That's actually a pretty good plan and I agree with that... nice to see someone to spend a little time up front to save wads later on.

Exactly... (and great explanation, by the way)   I know that and you know that and even some other folks know that.  And, you're absolutely...

1.  Doesn't matter if the header doesn't have the same number of delimiters.  BCP is picky that way.

2.  That's because of what I said before... you must delimit the end...

Man, I gotta remember to check for those...

Not sure what you're calling a base table but the correct covering indexes will allow for full length table access using index seeks across all tables in a join.  By...

Sorry Tomm... was just trying to lighten things up a bit.  And, maybe I'm taking you the wrong way, but "flame war" about what?  It takes two and I'm out...

Cool.  Thanks for the feedback

Heh... and it's true, too!

First problem is the header record...

NEATCM|8.07.00|867850731

...which does not have the same number of delimiters as the other rows.  Will always throw an error or cause a skip of the first...

Heh... I asked the same thing... and the answer is... crappola 3rd party software. 

So, Mike, you never hardcode anything?

Couldn't have said it better myself... thanks Steve.

You're not... the DEL command is being executed by xp_CmdShell and the >> is just part of any command line as a redirection of output from whatever command is on...

Wrong... correctly formulated covering indexes will be used no matter how much of the correct columns are used... this example uses 90 % of the data in two columns and...

If you're users want real time data when they open the spreadsheet, don't export the data as an "Excel file"... create a spreadsheet that uses "External Data" from a view. ...

SQL Portable?  Maybe for the GUI world but certainly not in the batch world.  For example, define some variables in SQL Server and do the same thing in Oracle... now, try...