 |
Chapter 7 |  |
|
Chapter 7 | |
Contents:
Object Serialization Basics
Writing Classes to Work with Serialization
Versioning of Classes
The object serialization mechanism in Java 1.1 provides a way for objects to be written as a stream of bytes and then later recreated from that stream of bytes. This facility supports a variety of interesting applications. For example, object serialization provides persistent storage for objects, whereby objects are stored in a file for later use. Also, a copy of an object can be sent through a socket to another Java program. Object serialization forms the basis for the remote method invocation mechanism in Java that facilitates distributed programs. Object serialization is supported by a number of new classes in the java.io package in Java 1.1.
If a class is designed to work with object serialization, reading and writing instances of that class is quite simple. The process of writing an object to a byte stream is called serialization. For example, here is how you can write a Color object to a file:
FileOutputStream out = new FileOutputStream("tmp");
ObjectOutput objOut = new ObjectOutputStream(out);
objOut.writeObject(Color.red);
All you need to do is create an ObjectOutputStream around another output stream and then pass the object to be written to the writeObject() method. If you are writing objects to a socket or any other destination that is time-sensitive, you should call the flush() method after you are finished passing objects to the ObjectOutputStream.
The process of reading an object from byte stream is called deserialization. Here is how you can read that Color object from its file:
FileInputStream in = new FileInputStream("tmp");
ObjectInputStream objIn = new ObjectInputStream(in);
Color c = (Color)objIn.readObject();
Here all you need to do is create an ObjectInputStream object around another input stream and call its readObject() method.
|
 |
|
| File Manipulation |  |
Writing Classes to Work with Serialization |
