LINQ order by "round robin"

Question

Seems like this should be an easy task but I can't figure out how to do this with LINQ. The only information I've been able to find so far is regarding the round robin tournament format, which isn't what I'm after. I may be searching wrong. Given the following list:

var items [] { "apple", "banana", "banana", "candy", "banana", "fruit", "apple" };

How can I sort this (preferably using linq) so that it comes out in "round robin" order, that is, select each unique item once before repeats. So the above list would come out like this (It's not important if it comes out in alphabetical order, even though this list does):

var sorted [] { "apple", "banana", "candy", "fruit", "apple", "banana", "banana" };

I know I can do this by iterating over it the hard way, I was just hoping for something easier. Does anyone have any insight how to do this? Thanks in advance!

Can you explain exactly what you mean by "round-robin" sorting? — Mathew Thompson, May 28 '12 at 11:07
He mean "round-robin" sorting http://en.wikipedia.org/wiki/Round-robin — Likurg, May 28 '12 at 11:11

score 10 · Accepted Answer · answered May 28 '12 at 11:34

10

var sorted = items.GroupBy(s => s)
    .SelectMany(grp => grp.Select((str, idx) => new { Index = idx, Value = str }))
    .OrderBy(v => v.Index).ThenBy(v => v.Value)
    .Select(v => v.Value)
    .ToArray();

answered May 28 '12 at 11:34

Lee

142,018
20
234
287

shoulda learn to copypaste properly. accept this, this works. – Alex May 28 '12 at 11:42
You sir are awesome. It works perfectly. Thank you! All I needed to change was the GroupBy for my project in order to group by the actual unique item I cared about, the rest was literally copy-paste. Thanks again! – Eric May 28 '12 at 18:29

score 0 · Answer 2 · answered May 28 '12 at 11:34

I did this once, dug up the code:

//Originially written for lists, all you need is prepend a .ToList() where needed to apply this to an array
List<string> src = new List<string> { "string1", "string2" }; //source
List<string> dst = new List<string>();

dst.AddRange(src.Distinct());
dst.ForEach(d => src.RemoveAt(src.FindIndex(i => i.Equals(d)))); //remove the first occurrence of each distinct element
dst.AddRange(src);

score 0 · Answer 3 · answered May 28 '12 at 11:58

Just saw that two answers popped up while I was writing this; oh well, here is another way:

var items [] { "apple", "banana", "banana", "candy", "banana", "fruit", "apple" };

var uniqueItems = items.Distinct().OrderBy(item => item); // alphabetical orderBy is optional

var duplicateItems = items
                     .GroupBy(item => item)
                     .SelectMany(group => group.Skip(1))
                     .OrderBy(item => item); // alphabetical orderBy is optional;

var sorted = uniqueItems.Append( duplicateItems ).ToArray();

LINQ order by "round robin"

3 Answers3

Linked