Lesson Weekend

Now that we've set up a database connection, we're ready to begin interacting with our database in our application.

In this lesson, we'll use MySQL and C# to turn information from a database into C# objects our application can work with. Specifically, we'll rewrite our GetAll() method to return all the rows from our items table.

Because our database will persist Items, we no longer need to store Items in the static list _instances. We'll start by removing all references to _instances from our code.

First, we'll remove them from our class declaration and constructor:

ToDoList/Models/Item.cs
...
namespace ToDoList.Models
{
  public class Item
  {
    public string Description { get; set; }
    public int Id { get; }

    public Item(string description)
    {
        Description = description;
    }

    ...
  }
  ...
}

Next, we'll remove _instances from the Find() and ClearAll() methods. Because we'll get a compiler error if Find() doesn't return an Item, we'll update the method to return a placeholder item for now. We'll refactor the method to return an actual Item from the database in a future lesson.

ToDoList/Models/Item.cs
...

  public static Item Find(int searchId)
  {
    // Temporarily returning placeholder item to get beyond compiler errors until we refactor to work with database.
    Item placeholderItem = new Item("placeholder item");
    return placeholderItem;
  }

...

Our ClearAll() method is void and doesn't return anything so it can just be empty for now:

ToDoList/Models/Item.cs
public static void ClearAll()
{
}

Now we're ready to rewrite our existing GetAll() method.

Returning All Records From the Database


Our new GetAll() method will present a lot of new concepts and boilerplate code, so it may feel overwhelming at first. We'll start by displaying the method in its entirety. Then we'll walk through the method line by line.

When we begin to use Entity Framework Core in this section, our lengthy boilerplate code (as seen below) will be abstracted into method calls to built-in .NET classes. So, the goal of writing this boilerplate code now is to demonstrate what is actually happening when a .NET application interacts with a database. While we will be writing our methods this way for the first class session, this will not be expected for this section's independent project.

Here's our new full GetAll() method:

ToDoList/Models/Item.cs
...
using MySql.Data.MySqlClient;
...

    public static List<Item> GetAll()
    {
        List<Item> allItems = new List<Item> { };
        MySqlConnection conn = DB.Connection();
        conn.Open();
        MySqlCommand cmd = conn.CreateCommand() as MySqlCommand;
        cmd.CommandText = "SELECT * FROM items;";
        MySqlDataReader rdr = cmd.ExecuteReader() as MySqlDataReader;
        while (rdr.Read())
        {
            int itemId = rdr.GetInt32(0);
            string itemDescription = rdr.GetString(1);
            Item newItem = new Item(itemDescription, itemId);
            allItems.Add(newItem);
        }
        conn.Close();
        if (conn != null)
        {
            conn.Dispose();
        }
        return allItems;
    }
...

That's a lot of new code! The first and last line should be familiar. We instantiate a new empty List and return that List after it's been populated. Before we cover the rest of the code line-by-line, here's a quick overview of what this method does:

  • Open a database connection;
  • Construct a SQL query;
  • Return the query results from the database;
  • Close the connection.

Before we continue, note that we also need to add using MySql.Data.MySqlClient; to the top of our file.

Opening a Database Connection

Each time we make a query, we need to open a new database connection:

MySqlConnection conn = DB.Connection();
conn.Open();

This part is relatively simple. We instantiate a new DB.connection() in a variable called conn. Then we Open() it. Our application will throw an exception if we try to make a SQL query without first opening a database connection.

Construct a SQL Query

Once our connection is open, we can construct our SQL query:

MySqlCommand cmd = conn.CreateCommand() as MySqlCommand;
cmd.CommandText = "SELECT * FROM items;";

When we make a SQL query in our application, it's not just a string of text. The query needs to be stored in a special object called a MySqlCommand.

In order to do this, we call the createCommand() method on our conn object. We include the expression as MySqlCommand at the end of this line. Using as creates an expression that casts cmd into a MySqlCommand object.

This casting is important because there are many different types of SQL databases and many different types of objects that can interact with them. Because our connection is a MySqlConnection type object, we cast it to send a corresponding MySqlCommand to the database.

