Infeasible Paths in the Context of Data Flow Based Testing Criteria: Identification, Classification and Prediction

Silvia R. VergilioJosé Carlos MaldonadoMário Jino

Infeasible paths constitute a bottleneck for the completeautomation of software testing, one of the mostexpensive activities of software quality assurance. Researchefforts have been spent on infeasible paths, basicallyon three main approaches: prediction, classificationand identification of infeasibility. This workreports the results of experiments on data flow basedcriteria and of studies aimed at the three approachesabove. Identification, classification, and prediction ofinfeasible paths are revisited in the context of dataflow based criteria (Potential Uses Criteria-PU). Additionally,these aspects are also addressed in thescope of integration and object-oriented testing. Implementationaspects of mechanisms and facilities todeal with infeasibility are presented taking into considerationPoketool - a tool that supports the applicationof the Potential Uses Criteria Family. Theresults and ideas presented contribute to reduce theefforts spent during the testing activity concerning infeasiblepaths.

Caso o link acima esteja inválido, faça uma busca pelo texto completo na Web: Buscar na Web

Biblioteca Digital Brasileira de Computação - Contato:
     Mantida por: