CyberCommandes

Delivery / tour-planning management application — Projet Intégrateur 2025-2026 (Université Grenoble Alpes).

A full-stack application built with an Angular 22 frontend and a Spring Boot 4 / Java 21 multi-module backend, backed by PostgreSQL. The repository is fully containerized and ships with CI/CD, image scanning, Kubernetes manifests and a Prometheus/Grafana monitoring stack.

Demo

Grid Strategy planning 399 deliveries around Grenoble with 3 vehicles (664 km, 18.45 h), with the cost/energy/CO₂ comparison table:

Every run is saved to the Historique tab for side-by-side comparison — here K-Means vs Clarke & Wright on the same delivery set:

Architecture

 ┌────────────┐ /api ┌────────────┐ JDBC ┌────────────┐
Browser │ frontend │ ────────────▶ │ backend │ ─────────────▶ │ PostgreSQL │
────────▶ (nginx) │ │(Spring Boot)│ │ │
└────────────┘ └─────┬──────┘ └────────────┘
│ /actuator/prometheus
▼
┌──────────────────┐
│ Prometheus + Grafana │
└──────────────────┘

Layer	Tech
Frontend	Angular 22, Leaflet, served by nginx
Backend	Spring Boot 4, Java 21, JPA, MapStruct, Actuator
Database	PostgreSQL 16
CI/CD	GitHub Actions, GHCR, Trivy, Dependabot
Orchestr.	Docker Compose, Kubernetes (kustomize)
Monitoring	Prometheus, Grafana

Quick start (Docker)

The whole stack runs with a single command:

cp .env.example .env # adjust values if needed
make up # or: docker compose up --build -d

Service	URL
Frontend	http://localhost:4200
Backend	http://localhost:8080
Swagger UI	http://localhost:8080/swagger-ui.html
Health	http://localhost:8080/actuator/health

With monitoring (make monitoring): Grafana on http://localhost:3000 (admin / GRAFANA_ADMIN_PASSWORD), Prometheus on http://localhost:9090.

Run make help to list every available command.

Seed data

Once the stack is up and the backend has created the schema, load the demo domain data (addresses, clients, orders, products):

make seed # loads docker/seed-data.sql into the running database

make seed waits until the backend has created the schema, then loads the data. The demo orders' delivery date is set to the day you run it (CURRENT_DATE), which is the date the UI queries on startup. The seed is re-runnable: run it again any day to wipe and reload the demo rows with fresh dates — the distance cache is left untouched.

Note on the distance matrix. The distance_cache_entity table is a precomputed, regenerable cache (~99% of the original dump, ~69 MB) and is intentionally not tracked in git — only the small domain seed (docker/seed-data.sql, ~460 KB) is. The application recomputes and caches distances on demand. If you have the full offline dump (docker/matrice-complete.sql, git-ignored), you can load it the same way: docker compose exec -T postgres psql -U postgres -d local < docker/matrice-complete.sql.

Kubernetes

Manifests live in k8s/ and are assembled with kustomize:

kubectl apply -k k8s/ # or: make k8s-deploy

Set the image tags produced by the CD pipeline before deploying:

cd k8s
kustomize edit set image \
ghcr.io/luunpl/cybercommandes-backend=ghcr.io/luunpl/cybercommandes-backend:<tag>

CI/CD

• CI (.github/workflows/ci.yml) — builds and tests the backend (mvn verify, H2) and the frontend (npm ci, unit tests, production build) on every push and pull request.
• CD (.github/workflows/cd.yml) — builds the backend and frontend Docker images, scans them with Trivy (results uploaded to the GitHub Security tab), and pushes them to GHCR on main and version tags.
• Dependabot keeps Maven, npm, GitHub Actions and Docker base images up to date.

---

Développement (FR)

Base de données locale

Vous pouvez créer une base de données locale via docker/docker-compose-local.yml.

Prérequis

1. Docker et docker-compose installés.

Démarrage / arrêt

docker compose -f docker/docker-compose-local.yml up --detach
docker compose -f docker/docker-compose-local.yml down

Variables d'environnement

Copiez .env.example vers .env. Les variables consommées par le backend :

DB_USERNAME=<username>DB_PASSWORD=<password>DB_URL=jdbc:postgresql://<server>:<port>/<database>DB_JPA_DDL_AUTO=<create-drop | create | update | validate | none>

Compiler et démarrer le serveur

mvn -f pom.xml -pl :server -am clean install # compiler
make backend # démarrer Spring Boot

Frontend

La doc d'installation du frontend est dans client/README-INSTALLATION.md.

cd client && npm install && npm start