Vertical Cinema (Kotlin)

Shows how to use Domain-Driven Design, Event Storming, Event Modeling and Event Sourcing in the small Cinema domain.

MIRO Board with the complete Event Modeling: CLICK (you will receive the password during the workshop)

🚀 How to run the project?

Locally (recommended)

1. Install any distribution of Java. At least version 21 (on MacOS I recommend SDKMAN to do that) on your machine and Docker.
2. Install Docker Compose
3. Clone the repository
4. Make sure that Docker Daemon is running on your machine executing: docker ps. If you don't see output like Cannot connect to the Docker daemon you're good to go.
5. Build the project executing: ./gradlew build -x test
6. Run tests after changes with ./gradlew test
7. If you want to run the application, execute: docker compose -f docker-compose.axon.yaml up and ./gradlew bootRun

To develop the application is the best to use the Intellij IDEA IDE.

• Ultimate version with 30-days trial: https://www.jetbrains.com/idea/download/
• Community version: https://www.jetbrains.com/idea/download

Devcontainers (experimental)

Development Containers - An open specification for enriching containers with development specific content and settings. If you don't want to install anything on your machine, you can use the devcontainer from the .devcontainer directory. But it gives you slower build/test times than running the project locally.

How to run devcontainer? It depends on your IDE:

• GitHub Codespaces - just go to the repository page, click the Code button, select Codespaces tab and run new environment.
• IntelliJ IDEA - open the project in IntelliJ IDEA, right click on .devcontainer/devcontainer.json and select Dev Containers -> Create Dev Container and Clone Sources
• Visual Studio Code ...

🌐 Interacting with the Application

REST API

Access the REST API documentation at: http://localhost:3883/swagger-ui/index.html

Model Context Protocol (MCP) Server

The application exposes a Model Context Protocol Server that allows AI assistants to interact directly with the domain. You can add it to Claude Code with:

claude mcp add --transport sse Cinema http://localhost:3883/sse

🗄️ Querying Events in PostgreSQL Event Store

When using PostgreSQL as the event store (instead of Axon Server), events are stored in the domain_event_entry table. The payload and meta_data columns use PostgreSQL's Large Object (LOB) storage with oid type.

Database Connection

# Connect to the PostgreSQL database
psql -h localhost -p 6556 -U cinema_user -d cinema_db

Basic Event Queries

View recent events with readable content:

SELECT
    global_index,
    event_identifier,
    payload_type,
    aggregate_identifier,
    sequence_number,
    time_stamp,
    convert_from(lo_get(payload), 'UTF8') as payload_content,
    convert_from(lo_get(meta_data), 'UTF8') as meta_data_content
FROM domain_event_entry
ORDER BY global_index DESC
LIMIT 10;

View raw binary data:

SELECT
    global_index,
    event_identifier,
    payload_type,
    lo_get(payload) as payload_bytes,
    lo_get(meta_data) as meta_data_bytes
FROM domain_event_entry
ORDER BY global_index DESC
LIMIT 5;

Filtering Events

Events by specific aggregate:

SELECT
    global_index,
    payload_type,
    convert_from(lo_get(payload), 'UTF8') as payload_content
FROM domain_event_entry
WHERE aggregate_identifier = 'your-aggregate-id'
ORDER BY sequence_number;

Events by type pattern:

SELECT
    global_index,
    event_identifier,
    payload_type,
    convert_from(lo_get(payload), 'UTF8') as payload_content
FROM domain_event_entry
WHERE payload_type LIKE '%SeatBlocked%'
ORDER BY global_index DESC;

Events within time range:

SELECT
    global_index,
    payload_type,
    time_stamp,
    convert_from(lo_get(payload), 'UTF8') as payload_content
FROM domain_event_entry
WHERE time_stamp >= '2025-01-01T00:00:00.000Z'
ORDER BY global_index DESC;

Understanding the Schema

• global_index: Unique sequential identifier across all events
• event_identifier: UUID for the specific event instance
• payload_type: Fully qualified class name of the event (e.g., com.dddheroes.cinema.seatsblocking.events.SeatBlocked)
• payload: Event data stored as Large Object (JSON serialized by Axon)
• meta_data: Event metadata stored as Large Object (timestamps, correlation IDs, etc.)
• aggregate_identifier: ID of the aggregate that produced the event
• sequence_number: Event sequence within the aggregate (starts from 0)
• time_stamp: ISO 8601 timestamp when event was stored

Large Object Functions

PostgreSQL Large Objects require special functions:

• lo_get(oid): Retrieves the binary content of a Large Object
• convert_from(bytea, encoding): Converts binary data to text using specified encoding
• Use 'UTF8' encoding since Axon stores JSON-serialized events