Viewing Software Artifacts for Different Software Maintenance Categories Using Graph Representations

Abstract

Information needed during an iterative process of a software maintenance process is much different from that of a software development process. Without up-dated documents, software maintainers need to gain information required to solve different maintenance categories through source codes hence consuming more time and effort. With the emergence of reverse engineering tools, the process of understanding source codes to solve maintenance tasks in different maintenance categories can be improved. Such tools employ diverse software visualisation methods that generate graph representations of parsed software artifacts. This paper discusses on how the graph representations provided by the proposed DocLike Modularised Graph (DMG) method employed in DocLike Viewer prototype tool can serve different levels of information needed by software maintainers in the case of corrective, adaptive and perfective maintenance category. It is observed that not only do software maintainers require diverse level of information; the necessity of the information is also not exactly of the same degree.