Re: How can I insert multiple rows in a single SQL statement
Date: Tue, 23 Jul 2002 01:33:53 +0000 (UTC)
Message-ID: <28fbfb9469fc21e376305cb9e1b66aae.54296_at_mygate.mailgate.org>
Hi,
Have you considered of bulk data copy/load directly to the database?
If you are using a commercial DBMS (like Oracle, Sybase or Informix on
the Unix), as I know all of them support bulk copy/load operations, and
C/C++ applications can utilize them using either DBMS' native API or
ODBC API.
Bulk copy/load architecture is specially intented to provide the best
perofrmance for massive data loads.
Also, if you want, bulk operations can be performed without "marking"
the transaction/change log, ie. without opening a transaction.
Hope this helps...
Alex Petrov
Sys analyst,mcdba,ocp
~~~~~~~~~ ~~~~~~~~~ ~~~~~~~~~ ~~~~~~~~~ ~~~~~~~~~
From: Ryan Hennessy <ryan.hennessy_at_alcatel.com>
Subject: How can I insert multiple rows in a single SQL statement
Date: Mon, 22 Jul 2002 10:52:52 -0400
Organization: Alcatel Canada
Good morning,
I'm having trouble coming up with a way to insert multiple rows worth of
arbitrary values into a table in a single SQL query. The values are
I am disappointed that insert into will only take a single row of values, and that similarly, a stored procedure that inserts a row of values can only be called once. I desperately need to reduce the number of transactions to one. Somebody has made the suggestion that I could select my values into a temporary table, and then use insert on that, but I am unsure if this is the best method available to me. I can't make a stored procedure to insert multiple rows at once, because I can't make the stored procedure take a variadic number of arguments, and the values are too random for simple loops to work.
Can somebody think of anything clever?
Thanks,
Ryan.
-- Posted via Mailgate.ORG Server - http://www.Mailgate.ORGReceived on Tue Jul 23 2002 - 03:33:53 CEST