On Sun, 2005-08-14 at 20:53 -0400, Tom Lane wrote:
> regression=# create type table_count_result as (table_name text, count bigint);
> CREATE TYPE
> regression=# create or replace function generate_table_count ()
> regression-# returns setof table_count_result as $$
> regression$# declare
> regression$# tname record;
> regression$# count table_count_result;
> regression$# begin
> regression$# for tname in select table_name from information_schema.tables
> regression$# where table_schema = 'public' loop
> regression$# for count in execute 'select ' || quote_literal(tname.table_name) || ', count(*) from ' || quote_ident(tname.table_name) loop
> regression$# return next count;
> regression$# end loop;
> regression$# end loop;
> regression$# return;
> regression$# end$$ language plpgsql;
> CREATE FUNCTION
> regression=# select * from generate_table_count();
Sweet mercy...success. Thanks...my forehead was a bit bloody from all
the head bashing going on over here...
I have much to learn about pl/pgsql programming!
Thanks again!
John
---------------------------(end of broadcast)---------------------------
TIP 9: In versions below 8.0, the planner will ignore your desire to
choose an index scan if your joining column's datatypes do not
match
