I want to update a column VIA insert trigger

I have a SQL query that creates a table, then creates triggers for the table. The triggers need to update a column in the new record from sums and counts in other tables. I'm pretty new to writing queries, so Im a bit confused. I just cannot seem to get the inserted record values to come up. I've tried using variabbles and using the inserted.values from the inserted table.
I need this working tomorrow so any help is appreciated.....

thanks
/* ---------- CREATE THE TABLE --------------------*/
CREATE TABLE [dwANALYZER].[CalData] (
	[F1] [smalldatetime] NULL ,
	[MasterAgencyID] [varchar] (20) COLLATE Latin1_General_BIN NULL ,
	[ProducerID] [varchar] (5) COLLATE Latin1_General_BIN NULL ,
	[Issue_State] [nvarchar] (50) COLLATE Latin1_General_BIN NULL ,
	[Category] [nvarchar] (50) COLLATE Latin1_General_BIN NULL ,
	[Line] [nvarchar] (50) COLLATE Latin1_General_BIN NULL ,
	[Form] [varchar] (4) COLLATE Latin1_General_BIN NULL ,
	[Origin] [nvarchar] (10) COLLATE Latin1_General_BIN NULL ,
	[Originator] [nvarchar] (80) COLLATE Latin1_General_BIN NULL ,
	[Policy_Type] [nvarchar] (50) COLLATE Latin1_General_BIN NULL ,
	[Quotes] [int] NULL ,
	[QuotePremium] [decimal](38, 2) NULL,
	[Issues] [int] NULL ,
	[IssuePremium] [decimal](38, 2) NULL,
	[Renewals] [int] NULL ,
	[RenewalPremium] [decimal](38, 2) NULL,
	[Cancels] [int] NULL ,
	[CancelPremium] [decimal](38, 2) NULL
) ON [PRIMARY]
GO
 
SET QUOTED_IDENTIFIER ON 
GO
SET ANSI_NULLS ON 
GO
 
/*----------------- create triggers here ----------------------*/
CREATE TRIGGER Q_INSERT ON CalData AFTER INSERT
AS
BEGIN
DECLARE @F1 smalldatetime 
DECLARE @MasterAgencyID varchar(20)
DECLARE	@ProducerID varchar(5) 
DECLARE	@Issue_State nvarchar(50)
DECLARE	@Category nvarchar (50)
DECLARE	@Line nvarchar (50) 
DECLARE	@Form varchar (4) 
DECLARE	@Origin nvarchar (10)
DECLARE	@Originator nvarchar (80) 
DECLARE	@Policy_Type nvarchar (50) 
 
SELECT 
@F1 = [F1],
@MasterAgencyID = [MasterAgencyID], 
@ProducerID = [ProducerID],
@Issue_State = [Issue_State], 
@Category = [Category],
@Line = [Line], 
@Form = [Form],
@Origin = [Origin], 
@Originator = [Originator],  
@Policy_Type = [Policy_Type] FROM inserted
 
 
/* -------ive tried it two ways ------------------------*/
 
UPDATE CalData  
Set [Quotes] = (SELECT     COUNT(pqi.POLICY_NUMBER) AS Policies
FROM         dbo.tblCalendarDate cal LEFT OUTER JOIN
                      dbo.A_PolicyQuoteAndIssueWithInspectionsPlusW1 pqi ON CONVERT(char(10), cal.F1, 102) = CONVERT(CHAR(10), pqi.QuoteDate, 102)
GROUP BY cal.F1, pqi.MasterAgencyID, pqi.ProducerID, pqi.Issue_State, pqi.Category, pqi.Line, pqi.Origin, pqi.Originator, pqi.Policy_Type, pqi.Form
HAVING     (cal.F1 = inserted.F1) AND (pqi.MasterAgencyID = inserted.MasterAgencyID) AND (pqi.ProducerID = inserted.ProducerID) AND (pqi.Issue_State = inserted.Issue_State) 
AND (pqi.Category = inserted.Category) AND (pqi.Line = inserted.Line) AND (pqi.Form = inserted.Form) 
AND (pqi.Origin = inserted.Origin) AND (pqi.Originator = inserted.Originator) AND (pqi.Policy_Type = inserted.Policy_Type))
/* -------------------------and this way as well ---------------*/
 
