以下是示例代码

import java.lang.reflect.InvocationTargetException;

import java.lang.reflect.Method;

import java.net.HttpURLConnection;

import java.net.URL;



import org.htmlparser.Node;

import org.htmlparser.PrototypicalNodeFactory;

import org.htmlparser.Tag;

import org.htmlparser.lexer.Lexer;

import org.htmlparser.lexer.Page;

import org.htmlparser.nodes.TagNode;

import org.htmlparser.nodes.TextNode;

import org.htmlparser.tags.ScriptTag;

import org.htmlparser.tags.StyleTag;

import org.htmlparser.util.ParserException;



import com.lietu.tag.CnTagMaker.WordWeight;



public class TagExt {

    private static StringBuffer body ;

    private static String title ;



    public static void main(String[] args) throws Exception {

       String path;



       if (0 >= args. length ) {

           path =

           /**   以下是测试地址   **/

           // "http://www.ibm.com/developerworks/cn/webservices/0901_haoxf_humantask/";

           // "http://developers.sun.com.cn/Java/xref_index.html";

           "http://mil.news.sina.com.cn/2009-01-09/0839538126.html" ;

           // "http://www.sina.com.cn/";

           // "http://www.ibm.com/";

           // "http://hao861002.javaeye.com/blog/301581";

       } else {

           path = args[0];

       }

      

       /**

         * 构造 URL ，并打开网络链接。

         */

       URL url = new URL(path);

       HttpURLConnection httpUrl = (HttpURLConnection) url.openConnection();

      

       /**   对该网页进行解析   **/

       parseHTML (httpUrl);



       /**   打印解析后的内容   **/

       System. out .println( "body=" + body );

    }

   

    /**

      * 解析网页内容

      * @param uc   传入一个 HttpURLConnection 链接对象

      * @throws ParserException

      */

    public static void parseHTML(HttpURLConnection uc) throws ParserException {

       /**   声明节点   **/

       Node node;



       String stringText;

       body = new StringBuffer();



       /**   从 head 头获取网页编码格式。该方式取决于服务器是否设置 charSet 值，如果没有，该方式将无法获取 charSet 值   **/

       String contentType = uc.getContentType();  

       String charSet = getCharset (contentType);

       Lexer lexer = null ;



       if (charSet == null ) {

           charSet = "UTF-8" ;

       }



       try {

           lexer = new Lexer( new Page(uc.getInputStream(), charSet));

       } catch (Exception e) {

           e.printStackTrace();

           return ;

       }

      

       /**   对网页内容进行解析   **/

       lexer.setNodeFactory( new PrototypicalNodeFactory());

      

       /**   设置开关，决定网页是否重新解析   **/

       boolean tryAgain = false ;

      

       while ( null != (node = lexer.nextNode())) {

           /**   以下是判断节点的类型，并作相应的处理   **/

           if (node instanceof ScriptTag) {

              while ( null != (node = lexer.nextNode())) {

                  if (node instanceof Tag) {

                     Tag tag = (Tag) node;

                    if (tag.isEndTag() && "SCRIPT" .equals(tag.getTagName())) {

                         break ;

                     }

                  }

              }

              if ( null == node)

                  break ;

           } else if (node instanceof StyleTag) {

              while ( null != (node = lexer.nextNode())) {

                  if (node instanceof Tag) {

                     Tag tag = (Tag) node;

                     if (tag.isEndTag())

                         break ;

                  }

              }

              if ( null == node)

                  break ;

           } else if (node instanceof TextNode) {

              stringText = node.toPlainTextString();

              if ( "" .equals( title ))

                  continue
else if (node instanceof TextNode) {

              stringText = node.toPlainTextString();

              if ( "" .equals( title ))

                  continue ;

              stringText = stringText.replaceAll( "[ \t\n\f\r 　 ]+" , " " );

              stringText = TextHtml.html2text (stringText.trim());



              if (! "" .equals(stringText)) {

                  body .append(stringText);

                  body .append( " " );

              }

           } else if (node instanceof TagNode) {

              TagNode tagNode = (TagNode) node;

              String name = ((TagNode) node).getTagName();

              if (name.equals( "TITLE" ) && !tagNode.isEndTag()) {

                  node = lexer.nextNode();

                  stringText = node.toPlainTextString().trim();

                  if (! "" .equals(stringText)) {

                     title = stringText;

                  }

              } else if (name.equals( "META" )) {

                  String contentCharSet = tagNode.getAttribute( "CONTENT" );

                  // System.out.println("contentCharset="+contentCharSet);

                  int b = contentCharSet.toLowerCase().indexOf( "charset" );

                  if (b > -1) {

                     String newCharSet = getCharset (contentCharSet);

