For API 2.0, we'd completely remove the magic of Hibernate outside of the API – e.g., return detached objects – sacrificing lazy loading for predictable, intuitive, and reliable behavior.
As it stands, I'm favoring that we (1) acknowledge how things work & strive to making the current behavior understandable/predictable instead of (2) trying to make it behave how we'd like with a caveat that "it might not work some of the time." -Burke 2012/3/9 Mark Goodrich <[email protected]> > It just seems like the way our API is written and is used, the assumption > is that changes made to a domain object are only persisted if a > saveObject() service method is called. But by default Hibernate persists > changes to any attached objects automatically, and we are struggling > without complete success to stop "premature" flushes, which make the prior > assumption false. I haven't read the API 2.0 notes, but my suggestion > would be in 2.0 we give up on that assumption instead of trying to enforce > it. > > Mark > _________________________________________ To unsubscribe from OpenMRS Developers' mailing list, send an e-mail to [email protected] with "SIGNOFF openmrs-devel-l" in the body (not the subject) of your e-mail. [mailto:[email protected]?body=SIGNOFF%20openmrs-devel-l]
