com.hp.hpl.sparta.Element Maven / Gradle / Ivy
Go to download
Show more of this group Show more artifacts with this name
Show all versions of pinyin4j Show documentation
Show all versions of pinyin4j Show documentation
Support Chinese character (both Simplified and Tranditional) to most popular Pinyin systems, including
Hanyu Pinyin, Tongyong Pinyin, Wade-Giles, MPS2, Yale and Gwoyeu Romatzyh. Support multiple pronounciations and
customized output.
The newest version!
package com.hp.hpl.sparta;
import java.io.*;
import java.util.*;
import com.hp.hpl.sparta.xpath.*;
/**
A type of Node with a particular tagName that has a set of attributes and
can contain other nodes as children. An example of its form in XML in the form
<tagName attr1="value1" attr2="value2">
text
<childTag...>
<childTag...>
text
<childTag...>
text
</tagName>
Copyright (C) 2002 Hewlett-Packard Company.
This file is part of Sparta, an XML Parser, DOM, and XPath library.
This library is free software; you can redistribute it and/or
modify it under the terms of the GNU Lesser General Public License
as published by the Free Software Foundation; either version 2.1 of
the License, or (at your option) any later version. This library
is distributed in the hope that it will be useful, but WITHOUT ANY
WARRANTY; without even the implied warranty of MERCHANTABILITY or
FITNESS FOR A PARTICULAR PURPOSE.
@see GNU Lesser General Public License
@version $Date: 2003/10/28 00:25:11 $ $Revision: 1.10 $
@author Eamonn O'Brien-Strain
@see org.w3c.dom.Element
* @stereotype container
*/
public class Element extends Node {
static private final boolean DEBUG = false;
public Element(String tagName) {
tagName_ = Sparta.intern(tagName);
}
Element() {}
/** Create a deep clone of this Element. It will have the tagname
* and attributes as this node. This method will be called
* recursively to copy the while subtree of child Elements and
* Text nodes. */
public Object clone() {
return cloneElement(true);
}
/** Create a shallow clone of this Element. It will have the
* tagname and attributes as this Element but will not have child
* Elements or Nodes. */
public Element cloneShallow() {
return cloneElement(false);
}
/** Create a clone of this node. It will
* have the tagname and attributes as this node. If deep is
* true, this method will be called recursively to copy the while
* subtree of child Elements and text nodes. */
public Element cloneElement(boolean deep) {
Element result = new Element(tagName_);
if (attributeNames_ != null)
for (Enumeration i = attributeNames_.elements(); i.hasMoreElements();) {
String name = (String) i.nextElement();
result.setAttribute(name, (String) attributes_.get(name));
}
if (deep) for (Node n = firstChild_; n != null; n = n.getNextSibling())
result.appendChild((Node) n.clone());
return result;
}
/** @return tag as interned string.*/
public String getTagName() {
return tagName_;
}
public void setTagName(String tagName) {
tagName_ = Sparta.intern(tagName);
notifyObservers();
}
/**
* @return either an Element or a Text node
*/
public Node getFirstChild() {
return firstChild_;
}
/**
* @return either an Element or a Text node
*/
public Node getLastChild() {
return lastChild_;
}
/** Return enumeration of Strings */
public Enumeration getAttributeNames() {
if (attributeNames_ == null)
return Document.EMPTY;
else
return attributeNames_.elements();
}
/**
* @return value of attribute that has this name or null if no such attribute.
* WARNING! Unlike in a previous version of this API, this string is not interned
* therefore you must do a.equals(b) instead of a==b when comparing
* attribute values.
*/
public String getAttribute(String name) {
return attributes_ == null ? null : (String) attributes_.get(name);
}
/** @param name attribute name which must be non-null, non empty
@param value attribue value.
@precondition non zero-length name
*/
public void setAttribute(String name, String value) {
if (attributes_ == null) {
attributes_ = new Hashtable();
attributeNames_ = new Vector();
}
if (attributes_.get(name) == null) attributeNames_.addElement(name);
attributes_.put(name, value);
notifyObservers();
if (DEBUG) checkInvariant();
}
/** remove this attribute if it exists, otherwise silently do nothing. */
public void removeAttribute(String name) {
if (attributes_ == null) return;
attributes_.remove(name);
attributeNames_.removeElement(name);
notifyObservers();
}
void appendChildNoChecking(Node addedChild) {
if (DEBUG) checkInvariant();
Element oldParent = addedChild.getParentNode();
if (oldParent != null) oldParent.removeChildNoChecking(addedChild);
addedChild.insertAtEndOfLinkedList(lastChild_);
if (firstChild_ == null) firstChild_ = addedChild;
addedChild.setParentNode(this);
//!children_.add/*Element*/( addedChild );
lastChild_ = addedChild;
addedChild.setOwnerDocument(getOwnerDocument());
if (DEBUG) checkInvariant();
}
/** Add node as child of this element, cloning node if it is this element or
* an ancestor.
