利用Sikuli進行以模型為基礎的圖形介面測試

Model-Based GUI Testing Using Sikuli

黃國僑(Kuo-Chiao Huang)

指導教授 ： 王凡

國立臺灣大學/電機資訊學院/電機工程學研究所/碩士(2010年)

https://doi.org/10.6342/NTU.2010.01605

摘要

本論文提出一個以模型為基礎的圖形介面測試架構， 並且測試的過程中採用圖形化測試腳本， 在這個架構中，待測物的規範由使用者經由網頁輸入的狀態轉移表(state-transition table)來描述， 我們開發了兩套自動化測試工具可以建立Sikuli的圖形化測試腳本並且利用特徵圖片去檢查圖形介面程式的錯誤。第一個工具會根據規範檢查待測物的頁面是否被定義、在這頁面中所有的按鈕是否被定義或是有缺失。第二個工具，首先經由狀態轉移表建立一個自動機， 以模型為基礎的方式經由特定的測試標準決定測試計畫， 我們實做了兩個測試計畫，隨機與深度優先來計算按鈕覆蓋率button coverage)來展示了這個可能性， 之後我們提出一些增加測試腳本效率的技巧， 最後我們將實驗進行在定點看護(point-of-care)的醫療系統之上，並成功得發現數個錯誤。

關鍵字

圖形介面測試 ； 圖形化測試腳本 ； 模型基礎 ； 軟體測試

並列摘要

We present a framework of model-based GUI (graphical user-interface) testing with visual test scripts. In the framework, the system-under-test (SUT) specifications are state-transition tables input from a web interface. We developed two automatic testing tools that construct visual test scripts in Sikuli and use feature images to check for bugs in GUI programs. The first tool explores the pages of the SUT and applies test cases to check if a page is defined, if all buttons in the pages are defined, and if some buttons are missing in a page according to the specification. The second tool constructs an automata from the state-transition table and then uses model-based techniques to construct test plans for certain testing criteria. We show the feasibility of the idea with two test strategies, random and depth-first search, and the button-coverage criterion. We then discuss some techniques for the construction of efficient visual test scripts. We report our experiment with a point-of-care medication administration system. Our tools successfully helped in discovering several bugs.

並列關鍵字

GUI testing ； visual test scripts ； experiment ； model-based ； event coverage ； software testing ； specification

參考文獻

[25] T. Yeh, T.-H. Chang, and R. C. Miller. Sikuli: using gui screenshots for search

[6] R. Alur and D. L. Dill. A theory of timed automata.

[7] F. Belina and D. Hogrefe. The ccitt-specification and description language sdl. Computer

[8] E. Borger and R. Stark. Abstract State Machines: A Method for High-Level System

[9] R. Brunelli. Template Matching Techniques in Computer Vision: Theory and Practice.