iaik.xml.filter.impl.dsig

Class Canonicalizer

java.lang.Object

iaik.xml.filter.impl.dsig.Canonicalizer

public class Canonicalizer
extends Object

A canonicalizer that supports the canonicalization of arbitrary node-sets according to Canonical XML Version 1.0 and Exclusive XML Canonicalization Version 1.0.

The Canonicalizer is initialized using an Iterator that MUST provide the Nodes of the input node-set in document order (according to XML Path Language (XPath) Version 1.0, Section 5) with the following weakening: The order of the namespace and attribute child nodes of an element may be arbitrary.

This Canonicalizer examines only the nodes in the input node-set (the nodes provided by the Iterator) and their ancestor-elements. No node is examined twice, as a stack is used to keep track of the ancestor-elements of each node in the input node-set. Therefore, the elements have to be provided in document order.

This Canonicalizer supports stream processing as processNextNode() processes the next node provided by the Iterator. The returned byte[] may be empty as long as further nodes of the input node-set have to be processed to be able to render a corresponding canonicalization result.

Use a CanonInputStream to implement streaming behaviour.

See Also:

CanonInputStream

Constructor Summary

Constructors

Constructor and Description
Canonicalizer(Iterator nodeIterator, boolean withComments, boolean exclusive, boolean c14n11, List inclusiveNSPrefixes) Creates a Canonicalizer.
Canonicalizer(Iterator nodeIterator, boolean withComments, boolean exclusive, List inclusiveNSPrefixes) Creates a Canonicalizer.

Method Summary

Modifier and Type	Method and Description
byte[]	processNextNode() Processes the next Nodeof the input node-set provided by the Iteratorused to initialize this Canonicalizer.
static void	setBackToCompatibilityPrior1_13() There has been a bug in XSECT prior to 1.13 that caused a wrong canonicalization result when tabs '#x9' or line breaks '#xA' where used in an attribute.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

Canonicalizer

public Canonicalizer(Iterator nodeIterator,
                     boolean withComments,
                     boolean exclusive,
                     List inclusiveNSPrefixes)

Creates a Canonicalizer.

Parameters:

nodeIterator - that MUST provide the Nodes of the input node-set in document order

withComments - true if canonicalization with comments is should be used, or false otherwise

exclusive - true if Exclusive XML Canonicalization Version 1.0 should be used, false if Canonical XML Version 1.0 should be used

inclusiveNSPrefixes - a list of namespace prefixes (Strings) that should be canonicalized according to Canonical XML when using Exclusive XML Canonicalization . (must be null if exclusive is false.

Canonicalizer

public Canonicalizer(Iterator nodeIterator,
                     boolean withComments,
                     boolean exclusive,
                     boolean c14n11,
                     List inclusiveNSPrefixes)

Creates a Canonicalizer.

Parameters:

nodeIterator - that MUST provide the Nodes of the input node-set in document order

withComments - true if canonicalization with comments is should be used, or false otherwise

exclusive - true if Exclusive XML Canonicalization Version 1.0 should be used, false if Canonical XML Version 1.0 should be used

c14n11 - This is to support C14n 1.1 (experimental).

inclusiveNSPrefixes - a list of namespace prefixes (Strings) that should be canonicalized according to Canonical XML when using Exclusive XML Canonicalization . (must be null if exclusive is false.

Method Detail

setBackToCompatibilityPrior1_13

public static void setBackToCompatibilityPrior1_13()

There has been a bug in XSECT prior to 1.13 that caused a wrong canonicalization result when tabs '#x9' or line breaks '#xA' where used in an attribute. The wrong canonicalization result should not be security critical, however it will harm inter-operability and cause false negatives. If you have an application built before 1.12 that uses tabs '#x9' or line breaks '#xA' in attributes you may want to reinstall the bug by calling this static function prior to processing.

Since:

XSECT 1.13

processNextNode

public byte[] processNextNode()

Processes the next Nodeof the input node-set provided by the Iteratorused to initialize this Canonicalizer. Returns a byte[] containing the result of rendering the next node of the input-node set. The returned byte[] may be empty until enough nodes of the input node-set have been processed to render a corresponding binary representation (e.g. the start-tag of an element is rendered not before all namespace and attribute children of this element have been processed). Returns null after the last node of the input node-set has been processed and no further output by the canonicalizer is produced.

Returns:

a byte[] containing the result of processing the next node of the input node-set (may be empty), or null after the processing of the last node of the input node-set