Re: Why you shouldn't be afraid to use VARCHAR in RPGLE

[Peter Dow <petercdow@xxxxxxxxx>]

   dcl-s boolVar IND inz;

   boolVar = (a = b ? '1': '0');

   It's your choice.  That is RPG valid code above.

I assume you mean that the DCL-S is valid RPG code, since the request 
was to have a construct like

boolVar = (a = b ? '1': '0');

in RPG.

RPG does not have ==, just =, but in the context of

boolVal = (a = b);

the (a = b) is the same as the C (a == b), correct?

In that case, unless RPG comes up with some operator that distinguishes 
between = (assignment) and == (test for logical equivalence), the 
closest we can get to the C version is

boolVal = (a = b); // the equivalent of the C expression boolVal = (a == 
b ? '1' : '0');

If the C expression was

resultString = (a == b ? 'Hello' : 'Goodbye');

RPG would require

if a = b;
    resultString = 'Hello';
else;
    resultString = 'Goodbye';
endif;

and if the C expression was

resultString = (a = b ? 'Hello' : 'Goodbye');

RPG would require

a = b;
if b <> 0;
    resultString = 'Hello';
else;
    resultString = 'Goodbye';
endif;

Meaning that the assignment of b to a is unconditional, but the 
assignment of 'Hello' or 'Goodbye' is conditioned upon the value of b 
being non-zero.

--
*Peter Dow* /
909 793-9050
petercdow@xxxxxxxxx
/
On 12/11/2025 6:56 PM, Javier Sanchez wrote:
> Simple:
> One line of code. The compiler would make the hard work with the return
> value.  It is your intellectual work that you want back.  A string?  A
> floating point? A char()? Whatever.  The single line is clear upon your
> intention.
> The == notation is "C", put that away in RPG.  Use only one = sign.
> The return value is your intellectual work. Say it is an IND:
> dcl-s boolVar IND inz;
> boolVar = (a = b ? '1': '0');
> It's your choice.  That is RPG valid code above.
>
> JS