Abstract: The computer-implemented method for producing a test code for automated testing a software comprises providing a high-level test model for modelling a set of requirements for the software to be tested; generating a plurality of high-level test cases on the basis of the high-level test model; providing a human tester with the software to be tested; executing the software; prompting the human tester to execute the high-level test cases by the software under test with using input values arbitrarily selected from an available set of input values of the software in accordance with the high-level test model; producing a plurality of low-level test models for the high-level test cases by the human tester; and generating a computer-executable test program code for the low-level test models.

Abstract: A system and method for determining execution trace differences in a computer-implemented software application is provided herein. A software application under analysis is executed at least twice, thereby generating first and second execution trace and associated first and second sets of execution data describing all the necessary data gained by program instrumentation at each program statement of source or bytecode level. These data are stored for at least two executions of the software application. This data is then compared to determine a set of differences between the first and second executions of the program. The set of differences may contain statement coverage data, execution trace data, variable values or influences among program instructions. The differences can be arranged according into historical order. The differences then can be analyzed to identify the location of the fault in the program or to map the related code to a feature in unknown code.