postgresql/src/test/regress/expected/create_procedure.out

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

396 lines
11 KiB
Plaintext
Raw Normal View History

CALL nonexistent(); -- error
ERROR: procedure nonexistent() does not exist
LINE 1: CALL nonexistent();
^
HINT: No procedure matches the given name and argument types. You might need to add explicit type casts.
CALL random(); -- error
ERROR: random() is not a procedure
LINE 1: CALL random();
^
HINT: To call a function, use SELECT.
Clean up duplicate table and function names in regression tests. Many of the objects we create during the regression tests are put in the public schema, so that using the same names in different regression tests creates a hazard of test failures if any two such scripts run concurrently. This patch cleans up a bunch of latent hazards of that sort, as well as two live hazards. The current situation in this regard is far worse than it was a year or two back, because practically all of the partitioning-related test cases have reused table names with enthusiasm. I despaired of cleaning up that mess within the five most-affected tests (create_table, alter_table, insert, update, inherit); fortunately those don't run concurrently. Other than partitioning problems, most of the issues boil down to using names like "foo", "bar", "tmp", etc, without thought for the fact that other test scripts might use similar names concurrently. I've made an effort to make all such names more specific. One of the live hazards was that commit 7421f4b8 caused with.sql to create a table named "test", conflicting with a similarly-named table in alter_table.sql; this was exposed in the buildfarm recently. The other one was that join.sql and transactions.sql both create tables named "foo" and "bar"; but join.sql's uses of those names date back only to December or so. Since commit 7421f4b8 was back-patched to v10, back-patch a minimal fix for that problem. The rest of this is just future-proofing. Discussion: https://postgr.es/m/4627.1521070268@sss.pgh.pa.us
2018-03-15 22:08:51 +01:00
CREATE FUNCTION cp_testfunc1(a int) RETURNS int LANGUAGE SQL AS $$ SELECT a $$;
CREATE TABLE cp_test (a int, b text);
CREATE PROCEDURE ptest1(x text)
LANGUAGE SQL
AS $$
INSERT INTO cp_test VALUES (1, x);
$$;
\df ptest1
List of functions
Schema | Name | Result data type | Argument data types | Type
--------+--------+------------------+---------------------+------
Reconsider the handling of procedure OUT parameters. Commit 2453ea142 redefined pg_proc.proargtypes to include the types of OUT parameters, for procedures only. While that had some advantages for implementing the SQL-spec behavior of DROP PROCEDURE, it was pretty disastrous from a number of other perspectives. Notably, since the primary key of pg_proc is name + proargtypes, this made it possible to have multiple procedures with identical names + input arguments and differing output argument types. That would make it impossible to call any one of the procedures by writing just NULL (or "?", or any other data-type-free notation) for the output argument(s). The change also seems likely to cause grave confusion for client applications that examine pg_proc and expect the traditional definition of proargtypes. Hence, revert the definition of proargtypes to what it was, and undo a number of complications that had been added to support that. To support the SQL-spec behavior of DROP PROCEDURE, when there are no argmode markers in the command's parameter list, we perform the lookup both ways (that is, matching against both proargtypes and proallargtypes), succeeding if we get just one unique match. In principle this could result in ambiguous-function failures that would not happen when using only one of the two rules. However, overloading of procedure names is thought to be a pretty rare usage, so this shouldn't cause many problems in practice. Postgres-specific code such as pg_dump can defend against any possibility of such failures by being careful to specify argmodes for all procedure arguments. This also fixes a few other bugs in the area of CALL statements with named parameters, and improves the documentation a little. catversion bump forced because the representation of procedures with OUT arguments changes. Discussion: https://postgr.es/m/3742981.1621533210@sss.pgh.pa.us
2021-06-10 23:11:36 +02:00
public | ptest1 | | IN x text | proc
(1 row)
SELECT pg_get_functiondef('ptest1'::regproc);
Reconsider the handling of procedure OUT parameters. Commit 2453ea142 redefined pg_proc.proargtypes to include the types of OUT parameters, for procedures only. While that had some advantages for implementing the SQL-spec behavior of DROP PROCEDURE, it was pretty disastrous from a number of other perspectives. Notably, since the primary key of pg_proc is name + proargtypes, this made it possible to have multiple procedures with identical names + input arguments and differing output argument types. That would make it impossible to call any one of the procedures by writing just NULL (or "?", or any other data-type-free notation) for the output argument(s). The change also seems likely to cause grave confusion for client applications that examine pg_proc and expect the traditional definition of proargtypes. Hence, revert the definition of proargtypes to what it was, and undo a number of complications that had been added to support that. To support the SQL-spec behavior of DROP PROCEDURE, when there are no argmode markers in the command's parameter list, we perform the lookup both ways (that is, matching against both proargtypes and proallargtypes), succeeding if we get just one unique match. In principle this could result in ambiguous-function failures that would not happen when using only one of the two rules. However, overloading of procedure names is thought to be a pretty rare usage, so this shouldn't cause many problems in practice. Postgres-specific code such as pg_dump can defend against any possibility of such failures by being careful to specify argmodes for all procedure arguments. This also fixes a few other bugs in the area of CALL statements with named parameters, and improves the documentation a little. catversion bump forced because the representation of procedures with OUT arguments changes. Discussion: https://postgr.es/m/3742981.1621533210@sss.pgh.pa.us
