java - UTF-8 works on Eclipse but fails to work with an exported jar

Question

I am working on a desktop application with Java Swing and save data on a MySQL database in arabic language with using UTF-8.

When I run the application from Eclipse everything work well but when I finish and I export my work to a runnable jar using Eclipse export, nothing related to the database works.

• Login doesn't work
• When I try to save date to my database in Arabic I get ?????? in the database

However, as I mentioned earlier, everything work correctly when I run it from Eclipse. Can any one help me, I must deliver my work

This is a sample of my work:

This is how I connect connect my database :

static Connection conn = null;
static String url      = "jdbc:mysql://localhost:3306/";
static String dbName   = "gestiondestock";
static String driver   = "com.mysql.jdbc.Driver";
static String userName = "root"; 
static String password = "";
static String unicode= "?useUnicode=yes&characterEncoding=UTF-8";

This is the code for buttonLogin ActionPerformed:

private void buttonLogin_ActionPerformed(ActionEvent e) {
    if(textField.getText().equals("") || passwordField.getText().equals(""))
    {
       jd =new JDialog();
        jd.applyComponentOrientation(ComponentOrientation.RIGHT_TO_LEFT);
        jd.setTitle("?????? ??? ??????");
        jd.setVisible(true);
        jd.setLocationRelativeTo(null);
        jd.setSize(400,200);
        jd.setContentPane(buildpp());   
    }
    else{
        if(textField.getText().equals("Adel91") || passwordField.getText().equals("Adel91"))
        {
            CardLayout cardLayout = (CardLayout) contentPane.getLayout();
            cardLayout.show(contentPane, "Panel_Home");
        }
        else{
            try{
                String usernamena = new String(textField.getText());
                String passwordlogin = new String(passwordField.getText());

                MessageDigest mdEnc4 = MessageDigest.getInstance("MD5");

                mdEnc4.update(passwordlogin.getBytes(), 0, passwordlogin.length());
                String passwordlogindmd5 = new BigInteger(1, mdEnc4.digest()).toString(16); // Encrypted 

                try{
                    Class.forName(driver).newInstance();
                    conn = DriverManager.getConnection(url+dbName+unicode,userName,password);
                    Statement st = conn.createStatement();

                    ResultSet res = st.executeQuery("SELECT username,password FROM client ");

                    String user = null;
                    String pass = null;

                    if(res.next()) {
                        user = new String( res.getBytes(1), "UTF-8");
                        pass =  new String( res.getBytes(2), "UTF-8");
                    }

                    if(usernamena.equals(user)&&passwordlogindmd5.equals(pass)){
                        CardLayout cardLayout = (CardLayout) contentPane.getLayout();
                        cardLayout.show(contentPane, "Panel_Home");
                    }
                    else{
                        jd =new JDialog();
                        jd.applyComponentOrientation(ComponentOrientation.RIGHT_TO_LEFT);
                        jd.setTitle("???? ?????? ?????? ??? ???????");
                        jd.setVisible(true);
                        jd.setLocationRelativeTo(null);
                        jd.setSize(430,200);
                        jd.setContentPane(buildwronglogin());
                    }
                } catch (Exception ee) {
                    ee.printStackTrace();
                }
            } catch (NoSuchAlgorithmException e1) {
                // TODO Auto-generated catch block
                e1.printStackTrace();
            } 
        }
    }   
}

Resulting MySQL variables using this cmd :SHOW VARIABLES LIKE 'c%'; abd every thins is uts8

Answer 1

First, you should never convert a String to bytes using an unspecified charset (here you are using the platform default, which could change):

passwordlogin.getBytes()

Since this is for the bytes you are going to hash, which charset you use doesn't really matter as long as it is consistent. something like passwordlogin.getBytes("UTF-8") would seem reasonable.

Also, this is certainly a problematic bit of code:

                if(res.next()) {
                    user = new String( res.getBytes(1), "UTF-8");
                    pass =  new String( res.getBytes(2), "UTF-8");
                }

it seems to imply that you have a characterset breakdown somewhere.

Lastly, how are you running your jar? Are you specifying a charset on the command line (e.g. java -Dfile.encoding="UTF-8" ...).