I would like to have a fairly generic function that takes an INSERT, UPDATE, or DELETE, which might contain a RETURNING clause, and return that result. A fairly contrived example might be something like this:
CREATE FUNCTION my_insert(sql_insert TEXT) RETURNS record AS $$
DECLARE
result record;
BEGIN
EXECUTE sql_insert INTO result;
RETURN result;
END;
$$ LANGUAGE plpgsql;
SELECT my_insert('INSERT INTO foo VALUES (1) RETURNING some_column, another_column');
While this works ok, I don't think that record is the right type here, because RETURNING typically returns a table type. I'd like to return the exact same format from my function as RETURNING does, but I'm not sure how.
Returning record is fine, as long as the SELECT from the function defines the table structure:
SELECT * FROM my_insert('INSERT INTO foo VALUES (1) RETURNING some_column, another_column)
AS f(some_column bigint, another_column timestamp);
If the return value from the my_insert function is always going to be of a certain type, a composite type is maybe a better way to go:
CREATE TYPE my_insert_result AS (
some_column bigint,
another_column timestamp
);
And then change the function to return that type:
CREATE FUNCTION my_insert(sql_insert TEXT) RETURNS my_insert_result AS $$
DECLARE
result my_insert_result;
BEGIN
EXECUTE sql_insert INTO result;
RETURN result;
END;
$$ LANGUAGE plpgsql;
SELECT * FROM my_insert('INSERT INTO foo VALUES (1) RETURNING some_column, another_column');
Related
I have a PL/PgSQL function like this:
CREATE FUNCTION get_value(firstval integer) RETURNS SETOF mytable AS
$func$
DECLARE
current mytable;
BEGIN
SELECT fv FROM mytable WHERE fv = fistval INTO current;
IF current IS NULL THEN
INSERT INTO mytable(fv) VALUES (firstval);
END IF;
RETURN current.fv;
END
$func$ LANGUAGE plpgsql;
I need to return a specific row (fv) of the current variable, but this code is not working (it does not return anything), so how I have to do this?
There are lot of issues:
current is declared like record. But you try to assign integer value to this composite variable SELECT INTO.
The return from table functions should be realised by RETURN TABLE statement. There are not any table with name mytable. In Postgres concept - RETURNS SETOF mytable means - returns rows of type like table "mytable".
The test IS NULL is not safe in this case. You want to check FOUND variable.
Maybe you don't want to return table - then there is badly used SETOF clause.
CREATE OR REPLACE FUNCTION get_value(firstval int)
RETURNS mytable AS $$
DECLARE
_r mytable;
BEGIN
SELECT * FROM mytable WHERE fv = firstval INTO _r;
IF NOT FOUND THEN
INSERT INTO mytable(fv) VALUES(firstval);
SELECT * FROM mytable WHERE fv = firstval INTO _r;
END IF;
RETURN _r;
END;
$$ LANGUAGE plpgsql;
Attention! - this code maybe does what you want, but it is not safe against race conditions. Better to use Postgres statement INSERT ON CONFLICT DO.
Please, try to read documentation first - it is designed differently than you are expecting.
I've got a pretty simple function defined like so:
CREATE OR REPLACE FUNCTION create_new_order(....) RETURNS integer AS
$BODY$
BEGIN
PERFORM add_points_to_usage(client_id_p, date_in_p, total_points_p);
INSERT INTO orders (...) VALUES (...)
RETURNING ident;
END;
$BODY$ LANGUAGE plpgsql;
Where I'm struggling is how to actually RETURN the value stored in the ident field back via the RETURNING clause. I've tried setting the value to a variable but that either doesn't work or I'm just messing up the syntax.
You're missing the variable declaration, the INTO clause and the final RETURN:
CREATE OR REPLACE FUNCTION create_new_order(....) RETURNS integer AS
$BODY$
DECLARE
var_ident int;
BEGIN
PERFORM add_points_to_usage(client_id_p, date_in_p, total_points_p);
INSERT INTO orders (...) VALUES (...)
RETURNING ident INTO var_ident;
RETURN var_ident;
END;
$BODY$ LANGUAGE plpgsql;
I'm using PostgreSQL 9.2.4.
postgres=# select version();
version
-------------------------------------------------------------
PostgreSQL 9.2.4, compiled by Visual C++ build 1600, 64-bit
(1 row)
sqlfiddle link
My Query executes the insertion safely. What i need is that my function should return something except the void datatype. Something like text("inserted into table") or integer(0-false,1-true) , it will be useful for me to validate whether it is inserted or not?
I need a syntax for a function that returns an integer or a text when an insertion is done. For validation purpose. Is there any way to solve this?
