Java stream reduce

Question

I have the following example data set that I want to transform / reduce using Java stream api based on direction\'s value

Direction    int[]
IN           1, 2
O         


        

Answer 1

I think your solution is pretty nice already, especially as using a reduction enables parallelism easily compared to collecting into a shared outside container. But it's easier to use collect instead of reduce as Holger pointed out. Furthermore, the conditions in the accumulator can be simplified a bit, and you forgot to merge the last and first elements in the combiner:

List reduce = tupleStream.collect(ArrayList::new, WDParser::add, WDParser::combine);

private static List combine(List list1, List list2)
{
    if (!list2.isEmpty())
    {
        add(list1, list2.remove(0)); // merge lists in the middle if necessary
        list1.addAll(list2);         // add all the rest
    }
    return list1;
}

private static List add(List list, Tuple t)
{
    int lastIndex = list.size() - 1;
    if (list.isEmpty() || list.get(lastIndex).getDirection() != t.getDirection())
    {
        list.add(t);
    }
    else
    {
        list.set(lastIndex, list.get(lastIndex).merge(t));
    }
    return list;
}

Instead of using indexes to access the first/last element you could even use LinkedList and the methods add/removeFirst/Last().