利用HtmlParser 完成含有html标签的字符截取并补全标签

截取一定长度带有HTML标签的文件内容，如果按普通文字用subtring进行截取的，当把html标签载断而非完整闭合时，整个页面都可能因此而变形， 利用HtmlParser可以很好的保留原HTML标签并将载段的标签补齐。

应该注意一点<strong>***</strong>标签需要重新定义，因为现在strong标签的getEndTag() 是null不知道是不是htmlparser的一个bug

package com.test.util;

import java.util.Stack;
import org.htmlparser.tags.CompositeTag;

import org.htmlparser.Node;
import org.htmlparser.Parser;
import org.htmlparser.Tag;
import org.htmlparser.nodes.TagNode;
import org.htmlparser.nodes.TextNode;
import org.htmlparser.util.NodeIterator;
import org.htmlparser.util.ParserException;

public class SubstringHTML {

	private final String CONTENT;
	private Parser parser;
	private Stack<TagNode> nodeStack;
	private int subLength;
	private int textLength = 0;
	private int pos = 0;

	public SubstringHTML(String content) {
		CONTENT = content;

		parser = Parser.createParser(content, "GBK");
		factory = new PrototypicalNodeFactory();
		factory.registerTag(new StrongTag());
		parser.setNodeFactory(factory);

		nodeStack = new Stack<TagNode>();
	}

	private void recusive(NodeIterator iterator) throws ParserException  {
		
		while (iterator.hasMoreNodes()) {
			Node node = iterator.nextNode();
			if (node instanceof TagNode) {
				TagNode tagNode = (TagNode)node;
				Tag tag = tagNode.getEndTag();
				
				if (tag != null) {
					nodeStack.push(tagNode);
				}
			}
			else if (node instanceof TextNode) {
				if (node.getText().trim().length() == 0) {
					continue;
				}
				
                if (node.getTagName.equals("SCRIPT") {
                	continue;      
                }

				String nodeText = node.getText();
				int tLen = nodeText.length();
				if ((textLength < subLength) && ((textLength + tLen) > subLength)) {
					pos = node.getStartPosition() + subLength - textLength;
					textLength = subLength;
					return;
				}
				else {
					textLength += tLen;
					pos = node.getEndPosition();
				}
			}
			
			if (node.getChildren() == null) {
				continue;
			}
			recusive(node.getChildren().elements());
			
			if (subLength <= textLength) {
				return;
			}
		}
	}

	public String subString(int length, String end) {
		if (length >= CONTENT.length() || length <= 0) {
			return CONTENT;
		}

		subLength = length;
		try {
			recusive(parser.elements());
		} catch (ParserException e) {
			System.out.println("parser error:" + e.getMessage());
			return CONTENT;
		}

		int size = nodeStack.size();
		StringBuffer buffer = new StringBuffer();
		buffer.append(CONTENT.substring(0, pos));

		while (size > 0) {
			TagNode node = nodeStack.pop();
			size--;

			if (node.getEndTag().getEndPosition() <= pos || node.getTagBegin() >= pos) {
				continue;
			}

			buffer.append("</");
			buffer.append(node.getTagName());
			buffer.append(">");
		}

		buffer.append(end);
		return buffer.toString();
	}

	private static String getContent() {
		byte[] con = null;
		InputStream in = SubstringHTML.class.getResourceAsStream("content.txt");
		try {
			int length = in.available();
			con = new byte[length];
			in.read(con, 0, length);
		} catch (IOException e) {
			e.printStackTrace();
		}

		try {
			return new String(con, "GBK");
		} catch (UnsupportedEncodingException e) {
			e.printStackTrace();
			return "";
		}
	}

	public static void main(String[] args) {
		String content = getContent();
		SubstringHTML app = new SubstringHTML(content);
		String str = app.subString(200, "");
		System.out.println(str);
	}
}

class StrongTag extends CompositeTag {
	private static final long serialVersionUID = 1L;
	private static final String[] mIds = new String[] { "STRONG" };
        private static final String[] mEndTagEnders = new String[] {"BODY", "HTML"};

	public String[] getIds() {
		return mIds;
	}

	public String[] getEndTagEnders ()
	{
		return (mEndTagEnders);
	}
}

评论

1 楼 javalucky 2010-03-03  

怎么不加一下注释啊