There are certain key points that are needed to be remembered before adhering forward where we will be discussing and implementing static and dynamic bindings in Java later concluding out the differences.
- private, final and static members (methods and variables) use static binding while for virtual methods (In Java methods are virtual by default) binding is done during run time based upon the run time object.
- The static binding uses Type information for binding while Dynamic binding uses Objects to resolve to bind.
- Overloaded methods are resolved (deciding which method to be called when there are multiple methods with the same name) using static binding while overridden methods use dynamic binding, i.e, at run time.
Static Binding
The binding which can be resolved at compile time by the compiler is known as static or early binding. The binding of all the static, private, and final methods is done at compile-time.
Example:
Java
// Java Program to Illustrate Static Binding // Main class class NewClass { // Static nested inner class // Class 1 public static class superclass { // Method of inner class static void print() { // Print statement System.out.println( "print() in superclass is called" ); } } // Static nested inner class // Class 2 public static class subclass extends superclass { // Method of inner class static void print() { // print statement System.out.println( "print() in subclass is called" ); } } // Method of main class // Main driver method public static void main(String[] args) { // Creating objects of static inner classes // inside main() method superclass A = new superclass(); superclass B = new subclass(); // Calling method over above objects A.print(); B.print(); } } |
print() in superclass is called print() in superclass is called
Output Explanation: As you can see, in both cases the print method of the superclass is called. Let us discuss how this happens
- We have created one object of subclass and one object of the superclass with the reference of the superclass.
- Since the print method of the superclass is static, the compiler knows that it will not be overridden in subclasses and hence compiler knows during compile time which print method to call and hence no ambiguity.
As an exercise, the reader can change the reference of object B to subclass and then check the output.
Dynamic Binding
In Dynamic binding compiler doesn’t decide the method to be called. Overriding is a perfect example of dynamic binding. In overriding both parent and child classes have the same method.
Example:
Java
// Java Program to Illustrate Dynamic Binding // Main class public class GFG { // Static nested inner class // Class 1 public static class superclass { // Method of inner class 1 void print() { // Print statement System.out.println( "print in superclass is called" ); } } // Static nested inner class // Class 2 public static class subclass extends superclass { // Method of inner class 2 @Override void print() { // Print statement System.out.println( "print in subclass is called" ); } } // Method inside main class public static void main(String[] args) { // Creating object of inner class 1 // with reference to constructor of super class superclass A = new superclass(); // Creating object of inner class 1 // with reference to constructor of sub class superclass B = new subclass(); // Calling print() method over above objects A.print(); B.print(); } } |
print in superclass is called print in subclass is called
Output Explanation: Here the output differs. But why? Let’s break down the code and understand it thoroughly.
- Methods are not static in this code.
- During compilation, the compiler has no idea as to which print has to be called since the compiler goes only by referencing variable not by the type of object, and therefore the binding would be delayed to runtime and therefore the corresponding version of the print will be called based on type on an object.
Tip: Geeks, now the question arises why binding of static, final, and private methods is always static binding?
Static binding is better performance-wise (no extra overhead is required). The compiler knows that all such methods cannot be overridden and will always be accessed by objects of the local class. Hence compiler doesn’t have any difficulty determining the object of the class (local class for sure). That’s the reason binding for such methods is static.
Let us now finally see the differences between static and dynamic binding that is as follows:
Static Binding | Dynamic Binding |
---|---|
It takes place at compile time for which is referred to as early binding | It takes place at runtime so do it is referred to as late binding. |
It uses overloading more precisely operator overloading method | It uses overriding methods. |
It takes place using normal functions | It takes place using virtual functions |
Static or const or private functions use real objects in static binding | Real objects use dynamic binding. |
This article is contributed by Rishabh Mahrsee. If you like Lazyroar and would like to contribute, you can also write an article and mail your article to review-team@geeksforgeeks.org. See your article appearing on the Lazyroar main page and help other Geeks.