Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/vk.py

   1 # encoding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5 import json
   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     parse_duration,
  20     remove_start,
  21     str_to_int,
  22     unescapeHTML,
  23     unified_strdate,
  24     urlencode_postdata,
  25 )
  26 from .vimeo import VimeoIE
  27 from .pladform import PladformIE
  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         cookies = url_handle.headers.get('Set-Cookie')
  56         if cookies:
  57             if sys.version_info[0] >= 3:
  58                 cookies = cookies.encode('iso-8859-1')
  59             cookies = cookies.decode('utf-8')
  60             remixlhk = re.search(r'remixlhk=(.+?);.*?\bdomain=(.+?)(?:[,;]|$)', cookies)
  61             if remixlhk:
  62                 value, domain = remixlhk.groups()
  63                 self._set_cookie(domain, 'remixlhk', value)
  64
  65         login_page = self._download_webpage(
  66             'https://login.vk.com/?act=login', None,
  67             note='Logging in as %s' % username,
  68             data=urlencode_postdata(login_form))
  69
  70         if re.search(r'onLoginFailed', login_page):
  71             raise ExtractorError(
  72                 'Unable to login, incorrect username and/or password', expected=True)
  73
  74     def _real_initialize(self):
  75         self._login()
  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': '0deae91935c54e00003c2a00646315f0',
 100             'info_dict': {
 101                 'id': '162222515',
 102                 'ext': 'flv',
 103                 'title': 'ProtivoGunz - Хуёвая песня',
 104                 'uploader': 're:(?:Noize MC|Alexander Ilyashenko).*',
 105                 'duration': 195,
 106                 'upload_date': '20120212',
 107                 'view_count': int,
 108             },
 109         },
 110         {
 111             'url': 'http://vk.com/video205387401_165548505',
 112             'md5': '6c0aeb2e90396ba97035b9cbde548700',
 113             'info_dict': {
 114                 'id': '165548505',
 115                 'ext': 'mp4',
 116                 'uploader': 'Tom Cruise',
 117                 'title': 'No name',
 118                 'duration': 9,
 119                 'upload_date': '20130721',
 120                 'view_count': int,
 121             }
 122         },
 123         {
 124             'note': 'Embedded video',
 125             'url': 'http://vk.com/video_ext.php?oid=32194266&id=162925554&hash=7d8c2e0d5e05aeaa&hd=1',
 126             'md5': 'c7ce8f1f87bec05b3de07fdeafe21a0a',
 127             'info_dict': {
 128                 'id': '162925554',
 129                 'ext': 'mp4',
 130                 'uploader': 'Vladimir Gavrin',
 131                 'title': 'Lin Dan',
 132                 'duration': 101,
 133                 'upload_date': '20120730',
 134                 'view_count': int,
 135             },
 136             'skip': 'This video has been removed from public access.',
 137         },
 138         {
 139             # VIDEO NOW REMOVED
 140             # please update if you find a video whose URL follows the same pattern
 141             'url': 'http://vk.com/video-8871596_164049491',
 142             'md5': 'a590bcaf3d543576c9bd162812387666',
 143             'note': 'Only available for registered users',
 144             'info_dict': {
 145                 'id': '164049491',
 146                 'ext': 'mp4',
 147                 'uploader': 'Триллеры',
 148                 'title': '► Бойцовский клуб / Fight Club 1999 [HD 720]',
 149                 'duration': 8352,
 150                 'upload_date': '20121218',
 151                 'view_count': int,
 152             },
 153             'skip': 'Requires vk account credentials',
 154         },
 155         {
 156             'url': 'http://vk.com/hd_kino_mania?z=video-43215063_168067957%2F15c66b9b533119788d',
