Purple A11y

Purple A11y is a customisable, automated accessibility testing tool that allows software development teams to assess whether their products are user-friendly to persons with disabilities (PWDs).

This is the engine and command-line interface (CLI) for Purple A11y. For a desktop graphical user interface (GUI), check out Purple A11y Desktop. The official application can only be downloaded at https://go.gov.sg/purple-a11y-cicd. We recommend that you download the software only from the official link, as other sources and/or third party links may pose risks and/or compromise your system.

Technology Stack

1. Crawlee
2. Axe-core
3. Node.js
4. Playwright
5. Pixelmatch
6. Corretto
7. VeraPDF

Using Purple A11y as a NodeJS module

If you wish to use Purple A11y as a NodeJS module that can be integrated with end-to-end testing frameworks, refer to the integration guide on how you can do so.

Prerequisites and Installations

Portable Purple A11y

Portable Purple A11y is the recommended way to run Purple A11y as it reduces the difficulty for installation. Refer to Installation Guide for step-by-step instructions for portable Purple A11y.

Manual Installation

Please ensure the following requirements are met:

Node.js

• A Node distribution of 20 (LTS) or above.
• To check your version of Node, go into terminal and paste the command bellow

node -v

• If you do not have node, or if there is a need to manage your node versions, you can consider using Node Version Manager (NVM). After NVM is installed, you can then install an LTS version of Node:

# Install NodeJS version with NVM
nvm install --lts

# For subsequent use, you will need to run the command below each time you open a new terminal
nvm use --lts

• Install the required NPM packages with npm install.
• Build the project with npm run build before you try to run it with npm start.

Java JRE/JDK

• A JRE/JDK distribution of version 11 or above.
• To check your version of Java, go into terminal and paste the command bellow

java --version

• If you do not have java, you can consider installing Corretto distribution of OpenJDK.

VeraPDF

• VeraPDF is used for scanning PDF files. Install VeraPDF by following the install guide. You may wish to use the Automated installation script provided on that page, and changing the XML installation script section <installpath>/tmp/verapdf-test</installpath> to a location of your choice.
• Ensure you have VeraPDF set in PATH then verify verapdf is installed correctly:

# Invoke the VeraPDF installer
# Windows users should use vera-install.bat instead of vera-install
./verapdf-install ./auto-install.xml

# Add VeraPDF to PATH. For subsequent use, you will need to run the command below each time you open a new terminal
export PATH="<location of verapdf>:$PATH"

# Verify verapdf is installed
verapdf --version

Facing issues?

Please refer to Troubleshooting section for more information.

Features

Purple A11y can perform the following to scan the target URL.

• To run Purple A11y in terminal, run npm start. Questions will be prompted to assist you in providing the right inputs.
• Results will be compiled in JSON format, followed by generating a HTML report.

NOTE: For your initial scan, there may be some loading time required before use. Purple-A11y will also ask for your name and email address and collect your app usage data to personalise your experience. Your information fully complies with GovTech’s Privacy Policy.

Delete/Edit Details

You may delete and edit your cached name and e-mail address by running the following command to delete userData.txt:

• Windows (PowerShell): rm "$env:APPDATA\Purple A11y\userData.txt"
• MacOS (Terminal): rm "$HOME/Library/Application Support/Purple A11y/userData.txt"

If userData.txt does not exists just run npm start.

Scan Selection

You can interact via your arrow keys.

% npm start
┌────────────────────────────────────────────────────────────┐
│  Purple A11y (ver      )                                   │
│  We recommend using Chrome browser for the best experience.│
│                                                            │
│ Welcome back User!                                       │
│ (Refer to readme.txt on how to change your profile)        │
└────────────────────────────────────────────────────────────┘
? What would you like to scan? (Use arrow keys)
❯ Sitemap
  Website
  Custom

Headless Mode

Headless mode would allow you to run the scan in the background. If you would like to observe the scraping process, please enter n

 % npm start
┌────────────────────────────────────────────────────────────┐
│ Purple A11y (ver      )                                    │
│ We recommend using Chrome browser for the best experience. │
│                                                            │
│ Welcome back User!                                         │
│ (Refer to readme.txt on how to change your profile)        │
└────────────────────────────────────────────────────────────┘
? What would you like to scan? Sitemap
? Do you want purple-a11y to run in the background? (Y/n) No

Sitemap Scan

% npm start
┌────────────────────────────────────────────────────────────┐
│ Purple A11y (ver      )                                     │
│ We recommend using Chrome browser for the best experience. │
│                                                            │
│ Welcome back User!                                         │
│ (Refer to readme.txt on how to change your profile)        │
└────────────────────────────────────────────────────────────┘
? What would you like to scan? Sitemap
? Do you want purple-a11y to run in the background? No
? Which screen size would you like to scan? (Use arrow keys) Desktop
? Please enter URL or file path to sitemap, or drag and drop a sitemap file here:  https://www.sitemaps.org/sitemap.xml


 Scanning website...


 Fetching URLs. This might take some time...


Scanning website...

#purple-a11y will then start scraping from the file link provided above.
#Console results

If the sitemap URL provided is invalid, an error message will be prompted for you to provide a valid input.

>> Invalid sitemap format. Please provide a URL with a valid sitemap.

Website Scan

% npm start
┌────────────────────────────────────────────────────────────┐
│ Purple A11y (ver      )                                    │
│ We recommend using Chrome browser for the best experience. │
│                                                            │
│ Welcome back User!                                         │
│ (Refer to readme.txt on how to change your profile)        │
└────────────────────────────────────────────────────────────┘
? What would you like to scan? Website
? Do you want purple-a11y to run in the background? Yes
? Which screen size would you like to scan? (Use arrow keys) Desktop
? Please enter URL of website:  https://www.domain.org

If the website URL provided is invalid, an error message will be prompted for you to provide a valid input.

>> Cannot resolve URL. Please provide a valid URL.

Customised Mobile Device Scan

% npm start
┌────────────────────────────────────────────────────────────┐
│ Purple A11y (ver      )                                   │
│ We recommend using Chrome browser for the best experience. │
│                                                            │
│ Welcome back User!                                         │
│ (Refer to readme.txt on how to change your profile)        │
└────────────────────────────────────────────────────────────┘
? What would you like to scan? Website
? Do you want purple-a11y to run in the background? No
? Which screen size would you like to scan? (Use arrow keys) (Use arrow keys)
❯ Desktop 
  Mobile
  Custom

Choose Mobile for a default mobile screen size scan and Custom to choose a device or specify viewport width options.

Custom flow

Custom flow allows you to specify a user journey by enabling you to click the scan button on each desired webpage on a browser to initiate scan. This is useful to scan websites that require user and form input.

1. Start by choosing the Custom flow in the menu selection.

% npm start
┌────────────────────────────────────────────────────────────┐
│ Purple A11y (ver      )                                   │
│ We recommend using Chrome browser for the best experience. │
│                                                            │
│ Welcome back User!                                         │
│ (Refer to readme.txt on how to change your profile)        │
└────────────────────────────────────────────────────────────┘
? What would you like to scan?
  Sitemap
  Website
❯ Custom

1. Specify the URL of the starting page you wish to scan
2. A Chrome window will appear. Navigate through the pages and click on the Scan this page button at the top right of the broswer's page to conduct an accessibility scan on the current page.
3. You may drag the top black bar to the bottom of the page in the event it prevents you from viewing / blocking certain page elements.
4. Close the Chrome window to end the scan.

Other options:

• You can specify sites to exclude from accessibility scan (e.g. login page) by adding a pattern of the domain to exclusions.txt. An example of exclusions.txt:

\.*login.singpass.gov.sg\.*

Known Issues

If the custom flow fails to start, you might be running multiple versions of Playwright. Re-install Playwright:

1. On Windows, delete the folder %USERPROFILE%\AppData\Local\ms-playwright where %USERPROFILE% is typically located at C:\Users\<username>.
2. On MacOS, delete the folder ~/Library/Caches/ms-playwright where ~ refers to /Users/<username>.
3. Within PowerShell (Windows) or Terminal (MacOS) app, run the following two commands to re-install Playwright:

npx [email protected] install

CLI Mode

CLI mode is designed to be run in continuous integration (CI) environment. Run npm run cli for a set of command-line parameters available.

