Start of Tutorial > Start of Trail > Start of Lesson

Search Feedback Form

Details of a Method Declaration

A method's declaration provides a lot of information about the method to the compiler, to the runtime system, and to other classes and objects. Included is not only the name of the method, but also such information as the return type of the method, the number and type of the arguments required by the method, and which other classes and objects can call the method.

While this may sound like writing a novel rather than simply declaring a method, most method attributes can be declared implicitly. The only required elements of a method declaration are the method's name, its return type, and a pair of parentheses ( ). This figure shows the elements of a method declaration.

Each element of a method declaration is further defined below:

accessLevel

As with member variables, you control which other classes have access to a method using one of four access levels: public, protected, package, and private. Controlling Access to Members of a Class covers access levels in detail.

static

As with member variables, static declares this method as a class method rather than an instance method. Understanding Instance and Class Members talks about declaring instance and class methods.

abstract

An abstract method has no implementation and must be a member of an abstract class. Refer to Writing Abstract Classes and Methods for information about why you might want to write an abstract method and how such methods affect subclasses.

final

A final method cannot be overridden by subclasses. Writing Final Classes and Methods discusses why you might want to write final methods, how they affect subclasses, and whether you might want to write a final class instead.

native

If you have a significant library of functions written in another language such as C, you may wish to preserve that investment and use those functions from Java. Methods implemented in a language other than Java are called native methods and are declared as such using the native keyword. Check out our Java Native Interface trail for information about writing native methods.

synchronized

Concurrently running threads often invoke methods that operate on the same data. These methods may be declared synchronized to ensure that the threads access information in a thread-safe manner. Synchronizing method calls is covered in Threads: Doing Two or More Tasks At Once. Take particular note of the section entitled Synchronizing Threads.

returnType

Java requires that a method declare the data type of the value that it returns. If your method does not return a value, use the keyword void for the return type. Returning a Value from a Method talks about the issues related to returning values from a method.

methodName

A method name can be any legal Java identifier. You need to consider several issues in regards to Java method names. These are covered in Method Names.

( paramlist )

You pass information into a method through its arguments. See the next section, Passing Information into a Method.

[throws exceptions]

If your method throws any checked exceptions, your method declaration must indicate the type of those exceptions. See Handling Errors with Exceptionsfor information. In particular, refer to Specifying the Exceptions Thrown by a Method.

You declare a method's return type in its method declaration. Within the body of the method, you use the return operator to return the value. Any method that is not declared void must contain a return statement. The Stack class declares the isEmpty method, which returns a boolean:

public boolean isEmpty() {
    if (items.size() == 0)
        return true;
    else
        return false;
}

The data type of the return value must match the method's return type; you can't return an Object type from a method declared to return an integer. The isEmpty method returns either the boolean value true or false, depending on the outcome of a test. A compiler error results if you try to write a method in which the return value doesn't match the return type.

The isEmpty method returns a primitive type. Methods also can return a reference type. For example, Stack declares the pop method that returns the Object reference type:

public synchronized Object pop() {
    int len = items.size();
    Object obj = null;
    if (len == 0)
        throw new EmptyStackException();
    obj = items.elementAt(len - 1);
    items.removeElementAt(len - 1);
    return obj;
}

When a method returns an object such as pop does, the class of the returned object must be either a subclass of or the exact class of the return type. This can be a source of confusion, so let's look at this more closely. Suppose you have a class hierarchy where ImaginaryNumber is a subclass of java.lang.Number, which is, in turn, a subclass of Object, as illustrated here:

Now suppose you have a method declared to return a Number:

public Number returnANumber() {
    . . .
}

The returnANumber method can return an ImaginaryNumber but not an Object. ImaginaryNumber "is a" Number because it's a subclass of Number. However, an Object is not necessarily a Number--it could be a String or some other type. You also can use interface names as return types. In this case, the object returned must implement the specified interface.

Java supports method name overloading so that multiple methods can share the same name. For example, suppose you are writing a class that can render various types of data (strings, integers, and so on) to its drawing area. You need to write a method that knows how to render each data type. In other languages, you have to think of a new name for each method, for example, drawString, drawInteger, drawFloat, and so on. In Java, you can use the same name for all of the drawing methods but pass a different type of parameter to each method. So, in your data rendering class, you can declare three methods named draw, each of which takes a different type of parameter:

class DataRenderer {
    void draw(String s) {
        . . .
    }
    void draw(int i) {
        . . .
    }
    void draw(float f) {
        . . .
    }
}

Overloaded methods are differentiated by the number and type of the arguments passed into the method. In the code sample, draw(String s) and draw(int i) are distinct and unique methods because they require different argument types. You cannot declare more than one method with the same name and the same number and type of arguments because the compiler cannot differentiate them. So, draw(String s) and draw(String t) are identical and result in a compiler error.

A class may override a method in its superclass. The overriding method must have the same name, return type, and parameter list as the method it overrides. Overriding Methods shows you how to override methods.

Start of Tutorial > Start of Trail > Start of Lesson

Search Feedback Form

Copyright 1995-2001 Sun Microsystems, Inc. All rights reserved.