Conference article

Unit testing of software requires the construction of large amounts of supporting software. Test drivers and stubs are needed to isolate the unit under test from other units; and test cases specifications need to be translated into compilable code. Constructing test software is as time-consuming and as fault-prone as any other software; and sometimes more so. To support testers there are tools that can automatically generate test code from test case specifications that are similar to the test case specification. These tools include a test case definition language; in which the test cases are stated; and a test script generator; that generates compilable code from the test case definitions. Even so; testers often have to resort to writing source code for test cases requiring features unavailable in the test case definition language. To learn more about how a better test case definition language could be constructed; we have designed and implemented a test script generator and a test case definition language. This paper gives an overview of some of our experiences and some of our ideas for future work.

No keywords available

[1] Marc J. Balcer, William M. Hasling, and Thomas J. Ostrand. Automatic generation of test scripts from formal test specfications. In Proceedings of the ACM SIGSOFT’89 Symposium on Software Testing Analysis and Verification (TAV3), number 5 Key West, Florida, December 1989. ACM SIG-SOFT.

[2] IPL Information Processing Ltd., Bath. Cantata Test Case Definition Language Reference Manual. See also http://www.iplath.co.uk/

[3] Thomas J. Ostrand and Marc J. Balcer. The category-partition method for specifying and generating functional tests. Communications of the ACM, 31(6):676-686, June 1988

[4] Sun Microsystems, Inc. ADL Translator User’s Guide. Sun Microsystems, Inc., 1996. See also http://ADL.opengroup.org/.