We Keep Coding

Why shouldn't i change the interface method names when implementing them?

Say I'm using a Runnable interface,
Runnable runnable = new Runnable() {
#Override
public void run() {
// some code
}
};
If the run () method in Runnable interface is empty , why should we use #Override and same name for the method in our code instead of just using a random name for our implementation like ,
Runnable runnable = new Runnable() {
public void myRun() {
// some code
}
};
The question might seem dumb but I'm a newbie to programming!

Runnable has only one method, thus, you may be tempted to think the method name is irrelevant. But most interfaces have more than one method. For example, java.util.Iterator which has 2 (well, 4, but 2 of those have default impls so you don't need to implement those. Point is, 2 that you must implement).
So, let's say that instead of naming your two methods next and hasNext, you name them jill and jane instead.
How is the compiler or any caller supposed to figure out which one performs the hasNext job and which one performs the next job? Look at the signatures? What if your interface has 2 different methods with the exact same signature?
Not possible. That's why the names need to match: They are used to look this up.
Yes, if the interface has just the one name you'd think the compiler/caller can just go: Yeah, well, the one and only method it has, that's the one. But that's not how it works - language design is simpler if you avoid making exceptions just for simplified cases.
Note that java has lambda syntax, which only works when an interface has one method, and which omit the method name entirely, as well as the argument types. Instead of your myRun method you could write this too:
Runnable r = () -> /* some code */;
No new Runnable() { public void myRun, and no }} required. If myRun had arguments, you don't even need to add their types, just their names. For example, Comparator, which has arguments:
long form:
Comparator<String> lengthComparator = new Comparator<String>() {
#Override public int compare(String a, String b) {
return a.length() - b.length();
}
};
can be written much shorter:
Comparator<String> lengthComparator = (a, b) -> a.length() - b.length();

Think about what it means to "implement an interface". If I have a variable of an interface type, I can call any method defined in the interface on that variable. For example, for the Runnable interface:
interface Runnable {
void run();
}
If I have a variable of type Runnable, I know I can call run on it:
Runnable r = ...;
r.run(); // I can guarantee this will work, because whatever "..." evaluates to, implements Runnable
So when you implement Runnable, you are really just saying that "this class has a run method". The Java compiler also checks that what you are saying is true, that the class actually has a run method. Saying false things about your class, intuitively, should not be allowed, right?
We can also think about what would happen if you were allowed to implement Runnable, but not provide a run method, then you would not be 100% sure that any variable of type Runnable has a run method.
class TrueRunnable implements Runnable {
#Override public void run() {}
}
class FakeRunnable implements Runnable { // suppose this is allowed
public void myRun() {}
}
...
Runnable r;
if (new Random().nextBoolean()) {
r = new TrueRunnable()
} else {
r = new FakeRunnable();
}
// does r have a run method? Maybe.
Then interfaces would not be very useful, because you are not 100% sure whether any Object has any method anyway.
Object r;
if (new Random().nextBoolean()) {
r = new TrueRunnable()
} else {
r = new FakeRunnable();
}
// does r have a run method? Maybe.
So in that case we would lose the whole point of using interfaces.

As to why an interface implementation method has to have the same name as the declaration method name in the interface , i think it boils down to allowing Java compiler to understand which method actually implements the interface. Otherwise its impossible for the Java compiler to know.
Another solution could have been to use a qualifier name in an annotation to signal which is the implementation method something like this.
public interface MyInterface {
void sayHello();
}
and
public class Hello implements MyInterface {
#Implements("MyInterface.sayHello")
public void someOtherName() {
print("Hello World);
}
}
This approach is not valid , but could have been an alternative if Java creators wanted to allow users to declare different names for interface implementation methods. I think this approach is uglier and prone to errors though.
Also this problem does not exist when using lambda expressions where the compiler can understand whats happening without the code writer having to specify the name of the interface method. Note that this only applies to functional interfaces.

How do I stop a functional interface from being a target for lambda expressions?

