how can i correctly parse a text file delimited by white space

Question

Below is my sample text file

{

Here is my schema file<

Answer 1

I would probably do this a different way. I would use a StreamReader, and read in the file line by line, break the string up into object properties, and store the objects in a list. Then you bind the list to the datagridviews datasource. I demonstrate two quick ways to do this.

1 -Tab separated data

If the file is tab separated, as it seems to be, split the line into an array and assign each index with to a property like so.

public partial class Form1 : Form
{
    private void Form1_Load(object sender, EventArgs e)
    {
        var rows = new List<Row>();
        var sr = new StreamReader(@"C:\so_test.txt");
        while (!sr.EndOfStream)
        {
            string s = sr.ReadLine();
            if (!String.IsNullOrEmpty(s.Trim()))
            {
                rows.Add(new Row(s));
            }
        }
        sr.Close();
        dataGridView1.DataSource = rows;
    }
}

public class Row
{
    public double Number1 { get; set; }
    public double Number2 { get; set; }
    public double Number3 { get; set; }
    public double Number4 { get; set; }
    public double Number5 { get; set; }
    public double Number6 { get; set; }
    public double Number7 { get; set; }
    public string Date1 { get; set; }

    public Row(string str)
    {
        string[] separator = { "\t" };
        var arr = str.Split(separator, StringSplitOptions.None);
        Number1 = Convert.ToDouble(arr[0]);
        Number2 = Convert.ToDouble(arr[1]);
        Number3 = Convert.ToDouble(arr[2]);
        Number4 = Convert.ToDouble(arr[3]);
        Number5 = Convert.ToDouble(arr[4]);
        Number6 = Convert.ToDouble(arr[5]);
        Number7 = Convert.ToDouble(arr[6]);
        Date1 = arr[7];
    }
}

2 -Hard Start points and lengths

If the data is tab separated, but conforms to strict start and endpoints for each column, you could declare the startpoints and lengths for each column as constants and get those via substring. This would only need a change in code in your Row class, like this. I have left of the constants from brevity, and just hardcoded them.

    public Row(string str)
    {
        Number1 = Convert.ToDouble(str.Substring(4, 6));
        Number2 = Convert.ToDouble(str.Substring(16, 6));
        Number3 = Convert.ToDouble(str.Substring(28, 7));
        Number4 = Convert.ToDouble(str.Substring(40, 7));
        Number5 = Convert.ToDouble(str.Substring(52, 6));
        Number6 = Convert.ToDouble(str.Substring(64, 6));
        Number7 = Convert.ToDouble(str.Substring(76, 6));
        Date1 = str.Substring(88, 24);
    }

Answer 2

This modified schema.ini file does seem to solve the problem.

[Sample File.txt]
Format=FixedLength
Format=TabDelimited
MaxScanRows=25
CharacterSet=ANSI
Col1=Col1 Memo Width 10
Col2=Col2 Memo Width 15
Col3=Col3 Memo Width 11
Col4=Col4 Memo Width 12
Col5=Col5 Memo Width 10
Col6=Col6 Memo Width 11
Col7=Col7 Memo Width 150