 157             'md5': '4d7a5ef8cf114dfa09577e57b2993202',
 158             'info_dict': {
 159                 'id': '168067957',
 160                 'ext': 'mp4',
 161                 'uploader': 'Киномания - лучшее из мира кино',
 162                 'title': ' ',
 163                 'duration': 7291,
 164                 'upload_date': '20140328',
 165             },
 166             'skip': 'Requires vk account credentials',
 167         },
 168         {
 169             'url': 'http://m.vk.com/video-43215063_169084319?list=125c627d1aa1cebb83&from=wall-43215063_2566540',
 170             'md5': '0c45586baa71b7cb1d0784ee3f4e00a6',
 171             'note': 'ivi.ru embed',
 172             'info_dict': {
 173                 'id': '60690',
 174                 'ext': 'mp4',
 175                 'title': 'Книга Илая',
 176                 'duration': 6771,
 177                 'upload_date': '20140626',
 178                 'view_count': int,
 179             },
 180             'skip': 'Only works from Russia',
 181         },
 182         {
 183             # video (removed?) only available with list id
 184             'url': 'https://vk.com/video30481095_171201961?list=8764ae2d21f14088d4',
 185             'md5': '091287af5402239a1051c37ec7b92913',
 186             'info_dict': {
 187                 'id': '171201961',
 188                 'ext': 'mp4',
 189                 'title': 'ТюменцевВВ_09.07.2015',
 190                 'uploader': 'Anton Ivanov',
 191                 'duration': 109,
 192                 'upload_date': '20150709',
 193                 'view_count': int,
 194             },
 195         },
 196         {
 197             # youtube embed
 198             'url': 'https://vk.com/video276849682_170681728',
 199             'info_dict': {
 200                 'id': 'V3K4mi0SYkc',
 201                 'ext': 'webm',
 202                 'title': "DSWD Awards 'Children's Joy Foundation, Inc.' Certificate of Registration and License to Operate",
 203                 'description': 'md5:d9903938abdc74c738af77f527ca0596',
 204                 'duration': 178,
 205                 'upload_date': '20130116',
 206                 'uploader': "Children's Joy Foundation",
 207                 'uploader_id': 'thecjf',
 208                 'view_count': int,
 209             },
 210         },
 211         {
 212             # video key is extra_data not url\d+
 213             'url': 'http://vk.com/video-110305615_171782105',
 214             'md5': 'e13fcda136f99764872e739d13fac1d1',
 215             'info_dict': {
 216                 'id': '171782105',
 217                 'ext': 'mp4',
 218                 'title': 'S-Dance, репетиции к The way show',
 219                 'uploader': 'THE WAY SHOW | 17 апреля',
 220                 'upload_date': '20160207',
 221                 'view_count': int,
 222             },
 223         },
 224         {
 225             # removed video, just testing that we match the pattern
 226             'url': 'http://vk.com/feed?z=video-43215063_166094326%2Fbb50cacd3177146d7a',
 227             'only_matching': True,
 228         },
 229         {
 230             # age restricted video, requires vk account credentials
 231             'url': 'https://vk.com/video205387401_164765225',
 232             'only_matching': True,
 233         },
 234         {
 235             # pladform embed
 236             'url': 'https://vk.com/video-76116461_171554880',
 237             'only_matching': True,
 238         },
 239         {
 240             'url': 'http://new.vk.com/video205387401_165548505',
 241             'only_matching': True,
 242         }
 243     ]
 244
 245     def _real_extract(self, url):
 246         mobj = re.match(self._VALID_URL, url)
 247         video_id = mobj.group('videoid')
 248
 249         if video_id:
 250             info_url = 'https://vk.com/al_video.php?act=show&al=1&module=video&video=%s' % video_id
