Practical Automated Assistance for Program Restructuring

该项目的目标是改善现有的技术， 语义驱动的全球程序重组，使其 降低软件维护成本的实际应用。 软件维护成本过高， 已经观察到，结构退化是主要的，但 不可避免的原因 唯一的选择是重新实现 或者是重组。重组更有吸引力 方法，虽然这是一项复杂的活动，需要工具的帮助， 管理分散在程序文本中的更改的一致性。 最近，已经开发了用于工具辅助的方法。 重组，原型已经建立和测试。 然而，在这方面， 目前的方法不能完全解决模块级的重组问题 或交互式操作语义的性能问题 信息. 为了使这些技术切实可行，本项目努力 设计和实现全局模块转换，提高速度 通过增量更新数据结构， 重组工具与结构分析工具和一个可视化的前端- 以支持更高级别的转换。 结果预计 是对现有重组成果的自然适应， 编译器、编程环境和程序理解。 研究的结果将是一个工具，采用这些 技术，允许程序的高层次交互式重组， 并表明结构不是软件的固有障碍 上维护

The goal of this project is to improve the current technology of semantically-driven global program restructuring, enabling its practical application in reducing software maintenance costs. Software maintenance is disproportionately costly, and it has been observed that structural degradation is the primary, but unavoidable, cause. The only alternatives are to reimplement the system or to restructure it. Restructuring is the more attractive approach, though it is a complex activity requiring tool assistance to manage consistency of changes dispersed through the program text. Recently, methods have been developed for tool-assisted restructuring, and a prototype has been built and tested. However, the current methods do not fully address module-level restructuring or performance issues of interactively manipulating semantic information. To make these techniques practical, this project endeavors to design and implement global module transformations, increase the speed of the tool by incrementally updating data structures, and enhance the restructuring tool with structural analysis tools and a visual front- end to support higher-level transformations. The results are expected to be natural adaptations of existing results in restructuring, compilers, programming environments, and program understanding. The result of the research will be a tool employing these techniques, allowing high-level interactive restructuring of programs, and showing that structure is not an inherent barrier to software maintenance.