Login to google workspace and check if links in csv are accessible

check_google_urls.js

'use strict'

const fs = require('fs')
const puppeteer = require('puppeteer')
const parse = require('csv-parse')


const import_csv = file => {
  return new Promise((resolve, reject) => {
    const parser = parse({}, (err , data) => {
      if(err) return reject(err)

      const entries = []
      const header = data[0]

      for(let i = 1; i < data.length; ++i){
        let r = data[i]
        let object = {}
        for(var n = 0; n < r.length; ++n){
          object[header[n]] = r[n]
        }
        entries.push(object)
      }
      resolve(entries)
    })
    const input = fs.createReadStream(file)
    input.pipe(parser)
  })
}


(async () => {
  const browser = await puppeteer.launch({
    //args: ['--no-sandbox', '--disable-setuid-sandbox'],
    headless: false,
  })

  const page = await browser.newPage()

  await page.goto('https://drive.google.com/a/hcpss.org')
  await page.waitForSelector('#accountstatussubmit')
  await page.type('#username', '')
  await page.type('#password', '')
  await page.click('#accountstatussubmit')
  await page.waitForSelector('#drive_hist_frame')

  import_csv('unique_google_urls.csv').then(async rows => {
    for(const row of rows){
      const url = row.old_google_url
      await page.goto(url, { waitUntil: 'domcontentloaded' })
      let button = (await page.$('#request-access-button')) || '';
      if(button){
        console.log(url + ' bad')
      }else{
        const title = await page.title()
        if(title == 'Error 404 (Not Found)!!1'){
          console.log(url + ' missing')
        }else{
          console.log(url + ' good')
        }
      }
    }
  })
  await browser.close()
})()

Works with old version of csv-parse

{
  "dependencies": {
    "csv-parse": "^2.5.0",
    "puppeteer": "^12.0.1"
  }
}