Text::Roman - Allows conversion between Roman and Arabic algarisms.


version 3.5


    #!/usr/bin/env perl
    use strict;
    use warnings;
    use Text::Roman qw(:all);

    print int2roman(123), "\n";

    my $roman = "XXXV";
    print roman2int($roman), "\n" if isroman($roman);

    my $milhar = 'L_X_XXIII'; # = 60,023
    print milhar2int($milhar), "\n" if ismilhar($milhar);


This package supports both conventional Roman algarisms (which range from 1 to 3999) and Milhar Romans, a variation which uses a bar across the algarism to indicate multiplication by 1_000. For the purposes of this module, acceptable syntax consists of an underscore suffixed to the algarism e.g. IV_V = 4_005. The term Milhar apparently derives from the Portuguese word for "thousands" and the range of this notation extends the range of Roman numbers to 3999 * 1000 + 3999 = 4_002_999.

Note: the functions in this package treat Roman algarisms in a case-insensitive manner such that "VI" == "vI" == "Vi" == "vi".

The following functions may be imported into the caller package by name:



Tests a string to be a valid Roman algarism. Returns a boolean value.


Converts an integer expressed in Arabic numerals, to its corresponding Roman algarism. If the integer provided is out of the range expressible in Roman notation, an undef is returned.


Does the converse of "int2roman", converting a Roman algarism to its integer value.


Determines whether a string qualifies as a Milhar Roman algarism.


Converts a Milhar Roman algarism to an integer.


These functions belong to the module's old interface and are considered deprecated. Do not use them in new code and they will eventually be discontinued; they map as follows:

  • ismroman => ismilhar

  • mroman2int => milhar2int

  • roman => int2roman


Some changes worth noting from this module's previous incarnation:

namespace imports

The call to use must now explicitly request function names imported into it's namespace.

argument defaults/void context

All functions now will operate on $_ when no arguments are passed, and will set $_ when called in a void context. This allows for writing code like:

    @x = qw/V III XI IV/;
    roman2int() for @x;
    print join("-", @x);

instead of the uglier:

    @x = qw/V III XI IV/;
    $_ = roman2int($_) for @x;
    print join("-", @x);


Roman algarisms may be described using the following BNF-like formula:

    a   = I{1,3}
    b   = V\a?|IV|\a
    e   = X{1,3}\b?|X{0,3}IX|\b
    ee  = IX|\b
    f   = L\e?|XL\ee?|\e
    g   = C{1,3}\f?|C{0,3}XC\ee?|\f
    gg  = XC\ee?|\f
    h   = D\g?|CD\gg?|\g
    j   = M{1,3}\h?|M{0,3}CM\gg?|\h


For a description of the Roman numeral system see: A reference to Milhar Roman alagarisms (in Portuguese) may be found at:


This module was originally written by Peter de Padua Krauss and submitted to CPAN by Stanislaw Pusep who has relinquished control to Erick Calder since the original author has never maintained it and can no longer be reached.

Erick have completely rewritten the module, implementing simpler algorithms to perform the same functionality, adding a test suite, a Changes file, etc. and providing more comprehensive documentation.

Ten years later, Stanislaw returned as a maintainer.


Stanislaw Pusep <>


This software is copyright (c) 2003 by Erick Calder <>.

This is free software; you can redistribute it and/or modify it under the same terms as the Perl 5 programming language system itself.