MySQL更新。 过程结果和phpmyadmin sql之间的区别

Question

I am having difficult getting a procedure to update a table in the way I require. I am using phpmyadmin on my local computer. In phpmyadmin I can put the following code into the SQL tab and one row will be updated:

SET `adjCost` = 22.05 WHERE `Name` LIKE CONCAT('magic', '%') AND `idKey` = '2016fulham02345';

As expected and wanted, IF the name begins with magic AND the idKey is '2016fulham02345' THEN the adjCost is updated to 22.05.

There will be between 2 and 50 rows with the same idKey. The Name will never be repeated in a set with the same idKey.

I created a procedure with the following parameters:

IN idK VARCHAR 255 Charset    
IN aName VARCHAR 255 Charset    
IN cost FLOAT 5,2

BEGIN 
    UPDATE `raceresults` SET `adjCost` = cost WHERE `Name` LIKE CONCAT(aName, '%') AND `idKey` = idK; 
END

When I run this procedure it updates ALL adjCost where the idKey = idk and (seems) to ignore the name parameter.

I have tried concatenating the name string first:

BEGIN 
    SELECT CONCAT(aName, '%') INTO @str; 
    UPDATE `raceresults` SET `adjCost` = cost WHERE `Name` = @str AND `idKey` = idK; 
END

but to no avail.

I looked through w3schools, stackoverflow and google and have not been able to find the answer.

My question is:

How can I correct my procedure to get it to work as I would like?

UPDATE: as requested.

    CREATE DEFINER=`root`@`localhost` PROCEDURE `importAltUpdateAjdCost`(IN `idK` VARCHAR(255), IN `aName` VARCHAR(255), IN `cost` FLOAT(5,2))
    NO SQL
BEGIN
UPDATE `costingPP`
SET `adjCost` = cost
WHERE 
`Name` LIKE CONCAT(aName, '%')
AND
`idKey` = idK;
END

To get this, I selected export on my list of procedures on phpmyadmin.

Answer 1

I'm not entirely sure what or how you did, but here's what I did and it instantly worked. Since you didn't specify MySQL version, I used 5.7.

EDIT: Now as I went back to see your procedure creation statement I realised that NO SQL was introduced in MySQL 8.0. Since your procedure clearly is SQL then please remove the NO SQL and re-create the procedure.

I'm leaving my MySQL 5.7 sample here for reference:

1) Created a simple table:

mysql> CREATE TABLE raceresults (
->   idKey VARCHAR(255),
->   Name VARCHAR(255),
->   adjCost FLOAT(5,2)
-> );
Query OK, 0 rows affected (0.06 sec)

2) Here we insert a sample data row:

mysql> INSERT INTO raceresults VALUES ('2016fulham02345', 'magicFlyingHorse', 0.00);
Query OK, 1 row affected (0.01 sec)

3) To create a (STORED) PROCEDURE we have to temporarily set a different delimiter, so query parser wouldn't terminate procedure creation on default semi-colon, as it's used inside the procedure. After delimiter's change we create the procedure and set the delimiter back to semi-colon

mysql> DELIMITER //
mysql> CREATE PROCEDURE update_test(IN idK VARCHAR(255), IN aName VARCHAR(255), IN cost FLOAT(5,2))
-> BEGIN
->     UPDATE `raceresults` SET `adjCost` = cost WHERE `Name` LIKE CONCAT(aName, '%') AND `idKey` = idK;
-> END//
mysql> DELIMITER ;
Query OK, 0 rows affected (0.00 sec)

4) Now let's see how it all works. Before and after the procedure call I'm selecting the rows from database. You can see the cost column value changing:

mysql> SELECT * FROM raceresults;
+-----------------+------------------+---------+
| idKey           | Name             | adjCost |
+-----------------+------------------+---------+
| 2016fulham02345 | magicFlyingHorse |    0.00 |
+-----------------+------------------+---------+
1 row in set (0.00 sec)

mysql> CALL update_test('2016fulham02345', 'magic', 1.23);
Query OK, 1 row affected (0.02 sec)

mysql> SELECT * FROM raceresults;
+-----------------+------------------+---------+
| idKey           | Name             | adjCost |
+-----------------+------------------+---------+
| 2016fulham02345 | magicFlyingHorse |    1.23 |
+-----------------+------------------+---------+
1 row in set (0.00 sec)

And now one piece of advise too: If possible, use only lower case table, column, indexes, functions, procedures, etc... names, while always writing all SQL commands in uppercase (which you did). This is kind of a de facto standard and makes life easier both for you and others reading your code.