Is there an easy way to make the test fail if a new screenshot is generated?

[harpet06]

browser.saveAndDiffScreenshot('test');

say the above command creates a diff, is there a way to make that fail the test easily?

[blueimp]

I've just added a new commit that is included in the new version 2.3.0 of wdio-screen-commands and updated the dependency in this project, so now you can do the following:

const ssim = browser.saveAndDiffScreenshot('screenshot name')
if (ssim && ssim.All < 1) {
  // Screenshot differs from previous run
  // See also: https://github.com/blueimp/node-ffmpeg-image-diff
}

In my experience, failing tests for screenshot differences might give you too many false negatives though, as e.g. it's easy to have differences that are due to

• browser rendering changes
• os font rendering changes
• changes due to animation states
• changes outside of the browser viewport, e.g. different clock times in screenshots of mobile devices

[harpet06]

Thanks for the quick response, I'll try that out.

I've wrote a quick command to strip out more volatile bits of the pages, so hopefully we won't get too many false negatives

[blueimp]

Thanks, please let me know if we can close the issue (or feel free to close it yourself).