File Coverage

blib/lib/Lingua/GLG/Word2Num.pm

Criterion	Covered	Total	%
statement	26	41	63.4
branch	0	22	0.0
condition	2	4	50.0
subroutine	9	10	90.0
pod	3	3	100.0
total	40	80	50.0

# For Emacs: -*- mode:cperl; eval: (folding-mode 1); coding:utf-8; -*-

package Lingua::GLG::Word2Num;
# ABSTRACT: Word to number conversion in Galician

use 5.16.0;
use utf8;
use warnings;

# {{{ use block

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

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

# }}}

# {{{ w2n                                         convert text to number

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

    $input = lc $input;
    $input =~ s/\s+/ /g;
    $input =~ s/^\s+|\s+$//g;

    return $parser->numeral($input);
}
# }}}
# {{{ glg_numerals                                 create parser for numerals

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



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

       number:  'un'           {  1 }
             |  'dous'         {  2 }
             |  'tres'         {  3 }
             |  'catro'        {  4 }
             |  'cinco'        {  5 }
             |  'seis'         {  6 }
             |  'sete'         {  7 }
             |  'oito'         {  8 }
             |  'nove'         {  9 }
             |  'dezaseis'     { 16 }
             |  'dezasete'     { 17 }
             |  'dezaoito'     { 18 }
             |  'dezanove'     { 19 }
             |  'dez'          { 10 }
             |  'once'         { 11 }
             |  'doce'         { 12 }
             |  'trece'        { 13 }
             |  'catorce'      { 14 }
             |  'quince'       { 15 }

         tens:  'vinte'        { 20 }
             |  'trinta'       { 30 }
             |  'corenta'      { 40 }
             |  'cincuenta'    { 50 }
             |  'sesenta'      { 60 }
             |  'setenta'      { 70 }
             |  'oitenta'      { 80 }
             |  'noventa'      { 90 }

     hundreds:  'cincocentos'        { 500 }
             |  'setecentos'         { 700 }
             |  'novecentos'         { 900 }
             |  'oitocentos'         { 800 }
             |  'seiscentos'         { 600 }
             |  'catrocentos'        { 400 }
             |  'trescentos'         { 300 }
             |  'douscentos'         { 200 }
             |  /cen(to)?/           { 100 }

         deca:  tens 'e' number      { $item[1] + $item[3] }
             |  tens
             |  number

        hecto:  hundreds 'e' deca     { $item[1] + $item[3] }
             |  hundreds deca        { $item[1] + $item[2] }
             |  hundreds

          hOd:  hecto
             |  deca

         kilo:  hOd milnotmeg 'e' hOd { $item[1] * 1000 + $item[4] }
             |  hOd milnotmeg hOd    { $item[1] * 1000 + $item[3] }
             |  hOd milnotmeg        { $item[1] * 1000 }
             |      milnotmeg 'e' hOd { 1000 + $item[3] }
             |      milnotmeg hOd    { 1000 + $item[2] }
             |      milnotmeg        { 1000 }

    milnotmeg:   ...!/mill(ón|óns)/ 'mil'

        kOhOd:  kilo
             |  hOd

         mega:  hOd /mill(óns|ón)/ 'e' kOhOd  { $item[1] * 1_000_000 + $item[4] }
             |  hOd /mill(óns|ón)/ kOhOd     { $item[1] * 1_000_000 + $item[3] }
             |  hOd /mill(óns|ón)/            { $item[1] * 1_000_000 }
    });
}

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

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

    # Galician ordinals 1-10 are fully irregular
    state $irregular = {
        'primeiro'  => 'un',      'primeira'  => 'un',
        'segundo'   => 'dous',    'segunda'   => 'dous',
        'terceiro'  => 'tres',    'terceira'  => 'tres',
        'cuarto'    => 'catro',   'cuarta'    => 'catro',
        'quinto'    => 'cinco',   'quinta'    => 'cinco',
        'sexto'     => 'seis',    'sexta'     => 'seis',
        'sétimo'    => 'sete',    'sétima'    => 'sete',
        'oitavo'    => 'oito',    'oitava'    => 'oito',
        'noveno'    => 'nove',    'novena'    => 'nove',
        'décimo'    => 'dez',     'décima'    => 'dez',
    };

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

    # Regular (11+): cardinal (drop final vowel) + "ésimo/ésima"
    $input =~ s{ésim[oa]\z}{}xms or return;

    # Galician drops the final vowel before adding -ésimo.  The dropped
    # vowel varies by word, so we restore it based on the stem ending.

    # cinco: cinc→cinco (must precede the general -c rule)
    if    ($input =~ m{\Acinc\z}xms)             { $input .= 'o' }
    # also "vinte e cinc" etc. — cinc at end of compound
    elsif ($input =~ m{cinc\z}xms)               { $input .= 'o' }
    # teens ending in -c: onc→once, doc→doce, trec→trece, catorc→catorce, quinc→quince
    elsif ($input =~ m{c\z}xms)                  { $input .= 'e' }
    # oito family: dezaoit→dezaoito, oit→oito
    elsif ($input =~ m{oit\z}xms)                { $input .= 'o' }
    # sete family: dezaset→dezasete, set→sete
    elsif ($input =~ m{set\z}xms)                { $input .= 'e' }
    # vinte: vint→vinte
    elsif ($input =~ m{vint\z}xms)               { $input .= 'e' }
    # decades (trinta, corenta, etc.): trint→trinta, corent→corenta
    elsif ($input =~ m{nt\z}xms)                 { $input .= 'a' }
    # nove family: dezanov→dezanove, nov→nove
    elsif ($input =~ m{ov\z}xms)                 { $input .= 'e' }
    # catro: catr→catro
    elsif ($input =~ m{tr\z}xms)                 { $input .= 'o' }

    return $input;
}

# }}}

1;

__END__