youtube_dl/extractor/vk.py

   1 # encoding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import collections
   5 import re
   6 import json
   7 import sys
   8
   9 from .common import InfoExtractor
  10 from ..compat import (
  11     compat_str,
  12     compat_urlparse,
  13 )
  14 from ..utils import (
  15     clean_html,
  16     ExtractorError,
  17     get_element_by_class,
  18     int_or_none,
  19     orderedSet,
  20     remove_start,
  21     str_to_int,
  22     unescapeHTML,
  23     unified_timestamp,
  24     urlencode_postdata,
  25 )
  26 from .dailymotion import DailymotionIE
  27 from .pladform import PladformIE
  28 from .vimeo import VimeoIE
  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         # https://new.vk.com/ serves two same remixlhk cookies in Set-Cookie header
  50         # and expects the first one to be set rather than second (see
  51         # https://github.com/rg3/youtube-dl/issues/9841#issuecomment-227871201).
  52         # As of RFC6265 the newer one cookie should be set into cookie store
  53         # what actually happens.
  54         # We will workaround this VK issue by resetting the remixlhk cookie to
  55         # the first one manually.
  56         for header, cookies in url_handle.headers.items():
  57             if header.lower() != 'set-cookie':
  58                 continue
  59             if sys.version_info[0] >= 3:
  60                 cookies = cookies.encode('iso-8859-1')
  61             cookies = cookies.decode('utf-8')
  62             remixlhk = re.search(r'remixlhk=(.+?);.*?\bdomain=(.+?)(?:[,;]|$)', cookies)
  63             if remixlhk:
  64                 value, domain = remixlhk.groups()
  65                 self._set_cookie(domain, 'remixlhk', value)
  66                 break
  67
  68         login_page = self._download_webpage(
  69             'https://login.vk.com/?act=login', None,
  70             note='Logging in as %s' % username,
  71             data=urlencode_postdata(login_form))
  72
  73         if re.search(r'onLoginFailed', login_page):
  74             raise ExtractorError(
  75                 'Unable to login, incorrect username and/or password', expected=True)
  76
  77     def _real_initialize(self):
  78         self._login()
  79
  80
  81 class VKIE(VKBaseIE):
  82     IE_NAME = 'vk'
  83     IE_DESC = 'VK'
  84     _VALID_URL = r'''(?x)
  85                     https?://
  86                         (?:
  87                             (?:
  88                                 (?:(?:m|new)\.)?vk\.com/video_|
  89                                 (?:www\.)?daxab.com/
  90                             )
  91                             ext\.php\?(?P<embed_query>.*?\boid=(?P<oid>-?\d+).*?\bid=(?P<id>\d+).*)|
  92                             (?:
  93                                 (?:(?:m|new)\.)?vk\.com/(?:.+?\?.*?z=)?video|
  94                                 (?:www\.)?daxab.com/embed/
  95                             )
  96                             (?P<videoid>-?\d+_\d+)(?:.*\blist=(?P<list_id>[\da-f]+))?
  97                         )
  98                     '''
  99     _TESTS = [
 100         {
 101             'url': 'http://vk.com/videos-77521?z=video-77521_162222515%2Fclub77521',
 102             'md5': '0deae91935c54e00003c2a00646315f0',
 103             'info_dict': {
 104                 'id': '162222515',
 105                 'ext': 'flv',
 106                 'title': 'ProtivoGunz - Хуёвая песня',
 107                 'uploader': 're:(?:Noize MC|Alexander Ilyashenko).*',
 108                 'duration': 195,
 109                 'timestamp': 1329060660,
 110                 'upload_date': '20120212',
 111                 'view_count': int,
 112             },
 113         },
 114         {
 115             'url': 'http://vk.com/video205387401_165548505',
 116             'md5': '6c0aeb2e90396ba97035b9cbde548700',
 117             'info_dict': {
 118                 'id': '165548505',
 119                 'ext': 'mp4',
 120                 'uploader': 'Tom Cruise',
 121                 'title': 'No name',
 122                 'duration': 9,
 123                 'timestamp': 1374374880,
 124                 'upload_date': '20130721',
 125                 'view_count': int,
 126             }
 127         },
 128         {
 129             'note': 'Embedded video',
 130             'url': 'http://vk.com/video_ext.php?oid=32194266&id=162925554&hash=7d8c2e0d5e05aeaa&hd=1',
 131             'md5': 'c7ce8f1f87bec05b3de07fdeafe21a0a',
 132             'info_dict': {
 133                 'id': '162925554',
 134                 'ext': 'mp4',
 135                 'uploader': 'Vladimir Gavrin',
 136                 'title': 'Lin Dan',
 137                 'duration': 101,
 138                 'upload_date': '20120730',
 139                 'view_count': int,
 140             },
 141             'skip': 'This video has been removed from public access.',
 142         },
 143         {
 144             # VIDEO NOW REMOVED
 145             # please update if you find a video whose URL follows the same pattern
 146             'url': 'http://vk.com/video-8871596_164049491',
 147             'md5': 'a590bcaf3d543576c9bd162812387666',
 148             'note': 'Only available for registered users',
 149             'info_dict': {
 150                 'id': '164049491',
 151                 'ext': 'mp4',
 152                 'uploader': 'Триллеры',
 153                 'title': '► Бойцовский клуб / Fight Club 1999 [HD 720]',
 154                 'duration': 8352,
 155                 'upload_date': '20121218',
 156                 'view_count': int,
 157             },
 158             'skip': 'Requires vk account credentials',
 159         },
 160         {
 161             'url': 'http://vk.com/hd_kino_mania?z=video-43215063_168067957%2F15c66b9b533119788d',
 162             'md5': '4d7a5ef8cf114dfa09577e57b2993202',
 163             'info_dict': {
 164                 'id': '168067957',
 165                 'ext': 'mp4',
 166                 'uploader': 'Киномания - лучшее из мира кино',
 167                 'title': ' ',
 168                 'duration': 7291,
 169                 'upload_date': '20140328',
 170             },
 171             'skip': 'Requires vk account credentials',
 172         },
 173         {
 174             'url': 'http://m.vk.com/video-43215063_169084319?list=125c627d1aa1cebb83&from=wall-43215063_2566540',
 175             'md5': '0c45586baa71b7cb1d0784ee3f4e00a6',
 176             'note': 'ivi.ru embed',
 177             'info_dict': {
 178                 'id': '60690',
 179                 'ext': 'mp4',
 180                 'title': 'Книга Илая',
 181                 'duration': 6771,
 182                 'upload_date': '20140626',
 183                 'view_count': int,
 184             },
 185             'skip': 'Only works from Russia',
 186         },
 187         {
 188             # video (removed?) only available with list id
 189             'url': 'https://vk.com/video30481095_171201961?list=8764ae2d21f14088d4',
 190             'md5': '091287af5402239a1051c37ec7b92913',
 191             'info_dict': {
 192                 'id': '171201961',
 193                 'ext': 'mp4',
 194                 'title': 'ТюменцевВВ_09.07.2015',
 195                 'uploader': 'Anton Ivanov',
