Skip to content

Reactive BBQ Case Study

Some people can learn RIDDL faster by looking at examples than by reading the informal definitions in the language reference or the formal definitions in the EBNF grammar. To support that mode of learning, this section decomposes the domain of a restaurant chain, Reactive BBQ.

Full Model Source

The complete RIDDL model is in the riddl-models repository. Every code snippet in this tutorial is copied verbatim from that source.

About Reactive BBQ

The Reactive BBQ domain is a familiar one for those who have taken the Lightbend Reactive Architecture Professional (LRA-P) course which uses this example throughout the course to good effect in the workshops. The same premises apply in this domain, but we have chosen to fully specify it in RIDDL.

As you might guess from the name, Reactive BBQ is a restaurant chain that serves spicy (reactive!) BBQ dishes. It doesn't exist of course, even if some people may or may not have mistaken R&R BBQ in Salt Lake City for the Reactive version in 2018.

The Reactive BBQ case study in the LRA-P course included interviews with several fictitious employees. Those interviews and the case study material have been replicated in the scenario description with thanks to Lightbend; we recommend that you read that scenario first.

What This Tutorial Teaches

This tutorial walks through a 3-domain, 11-context, 13-entity RIDDL model that demonstrates how to translate stakeholder interviews into a fully specified reactive system design. You'll learn:

  • How to decompose a business into domains and bounded contexts
  • How to model entity lifecycles with commands, events, state, and handlers
  • How adaptors route messages between contexts
  • How projectors provide CQRS read models
  • How repositories define persistence schemas
  • How to model external systems as context boundaries
  • How epics capture user journeys across contexts

Domain Structure

The model comprises three domains with eleven bounded contexts:

  • Reactive BBQ Domain — The top-level domain
    • Restaurant — 6 contexts covering dine-in, kitchen, bar, online ordering, delivery, and loyalty
    • BackOffice — 3 contexts for scheduling, inventory, and reporting
    • Corporate — 3 contexts for menu management, supply chain, and marketing

Plus external system contexts modeling third-party integrations.

Patterns Demonstrated

The Patterns page provides a cross-cutting reference to every RIDDL pattern used in the model, with links to the context pages where each pattern appears.

Personas

The case study includes interviews with key personnel that inform the domain model: