Tools for your docs: Docusaurus and MkDocs Material

In this article, I explain what tool you can select for your documentation. Specifically, I discuss the two tools that I recommend for technical writers: Docusaurus and MkDocs Material

The goal is to help you choose a documentation tool that uses docs-as-code, Markdown, and doesn’t require much time to deploy.

Introduction

While there are many documentation tools for technical writers, sometimes even experienced tech writers ask themselves: which tool should I use for my project? Your project may be API Reference or end-user documentation, online help for your product, or any other domain. In this article, I focus on the customer-facing docs, not the internal documentation that you write in wiki systems like Confluence or CMS like SharePoint Online. You can have authentication or password protection for your customer-facing docs site, of course.

Ok, so your manager or your test task requires you to come up with the best documentation solution for their product/API/you name it. They may even offer you some hints like Word, SharePoint, or even some unknown commercial tools for building docs sites. The first question I would ask, is whether they’ve been using this tool before and satisfied with it. If both answers are yes, maybe it’s not worth convincing them that your tool would do a better job. However, you can still create a demo site or show them available projects built with Docusaurus or MkDocs Material static site generators.

Docusaurus and MkDocs Material

Have you heard about Docusaurus and MkDocs Material? These are static site generators (SSGs) that build a static documentation site. These SSGs use the docs-as-code approach, Markdown, and git. They’re designed specifically for technical writers or developers who want to stand up a nice-looking docs site effortlessly and quickly. I wrote a comprehensive guide how to start working with Docusaurus a couple of years ago. As this tool is developed, some things could have changed. That’s why I’m going to go step by step again to deploy a Docusaurus documentation site.

MkDocs Material has long been on my list of the best SSGs for documentation sites. Material theme of MkDocs is designed for documentation writers. It has many features, you’d better consult their documentation.

Docusaurus

The goal is to build and deploy a test Docusaurus site. You can then reproduce the steps to build your own docs site and deploy it to public GitHub Pages.

Prerequisites

You need to have the following items installed on your computer.

Node.js

You can check if it’s already installed by typing node -v in the terminal or Command Prompt. You need v.18 or later.

Command Prompt with Node.js version

If you have an older version, remove it using Windows Add or remove programs. Then install the newest version from here.

Install Docusaurus package

Use Node.js command to install Docusaurus on your computer:

  1. Run npm init docusaurus.

  2. Type y when prompted and press Enter.

    NPM Init Docusaurus
  3. Type the name of your site (project) when prompted and press Enter.

    Project name
  4. Select the recommended classic template by pressing Enter.

    Classic template
  5. Select JavaScript by pressing Enter.

    JavaScript
  6. Type cd test-docusaurus-docs to go to the folder with installed Docusaurus.

  7. Type npm start to start a hot reload server for opening the docs site in your browser on local host.

    NPM start

Your site opens in the browser with this address: http://localhost:3000/

Docusaurus default site

Deploy Docusaurus to GitHub Pages

Now that you’ve built your site locally, you can start editing its content in Markdown and customize the site theme: CSS, logo, name, sidebar menu, etc. I don’t intend to show all these steps as I’ve described them here. Instead I will provide instructions on deploying your site to GitHub Pages, so it’s available on the internet.

  1. Use VS Code to open your Docusaurus project: File > Open Folder… and select your project name that you typed when installing Docusaurus. In my case, it’s test-docusaurus-docs.

  2. Select Source Control tab in VS Code left side panel.

    Source control panel

  3. Select Initialize Repository.

  4. Select Commit.

    Commit

  5. Enter the commit message. For example: first commit. Press Enter.

  6. Select Publish Branch.

  7. Select Publish to GitHub public repository.

    Commit

  8. Select Open on GitHub to open the project in the web version of GitHub.

    Open project on GitHub

To deploy your site on GitHub Pages:

  1. In VS Code, go to Explorer tab and select the docusaurus.config.js file that stores configuration of your Docusaurus site. In my case, the path is C:\Users\ivanc\test-docusaurus-docs\docusaurus.config.js.

  2. Change the values for the following parameters:

    • organizationName - In my case, it’s ivancheban, my GitHub account.

    • projectName - In my case, it’s test-docusaurus-docs, your Docusaurus project name you selected and published to GitHub.

    • url - In my case, it’s https://ivancheban.github.io.

    • baseUrl - In my case, it’s /test-docusaurus-docs/.

      Docusaurus.config.js values

  3. In the root folder of your Docusaurus project, create the deploy.yml file with this path: .github/workflows/deploy.yml. It means that first you create the .github folder, then workflows folder inside it, and only then the deploy.yml file. Paste the following code inside the deploy.yml file.

name: Deploy to GitHub Pages

on:
  push:
    branches:
      - main
    # Review gh actions docs if you want to further define triggers, paths, etc
    # https://docs.github.com/en/actions/using-workflows/workflow-syntax-for-github-actions#on

jobs:
  build:
    name: Build Docusaurus
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
        with:
          fetch-depth: 0
      - uses: actions/setup-node@v4
        with:
          node-version: 18
          cache: npm

      - name: Install dependencies
        run: npm ci
      - name: Build website
        run: npm run build

      - name: Upload Build Artifact
        uses: actions/upload-pages-artifact@v3
        with:
          path: build

  deploy:
    name: Deploy to GitHub Pages
    needs: build

    # Grant GITHUB_TOKEN the permissions required to make a Pages deployment
    permissions:
      pages: write # to deploy to Pages
      id-token: write # to verify the deployment originates from an appropriate source

