I am just started writing the PL/pgSQL Trigger function. I am having couple of tables called Student and Result. Student having the following columns. ID,name,subject,mark (ID is the primary key) and the Result table is having two columns like ID,Status
Whenever one record has added in the student table, I want to update the Result table by checking the mark in the Student table, If the mark entered is greater than 50 then one record should be inserted in the Result table with ID and Status = Pass and if it is less than 50 then status will be fail. I have the following Trigger function to achieve this
CREATE OR REPLACE FUNCTION "UpdateResult"() RETURNS trigger AS $BODY$
BEGIN
IF NEW.mark < 50 THEN
INSERT INTO "Result" SELECT 92,'fail';
RETURN NEW;
ELSE
INSERT INTO "Result" SELECT 92,'pass';
RETURN NEW;
END IF;
END;
$BODY$
LANGUAGE 'plpgsql' VOLATILE STRICT SECURITY DEFINER
COST 100;
ALTER FUNCTION "UpdateResult"() OWNER TO postgres;
CREATE TRIGGER "Result"
AFTER INSERT
ON "Student"
FOR EACH ROW
EXECUTE PROCEDURE "UpdateResult"();
By this function trigger has worked as expected since I have hard coded the primary key value. But When I modify the SQL inside Trigger function like the following
INSERT INTO "Result" SELECT NEW.ID,'fail'; (or) INSERT INTO "Result" SELECT NEW.ID,'pass';
It is throwing error like
> ***Record "new" has no field "id" Context : PL/pgSQL function
> "UpdateResult" line 3 at SQL statement***
Means it is able to take the values of non primary key values from NEW variable not the primary key value. Can any one please tell me is there a restriction in PL/pgSQL or Am I doing anything wrong !