I have an API with an interface with two (overloaded) methods of the same name that take different argument types. These different types are both technically functional interfaces, but the user should not be allowed to create instances of one of them. Here is a simplified example:
public class Example
{
#FunctionalInterface
public interface Computation
{
int compute();
}
public interface WrappedComputation
{
Computation unwrap();
}
public static class Solver
{
public static int solve(Computation a)
{
return a.compute();
}
public static int solve(WrappedComputation b)
{
return solve(b.unwrap());
}
}
public static void main(String... args)
{
// 'Computation' interface should be a lambda target
// so coder can make their own 'A' computation
Solver.solve( () -> { return 5 + 5; } );
// 'WrappedComputation' interface SHOULD NOT be a lambda target
// or else coder can cause runtime exceptions etc., like passing a null 'Computation' reference that will be computed
Solver.solve( () -> { Computation a = null; return a; } );
}
}
The only idea I have so far is to add a dummy/unused method to the interface that I don't want to be a lambda target, and implement it in all of the implementing classes. That seems a little sloppy/unneeded though... any other suggestions?

There is no way to prevent the usage of lambda expressions by API clients when your interface satisfies the criteria for it.
Your supposition that API clients can only cause trouble via lambdas is incorrect -- any bad thing you can do with a lambda, you can also do with an anonymous inner class or a named class. Consider:
Solver.solve( new WrappedComputation(){
public Computation unwrap(){
Computation a = null; return a;
}
});
This has exactly the same semantics and end result (a runtime exception) as your lambda.
Perhaps what you want instead is to prevent any uncontrolled creation of WrappedComputation? In that case, consider making it a final class whose constructor checks for error cases:
public final class WrappedComputation{
private final Computation _wrapped;
public WrappedComputation(#NonNull wrapped){
_wrapped = Objects.requireNonNull(wrapped);
}
}

Coder can always cause runtime exception, so that's a bad excuse.
If you don't want ambiguity, don't overload the method.
Name it something else, e.g. solveWrapped.

Why is this variable's type its own class? [duplicate]

What is the use of anonymous classes in Java? Can we say that usage of anonymous class is one of the advantages of Java?

By an "anonymous class", I take it you mean anonymous inner class.
An anonymous inner class can come useful when making an instance of an object with certain "extras" such as overriding methods, without having to actually subclass a class.
I tend to use it as a shortcut for attaching an event listener:
button.addActionListener(new ActionListener() {
#Override
public void actionPerformed(ActionEvent e) {
// do something
}
});
Using this method makes coding a little bit quicker, as I don't need to make an extra class that implements ActionListener -- I can just instantiate an anonymous inner class without actually making a separate class.
I only use this technique for "quick and dirty" tasks where making an entire class feels unnecessary. Having multiple anonymous inner classes that do exactly the same thing should be refactored to an actual class, be it an inner class or a separate class.

Anonymous inner classes are effectively closures, so they can be used to emulate lambda expressions or "delegates". For example, take this interface:
public interface F<A, B> {
B f(A a);
}
You can use this anonymously to create a first-class function in Java. Let's say you have the following method that returns the first number larger than i in the given list, or i if no number is larger:
public static int larger(final List<Integer> ns, final int i) {
for (Integer n : ns)
if (n > i)
return n;
return i;
}
And then you have another method that returns the first number smaller than i in the given list, or i if no number is smaller:
public static int smaller(final List<Integer> ns, final int i) {
for (Integer n : ns)
if (n < i)
return n;
return i;
}
These methods are almost identical. Using the first-class function type F, we can rewrite these into one method as follows:
public static <T> T firstMatch(final List<T> ts, final F<T, Boolean> f, T z) {
for (T t : ts)
if (f.f(t))
return t;
return z;
}
You can use an anonymous class to use the firstMatch method:
F<Integer, Boolean> greaterThanTen = new F<Integer, Boolean> {
Boolean f(final Integer n) {
return n > 10;
}
};
int moreThanMyFingersCanCount = firstMatch(xs, greaterThanTen, x);
This is a really contrived example, but its easy to see that being able to pass functions around as if they were values is a pretty useful feature. See "Can Your Programming Language Do This" by Joel himself.
A nice library for programming Java in this style: Functional Java.

