Yes... why have you defined both a delimiter and a row teminator when you are using a format file?

Try a single select with outer joins to MyTable...

So, you mean having all of my data for each database in a single name/value table really won't work unless I partition across multiple databases using views?  Dang...   

Heh... yeaup... and this time, the newbie started it

Can't flame if there's no fuel so I'm gonna let William think what he wants...

You're kidding...right?

Yes, Kavita... make it a DateTime datatype and most of your problems go away.  Convince your boss that it's the right thing to do.

No... but you could place the results of the proc in a temp table...

Heh... now THAT's important... so, lemme ask... is Patient_ID an autonumbering column (or, at least, an incrementing sequence number)?

Heh... just can't bring myself to do that, Serqiy... that would mean I'd have to demo saving the date as an integer that looks like an ISO date or storing...

If it's "inline" code, as you say, I've previously told the OP's to use the DATEADD/DATEDIFF method to ensure that there's no chance of an "anomoly" like what appeared in...

Bill,

I absolutely agree with writing consistant code and some of the methods to make the code consistant and easy.  My only problems with such a UDF are (putting aside speed for...

Not quite what I was after, Peter.  Sorry I wasn't clear... was looking to compare the dates in two tables for rows having occurred on the same day.

Serqiy reminded me...

You must remember that issue with implicit conversion of 0 to date which made that function nondeterministic.

THAT's what...

Now... THAT would be quite the trick... anyone know a way to get "date only" and still be able to use an Index SEEK?

Dang... sorry Peter... I keep forgetting to post the results...

===== DateDiff/DateAdd method ======

DBCC execution completed. If DBCC printed error messages, contact your system administrator.

DBCC execution completed. If DBCC...