Python ctypes - TypeError: int expected instead of float

Question

I try to convert a python list to a ctype array. But somehow I always get the error "TypeError: int expected instead of float" in the line with this code:

    self.cValues = (ctypes.c_int * len(self.values))(*self.values)

I created self.values a few lines before. It's a normal python list. I am 100% sure, len(self.values) is an integer, I tested it with the type() function and the result was class 'int'. So what else is the problem with this line? Thanks for help.

Look at what you pasted here again... it's not even a valid expression. — Jeff Mercado, Aug 27 '14 at 01:05
@JeffMercado How do you mean that? I did it like in here: http://stackoverflow.com/questions/4145775/how-do-i-convert-a-python-list-into-a-c-array-by-using-ctypes Or do you mean the selfs? — Jan Gimmler, Aug 27 '14 at 01:11
Ok, saw the mistake, I removed it. Sorry for that, copy & paste fail. But that problem is still the same. — Jan Gimmler, Aug 27 '14 at 01:15
I am suspicious about what self.values really contains: can you print out its repr()? For instance, if values is a List of ints, your expression would work fine: `values = [0,1,2,3,4,5]; cValues = (ctypes.c_int * len(values))(*values)`. But if values is a List of floats, or ints mixed with floats, etc. it will bomb out: `values = [0,1.0,2,3,4,5]; cValues = (ctypes.c_int * len(values))(*values)` gets the same error you posted. — Josh Kupershmidt, Aug 27 '14 at 01:21
There actually are floats in the list. I thought the problem would be the len() function. Is there a way to make a ctype array with floats? I won't print out the repr, because the array is a 100000 lines 3D object coord array ^^ — Jan Gimmler, Aug 27 '14 at 01:27

score 2 · Accepted Answer · answered Aug 27 '14 at 02:20

The problem is not with the len() function but with self.values being a List of floats not a List of ints. Try it with a small self-contained example to see:

import ctypes
values = [0.0, 1.0, 2.0, 3.0, 4.0, 5.0]
cValues = (ctypes.c_int * len(values))(*values)

bombs out with the "TypeError: int expected instead of float" message you got. If you change values to be ints, i.e. values = [0, 1, 2, 3, 4, 5] it works fine. Or, since you say in your last comment that self.values is a List of floats, you probably want to use ctypes.c_float like so:

values = [0.0, 1.0, 2.0, 3.0, 4.0, 5.0]
cValues = (ctypes.c_float * len(values))

Python ctypes - TypeError: int expected instead of float

1 Answers1