Architecture for automatic source code comment generation

Loading...
Thumbnail Image

Date

2022

Journal Title

Journal ISSN

Volume Title

Publisher

Abstract

This research tries to give a highly available software architecture to a source code commenting tool. Code commenting is helpful for many phases like code updates, bug fixes, etc. Developers need to know which task each code is performing, and also, adding comments for each code is essential. But most developers take more time to add comments which has created the problem of wasting time. Using an automated comment generator tool for source code, we can avoid that problem and save time for another task. Currently, there is not a properly defined architecture for an automated comment generator tool. Therefore, we are trying to give an evaluated architecture for that tool that supports overcoming the problem. In this architecture, we plan to follow the microservice architectural concepts. Then modularization of each component based on the service it is going to provide. Also, plan to use ActiveMQ for the queue technique, Scheduling techniques to reschedule things, No SQL databases for data saving, Caching techniques to store temporary data, etc. This procedure ensures that it will provide a highly available architecture. Also, this architecture will have the following quality attributes: Maintainability, Performance, Interoperability, Usability, Availability, Reliability, Testability, Modifiability, Scalability, and Reusability. Finally, we use presentations, proof of concept, and ATAM & CBAM methods to validate whether the architecture is commercially viable or not.

Description

Keywords

DEVELOPMENT, ZOOKEEPER, ACTIVEMQ, REST, DATABASE, QUEUE, ASYNCHRONOUS, CACHE, COMPUTER SCIENCE & ENGINEERING - Dissertation, COMPUTER SCIENCE- Dissertation

Citation

Samarasinghe, R.M.N.S. (2022). Architecture for automatic source code comment generation [Master's theses, University of Moratuwa]. Institutional Repository University of Moratuwa. http://dl.lib.uom.lk/handle/123/22367

DOI