JAXP处理XML文件的时候，把空格也当作text节点？

Question

有下面这样的Java程序：import java.io.File;
import java.io.IOException;
// JAXP
import javax.xml.parsers.FactoryConfigurationError;
import javax.xml.parsers.ParserConfigurationException;
import javax.xml.parsers.DocumentBuilderFactory;
import javax.xml.parsers.DocumentBuilder;
// DOM
import org.w3c.dom.Document;
import org.w3c.dom.Node;
import org.w3c.dom.NodeList;
import org.xml.sax.SAXException;public class ParseXml
{   public ParseXml()
   {
      // TODO Auto-generated constructor stub
   }   public static void main(String[] args)
   {
      try {
         // Get Document Builder Factory
         DocumentBuilderFactory factory = DocumentBuilderFactory.newInstance();
         // Turn on validation, and turn off namespaces
         factory.setValidating(false);
         factory.setNamespaceAware(false);
         factory.setIgnoringElementContentWhitespace(true);
         DocumentBuilder builder = factory.newDocumentBuilder();
         Document doc = builder.parse(new File(".\\src\\input.xml"));
         // Print the document from the DOM tree and
         // feed it an initial indentation of nothing
         printNode(doc, "");
      } catch (ParserConfigurationException e)
      {
         System.out.print("ParserConfigurationException occurred: ");
         System.out.println("The underlying parser does not " + "support the requested features.");
      } catch (FactoryConfigurationError e)
      {
         System.out.print("FactoryConfigurationError occurred: ");
         System.out.println("Error occurred obtaining Document " + "Builder Factory.");
      } catch (IOException e)
      {
         System.out.print("IOException occurred: ");
         e.printStackTrace();
      } catch (SAXException e)
      {
         System.out.print("SAXException occurred: ");
         e.printStackTrace();
      }
   }   private static void printNode(Node node, String indent)
   {
      // print the DOM tree
      short nodeType = node.getNodeType();
      switch(nodeType)
      {
         case Node.DOCUMENT_NODE:
            break;
            
         case Node.ELEMENT_NODE:
            System.out.println(indent + node.getNodeName());
            break;
            
         case Node.TEXT_NODE:
            String nodeValue = node.getNodeValue();
            if (nodeValue != null)
               System.out.println(indent + node.getNodeName() + ": " + nodeValue);
            break;
            
         default:
            break;
      }
      
      NodeList nl = node.getChildNodes();
      for(int index = 0; index < nl.getLength(); index ++)
      {
         printNode(nl.item(index), indent + "   ");
      }
   }}这段程序解析下面的XML文件，并打印出各个节点的名称和值：<?xml version="1.0" encoding="gbk" standalone="yes"?>
<family>
   <fathor>
      <name>kingfox</name>
      <age>40</age>
   </fathor>
   <mothor>
      <name>apple</name>
      <age>32</age>
   </mothor>
   <daughter>
      <name>ivy</name>
      <age>8</age>
   </daughter>
   <brothor type="text">
      <name>biu</name>
      <age>36</age>
   </brothor>
</family>输出确是这样的结果：   family
      #text: 
   
      fathor
         #text: 
      
         name
            #text: kingfox
         #text: 
      
         age
            #text: 40
         #text: 
   
      #text: 
   
      mothor
         #text: 
      
         name
            #text: apple
         #text: 
      
         age
            #text: 32
         #text: 
   
      #text: 
   
      daughter
         #text: 
      
         name
            #text: ivy
         #text: 
      
         age
            #text: 8
         #text: 
   
      #text: 
   
      brothor
         #text: 
      
         name
            #text: biu
         #text: 
      
         age
            #text: 36
         #text: 
   
      #text: 
那些#text节点和其后的空行，其实就是对应了每个XML节点前面的空格。难道Java的XML处理器会将用于XML文件格式化的那些空格也当作text节点来处理？如何设置能够忽略这些空格？

Answer 1

好像用JDOM或DOM4j来解析就不会有你说的问题了。

Answer 2

JDOM/DOM4J也一样，虽然在JDOM/DOM4J中调用node.getChildren()方法，不会返回Text信息；但如果需要节点的Text信息，可以调用node.getText()，这个方法也会返回\t\n等这些特殊字符。

Answer 3

节点之间的换行和空白符，始末节点之间的值都是Text node，之间输出就可以了
对Document节点，取得它的第一个子节点，也就是跟节点进行递归输出。
对Element节点可以对的下一个兄弟节点NextSibling进行递归输出。
其他节点就返回。private static void printNode(Node node)
{
// print the DOM tree
short nodeType = node.getNodeType();
switch(nodeType)
{
case Node.DOCUMENT_NODE:
Node n = node.getFirstChild(); printNode(n); return; case Node.ELEMENT_NODE:
System.out.print("<" + node.getNodeName()+">");

