use IO::File;
-use vars qw($badword @regex);
+use vars qw($badword $regexcode);
my $oldfn = "$main::data/badwords";
my $regex = "$main::data/badw_regex";
my $bwfn = "$main::data/badword";
# copy issue ones across
-filecopy("$regex.issue", $regex) unless -e $regex;
+filecopy("$regex.gb.issue", $regex) unless -e $regex;
filecopy("$bwfn.issue", $bwfn) unless -e $bwfn;
$badword = new DXHash "badword";
use vars qw($VERSION $BRANCH);
$VERSION = sprintf( "%d.%03d", q$Revision$ =~ /(\d+)\.(\d+)/ );
-$BRANCH = sprintf( "%d.%03d", q$Revision$ =~ /\d+\.\d+\.(\d+)\.(\d+)/ ) || 0;
+$BRANCH = sprintf( "%d.%03d", q$Revision$ =~ /\d+\.\d+\.(\d+)\.(\d+)/ || (0,0));
$main::build += $VERSION;
$main::branch += $BRANCH;
sub create_regex
{
my @out;
- @regex = ();
-
my $fh = new IO::File $regex;
if ($fh) {
+ my $s = "sub { my \$str = shift; my \@out; \n";
while (<$fh>) {
chomp;
next if /^\s*\#/;
# create a closure for each word so that it matches stuff with spaces/punctuation
# and repeated characters in it
my $w = uc $_;
- my @l = map { $_ eq 'I' ? '[I1]' : ($_ eq 'O' ? '[O0]' : $_) }split //, $w;
- my $e = join '+[\s\W]+', @l;
- my $s = eval qq{sub { return \$_[0] =~ /$e+/ ? '$w' : () } };
- push @regex, $s unless $@;
- dbg("create_regex: $@") if $@;
+ 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 $regex $!";
{
my $s = uc shift;
my @out;
-
- for (@regex) {
- push @out, &$_($s);
- }
+
+ push @out, &$regexcode($s) if $regexcode;
return @out if @out;
for (split(/\s+/, $s)) {
- s/[^\w]//g;
- push @out, $_ if $badword->in($_);
s/\'?S$//;
push @out, $_ if $badword->in($_);
}