> but a select like this takes ages (long time):
> # select * from file where id = 1921773;
>id | name
>-+
> 1921777 | icons
>
I believe the reason is this : the numeric literal is first considered an int4
becuase it falls within the range of int4 (-2147483648
On Friday 31 January 2003 14:21, Luis Magaña wrote:
> Hi,
>
> I have a question here:
>
> I have a table with this fields:
>
> month
> description
> amount
>
> now I have to write a query that retrieves the sum of the amount from
> the minimum month to the maximum month registered for each diferen
> I need to get the most recent transaction for each customer. I need only
> the transaction ID, but the entire row would be best.
>
Why don't you alter the customer table to hold the transaction ID of the most
recent transaction?
Some questions though:
Do you vacuum the database regularly?
On Monday 22 September 2003 09:41, Wei Weng wrote:
> The production server uses PostgreSQL 7.3.1 right now, so I am only
> looking at contrib/tsearch.
>
tsearch2 is compatible, and available for any version of PosgreSQL 7.3. It is
not included in the contrib directory. It is available in CVS he
On Monday 25 October 2004 05:20, Mike Rylander wrote:
> SELECT * FROM temp50 GROUP BY gc ORDER BY ora DESC;
You can not have have expressions (columns etc.) in the SELECT list that are
either not in a GROUP BY clause, or used in an aggregate function when you
use GROUP BY in the statement. By s
