From: minima <minima>
Date: Mon, 1 Oct 2001 14:30:18 +0000 (+0000)
Subject: Try each badword with an S on the end as well and also check for 'f.u   c'
X-Git-Tag: R_1_49~96
X-Git-Url: http://dxcluster.net/gitweb/gitweb.cgi?a=commitdiff_plain;h=7346d1f8129457eee2e602b36e27435141533049;p=spider.git

Try each badword with an S on the end as well and also check for 'f.u   c'
type things as well (this only works for a few wellknown english ones).
---

diff --git a/Changes b/Changes
index ed8b541b..470a8ece 100644
--- a/Changes
+++ b/Changes
@@ -3,6 +3,8 @@
 2. change the looping protoection for PC16/17/19/21 and also be more rigorous
 in who we accept which PC16/17/21's from.
 3. fix MAJOR disconnect bug introduced accidentally in the NP work.
+4. Try each badword with an S on the end as well and also check for 'f.u   c'
+type things as well (this only works for a few wellknown english ones).
 30Sep01=======================================================================
 1. made some small bug fixes in rspf checking and also messages. 
 23Sep01=======================================================================
diff --git a/perl/BadWords.pm b/perl/BadWords.pm
index e7d1169e..81e7ec90 100644
--- a/perl/BadWords.pm
+++ b/perl/BadWords.pm
@@ -56,7 +56,29 @@ sub load
 # check the text against the badwords list
 sub check
 {
-	return grep { $badword->in($_) } split(/\b/, lc shift);
+	my $s = uc shift;
+	
+	for (split(/\s+/, $s)) {
+		s/[^\w]//g;
+		return $_ if $badword->in($_);
+		s/\'?S$//;
+		return $_ if $badword->in($_);
+	}
+	
+	# look for a few of the common ones with spaces and stuff
+	if ($s =~ /F[\s\W]*U[\s\W]*C[\s\W]*K/) {
+		return "FUCK";
+	} elsif ($s =~ /C[\s\W]*U[\s\W]*N[\s\W]*T/) {
+		return "CUNT";
+	} elsif ($s =~ /W[\s\W]*A[\s\W]*N[\s\W]*K/) {
+		return "WANK";
+	} elsif ($s =~ /C[\s\W]*[0O][\s\W]*C[\s\W]*K/) {
+		return "COCK";
+	} elsif ($s =~ /S[\s\W]*H[\s\W]*[I1][\s\W]*T/) {
+		return "SHIT";
+	}
+	
+	return undef;
 }
 
 1;