Question

I have a stored procedure, which contains a simple select statement:

        ALTER PROCEDURE [dbo].[TransferAuditRecords] 
As
SET NOCOUNT ON
SET XACT_ABORT ON

Declare @UserCode As varchar(50)

DECLARE AuditCursor CURSOR FOR
        select top 1 UserCode from [AuditDatabaseServer].AuditDatabase.dbo.dbaudit where auditdate >= '2012-09-04'  
        Open AuditCursor
        FETCH NEXT FROM AuditCursor INTO @UserCode
        WHILE @@FETCH_STATUS = 0  
            BEGIN
                Print @Usercode
                FETCH NEXT FROM AuditCursor INTO @UserCode
            END

    Close AuditCursor 
    Deallocate AuditCursor 

There are a few lines of code missing, which I excluded as they are irrelevant to the question.

If I execute the SQL statement without the stored procedure i.e. in SQL Studio Manager 2005 then I get a different output than if I run the stored procedure i.e. the top reference returned is different. Why are the outputs different?

I am fairly sure that the reason for this is because SQL Server uses a different execution plan for compiled code is comparison to code being run in SQL Studio Manager. I wanted to check though.

Answer 1

Use an ORDER BY clause to predictably indicate which row is selected.

e.g. select top 1… order by auditdate

Unrelated:: is the use of a cursor required by part of the code you removed? I guess so, otherwise a simpler “select top 1 @UserCode = UserCode from…” would be enough.