Skip to Content

Time and again we face  a scenario where we receive frequent changes in files.The files are exported 

through Open Hub Destination or Infospokes.The changes in file may range from filters in data

(e.g. source systems,company codes,plant) .It may also apply for changes in calculation logic for

keyfigure values.This blog will try to deal with it by suggesting little change in the dataflow.

This will remove the need for changing the transformation or update rules.

 

In traditional data model we just have a simple infoprovider and a transformation to Open Hub

which is loaded via DTP or infospoke having BADIs written for filtering /manipulating the exported data.

Data Flow for traditional data model is like :

INFOPROVIDER–>TRANSFORMATION+DTP–>OPEN HUB DESTINATION

INFOPROVIDER–>INFOSPOKE(BADIs written for data manipulation like in end/start

routine of Transformation.)

 

Making slight changes in this dataflow and inserted the APD in between i.e.

INFOPROVIDER–>APD–>Direct Update DSO–>Transformation–>OPEN HUB DESTINATION

INFOPROVIDER–>APD–>Direct Update DSO–>INFOSPOKE 

Going Step by Step:

  • Create A Query On top of  the Infoprovider,
  • Put as much logic as possible in it with calculations and filters etc.
  • Select this query as source in APD process .

 test

PIC 1. APD with source query and direct update dso.

  (PIC 2. Select Query created on the source infoprovider)

 

  • Create direct update DSO resembling OPEN HUB DESTINATION/INFOSPOKE fields.

 

 PIC 3. Open Hub Destination,Click on Field Definitions for copying the fields to Direct update DSO

  • Now Join the Query with direct update DSO using the mapping rules in APD(Transaction RSANWB).

          Refer Pic 1.

 

 PIC 4.Mapping between Source Infoprovider Query and direct update DSO. 

  • Create one Transformation mapping the direct update dso with Open Hub or Infospoke.
  • If you are using an Infospoke,then in RSBO make rules mapping the source infoprovider to

         the Infospoke.(use BADIs is required).

  • Run the APDs using standard program RSAN_PROCESS_EXECUTE using variant name as APD.
  • Create a variant for the same in SE38.

       (you can use it while executing the APD process through process chain).

 

 PIC 5. Using Program RSAN_PROCESS_EXECUTE in Process Chain variant to run the APD

  • Now run the DTP to Export the files/or Execute the Infospoke to export the file
  • The whole process can be automated using the Process chain shown below.
  • The infospoke can be directly executed in Process Chain while the Open Hub DTP will be used

          for exporting files.

 

 

Pic 6. Automated process using Process Chains

To report this post you need to login first.

2 Comments

You must be Logged on to comment or reply to a post.

  1. Former Member
    Rakesh – This approach (using Query) will put unnecessary load on OLAP and will have issues when extracting high data volume. It is better from performance point of view to change transformations than to go for a query-APD.

    Frequent changes in the requirements can be handled “non-technically” by bringing in release & change management processes in the picture.

    Regards,
    Amol

    (0) 
    1. Former Member Post author
      thanks Amol,
      but ..I think running a single query for a data load won’t be that big load…(compare this with 1000 users running ur report).

      In anycase this not the generic solution/data model..it is meant for scenarios where you have frequest changes in filters etc…Also it has advantages like using same infospoke and D/U dso you can generate multiple files using different apds..with different soruce queries…the bottomline is flexibility…without writing code and transporting..you can achieve the same in least time and effort….
      regards
      rk

      (0) 

Leave a Reply