X-Git-Url: http://gb7djk.dxcluster.net/gitweb/gitweb.cgi?a=blobdiff_plain;f=perl%2FBadWords.pm;h=312a04082c705fac550f9b0fcb065388cca2e564;hb=6ec22e78a4a344ce675645fabf18b2a1971f364a;hp=63ec8c881aa5e9db936ecc7fc3582f4577ce3e2d;hpb=cc579a96816b0bae5b37dc132942fc1075449cf3;p=spider.git

diff --git a/perl/BadWords.pm b/perl/BadWords.pm
index 63ec8c88..312a0408 100644
--- a/perl/BadWords.pm
+++ b/perl/BadWords.pm
@@ -3,7 +3,7 @@
 #
 # Copyright (c) 2000 Dirk Koopman
 #
-# $Id$
+#
 #
 
 package BadWords;
@@ -13,43 +13,66 @@ use strict;
 use DXUtil;
 use DXVars;
 use DXHash;
-use IO::File;
+use DXDebug;
 
-use vars qw($badword);
+use IO::File;
 
-my $oldfn = "$main::data/badwords";
-$badword = new DXHash "badword";
+use vars qw($badword $regexcode);
 
-use vars qw($VERSION $BRANCH);
-$VERSION = sprintf( "%d.%03d", q$Revision$ =~ /(\d+)\.(\d+)/ );
-$BRANCH = sprintf( "%d.%03d", q$Revision$ =~ /\d+\.\d+\.(\d+)\.(\d+)/ ) || 0;
-$main::build += $VERSION;
-$main::branch += $BRANCH;
+our $regex;
 
 # load the badwords file
 sub load
 {
+	my $bwfn = localdata("badword");
+	filecopy("$main::data.issue", $bwfn) unless -e $bwfn;
+	
+	my @out;
+
+	$badword = new DXHash "badword";
+	
+	push @out, create_regex(); 
+	return @out;
+}
+
+sub create_regex
+{
+	$regex = localdata("badw_regex");
+	filecopy("$regex.gb.issue", $regex) unless -e $regex;
+	
 	my @out;
-	return unless -e $oldfn;
-	my $fh = new IO::File $oldfn;
+	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) {
-				$badword->add($_);
+				# 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;
+				$s .= qq{push \@out, \$1 if \$str =~ m|\\b($e+)|;\n};
 			}
 		}
+		$s .= "return \@out;\n}";
+		$regexcode = eval $s;
+		dbg($s) if isdbg('badword');
+		if ($@) {
+			@out = ($@);
+			dbg($@);
+			return @out;
+		}
 		$fh->close;
-		$badword->put;
-		unlink $oldfn;
 	} else {
-		my $l = "can't open $oldfn $!";
+		my $l = "can't open $regex $!";
 		dbg($l);
 		push @out, $l;
 	}
+	
 	return @out;
 }
 
@@ -57,28 +80,17 @@ sub load
 sub check
 {
 	my $s = uc shift;
+	my @out;
+
+	push @out, &$regexcode($s) if $regexcode;
 	
-	for (split(/\s+/, $s)) {
-		s/[^\w]//g;
-		return $_ if $badword->in($_);
-		s/\'?S$//;
-		return $_ if $badword->in($_);
-	}
+	return @out if @out;
 	
-	# 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";
+	for (split(/\b/, $s)) {
+		push @out, $_ if $badword->in($_);
 	}
-	
-	return ();
+
+	return @out;
 }
 
 1;