Skip to content

01-python.md

Latest commit

 

History

History
113 lines (79 loc) · 4.17 KB

01-python.md

File metadata and controls

113 lines (79 loc) · 4.17 KB

01: Python Backend Development

Scenario

Contoso is a company that sells products for various outdoor activities. A marketing department of Contoso would like to launch a micro social media website to promote their products for existing and potential customers.

As a Python developer, you're going to build a Python backend app using FastAPI. For now, you're using the in-memory feature of SQLite.

Prerequisites

Refer to the README doc for preparation.

Getting Started

Check GitHub Copilot Agent Mode

  1. Click the GitHub Copilot icon on the top of GitHub Codespace or VS Code and open GitHub Copilot window.

    Open GitHub Copilot Chat

  2. If you're asked to login or sign up, do it. It's free of charge.

  3. Make sure you're using GitHub Copilot Agent Mode.

    GitHub Copilot Agent Mode

  4. Select model to either GPT-4.1 or Claude Sonnet 4.

Prepare Custom Instructions

  1. Set the environment variable of $REPOSITORY_ROOT.

    # bash/zsh
REPOSITORY_ROOT=$(git rev-parse --show-toplevel)
    # PowerShell
$REPOSITORY_ROOT = git rev-parse --show-toplevel

  2. Copy custom instructions.

    # bash/zsh
cp -r $REPOSITORY_ROOT/docs/custom-instructions/python/. \
      $REPOSITORY_ROOT/.github/
    # PowerShell
Copy-Item -Path $REPOSITORY_ROOT/docs/custom-instructions/python/* `
          -Destination $REPOSITORY_ROOT/.github/ -Recurse -Force

Prepare Virtual Environment

  1. Make sure that you're using GitHub Copilot Agent Mode with the model of Claude Sonnet 4 or GPT-4.1.

  2. Use prompt like below to prepare virtual environment for Python app development.

    I'd like to write a Python application. But before that, I need to set up a virtual environment. Follow the instructions below.

- Your working directory is `python`.
- Identify all the steps first, which is you're going to do.
- Use `.venv` for the virtual environment.
- Use `uv` as the Python package manager.

Build FastAPI Backend App

  1. Make sure that you're using GitHub Copilot Agent Mode with the model of Claude Sonnet 4 or GPT-4.1.

  2. Add product-requirements.md and openapi.yaml to GitHub Copilot.

  3. Use prompt like below to build a FastAPI backend application.

    I'd like to build a FastAPI application as a backend API. Carefully read the entire PRD and `openapi.yaml`. Then, follow the instructions below.

- Your working directory is `python`.
- Identify all the steps first, which is you're going to do.
- Use FastAPI as the API app framework.
- Use SQLite as the database.
- Use in-memory feature of SQLite.
- The database should always be initialized whenever starting the app.
- Use `openapi.yaml` that describes all the endpoints and data schema.
- Use the port number of `8000`.
- Entrypoint is `main.py`.
- The API application should render Swagger UI page through a default endpoint.
- The API application should render exactly the same OpenAPI document through a default endpoint.
- DO NOT add anything not defined in `openapi.yaml`.
- DO NOT modify anything defined in `openapi.yaml`.

  4. Click the [keep] button of GitHub Copilot to take the changes.

  5. Once the application is built, verify if it's written properly or not.

    Run the FastAPI app and verify if the app is properly running. Also verify the OpenAPI endpoint renders exactly the same content as `openapi.yaml`.

If app running fails, analyze the issues and fix them.

  6. Open a web browser and navigate to http://localhost:8000.

  7. Click the [keep] button of GitHub Copilot to take the changes.

OK. You've completed the "Python" step. Let's move onto STEP 02: JavaScript Frontend Development.