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
   7 from .common import InfoExtractor
   8 from ..compat import (
   9     compat_str,
  10     compat_urllib_parse,
  11 )
  12 from ..utils import (
  13     ExtractorError,
  14     orderedSet,
  15     sanitized_Request,
  16     str_to_int,
  17     unescapeHTML,
  18     unified_strdate,
  19 )
  20 from .vimeo import VimeoIE
  21 from .pladform import PladformIE
  22
  23
  24 class VKIE(InfoExtractor):
  25     IE_NAME = 'vk'
  26     IE_DESC = 'VK'
  27     _VALID_URL = r'''(?x)
  28                     https?://
  29                         (?:
  30                             (?:m\.)?vk\.com/video_ext\.php\?.*?\boid=(?P<oid>-?\d+).*?\bid=(?P<id>\d+)|
  31                             (?:
  32                                 (?:m\.)?vk\.com/(?:.+?\?.*?z=)?video|
  33                                 (?:www\.)?biqle\.ru/watch/
  34                             )
  35                             (?P<videoid>[^s].*?)(?:\?(?:.*\blist=(?P<list_id>[\da-f]+))?|%2F|$)
  36                         )
  37                     '''
  38     _NETRC_MACHINE = 'vk'
  39
  40     _TESTS = [
  41         {
  42             'url': 'http://vk.com/videos-77521?z=video-77521_162222515%2Fclub77521',
  43             'md5': '0deae91935c54e00003c2a00646315f0',
  44             'info_dict': {
  45                 'id': '162222515',
  46                 'ext': 'flv',
  47                 'title': 'ProtivoGunz - Хуёвая песня',
  48                 'uploader': 're:(?:Noize MC|Alexander Ilyashenko).*',
  49                 'duration': 195,
  50                 'upload_date': '20120212',
  51                 'view_count': int,
  52             },
  53         },
  54         {
  55             'url': 'http://vk.com/video205387401_165548505',
  56             'md5': '6c0aeb2e90396ba97035b9cbde548700',
  57             'info_dict': {
  58                 'id': '165548505',
  59                 'ext': 'mp4',
  60                 'uploader': 'Tom Cruise',
  61                 'title': 'No name',
  62                 'duration': 9,
  63                 'upload_date': '20130721',
  64                 'view_count': int,
  65             }
  66         },
  67         {
  68             'note': 'Embedded video',
  69             'url': 'http://vk.com/video_ext.php?oid=32194266&id=162925554&hash=7d8c2e0d5e05aeaa&hd=1',
  70             'md5': 'c7ce8f1f87bec05b3de07fdeafe21a0a',
  71             'info_dict': {
  72                 'id': '162925554',
  73                 'ext': 'mp4',
  74                 'uploader': 'Vladimir Gavrin',
  75                 'title': 'Lin Dan',
  76                 'duration': 101,
  77                 'upload_date': '20120730',
  78                 'view_count': int,
  79             }
  80         },
  81         {
  82             # VIDEO NOW REMOVED
  83             # please update if you find a video whose URL follows the same pattern
  84             'url': 'http://vk.com/video-8871596_164049491',
  85             'md5': 'a590bcaf3d543576c9bd162812387666',
  86             'note': 'Only available for registered users',
  87             'info_dict': {
  88                 'id': '164049491',
  89                 'ext': 'mp4',
  90                 'uploader': 'Триллеры',
  91                 'title': '► Бойцовский клуб / Fight Club 1999 [HD 720]',
  92                 'duration': 8352,
  93                 'upload_date': '20121218',
  94                 'view_count': int,
  95             },
  96             'skip': 'Requires vk account credentials',
  97         },
  98         {
  99             'url': 'http://vk.com/hd_kino_mania?z=video-43215063_168067957%2F15c66b9b533119788d',
 100             'md5': '4d7a5ef8cf114dfa09577e57b2993202',
 101             'info_dict': {
 102                 'id': '168067957',
 103                 'ext': 'mp4',
 104                 'uploader': 'Киномания - лучшее из мира кино',
 105                 'title': ' ',
 106                 'duration': 7291,
 107                 'upload_date': '20140328',
 108             },
 109             'skip': 'Requires vk account credentials',
 110         },
 111         {
 112             'url': 'http://m.vk.com/video-43215063_169084319?list=125c627d1aa1cebb83&from=wall-43215063_2566540',
 113             'md5': '0c45586baa71b7cb1d0784ee3f4e00a6',
 114             'note': 'ivi.ru embed',
 115             'info_dict': {
 116                 'id': '60690',
 117                 'ext': 'mp4',
 118                 'title': 'Книга Илая',
 119                 'duration': 6771,
 120                 'upload_date': '20140626',
 121                 'view_count': int,
 122             },
 123             'skip': 'Only works from Russia',
 124         },
 125         {
