yt_dlp/extractor/vk.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import collections
   5 import functools
   6 import re
   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     OnDemandPagedList,
  16     orderedSet,
  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 .odnoklassniki import OdnoklassnikiIE
  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         # vk serves two same remixlhk cookies in Set-Cookie header and expects
  50         # first one to be actually set
  51         self._apply_first_set_cookie_header(url_handle, 'remixlhk')
  52
  53         login_page = self._download_webpage(
  54             'https://vk.com/login', None,
  55             note='Logging in',
  56             data=urlencode_postdata(login_form))
  57
  58         if re.search(r'onLoginFailed', login_page):
  59             raise ExtractorError(
  60                 'Unable to login, incorrect username and/or password', expected=True)
  61
  62     def _real_initialize(self):
  63         self._login()
  64
  65     def _download_payload(self, path, video_id, data, fatal=True):
  66         data['al'] = 1
  67         code, payload = self._download_json(
  68             'https://vk.com/%s.php' % path, video_id,
  69             data=urlencode_postdata(data), fatal=fatal,
  70             headers={'X-Requested-With': 'XMLHttpRequest'})['payload']
  71         if code == '3':
  72             self.raise_login_required()
  73         elif code == '8':
  74             raise ExtractorError(clean_html(payload[0][1:-1]), expected=True)
  75         return payload
  76
  77
  78 class VKIE(VKBaseIE):
  79     IE_NAME = 'vk'
  80     IE_DESC = 'VK'
  81     _VALID_URL = r'''(?x)
  82                     https?://
  83                         (?:
  84                             (?:
  85                                 (?:(?:m|new)\.)?vk\.com/video_|
  86                                 (?:www\.)?daxab.com/
  87                             )
  88                             ext\.php\?(?P<embed_query>.*?\boid=(?P<oid>-?\d+).*?\bid=(?P<id>\d+).*)|
  89                             (?:
  90                                 (?:(?:m|new)\.)?vk\.com/(?:.+?\?.*?z=)?video|
  91                                 (?:www\.)?daxab.com/embed/
  92                             )
  93                             (?P<videoid>-?\d+_\d+)(?:.*\blist=(?P<list_id>[\da-f]+))?
  94                         )
  95                     '''
  96     _TESTS = [
  97         {
  98             'url': 'http://vk.com/videos-77521?z=video-77521_162222515%2Fclub77521',
  99             'md5': '7babad3b85ea2e91948005b1b8b0cb84',
 100             'info_dict': {
 101                 'id': '-77521_162222515',
 102                 'ext': 'mp4',
 103                 'title': 'ProtivoGunz - Хуёвая песня',
 104                 'uploader': 're:(?:Noize MC|Alexander Ilyashenko).*',
 105                 'uploader_id': '-77521',
 106                 'duration': 195,
 107                 'timestamp': 1329049880,
 108                 'upload_date': '20120212',
 109             },
 110         },
 111         {
 112             'url': 'http://vk.com/video205387401_165548505',
 113             'info_dict': {
 114                 'id': '205387401_165548505',
 115                 'ext': 'mp4',
 116                 'title': 'No name',
 117                 'uploader': 'Tom Cruise',
 118                 'uploader_id': '205387401',
 119                 'duration': 9,
 120                 'timestamp': 1374364108,
 121                 'upload_date': '20130720',
 122             }
 123         },
 124         {
 125             'note': 'Embedded video',
 126             'url': 'https://vk.com/video_ext.php?oid=-77521&id=162222515&hash=87b046504ccd8bfa',
 127             'md5': '7babad3b85ea2e91948005b1b8b0cb84',
 128             'info_dict': {
 129                 'id': '-77521_162222515',
 130                 'ext': 'mp4',
 131                 'uploader': 're:(?:Noize MC|Alexander Ilyashenko).*',
 132                 'title': 'ProtivoGunz - Хуёвая песня',
 133                 'duration': 195,
 134                 'upload_date': '20120212',
 135                 'timestamp': 1329049880,
 136                 'uploader_id': '-77521',
 137             },
 138         },
 139         {
 140             # VIDEO NOW REMOVED
 141             # please update if you find a video whose URL follows the same pattern
 142             'url': 'http://vk.com/video-8871596_164049491',
 143             'md5': 'a590bcaf3d543576c9bd162812387666',
