◉ Setup
The examples in this article require the following test table.
DROP TABLE t1 PURGE;
CREATE TABLE t1 AS
SELECT 1 AS id
FROM dual
CONNECT BY level <= 1000000;
-- Gathering stats on a CTAS is no longer necessary in 12c,
-- provided the statement is issued by a non-SYS user.
-- EXEC DBMS_STATS.gather_table_stats(USER, 't1');
◉ Functions in the WITH Clause
The declaration section of the WITH clause can be used to define PL/SQL functions, as shown below.
WITH
FUNCTION with_function(p_id IN NUMBER) RETURN NUMBER IS
BEGIN
RETURN p_id;
END;
SELECT with_function(id)
FROM t1
WHERE rownum = 1
/
WITH_FUNCTION(ID)
-----------------
1
SQL>
Interestingly, the ";" does not seem to work as a terminator to the SQL statement when the PL/SQL declaration is included in the WITH clause. If we attempt to use it on its own, SQL*Plus waits for more text to be entered.
From a name resolution perspective, functions defined in the PL/SQL declaration section of the WITH clause take precedence over objects with the same name defined at the schema level.
◉ Procedures in the WITH Clause
We can also define procedures in the declaration section, even if they are not used.
SET SERVEROUTPUT ON
WITH
PROCEDURE with_procedure(p_id IN NUMBER) IS
BEGIN
DBMS_OUTPUT.put_line('p_id=' || p_id);
END;
SELECT id
FROM t1
WHERE rownum = 1
/
ID
----------
1
SQL>
In reality, you would only put a procedure into a WITH clause if you planned to call the procedure from a function in the declaration section.
WITH
PROCEDURE with_procedure(p_id IN NUMBER) IS
BEGIN
DBMS_OUTPUT.put_line('p_id=' || p_id);
END;
FUNCTION with_function(p_id IN NUMBER) RETURN NUMBER IS
BEGIN
with_procedure(p_id);
RETURN p_id;
END;
SELECT with_function(id)
FROM t1
WHERE rownum = 1
/
WITH_FUNCTION(ID)
-----------------
1
p_id=1
SQL>
◉ PL/SQL Support
There does not appear to be any PL/SQL support for this feature. Any attempt to use it results in compilation errors, as shown below.
BEGIN
FOR cur_rec IN (WITH
FUNCTION with_function(p_id IN NUMBER) RETURN NUMBER IS
BEGIN
RETURN p_id;
END;
SELECT with_function(id)
FROM t1
WHERE rownum = 1)
LOOP
NULL;
END LOOP;
END;
/
FUNCTION with_function(p_id IN NUMBER) RETURN NUMBER IS
*
ERROR at line 3:
ORA-06550: line 3, column 30:
PL/SQL: ORA-00905: missing keyword
ORA-06550: line 2, column 19:
PL/SQL: SQL Statement ignored
ORA-06550: line 5, column 34:
PLS-00103: Encountered the symbol ";" when expecting one of the following:
loop
SQL>
Using dynamic SQL allows you to work around this restriction.
SET SERVEROUTPUT ON
DECLARE
l_sql VARCHAR2(32767);
l_cursor SYS_REFCURSOR;
l_value NUMBER;
BEGIN
l_sql := 'WITH
FUNCTION with_function(p_id IN NUMBER) RETURN NUMBER IS
BEGIN
RETURN p_id;
END;
SELECT with_function(id)
FROM t1
WHERE rownum = 1';
OPEN l_cursor FOR l_sql;
FETCH l_cursor INTO l_value;
DBMS_OUTPUT.put_line('l_value=' || l_value);
CLOSE l_cursor;
END;
/
l_value=1
PL/SQL procedure successfully completed.
SQL>
Support for this feature using static SQL inside PL/SQL is due in a future release.
◉ Performance Advantages
The whole reason for defining the PL/SQL code inline is to improve performance. Create a regular function to use as a comparison.
CREATE OR REPLACE FUNCTION normal_function(p_id IN NUMBER) RETURN NUMBER IS
BEGIN
RETURN p_id;
END;
/
Run the following test, which measures the elapsed time and CPU usage of the query using the inline function definition.
SET SERVEROUTPUT ON
DECLARE
l_time PLS_INTEGER;
l_cpu PLS_INTEGER;
l_sql VARCHAR2(32767);
l_cursor SYS_REFCURSOR;
TYPE t_tab IS TABLE OF NUMBER;
l_tab t_tab;
BEGIN
l_time := DBMS_UTILITY.get_time;
l_cpu := DBMS_UTILITY.get_cpu_time;
l_sql := 'WITH
FUNCTION with_function(p_id IN NUMBER) RETURN NUMBER IS
BEGIN
RETURN p_id;
END;
SELECT with_function(id)
FROM t1';
OPEN l_cursor FOR l_sql;
FETCH l_cursor
BULK COLLECT INTO l_tab;
CLOSE l_cursor;
DBMS_OUTPUT.put_line('WITH_FUNCTION : ' ||
'Time=' || TO_CHAR(DBMS_UTILITY.get_time - l_time) || ' hsecs ' ||
'CPU Time=' || (DBMS_UTILITY.get_cpu_time - l_cpu) || ' hsecs ');
l_time := DBMS_UTILITY.get_time;
l_cpu := DBMS_UTILITY.get_cpu_time;
l_sql := 'SELECT normal_function(id)
FROM t1';
OPEN l_cursor FOR l_sql;
FETCH l_cursor
BULK COLLECT INTO l_tab;
CLOSE l_cursor;
DBMS_OUTPUT.put_line('NORMAL_FUNCTION: ' ||
'Time=' || TO_CHAR(DBMS_UTILITY.get_time - l_time) || ' hsecs ' ||
'CPU Time=' || (DBMS_UTILITY.get_cpu_time - l_cpu) || ' hsecs ');
END;
/
WITH_FUNCTION : Time=45 hsecs CPU Time=39 hsecs
NORMAL_FUNCTION: Time=129 hsecs CPU Time=113 hsecs
PL/SQL procedure successfully completed.
SQL>
From this we can see the inline function definition takes approximately one third the elapsed time and CPU time to complete.
◉ PRAGMA UDF
In a number of presentations prior to the official 12c release, speakers mentioned PRAGMA UDF (User Defined Function), which supposedly gives you the performance advantages of inline PL/SQL, whilst allowing you to define the PL/SQL object outside the SQL statement. The following code redefines the previous normal function to use this pragma.
CREATE OR REPLACE FUNCTION normal_function(p_id IN NUMBER) RETURN NUMBER IS
PRAGMA UDF;
BEGIN
RETURN p_id;
END;
/
Once the function is compiled, running the test from the previous section against this function produces rather interesting results.
SET SERVEROUTPUT ON
DECLARE
l_time PLS_INTEGER;
l_cpu PLS_INTEGER;
l_sql VARCHAR2(32767);
l_cursor SYS_REFCURSOR;
TYPE t_tab IS TABLE OF NUMBER;
l_tab t_tab;
BEGIN
l_time := DBMS_UTILITY.get_time;
l_cpu := DBMS_UTILITY.get_cpu_time;
l_sql := 'WITH
FUNCTION with_function(p_id IN NUMBER) RETURN NUMBER IS
BEGIN
RETURN p_id;
END;
SELECT with_function(id)
FROM t1';
OPEN l_cursor FOR l_sql;
FETCH l_cursor
BULK COLLECT INTO l_tab;
CLOSE l_cursor;
DBMS_OUTPUT.put_line('WITH_FUNCTION : ' ||
'Time=' || TO_CHAR(DBMS_UTILITY.get_time - l_time) || ' hsecs ' ||
'CPU Time=' || (DBMS_UTILITY.get_cpu_time - l_cpu) || ' hsecs ');
l_time := DBMS_UTILITY.get_time;
l_cpu := DBMS_UTILITY.get_cpu_time;
l_sql := 'SELECT normal_function(id)
FROM t1';
OPEN l_cursor FOR l_sql;
FETCH l_cursor
BULK COLLECT INTO l_tab;
CLOSE l_cursor;
DBMS_OUTPUT.put_line('NORMAL_FUNCTION: ' ||
'Time=' || TO_CHAR(DBMS_UTILITY.get_time - l_time) || ' hsecs ' ||
'CPU Time=' || (DBMS_UTILITY.get_cpu_time - l_cpu) || ' hsecs ');
END;
/
WITH_FUNCTION : Time=44 hsecs CPU Time=40 hsecs
NORMAL_FUNCTION: Time=33 hsecs CPU Time=29 hsecs
PL/SQL procedure successfully completed.
