Re: Should I replace all CHAIN, SETLL, and READs to SQL

[Charles Wilt <charles.wilt@xxxxxxxxx>]

Yes, SQL allows access via RRN, but you do not want to use it.  As
accessing via RRN forces a full table scan.

Unless you plan to add a unique key, I'd stick with RLA.

Furthermore, simply replacing RLA op-codes with the SQL equivalent is
usually a bad idea; since the program is still designed for record at a
time processing.  RPG RLA is optimized for record at a time, SQL is not.
 SQL is designed to process sets.  May not be such a big deal in an
interactive program...but for a batch program you'll end up with horrible
performance.

HTH,
Charles


On Fri, Nov 2, 2012 at 11:06 AM, John Allen <jallen@xxxxxxxxxxx> wrote:

> I have been thinking about changing all CHAINs SETLLs and
> READs in our software to SQL
>
>
>
> Our programs use these operations for various reasons such
> as:
>
> Simple SETLL to check if a value is valid
>
> Reading records for loading subfiles
>
> Chaining by RRN (used in subfile processing, the input file
> does not have a unique key so RRN is stored in Subfile then
> used for accessing the original record)
>
> Reading and chaining to several files for processing
> thousands of records and doing validations, calculations
> etc.
>
>
>
> My main concern would be:
>
> How this would affect performance
>
> Does SQL allow for accessing records by RRN.
>
>
>
> Does anyone have any thoughts on why this is a good or bad
> idea
>
>
>
> Thanks
>
>
>
> John
>
>
>
> --
> This is the RPG programming on the IBM i / System i (RPG400-L) mailing list
> To post a message email: RPG400-L@xxxxxxxxxxxx
> To subscribe, unsubscribe, or change list options,
> visit: http://lists.midrange.com/mailman/listinfo/rpg400-l
> or email: RPG400-L-request@xxxxxxxxxxxx
> Before posting, please take a moment to review the archives
> at http://archive.midrange.com/rpg400-l.