What you probably need
Most likely you need one function to return text and another one to return integer or a function that returns boolean to indicate success. All of this is trivial and I'll refer you to the excellent manual on CREATE FUNCTION or code examples in similar questions on SO.
What you actually asked
How to write a function that returns text or integer values?
... in the sense that we have one return type being either text or integer. Not as trivial, but also not impossible as has been suggested. The key word is: polymorphic types.
Building on this simple table:
CREATE TABLE tbl(
tbl_id int,
txt text,
nr int
);
This function returns either integer or text (or any other type if you allow it), depending on the input type.
CREATE FUNCTION f_insert_data(_id int, _data anyelement, OUT _result anyelement)
RETURNS anyelement AS
$func$
BEGIN
CASE pg_typeof(_data)
WHEN 'text'::regtype THEN
INSERT INTO tbl(tbl_id, txt) VALUES(_id, _data)
RETURNING txt
INTO _result;
WHEN 'integer'::regtype THEN
INSERT INTO tbl(tbl_id, nr) VALUES(_id, _data)
RETURNING nr
INTO _result;
ELSE
RAISE EXCEPTION 'Unexpected data type: %', pg_typeof(_data)::text;
END CASE;
END
$func$
LANGUAGE plpgsql;
Call:
SELECT f_insert_data(1, 'foo'::text); -- explicit cast needed.
SELECT f_insert_data(1, 7);
Simple case
One function that returns TRUE / FALSE to indicate whether a row has been inserted, only one input parameter of varying type:
CREATE FUNCTION f_insert_data2(_id int, _data anyelement)
RETURNS boolean AS
$func$
BEGIN
CASE pg_typeof(_data)
WHEN 'text'::regtype THEN
INSERT INTO tbl(tbl_id, txt) VALUES(_id, _data);
WHEN 'integer'::regtype THEN
INSERT INTO tbl(tbl_id, nr) VALUES(_id, _data);
ELSE
RAISE EXCEPTION 'Unexpected data type: >>%<<', pg_typeof(_data)::text;
END CASE;
IF FOUND THEN RETURN TRUE;
ELSE RETURN FALSE;
END IF;
END
$func$
LANGUAGE plpgsql;
The input type can be replaced with a text parameter for most purposes, which can be cast to and from any other type.
It sounds like you're solving a problem by creating a bigger problem.
You don't need a function for this at all. Do it on the client side by checking the affected rows count that's returned by every DML query, or use INSERT ... RETURNING.
You didn't mention your client language, so here's how to do it in Python with psycopg2. The same approach applies in other languages with syntax variations.
#!/usr/bin/env python
import psycopg2
# Connect to the db
conn = psycopg2.connect("dbname=regress")
curs = conn.cursor()
# Set up the table to use
curs.execute("""
DROP TABLE IF EXISTS so17587735;
CREATE TABLE so17587735 (
id serial primary key,
blah text not null
);
""");
# Approach 1: Do the insert and check the rowcount:
curs.execute("""
INSERT INTO so17587735(blah) VALUES ('whatever');
""");
if curs.rowcount != 1:
raise Exception("Argh, insert affected zero rows, wtf?")
print("Inserted {0} rows as expected".format(curs.rowcount))
# Approach 2: Use RETURNING
curs.execute("""
INSERT INTO so17587735(blah) VALUES ('bored') RETURNING id;
""");
returned_rows = curs.fetchall();
if len(returned_rows) != 1:
raise Exception("Got unexpected row count {0} from INSERT".format(len(returned_rows)))
print("Inserted row id is {0}".format(returned_rows[0][0]))
In the case of PL/PgSQL calling INSERT you can use the GET DIAGNOSTICS command, the FOUND variable, or RETURN QUERY EXECUTE INSERT ... RETURNING .... Using GET DIAGNOSTICS:
CREATE OR REPLACE FUNCTION blah() RETURNS void AS $$
DECLARE
inserted_rows integer;
BEGIN
INSERT INTO some_table VALUES ('whatever');
GET DIAGNOSTICS inserted_rows = ROW_COUNT;
IF inserted_rows <> 1 THEN
RAISE EXCEPTION 'Failed to insert rows; expected 1 row, got %', inserted_rows;
END IF;
END;
$$ LANGUAGE plpgsql VOLATILE;
or if you must return values and must for some reason use PL/PgSQL:
CREATE OR REPLACE FUNCTION blah() RETURNS SETOF integer AS $$
BEGIN
RETURN QUERY EXECUTE INSERT INTO some_table VALUES ('whatever') RETURNING id;
END;
$$ LANGUAGE plpgsql VOLATILE;
(assuming the key is id)
which would be the same as:
CREATE OR REPLACE FUNCTION blah() RETURNS SETOF integer AS $$
INSERT INTO some_table VALUES ('whatever') RETURNING id;
$$ LANGUAGE sql;
or just
INSERT INTO some_table VALUES ('whatever') RETURNING id;
In other words: Why wrap this in a function? It doesn't make sense. Just check the row-count client side, either with RETURNING or by using the client driver's affected-rows count for INSERT.
