Files
shopping-list-api/README.md
steve-w b2f7d0480a
All checks were successful
CI / test (push) Successful in 5m48s
CI / test (pull_request) Successful in 48s
CI / docker-build (push) Successful in 3m7s
CI / docker-build (pull_request) Successful in 45s
Migrate project to uv and add Docker CI
2026-04-06 16:51:36 +00:00

194 lines
3.8 KiB
Markdown

# Shopping List API
A simple internal API to track shopping list items using SQLite and FastAPI.
## Tech Stack
- Python 3.12
- FastAPI
- SQLite
- `uv` for dependency management and local workflows
- Docker for container packaging
- Gitea Actions for CI
## Local Development
### Prerequisites
- `uv` installed: https://docs.astral.sh/uv/
- Python 3.12 available locally
### Install dependencies
```bash
uv sync --dev
```
### Run the server
```bash
uv run uvicorn main:app --reload
```
The API will be available at http://localhost:8000.
## Running Tests
```bash
uv run pytest
```
Tests cover:
- Root endpoint
- Product CRUD
- List CRUD
- List items management (add, update, delete, cascade)
## Docker
### Build the image
```bash
docker build -t shopping-list-api:local .
```
### Run the container
```bash
docker run --rm -p 8000:8000 -v "$PWD/data:/app/data" -e DB_PATH=/app/data/shopping.db shopping-list-api:local
```
This stores the SQLite database on the host under `./data/shopping.db`.
## Database
SQLite database file: `shopping.db` by default (created automatically on first startup).
Schema:
- `products` (id, name, sku?, created_at)
- `lists` (id, name, created_at)
- `list_items` (id, list_id, product_id, quantity, added_at; foreign keys with cascade)
## API Reference
### Root
**GET /**
Returns a welcome message.
Response:
```json
{ "message": "Shopping List API" }
```
### Products
**POST /products**
Create a product.
Body:
```json
{ "name": "string", "sku": "optional string" }
```
**GET /products**
List all products.
**GET /products/{id}**
Get a product by ID.
**DELETE /products/{id}**
Delete a product.
### Lists
**POST /lists**
Create a shopping list.
Body:
```json
{ "name": "string" }
```
**GET /lists**
List all lists.
**GET /lists/{id}**
Get a list with its items (includes product details in items).
**DELETE /lists/{id}**
Delete a list (cascades to items).
### List Items
**POST /lists/{list_id}/items**
Add a product to a list.
Body:
```json
{ "product_id": 1, "quantity": 2 }
```
**PATCH /lists/{list_id}/items/{item_id}**
Update the quantity of an item in a list.
Body:
```json
{ "quantity": 5 }
```
**DELETE /lists/{list_id}/items/{item_id}**
Remove an item from a list.
**GET /lists/{list_id}/items**
List items in a list (includes product name and sku).
## Manual Testing
You can use curl or any HTTP client.
Example flow:
```bash
# Create products
curl -X POST http://localhost:8000/products -H "Content-Type: application/json" -d '{"name":"Paper Plates","sku":"PP-001"}'
curl -X POST http://localhost:8000/products -H "Content-Type: application/json" -d '{"name":"Bread"}'
# Create a list
curl -X POST http://localhost:8000/lists -H "Content-Type: application/json" -d '{"name":"Picnic List"}'
# Add items to the list (use IDs from previous responses)
curl -X POST http://localhost:8000/lists/1/items -H "Content-Type: application/json" -d '{"product_id":1,"quantity":2}'
curl -X POST http://localhost:8000/lists/1/items -H "Content-Type: application/json" -d '{"product_id":2,"quantity":1}'
# Get list with items
curl http://localhost:8000/lists/1
# Update quantity
curl -X PATCH http://localhost:8000/lists/1/items/1 -H "Content-Type: application/json" -d '{"quantity":5}'
# List items in list
curl http://localhost:8000/lists/1/items
# Delete item
curl -X DELETE http://localhost:8000/lists/1/items/2
# Delete list (cascades)
curl -X DELETE http://localhost:8000/lists/1
```
## CI
The repository includes a Gitea Actions workflow at `.gitea/workflows/ci.yml` that:
1. installs Python 3.12 and `uv`
2. syncs locked dependencies
3. runs the test suite
4. builds the Docker image
## Notes
- This is an internal API; security/auth is not implemented.
- For production use, consider adding stronger validation, structured logging, and a non-SQLite database if concurrency requirements grow.