 144             'note': 'Only available for registered users',
 145             'info_dict': {
 146                 'id': '-8871596_164049491',
 147                 'ext': 'mp4',
 148                 'uploader': 'Триллеры',
 149                 'title': '► Бойцовский клуб / Fight Club 1999 [HD 720]',
 150                 'duration': 8352,
 151                 'upload_date': '20121218',
 152                 'view_count': int,
 153             },
 154             'skip': 'Removed',
 155         },
 156         {
 157             'url': 'http://vk.com/hd_kino_mania?z=video-43215063_168067957%2F15c66b9b533119788d',
 158             'info_dict': {
 159                 'id': '-43215063_168067957',
 160                 'ext': 'mp4',
 161                 'uploader': 'Bro Mazter',
 162                 'title': ' ',
 163                 'duration': 7291,
 164                 'upload_date': '20140328',
 165                 'uploader_id': '223413403',
 166                 'timestamp': 1396018030,
 167             },
 168             'skip': 'Requires vk account credentials',
 169         },
 170         {
 171             'url': 'http://m.vk.com/video-43215063_169084319?list=125c627d1aa1cebb83&from=wall-43215063_2566540',
 172             'md5': '0c45586baa71b7cb1d0784ee3f4e00a6',
 173             'note': 'ivi.ru embed',
 174             'info_dict': {
 175                 'id': '-43215063_169084319',
 176                 'ext': 'mp4',
 177                 'title': 'Книга Илая',
 178                 'duration': 6771,
 179                 'upload_date': '20140626',
 180                 'view_count': int,
 181             },
 182             'skip': 'Removed',
 183         },
 184         {
 185             # video (removed?) only available with list id
 186             'url': 'https://vk.com/video30481095_171201961?list=8764ae2d21f14088d4',
 187             'md5': '091287af5402239a1051c37ec7b92913',
 188             'info_dict': {
 189                 'id': '30481095_171201961',
 190                 'ext': 'mp4',
 191                 'title': 'ТюменцевВВ_09.07.2015',
 192                 'uploader': 'Anton Ivanov',
 193                 'duration': 109,
 194                 'upload_date': '20150709',
 195                 'view_count': int,
 196             },
 197             'skip': 'Removed',
 198         },
 199         {
 200             # youtube embed
 201             'url': 'https://vk.com/video276849682_170681728',
 202             'info_dict': {
 203                 'id': 'V3K4mi0SYkc',
 204                 'ext': 'mp4',
 205                 'title': "DSWD Awards 'Children's Joy Foundation, Inc.' Certificate of Registration and License to Operate",
 206                 'description': 'md5:bf9c26cfa4acdfb146362682edd3827a',
 207                 'duration': 178,
 208                 'upload_date': '20130116',
 209                 'uploader': "Children's Joy Foundation Inc.",
 210                 'uploader_id': 'thecjf',
 211                 'view_count': int,
 212             },
 213         },
 214         {
 215             # dailymotion embed
 216             'url': 'https://vk.com/video-37468416_456239855',
 217             'info_dict': {
 218                 'id': 'k3lz2cmXyRuJQSjGHUv',
 219                 'ext': 'mp4',
 220                 'title': 'md5:d52606645c20b0ddbb21655adaa4f56f',
 221                 'description': 'md5:424b8e88cc873217f520e582ba28bb36',
 222                 'uploader': 'AniLibria.Tv',
 223                 'upload_date': '20160914',
 224                 'uploader_id': 'x1p5vl5',
 225                 'timestamp': 1473877246,
 226             },
 227             'params': {
 228                 'skip_download': True,
 229             },
 230         },
 231         {
 232             # video key is extra_data not url\d+
 233             'url': 'http://vk.com/video-110305615_171782105',
 234             'md5': 'e13fcda136f99764872e739d13fac1d1',
 235             'info_dict': {
 236                 'id': '-110305615_171782105',
 237                 'ext': 'mp4',
 238                 'title': 'S-Dance, репетиции к The way show',
 239                 'uploader': 'THE WAY SHOW | 17 апреля',
 240                 'uploader_id': '-110305615',
 241                 'timestamp': 1454859345,
 242                 'upload_date': '20160207',
 243             },
 244             'params': {
 245                 'skip_download': True,
 246             },
 247         },
 248         {
 249             # finished live stream, postlive_mp4
