🗣️ Week 02 Lecture

REST APIs and Introduction to FastAPI

Author

Published

27 January 2025

Last Updated: 26 January 2025, 22:20.

Welcome to Week 02 of DS205, where we will explore how to transition from data exploration to serving data through APIs.

📍 Session Details

Date: Monday, 27 January 2025
Time: 10:00 am - 12:00 pm
Location: KSW.1.01

🗣️ Lecture Content

1. Recap of Week 01

Discuss how easy/difficult it was to explore the ASCOR dataset independently.
Capture any sources of confusion about Python and pandas that needs to be reviewed in the course (if any is mentioned).

2. Introduction to REST APIs

What are APIs?
- Definition and purpose.
Key concepts:
- All the layers (client, server, etc.)
- RESTful design: Resources, endpoints, HTTP methods.
- JSON response structure: Flat vs nested.
- Status codes: 200, 404, 500.

Visual Aids: Diagram showing the structure of RESTful API communication.

3. FastAPI Overview

Key features: Type hints, automatic validation, async support.
Code examples:
- A simple "hello world" API.
- Launching a basic FastAPI app with Uvicorn.
- Demonstrating the auto-generated documentation at /docs.

4. Building blocks of a simple API

Endpoints: Define the structure of the API.
Models: Define the data structure for request and response bodies.
Routes: Define the URL paths for the endpoints.

5. Designing APIs for ASCOR Data

If time allows, a brainstorming session on designing an API for the ASCOR dataset.
Otherwise, group discusses the following API endpoints:
- /v1/indicators: Serve indicators across all countries.
- /v1/country-data: Return emission trends for a selected country.
Discuss user technical requirements and potential use cases.

🎬 Lecture Slides

Use keyboard arrows to navigate. Select the slides below or view fullscreen.

Or download the slides directly as a PDF:

👉 Next Steps

Gain more knowledge depth by browsing the FastAPI Official Tutorial.
Access the 💻 W02 Lab materials to see how we will practice building APIs

🎥 Session Recording

Let’s hope the lecture capture works this time! 🤞 If there are no issues, I’ll make the lecture recording available on Moodle on the afternoon of the lecture.