_ Git - youtube-dl/blob - 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         },
  28     }, {
  29         'url': 'http://vivo.sx/d7ddda0e78',
  30         'md5': '15b3af41be0b4fe01f4df075c2678b2c',
  31         'info_dict': {
  32             'id': 'd7ddda0e78',
  33             'ext': 'mp4',
  34             'title': 'Chicken',
  35             'filesize': 528031,
  36         },
  37     }]
  38
  39     def _real_extract(self, url):
  40         video_id = self._match_id(url)
  41         webpage = self._download_webpage(url, video_id)
  42
  43         if '>File does not exist<' in webpage:
  44             raise ExtractorError(
  45                 'Video %s does not exist' % video_id, expected=True)
  46
  47         download_form = self._hidden_inputs(webpage)
  48         request = compat_urllib_request.Request(
  49             url, compat_urllib_parse.urlencode(download_form))
  50         request.add_header('Content-Type', 'application/x-www-form-urlencoded')
  51
  52         video_page = self._download_webpage(
  53             request, video_id, 'Downloading video page')
  54
  55         video_url = self._html_search_regex(
  56             r'data-url="([^"]+)"', video_page, 'video URL')
  57         title = base64.b64decode(self._html_search_meta(
  58             'full:title', webpage, 'title').encode('utf-8')).decode('utf-8')
  59         filesize = int_or_none(self._html_search_meta(
  60             'full:size', webpage, 'file size', fatal=False))
  61         thumbnail = self._html_search_regex(
  62             r'data-poster="([^"]+)"', video_page, 'thumbnail', default=None)
  63
  64         return {
  65             'id': video_id,
  66             'url': video_url,
  67             'ext': 'mp4',
  68             'filesize': filesize,
  69             'title': title,
  70             'thumbnail': thumbnail,
  71         }