lib/AkInflexor.php

   1 <?php
   2 /* vim: set expandtab tabstop=4 shiftwidth=4 softtabstop=4: */
   3
   4 // +----------------------------------------------------------------------+
   5 // | Akelos Framework - http://www.akelos.org                             |
   6 // +----------------------------------------------------------------------+
   7 // | Copyright (c) 2002-2006, Akelos Media, S.L.  & Bermi Ferrer Martinez |
   8 // | Released under the GNU Lesser General Public License, see LICENSE.txt|
   9 // +----------------------------------------------------------------------+
  10
  11 /**
  12  * @package ActiveSupport
  13  * @subpackage Inflector
  14  * @author Bermi Ferrer <bermi a.t akelos c.om>
  15  * @copyright Copyright (c) 2002-2006, Akelos Media, S.L. http://www.akelos.org
  16  * @license GNU Lesser General Public License <http://www.gnu.org/copyleft/lesser.html>
  17  */
  18
  19 require_once(AK_LIB_DIR.DS.'AkInflector.php');
  20
  21 /**
  22 * Spanish Inflector
  23 */
  24 class AkInflexor extends AkInflector
  25 {
  26
  27     function pluralize($word)
  28     {
  29         $plural = array(
  30         '/([aeiou])x$/i'=> '\\1x', // This could fail if the word is oxytone.
  31         '/([áéíóú])([ns])$/i'=> '|1\\2es',
  32         '/(^[bcdfghjklmnñpqrstvwxyz]*)an$/i'=>'\\1anes', //clan->clanes
  33         '/([áéíóú])s$/i'=> '|1ses',
  34         '/(^[bcdfghjklmnñpqrstvwxyz]*)([aeiou])([ns])$/i'=>'\\1\\2\\3es', //tren->trenes
  35         '/([aeiouáéó])$/i'=> '\\1s', // casa->casas, padre->padres, papá->papás
  36         '/([aeiou])s$/i'=> '\\1s', // atlas->atlas, virus->virus, etc.
  37         '/([éí])(s)$/i'=> '|1\\2es', // inglés->ingleses
  38         '/z$/i'=> 'ces',  // luz->luces
  39         '/([íú])$/i' => '\\1es', // ceutí->ceutíes, tabú->tabúes
  40         '/(ng|[wckgtp])$/'=>'\\1s', // Anglicismos como puenting, frac, crack, show (En que casos podría fallar esto?)
  41         '/$/i'=> 'es',  // ELSE +es (v.g. árbol->árboles)
  42         );      // We should manage _orden_ -> _órdenes_, _joven_->_jóvenes_ and so.
  43
  44         $uncountable = array('tijeras','gafas', 'vacaciones','víveres','déficit');
  45         /* In fact these words have no singular form: you cannot say neither
  46         "una gafa" nor "un vívere". So we should change the variable name to
  47         $onlyplural or something alike.*/
  48         $irregular = array(
  49         'país'=>'países',
  50         'champú'=>'champús',
  51         'jersey'=>'jerséis',
  52         'carácter'=>'caracteres',
  53         'espécimen'=>'especímenes',
  54         'menú'=>'menús',
  55         'régimen'=>'regímenes',
  56         'curriculum' => 'currículos',
  57         'ultimátum' => 'ultimatos',
  58         'memorándum' => 'memorandos',
  59         'referéndum' => 'referendos'
  60         );
  61         $lowercased_word = strtolower($word);
  62
  63
  64         foreach ($uncountable as $_uncountable){
  65             if(substr($lowercased_word,(-1*strlen($_uncountable))) == $_uncountable){
  66                 return $word;
  67             }
  68         }
  69
  70         foreach ($irregular as $_plural=> $_singular){
  71             if (preg_match('/('.$_plural.')$/i', $word, $arr)) {
  72                 return preg_replace('/('.$_plural.')$/i', substr($arr[0],0,1).substr($_singular,1), $word);
  73             }
  74         }
  75
  76         foreach ($plural as $rule => $replacement) {
  77             if (preg_match($rule, $word, $match)) {
  78                 if(strstr($replacement,'|')){
  79                     foreach ($match as $k=>$v){
  80                         $replacement = str_replace("|$k",strtr($v,'ÁÉÍÓÚáéíóú','AEIOUaeiou'), $replacement);
  81                     }
  82                 }
  83                 $result = preg_replace($rule, $replacement, $word);
  84                 // Esto acentua los sustantivos que al pluralizarse se convierten en esdrújulos como esmóquines, jóvenes...
