_ 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 ..utils import (
   7     ExtractorError,
   8     int_or_none,
   9     sanitized_Request,
  10     urlencode_postdata,
  11 )
  12
  13
  14 class SharedIE(InfoExtractor):
  15     IE_DESC = 'shared.sx and vivo.sx'
  16     _VALID_URL = r'https?://(?:shared|vivo)\.sx/(?P<id>[\da-z]{10})'
  17
  18     _TESTS = [{
  19         'url': 'http://shared.sx/0060718775',
  20         'md5': '106fefed92a8a2adb8c98e6a0652f49b',
  21         'info_dict': {
  22             'id': '0060718775',
  23             'ext': 'mp4',
  24             'title': 'Bmp4',
  25             'filesize': 1720110,
  26         },
  27     }, {
  28         'url': 'http://vivo.sx/d7ddda0e78',
  29         'md5': '15b3af41be0b4fe01f4df075c2678b2c',
  30         'info_dict': {
  31             'id': 'd7ddda0e78',
  32             'ext': 'mp4',
  33             'title': 'Chicken',
  34             'filesize': 528031,
  35         },
  36     }]
  37
  38     def _real_extract(self, url):
  39         video_id = self._match_id(url)
  40
  41         webpage, urlh = self._download_webpage_handle(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
  49         request = sanitized_Request(
  50             urlh.geturl(), urlencode_postdata(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         }