RE: %DEC problems

I would say hard code the length and precision since that field will more
than likely not change in size or define constants based off the bifs %LEN()
and %DECPOS()

Here is what I have done in the past;
     D Y1              C                   %LEN(VM3VMIN)    
     D Z1              C                   %DECPOS(VM3VMIN)

     C                   EVAL      VM3VMIN  =  %DEC(VMVCHRMIN:Y1:Z1)

You cannot do this during run time.  

Thank you,
Matt Tyler
WinCo Foods, LLC
mattt@xxxxxxxxxxxxxx


-----Original Message-----
From: rpg400-l-bounces@xxxxxxxxxxxx [mailto:rpg400-l-bounces@xxxxxxxxxxxx]
On Behalf Of Dan
Sent: Thursday, July 21, 2005 2:30 PM
To: RPG400-L@xxxxxxxxxxxx
Subject: %DEC problems

Trying to figure out why the following /free code is blowing raspberries at
me:

    cuPD   = %dec( %char( %date : *ISO0 ) 
                 : %len( cuPD ) 
                 : %decpos( cuPD )        );

cuPD is defined as P(8,0) in a database file defined for output.  The
compiler errors I'm getting are related to the 2nd & 3rd parameters of
the %dec BIF:

RNF0328: The second parameter for %DEC or %DECH is not valid.
  The second parameter for built-in function %DEC or %DECH must be a
numeric literal; a built-in function that has a value known at
compile-time; or a numeric named constant. In addition, the value of
the parameter must be between 1 and 30 with no decimal digits. A float
literal is not allowed. The built-in function is ignored.

RNF0329: The third parameter for %DEC or %DECH is not valid.       
  The third parameter for built-in function %DEC or %DECH must be a
numeric literal; a built-in function that has a value known at
compile-time; or a numeric named constant. The value of the parameter
must be between 0 and 30 with no decimal digits. Also, the value must
be less than or equal to the value of the second parameter specified
for this built-in function. A float literal is not allowed. The
built-in function is ignored.

I would have thought that these are "known values at compile-time"
and, therefore, would have satisfied the second condition in the error
messages.  Previous to getting these errors, I had hardcoded the
numeric literals 8 and 0 for the 2nd & 3rd parameters of the %dec BIF,
respectively, and this compiled fine.

I am still playing catch-up after my layoff.  Is this the easiest/best
way in /free to populate an 8-digit numeric with the current date?  It
seems there should be an easier way to do this.

TIA, 
Dan