 251             # Some videos (removed?) can only be downloaded with list id specified
 252             list_id = mobj.group('list_id')
 253             if list_id:
 254                 info_url += '&list=%s' % list_id
 255         else:
 256             info_url = 'http://vk.com/video_ext.php?' + mobj.group('embed_query')
 257             video_id = '%s_%s' % (mobj.group('oid'), mobj.group('id'))
 258
 259         info_page = self._download_webpage(info_url, video_id)
 260
 261         error_message = self._html_search_regex(
 262             [r'(?s)<!><div[^>]+class="video_layer_message"[^>]*>(.+?)</div>',
 263                 r'(?s)<div[^>]+id="video_ext_msg"[^>]*>(.+?)</div>'],
 264             info_page, 'error message', default=None)
 265         if error_message:
 266             raise ExtractorError(error_message, expected=True)
 267
 268         if re.search(r'<!>/login\.php\?.*\bact=security_check', info_page):
 269             raise ExtractorError(
 270                 'You are trying to log in from an unusual location. You should confirm ownership at vk.com to log in with this IP.',
 271                 expected=True)
 272
 273         ERRORS = {
 274             r'>Видеозапись .*? была изъята из публичного доступа в связи с обращением правообладателя.<':
 275             'Video %s has been removed from public access due to rightholder complaint.',
 276
 277             r'<!>Please log in or <':
 278             'Video %s is only available for registered users, '
 279             'use --username and --password options to provide account credentials.',
 280
 281             r'<!>Unknown error':
 282             'Video %s does not exist.',
 283
 284             r'<!>Видео временно недоступно':
 285             'Video %s is temporarily unavailable.',
 286
 287             r'<!>Access denied':
 288             'Access denied to video %s.',
 289         }
 290
 291         for error_re, error_msg in ERRORS.items():
 292             if re.search(error_re, info_page):
 293                 raise ExtractorError(error_msg % video_id, expected=True)
 294
 295         youtube_url = self._search_regex(
 296             r'<iframe[^>]+src="((?:https?:)?//www.youtube.com/embed/[^"]+)"',
 297             info_page, 'youtube iframe', default=None)
 298         if youtube_url:
 299             return self.url_result(youtube_url, 'Youtube')
 300
 301         vimeo_url = VimeoIE._extract_vimeo_url(url, info_page)
 302         if vimeo_url is not None:
 303             return self.url_result(vimeo_url)
 304
 305         pladform_url = PladformIE._extract_url(info_page)
 306         if pladform_url:
 307             return self.url_result(pladform_url)
 308
 309         m_rutube = re.search(
 310             r'\ssrc="((?:https?:)?//rutube\.ru\\?/(?:video|play)\\?/embed(?:.*?))\\?"', info_page)
 311         if m_rutube is not None:
 312             rutube_url = self._proto_relative_url(
 313                 m_rutube.group(1).replace('\\', ''))
 314             return self.url_result(rutube_url)
 315
 316         m_opts = re.search(r'(?s)var\s+opts\s*=\s*({.+?});', info_page)
 317         if m_opts:
 318             m_opts_url = re.search(r"url\s*:\s*'((?!/\b)[^']+)", m_opts.group(1))
 319             if m_opts_url:
 320                 opts_url = m_opts_url.group(1)
 321                 if opts_url.startswith('//'):
 322                     opts_url = 'http:' + opts_url
 323                 return self.url_result(opts_url)
 324
 325         data_json = self._search_regex(r'var\s+vars\s*=\s*({.+?});', info_page, 'vars')
 326         data = json.loads(data_json)
 327
 328         # Extract upload date
 329         upload_date = None
 330         mobj = re.search(r'id="mv_date(?:_views)?_wrap"[^>]*>([a-zA-Z]+ [0-9]+), ([0-9]+) at', info_page)
 331         if mobj is not None:
 332             mobj.group(1) + ' ' + mobj.group(2)
 333             upload_date = unified_strdate(mobj.group(1) + ' ' + mobj.group(2))
 334
 335         view_count = None
 336         views = self._html_search_regex(
 337             r'"mv_views_count_number"[^>]*>(.+?\bviews?)<',
 338             info_page, 'view count', default=None)
 339         if views:
 340             view_count = str_to_int(self._search_regex(
 341                 r'([\d,.]+)', views, 'view count', fatal=False))
 342
 343         formats = []
 344         for k, v in data.items():
 345             if not k.startswith('url') and not k.startswith('cache') and k != 'extra_data' or not v:
 346                 continue
 347             height = int_or_none(self._search_regex(
 348                 r'^(?:url|cache)(\d+)', k, 'height', default=None))
 349             formats.append({
 350                 'format_id': k,
 351                 'url': v,
 352                 'height': height,
 353             })
 354         self._sort_formats(formats)
 355
 356         return {
 357             'id': compat_str(data['vid']),
 358             'formats': formats,
 359             'title': unescapeHTML(data['md_title']),
 360             'thumbnail': data.get('jpg'),
 361             'uploader': data.get('md_author'),
 362             'duration': data.get('duration'),
 363             'upload_date': upload_date,
 364             'view_count': view_count,
 365         }
 366
 367
 368 class VKUserVideosIE(VKBaseIE):
 369     IE_NAME = 'vk:uservideos'
 370     IE_DESC = "VK - User's Videos"
 371     _VALID_URL = r'https?://(?:(?:m|new)\.)?vk\.com/videos(?P<id>-?[0-9]+)(?!\?.*\bz=video)(?:[/?#&]|$)'
 372     _TEMPLATE_URL = 'https://vk.com/videos'
 373     _TESTS = [{
 374         'url': 'http://vk.com/videos205387401',
 375         'info_dict': {
 376             'id': '205387401',
 377             'title': "Tom Cruise's Videos",
 378         },
 379         'playlist_mincount': 4,
 380     }, {
 381         'url': 'http://vk.com/videos-77521',
 382         'only_matching': True,
 383     }, {
 384         'url': 'http://vk.com/videos-97664626?section=all',
 385         'only_matching': True,
 386     }, {
 387         'url': 'http://m.vk.com/videos205387401',
 388         'only_matching': True,
 389     }, {
 390         'url': 'http://new.vk.com/videos205387401',
 391         'only_matching': True,
 392     }]
 393
 394     def _real_extract(self, url):
 395         page_id = self._match_id(url)
 396
 397         webpage = self._download_webpage(url, page_id)
 398
 399         entries = [
 400             self.url_result(
 401                 'http://vk.com/video' + video_id, 'VK', video_id=video_id)
 402             for video_id in orderedSet(re.findall(r'href="/video(-?[0-9_]+)"', webpage))]
 403
 404         title = unescapeHTML(self._search_regex(
 405             r'<title>\s*([^<]+?)\s+\|\s+\d+\s+videos',
 406             webpage, 'title', default=page_id))
 407
 408         return self.playlist_result(entries, page_id, title)
 409
 410
 411 class VKWallPostIE(VKBaseIE):
 412     IE_NAME = 'vk:wallpost'
 413     _VALID_URL = r'https?://(?:(?:(?:(?:m|new)\.)?vk\.com/(?:[^?]+\?.*\bw=)?wall(?P<id>-?\d+_\d+)))'
 414     _TESTS = [{
 415         # public page URL, audio playlist
 416         'url': 'https://vk.com/bs.official?w=wall-23538238_35',
 417         'info_dict': {
 418             'id': '23538238_35',
 419             'title': 'Black Shadow - Wall post 23538238_35',
 420             'description': 'md5:3f84b9c4f9ef499731cf1ced9998cc0c',
 421         },
 422         'playlist': [{
 423             'md5': '5ba93864ec5b85f7ce19a9af4af080f6',
 424             'info_dict': {
 425                 'id': '135220665_111806521',
 426                 'ext': 'mp3',
 427                 'title': 'Black Shadow - Слепое Верование',
 428                 'duration': 370,
 429                 'uploader': 'Black Shadow',
 430                 'artist': 'Black Shadow',
 431                 'track': 'Слепое Верование',
 432             },
 433         }, {
 434             'md5': '4cc7e804579122b17ea95af7834c9233',
