RANCANG BANGUN EKSTENSI VISUAL STUDIO CODE (VSCODE) TERINTEGRASI AI UNTUK GENERASI UNIT TEST

Fansuri Fadel Fitrah Prakon, NIM.: 21106050091 (2025) RANCANG BANGUN EKSTENSI VISUAL STUDIO CODE (VSCODE) TERINTEGRASI AI UNTUK GENERASI UNIT TEST. Skripsi thesis, UIN SUNAN KALIJAGA YOGYAKARTA.

Preview	Text (RANCANG BANGUN EKSTENSI VISUAL STUDIO CODE (VSCODE) TERINTEGRASI AI UNTUK GENERASI UNIT TEST) 21106050091_BAB-I_IV-atau-V_DAFTAR-PUSTAKA.pdf - Published Version Download (4MB) | Preview
	Text (RANCANG BANGUN EKSTENSI VISUAL STUDIO CODE (VSCODE) TERINTEGRASI AI UNTUK GENERASI UNIT TEST) 21106050091_BAB-II_sampai_SEBELUM-BAB-TERAKHIR.pdf - Published Version Restricted to Registered users only Download (13MB) | Request a copy

Abstract

Software testing is a critical phase in the software development lifecycle to ensure system quality and reliability. However, manual unit test writing remains time-consuming and effort-intensive, often becoming a burden for developers. This study aims to design and develop a Visual Studio Code extension named Unittest Generator, integrated with artificial intelligence based on Large Language Models (LLMs) through Groq API and OpenRouter API, to automate unit test generation for multiple programming languages, namely Python, JavaScript, and Java. The development process applies the Extreme Programming (XP) methodology combined with a Research and Development (R&D) approach across two development iterations. The testing results in the second iteration indicate that all functional requirements of the system have been successfully fulfilled. The quality evaluation of the generated unit tests shows line coverage ranging from 85–92% and mutation scores between 70–86%, indicating that the generated tests are effective in detecting faults within the source code. Furthermore, evaluation using BLEU and ROUGE metrics demonstrates a strong structural and semantic alignment between the generated unit tests and reference implementations. In terms of performance, the system achieves an average unit test generation time of 4–8 seconds, which is significantly faster than manual test creation. Based on these results, this research concludes that the Unittest Generator extension effectively improves testing efficiency, quality, and developer productivity, and serves as a practical solution to support AI adoption in modern software testing workflows.

Share this knowledge with your friends :

Actions (login required)

View Item