 196                 'duration': 109,
 197                 'upload_date': '20150709',
 198                 'view_count': int,
 199             },
 200             'skip': 'Removed',
 201         },
 202         {
 203             # youtube embed
 204             'url': 'https://vk.com/video276849682_170681728',
 205             'info_dict': {
 206                 'id': 'V3K4mi0SYkc',
 207                 'ext': 'webm',
 208                 'title': "DSWD Awards 'Children's Joy Foundation, Inc.' Certificate of Registration and License to Operate",
 209                 'description': 'md5:d9903938abdc74c738af77f527ca0596',
 210                 'duration': 178,
 211                 'upload_date': '20130116',
 212                 'uploader': "Children's Joy Foundation",
 213                 'uploader_id': 'thecjf',
 214                 'view_count': int,
 215             },
 216         },
 217         {
 218             # dailymotion embed
 219             'url': 'https://vk.com/video-37468416_456239855',
 220             'info_dict': {
 221                 'id': 'k3lz2cmXyRuJQSjGHUv',
 222                 'ext': 'mp4',
 223                 'title': 'md5:d52606645c20b0ddbb21655adaa4f56f',
 224                 'description': 'md5:c651358f03c56f1150b555c26d90a0fd',
 225                 'uploader': 'AniLibria.Tv',
 226                 'upload_date': '20160914',
 227                 'uploader_id': 'x1p5vl5',
 228                 'timestamp': 1473877246,
 229             },
 230             'params': {
 231                 'skip_download': True,
 232             },
 233         },
 234         {
 235             # video key is extra_data not url\d+
 236             'url': 'http://vk.com/video-110305615_171782105',
 237             'md5': 'e13fcda136f99764872e739d13fac1d1',
 238             'info_dict': {
 239                 'id': '171782105',
 240                 'ext': 'mp4',
 241                 'title': 'S-Dance, репетиции к The way show',
 242                 'uploader': 'THE WAY SHOW | 17 апреля',
 243                 'timestamp': 1454870100,
 244                 'upload_date': '20160207',
 245                 'view_count': int,
 246             },
 247         },
 248         {
 249             # finished live stream, live_mp4
 250             'url': 'https://vk.com/videos-387766?z=video-387766_456242764%2Fpl_-387766_-2',
 251             'md5': '90d22d051fccbbe9becfccc615be6791',
 252             'info_dict': {
 253                 'id': '456242764',
 254                 'ext': 'mp4',
 255                 'title': 'ИгроМир 2016 — день 1',
 256                 'uploader': 'Игромания',
 257                 'duration': 5239,
 258                 'view_count': int,
 259             },
 260         },
 261         {
 262             # live stream, hls and rtmp links,most likely already finished live
 263             # stream by the time you are reading this comment
 264             'url': 'https://vk.com/video-140332_456239111',
 265             'only_matching': True,
 266         },
 267         {
 268             # removed video, just testing that we match the pattern
 269             'url': 'http://vk.com/feed?z=video-43215063_166094326%2Fbb50cacd3177146d7a',
 270             'only_matching': True,
 271         },
 272         {
 273             # age restricted video, requires vk account credentials
 274             'url': 'https://vk.com/video205387401_164765225',
 275             'only_matching': True,
 276         },
 277         {
 278             # pladform embed
 279             'url': 'https://vk.com/video-76116461_171554880',
 280             'only_matching': True,
 281         },
 282         {
 283             'url': 'http://new.vk.com/video205387401_165548505',
 284             'only_matching': True,
 285         }
 286     ]
 287
 288     def _real_extract(self, url):
 289         mobj = re.match(self._VALID_URL, url)
 290         video_id = mobj.group('videoid')
 291
 292         if video_id:
 293             info_url = 'https://vk.com/al_video.php?act=show&al=1&module=video&video=%s' % video_id
 294             # Some videos (removed?) can only be downloaded with list id specified
 295             list_id = mobj.group('list_id')
 296             if list_id:
 297                 info_url += '&list=%s' % list_id
 298         else:
 299             info_url = 'http://vk.com/video_ext.php?' + mobj.group('embed_query')
 300             video_id = '%s_%s' % (mobj.group('oid'), mobj.group('id'))
 301
 302         info_page = self._download_webpage(info_url, video_id)
 303
 304         error_message = self._html_search_regex(
 305             [r'(?s)<!><div[^>]+class="video_layer_message"[^>]*>(.+?)</div>',
 306                 r'(?s)<div[^>]+id="video_ext_msg"[^>]*>(.+?)</div>'],
 307             info_page, 'error message', default=None)
 308         if error_message:
 309             raise ExtractorError(error_message, expected=True)
 310
 311         if re.search(r'<!>/login\.php\?.*\bact=security_check', info_page):
 312             raise ExtractorError(
 313                 'You are trying to log in from an unusual location. You should confirm ownership at vk.com to log in with this IP.',
 314                 expected=True)
 315
 316         ERRORS = {
 317             r'>Видеозапись .*? была изъята из публичного доступа в связи с обращением правообладателя.<':
 318             'Video %s has been removed from public access due to rightholder complaint.',
 319
 320             r'<!>Please log in or <':
 321             'Video %s is only available for registered users, '
 322             'use --username and --password options to provide account credentials.',
 323
 324             r'<!>Unknown error':
 325             'Video %s does not exist.',
 326
 327             r'<!>Видео временно недоступно':
 328             'Video %s is temporarily unavailable.',
 329
 330             r'<!>Access denied':
 331             'Access denied to video %s.',
 332         }
 333
 334         for error_re, error_msg in ERRORS.items():
 335             if re.search(error_re, info_page):
 336                 raise ExtractorError(error_msg % video_id, expected=True)
 337
 338         youtube_url = self._search_regex(
 339             r'<iframe[^>]+src="((?:https?:)?//www.youtube.com/embed/[^"]+)"',
 340             info_page, 'youtube iframe', default=None)
 341         if youtube_url:
 342             return self.url_result(youtube_url, 'Youtube')
 343
 344         vimeo_url = VimeoIE._extract_vimeo_url(url, info_page)
 345         if vimeo_url is not None:
 346             return self.url_result(vimeo_url)
 347
 348         pladform_url = PladformIE._extract_url(info_page)
 349         if pladform_url:
 350             return self.url_result(pladform_url)
 351
 352         m_rutube = re.search(
 353             r'\ssrc="((?:https?:)?//rutube\.ru\\?/(?:video|play)\\?/embed(?:.*?))\\?"', info_page)
 354         if m_rutube is not None:
 355             rutube_url = self._proto_relative_url(
 356                 m_rutube.group(1).replace('\\', ''))
 357             return self.url_result(rutube_url)
 358
 359         dailymotion_urls = DailymotionIE._extract_urls(info_page)
 360         if dailymotion_urls:
 361             return self.url_result(dailymotion_urls[0], DailymotionIE.ie_key())
 362
 363         m_opts = re.search(r'(?s)var\s+opts\s*=\s*({.+?});', info_page)
 364         if m_opts:
 365             m_opts_url = re.search(r"url\s*:\s*'((?!/\b)[^']+)", m_opts.group(1))
 366             if m_opts_url:
 367                 opts_url = m_opts_url.group(1)
 368                 if opts_url.startswith('//'):
 369                     opts_url = 'http:' + opts_url
 370                 return self.url_result(opts_url)
 371
 372         data_json = self._search_regex(r'var\s+vars\s*=\s*({.+?});', info_page, 'vars')
 373         data = json.loads(data_json)
 374
 375         title = unescapeHTML(data['md_title'])
 376
 377         if data.get('live') == 2:
 378             title = self._live_title(title)
 379
 380         timestamp = unified_timestamp(self._html_search_regex(
 381             r'class=["\']mv_info_date[^>]+>([^<]+)(?:<|from)', info_page,
 382             'upload date', fatal=False))
 383
 384         view_count = str_to_int(self._search_regex(
 385             r'class=["\']mv_views_count[^>]+>\s*([\d,.]+)',
 386             info_page, 'view count', fatal=False))
 387
 388         formats = []
 389         for format_id, format_url in data.items():
