Views And Instead Of Update Trigger Limitation. Why?
Sep 19, 2007
I am attempting to use views and Instead of triggers to insert, update and delete rows from THE table that is exposed through a view. In my application, the views provide a filtered set of rows from the base table that a user has permission to access. To improve performance, I submit a set (i.e. multiple rows) to be acted upon in a single SQL Statement. The insert and delete triggers work great. But, for some reason, the Instead of Update trigger on the view will not accept multiple rows. The Error I receive is:
Msg 414, Level 16, State 1, Line 1
UPDATE is not allowed because the statement updates view "FooView" which participates in a join and has an INSTEAD OF UPDATE trigger.
The only hint I can find about this behavior is from BOL for the Transact-SQL Update statement:
"A view with an INSTEAD OF UPDATE trigger cannot be a target of an UPDATE with a FROM clause."
Why is this not allowed?
If the trigger is on a table and the update statement is executed against the table it works. Why is the statement not allowed to execute against a view that has exactly the same construction as the table?
Any help or explanations for this behavior is appreciated.
I've included a code sample that will replicate the behavior and problem.
CREATE TABLE Foo (
ID INT PRIMARY KEY,
aName VARCHAR (10),
aValue FLOAT
)
GO
-- The view and its triggers
CREATE VIEW FooView AS
SELECT [ID]
,[aName]
,[aValue]
FROM Foo
GO
-- Instead of Triggers on the view
CREATE TRIGGER FooViewInsertInsteadOf
ON FooView
INSTEAD OF INSERT
AS
BEGIN
INSERT INTO Foo
SELECT Id, aName, aValue
FROM INSERTED
END
GO
CREATE TRIGGER FooViewInsteadOfUpdate
ON FooView
INSTEAD OF UPDATE
AS
BEGIN
UPDATE Foo
SET
Foo.aName = Inserted.aName,
Foo.aValue = Inserted.aValue
FROM Foo
JOIN INSERTED on INSERTED.ID = Foo.Id
END
GO
-- Test the Insert Trigger
INSERT INTO FooView
SELECT 2, 'First', 2.0
UNION ALL
SELECT 3, 'First', 3.0
UNION ALL
SELECT 4, 'First', 4.0
GO
SELECT * FROM FooView
GO
-- Update a single row in the view. This works with the Instead of Update Trigger.
UPDATE FooView
SET
FooView.aName = 'Third',
FooView.aValue = 3.3
WHERE FooView.ID = 3
GO
SELECT * FROM FooView
GO
===============================
-- T-SQL will not support the following statement when executed against a view with an instead of trigger.
-- If this statement is executed against a table with an instead of trigger it works as desired.
-- Why can't I execute this against the view?
UPDATE FooView
SET
FooView.aName = tmp.f1,
FooView.aValue = tmp.f2
FROM FooView
JOIN (
SELECT 2 as [id], 'Second' as [F1], 2.2 as [F2]
UNION ALL
SELECT 4, 'Fourth', 4.33
)
as tmp on FooView.ID = tmp.[id]
Currently I have two tables T1 and T2. A view V1 is defined over T1 with an INSTEAD OF UPDATE trigger and another view V2 is defined over V1 and T2 with another INSTEAD OF UPDATE trigger. Unfortunately, inside V2’s trigger following update statement is not working: Update V1 set V1.name = i.name from inserted i because SQL server complains: View 'V1' has an INSTEAD OF UPDATE trigger and cannot be a target of an UPDATE FROM statement. Is there any way to get around this problem? I.e., how can I make the INSTEAD OF UPDATE trigger in V2 to work if I can’t reference “inserted”?
