利用httpclient模拟站点的登录发帖回复

利用httpclient模拟站点的登录发帖回复

   本文采用 HttpClient 来模拟站点的登录发帖回复，介绍 HttpClient 的用法和常见问题的解决方案。

HttpClient 是 Apache Jakarta Common 下的子项目，可以用来提供支持 HTTP 协议的客户端编程工具包，模拟浏览器的行为。它提供了很多的方法来简化网络的访问，虽然大部分的功能可以使用较底层的 java.net.HttpURLConnection 来实现。例如：

• 实现了所有 HTTP 的方法（ GET,POST 等）
• 支持 HTTPS 协议
• 支持代理服务器
• 自动维护 Cookies 等 

我们知道， http 协议是面向无连接的，要维持会话，现在基本上都是采用基于 Cookies 的方式（ Session 机制也是通过 Cookies 实现的），所以 HttpClient 的自动维护 Cookies 的方式对我们的登录发帖回复非常有用（一般网站都需要先登录再发帖回复）。

下面的 例子都是采用 commons-httpclient-3.1.jar 包来实现的（虽然 commons-httpclient-4.0 已经发布，但是代码发生了较大的重构，调用方式也发生了很大的改变）。

       下载 jar 包的路径为： http://hc.apache.org/downloads.cgi

        由于 httpclient 使用了 Apache Jakarta common 下的子项目 logging 和 codec ，所以也需要在 http://commons.apache.org/ 下载这两个包：

        commons-logging.jar

        commons-codec-1.3.jar

 

       为了更好地理解代码，设计的 UML 类图如下：
    

　　
    方法调用的时序图如下：

    其中， BrowserContext 类代表浏览器上下文对象，维护 HttpClient 链接和 Cookies 。 KaixinSitePost 是负责实现开心网的具体登录发帖回复逻辑的类。

    BrowserContext 的代码如下：

 

