Re: SQL Update in RPGLE.

Just a couple of off the cuff thoughts...

On 3/30/06, Mike Troxclaire <mike_troxclaire@xxxxxxxxxxx> wrote:
>
> Pros,
>
> =============================================================
>      FCUST      UF   E             DISK    RENAME(CUSTR:CSREC)
>
>       /FREE
>        DoW not  %eof(CUST);
>          Read CUST;
>          If not %eof(CUST);
>            Exsr UpdateCSREC;
>          EndIf;
>        EndDo;
>        dsply 'UPDATE SUCCESSFUL';
>       /END-FREE


Won't this message show every time you run the program?  It has no
conditioning...

      /FREE
>        BegSR UpdateCSREC;
>       /END-FREE
>
>
>      C/EXEC SQL
>      C+ UPDATE TESTING/CUST SET CUSSTA = 'TX'
>      C+ WHERE CUSSTA = 'TN'
>      C/END-EXEC
>       /FREE
>        *inlr = *on;
>        EndSR;


Why is your *INLR statement inside the update Subroutine? Is this supposed
to control the loop in the mainline?

How is all this logic any different than simply executing the UPDATE sql?  I
mean, the program opens the file for update (unnecessary), reads 1 record,
goes to the execute subroutine, updates all the necesary records, and then
sets on LR presumably to break out of the loop that calls the SR.  Try this
instead:

(notice no F-Spec)
===============================
    C/EXEC SQL
    C+ UPDATE TESTING/CUST SET CUSSTA = 'TX'
    C+ WHERE CUSSTA = 'TN'
    C/END-EXEC
     /FREE
      if sqlstt = '00000' ;
       dsply 'UPDATE SUCCESSFUL';
      else ;
       dsply 'UPDATE FAILED';
      endif ;
      *inlr = *on;
     /END-FREE
=============================================================

Just some quick thoughts...
--
Joel Cochran