The following code is a basic left join. It just doesn't return the result set you requested.

select

*

from

#date_dim d

...

Is there a reason you need all the duplicated rows where the dates have null values?

If you provide the DDL (create statements) for your tables, sample data for the source table (as insert statements that can be cut, paste, and executed in SSMS) to load...

Philip Horan (12/18/2008)

---

I am attempting to INSERT from a temporary table. I created the temporary table from an excel file. The data types for the temporary table do not match...

You want to use one of these two functions: suser_name(), suser_sname(). You might want to read up on them in Books Online to see what they do.

I did a little more testing adding a cast( as varchar(100)) around the replace, and it validated Jeffrey's assertions above.

That is essentially the question you asked originally. The problem was that the information you showed did not match reality. You showed values that were all numeric. ...

Just ran a test. Here is what I found. If you use REPLACE in a select into statement, since it needs to create the table on the fly...

Steve used to give away SSC Polo Shirts when you reached a certain level as a forum poster, but the heck if I know what level that is, as I...

Well, not knowing what you are working with, and deciding to take a swag at it, try out the following code and see if that gets you going in the...

gcopeland (12/17/2008)

---

Lynn Pettis (12/17/2008)

---

The .NET CLR was designed to work with any language, unfortunately SSIS and SSRS weren't written that way pre SQL 2008. In SQL Server 2005 and...

gcopeland (12/17/2008)

---

Jack Corbett (12/17/2008)

---

It also depends on what you are doing with the language as well. As I said in my original post, if you are using SSRS or...

Okay, I did say if you knew if your data was ALL numeric. You will need to analysis your data first and see what you have. Do you...

If this were the case, there would be no choice in language. Seems to be a bit of arrogance in this stance.

I am advocate...

If you are sure that the values are all numeric, you can do this:

ORDER BY

CAST(YourColumn as INT)