Skip to main content

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
AW: [mdt-bpmn2.dev] SubProcess as InteractionNode

Hi Anne,
from the semantic point of view a message flow to a collapsed call activity or collapsed sub process does not make sense (as it also doesn't make sense in the expanded case to have the message flow line ending at the border).
Note that none of the two are considered to be used as black-box - in contrast to pools, that can be used as black boxes and also support message flow to the border. A user would always expect that he (or she) could expand a collapsed call activity or sub process.
As the Di depiction has a way to visualize what you want, I suggest that you populate your sub process with some activities or events that can send/receive your messages. Let the message flow end there. The rest of the sub process does not need to be complete, e.g. you could skip sequence flow etc. Collapse your sub process and in BPMNDI BPMNEdge use source/target to point to the shape of the collapsed sub process. This is the way how the standard expects it to be.

Best regards,
 Reiner.

-----Ursprüngliche Nachricht-----
Von: mdt-bpmn2.dev-bounces@xxxxxxxxxxx [mailto:mdt-bpmn2.dev-bounces@xxxxxxxxxxx] Im Auftrag von Anne Baumgraß
Gesendet: Mittwoch, 4. August 2010 13:04
An: BPMN2 Developers Mailing List
Betreff: Re: [mdt-bpmn2.dev] SubProcess as InteractionNode

Hi Rainer,

thanks for the fast response. From my point of view: If the message flow 
from an embedded SubProcess makes no sense than also the message flow 
from/to a participant ("Black Box") would make no sense.

Even if the picture on page 44 can be interpreted as collapsed 
CallActivity or collapsed SubProcess, neither is possible in the MDT 
BPMN Modeler. I would like to connect via message flow a collapsed 
subprocess to another participant. I want to use the collapsed sub 
process to point out that the activity could contain more activities, 
like a "Black Box". Or do you know other possibilities to depict that an 
activity is not completely defined? Having another look at BPMN I even 
found such an example on page 113 in figure 9.4.

Regards,
Anne

Am 03.08.2010 10:51:
> Hi Anne,
> welcome!
> About your question: SubProcess (embedded subprocess) is indeed not a InteractionNode and cannot be connected with MessageFlow. The picture on page 44 could be either interpreted as collapsed CallActivity or collapsed embedded sub process (SubProcess). I would assume that the authors had in mind the first one. For me it does not make sense to have incoming or outgoing message flow from an embedded subprocess, as the subprocess does not send or receive anything. Anyway, the depiction could be valid, if you have incoming/outgoing message flow from/to one or more elements that are contained in the SubProcess - and the subprocess is visually collapsed. But in this case the semantic model still uses the contained element (e.g. event or send/receive task) as source/target), but the BPMNEdge (in DI) would point to the collapsed SubProcess.
>
> Regards,
>   Reiner.
>
> -----Ursprüngliche Nachricht-----
> Von: mdt-bpmn2.dev-bounces@xxxxxxxxxxx [mailto:mdt-bpmn2.dev-bounces@xxxxxxxxxxx] Im Auftrag von Anne Baumgraß
> Gesendet: Dienstag, 3. August 2010 10:34
> An: BPMN2 Developers Mailing List
> Betreff: [mdt-bpmn2.dev] SubProcess as InteractionNode
>
> Dear Community,
>
> recently I started to model BPMN 2.0 models in Eclipse with the MDT
> BPMN2.0 Modeler. Importing my own models I realized, that a MessageFlow
> between SubProcesses causes a problem (IllegalValueException). Looking
> at the cause it seems to be a ClassCastException:
> org.eclipse.bpmn2.impl.SubProcessImpl cannot be cast to
> org.eclipse.bpmn2.InteractionNode.
>
> If I understand the problem correctly, the message flow in MDT BPMN2.0
> Modeler is not allowed between subprocesses, but from the published BPMN
> 2.0 Beta 2 (dtc/2010-06-04) it is possible for all kind of activities:
> depicted in the table 7.4 on page 44 and documented on page 124. For
> this reason the meta-model on page 123 would also be erroneous (it
> states only the Task as InteractionNode).
>
> Regards,
> Anne
>
>    


-- 
-----------------------------------------
Dipl. Inf. Anne Baumgraß
Research and Teaching Assistent

WU - Wirtschaftsuniversität Wien
Vienna University of Economics and Business Institute for Information Systems and New Media

UZA II, Augasse 2-6, A-1090 Vienna, Austria

Tel: + 43 1 313 36 / 4354
Fax: + 43 1 313 36 / 746
Email: anne.baumgrass@xxxxxxxx
http://nm.wu.ac.at

_______________________________________________
mdt-bpmn2.dev mailing list
mdt-bpmn2.dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/mdt-bpmn2.dev


Back to the top