_ Git - youtube-dl/blob - youtube_dl/extractor/dumpert.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import base64
   5
   6 from .common import InfoExtractor
   7 from ..compat import compat_urllib_request
   8 from ..utils import qualities
   9
  10
  11 class DumpertIE(InfoExtractor):
  12     _VALID_URL = r'https?://(?:www\.)?dumpert\.nl/mediabase/(?P<id>[0-9]+/[0-9a-zA-Z]+)'
  13     _TEST = {
  14         'url': 'http://www.dumpert.nl/mediabase/6646981/951bc60f/',
  15         'md5': '1b9318d7d5054e7dcb9dc7654f21d643',
  16         'info_dict': {
  17             'id': '6646981/951bc60f',
  18             'ext': 'mp4',
  19             'title': 'Ik heb nieuws voor je',
  20             'description': 'Niet schrikken hoor',
  21             'thumbnail': 're:^https?://.*\.jpg$',
  22         }
  23     }
  24
  25     def _real_extract(self, url):
  26         video_id = self._match_id(url)
  27
  28         req = compat_urllib_request.Request(url)
  29         req.add_header('Cookie', 'nsfw=1; cpc=10')
  30         webpage = self._download_webpage(req, video_id)
  31
  32         files_base64 = self._search_regex(
  33             r'data-files="([^"]+)"', webpage, 'data files')
  34
  35         files = self._parse_json(
  36             base64.b64decode(files_base64.encode('utf-8')).decode('utf-8'),
  37             video_id)
  38
  39         quality = qualities(['flv', 'mobile', 'tablet', '720p'])
  40
  41         formats = [{
  42             'url': video_url,
  43             'format_id': format_id,
  44             'quality': quality(format_id),
  45         } for format_id, video_url in files.items() if format_id != 'still']
  46         self._sort_formats(formats)
  47
  48         title = self._html_search_meta(
  49             'title', webpage) or self._og_search_title(webpage)
  50         description = self._html_search_meta(
  51             'description', webpage) or self._og_search_description(webpage)
  52         thumbnail = files.get('still') or self._og_search_thumbnail(webpage)
  53
  54         return {
  55             'id': video_id,
  56             'title': title,
  57             'description': description,
  58             'thumbnail': thumbnail,
  59             'formats': formats
  60         }