Supporting Domain-Independent Model Management through Automated REST APIs Generation

The increasing complexity of modern software systems has made effective model management a crucial challenge in software engineering. Model-Driven Engineering (MDE), which places models at the core of the development process, has emerged as a promising approach to managing this complexity. The MDE paradigm is a software development method that aims to improve productivity and software quality by using models as primary artifacts in all aspects of the software engineering process. In this approach, models serve as abstract representations of a system, enabling data manipulation, property validation and ultimately transformed into application artifacts such as code, documentation, and tests. While MDE pretended to be the silver bullet for software engineering, its adoption in distributed, web-based environments has revealed significant limitations, particularly in scalability when managing large and complex models within industrial contexts, ensuring seamless interoperability across diverse modeling tools, and simplifying model querying and manipulation, which often requires domain-specific knowledge that poses challenges for non-expert users. Traditional model management tools often falter as models grow larger and more complex. These tools are generally ill-suited to handle the dynamic and rapidly evolving nature of modern systems, particularly in distributed, web-based environments, where managing and handling large-scale models efficiently is essential. Additionally, existing model-based tools often operate in isolation, relying on proprietary languages and repositories, which hinder seamless model exchange and collaborative development. The lack of interoperable solutions leads to model silos, making models inaccessible to other model-based tools and platforms that rely on them. Another key challenge in model management is model querying and manipulation, which often require domain-specific knowledge that may not be accessible to non-expert users. Many existing solutions demand familiarity with custom query languages and intricate model structures, making it difficult for users without deep MDE expertise to efficiently access and manipulate model elements. A more user-friendly and intuitive approach is needed to lower the barrier to model interaction and enhance accessibility. Current web-based model management approaches typically rely on static generative solutions to generate REST APIs. These solutions require predefined models and manual intervention, leading to inefficiencies and delays when adapting to evolving system requirements. Such limitations render them unsuitable for dynamic distributed environments, where models are not known in advance and REST APIs must be provisioned on-the-fly to support adaptive model management. This research addresses these challenges by developing a middleware solution that facilitates scalable, interoperable, and efficient model management in distributed, web-based environments. The middleware serves as a bridge between model repositories and modelling tools, enabling efficient model exchange, manipulation, and querying in an environment independent of any specific technology. This solution enables dynamic REST APIs provisioning for model management, ensures scalability for handling large and complex models, and enhances interoperability across diverse model-based tools and platforms. Specifically, the contributions of this work are threefold. (i) Dynamic REST API provisioning, by designing and implementing a middleware solution capable of generating REST APIs on-the-fly for any given domain model, while providing a more user-friendly approach to model querying and manipulation. (ii) Scalability and efficient model handling to efficiently query and manipulate large and complex models. (iii) Enhancing interoperability by enabling seamless model exchange across heterogeneous model-based tools and platforms. This work makes a significant contribution to advancing MDE practices by addressing critical limitations in existing model management solutions, enabling more efficient, adaptable, and scalable model management in distributed, web-based environments.