| Formal specification uses mathematical notation to describe in a precise way what an information system should do. In the process of developing a computer system, developers will interpret the specification and turn it into a software product. In order to ensure the product complies with the specification, Unit Testing needs to be carried out against the each component of the product. The purpose of this research is to examine the feasibility of automatically producing self-checking tests from a specification. The specification may be only partial and involve descriptive text, in which case, the programmer will have to provide additional input at test generation time. This thesis implemented a framework for Black-Box Unit Testing, which ensures a program meets its specification from a functional or behavioral perspective. The framework includes a parser to interpret the specification, and will allow the tester to specify input values for a test and will use the specification to determine the correct result. This result is then compared to the actual result obtained by running the test. The system will output testing result in a report format for each software component it has tested. The immediate application of this work is in improving the existing Unit Testing tool set to fill in the gap between functional specification and quality assurance. |
