[
https://issues.apache.org/activemq/browse/SM-1248?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Andrea Zoppello updated SM-1248:
--------------------------------
Attachment: JbiHelper.java.patch
DroolsEndpoint.java.patch
Hi Guillaume,
About fault, i totally agree with you, the only reason why i provided a
faultToDefaultTarget services, is that some user refers to "logical errors"
with the term fault also if this is not really a jbi fault, and they want to
manage the fault by hand
within the flow.
The fault in that case is a message with a particular content, that is for
example handled by a router, in most of my use case my users do the thing as i
described, they use faultTodefaultTargetService with a particular message
content and they manage with a router.
BTW now i'm going to inform them to use the routeToDefault(content) method.
I think we're just sayng the same thing, it's just terminology.
The new patch apply all suggestion you've made.
Take a look and let me know
Regards
> ServiceMix Drools Enhancements
> ------------------------------
>
> Key: SM-1248
> URL: https://issues.apache.org/activemq/browse/SM-1248
> Project: ServiceMix
> Issue Type: Improvement
> Components: servicemix-drools
> Affects Versions: 3.2.1
> Reporter: Andrea Zoppello
> Attachments: DroolsEndpoint.java.patch, DroolsEndpoint.java.patch,
> JbiHelper.java.patch, JbiHelper.java.patch
>
>
> To be complete the servicemix 3.2.1 drools components, require some
> enhancements:
> 1) There's the need to support a default destination, where exchange will be
> routed if none rules in the drools file is verified.
> To support this, we need to provide a way to specify default destinations.
> there will be a defaultTargetService ( QName ) attribute and a
> defaultTargetURI attribute in drools endpoint class. ( defaultTargetURI will
> have the precedence on defaultTargetServiceAttribute)
> 2) We must add the possibility to configure not only global "objects" for
> memory but also a list of "asserted objects" to be asserted in memory ( so
> it's possible to use that objects in LHS part of rules )
> 3) The actual code of jbi helper does not care about sender and correlation
> id.
> 4) We need a method in jbi helper object to support the concept of fault
> withn the flow, ( to default destination ) without generate a jbi fault. a
> fault in flow from jbi point of view is a "correct" normalized message.
> The attached patches solve all the above in a very generic way.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
