I'm working on a simple trigger that updates the Quantity on Hand of a product when adding a new invoice line.
I got that working fine; but I starting thinking that in a practical application it would be prudent to verify that the QOH is sufficient before allowing the update. I'm sure I could write a separate trigger for that, but I'd like to see if it's possible to join two tables for a conditional statement within a trigger.
This is as far as I've gotten; adding a SELECT statement anywhere in there causes all hell to break loose, so I'm a little stumped how I can declare the PRODUCT.P_QOH before calling the conditional.
CREATE OR REPLACE TRIGGER trg_prod_QOH_on_line_add
BEFORE INSERT ON LINE
FOR EACH ROW
BEGIN
IF :NEW.LINE_UNITS > PRODUCT.P_QOH THEN
RAISE_APPLICATION_ERROR(-20202, 'Insufficient quantity on hand');
ELSE
UPDATE PRODUCT
SET P_QOH = P+QOH - :NEW.LINE_UNITS;
WHERE PRODUC.P_CODE = :NEW.P_CODE;
END IF;
END;
/
This isn't a major problem for me, as I said there is probably a different way; I'm just starting to learn this stuff and would like to see what's possible. Thanks for your help.
You get into dangerous territory trying to enforce rules like this via triggers. The solution you are asking for is:
create or replace trigger trg_prod_qoh_on_line_add
before insert on line
for each row
declare
v_qoh product.p_qoh%type;
begin
select p_qoh
into v_qoh
from product
where product.p_code = :new.p_code;
if :new.line_units > v_qoh then
raise_application_error(-20202, 'Insufficient quantity on hand');
else
update product
set p_qoh = p_qoh - :new.line_units
where product.p_code = :new.p_code;
end if;
end;
However this is not a safe solution in a system with more than one concurrent user. Suppose product 'X' has p_qoh=10 and then 2 users do this:
user1> insert into line (p_code, line_units) values ('X', 7);
user2> insert into line (p_code, line_units) values ('X', 8);
user1> commit;
user2> commit;
Both sessions will see that 'X' has p_qoh = 10 so both will succeed and product.p_qoh will end up as -5. All is corrupt!
The safe solution would be to create a check constraint on product:
alter table product add constraint prod_qoh_chk check (p_qoh >= 0);
Now your trigger can be simply:
create or replace trigger trg_prod_qoh_on_line_add
before insert on line
for each row
begin
update product
set p_qoh = p+qoh - :new.line_units;
where produc.p_code = :new.p_code;
end;
This would raise a less friendly error message like:
ORA-02290: check constraint (MYSCHEMA.PROD_QOH_CHECK) violated
You can trap this in your trigger and give the message you want:
create or replace trigger trg_prod_qoh_on_line_add
before insert on line
for each row
begin
update product
set p_qoh = p+qoh - :new.line_units;
where produc.p_code = :new.p_code;
exception
when others then
if sqlerrm like 'ORA-02291:%(MYSCHEMA.PROD_QOH_CHECK)%' then
raise_application_error(-20202,'Insufficient quantity on hand');
else
raise;
end if;
end;
Now if we re-run the 2 user scenario above:
user1> insert into line (p_code, line_units) values ('X', 7);
user2> insert into line (p_code, line_units) values ('X', 8);
user1> commit;
At this point user2's insert fails with the error message:
ORA-20202: Insufficient quantity on hand
Related
I am new in learning SQL and I am trying to apply a trig on those two tables by using SQLcl:
prodc
( PRODC_NAME, DISCOUNTED CHAR(1) DEFAULT 'N',
PK PRIMARY KEY (PRODC_NAME));
peopleOrder
( ORDERID, PRODC_NAME,
PRIMARY KEY (ORDERID, PRODC_NAME),
FOREIGN KEY (PRODC_NAME) REFERENCES PRODC (PRODC_NAME));
I want my trig to do the following:
when I update or insert rows to peopleOrder, the trig should check if a row's product_name has value 'N' in the discounted column that is located in the product table; if it has no 'N' value, it should show an error message.
