Esther Guerra

Verification and validation of declarative model-to-model transformations through invariants

In this paper we propose a method to derive OCL invariants from declarative model-to-model transformations in order to enable their verification and analysis. For this purpose we have defined a number of invariant-based verification properties which provide increasing degrees of confidence about transformation correctness, such as whether a rule (or the whole transformation) is satisfiable by some model, executable or total. We also provide some heuristics for generating meaningful scenarios that can be used to semi-automatically validate the transformations. As a proof of concept, the method is instantiated for two prominent model-to-model transformation languages: Triple Graph Grammars and QVT.

Deep meta-modelling with METADEPTH

Meta-modelling is at the core of Model-Driven Engineering, where it is used for language engineering and domain modelling. The OMGs Meta-Object Facility is the standard framework for building and instantiating meta-models. However, in the last few years, several researchers have identified limitations and rigidities in such a scheme, most notably concerning the consideration of only two meta-modelling levels at the same time. In this paper we present METADEPTH, a novel framework that supports a dual linguistic/ontological instantiation and permits building systems with an arbitrary number of meta-levels through deep metamodelling. The framework implements advanced modelling concepts allowing the specification and evaluation of derived attributes and constraints across multiple meta-levels, linguistic extensions of ontological instance models, transactions, and hosting different constraint and action languages.

A research roadmap towards achieving scalability in model driven engineering

As Model-Driven Engineering (MDE) is increasingly applied to larger and more complex systems, the current generation of modelling and model management technologies are being pushed to their limits in terms of capacity and efficiency. Additional research and development is imperative in order to enable MDE to remain relevant with industrial practice and to continue delivering its widely recognised productivity, quality, and maintainability benefits. Achieving scalability in modelling and MDE involves being able to construct large models and domain-specific languages in a systematic manner, enabling teams of modellers to construct and refine large models in a collaborative manner, advancing the state of the art in model querying and transformations tools so that they can cope with large models (of the scale of millions of model elements), and providing an infrastructure for efficient storage, indexing and retrieval of large models. This paper attempts to provide a research roadmap for these aspects of scalability in MDE and outline directions for work in this emerging research area.

Automated verification of model transformations based on visual contracts

Model-Driven Engineering promotes the use of models to conduct the different phases of the software development. In this way, models are transformed between different languages and notations until code is generated for the final application. Hence, the construction of correct Model-to-Model (M2M) transformations becomes a crucial aspect in this approach.Even though many languages and tools have been proposed to build and execute M2M transformations, there is scarce support to specify correctness requirements for such transformations in an implementation-independent way, i.e., irrespective of the actual transformation language used.In this paper we fill this gap by proposing a declarative language for the specification of visual contracts, enabling the verification of transformations defined with any transformation language. The verification is performed by compiling the contracts into QVT to detect disconformities of transformation results with respect to the contracts. As a proof of concept, we also report on a graphical modeling environment for the specification of contracts, and on its use for the verification of transformations in several case studies.

Supporting user-oriented analysis for multi-view domain-specific visual languages

The integration of usable and flexible analysis support in modelling environments is a key success factor in Model-Driven Development. In this paradigm, models are the core asset from which code is automatically generated, and thus ensuring model correctness is a fundamental quality control activity. For this purpose, a common approach is to transform the system models into formal semantic domains for verification. However, if the analysis results are not shown in a proper way to the end-user (e.g. in terms of the original language) they may become useless. In this paper we present a novel DSVL called BaVeL that facilitates the flexible annotation of verification results obtained in semantic domains to different formats, including the context of the original language. BaVeL is used in combination with a consistency framework, providing support for all steps in a verification process: acquisition of additional input data, transformation of the system models into semantic domains, verification, and flexible annotation of analysis results. The approach has been validated analytically by the cognitive dimensions framework, and empirically by its implementation and application to several DSVLs. Here we present a case study of a notation in the area of Digital Libraries, where the analysis is performed by transformations into Petri nets and a process algebra.