 390             if not isinstance(format_url, compat_str) or not format_url.startswith(('http', '//', 'rtmp')):
 391                 continue
 392             if format_id.startswith(('url', 'cache')) or format_id in ('extra_data', 'live_mp4'):
 393                 height = int_or_none(self._search_regex(
 394                     r'^(?:url|cache)(\d+)', format_id, 'height', default=None))
 395                 formats.append({
 396                     'format_id': format_id,
 397                     'url': format_url,
 398                     'height': height,
 399                 })
 400             elif format_id == 'hls':
 401                 formats.extend(self._extract_m3u8_formats(
 402                     format_url, video_id, 'mp4', m3u8_id=format_id,
 403                     fatal=False, live=True))
 404             elif format_id == 'rtmp':
 405                 formats.append({
 406                     'format_id': format_id,
 407                     'url': format_url,
 408                     'ext': 'flv',
 409                 })
 410         self._sort_formats(formats)
 411
 412         return {
 413             'id': compat_str(data.get('vid') or video_id),
 414             'formats': formats,
 415             'title': title,
 416             'thumbnail': data.get('jpg'),
 417             'uploader': data.get('md_author'),
 418             'duration': data.get('duration'),
 419             'timestamp': timestamp,
 420             'view_count': view_count,
 421         }
 422
 423
 424 class VKUserVideosIE(VKBaseIE):
 425     IE_NAME = 'vk:uservideos'
 426     IE_DESC = "VK - User's Videos"
 427     _VALID_URL = r'https?://(?:(?:m|new)\.)?vk\.com/videos(?P<id>-?[0-9]+)(?!\?.*\bz=video)(?:[/?#&]|$)'
 428     _TEMPLATE_URL = 'https://vk.com/videos'
 429     _TESTS = [{
 430         'url': 'http://vk.com/videos205387401',
 431         'info_dict': {
 432             'id': '205387401',
 433             'title': "Tom Cruise's Videos",
 434         },
 435         'playlist_mincount': 4,
 436     }, {
 437         'url': 'http://vk.com/videos-77521',
 438         'only_matching': True,
 439     }, {
 440         'url': 'http://vk.com/videos-97664626?section=all',
 441         'only_matching': True,
 442     }, {
 443         'url': 'http://m.vk.com/videos205387401',
 444         'only_matching': True,
 445     }, {
 446         'url': 'http://new.vk.com/videos205387401',
 447         'only_matching': True,
 448     }]
 449
 450     def _real_extract(self, url):
 451         page_id = self._match_id(url)
 452
 453         webpage = self._download_webpage(url, page_id)
 454
 455         entries = [
 456             self.url_result(
 457                 'http://vk.com/video' + video_id, 'VK', video_id=video_id)
 458             for video_id in orderedSet(re.findall(r'href="/video(-?[0-9_]+)"', webpage))]
 459
 460         title = unescapeHTML(self._search_regex(
 461             r'<title>\s*([^<]+?)\s+\|\s+\d+\s+videos',
 462             webpage, 'title', default=page_id))
 463
 464         return self.playlist_result(entries, page_id, title)
 465
 466
 467 class VKWallPostIE(VKBaseIE):
 468     IE_NAME = 'vk:wallpost'
 469     _VALID_URL = r'https?://(?:(?:(?:(?:m|new)\.)?vk\.com/(?:[^?]+\?.*\bw=)?wall(?P<id>-?\d+_\d+)))'
 470     _TESTS = [{
 471         # public page URL, audio playlist
 472         'url': 'https://vk.com/bs.official?w=wall-23538238_35',
 473         'info_dict': {
 474             'id': '23538238_35',
 475             'title': 'Black Shadow - Wall post 23538238_35',
 476             'description': 'md5:3f84b9c4f9ef499731cf1ced9998cc0c',
 477         },
 478         'playlist': [{
 479             'md5': '5ba93864ec5b85f7ce19a9af4af080f6',
 480             'info_dict': {
 481                 'id': '135220665_111806521',
 482                 'ext': 'mp3',
 483                 'title': 'Black Shadow - Слепое Верование',
 484                 'duration': 370,
 485                 'uploader': 'Black Shadow',
 486                 'artist': 'Black Shadow',
 487                 'track': 'Слепое Верование',
 488             },
 489         }, {
 490             'md5': '4cc7e804579122b17ea95af7834c9233',
 491             'info_dict': {
 492                 'id': '135220665_111802303',
