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 ..utils import (
   9     ExtractorError,
  10     compat_urllib_request,
  11     compat_urllib_parse,
  12     compat_str,
  13     unescapeHTML,
  14     unified_strdate,
  15     orderedSet)
  16
  17
  18 class VKIE(InfoExtractor):
  19     IE_NAME = 'vk.com'
  20     _VALID_URL = r'https?://(?:m\.)?vk\.com/(?:video_ext\.php\?.*?\boid=(?P<oid>-?\d+).*?\bid=(?P<id>\d+)|(?:.+?\?.*?z=)?video(?P<videoid>[^s].*?)(?:\?|%2F|$))'
  21     _NETRC_MACHINE = 'vk'
  22
  23     _TESTS = [
  24         {
  25             'url': 'http://vk.com/videos-77521?z=video-77521_162222515%2Fclub77521',
  26             'md5': '0deae91935c54e00003c2a00646315f0',
  27             'info_dict': {
  28                 'id': '162222515',
  29                 'ext': 'flv',
  30                 'title': 'ProtivoGunz - Хуёвая песня',
  31                 'uploader': 're:Noize MC.*',
  32                 'duration': 195,
  33                 'upload_date': '20120212',
  34             },
  35         },
  36         {
  37             'url': 'http://vk.com/video205387401_165548505',
  38             'md5': '6c0aeb2e90396ba97035b9cbde548700',
  39             'info_dict': {
  40                 'id': '165548505',
  41                 'ext': 'mp4',
  42                 'uploader': 'Tom Cruise',
  43                 'title': 'No name',
  44                 'duration': 9,
  45                 'upload_date': '20130721'
  46             }
  47         },
  48         {
  49             'note': 'Embedded video',
  50             'url': 'http://vk.com/video_ext.php?oid=32194266&id=162925554&hash=7d8c2e0d5e05aeaa&hd=1',
  51             'md5': 'c7ce8f1f87bec05b3de07fdeafe21a0a',
  52             'info_dict': {
  53                 'id': '162925554',
  54                 'ext': 'mp4',
  55                 'uploader': 'Vladimir Gavrin',
  56                 'title': 'Lin Dan',
  57                 'duration': 101,
  58                 'upload_date': '20120730',
  59             }
  60         },
  61         {
  62             # VIDEO NOW REMOVED
  63             # please update if you find a video whose URL follows the same pattern
  64             'url': 'http://vk.com/video-8871596_164049491',
  65             'md5': 'a590bcaf3d543576c9bd162812387666',
  66             'note': 'Only available for registered users',
  67             'info_dict': {
  68                 'id': '164049491',
  69                 'ext': 'mp4',
  70                 'uploader': 'Триллеры',
  71                 'title': '► Бойцовский клуб / Fight Club 1999 [HD 720]',
  72                 'duration': 8352,
  73                 'upload_date': '20121218'
  74             },
  75             'skip': 'Requires vk account credentials',
  76         },
  77         {
  78             'url': 'http://vk.com/hd_kino_mania?z=video-43215063_168067957%2F15c66b9b533119788d',
  79             'md5': '4d7a5ef8cf114dfa09577e57b2993202',
  80             'info_dict': {
  81                 'id': '168067957',
  82                 'ext': 'mp4',
  83                 'uploader': 'Киномания - лучшее из мира кино',
  84                 'title': ' ',
  85                 'duration': 7291,
  86                 'upload_date': '20140328',
  87             },
  88             'skip': 'Requires vk account credentials',
  89         },
  90         {
  91             'url': 'http://m.vk.com/video-43215063_169084319?list=125c627d1aa1cebb83&from=wall-43215063_2566540',
  92             'md5': '0c45586baa71b7cb1d0784ee3f4e00a6',
  93             'note': 'ivi.ru embed',
  94             'info_dict': {
  95                 'id': '60690',
  96                 'ext': 'mp4',
  97                 'title': 'Книга Илая',
  98                 'duration': 6771,
  99                 'upload_date': '20140626',
 100             },
 101             'skip': 'Only works from Russia',
 102         },
 103         {
 104             # removed video, just testing that we match the pattern
 105             'url': 'http://vk.com/feed?z=video-43215063_166094326%2Fbb50cacd3177146d7a',
 106             'only_matching': True,
 107         },
 108     ]
 109
 110     def _login(self):
 111         (username, password) = self._get_login_info()
 112         if username is None:
 113             return
 114
 115         login_form = {
 116             'act': 'login',
 117             'role': 'al_frame',
 118             'expire': '1',
