better alternative to C# contains method

Question

Is there a faster alternative to using the Contains method in C#?

Here is a sample that looks like what I'm trying to do, but I have read that Contains is an expensive way to do it and I want to find a better alternative as speed is important.

    public List<Guid> getCumulativeListByUsername(string username)
    {
        List<Guid> CombinedList = new List<Guid>();
        List<Guid> ListA = new List<Guid>();
        List<Guid> ListB = new List<Guid>();

        ListA = getListAFromDatabase(username);
        ListB = getListBFromDatabase(username);

        CombinedList = ListA;

        foreach (Guid x in ListB)
        {
            if (!CombinedList.Contains(x))
            {
                CombinedList.Add(x);
            }
        }
        return CombinedList;
    }

If you could explain your reasoning as well, it would be appreciated.

_"I have read that Contains is an expensive way to do it"_ - do you notice that in practice? And where did you read that, did they offer alternatives? If not, what did your research show? The de facto answer to this is to use a HashSet or Dictionary, I'm sure you must have read that as well? — CodeCaster, Jun 16 '17 at 16:47
If you are pulling 2 lists from a database perhaps combine there, databases are good at set operations. — Alex K., Jun 16 '17 at 16:48
https://stackoverflow.com/questions/823860/c-listt-contains-too-slow — mrogers, Jun 16 '17 at 16:48
I searched for answer before I posted, not sure why I didn't find the ones you did. Anyways I only came here to learn @CodeCaster thanks for the downvote though. smh. — cmckeeth, Jun 16 '17 at 16:54

Andy Lamb · Answer 1 · 2017-06-16T16:59:25.517

1

Not sure about performance, you'd need to provide details on list length, likely hood of a clash, etc. But you can shorten the code with Linq:

return ListA.Union(ListB).ToList();

I'd want to prove a performance problem exists before trying to arbitrarily improve performance and more than likely making the code less readable.

edited Jun 16 '17 at 16:59

answered Jun 16 '17 at 16:50

Andy Lamb

2,151
20
22

voteup, But you no need Distinct there, Union do it inside – gabba Jun 16 '17 at 16:51
Most of the time Linq is slower that direct iteration. – Spencer Wieczorek Jun 16 '17 at 16:54
@CodeCaster, cause it more faster readable and short solution than in answer – gabba Jun 16 '17 at 16:55
2

@gabba that's not what the question asked for. – CodeCaster Jun 16 '17 at 16:56
@SpencerWieczorek Contains in list take o(n), Union uses hashes inside so it o(1) – gabba Jun 16 '17 at 16:56
@CodeCaster, I've read question again but still didn't see difference. Sorry for that – gabba Jun 16 '17 at 20:02
@gabba the question is asking for faster execution (fewer CPU cycles used in execution), not faster readability (fewer lines of code). – CodeCaster Jun 17 '17 at 06:03
@CodeCaster, this solution faster than scan in loop – gabba Jun 19 '17 at 10:33

better alternative to C# contains method

1 Answers1