UPDATE CalData  
Set [QuotePremium] = (SELECT     SUM(pqi.IssuedPremium) AS Premium
FROM         dbo.tblCalendarDate cal LEFT OUTER JOIN
                      dbo.A_PolicyQuoteAndIssueWithInspectionsPlusW1 pqi ON CONVERT(char(10), cal.F1, 102) = CONVERT(CHAR(10), pqi.QuoteDate, 102)
GROUP BY cal.F1, pqi.MasterAgencyID, pqi.ProducerID, pqi.Issue_State, pqi.Category, pqi.Line, pqi.Origin, pqi.Originator, pqi.Policy_Type, pqi.Form
HAVING     (cal.F1 = @F1) AND (pqi.MasterAgencyID = @MasterAgencyID) AND (pqi.ProducerID = @ProducerID) AND (pqi.Issue_State = @Issue_State) 
AND (pqi.Category = @Category) AND (pqi.Line = @Line) AND (pqi.Form = @Form) 
AND (pqi.Origin = @Origin) AND (pqi.Originator = @Originator) AND (pqi.Policy_Type = @Policy_Type))
 
end

Open in new window

mannymsAsked:
Who is Participating?

[Product update] Infrastructure Analysis Tool is now available with Business Accounts.Learn More

x
I wear a lot of hats...

"The solutions and answers provided on Experts Exchange have been extremely helpful to me over the last few years. I wear a lot of hats - Developer, Database Administrator, Help Desk, etc., so I know a lot of things but not a lot about one thing. Experts Exchange gives me answers from people who do know a lot about one thing, in a easy to use platform." -Todd S.

Bob HoffmanDeveloperCommented:
The first update wont work because alias inserted is not known in that context. The second on should work with a fiew adjustment... see below
Also you can test in in Mgmt Studio by executing the select and supplying values for your variables.  see further below
Hope this helps

UPDATE CalData  
Set [QuotePremium] = (SELECT SUM(pqi.IssuedPremium) AS Premium
						FROM dbo.tblCalendarDate cal INNER JOIN dbo.A_PolicyQuoteAndIssueWithInspectionsPlusW1 pqi 
							ON CONVERT(char(10), cal.F1, 102) = CONVERT(CHAR(10), pqi.QuoteDate, 102)
						WHERE (cal.F1 = @F1) AND (pqi.MasterAgencyID = @MasterAgencyID) AND (pqi.ProducerID = @ProducerID) 
						AND (pqi.Issue_State = @Issue_State) AND (pqi.Category = @Category) AND (pqi.Line = @Line) AND (pqi.Form = @Form) 
						AND (pqi.Origin = @Origin) AND (pqi.Originator = @Originator) AND (pqi.Policy_Type = @Policy_Type))
 
 
 
 
--To test
SET @F1 = ?
SET @MasterAgencyID = ?
SET	@ProducerID = ?
SET	@Issue_State = ?
SET	@Category = ?
SET	@Line = ?
SET	@Form = ?
SET	@Origin = ?
SET	@Originator = ?
SET	@Policy_Type = ?
 
SELECT SUM(pqi.IssuedPremium) AS Premium
						FROM dbo.tblCalendarDate cal INNER JOIN dbo.A_PolicyQuoteAndIssueWithInspectionsPlusW1 pqi 
							ON CONVERT(char(10), cal.F1, 102) = CONVERT(CHAR(10), pqi.QuoteDate, 102)
						WHERE (cal.F1 = @F1) AND (pqi.MasterAgencyID = @MasterAgencyID) AND (pqi.ProducerID = @ProducerID) 
						AND (pqi.Issue_State = @Issue_State) AND (pqi.Category = @Category) AND (pqi.Line = @Line) AND (pqi.Form = @Form) 
						AND (pqi.Origin = @Origin) AND (pqi.Originator = @Originator) AND (pqi.Policy_Type = @Policy_Type)

Open in new window

mannymsAuthor Commented:
Dont I need the GROUP BY statement in order for the SUM and COUNT to work? (original code snippet line 63 & 73)

The variables come from the newly inserted row. Therefore the AFTER INSERT should supply them, No?
(original snippet line 44-54)
Anthony PerkinsCommented:
Replace <PrimaryKey> with the Primary Key or unique column in CallData:
CREATE TRIGGER Q_INSERT ON CalData 
 
