Detailseite
Projekt Druckansicht

Regelgesteuerte invasive Softwarekomposition mit strategischer Port-Graphersetzung (RISCOS)

Fachliche Zuordnung Softwaretechnik und Programmiersprachen
Förderung Förderung von 2016 bis 2020
Projektkennung Deutsche Forschungsgemeinschaft (DFG) - Projektnummer 280611750
 
Erstellungsjahr 2021

Zusammenfassung der Projektergebnisse

The goal of RISCOS was to develop a simple weaving technology. This technology should employ port graphs and port-graph rewriting systems (PGRS) to weave program and model snippets (fragments), as well as fragment components, by port-graph rewriting. Unexpectedly, it turned out during the project that the employment of port-graph rewriting systems (PGRS) must be broadened to CC-brick graph rewriting and bigraph rewriting. In these approaches, bimodal ports of components can be modeled easier. Such port modes are very important for component modeling, because IN and OUT interfaces, or at least PROVIDED and REQUIRED interfaces, must be defined for a component’s communication with its environment. The main results of the RISCOS project are: • Identification of CC-bricks and bigraphs as general formalization approaches for component composition and weaving. • Definition of Relational RAG as extension of Reference Attributed Grammars (RAG), a lightweight approach to harvest the advantages of CC-bricks and bigraph analysis and rewriting, in particular, for the definition of crosscut graphs and weavings of program and model fragments. • Provision of mature, open-source Relational RAG toolset, already in use by other projects. • Conduction of several case studies to evaluate the Relational RAG toolset. • Interoperability of the toolset with a contemporary bigraph toolkit. • Definition of a (simple) algebraic framework for feature-term weaving, and thereby, for simple, language-generic weaving of model and program fragments. However, the project did not reach all of its goals: The definition and formalization of CC-brick rewrite systems and bigraph weaving systems is still future work. Also, the definition of Relational RAG weaving technology for feature terms could not be finished. Nevertheless, the basic goal of developing the foundations of a very simple weaving technology have been laid, and the group will continue to follow the lines of RISCOS in the next years, because several knowledge transfers have already started.

Projektbezogene Publikationen (Auswahl)

 
 

Zusatzinformationen

Textvergrößerung und Kontrastanpassung