Hi,I am new to using SQL. I want to be able to exucute a query that Iplace in a view. I want this views to be executed every time a valuechange in one of the table in a particular field. So my guess was touse a trigger that will call the views every time the data change inthe selected table. Is this the proper way of doing thing? Should Iuse other SQL tools to achive this. I search for exemple of triggerexecuting views but did not found anything as of yet. Let's use thisdummy name for the exemple:Database: DB1Table: Tbl1Special field in Tbl1: flagViews name: views_01Thank you.Philippe
Are there any limitations or gotchas to updating the same table whichfired a trigger from within the trigger?Some example code below. Hmmm.... This example seems to be workingfine so it must be something with my specific schema/code. We'reworking on running a SQL trace but if anybody has any input, fireaway.Thanks!create table x(Id int,Account varchar(25),Info int)GOinsert into x values ( 1, 'Smith', 15);insert into x values ( 2, 'SmithX', 25);/* Update trigger tu_x for table x */create trigger tu_xon xfor updateasbegindeclare @TriggerRowCount intset @TriggerRowCount = @@ROWCOUNTif ( @TriggerRowCount = 0 )returnif ( @TriggerRowCount > 1 )beginraiserror( 'tu_x: @@ROWCOUNT[%d] Trigger does not handle @@ROWCOUNT[color=blue]> 1 !', 17, 127, @TriggerRowCount) with seterror, nowait[/color]returnendupdate xsetAccount = left( i.Account, 24) + 'X',Info = i.Infofrom deleted, inserted iwhere x.Account = left( deleted.Account, 24) + 'X'endupdate x set Account = 'Blair', Info = 999 where Account = 'Smith'
I want to retrieve the last update time of database. Whenever any update or delete or insert happend to my database i want to store and retrieve that time.
I know one way is that i have to make a table that will store the datetime field and system trigger / trigger that can update this field record whenever any update insert or deletion occur in database.
But i don't know exactly how to do the coding for this?
I have a table where table row gets updated multiple times(each column will be filled) based on telephone call in data.  Initially, I have implemented after insert trigger on ROW level thinking that the whole row is inserted into table will all column values at a time. But the issue is all columns are values are not filled at once, but observed that while telephone call in data, there are multiple updates to the row (i.e multiple updates in the sense - column data in row is updated step by step),
I thought to implement after update trigger , but when it comes to the performance will be decreased for each and every hit while row update.
I need to implement after update trigger that should be fired on column level instead of Row level to improve the performance?
Hi, I am not sure if this is the right forum to post this question. I run an update statement like "Update mytable set status='S' " on the SQL 2005 management Studio. When I run "select * from mytable" for a few seconds all status = "S". After a few seconds all status turn to "H". This is a behaviour when you have an update trigger for the table. But I don't see any triggers under this table. What else would cause the database automatically change my update? Could there be any other place I should look for an update trigger on this table? Thanks,
VIEW behaviour depending on what operation is run on it.
VIEW syntax is CREATE VIEW xx AS SELECT ... It would be very convinient to know wether a select, update, insert or delete is the operation executed on it so the view could take action accordingly, ie CREATE VIEW xxx AS SELECT .. CASE UPDATE THEN do_this CASE SELECT THEN do_that etc. Does anyone know if this is possible?
My client want me to update a table and the view with a new column which should be hidden. I am kind of confused on how to go about this in order to get a good result. PLEASE HELP!!
I have an update trigger which fires from a transactiion table to update a parent record in another table. I am getting no errors, but also no update. Any help appreciated (see script below)
create trigger tr_cmsUpdt_meds on dbo.medisp for UPDATE as
if update(pstat) begin update med set REC_FLAG = 2 from deleted dt where med.uniq_id = dt.uniq_id and dt.pstat = 2 and dt.spec_flag = 'kop' end
I am trying to update a fields with an UPDATE statement but I keep getting the error message when I run the query.
Subquery returned more than 1 value. This is not permitted when the subquery follows =, !=, <, <= , >, >= or when the subquery is used as an expression.
I have this Update trigger that I know is causing the error message because I guess it's not built to manage multi-row updates.
Can someone help me re-write it. I also tried using the WHERE p.ID = p.ID but when I do that it modifies all rows in the modifieddate column instead of just the cells/rows that I'm updating
ALTER TRIGGER [dbo].[MultitrigCA] ON [dbo].[ProdDesc] AFTER UPDATE AS
SET NOCOUNT ON
IF UPDATE (codeabbreviation) UPDATE p sET p.ModifiedDate = GETDATE() FROM ProdDesc AS p WHERE p.ID = (SELECT ID FROM inserted)
Hi there,I'm a little stuck and would like some helpI need to create an update trigger which will run an update query onanother table.However, What I need to do is update the other table with the changedrecord value from the table which has the trigger.Can someone please show me how this is done please??I can write both queries, but am unsure as to how to get the value ofthe changed record for use in my trigger???Please helpM3ckon*** Sent via Developersdex http://www.developersdex.com ***Don't just participate in USENET...get rewarded for it!
i have a database with about 20 tables. i appended to each table a column "UpdatedOn", and i want to write a trigger to set the date of the update date in that column, using a trigger.
i want to avoid the trigger launching for the last column (UpdatedOn).
how can i detect the rows that changed, and modify only the update date/time? i read something about TableName_Inserted and TableName_Deleted, but i would prefer to copy as generic as possible the data from there, meaning, not to write column names in my script.
another idea i thought about was to prevent the trigger executing if no other column except for UpdatedOn changed, but... i encounter some trouble, when i try to pass column name (as string) to UPDATE() function.(Error: Expecting ID or QUOTED_ID)
If I install SQL 2005 Standard on Windows 2003 Standard, is SQL limited to 4 gigs of physical RAM?
