Automated class testing using threaded multi-way trees to represent the behaviour of state machines

  • Authors:

  • Bor-Yuan Tsai;Simon Stobart;Norman Parrington;Ian Mitchell

  • Affiliations:

  • School of Computing, Engineering and Technology, University of Sunderland, St. Peter's Way, Sunderland SR6 0DD, UK E-mail: bor‐yuan.tsai@sunderland.ac.uk;School of Computing, Engineering and Technology, University of Sunderland, St. Peter’s Way, Sunderland SR6 0DD, UK E-mail: bor‐yuan.tsai@sunderland.ac.uk;School of Computing, Engineering and Technology, University of Sunderland, St. Peter’s Way, Sunderland SR6 0DD, UK E-mail: bor‐yuan.tsai@sunderland.ac.uk;School of Computing, Engineering and Technology, University of Sunderland, St. Peter’s Way, Sunderland SR6 0DD, UK E-mail: bor‐yuan.tsai@sunderland.ac.uk

  • Venue:
  • Annals of Software Engineering
  • Year:
  • 1999

Quantified Score

Hi-index 0.00

Visualization

Abstract

Extensive test data is required to demonstrate that “few” errors exist in software. If the process of software testing could be carried out automatically, testing efficiency would increase and the cost of software development would be significantly reduced. In this paper, a tool for detecting errors in object oriented classes is proposed. The approach uses a state‐based testing method. The method utilises state machines in order to produce threaded multi‐way trees, which are referred to as inspection trees. Inspection trees can be used to generate test cases and parse test results files. This allows us to determine whether the classes under test contain errors. The algorithms for the creation of inspection trees and the examination of the test results file using an inspection tree are described in the paper.