RE: Somewhat simplistic question about record delete

What's so simplistic, it works great with multiple format files
... and you can work with whichever one you want, not just one at a time
... certainly gets rid of indicators
... with the way you described, you still have to identify the format name
........ or specifically read that format name
... in order to delete it

It's just another way
... no better, no worse

-----Original Message-----
From: rpg400-l-bounces@xxxxxxxxxxxx [mailto:rpg400-l-bounces@xxxxxxxxxxxx]
On Behalf Of Exterior Wood
Sent: Wednesday, October 14, 2009 2:34 PM
To: 'RPG programming on the IBM i / System i'
Subject: RE: Somewhat simplistic question about record delete

You could also setup an ... InfDs(FileNmDs) for the file

Then setup the ds for it (this is a system ds)

D*-------------------------------------------------------- Data Feedback

D* Standard RPG feedback area
D* ... format name is actual even if renamed

DFileNmDs Ds

D Io_Nme_Fmt 261 270 *
Record Fmt name

Then, when you are reading the file you just test it

If Io_Nme_Fmt = 'WHATEVER'
Do your delete
EndIf

-----Original Message-----
From: rpg400-l-bounces@xxxxxxxxxxxx [mailto:rpg400-l-bounces@xxxxxxxxxxxx]
On Behalf Of sjl
Sent: Wednesday, October 14, 2009 1:42 PM
To: rpg400-l@xxxxxxxxxxxx
Subject: Re: Somewhat simplistic question about record delete

John -

I prefer to use SETLL and READE on <FormatName>, followed by DELETE on
<FormatName>.
This eliminates the need for record-identifying indicators.

- sjl



John wrote:
I have a LF with two formats. The two formats are identified by I specs
that turn on an appropriate indicator when either of the formats is read.

So, I have this code

xxky setll <filename>
xxky reade <filename>

if *IN09 = *ON
xxky delete <format1>
endif

if *IN10 = *ON
xxky delete <format2>


My simple question is: Is this how the delete is coded for a LF with
multiple formats?
I RARELY use delete, as that is usually handled by the vendor code,
but I have a mess to clean up, and don't want to make it worse.

There is a lot more to the code than the above.
I left that out as the record selection logic >appears<
(over 13000 records to delete) to be correct.
A separate file is supplying the partial key for a loop.