incubator-clerezza-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Reto Bachmann-Gmür (JIRA) <j...@apache.org>
Subject [jira] [Commented] (CLEREZZA-483) allow jsr311 code to return normal Java Objects
Date Sat, 02 Apr 2011 17:11:05 GMT

    [ https://issues.apache.org/jira/browse/CLEREZZA-483?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13015054#comment-13015054
] 

Reto Bachmann-Gmür commented on CLEREZZA-483:
---------------------------------------------

The title of the issue is missleaning, there's nothing that prevents jax-rs (aka jsr311) resource
methods to return normal java objects. I think the issue should be: Create tool to generate
read-only GraphNodes from Java Objects.

> allow jsr311 code to return normal Java Objects
> -----------------------------------------------
>
>                 Key: CLEREZZA-483
>                 URL: https://issues.apache.org/jira/browse/CLEREZZA-483
>             Project: Clerezza
>          Issue Type: New Feature
>            Reporter: Henry Story
>
> Sometimes one has java objects and it is tedious to transform them into some graph notation.
In the case I am dealing with I have an X509Certificate object and a subject, and I need to
produce some test html pages. The SSP code should be able to just manipulate the java object,
though the object should somehow be disallowed from transforming the databases (ie it should
be read only).
> Of course for this to work with the Clerezza philosophy, the resource that object represents
should have an rdf/xml and other representations. So what is needed is a way to automatically
transform objects into rdf.
> It turns out that this is quite easy, and I showed how to do it a few years ago in a
blog post
>     http://blogs.sun.com/bblfish/entry/serialising_java_objects_to_rdf
> That shows how one does it with objects that have @rdf annotations on methods, fields
and classes. For objects on which that access is not possible, one will need to use blank
nodes and create made up uris of the form perhaps java:{version}:package or something for
class names and methods. Or one would want an java->rdf mapping file, which of course would
be easy to write with the above java urls.
> In any case having a quick mapper is what the sommer project I worked on a while ago
was about, and should certainly be useful in many ways.
> It is true that with Scala working with RDF is not that problematic: writing res/FOAF.knows
is pretty easy to the eye and to write. Doing the same in java was realy really really really
tedious. So it would be useful to work out when using scala object notation held an advantage.
My guess is that there are situations where one just receives a java object, and it's going
to be a lot faster to use it, then to go through a graph. That's the case here where I get
a build X509 certifiate from the OS. Perhaps that is just really a rare case.  Perhaps it's
worth listing those cases where we think it might be useful anyway. It may be that we are
mostly mistaken there....

--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira

Mime
View raw message