Sure... why not?  Perhaps they're trying to make a delivery route list to cut down on the number of stops that a truck has to make where street numbers and...

Since you have no Primary Key, I have to agree... might be a problem with the table design a bit...

Anyway... take the query you posted...

select count(1)

from CL_PRODUCT_ATTRIBUTE

where ATRRIBUTE_ID = @attrib_id

and...

Ok, thanks... isn't that the way?  Company buys something and that won't let you use it which forces you to revert to undocumented features in SQL Server...

This will work in...

Thanks John... appreciate the feedback!

This is pretty cool... we have several complex, multi-step store procedures and some folks have put some pretty complex logging procedures in them just so they can query a...

How simple... how elegant... I never knew you could do this... nicely done.

We sure don't need a cursor for this and I'm not sure you could do it any faster or easier in the client ...

What... replace one form of RBAR (pronounced "ree-bar" and is a "Modenism" for "Row By Agonizing Row") in the form of a correlated sub-query for another form?  I don't know...

Sure... right after you tell us why you can't use xp_filesize so we don't recommend something similar...

Nice article, David, as usual...

As you are aware, I suspect that any performance differences will only occur where joins of different data types occur due to implicit conversions... but...

Yep...  you're supposed to know what the bad rows are BEFORE you begin the transfer.

Heh... my turn Remi... how come no Tally table here?  "I'm surprised you didn't think of it"

Here... shoe-horn a call to this function into my previous code to get rid of all but the required "pattern" of characters...

CREATE FUNCTION dbo.CleanData(@sText AS VARCHAR(8000),@pattern varchar(8000))

RETURNS VARCHAR(8000)

    ...

Thanks for the feedback Jules... I always wonder if anyone actually runs these things

Like I said, how long does it take on...

David...

What do you call "huge?"   How many denormalized rows are we talking about?  I work with a denormalized input from a million customer IVR and there are currently almost...