Model Generation

Motivation

There are many model-oriented operations (MOs), such as model transformation, model querying, and model comparison. The qualities (especially the performance and the correctness) of an MO become an important factor when the operation is to be put into practice. Hence, a systematic testing should be carried out before adopting those MOs.

However, realistic models that are created by human are usually not available for the purpose of testing. Synthesized models are needed as the test input.

In this project, we developed two general purpose model generators to meet various needs for model generation.

Tools in This Project

• Random model generator is able to generate a random model according to a metamodel. The generator also supports a set of predefined model constraints. The generator is suitable for generating large-size models for performance testing and black-box testing.
• Template-based model generator is able to generate a model according to a set of user-defined templates. It provides a template language to specify templates. The generator turns the templates into a wrapper metamodel and invokes the random model generator to generate a wrapper model. Finally, the wrapper model is unfolded into the desired model.
• Test case management framework is able to manage test cases of MOs. It can invoke the random model generator to generate test inputs. It can also execute the test cases.

Application Scenario：

• Performance testing of MOs
• Benchmark generation
• Black-box testing of MOs, especially, as a baseline approach

Publications

1. Xiao He, Xing Chen, Sibo Cai, Ying Zhang, Gang Huang. Testing Bidirectional Model Transformation Using Metamorphic Testing. Information and Software Technology. (Accepted for publication). CCF B (More information)
2. He Xiao, Zhang Tian, Pan Minxue, Ma Zhiyi, Hu Changjun. Template-based Model Generation. Software and Systems Modeling. (Accept for publication). DOI:10.1007/s10270-017-0634-5
3. He Xiao, Zhang Tian, Hu Changjun, Ma Zhiyi, Weizhong Shao. An MDE Performance Testing Framework Based on Random Model Generation. Journal of Systems and Software, 2016, 121: 247-264. DOI: 10.1016/j.jss.2016.04.044
4. He Xiao, Li Wenfeng, Zhang Tian, Ma Zhiyi, Shao Weizhong, Hu Changjun. Randomized Approach to Software Model Generation. Journal of Software (RuanJianXueBao), 2017, 28(4):907-924 (in Chinese with English abstract). DOI: http://www.jos.org.cn/1000-9825/5055.htm
5. He Xiao, Li Wenfeng, Zhang Tian, Liu Yi. Towards Parallel Model Generation for Random Performance Testing of Model-Oriented Operations. The 10th International Symposium on Theoretical Aspects of Software Engineering, 2016, pp. 57-64.
6. He Xiao, Zhang Tian, Ma Zhiyi, Shao Weizhong. Randomized model generation for performance testing of model transformations. The 38th Annual International Computers, Software & Applications Conference (COMPSAC’14), 11-20. DOI:10.1109/COMPSAC.2014.103