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
  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/rg3/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 as %s' % username,
  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': '0deae91935c54e00003c2a00646315f0',
 102             'info_dict': {
 103                 'id': '162222515',
 104                 'ext': 'flv',
 105                 'title': 'ProtivoGunz - Хуёвая песня',
 106                 'uploader': 're:(?:Noize MC|Alexander Ilyashenko).*',
 107                 'duration': 195,
 108                 'timestamp': 1329060660,
 109                 'upload_date': '20120212',
 110                 'view_count': int,
 111             },
 112         },
 113         {
 114             'url': 'http://vk.com/video205387401_165548505',
 115             'md5': '6c0aeb2e90396ba97035b9cbde548700',
 116             'info_dict': {
 117                 'id': '165548505',
 118                 'ext': 'mp4',
 119                 'uploader': 'Tom Cruise',
 120                 'title': 'No name',
 121                 'duration': 9,
 122                 'timestamp': 1374374880,
 123                 'upload_date': '20130721',
 124                 'view_count': int,
 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': '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': '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': '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': '60690',
 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': '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': 'webm',
 207                 'title': "DSWD Awards 'Children's Joy Foundation, Inc.' Certificate of Registration and License to Operate",
 208                 'description': 'md5:d9903938abdc74c738af77f527ca0596',
 209                 'duration': 178,
 210                 'upload_date': '20130116',
 211                 'uploader': "Children's Joy Foundation",
 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                 'description': 'md5:c651358f03c56f1150b555c26d90a0fd',
 224                 'uploader': 'AniLibria.Tv',
 225                 'upload_date': '20160914',
 226                 'uploader_id': 'x1p5vl5',
 227                 'timestamp': 1473877246,
 228             },
 229             'params': {
 230                 'skip_download': True,
 231             },
 232         },
 233         {
 234             # video key is extra_data not url\d+
 235             'url': 'http://vk.com/video-110305615_171782105',
 236             'md5': 'e13fcda136f99764872e739d13fac1d1',
 237             'info_dict': {
 238                 'id': '171782105',
 239                 'ext': 'mp4',
 240                 'title': 'S-Dance, репетиции к The way show',
 241                 'uploader': 'THE WAY SHOW | 17 апреля',
 242                 'timestamp': 1454870100,
 243                 'upload_date': '20160207',
 244                 'view_count': int,
 245             },
 246         },
 247         {
 248             # finished live stream, postlive_mp4
 249             'url': 'https://vk.com/videos-387766?z=video-387766_456242764%2Fpl_-387766_-2',
 250             'md5': '90d22d051fccbbe9becfccc615be6791',
 251             'info_dict': {
 252                 'id': '456242764',
 253                 'ext': 'mp4',
 254                 'title': 'ИгроМир 2016 — день 1',
 255                 'uploader': 'Игромания',
 256                 'duration': 5239,
 257                 'view_count': int,
 258             },
 259         },
 260         {
 261             # live stream, hls and rtmp links, most likely already finished live
 262             # stream by the time you are reading this comment
 263             'url': 'https://vk.com/video-140332_456239111',
 264             'only_matching': True,
 265         },
 266         {
 267             # removed video, just testing that we match the pattern
 268             'url': 'http://vk.com/feed?z=video-43215063_166094326%2Fbb50cacd3177146d7a',
 269             'only_matching': True,
 270         },
 271         {
 272             # age restricted video, requires vk account credentials
 273             'url': 'https://vk.com/video205387401_164765225',
 274             'only_matching': True,
 275         },
 276         {
 277             # pladform embed
 278             'url': 'https://vk.com/video-76116461_171554880',
