[ http://opensource.atlassian.com/projects/hibernate/browse/HHH-2439?page=c... ] Christian Bauer commented on HHH-2439: -------------------------------------- This is risky, it might break existing applications. If applications rely on a non-null identifier value being available on an entity instance if its in persistent state, this guarantee no longer holds if they run with FlushMode.MANUAL. Limiting this behavior to the new persist() method _and_ requiring auto-commit mode is a good way of restricting exposure of this behavior to new applications, and to let only users who understand the behavior actually utilize it (we hope). -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: http://opensource.atlassian.com/projects/hibernate/secure/Administrators.... - For more information on JIRA, see: http://www.atlassian.com/software/jira

[ http://opensource.atlassian.com/projects/hibernate/browse/HHH-2439?page=c... ] Christian Bauer commented on HHH-2439: -------------------------------------- Sure it would be nice to have it in FlushMode.MANUAL as well, without auto-commit. I'm only worried about breaking stuff. About the spec: It's not in there. The specification is very clear: persist() brings an entity instance into managed state, an entity instance has an assigned identifier value if it is in managed state. This is a guarantee we are already breaking with persist() and auto-commit. The spec does not mention anything about "assigned at flush-time". I opened a JIRA case a year ago about it and was told by Gavin that this is fine within the spec. But really, it's not there and I carefully avoided saying that in he book. If someone found a section in the spec that allows it, I need the number. -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: http://opensource.atlassian.com/projects/hibernate/secure/Administrators.... - For more information on JIRA, see: http://www.atlassian.com/software/jira

[ http://opensource.atlassian.com/projects/hibernate/browse/HHH-2439?page=c... ] Christian Bauer commented on HHH-2439: -------------------------------------- 3.2, second item: "A managed entity instance is an instance with a persistent identity that is currently associated with a persistence context." Then, 3.2.1 states that an entity instance becomes managed when persist() is invoked. Of course you can interpret "with a persistent identity" as some obscure fact that is half-defned elsewhere (e.g. how it is used to define merging: "with the same persistent identity"), but most people would say "that means that the identifier value is not null". -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: http://opensource.atlassian.com/projects/hibernate/secure/Administrators.... - For more information on JIRA, see: http://www.atlassian.com/software/jira

[ http://opensource.atlassian.com/projects/hibernate/browse/HHH-2439?page=c... ] Christian Bauer commented on HHH-2439: -------------------------------------- I'm not saying it's OK. I'm saying the spec is not clear. So we can continue ignoring that and extend FlushMode.MANUAL behavior or we can try to fix or at least document it. If "instance with a persistent identity" does not mean the obvious thing, then document what it means. -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: http://opensource.atlassian.com/projects/hibernate/secure/Administrators.... - For more information on JIRA, see: http://www.atlassian.com/software/jira

[ http://opensource.atlassian.com/projects/hibernate/browse/HHH-2439?page=c... ] Steve Ebersole updated HHH-2439: -------------------------------- Fix Version/s: (was: 3.2.4) 3.3 -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: http://opensource.atlassian.com/projects/hibernate/secure/Administrators.... - For more information on JIRA, see: http://www.atlassian.com/software/jira

[ http://opensource.atlassian.com/projects/hibernate/browse/HHH-2439?page=c... ] Paul Benedict commented on HHH-2439: ------------------------------------ I hope I didn't miss the finer points, but to what persistence strategy is this issue targeting? Will the delay be applied to both save() and persist()? If it is persist(), then I think it is OKAY, but if the latter then here's my case against it.... The main difference between save() and persist() and is that save() immediately gets an identifier. See HHH-1682 and http://www.hibernate.org/hib_docs/v3/reference/en/html/objectstate.html#o..., which clearly detail the difference: "If Cat has a generated identifier, the identifier is generated and assigned to the cat when save() is called. " -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: http://opensource.atlassian.com/projects/hibernate/secure/Administrators.... - For more information on JIRA, see: http://www.atlassian.com/software/jira

[ http://opensource.atlassian.com/projects/hibernate/browse/HHH-2439?page=c... ] Steve Ebersole updated HHH-2439: -------------------------------- Fix Version/s: (was: 3.3.x) 3.4 So lets not do this in a point release. -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: http://opensource.atlassian.com/projects/hibernate/secure/Administrators.... - For more information on JIRA, see: http://www.atlassian.com/software/jira

[ http://opensource.atlassian.com/projects/hibernate/browse/HHH-2439?page=c... ] Steve Ebersole updated HHH-2439: -------------------------------- Fix Version/s: (was: 3.5.0.Beta-1) 3.5-tmp -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: http://opensource.atlassian.com/projects/hibernate/secure/Administrators.... - For more information on JIRA, see: http://www.atlassian.com/software/jira

[ http://opensource.atlassian.com/projects/hibernate/browse/HHH-2439?page=c... ] Steve Ebersole updated HHH-2439: -------------------------------- Fix Version/s: (was: 3.5.1) 3.5.x -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: http://opensource.atlassian.com/projects/hibernate/secure/Administrators.... - For more information on JIRA, see: http://www.atlassian.com/software/jira

[ http://opensource.atlassian.com/projects/hibernate/browse/HHH-2439?page=c... ] Gail Badner updated HHH-2439: ----------------------------- Fix Version/s: (was: 3.5.4) 3.5.x -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: http://opensource.atlassian.com/projects/hibernate/secure/Administrators.... - For more information on JIRA, see: http://www.atlassian.com/software/jira