2021-06-10 23:11:36 +02:00
pg_get_functiondef
------------------------------------------------------
CREATE OR REPLACE PROCEDURE public.ptest1(IN x text)+
LANGUAGE sql +
AS $procedure$ +
INSERT INTO cp_test VALUES (1, x); +
$procedure$ +
(1 row)
2018-07-14 12:17:49 +02:00
-- show only normal functions
\dfn public.*test*1
List of functions
Schema | Name | Result data type | Argument data types | Type
--------+--------------+------------------+---------------------+------
public | cp_testfunc1 | integer | a integer | func
(1 row)
-- show only procedures
\dfp public.*test*1
List of functions
Schema | Name | Result data type | Argument data types | Type
--------+--------+------------------+---------------------+------
Reconsider the handling of procedure OUT parameters. Commit 2453ea142 redefined pg_proc.proargtypes to include the types of OUT parameters, for procedures only. While that had some advantages for implementing the SQL-spec behavior of DROP PROCEDURE, it was pretty disastrous from a number of other perspectives. Notably, since the primary key of pg_proc is name + proargtypes, this made it possible to have multiple procedures with identical names + input arguments and differing output argument types. That would make it impossible to call any one of the procedures by writing just NULL (or "?", or any other data-type-free notation) for the output argument(s). The change also seems likely to cause grave confusion for client applications that examine pg_proc and expect the traditional definition of proargtypes. Hence, revert the definition of proargtypes to what it was, and undo a number of complications that had been added to support that. To support the SQL-spec behavior of DROP PROCEDURE, when there are no argmode markers in the command's parameter list, we perform the lookup both ways (that is, matching against both proargtypes and proallargtypes), succeeding if we get just one unique match. In principle this could result in ambiguous-function failures that would not happen when using only one of the two rules. However, overloading of procedure names is thought to be a pretty rare usage, so this shouldn't cause many problems in practice. Postgres-specific code such as pg_dump can defend against any possibility of such failures by being careful to specify argmodes for all procedure arguments. This also fixes a few other bugs in the area of CALL statements with named parameters, and improves the documentation a little. catversion bump forced because the representation of procedures with OUT arguments changes. Discussion: https://postgr.es/m/3742981.1621533210@sss.pgh.pa.us
2021-06-10 23:11:36 +02:00
public | ptest1 | | IN x text | proc
2018-07-14 12:17:49 +02:00
(1 row)
SELECT ptest1('x'); -- error
ERROR: ptest1(unknown) is a procedure
LINE 1: SELECT ptest1('x');
^
HINT: To call a procedure, use CALL.
CALL ptest1('a'); -- ok
CALL ptest1('xy' || 'zzy'); -- ok, constant-folded arg
CALL ptest1(substring(random()::numeric(20,15)::text, 1, 1)); -- ok, volatile arg
SELECT * FROM cp_test ORDER BY b COLLATE "C";
a | b
---+-------
1 | 0
1 | a
1 | xyzzy
(3 rows)
SQL-standard function body This adds support for writing CREATE FUNCTION and CREATE PROCEDURE statements for language SQL with a function body that conforms to the SQL standard and is portable to other implementations. Instead of the PostgreSQL-specific AS $$ string literal $$ syntax, this allows writing out the SQL statements making up the body unquoted, either as a single statement: CREATE FUNCTION add(a integer, b integer) RETURNS integer LANGUAGE SQL RETURN a + b; or as a block CREATE PROCEDURE insert_data(a integer, b integer) LANGUAGE SQL BEGIN ATOMIC INSERT INTO tbl VALUES (a); INSERT INTO tbl VALUES (b); END; The function body is parsed at function definition time and stored as expression nodes in a new pg_proc column prosqlbody. So at run time, no further parsing is required. However, this form does not support polymorphic arguments, because there is no more parse analysis done at call time. Dependencies between the function and the objects it uses are fully tracked. A new RETURN statement is introduced. This can only be used inside function bodies. Internally, it is treated much like a SELECT statement. psql needs some new intelligence to keep track of function body boundaries so that it doesn't send off statements when it sees semicolons that are inside a function body. Tested-by: Jaime Casanova <jcasanov@systemguards.com.ec> Reviewed-by: Julien Rouhaud <rjuju123@gmail.com> Discussion: https://www.postgresql.org/message-id/flat/1c11f1eb-f00c-43b7-799d-2d44132c02d7@2ndquadrant.com
2021-04-07 21:30:08 +02:00
-- SQL-standard body
CREATE PROCEDURE ptest1s(x text)
LANGUAGE SQL
BEGIN ATOMIC
INSERT INTO cp_test VALUES (1, x);
END;
\df ptest1s
List of functions
Schema | Name | Result data type | Argument data types | Type
--------+---------+------------------+---------------------+------
Reconsider the handling of procedure OUT parameters. Commit 2453ea142 redefined pg_proc.proargtypes to include the types of OUT parameters, for procedures only. While that had some advantages for implementing the SQL-spec behavior of DROP PROCEDURE, it was pretty disastrous from a number of other perspectives. Notably, since the primary key of pg_proc is name + proargtypes, this made it possible to have multiple procedures with identical names + input arguments and differing output argument types. That would make it impossible to call any one of the procedures by writing just NULL (or "?", or any other data-type-free notation) for the output argument(s). The change also seems likely to cause grave confusion for client applications that examine pg_proc and expect the traditional definition of proargtypes. Hence, revert the definition of proargtypes to what it was, and undo a number of complications that had been added to support that. To support the SQL-spec behavior of DROP PROCEDURE, when there are no argmode markers in the command's parameter list, we perform the lookup both ways (that is, matching against both proargtypes and proallargtypes), succeeding if we get just one unique match. In principle this could result in ambiguous-function failures that would not happen when using only one of the two rules. However, overloading of procedure names is thought to be a pretty rare usage, so this shouldn't cause many problems in practice. Postgres-specific code such as pg_dump can defend against any possibility of such failures by being careful to specify argmodes for all procedure arguments. This also fixes a few other bugs in the area of CALL statements with named parameters, and improves the documentation a little. catversion bump forced because the representation of procedures with OUT arguments changes. Discussion: https://postgr.es/m/3742981.1621533210@sss.pgh.pa.us
