RE: RPG IV release levels and complexity

[Joel Cochran <jrc@xxxxxxxxxx>]

On Tue, 2003-11-18 at 20:04, Bob cozzi wrote:
> Scott,
> Maybe IBM should drop support for the READ/WRITE/OPEN/CLOSE/UPDATE opcodes
> so that we are forced to upgrade and/or purchase SQL. <vbg>
> What I meant by fixed-format can be ugly is that any code can be ugly, but
> mixing SQL and /FREE is always ugly.
> 
> Bob Cozzi
> Cozzi Consulting
> www.rpgiv.com
> 

And as long as we're cross posting, this was the follow on to the
earlier ugly code:

=================================
A perfectly valid complaint, and one that I share, but let's acknowledge
that /free is only half the problem in this case.  Not only do I think
the /free.../end-free should be unnecessary, I also think there should
be a much less kludgy way of executing SQL than /exec sql.../end-exec. 
And as you point out, the combination is truly annoying.

And it is for this reason that I typically house all my SQL statements
in seperate procedures, so that the same mainline would look something
like this:

<psuedo-code>

d masterDS      e ds                  extname(CUSTMAST) prefix(m_)

/free
    myHostVariable = 'Select * from myLib/myFile' ;
    prepareCursor();
    fetchMaster();
    dow sqlStt <> '00000' ;
        // do stuff
       fetchMaster();
    enddo ;
    shutDown();
 /end-free

</psuedo-code>

At least from my point of view this is much more preferrable.

=====================================


Joel Cochran
http://www.rpgnext.com