windows环境下的Python HTTP请求模块

windows环境下的Python HTTP请求模块

2013年8月18日  2条评论  Article

网络上关于Python的HTTP请求大部分都是利用了urllib，httplib，pycurl，但是经过我试用之后，都没有对cookie以及ssl良好的支持，所以本模块主要使用了wininet API来发送HTTP请求，经过测试之后，wininet库很好的封装了ssl，cookie的支持，因此将wininet库在Python中的调用方式公开。

因为wininet是windows对Internet的封装，所以只支持windows系统。

在Linux下，我们可以选择curl库来执行http请求。 PS : chrome的网络请求就是使用了curl。

Python中对curl的封装有pycurl，但是pycurl的作者封装的curl是不带ssl版本，所以无法发起https请求，在这里就不提了，有兴趣的朋友可以访问pycurl的下载地址来试用。

上代码：

#coding:utf-8
import ctypes
from urlparse  import urlparse
import threading

INTERNET_OPEN_TYPE_PRECONFIG  = 0
INTERNET_OPEN_TYPE_PROXY  = 3
INTERNET_OPEN_TYPE_DIRECT  = 1
INTERNET_OPEN_TYPE_PRECONFIG_WITH_NO_AUTOPROXY  = 4

INTERNET_SERVICE_HTTP  = 3
INTERNET_FLAG_RELOAD  = 2147483648
HTTP_ADDREQ_FLAG_REPLACE  = 2147483648
HTTP_ADDREQ_FLAG_ADD = 536870912
INTERNET_OPEN_TYPE_DIRECT  = 1
INTERNET_FLAG_SECURE  = 8388608
HTTP_QUERY_STATUS_CODE  = 19

wininet  = ctypes.windll.LoadLibrary( 'wininet.dll' )

class wininetHttp():

    def httpRequest( self ,url,method = "GET" ,proxy = " ",username=" ",password=" ",headers=" ",data=" ",timeout = 20 ):
        """
        发出一个HTTP请求
        参数：
        url: 完整网址
        method: 请求方法，可以是GET或POST
        proxy: 代理地址
        username: 用户名
        password: 密码
        headers: 协议头
        data: 提交数据
        timeout: 请求超时值
        返回值：
        元组(header,html)
        """
        requestThread  = _httprequest(url,method,proxy,username,password,headers,data)
        requestThread.daemon  = True #线程随主线程一起退出
        requestThread.start()
        requestThread.join(timeout)
        ret  = requestThread.getResponse()
        return ret

class _httprequest(threading.Thread):

    response  = ( 0 ,'')

    def __init__( self ,url,method,proxy,username,password,headers,data):
        threading.Thread.__init__( self )
        self .url  = url
        self .url_info  = urlparse(url)
        self .method  = method
        self .proxy  = proxy
        self .username  = username
        self .password  = password
        self .headers  = headers
        self .data  = data

    def run( self ):
        dwAccessType  = self .proxy  and INTERNET_OPEN_TYPE_PROXY  or INTERNET_OPEN_TYPE_DIRECT
        lpszProxyName  = self .proxy  and 'http=' + self .proxy  or None
        lpszAgent  = 'Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/28.0.1500.72 Safari/537.36'
        hOpen  = wininet.InternetOpenA(lpszAgent,dwAccessType,lpszProxyName, None , 0 )
        if not hOpen:
            return
        nServerPort  = ( self .url_info.scheme  = = 'https' )  and 443 or 80
        hConn  = wininet.InternetConnectA(hOpen, self .url_info.netloc.lower(),nServerPort, self .username, self .password,INTERNET_SERVICE_HTTP, 0 , 0 )
        if not hConn:
            wininet.InternetCloseHandle(hOpen)
            return
        dwFlags  = ( self .url_info.scheme  = = 'https' )  and INTERNET_FLAG_SECURE  or INTERNET_FLAG_RELOAD
        hRequest  = wininet.HttpOpenRequestA(hConn, self .method, self .url_info.path, 'HTTP/1.1' , None , 0 ,dwFlags, 0 )
        if not hRequest:
            wininet.InternetCloseHandle(hConn)
            wininet.InternetCloseHandle(hOpen)
            return
        if not 'Accept: ' in self .headers:
            self .headers  + = 'Accept: text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8\r\n'
        if not 'Referer: ' in self .headers:
            self .headers  + = 'Referer: %s\r\n' % self .url
        if not 'Accept-Language: ' in self .headers:
            self .headers  + = 'Accept-Language: zh-CN,zh;q=0.8\r\n'
        if self .method.upper()  = = 'GET' :
            wininet.HttpSendRequestA(hRequest,  self .headers, len ( self .headers), None ,  0 )
        elif self .method.upper()  = = 'POST' :
            if self .data ! = '':
                if not 'Content-Type: ' in self .headers:
                    self .headers  = self .headers  + "Content-Type: application/x-www-form-urlencoded\r\n"
                self .headers  = self .headers  + "Content-Length: %s\r\n" % len ( self .data)
            wininet.HttpSendRequestA(hRequest,  self .headers, len ( self .headers), self .data, len ( self .data))
        else :
            wininet.InternetCloseHandle(hRequest)
            wininet.InternetCloseHandle(hConn)
            wininet.InternetCloseHandle(hOpen)
            return
        data_buf  = (ctypes.c_char  * 2048 )()
        ret_data  = ''
        x  = ctypes.c_int( 1 )
        while x.value ! = 0 :
            wininet.InternetReadFile(hRequest,data_buf, 2048 ,ctypes.byref(x))
            buf  = (ctypes.c_char  * x.value)()
            ctypes.memmove(buf,data_buf,x.value)
            ret_data  + = buf[:x.value]

        head_buf  = ctypes.create_string_buffer( 2048 )
        buf_size  = ctypes.c_int( 2048 )
        wininet.HttpQueryInfoA(hRequest, 22 ,head_buf,ctypes.byref(buf_size), 0 )

        wininet.InternetCloseHandle(hRequest)
        wininet.InternetCloseHandle(hConn)
        wininet.InternetCloseHandle(hOpen)

        self .response  = (head_buf.value,ret_data)
        return

    def getResponse( self ):
        return self .response

因为wininet无法设置超时，所以模块中的超时是利用Python threading模块来控制。

学Python才几天，如果代码中有写的不对的地方或者是有更好的解决方案，欢迎联系我一起讨论~

代码自由使用，转载需注明原出处

其实Python的学习，只要多看官方doc就可以了。(●´∀｀●)