 279             'only_matching': True,
 280         },
 281         {
 282             'url': 'http://new.vk.com/video205387401_165548505',
 283             'only_matching': True,
 284         }
 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&al=1&module=video&video=%s' % 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         ERRORS = {
 316             r'>Видеозапись .*? была изъята из публичного доступа в связи с обращением правообладателя.<':
 317             'Video %s has been removed from public access due to rightholder complaint.',
 318
 319             r'<!>Please log in or <':
 320             'Video %s is only available for registered users, '
 321             'use --username and --password options to provide account credentials.',
 322
 323             r'<!>Unknown error':
 324             'Video %s does not exist.',
 325
 326             r'<!>Видео временно недоступно':
 327             'Video %s is temporarily unavailable.',
 328
 329             r'<!>Access denied':
 330             'Access denied to video %s.',
 331         }
 332
 333         for error_re, error_msg in ERRORS.items():
 334             if re.search(error_re, info_page):
 335                 raise ExtractorError(error_msg % video_id, expected=True)
 336
 337         youtube_url = self._search_regex(
 338             r'<iframe[^>]+src="((?:https?:)?//www.youtube.com/embed/[^"]+)"',
 339             info_page, 'youtube iframe', default=None)
 340         if youtube_url:
 341             return self.url_result(youtube_url, 'Youtube')
 342
 343         vimeo_url = VimeoIE._extract_url(url, info_page)
 344         if vimeo_url is not None:
 345             return self.url_result(vimeo_url)
 346
 347         pladform_url = PladformIE._extract_url(info_page)
 348         if pladform_url:
 349             return self.url_result(pladform_url)
 350
 351         m_rutube = re.search(
 352             r'\ssrc="((?:https?:)?//rutube\.ru\\?/(?:video|play)\\?/embed(?:.*?))\\?"', info_page)
 353         if m_rutube is not None:
 354             rutube_url = self._proto_relative_url(
 355                 m_rutube.group(1).replace('\\', ''))
 356             return self.url_result(rutube_url)
 357
 358         dailymotion_urls = DailymotionIE._extract_urls(info_page)
 359         if dailymotion_urls:
 360             return self.url_result(dailymotion_urls[0], DailymotionIE.ie_key())
 361
 362         m_opts = re.search(r'(?s)var\s+opts\s*=\s*({.+?});', info_page)
 363         if m_opts:
 364             m_opts_url = re.search(r"url\s*:\s*'((?!/\b)[^']+)", m_opts.group(1))
 365             if m_opts_url:
 366                 opts_url = m_opts_url.group(1)
 367                 if opts_url.startswith('//'):
 368                     opts_url = 'http:' + opts_url
 369                 return self.url_result(opts_url)
 370
 371         # vars does not look to be served anymore since 24.10.2016
 372         data = self._parse_json(
 373             self._search_regex(
 374                 r'var\s+vars\s*=\s*({.+?});', info_page, 'vars', default='{}'),
 375             video_id, fatal=False)
 376
 377         # <!json> is served instead
 378         if not data:
 379             data = self._parse_json(
 380                 self._search_regex(
 381                     r'<!json>\s*({.+?})\s*<!>', info_page, 'json', default='{}'),
 382                 video_id)
 383             if data:
 384                 data = data['player']['params'][0]
 385
 386         if not data:
 387             data = self._parse_json(
 388                 self._search_regex(
 389                     r'var\s+playerParams\s*=\s*({.+?})\s*;\s*\n', info_page,
 390                     'player params'),
 391                 video_id)['params'][0]
 392
 393         title = unescapeHTML(data['md_title'])
 394
 395         # 2 = live
 396         # 3 = post live (finished live)
 397         is_live = data.get('live') == 2
 398         if is_live:
 399             title = self._live_title(title)
 400
 401         timestamp = unified_timestamp(self._html_search_regex(
 402             r'class=["\']mv_info_date[^>]+>([^<]+)(?:<|from)', info_page,
 403             'upload date', fatal=False))
 404
 405         view_count = str_to_int(self._search_regex(
 406             r'class=["\']mv_views_count[^>]+>\s*([\d,.]+)',
 407             info_page, 'view count', fatal=False))
 408
