recursion

A few weeks ago an interesting question was asked on the StackOverflow. Someone experimented with a recursion in Groovy and stepped into Closure.trampoline() ^[1]. It quickly turned out that using TrampolineClosure makes a recursive execution slower. Is this a valid behavior, or do we do something wrong?

Groovy has many useful functions built-in, and one of them is Iterable.combinations() that takes aggregated collections and finds all combinations of items. However, if we take a look its source code, we will find out that it was implemented using very imperative approach (nested for-loops + some if-statement). In this blog post I will show you how to implement the same function using Groovy and tail-recursion algorithm. Enjoy!

Most of the object-oriented programmers prefer constructing algorithms using imperative style over using recursion. This is pretty obvious in the JVM ecosystem, where imperative iteration is much more efficient than recursive function call chain. However, what if I tell you that in Groovy you can take advantage of clean tail-recursive functions without sacrificing performance? Interested? Let’s deep dive into it.