A Docker-ready FastAPI application implementing all five product CRUD operations with in-memory storage. This practical assignment does not use MySQL or any other database. Data is kept in the running process and resets when the container restarts.
- Python 3.11+
- FastAPI, Uvicorn, Pydantic
- Docker and Docker Compose
- pytest, Swagger UI, and Postman
- Git and GitHub
app/
config/settings.py
models/product_model.py
repositories/product_repository.py
routers/product_routes.py
schemas/product_schema.py
services/product_service.py
main.py
scripts/smoke_test.py
tests/test_product_api.py
Dockerfile
compose.yaml
requirements.txt
requirements-dev.txt
README.md
PROJECT_GUIDE.md
Routes, schemas, business logic, and storage are separate. See PROJECT_GUIDE.md for a file-by-file explanation, test procedure, and team discussion questions.
Base URL when running Docker: http://localhost:8003
| Method | Path | Purpose | Success |
|---|---|---|---|
| POST | /api/products/ |
Create a product | 201 |
| GET | /api/products/ |
List, filter, and paginate | 200 |
| GET | /api/products/{product_id} |
Get one product | 200 |
| PUT | /api/products/{product_id} |
Replace product details | 200 |
| DELETE | /api/products/{product_id} |
Delete a product | 204 |
Category filtering is case-insensitive. List pagination accepts skip and
limit. Unknown IDs return 404; invalid request data or non-positive IDs
return 422.
Example request for POST and PUT:
{
"name": "Headphones",
"description": "Wireless headphones",
"price": 25.5,
"stock_quantity": 20,
"category": "Accessories",
"status": "Active"
}Price must be positive with at most four decimal places; stock must be a
non-negative integer; status must be Active or Inactive. Names and
categories cannot be blank. Extra request fields are rejected.
From PowerShell in this project folder, with Docker Desktop running:
docker compose up -d --build
docker compose psOpen Swagger UI at http://localhost:8003/docs. To exercise all five CRUD endpoints on the Docker-hosted application (not just in an in-process test), run from the project folder:
.\.venv\Scripts\python.exe scripts\smoke_test.py --base-url http://localhost:8003The script sends real HTTP requests to the published container port and checks
health, create, list/filter/pagination, read, update, delete, invalid inputs,
and not-found responses. It exits with an error if a connection or assertion
fails. You can also use Postman with http://localhost:8003 as the base URL.
Useful checks:
curl.exe -i http://localhost:8003/
curl.exe -i http://localhost:8003/api/products/
docker compose logs --tail 50 product-serviceStop the service with docker compose down. Since storage is in memory, data
is intentionally temporary and is cleared when the application process
restarts.
python -m pip install -r requirements-dev.txt
python -m pytest -qThese tests verify API validation and behavior. The smoke-test script above separately verifies the API over HTTP through Docker.
The source repository should include application code, tests, Docker files,
documentation, and sanitized evidence. Never commit .env, passwords, access
tokens, .venv, or caches. Capture your own Docker and Swagger screenshots
after running the container, then add only those non-secret evidence files.
git status
git add README.md PROJECT_GUIDE.md app scripts tests Dockerfile compose.yaml requirements.txt requirements-dev.txt evidence reports
git commit -m "Complete in-memory Docker product CRUD"
git push