 435             'info_dict': {
 436                 'id': '135220665_111802303',
 437                 'ext': 'mp3',
 438                 'title': 'Black Shadow - Война - Негасимое Бездны Пламя!',
 439                 'duration': 423,
 440                 'uploader': 'Black Shadow',
 441                 'artist': 'Black Shadow',
 442                 'track': 'Война - Негасимое Бездны Пламя!',
 443             },
 444             'params': {
 445                 'skip_download': True,
 446             },
 447         }],
 448         'skip': 'Requires vk account credentials',
 449     }, {
 450         # single YouTube embed, no leading -
 451         'url': 'https://vk.com/wall85155021_6319',
 452         'info_dict': {
 453             'id': '85155021_6319',
 454             'title': 'Sergey Gorbunov - Wall post 85155021_6319',
 455         },
 456         'playlist_count': 1,
 457         'skip': 'Requires vk account credentials',
 458     }, {
 459         # wall page URL
 460         'url': 'https://vk.com/wall-23538238_35',
 461         'only_matching': True,
 462     }, {
 463         # mobile wall page URL
 464         'url': 'https://m.vk.com/wall-23538238_35',
 465         'only_matching': True,
 466     }]
 467
 468     def _real_extract(self, url):
 469         post_id = self._match_id(url)
 470
 471         wall_url = 'https://vk.com/wall%s' % post_id
 472
 473         post_id = remove_start(post_id, '-')
 474
 475         webpage = self._download_webpage(wall_url, post_id)
 476
 477         error = self._html_search_regex(
 478             r'>Error</div>\s*<div[^>]+class=["\']body["\'][^>]*>([^<]+)',
 479             webpage, 'error', default=None)
 480         if error:
 481             raise ExtractorError('VK said: %s' % error, expected=True)
 482
 483         description = clean_html(get_element_by_class('wall_post_text', webpage))
 484         uploader = clean_html(get_element_by_class(
 485             'fw_post_author', webpage)) or self._og_search_description(webpage)
 486         thumbnail = self._og_search_thumbnail(webpage)
 487
 488         entries = []
 489
 490         for audio in re.finditer(r'''(?sx)
 491                             <input[^>]+
 492                                 id=(?P<q1>["\'])audio_info(?P<id>\d+_\d+).*?(?P=q1)[^>]+
 493                                 value=(?P<q2>["\'])(?P<url>http.+?)(?P=q2)
 494                                 .+?
 495                             </table>''', webpage):
 496             audio_html = audio.group(0)
 497             audio_id = audio.group('id')
 498             duration = parse_duration(get_element_by_class('duration', audio_html))
 499             track = self._html_search_regex(
 500                 r'<span[^>]+id=["\']title%s[^>]*>([^<]+)' % audio_id,
 501                 audio_html, 'title', default=None)
 502             artist = self._html_search_regex(
 503                 r'>([^<]+)</a></b>\s*&ndash', audio_html,
 504                 'artist', default=None)
 505             entries.append({
 506                 'id': audio_id,
 507                 'url': audio.group('url'),
 508                 'title': '%s - %s' % (artist, track) if artist and track else audio_id,
 509                 'thumbnail': thumbnail,
 510                 'duration': duration,
 511                 'uploader': uploader,
 512                 'artist': artist,
 513                 'track': track,
 514             })
 515
 516         for video in re.finditer(
 517                 r'<a[^>]+href=(["\'])(?P<url>/video(?:-?[\d_]+).*?)\1', webpage):
 518             entries.append(self.url_result(
 519                 compat_urlparse.urljoin(url, video.group('url')), VKIE.ie_key()))
 520
 521         title = 'Wall post %s' % post_id
 522
 523         return self.playlist_result(
 524             orderedSet(entries), post_id,
 525             '%s - %s' % (uploader, title) if uploader else title,
 526             description)