A Rapid Prototyping Language Workbench for Textual DSLs based on Xtext: Vision and Progress

TL;DR

This paper proposes a visual, real-time, and example-based approach to rapidly prototype and evolve textual domain-specific languages using an integrated grammar optimization framework within Xtext.

Contribution

It introduces a language workbench that combines grammar optimization, visual rule configuration, real-time preview, and example-based grammar inference to enhance rapid prototyping of DSLs.

Findings

Supports visual configuration of optimization rules

Provides real-time preview of grammar changes

Enables grammar inference from model examples

Abstract

Metamodel-based DSL development in language workbenches like Xtext allows language engineers to focus more on metamodels and domain concepts rather than grammar details. However, the grammar generated from metamodels often requires manual modification, which can be tedious and time-consuming. Especially when it comes to rapid prototyping and language evolution, the grammar will be generated repeatedly, this means that language engineers need to repeat such manual modification back and forth. Previous work introduced GrammarOptimizer, which automatically improves the generated grammar using optimization rules. However, the optimization rules need to be configured manually, which lacks user-friendliness and convenience. In this paper, we present our vision for and current progress towards a language workbench that integrates GrammarOptimizer's grammar optimization rules to support rapid prototyping and evolution of metamodel-based languages. It provides a visual configuration of optimization rules and a real-time preview of the effects of grammar optimization to address the limitations of GrammarOptimizer. Furthermore, it supports the inference of a grammar based on examples from model instances and offers a selection of language styles. These features aim to enhance the automation level of metamodel-based DSL development with Xtext and assist language engineers in iterative development and rapid prototyping. Our paper discusses the potential and applications of this language workbench, as well as how it fills the gaps in existing language workbenches.