                     // System.out.println("newCharSet=" + newCharSet);

                     if (!charSet.equals(newCharSet)) {

                         tryAgain = true ;

                         charSet = newCharSet;

                         // System.out.println("charSet=" + charSet);

                         // System.out.println("newCharSet=" + newCharSet);

                         break ;

                     }

                  }

              }

           }

       }



       /**   如果在 Meta 信息中检测到新的字符编码，则需要按照 meta 信息中的编码再次解析网页。   **/

       if (tryAgain) {

           body = new StringBuffer();



           try {

                uc = (HttpURLConnection) uc.getURL().openConnection();

              lexer = new Lexer( new Page(uc.getInputStream(), charSet));

           } catch (Exception e) {

              e.printStackTrace();

           }



           lexer.setNodeFactory( new PrototypicalNodeFactory());



           while ( null != (node = lexer.nextNode())) {

              if (node instanceof TextNode) {

                  stringText = node.toPlainTextString();

                  if ( "" .equals( title ))

                     continue ;

                  stringText = stringText.replaceAll( "[ \t\n\f\r 　 ]+" , " " );

                  stringText = TextHtml.html2text (stringText.trim());

                  if (! "" .equals(stringText)) {

                     body .append(stringText);

                     body .append( " " );

                  }

              }

           }

       }

    }

   

    /**

      * 找出最终的网页编码

      * @param name 经过 getCharset 方法处理后 meta 标签的值

      * @param _default 默认的编码集

      * @return

      */

    public static String findCharset(String name, String _default) {

       String ret;



       try {

           Class<java.nio.charset.Charset> cls;

           Method method;

           Object object;



           cls = java.nio.charset.Charset. class ;

           method = cls.getMethod( "forName" , new Class[] { String. class });

           object = method.invoke( null , new Object[] { name });

           method = cls.getMethod( "name" , new Class[] {});

           object = method.invoke(object, new Object[] {});

           ret = (String) object;

       } catch (NoSuchMethodException nsme) {

           ret = name;

       } catch (IllegalAccessException ia) {

           ret = name;

       } catch (InvocationTargetException ita) {

           ret = _default;

           System. out

                  .println( "unable to determine cannonical charset name for "

                         + name + " - using " + _default);

       }



       return (ret);

    }
/**

      * 处理 meta 中的内容，并调用 findCharset() 方法获取编码值

      * @param content Meta 中的内容

      * @return

      */

    public static String getCharset(String content) {

       final String CHARSET_STRING = "charset" ;

       int index;

       String ret;



       ret = null ;

       if ( null != content) {

           index = content.indexOf(CHARSET_STRING);



           if (index != -1) {

              content = content.substring(index + CHARSET_STRING.length())

                     .trim();

              if (content.startsWith( "=" )) {

                  content = content.substring(1).trim();

                  index = content.indexOf( ";" );

                  if (index != -1)

                     content = content.substring(0, index);



                  if (content.startsWith( "\"" ) && content.endsWith( "\"" )

                         && (1 < content.length()))

                     content = content.substring(1, content.length() - 1);



                  if (content.startsWith( "'" ) && content.endsWith( "'" )

                         && (1 < content.length()))

                     content = content.substring(1, content.length() - 1);



                  ret = findCharset (content, ret);

              }

           }

       }



       return (ret);

    }



}



说明：


           该类使用 Boolean 型 tryAgain 来判断是否进行第二次解析！



       如果通过 String contentType = uc.getContentType() 来获得 http 协议中 Head 头中 contentType 中的值，并使用 String charSet = getCharset(contentType); 从 contentType 中获取 charSet 值。如果此时 charSet 获取的为空，则先用 charSet=UTF-8 （即：默认情况下， charSet 值为 UTF-8 ）进行解析。使用 org.htmlparser.tags 包中的类对比节点 node ，并对合适的节点的内容进行处理，本例中是对类型为 TextNode 的 node 节点，获取其中的内容，并将其添加到类型为 StringBuffer 的 body 当中。当解析到 <meta> 标签的时候，会试图从 <meta> 中获取 newCharSet 值，并将 newCharSet 和 charSet 进行比较，如果相同，则不再进行解析，主函数将 body 打印。如果不同，说明默认的的 charSet 值并不是网页的编码格式，刚才的解析可能不正确，会出现乱码。这时候，将 newCharSet 中的值赋给 charSet ，并将 tryAgain 值设为 true 。当程序进行到 if(tryAgain){ …… } 判断时，结果为真，则根据新得到的 charSet 值进行二次解析。这时候，我们就会得到正确的解析结果