Question

I have a datatable with three columns

Id name value value1

I want to create a string from this datatable so that

1. each row in datatable will create two rows
2. name will become one row
3. value 1 will become another row

4. If name and value1 are similar then include value 1 only otherwise include both name and value1 (this is done)

5. If name of one row is similar to name in any other row, then add following text infront of both duplicate rows in string

Id is similar to Id

This is what I have written so far:

Public Function GetGazetteer(dataTable As DataTable) As String
            Dim processedData = New List(Of String)
            Dim rowData = String.Empty
            Dim results = New StringBuilder()

            For Each row As DataRow In dataTable.Rows
                processedData.Add(row(1))
                If row(3) <> row(1) Then
                    processedData.Add(row(3))
                End If
            Next

            For Each row As String In processedData
                rowData = row

                If rowData.Trim <> String.Empty Then
                    If (processedData.Where(Function(d) d = rowData).Count = 1) Then
                        results.Append(rowData)
                        results.Append("<br />")
                    Else
                        results.Append(rowData)
                        results.Append("*")
                        results.Append("<br />")
                    End If
                End If

            Next

            Return results.ToString
        End Function

Currently, * is added ( please suggest how to add above text.)

Here is a sample

id   name           value  value1
1    this is string 1       abc    This is sample
2    this is string 2       abc    this is string 2
3    this is string 3       abc    this is string 4
4    this is string 3       abc    asasaasd

Here is desired output

this is string 1
This is sample
this is string 2
*3 is duplicate of 4* this is string 3
this is string 4
*4 is duplicate of 3* this is string 3
asasaasd

Answer 1

Edited based on output given on question

Here is a solution using C# (hope you can convert this to VB.NET). I have used a struct to hold id, value, col information.

The struct is

    public struct Container
    {
       public int Id;
       public int Col;
       public string Value;
    }

The method is

        public string GetGazetteer(DataTable dtInput)
        {
            string result = null;
            List<Container> containers = new List<Container>();
            List<Container> finalContainers = new List<Container>();
            StringBuilder sb = new StringBuilder();

            foreach (DataRow row in dtInput.Rows)
            {
                Container container;
                container = new Container() { Id = int.Parse(row[0].ToString()), Value = row[1].ToString(), Col = 1 };
                containers.Add(container);
                if (row[1] != row[3])
                {
                    container = new Container() { Id = int.Parse(row[0].ToString()), Value = row[3].ToString(), Col = 2 };
                    containers.Add(container);
                }
            }

            containers = containers.OrderBy(c => c.Value).ThenBy(c => c.Id).ToList();

            if (containers.Count > 0)
            {
                string initialVal = containers[0].Value;
                finalContainers.Add(containers[0]);
                for (int i = 1; i < containers.Count; i++)
                {
                    if (containers[i].Value == initialVal)
                    {
                        finalContainers.Remove(containers[i]);
                        finalContainers.Remove(containers[i - 1]);
                        finalContainers.Add(new Container() { Id = containers[i - 1].Id, Value = "*" + containers[i - 1].Id + " is duplicate of " + containers[i].Id + "* " + containers[i - 1].Value });
                        finalContainers.Add(new Container() { Id = containers[i].Id, Value = "*" + containers[i].Id + " is duplicate of " + containers[i - 1].Id + "* " + containers[i].Value });
                    }
                    else
                    {
                        finalContainers.Add(containers[i]);
                    }

                    initialVal = containers[i].Value;
                }

                finalContainers = finalContainers.OrderBy(c => c.Id).ThenBy(c => c.Col).ToList();

                foreach (Container container in finalContainers)
                {
                    sb.Append(container.Value + "</br>");
                }

                result = sb.ToString();
            }

            return result;
        } 

This is the sample datatable I used for testing

            DataTable dtInput = new DataTable();
            dtInput.Columns.Add("id");
            dtInput.Columns.Add("name");
            dtInput.Columns.Add("value");
            dtInput.Columns.Add("value1");

            DataRow drInput1 = dtInput.NewRow();
            drInput1[0] = "1";
            drInput1[1] = "this is string 1";
            drInput1[2] = "abc";
            drInput1[3] = "This is sample";
            dtInput.Rows.Add(drInput1);

            DataRow drInput2 = dtInput.NewRow();
            drInput2[0] = "2";
            drInput2[1] = "this is string 2";
            drInput2[2] = "abc";
            drInput2[3] = "this is string 2";
            dtInput.Rows.Add(drInput2);

            DataRow drInput3 = dtInput.NewRow();
            drInput3[0] = "3";
            drInput3[1] = "this is string 3";
            drInput3[2] = "abc";
            drInput3[3] = "this is string 4";
            dtInput.Rows.Add(drInput3);

            DataRow drInput4 = dtInput.NewRow();
            drInput4[0] = "4";
            drInput4[1] = "this is string 3";
            drInput4[2] = "abc";
            drInput4[3] = "asasaasd";
            dtInput.Rows.Add(drInput4);

output will be

this is string 1</br>This is sample</br>this is string 2</br>*3 is duplicate of 4* this is string 3</br>this is string 4</br>*4 is duplicate of 3* this is string 3</br>asasaasd</br>