Usage: npm run cli -- -c <crawler> -d <device> -w <view
port> -u <url> OPTIONS

Options:
      --help                         Show help                         [boolean]
  -c, --scanner                      Type of scan, 1) sitemap, 2) website crawl,
                                      3) custom flow, 4) intelligent
                                     elligent
  [required] [choices: "sitemap", "website", "custom", "intelligent"]
  -u, --url                          Website URL you want to scan
                                                             [string] [required]
  -d, --customDevice                 Device you want to scan            [string]
  -w, --viewportWidth                Viewport width (in pixels) you want to scan
                                                                        [number]
  -o, --zip                          Zip filename to save results       [string]
  -p, --maxpages                     Maximum number of pages to scan (default: 1
                                     00). Only available in website and sitemap
                                     scans                              [number]
  -f, --safeMode                     Disable dynamically clicking of page button
                                     s and links to find links, which resolve is
                                     sues on some websites. [yes / no]
                                                        [string] [default: "no"]
  -h, --headless                     Run the scan in headless mode. [yes / no]
                                                       [string] [default: "yes"]
  -b, --browserToRun                 Browser to run the scan on: 1) Chromium, 2)
                                      Chrome, 3) Edge. Defaults to Chromium.
                     [choices: "chromium", "chrome", "edge"] [default: "chrome"]
  -s, --strategy                     Crawls up to general (same parent) domains,
                                      or only specific hostname. Defaults to "sa
                                     me-domain".
                                       [choices: "same-domain", "same-hostname"]
  -e, --exportDirectory              Preferred directory to store scan results.
                                     Path is relative to your home directory.
                                                                        [string]
  -j, --customFlowLabel              Give Custom Flow Scan a label for easier re
                                     ference in the report              [string]
  -k, --nameEmail                    To personalise your experience, we will be
                                     collecting your name, email address and app
                                      usage data. Your information fully complie
                                     s with GovTech’s Privacy Policy. Please pro
                                     vide your name and email address in this fo
                                     rmat "John Doe:[email protected]".
                                                             [string] [required]
  -t, --specifiedMaxConcurrency      Maximum number of pages to scan concurrentl
                                     y. Use for sites with throttling. Defaults
                                     to 25.                             [number]
  -i, --fileTypes                    File types to include in the scan. Defaults
                                      to html-only.
       [string] [choices: "all", "pdf-only", "html-only"] [default: "html-only"]
  -x, --blacklistedPatternsFilename  Txt file that has a list of pattern of doma
                                     ins to exclude from accessibility scan sepa
                                     rated by new line
                                            [string] [default: "exclusions.txt"]
  -a, --additional                   Additional features to include in the repor
                                     t:
                                     screenshots - Include element screensho
                                     ts in the generated report
                                     none - Exclude
                                     all additional features in the generated re
                                     port
              [string] [choices: "screenshots", "none"] [default: "screenshots"]
  -q, --metadata                     Json string that contains additional scan m
                                     etadata for telemetry purposes. Defaults to
                                      "{}"              [string] [default: "{}"]
  -r, --followRobots                 Crawler adheres to robots.txt rules if it e
                                     xists. [yes / no]  [string] [default: "no"]
  -m, --header                       The HTTP authentication header keys and the
                                     ir respective values to enable crawler acce
                                     ss to restricted resources.        [string]

Examples:
  To scan sitemap of website:', 'npm run cli -- -c [ 1 | sitemap ] -u <url_lin
  k> [ -d <device> | -w <viewport_width> ]
  To scan a website', 'npm run cli -- -c [ 2 | website ] -u <url_link> [ -d <d
  evice> | -w <viewport_width> ]
  To start a custom flow scan', 'npm run cli -- -c [ 3 | custom ] -u <url_link
  > [ -d <device> | -w <viewport_width> ]

