Keywords:

Keywords:
Astract: In this paper, we discuss how to make a generic model versioning system language-specific by using various adaptation techniques. In particular, we recap some lessons learned during the AMOR project and outline some open challenges for adaptable model versioning systems.

Keywords:

Keywords: Model versioning, conflict resolution, Model-Driven Engineering
Astract: In most engineering disciplines, models are built as pragmatic, yet precise abstractions of huge systems. The model building process requires multiple people jointly elaborating on artifacts, which are analyzed, used to communicate among stakeholders, and act finally as construction plan for realizing the modeled system. In the field of software engineering, modeling languages such as the Unified Modeling Language (UML) provide multiple diagrams to describe various viewpoints of a system in a concrete graphical notation. While the code-centric software engineering discipline adopted those models as visual language for describing the system under study, the increasing complexity of modern software systems accompanied by ever shorter time to market constraints has asked for new techniques. The upcoming Model-Driven Engineering (MDE) approach aims at additionally exploiting models to automatically generate executable code. This paradigm shift lifts models to first-class citizens within the whole engineering process, effectively shaping the primary artifact of change undergoing the collaborative refinement from informal sketches to blueprints. This upgrowth intrinsically demands tool support for managing the models' history including merging of parallel evolved models. Optimistic versioning systems, which are already successfully applied for the management of source code, handle both issues. However, applying those systems to models fails due to the models' graph-based structure. Consequently, first dedicated model versioning systems emerged. Although current model versioning systems provide decent conflict detection facilities, they (1) ignore the graphical representation of the models, and (2) neglect conflict resolution by totally shifting the responsibility to the user.
Yet, the central role of models unifying the human-centric, collaborative abstraction and design process with the computation-centric process of generating executable systems, demands proper mechanisms to foster validity and quality of the merged model.
In this thesis, we first analyze specifics of model versioning and elaborate on the notion of conflict to improve conflict resolution respecting the central role of models. To cope with the human-centric aspect, we present a conflict aware merge strategy to calculate a tentatively merged conflict diagram as accelerator for conflict resolution retaining the graphical representation of the model. The conflict diagram unifies non-conflicting changes and materializes merge conflicts in form of annotations, rendering a coherent picture of the model's evolution. To further support the conflict resolution process, we elaborate on a conflict resolution recommender system on top of the conflict diagram, which recommends automatically executable conflict resolution patterns. Finally, to satisfy validity conditions of the computation-centric aspect, we establish a formal framework based on graph transformation theory, to showcase the feasibility of our approach.

Keywords: model-driven, modeling, versioning, model transformation, demonstration-based, software development
Astract: Model-driven engineering (MDE) is evermore adopted in academia and industry. In MDE, software models constitute the central artifacts in the software engineering process.
Developing a large software system entails the need for a large number of collaborating developers. Unfortunately, collaborative development of models is currently not sufficiently supported. Traditional versioning systems for code fail for models, because they neglect the graph-based nature of models.
A few dedicated model versioning approaches have been proposed. However, these approaches suffer from four major deficiencies. First, they either support only one modeling language or, if they are generic, they do not consider important specifics of a modeling language. Second, they do not allow the specification of composite operations such as refactorings and thus, third, they neglect the importance of respecting the original intention behind composite operations for detecting conflicts and constructing a merged model. Fourth, the types of detectable conflicts is insufficient and not extensible by users.
To tackle these deficiencies, we present an adaptable model versioning framework, which offers out-of-the-box support for all modeling languages, but also allows to be adapted for for specific modeling languages. For easily specifying language-specific composite operations, we propose a novel technique called model transformation by demonstration. These operation specifications constitute the adaptation artifacts for enabling the detection of applications of specified composite operations. Furthermore, we introduce techniques for detecting additional types of conflicts caused by concurrently applied composite operations as well as for revealing potentially unfavorable merge results.