飞云小侠的个人博客

在使用普通字符串时,使用substring就可以进行截取部分字符串 (当然还要考虑多国语言的问题)

但是对于Html字符串来说,如果采用同样的方法,则会破坏html标签,造成页面错乱,经过对HtmlParser的研究,写出了一个类,可以对Html字符串进行截取.

可以自己根据实际情况改进,如果考虑多国语言,也要修改字符串长度的计算方法. 总之要灵活使用,随机应变,而不是照搬照抄.

下面贴出此类的内容,使用了开源项目Html Parser. 

import org.apache.commons.logging.Log;
import org.apache.commons.logging.LogFactory;
import org.htmlparser.*;
import org.htmlparser.tags.CompositeTag;
import org.htmlparser.util.NodeIterator;
import org.htmlparser.util.NodeList;
 
/**
 * Functions for HTML.
 *
 * @author Scud http://www.javascud.org  Date: Nov 3, 2006 10:22:20 AM
 */
public class HtmlSubstring
{
    private static Log log = LogFactory.getLog(HtmlSubstring.class);
 
    /**
     * get parser for substring.
     * @return Parser
     */
    public static Parser getMyParser()
    {
        Parser parser = new Parser();
 
        PrototypicalNodeFactory factory = new PrototypicalNodeFactory();
 
        //register tags which htmlParser not have
        factory.registerTag(new StrongTag());
        factory.registerTag(new BoldTag());
        factory.registerTag(new ItalicTag());
        factory.registerTag(new UnderlineTag());
        factory.registerTag(new CenterTag());
        factory.registerTag(new FontTag());
 
        parser.setNodeFactory(factory);
 
        return parser;
    }
 
    /**
     * Substring for Html String.
     *
     * @param htmlString Html string
     * @param maxlength  maxlength
     * @return String
     */
    public static String substring(String htmlString, int maxlength)
    {
        StringBuffer htmlOut = new StringBuffer();
        StringBuffer stringOut = new StringBuffer();
 
        try
        {
            Parser parser = getMyParser();
 
            parser.setInputHTML(htmlString);
 
            NodeIterator nit = parser.elements();
 
            boolean breaked = false;
 
            while (nit.hasMoreNodes())
            {
                Node node = nit.nextNode();
                if (node instanceof Text)
                if (node instanceof Text)
                {
                    breaked = dealText(node, stringOut, htmlOut, maxlength);
                }
                else if (node instanceof Tag)
                {
                    Tag tag = (Tag) node;
                    breaked = dealTag(tag, stringOut, htmlOut, maxlength);
                }
                else if (node instanceof Remark)
                {
                    //nothing to do
                }
                if (breaked)
                {
                    break;
                }
            }
        }
        catch (Exception e)
        {
            log.error("Error occured when parse Html String", e);
        }
 
        return htmlOut.toString();
    }
 
    private static boolean dealText(Node node, StringBuffer stringOut, StringBuffer htmlOut, int maxlength)
    {
        String currentText = node.getText();
        int previousLength = stringOut.length();
 
        if (previousLength + currentText.length() >= maxlength)
        {
            String cutString = currentText.substring(0, maxlength - previousLength);
 
            stringOut.append(cutString);
            htmlOut.append(cutString);
 
            log.debug(cutString);
 
            return true;
        }
        else
        {
            stringOut.append(node.getText());
            htmlOut.append(node.getText());
            log.debug(node.getText());
        }
 
        return false;
    }
 
    private static boolean dealTag(Tag aTag, StringBuffer stringOut, StringBuffer htmlOut, int maxlength) throws Exception
    {
        NodeList list = aTag.getChildren();
 
        log.debug(getStartTagString(aTag));
 
        htmlOut.append(getStartTagString(aTag));
 
        boolean breaked = false;
 
        if (list != null)
        {
            NodeIterator it = list.elements();
 
            while (it.hasMoreNodes())
            {
                Node node = it.nextNode();
                if (node instanceof Text)
                {
                    breaked = dealText(node, stringOut, htmlOut, maxlength);
                }
                else if (node instanceof Tag)
                {
                    Tag tag = (Tag) node;
                    breaked = dealTag(tag, stringOut, htmlOut, maxlength);
                }
                else if (node instanceof Remark)
                {
                    //nothing to do
                }
 
                if (breaked)
                {
                    break;
                }
            }
        }
 
