The best way to show the Groovy compiler in action is to actually invoke it on some Groovy code.

Let's create a file named fizzbuzz.groovy and add the following code to it:

1.upto(100) {
   ans = ''
   if (it % 3 == 0) { ans ='Fizz' }
   if (it % 5 == 0) { ans += 'Buzz' }
   if (ans == '') {
     println it
   } else {
     println ans
   }
}

The code prints numbers from 1 to 100. For numbers that are multiples of three, it prints Fizz, and for the multiples of five, it prints Buzz. For numbers which are multiples of both three and five, it prints FizzBuzz. This little coding exercise was made popular by Jeff Atwood back in 2007 in his CODING HORROR blog http://www.codinghorror.com/blog/2007/02/why-cant-programmers-program.html.

Once the file is ready, let's make sure it prints Fizz and Buzz where it has to:

The reason groovyc produces two class files lies in the dynamic nature of Groovy and the usages of special constructs—closures—in our code snippet. The code block that appears after the each statement is actually a closure that can be stored and passed as a data structure in Groovy. Closures are discussed in more detail in the Defining code as data in Groovy recipe in Chapter 3, Using Groovy Language Features. Groovy creates a separate internal class for each such dynamic code block. The main script code is placed in fizzbuzz.class.

Also, as you probably noticed, we added all the Groovy distribution libraries to the classpath since compiled Groovy code relies on their functionality.

For single file compilation, groovyc is easy to use, though if you need to compile dozens or even hundreds of classes, then it would be better to use a build tool for that: