[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: JDO from Sun
On Wed, 03 May 2000, David Duddleston wrote:
> > We are already using Castor to map the XML class descriptor to
> > Java and vice
> > versa.
>
> That cool, didn't know that..
>
> > However, I'm not sure if this Castor feature is of any use
> > for ozone
> > because Castor produces 'static' Java code out of a XML schema to allow
> > programmers to manipulate XML resources via Java. What we need
> > for ozone is a
> > technology that produces a XML/DOM representation of any given
> > object on the
> > fly.
>
> Are you talking about converting a set of objects (nested) into a DOM
> representation or dumped to an XML file?
Both. When we have a DOM representation we may dump this to an XML file.
Actually we could use it to access/manipulate database content via standard XML
tools like XPath _and_ dumping the database content for backup/restore or to
migrate to a new ozone version.
>
> > Also, using ozone as a store back-end for Castor doesn't make to
> > much sense
> > IMO. Assaf, the maker of the persistency part of Castor, is strongly RDBMS
> > oriented... However, maybe the JDO spec itself leaves the door
> > open even for
> > OODBMSs. Ideas?
>
> I got the feeling this is the case. Castor should work will serval forms of
> storage, including OODBMS like ozone, but of course you have to write an
> adapter. I don't know the internal workings of Castor very well, but I would
> think that if ozone has some sort of supports for OQL/JDO like Castor, then
> writing the adaptor would be easy?
I don't have the time to look at this right now. However, ideas are as always
welcome.
Falko
--
______________________________________________________________________
Falko Braeutigam mailto:falko@softwarebuero.de
softwarebuero m&b (SMB) http://www.softwarebuero.de