test: refactor cloudevent_source_test.go

[maxcao13]

Refactors the cloudevent_source_test.go test to rely on events instead of sleeping in order to progress and also improves the cleanup between and after tests. We've noticed flakes with this test where the magic sleep timings don't play well on certain machines and platforms.

This commit also adds several more helper test functions that:

• Allow you to wait for Kubernetes events after calling a trigger function
• Allows you to execute commands on pods with TTY turned off (this change is due to several weird test flakes
  we've seen where the output of an exec command returned to the user is unexpected)

Checklist

• [-] When introducing a new scaler, I agree with the scaling governance policy
• I have verified that my change is according to the deprecations & breaking changes policy
• [-] Tests have been added
• [-] Changelog has been updated and is aligned with our changelog requirements
• [-] A PR is opened to update our Helm chart (repo) (if applicable, ie. when deployment manifests are modified)
• [-] A PR is opened to update the documentation on (repo) (if applicable)
• Commits are signed with Developer Certificate of Origin (DCO - learn more)

[wozniakjan]

/run-e2e internal
Update: You can check the progress here

[wozniakjan]

great effort! thank you so much for fixing these time.Sleeps with more deterministic behavior. I think it would benefit greatly from using gomega (no need to add ginkgo to the mix, I think that won't be necessary)

[maxcao13]

Thanks Jan! I've just force pushed some changes, lmk what you think when you have time. :-)

[JorTurFer]

This looks nice! just small nit inline

[maxcao13]

Thanks! Had to rebase to resolve the merge conflict.

[Copilot]

Pull Request Overview

This PR refactors the cloud event source test to improve its reliability by replacing fixed sleep timings with event watchers and by enhancing the test cleanup process. Key changes include:

• Replacing magic sleep durations with event-based triggers and consistency checks.
• Adding alternate execution functions for commands without TTY and updating logging messages.
• Improving cleanup by using t.Cleanup in the main test function.

Reviewed Changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated no comments.

File	Description
tests/internals/cloudevent_source/cloudevent_source_test.go	Refactored several test functions to use event watchers, deferred deletion, and improved logging to reduce flakiness
tests/helper/helper.go	Introduced ExecCommandOnSpecificPodWithoutTTY and implemented the KedaConsistently function to support event watching and polling

Comments suppressed due to low confidence (1)

tests/internals/cloudevent_source/cloudevent_source_test.go:623

• The log message in testErrEventSourceCreation mentions 'include filter' even though the function is focused on testing creation errors. Consider updating the log message to more accurately reflect its purpose (e.g., mention 'creation error') for better clarity.

t.Logf("--- test emitting eventsource about scaledobject err with include filter --- [isClusterScope: %t]", isClusterScope)

[rickbrouwer]

/run-e2e internal
Update: You can check the progress here

[wozniakjan]

lgtm, thank you!

I could see some minor code structure changes in the future to make the code a bit cleaner, but this fixes a real flaky test so I shouldn't dwell on nitpicks :)