youtube_dl/extractor/shared.py

   1 from __future__ import unicode_literals
   2
   3 import base64
   4
   5 from .common import InfoExtractor
   6 from ..compat import (
   7     compat_urllib_parse,
   8     compat_urllib_request,
   9 )
  10 from ..utils import (
  11     ExtractorError,
  12     int_or_none,
  13 )
  14
  15
  16 class SharedIE(InfoExtractor):
  17     IE_DESC = 'shared.sx and vivo.sx'
  18     _VALID_URL = r'http://(?:shared|vivo)\.sx/(?P<id>[\da-z]{10})'
  19
  20     _TESTS = [{
  21         'url': 'http://shared.sx/0060718775',
  22         'md5': '106fefed92a8a2adb8c98e6a0652f49b',
  23         'info_dict': {
  24             'id': '0060718775',
  25             'ext': 'mp4',
  26             'title': 'Bmp4',
  27             'filesize': 1720110,
  28         },
  29     }, {
  30         'url': 'http://vivo.sx/d7ddda0e78',
  31         'md5': '15b3af41be0b4fe01f4df075c2678b2c',
  32         'info_dict': {
  33             'id': 'd7ddda0e78',
  34             'ext': 'mp4',
  35             'title': 'Chicken',
  36             'filesize': 528031,
  37         },
  38     }]
  39
  40     def _real_extract(self, url):
  41         video_id = self._match_id(url)
  42         webpage = self._download_webpage(url, video_id)
  43
  44         if '>File does not exist<' in webpage:
  45             raise ExtractorError(
  46                 'Video %s does not exist' % video_id, expected=True)
  47
  48         download_form = self._hidden_inputs(webpage)
  49         request = compat_urllib_request.Request(
  50             url, compat_urllib_parse.urlencode(download_form))
  51         request.add_header('Content-Type', 'application/x-www-form-urlencoded')
  52
  53         video_page = self._download_webpage(
  54             request, video_id, 'Downloading video page')
  55
  56         video_url = self._html_search_regex(
  57             r'data-url="([^"]+)"', video_page, 'video URL')
  58         title = base64.b64decode(self._html_search_meta(
  59             'full:title', webpage, 'title').encode('utf-8')).decode('utf-8')
  60         filesize = int_or_none(self._html_search_meta(
  61             'full:size', webpage, 'file size', fatal=False))
  62         thumbnail = self._html_search_regex(
  63             r'data-poster="([^"]+)"', video_page, 'thumbnail', default=None)
  64
  65         return {
  66             'id': video_id,
  67             'url': video_url,
  68             'ext': 'mp4',
  69             'filesize': filesize,
  70             'title': title,
  71             'thumbnail': thumbnail,
  72         }