2021-06-10 23:11:36 +02:00
public | ptest1s | | IN x text | proc
SQL-standard function body This adds support for writing CREATE FUNCTION and CREATE PROCEDURE statements for language SQL with a function body that conforms to the SQL standard and is portable to other implementations. Instead of the PostgreSQL-specific AS $$ string literal $$ syntax, this allows writing out the SQL statements making up the body unquoted, either as a single statement: CREATE FUNCTION add(a integer, b integer) RETURNS integer LANGUAGE SQL RETURN a + b; or as a block CREATE PROCEDURE insert_data(a integer, b integer) LANGUAGE SQL BEGIN ATOMIC INSERT INTO tbl VALUES (a); INSERT INTO tbl VALUES (b); END; The function body is parsed at function definition time and stored as expression nodes in a new pg_proc column prosqlbody. So at run time, no further parsing is required. However, this form does not support polymorphic arguments, because there is no more parse analysis done at call time. Dependencies between the function and the objects it uses are fully tracked. A new RETURN statement is introduced. This can only be used inside function bodies. Internally, it is treated much like a SELECT statement. psql needs some new intelligence to keep track of function body boundaries so that it doesn't send off statements when it sees semicolons that are inside a function body. Tested-by: Jaime Casanova <jcasanov@systemguards.com.ec> Reviewed-by: Julien Rouhaud <rjuju123@gmail.com> Discussion: https://www.postgresql.org/message-id/flat/1c11f1eb-f00c-43b7-799d-2d44132c02d7@2ndquadrant.com
2021-04-07 21:30:08 +02:00
(1 row)
SELECT pg_get_functiondef('ptest1s'::regproc);
Reconsider the handling of procedure OUT parameters. Commit 2453ea142 redefined pg_proc.proargtypes to include the types of OUT parameters, for procedures only. While that had some advantages for implementing the SQL-spec behavior of DROP PROCEDURE, it was pretty disastrous from a number of other perspectives. Notably, since the primary key of pg_proc is name + proargtypes, this made it possible to have multiple procedures with identical names + input arguments and differing output argument types. That would make it impossible to call any one of the procedures by writing just NULL (or "?", or any other data-type-free notation) for the output argument(s). The change also seems likely to cause grave confusion for client applications that examine pg_proc and expect the traditional definition of proargtypes. Hence, revert the definition of proargtypes to what it was, and undo a number of complications that had been added to support that. To support the SQL-spec behavior of DROP PROCEDURE, when there are no argmode markers in the command's parameter list, we perform the lookup both ways (that is, matching against both proargtypes and proallargtypes), succeeding if we get just one unique match. In principle this could result in ambiguous-function failures that would not happen when using only one of the two rules. However, overloading of procedure names is thought to be a pretty rare usage, so this shouldn't cause many problems in practice. Postgres-specific code such as pg_dump can defend against any possibility of such failures by being careful to specify argmodes for all procedure arguments. This also fixes a few other bugs in the area of CALL statements with named parameters, and improves the documentation a little. catversion bump forced because the representation of procedures with OUT arguments changes. Discussion: https://postgr.es/m/3742981.1621533210@sss.pgh.pa.us
2021-06-10 23:11:36 +02:00
pg_get_functiondef
-------------------------------------------------------
CREATE OR REPLACE PROCEDURE public.ptest1s(IN x text)+
LANGUAGE sql +
BEGIN ATOMIC +
INSERT INTO cp_test (a, b) +
VALUES (1, ptest1s.x); +
END +
SQL-standard function body This adds support for writing CREATE FUNCTION and CREATE PROCEDURE statements for language SQL with a function body that conforms to the SQL standard and is portable to other implementations. Instead of the PostgreSQL-specific AS $$ string literal $$ syntax, this allows writing out the SQL statements making up the body unquoted, either as a single statement: CREATE FUNCTION add(a integer, b integer) RETURNS integer LANGUAGE SQL RETURN a + b; or as a block CREATE PROCEDURE insert_data(a integer, b integer) LANGUAGE SQL BEGIN ATOMIC INSERT INTO tbl VALUES (a); INSERT INTO tbl VALUES (b); END; The function body is parsed at function definition time and stored as expression nodes in a new pg_proc column prosqlbody. So at run time, no further parsing is required. However, this form does not support polymorphic arguments, because there is no more parse analysis done at call time. Dependencies between the function and the objects it uses are fully tracked. A new RETURN statement is introduced. This can only be used inside function bodies. Internally, it is treated much like a SELECT statement. psql needs some new intelligence to keep track of function body boundaries so that it doesn't send off statements when it sees semicolons that are inside a function body. Tested-by: Jaime Casanova <jcasanov@systemguards.com.ec> Reviewed-by: Julien Rouhaud <rjuju123@gmail.com> Discussion: https://www.postgresql.org/message-id/flat/1c11f1eb-f00c-43b7-799d-2d44132c02d7@2ndquadrant.com
2021-04-07 21:30:08 +02:00
(1 row)
CALL ptest1s('b');
SELECT * FROM cp_test ORDER BY b COLLATE "C";
a | b
---+-------
1 | 0
1 | a
1 | b
1 | xyzzy
(4 rows)
-- utility functions currently not supported here
SQL-standard function body This adds support for writing CREATE FUNCTION and CREATE PROCEDURE statements for language SQL with a function body that conforms to the SQL standard and is portable to other implementations. Instead of the PostgreSQL-specific AS $$ string literal $$ syntax, this allows writing out the SQL statements making up the body unquoted, either as a single statement: CREATE FUNCTION add(a integer, b integer) RETURNS integer LANGUAGE SQL RETURN a + b; or as a block CREATE PROCEDURE insert_data(a integer, b integer) LANGUAGE SQL BEGIN ATOMIC INSERT INTO tbl VALUES (a); INSERT INTO tbl VALUES (b); END; The function body is parsed at function definition time and stored as expression nodes in a new pg_proc column prosqlbody. So at run time, no further parsing is required. However, this form does not support polymorphic arguments, because there is no more parse analysis done at call time. Dependencies between the function and the objects it uses are fully tracked. A new RETURN statement is introduced. This can only be used inside function bodies. Internally, it is treated much like a SELECT statement. psql needs some new intelligence to keep track of function body boundaries so that it doesn't send off statements when it sees semicolons that are inside a function body. Tested-by: Jaime Casanova <jcasanov@systemguards.com.ec> Reviewed-by: Julien Rouhaud <rjuju123@gmail.com> Discussion: https://www.postgresql.org/message-id/flat/1c11f1eb-f00c-43b7-799d-2d44132c02d7@2ndquadrant.com
2021-04-07 21:30:08 +02:00
CREATE PROCEDURE ptestx()
LANGUAGE SQL
BEGIN ATOMIC
CREATE TABLE x (a int);
END;
ERROR: CREATE TABLE is not yet supported in unquoted SQL function body
CREATE PROCEDURE ptest2()
LANGUAGE SQL
AS $$
SELECT 5;
$$;
CALL ptest2();
-- nested CALL
TRUNCATE cp_test;
CREATE PROCEDURE ptest3(y text)
LANGUAGE SQL
AS $$
CALL ptest1(y);
CALL ptest1($1);
$$;
CALL ptest3('b');
SELECT * FROM cp_test;
a | b
---+---
1 | b
1 | b
(2 rows)
-- output arguments
CREATE PROCEDURE ptest4a(INOUT a int, INOUT b int)
LANGUAGE SQL
AS $$
SELECT 1, 2;
$$;
CALL ptest4a(NULL, NULL);
a | b
---+---
1 | 2
(1 row)
CREATE PROCEDURE ptest4b(INOUT b int, INOUT a int)
LANGUAGE SQL
AS $$
CALL ptest4a(a, b); -- error, not supported
$$;
ERROR: calling procedures with output arguments is not supported in SQL functions
CONTEXT: SQL function "ptest4b"
-- we used to get confused by a single output argument that is composite
CREATE PROCEDURE ptest4c(INOUT comp int8_tbl)
LANGUAGE SQL
AS $$
SELECT ROW(1, 2);
$$;
CALL ptest4c(NULL);
comp
-------
(1,2)
(1 row)
DROP PROCEDURE ptest4a, ptest4c;
-- named and default parameters
CREATE OR REPLACE PROCEDURE ptest5(a int, b text, c int default 100)
LANGUAGE SQL
AS $$
INSERT INTO cp_test VALUES(a, b);
INSERT INTO cp_test VALUES(c, b);
$$;
TRUNCATE cp_test;
CALL ptest5(10, 'Hello', 20);
CALL ptest5(10, 'Hello');
CALL ptest5(10, b => 'Hello');
CALL ptest5(b => 'Hello', a => 10);
SELECT * FROM cp_test;
a | b
-----+-------
10 | Hello
20 | Hello
10 | Hello
100 | Hello
10 | Hello
100 | Hello
10 | Hello
100 | Hello
(8 rows)
-- polymorphic types
CREATE PROCEDURE ptest6(a int, b anyelement)
LANGUAGE SQL
AS $$
SELECT NULL::int;
$$;
CALL ptest6(1, 2);
-- collation assignment
CREATE PROCEDURE ptest7(a text, b text)
LANGUAGE SQL
AS $$
SELECT a = b;
$$;
CALL ptest7(least('a', 'b'), 'a');
SQL-standard function body This adds support for writing CREATE FUNCTION and CREATE PROCEDURE statements for language SQL with a function body that conforms to the SQL standard and is portable to other implementations. Instead of the PostgreSQL-specific AS $$ string literal $$ syntax, this allows writing out the SQL statements making up the body unquoted, either as a single statement: CREATE FUNCTION add(a integer, b integer) RETURNS integer LANGUAGE SQL RETURN a + b; or as a block CREATE PROCEDURE insert_data(a integer, b integer) LANGUAGE SQL BEGIN ATOMIC INSERT INTO tbl VALUES (a); INSERT INTO tbl VALUES (b); END; The function body is parsed at function definition time and stored as expression nodes in a new pg_proc column prosqlbody. So at run time, no further parsing is required. However, this form does not support polymorphic arguments, because there is no more parse analysis done at call time. Dependencies between the function and the objects it uses are fully tracked. A new RETURN statement is introduced. This can only be used inside function bodies. Internally, it is treated much like a SELECT statement. psql needs some new intelligence to keep track of function body boundaries so that it doesn't send off statements when it sees semicolons that are inside a function body. Tested-by: Jaime Casanova <jcasanov@systemguards.com.ec> Reviewed-by: Julien Rouhaud <rjuju123@gmail.com> Discussion: https://www.postgresql.org/message-id/flat/1c11f1eb-f00c-43b7-799d-2d44132c02d7@2ndquadrant.com
2021-04-07 21:30:08 +02:00
-- empty body
CREATE PROCEDURE ptest8(x text)
BEGIN ATOMIC
END;
\df ptest8
List of functions
Schema | Name | Result data type | Argument data types | Type
--------+--------+------------------+---------------------+------
Reconsider the handling of procedure OUT parameters. Commit 2453ea142 redefined pg_proc.proargtypes to include the types of OUT parameters, for procedures only. While that had some advantages for implementing the SQL-spec behavior of DROP PROCEDURE, it was pretty disastrous from a number of other perspectives. Notably, since the primary key of pg_proc is name + proargtypes, this made it possible to have multiple procedures with identical names + input arguments and differing output argument types. That would make it impossible to call any one of the procedures by writing just NULL (or "?", or any other data-type-free notation) for the output argument(s). The change also seems likely to cause grave confusion for client applications that examine pg_proc and expect the traditional definition of proargtypes. Hence, revert the definition of proargtypes to what it was, and undo a number of complications that had been added to support that. To support the SQL-spec behavior of DROP PROCEDURE, when there are no argmode markers in the command's parameter list, we perform the lookup both ways (that is, matching against both proargtypes and proallargtypes), succeeding if we get just one unique match. In principle this could result in ambiguous-function failures that would not happen when using only one of the two rules. However, overloading of procedure names is thought to be a pretty rare usage, so this shouldn't cause many problems in practice. Postgres-specific code such as pg_dump can defend against any possibility of such failures by being careful to specify argmodes for all procedure arguments. This also fixes a few other bugs in the area of CALL statements with named parameters, and improves the documentation a little. catversion bump forced because the representation of procedures with OUT arguments changes. Discussion: https://postgr.es/m/3742981.1621533210@sss.pgh.pa.us
