A POJO is just a plain, old Java Bean with the restrictions removed. Java Beans must meet the following requirements:
- Default no-arg constructor
- Follow the Bean convention of getFoo (or isFoo for booleans) and setFoo methods for a mutable attribute named foo; leave off the setFoo if foo is immutable.
- Must implement java.io.Serializable
POJO does not mandate any of these. It's just what the name says: an object that compiles under JDK can be considered a Plain Old Java Object. No app server, no base classes, no interfaces required to use.
The acronym POJO was a reaction against EJB 2.0, which required several interfaces, extended base classes, and lots of methods just to do simple things. Some people, Rod Johnson and Martin Fowler among them, rebelled against the complexity and sought a way to implement enterprise scale solutions without having to write EJBs.
Martin Fowler coined a new acronym.
Rod Johnson wrote "J2EE Without EJBs", wrote Spring, influenced EJB enough so version 3.1 looks a great deal like Spring and Hibernate, and got a sweet IPO from VMWare out of it.
Here's an example that you can wrap your head around:
public class MyFirstPojo
{
private String name;
public static void main(String [] args)
{
for (String arg : args)
{
MyFirstPojo pojo = new MyFirstPojo(arg); // Here's how you create a POJO
System.out.println(pojo);
}
}
public MyFirstPojo(String name)
{
this.name = name;
}
public String getName() { return this.name; }
public String toString() { return this.name; }
}
与恶龙缠斗过久,自身亦成为恶龙;凝视深渊过久,深渊将回以凝视…