Extract URL paths from a HAR file

har2paths.js

// Useful for generating a list of paths for performance testing using `wrk`
// and https://github.com/timotta/wrk-scripts/blob/master/multiplepaths.lua.

const description = `Usage: node har2paths (file.har) > paths.txt

Prints URL paths of all server calls made while loading the first page
from the HAR file, which has been created for a SPA.`

const harName = process.argv[2]
if (!harName) {
  console.log(description)
  return process.exit(1)
}

const { readFile } = require('fs')
const { promisify } = require('util')

const readFileAsync = promisify(readFile)

function log (message) {
  process.stderr.write(message)
  process.stderr.write('\n')
}

async function main () {
  try {
    log(`Loading "${harName}"...`)
    const harString = await readFileAsync(harName, 'utf-8')
    log('Parsing the HAR content...')
    const { log:harContent } = JSON.parse(harString)
    const { title: pageURL, id: pageID } = harContent.pages[0]
    log(`Analysing the page ${pageURL}...`)
    const assetPaths = harContent.entries
      .filter(({ pageref }) => pageref === pageID)
      .map(({ request }) => new URL(request.url).pathname)
    log(`${assetPaths.length} static assets found.`)
    const printOutput = assetPaths.join('\n')
    console.log(printOutput)
  } catch (error) {
    console.error(error)
    process.exitCode = 1
  }
}

main()