The Research On Test Database Automatic Generation Base On RQP

With the popular database application system,software testing play an important role in the quality and reliability of database application system.It is necessary to generate some significant data in database application system before testing it, which also known as the test database generation. The quality of the test data generation will affect the efficiency and completeness of the testing.There are a number of commercial tools to automatically generate test databases. These tools take a database schema and table sizes as input in order to generate new tuples. However, these tools often generate test databases which do not reflect the semantics of the application logic that should be tested, not adequate for testing a database application.In order to solve the problem that the generation process of Multi-RQP testing database the integration of data may obey the referential integrity constraints, a TDAG-RQP test database generation algorithm based on RQP (Reverse Query Processing) has been proposed in this thesis. Using database integrity constraints, the proposal first obtains data dependence of test data generation between separate tables in database, and then applies RQP to a top-down test database generation according to tester specified queries. The referential values in top tables are involved in test data generation of down tables when deriving tester specified queries. Algorithm analysis and case studies show that the test data generated from the new algorithm satisfies the database integrity constraints. The new algorithm improves the efficiency of the test database generation significantly, as well as follows the requirements of tester specified queries.The domain knowledge heuristics algorithm is proposed in order to solve the problem that the value generated by the randomly method can't meet the semantic requirements of keys or contrary to attribute domain constraints and the primary key uniqueness constraints. In the algorithm, domain words database for each attribute is established,and each value must be obtain the domain words to satisfy semantics and meet the domain constraint. The algorithm can guarantee the primary key is unique in database. The algorithm generate the tuples for each query firstly,then insert into database if the tuples meet the unique constrain,otherwise ignore them and then regenerate. Cases analyses show that the algorithm can guarantee primary key is unique effectively,it satisfies domain constraints and the data generated meet the semantic requirements.