Python: convert 2 ints to 32 float

Question

How can I combine 2 ints to a single 32bit IEEE floating point ? (each of the 2 ints represent 16 bit) And in the opposite direction: How can I transform a python float into 2 16 bit ints?

(I need this because of modbus protocol - where 2x16 bit registers are treated as single 32 floating point number)

Answer 1

The standard struct module can be used to do this easily. Just be careful of your platform endianess but, other than that, it should be a pretty straight-forward application of pack() and unpack().

Answer 2

This code takes the 16 bits integers i1 and i2 and convert them to the floating point number 3.14, and vice versa.

from struct import *
# Two integers to a floating point
i1 = 0xC3F5
i2 = 0x4840
f = unpack('f',pack('>HH',i1,i2))[0]

# Floating point to two integers
i1, i2 = unpack('>HH',pack('f',3.14))