Creating a trigger for the row being updated

Hello group,
I have a question regarding an update trigger.
For example, I have the following fields in a table..
ItemCode
ShortDescription
Quantity
UpdatedDate
UpdatedBy
I want to create a trigger that updates the "UpdatedDate" and
"UpdatedBy" fields when either the "ShortDescription" or "Quantity"
fields get updated.
I have figured out how to use the "bitwise" operators so as to target
only columns 2 and 3 for chages, however, telling the trigger to only
update the records which has been changed is proving to be a
challenge.
Thanks in advance
Leenux_tuxTry:
update m
set
UpdatedDate = getdate ()
, UpdatedBy = CURRENT_USER
from
MyTable m
join
inserted i on i.ItemCode = m.ItemCode
join
deleted d on d.ItemCode = i.ItemCode
where
i.ShortDescription <> d.ShortDescription
or i.Quantity <> d.Quantity
--
Tom
----
Thomas A. Moreau, BSc, PhD, MCSE, MCDBA, MCITP, MCTS
SQL Server MVP
Toronto, ON Canada
https://mvp.support.microsoft.com/profile/Tom.Moreau
"leenux_tux" <nigel.henderson@.connecta.ltd.uk> wrote in message
news:1181137271.655691.191540@.m36g2000hse.googlegroups.com...
Hello group,
I have a question regarding an update trigger.
For example, I have the following fields in a table..
ItemCode
ShortDescription
Quantity
UpdatedDate
UpdatedBy
I want to create a trigger that updates the "UpdatedDate" and
"UpdatedBy" fields when either the "ShortDescription" or "Quantity"
fields get updated.
I have figured out how to use the "bitwise" operators so as to target
only columns 2 and 3 for chages, however, telling the trigger to only
update the records which has been changed is proving to be a
challenge.
Thanks in advance
Leenux_tux|||leenux_tux,
Assuming that ItemCode is an unchanging primary key:
UPDATE M
SET UpdatedDate = GETDATE(), UpdatedBy = SUSER_SNAME()
FROM MyTable M
JOIN inserted i
ON M.ItemCode = i.ItemCode
The inserted table contains inserted or updated rows and the deleted table
contains deleted rows. Experience says that this trigger may actually take
more time to run for a single row than the original update ran.
If you are updating the row through a stored procedure (recommended) or
through other controllable code, considering directly supplying the data and
name in the initial update rather than having the trigger do clean up.
RLF
"leenux_tux" <nigel.henderson@.connecta.ltd.uk> wrote in message
news:1181137271.655691.191540@.m36g2000hse.googlegroups.com...
> Hello group,
> I have a question regarding an update trigger.
> For example, I have the following fields in a table..
> ItemCode
> ShortDescription
> Quantity
> UpdatedDate
> UpdatedBy
> I want to create a trigger that updates the "UpdatedDate" and
> "UpdatedBy" fields when either the "ShortDescription" or "Quantity"
> fields get updated.
> I have figured out how to use the "bitwise" operators so as to target
> only columns 2 and 3 for chages, however, telling the trigger to only
> update the records which has been changed is proving to be a
> challenge.
> Thanks in advance
> Leenux_tux
>|||The problem with this code is that is does not exactly meet the spec:
'I want to create a trigger that updates the "UpdatedDate" and "UpdatedBy"
fields when either the "ShortDescription" or "Quantity" fields get updated.'
IOW, if both of these columns are set to their original values, the you are
updating the UpdatedDate and UpdatedBy columns, which you don't want to do.
