The Variant Data Type

                                                                                                          << Back to Index
A Variant variable is capable of storing all system-defined types of data. You don't have to convert between these types of data if you assign them to a Variant variable; Visual Basic automatically performs any necessary conversion. For example:
Dim SomeValue      ' Variant by default.
SomeValue = "17"   ' SomeValue contains "17" (a two-
                  ' character string).
SomeValue = SomeValue - 15       ' SomeValue now contains
                              ' the numeric value 2.
SomeValue = "U" & SomeValue      ' SomeValue now contains
                     ' "U2" (a two- character string).
While you can perform operations on Variant variables without much concern for the kind of data they contain, there are some traps you must avoid.
  • If you perform arithmetic operations or functions on a Variant, the Variant must contain something that is a number. For details, see the section, "Numeric Values Stored in Variants," in "Advanced Variant Topics."
  • If you are concatenating strings, use the & operator instead of the + operator. 
In addition to being able to act like the other standard data types, Variants can also contain three special values: Empty, Null, and Error.

The Empty Value

Sometimes you need to know if a value has ever been assigned to a created variable. A Variant variable has the Empty value before it is assigned a value. The Empty value is a special value different from 0, a zero-length string (""), or the Null value. You can test for the Empty value with the IsEmpty function:
If IsEmpty(Z) Then Z = 0
When a Variant contains the Empty value, you can use it in expressions; it is treated as either 0 or a zero-length string, depending on the expression.
The Empty value disappears as soon as any value (including 0, a zero-length string, or Null) is assigned to a Variant variable. You can set a Variant variable back to Empty by assigning the keyword Empty to the Variant.

The Null Value

The Variant data type can contain another special value: Null. Null is commonly used in database applications to indicate unknown or missing data. Because of the way it is used in databases, Null has some unique characteristics:
  • Expressions involving Null always result in Null. Thus, Null is said to "propagate" through expressions; if any part of the expression evaluates to Null, the entire expression evaluates to Null.
  • Passing Null, a Variant containing Null, or an expression that evaluates to Null as an argument to most functions causes the function to return Null.
  • Null values propagate through intrinsic functions that return Variant data types.
You can also assign Null with the Null keyword:
Z = Null
You can use the IsNull function to test if a Variant variable contains Null:
If IsNull(X) And IsNull(Y) Then
   Z = Null
Else
   Z = 0
End If
If you assign Null to a variable of any type other than Variant, a trappable error occurs. Assigning Null to a Variant variable doesn't cause an error, and Null will propagate through expressions involving Variant variables (though Null does not propagate through certain functions). You can return Null from any Function procedure with a Variant return value.
Variables are not set to Null unless you explicitly assign Null to them, so if you don't use Null in your application, you don't have to write code that tests for and handles it.

The Error Value

In a Variant, Error is a special value used to indicate that an error condition has occurred in a procedure. However, unlike for other kinds of errors, normal application-level error handling does not occur. This allows you, or the application itself, to take some alternative based on the error value. Error values are created by converting real numbers to error values using the CVErr function.
     
     
Subscribe to: Posts (Atom)