us.codecraft.webmagic.samples.NjuBBSProcessor Maven / Gradle / Ivy

Go to download

Show more of this group Show more artifacts with this name
Show all versions of webmagic-samples Show documentation

The newest version!

package us.codecraft.webmagic.samples;

import us.codecraft.webmagic.Page;
import us.codecraft.webmagic.Site;
import us.codecraft.webmagic.Spider;
import us.codecraft.webmagic.processor.PageProcessor;

import java.util.List;

/**
 * @author [email protected] 

 * Date: 13-4-21
 * Time: 下午8:08
 */
public class NjuBBSProcessor implements PageProcessor {
    @Override
    public void process(Page page) {
        List requests = page.getHtml().regex("]*href=(bbstcon\\?board=Pictures&file=[^>]*)").all();
        page.addTargetRequests(requests);
        page.putField("title",page.getHtml().xpath("//div[@id='content']//h2/a"));
        page.putField("content",page.getHtml().smartContent());
    }

    @Override
    public Site getSite() {
        return Site.me().setDomain("bbs.nju.edu.cn");
    }

    public static void main(String[] args) {
        Spider.create(new NjuBBSProcessor()).addUrl("http://bbs.nju.edu.cn/board?board=Pictures").run();
    }
}