Bit fields in python?

[Eli Bendersky]

>
>
> I'm trying to rewrite a c program in python & encountered several problems.
> I have some data structures in my c program like below:
>
> typedef struct
> {
>     unsigned short size;
>
>     unsigned short reserved:8;
>     unsigned short var_a1:2;
>     unsigned short var_a2:2;
>     unsigned short var_a3:2;
>     unsigned short var_a4:2;
>
>     unsigned int var_a5;
> }structa;
>
>  typedef struct
>  {
>     unsigned short size;
>
>     unsigned char reserved:4;
>     unsigned char var_b1:1;
>     unsigned char var_b2:1;
>     unsigned char var_b3:1;
>     unsigned char var_b4:1;
>
>     structa var_structa;
> }structb;
>
> I tried to code the above in python but only got this far:
>
> class StructA(object):
>     def __init__(self, size=0)
>     self.size = size
>
> class StructB(object):
>     def __init__(self, size=0)
>
> Any equivalent for c data structures & bit fields in python? And how do I
> define var_structa (in structb) in python?
>
>

Bitfields are most commonly used for extreme space optimization - i.e.
shoving several variables and flags with carefully limited ranges into a
single work. In Python you rarely work this way (where such an optimization
is warranted, Python isn't the best tool for the job). However, as in your
use case, it is sometimes needed in Python in order to communicate with
other devices over the network or some other link.

In my work with Python and embedded devices I've found the construct library
(http://construct.wikispaces.com/) very useful. It allows to you very easily
define complex formats for frames/messages on the bit and byte level. The
idea is to use construct to encode and decode messages being sent to an
embedded device. It works great.

If you have further questions about this approach, feel free to ask.

Eli
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20100908/2239ad6b/attachment.html>