Re: simple problem with string handling

On 2013/2/25 1:59 PM, CRPence wrote:

Odd that the RPG does not simply return the empty string for the result
of the %subst opcode in that example, and therefore evaluate to false
for the comparison.

I don't necessarily disagree that the behaviour you describe might have been better in the first place, but I'm wondering about other changes that would logically follow from such a change.

Consider this example:
varfld = 'abcde';
n = 4;
m = 5;

%subst(varfld : n : m)

Current rules say that 2 is the maximum value for m in this case. Should the "length" parameter actually just be interpreted as the "maximum length", and a value of 'de' be returned?

We did change the behaviour, without a syntax change, regarding the length parameter of %SUBST being zero. It used to be an error to have a length of zero, but we changed it to just return a zero-length subfield.

But returning a zero-length subfield when a 1-length subfield was specifically requested doesn't seem to me like something that can just be changed without some new syntax requesting that behaviour.

Joep's specific case could be solved by adding an extra check to the if:

if %len(path) > 0 and %subst(path :1 :1) = '/';
// absolute path
else;
// relative path
endif;