File Coverage

blib/lib/Lingua/RON/Word2Num.pm

Criterion	Covered	Total	%
statement	23	35	65.7
branch	0	12	0.0
condition	2	4	50.0
subroutine	9	10	90.0
pod	3	3	100.0
total	37	64	57.8

# For Emacs: -*- mode:cperl; eval: (folding-mode 1); coding:utf-8 -*-
package Lingua::RON::Word2Num;
# ABSTRACT: Word to number conversion in Romanian

use 5.16.0;
use utf8;
use warnings;

# {{{ use block

use Export::Attrs;
use Parse::RecDescent;

# }}}
# {{{ var block
our $VERSION = '0.2603300';
my $parser   = ron_numerals();

# }}}

# {{{ w2n                       convert text to number

sub w2n :Export {
    my $input = shift // return;

    return $parser->numeral($input);
}

# }}}
# {{{ ron_numerals              create parser for romanian numerals

sub ron_numerals {
    return Parse::RecDescent->new(q{


      numeral:      mega
                  | kOhOd
                  | 'zero'        {  0 }
                  |               {    }

      number:       'unsprezece'          { 11 }
                  | 'doisprezece'         { 12 }
                  | 'treisprezece'        { 13 }
                  | 'paisprezece'         { 14 }
                  | 'cincisprezece'       { 15 }
                  | /[sș]aisprezece/      { 16 }
                  | /[sș]aptesprezece/    { 17 }
                  | 'optsprezece'         { 18 }
                  | /nou[aă]sprezece/     { 19 }
                  | 'zece'                { 10 }
                  | /un[ua]?/             {  1 }
                  | /dou[aă]/             {  2 }
                  | 'doi'                 {  2 }
                  | 'trei'                {  3 }
                  | 'patru'               {  4 }
                  | 'cinci'               {  5 }
                  | /[sș]ase/             {  6 }
                  | /[sș]apte/            {  7 }
                  | 'opt'                 {  8 }
                  | /nou[aă]/             {  9 }

      tens:         /dou[aă]zeci/         { 20 }
                  | 'treizeci'            { 30 }
                  | 'patruzeci'           { 40 }
                  | 'cincizeci'           { 50 }
                  | /[sș]aizeci/          { 60 }
                  | /[sș]aptezeci/        { 70 }
                  | 'optzeci'             { 80 }
                  | /nou[aă]zeci/         { 90 }

      deca:         tens /[sș]i/ number   { $item[1] + $item[3] }
                  | tens
                  | number

      hecto:        'o' /sut[aă]/ deca    { 100 + $item[3]            }
                  | 'o' /sut[aă]/         { 100                       }
                  | /dou[aă]/ 'sute' deca { 200 + $item[3]            }
                  | /dou[aă]/ 'sute'      { 200                       }
                  | number 'sute' deca    { $item[1] * 100 + $item[3] }
                  | number 'sute'         { $item[1] * 100            }

      hOd:        hecto
                | deca

      kilo:       'o' 'mie' hOd          { 1000 + $item[3]           }
                | 'o' 'mie'              { 1000                      }
                | /dou[aă]/ 'mii' hOd    { 2000 + $item[3]           }
                | /dou[aă]/ 'mii'        { 2000                      }
                | hOd 'mii' hOd          { $item[1] * 1000 + $item[3] }
                | hOd 'mii'             { $item[1] * 1000            }

      kOhOd:      kilo
                | hOd

      mega:       'un' 'milion' kOhOd    { 1_000_000 + $item[3]               }
                | 'un' 'milion'          { 1_000_000                           }
                | /dou[aă]/ 'milioane' kOhOd { 2_000_000 + $item[3]           }
                | /dou[aă]/ 'milioane'       { 2_000_000                      }
                | hOd 'milioane' kOhOd   { $item[1] * 1_000_000 + $item[3]   }
                | hOd 'milioane'         { $item[1] * 1_000_000              }
    });
}

# }}}
# {{{ ordinal2cardinal                              convert ordinal text to cardinal text

sub ordinal2cardinal :Export {
    my $input = shift // return;

    # Romanian ordinals:
    #   "primul/prima" → suppletive for 1st (cardinal: "unu"/"una")
    #   2-10: "al doilea/a doua" etc. — irregular stems
    #   11+:  "al " + cardinal + "lea" / "a " + cardinal + "a"

    # Masculine suppletive 1st
    return 'unu'  if $input =~ m{\A primul \z}xms;
    return 'una'  if $input =~ m{\A prima  \z}xms;

    # Feminine ordinals: "a a"
    # (must check before masculine to avoid ambiguity)
    state $fem_irregular = {
        'a doua'    => 'două',
        'a treia'   => 'trei',
        'a patra'   => 'patru',
        'a cincea'  => 'cinci',
        'a șasea'   => 'șase',
        'a șaptea'  => 'șapte',
        'a opta'    => 'opt',
        'a noua'    => 'nouă',
        'a zecea'   => 'zece',
    };

    return $fem_irregular->{$input} if exists $fem_irregular->{$input};

    # Masculine ordinals: "al lea"
    state $masc_irregular = {
        'al doilea'   => 'doi',
        'al treilea'  => 'trei',
        'al patrulea' => 'patru',
        'al cincilea' => 'cinci',
        'al șaselea'  => 'șase',
        'al șaptelea' => 'șapte',
        'al optulea'  => 'opt',
        'al nouălea'  => 'nouă',
        'al zecelea'  => 'zece',
    };

    return $masc_irregular->{$input} if exists $masc_irregular->{$input};

    # Regular masculine (11+): "al " + cardinal + "lea"
    if ($input =~ m{\A al \s+ (?.+?) lea \z}xms) {
        return $+{card};
    }

    # Regular feminine (11+): "a " + cardinal + "a"
    if ($input =~ m{\A a \s+ (?.+?) a \z}xms) {
        return $+{card};
    }

    return;  # not an ordinal
}

# }}}

1;

__END__