RE: Omitting retrieve values on a command -- MIDRANGE-L

A short while ago, Scott K mentioned the CEETSTA (Test for Omitted Argument) API, which can be called before you try to work with a parameter. This could be very useful and will take care of your MCH3601 problem. Check InfoCenter for documentation - it can be called with no problem in ILE CL (CALLPRC), as the parameters are pointers. Use *CHAR(4) variables for the first 2 parameters, *OMIT for the 3rd.

HTH
Vern

At 09:21 AM 9/22/2004, you wrote:

Dan,

 The problem is that MCH3601 Pointer not set for location referenced is
thrown.  Here are some code snippents:

Compile:
CRTCMD CMD(XTEST )
       PGM(XTEST )
       ALLOW(*IPGM *BPGM *IMOD *BMOD *IREXX *BREXX)

CMD        PROMPT( 'Testing retrieve parms' )

PARM       P1        TYPE( *CHAR ) LEN( 1 ) MIN( 1 )     +
                     PROMPT( 'Parm 1' )                  +
                     EXPR( *YES )

PARM       P2        TYPE( *CHAR ) LEN( 1 )              +
                     RTNVAL( *YES )                      +
                     PROMPT( 'Return Parm - P2' )

PARM       P3        TYPE( *CHAR ) LEN( 1 )              +
                     RTNVAL( *YES )                      +
                     PROMPT( 'Return Parm - P3' )

PARM       P4        TYPE( *CHAR ) LEN( 1 )              +
                     RTNVAL( *YES )                      +
                     PROMPT( 'Return Parm - P4' )


/*  Test return values in command - CPP */
XTEST: PGM      ( &P1 &P2 &P3 &P4)

       dcl      &p1   *char 1
       dcl      &p2   *char 1
       dcl      &p3   *char 1
       dcl      &p4   *char 1

       chgvar   &p2  '2'
       chgvar   &p3  '3'
       chgvar   &p4  '4'

endpgm
-------------
xtestc:  pgm

       dcl      &p1   *char 1  'a'
       dcl      &p2   *char 1
       dcl      &p3   *char 1

             XTEST      P1( &p1 )  P2( &p2 ) P3( &p3 )

             SNDPGMMSG  MSG(&p1 |> &p2 |> &p3)

endpgm
-----------

 I origianlly had through parm &p3.  When I added parm &P4, MCH3601 is
generated.  I hope that this makes it clearer.

 -mark

Original Message:
-----------------
From: Dan Bale dbale@xxxxxxxxxxxxx
Date: Wed, 22 Sep 2004 08:59:16 -0400
To: midrange-l@xxxxxxxxxxxx
Subject: RE: Omitting retrieve values on a command


> -----Original Message-----
> From: midrange-l-bounces@xxxxxxxxxxxx / M. Lazarus
> Sent: Tuesday, September 21, 2004 9:47 PM
>
>   I have created a "retrieve" command as a front end to some values in a
> file.  Is there a way to shield existing programs from a change to the
> command?  Normally it's easy to add parms to a command without affecting
> existing programs - as long as the keywords used in those programs still
> exist.  But on a retrieve type command, if I add a parm then the pointer
> has nowhere to go!
>
>   Is there an elegant way to get around this limitation or do I
> have to go
> into all the programs, add the parms and associated variables and
> recompile?  TIA.

I guess I'm not seeing the problem, Mark.  Are the new parms "required
parms"?  If you add parms to any command that is already being used in your
apps, those parms just won't be "retrieved" for those programs.  I use
RTVMBRD in many applications, but only use a small number of its parms.
What happens if IBM adds another retrievable parm to that command?  Will I
have to recompile all my programs?

db

--------------------------------------------------------------------
mail2web - Check your email from the web at
http://mail2web.com/ .

--
This is the Midrange Systems Technical Discussion (MIDRANGE-L) mailing list
To post a message email: MIDRANGE-L@xxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: http://lists.midrange.com/mailman/listinfo/midrange-l
or email: MIDRANGE-L-request@xxxxxxxxxxxx
Before posting, please take a moment to review the archives
at http://archive.midrange.com/midrange-l.