The First one (Code Type 1) is just a basic insertion using SqlCommand and calling stored procedure and calling that insert method under Insert Click Event Handler.
The First one
Code Type 1:
public string InsertUserInformation()
{
SqlConnection con = new SqlConnection(ConnectionString);
con.Open();
SqlCommand cmd = new SqlCommand("sp_userinformation", con);
cmd.CommandType = CommandType.StoredProcedure;
try
{
cmd.Parameters.AddWithValue("@UserName", TextBox1.Text);
cmd.Parameters.AddWithValue("@Password", TextBox2.Text);
cmd.Parameters.AddWithValue("@FirstName", TextBox3.Text);
cmd.Parameters.AddWithValue("@LastName", TextBox4.Text);
cmd.Parameters.AddWithValue("@Email", TextBox5.Text);
cmd.Parameters.AddWithValue("@PhoneNo", TextBox6.Text);
cmd.Parameters.AddWithValue("@Location", TextBox7.Text);
cmd.Parameters.AddWithValue("@Created_By", TextBox8.Text);
cmd.Parameters.Add("@ERROR", SqlDbType.Char, 500);
cmd.Parameters["@ERROR"].Direction = ParameterDirection.Output;
cmd.ExecuteNonQuery();
string strMessage = (string)cmd.Parameters["@ERROR"].Value;
con.Close();
return strMessage;
}
catch (Exception ex)
{
throw ex;
}
finally
{
cmd.Dispose();
con.Close();
con.Dispose();
}
}
protected void btnSubmit_Click(object sender, EventArgs e)
{
InsertUserInformation();
}
The Second one(Code Type 2) :
http://www.aspdotnet-suresh.com/2010/05/introduction-to-3-tier-architecture-in_17.html
public string InsertUserInformation(BEL objBELUserDetails)
{
SqlConnection con = new SqlConnection(ConnectionString);
con.Open();
SqlCommand cmd = new SqlCommand("sp_userinformation", con);
cmd.CommandType = CommandType.StoredProcedure;
try
{
cmd.Parameters.AddWithValue("@UserName",objBELUserDetails.UserName);
cmd.Parameters.AddWithValue("@Password", objBELUserDetails.Password);
cmd.Parameters.AddWithValue("@FirstName", objBELUserDetails.FirstName);
cmd.Parameters.AddWithValue("@LastName", objBELUserDetails.LastName);
cmd.Parameters.AddWithValue("@Email", objBELUserDetails.Email);
cmd.Parameters.AddWithValue("@PhoneNo", objBELUserDetails.Phoneno);
cmd.Parameters.AddWithValue("@Location", objBELUserDetails.Location);
cmd.Parameters.AddWithValue("@Created_By", objBELUserDetails.Created_By);
cmd.Parameters.Add("@ERROR", SqlDbType.Char, 500);
cmd.Parameters["@ERROR"].Direction = ParameterDirection.Output;
cmd.ExecuteNonQuery();
string strMessage = (string) cmd.Parameters["@ERROR"].Value;
con.Close();
return strMessage;
}
catch (Exception ex)
{
throw ex;
}
finally
{
cmd.Dispose();
con.Close();
con.Dispose();
}
}
this is also an insertion but uses BEL,BLL,DAL and finally the BLL layer's insert_method is called under button insert event handler of PL.
As far as i know, 3 tier helps to organize the code into different layers , so that maintenance would be easier and other x factors are there.
But, the First One(Code Type 1) seems to be much simpler to me.
So, please advise me ,
1. why would they refer the Second one, when First one is more simple and easy?
2. what about Performance among two, which one is faster?
Thanks in advance.
The first example tightly couples data storage code with UI code. This code can't be used elsewhere in the same project (or other projects) without copying/pasting followed by editing of the textbox names. If there is a new required parameter added to sp_userinformation, and this code has been copied in several places (or other projects), then each usage must be found and updated. The compiler can't help with missing parameters.
Compare this with the second example which uses an object (sometimes referred to as a data transfer object). The second example doesn't directly refer to user interface objects and therefore can be isolated in a seperate class. This same code could be called from other parts of the application or from another application encouraging code reuse. If a new required parameter is added to sp_userinformation then this could also be added to the objBELUserDetails constructor. If some other code doesn't pass in the required parameter with the constructor then the code will not compile. This allows for bugs to be found before deployment. There are many other reasons but you've already hit on some of them, organization, ease of maintenance. Too numerous to discuss here.
As for performance, the first code is likely more perfomant as it simply makes a database call. However, performance typically doesn't take a big hit passing around data transfer objects. It probably takes far longer for the database call to complete than any of the c# code. But, you can always measure to make sure.
Anyways, hope this helps!