That seems a little extreme to me, although I do see your point. I'm not convinced that we should disallow overrides of logical annotations.
The use case that comes to mind is the typical one of needing to modify behavior without having to recompile. Without overrides, you couldn't do it. Lastly, it makes overriding inconsistent between logical and mapping metadata. Mapping metadata would be overridable but logical metadata wouldn't -- makes for a higher learning curve to me. -matthew -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Sent: Wednesday, June 20, 2007 3:23 PM To: Apache JDO project; JDO Expert Group Subject: Annotation overrides Hi, I'm looking at how annotations can be overridden by the xml metadata files. It seems that the .jdo metadata that's not .orm metadata is an exact duplicate of the annotation metadata. Is there ever a case where you would want to override the annotation with jdo metadata? Seems that it's really only the orm metadata in the annotations that you want to override. We did carefully consider that the orm metadata is just mapping, and doesn't include information that would affect the object model. In this case, it seems that you would want to support these combinations: no annotations + .jdo (with jdo and orm metadata) no annotations + .jdo (with jdo and orm metadata) + .orm (overrides all orm in the jdo) annotations alone (includes jdo and orm metadata) annotations + .orm (overrides all orm in the annotations) So I'm proposing that we disallow annotations + jdo metadata file. In other words, if you make the decision to provide annotations, you can't override the jdo metadata. The annotations are closer to the Java file which is what the jdo metadata is for. Craig Craig Russell Architect, Sun Java Enterprise System http://java.sun.com/products/jdo 408 276-5638 mailto:[EMAIL PROTECTED] P.S. A good JDO? O, Gasp!
