webmagic ajax,webmagic 爬虫分页

全栈程序员-站长 • 2025年8月15日上午11:15 • 未分类 • 阅读 2

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.JsonPathSelector;

import us.codecraft.webmagic.selector.Selectors;

import java.util.List;

* @author

* @since 0.5.0

public class YidianzixunProcessor implements PageProcessor {

private Site site = Site.me();

// . ? 等特殊符号需要加\\引用符

private static final String ARTICLE_URL = “http://www\\.yidianzixun\\.com/home\\?page=article&id=\\w+”;

public void process(Page page) {

if (page.getUrl().regex(ARTICLE_URL).match()) {

String content = page.getHtml().xpath(“//div[@class=’content’]”).toString();

//1)文章头部:1、标题、2、发表时间、3、文章来源

String content_hd = Selectors.xpath(“//div[@class=’content-hd’]”).select(content);

String title = Selectors.xpath(“//h2/text()”).select(content_hd);

String date = Selectors.xpath(“//div[@class=’meta’]/span[@class=’date’]/text()”).select(content_hd);

String source = Selectors.xpath(“//a[@id=’source-name’]/text()/text()”).select(content_hd);

//2)文章主题

String body = Selectors.xpath(“//div[@class=’content-bd’]”).select(content);

System.out.println(body);

} else {

List results = new JsonPathSelector(“$.result[*]”).selectList(page.getRawText());

for(String result:results){

// /home?page=article&id={
{docid}}&up={
{up}}#comment

// /home?page=article&id={
{docid}} 也是可以支持

String docid = new JsonPathSelector(“docid”).select(result);

String targetUrl = “http://www.yidianzixun.com/home?page=article&id=”+docid;

page.addTargetRequest(targetUrl);

}

public Site getSite() {

return site;

}

public static void main(String[] args) {

// url的特殊符号 ‘|’ : %7C

//通过ajax判断请求的参数

String targetUrl = “http://www.yidianzixun.com/api/q/?path=channel%7Cnews-list-for-keyword&display=%E8%8A%B1%E8%8C%B6&word_type=token&fields=docid&fields=category&fields=date&fields=image&fields=image_urls&fields=like&fields=source&fields=title&fields=url&fields=comment_count&fields=summary&fields=up&cstart=10&cend=20&version=&infinite=true”;

Spider.create(new YidianzixunProcessor()).addUrl(targetUrl).thread(5).run();

}