+)(?:<|from)', info_page,
428 'upload date', default=None)) or int_or_none(data.get('date'))
429
430 view_count = str_to_int(self._search_regex(
431 r'class=["\']mv_views_count[^>]+>\s*([\d,.]+)',
432 info_page, 'view count', default=None))
433
434 formats = []
435 for format_id, format_url in data.items():
436 format_url = url_or_none(format_url)
437 if not format_url or not format_url.startswith(('http', '//', 'rtmp')):
438 continue
439 if (format_id.startswith(('url', 'cache')) or
440 format_id in ('extra_data', 'live_mp4', 'postlive_mp4')):
441 height = int_or_none(self._search_regex(
442 r'^(?:url|cache)(\d+)', format_id, 'height', default=None))
443 formats.append({
444 'format_id': format_id,
445 'url': format_url,
446 'height': height,
447 })
448 elif format_id == 'hls':
449 formats.extend(self._extract_m3u8_formats(
450 format_url, video_id, 'mp4', 'm3u8_native',
451 m3u8_id=format_id, fatal=False, live=is_live))
452 elif format_id == 'rtmp':
453 formats.append({
454 'format_id': format_id,
455 'url': format_url,
456 'ext': 'flv',
457 })
458 self._sort_formats(formats)
459
460 return {
461 'id': compat_str(data.get('vid') or video_id),
462 'formats': formats,
463 'title': title,
464 'thumbnail': data.get('jpg'),
465 'uploader': data.get('md_author'),
466 'uploader_id': str_or_none(data.get('author_id')),
467 'duration': data.get('duration'),
468 'timestamp': timestamp,
469 'view_count': view_count,
470 'like_count': int_or_none(data.get('liked')),
471 'dislike_count': int_or_none(data.get('nolikes')),
472 'is_live': is_live,
473 }
474
475
476 class VKUserVideosIE(VKBaseIE):
477 IE_NAME = 'vk:uservideos'
478 IE_DESC = "VK - User's Videos"
479 _VALID_URL = r'https?://(?:(?:m|new)\.)?vk\.com/videos(?P<id>-?[0-9]+)(?!\?.*\bz=video)(?:[/?#&]|$)'
480 _TEMPLATE_URL = 'https://vk.com/videos'
481 _TESTS = [{
482 'url': 'http://vk.com/videos205387401',
483 'info_dict': {
484 'id': '205387401',
485 'title': "Tom Cruise's Videos",
486 },
487 'playlist_mincount': 4,
488 }, {
489 'url': 'http://vk.com/videos-77521',
490 'only_matching': True,
491 }, {
492 'url': 'http://vk.com/videos-97664626?section=all',
493 'only_matching': True,
494 }, {
495 'url': 'http://m.vk.com/videos205387401',
496 'only_matching': True,
497 }, {
498 'url': 'http://new.vk.com/videos205387401',
499 'only_matching': True,
500 }]
501
502 def _real_extract(self, url):
503 page_id = self._match_id(url)
504
505 webpage = self._download_webpage(url, page_id)
506
507 entries = [
508 self.url_result(
509 'http://vk.com/video' + video_id, 'VK', video_id=video_id)
510 for video_id in orderedSet(re.findall(r'href="/video(-?[0-9_]+)"', webpage))]
511
512 title = unescapeHTML(self._search_regex(
513 r'<title>\s*([^<]+?)\s+\|\s+\d+\s+videos',
514 webpage, 'title', default=page_id))
515
516 return self.playlist_result(entries, page_id, title)
517
518
519 class VKWallPostIE(VKBaseIE):
520 IE_NAME = 'vk:wallpost'
521 _VALID_URL = r'https?://(?:(?:(?:(?:m|new)\.)?vk\.com/(?:[^?]+\?.*\bw=)?wall(?P<id>-?\d+_\d+)))'
522 _TESTS = [{
523 # public page URL, audio playlist
524 'url': 'https://vk.com/bs.official?w=wall-23538238_35',
525 'info_dict': {
526 'id': '23538238_35',
527 'title': 'Black Shadow - Wall post 23538238_35',
528 'description': 'md5:3f84b9c4f9ef499731cf1ced9998cc0c',
529 },
530 'playlist': [{
531 'md5': '5ba93864ec5b85f7ce19a9af4af080f6',
