Detailed explanation of Python struct (byte stream, grouping and unpacking implementation) module

Everyone should be familiar with the English word struct, because there is struct in c/c++, where struct is called a structure. Struct is also used in Python, which fully demonstrates that this struct should have a deep origin with the struct in c/c++. Python uses the struct module to perform conversions between Python values ​​and C structures to form Python bytes objects. It uses a format string as a compact description of the underlying C structure and converts it into a Python value based on this format string.

We know that python only defines 6 data types, strings, integers, floating point numbers, lists, tuples, and dictionaries. But there are some byte-type variables in C language. How to implement them in python? This is quite important, especially if data is to be transmitted over the network.

Python provides a struct module to provide conversion. Here are some methods in this module.

pack(fmt, v1, v2, ...) ------ Convert the values ​​v1, v2, ... into a string according to the format described by the given fmt.

unpack(fmt, bytes) ------ Reversely parse the bytes according to the format described by the given fmt and return a tuple.

calcsize(fmt) ------ Returns the size of the structure according to the format described by the given fmt.

struct.pack():

struct.pack is used to convert the Python value into a string according to the format character (because there is no byte (Byte) type in Python, the string here can be understood as a byte stream , or byte array). The function prototype is: struct.pack(fmt, v1, v2, ...), the parameter fmt is the format string, and the relevant information about the format string is introduced below. v1, v2, ... represent the Python values ​​to be converted. The following example converts two integers into strings (byte streams):

>>> import struct

>>> a=20

>>> b=400

>>> str=struct.pack ('ii',a,b) #Convert to a byte stream, although it is still a string, it can be transmitted on the network

>>> print len(str)             #ii represents two ints

8                                                                      The length is 8 bytes, which is exactly the length of two int type data

>>> print str

                                                                                                                                                                                                   #Where hexadecimal 0x00000014, 0x00001009 respectively represent 20 and 400

>>>>

From this we can arbitrarily group packages, such as the following packaging example, which only introduces the pack

format = "!HH% ds" % len(data)

buffer = struct.pack(format,opcode,blocknumber,data)

We need to pack a data and add some headers. Based on the following format character information, we know that H is unsigned short is 2 bytes, and s is char type. So this buffer is 2 bytes of opcode, 2 bytes of blocknumber, and len long char.

Python module - struct (byte stream, grouping and unpacking implementation)

struct.unpack():

Let’s continue the above example and run:

>>> a1,a2=struct.unpack( 'ii',str)

>>> print 'a1',a1

a1 20

>>> print 'a2=',a2

a2= 400

You can see that "ii" starts with four characters Section is the demarcation point, and the 8-byte str is divided into two int type integers.

struct.calcsize(): Used to calculate the size of the output in a specific format, which is a few bytes, such as:

>>> struct.calcsize('HH4s')

8

>>> struct.calcsize ('ii')

8

>>>>

>>> format='!HH%ds' % len('hello python')

>>> struct.calcsize(format)

16

>>>