On Tue, Jul 15, 2008 at 7:15 PM, Mark Roberts
<mailing_lists@pandapocket.com> wrote:
>
> On Tue, 2008-07-15 at 14:31 +0200, A. Kretschmer wrote:
>> am Tue, dem 15.07.2008, um 13:12:39 +0100 mailte Oliveiros Cristina folgendes:
>> > Howdy, all,
>> >
>> > I have a problem.
>> >
>> > I have a table which one of the fields is of type date.
>> >
>> > I need to obtain the totals of the other fields in a by-month basis
>> > IS there any easy way to do this using the GROUP BY or any other construct?
>>
>> ... group by extract(month from date)
>>
>>
>> Andreas
>
> It's worth noting that extract(month from timestamp) returns a month_no, and thus will not be suitable for grouping
queriesthat span years.
>
> I recommend group by date_trunc('month', <<timestamp field>>)
Both have their uses. If you're viewing the last 5 decembers versus
the last 5 novembers, then extract would be a good choice. But mostly
date_trunc is more useful.