532 'info_dict': {
533 'id': '135220665_111806521',
534 'ext': 'mp3',
535 'title': 'Black Shadow - Слепое Верование',
536 'duration': 370,
537 'uploader': 'Black Shadow',
538 'artist': 'Black Shadow',
539 'track': 'Слепое Верование',
540 },
541 }, {
542 'md5': '4cc7e804579122b17ea95af7834c9233',
543 'info_dict': {
544 'id': '135220665_111802303',
545 'ext': 'mp3',
546 'title': 'Black Shadow - Война - Негасимое Бездны Пламя!',
547 'duration': 423,
548 'uploader': 'Black Shadow',
549 'artist': 'Black Shadow',
550 'track': 'Война - Негасимое Бездны Пламя!',
551 },
552 'params': {
553 'skip_download': True,
554 },
555 }],
556 'params': {
557 'usenetrc': True,
558 },
559 'skip': 'Requires vk account credentials',
560 }, {
561 # single YouTube embed, no leading -
562 'url': 'https://vk.com/wall85155021_6319',
563 'info_dict': {
564 'id': '85155021_6319',
565 'title': 'Sergey Gorbunov - Wall post 85155021_6319',
566 },
567 'playlist_count': 1,
568 'params': {
569 'usenetrc': True,
570 },
571 'skip': 'Requires vk account credentials',
572 }, {
573 # wall page URL
574 'url': 'https://vk.com/wall-23538238_35',
575 'only_matching': True,
576 }, {
577 # mobile wall page URL
578 'url': 'https://m.vk.com/wall-23538238_35',
579 'only_matching': True,
580 }]
581
582 def _real_extract(self, url):
583 post_id = self._match_id(url)
584
585 wall_url = 'https://vk.com/wall%s' % post_id
586
587 post_id = remove_start(post_id, '-')
588
589 webpage = self._download_webpage(wall_url, post_id)
590
591 error = self._html_search_regex(
592 r'>Error</div>\s*<div[^>]+class=["\']body["\'][^>]*>([^<]+)',
593 webpage, 'error', default=None)
594 if error:
595 raise ExtractorError('VK said: %s' % error, expected=True)
596
597 description = clean_html(get_element_by_class('wall_post_text', webpage))
598 uploader = clean_html(get_element_by_class('author', webpage))
599 thumbnail = self._og_search_thumbnail(webpage)
600
601 entries = []
602
603 audio_ids = re.findall(r'data-full-id=["\'](\d+_\d+)', webpage)
604 if audio_ids:
605 al_audio = self._download_webpage(
606 'https://vk.com/al_audio.php', post_id,
607 note='Downloading audio info', fatal=False,
608 data=urlencode_postdata({
609 'act': 'reload_audio',
610 'al': '1',
611 'ids': ','.join(audio_ids)
612 }))
613 if al_audio:
614 Audio = collections.namedtuple(
615 'Audio', ['id', 'user_id', 'url', 'track', 'artist', 'duration'])
616 audios = self._parse_json(
617 self._search_regex(
618 r'<!json>(.+?)<!>', al_audio, 'audios', default='[]'),
619 post_id, fatal=False, transform_source=unescapeHTML)
620 if isinstance(audios, list):
621 for audio in audios:
622 a = Audio._make(audio[:6])
623 entries.append({
624 'id': '%s_%s' % (a.user_id, a.id),
625 'url': a.url,
626 'title': '%s - %s' % (a.artist, a.track) if a.artist and a.track else a.id,
627 'thumbnail': thumbnail,
628 'duration': a.duration,
629 'uploader': uploader,
630 'artist': a.artist,
631 'track': a.track,
632 })
633
634 for video in re.finditer(
635 r'<a[^>]+href=(["\'])(?P<url>/video(?:-?[\d_]+).*?)\1', webpage):
636 entries.append(self.url_result(
637 compat_urlparse.urljoin(url, video.group('url')), VKIE.ie_key()))
638
639 title = 'Wall post %s' % post_id
640
641 return self.playlist_result(
642 orderedSet(entries), post_id,
643 '%s - %s' % (uploader, title) if uploader else title,
644 description)