2021-06-10 23:11:36 +02:00
public | ptest8 | | IN x text | proc
SQL-standard function body This adds support for writing CREATE FUNCTION and CREATE PROCEDURE statements for language SQL with a function body that conforms to the SQL standard and is portable to other implementations. Instead of the PostgreSQL-specific AS $$ string literal $$ syntax, this allows writing out the SQL statements making up the body unquoted, either as a single statement: CREATE FUNCTION add(a integer, b integer) RETURNS integer LANGUAGE SQL RETURN a + b; or as a block CREATE PROCEDURE insert_data(a integer, b integer) LANGUAGE SQL BEGIN ATOMIC INSERT INTO tbl VALUES (a); INSERT INTO tbl VALUES (b); END; The function body is parsed at function definition time and stored as expression nodes in a new pg_proc column prosqlbody. So at run time, no further parsing is required. However, this form does not support polymorphic arguments, because there is no more parse analysis done at call time. Dependencies between the function and the objects it uses are fully tracked. A new RETURN statement is introduced. This can only be used inside function bodies. Internally, it is treated much like a SELECT statement. psql needs some new intelligence to keep track of function body boundaries so that it doesn't send off statements when it sees semicolons that are inside a function body. Tested-by: Jaime Casanova <jcasanov@systemguards.com.ec> Reviewed-by: Julien Rouhaud <rjuju123@gmail.com> Discussion: https://www.postgresql.org/message-id/flat/1c11f1eb-f00c-43b7-799d-2d44132c02d7@2ndquadrant.com
2021-04-07 21:30:08 +02:00
(1 row)
SELECT pg_get_functiondef('ptest8'::regproc);
Reconsider the handling of procedure OUT parameters. Commit 2453ea142 redefined pg_proc.proargtypes to include the types of OUT parameters, for procedures only. While that had some advantages for implementing the SQL-spec behavior of DROP PROCEDURE, it was pretty disastrous from a number of other perspectives. Notably, since the primary key of pg_proc is name + proargtypes, this made it possible to have multiple procedures with identical names + input arguments and differing output argument types. That would make it impossible to call any one of the procedures by writing just NULL (or "?", or any other data-type-free notation) for the output argument(s). The change also seems likely to cause grave confusion for client applications that examine pg_proc and expect the traditional definition of proargtypes. Hence, revert the definition of proargtypes to what it was, and undo a number of complications that had been added to support that. To support the SQL-spec behavior of DROP PROCEDURE, when there are no argmode markers in the command's parameter list, we perform the lookup both ways (that is, matching against both proargtypes and proallargtypes), succeeding if we get just one unique match. In principle this could result in ambiguous-function failures that would not happen when using only one of the two rules. However, overloading of procedure names is thought to be a pretty rare usage, so this shouldn't cause many problems in practice. Postgres-specific code such as pg_dump can defend against any possibility of such failures by being careful to specify argmodes for all procedure arguments. This also fixes a few other bugs in the area of CALL statements with named parameters, and improves the documentation a little. catversion bump forced because the representation of procedures with OUT arguments changes. Discussion: https://postgr.es/m/3742981.1621533210@sss.pgh.pa.us
2021-06-10 23:11:36 +02:00
pg_get_functiondef
------------------------------------------------------
CREATE OR REPLACE PROCEDURE public.ptest8(IN x text)+
LANGUAGE sql +
BEGIN ATOMIC +
END +
SQL-standard function body This adds support for writing CREATE FUNCTION and CREATE PROCEDURE statements for language SQL with a function body that conforms to the SQL standard and is portable to other implementations. Instead of the PostgreSQL-specific AS $$ string literal $$ syntax, this allows writing out the SQL statements making up the body unquoted, either as a single statement: CREATE FUNCTION add(a integer, b integer) RETURNS integer LANGUAGE SQL RETURN a + b; or as a block CREATE PROCEDURE insert_data(a integer, b integer) LANGUAGE SQL BEGIN ATOMIC INSERT INTO tbl VALUES (a); INSERT INTO tbl VALUES (b); END; The function body is parsed at function definition time and stored as expression nodes in a new pg_proc column prosqlbody. So at run time, no further parsing is required. However, this form does not support polymorphic arguments, because there is no more parse analysis done at call time. Dependencies between the function and the objects it uses are fully tracked. A new RETURN statement is introduced. This can only be used inside function bodies. Internally, it is treated much like a SELECT statement. psql needs some new intelligence to keep track of function body boundaries so that it doesn't send off statements when it sees semicolons that are inside a function body. Tested-by: Jaime Casanova <jcasanov@systemguards.com.ec> Reviewed-by: Julien Rouhaud <rjuju123@gmail.com> Discussion: https://www.postgresql.org/message-id/flat/1c11f1eb-f00c-43b7-799d-2d44132c02d7@2ndquadrant.com
2021-04-07 21:30:08 +02:00
(1 row)
CALL ptest8('');
-- OUT parameters
CREATE PROCEDURE ptest9(OUT a int)
LANGUAGE SQL
AS $$
INSERT INTO cp_test VALUES (1, 'a');
SELECT 1;
$$;
