模拟ajax实现网络爬虫——HtmlUnit

【推荐】2019 Java 开发者跳槽指南.pdf(吐血整理) >>>

最近在用Jsoup抓取某网站数据，可有些页面是ajax请求动态生成的，去群里问了一下，大神说模拟ajax请求即可。去网上搜索了一下，发现了这篇文章，拿过来先用着试试。转帖如下：

网上关于网络爬虫实现方式有很多种，但是很多都不支持Ajax，李兄说：模拟才是王道。确实，如果能够模拟一个没有界面的浏览器，还有什么不能做到的呢？关于解析Ajax网站的框架也有不少，我选择了HtmlUnit,官方网站：http://htmlunit.sourceforge.net /,htmlunit可以说是一个Java版本的无界面浏览器，几乎无所不能，而且很多东西都封装得特别完美。这是这几天来积累下来的心血，记录一下。
package com.lanyotech.www.wordbank;

import java.io.FileOutputStream;
import java.io.IOException;
import java.io.InputStream;
import java.io.OutputStream;
import java.net.MalformedURLException;
import java.util.List;
import com.gargoylesoftware.htmlunit.FailingHttpStatusCodeException;
import com.gargoylesoftware.htmlunit.NicelyResynchronizingAjaxController;
import com.gargoylesoftware.htmlunit.ScriptResult;
import com.gargoylesoftware.htmlunit.WebClient;
import com.gargoylesoftware.htmlunit.html.HtmlOption;
import com.gargoylesoftware.htmlunit.html.HtmlPage;
import com.gargoylesoftware.htmlunit.html.HtmlSelect;

public class WorldBankCrawl {

    private static String TARGET_URL = "http://databank.worldbank.org/ddp/home.do";

    public static void main(String[] args) throws FailingHttpStatusCodeException, MalformedURLException, IOException {
        //模拟一个浏览器
        WebClient webClient = new WebClient();
        //设置webClient的相关参数
        webClient.setJavaScriptEnabled(true);
        webClient.setCssEnabled(false);
        webClient.setAjaxController(new NicelyResynchronizingAjaxController());
        webClient.setTimeout(35000);
        webClient.setThrowExceptionOnScriptError(false);
        //模拟浏览器打开一个目标网址
        HtmlPage rootPage= webClient.getPage(TARGET_URL);
        //获取第一个数据库
        HtmlSelect hs = (HtmlSelect) rootPage.getElementById("lstCubes");
        //按要求选择第一个数据库
        hs.getOption(0).setSelected(true);
        //模拟点击Next按钮，跳转到第二个页面
        System.out.println("正在跳转…");
        //执行按钮出发的js事件
        ScriptResult sr = rootPage.executeJavaScript("javascript:setCubeData(2,-1,4,'/ddp');");

        //跳转到第二个页面，选择国家
        HtmlPage countrySelect = (HtmlPage) sr.getNewPage();
        //获得包含全部国家信息的选择框页面
        HtmlPage framePage=(HtmlPage)countrySelect.getFrameByName("frmTree1″).getEnclosedPage();
        //获得selectAll按钮，触发js事件
        framePage.executeJavaScript("javascript:TransferListAll(‘countrylst','countrylstselected','no');SetSelectedCount(‘countrylstselected','tdcount');");
        //获取Next按钮，触发js事件
        ScriptResult electricityScriptResult = framePage.executeJavaScript("javascript:wrapperSetCube('/ddp')");

        System.out.println("正在跳转…");
        //跳转到下一个页面electricitySelect
        HtmlPage electricitySelect = (HtmlPage) electricityScriptResult.getNewPage();
        //获得electricity选择的iframe
        HtmlPage electricityFrame = (HtmlPage) electricitySelect.getFrameByName("frmTree1″).getEnclosedPage();
        //获得选择框
        HtmlSelect seriesSelect = (HtmlSelect) electricityFrame.getElementById("countrylst");
        //获得所有的选择框内容
        List optionList = seriesSelect.getOptions();
        //将指定的选项选中
        optionList.get(1).setSelected(true);
        //模拟点击select按钮                       electricityFrame.executeJavaScript("javascript:TransferList('countrylst','countrylstselected','no');SetSelectedCount('countrylstselected','tdcount');");
        //获取选中后，下面的选择框
        HtmlSelect electricitySelected = (HtmlSelect) electricityFrame.getElementById("countrylstselected");
        List list = electricitySelected.getOptions();
        //模拟点击Next按钮，跳转到选择时间的页面
        ScriptResult timeScriptResult = electricityFrame.executeJavaScript("javascript:wrapperSetCube('/ddp')");

        System.out.println("正在跳转…");
        HtmlPage timeSelectPage = (HtmlPage) timeScriptResult.getNewPage();
        //获取选中时间的选择框
        timeSelectPage = (HtmlPage) timeSelectPage.getFrameByName("frmTree1″).getEnclosedPage();
        //选中所有的时间                timeSelectPage.executeJavaScript("javascript:TransferListAll('countrylst','countrylstselected','no');SetSelectedCount('countrylstselected','tdcount');");
        //点击Next按钮
        ScriptResult exportResult = timeSelectPage.executeJavaScript("javascript:wrapperSetCube('/ddp')");

        System.out.println("正在跳转…");
        //转到export页面
        HtmlPage exportPage = (HtmlPage) exportResult.getNewPage();
        //点击页面上的Export按钮,进入下载页面
        ScriptResult downResult = exportPage.executeJavaScript("javascript:exportData('/ddp' ,'EXT_BULK' ,'WDI_Time=51||WDI_Series=1||WDI_Ctry=244||' );");

        System.out.println("正在跳转…");
        HtmlPage downLoadPage = (HtmlPage) downResult.getNewPage();
        //点击Excel图标，开始下载
        ScriptResult downLoadResult = downLoadPage.executeJavaScript("javascript:exportData('/ddp','BULKEXCEL');");
        //下载Excel文件
        InputStream is = downLoadResult.getNewPage().getWebResponse().getContentAsStream();

        OutputStream fos = new FileOutputStream("d://test.xls");
        byte[] buffer=new byte[1024*30];
        int len=-1;
        while((len=is.read(buffer))>0){
            fos.write(buffer, 0, len);
        }
        fos.close();
        fos.close();
        System.out.println("Success!");
    }
}

来源：oschina

链接：https://my.oschina.net/u/731762/blog/140729

标签

ajax

网络爬虫

javascript

htmlunit