Special section on Mutation testing (Mutation 2010).

L Du Bousquet, JS Bradbury, G Fraser - Sci. Comput. Program., 2013 - academia.edu
Sci. Comput. Program., 2013academia.edu
How good are my test cases? Mutation testing has been used for over 30 years to answer
this question and to provide confidence in the testing process. In mutation testing, many
versions of a program-under-test are created, each with a syntactic variation. Mutation
testing measures how sensitive the test cases are to these syntactic changes by determining
how many test cases can detect the difference between the original program and the
''mutant''program versions. Within the Software Engineering research community, mutation …
How good are my test cases? Mutation testing has been used for over 30 years to answer this question and to provide confidence in the testing process. In mutation testing, many versions of a program-under-test are created, each with a syntactic variation. Mutation testing measures how sensitive the test cases are to these syntactic changes by determining how many test cases can detect the difference between the original program and the ‘‘mutant’’program versions. Within the Software Engineering research community, mutation testing continues to see an increase in popularity with contributions and applications of mutation analysis appearing in many major conference proceedings and journals. The premiere venue dedicated to mutation testing research is the International Mutation Analysis Workshop Series, an annual event with 7 workshops since 2000. The Mutation workshops examine all aspects of mutation testing and analysis, providing a successful venue for discussing the state-of-the-art and future directions of this growing research area. The 5th International Workshop on Mutation Analysis (Mutation 2010) was held in Paris, France and received 12 submissions. After a rigorous reviewing process in which each paper was subjected to at least three independent reviews followed by program committee discussion, 9 papers were accepted for presentation and publication. The two papers selected by the guest editors for publication in this special section are significantly extended versions of those papers that received the strongest support from the workshop audience and the journal referees. Each of the extended papers was reviewed by at least three expert reviewers and has undergone additional revisions. Although both papers differ in their subject matter, they share a common underlining theme—the evolution of the traditional definition of mutation testing in new and novel ways.
Mutation testing traditionally involves the generation of syntactic mutant variations of a program. The first paper by John Clark, Haitao Dan, and Rob Hierons presents the new idea of Semantic Mutation Testing—rather than introducing syntactic variations, semantic mutation testing varies the semantics of the programming language. Such changes represent possible misunderstandings of the language semantics, such as floating point comparisons. The paper discusses semantic mutation testing for Statecharts models and C programs, and presents the tool SMT-C that performs semantic mutation testing for C programs. The empirical results presented in the paper are promising and indicate that semantic mutation operators represent different types of faults than syntactic mutation operators. The implication of this result is that both semantic and syntactic mutants should be used in the future.
academia.edu
Showing the best result for this search. See all results