use HTML::TokeParser; use Data::Dumper; my $p = HTML::TokeParser->new( \" hada "); while($_=$p->get_token){ print Dumper($_); } __END__ $VAR1 = [ 'T', ' ', '' ]; $VAR1 = [ 'PI', 'php is=1 2 3 four 5 times the=\'something\' ?', '' ]; $VAR1 = [ 'T', ' hada', '' ]; $VAR1 = [ 'T', ' ', '' ];