Question

I want to catch duplicates mutually exclusively, that is, I need to show that both the first and the third items are duplicates, and that the first and the fourth items are duplicates.

public class Foo
{
    public String Name { get; set; }
    public String SName { get; set; }
}

class Program
{
    static void Main(string[] args)
    {
        var list = new List<Foo>();
        list.Add(new Foo { Name = "a", SName = "d" });
        list.Add(new Foo { Name = "b", SName = "e" });
        list.Add(new Foo { Name = "c", SName = "a" });
        list.Add(new Foo { Name = "a", SName = "f" });

        // only groups by 1 name
        var duplicates = list.GroupBy(i => i.Name).Where(g => g.Count() > 1).Select(g => g.Key);
    }
}

I know this can be done trivially with foreach, I want to learn.

Answer 1

So we’ll start out by getting all of the repeated name values from either column, this is fairly easy:

IEnumerable<string> repeatedNames = list.SelectMany(foo => new[] { foo.Name, foo.SName })
    .GroupBy(name => name)
    .Where (g => g.Count () > 1)
    .Select(g => g.Key);

Next we take each of those names and find all of the items that contain that value. The end result is a sequence for each distinct name where the sequence is all of the Foo items containing that value.

IEnumerable<List<Foo>> groupings = repeatedNames .Select(name =>
    list.Where(foo => foo.Name == name || foo.SName == name).ToList());

If you want it to be, instead of a sequence of Lists of Foos, a sequence of an item with both the distinct value and a sequence then it’s easy enough to add in:

var groupings = repeatedNames .Select(name => new
    {
        Name = name,
        Foos = list.Where(foo => foo.Name == name || foo.SName == name).ToList()
    });