Systematic Component-Oriented Language Reuse (SCOLAR)

• Applicant: Professor Dr. Andreas Wortmann
• Subject Area: Software Engineering and Programming Languages
• Term: since 2020
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 441207927

Software-intensive systems are developed with experts of different domains. This requires reifying their domain expertise in software, which raises the need for domain-specific languages (DSLs) to bridge the gap between the problem space of the experts’ experience and software development. Developing suitable DSLs still is prohibitively complex due to the lack of viable concepts for DSL reuse and customisation. Existing concepts either give rise to a conceptual gap between their abstractions and language definition constituents or are tied to specific technological spaces.We will conceive a novel framework for the systematic, constituent-complete (i.e., supporting concrete syntax, abstract syntax, static semantics, and transformations), reuse of textual, external DSLs with translational behaviour (txtDSLs). It will promote systematic modularity and reusability through arranging these in language families of composable txtDSL components. At the core of this framework will be an extensible conceptual model for txtDSL families over components with explicit interfaces yielding extensions that describe provided constituents meant for reuse by other language components, extension points that specify elements required by the language component, and parameters that support additional, language constituent-independent, configuration. For such txtDSL components, we will conceive a composition operator that supports the pairwise additive embedding of two language components to produce a new language component. Leveraging this operator, we will explain resolving variability of txtDSL language families and customisation of txtDSL components. Research in SCOLAR starts from the state-of-the-art in software language engineering and refines the latest conceptual models for DSL reuse into viable foundations for the systematic reuse of txtDSLs. Based on these and our contributions to engineering reusable textual DSLs, DSL variability, and systematic DSL extension, we will conceptualize language families for txtDSLs. For ongoing evaluation, we will iteratively develop a txtDSL language family modelling demonstrator that, ultimately, will be used for horizontal evaluation of our concepts in different technological spaces and for vertical evaluation by decomposing a complex language into a txtDSL language family. To this end, the demonstrator will feature modelling languages to represent families, components, and interfaces of txtDSLs, analyses over these, and control their composition.The SCOLAR conceptual framework thus enables capturing the variability and customisability of DSLs to facilitate their reuse in multi-disciplinary contexts with experts of different domains. This, ultimately, enables unprecedented systematic reuse of txtDSLs for compatible technological spaces from which researchers, DSL engineers, and educators in many domains can greatly benefit.

• DFG Programme: Research Grants