Task Manager REST API

A production-style REST API built in Python with FastAPI and SQLAlchemy. Demonstrates clean architecture, full CRUD operations, input validation, persistence, dependency injection, and automated integration testing.

---

Project Overview

A complete backend service for managing tasks (a to-do style domain), built to demonstrate professional software engineering practices in Python:

• Clean separation of concerns (models, database, routes, application entry point)
• Auto-generated interactive API documentation (OpenAPI / Swagger)
• Strict input validation via Pydantic
• ORM-based persistence via SQLAlchemy (SQLite locally, easily swappable for PostgreSQL or Azure SQL)
• Automated integration tests with pytest — every endpoint covered

Stack: Python · FastAPI · SQLAlchemy · Pydantic · Pytest · Uvicorn

---

Endpoints

Method	Path	Purpose
GET	/health	Liveness probe for monitoring / load balancers
POST	/tasks	Create a new task
GET	/tasks	List all tasks (filterable by completed and priority)
GET	/tasks/{id}	Retrieve a single task
PATCH	/tasks/{id}	Partially update a task
DELETE	/tasks/{id}	Delete a task
GET	/stats	Aggregate statistics across all tasks
GET	/docs	Auto-generated interactive API documentation

---

Demo

Creating a task (POST /tasks)

The API validates input via Pydantic, persists the task to SQLite, and returns the created resource with 201 Created status.

Automated test suite (7/7 passing)

Every endpoint is covered by an integration test. Tests run against a fresh temporary database to keep them fast and deterministic.

---

Architecture

app/
├── main.py        # FastAPI application entry point and lifespan management
├── routes.py      # HTTP routes (CRUD + filtering + stats)
├── models.py      # SQLAlchemy ORM model + Pydantic request/response schemas
└── database.py    # Engine, session factory, dependency-injected get_db()

tests/
└── test_api.py    # 7 integration tests covering every endpoint

Key design decisions:

• Dependency injection for the database session — each request gets its own session, automatically closed afterwards (FastAPI's Depends(get_db) pattern).
• Separate Pydantic schemas for create / update / response — different operations have different validation needs (e.g., id is set by the database, not the user).
• PATCH for updates — allows partial updates without resending unchanged fields, more REST-correct than PUT.
• Cloud-ready persistence layer — DATABASE_URL is the only thing that changes to deploy on PostgreSQL or Azure SQL.

---

Running Locally

# Create and activate a virtual environment
python -m venv venv
.\venv\Scripts\Activate.ps1   # Windows PowerShell
# source venv/bin/activate    # macOS / Linux

# Install dependencies
pip install -r requirements.txt

# Start the API
uvicorn app.main:app --reload

Then open:

• API root: http://127.0.0.1:8000
• Interactive docs: http://127.0.0.1:8000/docs

---

Running Tests

pytest -v

All 7 integration tests should pass, covering health checks, CRUD operations, filtering, validation, and error handling.

---

Skills Demonstrated

Backend & APIs: REST API design, HTTP semantics, status codes, OpenAPI / Swagger Python: FastAPI, Pydantic, SQLAlchemy, async lifespan handlers, type hints Persistence: ORM modelling, dependency-injected sessions, environment-portable database layer Testing: Pytest, integration testing, fixtures, dependency overrides Software practices: Modular architecture, separation of concerns, version control, dependency management

---

Author

Kumail Janjua