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|embed)/(?P<id>[0-9]+/[0-9a-zA-Z]+)'
  13     _TESTS = [{
  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         'url': 'http://www.dumpert.nl/embed/6675421/dc440fe7/',
  25         'only_matching': True,
  26     }]
  27
  28     def _real_extract(self, url):
  29         video_id = self._match_id(url)
  30
  31         url = 'https://www.dumpert.nl/mediabase/' + video_id
  32         req = compat_urllib_request.Request(url)
  33         req.add_header('Cookie', 'nsfw=1; cpc=10')
  34         webpage = self._download_webpage(req, video_id)
  35
  36         files_base64 = self._search_regex(
  37             r'data-files="([^"]+)"', webpage, 'data files')
  38
  39         files = self._parse_json(
  40             base64.b64decode(files_base64.encode('utf-8')).decode('utf-8'),
  41             video_id)
  42
  43         quality = qualities(['flv', 'mobile', 'tablet', '720p'])
  44
  45         formats = [{
  46             'url': video_url,
  47             'format_id': format_id,
  48             'quality': quality(format_id),
  49         } for format_id, video_url in files.items() if format_id != 'still']
  50         self._sort_formats(formats)
  51
  52         title = self._html_search_meta(
  53             'title', webpage) or self._og_search_title(webpage)
  54         description = self._html_search_meta(
  55             'description', webpage) or self._og_search_description(webpage)
  56         thumbnail = files.get('still') or self._og_search_thumbnail(webpage)
  57
  58         return {
  59             'id': video_id,
  60             'title': title,
  61             'description': description,
  62             'thumbnail': thumbnail,
  63             'formats': formats
  64         }