Raphaël G. Git Repositories - youtubedl/blob - devscripts/check-porn.py

   1 #!/usr/bin/env python
   2
   3 """
   4 This script employs a VERY basic heuristic ('porn' in webpage.lower()) to check
   5 if we are not 'age_limit' tagging some porn site
   6 """
   7
   8 # Allow direct execution
   9 import os
  10 import sys
  11 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
  12
  13 from test.helper import get_testcases
  14 from youtube_dl.utils import compat_urllib_request
  15
  16 for test in get_testcases():
  17     try:
  18         webpage = compat_urllib_request.urlopen(test['url'], timeout=10).read()
  19     except:
  20         print('\nFail: {0}'.format(test['name']))
  21         continue
  22
  23     webpage = webpage.decode('utf8', 'replace')
  24
  25     if 'porn' in webpage.lower() and ('info_dict' not in test
  26                                       or 'age_limit' not in test['info_dict']
  27                                       or test['info_dict']['age_limit'] != 18):
  28         print('\nPotential missing age_limit check: {0}'.format(test['name']))
  29
  30     elif 'porn' not in webpage.lower() and ('info_dict' in test and
  31                                             'age_limit' in test['info_dict'] and
  32                                             test['info_dict']['age_limit'] == 18):
  33         print('\nPotential false negative: {0}'.format(test['name']))
  34
  35     else:
  36         sys.stdout.write('.')
  37     sys.stdout.flush()
  38
  39 print()