I use TRY/CATCH and looks like Mel Sansone[/url] finds it useful too...

You need to use an aggregate function e.g.

SELECT Agent

,[Jan]

,[Feb]

FROM (SELECT *

...

For those on SSIS 2008: you might consider using the MERGE statement offering you less work and better performance.

Basically, your package will look like this:

Data Flow Task: Read Data From...

Here's a great article to get you started.

Use an Execute SQL Task to read the list with filenames (from a table) and put the result set in a variable of type Object (you can already filter on...

Did you try sp_recompile?

Take the Estimated Execution Plan (rather than the Actual Execution Plan) to begin with.

Jason,

You will find some information on the conditional split transformation (CST) in BOL.

You don't have to use an output downstream of the CST, so basically those rows won't go anywhere...

UPDATE mytable

SET gender = CASE WHEN gender = 'F' THEN 'M'

...

You need to do an outer join with a table that contains all dates on which you want to report.

Check this topic for some inspiration.

Assuming the identity column in your table is called version_id, your SP will look like

BEGIN

-- SET NOCOUNT ON added to prevent extra...

Dave is correct. Use an outer join and don't forget to put the filter on PG_ID in the join condition, not in your WHERE clause!

If you want to replace the...

Check the following articles:

Jamie Thomson

Simon Sabin

Jason,

Use a conditional split to eliminate the rows with NULLs.

Use the IGNORE_DUP_KEY = ON option, see BOL.

From my own experience, the overhead of the mutlicast and...