Socket programming struct variable

Sockaddr struct
Sockaddr's defect: sa_data mixed the target address and port information

Struct sockaddr { Unsigned short sa_family; Char sa_data [14]; };

Sa_family is the communication type, and the most common value is "af_inet" Sa_data14 bytes, including the destination address and port information in the socket

Sockaddr_in struct
The sockaddr_in struct solves the sockaddr defect and stores the port and ADDR separately in two variables.

Struct sockaddr_in { Short int sin_family; Unsigned short int sin_port; Struct in_addr sin_addr; Struct in_addr { Unsigned long s_addr; } Unsigned char sin_zero [8]; }

Both sin_port and sin_addr must be nbo
Generally, visualized numbers are HBO (local byte order)

The initial value of sin_zero should be set to zero using the function bzero.
The following statement is generally used:

Struct sockaddr_in cliaddr; Bzero (& cliaddr, sizeof (cliaddr ));

Basic configuration of sockaddr_in struct Variables

Struct sockaddr_in ina; Bzero (& Ina, sizeof (INA )); Ina. sin_family = af_inet; Ina. sin_port = htons (23 ); Ina. sin_addr.s_addr = inet_addr ("132.241.5.10 ");

Relationship between sockaddr and sockaddr_in
Generally, after assigning values to the sockaddr_in variable, forcibly convert the type and pass in the function that uses sockaddr for parameters.

• Sockaddr_in is used for socket definition and assignment.
• Sockaddr is used for function parameters.

The most typical source and target node socket Definition
For the source, destination, and Source and Destination Address ports, two socket variables need to be created.
Bind the cliaddr source address and Source Port
Set the destination address and destination port for connect and sendto.

Struct sockaddr_in servaddr, cliaddr; Create_socket (char * server_addr_string, unsigned int server_port) { Source socket assignment Bzero (& cliaddr, sizeof (cliaddr )); Cliaddr. sin_family = af_inet; Generally, the source address and port of TCP/UDP protocol are random. Cliaddr. sin_addr.s_addr = htons (inaddr_any ); Cliaddr. sin_port = htons (0 ); Destination socket assignment Bzero (& servaddr, sizeof (servaddr )); Servaddr. sin_family = af_inet; Inet_aton (server_addr_string, & servaddr. sin_addr ); Servaddr. sin_port = htons (server_port ); }

Network byte order (nbo)
Both sin_port and sin_addr of the struct must be nbo.

Host byte order HBO
Generally, visual numbers are HBO.

Nbo and HBO Conversion
Inet_addr () converts the string point format address to an unsigned long integer (unsigned long s_addr ;)
Inet_aton () converts the string point format address to nbo
Inet_ntoa () converts the nbo address into the string point format
Htons () "host to network short"
Htonl () "host to network long"
Ntohs () "network to host short"
Ntohl () "network to host long"
Htons () is commonly used. inet_addr () exactly corresponds to the port type and address type of the struct.

Three methods to assign an address to a socket

Inet_aton (server_addr_string, & myaddr. sin_addr );

Myaddr. sin_addr.s_addr = inet_addr ("132.241.5.10 ");

Inaddr_any Myaddr. sin_addr.s_addr = htons (inaddr_any ); Myaddr. sin_addr.s_addr = inaddr_any;

Two methods to assign port values to socket

# Define myport 3490 Myaddr. sin_port = htons (myport );

0 (random port) switch to nbo casually Myaddr. sin_port = htons (0 ); Myaddr. sin_port = 0;

Numbers such as htons/L and ntohs/l cannot be used for address translation. Because the addresses are in point format, the addresses can only be converted using numbers/strings, such as inet_aton and inet_ntoa;
The only htons that can be used for address translation is for inaddr_any.
Cliaddr. sin_addr.s_addr = htons (inaddr_any)

Differences between inet_addr () and inet_aton ()

• Inet_addr () is the return value type.

Struct sockaddr_in ina; Ina. sin_addr.s_addr = inet_addr ("132.241.5.10 ");

• Inet_aton () is a parameter pointer type

Struct sockaddr_in ina; Inet_aton ("132.241.5.10", & Ina. sin_addr );

Inet_ntoa converts the nbo address into the string point format
Parameter: struct variable. sinaddr
Return Value: String pointer

A1 = inet_ntoa (INA. sin_addr ); Printf ("address 1: % Sn", A1 );

Address 1: 132.241.5.10

Inet_addr () defect:-1 must be detected.
Because the result of inet_addr () is an integer,-1 is returned when an error occurs.
Ina. sin_addr.s_addr is of the unsigned long type.
-1 is displayed as 111111111 in long short, which is consistent with the IP address 255.255.255.255! The broadcast address is mistaken!