Aspect-Oriented Approach for Testing Software Applications and Automatic Aspect Creation

The existing techniques for software testing can be used to perform only a particular type of testing, and moreover proficiency is required to write the automation test scripts using these techniques. This paper proposes a novel software testing approach using Aspect-Oriented Programming (AOP) that alone suffices for carrying out most of the types of software testing and thus obliterates the need of using distinctive tools for different types of testing. Nevertheless, AOP is a new programming paradigm and not all testers have the proficiency of working with it. Hence, a domain-specific language named Testing Aspect Generator Language (TAGL) was developed which has got a very low learning curve. Using TAGL, testers can write the testing code in the form of natural language-like statements. Further, the lexical analyzer and parser, written using lex and yacc, convert the TAGL statements into actual testing code in the form of AOP. The proposed approach was applied for the testing of widely used open source projects and remarkable bugs were detected into them. A detailed comparison as to how our approach is effective than the conventional testing techniques is provided.