I'm not very familiar with MySQL stored procedures, but am attempting to write one for the first time. In my procedure, I have 2 in parameters and one or both of them could be null. I need to create a cursor to loop over, but my cursor needs to be based on the in parameters. Where if 1 is null and the other isn't, my cursor query is different.
For example:
CREATE PROCEDURE test (IN date timestamp, IN id int(11))
BEGIN
DECLARE cur CURSOR FOR
IF timestamp IS NOT NULL THEN
IF id IS NULL THEN
SELECT columns FROM Table WHERE modified_on <= timestamp
ELSE
SELECT columns FROM Table WHERE userid = id AND modified_on <= timestamp
ELSE
/* Logic here to pull query without the timestamp and where userid matches if the id passed in is not null */
END IF
END
Could someone show me a simple example of how to achieve this?
issue
syntax error, the declare cursor statement requires to be associated with exactly one select query :
DECLARE cursor_name CURSOR FOR select_statement
also Table is a reserved keyword and needs to be escaped ( or use a different name ) in :
SELECT columns FROM Table
to fix, either create two cursors one for each scenario or embed both query paths in one select query
setup
create table `Table`
(
id integer primary key auto_increment not null,
userid integer not null,
modified_on datetime not null
);
fix
-- option one : combine queries into one
drop procedure if exists test;
delimiter $$
CREATE PROCEDURE test (IN date timestamp, IN id int(11))
BEGIN
DECLARE cur CURSOR FOR SELECT columns FROM `Table` WHERE ( id is null or userid = id ) and modified_on <= timestamp;
-- open and read from cursor..
END$$
delimiter ;
-- option two define two cursors and use conditional logic below to decide which to read from
drop procedure if exists test;
delimiter $$
CREATE PROCEDURE test (IN date timestamp, IN id int(11))
BEGIN
DECLARE cur1 CURSOR FOR SELECT columns FROM `Table` WHERE modified_on <= timestamp;
DECLARE cur2 CURSOR FOR SELECT columns FROM `Table` WHERE userid = id AND modified_on <= timestamp;
-- evaluate if/else logic here to decide which cursor to open and use..
END$$
delimiter ;
note: not sure what you're planning todo for each cursor fetch. depending on your use case, its possible you can do this without a cursor. if this is the case, dont use a cursor and keep the processing closer to the natural sql set-based processing
reference
mysql declare cursor syntax
ansi sql reserved keywords
mysql cursors
Related
I have a stored procedure like so:
$connection->query('
drop procedure if exists listing_count;
create procedure listing_count(IN parent int(11))
begin
declare count1 int(11) default 0;
declare count2 int(11) default 1;
create temporary table ids as (select id from category where id=parent);
while count1<>count2 do
set count1=(select count(id) from ids);
insert into ids(id) select id from category where id not in(select id from ids) and related in(select id from ids);
set count2=(select count(id) from ids);
end while;
(select count(*) from listing_category where category in(select id from ids));
end');
$fetch=$connection->query('select *,listing_count(id) as listing_count from category')->fetchall(pdo::FETCH_UNIQUE|pdo::FETCH_ASSOC);
I would like to use my procedure like a function. So that listing_count gets the count so that I can use it. Do I need to create a separate function? Can a procedure get my count and return it?
Turning it into a function like so:
drop function if exists listing_count;
create function listing_count(parent int(11)) returns int(11) deterministic
begin
declare count1 int(11) default 0;
declare count2 int(11) default 1;
create temporary table ids as (select id from category where id=parent);
while count1<>count2 do
set count1=(select count(id) from ids);
insert into ids(id) select id from category where id not in(select id from ids) and related in(select id from ids);
set count2=(select count(id) from ids);
end while;
return (select count(*) from listing_category where category in(select id from ids));
end
But this does not work. I am not very familiar with procedures vs functions but I assume that I can't add all the functionality into a function as I can in a procedure.
I would like to use my procedure like a function.
You Can't Do Thatâ„¢.
I suggest you convert your sp to a stored function. That's a good idea in any case because it returns a single value. The way you have it now, it returns a one-column one-row result set. If it were a function it would work easily in every context you need it. In contrast, stored procedures returning result sets are not nearly as easy to use. For example, see this. How to use Table output from stored MYSQL Procedure
Or you could write a stored function to wrap your stored procedure and return the value. In my opinion that is an inferior solution, just because it has extra complexity.
