Here's a template... I'll be waiting for your questions .

IF Object_id('Split') > 0

DROP FUNCTION SPLIT

GO

CREATE FUNCTION [dbo].[Split] (@vcDelimitedString nVarChar(4000),

@vcDelimiternVarChar(100) )

/**************************************************************************

DESCRIPTION: Accepts a delimited string...

Care to enumerate what they all mean??

You could probabely use dts o transfer data insted of looping... will make the whole thing much faster.

make the tTransToDate the clustered index (and primary if possible), and see how you're query will speed up .

SELECT mile_code,

AVG(Cast(CompAct as decimal(18,2)))

FROM opvw_actPerMile

GROUP BY mile_code

ORDER BY mile_code

maybe you'd be better off using DTS bulk insert task, and just start that instead. You wouldn't hog your application for hours that way...

Why not do it all client side?

I'd say... Did you put additional fail-safe on that switch???

Can you tell us what you're trying to do?

Unless you really wanna piss off some dbas .

I'll still stick to Noeld's solution... extremely simple and easy to understand.. Mine is just required if you query this field often and need performance.

Finally one good utilisation of dynamic sql .

Group by or order by???

Can we see the table definition and some sample data with the expected results from the query?

Is sp_renamedb forbiden too?

Time to think about it at least...