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 compat_str
  10 from ..utils import (
  11     ExtractorError,
  12     int_or_none,
  13     orderedSet,
  14     str_to_int,
  15     unescapeHTML,
  16     unified_strdate,
  17     urlencode_postdata,
  18 )
  19 from .vimeo import VimeoIE
  20 from .pladform import PladformIE
  21
  22
  23 class VKIE(InfoExtractor):
  24     IE_NAME = 'vk'
  25     IE_DESC = 'VK'
  26     _VALID_URL = r'''(?x)
  27                     https?://
  28                         (?:
  29                             (?:
  30                                 (?:(?:m|new)\.)?vk\.com/video_|
  31                                 (?:www\.)?daxab.com/
  32                             )
  33                             ext\.php\?(?P<embed_query>.*?\boid=(?P<oid>-?\d+).*?\bid=(?P<id>\d+).*)|
  34                             (?:
  35                                 (?:(?:m|new)\.)?vk\.com/(?:.+?\?.*?z=)?video|
  36                                 (?:www\.)?daxab.com/embed/
  37                             )
  38                             (?P<videoid>-?\d+_\d+)(?:.*\blist=(?P<list_id>[\da-f]+))?
  39                         )
  40                     '''
  41     _NETRC_MACHINE = 'vk'
  42
  43     _TESTS = [
  44         {
  45             'url': 'http://vk.com/videos-77521?z=video-77521_162222515%2Fclub77521',
  46             'md5': '0deae91935c54e00003c2a00646315f0',
  47             'info_dict': {
  48                 'id': '162222515',
  49                 'ext': 'flv',
  50                 'title': 'ProtivoGunz - Хуёвая песня',
  51                 'uploader': 're:(?:Noize MC|Alexander Ilyashenko).*',
  52                 'duration': 195,
  53                 'upload_date': '20120212',
  54                 'view_count': int,
  55             },
  56         },
  57         {
  58             'url': 'http://vk.com/video205387401_165548505',
  59             'md5': '6c0aeb2e90396ba97035b9cbde548700',
  60             'info_dict': {
  61                 'id': '165548505',
  62                 'ext': 'mp4',
  63                 'uploader': 'Tom Cruise',
  64                 'title': 'No name',
  65                 'duration': 9,
  66                 'upload_date': '20130721',
  67                 'view_count': int,
  68             }
  69         },
  70         {
  71             'note': 'Embedded video',
  72             'url': 'http://vk.com/video_ext.php?oid=32194266&id=162925554&hash=7d8c2e0d5e05aeaa&hd=1',
  73             'md5': 'c7ce8f1f87bec05b3de07fdeafe21a0a',
  74             'info_dict': {
  75                 'id': '162925554',
  76                 'ext': 'mp4',
  77                 'uploader': 'Vladimir Gavrin',
  78                 'title': 'Lin Dan',
  79                 'duration': 101,
  80                 'upload_date': '20120730',
  81                 'view_count': int,
  82             },
  83             'skip': 'This video has been removed from public access.',
  84         },
  85         {
  86             # VIDEO NOW REMOVED
  87             # please update if you find a video whose URL follows the same pattern
  88             'url': 'http://vk.com/video-8871596_164049491',
  89             'md5': 'a590bcaf3d543576c9bd162812387666',
  90             'note': 'Only available for registered users',
  91             'info_dict': {
  92                 'id': '164049491',
  93                 'ext': 'mp4',
  94                 'uploader': 'Триллеры',
  95                 'title': '► Бойцовский клуб / Fight Club 1999 [HD 720]',
  96                 'duration': 8352,
  97                 'upload_date': '20121218',
  98                 'view_count': int,
  99             },
 100             'skip': 'Requires vk account credentials',
 101         },
 102         {
 103             'url': 'http://vk.com/hd_kino_mania?z=video-43215063_168067957%2F15c66b9b533119788d',
 104             'md5': '4d7a5ef8cf114dfa09577e57b2993202',
 105             'info_dict': {
 106                 'id': '168067957',
 107                 'ext': 'mp4',
 108                 'uploader': 'Киномания - лучшее из мира кино',
 109                 'title': ' ',
 110                 'duration': 7291,
 111                 'upload_date': '20140328',
 112             },
 113             'skip': 'Requires vk account credentials',
 114         },
 115         {
 116             'url': 'http://m.vk.com/video-43215063_169084319?list=125c627d1aa1cebb83&from=wall-43215063_2566540',
 117             'md5': '0c45586baa71b7cb1d0784ee3f4e00a6',
 118             'note': 'ivi.ru embed',
 119             'info_dict': {
 120                 'id': '60690',
 121                 'ext': 'mp4',
 122                 'title': 'Книга Илая',
 123                 'duration': 6771,
 124                 'upload_date': '20140626',
 125                 'view_count': int,
 126             },
 127             'skip': 'Only works from Russia',
 128         },
 129         {
 130             # video (removed?) only available with list id
 131             'url': 'https://vk.com/video30481095_171201961?list=8764ae2d21f14088d4',
 132             'md5': '091287af5402239a1051c37ec7b92913',
 133             'info_dict': {
 134                 'id': '171201961',
 135                 'ext': 'mp4',
 136                 'title': 'ТюменцевВВ_09.07.2015',
 137                 'uploader': 'Anton Ivanov',
 138                 'duration': 109,
 139                 'upload_date': '20150709',
 140                 'view_count': int,
 141             },
 142         },
 143         {
 144             # youtube embed
 145             'url': 'https://vk.com/video276849682_170681728',
 146             'info_dict': {
 147                 'id': 'V3K4mi0SYkc',
 148                 'ext': 'webm',
 149                 'title': "DSWD Awards 'Children's Joy Foundation, Inc.' Certificate of Registration and License to Operate",
 150                 'description': 'md5:d9903938abdc74c738af77f527ca0596',
 151                 'duration': 178,
 152                 'upload_date': '20130116',
 153                 'uploader': "Children's Joy Foundation",
 154                 'uploader_id': 'thecjf',
 155                 'view_count': int,
 156             },
 157         },
 158         {
 159             # video key is extra_data not url\d+
 160             'url': 'http://vk.com/video-110305615_171782105',
 161             'md5': 'e13fcda136f99764872e739d13fac1d1',
 162             'info_dict': {
 163                 'id': '171782105',
 164                 'ext': 'mp4',
 165                 'title': 'S-Dance, репетиции к The way show',
 166                 'uploader': 'THE WAY SHOW | 17 апреля',
 167                 'upload_date': '20160207',
 168                 'view_count': int,
 169             },
 170         },
 171         {
 172             # removed video, just testing that we match the pattern
 173             'url': 'http://vk.com/feed?z=video-43215063_166094326%2Fbb50cacd3177146d7a',
 174             'only_matching': True,
 175         },
 176         {
 177             # age restricted video, requires vk account credentials
 178             'url': 'https://vk.com/video205387401_164765225',
 179             'only_matching': True,
 180         },
 181         {
 182             # pladform embed
 183             'url': 'https://vk.com/video-76116461_171554880',
 184             'only_matching': True,
 185         },
 186         {
 187             'url': 'http://new.vk.com/video205387401_165548505',
 188             'only_matching': True,
 189         }
 190     ]
 191
 192     def _login(self):
 193         (username, password) = self._get_login_info()
 194         if username is None:
 195             return
 196
 197         login_page, url_handle = self._download_webpage_handle(
 198             'https://vk.com', None, 'Downloading login page')
 199
 200         login_form = self._hidden_inputs(login_page)
 201
 202         login_form.update({
 203             'email': username.encode('cp1251'),
 204             'pass': password.encode('cp1251'),
 205         })
 206
 207         # https://new.vk.com/ serves two same remixlhk cookies in Set-Cookie header
 208         # and expects the first one to be set rather than second (see
 209         # https://github.com/rg3/youtube-dl/issues/9841#issuecomment-227871201).
 210         # As of RFC6265 the newer one cookie should be set into cookie store
 211         # what actually happens.
 212         # We will workaround this VK issue by resetting the remixlhk cookie to
 213         # the first one manually.
