Fault-based refinement-testing for CSP
Fault-based refinement-testing for CSP
复制标题
CSP 基于故障的细化测试
DOI：
10.1007/s11219-018-9431-9
复制
发表时间：
2019
影响因子：
1.9
通讯作者：
Cavalcanti A
中科院分区：
文献类型：
--
作者：
Cavalcanti A
进程代数CSP已被作为基于模型的测试的符号进行了研究。理论和实践工作都是利用它的踪迹和故障语义，以及它们作为一致性关系的精化概念来展开的。已经定义并证明了两组测试是详尽的，在某种意义上，它们可以识别与相关细化关系不一致的任何SUT。然而，这些集合通常是无限的，在这种情况下，显然不可能应用它们来验证SUT的符合性。研究了一些经典的基于模型的选择准则。在本文中，我们提出了一种在线测试生成过程，用于从CSP模型中选择用于轨迹求精的有限测试集。它基于故障域的概念，专注于感兴趣的一组错误实现。我们调查在有限数量的测试执行后可以得出测试活动的裁决的情况。我们通过一些案例研究来说明该方法的使用。
The process algebra CSP has been studied as a notation for model-based testing. Theoretical and practical work has been developed using its trace and failure semantics, and their refinement notions as conformance relations. Two sets of tests have been defined and proved to be exhaustive, in the sense that they can identify any SUT that is non-conforming with respect to the relevant refinement relation. However, these sets are usually infinite, and in this case, it is obviously not possible to apply them to verify the conformity of an SUT. Some classical selection criteria based on models have been studied. In this paper, we propose a procedure for online test generation for selection of finite test sets for traces refinement from CSP models. It is based on the notion of fault domains, focusing on the set of faulty implementations of interest. We investigate scenarios where the verdict of a test campaign can be reached after a finite number of test executions. We illustrate the usage of the procedure with some case studies.