0 Replies Latest reply: Nov 19, 2012 12:01 PM by Rafal Wardas RSS

    Cascade processing of QVD files.

    Rafal Wardas

      Hello,

       

      Let's say I have 2 views on database side ( Customer , Customer_LifeTime_Activity ) with following data :

       

      Customer:

       

      [c_id]

      A

      B

      C

       

      Customer_LifeTime_Activity:

       

      [c_id], [value]

      A 100

      A 200

      A 300

      B 100

      B 400

      B 855

      C 100

       

      Extracts are generated from database views weekly. I use FastExport dump to file, and file import it's quicker then ODBC for Teradata. So in the end we have 2 calls of load utility :

       

      LOAD

           @1 as c_id,

           @2 as value 

      FROM

      C:\Users\sg0216851\Desktop\Travelocity\04_QlickView\Dir_struct\Prod\data\in\tableA.txt

      (txt, codepage is 1252, explicit labels, delimiter is \x7, msq);

       

      which we will save into QVD files to get "Customer.qvd" and "Customer_LifeTime_Activity.qvd". Now I want to move some logic implemented

      in Teradata side into QlikView, for example :

       

      SELECT

      c.C_id, Count(*)

      FROM

      Customer c  inner join Customer_LifeTime_Activity ca on c.c_id = ca.c_id

      group by 1 ;

       

      I now it's possible with QV, but it's not the point of discussion. I guess I need some back end process which will generate second level of  QVD

      files based on QVD extracts calculation. How can I implement that for command line purpose ?  Second thing is error tracking, I want to return

      log of QVDs processing into Teradata database via ODBC, is it possible with QV?   Third thing which I want is implementation of easy QV report to

      manage jobs ( QVD processing  ). ;-)

       

      Thanks,

      -R