SQL>
The standalone function using PRAGMA UDF seems to consistently out perform the inline function.
I was under the impression calling a function defined with PRAGMA UDF directly from PL/SQL would fail. This does not seem to be the case.
DECLARE
l_number NUMBER;
BEGIN
l_number := normal_function(1);
END;
/
PL/SQL procedure successfully completed.
SQL>
◉ WITH_PLSQL Hint
If the query containing the PL/SQL declaration section is not the top level query, the top-level query must include the WITH_PLSQL hint. Without this hint, the statement will fail to compile, as shown by the following update statement.
UPDATE t1 a
SET a.id = (WITH
FUNCTION with_function(p_id IN NUMBER) RETURN NUMBER IS
BEGIN
RETURN p_id;
END;
SELECT with_function(a.id)
FROM dual);
/
SET a.id = (WITH
*
ERROR at line 2:
ORA-32034: unsupported use of WITH claus
SQL>
Adding the WITH_PLSQL hint allows the statement to compile and run as expected.
UPDATE /*+ WITH_PLSQL */ t1 a
SET a.id = (WITH
FUNCTION with_function(p_id IN NUMBER) RETURN NUMBER IS
BEGIN
RETURN p_id;
END;
SELECT with_function(a.id)
FROM dual);
/
1000000 rows updated.
SQL>
Remember, this is true of queries and DML.
◉ DETERMINISTIC Hint
The use of functions in the WITH clause prevent the DETERMINISTIC optimizations from taking place in the initial release.
SET TIMING ON ARRAYSIZE 15
WITH
FUNCTION slow_function(p_id IN NUMBER) RETURN NUMBER DETERMINISTIC IS
BEGIN
DBMS_LOCK.sleep(1);
RETURN p_id;
END;
SELECT slow_function(id)
FROM t1
WHERE ROWNUM <= 10;
/
SLOW_FUNCTION(ID)
-----------------
1
1
1
1
1
1
1
1
1
1
10 rows selected.
Elapsed: 00:00:10.07
SQL>
The behaviour was fixed in 18c onward. Notice below the elapsed time is down to approximately 1 second. Thanks to Mikhail L for pointing that out.
SET TIMING ON ARRAYSIZE 15
WITH
FUNCTION slow_function(p_id IN NUMBER) RETURN NUMBER DETERMINISTIC IS
BEGIN
DBMS_LOCK.sleep(1);
RETURN p_id;
END;
SELECT slow_function(id)
FROM t1
WHERE ROWNUM <= 10;
/
SLOW_FUNCTION(ID)
-----------------
1
1
1
1
1
1
1
1
1
1
10 rows selected.
Elapsed: 00:00:01.071
SQL>
◉ Scalar Subquery Caching
In the previous section we saw the negative impact of inline function definitions on the DETERMINISTIC hint optimizations. Fortunately, scalar subquery caching is not adversely affected in the same way.
SET TIMING ON
WITH
FUNCTION slow_function(p_id IN NUMBER) RETURN NUMBER DETERMINISTIC IS
BEGIN
DBMS_LOCK.sleep(1);
RETURN p_id;
END;
SELECT (SELECT slow_function(id) FROM dual)
FROM t1
WHERE ROWNUM <= 10;
/
(SELECTSLOW_FUNCTION(ID)FROMDUAL)
---------------------------------
1
1
1
1
1
1
1
1
1
1
10 rows selected.
Elapsed: 00:00:01.04
SQL>
Posts
0 comments:
Post a Comment