[Summary view] [Print] [Text view]

   1  <?php
   2  
   3  /**
   4   * @package     Joomla.Administrator
   5   * @subpackage  com_finder
   6   *
   7   * @copyright   (C) 2018 Open Source Matters, Inc. <https://www.joomla.org>
   8   * @license     GNU General Public License version 2 or later; see LICENSE.txt
   9   */
  10  
  11  namespace Joomla\Component\Finder\Administrator\Indexer\Language;
  12  
  13  use Joomla\Component\Finder\Administrator\Indexer\Language;
  14  use Joomla\String\StringHelper;
  15  
  16  // phpcs:disable PSR1.Files.SideEffects
  17  \defined('_JEXEC') or die;
  18  // phpcs:enable PSR1.Files.SideEffects
  19  
  20  /**
  21   * Chinese (simplified) language support class for the Finder indexer package.
  22   *
  23   * @since  4.0.0
  24   */
  25  class Zh extends Language
  26  {
  27      /**
  28       * Language locale of the class
  29       *
  30       * @var    string
  31       * @since  4.0.0
  32       */
  33      public $language = 'zh';
  34  
  35      /**
  36       * Spacer between terms
  37       *
  38       * @var    string
  39       * @since  4.0.0
  40       */
  41      public $spacer = '';
  42  
  43      /**
  44       * Method to construct the language object.
  45       *
  46       * @since   4.0.0
  47       */
  48      public function __construct($locale = null)
  49      {
  50          // Override parent constructor since we don't need to load an external stemmer
  51      }
  52  
  53      /**
  54       * Method to tokenise a text string.
  55       *
  56       * @param   string  $input  The input to tokenise.
  57       *
  58       * @return  array  An array of term strings.
  59       *
  60       * @since   4.0.0
  61       */
  62      public function tokenise($input)
  63      {
  64          $terms = parent::tokenise($input);
  65  
  66          // Iterate through the terms and test if they contain Chinese.
  67          for ($i = 0, $n = count($terms); $i < $n; $i++) {
  68              $charMatches = array();
  69              $charCount = preg_match_all('#[\p{Han}]#mui', $terms[$i], $charMatches);
  70  
  71              // Split apart any groups of Chinese characters.
  72              for ($j = 0; $j < $charCount; $j++) {
  73                  $tSplit = StringHelper::str_ireplace($charMatches[0][$j], '', $terms[$i], false);
  74  
  75                  if (!empty($tSplit)) {
  76                      $terms[$i] = $tSplit;
  77                  } else {
  78                      unset($terms[$i]);
  79                  }
  80  
  81                  $terms[] = $charMatches[0][$j];
  82              }
  83          }
  84  
  85          return $terms;
  86      }
  87  }