I'm planning a new system that will run SQL 2005 Standard edition on a Windows 2003 Standard platform. The spec calls for 8 GB of RAM. My experience would lead me to suspect I need to install Windows 2003 Enterprise to take advantage of all the installed memory.
Fellow database developers,I would like to draw on your experience with views. I have a databasethat includes many views. Sometimes, views contains other views, andthose views in turn may contain views. In fact, I have some views inmy database that are a product of nested views of up to 6 levels deep!The reason we did this was.1. Object-oriented in nature. Makes it easy to work with them.2. Changing an underlying view (adding new fields, removing etc),automatically the higher up views inherit this new information. Thismake maintenance very easy.3. These nested views are only ever used for the reporting side of ourapplication, not for the day-to-day database use by the application.We use Crystal Reports and Crystal is smart enough (can't believe Ijust said that about Crystal) to only pull back the fields that arebeing accessed by the report. In other words, Crystal will issue aSelect field1, field2, field3 from ReportingView Where .... eventhough "ReportingView" contains a long list of fields.Problems I can see.1. Parent views generally use "Select * From childview". This meansthat we have to execute a "sp_refreshview" command against all viewswhenever child views are altered.2. Parent views return a lot of information that isn't necessarilyused.3. Makes it harder to track down exactly where the information iscoming from. You have to drill right through to the child view to seethe raw table joins etc.Does anyone have any comments on this database design? I would love tohear your opinions and tales from the trenches.Best regards,Rod.
Hello All, I've been adviced without much detail that the best way to handle updates is with an Update Trigger to write to an AuditTable. I always thought it was better to avoid triggers. I'm getting confused very fast in reading about triggers. Could someone please point me in the right direction?
i am using address table to store the address of employees, agents, students; so can't use cascade update and delete. so i decided to use trigger but i got stuck here as well. Here employeeid is the PK of employee id. so when it is updated, i want to update that id in corresponding address table as well. so what i need is to find out the previous value of EmployeeID before updating to use in WHERE clause and new EmployeeID to use in SET clause.
<code> CREATE TRIGGER [AddressUpdate] ON [dbo].[MCS_Employee] FOR UPDATE AS IF UPDATE(EmployeeID) BEGIN UPDATE MCS_Address SET EmployeeID = 'i am stuck here; how to know the previous value? WHERE EmployeeID = ??? END </end>
ironically, PK in my table can be updateable as well.
Hello all. I have a table which is the exact replication of a different table. Now if an update takes place in the original table I want to copy that row into the new table without the update.I have this but I am not sure if it works.Create Trigger Content_Archive on content For Update As Insert Into content_audit Select * From DeletedI dont know if Updated exists so I am using Deleted. It is returning both rows (before and after update). And the insert should take place as soon as an update takes place in any field in the content table.
THREAD ------- THREAD_ID TOPIC_ID STATUS_ID THREAD THREAD_DATE NAME
I want to update STATUS_ID in TOPIC when I post a new THREAD. I guess the best would be to use a trigger? Don't know much how you write them so if someone please could help or point me in right direction. My plan is to always show the updated STATUS_ID in TOPIC while I have history in the THREAD TBL. I am using MS SQL 7.0
I am new to this and on my way to learning. I hope someone can help me with a trigger. I want to update status of one table when I insert a new threat in another. This is what I have come up with so far.
CREATE TRIGGER [trg_Update_Status] ON tForumThread FOR INSERT AS BEGIN UPDATE tJournalTopic (STATUS_ID) SELECT STATUS_ID FROM Inserted
I need to be able to select the row or specific field that was updated in an update trigger. I dont have any time-stamp or before-after value columns to compare. Please help!
