C Extension Module using Python

Writing a simple C extension module directly using Python’s extension API and no other tools. It is straightforward to make a handcrafted extension module for a simple C code. But first, we have to make sure that the C code has a proper header file.

Code #1 :

#include <math.h>

extern int gcd(int, int);
extern int in_mandel(double x0, double y0, int n);
extern int divide(int a, int b, int *remainder);
extern double avg(double *a, int n);

typedef struct Point
{
    double x, y;
} Point;

extern double distance(Point *p1, Point *p2);

The header would correspond to a library that has been separately compiled. The code below illustrates the basics of writing extension functions, following this assumption.

Code #2:

# include "Python.h"
# include "sample.h"

/* int gcd(int, int) */
static PyObject * py_gcd(PyObject * self, PyObject * args)
{
    int x, y, result;
    if (! PyArg_ParseTuple(args, "ii", &x, &y))
    {
        return NULL;
    }
    result = gcd(x, y);
    return Py_BuildValue("i", result);
}

/* int divide(int, int, int *) */
static PyObject * py_divide(PyObject * self, PyObject * args)
{
    int a, b, quotient, remainder;
    if (! PyArg_ParseTuple(args, "ii", &a, &b))
    {
        return NULL;
    }
    quotient = divide(a, b, &remainder);
    return Py_BuildValue("(ii)", quotient, remainder);
}

Code #3 : Module method table and structure

/* Module method table */
static PyMethodDef SampleMethods[] =
{
    {"gcd", py_gcd, METH_VARARGS, "Greatest common divisor"},
    {"divide", py_divide, METH_VARARGS, "Integer division"},
    { NULL, NULL, 0, NULL}
};

/* Module structure */
static struct PyModuleDef samplemodule =
{
    PyModuleDef_HEAD_INIT,
    "sample", /* name of module */
    "A sample module", /* Doc string (may be NULL) */
    -1, /* Size of per-interpreter state or -1 */
    SampleMethods /* Method table */
};

/* Module initialization function */
PyMODINIT_FUNC
PyInit_sample(void)
{
    return PyModule_Create(&samplemodule);
}

Code #4: Creating a setup.py python file for building the extension module.

# setup.py
from distutils.core import setup, Extension

setup(name='sample',
    ext_modules=[
            Extension('sample',
                    ['pysample.c'],
                    include_dirs = ['/some/dir'],
                    define_macros = [('FOO','1')],
                    undef_macros = ['BAR'],
                    library_dirs = ['/usr/local/lib'],
                    libraries = ['sample']
                    )
            ]
)

Code #5: Now simply use python3 buildlib.py build_ext --inplace, to build the resulting library.

bash% python3 setup.py build_ext --inplace
running build_ext
building 'sample' extension
gcc -fno-strict-aliasing -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes
  -I/usr/local/include/python3.3m -c pysample.c
  -o build/temp.macosx-10.6-x86_64-3.3/pysample.o
gcc -bundle -undefined dynamic_lookup
build/temp.macosx-10.6-x86_64-3.3/pysample.o \
  -L/usr/local/lib -lsample -o sample.so
bash %

The above code will create a shared library called sample.so.

Code #6 :

import sample

print ("gcd = ", sample.gcd(35, 42))

print ("\ndistance : ", sample.divide(42, 8))

Output :

gcd = 7

distance = (5, 2)

“Extending and Embedding the Python Interpreter” is a Python’s documentation that can be consulted before attempting any kind of handwritten extension.

In extension modules, functions can be written as shown in code snippet below.

Code #4 :

static PyObject *py_func(PyObject *self, PyObject *args)
{
    ...
}

• PyObject – C data type that represents any Python object. At a very high level, an extension function is a C function that receives a tuple of Python objects (in PyObject *args) and returns a new Python object as a result. The self argument to the function is unused for simple extension functions, but comes into play should you want to define new classes or object types in C.
• The PyArg_ParseTuple() function is used to convert values from Python to a C representation. As input, it takes a format string that indicates the required values, such as “i” for integer and “d” for double, as well as the addresses of C variables in which to place the converted results.
• Py_BuildValue() function is used to create Python objects from C data types. It also accepts a format code to indicate the desired type. In the extension functions, it is used to return results back to Python. One feature of Py_BuildValue() is that it can build more complicated kinds of objects, such as tuples and dictionaries.