DEVELOPMENT OF ABSTRACT AUTOGRADER INTEGRATION SYSTEM WITH MOODLE AND GITLAB

The use of LMS (Learning Management System) for learning activities still requires students to upload program files, including in ITB itself, be it program files directly or in zip form. While programming in general uses VCS (Version Control System) to record changes to the developed program, it...

Full description

Saved in:
Bibliographic Details
Main Author: Wahyu Langkawi, Dimas
Format: Final Project
Language:Indonesia
Online Access:https://digilib.itb.ac.id/gdl/view/67772
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Institut Teknologi Bandung
Language: Indonesia
Description
Summary:The use of LMS (Learning Management System) for learning activities still requires students to upload program files, including in ITB itself, be it program files directly or in zip form. While programming in general uses VCS (Version Control System) to record changes to the developed program, it would be nice if students could learn and adapt early on to this kind of environment. Therefore, a system that integrates learning media or LMS with VCS is needed. There has been research on system integration by Panjaitan et al. (2021), but from the results of this research there are still several things that can be improved. In this study, the system is limited to using black box autograders and does not support other autograders to be integrated in the system. This can be improved by generalizing communication with the autograder so that the system can be integrated with abstract autograders, be it the white box autograder studied in this capstone final project or autograders that will be developed in the future. Since the number, location and availability of abstract autograders will inevitably change dynamically during the running of the system, a way to discover, locate and manage autograders is required. To solve this problem, a service discovery pattern is used to enable abstract autograders to be integrated with the system automatically while the system is running. This pattern was successfully implemented within the constraints of this paper and successfully detected and managed the autograders registered with the system. The system has also generalized the communication with the autograder so that the system is able to integrate with abstract autograder.