AJP12 InputStream has some problems in handing string.
It makes every byte buffer into UTF-8 String.
But in Korean( I lives in Korea.!) lanugage environment!,
Query CAN String has non-escaped EUC-KR String.
EUC-KR byte sequences are not converted into UTF-8 encoding.
So your creating string encoding must be changed to "8859-1"..
Um every 8859-1 can be another Locale's encoding.
Using the following!!
String myEncoding = "EUC-JP"; // if you are a Japanese.
Stirng value = request.getParameter("param-name");
String myvalue = new String(value.getBytes("ISO8859-1", myEncoding);
Above, simple string conversion make it correct!!
But,
class Ajp12InputStream in Ajp12ConnectionHandler.java handles
above byte sequences as UTF-8.
Um I guess CJK encoded byte sequences failed to converting UTF-8 Directly..
Please correct it before releasing 3.2.2 tomcat.
