Http代理服务器

来源：喜迪情感

最近打算好好深入研究下python的socket编程，于是打算学习下，仿写了一下，发现写好还真不容易，中途出现很多问题，果真是看的容易，做起来难啊

import socket
import thread
import urlparse
import select
 
BUFLEN=8192
 
 
class Proxy(object):
 def __init__(self,conn,addr):
 self.source=conn
 self.request=""
 self.headers={}
 self.destnation=socket.socket(socket.AF_INET,socket.SOCK_STREAM)
 self.run()
 
 def get_headers(self):
 header=''
 while True:
 header+=self.source.recv(BUFLEN)
 index=header.find('
')
 if index >0:
 break
 #firstLine,self.request=header.split('
',1) 
 firstLine=header[:index]
 self.request=header[index+1:]
 self.headers['method'],self.headers['path'],self.headers['protocol']=firstLine.split()
 
 def conn_destnation(self):
 url=urlparse.urlparse(self.headers['path'])
 hostname=url[1]
 port="80"
 if hostname.find(':') >0:
 addr,port=hostname.split(':')
 else:
 addr=hostname
 port=int(port)
 ip=socket.gethostbyname(addr)
 print ip,port
 self.destnation.connect((ip,port))
 data="%s %s %s
" %(self.headers['method'],self.headers['path'],self.headers['protocol'])
 self.destnation.send(data+self.request)
 print data+self.request
 
 
 def renderto(self):
 readsocket=[self.destnation]
 while True:
 data=''
 (rlist,wlist,elist)=select.select(readsocket,[],[],3)
 if rlist:
 data=rlist[0].recv(BUFLEN)
 if len(data)>0:
 self.source.send(data)
 else:
 break
 def run(self):
 self.get_headers()
 self.conn_destnation()
 self.renderto()
 
 
 
class Server(object):
 
 def __init__(self,host,port,handler=Proxy):
 self.host=host
 self.port=port
 self.server=socket.socket(socket.AF_INET,socket.SOCK_STREAM)
 self.server.setsockopt(socket.SOL_SOCKET, socket.SO_REUSEADDR, 1)
 self.server.bind((host,port))
 self.server.listen(5)
 self.handler=handler
 
 def start(self):
 while True:
 try:
 conn,addr=self.server.accept()
 thread.start_new_thread(self.handler,(conn,addr))
 except:
 pass
 
 
if __name__=='__main__':
 s=Server('127.0.0.1',8080)
 s.start()
import socket
import thread
import urlparse
import select
BUFLEN=8192
class Proxy(object):
 def __init__(self,conn,addr):
 self.source=conn
 self.request=""
 self.headers={}
 self.destnation=socket.socket(socket.AF_INET,socket.SOCK_STREAM)
 self.run()
 def get_headers(self):
 header=''
 while True:
 header+=self.source.recv(BUFLEN)
 index=header.find('
')
 if index >0:
 break
 #firstLine,self.request=header.split('
',1)
 firstLine=header[:index]
 self.request=header[index+1:]
 self.headers['method'],self.headers['path'],self.headers['protocol']=firstLine.split()
 def conn_destnation(self):
 url=urlparse.urlparse(self.headers['path'])
 hostname=url[1]
 port="80"
 if hostname.find(':') >0:
 addr,port=hostname.split(':')
 else:
 addr=hostname
 port=int(port)
 ip=socket.gethostbyname(addr)
 print ip,port
 self.destnation.connect((ip,port))
 data="%s %s %s
" %(self.headers['method'],self.headers['path'],self.headers['protocol'])
 self.destnation.send(data+self.request)
 print data+self.request
 def renderto(self):
 readsocket=[self.destnation]
 while True:
 data=''
 (rlist,wlist,elist)=select.select(readsocket,[],[],3)
 if rlist:
 data=rlist[0].recv(BUFLEN)
 if len(data)>0:
 self.source.send(data)
 else:
 break
 def run(self):
 self.get_headers()
 self.conn_destnation()
 self.renderto()
 
class Server(object):
 def __init__(self,host,port,handler=Proxy):
 self.host=host
 self.port=port
 self.server=socket.socket(socket.AF_INET,socket.SOCK_STREAM)
 self.server.setsockopt(socket.SOL_SOCKET, socket.SO_REUSEADDR, 1)
 self.server.bind((host,port))
 self.server.listen(5)
 self.handler=handler
 def start(self):
 while True:
 try:
 conn,addr=self.server.accept()
 thread.start_new_thread(self.handler,(conn,addr))
 except:
 pass
if __name__=='__main__':
 s=Server('127.0.0.1',8080)
 s.start()

其实Http代理服务器本身不难，但写出来还是挺费事的，这里就不细说源代码了，很简单。主要说说，我遇到的问题。

一：我本来只知道，thread.start_new_thread的第一个参数是函数对象，但当我看到上面的博文时，心里一愣，这样也可以，于是我迅速的测试了一下：

import thread
 
class Hello:
 def __init__(self,content):
 print content
 
def cs():
 thread.start_new_thread(Hello, ("Hello World",))
 
if __name__=='__main__':
 cs()
import thread
class Hello:
 def __init__(self,content):
 print content
def cs():
 thread.start_new_thread(Hello, ("Hello World",))
if __name__=='__main__':
 cs()

Unhandled exception in thread started by
Error in sys.excepthook:

Original exception was:

Unhandled exception in thread started by
Error in sys.excepthook:

Original exception was:

一看，我说嘛，第一个参数怎么可以是对象，我呵呵一笑，稍微鄙视了一下作者。于是，我洗洗睡了，第二天，我还是不死心，于是把代码下下来，本地实验了一下，可以的，立马意识到是我2了，于是立马百度。

原来thread模块中，主线程如果比子线程先结束，就会抛出这个异常，所以我们必须让子线程先结束，最简单的方法就是让主线程sleep足够长的时间，至于多长时间，貌似不知道，那到底怎么解决呢？

比较好的解决办法就是，主线程给每个子线程都加一把锁，子线程在结束前将锁释放掉，主线程一直循环检查锁的状态。代码如下：

import thread
 
class Hello:
 def __init__(self,content,lock):
 print content
 """
 do something
 ....
 At the end,release the lock
 """
 lock.release()
 
def cs():
 lock=thread.allocate_lock()
 lock.acquire()
 thread.start_new_thread(Hello, ("Hello World",lock))
 while True:
 if not lock.locked():
 break
 print "lock release"
 
if __name__=='__main__':
 cs()
import thread
class Hello:
 def __init__(self,content,lock):
 print content
 """
 do something
 ....
 At the end,release the lock
 """
 lock.release()
def cs():
 lock=thread.allocate_lock()
 lock.acquire()
 thread.start_new_thread(Hello, ("Hello World",lock))
 while True:
 if not lock.locked():
 break
 print "lock release"
if __name__=='__main__':
 cs()

二.第二个错误就是比较2的了

self.source.send[data]

peError: 'builtin_function_or_method' object is unsubscriptable

self.source.send[data]

TypeError: 'builtin_function_or_method' object is unsubscriptable

主要意思就是说，内置函数或方法无法拥有下标，你懂的