Reconsider the handling of procedure OUT parameters. Commit 2453ea142 redefined pg_proc.proargtypes to include the types of OUT parameters, for procedures only. While that had some advantages for implementing the SQL-spec behavior of DROP PROCEDURE, it was pretty disastrous from a number of other perspectives. Notably, since the primary key of pg_proc is name + proargtypes, this made it possible to have multiple procedures with identical names + input arguments and differing output argument types. That would make it impossible to call any one of the procedures by writing just NULL (or "?", or any other data-type-free notation) for the output argument(s). The change also seems likely to cause grave confusion for client applications that examine pg_proc and expect the traditional definition of proargtypes. Hence, revert the definition of proargtypes to what it was, and undo a number of complications that had been added to support that. To support the SQL-spec behavior of DROP PROCEDURE, when there are no argmode markers in the command's parameter list, we perform the lookup both ways (that is, matching against both proargtypes and proallargtypes), succeeding if we get just one unique match. In principle this could result in ambiguous-function failures that would not happen when using only one of the two rules. However, overloading of procedure names is thought to be a pretty rare usage, so this shouldn't cause many problems in practice. Postgres-specific code such as pg_dump can defend against any possibility of such failures by being careful to specify argmodes for all procedure arguments. This also fixes a few other bugs in the area of CALL statements with named parameters, and improves the documentation a little. catversion bump forced because the representation of procedures with OUT arguments changes. Discussion: https://postgr.es/m/3742981.1621533210@sss.pgh.pa.us
2021-06-10 23:11:36 +02:00
-- standard way to do a call:
CALL ptest9(NULL);
a
---
1
(1 row)
Reconsider the handling of procedure OUT parameters. Commit 2453ea142 redefined pg_proc.proargtypes to include the types of OUT parameters, for procedures only. While that had some advantages for implementing the SQL-spec behavior of DROP PROCEDURE, it was pretty disastrous from a number of other perspectives. Notably, since the primary key of pg_proc is name + proargtypes, this made it possible to have multiple procedures with identical names + input arguments and differing output argument types. That would make it impossible to call any one of the procedures by writing just NULL (or "?", or any other data-type-free notation) for the output argument(s). The change also seems likely to cause grave confusion for client applications that examine pg_proc and expect the traditional definition of proargtypes. Hence, revert the definition of proargtypes to what it was, and undo a number of complications that had been added to support that. To support the SQL-spec behavior of DROP PROCEDURE, when there are no argmode markers in the command's parameter list, we perform the lookup both ways (that is, matching against both proargtypes and proallargtypes), succeeding if we get just one unique match. In principle this could result in ambiguous-function failures that would not happen when using only one of the two rules. However, overloading of procedure names is thought to be a pretty rare usage, so this shouldn't cause many problems in practice. Postgres-specific code such as pg_dump can defend against any possibility of such failures by being careful to specify argmodes for all procedure arguments. This also fixes a few other bugs in the area of CALL statements with named parameters, and improves the documentation a little. catversion bump forced because the representation of procedures with OUT arguments changes. Discussion: https://postgr.es/m/3742981.1621533210@sss.pgh.pa.us
2021-06-10 23:11:36 +02:00
-- you can write an expression, but it's not evaluated
CALL ptest9(1/0); -- no error
a
---
1
(1 row)
-- ... and it had better match the type of the parameter
CALL ptest9(1./0.); -- error
ERROR: procedure ptest9(numeric) does not exist
LINE 1: CALL ptest9(1./0.);
^
HINT: No procedure matches the given name and argument types. You might need to add explicit type casts.
-- check named-parameter matching
CREATE PROCEDURE ptest10(OUT a int, IN b int, IN c int)
LANGUAGE SQL AS $$ SELECT b - c $$;
CALL ptest10(null, 7, 4);
a
---
3
(1 row)
CALL ptest10(a => null, b => 8, c => 2);
a
---
6
(1 row)
CALL ptest10(null, 7, c => 2);
a
---
5
(1 row)
CALL ptest10(null, c => 4, b => 11);
a
---
7
(1 row)
CALL ptest10(b => 8, c => 2, a => 0);
a
---
6
(1 row)
CREATE PROCEDURE ptest11(a OUT int, VARIADIC b int[]) LANGUAGE SQL
AS $$ SELECT b[1] + b[2] $$;
CALL ptest11(null, 11, 12, 13);
a
----
23
(1 row)
-- check resolution of ambiguous DROP commands
CREATE PROCEDURE ptest10(IN a int, IN b int, IN c int)
LANGUAGE SQL AS $$ SELECT a + b - c $$;
\df ptest10
List of functions
Schema | Name | Result data type | Argument data types | Type
--------+---------+------------------+-------------------------------------------+------
public | ptest10 | | IN a integer, IN b integer, IN c integer | proc
public | ptest10 | | OUT a integer, IN b integer, IN c integer | proc
(2 rows)
drop procedure ptest10; -- fail
ERROR: procedure name "ptest10" is not unique
HINT: Specify the argument list to select the procedure unambiguously.
drop procedure ptest10(int, int, int); -- fail
ERROR: procedure name "ptest10" is not unique
begin;
drop procedure ptest10(out int, int, int);
\df ptest10
List of functions
Schema | Name | Result data type | Argument data types | Type
--------+---------+------------------+------------------------------------------+------
public | ptest10 | | IN a integer, IN b integer, IN c integer | proc
(1 row)
drop procedure ptest10(int, int, int); -- now this would work
rollback;
begin;
drop procedure ptest10(in int, int, int);
\df ptest10
List of functions
Schema | Name | Result data type | Argument data types | Type
--------+---------+------------------+-------------------------------------------+------
public | ptest10 | | OUT a integer, IN b integer, IN c integer | proc
(1 row)
drop procedure ptest10(int, int, int); -- now this would work
rollback;
-- various error cases
CALL version(); -- error: not a procedure
ERROR: version() is not a procedure
LINE 1: CALL version();
^
HINT: To call a function, use SELECT.
