All the examples for this lesson are based on Microsoft SQL Server Management Studio and the AdventureWorks2012 database. You can get started using these free tools using my Guide Getting Started Using SQL Server.
Using Subqueries in the WHERE Clause
a In some cases it may make sense to rethink the query and use a JOIN, but you should really study both forms via the query optimizer before making a final decision.
The comparison modifiers ANY and ALL can be used with greater than, less than, or equals operators. Doing so provides a means to compare a single value, such as a column, to one or more results returned from a subquery.
Let’s now explore these in detail
Exist and Not Exists
The EXISTS condition is used in combination with a subquery. It returns TRUE whenever the subquery returns one or more values.
In its simplest form the syntax for the EXISTS condition is
WHERE EXISTS (sub query)
Suppose we need to return all sales orders written by salespeople with sales year to date greater than three million dollars. To do so we can use the EXISTS clause as shown in this example:
SELECT SalesOrderID, RevisionNumber, OrderDate FROM Sales.SalesOrderHeader WHERE EXISTS (SELECT 1 FROM sales.SalesPerson WHERE SalesYTD > 3000000 AND SalesOrderHeader.SalesPersonID = Sales.SalesPerson.BusinessEntityID)
When this SQL executes the following comparisons are made:
- The WHERE clause returns all records where the EXISTS clause is TRUE.
- The EXIST clause uses a correlated subquery. The outer query is correlated to the inner query by SalesPersonID.
- Only SalesPersons with SalesYTD greater than three million are included in the results.
- The EXISTS clause returns TRUE if one or more rows are returned by the subquery.
The EXISTS condition is a membership condition in the sense it only returns TRUE if a result is returned. Conversely, if we want to test for non-membership we can use NOT EXISTS.
NOT EXISTS returns TRUE if zero rows are returned. So, if we want to find all sales orders that were written by salespeople that didn’t have 3,000,000 in year-to-date sales, we can use the following query:
SELECT SalesOrderID, RevisionNumber, OrderDate FROM Sales.SalesOrderHeader WHERE NOT EXISTS (SELECT 1 FROM sales.SalesPerson WHERE SalesYTD > 3000000 AND SalesOrderHeader.SalesPersonID = Sales.SalesPerson.BusinessEntityID)
WHAT happens to NULL?
When the subquery returns a null value what does EXIST return: NULL, TRUE, or FALSE?
To be honest I was surprised.
I was sure it would return NULL, but to my surprise, I learned it returns TRUE. Therefore, if your subquery returns a NULL value, the EXISTS statement resolves to TRUE. In the following example all the SalesOrderHeader rows are returned as the WHERE clause essentially resolved to TRUE:
SELECT SalesOrderID, RevisionNumber, OrderDate FROM Sales.SalesOrderHeader WHERE EXISTS (SELECT NULL)
As we study the IN operator, we’ll see this behavior is unique to the EXISTS clause.
IN and NOT IN
We first studied the IN operator back in the lesson How to Filter Your Query Results. When used in subqueries, the mechanics of the IN and NOT IN clause are the same. Here is a summary of that article.
IN and NOT IN Review
The IN operator is considered a membership type. The membership type allows you to conduct multiple match tests compactly in one statement. For instance, consider if you have a couple spelling variations for the leader of the company such as ‘Owner’, ‘President’, and ‘CEO.’ In c case like this you could use the in operator to find all matches
ContactTitle IN ('CEO', 'Owner', 'President')
The above will math or return turn if the contact title is either ‘CEO’, ‘Owner’, or ‘President.’ To use the IN comparison operator separate the items you which to test for with commas and be sure to enclose them in parenthesis. The full SQL statement for our example is
SELECT CompanyName, ContactName, ContactTitle FROM Customers WHERE ContactTitle IN ('CEO', 'Owner', 'President'); Note: The above query isn't meant for the adventure works database
Using IN with a Subquery
When used with subqueries, the list of values is replaced with a subquery. The advantage of using a subquery, in this case, is that it helps to make your queries more data-driven and less brittle.
What I mean is you don’t have to hard code values.
If for instance you’re doing a query to find sales order by top salespeople, the non-sub query way to use the IN statement is
SELECT SalesOrderID, OrderDate, AccountNumber, CustomerID, SalesPersonID, TotalDue FROM Sales.SalesOrderHeader WHERE SalesPersonID IN (279, 286, 289) whose results are
But now since we know about subqueries, we can use the following to obtain the same list
SELECT SalesOrderID, OrderDate, AccountNumber, CustomerID, SalesPersonID, TotalDue FROM Sales.SalesOrderHeader WHERE SalesPersonID IN (SELECT BusinessEntityID FROM Sales.SalesPerson WHERE Bonus > 5000)
The advantage is that as salespersons sell more or less, the list of salesperson ID’s returned adjusts.
Just like with other queries you can create a correlated subquery to be used with the IN clause. Here is the same query we used with the EXIST clause.