 250             'url': 'https://vk.com/videos-387766?z=video-387766_456242764%2Fpl_-387766_-2',
 251             'info_dict': {
 252                 'id': '-387766_456242764',
 253                 'ext': 'mp4',
 254                 'title': 'ИгроМир 2016 День 1 — Игромания Утром',
 255                 'uploader': 'Игромания',
 256                 'duration': 5239,
 257                 # TODO: use act=show to extract view_count
 258                 # 'view_count': int,
 259                 'upload_date': '20160929',
 260                 'uploader_id': '-387766',
 261                 'timestamp': 1475137527,
 262             },
 263             'params': {
 264                 'skip_download': True,
 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     @staticmethod
 304     def _extract_sibnet_urls(webpage):
 305         # https://help.sibnet.ru/?sibnet_video_embed
 306         return [unescapeHTML(mobj.group('url')) for mobj in re.finditer(
 307             r'<iframe\b[^>]+\bsrc=(["\'])(?P<url>(?:https?:)?//video\.sibnet\.ru/shell\.php\?.*?\bvideoid=\d+.*?)\1',
 308             webpage)]
 309
 310     def _real_extract(self, url):
 311         mobj = self._match_valid_url(url)
 312         video_id = mobj.group('videoid')
 313
 314         mv_data = {}
 315         if video_id:
 316             data = {
 317                 'act': 'show_inline',
 318                 'video': video_id,
 319             }
 320             # Some videos (removed?) can only be downloaded with list id specified
 321             list_id = mobj.group('list_id')
 322             if list_id:
 323                 data['list'] = list_id
 324
 325             payload = self._download_payload('al_video', video_id, data)
 326             info_page = payload[1]
 327             opts = payload[-1]
 328             mv_data = opts.get('mvData') or {}
 329             player = opts.get('player') or {}
 330         else:
 331             video_id = '%s_%s' % (mobj.group('oid'), mobj.group('id'))
 332
 333             info_page = self._download_webpage(
 334                 'http://vk.com/video_ext.php?' + mobj.group('embed_query'), video_id)
 335
 336             error_message = self._html_search_regex(
 337                 [r'(?s)<!><div[^>]+class="video_layer_message"[^>]*>(.+?)</div>',
 338                     r'(?s)<div[^>]+id="video_ext_msg"[^>]*>(.+?)</div>'],
 339                 info_page, 'error message', default=None)
 340             if error_message:
 341                 raise ExtractorError(error_message, expected=True)
 342
 343             if re.search(r'<!>/login\.php\?.*\bact=security_check', info_page):
 344                 raise ExtractorError(
 345                     'You are trying to log in from an unusual location. You should confirm ownership at vk.com to log in with this IP.',
 346                     expected=True)
 347
 348             ERROR_COPYRIGHT = 'Video %s has been removed from public access due to rightholder complaint.'
 349
 350             ERRORS = {
 351                 r'>Видеозапись .*? была изъята из публичного доступа в связи с обращением правообладателя.<':
 352                 ERROR_COPYRIGHT,
 353
 354                 r'>The video .*? was removed from public access by request of the copyright holder.<':
 355                 ERROR_COPYRIGHT,
 356
 357                 r'<!>Please log in or <':
 358                 'Video %s is only available for registered users, '
 359                 'use --username and --password options to provide account credentials.',
 360
 361                 r'<!>Unknown error':
 362                 'Video %s does not exist.',
 363
 364                 r'<!>Видео временно недоступно':
 365                 'Video %s is temporarily unavailable.',
 366
 367                 r'<!>Access denied':
 368                 'Access denied to video %s.',
 369
 370                 r'<!>Видеозапись недоступна, так как её автор был заблокирован.':
 371                 'Video %s is no longer available, because its author has been blocked.',
 372
 373                 r'<!>This video is no longer available, because its author has been blocked.':
 374                 'Video %s is no longer available, because its author has been blocked.',
 375
 376                 r'<!>This video is no longer available, because it has been deleted.':
 377                 'Video %s is no longer available, because it has been deleted.',
 378
 379                 r'<!>The video .+? is not available in your region.':
 380                 'Video %s is not available in your region.',
