AutoHotkey Community

DELETE FROM <TABLE_NAME>
WHERE rid(<TABLE_NAME>) <= (SELECT max(rid) FROM (
						        SELECT rid(<TABLE_NAME>) AS rid
						        FROM <TABLE_NAME>
						        ORDER BY rid(<TABLE_NAME>)
						        FETCH FIRST <N> ROWS ONLY) i)

So, I was just provided this SQL today, though I'm not completely sure it's fool-proof:
[code=sql file=Script.ahk]DELETE FROM <TABLE_NAME>
WHERE rid(<TABLE_NAME>) <= (SELECT max(rid) FROM (
SELECT rid(<TABLE_NAME>) AS rid
FROM <TABLE_NAME>
ORDER BY rid(<TABLE_NAME>)
FETCH FIRST <N> ROWS ONLY) i)[/code]

hmmm im no expert in SQL, but have you tried [url=http://stackoverflow.com/]StackOverflow[/url]? they answer quickly, and im sure youll find your answer there... ;)

delete from <TABLE_NAME>
where (<KEY>) in (
        select <KEY> from <TABLE_NAME>
        where <CONDITION>
        FETCH first <N> rows only     )

Thanks for the input, but that will not satisfy my request. This is an automated process in production, so I cannot just execute a delete statement as such - since there could be millions+ of deletes - which could bog down the server. That's why I'm attempting to create a commit threshold.

The following works:
[code=SQL]delete from <TABLE_NAME>
where (<KEY>) in (
select <KEY> from <TABLE_NAME>
where <CONDITION>
FETCH first <N> rows only )[/code]
The problem with this, however, is there is a limit not only on the IN-statement, but there's also a Lock limit. Again, my goal is to create a generic SQL statement to consolidate code for a purge process on multiple tables.

DELETE FROM table_name
WHERE some_column=some_value;

[code=sql file=Untitled.txt]DELETE FROM table_name
WHERE some_column=some_value;[/code]

????

delete from
(select * from <TABLE> fetch first 2 rows only)

Anyone happen to know how to delete the first n rows from a table in Mainframe DB2? By first, I mean the way you can utilize fetch first n rows in a select statement. The order doesn't actually matter - I'm just trying to limit the result set so I can generically create a commit limit for deleting. I'd rather not find the key & try to create a condition to return a given n rows because I'd like this to be applied to several tables, and not become too complex. The following works of distributed DB2, which is quite frustrating:[code=autohotkey file=Script.ahk]delete from
(select * from <TABLE> fetch first 2 rows only)[/code]
Executing this in Mainframe DB2 gives the folloiwng:[quote]Error: SQL0104N An unexpected token "(" was found following "(" ...[/quote]