Model Based Testing Automation Framework Of Data Pocessing System

With the rapid development of computer and network technology, data processing system is becoming more complex which increases the cost and difficulty for high quality software testing. Therefore, it’s very important to improve the automation of software testing process to ensure the test quality and reduce the test cost. Especially enhancing the automation of test case generation takes the key role. Based on this, this thesis used model based testing to design and implements an automation testing framework for data processing system function testing.At first, the status quo of the data processing automation, MBT (model based testing) and data mapping are analyzed, and then briefly introduces the related theory and technology of automated testing, MBT, MDA (Model Driven Architecture) and the key technologies used by the test framework. Secondly, through the analysis, the automation of test case generation, especially for expected result is not enough at present for data processing system. On the basis of the above analysis, the thesis uses the MBT thinking to get2abstract models to describe the data processing system. The first one is a data model which describes the metadata with XML Schema syntax and used to generate testing data. And the other one is a data mapping model which describe the data mapping with XPath (XML Path Language) grammar and PL/SQL(Procedural Language/SQL) functions and used to generate the expected result. XPath grammar is used to search and locate data and PL/SQL functions are used to operate the data. Thirdly, this thesis introduces design objective and function requirement. The architecture and transaction flows are designed accordingly. Finally, the automation testing framework is realized. The framework consists of test model creation, test case automatic generation and execution, and test result automatic comparison.This framework has already been deployed on data process system of a financial information company. With the use, it shows that the framework can improve the automation in the testing process which leads to the test efficiency improve and test cost reduces. Because of MBT thinking, it can continue to respond to frequent requirement changes quickly with easy model change instead of huge test case change. It’s a suitable testing framework for current popular agile pattern.