【学习笔记:Python-网络编程】Socket 之初见

Socket 是任何一种计算机网络通讯中最基础的内容。当你在浏览器地址栏中输入一个地址时,你会打开一个套接字,可以说任何网络通讯都是通过 Socket 来完成的。

Socket 的 python 官方函数 http://docs.python.org/library/socket.html

socket和file的区别:

  1、file模块是针对某个指定文件进行【打开】【读写】【关闭】

  2、socket模块是针对 服务器端 和 客户端Socket 进行【打开】【读写】【关闭】

基本流程:

简单的一个端对端单线通信代码如下:

Server:

 1 # -*- coding:utf-8
 2 import socket
 3 sock = socket.socket(socket.AF_INET,socket.SOCK_STREAM)
 4 sock.bind(("0.0.0.0",12800))
 5 sock.listen(2)
 6 while True:
 7     conn,sockname = sock.accept()
 8     print("Now,We have accepted a connection from:",sockname)
 9     print("Socket Name is:",conn.getsockname())
10     print("Socket Peer is:",conn.getpeername())
11     message = conn.recv(1024)
12     print("The message we have received is:\n%s"%message)

Client:

 
# -*- coding:utf-8
import socket
sock = socket.socket(socket.AF_INET,socket.SOCK_STREAM)
sock.connect(("127.0.0.1",12800))
sock.send(b"Hello World!")
 

参数解释:

 socket.socket(family=AF_INET, type=SOCK_STREAM, proto=0, fileno=None): 创建一个套接字对象,所有套接字操作基于此对象进行
  family:套接字协议族,包括:有两种类型的套接字:基于文件的和面向网络的。AF表示地址家族(address family)
    AF_UNIX该套接字是基于文件的,一般用于本机通信
    
AF_INET:该套接字是基于网络的,这也是最常用的(默认)   
    AF_INET6 用于第6 版因特网协议(IPv6)寻址
  type:套接字类型,常用的有两种:
    SOCK_STREAM:流式socket ,
for TCP (默认)   
    SOCK_DGRAM:数据报式socket , for UDP
    SOCK_RAW 原始套接字,普通的套接字无法处理ICMP、IGMP等网络报文,而SOCK_RAW可以;其次,SOCK_RAW也可以处理特殊的IPv4报文;此外,利用原始套接字,可以通过IP_HDRINCL套接字选项由用户构造IP头。
    SOCK_RDM 是一种可靠的UDP形式,即保证交付数据报但不保证顺序。SOCK_RAM用来提供对原始协议的低级访问,在需要执行某些特殊操作时使用,如发送ICMP报文。SOCK_RAM通常仅限于高级用户或管理员运行的程序使用。
    SOCK_SEQPACKET 可靠的连续数据包服务

  proto:协议,与特定的地址家族相关的协议,如果是 0 ,则系统就会根据地址格式和套接类别,自动选择一个合适的协议


sock.bind(address)

  sock.bind(address) 将套接字绑定到地址。address地址的格式取决于地址族。在AF_INET下,以元组(host,port)的形式表示地址。

 

sock.listen(backlog)-- 服务端

  开始监听传入连接。backlog指定在拒绝连接之前,可以挂起的最大连接数量。

    backlog等于5,表示内核已经接到了连接请求,但服务器还没有调用accept进行处理的连接个数最大为5
    这个值不能无限大,因为要在内核中维护连接队列,一般默认值为5

 

 

sock.settimeout(timeout)

 

  设置套接字操作的超时期,timeout是一个浮点数,单位是秒。值为None表示没有超时期。一般,超时期应该在刚创建套接字时设置,因为它们可能用于连接的操作(如 client 连接最多等待5s )

 

sock.getpeername()

 

  返回连接套接字的远程地址。返回值通常是元组(ipaddr,port)。

 

sock.getsockname()

 

  返回套接字自己的地址。通常是一个元组(ipaddr,port)

 

sock.fileno()

 

  套接字的文件描述符

 

 

sock.setblocking(bool)

  是否阻塞(默认True),如果设置False,那么accept和recv时一旦无数据,则报错。

sock.accept()-- 服务端

  接受连接并返回(conn,address),其中conn是新的套接字对象,可以用来接收和发送数据。address是连接客户端的地址。

  接收TCP 客户的连接(阻塞式)等待连接的到来

sock.connect(address)-- 客户端

  连接到address处的套接字。一般,address的格式为元组(hostname,port),如果连接出错,返回socket.error错误。

sock.connect_ex(address)

  同上,只不过会有返回值,连接成功时返回 0 ,连接失败时候返回编码,例如:10061

sock.close()

  关闭套接字

 

 

sock.recv(bufsize[,flag])

  接受套接字的数据。数据以字符串形式返回,bufsize指定最多可以接收的数量。flag提供有关消息的其他信息,通常可以忽略。

sock.recvfrom(bufsize[.flag])