Anonymous inner class is used in following scenario:
1.) For Overriding(subclassing), when class definition is not usable except current case:
class A{
public void methodA() {
System.out.println("methodA");
}
}
class B{
A a = new A() {
public void methodA() {
System.out.println("anonymous methodA");
}
};
}
2.) For implementing an interface, when implementation of interface is required only for current case:
interface InterfaceA{
public void methodA();
}
class B{
InterfaceA a = new InterfaceA() {
public void methodA() {
System.out.println("anonymous methodA implementer");
}
};
}
3.) Argument Defined Anonymous inner class:
interface Foo {
void methodFoo();
}
class B{
void do(Foo f) { }
}
class A{
void methodA() {
B b = new B();
b.do(new Foo() {
public void methodFoo() {
System.out.println("methodFoo");
}
});
}
}

I use them sometimes as a syntax hack for Map instantiation:
Map map = new HashMap() {{
put("key", "value");
}};
vs
Map map = new HashMap();
map.put("key", "value");
It saves some redundancy when doing a lot of put statements. However, I have also run into problems doing this when the outer class needs to be serialized via remoting.

They're commonly used as a verbose form of callback.
I suppose you could say they're an advantage compared to not having them, and having to create a named class every time, but similar concepts are implemented much better in other languages (as closures or blocks)
Here's a swing example
myButton.addActionListener(new ActionListener(){
public void actionPerformed(ActionEvent e) {
// do stuff here...
}
});
Although it's still messily verbose, it's a lot better than forcing you to define a named class for every throw away listener like this (although depending on the situation and reuse, that may still be the better approach)

You use it in situations where you need to create a class for a specific purpose inside another function, e.g., as a listener, as a runnable (to spawn a thread), etc.
The idea is that you call them from inside the code of a function so you never refer to them elsewhere, so you don't need to name them. The compiler just enumerates them.
They are essentially syntactic sugar, and should generally be moved elsewhere as they grow bigger.
I'm not sure if it is one of the advantages of Java, though if you do use them (and we all frequently use them, unfortunately), then you could argue that they are one.

GuideLines for Anonymous Class.
Anonymous class is declared and initialized simultaneously.
Anonymous class must extend or implement to one and only one class or interface resp.
As anonymouse class has no name, it can be used only once.
eg:
button.addActionListener(new ActionListener(){
public void actionPerformed(ActionEvent arg0) {
// TODO Auto-generated method stub
}
});

Yes, anonymous inner classes is definitely one of the advantages of Java.
With an anonymous inner class you have access to final and member variables of the surrounding class, and that comes in handy in listeners etc.
But a major advantage is that the inner class code, which is (at least should be) tightly coupled to the surrounding class/method/block, has a specific context (the surrounding class, method, and block).

new Thread() {
public void run() {
try {
Thread.sleep(300);
} catch (InterruptedException e) {
System.out.println("Exception message: " + e.getMessage());
System.out.println("Exception cause: " + e.getCause());
}
}
}.start();
This is also one of the example for anonymous inner type using thread

An inner class is associated with an instance of the outer class and there are two special kinds: Local class and Anonymous class. An anonymous class enables us to declare and instantiate a class at same time, hence makes the code concise. We use them when we need a local class only once as they don't have a name.
Consider the example from doc where we have a Person class:
public class Person {
public enum Sex {
MALE, FEMALE
}
String name;
LocalDate birthday;
Sex gender;
String emailAddress;
public int getAge() {
// ...
}
public void printPerson() {
// ...
}
}
and we have a method to print members that match search criteria as:
public static void printPersons(
List<Person> roster, CheckPerson tester) {
for (Person p : roster) {
if (tester.test(p)) {
p.printPerson();
}
}
}
where CheckPerson is an interface like:
interface CheckPerson {
boolean test(Person p);
}
Now we can make use of anonymous class which implements this interface to specify search criteria as:
printPersons(
roster,
new CheckPerson() {
public boolean test(Person p) {
return p.getGender() == Person.Sex.MALE
&& p.getAge() >= 18
&& p.getAge() <= 25;
}
}
);
Here the interface is very simple and the syntax of anonymous class seems unwieldy and unclear.
Java 8 has introduced a term Functional Interface which is an interface with only one abstract method, hence we can say CheckPerson is a functional interface. We can make use of Lambda Expression which allows us to pass the function as method argument as:
printPersons(
roster,
(Person p) -> p.getGender() == Person.Sex.MALE
&& p.getAge() >= 18
&& p.getAge() <= 25
);
We can use a standard functional interface Predicate in place of the interface CheckPerson, which will further reduce the amount of code required.

i use anonymous objects for calling new Threads..
new Thread(new Runnable() {
public void run() {
// you code
}
}).start();

Anonymous inner class can be beneficial while giving different implementations for different objects. But should be used very sparingly as it creates problem for program readability.

