let's assume I have a task to create a Set of class names. To remove duplication of .getName() method calls for each class, I used org.apache.commons.collections.CollectionUtils and org.apache.commons.collections.Transformer as follows:

Snippet 1:

Set<String> myNames = new HashSet<String>();
CollectionUtils.collect(
        Arrays.<Class<?>>asList(My1.class, My2.class, My3.class, My4.class, My5.class),
        new Transformer() {
            public Object transform(Object o) {
                return ((Class<?>) o).getName();
            }
        }, myNames);

An alternative would be this code:

Snippet 2:

Collections.addAll(myNames, My1.class.getName(), My2.class.getName(), My3.class.getName(), My4.class.getName(), My5.class.getName());

So, when using functional programming approach is overhead and when it's not and why?

Isn't my usage of functional programming approach in snippet 1 is an overhead and why?