 126             # video (removed?) only available with list id
 127             'url': 'https://vk.com/video30481095_171201961?list=8764ae2d21f14088d4',
 128             'md5': '091287af5402239a1051c37ec7b92913',
 129             'info_dict': {
 130                 'id': '171201961',
 131                 'ext': 'mp4',
 132                 'title': 'ТюменцевВВ_09.07.2015',
 133                 'uploader': 'Anton Ivanov',
 134                 'duration': 109,
 135                 'upload_date': '20150709',
 136                 'view_count': int,
 137             },
 138         },
 139         {
 140             # youtube embed
 141             'url': 'https://vk.com/video276849682_170681728',
 142             'info_dict': {
 143                 'id': 'V3K4mi0SYkc',
 144                 'ext': 'mp4',
 145                 'title': "DSWD Awards 'Children's Joy Foundation, Inc.' Certificate of Registration and License to Operate",
 146                 'description': 'md5:bf9c26cfa4acdfb146362682edd3827a',
 147                 'duration': 179,
 148                 'upload_date': '20130116',
 149                 'uploader': "Children's Joy Foundation",
 150                 'uploader_id': 'thecjf',
 151                 'view_count': int,
 152             },
 153         },
 154         {
 155             # removed video, just testing that we match the pattern
 156             'url': 'http://vk.com/feed?z=video-43215063_166094326%2Fbb50cacd3177146d7a',
 157             'only_matching': True,
 158         },
 159         {
 160             # age restricted video, requires vk account credentials
 161             'url': 'https://vk.com/video205387401_164765225',
 162             'only_matching': True,
 163         },
 164         {
 165             # vk wrapper
 166             'url': 'http://www.biqle.ru/watch/847655_160197695',
 167             'only_matching': True,
 168         },
 169         {
 170             # pladform embed
 171             'url': 'https://vk.com/video-76116461_171554880',
 172             'only_matching': True,
 173         }
 174     ]
 175
 176     def _login(self):
 177         (username, password) = self._get_login_info()
 178         if username is None:
 179             return
 180
 181         login_page = self._download_webpage(
 182             'https://vk.com', None, 'Downloading login page')
 183
 184         login_form = self._hidden_inputs(login_page)
 185
 186         login_form.update({
 187             'email': username.encode('cp1251'),
 188             'pass': password.encode('cp1251'),
 189         })
 190
 191         request = sanitized_Request(
 192             'https://login.vk.com/?act=login',
 193             compat_urllib_parse.urlencode(login_form).encode('utf-8'))
 194         login_page = self._download_webpage(
 195             request, None, note='Logging in as %s' % username)
 196
 197         if re.search(r'onLoginFailed', login_page):
 198             raise ExtractorError(
 199                 'Unable to login, incorrect username and/or password', expected=True)
 200
 201     def _real_initialize(self):
 202         self._login()
 203
 204     def _real_extract(self, url):
 205         mobj = re.match(self._VALID_URL, url)
 206         video_id = mobj.group('videoid')
 207
 208         if not video_id:
 209             video_id = '%s_%s' % (mobj.group('oid'), mobj.group('id'))
 210
 211         info_url = 'https://vk.com/al_video.php?act=show&al=1&module=video&video=%s' % video_id
 212
 213         # Some videos (removed?) can only be downloaded with list id specified
 214         list_id = mobj.group('list_id')
 215         if list_id:
 216             info_url += '&list=%s' % list_id
 217
 218         info_page = self._download_webpage(info_url, video_id)
 219
 220         error_message = self._html_search_regex(
 221             r'(?s)<!><div[^>]+class="video_layer_message"[^>]*>(.+?)</div>',
 222             info_page, 'error message', default=None)
 223         if error_message:
 224             raise ExtractorError(error_message, expected=True)
 225
 226         if re.search(r'<!>/login\.php\?.*\bact=security_check', info_page):
 227             raise ExtractorError(
 228                 'You are trying to log in from an unusual location. You should confirm ownership at vk.com to log in with this IP.',
 229                 expected=True)
 230
 231         ERRORS = {
 232             r'>Видеозапись .*? была изъята из публичного доступа в связи с обращением правообладателя.<':
 233             'Video %s has been removed from public access due to rightholder complaint.',
 234
 235             r'<!>Please log in or <':
 236             'Video %s is only available for registered users, '
 237             'use --username and --password options to provide account credentials.',