  1 /** */ /**
  2 *  Copyright (C): 2009
  3 *  @author 陈新汉
  4 *  Aug 24, 2009 3:09:00 PM
  5   */
  6
  7 /** */ /**
  8 * 浏览器进程上下文
  9   */
10 public   class BrowserContext
11 {
12     private HttpClient client; // 注意：每个站点和每个用户，对应一个单独的BrowserContext对象
13     private Cookie[] cookies = new Cookie[ 0 ]; // 维护Cookies
14     private Proxyips proxyip = null ; // 当前的代理IP
15     private Siteusers user = null ; // 当前的登录用户
16    
17     public Cookie[] getCookies() {
18         return cookies;
19     }
20    
21     public   void setCookies(Cookie[] cookies) {
22         this .cookies = cookies;
23     }
24    
25     public   void addCookie(Cookie c) {
26         if (cookies != null   && cookies.length > 0 ) {
27             Cookie[] others = new Cookie[cookies.length + 1 ];
28             System.arraycopy(cookies, 0 , others, 0 , cookies.length);
29             others[others.length - 1 ] = c;
30             cookies = others;
31         } else {
32             cookies = new Cookie[ 1 ];
33             cookies[ 0 ] = c;
34         }
35     }
36    
37     public Proxyips getProxyip() {
38         return proxyip;
39     }
40    
41     public   void setProxyip(Proxyips proxyip) {
42         this .proxyip = proxyip;
43         if ( this .proxyip != null ) {
44             client.getHostConfiguration().setProxy(proxyip.getIp(),proxyip.getPort());  
45             client.getParams().setAuthenticationPreemptive( true );  
46             // 如果代理需要密码验证，这里设置用户名密码  
47             // client.getState().setProxyCredentials(AuthScope.ANY, new UsernamePasswordCredentials("","")); 
48         }
49     }
50    
51     public HttpClient getClient() {
52         return client;
53     }
54    
55     public Siteusers getUser() {
56         return user;
57     }
58
59     public   void setUser(Siteusers user) {
60         this .user = user;
61     }
62
63     private BrowserContext(Site site) {
64         super ();
65         Protocol myhttps =   new Protocol( " https " , new MySecureProtocolSocketFactory(), 443 );
66         Protocol.registerProtocol( " https " , myhttps);
67         client =   new HttpClient();
68         client.getParams().setCookiePolicy(CookiePolicy.BROWSER_COMPATIBILITY);
69         HttpConnectionManagerParams managerParams = client.getHttpConnectionManager().getParams();
70         // 设置连接超时时间(单位毫秒)
71         // managerParams.setConnectionTimeout(50000);
72         // 设置读数据超时时间(单位毫秒)
73         // managerParams.setSoTimeout(120000);
74         initForSiteVisit(site.getSite(),site.getPort(),site.getCharset());
75     }
76    
77     public BrowserContext(Site site,Proxyips proxyip) {
78         this (site);
79         this .setProxyip(proxyip);
80     }
81
82     private   void initForSiteVisit(String siteurl, int port,String charset) {
83         client.getHostConfiguration().setHost(siteurl, port, " http " );
84         // 解决中文乱码问题，和指定网站的页面编码一致
85         client.getParams().setParameter(HttpMethodParams.HTTP_CONTENT_CHARSET, charset);
86     }
87    
88         // 查看cookie信息
89     public   void printCookies()
90     {
91         System.out.println( " ---------------Cookie---------------- " );
92         if (cookies != null ) {
93             for (Cookie c:cookies) {
94                 System.out.println(c.getName() + " : " + c.getValue());
95             }
96         } else {
97             System.out.println( " 没有设置Cookies " );
98         }
99         System.out.println( " ---------------Cookie---------------- " );
100     }
101    
102     public   void setCommonMethodRequestHeaders(HttpMethodBase method)
103     {
104         method.setRequestHeader( " Accept " , " */* " );
105         // method.setRequestHeader("Accept-Language", "zh-cn");
106         // method.setRequestHeader("Accept-Encoding", "gzip,deflate");
107         method.setRequestHeader( " User-Agent " , " Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1;) " );
108         // 设置非常重要
109         method.setRequestHeader( " Connection " , " Keep-Alive " );
110     }
111    
112     public String redirectToURL(String url) throws IOException
113     { 
114         if (url != null ) {
115             try {
116                 System.out.println( " 页面重定向到： " + url);
117                 String responseString = this .doCommonVisitWithURL(url);
118                 // System.out.println(responseString);
119                 return responseString;
120             } catch (IOException e) {
121                 System.out.println( " 重定向： " + url + " 出错 " );
122             }
123         } else {
124             System.out.println( " redirect url is null " );
125         }
126         return   null ;
127     }
128    
129     public String doCommonVisitWithURL(String url) throws IOException {
130         GetMethod get =   new GetMethod(url);
131         return   this .doGet(get);
132     }
133    
134     public String doPost(ExpectContinueMethod post) throws IOException
135     {
136         if (post == null )
137             return   null ;
138         try  
139         {
140             if (getCookies() != null ) {
141                 // printCookies();
142                 client.getState().addCookies(cookies);
143                 post.addRequestHeader( " Cookie " ,getCookies().toString());
144                 // System.out.println(post.getRequestHeader("Cookie").getValue());
145             }
146             setCommonMethodRequestHeaders(post);
147             int statusCode = client.executeMethod(post);
148             cookies = client.getState().getCookies();
149             System.out.println(statusCode);
150             // System.out.println(post.getResponseHeader("Location"));
151             String responseString = post.getResponseBodyAsString();
152             System.out.println(responseString);
153             printCookies();
154             post.releaseConnection();
155             if (statusCode ==   301   || statusCode == 302 ) {
156                 redirectToURL(post.getResponseHeader( " Location " ).getValue());
157             }
158             return responseString;
159         }  
160         finally   {
161             if (post !=   null )
162                 post.releaseConnection();
163         }
164     }
165    
166     public String doGet(GetMethod get) throws IOException
167     {
168         if (get == null )
169             return   null ;
170         if (cookies != null ) {
171             // printCookies();
172             client.getState().addCookies(cookies);
173             get.addRequestHeader( " Cookie " ,cookies.toString());
174         }
175         try {
176             setCommonMethodRequestHeaders(get);
177             int statusCode = client.executeMethod(get);
178             cookies = client.getState().getCookies();  // 重新保存Cookies
179             printCookies();
180             System.out.println(statusCode);
181             if (statusCode ==   301   || statusCode == 302 ) {
182                 redirectToURL(get.getResponseHeader( " Location " ).getValue());
183             }
184             String responseString = get.getResponseBodyAsString();
185             // System.out.println(responseString);
186             return responseString;
187         }
188         finally {
189             if (get != null )
190                 get.releaseConnection();
191         }
192     }
193    
194     public String getRedirectURL(String content)
195     {
196         if (content != null   && content.indexOf( " window.location=\ "" )!=-1){
197             int begin = content.indexOf( " window.location=\ "" );
198             int end = content.indexOf( " \ "" , begin+17);
199             return content.substring(begin + 17 ,end);
200         }
201         return   null ;
202     }
203 }