 381             }
 382
 383             for error_re, error_msg in ERRORS.items():
 384                 if re.search(error_re, info_page):
 385                     raise ExtractorError(error_msg % video_id, expected=True)
 386
 387             player = self._parse_json(self._search_regex(
 388                 r'var\s+playerParams\s*=\s*({.+?})\s*;\s*\n',
 389                 info_page, 'player params'), video_id)
 390
 391         youtube_url = YoutubeIE._extract_url(info_page)
 392         if youtube_url:
 393             return self.url_result(youtube_url, YoutubeIE.ie_key())
 394
 395         vimeo_url = VimeoIE._extract_url(url, info_page)
 396         if vimeo_url is not None:
 397             return self.url_result(vimeo_url, VimeoIE.ie_key())
 398
 399         pladform_url = PladformIE._extract_url(info_page)
 400         if pladform_url:
 401             return self.url_result(pladform_url, PladformIE.ie_key())
 402
 403         m_rutube = re.search(
 404             r'\ssrc="((?:https?:)?//rutube\.ru\\?/(?:video|play)\\?/embed(?:.*?))\\?"', info_page)
 405         if m_rutube is not None:
 406             rutube_url = self._proto_relative_url(
 407                 m_rutube.group(1).replace('\\', ''))
 408             return self.url_result(rutube_url)
 409
 410         dailymotion_urls = DailymotionIE._extract_urls(info_page)
 411         if dailymotion_urls:
 412             return self.url_result(dailymotion_urls[0], DailymotionIE.ie_key())
 413
 414         odnoklassniki_url = OdnoklassnikiIE._extract_url(info_page)
 415         if odnoklassniki_url:
 416             return self.url_result(odnoklassniki_url, OdnoklassnikiIE.ie_key())
 417
 418         sibnet_urls = self._extract_sibnet_urls(info_page)
 419         if sibnet_urls:
 420             return self.url_result(sibnet_urls[0])
 421
 422         m_opts = re.search(r'(?s)var\s+opts\s*=\s*({.+?});', info_page)
 423         if m_opts:
 424             m_opts_url = re.search(r"url\s*:\s*'((?!/\b)[^']+)", m_opts.group(1))
 425             if m_opts_url:
 426                 opts_url = m_opts_url.group(1)
 427                 if opts_url.startswith('//'):
 428                     opts_url = 'http:' + opts_url
 429                 return self.url_result(opts_url)
 430
 431         data = player['params'][0]
 432         title = unescapeHTML(data['md_title'])
 433
 434         # 2 = live
 435         # 3 = post live (finished live)
 436         is_live = data.get('live') == 2
 437         if is_live:
 438             title = self._live_title(title)
 439
 440         timestamp = unified_timestamp(self._html_search_regex(
 441             r'class=["\']mv_info_date[^>]+>([^<]+)(?:<|from)', info_page,
 442             'upload date', default=None)) or int_or_none(data.get('date'))
 443
 444         view_count = str_to_int(self._search_regex(
 445             r'class=["\']mv_views_count[^>]+>\s*([\d,.]+)',
 446             info_page, 'view count', default=None))
 447
 448         formats = []
 449         for format_id, format_url in data.items():
 450             format_url = url_or_none(format_url)
 451             if not format_url or not format_url.startswith(('http', '//', 'rtmp')):
 452                 continue
 453             if (format_id.startswith(('url', 'cache'))
 454                     or format_id in ('extra_data', 'live_mp4', 'postlive_mp4')):
 455                 height = int_or_none(self._search_regex(
 456                     r'^(?:url|cache)(\d+)', format_id, 'height', default=None))
 457                 formats.append({
 458                     'format_id': format_id,
 459                     'url': format_url,
 460                     'height': height,
 461                 })
 462             elif format_id == 'hls':
 463                 formats.extend(self._extract_m3u8_formats(
 464                     format_url, video_id, 'mp4', 'm3u8_native',
 465                     m3u8_id=format_id, fatal=False, live=is_live))
 466             elif format_id == 'rtmp':
 467                 formats.append({
 468                     'format_id': format_id,
 469                     'url': format_url,
 470                     'ext': 'flv',
 471                 })
 472         self._sort_formats(formats)
 473
 474         subtitles = {}
 475         for sub in data.get('subs') or {}:
 476             subtitles.setdefault(sub.get('lang', 'en'), []).append({
 477                 'ext': sub.get('title', '.srt').split('.')[-1],
 478                 'url': url_or_none(sub.get('url')),
