I am learning to write MySQL stored procedures and I have encountered some difficulties. Here I have two stored procedures:
First stored procedure:
CREATE PROCEDURE sp1 (IN `username` TEXT, OUT `user_id` INT)
BEGIN
DECLARE rowcount INT;
SELECT count(`User ID`) INTO rowcount FROM user WHERE `Username`=username;
SET user_id = rowcount;
END|
Second stored procedure:
CREATE PROCEDURE sp2 (IN `doc_id` INT, IN `content` LONGTEXT)
BEGIN
UPDATE doc SET `Content`=content WHERE `Doc ID`=doc_id;
END|
(Delimiter is |.)
Question:
I observe that the result of the first stored procedure is the same as calling SELECT count(`User ID`) FROM user;. However, the second stored procedure does its job and gets the content updated with the new content.
So why does the first stored procedure treat `Username` and username as equal identifiers but the second stored procedure treats `Content` and content as different identifiers? The two identifiers in both cases are the same except the capitalization of the first letter.
I figure it out after reading the official MySQL documentation about the scope of local variables.
It states that:
A local variable should not have the same name as a table column. If an SQL statement, such as a SELECT ... INTO statement, contains a reference to a column and a declared local variable with the same name, MySQL currently interprets the reference as the name of a variable.
Related
Is it possible to use a procedure inside a function? For example, I would like to gather all my rows related to an id but I would also like to count the rows and use it in a select statement. This is not working:
drop procedure if exists relatives;
create procedure relatives(in parent int(11),out counted int(11))
begin
set counted=(select count(*) from category where related=parent);
end;
drop function if exists relatives_count;
create function relatives_count(parent parent(11)) returns int(11)
begin
declare count int(11);
call relatives(parent,counted);
return counted;
end;
So that I can use the count
select relatives_count(id) from category
This is just for curiosity purposes. It may look senseless since I can just call a single select query and get the same results but I want to know how I can use my procedure out variable in a function.
Yes, a MySQL FUNCTION can call a MySQL PROCEDURE.
But... the operations the procedure performs will be limited to the operations allowed by a function. (We can't use a procedure to workaround the limitations placed on a function.)
"is not working" is so nebulously vague as to be practically useless in debugging the issue. What exact behavior is being observed?
My suspicion is that the SQL statements shown are failing, because there is no override for the default statement delimiter.
Also, parent(11) is not a valid datatype.
Be aware that when an identifier for a column in a SQL statement in a MySQL stored program matches an identifier used for an argument or local variable, MySQL follows a rule about which (the column name or the variable) that is being referenced.
Best practice is to adopt a naming convention for arguments and local variables that do not match column names, and to qualify all column references with a table name or table alias.
Personally, I use a prefix for arguments and local variables (a for argument, l for local, followed by a datatype i for integer, d for date/datetime, n for decimal, ...
DELIMITER $$
DROP PROCEDURE IF EXISTS relatives$$
CREATE PROCEDURE relatives(IN ai_parent INT(11),OUT ai_counted INT(11))
BEGIN
SELECT COUNT(*)
INTO ai_counted
FROM category c
WHERE c.related = ai_parent
;
END$$
DROP FUNCTION IF EXISTS relatives_count$$
CREATE FUNCTION relatives_count(ai_parent INT(11))
RETURNS INT(11)
BEGIN
DECLARE li_counted INT(11);
CALL relatives(ai_parent,li_counted);
RETURN li_counted;
END$$
DELIMITER ;
Please identify the exact behavior you observe. Error message when creating the procedure? Error message when executing the function? Unexpected behavior. That's much more precise and informative than telling us something "is not working".
I try to understand how functions work. I can make the equivalent in procedure but I can't create a simple function with select.
element is UNIQUE and
thing is PRIMARY
CREATE DEFINER=`root`#`localhost`
FUNCTION `get_element_by_thing`(`thing` VARCHAR(255))
RETURNS VARCHAR(255)
CHARSET utf8
NOT DETERMINISTIC
READS SQL DATA
SQL SECURITY DEFINER
DECLARE #return_element VARCHAR(255);
SET #return_element = (
SELECT
`element`
FROM
`table1`
WHERE
`thing` = thing
);
RETURN #return_element;
I use the phpmyadmin interface.
1) Don't declare user-defined variables.
The name of a local variable in MySQL stored program does not start with an at sign #. As an example:
DECLARE stored_program_local_variable VARCHAR(255);
SET stored_program_local_variable = 'somevalue';
The name of a user-defined variables start with an at sign #. (The at sign character is what distinguishes user-defined variables from other identifiers.) It's not valid to declare a user-defined variable in a stored program. To create a user-defined variable, just assign a value to it. For example:
SET #user_defined_variable = 'somevalue';
2) If we don't need to persist variables beyond the scope of a stored program, we typically use local variables, which exist only for the duration of the stored program execution. (Which is different behavior than user-defined variables which are at the session level.)
3) Use the SELECT ... INTO syntax to retrieve scalar values into user-defined or local variables. https://dev.mysql.com/doc/refman/5.7/en/select-into.html
Try:
DELIMITER $$
CREATE DEFINER=`root`#`localhost`
FUNCTION `get_element_by_thing`(`thing` VARCHAR(255))
RETURNS VARCHAR(255)
...
BEGIN
DECLARE return_element VARCHAR(255) ;
SELECT t1.element
INTO return_element
FROM table1 t1
WHERE t1.thing = thing
LIMIT 1 ;
RETURN return_element ;
END $$
DELIMITER ;
Note: with ambiguous identifiers (i.e. routine parameter and column with the same name in a SQL statement, the routine parameter takes precedence over the column name. Qualify the column reference with the table name or table alias so it's not ambiguous. I prefer to assign routine parameters (and local variables) names that do not match column names.
If for some reason you need to assign a value to a user-defined variable in a SQL statement, you can use the := assignment operator. This is also valid outside the context of a stored program.
SELECT #user_defined_variable := t.somecolumn
FROM mytable t
WHERE somecondition
ORDER BY someexpression
LIMIT 1
I'm trying to write a stored procedure that inserts into one table (A), then queries another table (B), then finally inserts into table (C) the last insert id, along with the result from table B. I have written a stored procedure named VetIdFromCode to do the selecting from table B, which works fine in isolation. When I run the query in isolation, subbing in value for the IN parameters then it runs fine, but when I try and save it as a stored procedure it tells me invalid SQL near 'SET #LIID...'
Many thanks for any help.
CREATE PROCEDURE `NewClientUser`(
IN `uemail` VARCHAR(60),
IN `uphash` CHAR(40),
IN `uvcode` VARCHAR(11))
DETERMINISTIC
MODIFIES SQL DATA
SQL SECURITY INVOKER
INSERT INTO users (user_id,user_email,user_hash,user_role)
VALUES (NULL,uemail,uphash,'1');
SET #LIID = LAST_INSERT_ID();
CALL `VetIdFromCode`(uvcode, #VID);
INSERT INTO user_vet_lookup(user_id,vet_id)
VALUES (#LIID,#VID);
You need to start the "code" of the procedure with the key word "BEGIN" and put an "END" at the end. Like:
CREATE PROCEDURE `NewClientUser`(
IN `uemail` VARCHAR(60),
IN `uphash` CHAR(40),
IN `uvcode` VARCHAR(11))
BEGIN
DETERMINISTIC
MODIFIES SQL DATA
SQL SECURITY INVOKER
INSERT INTO users (user_id,user_email,user_hash,user_role)
VALUES (NULL,uemail,uphash,'1');
SET #LIID = LAST_INSERT_ID();
CALL `VetIdFromCode`(uvcode, #VID);
INSERT INTO user_vet_lookup(user_id,vet_id)
VALUES (#LIID,#VID);
END
Check out the documentation: http://dev.mysql.com/doc/refman/5.7/en/create-procedure.html
I am writing a stored procedure in mysql which simply returns the row with ID provided or return all table when no ID is provided.
