Mapping Java Objects to a Database with Castor-JDO

As a Java-programmer you’ll probably find it easy to work with objects. To create an instance of a class one can simply use “new” and maybe a couple of setter-methods. To save your instances, so you can continue working with them tomorrow might be trickier. One solution would be to serialize your objects and store them in XML format in a flat file. If you’ve got many objects with complex relations then you’ll soon find out that this is not the way to go. If you’ve got a database system and you know your SQL, then you have a solution that works. On the other hand: what if you could simply say “save my objects”, and they’d be in your data base? That’d be nice, right?

If you continue reading I’ll disrobe the tool that will enable the “save my objects” slight of hand. To save an instance of class “Media” to a database you’d have to code:

      db.begin();
      Media m = new Media(1,"DVD");
      db.create(m);
      db.commit();
      db.close();

Harry Potter would love this spell and the tool that can make this come true is Castor-JDO. 

JDO – the history

JDO spelled out means Java Data Objects, and Sun has an API and a specification (JSR12) with this name. This API addresses how to persist (store permanently) Java objects on a back-end, which could be a mainframe, a database or something else. Castor-JDO, which was released in December 1999, before Sun’s API, has a more focused approach. Its purpose is to automate binding Java objects to a relational database. Since Castor and Sun use the same name, JDO, for their APIs, there have been some misunderstandings and discussions about who has the rights to the name, and what comprises “true JDO”. The resources section of the article has some links where you can follow some of the arguments.

This article will focus only on Castor-JDO, which in my opinion is the simplest API to learn, and therefore a good way to be introduced to the object-relational mapping world. By the way: Castor-JDO has a twin brother, Castor-XML, which can serialize a Java bean object into XML, and back again. See the resources section for more on this useful feature. When I use “Castor” in this article I refer to Castor-JDO.   

So this is about how you automate the persistence of Java objects into a relational database. What we’d really like is to let someone else write the SQL and setup the JDBC API to manage the database, thereby allowing us to focus on handling the Java objects and the business logic.

If you’re like me, you’re excited about this possibility, then I think it’s fair to stress the following. You will need to know about database tables, columns, foreign keys etc. in order to set up Castor properly. The more you know about the good, the bad and the ugly of database design and SQL coding the better you will be able to take benefit from what Castor offers. Don’t be scared, in this article I’ll go step-by-step through the basics of setting up your application for Castor, and when it’s done and the train is on the rails, you’ll be halfway to Hogwart’s School of Wizardry before you know it!

Introduction
Installation of MySQL
Installation of Castor-JDO
Transient and Persistent Objects,The database configuration file,The mapping file
Handling one-to-many relations
Handling many-to-many relations
Reading from the database
The Source of Magic
Conclusion