Fix regression test - #71
Conversation
|
On a cursory look, the change appears ok. However, I think the reason why this wasn't noticed by me when I proposed the addition of the tests to begin with is because the outputs may differ depending on the OS, and my tests might not have covered the same as yours. If so, applying it may cause tests to fail elsewhere. It is important that we build and run the tests across all our supported platforms, including macOS and Windows, and check where they succeed and fail. Because if they all fail with regression.diffs similar to this PR proposal, then the PR may be applied as is. But if some succeed, then we should change the proposal so that instead of it changing @neel5481 , @zainedb , @agarwal-kritika I see you are all involved with the development of this, including the workflows that build and test it. We must make sure that the workflows that build all our packages also test them, and only upload to cloudsmith if the tests pass. Whatever tests fail will be input to deciding what to do with this PR. |
This small patch fixes the only failing regression test. Not sure if it is the right fix or not though.
Per: pgdg-packaging/pgdg-rpms#213