#!/usr/bin/perl -T use strict; use warnings; use v5.10; use Scalar::Util; use locale; use POSIX qw(locale_h); setlocale(LC_CTYPE, "de_DE"); sub filter_string { my $s = shift; warn("raw string '$s' is tainted") if Scalar::Util::tainted($s); $s =~ m:^(.+?) \([0-9]{4}(/[IVX]+)?\).*?$: or die "String '$s' doesn't match"; $s = $1; warn("filtered string '$s' is tainted") if Scalar::Util::tainted($s); $s; } my @strings = ('ABC1 (1945/IXV)', 'ABC2 (1945)', 'ABC3 (1945)', 'ABC4 (1945) {DEF()}', 'ABC5 (1945) {D ()}', 'ABC6 (1945) {D(#)}', 'ABC7 (1945) {D (#)}', 'ABC8 (1945) {()}', 'ABC9 (1945) {D ()}', 'ABC10 (1945) ()', 'ABC11 (1945) {()', 'ABC12 (1945) { ()}', 'ABC13 (1945) { ()', 'ABC14 (1945) x(/', 'ABC15 (1945) ( ', 'ABC16 (1945) (', 'ABC17 (1945) { (', 'ABC18 (1945) { (#', ); foreach my $s (@strings) { say filter_string($s); }