文件名称:
Jsoup+httpclient 模拟登陆和抓取
开发工具:
文件大小: 406kb
下载次数: 0
上传时间: 2013-07-12
详细说明： Jsoup+httpclient 模拟登陆和抓取页面 package com.app.html; import java.io.BufferedReader; import java.io.BufferedWriter; import java.io.File; import java.io.FileOutputStream; import java.io.FileReader; import java.io.IOException; import java.io.OutputStreamWriter; import java.io.Writer; import java.text.SimpleDateFormat; import java.util.Date; import org.apache.commons.httpclient.Cookie; import org.apache.commons.httpclient.HttpClient; import org.apache.commons.httpclient.NameValuePair; import org.apache.commons.httpclient.cookie.Cookie Policy; import org.apache.commons.httpclient.cookie.CookieSpec; import org.apache.commons.httpclient.methods.PostMethod; import org.apache.commons.httpclient.params.HttpMethodParams; import org.jsoup.Jsoup; import org.jsoup.nodes.Document; import org.jsoup.nodes.Element; import org.jsoup.select.Elements; import com.app.comom.FileUtil; public class HttpClientHtml { private static final String SITE = "login.goodjobs.cn"; private static final int PORT = 80; private static final String loginAction = "/index.php/action/UserLogin"; private static final String forwardURL = "http://user.goodjobs.cn/dispatcher.php/module/Personal/?skip_fill=1"; private static final String toUrl = "d:\\test\\"; private static final String css = "http://user.goodjobs.cn/personal.css"; private static final String Img = "http://user.goodjobs.cn/images"; private static final String _JS = "http://user.goodjobs.cn/scripts/fValidate/fValidate.one.js"; /** * 模拟等录 * @param LOGON_SITE * @param LOGON_PORT * @param login_Action * @param params * @throws Exception */ private static HttpClient loginHtml(String LOGON_SITE, int LOGON_PORT,String login_Action,String ...params) throws Exception { HttpClient client = new HttpClient(); client.getHostConfiguration().setHost(LOGON_SITE, LOGON_PORT); // 模拟登录页面 PostMethod post = new PostMethod(login_Action); NameValuePair userName = new NameValuePair("memberName",params[0] ); NameValuePair password = new NameValuePair("password",params[1] ); post.setRequestBody(new NameValuePair[] { userName, password }); client.executeMethod(post); post.releaseConnection(); // 查看cookie信息 CookieSpec cookiespec = CookiePolicy.getDefaultSpec(); Cookie[] cookies = cookiespec.match(LOGON_SITE, LOGON_PORT, "/", false, client.getState().getCookies()); if (cookies != null) if (cookies.length == 0) { System.out.println("Cookies is not Exists "); } else { for (int i = 0; i < cookies.length; i++) { System.out.println(cookies[i].toString()); } } return client; } /** * 模拟等录 后获取所需要的页面 * @param client * @param newUrl * @throws Exception */ private static String createHtml(HttpClient client, String newUrl) throws Exception { SimpleDateFormat format = new SimpleDateFormat("yyyy-MM-dd"); String filePath = toUrl + format.format(new Date() )+ "_" + 1 + ".html"; PostMethod post = new PostMethod(newUrl); client.executeMethod(post); //设置编码 post.getParams().setParameter(HttpMethodParams.HTTP_CONTENT_CHARSET, "GBK"); String content= post.getResponseBodyAsString(); FileUtil.write(content, filePath); System.out.println("\n写入文件成功!"); post.releaseConnection(); return filePath; } /** * 解析html代码 * @param filePath * @param random * @return */ private static String JsoupFile(String filePath, int random) { SimpleDateFormat format = new SimpleDateFormat("yyyy-MM-dd"); File infile = new File(filePath); String url = toUrl + format.format(new Date()) + "_new_" + random+ ".html"; try { File outFile = new File(url); Document doc = Jsoup.parse(infile, "GBK"); String html=""; StringBuffer sb = new StringBuffer(); sb.append(html).append("\n"); sb.append("").append("\n"); sb.append("").append("\n"); sb.append("欢迎使用新安人才网个人专区").append("\n"); Elements meta = doc.getElementsByTag("meta"); sb.append(meta.toString()).append("\n"); ////////////////////////////body////////////////////////// Elements body = doc.getElementsByTag("body"); ////////////////////////////link////////////////////////// Elements links = doc.select("link");//对link标签有href的路径都作处理 for (Element link : links) { String hrefAttr = link.attr("href"); if (hrefAttr.contains("/personal.css")) { hrefAttr = hrefAttr.replace("/personal.css",css); Element hrefVal=link.attr("href", hrefAttr);//修改href的属性值 sb.append(hrefVal.toString()).append("\n"); } } ////////////////////////////script////////////////////////// Elements scripts = doc.select("script");//对script标签 for (Element js : scripts) { String jsrc = js.attr("src"); if (jsrc.contains("/fValidate.one.js")) { String oldJS="/scripts/fValidate/fValidate.one.js";//之前的css jsrc = jsrc.replace(oldJS,_JS); Element val=js.attr("src", jsrc);//修改href的属性值 sb.append(val.toString()).append("\n").append(""); } } ////////////////////////////script////////////////////////// Elements tags = body.select("*");//对所有标签有src的路径都作处理 for (Element tag : tags) { String src = tag.attr("src"); if (src.contains("/images")) { src = src.replace("/images",Img); tag.attr("src", src);//修改src的属性值 } } sb.append(body.toString()); sb.append(""); BufferedReader in = new BufferedReader(new FileReader(infile)); Writer out = new BufferedWriter(new OutputStreamWriter( new FileOutputStream(outFile), "gbk")); String content = sb.toString(); out.write(content); in.close(); System.out.println("页面已经爬完"); out.close(); } catch (IOException e) { e.printStackTrace(); } return url; } public static void main(String[] args) throws Exception { String [] params={"admin","admin123"}; HttpClient client = loginHtml(SITE, PORT, loginAction,params); // 访问所需的页面 String path=createHtml(client, forwardURL); System.out.println( JsoupFile(path,1)); } } ...展开收缩
(系统自动生成,下载前可以参看下载内容)
下载文件列表
相关说明
- 本站资源为会员上传分享交流与学习,如有侵犯您的权益,请联系我们删除.
- 本站是交换下载平台，提供交流渠道，下载内容来自于网络，除下载问题外，其它问题请自行百度。
- 本站已设置防盗链，请勿用迅雷、QQ旋风等多线程下载软件下载资源，下载后用WinRAR最新版进行解压.
- 如果您发现内容无法下载，请稍后再次尝试；或者到消费记录里找到下载记录反馈给我们.
- 下载后发现下载的内容跟说明不相乎，请到消费记录里找到下载记录反馈给我们，经确认后退回积分.
- 如下载前有疑问，可以通过点击"提供者"的名字，查看对方的联系方式，联系对方咨询.