A function can only return one type. In your case, you could create a composite type with two fields, one integer and one text, and return that.
I have spent good amount of time trying to figure it out and I haven't been able to resolve it. So, I need your help please.
I am trying to write a PL/pgSQL function that returns multiple rows. The function I wrote is shown below. But it is not working.
CREATE OR REPLACE FUNCTION get_object_fields()
RETURNS SETOF RECORD
AS
$$
DECLARE result_record keyMetrics;
BEGIN
return QUERY SELECT department_id into result_record.visits
from fact_department_daily
where report_date='2013-06-07';
--return result_record;
END
$$ LANGUAGE plpgsql;
SELECT * FROM get_object_fields;
It is returning this error:
ERROR: RETURN cannot have a parameter in function returning set;
use RETURN NEXT at or near "QUERY"
After fixing the bugs #Pavel pointed out, also define your return type properly, or you have to provide a column definition list with every call.
This call:
SELECT * FROM get_object_fields()
... assumes that Postgres knows how to expand *. Since you are returning anonymous records, you get an exception:
ERROR: a column definition list is required for functions returning "record"
One way (of several) to fix this is with RETURNS TABLE (Postgres 8.4+):
CREATE OR REPLACE FUNCTION get_object_fields()
RETURNS TABLE (department_id int) AS
$func$
BEGIN
RETURN QUERY
SELECT department_id
FROM fact_department_daily
WHERE report_date = '2013-06-07';
END
$func$ LANGUAGE plpgsql;
Works for SQL functions just the same.
Related:
PostgreSQL: ERROR: 42601: a column definition list is required for functions returning "record"
I see more bugs:
first, a SET RETURNING FUNCTIONS call has following syntax
SELECT * FROM get_object_fields()
second - RETURN QUERY forwards query result to output directly. You cannot store this result to variable - it is not possible ever in PostgreSQL now.
BEGIN
RETURN QUERY SELECT ....; -- result is forwarded to output directly
RETURN; -- there will not be any next result, finish execution
END;
third - these simple functions is better to implement in SQL languages
CREATE OR REPLACE FUNCTION get_object_fields()
RETURNS SETOF RECORD AS $$
SELECT department_id WHERE ...
$$ LANGUAGE sql STABLE;
Here's one way
drop function if exists get_test_type();
drop type if exists test_comp;
drop type if exists test_type;
drop type if exists test_person;
create type test_type as (
foo int,
bar int
);
create type test_person as (
first_name text,
last_name text
);
create type test_comp as
(
prop_a test_type[],
prop_b test_person[]
);
create or replace function get_test_type()
returns test_comp
as $$
declare
a test_type[];
b test_person[];
x test_comp;
begin
a := array(
select row (m.message_id, m.message_id)
from message m
);
-- alternative 'strongly typed'
b := array[
row('Bob', 'Jones')::test_person,
row('Mike', 'Reid')::test_person
]::test_person[];
-- alternative 'loosely typed'
b := array[
row('Bob', 'Jones'),
row('Mike', 'Reid')
];
-- using a select
b := array (
select row ('Jake', 'Scott')
union all
select row ('Suraksha', 'Setty')
);
x := row(a, b);
return x;
end;
$$
language 'plpgsql' stable;
select * from get_test_type();
CREATE OR REPLACE FUNCTION get_object_fields()
RETURNS table (department_id integer)
AS
$$
DECLARE result_record keyMetrics;
BEGIN
return QUERY
SELECT department_id
from fact_department_daily
where report_date='2013-06-07';
--return result_record;
END;
$$ LANGUAGE plpgsql;
SELECT * FROM get_object_fields()
I can't find anything in the PostgreSQL documentation that shows how to declare a record, or row, while declaring the tuple structure at the same time. If you don't define you tuple structure you get the error "The tuple structure of a not-yet-assigned record is indeterminate".
This is what I'm doing now, which works fine, but there must be a better way to do it.
CREATE OR REPLACE FUNCTION my_func()
RETURNS TABLE (
"a" integer,
"b" varchar
) AS $$
DECLARE r record;
BEGIN
CREATE TEMP TABLE tmp_t (
"a" integer,
"b" varchar
);
-- Define the tuple structure of r by SELECTing an empty row into it.
-- Is there a more straight-forward way of doing this?
SELECT * INTO r
FROM tmp_t;
-- Now I can assign values to the record.
r.a := at.something FROM "another_table" at
WHERE at.some_id = 1;
-- A related question is - how do I return the single record 'r' from
-- this function?
