Re: MCH3601 on Fetch

[Scott Klement <rpg400-l@xxxxxxxxxxxxxxxx>]

Hi Pete,

Is one of your host variables (ORDER#,ITEM,QTY,VIN#) declared as a 
parameter that you forgot to pass, by chance?

If that's not the case, then... are you using pointers?  Any of these a 
BASED field?

This probably isn't related to the MCH3601, but...  is there some reason 
you're using dynamic SQL instead of static SQL?  I can't see a reason 
(in the part you posted) why you'd want to do extra work, and have your 
program run slower?  Seems like static SQL would be easier to code, and 
run faster...


On 2/9/2011 4:42 PM, Pete Helgren wrote:
> This one has me stumped.  I have a prepared statement that selects a row
> based on parameter values.  The sql looks like this:
>
>    gSQLStmnt2 = 'select distinct extord, lprod, lqord, ' +
>                       'substr(extvin,1,10) from ecl ' +
>                       'inner join eclx01 on lord=extord and lline=exline ' +
>                       'where lprod = ? and extvin = ? ';
>
> I then do the following in another sub procedure:
>
>        c/free
>          EXEC SQL
>           PREPARE S2 FROM :gSQLStmnt2;
>          EXEC SQL
>               DECLARE C2 CURSOR FOR S2;
>          EXEC SQL
>               OPEN C2 USING :ItemNum, :Vin;
>         /end-free
>
> When I run the program the joblog shows this:
> .......
> Blocking used for query.
> Cursor C2 opened.
> 1 rows fetched from cursor C2.
> Pointer not set for location referenced.
>
> That MCH3601 occurs on this fetch:
>
>             FETCH FROM C2 INTO :order#,:Item,:QTY,:vin#;
>
> This is the first time I have used a parameterized SQL statement in an
> RPG program so I have probably done something wrong.  The compiler
> listing shows this at the statement the error points to:
>
> 451 C     SQL_00032     IFEQ      '1'
> 452 C                   EVAL      ORDER# = SQL_00034
> 453 C                   EVAL      ITEM = SQL_00035
> 454 C                   EVAL      QTY = SQL_00036
> 455 C                   EVAL      VIN# = SQL_00037
> 456 C                   END
>
> It is almost as though it doesn't like the host variables.  I double
> checked the sizes and types of the variables and they *seem* OK.
>
> Anybody seen something like this before and can point me to my error?