How to create "fill in" rows for date ranges that have gaps?

I need some help building some SQL that will create "fill in" rows for date ranges that have gaps. The data I'm working with looks something like this:

Student_ID EFF_DATE EXP_DATE
100 1/1/2009 5/1/2009
100 8/1/2009 12/31/2009
101 12/1/2009 4/1/2010
101 7/1/2010 9/1/2010
101 11/1/2010 12/1/2010
102 1/1/2010 6/1/2010

In this example students 100 and 101, both have gaps for their respective date ranges. I'll need to come up with some SQL that will return the following 3 rows:

Student ID 100: 5/2/2009-7/31/2009
Student ID 101: 4/2/2010-6/30/2010
Student ID 101: 9/2/2010-10/31/2010

Any ideas how I'll be able to code this in SQL?

The following is based on code from a design tip I wrote last year on assigning end dates to historical rows in a dimension. It is Oracle based, and it uses a Customer_Master table, but I think it does what you are looking for.

INSERT INTO Customer_master
SELECT TabA.Customer_Key, TabA.Source_Cust_ID, TabA.First_Name, TabA.Last_Name,
TabA.Address, TabA.City, TabA.State, TabA.Zip,
TabA.End_Date + 1 AS Eff_Date, TabB.Eff_Date -1 AS End_Date, 'N', 'INS'
FROM
(SELECT ROW_NUMBER() OVER(Partition by Source_Cust_ID Order by Eff_Date) AS RowNumA,
Customer_Key, Source_Cust_ID, First_Name, Last_Name, Address, City, State,
Zip, Eff_Date, End_Date, Current_Flag, Change_Reason
from Customer_master ) TabA -- First row for a customer
LEFT OUTER JOIN
(SELECT ROW_NUMBER() OVER(Partition by Source_Cust_ID Order by Eff_Date) AS RowNumB,
Source_Cust_ID, Eff_Date, End_Date
from Customer_master) TabB -- Second row for the same customer
ON TabA.RowNumA = TabB.RowNumB - 1
AND TabA. Source_Cust_ID = TabB. Source_Cust_ID
WHERE TabB.Eff_Date - TabA.End_Date > 1; -- Must have more than one day difference

Hope this helps,
--Warren

Thanks Warren! This works perfectly!