Question

I’ve just come across some code that’s confusing me slightly; there are really 2 variations that I’d like to clarify.

Example 1:

public String getFilepath(){
        final File file = new File(this.folder, this.filename);
        return file.getAbsolutePath();
    }

What would be the purpose of declaring file “final“? Since Java primitives are passed by value, and getAbsolutePath() is just returning a String, the variable won’t be final on the other side (calling method), will it? And since the file variable only exists within the scope of these 2 lines, I can’t really see any purpose of the final keyword. Is there something I’m missing? Anyone see a reason to do this?

Example 2:

public String getFilepath(){
        final File file = new File(this.folder, this.filename);
        return file;
    }

Since here the actual object is being returned… Does that mean the file variable will be constant/final on the other side…? It doesn’t seem to make sense.

In general, it seems to me that you pass a variable, without it’s access type. As in, I can have a private variable in a function with a public get function that returns it – but the variable that receives it by calling the function has to specify an access modifier. So if it specifies public, the returned variable will be public in that scope. If it specifies private, the returned variable will be private in that scope. Is there a difference with final? Is the “constancy” of a variable something that can be passed? This strikes me as rather improbable, considering what I know of Java.

Or am I missing the point entirely and there’s some other purpose of the final keyword in the above code?

Edit:

I checked back with the original developer who wrote the code, and he said he only put the final keyword in because he had originally thought the method would be a lot longer and wanted to ensure that the file stayed constant throughout. He also said that he generally declares variables that should not be changed as final, as a rule across the board and sort of on principle – a point that both the answers below mentioned. So it seems I was reading too much into a simple extra keyword included for standards reasons. Thanks everyone!

Answer 1

Some people might tell you that there’s a performance benefit to using final, but that is, in no way, conclusively proven.

The primary benefit of the final keyword is for the programmer to indicate that a class, method, or field should not be changed.

Bear in mind that declaring a variable final does not make the referenced object immutable. It just means that the variable cannot have its value reassigned. You can still run methods of the variable file that could change the File object internally.

In the two methods you give, I see no value in making the file variable final. Some code conventions advocate making all variable final unless the need to be modified. Some people don’t subscribe to that. I consider it a variation on the precautionary principle.