One of the major usage of anonymous classes in class-finalization which called finalizer guardian. In Java world using the finalize methods should be avoided until you really need them. You have to remember, when you override the finalize method for sub-classes, you should always invoke super.finalize() as well, because the finalize method of super class won't invoke automatically and you can have trouble with memory leaks.
so considering the fact mentioned above, you can just use the anonymous classes like:
public class HeavyClass{
private final Object finalizerGuardian = new Object() {
#Override
protected void finalize() throws Throwable{
//Finalize outer HeavyClass object
}
};
}
Using this technique you relieved yourself and your other developers to call super.finalize() on each sub-class of the HeavyClass which needs finalize method.

You can use anonymous class this way
TreeSet treeSetObj = new TreeSet(new Comparator()
{
public int compare(String i1,String i2)
{
return i2.compareTo(i1);
}
});

Seems nobody mentioned here but you can also use anonymous class to hold generic type argument (which normally lost due to type erasure):
public abstract class TypeHolder<T> {
private final Type type;
public TypeReference() {
// you may do do additional sanity checks here
final Type superClass = getClass().getGenericSuperclass();
this.type = ((ParameterizedType) superClass).getActualTypeArguments()[0];
}
public final Type getType() {
return this.type;
}
}
If you'll instantiate this class in anonymous way
TypeHolder<List<String>, Map<Ineger, Long>> holder =
new TypeHolder<List<String>, Map<Ineger, Long>>() {};
then such holder instance will contain non-erasured definition of passed type.
Usage
This is very handy for building validators/deserializators. Also you can instantiate generic type with reflection (so if you ever wanted to do new T() in parametrized type - you are welcome!).
Drawbacks/Limitations
You should pass generic parameter explicitly. Failing to do so will lead to type parameter loss
Each instantiation will cost you additional class to be generated by compiler which leads to classpath pollution/jar bloating

An Anonymous Inner Class is used to create an object that will never be referenced again. It has no name and is declared and created in the same statement.
This is used where you would normally use an object's variable. You replace the variable with the new keyword, a call to a constructor and the class definition inside { and }.
When writing a Threaded Program in Java, it would usually look like this
ThreadClass task = new ThreadClass();
Thread runner = new Thread(task);
runner.start();
The ThreadClass used here would be user defined. This class will implement the Runnable interface which is required for creating threads. In the ThreadClass the run() method (only method in Runnable) needs to be implemented as well.
It is clear that getting rid of ThreadClass would be more efficient and that's exactly why Anonymous Inner Classes exist.
Look at the following code
Thread runner = new Thread(new Runnable() {
public void run() {
//Thread does it's work here
}
});
runner.start();
This code replaces the reference made to task in the top most example. Rather than having a separate class, the Anonymous Inner Class inside the Thread() constructor returns an unnamed object that implements the Runnable interface and overrides the run() method. The method run() would include statements inside that do the work required by the thread.
Answering the question on whether Anonymous Inner Classes is one of the advantages of Java, I would have to say that I'm not quite sure as I am not familiar with many programming languages at the moment. But what I can say is it is definitely a quicker and easier method of coding.
References: Sams Teach Yourself Java in 21 Days Seventh Edition

The best way to optimize code. also, We can use for an overriding method of a class or interface.
import java.util.Scanner;
abstract class AnonymousInner {
abstract void sum();
}
class AnonymousInnerMain {
public static void main(String []k){
Scanner sn = new Scanner(System.in);
System.out.println("Enter two vlaues");
int a= Integer.parseInt(sn.nextLine());
int b= Integer.parseInt(sn.nextLine());
AnonymousInner ac = new AnonymousInner(){
void sum(){
int c= a+b;
System.out.println("Sum of two number is: "+c);
}
};
ac.sum();
}
}

One more advantage:
As you know that Java doesn't support multiple inheritance, so if you use "Thread" kinda class as anonymous class then the class still has one space left for any other class to extend.

Can we write a function in Java that takes another function signature as a parameter and executes it? [duplicate]

This may be something common and trivial, but I seem to be having trouble finding a concrete answer. In C# there is a concept of delegates, which relates strongly to the idea of function pointers from C++. Is there a similar functionality in Java? Given that pointers are somewhat absent, what is the best way about this? And to be clear, we're talking first class here.