I HAVE TWO TABLES IN THE DATABSE T1 AND THE SECOND TABLE(T2) IS FOR AUDITING. THE STRUCTURE OF BOTH THE TABLES IS SIMILAR WITH ONE MORE COLUMN 'STATUS' IN THE T2. I HAVE ALREADY WRITTEN THE TRIGGERS FOR THE INSERT AND DELETE TRIGGERS SO THAT I CAN SAVE THE RECORD WITH STATUS OF 'NEW INSERT' OR 'DELETE' OR 'UPDATE' IN THE AUDIT TABLE(T2). HELP ME IN WRITING THE TRIGGER FOR UPDATE. MY PROBLEM IS I DON'T KNOW HOW TO STORE THE UPDATED COLUMNS VALUES INTO VARIABLES AND SO THAT I CAN STORE THE OLD VALUES INTO THE AUDIT TABLE(T2). I M USING SQL SERVER 6.5, SO WE CAN NOT USE COLUMNS_UPDATED() IN THE TRIGGER. PLEASE SUGGEST ME THE CODE SO THAT I CAN STORE THE OLD ROW IN THE TABLE(T1) TO TABLE (T2) WITH STATUS 'UPDATE' IN THE T2.
PLEASE SUGGEST ME..ITS URGENT.
THANKS IN ADVANCE HARISH
============================= /*test trigger for insert status */ if exists (select * from sysobjects where id = object_id('dbo.tri_t1_insert') and sysstat & 0xf = 8) drop trigger dbo.tri_t1_insert GO
CREATE TRIGGER tri_t1_insert ON dbo.t1 FOR INSERT AS declare @v1 binary(20), @v2 varchar(255) Begin select @v1=stamp,@v2=name from inserted insert into t2(stamp,name,status) values(@v1,@v2,'NEW INSERT') end GO ======================================== /*test trigger for delete status */ if exists (select * from sysobjects where id = object_id('dbo.tri_t1_delete') and sysstat & 0xf = 8) drop trigger dbo.tri_t1_delete GO
CREATE TRIGGER tri_t1_delete ON dbo.t1 FOR delete AS declare @v1 binary(20), @v2 varchar(255) Begin select @v1=stamp,@v2=name from deleted insert into t2(stamp,name,status) values(@v1,@v2,'DELETE')
I have an update trigger on a table on my transactional database that inserts a row of data into another database (audit database)for any modification made on the transactional database. So if i modify a row on tran db it will write the data modified as a new row in the audit db.
This works fine if I am updating only 1 row with each Update statement. However if I update more than 1 row (multiple rows) with the same Update statement, the update trigger only inserts the last row modified in the audit database. So I lose record of any other rows modified with the same update statement.
Is there a way by which i can change my Update trigger or something, so I get all the rows updated by 1 update statement inmy audit database???
these columns can allow nulls. What i whant to do with my trigger is do a after trigger and check after the load if a certain column is NULL place a X instead like a flag but only on the columns that are NULL how would i do this.
Is there a way I can get the old value of a specific field when using an update trigger? e.g. I want to use the condition 'If Update(Column) ' in odrer to create a logfile which stores the old and new value of a field. The new value can be get from a Select from Inserted table. Is there a way I can get the old value (before update) also??
I have created a table with the following columns Jobnumber varchar(20), weight real(4), freightcost money(8), trackingnumber vchar(50), comments varchar(2000) and voidid varchar(3)
I wrote a trigger that updates this data based on the voidid to update the package table as followed: CREATE TIGGER [UPS] ON dbo.UPSSHIPMENT FOR INSERT AS DECLARE @JOBNUMBER CHAR(20) DECLARE @WEIGHT REAL(4) DECLARE @FREIGHTCOST MONEY(8) DECLARE @TRACKINGNUMBER CHAR(25) DECLARE @SHIPMETHOD CHAR(50) DECLARE @voidid char(2)
SELECT @JOBNUMBER=JOBNUMBER, @WEIGHT=WEIGHT, @FREIGHTCOST=FREIGHTCOST, @TRACKINGNUMBER=TRACKINGNUMBER, @SHIPMETHOD=SHIPMETHOD, @VOIDID=VOIDID FROM INSERTED
UPDATE PACKAGE SET PACKAGE.WEIGHT = @WEIGHT, PACKAGE.FREIGHTCOST = @FREIGHTCOST, PACKAGE.TRACKINGNUMBER = @TRACKINGNUMBER, PACKAGE.COMMENTS = @SHIPMETHOD WHERE PACKAGE.JOBNUMBER = @JOBNUMBER AND @VOIDID = 'N'
UPDATE PACKAGE SET PACKAGE.WEIGHT = '', PACKAGE.TRACKINGNUMBER = '', PACKAGE.COMMENTS = 'UPS VOID', PACKAGE.FREIGHTCOST = '' WHERE PACKAGE.JOBNUMBER = @JOBNUMBER AND @VOIDID = 'Y'
I am getting the following error see attached. Any help would be great Thank you!