CALL sum(1); -- error: not a procedure
ERROR: sum(integer) is not a procedure
LINE 1: CALL sum(1);
^
HINT: To call a function, use SELECT.
CREATE PROCEDURE ptestx() LANGUAGE SQL WINDOW AS $$ INSERT INTO cp_test VALUES (1, 'a') $$;
ERROR: invalid attribute in procedure definition
LINE 1: CREATE PROCEDURE ptestx() LANGUAGE SQL WINDOW AS $$ INSERT I...
^
CREATE PROCEDURE ptestx() LANGUAGE SQL STRICT AS $$ INSERT INTO cp_test VALUES (1, 'a') $$;
ERROR: invalid attribute in procedure definition
LINE 1: CREATE PROCEDURE ptestx() LANGUAGE SQL STRICT AS $$ INSERT I...
^
Reconsider the handling of procedure OUT parameters. Commit 2453ea142 redefined pg_proc.proargtypes to include the types of OUT parameters, for procedures only. While that had some advantages for implementing the SQL-spec behavior of DROP PROCEDURE, it was pretty disastrous from a number of other perspectives. Notably, since the primary key of pg_proc is name + proargtypes, this made it possible to have multiple procedures with identical names + input arguments and differing output argument types. That would make it impossible to call any one of the procedures by writing just NULL (or "?", or any other data-type-free notation) for the output argument(s). The change also seems likely to cause grave confusion for client applications that examine pg_proc and expect the traditional definition of proargtypes. Hence, revert the definition of proargtypes to what it was, and undo a number of complications that had been added to support that. To support the SQL-spec behavior of DROP PROCEDURE, when there are no argmode markers in the command's parameter list, we perform the lookup both ways (that is, matching against both proargtypes and proallargtypes), succeeding if we get just one unique match. In principle this could result in ambiguous-function failures that would not happen when using only one of the two rules. However, overloading of procedure names is thought to be a pretty rare usage, so this shouldn't cause many problems in practice. Postgres-specific code such as pg_dump can defend against any possibility of such failures by being careful to specify argmodes for all procedure arguments. This also fixes a few other bugs in the area of CALL statements with named parameters, and improves the documentation a little. catversion bump forced because the representation of procedures with OUT arguments changes. Discussion: https://postgr.es/m/3742981.1621533210@sss.pgh.pa.us
2021-06-10 23:11:36 +02:00
CREATE PROCEDURE ptestx(a VARIADIC int[], b OUT int) LANGUAGE SQL
AS $$ SELECT a[1] $$;
ERROR: VARIADIC parameter must be the last parameter
CREATE PROCEDURE ptestx(a int DEFAULT 42, b OUT int) LANGUAGE SQL
AS $$ SELECT a $$;
ERROR: procedure OUT parameters cannot appear after one with a default value
ALTER PROCEDURE ptest1(text) STRICT;
ERROR: invalid attribute in procedure definition
LINE 1: ALTER PROCEDURE ptest1(text) STRICT;
^
ALTER FUNCTION ptest1(text) VOLATILE; -- error: not a function
ERROR: ptest1(text) is not a function
Clean up duplicate table and function names in regression tests. Many of the objects we create during the regression tests are put in the public schema, so that using the same names in different regression tests creates a hazard of test failures if any two such scripts run concurrently. This patch cleans up a bunch of latent hazards of that sort, as well as two live hazards. The current situation in this regard is far worse than it was a year or two back, because practically all of the partitioning-related test cases have reused table names with enthusiasm. I despaired of cleaning up that mess within the five most-affected tests (create_table, alter_table, insert, update, inherit); fortunately those don't run concurrently. Other than partitioning problems, most of the issues boil down to using names like "foo", "bar", "tmp", etc, without thought for the fact that other test scripts might use similar names concurrently. I've made an effort to make all such names more specific. One of the live hazards was that commit 7421f4b8 caused with.sql to create a table named "test", conflicting with a similarly-named table in alter_table.sql; this was exposed in the buildfarm recently. The other one was that join.sql and transactions.sql both create tables named "foo" and "bar"; but join.sql's uses of those names date back only to December or so. Since commit 7421f4b8 was back-patched to v10, back-patch a minimal fix for that problem. The rest of this is just future-proofing. Discussion: https://postgr.es/m/4627.1521070268@sss.pgh.pa.us
2018-03-15 22:08:51 +01:00
ALTER PROCEDURE cp_testfunc1(int) VOLATILE; -- error: not a procedure
ERROR: cp_testfunc1(integer) is not a procedure
ALTER PROCEDURE nonexistent() VOLATILE;
ERROR: procedure nonexistent() does not exist
DROP FUNCTION ptest1(text); -- error: not a function
ERROR: ptest1(text) is not a function
Clean up duplicate table and function names in regression tests. Many of the objects we create during the regression tests are put in the public schema, so that using the same names in different regression tests creates a hazard of test failures if any two such scripts run concurrently. This patch cleans up a bunch of latent hazards of that sort, as well as two live hazards. The current situation in this regard is far worse than it was a year or two back, because practically all of the partitioning-related test cases have reused table names with enthusiasm. I despaired of cleaning up that mess within the five most-affected tests (create_table, alter_table, insert, update, inherit); fortunately those don't run concurrently. Other than partitioning problems, most of the issues boil down to using names like "foo", "bar", "tmp", etc, without thought for the fact that other test scripts might use similar names concurrently. I've made an effort to make all such names more specific. One of the live hazards was that commit 7421f4b8 caused with.sql to create a table named "test", conflicting with a similarly-named table in alter_table.sql; this was exposed in the buildfarm recently. The other one was that join.sql and transactions.sql both create tables named "foo" and "bar"; but join.sql's uses of those names date back only to December or so. Since commit 7421f4b8 was back-patched to v10, back-patch a minimal fix for that problem. The rest of this is just future-proofing. Discussion: https://postgr.es/m/4627.1521070268@sss.pgh.pa.us
