Evaluating cyclomatic complexity on functional JavaScript

Bugs in software is a very common problem, code reviews can help to catch bugs early on and detect which code is the most complex and may introduce bugs but when the code base is very large it can be costly to review all the code. Cyclomatic complexity can be used to give an indication of how complex the system source code is and help the developers to select which code they should review. But when measuring cyclomatic complexity on code written according to the functional paradigm, McCabe’s formula will not be sufficient since it is a formula most suitable for imperative code. Therefore we are making adaptations to a formula suited for pure functional languages in order to fit functional JavaScript. We are using an inductive empirical quantitative measurement method to calculate cyclomatic complexity on a directed graph implementation in order to define adaptations for functional JavaScript. Our results show a working adapted version of the formula. We have measured on a graph implemented in Haskell and on a corresponding functional JavaScript version which results in a cyclomatic complexity difference at only 0.375