 214         cookies = url_handle.headers.get('Set-Cookie')
 215         if sys.version_info[0] >= 3:
 216             cookies = cookies.encode('iso-8859-1')
 217         cookies = cookies.decode('utf-8')
 218         remixlhk = re.search(r'remixlhk=(.+?);.*?\bdomain=(.+?)(?:[,;]|$)', cookies)
 219         if remixlhk:
 220             value, domain = remixlhk.groups()
 221             self._set_cookie(domain, 'remixlhk', value)
 222
 223         login_page = self._download_webpage(
 224             'https://login.vk.com/?act=login', None,
 225             note='Logging in as %s' % username,
 226             data=urlencode_postdata(login_form))
 227
 228         if re.search(r'onLoginFailed', login_page):
 229             raise ExtractorError(
 230                 'Unable to login, incorrect username and/or password', expected=True)
 231
 232     def _real_initialize(self):
 233         self._login()
 234
 235     def _real_extract(self, url):
 236         mobj = re.match(self._VALID_URL, url)
 237         video_id = mobj.group('videoid')
 238
 239         if video_id:
 240             info_url = 'https://vk.com/al_video.php?act=show&al=1&module=video&video=%s' % video_id
 241             # Some videos (removed?) can only be downloaded with list id specified
 242             list_id = mobj.group('list_id')
 243             if list_id:
 244                 info_url += '&list=%s' % list_id
 245         else:
 246             info_url = 'http://vk.com/video_ext.php?' + mobj.group('embed_query')
 247             video_id = '%s_%s' % (mobj.group('oid'), mobj.group('id'))
 248
 249         info_page = self._download_webpage(info_url, video_id)
 250
 251         error_message = self._html_search_regex(
 252             [r'(?s)<!><div[^>]+class="video_layer_message"[^>]*>(.+?)</div>',
 253                 r'(?s)<div[^>]+id="video_ext_msg"[^>]*>(.+?)</div>'],
 254             info_page, 'error message', default=None)
 255         if error_message:
 256             raise ExtractorError(error_message, expected=True)
 257
 258         if re.search(r'<!>/login\.php\?.*\bact=security_check', info_page):
 259             raise ExtractorError(
 260                 'You are trying to log in from an unusual location. You should confirm ownership at vk.com to log in with this IP.',
 261                 expected=True)
 262
 263         ERRORS = {
 264             r'>Видеозапись .*? была изъята из публичного доступа в связи с обращением правообладателя.<':
 265             'Video %s has been removed from public access due to rightholder complaint.',
 266
 267             r'<!>Please log in or <':
 268             'Video %s is only available for registered users, '
 269             'use --username and --password options to provide account credentials.',
 270
 271             r'<!>Unknown error':
 272             'Video %s does not exist.',
 273
 274             r'<!>Видео временно недоступно':
 275             'Video %s is temporarily unavailable.',
 276
 277             r'<!>Access denied':
 278             'Access denied to video %s.',
 279         }
 280
 281         for error_re, error_msg in ERRORS.items():
 282             if re.search(error_re, info_page):
 283                 raise ExtractorError(error_msg % video_id, expected=True)
 284
 285         youtube_url = self._search_regex(
 286             r'<iframe[^>]+src="((?:https?:)?//www.youtube.com/embed/[^"]+)"',
 287             info_page, 'youtube iframe', default=None)
 288         if youtube_url:
 289             return self.url_result(youtube_url, 'Youtube')
 290
 291         vimeo_url = VimeoIE._extract_vimeo_url(url, info_page)
 292         if vimeo_url is not None:
 293             return self.url_result(vimeo_url)
 294
 295         pladform_url = PladformIE._extract_url(info_page)
 296         if pladform_url:
 297             return self.url_result(pladform_url)
 298
 299         m_rutube = re.search(
 300             r'\ssrc="((?:https?:)?//rutube\.ru\\?/(?:video|play)\\?/embed(?:.*?))\\?"', info_page)
