Implementation of Discretionary Behaviors in the XslTransform Class
Note
The XslTransform class is obsolete in the .NET Framework 2.0. You can perform Extensible Stylesheet Language for Transformations (XSLT) transformations using the XslCompiledTransform class. See Using the XslCompiledTransform Class and Migrating From the XslTransform Class for more information.
Discretionary behaviors are described as behaviors listed in the World Wide Web Consortium (W3C) XSL Transformations (XSLT) Version 1.0 Recommendation, in which the implementation provider chooses one of several possible options as a way to handle a situation. For example, in section 7.3 Creating Processing Instructions, the W3C Recommendation says that it is an error if instantiating the content of xsl:processing-instruction creates nodes other than text nodes. For some problems, the W3C tells what decision should be made if the processor decides to recover from the error. For the problem given in section 7.3, the W3C says that the implementation can recover from this error by ignoring the nodes and their content.
Therefore, for each of the discretionary behaviors allowed by the W3C, the table below lists the discretionary behaviors implemented for the .NET Framework implementation of the XslTransform class, and what section in the W3C XSLT 1.0 Recommendation that this problem is discussed.
| Problem | Behavior | Section |
|---|---|---|
A text node matches both xsl:strip-space and xsl:preserve-space. |
Recover | 3.4 |
| A source node matches more than one template rule. | Recover | 5.5 |
| A namespace Uniform Resource Identifier (URI) is declared to be an alias for multiple namespace URIs, all having the same import precedence. | Recover | 7.1.1 |
The name attribute in xsl:attribute and xsl:element generated from an attribute value template is not a valid qualified name (QName). |
Exception thrown | 7.1.2 and 7.1.3 |
| Adding an attribute to an element after child nodes have already been added to the element node. | Recover | 7.1.3 |
| Adding an attribute to anything other than an element node. | Recover | 7.1.3 |
Instantiation of the content of the xsl:attribute element is not a text node. |
Recover | 7.1.3 |
| Two attribute sets have the same import precedence and expanded name. Both have the same attribute, and there is no other attribute set containing the common attribute with the same name with higher importance. | Recover | 7.1.4 |
xsl:processing-instruction name attribute does not yield both a no-colon name (NCName) and a Processing Instruction target. |
Recover | 7.3 |
Instantiating the content of xsl:processing-instruction creates nodes other than text nodes. |
Recover | 7.3 |
Results of instantiating the content of the xsl:processing-instruction contains the string "?>". |
Recover | 7.3 |
Results of instantiating the content of the xsl:comment contains the string "--", or ends with "-". |
Recover | 7.4 |
Results of instantiating the content of the xsl:comment creates nodes other than text nodes. |
Recover | 7.4 |
| The template within a variable-binding element returns an attribute node or a namespace node. | Recover | 11.2 |
| There is an error retrieving the resource from the URI passed into the document function. | Exception thrown | 12.1 |
| The URI reference in the document function contains a fragment identifier, and there is an error processing the fragment identifier. | Exception thrown | 12.1 |
There are multiple attributes with the same name that are not named cdata-section-elements in xls:output, and these attributes have the same import precedence. |
Recover | 16 |
The processor does not support the character encoding value given in the encoding attribute of the xsl:output element. |
Recover | 16.1 |
disable-output-escaping is used for a text node, and that text node is used to create something other than a text node in the result tree. |
disable-output-escaping attribute is ignored |
16.4 |
| Converting a result tree fragment to a number or string if the result tree fragment contains a text node with output escaping enabled. | Ignored | 16.4 |
| Output escaping is disabled for characters that cannot be represented in the encoding that the XSLT processor is using for output. | Ignored | 16.4 |
| Adding a namespace node to an element after children have been added to it or after attributes have been added to it | Recover | Errata e25 |
xsl:number is NaN, infinite, or less than 0.5. |
Recover | Errata e24 |
| The second argument node-set to the document function is empty and the URI reference is relative | Recover | Errata e14 |
Sections from the errata can be found in the W3C XSL Transformations (XSLT) Version 1.0 Specification Errata.
Custom-Defined Implementation Behaviors
There are behaviors unique to the XslTransform class implementation. This section discusses the provider-specific implementation of the xsl:sort and optional features that are supported by the XslTransform class.
xsl:sort
When using a transformation to sort, the W3C XSLT 1.0 Recommendation makes some observations. They are:
Two XSLT processors may be conforming processors, but still may sort differently.
Not all XSLT processors support the same languages.
With regard to languages, different processors may vary on their sorting on a particular language not specified on the
xsl:sort.
The following table shows the sorting behavior implemented for each data type in the .NET Framework implementation of a transformation using XslTransform:
| Data type | Sorting behavior |
|---|---|
| Text | Data is sorted using the common language runtime (CLR) String.Compare method, and the cultural locale. When the data type equals "text", sorting in the XslTransform class behaves identically to the string comparison behaviors of the CLR. |
| Number | Numeric values are treated as XML Path Language (XPath) numbers and are sorted according to the details outlined in the W3C XML Path Language (XPath) Version 1.0 Recommendation, Section 3.5. |
Optional Features Supported
The following table shows the features that are optional for an XSLT processor to implement and are implemented in the XslTransform class.
| Feature | Reference location | Notes |
|---|---|---|
disable-output-escaping attribute on <xsl:text...> and <xsl:value-of...> tags. |
W3C XSLT 1.0 Recommendation, Section 16.4 |
The disable-output-escaping attribute is ignored when the xsl:text or xsl:value-of elements are used in an xsl:comment, xsl:processing-instruction, or xsl:attribute element.Result tree fragments that contain text and the text output that has been escaped are not supported. The disable-output-escaping attribute is ignored when transforming to an XmlReader or XmlWriter object. |