8 Replies Latest reply: Sep 13, 2010 1:08 AM by Jan Vervecken RSS

    mutual dependent task-flows

    Jan Vervecken
      hi

      Please consider the example application
      in http://orclsamples.googlecode.com/files/TFParameters.zip
      referred to in the blog post "Passing Parameters between Bounded Task flows" by Zeeshan Baig
      at http://baigsorcl.blogspot.com/2010/07/passing-parameters-between-bounded-task.html

      It has a task-flow "flowA" and a task-flow "flowB" that both call each other, as can be seen in "mutual-dependant-flowAflowB.png".

      question:
      - (q1) Is it in general a good idea to have such mutual dependent task-flows that call each other? If not, please explain. If it is no problem, please suggest a use-case where it applies.

      thanks
      Jan Vervecken
        • 1. Re: mutual dependent task-flows
          Jang-Vijay Singh
          - (q1) Is it in general a good idea to have such mutual dependent task-flows that call each other? If not, please explain. If it is no problem, please suggest a use-case where it applies.
          Taskflows should be completely decoupled.
          A 'called' TF shouldn't need to know anything about its caller(s).
          I would rather have:
          TFa calls TFb.
          +TFb with a general 'task-flow return' activity to return to whoever the caller is.+

          Perhaps the example you linked to is just meant to illustrate a technical concept without promoting any specific design pattern.
          • 2. Re: mutual dependent task-flows
            Jan Vervecken
            Thanks for your reply Jang Vijay Singh.
            ... promoting any specific design pattern
            makes me wonder ...

            - (q2) Does any documentation on task-flow related design patterns or anti-patterns exist?

            (I know there is "An Oracle White Paper - June 2010 - Task Flow Design Fundamentals " ...
            at http://www.oracle.com/technology/products/jdev/11/collateral/adf_task_flow_design.pdf
            ... but that is not really about task-flow design patterns or anti-patterns,
            e.g. these "mutual dependent task-flows " seem to be a good candidate to be documented as "do" or "don't".)

            Looking forward to more feedback on (q1) and (q2).

            regards
            Jan
            • 3. Re: mutual dependent task-flows
              Jang-Vijay Singh
              - (q2) Does any documentation on task-flow related design patterns or anti-patterns exist?
              I did start on general ADF antipatterns
              (http://groups.google.com/group/adf-methodology/browse_thread/thread/ec5ea52229b92ca9/d5fd3a07d8a0a77a?lnk=gst&q=anti-patterns#d5fd3a07d8a0a77a )
              on the ADF EMG and was going to add to that document 'soon'.
              Do add to it as i've seen you post on that group recently.

              Jang Vijay
              • 4. Re: mutual dependent task-flows
                Frank Nimphius-Oracle
                Hi,

                a good way to look at task flows is as a service with a UI attached to it. The task flow input parameters and output parameters are the contract that is defined between the client (consumer) and the service provider (the task flows). The implementation of the task flow - like in Web Services - is a detail that only the developer of the task flow need to know

                Frank
                • 5. Re: mutual dependent task-flows
                  Jan Vervecken
                  Thanks for your reply Jang Vijay.

                  Thank you for referring to the discussion thread "How about a document on Anti Patterns?"
                  at http://groups.google.com/group/adf-methodology/browse_thread/thread/ec5ea52229b92ca9

                  regards
                  Jan
                  • 6. Re: mutual dependent task-flows
                    Jan Vervecken
                    Thanks for your reply Frank.

                    I am not sure how to interpret your contract/implementation argument in the context of question (q1). It seems to suggest that "mutual dependent task-flows" should not be a problem in that respect, but it also does not confirm that it is a good idea in general to have such mutual dependent task-flows that call each other.

                    regards
                    Jan
                    • 7. Re: mutual dependent task-flows
                      Jan Vervecken
                      fyi

                      see also discussion thread "(anti-)patterns [was: How about a document on Anti Patterns?] "
                      at http://groups.google.com/group/adf-methodology/browse_thread/thread/990e41c2ffc2c174

                      regards
                      Jan