Program testing requires a series of tasks such as preparing drivers and stubs, creating test cases, and executing unit tests. To reduce manual effort of performing such tasks for testing Java programs, we developed a tool that fully integrates and automates all of these processes, by using JPF with extensions as a symbolic execution engine for automatically generating unit test cases. In this paper, we present this tool and its application to real projects to evaluate its efficacy. The evaluation results demonstrate that the tool performs well in terms of the test time reduction compared with manual test as it eliminates the total amount of manual effort, while largely preserving a high coverage of greater than 90 % as our expected borderline.
In the social reality, objects communicate with each other by means of assuming roles to establish collaboration, and then can adaptively change their roles to obtain other interaction possibilities. To achieve the goal of supporting and realizing such object collaboration and adaptation in the object-oriented technology, especially in Java, a new adaptive role-based model Epsilon and a corresponding language EpsilonJ have been proposed. In this paper, we present the background of adaptive role-based models, and then focus on the design of this Epsilon model and its language. A program written in EpsilonJ must be translated into executable code to execute. We propose a translation scheme of mapping EpsilonJ syntax to the standard Java. With this translation scheme, we implemented a practical syntax translator as a preprocessor of EpsilonJ program, through lexical analysis and parsing. Evaluation shows that our translator can effectively perform transformation in high accuracy, and translated programs can be executed more efficiently than the existing implementation of EpsilonJ.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.