I have a regular class, implements serializable, private variables with
getter/setter.

Nothing prints - despite being populated - ahhh driving me nuts

The plot thickens when I use hardcoded strings to set the properties
i.e. instead of cp.getUserName() I used "username"

If I put println in the method it proves there is data...

public void saveConnection(ConnProperty cp)
ConnProperty conn = new ConnProperty();
conn.setUsername(cp.getUserName());
conn.setDriverClass(cp.getDriver());
conn.setPassword(cp.getPassword);
conn.setUrl(cp.getUrl());
ConnectionList.setConfigFileName("test.xml");
ConnectionList list =new ConnectionList();
list.add(conn);
SerializeUtility.serializeXML(new File(filename),
list.elements());

public static void serializeXML(File file, Enumeration e1) throws
IOException {
XMLEncoder encoder = null;
try
{
encoder = new XMLEncoder(
new BufferedOutputStream(
new FileOutputStream(file)));
while (e1.hasMoreElements()) {
encoder.writeObject(e1.nextElement());
}
} catch (Exception ex) {
Debug.LogException("SerializeUtility", ex);
}
finally
{
if (encoder != null) {
encoder.close();
}
}
}

package com.osframework.datalibrary.common;

import java.io.Serializable;

public class ConnProperty implements Serializable {

private static final long serialVersionUID = 0L;
/**
* False for MS Access, True for almost all other databases.
*/
private boolean connectToURL = true;

/**
* URL to connect to remote database (or MS Access local file)
* e.g. jdbcracle:thin:@localhost:1521rod1
*/
private String url = "jdbcdbcRIVER={Microsoft Access Driver
(*.mdb)};DBQ=osrmt.mdb";

/**
* Driver implementation - class name - must be in the classpath!
*/
private String driverClass = "sun.jdbc.odbc.JdbcOdbcDriver";

/**
* Database connections of the same environment name
* will access URLs in this order
*/
private int accessSequence = 0;

private String username = "admin";

private String password = "";

public int getAccessSequence() {
return accessSequence;
}

public void setAccessSequence(int accessSequence) {
this.accessSequence = accessSequence;
}

public boolean isConnectToURL() {
return connectToURL;
}

public void setConnectToURL(boolean connectToURL) {
this.connectToURL = connectToURL;
}

public String getDriverClass() {
return driverClass;
}

public void setDriverClass(String driverClass) {
this.driverClass = driverClass;
}


public String getPassword() {
return password;
}

public void setPassword(String password) {
this.password = password;
}

public String getUrl() {
return url;
}

public void setUrl(String url) {
this.url = url;
}

public String getUsername() {
return username;
}

public void setUsername(String username) {
this.username = username;
}

}

chris riesbeck

Not sure when you say "nothing prints" if you mean the file is empty, or
has an empty XML element. The former implies I/O error -- any errors in
your log? The latter can be a number of things.

But here are some points to consider.

First, this isn't real test code since it calls getDriver() instead of
getDriverClass(). Always post the real code you ran.

Second, I don't think you can write more than one object. The output is
a complete XML root and only one root is allowed per file. If you want
the file to contain a list of ConnProperty objects, just write the list
itself. The Encoder will dump the list in XML and all it's parts. Very
neat and simple. No loop on your part.

Third, why are you copying the object before serializing? Just curious.

Fourth, don't forget that if your object only has default data, nothing


default constructor can take care of. So a default object is not a good
test case.

Hope this helps. timasmith@hotmail.com wrote:

ooooohhhhhh mannnn

I've been banging my head against the wall for 3 hours - and all the
time it doesnt write it because my private variables are initialized to
that value????

Well I say that is absurd ... so I persist the object to XML, later
change the default value of my *private* variables - and later
XML-->Object = wrong values

icky

Obviously the workaround is initiliaze nothing but integers and
boolean.

anyway - THANKS!

chris riesbeck

Design trade-off. The API says it uses a "redundancy elimination
algorithm," which is summarized in 2 sentences at

http://java.sun.com/products/jfc/tsc...stence4/#intro

The goal is to reduce how much is read and written.

There are lots of ways changing your class could break things later on
when decoding, so they may have felt it was a lost cause trying to
prevent that.