Jason Frey
Jason Frey
Kind of a tangent, but one thing I do find strange is that when calling .subclasses after remove_const it presents the class as if the constant still exists (e.g. `[D]`)....
This is a cool idea. We had a similar problem recently with building ManageIQ where I accidentally committed my overrides to the lock. @bdunne What do you think?
@d-lebed Can you add some specs for these cases (particularly for the "not able to be resolved" case, which is pretty interesting)?
No problem 😊
@Whirlwind Yeah, I think `eval_gemfile` is more the way to go. bundler-inject is more about overriding things in existing Gemfiles, but it sounds like your request is more to completely...
No, this is a different issue from ensure_gem.
Though that being said, maybe ensure_gem would work for this case? If so, we should document that.
Ok, interesting - yeah then we should at least document it in the README, and then we can close this, or change it to a feature request.
@miq-bot cross-repo-test /all
Backported to `radjabov` in commit 606376ff47bbd39a7286c3fc374080b3e6c871d2. ```text commit 606376ff47bbd39a7286c3fc374080b3e6c871d2 Author: Jason Frey Date: Thu May 23 10:38:21 2024 -0400 Merge pull request #23042 from kbrock/service_reconfigure_request_type service reconfigure task needs a...