Testing object-oriented industrial software without precise oracles or results

Research output: Journal Publications and Reviews (RGC: 21, 22, 62)21_Publication in refereed journalpeer-review

14 Scopus Citations
View graph of relations

Author(s)

  • T. H. Tse
  • Francis C. M. Lau
  • W. K. Chan
  • Peter C. K. Liu
  • Colin K. F. Luk

Detail(s)

Original languageEnglish
Pages (from-to)78-85
Journal / PublicationCommunications of the ACM
Volume50
Issue number8
Publication statusPublished - 1 Aug 2007
Externally publishedYes

Abstract

Software testing such as object-oriented software Testing At the Class and Cluster LEvels, or TACCLE, can be achieved by defining the test objectives, selecting and executing test cases, and checking results. The software specifications are extracted from technical drawings of mechanical and electronic hardware designed by process engineers using an in-house technique. TACCLE enables software engineers to test each individual class independently, then test the interaction among classes. The TACCLE methodology is described in three sections that include class-level testing, cluster-level testing, and test case generation and test script translation. The notion of testing observational equivalence and bypassing the need for oracles is important in test automation for industrial projects where it is impractical to define a precise relationship between specification and the software under test.

Citation Format(s)

Testing object-oriented industrial software without precise oracles or results. / Tse, T. H.; Lau, Francis C. M.; Chan, W. K. et al.

In: Communications of the ACM, Vol. 50, No. 8, 01.08.2007, p. 78-85.

Research output: Journal Publications and Reviews (RGC: 21, 22, 62)21_Publication in refereed journalpeer-review