Re: SQL Cursor Not Open

[Sam_L <lennon_s_j@xxxxxxxxxxx>]

Check SQLCOD or SQLSTT after every SQL statement, except DECLARE CURSOR, 
because it is not an executable statement.

If you look in the job log you should find the statement that failed 
before you get the cursor not open error.

Sam


On 10/25/2012 2:42 PM, Jerry C. Adams wrote:
> First: V5R1
>
>
>
> I have a simple (I thought) SQLRPGLE program that is supposed to select all
> of the records from our Item Master, put them in order, and then use the
> retrieved records to print a report.  Done this a few dozen times in the
> past, but this time no dice.  Under debug the program ends (normally, not an
> abend) when I test the value of SQLSTT, which is 24501 (Cursor is not open).
>
>
>
> Relevant, I think, code:
>
>
>
> <code>
>
>           $BuildSQL();
>
>
>
>        /end-free
>
>
>
>       C/Exec SQL
>
>       C+  Prepare MySQLStmt from :SQLString
>
>       C/End-Exec
>
>       C/Exec SQL
>
>       C+  Declare NVList Cursor for MySQLStmt
>
>       C/End-Exec
>
>       C/Exec SQL
>
>       C+  Open NVList
>
>       C/End-Exec
>
>
>
>        /free
>
>
>
>           DOW 1 = 1;
>
>
>
>        /end-free
>
>
>
>       C/Exec SQL
>
>       C+  Fetch Next from NVList into :NVLine
>
>      C/End-Exec
>
>
>
>        /free
>
>
>
>             IF  SQLStt  = '02000' OR  SQLCOD  < *Zeros;
>
> </endcode>
>
>
>
> The $BuildSQL subprocedure successfully builds the SQLString value:
>
> SELECT imloc, imdesc, imumi, imbal, impidx, impitm, imclas,
>
> FROM "A.IVMAST" ORDER BY imloc, impitm FOR Read Only
>
>
>
> NVLine is a data structure with the field names equivalent to the SELECT
> with the proper attributes and order.
>
>
>
> SQLSST = '24501' and SQLCOD = -501 under debug.  Both of which mean that the
> sursor is not open.
>
>
>
> This is a single mod program; the only subprocedure is $BuildSQL which only
> creates the SQLString.  The code snippet is from a subroutine.  So I don't
> see that the cursor could be closed by ending the activation group, which
> was the compile option.
>
>
>
> I have compared this to similar programs I wrote using the same methodology.
> I, also, know/suspect that I am overlooking the obvious and will be
> embarrassed when we find out what is wrong, but I have looked at this and
> debugged it for the last hour.
>
>
>
> Thanks.
>
>
>
> Jerry C. Adams
>
> IBM i Programmer/Analyst
>
> Never argue with a women when she's tired -- or rested.