Device Options

• "Desktop" (defaults to a 1280x720 viewport)
• "Mobile" (defaults to iPhone 11 viewport)
• "Desktop Chrome HiDPI"
• "Desktop Edge HiDPI"
• "Desktop Firefox HiDPI"
• "Desktop Safari"
• "Desktop Chrome"
• "Desktop Edge"
• "Desktop Firefox"
• "Blackberry PlayBook"
• "Blackberry PlayBook landscape"
• "BlackBerry Z30"
• "BlackBerry Z30 landscape"
• "Galaxy Note 3"
• "Galaxy Note 3 landscape"
• "Galaxy Note II"
• "Galaxy Note II landscape"
• "Galaxy S III"
• "Galaxy S III landscape"
• "Galaxy S5"
• "Galaxy S5 landscape"
• "Galaxy S8"
• "Galaxy S8 landscape"
• "Galaxy S9+"
• "Galaxy S9+ landscape"
• "Galaxy Tab S4"
• "Galaxy Tab S4 landscape"
• "iPad (gen 6)"
• "iPad (gen 6) landscape"
• "iPad (gen 7)"
• "iPad (gen 7) landscape"
• "iPad Mini"
• "iPad Mini landscape"
• "iPad Pro 11"
• "iPad Pro 11 landscape"
• "iPhone 6"
• "iPhone 6 landscape"
• "iPhone 6 Plus"
• "iPhone 6 Plus landscape"
• "iPhone 7"
• "iPhone 7 landscape"
• "iPhone 7 Plus"
• "iPhone 7 Plus landscape"
• "iPhone 8"
• "iPhone 8 landscape"
• "iPhone 8 Plus"
• "iPhone 8 Plus landscape"
• "iPhone SE"
• "iPhone SE landscape"
• "iPhone X"
• "iPhone X landscape"
• "iPhone XR"
• "iPhone XR landscape"
• "iPhone 11"
• "iPhone 11 landscape"
• "iPhone 11 Pro"
• "iPhone 11 Pro landscape"
• "iPhone 11 Pro Max"
• "iPhone 11 Pro Max landscape"
• "iPhone 12"
• "iPhone 12 landscape"
• "iPhone 12 Pro"
• "iPhone 12 Pro landscape"
• "iPhone 12 Pro Max"
• "iPhone 12 Pro Max landscape"
• "iPhone 12 Mini"
• "iPhone 12 Mini landscape"
• "iPhone 13"
• "iPhone 13 landscape"
• "iPhone 13 Pro"
• "iPhone 13 Pro landscape"
• "iPhone 13 Pro Max"
• "iPhone 13 Pro Max landscape"
• "iPhone 13 Mini"
• "iPhone 13 Mini landscape"
• "Kindle Fire HDX"
• "Kindle Fire HDX landscape"
• "LG Optimus L70"
• "LG Optimus L70 landscape"
• "Microsoft Lumia 550"
• "Microsoft Lumia 550 landscape"
• "Microsoft Lumia 950"
• "Microsoft Lumia 950 landscape"
• "Nexus 10"
• "Nexus 10 landscape"
• "Nexus 4"
• "Nexus 4 landscape"
• "Nexus 5"
• "Nexus 5 landscape"
• "Nexus 5X"
• "Nexus 5X landscape"
• "Nexus 6"
• "Nexus 6 landscape"
• "Nexus 6P"
• "Nexus 6P landscape"
• "Nexus 7"
• "Nexus 7 landscape"
• "Nokia Lumia 520"
• "Nokia Lumia 520 landscape"
• "Nokia N9"
• "Nokia N9 landscape"
• "Pixel 2"
• "Pixel 2 landscape"
• "Pixel 2 XL"
• "Pixel 2 XL landscape"
• "Pixel 3"
• "Pixel 3 landscape"
• "Pixel 4"
• "Pixel 4 landscape"
• "Pixel 4a (5G)"
• "Pixel 4a (5G) landscape"
• "Pixel 5"
• "Pixel 5 landscape"
• "Moto G4"
• "Moto G4 landscape"

If the device name contains ( and ), wrap the device name in single quotes when entered into the CLI. Please note that -d and -w are mutually exclusive. If none are specified, the default device used for the CLI scan is Desktop.

For example, to conduct a website scan to the URL "http://localhost:8000" and write to "a11y-scan-results.zip" with an 'iPad (gen 7) landscape' screen, run

npm run cli -- -c 2 -o a11y-scan-results.zip -u http://localhost:8000 -d 'iPad (gen 7) landscape'

If the site you want to scan has a query string wrap the link in single quotes when entered into the CLI.

