Manpower API

Purpose

Technical Test for Developer From PylonAI

Requirements

Recommended (Docker)

• Docker ≥ 20.10
• Docker Compose ≥ 2.0 (comes bundled with recent Docker Desktop versions)

Running in containers means you do not need to install Python, database drivers or system packages locally.

If you prefer to run natively

• Python ≥ 3.12 — create a virtual environment.
• Microsoft ODBC Driver 18 for SQL Server — installation guide.
• Install Python dependencies: pip install -r requirements.txt.

Update Environment Variable

Copy .env.example

cp .env.example .env

and update

DB_USERNAME=sa
DB_PASSWORD=pw
DB_SERVER=localhost
DB_NAME=test
DB_TABLE_PREFIX=test

Warning: In case got error cannot get env please restart terminal

Seed Dump Data for Testing

python seed.py

Manual run (without Docker – optional)

To launch uvicorn:

uvicorn manpower_api.main:app --reload --port=8000

Access Flutter Web Built

http://localhost:8000

API Documentation Swagger

Then load the fancy interactive docs page at

http://localhost:8000/docs

How to Run Frontend with Backend (MacOS and Linux Only):

Clone Frontend Project

git clone https://github.com/mengtongun/manpower_app.git

Setup and Install Dependencies

cd manpower_app

flutter pub get

Build Flutter Web

fluter build web

Replace Static Web File in Backend)

cd ..
rm -rf web && cp manpower_app/build/web web

Run in Local

uvicorn manpower_api.main:app --port=8000

Project Structure

manpower_api
├── manpower_api
│   ├── auth
│   │   ├── router.py
│   │   ├── schemas.py  # pydantic models
│   │   ├── models.py  # db models
│   │   ├── config.py  # local configs
│   │   ├── constants.py
│   │   ├── exceptions.py
│   │   ├── security.py
│   │   ├── service.py
│   └── employee
│   │   ├── router.py
│   │   ├── schemas.py
│   │   ├── models.py
│   │   ├── constants.py
│   │   ├── exceptions.py
│   │   ├── service.py
│   ├── config.py  # global configs
│   ├── models.py  # global models
│   ├── exceptions.py  # global exceptions
│   ├── database.py  # db connection related stuff
│   └── main.py
├── web # static build file from flutter web
│   └── index.html
│   └── ...
├── requirements.txt
├── README.md
├── .env
├── .gitignore

Details at

https://fastapi.tiangolo.com/tutorial/

and

https://fastapi.tiangolo.com/tutorial/sql-databases/

Features

• Authentication: Basic HTTP authentication with secure password hashing (bcrypt).
• Employee Management:
  • List employees with pagination: GET /v1/api/employees
  • Update employee details: PATCH /v1/api/employees/{manpower_id}
  • Export employees as CSV: GET /v1/api/employees/csv
• Health Check: GET /health verifies application ↔ database connectivity.
• Interactive API Docs: Swagger UI available at http://localhost:8000/docs.
• Static Front-End: Compiled Flutter web application served from the root path /.

Technology Stack

Layer	Technology
Language	Python ≥ 3.12
Web Framework	FastAPI
ASGI Server	Uvicorn
ORM	SQLAlchemy
Database Driver	pyodbc (Microsoft SQL Server)
Authentication	bcrypt + HTTP Basic Auth
Data Utilities	pandas (CSV export)
Containerisation	Docker & Docker Compose
Front-End	Flutter Web (static files)

Local Development

Option 1 – Docker (recommended)

cp .env.example .env   # customise credentials if needed
make init-db
make seed
make build
make up

The API will now be reachable at http://localhost:8000.

Makefile shortcuts

Command	Description
make build	Build (or rebuild) the Docker images. Run this after updating the Dockerfile or dependencies.
make up	Start all containers in detached mode.
make down	Stop and remove containers, networks and volumes (drops DB data).
make init-db	Create the database schema inside the SQL Server container. Run once after the first make up.
make seed	Execute seed.py inside the API container to populate sample data.
make logs	Follow logs from all running containers (Ctrl+C to stop).
make ps	Show current container status.

🔄 Quick rebuild cycle: when you change backend code or dependencies run make build && make up to recreate the manpower_api image and restart the stack.

🧹 Start from scratch: make down removes containers and the persisted mssql_data volume, giving you a clean slate.

Option 2 – Manual (native Python)

Create a virtual environment, install dependencies and run the server:

python -m venv .venv && source .venv/bin/activate
pip install -r requirements.txt
uvicorn manpower_api.main:app --reload --port 8000

Environment Variables

Name	Description	Default
DB_USERNAME	Database username	sa
DB_PASSWORD	Database password	pw
DB_SERVER	Hostname / IP of SQL Server	localhost
DB_NAME	Database name	test
DB_TABLE_PREFIX	Prefix prepended to every table name	test

Seed Sample Data

Populate the database with sample records for quick testing:

python seed.py

---