AI-powered Car Selling Store

An AI-powered Car Selling Store App Written in FastAPI, Svelte and MongoDB.

Database

Setting your free Mongo Atlas account at https://cloud.mongodb.com/. Then, create an instance, a database, and a collection. After that, create a .env file at the root folder of the project and fill in the connection string (should be kept private), the database and collection name like below

DB_URL=""
DB_NAME=""
CARS_COLLECTION_NAME=""

In the backend/data/data_script.py, we have the code needed to upload the data in backend/data/cars_data.csv to MongoDB Atlas according to the data model defined in backend/app/models/car_model.py. Run it with

python -m backend.data.data_script

Note: If you get the pymongo.errors.ServerSelectionTimeoutError: SSL handshake failed, it is probably because you try to connect to the database from an IP address which is not on the list of allowed IPs for accessing your MongoDB database. To fix this, you'll need to add your current IP address to the IP whitelist in your MongoDB Atlas account. Or you can allow access from anywhere in the Network Access tab of MongoDB Atlas

Backend

The backend is written in FastAPI and requires Python 3.11

conda create -n carstore python=3.11
conda activate carstore
pip install -r backend/requirements.txt

To run the backend server, run the following in the terminal

bash backend.sh

Assuming the backend server is run at http://localhost:8000/, we can test it with the installed HTTPie Python package like below

http "http://localhost:8000/"

The backend is deployed on AWS Elastic Beanstalk at the URL http://car-store-fasm-dev.ap-southeast-1.elasticbeanstalk.com/

Frontend

The frontend is built using Svelte 5.0 and SvelteKit 2.0 To run the frontend do

bash frontend.sh

Tests

To run the backend tests, do pytest backend/tests

References

• Full Stack FastAPI, React, and MongoDB, published by Packt