Re: CL pgm and pointer variables

[Hans Boldt <boldt@xxxxxxxxxx>]

Steve Richter wrote:
> -----Original Message-----
> From: rpg400-l-admin=Zwy7GipZuJhWk0Htik3J/w@public.gmane.org [mailto:rpg400-l-admin@midrange.com]On
> Behalf Of Hans Boldt
>
> > You want data pointers? Program in MI. But in practice, over the
> > past 20 years, I can only think of one place where I've ever seen
> > data pointers put to use.
> Hi Hans,
>
> Would that one place be a web appl where the requirement is to parse form
> posted StdIn variables and copy them to an array of data pointers that
> represent variables of various types in the html page processing program?
> Possibly two arrays are passed to a common procedure. The first is an array
> of data pointers, the 2nd is an array of form field names that the stdin
> stream is to be searched for.
No. The one place I've seen data pointers used was in the
implementation of the run-time routine for handling GDDM calls in
RPG III. If it weren't for the fact that RPG III didn't support
float numeric data, allowing for conversion of all numeric parms on
GDDM calls would have been a bit of overkill.

> If this scenario is not what you have in mind, do you consider it to have
> merit?
No. You still have to validate and store each value individually
anyways. And since the compiler knows the attributes of the target
variables at compile-time anyways, there is little to be gained by
deferring the type determination to run-time.

I think I understand what you're suggesting. But since the
validation for each input value is unique anyways (generally), you
still have to handle the inputs one by one. (You *are* validating
*all* of your CGI inputs, aren't you?)

Cheers!  Hans  (2 days left to my LOA!)