Debugging is the most important task related to the testing activity. It has the goal of locating and removing a fault after a failure occurred during test. However, it is not a trivial task and generally consumes effort and time. Debugging techniques generally use testing information but usually they are very specific for certain domains, languages and development paradigms. Because of this, a Neural Network (NN) approach has been investigated with this goal. It is independent of the context and presented promising results for procedural code. However it was not validated in the context of Object-Oriented (OO) applications. In addition to this, the use of other Machine Learning techniques is also interesting, because they can be more efficient. With this in mind, the present work adapts the NN approach to the OO context and also explores the use of Support Vector Machines (SVMs). Results from the use of both techniques are presented and analysed. They show that their use contributes for easing the fault localization task.

We present in this paper a systematic review, using approach, of methods, techniques and tools regarding to concolic testing with application of test criteria. The test activity is the process of running a program with the intent of discovering defects. The search for test cases to increase the coverage of structural tests is being addressed by approaches that generate test cases using symbolic and concolic execution. Concolic testing is an effective technique for automated software testing, that aims to generate test inputs to locate failures of implementation in a program. Application of a test criterion is very important to ensure the quality of the test cases used. The number of elements exercised provides a measure of coverage that can be used to evaluate the test data set and consider the test activity to be closed.

No abstract

O teste evolutivo de software orientado a objeto é uma área de pesquisa emergente. Algumas abordagens promissoras sobre o assunto são encontradas na literatura, entretanto, estas não consideram critérios propostos recentemente que utilizam o código objeto Java para obter os requisitos de teste. Além disso, os trabalhos geralmente não estão integrados a uma ferramenta de teste. Neste artigo, um framework, chamado TDSGen/OO para geração de dados de teste é descrito. TDSGen/OO utiliza Algoritmos Genéticos e trabalha de maneira integrada com a ferramenta JaBUTi, que implementa diferentes critérios de teste baseados no bytecode e em mecanismos de tratamento de exceções, permitindo o teste de componentes mesmo que o código fonte não esteja disponível. Alguns resultados preliminares são também apresentados que mostram benefícios no uso do framework.