Daniel Strüber

RuleMerger: Automatic Construction of Variability-Based Model Transformation Rules

Unifying similar model transformation rules into variability-based ones can improve both the maintainability and the performance of a model transformation system. Yet, manual identification and unification of such similar rules is a tedious and error-prone task. In this paper, we propose a novel merge-refactoring approach for automating this task. The approach employs clone detection for identifying overlapping rule portions and clustering for selecting groups of rules to be unified. Our instantiation of the approach harnesses state-of-the-art clone detection and clustering techniques and includes a specialized merge construction algorithm. We formally prove correctness of the approach and demonstrate its ability to produce high-quality outcomes in twoi¾źreal-lifei¾źcase-studies.

Splitting Models Using Information Retrieval and Model Crawling Techniques

In team environments, models are often shared and edited by multiple developers. To allow modularity and facilitate developer independence, we consider the problem of splitting a large monolithic model into sub-models. We propose an approach that assists users in incrementally discovering the set of desired sub-models. Our approach is supported by an automated tool that performs model splitting using information retrieval and model crawling techniques. We demonstrate the effectiveness of our approach on a set of real-life case studies, involving UML class models and EMF meta-models.

Henshin: A Usability-Focused Framework for EMF Model Transformation Development

Improved usability of tools is a fundamental prerequisite for a more widespread industrial adoption of Model-Driven Engineering. We present the current state of Henshin, a model transformation language and framework based on algebraic graph transformations. Our demonstration focuses on Henshin’s novel usability-oriented features, specifically: (i) a textual syntax, complementing the existing graphical one by improved support for rapid transformation development, (ii) extended static validation, including checks for correct integration with general-purpose-language code, (iii) advanced refactoring support, in particular, for splitting large transformation programs, (iv) editing utilities for facilitating recurring tasks in model transformation development. We demonstrate the usefulness of these features using a running example.

A Variability-Based Approach to Reusable and Efficient Model Transformations

Large model transformation systems often contain transformation rules that are substantially similar to each other, causing performance bottlenecks for systems in which rules are applied nondeterministically, as long as one of them is applicable. We tackle this problem by introducing variability-based graph transformations. We formally define variability-based rules and contribute a novel match-finding algorithm for applying them. We prove correctness of our approach by showing its equivalence to the classic one of applying the rules individually, and demonstrate the achieved performance speed-up on a realistic transformation scenario.

Towards a distributed modeling process based on composite models

The rising impact of software development in globally distributed teams strengthens the need for strategies that establish a clear separation of concerns in software models. Dealing with large, weakly modularized models and conflicting changes on interrelated models are typical obstacles to be witnessed. This paper proposes a structured process for distributed modeling based on the modularization technique provided by composite models with explicit interfaces. It provides a splitting activity for decomposing large models, discusses asynchronous and synchronous editing steps in relation to consistency management and provides a merge activity allowing the reuse of code generators. All main concepts of composite modeling are precisely defined based on category theory.

Tool support for clustering large meta-models

Ever-growing requirements, long-term evolution and modernization of software projects lead to meta-models of remarkable size, being difficult to comprehend and maintain. This paper presents a tool that supports the decomposition of a meta-model into clusters of model elements. Methods proposed in the research area of graph clustering, aiming at the desired properties of high cohesion and low coupling, have been integrated in the tool. The methods are customized not only to utilize the underlying graph structure, but also the semantic information given in meta-models. An evaluation of the tool is provided in terms of a case study.

A Tool Environment for Managing Families of Model Transformation Rules

Model transformation systems often contain families of rules that are substantially similar to each other. Variability-based rules are a recent approach to express such families of rules in a compact representation, enabling the convenient editing of multiple rule variants at once. On the downside, this approach gives rises to distinct maintenance drawbacks: Users are required to view and edit presence conditions. The complexity and size of the resulting rules may impair their readability.

Granularity of Conflicts and Dependencies in Graph Transformation Systems

Conflict and dependency analysis (CDA) is a static analysis for the detection of conflicting and dependent rule applications in a graph transformation system. The state-of-the-art CDA technique, critical pair analysis, provides its users the benefits of completeness, i.e., its output contains a precise representation of each potential conflict and dependency in a minimal context, called critical pair. Yet, user feedback has shown that critical pairs can be hard to understand; users are interested in core information about conflicts and dependencies occurring in various combinations. In this paper, we investigate the granularity of conflicts and dependencies in graph transformation systems. We introduce a variety of new concepts on different granularity levels: We start with conflict atoms, representing individual graph elements as smallest building bricks that may cause a conflict. We show that each conflict atom can be extended to at least one conflict reason and, conversely, each conflict reason is covered by atoms. Moreover, we relate conflict atoms to minimal conflict reasons, representing smallest element sets to be overlapped in order to obtain a pair of conflicting transformations. We show how conflict reasons are related to critical pairs. Finally, we introduce dual concepts for dependency analysis. As we discuss in a running example, our concepts pave the way for an improved CDA technique.

Clone Detection for Graph-Based Model Transformation Languages

Cloning is a convenient mechanism to enable reuse acrossi¾źand within software artifacts. On the downside, it is also a practice relatedi¾źto significant long-term maintainability impediments, thus generating a need to identify clones in affected artifacts. A large variety of clone detection techniques has been proposed for programming and modeling languages; yet no specific ones have emerged for model transformation languages. In this paper, we explore clone detection for graph-based model transformation languages. We introduce potential use cases for such techniques in the context of constructive and analytical quality assurance. From these use cases, we derive a set of key requirements. We describe our customization of existing model clone detection techniques allowing us to address these requirements. Finally, we provide an experimental evaluation, indicating that our customization of ConQAT, one of the existing techniques, is well-suited to satisfy all identified requirements.

Model-Based Privacy Analysis in Industrial Ecosystems

Article 25 of Regulation (EU) 2016/679 on the protection of natural persons with regard to the processing and the free movement of personal data, refers to data protection by design and by default. Privacy and data protection by design implies that IT systems need to be adapted or focused to technically support privacy and data protection. To this end, we need to verify whether security and privacy are supported by a system, or any change in the design of the system is required. In this paper, we provide a model-based privacy analysis approach to analyze IT systems that provide IT services to service customers. An IT service may rely on different enterprises to process the data that is provided by service customers. Therefore, our approach is modular in the sense that it analyzes the system design of each enterprise individually. The approach is based on the four privacy fundamental elements, namely purpose, visibility, granularity, and retention. We present an implementation of the approach based on the CARiSMA tool. To evaluate our approach, we apply it to an industrial case study.