KaixinSitePost类的代码：

  1 /** */ /**
  2 *  Copyright (C): 2009
  3 *  @author 陈新汉
  4 *  Aug 14, 2009 11:16:32 AM
  5 */
  6
  7 /** */ /**
  8 * 模拟测试网站（不需要验证码）
  9 * 开心网（www.kaixin.com）
10 */
11 public   class KaixinSitePost implements ISitePost
12 {
13    private static final String LOGON_SITE = "www.kaixin.com";
14    private static final int LOGON_PORT = 80;
15    private static final String CHARSET="UTF-8";
16    private BrowserContext context=null;
17
18   
19    //单个用户登录
20    public String login(Siteusers userinfo, Proxyips ip)
21    {
22        if(userinfo!=null)
23        {
24            SiteLogin login=new SiteLogin(context,"http://login.kaixin.com/Login.do");
25            if(ip!=null)
26                login.getContext().setProxyip(ip);
27            Map<String,String> params=new HashMap<String,String>();
28            params.put("ss", "10106");
29            params.put("loginregFrom", "index");
30            params.put("origURL", "http://www.kaixin.com/SysHome.do");
31            params.put("email", userinfo.getUsername());
32            params.put("password", userinfo.getUserpwd());
33            login.addRequestParameters(params);
34            return login.login(userinfo);
35        }
36        return null;
37    }
38   
39    public List<Siteboards> parseBoard(Siteboards data) {
40       
41        return null;
42    }
43
44    public String post(Postinfos postinfo,List<Siteboards> siteboards)
45    {
46        if(postinfo!=null && siteboards!=null){
47            SitePost sport=new SitePost(context);
48            context.getClient().getHostConfiguration().setHost("blog.kaixin.com");
49            Map<String,String> params=new HashMap<String,String>();
50            params.put("categoryId", "0");
51            params.put("blogControl", "1");
52            params.put("title", postinfo.getTitle());
53            params.put("body",postinfo.getContent());
54            sport.addRequestParameters(params);
55           
56            for(Siteboards sb:siteboards){
57                sb.setPostUrl("http://blog.kaixin.com/NewEntry.do");
58                try{
59                    sport.post(postinfo, sb);
60                }catch(IOException e){
61                    e.printStackTrace();
62                }
63            }
64        }
65        return null;
66    }
67
68
69    public String reply(Postinfos postinfo,List<Articleinfos> arts)
70    {
71       
72        return null;
73    }
74
75    /** *//**
76     * @param args
77     */
78    public static void main(String[] args)
79    {
80        try
81        {
82            Siteusers userinfo=new Siteusers();
83            userinfo.setUsername("xxxx");
84            userinfo.setUserpwd("xxxx");
85            Proxyips ips = new Proxyips();
86            ips.setIp("218.56.64.210");
87            ips.setPort(8080);
88            KaixinSitePost sp=new KaixinSitePost();
89            sp.login(userinfo,ips);
90            Postinfos post=new Postinfos();
91            post.setContent("<p>lllllllllllllllllllllll</p>");
92            post.setTitle("中文测试");
93            List<Siteboards> siteboards=new ArrayList<Siteboards>();
94            siteboards.add(new Siteboards());
95            siteboards.add(new Siteboards());
96            sp.post(post,siteboards);
97        }
98        catch(Exception e){
99            e.printStackTrace();
100        }
101    }
102}

