X-Git-Url: http://dxcluster.org/gitweb/gitweb.cgi?a=blobdiff_plain;f=perl%2FBadWords.pm;h=7ae0373f436f007edcd259af76b2699f3f210a6d;hb=21ee7b1845f8fc5ab10c6a33a058cfbd11d7a705;hp=63ec8c881aa5e9db936ecc7fc3582f4577ce3e2d;hpb=cc579a96816b0bae5b37dc132942fc1075449cf3;p=spider.git

diff --git a/perl/BadWords.pm b/perl/BadWords.pm
index 63ec8c88..7ae0373f 100644
--- a/perl/BadWords.pm
+++ b/perl/BadWords.pm
@@ -13,11 +13,20 @@ use strict;
 use DXUtil;
 use DXVars;
 use DXHash;
+use DXDebug;
+
 use IO::File;
 
-use vars qw($badword);
+use vars qw($badword $regexcode);
 
 my $oldfn = "$main::data/badwords";
+my $regex = "$main::data/badw_regex";
+my $bwfn = "$main::data/badword";
+
+# copy issue ones across
+filecopy("$regex.gb.issue", $regex) unless -e $regex;
+filecopy("$bwfn.issue", $bwfn) unless -e $bwfn;
+
 $badword = new DXHash "badword";
 
 use vars qw($VERSION $BRANCH);
@@ -30,7 +39,6 @@ $main::branch += $BRANCH;
 sub load
 {
 	my @out;
-	return unless -e $oldfn;
 	my $fh = new IO::File $oldfn;
 	
 	if ($fh) {
@@ -45,11 +53,46 @@ sub load
 		$fh->close;
 		$badword->put;
 		unlink $oldfn;
+	}
+	push @out, create_regex(); 
+	return @out;
+}
+
+sub create_regex
+{
+	my @out;
+	my $fh = new IO::File $regex;
+	
+	if ($fh) {
+		my $s = "sub { my \$str = shift; my \@out; \n";
+		while (<$fh>) {
+			chomp;
+			next if /^\s*\#/;
+			my @list = split " ";
+			for (@list) {
+				# create a closure for each word so that it matches stuff with spaces/punctuation
+				# and repeated characters in it
+				my $w = uc $_;
+				my @l = split //, $w;
+				my $e = join '+[\s\W]*', @l;
+				$s .= "push \@out, \$1 if \$str =~ /($e)/;\n";
+			}
+		}
+		$s .= "return \@out;\n}";
+		$regexcode = eval $s;
+		dbg($s) if isdbg('badword');
+		if ($@) {
+			@out = ($@);
+			dbg($@);
+			return @out;
+		}
+		$fh->close;
 	} else {
-		my $l = "can't open $oldfn $!";
+		my $l = "can't open $regex $!";
 		dbg($l);
 		push @out, $l;
 	}
+	
 	return @out;
 }
 
@@ -57,28 +100,18 @@ sub load
 sub check
 {
 	my $s = uc shift;
+	my @out;
+
+	push @out, &$regexcode($s) if $regexcode;
+	
+	return @out if @out;
 	
 	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";
+		push @out, $_ if $badword->in($_);
 	}
-	
-	return ();
+
+	return @out;
 }
 
 1;