Exporting Confluence page as PDF using NodeJS

exportConfluencePdf.js

import { writeFileSync } from 'fs'

/**
 * Example of how to export a Confluence Page as a PDF and writing it to disk
 * The code has poor error handling, but is hopefully easy to understand
 * 
 * The basic steps are:
 *  - Load PDF export page
 *  - Find task id
 *  - Wait for task to complete and get result path in return
 *  - Get task result which is a signed AWS S3 url allowing you to download the PDF
 *  - Call the S3 url and do whatever you want with the returned value
 * 
 * Logic was extracted from https://github.com/atlassian-api/atlassian-python-api/blob/master/atlassian/confluence.py
 */

const confluenceDomain = ''
const username = ''
// password probably works as well
const apiToken = ''
// id of page you want to export as PDF, found in the url of the page itself. Always 10 digits from what I can tell
const pageId = ''

const sleep = ms => new Promise(resolve => setTimeout(resolve, ms))
const waitForTaskCompletion = async (taskId) => {
  while (true) {
    await sleep(2000)
    const progress = await fetch(`${apiDomain}/wiki/services/api/v1/task/${taskId}/progress`, { headers })
      .then(res => res.json())

    console.log(progress)
    switch (progress.state) {
      case 'PDF_GENERATED':
      case 'IN_PROGRESS':
        break
      case 'UPLOADED_TO_S3':
        return progress.result
      default:
        throw new Error('Unknown state')
    }
  }
}

const apiDomain = `https://${confluenceDomain}.atlassian.net`
const headers = new Headers()
headers.append('Authorization', `Basic ${Buffer.from(username + ':' + apiToken).toString('base64')})}`)
headers.append('X-Atlassian-Token', 'no-check')

// This serves two purposes. It triggers the PDF generation and returns the HTML page which contains the task id
const pdfDownloadPage = await fetch(`${apiDomain}/wiki/spaces/flyingpdf/pdfpageexport.action?pageId=${pageId}`, { headers })
  .then(res => res.text())

// The line we are looking for is of format "            <meta name="ajs-taskId" content="54b4d158-2a34-4bf4-acc1-58da7f27b3dd">"
const taskId = pdfDownloadPage
  .split('\n')
  .find(line => line.includes('ajs-taskId'))
  .match(/[a-f0-9]{8}-[a-f0-9-]+/)[0]

const taskResult = await waitForTaskCompletion(taskId)

const downloadUrl = await fetch(`${apiDomain}${taskResult}`, { headers })
  .then(res => res.text())

const pdfBuffer = await fetch(downloadUrl)
  .then(res => res.arrayBuffer())
writeFileSync('./page.pdf', Buffer.from(pdfBuffer))