I need a little help with a trigger. Where I'm in wrong:
CREATE TRIGGER [dbo].[tr_ins_MyTriggerName]
ON [dbo].MyTableName
FOR INSERT
AS
DECLARE
@Number nvarchar(20) ,
@Date datetime ,
@Supplier nvarchar(12);
SET @Number = (SELECT Number FROM INSERTED);
SET @Date = (SELECT Date FROM INSERTED);
SET @Supplier = (SELECT Supplier FROM INSERTED);
BEGIN
IF EXISTS (
SELECT TOP 1 1 FROM MyTableName x
WHERE x.Number = @Number
AND x.Date = @Date
AND x.Supplier = @Supplier
)
BEGIN
RAISERROR ('There is the same Number for the same Supplier!' ,11,1)
ROLLBACK
END
END
This trigger runs in all cases even the the EXITS
is not true!
2 Answers 2
I'd suggest you to create a unique index on the 3 columns, i.e. date / number / supplier.
If this is not applicable, I'd say you modify your trigger as follows
CREATE TRIGGER [dbo].[tr_ins_MyTriggerName]
ON [dbo].MyTableName
FOR INSERT
AS
DECLARE
@Number nvarchar(20) ,
@Date datetime ,
@Supplier nvarchar(12);
SET @Number = (SELECT Number FROM INSERTED);
SET @Date = (SELECT Date FROM INSERTED);
SET @Supplier = (SELECT Supplier FROM INSERTED);
BEGIN
IF (
SELECT count(*) FROM MyTableName x
WHERE x.Number = @Number
AND x.Date = @Date
AND x.Supplier = @Supplier
) > 1
BEGIN
RAISERROR ('There is the same Number for the same Supplier!' ,11,1)
ROLLBACK
END
END
However, this trigger is problematic when you insert multiple values in one insert statement.
This trigger runs in all cases even the the EXITS is not true!
The condition will always be true since this trigger fires after the record is inserted. FOR
is synonymous with AFTER
for TRIGGERS
, thus the EXISTS
would never not be true.
FOR | AFTER
Specifies that the DML trigger is fired only when all operations specified in the triggering SQL statement have executed successfully. All referential cascade actions and constraint checks also must succeed before this trigger fires.
AFTER is the default when FOR is the only keyword specified.
When you say it runs do you mean the trigger fires, or that the error is always raised? I'd expect both, as explained above.
It seems like you are trying to do error handling for an INSERT
in which I'd do it in the procedure or code block that you are doing the INSERT
via a TRY / CATCH
block, or just add a constraint on the table to prevent this type of duplicate insertion. Erland Sommarskog has an extensive blog on error handling that's very helpful. Here's a basic framework via one of this posts, but be sure to read it!
Here is a stored procedure that showcases how you should work with errors and transactions.
CREATE PROCEDURE insert_data @a int, @b int AS
SET XACT_ABORT, NOCOUNT ON
BEGIN TRY
BEGIN TRANSACTION
INSERT sometable(a, b) VALUES (@a, @b)
INSERT sometable(a, b) VALUES (@b, @a)
COMMIT TRANSACTION
END TRY
BEGIN CATCH
IF @@trancount > 0 ROLLBACK TRANSACTION
DECLARE @msg nvarchar(2048) = error_message()
RAISERROR (@msg, 16, 1)
RETURN 55555
END CATCH
-
Hi, @scsimon, Thank you very match for your answer! It is very useful. Details are: 1. I do not have access to source code. 2. We have many old records and that is the reason that I want to add trigger to stop the new errors. :)StanYan– StanYan2019年01月09日 05:50:42 +00:00Commented Jan 9, 2019 at 5:50