 479             })
 480
 481         return {
 482             'id': video_id,
 483             'formats': formats,
 484             'title': title,
 485             'thumbnail': data.get('jpg'),
 486             'uploader': data.get('md_author'),
 487             'uploader_id': str_or_none(data.get('author_id') or mv_data.get('authorId')),
 488             'duration': int_or_none(data.get('duration') or mv_data.get('duration')),
 489             'timestamp': timestamp,
 490             'view_count': view_count,
 491             'like_count': int_or_none(mv_data.get('likes')),
 492             'comment_count': int_or_none(mv_data.get('commcount')),
 493             'is_live': is_live,
 494             'subtitles': subtitles,
 495         }
 496
 497
 498 class VKUserVideosIE(VKBaseIE):
 499     IE_NAME = 'vk:uservideos'
 500     IE_DESC = "VK - User's Videos"
 501     _VALID_URL = r'https?://(?:(?:m|new)\.)?vk\.com/videos(?P<id>-?[0-9]+)(?!\?.*\bz=video)(?:[/?#&](?:.*?\bsection=(?P<section>\w+))?|$)'
 502     _TEMPLATE_URL = 'https://vk.com/videos'
 503     _TESTS = [{
 504         'url': 'https://vk.com/videos-767561',
 505         'info_dict': {
 506             'id': '-767561_all',
 507         },
 508         'playlist_mincount': 1150,
 509     }, {
 510         'url': 'https://vk.com/videos-767561?section=uploaded',
 511         'info_dict': {
 512             'id': '-767561_uploaded',
 513         },
 514         'playlist_mincount': 425,
 515     }, {
 516         'url': 'http://vk.com/videos205387401',
 517         'only_matching': True,
 518     }, {
 519         'url': 'http://vk.com/videos-77521',
 520         'only_matching': True,
 521     }, {
 522         'url': 'http://vk.com/videos-97664626?section=all',
 523         'only_matching': True,
 524     }, {
 525         'url': 'http://m.vk.com/videos205387401',
 526         'only_matching': True,
 527     }, {
 528         'url': 'http://new.vk.com/videos205387401',
 529         'only_matching': True,
 530     }]
 531     _PAGE_SIZE = 1000
 532     _VIDEO = collections.namedtuple('Video', ['owner_id', 'id'])
 533
 534     def _fetch_page(self, page_id, section, page):
 535         l = self._download_payload('al_video', page_id, {
 536             'act': 'load_videos_silent',
 537             'offset': page * self._PAGE_SIZE,
 538             'oid': page_id,
 539             'section': section,
 540         })[0][section]['list']
 541
 542         for video in l:
 543             v = self._VIDEO._make(video[:2])
 544             video_id = '%d_%d' % (v.owner_id, v.id)
 545             yield self.url_result(
 546                 'http://vk.com/video' + video_id, VKIE.ie_key(), video_id)
 547
 548     def _real_extract(self, url):
 549         page_id, section = self._match_valid_url(url).groups()
 550         if not section:
 551             section = 'all'
 552
 553         entries = OnDemandPagedList(
 554             functools.partial(self._fetch_page, page_id, section),
 555             self._PAGE_SIZE)
 556
 557         return self.playlist_result(entries, '%s_%s' % (page_id, section))
 558
 559
 560 class VKWallPostIE(VKBaseIE):
 561     IE_NAME = 'vk:wallpost'
 562     _VALID_URL = r'https?://(?:(?:(?:(?:m|new)\.)?vk\.com/(?:[^?]+\?.*\bw=)?wall(?P<id>-?\d+_\d+)))'
 563     _TESTS = [{
 564         # public page URL, audio playlist
 565         'url': 'https://vk.com/bs.official?w=wall-23538238_35',
 566         'info_dict': {
 567             'id': '-23538238_35',
 568             'title': 'Black Shadow - Wall post -23538238_35',
 569             'description': 'md5:3f84b9c4f9ef499731cf1ced9998cc0c',
 570         },
 571         'playlist': [{
 572             'md5': '5ba93864ec5b85f7ce19a9af4af080f6',
 573             'info_dict': {
 574                 'id': '135220665_111806521',
 575                 'ext': 'mp4',
 576                 'title': 'Black Shadow - Слепое Верование',
 577                 'duration': 370,
 578                 'uploader': 'Black Shadow',
 579                 'artist': 'Black Shadow',
 580                 'track': 'Слепое Верование',
 581             },
 582         }, {
 583             'md5': '4cc7e804579122b17ea95af7834c9233',
 584             'info_dict': {
 585                 'id': '135220665_111802303',
 586                 'ext': 'mp4',
 587                 'title': 'Black Shadow - Война - Негасимое Бездны Пламя!',
 588                 'duration': 423,
 589                 'uploader': 'Black Shadow',
 590                 'artist': 'Black Shadow',
