Simple DataReader in C#

If you are an experienced ADO.NET developer then this article will be a complete bore for you. But, believe it or not, there are people who are trying to learn how to work with databases in C#. So maybe I can help out at least one person!
What is an SQL DataReader? DataReaders are a fast way to pull records from a database when all you want to do is simply READ. You may have heard the term "Firehose Cursor" used to describe a DataReader. A firehose is a good comparison because the water (data) only flows one way and it flows fast. DataReaders can not be used to update data, delete data, or anything else other than reading. A good example of when to use a DataReader would be cities in a state. You may want to read out all cities in New York and since they aren't exactly changing every day, you would want to pull them down as fast as possible.

Ok, I promised fast and easy so here goes.
First, you must instantiate (create) a new database connection. Now, I am only working with Microsoft's SQL server today. If you need help converting this article to other database platforms like Oracle or MySQL then please let me know.

Make sure you are also using the needed namespaces before you begin.

using System.Data;
using System.Data.SqlClient;

SqlConnection adoConn = new SqlConnection("Data Source=server;Initial Catalog=database;Persist Security Info=True;User ID=username;Password=password");

Database is now created and opened. The string that we passed is called the "Connection String". All it does is tell the database how and where to open the connection. Substitute "server", "Initial Catalog", and "User ID/Password" with your database information. Remember, this is ONLY an open connection. The database is sitting there waiting on a command. And that is exactly what we setup next. A command. Think of a command as a direct order you give the server (even though it may or may not listen!).

// new command
string sql = "SELECT CustomerName FROM MyTable";
SqlCommand adoCmd = new SqlCommand(sql, adoConn);

The sql string is simply a SQL command we are passing. The adoConn is telling the command which connection to use. Simple, huh?

Ok, now we have an open connection and a command (using the sql string). Our next move is to create the DataReader and display some data.

SqlDataReader adoDR = adoCmd.ExecuteReader();

if (adoDR.HasRows)
while (adoDR.Read())

The ExecuteReader() method sends the SQL data from the command (our SELECT statement) and if there are records, brings them one at a time down to the DataReader (adoDR).

You'll notice that we first called the .HasRows condition. It's always good to first make sure there is data returned before you do anything with it. The next statement might look a little confusing. This while loop brings each record down one at a time. See, when you call the ExecuteReader and assuming there are rows, you actually start at position "-1". Strange, huh? For example, let's say that SELECT statement returned 50 rows of data. The first record number would be 0, the next would be 1, then so on until record 49. 0-49 records. Everytime you call the .Read() on the DataReader, you advance a record. So, if you started at -1 and advanced a record you would be at the beginning. Record 0. Calling .Read() will continue to return TRUE until you reach the last record. So as you can see, this makes it convenient to cycle through all records. Also I should mention you HAVE to call it at least once to advance to the first record.

The Response.Write command simply sends the data to the web page. This could have been Console.WriteLine, etc. Notice how the "CustomerName" was used. Be careful here because you want to make sure you don't try to call a field in a table that you didn't SELECT.

Ok, the last thing to do is close connections and dispose so that we don't create memory leaks on the server.


Noticed I reversed the order that I used when creating the objects. DataReaders are opened when you call the ExecuteReader() and when you open something, you should close it. Calling .Dispose() on these objects would also close them but closing them myself has always been a habbit of mine. Command objects aren't opened or closed so no Close() is needed. And finally we close/dispose of the database connection.

There. Was that so hard? We created a database connection, opened it, created a command (using a custom SQL query) and executed the DataReader. Then, we looped through the records. Finally, we closed and disposed of all the objects.

There you have it. Simple. ADO.NET has made it really easy to display data. This is just a tiny scratch on the Titanic. ADO.NET could fill 50,000 pages!

I hope you enjoyed this article. I have to admit, I'm not much of a writer but I remember the first time I pulled data from a database and I wished I had someone telling me in plain English how to get right to the point.
Obviously, we didn't cover other topics like error trapping, DataGrids, DataSets, etc. Those will come in time!

No comments: