On Mon, Sep 28, 2009 at 10:38:05AM -0500, Dave Huber wrote:
> Using COPY is out of the question as the file is not formatted for
> that and since other operations need to occur, the file needs to be
> read sequentially anyway.
Just to expand on what Martin said; if you can generate a set of EXECUTE
commands, you can certainly generate a COPY command to insert the same
data. The advantage is a large drop in parse time for inserting larger
numbers of rows. As you're saying you want to insert 500 rows, I'd
suggest at least trying to get COPY working.
In response to
pgsql-general by date
|Next:||From: Magnus Hagander||Date: 2009-09-28 20:59:11|
|Subject: pgday.eu 2009: Schedule available and registration open|
|Previous:||From: Scott Marlowe||Date: 2009-09-28 20:48:39|
|Subject: Re: PostgreSQL reads each 8k block - no larger blocks are used - even on sequential scans|