Re: Embedded sql how to test for no lines

I use the search-for-one-row technique too.
Though Buck's solution is far more elegant when it comes to knowing if a
row exists I find this solution more practical.
I usually use an indicator named exists in this way:

exists = *off;
exec SQL select '1' into : exists from xxx where xxx limit 1;
if (exists);


Il giorno mer 28 ago 2024 alle ore 20:47 Jonathan Ball <jonball52@xxxxxxxxx>
ha scritto:

On 8/28/2024 1:48 AM, Birgitta Hauser wrote:

SELECT DISTINCT is a grouping, i.e. costs additional performance.
Just a simple SELECT to read the first row that machtes the search

criteria

is the best solution!!!

Correct. You should not use COUNT() as an existence test. You should only
use
COUNT() if one non-zero value has a different meaning in terms of business
logic
than another non-zero value. If all you want to know is if there is at
least one
row, then do SELECT...LIMIT 1. If you get a return value, you do one
thing; if
not, you do something else.
--
This is the RPG programming on IBM i (RPG400-L) mailing list
To post a message email: RPG400-L@xxxxxxxxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: https://lists.midrange.com/mailman/listinfo/rpg400-l
or email: RPG400-L-request@xxxxxxxxxxxxxxxxxx
Before posting, please take a moment to review the archives
at https://archive.midrange.com/rpg400-l.

Please contact support@xxxxxxxxxxxxxxxxxxxx for any subscription related
questions.