From: | Jamie Tufnell <diesql(at)googlemail(dot)com> |
---|---|
To: | pgsql-sql <pgsql-sql(at)postgresql(dot)org> |
Subject: | current_date vs 'now' |
Date: | 2009-02-05 06:05:28 |
Message-ID: | b0a4f3350902042205l6ba4f2c0u2545992fbb7e9088@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-sql |
Hi,
I was doing EXPLAIN ANALYZE on a query where I compare against
current_date and noticed the following:
Filter: (date <= ('now'::text)::date)
I knew about now() but did not know about 'now' and have since learnt
of 'today', 'tomorrow', etc. Great!
So, I changed my condition to <= 'now' to see if there would be any
improvement, which changed the filter line to:
Filter: (date <= '2009-02-05'::date)
and without fail this query is 2ms quicker every time. I alternated
between the two and ran them both several times to make sure it wasn't
just a cache difference.
AFAIK current_date is standard and 'now' and 'today', etc are not...
so that's one reason to continuing using current_date. However, I
wonder why 'today' and current_date don't generate the same query
plan?
Jamie
From | Date | Subject | |
---|---|---|---|
Next Message | Tom Lane | 2009-02-05 15:34:06 | Re: current_date vs 'now' |
Previous Message | Chris | 2009-02-04 22:42:54 | Re: postgre2postgre |