Could you include some unit tests for this change?

Could you include some unit tests for this change?

0732136
I added a test for that change. If you need anything else, please let me know. Thanks. 😊

thanks, let me discuss this issue on the issue thread.

Hi @bluebill1049 I checked the your comment on the issue.
totally agree that skipping computation when not subscribed is important for performance.
What do you think about syncing isDirty internally based on dirtyFields instead?
It wouldn’t call _getDirty() or add any extra work, but would keep the internal state consistent so late subscribers still get the correct value without extra cost.

Hi @bluebill1049 I checked the your comment on the issue. totally agree that skipping computation when not subscribed is important for performance. What do you think about syncing isDirty internally based on dirtyFields instead? It wouldn’t call _getDirty() or add any extra work, but would keep the internal state consistent so late subscribers still get the correct value without extra cost.

sounds good 👍

Hi @bluebill1049 I checked the your comment on the issue. totally agree that skipping computation when not subscribed is important for performance. What do you think about syncing isDirty internally based on dirtyFields instead? It wouldn’t call _getDirty() or add any extra work, but would keep the internal state consistent so late subscribers still get the correct value without extra cost.

sounds good 👍

thanks for review

commit: e7d3fca

This Commit proposes syncing isDirty internally based on dirtyFields to keep form state consistent for late subscriptions.
It doesn’t call _getDirty() or introduce extra computation when isDirty isn’t subscribed — maintaining the existing performance behavior while ensuring correct state visibility for new subscribers.