After the interface of several clients has been processed, I have encountered many pits and have time to summarize.
Now let's start working on the server side, let's learn first
Source: http://acm.tzc.edu.cn/acmhome/projectList.do?method=projectNewsDetail&nid=2
Author: carter
Read your article, transfer it to save it, and if it offends, please contact me to delete it.
1. Introduction to Socket
Socket is a way of process communication, that is, calling some API functions of this network library to realize data exchange between related processes distributed on different hosts.
A few definitions:
(1) IP address: that is, the network address assigned to the local host according to the TCP/IP protocol. Two processes need to communicate. Either process must first know the location of the other party, that is, the IP of the other party.
(2) Port number: It is used to identify the local communication process. A local process will occupy a port number when communicating. Different processes have different port numbers. Therefore, a port number that has not been accessed must be assigned before communication.
(3) Connection: refers to the communication link between two processes.
(4) Semi-correlation: A triplet in the network can uniquely mark a process globally:
(protocol, local address, local port number)
Such a triple, called a half-correlation, specifies each half of the connection.
(4) Full correlation: A complete inter-process communication needs to be composed of two processes, and only the same high-level protocol can be used. That is, one end that cannot communicate uses the TCP protocol, and the other end uses the UDP protocol. Therefore, a complete Internet communication requires a five-tuple to identify:
(protocol, local address, local port number, remote address, remote port number)
Such a quintuple is called an association, that is, two semi-associations with the same protocol can be combined into a suitable association, or fully specified to form a connection.
2. Client/Server Mode
In TCP/IP network applications, the main mode of interaction between the two processes of communication is the client/server (Client/Server, C/S) mode, that is, the client sends a service request to the server, and the server provides the corresponding service after receiving the request. service. The establishment of the client/server model is based on the following two points:
(1) First of all, the reason for establishing a network is that the hardware and software resources, computing power and information in the network are unequal and need to be shared, so that hosts with many resources provide services and clients with fewer resources request services.
(2) Secondly, the communication between the processes in the network is completely asynchronous. There is neither a parent-child relationship nor a shared memory buffer between the processes that communicate with each other. Therefore, a mechanism is needed to establish a connection between the processes that want to communicate and for the two. Data exchange provides synchronization, which is TCP/IP based on the client/server model.
Service-Terminal:
The process is to start the server first and provide corresponding services according to the request:
(1) Open a communication channel and inform the local host that it is willing to receive client requests on a port on a recognized address (for example, the port of FTP may be 21);
(2) Wait for the client request to arrive at the port;
