[SOLVED] DBF Large Char Field

DBF Large Char Field

I have a database file that I beleive was created with Clipper but can't say for sure (I have .ntx files for indexes which I understand is what Clipper uses). I am trying to create a C# application that will read this database using the System.Data.OleDB namespace.

For the most part I can sucessfully read the contents of the tables there is one field that I cannot. This field called CTRLNUMS that is defined as a CHAR(750). I have read various articles found through Google searches that suggest field larger than 255 chars have to be read through a different process than the normal assignment to a string variable. So far I have not been successful in an approach that I have found.

The following is a sample code snippet I am using to read the table and includes two options I used to read the CTRLNUMS field. Both options resulted in 238 characters being returned even though there is 750 characters stored in the field.

Here is my connection string:

Provider=Microsoft.Jet.OLEDB.4.0;Data Source=c:\datadir;Extended Properties=DBASE IV;

Can anyone tell me the secret to reading larger fields from a DBF file?

using (OleDbConnection conn = new OleDbConnection(connectionString))
{
    conn.Open();

    using (OleDbCommand cmd = new OleDbCommand())
    {
        cmd.Connection = conn;
        cmd.CommandType = CommandType.Text;
        cmd.CommandText = string.Format("SELECT ITEM,CTRLNUMS FROM STUFF WHERE ITEM = '{0}'", stuffId);

        using (OleDbDataReader dr = cmd.ExecuteReader())
        {
            if (dr.Read())
            {
                stuff.StuffId = dr["ITEM"].ToString();

                // OPTION 1
                string ctrlNums = dr["CTRLNUMS"].ToString();

                // OPTION 2
                char[] buffer = new char[750];
                int index = 0;
                int readSize = 5;
                while (index < 750)
                {
                    long charsRead = dr.GetChars(dr.GetOrdinal("CTRLNUMS"), index, buffer, index, readSize);

                    index += (int)charsRead;

                    if (charsRead < readSize)
                    {
                        break;
                    }
                }
            }
        }
    }
}

Solution

You can find a description of the DBF structure here: http://www.dbf2002.com/dbf-file-format.html

What I think Clipper used to do was modify the Field structure so that, in Character fields, the Decimal Places held the high-order byte of the size, so Character field sizes were really 256*Decimals+Size.

I may have a C# class that reads dbfs (natively, not ADO/DAO), it could be modified to handle this case. Let me know if you're interested.