*/
public void appendChild(Node addedChild) {
if (DEBUG) checkInvariant();
if (!canHaveAsDescendent(addedChild)) addedChild = (Element) addedChild.clone();
appendChildNoChecking(addedChild);
notifyObservers();
if (DEBUG) checkInvariant();
}
boolean canHaveAsDescendent(Node node) {
if (node == this) return false;
Element parent = getParentNode();
if (parent == null) return true;
return parent.canHaveAsDescendent(node);
}
private boolean removeChildNoChecking(Node childToRemove) {
if (DEBUG) checkInvariant();
int i = 0;
for (Node child = firstChild_; child != null; child = child.getNextSibling()) {
if (child.equals(childToRemove)) {
//Fix up list endpoints if necessary
if (firstChild_ == child) firstChild_ = child.getNextSibling();
if (lastChild_ == child) lastChild_ = child.getPreviousSibling();
child.removeFromLinkedList();
child.setParentNode(null);
child.setOwnerDocument(null);
if (DEBUG) checkInvariant();
return true;
}
++i;
}
if (DEBUG) checkInvariant();
return false;
}
public void removeChild(Node childToRemove) throws DOMException {
boolean found = removeChildNoChecking(childToRemove);
if (!found)
throw new DOMException(DOMException.NOT_FOUND_ERR, "Cannot find " + childToRemove + " in "
+ this);
notifyObservers();
if (DEBUG) checkInvariant();
}
/** Replace oldChild with newChild.
* @throws DOMException if oldChild object is not a child.
* */
public void replaceChild(Element newChild, Node oldChild) throws DOMException {
replaceChild_(newChild, oldChild);
notifyObservers();
}
/** Replace oldChild with newChild.
* @throws DOMException if oldChild object is not a child.
* */
public void replaceChild(Text newChild, Node oldChild) throws DOMException {
replaceChild_(newChild, oldChild);
notifyObservers();
}
private void replaceChild_(Node newChild, Node oldChild) throws DOMException {
int i = 0;
for (Node child = firstChild_; child != null; child = child.getNextSibling()) {
if (child == oldChild) {
//Fix up list endpoints if necessary
if (firstChild_ == oldChild) firstChild_ = newChild;
if (lastChild_ == oldChild) lastChild_ = newChild;
//Make oldChild's neighbouring siblings point to newChild
oldChild.replaceInLinkedList(newChild);
//Fix parent pointers
newChild.setParentNode(this);
oldChild.setParentNode(null);
return;
}
++i;
}
throw new DOMException(DOMException.NOT_FOUND_ERR, "Cannot find " + oldChild + " in " + this);
}
/** Accumlate text nodes hierarchically. */
void toString(Writer writer) throws IOException {
for (Node i = firstChild_; i != null; i = i.getNextSibling())
i.toString(writer);
}
/** Write XML representation to character stream. */
public void toXml(Writer writer) throws IOException {
writer.write("<" + tagName_);
if (attributeNames_ != null)
for (Enumeration i = attributeNames_.elements(); i.hasMoreElements();) {
String name = (String) i.nextElement();
String value = (String) attributes_.get(name);
writer.write(" " + name + "=\"");
htmlEncode(writer, value);
writer.write("\"");
}
if (firstChild_ == null)
writer.write("/>");
else {
writer.write(">");
for (Node i = firstChild_; i != null; i = i.getNextSibling())
i.toXml(writer);
writer.write("");
}
}
private XPathVisitor visitor(String xpath, boolean expectStringValue) throws XPathException {
XPath parseTree = XPath.get(xpath);
if (parseTree.isStringValue() != expectStringValue) {
String msg =
expectStringValue ? "evaluates to element not string" : "evaluates to string not element";
throw new XPathException(parseTree, "\"" + parseTree + "\" evaluates to " + msg);
}
return new XPathVisitor(this, parseTree);
}
/** Select all the elements that match the relative XPath
expression with respect to this element. */
public Enumeration xpathSelectElements(String xpath) throws ParseException {
try {
if (DEBUG) {
Document doc = getOwnerDocument();
if (doc != null && this == doc.getDocumentElement()) {
XPath parseTree = XPath.get(xpath);
doc.monitor(parseTree);
}
}
return visitor(xpath, false).getResultEnumeration();
} catch (XPathException e) {
throw new ParseException("XPath problem", e);
}
}
/** Select all the strings that match the relative XPath
expression with respect to this element. */
public Enumeration xpathSelectStrings(String xpath) throws ParseException {
try {
return visitor(xpath, true).getResultEnumeration();
} catch (XPathException e) {
throw new ParseException("XPath problem", e);
}
}
/** Make sure this XPath exists, creating nodes if necessary,
* returning true if any nodes created. Xpath must of the type that
* returns an element (not a string).
