Enumeration Octet​Arrangement

public enum OctetArrangement 

Applications should prefer to work with individual octets wherever possible, as octets have a consistent numeric interpretation and binary representation across machines of different endianness.

When combining octets in to larger (e.g. 16- or 32-bit) integers, we have to consider that machines have a choice in how their octets are arranged in memory. For instance, the first piece of the IPv6 address 2608::3:5 consists of 2 octets, 0x26 and 0x08; if we created a 16-bit integer with those same octets arranged in that order, a big-endian machine would read this as having numeric value 9736 (for the purposes of integer-level operations, such as addition), whereas a little-endian machine would consider the same octets to contain the numeric value 2086.

Hence there are 2 ways to combine octets in to larger integers:

1. With the same octets in the same order in memory. We call this the binary interpretation, although it is more-commonly known as "network" byte order, or big-endian. As noted above, integers derived from the same address using the binary interpretation may have different numeric values on different machines.

2. By rearranging octets to give a consistent numeric value. We call this the numeric interpretation, although it is more-comonly known as "host" byte-order. For instance, when reading the first 16-bit piece of the above address on a little-endian machine, the octets 0x26 0x08 will be reordered to 0x08 0x26, so that the numeric value (9736) is the same as the hexadecimal number 0x2608. Assigning a group of octets using the numeric integer 9736 will similarly reorder the octets, so that they appear as the octet sequence 0x26 0x08 in the address.

case numeric

case binary

@inlinable public static var hostOrder: Self 

@inlinable public static var networkOrder: Self