question (blocking):

why is this expectation changed? I was expecting it to remain same.

Harsh Sagar (cfed4021) at 17 Mar 13:56

This feels like a verify specific dependency, I think I'll forget this next time.

@dstull - calling the last event assignment doesn't make sense, could you recommend a better name for it?

source ee/app/services/registrations/base_namespace_create_service.rb , ee/app/services/gitlab_subscriptions/trials/welcome_create_service.rb

@dstull - it looks like the matcher looks for unique event names, the second render in the above example is failing(I promise it was passing in our pairing session 😭 ). If you want, we can name them uniquely.

source https://gitlab.com/gitlab-org/gitlab/-/blob/master/spec/support/helpers/features/experiment_tracking_helper.rb#L81

Harsh Sagar (e6229bcf) at 17 Mar 11:57

Harsh Sagar (92354605) at 17 Mar 11:15

question (blocking):

why did we remove this test?

@cngo is that okay if I do it tomorrow?

Agree.

@dstull - can this be done as a follow up?

@dstull - I added the screenshots now.

@dstull - I don't think we need a DB review here, but let me know if you think otherwise.

I've updated the logic.

@kgreif 🏓

In this case, the requirement is specifically for free groups. If I introduce a limit here, it could interfere with the logic I’m using to check for the existence of non-free groups.

I could run a separate query to check whether the user has any paid groups, but comparing the counts seems cheaper. That said, I’m open to better suggestions if there’s a cleaner approach.

more context !225536 (comment 3138865722)

Harsh Sagar (63393734) at 13 Mar 10:08

... and 1 more commit