Skip to content

[Task] - Write tests that confirm failure scene output #46

@matthew-carroll

Description

@matthew-carroll

How we output failure scenes is as important as how we output original golden scenes. We need tests that lock down our failure output, too.

First, generate a number of failure scenarios. Here are some ideas:

  • A button that has different text
  • A button that changes color
  • A button with a slightly different shadow
  • Something that causes the new golden to have a physically larger size
  • Something that causes the new golden to have a physically smaller size

These failure goldens, which need to include two images each, should be placed in the test directory, e.g,. test_goldens/reference_images/failures/gets_bigger/

Then, write tests that run a comparison for each of the two images, produces a failure image, and then ensure that failure image matches the existing failure image that we have on file. In other words, write golden tests for the failure images from golden tests.

Metadata

Metadata

Labels

taskA work item that isn't a bug, feature, or infrastructure

Type

No type

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions