XSLT 2.0 and XPath 2.0 Programmer's Reference, 4th Edition (40 page)

The Element

The

element (or

, which is a synonym) is the outermost element of every stylesheet module.

The name

is a conventional name. The first part,
xsl
, is a prefix that identifies the namespace to which the element name belongs. Any prefix can be used so long as it is mapped, using a namespace declaration, to the URI
http://www.w3.org/1999/XSL/Transform
. There is also a mandatory
version
attribute. So the start tag of the

element will usually look like this:

The

element can also be written as

if you prefer to think of XSLT as doing transformations rather than styling. The two names are completely interchangeable, but I usually use

for familiarity. Everything I say about the

element applies equally to

.

As a general principle, it's advisable to specify
version = “2.0”
if the stylesheet module uses any facilities from XSLT 2.0 or XPath 2.0, and
version = “1.0”
if it relies only on XSLT 1.0 and XPath 1.0 features. This isn't purely documentary; as we'll see later (see page 128), XSLT processors have some subtle differences in behavior, depending on the setting of this attribute.

The other attributes that may appear on this element are described under

in Chapter 6, page 465. Specifically, they are as follows:

• id
  identifies the stylesheet when it appears as an embedded stylesheet within another document. Embedded stylesheets are described in the next section.
  
• extension-element-prefixes
  is a list of namespace prefixes that denote elements used for vendor-defined or user-defined extensions to the XSLT language.
  
• exclude-result-prefixes
  is a list of namespaces used in the stylesheet that should not be copied to the result tree unless they are actually needed. I'll explain how this works in the section
  Literal Result Elements
  on page 112.
  
• xpath-default-namespace
  is a namespace URI, which is used as the default namespace for unprefixed element names used in path expressions within the stylesheet, and also for unprefixed type names. This attribute is handy when all the elements in your source document are in a particular namespace, because it saves you having to use a namespace prefix every time you refer to an element in this namespace. Without this attribute, element names with no prefix are assumed to refer to names in the null namespace (neither the default namespace declared using
  xmlns = “uri”
  in the stylesheet, nor the default namespace declared in the source document, has any effect on names used in path expressions).
  
• default-validation
  takes one of the values
  preserve
  or
  strip
  . This attribute is used by a schema-aware processor to determine whether any schema-derived type annotations are retained when elements and attributes are copied. You will find more details of how result trees are validated against a schema in Chapter 4.
  
• default-collation
  defines how strings are to be compared and sorted. There are many different conventions for sorting strings, depending on the language and sometimes on the whims of a publisher. The value of this attribute is simply a URI that acts as a name for the set of conventions to be used; the values you can use for this attribute will vary from one XSLT processor to another. For more details on collations, see
  
  on page 459 in Chapter 6.
  
• input-type-annotations
  takes one of the values
  preserve
  ,
  strip
  ,
  or
  unspecified
  . This attribute is used by a schema-aware processor to determine whether any schema-derived type annotations are retained in the input document.