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

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6
   7
   8 class KeekIE(InfoExtractor):
   9     _VALID_URL = r'https?://(?:www\.)?keek\.com/(?:!|\w+/keeks/)(?P<id>\w+)'
  10     IE_NAME = 'keek'
  11     _TEST = {
  12         'url': 'https://www.keek.com/ytdl/keeks/NODfbab',
  13         'md5': '09c5c109067536c1cec8bac8c21fea05',
  14         'info_dict': {
  15             'id': 'NODfbab',
  16             'ext': 'mp4',
  17             'uploader': 'youtube-dl project',
  18             'uploader_id': 'ytdl',
  19             'title': 'test chars: "\'/\\\u00e4<>This is a test video for youtube-dl.For more information, contact phihag@phihag.de .',
  20         },
  21     }
  22
  23     def _real_extract(self, url):
  24         video_id = self._match_id(url)
  25
  26         video_url = 'http://cdn.keek.com/keek/video/%s' % video_id
  27         thumbnail = 'http://cdn.keek.com/keek/thumbnail/%s/w100/h75' % video_id
  28         webpage = self._download_webpage(url, video_id)
  29
  30         raw_desc = self._html_search_meta('description', webpage)
  31         if raw_desc:
  32             uploader = self._html_search_regex(
  33                 r'Watch (.*?)\s+\(', raw_desc, 'uploader', fatal=False)
  34             uploader_id = self._html_search_regex(
  35                 r'Watch .*?\(@(.+?)\)', raw_desc, 'uploader_id', fatal=False)
  36         else:
  37             uploader = None
  38             uploader_id = None
  39
  40         return {
  41             'id': video_id,
  42             'url': video_url,
  43             'ext': 'mp4',
  44             'title': self._og_search_title(webpage),
  45             'thumbnail': thumbnail,
  46             'uploader': uploader,
  47             'uploader_id': uploader_id,
  48         }