 591                 'track': 'Война - Негасимое Бездны Пламя!',
 592             },
 593         }],
 594         'params': {
 595             'skip_download': True,
 596             'usenetrc': True,
 597         },
 598         'skip': 'Requires vk account credentials',
 599     }, {
 600         # single YouTube embed, no leading -
 601         'url': 'https://vk.com/wall85155021_6319',
 602         'info_dict': {
 603             'id': '85155021_6319',
 604             'title': 'Сергей Горбунов - Wall post 85155021_6319',
 605         },
 606         'playlist_count': 1,
 607         'params': {
 608             'usenetrc': True,
 609         },
 610         'skip': 'Requires vk account credentials',
 611     }, {
 612         # wall page URL
 613         'url': 'https://vk.com/wall-23538238_35',
 614         'only_matching': True,
 615     }, {
 616         # mobile wall page URL
 617         'url': 'https://m.vk.com/wall-23538238_35',
 618         'only_matching': True,
 619     }]
 620     _BASE64_CHARS = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMN0PQRSTUVWXYZO123456789+/='
 621     _AUDIO = collections.namedtuple('Audio', ['id', 'owner_id', 'url', 'title', 'performer', 'duration', 'album_id', 'unk', 'author_link', 'lyrics', 'flags', 'context', 'extra', 'hashes', 'cover_url', 'ads'])
 622
 623     def _decode(self, enc):
 624         dec = ''
 625         e = n = 0
 626         for c in enc:
 627             r = self._BASE64_CHARS.index(c)
 628             cond = n % 4
 629             e = 64 * e + r if cond else r
 630             n += 1
 631             if cond:
 632                 dec += chr(255 & e >> (-2 * n & 6))
 633         return dec
 634
 635     def _unmask_url(self, mask_url, vk_id):
 636         if 'audio_api_unavailable' in mask_url:
 637             extra = mask_url.split('?extra=')[1].split('#')
 638             func, base = self._decode(extra[1]).split(chr(11))
 639             mask_url = list(self._decode(extra[0]))
 640             url_len = len(mask_url)
 641             indexes = [None] * url_len
 642             index = int(base) ^ vk_id
 643             for n in range(url_len - 1, -1, -1):
 644                 index = (url_len * (n + 1) ^ index + n) % url_len
 645                 indexes[n] = index
 646             for n in range(1, url_len):
 647                 c = mask_url[n]
 648                 index = indexes[url_len - 1 - n]
 649                 mask_url[n] = mask_url[index]
 650                 mask_url[index] = c
 651             mask_url = ''.join(mask_url)
 652         return mask_url
 653
 654     def _real_extract(self, url):
 655         post_id = self._match_id(url)
 656
 657         webpage = self._download_payload('wkview', post_id, {
 658             'act': 'show',
 659             'w': 'wall' + post_id,
 660         })[1]
 661
 662         description = clean_html(get_element_by_class('wall_post_text', webpage))
 663         uploader = clean_html(get_element_by_class('author', webpage))
 664
 665         entries = []
 666
 667         for audio in re.findall(r'data-audio="([^"]+)', webpage):
 668             audio = self._parse_json(unescapeHTML(audio), post_id)
 669             a = self._AUDIO._make(audio[:16])
 670             if not a.url:
 671                 continue
 672             title = unescapeHTML(a.title)
 673             performer = unescapeHTML(a.performer)
 674             entries.append({
 675                 'id': '%s_%s' % (a.owner_id, a.id),
 676                 'url': self._unmask_url(a.url, a.ads['vk_id']),
 677                 'title': '%s - %s' % (performer, title) if performer else title,
 678                 'thumbnails': [{'url': c_url} for c_url in a.cover_url.split(',')] if a.cover_url else None,
 679                 'duration': int_or_none(a.duration),
 680                 'uploader': uploader,
 681                 'artist': performer,
 682                 'track': title,
 683                 'ext': 'mp4',
 684                 'protocol': 'm3u8',
 685             })
 686
 687         for video in re.finditer(
 688                 r'<a[^>]+href=(["\'])(?P<url>/video(?:-?[\d_]+).*?)\1', webpage):
 689             entries.append(self.url_result(
 690                 compat_urlparse.urljoin(url, video.group('url')), VKIE.ie_key()))
 691
 692         title = 'Wall post %s' % post_id
 693
 694         return self.playlist_result(
 695             orderedSet(entries), post_id,
 696             '%s - %s' % (uploader, title) if uploader else title,
 697             description)