#!/usr/bin/env perl

# demonstration of a simple lexer that uses the following features of Perl:
#           (?{ code })     - run a piece of code, in the middle of running a regexp
#           $^N             - the most recent capture group
#           $^R             - the return value of the last (?{ ... })

    use strict;
    use warnings;

    use Data::Dumper;


my $str = '(3 + 2) * 5';


my @tokens;

$str =~ /
  ^
  (
    (?:
            \s+             |
            ( \(        )       (?{ [OPEN_PAREN  => $^N]  })    |
            ( \)        )       (?{ [CLOSE_PAREN => $^N]  })    |
            ( \d+       )       (?{ [NUMBER      => $^N]  })    |
            ( [-+*\/]   )       (?{ [OPERATOR    => $^N]  })
    )
    (?{ push @tokens, $^R   if $^R;
        undef;                      # reset $^R
    })
  )*
  $
/x  or die "syntax error\n";


print Dumper \@tokens;