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