I think that table variables might give you what you need - they can be passed as parameters to/from stored procedures - try reading the on-line help and see how...

What do you mean by 'linked table'?  Are you talking about a table residing in a linked server?  Or perhaps you're using Access and have linked to a SQL Server...

I still think that the error is happening before processing gets to the

Set Excel_WorkBook = Excel_Application.Workbooks.Open(sFilename)

line.  To prove it one way or the other, try changing sFilename to...

Why do the CONVERT when the ROUND works without it?

No problem.  The other thing you should consider doing is creating a lookup table:

DayNo (int, PK), Day (varchar(10))

and then populating it as follows:

1     Monday

2     Tuesday

........

7     Sunday

Then you can just do...

This can be achieved in many fewer lines, try this:

declare @days char(100), @classday varchar(10), @dayno int

set @days = '         Monday    Tuesday   Wednesday Thursday  Friday    Saturday  Sunday'

set @classday = 'Sunday'

set @dayno...

Eh?  Was this supposed to be part of another thread?  If not: "Which stored procedure?"

If your question is just: "Can stored procedures be called from VB6?", the answer is yes. ...

Sorry if this is obvious, but is Excel installed on the SQL Server server itself (if you know what I mean)?

Your error is happening at the CreateObject line, before the...

But I've just eaten 3 pieces of fried chicken & need to remove the grease somehow - not on my screen either

A fixed-length file has ALL records the same length!  Yours does not, so it is not.

If you've got less than 65,535 rows you can do it in Excel in 1...

Here's another way that might be faster as it's not doing the CONVERTs

SELECT floor(0.4799608993157 * 1000) / 1000

Following your own explanation, surely the formula is

UNION ALL = UNION + duplicates

?

Somethink like this might be what you need:

ALTER DATABASE <dbname> SET SINGLE_USER WITH ROLLBACK IMMEDIATE

go

Regards

I don't think that this is enough detail.

Can you also possibly include some sample source data and what you are seeing - highlighting the spurious rows.

Regards

OK - can you provide a small sample of the data you have and what you want to be returned.  No need to include all of the existing GROUP BY...