(3) When receiving a service request from the client, process the request and send a response signal. When a concurrent service request is received, a new process must be activated to handle the client request (eg, fork, exec in UNIX systems). The new process handles this client request and does not need to respond to other requests. After the service is complete, the communication link between this new process and the client is closed and terminated.
(4) Return to step (2) and wait for another client request.
(5) Shut down the server
Client:
(1) Open a communication channel and connect to the specific port of the host where the server is located;
(2) Send a service request message to the server, wait for and receive a response; continue to make requests...
(3) After the request ends, the communication channel is closed and terminated.
From the process described above, it can be seen that:
(1) The roles of the client and server processes are asymmetric, so the codes are different.
(2) The server process is generally started first. As long as the system is running, the service process exists until it is terminated normally or forcibly.
After introducing the basic knowledge, here are some API functions:
Create a socket - socket()
Before an application uses a socket, it must first have a socket. The system call socket() provides a means for the application to create a socket. The calling format is as follows:
SOCKET PASCAL FAR socket(int af, int type, int protocol);
该调用要接收三个参数：af、type、protocol。参数af指定通信发生的区域：AF_UNIX、AF_INET、AF_NS等，而DOS、WINDOWS中仅支持AF_INET，它是网际网区域。因此，地址族与协议族相同。参数type 描述要建立的套接字的类型。这里分三种：
（1）一是TCP流式套接字(SOCK_STREAM)提供了一个面向连接、可靠的数据传输服务，数据无差错、无重复地发送，且按发送顺序接收。内设流量控制，避免数据流超限；数据被看作是字节流，无长度限制。文件传送协议（FTP）即使用流式套接字。
（2）二是数据报式套接字(SOCK_DGRAM)提供了一个无连接服务。数据包以独立包形式被发送，不提供无错保证，数据可能丢失或重复，并且接收顺序混乱。网络文件系统（NFS）使用数据报式套接字。
（3）三是原始式套接字(SOCK_RAW)该接口允许对较低层协议，如IP、ICMP直接访问。常用于检验新的协议实现或访问现有服务中配置的新设备。
参数protocol说明该套接字使用的特定协议，如果调用者不希望特别指定使用的协议，则置为0，使用默认的连接模式。根据这三个参数建立一个套接字，并将相应的资源分配给它，同时返回一个整型套接字号。因此，socket()系统调用实际上指定了相关五元组中的“协议”这一元。
指定本地地址──bind()
当一个套接字用socket()创建后，存在一个名字空间(地址族)，但它没有被命名。bind()将套接字地址（包括本地主机地址和本地端口地址）与所创建的套接字号联系起来，即将名字赋予套接字，以指定本地半相关。其调用格式如下：
int PASCAL FAR bind(SOCKET s, const struct sockaddr FAR * name, int namelen);
参数s是由socket()调用返回的并且未作连接的套接字描述符(套接字号)。参数name 是赋给套接字s的本地地址（名字），其长度可变，结构随通信域的不同而不同。namelen表明了name的长度。如果没有错误发生，bind()返回0。否则返回SOCKET_ERROR。
建立套接字连接──connect()与accept()
这两个系统调用用于完成一个完整相关的建立，其中connect()用于建立连接。accept()用于使服务器等待来自某客户进程的实际连接。
connect()的调用格式如下：
int PASCAL FAR connect(SOCKET s, const struct sockaddr FAR * name, int namelen);
参数s是欲建立连接的本地套接字描述符。参数name指出说明对方套接字地址结构的指针。对方套接字地址长度由namelen说明。
如果没有错误发生，connect()返回0。否则返回值SOCKET_ERROR。在面向连接的协议中，该调用导致本地系统和外部系统之间连接实际建立。
由于地址族总被包含在套接字地址结构的前两个字节中，并通过socket()调用与某个协议族相关。因此bind()和connect()无须协议作为参数。
accept()的调用格式如下：
SOCKET PASCAL FAR accept(SOCKET s, struct sockaddr FAR* addr, int FAR* addrlen);
参数s为本地套接字描述符，在用做accept()调用的参数前应该先调用过listen()。addr 指向客户方套接字地址结构的指针，用来接收连接实体的地址。addr的确切格式由套接字创建时建立的地址族决定。addrlen 为客户方套接字地址的长度（字节数）。如果没有错误发生，accept()返回一个SOCKET类型的值，表示接收到的套接字的描述符。否则返回值INVALID_SOCKET。
accept()用于面向连接服务器。参数addr和addrlen存放客户方的地址信息。调用前，参数addr 指向一个初始值为空的地址结构，而addrlen 的初始值为0；调用accept()后，服务器等待从编号为s的套接字上接受客户连接请求，而连接请求是由客户方的connect()调用发出的。当有连接请求到达时，accept()调用将请求连接队列上的第一个客户方套接字地址及长度放入addr 和addrlen，并创建一个与s有相同特性的新套接字号。新的套接字可用于处理服务器并发请求。
四个套接字系统调用，socket()、bind()、connect()、accept()，可以完成一个完全五元相关的建立。socket()指定五元组中的协议元，它的用法与是否为客户或服务器、是否面向连接无关。bind()指定五元组中的本地二元，即本地主机地址和端口号，其用法与是否面向连接有关：在服务器方，无论是否面向连接，均要调用bind()，若采用面向连接，则可以不调用bind()，而通过connect()自动完成。若采用无连接，客户方必须使用bind()以获得一个唯一的地址。
监听连接──listen()
此调用用于面向连接服务器，表明它愿意接收连接。listen()需在accept()之前调用，其调用格式如下：
int PASCAL FAR listen(SOCKET s, int backlog);
参数s标识一个本地已建立、尚未连接的套接字号，服务器愿意从它上面接收请求。backlog表示请求连接队列的最大长度，用于限制排队请求的个数，目前允许的最大值为5。如果没有错误发生，listen()返回0。否则它返回SOCKET_ERROR。
