package org.andromda.utils.inflector;
   
   import java.util.LinkedHashMap;
   import java.util.Map;
   
   /**
    * Language utility for transforming English words.
    * See also <a href="http://www.csse.monash.edu.au/~damian/papers/HTML/Plurals.html">
    * http://www.csse.monash.edu.au/~damian/papers/HTML/Plurals.html</a>;
   *
   * @author maetl@coretxt.net.nz
   * @author wouter@andromda.org
   * @author bob@andromda.org
   */
  public class EnglishInflector
  {
      /**
       * Converts an English word to plural form. Leaves it alone if already plural.
       *
       * @param word an English word
       * @return the pluralization of the argument English word, or the argument in case it is <code>null</code>
       */
      public static String pluralize(String word)
      {
          if (word == null) return null;
  
          final Map<String, String> rules = EnglishInflector.getPluralRules();
          for (final Map.Entry<String, String> rule : rules.entrySet())
          {
              final String pattern = rule.getKey();
              final String replace = rule.getValue();
              if (word.matches(pattern))
              {
                  return word.replaceFirst(pattern, replace);
              }
          }
          return word.replaceFirst("([\\w]+)([^s])$", "$1$2s");
      }
  
      /**
       * Returns map of plural patterns
       */
      private static Map<String, String> getPluralRules()
      {
          // Rules are checked in order through LinkedHashMap.
          final Map<String, String> rules = new LinkedHashMap<String, String>();
          // Check first if special case word is already plural
          rules.put("(\\w*)people$", "$1people");
          rules.put("(\\w*)children$", "$1children");
          rules.put("(\\w*)feet$", "$1feet");
          rules.put("(\\w*)teeth$", "$1teeth");
          rules.put("(\\w*)men$", "$1men");
          rules.put("(\\w*)equipment$", "$1equipment");
          rules.put("(\\w*)information$", "$1information");
          rules.put("(\\w*)rice$", "$1rice");
          rules.put("(\\w*)money$", "$1money");
          rules.put("(\\w*)fish$", "$fish");
          rules.put("(\\w*)sheep$", "$1sheep");
          rules.put("(\\w+)(es)$", "$1es");
          // Check exception special case words
          rules.put("(\\w*)person$", "$1people");
          rules.put("(\\w*)child$", "$1children");
          rules.put("(\\w*)foot$", "$1feet");
          rules.put("(\\w*)tooth$", "$1teeth");
          rules.put("(\\w*)bus$", "$1buses");
          rules.put("(\\w*)man$", "$1men");
          rules.put("(\\w*)(ox|oxen)$", "$1$2");
          rules.put("(\\w*)(buffal|tomat)o$", "$1$2oes");
          rules.put("(\\w*)quiz$", "$1$2zes");
          // Greek endings
          rules.put("(\\w+)(matr|vert|ind)ix|ex$", "$1$2ices");
          rules.put("(\\w+)(sis)$", "$1ses");
          rules.put("(\\w+)(um)$", "$1a");
          // Old English. hoof -> hooves, leaf -> leaves
          rules.put("(\\w*)(fe)$", "$1ves");
          rules.put("(\\w*)(f)$", "$1ves");
          rules.put("(\\w*)([m|l])ouse$", "$1$2ice");
          // Y preceded by a consonant changes to ies
          rules.put("(\\w+)([^aeiou]|qu)y$", "$1$2ies");
          // Voiced consonants add es instead of s
          rules.put("(\\w+)(z|ch|sh|as|ss|us|x)$", "$1$2es");
          // Check exception special case words
          rules.put("(\\w*)cactus$", "$1cacti");
          rules.put("(\\w*)focus$", "$1foci");
          rules.put("(\\w*)fungus$", "$1fungi");
          rules.put("(\\w*)octopus$", "$1octopi");
          rules.put("(\\w*)radius$", "$1radii");
          // If nothing else matches, and word ends in s, assume plural already
          rules.put("(\\w+)(s)$", "$1s");
          // Otherwise, just add s at the end in pluralize()
          return rules;
      }
  
      /**
       * Converts an English word to singular form. Leaves it alone if already singular.
       *
       * @param word an English word
       * @return the singularization of the argument English word, or the argument in case it is <code>null</code>
       */
     public static String singularize(String word)
     {
         if (word == null) return null;
 
         final Map<String, String> rules = EnglishInflector.getSingularRules();
         for (final Map.Entry<String, String> rule : rules.entrySet())
         {
             final String pattern = rule.getKey();
             final String replace = rule.getValue();
             if (word.matches(pattern))
             {
                 return word.replaceFirst(pattern, replace);
             }
         }
         return word.replaceFirst("([\\w]+)s$", "$1");
     }
 
     /**
      * Returns map of singular patterns
      */
     private static Map<String, String> getSingularRules()
     {
         final Map<String, String> rules = new LinkedHashMap<String, String>();
         rules.put("(\\w*)people$", "$1person");
         rules.put("(\\w*)children$", "$1child");
         rules.put("(\\w*)series$", "$1series");
         rules.put("(\\w*)feet$", "$1foot");
         rules.put("(\\w*)teeth$", "$1tooth");
         rules.put("(\\w*)buses$", "$1bus");
         rules.put("(\\w*)men$", "$1man");
         rules.put("(\\w*)person$", "$1person");
         rules.put("(\\w*)child$", "$1child");
         rules.put("(\\w*)foot$", "$1foot");
         rules.put("(\\w*)tooth$", "$1tooth");
         rules.put("(\\w*)bus$", "$1bus");
         rules.put("(\\w*)man$", "$1man");
         rules.put("(\\w+)(sis)$", "$1sis");
         rules.put("(\\w+)([ll])f", "$1$2f");
         rules.put("(\\w+)([^l])fe", "$1$2fe");
         rules.put("(\\w+)(ses)$", "$1sis");
         rules.put("(\\w+)([ll])ves", "$1$2f");
         rules.put("(\\w+)([^l])ves", "$1$2fe");
         rules.put("(\\w+)([^aeiou])y", "$1$2y");
         rules.put("(\\w+)([^aeiou])ies", "$1$2y");
         rules.put("(\\w+)(z|ch|ss|sh|x)es$", "$1$2");
         return rules;
     }
 }