Skip to content

[String] Add Spanish inflector with some rules #58228

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Sep 18, 2024
Merged

[String] Add Spanish inflector with some rules #58228

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
126 changes: 126 additions & 0 deletions src/Symfony/Component/String/Inflector/SpanishInflector.php
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,126 @@
<?php

/*
* This file is part of the Symfony package.
*
* (c) Fabien Potencier <fabien@symfony.com>
*
* For the full copyright and license information, please view the LICENSE
* file that was distributed with this source code.
*/

namespace Symfony\Component\String\Inflector;

final class SpanishInflector implements InflectorInterface
{
/**
* A list of all rules for pluralise.
*
* @see https://www.spanishdict.com/guide/spanish-plural-noun-forms
* @see https://www.rae.es/gram%C3%A1tica/morfolog%C3%ADa/la-formaci%C3%B3n-del-plural-plurales-en-s-y-plurales-en-es-reglas-generales
*/
// First entry: regex
// Second entry: replacement
private const PLURALIZE_REGEXP = [
// Specials sí, no
['/(sí|no)$/i', '\1es'],

// Words ending with vowel must use -s (RAE 3.2a, 3.2c)
['/(a|e|i|o|u|á|é|í|ó|ú)$/i', '\1s'],

// Word ending in s or x and the previous letter is accented (RAE 3.2n)
['/ás$/i', 'ases'],
['/és$/i', 'eses'],
['/ís$/i', 'ises'],
['/ós$/i', 'oses'],
['/ús$/i', 'uses'],

// Words ending in -ión must changed to -iones
['/ión$/i', '\1iones'],

// Words ending in some consonants must use -es (RAE 3.2k)
['/(l|r|n|d|j|s|x|ch|y)$/i', '\1es'],

// Word ending in z, must changed to ces
['/(z)$/i', 'ces'],
];

/**
* A list of all rules for singularize.
*/
private const SINGULARIZE_REGEXP = [
// Specials sí, no
['/(sí|no)es$/i', '\1'],

// Words ending in -ión must changed to -iones
['/iones$/i', '\1ión'],

// Word ending in z, must changed to ces
['/ces$/i', 'z'],

// Word ending in s or x and the previous letter is accented (RAE 3.2n)
['/(\w)ases$/i', '\1ás'],
['/eses$/i', 'és'],
['/ises$/i', 'ís'],
['/(\w{2,})oses$/i', '\1ós'],
['/(\w)uses$/i', '\1ús'],

// Words ending in some consonants and -es, must be the consonants
['/(l|r|n|d|j|s|x|ch|y)e?s$/i', '\1'],

// Words ended with vowel and s, must be vowel
['/(a|e|i|o|u|á|é|ó|í|ú)s$/i', '\1'],
];

private const UNINFLECTED_RULES = [
// Words ending with pies (RAE 3.2n)
'/.*(piés)$/i',
];

private const UNINFLECTED = '/^(lunes|martes|miércoles|jueves|viernes|análisis|torax|yo|pies)$/i';

public function singularize(string $plural): array
{
if ($this->isInflectedWord($plural)) {
return [$plural];
}

foreach (self::SINGULARIZE_REGEXP as $rule) {
[$regexp, $replace] = $rule;

if (1 === preg_match($regexp, $plural)) {
return [preg_replace($regexp, $replace, $plural)];
}
}

return [$plural];
}

public function pluralize(string $singular): array
{
if ($this->isInflectedWord($singular)) {
return [$singular];
}

foreach (self::PLURALIZE_REGEXP as $rule) {
[$regexp, $replace] = $rule;

if (1 === preg_match($regexp, $singular)) {
return [preg_replace($regexp, $replace, $singular)];
}
}

return [$singular.'s'];
}

private function isInflectedWord(string $word): bool
{
foreach (self::UNINFLECTED_RULES as $rule) {
if (1 === preg_match($rule, $word)) {
return true;
}
}

return 1 === preg_match(self::UNINFLECTED, $word);
}
}
158 changes: 158 additions & 0 deletions src/Symfony/Component/String/Tests/Inflector/SpanishInflectorTest.php
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,158 @@
<?php

/*
* This file is part of the Symfony package.
*
* (c) Fabien Potencier <fabien@symfony.com>
*
* For the full copyright and license information, please view the LICENSE
* file that was distributed with this source code.
*/

namespace Symfony\Component\String\Tests\Inflector;

use PHPUnit\Framework\TestCase;
use Symfony\Component\String\Inflector\SpanishInflector;

class SpanishInflectorTest extends TestCase
{
public static function singularizeProvider(): array
{
return [
// vowels (RAE 3.2a, 3.2c)
['peras', 'pera'],
['especies', 'especie'],
['álcalis', 'álcali'],
['códigos', 'código'],
['espíritus', 'espíritu'],

// accented (RAE 3.2a, 3.2c)
['papás', 'papá'],
['cafés', 'café'],
['isrealís', 'isrealí'],
['burós', 'buró'],
['tisús', 'tisú'],

// ending in -ión
['aviones', 'avión'],
['camiones', 'camión'],

// ending in some letters (RAE 3.2k)
['amores', 'amor'],
['antifaces', 'antifaz'],
['atriles', 'atril'],
['fácsimiles', 'fácsimil'],
['vides', 'vid'],
['reyes', 'rey'],
['relojes', 'reloj'],
['faxes', 'fax'],
['sándwiches', 'sándwich'],
['cánones', 'cánon'],

// (RAE 3.2n)
['adioses', 'adiós'],
['aguarrases', 'aguarrás'],
['arneses', 'arnés'],
['autobuses', 'autobús'],
['kermeses', 'kermés'],
['palmareses', 'palmarés'],
['toses', 'tos'],

// Special
['síes', 'sí'],
['noes', 'no'],
];
}

public static function pluralizeProvider(): array
{
return [
// vowels (RAE 3.2a, 3.2c)
['pera', 'peras'],
['especie', 'especies'],
['álcali', 'álcalis'],
['código', 'códigos'],
['espíritu', 'espíritus'],

// accented (RAE 3.2a, 3.2c)
['papá', 'papás'],
['café', 'cafés'],
['isrealí', 'isrealís'],
['buró', 'burós'],
['tisú', 'tisús'],

// ending in -ión
['avión', 'aviones'],
['camión', 'camiones'],

// ending in some letters (RAE 3.2k)
['amor', 'amores'],
['antifaz', 'antifaces'],
['atril', 'atriles'],
['fácsimil', 'fácsimiles'],
['vid', 'vides'],
['rey', 'reyes'],
['reloj', 'relojes'],
['fax', 'faxes'],
['sándwich', 'sándwiches'],
['cánon', 'cánones'],

// (RAE 3.2n)
['adiós', 'adioses'],
['aguarrás', 'aguarrases'],
['arnés', 'arneses'],
['autobús', 'autobuses'],
['kermés', 'kermeses'],
['palmarés', 'palmareses'],
['tos', 'toses'],

// Specials
['sí', 'síes'],
['no', 'noes'],
];
}

public static function uninflectedProvider(): array
{
return [
['lunes'],
['rodapiés'],
['reposapiés'],
['miércoles'],
['pies'],
];
}

/**
* @dataProvider singularizeProvider
*/
public function testSingularize(string $plural, $singular)
{
$this->assertSame(
\is_array($singular) ? $singular : [$singular],
(new SpanishInflector())->singularize($plural)
);
}

/**
* @dataProvider pluralizeProvider
*/
public function testPluralize(string $singular, $plural)
{
$this->assertSame(
\is_array($plural) ? $plural : [$plural],
(new SpanishInflector())->pluralize($singular)
);
}

/**
* @dataProvider uninflectedProvider
*/
public function testUninflected(string $word)
{
$this->assertSame(
\is_array($word) ? $word : [$word],
(new SpanishInflector())->pluralize($word)
);
}
}
Loading