Erlang -- bit syntax

(1) Basic concepts:

1. Binary: Raw binary data

2. the Erlang Virtual Machine optimizes binary input, output, and message transmission, which is very efficient.

3. If the number of digits in binary is a multiple of 8, it is called binary)

4. If the number of digits in the binary is not a multiple of 8, it is called a bitstring)

(2) Basic operations:

1. Binary: it is a column INTEGER (0 ~ 255) or string. Eg: <5, 10, 20>

2. Two important functions in binary operations:

1. term_to_binary (TERM)-> bin. % this function can convert any Erlang data type into a binary type.

2. binary_to_term (BIN)-> Inverse Operation of the term. % term_to_binary Function

3. Bit syntax expression:

<<E1, E2, ...，En>>Ei = Value |     Value:Size |     Value/TypeSpecifierList |     Value:Size/TypeSpecifierList

1. value, which can be any value that can be determined

2. Size, an integer, indicating the bit size occupied by value

3. typespecifierlist (Type-specified list) is a list separated by a hyphen, in the formEnd-sign-type-Unit

1. End: it specifies the byte sequence of the machine.

1. Big, large-end, default, network byte order

2. Little

3. Native is determined by the machine's CPU at runtime,

For example:

{<16 #12345678: 32/big, 16 #12345678: 32/little, 16 #12345678: 32/native, 16 #12345678: 32 >> }. the result is: {<86,120,120, 18,120, 120, >}. note that the hexadecimal format is not 10.

2. Sign: used for pattern matching to mark the data type, because some data is signed:

1. Sign

2. unsigned, default value

3. type indicates the type of the value: Yes

Integer | float | binary | bytes | bitstring | bits | utf8 | UTF16 | UTF32; default value: integer

4. Unit indicates the number of digits occupied by the Unit. Format: Unit: INTEGER:

1. The default unit values of integer, float, and bitstring are 1.

2. Binary, 8

3. No value is required for the utf8, UTF16, and UTF32 types,

For example:

<X: 4/little-signed-integer-unit: 8> % indicates a 32-digit integer that contains a signed integer and uses the binary data represented by a small end.

4. binary matching:

1. Package data:

2. parse data:

For example:

Red = 2 #10000001, Green = 2 #1000000000000010, Blue = 2 #10000100, IO: Format ("before unpack, red is ~ P, green is ~ P, blue is ~ P ~ N ", [red, green, blue]), RGB = <RED: 8/little-signed-integer-unit: 1, Green: 16/little-signed-integer-unit: 1, Blue: 8/little-signed-integer-unit: 1 >>,% RGB is the packaged data IO: format ("The package bin is ~ W ~ N ", [RGB]), <r1: 8/little-signed-integer-unit: 1, G1: 16/little-signed-integer-unit: 1, B1: 8/little-signed-integer-unit: 1 >>= RGB, % R1, G1, B1 is the parsed data IO: Format ("after unpack, red is ~ P, green is ~ P, blue is ~ P ~ N ", [R1, G1, B1]).

Note:

1. Because the default value is unsigned, I use signed here. You can calculate it.

Erlang -- bit syntax