Tom Baeyens wrote:
>> I did try to model resource names from the jbpm docs. What did I miss?
>
> would be nice if the urls would correspond closer to the concepts in
> the api service methods. like e.g.
>
> http://bpm.org/repository/deployments
> http://bpm.org/repository/processdefinitions
> http://bpm.org/execution/processinstances
> http://bpm.org/execution/executions
> http://bpm.org/management/jobs
>
>> I really need you guys to help flush out use cases and whether or not the distributed model (not the media type of a PD) can handle it. It would also be cool if you were open to make changes to jbpm itself. I don't mind writing the RESTful implementation though.
>
> i don't get this yet: "the distributed model (not the media type of a
> PD) can handle it". can you elaborate
>
I need help defining the requirements and use cases the spec needs to
address.
> of course we're open to changes.
>
>
> Another question that I have is about reading the interface
> specification.
>
> HEAD http://bpm.org
>
> àResponse
> HTTP/1.1 OK 200
> Link: <http://bpm.org/definitions>, rel=”definitions”
> Link: <http://bpm.org/definitions>, rel=”create definition”
>
> Does that mean that user should use the rel parameter like this?
> http://bpm.org/definitions?rel=definitions
>
No. URLs are opaque in REST. Rel just defines the logial name of the
link. An analogy is HTML links:
<a href="http://jboss.org/jbpm">The JBPM Project</a>
The href is the URL. "The JBPM Project" is the "rel" (relationship).
What a Link describes is what URL you should follow to retrieve
information, or what URL your should POST or PUT representations (XML
documents, JSON documents, etc.) to change the state of a particular
thing in your system.
Am I making sense? Look at the examples I gave in the specification to
see how it works. Read in with this in mind though: *URLs ARE OPAQUE*.
The structure of the URL doesn't matter.
> Is there a pointer that describes the format: for Link: <http://
> bpm.org/definitions>, rel=”definitions”
http://tools.ietf.org/html/draft-nottingham-http-link-header-03
Atom links are a good definition too.
http://tools.ietf.org/html/rfc4287#section-4.2.7
--
Bill Burke
JBoss, a division of Red Hat
http://bill.burkecentral.com