When I call this stored procedure it shows error: unknown column...
BEGIN
if (
`LastRow.Transaction`=4 and `LastRow.Xre`>1)
then
SELECT
sleep(2);
END if;
end
Please note that sleep(2) is just to demonstrate to do something if condition is true. What would be the proper way to accomplish a test based on value of a specific record? In the above example the table (actually a View) has only one row.
Q: What would be the proper way to accomplish a test based on value of a specific record?
If you mean, based on values in columns stored in one row of a table... it seems like we would need a query that references the table that retrieve the values stored in the row. And then we can have those values available in the procedure.
As an example
BEGIN
-- local procedure variables, specify appropriate datatypes
DECLARE lr_transaction BIGINT DEFAULT NULL;
DECLARE lr_xre BIGINT DEFAULT NULL;
-- retrieve values from columns into local procedure variables
SELECT `LastRow`.`Transaction`
, `LastRow`.`Xre`
INTO lr_transaction
, lr_xre
FROM `LastRow`
WHERE someconditions
ORDER BY someexpressions
LIMIT 1
;
IF ( lr_transaction = 4 AND lr_xre > 1 ) THEN
-- do something
END IF;
END$$
That's an example of how we can retrieve a row from a table, and do some check. We could also do the check with SQL and just return a boolean
BEGIN
-- local procedure variables, specify appropriate datatypes
DECLARE lb_check TINYINT(1) UNSIGNED DEFAULT 0;
-- retrieve values from columns into local procedure variables
SELECT IF(`LastRow`.`Transaction` = 4 AND `LastRow`.`Xre` > 1,1,0)
INTO lb_check
FROM `LastRow`
WHERE someconditions
ORDER BY someexpressions
LIMIT 1
;
IF ( lb_check ) THEN
-- do something
END IF;
END$$
I have a table called Contacts with a field called person_id that I have connected to a java application.
If no value is specified for person_id in the application, I want to select everything from the contacts table using a stored procedure.
The operation I want to perform is this:
Select * from Contacts where (person_id like "%")
For this I have written a stored procedure shown below:
Delimiter $$
CREATE DEFINER=`root`#`localhost` PROCEDURE `selectTest2`(In p_id int(11))
BEGIN
if p_id = null then
set p_id = "%";
end if;
select * from Contacts where (person_id like p_id);
END $$
Delimiter ;
However when I run this procedure in my sql using the following
call selectTest2(null)
The table that is returned is blank. How do I make it show all the values in the table?
The parameter p_id gets its value from a text box in the application. If the user has entered an id, I want the procedure to show only that particular record else I want it to show all records.
What have I done wrong and how do I correct it? I am aware that p_id is an int however I tried the same thing with other fields of type varchar and the table failed to return any value.
Try using case statement in where clause like below
WHERE CASE WHEN p_id IS NOT NULL THEN person_id = p_id ELSE TRUE END
Hope this should solve your problem
I wrote store procedure in mysql. Step were followed this website http://www.mysqltutorial.org/mysql-cursor/
But it doesn't work. Here is code
DELIMITER $$
USE `hr`$$
DROP PROCEDURE IF EXISTS `at_getShift`$$
CREATE DEFINER=`root`#`%` PROCEDURE `at_getShift`()
BEGIN
DECLARE finished BOOLEAN DEFAULT FALSE;
DECLARE employeeID VARCHAR(255);-- Default "";
-- declare cursor for employee email
DECLARE hrEmployee CURSOR FOR SELECT EmployeeID FROM h_employees WHERE EmployeeID IN ('100013', '100014');
-- declare NOT FOUND handler
DECLARE CONTINUE HANDLER FOR NOT FOUND SET finished = TRUE;
DROP TABLE IF EXISTS temp;
CREATE TABLE IF NOT EXISTS temp(
`Code` VARCHAR(255)
);
OPEN hrEmployee;
get_employee: LOOP
FETCH hrEmployee INTO employeeID;
INSERT INTO temp(`Code`) VALUE (employeeID);
-- If no any row, leave loop
IF finished THEN
INSERT INTO temp(`Code`) VALUE ("112");
CLOSE hrEmployee;
LEAVE get_employee;
END IF;
-- insert temp
INSERT INTO temp(`Code`) VALUE ("111");
END LOOP get_employee;
SELECT * FROM temp;
END$$
DELIMITER ;
Execute: CALL at_getShift();
Result is:
2 rows in temp table ( 1 null, 1 is 112)
Please kindly help me to resolve this trouble.
