Timeless jokes are always the best. Thanks, Steve!

And, sorry about the triple spaced code... the IFCODE's used to work correctly and they don't seem to, anymore... can't even color the damned errors anymore... they really messed something...

I'm thinking there's a couple of bugs... I have a table that looks like this...

--===== Create and populate a 1,000,000 row test table.

-- Column RowNum...

Several hundred developers? What you really need is a good set of development standards... 😉

Pass it forward, Jon... post the code you solved it with.

And, by the way, saving a timestamp (datetime) in a table as any character or integer based data is death...

Spot on! 🙂

Phil's method only works if the ID is contiguous and sequential.  However, you can make that true if you copy the data you want into a temp table that has...

SELECT

(SELECT COUNT(*) FROM A) AS [Count From A],

(SELECT COUNT(*) FROM B) AS [Count From B]

Heh... now that's funny.  Near his avatar, it says that Phil Factor is a "Rookie" 😀

If you take a look at the FROM clause in Books Online, you'll see that using a variable is NOT one of the options for a table name... in order...

p.s.  The best way to fix this is to pummel the vendor providing the data into submitting properly formatted data.  The first row should be in a separate "control" file...

Victoria,

The fastest way to do this is to build a BCP format file and then use Bulk Insert to import the data into a table.  Of course, all that information...

And, how do you do THAT? :Whistling:

Heh... let's see your "low knowledge" do it 😉

Yeaup... quite the mess.

This could be done all in T-SQL by importing the whole row into a wide column and doing a split (using a function or some clever code)...