Refinement in statechart testing

Abstract

An existing statechart testing method developed by the authors allows a tester to show the equivalence of the behaviour of an implementation of some software system to its Harel statechart specification, by testing. This requires a rather large test set and satisfaction of certain requirements for both a specification and an implementation. Introduction of additional constraints permits reduction in a test set size, without weakening of the conclusions obtained by testing. This paper describes how a test set can be derived from a relatively big statechart and investigates the role of a few specific constraints in the reduction of test set size, depending on where in the system structure these constraints are used. The main contribution of the paper is a demonstration that enforcing relatively weak constraints in the right place of a system's structure makes it possible to reduce the size of a test set by many orders of magnitude. The practical use of the contribution is to help developers structure a system under construction in such a way that it is completely testable using the described testing method with few additional constraints. Copyright © 2004 John Wiley & Sons, Ltd.

Identification and Selection of Interaction Test Scenarios for Integration Testing

“…Approaches, in this level, are of ad-hoc nature [14] without associations between the testing levels. The test models are, for instance, generated from the system design specification which is described with Finite State Machines (FSM) [17], UML [3], or mathematical notations.…”