The Java idiom for function-pointer-like functionality is an an anonymous class implementing an interface, e.g.
Collections.sort(list, new Comparator<MyClass>(){
public int compare(MyClass a, MyClass b)
{
// compare objects
}
});
Update: the above is necessary in Java versions prior to Java 8. Now we have much nicer alternatives, namely lambdas:
list.sort((a, b) -> a.isGreaterThan(b));
and method references:
list.sort(MyClass::isGreaterThan);

You can substitue a function pointer with an interface. Lets say you want to run through a collection and do something with each element.
public interface IFunction {
public void execute(Object o);
}
This is the interface we could pass to some say CollectionUtils2.doFunc(Collection c, IFunction f).
public static void doFunc(Collection c, IFunction f) {
for (Object o : c) {
f.execute(o);
}
}
As an example say we have a collection of numbers and you would like to add 1 to every element.
CollectionUtils2.doFunc(List numbers, new IFunction() {
public void execute(Object o) {
Integer anInt = (Integer) o;
anInt++;
}
});

You can use reflection to do it.
Pass as parameter the object and the method name (as a string) and then invoke the method. For example:
Object methodCaller(Object theObject, String methodName) {
return theObject.getClass().getMethod(methodName).invoke(theObject);
// Catch the exceptions
}
And then use it as in:
String theDescription = methodCaller(object1, "toString");
Class theClass = methodCaller(object2, "getClass");
Of course, check all exceptions and add the needed casts.

No, functions are not first class objects in java. You can do the same thing by implementing a handler class - this is how callbacks are implemented in the Swing etc.
There are however proposals for closures (the official name for what you're talking about) in future versions of java - Javaworld has an interesting article.

This brings to mind Steve Yegge's Execution in the Kingdom of Nouns. It basically states that Java needs an object for every action, and therefore does not have "verb-only" entities like function pointers.

To achieve similar functionality you could use anonymous inner classes.
If you were to define a interface Foo:
interface Foo {
Object myFunc(Object arg);
}
Create a method bar which will receive a 'function pointer' as an argument:
public void bar(Foo foo) {
// .....
Object object = foo.myFunc(argValue);
// .....
}
Finally call the method as follows:
bar(new Foo() {
public Object myFunc(Object arg) {
// Function code.
}
}

Java8 has introduced lambdas and method references. So if your function matches a functional interface (you can create your own) you can use a method reference in this case.
Java provides a set of common functional interfaces. whereas you could do the following:
public class Test {
public void test1(Integer i) {}
public void test2(Integer i) {}
public void consumer(Consumer<Integer> a) {
a.accept(10);
}
public void provideConsumer() {
consumer(this::test1); // method reference
consumer(x -> test2(x)); // lambda
}
}

There is no such thing in Java. You will need to wrap your function into some object and pass the reference to that object in order to pass the reference to the method on that object.
Syntactically, this can be eased to a certain extent by using anonymous classes defined in-place or anonymous classes defined as member variables of the class.
Example:
class MyComponent extends JPanel {
private JButton button;
public MyComponent() {
button = new JButton("click me");
button.addActionListener(buttonAction);
add(button);
}
private ActionListener buttonAction = new ActionListener() {
public void actionPerformed(ActionEvent e) {
// handle the event...
// note how the handler instance can access
// members of the surrounding class
button.setText("you clicked me");
}
}
}

I have implemented callback/delegate support in Java using reflection. Details and working source are available on my website.
How It Works
We have a principle class named Callback with a nested class named WithParms. The API which needs the callback will take a Callback object as a parameter and, if neccessary, create a Callback.WithParms as a method variable. Since a great many of the applications of this object will be recursive, this works very cleanly.
With performance still a high priority to me, I didn't want to be required to create a throwaway object array to hold the parameters for every invocation - after all in a large data structure there could be thousands of elements, and in a message processing scenario we could end up processing thousands of data structures a second.
In order to be threadsafe the parameter array needs to exist uniquely for each invocation of the API method, and for efficiency the same one should be used for every invocation of the callback; I needed a second object which would be cheap to create in order to bind the callback with a parameter array for invocation. But, in some scenarios, the invoker would already have a the parameter array for other reasons. For these two reasons, the parameter array did not belong in the Callback object. Also the choice of invocation (passing the parameters as an array or as individual objects) belongs in the hands of the API using the callback enabling it to use whichever invocation is best suited to it's inner workings.
The WithParms nested class, then, is optional and serves two purposes, it contains the parameter object array needed for the callback invocations, and it provides 10 overloaded invoke() methods (with from 1 to 10 parameters) which load the parameter array and then invoke the callback target.