 409         formats = []
 410         for format_id, format_url in data.items():
 411             if not isinstance(format_url, compat_str) or not format_url.startswith(('http', '//', 'rtmp')):
 412                 continue
 413             if (format_id.startswith(('url', 'cache')) or
 414                     format_id in ('extra_data', 'live_mp4', 'postlive_mp4')):
 415                 height = int_or_none(self._search_regex(
 416                     r'^(?:url|cache)(\d+)', format_id, 'height', default=None))
 417                 formats.append({
 418                     'format_id': format_id,
 419                     'url': format_url,
 420                     'height': height,
 421                 })
 422             elif format_id == 'hls':
 423                 formats.extend(self._extract_m3u8_formats(
 424                     format_url, video_id, 'mp4',
 425                     entry_protocol='m3u8' if is_live else 'm3u8_native',
 426                     m3u8_id=format_id, fatal=False, live=is_live))
 427             elif format_id == 'rtmp':
 428                 formats.append({
 429                     'format_id': format_id,
 430                     'url': format_url,
 431                     'ext': 'flv',
 432                 })
 433         self._sort_formats(formats)
 434
 435         return {
 436             'id': compat_str(data.get('vid') or video_id),
 437             'formats': formats,
 438             'title': title,
 439             'thumbnail': data.get('jpg'),
 440             'uploader': data.get('md_author'),
 441             'duration': data.get('duration'),
 442             'timestamp': timestamp,
 443             'view_count': view_count,
 444             'is_live': is_live,
 445         }
 446
 447
 448 class VKUserVideosIE(VKBaseIE):
 449     IE_NAME = 'vk:uservideos'
 450     IE_DESC = "VK - User's Videos"
 451     _VALID_URL = r'https?://(?:(?:m|new)\.)?vk\.com/videos(?P<id>-?[0-9]+)(?!\?.*\bz=video)(?:[/?#&]|$)'
 452     _TEMPLATE_URL = 'https://vk.com/videos'
 453     _TESTS = [{
 454         'url': 'http://vk.com/videos205387401',
 455         'info_dict': {
 456             'id': '205387401',
 457             'title': "Tom Cruise's Videos",
 458         },
 459         'playlist_mincount': 4,
 460     }, {
 461         'url': 'http://vk.com/videos-77521',
 462         'only_matching': True,
 463     }, {
 464         'url': 'http://vk.com/videos-97664626?section=all',
 465         'only_matching': True,
 466     }, {
 467         'url': 'http://m.vk.com/videos205387401',
 468         'only_matching': True,
 469     }, {
 470         'url': 'http://new.vk.com/videos205387401',
 471         'only_matching': True,
 472     }]
 473
 474     def _real_extract(self, url):
 475         page_id = self._match_id(url)
 476
 477         webpage = self._download_webpage(url, page_id)
 478
 479         entries = [
 480             self.url_result(
 481                 'http://vk.com/video' + video_id, 'VK', video_id=video_id)
 482             for video_id in orderedSet(re.findall(r'href="/video(-?[0-9_]+)"', webpage))]
 483
 484         title = unescapeHTML(self._search_regex(
 485             r'<title>\s*([^<]+?)\s+\|\s+\d+\s+videos',
 486             webpage, 'title', default=page_id))
 487
 488         return self.playlist_result(entries, page_id, title)
 489
 490
 491 class VKWallPostIE(VKBaseIE):
 492     IE_NAME = 'vk:wallpost'
 493     _VALID_URL = r'https?://(?:(?:(?:(?:m|new)\.)?vk\.com/(?:[^?]+\?.*\bw=)?wall(?P<id>-?\d+_\d+)))'
 494     _TESTS = [{
 495         # public page URL, audio playlist
 496         'url': 'https://vk.com/bs.official?w=wall-23538238_35',
 497         'info_dict': {
 498             'id': '23538238_35',
 499             'title': 'Black Shadow - Wall post 23538238_35',
 500             'description': 'md5:3f84b9c4f9ef499731cf1ced9998cc0c',
 501         },
 502         'playlist': [{
 503             'md5': '5ba93864ec5b85f7ce19a9af4af080f6',
 504             'info_dict': {
 505                 'id': '135220665_111806521',
 506                 'ext': 'mp3',
 507                 'title': 'Black Shadow - Слепое Верование',
 508                 'duration': 370,
 509                 'uploader': 'Black Shadow',
 510                 'artist': 'Black Shadow',
 511                 'track': 'Слепое Верование',
 512             },
 513         }, {
