在socket.recv_into中使用bytearray [英] using bytearray with socket.recv_into

问题描述

我正在做一些套接字IO，并使用bytearray对象作为缓冲区.我想使用csock.recv_into接收带有偏移量的数据到此缓冲区中，如下所示，以避免创建中间字符串对象.不幸的是，似乎字节数组不能以这种方式使用，并且下面的代码不起作用.

I am doing some socket IO, and using a bytearray object as a buffer. I would like to receive data with an offset into this buffer using csock.recv_into as shown below in order to avoid creating intermediary string objects. Unfortunately, it seems bytearrays can't be used this way, and the code below doesn't work.

buf    = bytearray(b" " * toread)
read   = 0
while(toread):
  nbytes = csock.recv_into(buf[read:],toread)
  toread -= nbytes
  read   += nbytes

所以我改用下面的代码，确实使用一个临时字符串(并且可以工作)...

So instead I am using the code below, which does use a temporary string (and works)...

buf    = bytearray(b" " * toread)
read   = 0
while(toread):
  tmp = csock.recv(toread)
  nbytes = len(tmp)
  buf[read:] = tmp
  toread -= nbytes
  read   += nbytes

是否有一种更优雅的方法不需要复制中间字符串?

Is there a more elegant way to do this that doesn't require copying intermediate strings around?

推荐答案

使用memoryview包裹bytearray:

buf = bytearray(toread)
view = memoryview(buf)
while toread:
    nbytes = sock.recv_into(view, toread)
    view = view[nbytes:] # slicing views is cheap
    toread -= nbytes

这篇关于在socket.recv_into中使用bytearray的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！