Re: CHAIN Versus SETLL and READ When Data Needed

[Jerry Adams <jerry@xxxxxxxxxxxxxxx>]

While I've seen those kinds of constructs used, I personally don't like 
them because they usually don't tell me anything.


I do use LEAVE, though, in subroutines that read a display file format 
over and over, such as:

DOW F3 <> *On;
   EXFMT   FORMAT;
   IF   F3 = *On;
      LEAVE;
   ENDIF;
   [stuff]
ENDDO;


Others probably have a cleaner way of doing this. 


As for the ITER, I try to avoid it, but don't always.  I may be in a DOx 
loop that has general test criteria, such as account_number = 
something.  But within that construct, I may want to test further for 
other conditions (such as Booth's red-haired girls).  This can be 
handled with IF, of course.  But ITER can work, too, to avoid deep 
nesting.  I once had to maintain a program (not here) where I identified 
53 (count 'em!) nests before I quit counting.  The chief programmer (a 
CDP) considered this a structured program.


        * Jerry C. Adams
*IBM System i5/iSeries Programmer/Analyst
B&W Wholesale Distributors, Inc.* *
voice
        615.995.7024
fax
        615.995.1201
email
        jerry@xxxxxxxxxxxxxxx <mailto:jerry@xxxxxxxxxxxxxxx>



Wes Reinhold wrote:
> There was a lot of 'bad mouthing' of iter, leave, etc. yesterday.  How
> would you get out of a DOW 0=0 without using them? 
>
> -----Original Message-----
> From: rpg400-l-bounces@xxxxxxxxxxxx
> [mailto:rpg400-l-bounces@xxxxxxxxxxxx] On Behalf Of rob@xxxxxxxxx
> Sent: Thursday, March 01, 2007 8:44 AM
> To: RPG programming on the AS400 / iSeries
> Subject: Re: CHAIN Versus SETLL and READ When Data Needed
>
> Interesting.  Haven't used procedure pointers.  So, do you think your
> coworkers would find the use of procedure pointers clearer than the use
> of ITER or LEAVE?
>
>
> Rob Berendt
> --
> Group Dekko Services, LLC
> Dept 01.073
> PO Box 2000
> Dock 108
> 6928N 400E
> Kendallville, IN 46755
> http://www.dekko.com
>
>
>
>
>
> "Chris Pando" <chris@xxxxxxxxx> 
> Sent by: rpg400-l-bounces@xxxxxxxxxxxx
> 02/28/2007 05:04 PM
> Please respond to
> RPG programming on the AS400 / iSeries <rpg400-l@xxxxxxxxxxxx>
>
>
> To
> "RPG programming on the AS400 / iSeries" <rpg400-l@xxxxxxxxxxxx>
> cc
>
> Fax to
>
> Subject
> Re: CHAIN Versus SETLL and READ When Data Needed
>
>
>
>
>
>
> On 2/28/07, rob@xxxxxxxxx <rob@xxxxxxxxx> wrote:
>   
> > Yes, actually, it is too much to expect.  I tried to follow that rule 
> >     
> but
>   
> > it got too ugly with nested if's and stuff to put logic around each
> >     
> part
>   
> > to see if it's time to exit
> > Dou KissOff;
> >  if this;
> >    //
> >  else;
> >    //
> >  EndIf;
> >  If not KissOff;
> >    if ...;
> >      //
> >    EndIf;
> >  EndIf;
> >  If not KissOff;
> >    If ...;
> >      //
> >    EndIf;
> >  EndIf;
> >  If not KissOff;
> >    If ...;
> >      //
> >    EndIf;
> >  EndIf;
> > EndDo;
> >
> >     
>
> I have to deal with similar circustances all the time - as an example, I
>
> have
> a program with ten (10) consecutive procedures to execute. If any of
> them
> fail, I want to stop processing.
>
> I create an array (@procProxy@) containing pointers to the procedures,
> each of which returns KissOff.
>
> I then create a proxy procedure (procProxy) based upon the procedure 
> pointer
> procProxy@.
>
> Then my code looks like this:
>
> $I = 0;
> DoU ( KissOff Or $I = %Elem(@procProxy@) );
>   $I = $I + 1;
>   procProxy@ = @procProxy@($I));
>   KissOff = procProxy();
> EndDo;
>
>
> The Do loop has a true invariant, and GOTOs aren't necessary.
>
>
> Chris "IMHFO" Pando
> --
> chris@xxxxxxxxx           | Every normal man must be tempted at
> www.pando.org             | imes to spit on his hands, hoist
>                           | the black flag, and begin slitting
>                           | throats. H. L. Mencken
>