listen()在执行调用过程中可为没有调用过bind()的套接字s完成所必须的连接，并建立长度为backlog的请求连接队列。
调用listen()是服务器接收一个连接请求的四个步骤中的第三步。它在调用socket()分配一个流套接字，且调用bind()给s赋于一个名字之后调用，而且一定要在accept()之前调用。
数据传输──send()与recv()
当一个连接建立以后，就可以传输数据了。常用的系统调用有send()和recv()。
send()调用用于s指定的已连接的数据报或流套接字上发送输出数据，格式如下：
int PASCAL FAR send(SOCKET s, const char FAR *buf, int len, int flags);
参数s为已连接的本地套接字描述符。buf 指向存有发送数据的缓冲区的指针，其长度由len 指定。flags 指定传输控制方式，如是否发送带外数据等。如果没有错误发生，send()返回总共发送的字节数。否则它返回SOCKET_ERROR。
recv()调用用于s指定的已连接的数据报或流套接字上接收输入数据，格式如下：
int PASCAL FAR recv(SOCKET s, char FAR *buf, int len, int flags);
参数s 为已连接的套接字描述符。buf指向接收输入数据缓冲区的指针，其长度由len 指定。flags 指定传输控制方式，如是否接收带外数据等。如果没有错误发生，recv()返回总共接收的字节数。如果连接被关闭，返回0。否则它返回SOCKET_ERROR。
输入/输出多路复用──select()
select()调用用来检测一个或多个套接字的状态。对每一个套接字来说，这个调用可以请求读、写或错误状态方面的信息。请求给定状态的套接字集合由一个fd_set结构指示。在返回时，此结构被更新，以反映那些满足特定条件的套接字的子集，同时， select()调用返回满足条件的套接字的数目，其调用格式如下：
int PASCAL FAR select(int nfds, fd_set FAR * readfds, fd_set FAR * writefds, fd_set FAR * exceptfds, const struct timeval FAR * timeout);
参数nfds指明被检查的套接字描述符的值域，此变量一般被忽略。
参数readfds指向要做读检测的套接字描述符集合的指针，调用者希望从中读取数据。参数writefds 指向要做写检测的套接字描述符集合的指针。exceptfds指向要检测是否出错的套接字描述符集合的指针。timeout指向select()函数等待的最大时间，如果设为NULL则为阻塞操作。select()返回包含在fd_set结构中已准备好的套接字描述符的总数目，或者是发生错误则返回SOCKET_ERROR。
关闭套接字──closesocket()
closesocket()关闭套接字s，并释放分配给该套接字的资源；如果s涉及一个打开的TCP连接，则该连接被释放。closesocket()的调用格式如下：
BOOL PASCAL FAR closesocket(SOCKET s);
参数s待关闭的套接字描述符。如果没有错误发生，closesocket()返回0。否则返回值SOCKET_ERROR。
以上就是SOCKET API一些常用的API函数，下面是一段代码：
//客户端代码：
#include <WINSOCK2.H>
#include <stdio.h>
#pragma comment(lib,"ws2_32.lib")
int main()
{
int err;
WORD versionRequired;
WSADATA wsaData;
versionRequired=MAKEWORD(1,1);
err=WSAStartup(versionRequired,&wsaData);//协议库的版本信息
if (!err)
{
printf("客户端嵌套字已经打开!\n");
}
else
{
printf("客户端的嵌套字打开失败!\n");
return 0;//结束
}
SOCKET clientSocket=socket(AF_INET,SOCK_STREAM,0);
SOCKADDR_IN clientsock_in;
clientsock_in.sin_addr.S_un.S_addr=inet_addr("127.0.0.1");
clientsock_in.sin_family=AF_INET;
clientsock_in.sin_port=htons(6000);
//bind(clientSocket,(SOCKADDR*)&clientsock_in,strlen(SOCKADDR));//注意第三个参数
//listen(clientSocket,5);
connect(clientSocket,(SOCKADDR*)&clientsock_in,sizeof(SOCKADDR));//开始连接
char receiveBuf[100];
recv(clientSocket,receiveBuf,101,0);
printf("%s\n",receiveBuf);
send(clientSocket,"hello,this is client",strlen("hello,this is client")+1,0);
closesocket(clientSocket);
WSACleanup();
return 0;
}
///////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
//服务器端代码：
#include <WINSOCK2.H>
#include <stdio.h>
#pragma comment(lib,"ws2_32.lib")
int main()
{
//创建套接字
WORD myVersionRequest;
WSADATA wsaData;
myVersionRequest=MAKEWORD(1,1);
int err;
err=WSAStartup(myVersionRequest,&wsaData);
if (!err)
{
printf("已打开套接字\n");
}
else
{
//进一步绑定套接字
printf("嵌套字未打开!");
return 0;
}
SOCKET serSocket=socket(AF_INET,SOCK_STREAM,0);//创建了可识别套接字
//需要绑定的参数
SOCKADDR_IN addr;
addr.sin_family=AF_INET;
addr.sin_addr.S_un.S_addr=htonl(INADDR_ANY);//ip地址
addr.sin_port=htons(6000);//绑定端口
bind(serSocket,(SOCKADDR*)&addr,sizeof(SOCKADDR));//绑定完成
listen(serSocket,5);//其中第二个参数代表能够接收的最多的连接数
//////////////////////////////////////////////////////////////////////////
//开始进行监听
//////////////////////////////////////////////////////////////////////////
SOCKADDR_IN clientsocket;
int len=sizeof(SOCKADDR);
while (1)
{
SOCKET serConn=accept(serSocket,(SOCKADDR*)&clientsocket,&len);//如果这里不是accept而是conection的话。。就会不断的监听
char sendBuf[100];
sprintf(sendBuf,"welcome %s to bejing",inet_ntoa(clientsocket.sin_addr));//找对对应的IP并且将这行字打印到那里
send(serConn,sendBuf,strlen(sendBuf)+1,0);
char receiveBuf[100];//接收
recv(serConn,receiveBuf,strlen(receiveBuf)+1,0);
printf("%s\n",receiveBuf);
closesocket(serConn);//关闭
WSACleanup();//释放资源的操作
}
return 0;
}