Engineering model transformations with transML

Model transformation is one of the pillars of model-driven engineering (MDE). The increasing complexity of systems and modelling languages has dramatically raised the complexity and size of model transformations as well. Even though many transformation languages and tools have been proposed in the last few years, most of them are directed to the implementation phase of transformation development. In this way, even though transformations should be built using sound engineering principles—just like any other kind of software—there is currently a lack of cohesive support for the other phases of the transformation development, like requirements, analysis, design and testing. In this paper, we propose a unified family of languages to cover the life cycle of transformation development enabling the engineering of transformations. Moreover, following an MDE approach, we provide tools to partially automate the progressive refinement of models between the different phases and the generation of code for several transformation implementation languages.

Analyzing Rule-Based Behavioral Semantics of Visual Modeling Languages with Maude

There is a growing need to explicitly represent the behavioral semantics of Modeling Languages in a precise way, something especially important in industrial environments in which simulation and verification are critical issues. Graph transformation provides one way to specify the semantics of Domain Specific Visual Languages (DSVLs), with the advantage of being intuitive and easy to use for the system designer. Even though its theory has been extensively developed during the last 30 years, it has some limitations concerning specific analysis capabilities. On the contrary, Maude is a rewriting logic-based language with very good formal analysis support, but which requires specialized knowledge. In this paper we show how a mapping between graph transformation-based specifications of DSVL semantics and Maude is possible. This allows performing simulation, reachability and model-checking analysis on the models, using the tools and techniques that Maude provides.

Bottom-up meta-modelling: an interactive approach

The intensive use of models in Model-Driven Engineering (MDE) raises the need to develop meta-models with different aims, like the construction of textual and visual modelling languages and the specification of source and target ends of model-to-model transformations. While domain experts have the knowledge about the concepts of the domain, they usually lack the skills to build meta-models. These should be tailored according to their future usage and specific implementation platform, which demands knowledge available only to engineers with great expertise in MDE platforms. These issues hinder a wider adoption of MDE both by domain experts and software engineers. In order to alleviate this situation we propose an interactive, iterative approach to meta-model construction enabling the specification of model fragments by domain experts, with the possibility of using informal drawing tools like Dia. These fragments can be annotated with hints about the intention or needs for certain elements. A meta-model is automatically induced, which can be refactored in an interactive way, and then compiled into an implementation meta-model using profiles and patterns for different platforms and purposes.

When and How to Use Multilevel Modelling

Model-Driven Engineering (MDE) promotes models as the primary artefacts in the software development process, from which code for the final application is derived. Standard approaches to MDE (like those based on MOF or EMF) advocate a two-level metamodelling setting where Domain-Specific Modelling Languages (DSMLs) are defined through a metamodel that is instantiated to build models at the metalevel below. Multilevel modelling (also called deep metamodelling) extends the standard approach to metamodelling by enabling modelling at an arbitrary number of metalevels, not necessarily two. Proposers of multilevel modelling claim this leads to simpler model descriptions in some situations, although its applicability has been scarcely evaluated. Thus, practitioners may find it difficult to discern when to use it and how to implement multilevel solutions in practice. In this article, we discuss those situations where the use of multilevel modelling is beneficial, and identify recurring patterns and idioms. Moreover, in order to assess how often the identified patterns arise in practice, we have analysed a wide range of existing two-level DSMLs from different sources and domains, to detect when their elements could be rearranged in more than two metalevels. The results show this scenario is not uncommon, while in some application domains (like software architecture and enterprise/process modelling) pervasive, with a high average number of pattern occurrences per metamodel.

Event-driven grammars: Relating abstract and concrete levels of visual languages

In this work we introduce event-driven grammars, a kind of graph grammars that are especially suited for visual modelling environments generated by meta-modelling. Rules in these grammars may be triggered by user actions (such as creating, editing or connecting elements) and in their turn may trigger other user-interface events. Their combination with triple graph transformation systems allows constructing and checking the consistency of the abstract syntax graph while the user is building the concrete syntax model, as well as managing the layout of the concrete syntax representation. As an example of these concepts, we show the definition of a modelling environment for UML sequence diagrams. A discussion is also presented of methodological aspects for the generation of environments for visual languages with multiple views, its connection with triple graph grammars, the formalization of the latter in the double pushout approach and its extension with an inheritance concept.