NodeList nl = node.getChildNodes();
for(int index = 0; index < nl.getLength(); index ++)
{
printNode(nl.item(index));
} System.out.print("<"+node.getNodeName()+"/>");
break;
case Node.TEXT_NODE:
System.out.print(node.getNodeValue());
return; default:
System.out.print("--" + node.getNodeType());
return;
}
Node ns = node.getNextSibling();
if(ns!=null)
printNode(ns); }<family>
   <fathor>
      <name>kingfox<name/>
      
      <age>40<age/>
   
   <fathor/>
   
   <mothor>
      <name>apple<name/>
      
      <age>32<age/>
   
   <mothor/>
   
   <daughter>
      <name>ivy<name/>
      
      <age>8<age/>
   
   <daughter/>
   
   <brothor>
      <name>biu<name/>
      
      <age>36<age/>
   
   <brothor/><family/>

Answer 4

谢谢5楼，不过这种方法还是没有解决那些空格和回车字符的问题。后来找到了一种办法，对于text节点，取出其nodeValue之后，调用trim函数即可去掉头尾的空格和回车换行符号。现在的代码如下：
import java.io.File;
import java.io.IOException;
// JAXP
import javax.xml.parsers.FactoryConfigurationError;
import javax.xml.parsers.ParserConfigurationException;
import javax.xml.parsers.DocumentBuilderFactory;
import javax.xml.parsers.DocumentBuilder;
// DOM
import org.w3c.dom.Document;
import org.w3c.dom.Element;
import org.w3c.dom.Node;
import org.w3c.dom.NodeList;
import org.xml.sax.*;public class ParseXml
{   public ParseXml()
   {
   }   public static void main(String[] args)
   {
      ParseXml px = new ParseXml();
      px.run();
   }
   
   public void run()
   {
      try {
         // Get Document Builder Factory
         DocumentBuilderFactory factory = DocumentBuilderFactory.newInstance();
         // Turn on validation, and turn off namespaces
         factory.setValidating(false);
         factory.setNamespaceAware(false);
         factory.setIgnoringElementContentWhitespace(true);
         DocumentBuilder builder = factory.newDocumentBuilder();
         Document doc = builder.parse(new File(".\\src\\input.xml"));
         Element rootElement = doc.getDocumentElement();
         // Print the document from the DOM tree and
         // feed it an initial indentation of nothing
         printNode(rootElement, "");
      } catch (ParserConfigurationException e)
      {
         System.out.print("ParserConfigurationException occurred: ");
         System.out.println("The underlying parser does not " + "support the requested features.");
      } catch (FactoryConfigurationError e)
      {
         System.out.print("FactoryConfigurationError occurred: ");
         System.out.println("Error occurred obtaining Document " + "Builder Factory.");
      } catch (IOException e)
      {
         System.out.print("IOException occurred: ");
         e.printStackTrace();
      } catch (SAXException e)
      {
         System.out.print("SAXException occurred: ");
         e.printStackTrace();
      }
   }   private void printNode(Node node, String indent)
   {
      // print the DOM tree
      short nodeType = node.getNodeType();      switch (nodeType)
      {
         case Node.DOCUMENT_NODE:
            break;         case Node.ELEMENT_NODE:
            String nodeName = node.getNodeName();
            System.out.println(indent + "<" + nodeName + ">");
            if (node.hasChildNodes())
            {
               Node childNode = node.getFirstChild();
               do {
                  printNode(childNode, indent + "   ");
                  childNode = childNode.getNextSibling();
               } while(childNode != null);
            }
            System.out.println(indent + "</" + nodeName + ">");
            break;         case Node.TEXT_NODE:
            String nodeValue = node.getNodeValue();
            if (!nodeValue.trim().equals(""))
            {
               System.out.println(indent + "   " + nodeValue);
            }
            break;         default:
            break;
      }
   }}输出如下：<family>
   <fathor>
      <name>
            kingfox
      </name>
      <age>
            40
      </age>
      <favourite>
      </favourite>
   </fathor>
   <mothor>
      <name>
            apple
      </name>
      <age>
            32
      </age>
   </mothor>
   <daughter>
      <name>
            ivy
      </name>
      <age>
            8
      </age>
   </daughter>
   <brothor>
      <name>
            biu
      </name>
      <age>
            36
      </age>
   </brothor>
</family>