> -Original Message-
> From: Patrik Nordwall [mailto:patrik.nordw...@gmail.com]
>
> Intersting, any idea of how we should solve it?
> Do you know if it is the same when using @javax.persistence.Lob ?
Yes, the annotation does nothing to the merge process.
> Feels like a stupid limitatio
After some investigation I found bug HHH-2680
http://opensource.atlassian.com/projects/hibernate/browse/HHH-2680
Here is patch from 2008 for this issue and also technique how to avoid
it. Instead of:
session.merge(domain); // changes blob value to 44
Changing it to this avoids the problem:
D
Intersting, any idea of how we should solve it?
Do you know if it is the same when using @javax.persistence.Lob ?
Feels like a stupid limitation in Hibernate. Is it the same in EclipseLink?
Maybe we could check if the id is not assigned and then use persist instead
of merge.
You can of course do
Hi,
as I discovered, the Blob will not be saved for new entities in Hibernate
because the save operation uses merge, which leaves the Blob behind, as
org.hibernate.type.BlobType says: "//Blobs are ignored by merge()".
It seems to be in the EJB3 spec, as a forum entry hints:
>>As per the EJB3 s
