getracedate.java

java写的读取香港马术比赛数据,分析并导入到EXCEL的源码,可供研究用。

package hkjc2.logic;

import java.util.ArrayList;
import java.util.List;

import org.htmlparser.Node;
import org.htmlparser.NodeFilter;
import org.htmlparser.Parser;
import org.htmlparser.Tag;
import org.htmlparser.tags.OptionTag;
import org.htmlparser.tags.SelectTag;
import org.htmlparser.util.NodeList;

public class GetRaceDate {

	/**
	 * 读取赛事日期
	 * @param client
	 * @return
	 * @throws Exception 
	 */
	public List<String> getRaceDate(HttpUtil client) throws Exception {
		List<String> result = new ArrayList<String>();
		String url = "http://www.hkjc.com/chinese/racing/Results.asp";
		String html = client.getHTML(url);
		Parser parser = new Parser();
		parser.setInputHTML(html);
        NodeList nl = parser.extractAllNodesThatMatch(racedateFilter);
        if (nl.size() > 0) {
        	SelectTag select = (SelectTag)nl.elementAt(0);
        	OptionTag[] options = select.getOptionTags();
        	for (int i=0;i<options.length;i++) {
        		result.add(options[i].getValue());
        	}
        }
		return result;
	}
	
	private NodeFilter racedateFilter = new NodeFilter() {
		private static final long serialVersionUID = 7301987153616374408L;
		public boolean accept(Node node) {
            if (node instanceof Tag) {
                Tag tag = (Tag) node;
                if (tag instanceof SelectTag && "raceDate".equals(tag.getAttribute("name"))) {
                    return true;
                }
            }
            return false;
        }
    };
    
    public static void main(String[] args) throws Exception {
    	GetRaceDate get = new GetRaceDate();
    	HttpUtil client = new HttpUtil();
    	List<String> result = get.getRaceDate(client);
    	for (int i=0;i<result.size();i++) {
    		System.out.println(result.get(i));
    	}
    }
}