In that case, either an IIF statement or REGEX should help. (Google regex or IIF in ssis for details.)

You'd basically use a script component after your data import to remove...

You might want to search this site for "string concatenate for xml path".

I'm sure FOR XML PATH will do the trick.

Something like this?

SELECT CASE WHEN LEFT(YourCol,1) ='|' THEN STUFF(YourCol,1,1,'') ELSE YourCol END

I had a look at the Execution plan you posted.

It looks pretty straight forward.

If 50% of the time it takes is sorting 2 (two) rows in a table variable I...

tosscrosby (12/2/2010)

---

Pffftttt.....You must not be in the IT arena. Seems more like HR or accounting.....more like HR...

You could try

SUM(CASE WHEN [Year Purchased]>YEAR(DateAdd(Year, - 1, GetDate()))

THEN 1 ELSE 0 END) AS [1yr PO Line Items]

It might be a good idea...

You can simply expand your current CHECK constraint:

CHECK ((actionId IN(1,2) and associatedId is null) OR (actionId =3 and associatedId is not null))

As a side note: You should always use a...

If possible for you could you please post the basic concept of your sproc?

Reason: Having a single sproc with a 1000 update statements calls for a review. Either the sproc...

ALZDBA (11/30/2010)

---

You're getting the arithmatic overflow, because checksum generates:

declare @a int, @b-2 int, @c int

select @a=checksum(1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16)

, @b-2=checksum(10,2,3,4,5,6,7,8,9,1,11,12,13,14,15,16)

,@c=checksum(14,2,3,4,5,6,7,8,9,10,11,12,13,1,15,16)

-- Which one is closer to @a? @b-2 or @c?

-- I'm...

looks like a missing comma before the CASE statement.

I'd also recommend to add an alias to your CASE .. END. 😉

ALZDBA (11/30/2010)

---

did you experiment with CheckSum computed columns to detect perfect matches quickly ?

, RwCheckSum as checksum(col1, col2, col3, col4, col5, col6, col7, col8, col9, col10,...

Would it be possible to add a persisted computed column to each table based on col1*2^(1/8) + col2*2^(1/7)+...col16*2^7 or any other calculation that'd take the ordinal position into account and...

and here's the set based solution (I knew there would be at least one...)

CREATE TABLE #tempWeirdExcelList

( id INT IDENTITY(1,1),

list VARCHAR(50)

)

INSERT INTO #tempWeirdExcelList(list)

SELECT ' 1400.00,"" ...

Here's a *cough* loop version I just tried to solve that little puzzle. I'm more than confident there are much better solutions available...

Edit: a recursive CTE would be possible, too....

In general, both options you described can be used.

However, the xml sample you provided cannot be shredded into separate columns.

The reason is the multiple declaration of the default namespace. (xmlns="http://guidewire.com/typelists"...