-- This works:
RETURN QUERY
SELECT * FROM tmp_t;
-- But this doesn't:
RETURN r;
-- ERROR: RETURN cannot have a parameter in function returning set
END; $$ LANGUAGE plpgsql;
You are mixing the syntax for returning SETOF values with syntax for returning a single row or value.
-- A related question is - how do I return the single record 'r' from
When you declare a function with RETURNS TABLE, you have to use RETURN NEXT in the body to return a row (or scalar value). And if you want to use a record variable with that it has to match the return type. Refer to the code examples further down.
Return a single value or row
If you just want to return a single row, there is no need for a record of undefined type. #Kevin already demonstrated two ways. I'll add a simplified version with OUT parameters:
CREATE OR REPLACE FUNCTION my_func(OUT a integer, OUT b text)
AS
$func$
BEGIN
a := ...;
b := ...;
END
$func$ LANGUAGE plpgsql;
You don't even need to add RETURN; in the function body, the value of the declared OUT parameters will be returned automatically at the end of the function - NULL for any parameter that has not been assigned.
And you don't need to declare RETURNS RECORD because that's already clear from the OUT parameters.
Return a set of rows
If you actually want to return multiple rows (including the possibility for 0 or 1 row), you can define the return type as RETURNS ...
SETOF some_type, where some_type can be any registered scalar or composite type.
TABLE (col1 type1, col2 type2) - an ad-hoc row type definition.
SETOF record plus OUT parameters to define column names andtypes.
100% equivalent to RETURNS TABLE.
SETOF record without further definition. But then the returned rows are undefined and you need to include a column definition list with every call (see example).
The manual about the record type:
Record variables are similar to row-type variables, but they have no
predefined structure. They take on the actual row structure of the
row they are assigned during a SELECT or FOR command.
There is more, read the manual.
You can use a record variable without assigning a defined type, you can even return such undefined records:
CREATE OR REPLACE FUNCTION my_func()
RETURNS SETOF record AS
$func$
DECLARE
r record;
BEGIN
r := (1::int, 'foo'::text); RETURN NEXT r; -- works with undefined record
r := (2::int, 'bar'::text); RETURN NEXT r;
END
$func$ LANGUAGE plpgsql;
Call:
SELECT * FROM my_func() AS x(a int, b text);
But this is very unwieldy as you have to provide the column definition list with every call. It can generally be replaced with something more elegant:
If you know the type at time of function creation, declare it right away (RETURNS TABLE or friends).
CREATE OR REPLACE FUNCTION my_func()
RETURNS SETOF tbl_or_type AS
$func$
DECLARE
r tbl_or_type;
BEGIN
SELECT INTO tbl_or_type * FROM tbl WHERE id = 10;
RETURN NEXT r; -- type matches
SELECT INTO tbl_or_type * FROM tbl WHERE id = 12;
RETURN NEXT r;
-- Or simpler:
RETURN QUERY
SELECT * FROM tbl WHERE id = 14;
END
$func$ LANGUAGE plpgsql;
If you know the type at time of the function call, there are more elegant ways using polymorphic types:
Refactor a PL/pgSQL function to return the output of various SELECT queries
Your question is unclear as to what you need exactly.
There might be some way that avoids the explicit type declaration, but offhand the best I can come up with is:
CREATE TYPE my_func_return AS (
a integer,
b varchar
);
CREATE OR REPLACE FUNCTION my_func()
RETURNS my_func_return AS $$
DECLARE
r my_func_return;
BEGIN
SELECT 1, 'one' INTO r.a, r.b;
RETURN r;
END; $$ LANGUAGE plpgsql;
Oh, I almost forgot the simplest way to do this:
CREATE OR REPLACE FUNCTION my_func2(out a int, out b text)
RETURNS RECORD AS $$
BEGIN
SELECT 1, 'one' INTO a, b;
RETURN;
END; $$ LANGUAGE plpgsql;
It is much easier to use OUT parameters rather than a record. If iteratively building a set of records (a table) use RETURN NEXT. If generating from a query, use RETURN QUERY. See:
https://stackoverflow.com/a/955289/398670
and:
http://www.postgresql.org/docs/current/static/plpgsql-declarations.html
http://www.postgresql.org/docs/current/static/sql-createfunction.html
http://www.postgresonline.com/journal/archives/129-Use-of-OUT-and-INOUT-Parameters.html
Think:
CREATE OR REPLACE FUNCTION my_func(OUT a integer, OUT b varchar) RETURNS SETOF RECORD AS $$
BEGIN
-- Assign a and b, RETURN NEXT, repeat. when done, RETURN.
END;
$$ LANGUAGE 'plpgsql';