Towards a Framework for Automated Random Testing of Aspect-Oriented Programs.
Reza M. Parizi, Abdul Ghani, Rusli Abdullah, Rodziah binti Atan · Software Engineering and Data Engineering · 2009
Abstract The recent past has seen the emergence of several techniques for testing aspect-oriented programs; well-known among them is systematic testing such as structural code-based testing using data flow or control flow analysis and specification-based testing by means of requirements or models. It is intuitively conjectured that applying recent techniques on testing and verification to AO-based systems is crucially important for making AOSD (Aspect-Oriented Software Development) success. As in the sequel, there is much advancement on testing and verification, like automated test generation using random testing, coverage analysis, mutation analysis and so forth. Many of those techniques are not yet applied to AO-based programs. Some would be trivial, but some would need some efforts to make them possible and lift to testing of AOP. Random testing is an active research topic in software testing, which has also a niche in practical settings due to the merits it offers, e.g. fault-detection capacities at low cost, ease of implementation, reliability estimation, facility for automation and so forth. However, the idea behind random testing can intuitively be worthwhile and attractive for testing aspect-oriented programs since current research on testing of AOP, especially automated has not been adequately performed and is still in infancy. So far, there is no testing approach, in the context of AOSD, taking the random testing techniques into account for aspect-oriented programs which can make this study as first and unique in this sense. In this position paper, we propose a framework to random testing of aspect-oriented programs. The major objective of this work is to combine the idea of random testing with the AOP testing in which randomness is brought into the AOP testing to facilitate test case generation and execution and finally contribute to the testing AOP success.