Understanding dynamic program behaviour

了解动态程序行为

• 批准号: 402446-2011
• 负责人: Holmes, Reid
• 金额: $ 2.11万
• 依托单位: University of British Columbia
• 依托单位国家: 加拿大
• 项目类别: Discovery Grants Program - Individual
• 财政年份: 2015
• 资助国家: 加拿大
• 起止时间: 2015-01-01 至 2016-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=573286
• 关键词: Understanding; dynamic; program; behaviour

Large software systems are notoriously difficult to create and evolve, and yet the pressure for change is unrelenting: new features must be added, existing defects must be repaired, and new platforms and environments must be supported, or systems risk obsolescence. Unfortunately, for those software engineers tasked with creating, evolving, and maintaining software systems, there is a conceptual gap between the modifications they make on a system and how the system's users perceive the success of those modifications. This gap arises because developers work with software's source code, which is a structured collection of textual artifacts. In contrast, users interact with the software by executing it and observing its behaviour. This gap makes it more difficult for developers to ensure that their systems exhibit the behavioural properties they desire without any undesirable defect-inducing behaviours from being introduced. The universal acceptance of software debuggers demonstrates that developers desire a link between the executable program and its source code; unfortunately, debuggers provide only a limited window into this linkage. The thesis of this research program is that by decreasing the conceptual gap between the static representation of a system and its corresponding dynamic behaviour, developers can more effectively evolve their systems. This program will help developers ensure that their source code changes induce the behavioural modifications they expect, while helping them avoid unexpected behavioural modifications. Ultimately, this research program will help developers more efficiently create and evolve their systems while reducing the likelihood of unintended defects being introduced.

众所周知，大型软件系统很难创建和发展，但变革的压力却是无情的：必须添加新功能，必须修复现有缺陷，必须支持新平台和环境，否则系统就有被淘汰的风险。不幸的是，对于那些负责创建、发展和维护软件系统的软件工程师来说，他们对系统所做的修改与系统用户如何看待这些修改的成功之间存在概念上的差距。 这种差距的出现是因为开发人员使用软件的源代码，而软件的源代码是文本工件的结构化集合。相比之下，用户通过执行软件并观察其行为来与软件进行交互。这种差距使得开发人员更难以确保他们的系统表现出他们想要的行为属性，而不会引入任何不良的缺陷诱发行为。软件调试器的普遍接受表明开发人员希望在可执行程序及其源代码之间建立链接；不幸的是，调试器只提供了一个有限的窗口来了解这种联系。 该研究计划的主题是，通过减少系统的静态表示与其相应的动态行为之间的概念差距，开发人员可以更有效地发展他们的系统。该计划将帮助开发人员确保他们的源代码更改引起他们期望的行为修改，同时帮助他们避免意外的行为修改。最终，该研究计划将帮助开发人员更有效地创建和发展他们的系统，同时减少引入意外缺陷的可能性。