 301         if m_rutube is not None:
 302             rutube_url = self._proto_relative_url(
 303                 m_rutube.group(1).replace('\\', ''))
 304             return self.url_result(rutube_url)
 305
 306         m_opts = re.search(r'(?s)var\s+opts\s*=\s*({.+?});', info_page)
 307         if m_opts:
 308             m_opts_url = re.search(r"url\s*:\s*'((?!/\b)[^']+)", m_opts.group(1))
 309             if m_opts_url:
 310                 opts_url = m_opts_url.group(1)
 311                 if opts_url.startswith('//'):
 312                     opts_url = 'http:' + opts_url
 313                 return self.url_result(opts_url)
 314
 315         data_json = self._search_regex(r'var\s+vars\s*=\s*({.+?});', info_page, 'vars')
 316         data = json.loads(data_json)
 317
 318         # Extract upload date
 319         upload_date = None
 320         mobj = re.search(r'id="mv_date(?:_views)?_wrap"[^>]*>([a-zA-Z]+ [0-9]+), ([0-9]+) at', info_page)
 321         if mobj is not None:
 322             mobj.group(1) + ' ' + mobj.group(2)
 323             upload_date = unified_strdate(mobj.group(1) + ' ' + mobj.group(2))
 324
 325         view_count = None
 326         views = self._html_search_regex(
 327             r'"mv_views_count_number"[^>]*>(.+?\bviews?)<',
 328             info_page, 'view count', default=None)
 329         if views:
 330             view_count = str_to_int(self._search_regex(
 331                 r'([\d,.]+)', views, 'view count', fatal=False))
 332
 333         formats = []
 334         for k, v in data.items():
 335             if not k.startswith('url') and not k.startswith('cache') and k != 'extra_data' or not v:
 336                 continue
 337             height = int_or_none(self._search_regex(
 338                 r'^(?:url|cache)(\d+)', k, 'height', default=None))
 339             formats.append({
 340                 'format_id': k,
 341                 'url': v,
 342                 'height': height,
 343             })
 344         self._sort_formats(formats)
 345
 346         return {
 347             'id': compat_str(data['vid']),
 348             'formats': formats,
 349             'title': unescapeHTML(data['md_title']),
 350             'thumbnail': data.get('jpg'),
 351             'uploader': data.get('md_author'),
 352             'duration': data.get('duration'),
 353             'upload_date': upload_date,
 354             'view_count': view_count,
 355         }
 356
 357
 358 class VKUserVideosIE(InfoExtractor):
 359     IE_NAME = 'vk:uservideos'
 360     IE_DESC = "VK - User's Videos"
 361     _VALID_URL = r'https?://(?:(?:m|new)\.)?vk\.com/videos(?P<id>-?[0-9]+)(?!\?.*\bz=video)(?:[/?#&]|$)'
 362     _TEMPLATE_URL = 'https://vk.com/videos'
 363     _TESTS = [{
 364         'url': 'http://vk.com/videos205387401',
 365         'info_dict': {
 366             'id': '205387401',
 367             'title': "Tom Cruise's Videos",
 368         },
 369         'playlist_mincount': 4,
 370     }, {
 371         'url': 'http://vk.com/videos-77521',
 372         'only_matching': True,
 373     }, {
 374         'url': 'http://vk.com/videos-97664626?section=all',
 375         'only_matching': True,
 376     }, {
 377         'url': 'http://m.vk.com/videos205387401',
 378         'only_matching': True,
 379     }, {
 380         'url': 'http://new.vk.com/videos205387401',
 381         'only_matching': True,
 382     }]
 383
 384     def _real_extract(self, url):
 385         page_id = self._match_id(url)
 386
 387         webpage = self._download_webpage(url, page_id)
 388
 389         entries = [
 390             self.url_result(
 391                 'http://vk.com/video' + video_id, 'VK', video_id=video_id)
 392             for video_id in orderedSet(re.findall(r'href="/video(-?[0-9_]+)"', webpage))]
 393
 394         title = unescapeHTML(self._search_regex(
 395             r'<title>\s*([^<]+?)\s+\|\s+\d+\s+videos',
 396             webpage, 'title', default=page_id))
 397
 398         return self.playlist_result(entries, page_id, title)