package com.canrd.webmagic.processor;

import com.canrd.webmagic.processor.config.Agent;
import com.canrd.webmagic.processor.download.SeleniumDownloader;
import com.canrd.webmagic.processor.pipeline.ArticlePipeline;
import lombok.extern.slf4j.Slf4j;
import org.apache.logging.log4j.core.util.UuidUtil;
import org.springframework.stereotype.Component;
import us.codecraft.webmagic.Page;
import us.codecraft.webmagic.Site;
import us.codecraft.webmagic.Spider;
import us.codecraft.webmagic.processor.PageProcessor;
import us.codecraft.webmagic.selector.Selectable;
import us.codecraft.webmagic.selector.XpathSelector;

import javax.annotation.Resource;
import java.util.List;

/**
 * https://www.univie.ac.at/suche/?q=battery
 *
 * @author: xms
 * @description: TODO
 * @date: 2024/4/1 14:19
 * @version: 1.0
 */
@Slf4j
@Component
public class UnivieSearchPageProcessor implements PageProcessor {

    @Resource
    private Univie4PhysnanoArticlePageProcessor univie4PhysnanoArticlePageProcessor;

    @Resource
    private SeleniumDownloader seleniumDownloader;

    @Resource
    private ArticlePipeline articlePipeline;

    /**
     * 抓取网站的相关配置，包括编码、抓取间隔、重试次数等
     */
    private Site site = Site.me().setRetryTimes(3).setSleepTime(5).setUserAgent(Agent.getRandom());

    /**
     * 定制爬虫逻辑的核心接口，在这里编写抽取逻辑
     *
     * @param page
     */
    @Override
    public void process(Page page) {
        doSearch(page);
    }

    /**
     * @param page
     */
    private void doSearch(Page page) {
        String url = page.getUrl().get();
        /**
         * 通过page.getHtml()可以获取到main函数中Spider.create(new BaiduHotSearchPageProcessor()).addUrl中的地址的网页内容
         * 1、通过$或css()方法获取到该page html下某元素dom
         */
        Selectable selectable = page.getHtml().xpath("//div[@class='univie-search']").xpath("//div[@class='content-element-margin-small yacy-result']");
        List<Selectable> nodes = selectable.nodes();

        /**
         * 获取到指定的dom后，从这些dom中提取元素内容。
         */
        for (int i = 0; i <= nodes.size() - 1; i++) {
            String link = nodes.get(i).links().get();
            String title = nodes.get(i).xpath("//a/h2/text()").get();
            if (link.contains(".pdf") || link.contains(".docx")) {
                continue;
            }
            if (link.contains("physnano.univie.ac.at/publications/publication-detail/pure")) {
                log.info("关键字文章列表链接：{},标题:{},文章链接：{}", url, title, link);
                Spider.create(univie4PhysnanoArticlePageProcessor)
                        .addUrl(link)
                        .addPipeline(articlePipeline)
//                        .setDownloader(seleniumDownloader)
                        .setUUID(UuidUtil.getTimeBasedUuid().toString())
                        // 开启5个线程执行，并开始爬取
                        .thread(1).run();
            }
        }

    }


    @Override
    public Site getSite() {
        return site;
    }
}