The Research On Test Case Sequence Optimization Of Database Application System

Software testing plays an important role in the software life cycle,which is an important technical measures to ensure the software quality,plays an irreplaceable role in finding software defects and improving the efficiency of the test.According to incomplete statistics,in the software application,about 80% is the application of database system.A typical feature of the database application system test is the status of the database,which may change with the implementation of test cases.Most of test cases of database application system will interact with the data in the database and change the state of database.And the change of the database's state likely have an impact on the next test case,leading to test failure.Therefore,there is a dependency relationship between test cases.When the change of the database' state lead to the failure in continuing execution of the next test cases,we need to revert to the previous state of the database by resetting operation.This operatio n,however,is very time consuming,and reduces the test efficiency.So,how to avoid resetting operation because of the dependency relationship between test cases is a problem faced by the database application system test.To solve problems above,this paper presents a test case sequence optimization algorithm based on database application system,details are as follows:(1)it will analyze and summarize the reason of dependencies between database operations generated,with the operation of relational database tables as a unit,and give a dependent recognition algorithm to extract the dependencies between database operations,and then map the test cases to specific operations of specific tables,extract dependencies between test cases indirectly based on the front dependencies between database operations.(2)according to the dependencies between test cases mentioned previously,this paper presents a test case sequence optimization algorithm,which is based on the idea of graph theory,transform the test set and its dependencies into a directed graph,break the loops that may exist in the directed graph by dividing the test suite into several sections of test cases.By this method,dependencies between test cases become dependencies between slices of test cases,which will reduce conflicts with implementation of test cases occurred,thereby reduce the number of times to reset the database.Finally,on the basis of the above,this paper presents a test case sequence optimization framework based on a database application system.The specific project example shows that the framework can optimize the sort of test case sequence,avoid dependencies between test cases effectively,reduce the number of database resets and improve the test efficiency.