The reuse of tests for configured software products

New efficient software production techniques are important for improving the time-to-market of software products. One example of such advanced techniques is the so-called feature-based software production which employs high-level requirements or features in finding and selecting reusable software components for the development of new products. This kind of model-driven software development shortens the production time, but the validation of configured products still remains a bottleneck. An effort to apply regression testing techniques to configured software products shows that these techniques are not very well suited to meeting the new testing challenges. It is obvious that retesting an entire program, containing possibly only a few minor changes, is expensive. Therefore, an efficient testing approach is required for optimizing the size of the test suite. Other important issues concerning the testing approach are the design of reusable tests, the configuration and management of tests, and the automation of test execution. In the research, the testing efficiency problem is solved by using the idea of reusable software components from the feature-based production. In software testing the idea converts into a set of reusable test components designed for a product family. From a test material repository a suitable subset of tests is selected, modified, and configured to cover the characteristics of the product being tested. In addition, the repository may contain other relevant test data to be used for configuration, such as the test plans and the test environment configuration. The technique is called the feature-based testing approach. For identifying the relevant test data, the method proposes links to be created between the product features and the test material. The result of the test configuration depends on the automation degree, varying from a simple test

[1]  Etsi Tc-Mts Methods for Testing and Specification (MTS); Protocol and profile conformance testing specifications The Tree and Tabular Combined Notation (TTCN) style guide , 1994 .

[2]  U Pulkkinen,et al.  Usage models in reliability assessment of software based systems , 1997 .

[3]  R. Lewis,et al.  Assay - A Tool to Support Regression Testing , 1989, ESEC.

[4]  R. Atkins,et al.  TRSL standard supports current and future test processes , 1994, Proceedings of AUTOTESTCON '94.

[5]  Gregg Rothermel,et al.  Analyzing Regression Test Selection Techniques , 1996, IEEE Trans. Software Eng..

[6]  Hareton K. N. Leung,et al.  Insights into testing and regression testing global variables , 1990, J. Softw. Maintenance Res. Pract..

[7]  Rajiv Gupta,et al.  A methodology for controlling the size of a test suite , 1990, Proceedings. Conference on Software Maintenance 1990.

[8]  Paul Kogut,et al.  Component Provider's and Tool Developer's Handbook. Central Archive for Reusable Defense Software (CARDS) , 1994 .

[9]  David Wright,et al.  A Bayesian Model that Combines Disparate Evidence for the Quantitative Assessment of System Dependability , 1998, SAFECOMP.

[10]  Gregg Rothermel,et al.  A safe, efficient regression test selection technique , 1997, TSEM.

[11]  Anneliese Amschler Andrews,et al.  A knowledge-based approach to regression testing , 1994, Proceedings of 1st Asia-Pacific Software Engineering Conference.

[12]  Anneliese Amschler Andrews,et al.  Efficient testing of software modifications , 1993, Proceedings of IEEE International Test Conference - (ITC).

[13]  H. D. Desai Test Case Management System (TCMS) , 1994, 1994 IEEE GLOBECOM. Communications: The Global Bridge.

[14]  Robert W. Krut,et al.  Domain Analysis Workshop Report for the Automated Prompt and Response System Domain , 1996 .