A biting‐down approach to hierarchical decomposition of object‐oriented systems based on structure analysis

Abstract

System decomposition has been widely viewed as an effective means to facilitate the comprehension of complex software systems and/or capture potentially reusable components in them. In fact, various approaches to system decomposition have been intensively documented in the literature. However, during the process of system decomposition, only a few of them can also capture the target system's hierarchical organization structure, which is essential when the target system is very complex. In this paper, we present a biting‐down approach to hierarchical decomposition of object‐oriented systems. Compared with the previous hierarchical approaches, the distinct features of this approach are as follows. First, our approach does not rely on agglomeration, and thus can avoid some unnecessary calculations. Second, our approach does not require merging nodes when performing high‐level decomposition, and thus can avoid imprecision induced by the merging. To evaluate our approach, we conducted a case study and an experimental study on our approach. The results of these studies can confirm its effectiveness and its superiority over our previous approach. Copyright © 2009 John Wiley & Sons, Ltd.

Using reverse engineering techniques to infer a system use case model

“…Zhang et al,() Si et al, and Li et al employ similar approaches to obtain use case models. Zhang et al and Si et al build a special kind of graph that considers branches in source code ( branch graph ), and then, using static code analysis techniques, they associate source code entities to use cases.…”

“…The resulting weight vector for Cyclomatic Complexity, Number of LOCs, Visibility, and Type is [50,20,15,15]. By adopting this technique to obtain a weighting scheme for each operator, we could check the effective consistency among the criteria compared.…”

“…When relevant patterns are detected, the software engineer must enrich them with semantic information. These patterns are taken as a base to obtain system use cases.Zhang et al, 14,15 Si et al, 16 and Li et al 17 employ similar approaches to obtain use case models. Zhang et al and Si et al build a special kind of graph that considers branches in source code (branch graph), and then, using static code analysis techniques, they associate source code entities to use cases.…”

Abc: 一种全生命周期软件体系结构建模方法

“…完成从代码到设计的逆向追踪的首要目标是逆向恢复遗产软件系统的体系结构. 针对这一问题, 需先识别代码中的构件, 并在此基础上恢复软件体系结构 [34,35] 为了帮助遗产软件系统的维护与演化, 需要挖掘构件的使用规约 [36,37] , 以便在维护和演化过程中 避免开发人员对构件的误用. 由于构件使用规约蕴含在构件的使用代码中, 如果一个遗产软件系统中 的构件使用代码不够充足, 就需要联合多个遗产软件系统进行挖掘.…”

Recuperación de Arquitecturas de Software: Un Mapeo Sistemático de la Literatura

“…Además, hace más eficiente el trabajo de investigación en este campo, porque brinda la oportunidad de integrar las propuestas existentes; al mismo tiempo que facilita la definición de lineamientos y criterios que hagan posible la unificación de los mecanismos de representación de los elementos recuperados por cada técnica (Favre, 2004). Esto también contribuye a la definición de técnicas de análisis arquitectónico sobre los resultados obtenidos en los procesos de recuperación de arquitecturas (Zhang et al, 2009;Peake et al, 2013).…”