There is provided a computer-implemented method of testing an application. The method obtains first temporary test scripts for testing at least one test case of a first version of the application, and the first temporary test scripts are recorded with first mark data used for testing the first version of the application. The method obtains a first correspondence between the first mark data and test data. The method substitutes the test data for the first mark data in the first temporary test scripts based on the first correspondence to obtain first test scripts for testing the at least one test case of the first version of the application.