from_buffer_copy（）的语法

Question

我正在读一本名为 Black Hat的书 Python Python编程针对黑客和五旬节的编程 贾斯汀·塞兹（Justin Seitz）。

这是本书中的代码：

import socket
import os
import struct
from ctypes import *

# host to listen
host = "10.0.2.15"


# our IP header
class IP(Structure):
    _fields_ = [
        ("ihl", c_ubyte, 4),
        ("version", c_ubyte,4),
        ("tos", c_ubyte),
        ("len", c_ushort),
        ("id", c_ushort),
        ("offset", c_ubyte),
        ("ttl", c_ubyte),
        ("protocol_num", c_ubyte),
        ("sum", c_ubyte),
        ("src", c_ulong),
        ("dst", c_ulong)
    ]
    
    def __new__(cls,socket_buffer=None):
        return cls.from_buffer_copy(socket_buffer)

在此代码中，在函数新的内部，有名为from_buffer_copy（）的函数。它是ctypes python库的功能。

我想知道它的作用以及使用它的语法是什么？

Python文档说该语法是from_buffer_copy（source [，offset]）。我发现了使用此语法的示例，但是文档提供的用法定义对我来说是模糊的。

：

Python文档 仅用一个参数看到了此功能，那么它如何实现？

任何帮助将不胜感激。

Answer 1

在文档from_buffer_copy（源[，offset]）中，方括号表示第二个参数是可选的。在代码中，它如下所示，其中给出了可选第二个参数的默认值：

def from_buffer_copy(self, source, offset=0): ...

这是.from_buffer_copy 和.from_buffer带有说明的工作示例。 。这些功能用于创建基于接收到字节数据的内存结构的python类：

import ctypes as ct

# A simple ctypes structure with two 4-byte fields.
# It will need at least 8 bytes of data to build this structure.
class Demo(ct.Structure):

    _fields_ = (('first', ct.c_uint32),
                ('second', ct.c_uint32))

    # defines the print representation to easily look at the fields
    def __repr__(self):
        return f'Demo(first=0x{self.first:08x}. second=0x{self.second:08x})'

# A 10-byte read-only bytes buffer.
# Extra bytes to demonstrate the offset parameter.
buf = bytes([1,2,3,4,5,6,7,8,9,10])

try:
    # This syntax constructs a Demo object that refers to the *same*
    # memory that buffer occupies, starting from the default offset of 0,
    # but to do so requires a writeable buffer
    demo1 = Demo.from_buffer(buf)
except TypeError as e:
    print(e)  # This error will print

# This builds the Demo object from a copy of the memory,
# so a read-only buffer can be used, starting from default offset 0.
demo2 = Demo.from_buffer_copy(buf)
print(demo2)

# This builds the object from offset 2, so the values of "first" and
# "second" will be different.
demo3 = Demo.from_buffer_copy(buf,2)
print(demo3)

# This is a writable buffer, so from_buffer will work
buf2 = bytearray([1,2,3,4,5,6,7,8])
demo4 = Demo.from_buffer(buf2)
print(demo4)

# Since demo4 references the *same* memory, changing writable buf2
# also changes demo4
buf2[1] = 0xff
print(demo4)

输出：

underlying buffer is not writable               # error message from demo1 attempt
Demo(first=0x04030201. second=0x08070605)       # demo2 copied from offset 0
Demo(first=0x06050403. second=0x0a090807)       # demo3 copied from offset 2
Demo(first=0x04030201. second=0x08070605)       # demo4
Demo(first=0x0403ff01. second=0x08070605)       # demo4 after write to buf2