It returns all sales orders written by salespeople with sales year to date greater than three million dollars, but now we use the IN clause:
SELECT SalesOrderID, RevisionNumber, OrderDate FROM Sales.SalesOrderHeader WHERE SalesPersonID IN (SELECT SalesPerson.BusinessEntityID FROM sales.SalesPerson WHERE SalesYTD > 3000000 AND SalesOrderHeader.SalesPersonID = Sales.SalesPerson.BusinessEntityID)
As IN returns TRUE if the tested value is found in the comparison list, NOT IN returns TRUE if the tested value is not found. Taking the same query from above, we can find all Sales orders that were written by salespeople that didn’t write 3,000,000 in year-to-date sales, we can write the following query:
SELECT SalesOrderID, RevisionNumber, OrderDate FROM Sales.SalesOrderHeader WHERE SalesPersonID NOT IN (SELECT SalesPerson.BusinessEntityID FROM sales.SalesPerson WHERE SalesYTD > 3000000 AND SalesOrderHeader.SalesPersonID = Sales.SalesPerson.BusinessEntityID)
WHAT happens to NULL with IN?
When the comparison list only contains the NULL value, then any value compared to that list returns false.
SELECT SalesOrderID, RevisionNumber, OrderDate FROM Sales.SalesOrderHeader WHERE SalesOrderID IN (SELECT NULL)
returns zero rows. This is because the IN clause always returns false. Contrast this to EXISTS, which returns TRUE even when the subquery returns NULL.
Comparison operators such as greater than, less than, equal, and not equal can be modified in interesting ways to enhance comparisons done in conjunction with WHERE clauses.
Rather than using >, which only makes sense when comparing to a single (scalar) value, you can use > ANY or > ALL to compare a column value to a list results returned from the subquery.
Using the > ANY Modifier
The comparison operator > ANY means greater than one or more items in the list. This is the same as saying it greater than the MIN value of the list. So the expression
Sales > ANY (1000, 2000, 2500)
returns TRUE if Sales are greater than 1000 as this expression is equivalent to
Sales > MIN(1000, 2000, 2500)
Which simplifies to
Sales > 1000
Note: You may see some queries using SOME. Queries using SOME return the same result as those using ANY. Simply said > ANY is the same as > SOME.
Let’s do an example using the adventure works database. We’re going to find all products which may have a high safety stock level. To do so, we’ll look for all products that have a SafetyStockLevel that is greater than the average SafetyStockLevel for various DaysToManufacture.
The query to do this is:
SELECT ProductID, Name, SafetyStockLevel, DaysToManufacture FROM Production.Product WHERE SafetyStockLevel > ANY (SELECT AVG(SafetyStockLevel) FROM Production.Product GROUP BY DaysToManufacture)
When this subquery is run it first calculates the Average SafetyStockLevel. This returns a list of numbers. Then for each product row in the outer query SafetyStockLevel is compared. If it is greater than one or more from the list, then include it in the results.
Like me, you may at first think that > ANY is redundant, and not needed. It is equivalent to > MIN(…) right?
What I found out is that though it is equivalent in principle, you can’t use MIN. The statement
SELECT ProductID, Name, SafetyStockLevel, DaysToManufacture FROM Production.Product WHERE SafetyStockLevel > MIN((SELECT AVG(SafetyStockLevel) FROM Production.Product GROUP BY DaysToManufacture))
Won’t run. It return the error, “Cannot perform an aggregate function on an expression containing an aggregate or a subquery.”
Using the > ALL Modifier
The > ALL modifier works in a similar fashion except it returns the outer row if it’s comparison value is greater than every value returned by the inner query.
The comparison operator > ALL means greater than the MAX value of the list. Using the example above, then
Sales > ALL (1000, 2000, 2500)
is equivalent to
Sales > MAX(1000, 2000, 2500)
Which returns TRUE if Sales > 2500
In this example, we’ll return all SalesPeople that have a bonus greater than ALL salespeople whose year-to-date sales were less than a million dollars.
SELECT p.BusinessEntityID, p.FirstName, p.LastName, s.Bonus, s.SalesYTD FROM Person.Person AS p INNER JOIN Sales.SalesPerson AS s ON p.BusinessEntityID = s.BusinessEntityID WHERE s.Bonus > ALL (SELECT Bonus FROM Sales.SalesPerson WHERE Sales.SalesPerson.SalesYTD < 1000000)
Summary of Various Comparison Modifiers
You can use comparison modifiers with other operators, such as equals. Use the chart below to get a better understanding of the examples. I’ve listed all the combinations, even those that don’t make too much sense.
When reviewing the example assume the subquery returns a list of three numbers: 1,2,3.
Some combinations of these comparison modifiers are downright goofy. For instance, I can’t imagine using “= ALL” or “<> ANY.” The others make sense, and as we have shown you can really use MAX or MIN as legal equivalent statements. ANY and ALL do have their places!
Out of all of the items we discussed today I’ve used EXISTS and NOT EXISTS the most with subqueries. I use IN quite a bit, but usually with a static list, not with subqueries.