  与recv()类似,但返回值是(data,address)。其中data是包含接收数据的字符串,address是发送数据的套接字地址。

sock.send(string[,flag])

  将string中的数据发送到连接的套接字。返回值是要发送的字节数量,该数量可能小于string的字节大小。即:可能未将指定内容全部发送。

sock.sendall(string[,flag])

  将string中的数据发送到连接的套接字,但在返回之前会尝试发送所有数据。成功返回None,失败则抛出异常。

      内部通过递归调用send,将所有内容发送出去。

sock.sendto(string[,flag],address)

  将数据发送到套接字,address是形式为(ipaddr,port)的元组,指定远程地址。返回值是发送的字节数。该函数主要用于UDP协议。

SocketType.__doc__

  1 class SocketType(__builtin__.object)
  2  |  socket([family[, type[, proto]]]) -> socket object
  3  |  
  4  |  Open a socket of the given type.  The family argument specifies the
  5  |  address family; it defaults to AF_INET.  The type argument specifies
  6  |  whether this is a stream (SOCK_STREAM, this is the default)
  7  |  or datagram (SOCK_DGRAM) socket.  The protocol argument defaults to 0,
  8  |  specifying the default protocol.  Keyword arguments are accepted.
  9  |  
 10  |  A socket object represents one endpoint of a network connection.
 11  |  
 12  |  Methods of socket objects (keyword arguments not allowed):
 13  |  
 14  |  accept() -- accept a connection, returning new socket and client address
 15  |  bind(addr) -- bind the socket to a local address
 16  |  close() -- close the socket
 17  |  connect(addr) -- connect the socket to a remote address
 18  |  connect_ex(addr) -- connect, return an error code instead of an exception
 19  |  dup() -- return a new socket object identical to the current one [*]
 20  |  fileno() -- return underlying file descriptor
 21  |  getpeername() -- return remote address [*]
 22  |  getsockname() -- return local address
 23  |  getsockopt(level, optname[, buflen]) -- get socket options
 24  |  gettimeout() -- return timeout or None
 25  |  listen(n) -- start listening for incoming connections
 26  |  makefile([mode, [bufsize]]) -- return a file object for the socket [*]
 27  |  recv(buflen[, flags]) -- receive data
 28  |  recv_into(buffer[, nbytes[, flags]]) -- receive data (into a buffer)
 29  |  recvfrom(buflen[, flags]) -- receive data and sender's address
 30  |  recvfrom_into(buffer[, nbytes, [, flags])
 31  |    -- receive data and sender's address (into a buffer)
 32  |  sendall(data[, flags]) -- send all data
 33  |  send(data[, flags]) -- send data, may not send all of it
 34  |  sendto(data[, flags], addr) -- send data to a given address
 35  |  setblocking(0 | 1) -- set or clear the blocking I/O flag
 36  |  setsockopt(level, optname, value) -- set socket options
 37  |  settimeout(None | float) -- set or clear the timeout
 38  |  shutdown(how) -- shut down traffic in one or both directions
 39  |  
 40  |   [*] not available on all platforms!
 41  |  
 42  |  Methods defined here:
 43 
 44  |  
 45  |  accept(self)
 46  |      accept() -> (socket object, address info)
 47  |      
 48  |      Wait for an incoming connection.  Return a new socket representing the
 49  |      connection, and the address of the client.  For IP sockets, the address
 50  |      info is a pair (hostaddr, port).
 51  |  
 52  |  bind(self, address)
 53     
 54  |      bind(address)
 55  |      
 56  |      Bind the socket to a local address.  For IP sockets, the address is a
 57  |      pair (host, port); the host must refer to the local host. For raw packet
 58  |      sockets the address is a tuple (ifname, proto [,pkttype [,hatype]])
 59  |  
 60  |  close(self)
 61  |      close()
 62  |      
 63  |      Close the socket.  It cannot be used after this call.
 64  |  
 65  |  connect(self, address)
 66  |      connect(address)
 67  |      
 68  |      Connect the socket to a remote address.  For IP sockets, the address
 69  |      is a pair (host, port).
 70  |  
 71  |  connect_ex(self, address)
 72  |      connect_ex(address) -> errno
 73  |      
 74  |      This is like connect(address), but returns an error code (the errno value)
 75  |      instead of raising an exception when an error occurs.
 76  |  
 77  |  fileno(self)
 78  |      fileno() -> integer
 79  |      
 80  |      Return the integer file descriptor of the socket.
 81  |  
 82  |  getpeername(self)
 83  |      getpeername() -> address info
 84  |      
 85  |      Return the address of the remote endpoint.  For IP sockets, the address
 86  |      info is a pair (hostaddr, port).
 87  |  
 88  |  getsockname(self)
 89  |      getsockname() -> address info
 90  |      
 91  |      Return the address of the local endpoint.  For IP sockets, the address
 92  |      info is a pair (hostaddr, port).
 93  |  
 94  |  getsockopt(self, level, option, buffersize=None)
 95  |      getsockopt(level, option[, buffersize]) -> value
 96  |      
 97  |      Get a socket option.  See the Unix manual for level and option.
