Updates for Japanese romanization support FS#1363

This patch adds some fixes for the romanization lookup table and a test case for more
than 20000 phrases and their correct romanization. About 2100 tests currently fail.

darcs-hash:20080506203707-7ad00-9d95b8af459fa44c8d3e95560c7e1c116b8ffc48.gz

1 files changed, 24 insertions, 0 deletions

diff --git a/_test/cases/inc/utf8_romanize.test.php b/_test/cases/inc/utf8_romanize.test.php
new file mode 100644
index 000000000..e1c1be7c2
--- /dev/null
+++ b/_test/cases/inc/utf8_romanize.test.php
@@ -0,0 +1,24 @@
+<?php
+// use no mbstring help here
+if(!defined('UTF8_NOMBSTRING')) define('UTF8_NOMBSTRING',1);
+require_once DOKU_INC.'inc/utf8.php';
+
+class utf8_substr_test extends UnitTestCase {
+
+    /**
+     * Check Japanese romanization
+     *
+     * @author Denis Scheither <amorphis@uni-bremen.de>
+     */
+    function test_japanese(){
+        $tests = file(dirname(__FILE__).'/utf8_kanaromaji.txt');
+        foreach($tests as $test){
+            list($jap,$rom) = explode(';',trim($test));
+
+            $chk = utf8_romanize($jap);
+            #if($chk != $rom) echo "$jap\t->\t$chk\t!=\t$rom\n";
+            $this->assertEqual($chk,$rom);
+        }
+    }
+}
+//Setup VIM: ex: et ts=4 enc=utf-8 :