In a SQL statement in MySQL stored program, the references to procedure variables take precedence over references to columns.
That is, when an identifier in a SQL statement matches a procedure variable, the SQL statement references the procedure variable.
References that are qualified with the table name or table alias reference columns from the table, even when there is a procedure variable with the same name.
Demonstration:
CREATE TABLE emp (id INT);
INSERT INTO emp (id) VALUES (101),(102);
DELIMITER $$
CREATE PROCEDURE foo()
BEGIN
DECLARE id INT DEFAULT 3;
-- this query returns 3 for all rows in emp
-- because "id" is a reference to the procedure variable
SELECT id FROM emp WHERE id = 3;
-- this query returns no rows
-- because "id" is a reference to the procedure variable
SELECT id FROM emp WHERE id = 101;
-- this query references columns in the table because
-- references to "id" are qualified
SELECT t.id FROM emp t WHERE t.id = 101;
END$$
DELIMITER ;
CALL foo;
The first query returns value of procedure variable for all rows from emp
id
-----
3
3
second query returns no rows
id
-----
third query returns references "id" column in table:
id
-----
101
The takeaway are two "best practices":
qualify all column references in a SQL statement in a procedure
and
procedure variable names should differ from names of columns, the usual pattern is to use a distinctive prefix on variables. As a trivial example: v_id, v_name, etc.
Both of these practices make it easier for a human reader to decipher a procedure.
Distinctive naming of procedure variables does reduce the chances of collisions, but does not invalidate the "best practice" of qualifying all column references in SQL statements. Both of those serve to make the author's intent more clear to the human reader.
EDIT:
I attempted to answer the question I thought you were asking... "Why is my procedure not doing what I expect it to?".
Beyond the answer to the question you asked... the operation that your procedure appears to be performing (populating a temporary table with a set of rows) that operation could be performed much faster and more efficiently by processing the rows as a set, rather than issuing painfully inefficient individual insert statements for each row. In terms of performance, a cursor loop processing RBAR (row-by-agonizing-row) is going to eat your lunch. And your lunch box.
DELIMITER $$
CREATE PROCEDURE `at_getShift_faster`()
BEGIN
-- ignore warning message when dropping a table that does not exist
DECLARE CONTINUE HANDLER FOR 1305 BEGIN END;
DROP TABLE IF EXISTS temp;
CREATE TABLE IF NOT EXISTS temp(`Code` VARCHAR(255));
INSERT INTO temp (`Code`)
SELECT h.EmployeeID
FROM h_employees h
WHERE h.EmployeeID IN ('100013', '100014')
;
SELECT * FROM temp;
END$$
DELIMITER ;
I'm writing a stored procedure to update a table:
UPDATE st SET somedate = NOW();
The client of the SP must know the exact date and time generated by the NOW function.
There are two options:
1) the client passes an input parameter (called _now) to the SP giving it the current date and time
UPDATE st SET somedate = _now;
2) the SP returns back the NOW's output to the client into an out parameter
UPDATE st SET somedate = NOW();
SELECT somedate FROM st INTO _now;
What do you think is the best option?
Are other options possible?
varnow = now()
UPDATE st set somedate = varnow
return varnow
i would do something like this:
drop table if exists users;
create table users
(
user_id int unsigned not null auto_increment primary key,
username varchar(32) unique not null,
created_date datetime not null
)
engine=innodb;
delimiter ;
drop procedure if exists insert_user;
delimiter #
create procedure insert_user
(
in uname varchar(32)
)
proc_main:begin
declare id int unsigned default null;
declare created datetime default null;
set created = now();
insert into users (username, created_date) values (uname, created);
set id = last_insert_id();
-- use id elsewhere maybe...
select id as user_id, created as created_date;
end proc_main #
delimiter ;
call insert_user('f00');
call insert_user('bar');
select * from users;
I suspect that both approaches are wrong.
client of the SP must know the exact date and time
Why? I suspect you really men that the client must be able to identify the records affected by a transaction - but using a timestamp to do that will not be accurate. And its not just a transaction spanning more than 1 second that is the problem. Potentially two such operations may occur in the same second.
If you've got a set of records which you need to identify as belonging to some group then that must be expressed in the schema - the timestamp of the most transaction is obviously not reliable even assuming that you never have further updates on the table.
Add another column or another table and generate a surrogate key to describe the transaction.
C.