Testing Compositionality

TL;DR

This paper develops algorithms for verifying and utilizing mutual acceptance to enable compositional testing of large systems by testing individual components and their environments, improving efficiency and scalability.

Contribution

It introduces three algorithms for applying mutual acceptance in model-based testing, including verification, exhaustive testing, and optimization techniques.

Findings

Verification of mutual acceptance guarantees system correctness.

Component testing reduces large system testing complexity.

Optimized algorithms improve testing efficiency in specific contexts.

Abstract

Compositionality supports the manipulation of large systems by working on their components. For model-based testing, this means that large systems can be tested by modelling and testing their components: passing tests for all components implies passing tests for the whole system. In previous work, we defined mutual acceptance for specification models and proved that this property is a sufficient condition for compositionality in model-based testing. In this paper, we present three main algorithms for using mutual acceptance in practice. First, we can verify mutual acceptance on specifications, proving compositionality for all valid implementations. Second, we give a sound and exhaustive model-based testing procedure which checks mutual acceptance on a specific black-box implementation. The result is that testing the correctness of large systems can be decomposed into testing the component implementations for uioco conformance to their specifications, and testing for environmental conformance to the specifications of their environment. Finally, we optimise this procedure further by utilizing the constraints imposed by multiple specifications at the same time. These three algorithms together allow picking the most suitable approach for a given situation, trading in more generalizable results for faster runtime by optimising for a specific context as desired.