Check the closures how they have been implemented in the lambdaj library. They actually have a behavior very similar to C# delegates:
http://code.google.com/p/lambdaj/wiki/Closures

Relative to most people here I am new to java but since I haven't seen a similar suggestion I have another alternative to suggest. Im not sure if its a good practice or not, or even suggested before and I just didn't get it. I just like it since I think its self descriptive.
/*Just to merge functions in a common name*/
public class CustomFunction{
public CustomFunction(){}
}
/*Actual functions*/
public class Function1 extends CustomFunction{
public Function1(){}
public void execute(){...something here...}
}
public class Function2 extends CustomFunction{
public Function2(){}
public void execute(){...something here...}
}
.....
/*in Main class*/
CustomFunction functionpointer = null;
then depending on the application, assign
functionpointer = new Function1();
functionpointer = new Function2();
etc.
and call by
functionpointer.execute();

Use of class definitions inside a method in Java

Example:
public class TestClass {
public static void main(String[] args) {
TestClass t = new TestClass();
}
private static void testMethod() {
abstract class TestMethod {
int a;
int b;
int c;
abstract void implementMe();
}
class DummyClass extends TestMethod {
void implementMe() {}
}
DummyClass dummy = new DummyClass();
}
}
I found out that the above piece of code is perfectly legal in Java. I have the following questions.
What is the use of ever having a class definition inside a method?
Will a class file be generated for DummyClass
It's hard for me to imagine this concept in an Object Oriented manner. Having a class definition inside a behavior. Probably can someone tell me with equivalent real world examples.
Abstract classes inside a method sounds a bit crazy to me. But no interfaces allowed. Is there any reason behind this?

This is called a local class.
2 is the easy one: yes, a class file will be generated.
1 and 3 are kind of the same question. You would use a local class where you never need to instantiate one or know about implementation details anywhere but in one method.
A typical use would be to create a throw-away implementation of some interface. For example you'll often see something like this:
//within some method
taskExecutor.execute( new Runnable() {
public void run() {
classWithMethodToFire.doSomething( parameter );
}
});
If you needed to create a bunch of these and do something with them, you might change this to
//within some method
class myFirstRunnableClass implements Runnable {
public void run() {
classWithMethodToFire.doSomething( parameter );
}
}
class mySecondRunnableClass implements Runnable {
public void run() {
classWithMethodToFire.doSomethingElse( parameter );
}
}
taskExecutor.execute(new myFirstRunnableClass());
taskExecutor.execute(new mySecondRunnableClass());
Regarding interfaces: I'm not sure if there's a technical issue that makes locally-defined interfaces a problem for the compiler, but even if there isn't, they wouldn't add any value. If a local class that implements a local interface were used outside the method, the interface would be meaningless. And if a local class was only going to be used inside the method, both the interface and the class would be implemented within that method, so the interface definition would be redundant.

Those are called local classes. You can find a detailed explanation and an example here. The example returns a specific implementation which we doesn't need to know about outside the method.

The class can't be seen (i.e. instantiated, its methods accessed without Reflection) from outside the method. Also, it can access the local variables defined in testMethod(), but before the class definition.
I actually thought: "No such file will be written." until I just tried it: Oh yes, such a file is created! It will be called something like A$1B.class, where A is the outer class, and B is the local class.
Especially for callback functions (event handlers in GUIs, like onClick() when a Button is clicked etc.), it's quite usual to use "anonymous classes" - first of all because you can end up with a lot of them. But sometimes anonymous classes aren't good enough - especially, you can't define a constructor on them. In these cases, these method local classes can be a good alternative.

