RE: EXCEPT vs. UPDATE (was Re: rpg400-l-digest V1 #241)

I too felt this way for awhile. But reality is that in today's Object
Oriented environment, this should not be a concern. Recently I evaluated
a data base trigger performance and discovered that it even handled the
'dummy update' methods we've used for years to release records. By
simply not creating the output specs manually saves many hours of coding
during programming and enhancement phases.

Ken Slaugh
(707) 795-1512 Ext 118
Programmer/System Analyst
Certified Network Administrator MSE
Specialist - Client Access/400

Chouinard & Myhre, Inc.
An IBM Premier Business Partner
P.O. Box 636
Cotati, Ca 94931
http://www.cm-inc.com/


> -----Original Message-----
> From: Dan Bale [SMTP:dbale@genfast.com]
> Sent: Friday, October 01, 1999 9:22 AM
> To:   rpg400-l@midrange.com
> Subject:      EXCEPT vs. UPDATE (was Re: rpg400-l-digest V1 #241)
> 
> 
> 
> When I write an app from scratch, my rule is: NEVER modify a database
> field
> unless you intend to update it in the database, and don't modify it
> until just
> before the UPDATE.  This is done by using a subroutine that does
> nothing but
> update the database fields and then does the UPDATE.  This way, I am
> assured
> that I can use UPDATE without updating a field that was not intended
> to be
> updated.
> 
> - Dan Bale
> 
> From: Joel Fritz <JFritz@sharperimage.com>
> Subject: RE: rpg400-l-digest V1 #241
> 
> <SNIP>
> Point 2:  I didn't always do much exception output to files.  For a
> batch
> job that runs several hours, exception output can give you a
> performance
> boost that makes a difference provided you don't have the option of
> unkeyed
> sequential I/O.  For programs that run in a few minutes, or
> interactive
> programs I rarely use it.
> <SNIP>
> > From: Pete [mailto:prmoore@mail.globalnet.co.uk]
> <SNIP>
> > > 2.  Exception output has some legitimate uses.  If you're
> > updating a few
> > > fields in a large record format in a batch process on a
> > large file, you can
> > > get a performance boost by using exception output.
> > Obviously the best
> > > performance comes from using sequential input and output to
> > a new file that
> > > replaces the original (using write), but you don't always
> > have that luxury.
> > I'm not persuaded by most machine-efficiency arguments.
> > Anyway are you saying that the compiler generates less code
> > via O-specs
> > than field moves?
> <SNIP>
> 
> 
> 
> 
> 
> +---
> | This is the RPG/400 Mailing List!
> | To submit a new message, send your mail to RPG400-L@midrange.com.
> | To subscribe to this list send email to RPG400-L-SUB@midrange.com.
> | To unsubscribe from this list send email to
> RPG400-L-UNSUB@midrange.com.
> | Questions should be directed to the list owner/operator:
> david@midrange.com
> +---
+---
| This is the RPG/400 Mailing List!
| To submit a new message, send your mail to RPG400-L@midrange.com.
| To subscribe to this list send email to RPG400-L-SUB@midrange.com.
| To unsubscribe from this list send email to RPG400-L-UNSUB@midrange.com.
| Questions should be directed to the list owner/operator: david@midrange.com
+---