Example:

create table table_name

(a datetime

,b datetime

,CONSTRAINT constraint_name CHECK ( a > b )

)

If you are talking to a text file through ADO, then you are using an OLE DB (or ODBC) driver for text files. In a way, you are using the...

This would work:

DECLARE @NAME varchar(30)

SELECT EXAMINER_ID

FROM USER_GROUP

WHERE PATINDEX('%' + @NAME + '%',EXAMINER_ID) > 0

Michael

To accomplish this is one SELECT statement, something like this would work:

SELECTSUM( CASE iDEAS_sTATUS WHEN 'implemented' THEN 1. ELSE 0. END ) / COUNT(*)

FROMIdeas

WHEREdateSubmitted BETWEEN ...

Edited by - mromm on...

The other thing that might cause slow changes to the table is if you have foreign key defined on the table.

Also, if this is OLTP, why so many indexes?

Those...

Multiple instances of SQL Server on one machine is a feature of SQL Server 2000 (already forgot if it was in 7). Even then it should not be used in...

I think you can do it without cursor in one table scan like this (only replace NULL values with, say, '' using ISNULL function):

SELECT PersonId

, CASE t1.FirstName WHEN...

May be MSSQLSERVER service is not running? Go to the Administrative Tool --> Services and check there.

Ron, I realize that about clustered indexes. I only meant there is the option but, of course, you use it only with non-clustered indexes.

Just make sure you do not get too excited about UDF. It is a good replacement for parameterized views not supported by SQL Server. However if you start putting UDF...

Good idea: eBay for database contracts with the bids for hourly rate strting high and going down... Just kidding.

There is a version column in the same table. If you record the current version, you can monitor if it was changed. It inrements every time you alter the object....

You can create a variable of the type TABLE only in SQL Server 2000, not in version 7.

Edited by - mromm on 03/05/2003 2:30:01 PM

Consider replacing a stored procedure with a user-defined function that returns a table. It may satisfy your need. Something like this:

CREATE FUNCTION fn_abc ( <parameter list> )

AS TABLE

BEGIN

...

END

I think there is no performance difference because a unique constraint is implemented in MS SQL Server through a unique index creation. In the same manner, a PRIMARY KEY constraint...