Re: %editc with leading 0s and decimal

[Hans Boldt <boldt@xxxxxxxxxx>]

Buck Calabro wrote:
> Hans said:
>
>
> > Well, if %EDITW does what you want, then
> > why not just use that?
> >
> > But if you want to avoid %EDITW, then
> > you can always try:
> > %xlate(' ':'0':%editc(num:'3'))
>
> Which requires a V5 OS/400 release.  The request didn't mention them, but
> %xlate doesn't handle negative numbers.  If, for instance the input number
> was -12.34, you would probably use the L edit code, which would result in
> 00012.34-  If you use an L edit code on a positive number like 12.34 you
> would get 00012.340, because of the trailing space where the sign would go.
> Of course you could wrap that up in a test and use edit code 3 for positive
> numbers and L for negative numbers, but why go through that?
Well, note that I specified edit code 3 in my example, which drops
the sign.  If a leading sign is still needed, some extra step would
of course be needed.  If you wanted, you could still do it all in
one single expression, such as:

signs(%int(number<0)+1)+%xlate(' ':'0':%editc(number:'3'))

where SIGNS is a 2 element character varying array initialized with
the values '' and '-'.

(OK, that needs V5R2.  In V5R1, just use your favorite method to
convert character to numeric instead of %INT.)

> Hans' comment about %editw makes sense to me.
I understand Seth's relucatance to use %EDITW.  If the length of the
database field changes, he would have to change all his edit words.

Cheers!  Hans