@TheJaredWilcurt TheJaredWilcurt released this Oct 27, 2017 · 5 commits to 3.0 since this release

Assets 6

Happy Koalaween! Featuring our Koa11y koala mascot in a pumpkin mask

This release marks a major rewrite for the underlying structure of Koa11y, to make it much cleaner and easier to move forward with the big plans we have for Version 4. The best new feature of version 3 is the improved support for screenreaders.

Contributors to this release

Change log

  • Complete removal of UGUI dependency
  • Rewrote most of the Code to use Vue.js
  • Fixed line ending issue to make cross-platform developing easy for team members
  • Created modularized logic components
  • Most references to jQuery have been removed, rest will be gutted in v4
  • Build script has been created to handle automating builds for each platform.
  • Updated Dependencies: NW.js, Pa11y, fs-extra, semver
Assets 6

Koa11y Logo

I've been using this personally for about a year, but today marks the first public release of Koa11y; and with it, lots of polish to allow for easy usage by the general public.

Screenshot of Koa11y running on Windows

Major features:

  • Runs on Windows, Linux (32 and 64-bit), and OSX.
  • A simple UI to interact with the Pa11y command line application.
  • Generates reports in HTML, JSON, CSV, Markdown, and XML.
  • HTML reports are completely self contained and make only one network request for a webfont. All styles, scripts, and icons are embedded in the report, making the document easy to transfer/send/share.
  • HTML reports allow you to show/hide Errors/Warnings/Notices easily.
  • Supports WCAG2 A, WCAG2 AA, WCAG2 AAA, and Section 508 accessibility standards/guidelines.
  • Produces image statistics such as total file size, alt text length, failed-to-load, and allows for human interaction to verify if alt text is descriptive.

Missing Features

  • Image stats are exported with HTML, JSON, Markdown, and XML, however I couldn't find a way to structure the data to work with CSV. Will accept pull request.
  • When verifying if the images have descriptive alt text, the process feels slow and could be improved, allowing for faster confirmation of images. The current system is function but it's UX could be improved. A discussion should be had about how to proceed on this in the GitHub Issues page if we were to attempt to improve this.