my $token_grammar = qr {
        
        <nocontext:>
        <debug: on>
        
        <[unit]>+
        
        <token: unit>                  <word> | <punctuation>
        <token: word>             \b \w+ \b   |  <hyphenword>  |  <contraction>
        <token: hyphenword>  \b \w+ ( - | -- ) \w+ \b
        <token: contraction>    \b \w+ ' \w \b
        <token: punctuation>   ( \.+ | ,+ | ?+ | !+ | "+ | '+ | ;+ | :+ | -+ | &+ | \(+ | \)+ | \[+ | \]+ )
        
    };