IMO, mailing list is probably good for discussing strategies and
techniques. JIRA reports are good for providing implementation
summaries and proposed patches.
On May 16, 2007, at 4:58 PM, Gokhan Ergul wrote:
Marc, Patrick,
What would be the correct medium to discuss various design/
implementation details? This mailing list / jira or somewhere else?
Gokhan.
Marc Prud'hommeaux wrote:
I've gone ahead and created https://issues.apache.org/jira/browse/
OPENJPA-239 in order to track comments and patches for this issue.
On May 16, 2007, at 12:56 PM, Patrick Linskey wrote:
Would you guys be interested in getting donations in that area?
Certainly!
We do have our homegrown solution that does that (along with
some other
goodies as well). It's currently based on our metadata
representation
reverse-engineered from the database schema, it'd be nice (read:
more
flexible and maintainable) to have it based on openjpa metadata and
integrated in openjpa itself for the community to use.
The basics could be implemented as a new subtype of
ReverseMappingTool.ReverseCodeGenerator that did more intelligent
things in the various field / method generators; ideally, we'd
want to
provide options for things like leaving ORM metadata in XML but
putting @Entity and @Id etc. in classes, and obviously the
ability to
not write any XML at all and put everything in annotations instead.
-Patrick