You need to join also on the deleted virtual table and compare the values in
inserted vs. deleted.
--
Tom
----
Thomas A. Moreau, BSc, PhD, MCSE, MCDBA, MCITP, MCTS
SQL Server MVP
Toronto, ON Canada
https://mvp.support.microsoft.com/profile/Tom.Moreau
"Russell Fields" <russellfields@.nomail.com> wrote in message
news:uO23zPEqHHA.3948@.TK2MSFTNGP05.phx.gbl...
leenux_tux,
Assuming that ItemCode is an unchanging primary key:
UPDATE M
SET UpdatedDate = GETDATE(), UpdatedBy = SUSER_SNAME()
FROM MyTable M
JOIN inserted i
ON M.ItemCode = i.ItemCode
The inserted table contains inserted or updated rows and the deleted table
contains deleted rows. Experience says that this trigger may actually take
more time to run for a single row than the original update ran.
If you are updating the row through a stored procedure (recommended) or
through other controllable code, considering directly supplying the data and
name in the initial update rather than having the trigger do clean up.
RLF
"leenux_tux" <nigel.henderson@.connecta.ltd.uk> wrote in message
news:1181137271.655691.191540@.m36g2000hse.googlegroups.com...
> Hello group,
> I have a question regarding an update trigger.
> For example, I have the following fields in a table..
> ItemCode
> ShortDescription
> Quantity
> UpdatedDate
> UpdatedBy
> I want to create a trigger that updates the "UpdatedDate" and
> "UpdatedBy" fields when either the "ShortDescription" or "Quantity"
> fields get updated.
> I have figured out how to use the "bitwise" operators so as to target
> only columns 2 and 3 for chages, however, telling the trigger to only
> update the records which has been changed is proving to be a
> challenge.
> Thanks in advance
> Leenux_tux
>|||Tom,
You are correct. My mistake.
Of course, if Item_Code is NOT an unchanging primary key, then things still
get interesting.
RLF
"Tom Moreau" <tom@.dont.spam.me.cips.ca> wrote in message
news:%23qS6J8HqHHA.1216@.TK2MSFTNGP03.phx.gbl...
> The problem with this code is that is does not exactly meet the spec:
> 'I want to create a trigger that updates the "UpdatedDate" and "UpdatedBy"
> fields when either the "ShortDescription" or "Quantity" fields get
> updated.'
> IOW, if both of these columns are set to their original values, the you
> are
> updating the UpdatedDate and UpdatedBy columns, which you don't want to
> do.
> You need to join also on the deleted virtual table and compare the values
> in
> inserted vs. deleted.
> --
> Tom
> ----
> Thomas A. Moreau, BSc, PhD, MCSE, MCDBA, MCITP, MCTS
> SQL Server MVP
> Toronto, ON Canada
> https://mvp.support.microsoft.com/profile/Tom.Moreau
>
> "Russell Fields" <russellfields@.nomail.com> wrote in message
> news:uO23zPEqHHA.3948@.TK2MSFTNGP05.phx.gbl...
> leenux_tux,
> Assuming that ItemCode is an unchanging primary key:
> UPDATE M
> SET UpdatedDate = GETDATE(), UpdatedBy = SUSER_SNAME()
> FROM MyTable M
> JOIN inserted i
> ON M.ItemCode = i.ItemCode
> The inserted table contains inserted or updated rows and the deleted table
> contains deleted rows. Experience says that this trigger may actually
> take
> more time to run for a single row than the original update ran.
> If you are updating the row through a stored procedure (recommended) or
> through other controllable code, considering directly supplying the data
> and
> name in the initial update rather than having the trigger do clean up.
> RLF
> "leenux_tux" <nigel.henderson@.connecta.ltd.uk> wrote in message
> news:1181137271.655691.191540@.m36g2000hse.googlegroups.com...
>> Hello group,
>> I have a question regarding an update trigger.
>> For example, I have the following fields in a table..
>> ItemCode
>> ShortDescription
>> Quantity
>> UpdatedDate
>> UpdatedBy
>> I want to create a trigger that updates the "UpdatedDate" and
>> "UpdatedBy" fields when either the "ShortDescription" or "Quantity"
>> fields get updated.
>> I have figured out how to use the "bitwise" operators so as to target
>> only columns 2 and 3 for chages, however, telling the trigger to only
>> update the records which has been changed is proving to be a
>> challenge.
>> Thanks in advance
>> Leenux_tux
>|||That emphasizes the fact that a PK should never change. There are
exceptions, though.
--
Tom
----
Thomas A. Moreau, BSc, PhD, MCSE, MCDBA, MCITP, MCTS
SQL Server MVP
Toronto, ON Canada
https://mvp.support.microsoft.com/profile/Tom.Moreau
"Russell Fields" <russellfields@.nomail.com> wrote in message
news:uh5I%23VIqHHA.3892@.TK2MSFTNGP05.phx.gbl...
Tom,
You are correct. My mistake.
Of course, if Item_Code is NOT an unchanging primary key, then things still
get interesting.
RLF
"Tom Moreau" <tom@.dont.spam.me.cips.ca> wrote in message
news:%23qS6J8HqHHA.1216@.TK2MSFTNGP03.phx.gbl...
> The problem with this code is that is does not exactly meet the spec:
> 'I want to create a trigger that updates the "UpdatedDate" and "UpdatedBy"
> fields when either the "ShortDescription" or "Quantity" fields get
> updated.'
> IOW, if both of these columns are set to their original values, the you
> are
> updating the UpdatedDate and UpdatedBy columns, which you don't want to
> do.
> You need to join also on the deleted virtual table and compare the values
> in
> inserted vs. deleted.
> --
> Tom
> ----
> Thomas A. Moreau, BSc, PhD, MCSE, MCDBA, MCITP, MCTS
> SQL Server MVP
> Toronto, ON Canada
> https://mvp.support.microsoft.com/profile/Tom.Moreau
>
> "Russell Fields" <russellfields@.nomail.com> wrote in message
> news:uO23zPEqHHA.3948@.TK2MSFTNGP05.phx.gbl...
> leenux_tux,
> Assuming that ItemCode is an unchanging primary key:
> UPDATE M
> SET UpdatedDate = GETDATE(), UpdatedBy = SUSER_SNAME()
> FROM MyTable M
> JOIN inserted i
> ON M.ItemCode = i.ItemCode
> The inserted table contains inserted or updated rows and the deleted table
> contains deleted rows. Experience says that this trigger may actually
> take
> more time to run for a single row than the original update ran.
> If you are updating the row through a stored procedure (recommended) or
> through other controllable code, considering directly supplying the data
> and
> name in the initial update rather than having the trigger do clean up.
> RLF
> "leenux_tux" <nigel.henderson@.connecta.ltd.uk> wrote in message
> news:1181137271.655691.191540@.m36g2000hse.googlegroups.com...
>> Hello group,
>> I have a question regarding an update trigger.
>> For example, I have the following fields in a table..
>> ItemCode
>> ShortDescription
>> Quantity
>> UpdatedDate
>> UpdatedBy
>> I want to create a trigger that updates the "UpdatedDate" and
>> "UpdatedBy" fields when either the "ShortDescription" or "Quantity"
>> fields get updated.
>> I have figured out how to use the "bitwise" operators so as to target
>> only columns 2 and 3 for chages, however, telling the trigger to only
>> update the records which has been changed is proving to be a
>> challenge.
>> Thanks in advance
>> Leenux_tux
>

Creating "columns" from transaction data

Hi,
I have a transaction table that basically has the following fields
RecId, PeriodId, Quantity (a single RecId can have multiple records, i.e.
quantities in multiple periods)
I need to convert an entire table of these records to one that looks like
this...
RedId , P1Qty, P2Qty, P3Qty, P4Qty etc...
Which has one row per RecId and places the quantity (quantities) in the
appropriate "period" column(s) based on the value(s) of "Period" in
the transaction file for each record.
I've done this before in Access, using the IIF function for each of P1...P4
columns (IIF(Period = 1, Quantity, 0), IIF(Period = 2, Quantity, 0) for
each of the columns of the derived table I was making. This doesn't seem to
work for SQL Server. IIF exists, but I can't get the computed columns to
work properly (Syntax error near "=").
So, I've thought about...
1) Use CreateTable to create my derived table with periods as columns,
2) Write a series of INSERT queries that reads the transaction file for
each possible individual value for "Period" and populates the appropriate
column in the derived table
3) Sum the derived table on every column by RecId
4) Run the whole batch as the SelectCommand of my DataAdapter. The last
command in the batch is Select * from DerivedTable and this is the table
that the DataSet gets.
There has to be a better way to do this?
Thanks.
BBMYou can use CASE expressions instead of IIF. But why would you ever create a
table like this? What you are asking for is a report not a table. Any
reporting tool will construct a cross tab report for you.
David Portas
SQL Server MVP
--|||Do you have any idea what First Normal Form is? You might want to
learn about RDBMS before you write any code.|||Thanks David, CASE was just what I'm looking for.
In this instance, this result set is used as one of the tables in
multi-table DataSet used on a fairly complex display.
Thanks again.
BBM
"David Portas" wrote:

> You can use CASE expressions instead of IIF. But why would you ever create
a
> table like this? What you are asking for is a report not a table. Any
> reporting tool will construct a cross tab report for you.
> --
> David Portas
> SQL Server MVP
> --
>
>|||Yes, in fact I do. I simplified the underlying data structure in my questio
n
to hopefully make it easier to reply to. I was only using the "extra" table
,
because I couldn't figure out how to get the result set I wanted in one pass
.
Thanks for your response anyway.
"--CELKO--" wrote:

> Do you have any idea what First Normal Form is? You might want to
> learn about RDBMS before you write any code.
>

Create View with 2 Select Queries

Hi,

I don't even know if this is possible, but here goes.

I have 2 select queries.

1) SELECT w.Item AS [WareHouse Item], w.Quantity AS [WareHouse Quantity],

w.RestockLevel AS [WareHouse Restock], w.ReorderPoint AS [WareHouse ReorderPoint],

SUM(Wp.QuantityOrdered - Wp.QuantityReceivedToDate) AS WareHouseOnOrder

FROM WH.dbo.PurchaseOrderEntry AS wp INNER JOIN

WH.dbo.Item AS w ON wp.ItemID = w.ID INNER JOIN

WH.dbo.PurchaseOrder AS Wpo ON Wp.PurchaseOrderID = WPO.ID

Where (wPO.POType < 2) AND (wPO.Status = '0')

Group By w.Item, w.Quantity, w.RestockLevel, w.ReorderPoint

Order By w.Item

My socond query is the same, but from a different database on the same Server. If I join the 2 queries I get duplicate Items in my results, because of the 2 databases.

My question is:

Can I run both queries to create one View so that I can create my Crystal report from that?

Any better Ideas would be appretiated!

Thanks

what are you joining the two queries on?|||

You can use UNION. (If the column structure and order is the same.)

SELECT ... FROM dbo.MyTable WHERE ...

UNION

SELECT ... FROM MyOtherDatabase.dbo.MyTable WHERE ...

|||

Arnie Rowland wrote:

You can use UNION. (If the column structure and order is the same.)

SELECT ... FROM dbo.MyTable WHERE ...

UNION

SELECT ... FROM MyOtherDatabase.dbo.MyTable WHERE ...

or "UNION ALL" if you want the duplicates.

|||

Thanks Arnie and Michael

That was exactly what I needed.

Thanks!!!

|||

Hi,

I am not sure if view is your requirement or not. Aa far as the data from query to a report is concerned you can you use a temp table in stored procedure to do that or use UNION to combine these results.

Thanks,

Paraclete

|||

That is also VERY true.

After seeing that I can actually get all the information using the union that Arnie suggested I will probably create a stored procedure to run my select Statement.

Thanks Again!!