Overflow checking with ctypes

Question

Is it possible to detect an integer overflow with the Python ctypes module?

The docs suggest that none of the primitives support overflow checking, but I'd like to write my own so I can use native c types for my CPU emulator.

Lawrence D'Oliveiro · Accepted Answer · 2012-02-07 11:36:21Z

1

ctypes isn’t for doing operations on integers, it’s for calling functions in shareable libraries. Any relevant integer arithmetic would still have to be done in Python.

Unless the sharable library is itself providing integer-arithmetic functions, in which case it would have to do the overflow detection.

answered Feb 7, 2012 at 11:36

Lawrence D'Oliveiro

2,8141 gold badge17 silver badges13 bronze badges

Sign up to request clarification or add additional context in comments.

2 Comments

mara004 Over a year ago

Well, the question is, what would happen if we have a python integer that is too big for the C type in question? Say we have an FFI function with argtype c_double, and call it with a python float that is bigger than what c_double can hold. The docs suggest to me this would silently cause an overflow.

mara004 Over a year ago

The next problem is, how should the caller check against overflows in a portable manner (given that max sizes may differ across platforms)? And then, do we have to remember doing this individually on each FFI function that may receive a previously-unchecked python int? Or should we wrap the c_* types to generally do overflow checking on parameter conversion?

Collectives™ on Stack Overflow

Overflow checking with ctypes

1 Answer 1

2 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

2 Comments

Your Answer

Sign up or log in

Post as a guest

Related