Are simple-openai tests flaky?

[the-gigi]

We recently forked simple-openai due these deserialization issues:
sashirestela/cleverclient#77
#205

The only change is that instead of using Cleverclient's jsonToObject() we provide a local implementation with an ObjectMapper that can handle dates

We noticed that many of the tests fail intermittently. This happens mostly in our CI/CD system (Github Actions). I don't remember such intermittent failures when I was running the tests locally. In the interest of minimizing noise we disabled all the tests for now.

Is that a known phenomenon?

[sashirestela]

Hi @the-gigi. No, it is not. Tests should not fail.