Protocol-Based Interactive Debugging for Domain-Specific Languages
Nominated for Best Paper
Interactive debuggers are established tools used by developers to understand programs and localize faults. They are equally valuable in the context of model-driven development, when working on executable behavioral models.
However, development costs of interactive debuggers for Domain-Specific Languages (DSLs) can be significant. In order to mitigate these costs, several reusable DSL-agnostic debugging solutions have been proposed. We argue that the applicability of these solutions is limited by being tied to a fixed set of debugging services, a specific language engineering approach, or a particular user interface.
In this paper, we present a novel approach to provide interactive debugging services for executable DSLs through a reusable generic architecture. We propose a protocol allowing a generic interactive debugger to communicate with heterogeneous DSL runtimes, both for controlling the execution and for configuring the debugger with domain-specific breakpoints. The proposed debugger can itself be controlled using a reinterpretation of the Debug Adapter Protocol (DAP), for an effortless integration in existing Integrated Development Environments (IDEs) that support it.
Using a prototype implementation based on JSON-RPC and two heterogeneous DSL runtimes, we show that our approach provides an off-the-shelf reusable interactive debugger that supports meaningful domain-specific breakpoints, and that can be used with minimal effort within an IDE such as Visual Studio Code.
Thu 20 JulDisplayed time zone: London change
13:45 - 15:15
|Model-Driven Engineering for Augmented Reality
Rubén Campos-López Universidad Autónoma de Madrid, Esther Guerra Universidad Autónoma de Madrid, P: Juan de Lara Autonomous University of Madrid, Alessandro Colantoni Johannes Kepler University Linz, Antonio Garmendia Universidad Autónoma de MadridDOI Pre-print Media Attached
|Protocol-Based Interactive Debugging for Domain-Specific LanguagesNominated for Best Paper
P: Josselin Enet Nantes Université, Erwan Bousse Nantes Université, Massimo Tisi IMT Atlantique, LS2N (UMR CNRS 6004), Gerson Sunyé Nantes UniversityDOI
|Streamlining the Development of Hybrid Graphical-Textual Model Editors for Domain-Specific LanguagesNominated for Best Paper
P: Ionut Predoaia University of York, Dimitris Kolovos University of York, Matthias Lenk NetApp, Antonio Garcia-Dominguez University of YorkDOI