* */
public boolean xpathEnsure(String xpath) throws ParseException {
try {
//Quick exit for common case
if (xpathSelectElement(xpath) != null) return false;
//Split XPath into parent steps and last step
final XPath parseTree = XPath.get(xpath);
int stepCount = 0;
for (Enumeration i = parseTree.getSteps(); i.hasMoreElements();) {
i.nextElement();
++stepCount;
}
Step[] parentSteps = new Step[stepCount - 1];
Enumeration i = parseTree.getSteps();
for (int j = 0; j < parentSteps.length; ++j)
parentSteps[j] = (Step) i.nextElement();
Step lastStep = (Step) i.nextElement();
Element parent;
if (parentSteps.length == 0) {
parent = this;
} else {
String parentXPath = XPath.get(parseTree.isAbsolute(), parentSteps).toString();
xpathEnsure(parentXPath.toString()); //recursion
parent = xpathSelectElement(parentXPath);
}
Element newChild = makeMatching(parent, lastStep, xpath);
parent.appendChildNoChecking(newChild);
return true;
} catch (XPathException e) {
throw new ParseException(xpath, e);
}
}
/** Select the first element that matches the relative XPath
expression with respect to this element, or null if
there is no match.
@todo make more efficient by short-circuiting the search.*/
public Element xpathSelectElement(String xpath) throws ParseException {
try {
if (DEBUG) {
Document doc = getOwnerDocument();
if (doc != null && this == doc.getDocumentElement()) {
XPath parseTree = XPath.get(xpath);
doc.monitor(parseTree);
}
}
return visitor(xpath, false).getFirstResultElement();
} catch (XPathException e) {
throw new ParseException("XPath problem", e);
}
}
/** Select the first element that matches the relative XPath
expression with respect to this element, or null if
there is no match. */
public String xpathSelectString(String xpath) throws ParseException {
try {
return visitor(xpath, true).getFirstResultString();
} catch (XPathException e) {
throw new ParseException("XPath problem", e);
}
}
/** To be equal elements must have the same tagname, they must
* have the same children (applying equals recursivly) in the
* same order and they must have the same attributes in any
* order. Elements can be equal even if they are in different
* documents, have different parents, have different siblings, or
* have different annotations.
* */
public boolean equals(Object thatO) {
//Do cheap tests first
if (this == thatO) return true;
if (!(thatO instanceof Element)) return false;
Element that = (Element) thatO;
if (!this.tagName_.equals(that.tagName_)) return false;
//!if( this.children_.size() != that.children_.size() )
//! return false;
int thisAttrCount = this.attributes_ == null ? 0 : this.attributes_.size();
int thatAttrCount = that.attributes_ == null ? 0 : that.attributes_.size();
if (thisAttrCount != thatAttrCount) return false;
//Compare attributes ignoring order (we already know the
//number is the same)
if (attributes_ != null) for (Enumeration i = attributes_.keys(); i.hasMoreElements();) {
String key = (String) i.nextElement();
String thisValue = (String) this.attributes_.get(key);
//non-null
String thatValue = (String) that.attributes_.get(key);
//maybe null
if (!thisValue.equals(thatValue)) return false;
}
//Compare children in order (we already know the number is the same)
Node thisChild = this.firstChild_;
Node thatChild = that.firstChild_;
while (thisChild != null) {
if (!thisChild.equals(thatChild)) return false;
thisChild = thisChild.getNextSibling();
thatChild = thatChild.getNextSibling();
}
return true;
}
/** Called whenever cached version of hashCode needs to be regenerated. */
protected int computeHashCode() {
int hash = tagName_.hashCode();
if (attributes_ != null) for (Enumeration i = attributes_.keys(); i.hasMoreElements();) {
String key = (String) i.nextElement();
hash = 31 * hash + key.hashCode();
String value = (String) attributes_.get(key);
hash = 31 * hash + value.hashCode();
}
for (Node i = firstChild_; i != null; i = i.getNextSibling())
hash = 31 * hash + i.hashCode();
return hash;
}
private void checkInvariant() {
if (DEBUG) {
if (tagName_ != Sparta.intern(tagName_)) throw new Error("tagname not interned");
if (attributeNames_ != null)
for (Enumeration i = attributeNames_.elements(); i.hasMoreElements();) {
String name = (String) i.nextElement();
if (name.indexOf(' ') != -1) throw new Error("Bad attribute " + name);
}
}
}
/**
* @link aggregation
* @label firstChild
*/
private Node firstChild_ = null;
/**
* @link aggregation
* @label lastChild
*/
private Node lastChild_ = null;
//Profiling found that these hashtables were using a lot of memory, so create them lazily
private Hashtable attributes_ = null; //create on first setAttribute
private Vector attributeNames_ = null; //create on first setAttribute
private String tagName_ = null;
}
// $Log: Element.java,v $
// Revision 1.10 2003/10/28 00:25:11 eobrain
// Whitespace change only.
