一个使用Neko提取HTML纯文本内容的程序例子

import org.apache.html.dom.HTMLDocumentImpl;import org.cyberneko.html.parsers.DOMFragmentParser;import org.w3c.dom.Document;import org.w3c.dom.Node;import org.w3c.dom.DocumentFragment;import org.w3c.dom.NodeList;import org.w3c.dom.html.HTMLDocument;import org.xml.sax.InputSource;import java.net.*;

public class NekoTextExtractor { public DocumentFragment getDocument(InputStream is){  DocumentFragment node=null;  try{   DOMFragmentParser parser=new DOMFragmentParser();   node=new HTMLDocumentImpl().createDocumentFragment();   parser.parse(new InputSource(is),node);  }catch(Exception ex){   ex.printStackTrace();  }  return node; }  public DocumentFragment getDocument(Reader reader){  DocumentFragment node=null;  try{   DOMFragmentParser parser=new DOMFragmentParser();   node=new HTMLDocumentImpl().createDocumentFragment();   parser.parse(new InputSource(reader),node);  }catch(Exception ex){   ex.printStackTrace();  }  return node; }      public void getText(StringBuffer sb,Node node)throws Exception{  if(enabledNode(node)){   if(node.getNodeType()==Node.TEXT_NODE){    String value=node.getNodeValue();    sb.append(value.trim());   }   else{    NodeList nodes=node.getChildNodes();    for(int i=0;i<nodes.getLength();i++){     Node n=nodes.item(i);     getText(sb,n);    }   }  } }  public String extractText(String source)throws Exception{  StringReader reader=new StringReader(source);  return extractText(reader); }  public String extractText(InputStream is)throws Exception{  Node node=getDocument(is);  StringBuffer sb=new StringBuffer();  getText(sb,node);  return sb.toString(); }  public String extractText(Reader reader)throws Exception{  Node node=getDocument(reader);  StringBuffer sb=new StringBuffer();  getText(sb,node);  return sb.toString(); }  public static void main(String[] args)throws Exception {  URL url=new URL("http://news.sina.com.cn");  HttpURLConnection uc=(HttpURLConnection)url.openConnection();  StringBuffer sb=new StringBuffer();  NekoTextExtractor nte=new NekoTextExtractor();  Node node=nte.getDocument(uc.getInputStream());  nte.getText(sb,node);  System.out.println(sb.toString()); }  public boolean enabledNode(Node node){  if(node.getNodeName().equalsIgnoreCase("script"))   return false;  else if(node.getNodeName().equalsIgnoreCase("link"))   return false;  else if(node.getNodeName().equalsIgnoreCase("style"))   return false;  else if(node.getNodeType()==Node.COMMENT_NODE)   return false;  else   return true; }  }