 119             'email': username,
 120             'pass': password,
 121         }
 122
 123         request = compat_urllib_request.Request('https://login.vk.com/?act=login',
 124                                                 compat_urllib_parse.urlencode(login_form).encode('utf-8'))
 125         login_page = self._download_webpage(request, None, note='Logging in as %s' % username)
 126
 127         if re.search(r'onLoginFailed', login_page):
 128             raise ExtractorError('Unable to login, incorrect username and/or password', expected=True)
 129
 130     def _real_initialize(self):
 131         self._login()
 132
 133     def _real_extract(self, url):
 134         mobj = re.match(self._VALID_URL, url)
 135         video_id = mobj.group('videoid')
 136
 137         if not video_id:
 138             video_id = '%s_%s' % (mobj.group('oid'), mobj.group('id'))
 139
 140         info_url = 'http://vk.com/al_video.php?act=show&al=1&video=%s' % video_id
 141         info_page = self._download_webpage(info_url, video_id)
 142
 143         ERRORS = {
 144             r'>Видеозапись .*? была изъята из публичного доступа в связи с обращением правообладателя.<':
 145             'Video %s has been removed from public access due to rightholder complaint.',
 146
 147             r'<!>Please log in or <':
 148             'Video %s is only available for registered users, '
 149             'use --username and --password options to provide account credentials.',
 150
 151             r'<!>Unknown error':
 152             'Video %s does not exist.'
 153         }
 154
 155         for error_re, error_msg in ERRORS.items():
 156             if re.search(error_re, info_page):
 157                 raise ExtractorError(error_msg % video_id, expected=True)
 158
 159         m_yt = re.search(r'src="(http://www.youtube.com/.*?)"', info_page)
 160         if m_yt is not None:
 161             self.to_screen('Youtube video detected')
 162             return self.url_result(m_yt.group(1), 'Youtube')
 163
 164         m_opts = re.search(r'(?s)var\s+opts\s*=\s*({.*?});', info_page)
 165         if m_opts:
 166             m_opts_url = re.search(r"url\s*:\s*'([^']+)", m_opts.group(1))
 167             if m_opts_url:
 168                 opts_url = m_opts_url.group(1)
 169                 if opts_url.startswith('//'):
 170                     opts_url = 'http:' + opts_url
 171                 return self.url_result(opts_url)
 172
 173         data_json = self._search_regex(r'var vars = ({.*?});', info_page, 'vars')
 174         data = json.loads(data_json)
 175
 176         # Extract upload date
 177         upload_date = None
 178         mobj = re.search(r'id="mv_date_wrap".*?Added ([a-zA-Z]+ [0-9]+), ([0-9]+) at', info_page)
 179         if mobj is not None:
 180             mobj.group(1) + ' ' + mobj.group(2)
 181             upload_date = unified_strdate(mobj.group(1) + ' ' + mobj.group(2))
 182
 183         formats = [{
 184             'format_id': k,
 185             'url': v,
 186             'width': int(k[len('url'):]),
 187         } for k, v in data.items()
 188             if k.startswith('url')]
 189         self._sort_formats(formats)
 190
 191         return {
 192             'id': compat_str(data['vid']),
 193             'formats': formats,
 194             'title': unescapeHTML(data['md_title']),
 195             'thumbnail': data.get('jpg'),
 196             'uploader': data.get('md_author'),
 197             'duration': data.get('duration'),
 198             'upload_date': upload_date,
 199         }
 200
 201
 202 class VKUserVideosIE(InfoExtractor):
 203     IE_NAME = 'vk.com:user-videos'
 204     IE_DESC = 'vk.com:All of a user\'s videos'
 205     _VALID_URL = r'https?://vk\.com/videos(?P<id>[0-9]+)(?:m\?.*)?'
 206     _TEMPLATE_URL = 'https://vk.com/videos'
 207     _TEST = {
 208         'url': 'http://vk.com/videos205387401',
 209         'playlist_mincount': 4,
 210     }
 211
 212     def _real_extract(self, url):
 213         page_id = self._match_id(url)
 214         page = self._download_webpage(url, page_id)
 215         video_ids = orderedSet(
 216             m.group(1) for m in re.finditer(r'href="/video([0-9_]+)"', page))
 217         url_entries = [
 218             self.url_result(
 219                 'http://vk.com/video' + video_id, 'VK', video_id=video_id)
 220             for video_id in video_ids]
 221         return self.playlist_result(url_entries, page_id)