2 from __future__ import unicode_literals
5 This script employs a VERY basic heuristic ('porn' in webpage.lower()) to check
6 if we are not 'age_limit' tagging some porn site
8 A second approach implemented relies on a list of porn domains, to activate it
9 pass the list filename as the only argument
12 # Allow direct execution
15 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
17 from test.helper import get_testcases
18 from youtube_dl.utils import compat_urllib_parse_urlparse
19 from youtube_dl.utils import compat_urllib_request
23 LIST = open(sys.argv[1]).read().decode('utf8').strip()
27 for test in get_testcases():
28 if METHOD == 'EURISTIC':
30 webpage = compat_urllib_request.urlopen(test['url'], timeout=10).read()
32 print('\nFail: {0}'.format(test['name']))
35 webpage = webpage.decode('utf8', 'replace')
37 RESULT = 'porn' in webpage.lower()
39 elif METHOD == 'LIST':
40 domain = compat_urllib_parse_urlparse(test['url']).netloc
42 print('\nFail: {0}'.format(test['name']))
44 domain = '.'.join(domain.split('.')[-2:])
46 RESULT = ('.' + domain + '\n' in LIST or '\n' + domain + '\n' in LIST)
48 if RESULT and ('info_dict' not in test or 'age_limit' not in test['info_dict'] or
49 test['info_dict']['age_limit'] != 18):
50 print('\nPotential missing age_limit check: {0}'.format(test['name']))
52 elif not RESULT and ('info_dict' in test and 'age_limit' in test['info_dict'] and
53 test['info_dict']['age_limit'] == 18):
54 print('\nPotential false negative: {0}'.format(test['name']))