2018-03-15 22:08:51 +01:00
DROP PROCEDURE cp_testfunc1(int); -- error: not a procedure
ERROR: cp_testfunc1(integer) is not a procedure
DROP PROCEDURE nonexistent();
ERROR: procedure nonexistent() does not exist
-- privileges
CREATE USER regress_cp_user1;
GRANT INSERT ON cp_test TO regress_cp_user1;
REVOKE EXECUTE ON PROCEDURE ptest1(text) FROM PUBLIC;
SET ROLE regress_cp_user1;
CALL ptest1('a'); -- error
ERROR: permission denied for procedure ptest1
RESET ROLE;
GRANT EXECUTE ON PROCEDURE ptest1(text) TO regress_cp_user1;
SET ROLE regress_cp_user1;
CALL ptest1('a'); -- ok
RESET ROLE;
-- ROUTINE syntax
Clean up duplicate table and function names in regression tests. Many of the objects we create during the regression tests are put in the public schema, so that using the same names in different regression tests creates a hazard of test failures if any two such scripts run concurrently. This patch cleans up a bunch of latent hazards of that sort, as well as two live hazards. The current situation in this regard is far worse than it was a year or two back, because practically all of the partitioning-related test cases have reused table names with enthusiasm. I despaired of cleaning up that mess within the five most-affected tests (create_table, alter_table, insert, update, inherit); fortunately those don't run concurrently. Other than partitioning problems, most of the issues boil down to using names like "foo", "bar", "tmp", etc, without thought for the fact that other test scripts might use similar names concurrently. I've made an effort to make all such names more specific. One of the live hazards was that commit 7421f4b8 caused with.sql to create a table named "test", conflicting with a similarly-named table in alter_table.sql; this was exposed in the buildfarm recently. The other one was that join.sql and transactions.sql both create tables named "foo" and "bar"; but join.sql's uses of those names date back only to December or so. Since commit 7421f4b8 was back-patched to v10, back-patch a minimal fix for that problem. The rest of this is just future-proofing. Discussion: https://postgr.es/m/4627.1521070268@sss.pgh.pa.us
2018-03-15 22:08:51 +01:00
ALTER ROUTINE cp_testfunc1(int) RENAME TO cp_testfunc1a;
ALTER ROUTINE cp_testfunc1a RENAME TO cp_testfunc1;
ALTER ROUTINE ptest1(text) RENAME TO ptest1a;
ALTER ROUTINE ptest1a RENAME TO ptest1;
Clean up duplicate table and function names in regression tests. Many of the objects we create during the regression tests are put in the public schema, so that using the same names in different regression tests creates a hazard of test failures if any two such scripts run concurrently. This patch cleans up a bunch of latent hazards of that sort, as well as two live hazards. The current situation in this regard is far worse than it was a year or two back, because practically all of the partitioning-related test cases have reused table names with enthusiasm. I despaired of cleaning up that mess within the five most-affected tests (create_table, alter_table, insert, update, inherit); fortunately those don't run concurrently. Other than partitioning problems, most of the issues boil down to using names like "foo", "bar", "tmp", etc, without thought for the fact that other test scripts might use similar names concurrently. I've made an effort to make all such names more specific. One of the live hazards was that commit 7421f4b8 caused with.sql to create a table named "test", conflicting with a similarly-named table in alter_table.sql; this was exposed in the buildfarm recently. The other one was that join.sql and transactions.sql both create tables named "foo" and "bar"; but join.sql's uses of those names date back only to December or so. Since commit 7421f4b8 was back-patched to v10, back-patch a minimal fix for that problem. The rest of this is just future-proofing. Discussion: https://postgr.es/m/4627.1521070268@sss.pgh.pa.us
2018-03-15 22:08:51 +01:00
DROP ROUTINE cp_testfunc1(int);
-- cleanup
DROP PROCEDURE ptest1;
SQL-standard function body This adds support for writing CREATE FUNCTION and CREATE PROCEDURE statements for language SQL with a function body that conforms to the SQL standard and is portable to other implementations. Instead of the PostgreSQL-specific AS $$ string literal $$ syntax, this allows writing out the SQL statements making up the body unquoted, either as a single statement: CREATE FUNCTION add(a integer, b integer) RETURNS integer LANGUAGE SQL RETURN a + b; or as a block CREATE PROCEDURE insert_data(a integer, b integer) LANGUAGE SQL BEGIN ATOMIC INSERT INTO tbl VALUES (a); INSERT INTO tbl VALUES (b); END; The function body is parsed at function definition time and stored as expression nodes in a new pg_proc column prosqlbody. So at run time, no further parsing is required. However, this form does not support polymorphic arguments, because there is no more parse analysis done at call time. Dependencies between the function and the objects it uses are fully tracked. A new RETURN statement is introduced. This can only be used inside function bodies. Internally, it is treated much like a SELECT statement. psql needs some new intelligence to keep track of function body boundaries so that it doesn't send off statements when it sees semicolons that are inside a function body. Tested-by: Jaime Casanova <jcasanov@systemguards.com.ec> Reviewed-by: Julien Rouhaud <rjuju123@gmail.com> Discussion: https://www.postgresql.org/message-id/flat/1c11f1eb-f00c-43b7-799d-2d44132c02d7@2ndquadrant.com
2021-04-07 21:30:08 +02:00
DROP PROCEDURE ptest1s;
DROP PROCEDURE ptest2;
DROP TABLE cp_test;
DROP USER regress_cp_user1;