Spicy.ai - Full Stack

 🌶 Live demo: https://spicy.pascalgiguere.dev

This project is a fork of the Zesty.ai engineering test. It's a modified version of the backend-test branch that incorporates features from the Zesty.ai full-stack engineering test.

The back-end test assignment instructions can be found in backend-assignment.md. The full-stack test assignment instructions can be found in fullstack-assignment.md.

The README of the back-end test can be found in backend-test.md.

Implemented features (back-end)

See the features previously implemented as part of the backend-test branch in backend-test.md.

Implemented features (front-end)

• Search
  • /search
• Save for Later
  • /favorites
  • Implemented using local storage (no user authentication)
• Freestyle
  • Display all properties
    • /
  • Display fake property names and risk levels
    • Pseudo-random, deterministic formulas based on property ID, so that the name and risk level are always the same for a given property
  • Cool logo
    • Why not?

Setup

A single Express app is used to serve both the API and the static front-end files. As such, a single Docker image is used for both the back-end and front-end.

You can start Spicy.ai using either of these three methods:

1. Build and run locally (with Docker)

This project requires Docker and docker-compose to be installed.

From the project root directory:

PORT=3031 docker-compose up

Note: This project requires a lot of memory to build. If you get a "Heap out of memory" error at build time, you may need to increase your max Docker memory from 2 GB to 4 GB in your Docker settings.

2. Build and run locally (without Docker)

This project requires Node 12, Yarn 1, PostgreSQL 9 and PostGIS 2 to be installed.

Once the database has been initialized and is running, start the API from the project root directory:

export PORT=3031
export DATABASE_URL=postgres://postgres@localhost:5432/zesty
yarn
cd backend/ && yarn start

3. Run a pre-built Docker image

If not already done, login to GitHub with your username and a personal access token

docker login docker.pkg.github.com --username <your_username>

From the project root directory:

PORT=3031 docker-compose -f docker-compose-prebuilt.yml up

Usage

Once the API and database are running, you'll be able to access Spicy.ai from your host machine on the port defined above:

# Front-end
curl http://localhost:3031/
# -> <index.html>

# Back-end
curl http://localhost:3031/api
# -> "Spicy.ai API"