Bachelor 2022 - Anders Larsen & Theis Tengs

This action converts your markdown files into the specified Atlassian Confluence space.

First you need to create an API token for the user, you want to use for the action. We recommend that you create a new user, that is only used for this action, in order to get the full benefits from the action. This user should also have full permission to make changes on the space. See here for how to set up your API token.

fileslocation

Optional

The path from the root of your repository, to the folder containing the markdown documentation. Default ./.

should_upload

Optional

A boolean value determining whether the action if run on a push event should upload. Defaults to 'true'

confluence_url

Required

The base URL for the Atlassian network. Follows the form: 'https://[your-network-name].atlassian.net'

confluence_space_key

Required

The key of the space that is being uploaded to. Can be found in the URL for your space. More info here

auth_username

Required

The email used for the user connected to the API token. We recommend setting this as a GitHub secret.

auth_api_token

Required

The API token generated for the user. We recommend setting this as a GitHub secret.

is_preview

Optional

Determines weather or not this worklow should only upload a preview. When uploading a preview, make sure to use a dedicated preview space. A preview only showcases the nescesarry pages to show the pages that have been made changes to.

Ex. If a the only change that has been made, is to a child page of some other page, then the parent page and child page will be uploaded, and the rest of the space will be empty.

With GitHub secrets

jobs:
  markdown_to_confluence_job:
    runs-on: ubuntu-latest
    name: Converting Markdown to Confluence
    steps:
      - name: Checkout
        uses: actions/checkout@v3
      - name: Conversion step
        uses: TTengs/MarkdownToConfluence@v3
        with:
          fileslocation: './documentation'
          should_upload: true
          confluence_url: 'https://network.atlassian.net/wiki'
          confluence_space_key: 'spaceKey'
          auth_username: ${{ secrets.AUTH_USERNAME }}
          auth_api_token: ${{ secrets.AUTH_API_TOKEN }}
          is_preview: false

If you want to create a pipeline, that publishes a preview on a pull-request, and then update the public space once the pull-request is accepted, you will have to set up these two workflows:

preview.yaml

on:
  pull_request:
    branches:
      - "main"
    paths:
      - "documentation/**"
      
jobs:
  create-preview:
    runs-on: ubuntu-latest
    name: Uploading Preview
    steps:
      - name: Checkout
        uses: actions/checkout@v3
      - name: Conversion step
        uses: TTengs/MarkdownToConfluence@v3
        with:
          fileslocation: 'documentation'
          confluence_url: 'https://network.atlassian.net'
          confluence_space_key: 'PreviewSpaceKey'
          auth_username: ${{ secrets.AUTH_USERNAME }}
          auth_api_token: ${{ secrets.AUTH_API_TOKEN }}
          is_preview: true

upload.yaml

on:
  push:
    branches:
      - "main"
    paths:
      - "documentation/**"

jobs:
  push_to_confluence:
    runs-on: ubuntu-latest
    name: Converting Markdown to Confluence
    steps:
      # To use this repository's private action,
      # you must check out the repository
      - name: Checkout
        uses: actions/checkout@v3
      - name: Conversion step
        uses: TTengs/MarkdownToConfluence@v3
        with:
          fileslocation: 'documentation'
          confluence_url: 'https://network.atlassian.net'
          confluence_space_key: 'PublicSpaceKey'
          auth_username: ${{ secrets.AUTH_USERNAME }}
          auth_api_token: ${{ secrets.AUTH_API_TOKEN }}

The structure of the markdown files in the filesystem, dictates the structure of the pages on confluence.

All files should be located in a single folder, which is the one specified in fileslocation. This folder will not contain any markdown files (if so, they won't be recognized as pages), but only folders, a prefix- and settings-file. Every subfolder under this folder, will be considered a page, these are referenced later as a "folderpage".

A folderpage can contain other folderpages, as well as markdown files. The file containing the content of the folderpage, MUST be named "index.md". Any other markdown files inside the folderpage, will be considered as a child page.

A child page can also be defined by another folderpage inside a folderpage. This is only recommended, if you need child pages for that folderpage as well. Otherwise, we recommend simply using multiple markdown files.

The name of a folderpage dicates the name of the page in confluence.

Folderpages with multiple markdown files, will contain child pages named after the name of the markdown file.

Any folderpage can contain a prefix.txt file, dictating the prefix for all it's child pages.

The following filestructure:

Will look like this, once uploaded to Confluence:

This tool comes with a number of modules, which add extra markdown syntax, specialized for Confluence and other tools.

• Jira Tickets
• Attachments
• Mermaid.js
• Table of Contents
• Trello Boards