integration-catalog

The integration catalog contains a list of connectors that represent possible application integrations and provides a set of available operations that can be performed on them. It serves as a central point for managing application integrations, allowing users to easily browse, upload, or download existing connectors.

---

Running with Docker

Prerequisites

Make sure you have installed:

• Docker
• Docker Compose (v2+)

Also, verify that the following secret files exist in docker/secrets/:

• db_password.txt
• github_token.txt
• jenkins_token.txt
• jenkins_username.txt

1️⃣ Navigate to the Docker folder

cd docker

This folder contains docker-compose.yml and entrypoint scripts.

2️⃣ Build and start the containers

docker-compose up --build

• --build ensures that Angular frontend and Spring Boot backend are rebuilt.
• Two containers will start:
  • db → PostgreSQL database (with init scripts executed automatically)
  • integration-catalog → Spring Boot application with Angular 19.2.11 frontend

3️⃣ Verify the services

• Spring Boot app: http://localhost:8080
• PostgreSQL database: port 5432

Check logs:

docker-compose logs -f integration-catalog
docker-compose logs -f db

4️⃣ Stopping and cleaning up

• Stop containers:

docker-compose down

• To remove persistent DB data and restart clean:

docker-compose down -v
docker-compose up --build

Notes

• The application reads secrets from the mounted secret files and sets them as environment variables automatically.
• SQL scripts in config/sql/ is executed only on the first database initialization.

---

Database schema versioning

The database schema version is tracked in the database_version table (the current version is the highest version row). On startup the application compares it with the version it requires (DatabaseSchemaVersionValidator.REQUIRED_VERSION) and refuses to start with a clear error message when the database is outdated (or newer than the application build).

• Fresh database: config/sql/postgres.sql creates the schema already at the current version. In Docker this happens automatically on the first start of the db container.
• Existing database: re-run the whole cumulative upgrade script:

psql -v ON_ERROR_STOP=1 -U integration_catalog -d integration_catalog -f config/sql/postgres-upgrade.sql

The script has one section per schema version and is safe to run repeatedly. From version 5 on, each section is a call apply_change(N, ...) that executes only when the database version is lower than N, records the version row and commits — the SQL inside no longer has to be idempotent. Use plain psql (not pgAdmin or other tools with their own transaction handling): apply_change commits internally, which fails inside a wrapping transaction block.

Adding a schema change

1. Append a new -- region version N: <name> section to config/sql/postgres-upgrade.sql:

   call apply_change(N, '<short description>', $aa$
   <any SQL, does not have to be idempotent>
   $aa$);

   If a statement depends on a previous one being committed (typically ALTER TYPE ... ADD VALUE followed by a use of the new value), split them into two apply_change calls (two versions).
2. Make the same change in config/sql/postgres.sql and bump the version inserted at the end of that script.
3. Bump REQUIRED_VERSION in DatabaseSchemaVersionValidator.

---

License

This project is licensed under European Union Public License.