Shoot... here we go... this is a perfect example of what I've spoken of in the past... You had to write a little VB...

Using ISNUMERIC to determine if something is all numeric digits is a form of "Death by SQL"...

IF IsNumeric(RIGHT(RTRIM('ABCDEF$-1,'), 4)) = 1

print 'yay'

else

print 'boo'

IF IsNumeric(RIGHT(RTRIM('ABCDEF0D01'), 4)) = 1

print 'yay'

else

print 'boo'

IF IsNumeric(RIGHT(RTRIM('ABCDEF23E2'),...

Although I agree that doing an ORDER BY in a view is a bit crazy, I'd have to disagree with the author of the blog because I tried the methods...

Sorry for my obvious post, Jason... I pulled to reply to Mick's post and got distracted... in that time, you answered and I didn't see it.

True but you have to be careful not to allow things so large that you have the potential for trying to store more than 8060 bytes (or less) in a...

Which one worked?  The merge or the reregister?

You cannot say SELECT @somevariable = * and get away with it...

While OPENROWSET is fine, it exposes passwords and logins... it would be far better to create a Linked Server.

I don't know if DTS has a way to modify that output but, if it can, replace the 1234567 in the following code with whatever DTS uses...

SELECT CONVERT(CHAR(8),DATEADD(ms,1234567,0),108)

Converts 1234567 milliseconds...

>>[Subquery returned more than 1 value] error

That usually has more to do with what you're doing in the code... you should post it so we can take a look for...

Peter used two "AND"s just due to a cut'n'paste error... remove one of the "AND"s and it'll probably work just fine.

Yes,

It "pukes" because you first compare a variable to null so it must first try to compare every record to a condition that does not exist.

Use the first form...

Although I have an extreme dislike for "sequence tables", this would be a good place for one.

If all you want to do is have GETDATE() return a time of midnight no matter what the time is, this will do...

SELECT DATEADD(dd,DATEDIFF(dd,0,GETDATE()),0)

I agree with JeffB... outer join, search for the nulls... haven't tested for speed but seems that it would be faster than either the correlated subquery or the concatenate method......

...and it's not just LEN... Try these in your order by...

'     '+CtrlCode

... or ...

REPLACE(CtrlCode,'T','')

Looks like a genuine Microsoft Bug... gee... that's never happened before... bet the boys in Redmond will...