Next, we'll add the actual text of our SQL command:

cmd.CommandText = "SELECT * FROM items;";

Remember that cmd is a MySqlCommand object. A MySQLCommand object has a number of different properties we can set. We won't cover most of them, but the CommandText property is essential because it's where we'll store our actual SQL query.

We set this property value to SELECT * FROM items;.

Returning Results from the Database

Next, we need to create a Data Reader Object. It will be responsible for reading the data returned by our database in response to the SELECT * FROM items; command:

MySqlDataReader rdr = cmd.ExecuteReader() as MySqlDataReader;

We'll cast its type for use with MySQL just like we did with conn. We call this Data Reader rdr and use the as keyword to cast it into a MySqlDataReader object.

The rdr object represents the actual reading of the database. However, we will need to call other methods on the rdr object in order to display the results of the query in our application:

while(rdr.Read())
{
  // Code temporarily omitted.
}

A MySqlDataReader object has a built-in Read() method that reads results from the database one at a time and then advances to the next record. This method returns a boolean. If the method advances to the next object in the database, it returns true. If it reaches the end of the records that our query has returned, it returns false and our while loop ends.

In the while loop, we'll take each individual record from our database and translate that record into an Item object our application understands:

    while(rdr.Read())
    {
      int itemId = rdr.GetInt32(0);
      string itemDescription = rdr.GetString(1);
      Item newItem = new Item(itemDescription, itemId);
      allItems.Add(newItem);
    }

Our MySQLDataReader rdr object has many methods available to it. Many of these methods are specifically for extracting data from a record. GetInt32() returns a 32 bit integer. GetString() is self-explanatory.

We also pass in a number value as an argument to both methods. This is because rows from the database are returned by the rdr.Read() method as indexed arrays. Let's use the following table as an example to demonstrate:

 id | description
 ---+---------
 1  | Mow the lawn
 2  | Walk the dog
 3  | Make dinner

When the reader object returns the first entry in this example database, it'll look like this:

{ 1, "Mow the lawn" };

The second object will look like this:

{ 2, "Walk the dog" };

The id column is at index 0 while the description column is at index 1. If we had a third column, it'd be at index 2.

In our while loop above, we define our itemId as rdr.GetInt32(0); because this will return the integer at the 0th index of the array returned by the reader. Similarly, we define itemDescription as rdr.GetString(1) because our item description will be at the 1st index of the array returned by the reader.

Once we've collected the data, we can use it to instantiate new Item objects and add them to our allItems list. Now each row in our database is an Item stored in a List that our application understands.

Closing the Connection

Communicating with a database is a resource-intensive process. For this reason, it's important to close our database connection when we're done. This allows the database to reallocate resources to respond to requests from other users. We can use a built-in Close() method to do this.

conn.Close();

if (conn != null)
{
  conn.Dispose();
}

The Close() method is self-explanatory. We also include a conditional because on rare occasions, our database connection will fail to close properly. It's considered best practice to confirm it's fully closed. That's why we put the Dispose() method inside a conditional. This method will only run if conn is not null.

Overloading the Item Constructor


Now that we've added a GetAll() method, we'll need to make one more change to our code so our application can properly interact with our new method. Our GetAll() method includes the following line:

Item newItem = new Item(itemDescription, itemId);

Currently, our constructor only accepts description as an argument. Whenever we create a new object in our application, it should only have a description. However, when we retrieve a record from the database, we want its id, too. We can add an overloaded constructor so our application can instantiate an Item either way:

ToDoList/Models/Item.cs
        public Item(string description, int id)
        {
            Description = description;
            Id = id;
        }

Summary


We've covered a lot of ground here. In the process of rewriting our GetAll() method, we needed to open a database connection, create a special object to hold our SQL query, and then create another special object to actually read through the results. Finally, we had to translate our database rows into C# objects and close the connection. It's a lot of work for a simple query. Entity will make our lives much easier later this section, but for now, it's important to solidify our understanding of how our application actually queries a database.

At this point, our to do list has the code necessary to query and return database-backed information. However, we have some more work to do before our database functionality is fully implemented. In the next lesson, we'll add functionality for testing our new database-backed methods.

Repository Reference

Follow the link below to view how a sample version of the project should look at this point. Note that this is a link to a specific commit in the repository.

Example GitHub Repo for To Do List

Lesson 6 of 36
Last updated February 24, 2022