问题
I am trying to automate the process of running a PLPGSQL function for a range of dates.
Typically I have to run the following code that generates a single table per day per function call:
SELECT dhcp.singleday('2012-11-24'::date, '2012-11-25'::date);
SELECT dhcp.singleday('2012-11-25'::date, '2012-11-26'::date);
SELECT dhcp.singleday('2012-11-26'::date, '2012-11-27'::date);
SELECT dhcp.singleday('2012-11-27'::date, '2012-11-28'::date);
SELECT dhcp.singleday('2012-11-28'::date, '2012-11-29'::date);
SELECT dhcp.singleday('2012-11-29'::date, '2012-11-30'::date);
SELECT dhcp.singleday('2012-11-30'::date, '2012-12-01'::date);
SELECT dhcp.singleday('2012-12-01'::date, '2012-12-02'::date);
SELECT dhcp.singleday('2012-12-02'::date, '2012-12-03'::date);
SELECT dhcp.singleday('2012-12-03'::date, '2012-12-04'::date);
Is there a good way to automate this sort of thing with a simple loop or function for an arbitrary date range?
I am thinking it might be hard to handle the cases of going month to month so I suppose it is better assume the date range is for a single month.
回答1:
No need for functions:
select dhcp.singleday(a::date, a::date + 1)
from generate_series(
'2012-11-24'::date,
'2012-12-03',
'1 day'
) s(a)
This will work for any date range. Not only an inside month one.
回答2:
Simple plpgsql function:
CREATE OR REPLACE FUNCTION f_machine_gun_sally(date, date)
RETURNS void AS
$func$
DECLARE
d date := $1;
BEGIN
LOOP
PERFORM dhcp.singleday(d, d+1);
d := d + 1;
EXIT WHEN d > $2;
END LOOP;
END
$func$ LANGUAGE plpgsql;
- Use PERFORM when you want don't care about the return value.
- You can just add an
integerto adateto increment. Month or year boundaries are irrelevant this way.
来源:https://stackoverflow.com/questions/14413203/writing-a-function-in-sql-to-loop-through-a-date-range-in-a-udf