instance factory methods Vs Static factory methods

Assuming that by "instance factory method" you're actually saying about the GoF "factory method", the term "static factory method" is described by Joshua Bloch in his book "Effective Java". Googling around I reached at these sites:

  • Factory Method: http://sourcemaking.com/design_patterns/factory_method
  • Static Factory Method: http://www.informit.com/articles/article.aspx?p=1216151

Hope that it helped to make the difference a little bit clearer.

Following Marvo's advice:

  • Factory Method as stated in GoF:

    define an interface for creating an object, but let subclasses decide which class to instantiate. Factory Method lets a class defer instantiation to subclasses.

Example (Java code):

    public abstract class Product { ... }

    public class ConcreteProduct extends Product { ... }

    public abstract class Creator {
      public void anOperation() { ... product = factoryMethod(); ... }
      public abstract Product factoryMethod();
    }

    public class ConcreteCreator extends Creator {
      public Product factoryMethod() { return new ConcreteProduct(); }
    }
  • Static Factory Method as stated in Effective Java:

    A class can provide a public static factory method, which is simply a static method that returns an instance of the class. (...) One advantage of static factory methods is that, unlike constructors, they have names. (...) A second advantage of static factory methods is that, unlike constructors, they are not required to create a new object each time they’re invoked. (...) A third advantage of static factory methods is that, unlike constructors, they can return an object of any subtype of their return type. (...) The main disadvantage of providing only static factory methods is that classes without public or protected constructors cannot be subclassed.

Example (Java code):

    public class Boolean {
      ...
      public static Boolean valueOf(boolean b) {
        return b ? Boolean.TRUE : Boolean.FALSE;
      }
      ...
    }

My current preference is to make factory methods not static for the sake of easier testing. You can't change a static factory method call at runtime, whereas if I could provide a factory implementation to the object then I could test it more thoroughly as I'm in more control of the context and object graph.


If you are returning a Singleton from your factory then you are going to need make sure you only have one instance, if you are going to create a new instance every time you call the factory then make it static.


It depends, for instance you can have a factory that will only produce a certain amount of objects in which case there will be an associated state. For the most part factory methods can be static as long as they don't rely on any non-static variables (such as non-static globals and such) for their creation. It also tends to differentiate different factories for different parts of an application so depending on whether there is a state or not in your factory is what you would go with it's not set in stone that all factory methods should be static, check what situation applies to you and write it appropriately.

Another consideration is the static keyword, this causes whatever is static to be instantiated only once in memory but a drawback is that it resides in the process memory always (which increases working set size). This may be something that you don't want as your factory might have very high locality in certain areas and it would otherwise just be using up memory somewhere else but this is usually an optimization issue that should be looked at only if the issue of memory pressure in your application arises.