Getting the absolute position from relative index [^x] in array, list or dictionary? (C#)

Question

Since C# version 8, there is a new way of indexing, as shown in the example:

void Main()
{
    Index idx = ^5; // counted from last to first element

    var sentence = new string[]
    {
        "The", "quick", "brown", "fox",         // position 0 ... 3
        "jumped", "over", "the", "lazy", "dog"  // position 4 ... 8
    };
    var lstSentence = sentence.ToList(); // applicable to list as well

    Console.WriteLine(sentence[idx]);
    Console.WriteLine(lstSentence[idx]);
}

It returns jumped.

So far so good. But how can you get the absolute position representing ^5 efficiently ?
I saw .NET internally is using Index (which I added to this example), and there is a value property.

How can I get the position of sentence[^5] ? Something like

var absPos = sentence[^5].Index.Value; // get absolute position (which is = 4)

(but unfortunately this doesn't seem to exist)

I know there is a formula (number of elements - position from right), but the compiler / .net calculates and probably stores this somewhere.

Update: idx.Value isn't the solution, it just stores the relative value 5 together with .IsFromEnd, which is true. But you can't access it directly, like sentence[^5].Index.Value which is what I wanted to know.

Note: This is also useful for Dictionary, if you take a look at the RudimentaryMultiValuedDictionary Example - if you want to add an additional indexer like public List<TValue> this[Index idx] you will require that. internalDictionary.ElementAtOrDefault(i), which you need for this purpose, does only work with integer values (absolute positions).

Answer 1

If you use a decompiler to see what code is generated for your example, it looks like this:

private static void Main()
{
    Index index = new Index(5, true);
    string[] strArray = new string[] { "The", "quick", "brown", "fox", "jumped", "over", "the", "lazy", "dog" };
    Console.WriteLine(strArray[index.GetOffset((int)strArray.Length)]);
}

As you can see, it just uses index.GetOffset((int)strArray1.Length) to calculate the offset. There's no way to do this yourself other than by doing it the same way. There's no other language construct that you can use.

(I'm not sure why it casts the strArray1.Length property to int, since it's already an int.)

Also note that if you put the index inline using code like this:

var sentence = new string[]
{
    "The", "quick", "brown", "fox",
    "jumped", "over", "the", "lazy", "dog"
};

Console.WriteLine(sentence[^5]);

That gets compiled to:

string[] strArray = new string[] { "The", "quick", "brown", "fox", "jumped", "over", "the", "lazy", "dog" };
Console.WriteLine(strArray[(int)strArray.Length - 5]);

Which has optimised away the use of an Index object altogether.

---

As I mentioned in the comments below, you could write an extension method which accepts an IList. This would then work with anything that implements IList such as List<T>, arrays, ImmutableSortedSet<T> and so on.

For example:

public static class IndexExt
{
    public static int AbsIndexFor(this IList self, Index index)
    {
        return index.GetOffset(self.Count);
    }
}

Then you can do:

var array = new string[]
{
    "The", "quick", "brown", "fox",
    "jumped", "over", "the", "lazy", "dog"
};

Console.WriteLine(array.AbsIndexFor(^5));

var list = new List<string>
{
    "The", "quick", "brown", "fox",
    "jumped", "over", "the", "lazy", "dog"
};

Console.WriteLine(list.AbsIndexFor(^5));