Wiki Support for Automated Definition of Software Test Cases

Abstract

The design of tests is a very important step in the software development process since it allows us to match the users’ expectations with the finished product. Considered as a cumbersome activity, efforts have been made to automatize and alleviate the burden of test generation, but it is still a largely neglected step. We propose taking advantage of existing requirement artifacts, like Scenarios that describe the dynamic of the domain in a very early stage of software development, to obtain tests from them. In particular, the approach proposed complement the Scenarios that are textually described with a glossary, the Language Extended Lexicon. Thus, a set of rules to derive tests from Scenarios is also proposed. The tests are then described using the Task/Method model. The main findings of this work consist of an extension of a previously presented set of rules. And a tool based on a media wiki platform that makes possible to record Scenarios and the Language Extended Lexicon and implement the rules to obtain the tests. The main originality of this work is the glossary which complements Scenarios, the semantic support to obtain tests and the tool to automatize the approach.