[
Date Prev][
Date Next][
Thread Prev][
Thread Next][
Date Index][
Thread Index]
[
List Home]
|
Re: AW: [jwt-dev] BPEL Generation
|
Le Tuesday 05 August 2008 16:31:09 Florian Lautenbacher, vous avez écrit :
> Hi Pierre,
>
> yes, you are right. There is an underlying framework that is used right now
> in another project called AgilPro integration framework that builds on top
> of the JBoss jBPM engine and that needs all the specific invoke calls
> (start_action, and so on).
Something remains unclear here: do you use a BPEL engine (from JBPM suite)
or the workflow engine (also from JBPM suite) (or both maybe: one for BPEL interpretation,
the other the start_action, and so on ;-)))?
> Therefore, all the additional variables and
> porttypes are defined. It is quite easy to change these templates in order
> to support any other BPEL engine. Our strategy currently was to create an
> invoke call for each data, for each application, for each role, etc. that is
> referenced from an action, before execution the action itself. But that is
> only one strategy. It would of course also be possible to use the data
> simply as parameters and have one invoke call for each action instead. Your
> freedom of choice here!
Fine.
> Concerning the algorithm that is used to change the graph-based model into
> the block-based language, please have a look at [1] that describes the
> workflow codegeneration framework in principal and [2] to see how the
> transformation algorithm internally exact works.
> [1] http://www.dsmforum.org/events/DSM07/papers/roser.pdf
> [2] http://www.ds-lab.org/publications/reports/2008-08.html
I read those papers. From 1, I don't really get what are the constraints: can you generate BPEL from
any JWT model, even in the case of unstructured ones (see the one sent in my previous post)?
From 2, I understand that it helps in the finding of components used in the BPMN2BPEL transformation
from Van der Aalst et al.
Anyway, if I am not wrong, the BPEL file sent to you in my previous post which is the result of the
JWT/AgilPro BPEL transformation algorithm is not semantically equivalent to my unstructured JWT
model? In the model, Activities B and C can be executed in parallel only after A. Activity D should
wait until B and C are completed. E can start as soon as C finished. Finally, F should be executed
only after D and E. From the BPEL file, I can't find where those constraints are expressed, if they are.
Can you tell me:
1. Is the resulting BPEL file correct (means: semantically equivalent to my JWT model)?
If yes, how? If no, why?
2. What are the constraints at the JWT level for the correct generation of BPEL code?
Best regards and thanks for your time. It is highly appreciated.
--
Pierre Vignéras
Bull, Architect of an Open World TM
*BPM Team*, Bull R&D
1, rue de Provence
38130 Echirolles (France)
Direct Line: +33-4-76-29-74-06
*Orchestra*, The BPEL open source project: http://orchestra.objectweb.org
*Bonita*, The XPDL open source project: http://bonita.objectweb.org
