Bean introspection in Java

Question

I'm learning for a job interview in Java. They told me to learn the concepts of Beans introspection, so I searched the web and read in a couple of sites including the next posts:

• Java introspection and reflection
• Places where JavaBeans are used?
• What is a Java Bean exactly?
• Java doc - Reflection
• Introspection in Java

As far as I understood: Bean is like any other object class in Java, but this class must have the next features:

1. All properties private (use getters/setters)
2. A public no-argument constructor
3. Implements Serializable.

General things:

4. Introspection is giving me the possibility to "examine" an object during run-time, and that way I can get the class properties names, methods names constructors etc.
5. Introspection uses Reflection to get the Information of a class.

I still have some questions:

1. Why do I need this kind of a mechanism, meaning, in which cases should I use introspection instead of using any other thing?
2. Is there any difference between bean's introspection and a regular introspection?
3. How it's working besides the methods I can use?

I would be happy if someone could give me his own prospective about this subject, or to give me some kind of a link for useful information.

Answer 1

I'm not sure what they want to know.
You can use frameworks such as apache-bean-utils to query information about bean structure.
I developed such a code manually (big mistake! :) ) -
I used a recursive mechanism based on java bean notation (i.e - setters must begin with "set",
getters begin with "is" for boolean or "get" for all types)
You then can us this code to automate some behavior -
At my case fo example I wrote a tool that parses WSDL, and creates binding between WS calls and our application entitites via code.
The user of our application provided an XML indicating how to perform a mapping -
i.e - let's say that a WS call returned a Person object, but in our application we had a student entity
so the XML defined how to perform the mapping, and I used code like apache-bean-utils to perform introspection
and to understand what setters and getters to invoke.
This was done in contrast to what is done usually in java applications:
1. Generate Java clients (i.e - use wsdl2java) from WSDL
2. Compile the application with the client code.

I can assume introspection can be used in profilers code - for example,
Since there are many frameworks that use getters and setters , it is very improtant that these methods will be efficient,
so it's something that mabye profiles should first look into.

Feel free to add more questions