Структурно-функціональне моделювання системи автоматизованого тестування адаптивних веб-інтерфейсів

• Дата: 2025
• Автори: Іванюк, Віталій, Верлань, Андрій, Мястковська, Марина, Косінов, Михайло
• Формат: Стаття
• Мова: Англійська
• Опубліковано: Kamianets-Podilskyi National Ivan Ohiienko University 2025
• Онлайн доступ: http://mcm-tech.kpnu.edu.ua/article/view/347807
• Назва журналу: Mathematical and computer modelling. Series: Technical sciences

This study addresses a critical scientific and applied problem: enhancing the efficiency of automated testing for web applications developed using Responsive Web Design (RWD). The relevance of this research arises from the widespread adoption of the Mobile First paradigm, which imposes significant challenges on Quality Assurance (QA) processes due to the need for consistent functionality across heterogeneous platforms. An analysis of conventional approaches, particularly the Page Object Model (POM) pattern, revealed their limitations in multi-platform environments. These shortcomings include exponential growth in code volume, violations of Clean Code principles, and difficulties in adapting to structural discrepancies in Document Object Model (DOM) elements across different viewports. The objective of this work is to improve testing efficiency by designing an architecture that clearly separates scenario-level business logic from interface-level technical implementation. To achieve this, a structural-functional system model is proposed, formalized as a set-theoretic tuple S = <D, P, B, T, C> where D represents test data, P page objects, B business steps, T specifications, and C the context state space. Introducing a dynamic context enables modeling page behavior as a function of a configuration vector, ensuring adaptability to varying execution conditions. The theoretical framework was implemented in practice through the development of a four-layer architecture based on the Playwright tool. A novel algorithm for dynamic context injection was introduced, enabling automatic selection of locator strategies and interaction modes (Touch/Mouse) during runtime. Experimental evaluation demonstrated that the proposed approach ensures architectural invariance of tests, facilitates efficient matrix builds in CI/CD pipelines (GitHub Actions) for parallel execution in isolated containers, and reduces code maintenance costs by 35-40%. Furthermore, it eliminates the need to duplicate scenarios for new devices, thereby significantly improving scalability and maintainability.