I tried many ways and the following trig, but when I insert rows, the trig seems to not be working and have no effect on the rows:
CREATE OR REPLACE TRIGGER constraint_1
After UPDATE OR INSERT on peopleOrder
for each row
begin
if not (:new.prodc_name in
(select prodc_name from prodc where discounted = 'N' )) then
RAISE_ERROR(-30001, 'No product can be ordered
!');
END IF;
INSERT INTO peopleOrder VALUES (:new.ORDERID, :new.PRODC_NAME);
END;
/
My insert command is:
INSERT INTO peopleOder (ORDERID, PRODC_NAME)
VALUES (251, 'Puton');
and 'Puton' has value 'N' in the discounted column in the prodc table.
The first thing you need is to understand what a trigger is. It is a piece of code that runs as part of the statement causing it to fire. For this reason you cannot reference the table causing it to fire. Normally that is completely unnecessary anyway. Every row trigger has access to to 2 rows: a copy of the existing data (old) and the resulting data (new). Update having both filled and Insert/Delete just new/old as appropriate for the action. Before statement triggers can modify the new data, after triggers cannot. Those rows allow you to process the columns in the table without DML or select on it. (However you cannot 'scan' the table). At any level a trigger may raise an exception which halts the entire process and the invoking statement. If no exception is raised then processing the invoking proceeds. For a more complete description see PL/SQL Language Reference or the version of Oracle you are running.
In this case your trigger breaks down to a single if
statement.
-- Revised. Orig missed that flag comming from different table.
create or replace trigger constraint_1
before update or insert on peopleOrder
for each row
declare
dis_num integer;
begin
select count(*)
into dis_num
from product
where discounted != 'N'
and product_name = :new.product_name
and rownum < 2;
if dis_num != 0 then
raise_application_error(-20001, 'No discounted product can be ordered!');
end if;
end constraint_1;
-- Orig ==============================================
CREATE OR REPLACE TRIGGER constraint_1
before UPDATE OR INSERT on peopleOrder
for each row
begin
if :new.discounted = 'N' then
RAISE_APPLICATION_ERROR(-20001, 'No discounted product can be ordered')
end if;
END;
/
I don't know what my problem when created new trigger.
Is my syntax correct? Thanks! Console Logging pane
p/s: This my console display when I try to insert values
CREATE OR REPLACE TRIGGER EX03_3
BEFORE INSERT ON HR.CHITIETDATHANG
FOR EACH ROW
DECLARE
TONGHANG NUMBER; -- Total Items
HANGHIENCO NUMBER; -- Items present
HANGDABAN NUMBER; -- Items was sales.
BEGIN
-- Get total Items
SELECT SUM(MH.SOLUONG) INTO TONGHANG
FROM HR.MATHANG MH;
-- Get total Items was sales
SELECT SUM(CTDH.SOLUONG) INTO HANGDABAN
FROM HR.CHITIETDATHANG CTDH;
-- Items present
HANGHIENCO := TONGHANG - HANGDABAN;
IF(HANGHIENCO >= HANGDABAN) THEN
HANGHIENCO := HANGHIENCO-1;
INSERT INTO HR.CHITIETDATHANG VALUES(:NEW.SOHOADON,:NEW.MAHANG,
:NEW.GIABAN,:NEW.SOLUONG,:NEW.MUCGIAMGIA);
ROLLBACK;
END IF;
NULL;
END;
Seem two critical mistakes
1) trigger tries to insert into HR.CHITIETDATHANG in the body of insert trigger of HR.CHITIETDATHANG.
2) to use rollback after an insert statement is useless.
Note : I can see nothing relevant to make raise no_data_found in those select statements. EX03_3 and EX04_4 are confused, as Kaushik Nayak says.
I need to Alter the Product table to have a varchar2(3) reorder column and modify the statement level trigger from the notes to set the product reorder field to “yes” if the quantity on hand is less than twice the product min quantity or if the quantity on hand is less than 10. The value should be “no” otherwise. The trigger is fired after an insert or an update to p_min or p_qoh. Debug by checking the data beforehand, making a change, then checking afterwards.
