Automatic test generation: a use case driven approach

Abstract

Use cases are believed to be a good basis for system testing. Yet, to automate the test generation process, there is a large gap to bridge between high-level use cases and concrete test cases. We propose a new approach for automating the generation of system test scenarios in the context of object-oriented embedded software, taking into account traceability problems between highlevel views and concrete test case execution. Starting from a formalization of the requirements based on use cases extended with contracts, we automatically build a transition system from which we synthesize test cases. Our objective is to cover the system in terms of statement coverage with those generated tests: An empirical evaluation of our approach is given based on this objective and several case studies. We briefly discuss the experimental deployment of our approach in the field at Thalè s Airborne Systems.

Model-Driven Engineering for Requirements Analysis

“…Besides analysis, other works aim at generating test cases from requirements. In particular, several approaches [4,1,13] use requirements expressed with extended use cases to generate test cases or at least test objectives. This kind of work shows the benefits that are obtained when the requirements take the form of use cases and emphasizes the need for validated use cases, for example using the simulation mechanism we propose.…”

“…The implemented test criterion is robustness (rejection paths), while we implement more possibilities (robustness and nominal behavior). The test criteria we have implemented are explained in [13].…”

Derivation of Test Objectives Automatically

“…The first group includes approaches that generate test objectives directly from use cases, like [2] and [11]. The second group includes approaches that generate a behavioural model from the use cases and derive test objectives from them, like [18], [13] or [17]. Some of the notations used in this group are: activity diagrams, state-machines diagrams, use case transitions systems or scenario trees.…”

“…In the first group, we find approaches that generate test objectives from isolated use cases, like [2], [11] or [17]. The second group includes approaches that generate test objectives from sequences of use cases, like [18] or [13]. However, none of the approaches of the first group might automatically derive behavioural models and test objectives from use cases.…”

“…We can also point out a lack of supporting tools. In the second group, the reference [13] generates test objectives in an automatic way for sequences of use cases, not for isolated use cases. These facts justified a new approach to generate a behavioural model and to derive test objectives in an automatic way.…”