소켓 API 소개

Welcome to socket programming! A socket is a crucial concept in network communication. Think of it as one endpoint of a two-way communication link between two programs running on the network.

It's like a phone jack on a wall. You plug a phone into it to connect to the phone network. Similarly, programs use sockets to connect and talk over a computer network.

Python makes socket programming accessible through its built-in socket module. This module provides all the necessary functions and classes to create and manage sockets.

You'll typically import it at the beginning of your script. Let's see how to get started:

Before creating a socket, we need to specify its address family. This determines the type of addresses the socket can work with.

• socket.AF_INET: Used for IPv4 addresses (e.g., 192.168.1.1). This is the most common.
• socket.AF_INET6: Used for IPv6 addresses.
• socket.AF_UNIX: Used for communication between processes on the same machine.

We'll primarily use AF_INET for standard internet communication.

Next, we choose the socket type, which defines the communication style.

• socket.SOCK_STREAM: This type uses TCP (Transmission Control Protocol). It's reliable, ordered, and connection-oriented, like a phone call. Data arrives exactly as sent.
• socket.SOCK_DGRAM: This type uses UDP (User Datagram Protocol). It's connectionless and less reliable, like sending postcards. Data might arrive out of order or not at all.

Most common applications, especially web browsing, use SOCK_STREAM.

Now let's put it together and create a basic socket object. We'll specify the address family (IPv4) and the socket type (TCP stream).

This line creates an inactive socket that's ready to be configured for either client or server roles.

For a server, the socket needs an address to listen on. This is called binding. You assign an IP address and a port number to your socket.

Clients will use this address and port to connect to your server. A port is a number that identifies a specific application or service on a host.

After binding, a server socket must listen for incoming client connections. The listen() method puts the socket into server mode.

The argument to listen() is the backlog, which is the maximum number of unaccepted connections that the system will allow before refusing new connections.

When a client tries to connect, the server uses the accept() method. This method blocks (waits) until a client connects.

It then returns a new socket object (conn) representing the connection to the client, and the client's address (addr). All communication with the client happens via this new socket.

A client socket doesn't bind; instead, it uses the connect() method to establish a connection with a server that is listening on a specific IP address and port.

Once connected, the client can immediately start sending and receiving data. Data is always exchanged as bytes.

Once a connection is established, data can be exchanged using sendall() and recv().

• socket.sendall(data): Sends all of the byte-like data to the connected socket. It keeps sending until all data is sent or an error occurs.
• socket.recv(bufsize): Receives data from the socket. bufsize is the maximum number of bytes to be received at once. It returns an empty bytes object (b'') when the connection is closed.

Which of the following statements about Python's socket module and socket types is TRUE?

Great job! In this lesson, you've taken your first steps into socket programming with Python. You learned:

• What a network socket is and its role in communication.
• About Python's socket module.
• The difference between address families (AF_INET, AF_INET6) and socket types (SOCK_STREAM for TCP, SOCK_DGRAM for UDP).
• The basic API calls: socket() for creation, bind() and listen() for servers, and connect() for clients.
• How accept() handles incoming connections and how sendall()/recv() work.

Next, we'll dive deeper into building full TCP client-server applications!