// Connect to database using a org.apache.commons.dbcp.BasicDataSource
    BasicDataSource ds = new BasicDataSource();
    ds.setDriverClassName( "com.mysql.jdbc.Driver" );
    ds.setUsername( username );
    ds.setPassword( password );
    ds.setUrl( "jdbc:mysql://" + hostpath + ":" + port + "/" + database + "?" + dbparams );
    ds.setMaxActive( 10);

    /**
    * Create Data Manager, this example for creating a connection to an sql rdbms
    * There is also an option to use Amazon SimpleDB as a backend and more backends 
    * may be supported in the future
    **/
    SqlDataManager dm = new SqlDataManager( ds );
    dm.setLoggingLevel( org.apache.log4j.Level.DEBUG ); //Debug mode will log out the sql queries generated
  
    /**
    * Build schemas reads the database structure from the target database so that the library
    * can build the appropriate queries and add in foreign keys
    **/
    dm.buildSchemas();

    /**
    * The following settings described how data updates are dealt with in cases where the data we are 
    * trying to write matches a unique key for existing data.
    * Tble Summary of what each setting does
    *                                 UPDATE
    *                      false                        true
    *                 --------------------------------------------------------------
    *           false |  No updates are performed  | Only NOT NULL database values |
    *                 | (All queries INSERT IGNORE)| are updated                   |
    *  MERGE          |------------------------------------------------------------| 
    *           true  | Only NULL database values  | All existing data is updated, |
    *                 | are updated.               | but is never updated to NULL  |
    *                 --------------------------------------------------------------
    **/
    
    dm.setGlobalUpdateEnabled( true );
    dm.setGlobalMergeEnabled( true );

    /**
    * We are now ready to save data, for this example we will save a person and a couple of addresses
    * Related to them.  This assumes your database has a 'person' table and an 'address' table.
    *  table, column, value 
    **/
    dm.addData("person","name", "John Doe");

    // Or we can accept a Map of values
    Map data = new HashMap();
    data.put("address", "123 Main Street");
    data.put("city", "Townsvilleburgton");
    data.put("state", "WV");
    dm.addData("address", data);

    /**
    * Calling commit closes off the record for writing so the next time dm.addData is called
    * for the table it will correspond to a new database row.  This does not cause the actuall
    * write to happen yet, however.
    *//
    dm.commit("address");

    //Add another address for the same person, it is case insensitve for table/column names
    dm.addData("ADDRESS", "address", "457 Minor Dr.");
    dm.addDAta("address", "City", "Megacity");
    dm.addData("address", "STATE", "NC");
    dm.commit("address");

    /**
    * We now have a person and two addresses.  The library knows the structure of the database and 
    * that we have a foreign key relationship between 'person' and 'address' so it will connect them 
    * in the database.  For this to happen all of the children of a row need to be added before the parent
    * is committed.  Since we added the children we can now commit 'person'.
    **/
    dm.commit("person");
    
    /**
    * We have all of our data added, so lets write it out by calling flush().  If we wanted we could add     
    * multiple 'person' entries and write them all out at once, or we could write them out once we get a
    * complete record. 
    **/
    dm.flush(); 
    
    //Calling dm.close() closes the underlying BasicDataSource
    dm.close();