Re: what say you about subfiles?

["James W. Kilgore" <qappdsn@xxxxxxx>]

boothm@ibm.net wrote:

> I just was looking over IBM's sample subfile program in VaRPG.

<<snip>>

>
>
> One of the downsides of the technique described is that it appears to work
> on a physical file only, disregarding all logicals, even if specified.
>

Booth,

Looking for more to do ;-)

Can't speak from vaRPG experience, but this technique does work for logicals on
V3R2.

At least in my tests it works on existing logicals, but not on files opened with
OPNQRYF (don't get the proper feed back of record count).  Just a thought
though....you could do a CPYFRMQRYF to a physical, but the query lag time may
make startup too long. You might as well load the subfile if you're going to
wait.

But anyway, IMO for a file that size, where potentially all entries would be
displayed, a page at a time technique would be called for.  A "position to" 
field
would perform a DO of choice load.  Smaller files of let's say <5000 records or
whatever your particular configuration/workload would permit we would load the
whole subfile showing a flashing "working...." message during the load.

BTW, I have always thought that if you had subfile size greater than page size
and could get the machine to use up a whole display line for the word "more..."
or "bottom", it could also show "top" or page x of y. Although this is easy
enough to program yourself so long as you are willing to stick with page at a
time.

My two cents in the hat..
begin:          vcard
fn:             James Kilgore
n:              Kilgore;James
org:            Progressive Data Systems, Inc.
adr:            311 31st Ave SE;;;Puyallup;WA;98374;USA
email;internet: qappdsn@ibm.net
title:          President
tel;work:       (253) 848-2567
tel;fax:        (253) 770-1424
x-mozilla-cpt:  ;0
x-mozilla-html: TRUE
version:        2.1
end:            vcard