AFTER INSERT
 
AS
 
BEGIN
 
UPDATE	c
Set		Quotes = d.Policies
FROM	CalData  c
		Inner Join (
			Select	i.<PrimaryKey>,
					COUNT(*) Policies
			From	dbo.tblCalendarDate cal
					Inner Join Inserted i On cal.F1 = i.F1
					Inner Join dbo.A_PolicyQuoteAndIssueWithInspectionsPlusW1 pqi ON cal.F1 = pqi.QuoteDate
											And i.MasterAgencyID = pqi.MasterAgencyID
											And i.ProducerID = pqi.ProducerID
											And i.Issue_State = pqi.Issue_State
											And i.Category = pqi.Category
											And i.Line = pqi.Line
											And i.Form = pqi.Form
											And i.Origin = pqi.Origin
											And i.Originator = pqi.Originator
											AND i.Policy_Type = pqi.Policy_Type
			Group By
					i.<PrimaryKey>
				) d On c.PrimaryKey = d.<PrimaryKey>
 
END

Open in new window

SolarWinds® VoIP and Network Quality Manager(VNQM)

WAN and VoIP monitoring tools that can help with troubleshooting via an intuitive web interface. Review quality of service data, including jitter, latency, packet loss, and MOS. Troubleshoot call performance and correlate call issues with WAN performance for Cisco and Avaya calls

Learn More
Anthony PerkinsCommented:
This might be easier to read:
CREATE TRIGGER Q_INSERT ON CalData

AFTER INSERT

AS

BEGIN
 
UPDATE      c
Set            Quotes = d.Policies
FROM      CalData  c
            Inner Join (
                  Select      i.<PrimaryKey>,
                              COUNT(*) Policies
                  From      dbo.tblCalendarDate cal
                              Inner Join Inserted i On cal.F1 = i.F1
                              Inner Join dbo.A_PolicyQuoteAndIssueWithInspectionsPlusW1 pqi ON cal.F1 = pqi.QuoteDate
                                                                  And i.MasterAgencyID = pqi.MasterAgencyID
                                                                  And i.ProducerID = pqi.ProducerID
                                                                  And i.Issue_State = pqi.Issue_State
                                                                  And i.Category = pqi.Category
                                                                  And i.Line = pqi.Line
                                                                  And i.Form = pqi.Form
                                                                  And i.Origin = pqi.Origin
                                                                  And i.Originator = pqi.Originator
                                                                  AND i.Policy_Type = pqi.Policy_Type
                  Group By
                              i.<PrimaryKey>
                        ) d On c.PrimaryKey = d.<PrimaryKey>

END
mannymsAuthor Commented:
I see that you have used <PrimaryKey> .
Is this a variable or are you telling me to enter some data here?
There is no primary key for this table as many of the matching fields will contain nulls.
Bob HoffmanDeveloperCommented:
You don't need a group by because you're only returing the one value SUM(pqi.IssuedPremium). The where clause will define which row make up the sum.
Yes, since the trigger is Insert After you will get the values when you do the select.
Anthony PerkinsCommented:
>>Is this a variable or are you telling me to enter some data here? <<
Did you not see my comment:
"Replace <PrimaryKey> with the Primary Key or unique column in CallData:"

>>There is no primary key for this table as many of the matching fields will contain nulls.<<
That is too bad.  Can you find some combination that is unique or failiing that add a unique column (an IDENTITY column for example)

Here is the problem with your approach:
Let's suppose that your INSERT statement adds 100 rows, then the following local variables will contain just one value (the last one in the Inserted logical table):
SELECT
@F1 = [F1],
@MasterAgencyID = [MasterAgencyID],
@ProducerID = [ProducerID],
@Issue_State = [Issue_State],
@Category = [Category],
@Line = [Line],
@Form = [Form],
@Origin = [Origin],
@Originator = [Originator],  
@Policy_Type = [Policy_Type] FROM inserted

The Trigger fires once per statement and NOT once per row inserted.
Anthony PerkinsCommented:
HBHoffman,

>>You don't need a group by because you're only returing the one value SUM(pqi.IssuedPremium).<<
How do you know? See my previous comments

>>Yes, since the trigger is Insert After you will get the values when you do the select.<<
That is only true if a single row is Inserted.  Again, read up on how Triggers really work.
mannymsAuthor Commented:
Ok, so maybe I'm going about this wrong.
I have the table created above that have fields from [F1] to [PolicyType] making the unique key. However they can contain Nulls. Each of the following fields are a COUNT or a SUM based on date JOINS to the tblCalendarDate table. I DO however have views for each of these calculations (Each view contains a COUNT and SUM, 4 views: Quotes, Issues, Cancels and Renewals) .
So maybe instead of a trigger, could I create the table and use a select statement for each of the fields that need calculations?

Here is the view SQL for quotes 
SELECT     TOP 100 PERCENT cal.F1, pqi.MasterAgencyID, pqi.ProducerID, pqi.Issue_State, pqi.Category, pqi.Line, pqi.Form, pqi.Origin, pqi.Originator, 
                      pqi.Policy_Type, COUNT(pqi.POLICY_NUMBER) AS Policies, SUM(pqi.IssuedPremium) AS Premium
FROM         dbo.tblCalendarDate cal LEFT OUTER JOIN
                      dbo.A_PolicyQuoteAndIssueWithInspectionsPlusW1 pqi ON CONVERT(char(10), cal.F1, 102) = CONVERT(CHAR(10), pqi.QuoteDate, 102)
GROUP BY cal.F1, pqi.MasterAgencyID, pqi.ProducerID, pqi.Issue_State, pqi.Category, pqi.Line, pqi.Origin, pqi.Originator, pqi.Policy_Type, pqi.Form
HAVING      (cal.F1 < GETDATE() - 1)
ORDER BY cal.F1 DESC, pqi.MasterAgencyID, pqi.ProducerID

Open in new window

mannymsAuthor Commented:
BTW
Here is the rest of the query. This is where it inserts the records

/*-----------------------------------------------------------------------------*/
INSERT INTO CalData
(F1,
MasterAgencyID ,
ProducerID,
Issue_State,
Category,
Line,
Form,
Origin,
Originator ,
Policy_Type 
)
SELECT
cal.F1,  
pqi.MasterAgencyID, 
pqi.ProducerID, 
pqi.Issue_State, 
pqi.Category, 
pqi.Line, 
pqi.Form, 
pqi.Origin, 
pqi.Originator, 
pqi.Policy_Type
FROM         
dbo.tblCalendarDate cal LEFT OUTER JOIN
dbo.A_PolicyQuoteAndIssueWithInspectionsPlusW1 pqi 
ON CONVERT(char(10), cal.F1, 102) = CONVERT(CHAR(10), pqi.QuoteDate, 102)
GROUP BY cal.F1, pqi.MasterAgencyID, pqi.ProducerID, pqi.Issue_State, pqi.Category, pqi.Line, pqi.Origin, pqi.Originator, pqi.Policy_Type, pqi.Form
HAVING      (cal.F1 > GETDATE() - 90)

Open in new window

Anthony PerkinsCommented:
>>Ok, so maybe I'm going about this wrong.<<
No necessarily.  Did you try my code?

>>I have the table created above that have fields from [F1] to [PolicyType] making the unique key. However they can contain Nulls. <<
It would certainly help if you had a primary key.  But the code I originally posted should also work.
mannymsAuthor Commented:
Is there a way to create the table and insert/replace Spaces instead of NULL. Wont that allow me to create a PRIMARY key?
And I did try your solution but without a primary key, It wont work.
Anthony PerkinsCommented:
>>And I did try your solution but without a primary key, It wont work.<<
How do you know?  I did not say you had to ahve one, I said it would help.
mannymsAuthor Commented:
Onlines 13, 28 and 29 you reference <PrimaryKey> should I just reference the Columns that are unique?
Since there was no Primary key I didnt replace them with anything.

Can you post how you would write it without the primary key reference?
mannymsAuthor Commented:
One more thing, Is this going to work on all the records?
You said a trigger only works when a single row is inserted. Does a "Select into" qualify as multiple inserts?
I'm back at the office this morning (and overdue obviously). I'll try all of your suggestions and update you.
Thanks for all the time with this.

Anthony PerkinsCommented:
>>Can you post how you would write it without the primary key reference?<<