    # Deploy to the github-pages environment
    environment:
      name: github-pages
      url: ${{ steps.deployment.outputs.page_url }}

    runs-on: ubuntu-latest
    steps:
      - name: Deploy to GitHub Pages
        id: deployment
        uses: actions/deploy-pages@v4

Continue to deploy your site to GitHub Pages:

  1. Commit and push your changes:

    • Ctrl + Shift + P.
    • Select Git: Commit All.
    • Add the commit message.
    • Ctrl + Shift + P.
    • Select Git: Push.
  2. Create a gh-pages branch in your Docusaurus project. Although you commit and push to the main branch, the gh-pages branch will be used for deployment of your site on GitHub Pages.

  3. Go to Settings in GitHub page of your project.

    Settings in GitHub

  4. Select Pages and select the gh-pages branch. Save the changes.

    GitHub Pages

  5. Go to Settings > Environments and remove the gh-pages from the limitation.

    GitHub Pages

  6. Change anything in your local files, commit and push changes. The commit to the main branch starts the site deployment. Wait while the pipeline finishes building and deploying your site. Check the built site. In my case, it’s: https://ivancheban.github.io/test-docusaurus-docs/.

MkDocs Material

The goal is to build and deploy a test MkDocs Material site. You can then reproduce the steps to build your own docs site and deploy it to public GitHub Pages.

Prerequisites

You need to have Python with pip for MkDocs. Then you can install MkDocs and the MkDocs Material packages using pip.

  1. Ensure Python is installed: You can check if Python is installed on your system by opening a command prompt and typing python --version. If Python is installed, you will see something like Python 3.11.3. If you don’t have Python installed, install it from their official website.

  2. Ensure pip is installed: You can check if pip is installed by typing pip --version in the command prompt. If pip is installed, it will display the version.

  3. Install MkDocs: Type pip install mkdocs in the command prompt. Make sure MkDocs is installed by typing mkdocs --version.

  4. Install MkDocs Material: Type pip install mkdocs-material in the command prompt. To check if MkDocs Material is installed, type mkdocs serve --help. This command should list material as an option under the --theme. If material is listed, it means that Material for MkDocs is installed correctly.

    Material theme

For more information, see MkDocs Installation and MkDocs Material Installation.

Install the MkDocs site

You can continue creating a brand new MkDocs Material site using these instructions. Or, you can fork my repo with the ready configuration:

  1. Fork or download the zipped project from here: https://github.com/ivancheban/my-project.

  2. Open the mkdocs.yml file to edit the configuration of your site.

site_name: Docs site
site_url: https://ivancheban.github.io/my-project/
nav:
    - Introduction: 'index.md'
    - User Guide:
        - 'Test': 'test-folder/test.md'
        - 'Test 1': 'test-folder/test1.md'
        - 'Test 2': 'test-folder/test2.md'
    - About:
        - 'About this site': 'about.md'
theme:
  features:
    - navigation.footer
  name: material
  custom_dir: overrides
  logo: img/logo.svg
  favicon: img/favicon.ico
  palette: 
    scheme: default
    accent: light blue
  
extra_css:
  - stylesheets/extra.css

plugins:
  - search
  - mike

extra:
  version:
    provider: mike
  social:
    - icon: fontawesome/brands/github
      link: https://github.com/ivancheban
    - icon: fontawesome/brands/linkedin
      link: https://linkedin.com/in/ivan-cheban-a24b576
  generator: false

markdown_extensions:
  - pymdownx.superfences:
      custom_fences:
        - name: mermaid
          class: mermaid
          format: !!python/name:pymdownx.superfences.fence_code_format
  - admonition
  - pymdownx.details
  - pymdownx.tabbed:
      alternate_style: true
copyright: Copyright © 2023 Ivan Cheban

To run the site on your local host, type: mkdocs serve. This starts the site in your browser with this address: http://127.0.0.1:8000/my-project/.

MkDocs local site

Deploy MkDocs Material to GitHub Pages

Now that you’ve checked that your MkDocs Material site works locally, it’s time to deploy it on GitHub as a public site.

  1. Use the steps 1–8 from deploying a Docusaurus site to GitHub for committing and pushing your MkDocs project to a GitHub repository.

  2. Create a gh-pages branch in your repository.

  3. In the web interface of your repository, Go to Settings > Pages and selected gh-pages as a branch to deploy your site from. Save the changes.

  4. At the root of your MkDocs project, create a new GitHub Actions workflow file: .github/workflows/ci.yml, and copy and paste the following contents:

name: ci 
on:
  push:
    branches:
      - master 
      - main
permissions:
  contents: write
jobs:
  deploy:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v3
      - uses: actions/setup-python@v4
        with:
          python-version: 3.x
      - run: echo "cache_id=$(date --utc '+%V')" >> $GITHUB_ENV 
      - uses: actions/cache@v3
        with:
          key: mkdocs-material-${{ env.cache_id }}
          path: .cache
          restore-keys: |
                        mkdocs-material-
      - run: pip install mkdocs-material 
      - run: mkdocs gh-deploy --force

Commit and push your changes.

Ask chatbot

Last modified December 11, 2024: added chatbot to all pages (b123fec)