Linq 查询，其中 DataTable 的第一列与重复的第二列进行比较

Question

我已合并在两列中都包含重复项的 DataTable。

DataTable 结构：“Worknumber 用户名”

1234 John
1235 Mike
1235 Mike
1236 Donald
1236 Jack

我要赶上

1236 Donald
1236 Jack

最好以格式输出

1236 Donald Jack

我一直在尝试为此目的创建 Linq 查询一段时间（相当长的时间），但我无法找到正确的方法。

这是我目前所拥有的。这行得通，我刚刚用dataRow.Field<string>("WORKNUMBER") 测试了列名是否正确：

        public static DataTable ProjectsInLiinos = new DataTable();

        ...

        MainProcess.MergedTable();

        foreach (DataRow dataRow in MainProcess.ProjectsInLiinos.Rows)
        {
                Console.WriteLine(dataRow.Field<string>("WORKNUMBER") + dataRow.Field<string>("USERNAME"));
        }

这是 Linq 查询：

var queryCustomers = from data in MainProcess.ProjectsInLiinos.AsEnumerable()
                                 where data.Field<string>("WORKNUMBER").Count() > 1 && data.Field<string>("USERNAME").Count() < 1
                                 select data;

它不会产生任何东西。我不知道如何将第一列重复项与第二列重复项进行比较？

否则我可以将 dataRow.Field<string>("WORKNUMBER") + dataRow.Field<string>("USERNAME") 分组，然后检查列表是否有重复项，但它会生成：

1235Mike
1235Mike

反过来

1234John
1236Donald
1236Jack

Answer 1

尝试以下：

            DataTable dt = new DataTable();
            dt.Columns.Add("Worknumber",typeof(int));
            dt.Columns.Add("Username",typeof(string));
            
            dt.Rows.Add(new object[] {1234, "John"});
            dt.Rows.Add(new object[] {1235, "Mike"});
            dt.Rows.Add(new object[] {1235, "Mike"});
            dt.Rows.Add(new object[] {1236, "Donald"});
            dt.Rows.Add(new object[] {1236, "Jack"});

            var distinct = dt.AsEnumerable()
                .GroupBy(x => x.Field<int>("Worknumber"))
                .Select(x => new { workNumber = x.Key, user = x.Select(y => y.Field<string>("Username")).Distinct().ToList() })
                .Where(x => x.user.Count > 1)
                .Select(x => new { workNumber = x.workNumber, user = string.Join(" ", x.user) })
                .ToList();

Answer 2

查看这个简单的演示

public class C
{
    public int Worknumber { get; set; }
    public string Username { get; set; }

    public C(int worknumber, string username)
    {
        Worknumber = worknumber;
        Username = username;
    }
}

void Main()
{
    var lst = new List<C>(){
        new C(1234, "Mike"),
        new C(1235, "Mike"),
        new C(1235, "Mike"),
        new C(1236, "Donald"),
        new C(1236, "Jake")
    };

    var p = lst.GroupBy(t => t.Worknumber)
        .Select(t => new { Key = t.Key, l = t.Select(x => x.Username).ToList() })
        .Where(x => x.l.Distinct().Count() > 1)
        .Select(x => String.Format("{0} {1}", x.Key,  string.Join(' ', x.l)))
        .Dump();
}

Answer 3

您可以使用分组方式：

void Main()
{
    DataTable tbl = new DataTable();
    tbl.Columns.Add("Id", typeof(int));
    tbl.Columns.Add("Name", typeof(string));
    tbl.Rows.Add(1234, "John");
    tbl.Rows.Add(1235, "Mike");
    tbl.Rows.Add(1235, "Mike");
    tbl.Rows.Add(1236, "Donald");
    tbl.Rows.Add(1236, "Jack");

    var dupes = tbl.AsEnumerable()
    .GroupBy(t => t.Field<int>("Id"))
    .Where(x => x.Select(y => y.Field<string>("Name")).Distinct().Count() > 1)
    .Select(x => new
    {
        Id = x.Key,
        Names = string.Join(",", x.Select(y => y.Field<string>("Name")))
    });
    foreach (var dupe in dupes)
    {
        Console.WriteLine($"{dupe.Id}:{dupe.Names}");
    }
}

我认为无论如何使用源代码会好得多，而不是介于两者之间的 DataTable。

Answer 4

正如已经建议的那样，首先使用第一列分组。然后，您可以通过 Where 过滤器中的第二个分组找到不同的用户名：

void Main()
{
    string column1 = "Worknumber";
    string column2 = "Username";
    DataTable table = new DataTable();
    table.Columns.Add(column1, typeof(int));
    table.Columns.Add(column2, typeof(string));

    table.Rows.Add(1234, "john");
    table.Rows.Add(1235, "Mike");
    table.Rows.Add(1235, "Mike");
    table.Rows.Add(1236, "Donald");
    table.Rows.Add(1236, "Jack");
    table.Rows.Add(1237, "Val");
    table.Rows.Add(1237, "Nick");
    var result = table.AsEnumerable()
         .GroupBy(x => x.Field<int>(column1))
         .Where(g => g.Count() > 1 && g.GroupBy(x => x.Field<string>(column2)).Count() > 1)
         .Select(x => new {WorkNumber = x.Key, AllUsers = string.Join(" ", x.Select(u => u.Field<string>(column2)))}).Dump();       

}

我在这里使用了匿名类型，但您可以轻松编写自定义类，并使用适当的构造函数来适应用户名值。

Answer 5

假设你有

class MyClass : IEquatable<MyClass> // IEquatable is required to perform Dictinct()
{
    public string Worknumber { get; set; }
    public string Username { get; set; }

    public bool Equals(MyClass other)
    {
        if (this.Username != other.Username)
            return false;

        if (this.Worknumber != other.Worknumber)
            return false;

        return true;
    }

    public override int GetHashCode() // required for objects comparison
    {
        int hashWorknumber = Worknumber == null ? 0 : Worknumber.GetHashCode();
        int hashUsername = Username == null ? 0 : Username.GetHashCode();

        return hashWorknumber ^ hashUsername;
    }
}

还有这样的模拟

List<MyClass> myclasses = new List<MyClass>();

myclasses.Add(new MyClass { Worknumber = "1234", Username = "John" });
myclasses.Add(new MyClass { Worknumber = "1235", Username = "Mike" });
myclasses.Add(new MyClass { Worknumber = "1235", Username = "Mike" });
myclasses.Add(new MyClass { Worknumber = "1236", Username = "Donald" });
myclasses.Add(new MyClass { Worknumber = "1236", Username = "Jack" });

我们做一个简单的 lambda 查询

var result = myclasses
    .Distinct()
    .GroupBy(e => e.Worknumber) // grouping
    .Select(e => new
    {
        Worknumber = e.Key,
        Count = e.Count(),
        Users = string.Join(" ", e.Select(p => p.Username)) // joining names
    })
    .Where(e => e.Count > 1) // filtering duplicates only
    .ToList();