Question

I’m writing a method that doubles the size of the data in every node in a tree. I think I have the algorithm down for writing it, but I’m having trouble figuring out how to multiply a generic type by 2.

So essentially I want to take a data that is type T and convert it to an integer, multiply it by two, and store it as type T. I’ve played around with some options but none have worked. Any help on how I should do this would be helpful. If you need any other info let me know. Thanks.

Answer 1

Because of type erasure, you cannot do this directly. One approach is is to define a generic interface for objects that “know” how to double themselves:

public interface Doubler<T> {
    T doubled(T arg);
}

Then you can make each node (that knows how to double itself) implement the Doubler interface.

There are other approaches (factory methods, factory objects, passing around Class<T> instances, etc.) that might work just as well (or better). But the one thing you can’t do is create an instance of a generic type in code that just knows the generic type and nothing else.