XSLT 2.0 and XPath 2.0 Programmer's Reference, 4th Edition (559 page)

The function has been generalized to work with all numeric types.

Signature

Effect

If the number is an
xs:integer
, or is equal to an
xs:integer
, then it is returned unchanged.

Otherwise, it is rounded down to the next lowest whole number. If the supplied value is an
xs:decimal
, the result will be an
xs:decimal
, if it is an
xs:double
, the result will be an
xs:double
, and if it is an
xs:float
, the result will be an
xs:float
. In the case of negative numbers, the rounding is away from zero.

The
xs:double
and
xs:float
types in XPath support special values such as infinity, negative zero and NaN (not-a-number), which are described on page 199 in Chapter 5.

If the argument is NaN (not-a-number), the result will be NaN. Similarly, when the argument is positive or negative infinity, the function will return the value of the argument unchanged.

Examples

Expression	Result
floor(1.0)	xs:decimal 1.0, displayed as 1
floor(1.6e0)	xs:double 1.0e0, displayed as 1
floor(17 div 3)	xs:decimal 5.0, displayed as 5
floor(-3.0)	xs:decimal −3.0, displayed as -3
floor(-8.2e0)	xs:double −9.0e0, displayed as -9
floor(number(‘NaN’))	xs:double NaN

Usage

Like
round()
and
ceiling()
, this function is useful when calculating sizes of HTML tables.

Two alternatives you may want to consider are:

• Using the
  xs:integer()
  constructor function. This differs from
  floor()
  in that it always truncates (rounds toward zero); also, it returns an actual integer, rather than returning a value of the same type as the argument.
  
• Using the expression
  $x idiv 1
  . This produces the same result as the
  xs:integer()
  constructor function, but saves you from having to declare the XML Schema namespace.