 493                 'ext': 'mp3',
 494                 'title': 'Black Shadow - Война - Негасимое Бездны Пламя!',
 495                 'duration': 423,
 496                 'uploader': 'Black Shadow',
 497                 'artist': 'Black Shadow',
 498                 'track': 'Война - Негасимое Бездны Пламя!',
 499             },
 500             'params': {
 501                 'skip_download': True,
 502             },
 503         }],
 504         'params': {
 505             'usenetrc': True,
 506         },
 507         'skip': 'Requires vk account credentials',
 508     }, {
 509         # single YouTube embed, no leading -
 510         'url': 'https://vk.com/wall85155021_6319',
 511         'info_dict': {
 512             'id': '85155021_6319',
 513             'title': 'Sergey Gorbunov - Wall post 85155021_6319',
 514         },
 515         'playlist_count': 1,
 516         'params': {
 517             'usenetrc': True,
 518         },
 519         'skip': 'Requires vk account credentials',
 520     }, {
 521         # wall page URL
 522         'url': 'https://vk.com/wall-23538238_35',
 523         'only_matching': True,
 524     }, {
 525         # mobile wall page URL
 526         'url': 'https://m.vk.com/wall-23538238_35',
 527         'only_matching': True,
 528     }]
 529
 530     def _real_extract(self, url):
 531         post_id = self._match_id(url)
 532
 533         wall_url = 'https://vk.com/wall%s' % post_id
 534
 535         post_id = remove_start(post_id, '-')
 536
 537         webpage = self._download_webpage(wall_url, post_id)
 538
 539         error = self._html_search_regex(
 540             r'>Error</div>\s*<div[^>]+class=["\']body["\'][^>]*>([^<]+)',
 541             webpage, 'error', default=None)
 542         if error:
 543             raise ExtractorError('VK said: %s' % error, expected=True)
 544
 545         description = clean_html(get_element_by_class('wall_post_text', webpage))
 546         uploader = clean_html(get_element_by_class('author', webpage))
 547         thumbnail = self._og_search_thumbnail(webpage)
 548
 549         entries = []
 550
 551         audio_ids = re.findall(r'data-full-id=["\'](\d+_\d+)', webpage)
 552         if audio_ids:
 553             al_audio = self._download_webpage(
 554                 'https://vk.com/al_audio.php', post_id,
 555                 note='Downloading audio info', fatal=False,
 556                 data=urlencode_postdata({
 557                     'act': 'reload_audio',
 558                     'al': '1',
 559                     'ids': ','.join(audio_ids)
 560                 }))
 561             if al_audio:
 562                 Audio = collections.namedtuple(
 563                     'Audio', ['id', 'user_id', 'url', 'track', 'artist', 'duration'])
 564                 audios = self._parse_json(
 565                     self._search_regex(
 566                         r'<!json>(.+?)<!>', al_audio, 'audios', default='[]'),
 567                     post_id, fatal=False, transform_source=unescapeHTML)
 568                 if isinstance(audios, list):
 569                     for audio in audios:
 570                         a = Audio._make(audio[:6])
 571                         entries.append({
 572                             'id': '%s_%s' % (a.user_id, a.id),
 573                             'url': a.url,
 574                             'title': '%s - %s' % (a.artist, a.track) if a.artist and a.track else a.id,
 575                             'thumbnail': thumbnail,
 576                             'duration': a.duration,
 577                             'uploader': uploader,
 578                             'artist': a.artist,
 579                             'track': a.track,
 580                         })
 581
 582         for video in re.finditer(
 583                 r'<a[^>]+href=(["\'])(?P<url>/video(?:-?[\d_]+).*?)\1', webpage):
 584             entries.append(self.url_result(
 585                 compat_urlparse.urljoin(url, video.group('url')), VKIE.ie_key()))
 586
 587         title = 'Wall post %s' % post_id
 588
 589         return self.playlist_result(
 590             orderedSet(entries), post_id,
 591             '%s - %s' % (uploader, title) if uploader else title,
 592             description)