You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Currently, failure screenshots bypass the pixel-checking we do on regular screenshots for the sake of speed. In most cases, this is fine since most tests aren't going to have screenshots in them. However, in the case of a test having a cy.screenshot() command in it and failing, the failure screenshot can randomly not display the runner UI because the browser hasn't had time to paint it yet.
I think we need to re-introduce the pixel-checking on failure screenshots to make them 100% reliable. This, of course, will add some latency to them.
The text was updated successfully, but these errors were encountered:
It's possible to keep track of whether a screenshot is taken in a test and only do the pixel-checking on failures in that case only. So for the majority use-case, there will be no added latency.
Currently, failure screenshots bypass the pixel-checking we do on regular screenshots for the sake of speed. In most cases, this is fine since most tests aren't going to have screenshots in them. However, in the case of a test having a
cy.screenshot()
command in it and failing, the failure screenshot can randomly not display the runner UI because the browser hasn't had time to paint it yet.I think we need to re-introduce the pixel-checking on failure screenshots to make them 100% reliable. This, of course, will add some latency to them.
The text was updated successfully, but these errors were encountered: