Re: Pointer vs. BIF performance (was: Getting into pointers)

I would expect the pointer implementation to be a bit faster than substring operations with a variable (unless Barbara and/or wcode has really gotten into some optimization work). I would have used the pointer approach (for what that's worth) with a slight change:

charPtr = %addr(SRCDATA);
for p = 0 to %len(%trimr(SRCDATA));
if char < x'40';
char = *blank;
endif;
charPtr += 1;
endfor;

Bruce

----- Original Message ----- From: "David Gibbs" <david@xxxxxxxxxxxx>
To: "RPG programming on the IBM i / System i" <rpg400-l@xxxxxxxxxxxx>
Sent: Wednesday, August 12, 2009 9:42 AM
Subject: Pointer vs. BIF performance (was: Getting into pointers)

This question is mostly out of idle curiosity ...

I've got a simple routine that is designed to replace any 5250 control characters in a RPG source file with blank.

for p = 1 to %len(%trim(SRCDATA));
if %subst(SRCDATA:p:1) < x'40';
%subst(SRCDATA:p:1) = *blank;
endif;
endfor;

Would there be any performance benefit if I used a pointer to update the field?

D charPtr S *
D char S 1A based(charPtr)

.
.
.

for p = 0 to %len(%trim(SRCDATA));
charPtr = %addr(SRCDATA)+p;
if char < x'40';
char = *blank;
endif;
endfor;

Thanks!

david

--
IBM i on Power - For when you can't afford to be out of business
--
This is the RPG programming on the IBM i / System i (RPG400-L) mailing list
To post a message email: RPG400-L@xxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: http://lists.midrange.com/mailman/listinfo/rpg400-l
or email: RPG400-L-request@xxxxxxxxxxxx
Before posting, please take a moment to review the archives
at http://archive.midrange.com/rpg400-l.