4th Increment – MIDAS version 0.4
On September 30th the MIDAS 0.4, according to the 4th increment, was released.
The objectives of the 4th increment were the following:
- Complete prototype with all testing methods developed as part of MIDAS.
- Full support for all scenarios defined for the three major MIDAS use cases.
- Possibility to dynamically orchestrate test scenarios through the definition of a MIDAS test method.
- Usage of first version of the pilots for verification.
- Full adoption of the MIDAS DSL.
- Integration steps 4, 5, and 6 are completed.
The developed prototype did met most of the objectives. Dynamic orchestrations were already available in the 3rd increment. Furthermore, the end user services for file management, the identity and authentication, and test status handling were extended to trace user activities inside a tenancy. This enabled the implementation of the new accounting and billing service, which was added in this increment. The TTCN-3 generator was validated with a first and simplified version of the health care pilot. However, the compilation of TTCN-3 as well as the generation of test schedulings in TTCN-3 from MIDAS DSL compliant UML activities was postponed to the third year of the project.
The usage-based testing implements all mandatory as well as some optional requirements for both test generation and test scheduling. All generated test cases and test schedules are fully MIDAS DSL compliant and can be used together with the TTCN-3 generator. This means that all defined scenarios for usage-based testing are fully supported. The components for usage-based testing were verified using a first and simplified version of the health care pilot. Hence, the usage-based testing components fully meet all objectives of the increment. The implementation of optional requirements on the usage-based testing will be considered in the third year of the project.
To support the security testing, the TTCN-3 generator was extended with functionality to support data fuzzing, which enabled the implementation of a TTCN-3 adaption of FF’s data fuzzing library. Subsequently, the data fuzzing and the behavioral fuzzing test generation were implemented, including the selection of fuzzing heuristics based on data type constraints, generational and mutational data fuzzing. Additionally, the fuzzing generators support repeatability of the randomized generation and the specification of test strategies for data fuzzing. The security testing is fully compliant to the MIDAS DSL and can be used together with the TTCN-3 generator. The combination of the data and behavioral fuzzing, as well as security-oriented scheduling of tests were postponed to the third year.