This may also work for you.

DECLARE @word VARCHAR(MAX)

SET @word = 'ABCDEFG'

DECLARE @t TABLE (words VARCHAR(max))

INSERT INTO @t

SELECT 'ABCDEFG (9105) CDD ABCDEFG (1690) ENGINEERING'

UNION ALL SELECT 'ABCDEQG (9105) CDD ABCDEFG...

Normally I try to avoid self joins because they can be slow performance-wise with many rows in the table and there's usually another approach that is faster.

In this...

John,

That's exactly where I looked! And I can read and construct a working example too! 🙂

DECLARE @sql nvarchar(500)

DECLARE @AddItemVal INT

DECLARE @MyColumn NVARCHAR(25)

DECLARE @ParmDefinition nvarchar(500)

CREATE TABLE #myTable (Col1 INT, Col2...

There is no easy way to do what you're looking to do. Too bad, I'd love to do this myself.

The only way to query out columns dynamically would be...

The schemas don't need to be the same, but presumably there is some similarity in the columnsyou're updating to each table?

You can do all kinds of transformations on the OUTPUT...

From GPO's last suggestion, wouldn't this be expressed a little more elegantly as follows:

SELECT WEB.DIM

,s.[GROUP]

,COUNT(WEB.ACTIVITY) AS [COUNT]...

Possibly something like this may work for you:

DECLARE @t TABLE ([From] DATETIME, [To] DATETIME)

DECLARE @FromDateDATETIME

,@ToDateDATETIME

,@monthsINT

SELECT @FromDate = '2011-01-01', @ToDate = '2012-01-20'

SELECT @Months = 1 + DATEDIFF(month, DATEADD(month, DATEDIFF(month, 0, @FromDate),...

If you've got 12 tables to INSERT into within the transaction, you can try reducing the number of INSERT statements to 6 using the OUTPUT clause.

CREATE TABLE #T1 (num INT,...

You can also fudge round using this technique if you don't want to mess with your precision:

http://www.sqlservercentral.com/articles/Financial+Rounding/88067/

When I saw you had posted to this thread,...

Brubin,

I believe the following will get you what you want although I played a bit fast and loose with your problem description.

--Create a table to hold the data.

DECLARE @tbl_Data TABLE

(TimePunchID...

Just in case anyone wants to try to be my hero, this one is probably the one to beat because it has yet a better query plan cost than the...

And then of course, since someone just had to bring up that pesky question about performance, I simply had to offer something better than my original.

You see, what I was...

I also changed the ORDER BY on the PARTITION to be Change_Date instead of revision, because I noticed one case where change date decreased on an increasing revision number. ...

Dwains does pretty much exactly what needs to be done even with the additional business rules, with the exception that it errors in two places:

WA_Traffic Proposed should be 30-60 days,...