封装登录的类SiteLogin：

1 /** */ /**
2 *  Copyright (C): 2009
3 *  @author 陈新汉
4 *  Aug 24, 2009 3:03:00 PM
5 */
6
7 /** */ /**
8 * 站点登录
9 */
10 public   class SiteLogin extends AbstractMethodAdapter
11 {
12    private HttpMethodBase method;
13    private boolean ispost=true;
14    protected BrowserContext context; //当前的浏览器进程上下文
15   
16    public BrowserContext getContext() {
17        return context;
18    }
19   
20    /** *//**
21     * 构造函数
22     * @param context
23     * @param url
24     * @param ispost 设置是否POST方式提交，默认为POST
25     */
26    public SiteLogin(BrowserContext context,String url,boolean ispost) {
27        super();
28        this.context = context;
29        this.ispost=ispost;
30        method = this.ispost?new PostMethod(url):new GetMethod(url);
31    }
32   
33    public SiteLogin(BrowserContext context,String url) {
34        this(context,url,true);
35    }
36
37    public String login(Siteusers user)
38    {
39        int statusCode=0;
40        if(this.ispost && this.hasRequestParameters()){
41            ((PostMethod)method).setRequestBody(this.getRequestParams());
42        }
43       
44        if(this.hasExtraRequestHeaders()){
45            this.addExtraRequestHeaders(method,this.getExtraRequestHeaders());
46        }
47        context.setCommonMethodRequestHeaders(method);
48        try
49        {
50            if(context.getCookies()!=null){
51                //printCookies();
52                context.getClient().getState().addCookies(context.getCookies());
53                method.addRequestHeader("Cookie", context.getCookies().toString());
54            }
55            statusCode = context.getClient().executeMethod(method);
56            context.setCookies(context.getClient().getState().getCookies());
57            String responseString = method.getResponseBodyAsString();
58            //System.out.println(responseString);
59            method.releaseConnection();
60            if(statusCode==HttpStatus.SC_OK){
61                System.out.println("登录成功");
62                return responseString;
63            }
64            else if(statusCode==302 ||statusCode==301){
65                System.out.println("登录成功，页面重定向");
66                String url=method.getResponseHeader("Location").getValue();
67                return context.redirectToURL(url);
68            }
69            else{
70                System.out.println("登录失败,状态码："+statusCode);
71            }
72        }catch(Exception e){
73            e.printStackTrace();
74        }finally{
75            if(method!=null)
76                method.releaseConnection();
77        }
78        return null;
79    }
80}
81

封装站点发帖的类SitePost：

1 /** */ /**
2 *  Copyright (C): 2009
3 *  @author 陈新汉
4 *  Aug 24, 2009 5:05:55 PM
5 */
6
7 /** */ /**
8 * 站点发帖新帖
9 */
10 public   class SitePost extends CommonSitePost
11 {
12   
13    public SitePost(BrowserContext context) {
14        super();
15        this.context=context;
16    }
17   
18    public String post(Postinfos postinfo,Siteboards siteboard) throws IOException
19    {
20        if (postinfo != null && siteboard != null) {
21            if (StringUtils.isNotEmpty(siteboard.getPostUrl())) {
22                PostMethod post = new PostMethod(siteboard.getPostUrl());
23                if(this.hasRequestParameters()){
24                    post.setRequestBody(this.getRequestParams());
25                }
26                if(this.hasExtraRequestHeaders()){
27                    this.addExtraRequestHeaders(post,this.getExtraRequestHeaders());
28                }
29                context.setCommonMethodRequestHeaders(post);
30                this.context.doPost(post);
31            }else{
32                System.out.println("版面的新帖提交地址不能为空！");
33            }
34        }else{
35            System.out.println("帖子或者版面信息输入都不能为空");
36        }
37        return null;
38    }
39}
40

    友情提醒：本博文章欢迎转载，但请注明出处： 陈新汉