Hi Christian,

After further discussion I ultimately think that per the javadoc the method’s return value likely should not be a fuzzy hit for that specific method where you supply the primary-key and the revision number. What I think may work for the short-term would be to introduce a configuration option that forces the method to act in an exact match form without changing the API. This would allow your spring-data-envers solution to work as you desire and would maintain backward compatibility which is critical for us.

I’m currently working on some release tasks for another project today but once I get that completed I’ll try to jump on this and submit a PR for the next 5.4.x release. You should be able to snag the latest master & merge the fix locally if you want to test it out for me and let me know if you run into any problems.

I’ll update this issue once I have done that so you can test it if you want.