Hi,
The following Postgres 7.1 query extracts aggregated data for an
arbitrary period within each year for sites in a table containing
30 years of temperature data.
topo=> \d longterm
Table "longterm"
Attribute | Type | Modifier
-----------+--------------+----------
site | character(5) | not null
obs | date | not null
lo | numeric(3,1) |
hi | numeric(3,1) |
topo=> select site, extract(year from obs) as year, sum((hi+lo)/2-4) as
gdd4
topo=> from temperature
topo=> where extract(doy from obs) >= 1
topo=> and extract(doy from obs) <= 5
topo=> group by site, extract(year from obs);
site | year | gdd4
-------+------+--------
01001 | 1973 | 51.7
01001 | 1974 | 39.5
01001 | 1975 | 67.9
. . .
. . .
My question is, how can this type of query be contructed to do
the same sort of thing for a period that straddles the calendar year
boundary?
--
Regards,
+----------------------+------------------------------------------+
Ray Jackson email: [EMAIL PROTECTED]
Computing Coordinator phone: +64-3-479-8768
Dept. Geography/Te Ihowhenua fax: +64-3-479-8706
Otago University postal: Box 56, Dunedin
Te Whare Wananga o Otago AOTEAROA/NEW ZEALAND
+----------------------+------------------------------------------+
---------------------------(end of broadcast)---------------------------
TIP 4: Don't 'kill -9' the postmaster
