使用 Ruby r实现如何登录新浪微博

看了网上有用 Python 登录的，不过有点过时了，现在加密算法也变了，所以就想自己写一个登录程序。

然后把 ruby 也移植到我的 Linux 开发板上，然后就可以定时登录....... ,

原来一直想用 PHP 实现，可以PHP 操作 http 用 curl 库，总觉着别扭，不够强大！

 

 

登陆分为 三步

1：prelogin

发送用户名，密码前，先从新浪服务器获取 三个主要的变量

 

• servertime
• nonce
• rsakv(RSA的Public Key，下一步用这个public key 加密密码)

 

#pre login
preloginurl = 'http://login.sina.com.cn/sso/prelogin.php?entry=sso&' +
'callback=sinaSSOController.preloginCallBack&su=' +
'dW5kZWZpbmVk' + '&rsakt=mod&client=ssologin.js(v1.4.2)' +
"&_=" + Time.now.to_i.to_s
#puts preloginurl
res = HttpGet(preloginurl)
keys = res.body[/\{\"retcode[^\0]*\}/]
returnfalseif keys == nil
keys = JSON.parse(keys)
returnfalseif keys == nil
returnfalseif keys['retcode'] != 0

                #pre login
		preloginurl = 'http://login.sina.com.cn/sso/prelogin.php?entry=sso&' +
			'callback=sinaSSOController.preloginCallBack&su=' + 
			'dW5kZWZpbmVk' + '&rsakt=mod&client=ssologin.js(v1.4.2)' + 
				 "&_=" + Time.now.to_i.to_s
		#puts preloginurl
		res = HttpGet(preloginurl)
		keys = res.body[/\{\"retcode[^\0]*\}/]
		return false if keys == nil
		keys = JSON.parse(keys)
		return false if keys == nil
		return false if keys['retcode'] != 0

2：login

 

login 过程中使用 Http Post,Post 的数据主要有：

 

logindata={'entry'=>'weibo', 'gateway'=>'1',
'from'=>'', 'savestate'=>'7', 'userticket'=>'1',
'ssosimplelogin'=>'1', 'vsnf'=>'1', 'su'=>'',
'service'=>'miniblog', 'servertime'=>'', 'nonce'=>'',
'pwencode'=>'rsa2', 'rsakv'=>'', 'sp'=>'',
'encoding'=>'UTF-8', 'prelt'=>'115',
'returntype'=>'META',
'url'=>'http://weibo.com/ajaxlogin.php?framelogin=1&callback=parent.sinaSSOController.feedBackUrlCallBack'
}

               logindata={'entry'=>'weibo', 'gateway'=>'1',
			'from'=>'', 'savestate'=>'7', 'userticket'=>'1',
			'ssosimplelogin'=>'1', 'vsnf'=>'1', 'su'=>'', 
			'service'=>'miniblog', 'servertime'=>'', 'nonce'=>'',
			'pwencode'=>'rsa2', 'rsakv'=>'', 'sp'=>'',
			'encoding'=>'UTF-8', 'prelt'=>'115',
			'returntype'=>'META',
			'url'=>'http://weibo.com/ajaxlogin.php?framelogin=1&callback=parent.sinaSSOController.feedBackUrlCallBack'
		}

 

2.1 先使用第一步得到的数据 keys 填充 logindata

 

logindata['servertime'] = keys['servertime']
logindata['nonce'] = keys['nonce']
logindata['rsakv'] = keys['rsakv']
logindata['su'] = Base64.strict_encode64(uname.sub("@","%40"))

	logindata['servertime'] = keys['servertime'] 
	logindata['nonce'] =  keys['nonce'] 
	logindata['rsakv'] =  keys['rsakv'] 
	logindata['su'] = Base64.strict_encode64(uname.sub("@","%40"))

 

logindata['su'] 中是经过 Base64 编码过的用户名

其中 Base64.strict_encode64 是 ruby1.9 以后出现的，使用 ruby1.8 中的 Base64.encode64 会多出来换行符，需要自己去掉

