Package de.wwu.scdh.annotation.selection.rewriter

Class XPathNormalizer

java.lang.Object

de.wwu.scdh.annotation.selection.rewriter.XPathRewriterBase

de.wwu.scdh.annotation.selection.rewriter.XPathNormalizer

All Implemented Interfaces:

Rewriter<DOMResource,XPathRefinedByRFC5147CharScheme,XPathRefinedByRFC5147CharScheme>

Direct Known Subclasses:

XPathNormalizerWithXPath

public abstract class XPathNormalizer
extends XPathRewriterBase
implements Rewriter<DOMResource,XPathRefinedByRFC5147CharScheme,XPathRefinedByRFC5147CharScheme>

The same position (or range) in a DOMResource can be selected by different pairs of XPaths and RFC 5147 character schemes (or range schemes), i.e. *referentially equal* selectors may have different values. Normalization maps referentially equal selectors to the same selector. This XPathNormalizer is a base class for such normalization tasks.

Normalization of selectors is a 2-stage process: 1) In the first stage, the text position (or range) which is referenced by the selector, has to be found. A pair containing a text node and an RFC 5147 character scheme position is returned. 2) In the second stage, this position is expressed as a selector again, i.e., the node is referenced with an XPath, where the XPath may be written as a path expression descending from the root element, or from the deepest element with an XML-ID, etc., and even the character scheme component of the selector may be recalculated.

There is **not** the one and only normalization. Both stages of the normalization process may be implemented differently, leading to different results. For corner cases, even the first stage may lead to different results.

This class implements the first stage of the normalization process. The algorithm is selected by values of the Mode enum type, which is passed to the normalization methods. The second stage of the normalization process has to be implemented by subclasses of the abstract base class.

XPath expressions to be normalized may be arbitrary XPath 4.0 expressions which select a single node from the DOMResource. Expressions selecting not exactly one node result in an SelectorException.

Nested Class Summary

Nested classes/interfaces inherited from interface de.wwu.scdh.annotation.selection.Rewriter

Rewriter.Direction

Field Summary

Fields inherited from class de.wwu.scdh.annotation.selection.rewriter.XPathRewriterBase

EMPTY_NAMESPACE, TEXT_LEAF, TRACE_TEXT_LEAF, xpath, xPathCompiler

Constructor Summary

Constructors

Constructor	Description
XPathNormalizer(net.sf.saxon.s9api.XPathCompiler xPathCompiler, String xpath)

Method Summary

Modifier and Type	Method	Description
protected abstract String	getNormalizedXPath(DOMResource resource, net.sf.saxon.s9api.XdmNode node, boolean escaped)	This abstract method must be implemented by normalizers.
List<XPathRefinedByRFC5147CharScheme>	rewrite(DOMResource resource, XPathRefinedByRFC5147CharScheme point, RewriterConfig config)	Normalize an XPath Selector refined by an RFC5147 character scheme.

Methods inherited from class de.wwu.scdh.annotation.selection.rewriter.XPathRewriterBase

getDeepTextNodeAtPositionStepOverEnd, getDeepTextNodeAtPositionStopAtEnd, getDepth, getDepth, getDescendantTextNodesWithPosition, getFirstNodeAtPosition, getFirstOfDeepestNodesAtPosition, getLastOfDeepestNodesAtPosition, getNode, getNode, getNode, getSecondNodeAtPosition, getTextNodeAtPosition, getTextNodesWithPosition, getXPath, pathExpressionWithXPath, posInNormalizedNode, postProcForwardPaths, replaceEmptyNamespace, replaceTextLeaf, replaceTextLeaf, replaceTraceTextLeaf, reportNotFound, unespace

Methods inherited from class java.lang.Object

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

Constructor Details

XPathNormalizer

public XPathNormalizer(net.sf.saxon.s9api.XPathCompiler xPathCompiler,
 String xpath)

Method Details

rewrite

public List<XPathRefinedByRFC5147CharScheme> rewrite(DOMResource resource,
 XPathRefinedByRFC5147CharScheme point,
 RewriterConfig config)
                                              throws SelectorException

Normalize an XPath Selector refined by an RFC5147 character scheme. The position must be a valid character position *inside* the fragment selected by the XPath expression. If the fragment's text (its concatenated text nodes) is shorter than the position's value, a SelectorException is thrown.

In general, this method will result in a recalculated XPath *and* position.

Specified by:

rewrite in interface Rewriter<DOMResource,XPathRefinedByRFC5147CharScheme,XPathRefinedByRFC5147CharScheme>

Parameters:

resource - the DOMResource on the base of which the normalization is done

point - an XPathRefinedByRFC5147CharScheme record representing point in the resource

config - a record with configuration options

Returns:

a recalculated XPathRefinedByRFC5147CharScheme

Throws:

SelectorException - when the position cannot be found in the resource //

getNormalizedXPath

protected abstract String getNormalizedXPath(DOMResource resource,
 net.sf.saxon.s9api.XdmNode node,
 boolean escaped)
                                      throws SelectorException

This abstract method must be implemented by normalizers. It returns the normalized path to the node given as argument.

Parameters:

node - an XdmNode to which the normalized path must be generated to

escaped - whether or not the normalized XPath is to be escaped for X-processing, e.g., ' escaped to '.

Returns:

the normalized XPath expression as a String

Throws:

SelectorException

See Also:

• XPathRewriterBase.getTextNodeAtPosition(S9ApiResource, String, int, Mode)