Re: Hash Table Probe

Greg, in regard to the meaning of a fetch loop, many RPG programmers will
read an SQL result set (cursor) one row at a time using a fetch statement
such as:

FETCH NEXT FROM cursor-name into :data-structure ...

However, I prefer to use IBM i CLI Procedures such as sqlFetchScroll(),
which you can review in the IBM i SQL CLI Reference. Actually, I wrote an
RPG wrapper procedure named csrGoto(), which streamlines the API. Note how
row_count is incremented by each fetch?

--- begin RPG code ---

clear row_count;

dow csrGoto(csr_next) and row_count < 10;

...

row_count += 1;
enddo;

--- end RPG code ---



On Wed, Jun 17, 2020 at 3:02 PM Greg Wilburn <
gwilburn@xxxxxxxxxxxxxxxxxxxxxxx> wrote:

Nathan...

I don't have that kind of control (I also don't know what you mean by a
"fetch loop"). I am an RPG programmer that has learned SQL as went along.

The application I'm using runs two separate SQL statements.

1. Select count(*) from my_sql_view where...
Then
2. Select field1, field2, field3, etc. from my_sql_view where...

The file and where clause are always the same.
The app creates a grid in the browser and displays a row count at the
bottom.
The app has filters that likely add to the SQL statement.

I'm just trying to get the "count" statement to not take 6x as long as the
select rows statement.
But I can't really change the fundamental way the software works... it
runs two separate statements.

Thx,
Greg

-----Original Message-----
From: MIDRANGE-L [mailto:midrange-l-bounces@xxxxxxxxxxxxxxxxxx] On Behalf
Of Nathan Andelin
Sent: Wednesday, June 17, 2020 2:17 PM
To: Midrange Systems Technical Discussion <midrange-l@xxxxxxxxxxxxxxxxxx>
Subject: Re: Hash Table Probe

Greg,

Are we to understand that you're running two select statements? One that
generates a result set? Another that returns a count of 20-40?

Since the count is so low, it might be quicker to count the rows yourself
during a "fetch" loop as opposed to running "select count(*) ..."



On Wed, Jun 17, 2020 at 12:06 PM Greg Wilburn <
gwilburn@xxxxxxxxxxxxxxxxxxxxxxx> wrote:

Sorry... I should have mentioned that I've already been through some of
these explanations.

It's returning a count of 20-40 (that's how many rows meet the
criteria)... The Select is using
OPTIMIZE FOR 250 ROWS (as suggested by the IBM DB group)

The SQL statements are executing within a CGI application... so I can

only

really control the source (view) and the indexes.

--
This is the Midrange Systems Technical Discussion (MIDRANGE-L) mailing list
To post a message email: MIDRANGE-L@xxxxxxxxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: https://lists.midrange.com/mailman/listinfo/midrange-l
or email: MIDRANGE-L-request@xxxxxxxxxxxxxxxxxx
Before posting, please take a moment to review the archives
at https://archive.midrange.com/midrange-l.

Please contact support@xxxxxxxxxxxxxxxxxxxx for any subscription related
questions.

Help support midrange.com by shopping at amazon.com with our affiliate
link: https://amazon.midrange.com
--
This is the Midrange Systems Technical Discussion (MIDRANGE-L) mailing list
To post a message email: MIDRANGE-L@xxxxxxxxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: https://lists.midrange.com/mailman/listinfo/midrange-l
or email: MIDRANGE-L-request@xxxxxxxxxxxxxxxxxx
Before posting, please take a moment to review the archives
at https://archive.midrange.com/midrange-l.

Please contact support@xxxxxxxxxxxxxxxxxxxx for any subscription related
questions.

Help support midrange.com by shopping at amazon.com with our affiliate
link: https://amazon.midrange.com