2.1 生成RSA 加密过的密码

 

pwdkey = keys['servertime'].to_s + "\t" + keys['nonce'].to_s + "\n" + pwd.to_s

pub = OpenSSL::PKey::RSA::new
pub.e = 65537
pub.n = OpenSSL::BN.new(keys['pubkey'],16)
logindata['sp'] = pub.public_encrypt(pwdkey).unpack('H*').first

        pwdkey = keys['servertime'].to_s + "\t" + keys['nonce'].to_s + "\n" + pwd.to_s

	pub = OpenSSL::PKey::RSA::new
	pub.e = 65537
	pub.n = OpenSSL::BN.new(keys['pubkey'],16)
	logindata['sp'] = pub.public_encrypt(pwdkey).unpack('H*').first

2.2 Post 数据，并检查是否登录成功

 

 

res = HttpPost(uri,logindata,hdrs)
#puts 'save cookie'
redrecturi = res.body[/http:\/\/weibo.com\/ajax[^'"]*/]
retcode = redrecturi.match(/retcode=([\d]+)/)
retcode = retcode[1] if retcode != nil
returnfalseif redrecturi == nil
retcode = retcode.to_s
if retcode.to_s != '0'
reason = redrecturi.match(/reason=([^&]+)/)
reason = reason[1] if reason != nil
reason = URI.unescape(reason)
reason = @iconvFromGB2312.iconv(reason)
puts reason

                res = HttpPost(uri,logindata,hdrs)
		#puts 'save cookie'
		redrecturi = res.body[/http:\/\/weibo.com\/ajax[^'"]*/]
		retcode = redrecturi.match(/retcode=([\d]+)/)
		retcode = retcode[1] if retcode != nil
		return false if redrecturi == nil
		retcode = retcode.to_s
		if retcode.to_s != '0'
			reason = redrecturi.match(/reason=([^&]+)/)
			reason = reason[1] if reason != nil
			reason = URI.unescape(reason)
			reason = @iconvFromGB2312.iconv(reason)
			puts reason

end

		end

<SPAN style="WHITE-SPACE: pre"> </SPAN>return retcode if retcode != '0'
<SPAN style="WHITE-SPACE: pre"> </SPAN>returnfalseif !@cookies.GetCookie('SUP','sina.com.cn')

return retcode if retcode != '0'
return false if !@cookies.GetCookie('SUP','sina.com.cn')

[ruby] view plaincopyprint?

其中 @iconvFromGB2312 只字符转码，Ruby1.9.3 以后的版本可能要移除，转移到String 类下面了，定义如下

 

 

@iconvFromGB2312=Iconv.new('UTF-8','GB2312')

@iconvFromGB2312=Iconv.new('UTF-8','GB2312') 

@cookies 是一个管理 Http 中 cookie 的自定义类， 在 HttpPost 和 HttpGet 中自动生成、记录了http 传输过程中需要的cookie

 

3： 完成登录

上步登录成功以后，会在返回的 body 中包含要转到的链接，这一步骤跟随该链接，获取用户基本的信息，并最终访问 http://weibo.com/login.php 完成登录

 

<SPAN style="WHITE-SPACE: pre"> </SPAN>res = HttpGet(redrecturi)
ujson = res.body[/\{\"result\"\:[^)]+\}/]
returnfalseif ujson == nil
rval = JSON.parse(ujson);
returnfalseif rval.has_key?('userinfo') == false
returnfalseif rval.has_key?('result') == false
returnfalseif rval['result'] == false
@userInfo = rval['userinfo']
HttpGet('http://weibo.com/login.php');
returntrue

res = HttpGet(redrecturi)
		ujson = res.body[/\{\"result\"\:[^)]+\}/]
		return false if ujson == nil
		rval = JSON.parse(ujson);
		return false if rval.has_key?('userinfo') == false
		return false if rval.has_key?('result') == false
		return false if rval['result'] == false
		@userInfo = rval['userinfo']
		HttpGet('http://weibo.com/login.php');
		return true

 

Ruby 还用不熟悉，只是也是网上东拼西凑的，不系统，所以用着Ruby 语言，却不是使用 Ruby Way 实现的，有点惭愧，各位觉着有用也就大概看下微博登录的方法吧，编程风格是完全不可取的！再此申明，免担误导之责 ，不过还是真心希望 Ruby 大牛，能指正一下！