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 )
  15
  16
  17 class VKIE(InfoExtractor):
  18     IE_NAME = 'vk.com'
  19     _VALID_URL = r'https?://vk\.com/(?:videos.*?\?.*?z=)?video(?P<id>.*?)(?:\?|%2F|$)'
  20     _NETRC_MACHINE = 'vk'
  21
  22     _TESTS = [
  23         {
  24             'url': 'http://vk.com/videos-77521?z=video-77521_162222515%2Fclub77521',
  25             'md5': '0deae91935c54e00003c2a00646315f0',
  26             'info_dict': {
  27                 'id': '162222515',
  28                 'ext': 'flv',
  29                 'title': 'ProtivoGunz - Хуёвая песня',
  30                 'uploader': 'Noize MC',
  31                 'duration': 195,
  32             },
  33         },
  34         {
  35             'url': 'http://vk.com/video4643923_163339118',
  36             'md5': 'f79bccb5cd182b1f43502ca5685b2b36',
  37             'info_dict': {
  38                 'id': '163339118',
  39                 'ext': 'mp4',
  40                 'uploader': 'Elvira Dzhonik',
  41                 'title': 'Dream Theater - Hollow Years Live at Budokan 720*',
  42                 'duration': 558,
  43             }
  44         },
  45         {
  46             'url': 'http://vk.com/video-8871596_164049491',
  47             'md5': 'a590bcaf3d543576c9bd162812387666',
  48             'note': 'Only available for registered users',
  49             'info_dict': {
  50                 'id': '164049491',
  51                 'ext': 'mp4',
  52                 'uploader': 'Триллеры',
  53                 'title': '► Бойцовский клуб / Fight Club 1999 [HD 720]\u00a0',
  54                 'duration': 8352,
  55             },
  56             'skip': 'Requires vk account credentials',
  57         }
  58     ]
  59
  60     def _login(self):
  61         (username, password) = self._get_login_info()
  62         if username is None:
  63             return
  64
  65         login_form = {
  66             'act': 'login',
  67             'role': 'al_frame',
  68             'expire': '1',
  69             'email': username,
  70             'pass': password,
  71         }
  72
  73         request = compat_urllib_request.Request('https://login.vk.com/?act=login',
  74             compat_urllib_parse.urlencode(login_form).encode('utf-8'))
  75         login_page = self._download_webpage(request, None, note='Logging in as %s' % username)
  76
  77         if re.search(r'onLoginFailed', login_page):
  78             raise ExtractorError('Unable to login, incorrect username and/or password', expected=True)
  79
  80     def _real_initialize(self):
  81         self._login()
  82
  83     def _real_extract(self, url):
  84         mobj = re.match(self._VALID_URL, url)
  85         video_id = mobj.group('id')
  86
  87         info_url = 'http://vk.com/al_video.php?act=show&al=1&video=%s' % video_id
  88         info_page = self._download_webpage(info_url, video_id)
  89
  90         if re.search(r'<!>Please log in or <', info_page):
  91             raise ExtractorError('This video is only available for registered users, '
  92                 'use --username and --password options to provide account credentials.', expected=True)
  93
  94         m_yt = re.search(r'src="(http://www.youtube.com/.*?)"', info_page)
  95         if m_yt is not None:
  96             self.to_screen(u'Youtube video detected')
  97             return self.url_result(m_yt.group(1), 'Youtube')
  98         data_json = self._search_regex(r'var vars = ({.*?});', info_page, 'vars')
  99         data = json.loads(data_json)
 100
 101         formats = [{
 102             'format_id': k,
 103             'url': v,
 104             'width': int(k[len('url'):]),
 105         } for k, v in data.items()
 106             if k.startswith('url')]
 107         self._sort_formats(formats)
 108
 109         return {
 110             'id': compat_str(data['vid']),
 111             'formats': formats,
 112             'title': unescapeHTML(data['md_title']),
 113             'thumbnail': data.get('jpg'),
 114             'uploader': data.get('md_author'),
 115             'duration': data.get('duration')
 116         }