tuscany-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Kris De Waele (JIRA)" <tuscany-...@ws.apache.org>
Subject [jira] Updated: (TUSCANY-1493) Snapshot mapping framework to convert DataObjects to and from Java objects
Date Thu, 22 Nov 2007 13:36:45 GMT

     [ https://issues.apache.org/jira/browse/TUSCANY-1493?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel

Kris De Waele updated TUSCANY-1493:


apparently there is no public maven repo that hosts these artifacts.

We added the eclipse artifacts to our own repo using (something almost identical to) the maven
eclipse plugin ( http://maven.apache.org/plugins/maven-eclipse-plugin/to-maven-mojo.html )

> Snapshot mapping framework to convert DataObjects to and from Java objects
> --------------------------------------------------------------------------
>                 Key: TUSCANY-1493
>                 URL: https://issues.apache.org/jira/browse/TUSCANY-1493
>             Project: Tuscany
>          Issue Type: New Feature
>          Components: Java SDO Implementation
>            Reporter: bert.robben
>             Fix For: Java-SDO-Next
>         Attachments: agfasdo.tar.gz, sdo-snapshot.zip, sdo.zip
> We're developing 3-tier applications with a  swing client, JBoss app server and a couple
of databases in the back-end. We use sdo as mechanism to exchange data between our client
and our server. On the server side we have a fairly complex implementation based partially
on Hibernate pojo's, partially on an inhouse legacy persistency system. The legacy part (although
written in Java) is very hard to change. In this environment we often need to convert between
data objects and server side objects (typically, but not always at a transition from server
to client or vice versa). 
> To support this we developed a mapping framework that allows us to convert data between
SDO and ordinary Java objects. This framework defines a number of important concepts.
> - A snapshot is an opaque collection of structured data at a given moment in time. It
is opaque in the sense that the data inside can't be accessed directly.
> - A Mapper is an interface that defined how data can be accessed from an object. We have
implementations for SDO DataObjects, normal Java POJO's (following java beans convention).
hibernate pojos, and support for customizing this to access any kind of object (as we need
for instance for our legacy objects).
> - We defined a DataAccessService (sorry for the confusing name) that given a mapper and
some objects can create a snapshot. Given a snapshot and a mapper it can instantiate new objects.
As such we can convert data to and from data objects very easily.
> This framework is part of our in-house developed implementation of the SDO spec. We want
to share our code and experience with the open-source community. As such, Frank Budinsky proposed
that we make JIRA request for this to start the discussion. 
> I'll attach the core classes for this feature to this JIRA. At this moment I'm not making
the entire code available because we implemented more than one additional feature (I'll add
some more JIRA over the next days) and I'd like to have a more focussed discussion. Also,
at this moment in time some smaller parts of implementation (for instance the hibernate integration)
are still implemented as a specialization of our SDO implementation. As such they can't be
built outside of the rest of our application (which is not open-sourced). However this is
just a matter of finding enough time to move them to our sdo component.

This message is automatically generated by JIRA.
You can reply to this email to add a comment to the issue online.

To unsubscribe, e-mail: tuscany-dev-unsubscribe@ws.apache.org
For additional commands, e-mail: tuscany-dev-help@ws.apache.org

View raw message