[DEPRECATED] axe-puppeteer

This repository has been deprecated. The package has been moved to axe-core-npm. The package will be available via NPM as @axe-core/puppeteer.

---

Provides a chainable axe API for Puppeteer and automatically injects into all frames.

Getting Started

Install Node.js if you haven't already. For running axe-puppeteer tests read more about setting up your environment.

Install Puppeteer: npm install puppeteer --no-save

Install axe-puppeteer and its dependencies: npm install axe-puppeteer

Usage

This module uses a chainable API to assist in injecting, configuring and analyzing using axe with Puppeteer. As such, it is required to pass an instance of a Puppeteer Page or Frame.

Here is an example of a script that will drive Puppeteer to this repository, perform analysis and then log results to the console.

1const { AxePuppeteer } = require('axe-puppeteer')
2const puppeteer = require('puppeteer')
3
4;(async () => {
5  const browser = await puppeteer.launch()
6  const page = await browser.newPage()
7  await page.setBypassCSP(true)
8
9  await page.goto('https://dequeuniversity.com/demo/mars/')
10
11  const results = await new AxePuppeteer(page).analyze()
12  console.log(results)
13
14  await page.close()
15  await browser.close()
16})()

Note: Usage examples make use of ES2017 async/await. Use of await can only be done in a function declared async. If your project does not support async/await, you can just directly use the promise async functions return. Check here for more information.

Bypassing Content Security Policy

When trying to run axe, you might run into issues if the page you are checking has Content Security Policy enabled. To get around this, you must disable it through Page#setBypassCSP before navigating to the site.

loadPage(browser: Browser, url: string, { opts, source }: { opts?: any, source?: string } = {})

An alternate constructor is available which opens a page and performs the CSP bypass for you.

It closes the page after analyze is called.

1const { loadPage } = require('axe-puppeteer')
2const puppeteer = require('puppeteer')
3
4;(async () => {
5  const browser = await puppeteer.launch()
6  const axeBuilder = await loadPage(
7    browser,
8    'https://dequeuniversity.com/demo/mars/'
9  )
10  const results = await axeBuilder.analyze()
11  console.log(results)
12
13  await browser.close()
14})()

AxePuppeteer(page: Frame | Page[, axeSource: string])

Constructor for the AxePuppeteer helper. You must pass an instance of a Puppeteer Frame or Page as the first argument. Cannot be called without the new keyword.

1const builder = new AxePuppeteer(page)

If you wish to run a specific version of axe-core, you can pass the source axe-core source file in as a string. Doing so will mean axe-puppeteer runs this version of axe-core, instead of the one installed as a dependency of axe-puppeteer.

1const axeSource = fs.readFileSync('./axe-3.0.js', 'utf8')
2const builder = new AxePuppeteer(page, axeSource)

Note that you might need to bypass the Content Security Policy in some cases.

AxePuppeteer#include(selector: string | string[])

Adds a CSS selector to the list of elements to include in analysis

1new AxePuppeteer(page).include('.results-panel')

AxePuppeteer#exclude(selector: string | string[])

Add a CSS selector to the list of elements to exclude from analysis

1new AxePuppeteer(page)
2  .include('.results-panel')
3  .exclude('.results-panel h2')

AxePuppeteer#options(options: Axe.RunOptions)

Specifies options to be used by axe.run. Will override any other configured options, including calls to withRules and withTags. See axe-core API documentation for information on its structure.

1new AxePuppeteer(page).options({
2  checks: { 'valid-lang': ['orcish'] }
3})

AxePuppeteer#withRules(rules: string | string[])

Limits analysis to only those with the specified rule IDs. Accepts a String of a single rule ID or an Array of multiple rule IDs. Subsequent calls to AxePuppeteer#options, AxePuppeteer#withRules or AxePuppeteer#withRules will override specified options.

1new AxePuppeteer(page).withRules('html-lang')

1new AxePuppeteer(page).withRules(['html-lang', 'image-alt'])

AxePuppeteer#withTags(tags: string | string[])

Limits analysis to only those with the specified rule IDs. Accepts a String of a single tag or an Array of multiple tags. Subsequent calls to AxePuppeteer#options, AxePuppeteer#withRules or AxePuppeteer#withRules will override specified options.

1new AxePuppeteer(page).withTags('wcag2a')

1new AxePuppeteer(page).withTags(['wcag2a', 'wcag2aa'])

AxePuppeteer#disableRules(rules: string | string[])

Skips verification of the rules provided. Accepts a String of a single rule ID or an Array of multiple rule IDs. Subsequent calls to AxePuppeteer#options, AxePuppeteer#disableRules will override specified options.

1new AxePuppeteer(page).disableRules('color-contrast')

or use it combined with some specified tags:

1new AxePuppeteer(page)
2  .withTags(['wcag2a', 'wcag2aa'])
3  .disableRules('color-contrast')

AxePuppeteer#configure(config: Axe.Spec)

Inject an axe configuration object to modify the ruleset before running Analyze. Subsequent calls to this method will invalidate previous ones by calling axe.configure and replacing the config object. See axe-core API documentation for documentation on the object structure.

1const config = {
2  checks: [Object],
3  rules: [Object]
4}
5const results = await new AxePuppeteer(page)
6  .configure(config)
7  .analyze()
8console.log(results)

AxePuppeteer#analyze([callback: (Error | null[, Object]) => void])

Performs analysis and passes any encountered error and/or the result object to the provided callback function or promise function. Does not chain as the operation is asynchronous

Using the returned promise (optional):

1new AxePuppeteer(page)
2  .analyze()
3  .then(function(results) {
4    console.log(results)
5  })
6  .catch(err => {
7    // Handle error somehow
8  })

Using a callback function

1new AxePuppeteer(page).analyze(function(err, results) {
2  if (err) {
3    // Handle error somehow
4  }
5  console.log(results)
6})