Struct++

Introduction

A 100% drop in replacement for the python struct module that adds support for decoding binary structures into namedtuples and decoding bitfields and other types such as fixed point numbers.

The two main use cases for structpp are:

• Naming the results from struct.unpack
• Decoding bitfields

Basic Usage

Installation

pip install structpp

Naming Results

To name the results of an unpacked binary structure, put the name in curly brackets after the format code:

import struct
import structpp

packed = struct.pack("<HH", 0xFFAB, 1)
unpacked = structpp.unpack("<H{val1}H{val2}", packed)

assert unpacked.val1 == 0xFFAB
assert unpacked.val2 == 1

You can also ask for the results as a dictionary or list instead of the default namedtuple:

unpacked = structpp.unpack("<H{val1}H{val2}", packed, asdict=True)
assert unpacked['val1'] == 0xFFAB
assert unpacked['val2'] == 1

unpacked = structpp.unpack("<H{val1}H{val2}", packed, aslist=True)
assert unpacked[0] == 0xFFAB
assert unpacked[1] == 1

You can freely mix name based and index based accessed when using the default namedtuple output:

unpacked = structpp.unpack("<H{val1}H{val2}", packed)
assert unpacked.val1 == 0xFFAB
assert unpacked[1] == 1

Decoding Bitfields

Using the same packed structure as above, we can decode the first 16-bit value into three values of width 4, 4, and 8 bits:

packed = struct.pack("<HH", 0xFFAB, 1)
unpacked = structpp.unpack("<H{val1: 4, val2: 4, val3: 8}H{val4}", packed)
assert unpacked.val1 == 0xB
assert unpacked.val2 == 0xA
assert unpacked.val3 == 0xFF
assert unpacked.val4 == 1