CREATE DEFINER=`root`#`localhost` PROCEDURE `SLICE_GET`(`slice_id` int)
BEGIN
SELECT *
FROM `thesis_db`.`SLICE_INFO`
WHERE (SLICE_ID = `slice_id` OR `slice_id` IS NULL);
END
I have used the same idea in ms-sql for years yet it doesn't seem to work for mysql since no matter which ID is passed, the procedure returns entire table.
What am I missing here ?
This is a way to write procedures in mysql
DELIMITER $$
CREATE PROCEDURE `name of procedure` (x CHAR(1), D1 DATE, D2 DATE)
BEGIN
SELECT name of columns you want to display
FROM table name
WHERE SLICE_ID= x
OR SLICE_ID IS NULL;
END
$$
Note: Moreover mysql is not case sensitive means all caps or all small will not effect it.
delimiter is used to:
If you use the mysql client program to define a stored program containing semicolon characters, a problem arises.
By default, mysql itself recognizes the semicolon as a statement delimiter, so you must redefine the delimiter temporarily to cause mysql to pass the entire stored program definition to the server.
In MySql
UPDATE `inventoryentry` SET `Status` = 1 WHERE `InventoryID`=92 AND `ItemID`=28;
It successfully update only one row , where inventoryID = 92 and itemID=28 , the following message displayed.
1 row(s) affected
when I put this on stored procedure, as follow
CREATE DEFINER=`root`#`localhost` PROCEDURE `Sample`(IN itemId INT, IN itemQnty
DOUBLE, IN invID INT)
BEGIN
DECLARE crntQnty DOUBLE;
DECLARE nwQnty DOUBLE;
SET crntQnty=(SELECT `QuantityOnHand` FROM `item` WHERE id=itemId);
SET nwQnty=itemQnty+crntQnty;
UPDATE `item` SET `QuantityOnHand`=nwQnty WHERE `Id`=itemId;
UPDATE `inventoryentry` SET `Status` = 1 WHERE `InventoryID`=invID AND
`ItemID`=itemId;
END$$
calling stored procedures
CALL Sample(28,10,92)
It update all the status = 1 in inventoryentry against InventoryID (i.e. 92) ignoring ItemID, instead of updating only one row. The following message displayed!
5 row(s) affected
Why Stored procedure ignoring itemID in update statement ? or Why Stored procedure updating more than one time? But without Stored procedure it working fine.
You need to use different variable names apart from your field name, also use the table name with the columns for better understanding like i used in following:
CREATE DEFINER=`root`#`localhost` PROCEDURE `Sample`(IN itemID INT, IN itemQnty
DOUBLE, IN invID INT)
BEGIN
DECLARE crntQnty DOUBLE;
DECLARE nwQnty DOUBLE;
SET crntQnty=(SELECT `QuantityOnHand` FROM `item` WHERE id=itemID);
SET nwQnty=itemQnty+crntQnty;
UPDATE `item` SET `QuantityOnHand`=nwQnty WHERE `QuantityOnHand`.`Id`=itemID;
UPDATE `inventoryentry` SET `Status` = 1 WHERE `InventoryID`=invID AND
`inventoryentry`.`ItemID`=itemID;
END$$
because of
update inventoryentry ... WHERE ... AND `ItemID`=itemId
You are saying that column itemid should be the same as column itemid which is always true
Try renaming your parameter to a name that differs from your column name
Using same names for columns and variable names has some issues.
Semantics of Stored procedure code is not checked at CREATE time. At runtime, undeclared variables are detected, and an error message is generated for each reference to an undeclared variable. However, SP's seem to believe any reference denotes a column, even though the syntactic context excludes that. This leads to a very confusing error message in case the procedure.
Your column name ItemID matches with input variable name itemId, and hence is the issue.
Please look at my answer to a similar query here.