 98  |      If a nonzero buffersize argument is given, the return value is a
 99  |      string of that length; otherwise it is an integer.
100  |  
101  |  gettimeout(self)
102  |      gettimeout() -> timeout
103  |      
104  |      Returns the timeout in seconds (float) associated with socket 
105  |      operations. A timeout of None indicates that timeouts on socket 
106  |      operations are disabled.
107  |  
108  |  ioctl(self, cmd, option)
109  |      ioctl(cmd, option) -> long
110  |      
111  |      Control the socket with WSAIoctl syscall. Currently supported 'cmd' values are
112  |      SIO_RCVALL:  'option' must be one of the socket.RCVALL_* constants.
113  |      SIO_KEEPALIVE_VALS:  'option' is a tuple of (onoff, timeout, interval).
114  |  
115  |  listen(self, backlog)
116  |      listen(backlog)
117  |      
118  |      Enable a server to accept connections.  The backlog argument must be at
119  |      least 0 (if it is lower, it is set to 0); it specifies the number of
120  |      unaccepted connections that the system will allow before refusing new
121  |      connections.
122  |  
123  |  recv(self, buffersize, flags=None)
124  |      recv(buffersize[, flags]) -> data
125  |      
126  |      Receive up to buffersize bytes from the socket.  For the optional flags
127  |      argument, see the Unix manual.  When no data is available, block until
128  |      at least one byte is available or until the remote end is closed.  When
129  |      the remote end is closed and all data is read, return the empty string.
130  |  
131  |  recv_into(self, buffer, nbytes=None, flags=None)
132  |      recv_into(buffer, [nbytes[, flags]]) -> nbytes_read
133  |      
134  |      A version of recv() that stores its data into a buffer rather than creating 
135  |      a new string.  Receive up to buffersize bytes from the socket.  If buffersize 
136  |      is not specified (or 0), receive up to the size available in the given buffer.
137  |      
138  |      See recv() for documentation about the flags.
139  |  
140  |  recvfrom(self, buffersize, flags=None)
141  |      recvfrom(buffersize[, flags]) -> (data, address info)
142  |      
143  |      Like recv(buffersize, flags) but also return the sender's address info.
144  |  
145  |  recvfrom_into(self, buffer, nbytes=None, flags=None)
146  |      recvfrom_into(buffer[, nbytes[, flags]]) -> (nbytes, address info)
147  |      
148  |      Like recv_into(buffer[, nbytes[, flags]]) but also return the sender's address info.
149  |  
150  |  send(self, data, flags=None)
151  |      send(data[, flags]) -> count
152  |      
153  |      Send a data string to the socket.  For the optional flags
154  |      argument, see the Unix manual.  Return the number of bytes
155  |      sent; this may be less than len(data) if the network is busy.
156  |  
157  |  sendall(self, data, flags=None)
158  |      sendall(data[, flags])
159  |      
160  |      Send a data string to the socket.  For the optional flags
161  |      argument, see the Unix manual.  This calls send() repeatedly
162  |      until all data is sent.  If an error occurs, it's impossible
163  |      to tell how much data has been sent.
164  |  
165  |  sendto(self, data, flags=None, *args, **kwargs)
166  |      sendto(data[, flags], address) -> count
167  |      
168  |      Like send(data, flags) but allows specifying the destination address.
169  |      For IP sockets, the address is a pair (hostaddr, port).
170  |  
171  |  setblocking(self, flag)
172  |      setblocking(flag)
173  |      
174  |      Set the socket to blocking (flag is true) or non-blocking (false).
175  |      setblocking(True) is equivalent to settimeout(None);
176  |      setblocking(False) is equivalent to settimeout(0.0).
177  |  
178  |  setsockopt(self, level, option, value)
179  |      setsockopt(level, option, value)
180  |      
181  |      Set a socket option.  See the Unix manual for level and option.
182  |      The value argument can either be an integer or a string.
183  |  
184  |  settimeout(self, timeout)
185  |      settimeout(timeout)
186  |      
187  |      Set a timeout on socket operations.  'timeout' can be a float,
188  |      giving in seconds, or None.  Setting a timeout of None disables
189  |      the timeout feature and is equivalent to setblocking(1).
190  |      Setting a timeout of zero is the same as setblocking(0).
191  |  
192  |  shutdown(self, flag)
193  |      shutdown(flag)
194  |      
195  |      Shut down the reading side of the socket (flag == SHUT_RD), the writing side
196  |      of the socket (flag == SHUT_WR), or both ends (flag == SHUT_RDWR).
197  
View Code

猜你喜欢

转载自www.cnblogs.com/halfleaf/p/9584645.html