Groovy Tutorial for Java Developers – Part 2: Closures

Groovy has support for closures, which work much like Java 8 lambdas. A closure is an anonymous block of executable code (let’s say a function) which can be passed to variables and has access to data in the context where it was defined. Let’s look at this in a bit more detail. To define a closure and assign it to a variable you can write as follows:

def helloWorld = {
  println "Hello World"
}

The variable helloWorld now holds a reference to the closure and you can execute it by “calling the variable”:

helloWorld()

Parameters

You can also expect parameters in closures and pass them when calling:

def power = { int x, int y ->
  return Math.pow(x, y)
}
println power(2, 3) // Will print 8.0

Type definition of parameters is the same like variables. If you define a type you can only use this type, but you can also skip the type of parameters and pass in anything you want:

def say = { what ->
  println what
}
say "Hello World"

If you don’t specify any parameters (and no arrow) the closure can accept one parameter which is available by the name it inside of the closure. That way you can make often used one parameter closures even shorter, e.g. you can shorten the above to the following (which will produce exactly the same result):

def say = { println it }
say "Hello World"

If you don’t pass the parameter, it will just be null. If you really want to create a closure that won’t be able to take an argument, you have to write the following:

def clos = { ->
  println "This closure does not take any arguments."
}

Optional Return

In Groovy the last statement of a method block (or closure) is returned implicitly without you needing to write the return statement. This can be useful to reduce boilerplate code to a minimum. A closure that calculates the square of a number, can be shortened as follows:

def square = { it * it }
println square(4) // Will print out 16

This closure makes usage of the implicit parameter it and the optional return statement.

Passing Closures around

The power of being able to assign closures to variable is that you can also pass them around to methods. Let’s write a closure (you could of course do the same to class methods) which will expect a closure as a parameter and a string:

def transform = { str, transformation ->
  transformation(str)
}

Your transform function now can be called with a string and a closure. The transform function will just call the passed closure with the passed string (what makes it a pretty useless example).

println transform("Hello World", { it.toUpperCase() })

This will print out HELLO WORLD, since it will call the closure with Hello World as a parameter. The closure itself just transforms it one parameter it to upper case and implicitly return it.

Because Groovy brims with syntactic sugar, it is allowed to close the method parentheses and place the closure behind it:

println transform("Hello World") { it.toUpperCase() }

This syntax should increase readability (especially when the closure does not simply fit into one line. This optimization also works when the closure is the only parameter to the method. You can still place the closure behind the parantheses (or just skip the parantheses, since they are anyway optional in the most cases).

Closing of variables

Everything we’ve seen so far is specific to anonymous first-class functions (often called lambda functions), but what is the specific about closures?

Closures have access to the variables in the context they were defined. Let this demonstrate with an easy sample:

def createGreeter = { name ->
  return {
    def day = new Date().getDay()
    if (day == 0 || day == 6) {
      println "Nice Weekend, $name"
    } else {
      println "Hello, $name"
    }
  }
}
def greetWorld = createGreeter("World")
greetWorld()

We call the createGreeter closure and pass the name World to it. This closure will basically return a new closure, which we store in the variable greetWorld. Whenever we call greetWorld now we will get a nice message depending on whether it’s weekend or not. We never passed the content of the name variable to the inner closure, that we return and store in greetWorld. We still can access it from inside the closure, because that is what being a closure means. It has access to the variables of the context it was defined in. In this case it was defined somewhere where a variable name exists. The closure can use this variable even once it got returned and called somewhere completely different.

Please note that the usage of Date.getDay() is deprecated in Java and I just used it for the sake of readability in this snippet!

What’s next?

Besides the basics we covered here, Groovy offers support for currying and memoization. To read more about Groovys closure head over to the official docs.

The next part of this tutorial covers usage of collection types like lists and maps in Groovy.