[jboss-user] [jBPM] - Re: Instantiating the KnowledgeBase

Esteban Aliverti do-not-reply at jboss.com
Thu Apr 7 09:18:43 EDT 2011

Esteban Aliverti [http://community.jboss.org/people/eaa] created the discussion

"Re: Instantiating the KnowledgeBase"

To view the discussion, visit: http://community.jboss.org/message/598503#598503

We need to make a difference between 3 things:
1. Process Definition (human-readable): this is the xml file where you define your process using BPMN2
2. Process Definition (binary): this is the representation of your process definition using Objects of classes like Process, Node, etc. This is what you will find inside your knowledge base
3. Process Instance (also binary): when you start or resume a process, you get the instance of it containing the state of the process.

So, if you need your human-readable-process-definition in your application, then you can download it from guvnor using webdav. 
If you need the binary representation of your process, or maybe you need to get all the process names inside the kbase, then you will need to use kbase object to retrieve that information (Of course, this information will be available only after the agent processes the changeset).

What we usually do, is to point the agent to the binary package created by guvnor (i.e.  http://localhost:8080/drools-guvnor/org.drools.guvnor.Guvnor/package/defaultPackage/LATEST http://localhost:8080/drools-guvnor/org.drools.guvnor.Guvnor/package/defaultPackage/LATEST). This package will already contain all the rules and process definitions inside.

Reply to this message by going to Community

Start a new discussion in jBPM at Community

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.jboss.org/pipermail/jboss-user/attachments/20110407/c4306fc6/attachment.html 

More information about the jboss-user mailing list