登录校内网并读取新鲜事

今天想到不登录校内，看看有什么新鲜事，没有感兴趣的就不想到网页登录了，以后可以慢慢扩展更多功能，顺便再用一下HttpUrlConnection这个东东.

还用到了htmparser这个包包

 

我登录后目前是把home.do和feed保存到一个.html文件中了。

 

代码如下：

 

import java.io.BufferedReader; import java.io.FileWriter; import java.io.IOException; import java.io.InputStream; import java.io.InputStreamReader; import java.io.PrintStream; import java.net.HttpURLConnection; import java.net.URL; import java.net.URLEncoder; import org.htmlparser.Node; import org.htmlparser.NodeFilter; import org.htmlparser.Parser; import org.htmlparser.filters.NodeClassFilter; import org.htmlparser.filters.OrFilter; import org.htmlparser.nodes.TextNode; import org.htmlparser.tags.Div; import org.htmlparser.tags.LinkTag; import org.htmlparser.util.NodeList; /** * This is a demo class to login xiaonei.com * @author chao yin * Feb 20 2009 */ public class HttpLogin { public static final String feedFile = "d:/feed.html"; public static final String homeFile = "d:/home.html"; /** * @param args * @throws IOException */ public static void main(String[] args) throws IOException { System.out.println("login..."); String urlStr = "http://login.xiaonei.com/Login.do"; String post = "email=" + URLEncoder.encode("xxxx@yahoo.com.cn", "UTF-8") + "&password=" + "xxxxxxxxxxx"; Login(urlStr, post, ""); System.out.println("login success."); } /** * Use HttpUrlConnection Login * @param urlStr index page * @param param parameter * @param homeUrl after login redirect page * @param outPath output file path */ public static void Login(String urlStr, String param, String homeUrl) { URL indexUrl = null; HttpURLConnection httpUrlConn = null; InputStream in = null; try { indexUrl = new URL(urlStr); httpUrlConn = (HttpURLConnection) indexUrl.openConnection(); httpUrlConn.setRequestMethod("POST"); HttpURLConnection.setFollowRedirects(false); httpUrlConn.setInstanceFollowRedirects(false); httpUrlConn.setDoOutput(true); httpUrlConn.setDoInput(true); httpUrlConn.setRequestProperty("User-Agent", "Mozilla/5.0 (compatible; MSIE 6.0; Windows NT)"); httpUrlConn.setRequestProperty("Content-Type", "application/x-www-form-urlencoded"); //post parameter PrintStream sendPost = new PrintStream(httpUrlConn.getOutputStream()); sendPost.print(param); sendPost.close(); httpUrlConn.disconnect(); URL homeURL = null; if (httpUrlConn.getHeaderField("Location") != null) { homeURL = new URL(httpUrlConn.getHeaderField("Location")); } else { if (homeUrl != null || homeUrl != "") { homeURL = new URL(homeUrl); } } //get cookie String cookies = getCookies(httpUrlConn); System.out.println("Cookie: " + cookies); //reconnect home page httpUrlConn = (HttpURLConnection) homeURL.openConnection(); httpUrlConn.setRequestProperty("User-Agent", "Mozilla/5.0 (compatible; MSIE 6.0; Windows NT)"); httpUrlConn.setRequestProperty("Cookie", cookies); httpUrlConn.setDoInput(true); System.out.println(httpUrlConn.getResponseCode() + " : "+ httpUrlConn.getResponseMessage()); in = httpUrlConn.getInputStream(); BufferedReader reader = new BufferedReader(new InputStreamReader( in, "utf-8")); StringBuffer sb = new StringBuffer(); String inputLine; while ((inputLine = reader.readLine()) != null) { sb.append(inputLine); sb.append("/n"); } //write all the home page to file writeToFile(homeFile, sb.toString()); //write feed to file writeToFile(feedFile, (readFeed(sb.toString()))); } catch (Exception e) { e.printStackTrace(); } } /** * * @param http * @return */ public static String getCookies(HttpURLConnection http) { String cookieVal = null; String key = null; String sessionId = null; for (int i = 1; (key = http.getHeaderFieldKey(i)) != null; i++) { if (key.equalsIgnoreCase("set-cookie")) { cookieVal = http.getHeaderField(i); cookieVal = cookieVal.substring(0, cookieVal.indexOf(";")); sessionId = sessionId + cookieVal + ";"; } } return sessionId.toString(); } /** * * @param path * @param content * @throws IOException */ public static void writeToFile(String path, String content) throws IOException { FileWriter fileWriter = new FileWriter(path); System.out.println("write "+path); fileWriter.write(content); fileWriter.flush(); fileWriter.close(); } /** * * @param result * @return * @throws Exception */ public static String readFeed(String result) throws Exception { Parser parser; NodeList nodelist; parser = Parser.createParser(result, "utf8"); NodeFilter divFilter = new NodeClassFilter(Div.class); NodeFilter textFilter = new NodeClassFilter(TextNode.class); NodeFilter linkFilter = new NodeClassFilter(LinkTag.class); OrFilter lastFilter = new OrFilter(); lastFilter.setPredicates(new NodeFilter[] { divFilter, textFilter, linkFilter }); nodelist = parser.parse(lastFilter); Node[] nodes = nodelist.toNodeArray(); String feed = ""; for (int i = 0; i < nodes.length; i++) { Node node = nodes[i]; if (node instanceof Div) { Div divNode = (Div) node; String className = divNode.getAttribute("class"); if (className != null) { if (className.equals("title")) { feed += divNode.getChildrenHTML() + "/r/n"; } } } } return feed; } }

 

 

登录过程控制台信息如下：

 

login...
Cookie: nullkl=null;societyguester=null;login_email=null;_de=D84BCAA0A66D8BEEF2C59733A8D819731967A5E9EF81591C696BF75400CE19CC;userid=25645639677;univid=16004;gender=1;univyear=2005;WebOnLineNotice_22755546345967=1;societyguester=75fdddf76604368d0015c9136d446cda7;kl=41948313cfa227dc6e136fcf8a9b27a4_225439677;hostid=22657639677;BIGipServerpool_profile=3737916938.20480.0000;
200 : OK
write d:/home.html
write d:/feed.html
login success.