Multi-Language Program Understanding Tool

Abstract

Open-source programs have gained popularity due to their decentralized, quick development cycles and accessibility to everyone. Program understanding is vital for open-source software developers to modify or improve the code. However, one problem open-source developers face is the difficulty in understanding the programs as the program grows large and becomes complex. The current program understanding tool is inefficient because it only supports one programming language, while open-source programs are written in various languages. This paper discusses a new program understanding technique that facilitates multi-language program understanding. The proposed technique helps developers to understand open-source programs by supporting two unique features: multimedia and additional comments. We carried out this study in four stages. First, we examined available tools and techniques in software understanding to identify their strengths and weaknesses. Second, we proposed a new technique. Third, we designed a new tool to implement the new technique. Lastly, we evaluated the tool using a survey. We invited twenty users, including students and programmers, to use the system and ask for their feedback. The evaluation of the proposed techniques shows that the respondents have a positive perception as they agree that the technique helped them better understand the program. The multimedia support and an additional comment provided by the tool significantly improve user understanding of the program. For future work, we would like to explore the possibility of utilizing some machine-learning techniques to enhance the process of program understanding.