Re: substitute variable in sql?

> On Tue, Oct 29, 2002 at 11:35:42 -0500,
> "Johnson, Shaunn" <SJohnson6(at)bcbsm(dot)com> wrote:
> > --thanks for the reply
> >
> > --i just have an sql script that i launch
> > --by hand (psql -U <user> -d <database> -f ./<script>)
> >
> > --the date is a rolling date which is supposed to
> > --change every 3 months. and as variable 1 is moved
> > --up every month, so should variable 2 move up every
> > --month.

You could create a template script which contains
DATA_1 for the first parameter and DATA_2 for the second
parameter.

Then, use sed to replace DATA_1 and DATA_2
with the actual values.

--- edit a.sql :
select * from tbl where ddata between DATA_1 and DATA_2;

--- run sed :
sed -e 's/DATA_1/10-09-2002/' -e 's/DATA_2/10-12-2002/' a.sql > b.sql

--- cat b.sql :
select * from tbl where ddata between 10-09-2002 and 10-12-2002;

--- run the script :
psql -f b.sql

( I think this solution could be improved by using pipes ,
in order to avoid the temporary b.sql file ... )

Adrian Maier
(am(at)fx(dot)ro)