For example, to conduct a website scan to the URL "http://localhost:8000" and write to "a11y-scan-results.zip" with a custom screen width '360', run

npm run cli -- -c 2 -o a11y-scan-results.zip -u "http://localhost:8000" -w 360

Report

Once a scan of the site is completed.

A report will be downloaded into the current working directory.

Accessibility Scan Results

Each Issue has its own severity "Must Fix" / "Good to Fix" based on the WCAG 2.2 Conformance.

For details on which accessibility scan results triggers a "Must Fix" / "Good to Fix" findings, you may refer to Scan Issue Details.

Troubleshooting

Please refer to the information below to assist in debugging. Most errors below are due to the switching between Node.js versions.

Incompatible Node.js versions

Issue: When your Node.js version is incompatible, you may face the following syntax error.

const URL_NO_COMMAS_REGEX = RegExp('https?://(www\\.)?[\\p{L}0-9][-\\p{L}0-9@:%._\\+~#=]{0,254}[\\p{L}0-9]\\.[a-z]{2,63}(:\\d{1,5})?(/[-\\p{L}0-9@:%_\\+.~#?&//=\\(\\)]*)?', 'giu'); // eslint-disable-line
                            ^
SyntaxError: Invalid regular expression: /https?://(www\.)?[\p{L}0-9][-\p{L}0-9@:%\._\+~#=]{0,254}[\p{L}0-9]\.[a-z]{2,63}(:\d{1,5})?(/[-\p{L}0-9@:%_\+.~#?&//=\(\)]*)?/: Invalid escape

Solution: Install Node.js versions >= 20, i.e. Node.js v20 and above.

Compiled against a different Node.js version

Issue: When you switch between different versions of Node.js in your environment, you may face the following error.

<user_path>/purple-a11y/node_modules/bindings/bindings.js:91
        throw e
        ^

Error: The module '<module_file_path>'
was compiled against a different Node.js version using
NODE_MODULE_VERSION 57. This version of Node.js requires
NODE_MODULE_VERSION 88. Please try re-compiling or re-installing
the module (for instance, using `npm rebuild` or `npm install`).

Solution: As recommended in the error message, run npm rebuild or npm install

dyld Error

Issue: Not able to run Purple A11y due to the following error shown below

dyld: lazy symbol binding failed: Symbol not found: __ZN2v87Isolate37AdjustAmountOfExternalAllocatedMemoryEx
  Referenced from: <user_path>/purple-a11y/node_modules/libxmljs/build/Release/xmljs.node
  Expected in: flat namespace

dyld: Symbol not found: __ZN2v87Isolate37AdjustAmountOfExternalAllocatedMemoryEx
  Referenced from: <user_path>/PURPLE_A11y/purple-a11y/node_modules/libxmljs/build/Release/xmljs.node
  Expected in: flat namespace

zsh: abort      node index.js

Solutions:

1. Delete existing node_modules folder and re-install the NPM packages with npm install.
2. Refer to this GitHub issue for more alternative solutions

Element Screenshot Limitation

Limitation: Due to animations causing elements to shift out of the viewport after an Axe scan, there's a risk of element screenshots timing out within 5 seconds if the element is not found. This known issue is particularly prevalent in scenarios like carousels with interval-based transitions.

FAQ

How do I limit number of pages scanned?

If you find a scan takes too long to complete due to large website, or there are too many pages in a sitemap to scan, you may choose to limit number of pages scanned.

To do this, run CLI mode npm run cli -- with the needed settings and specify -p 10 where 10 is the number of pages you wish to scan.

I am a new developer and I have some knowledge gap.

We recommend looking at our Technology Stack to understand the usage of each component. Take your time to understand.

Additional Information on Data

Purple A11y uses third-party open-source tools that may be downloaded over the Internet during the installation process of Purple A11y. Users should be aware of the libraries used by examining package.json.

Purple A11y may send information to the website, URL and HTML code snippets where the user chooses to initiate a Purple A11y scan for telemetry purposes.