hi there,

i have a problem that is not directly related to hibernate.
only in such a way that i use hibernate and that there maybe
a way to make a workaround in hibernate.

i want to use unicode in my app and in my database. which
does work to a degree. what happens when i write to
my db (mysql 4.1, postgres 7.4) is that the data is being
encoded twice. this is not so bad at all as when i read the
data from the db the decoding seems to happen two
times too.
for a more in depth description of what happens see here:
http://lists.mysql.com/mysql/166280.
this happens regardless of the way i write into the db.
i can use hibernate or i can use a direct jdbc approach. see
the test-app below that i wrote in order to check if this is
hibernate problem or not.
i also wrote a python script which does excatly the same but
which does not exhibit this error - meaning if i look into the
db via the mysql/psql console all data is displayed correctly
(given that my os-console is configured correctly and can
display non-latin chars).

well, the usuall advice is
a) check your DB settings!
belive me they are all set to UTF-8 (mysql as well as
postgress)
b) check your connection url!
this i did too. the interessting part is: it does not matter.
...?useUnicode=True&characterEncoding=UTF-8 result
in the same output as leaving them out. the same
applies for the postgres url (encoding=UTF-8 or
charSet=UTF-8)
3) check your os-settings!
i did try different ones (de_DE.uf8, de_DE, de_DE@euro
on linux and some others on a windows machine). no
difference.

as long as i solely write and read from my java-app.
everything is ok as my output and my input are at least
consistent. it only gets bad if others (python or so try
to read what i have writen). if i want to read data written
by python via jdbc i also get garbage. but there is a work
around. if i use
String s = new String(a_result_set.getBytes());
instead of
String s = a_result_set.getString();
i can read others data perfectly well.

ok so i know that this is no problem of hibernate. never
theless: is there a way to workaround this via hibernate?
could i for example define that strings should be written and
read to the db via setBytes/getBytes?


ciao roberj
ps: here the test app with direct jdbc access. i hope
the cyrilic characters are being display correctly on the
forum.

[code]package org.pragmatico.ctpe.test.service.umlaut;

import java.sql.Connection;
import java.sql.DriverManager;
import java.sql.PreparedStatement;
import java.sql.ResultSet;
import java.sql.SQLException;
import java.sql.Statement;
import java.util.Properties;

import junit.framework.TestCase;
import junit.framework.Test;
import junit.framework.TestSuite;

public class TestUTF8 extends TestCase
{
public TestUTF8(String _name)
{
super(_name);
}

public static Test suite()
{
TestSuite suite;
suite = new TestSuite(TestUTF8.class);

suite = new TestSuite();
suite.addTest(new TestUTF8("testPostgres"));

return suite;
}

public void testPostgres()
{
String url = "jdbc:postgresql://192.168.1.110:5432/test_db?encoding=UTF-8";

String cls = "org.postgresql.Driver";
String user = "";
String pwd = "";
createAndRead(url, user, pwd, cls, "POSTGRES");
}

public void createAndRead(String _url, String _user, String _pwd, String _class, String _dbtype)
{
Connection con = null;
Statement stmt;

try
{
Class.forName(_class);
Properties props = new Properties();
props.put("user", _user);
props.put("password", _pwd);
//props.put("charSet", "UTF-8");
con = DriverManager.getConnection(_url, props);

stmt = con.createStatement();

try
{
stmt.executeUpdate("drop schema UTF8_TEST cascade");
}
catch (Exception ex)
{
ex.printStackTrace();
}
stmt.executeUpdate("create schema UTF8_TEST");

String create_stmt = "create table UTF8_TEST.UTF8_TEST" +
"( ID CHAR (02)," +
" FAMILY_NAME CHAR (32)," +
" GIVEN_NAME CHAR (32)) ";

stmt.executeUpdate(create_stmt);

stmt.executeUpdate("insert into UTF8_TEST.UTF8_TEST (ID, FAMILY_NAME, GIVEN_NAME) "
+ "values ('01', 'K