For this I have
CREATE or REPLACE TRIGGER
TRG_Product_Reorder
AFTER INSERT OR UPDATE of P_min, P_qoh ON lab9_Product
BEGIN
UPDATE lab9_Product SET REORDER = 'yes'
WHERE P_qoh < P_min*2 or p_qoh < 10;
UPDATE lab9_Product SET REORDER = 'no'
WHERE P_qoh >= p_min*2;
END;
/
I get the error:
SQL statement ignored
"REORDER": invalid identifier
I think you might want a before trigger rather than an after trigger. It would be something like this:
CREATE or REPLACE TRIGGER
TRG_Product_Reorder
BEFORE INSERT OR UPDATE of P_min, P_qoh ON
lab9_Product
BEGIN
IF :OLD.P_qoh < :OLD.P_min*2 AND :OLD.p_qoh < 10 THEN
:NEW.REORDER := 'yes';
ELSE :NEW.REORDER := 'no';
END IF;
END;
I'm creating a trigger within my database, I came across two error that I am not able to fix, I'm pretty sure that those two are relating to my use of DBMS_OUTPUT.PUT_LINE, the rest of the statement does not cause any errors, although it had before.
Errors:
Error(5,3): PL/SQL: SQL Statement ignored
Error(5,15): PL/SQL: ORA-00903: invalid table name
Code:
CREATE TRIGGER INVOICES
BEFORE INSERT OR UPDATE ON BRUINVOICE
FOR EACH ROW
BEGIN
IF :new.BRU_DATE < :new.BRU_PAID_DATE THEN
DBMS_OUTPUT.PUT_LINE('You cannot do that');
ELSE
INSERT INTO table BRUINVOICE
values
from inserted;
END IF;
END;
Check constraints are a better choice (performance-wise) than triggers when it comes to record level validation:
ALTER TABLE bruinvoice
ADD CONSTRAINT validate_bru_date CHECK (BRU_DATE < BRU_PAID_DATE);
Inserting invalid data will raise an error message like the following:
scott#ORCL> insert into bruinvoice values ('21-DEC-14','20-DEC-14');
insert into bruinvoice values ('21-DEC-14','20-DEC-14')
*
ERROR at line 1:
ORA-02290: check constraint (SCOTT.VALIDATE_BRU_DATE) violated
I fully agree with cstotzer, a check constraint is much better in your situation at should be the preferred way of doing it. However, just for information this would be the trigger syntax:
CREATE TRIGGER INVOICES
BEFORE INSERT OR UPDATE ON BRUINVOICE
FOR EACH ROW
BEGIN
IF :new.BRU_DATE < :new.BRU_PAID_DATE THEN
RAISE_APPLICATION_ERROR(-20001, 'You cannot do that');
END IF;
END;
You don't need any ELSE, your INSERT or UPDATE will be simply executed in this case.
I have two Tables.
I want to insert the same record on both tables at the same time.
I.e., while I insert a record for the first table, this same record also is inserted in the second table using a trigger.
Do you have any experience/advice in this process ?
if you're using stored procedures you can easily manage this
CREATE PROCEDURE sp_Insert
#Value varchar(10)
AS
insert into table1 (...,...) values (#value,...)
insert into table2 (...,...) values (#value,...)
I would suggest using Erik's method over a trigger. Triggers tend to cause performance issues, and a lot of times, you forget that the trigger exists, and get unexpected behavior. If you do want to use a trigger however, it will work. here is an example:
CREATE TRIGGER trgTest ON Test
FOR INSERT
AS
INSERT Test2
(Id, value)
SELECT Id, Value
FROM Inserted
Can Use Cursor Concept!
CREATE OR REPLACE TRIGGER bi_order
BEFORE INSERT
ON ord
REFERENCING OLD AS OLD NEW AS NEW
FOR EACH ROW
WHEN (NEW.payment_type = 'CREDIT')
DECLARE
CURSOR cur_check_customer IS
SELECT 'x'
FROM customer
WHERE customer_id = :NEW.customer_id
AND credit_rating = 'POOR';
lv_temp_txt VARCHAR2(1);
lv_poor_credit_excep EXCEPTION;
BEGIN
OPEN cur_check_customer;
FETCH cur_check_customer INTO lv_temp_txt;
IF (cur_check_customer%FOUND) THEN
CLOSE cur_check_customer;
RAISE lv_poor_credit_excep;
ELSE
CLOSE cur_check_customer;
END IF;
EXCEPTION
WHEN lv_poor_credit_excep THEN
RAISE_APPLICATION_ERROR(-20111, 'Cannot process CREDIT ' ||
'order for a customer with a POOR credit rating.');
WHEN OTHERS THEN
RAISE_APPLICATION_ERROR(-20122, 'Unhandled error occurred in' ||
' BI_ORDER trigger for order#:' || TO_CHAR(:NEW.ORDER_ID));
END bi_order;