There is a related question to this:
What's the best method to pass parameters to SQLCommand?
But I am wanting to know what the differences are and if there are any problems with the different ways.
I usually use a structure something like this:
using (SqlConnection conn = new SqlConnection(connectionString))
using (SqlCommand cmd = new SqlCommand(SQL, conn))
{
cmd.CommandType = CommandType.Text;
cmd.CommandTimeout = Settings.Default.reportTimeout;
cmd.Parameters.Add("type", SqlDbType.VarChar, 4).Value = type;
cmd.Connection.Open();
using (SqlDataAdapter adapter = new SqlDataAdapter(cmd))
{
adapter.Fill(ds);
}
//use data
}
Now there are several ways to add the cmd parameters and I am wondering which is best:
cmd.Parameters.Add("@Name", SqlDbType.VarChar, 20).Value = "Bob";
cmd.Parameters.Add("@Name", SqlDbType.VarChar).Value = "Bob";
cmd.Parameters.Add("@Name").Value = "Bob";
cmd.Parameters.AddWithValue("@Name", "Bob");
Having the length of the field in the passing of the varchars I assume is not preferable as it's a magic value which may be changed later at the database. Is this correct? Does it cause any issue passing a varchar in this way (performance or other), I assume it defaults to varchar(max) or the database equivalent. I am reasonably happy this will work.
The part that concerns me more is the loss of the SqlDbType enum if I am using the third or fourth options I listed above I am not supplying a type at all. Are there cases where this won't work I can imagine problems with varchar being incorrectly cast to char or vice versa or perhaps issues with decimal to money....
In terms of the database the field type I would say is much less likely to change than the length so is it worth retaining?
SqlDataAdapter.Fill()
, and that there is benefit in not doing so? – Johnnajohnnie