X-Git-Url: http://gb7djk.dxcluster.net/gitweb/gitweb.cgi?a=blobdiff_plain;f=perl%2FBadWords.pm;h=141b3e9a729b6a933149cb0bf71bf4a942938285;hb=refs%2Fheads%2Fnew-spawn;hp=9814e3fadd5ef712d25315887d1582e2c509d338;hpb=e0cfa6eebc0cddc92fe3c45636e6f1f641edda1b;p=spider.git

diff --git a/perl/BadWords.pm b/perl/BadWords.pm
index 9814e3fa..141b3e9a 100644
--- a/perl/BadWords.pm
+++ b/perl/BadWords.pm
@@ -3,7 +3,7 @@
 #
 # Copyright (c) 2000 Dirk Koopman
 #
-# $Id$
+#
 #
 
 package BadWords;
@@ -13,24 +13,26 @@ 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";
-$badword = new DXHash "badword";
+my $oldfn = localdata("badwords");
+my $regex = localdata("badw_regex");
+my $bwfn = localdata("badword");
+
+# copy issue ones across
+filecopy("$regex.gb.issue", $regex) unless -e $regex;
+filecopy("$bwfn.issue", $bwfn) unless -e $bwfn;
 
-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 += $BRANCH;
+$badword = new DXHash "badword";
 
 # load the badwords file
 sub load
 {
 	my @out;
-	return unless -e $oldfn;
 	my $fh = new IO::File $oldfn;
 	
 	if ($fh) {
@@ -45,11 +47,46 @@ 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;
+				$s .= "push \@out, \$1 if \$str =~ /\\b($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;
 }
 
@@ -57,32 +94,17 @@ 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 ($s =~ /S[\s\W]*H[\s\W]*[I1][\s\W]*T/) {
-		return "SHIT";
-	} elsif ($s =~ /P[\s\W]*[I1][\s\W]*S[\s\W]*S/) {
-		return "PISS";
-	} elsif ($s =~ /B[\s\W]*[O0][\s\W]*L[\s\W]*L[\s\W]*[O0][\s\W]*[CK]/) {
-		return "BOLLOCKS";
+	for (split(/\b/, $s)) {
+		push @out, $_ if $badword->in($_);
 	}
-	
-	return ();
+
+	return @out;
 }
 
 1;