Hi,

I am trying to chain a value and display it as per below condition but despite the value exists in this file i get wrong result any idea why is it happening , already checked my lib. list,data using sql and through debug mode nothing worked yet.

[code]

            // Check if it's a valid pallet (exists in FILE1)     
                              Chain(N) (LdWhse:' ':WSCASN) FILE1;       
CR01        //          Chain(N) (%TRIM(WSCASN)) FILE1;           
                        If  Not %Found(IDCASE10);                    
                     WsMsg = 'Pallet does not exist in FILE1';    
                              Flg_Error = 'Y';                       
SR01                          LeaveSr;                               
                        EndIf;                                       

[/code]

Note i tried both the ways like above first by this chain condition :-1)Chain(N) (LdWhse:' ':WSCASN) FILE1;

2nd by this condition :- CR01        //          Chain(N) (%TRIM(WSCASN)) FILE1;           

i have a field WSCASN defined in my display file let's say 'D1' where this field is defined as 20 A and same definition inside for File1 only it's name is IDCASN in this file for example when i tested this code for key value '305446C003871A'  then though this value does exist in IDCASN field of file1 and debug mode also i can see it's assigned to this WSCASN through display file screen still above  program code displays wrongly as ''Pallet does not exist in FILE1'; 

Could someone please advise why is it happening ?

Thanks much...

------------------------------
Kevin Steve
------------------------------

You should check %EQUAL() (Excat Match) instead of %FOUND() .

From the IBM documentation:

%FOUND returns '1' if the most recent relevant file operation found a record, a string operation found a
match, or a search operation found an element. Otherwise, this function returns '0'.

%EQUAL returns '1' if the most recent relevant operation found an exact match; otherwise, it returns '0'.

------------------------------
Birgitta Hauser
Database and Software Engineer
Selfemployed - Modernization-Education-Consulting on IBM i
Kaufering
+49 170 5269964
------------------------------

The trim is doing nothing and should be removed.

The %Found is applied to file IDCASE10 - but you are chaining to File1

The version of the Chain is not valid - copy/paste error? Chain(N) (LdWhse:' ':WSCASN) FILE1;  

------------------------------
Jon Paris
------------------------------

Chain(N) (LdWhse:' ':WSCASN) FILE1;       
Chain(N) (%TRIM(WSCASN)) FILE1;           

Looks like the file has 3 fields as key (per the 1st statement, it is searching with 3 arguments, whse, space and case)

The second statement is searching using only one (the case)

------------------------------
Carlos Romero
------------------------------

Good point - I missed that. Looks what may be needed is SETLL rather than Chain.

What is still unclear is whether the OP is thinking that the ‘’ will give a null key or what is intended. I’m thinking that it will result in a second key of blanks - probably not what was wanted.

Jon P


Original Message:
Sent: 10/10/2023 9:19:00 AM
From: Carlos Romero
Subject: RE: Chaining a value

Chain(N) (LdWhse:' ':WSCASN) FILE1;       
Chain(N) (%TRIM(WSCASN)) FILE1;           

Looks like the file has 3 fields as key (per the 1st statement, it is searching with 3 arguments, whse, space and case)

The second statement is searching using only one (the case)

------------------------------
Carlos Romero
------------------------------

Kevin, 

  Birgitta is correct, but our code has hundreds of examples of %Found doing this same type of thing just fine.  For us, %Equal is usually used in a SETLL situation.

  You really don't give enough information to draw  exact conclusions.  So I'm going to make assumptions, and show an example I think fits your issue.

  Assuming your file IDCASE10 is keyed and there are 3 fields in the key.  Attributes and names are unknown, and the record format is also unknown.  The declare should look like this, and I'm stepping it up a bit with a key list for the chain that is a data structure modeled on the file key list.  We as a rule code the record format as the I/O for single record format files, and the file name in the "test" for success.  See what you think of this: 

       Dcl-f IDCASE10 keyed;
       Dcl-ds CseKey LikeRec(IDCASEREC: *Key) inz;
   ‚   // real record format unknown...
       Clear CseKey;

       IDWHSE = LdWhse;
       IDCASN = WACASN;
   ‚   // real key field names of record format unknown...

       Chain(n) %Kds(CseKey) IDCASEREC;
 1b    If Not %Found(IDCASE10);
               WSMSG = 'Pallet does not exist in IDCASE10';
               Flg_Error = 'Y';
 LV          LeaveSr;
 1e    Endif; 

  All ambiguity removed, very clean, and things line up easily.  There are other techniques you can use to tighten this up as well, we often use a data structure over the display that can move many fields into other data structures like the key data structure with the EVAL-CORR op-code, as long as names line up.

------------------------------
Mike Overlander
------------------------------