searcher.java~2~

来自「搜索引擎的研究与实现(Java)(含源码)」· JAVA~2~ 代码 · 共 63 行

package news;

/**
 * <p>Title: 新闻搜索引擎</p>
 * <p>Description: 毕业设计</p>
 * <p>Copyright: Copyright (c) 2003</p>
 * <p>Company: </p>
 * @author 计算机99630 沈晨
 * @version 1.0
 */

import com.heaton.bot.HTTP;
import com.heaton.bot.HTTPSocket;
import com.heaton.bot.ISpiderReportable;
import com.heaton.bot.IWorkloadStorable;
import com.heaton.bot.Spider;
import com.heaton.bot.SpiderInternalWorkload;

public class Searcher
    implements ISpiderReportable {
  public static void main(String[] args) throws Exception {

   //队列在内存中
   IWorkloadStorable wl = new SpiderInternalWorkload();

    Searcher _searcher = new Searcher();

    //构造Spider程序
    Spider _spider
        = new Spider(_searcher, "http://www.jamessc.com/news.htm",
                     new HTTPSocket(), 100, wl);
    _spider.setMaxBody(100);
    _spider.start();
  }

  public boolean foundInternalLink(String url) {
    return false;
  }

  public boolean foundExternalLink(String url) {
    return false;
  }

  public boolean foundOtherLink(String url) {
    return false;
  }

  public void processPage(HTTP http) {
    System.out.println(http.getURL());
    //new HTMLParse(http).start();
  }

  public void completePage(HTTP http, boolean error) {
  }

  public boolean getRemoveQuery() {
    return true;
  }

  public void spiderComplete() {
  }
}