get multiple result set for a Procedure in PostgreSQL - postgresql

Question

My requirement is that I want to create a generic function where I can pass any other function and its params and it should return appropriate output (i.e It may be a table result or single result etc.) and it should be with in single statement there.
This Is what I have searched and tried but I don't want to run any multiple statements.
CREATE FUNCTION CustomerWithOrdersByState() RETURNS SETOF refcursor AS $$
DECLARE
ref1 refcursor; -- Declare cursor variables
ref2 refcursor;
BEGIN
OPEN ref1 FOR SELECT * FROM "table1" limit 10;
RETURN NEXT ref1;
OPEN ref2 FOR SELECT * FROM "table2" limit 10;
RETURN NEXT ref2;
END;
$$ LANGUAGE plpgsql;
==================================================================
begin;
select * from CustomerWithOrdersByState();
FETCH ALL FROM "<unnamed portal 31>";
-- FETCH ALL FROM "<unnamed portal 30>";
commit;
I am using Postgres 11.4 version..

Answer 1

I've had what I believe is a similar issue where I wanted a way to execute a script with multiple result sets in a single batch.
As pointed out above, PGAdmin4 (and many other clients I've tried) only seem to process a single command at a time, meaning that you have to select a row, execute, select the next, execute... etc.
One quick way I found which appears to work is to save the script as a single file, then execute it on the CLI via PSQL.
So, for an example, I created a file called myscript.sql as follows:
DROP TABLE IF EXISTS sampledata;
CREATE TABLE if not exists sampledata as select x,1 as c2,2 as c3, md5(random()::text) from generate_series(1,5) x;
CREATE OR REPLACE FUNCTION GET_RECORDS(ref refcursor) RETURNS REFCURSOR AS $$
BEGIN
OPEN ref FOR SELECT * FROM SAMPLEDATA; -- OPEN A CURSOR
RETURN ref; -- RETURN THE CURSOR TO THE CALLER
END;
$$ LANGUAGE PLPGSQL;
/*
In PGManage, you would need to execute this commands one at a time (ie, 4 times).
*/
BEGIN;
SELECT get_records('r1');
FETCH ALL IN "r1";
COMMIT;
I then created a bash script (runscript.sh) which allowed for easy execution of different files.
#!/bin/bash
# Can be used to execute scripts.
# Like this ./runfile.sh hello.sql
psql -U xuser -d postgres < "$1"
I set the script to be executable:
chmod a+x runscript.sh
And then execute as follows:
./runscript.sh myscript.sql
The script executes and I see the results in the CLI. I can iterate quickly on the file, save it and execute it in the shell.