 238
 239             r'<!>Unknown error':
 240             'Video %s does not exist.',
 241
 242             r'<!>Видео временно недоступно':
 243             'Video %s is temporarily unavailable.',
 244
 245             r'<!>Access denied':
 246             'Access denied to video %s.',
 247         }
 248
 249         for error_re, error_msg in ERRORS.items():
 250             if re.search(error_re, info_page):
 251                 raise ExtractorError(error_msg % video_id, expected=True)
 252
 253         youtube_url = self._search_regex(
 254             r'<iframe[^>]+src="((?:https?:)?//www.youtube.com/embed/[^"]+)"',
 255             info_page, 'youtube iframe', default=None)
 256         if youtube_url:
 257             return self.url_result(youtube_url, 'Youtube')
 258
 259         vimeo_url = VimeoIE._extract_vimeo_url(url, info_page)
 260         if vimeo_url is not None:
 261             return self.url_result(vimeo_url)
 262
 263         pladform_url = PladformIE._extract_url(info_page)
 264         if pladform_url:
 265             return self.url_result(pladform_url)
 266
 267         m_rutube = re.search(
 268             r'\ssrc="((?:https?:)?//rutube\.ru\\?/(?:video|play)\\?/embed(?:.*?))\\?"', info_page)
 269         if m_rutube is not None:
 270             rutube_url = self._proto_relative_url(
 271                 m_rutube.group(1).replace('\\', ''))
 272             return self.url_result(rutube_url)
 273
 274         m_opts = re.search(r'(?s)var\s+opts\s*=\s*({.+?});', info_page)
 275         if m_opts:
 276             m_opts_url = re.search(r"url\s*:\s*'((?!/\b)[^']+)", m_opts.group(1))
 277             if m_opts_url:
 278                 opts_url = m_opts_url.group(1)
 279                 if opts_url.startswith('//'):
 280                     opts_url = 'http:' + opts_url
 281                 return self.url_result(opts_url)
 282
 283         data_json = self._search_regex(r'var\s+vars\s*=\s*({.+?});', info_page, 'vars')
 284         data = json.loads(data_json)
 285
 286         # Extract upload date
 287         upload_date = None
 288         mobj = re.search(r'id="mv_date(?:_views)?_wrap"[^>]*>([a-zA-Z]+ [0-9]+), ([0-9]+) at', info_page)
 289         if mobj is not None:
 290             mobj.group(1) + ' ' + mobj.group(2)
 291             upload_date = unified_strdate(mobj.group(1) + ' ' + mobj.group(2))
 292
 293         view_count = None
 294         views = self._html_search_regex(
 295             r'"mv_views_count_number"[^>]*>(.+?\bviews?)<',
 296             info_page, 'view count', fatal=False)
 297         if views:
 298             view_count = str_to_int(self._search_regex(
 299                 r'([\d,.]+)', views, 'view count', fatal=False))
 300
 301         formats = [{
 302             'format_id': k,
 303             'url': v,
 304             'width': int(k[len('url'):]),
 305         } for k, v in data.items()
 306             if k.startswith('url')]
 307         self._sort_formats(formats)
 308
 309         return {
 310             'id': compat_str(data['vid']),
 311             'formats': formats,
 312             'title': unescapeHTML(data['md_title']),
 313             'thumbnail': data.get('jpg'),
 314             'uploader': data.get('md_author'),
 315             'duration': data.get('duration'),
 316             'upload_date': upload_date,
 317             'view_count': view_count,
 318         }
 319
 320
 321 class VKUserVideosIE(InfoExtractor):
 322     IE_NAME = 'vk:uservideos'
 323     IE_DESC = "VK - User's Videos"
 324     _VALID_URL = r'https?://vk\.com/videos(?P<id>-?[0-9]+)(?!\?.*\bz=video)(?:[/?#&]|$)'
 325     _TEMPLATE_URL = 'https://vk.com/videos'
 326     _TESTS = [{
 327         'url': 'http://vk.com/videos205387401',
 328         'info_dict': {
 329             'id': '205387401',
 330             'title': "Tom Cruise's Videos",
 331         },
 332         'playlist_mincount': 4,
 333     }, {
 334         'url': 'http://vk.com/videos-77521',
 335         'only_matching': True,
 336     }, {
 337         'url': 'http://vk.com/videos-97664626?section=all',
 338         'only_matching': True,
 339     }]
 340
 341     def _real_extract(self, url):
 342         page_id = self._match_id(url)
 343
 344         webpage = self._download_webpage(url, page_id)
 345
 346         entries = [
 347             self.url_result(
 348                 'http://vk.com/video' + video_id, 'VK', video_id=video_id)
 349             for video_id in orderedSet(re.findall(r'href="/video(-?[0-9_]+)"', webpage))]
 350
 351         title = unescapeHTML(self._search_regex(
 352             r'<title>\s*([^<]+?)\s+\|\s+\d+\s+videos',
 353             webpage, 'title', default=page_id))
 354
 355         return self.playlist_result(entries, page_id, title)