Goodday, I need to port the following C code in Python, but I can't find a good way to do the unsigned char arithmetic that ignores the overflow bits in python, ie. 255+1=0; 255+2=1 etc. The code below is used in a checksum calculation for a protocol that is implemented in C on the Arduino that I need to interface with.
unsigned char b[length];
unsigned char c1=0,c2=0,c3=0,c4=0;
for (i=0; i<length;i++)
{
c1+=b[i];
c2+=c1;
c3+=c2;
c4+=c3;
}
You may use % 256:
>>> (255 + 1) % 256
0
>>> (255 + 2) % 256
1
With your example, if b is a Python string:
>>> b = "thisisatest"
>>> c1, c2, c3, c4 = 0, 0, 0, 0
>>> for c in b:
... c1 = (c1 + ord(c)) % 256
... c2 = (c2 + c1) % 256
... c3 = (c3 + c2) % 256
... c4 = (c4 + c3) % 256
...
>>> c1, c2, c3, c4
... (181, 36, 46, 174)
ord() will work with "raw non-printable data": ord('\x00') == 0 (\x00 is a null byte).You can also use bitwise AND, which might be clearer. I also like hex notation for this.
255 & 0xff
256 & 0xff
257 & 0xff
-1 & 0xff
In Python there is no notion of signed/unsigned. If you want to handle overflow, you can simply use modular arithmetic.
(255 + 1) % 256 = 0
(250 + 2) % 256 = 252
(255 + 5) % 256 = 4
