gaogaotiantian commented on PR #53042: URL: https://github.com/apache/spark/pull/53042#issuecomment-3529538649
Those are all commit triggered CIs which could happen multiple times every day. Of course it makes sense to try to do less workflows on those. Coverage run only runs once per day and that provides our coverage number. If we don't care about a correct coverage number, why do coverage run at all? We can just disable that workflow for more resources. A wrong coverage number is just not useful. I think that might be the only normal python3.11 run? If that's the case, we are missing some tests on python3.11 + Linux + pandas for our daily CI. I just want to be crystal clear that I'm talking about this number: <img width="126" height="25" alt="image" src="https://github.com/user-attachments/assets/293c67c2-a23e-46a1-b52c-615a7662a682"; /> and the specific daily run workflow https://github.com/apache/spark/blob/master/.github/workflows/build_coverage.yml . Almost all the other daily workflows run pandas tests (https://github.com/apache/spark/blob/master/.github/workflows/build_python_3.12.yml for example), I don't know why this specific coverage run would be an exception. It's just so arbitrary. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