I'm totally new to writing triggers, but managed to get a trigger to update a specific column (ufINCTcost) on change of another column (ufINCKm)
My problem is that the trigger performs this update on ALL rows, which makes it very slow. How can I get it to only update the column on the row where the change was made ?
My trigger looks as follows:
CREATE TRIGGER updateincidents ON [dbo].[_rtblIncidents] FOR UPDATE, INSERT AS IF UPDATE (ufINCKm) BEGIN UPDATE dbo._rtblIncidents SET ufINCTcost=dbo._rtblIncidents.ufINCKm+dbo._rtblInc idents.ufINCToll+dbo._rtblIncidents.ufINCParking+d bo._rtblIncidents.ufINCFlight+dbo._rtblIncidents.u fINCRental
The EMPLOYEE table holds all the basic stuff you would want to know about an employee and it contains a STORE_CODE attribute that points to the same attribute in the STORE table (to keep track of the store they work at). The STORE table has the STORE_CODE attribute, a NUM_EMP attribute that keeps track of the number of employees at that store, and some other information that is of no relevance to the question.
I've written the following TRIGGER to update the NUM_EMP attribute in STORE everytime a row is inserted or deleted from EMPLOYEE. It works fine for inserts and deletes but I am clueless as how to make it work for updates (an EMPLOYEE transfers to another store). I appreciate any feedback and please feel free to tell where I've gone wrong so far.
Thanks!
CREATE TRIGGER [UPDATENUMEMP] ON [EMPLOYEE] FOR INSERT,UPDATE,DELETE AS
DECLARE @STORECODE INT DECLARE @NUMEMP INT
/*DELETE CASE*/ IF (NOT EXISTS (SELECT * FROM INSERTED) AND EXISTS (SELECT * FROM DELETED)) BEGIN DECLARE DELETECURSOR CURSOR FOR SELECTSTORE_CODE, COUNT(*) AS NUMEMP FROMDELETED GROUP BY STORE_CODE
OPEN DELETECURSOR
FETCH NEXT FROM DELETECURSOR INTO @STORECODE, @NUMEMP WHILE(@@FETCH_STATUS = 0) BEGIN UPDATESTORE SETNUM_EMP = NUM_EMP - @NUMEMP WHERESTORE_CODE = @STORECODE FETCH NEXT FROM DELETECURSOR INTO @STORECODE, @NUMEMP END CLOSE DELETECURSOR DEALLOCATE DELETECURSOR END
/*INSERT CASE*/ IF(EXISTS (SELECT * FROM INSERTED) AND NOT EXISTS (SELECT * FROM DELETED)) BEGIN DECLARE INSERTCURSOR CURSOR FOR SELECTSTORE_CODE, COUNT(*) AS NUMEMP FROMINSERTED GROUP BY STORE_CODE
OPEN INSERTCURSOR
FETCH NEXT FROM INSERTCURSOR INTO @STORECODE, @NUMEMP WHILE(@@FETCH_STATUS = 0) BEGIN UPDATESTORE SETNUM_EMP = NUM_EMP + @NUMEMP WHERESTORE_CODE = @STORECODE FETCH NEXT FROM INSERTCURSOR INTO @STORECODE, @NUMEMP END CLOSE INSERTCURSOR DEALLOCATE INSERTCURSOR END GO
This is something I have limited experience on. I need to create an update trigger after insert on one table that updates a completely different production table. My development server is being re-done and I'm not sure when it will be back on line. I have created some audit tables using triggers but they a simple inserts into a new table.
I believe the trigger statement should look like this, any advice would be appreciated.
Thanks in advance.
CREATE TRIGGER OB$1InsertAudit ON OB$0001 AFTER INSERT AS Begin Update AX$0001 Set AX$0001.receiveddate=getdate(), AX$0001.docreviewstatus=null where Ob$0001.objid=AX$0001.refobjid and OB$0001.recordtype=6 and OB$0001.descriptor in('Ancillary Services Tracking Form', 'FCE Referral Request', 'Remain At Work', 'Voc Rehab Job Retention')
If I have a situation where a particular SQL statement updates three records on a table and there is an Update trigger defined on the table, then how many times will the trigger fire - three times or one time.