use DXUtil;
use DXVars;
use DXHash;
+use DXDebug;
+
use IO::File;
-use vars qw($badword);
+use vars qw($badword @regex);
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 $regex;
+filecopy("$bwfn.issue", $bwfn) unless -e $bwfn;
+
$badword = new DXHash "badword";
use vars qw($VERSION $BRANCH);
sub load
{
my @out;
- return unless -e $oldfn;
my $fh = new IO::File $oldfn;
if ($fh) {
$fh->close;
$badword->put;
unlink $oldfn;
+ }
+ push @out, create_regex();
+ return @out;
+}
+
+sub create_regex
+{
+ my @out;
+ @regex = ();
+
+ my $fh = new IO::File $regex;
+
+ if ($fh) {
+ 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 = 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 $@;
+ }
+ }
+ $fh->close;
} else {
- my $l = "can't open $oldfn $!";
+ my $l = "can't open $regex $!";
dbg($l);
push @out, $l;
}
+
return @out;
}
# check the text against the badwords list
sub check
{
- return grep { $badword->in($_) } split(/\b/, lc shift);
+ my $s = uc shift;
+ my @out;
+
+ for (@regex) {
+ push @out, &$_($s);
+ }
+
+ return @out if @out;
+
+ for (split(/\s+/, $s)) {
+ s/[^\w]//g;
+ push @out, $_ if $badword->in($_);
+ s/\'?S$//;
+ push @out, $_ if $badword->in($_);
+ }
+
+ return @out;
}
1;