如何用数组_c＃Linq的最后一个索引替换数组的第二个和第三个索引？

Question

Actually I have a list like

oldList= ['First','Second','Third',.....'Seventh'];

newArray should be like that

newArray=['First',.........'Seventh','Second','Third']

so How can I provide ordering like that with linq c#?

I mean my first element should be kept at old place. but I have to skip second and third elements to end of the list

Answer 1

Weird requirement, but you can use OrderBy even in this case:

var newArray = oldList.Select((t, index) => new { t, index })
                      .OrderBy(x => x.index == 1 || x.index == 2)
                      .ThenBy(x => x.index)
                      .Select(x => x.t)
                      .ToArray();

Here's a demo: http://ideone.com/qJOlv

Answer 2

Yes, you can do this in LINQ, though it might not be the most efficient:

var newArray = oldList.Take(1)
    .Concat(oldList.Skip(3).Take(4))
    .Concat(oldList.Skip(1).Take(2))
    .ToArray();

However, if your original container is a List<T> you can do things a little more efficiently:

var newList = oldList.GetRange(0, 1)
    .Concat(oldList.GetRange(3, 4))
    .Concat(oldList.GetRange(1, 2))
    .ToList();

UPDATE :

For kicks and grins I went ahead and timed several of the proposed methods, and it looks like @LB's Get/Add/RemoveRange() comes on top:

Over 100,000 iterations:

• Skip() / Take() took 82 ms (0.00082 ms / call)
• GetRange() / Concat() took 69 ms (0.00069 ms / call)
• OrderBy() / ThenBy() took 145 ms (0.00145 ms / call)
• Add/Get/RemoveRange() took 21 ms (0.00021 ms / call)

Of course, we're getting into microoptimization here, so I'd say go with the one that makes the most sense to you and is the most maintainable. But it looks like LB's solution is the most performant (and is easy to read as well).

Answer 3

If your collection is already a list, you can skip the first line

List<string> oldList2 = new List<string>(oldList);
var tmp = oldList2.GetRange(1, 2);
oldList2.RemoveRange(1, 2);
oldList2.AddRange(tmp);

Answer 4

LINQ is designed for querying data sources to produce new results, not mutating existing data.

The general approach I would use to solve this problem:

var second = array[1];
var third = array[2];

for(int i = 3; i < array.Length; i++)
{
    array[i-2] = array[i];
}

array[array.Length-2] = second;
array[array.Length-1] = third;

If you really, really need to use LINQ, I'd do something like this:

var newArray = array.Take(1)
          .Concat(array.Skip(3))
          .Concat(array.Skip(1).Take(2))
          .ToArray();

Note that this approach iterates the data source 3 times; that's just fine if it's actually an array or list, but if, in reality, it's just an IEnumerable<T> thats the result of some query or complex operation that would be bad.