On Insert- Create new entry? Trigger - mysql

Question

I currently have two tables for which I have listed the codes below:
CREATE TABLE `class` (
`id` int(11) NOT NULL,
`class_no` int(11) NOT NULL,
`class_title` varchar(11) NOT NULL,
`no_of_students` int(11) NOT NULL
) ENGINE=InnoDB DEFAULT CHARSET=latin1;
CREATE TABLE `enrolment` (
`id` int(11) NOT NULL,
`ssn` varchar(11) NOT NULL,
`class_no` int(11) NOT NULL,
`grade` varchar(11) NOT NULL
) ENGINE=InnoDB DEFAULT CHARSET=latin1;
I currently have the following statement, which when you enter something into 'Enrolment', and the class number you type in already matches a class_no in the table 'Class', then it adds whatever is typed into grade to no_of_students. What I now need it to do is if someone say typed into enrollment (2, 4, 6, 8) and the class_no 6 didn't already exist in the 'Class' table, it would add into 'Class' with the class_no as 6 and the no_of_students as 8.
DELIMITER $
CREATE TRIGGER `after_insert_add_one`
AFTER INSERT ON `enrolment`
FOR EACH ROW UPDATE class
SET class.no_of_students = class.no_of_students + NEW.grade
WHERE class_no = NEW.class_no
$
DELIMITER ;

Answer 1

(When you add an enrolment row do you really want to increment class.no_of_students by the value of grade? Seems like you might want to increment it by one. I wrote my answer assuming that.)
(I suspect you didn't show, in your question, the indexes definitions for your tables. I suspect you didn't show how your id columns get set. Presumably they are set with autoincrement. If that's not true my answer is probably worthless.)
MySQL offers an INSERT ... ON DUPLICATE KEY UPDATE statement. You can change your trigger to use it, something like this (not debugged).
DELIMITER $
CREATE TRIGGER `after_insert_add_one`
AFTER INSERT ON `enrolment`
FOR EACH ROW
INSERT INTO class
(class_no, no_of_students, grade)
VALUES (NEW.class_no, 1, NEW.grade)
ON DUPLICATE KEY
UPDATE SET class.no_of_students = class.no_of_students + 1$
DELIMITER ;
For this to work correctly you need a unique index on class.class_no. You can create it as follows:
CREATE UNIQUE INDEX class_no ON class (class_no);
By the way, you need some sort of index on class.class_no. Without it neither your original trigger nor the new one will perform very well. And, your "business rules" seem to require each value of class_no to have just one row in the class table. So you may as well use a unique index.
Pro tip: Giving your columns accurate and descriptive names is almost always worth a lot of trouble early in your project. If, for example, the column grade actually is a count of new students, you'll create years of confusion. The more successful your project is, the more confusion you'll create.

Answer 2

This can be done in app logic or with a trigger, the later below:
CREATE TRIGGER after_insert_add_one AFTER INSERT ON enrolment FOR EACH ROW
BEGIN
DECLARE counter int DEFAULT -1;
SELECT COUNT() INTO counter
FROM class
WHERE class_no = NEW.class_no;
IF counter = 0 THEN
# no classes exist with enrolment ID
INSERT INTO class (class_no) VALUES (NEW.class_no);
ELSE:
# class exists ++
UPDATE class
SET class.no_of_students = class.no_of_students + NEW.grade
WHERE class_no = NEW.class_no;
END IF;
END;