cinc 2.0.post1

Usage

Signed types are named “intN” and unsigned types are named “uintN”, where “N” is the number of bits. Types for 8, 16, 32 and 64 bits are provided.

cinc integers can be constructed from Python ints or cast from another cinc type. If the source integer has more bits than the cinc type then it is truncated:

>>> x = uint32(0xFFFFFFFF)
>>> x
cinc.uint32(4294967295)
>>> uint16(x)
cinc.uint16(65535)

Arithmetic

cinc integers are compatible with python ints. Comparisons and operators cast the integer to their cinc type before performing the comparison or operation.

The result of the operation has the same type as the left operand:

>>> x = uint32(2)
>>> y = int32(2)
>>> x + y
cinc.uint32(4)
>>> y + x
cinc.int32(4)

This applies to operations with python ints too:

>>> x = uint32(2)
>>> x + 2
cinc.uint32(4)
>>> 2 + x
4

There are also methods for bit rotate operations and bit insertion and extraction:

>>> x = uint32(0x000000ff)
>>> hex(x.rrotate(8))
'0xff000000'
>>> y = uint32(0xff000000)
>>> hex(y.insert(0xff, 8, 8))
'0xff00ff00'
>>> hex(y.extract(16, 16))
'0xff00'

Python Fallback

If no C++ compiler is available during installation, then a Python source module based on ctypes is used instead. The Python version is significantly slower than the compiled C extension.

Changelog

2.0

• cinc types are now compatible with python ints.

• A pure Python module was added.