A graph-based algorithm for consistency maintenance in incremental and interactive integration tools

Abstract

Development processes in engineering disciplines are inherently complex. Throughout the development process, the system to be built is modeled from different perspectives, on different levels of abstraction, and with different intents. Since state-ofthe-art development processes are highly incremental and iterative, models of the system are not constructed in one shot; rather, they are extended and improved repeatedly. Furthermore, models are related by manifold dependencies and need to be maintained mutually consistent with respect to these dependencies. Thus, tools are urgently needed which assist developers in maintaining consistency between inter-dependent and evolving models. These tools have to operate incrementally, i.e., they have to propagate changes performed on one model into related models which are affected by these changes. In addition, they need to support user interactions in settings where the effects of changes cannot be determined automatically and deterministically. We present an algorithm for incremental and interactive consistency maintenance which meets these requirements. The algorithm is based on graphs, which are used as the data model for representing the models to be integrated, and graph transformation rules, which describe the modifications of the graphs to be performed on a high level of abstraction.

Uncertainty in bidirectional transformations

“…Concerning the multiplicity of solutions in bidirectional transformations, most of the existing languages are deterministic. However, in [1] the authors propose PROGRES, a bidirectional transformation solution based on Triple Graph Grammars (TGGs) which is able to recognize ambiguous mappings and in case resolve them by interactively asking the user. In [24] the QVT-R bidirectional transformation language is discussed.…”

Checking architectural compliance in component-based systems

“…For example, [26] proposes an approach based on Communicating Sequential Processes (CSP), which might be more appropriate for behavioral consistency aspects. Other approaches are based on graphs and graph transformations, as for example [27] and [28]. However, examples mostly do not deal with the consistency or views on the same or a similar level of abstraction, and not with consistency towards an intensional model.…”

Feature-based classification of bidirectional transformation approaches

“…These tools can come with artifact editors that allow the definition of these changes and check their correctness, before activating the synchronization engine. We divide this feature ( The (information-preserving) bidirectional engine for TGG in [40] supports all possible changes in the correspondent artifacts, while the ATL engine in [60] does not support addition of elements to the target 4 The kind of change on the other artifacts that reflect the change made by the user is not necessarily the same as the ification Support). Note that in systems like Vdl [45] distinction between addition, removal and modification is somehow meaningless (being the system based on constructors like cons and nil in Lisp).…”