CREATE TRIGGER Q_INSERT ON CalData

AFTER INSERT

AS

BEGIN
 
UPDATE      c
Set      Quotes = d.Policies
FROM      CalData  c
      Inner Join (
          Select  i.MasterAgencyID,
                i.ProducerID,
                i.Issue_State,
                i.Category,
                i.Line,
                i.Form,
                i.Origin,
                i.Originator,
                i.Policy_Type,
                COUNT(*) Policies
          From    dbo.tblCalendarDate cal
                Inner Join Inserted i On cal.F1 = i.F1
                Inner Join dbo.A_PolicyQuoteAndIssueWithInspectionsPlusW1 pqi ON cal.F1 = pqi.QuoteDate
                        And ISNULL(i.MasterAgencyID, '') = ISNULL(pqi.MasterAgencyID, '')
                        And ISNULL(i.ProducerID, '') = ISNULL(pqi.ProducerID, '')
                        And ISNULL(i.Issue_State, '') = ISNULL(pqi.Issue_State, '')
                        And ISNULL(i.Category, '') = ISNULL(pqi.Category, '')
                        And ISNULL(i.Line, '') = ISNULL(pqi.Line, '')
                        And ISNULL(i.Form, '') = ISNULL(pqi.Form, '')
                        And ISNULL(i.Origin, '') = ISNULL(pqi.Origin, '')
                        And ISNULL(i.Originator, '') = ISNULL(pqi.Originator, '')
                        AND ISNULL(i.Policy_Type, '') = ISNULL(pqi.Policy_Type, '')
          Group By
                i.MasterAgencyID,
                i.ProducerID,
                i.Issue_State,
                i.Category,
                i.Line,
                i.Form,
                i.Origin,
                i.Originator,
                i.Policy_Type
          ) d On ISNULL(i.MasterAgencyID, '') = ISNULL(d.MasterAgencyID, '')
                And ISNULL(i.ProducerID, '') = ISNULL(d.ProducerID, '')
                And ISNULL(i.Issue_State, '') = ISNULL(d.Issue_State, '')
                And ISNULL(i.Category, '') = ISNULL(d.Category, '')
                And ISNULL(i.Line, '') = ISNULL(d.Line, '')
                And ISNULL(i.Form, '') = ISNULL(d.Form, '')
                And ISNULL(i.Origin, '') = ISNULL(d.Origin, '')
                And ISNULL(i.Originator, '') = ISNULL(d.Originator, '')
                AND ISNULL(i.Policy_Type, '') = ISNULL(d.Policy_Type, '')
          END

>>Is this going to work on all the records?<<
That was the point I was trying to make previously (#24365600) and why you have to join using the logical table Inserted as opposed to assiging the values to local variables.

But whether it ultimately "works" I have no clue.  To be totally honest, I cannot follow entirely what you are trying to achieve.


Experts Exchange Solution brought to you by

Your issues matter to us.

Facing a tech roadblock? Get the help and guidance you need from experienced professionals who care. Ask your question anytime, anywhere, with no hassle.

Start your 7-day free trial
mannymsAuthor Commented:
OK So I modified the Code to use the unique columns instead of a primary key. it did not work.  Attached is zip file of SQL. (2a)
A Second option is that I have views that have already summed the counts and sums that only require a single row be selected to update the inserted record. I tried to replicate the structure you've used but it hangs for a very long time. See Zip (2b)
The view is created by the code I posted 6-7 messages back.
2a-create-caldata-table.zip
2b-create-caldata-table.zip
mannymsAuthor Commented:
OK here is the objective:
I have two tables
1) A table of calendar dates (cal), containing all the dates from 1/1/2006 to 12/31/2012
2) a table (PQI) that contains the following:
MasterAgencyID, ProducerID, Issue_State, Category, Line, Form, Origin, Originator, Policy_Type, QuoteDate, IssueDate, EffDate, CancelDate, Premium

I need to create a single table that groups the first 9 fields uniquelly, Counts the records and Sums the premium, for each of the Dates by way of the JOIN to the Calendar table. (therfore 4 updates) Naturally the dates will all be different therefore the Counts and Sums will be different.

Thast why I'm only trying to get the first one to work, figureing I can replicate the code to use the other dates once the first works.

I hope this helps

Thanks again.

I've tried the last code you sent and it cannot find the reference "i"
Anthony PerkinsCommented:
>>it did not work.<<
Unfortunately that tell me nothing.

I suggest you ask for a refund and re-post.  Hopefully someone will step up to the plate.

Good luck.
mannymsAuthor Commented:
It's hard to explain a problem sometimes, so I can see how it's even harder to explain a solution to a problem that hasnt been communicated completely.  Even though it wasn't the answer I was looking for, you were thorough and helpful. It helped me understand how triggers work a bit better and, most of all, it helped me realize I was going about it wrong and made me look into other options.  I actually got it done with cursors and updates/inserts. So you deserve the points for all your help. Thank you very much.
It's more than this solution.Get answers and train to solve all your tech problems - anytime, anywhere.Try it for free Edge Out The Competitionfor your dream job with proven skills and certifications.Get started today Stand Outas the employee with proven skills.Start learning today for free Move Your Career Forwardwith certification training in the latest technologies.Start your trial today
Microsoft SQL Server

From novice to tech pro — start learning today.