I think there are two uses of jBPM. The traditional use is a long transaction where multiple nodes occur within one transaction. I see the validity of this, but this isn't the only use for jBPM.
The second use for jBPM is to save state when each node is complete. Yes, this means the node is doing stuff with the transaction, but this is what I personally wanted from the get go. Without this, you will be UNABLE to monitor jBPM processes externally because the results are in memory until committed to the database. Without this second solution, jBPM will never be able to handle asynchronous processing. I have essentially been hacking jBPM to do this and I am about finished. There are obviously others that want this functionality as well. My coworkers were a bit dumbfounded to discover I had to do this manually. You may say that this makes it flexible. I say it makes it cumbersome. By saying use two is invalid, I think you are turning people away from a valid use of the process engine. To do this, you simply need to add three nodes (at a minimum): StateAsync, ForkAsync, JoinAsync. This does not break jBPM. It just provides a different use where the nodes themselves do work with the jBPMSession. Sean View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3911921#3911921 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=3911921 ------------------------------------------------------- This SF.net email is sponsored by: Splunk Inc. Do you grep through log files for problems? Stop! Download the new AJAX search engine that makes searching your log files as easy as surfing the web. DOWNLOAD SPLUNK! http://ads.osdn.com/?ad_id=7637&alloc_id=16865&op=click _______________________________________________ JBoss-user mailing list JBoss-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/jboss-user