//
// Revision 1.9 2003/07/17 23:53:19 eobrain
// Make compatiblie with J2ME. For example do not use "new"
// java.util classes.
//
// Revision 1.8 2003/06/24 23:54:32 eobrain
// Fix bug that was causing duplicate identical text-node children to be deleted.
// Remove unnecessary hashset of children.
//
// Revision 1.7 2003/06/19 20:28:20 eobrain
// Hash code optimization.
// Add monitoring (in debug mode) to detect when indexing could optimize.
//
// Revision 1.6 2003/05/12 19:56:36 eobrain
// Performance improvements including interning of tagname and attribute values.
//
// Revision 1.5 2003/01/27 23:30:58 yuhongx
// Replaced Hashtable with HashMap.
//
// Revision 1.4 2002/12/13 23:09:24 eobrain
// Fix javadoc.
//
// Revision 1.3 2002/12/13 18:12:17 eobrain
// Fix xpathEnsure to handle case when the XPath given specifies a root node tagname that conflicts with the existing root node. Extend xpathEnsure to work with any type of predicate. Replace hacky string manipulation code with code that works on the XPath parse tree.
//
// Revision 1.2 2002/10/30 16:37:25 eobrain
// Fix Element.appendChild so that it cannot create invalid loop by
// adding ancestor. No longer throw DOMException. (WARNING: This breaks
// backwards compatibility. Client code that catches or propagates
// DOMException because of this call will now give a compile error.)
//
// Revision 1.1.1.1 2002/08/19 05:03:55 eobrain
// import from HP Labs internal CVS
//
// Revision 1.23 2002/08/18 04:20:52 eob
// Sparta no longer throws XPathException -- it throws ParseException
// instead.
//
// Revision 1.22 2002/08/15 23:40:22 sermarti
//
// Revision 1.21 2002/08/15 21:29:00 eob
// Constructor no longer needs documenent.
//
// Revision 1.20 2002/08/15 05:08:21 eob
// Notify observers.
//
// Revision 1.19 2002/07/25 21:10:15 sermarti
// Adding files that mysteriously weren't added from Sparta before.
//
// Revision 1.18 2002/06/21 00:25:47 eob
// Make work with old JDK 1.1.*
//
// Revision 1.17 2002/06/15 22:16:49 eob
// Comment change only. Fix javadoc problem.
//
// Revision 1.16 2002/06/14 19:37:15 eob
// Make toString of Node do the same as in XSLT -- recursive
// concatenation of all text in text nodes.
//
// Revision 1.15 2002/05/23 21:22:44 eob
// Better error reporting.
//
// Revision 1.14 2002/05/11 00:10:18 eob
// Remove stub method that is now implemented with a slightly different
// name in Node.
//
// Revision 1.13 2002/05/10 21:03:19 eob
// Add equals method.
//
// Revision 1.12 2002/05/09 16:48:40 eob
// Add replaceChild. Fix cloneChild.
//
// Revision 1.11 2002/03/28 01:23:18 jrowson
// fixed bugs related to client side caching
//
// Revision 1.10 2002/03/26 01:45:39 eob
// Deprecate XPathAPI
//
// Revision 1.9 2002/02/23 02:04:44 eob
// Add clone method. Tweak toXml API.
//
// Revision 1.8 2002/02/15 21:20:10 eob
// Rename xpath* methods to xpathSelect* to make more obvious.
//
// Revision 1.7 2002/02/15 21:05:55 eob
// Add convenient xpath* methods, allowing a more object-oriented use than
// XPathAPI.
//
// Revision 1.6 2002/02/01 21:50:27 eob
// Move toXml up to Node
//
// Revision 1.5 2002/01/05 07:53:28 eob
// Factor out some functionality into Node.
//
// Revision 1.4 2002/01/04 00:37:50 eob
// add annotation
//
// Revision 1.3 2002/01/04 16:49:17 eob
// Fix indentation.
//
// Revision 1.2 2002/01/04 16:47:59 eob
// Move parse functionality functionality to ParseSource.
//
// Revision 1.1 2001/12/19 05:52:38 eob
// initial