Question

I previously asked the question and got answer to Best approach to write query but the problem is that if you have to save this result in a list then there duplication of records. For example
the resultant table of the join given EXAMPLE

See there are duplicate rows. How can you filter them out, and yet save the data of order number?
Of course there may be some ways but I am looking for some great ways

How can we store the data in list and not create duplicate rows in list?

My current code for my tables is

  int lastUserId = 0;
  sql_cmd = new SqlCommand();
  sql_cmd.Connection = sql_con;

  sql_cmd.CommandText = "SELECT * FROM AccountsUsers LEFT JOIN Accounts ON AccountsUsers.Id = Accounts.userId ORDER BY AccountsUsers.accFirstName";
  SqlDataReader reader = sql_cmd.ExecuteReader();

  if (reader.HasRows == true)
  {
      Users userToAdd = new Users();

      while (reader.Read())
      {
           userToAdd = new Users();
           userToAdd.userId = int.Parse(reader["Id"].ToString());
           userToAdd.firstName = reader["accFirstName"].ToString();
           userToAdd.lastName = reader["accLastName"].ToString();
           lastUserId = userToAdd.userId;

           Websites domainData = new Websites();
           domainData.domainName = reader["accDomainName"].ToString();
           domainData.userName = reader["accUserName"].ToString();
           domainData.password = reader["accPass"].ToString();
           domainData.URL = reader["accDomain"].ToString();
           userToAdd.DomainData.Add(domainData);

           allUsers.Add(userToAdd);
       }
    }

For second table I have custom list that will hold the entries of all the data in second table.

The table returned is table having joins and have multiple rows for same

Answer 1

Besides using the Dictionary idea as answered by Antonio Bakula…

If you persist the dictionary of users and call the code in your sample multiple times you should consider that a user account is either new, modifed, or deleted.

The algorithm to use is the following when executing your SQL query:

1. If row in query result is not in dictionary create and add new user to the dictionary.
2. If row in query result is in dictionary update the user information.
3. If dictionary item not in query result delete the user from the dictionary.

I’d also recommend not using SELECT *
Use only the table columns your code needs, this improves the performance of your code, and prevents a potential security breach by returning private user information.