db-jdo-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Craig L Russell <Craig.Russ...@Sun.COM>
Subject Re: Relationships mapped with mapped-by (long)
Date Sat, 09 Sep 2006 18:13:16 GMT
Hi David,

Thanks for your comments on this issue.

On Sep 9, 2006, at 1:34 AM, David Bullock wrote:

> On 9/6/06, Craig L Russell <Craig.Russell@sun.com> wrote:
> We have an issue with regard to the behavior of relationships  
> mapped using mapped-by, where there is a bidirectional relationship  
> instantiated by a single database artifact, either a foreign key or  
> a row in a join table.
> The spec currently defines a subset of behavior where relationships  
> are updated by the program, and the memory model must be updated to  
> reflect the change in the datastore after flush (including flush  
> for commit).
> You already know that as a user of JDO, I view this behaviour as  
> obnoxious - the JDO implementation should not alter the semantic  
> state of my objects, and it should by no means attempt to guess my  
> intent when I have made ambiguous alterations (an exception telling  
> me to get my act together is all I need).  Interesting that this  
> EJB3-envy feature is introducing new issues.

If you choose to completely manage both sides of the relationships  
via your own user code, then these changes will not affect you at all.

JDO has always had as the primary objective the transparent  
synchronization of memory and datastore models. I think that the only  
thing we're solving with this round of changes is more cases where we  
need to keep the memory model consistent with the datastore model.  
These cases arise when mapping two different parts of the memory  
model to the same datastore artifact.

> I would much prefer that you dropped it and tackled the issues of  
> relationship-management comprehensively in a separate JSR, which  
> used to be your entirely logical stance on the matter.  I regard  
> the need for maintenance in this area as informal evidence of an  
> iceberg waiting beneath the tip.

I still believe that a comprehensive relationship management JSR is  
useful. I just don't have the time necessary to manage it.


> sincerely,
> David Bullock.

Craig Russell
Architect, Sun Java Enterprise System http://java.sun.com/products/jdo
408 276-5638 mailto:Craig.Russell@sun.com
P.S. A good JDO? O, Gasp!

View raw message