Re: Alphabetizing detail records to a header file

Performance sucks.  It still has to read every record in the big file, and 
with a 500:1 ratio that is a big hit. 

Back to the original plan I guess: load an array as I load one of the 
other logicals, then sort the array by name.   It'll be fast to use, and 
easy to implement. 

Thanks for the info on the join logical.  I like the idea now.  My only 
other experience with them was when we first went  to the AS/400 and 
performance was so horrible that we scrapped them and never revisited the 
idea.
_______________________
Booth Martin
boothm@earth.goddard.edu
http://www.spy.net/~booth
_______________________




Bob Larkin <blarkin@wt.net>
Sent by: owner-rpg400-l@midrange.com
12/14/1999 12:53 AM
Please respond to RPG400-L

 
        To:     RPG400-L@midrange.com
        cc: 
        Subject:        Re: Alphabetizing detail records to a header file


OOPs,
You are correct. Swap the header detail files names, and don't use JDFTVAL 
keyword. you will only have records
that exist in both files. then you should be able to specify the alpha 
field as the key. Performance should not be
bad, as the max number of records will be the number in the detail file.

Bob
boothm@earth.goddard.edu wrote:

> Did I miss something here?  When I  "Select the Alpha Name from the 
secondary file (Header) as the key" I  got
> a halt saying the key must be from the first file on the left.  If I 
move
> the huge file to the left side I am expecting huge performance problems
> although I admit I haven't tried it yet.  Any ideas what I missed?
> _______________________
> Booth Martin
> boothm@earth.goddard.edu
> http://www.spy.net/~booth
> _______________________
>
> Bob Larkin <blarkin@wt.net>
> Sent by: owner-rpg400-l@midrange.com
> 12/09/1999 12:46 AM
> Please respond to RPG400-L
>
>
>         To:     RPG400-L@midrange.com
>         cc:
>         Subject:        Re: Alphabetizing detail records to a header 
file
>
> If I understand the question, create a logical or view withthe DETAIL 
file
> as the
> primary file joining to the Header file. The join would be on the 
"arrival
> sequence Unique ID number field". Select the Alpha Name from the 
secondary
> file
> (Header) as the key. The max reads would be the total number of records 
in
> the
> detail file. It sounds like you have used the JDFTVAL keyword in your
> join, in
> which case  the join is backwards, resulting in a view with the total
> number of
> records equal to the number of records in the Header file (unless you 
use
> an
> Inner Join which would result from no JDFTVAL keyword.)
>
> The logical below should work. Let us know if it does the trick.
>
> 
|...+....1....+....2....+....3....+....4....+....5....+....6....+....7....+....8
>       A          R JOINREC                   JFILE(DETAIL HEADER)
>       A          J                           JOIN(DETAIL HEADER
>       A                                      JFLD(DetailUniqueID
> HeaderUniqueID)
>       A            UniqueID         JREF(DetailUniqueID)
>       A            NAME
>       A           ..... fields .......
>       A         K NAME

+---
| 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
+---