c# - 在数据表中查找重复项然后比较重复项

Question

我有一个包含重复行的数据表。我需要获取重复项并比较重复行以获得某些列中的最佳值。

DataTable dt = new DataTable();

dt.Rows.Add(1, "Test1", "584", 12);
dt.Rows.Add(2, "Test2", "32", 123);
dt.Rows.Add(3, "Test3", "425", 54);
dt.Rows.Add(4, "Test1", "4", 755);
dt.Rows.Add(5, "Test5", "854", 879);
dt.Rows.Add(6, "Test2", "1", null);
dt.Rows.Add(7, "Test2", "999", 3);

注意测试 1 和 2 有重复项。

(1, "Test1", "584", 12)
(4, "Test1", "4", 755)

(2, "Test2", "32", 123)
(6, "Test2", "1", null)
(7, "Test2", "999", 3)

现在我有了重复项。我需要制作一条具有最佳值(value)的线路。新数据表应显示:

Test1 = "Test1", "584", 755
Test2 = "Test2", "999", 123
Test3 = "Test3", "425", 54
Test5 = "Test5", "854", 879

Answer 1

//GroupBy(x => x[1]) = groupby the second column
//Where(x => x.Count() > 1) = only get groups that have a count greater than 1, so duplicates
var duplicates = dt.Rows.OfType<DataRow>().GroupBy(x => x[1]).Where(x => x.Count() > 1).ToList();

//enumerate all duplicates
foreach (var duplicate in duplicates)
{
    //enumerate each row of the duplicate
    foreach (var dataRow in duplicate)
    {
        //do something…
        //I don't know your rules why a row is better than the other, so that part you have to figure out yourself, or extend your question
    }
}

也许您正在寻找这个:

DataTable dt = new DataTable();
dt.Columns.Add("Id", typeof(int));
dt.Columns.Add("Text", typeof(string));
dt.Columns.Add("Value1", typeof(string));
dt.Columns.Add("Value2", typeof(int));

dt.Rows.Add(1, "Test1", "584", 12);
dt.Rows.Add(2, "Test2", "32", 123);
dt.Rows.Add(3, "Test3", "425", 54);
dt.Rows.Add(4, "Test1", "4", 755);
dt.Rows.Add(5, "Test5", "854", 879);
dt.Rows.Add(6, "Test2", "1", null);
dt.Rows.Add(7, "Test2", "999", 3);

var duplicates = dt.Rows.OfType<DataRow>().GroupBy(x => x[1]).Where(x => x.Count() > 1).ToList();

//get the current highestId (first column) so that when we remove duplicates and a new row the new row will get the next available id
var highestId = dt.Rows.OfType<DataRow>().Max(x => (int)x[0]);

//enumerate all duplicates
foreach (var duplicate in duplicates)
{
    //get the highest value of each column
    var newId = ++highestId;
    var newText = duplicate.Key;
    var newValue1 = duplicate.Max(x => x[2]); //this does a string comparison, instead of a numeric one, this means that for example that 2 is bigger then 10

    // use this if you need numeric comparison
    var newValue1AsNumeric = duplicate.Select(x =>
    {
        if (int.TryParse(Convert.ToString(x[2]), out var value))
            return value;

        return (int?)null;
    }).OfType<int>().Max(); 

    var newValue2 = duplicate.Select(x => x[3]).OfType<int>().Max();

    //enumerate each row of the duplicate
    foreach (var dataRow in duplicate)
        dt.Rows.Remove(dataRow);

    dt.Rows.Add(newId, newText, newValue1, newValue2);
}

您可以在此处查看运行中的代码: https://dotnetfiddle.net/rp1DUc