The real purpose of this is to allow us to create classes inline in function calls to console those of us who like to pretend that we're writing in a functional language ;)

The only case when you would like to have a full blown function inner class vs anonymous class ( a.k.a. Java closure ) is when the following conditions are met
you need to supply an interface or abstract class implementation
you want to use some final parameters defined in calling function
you need to record some state of execution of the interface call.
E.g. somebody wants a Runnable and you want to record when the execution has started and ended.
With anonymous class it is not possible to do, with inner class you can do this.
Here is an example do demonstrate my point
private static void testMethod (
final Object param1,
final Object param2
)
{
class RunnableWithStartAndEnd extends Runnable{
Date start;
Date end;
public void run () {
start = new Date( );
try
{
evalParam1( param1 );
evalParam2( param2 );
...
}
finally
{
end = new Date( );
}
}
}
final RunnableWithStartAndEnd runnable = new RunnableWithStartAndEnd( );
final Thread thread = new Thread( runnable );
thread.start( );
thread.join( );
System.out.println( runnable.start );
System.out.println( runnable.end );
}
Before using this pattern though, please evaluate if plain old top-level class, or inner class, or static inner class are better alternatives.

The main reason to define inner classes (within a method or a class) is to deal with accessibility of members and variables of the enclosing class and method.
An inner class can look up private data members and operate on them. If within a method it can deal with final local variable as well.
Having inner classes does help in making sure this class is not accessible to outside world. This holds true especially for cases of UI programming in GWT or GXT etc where JS generating code is written in java and behavior for each button or event has to be defined by creating anonymous classes

I've came across a good example in the Spring. The framework is using concept of local class definitions inside of the method to deal with various database operations in a uniform way.
Assume you have a code like this:
JdbcTemplate jdbcOperations = new JdbcTemplate(this.myDataSource);
jdbcOperations.execute("call my_stored_procedure()")
jdbcOperations.query(queryToRun, new MyCustomRowMapper(), withInputParams);
jdbcOperations.update(queryToRun, withInputParams);
Let's first look at the implementation of the execute():
#Override
public void execute(final String sql) throws DataAccessException {
if (logger.isDebugEnabled()) {
logger.debug("Executing SQL statement [" + sql + "]");
}
/**
* Callback to execute the statement.
(can access method local state like sql input parameter)
*/
class ExecuteStatementCallback implements StatementCallback<Object>, SqlProvider {
#Override
#Nullable
public Object doInStatement(Statement stmt) throws SQLException {
stmt.execute(sql);
return null;
}
#Override
public String getSql() {
return sql;
}
}
//transforms method input into a functional Object
execute(new ExecuteStatementCallback());
}
Please note the last line. Spring does this exact "trick" for the rest of the methods as well:
//uses local class QueryStatementCallback implements StatementCallback<T>, SqlProvider
jdbcOperations.query(...)
//uses local class UpdateStatementCallback implements StatementCallback<Integer>, SqlProvider
jdbcOperations.update(...)
The "trick" with local classes allows the framework to deal with all of those scenarios in a single method which accept those classes via StatementCallback interface.
This single method acts as a bridge between actions (execute, update) and common operations around them (e.g execution, connection management, error translation and dbms console output)
public <T> T execute(StatementCallback<T> action) throws DataAccessException {
Assert.notNull(action, "Callback object must not be null");
Connection con = DataSourceUtils.getConnection(obtainDataSource());
Statement stmt = null;
try {
stmt = con.createStatement();
applyStatementSettings(stmt);
//
T result = action.doInStatement(stmt);
handleWarnings(stmt);
return result;
}
catch (SQLException ex) {
// Release Connection early, to avoid potential connection pool deadlock
// in the case when the exception translator hasn't been initialized yet.
String sql = getSql(action);
JdbcUtils.closeStatement(stmt);
stmt = null;
DataSourceUtils.releaseConnection(con, getDataSource());
con = null;
throw translateException("StatementCallback", sql, ex);
}
finally {
JdbcUtils.closeStatement(stmt);
DataSourceUtils.releaseConnection(con, getDataSource());
}
}

Everything is clear here but I wanted to place another example of reasonable use case for this definition type of class for the next readers.
Regarding #jacob-mattison 's answer, If we assume we have some common actions in these throw-away implementations of the interface, So, it's better to write it once but keep the implementations anonymous too:
//within some method
abstract class myRunnableClass implements Runnable {
protected abstract void DO_AN_SPECIFIC_JOB();
public void run() {
someCommonCode();
//...
DO_AN_SPECIFIC_JOB();
//..
anotherCommonCode();
}
}
Then it's easy to use this defined class and just implement the specific task separately:
taskExecutor.execute(new myRunnableClass() {
protected void DO_AN_SPECIFIC_JOB() {
// Do something
}
});
taskExecutor.execute(new myRunnableClass() {
protected void DO_AN_SPECIFIC_JOB() {
// Do another thing
}
});