]> jfr.im git - yt-dlp.git/blob - devscripts/check-porn.py
[ie/youtube] Fix comments extraction (#9775)
[yt-dlp.git] / devscripts / check-porn.py
1 #!/usr/bin/env python3
2 """
3 This script employs a VERY basic heuristic ('porn' in webpage.lower()) to check
4 if we are not 'age_limit' tagging some porn site
5
6 A second approach implemented relies on a list of porn domains, to activate it
7 pass the list filename as the only argument
8 """
9
10 # Allow direct execution
11 import os
12 import sys
13
14 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
15
16
17 import urllib.parse
18 import urllib.request
19
20 from test.helper import gettestcases
21
22 if len(sys.argv) > 1:
23 METHOD = 'LIST'
24 LIST = open(sys.argv[1]).read().decode('utf8').strip()
25 else:
26 METHOD = 'EURISTIC'
27
28 for test in gettestcases():
29 if METHOD == 'EURISTIC':
30 try:
31 webpage = urllib.request.urlopen(test['url'], timeout=10).read()
32 except Exception:
33 print('\nFail: {}'.format(test['name']))
34 continue
35
36 webpage = webpage.decode('utf8', 'replace')
37
38 RESULT = 'porn' in webpage.lower()
39
40 elif METHOD == 'LIST':
41 domain = urllib.parse.urlparse(test['url']).netloc
42 if not domain:
43 print('\nFail: {}'.format(test['name']))
44 continue
45 domain = '.'.join(domain.split('.')[-2:])
46
47 RESULT = ('.' + domain + '\n' in LIST or '\n' + domain + '\n' in LIST)
48
49 if RESULT and ('info_dict' not in test or 'age_limit' not in test['info_dict']
50 or test['info_dict']['age_limit'] != 18):
51 print('\nPotential missing age_limit check: {}'.format(test['name']))
52
53 elif not RESULT and ('info_dict' in test and 'age_limit' in test['info_dict']
54 and test['info_dict']['age_limit'] == 18):
55 print('\nPotential false negative: {}'.format(test['name']))
56
57 else:
58 sys.stdout.write('.')
59 sys.stdout.flush()
60
61 print()