 514             'md5': '4cc7e804579122b17ea95af7834c9233',
 515             'info_dict': {
 516                 'id': '135220665_111802303',
 517                 'ext': 'mp3',
 518                 'title': 'Black Shadow - Война - Негасимое Бездны Пламя!',
 519                 'duration': 423,
 520                 'uploader': 'Black Shadow',
 521                 'artist': 'Black Shadow',
 522                 'track': 'Война - Негасимое Бездны Пламя!',
 523             },
 524             'params': {
 525                 'skip_download': True,
 526             },
 527         }],
 528         'params': {
 529             'usenetrc': True,
 530         },
 531         'skip': 'Requires vk account credentials',
 532     }, {
 533         # single YouTube embed, no leading -
 534         'url': 'https://vk.com/wall85155021_6319',
 535         'info_dict': {
 536             'id': '85155021_6319',
 537             'title': 'Sergey Gorbunov - Wall post 85155021_6319',
 538         },
 539         'playlist_count': 1,
 540         'params': {
 541             'usenetrc': True,
 542         },
 543         'skip': 'Requires vk account credentials',
 544     }, {
 545         # wall page URL
 546         'url': 'https://vk.com/wall-23538238_35',
 547         'only_matching': True,
 548     }, {
 549         # mobile wall page URL
 550         'url': 'https://m.vk.com/wall-23538238_35',
 551         'only_matching': True,
 552     }]
 553
 554     def _real_extract(self, url):
 555         post_id = self._match_id(url)
 556
 557         wall_url = 'https://vk.com/wall%s' % post_id
 558
 559         post_id = remove_start(post_id, '-')
 560
 561         webpage = self._download_webpage(wall_url, post_id)
 562
 563         error = self._html_search_regex(
 564             r'>Error</div>\s*<div[^>]+class=["\']body["\'][^>]*>([^<]+)',
 565             webpage, 'error', default=None)
 566         if error:
 567             raise ExtractorError('VK said: %s' % error, expected=True)
 568
 569         description = clean_html(get_element_by_class('wall_post_text', webpage))
 570         uploader = clean_html(get_element_by_class('author', webpage))
 571         thumbnail = self._og_search_thumbnail(webpage)
 572
 573         entries = []
 574
 575         audio_ids = re.findall(r'data-full-id=["\'](\d+_\d+)', webpage)
 576         if audio_ids:
 577             al_audio = self._download_webpage(
 578                 'https://vk.com/al_audio.php', post_id,
 579                 note='Downloading audio info', fatal=False,
 580                 data=urlencode_postdata({
 581                     'act': 'reload_audio',
 582                     'al': '1',
 583                     'ids': ','.join(audio_ids)
 584                 }))
 585             if al_audio:
 586                 Audio = collections.namedtuple(
 587                     'Audio', ['id', 'user_id', 'url', 'track', 'artist', 'duration'])
 588                 audios = self._parse_json(
 589                     self._search_regex(
 590                         r'<!json>(.+?)<!>', al_audio, 'audios', default='[]'),
 591                     post_id, fatal=False, transform_source=unescapeHTML)
 592                 if isinstance(audios, list):
 593                     for audio in audios:
 594                         a = Audio._make(audio[:6])
 595                         entries.append({
 596                             'id': '%s_%s' % (a.user_id, a.id),
 597                             'url': a.url,
 598                             'title': '%s - %s' % (a.artist, a.track) if a.artist and a.track else a.id,
 599                             'thumbnail': thumbnail,
 600                             'duration': a.duration,
 601                             'uploader': uploader,
 602                             'artist': a.artist,
 603                             'track': a.track,
 604                         })
 605
 606         for video in re.finditer(
 607                 r'<a[^>]+href=(["\'])(?P<url>/video(?:-?[\d_]+).*?)\1', webpage):
 608             entries.append(self.url_result(
 609                 compat_urlparse.urljoin(url, video.group('url')), VKIE.ie_key()))
 610
 611         title = 'Wall post %s' % post_id
 612
 613         return self.playlist_result(
 614             orderedSet(entries), post_id,
 615             '%s - %s' % (uploader, title) if uploader else title,
 616             description)