Re: INZPFM replacement ???

[Vern Hamberg <vhamberg@xxxxxxxxxxxxxxxxxxxxxxxxx>]

I tried this using SQL. INSERT INTO followed by an immediate DELETE FROM 
using RRN. This was on a file with a PRIMARY KEY (has to be unique).

I think you're right, if you write all the records, then delete them all. 
My thought was to write one, then delete it immediately, then write the 
next one, delete IT right away.

I ran into a little thing, since the file reused deleted records. If you 
cannot WRITE to a new RRN, then you may need to CHGPF first to 
REUSEDLT(*NO), then set back, if that's what you want.

Vern

At 04:01 PM 5/27/2003 -0700, you wrote:

> Vern..
>
>       The problem with writing to the deleted records to consume them is
> that unique keyed files will prevent you from doing so.
>
> Ken Slaugh (707) 795-1512 x118
> Chouinard & Myhre, Inc.
> CA/400 Certified Specialist
> iSeries Network/MSE Administrator
> http://www.cm-inc.com/
>
>
>                       Vern 
> Hamberg 
>
>                       <vhamberg@xxxxxxxxxxxxxxx        To:       Midrange 
> Systems Technical Discussion <midrange-l@xxxxxxxxxxxx>
>                       nology.com>                      cc: 
>
>                       Sent by:                         Subject:  Re: 
> INZPFM replacement ???
>                       midrange-l-bounces@xxxxxx 
>
>                       ge.com 
>
>
>
>
>
>                       05/27/2003 03:38 
> PM 
>
>                       Please respond 
> to 
>
>                       Midrange 
> Systems 
>
>                       Technical 
> Discussion 
>
>
> Probably not directly. The delete status bit is in a part of the record not
>
> available to mortals. But how about writing new records, then checking the
> INFDS for RRN just added (or however you can do this), then delete by RRN.
> The delete ought to be fast, as it is only changing the value of a single
> bit in the header of the record.
>
> Regards
>
> Vern