Null Object Pattern


Null Object Pattern

Motivation

The Null Object pattern can be used to avoid having checks that objects are not null before using them to avoid throwing NullReferenceExceptions, e.g.

if (myObject == null)
{
   return;
}
// do something with myObject…

Example

For example, say we have a device object with a Connect() method that a client wishes to call. If we're getting the device from a repository we have to do a check for null before calling Connect(). If however the repository returns the null object, no null check is required. The null check is done within the repository so the client code does have to worry about it.

Null object code

public interface IDevice
    {
        string Name { get; set; }

        void Connect ();
    }

    public class RealDevice : IDevice
    {
        public string Name { get; set; }

        public void Connect ()
        {
            Console.WriteLine("Device is connecting...");
        }
    }

    public class NullDevice : IDevice
    {
        public string Name
        {
            // Return something making it obvious this is the null object.
            get { return "null device"; }

            set { /* do nothing */ }
        }

        public void Connect ()
        {
            // Normally the method in the null object will just do nothing.
            Console.WriteLine("Device is null so do nothing");
        }

    }

Repository code

public interface IDeviceRepository
    {
        IDevice GetDeviceByName (string name);
    }

    public class DeviceRepository : IDeviceRepository
    {
        private List<IDevice> _devicesList;

        public DeviceRepository ()
        {
            _devicesList = new List<IDevice>
            {
                new RealDevice() { Name = "Camera" },
                new RealDevice() { Name = "Gripper" }
            };
        }

        public IDevice GetDeviceByName (string name)
        {
            IDevice device = _devicesList.SingleOrDefault(dev => dev.Name == name);

            // Do the null check here instead of in the client code.
            if (device == null)
            {
                device = new NullDevice();
            }

            return device;
        }
    }

Client code

class Program
    {
        private static IDeviceRepository _deviceRepository = new DeviceRepository();

        static void Main (string[] args)
        {
            // Valid will talk to real object.
            ConnectToDevice("Camera");

            // Invalid so will talk to null object.
            ConnectToDevice("any old name");

            Console.Read();
        }

        public static void ConnectToDevice(string deviceName)
        {
            IDevice device = _deviceRepository.GetDeviceByName(deviceName);
            device.Connect();
        }
    }

UML class diagram





Code

The example code is on Github.

Comments

Post a Comment

Popular posts from this blog

Path environmental variable on mac

To view your paths in terminal type: $ echo $PATH To add a path for the duration of this terminal session type: $ export PATH=$PATH:/my/new/path Note: the $PATH: is important as it adds the new path to your existing one. Without $PATH: the existing path is overwritten with the new one. The disadvantage with this approach is that you modification to the path is lost when the terminal session ends. To add to the path for all users permanently you need to edit the paths file in /etc.  In terminal navigate to /etc. Type $ sudo pico paths Add your path. Type ^x (note: ^ = ctrl key). You'll be prompted to save. Type Y . To view your paths using echo $PATH you'll need to reopen the terminal to see your changes.
Read more

Setting up node-sass on a mac

I followed the installation instructions for node-sass but it would not run from the terminal. The message was something like 'command not found'. Eventually I worked out that I needed to add  /Users/my_user_name/node_modules/node-sass/bin to the path. To do that see my post on setting the path on macs .
Read more