将固定宽度的文本文件逐行读取到数据表中-C＃

Question

I have a large fixed width text file, I want to read it row by row and insert into a datatable, or just read it into a datatable. how can I assign that from which to which position one column will be?

I also have to perform some action like trim the extra space of columns for every record.

I tried this code but it gives me datatable with just 3 columns instead of 15

string sourcePath = @"c:\\";
string filename = "file.txt";

DataTable dt;

//Create OleDb connection object
using (OleDbConnection cn = new OleDbConnection(@"Provider=Microsoft.Jet.OleDb.4.0;" +
            "Data Source=" + sourcePath + ";Extended Properties=\"Text;\""))
{
    // Open connection 
    cn.Open();

    // Create OleDb Adapter object 
    using (OleDbDataAdapter adapter = new OleDbDataAdapter("SELECT * FROM " + filename, cn))
    {
        dt = new DataTable("Records");
        adapter.Fill(dt);

        // Display results
        dataGridView1.DataSource = dt;
    }
}

Sample Data

Sample Name.......~Address~City~State~12345 Contact Name 123-456-7890 Type Payment Sample Name.......~Address~City~State~12345 Contact Name 123-456-7890 Type Payment Sample Name.......~Address City~State~12345 Contact Name 123-456-7890 Type Payment Sample Name.......~Address City~State~12345 Contact Name 123-456-7890 Type Payment

There is a lot spaces in between..

Answer 1

If you want to read your text file using OleDb Text Driver you need to provide a Schema.ini file that describe the structure of one of the "records" (a line) of your text file

So supposing that your text file contains records in this format:

XXXXXYYYYYKKKKKZZZZ

(Four fields, each of 5 char) you need to add in the same folder where is located your text file a Schema.ini file with this structure

[file.txt]
Format=FixedLength
Col1=CustomerNumber Text Width 5
Col2=CustomerName Text Width 5
Col3=CustomerCity Text Width 5
Col4=CustomerZIP Text Width 5

You could find more detail at the MSDN page linked above

Without using OleDb you need to use something already available in the NET Framework, or write your own splitting code, or use a thirdy party library

For your own code approach I would start defining a class for your record.
For example an Customer class could be

public class Customer
{
    public string Name {get;set;}
    public string Address {get;set;}
    ... other fields omitted ....
}

then you define a List<Customer> and start reading your file with

List<Customer> customers = new List<Customer>();
foreach(string line in File.ReadLines(filename)
{
    Customer c = new Customer();
    c.Name = line.Substring(0, 10).Trim();
    c.Address = line.Substring(10, 15).Trim();
    ... and so on for all the other fields....

    customers.Add(c);        
}

At the end of the loop you have your list filled with your data. Of course if you just need to write everything in another table on a database server you could write the code that inserts/updates the database table directly inside the loop and avoid the loading of the list.