【转】 python socket向百度发送http长连接请求 并做搜索

http://hi.baidu.com/leejun_2005/blog/item/30fe9bd23a396c28960a1640.html

【转】 python socket向百度发送http长连接请求 并做搜索

2011-10-08 22:13

转载自 冷子健

最终编辑 leejun_2005

 

#coding=gbk
'''
socket 给百度发送http请求

连接成功后，发送http的get请求，所搜索功能

'''
import socket
import sys
import time
if __name__=='__main__':
    #创建套接字
    try :
        sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
    except socket.eorror,e:
        print 'socket false:%s'%e
    print 'socket ...'
   
    #连接百度ip
    try :
        sock.connect(('220.181.111.148',80))
    except socket.error,e:
        print 'connect false %s'%e
        sock.close()
    print 'connect ...'
   
    #发送百度首页面请求并且保持连接
    try :
        print 'send start...'
        str='GET / HTTP/1.1\r\nHost:www.baidu.com\r\nConnection:keep-alive\r\n\r\n'
        sock.send(str)
    except socket.eorror,e:
        print 'send false'
        sock.close()
   
    data=''
    data = sock.recv(1024)
    while (1):       '''如何判断数据接收完毕，在发送http 最前端,
                        包含发送数据文件大小属性Content-Length，
                        用字符匹配方式取得文件大小,
                        同过大小判断是否接收完毕。
                        '''
        print data
        beg = data.find('Content-Length:',0,len(data))
        end = data.find('Content-Type:',0,len(data))
        print beg
        print end
        if(beg == end):
            print 'connecting closed'
            break
        num = long(data[beg+16:end-2])
        print num
        nums = 0
        while (1):
            data=sock.recv(1024)
            print data
            nums +=len(data)
            if(nums >= num):
                break
        word = raw_input('please input your word----->')
        str='''GET /s?wd=''' + word + ''' HTTP/1.1
Host:www.baidu.com
Connection: Keep-Alive

'''
        print str
        sock.send(str)
        data = ''
        data = sock.recv(1024)   
    sock.close()
    print data

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
用python搜索google
from：http://ipython.iteye.com/blog/336304

如果直接用urllib.urlopen(url)会返回一个forbidden的，其实加一个add_header就可以了.
给出一个关键字
word 就会返回搜索结果了。

    import urllib  
    import urllib2  
    def search(word):  
     url=r'http://www.google.cn/search?hl=zh-CN&newwindow=1&q='+word+"&start=10&sa=N"  
     req=urllib2.Request(url)  
     req.add_header("User-Agent",'Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.2; .NET CLR 1.1.4322)')  
     opener=urllib2.build_opener()  
     text=opener.open(req).read()  
     return text 

python通过socket查询whois脚本

from:http://www.oschina.net/code/snippet_54100_2870

import socket
s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
s.connect(("whois.networksolutions.com", 43))
s.send("sina.com.cn \r\n")
while 1:
        v = s.recv(1024)
        if v == "" or v == None:
                break
        print v
s.close()

 

基于Apache Mina实现的TCP长连接和短连接实例

http://blog.csdn.net/peterwanghao/article/details/6900523