This a simple booking service that allows you to book or block properties.

It is a RESTful Java 17 application with Spring Boot and H2 database using JPA.

Its developed using layered architecture, with an app, service, repository layers. Divided into modules.

For overlapping considerations, the application allow an start date in the same day of an end date. Considering for example check-in and check-out hours with proper distances to use the same day.

You can use docker-compose provided in this repo to execute the application. Just run the following command:

docker-compose up

This will serve the application on http://localhost:8888.

If you need to build the application again for some reason, you must use the --build option:

docker-compose up --build

To execute the tests you can use

docker-compose run --entrypoint "mvn test" app  

The application has the following endpoints:

GET /properties
GET /properties/{id}
POST /properties
DELETE /properties/{id}
PUT /properties/{id}

Body example:

{
    "address": "123 Main St",
    "city": "Springfield",
    "ownerName": "Sly Stone"
}

GET /properties/{id}/bookings
GET /properties/{id}/bookings/{id}
POST /properties/{id}/bookings
DELETE /properties/{id}/bookings/{id}
PUT /properties/{id}/bookings/{id}

Body example:

{
  "startDate": "2022-12-10",
  "endDate": "2022-12-11",
  "guestName": "Stevie Wonder",
  "numberOfGuests": "2",
  "bookingState": "ACTIVE"
}

GET /properties/{id}/blocks
GET /properties/{id}/blocks/{id}
POST /properties/{id}/blocks
DELETE /properties/{id}/blocks/{id}
PUT /properties/{id}/blocks/{id}

Body example:

{
  "startDate": "2022-12-01",
  "endDate": "2022-12-10",
  "reason" : "Painting"
}

A postman collection is provided in the postman folder for testing purposes.

• Don't use memory database, having a relational with some migration tool to version and execute structure changes(in pipeline or integration tests).
• Use proper environment variables for configuration.
• Use some APM tool to monitor the application. (Like datadog, newrelic, etc)
• Use some log management tool to centralize logs. (Like ELK)

• Include pagination in the list responses
• Use better Lombok annotations to avoid boilerplate code
• Track history of bookings and blocks (probably using events to be consumed by other microservices)
• Use a linter to enforce code style
• Use Richardson Maturity Model level 3, Adding HATEOAS to the responses, making it easier to consume the API. Linking related resources.

• It's possible to deal properly with booking cancelling and rebooking using the /properties/{id}/bookings/{id} endpoint, but it would be better to have a specific endpoint for those, isolating rules and making it easier to consumers.
• Create a proper endpoint for all reservation in a property (bookings and blocks).
• Don't allow to book or block a property past dates (maybe isolating booking history to a different microservice)
• Have specific entities and endpoints for property owners and guests.
• Create unit testing for validations
• Create specific unit tests for entity mappers
• Proper API documentation