e-Informatica Software Engineering Journal Automatic Test Cases Generation from Software Specifications

Automatic Test Cases Generation from Software Specifications

Aysh Alhroob, Keshav Dahal, Alamgir Hossain
Abstract
A new technique is proposed in this paper to extend the Integrated Classification Tree Methodology (ICTM) developed by Chen et al. [13] This software assists testers to construct test cases from functional specifications. A Unified Modelling Language (UML) class diagram and Object Constraint Language (OCL) are used in this paper to represent the software specifications. Each classification and associated class in the software specification is represented by classes and attributes in the class diagram. Software specification relationships are represented by associated and hierarchical relationships in the class diagram. To ensure that relationships are consistent, an automatic methodology is proposed to capture and control the class relationships in a systematic way. This can help to reduce duplication and illegitimate test cases, which improves the testing efficiency and minimises the time and cost of the testing. The methodology introduced in this paper extracts only the legitimate test cases, by removing the duplicate test cases and those incomputable with the software specifications. Large amounts of time would have been needed to execute all of the test cases; therefore, a methodology was proposed which aimed to select a best testing path. This path guarantees the highest coverage of system units and avoids using all generated test cases. This path reduces the time and cost of the testing.
[1]Aysh Alhroob, Keshav Dahal and Alamgir Hossain, "Automatic Test Cases Generation from Software Specifications", In e-Informatica Software Engineering Journal, vol. 4, no. 1, pp. 109-121, 2010. DOI: .

Get article (PDF)View article entry (BibTeX)

©2015 e-Informatyka.pl, All rights reserved.

Built on WordPress Theme: Mediaphase Lite by ThemeFurnace.