I’ve noticed that a number of tests, include a “Featured Answer” that’s been edited, fail when I try them. As an Upcase member, what are your thoughts on seeing if the submitted answer’s specs are passing automatically?
Hi Andrew, I think this could be a nice addition, but really the focus of the system is on the code changes and work under the assumption that the tests are passing. Unfortunately, building out essentially a CI service to back the exercise system is likely far too complicated for us to be able to work on.
That said, if any of the featured solutions have failing tests, I’d love to know that. Feel free to reply here, or send notes to help@upcase.com.
Here’s the feature solution that fails for me (see link).
I 100% agree that the focus is on the code changes and that we should be under the assumption that the tests are passing. However, don’t the basis assumptions need to be tested as it’s good to peer-review code that’s passing rather than basing (which I unfortunately did) my initial review on a test that failed? Hence, it adds more code unfortunately but a pre-commit hook could run bundle exec rspec.