  85                 if(preg_match('/([aeiou]).{1,3}[aeiou]nes$/i',$result,$match) && !preg_match('/[áéíóú]/i',$word)){
  86                     $result = str_replace($match[0], strtr($match[1],'AEIOUaeiou','ÁÉÍÓÚáéíóú').substr($match[0],1), $result);
  87                 }
  88                 return $result;
  89             }
  90         }
  91         return false;
  92
  93     }
  94
  95     function singularize($word)
  96     {
  97         $singular = array (
  98         '/^([bcdfghjklmnñpqrstvwxyz]*)([aeiou])([ns])es$/i'=> '\\1\\2\\3',
  99         '/([aeiou])([ns])es$/i'=> '~1\\2',
 100         '/oides$/i'=> 'oide', //androides->androide
 101         '/(ces)$/i' => 'z',
 102         '/(sis|tis|xis)+$/i'=> '\\1', //crisis, apendicitis, praxis
 103         '/(é)s$/i'=> '\\1', // bebés->bebé
 104         '/([^e])s$/i'=> '\\1', // casas->casa
 105         '/([bcdfghjklmnñprstvwxyz]{2,}e)s$/i'=>'\\1', // cofres->cofre
 106         '/([ghñpv]e)s$/i'=>'\\1', // 24-01 llaves->llave
 107         '/es$/i'=>'', // ELSE remove _es_  monitores->monitor
 108         );
 109
 110
 111         $uncountable = array('paraguas','tijeras','gafas', 'vacaciones','víveres','lunes','martes','miércoles','jueves','viernes','cumpleaños','virus','atlas','sms');
 112         $irregular = array(
 113         'jersey'=>'jerséis',
 114         'espécimen'=>'especímenes',
 115         'carácter'=>'caracteres',
 116         'régimen'=>'regímenes',
 117         'menú'=>'menús',
 118         'régimen'=>'regímenes',
 119         'curriculum' => 'currículos',
 120         'ultimátum' => 'ultimatos',
 121         'memorándum' => 'memorandos',
 122         'referéndum' => 'referendos',
 123         'sándwich' => 'sándwiches'
 124         );
 125
 126         $lowercased_word = strtolower($word);
 127         foreach ($uncountable as $_uncountable){
 128             if(substr($lowercased_word,(-1*strlen($_uncountable))) == $_uncountable){
 129                 return $word;
 130             }
 131         }
 132
 133         foreach ($irregular as $_plural=> $_singular){
 134             if (preg_match('/('.$_singular.')$/i', $word, $arr)) {
 135                 return preg_replace('/('.$_singular.')$/i', substr($arr[0],0,1).substr($_plural,1), $word);
 136             }
 137         }
 138
 139         foreach ($singular as $rule => $replacement) {
 140             if (preg_match($rule, $word, $match)) {
 141                 if(strstr($replacement,'~')){
 142                     foreach ($match as $k=>$v){
 143                         $replacement = str_replace("~$k",strtr($v,'AEIOUaeiou','ÁÉÍÓÚáéíóú'), $replacement);
 144                     }
 145                 }
 146
 147                 $result = preg_replace($rule, $replacement, $word);
 148                 // Esta es una posible solución para el problema de dobles acentos. Un poco sucio pero funciona
 149                 $result = preg_match('/([áéíóú]).*([áéíóú])/',$result) ? strtr($result,'ÁÉÍÓÚáéíóú','AEIOUaeiou') : $result;
 150
 151                 return $result;
 152             }
 153         }
 154
 155         return $word;
 156     }
 157 }
 158
 159 ?>