IIF and Choose Functions in SQL Server 2012

Here, I have provided an article showing you how to utilize the two new logical functions Choose and IIF in SQL Server. The Choose function works like an array kind of thing and the IIF function is used to check a condition. In this article we will see both functions with examples. These functions are also called new logical functions in SQL Server 2012. So let's take a look at a practical example of how to use the Choose and IIF functions in SQL Server. The example is developed in SQL Server 2012 using the SQL Server Management Studio.

These are the two logical functions:

  1. IIF() Function
  2. Choose() Function

IIF() Function

The IIF function is used to check a condition. Suppose X>Y. In this condition a is the first expression and b is the second expression. If the first expression evaluates to TRUE then the first value is displayed, if not the second value is displayed.

Syntax

IIF ( boolean_expression, true_value, false_value )

Example

DECLARE @X INT;

SET @X=50;

DECLARE @Y INT;

SET @Y=60;

Select iif(@X>@Y, 50, 60) As IIFResult

In this example X=50 and Y=60; in other words the condition is false.  Select iif(@X>@Y, 50, 60) As IIFResult. It returns false value that is 60.

Output

IIF-Function.jpg

Choose() Function

This function is used to return the value out of a list based on its index number. You can think of it as an array kind of thing. The Index number here starts from 1.

Syntax

CHOOSE ( index, value1, value2.... [, valueN ] )

CHOOSE() Function excepts two parameters,

Index: Index is an integer expression that represents an index into the list of the items. The list index always starts at 1. 

Value: List of values of any data type.

Now some facts related to the Choose Function

1. Item index starts from 1

DECLARE @ShowIndex INT;

SET @ShowIndex =5;

Select Choose(@ShowIndex, 'M','N','H','P','T','L','S','H') As ChooseResult 

 

In the preceding example we take index=5. It will start at 1. Choose() returns T as output since T is present at @Index location 5.

Output

Coose-Function1-in-sql-server.jpg

2.  When passed a set of types to the function it returns the data type with the highest precedence; see:

DECLARE @ShowIndex INT;

SET @ShowIndex =5;

Select Choose(@ShowIndex ,35,42,12.6,14,15,18.7)  As CooseResult

In this example we use index=5. It will start at 1. Choose() returns 15.0 as output since 15 is present at @ShowIndex location 5 because in the item list, fractional numbers have higher precedence than integers.

Coose-Function2-in-sql-server.jpg

3. If an index value exceeds the bound of the array it returns NULL

DECLARE @ShowIndex INT;

SET @ShowIndex =9;

Select Choose(@ShowIndex , 'M','N','H','P','T','L','S','H')  As CooseResult

In this example we take index=9. It will start at 1. Choose() returns Null as output because in the item list the index value exceeds the bounds of the array; the last Index=8.

Output

Coose-Function3-in-sql-server.jpg

4. If the index value is negative then that exceeds the bounds of the array therefore it returns NULL; see:

DECLARE @ShowIndex INT;

SET @ShowIndex =-1;

Select Choose(@ShowIndex, 'M','N','H','P','T','L','S','H')  As CooseResult

In this example we take index= -1. It will start at 1. Choose() returns Null as output because in the item list the index value exceeds the bounds of the array.

Output

Coose-Function3-in-sql-server.jpg

5. If the provided index value has a float data type other than int, then the value is implicitly converted to an integer; see:

DECLARE @ShowIndex  INT;

SET @ShowIndex =4.5;

Select Choose(@ShowIndex ,35,42,12.6,13,15,20) As CooseResult

In this example we take index= 4.5. It will start at 1.  If the specified index value has a float data type other than int, then the value is implicitly converted to an integer. It returns the 13.0 as output since 15 is present at @ShowIndex=4.5 which means index is 4.

Output

Coose-Function4-in-sql-server.jpg

Up Next
    Ebook Download
    View all
    Learn
    View all