AmPyfier: Test Amplification in Python

TL;DR

AmPyfier introduces a novel approach to test amplification tailored for Python, a dynamically typed language, demonstrating its effectiveness on open-source projects and expanding testing capabilities in such languages.

Contribution

This paper presents AmPyfier, the first tool to enable test amplification in Python, overcoming challenges posed by dynamic typing and broadening the applicability of test amplification techniques.

Findings

Successfully strengthened 70% of tested classes

Proved feasibility of test amplification in Python

Extended testing methods to dynamically typed languages

Abstract

Test Amplification is a method to extend handwritten tests into a more rigorous test suite covering corner cases in the system under test. Unfortunately, the current state-of-the-art for test amplification heavily relies on program analysis techniques which benefit a lot from explicit type declarations present in statically typed languages like Java and C++. In dynamically typed languages, such type declarations are not available and as a consequence test amplification has yet to find its way to programming languages like Python, Ruby and Javascript. In this paper, we present AmPyfier, a proof-of-concept tool, which brings test amplification to the dynamically typed, interpreted language Python. We evaluated this approach on 7 open-source projects, and found that AmPyfier could successfully strengthen 7 out of 10 test classes (70%). As such we demonstrate that test amplification is feasible for one of the most popular programming languages in use today.