数据驱动的测试用例排序实用理论研究

Test case prioritization, aims to schedule the execution order of test cases so as to improve the efficiency of software testing. Although test case prioritization has been studied for long, its techniques have rarely been used in practice due to the gap between research and practice. In this project, we will present an application methodology of test case prioritization driven by history data, which targets at bridging the gap between the research on test case prioritization and the practice of software testing. In particular, we will first conduct an empirical study to investigate the factors that influence the application of test case prioritization. Then we classify these factors from three aspects. The first one is technique choice, which refers to the selection of prioritization techniques. The second one refers to the testing scenario like the modification on the source code and test code. The third one refers to the hardware that software testing is performed on, e.g., the processors. Regarding to each factor influencing the practical usage of test case prioritization, we present a data-driven technique accordingly. Finally, we will evaluate the proposed application methodology of test case prioritization and its techniques through open-source projects and commercial projects.

测试用例排序，旨在通过优化测试用例的执行顺序，来提高软件测试的效率。测试用例排序研究历史悠久，形成了大量的研究成果。另一方面，软件产业的实际测试过程中虽然存在着明显的测试效率问题，但却鲜少使用已有的排序研究成果。换而言之，测试用例排序的学术研究和实际应用之间存在着明显间隙。本项目拟针对这个间隙，提出一套测试用例排序的实用理论，以开源闭源项目数据为基础，解决测试用例排序实用化中的关键问题。具体地，本项目拟首先通过实证研究明确测试用例排序技术在实际应用中的关键问题，然后从排序技术选择、软件环境和硬件环境三方面入手，提出数据驱动的测试用例排序选择技术、考虑软件环境（如测试用例和程序代码变化）的测试用例排序技术，以及考虑硬件环境（如多核处理器）的测试用例排序技术。最后，本项目拟在开源项目和闭源项目上验证上述技术的有效性。

测试用例排序，旨在通过优化测试用例的执行顺序，来提高软件测试的效率。测试用例排序研究历史悠久，形成了大量的研究成果，但这些研究鲜少被业界使用。本项目拟从测试用例排序的研究现状和实际测试过程的显著差异为入手点，以开源闭源项目数据为基础，提出一种数据驱动的测试用例排序实用理论，解决测试用例排序实用化中的关键问题。具体地，本项目的代表性研究成果包括（但不限于）：并行测试用例排序、持续集成环境下的测试用例排序和测试用例排序的加速算法。研究成果共发表学术论文16篇，其中国际期刊4篇，国际会议9篇，国内期刊3篇：其中，在国际软件工程顶级权威期刊《IEEE Transactions on Software Engineering》和《ACM Transactions on Software Methodology and Engineering》（CCF A类期刊）发表论文3篇，计算机领域重要综述期刊《ACM Computing Survey》发表论文1篇，国际（国内）期刊《Science China: Information Sciences》发表论文1篇；在国际软件工程领域的顶级会议（CCF A类会议）International Conference on Software Engineering、ACM SIGSOFT International Symposium on the Foundations of Software Engineering、International Conference on Automated Software Engineering和International Symposium of Software Testing and Analysis上发表学术论文6篇、新想法类短论文1篇、产业应用实践论文1篇。申请国际专利一项，获得国内发明专利一项。本项目的课题负责人郝丹于2021年获聘教育部长江学者“特聘教授”，参与人员博士生陈俊洁获得2019年度中国计算机优秀博士论文。