使用List.Distinct（）时返回结果List中的项的顺序

Question

What items will be removed from the resultant list when DISTINCT.ToList() is applied in my following illustration? Is the first entry (ie which was first added to the list) among duplicates be preserved in the new list being returned? If not, is there a way to make the DISTINCT.ToList() to preserve the first entry among duplicates in the new list being returned?

    Dim values As List(Of Integer) = New List(Of Integer)
    values.Add(1)
    values.Add(5)
    values.Add(2)
    values.Add(3)
    values.Add(2)
    values.Add(3)
    values.Add(4)
    values.Add(2)
    values.Add(2)
    values.Add(3)
    values.Add(3)
    values.Add(3)

    Dim items As List(Of Integer) = values.Distinct().ToList

    ' Display result.
    For Each i As Integer In items
        Console.WriteLine(i)
    Next

Expected output:
1
5
2
3
4

This MSDN page says " The Distinct(Of TSource)(IEnumerable(Of TSource)) method returns an unordered sequence that contains no duplicate values ". Is there a way to get around this?

Answer 1

No you can't use Distinct to work around that. As it happens it works exactly as you expect but the documentation explicitly states that it is not guaranteed. Therefore the implementation can change in future versions of the framework so you cannot rely on it. The method is trivial to write. In fact you can even copy the framework implementation .

Again - it currently works as you want it to but it is not guaranteed to do so in the future.

On the other hand I am pretty confident that this implementation will never change as I cannot imagine that more efficient implementation exists.

Here is an implementation for completeness (sorry it's C# and not VB.NET)

public static class MyEnumerable
{
    public static IEnumerable<T> Distinct<T>(this IEnumerable<T> source)
    {
        if (source == null)
        {
            throw new ArgumentNullException(nameof(source));
        }

        var items = new HashSet<T>();

        foreach (T item in source)
        {
            if (items.Add(item))
            {
                yield return item;
            }
        }
    }
}

Answer 2

No you can't get around it with standard methods provided by the framework. You can go around it by coding it yourself like Stilgar suggested.

With your example provided selecting the first item by index is technically irrelevant as you wont be able to know if it was the first or 100th occurrence that in the list since Int are structure.

But that said i am guessing you are using a custom object. In that case your order comes from some sort of sorting. In that case i suggest you instead use a GroupBy<> and then for each group order the items by your OrderBy<> statement and do a First<> on that.

Group by and Distinct are very close. distinct can be replaced by a group by and then first on each group. Indeed it's much slower than the real implementation but the goal here is to explain how you can used that to customize the output if you eventually need more than simply the first item.