Skip to Content

Contributors

Re: apologies for mass PR creation

Thanks for the apologies. I think this kind of massive changes that has nothing to do with the community and that involves a lot of repositories, should be planned with the community (like now you are doing with this message).

Following that, anything done massively should have a pilot test to check that everything is OK instead of blindly performing 100 PRs. The commit message was not following the conventions, and that meant to double the number of pipelines to execute. I know this is not something that costs us directly (apart from a delay in all the real contributions checks), but we should also see for the environment not wasting CPU resources.

Well, I don't want to dig more into something that has no remedy. Just for future actions. Now, a less impacting way to do this is as some proposals:

- To be changed only for newer branches or when an update is done, like ForgeFlow did to pass from Eficent.
- Do it progressively.
- Do it directly for not having 3 checks: the PR one + the ocabot one + the merge one.

And I add another one: you can put `[ci skip]` at the end of the commit message for avoiding the check at all, but this only serves if you do directly the write or merge the PR without ocabot.

Regards.

by Pedro M. Baeza - 01:51 - 14 Jun 2024

Reference

  • apologies for mass PR creation
    Hello everyone,
    
    I asked yesterday Vincent to fix the name of Camptocamp in the manifest 
    of the OCA addons, without realizing the impact this would have on the CI.
    
    We are sorry for the disruption caused, and I should have checked before 
    acting about the proper way to do this. Please accept our humble apologies.
    
    Now if you have a recommendation on how to deal with the update in the 
    proper way, we will be happy to get your feedback.
    
    Kind regards,
    
    
    -- 
    Alexandre Fayolle
    

    by Alexandre Fayolle - 12:21 - 14 Jun 2024