Question

I am using SQL Server 2008.

I have a query that pulls two random BikeIDs out of a table called Bikes. The table looks like this:

 BikeID     BikeName
 1          Bob
 2          Cindy
 3          Carl
 4          Joe
 5          Jane

I use the NEWID() function to pull two random BikeIDs out like this:

SELECT     TOP 2 BikeID
FROM       Bikes
ORDER BY   NEWID()

My results look like this:

 Row - BikeID
 1     5
 2     1

 Row - BikeID
 1     3
 2     4

 Row - BikeID
 1     2
 2     5

 Row - BikeID
 1     3
 2     3

My problem is that I am coming up with duplicates, meaning I get the same number returned twice, see the last example resultset. I am NOT SURE whether NEWID() can return the exact same number twice or not. If not, then I must have a glitch elsewhere.

These are not actually the results that I am producing. These kinds of results appear on my web page that calls this query. So, I get duplicates on the web page. I have not reproduced these results using just SQL in the database query writer.

Is this the best or right way to get two random rows from a query? Does NEWID() eliminate the possibility of return duplicates?

Answer 1

You shouldn’t get duplicates because you can’t select the same row twice in such a simple query.

So, the simple answer is you have 2 rows with BikeID = 3