Scholarlink

This paper is an extended revision of a previous work that examined the task of recovering design models in UML from existing code: the code basis needs to be imported into the modeling language. However, there is an abstraction gap between the programming language (C++) and the modeling language, implying that the model obtained via reverse engineering directly mirrors the object-oriented implementation structures. It does not use higher-level modeling mechanisms such as component-based concepts or state machines. In the previous work, we described problems related to specific language concepts, such as advanced templates, and gave the perspective that AI can help to recover the high-level model. In this paper, we show the first results of AI-assisted reverse engineering for small examples and plan to add these to improve the reverse engineering capabilities of an existing tool.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

AI-Based Recovery of High-Level Models in the Context of Reverse Engineering of C++ to UML

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Ansgar Radermacher,
  • Marcos Didonet Del Fabro,
  • Shebli Anvar,
  • Frédéric Chateau,
  • Yassine Abdeljalil

摘要

This paper is an extended revision of a previous work that examined the task of recovering design models in UML from existing code: the code basis needs to be imported into the modeling language. However, there is an abstraction gap between the programming language (C++) and the modeling language, implying that the model obtained via reverse engineering directly mirrors the object-oriented implementation structures. It does not use higher-level modeling mechanisms such as component-based concepts or state machines. In the previous work, we described problems related to specific language concepts, such as advanced templates, and gave the perspective that AI can help to recover the high-level model. In this paper, we show the first results of AI-assisted reverse engineering for small examples and plan to add these to improve the reverse engineering capabilities of an existing tool.