Closed rambleraptor closed 9 years ago
@astephen2: The fact that no test failed with the change indicates that it wasn't covered by tests. It means I don't have confidence that this change is ok.
We need to know what happens when this scenario occurs: two runners check out the test then they both return results.
I'm currently building a test for this. The code makes it appear as if it will only remember the last result, unless the results differ in success (one success, one fail). In this case, it'll remember the last success and the last fail. Also, it'll record all failures towards the failure count. Which runs do you think should be counted?
Makes sense to me.
--phone is hard. On Jul 18, 2014 2:00 PM, "Alex Stephen" notifications@github.com wrote:
I'm currently building a test for this. The code makes it appear as if it will only take the result of the final run. I personally think that it should only matter if the two results differ. In this case it will record it as a fail, regardless of which run (first, second, nth) is successful/failed. Thoughts?
— Reply to this email directly or view it on GitHub https://github.com/Yelp/Testify/pull/273#issuecomment-49479165.
ping: 18 days
This PR still needs a test. As @astephen2 doesn't appear to have the available time for this, anyone can takeover this PR.
We will be able to delete the client/server stuff entirely in the near term I think. I'm declaring this branch officially dead.
270
Deleted the runner check. Test suite still works.