Mercurial > public > sg101

annotate gpp/antispam/utils.py @ 334:6805d15cda13

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Adding a script I had to write on the fly to filter out posts from the posts csv file that had no parent topics. MyISAM let me get away with that, but InnoDB won't.
author Brian Neal <bgneal@gmail.com>
date Sat, 26 Feb 2011 01:28:22 +0000
parents 767cedc7d12a
children 98b373ca09f3
rev   line source
bgneal@214 1 """Antispam utility functions other apps can use."""
bgneal@215 2 import datetime
bgneal@215 3 import logging
bgneal@215 4 import textwrap
bgneal@215 5
bgneal@214 6 from django.core.cache import cache
bgneal@214 7
bgneal@214 8 from antispam import SPAM_PHRASE_KEY
bgneal@214 9 from antispam.models import SpamPhrase
bgneal@215 10 from core.functions import email_admins
bgneal@215 11 from bio.models import STA_SUSPENDED
bgneal@214 12
bgneal@214 13
bgneal@214 14 def contains_spam(s):
bgneal@214 15 """This function returns True if the supplied string s contains any spam
bgneal@214 16 phrases and False otherwise.
bgneal@214 17 """
bgneal@214 18 phrases = _get_spam_phrases()
bgneal@215 19 s = s.lower()
bgneal@214 20 for spam in phrases:
bgneal@214 21 if spam in s:
bgneal@214 22 return True
bgneal@214 23
bgneal@214 24 return False
bgneal@214 25
bgneal@214 26
bgneal@215 27 def spam_check(request, content):
bgneal@215 28 """This function checks the supplied content for spam if the user from the
bgneal@215 29 supplied request is a stranger (new to the site). If spam is found, the
bgneal@215 30 function makes a log entry, emails the admins, suspends the user's account
bgneal@215 31 and returns True. If spam is not found, False is returned.
bgneal@215 32 It is assumed that request.user is an authenticated user and thus has a
bgneal@215 33 user profile.
bgneal@215 34 """
bgneal@215 35 user = request.user
bgneal@215 36 if user.get_profile().is_stranger() and contains_spam(content):
bgneal@215 37
bgneal@215 38 ip = request.META.get('REMOTE_ADDR', "unknown")
bgneal@215 39
bgneal@215 40 msg = textwrap.dedent("""\
bgneal@215 41 SPAM FILTER TRIPPED by %s
bgneal@215 42 PATH: %s
bgneal@215 43 IP: %s
bgneal@215 44 Message:
bgneal@215 45 %s
bgneal@215 46 """ % (user.username, request.path, ip, content))
bgneal@215 47
bgneal@215 48 logging.info(msg)
bgneal@215 49 email_admins("SPAM FILTER TRIPPED BY %s" % user.username, msg)
bgneal@215 50 suspend_user(user)
bgneal@215 51 return True
bgneal@215 52
bgneal@215 53 return False
bgneal@215 54
bgneal@215 55
bgneal@215 56 def suspend_user(user):
bgneal@215 57 """This function marks the user as suspended."""
bgneal@215 58 user.is_active = False
bgneal@215 59 user.save()
bgneal@215 60 profile = user.get_profile()
bgneal@215 61 profile.status = STA_SUSPENDED
bgneal@215 62 profile.status_date = datetime.datetime.now()
bgneal@215 63 profile.save()
bgneal@316 64
bgneal@215 65
bgneal@214 66 def _get_spam_phrases():
bgneal@214 67 """This function returns the current list of spam phrase strings.
bgneal@214 68 The strings are cached to avoid hitting the database.
bgneal@214 69 """
bgneal@214 70 phrases = cache.get(SPAM_PHRASE_KEY)
bgneal@214 71 if phrases:
bgneal@214 72 return phrases
bgneal@214 73
bgneal@214 74 phrases = SpamPhrase.objects.values_list('phrase', flat=True)
bgneal@214 75 cache.set(SPAM_PHRASE_KEY, phrases)
bgneal@214 76 return phrases