Understanding java generic methods

Question

I'm trying to understand generic methods in Java. Given the following code:

public class GenericTest {

    interface Shape {
        public long area();
    }

    public static class Triangle implements Shape
    {
        private long base, height;
        public long area() { return (base * height) / 2; }
    }

    public static  class Rectangle implements Shape
    {
        private long width, height;
        public long area() { return width * height; }
    }

    public <T extends Shape> long area1(T shape)
    {
        return shape.area();
    }

    public long area2(Shape shape)
    {
        return shape.area();
    }

}

I can't see/understand why I should use/implement area1 instead of area2 (or vice versa). Am I missing something? Don't both methods do the same thing?

It has left me a bit confused regarding generics in Java

Answer 1

In your example, since the T type parameter isn't used in any return values, there is no difference.

However, imagine you had the following methods:

public <T extends Shape> T movedShape1(T shape) {
    return shape.move();
}

public Shape movedShape2(Shape shape) {
    return shape.move();
}

Here you can see a clear advantage to using movedShape1(). You get a more specific type for your return value, without losing any type safety.

Answer 2

There is no good reason for creating the area1 method. The area2 method is preferable. Generics are to be used when there is a relationship with a specific yet unknown type. Here, there is nothing specific about the parameter shape. The interface Shape already lets us use the area method, so we don't care which specific implementation of Shape is passed in as shape. So generics aren't needed here.

Use

public long area2(Shape shape)