Question

I’ve a data table and interval size as 10

Input table:

 DateTime    ChNo   Data
--------------------------
6/10 10:10    1      0.1
6/10 10:20    1      0.1
6/10 11:05    2      0.1
6/10 11:06    2      0.1
6/10 11:07    2      0.1
6/10 11:08    2      0.1
6/10 11:09    2      0.1
6/10 11:10    2      0.1
6/10 11:11    2      0.1
6/10 11:12    2      0.1
6/10 11:13    2      0.1

I need to convert above table based on interval size, as i explained above if interval size is 10 my target table should look like

DateTime    ChNo   Data
--------------------------
6/10 10:10    1      0.1
6/10 10:20    1      0.1
6/10 11:10    2      0.6 -----> sum of intervals from 11:05 to 11:10
6/10 11:13    2      0.3 -----> sum of intervals from 11:11 to 11:13

What is the best way to get desired output like above using linqs

Thanks,
Murali.

Answer 1

I haven’t found the Linq solution but I can offer you a programmatic one. Note that I did this using full DateTime representation because of parsing instead of your that doesn’t have year. Check the code:

                //DataTable initialisation
                dt = new DataTable();
                dt.Columns.Add("Column1");
                dt.Columns.Add("Column2");
                dt.Columns.Add("Column3");


                dt.Rows.Add(new object[] { DateTime.Parse("6/10/2012 10:10"), 1, 0.1 });
                dt.Rows.Add(new object[] { DateTime.Parse("6/10/2012 10:20"), 1, 0.1 });
                dt.Rows.Add(new object[] { DateTime.Parse("6/10/2012 11:05"), 2, 0.1 });
                dt.Rows.Add(new object[] { DateTime.Parse("6/10/2012 11:06"), 2, 0.1 });
                dt.Rows.Add(new object[] { DateTime.Parse("6/10/2012 11:07"), 2, 0.1 });
                dt.Rows.Add(new object[] { DateTime.Parse("6/10/2012 11:08"), 2, 0.1 });
                dt.Rows.Add(new object[] { DateTime.Parse("6/10/2012 11:09"), 2, 0.1 });
                dt.Rows.Add(new object[] { DateTime.Parse("6/10/2012 11:10"), 2, 0.1 });
                dt.Rows.Add(new object[] { DateTime.Parse("6/10/2012 11:11"), 2, 0.1 });
                dt.Rows.Add(new object[] { DateTime.Parse("6/10/2012 11:12"), 2, 0.1 });
                dt.Rows.Add(new object[] { DateTime.Parse("6/10/2012 11:13"), 2, 0.1 });


//Method that groups data
    void GroupByDate()
    {
                DateTime startDate = DateTime.Parse(dt.Rows[0].ItemArray[0].ToString());
                DateTime endDate = DateTime.Parse(dt.Rows[dt.Rows.Count - 1].ItemArray[0].ToString());
                StringBuilder sb = new StringBuilder();
                int rowIndex = 0;
                for (DateTime d = startDate.AddMinutes(-(startDate.Minute - 1) % 10); rowIndex < dt.Rows.Count && d < endDate.AddMinutes(10 - endDate.Minute % 10); d = d.AddMinutes(10))
                {
                    double sum = 0;
                    DateTime lastDateInSequence = new DateTime();
                    for (DateTime md = d;rowIndex < dt.Rows.Count && md < d.AddMinutes(10); md = md.AddMinutes(1))
                    {
                        DateTime inbetween = DateTime.Parse(dt.Rows[rowIndex].ItemArray[0].ToString());
                        if ( inbetween == md)
                        {
                            sum += double.Parse(dt.Rows[rowIndex].ItemArray[2].ToString());
                            lastDateInSequence = md;
                            rowIndex++;
                        }
                    }
                    if (sum > 0.0)
                    {
                        // you can add this results to the new DataTable like dt1.Rows.Add(lastDateInSequence.ToString("dd/MM hh:mm"), dt.Rows[rowIndex - 1].ItemArray[1], sum);
                        sb.Append(lastDateInSequence.ToString("dd/MM hh:mm") + " " + dt.Rows[rowIndex - 1].ItemArray[1].ToString() + " " + sum.ToString() + Environment.NewLine);
                    }
                }
                MessageBox.Show(sb.ToString());
    }