        Tag endTag = aTag.getEndTag();
        if (endTag != null)
        {
            htmlOut.append(aTag.getEndTag().toHtml());
            log.debug(aTag.getEndTag().toHtml());
        }
 
        return breaked;
    }
 
    private static String getStartTagString(Tag aTag)
    {
        StringBuffer start = new StringBuffer("<");
 
        for (Object o : aTag.getAttributesEx())
        {
            Attribute ab = (Attribute) o;
            start.append(ab.toString());
        }
 
        start.append(">");
        return start.toString();
    }
 
 
}
 
class StrongTag extends CompositeTag
{
    private static final String[] mIds = new String[]{"STRONG"};
 
    public StrongTag()
    {
    }
 
    public String[] getIds()
    {
        return (mIds);
    }
 
    public String[] getEnders()
    {
        return (mIds);
    }                            
 
    public String[] getEndTagEnders()
    {
        return (new String[0]);
    }
}
 
class BoldTag extends CompositeTag
{
    private static final String[] mIds = new String[]{"B"};
 
    public BoldTag()
    {
    }
 
    public String[] getIds()
    {
        return (mIds);
    }
 
    public String[] getEnders()
    {
        return (mIds);
    }
 
    public String[] getEndTagEnders()
    {
        return (new String[0]);
    }
}
 
class ItalicTag extends CompositeTag
{
    private static final String[] mIds = new String[]{"I"};
 
    public ItalicTag()
    {
    }
 
    public String[] getIds()
    {
        return (mIds);
    }
 
    public String[] getEnders()
    {
        return (mIds);
    }
 
    public String[] getEndTagEnders()
    {
        return (new String[0]);
    }
}
 
class UnderlineTag extends CompositeTag
{
    private static final String[] mIds = new String[]{"U"};
 
    public UnderlineTag()
    {
    }
 
    public String[] getIds()
    {
        return (mIds);
    }
 
    public UnderlineTag()
    {
    }
 
    public String[] getIds()
    {
        return (mIds);
    }
 
    public String[] getEnders()
    {
        return (mIds);
    }
 
    public String[] getEndTagEnders()
    {
        return (new String[0]);
    }
}
 
class CenterTag extends CompositeTag
{
    private static final String[] mIds = new String[]{"CENTER"};
 
    public CenterTag()
    {
    }
 
    public String[] getIds()
    {
        return (mIds);
    }
 
    public String[] getEnders()
    {
        return (mIds);
    }
 
    public String[] getEndTagEnders()
    {
        return (new String[0]);
    }
}
 
class FontTag extends CompositeTag
{
    private static final String[] mIds = new String[]{"FONT"};
 
    public FontTag()
    {
    }
 
    public String[] getIds()
    {
        return (mIds);
    }
 
    public String[] getEnders()
    {
        return (mIds);
    }
 
    public String[] getEndTagEnders()
    {
        return (new String[0]);
    }
} 

• 二月 2008 [1]
• 十二月 2007 [2]
• 十月 2007 [1]
• 九月 2007 [1]
• 六月 2007 [1]
• 五月 2007 [2]
• 十一月 2006 [2]
• 九月 2006 [1]
• 七月 2006 [2]
• 六月 2006 [4]
• 五月 2006 [13]
• 四月 2006 [6]
• 十一月 2005 [1]
• 九月 2005 [1]

• 开发手记 [20] 
• WebWork [3] 
• 其他 [10] 
• LifeType [3] 
• 服务配置 [2] 

• 硅谷创业公司 BeyondMedia 高薪招聘 (北京)
• Resin在Windows下设置内存参数
• Apache加Resin时配置仅部分URL使用Resin
• Thinkpad T61 感觉不错
• 股市有风险, 投资须谨慎
• 3G 内存
• 植物园的郁金香 Tulip in Arboretum(2007-4-21)
• Freemarker + Tiles/SiteMesh

• 风吹散发猫: 打游戏或许会提高性能；也可以玩VMWare虚拟机运行N个linux系统。。。
  
• 风吹散发猫: 都不要说啦，我作java开发6年啦～～...
  
• CAnca: 不能截取啊。。。
  
• 小柔: 在那里可以下载FreeMarker-IDE。我找不到啊
  
• ss: 我也喜欢，
  

• 使用FreeMarker-IDE,辅助提示WebWork的Tag语法(6545)
• Webwork in action 中文版面世了(4320)
• 紫光输入法5.0发布了(3953)
• WebWork 2.2在Apache+Resin上发布的一点注意事项(3662)
• 买了一个PDA:HP 6828(3319)

• 我的个人网站
• JavaScud开源平台