Small differences in scolling cause the tests to fail about cypress-visual-regression

[mjhea0]

Have you tried increasing the error threshold?

[dariuszteooh]

As you can see, the differences in scrolling make the diff quite significant, so increasing the threshold high enough to ignore that will just render the tests useless, as they won't pick up actual changes.

…

On Fri, 9 Jul 2021, 13:14 Michael Herman, ***@***.***> wrote: Have you tried increasing the error threshold? — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#88 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/APOY6MH4BPLYNJCH7MQUT6DTW3RZDANCNFSM5ACNMEKA> .

[mjhea0]

Why are you testing the full page? I would isolate it to specific elements. The larger the image, the greater the chance something like this is going to happen.

I'm going to close this since it's unrelated to the tool.

[adamdehaven]

This seems like a valid issue to me -- taking a screenshot of the entire page is a valid test and a fixed scrollbar shouldn't break the usage. I don't want to test specific elements on a page.

Any update or workaround for this?

[bmaupin]

If I'm not mistaken, this seems to be an issue with Cypress itself and how it takes screenshots: cypress-io/cypress#2681