为什么在IN6_ADDR中用union代表IP

Question

I am learning networking on Windows using C. I came across IN6_ADDR structure that represents an IPv6 address as a union of 2 arrays:

typedef struct in6_addr {
  union {
    UCHAR  Byte[16];
    USHORT Word[8];
  } u;
} IN6_ADDR, *PIN6_ADDR, *LPIN6_ADDR;

I can't wrap my head around why would someone desire a union of 2 arrays instead of just 1 array. What is the reason for such declaration? Note that both arrays are 128 bits long.

Answer 1

IPv6 addresses are generally written in a format that uses 2-byte groups, eg

2001:0db8:85a3:0000:0000:8a2e:0370:7334

Each of those groups corresponds to an element of the Word array.

But sometimes it's also useful to process each byte of the address. In this case, you would use the Byte array, rather than having to shift and mask elements of Word .

Answer 2

IPv6 addresses are typically represented as a group of 16-bit values, for example:

2001:0db8:0000:0000:0000:8a2e:0370:7334

The Word member allows you to read these 16 bit words individually instead of reading a single byte at a time.