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