Merge "Remove meaningless default action name"
[lhc/web/wiklou.git] / languages / classes / LanguageYue.php
index f00ac31..3293cc6 100644 (file)
@@ -1,26 +1,66 @@
 <?php
 /**
+ * Cantonese (粵語) specific code.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License along
+ * with this program; if not, write to the Free Software Foundation, Inc.,
+ * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
+ * http://www.gnu.org/copyleft/gpl.html
+ *
+ * @file
+ * @ingroup Language
+ */
+
+/**
+ * Cantonese (粵語)
+ *
  * @ingroup Language
  */
 class LanguageYue extends Language {
-       function stripForSearch( $string, $doStrip = true ) {
+
+       /**
+        * @return bool
+        */
+       function hasWordBreaks() {
+               return false;
+       }
+
+       /**
+        * Eventually this should be a word segmentation;
+        * for now just treat each character as a word.
+        * @todo FIXME: Only do this for Han characters...
+        *
+        * @param string $string
+        * @return string
+        */
+       function segmentByWord( $string ) {
+               $reg = "/([\\xc0-\\xff][\\x80-\\xbf]*)/";
+               $s = self::insertSpace( $string, $reg );
+               return $s;
+       }
+
+       /**
+        * @param string $string
+        * @return string
+        */
+       function normalizeForSearch( $string ) {
                wfProfileIn( __METHOD__ );
 
                // Double-width roman characters
                $s = self::convertDoubleWidth( $string );
-
-               if ( $doStrip == true ) {
-                       // eventually this should be a word segmentation;
-                       // for now just treat each character as a word.
-                       // @todo Fixme: only do this for Han characters...
-                       $reg = "/([\\xc0-\\xff][\\x80-\\xbf]*)/";
-                       $s = self::wordSegmentation( $s, $reg );
-               }
-
                $s = trim( $s );
+               $s = parent::normalizeForSearch( $s );
 
-               // Do general case folding and UTF-8 armoring
-               $s = parent::stripForSearch( $s, $doStrip );
                wfProfileOut( __METHOD__ );
                return $s;
        }