Repository navigation
Conversation
|
I don't think your logic is sound here. If it was sound then we wouldn't do any tests on merge-to-master. We test on merge-to-master in case there are interactions with whatever is on master already. Particularly for very old PRs this can become a big problem. With the workflow filtering we are now doing the first we might know if it is a user report, which is no good. |
|
I would agree with you but not for for the env test. even if we merge the env change on master, nothing garantee it isn't broken 1s after it is merged on master as those depends on external depencies. And we got the weekly check for this to get inputs on this. I removed this from merge as this workflow is really slow so it shouldn't hold up the CI for low value I can put as ruling that this workflow should be always rebase against files on master so we got the true garantee the PR check is correct. Would that be ok with you ? |
Summary
We shouldn't redo the tests on merge on master branch as they already passed on PR
Restrict the PR to those that target master, so it discard release branch
Classification & Testing (check all that apply and add your own)
Description