CODEWITHSUNDEEP
c#linqlistrange
Punit Vora
Punit Vora

Reputation: 5188

How to select values within a provided index range from a List using LINQ

I am a LINQ newbie trying to use it to acheive the following:

I have a list of ints:-

List<int> intList = new List<int>(new int[]{1,2,3,3,2,1});

Now, I want to compare the sum of the first three elements [index range 0-2] with the last three [index range 3-5] using LINQ. I tried the LINQ Select and Take extension methods as well as the SelectMany method, but I cannot figure out how to say something like 

(from p in intList  
where p in  Take contiguous elements of intList from index x to x+n  
select p).sum()

I looked at the Contains extension method too, but that doesn't see to get me what I want. Any suggestions? Thanks.

Upvotes: 75

Views: 83647

Answers (5)

Jaroslav Kubacek
Jaroslav Kubacek

Reputation: 1447

Starting from .NET 6 it is possible to use range syntax for Take method.

List<int> intList = new List<int>(new int[]{1, 2, 3, 3, 2, 1});

// Starting from index 0 (including) to index 3 (excluding) will select indexes (0, 1, 2)
Console.WriteLine(intList.Take(0..3).Sum()); // {1, 2, 3} -> 6

// By default is first index 0 and can be used following shortcut.
Console.WriteLine(intList.Take(..3).Sum());  // {1, 2, 3} -> 6      


// Starting from index 3 (including) to index 6 (excluding) will select indexes (3, 4, 5)
Console.WriteLine(intList.Take(3..6).Sum()); // {3, 2, 1} -> 6  

// By default is last index lent -1 and can be used following shortcut.
Console.WriteLine(intList.Take(3..).Sum());  // {3, 4, 5} -> 6

// Reverse index syntax can be used. Take last 3 items.
Console.WriteLine(intList.Take(^3..).Sum()); // {3, 2, 1} -> 6

// No exception will be raised in case of range is exceeded.
Console.WriteLine(intList.Take(^100..1000).Sum());

So simply put, intList.Take(..3).Sum() and intList.Take(3..).Sum() can be used with .NET 6.

Upvotes: 5

stomy
stomy

Reputation: 1988

To filter by specific indexes (not from-to):

public static class ListExtensions
{
   public static IEnumerable<TSource> ByIndexes<TSource>(this IList<TSource> source, params int[] indexes)
   {        
        if (indexes == null || indexes.Length == 0)
        {
            foreach (var item in source)
            {
                yield return item;
            }
        }
        else
        {
            foreach (var i in indexes)
            {
                if (i >= 0 && i < source.Count)
                    yield return source[i];
            }
        }
   }
}

For example:

string[] list = {"a1", "b2", "c3", "d4", "e5", "f6", "g7", "h8", "i9"};
var filtered = list.ByIndexes(5, 8, 100, 3, 2); // = {"f6", "i9", "d4", "c3"};

Upvotes: 1

Tao
Tao

Reputation: 13996

For larger lists, a separate extension method could be more appropriate for performance. I know this isn't necessary for the initial case, but the Linq (to objects) implementation relies on iterating the list, so for large lists this could be (pointlessly) expensive. A simple extension method to achieve this could be:

public static IEnumerable<TSource> IndexRange<TSource>(
    this IList<TSource> source,
    int fromIndex, 
    int toIndex)
{
    int currIndex = fromIndex;
    while (currIndex <= toIndex)
    {
        yield return source[currIndex];
        currIndex++;
    }
}

Upvotes: 18

onuralp
onuralp

Reputation: 910

You can use GetRange()

list.GetRange(index, count);

Upvotes: 41

Adam Sills
Adam Sills

Reputation: 17062

Use Skip then Take.

yourEnumerable.Skip(4).Take(3).Select( x=>x )

(from p in intList.Skip(x).Take(n) select p).sum()

Upvotes: 109

Related Questions