Fixes folders or resources with missing title remain 'incomplete' after editing just the title

[AllanOXDi]

Summary

This PR fixes folders or resources with missing title remain 'incomplete' after editing just the title
Closes #5347

References

#5347

Before

2025-09-01_17-24-52.mp4

After

Screen.Recording.2025-11-06.at.18.45.55.mov

Reviewer guidance

1. Sign in to Studio
2. Go to a channel and add a folder or a resource without a title
3. Edit just the title of the folder or resource and observe that it's still marked as incomplete

[akolson]

Hi @AllanOXDi! I left a general comment about the implementation. Thanks

[akolson]

Change looks correct to me! Manual QA also checks out. Thanks @AllanOXDi

[akolson]

@AllanOXDi can we re-target this to hotfixes? Thanks

[akolson]

@AllanOXDi, !t looks like the branch is from unstable so we might have to squash all the commits from unstable. I think cherypicking the commit instead should do the trick? or rebasing?

[AllanOXDi]

working on that

[AllanOXDi]

Fixed!

[akolson]

Left a blocking comment. All else looks correct to me!

[akolson]

All should be good now. Thanks @AllanOXDi

General comment on the tests updates: I wonder if making the tests more liberal using the objectContaining function could increase the risk of masking bugs as it only asserts that certain keys exist, not that the whole payload is exactly what is expected (which I think is the intention of the previous test). Not a blocker but something to think about

[AllanOXDi]

Totally fair point ! objectContaining is more permissive.
In this case though, that’s intentional: the test’s job is just to ensure the important fields (id, title, description ) are sent correctly, not to lock down the entire payload shape. Using objectContaining keeps the test focused on behavior rather than structure, and avoids breaking anytime we add non-essential fields. So it’s still a safe assertion for the contract we care about here?

[akolson]

I agree! However, the intent of the tests is lost--we've added the checkComplete to the payload in the updateContentNode action, so using objectContaining would bypass the added field. I think the test failures could have been resolved by just updating the tests to have the checkComplete: true,. I suspect the two tests are failing because they default checkComplete to undefined thus the difference in the expected payload.

expect(updateContentNode).toHaveBeenCalledWith({
  id: nodeId,
  title: newTitle,
  description: newDescription // or '',
  checkComplete: true,
});

[AllanOXDi]

Updated! I leaned toward using objectContaininghis because I see that this modal isn’t the one adding `checkComplete that happens inside the action logic.

[akolson]

Correct! It however makes use of it while calling the updateContentNode function.

Thanks @AllanOXDi for the fix. Will merge this